diff options
Diffstat (limited to 'drivers/net/ethernet')
546 files changed, 43263 insertions, 15106 deletions
diff --git a/drivers/net/ethernet/3com/3c509.c b/drivers/net/ethernet/3com/3c509.c index b762176a1406..139d0120f511 100644 --- a/drivers/net/ethernet/3com/3c509.c +++ b/drivers/net/ethernet/3com/3c509.c @@ -85,7 +85,6 @@ #include <linux/device.h> #include <linux/eisa.h> #include <linux/bitops.h> -#include <linux/vermagic.h> #include <linux/uaccess.h> #include <asm/io.h> diff --git a/drivers/net/ethernet/3com/3c515.c b/drivers/net/ethernet/3com/3c515.c index 90312fcd6319..47b4215bb93b 100644 --- a/drivers/net/ethernet/3com/3c515.c +++ b/drivers/net/ethernet/3com/3c515.c @@ -22,7 +22,6 @@ */ -#include <linux/vermagic.h> #define DRV_NAME "3c515" #define CORKSCREW 1 diff --git a/drivers/net/ethernet/3com/3c59x.c b/drivers/net/ethernet/3com/3c59x.c index a2b7f7ab8170..5984b7033999 100644 --- a/drivers/net/ethernet/3com/3c59x.c +++ b/drivers/net/ethernet/3com/3c59x.c @@ -1149,7 +1149,7 @@ static int vortex_probe1(struct device *gendev, void __iomem *ioaddr, int irq, print_info = (vortex_debug > 1); if (print_info) - pr_info("See Documentation/networking/device_drivers/3com/vortex.txt\n"); + pr_info("See Documentation/networking/device_drivers/3com/vortex.rst\n"); pr_info("%s: 3Com %s %s at %p.\n", print_name, @@ -1954,7 +1954,7 @@ vortex_error(struct net_device *dev, int status) dev->name, tx_status); if (tx_status == 0x82) { pr_err("Probably a duplex mismatch. See " - "Documentation/networking/device_drivers/3com/vortex.txt\n"); + "Documentation/networking/device_drivers/3com/vortex.rst\n"); } dump_tx_ring(dev); } diff --git a/drivers/net/ethernet/3com/Kconfig b/drivers/net/ethernet/3com/Kconfig index 3a6fc99c6f32..7cc259893cb9 100644 --- a/drivers/net/ethernet/3com/Kconfig +++ b/drivers/net/ethernet/3com/Kconfig @@ -76,7 +76,7 @@ config VORTEX "Hurricane" (3c555/3cSOHO) PCI If you have such a card, say Y here. More specific information is in - <file:Documentation/networking/device_drivers/3com/vortex.txt> and + <file:Documentation/networking/device_drivers/3com/vortex.rst> and in the comments at the beginning of <file:drivers/net/ethernet/3com/3c59x.c>. diff --git a/drivers/net/ethernet/8390/ne2k-pci.c b/drivers/net/ethernet/8390/ne2k-pci.c index 42985a82321a..77d78b4c59c4 100644 --- a/drivers/net/ethernet/8390/ne2k-pci.c +++ b/drivers/net/ethernet/8390/ne2k-pci.c @@ -1,39 +1,43 @@ -/* ne2k-pci.c: A NE2000 clone on PCI bus driver for Linux. */ -/* - A Linux device driver for PCI NE2000 clones. - - Authors and other copyright holders: - 1992-2000 by Donald Becker, NE2000 core and various modifications. - 1995-1998 by Paul Gortmaker, core modifications and PCI support. - Copyright 1993 assigned to the United States Government as represented - by the Director, National Security Agency. - - This software may be used and distributed according to the terms of - the GNU General Public License (GPL), incorporated herein by reference. - Drivers based on or derived from this code fall under the GPL and must - retain the authorship, copyright and license notice. This file is not - a complete program and may only be used when the entire operating - system is licensed under the GPL. - - The author may be reached as becker@scyld.com, or C/O - Scyld Computing Corporation - 410 Severn Ave., Suite 210 - Annapolis MD 21403 - - Issues remaining: - People are making PCI ne2000 clones! Oh the horror, the horror... - Limited full-duplex support. -*/ +/* A Linux device driver for PCI NE2000 clones. + * + * Authors and other copyright holders: + * 1992-2000 by Donald Becker, NE2000 core and various modifications. + * 1995-1998 by Paul Gortmaker, core modifications and PCI support. + * Copyright 1993 assigned to the United States Government as represented + * by the Director, National Security Agency. + * + * This software may be used and distributed according to the terms of + * the GNU General Public License (GPL), incorporated herein by reference. + * Drivers based on or derived from this code fall under the GPL and must + * retain the authorship, copyright and license notice. This file is not + * a complete program and may only be used when the entire operating + * system is licensed under the GPL. + * + * The author may be reached as becker@scyld.com, or C/O + * Scyld Computing Corporation + * 410 Severn Ave., Suite 210 + * Annapolis MD 21403 + * + * Issues remaining: + * People are making PCI NE2000 clones! Oh the horror, the horror... + * Limited full-duplex support. + */ #define DRV_NAME "ne2k-pci" +#define DRV_DESCRIPTION "PCI NE2000 clone driver" +#define DRV_AUTHOR "Donald Becker / Paul Gortmaker" #define DRV_VERSION "1.03" #define DRV_RELDATE "9/22/2003" +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt /* The user-configurable values. - These may be modified when a driver module is loaded.*/ + * These may be modified when a driver module is loaded. + */ + +/* More are supported, limit only on options */ +#define MAX_UNITS 8 -#define MAX_UNITS 8 /* More are supported, limit only on options */ /* Used to pass the full-duplex flag, etc. */ static int full_duplex[MAX_UNITS]; static int options[MAX_UNITS]; @@ -52,7 +56,7 @@ static int options[MAX_UNITS]; #include <linux/netdevice.h> #include <linux/etherdevice.h> -#include <asm/io.h> +#include <linux/io.h> #include <asm/irq.h> #include <linux/uaccess.h> @@ -60,20 +64,14 @@ static int options[MAX_UNITS]; static u32 ne2k_msg_enable; -/* These identify the driver base version and may not be removed. */ -static const char version[] = - KERN_INFO DRV_NAME ".c:v" DRV_VERSION " " DRV_RELDATE - " D. Becker/P. Gortmaker\n"; - #if defined(__powerpc__) #define inl_le(addr) le32_to_cpu(inl(addr)) #define inw_le(addr) le16_to_cpu(inw(addr)) #endif -#define PFX DRV_NAME ": " - -MODULE_AUTHOR("Donald Becker / Paul Gortmaker"); -MODULE_DESCRIPTION("PCI NE2000 clone driver"); +MODULE_AUTHOR(DRV_AUTHOR); +MODULE_DESCRIPTION(DRV_DESCRIPTION); +MODULE_VERSION(DRV_VERSION); MODULE_LICENSE("GPL"); module_param_named(msg_enable, ne2k_msg_enable, uint, 0444); @@ -83,7 +81,8 @@ MODULE_PARM_DESC(msg_enable, "Debug message level (see linux/netdevice.h for bit MODULE_PARM_DESC(options, "Bit 5: full duplex"); MODULE_PARM_DESC(full_duplex, "full duplex setting(s) (1)"); -/* Some defines that people can play with if so inclined. */ +/* Some defines that people can play with if so inclined. + */ /* Use 32 bit data-movement operations instead of 16 bit. */ #define USE_LONGIO @@ -91,14 +90,18 @@ MODULE_PARM_DESC(full_duplex, "full duplex setting(s) (1)"); /* Do we implement the read before write bugfix ? */ /* #define NE_RW_BUGFIX */ -/* Flags. We rename an existing ei_status field to store flags! */ -/* Thus only the low 8 bits are usable for non-init-time flags. */ +/* Flags. We rename an existing ei_status field to store flags! + * Thus only the low 8 bits are usable for non-init-time flags. + */ #define ne2k_flags reg0 + enum { - ONLY_16BIT_IO=8, ONLY_32BIT_IO=4, /* Chip can do only 16/32-bit xfers. */ - FORCE_FDX=0x20, /* User override. */ - REALTEK_FDX=0x40, HOLTEK_FDX=0x80, - STOP_PG_0x60=0x100, + /* Chip can do only 16/32-bit xfers. */ + ONLY_16BIT_IO = 8, ONLY_32BIT_IO = 4, + /* User override. */ + FORCE_FDX = 0x20, + REALTEK_FDX = 0x40, HOLTEK_FDX = 0x80, + STOP_PG_0x60 = 0x100, }; enum ne2k_pci_chipsets { @@ -120,7 +123,7 @@ static struct { char *name; int flags; } pci_clone_list[] = { - {"RealTek RTL-8029", REALTEK_FDX}, + {"RealTek RTL-8029(AS)", REALTEK_FDX}, {"Winbond 89C940", 0}, {"Compex RL2000", 0}, {"KTI ET32P2", 0}, @@ -149,13 +152,14 @@ static const struct pci_device_id ne2k_pci_tbl[] = { { 0x8c4a, 0x1980, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CH_Winbond_89C940_8c4a }, { 0, } }; + MODULE_DEVICE_TABLE(pci, ne2k_pci_tbl); /* ---- No user-serviceable parts below ---- */ #define NE_BASE (dev->base_addr) -#define NE_CMD 0x00 +#define NE_CMD 0x00 #define NE_DATAPORT 0x10 /* NatSemi-defined port window offset. */ #define NE_RESET 0x1f /* Issue a read to reset, a write to clear. */ #define NE_IO_EXTENT 0x20 @@ -168,18 +172,20 @@ static int ne2k_pci_open(struct net_device *dev); static int ne2k_pci_close(struct net_device *dev); static void ne2k_pci_reset_8390(struct net_device *dev); -static void ne2k_pci_get_8390_hdr(struct net_device *dev, struct e8390_pkt_hdr *hdr, - int ring_page); +static void ne2k_pci_get_8390_hdr(struct net_device *dev, + struct e8390_pkt_hdr *hdr, int ring_page); static void ne2k_pci_block_input(struct net_device *dev, int count, - struct sk_buff *skb, int ring_offset); + struct sk_buff *skb, int ring_offset); static void ne2k_pci_block_output(struct net_device *dev, const int count, - const unsigned char *buf, const int start_page); + const unsigned char *buf, + const int start_page); static const struct ethtool_ops ne2k_pci_ethtool_ops; /* There is no room in the standard 8390 structure for extra info we need, - so we build a meta/outer-wrapper structure.. */ + * so we build a meta/outer-wrapper structure.. + */ struct ne2k_pci_card { struct net_device *dev; struct pci_dev *pci_dev; @@ -187,18 +193,17 @@ struct ne2k_pci_card { -/* - NEx000-clone boards have a Station Address (SA) PROM (SAPROM) in the packet - buffer memory space. By-the-spec NE2000 clones have 0x57,0x57 in bytes - 0x0e,0x0f of the SAPROM, while other supposed NE2000 clones must be - detected by their SA prefix. - - Reading the SAPROM from a word-wide card with the 8390 set in byte-wide - mode results in doubled values, which can be detected and compensated for. - - The probe is also responsible for initializing the card and filling - in the 'dev' and 'ei_status' structures. -*/ +/* NEx000-clone boards have a Station Address (SA) PROM (SAPROM) in the packet + * buffer memory space. By-the-spec NE2000 clones have 0x57,0x57 in bytes + * 0x0e,0x0f of the SAPROM, while other supposed NE2000 clones must be + * detected by their SA prefix. + * + * Reading the SAPROM from a word-wide card with the 8390 set in byte-wide + * mode results in doubled values, which can be detected and compensated for. + * + * The probe is also responsible for initializing the card and filling + * in the 'dev' and 'ei_status' structures. + */ static const struct net_device_ops ne2k_netdev_ops = { .ndo_open = ne2k_pci_open, @@ -208,7 +213,7 @@ static const struct net_device_ops ne2k_netdev_ops = { .ndo_get_stats = ei_get_stats, .ndo_set_rx_mode = ei_set_multicast_list, .ndo_validate_addr = eth_validate_addr, - .ndo_set_mac_address = eth_mac_addr, + .ndo_set_mac_address = eth_mac_addr, #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = ei_poll, #endif @@ -227,28 +232,21 @@ static int ne2k_pci_init_one(struct pci_dev *pdev, int flags = pci_clone_list[chip_idx].flags; struct ei_device *ei_local; -/* when built into the kernel, we only print version if device is found */ -#ifndef MODULE - static int printed_version; - if (!printed_version++) - printk(version); -#endif - fnd_cnt++; - i = pci_enable_device (pdev); + i = pci_enable_device(pdev); if (i) return i; - ioaddr = pci_resource_start (pdev, 0); + ioaddr = pci_resource_start(pdev, 0); irq = pdev->irq; - if (!ioaddr || ((pci_resource_flags (pdev, 0) & IORESOURCE_IO) == 0)) { + if (!ioaddr || ((pci_resource_flags(pdev, 0) & IORESOURCE_IO) == 0)) { dev_err(&pdev->dev, "no I/O resource at PCI BAR #0\n"); goto err_out; } - if (request_region (ioaddr, NE_IO_EXTENT, DRV_NAME) == NULL) { + if (!request_region(ioaddr, NE_IO_EXTENT, DRV_NAME)) { dev_err(&pdev->dev, "I/O resource 0x%x @ 0x%lx busy\n", NE_IO_EXTENT, ioaddr); goto err_out; @@ -261,14 +259,17 @@ static int ne2k_pci_init_one(struct pci_dev *pdev, /* Do a preliminary verification that we have a 8390. */ { int regd; - outb(E8390_NODMA+E8390_PAGE1+E8390_STOP, ioaddr + E8390_CMD); + + outb(E8390_NODMA + E8390_PAGE1 + E8390_STOP, ioaddr + E8390_CMD); regd = inb(ioaddr + 0x0d); outb(0xff, ioaddr + 0x0d); - outb(E8390_NODMA+E8390_PAGE0, ioaddr + E8390_CMD); - inb(ioaddr + EN0_COUNTER0); /* Clear the counter by reading. */ + outb(E8390_NODMA + E8390_PAGE0, ioaddr + E8390_CMD); + /* Clear the counter by reading. */ + inb(ioaddr + EN0_COUNTER0); if (inb(ioaddr + EN0_COUNTER0) != 0) { outb(reg0, ioaddr); - outb(regd, ioaddr + 0x0d); /* Restore the old values. */ + /* Restore the old values. */ + outb(regd, ioaddr + 0x0d); goto err_out_free_res; } } @@ -291,9 +292,9 @@ static int ne2k_pci_init_one(struct pci_dev *pdev, outb(inb(ioaddr + NE_RESET), ioaddr + NE_RESET); - /* This looks like a horrible timing loop, but it should never take - more than a few cycles. - */ + /* This looks like a horrible timing loop, but it should never + * take more than a few cycles. + */ while ((inb(ioaddr + EN0_ISR) & ENISR_RESET) == 0) /* Limit wait: '2' avoids jiffy roll-over. */ if (jiffies - reset_start_time > 2) { @@ -301,42 +302,53 @@ static int ne2k_pci_init_one(struct pci_dev *pdev, "Card failure (no reset ack).\n"); goto err_out_free_netdev; } - - outb(0xff, ioaddr + EN0_ISR); /* Ack all intr. */ + /* Ack all intr. */ + outb(0xff, ioaddr + EN0_ISR); } /* Read the 16 bytes of station address PROM. - We must first initialize registers, similar to NS8390_init(eifdev, 0). - We can't reliably read the SAPROM address without this. - (I learned the hard way!). */ + * We must first initialize registers, similar + * to NS8390_init(eifdev, 0). + * We can't reliably read the SAPROM address without this. + * (I learned the hard way!). + */ { struct {unsigned char value, offset; } program_seq[] = { - {E8390_NODMA+E8390_PAGE0+E8390_STOP, E8390_CMD}, /* Select page 0*/ - {0x49, EN0_DCFG}, /* Set word-wide access. */ - {0x00, EN0_RCNTLO}, /* Clear the count regs. */ + /* Select page 0 */ + {E8390_NODMA + E8390_PAGE0 + E8390_STOP, E8390_CMD}, + /* Set word-wide access */ + {0x49, EN0_DCFG}, + /* Clear the count regs. */ + {0x00, EN0_RCNTLO}, + /* Mask completion IRQ */ {0x00, EN0_RCNTHI}, - {0x00, EN0_IMR}, /* Mask completion irq. */ + {0x00, EN0_IMR}, {0xFF, EN0_ISR}, - {E8390_RXOFF, EN0_RXCR}, /* 0x20 Set to monitor */ - {E8390_TXOFF, EN0_TXCR}, /* 0x02 and loopback mode. */ + /* 0x20 Set to monitor */ + {E8390_RXOFF, EN0_RXCR}, + /* 0x02 and loopback mode */ + {E8390_TXOFF, EN0_TXCR}, {32, EN0_RCNTLO}, {0x00, EN0_RCNTHI}, - {0x00, EN0_RSARLO}, /* DMA starting at 0x0000. */ + /* DMA starting at 0x0000 */ + {0x00, EN0_RSARLO}, {0x00, EN0_RSARHI}, {E8390_RREAD+E8390_START, E8390_CMD}, }; for (i = 0; i < ARRAY_SIZE(program_seq); i++) - outb(program_seq[i].value, ioaddr + program_seq[i].offset); + outb(program_seq[i].value, + ioaddr + program_seq[i].offset); } /* Note: all PCI cards have at least 16 bit access, so we don't have - to check for 8 bit cards. Most cards permit 32 bit access. */ + * to check for 8 bit cards. Most cards permit 32 bit access. + */ if (flags & ONLY_32BIT_IO) { for (i = 0; i < 4 ; i++) ((u32 *)SA_prom)[i] = le32_to_cpu(inl(ioaddr + NE_DATAPORT)); } else - for(i = 0; i < 32 /*sizeof(SA_prom)*/; i++) + for (i = 0; i < 32 /* sizeof(SA_prom )*/; i++) SA_prom[i] = inb(ioaddr + NE_DATAPORT); /* We always set the 8390 registers for word mode. */ @@ -356,7 +368,7 @@ static int ne2k_pci_init_one(struct pci_dev *pdev, ei_status.word16 = 1; ei_status.ne2k_flags = flags; if (fnd_cnt < MAX_UNITS) { - if (full_duplex[fnd_cnt] > 0 || (options[fnd_cnt] & FORCE_FDX)) + if (full_duplex[fnd_cnt] > 0 || (options[fnd_cnt] & FORCE_FDX)) ei_status.ne2k_flags |= FORCE_FDX; } @@ -388,16 +400,15 @@ static int ne2k_pci_init_one(struct pci_dev *pdev, return 0; err_out_free_netdev: - free_netdev (dev); + free_netdev(dev); err_out_free_res: - release_region (ioaddr, NE_IO_EXTENT); + release_region(ioaddr, NE_IO_EXTENT); err_out: pci_disable_device(pdev); return -ENODEV; } -/* - * Magic incantation sequence for full duplex on the supported cards. +/* Magic incantation sequence for full duplex on the supported cards. */ static inline int set_realtek_fdx(struct net_device *dev) { @@ -431,7 +442,9 @@ static int ne2k_pci_set_fdx(struct net_device *dev) static int ne2k_pci_open(struct net_device *dev) { - int ret = request_irq(dev->irq, ei_interrupt, IRQF_SHARED, dev->name, dev); + int ret = request_irq(dev->irq, ei_interrupt, IRQF_SHARED, + dev->name, dev); + if (ret) return ret; @@ -450,7 +463,8 @@ static int ne2k_pci_close(struct net_device *dev) } /* Hard reset the card. This used to pause for the same period that a - 8390 reset command required, but that shouldn't be necessary. */ + * 8390 reset command required, but that shouldn't be necessary. + */ static void ne2k_pci_reset_8390(struct net_device *dev) { unsigned long reset_start_time = jiffies; @@ -467,31 +481,34 @@ static void ne2k_pci_reset_8390(struct net_device *dev) /* This check _should_not_ be necessary, omit eventually. */ while ((inb(NE_BASE+EN0_ISR) & ENISR_RESET) == 0) if (jiffies - reset_start_time > 2) { - netdev_err(dev, "ne2k_pci_reset_8390() did not complete.\n"); + netdev_err(dev, "%s did not complete.\n", __func__); break; } - outb(ENISR_RESET, NE_BASE + EN0_ISR); /* Ack intr. */ + /* Ack intr. */ + outb(ENISR_RESET, NE_BASE + EN0_ISR); } /* Grab the 8390 specific header. Similar to the block_input routine, but - we don't need to be concerned with ring wrap as the header will be at - the start of a page, so we optimize accordingly. */ + * we don't need to be concerned with ring wrap as the header will be at + * the start of a page, so we optimize accordingly. + */ -static void ne2k_pci_get_8390_hdr(struct net_device *dev, struct e8390_pkt_hdr *hdr, int ring_page) +static void ne2k_pci_get_8390_hdr(struct net_device *dev, + struct e8390_pkt_hdr *hdr, int ring_page) { long nic_base = dev->base_addr; - /* This *shouldn't* happen. If it does, it's the last thing you'll see */ + /* This *shouldn't* happen. If it does, it's the last thing you'll see + */ if (ei_status.dmaing) { - netdev_err(dev, "DMAing conflict in ne2k_pci_get_8390_hdr " - "[DMAstat:%d][irqlock:%d].\n", - ei_status.dmaing, ei_status.irqlock); + netdev_err(dev, "DMAing conflict in %s [DMAstat:%d][irqlock:%d].\n", + __func__, ei_status.dmaing, ei_status.irqlock); return; } ei_status.dmaing |= 0x01; - outb(E8390_NODMA+E8390_PAGE0+E8390_START, nic_base+ NE_CMD); + outb(E8390_NODMA + E8390_PAGE0 + E8390_START, nic_base + NE_CMD); outb(sizeof(struct e8390_pkt_hdr), nic_base + EN0_RCNTLO); outb(0, nic_base + EN0_RCNTHI); outb(0, nic_base + EN0_RSARLO); /* On page boundary */ @@ -499,20 +516,22 @@ static void ne2k_pci_get_8390_hdr(struct net_device *dev, struct e8390_pkt_hdr * outb(E8390_RREAD+E8390_START, nic_base + NE_CMD); if (ei_status.ne2k_flags & ONLY_16BIT_IO) { - insw(NE_BASE + NE_DATAPORT, hdr, sizeof(struct e8390_pkt_hdr)>>1); + insw(NE_BASE + NE_DATAPORT, hdr, + sizeof(struct e8390_pkt_hdr) >> 1); } else { - *(u32*)hdr = le32_to_cpu(inl(NE_BASE + NE_DATAPORT)); + *(u32 *)hdr = le32_to_cpu(inl(NE_BASE + NE_DATAPORT)); le16_to_cpus(&hdr->count); } - - outb(ENISR_RDC, nic_base + EN0_ISR); /* Ack intr. */ + /* Ack intr. */ + outb(ENISR_RDC, nic_base + EN0_ISR); ei_status.dmaing &= ~0x01; } /* Block input and output, similar to the Crynwr packet driver. If you - are porting to a new ethercard, look at the packet driver source for hints. - The NEx000 doesn't share the on-board packet memory -- you have to put - the packet out through the "remote DMA" dataport using outb. */ + *are porting to a new ethercard, look at the packet driver source for hints. + *The NEx000 doesn't share the on-board packet memory -- you have to put + *the packet out through the "remote DMA" dataport using outb. + */ static void ne2k_pci_block_input(struct net_device *dev, int count, struct sk_buff *skb, int ring_offset) @@ -520,30 +539,30 @@ static void ne2k_pci_block_input(struct net_device *dev, int count, long nic_base = dev->base_addr; char *buf = skb->data; - /* This *shouldn't* happen. If it does, it's the last thing you'll see */ + /* This *shouldn't* happen. + * If it does, it's the last thing you'll see. + */ if (ei_status.dmaing) { - netdev_err(dev, "DMAing conflict in ne2k_pci_block_input " - "[DMAstat:%d][irqlock:%d].\n", - ei_status.dmaing, ei_status.irqlock); + netdev_err(dev, "DMAing conflict in %s [DMAstat:%d][irqlock:%d]\n", + __func__, ei_status.dmaing, ei_status.irqlock); return; } ei_status.dmaing |= 0x01; if (ei_status.ne2k_flags & ONLY_32BIT_IO) count = (count + 3) & 0xFFFC; - outb(E8390_NODMA+E8390_PAGE0+E8390_START, nic_base+ NE_CMD); + outb(E8390_NODMA + E8390_PAGE0 + E8390_START, nic_base + NE_CMD); outb(count & 0xff, nic_base + EN0_RCNTLO); outb(count >> 8, nic_base + EN0_RCNTHI); outb(ring_offset & 0xff, nic_base + EN0_RSARLO); outb(ring_offset >> 8, nic_base + EN0_RSARHI); - outb(E8390_RREAD+E8390_START, nic_base + NE_CMD); + outb(E8390_RREAD + E8390_START, nic_base + NE_CMD); if (ei_status.ne2k_flags & ONLY_16BIT_IO) { - insw(NE_BASE + NE_DATAPORT,buf,count>>1); - if (count & 0x01) { + insw(NE_BASE + NE_DATAPORT, buf, count >> 1); + if (count & 0x01) buf[count-1] = inb(NE_BASE + NE_DATAPORT); - } } else { - insl(NE_BASE + NE_DATAPORT, buf, count>>2); + insl(NE_BASE + NE_DATAPORT, buf, count >> 2); if (count & 3) { buf += count & ~3; if (count & 2) { @@ -556,30 +575,32 @@ static void ne2k_pci_block_input(struct net_device *dev, int count, *buf = inb(NE_BASE + NE_DATAPORT); } } - - outb(ENISR_RDC, nic_base + EN0_ISR); /* Ack intr. */ + /* Ack intr. */ + outb(ENISR_RDC, nic_base + EN0_ISR); ei_status.dmaing &= ~0x01; } static void ne2k_pci_block_output(struct net_device *dev, int count, - const unsigned char *buf, const int start_page) + const unsigned char *buf, const int start_page) { long nic_base = NE_BASE; unsigned long dma_start; /* On little-endian it's always safe to round the count up for - word writes. */ + * word writes. + */ if (ei_status.ne2k_flags & ONLY_32BIT_IO) count = (count + 3) & 0xFFFC; else if (count & 0x01) count++; - /* This *shouldn't* happen. If it does, it's the last thing you'll see */ + /* This *shouldn't* happen. + * If it does, it's the last thing you'll see. + */ if (ei_status.dmaing) { - netdev_err(dev, "DMAing conflict in ne2k_pci_block_output." - "[DMAstat:%d][irqlock:%d]\n", - ei_status.dmaing, ei_status.irqlock); + netdev_err(dev, "DMAing conflict in %s [DMAstat:%d][irqlock:%d]\n", + __func__, ei_status.dmaing, ei_status.irqlock); return; } ei_status.dmaing |= 0x01; @@ -588,9 +609,10 @@ static void ne2k_pci_block_output(struct net_device *dev, int count, #ifdef NE8390_RW_BUGFIX /* Handle the read-before-write bug the same way as the - Crynwr packet driver -- the NatSemi method doesn't work. - Actually this doesn't always work either, but if you have - problems with your NEx000 this is better than nothing! */ + * Crynwr packet driver -- the NatSemi method doesn't work. + * Actually this doesn't always work either, but if you have + * problems with your NEx000 this is better than nothing! + */ outb(0x42, nic_base + EN0_RCNTLO); outb(0x00, nic_base + EN0_RCNTHI); outb(0x42, nic_base + EN0_RSARLO); @@ -599,16 +621,16 @@ static void ne2k_pci_block_output(struct net_device *dev, int count, #endif outb(ENISR_RDC, nic_base + EN0_ISR); - /* Now the normal output. */ + /* Now the normal output. */ outb(count & 0xff, nic_base + EN0_RCNTLO); outb(count >> 8, nic_base + EN0_RCNTHI); outb(0x00, nic_base + EN0_RSARLO); outb(start_page, nic_base + EN0_RSARHI); outb(E8390_RWRITE+E8390_START, nic_base + NE_CMD); if (ei_status.ne2k_flags & ONLY_16BIT_IO) { - outsw(NE_BASE + NE_DATAPORT, buf, count>>1); + outsw(NE_BASE + NE_DATAPORT, buf, count >> 1); } else { - outsl(NE_BASE + NE_DATAPORT, buf, count>>2); + outsl(NE_BASE + NE_DATAPORT, buf, count >> 2); if (count & 3) { buf += count & ~3; if (count & 2) { @@ -623,14 +645,15 @@ static void ne2k_pci_block_output(struct net_device *dev, int count, dma_start = jiffies; while ((inb(nic_base + EN0_ISR) & ENISR_RDC) == 0) - if (jiffies - dma_start > 2) { /* Avoid clock roll-over. */ + /* Avoid clock roll-over. */ + if (jiffies - dma_start > 2) { netdev_warn(dev, "timeout waiting for Tx RDC.\n"); ne2k_pci_reset_8390(dev); - NS8390_init(dev,1); + NS8390_init(dev, 1); break; } - - outb(ENISR_RDC, nic_base + EN0_ISR); /* Ack intr. */ + /* Ack intr. */ + outb(ENISR_RDC, nic_base + EN0_ISR); ei_status.dmaing &= ~0x01; } @@ -640,9 +663,9 @@ static void ne2k_pci_get_drvinfo(struct net_device *dev, struct ei_device *ei = netdev_priv(dev); struct pci_dev *pci_dev = (struct pci_dev *) ei->priv; - strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); - strlcpy(info->version, DRV_VERSION, sizeof(info->version)); - strlcpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info)); + strscpy(info->driver, DRV_NAME, sizeof(info->driver)); + strscpy(info->version, DRV_VERSION, sizeof(info->version)); + strscpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info)); } static u32 ne2k_pci_get_msglevel(struct net_device *dev) @@ -677,9 +700,9 @@ static void ne2k_pci_remove_one(struct pci_dev *pdev) } #ifdef CONFIG_PM -static int ne2k_pci_suspend (struct pci_dev *pdev, pm_message_t state) +static int ne2k_pci_suspend(struct pci_dev *pdev, pm_message_t state) { - struct net_device *dev = pci_get_drvdata (pdev); + struct net_device *dev = pci_get_drvdata(pdev); netif_device_detach(dev); pci_save_state(pdev); @@ -689,9 +712,9 @@ static int ne2k_pci_suspend (struct pci_dev *pdev, pm_message_t state) return 0; } -static int ne2k_pci_resume (struct pci_dev *pdev) +static int ne2k_pci_resume(struct pci_dev *pdev) { - struct net_device *dev = pci_get_drvdata (pdev); + struct net_device *dev = pci_get_drvdata(pdev); int rc; pci_set_power_state(pdev, PCI_D0); @@ -718,24 +741,20 @@ static struct pci_driver ne2k_driver = { #ifdef CONFIG_PM .suspend = ne2k_pci_suspend, .resume = ne2k_pci_resume, -#endif /* CONFIG_PM */ +#endif }; static int __init ne2k_pci_init(void) { -/* when a module, this is printed whether or not devices are found in probe */ -#ifdef MODULE - printk(version); -#endif return pci_register_driver(&ne2k_driver); } static void __exit ne2k_pci_cleanup(void) { - pci_unregister_driver (&ne2k_driver); + pci_unregister_driver(&ne2k_driver); } module_init(ne2k_pci_init); diff --git a/drivers/net/ethernet/adaptec/starfire.c b/drivers/net/ethernet/adaptec/starfire.c index 2db42211329f..a64191fc2af9 100644 --- a/drivers/net/ethernet/adaptec/starfire.c +++ b/drivers/net/ethernet/adaptec/starfire.c @@ -45,7 +45,6 @@ #include <asm/processor.h> /* Processor type for cache alignment. */ #include <linux/uaccess.h> #include <asm/io.h> -#include <linux/vermagic.h> /* * The current frame processor firmware fails to checksum a fragment diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c index 1b19385ad8a9..865892c1f23f 100644 --- a/drivers/net/ethernet/agere/et131x.c +++ b/drivers/net/ethernet/agere/et131x.c @@ -714,11 +714,11 @@ static int et131x_init_eeprom(struct et131x_adapter *adapter) * gather additional information that normally would * come from the eeprom, like MAC Address */ - adapter->has_eeprom = 0; + adapter->has_eeprom = false; return -EIO; } } - adapter->has_eeprom = 1; + adapter->has_eeprom = true; /* Read the EEPROM for information regarding LED behavior. Refer to * et131x_xcvr_init() for its use. diff --git a/drivers/net/ethernet/allwinner/sun4i-emac.c b/drivers/net/ethernet/allwinner/sun4i-emac.c index 18d3b4340bd4..b3b8a8010142 100644 --- a/drivers/net/ethernet/allwinner/sun4i-emac.c +++ b/drivers/net/ethernet/allwinner/sun4i-emac.c @@ -417,7 +417,7 @@ static void emac_timeout(struct net_device *dev, unsigned int txqueue) /* Hardware start transmission. * Send a packet to media from the upper layer. */ -static int emac_start_xmit(struct sk_buff *skb, struct net_device *dev) +static netdev_tx_t emac_start_xmit(struct sk_buff *skb, struct net_device *dev) { struct emac_board_info *db = netdev_priv(dev); unsigned long channel; @@ -425,7 +425,7 @@ static int emac_start_xmit(struct sk_buff *skb, struct net_device *dev) channel = db->tx_fifo_stat & 3; if (channel == 3) - return 1; + return NETDEV_TX_BUSY; channel = (channel == 1 ? 1 : 0); diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c index 1671c1f36691..907125abef2c 100644 --- a/drivers/net/ethernet/altera/altera_tse_main.c +++ b/drivers/net/ethernet/altera/altera_tse_main.c @@ -554,7 +554,7 @@ static irqreturn_t altera_isr(int irq, void *dev_id) * physically contiguous fragment starting at * skb->data, for length of skb_headlen(skb). */ -static int tse_start_xmit(struct sk_buff *skb, struct net_device *dev) +static netdev_tx_t tse_start_xmit(struct sk_buff *skb, struct net_device *dev) { struct altera_tse_private *priv = netdev_priv(dev); unsigned int txsize = priv->tx_ring_size; @@ -562,7 +562,7 @@ static int tse_start_xmit(struct sk_buff *skb, struct net_device *dev) struct tse_buffer *buffer = NULL; int nfrags = skb_shinfo(skb)->nr_frags; unsigned int nopaged_len = skb_headlen(skb); - enum netdev_tx ret = NETDEV_TX_OK; + netdev_tx_t ret = NETDEV_TX_OK; dma_addr_t dma_addr; spin_lock_bh(&priv->tx_lock); diff --git a/drivers/net/ethernet/amazon/ena/ena_admin_defs.h b/drivers/net/ethernet/amazon/ena/ena_admin_defs.h index 8baf847e8622..336742f6e3c3 100644 --- a/drivers/net/ethernet/amazon/ena/ena_admin_defs.h +++ b/drivers/net/ethernet/amazon/ena/ena_admin_defs.h @@ -404,6 +404,10 @@ struct ena_admin_basic_stats { u32 rx_drops_low; u32 rx_drops_high; + + u32 tx_drops_low; + + u32 tx_drops_high; }; struct ena_admin_acq_get_stats_resp { @@ -764,8 +768,8 @@ enum ena_admin_os_type { ENA_ADMIN_OS_DPDK = 3, ENA_ADMIN_OS_FREEBSD = 4, ENA_ADMIN_OS_IPXE = 5, - ENA_ADMIN_OS_ESXI = 6, - ENA_ADMIN_OS_GROUPS_NUM = 6, + ENA_ADMIN_OS_ESXI = 6, + ENA_ADMIN_OS_GROUPS_NUM = 6, }; struct ena_admin_host_info { @@ -809,7 +813,8 @@ struct ena_admin_host_info { u16 reserved; - /* 1 :0 : reserved + /* 0 : reserved + * 1 : rx_offset * 2 : interrupt_moderation * 31:3 : reserved */ @@ -1017,6 +1022,10 @@ struct ena_admin_aenq_keep_alive_desc { u32 rx_drops_low; u32 rx_drops_high; + + u32 tx_drops_low; + + u32 tx_drops_high; }; struct ena_admin_ena_mmio_req_read_less_resp { @@ -1116,6 +1125,8 @@ struct ena_admin_ena_mmio_req_read_less_resp { #define ENA_ADMIN_HOST_INFO_DEVICE_MASK GENMASK(7, 3) #define ENA_ADMIN_HOST_INFO_BUS_SHIFT 8 #define ENA_ADMIN_HOST_INFO_BUS_MASK GENMASK(15, 8) +#define ENA_ADMIN_HOST_INFO_RX_OFFSET_SHIFT 1 +#define ENA_ADMIN_HOST_INFO_RX_OFFSET_MASK BIT(1) #define ENA_ADMIN_HOST_INFO_INTERRUPT_MODERATION_SHIFT 2 #define ENA_ADMIN_HOST_INFO_INTERRUPT_MODERATION_MASK BIT(2) @@ -1125,4 +1136,4 @@ struct ena_admin_ena_mmio_req_read_less_resp { /* aenq_link_change_desc */ #define ENA_ADMIN_AENQ_LINK_CHANGE_DESC_LINK_STATUS_MASK BIT(0) -#endif /*_ENA_ADMIN_H_ */ +#endif /* _ENA_ADMIN_H_ */ diff --git a/drivers/net/ethernet/amazon/ena/ena_com.c b/drivers/net/ethernet/amazon/ena/ena_com.c index a250046b8e18..432f143559a1 100644 --- a/drivers/net/ethernet/amazon/ena/ena_com.c +++ b/drivers/net/ethernet/amazon/ena/ena_com.c @@ -62,7 +62,9 @@ #define ENA_REGS_ADMIN_INTR_MASK 1 -#define ENA_POLL_MS 5 +#define ENA_MIN_ADMIN_POLL_US 100 + +#define ENA_MAX_ADMIN_POLL_US 5000 /*****************************************************************************/ /*****************************************************************************/ @@ -200,17 +202,17 @@ static void comp_ctxt_release(struct ena_com_admin_queue *queue, static struct ena_comp_ctx *get_comp_ctxt(struct ena_com_admin_queue *queue, u16 command_id, bool capture) { - if (unlikely(!queue->comp_ctx)) { - pr_err("Completion context is NULL\n"); - return NULL; - } - if (unlikely(command_id >= queue->q_depth)) { pr_err("command id is larger than the queue size. cmd_id: %u queue size %d\n", command_id, queue->q_depth); return NULL; } + if (unlikely(!queue->comp_ctx)) { + pr_err("Completion context is NULL\n"); + return NULL; + } + if (unlikely(queue->comp_ctx[command_id].occupied && capture)) { pr_err("Completion context is occupied\n"); return NULL; @@ -375,7 +377,7 @@ static int ena_com_init_io_sq(struct ena_com_dev *ena_dev, io_sq->bounce_buf_ctrl.next_to_use = 0; size = io_sq->bounce_buf_ctrl.buffer_size * - io_sq->bounce_buf_ctrl.buffers_num; + io_sq->bounce_buf_ctrl.buffers_num; dev_node = dev_to_node(ena_dev->dmadev); set_dev_node(ena_dev->dmadev, ctx->numa_node); @@ -523,9 +525,6 @@ static int ena_com_comp_status_to_errno(u8 comp_status) if (unlikely(comp_status != 0)) pr_err("admin command failed[%u]\n", comp_status); - if (unlikely(comp_status > ENA_ADMIN_UNKNOWN_ERROR)) - return -EINVAL; - switch (comp_status) { case ENA_ADMIN_SUCCESS: return 0; @@ -540,7 +539,14 @@ static int ena_com_comp_status_to_errno(u8 comp_status) return -EINVAL; } - return 0; + return -EINVAL; +} + +static void ena_delay_exponential_backoff_us(u32 exp, u32 delay_us) +{ + delay_us = max_t(u32, ENA_MIN_ADMIN_POLL_US, delay_us); + delay_us = min_t(u32, delay_us * (1U << exp), ENA_MAX_ADMIN_POLL_US); + usleep_range(delay_us, 2 * delay_us); } static int ena_com_wait_and_process_admin_cq_polling(struct ena_comp_ctx *comp_ctx, @@ -549,6 +555,7 @@ static int ena_com_wait_and_process_admin_cq_polling(struct ena_comp_ctx *comp_c unsigned long flags = 0; unsigned long timeout; int ret; + u32 exp = 0; timeout = jiffies + usecs_to_jiffies(admin_queue->completion_timeout); @@ -572,7 +579,8 @@ static int ena_com_wait_and_process_admin_cq_polling(struct ena_comp_ctx *comp_c goto err; } - msleep(ENA_POLL_MS); + ena_delay_exponential_backoff_us(exp++, + admin_queue->ena_dev->ena_min_poll_delay_us); } if (unlikely(comp_ctx->status == ENA_CMD_ABORTED)) { @@ -702,8 +710,7 @@ static int ena_com_config_llq_info(struct ena_com_dev *ena_dev, /* The desc list entry size should be whole multiply of 8 * This requirement comes from __iowrite64_copy() */ - pr_err("illegal entry size %d\n", - llq_info->desc_list_entry_size); + pr_err("illegal entry size %d\n", llq_info->desc_list_entry_size); return -EINVAL; } @@ -775,7 +782,7 @@ static int ena_com_wait_and_process_admin_cq_interrupts(struct ena_comp_ctx *com if (admin_queue->auto_polling) admin_queue->polling = true; } else { - pr_err("The ena device doesn't send a completion for the admin cmd %d status %d\n", + pr_err("The ena device didn't send a completion for the admin cmd %d status %d\n", comp_ctx->cmd_opcode, comp_ctx->status); } /* Check if shifted to polling mode. @@ -943,12 +950,13 @@ static void ena_com_io_queue_free(struct ena_com_dev *ena_dev, static int wait_for_reset_state(struct ena_com_dev *ena_dev, u32 timeout, u16 exp_state) { - u32 val, i; + u32 val, exp = 0; + unsigned long timeout_stamp; - /* Convert timeout from resolution of 100ms to ENA_POLL_MS */ - timeout = (timeout * 100) / ENA_POLL_MS; + /* Convert timeout from resolution of 100ms to us resolution. */ + timeout_stamp = jiffies + usecs_to_jiffies(100 * 1000 * timeout); - for (i = 0; i < timeout; i++) { + while (1) { val = ena_com_reg_bar_read32(ena_dev, ENA_REGS_DEV_STS_OFF); if (unlikely(val == ENA_MMIO_READ_TIMEOUT)) { @@ -960,10 +968,11 @@ static int wait_for_reset_state(struct ena_com_dev *ena_dev, u32 timeout, exp_state) return 0; - msleep(ENA_POLL_MS); - } + if (time_is_before_jiffies(timeout_stamp)) + return -ETIME; - return -ETIME; + ena_delay_exponential_backoff_us(exp++, ena_dev->ena_min_poll_delay_us); + } } static bool ena_com_check_supported_feature_id(struct ena_com_dev *ena_dev, @@ -1067,16 +1076,10 @@ static void ena_com_hash_key_fill_default_key(struct ena_com_dev *ena_dev) static int ena_com_hash_key_allocate(struct ena_com_dev *ena_dev) { struct ena_rss *rss = &ena_dev->rss; - struct ena_admin_get_feat_resp get_resp; - int rc; - rc = ena_com_get_feature_ex(ena_dev, &get_resp, - ENA_ADMIN_RSS_HASH_FUNCTION, - ena_dev->rss.hash_key_dma_addr, - sizeof(ena_dev->rss.hash_key), 0); - if (unlikely(rc)) { + if (!ena_com_check_supported_feature_id(ena_dev, + ENA_ADMIN_RSS_HASH_FUNCTION)) return -EOPNOTSUPP; - } rss->hash_key = dma_alloc_coherent(ena_dev->dmadev, sizeof(*rss->hash_key), @@ -1290,13 +1293,9 @@ static int ena_com_ind_tbl_convert_to_device(struct ena_com_dev *ena_dev) static void ena_com_update_intr_delay_resolution(struct ena_com_dev *ena_dev, u16 intr_delay_resolution) { - /* Initial value of intr_delay_resolution might be 0 */ - u16 prev_intr_delay_resolution = - ena_dev->intr_delay_resolution ? - ena_dev->intr_delay_resolution : - ENA_DEFAULT_INTR_DELAY_RESOLUTION; + u16 prev_intr_delay_resolution = ena_dev->intr_delay_resolution; - if (!intr_delay_resolution) { + if (unlikely(!intr_delay_resolution)) { pr_err("Illegal intr_delay_resolution provided. Going to use default 1 usec resolution\n"); intr_delay_resolution = ENA_DEFAULT_INTR_DELAY_RESOLUTION; } @@ -1450,11 +1449,13 @@ void ena_com_wait_for_abort_completion(struct ena_com_dev *ena_dev) { struct ena_com_admin_queue *admin_queue = &ena_dev->admin_queue; unsigned long flags = 0; + u32 exp = 0; spin_lock_irqsave(&admin_queue->q_lock, flags); while (atomic_read(&admin_queue->outstanding_cmds) != 0) { spin_unlock_irqrestore(&admin_queue->q_lock, flags); - msleep(ENA_POLL_MS); + ena_delay_exponential_backoff_us(exp++, + ena_dev->ena_min_poll_delay_us); spin_lock_irqsave(&admin_queue->q_lock, flags); } spin_unlock_irqrestore(&admin_queue->q_lock, flags); @@ -1802,6 +1803,7 @@ int ena_com_admin_init(struct ena_com_dev *ena_dev, if (ret) goto error; + admin_queue->ena_dev = ena_dev; admin_queue->running_state = true; return 0; @@ -2009,7 +2011,7 @@ void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data) struct ena_admin_aenq_entry *aenq_e; struct ena_admin_aenq_common_desc *aenq_common; struct ena_com_aenq *aenq = &dev->aenq; - unsigned long long timestamp; + u64 timestamp; ena_aenq_handler handler_cb; u16 masked_head, processed = 0; u8 phase; @@ -2027,9 +2029,8 @@ void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data) */ dma_rmb(); - timestamp = - (unsigned long long)aenq_common->timestamp_low | - ((unsigned long long)aenq_common->timestamp_high << 32); + timestamp = (u64)aenq_common->timestamp_low | + ((u64)aenq_common->timestamp_high << 32); pr_debug("AENQ! Group[%x] Syndrom[%x] timestamp: [%llus]\n", aenq_common->group, aenq_common->syndrom, timestamp); @@ -2059,8 +2060,7 @@ void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data) /* write the aenq doorbell after all AENQ descriptors were read */ mb(); - writel_relaxed((u32)aenq->head, - dev->reg_bar + ENA_REGS_AENQ_HEAD_DB_OFF); + writel_relaxed((u32)aenq->head, dev->reg_bar + ENA_REGS_AENQ_HEAD_DB_OFF); } int ena_com_dev_reset(struct ena_com_dev *ena_dev, @@ -2282,12 +2282,14 @@ int ena_com_fill_hash_function(struct ena_com_dev *ena_dev, enum ena_admin_hash_functions func, const u8 *key, u16 key_len, u32 init_val) { - struct ena_rss *rss = &ena_dev->rss; + struct ena_admin_feature_rss_flow_hash_control *hash_key; struct ena_admin_get_feat_resp get_resp; - struct ena_admin_feature_rss_flow_hash_control *hash_key = - rss->hash_key; + enum ena_admin_hash_functions old_func; + struct ena_rss *rss = &ena_dev->rss; int rc; + hash_key = rss->hash_key; + /* Make sure size is a mult of DWs */ if (unlikely(key_len & 0x3)) return -EINVAL; @@ -2299,7 +2301,7 @@ int ena_com_fill_hash_function(struct ena_com_dev *ena_dev, if (unlikely(rc)) return rc; - if (!((1 << func) & get_resp.u.flow_hash_func.supported_func)) { + if (!(BIT(func) & get_resp.u.flow_hash_func.supported_func)) { pr_err("Flow hash function %d isn't supported\n", func); return -EOPNOTSUPP; } @@ -2325,26 +2327,27 @@ int ena_com_fill_hash_function(struct ena_com_dev *ena_dev, return -EINVAL; } + old_func = rss->hash_func; rss->hash_func = func; rc = ena_com_set_hash_function(ena_dev); /* Restore the old function */ if (unlikely(rc)) - ena_com_get_hash_function(ena_dev, NULL, NULL); + rss->hash_func = old_func; return rc; } int ena_com_get_hash_function(struct ena_com_dev *ena_dev, - enum ena_admin_hash_functions *func, - u8 *key) + enum ena_admin_hash_functions *func) { struct ena_rss *rss = &ena_dev->rss; struct ena_admin_get_feat_resp get_resp; - struct ena_admin_feature_rss_flow_hash_control *hash_key = - rss->hash_key; int rc; + if (unlikely(!func)) + return -EINVAL; + rc = ena_com_get_feature_ex(ena_dev, &get_resp, ENA_ADMIN_RSS_HASH_FUNCTION, rss->hash_key_dma_addr, @@ -2357,8 +2360,15 @@ int ena_com_get_hash_function(struct ena_com_dev *ena_dev, if (rss->hash_func) rss->hash_func--; - if (func) - *func = rss->hash_func; + *func = rss->hash_func; + + return 0; +} + +int ena_com_get_hash_key(struct ena_com_dev *ena_dev, u8 *key) +{ + struct ena_admin_feature_rss_flow_hash_control *hash_key = + ena_dev->rss.hash_key; if (key) memcpy(key, hash_key->key, (size_t)(hash_key->keys_num) << 2); @@ -2641,10 +2651,10 @@ int ena_com_rss_init(struct ena_com_dev *ena_dev, u16 indr_tbl_log_size) * ignore this error and have indirection table support only. */ rc = ena_com_hash_key_allocate(ena_dev); - if (unlikely(rc) && rc != -EOPNOTSUPP) - goto err_hash_key; - else if (rc != -EOPNOTSUPP) + if (likely(!rc)) ena_com_hash_key_fill_default_key(ena_dev); + else if (rc != -EOPNOTSUPP) + goto err_hash_key; rc = ena_com_hash_ctrl_init(ena_dev); if (unlikely(rc)) diff --git a/drivers/net/ethernet/amazon/ena/ena_com.h b/drivers/net/ethernet/amazon/ena/ena_com.h index 469f298199a7..bc187adf54e4 100644 --- a/drivers/net/ethernet/amazon/ena/ena_com.h +++ b/drivers/net/ethernet/amazon/ena/ena_com.h @@ -54,9 +54,9 @@ #undef pr_fmt #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt -#define ENA_MAX_NUM_IO_QUEUES 128U +#define ENA_MAX_NUM_IO_QUEUES 128U /* We need to queues for each IO (on for Tx and one for Rx) */ -#define ENA_TOTAL_NUM_QUEUES (2 * (ENA_MAX_NUM_IO_QUEUES)) +#define ENA_TOTAL_NUM_QUEUES (2 * (ENA_MAX_NUM_IO_QUEUES)) #define ENA_MAX_HANDLERS 256 @@ -73,13 +73,15 @@ /*****************************************************************************/ /* ENA adaptive interrupt moderation settings */ -#define ENA_INTR_INITIAL_TX_INTERVAL_USECS 64 -#define ENA_INTR_INITIAL_RX_INTERVAL_USECS 0 -#define ENA_DEFAULT_INTR_DELAY_RESOLUTION 1 +#define ENA_INTR_INITIAL_TX_INTERVAL_USECS 64 +#define ENA_INTR_INITIAL_RX_INTERVAL_USECS 0 +#define ENA_DEFAULT_INTR_DELAY_RESOLUTION 1 -#define ENA_HW_HINTS_NO_TIMEOUT 0xFFFF +#define ENA_HASH_KEY_SIZE 40 -#define ENA_FEATURE_MAX_QUEUE_EXT_VER 1 +#define ENA_HW_HINTS_NO_TIMEOUT 0xFFFF + +#define ENA_FEATURE_MAX_QUEUE_EXT_VER 1 struct ena_llq_configurations { enum ena_admin_llq_header_location llq_header_location; @@ -237,6 +239,7 @@ struct ena_com_stats_admin { struct ena_com_admin_queue { void *q_dmadev; + struct ena_com_dev *ena_dev; spinlock_t q_lock; /* spinlock for the admin queue */ struct ena_comp_ctx *comp_ctx; @@ -349,6 +352,8 @@ struct ena_com_dev { struct ena_intr_moder_entry *intr_moder_tbl; struct ena_com_llq_info llq_info; + + u32 ena_min_poll_delay_us; }; struct ena_com_dev_get_features_ctx { @@ -393,7 +398,7 @@ struct ena_aenq_handlers { */ int ena_com_mmio_reg_read_request_init(struct ena_com_dev *ena_dev); -/* ena_com_set_mmio_read_mode - Enable/disable the mmio reg read mechanism +/* ena_com_set_mmio_read_mode - Enable/disable the indirect mmio reg read mechanism * @ena_dev: ENA communication layer struct * @readless_supported: readless mode (enable/disable) */ @@ -501,18 +506,6 @@ bool ena_com_get_admin_running_state(struct ena_com_dev *ena_dev); */ void ena_com_set_admin_polling_mode(struct ena_com_dev *ena_dev, bool polling); -/* ena_com_set_admin_polling_mode - Get the admin completion queue polling mode - * @ena_dev: ENA communication layer struct - * - * Get the admin completion mode. - * If polling mode is on, ena_com_execute_admin_command will perform a - * polling on the admin completion queue for the commands completion, - * otherwise it will wait on wait event. - * - * @return state - */ -bool ena_com_get_ena_admin_polling_mode(struct ena_com_dev *ena_dev); - /* ena_com_set_admin_auto_polling_mode - Enable autoswitch to polling mode * @ena_dev: ENA communication layer struct * @polling: Enable/Disable polling mode @@ -527,7 +520,7 @@ void ena_com_set_admin_auto_polling_mode(struct ena_com_dev *ena_dev, /* ena_com_admin_q_comp_intr_handler - admin queue interrupt handler * @ena_dev: ENA communication layer struct * - * This method go over the admin completion queue and wake up all the pending + * This method goes over the admin completion queue and wakes up all the pending * threads that wait on the commands wait event. * * @note: Should be called after MSI-X interrupt. @@ -537,7 +530,7 @@ void ena_com_admin_q_comp_intr_handler(struct ena_com_dev *ena_dev); /* ena_com_aenq_intr_handler - AENQ interrupt handler * @ena_dev: ENA communication layer struct * - * This method go over the async event notification queue and call the proper + * This method goes over the async event notification queue and calls the proper * aenq handler. */ void ena_com_aenq_intr_handler(struct ena_com_dev *dev, void *data); @@ -554,14 +547,14 @@ void ena_com_abort_admin_commands(struct ena_com_dev *ena_dev); /* ena_com_wait_for_abort_completion - Wait for admin commands abort. * @ena_dev: ENA communication layer struct * - * This method wait until all the outstanding admin commands will be completed. + * This method waits until all the outstanding admin commands are completed. */ void ena_com_wait_for_abort_completion(struct ena_com_dev *ena_dev); /* ena_com_validate_version - Validate the device parameters * @ena_dev: ENA communication layer struct * - * This method validate the device parameters are the same as the saved + * This method verifies the device parameters are the same as the saved * parameters in ena_dev. * This method is useful after device reset, to validate the device mac address * and the device offloads are the same as before the reset. @@ -695,23 +688,32 @@ int ena_com_fill_hash_function(struct ena_com_dev *ena_dev, */ int ena_com_set_hash_function(struct ena_com_dev *ena_dev); -/* ena_com_get_hash_function - Retrieve the hash function and the hash key - * from the device. +/* ena_com_get_hash_function - Retrieve the hash function from the device. * @ena_dev: ENA communication layer struct * @func: hash function - * @key: hash key * - * Retrieve the hash function and the hash key from the device. + * Retrieve the hash function from the device. * - * @note: If the caller called ena_com_fill_hash_function but didn't flash + * @note: If the caller called ena_com_fill_hash_function but didn't flush * it to the device, the new configuration will be lost. * * @return: 0 on Success and negative value otherwise. */ int ena_com_get_hash_function(struct ena_com_dev *ena_dev, - enum ena_admin_hash_functions *func, - u8 *key); + enum ena_admin_hash_functions *func); +/* ena_com_get_hash_key - Retrieve the hash key + * @ena_dev: ENA communication layer struct + * @key: hash key + * + * Retrieve the hash key. + * + * @note: If the caller called ena_com_fill_hash_key but didn't flush + * it to the device, the new configuration will be lost. + * + * @return: 0 on Success and negative value otherwise. + */ +int ena_com_get_hash_key(struct ena_com_dev *ena_dev, u8 *key); /* ena_com_fill_hash_ctrl - Fill RSS hash control * @ena_dev: ENA communication layer struct. * @proto: The protocol to configure. @@ -746,7 +748,7 @@ int ena_com_set_hash_ctrl(struct ena_com_dev *ena_dev); * * Retrieve the hash control from the device. * - * @note, If the caller called ena_com_fill_hash_ctrl but didn't flash + * @note: If the caller called ena_com_fill_hash_ctrl but didn't flush * it to the device, the new configuration will be lost. * * @return: 0 on Success and negative value otherwise. @@ -798,7 +800,7 @@ int ena_com_indirect_table_set(struct ena_com_dev *ena_dev); * * Retrieve the RSS indirection table from the device. * - * @note: If the caller called ena_com_indirect_table_fill_entry but didn't flash + * @note: If the caller called ena_com_indirect_table_fill_entry but didn't flush * it to the device, the new configuration will be lost. * * @return: 0 on Success and negative value otherwise. @@ -824,14 +826,14 @@ int ena_com_allocate_debug_area(struct ena_com_dev *ena_dev, /* ena_com_delete_debug_area - Free the debug area resources. * @ena_dev: ENA communication layer struct * - * Free the allocate debug area. + * Free the allocated debug area. */ void ena_com_delete_debug_area(struct ena_com_dev *ena_dev); /* ena_com_delete_host_info - Free the host info resources. * @ena_dev: ENA communication layer struct * - * Free the allocate host info. + * Free the allocated host info. */ void ena_com_delete_host_info(struct ena_com_dev *ena_dev); @@ -872,9 +874,9 @@ int ena_com_destroy_io_cq(struct ena_com_dev *ena_dev, * @cmd_completion: command completion return value. * @cmd_comp_size: command completion size. - * Submit an admin command and then wait until the device will return a + * Submit an admin command and then wait until the device returns a * completion. - * The completion will be copyed into cmd_comp. + * The completion will be copied into cmd_comp. * * @return - 0 on success, negative value on failure. */ @@ -937,7 +939,7 @@ unsigned int ena_com_get_nonadaptive_moderation_interval_rx(struct ena_com_dev * /* ena_com_config_dev_mode - Configure the placement policy of the device. * @ena_dev: ENA communication layer struct * @llq_features: LLQ feature descriptor, retrieve via - * ena_com_get_dev_attr_feat. + * ena_com_get_dev_attr_feat. * @ena_llq_config: The default driver LLQ parameters configurations */ int ena_com_config_dev_mode(struct ena_com_dev *ena_dev, @@ -963,7 +965,7 @@ static inline void ena_com_disable_adaptive_moderation(struct ena_com_dev *ena_d * @intr_reg: interrupt register to update. * @rx_delay_interval: Rx interval in usecs * @tx_delay_interval: Tx interval in usecs - * @unmask: unask enable/disable + * @unmask: unmask enable/disable * * Prepare interrupt update register with the supplied parameters. */ diff --git a/drivers/net/ethernet/amazon/ena/ena_common_defs.h b/drivers/net/ethernet/amazon/ena/ena_common_defs.h index 23beb7e7ed7b..8a8ded0de9ac 100644 --- a/drivers/net/ethernet/amazon/ena/ena_common_defs.h +++ b/drivers/net/ethernet/amazon/ena/ena_common_defs.h @@ -45,4 +45,4 @@ struct ena_common_mem_addr { u16 reserved16; }; -#endif /*_ENA_COMMON_H_ */ +#endif /* _ENA_COMMON_H_ */ diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_com.c b/drivers/net/ethernet/amazon/ena/ena_eth_com.c index 2845ac277724..ec8ea25e988d 100644 --- a/drivers/net/ethernet/amazon/ena/ena_eth_com.c +++ b/drivers/net/ethernet/amazon/ena/ena_eth_com.c @@ -519,7 +519,7 @@ int ena_com_rx_pkt(struct ena_com_io_cq *io_cq, struct ena_eth_io_rx_cdesc_base *cdesc = NULL; u16 cdesc_idx = 0; u16 nb_hw_desc; - u16 i; + u16 i = 0; WARN(io_cq->direction != ENA_COM_IO_QUEUE_DIRECTION_RX, "wrong Q type"); @@ -538,13 +538,19 @@ int ena_com_rx_pkt(struct ena_com_io_cq *io_cq, return -ENOSPC; } - for (i = 0; i < nb_hw_desc; i++) { + cdesc = ena_com_rx_cdesc_idx_to_ptr(io_cq, cdesc_idx); + ena_rx_ctx->pkt_offset = cdesc->offset; + + do { + ena_buf[i].len = cdesc->length; + ena_buf[i].req_id = cdesc->req_id; + + if (++i >= nb_hw_desc) + break; + cdesc = ena_com_rx_cdesc_idx_to_ptr(io_cq, cdesc_idx + i); - ena_buf->len = cdesc->length; - ena_buf->req_id = cdesc->req_id; - ena_buf++; - } + } while (1); /* Update SQ head ptr */ io_sq->next_to_comp += nb_hw_desc; @@ -578,10 +584,10 @@ int ena_com_add_single_rx_desc(struct ena_com_io_sq *io_sq, desc->length = ena_buf->len; - desc->ctrl = ENA_ETH_IO_RX_DESC_FIRST_MASK; - desc->ctrl |= ENA_ETH_IO_RX_DESC_LAST_MASK; - desc->ctrl |= io_sq->phase & ENA_ETH_IO_RX_DESC_PHASE_MASK; - desc->ctrl |= ENA_ETH_IO_RX_DESC_COMP_REQ_MASK; + desc->ctrl = ENA_ETH_IO_RX_DESC_FIRST_MASK | + ENA_ETH_IO_RX_DESC_LAST_MASK | + (io_sq->phase & ENA_ETH_IO_RX_DESC_PHASE_MASK) | + ENA_ETH_IO_RX_DESC_COMP_REQ_MASK; desc->req_id = req_id; diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_com.h b/drivers/net/ethernet/amazon/ena/ena_eth_com.h index 77986c0ea52c..8b1afd3b32f2 100644 --- a/drivers/net/ethernet/amazon/ena/ena_eth_com.h +++ b/drivers/net/ethernet/amazon/ena/ena_eth_com.h @@ -73,6 +73,7 @@ struct ena_com_rx_ctx { u32 hash; u16 descs; int max_bufs; + u8 pkt_offset; }; int ena_com_prepare_tx(struct ena_com_io_sq *io_sq, @@ -95,7 +96,7 @@ static inline void ena_com_unmask_intr(struct ena_com_io_cq *io_cq, writel(intr_reg->intr_control, io_cq->unmask_reg); } -static inline int ena_com_free_desc(struct ena_com_io_sq *io_sq) +static inline int ena_com_free_q_entries(struct ena_com_io_sq *io_sq) { u16 tail, next_to_comp, cnt; @@ -113,7 +114,7 @@ static inline bool ena_com_sq_have_enough_space(struct ena_com_io_sq *io_sq, int temp; if (io_sq->mem_queue_type == ENA_ADMIN_PLACEMENT_POLICY_HOST) - return ena_com_free_desc(io_sq) >= required_buffers; + return ena_com_free_q_entries(io_sq) >= required_buffers; /* This calculation doesn't need to be 100% accurate. So to reduce * the calculation overhead just Subtract 2 lines from the free descs @@ -122,7 +123,7 @@ static inline bool ena_com_sq_have_enough_space(struct ena_com_io_sq *io_sq, */ temp = required_buffers / io_sq->llq_info.descs_per_entry + 2; - return ena_com_free_desc(io_sq) > temp; + return ena_com_free_q_entries(io_sq) > temp; } static inline bool ena_com_meta_desc_changed(struct ena_com_io_sq *io_sq, diff --git a/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h b/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h index 00e0f056a741..d105c9c56192 100644 --- a/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h +++ b/drivers/net/ethernet/amazon/ena/ena_eth_io_defs.h @@ -264,7 +264,9 @@ struct ena_eth_io_rx_cdesc_base { u16 sub_qid; - u16 reserved; + u8 offset; + + u8 reserved; }; /* 8-word format */ @@ -412,4 +414,4 @@ struct ena_eth_io_numa_node_cfg_reg { #define ENA_ETH_IO_NUMA_NODE_CFG_REG_ENABLED_SHIFT 31 #define ENA_ETH_IO_NUMA_NODE_CFG_REG_ENABLED_MASK BIT(31) -#endif /*_ENA_ETH_IO_H_ */ +#endif /* _ENA_ETH_IO_H_ */ diff --git a/drivers/net/ethernet/amazon/ena/ena_ethtool.c b/drivers/net/ethernet/amazon/ena/ena_ethtool.c index 9cc28b4b2627..e340b65af08c 100644 --- a/drivers/net/ethernet/amazon/ena/ena_ethtool.c +++ b/drivers/net/ethernet/amazon/ena/ena_ethtool.c @@ -83,6 +83,7 @@ static const struct ena_stats ena_stats_tx_strings[] = { ENA_STAT_TX_ENTRY(bad_req_id), ENA_STAT_TX_ENTRY(llq_buffer_copy), ENA_STAT_TX_ENTRY(missed_tx), + ENA_STAT_TX_ENTRY(unmask_interrupt), }; static const struct ena_stats ena_stats_rx_strings[] = { @@ -205,7 +206,7 @@ int ena_get_sset_count(struct net_device *netdev, int sset) if (sset != ETH_SS_STATS) return -EOPNOTSUPP; - return adapter->num_io_queues * (ENA_STATS_ARRAY_TX + ENA_STATS_ARRAY_RX) + return adapter->num_io_queues * (ENA_STATS_ARRAY_TX + ENA_STATS_ARRAY_RX) + ENA_STATS_ARRAY_GLOBAL + ENA_STATS_ARRAY_ENA_COM; } @@ -259,7 +260,6 @@ static void ena_get_strings(struct net_device *netdev, u32 sset, u8 *data) for (i = 0; i < ENA_STATS_ARRAY_GLOBAL; i++) { ena_stats = &ena_stats_global_strings[i]; - memcpy(data, ena_stats->name, ETH_GSTRING_LEN); data += ETH_GSTRING_LEN; } @@ -306,10 +306,8 @@ static int ena_get_coalesce(struct net_device *net_dev, struct ena_adapter *adapter = netdev_priv(net_dev); struct ena_com_dev *ena_dev = adapter->ena_dev; - if (!ena_com_interrupt_moderation_supported(ena_dev)) { - /* the devie doesn't support interrupt moderation */ + if (!ena_com_interrupt_moderation_supported(ena_dev)) return -EOPNOTSUPP; - } coalesce->tx_coalesce_usecs = ena_com_get_nonadaptive_moderation_interval_tx(ena_dev) * @@ -325,7 +323,7 @@ static int ena_get_coalesce(struct net_device *net_dev, return 0; } -static void ena_update_tx_rings_intr_moderation(struct ena_adapter *adapter) +static void ena_update_tx_rings_nonadaptive_intr_moderation(struct ena_adapter *adapter) { unsigned int val; int i; @@ -336,7 +334,7 @@ static void ena_update_tx_rings_intr_moderation(struct ena_adapter *adapter) adapter->tx_ring[i].smoothed_interval = val; } -static void ena_update_rx_rings_intr_moderation(struct ena_adapter *adapter) +static void ena_update_rx_rings_nonadaptive_intr_moderation(struct ena_adapter *adapter) { unsigned int val; int i; @@ -354,24 +352,22 @@ static int ena_set_coalesce(struct net_device *net_dev, struct ena_com_dev *ena_dev = adapter->ena_dev; int rc; - if (!ena_com_interrupt_moderation_supported(ena_dev)) { - /* the devie doesn't support interrupt moderation */ + if (!ena_com_interrupt_moderation_supported(ena_dev)) return -EOPNOTSUPP; - } rc = ena_com_update_nonadaptive_moderation_interval_tx(ena_dev, coalesce->tx_coalesce_usecs); if (rc) return rc; - ena_update_tx_rings_intr_moderation(adapter); + ena_update_tx_rings_nonadaptive_intr_moderation(adapter); rc = ena_com_update_nonadaptive_moderation_interval_rx(ena_dev, coalesce->rx_coalesce_usecs); if (rc) return rc; - ena_update_rx_rings_intr_moderation(adapter); + ena_update_rx_rings_nonadaptive_intr_moderation(adapter); if (coalesce->use_adaptive_rx_coalesce && !ena_com_get_adaptive_moderation_enabled(ena_dev)) @@ -635,6 +631,32 @@ static u32 ena_get_rxfh_key_size(struct net_device *netdev) return ENA_HASH_KEY_SIZE; } +static int ena_indirection_table_set(struct ena_adapter *adapter, + const u32 *indir) +{ + struct ena_com_dev *ena_dev = adapter->ena_dev; + int i, rc; + + for (i = 0; i < ENA_RX_RSS_TABLE_SIZE; i++) { + rc = ena_com_indirect_table_fill_entry(ena_dev, + i, + ENA_IO_RXQ_IDX(indir[i])); + if (unlikely(rc)) { + netif_err(adapter, drv, adapter->netdev, + "Cannot fill indirect table (index is too large)\n"); + return rc; + } + } + + rc = ena_com_indirect_table_set(ena_dev); + if (rc) { + netif_err(adapter, drv, adapter->netdev, + "Cannot set indirect table\n"); + return rc == -EPERM ? -EOPNOTSUPP : rc; + } + return rc; +} + static int ena_indirection_table_get(struct ena_adapter *adapter, u32 *indir) { struct ena_com_dev *ena_dev = adapter->ena_dev; @@ -672,17 +694,18 @@ static int ena_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, /* We call this function in order to check if the device * supports getting/setting the hash function. */ - rc = ena_com_get_hash_function(adapter->ena_dev, &ena_func, key); + rc = ena_com_get_hash_function(adapter->ena_dev, &ena_func); if (rc) { - if (rc == -EOPNOTSUPP) { - key = NULL; - hfunc = NULL; + if (rc == -EOPNOTSUPP) rc = 0; - } return rc; } + rc = ena_com_get_hash_key(adapter->ena_dev, key); + if (rc) + return rc; + switch (ena_func) { case ENA_ADMIN_TOEPLITZ: func = ETH_RSS_HASH_TOP; @@ -699,7 +722,7 @@ static int ena_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, if (hfunc) *hfunc = func; - return rc; + return 0; } static int ena_set_rxfh(struct net_device *netdev, const u32 *indir, @@ -707,27 +730,13 @@ static int ena_set_rxfh(struct net_device *netdev, const u32 *indir, { struct ena_adapter *adapter = netdev_priv(netdev); struct ena_com_dev *ena_dev = adapter->ena_dev; - enum ena_admin_hash_functions func; - int rc, i; + enum ena_admin_hash_functions func = 0; + int rc; if (indir) { - for (i = 0; i < ENA_RX_RSS_TABLE_SIZE; i++) { - rc = ena_com_indirect_table_fill_entry(ena_dev, - i, - ENA_IO_RXQ_IDX(indir[i])); - if (unlikely(rc)) { - netif_err(adapter, drv, netdev, - "Cannot fill indirect table (index is too large)\n"); - return rc; - } - } - - rc = ena_com_indirect_table_set(ena_dev); - if (rc) { - netif_err(adapter, drv, netdev, - "Cannot set indirect table\n"); - return rc == -EPERM ? -EOPNOTSUPP : rc; - } + rc = ena_indirection_table_set(adapter, indir); + if (rc) + return rc; } switch (hfunc) { @@ -746,7 +755,7 @@ static int ena_set_rxfh(struct net_device *netdev, const u32 *indir, return -EOPNOTSUPP; } - if (key) { + if (key || func) { rc = ena_com_fill_hash_function(ena_dev, func, key, ENA_HASH_KEY_SIZE, 0xFFFFFFFF); diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c index 2cc765df8da3..a0af74c93971 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.c +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c @@ -263,7 +263,7 @@ static int ena_xdp_tx_map_buff(struct ena_ring *xdp_ring, dma_addr_t dma = 0; u32 size; - tx_info->xdpf = convert_to_xdp_frame(xdp); + tx_info->xdpf = xdp_convert_buff_to_frame(xdp); size = tx_info->xdpf->len; ena_buf = tx_info->bufs; @@ -1435,6 +1435,8 @@ static struct sk_buff *ena_rx_skb(struct ena_ring *rx_ring, skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_info->page, rx_info->page_offset, len, ENA_PAGE_SIZE); + /* The offset is non zero only for the first buffer */ + rx_info->page_offset = 0; netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev, "rx skb updated. len %d. data_len %d\n", @@ -1590,6 +1592,7 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, { u16 next_to_clean = rx_ring->next_to_clean; struct ena_com_rx_ctx ena_rx_ctx; + struct ena_rx_buffer *rx_info; struct ena_adapter *adapter; u32 res_budget, work_done; int rx_copybreak_pkt = 0; @@ -1606,6 +1609,7 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, "%s qid %d\n", __func__, rx_ring->qid); res_budget = budget; xdp.rxq = &rx_ring->xdp_rxq; + xdp.frame_sz = ENA_PAGE_SIZE; do { xdp_verdict = XDP_PASS; @@ -1613,6 +1617,7 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, ena_rx_ctx.ena_bufs = rx_ring->ena_bufs; ena_rx_ctx.max_bufs = rx_ring->sgl_size; ena_rx_ctx.descs = 0; + ena_rx_ctx.pkt_offset = 0; rc = ena_com_rx_pkt(rx_ring->ena_com_io_cq, rx_ring->ena_com_io_sq, &ena_rx_ctx); @@ -1622,6 +1627,9 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, if (unlikely(ena_rx_ctx.descs == 0)) break; + rx_info = &rx_ring->rx_buffer_info[rx_ring->ena_bufs[0].req_id]; + rx_info->page_offset = ena_rx_ctx.pkt_offset; + netif_dbg(rx_ring->adapter, rx_status, rx_ring->netdev, "rx_poll: q %d got packet from ena. descs #: %d l3 proto %d l4 proto %d hash: %x\n", rx_ring->qid, ena_rx_ctx.descs, ena_rx_ctx.l3_proto, @@ -1683,7 +1691,7 @@ static int ena_clean_rx_irq(struct ena_ring *rx_ring, struct napi_struct *napi, rx_ring->next_to_clean = next_to_clean; - refill_required = ena_com_free_desc(rx_ring->ena_com_io_sq); + refill_required = ena_com_free_q_entries(rx_ring->ena_com_io_sq); refill_threshold = min_t(int, rx_ring->ring_size / ENA_RX_REFILL_THRESH_DIVIDER, ENA_RX_REFILL_THRESH_PACKET); @@ -1762,6 +1770,9 @@ static void ena_unmask_interrupt(struct ena_ring *tx_ring, tx_ring->smoothed_interval, true); + u64_stats_update_begin(&tx_ring->syncp); + tx_ring->tx_stats.unmask_interrupt++; + u64_stats_update_end(&tx_ring->syncp); /* It is a shared MSI-X. * Tx and Rx CQ have pointer to it. * So we use one of them to reach the intr reg @@ -2231,7 +2242,7 @@ static int ena_rss_configure(struct ena_adapter *adapter) rc = ena_rss_init_default(adapter); if (rc && (rc != -EOPNOTSUPP)) { netif_err(adapter, ifup, adapter->netdev, - "Failed to init RSS rc: %d\n", rc); + "Failed to init RSS rc: %d\n", rc); return rc; } } @@ -2304,7 +2315,7 @@ static int ena_create_io_tx_queue(struct ena_adapter *adapter, int qid) if (rc) { netif_err(adapter, ifup, adapter->netdev, "Failed to create I/O TX queue num %d rc: %d\n", - qid, rc); + qid, rc); return rc; } @@ -2453,7 +2464,7 @@ static int create_queues_with_size_backoff(struct ena_adapter *adapter) * ones due to past queue allocation failures. */ set_io_rings_size(adapter, adapter->requested_tx_ring_size, - adapter->requested_rx_ring_size); + adapter->requested_rx_ring_size); while (1) { if (ena_xdp_present(adapter)) { @@ -2494,7 +2505,7 @@ err_setup_tx: if (rc != -ENOMEM) { netif_err(adapter, ifup, adapter->netdev, "Queue creation failed with error code %d\n", - rc); + rc); return rc; } @@ -2517,7 +2528,7 @@ err_setup_tx: new_rx_ring_size = cur_rx_ring_size / 2; if (new_tx_ring_size < ENA_MIN_RING_SIZE || - new_rx_ring_size < ENA_MIN_RING_SIZE) { + new_rx_ring_size < ENA_MIN_RING_SIZE) { netif_err(adapter, ifup, adapter->netdev, "Queue creation failed with the smallest possible queue size of %d for both queues. Not retrying with smaller queues\n", ENA_MIN_RING_SIZE); @@ -3076,8 +3087,7 @@ static u16 ena_select_queue(struct net_device *dev, struct sk_buff *skb, return qid; } -static void ena_config_host_info(struct ena_com_dev *ena_dev, - struct pci_dev *pdev) +static void ena_config_host_info(struct ena_com_dev *ena_dev, struct pci_dev *pdev) { struct ena_admin_host_info *host_info; int rc; @@ -3107,6 +3117,7 @@ static void ena_config_host_info(struct ena_com_dev *ena_dev, host_info->num_cpus = num_online_cpus(); host_info->driver_supported_features = + ENA_ADMIN_HOST_INFO_RX_OFFSET_MASK | ENA_ADMIN_HOST_INFO_INTERRUPT_MODERATION_MASK; rc = ena_com_set_host_attributes(ena_dev); @@ -3169,6 +3180,7 @@ static void ena_get_stats64(struct net_device *netdev, struct ena_ring *rx_ring, *tx_ring; unsigned int start; u64 rx_drops; + u64 tx_drops; int i; if (!test_bit(ENA_FLAG_DEV_UP, &adapter->flags)) @@ -3203,9 +3215,11 @@ static void ena_get_stats64(struct net_device *netdev, do { start = u64_stats_fetch_begin_irq(&adapter->syncp); rx_drops = adapter->dev_stats.rx_drops; + tx_drops = adapter->dev_stats.tx_drops; } while (u64_stats_fetch_retry_irq(&adapter->syncp, start)); stats->rx_dropped = rx_drops; + stats->tx_dropped = tx_drops; stats->multicast = 0; stats->collisions = 0; @@ -3433,6 +3447,7 @@ static void ena_destroy_device(struct ena_adapter *adapter, bool graceful) ena_com_mmio_reg_read_request_destroy(ena_dev); + /* return reset reason to default value */ adapter->reset_reason = ENA_REGS_RESET_NORMAL; clear_bit(ENA_FLAG_TRIGGER_RESET, &adapter->flags); @@ -3678,8 +3693,7 @@ static void check_for_empty_rx_ring(struct ena_adapter *adapter) for (i = 0; i < adapter->num_io_queues; i++) { rx_ring = &adapter->rx_ring[i]; - refill_required = - ena_com_free_desc(rx_ring->ena_com_io_sq); + refill_required = ena_com_free_q_entries(rx_ring->ena_com_io_sq); if (unlikely(refill_required == (rx_ring->ring_size - 1))) { rx_ring->empty_rx_queue++; @@ -3817,11 +3831,11 @@ static void ena_timer_service(struct timer_list *t) mod_timer(&adapter->timer_service, round_jiffies(jiffies + HZ)); } -static int ena_calc_max_io_queue_num(struct pci_dev *pdev, +static u32 ena_calc_max_io_queue_num(struct pci_dev *pdev, struct ena_com_dev *ena_dev, struct ena_com_dev_get_features_ctx *get_feat_ctx) { - int io_tx_sq_num, io_tx_cq_num, io_rx_num, max_num_io_queues; + u32 io_tx_sq_num, io_tx_cq_num, io_rx_num, max_num_io_queues; if (ena_dev->supported_features & BIT(ENA_ADMIN_MAX_QUEUES_EXT)) { struct ena_admin_queue_ext_feature_fields *max_queue_ext = @@ -3991,7 +4005,7 @@ static int ena_rss_init_default(struct ena_adapter *adapter) } } - rc = ena_com_fill_hash_function(ena_dev, ENA_ADMIN_CRC32, NULL, + rc = ena_com_fill_hash_function(ena_dev, ENA_ADMIN_TOEPLITZ, NULL, ENA_HASH_KEY_SIZE, 0xFFFFFFFF); if (unlikely(rc && (rc != -EOPNOTSUPP))) { dev_err(dev, "Cannot fill hash function\n"); @@ -4107,8 +4121,8 @@ static int ena_calc_io_queue_size(struct ena_calc_queue_size_ctx *ctx) */ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) { - struct ena_com_dev_get_features_ctx get_feat_ctx; struct ena_calc_queue_size_ctx calc_queue_ctx = { 0 }; + struct ena_com_dev_get_features_ctx get_feat_ctx; struct ena_llq_configurations llq_config; struct ena_com_dev *ena_dev = NULL; struct ena_adapter *adapter; @@ -4152,6 +4166,8 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) goto err_free_region; } + ena_dev->ena_min_poll_delay_us = ENA_ADMIN_POLL_DELAY_US; + ena_dev->dmadev = &pdev->dev; rc = ena_device_init(ena_dev, pdev, &get_feat_ctx, &wd_state); @@ -4175,7 +4191,7 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) calc_queue_ctx.get_feat_ctx = &get_feat_ctx; calc_queue_ctx.pdev = pdev; - /* Initial Tx and RX interrupt delay. Assumes 1 usec granularity. + /* Initial TX and RX interrupt delay. Assumes 1 usec granularity. * Updated during device initialization with the real granularity */ ena_dev->intr_moder_tx_interval = ENA_INTR_INITIAL_TX_INTERVAL_USECS; @@ -4219,12 +4235,11 @@ static int ena_probe(struct pci_dev *pdev, const struct pci_device_id *ent) adapter->num_io_queues = max_num_io_queues; adapter->max_num_io_queues = max_num_io_queues; + adapter->last_monitored_tx_qid = 0; adapter->xdp_first_ring = 0; adapter->xdp_num_queues = 0; - adapter->last_monitored_tx_qid = 0; - adapter->rx_copybreak = ENA_DEFAULT_RX_COPYBREAK; adapter->wd_state = wd_state; @@ -4356,6 +4371,7 @@ static void __ena_shutoff(struct pci_dev *pdev, bool shutdown) cancel_work_sync(&adapter->reset_task); rtnl_lock(); /* lock released inside the below if-else block */ + adapter->reset_reason = ENA_REGS_RESET_SHUTDOWN; ena_destroy_device(adapter, true); if (shutdown) { netif_device_detach(netdev); @@ -4514,14 +4530,17 @@ static void ena_keep_alive_wd(void *adapter_data, struct ena_adapter *adapter = (struct ena_adapter *)adapter_data; struct ena_admin_aenq_keep_alive_desc *desc; u64 rx_drops; + u64 tx_drops; desc = (struct ena_admin_aenq_keep_alive_desc *)aenq_e; adapter->last_keep_alive_jiffies = jiffies; rx_drops = ((u64)desc->rx_drops_high << 32) | desc->rx_drops_low; + tx_drops = ((u64)desc->tx_drops_high << 32) | desc->tx_drops_low; u64_stats_update_begin(&adapter->syncp); adapter->dev_stats.rx_drops = rx_drops; + adapter->dev_stats.tx_drops = tx_drops; u64_stats_update_end(&adapter->syncp); } diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.h b/drivers/net/ethernet/amazon/ena/ena_netdev.h index 9e1860d81908..ba030d260940 100644 --- a/drivers/net/ethernet/amazon/ena/ena_netdev.h +++ b/drivers/net/ethernet/amazon/ena/ena_netdev.h @@ -50,12 +50,6 @@ #define DRV_MODULE_GEN_SUBMINOR 0 #define DRV_MODULE_NAME "ena" -#ifndef DRV_MODULE_GENERATION -#define DRV_MODULE_GENERATION \ - __stringify(DRV_MODULE_GEN_MAJOR) "." \ - __stringify(DRV_MODULE_GEN_MINOR) "." \ - __stringify(DRV_MODULE_GEN_SUBMINOR) "K" -#endif #define DEVICE_NAME "Elastic Network Adapter (ENA)" @@ -104,8 +98,6 @@ #define ENA_RX_RSS_TABLE_LOG_SIZE 7 #define ENA_RX_RSS_TABLE_SIZE (1 << ENA_RX_RSS_TABLE_LOG_SIZE) -#define ENA_HASH_KEY_SIZE 40 - /* The number of tx packet completions that will be handled each NAPI poll * cycle is ring_size / ENA_TX_POLL_BUDGET_DIVIDER. */ @@ -137,6 +129,8 @@ #define ENA_IO_IRQ_FIRST_IDX 1 #define ENA_IO_IRQ_IDX(q) (ENA_IO_IRQ_FIRST_IDX + (q)) +#define ENA_ADMIN_POLL_DELAY_US 100 + /* ENA device should send keep alive msg every 1 sec. * We wait for 6 sec just to be on the safe side. */ @@ -151,8 +145,9 @@ * The buffer size we share with the device is defined to be ENA_PAGE_SIZE */ -#define ENA_XDP_MAX_MTU (ENA_PAGE_SIZE - ETH_HLEN - ETH_FCS_LEN - \ - VLAN_HLEN - XDP_PACKET_HEADROOM) +#define ENA_XDP_MAX_MTU (ENA_PAGE_SIZE - ETH_HLEN - ETH_FCS_LEN - \ + VLAN_HLEN - XDP_PACKET_HEADROOM - \ + SKB_DATA_ALIGN(sizeof(struct skb_shared_info))) #define ENA_IS_XDP_INDEX(adapter, index) (((index) >= (adapter)->xdp_first_ring) && \ ((index) < (adapter)->xdp_first_ring + (adapter)->xdp_num_queues)) @@ -248,6 +243,7 @@ struct ena_stats_tx { u64 bad_req_id; u64 llq_buffer_copy; u64 missed_tx; + u64 unmask_interrupt; }; struct ena_stats_rx { @@ -333,6 +329,7 @@ struct ena_stats_dev { u64 interface_down; u64 admin_q_pause; u64 rx_drops; + u64 tx_drops; }; enum ena_flags_t { diff --git a/drivers/net/ethernet/amazon/ena/ena_regs_defs.h b/drivers/net/ethernet/amazon/ena/ena_regs_defs.h index 04fcafcc059c..b514bb1b855d 100644 --- a/drivers/net/ethernet/amazon/ena/ena_regs_defs.h +++ b/drivers/net/ethernet/amazon/ena/ena_regs_defs.h @@ -154,4 +154,4 @@ enum ena_regs_reset_reason_types { #define ENA_REGS_RSS_IND_ENTRY_UPDATE_CQ_IDX_SHIFT 16 #define ENA_REGS_RSS_IND_ENTRY_UPDATE_CQ_IDX_MASK 0xffff0000 -#endif /*_ENA_REGS_H_ */ +#endif /* _ENA_REGS_H_ */ diff --git a/drivers/net/ethernet/amd/7990.c b/drivers/net/ethernet/amd/7990.c index cf3562e82ca9..50fb66369415 100644 --- a/drivers/net/ethernet/amd/7990.c +++ b/drivers/net/ethernet/amd/7990.c @@ -536,7 +536,7 @@ void lance_tx_timeout(struct net_device *dev, unsigned int txqueue) } EXPORT_SYMBOL_GPL(lance_tx_timeout); -int lance_start_xmit(struct sk_buff *skb, struct net_device *dev) +netdev_tx_t lance_start_xmit(struct sk_buff *skb, struct net_device *dev) { struct lance_private *lp = netdev_priv(dev); volatile struct lance_init_block *ib = lp->init_block; diff --git a/drivers/net/ethernet/amd/7990.h b/drivers/net/ethernet/amd/7990.h index 8266b3c1fefc..e53551daeea1 100644 --- a/drivers/net/ethernet/amd/7990.h +++ b/drivers/net/ethernet/amd/7990.h @@ -241,7 +241,7 @@ struct lance_private { /* Now the prototypes we export */ int lance_open(struct net_device *dev); int lance_close(struct net_device *dev); -int lance_start_xmit(struct sk_buff *skb, struct net_device *dev); +netdev_tx_t lance_start_xmit(struct sk_buff *skb, struct net_device *dev); void lance_set_multicast(struct net_device *dev); void lance_tx_timeout(struct net_device *dev, unsigned int txqueue); #ifdef CONFIG_NET_POLL_CONTROLLER diff --git a/drivers/net/ethernet/amd/atarilance.c b/drivers/net/ethernet/amd/atarilance.c index 4e36122609a3..961796abab35 100644 --- a/drivers/net/ethernet/amd/atarilance.c +++ b/drivers/net/ethernet/amd/atarilance.c @@ -156,7 +156,7 @@ struct lance_memory { struct lance_init_block init; struct lance_tx_head tx_head[TX_RING_SIZE]; struct lance_rx_head rx_head[RX_RING_SIZE]; - char packet_area[0]; /* packet data follow after the + char packet_area[]; /* packet data follow after the * init block and the ring * descriptors and are located * at runtime */ diff --git a/drivers/net/ethernet/apple/bmac.c b/drivers/net/ethernet/apple/bmac.c index a58185b1d8bf..3e3711b60d01 100644 --- a/drivers/net/ethernet/apple/bmac.c +++ b/drivers/net/ethernet/apple/bmac.c @@ -1182,7 +1182,7 @@ bmac_get_station_address(struct net_device *dev, unsigned char *ea) int i; unsigned short data; - for (i = 0; i < 6; i++) + for (i = 0; i < 3; i++) { reset_and_select_srom(dev); data = read_srom(dev, i + EnetAddressOffset/2, SROMAddressBits); diff --git a/drivers/net/ethernet/aquantia/atlantic/Makefile b/drivers/net/ethernet/aquantia/atlantic/Makefile index 8b555665a33a..130a105d03f3 100644 --- a/drivers/net/ethernet/aquantia/atlantic/Makefile +++ b/drivers/net/ethernet/aquantia/atlantic/Makefile @@ -25,6 +25,10 @@ atlantic-objs := aq_main.o \ hw_atl/hw_atl_utils.o \ hw_atl/hw_atl_utils_fw2x.o \ hw_atl/hw_atl_llh.o \ + hw_atl2/hw_atl2.o \ + hw_atl2/hw_atl2_utils.o \ + hw_atl2/hw_atl2_utils_fw.o \ + hw_atl2/hw_atl2_llh.o \ macsec/macsec_api.o atlantic-$(CONFIG_MACSEC) += aq_macsec.o diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_cfg.h b/drivers/net/ethernet/aquantia/atlantic/aq_cfg.h index 7560f5506e55..52b9833fda99 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_cfg.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_cfg.h @@ -80,8 +80,8 @@ #define AQ_CFG_LOCK_TRYS 100U -#define AQ_CFG_DRV_AUTHOR "aQuantia" -#define AQ_CFG_DRV_DESC "aQuantia Corporation(R) Network Driver" +#define AQ_CFG_DRV_AUTHOR "Marvell" +#define AQ_CFG_DRV_DESC "Marvell (Aquantia) Corporation(R) Network Driver" #define AQ_CFG_DRV_NAME "atlantic" #endif /* AQ_CFG_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_common.h b/drivers/net/ethernet/aquantia/atlantic/aq_common.h index c8c402b013bb..52ad9433cabc 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_common.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_common.h @@ -1,7 +1,8 @@ /* SPDX-License-Identifier: GPL-2.0-only */ -/* - * aQuantia Corporation Network Driver - * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved +/* Atlantic Network Driver + * + * Copyright (C) 2014-2019 aQuantia Corporation + * Copyright (C) 2019-2020 Marvell International Ltd. */ /* File aq_common.h: Basic includes for all files in project. */ @@ -37,22 +38,31 @@ #define AQ_DEVICE_ID_AQC111S 0x91B1 #define AQ_DEVICE_ID_AQC112S 0x92B1 -#define HW_ATL_NIC_NAME "aQuantia AQtion 10Gbit Network Adapter" +#define AQ_DEVICE_ID_AQC113DEV 0x00C0 +#define AQ_DEVICE_ID_AQC113CS 0x94C0 +#define AQ_DEVICE_ID_AQC114CS 0x93C0 +#define AQ_DEVICE_ID_AQC113 0x04C0 +#define AQ_DEVICE_ID_AQC113C 0x14C0 +#define AQ_DEVICE_ID_AQC115C 0x12C0 + +#define HW_ATL_NIC_NAME "Marvell (aQuantia) AQtion 10Gbit Network Adapter" #define AQ_HWREV_ANY 0 #define AQ_HWREV_1 1 #define AQ_HWREV_2 2 -#define AQ_NIC_RATE_10G BIT(0) -#define AQ_NIC_RATE_5G BIT(1) -#define AQ_NIC_RATE_5GSR BIT(2) -#define AQ_NIC_RATE_2GS BIT(3) -#define AQ_NIC_RATE_1G BIT(4) -#define AQ_NIC_RATE_100M BIT(5) - -#define AQ_NIC_RATE_EEE_10G BIT(6) -#define AQ_NIC_RATE_EEE_5G BIT(7) -#define AQ_NIC_RATE_EEE_2GS BIT(8) -#define AQ_NIC_RATE_EEE_1G BIT(9) +#define AQ_NIC_RATE_10G BIT(0) +#define AQ_NIC_RATE_5G BIT(1) +#define AQ_NIC_RATE_5GSR BIT(2) +#define AQ_NIC_RATE_2G5 BIT(3) +#define AQ_NIC_RATE_1G BIT(4) +#define AQ_NIC_RATE_100M BIT(5) +#define AQ_NIC_RATE_10M BIT(6) + +#define AQ_NIC_RATE_EEE_10G BIT(7) +#define AQ_NIC_RATE_EEE_5G BIT(8) +#define AQ_NIC_RATE_EEE_2G5 BIT(9) +#define AQ_NIC_RATE_EEE_1G BIT(10) +#define AQ_NIC_RATE_EEE_100M BIT(11) #endif /* AQ_COMMON_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c index 7241cf92b43a..743d3b13b39d 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c @@ -88,13 +88,13 @@ static const char aq_ethtool_stat_names[][ETH_GSTRING_LEN] = { "InDroppedDma", }; -static const char aq_ethtool_queue_stat_names[][ETH_GSTRING_LEN] = { - "Queue[%d] InPackets", - "Queue[%d] OutPackets", - "Queue[%d] Restarts", - "Queue[%d] InJumboPackets", - "Queue[%d] InLroPackets", - "Queue[%d] InErrors", +static const char * const aq_ethtool_queue_stat_names[] = { + "%sQueue[%d] InPackets", + "%sQueue[%d] OutPackets", + "%sQueue[%d] Restarts", + "%sQueue[%d] InJumboPackets", + "%sQueue[%d] InLroPackets", + "%sQueue[%d] InErrors", }; #if IS_ENABLED(CONFIG_MACSEC) @@ -166,7 +166,8 @@ static u32 aq_ethtool_n_stats(struct net_device *ndev) struct aq_nic_s *nic = netdev_priv(ndev); struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(nic); u32 n_stats = ARRAY_SIZE(aq_ethtool_stat_names) + - ARRAY_SIZE(aq_ethtool_queue_stat_names) * cfg->vecs; + ARRAY_SIZE(aq_ethtool_queue_stat_names) * cfg->vecs * + cfg->tcs; #if IS_ENABLED(CONFIG_MACSEC) if (nic->macsec_cfg) { @@ -223,7 +224,7 @@ static void aq_ethtool_get_drvinfo(struct net_device *ndev, static void aq_ethtool_get_strings(struct net_device *ndev, u32 stringset, u8 *data) { - struct aq_nic_s *aq_nic = netdev_priv(ndev); + struct aq_nic_s *nic = netdev_priv(ndev); struct aq_nic_cfg_s *cfg; u8 *p = data; int i, si; @@ -231,24 +232,35 @@ static void aq_ethtool_get_strings(struct net_device *ndev, int sa; #endif - cfg = aq_nic_get_cfg(aq_nic); + cfg = aq_nic_get_cfg(nic); switch (stringset) { - case ETH_SS_STATS: + case ETH_SS_STATS: { + const int stat_cnt = ARRAY_SIZE(aq_ethtool_queue_stat_names); + char tc_string[8]; + int tc; + + memset(tc_string, 0, sizeof(tc_string)); memcpy(p, aq_ethtool_stat_names, sizeof(aq_ethtool_stat_names)); p = p + sizeof(aq_ethtool_stat_names); - for (i = 0; i < cfg->vecs; i++) { - for (si = 0; - si < ARRAY_SIZE(aq_ethtool_queue_stat_names); - si++) { - snprintf(p, ETH_GSTRING_LEN, - aq_ethtool_queue_stat_names[si], i); - p += ETH_GSTRING_LEN; + + for (tc = 0; tc < cfg->tcs; tc++) { + if (cfg->is_qos) + snprintf(tc_string, 8, "TC%d ", tc); + + for (i = 0; i < cfg->vecs; i++) { + for (si = 0; si < stat_cnt; si++) { + snprintf(p, ETH_GSTRING_LEN, + aq_ethtool_queue_stat_names[si], + tc_string, + AQ_NIC_CFG_TCVEC2RING(cfg, tc, i)); + p += ETH_GSTRING_LEN; + } } } #if IS_ENABLED(CONFIG_MACSEC) - if (!aq_nic->macsec_cfg) + if (!nic->macsec_cfg) break; memcpy(p, aq_macsec_stat_names, sizeof(aq_macsec_stat_names)); @@ -256,7 +268,7 @@ static void aq_ethtool_get_strings(struct net_device *ndev, for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { struct aq_macsec_txsc *aq_txsc; - if (!(test_bit(i, &aq_nic->macsec_cfg->txsc_idx_busy))) + if (!(test_bit(i, &nic->macsec_cfg->txsc_idx_busy))) continue; for (si = 0; @@ -266,7 +278,7 @@ static void aq_ethtool_get_strings(struct net_device *ndev, aq_macsec_txsc_stat_names[si], i); p += ETH_GSTRING_LEN; } - aq_txsc = &aq_nic->macsec_cfg->aq_txsc[i]; + aq_txsc = &nic->macsec_cfg->aq_txsc[i]; for (sa = 0; sa < MACSEC_NUM_AN; sa++) { if (!(test_bit(sa, &aq_txsc->tx_sa_idx_busy))) continue; @@ -283,10 +295,10 @@ static void aq_ethtool_get_strings(struct net_device *ndev, for (i = 0; i < AQ_MACSEC_MAX_SC; i++) { struct aq_macsec_rxsc *aq_rxsc; - if (!(test_bit(i, &aq_nic->macsec_cfg->rxsc_idx_busy))) + if (!(test_bit(i, &nic->macsec_cfg->rxsc_idx_busy))) continue; - aq_rxsc = &aq_nic->macsec_cfg->aq_rxsc[i]; + aq_rxsc = &nic->macsec_cfg->aq_rxsc[i]; for (sa = 0; sa < MACSEC_NUM_AN; sa++) { if (!(test_bit(sa, &aq_rxsc->rx_sa_idx_busy))) continue; @@ -302,6 +314,7 @@ static void aq_ethtool_get_strings(struct net_device *ndev, } #endif break; + } case ETH_SS_PRIV_FLAGS: memcpy(p, aq_ethtool_priv_flag_names, sizeof(aq_ethtool_priv_flag_names)); @@ -605,12 +618,15 @@ static enum hw_atl_fw2x_rate eee_mask_to_ethtool_mask(u32 speed) if (speed & AQ_NIC_RATE_EEE_10G) rate |= SUPPORTED_10000baseT_Full; - if (speed & AQ_NIC_RATE_EEE_2GS) + if (speed & AQ_NIC_RATE_EEE_2G5) rate |= SUPPORTED_2500baseX_Full; if (speed & AQ_NIC_RATE_EEE_1G) rate |= SUPPORTED_1000baseT_Full; + if (speed & AQ_NIC_RATE_EEE_100M) + rate |= SUPPORTED_100baseT_Full; + return rate; } @@ -777,8 +793,6 @@ static int aq_set_ringparam(struct net_device *ndev, dev_close(ndev); } - aq_nic_free_vectors(aq_nic); - cfg->rxds = max(ring->rx_pending, hw_caps->rxds_min); cfg->rxds = min(cfg->rxds, hw_caps->rxds_max); cfg->rxds = ALIGN(cfg->rxds, AQ_HW_RXD_MULTIPLE); @@ -787,15 +801,10 @@ static int aq_set_ringparam(struct net_device *ndev, cfg->txds = min(cfg->txds, hw_caps->txds_max); cfg->txds = ALIGN(cfg->txds, AQ_HW_TXD_MULTIPLE); - for (aq_nic->aq_vecs = 0; aq_nic->aq_vecs < cfg->vecs; - aq_nic->aq_vecs++) { - aq_nic->aq_vec[aq_nic->aq_vecs] = - aq_vec_alloc(aq_nic, aq_nic->aq_vecs, cfg); - if (unlikely(!aq_nic->aq_vec[aq_nic->aq_vecs])) { - err = -ENOMEM; - goto err_exit; - } - } + err = aq_nic_realloc_vectors(aq_nic); + if (err) + goto err_exit; + if (ndev_running) err = dev_open(ndev, NULL); diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_filters.c b/drivers/net/ethernet/aquantia/atlantic/aq_filters.c index 03ff92bc4a7f..1bc4d33a0ce5 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_filters.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_filters.c @@ -153,6 +153,8 @@ aq_check_approve_fvlan(struct aq_nic_s *aq_nic, struct aq_hw_rx_fltrs_s *rx_fltrs, struct ethtool_rx_flow_spec *fsp) { + struct aq_nic_cfg_s *cfg = &aq_nic->aq_nic_cfg; + if (fsp->location < AQ_RX_FIRST_LOC_FVLANID || fsp->location > AQ_RX_LAST_LOC_FVLANID) { netdev_err(aq_nic->ndev, @@ -170,10 +172,10 @@ aq_check_approve_fvlan(struct aq_nic_s *aq_nic, return -EINVAL; } - if (fsp->ring_cookie > aq_nic->aq_nic_cfg.num_rss_queues) { + if (fsp->ring_cookie > cfg->num_rss_queues * cfg->tcs) { netdev_err(aq_nic->ndev, "ethtool: queue number must be in range [0, %d]", - aq_nic->aq_nic_cfg.num_rss_queues - 1); + cfg->num_rss_queues * cfg->tcs - 1); return -EINVAL; } return 0; @@ -262,6 +264,7 @@ static bool __must_check aq_rule_is_not_correct(struct aq_nic_s *aq_nic, struct ethtool_rx_flow_spec *fsp) { + struct aq_nic_cfg_s *cfg = &aq_nic->aq_nic_cfg; bool rule_is_not_correct = false; if (!aq_nic) { @@ -274,11 +277,11 @@ aq_rule_is_not_correct(struct aq_nic_s *aq_nic, } else if (aq_check_filter(aq_nic, fsp)) { rule_is_not_correct = true; } else if (fsp->ring_cookie != RX_CLS_FLOW_DISC) { - if (fsp->ring_cookie >= aq_nic->aq_nic_cfg.num_rss_queues) { + if (fsp->ring_cookie >= cfg->num_rss_queues * cfg->tcs) { netdev_err(aq_nic->ndev, "ethtool: The specified action is invalid.\n" "Maximum allowable value action is %u.\n", - aq_nic->aq_nic_cfg.num_rss_queues - 1); + cfg->num_rss_queues * cfg->tcs - 1); rule_is_not_correct = true; } } diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h index 7d71bc7dc500..ed5b465bc664 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h @@ -18,6 +18,12 @@ #define AQ_HW_MAC_COUNTER_HZ 312500000ll #define AQ_HW_PHY_COUNTER_HZ 160000000ll +enum aq_tc_mode { + AQ_TC_MODE_INVALID = -1, + AQ_TC_MODE_8TCS, + AQ_TC_MODE_4TCS, +}; + #define AQ_RX_FIRST_LOC_FVLANID 0U #define AQ_RX_LAST_LOC_FVLANID 15U #define AQ_RX_FIRST_LOC_FETHERT 16U @@ -29,6 +35,9 @@ (AQ_RX_LAST_LOC_FVLANID - AQ_RX_FIRST_LOC_FVLANID + 1U) #define AQ_RX_QUEUE_NOT_ASSIGNED 0xFFU +/* Used for rate to Mbps conversion */ +#define AQ_MBPS_DIVISOR 125000 /* 1000000 / 8 */ + /* NIC H/W capabilities */ struct aq_hw_caps_s { u64 hw_features; @@ -46,7 +55,7 @@ struct aq_hw_caps_s { u32 mac_regs_count; u32 hw_alive_check_addr; u8 msix_irqs; - u8 tcs; + u8 tcs_max; u8 rxd_alignment; u8 rxd_size; u8 txd_alignment; @@ -55,6 +64,7 @@ struct aq_hw_caps_s { u8 rx_rings; bool flow_control; bool is_64_dma; + u32 priv_data_len; }; struct aq_hw_link_status_s { @@ -117,8 +127,11 @@ struct aq_stats_s { #define AQ_HW_TXD_MULTIPLE 8U #define AQ_HW_RXD_MULTIPLE 8U +#define AQ_HW_QUEUES_MAX 32U #define AQ_HW_MULTICAST_ADDRESS_MAX 32U +#define AQ_HW_PTP_TC 2U + #define AQ_HW_LED_BLINK 0x2U #define AQ_HW_LED_DEFAULT 0x0U @@ -136,6 +149,19 @@ enum aq_priv_flags { BIT(AQ_HW_LOOPBACK_PHYINT_SYS) |\ BIT(AQ_HW_LOOPBACK_PHYEXT_SYS)) +#define ATL_HW_CHIP_MIPS 0x00000001U +#define ATL_HW_CHIP_TPO2 0x00000002U +#define ATL_HW_CHIP_RPF2 0x00000004U +#define ATL_HW_CHIP_MPI_AQ 0x00000010U +#define ATL_HW_CHIP_ATLANTIC 0x00800000U +#define ATL_HW_CHIP_REVISION_A0 0x01000000U +#define ATL_HW_CHIP_REVISION_B0 0x02000000U +#define ATL_HW_CHIP_REVISION_B1 0x04000000U +#define ATL_HW_CHIP_ANTIGUA 0x08000000U + +#define ATL_HW_IS_CHIP_FEATURE(_HW_, _F_) (!!(ATL_HW_CHIP_##_F_ & \ + (_HW_)->chip_features)) + struct aq_hw_s { atomic_t flags; u8 rbl_enabled:1; @@ -159,6 +185,7 @@ struct aq_hw_s { struct hw_atl_utils_fw_rpc rpc; s64 ptp_clk_offset; u16 phy_id; + void *priv; }; struct aq_ring_s; @@ -182,6 +209,11 @@ struct aq_hw_ops { int (*hw_set_mac_address)(struct aq_hw_s *self, u8 *mac_addr); + int (*hw_soft_reset)(struct aq_hw_s *self); + + int (*hw_prepare)(struct aq_hw_s *self, + const struct aq_fw_ops **fw_ops); + int (*hw_reset)(struct aq_hw_s *self); int (*hw_init)(struct aq_hw_s *self, u8 *mac_addr); @@ -248,21 +280,19 @@ struct aq_hw_ops { int (*hw_rss_hash_set)(struct aq_hw_s *self, struct aq_rss_parameters *rss_params); + int (*hw_tc_rate_limit_set)(struct aq_hw_s *self); + int (*hw_get_regs)(struct aq_hw_s *self, const struct aq_hw_caps_s *aq_hw_caps, u32 *regs_buff); struct aq_stats_s *(*hw_get_hw_stats)(struct aq_hw_s *self); - int (*hw_get_fw_version)(struct aq_hw_s *self, u32 *fw_version); + u32 (*hw_get_fw_version)(struct aq_hw_s *self); int (*hw_set_offload)(struct aq_hw_s *self, struct aq_nic_cfg_s *aq_nic_cfg); - int (*hw_tx_tc_mode_get)(struct aq_hw_s *self, u32 *tc_mode); - - int (*hw_rx_tc_mode_get)(struct aq_hw_s *self, u32 *tc_mode); - int (*hw_ring_hwts_rx_fill)(struct aq_hw_s *self, struct aq_ring_s *aq_ring); diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.c b/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.c index 7dbf49adcea6..342c5179f846 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.c @@ -79,3 +79,29 @@ int aq_hw_err_from_flags(struct aq_hw_s *hw) err_exit: return err; } + +int aq_hw_num_tcs(struct aq_hw_s *hw) +{ + switch (hw->aq_nic_cfg->tc_mode) { + case AQ_TC_MODE_8TCS: + return 8; + case AQ_TC_MODE_4TCS: + return 4; + default: + break; + } + + return 1; +} + +int aq_hw_q_per_tc(struct aq_hw_s *hw) +{ + switch (hw->aq_nic_cfg->tc_mode) { + case AQ_TC_MODE_8TCS: + return 4; + case AQ_TC_MODE_4TCS: + return 8; + default: + return 4; + } +} diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.h b/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.h index 9ef82d487e01..32aa5f2fb840 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw_utils.h @@ -34,5 +34,7 @@ u32 aq_hw_read_reg(struct aq_hw_s *hw, u32 reg); void aq_hw_write_reg(struct aq_hw_s *hw, u32 reg, u32 value); u64 aq_hw_read_reg64(struct aq_hw_s *hw, u32 reg); int aq_hw_err_from_flags(struct aq_hw_s *hw); +int aq_hw_num_tcs(struct aq_hw_s *hw); +int aq_hw_q_per_tc(struct aq_hw_s *hw); #endif /* AQ_HW_UTILS_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c b/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c index 0b3e234a54aa..4a6dfac857ca 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_macsec.c @@ -401,7 +401,7 @@ static u32 aq_sc_idx_max(const enum aq_macsec_sc_sa sc_sa) break; default: break; - }; + } return result; } @@ -417,7 +417,7 @@ static u32 aq_to_hw_sc_idx(const u32 sc_idx, const enum aq_macsec_sc_sa sc_sa) return sc_idx; default: WARN_ONCE(true, "Invalid sc_sa"); - }; + } return sc_idx; } @@ -478,7 +478,7 @@ static int aq_mdo_add_secy(struct macsec_context *ctx) set_bit(txsc_idx, &cfg->txsc_idx_busy); - return 0; + return ret; } static int aq_mdo_upd_secy(struct macsec_context *ctx) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_main.c b/drivers/net/ethernet/aquantia/atlantic/aq_main.c index 9fcab646cbd5..8a1da044e908 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_main.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_main.c @@ -12,11 +12,13 @@ #include "aq_ethtool.h" #include "aq_ptp.h" #include "aq_filters.h" +#include "aq_hw_utils.h" #include <linux/netdevice.h> #include <linux/module.h> #include <linux/ip.h> #include <linux/udp.h> +#include <net/pkt_cls.h> MODULE_LICENSE("GPL v2"); MODULE_AUTHOR(AQ_CFG_DRV_AUTHOR); @@ -38,7 +40,7 @@ struct net_device *aq_ndev_alloc(void) struct net_device *ndev = NULL; struct aq_nic_s *aq_nic = NULL; - ndev = alloc_etherdev_mq(sizeof(struct aq_nic_s), AQ_CFG_VECS_MAX); + ndev = alloc_etherdev_mq(sizeof(struct aq_nic_s), AQ_HW_QUEUES_MAX); if (!ndev) return NULL; @@ -330,6 +332,73 @@ static int aq_ndo_vlan_rx_kill_vid(struct net_device *ndev, __be16 proto, return 0; } +static int aq_validate_mqprio_opt(struct aq_nic_s *self, + struct tc_mqprio_qopt_offload *mqprio, + const unsigned int num_tc) +{ + const bool has_min_rate = !!(mqprio->flags & TC_MQPRIO_F_MIN_RATE); + struct aq_nic_cfg_s *aq_nic_cfg = aq_nic_get_cfg(self); + const unsigned int tcs_max = min_t(u8, aq_nic_cfg->aq_hw_caps->tcs_max, + AQ_CFG_TCS_MAX); + + if (num_tc > tcs_max) { + netdev_err(self->ndev, "Too many TCs requested\n"); + return -EOPNOTSUPP; + } + + if (num_tc != 0 && !is_power_of_2(num_tc)) { + netdev_err(self->ndev, "TC count should be power of 2\n"); + return -EOPNOTSUPP; + } + + if (has_min_rate && !ATL_HW_IS_CHIP_FEATURE(self->aq_hw, ANTIGUA)) { + netdev_err(self->ndev, "Min tx rate is not supported\n"); + return -EOPNOTSUPP; + } + + return 0; +} + +static int aq_ndo_setup_tc(struct net_device *dev, enum tc_setup_type type, + void *type_data) +{ + struct tc_mqprio_qopt_offload *mqprio = type_data; + struct aq_nic_s *aq_nic = netdev_priv(dev); + bool has_min_rate; + bool has_max_rate; + int err; + int i; + + if (type != TC_SETUP_QDISC_MQPRIO) + return -EOPNOTSUPP; + + has_min_rate = !!(mqprio->flags & TC_MQPRIO_F_MIN_RATE); + has_max_rate = !!(mqprio->flags & TC_MQPRIO_F_MAX_RATE); + + err = aq_validate_mqprio_opt(aq_nic, mqprio, mqprio->qopt.num_tc); + if (err) + return err; + + for (i = 0; i < mqprio->qopt.num_tc; i++) { + if (has_max_rate) { + u64 max_rate = mqprio->max_rate[i]; + + do_div(max_rate, AQ_MBPS_DIVISOR); + aq_nic_setup_tc_max_rate(aq_nic, i, (u32)max_rate); + } + + if (has_min_rate) { + u64 min_rate = mqprio->min_rate[i]; + + do_div(min_rate, AQ_MBPS_DIVISOR); + aq_nic_setup_tc_min_rate(aq_nic, i, (u32)min_rate); + } + } + + return aq_nic_setup_tc_mqprio(aq_nic, mqprio->qopt.num_tc, + mqprio->qopt.prio_tc_map); +} + static const struct net_device_ops aq_ndev_ops = { .ndo_open = aq_ndev_open, .ndo_stop = aq_ndev_close, @@ -341,6 +410,7 @@ static const struct net_device_ops aq_ndev_ops = { .ndo_do_ioctl = aq_ndev_ioctl, .ndo_vlan_rx_add_vid = aq_ndo_vlan_rx_add_vid, .ndo_vlan_rx_kill_vid = aq_ndo_vlan_rx_kill_vid, + .ndo_setup_tc = aq_ndo_setup_tc, }; static int __init aq_ndev_init_module(void) diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c index a369705a786a..4435c6374f7e 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c @@ -1,7 +1,8 @@ // SPDX-License-Identifier: GPL-2.0-only -/* - * aQuantia Corporation Network Driver - * Copyright (C) 2014-2019 aQuantia Corporation. All rights reserved +/* Atlantic Network Driver + * + * Copyright (C) 2014-2019 aQuantia Corporation + * Copyright (C) 2019-2020 Marvell International Ltd. */ /* File aq_nic.c: Definition of common code for NIC. */ @@ -25,6 +26,7 @@ #include <linux/ip.h> #include <linux/tcp.h> #include <net/ip.h> +#include <net/pkt_cls.h> static unsigned int aq_itr = AQ_CFG_INTERRUPT_MODERATION_AUTO; module_param_named(aq_itr, aq_itr, uint, 0644); @@ -63,10 +65,38 @@ static void aq_nic_rss_init(struct aq_nic_s *self, unsigned int num_rss_queues) rss_params->indirection_table[i] = i & (num_rss_queues - 1); } +/* Recalculate the number of vectors */ +static void aq_nic_cfg_update_num_vecs(struct aq_nic_s *self) +{ + struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg; + + cfg->vecs = min(cfg->aq_hw_caps->vecs, AQ_CFG_VECS_DEF); + cfg->vecs = min(cfg->vecs, num_online_cpus()); + if (self->irqvecs > AQ_HW_SERVICE_IRQS) + cfg->vecs = min(cfg->vecs, self->irqvecs - AQ_HW_SERVICE_IRQS); + /* cfg->vecs should be power of 2 for RSS */ + cfg->vecs = rounddown_pow_of_two(cfg->vecs); + + if (ATL_HW_IS_CHIP_FEATURE(self->aq_hw, ANTIGUA)) { + if (cfg->tcs > 2) + cfg->vecs = min(cfg->vecs, 4U); + } + + if (cfg->vecs <= 4) + cfg->tc_mode = AQ_TC_MODE_8TCS; + else + cfg->tc_mode = AQ_TC_MODE_4TCS; + + /*rss rings */ + cfg->num_rss_queues = min(cfg->vecs, AQ_CFG_NUM_RSS_QUEUES_DEF); + aq_nic_rss_init(self, cfg->num_rss_queues); +} + /* Checks hw_caps and 'corrects' aq_nic_cfg in runtime */ void aq_nic_cfg_start(struct aq_nic_s *self) { struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg; + int i; cfg->tcs = AQ_CFG_TCS_DEF; @@ -78,7 +108,6 @@ void aq_nic_cfg_start(struct aq_nic_s *self) cfg->rxpageorder = AQ_CFG_RX_PAGEORDER; cfg->is_rss = AQ_CFG_IS_RSS_DEF; - cfg->num_rss_queues = AQ_CFG_NUM_RSS_QUEUES_DEF; cfg->aq_rss.base_cpu_number = AQ_CFG_RSS_BASE_CPU_NUM_DEF; cfg->fc.req = AQ_CFG_FC_MODE; cfg->wol = AQ_CFG_WOL_MODES; @@ -88,29 +117,13 @@ void aq_nic_cfg_start(struct aq_nic_s *self) cfg->is_autoneg = AQ_CFG_IS_AUTONEG_DEF; cfg->is_lro = AQ_CFG_IS_LRO_DEF; + cfg->is_ptp = true; /*descriptors */ cfg->rxds = min(cfg->aq_hw_caps->rxds_max, AQ_CFG_RXDS_DEF); cfg->txds = min(cfg->aq_hw_caps->txds_max, AQ_CFG_TXDS_DEF); - /*rss rings */ - cfg->vecs = min(cfg->aq_hw_caps->vecs, AQ_CFG_VECS_DEF); - cfg->vecs = min(cfg->vecs, num_online_cpus()); - if (self->irqvecs > AQ_HW_SERVICE_IRQS) - cfg->vecs = min(cfg->vecs, self->irqvecs - AQ_HW_SERVICE_IRQS); - /* cfg->vecs should be power of 2 for RSS */ - if (cfg->vecs >= 8U) - cfg->vecs = 8U; - else if (cfg->vecs >= 4U) - cfg->vecs = 4U; - else if (cfg->vecs >= 2U) - cfg->vecs = 2U; - else - cfg->vecs = 1U; - - cfg->num_rss_queues = min(cfg->vecs, AQ_CFG_NUM_RSS_QUEUES_DEF); - - aq_nic_rss_init(self, cfg->num_rss_queues); + aq_nic_cfg_update_num_vecs(self); cfg->irq_type = aq_pci_func_get_irq_type(self); @@ -135,6 +148,9 @@ void aq_nic_cfg_start(struct aq_nic_s *self) cfg->is_vlan_rx_strip = !!(cfg->features & NETIF_F_HW_VLAN_CTAG_RX); cfg->is_vlan_tx_insert = !!(cfg->features & NETIF_F_HW_VLAN_CTAG_TX); cfg->is_vlan_force_promisc = true; + + for (i = 0; i < sizeof(cfg->prio_tc_map); i++) + cfg->prio_tc_map[i] = cfg->tcs * i / 8; } static int aq_nic_update_link_status(struct aq_nic_s *self) @@ -180,6 +196,9 @@ static int aq_nic_update_link_status(struct aq_nic_s *self) #if IS_ENABLED(CONFIG_MACSEC) aq_macsec_enable(self); #endif + if (self->aq_hw_ops->hw_tc_rate_limit_set) + self->aq_hw_ops->hw_tc_rate_limit_set(self->aq_hw); + netif_tx_wake_all_queues(self->ndev); } if (netif_carrier_ok(self->ndev) && !self->link_status.mbps) { @@ -257,6 +276,28 @@ static void aq_nic_polling_timer_cb(struct timer_list *t) AQ_CFG_POLLING_TIMER_INTERVAL); } +static int aq_nic_hw_prepare(struct aq_nic_s *self) +{ + int err = 0; + + err = self->aq_hw_ops->hw_soft_reset(self->aq_hw); + if (err) + goto exit; + + err = self->aq_hw_ops->hw_prepare(self->aq_hw, &self->aq_fw_ops); + +exit: + return err; +} + +static bool aq_nic_is_valid_ether_addr(const u8 *addr) +{ + /* Some engineering samples of Aquantia NICs are provisioned with a + * partially populated MAC, which is still invalid. + */ + return !(addr[0] == 0 && addr[1] == 0 && addr[2] == 0); +} + int aq_nic_ndev_register(struct aq_nic_s *self) { int err = 0; @@ -266,7 +307,7 @@ int aq_nic_ndev_register(struct aq_nic_s *self) goto err_exit; } - err = hw_atl_utils_initfw(self->aq_hw, &self->aq_fw_ops); + err = aq_nic_hw_prepare(self); if (err) goto err_exit; @@ -281,6 +322,12 @@ int aq_nic_ndev_register(struct aq_nic_s *self) if (err) goto err_exit; + if (!is_valid_ether_addr(self->ndev->dev_addr) || + !aq_nic_is_valid_ether_addr(self->ndev->dev_addr)) { + netdev_warn(self->ndev, "MAC is invalid, will use random."); + eth_hw_addr_random(self->ndev); + } + #if defined(AQ_CFG_MAC_ADDR_PERMANENT) { static u8 mac_addr_permanent[] = AQ_CFG_MAC_ADDR_PERMANENT; @@ -364,26 +411,35 @@ int aq_nic_init(struct aq_nic_s *self) if (err < 0) goto err_exit; - if (self->aq_nic_cfg.aq_hw_caps->media_type == AQ_HW_MEDIA_TYPE_TP) { + if (ATL_HW_IS_CHIP_FEATURE(self->aq_hw, ATLANTIC) && + self->aq_nic_cfg.aq_hw_caps->media_type == AQ_HW_MEDIA_TYPE_TP) { self->aq_hw->phy_id = HW_ATL_PHY_ID_MAX; err = aq_phy_init(self->aq_hw); } - for (i = 0U, aq_vec = self->aq_vec[0]; - self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i]) + for (i = 0U; i < self->aq_vecs; i++) { + aq_vec = self->aq_vec[i]; + err = aq_vec_ring_alloc(aq_vec, self, i, + aq_nic_get_cfg(self)); + if (err) + goto err_exit; + aq_vec_init(aq_vec, self->aq_hw_ops, self->aq_hw); + } - err = aq_ptp_init(self, self->irqvecs - 1); - if (err < 0) - goto err_exit; + if (aq_nic_get_cfg(self)->is_ptp) { + err = aq_ptp_init(self, self->irqvecs - 1); + if (err < 0) + goto err_exit; - err = aq_ptp_ring_alloc(self); - if (err < 0) - goto err_exit; + err = aq_ptp_ring_alloc(self); + if (err < 0) + goto err_exit; - err = aq_ptp_ring_init(self); - if (err < 0) - goto err_exit; + err = aq_ptp_ring_init(self); + if (err < 0) + goto err_exit; + } netif_carrier_off(self->ndev); @@ -394,9 +450,12 @@ err_exit: int aq_nic_start(struct aq_nic_s *self) { struct aq_vec_s *aq_vec = NULL; + struct aq_nic_cfg_s *cfg; unsigned int i = 0U; int err = 0; + cfg = aq_nic_get_cfg(self); + err = self->aq_hw_ops->hw_multicast_list_set(self->aq_hw, self->mc_list.ar, self->mc_list.count); @@ -434,7 +493,7 @@ int aq_nic_start(struct aq_nic_s *self) timer_setup(&self->service_timer, aq_nic_service_timer_cb, 0); aq_nic_service_timer_cb(&self->service_timer); - if (self->aq_nic_cfg.is_polling) { + if (cfg->is_polling) { timer_setup(&self->polling_timer, aq_nic_polling_timer_cb, 0); mod_timer(&self->polling_timer, jiffies + AQ_CFG_POLLING_TIMER_INTERVAL); @@ -452,16 +511,16 @@ int aq_nic_start(struct aq_nic_s *self) if (err < 0) goto err_exit; - if (self->aq_nic_cfg.link_irq_vec) { + if (cfg->link_irq_vec) { int irqvec = pci_irq_vector(self->pdev, - self->aq_nic_cfg.link_irq_vec); + cfg->link_irq_vec); err = request_threaded_irq(irqvec, NULL, aq_linkstate_threaded_isr, IRQF_SHARED | IRQF_ONESHOT, self->ndev->name, self); if (err < 0) goto err_exit; - self->msix_entry_mask |= (1 << self->aq_nic_cfg.link_irq_vec); + self->msix_entry_mask |= (1 << cfg->link_irq_vec); } err = self->aq_hw_ops->hw_irq_enable(self->aq_hw, @@ -470,14 +529,21 @@ int aq_nic_start(struct aq_nic_s *self) goto err_exit; } - err = netif_set_real_num_tx_queues(self->ndev, self->aq_vecs); + err = netif_set_real_num_tx_queues(self->ndev, + self->aq_vecs * cfg->tcs); if (err < 0) goto err_exit; - err = netif_set_real_num_rx_queues(self->ndev, self->aq_vecs); + err = netif_set_real_num_rx_queues(self->ndev, + self->aq_vecs * cfg->tcs); if (err < 0) goto err_exit; + for (i = 0; i < cfg->tcs; i++) { + u16 offset = self->aq_vecs * i; + + netdev_set_tc_queue(self->ndev, i, self->aq_vecs, offset); + } netif_tx_start_all_queues(self->ndev); err_exit: @@ -488,6 +554,8 @@ unsigned int aq_nic_map_skb(struct aq_nic_s *self, struct sk_buff *skb, struct aq_ring_s *ring) { unsigned int nr_frags = skb_shinfo(skb)->nr_frags; + struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(self); + struct device *dev = aq_nic_get_dev(self); struct aq_ring_buff_s *first = NULL; u8 ipver = ip_hdr(skb)->version; struct aq_ring_buff_s *dx_buff; @@ -529,7 +597,7 @@ unsigned int aq_nic_map_skb(struct aq_nic_s *self, struct sk_buff *skb, need_context_tag = true; } - if (self->aq_nic_cfg.is_vlan_tx_insert && skb_vlan_tag_present(skb)) { + if (cfg->is_vlan_tx_insert && skb_vlan_tag_present(skb)) { dx_buff->vlan_tx_tag = skb_vlan_tag_get(skb); dx_buff->len_pkt = skb->len; dx_buff->is_vlan = 1U; @@ -544,12 +612,12 @@ unsigned int aq_nic_map_skb(struct aq_nic_s *self, struct sk_buff *skb, } dx_buff->len = skb_headlen(skb); - dx_buff->pa = dma_map_single(aq_nic_get_dev(self), + dx_buff->pa = dma_map_single(dev, skb->data, dx_buff->len, DMA_TO_DEVICE); - if (unlikely(dma_mapping_error(aq_nic_get_dev(self), dx_buff->pa))) { + if (unlikely(dma_mapping_error(dev, dx_buff->pa))) { ret = 0; goto exit; } @@ -581,13 +649,13 @@ unsigned int aq_nic_map_skb(struct aq_nic_s *self, struct sk_buff *skb, else buff_size = frag_len; - frag_pa = skb_frag_dma_map(aq_nic_get_dev(self), + frag_pa = skb_frag_dma_map(dev, frag, buff_offset, buff_size, DMA_TO_DEVICE); - if (unlikely(dma_mapping_error(aq_nic_get_dev(self), + if (unlikely(dma_mapping_error(dev, frag_pa))) goto mapping_error; @@ -621,12 +689,12 @@ mapping_error: if (!(dx_buff->is_gso_tcp || dx_buff->is_gso_udp) && !dx_buff->is_vlan && dx_buff->pa) { if (unlikely(dx_buff->is_sop)) { - dma_unmap_single(aq_nic_get_dev(self), + dma_unmap_single(dev, dx_buff->pa, dx_buff->len, DMA_TO_DEVICE); } else { - dma_unmap_page(aq_nic_get_dev(self), + dma_unmap_page(dev, dx_buff->pa, dx_buff->len, DMA_TO_DEVICE); @@ -640,15 +708,16 @@ exit: int aq_nic_xmit(struct aq_nic_s *self, struct sk_buff *skb) { - unsigned int vec = skb->queue_mapping % self->aq_nic_cfg.vecs; + struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(self); + unsigned int vec = skb->queue_mapping % cfg->vecs; + unsigned int tc = skb->queue_mapping / cfg->vecs; struct aq_ring_s *ring = NULL; unsigned int frags = 0U; int err = NETDEV_TX_OK; - unsigned int tc = 0U; frags = skb_shinfo(skb)->nr_frags + 1; - ring = self->aq_ring_tx[AQ_NIC_TCVEC2RING(self, tc, vec)]; + ring = self->aq_ring_tx[AQ_NIC_CFG_TCVEC2RING(cfg, tc, vec)]; if (frags > AQ_CFG_SKB_FRAGS_MAX) { dev_kfree_skb_any(skb); @@ -657,13 +726,14 @@ int aq_nic_xmit(struct aq_nic_s *self, struct sk_buff *skb) aq_ring_update_queue_state(ring); - if (self->aq_nic_cfg.priv_flags & BIT(AQ_HW_LOOPBACK_DMA_NET)) { + if (cfg->priv_flags & BIT(AQ_HW_LOOPBACK_DMA_NET)) { err = NETDEV_TX_BUSY; goto err_exit; } /* Above status update may stop the queue. Check this. */ - if (__netif_subqueue_stopped(self->ndev, ring->idx)) { + if (__netif_subqueue_stopped(self->ndev, + AQ_NIC_RING2QMAP(self, ring->idx))) { err = NETDEV_TX_BUSY; goto err_exit; } @@ -764,6 +834,9 @@ int aq_nic_get_regs(struct aq_nic_s *self, struct ethtool_regs *regs, void *p) u32 *regs_buff = p; int err = 0; + if (unlikely(!self->aq_hw_ops->hw_get_regs)) + return -EOPNOTSUPP; + regs->version = 1; err = self->aq_hw_ops->hw_get_regs(self->aq_hw, @@ -778,6 +851,9 @@ err_exit: int aq_nic_get_regs_count(struct aq_nic_s *self) { + if (unlikely(!self->aq_hw_ops->hw_get_regs)) + return 0; + return self->aq_nic_cfg.aq_hw_caps->mac_regs_count; } @@ -787,6 +863,7 @@ u64 *aq_nic_get_stats(struct aq_nic_s *self, u64 *data) struct aq_stats_s *stats; unsigned int count = 0U; unsigned int i = 0U; + unsigned int tc; if (self->aq_fw_ops->update_stats) { mutex_lock(&self->fwreq_mutex); @@ -825,10 +902,13 @@ u64 *aq_nic_get_stats(struct aq_nic_s *self, u64 *data) data += i; - for (i = 0U, aq_vec = self->aq_vec[0]; - aq_vec && self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i]) { - data += count; - aq_vec_get_sw_stats(aq_vec, data, &count); + for (tc = 0U; tc < self->aq_nic_cfg.tcs; tc++) { + for (i = 0U, aq_vec = self->aq_vec[0]; + aq_vec && self->aq_vecs > i; + ++i, aq_vec = self->aq_vec[i]) { + data += count; + aq_vec_get_sw_stats(aq_vec, tc, data, &count); + } } data += count; @@ -873,7 +953,7 @@ void aq_nic_get_link_ksettings(struct aq_nic_s *self, ethtool_link_ksettings_add_link_mode(cmd, supported, 5000baseT_Full); - if (self->aq_nic_cfg.aq_hw_caps->link_speed_msk & AQ_NIC_RATE_2GS) + if (self->aq_nic_cfg.aq_hw_caps->link_speed_msk & AQ_NIC_RATE_2G5) ethtool_link_ksettings_add_link_mode(cmd, supported, 2500baseT_Full); @@ -885,6 +965,10 @@ void aq_nic_get_link_ksettings(struct aq_nic_s *self, ethtool_link_ksettings_add_link_mode(cmd, supported, 100baseT_Full); + if (self->aq_nic_cfg.aq_hw_caps->link_speed_msk & AQ_NIC_RATE_10M) + ethtool_link_ksettings_add_link_mode(cmd, supported, + 10baseT_Full); + if (self->aq_nic_cfg.aq_hw_caps->flow_control) { ethtool_link_ksettings_add_link_mode(cmd, supported, Pause); @@ -912,7 +996,7 @@ void aq_nic_get_link_ksettings(struct aq_nic_s *self, ethtool_link_ksettings_add_link_mode(cmd, advertising, 5000baseT_Full); - if (self->aq_nic_cfg.link_speed_msk & AQ_NIC_RATE_2GS) + if (self->aq_nic_cfg.link_speed_msk & AQ_NIC_RATE_2G5) ethtool_link_ksettings_add_link_mode(cmd, advertising, 2500baseT_Full); @@ -924,6 +1008,10 @@ void aq_nic_get_link_ksettings(struct aq_nic_s *self, ethtool_link_ksettings_add_link_mode(cmd, advertising, 100baseT_Full); + if (self->aq_nic_cfg.link_speed_msk & AQ_NIC_RATE_10M) + ethtool_link_ksettings_add_link_mode(cmd, advertising, + 10baseT_Full); + if (self->aq_nic_cfg.fc.cur & AQ_NIC_FC_RX) ethtool_link_ksettings_add_link_mode(cmd, advertising, Pause); @@ -954,6 +1042,10 @@ int aq_nic_set_link_ksettings(struct aq_nic_s *self, speed = cmd->base.speed; switch (speed) { + case SPEED_10: + rate = AQ_NIC_RATE_10M; + break; + case SPEED_100: rate = AQ_NIC_RATE_100M; break; @@ -963,7 +1055,7 @@ int aq_nic_set_link_ksettings(struct aq_nic_s *self, break; case SPEED_2500: - rate = AQ_NIC_RATE_2GS; + rate = AQ_NIC_RATE_2G5; break; case SPEED_5000: @@ -1006,11 +1098,7 @@ struct aq_nic_cfg_s *aq_nic_get_cfg(struct aq_nic_s *self) u32 aq_nic_get_fw_version(struct aq_nic_s *self) { - u32 fw_version = 0U; - - self->aq_hw_ops->hw_get_fw_version(self->aq_hw, &fw_version); - - return fw_version; + return self->aq_hw_ops->hw_get_fw_version(self->aq_hw); } int aq_nic_set_loopback(struct aq_nic_s *self) @@ -1101,9 +1189,11 @@ void aq_nic_deinit(struct aq_nic_s *self, bool link_down) if (!self) goto err_exit; - for (i = 0U, aq_vec = self->aq_vec[0]; - self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i]) + for (i = 0U; i < self->aq_vecs; i++) { + aq_vec = self->aq_vec[i]; aq_vec_deinit(aq_vec); + aq_vec_ring_free(aq_vec); + } aq_ptp_unregister(self); aq_ptp_ring_deinit(self); @@ -1136,6 +1226,22 @@ void aq_nic_free_vectors(struct aq_nic_s *self) err_exit:; } +int aq_nic_realloc_vectors(struct aq_nic_s *self) +{ + struct aq_nic_cfg_s *cfg = aq_nic_get_cfg(self); + + aq_nic_free_vectors(self); + + for (self->aq_vecs = 0; self->aq_vecs < cfg->vecs; self->aq_vecs++) { + self->aq_vec[self->aq_vecs] = aq_vec_alloc(self, self->aq_vecs, + cfg); + if (unlikely(!self->aq_vec[self->aq_vecs])) + return -ENOMEM; + } + + return 0; +} + void aq_nic_shutdown(struct aq_nic_s *self) { int err = 0; @@ -1201,3 +1307,98 @@ void aq_nic_release_filter(struct aq_nic_s *self, enum aq_rx_filter_type type, break; } } + +int aq_nic_setup_tc_mqprio(struct aq_nic_s *self, u32 tcs, u8 *prio_tc_map) +{ + struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg; + const unsigned int prev_vecs = cfg->vecs; + bool ndev_running; + int err = 0; + int i; + + /* if already the same configuration or + * disable request (tcs is 0) and we already is disabled + */ + if (tcs == cfg->tcs || (tcs == 0 && !cfg->is_qos)) + return 0; + + ndev_running = netif_running(self->ndev); + if (ndev_running) + dev_close(self->ndev); + + cfg->tcs = tcs; + if (cfg->tcs == 0) + cfg->tcs = 1; + if (prio_tc_map) + memcpy(cfg->prio_tc_map, prio_tc_map, sizeof(cfg->prio_tc_map)); + else + for (i = 0; i < sizeof(cfg->prio_tc_map); i++) + cfg->prio_tc_map[i] = cfg->tcs * i / 8; + + cfg->is_qos = (tcs != 0 ? true : false); + cfg->is_ptp = (cfg->tcs <= AQ_HW_PTP_TC); + if (!cfg->is_ptp) + netdev_warn(self->ndev, "%s\n", + "PTP is auto disabled due to requested TC count."); + + netdev_set_num_tc(self->ndev, cfg->tcs); + + /* Changing the number of TCs might change the number of vectors */ + aq_nic_cfg_update_num_vecs(self); + if (prev_vecs != cfg->vecs) { + err = aq_nic_realloc_vectors(self); + if (err) + goto err_exit; + } + + if (ndev_running) + err = dev_open(self->ndev, NULL); + +err_exit: + return err; +} + +int aq_nic_setup_tc_max_rate(struct aq_nic_s *self, const unsigned int tc, + const u32 max_rate) +{ + struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg; + + if (tc >= AQ_CFG_TCS_MAX) + return -EINVAL; + + if (max_rate && max_rate < 10) { + netdev_warn(self->ndev, + "Setting %s to the minimum usable value of %dMbps.\n", + "max rate", 10); + cfg->tc_max_rate[tc] = 10; + } else { + cfg->tc_max_rate[tc] = max_rate; + } + + return 0; +} + +int aq_nic_setup_tc_min_rate(struct aq_nic_s *self, const unsigned int tc, + const u32 min_rate) +{ + struct aq_nic_cfg_s *cfg = &self->aq_nic_cfg; + + if (tc >= AQ_CFG_TCS_MAX) + return -EINVAL; + + if (min_rate) + set_bit(tc, &cfg->tc_min_rate_msk); + else + clear_bit(tc, &cfg->tc_min_rate_msk); + + if (min_rate && min_rate < 20) { + netdev_warn(self->ndev, + "Setting %s to the minimum usable value of %dMbps.\n", + "min rate", 20); + cfg->tc_min_rate[tc] = 20; + } else { + cfg->tc_min_rate[tc] = min_rate; + } + + return 0; +} diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h index 0663b8d0220d..2ab003065e62 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h @@ -59,8 +59,15 @@ struct aq_nic_cfg_s { bool is_polling; bool is_rss; bool is_lro; + bool is_qos; + bool is_ptp; + enum aq_tc_mode tc_mode; u32 priv_flags; u8 tcs; + u8 prio_tc_map[8]; + u32 tc_max_rate[AQ_CFG_TCS_MAX]; + unsigned long tc_min_rate_msk; + u32 tc_min_rate[AQ_CFG_TCS_MAX]; struct aq_rss_parameters aq_rss; u32 eee_speeds; }; @@ -77,8 +84,16 @@ struct aq_nic_cfg_s { #define AQ_NIC_WOL_MODES (WAKE_MAGIC |\ WAKE_PHY) -#define AQ_NIC_TCVEC2RING(_NIC_, _TC_, _VEC_) \ - ((_TC_) * AQ_CFG_TCS_MAX + (_VEC_)) +#define AQ_NIC_CFG_RING_PER_TC(_NIC_CFG_) \ + (((_NIC_CFG_)->tc_mode == AQ_TC_MODE_4TCS) ? 8 : 4) + +#define AQ_NIC_CFG_TCVEC2RING(_NIC_CFG_, _TC_, _VEC_) \ + ((_TC_) * AQ_NIC_CFG_RING_PER_TC(_NIC_CFG_) + (_VEC_)) + +#define AQ_NIC_RING2QMAP(_NIC_, _ID_) \ + ((_ID_) / AQ_NIC_CFG_RING_PER_TC(&(_NIC_)->aq_nic_cfg) * \ + (_NIC_)->aq_vecs + \ + ((_ID_) % AQ_NIC_CFG_RING_PER_TC(&(_NIC_)->aq_nic_cfg))) struct aq_hw_rx_fl2 { struct aq_rx_filter_vlan aq_vlans[AQ_VLAN_MAX_FILTERS]; @@ -104,7 +119,7 @@ struct aq_nic_s { atomic_t flags; u32 msg_enable; struct aq_vec_s *aq_vec[AQ_CFG_VECS_MAX]; - struct aq_ring_s *aq_ring_tx[AQ_CFG_VECS_MAX * AQ_CFG_TCS_MAX]; + struct aq_ring_s *aq_ring_tx[AQ_HW_QUEUES_MAX]; struct aq_hw_s *aq_hw; struct net_device *ndev; unsigned int aq_vecs; @@ -164,6 +179,7 @@ void aq_nic_deinit(struct aq_nic_s *self, bool link_down); void aq_nic_set_power(struct aq_nic_s *self); void aq_nic_free_hot_resources(struct aq_nic_s *self); void aq_nic_free_vectors(struct aq_nic_s *self); +int aq_nic_realloc_vectors(struct aq_nic_s *self); int aq_nic_set_mtu(struct aq_nic_s *self, int new_mtu); int aq_nic_set_mac(struct aq_nic_s *self, struct net_device *ndev); int aq_nic_set_packet_filter(struct aq_nic_s *self, unsigned int flags); @@ -181,4 +197,9 @@ void aq_nic_shutdown(struct aq_nic_s *self); u8 aq_nic_reserve_filter(struct aq_nic_s *self, enum aq_rx_filter_type type); void aq_nic_release_filter(struct aq_nic_s *self, enum aq_rx_filter_type type, u32 location); +int aq_nic_setup_tc_mqprio(struct aq_nic_s *self, u32 tcs, u8 *prio_tc_map); +int aq_nic_setup_tc_max_rate(struct aq_nic_s *self, const unsigned int tc, + const u32 max_rate); +int aq_nic_setup_tc_min_rate(struct aq_nic_s *self, const unsigned int tc, + const u32 min_rate); #endif /* AQ_NIC_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c b/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c index 8a70ffe1d326..41c0f560f95b 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_pci_func.c @@ -16,6 +16,7 @@ #include "aq_pci_func.h" #include "hw_atl/hw_atl_a0.h" #include "hw_atl/hw_atl_b0.h" +#include "hw_atl2/hw_atl2.h" #include "aq_filters.h" #include "aq_drvinfo.h" #include "aq_macsec.h" @@ -41,6 +42,13 @@ static const struct pci_device_id aq_pci_tbl[] = { { PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC111S), }, { PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC112S), }, + { PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC113DEV), }, + { PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC113CS), }, + { PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC114CS), }, + { PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC113), }, + { PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC113C), }, + { PCI_VDEVICE(AQUANTIA, AQ_DEVICE_ID_AQC115C), }, + {} }; @@ -70,6 +78,13 @@ static const struct aq_board_revision_s hw_atl_boards[] = { { AQ_DEVICE_ID_AQC109S, AQ_HWREV_ANY, &hw_atl_ops_b1, &hw_atl_b0_caps_aqc109s, }, { AQ_DEVICE_ID_AQC111S, AQ_HWREV_ANY, &hw_atl_ops_b1, &hw_atl_b0_caps_aqc111s, }, { AQ_DEVICE_ID_AQC112S, AQ_HWREV_ANY, &hw_atl_ops_b1, &hw_atl_b0_caps_aqc112s, }, + + { AQ_DEVICE_ID_AQC113DEV, AQ_HWREV_ANY, &hw_atl2_ops, &hw_atl2_caps_aqc113, }, + { AQ_DEVICE_ID_AQC113, AQ_HWREV_ANY, &hw_atl2_ops, &hw_atl2_caps_aqc113, }, + { AQ_DEVICE_ID_AQC113CS, AQ_HWREV_ANY, &hw_atl2_ops, &hw_atl2_caps_aqc113, }, + { AQ_DEVICE_ID_AQC114CS, AQ_HWREV_ANY, &hw_atl2_ops, &hw_atl2_caps_aqc113, }, + { AQ_DEVICE_ID_AQC113C, AQ_HWREV_ANY, &hw_atl2_ops, &hw_atl2_caps_aqc113, }, + { AQ_DEVICE_ID_AQC115C, AQ_HWREV_ANY, &hw_atl2_ops, &hw_atl2_caps_aqc113, }, }; MODULE_DEVICE_TABLE(pci, aq_pci_tbl); @@ -104,10 +119,8 @@ int aq_pci_func_init(struct pci_dev *pdev) int err; err = pci_set_dma_mask(pdev, DMA_BIT_MASK(64)); - if (!err) { + if (!err) err = pci_set_consistent_dma_mask(pdev, DMA_BIT_MASK(64)); - - } if (err) { err = pci_set_dma_mask(pdev, DMA_BIT_MASK(32)); if (!err) @@ -237,6 +250,15 @@ static int aq_pci_probe(struct pci_dev *pdev, goto err_ioremap; } self->aq_hw->aq_nic_cfg = aq_nic_get_cfg(self); + if (self->aq_hw->aq_nic_cfg->aq_hw_caps->priv_data_len) { + int len = self->aq_hw->aq_nic_cfg->aq_hw_caps->priv_data_len; + + self->aq_hw->priv = kzalloc(len, GFP_KERNEL); + if (!self->aq_hw->priv) { + err = -ENOMEM; + goto err_free_aq_hw; + } + } for (bar = 0; bar < 4; ++bar) { if (IORESOURCE_MEM & pci_resource_flags(pdev, bar)) { @@ -245,19 +267,19 @@ static int aq_pci_probe(struct pci_dev *pdev, mmio_pa = pci_resource_start(pdev, bar); if (mmio_pa == 0U) { err = -EIO; - goto err_free_aq_hw; + goto err_free_aq_hw_priv; } reg_sz = pci_resource_len(pdev, bar); if ((reg_sz <= 24 /*ATL_REGS_SIZE*/)) { err = -EIO; - goto err_free_aq_hw; + goto err_free_aq_hw_priv; } self->aq_hw->mmio = ioremap(mmio_pa, reg_sz); if (!self->aq_hw->mmio) { err = -EIO; - goto err_free_aq_hw; + goto err_free_aq_hw_priv; } break; } @@ -265,7 +287,7 @@ static int aq_pci_probe(struct pci_dev *pdev, if (bar == 4) { err = -EIO; - goto err_free_aq_hw; + goto err_free_aq_hw_priv; } numvecs = min((u8)AQ_CFG_VECS_DEF, @@ -305,6 +327,8 @@ err_register: aq_pci_free_irq_vectors(self); err_hwinit: iounmap(self->aq_hw->mmio); +err_free_aq_hw_priv: + kfree(self->aq_hw->priv); err_free_aq_hw: kfree(self->aq_hw); err_ioremap: @@ -332,6 +356,7 @@ static void aq_pci_remove(struct pci_dev *pdev) aq_nic_free_vectors(self); aq_pci_free_irq_vectors(self); iounmap(self->aq_hw->mmio); + kfree(self->aq_hw->priv); kfree(self->aq_hw); pci_release_regions(pdev); free_netdev(self->ndev); @@ -406,6 +431,9 @@ static int atl_resume_common(struct device *dev, bool deep) netif_tx_start_all_queues(nic->ndev); err_exit: + if (ret < 0) + aq_nic_deinit(nic, true); + rtnl_unlock(); return ret; diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ptp.c b/drivers/net/ethernet/aquantia/atlantic/aq_ptp.c index 58e8c641e8b3..599ced261b2a 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ptp.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ptp.c @@ -945,26 +945,29 @@ void aq_ptp_ring_deinit(struct aq_nic_s *aq_nic) #define PTP_4TC_RING_IDX 16 #define PTP_HWST_RING_IDX 31 +/* Index must be 8 (8 TCs) or 16 (4 TCs). + * It depends on Traffic Class mode. + */ +static unsigned int ptp_ring_idx(const enum aq_tc_mode tc_mode) +{ + if (tc_mode == AQ_TC_MODE_8TCS) + return PTP_8TC_RING_IDX; + + return PTP_4TC_RING_IDX; +} + int aq_ptp_ring_alloc(struct aq_nic_s *aq_nic) { struct aq_ptp_s *aq_ptp = aq_nic->aq_ptp; unsigned int tx_ring_idx, rx_ring_idx; struct aq_ring_s *hwts; - u32 tx_tc_mode, rx_tc_mode; struct aq_ring_s *ring; int err; if (!aq_ptp) return 0; - /* Index must to be 8 (8 TCs) or 16 (4 TCs). - * It depends from Traffic Class mode. - */ - aq_nic->aq_hw_ops->hw_tx_tc_mode_get(aq_nic->aq_hw, &tx_tc_mode); - if (tx_tc_mode == 0) - tx_ring_idx = PTP_8TC_RING_IDX; - else - tx_ring_idx = PTP_4TC_RING_IDX; + tx_ring_idx = ptp_ring_idx(aq_nic->aq_nic_cfg.tc_mode); ring = aq_ring_tx_alloc(&aq_ptp->ptp_tx, aq_nic, tx_ring_idx, &aq_nic->aq_nic_cfg); @@ -973,11 +976,7 @@ int aq_ptp_ring_alloc(struct aq_nic_s *aq_nic) goto err_exit; } - aq_nic->aq_hw_ops->hw_rx_tc_mode_get(aq_nic->aq_hw, &rx_tc_mode); - if (rx_tc_mode == 0) - rx_ring_idx = PTP_8TC_RING_IDX; - else - rx_ring_idx = PTP_4TC_RING_IDX; + rx_ring_idx = ptp_ring_idx(aq_nic->aq_nic_cfg.tc_mode); ring = aq_ring_rx_alloc(&aq_ptp->ptp_rx, aq_nic, rx_ring_idx, &aq_nic->aq_nic_cfg); diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c index bae95a618560..68fdb3994088 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c @@ -232,8 +232,11 @@ void aq_ring_queue_wake(struct aq_ring_s *ring) { struct net_device *ndev = aq_nic_get_ndev(ring->aq_nic); - if (__netif_subqueue_stopped(ndev, ring->idx)) { - netif_wake_subqueue(ndev, ring->idx); + if (__netif_subqueue_stopped(ndev, + AQ_NIC_RING2QMAP(ring->aq_nic, + ring->idx))) { + netif_wake_subqueue(ndev, + AQ_NIC_RING2QMAP(ring->aq_nic, ring->idx)); ring->stats.tx.queue_restarts++; } } @@ -242,8 +245,11 @@ void aq_ring_queue_stop(struct aq_ring_s *ring) { struct net_device *ndev = aq_nic_get_ndev(ring->aq_nic); - if (!__netif_subqueue_stopped(ndev, ring->idx)) - netif_stop_subqueue(ndev, ring->idx); + if (!__netif_subqueue_stopped(ndev, + AQ_NIC_RING2QMAP(ring->aq_nic, + ring->idx))) + netif_stop_subqueue(ndev, + AQ_NIC_RING2QMAP(ring->aq_nic, ring->idx)); } bool aq_ring_tx_clean(struct aq_ring_s *self) @@ -466,7 +472,10 @@ int aq_ring_rx_clean(struct aq_ring_s *self, buff->is_hash_l4 ? PKT_HASH_TYPE_L4 : PKT_HASH_TYPE_NONE); /* Send all PTP traffic to 0 queue */ - skb_record_rx_queue(skb, is_ptp_ring ? 0 : self->idx); + skb_record_rx_queue(skb, + is_ptp_ring ? 0 + : AQ_NIC_RING2QMAP(self->aq_nic, + self->idx)); ++self->stats.rx.packets; self->stats.rx.bytes += skb->len; diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_vec.c b/drivers/net/ethernet/aquantia/atlantic/aq_vec.c index f40a427970dc..d1d43c8ce400 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_vec.c +++ b/drivers/net/ethernet/aquantia/atlantic/aq_vec.c @@ -103,16 +103,11 @@ err_exit: struct aq_vec_s *aq_vec_alloc(struct aq_nic_s *aq_nic, unsigned int idx, struct aq_nic_cfg_s *aq_nic_cfg) { - struct aq_ring_s *ring = NULL; struct aq_vec_s *self = NULL; - unsigned int i = 0U; - int err = 0; self = kzalloc(sizeof(*self), GFP_KERNEL); - if (!self) { - err = -ENOMEM; + if (!self) goto err_exit; - } self->aq_nic = aq_nic; self->aq_ring_param.vec_idx = idx; @@ -128,10 +123,20 @@ struct aq_vec_s *aq_vec_alloc(struct aq_nic_s *aq_nic, unsigned int idx, netif_napi_add(aq_nic_get_ndev(aq_nic), &self->napi, aq_vec_poll, AQ_CFG_NAPI_WEIGHT); +err_exit: + return self; +} + +int aq_vec_ring_alloc(struct aq_vec_s *self, struct aq_nic_s *aq_nic, + unsigned int idx, struct aq_nic_cfg_s *aq_nic_cfg) +{ + struct aq_ring_s *ring = NULL; + unsigned int i = 0U; + int err = 0; + for (i = 0; i < aq_nic_cfg->tcs; ++i) { - unsigned int idx_ring = AQ_NIC_TCVEC2RING(self->nic, - self->tx_rings, - self->aq_ring_param.vec_idx); + const unsigned int idx_ring = AQ_NIC_CFG_TCVEC2RING(aq_nic_cfg, + i, idx); ring = aq_ring_tx_alloc(&self->ring[i][AQ_VEC_TX_ID], aq_nic, idx_ring, aq_nic_cfg); @@ -156,11 +161,11 @@ struct aq_vec_s *aq_vec_alloc(struct aq_nic_s *aq_nic, unsigned int idx, err_exit: if (err < 0) { - aq_vec_free(self); + aq_vec_ring_free(self); self = NULL; } - return self; + return err; } int aq_vec_init(struct aq_vec_s *self, const struct aq_hw_ops *aq_hw_ops, @@ -270,6 +275,18 @@ err_exit:; void aq_vec_free(struct aq_vec_s *self) { + if (!self) + goto err_exit; + + netif_napi_del(&self->napi); + + kfree(self); + +err_exit:; +} + +void aq_vec_ring_free(struct aq_vec_s *self) +{ struct aq_ring_s *ring = NULL; unsigned int i = 0U; @@ -279,13 +296,12 @@ void aq_vec_free(struct aq_vec_s *self) for (i = 0U, ring = self->ring[0]; self->tx_rings > i; ++i, ring = self->ring[i]) { aq_ring_free(&ring[AQ_VEC_TX_ID]); - aq_ring_free(&ring[AQ_VEC_RX_ID]); + if (i < self->rx_rings) + aq_ring_free(&ring[AQ_VEC_RX_ID]); } - netif_napi_del(&self->napi); - - kfree(self); - + self->tx_rings = 0; + self->rx_rings = 0; err_exit:; } @@ -333,16 +349,14 @@ cpumask_t *aq_vec_get_affinity_mask(struct aq_vec_s *self) return &self->aq_ring_param.affinity_mask; } -void aq_vec_add_stats(struct aq_vec_s *self, - struct aq_ring_stats_rx_s *stats_rx, - struct aq_ring_stats_tx_s *stats_tx) +static void aq_vec_add_stats(struct aq_vec_s *self, + const unsigned int tc, + struct aq_ring_stats_rx_s *stats_rx, + struct aq_ring_stats_tx_s *stats_tx) { - struct aq_ring_s *ring = NULL; - unsigned int r = 0U; + struct aq_ring_s *ring = self->ring[tc]; - for (r = 0U, ring = self->ring[0]; - self->tx_rings > r; ++r, ring = self->ring[r]) { - struct aq_ring_stats_tx_s *tx = &ring[AQ_VEC_TX_ID].stats.tx; + if (tc < self->rx_rings) { struct aq_ring_stats_rx_s *rx = &ring[AQ_VEC_RX_ID].stats.rx; stats_rx->packets += rx->packets; @@ -353,6 +367,10 @@ void aq_vec_add_stats(struct aq_vec_s *self, stats_rx->pg_losts += rx->pg_losts; stats_rx->pg_flips += rx->pg_flips; stats_rx->pg_reuses += rx->pg_reuses; + } + + if (tc < self->tx_rings) { + struct aq_ring_stats_tx_s *tx = &ring[AQ_VEC_TX_ID].stats.tx; stats_tx->packets += tx->packets; stats_tx->bytes += tx->bytes; @@ -361,7 +379,8 @@ void aq_vec_add_stats(struct aq_vec_s *self, } } -int aq_vec_get_sw_stats(struct aq_vec_s *self, u64 *data, unsigned int *p_count) +int aq_vec_get_sw_stats(struct aq_vec_s *self, const unsigned int tc, u64 *data, + unsigned int *p_count) { struct aq_ring_stats_rx_s stats_rx; struct aq_ring_stats_tx_s stats_tx; @@ -369,7 +388,8 @@ int aq_vec_get_sw_stats(struct aq_vec_s *self, u64 *data, unsigned int *p_count) memset(&stats_rx, 0U, sizeof(struct aq_ring_stats_rx_s)); memset(&stats_tx, 0U, sizeof(struct aq_ring_stats_tx_s)); - aq_vec_add_stats(self, &stats_rx, &stats_tx); + + aq_vec_add_stats(self, tc, &stats_rx, &stats_tx); /* This data should mimic aq_ethtool_queue_stat_names structure */ diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_vec.h b/drivers/net/ethernet/aquantia/atlantic/aq_vec.h index 0fe8e0904c7f..541af85e6510 100644 --- a/drivers/net/ethernet/aquantia/atlantic/aq_vec.h +++ b/drivers/net/ethernet/aquantia/atlantic/aq_vec.h @@ -25,17 +25,17 @@ irqreturn_t aq_vec_isr(int irq, void *private); irqreturn_t aq_vec_isr_legacy(int irq, void *private); struct aq_vec_s *aq_vec_alloc(struct aq_nic_s *aq_nic, unsigned int idx, struct aq_nic_cfg_s *aq_nic_cfg); +int aq_vec_ring_alloc(struct aq_vec_s *self, struct aq_nic_s *aq_nic, + unsigned int idx, struct aq_nic_cfg_s *aq_nic_cfg); int aq_vec_init(struct aq_vec_s *self, const struct aq_hw_ops *aq_hw_ops, struct aq_hw_s *aq_hw); void aq_vec_deinit(struct aq_vec_s *self); void aq_vec_free(struct aq_vec_s *self); +void aq_vec_ring_free(struct aq_vec_s *self); int aq_vec_start(struct aq_vec_s *self); void aq_vec_stop(struct aq_vec_s *self); cpumask_t *aq_vec_get_affinity_mask(struct aq_vec_s *self); -int aq_vec_get_sw_stats(struct aq_vec_s *self, u64 *data, +int aq_vec_get_sw_stats(struct aq_vec_s *self, const unsigned int tc, u64 *data, unsigned int *p_count); -void aq_vec_add_stats(struct aq_vec_s *self, - struct aq_ring_stats_rx_s *stats_rx, - struct aq_ring_stats_tx_s *stats_tx); #endif /* AQ_VEC_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c index 9b1062b8af64..a312864969af 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_a0.c @@ -1,7 +1,8 @@ // SPDX-License-Identifier: GPL-2.0-only -/* - * aQuantia Corporation Network Driver - * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved +/* Atlantic Network Driver + * + * Copyright (C) 2014-2019 aQuantia Corporation + * Copyright (C) 2019-2020 Marvell International Ltd. */ /* File hw_atl_a0.c: Definition of Atlantic hardware specific functions. */ @@ -20,7 +21,7 @@ .msix_irqs = 4U, \ .irq_mask = ~0U, \ .vecs = HW_ATL_A0_RSS_MAX, \ - .tcs = HW_ATL_A0_TC_MAX, \ + .tcs_max = HW_ATL_A0_TC_MAX, \ .rxd_alignment = 1U, \ .rxd_size = HW_ATL_A0_RXD_SIZE, \ .rxds_max = HW_ATL_A0_MAX_RXD, \ @@ -47,7 +48,7 @@ const struct aq_hw_caps_s hw_atl_a0_caps_aqc100 = { DEFAULT_A0_BOARD_BASIC_CAPABILITIES, .media_type = AQ_HW_MEDIA_TYPE_FIBRE, .link_speed_msk = AQ_NIC_RATE_5G | - AQ_NIC_RATE_2GS | + AQ_NIC_RATE_2G5 | AQ_NIC_RATE_1G | AQ_NIC_RATE_100M, }; @@ -57,7 +58,7 @@ const struct aq_hw_caps_s hw_atl_a0_caps_aqc107 = { .media_type = AQ_HW_MEDIA_TYPE_TP, .link_speed_msk = AQ_NIC_RATE_10G | AQ_NIC_RATE_5G | - AQ_NIC_RATE_2GS | + AQ_NIC_RATE_2G5 | AQ_NIC_RATE_1G | AQ_NIC_RATE_100M, }; @@ -66,7 +67,7 @@ const struct aq_hw_caps_s hw_atl_a0_caps_aqc108 = { DEFAULT_A0_BOARD_BASIC_CAPABILITIES, .media_type = AQ_HW_MEDIA_TYPE_TP, .link_speed_msk = AQ_NIC_RATE_5G | - AQ_NIC_RATE_2GS | + AQ_NIC_RATE_2G5 | AQ_NIC_RATE_1G | AQ_NIC_RATE_100M, }; @@ -74,7 +75,7 @@ const struct aq_hw_caps_s hw_atl_a0_caps_aqc108 = { const struct aq_hw_caps_s hw_atl_a0_caps_aqc109 = { DEFAULT_A0_BOARD_BASIC_CAPABILITIES, .media_type = AQ_HW_MEDIA_TYPE_TP, - .link_speed_msk = AQ_NIC_RATE_2GS | + .link_speed_msk = AQ_NIC_RATE_2G5 | AQ_NIC_RATE_1G | AQ_NIC_RATE_100M, }; @@ -135,10 +136,10 @@ static int hw_atl_a0_hw_qos_set(struct aq_hw_s *self) hw_atl_tps_tx_pkt_shed_desc_tc_arb_mode_set(self, 0U); hw_atl_tps_tx_pkt_shed_data_arb_mode_set(self, 0U); - hw_atl_tps_tx_pkt_shed_tc_data_max_credit_set(self, 0xFFF, 0U); - hw_atl_tps_tx_pkt_shed_tc_data_weight_set(self, 0x64, 0U); - hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(self, 0x50, 0U); - hw_atl_tps_tx_pkt_shed_desc_tc_weight_set(self, 0x1E, 0U); + hw_atl_tps_tx_pkt_shed_tc_data_max_credit_set(self, 0U, 0xFFF); + hw_atl_tps_tx_pkt_shed_tc_data_weight_set(self, 0U, 0x64); + hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(self, 0U, 0x50); + hw_atl_tps_tx_pkt_shed_desc_tc_weight_set(self, 0U, 0x1E); /* Tx buf size */ buff_size = HW_ATL_A0_TXBUF_MAX; @@ -267,8 +268,7 @@ static int hw_atl_a0_hw_init_tx_path(struct aq_hw_s *self) hw_atl_tdm_tx_desc_wr_wb_irq_en_set(self, 1U); /* misc */ - aq_hw_write_reg(self, 0x00007040U, IS_CHIP_FEATURE(TPO2) ? - 0x00010000U : 0x00000000U); + aq_hw_write_reg(self, 0x00007040U, 0x00000000U); hw_atl_tdm_tx_dca_en_set(self, 0U); hw_atl_tdm_tx_dca_mode_set(self, 0U); @@ -886,6 +886,8 @@ static int hw_atl_a0_hw_ring_rx_stop(struct aq_hw_s *self, } const struct aq_hw_ops hw_atl_ops_a0 = { + .hw_soft_reset = hw_atl_utils_soft_reset, + .hw_prepare = hw_atl_utils_initfw, .hw_set_mac_address = hw_atl_a0_hw_mac_addr_set, .hw_init = hw_atl_a0_hw_init, .hw_reset = hw_atl_a0_hw_reset, diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c index d20d91cdece8..14d79f70cad7 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c @@ -1,7 +1,8 @@ // SPDX-License-Identifier: GPL-2.0-only -/* - * aQuantia Corporation Network Driver - * Copyright (C) 2014-2019 aQuantia Corporation. All rights reserved +/* Atlantic Network Driver + * + * Copyright (C) 2014-2019 aQuantia Corporation + * Copyright (C) 2019-2020 Marvell International Ltd. */ /* File hw_atl_b0.c: Definition of Atlantic hardware specific functions. */ @@ -22,7 +23,7 @@ .msix_irqs = 8U, \ .irq_mask = ~0U, \ .vecs = HW_ATL_B0_RSS_MAX, \ - .tcs = HW_ATL_B0_TC_MAX, \ + .tcs_max = HW_ATL_B0_TC_MAX, \ .rxd_alignment = 1U, \ .rxd_size = HW_ATL_B0_RXD_SIZE, \ .rxds_max = HW_ATL_B0_MAX_RXD, \ @@ -45,7 +46,8 @@ NETIF_F_HW_VLAN_CTAG_RX | \ NETIF_F_HW_VLAN_CTAG_TX | \ NETIF_F_GSO_UDP_L4 | \ - NETIF_F_GSO_PARTIAL, \ + NETIF_F_GSO_PARTIAL | \ + NETIF_F_HW_TC, \ .hw_priv_flags = IFF_UNICAST_FLT, \ .flow_control = true, \ .mtu = HW_ATL_B0_MTU_JUMBO, \ @@ -59,7 +61,7 @@ const struct aq_hw_caps_s hw_atl_b0_caps_aqc100 = { .media_type = AQ_HW_MEDIA_TYPE_FIBRE, .link_speed_msk = AQ_NIC_RATE_10G | AQ_NIC_RATE_5G | - AQ_NIC_RATE_2GS | + AQ_NIC_RATE_2G5 | AQ_NIC_RATE_1G | AQ_NIC_RATE_100M, }; @@ -69,7 +71,7 @@ const struct aq_hw_caps_s hw_atl_b0_caps_aqc107 = { .media_type = AQ_HW_MEDIA_TYPE_TP, .link_speed_msk = AQ_NIC_RATE_10G | AQ_NIC_RATE_5G | - AQ_NIC_RATE_2GS | + AQ_NIC_RATE_2G5 | AQ_NIC_RATE_1G | AQ_NIC_RATE_100M, }; @@ -78,7 +80,7 @@ const struct aq_hw_caps_s hw_atl_b0_caps_aqc108 = { DEFAULT_B0_BOARD_BASIC_CAPABILITIES, .media_type = AQ_HW_MEDIA_TYPE_TP, .link_speed_msk = AQ_NIC_RATE_5G | - AQ_NIC_RATE_2GS | + AQ_NIC_RATE_2G5 | AQ_NIC_RATE_1G | AQ_NIC_RATE_100M, }; @@ -86,7 +88,7 @@ const struct aq_hw_caps_s hw_atl_b0_caps_aqc108 = { const struct aq_hw_caps_s hw_atl_b0_caps_aqc109 = { DEFAULT_B0_BOARD_BASIC_CAPABILITIES, .media_type = AQ_HW_MEDIA_TYPE_TP, - .link_speed_msk = AQ_NIC_RATE_2GS | + .link_speed_msk = AQ_NIC_RATE_2G5 | AQ_NIC_RATE_1G | AQ_NIC_RATE_100M, }; @@ -113,12 +115,34 @@ static int hw_atl_b0_set_fc(struct aq_hw_s *self, u32 fc, u32 tc) return 0; } +static int hw_atl_b0_tc_ptp_set(struct aq_hw_s *self) +{ + /* Init TC2 for PTP_TX */ + hw_atl_tpb_tx_pkt_buff_size_per_tc_set(self, HW_ATL_B0_PTP_TXBUF_SIZE, + AQ_HW_PTP_TC); + + /* Init TC2 for PTP_RX */ + hw_atl_rpb_rx_pkt_buff_size_per_tc_set(self, HW_ATL_B0_PTP_RXBUF_SIZE, + AQ_HW_PTP_TC); + /* No flow control for PTP */ + hw_atl_rpb_rx_xoff_en_per_tc_set(self, 0U, AQ_HW_PTP_TC); + + return aq_hw_err_from_flags(self); +} + static int hw_atl_b0_hw_qos_set(struct aq_hw_s *self) { - unsigned int i_priority = 0U; - u32 buff_size = 0U; + struct aq_nic_cfg_s *cfg = self->aq_nic_cfg; + u32 tx_buff_size = HW_ATL_B0_TXBUF_MAX; + u32 rx_buff_size = HW_ATL_B0_RXBUF_MAX; + unsigned int prio = 0U; u32 tc = 0U; + if (cfg->is_ptp) { + tx_buff_size -= HW_ATL_B0_PTP_TXBUF_SIZE; + rx_buff_size -= HW_ATL_B0_PTP_RXBUF_SIZE; + } + /* TPS Descriptor rate init */ hw_atl_tps_tx_pkt_shed_desc_rate_curr_time_res_set(self, 0x0U); hw_atl_tps_tx_pkt_shed_desc_rate_lim_set(self, 0xA); @@ -126,69 +150,45 @@ static int hw_atl_b0_hw_qos_set(struct aq_hw_s *self) /* TPS VM init */ hw_atl_tps_tx_pkt_shed_desc_vm_arb_mode_set(self, 0U); - /* TPS TC credits init */ - hw_atl_tps_tx_pkt_shed_desc_tc_arb_mode_set(self, 0U); - hw_atl_tps_tx_pkt_shed_data_arb_mode_set(self, 0U); - - tc = 0; - - /* TX Packet Scheduler Data TC0 */ - hw_atl_tps_tx_pkt_shed_tc_data_max_credit_set(self, 0xFFF, tc); - hw_atl_tps_tx_pkt_shed_tc_data_weight_set(self, 0x64, tc); - hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(self, 0x50, tc); - hw_atl_tps_tx_pkt_shed_desc_tc_weight_set(self, 0x1E, tc); - - /* Tx buf size TC0 */ - buff_size = HW_ATL_B0_TXBUF_MAX - HW_ATL_B0_PTP_TXBUF_SIZE; - - hw_atl_tpb_tx_pkt_buff_size_per_tc_set(self, buff_size, tc); - hw_atl_tpb_tx_buff_hi_threshold_per_tc_set(self, - (buff_size * - (1024 / 32U) * 66U) / - 100U, tc); - hw_atl_tpb_tx_buff_lo_threshold_per_tc_set(self, - (buff_size * - (1024 / 32U) * 50U) / - 100U, tc); - /* Init TC2 for PTP_TX */ - tc = 2; + tx_buff_size /= cfg->tcs; + rx_buff_size /= cfg->tcs; + for (tc = 0; tc < cfg->tcs; tc++) { + u32 threshold = 0U; - hw_atl_tpb_tx_pkt_buff_size_per_tc_set(self, HW_ATL_B0_PTP_TXBUF_SIZE, - tc); + /* Tx buf size TC0 */ + hw_atl_tpb_tx_pkt_buff_size_per_tc_set(self, tx_buff_size, tc); - /* QoS Rx buf size per TC */ - tc = 0; - buff_size = HW_ATL_B0_RXBUF_MAX - HW_ATL_B0_PTP_RXBUF_SIZE; + threshold = (tx_buff_size * (1024 / 32U) * 66U) / 100U; + hw_atl_tpb_tx_buff_hi_threshold_per_tc_set(self, threshold, tc); - hw_atl_rpb_rx_pkt_buff_size_per_tc_set(self, buff_size, tc); - hw_atl_rpb_rx_buff_hi_threshold_per_tc_set(self, - (buff_size * - (1024U / 32U) * 66U) / - 100U, tc); - hw_atl_rpb_rx_buff_lo_threshold_per_tc_set(self, - (buff_size * - (1024U / 32U) * 50U) / - 100U, tc); + threshold = (tx_buff_size * (1024 / 32U) * 50U) / 100U; + hw_atl_tpb_tx_buff_lo_threshold_per_tc_set(self, threshold, tc); - hw_atl_b0_set_fc(self, self->aq_nic_cfg->fc.req, tc); + /* QoS Rx buf size per TC */ + hw_atl_rpb_rx_pkt_buff_size_per_tc_set(self, rx_buff_size, tc); - /* Init TC2 for PTP_RX */ - tc = 2; + threshold = (rx_buff_size * (1024U / 32U) * 66U) / 100U; + hw_atl_rpb_rx_buff_hi_threshold_per_tc_set(self, threshold, tc); - hw_atl_rpb_rx_pkt_buff_size_per_tc_set(self, HW_ATL_B0_PTP_RXBUF_SIZE, - tc); - /* No flow control for PTP */ - hw_atl_rpb_rx_xoff_en_per_tc_set(self, 0U, tc); + threshold = (rx_buff_size * (1024U / 32U) * 50U) / 100U; + hw_atl_rpb_rx_buff_lo_threshold_per_tc_set(self, threshold, tc); + + hw_atl_b0_set_fc(self, self->aq_nic_cfg->fc.req, tc); + } + + if (cfg->is_ptp) + hw_atl_b0_tc_ptp_set(self); /* QoS 802.1p priority -> TC mapping */ - for (i_priority = 8U; i_priority--;) - hw_atl_rpf_rpb_user_priority_tc_map_set(self, i_priority, 0U); + for (prio = 0; prio < 8; ++prio) + hw_atl_rpf_rpb_user_priority_tc_map_set(self, prio, + cfg->prio_tc_map[prio]); return aq_hw_err_from_flags(self); } -static int hw_atl_b0_hw_rss_hash_set(struct aq_hw_s *self, - struct aq_rss_parameters *rss_params) +int hw_atl_b0_hw_rss_hash_set(struct aq_hw_s *self, + struct aq_rss_parameters *rss_params) { struct aq_nic_cfg_s *cfg = self->aq_nic_cfg; unsigned int addr = 0U; @@ -251,9 +251,10 @@ err_exit: return err; } -static int hw_atl_b0_hw_offload_set(struct aq_hw_s *self, - struct aq_nic_cfg_s *aq_nic_cfg) +int hw_atl_b0_hw_offload_set(struct aq_hw_s *self, + struct aq_nic_cfg_s *aq_nic_cfg) { + u64 rxcsum = !!(aq_nic_cfg->features & NETIF_F_RXCSUM); unsigned int i; /* TX checksums offloads*/ @@ -261,10 +262,8 @@ static int hw_atl_b0_hw_offload_set(struct aq_hw_s *self, hw_atl_tpo_tcp_udp_crc_offload_en_set(self, 1); /* RX checksums offloads*/ - hw_atl_rpo_ipv4header_crc_offload_en_set(self, !!(aq_nic_cfg->features & - NETIF_F_RXCSUM)); - hw_atl_rpo_tcp_udp_crc_offload_en_set(self, !!(aq_nic_cfg->features & - NETIF_F_RXCSUM)); + hw_atl_rpo_ipv4header_crc_offload_en_set(self, rxcsum); + hw_atl_rpo_tcp_udp_crc_offload_en_set(self, rxcsum); /* LSO offloads*/ hw_atl_tdm_large_send_offload_en_set(self, 0xFFFFFFFFU); @@ -272,7 +271,7 @@ static int hw_atl_b0_hw_offload_set(struct aq_hw_s *self, /* Outer VLAN tag offload */ hw_atl_rpo_outer_vlan_tag_mode_set(self, 1U); -/* LRO offloads */ + /* LRO offloads */ { unsigned int val = (8U < HW_ATL_B0_LRO_RXD_MAX) ? 0x3U : ((4U < HW_ATL_B0_LRO_RXD_MAX) ? 0x2U : @@ -311,10 +310,124 @@ static int hw_atl_b0_hw_offload_set(struct aq_hw_s *self, return aq_hw_err_from_flags(self); } +static int hw_atl_b0_hw_init_tx_tc_rate_limit(struct aq_hw_s *self) +{ + static const u32 max_weight = BIT(HW_ATL_TPS_DATA_TCTWEIGHT_WIDTH) - 1; + /* Scale factor is based on the number of bits in fractional portion */ + static const u32 scale = BIT(HW_ATL_TPS_DESC_RATE_Y_WIDTH); + static const u32 frac_msk = HW_ATL_TPS_DESC_RATE_Y_MSK >> + HW_ATL_TPS_DESC_RATE_Y_SHIFT; + const u32 link_speed = self->aq_link_status.mbps; + struct aq_nic_cfg_s *nic_cfg = self->aq_nic_cfg; + unsigned long num_min_rated_tcs = 0; + u32 tc_weight[AQ_CFG_TCS_MAX]; + u32 fixed_max_credit; + u8 min_rate_msk = 0; + u32 sum_weight = 0; + int tc; + + /* By default max_credit is based upon MTU (in unit of 64b) */ + fixed_max_credit = nic_cfg->aq_hw_caps->mtu / 64; + + if (link_speed) { + min_rate_msk = nic_cfg->tc_min_rate_msk & + (BIT(nic_cfg->tcs) - 1); + num_min_rated_tcs = hweight8(min_rate_msk); + } + + /* First, calculate weights where min_rate is specified */ + if (num_min_rated_tcs) { + for (tc = 0; tc != nic_cfg->tcs; tc++) { + if (!nic_cfg->tc_min_rate[tc]) { + tc_weight[tc] = 0; + continue; + } + + tc_weight[tc] = (-1L + link_speed + + nic_cfg->tc_min_rate[tc] * + max_weight) / + link_speed; + tc_weight[tc] = min(tc_weight[tc], max_weight); + sum_weight += tc_weight[tc]; + } + } + + /* WSP, if min_rate is set for at least one TC. + * RR otherwise. + */ + hw_atl_tps_tx_pkt_shed_data_arb_mode_set(self, min_rate_msk ? 1U : 0U); + /* Data TC Arbiter takes precedence over Descriptor TC Arbiter, + * leave Descriptor TC Arbiter as RR. + */ + hw_atl_tps_tx_pkt_shed_desc_tc_arb_mode_set(self, 0U); + + hw_atl_tps_tx_desc_rate_mode_set(self, nic_cfg->is_qos ? 1U : 0U); + + for (tc = 0; tc != nic_cfg->tcs; tc++) { + const u32 en = (nic_cfg->tc_max_rate[tc] != 0) ? 1U : 0U; + const u32 desc = AQ_NIC_CFG_TCVEC2RING(nic_cfg, tc, 0); + u32 weight, max_credit; + + hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(self, tc, + fixed_max_credit); + hw_atl_tps_tx_pkt_shed_desc_tc_weight_set(self, tc, 0x1E); + + if (num_min_rated_tcs) { + weight = tc_weight[tc]; + + if (!weight && sum_weight < max_weight) + weight = (max_weight - sum_weight) / + (nic_cfg->tcs - num_min_rated_tcs); + else if (!weight) + weight = 0x64; + + max_credit = max(8 * weight, fixed_max_credit); + } else { + weight = 0x64; + max_credit = 0xFFF; + } + + hw_atl_tps_tx_pkt_shed_tc_data_weight_set(self, tc, weight); + hw_atl_tps_tx_pkt_shed_tc_data_max_credit_set(self, tc, + max_credit); + + hw_atl_tps_tx_desc_rate_en_set(self, desc, en); + + if (en) { + /* Nominal rate is always 10G */ + const u32 rate = 10000U * scale / + nic_cfg->tc_max_rate[tc]; + const u32 rate_int = rate >> + HW_ATL_TPS_DESC_RATE_Y_WIDTH; + const u32 rate_frac = rate & frac_msk; + + hw_atl_tps_tx_desc_rate_x_set(self, desc, rate_int); + hw_atl_tps_tx_desc_rate_y_set(self, desc, rate_frac); + } else { + /* A value of 1 indicates the queue is not + * rate controlled. + */ + hw_atl_tps_tx_desc_rate_x_set(self, desc, 1U); + hw_atl_tps_tx_desc_rate_y_set(self, desc, 0U); + } + } + for (tc = nic_cfg->tcs; tc != AQ_CFG_TCS_MAX; tc++) { + const u32 desc = AQ_NIC_CFG_TCVEC2RING(nic_cfg, tc, 0); + + hw_atl_tps_tx_desc_rate_en_set(self, desc, 0U); + hw_atl_tps_tx_desc_rate_x_set(self, desc, 1U); + hw_atl_tps_tx_desc_rate_y_set(self, desc, 0U); + } + + return aq_hw_err_from_flags(self); +} + static int hw_atl_b0_hw_init_tx_path(struct aq_hw_s *self) { + struct aq_nic_cfg_s *nic_cfg = self->aq_nic_cfg; + /* Tx TC/Queue number config */ - hw_atl_rpb_tps_tx_tc_mode_set(self, 1U); + hw_atl_tpb_tps_tx_tc_mode_set(self, nic_cfg->tc_mode); hw_atl_thm_lso_tcp_flag_of_first_pkt_set(self, 0x0FF6U); hw_atl_thm_lso_tcp_flag_of_middle_pkt_set(self, 0x0FF6U); @@ -324,7 +437,7 @@ static int hw_atl_b0_hw_init_tx_path(struct aq_hw_s *self) hw_atl_tdm_tx_desc_wr_wb_irq_en_set(self, 1U); /* misc */ - aq_hw_write_reg(self, 0x00007040U, IS_CHIP_FEATURE(TPO2) ? + aq_hw_write_reg(self, 0x00007040U, ATL_HW_IS_CHIP_FEATURE(self, TPO2) ? 0x00010000U : 0x00000000U); hw_atl_tdm_tx_dca_en_set(self, 0U); hw_atl_tdm_tx_dca_mode_set(self, 0U); @@ -334,20 +447,32 @@ static int hw_atl_b0_hw_init_tx_path(struct aq_hw_s *self) return aq_hw_err_from_flags(self); } +void hw_atl_b0_hw_init_rx_rss_ctrl1(struct aq_hw_s *self) +{ + struct aq_nic_cfg_s *cfg = self->aq_nic_cfg; + u32 rss_ctrl1 = HW_ATL_RSS_DISABLED; + + if (cfg->is_rss) + rss_ctrl1 = (cfg->tc_mode == AQ_TC_MODE_8TCS) ? + HW_ATL_RSS_ENABLED_8TCS_2INDEX_BITS : + HW_ATL_RSS_ENABLED_4TCS_3INDEX_BITS; + + hw_atl_reg_rx_flr_rss_control1set(self, rss_ctrl1); +} + static int hw_atl_b0_hw_init_rx_path(struct aq_hw_s *self) { struct aq_nic_cfg_s *cfg = self->aq_nic_cfg; int i; /* Rx TC/RSS number config */ - hw_atl_rpb_rpf_rx_traf_class_mode_set(self, 1U); + hw_atl_rpb_rpf_rx_traf_class_mode_set(self, cfg->tc_mode); /* Rx flow control */ hw_atl_rpb_rx_flow_ctl_mode_set(self, 1U); /* RSS Ring selection */ - hw_atl_reg_rx_flr_rss_control1set(self, cfg->is_rss ? - 0xB3333333U : 0x00000000U); + hw_atl_b0_hw_init_rx_rss_ctrl1(self); /* Multicast filters */ for (i = HW_ATL_B0_MAC_MAX; i--;) { @@ -372,8 +497,8 @@ static int hw_atl_b0_hw_init_rx_path(struct aq_hw_s *self) hw_atl_rdm_rx_desc_wr_wb_irq_en_set(self, 1U); /* misc */ - aq_hw_write_reg(self, 0x00005040U, - IS_CHIP_FEATURE(RPF2) ? 0x000F0000U : 0x00000000U); + aq_hw_write_reg(self, 0x00005040U, ATL_HW_IS_CHIP_FEATURE(self, RPF2) ? + 0x000F0000U : 0x00000000U); hw_atl_rpfl2broadcast_flr_act_set(self, 1U); hw_atl_rpfl2broadcast_count_threshold_set(self, 0xFFFFU & (~0U / 256U)); @@ -384,7 +509,7 @@ static int hw_atl_b0_hw_init_rx_path(struct aq_hw_s *self) return aq_hw_err_from_flags(self); } -static int hw_atl_b0_hw_mac_addr_set(struct aq_hw_s *self, u8 *mac_addr) +int hw_atl_b0_hw_mac_addr_set(struct aq_hw_s *self, u8 *mac_addr) { unsigned int h = 0U; unsigned int l = 0U; @@ -479,23 +604,21 @@ err_exit: return err; } -static int hw_atl_b0_hw_ring_tx_start(struct aq_hw_s *self, - struct aq_ring_s *ring) +int hw_atl_b0_hw_ring_tx_start(struct aq_hw_s *self, struct aq_ring_s *ring) { hw_atl_tdm_tx_desc_en_set(self, 1, ring->idx); return aq_hw_err_from_flags(self); } -static int hw_atl_b0_hw_ring_rx_start(struct aq_hw_s *self, - struct aq_ring_s *ring) +int hw_atl_b0_hw_ring_rx_start(struct aq_hw_s *self, struct aq_ring_s *ring) { hw_atl_rdm_rx_desc_en_set(self, 1, ring->idx); return aq_hw_err_from_flags(self); } -static int hw_atl_b0_hw_start(struct aq_hw_s *self) +int hw_atl_b0_hw_start(struct aq_hw_s *self) { hw_atl_tpb_tx_buff_en_set(self, 1); hw_atl_rpb_rx_buff_en_set(self, 1); @@ -511,9 +634,8 @@ static int hw_atl_b0_hw_tx_ring_tail_update(struct aq_hw_s *self, return 0; } -static int hw_atl_b0_hw_ring_tx_xmit(struct aq_hw_s *self, - struct aq_ring_s *ring, - unsigned int frags) +int hw_atl_b0_hw_ring_tx_xmit(struct aq_hw_s *self, struct aq_ring_s *ring, + unsigned int frags) { struct aq_ring_buff_s *buff = NULL; struct hw_atl_txd_s *txd = NULL; @@ -600,9 +722,8 @@ static int hw_atl_b0_hw_ring_tx_xmit(struct aq_hw_s *self, return aq_hw_err_from_flags(self); } -static int hw_atl_b0_hw_ring_rx_init(struct aq_hw_s *self, - struct aq_ring_s *aq_ring, - struct aq_ring_param_s *aq_ring_param) +int hw_atl_b0_hw_ring_rx_init(struct aq_hw_s *self, struct aq_ring_s *aq_ring, + struct aq_ring_param_s *aq_ring_param) { u32 dma_desc_addr_msw = (u32)(((u64)aq_ring->dx_ring_pa) >> 32); u32 vlan_rx_stripping = self->aq_nic_cfg->is_vlan_rx_strip; @@ -643,9 +764,8 @@ static int hw_atl_b0_hw_ring_rx_init(struct aq_hw_s *self, return aq_hw_err_from_flags(self); } -static int hw_atl_b0_hw_ring_tx_init(struct aq_hw_s *self, - struct aq_ring_s *aq_ring, - struct aq_ring_param_s *aq_ring_param) +int hw_atl_b0_hw_ring_tx_init(struct aq_hw_s *self, struct aq_ring_s *aq_ring, + struct aq_ring_param_s *aq_ring_param) { u32 dma_desc_msw_addr = (u32)(((u64)aq_ring->dx_ring_pa) >> 32); u32 dma_desc_lsw_addr = (u32)aq_ring->dx_ring_pa; @@ -673,9 +793,8 @@ static int hw_atl_b0_hw_ring_tx_init(struct aq_hw_s *self, return aq_hw_err_from_flags(self); } -static int hw_atl_b0_hw_ring_rx_fill(struct aq_hw_s *self, - struct aq_ring_s *ring, - unsigned int sw_tail_old) +int hw_atl_b0_hw_ring_rx_fill(struct aq_hw_s *self, struct aq_ring_s *ring, + unsigned int sw_tail_old) { for (; sw_tail_old != ring->sw_tail; sw_tail_old = aq_ring_next_dx(ring, sw_tail_old)) { @@ -734,8 +853,8 @@ static int hw_atl_b0_hw_ring_hwts_rx_receive(struct aq_hw_s *self, return aq_hw_err_from_flags(self); } -static int hw_atl_b0_hw_ring_tx_head_update(struct aq_hw_s *self, - struct aq_ring_s *ring) +int hw_atl_b0_hw_ring_tx_head_update(struct aq_hw_s *self, + struct aq_ring_s *ring) { unsigned int hw_head_; int err = 0; @@ -753,8 +872,7 @@ err_exit: return err; } -static int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self, - struct aq_ring_s *ring) +int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self, struct aq_ring_s *ring) { for (; ring->hw_head != ring->sw_tail; ring->hw_head = aq_ring_next_dx(ring, ring->hw_head)) { @@ -854,14 +972,14 @@ static int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self, return aq_hw_err_from_flags(self); } -static int hw_atl_b0_hw_irq_enable(struct aq_hw_s *self, u64 mask) +int hw_atl_b0_hw_irq_enable(struct aq_hw_s *self, u64 mask) { hw_atl_itr_irq_msk_setlsw_set(self, LODWORD(mask)); return aq_hw_err_from_flags(self); } -static int hw_atl_b0_hw_irq_disable(struct aq_hw_s *self, u64 mask) +int hw_atl_b0_hw_irq_disable(struct aq_hw_s *self, u64 mask) { hw_atl_itr_irq_msk_clearlsw_set(self, LODWORD(mask)); hw_atl_itr_irq_status_clearlsw_set(self, LODWORD(mask)); @@ -871,7 +989,7 @@ static int hw_atl_b0_hw_irq_disable(struct aq_hw_s *self, u64 mask) return aq_hw_err_from_flags(self); } -static int hw_atl_b0_hw_irq_read(struct aq_hw_s *self, u64 *mask) +int hw_atl_b0_hw_irq_read(struct aq_hw_s *self, u64 *mask) { *mask = hw_atl_itr_irq_statuslsw_get(self); @@ -880,8 +998,8 @@ static int hw_atl_b0_hw_irq_read(struct aq_hw_s *self, u64 *mask) #define IS_FILTER_ENABLED(_F_) ((packet_filter & (_F_)) ? 1U : 0U) -static int hw_atl_b0_hw_packet_filter_set(struct aq_hw_s *self, - unsigned int packet_filter) +int hw_atl_b0_hw_packet_filter_set(struct aq_hw_s *self, + unsigned int packet_filter) { struct aq_nic_cfg_s *cfg = self->aq_nic_cfg; unsigned int i = 0U; @@ -1071,34 +1189,20 @@ err_exit: return err; } -static int hw_atl_b0_hw_ring_tx_stop(struct aq_hw_s *self, - struct aq_ring_s *ring) +int hw_atl_b0_hw_ring_tx_stop(struct aq_hw_s *self, struct aq_ring_s *ring) { hw_atl_tdm_tx_desc_en_set(self, 0U, ring->idx); return aq_hw_err_from_flags(self); } -static int hw_atl_b0_hw_ring_rx_stop(struct aq_hw_s *self, - struct aq_ring_s *ring) +int hw_atl_b0_hw_ring_rx_stop(struct aq_hw_s *self, struct aq_ring_s *ring) { hw_atl_rdm_rx_desc_en_set(self, 0U, ring->idx); return aq_hw_err_from_flags(self); } -static int hw_atl_b0_tx_tc_mode_get(struct aq_hw_s *self, u32 *tc_mode) -{ - *tc_mode = hw_atl_rpb_tps_tx_tc_mode_get(self); - return aq_hw_err_from_flags(self); -} - -static int hw_atl_b0_rx_tc_mode_get(struct aq_hw_s *self, u32 *tc_mode) -{ - *tc_mode = hw_atl_rpb_rpf_rx_traf_class_mode_get(self); - return aq_hw_err_from_flags(self); -} - #define get_ptp_ts_val_u64(self, indx) \ ((u64)(hw_atl_pcs_ptp_clock_get(self, indx) & 0xffff)) @@ -1478,6 +1582,8 @@ static int hw_atl_b0_set_loopback(struct aq_hw_s *self, u32 mode, bool enable) } const struct aq_hw_ops hw_atl_ops_b0 = { + .hw_soft_reset = hw_atl_utils_soft_reset, + .hw_prepare = hw_atl_utils_initfw, .hw_set_mac_address = hw_atl_b0_hw_mac_addr_set, .hw_init = hw_atl_b0_hw_init, .hw_reset = hw_atl_b0_hw_reset, @@ -1510,13 +1616,11 @@ const struct aq_hw_ops hw_atl_ops_b0 = { .hw_interrupt_moderation_set = hw_atl_b0_hw_interrupt_moderation_set, .hw_rss_set = hw_atl_b0_hw_rss_set, .hw_rss_hash_set = hw_atl_b0_hw_rss_hash_set, + .hw_tc_rate_limit_set = hw_atl_b0_hw_init_tx_tc_rate_limit, .hw_get_regs = hw_atl_utils_hw_get_regs, .hw_get_hw_stats = hw_atl_utils_get_hw_stats, .hw_get_fw_version = hw_atl_utils_get_fw_version, - .hw_tx_tc_mode_get = hw_atl_b0_tx_tc_mode_get, - .hw_rx_tc_mode_get = hw_atl_b0_rx_tc_mode_get, - .hw_ring_hwts_rx_fill = hw_atl_b0_hw_ring_hwts_rx_fill, .hw_ring_hwts_rx_receive = hw_atl_b0_hw_ring_hwts_rx_receive, diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.h index 09af1683034b..30f468f2084d 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.h +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.h @@ -1,7 +1,8 @@ /* SPDX-License-Identifier: GPL-2.0-only */ -/* - * aQuantia Corporation Network Driver - * Copyright (C) 2014-2017 aQuantia Corporation. All rights reserved +/* Atlantic Network Driver + * + * Copyright (C) 2014-2019 aQuantia Corporation + * Copyright (C) 2019-2020 Marvell International Ltd. */ /* File hw_atl_b0.h: Declaration of abstract interface for Atlantic hardware @@ -33,4 +34,41 @@ extern const struct aq_hw_ops hw_atl_ops_b0; #define hw_atl_ops_b1 hw_atl_ops_b0 +int hw_atl_b0_hw_rss_hash_set(struct aq_hw_s *self, + struct aq_rss_parameters *rss_params); +int hw_atl_b0_hw_offload_set(struct aq_hw_s *self, + struct aq_nic_cfg_s *aq_nic_cfg); + +int hw_atl_b0_hw_ring_tx_start(struct aq_hw_s *self, struct aq_ring_s *ring); +int hw_atl_b0_hw_ring_rx_start(struct aq_hw_s *self, struct aq_ring_s *ring); + +int hw_atl_b0_hw_ring_rx_init(struct aq_hw_s *self, struct aq_ring_s *aq_ring, + struct aq_ring_param_s *aq_ring_param); +int hw_atl_b0_hw_ring_rx_fill(struct aq_hw_s *self, struct aq_ring_s *ring, + unsigned int sw_tail_old); +int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self, struct aq_ring_s *ring); + +int hw_atl_b0_hw_ring_tx_init(struct aq_hw_s *self, struct aq_ring_s *aq_ring, + struct aq_ring_param_s *aq_ring_param); +int hw_atl_b0_hw_ring_tx_xmit(struct aq_hw_s *self, struct aq_ring_s *ring, + unsigned int frags); +int hw_atl_b0_hw_ring_tx_head_update(struct aq_hw_s *self, + struct aq_ring_s *ring); + +int hw_atl_b0_hw_ring_tx_stop(struct aq_hw_s *self, struct aq_ring_s *ring); +int hw_atl_b0_hw_ring_rx_stop(struct aq_hw_s *self, struct aq_ring_s *ring); + +void hw_atl_b0_hw_init_rx_rss_ctrl1(struct aq_hw_s *self); + +int hw_atl_b0_hw_mac_addr_set(struct aq_hw_s *self, u8 *mac_addr); + +int hw_atl_b0_hw_start(struct aq_hw_s *self); + +int hw_atl_b0_hw_irq_enable(struct aq_hw_s *self, u64 mask); +int hw_atl_b0_hw_irq_disable(struct aq_hw_s *self, u64 mask); +int hw_atl_b0_hw_irq_read(struct aq_hw_s *self, u64 *mask); + +int hw_atl_b0_hw_packet_filter_set(struct aq_hw_s *self, + unsigned int packet_filter); + #endif /* HW_ATL_B0_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0_internal.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0_internal.h index 7ab23a1751d3..cf460d61a45e 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0_internal.h +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0_internal.h @@ -75,7 +75,7 @@ #define HW_ATL_B0_RSS_HASHKEY_BITS 320U #define HW_ATL_B0_TCRSS_4_8 1 -#define HW_ATL_B0_TC_MAX 1U +#define HW_ATL_B0_TC_MAX 8U #define HW_ATL_B0_RSS_MAX 8U #define HW_ATL_B0_LRO_RXD_MAX 16U @@ -151,6 +151,10 @@ #define HW_ATL_B0_MAX_RXD 8184U #define HW_ATL_B0_MAX_TXD 8184U +#define HW_ATL_RSS_DISABLED 0x00000000U +#define HW_ATL_RSS_ENABLED_8TCS_2INDEX_BITS 0xA2222222U +#define HW_ATL_RSS_ENABLED_4TCS_3INDEX_BITS 0x80003333U + /* HW layer capabilities */ #endif /* HW_ATL_B0_INTERNAL_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c index d1f68fc16291..3c8e8047ea1e 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c @@ -693,6 +693,13 @@ void hw_atl_rpfl2multicast_flr_en_set(struct aq_hw_s *aq_hw, HW_ATL_RPFL2MC_ENF_SHIFT, l2multicast_flr_en); } +u32 hw_atl_rpfl2promiscuous_mode_en_get(struct aq_hw_s *aq_hw) +{ + return aq_hw_read_reg_bit(aq_hw, HW_ATL_RPFL2PROMIS_MODE_ADR, + HW_ATL_RPFL2PROMIS_MODE_MSK, + HW_ATL_RPFL2PROMIS_MODE_SHIFT); +} + void hw_atl_rpfl2promiscuous_mode_en_set(struct aq_hw_s *aq_hw, u32 l2promiscuous_mode_en) { @@ -747,7 +754,7 @@ void hw_atl_rpfl2_accept_all_mc_packets_set(struct aq_hw_s *aq_hw, } void hw_atl_rpf_rpb_user_priority_tc_map_set(struct aq_hw_s *aq_hw, - u32 user_priority_tc_map, u32 tc) + u32 user_priority, u32 tc) { /* register address for bitfield rx_tc_up{t}[2:0] */ static u32 rpf_rpb_rx_tc_upt_adr[8] = { @@ -766,10 +773,9 @@ void hw_atl_rpf_rpb_user_priority_tc_map_set(struct aq_hw_s *aq_hw, 0U, 4U, 8U, 12U, 16U, 20U, 24U, 28U }; - aq_hw_write_reg_bit(aq_hw, rpf_rpb_rx_tc_upt_adr[tc], - rpf_rpb_rx_tc_upt_msk[tc], - rpf_rpb_rx_tc_upt_shft[tc], - user_priority_tc_map); + aq_hw_write_reg_bit(aq_hw, rpf_rpb_rx_tc_upt_adr[user_priority], + rpf_rpb_rx_tc_upt_msk[user_priority], + rpf_rpb_rx_tc_upt_shft[user_priority], tc); } void hw_atl_rpf_rss_key_addr_set(struct aq_hw_s *aq_hw, u32 rss_key_addr) @@ -867,6 +873,13 @@ void hw_atl_rpf_vlan_prom_mode_en_set(struct aq_hw_s *aq_hw, vlan_prom_mode_en); } +u32 hw_atl_rpf_vlan_prom_mode_en_get(struct aq_hw_s *aq_hw) +{ + return aq_hw_read_reg_bit(aq_hw, HW_ATL_RPF_VL_PROMIS_MODE_ADR, + HW_ATL_RPF_VL_PROMIS_MODE_MSK, + HW_ATL_RPF_VL_PROMIS_MODE_SHIFT); +} + void hw_atl_rpf_vlan_accept_untagged_packets_set(struct aq_hw_s *aq_hw, u32 vlan_acc_untagged_packets) { @@ -1304,14 +1317,14 @@ void hw_atl_tpb_tx_buff_en_set(struct aq_hw_s *aq_hw, u32 tx_buff_en) HW_ATL_TPB_TX_BUF_EN_SHIFT, tx_buff_en); } -u32 hw_atl_rpb_tps_tx_tc_mode_get(struct aq_hw_s *aq_hw) +u32 hw_atl_tpb_tps_tx_tc_mode_get(struct aq_hw_s *aq_hw) { return aq_hw_read_reg_bit(aq_hw, HW_ATL_TPB_TX_TC_MODE_ADDR, HW_ATL_TPB_TX_TC_MODE_MSK, HW_ATL_TPB_TX_TC_MODE_SHIFT); } -void hw_atl_rpb_tps_tx_tc_mode_set(struct aq_hw_s *aq_hw, +void hw_atl_tpb_tps_tx_tc_mode_set(struct aq_hw_s *aq_hw, u32 tx_traf_class_mode) { aq_hw_write_reg_bit(aq_hw, HW_ATL_TPB_TX_TC_MODE_ADDR, @@ -1450,8 +1463,8 @@ void hw_atl_tps_tx_pkt_shed_desc_tc_arb_mode_set(struct aq_hw_s *aq_hw, } void hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(struct aq_hw_s *aq_hw, - u32 max_credit, - u32 tc) + const u32 tc, + const u32 max_credit) { aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_TCTCREDIT_MAX_ADR(tc), HW_ATL_TPS_DESC_TCTCREDIT_MAX_MSK, @@ -1460,13 +1473,13 @@ void hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(struct aq_hw_s *aq_hw, } void hw_atl_tps_tx_pkt_shed_desc_tc_weight_set(struct aq_hw_s *aq_hw, - u32 tx_pkt_shed_desc_tc_weight, - u32 tc) + const u32 tc, + const u32 weight) { aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_TCTWEIGHT_ADR(tc), HW_ATL_TPS_DESC_TCTWEIGHT_MSK, HW_ATL_TPS_DESC_TCTWEIGHT_SHIFT, - tx_pkt_shed_desc_tc_weight); + weight); } void hw_atl_tps_tx_pkt_shed_desc_vm_arb_mode_set(struct aq_hw_s *aq_hw, @@ -1479,8 +1492,8 @@ void hw_atl_tps_tx_pkt_shed_desc_vm_arb_mode_set(struct aq_hw_s *aq_hw, } void hw_atl_tps_tx_pkt_shed_tc_data_max_credit_set(struct aq_hw_s *aq_hw, - u32 max_credit, - u32 tc) + const u32 tc, + const u32 max_credit) { aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DATA_TCTCREDIT_MAX_ADR(tc), HW_ATL_TPS_DATA_TCTCREDIT_MAX_MSK, @@ -1489,13 +1502,49 @@ void hw_atl_tps_tx_pkt_shed_tc_data_max_credit_set(struct aq_hw_s *aq_hw, } void hw_atl_tps_tx_pkt_shed_tc_data_weight_set(struct aq_hw_s *aq_hw, - u32 tx_pkt_shed_tc_data_weight, - u32 tc) + const u32 tc, + const u32 weight) { aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DATA_TCTWEIGHT_ADR(tc), HW_ATL_TPS_DATA_TCTWEIGHT_MSK, HW_ATL_TPS_DATA_TCTWEIGHT_SHIFT, - tx_pkt_shed_tc_data_weight); + weight); +} + +void hw_atl_tps_tx_desc_rate_mode_set(struct aq_hw_s *aq_hw, + const u32 rate_mode) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_TX_DESC_RATE_MODE_ADR, + HW_ATL_TPS_TX_DESC_RATE_MODE_MSK, + HW_ATL_TPS_TX_DESC_RATE_MODE_SHIFT, + rate_mode); +} + +void hw_atl_tps_tx_desc_rate_en_set(struct aq_hw_s *aq_hw, const u32 desc, + const u32 enable) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_RATE_EN_ADR(desc), + HW_ATL_TPS_DESC_RATE_EN_MSK, + HW_ATL_TPS_DESC_RATE_EN_SHIFT, + enable); +} + +void hw_atl_tps_tx_desc_rate_x_set(struct aq_hw_s *aq_hw, const u32 desc, + const u32 rate_int) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_RATE_X_ADR(desc), + HW_ATL_TPS_DESC_RATE_X_MSK, + HW_ATL_TPS_DESC_RATE_X_SHIFT, + rate_int); +} + +void hw_atl_tps_tx_desc_rate_y_set(struct aq_hw_s *aq_hw, const u32 desc, + const u32 rate_frac) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL_TPS_DESC_RATE_Y_ADR(desc), + HW_ATL_TPS_DESC_RATE_Y_MSK, + HW_ATL_TPS_DESC_RATE_Y_SHIFT, + rate_frac); } /* tx */ diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h index 62992b23c0e8..61a6f70c51cd 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h @@ -349,6 +349,9 @@ void hw_atl_rpfl2multicast_flr_en_set(struct aq_hw_s *aq_hw, u32 l2multicast_flr_en, u32 filter); +/* get l2 promiscuous mode enable */ +u32 hw_atl_rpfl2promiscuous_mode_en_get(struct aq_hw_s *aq_hw); + /* set l2 promiscuous mode enable */ void hw_atl_rpfl2promiscuous_mode_en_set(struct aq_hw_s *aq_hw, u32 l2promiscuous_mode_en); @@ -420,6 +423,9 @@ void hw_atl_rpf_vlan_outer_etht_set(struct aq_hw_s *aq_hw, u32 vlan_outer_etht); void hw_atl_rpf_vlan_prom_mode_en_set(struct aq_hw_s *aq_hw, u32 vlan_prom_mode_en); +/* Get VLAN promiscuous mode enable */ +u32 hw_atl_rpf_vlan_prom_mode_en_get(struct aq_hw_s *aq_hw); + /* Set VLAN untagged action */ void hw_atl_rpf_vlan_untagged_act_set(struct aq_hw_s *aq_hw, u32 vlan_untagged_act); @@ -610,11 +616,11 @@ void hw_atl_thm_lso_tcp_flag_of_middle_pkt_set(struct aq_hw_s *aq_hw, /* tpb */ /* set TX Traffic Class Mode */ -void hw_atl_rpb_tps_tx_tc_mode_set(struct aq_hw_s *aq_hw, +void hw_atl_tpb_tps_tx_tc_mode_set(struct aq_hw_s *aq_hw, u32 tx_traf_class_mode); /* get TX Traffic Class Mode */ -u32 hw_atl_rpb_tps_tx_tc_mode_get(struct aq_hw_s *aq_hw); +u32 hw_atl_tpb_tps_tx_tc_mode_get(struct aq_hw_s *aq_hw); /* set tx buffer enable */ void hw_atl_tpb_tx_buff_en_set(struct aq_hw_s *aq_hw, u32 tx_buff_en); @@ -682,13 +688,13 @@ void hw_atl_tps_tx_pkt_shed_desc_tc_arb_mode_set(struct aq_hw_s *aq_hw, /* set tx packet scheduler descriptor tc max credit */ void hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(struct aq_hw_s *aq_hw, - u32 max_credit, - u32 tc); + const u32 tc, + const u32 max_credit); /* set tx packet scheduler descriptor tc weight */ void hw_atl_tps_tx_pkt_shed_desc_tc_weight_set(struct aq_hw_s *aq_hw, - u32 tx_pkt_shed_desc_tc_weight, - u32 tc); + const u32 tc, + const u32 weight); /* set tx packet scheduler descriptor vm arbitration mode */ void hw_atl_tps_tx_pkt_shed_desc_vm_arb_mode_set(struct aq_hw_s *aq_hw, @@ -696,13 +702,29 @@ void hw_atl_tps_tx_pkt_shed_desc_vm_arb_mode_set(struct aq_hw_s *aq_hw, /* set tx packet scheduler tc data max credit */ void hw_atl_tps_tx_pkt_shed_tc_data_max_credit_set(struct aq_hw_s *aq_hw, - u32 max_credit, - u32 tc); + const u32 tc, + const u32 max_credit); /* set tx packet scheduler tc data weight */ void hw_atl_tps_tx_pkt_shed_tc_data_weight_set(struct aq_hw_s *aq_hw, - u32 tx_pkt_shed_tc_data_weight, - u32 tc); + const u32 tc, + const u32 weight); + +/* set tx descriptor rate mode */ +void hw_atl_tps_tx_desc_rate_mode_set(struct aq_hw_s *aq_hw, + const u32 rate_mode); + +/* set tx packet scheduler descriptor rate enable */ +void hw_atl_tps_tx_desc_rate_en_set(struct aq_hw_s *aq_hw, const u32 desc, + const u32 enable); + +/* set tx packet scheduler descriptor rate integral value */ +void hw_atl_tps_tx_desc_rate_x_set(struct aq_hw_s *aq_hw, const u32 desc, + const u32 rate_int); + +/* set tx packet scheduler descriptor rate fractional value */ +void hw_atl_tps_tx_desc_rate_y_set(struct aq_hw_s *aq_hw, const u32 desc, + const u32 rate_frac); /* tx */ diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h index 18de2f7b8959..06220792daf1 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h @@ -2038,6 +2038,42 @@ /* default value of bitfield lso_tcp_flag_mid[b:0] */ #define HW_ATL_THM_LSO_TCP_FLAG_MID_DEFAULT 0x0 +/* tx tx_tc_mode bitfield definitions + * preprocessor definitions for the bitfield "tx_tc_mode". + * port="pif_tpb_tx_tc_mode_i,pif_tps_tx_tc_mode_i" + */ + +/* register address for bitfield tx_tc_mode */ +#define HW_ATL_TPB_TX_TC_MODE_ADDR 0x00007900 +/* bitmask for bitfield tx_tc_mode */ +#define HW_ATL_TPB_TX_TC_MODE_MSK 0x00000100 +/* inverted bitmask for bitfield tx_tc_mode */ +#define HW_ATL_TPB_TX_TC_MODE_MSKN 0xFFFFFEFF +/* lower bit position of bitfield tx_tc_mode */ +#define HW_ATL_TPB_TX_TC_MODE_SHIFT 8 +/* width of bitfield tx_tc_mode */ +#define HW_ATL_TPB_TX_TC_MODE_WIDTH 1 +/* default value of bitfield tx_tc_mode */ +#define HW_ATL_TPB_TX_TC_MODE_DEFAULT 0x0 + +/* tx tx_desc_rate_mode bitfield definitions + * preprocessor definitions for the bitfield "tx_desc_rate_mode". + * port="pif_tps_desc_rate_mode_i" + */ + +/* register address for bitfield tx_desc_rate_mode */ +#define HW_ATL_TPS_TX_DESC_RATE_MODE_ADR 0x00007900 +/* bitmask for bitfield tx_desc_rate_mode */ +#define HW_ATL_TPS_TX_DESC_RATE_MODE_MSK 0x00000080 +/* inverted bitmask for bitfield tx_desc_rate_mode */ +#define HW_ATL_TPS_TX_DESC_RATE_MODE_MSKN 0xFFFFFF7F +/* lower bit position of bitfield tx_desc_rate_mode */ +#define HW_ATL_TPS_TX_DESC_RATE_MODE_SHIFT 7 +/* width of bitfield tx_desc_rate_mode */ +#define HW_ATL_TPS_TX_DESC_RATE_MODE_WIDTH 1 +/* default value of bitfield tx_desc_rate_mode */ +#define HW_ATL_TPS_TX_DESC_RATE_MODE_DEFAULT 0x0 + /* tx tx_buf_en bitfield definitions * preprocessor definitions for the bitfield "tx_buf_en". * port="pif_tpb_tx_buf_en_i" @@ -2056,19 +2092,6 @@ /* default value of bitfield tx_buf_en */ #define HW_ATL_TPB_TX_BUF_EN_DEFAULT 0x0 -/* register address for bitfield tx_tc_mode */ -#define HW_ATL_TPB_TX_TC_MODE_ADDR 0x00007900 -/* bitmask for bitfield tx_tc_mode */ -#define HW_ATL_TPB_TX_TC_MODE_MSK 0x00000100 -/* inverted bitmask for bitfield tx_tc_mode */ -#define HW_ATL_TPB_TX_TC_MODE_MSKN 0xFFFFFEFF -/* lower bit position of bitfield tx_tc_mode */ -#define HW_ATL_TPB_TX_TC_MODE_SHIFT 8 -/* width of bitfield tx_tc_mode */ -#define HW_ATL_TPB_TX_TC_MODE_WIDTH 1 -/* default value of bitfield tx_tc_mode */ -#define HW_ATL_TPB_TX_TC_MODE_DEFAULT 0x0 - /* tx tx{b}_hi_thresh[c:0] bitfield definitions * preprocessor definitions for the bitfield "tx{b}_hi_thresh[c:0]". * parameter: buffer {b} | stride size 0x10 | range [0, 7] @@ -2270,6 +2293,58 @@ /* default value of bitfield data_tc_arb_mode */ #define HW_ATL_TPS_DATA_TC_ARB_MODE_DEFAULT 0x0 +/* tx desc{r}_rate_en bitfield definitions + * preprocessor definitions for the bitfield "desc{r}_rate_en". + * port="pif_tps_desc_rate_en_i[0]" + */ + +/* register address for bitfield desc{r}_rate_en */ +#define HW_ATL_TPS_DESC_RATE_EN_ADR(desc) (0x00007408 + (desc) * 0x10) +/* bitmask for bitfield desc{r}_rate_en */ +#define HW_ATL_TPS_DESC_RATE_EN_MSK 0x80000000 +/* inverted bitmask for bitfield desc{r}_rate_en */ +#define HW_ATL_TPS_DESC_RATE_EN_MSKN 0x7FFFFFFF +/* lower bit position of bitfield desc{r}_rate_en */ +#define HW_ATL_TPS_DESC_RATE_EN_SHIFT 31 +/* width of bitfield desc{r}_rate_en */ +#define HW_ATL_TPS_DESC_RATE_EN_WIDTH 1 +/* default value of bitfield desc{r}_rate_en */ +#define HW_ATL_TPS_DESC_RATE_EN_DEFAULT 0x0 + +/* tx desc{r}_rate_x bitfield definitions + * preprocessor definitions for the bitfield "desc{r}_rate_x". + * port="pif_tps_desc0_rate_x" + */ +/* register address for bitfield desc{r}_rate_x */ +#define HW_ATL_TPS_DESC_RATE_X_ADR(desc) (0x00007408 + (desc) * 0x10) +/* bitmask for bitfield desc{r}_rate_x */ +#define HW_ATL_TPS_DESC_RATE_X_MSK 0x03FF0000 +/* inverted bitmask for bitfield desc{r}_rate_x */ +#define HW_ATL_TPS_DESC_RATE_X_MSKN 0xFC00FFFF +/* lower bit position of bitfield desc{r}_rate_x */ +#define HW_ATL_TPS_DESC_RATE_X_SHIFT 16 +/* width of bitfield desc{r}_rate_x */ +#define HW_ATL_TPS_DESC_RATE_X_WIDTH 10 +/* default value of bitfield desc{r}_rate_x */ +#define HW_ATL_TPS_DESC_RATE_X_DEFAULT 0x0 + +/* tx desc{r}_rate_y bitfield definitions + * preprocessor definitions for the bitfield "desc{r}_rate_y". + * port="pif_tps_desc0_rate_y" + */ +/* register address for bitfield desc{r}_rate_y */ +#define HW_ATL_TPS_DESC_RATE_Y_ADR(desc) (0x00007408 + (desc) * 0x10) +/* bitmask for bitfield desc{r}_rate_y */ +#define HW_ATL_TPS_DESC_RATE_Y_MSK 0x00003FFF +/* inverted bitmask for bitfield desc{r}_rate_y */ +#define HW_ATL_TPS_DESC_RATE_Y_MSKN 0xFFFFC000 +/* lower bit position of bitfield desc{r}_rate_y */ +#define HW_ATL_TPS_DESC_RATE_Y_SHIFT 0 +/* width of bitfield desc{r}_rate_y */ +#define HW_ATL_TPS_DESC_RATE_Y_WIDTH 14 +/* default value of bitfield desc{r}_rate_y */ +#define HW_ATL_TPS_DESC_RATE_Y_DEFAULT 0x0 + /* tx desc_rate_ta_rst bitfield definitions * preprocessor definitions for the bitfield "desc_rate_ta_rst". * port="pif_tps_desc_rate_ta_rst_i" diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c index 354705f9bc49..73c0f41df8d8 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.c @@ -1,7 +1,8 @@ // SPDX-License-Identifier: GPL-2.0-only -/* - * aQuantia Corporation Network Driver - * Copyright (C) 2014-2019 aQuantia Corporation. All rights reserved +/* Atlantic Network Driver + * + * Copyright (C) 2014-2019 aQuantia Corporation + * Copyright (C) 2019-2020 Marvell International Ltd. */ /* File hw_atl_utils.c: Definition of common functions for Atlantic hardware @@ -53,7 +54,6 @@ enum mcp_area { MCP_AREA_SETTINGS = 0x20000000, }; -static int hw_atl_utils_ver_match(u32 ver_expected, u32 ver_actual); static int hw_atl_utils_mpi_set_state(struct aq_hw_s *self, enum hal_atl_utils_fw_state_e state); static u32 hw_atl_utils_get_mpi_mbox_tid(struct aq_hw_s *self); @@ -67,14 +67,10 @@ int hw_atl_utils_initfw(struct aq_hw_s *self, const struct aq_fw_ops **fw_ops) { int err = 0; - err = hw_atl_utils_soft_reset(self); - if (err) - return err; - hw_atl_utils_hw_chip_features_init(self, &self->chip_features); - hw_atl_utils_get_fw_version(self, &self->fw_ver_actual); + self->fw_ver_actual = hw_atl_utils_get_fw_version(self); if (hw_atl_utils_ver_match(HW_ATL_FW_VER_1X, self->fw_ver_actual) == 0) { @@ -313,7 +309,7 @@ int hw_atl_utils_fw_downld_dwords(struct aq_hw_s *self, u32 a, for (++cnt; --cnt && !err;) { aq_hw_write_reg(self, HW_ATL_MIF_CMD, 0x00008000U); - if (IS_CHIP_FEATURE(REVISION_B1)) + if (ATL_HW_IS_CHIP_FEATURE(self, REVISION_B1)) err = readx_poll_timeout_atomic(hw_atl_utils_mif_addr_get, self, val, val != a, 1U, 1000U); @@ -409,7 +405,7 @@ static int hw_atl_utils_fw_upload_dwords(struct aq_hw_s *self, u32 addr, u32 *p, if (err < 0) goto err_exit; - if (IS_CHIP_FEATURE(REVISION_B1)) + if (ATL_HW_IS_CHIP_FEATURE(self, REVISION_B1)) err = hw_atl_utils_write_b1_mbox(self, addr, p, cnt, area); else err = hw_atl_utils_write_b0_mbox(self, addr, p, cnt); @@ -438,7 +434,7 @@ int hw_atl_write_fwsettings_dwords(struct aq_hw_s *self, u32 offset, u32 *p, p, cnt, MCP_AREA_SETTINGS); } -static int hw_atl_utils_ver_match(u32 ver_expected, u32 ver_actual) +int hw_atl_utils_ver_match(u32 ver_expected, u32 ver_actual) { const u32 dw_major_mask = 0xff000000U; const u32 dw_minor_mask = 0x00ffffffU; @@ -501,7 +497,7 @@ int hw_atl_utils_fw_rpc_call(struct aq_hw_s *self, unsigned int rpc_size) struct aq_hw_atl_utils_fw_rpc_tid_s sw; int err = 0; - if (!IS_CHIP_FEATURE(MIPS)) { + if (!ATL_HW_IS_CHIP_FEATURE(self, MIPS)) { err = -1; goto err_exit; } @@ -607,7 +603,7 @@ void hw_atl_utils_mpi_read_stats(struct aq_hw_s *self, if (err < 0) goto err_exit; - if (IS_CHIP_FEATURE(REVISION_A0)) { + if (ATL_HW_IS_CHIP_FEATURE(self, REVISION_A0)) { unsigned int mtu = self->aq_nic_cfg ? self->aq_nic_cfg->mtu : 1514U; pmbox->stats.ubrc = pmbox->stats.uprc * mtu; @@ -692,7 +688,7 @@ int hw_atl_utils_mpi_get_link_status(struct aq_hw_s *self) link_status->mbps = 5000U; break; - case HAL_ATLANTIC_RATE_2GS: + case HAL_ATLANTIC_RATE_2G5: link_status->mbps = 2500U; break; @@ -806,22 +802,24 @@ void hw_atl_utils_hw_chip_features_init(struct aq_hw_s *self, u32 *p) u32 mif_rev = val & 0xFFU; u32 chip_features = 0U; + chip_features |= ATL_HW_CHIP_ATLANTIC; + if ((0xFU & mif_rev) == 1U) { - chip_features |= HAL_ATLANTIC_UTILS_CHIP_REVISION_A0 | - HAL_ATLANTIC_UTILS_CHIP_MPI_AQ | - HAL_ATLANTIC_UTILS_CHIP_MIPS; + chip_features |= ATL_HW_CHIP_REVISION_A0 | + ATL_HW_CHIP_MPI_AQ | + ATL_HW_CHIP_MIPS; } else if ((0xFU & mif_rev) == 2U) { - chip_features |= HAL_ATLANTIC_UTILS_CHIP_REVISION_B0 | - HAL_ATLANTIC_UTILS_CHIP_MPI_AQ | - HAL_ATLANTIC_UTILS_CHIP_MIPS | - HAL_ATLANTIC_UTILS_CHIP_TPO2 | - HAL_ATLANTIC_UTILS_CHIP_RPF2; + chip_features |= ATL_HW_CHIP_REVISION_B0 | + ATL_HW_CHIP_MPI_AQ | + ATL_HW_CHIP_MIPS | + ATL_HW_CHIP_TPO2 | + ATL_HW_CHIP_RPF2; } else if ((0xFU & mif_rev) == 0xAU) { - chip_features |= HAL_ATLANTIC_UTILS_CHIP_REVISION_B1 | - HAL_ATLANTIC_UTILS_CHIP_MPI_AQ | - HAL_ATLANTIC_UTILS_CHIP_MIPS | - HAL_ATLANTIC_UTILS_CHIP_TPO2 | - HAL_ATLANTIC_UTILS_CHIP_RPF2; + chip_features |= ATL_HW_CHIP_REVISION_B1 | + ATL_HW_CHIP_MPI_AQ | + ATL_HW_CHIP_MIPS | + ATL_HW_CHIP_TPO2 | + ATL_HW_CHIP_RPF2; } *p = chip_features; @@ -919,11 +917,9 @@ int hw_atl_utils_hw_get_regs(struct aq_hw_s *self, return 0; } -int hw_atl_utils_get_fw_version(struct aq_hw_s *self, u32 *fw_version) +u32 hw_atl_utils_get_fw_version(struct aq_hw_s *self) { - *fw_version = aq_hw_read_reg(self, 0x18U); - - return 0; + return aq_hw_read_reg(self, HW_ATL_MPI_FW_VERSION); } static int aq_fw1x_set_wake_magic(struct aq_hw_s *self, bool wol_enabled, diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h index b15513914636..0b4b54fc1de0 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils.h @@ -1,7 +1,8 @@ /* SPDX-License-Identifier: GPL-2.0-only */ -/* - * aQuantia Corporation Network Driver - * Copyright (C) 2014-2019 aQuantia Corporation. All rights reserved +/* Atlantic Network Driver + * + * Copyright (C) 2014-2019 aQuantia Corporation + * Copyright (C) 2019-2020 Marvell International Ltd. */ /* File hw_atl_utils.h: Declaration of common functions for Atlantic hardware @@ -360,6 +361,8 @@ struct aq_rx_filter_vlan { u8 queue; }; +#define HW_ATL_VLAN_MAX_FILTERS 16U + struct aq_rx_filter_l2 { s8 queue; u8 location; @@ -406,17 +409,6 @@ enum hw_atl_rx_ctrl_registers_l3l4 { #define HW_ATL_GET_REG_LOCATION_FL3L4(location) \ ((location) - AQ_RX_FIRST_LOC_FL3L4) -#define HAL_ATLANTIC_UTILS_CHIP_MIPS 0x00000001U -#define HAL_ATLANTIC_UTILS_CHIP_TPO2 0x00000002U -#define HAL_ATLANTIC_UTILS_CHIP_RPF2 0x00000004U -#define HAL_ATLANTIC_UTILS_CHIP_MPI_AQ 0x00000010U -#define HAL_ATLANTIC_UTILS_CHIP_REVISION_A0 0x01000000U -#define HAL_ATLANTIC_UTILS_CHIP_REVISION_B0 0x02000000U -#define HAL_ATLANTIC_UTILS_CHIP_REVISION_B1 0x04000000U - -#define IS_CHIP_FEATURE(_F_) (HAL_ATLANTIC_UTILS_CHIP_##_F_ & \ - self->chip_features) - enum hal_atl_utils_fw_state_e { MPI_DEINIT = 0, MPI_RESET = 1, @@ -427,7 +419,7 @@ enum hal_atl_utils_fw_state_e { #define HAL_ATLANTIC_RATE_10G BIT(0) #define HAL_ATLANTIC_RATE_5G BIT(1) #define HAL_ATLANTIC_RATE_5GSR BIT(2) -#define HAL_ATLANTIC_RATE_2GS BIT(3) +#define HAL_ATLANTIC_RATE_2G5 BIT(3) #define HAL_ATLANTIC_RATE_1G BIT(4) #define HAL_ATLANTIC_RATE_100M BIT(5) #define HAL_ATLANTIC_RATE_INVALID BIT(6) @@ -622,7 +614,7 @@ int hw_atl_utils_hw_set_power(struct aq_hw_s *self, int hw_atl_utils_hw_deinit(struct aq_hw_s *self); -int hw_atl_utils_get_fw_version(struct aq_hw_s *self, u32 *fw_version); +u32 hw_atl_utils_get_fw_version(struct aq_hw_s *self); int hw_atl_utils_update_stats(struct aq_hw_s *self); @@ -643,6 +635,8 @@ int hw_atl_utils_fw_rpc_call(struct aq_hw_s *self, unsigned int rpc_size); int hw_atl_utils_fw_rpc_wait(struct aq_hw_s *self, struct hw_atl_utils_fw_rpc **rpc); +int hw_atl_utils_ver_match(u32 ver_expected, u32 ver_actual); + extern const struct aq_fw_ops aq_fw_1x_ops; extern const struct aq_fw_ops aq_fw_2x_ops; diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c index 1ad10cc14918..eeedd8c90067 100644 --- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c @@ -1,7 +1,8 @@ // SPDX-License-Identifier: GPL-2.0-only -/* - * aQuantia Corporation Network Driver - * Copyright (C) 2014-2019 aQuantia Corporation. All rights reserved +/* Atlantic Network Driver + * + * Copyright (C) 2014-2019 aQuantia Corporation + * Copyright (C) 2019-2020 Marvell International Ltd. */ /* File hw_atl_utils_fw2x.c: Definition of firmware 2.x functions for @@ -134,7 +135,7 @@ static enum hw_atl_fw2x_rate link_speed_mask_2fw2x_ratemask(u32 speed) if (speed & AQ_NIC_RATE_5GSR) rate |= FW2X_RATE_5G; - if (speed & AQ_NIC_RATE_2GS) + if (speed & AQ_NIC_RATE_2G5) rate |= FW2X_RATE_2G5; if (speed & AQ_NIC_RATE_1G) @@ -155,7 +156,7 @@ static u32 fw2x_to_eee_mask(u32 speed) if (speed & HW_ATL_FW2X_CAP_EEE_5G_MASK) rate |= AQ_NIC_RATE_EEE_5G; if (speed & HW_ATL_FW2X_CAP_EEE_2G5_MASK) - rate |= AQ_NIC_RATE_EEE_2GS; + rate |= AQ_NIC_RATE_EEE_2G5; if (speed & HW_ATL_FW2X_CAP_EEE_1G_MASK) rate |= AQ_NIC_RATE_EEE_1G; @@ -170,7 +171,7 @@ static u32 eee_mask_to_fw2x(u32 speed) rate |= HW_ATL_FW2X_CAP_EEE_10G_MASK; if (speed & AQ_NIC_RATE_EEE_5G) rate |= HW_ATL_FW2X_CAP_EEE_5G_MASK; - if (speed & AQ_NIC_RATE_EEE_2GS) + if (speed & AQ_NIC_RATE_EEE_2G5) rate |= HW_ATL_FW2X_CAP_EEE_2G5_MASK; if (speed & AQ_NIC_RATE_EEE_1G) rate |= HW_ATL_FW2X_CAP_EEE_1G_MASK; @@ -282,8 +283,6 @@ static int aq_fw2x_get_mac_permanent(struct aq_hw_s *self, u8 *mac) u32 efuse_addr = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_EFUSE_ADDR); u32 mac_addr[2] = { 0 }; int err = 0; - u32 h = 0U; - u32 l = 0U; if (efuse_addr != 0) { err = hw_atl_utils_fw_downld_dwords(self, @@ -298,26 +297,6 @@ static int aq_fw2x_get_mac_permanent(struct aq_hw_s *self, u8 *mac) ether_addr_copy(mac, (u8 *)mac_addr); - if ((mac[0] & 0x01U) || ((mac[0] | mac[1] | mac[2]) == 0x00U)) { - unsigned int rnd = 0; - - get_random_bytes(&rnd, sizeof(unsigned int)); - - l = 0xE3000000U | (0xFFFFU & rnd) | (0x00 << 16); - h = 0x8001300EU; - - mac[5] = (u8)(0xFFU & l); - l >>= 8; - mac[4] = (u8)(0xFFU & l); - l >>= 8; - mac[3] = (u8)(0xFFU & l); - l >>= 8; - mac[2] = (u8)(0xFFU & l); - mac[1] = (u8)(0xFFU & h); - h >>= 8; - mac[0] = (u8)(0xFFU & h); - } - return err; } diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2.c new file mode 100644 index 000000000000..8df9d4ef36f0 --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2.c @@ -0,0 +1,841 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#include "aq_hw.h" +#include "aq_hw_utils.h" +#include "aq_ring.h" +#include "aq_nic.h" +#include "hw_atl/hw_atl_b0.h" +#include "hw_atl/hw_atl_utils.h" +#include "hw_atl/hw_atl_llh.h" +#include "hw_atl/hw_atl_llh_internal.h" +#include "hw_atl2_utils.h" +#include "hw_atl2_llh.h" +#include "hw_atl2_internal.h" +#include "hw_atl2_llh_internal.h" + +static int hw_atl2_act_rslvr_table_set(struct aq_hw_s *self, u8 location, + u32 tag, u32 mask, u32 action); + +#define DEFAULT_BOARD_BASIC_CAPABILITIES \ + .is_64_dma = true, \ + .msix_irqs = 8U, \ + .irq_mask = ~0U, \ + .vecs = HW_ATL2_RSS_MAX, \ + .tcs_max = HW_ATL2_TC_MAX, \ + .rxd_alignment = 1U, \ + .rxd_size = HW_ATL2_RXD_SIZE, \ + .rxds_max = HW_ATL2_MAX_RXD, \ + .rxds_min = HW_ATL2_MIN_RXD, \ + .txd_alignment = 1U, \ + .txd_size = HW_ATL2_TXD_SIZE, \ + .txds_max = HW_ATL2_MAX_TXD, \ + .txds_min = HW_ATL2_MIN_TXD, \ + .txhwb_alignment = 4096U, \ + .tx_rings = HW_ATL2_TX_RINGS, \ + .rx_rings = HW_ATL2_RX_RINGS, \ + .hw_features = NETIF_F_HW_CSUM | \ + NETIF_F_RXCSUM | \ + NETIF_F_RXHASH | \ + NETIF_F_SG | \ + NETIF_F_TSO | \ + NETIF_F_TSO6 | \ + NETIF_F_LRO | \ + NETIF_F_NTUPLE | \ + NETIF_F_HW_VLAN_CTAG_FILTER | \ + NETIF_F_HW_VLAN_CTAG_RX | \ + NETIF_F_HW_VLAN_CTAG_TX | \ + NETIF_F_GSO_UDP_L4 | \ + NETIF_F_GSO_PARTIAL | \ + NETIF_F_HW_TC, \ + .hw_priv_flags = IFF_UNICAST_FLT, \ + .flow_control = true, \ + .mtu = HW_ATL2_MTU_JUMBO, \ + .mac_regs_count = 72, \ + .hw_alive_check_addr = 0x10U, \ + .priv_data_len = sizeof(struct hw_atl2_priv) + +const struct aq_hw_caps_s hw_atl2_caps_aqc113 = { + DEFAULT_BOARD_BASIC_CAPABILITIES, + .media_type = AQ_HW_MEDIA_TYPE_TP, + .link_speed_msk = AQ_NIC_RATE_10G | + AQ_NIC_RATE_5G | + AQ_NIC_RATE_2G5 | + AQ_NIC_RATE_1G | + AQ_NIC_RATE_100M | + AQ_NIC_RATE_10M, +}; + +static u32 hw_atl2_sem_act_rslvr_get(struct aq_hw_s *self) +{ + return hw_atl_reg_glb_cpu_sem_get(self, HW_ATL2_FW_SM_ACT_RSLVR); +} + +static int hw_atl2_hw_reset(struct aq_hw_s *self) +{ + struct hw_atl2_priv *priv = (struct hw_atl2_priv *)self->priv; + int err; + + err = hw_atl2_utils_soft_reset(self); + if (err) + return err; + + memset(priv, 0, sizeof(*priv)); + + self->aq_fw_ops->set_state(self, MPI_RESET); + + err = aq_hw_err_from_flags(self); + + return err; +} + +static int hw_atl2_hw_queue_to_tc_map_set(struct aq_hw_s *self) +{ + struct aq_nic_cfg_s *cfg = self->aq_nic_cfg; + unsigned int tcs, q_per_tc; + unsigned int tc, q; + u32 rx_map = 0; + u32 tx_map = 0; + + hw_atl2_tpb_tx_tc_q_rand_map_en_set(self, 1U); + + switch (cfg->tc_mode) { + case AQ_TC_MODE_8TCS: + tcs = 8; + q_per_tc = 4; + break; + case AQ_TC_MODE_4TCS: + tcs = 4; + q_per_tc = 8; + break; + default: + return -EINVAL; + } + + for (tc = 0; tc != tcs; tc++) { + unsigned int tc_q_offset = tc * q_per_tc; + + for (q = tc_q_offset; q != tc_q_offset + q_per_tc; q++) { + rx_map |= tc << HW_ATL2_RX_Q_TC_MAP_SHIFT(q); + if (HW_ATL2_RX_Q_TC_MAP_ADR(q) != + HW_ATL2_RX_Q_TC_MAP_ADR(q + 1)) { + aq_hw_write_reg(self, + HW_ATL2_RX_Q_TC_MAP_ADR(q), + rx_map); + rx_map = 0; + } + + tx_map |= tc << HW_ATL2_TX_Q_TC_MAP_SHIFT(q); + if (HW_ATL2_TX_Q_TC_MAP_ADR(q) != + HW_ATL2_TX_Q_TC_MAP_ADR(q + 1)) { + aq_hw_write_reg(self, + HW_ATL2_TX_Q_TC_MAP_ADR(q), + tx_map); + tx_map = 0; + } + } + } + + return aq_hw_err_from_flags(self); +} + +static int hw_atl2_hw_qos_set(struct aq_hw_s *self) +{ + struct aq_nic_cfg_s *cfg = self->aq_nic_cfg; + u32 tx_buff_size = HW_ATL2_TXBUF_MAX; + u32 rx_buff_size = HW_ATL2_RXBUF_MAX; + unsigned int prio = 0U; + u32 tc = 0U; + + /* TPS Descriptor rate init */ + hw_atl_tps_tx_pkt_shed_desc_rate_curr_time_res_set(self, 0x0U); + hw_atl_tps_tx_pkt_shed_desc_rate_lim_set(self, 0xA); + + /* TPS VM init */ + hw_atl_tps_tx_pkt_shed_desc_vm_arb_mode_set(self, 0U); + + tx_buff_size /= cfg->tcs; + rx_buff_size /= cfg->tcs; + for (tc = 0; tc < cfg->tcs; tc++) { + u32 threshold = 0U; + + /* Tx buf size TC0 */ + hw_atl_tpb_tx_pkt_buff_size_per_tc_set(self, tx_buff_size, tc); + + threshold = (tx_buff_size * (1024 / 32U) * 66U) / 100U; + hw_atl_tpb_tx_buff_hi_threshold_per_tc_set(self, threshold, tc); + + threshold = (tx_buff_size * (1024 / 32U) * 50U) / 100U; + hw_atl_tpb_tx_buff_lo_threshold_per_tc_set(self, threshold, tc); + + /* QoS Rx buf size per TC */ + hw_atl_rpb_rx_pkt_buff_size_per_tc_set(self, rx_buff_size, tc); + + threshold = (rx_buff_size * (1024U / 32U) * 66U) / 100U; + hw_atl_rpb_rx_buff_hi_threshold_per_tc_set(self, threshold, tc); + + threshold = (rx_buff_size * (1024U / 32U) * 50U) / 100U; + hw_atl_rpb_rx_buff_lo_threshold_per_tc_set(self, threshold, tc); + } + + /* QoS 802.1p priority -> TC mapping */ + for (prio = 0; prio < 8; ++prio) + hw_atl_rpf_rpb_user_priority_tc_map_set(self, prio, + cfg->prio_tc_map[prio]); + + /* ATL2 Apply ring to TC mapping */ + hw_atl2_hw_queue_to_tc_map_set(self); + + return aq_hw_err_from_flags(self); +} + +static int hw_atl2_hw_rss_set(struct aq_hw_s *self, + struct aq_rss_parameters *rss_params) +{ + u8 *indirection_table = rss_params->indirection_table; + const u32 num_tcs = aq_hw_num_tcs(self); + u32 rpf_redir2_enable; + int tc; + int i; + + rpf_redir2_enable = num_tcs > 4 ? 1 : 0; + + hw_atl2_rpf_redirection_table2_select_set(self, rpf_redir2_enable); + + for (i = HW_ATL2_RSS_REDIRECTION_MAX; i--;) { + for (tc = 0; tc != num_tcs; tc++) { + hw_atl2_new_rpf_rss_redir_set(self, tc, i, + tc * + aq_hw_q_per_tc(self) + + indirection_table[i]); + } + } + + return aq_hw_err_from_flags(self); +} + +static int hw_atl2_hw_init_tx_tc_rate_limit(struct aq_hw_s *self) +{ + static const u32 max_weight = BIT(HW_ATL2_TPS_DATA_TCTWEIGHT_WIDTH) - 1; + /* Scale factor is based on the number of bits in fractional portion */ + static const u32 scale = BIT(HW_ATL_TPS_DESC_RATE_Y_WIDTH); + static const u32 frac_msk = HW_ATL_TPS_DESC_RATE_Y_MSK >> + HW_ATL_TPS_DESC_RATE_Y_SHIFT; + const u32 link_speed = self->aq_link_status.mbps; + struct aq_nic_cfg_s *nic_cfg = self->aq_nic_cfg; + unsigned long num_min_rated_tcs = 0; + u32 tc_weight[AQ_CFG_TCS_MAX]; + u32 fixed_max_credit_4b; + u32 fixed_max_credit; + u8 min_rate_msk = 0; + u32 sum_weight = 0; + int tc; + + /* By default max_credit is based upon MTU (in unit of 64b) */ + fixed_max_credit = nic_cfg->aq_hw_caps->mtu / 64; + /* in unit of 4b */ + fixed_max_credit_4b = nic_cfg->aq_hw_caps->mtu / 4; + + if (link_speed) { + min_rate_msk = nic_cfg->tc_min_rate_msk & + (BIT(nic_cfg->tcs) - 1); + num_min_rated_tcs = hweight8(min_rate_msk); + } + + /* First, calculate weights where min_rate is specified */ + if (num_min_rated_tcs) { + for (tc = 0; tc != nic_cfg->tcs; tc++) { + if (!nic_cfg->tc_min_rate[tc]) { + tc_weight[tc] = 0; + continue; + } + + tc_weight[tc] = (-1L + link_speed + + nic_cfg->tc_min_rate[tc] * + max_weight) / + link_speed; + tc_weight[tc] = min(tc_weight[tc], max_weight); + sum_weight += tc_weight[tc]; + } + } + + /* WSP, if min_rate is set for at least one TC. + * RR otherwise. + */ + hw_atl2_tps_tx_pkt_shed_data_arb_mode_set(self, min_rate_msk ? 1U : 0U); + /* Data TC Arbiter takes precedence over Descriptor TC Arbiter, + * leave Descriptor TC Arbiter as RR. + */ + hw_atl_tps_tx_pkt_shed_desc_tc_arb_mode_set(self, 0U); + + hw_atl_tps_tx_desc_rate_mode_set(self, nic_cfg->is_qos ? 1U : 0U); + + for (tc = 0; tc != nic_cfg->tcs; tc++) { + const u32 en = (nic_cfg->tc_max_rate[tc] != 0) ? 1U : 0U; + const u32 desc = AQ_NIC_CFG_TCVEC2RING(nic_cfg, tc, 0); + u32 weight, max_credit; + + hw_atl_tps_tx_pkt_shed_desc_tc_max_credit_set(self, tc, + fixed_max_credit); + hw_atl_tps_tx_pkt_shed_desc_tc_weight_set(self, tc, 0x1E); + + if (num_min_rated_tcs) { + weight = tc_weight[tc]; + + if (!weight && sum_weight < max_weight) + weight = (max_weight - sum_weight) / + (nic_cfg->tcs - num_min_rated_tcs); + else if (!weight) + weight = 0x640; + + max_credit = max(2 * weight, fixed_max_credit_4b); + } else { + weight = 0x640; + max_credit = 0xFFF0; + } + + hw_atl2_tps_tx_pkt_shed_tc_data_weight_set(self, tc, weight); + hw_atl2_tps_tx_pkt_shed_tc_data_max_credit_set(self, tc, + max_credit); + + hw_atl_tps_tx_desc_rate_en_set(self, desc, en); + + if (en) { + /* Nominal rate is always 10G */ + const u32 rate = 10000U * scale / + nic_cfg->tc_max_rate[tc]; + const u32 rate_int = rate >> + HW_ATL_TPS_DESC_RATE_Y_WIDTH; + const u32 rate_frac = rate & frac_msk; + + hw_atl_tps_tx_desc_rate_x_set(self, desc, rate_int); + hw_atl_tps_tx_desc_rate_y_set(self, desc, rate_frac); + } else { + /* A value of 1 indicates the queue is not + * rate controlled. + */ + hw_atl_tps_tx_desc_rate_x_set(self, desc, 1U); + hw_atl_tps_tx_desc_rate_y_set(self, desc, 0U); + } + } + for (tc = nic_cfg->tcs; tc != AQ_CFG_TCS_MAX; tc++) { + const u32 desc = AQ_NIC_CFG_TCVEC2RING(nic_cfg, tc, 0); + + hw_atl_tps_tx_desc_rate_en_set(self, desc, 0U); + hw_atl_tps_tx_desc_rate_x_set(self, desc, 1U); + hw_atl_tps_tx_desc_rate_y_set(self, desc, 0U); + } + + return aq_hw_err_from_flags(self); +} + +static int hw_atl2_hw_init_tx_path(struct aq_hw_s *self) +{ + struct aq_nic_cfg_s *nic_cfg = self->aq_nic_cfg; + + /* Tx TC/RSS number config */ + hw_atl_tpb_tps_tx_tc_mode_set(self, nic_cfg->tc_mode); + + hw_atl_thm_lso_tcp_flag_of_first_pkt_set(self, 0x0FF6U); + hw_atl_thm_lso_tcp_flag_of_middle_pkt_set(self, 0x0FF6U); + hw_atl_thm_lso_tcp_flag_of_last_pkt_set(self, 0x0F7FU); + + /* Tx interrupts */ + hw_atl_tdm_tx_desc_wr_wb_irq_en_set(self, 1U); + + /* misc */ + hw_atl_tdm_tx_dca_en_set(self, 0U); + hw_atl_tdm_tx_dca_mode_set(self, 0U); + + hw_atl_tpb_tx_path_scp_ins_en_set(self, 1U); + + hw_atl2_tpb_tx_buf_clk_gate_en_set(self, 0U); + + return aq_hw_err_from_flags(self); +} + +static void hw_atl2_hw_init_new_rx_filters(struct aq_hw_s *self) +{ + struct hw_atl2_priv *priv = (struct hw_atl2_priv *)self->priv; + u8 *prio_tc_map = self->aq_nic_cfg->prio_tc_map; + u16 action; + u8 index; + int i; + + /* Action Resolver Table (ART) is used by RPF to decide which action + * to take with a packet based upon input tag and tag mask, where: + * - input tag is a combination of 3-bit VLan Prio (PTP) and + * 29-bit concatenation of all tags from filter block; + * - tag mask is a mask used for matching against input tag. + * The input_tag is compared with the all the Requested_tags in the + * Record table to find a match. Action field of the selected matched + * REC entry is used for further processing. If multiple entries match, + * the lowest REC entry, Action field will be selected. + */ + hw_atl2_rpf_act_rslvr_section_en_set(self, 0xFFFF); + hw_atl2_rpfl2_uc_flr_tag_set(self, HW_ATL2_RPF_TAG_BASE_UC, + HW_ATL2_MAC_UC); + hw_atl2_rpfl2_bc_flr_tag_set(self, HW_ATL2_RPF_TAG_BASE_UC); + + /* FW reserves the beginning of ART, thus all driver entries must + * start from the offset specified in FW caps. + */ + index = priv->art_base_index + HW_ATL2_RPF_L2_PROMISC_OFF_INDEX; + hw_atl2_act_rslvr_table_set(self, index, 0, + HW_ATL2_RPF_TAG_UC_MASK | + HW_ATL2_RPF_TAG_ALLMC_MASK, + HW_ATL2_ACTION_DROP); + + index = priv->art_base_index + HW_ATL2_RPF_VLAN_PROMISC_OFF_INDEX; + hw_atl2_act_rslvr_table_set(self, index, 0, + HW_ATL2_RPF_TAG_VLAN_MASK | + HW_ATL2_RPF_TAG_UNTAG_MASK, + HW_ATL2_ACTION_DROP); + + /* Configure ART to map given VLan Prio (PCP) to the TC index for + * RSS redirection table. + */ + for (i = 0; i < 8; i++) { + action = HW_ATL2_ACTION_ASSIGN_TC(prio_tc_map[i]); + + index = priv->art_base_index + HW_ATL2_RPF_PCP_TO_TC_INDEX + i; + hw_atl2_act_rslvr_table_set(self, index, + i << HW_ATL2_RPF_TAG_PCP_OFFSET, + HW_ATL2_RPF_TAG_PCP_MASK, action); + } +} + +static void hw_atl2_hw_new_rx_filter_vlan_promisc(struct aq_hw_s *self, + bool promisc) +{ + u16 off_action = (!promisc && + !hw_atl_rpfl2promiscuous_mode_en_get(self)) ? + HW_ATL2_ACTION_DROP : HW_ATL2_ACTION_DISABLE; + struct hw_atl2_priv *priv = (struct hw_atl2_priv *)self->priv; + u8 index; + + index = priv->art_base_index + HW_ATL2_RPF_VLAN_PROMISC_OFF_INDEX; + hw_atl2_act_rslvr_table_set(self, index, 0, + HW_ATL2_RPF_TAG_VLAN_MASK | + HW_ATL2_RPF_TAG_UNTAG_MASK, off_action); +} + +static void hw_atl2_hw_new_rx_filter_promisc(struct aq_hw_s *self, bool promisc) +{ + u16 off_action = promisc ? HW_ATL2_ACTION_DISABLE : HW_ATL2_ACTION_DROP; + struct hw_atl2_priv *priv = (struct hw_atl2_priv *)self->priv; + bool vlan_promisc_enable; + u8 index; + + index = priv->art_base_index + HW_ATL2_RPF_L2_PROMISC_OFF_INDEX; + hw_atl2_act_rslvr_table_set(self, index, 0, + HW_ATL2_RPF_TAG_UC_MASK | + HW_ATL2_RPF_TAG_ALLMC_MASK, + off_action); + + /* turn VLAN promisc mode too */ + vlan_promisc_enable = hw_atl_rpf_vlan_prom_mode_en_get(self); + hw_atl2_hw_new_rx_filter_vlan_promisc(self, promisc | + vlan_promisc_enable); +} + +static int hw_atl2_act_rslvr_table_set(struct aq_hw_s *self, u8 location, + u32 tag, u32 mask, u32 action) +{ + u32 val; + int err; + + err = readx_poll_timeout_atomic(hw_atl2_sem_act_rslvr_get, + self, val, val == 1, + 1, 10000U); + if (err) + return err; + + hw_atl2_rpf_act_rslvr_record_set(self, location, tag, mask, + action); + + hw_atl_reg_glb_cpu_sem_set(self, 1, HW_ATL2_FW_SM_ACT_RSLVR); + + return err; +} + +static int hw_atl2_hw_init_rx_path(struct aq_hw_s *self) +{ + struct aq_nic_cfg_s *cfg = self->aq_nic_cfg; + int i; + + /* Rx TC/RSS number config */ + hw_atl_rpb_rpf_rx_traf_class_mode_set(self, cfg->tc_mode); + + /* Rx flow control */ + hw_atl_rpb_rx_flow_ctl_mode_set(self, 1U); + + hw_atl2_rpf_rss_hash_type_set(self, HW_ATL2_RPF_RSS_HASH_TYPE_ALL); + + /* RSS Ring selection */ + hw_atl_b0_hw_init_rx_rss_ctrl1(self); + + /* Multicast filters */ + for (i = HW_ATL2_MAC_MAX; i--;) { + hw_atl_rpfl2_uc_flr_en_set(self, (i == 0U) ? 1U : 0U, i); + hw_atl_rpfl2unicast_flr_act_set(self, 1U, i); + } + + hw_atl_reg_rx_flr_mcst_flr_msk_set(self, 0x00000000U); + hw_atl_reg_rx_flr_mcst_flr_set(self, HW_ATL_MCAST_FLT_ANY_TO_HOST, 0U); + + /* Vlan filters */ + hw_atl_rpf_vlan_outer_etht_set(self, ETH_P_8021AD); + hw_atl_rpf_vlan_inner_etht_set(self, ETH_P_8021Q); + + hw_atl_rpf_vlan_prom_mode_en_set(self, 1); + + /* Always accept untagged packets */ + hw_atl_rpf_vlan_accept_untagged_packets_set(self, 1U); + hw_atl_rpf_vlan_untagged_act_set(self, 1U); + + hw_atl2_hw_init_new_rx_filters(self); + + /* Rx Interrupts */ + hw_atl_rdm_rx_desc_wr_wb_irq_en_set(self, 1U); + + hw_atl_rpfl2broadcast_flr_act_set(self, 1U); + hw_atl_rpfl2broadcast_count_threshold_set(self, 0xFFFFU & (~0U / 256U)); + + hw_atl_rdm_rx_dca_en_set(self, 0U); + hw_atl_rdm_rx_dca_mode_set(self, 0U); + + return aq_hw_err_from_flags(self); +} + +static int hw_atl2_hw_init(struct aq_hw_s *self, u8 *mac_addr) +{ + static u32 aq_hw_atl2_igcr_table_[4][2] = { + [AQ_HW_IRQ_INVALID] = { 0x20000000U, 0x20000000U }, + [AQ_HW_IRQ_LEGACY] = { 0x20000080U, 0x20000080U }, + [AQ_HW_IRQ_MSI] = { 0x20000021U, 0x20000025U }, + [AQ_HW_IRQ_MSIX] = { 0x20000022U, 0x20000026U }, + }; + + struct hw_atl2_priv *priv = (struct hw_atl2_priv *)self->priv; + struct aq_nic_cfg_s *aq_nic_cfg = self->aq_nic_cfg; + u8 base_index, count; + int err; + + err = hw_atl2_utils_get_action_resolve_table_caps(self, &base_index, + &count); + if (err) + return err; + + priv->art_base_index = 8 * base_index; + + hw_atl2_init_launchtime(self); + + hw_atl2_hw_init_tx_path(self); + hw_atl2_hw_init_rx_path(self); + + hw_atl_b0_hw_mac_addr_set(self, mac_addr); + + self->aq_fw_ops->set_link_speed(self, aq_nic_cfg->link_speed_msk); + self->aq_fw_ops->set_state(self, MPI_INIT); + + hw_atl2_hw_qos_set(self); + hw_atl2_hw_rss_set(self, &aq_nic_cfg->aq_rss); + hw_atl_b0_hw_rss_hash_set(self, &aq_nic_cfg->aq_rss); + + hw_atl2_rpf_new_enable_set(self, 1); + + /* Reset link status and read out initial hardware counters */ + self->aq_link_status.mbps = 0; + self->aq_fw_ops->update_stats(self); + + err = aq_hw_err_from_flags(self); + if (err < 0) + goto err_exit; + + /* Interrupts */ + hw_atl_reg_irq_glb_ctl_set(self, + aq_hw_atl2_igcr_table_[aq_nic_cfg->irq_type] + [(aq_nic_cfg->vecs > 1U) ? + 1 : 0]); + + hw_atl_itr_irq_auto_masklsw_set(self, aq_nic_cfg->aq_hw_caps->irq_mask); + + /* Interrupts */ + hw_atl_reg_gen_irq_map_set(self, + ((HW_ATL2_ERR_INT << 0x18) | + (1U << 0x1F)) | + ((HW_ATL2_ERR_INT << 0x10) | + (1U << 0x17)), 0U); + + hw_atl_b0_hw_offload_set(self, aq_nic_cfg); + +err_exit: + return err; +} + +static int hw_atl2_hw_ring_rx_init(struct aq_hw_s *self, + struct aq_ring_s *aq_ring, + struct aq_ring_param_s *aq_ring_param) +{ + return hw_atl_b0_hw_ring_rx_init(self, aq_ring, aq_ring_param); +} + +static int hw_atl2_hw_ring_tx_init(struct aq_hw_s *self, + struct aq_ring_s *aq_ring, + struct aq_ring_param_s *aq_ring_param) +{ + return hw_atl_b0_hw_ring_tx_init(self, aq_ring, aq_ring_param); +} + +#define IS_FILTER_ENABLED(_F_) ((packet_filter & (_F_)) ? 1U : 0U) + +static int hw_atl2_hw_packet_filter_set(struct aq_hw_s *self, + unsigned int packet_filter) +{ + hw_atl2_hw_new_rx_filter_promisc(self, IS_FILTER_ENABLED(IFF_PROMISC)); + + return hw_atl_b0_hw_packet_filter_set(self, packet_filter); +} + +#undef IS_FILTER_ENABLED + +static int hw_atl2_hw_multicast_list_set(struct aq_hw_s *self, + u8 ar_mac + [AQ_HW_MULTICAST_ADDRESS_MAX] + [ETH_ALEN], + u32 count) +{ + struct aq_nic_cfg_s *cfg = self->aq_nic_cfg; + int err = 0; + + if (count > (HW_ATL2_MAC_MAX - HW_ATL2_MAC_MIN)) { + err = -EBADRQC; + goto err_exit; + } + for (cfg->mc_list_count = 0U; + cfg->mc_list_count < count; + ++cfg->mc_list_count) { + u32 i = cfg->mc_list_count; + u32 h = (ar_mac[i][0] << 8) | (ar_mac[i][1]); + u32 l = (ar_mac[i][2] << 24) | (ar_mac[i][3] << 16) | + (ar_mac[i][4] << 8) | ar_mac[i][5]; + + hw_atl_rpfl2_uc_flr_en_set(self, 0U, HW_ATL2_MAC_MIN + i); + + hw_atl_rpfl2unicast_dest_addresslsw_set(self, l, + HW_ATL2_MAC_MIN + i); + + hw_atl_rpfl2unicast_dest_addressmsw_set(self, h, + HW_ATL2_MAC_MIN + i); + + hw_atl2_rpfl2_uc_flr_tag_set(self, 1, HW_ATL2_MAC_MIN + i); + + hw_atl_rpfl2_uc_flr_en_set(self, (cfg->is_mc_list_enabled), + HW_ATL2_MAC_MIN + i); + } + + err = aq_hw_err_from_flags(self); + +err_exit: + return err; +} + +static int hw_atl2_hw_interrupt_moderation_set(struct aq_hw_s *self) +{ + unsigned int i = 0U; + u32 itr_tx = 2U; + u32 itr_rx = 2U; + + switch (self->aq_nic_cfg->itr) { + case AQ_CFG_INTERRUPT_MODERATION_ON: + case AQ_CFG_INTERRUPT_MODERATION_AUTO: + hw_atl_tdm_tx_desc_wr_wb_irq_en_set(self, 0U); + hw_atl_tdm_tdm_intr_moder_en_set(self, 1U); + hw_atl_rdm_rx_desc_wr_wb_irq_en_set(self, 0U); + hw_atl_rdm_rdm_intr_moder_en_set(self, 1U); + + if (self->aq_nic_cfg->itr == AQ_CFG_INTERRUPT_MODERATION_ON) { + /* HW timers are in 2us units */ + int tx_max_timer = self->aq_nic_cfg->tx_itr / 2; + int tx_min_timer = tx_max_timer / 2; + + int rx_max_timer = self->aq_nic_cfg->rx_itr / 2; + int rx_min_timer = rx_max_timer / 2; + + tx_max_timer = min(HW_ATL2_INTR_MODER_MAX, + tx_max_timer); + tx_min_timer = min(HW_ATL2_INTR_MODER_MIN, + tx_min_timer); + rx_max_timer = min(HW_ATL2_INTR_MODER_MAX, + rx_max_timer); + rx_min_timer = min(HW_ATL2_INTR_MODER_MIN, + rx_min_timer); + + itr_tx |= tx_min_timer << 0x8U; + itr_tx |= tx_max_timer << 0x10U; + itr_rx |= rx_min_timer << 0x8U; + itr_rx |= rx_max_timer << 0x10U; + } else { + static unsigned int hw_atl2_timers_table_tx_[][2] = { + {0xfU, 0xffU}, /* 10Gbit */ + {0xfU, 0x1ffU}, /* 5Gbit */ + {0xfU, 0x1ffU}, /* 5Gbit 5GS */ + {0xfU, 0x1ffU}, /* 2.5Gbit */ + {0xfU, 0x1ffU}, /* 1Gbit */ + {0xfU, 0x1ffU}, /* 100Mbit */ + }; + static unsigned int hw_atl2_timers_table_rx_[][2] = { + {0x6U, 0x38U},/* 10Gbit */ + {0xCU, 0x70U},/* 5Gbit */ + {0xCU, 0x70U},/* 5Gbit 5GS */ + {0x18U, 0xE0U},/* 2.5Gbit */ + {0x30U, 0x80U},/* 1Gbit */ + {0x4U, 0x50U},/* 100Mbit */ + }; + unsigned int mbps = self->aq_link_status.mbps; + unsigned int speed_index; + + speed_index = hw_atl_utils_mbps_2_speed_index(mbps); + + /* Update user visible ITR settings */ + self->aq_nic_cfg->tx_itr = hw_atl2_timers_table_tx_ + [speed_index][1] * 2; + self->aq_nic_cfg->rx_itr = hw_atl2_timers_table_rx_ + [speed_index][1] * 2; + + itr_tx |= hw_atl2_timers_table_tx_ + [speed_index][0] << 0x8U; + itr_tx |= hw_atl2_timers_table_tx_ + [speed_index][1] << 0x10U; + + itr_rx |= hw_atl2_timers_table_rx_ + [speed_index][0] << 0x8U; + itr_rx |= hw_atl2_timers_table_rx_ + [speed_index][1] << 0x10U; + } + break; + case AQ_CFG_INTERRUPT_MODERATION_OFF: + hw_atl_tdm_tx_desc_wr_wb_irq_en_set(self, 1U); + hw_atl_tdm_tdm_intr_moder_en_set(self, 0U); + hw_atl_rdm_rx_desc_wr_wb_irq_en_set(self, 1U); + hw_atl_rdm_rdm_intr_moder_en_set(self, 0U); + itr_tx = 0U; + itr_rx = 0U; + break; + } + + for (i = HW_ATL2_RINGS_MAX; i--;) { + hw_atl2_reg_tx_intr_moder_ctrl_set(self, itr_tx, i); + hw_atl_reg_rx_intr_moder_ctrl_set(self, itr_rx, i); + } + + return aq_hw_err_from_flags(self); +} + +static int hw_atl2_hw_stop(struct aq_hw_s *self) +{ + hw_atl_b0_hw_irq_disable(self, HW_ATL2_INT_MASK); + + return 0; +} + +static struct aq_stats_s *hw_atl2_utils_get_hw_stats(struct aq_hw_s *self) +{ + return &self->curr_stats; +} + +static int hw_atl2_hw_vlan_set(struct aq_hw_s *self, + struct aq_rx_filter_vlan *aq_vlans) +{ + struct hw_atl2_priv *priv = (struct hw_atl2_priv *)self->priv; + u32 queue; + u8 index; + int i; + + hw_atl_rpf_vlan_prom_mode_en_set(self, 1U); + + for (i = 0; i < HW_ATL_VLAN_MAX_FILTERS; i++) { + queue = HW_ATL2_ACTION_ASSIGN_QUEUE(aq_vlans[i].queue); + + hw_atl_rpf_vlan_flr_en_set(self, 0U, i); + hw_atl_rpf_vlan_rxq_en_flr_set(self, 0U, i); + index = priv->art_base_index + HW_ATL2_RPF_VLAN_USER_INDEX + i; + hw_atl2_act_rslvr_table_set(self, index, 0, 0, + HW_ATL2_ACTION_DISABLE); + if (aq_vlans[i].enable) { + hw_atl_rpf_vlan_id_flr_set(self, + aq_vlans[i].vlan_id, i); + hw_atl_rpf_vlan_flr_act_set(self, 1U, i); + hw_atl_rpf_vlan_flr_en_set(self, 1U, i); + + if (aq_vlans[i].queue != 0xFF) { + hw_atl_rpf_vlan_rxq_flr_set(self, + aq_vlans[i].queue, + i); + hw_atl_rpf_vlan_rxq_en_flr_set(self, 1U, i); + + hw_atl2_rpf_vlan_flr_tag_set(self, i + 2, i); + + index = priv->art_base_index + + HW_ATL2_RPF_VLAN_USER_INDEX + i; + hw_atl2_act_rslvr_table_set(self, index, + (i + 2) << HW_ATL2_RPF_TAG_VLAN_OFFSET, + HW_ATL2_RPF_TAG_VLAN_MASK, queue); + } else { + hw_atl2_rpf_vlan_flr_tag_set(self, 1, i); + } + } + } + + return aq_hw_err_from_flags(self); +} + +static int hw_atl2_hw_vlan_ctrl(struct aq_hw_s *self, bool enable) +{ + /* set promisc in case of disabing the vlan filter */ + hw_atl_rpf_vlan_prom_mode_en_set(self, !enable); + hw_atl2_hw_new_rx_filter_vlan_promisc(self, !enable); + + return aq_hw_err_from_flags(self); +} + +const struct aq_hw_ops hw_atl2_ops = { + .hw_soft_reset = hw_atl2_utils_soft_reset, + .hw_prepare = hw_atl2_utils_initfw, + .hw_set_mac_address = hw_atl_b0_hw_mac_addr_set, + .hw_init = hw_atl2_hw_init, + .hw_reset = hw_atl2_hw_reset, + .hw_start = hw_atl_b0_hw_start, + .hw_ring_tx_start = hw_atl_b0_hw_ring_tx_start, + .hw_ring_tx_stop = hw_atl_b0_hw_ring_tx_stop, + .hw_ring_rx_start = hw_atl_b0_hw_ring_rx_start, + .hw_ring_rx_stop = hw_atl_b0_hw_ring_rx_stop, + .hw_stop = hw_atl2_hw_stop, + + .hw_ring_tx_xmit = hw_atl_b0_hw_ring_tx_xmit, + .hw_ring_tx_head_update = hw_atl_b0_hw_ring_tx_head_update, + + .hw_ring_rx_receive = hw_atl_b0_hw_ring_rx_receive, + .hw_ring_rx_fill = hw_atl_b0_hw_ring_rx_fill, + + .hw_irq_enable = hw_atl_b0_hw_irq_enable, + .hw_irq_disable = hw_atl_b0_hw_irq_disable, + .hw_irq_read = hw_atl_b0_hw_irq_read, + + .hw_ring_rx_init = hw_atl2_hw_ring_rx_init, + .hw_ring_tx_init = hw_atl2_hw_ring_tx_init, + .hw_packet_filter_set = hw_atl2_hw_packet_filter_set, + .hw_filter_vlan_set = hw_atl2_hw_vlan_set, + .hw_filter_vlan_ctrl = hw_atl2_hw_vlan_ctrl, + .hw_multicast_list_set = hw_atl2_hw_multicast_list_set, + .hw_interrupt_moderation_set = hw_atl2_hw_interrupt_moderation_set, + .hw_rss_set = hw_atl2_hw_rss_set, + .hw_rss_hash_set = hw_atl_b0_hw_rss_hash_set, + .hw_tc_rate_limit_set = hw_atl2_hw_init_tx_tc_rate_limit, + .hw_get_hw_stats = hw_atl2_utils_get_hw_stats, + .hw_get_fw_version = hw_atl2_utils_get_fw_version, + .hw_set_offload = hw_atl_b0_hw_offload_set, +}; diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2.h new file mode 100644 index 000000000000..de8723f1c28a --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2.h @@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#ifndef HW_ATL2_H +#define HW_ATL2_H + +#include "aq_common.h" + +extern const struct aq_hw_caps_s hw_atl2_caps_aqc113; +extern const struct aq_hw_ops hw_atl2_ops; + +#endif /* HW_ATL2_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_internal.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_internal.h new file mode 100644 index 000000000000..5a89bb8722f9 --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_internal.h @@ -0,0 +1,127 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#ifndef HW_ATL2_INTERNAL_H +#define HW_ATL2_INTERNAL_H + +#include "hw_atl2_utils.h" + +#define HW_ATL2_MTU_JUMBO 16352U +#define HW_ATL2_MTU 1514U + +#define HW_ATL2_TX_RINGS 4U +#define HW_ATL2_RX_RINGS 4U + +#define HW_ATL2_RINGS_MAX 32U +#define HW_ATL2_TXD_SIZE (16U) +#define HW_ATL2_RXD_SIZE (16U) + +#define HW_ATL2_MAC_UC 0U +#define HW_ATL2_MAC_MIN 1U +#define HW_ATL2_MAC_MAX 38U + +/* interrupts */ +#define HW_ATL2_ERR_INT 8U +#define HW_ATL2_INT_MASK (0xFFFFFFFFU) + +#define HW_ATL2_TXBUF_MAX 128U +#define HW_ATL2_RXBUF_MAX 192U + +#define HW_ATL2_RSS_REDIRECTION_MAX 64U + +#define HW_ATL2_TC_MAX 8U +#define HW_ATL2_RSS_MAX 8U + +#define HW_ATL2_INTR_MODER_MAX 0x1FF +#define HW_ATL2_INTR_MODER_MIN 0xFF + +#define HW_ATL2_MIN_RXD \ + (ALIGN(AQ_CFG_SKB_FRAGS_MAX + 1U, AQ_HW_RXD_MULTIPLE)) +#define HW_ATL2_MIN_TXD \ + (ALIGN(AQ_CFG_SKB_FRAGS_MAX + 1U, AQ_HW_TXD_MULTIPLE)) + +#define HW_ATL2_MAX_RXD 8184U +#define HW_ATL2_MAX_TXD 8184U + +#define HW_ATL2_FW_SM_ACT_RSLVR 0x3U + +#define HW_ATL2_RPF_TAG_UC_OFFSET 0x0 +#define HW_ATL2_RPF_TAG_ALLMC_OFFSET 0x6 +#define HW_ATL2_RPF_TAG_ET_OFFSET 0x7 +#define HW_ATL2_RPF_TAG_VLAN_OFFSET 0xA +#define HW_ATL2_RPF_TAG_UNTAG_OFFSET 0xE +#define HW_ATL2_RPF_TAG_L3_V4_OFFSET 0xF +#define HW_ATL2_RPF_TAG_L3_V6_OFFSET 0x12 +#define HW_ATL2_RPF_TAG_L4_OFFSET 0x15 +#define HW_ATL2_RPF_TAG_L4_FLEX_OFFSET 0x18 +#define HW_ATL2_RPF_TAG_FLEX_OFFSET 0x1B +#define HW_ATL2_RPF_TAG_PCP_OFFSET 0x1D + +#define HW_ATL2_RPF_TAG_UC_MASK (0x0000003F << HW_ATL2_RPF_TAG_UC_OFFSET) +#define HW_ATL2_RPF_TAG_ALLMC_MASK (0x00000001 << HW_ATL2_RPF_TAG_ALLMC_OFFSET) +#define HW_ATL2_RPF_TAG_UNTAG_MASK (0x00000001 << HW_ATL2_RPF_TAG_UNTAG_OFFSET) +#define HW_ATL2_RPF_TAG_VLAN_MASK (0x0000000F << HW_ATL2_RPF_TAG_VLAN_OFFSET) +#define HW_ATL2_RPF_TAG_ET_MASK (0x00000007 << HW_ATL2_RPF_TAG_ET_OFFSET) +#define HW_ATL2_RPF_TAG_L3_V4_MASK (0x00000007 << HW_ATL2_RPF_TAG_L3_V4_OFFSET) +#define HW_ATL2_RPF_TAG_L3_V6_MASK (0x00000007 << HW_ATL2_RPF_TAG_L3_V6_OFFSET) +#define HW_ATL2_RPF_TAG_L4_MASK (0x00000007 << HW_ATL2_RPF_TAG_L4_OFFSET) +#define HW_ATL2_RPF_TAG_PCP_MASK (0x00000007 << HW_ATL2_RPF_TAG_PCP_OFFSET) + +#define HW_ATL2_RPF_TAG_BASE_UC BIT(HW_ATL2_RPF_TAG_UC_OFFSET) +#define HW_ATL2_RPF_TAG_BASE_ALLMC BIT(HW_ATL2_RPF_TAG_ALLMC_OFFSET) +#define HW_ATL2_RPF_TAG_BASE_UNTAG BIT(HW_ATL2_RPF_TAG_UNTAG_OFFSET) +#define HW_ATL2_RPF_TAG_BASE_VLAN BIT(HW_ATL2_RPF_TAG_VLAN_OFFSET) + +enum HW_ATL2_RPF_ART_INDEX { + HW_ATL2_RPF_L2_PROMISC_OFF_INDEX, + HW_ATL2_RPF_VLAN_PROMISC_OFF_INDEX, + HW_ATL2_RPF_L3L4_USER_INDEX = 8, + HW_ATL2_RPF_ET_PCP_USER_INDEX = HW_ATL2_RPF_L3L4_USER_INDEX + 16, + HW_ATL2_RPF_VLAN_USER_INDEX = HW_ATL2_RPF_ET_PCP_USER_INDEX + 16, + HW_ATL2_RPF_PCP_TO_TC_INDEX = HW_ATL2_RPF_VLAN_USER_INDEX + + HW_ATL_VLAN_MAX_FILTERS, +}; + +#define HW_ATL2_ACTION(ACTION, RSS, INDEX, VALID) \ + ((((ACTION) & 0x3U) << 8) | \ + (((RSS) & 0x1U) << 7) | \ + (((INDEX) & 0x3FU) << 2) | \ + (((VALID) & 0x1U) << 0)) + +#define HW_ATL2_ACTION_DROP HW_ATL2_ACTION(0, 0, 0, 1) +#define HW_ATL2_ACTION_DISABLE HW_ATL2_ACTION(0, 0, 0, 0) +#define HW_ATL2_ACTION_ASSIGN_QUEUE(QUEUE) HW_ATL2_ACTION(1, 0, (QUEUE), 1) +#define HW_ATL2_ACTION_ASSIGN_TC(TC) HW_ATL2_ACTION(1, 1, (TC), 1) + +enum HW_ATL2_RPF_RSS_HASH_TYPE { + HW_ATL2_RPF_RSS_HASH_TYPE_NONE = 0, + HW_ATL2_RPF_RSS_HASH_TYPE_IPV4 = BIT(0), + HW_ATL2_RPF_RSS_HASH_TYPE_IPV4_TCP = BIT(1), + HW_ATL2_RPF_RSS_HASH_TYPE_IPV4_UDP = BIT(2), + HW_ATL2_RPF_RSS_HASH_TYPE_IPV6 = BIT(3), + HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_TCP = BIT(4), + HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_UDP = BIT(5), + HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_EX = BIT(6), + HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_EX_TCP = BIT(7), + HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_EX_UDP = BIT(8), + HW_ATL2_RPF_RSS_HASH_TYPE_ALL = HW_ATL2_RPF_RSS_HASH_TYPE_IPV4 | + HW_ATL2_RPF_RSS_HASH_TYPE_IPV4_TCP | + HW_ATL2_RPF_RSS_HASH_TYPE_IPV4_UDP | + HW_ATL2_RPF_RSS_HASH_TYPE_IPV6 | + HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_TCP | + HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_UDP | + HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_EX | + HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_EX_TCP | + HW_ATL2_RPF_RSS_HASH_TYPE_IPV6_EX_UDP, +}; + +#define HW_ATL_MCAST_FLT_ANY_TO_HOST 0x00010FFFU + +struct hw_atl2_priv { + struct statistics_s last_stats; + unsigned int art_base_index; +}; + +#endif /* HW_ATL2_INTERNAL_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh.c new file mode 100644 index 000000000000..cd954b11d24a --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh.c @@ -0,0 +1,234 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#include "hw_atl2_llh.h" +#include "hw_atl2_llh_internal.h" +#include "aq_hw_utils.h" + +void hw_atl2_rpf_redirection_table2_select_set(struct aq_hw_s *aq_hw, + u32 select) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_ADR, + HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_MSK, + HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_SHIFT, select); +} + +void hw_atl2_rpf_rss_hash_type_set(struct aq_hw_s *aq_hw, u32 rss_hash_type) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_ADR, + HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_MSK, + HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_SHIFT, + rss_hash_type); +} + +/* rpf */ + +void hw_atl2_rpf_new_enable_set(struct aq_hw_s *aq_hw, u32 enable) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_NEW_EN_ADR, + HW_ATL2_RPF_NEW_EN_MSK, + HW_ATL2_RPF_NEW_EN_SHIFT, + enable); +} + +void hw_atl2_rpfl2_uc_flr_tag_set(struct aq_hw_s *aq_hw, u32 tag, u32 filter) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPFL2UC_TAG_ADR(filter), + HW_ATL2_RPFL2UC_TAG_MSK, + HW_ATL2_RPFL2UC_TAG_SHIFT, + tag); +} + +void hw_atl2_rpfl2_bc_flr_tag_set(struct aq_hw_s *aq_hw, u32 tag) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_L2_BC_TAG_ADR, + HW_ATL2_RPF_L2_BC_TAG_MSK, + HW_ATL2_RPF_L2_BC_TAG_SHIFT, + tag); +} + +void hw_atl2_new_rpf_rss_redir_set(struct aq_hw_s *aq_hw, u32 tc, u32 index, + u32 queue) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_RSS_REDIR_ADR(tc, index), + HW_ATL2_RPF_RSS_REDIR_MSK(tc), + HW_ATL2_RPF_RSS_REDIR_SHIFT(tc), + queue); +} + +void hw_atl2_rpf_vlan_flr_tag_set(struct aq_hw_s *aq_hw, u32 tag, u32 filter) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_VL_TAG_ADR(filter), + HW_ATL2_RPF_VL_TAG_MSK, + HW_ATL2_RPF_VL_TAG_SHIFT, + tag); +} + +/* TX */ + +void hw_atl2_tpb_tx_tc_q_rand_map_en_set(struct aq_hw_s *aq_hw, + const u32 tc_q_rand_map_en) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_ADR, + HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_MSK, + HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_SHIFT, + tc_q_rand_map_en); +} + +void hw_atl2_tpb_tx_buf_clk_gate_en_set(struct aq_hw_s *aq_hw, u32 clk_gate_en) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_ADR, + HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_MSK, + HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_SHIFT, + clk_gate_en); +} + +void hw_atl2_reg_tx_intr_moder_ctrl_set(struct aq_hw_s *aq_hw, + u32 tx_intr_moderation_ctl, + u32 queue) +{ + aq_hw_write_reg(aq_hw, HW_ATL2_TX_INTR_MODERATION_CTL_ADR(queue), + tx_intr_moderation_ctl); +} + +void hw_atl2_tps_tx_pkt_shed_data_arb_mode_set(struct aq_hw_s *aq_hw, + const u32 data_arb_mode) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_TPS_DATA_TC_ARB_MODE_ADR, + HW_ATL2_TPS_DATA_TC_ARB_MODE_MSK, + HW_ATL2_TPS_DATA_TC_ARB_MODE_SHIFT, + data_arb_mode); +} + +void hw_atl2_tps_tx_pkt_shed_tc_data_max_credit_set(struct aq_hw_s *aq_hw, + const u32 tc, + const u32 max_credit) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_TPS_DATA_TCTCREDIT_MAX_ADR(tc), + HW_ATL2_TPS_DATA_TCTCREDIT_MAX_MSK, + HW_ATL2_TPS_DATA_TCTCREDIT_MAX_SHIFT, + max_credit); +} + +void hw_atl2_tps_tx_pkt_shed_tc_data_weight_set(struct aq_hw_s *aq_hw, + const u32 tc, + const u32 weight) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_TPS_DATA_TCTWEIGHT_ADR(tc), + HW_ATL2_TPS_DATA_TCTWEIGHT_MSK, + HW_ATL2_TPS_DATA_TCTWEIGHT_SHIFT, + weight); +} + +u32 hw_atl2_get_hw_version(struct aq_hw_s *aq_hw) +{ + return aq_hw_read_reg(aq_hw, HW_ATL2_FPGA_VER_ADR); +} + +void hw_atl2_init_launchtime(struct aq_hw_s *aq_hw) +{ + u32 hw_ver = hw_atl2_get_hw_version(aq_hw); + + aq_hw_write_reg_bit(aq_hw, HW_ATL2_LT_CTRL_ADR, + HW_ATL2_LT_CTRL_CLK_RATIO_MSK, + HW_ATL2_LT_CTRL_CLK_RATIO_SHIFT, + hw_ver < HW_ATL2_FPGA_VER_U32(1, 0, 0, 0) ? + HW_ATL2_LT_CTRL_CLK_RATIO_FULL_SPEED : + hw_ver >= HW_ATL2_FPGA_VER_U32(1, 0, 85, 2) ? + HW_ATL2_LT_CTRL_CLK_RATIO_HALF_SPEED : + HW_ATL2_LT_CTRL_CLK_RATIO_QUATER_SPEED); +} + +/* set action resolver record */ +void hw_atl2_rpf_act_rslvr_record_set(struct aq_hw_s *aq_hw, u8 location, + u32 tag, u32 mask, u32 action) +{ + aq_hw_write_reg(aq_hw, + HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_ADR(location), + tag); + aq_hw_write_reg(aq_hw, + HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_ADR(location), + mask); + aq_hw_write_reg(aq_hw, + HW_ATL2_RPF_ACT_RSLVR_ACTN_ADR(location), + action); +} + +void hw_atl2_rpf_act_rslvr_section_en_set(struct aq_hw_s *aq_hw, u32 sections) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_RPF_REC_TAB_EN_ADR, + HW_ATL2_RPF_REC_TAB_EN_MSK, + HW_ATL2_RPF_REC_TAB_EN_SHIFT, + sections); +} + +void hw_atl2_mif_shared_buf_get(struct aq_hw_s *aq_hw, int offset, u32 *data, + int len) +{ + int j = 0; + int i; + + for (i = offset; i < offset + len; i++, j++) + data[j] = aq_hw_read_reg(aq_hw, + HW_ATL2_MIF_SHARED_BUFFER_IN_ADR(i)); +} + +void hw_atl2_mif_shared_buf_write(struct aq_hw_s *aq_hw, int offset, u32 *data, + int len) +{ + int j = 0; + int i; + + for (i = offset; i < offset + len; i++, j++) + aq_hw_write_reg(aq_hw, HW_ATL2_MIF_SHARED_BUFFER_IN_ADR(i), + data[j]); +} + +void hw_atl2_mif_shared_buf_read(struct aq_hw_s *aq_hw, int offset, u32 *data, + int len) +{ + int j = 0; + int i; + + for (i = offset; i < offset + len; i++, j++) + data[j] = aq_hw_read_reg(aq_hw, + HW_ATL2_MIF_SHARED_BUFFER_OUT_ADR(i)); +} + +void hw_atl2_mif_host_finished_write_set(struct aq_hw_s *aq_hw, u32 finish) +{ + aq_hw_write_reg_bit(aq_hw, HW_ATL2_MIF_HOST_FINISHED_WRITE_ADR, + HW_ATL2_MIF_HOST_FINISHED_WRITE_MSK, + HW_ATL2_MIF_HOST_FINISHED_WRITE_SHIFT, + finish); +} + +u32 hw_atl2_mif_mcp_finished_read_get(struct aq_hw_s *aq_hw) +{ + return aq_hw_read_reg_bit(aq_hw, HW_ATL2_MIF_MCP_FINISHED_READ_ADR, + HW_ATL2_MIF_MCP_FINISHED_READ_MSK, + HW_ATL2_MIF_MCP_FINISHED_READ_SHIFT); +} + +u32 hw_atl2_mif_mcp_boot_reg_get(struct aq_hw_s *aq_hw) +{ + return aq_hw_read_reg(aq_hw, HW_ATL2_MIF_BOOT_REG_ADR); +} + +void hw_atl2_mif_mcp_boot_reg_set(struct aq_hw_s *aq_hw, u32 val) +{ + return aq_hw_write_reg(aq_hw, HW_ATL2_MIF_BOOT_REG_ADR, val); +} + +u32 hw_atl2_mif_host_req_int_get(struct aq_hw_s *aq_hw) +{ + return aq_hw_read_reg(aq_hw, HW_ATL2_MCP_HOST_REQ_INT_ADR); +} + +void hw_atl2_mif_host_req_int_clr(struct aq_hw_s *aq_hw, u32 val) +{ + return aq_hw_write_reg(aq_hw, HW_ATL2_MCP_HOST_REQ_INT_CLR_ADR, + val); +} diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh.h new file mode 100644 index 000000000000..98c7a4621297 --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh.h @@ -0,0 +1,102 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#ifndef HW_ATL2_LLH_H +#define HW_ATL2_LLH_H + +#include <linux/types.h> + +struct aq_hw_s; + +/* Set TX Interrupt Moderation Control Register */ +void hw_atl2_reg_tx_intr_moder_ctrl_set(struct aq_hw_s *aq_hw, + u32 tx_intr_moderation_ctl, + u32 queue); + +/* Set Redirection Table 2 Select */ +void hw_atl2_rpf_redirection_table2_select_set(struct aq_hw_s *aq_hw, + u32 select); + +/** Set RSS HASH type */ +void hw_atl2_rpf_rss_hash_type_set(struct aq_hw_s *aq_hw, u32 rss_hash_type); + +/* set new RPF enable */ +void hw_atl2_rpf_new_enable_set(struct aq_hw_s *aq_hw, u32 enable); + +/* set l2 unicast filter tag */ +void hw_atl2_rpfl2_uc_flr_tag_set(struct aq_hw_s *aq_hw, u32 tag, u32 filter); + +/* set l2 broadcast filter tag */ +void hw_atl2_rpfl2_bc_flr_tag_set(struct aq_hw_s *aq_hw, u32 tag); + +/* set new rss redirection table */ +void hw_atl2_new_rpf_rss_redir_set(struct aq_hw_s *aq_hw, u32 tc, u32 index, + u32 queue); + +/* Set VLAN filter tag */ +void hw_atl2_rpf_vlan_flr_tag_set(struct aq_hw_s *aq_hw, u32 tag, u32 filter); + +/* set tx random TC-queue mapping enable bit */ +void hw_atl2_tpb_tx_tc_q_rand_map_en_set(struct aq_hw_s *aq_hw, + const u32 tc_q_rand_map_en); + +/* set tx buffer clock gate enable */ +void hw_atl2_tpb_tx_buf_clk_gate_en_set(struct aq_hw_s *aq_hw, u32 clk_gate_en); + +void hw_atl2_tps_tx_pkt_shed_data_arb_mode_set(struct aq_hw_s *aq_hw, + const u32 data_arb_mode); + +/* set tx packet scheduler tc data max credit */ +void hw_atl2_tps_tx_pkt_shed_tc_data_max_credit_set(struct aq_hw_s *aq_hw, + const u32 tc, + const u32 max_credit); + +/* set tx packet scheduler tc data weight */ +void hw_atl2_tps_tx_pkt_shed_tc_data_weight_set(struct aq_hw_s *aq_hw, + const u32 tc, + const u32 weight); + +u32 hw_atl2_get_hw_version(struct aq_hw_s *aq_hw); + +void hw_atl2_init_launchtime(struct aq_hw_s *aq_hw); + +/* set action resolver record */ +void hw_atl2_rpf_act_rslvr_record_set(struct aq_hw_s *aq_hw, u8 location, + u32 tag, u32 mask, u32 action); + +/* set enable action resolver section */ +void hw_atl2_rpf_act_rslvr_section_en_set(struct aq_hw_s *aq_hw, u32 sections); + +/* get data from firmware shared input buffer */ +void hw_atl2_mif_shared_buf_get(struct aq_hw_s *aq_hw, int offset, u32 *data, + int len); + +/* set data into firmware shared input buffer */ +void hw_atl2_mif_shared_buf_write(struct aq_hw_s *aq_hw, int offset, u32 *data, + int len); + +/* get data from firmware shared output buffer */ +void hw_atl2_mif_shared_buf_read(struct aq_hw_s *aq_hw, int offset, u32 *data, + int len); + +/* set host finished write shared buffer indication */ +void hw_atl2_mif_host_finished_write_set(struct aq_hw_s *aq_hw, u32 finish); + +/* get mcp finished read shared buffer indication */ +u32 hw_atl2_mif_mcp_finished_read_get(struct aq_hw_s *aq_hw); + +/* get mcp boot register */ +u32 hw_atl2_mif_mcp_boot_reg_get(struct aq_hw_s *aq_hw); + +/* set mcp boot register */ +void hw_atl2_mif_mcp_boot_reg_set(struct aq_hw_s *aq_hw, u32 val); + +/* get host interrupt request */ +u32 hw_atl2_mif_host_req_int_get(struct aq_hw_s *aq_hw); + +/* clear host interrupt request */ +void hw_atl2_mif_host_req_int_clr(struct aq_hw_s *aq_hw, u32 val); + +#endif /* HW_ATL2_LLH_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh_internal.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh_internal.h new file mode 100644 index 000000000000..e34c5cda061e --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_llh_internal.h @@ -0,0 +1,391 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#ifndef HW_ATL2_LLH_INTERNAL_H +#define HW_ATL2_LLH_INTERNAL_H + +/* RX pif_rpf_redir_2_en_i Bitfield Definitions + * PORT="pif_rpf_redir_2_en_i" + */ +#define HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_ADR 0x000054C8 +#define HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_MSK 0x00001000 +#define HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_MSKN 0xFFFFEFFF +#define HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_SHIFT 12 +#define HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_WIDTH 1 +#define HW_ATL2_RPF_PIF_RPF_REDIR2_ENI_DEFAULT 0x0 + +/* RX pif_rpf_rss_hash_type_i Bitfield Definitions + */ +#define HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_ADR 0x000054C8 +#define HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_MSK 0x000001FF +#define HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_MSKN 0xFFFFFE00 +#define HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_SHIFT 0 +#define HW_ATL2_RPF_PIF_RPF_RSS_HASH_TYPEI_WIDTH 9 + +/* rx rpf_new_rpf_en bitfield definitions + * preprocessor definitions for the bitfield "rpf_new_rpf_en_i". + * port="pif_rpf_new_rpf_en_i + */ + +/* register address for bitfield rpf_new_rpf_en */ +#define HW_ATL2_RPF_NEW_EN_ADR 0x00005104 +/* bitmask for bitfield rpf_new_rpf_en */ +#define HW_ATL2_RPF_NEW_EN_MSK 0x00000800 +/* inverted bitmask for bitfield rpf_new_rpf_en */ +#define HW_ATL2_RPF_NEW_EN_MSKN 0xfffff7ff +/* lower bit position of bitfield rpf_new_rpf_en */ +#define HW_ATL2_RPF_NEW_EN_SHIFT 11 +/* width of bitfield rpf_new_rpf_en */ +#define HW_ATL2_RPF_NEW_EN_WIDTH 1 +/* default value of bitfield rpf_new_rpf_en */ +#define HW_ATL2_RPF_NEW_EN_DEFAULT 0x0 + +/* rx l2_uc_req_tag0{f}[5:0] bitfield definitions + * preprocessor definitions for the bitfield "l2_uc_req_tag0{f}[7:0]". + * parameter: filter {f} | stride size 0x8 | range [0, 37] + * port="pif_rpf_l2_uc_req_tag0[5:0]" + */ + +/* register address for bitfield l2_uc_req_tag0{f}[2:0] */ +#define HW_ATL2_RPFL2UC_TAG_ADR(filter) (0x00005114 + (filter) * 0x8) +/* bitmask for bitfield l2_uc_req_tag0{f}[2:0] */ +#define HW_ATL2_RPFL2UC_TAG_MSK 0x0FC00000 +/* inverted bitmask for bitfield l2_uc_req_tag0{f}[2:0] */ +#define HW_ATL2_RPFL2UC_TAG_MSKN 0xF03FFFFF +/* lower bit position of bitfield l2_uc_req_tag0{f}[2:0] */ +#define HW_ATL2_RPFL2UC_TAG_SHIFT 22 +/* width of bitfield l2_uc_req_tag0{f}[2:0] */ +#define HW_ATL2_RPFL2UC_TAG_WIDTH 6 +/* default value of bitfield l2_uc_req_tag0{f}[2:0] */ +#define HW_ATL2_RPFL2UC_TAG_DEFAULT 0x0 + +/* rpf_l2_bc_req_tag[5:0] bitfield definitions + * preprocessor definitions for the bitfield "rpf_l2_bc_req_tag[5:0]". + * port="pifrpf_l2_bc_req_tag_i[5:0]" + */ + +/* register address for bitfield rpf_l2_bc_req_tag */ +#define HW_ATL2_RPF_L2_BC_TAG_ADR 0x000050F0 +/* bitmask for bitfield rpf_l2_bc_req_tag */ +#define HW_ATL2_RPF_L2_BC_TAG_MSK 0x0000003F +/* inverted bitmask for bitfield rpf_l2_bc_req_tag */ +#define HW_ATL2_RPF_L2_BC_TAG_MSKN 0xffffffc0 +/* lower bit position of bitfield rpf_l2_bc_req_tag */ +#define HW_ATL2_RPF_L2_BC_TAG_SHIFT 0 +/* width of bitfield rpf_l2_bc_req_tag */ +#define HW_ATL2_RPF_L2_BC_TAG_WIDTH 6 +/* default value of bitfield rpf_l2_bc_req_tag */ +#define HW_ATL2_RPF_L2_BC_TAG_DEFAULT 0x0 + +/* rx rpf_rss_red1_data_[4:0] bitfield definitions + * preprocessor definitions for the bitfield "rpf_rss_red1_data[4:0]". + * port="pif_rpf_rss_red1_data_i[4:0]" + */ + +/* register address for bitfield rpf_rss_red1_data[4:0] */ +#define HW_ATL2_RPF_RSS_REDIR_ADR(TC, INDEX) (0x00006200 + \ + (0x100 * !!((TC) > 3)) + (INDEX) * 4) +/* bitmask for bitfield rpf_rss_red1_data[4:0] */ +#define HW_ATL2_RPF_RSS_REDIR_MSK(TC) (0x00000001F << (5 * ((TC) % 4))) +/* lower bit position of bitfield rpf_rss_red1_data[4:0] */ +#define HW_ATL2_RPF_RSS_REDIR_SHIFT(TC) (5 * ((TC) % 4)) +/* width of bitfield rpf_rss_red1_data[4:0] */ +#define HW_ATL2_RPF_RSS_REDIR_WIDTH 5 +/* default value of bitfield rpf_rss_red1_data[4:0] */ +#define HW_ATL2_RPF_RSS_REDIR_DEFAULT 0x0 + +/* rx vlan_req_tag0{f}[3:0] bitfield definitions + * preprocessor definitions for the bitfield "vlan_req_tag0{f}[3:0]". + * parameter: filter {f} | stride size 0x4 | range [0, 15] + * port="pif_rpf_vlan_req_tag0[3:0]" + */ + +/* register address for bitfield vlan_req_tag0{f}[3:0] */ +#define HW_ATL2_RPF_VL_TAG_ADR(filter) (0x00005290 + (filter) * 0x4) +/* bitmask for bitfield vlan_req_tag0{f}[3:0] */ +#define HW_ATL2_RPF_VL_TAG_MSK 0x0000F000 +/* inverted bitmask for bitfield vlan_req_tag0{f}[3:0] */ +#define HW_ATL2_RPF_VL_TAG_MSKN 0xFFFF0FFF +/* lower bit position of bitfield vlan_req_tag0{f}[3:0] */ +#define HW_ATL2_RPF_VL_TAG_SHIFT 12 +/* width of bitfield vlan_req_tag0{f}[3:0] */ +#define HW_ATL2_RPF_VL_TAG_WIDTH 4 +/* default value of bitfield vlan_req_tag0{f}[3:0] */ +#define HW_ATL2_RPF_VL_TAG_DEFAULT 0x0 + +/* RX rx_q{Q}_tc_map[2:0] Bitfield Definitions + * Preprocessor definitions for the bitfield "rx_q{Q}_tc_map[2:0]". + * Parameter: Queue {Q} | bit-level stride | range [0, 31] + * PORT="pif_rx_q0_tc_map_i[2:0]" + */ + +/* Register address for bitfield rx_q{Q}_tc_map[2:0] */ +#define HW_ATL2_RX_Q_TC_MAP_ADR(queue) \ + (((queue) < 32) ? 0x00005900 + ((queue) / 8) * 4 : 0) +/* Lower bit position of bitfield rx_q{Q}_tc_map[2:0] */ +#define HW_ATL2_RX_Q_TC_MAP_SHIFT(queue) \ + (((queue) < 32) ? ((queue) * 4) % 32 : 0) +/* Width of bitfield rx_q{Q}_tc_map[2:0] */ +#define HW_ATL2_RX_Q_TC_MAP_WIDTH 3 +/* Default value of bitfield rx_q{Q}_tc_map[2:0] */ +#define HW_ATL2_RX_Q_TC_MAP_DEFAULT 0x0 + +/* tx tx_tc_q_rand_map_en bitfield definitions + * preprocessor definitions for the bitfield "tx_tc_q_rand_map_en". + * port="pif_tpb_tx_tc_q_rand_map_en_i" + */ + +/* register address for bitfield tx_tc_q_rand_map_en */ +#define HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_ADR 0x00007900 +/* bitmask for bitfield tx_tc_q_rand_map_en */ +#define HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_MSK 0x00000200 +/* inverted bitmask for bitfield tx_tc_q_rand_map_en */ +#define HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_MSKN 0xFFFFFDFF +/* lower bit position of bitfield tx_tc_q_rand_map_en */ +#define HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_SHIFT 9 +/* width of bitfield tx_tc_q_rand_map_en */ +#define HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_WIDTH 1 +/* default value of bitfield tx_tc_q_rand_map_en */ +#define HW_ATL2_TPB_TX_TC_Q_RAND_MAP_EN_DEFAULT 0x0 + +/* tx tx_buffer_clk_gate_en bitfield definitions + * preprocessor definitions for the bitfield "tx_buffer_clk_gate_en". + * port="pif_tpb_tx_buffer_clk_gate_en_i" + */ + +/* register address for bitfield tx_buffer_clk_gate_en */ +#define HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_ADR 0x00007900 +/* bitmask for bitfield tx_buffer_clk_gate_en */ +#define HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_MSK 0x00000020 +/* inverted bitmask for bitfield tx_buffer_clk_gate_en */ +#define HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_MSKN 0xffffffdf +/* lower bit position of bitfield tx_buffer_clk_gate_en */ +#define HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_SHIFT 5 +/* width of bitfield tx_buffer_clk_gate_en */ +#define HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_WIDTH 1 +/* default value of bitfield tx_buffer_clk_gate_en */ +#define HW_ATL2_TPB_TX_BUF_CLK_GATE_EN_DEFAULT 0x0 + +/* tx tx_q_tc_map{q} bitfield definitions + * preprocessor definitions for the bitfield "tx_q_tc_map{q}". + * parameter: queue {q} | bit-level stride | range [0, 31] + * port="pif_tpb_tx_q_tc_map0_i[2:0]" + */ + +/* register address for bitfield tx_q_tc_map{q} */ +#define HW_ATL2_TX_Q_TC_MAP_ADR(queue) \ + (((queue) < 32) ? 0x0000799C + ((queue) / 4) * 4 : 0) +/* lower bit position of bitfield tx_q_tc_map{q} */ +#define HW_ATL2_TX_Q_TC_MAP_SHIFT(queue) \ + (((queue) < 32) ? ((queue) * 8) % 32 : 0) +/* width of bitfield tx_q_tc_map{q} */ +#define HW_ATL2_TX_Q_TC_MAP_WIDTH 3 +/* default value of bitfield tx_q_tc_map{q} */ +#define HW_ATL2_TX_Q_TC_MAP_DEFAULT 0x0 + +/* tx data_tc_arb_mode bitfield definitions + * preprocessor definitions for the bitfield "data_tc_arb_mode". + * port="pif_tps_data_tc_arb_mode_i" + */ + +/* register address for bitfield data_tc_arb_mode */ +#define HW_ATL2_TPS_DATA_TC_ARB_MODE_ADR 0x00007100 +/* bitmask for bitfield data_tc_arb_mode */ +#define HW_ATL2_TPS_DATA_TC_ARB_MODE_MSK 0x00000003 +/* inverted bitmask for bitfield data_tc_arb_mode */ +#define HW_ATL2_TPS_DATA_TC_ARB_MODE_MSKN 0xfffffffc +/* lower bit position of bitfield data_tc_arb_mode */ +#define HW_ATL2_TPS_DATA_TC_ARB_MODE_SHIFT 0 +/* width of bitfield data_tc_arb_mode */ +#define HW_ATL2_TPS_DATA_TC_ARB_MODE_WIDTH 2 +/* default value of bitfield data_tc_arb_mode */ +#define HW_ATL2_TPS_DATA_TC_ARB_MODE_DEFAULT 0x0 + +/* tx data_tc{t}_credit_max[f:0] bitfield definitions + * preprocessor definitions for the bitfield "data_tc{t}_credit_max[f:0]". + * parameter: tc {t} | stride size 0x4 | range [0, 7] + * port="pif_tps_data_tc0_credit_max_i[15:0]" + */ + +/* register address for bitfield data_tc{t}_credit_max[f:0] */ +#define HW_ATL2_TPS_DATA_TCTCREDIT_MAX_ADR(tc) (0x00007110 + (tc) * 0x4) +/* bitmask for bitfield data_tc{t}_credit_max[f:0] */ +#define HW_ATL2_TPS_DATA_TCTCREDIT_MAX_MSK 0xffff0000 +/* inverted bitmask for bitfield data_tc{t}_credit_max[f:0] */ +#define HW_ATL2_TPS_DATA_TCTCREDIT_MAX_MSKN 0x0000ffff +/* lower bit position of bitfield data_tc{t}_credit_max[f:0] */ +#define HW_ATL2_TPS_DATA_TCTCREDIT_MAX_SHIFT 16 +/* width of bitfield data_tc{t}_credit_max[f:0] */ +#define HW_ATL2_TPS_DATA_TCTCREDIT_MAX_WIDTH 16 +/* default value of bitfield data_tc{t}_credit_max[f:0] */ +#define HW_ATL2_TPS_DATA_TCTCREDIT_MAX_DEFAULT 0x0 + +/* tx data_tc{t}_weight[e:0] bitfield definitions + * preprocessor definitions for the bitfield "data_tc{t}_weight[e:0]". + * parameter: tc {t} | stride size 0x4 | range [0, 7] + * port="pif_tps_data_tc0_weight_i[14:0]" + */ + +/* register address for bitfield data_tc{t}_weight[e:0] */ +#define HW_ATL2_TPS_DATA_TCTWEIGHT_ADR(tc) (0x00007110 + (tc) * 0x4) +/* bitmask for bitfield data_tc{t}_weight[e:0] */ +#define HW_ATL2_TPS_DATA_TCTWEIGHT_MSK 0x00007fff +/* inverted bitmask for bitfield data_tc{t}_weight[e:0] */ +#define HW_ATL2_TPS_DATA_TCTWEIGHT_MSKN 0xffff8000 +/* lower bit position of bitfield data_tc{t}_weight[e:0] */ +#define HW_ATL2_TPS_DATA_TCTWEIGHT_SHIFT 0 +/* width of bitfield data_tc{t}_weight[e:0] */ +#define HW_ATL2_TPS_DATA_TCTWEIGHT_WIDTH 15 +/* default value of bitfield data_tc{t}_weight[e:0] */ +#define HW_ATL2_TPS_DATA_TCTWEIGHT_DEFAULT 0x0 + +/* tx interrupt moderation control register definitions + * Preprocessor definitions for TX Interrupt Moderation Control Register + * Base Address: 0x00007c28 + * Parameter: queue {Q} | stride size 0x4 | range [0, 31] + */ + +#define HW_ATL2_TX_INTR_MODERATION_CTL_ADR(queue) (0x00007c28u + (queue) * 0x40) + +/* Launch time control register */ +#define HW_ATL2_LT_CTRL_ADR 0x00007a1c + +#define HW_ATL2_LT_CTRL_AVB_LEN_CMP_TRSHLD_MSK 0xFFFF0000 +#define HW_ATL2_LT_CTRL_AVB_LEN_CMP_TRSHLD_SHIFT 16 + +#define HW_ATL2_LT_CTRL_CLK_RATIO_MSK 0x0000FF00 +#define HW_ATL2_LT_CTRL_CLK_RATIO_SHIFT 8 +#define HW_ATL2_LT_CTRL_CLK_RATIO_QUATER_SPEED 4 +#define HW_ATL2_LT_CTRL_CLK_RATIO_HALF_SPEED 2 +#define HW_ATL2_LT_CTRL_CLK_RATIO_FULL_SPEED 1 + +#define HW_ATL2_LT_CTRL_25G_MODE_SUPPORT_MSK 0x00000008 +#define HW_ATL2_LT_CTRL_25G_MODE_SUPPORT_SHIFT 3 + +#define HW_ATL2_LT_CTRL_LINK_SPEED_MSK 0x00000007 +#define HW_ATL2_LT_CTRL_LINK_SPEED_SHIFT 0 + +/* FPGA VER register */ +#define HW_ATL2_FPGA_VER_ADR 0x000000f4 +#define HW_ATL2_FPGA_VER_U32(mj, mi, bl, rv) \ + ((((mj) & 0xff) << 24) | \ + (((mi) & 0xff) << 16) | \ + (((bl) & 0xff) << 8) | \ + (((rv) & 0xff) << 0)) + +/* ahb_mem_addr{f}[31:0] Bitfield Definitions + * Preprocessor definitions for the bitfield "ahb_mem_addr{f}[31:0]". + * Parameter: filter {f} | stride size 0x10 | range [0, 127] + * PORT="ahb_mem_addr{f}[31:0]" + */ + +/* Register address for bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_ADR(filter) \ + (0x00014000u + (filter) * 0x10) +/* Bitmask for bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_MSK 0xFFFFFFFFu +/* Inverted bitmask for bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_MSKN 0x00000000u +/* Lower bit position of bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_SHIFT 0 +/* Width of bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_WIDTH 31 +/* Default value of bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_REQ_TAG_DEFAULT 0x0 + +/* Register address for bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_ADR(filter) \ + (0x00014004u + (filter) * 0x10) +/* Bitmask for bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_MSK 0xFFFFFFFFu +/* Inverted bitmask for bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_MSKN 0x00000000u +/* Lower bit position of bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_SHIFT 0 +/* Width of bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_WIDTH 31 +/* Default value of bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_TAG_MASK_DEFAULT 0x0 + +/* Register address for bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_ACTN_ADR(filter) \ + (0x00014008u + (filter) * 0x10) +/* Bitmask for bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_ACTN_MSK 0x000007FFu +/* Inverted bitmask for bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_ACTN_MSKN 0xFFFFF800u +/* Lower bit position of bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_ACTN_SHIFT 0 +/* Width of bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_ACTN_WIDTH 10 +/* Default value of bitfield ahb_mem_addr{f}[31:0] */ +#define HW_ATL2_RPF_ACT_RSLVR_ACTN_DEFAULT 0x0 + +/* rpf_rec_tab_en[15:0] Bitfield Definitions + * Preprocessor definitions for the bitfield "rpf_rec_tab_en[15:0]". + * PORT="pif_rpf_rec_tab_en[15:0]" + */ +/* Register address for bitfield rpf_rec_tab_en[15:0] */ +#define HW_ATL2_RPF_REC_TAB_EN_ADR 0x00006ff0u +/* Bitmask for bitfield rpf_rec_tab_en[15:0] */ +#define HW_ATL2_RPF_REC_TAB_EN_MSK 0x0000FFFFu +/* Inverted bitmask for bitfield rpf_rec_tab_en[15:0] */ +#define HW_ATL2_RPF_REC_TAB_EN_MSKN 0xFFFF0000u +/* Lower bit position of bitfield rpf_rec_tab_en[15:0] */ +#define HW_ATL2_RPF_REC_TAB_EN_SHIFT 0 +/* Width of bitfield rpf_rec_tab_en[15:0] */ +#define HW_ATL2_RPF_REC_TAB_EN_WIDTH 16 +/* Default value of bitfield rpf_rec_tab_en[15:0] */ +#define HW_ATL2_RPF_REC_TAB_EN_DEFAULT 0x0 + +/* Register address for firmware shared input buffer */ +#define HW_ATL2_MIF_SHARED_BUFFER_IN_ADR(dword) (0x00012000U + (dword) * 0x4U) +/* Register address for firmware shared output buffer */ +#define HW_ATL2_MIF_SHARED_BUFFER_OUT_ADR(dword) (0x00013000U + (dword) * 0x4U) + +/* pif_host_finished_buf_wr_i Bitfield Definitions + * Preprocessor definitions for the bitfield "pif_host_finished_buf_wr_i". + * PORT="pif_host_finished_buf_wr_i" + */ +/* Register address for bitfield rpif_host_finished_buf_wr_i */ +#define HW_ATL2_MIF_HOST_FINISHED_WRITE_ADR 0x00000e00u +/* Bitmask for bitfield pif_host_finished_buf_wr_i */ +#define HW_ATL2_MIF_HOST_FINISHED_WRITE_MSK 0x00000001u +/* Inverted bitmask for bitfield pif_host_finished_buf_wr_i */ +#define HW_ATL2_MIF_HOST_FINISHED_WRITE_MSKN 0xFFFFFFFEu +/* Lower bit position of bitfield pif_host_finished_buf_wr_i */ +#define HW_ATL2_MIF_HOST_FINISHED_WRITE_SHIFT 0 +/* Width of bitfield pif_host_finished_buf_wr_i */ +#define HW_ATL2_MIF_HOST_FINISHED_WRITE_WIDTH 1 +/* Default value of bitfield pif_host_finished_buf_wr_i */ +#define HW_ATL2_MIF_HOST_FINISHED_WRITE_DEFAULT 0x0 + +/* pif_mcp_finished_buf_rd_i Bitfield Definitions + * Preprocessor definitions for the bitfield "pif_mcp_finished_buf_rd_i". + * PORT="pif_mcp_finished_buf_rd_i" + */ +/* Register address for bitfield pif_mcp_finished_buf_rd_i */ +#define HW_ATL2_MIF_MCP_FINISHED_READ_ADR 0x00000e04u +/* Bitmask for bitfield pif_mcp_finished_buf_rd_i */ +#define HW_ATL2_MIF_MCP_FINISHED_READ_MSK 0x00000001u +/* Inverted bitmask for bitfield pif_mcp_finished_buf_rd_i */ +#define HW_ATL2_MIF_MCP_FINISHED_READ_MSKN 0xFFFFFFFEu +/* Lower bit position of bitfield pif_mcp_finished_buf_rd_i */ +#define HW_ATL2_MIF_MCP_FINISHED_READ_SHIFT 0 +/* Width of bitfield pif_mcp_finished_buf_rd_i */ +#define HW_ATL2_MIF_MCP_FINISHED_READ_WIDTH 1 +/* Default value of bitfield pif_mcp_finished_buf_rd_i */ +#define HW_ATL2_MIF_MCP_FINISHED_READ_DEFAULT 0x0 + +/* Register address for bitfield pif_mcp_boot_reg */ +#define HW_ATL2_MIF_BOOT_REG_ADR 0x00003040u + +#define HW_ATL2_MCP_HOST_REQ_INT_READY BIT(0) + +#define HW_ATL2_MCP_HOST_REQ_INT_ADR 0x00000F00u +#define HW_ATL2_MCP_HOST_REQ_INT_SET_ADR 0x00000F04u +#define HW_ATL2_MCP_HOST_REQ_INT_CLR_ADR 0x00000F08u + +#endif /* HW_ATL2_LLH_INTERNAL_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils.c new file mode 100644 index 000000000000..f3766780e975 --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils.c @@ -0,0 +1,131 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#include <linux/iopoll.h> + +#include "aq_hw_utils.h" +#include "hw_atl/hw_atl_utils.h" +#include "hw_atl2_utils.h" +#include "hw_atl2_llh.h" +#include "hw_atl2_llh_internal.h" + +#define HW_ATL2_FW_VER_1X 0x01000000U + +#define AQ_A2_BOOT_STARTED BIT(0x18) +#define AQ_A2_CRASH_INIT BIT(0x1B) +#define AQ_A2_BOOT_CODE_FAILED BIT(0x1C) +#define AQ_A2_FW_INIT_FAILED BIT(0x1D) +#define AQ_A2_FW_INIT_COMP_SUCCESS BIT(0x1F) + +#define AQ_A2_FW_BOOT_FAILED_MASK (AQ_A2_CRASH_INIT | \ + AQ_A2_BOOT_CODE_FAILED | \ + AQ_A2_FW_INIT_FAILED) +#define AQ_A2_FW_BOOT_COMPLETE_MASK (AQ_A2_FW_BOOT_FAILED_MASK | \ + AQ_A2_FW_INIT_COMP_SUCCESS) + +#define AQ_A2_FW_BOOT_REQ_REBOOT BIT(0x0) +#define AQ_A2_FW_BOOT_REQ_HOST_BOOT BIT(0x8) +#define AQ_A2_FW_BOOT_REQ_MAC_FAST_BOOT BIT(0xA) +#define AQ_A2_FW_BOOT_REQ_PHY_FAST_BOOT BIT(0xB) + +int hw_atl2_utils_initfw(struct aq_hw_s *self, const struct aq_fw_ops **fw_ops) +{ + int err; + + self->fw_ver_actual = hw_atl2_utils_get_fw_version(self); + + if (hw_atl_utils_ver_match(HW_ATL2_FW_VER_1X, + self->fw_ver_actual) == 0) { + *fw_ops = &aq_a2_fw_ops; + } else { + aq_pr_err("Bad FW version detected: %x, but continue\n", + self->fw_ver_actual); + *fw_ops = &aq_a2_fw_ops; + } + aq_pr_trace("Detect ATL2FW %x\n", self->fw_ver_actual); + self->aq_fw_ops = *fw_ops; + err = self->aq_fw_ops->init(self); + + self->chip_features |= ATL_HW_CHIP_ANTIGUA; + + return err; +} + +static bool hw_atl2_mcp_boot_complete(struct aq_hw_s *self) +{ + u32 rbl_status; + + rbl_status = hw_atl2_mif_mcp_boot_reg_get(self); + if (rbl_status & AQ_A2_FW_BOOT_COMPLETE_MASK) + return true; + + /* Host boot requested */ + if (hw_atl2_mif_host_req_int_get(self) & HW_ATL2_MCP_HOST_REQ_INT_READY) + return true; + + return false; +} + +int hw_atl2_utils_soft_reset(struct aq_hw_s *self) +{ + bool rbl_complete = false; + u32 rbl_status = 0; + u32 rbl_request; + int err; + + hw_atl2_mif_host_req_int_clr(self, 0x01); + rbl_request = AQ_A2_FW_BOOT_REQ_REBOOT; +#ifdef AQ_CFG_FAST_START + rbl_request |= AQ_A2_FW_BOOT_REQ_MAC_FAST_BOOT; +#endif + hw_atl2_mif_mcp_boot_reg_set(self, rbl_request); + + /* Wait for RBL boot */ + err = readx_poll_timeout_atomic(hw_atl2_mif_mcp_boot_reg_get, self, + rbl_status, + ((rbl_status & AQ_A2_BOOT_STARTED) && + (rbl_status != 0xFFFFFFFFu)), + 10, 200000); + if (err) { + aq_pr_err("Boot code hanged"); + goto err_exit; + } + + err = readx_poll_timeout_atomic(hw_atl2_mcp_boot_complete, self, + rbl_complete, + rbl_complete, + 10, 2000000); + + if (err) { + aq_pr_err("FW Restart timed out"); + goto err_exit; + } + + rbl_status = hw_atl2_mif_mcp_boot_reg_get(self); + + if (rbl_status & AQ_A2_FW_BOOT_FAILED_MASK) { + err = -EIO; + aq_pr_err("FW Restart failed"); + goto err_exit; + } + + if (hw_atl2_mif_host_req_int_get(self) & + HW_ATL2_MCP_HOST_REQ_INT_READY) { + err = -EIO; + aq_pr_err("No FW detected. Dynamic FW load not implemented"); + goto err_exit; + } + + if (self->aq_fw_ops) { + err = self->aq_fw_ops->init(self); + if (err) { + aq_pr_err("FW Init failed"); + goto err_exit; + } + } + +err_exit: + return err; +} diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils.h new file mode 100644 index 000000000000..b66fa346581c --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils.h @@ -0,0 +1,606 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#ifndef HW_ATL2_UTILS_H +#define HW_ATL2_UTILS_H + +#include "aq_hw.h" + +/* F W A P I */ + +struct link_options_s { + u8 link_up:1; + u8 link_renegotiate:1; + u8 minimal_link_speed:1; + u8 internal_loopback:1; + u8 external_loopback:1; + u8 rate_10M_hd:1; + u8 rate_100M_hd:1; + u8 rate_1G_hd:1; + + u8 rate_10M:1; + u8 rate_100M:1; + u8 rate_1G:1; + u8 rate_2P5G:1; + u8 rate_N2P5G:1; + u8 rate_5G:1; + u8 rate_N5G:1; + u8 rate_10G:1; + + u8 eee_100M:1; + u8 eee_1G:1; + u8 eee_2P5G:1; + u8 eee_5G:1; + u8 eee_10G:1; + u8 rsvd3:3; + + u8 pause_rx:1; + u8 pause_tx:1; + u8 rsvd4:1; + u8 downshift:1; + u8 downshift_retry:4; +}; + +struct link_control_s { + u8 mode:4; + u8 disable_crc_corruption:1; + u8 discard_short_frames:1; + u8 flow_control_mode:1; + u8 disable_length_check:1; + + u8 discard_errored_frames:1; + u8 control_frame_enable:1; + u8 enable_tx_padding:1; + u8 enable_crc_forwarding:1; + u8 enable_frame_padding_removal_rx: 1; + u8 promiscuous_mode: 1; + u8 rsvd:2; + + u16 rsvd2; +}; + +struct thermal_shutdown_s { + u8 enable:1; + u8 warning_enable:1; + u8 rsvd:6; + + u8 shutdown_temperature; + u8 cold_temperature; + u8 warning_temperature; +}; + +struct mac_address_s { + u8 mac_address[6]; +}; + +struct mac_address_aligned_s { + struct mac_address_s aligned; + u16 rsvd; +}; + +struct sleep_proxy_s { + struct wake_on_lan_s { + u8 wake_on_magic_packet:1; + u8 wake_on_pattern:1; + u8 wake_on_link_up:1; + u8 wake_on_link_down:1; + u8 wake_on_ping:1; + u8 wake_on_timer:1; + u8 rsvd:2; + + u8 rsvd2; + u16 rsvd3; + + u32 link_up_timeout; + u32 link_down_timeout; + u32 timer; + } wake_on_lan; + + struct { + u32 mask[4]; + u32 crc32; + } wake_up_pattern[8]; + + struct __packed { + u8 arp_responder:1; + u8 echo_responder:1; + u8 igmp_client:1; + u8 echo_truncate:1; + u8 address_guard:1; + u8 ignore_fragmented:1; + u8 rsvd:2; + + u16 echo_max_len; + u8 rsvd2; + } ipv4_offload; + + u32 ipv4_offload_addr[8]; + u32 reserved[8]; + + struct __packed { + u8 ns_responder:1; + u8 echo_responder:1; + u8 mld_client:1; + u8 echo_truncate:1; + u8 address_guard:1; + u8 rsvd:3; + + u16 echo_max_len; + u8 rsvd2; + } ipv6_offload; + + u32 ipv6_offload_addr[16][4]; + + struct { + u16 port[16]; + } tcp_port_offload; + + struct { + u16 port[16]; + } udp_port_offload; + + struct { + u32 retry_count; + u32 retry_interval; + } ka4_offload; + + struct { + u32 timeout; + u16 local_port; + u16 remote_port; + u8 remote_mac_addr[6]; + u16 rsvd; + u32 rsvd2; + u32 rsvd3; + u16 rsvd4; + u16 win_size; + u32 seq_num; + u32 ack_num; + u32 local_ip; + u32 remote_ip; + } ka4_connection[16]; + + struct { + u32 retry_count; + u32 retry_interval; + } ka6_offload; + + struct { + u32 timeout; + u16 local_port; + u16 remote_port; + u8 remote_mac_addr[6]; + u16 rsvd; + u32 rsvd2; + u32 rsvd3; + u16 rsvd4; + u16 win_size; + u32 seq_num; + u32 ack_num; + u32 local_ip[4]; + u32 remote_ip[4]; + } ka6_connection[16]; + + struct { + u32 rr_count; + u32 rr_buf_len; + u32 idx_offset; + u32 rr__offset; + } mdns_offload; +}; + +struct pause_quanta_s { + u16 quanta_10M; + u16 threshold_10M; + u16 quanta_100M; + u16 threshold_100M; + u16 quanta_1G; + u16 threshold_1G; + u16 quanta_2P5G; + u16 threshold_2P5G; + u16 quanta_5G; + u16 threshold_5G; + u16 quanta_10G; + u16 threshold_10G; +}; + +struct data_buffer_status_s { + u32 data_offset; + u32 data_length; +}; + +struct device_caps_s { + u8 finite_flashless:1; + u8 cable_diag:1; + u8 ncsi:1; + u8 avb:1; + u8 rsvd:4; + + u8 rsvd2; + u16 rsvd3; + u32 rsvd4; +}; + +struct version_s { + struct bundle_version_t { + u8 major; + u8 minor; + u16 build; + } bundle; + struct mac_version_t { + u8 major; + u8 minor; + u16 build; + } mac; + struct phy_version_t { + u8 major; + u8 minor; + u16 build; + } phy; + u32 rsvd; +}; + +struct link_status_s { + u8 link_state:4; + u8 link_rate:4; + + u8 pause_tx:1; + u8 pause_rx:1; + u8 eee:1; + u8 duplex:1; + u8 rsvd:4; + + u16 rsvd2; +}; + +struct wol_status_s { + u8 wake_count; + u8 wake_reason; + + u16 wake_up_packet_length :12; + u16 wake_up_pattern_number :3; + u16 rsvd:1; + + u32 wake_up_packet[379]; +}; + +struct mac_health_monitor_s { + u8 mac_ready:1; + u8 mac_fault:1; + u8 mac_flashless_finished:1; + u8 rsvd:5; + + u8 mac_temperature; + u16 mac_heart_beat; + u16 mac_fault_code; + u16 rsvd2; +}; + +struct phy_health_monitor_s { + u8 phy_ready:1; + u8 phy_fault:1; + u8 phy_hot_warning:1; + u8 rsvd:5; + + u8 phy_temperature; + u16 phy_heart_beat; + u16 phy_fault_code; + u16 rsvd2; +}; + +struct device_link_caps_s { + u8 rsvd:3; + u8 internal_loopback:1; + u8 external_loopback:1; + u8 rate_10M_hd:1; + u8 rate_100M_hd:1; + u8 rate_1G_hd:1; + + u8 rate_10M:1; + u8 rate_100M:1; + u8 rate_1G:1; + u8 rate_2P5G:1; + u8 rate_N2P5G:1; + u8 rate_5G:1; + u8 rate_N5G:1; + u8 rate_10G:1; + + u8 rsvd3:1; + u8 eee_100M:1; + u8 eee_1G:1; + u8 eee_2P5G:1; + u8 rsvd4:1; + u8 eee_5G:1; + u8 rsvd5:1; + u8 eee_10G:1; + + u8 pause_rx:1; + u8 pause_tx:1; + u8 pfc:1; + u8 downshift:1; + u8 downshift_retry:4; +}; + +struct sleep_proxy_caps_s { + u8 ipv4_offload:1; + u8 ipv6_offload:1; + u8 tcp_port_offload:1; + u8 udp_port_offload:1; + u8 ka4_offload:1; + u8 ka6_offload:1; + u8 mdns_offload:1; + u8 wake_on_ping:1; + + u8 wake_on_magic_packet:1; + u8 wake_on_pattern:1; + u8 wake_on_timer:1; + u8 wake_on_link:1; + u8 wake_patterns_count:4; + + u8 ipv4_count; + u8 ipv6_count; + + u8 tcp_port_offload_count; + u8 udp_port_offload_count; + + u8 tcp4_ka_count; + u8 tcp6_ka_count; + + u8 igmp_offload:1; + u8 mld_offload:1; + u8 rsvd:6; + + u8 rsvd2; + u16 rsvd3; +}; + +struct lkp_link_caps_s { + u8 rsvd:5; + u8 rate_10M_hd:1; + u8 rate_100M_hd:1; + u8 rate_1G_hd:1; + + u8 rate_10M:1; + u8 rate_100M:1; + u8 rate_1G:1; + u8 rate_2P5G:1; + u8 rate_N2P5G:1; + u8 rate_5G:1; + u8 rate_N5G:1; + u8 rate_10G:1; + + u8 rsvd2:1; + u8 eee_100M:1; + u8 eee_1G:1; + u8 eee_2P5G:1; + u8 rsvd3:1; + u8 eee_5G:1; + u8 rsvd4:1; + u8 eee_10G:1; + + u8 pause_rx:1; + u8 pause_tx:1; + u8 rsvd5:6; +}; + +struct core_dump_s { + u32 reg0; + u32 reg1; + u32 reg2; + + u32 hi; + u32 lo; + + u32 regs[32]; +}; + +struct trace_s { + u32 sync_counter; + u32 mem_buffer[0x1ff]; +}; + +struct cable_diag_control_s { + u8 toggle :1; + u8 rsvd:7; + + u8 wait_timeout_sec; + u16 rsvd2; +}; + +struct cable_diag_lane_data_s { + u8 result_code; + u8 dist; + u8 far_dist; + u8 rsvd; +}; + +struct cable_diag_status_s { + struct cable_diag_lane_data_s lane_data[4]; + u8 transact_id; + u8 status:4; + u8 rsvd:4; + u16 rsvd2; +}; + +struct statistics_s { + struct { + u32 link_up; + u32 link_down; + } link; + + struct { + u64 tx_unicast_octets; + u64 tx_multicast_octets; + u64 tx_broadcast_octets; + u64 rx_unicast_octets; + u64 rx_multicast_octets; + u64 rx_broadcast_octets; + + u32 tx_unicast_frames; + u32 tx_multicast_frames; + u32 tx_broadcast_frames; + u32 tx_errors; + + u32 rx_unicast_frames; + u32 rx_multicast_frames; + u32 rx_broadcast_frames; + u32 rx_dropped_frames; + u32 rx_error_frames; + + u32 tx_good_frames; + u32 rx_good_frames; + u32 reserve_fw_gap; + } msm; + u32 main_loop_cycles; + u32 reserve_fw_gap; +}; + +struct filter_caps_s { + u8 l2_filters_base_index:6; + u8 flexible_filter_mask:2; + u8 l2_filter_count; + u8 ethertype_filter_base_index; + u8 ethertype_filter_count; + + u8 vlan_filter_base_index; + u8 vlan_filter_count; + u8 l3_ip4_filter_base_index:4; + u8 l3_ip4_filter_count:4; + u8 l3_ip6_filter_base_index:4; + u8 l3_ip6_filter_count:4; + + u8 l4_filter_base_index:4; + u8 l4_filter_count:4; + u8 l4_flex_filter_base_index:4; + u8 l4_flex_filter_count:4; + u8 rslv_tbl_base_index; + u8 rslv_tbl_count; +}; + +struct request_policy_s { + struct { + u8 all:1; + u8 mcast:1; + u8 rx_queue_tc_index:5; + u8 queue_or_tc:1; + } promisc; + + struct { + u8 accept:1; + u8 rsvd:1; + u8 rx_queue_tc_index:5; + u8 queue_or_tc:1; + } bcast; + + struct { + u8 accept:1; + u8 rsvd:1; + u8 rx_queue_tc_index:5; + u8 queue_or_tc:1; + } mcast; + + u8 rsvd:8; +}; + +struct fw_interface_in { + u32 mtu; + u32 rsvd1; + struct mac_address_aligned_s mac_address; + struct link_control_s link_control; + u32 rsvd2; + struct link_options_s link_options; + u32 rsvd3; + struct thermal_shutdown_s thermal_shutdown; + u32 rsvd4; + struct sleep_proxy_s sleep_proxy; + u32 rsvd5; + struct pause_quanta_s pause_quanta[8]; + struct cable_diag_control_s cable_diag_control; + u32 rsvd6; + struct data_buffer_status_s data_buffer_status; + u32 rsvd7; + struct request_policy_s request_policy; +}; + +struct transaction_counter_s { + u16 transaction_cnt_a; + u16 transaction_cnt_b; +}; + +struct management_status_s { + struct mac_address_s mac_address; + u16 vlan; + + struct{ + u32 enable : 1; + u32 rsvd:31; + } flags; + + u32 rsvd1; + u32 rsvd2; + u32 rsvd3; + u32 rsvd4; + u32 rsvd5; +}; + +struct fw_interface_out { + struct transaction_counter_s transaction_id; + struct version_s version; + struct link_status_s link_status; + struct wol_status_s wol_status; + u32 rsvd; + u32 rsvd2; + struct mac_health_monitor_s mac_health_monitor; + u32 rsvd3; + u32 rsvd4; + struct phy_health_monitor_s phy_health_monitor; + u32 rsvd5; + u32 rsvd6; + struct cable_diag_status_s cable_diag_status; + u32 rsvd7; + struct device_link_caps_s device_link_caps; + u32 rsvd8; + struct sleep_proxy_caps_s sleep_proxy_caps; + u32 rsvd9; + struct lkp_link_caps_s lkp_link_caps; + u32 rsvd10; + struct core_dump_s core_dump; + u32 rsvd11; + struct statistics_s stats; + u32 rsvd12; + struct filter_caps_s filter_caps; + struct device_caps_s device_caps; + u32 rsvd13; + struct management_status_s management_status; + u32 reserve[21]; + struct trace_s trace; +}; + +#define AQ_A2_FW_LINK_RATE_INVALID 0 +#define AQ_A2_FW_LINK_RATE_10M 1 +#define AQ_A2_FW_LINK_RATE_100M 2 +#define AQ_A2_FW_LINK_RATE_1G 3 +#define AQ_A2_FW_LINK_RATE_2G5 4 +#define AQ_A2_FW_LINK_RATE_5G 5 +#define AQ_A2_FW_LINK_RATE_10G 6 + +#define AQ_HOST_MODE_INVALID 0U +#define AQ_HOST_MODE_ACTIVE 1U +#define AQ_HOST_MODE_SLEEP_PROXY 2U +#define AQ_HOST_MODE_LOW_POWER 3U +#define AQ_HOST_MODE_SHUTDOWN 4U + +int hw_atl2_utils_initfw(struct aq_hw_s *self, const struct aq_fw_ops **fw_ops); + +int hw_atl2_utils_soft_reset(struct aq_hw_s *self); + +u32 hw_atl2_utils_get_fw_version(struct aq_hw_s *self); + +int hw_atl2_utils_get_action_resolve_table_caps(struct aq_hw_s *self, + u8 *base_index, u8 *count); + +extern const struct aq_fw_ops aq_a2_fw_ops; + +#endif /* HW_ATL2_UTILS_H */ diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils_fw.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils_fw.c new file mode 100644 index 000000000000..0ffc33bd67d0 --- /dev/null +++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl2/hw_atl2_utils_fw.c @@ -0,0 +1,320 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Atlantic Network Driver + * Copyright (C) 2020 Marvell International Ltd. + */ + +#include <linux/iopoll.h> + +#include "aq_hw.h" +#include "aq_hw_utils.h" +#include "hw_atl/hw_atl_llh.h" +#include "hw_atl2_utils.h" +#include "hw_atl2_llh.h" +#include "hw_atl2_internal.h" + +#define AQ_A2_FW_READ_TRY_MAX 1000 + +#define hw_atl2_shared_buffer_write(HW, ITEM, VARIABLE) \ + hw_atl2_mif_shared_buf_write(HW,\ + (offsetof(struct fw_interface_in, ITEM) / sizeof(u32)),\ + (u32 *)&(VARIABLE), sizeof(VARIABLE) / sizeof(u32)) + +#define hw_atl2_shared_buffer_get(HW, ITEM, VARIABLE) \ + hw_atl2_mif_shared_buf_get(HW, \ + (offsetof(struct fw_interface_in, ITEM) / sizeof(u32)),\ + (u32 *)&(VARIABLE), \ + sizeof(VARIABLE) / sizeof(u32)) + +/* This should never be used on non atomic fields, + * treat any > u32 read as non atomic. + */ +#define hw_atl2_shared_buffer_read(HW, ITEM, VARIABLE) \ +{\ + BUILD_BUG_ON_MSG((offsetof(struct fw_interface_out, ITEM) % \ + sizeof(u32)) != 0,\ + "Non aligned read " # ITEM);\ + BUILD_BUG_ON_MSG(sizeof(VARIABLE) > sizeof(u32),\ + "Non atomic read " # ITEM);\ + hw_atl2_mif_shared_buf_read(HW, \ + (offsetof(struct fw_interface_out, ITEM) / sizeof(u32)),\ + (u32 *)&(VARIABLE), sizeof(VARIABLE) / sizeof(u32));\ +} + +#define hw_atl2_shared_buffer_read_safe(HW, ITEM, DATA) \ + hw_atl2_shared_buffer_read_block((HW), \ + (offsetof(struct fw_interface_out, ITEM) / sizeof(u32)),\ + sizeof(((struct fw_interface_out *)0)->ITEM) / sizeof(u32),\ + (DATA)) + +static int hw_atl2_shared_buffer_read_block(struct aq_hw_s *self, + u32 offset, u32 dwords, void *data) +{ + struct transaction_counter_s tid1, tid2; + int cnt = 0; + + do { + do { + hw_atl2_shared_buffer_read(self, transaction_id, tid1); + cnt++; + if (cnt > AQ_A2_FW_READ_TRY_MAX) + return -ETIME; + if (tid1.transaction_cnt_a != tid1.transaction_cnt_b) + udelay(1); + } while (tid1.transaction_cnt_a != tid1.transaction_cnt_b); + + hw_atl2_mif_shared_buf_read(self, offset, (u32 *)data, dwords); + + hw_atl2_shared_buffer_read(self, transaction_id, tid2); + + cnt++; + if (cnt > AQ_A2_FW_READ_TRY_MAX) + return -ETIME; + } while (tid2.transaction_cnt_a != tid2.transaction_cnt_b || + tid1.transaction_cnt_a != tid2.transaction_cnt_a); + + return 0; +} + +static inline int hw_atl2_shared_buffer_finish_ack(struct aq_hw_s *self) +{ + u32 val; + int err; + + hw_atl2_mif_host_finished_write_set(self, 1U); + err = readx_poll_timeout_atomic(hw_atl2_mif_mcp_finished_read_get, + self, val, val == 0U, + 100, 100000U); + WARN(err, "hw_atl2_shared_buffer_finish_ack"); + + return err; +} + +static int aq_a2_fw_init(struct aq_hw_s *self) +{ + struct link_control_s link_control; + u32 mtu; + u32 val; + int err; + + hw_atl2_shared_buffer_get(self, link_control, link_control); + link_control.mode = AQ_HOST_MODE_ACTIVE; + hw_atl2_shared_buffer_write(self, link_control, link_control); + + hw_atl2_shared_buffer_get(self, mtu, mtu); + mtu = HW_ATL2_MTU_JUMBO; + hw_atl2_shared_buffer_write(self, mtu, mtu); + + hw_atl2_mif_host_finished_write_set(self, 1U); + err = readx_poll_timeout_atomic(hw_atl2_mif_mcp_finished_read_get, + self, val, val == 0U, + 100, 5000000U); + WARN(err, "hw_atl2_shared_buffer_finish_ack"); + + return err; +} + +static int aq_a2_fw_deinit(struct aq_hw_s *self) +{ + struct link_control_s link_control; + + hw_atl2_shared_buffer_get(self, link_control, link_control); + link_control.mode = AQ_HOST_MODE_SHUTDOWN; + hw_atl2_shared_buffer_write(self, link_control, link_control); + + return hw_atl2_shared_buffer_finish_ack(self); +} + +static void a2_link_speed_mask2fw(u32 speed, + struct link_options_s *link_options) +{ + link_options->rate_10G = !!(speed & AQ_NIC_RATE_10G); + link_options->rate_5G = !!(speed & AQ_NIC_RATE_5G); + link_options->rate_N5G = !!(speed & AQ_NIC_RATE_5GSR); + link_options->rate_2P5G = !!(speed & AQ_NIC_RATE_2G5); + link_options->rate_N2P5G = link_options->rate_2P5G; + link_options->rate_1G = !!(speed & AQ_NIC_RATE_1G); + link_options->rate_100M = !!(speed & AQ_NIC_RATE_100M); + link_options->rate_10M = !!(speed & AQ_NIC_RATE_10M); +} + +static int aq_a2_fw_set_link_speed(struct aq_hw_s *self, u32 speed) +{ + struct link_options_s link_options; + + hw_atl2_shared_buffer_get(self, link_options, link_options); + link_options.link_up = 1U; + a2_link_speed_mask2fw(speed, &link_options); + hw_atl2_shared_buffer_write(self, link_options, link_options); + + return hw_atl2_shared_buffer_finish_ack(self); +} + +static int aq_a2_fw_set_state(struct aq_hw_s *self, + enum hal_atl_utils_fw_state_e state) +{ + struct link_options_s link_options; + + hw_atl2_shared_buffer_get(self, link_options, link_options); + + switch (state) { + case MPI_INIT: + link_options.link_up = 1U; + break; + case MPI_DEINIT: + link_options.link_up = 0U; + break; + case MPI_RESET: + case MPI_POWER: + /* No actions */ + break; + } + + hw_atl2_shared_buffer_write(self, link_options, link_options); + + return hw_atl2_shared_buffer_finish_ack(self); +} + +static int aq_a2_fw_update_link_status(struct aq_hw_s *self) +{ + struct link_status_s link_status; + + hw_atl2_shared_buffer_read(self, link_status, link_status); + + switch (link_status.link_rate) { + case AQ_A2_FW_LINK_RATE_10G: + self->aq_link_status.mbps = 10000; + break; + case AQ_A2_FW_LINK_RATE_5G: + self->aq_link_status.mbps = 5000; + break; + case AQ_A2_FW_LINK_RATE_2G5: + self->aq_link_status.mbps = 2500; + break; + case AQ_A2_FW_LINK_RATE_1G: + self->aq_link_status.mbps = 1000; + break; + case AQ_A2_FW_LINK_RATE_100M: + self->aq_link_status.mbps = 100; + break; + case AQ_A2_FW_LINK_RATE_10M: + self->aq_link_status.mbps = 10; + break; + default: + self->aq_link_status.mbps = 0; + } + + return 0; +} + +static int aq_a2_fw_get_mac_permanent(struct aq_hw_s *self, u8 *mac) +{ + struct mac_address_aligned_s mac_address; + + hw_atl2_shared_buffer_get(self, mac_address, mac_address); + ether_addr_copy(mac, (u8 *)mac_address.aligned.mac_address); + + return 0; +} + +static int aq_a2_fw_update_stats(struct aq_hw_s *self) +{ + struct hw_atl2_priv *priv = (struct hw_atl2_priv *)self->priv; + struct statistics_s stats; + + hw_atl2_shared_buffer_read_safe(self, stats, &stats); + +#define AQ_SDELTA(_N_, _F_) (self->curr_stats._N_ += \ + stats.msm._F_ - priv->last_stats.msm._F_) + + if (self->aq_link_status.mbps) { + AQ_SDELTA(uprc, rx_unicast_frames); + AQ_SDELTA(mprc, rx_multicast_frames); + AQ_SDELTA(bprc, rx_broadcast_frames); + AQ_SDELTA(erpr, rx_error_frames); + + AQ_SDELTA(uptc, tx_unicast_frames); + AQ_SDELTA(mptc, tx_multicast_frames); + AQ_SDELTA(bptc, tx_broadcast_frames); + AQ_SDELTA(erpt, tx_errors); + + AQ_SDELTA(ubrc, rx_unicast_octets); + AQ_SDELTA(ubtc, tx_unicast_octets); + AQ_SDELTA(mbrc, rx_multicast_octets); + AQ_SDELTA(mbtc, tx_multicast_octets); + AQ_SDELTA(bbrc, rx_broadcast_octets); + AQ_SDELTA(bbtc, tx_broadcast_octets); + } +#undef AQ_SDELTA + self->curr_stats.dma_pkt_rc = + hw_atl_stats_rx_dma_good_pkt_counter_get(self); + self->curr_stats.dma_pkt_tc = + hw_atl_stats_tx_dma_good_pkt_counter_get(self); + self->curr_stats.dma_oct_rc = + hw_atl_stats_rx_dma_good_octet_counter_get(self); + self->curr_stats.dma_oct_tc = + hw_atl_stats_tx_dma_good_octet_counter_get(self); + self->curr_stats.dpc = hw_atl_rpb_rx_dma_drop_pkt_cnt_get(self); + + memcpy(&priv->last_stats, &stats, sizeof(stats)); + + return 0; +} + +static int aq_a2_fw_renegotiate(struct aq_hw_s *self) +{ + struct link_options_s link_options; + int err; + + hw_atl2_shared_buffer_get(self, link_options, link_options); + link_options.link_renegotiate = 1U; + hw_atl2_shared_buffer_write(self, link_options, link_options); + + err = hw_atl2_shared_buffer_finish_ack(self); + + /* We should put renegotiate status back to zero + * after command completes + */ + link_options.link_renegotiate = 0U; + hw_atl2_shared_buffer_write(self, link_options, link_options); + + return err; +} + +u32 hw_atl2_utils_get_fw_version(struct aq_hw_s *self) +{ + struct version_s version; + + hw_atl2_shared_buffer_read_safe(self, version, &version); + + /* A2 FW version is stored in reverse order */ + return version.mac.major << 24 | + version.mac.minor << 16 | + version.mac.build; +} + +int hw_atl2_utils_get_action_resolve_table_caps(struct aq_hw_s *self, + u8 *base_index, u8 *count) +{ + struct filter_caps_s filter_caps; + int err; + + err = hw_atl2_shared_buffer_read_safe(self, filter_caps, &filter_caps); + if (err) + return err; + + *base_index = filter_caps.rslv_tbl_base_index; + *count = filter_caps.rslv_tbl_count; + return 0; +} + +const struct aq_fw_ops aq_a2_fw_ops = { + .init = aq_a2_fw_init, + .deinit = aq_a2_fw_deinit, + .reset = NULL, + .renegotiate = aq_a2_fw_renegotiate, + .get_mac_permanent = aq_a2_fw_get_mac_permanent, + .set_link_speed = aq_a2_fw_set_link_speed, + .set_state = aq_a2_fw_set_state, + .update_link_status = aq_a2_fw_update_link_status, + .update_stats = aq_a2_fw_update_stats, +}; diff --git a/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c index fbe9d88b13c7..36c7cf05630a 100644 --- a/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c +++ b/drivers/net/ethernet/aquantia/atlantic/macsec/macsec_api.c @@ -846,8 +846,7 @@ static int get_ingress_sakey_record(struct aq_hw_s *hw, rec->key[7] = packed_record[14]; rec->key[7] |= packed_record[15] << 16; - rec->key_len = (rec->key_len & 0xFFFFFFFC) | - (packed_record[16] & 0x3); + rec->key_len = packed_record[16] & 0x3; return 0; } @@ -1158,6 +1157,7 @@ static int set_egress_ctlf_record(struct aq_hw_s *hw, packed_record[0] = rec->sa_da[0] & 0xFFFF; packed_record[1] = (rec->sa_da[0] >> 16) & 0xFFFF; + packed_record[2] = rec->sa_da[1] & 0xFFFF; packed_record[3] = rec->eth_type & 0xFFFF; @@ -1552,7 +1552,7 @@ static int set_egress_sc_record(struct aq_hw_s *hw, packed_record[5] |= (rec->sak_len & 0x3) << 4; - packed_record[7] |= (rec->valid & 0x1) << 15; + packed_record[7] = (rec->valid & 0x1) << 15; return set_raw_egress_record(hw, packed_record, 8, 2, ROWOFFSET_EGRESSSCRECORD + table_index); diff --git a/drivers/net/ethernet/atheros/ag71xx.c b/drivers/net/ethernet/atheros/ag71xx.c index 02b7705393ca..112edbd30823 100644 --- a/drivers/net/ethernet/atheros/ag71xx.c +++ b/drivers/net/ethernet/atheros/ag71xx.c @@ -871,13 +871,40 @@ static void ag71xx_mac_validate(struct phylink_config *config, unsigned long *supported, struct phylink_link_state *state) { + struct ag71xx *ag = netdev_priv(to_net_dev(config->dev)); __ETHTOOL_DECLARE_LINK_MODE_MASK(mask) = { 0, }; - if (state->interface != PHY_INTERFACE_MODE_NA && - state->interface != PHY_INTERFACE_MODE_GMII && - state->interface != PHY_INTERFACE_MODE_MII) { - bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS); - return; + switch (state->interface) { + case PHY_INTERFACE_MODE_NA: + break; + case PHY_INTERFACE_MODE_MII: + if ((ag71xx_is(ag, AR9330) && ag->mac_idx == 0) || + ag71xx_is(ag, AR9340) || + ag71xx_is(ag, QCA9530) || + (ag71xx_is(ag, QCA9550) && ag->mac_idx == 1)) + break; + goto unsupported; + case PHY_INTERFACE_MODE_GMII: + if ((ag71xx_is(ag, AR9330) && ag->mac_idx == 1) || + (ag71xx_is(ag, AR9340) && ag->mac_idx == 1) || + (ag71xx_is(ag, QCA9530) && ag->mac_idx == 1)) + break; + goto unsupported; + case PHY_INTERFACE_MODE_SGMII: + if (ag71xx_is(ag, QCA9550) && ag->mac_idx == 0) + break; + goto unsupported; + case PHY_INTERFACE_MODE_RMII: + if (ag71xx_is(ag, AR9340) && ag->mac_idx == 0) + break; + goto unsupported; + case PHY_INTERFACE_MODE_RGMII: + if ((ag71xx_is(ag, AR9340) && ag->mac_idx == 0) || + (ag71xx_is(ag, QCA9550) && ag->mac_idx == 1)) + break; + goto unsupported; + default: + goto unsupported; } phylink_set(mask, MII); @@ -889,6 +916,8 @@ static void ag71xx_mac_validate(struct phylink_config *config, phylink_set(mask, 100baseT_Full); if (state->interface == PHY_INTERFACE_MODE_NA || + state->interface == PHY_INTERFACE_MODE_SGMII || + state->interface == PHY_INTERFACE_MODE_RGMII || state->interface == PHY_INTERFACE_MODE_GMII) { phylink_set(mask, 1000baseT_Full); phylink_set(mask, 1000baseX_Full); @@ -898,6 +927,10 @@ static void ag71xx_mac_validate(struct phylink_config *config, __ETHTOOL_LINK_MODE_MASK_NBITS); bitmap_and(state->advertising, state->advertising, mask, __ETHTOOL_LINK_MODE_MASK_NBITS); + + return; +unsupported: + bitmap_zero(supported, __ETHTOOL_LINK_MODE_MASK_NBITS); } static void ag71xx_mac_pcs_get_state(struct phylink_config *config, diff --git a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c index 00bd7bd55794..decab9a8e4a8 100644 --- a/drivers/net/ethernet/atheros/atl1c/atl1c_main.c +++ b/drivers/net/ethernet/atheros/atl1c/atl1c_main.c @@ -1186,7 +1186,7 @@ static void atl1c_start_mac(struct atl1c_adapter *adapter) struct atl1c_hw *hw = &adapter->hw; u32 mac, txq, rxq; - hw->mac_duplex = adapter->link_duplex == FULL_DUPLEX ? true : false; + hw->mac_duplex = adapter->link_duplex == FULL_DUPLEX; hw->mac_speed = adapter->link_speed == SPEED_1000 ? atl1c_mac_speed_1000 : atl1c_mac_speed_10_100; @@ -2449,12 +2449,6 @@ static int atl1c_resume(struct device *dev) atl1c_reset_mac(&adapter->hw); atl1c_phy_init(&adapter->hw); -#if 0 - AT_READ_REG(&adapter->hw, REG_PM_CTRLSTAT, &pm_data); - pm_data &= ~PM_CTRLSTAT_PME_EN; - AT_WRITE_REG(&adapter->hw, REG_PM_CTRLSTAT, pm_data); -#endif - netif_device_attach(netdev); if (netif_running(netdev)) atl1c_up(adapter); diff --git a/drivers/net/ethernet/atheros/atlx/atl1.c b/drivers/net/ethernet/atheros/atlx/atl1.c index 271e7034fa70..b35fcfcd692d 100644 --- a/drivers/net/ethernet/atheros/atlx/atl1.c +++ b/drivers/net/ethernet/atheros/atlx/atl1.c @@ -1042,7 +1042,7 @@ static s32 atl1_setup_ring_resources(struct atl1_adapter *adapter) * each ring/block may need up to 8 bytes for alignment, hence the * additional 40 bytes tacked onto the end. */ - ring_header->size = size = + ring_header->size = sizeof(struct tx_packet_desc) * tpd_ring->count + sizeof(struct rx_free_desc) * rfd_ring->count + sizeof(struct rx_return_desc) * rrd_ring->count diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c index 2c6ba046d2a8..17ae6df90723 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c @@ -1145,7 +1145,7 @@ static void bnx2x_dcbx_get_num_pg_traf_type(struct bnx2x *bp, break; } } - if (false == pg_found) { + if (!pg_found) { data[help_data->num_of_pg].pg = add_pg; data[help_data->num_of_pg].pg_priority = (1 << ttp[add_traf_type]); @@ -1155,7 +1155,7 @@ static void bnx2x_dcbx_get_num_pg_traf_type(struct bnx2x *bp, } DP(BNX2X_MSG_DCB, "add_traf_type %d pg_found %s num_of_pg %d\n", - add_traf_type, (false == pg_found) ? "NO" : "YES", + add_traf_type, !pg_found ? "NO" : "YES", help_data->num_of_pg); } } @@ -1544,8 +1544,7 @@ static void bnx2x_dcbx_2cos_limit_cee_three_pg_to_cos_params( if (pg_entry < DCBX_MAX_NUM_PG_BW_ENTRIES) { entry = 0; - if (i == (num_of_pri-1) && - false == b_found_strict) + if (i == (num_of_pri-1) && !b_found_strict) /* last entry will be handled separately * If no priority is strict than last * entry goes to last queue. diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c index 517caedc0a87..1426c691c7c4 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c @@ -3085,6 +3085,7 @@ static int bnx2x_bsc_read(struct link_params *params, u8 xfer_cnt, u32 *data_array) { + u64 t0, delta; u32 val, i; int rc = 0; @@ -3114,17 +3115,18 @@ static int bnx2x_bsc_read(struct link_params *params, REG_WR(bp, MCP_REG_MCPR_IMC_COMMAND, val); /* Poll for completion */ - i = 0; + t0 = ktime_get_ns(); val = REG_RD(bp, MCP_REG_MCPR_IMC_COMMAND); while (((val >> MCPR_IMC_COMMAND_IMC_STATUS_BITSHIFT) & 0x3) != 1) { - udelay(10); - val = REG_RD(bp, MCP_REG_MCPR_IMC_COMMAND); - if (i++ > 1000) { - DP(NETIF_MSG_LINK, "wr 0 byte timed out after %d try\n", - i); + delta = ktime_get_ns() - t0; + if (delta > 10 * NSEC_PER_MSEC) { + DP(NETIF_MSG_LINK, "wr 0 byte timed out after %Lu ns\n", + delta); rc = -EFAULT; break; } + usleep_range(10, 20); + val = REG_RD(bp, MCP_REG_MCPR_IMC_COMMAND); } if (rc == -EFAULT) return rc; @@ -3138,16 +3140,18 @@ static int bnx2x_bsc_read(struct link_params *params, REG_WR(bp, MCP_REG_MCPR_IMC_COMMAND, val); /* Poll for completion */ - i = 0; + t0 = ktime_get_ns(); val = REG_RD(bp, MCP_REG_MCPR_IMC_COMMAND); while (((val >> MCPR_IMC_COMMAND_IMC_STATUS_BITSHIFT) & 0x3) != 1) { - udelay(10); - val = REG_RD(bp, MCP_REG_MCPR_IMC_COMMAND); - if (i++ > 1000) { - DP(NETIF_MSG_LINK, "rd op timed out after %d try\n", i); + delta = ktime_get_ns() - t0; + if (delta > 10 * NSEC_PER_MSEC) { + DP(NETIF_MSG_LINK, "rd op timed out after %Lu ns\n", + delta); rc = -EFAULT; break; } + usleep_range(10, 20); + val = REG_RD(bp, MCP_REG_MCPR_IMC_COMMAND); } if (rc == -EFAULT) return rc; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c index 5097a44686b3..b4476f44e386 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c @@ -331,27 +331,6 @@ bnx2x_vf_set_igu_info(struct bnx2x *bp, u8 igu_sb_id, u8 abs_vfid) BP_VFDB(bp)->vf_sbs_pool++; } -static inline void bnx2x_vf_vlan_credit(struct bnx2x *bp, - struct bnx2x_vlan_mac_obj *obj, - atomic_t *counter) -{ - struct list_head *pos; - int read_lock; - int cnt = 0; - - read_lock = bnx2x_vlan_mac_h_read_lock(bp, obj); - if (read_lock) - DP(BNX2X_MSG_SP, "Failed to take vlan mac read head; continuing anyway\n"); - - list_for_each(pos, &obj->head) - cnt++; - - if (!read_lock) - bnx2x_vlan_mac_h_read_unlock(bp, obj); - - atomic_set(counter, cnt); -} - static int bnx2x_vf_vlan_mac_clear(struct bnx2x *bp, struct bnx2x_virtf *vf, int qid, bool drv_only, int type) { diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index d1a83716d934..c62589c266b2 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -1766,7 +1766,7 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr, rc = -EIO; if (rx_err & RX_CMPL_ERRORS_BUFFER_ERROR_MASK) { - bnapi->cp_ring.rx_buf_errors++; + bnapi->cp_ring.sw_stats.rx.rx_buf_errors++; if (!(bp->flags & BNXT_FLAG_CHIP_P5)) { netdev_warn(bp->dev, "RX buffer error %x\n", rx_err); @@ -1849,7 +1849,7 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr, } else { if (rxcmp1->rx_cmp_cfa_code_errors_v2 & RX_CMP_L4_CS_ERR_BITS) { if (dev->features & NETIF_F_RXCSUM) - bnapi->cp_ring.rx_l4_csum_errors++; + bnapi->cp_ring.sw_stats.rx.rx_l4_csum_errors++; } } @@ -4176,14 +4176,12 @@ static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len, int i, intr_process, rc, tmo_count; struct input *req = msg; u32 *data = msg; - __le32 *resp_len; u8 *valid; u16 cp_ring_id, len = 0; struct hwrm_err_output *resp = bp->hwrm_cmd_resp_addr; u16 max_req_len = BNXT_HWRM_MAX_REQ_LEN; struct hwrm_short_input short_input = {0}; u32 doorbell_offset = BNXT_GRCPF_REG_CHIMP_COMM_TRIGGER; - u8 *resp_addr = (u8 *)bp->hwrm_cmd_resp_addr; u32 bar_offset = BNXT_GRCPF_REG_CHIMP_COMM; u16 dst = BNXT_HWRM_CHNL_CHIMP; @@ -4201,7 +4199,6 @@ static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len, bar_offset = BNXT_GRCPF_REG_KONG_COMM; doorbell_offset = BNXT_GRCPF_REG_KONG_COMM_TRIGGER; resp = bp->hwrm_cmd_kong_resp_addr; - resp_addr = (u8 *)bp->hwrm_cmd_kong_resp_addr; } memset(resp, 0, PAGE_SIZE); @@ -4270,7 +4267,6 @@ static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len, tmo_count = HWRM_SHORT_TIMEOUT_COUNTER; timeout = timeout - HWRM_SHORT_MIN_TIMEOUT * HWRM_SHORT_TIMEOUT_COUNTER; tmo_count += DIV_ROUND_UP(timeout, HWRM_MIN_TIMEOUT); - resp_len = (__le32 *)(resp_addr + HWRM_RESP_LEN_OFFSET); if (intr_process) { u16 seq_id = bp->hwrm_intr_seq_id; @@ -4298,9 +4294,8 @@ static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len, le16_to_cpu(req->req_type)); return -EBUSY; } - len = (le32_to_cpu(*resp_len) & HWRM_RESP_LEN_MASK) >> - HWRM_RESP_LEN_SFT; - valid = resp_addr + len - 1; + len = le16_to_cpu(resp->resp_len); + valid = ((u8 *)resp) + len - 1; } else { int j; @@ -4311,8 +4306,7 @@ static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len, */ if (test_bit(BNXT_STATE_FW_FATAL_COND, &bp->state)) return -EBUSY; - len = (le32_to_cpu(*resp_len) & HWRM_RESP_LEN_MASK) >> - HWRM_RESP_LEN_SFT; + len = le16_to_cpu(resp->resp_len); if (len) break; /* on first few passes, just barely sleep */ @@ -4334,7 +4328,7 @@ static int bnxt_hwrm_do_send_msg(struct bnxt *bp, void *msg, u32 msg_len, } /* Last byte of resp contains valid bit */ - valid = resp_addr + len - 1; + valid = ((u8 *)resp) + len - 1; for (j = 0; j < HWRM_VALID_BIT_DELAY_USEC; j++) { /* make sure we read from updated DMA memory */ dma_rmb(); @@ -5045,8 +5039,7 @@ int bnxt_hwrm_vnic_cfg(struct bnxt *bp, u16 vnic_id) req.dflt_ring_grp = cpu_to_le16(bp->grp_info[grp_idx].fw_grp_id); req.lb_rule = cpu_to_le16(0xffff); vnic_mru: - req.mru = cpu_to_le16(bp->dev->mtu + ETH_HLEN + ETH_FCS_LEN + - VLAN_HLEN); + req.mru = cpu_to_le16(bp->dev->mtu + ETH_HLEN + VLAN_HLEN); req.vnic_id = cpu_to_le16(vnic->fw_vnic_id); #ifdef CONFIG_BNXT_SRIOV @@ -5356,9 +5349,9 @@ static void bnxt_set_db(struct bnxt *bp, struct bnxt_db_info *db, u32 ring_type, { if (bp->flags & BNXT_FLAG_CHIP_P5) { if (BNXT_PF(bp)) - db->doorbell = bp->bar1 + 0x10000; + db->doorbell = bp->bar1 + DB_PF_OFFSET_P5; else - db->doorbell = bp->bar1 + 0x4000; + db->doorbell = bp->bar1 + DB_VF_OFFSET_P5; switch (ring_type) { case HWRM_RING_ALLOC_TX: db->db_key64 = DBR_PATH_L2 | DBR_TYPE_SQ; @@ -6365,6 +6358,7 @@ static int bnxt_hwrm_func_qcfg(struct bnxt *bp) { struct hwrm_func_qcfg_input req = {0}; struct hwrm_func_qcfg_output *resp = bp->hwrm_cmd_resp_addr; + u32 min_db_offset = 0; u16 flags; int rc; @@ -6413,6 +6407,21 @@ static int bnxt_hwrm_func_qcfg(struct bnxt *bp) if (!bp->max_mtu) bp->max_mtu = BNXT_MAX_MTU; + if (bp->db_size) + goto func_qcfg_exit; + + if (bp->flags & BNXT_FLAG_CHIP_P5) { + if (BNXT_PF(bp)) + min_db_offset = DB_PF_OFFSET_P5; + else + min_db_offset = DB_VF_OFFSET_P5; + } + bp->db_size = PAGE_ALIGN(le16_to_cpu(resp->l2_doorbell_bar_size_kb) * + 1024); + if (!bp->db_size || bp->db_size > pci_resource_len(bp->pdev, 2) || + bp->db_size <= min_db_offset) + bp->db_size = pci_resource_len(bp->pdev, 2); + func_qcfg_exit: mutex_unlock(&bp->hwrm_cmd_lock); return rc; @@ -6434,23 +6443,13 @@ static int bnxt_hwrm_func_backing_store_qcaps(struct bnxt *bp) if (!rc) { struct bnxt_ctx_pg_info *ctx_pg; struct bnxt_ctx_mem_info *ctx; - int i; + int i, tqm_rings; ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); if (!ctx) { rc = -ENOMEM; goto ctx_err; } - ctx_pg = kzalloc(sizeof(*ctx_pg) * (bp->max_q + 1), GFP_KERNEL); - if (!ctx_pg) { - kfree(ctx); - rc = -ENOMEM; - goto ctx_err; - } - for (i = 0; i < bp->max_q + 1; i++, ctx_pg++) - ctx->tqm_mem[i] = ctx_pg; - - bp->ctx = ctx; ctx->qp_max_entries = le32_to_cpu(resp->qp_max_entries); ctx->qp_min_qp1_entries = le16_to_cpu(resp->qp_min_qp1_entries); ctx->qp_max_l2_entries = le16_to_cpu(resp->qp_max_l2_entries); @@ -6483,6 +6482,20 @@ static int bnxt_hwrm_func_backing_store_qcaps(struct bnxt *bp) ctx->tim_entry_size = le16_to_cpu(resp->tim_entry_size); ctx->tim_max_entries = le32_to_cpu(resp->tim_max_entries); ctx->ctx_kind_initializer = resp->ctx_kind_initializer; + ctx->tqm_fp_rings_count = resp->tqm_fp_rings_count; + if (!ctx->tqm_fp_rings_count) + ctx->tqm_fp_rings_count = bp->max_q; + + tqm_rings = ctx->tqm_fp_rings_count + 1; + ctx_pg = kcalloc(tqm_rings, sizeof(*ctx_pg), GFP_KERNEL); + if (!ctx_pg) { + kfree(ctx); + rc = -ENOMEM; + goto ctx_err; + } + for (i = 0; i < tqm_rings; i++, ctx_pg++) + ctx->tqm_mem[i] = ctx_pg; + bp->ctx = ctx; } else { rc = 0; } @@ -6735,7 +6748,7 @@ static void bnxt_free_ctx_mem(struct bnxt *bp) return; if (ctx->tqm_mem[0]) { - for (i = 0; i < bp->max_q + 1; i++) + for (i = 0; i < ctx->tqm_fp_rings_count + 1; i++) bnxt_free_ctx_pg_tbls(bp, ctx->tqm_mem[i]); kfree(ctx->tqm_mem[0]); ctx->tqm_mem[0] = NULL; @@ -6756,6 +6769,7 @@ static int bnxt_alloc_ctx_mem(struct bnxt *bp) struct bnxt_ctx_pg_info *ctx_pg; struct bnxt_ctx_mem_info *ctx; u32 mem_size, ena, entries; + u32 entries_sp, min; u32 num_mr, num_ah; u32 extra_srqs = 0; u32 extra_qps = 0; @@ -6845,14 +6859,17 @@ static int bnxt_alloc_ctx_mem(struct bnxt *bp) ena |= FUNC_BACKING_STORE_CFG_REQ_ENABLES_TIM; skip_rdma: - entries = ctx->qp_max_l2_entries + extra_qps; + min = ctx->tqm_min_entries_per_ring; + entries_sp = ctx->vnic_max_vnic_entries + ctx->qp_max_l2_entries + + 2 * (extra_qps + ctx->qp_min_qp1_entries) + min; + entries_sp = roundup(entries_sp, ctx->tqm_entries_multiple); + entries = ctx->qp_max_l2_entries + extra_qps + ctx->qp_min_qp1_entries; entries = roundup(entries, ctx->tqm_entries_multiple); - entries = clamp_t(u32, entries, ctx->tqm_min_entries_per_ring, - ctx->tqm_max_entries_per_ring); - for (i = 0; i < bp->max_q + 1; i++) { + entries = clamp_t(u32, entries, min, ctx->tqm_max_entries_per_ring); + for (i = 0; i < ctx->tqm_fp_rings_count + 1; i++) { ctx_pg = ctx->tqm_mem[i]; - ctx_pg->entries = entries; - mem_size = ctx->tqm_entry_size * entries; + ctx_pg->entries = i ? entries : entries_sp; + mem_size = ctx->tqm_entry_size * ctx_pg->entries; rc = bnxt_alloc_ctx_pg_tbls(bp, ctx_pg, mem_size, 1, false); if (rc) return rc; @@ -9310,7 +9327,7 @@ static void __bnxt_close_nic(struct bnxt *bp, bool irq_re_init, bnxt_free_skbs(bp); /* Save ring stats before shutdown */ - if (bp->bnapi) + if (bp->bnapi && irq_re_init) bnxt_get_ring_stats(bp, &bp->net_stats_prev); if (irq_re_init) { bnxt_free_irq(bp); @@ -10265,7 +10282,7 @@ static void bnxt_chk_missed_irq(struct bnxt *bp) bnxt_dbg_hwrm_ring_info_get(bp, DBG_RING_INFO_GET_REQ_RING_TYPE_L2_CMPL, fw_ring_id, &val[0], &val[1]); - cpr->missed_irqs++; + cpr->sw_stats.cmn.missed_irqs++; } } } @@ -10894,6 +10911,9 @@ static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev) bp->dev = dev; bp->pdev = pdev; + /* Doorbell BAR bp->bar1 is mapped after bnxt_fw_init_one_p2() + * determines the BAR size. + */ bp->bar0 = pci_ioremap_bar(pdev, 0); if (!bp->bar0) { dev_err(&pdev->dev, "Cannot map device registers, aborting\n"); @@ -10901,13 +10921,6 @@ static int bnxt_init_board(struct pci_dev *pdev, struct net_device *dev) goto init_err_release; } - bp->bar1 = pci_ioremap_bar(pdev, 2); - if (!bp->bar1) { - dev_err(&pdev->dev, "Cannot map doorbell registers, aborting\n"); - rc = -ENOMEM; - goto init_err_release; - } - bp->bar2 = pci_ioremap_bar(pdev, 4); if (!bp->bar2) { dev_err(&pdev->dev, "Cannot map bar4 registers, aborting\n"); @@ -11829,6 +11842,16 @@ static int bnxt_pcie_dsn_get(struct bnxt *bp, u8 dsn[]) return 0; } +static int bnxt_map_db_bar(struct bnxt *bp) +{ + if (!bp->db_size) + return -ENODEV; + bp->bar1 = pci_iomap(bp->pdev, 2, bp->db_size); + if (!bp->bar1) + return -ENOMEM; + return 0; +} + static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { struct net_device *dev; @@ -11889,6 +11912,13 @@ static int bnxt_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (rc) goto init_err_pci_clean; + rc = bnxt_map_db_bar(bp); + if (rc) { + dev_err(&pdev->dev, "Cannot map doorbell BAR rc = %d, aborting\n", + rc); + goto init_err_pci_clean; + } + dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM | NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 | NETIF_F_GSO_UDP_TUNNEL | NETIF_F_GSO_GRE | diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h index f6a3250ef1c5..9e173d74b72a 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h @@ -537,6 +537,9 @@ struct nqe_cn { #define DBR_TYPE_NQ_ARM (0xbULL << 60) #define DBR_TYPE_NULL (0xfULL << 60) +#define DB_PF_OFFSET_P5 0x10000 +#define DB_VF_OFFSET_P5 0x4000 + #define INVALID_HW_RING_ID ((u16)-1) /* The hardware supports certain page sizes. Use the supported page sizes @@ -656,11 +659,6 @@ struct nqe_cn { #define HWRM_CMD_TIMEOUT (bp->hwrm_cmd_timeout) #define HWRM_RESET_TIMEOUT ((HWRM_CMD_TIMEOUT) * 4) #define HWRM_COREDUMP_TIMEOUT ((HWRM_CMD_TIMEOUT) * 12) -#define HWRM_RESP_ERR_CODE_MASK 0xffff -#define HWRM_RESP_LEN_OFFSET 4 -#define HWRM_RESP_LEN_MASK 0xffff0000 -#define HWRM_RESP_LEN_SFT 16 -#define HWRM_RESP_VALID_MASK 0xff000000 #define BNXT_HWRM_REQ_MAX_SIZE 128 #define BNXT_HWRM_REQS_PER_PAGE (BNXT_PAGE_SIZE / \ BNXT_HWRM_REQ_MAX_SIZE) @@ -907,6 +905,20 @@ struct bnxt_rx_ring_info { struct page_pool *page_pool; }; +struct bnxt_rx_sw_stats { + u64 rx_l4_csum_errors; + u64 rx_buf_errors; +}; + +struct bnxt_cmn_sw_stats { + u64 missed_irqs; +}; + +struct bnxt_sw_stats { + struct bnxt_rx_sw_stats rx; + struct bnxt_cmn_sw_stats cmn; +}; + struct bnxt_cp_ring_info { struct bnxt_napi *bnapi; u32 cp_raw_cons; @@ -934,9 +946,8 @@ struct bnxt_cp_ring_info { struct ctx_hw_stats *hw_stats; dma_addr_t hw_stats_map; u32 hw_stats_ctx_id; - u64 rx_l4_csum_errors; - u64 rx_buf_errors; - u64 missed_irqs; + + struct bnxt_sw_stats sw_stats; struct bnxt_ring_struct cp_ring_struct; @@ -1356,6 +1367,7 @@ struct bnxt_ctx_mem_info { u16 mrav_num_entries_units; u8 tqm_entries_multiple; u8 ctx_kind_initializer; + u8 tqm_fp_rings_count; u32 flags; #define BNXT_CTX_FLAG_INITED 0x01 @@ -1815,6 +1827,7 @@ struct bnxt { /* ensure atomic 64-bit doorbell writes on 32-bit systems. */ spinlock_t db_lock; #endif + int db_size; #define BNXT_NTP_FLTR_MAX_FLTR 4096 #define BNXT_NTP_FLTR_HASH_SIZE 512 @@ -1857,7 +1870,6 @@ struct bnxt { u8 dsn[8]; struct bnxt_tc_info *tc_info; struct list_head tc_indr_block_list; - struct notifier_block tc_netdev_nb; struct dentry *debugfs_pdev; struct device *hwmon_dev; }; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c index 34046a6286e8..6b88143af5ea 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c @@ -137,7 +137,7 @@ reset_coalesce: return rc; } -static const char * const bnxt_ring_stats_str[] = { +static const char * const bnxt_ring_rx_stats_str[] = { "rx_ucast_packets", "rx_mcast_packets", "rx_bcast_packets", @@ -146,6 +146,9 @@ static const char * const bnxt_ring_stats_str[] = { "rx_ucast_bytes", "rx_mcast_bytes", "rx_bcast_bytes", +}; + +static const char * const bnxt_ring_tx_stats_str[] = { "tx_ucast_packets", "tx_mcast_packets", "tx_bcast_packets", @@ -171,9 +174,12 @@ static const char * const bnxt_ring_tpa2_stats_str[] = { "rx_tpa_errors", }; -static const char * const bnxt_ring_sw_stats_str[] = { +static const char * const bnxt_rx_sw_stats_str[] = { "rx_l4_csum_errors", "rx_buf_errors", +}; + +static const char * const bnxt_cmn_sw_stats_str[] = { "missed_irqs", }; @@ -303,6 +309,11 @@ static struct { {0, "tx_total_discard_pkts"}, }; +#define NUM_RING_RX_SW_STATS ARRAY_SIZE(bnxt_rx_sw_stats_str) +#define NUM_RING_CMN_SW_STATS ARRAY_SIZE(bnxt_cmn_sw_stats_str) +#define NUM_RING_RX_HW_STATS ARRAY_SIZE(bnxt_ring_rx_stats_str) +#define NUM_RING_TX_HW_STATS ARRAY_SIZE(bnxt_ring_tx_stats_str) + static const struct { long offset; char string[ETH_GSTRING_LEN]; @@ -482,12 +493,21 @@ static int bnxt_get_num_tpa_ring_stats(struct bnxt *bp) static int bnxt_get_num_ring_stats(struct bnxt *bp) { - int num_stats; + int rx, tx, cmn; + bool sh = false; - num_stats = ARRAY_SIZE(bnxt_ring_stats_str) + - ARRAY_SIZE(bnxt_ring_sw_stats_str) + - bnxt_get_num_tpa_ring_stats(bp); - return num_stats * bp->cp_nr_rings; + if (bp->flags & BNXT_FLAG_SHARED_RINGS) + sh = true; + + rx = NUM_RING_RX_HW_STATS + NUM_RING_RX_SW_STATS + + bnxt_get_num_tpa_ring_stats(bp); + tx = NUM_RING_TX_HW_STATS; + cmn = NUM_RING_CMN_SW_STATS; + if (sh) + return (rx + tx + cmn) * bp->cp_nr_rings; + else + return rx * bp->rx_nr_rings + tx * bp->tx_nr_rings + + cmn * bp->cp_nr_rings; } static int bnxt_get_num_stats(struct bnxt *bp) @@ -528,13 +548,29 @@ static int bnxt_get_sset_count(struct net_device *dev, int sset) } } +static bool is_rx_ring(struct bnxt *bp, int ring_num) +{ + return ring_num < bp->rx_nr_rings; +} + +static bool is_tx_ring(struct bnxt *bp, int ring_num) +{ + int tx_base = 0; + + if (!(bp->flags & BNXT_FLAG_SHARED_RINGS)) + tx_base = bp->rx_nr_rings; + + if (ring_num >= tx_base && ring_num < (tx_base + bp->tx_nr_rings)) + return true; + return false; +} + static void bnxt_get_ethtool_stats(struct net_device *dev, struct ethtool_stats *stats, u64 *buf) { u32 i, j = 0; struct bnxt *bp = netdev_priv(dev); - u32 stat_fields = ARRAY_SIZE(bnxt_ring_stats_str) + - bnxt_get_num_tpa_ring_stats(bp); + u32 tpa_stats; if (!bp->bnapi) { j += bnxt_get_num_ring_stats(bp) + BNXT_NUM_SW_FUNC_STATS; @@ -544,17 +580,42 @@ static void bnxt_get_ethtool_stats(struct net_device *dev, for (i = 0; i < BNXT_NUM_SW_FUNC_STATS; i++) bnxt_sw_func_stats[i].counter = 0; + tpa_stats = bnxt_get_num_tpa_ring_stats(bp); for (i = 0; i < bp->cp_nr_rings; i++) { struct bnxt_napi *bnapi = bp->bnapi[i]; struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring; __le64 *hw_stats = (__le64 *)cpr->hw_stats; + u64 *sw; int k; - for (k = 0; k < stat_fields; j++, k++) + if (is_rx_ring(bp, i)) { + for (k = 0; k < NUM_RING_RX_HW_STATS; j++, k++) + buf[j] = le64_to_cpu(hw_stats[k]); + } + if (is_tx_ring(bp, i)) { + k = NUM_RING_RX_HW_STATS; + for (; k < NUM_RING_RX_HW_STATS + NUM_RING_TX_HW_STATS; + j++, k++) + buf[j] = le64_to_cpu(hw_stats[k]); + } + if (!tpa_stats || !is_rx_ring(bp, i)) + goto skip_tpa_ring_stats; + + k = NUM_RING_RX_HW_STATS + NUM_RING_TX_HW_STATS; + for (; k < NUM_RING_RX_HW_STATS + NUM_RING_TX_HW_STATS + + tpa_stats; j++, k++) buf[j] = le64_to_cpu(hw_stats[k]); - buf[j++] = cpr->rx_l4_csum_errors; - buf[j++] = cpr->rx_buf_errors; - buf[j++] = cpr->missed_irqs; + +skip_tpa_ring_stats: + sw = (u64 *)&cpr->sw_stats.rx; + if (is_rx_ring(bp, i)) { + for (k = 0; k < NUM_RING_RX_SW_STATS; j++, k++) + buf[j] = sw[k]; + } + + sw = (u64 *)&cpr->sw_stats.cmn; + for (k = 0; k < NUM_RING_CMN_SW_STATS; j++, k++) + buf[j] = sw[k]; bnxt_sw_func_stats[RX_TOTAL_DISCARDS].counter += le64_to_cpu(cpr->hw_stats->rx_discard_pkts); @@ -632,31 +693,48 @@ static void bnxt_get_strings(struct net_device *dev, u32 stringset, u8 *buf) switch (stringset) { case ETH_SS_STATS: for (i = 0; i < bp->cp_nr_rings; i++) { - num_str = ARRAY_SIZE(bnxt_ring_stats_str); - for (j = 0; j < num_str; j++) { - sprintf(buf, "[%d]: %s", i, - bnxt_ring_stats_str[j]); - buf += ETH_GSTRING_LEN; + if (is_rx_ring(bp, i)) { + num_str = NUM_RING_RX_HW_STATS; + for (j = 0; j < num_str; j++) { + sprintf(buf, "[%d]: %s", i, + bnxt_ring_rx_stats_str[j]); + buf += ETH_GSTRING_LEN; + } } - if (!BNXT_SUPPORTS_TPA(bp)) + if (is_tx_ring(bp, i)) { + num_str = NUM_RING_TX_HW_STATS; + for (j = 0; j < num_str; j++) { + sprintf(buf, "[%d]: %s", i, + bnxt_ring_tx_stats_str[j]); + buf += ETH_GSTRING_LEN; + } + } + num_str = bnxt_get_num_tpa_ring_stats(bp); + if (!num_str || !is_rx_ring(bp, i)) goto skip_tpa_stats; - if (bp->max_tpa_v2) { - num_str = ARRAY_SIZE(bnxt_ring_tpa2_stats_str); + if (bp->max_tpa_v2) str = bnxt_ring_tpa2_stats_str; - } else { - num_str = ARRAY_SIZE(bnxt_ring_tpa_stats_str); + else str = bnxt_ring_tpa_stats_str; - } + for (j = 0; j < num_str; j++) { sprintf(buf, "[%d]: %s", i, str[j]); buf += ETH_GSTRING_LEN; } skip_tpa_stats: - num_str = ARRAY_SIZE(bnxt_ring_sw_stats_str); + if (is_rx_ring(bp, i)) { + num_str = NUM_RING_RX_SW_STATS; + for (j = 0; j < num_str; j++) { + sprintf(buf, "[%d]: %s", i, + bnxt_rx_sw_stats_str[j]); + buf += ETH_GSTRING_LEN; + } + } + num_str = NUM_RING_CMN_SW_STATS; for (j = 0; j < num_str; j++) { sprintf(buf, "[%d]: %s", i, - bnxt_ring_sw_stats_str[j]); + bnxt_cmn_sw_stats_str[j]); buf += ETH_GSTRING_LEN; } } @@ -1749,8 +1827,8 @@ static int bnxt_flash_nvram(struct net_device *dev, return rc; } -static int bnxt_firmware_reset(struct net_device *dev, - u16 dir_type) +static int bnxt_hwrm_firmware_reset(struct net_device *dev, u8 proc_type, + u8 self_reset, u8 flags) { struct hwrm_fw_reset_input req = {0}; struct bnxt *bp = netdev_priv(dev); @@ -1758,48 +1836,77 @@ static int bnxt_firmware_reset(struct net_device *dev, bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FW_RESET, -1, -1); + req.embedded_proc_type = proc_type; + req.selfrst_status = self_reset; + req.flags = flags; + + if (proc_type == FW_RESET_REQ_EMBEDDED_PROC_TYPE_AP) { + rc = hwrm_send_message_silent(bp, &req, sizeof(req), + HWRM_CMD_TIMEOUT); + } else { + rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); + if (rc == -EACCES) + bnxt_print_admin_err(bp); + } + return rc; +} + +static int bnxt_firmware_reset(struct net_device *dev, + enum bnxt_nvm_directory_type dir_type) +{ + u8 self_reset = FW_RESET_REQ_SELFRST_STATUS_SELFRSTNONE; + u8 proc_type, flags = 0; + /* TODO: Address self-reset of APE/KONG/BONO/TANG or ungraceful reset */ /* (e.g. when firmware isn't already running) */ switch (dir_type) { case BNX_DIR_TYPE_CHIMP_PATCH: case BNX_DIR_TYPE_BOOTCODE: case BNX_DIR_TYPE_BOOTCODE_2: - req.embedded_proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_BOOT; + proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_BOOT; /* Self-reset ChiMP upon next PCIe reset: */ - req.selfrst_status = FW_RESET_REQ_SELFRST_STATUS_SELFRSTPCIERST; + self_reset = FW_RESET_REQ_SELFRST_STATUS_SELFRSTPCIERST; break; case BNX_DIR_TYPE_APE_FW: case BNX_DIR_TYPE_APE_PATCH: - req.embedded_proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_MGMT; + proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_MGMT; /* Self-reset APE upon next PCIe reset: */ - req.selfrst_status = FW_RESET_REQ_SELFRST_STATUS_SELFRSTPCIERST; + self_reset = FW_RESET_REQ_SELFRST_STATUS_SELFRSTPCIERST; break; case BNX_DIR_TYPE_KONG_FW: case BNX_DIR_TYPE_KONG_PATCH: - req.embedded_proc_type = - FW_RESET_REQ_EMBEDDED_PROC_TYPE_NETCTRL; + proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_NETCTRL; break; case BNX_DIR_TYPE_BONO_FW: case BNX_DIR_TYPE_BONO_PATCH: - req.embedded_proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_ROCE; - break; - case BNXT_FW_RESET_CHIP: - req.embedded_proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_CHIP; - req.selfrst_status = FW_RESET_REQ_SELFRST_STATUS_SELFRSTASAP; - if (bp->fw_cap & BNXT_FW_CAP_HOT_RESET) - req.flags = FW_RESET_REQ_FLAGS_RESET_GRACEFUL; - break; - case BNXT_FW_RESET_AP: - req.embedded_proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_AP; + proc_type = FW_RESET_REQ_EMBEDDED_PROC_TYPE_ROCE; break; default: return -EINVAL; } - rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT); - if (rc == -EACCES) - bnxt_print_admin_err(bp); - return rc; + return bnxt_hwrm_firmware_reset(dev, proc_type, self_reset, flags); +} + +static int bnxt_firmware_reset_chip(struct net_device *dev) +{ + struct bnxt *bp = netdev_priv(dev); + u8 flags = 0; + + if (bp->fw_cap & BNXT_FW_CAP_HOT_RESET) + flags = FW_RESET_REQ_FLAGS_RESET_GRACEFUL; + + return bnxt_hwrm_firmware_reset(dev, + FW_RESET_REQ_EMBEDDED_PROC_TYPE_CHIP, + FW_RESET_REQ_SELFRST_STATUS_SELFRSTASAP, + flags); +} + +static int bnxt_firmware_reset_ap(struct net_device *dev) +{ + return bnxt_hwrm_firmware_reset(dev, FW_RESET_REQ_EMBEDDED_PROC_TYPE_AP, + FW_RESET_REQ_SELFRST_STATUS_SELFRSTNONE, + 0); } static int bnxt_flash_firmware(struct net_device *dev, @@ -1988,9 +2095,9 @@ static int bnxt_flash_firmware_from_file(struct net_device *dev, rc, filename); return rc; } - if (bnxt_dir_type_is_ape_bin_format(dir_type) == true) + if (bnxt_dir_type_is_ape_bin_format(dir_type)) rc = bnxt_flash_firmware(dev, dir_type, fw->data, fw->size); - else if (bnxt_dir_type_is_other_exec_format(dir_type) == true) + else if (bnxt_dir_type_is_other_exec_format(dir_type)) rc = bnxt_flash_microcode(dev, dir_type, fw->data, fw->size); else rc = bnxt_flash_nvram(dev, dir_type, BNX_DIR_ORDINAL_FIRST, @@ -2012,11 +2119,12 @@ int bnxt_flash_package_from_file(struct net_device *dev, const char *filename, bnxt_hwrm_fw_set_time(bp); - if (bnxt_find_nvram_item(dev, BNX_DIR_TYPE_UPDATE, - BNX_DIR_ORDINAL_FIRST, BNX_DIR_EXT_NONE, - &index, &item_len, NULL) != 0) { + rc = bnxt_find_nvram_item(dev, BNX_DIR_TYPE_UPDATE, + BNX_DIR_ORDINAL_FIRST, BNX_DIR_EXT_NONE, + &index, &item_len, NULL); + if (rc) { netdev_err(dev, "PKG update area not created in nvram\n"); - return -ENOBUFS; + return rc; } rc = request_firmware(&fw, filename, &dev->dev); @@ -2377,7 +2485,7 @@ static int bnxt_set_eeprom(struct net_device *dev, } /* Create or re-write an NVM item: */ - if (bnxt_dir_type_is_executable(type) == true) + if (bnxt_dir_type_is_executable(type)) return -EOPNOTSUPP; ext = eeprom->magic & 0xffff; ordinal = eeprom->offset >> 16; @@ -2975,7 +3083,11 @@ static void bnxt_self_test(struct net_device *dev, struct ethtool_test *etest, static int bnxt_reset(struct net_device *dev, u32 *flags) { struct bnxt *bp = netdev_priv(dev); - int rc = 0; + bool reload = false; + u32 req = *flags; + + if (!req) + return -EINVAL; if (!BNXT_PF(bp)) { netdev_err(dev, "Reset is not supported from a VF\n"); @@ -2989,33 +3101,37 @@ static int bnxt_reset(struct net_device *dev, u32 *flags) return -EBUSY; } - if (*flags == ETH_RESET_ALL) { + if ((req & BNXT_FW_RESET_CHIP) == BNXT_FW_RESET_CHIP) { /* This feature is not supported in older firmware versions */ - if (bp->hwrm_spec_code < 0x10803) - return -EOPNOTSUPP; - - rc = bnxt_firmware_reset(dev, BNXT_FW_RESET_CHIP); - if (!rc) { - netdev_info(dev, "Reset request successful.\n"); - if (!(bp->fw_cap & BNXT_FW_CAP_HOT_RESET)) - netdev_info(dev, "Reload driver to complete reset\n"); - *flags = 0; + if (bp->hwrm_spec_code >= 0x10803) { + if (!bnxt_firmware_reset_chip(dev)) { + netdev_info(dev, "Firmware reset request successful.\n"); + if (!(bp->fw_cap & BNXT_FW_CAP_HOT_RESET)) + reload = true; + *flags &= ~BNXT_FW_RESET_CHIP; + } + } else if (req == BNXT_FW_RESET_CHIP) { + return -EOPNOTSUPP; /* only request, fail hard */ } - } else if (*flags == ETH_RESET_AP) { - /* This feature is not supported in older firmware versions */ - if (bp->hwrm_spec_code < 0x10803) - return -EOPNOTSUPP; + } - rc = bnxt_firmware_reset(dev, BNXT_FW_RESET_AP); - if (!rc) { - netdev_info(dev, "Reset Application Processor request successful.\n"); - *flags = 0; + if (req & BNXT_FW_RESET_AP) { + /* This feature is not supported in older firmware versions */ + if (bp->hwrm_spec_code >= 0x10803) { + if (!bnxt_firmware_reset_ap(dev)) { + netdev_info(dev, "Reset application processor successful.\n"); + reload = true; + *flags &= ~BNXT_FW_RESET_AP; + } + } else if (req == BNXT_FW_RESET_AP) { + return -EOPNOTSUPP; /* only request, fail hard */ } - } else { - rc = -EINVAL; } - return rc; + if (reload) + netdev_info(dev, "Reload driver to complete reset\n"); + + return 0; } static int bnxt_hwrm_dbg_dma_data(struct bnxt *bp, void *msg, int msg_len, diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h index 3576d951727b..ce7585ff9e4d 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.h @@ -77,8 +77,12 @@ struct hwrm_dbg_cmn_output { #define BNXT_LED_DFLT_ENABLES(x) \ cpu_to_le32(BNXT_LED_DFLT_ENA << (BNXT_LED_DFLT_ENA_SHIFT * (x))) -#define BNXT_FW_RESET_AP 0xfffe -#define BNXT_FW_RESET_CHIP 0xffff +#define BNXT_FW_RESET_AP (ETH_RESET_AP << ETH_RESET_SHARED_SHIFT) +#define BNXT_FW_RESET_CHIP ((ETH_RESET_MGMT | ETH_RESET_IRQ | \ + ETH_RESET_DMA | ETH_RESET_FILTER | \ + ETH_RESET_OFFLOAD | ETH_RESET_MAC | \ + ETH_RESET_PHY | ETH_RESET_RAM) \ + << ETH_RESET_SHARED_SHIFT) extern const struct ethtool_ops bnxt_ethtool_ops; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h index 7cf27dffadb5..7e9235c8d21e 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_hsi.h @@ -2,7 +2,7 @@ * * Copyright (c) 2014-2016 Broadcom Corporation * Copyright (c) 2014-2018 Broadcom Limited - * Copyright (c) 2018-2019 Broadcom Inc. + * Copyright (c) 2018-2020 Broadcom Inc. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -207,6 +207,8 @@ struct cmd_nums { #define HWRM_PORT_PHY_MDIO_READ 0xb6UL #define HWRM_PORT_PHY_MDIO_BUS_ACQUIRE 0xb7UL #define HWRM_PORT_PHY_MDIO_BUS_RELEASE 0xb8UL + #define HWRM_PORT_QSTATS_EXT_PFC_WD 0xb9UL + #define HWRM_PORT_ECN_QSTATS 0xbaUL #define HWRM_FW_RESET 0xc0UL #define HWRM_FW_QSTATUS 0xc1UL #define HWRM_FW_HEALTH_CHECK 0xc2UL @@ -220,6 +222,8 @@ struct cmd_nums { #define HWRM_FW_SET_STRUCTURED_DATA 0xcaUL #define HWRM_FW_GET_STRUCTURED_DATA 0xcbUL #define HWRM_FW_IPC_MAILBOX 0xccUL + #define HWRM_FW_ECN_CFG 0xcdUL + #define HWRM_FW_ECN_QCFG 0xceUL #define HWRM_EXEC_FWD_RESP 0xd0UL #define HWRM_REJECT_FWD_RESP 0xd1UL #define HWRM_FWD_RESP 0xd2UL @@ -233,6 +237,7 @@ struct cmd_nums { #define HWRM_TEMP_MONITOR_QUERY 0xe0UL #define HWRM_REG_POWER_QUERY 0xe1UL #define HWRM_CORE_FREQUENCY_QUERY 0xe2UL + #define HWRM_REG_POWER_HISTOGRAM 0xe3UL #define HWRM_WOL_FILTER_ALLOC 0xf0UL #define HWRM_WOL_FILTER_FREE 0xf1UL #define HWRM_WOL_FILTER_QCFG 0xf2UL @@ -331,6 +336,7 @@ struct cmd_nums { #define HWRM_FUNC_VF_BW_CFG 0x195UL #define HWRM_FUNC_VF_BW_QCFG 0x196UL #define HWRM_FUNC_HOST_PF_IDS_QUERY 0x197UL + #define HWRM_FUNC_QSTATS_EXT 0x198UL #define HWRM_SELFTEST_QLIST 0x200UL #define HWRM_SELFTEST_EXEC 0x201UL #define HWRM_SELFTEST_IRQ 0x202UL @@ -341,6 +347,31 @@ struct cmd_nums { #define HWRM_MFG_OTP_CFG 0x207UL #define HWRM_MFG_OTP_QCFG 0x208UL #define HWRM_MFG_HDMA_TEST 0x209UL + #define HWRM_MFG_FRU_EEPROM_WRITE 0x20aUL + #define HWRM_MFG_FRU_EEPROM_READ 0x20bUL + #define HWRM_TF 0x2bcUL + #define HWRM_TF_VERSION_GET 0x2bdUL + #define HWRM_TF_SESSION_OPEN 0x2c6UL + #define HWRM_TF_SESSION_ATTACH 0x2c7UL + #define HWRM_TF_SESSION_CLOSE 0x2c8UL + #define HWRM_TF_SESSION_QCFG 0x2c9UL + #define HWRM_TF_SESSION_RESC_QCAPS 0x2caUL + #define HWRM_TF_SESSION_RESC_ALLOC 0x2cbUL + #define HWRM_TF_SESSION_RESC_FREE 0x2ccUL + #define HWRM_TF_SESSION_RESC_FLUSH 0x2cdUL + #define HWRM_TF_TBL_TYPE_GET 0x2d0UL + #define HWRM_TF_TBL_TYPE_SET 0x2d1UL + #define HWRM_TF_CTXT_MEM_RGTR 0x2daUL + #define HWRM_TF_CTXT_MEM_UNRGTR 0x2dbUL + #define HWRM_TF_EXT_EM_QCAPS 0x2dcUL + #define HWRM_TF_EXT_EM_OP 0x2ddUL + #define HWRM_TF_EXT_EM_CFG 0x2deUL + #define HWRM_TF_EXT_EM_QCFG 0x2dfUL + #define HWRM_TF_TCAM_SET 0x2eeUL + #define HWRM_TF_TCAM_GET 0x2efUL + #define HWRM_TF_TCAM_MOVE 0x2f0UL + #define HWRM_TF_TCAM_FREE 0x2f1UL + #define HWRM_SV 0x400UL #define HWRM_DBG_READ_DIRECT 0xff10UL #define HWRM_DBG_READ_INDIRECT 0xff11UL #define HWRM_DBG_WRITE_DIRECT 0xff12UL @@ -356,6 +387,10 @@ struct cmd_nums { #define HWRM_DBG_RING_INFO_GET 0xff1cUL #define HWRM_DBG_CRASHDUMP_HEADER 0xff1dUL #define HWRM_DBG_CRASHDUMP_ERASE 0xff1eUL + #define HWRM_DBG_DRV_TRACE 0xff1fUL + #define HWRM_DBG_QCAPS 0xff20UL + #define HWRM_DBG_QCFG 0xff21UL + #define HWRM_DBG_CRASHDUMP_MEDIUM_CFG 0xff22UL #define HWRM_NVM_FACTORY_DEFAULTS 0xffeeUL #define HWRM_NVM_VALIDATE_OPTION 0xffefUL #define HWRM_NVM_FLUSH 0xfff0UL @@ -429,8 +464,8 @@ struct hwrm_err_output { #define HWRM_VERSION_MAJOR 1 #define HWRM_VERSION_MINOR 10 #define HWRM_VERSION_UPDATE 1 -#define HWRM_VERSION_RSVD 12 -#define HWRM_VERSION_STR "1.10.1.12" +#define HWRM_VERSION_RSVD 33 +#define HWRM_VERSION_STR "1.10.1.33" /* hwrm_ver_get_input (size:192b/24B) */ struct hwrm_ver_get_input { @@ -482,6 +517,7 @@ struct hwrm_ver_get_output { #define VER_GET_RESP_DEV_CAPS_CFG_CFA_EEM_SUPPORTED 0x800UL #define VER_GET_RESP_DEV_CAPS_CFG_CFA_ADV_FLOW_MGNT_SUPPORTED 0x1000UL #define VER_GET_RESP_DEV_CAPS_CFG_CFA_TFLIB_SUPPORTED 0x2000UL + #define VER_GET_RESP_DEV_CAPS_CFG_CFA_TRUFLOW_SUPPORTED 0x4000UL u8 roce_fw_maj_8b; u8 roce_fw_min_8b; u8 roce_fw_bld_8b; @@ -647,6 +683,7 @@ struct hwrm_async_event_cmpl { #define ASYNC_EVENT_CMPL_EVENT_ID_TFLIB_LINK_STATUS_CHANGE 0x3eUL #define ASYNC_EVENT_CMPL_EVENT_ID_QUIESCE_DONE 0x3fUL #define ASYNC_EVENT_CMPL_EVENT_ID_DEFERRED_RESPONSE 0x40UL + #define ASYNC_EVENT_CMPL_EVENT_ID_PFC_WATCHDOG_CFG_CHANGE 0x41UL #define ASYNC_EVENT_CMPL_EVENT_ID_FW_TRACE_MSG 0xfeUL #define ASYNC_EVENT_CMPL_EVENT_ID_HWRM_ERROR 0xffUL #define ASYNC_EVENT_CMPL_EVENT_ID_LAST ASYNC_EVENT_CMPL_EVENT_ID_HWRM_ERROR @@ -1089,7 +1126,7 @@ struct hwrm_func_qcaps_input { u8 unused_0[6]; }; -/* hwrm_func_qcaps_output (size:640b/80B) */ +/* hwrm_func_qcaps_output (size:704b/88B) */ struct hwrm_func_qcaps_output { __le16 error_code; __le16 req_type; @@ -1126,6 +1163,10 @@ struct hwrm_func_qcaps_output { #define FUNC_QCAPS_RESP_FLAGS_ERR_RECOVER_RELOAD 0x2000000UL #define FUNC_QCAPS_RESP_FLAGS_NOTIFY_VF_DEF_VNIC_CHNG_SUPPORTED 0x4000000UL #define FUNC_QCAPS_RESP_FLAGS_VLAN_ACCELERATION_TX_DISABLED 0x8000000UL + #define FUNC_QCAPS_RESP_FLAGS_COREDUMP_CMD_SUPPORTED 0x10000000UL + #define FUNC_QCAPS_RESP_FLAGS_CRASHDUMP_CMD_SUPPORTED 0x20000000UL + #define FUNC_QCAPS_RESP_FLAGS_PFC_WD_STATS_SUPPORTED 0x40000000UL + #define FUNC_QCAPS_RESP_FLAGS_DBG_QCAPS_CMD_SUPPORTED 0x80000000UL u8 mac_address[6]; __le16 max_rsscos_ctx; __le16 max_cmpl_rings; @@ -1146,7 +1187,12 @@ struct hwrm_func_qcaps_output { __le32 max_flow_id; __le32 max_hw_ring_grps; __le16 max_sp_tx_rings; - u8 unused_0; + u8 unused_0[2]; + __le32 flags_ext; + #define FUNC_QCAPS_RESP_FLAGS_EXT_ECN_MARK_SUPPORTED 0x1UL + #define FUNC_QCAPS_RESP_FLAGS_EXT_ECN_STATS_SUPPORTED 0x2UL + #define FUNC_QCAPS_RESP_FLAGS_EXT_EXT_HW_STATS_SUPPORTED 0x4UL + u8 unused_1[3]; u8 valid; }; @@ -1161,7 +1207,7 @@ struct hwrm_func_qcfg_input { u8 unused_0[6]; }; -/* hwrm_func_qcfg_output (size:704b/88B) */ +/* hwrm_func_qcfg_output (size:768b/96B) */ struct hwrm_func_qcfg_output { __le16 error_code; __le16 req_type; @@ -1267,7 +1313,11 @@ struct hwrm_func_qcfg_output { u8 always_1; __le32 reset_addr_poll; __le16 legacy_l2_db_size_kb; - u8 unused_2[1]; + __le16 svif_info; + #define FUNC_QCFG_RESP_SVIF_INFO_SVIF_MASK 0x7fffUL + #define FUNC_QCFG_RESP_SVIF_INFO_SVIF_SFT 0 + #define FUNC_QCFG_RESP_SVIF_INFO_SVIF_VALID 0x8000UL + u8 unused_2[7]; u8 valid; }; @@ -1420,9 +1470,10 @@ struct hwrm_func_qstats_input { __le64 resp_addr; __le16 fid; u8 flags; - #define FUNC_QSTATS_REQ_FLAGS_UNUSED 0x0UL - #define FUNC_QSTATS_REQ_FLAGS_ROCE_ONLY 0x1UL - #define FUNC_QSTATS_REQ_FLAGS_LAST FUNC_QSTATS_REQ_FLAGS_ROCE_ONLY + #define FUNC_QSTATS_REQ_FLAGS_UNUSED 0x0UL + #define FUNC_QSTATS_REQ_FLAGS_ROCE_ONLY 0x1UL + #define FUNC_QSTATS_REQ_FLAGS_COUNTER_MASK 0x2UL + #define FUNC_QSTATS_REQ_FLAGS_LAST FUNC_QSTATS_REQ_FLAGS_COUNTER_MASK u8 unused_0[5]; }; @@ -1456,6 +1507,53 @@ struct hwrm_func_qstats_output { u8 valid; }; +/* hwrm_func_qstats_ext_input (size:192b/24B) */ +struct hwrm_func_qstats_ext_input { + __le16 req_type; + __le16 cmpl_ring; + __le16 seq_id; + __le16 target_id; + __le64 resp_addr; + __le16 fid; + u8 flags; + #define FUNC_QSTATS_EXT_REQ_FLAGS_UNUSED 0x0UL + #define FUNC_QSTATS_EXT_REQ_FLAGS_ROCE_ONLY 0x1UL + #define FUNC_QSTATS_EXT_REQ_FLAGS_COUNTER_MASK 0x2UL + #define FUNC_QSTATS_EXT_REQ_FLAGS_LAST FUNC_QSTATS_EXT_REQ_FLAGS_COUNTER_MASK + u8 unused_0[5]; +}; + +/* hwrm_func_qstats_ext_output (size:1472b/184B) */ +struct hwrm_func_qstats_ext_output { + __le16 error_code; + __le16 req_type; + __le16 seq_id; + __le16 resp_len; + __le64 rx_ucast_pkts; + __le64 rx_mcast_pkts; + __le64 rx_bcast_pkts; + __le64 rx_discard_pkts; + __le64 rx_drop_pkts; + __le64 rx_ucast_bytes; + __le64 rx_mcast_bytes; + __le64 rx_bcast_bytes; + __le64 tx_ucast_pkts; + __le64 tx_mcast_pkts; + __le64 tx_bcast_pkts; + __le64 tx_discard_pkts; + __le64 tx_drop_pkts; + __le64 tx_ucast_bytes; + __le64 tx_mcast_bytes; + __le64 tx_bcast_bytes; + __le64 rx_tpa_eligible_pkt; + __le64 rx_tpa_eligible_bytes; + __le64 rx_tpa_pkt; + __le64 rx_tpa_bytes; + __le64 rx_tpa_errors; + u8 unused_0[7]; + u8 valid; +}; + /* hwrm_func_clr_stats_input (size:192b/24B) */ struct hwrm_func_clr_stats_input { __le16 req_type; @@ -1808,7 +1906,7 @@ struct hwrm_func_backing_store_qcaps_output { u8 ctx_kind_initializer; __le32 rsvd; __le16 rsvd1; - u8 rsvd2; + u8 tqm_fp_rings_count; u8 valid; }; @@ -2231,7 +2329,17 @@ struct hwrm_error_recovery_qcfg_output { #define ERROR_RECOVERY_QCFG_RESP_RESET_REG_ADDR_SFT 2 __le32 reset_reg_val[16]; u8 delay_after_reset[16]; - u8 unused_1[7]; + __le32 err_recovery_cnt_reg; + #define ERROR_RECOVERY_QCFG_RESP_ERR_RECOVERY_CNT_REG_ADDR_SPACE_MASK 0x3UL + #define ERROR_RECOVERY_QCFG_RESP_ERR_RECOVERY_CNT_REG_ADDR_SPACE_SFT 0 + #define ERROR_RECOVERY_QCFG_RESP_ERR_RECOVERY_CNT_REG_ADDR_SPACE_PCIE_CFG 0x0UL + #define ERROR_RECOVERY_QCFG_RESP_ERR_RECOVERY_CNT_REG_ADDR_SPACE_GRC 0x1UL + #define ERROR_RECOVERY_QCFG_RESP_ERR_RECOVERY_CNT_REG_ADDR_SPACE_BAR0 0x2UL + #define ERROR_RECOVERY_QCFG_RESP_ERR_RECOVERY_CNT_REG_ADDR_SPACE_BAR1 0x3UL + #define ERROR_RECOVERY_QCFG_RESP_ERR_RECOVERY_CNT_REG_ADDR_SPACE_LAST ERROR_RECOVERY_QCFG_RESP_ERR_RECOVERY_CNT_REG_ADDR_SPACE_BAR1 + #define ERROR_RECOVERY_QCFG_RESP_ERR_RECOVERY_CNT_REG_ADDR_MASK 0xfffffffcUL + #define ERROR_RECOVERY_QCFG_RESP_ERR_RECOVERY_CNT_REG_ADDR_SFT 2 + u8 unused_1[3]; u8 valid; }; @@ -2934,7 +3042,11 @@ struct hwrm_port_qstats_input { __le16 target_id; __le64 resp_addr; __le16 port_id; - u8 unused_0[6]; + u8 flags; + #define PORT_QSTATS_REQ_FLAGS_UNUSED 0x0UL + #define PORT_QSTATS_REQ_FLAGS_COUNTER_MASK 0x1UL + #define PORT_QSTATS_REQ_FLAGS_LAST PORT_QSTATS_REQ_FLAGS_COUNTER_MASK + u8 unused_0[5]; __le64 tx_stat_host_addr; __le64 rx_stat_host_addr; }; @@ -3058,7 +3170,11 @@ struct hwrm_port_qstats_ext_input { __le16 port_id; __le16 tx_stat_size; __le16 rx_stat_size; - u8 unused_0[2]; + u8 flags; + #define PORT_QSTATS_EXT_REQ_FLAGS_UNUSED 0x0UL + #define PORT_QSTATS_EXT_REQ_FLAGS_COUNTER_MASK 0x1UL + #define PORT_QSTATS_EXT_REQ_FLAGS_LAST PORT_QSTATS_EXT_REQ_FLAGS_COUNTER_MASK + u8 unused_0; __le64 tx_stat_host_addr; __le64 rx_stat_host_addr; }; @@ -3840,14 +3956,22 @@ struct hwrm_queue_pfcenable_qcfg_output { __le16 seq_id; __le16 resp_len; __le32 flags; - #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI0_PFC_ENABLED 0x1UL - #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI1_PFC_ENABLED 0x2UL - #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI2_PFC_ENABLED 0x4UL - #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI3_PFC_ENABLED 0x8UL - #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI4_PFC_ENABLED 0x10UL - #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI5_PFC_ENABLED 0x20UL - #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI6_PFC_ENABLED 0x40UL - #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI7_PFC_ENABLED 0x80UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI0_PFC_ENABLED 0x1UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI1_PFC_ENABLED 0x2UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI2_PFC_ENABLED 0x4UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI3_PFC_ENABLED 0x8UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI4_PFC_ENABLED 0x10UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI5_PFC_ENABLED 0x20UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI6_PFC_ENABLED 0x40UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI7_PFC_ENABLED 0x80UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI0_PFC_WATCHDOG_ENABLED 0x100UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI1_PFC_WATCHDOG_ENABLED 0x200UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI2_PFC_WATCHDOG_ENABLED 0x400UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI3_PFC_WATCHDOG_ENABLED 0x800UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI4_PFC_WATCHDOG_ENABLED 0x1000UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI5_PFC_WATCHDOG_ENABLED 0x2000UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI6_PFC_WATCHDOG_ENABLED 0x4000UL + #define QUEUE_PFCENABLE_QCFG_RESP_FLAGS_PRI7_PFC_WATCHDOG_ENABLED 0x8000UL u8 unused_0[3]; u8 valid; }; @@ -3860,14 +3984,22 @@ struct hwrm_queue_pfcenable_cfg_input { __le16 target_id; __le64 resp_addr; __le32 flags; - #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI0_PFC_ENABLED 0x1UL - #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI1_PFC_ENABLED 0x2UL - #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI2_PFC_ENABLED 0x4UL - #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI3_PFC_ENABLED 0x8UL - #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI4_PFC_ENABLED 0x10UL - #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI5_PFC_ENABLED 0x20UL - #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI6_PFC_ENABLED 0x40UL - #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI7_PFC_ENABLED 0x80UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI0_PFC_ENABLED 0x1UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI1_PFC_ENABLED 0x2UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI2_PFC_ENABLED 0x4UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI3_PFC_ENABLED 0x8UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI4_PFC_ENABLED 0x10UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI5_PFC_ENABLED 0x20UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI6_PFC_ENABLED 0x40UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI7_PFC_ENABLED 0x80UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI0_PFC_WATCHDOG_ENABLED 0x100UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI1_PFC_WATCHDOG_ENABLED 0x200UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI2_PFC_WATCHDOG_ENABLED 0x400UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI3_PFC_WATCHDOG_ENABLED 0x800UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI4_PFC_WATCHDOG_ENABLED 0x1000UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI5_PFC_WATCHDOG_ENABLED 0x2000UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI6_PFC_WATCHDOG_ENABLED 0x4000UL + #define QUEUE_PFCENABLE_CFG_REQ_FLAGS_PRI7_PFC_WATCHDOG_ENABLED 0x8000UL __le16 port_id; u8 unused_0[2]; }; @@ -5287,7 +5419,11 @@ struct hwrm_ring_cmpl_ring_qaggint_params_input { __le16 target_id; __le64 resp_addr; __le16 ring_id; - u8 unused_0[6]; + __le16 flags; + #define RING_CMPL_RING_QAGGINT_PARAMS_REQ_FLAGS_UNUSED_0_MASK 0x3UL + #define RING_CMPL_RING_QAGGINT_PARAMS_REQ_FLAGS_UNUSED_0_SFT 0 + #define RING_CMPL_RING_QAGGINT_PARAMS_REQ_FLAGS_IS_NQ 0x4UL + u8 unused_0[4]; }; /* hwrm_ring_cmpl_ring_qaggint_params_output (size:256b/32B) */ @@ -7618,7 +7754,9 @@ struct hwrm_nvm_modify_input { __le64 resp_addr; __le64 host_src_addr; __le16 dir_idx; - u8 unused_0[2]; + __le16 flags; + #define NVM_MODIFY_REQ_FLAGS_BATCH_MODE 0x1UL + #define NVM_MODIFY_REQ_FLAGS_BATCH_LAST 0x2UL __le32 offset; __le32 len; u8 unused_1[4]; @@ -8027,4 +8165,18 @@ struct hwrm_selftest_irq_output { u8 valid; }; +/* fw_status_reg (size:32b/4B) */ +struct fw_status_reg { + u32 fw_status; + #define FW_STATUS_REG_CODE_MASK 0xffffUL + #define FW_STATUS_REG_CODE_SFT 0 + #define FW_STATUS_REG_CODE_READY 0x8000UL + #define FW_STATUS_REG_CODE_LAST FW_STATUS_REG_CODE_READY + #define FW_STATUS_REG_IMAGE_DEGRADED 0x10000UL + #define FW_STATUS_REG_RECOVERABLE 0x20000UL + #define FW_STATUS_REG_CRASHDUMP_ONGOING 0x40000UL + #define FW_STATUS_REG_CRASHDUMP_COMPLETE 0x80000UL + #define FW_STATUS_REG_SHUTDOWN 0x100000UL +}; + #endif /* _BNXT_HSI_H_ */ diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c index cea2f9958a1d..3a9a51f7063a 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_sriov.c @@ -645,7 +645,7 @@ static int bnxt_hwrm_func_cfg(struct bnxt *bp, int num_vfs) FUNC_CFG_REQ_ENABLES_NUM_VNICS | FUNC_CFG_REQ_ENABLES_NUM_HW_RING_GRPS); - mtu = bp->dev->mtu + ETH_HLEN + ETH_FCS_LEN + VLAN_HLEN; + mtu = bp->dev->mtu + ETH_HLEN + VLAN_HLEN; req.mru = cpu_to_le16(mtu); req.mtu = cpu_to_le16(mtu); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c index 782ea0771221..0eef4f5e4a46 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_tc.c @@ -1939,53 +1939,25 @@ static int bnxt_tc_setup_indr_block(struct net_device *netdev, struct bnxt *bp, return 0; } -static int bnxt_tc_setup_indr_cb(struct net_device *netdev, void *cb_priv, - enum tc_setup_type type, void *type_data) -{ - switch (type) { - case TC_SETUP_BLOCK: - return bnxt_tc_setup_indr_block(netdev, cb_priv, type_data); - default: - return -EOPNOTSUPP; - } -} - static bool bnxt_is_netdev_indr_offload(struct net_device *netdev) { return netif_is_vxlan(netdev); } -static int bnxt_tc_indr_block_event(struct notifier_block *nb, - unsigned long event, void *ptr) +static int bnxt_tc_setup_indr_cb(struct net_device *netdev, void *cb_priv, + enum tc_setup_type type, void *type_data) { - struct net_device *netdev; - struct bnxt *bp; - int rc; - - netdev = netdev_notifier_info_to_dev(ptr); if (!bnxt_is_netdev_indr_offload(netdev)) - return NOTIFY_OK; - - bp = container_of(nb, struct bnxt, tc_netdev_nb); + return -EOPNOTSUPP; - switch (event) { - case NETDEV_REGISTER: - rc = __flow_indr_block_cb_register(netdev, bp, - bnxt_tc_setup_indr_cb, - bp); - if (rc) - netdev_info(bp->dev, - "Failed to register indirect blk: dev: %s\n", - netdev->name); - break; - case NETDEV_UNREGISTER: - __flow_indr_block_cb_unregister(netdev, - bnxt_tc_setup_indr_cb, - bp); + switch (type) { + case TC_SETUP_BLOCK: + return bnxt_tc_setup_indr_block(netdev, cb_priv, type_data); + default: break; } - return NOTIFY_DONE; + return -EOPNOTSUPP; } static const struct rhashtable_params bnxt_tc_flow_ht_params = { @@ -2074,8 +2046,8 @@ int bnxt_init_tc(struct bnxt *bp) /* init indirect block notifications */ INIT_LIST_HEAD(&bp->tc_indr_block_list); - bp->tc_netdev_nb.notifier_call = bnxt_tc_indr_block_event; - rc = register_netdevice_notifier(&bp->tc_netdev_nb); + + rc = flow_indr_dev_register(bnxt_tc_setup_indr_cb, bp); if (!rc) return 0; @@ -2101,7 +2073,8 @@ void bnxt_shutdown_tc(struct bnxt *bp) if (!bnxt_tc_flower_enabled(bp)) return; - unregister_netdevice_notifier(&bp->tc_netdev_nb); + flow_indr_dev_unregister(bnxt_tc_setup_indr_cb, bp, + bnxt_tc_setup_indr_block_cb); rhashtable_destroy(&tc_info->flow_table); rhashtable_destroy(&tc_info->l2_table); rhashtable_destroy(&tc_info->decap_l2_table); diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c index 4a316c4b3fa8..8c8368c2f335 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c @@ -104,7 +104,13 @@ static void bnxt_fill_msix_vecs(struct bnxt *bp, struct bnxt_msix_entry *ent) for (i = 0; i < num_msix; i++) { ent[i].vector = bp->irq_tbl[idx + i].vector; ent[i].ring_idx = idx + i; - ent[i].db_offset = (idx + i) * 0x80; + if (bp->flags & BNXT_FLAG_CHIP_P5) { + ent[i].db_offset = DB_PF_OFFSET_P5; + if (BNXT_VF(bp)) + ent[i].db_offset = DB_VF_OFFSET_P5; + } else { + ent[i].db_offset = (idx + i) * 0x80; + } } } @@ -475,6 +481,8 @@ struct bnxt_en_dev *bnxt_ulp_probe(struct net_device *dev) edev->flags |= BNXT_EN_FLAG_ROCEV2_CAP; edev->net = dev; edev->pdev = bp->pdev; + edev->l2_db_size = bp->db_size; + edev->l2_db_size_nc = bp->db_size; bp->edev = edev; } return bp->edev; diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h index 9895406b9830..6b4d2556a6df 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.h @@ -67,6 +67,14 @@ struct bnxt_en_dev { #define BNXT_EN_FLAG_ULP_STOPPED 0x8 const struct bnxt_en_ops *en_ops; struct bnxt_ulp ulp_tbl[BNXT_MAX_ULP]; + int l2_db_size; /* Doorbell BAR size in + * bytes mapped by L2 + * driver. + */ + int l2_db_size_nc; /* Doorbell BAR size in + * bytes mapped as non- + * cacheable. + */ }; struct bnxt_en_ops { diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c index c6f6f2033880..5e3b4a3b69ea 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_xdp.c @@ -138,6 +138,7 @@ bool bnxt_rx_xdp(struct bnxt *bp, struct bnxt_rx_ring_info *rxr, u16 cons, xdp_set_data_meta_invalid(&xdp); xdp.data_end = *data_ptr + *len; xdp.rxq = &rxr->xdp_rxq; + xdp.frame_sz = PAGE_SIZE; /* BNXT_RX_PAGE_MODE(bp) when XDP enabled */ orig_data = xdp.data; rcu_read_lock(); diff --git a/drivers/net/ethernet/broadcom/cnic.c b/drivers/net/ethernet/broadcom/cnic.c index 61ab7d21f6bd..c5cca63b8571 100644 --- a/drivers/net/ethernet/broadcom/cnic.c +++ b/drivers/net/ethernet/broadcom/cnic.c @@ -1918,7 +1918,6 @@ static int cnic_bnx2x_iscsi_ofld1(struct cnic_dev *dev, struct kwqe *wqes[], ret = cnic_alloc_bnx2x_conn_resc(dev, l5_cid); if (ret) { atomic_dec(&cp->iscsi_conn); - ret = 0; goto done; } ret = cnic_setup_bnx2x_ctx(dev, wqes, num); diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c index 79636c78127c..ff31da0ed846 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c @@ -2,7 +2,7 @@ /* * Broadcom GENET (Gigabit Ethernet) controller driver * - * Copyright (c) 2014-2019 Broadcom + * Copyright (c) 2014-2020 Broadcom */ #define pr_fmt(fmt) "bcmgenet: " fmt @@ -23,11 +23,6 @@ #include <linux/dma-mapping.h> #include <linux/pm.h> #include <linux/clk.h> -#include <linux/of.h> -#include <linux/of_address.h> -#include <linux/of_irq.h> -#include <linux/of_net.h> -#include <linux/of_platform.h> #include <net/arp.h> #include <linux/mii.h> @@ -70,6 +65,9 @@ #define GENET_RDMA_REG_OFF (priv->hw_params->rdma_offset + \ TOTAL_DESC * DMA_DESC_SIZE) +/* Forward declarations */ +static void bcmgenet_set_rx_mode(struct net_device *dev); + static inline void bcmgenet_writel(u32 value, void __iomem *offset) { /* MIPS chips strapped for BE will automagically configure the @@ -461,6 +459,384 @@ static inline void bcmgenet_rdma_ring_writel(struct bcmgenet_priv *priv, genet_dma_ring_regs[r]); } +static bool bcmgenet_hfb_is_filter_enabled(struct bcmgenet_priv *priv, + u32 f_index) +{ + u32 offset; + u32 reg; + + offset = HFB_FLT_ENABLE_V3PLUS + (f_index < 32) * sizeof(u32); + reg = bcmgenet_hfb_reg_readl(priv, offset); + return !!(reg & (1 << (f_index % 32))); +} + +static void bcmgenet_hfb_enable_filter(struct bcmgenet_priv *priv, u32 f_index) +{ + u32 offset; + u32 reg; + + offset = HFB_FLT_ENABLE_V3PLUS + (f_index < 32) * sizeof(u32); + reg = bcmgenet_hfb_reg_readl(priv, offset); + reg |= (1 << (f_index % 32)); + bcmgenet_hfb_reg_writel(priv, reg, offset); + reg = bcmgenet_hfb_reg_readl(priv, HFB_CTRL); + reg |= RBUF_HFB_EN; + bcmgenet_hfb_reg_writel(priv, reg, HFB_CTRL); +} + +static void bcmgenet_hfb_disable_filter(struct bcmgenet_priv *priv, u32 f_index) +{ + u32 offset, reg, reg1; + + offset = HFB_FLT_ENABLE_V3PLUS; + reg = bcmgenet_hfb_reg_readl(priv, offset); + reg1 = bcmgenet_hfb_reg_readl(priv, offset + sizeof(u32)); + if (f_index < 32) { + reg1 &= ~(1 << (f_index % 32)); + bcmgenet_hfb_reg_writel(priv, reg1, offset + sizeof(u32)); + } else { + reg &= ~(1 << (f_index % 32)); + bcmgenet_hfb_reg_writel(priv, reg, offset); + } + if (!reg && !reg1) { + reg = bcmgenet_hfb_reg_readl(priv, HFB_CTRL); + reg &= ~RBUF_HFB_EN; + bcmgenet_hfb_reg_writel(priv, reg, HFB_CTRL); + } +} + +static void bcmgenet_hfb_set_filter_rx_queue_mapping(struct bcmgenet_priv *priv, + u32 f_index, u32 rx_queue) +{ + u32 offset; + u32 reg; + + offset = f_index / 8; + reg = bcmgenet_rdma_readl(priv, DMA_INDEX2RING_0 + offset); + reg &= ~(0xF << (4 * (f_index % 8))); + reg |= ((rx_queue & 0xF) << (4 * (f_index % 8))); + bcmgenet_rdma_writel(priv, reg, DMA_INDEX2RING_0 + offset); +} + +static void bcmgenet_hfb_set_filter_length(struct bcmgenet_priv *priv, + u32 f_index, u32 f_length) +{ + u32 offset; + u32 reg; + + offset = HFB_FLT_LEN_V3PLUS + + ((priv->hw_params->hfb_filter_cnt - 1 - f_index) / 4) * + sizeof(u32); + reg = bcmgenet_hfb_reg_readl(priv, offset); + reg &= ~(0xFF << (8 * (f_index % 4))); + reg |= ((f_length & 0xFF) << (8 * (f_index % 4))); + bcmgenet_hfb_reg_writel(priv, reg, offset); +} + +static int bcmgenet_hfb_find_unused_filter(struct bcmgenet_priv *priv) +{ + u32 f_index; + + /* First MAX_NUM_OF_FS_RULES are reserved for Rx NFC filters */ + for (f_index = MAX_NUM_OF_FS_RULES; + f_index < priv->hw_params->hfb_filter_cnt; f_index++) + if (!bcmgenet_hfb_is_filter_enabled(priv, f_index)) + return f_index; + + return -ENOMEM; +} + +static int bcmgenet_hfb_validate_mask(void *mask, size_t size) +{ + while (size) { + switch (*(unsigned char *)mask++) { + case 0x00: + case 0x0f: + case 0xf0: + case 0xff: + size--; + continue; + default: + return -EINVAL; + } + } + + return 0; +} + +#define VALIDATE_MASK(x) \ + bcmgenet_hfb_validate_mask(&(x), sizeof(x)) + +static int bcmgenet_hfb_insert_data(u32 *f, int offset, + void *val, void *mask, size_t size) +{ + int index; + u32 tmp; + + index = offset / 2; + tmp = f[index]; + + while (size--) { + if (offset++ & 1) { + tmp &= ~0x300FF; + tmp |= (*(unsigned char *)val++); + switch ((*(unsigned char *)mask++)) { + case 0xFF: + tmp |= 0x30000; + break; + case 0xF0: + tmp |= 0x20000; + break; + case 0x0F: + tmp |= 0x10000; + break; + } + f[index++] = tmp; + if (size) + tmp = f[index]; + } else { + tmp &= ~0xCFF00; + tmp |= (*(unsigned char *)val++) << 8; + switch ((*(unsigned char *)mask++)) { + case 0xFF: + tmp |= 0xC0000; + break; + case 0xF0: + tmp |= 0x80000; + break; + case 0x0F: + tmp |= 0x40000; + break; + } + if (!size) + f[index] = tmp; + } + } + + return 0; +} + +static void bcmgenet_hfb_set_filter(struct bcmgenet_priv *priv, u32 *f_data, + u32 f_length, u32 rx_queue, int f_index) +{ + u32 base = f_index * priv->hw_params->hfb_filter_size; + int i; + + for (i = 0; i < f_length; i++) + bcmgenet_hfb_writel(priv, f_data[i], (base + i) * sizeof(u32)); + + bcmgenet_hfb_set_filter_length(priv, f_index, 2 * f_length); + bcmgenet_hfb_set_filter_rx_queue_mapping(priv, f_index, rx_queue); +} + +static int bcmgenet_hfb_create_rxnfc_filter(struct bcmgenet_priv *priv, + struct bcmgenet_rxnfc_rule *rule) +{ + struct ethtool_rx_flow_spec *fs = &rule->fs; + int err = 0, offset = 0, f_length = 0; + u16 val_16, mask_16; + u8 val_8, mask_8; + size_t size; + u32 *f_data; + + f_data = kcalloc(priv->hw_params->hfb_filter_size, sizeof(u32), + GFP_KERNEL); + if (!f_data) + return -ENOMEM; + + if (fs->flow_type & FLOW_MAC_EXT) { + bcmgenet_hfb_insert_data(f_data, 0, + &fs->h_ext.h_dest, &fs->m_ext.h_dest, + sizeof(fs->h_ext.h_dest)); + } + + if (fs->flow_type & FLOW_EXT) { + if (fs->m_ext.vlan_etype || + fs->m_ext.vlan_tci) { + bcmgenet_hfb_insert_data(f_data, 12, + &fs->h_ext.vlan_etype, + &fs->m_ext.vlan_etype, + sizeof(fs->h_ext.vlan_etype)); + bcmgenet_hfb_insert_data(f_data, 14, + &fs->h_ext.vlan_tci, + &fs->m_ext.vlan_tci, + sizeof(fs->h_ext.vlan_tci)); + offset += VLAN_HLEN; + f_length += DIV_ROUND_UP(VLAN_HLEN, 2); + } + } + + switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) { + case ETHER_FLOW: + f_length += DIV_ROUND_UP(ETH_HLEN, 2); + bcmgenet_hfb_insert_data(f_data, 0, + &fs->h_u.ether_spec.h_dest, + &fs->m_u.ether_spec.h_dest, + sizeof(fs->h_u.ether_spec.h_dest)); + bcmgenet_hfb_insert_data(f_data, ETH_ALEN, + &fs->h_u.ether_spec.h_source, + &fs->m_u.ether_spec.h_source, + sizeof(fs->h_u.ether_spec.h_source)); + bcmgenet_hfb_insert_data(f_data, (2 * ETH_ALEN) + offset, + &fs->h_u.ether_spec.h_proto, + &fs->m_u.ether_spec.h_proto, + sizeof(fs->h_u.ether_spec.h_proto)); + break; + case IP_USER_FLOW: + f_length += DIV_ROUND_UP(ETH_HLEN + 20, 2); + /* Specify IP Ether Type */ + val_16 = htons(ETH_P_IP); + mask_16 = 0xFFFF; + bcmgenet_hfb_insert_data(f_data, (2 * ETH_ALEN) + offset, + &val_16, &mask_16, sizeof(val_16)); + bcmgenet_hfb_insert_data(f_data, 15 + offset, + &fs->h_u.usr_ip4_spec.tos, + &fs->m_u.usr_ip4_spec.tos, + sizeof(fs->h_u.usr_ip4_spec.tos)); + bcmgenet_hfb_insert_data(f_data, 23 + offset, + &fs->h_u.usr_ip4_spec.proto, + &fs->m_u.usr_ip4_spec.proto, + sizeof(fs->h_u.usr_ip4_spec.proto)); + bcmgenet_hfb_insert_data(f_data, 26 + offset, + &fs->h_u.usr_ip4_spec.ip4src, + &fs->m_u.usr_ip4_spec.ip4src, + sizeof(fs->h_u.usr_ip4_spec.ip4src)); + bcmgenet_hfb_insert_data(f_data, 30 + offset, + &fs->h_u.usr_ip4_spec.ip4dst, + &fs->m_u.usr_ip4_spec.ip4dst, + sizeof(fs->h_u.usr_ip4_spec.ip4dst)); + if (!fs->m_u.usr_ip4_spec.l4_4_bytes) + break; + + /* Only supports 20 byte IPv4 header */ + val_8 = 0x45; + mask_8 = 0xFF; + bcmgenet_hfb_insert_data(f_data, ETH_HLEN + offset, + &val_8, &mask_8, + sizeof(val_8)); + size = sizeof(fs->h_u.usr_ip4_spec.l4_4_bytes); + bcmgenet_hfb_insert_data(f_data, + ETH_HLEN + 20 + offset, + &fs->h_u.usr_ip4_spec.l4_4_bytes, + &fs->m_u.usr_ip4_spec.l4_4_bytes, + size); + f_length += DIV_ROUND_UP(size, 2); + break; + } + + if (!fs->ring_cookie || fs->ring_cookie == RX_CLS_FLOW_WAKE) { + /* Ring 0 flows can be handled by the default Descriptor Ring + * We'll map them to ring 0, but don't enable the filter + */ + bcmgenet_hfb_set_filter(priv, f_data, f_length, 0, + fs->location); + rule->state = BCMGENET_RXNFC_STATE_DISABLED; + } else { + /* Other Rx rings are direct mapped here */ + bcmgenet_hfb_set_filter(priv, f_data, f_length, + fs->ring_cookie, fs->location); + bcmgenet_hfb_enable_filter(priv, fs->location); + rule->state = BCMGENET_RXNFC_STATE_ENABLED; + } + + kfree(f_data); + + return err; +} + +/* bcmgenet_hfb_add_filter + * + * Add new filter to Hardware Filter Block to match and direct Rx traffic to + * desired Rx queue. + * + * f_data is an array of unsigned 32-bit integers where each 32-bit integer + * provides filter data for 2 bytes (4 nibbles) of Rx frame: + * + * bits 31:20 - unused + * bit 19 - nibble 0 match enable + * bit 18 - nibble 1 match enable + * bit 17 - nibble 2 match enable + * bit 16 - nibble 3 match enable + * bits 15:12 - nibble 0 data + * bits 11:8 - nibble 1 data + * bits 7:4 - nibble 2 data + * bits 3:0 - nibble 3 data + * + * Example: + * In order to match: + * - Ethernet frame type = 0x0800 (IP) + * - IP version field = 4 + * - IP protocol field = 0x11 (UDP) + * + * The following filter is needed: + * u32 hfb_filter_ipv4_udp[] = { + * Rx frame offset 0x00: 0x00000000, 0x00000000, 0x00000000, 0x00000000, + * Rx frame offset 0x08: 0x00000000, 0x00000000, 0x000F0800, 0x00084000, + * Rx frame offset 0x10: 0x00000000, 0x00000000, 0x00000000, 0x00030011, + * }; + * + * To add the filter to HFB and direct the traffic to Rx queue 0, call: + * bcmgenet_hfb_add_filter(priv, hfb_filter_ipv4_udp, + * ARRAY_SIZE(hfb_filter_ipv4_udp), 0); + */ +int bcmgenet_hfb_add_filter(struct bcmgenet_priv *priv, u32 *f_data, + u32 f_length, u32 rx_queue) +{ + int f_index; + + f_index = bcmgenet_hfb_find_unused_filter(priv); + if (f_index < 0) + return -ENOMEM; + + if (f_length > priv->hw_params->hfb_filter_size) + return -EINVAL; + + bcmgenet_hfb_set_filter(priv, f_data, f_length, rx_queue, f_index); + bcmgenet_hfb_enable_filter(priv, f_index); + + return 0; +} + +/* bcmgenet_hfb_clear + * + * Clear Hardware Filter Block and disable all filtering. + */ +static void bcmgenet_hfb_clear(struct bcmgenet_priv *priv) +{ + u32 i; + + bcmgenet_hfb_reg_writel(priv, 0x0, HFB_CTRL); + bcmgenet_hfb_reg_writel(priv, 0x0, HFB_FLT_ENABLE_V3PLUS); + bcmgenet_hfb_reg_writel(priv, 0x0, HFB_FLT_ENABLE_V3PLUS + 4); + + for (i = DMA_INDEX2RING_0; i <= DMA_INDEX2RING_7; i++) + bcmgenet_rdma_writel(priv, 0x0, i); + + for (i = 0; i < (priv->hw_params->hfb_filter_cnt / 4); i++) + bcmgenet_hfb_reg_writel(priv, 0x0, + HFB_FLT_LEN_V3PLUS + i * sizeof(u32)); + + for (i = 0; i < priv->hw_params->hfb_filter_cnt * + priv->hw_params->hfb_filter_size; i++) + bcmgenet_hfb_writel(priv, 0x0, i * sizeof(u32)); +} + +static void bcmgenet_hfb_init(struct bcmgenet_priv *priv) +{ + int i; + + if (GENET_IS_V1(priv) || GENET_IS_V2(priv)) + return; + + INIT_LIST_HEAD(&priv->rxnfc_list); + for (i = 0; i < MAX_NUM_OF_FS_RULES; i++) { + INIT_LIST_HEAD(&priv->rxnfc_rules[i].list); + priv->rxnfc_rules[i].state = BCMGENET_RXNFC_STATE_UNUSED; + } + + bcmgenet_hfb_clear(priv); +} + static int bcmgenet_begin(struct net_device *dev) { struct bcmgenet_priv *priv = netdev_priv(dev); @@ -1045,6 +1421,229 @@ static int bcmgenet_set_eee(struct net_device *dev, struct ethtool_eee *e) return phy_ethtool_set_eee(dev->phydev, e); } +static int bcmgenet_validate_flow(struct net_device *dev, + struct ethtool_rxnfc *cmd) +{ + struct ethtool_usrip4_spec *l4_mask; + struct ethhdr *eth_mask; + + if (cmd->fs.location >= MAX_NUM_OF_FS_RULES) { + netdev_err(dev, "rxnfc: Invalid location (%d)\n", + cmd->fs.location); + return -EINVAL; + } + + switch (cmd->fs.flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) { + case IP_USER_FLOW: + l4_mask = &cmd->fs.m_u.usr_ip4_spec; + /* don't allow mask which isn't valid */ + if (VALIDATE_MASK(l4_mask->ip4src) || + VALIDATE_MASK(l4_mask->ip4dst) || + VALIDATE_MASK(l4_mask->l4_4_bytes) || + VALIDATE_MASK(l4_mask->proto) || + VALIDATE_MASK(l4_mask->ip_ver) || + VALIDATE_MASK(l4_mask->tos)) { + netdev_err(dev, "rxnfc: Unsupported mask\n"); + return -EINVAL; + } + break; + case ETHER_FLOW: + eth_mask = &cmd->fs.m_u.ether_spec; + /* don't allow mask which isn't valid */ + if (VALIDATE_MASK(eth_mask->h_source) || + VALIDATE_MASK(eth_mask->h_source) || + VALIDATE_MASK(eth_mask->h_proto)) { + netdev_err(dev, "rxnfc: Unsupported mask\n"); + return -EINVAL; + } + break; + default: + netdev_err(dev, "rxnfc: Unsupported flow type (0x%x)\n", + cmd->fs.flow_type); + return -EINVAL; + } + + if ((cmd->fs.flow_type & FLOW_EXT)) { + /* don't allow mask which isn't valid */ + if (VALIDATE_MASK(cmd->fs.m_ext.vlan_etype) || + VALIDATE_MASK(cmd->fs.m_ext.vlan_tci)) { + netdev_err(dev, "rxnfc: Unsupported mask\n"); + return -EINVAL; + } + if (cmd->fs.m_ext.data[0] || cmd->fs.m_ext.data[1]) { + netdev_err(dev, "rxnfc: user-def not supported\n"); + return -EINVAL; + } + } + + if ((cmd->fs.flow_type & FLOW_MAC_EXT)) { + /* don't allow mask which isn't valid */ + if (VALIDATE_MASK(cmd->fs.m_ext.h_dest)) { + netdev_err(dev, "rxnfc: Unsupported mask\n"); + return -EINVAL; + } + } + + return 0; +} + +static int bcmgenet_insert_flow(struct net_device *dev, + struct ethtool_rxnfc *cmd) +{ + struct bcmgenet_priv *priv = netdev_priv(dev); + struct bcmgenet_rxnfc_rule *loc_rule; + int err; + + if (priv->hw_params->hfb_filter_size < 128) { + netdev_err(dev, "rxnfc: Not supported by this device\n"); + return -EINVAL; + } + + if (cmd->fs.ring_cookie > priv->hw_params->rx_queues && + cmd->fs.ring_cookie != RX_CLS_FLOW_WAKE) { + netdev_err(dev, "rxnfc: Unsupported action (%llu)\n", + cmd->fs.ring_cookie); + return -EINVAL; + } + + err = bcmgenet_validate_flow(dev, cmd); + if (err) + return err; + + loc_rule = &priv->rxnfc_rules[cmd->fs.location]; + if (loc_rule->state == BCMGENET_RXNFC_STATE_ENABLED) + bcmgenet_hfb_disable_filter(priv, cmd->fs.location); + if (loc_rule->state != BCMGENET_RXNFC_STATE_UNUSED) + list_del(&loc_rule->list); + loc_rule->state = BCMGENET_RXNFC_STATE_UNUSED; + memcpy(&loc_rule->fs, &cmd->fs, + sizeof(struct ethtool_rx_flow_spec)); + + err = bcmgenet_hfb_create_rxnfc_filter(priv, loc_rule); + if (err) { + netdev_err(dev, "rxnfc: Could not install rule (%d)\n", + err); + return err; + } + + list_add_tail(&loc_rule->list, &priv->rxnfc_list); + + return 0; +} + +static int bcmgenet_delete_flow(struct net_device *dev, + struct ethtool_rxnfc *cmd) +{ + struct bcmgenet_priv *priv = netdev_priv(dev); + struct bcmgenet_rxnfc_rule *rule; + int err = 0; + + if (cmd->fs.location >= MAX_NUM_OF_FS_RULES) + return -EINVAL; + + rule = &priv->rxnfc_rules[cmd->fs.location]; + if (rule->state == BCMGENET_RXNFC_STATE_UNUSED) { + err = -ENOENT; + goto out; + } + + if (rule->state == BCMGENET_RXNFC_STATE_ENABLED) + bcmgenet_hfb_disable_filter(priv, cmd->fs.location); + if (rule->state != BCMGENET_RXNFC_STATE_UNUSED) + list_del(&rule->list); + rule->state = BCMGENET_RXNFC_STATE_UNUSED; + memset(&rule->fs, 0, sizeof(struct ethtool_rx_flow_spec)); + +out: + return err; +} + +static int bcmgenet_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd) +{ + struct bcmgenet_priv *priv = netdev_priv(dev); + int err = 0; + + switch (cmd->cmd) { + case ETHTOOL_SRXCLSRLINS: + err = bcmgenet_insert_flow(dev, cmd); + break; + case ETHTOOL_SRXCLSRLDEL: + err = bcmgenet_delete_flow(dev, cmd); + break; + default: + netdev_warn(priv->dev, "Unsupported ethtool command. (%d)\n", + cmd->cmd); + return -EINVAL; + } + + return err; +} + +static int bcmgenet_get_flow(struct net_device *dev, struct ethtool_rxnfc *cmd, + int loc) +{ + struct bcmgenet_priv *priv = netdev_priv(dev); + struct bcmgenet_rxnfc_rule *rule; + int err = 0; + + if (loc < 0 || loc >= MAX_NUM_OF_FS_RULES) + return -EINVAL; + + rule = &priv->rxnfc_rules[loc]; + if (rule->state == BCMGENET_RXNFC_STATE_UNUSED) + err = -ENOENT; + else + memcpy(&cmd->fs, &rule->fs, + sizeof(struct ethtool_rx_flow_spec)); + + return err; +} + +static int bcmgenet_get_num_flows(struct bcmgenet_priv *priv) +{ + struct list_head *pos; + int res = 0; + + list_for_each(pos, &priv->rxnfc_list) + res++; + + return res; +} + +static int bcmgenet_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, + u32 *rule_locs) +{ + struct bcmgenet_priv *priv = netdev_priv(dev); + struct bcmgenet_rxnfc_rule *rule; + int err = 0; + int i = 0; + + switch (cmd->cmd) { + case ETHTOOL_GRXRINGS: + cmd->data = priv->hw_params->rx_queues ?: 1; + break; + case ETHTOOL_GRXCLSRLCNT: + cmd->rule_cnt = bcmgenet_get_num_flows(priv); + cmd->data = MAX_NUM_OF_FS_RULES; + break; + case ETHTOOL_GRXCLSRULE: + err = bcmgenet_get_flow(dev, cmd, cmd->fs.location); + break; + case ETHTOOL_GRXCLSRLALL: + list_for_each_entry(rule, &priv->rxnfc_list, list) + if (i < cmd->rule_cnt) + rule_locs[i++] = rule->fs.location; + cmd->rule_cnt = i; + cmd->data = MAX_NUM_OF_FS_RULES; + break; + default: + err = -EOPNOTSUPP; + break; + } + + return err; +} + /* standard ethtool support functions. */ static const struct ethtool_ops bcmgenet_ethtool_ops = { .supported_coalesce_params = ETHTOOL_COALESCE_RX_USECS | @@ -1069,6 +1668,8 @@ static const struct ethtool_ops bcmgenet_ethtool_ops = { .get_link_ksettings = bcmgenet_get_link_ksettings, .set_link_ksettings = bcmgenet_set_link_ksettings, .get_ts_info = ethtool_op_get_ts_info, + .get_rxnfc = bcmgenet_get_rxnfc, + .set_rxnfc = bcmgenet_set_rxnfc, }; /* Power down the unimac, based on mode. */ @@ -2669,10 +3270,7 @@ static irqreturn_t bcmgenet_isr0(int irq, void *dev_id) static irqreturn_t bcmgenet_wol_isr(int irq, void *dev_id) { - struct bcmgenet_priv *priv = dev_id; - - pm_wakeup_event(&priv->pdev->dev, 0); - + /* Acknowledge the interrupt */ return IRQ_HANDLED; } @@ -2710,9 +3308,8 @@ static void bcmgenet_umac_reset(struct bcmgenet_priv *priv) static void bcmgenet_set_hw_addr(struct bcmgenet_priv *priv, unsigned char *addr) { - bcmgenet_umac_writel(priv, (addr[0] << 24) | (addr[1] << 16) | - (addr[2] << 8) | addr[3], UMAC_MAC0); - bcmgenet_umac_writel(priv, (addr[4] << 8) | addr[5], UMAC_MAC1); + bcmgenet_umac_writel(priv, get_unaligned_be32(&addr[0]), UMAC_MAC0); + bcmgenet_umac_writel(priv, get_unaligned_be16(&addr[4]), UMAC_MAC1); } static void bcmgenet_get_hw_addr(struct bcmgenet_priv *priv, @@ -2721,13 +3318,9 @@ static void bcmgenet_get_hw_addr(struct bcmgenet_priv *priv, u32 addr_tmp; addr_tmp = bcmgenet_umac_readl(priv, UMAC_MAC0); - addr[0] = addr_tmp >> 24; - addr[1] = (addr_tmp >> 16) & 0xff; - addr[2] = (addr_tmp >> 8) & 0xff; - addr[3] = addr_tmp & 0xff; + put_unaligned_be32(addr_tmp, &addr[0]); addr_tmp = bcmgenet_umac_readl(priv, UMAC_MAC1); - addr[4] = (addr_tmp >> 8) & 0xff; - addr[5] = addr_tmp & 0xff; + put_unaligned_be16(addr_tmp, &addr[4]); } /* Returns a reusable dma control register value */ @@ -2766,43 +3359,12 @@ static void bcmgenet_enable_dma(struct bcmgenet_priv *priv, u32 dma_ctrl) bcmgenet_tdma_writel(priv, reg, DMA_CTRL); } -/* bcmgenet_hfb_clear - * - * Clear Hardware Filter Block and disable all filtering. - */ -static void bcmgenet_hfb_clear(struct bcmgenet_priv *priv) -{ - u32 i; - - bcmgenet_hfb_reg_writel(priv, 0x0, HFB_CTRL); - bcmgenet_hfb_reg_writel(priv, 0x0, HFB_FLT_ENABLE_V3PLUS); - bcmgenet_hfb_reg_writel(priv, 0x0, HFB_FLT_ENABLE_V3PLUS + 4); - - for (i = DMA_INDEX2RING_0; i <= DMA_INDEX2RING_7; i++) - bcmgenet_rdma_writel(priv, 0x0, i); - - for (i = 0; i < (priv->hw_params->hfb_filter_cnt / 4); i++) - bcmgenet_hfb_reg_writel(priv, 0x0, - HFB_FLT_LEN_V3PLUS + i * sizeof(u32)); - - for (i = 0; i < priv->hw_params->hfb_filter_cnt * - priv->hw_params->hfb_filter_size; i++) - bcmgenet_hfb_writel(priv, 0x0, i * sizeof(u32)); -} - -static void bcmgenet_hfb_init(struct bcmgenet_priv *priv) -{ - if (GENET_IS_V1(priv) || GENET_IS_V2(priv)) - return; - - bcmgenet_hfb_clear(priv); -} - static void bcmgenet_netif_start(struct net_device *dev) { struct bcmgenet_priv *priv = netdev_priv(dev); /* Start the network engine */ + bcmgenet_set_rx_mode(dev); bcmgenet_enable_rx_napi(priv); umac_enable_set(priv, CMD_TX_EN | CMD_RX_EN, true); @@ -3421,8 +3983,6 @@ MODULE_DEVICE_TABLE(of, bcmgenet_match); static int bcmgenet_probe(struct platform_device *pdev) { struct bcmgenet_platform_data *pd = pdev->dev.platform_data; - struct device_node *dn = pdev->dev.of_node; - const struct of_device_id *of_id = NULL; const struct bcmgenet_plat_data *pdata; struct bcmgenet_priv *priv; struct net_device *dev; @@ -3437,12 +3997,6 @@ static int bcmgenet_probe(struct platform_device *pdev) return -ENOMEM; } - if (dn) { - of_id = of_match_node(bcmgenet_match, dn); - if (!of_id) - return -EINVAL; - } - priv = netdev_priv(dev); priv->irq0 = platform_get_irq(pdev, 0); if (priv->irq0 < 0) { @@ -3504,13 +4058,16 @@ static int bcmgenet_probe(struct platform_device *pdev) priv->dma_max_burst_length = DMA_MAX_BURST_LENGTH; } - priv->clk = devm_clk_get(&priv->pdev->dev, "enet"); + priv->clk = devm_clk_get_optional(&priv->pdev->dev, "enet"); if (IS_ERR(priv->clk)) { dev_dbg(&priv->pdev->dev, "failed to get enet clock\n"); - priv->clk = NULL; + err = PTR_ERR(priv->clk); + goto err; } - clk_prepare_enable(priv->clk); + err = clk_prepare_enable(priv->clk); + if (err) + goto err; bcmgenet_set_hw_params(priv); @@ -3528,16 +4085,18 @@ static int bcmgenet_probe(struct platform_device *pdev) priv->rx_buf_len = RX_BUF_LENGTH; INIT_WORK(&priv->bcmgenet_irq_work, bcmgenet_irq_task); - priv->clk_wol = devm_clk_get(&priv->pdev->dev, "enet-wol"); + priv->clk_wol = devm_clk_get_optional(&priv->pdev->dev, "enet-wol"); if (IS_ERR(priv->clk_wol)) { dev_dbg(&priv->pdev->dev, "failed to get enet-wol clock\n"); - priv->clk_wol = NULL; + err = PTR_ERR(priv->clk_wol); + goto err; } - priv->clk_eee = devm_clk_get(&priv->pdev->dev, "enet-eee"); + priv->clk_eee = devm_clk_get_optional(&priv->pdev->dev, "enet-eee"); if (IS_ERR(priv->clk_eee)) { dev_dbg(&priv->pdev->dev, "failed to get enet-eee clock\n"); - priv->clk_eee = NULL; + err = PTR_ERR(priv->clk_eee); + goto err; } /* If this is an internal GPHY, power it on now, before UniMAC is @@ -3546,7 +4105,7 @@ static int bcmgenet_probe(struct platform_device *pdev) if (device_get_phy_mode(&pdev->dev) == PHY_INTERFACE_MODE_INTERNAL) bcmgenet_power_up(priv, GENET_POWER_PASSIVE); - if ((pd) && (!IS_ERR_OR_NULL(pd->mac_address))) + if (pd && !IS_ERR_OR_NULL(pd->mac_address)) ether_addr_copy(dev->dev_addr, pd->mac_address); else if (!device_get_mac_address(&pdev->dev, dev->dev_addr, ETH_ALEN)) @@ -3612,11 +4171,10 @@ static void bcmgenet_shutdown(struct platform_device *pdev) } #ifdef CONFIG_PM_SLEEP -static int bcmgenet_resume(struct device *d) +static int bcmgenet_resume_noirq(struct device *d) { struct net_device *dev = dev_get_drvdata(d); struct bcmgenet_priv *priv = netdev_priv(dev); - unsigned long dma_ctrl; int ret; u32 reg; @@ -3628,6 +4186,38 @@ static int bcmgenet_resume(struct device *d) if (ret) return ret; + if (device_may_wakeup(d) && priv->wolopts) { + /* Account for Wake-on-LAN events and clear those events + * (Some devices need more time between enabling the clocks + * and the interrupt register reflecting the wake event so + * read the register twice) + */ + reg = bcmgenet_intrl2_0_readl(priv, INTRL2_CPU_STAT); + reg = bcmgenet_intrl2_0_readl(priv, INTRL2_CPU_STAT); + if (reg & UMAC_IRQ_WAKE_EVENT) + pm_wakeup_event(&priv->pdev->dev, 0); + } + + bcmgenet_intrl2_0_writel(priv, UMAC_IRQ_WAKE_EVENT, INTRL2_CPU_CLEAR); + + return 0; +} + +static int bcmgenet_resume(struct device *d) +{ + struct net_device *dev = dev_get_drvdata(d); + struct bcmgenet_priv *priv = netdev_priv(dev); + unsigned long dma_ctrl; + u32 offset, reg; + int ret; + + if (!netif_running(dev)) + return 0; + + /* From WOL-enabled suspend, switch to regular clock */ + if (device_may_wakeup(d) && priv->wolopts) + bcmgenet_power_up(priv, GENET_POWER_WOL_MAGIC); + /* If this is an internal GPHY, power it back on now, before UniMAC is * brought out of reset as absolutely no UniMAC activity is allowed */ @@ -3638,10 +4228,6 @@ static int bcmgenet_resume(struct device *d) init_umac(priv); - /* From WOL-enabled suspend, switch to regular clock */ - if (priv->wolopts) - clk_disable_unprepare(priv->clk_wol); - phy_init_hw(dev->phydev); /* Speed settings must be restored */ @@ -3653,15 +4239,17 @@ static int bcmgenet_resume(struct device *d) bcmgenet_set_hw_addr(priv, dev->dev_addr); + offset = HFB_FLT_ENABLE_V3PLUS; + bcmgenet_hfb_reg_writel(priv, priv->hfb_en[1], offset); + bcmgenet_hfb_reg_writel(priv, priv->hfb_en[2], offset + sizeof(u32)); + bcmgenet_hfb_reg_writel(priv, priv->hfb_en[0], HFB_CTRL); + if (priv->internal_phy) { reg = bcmgenet_ext_readl(priv, EXT_EXT_PWR_MGMT); reg |= EXT_ENERGY_DET_MASK; bcmgenet_ext_writel(priv, reg, EXT_EXT_PWR_MGMT); } - if (priv->wolopts) - bcmgenet_power_up(priv, GENET_POWER_WOL_MAGIC); - /* Disable RX/TX DMA and flush TX queues */ dma_ctrl = bcmgenet_dma_disable(priv); @@ -3698,7 +4286,7 @@ static int bcmgenet_suspend(struct device *d) { struct net_device *dev = dev_get_drvdata(d); struct bcmgenet_priv *priv = netdev_priv(dev); - int ret = 0; + u32 offset; if (!netif_running(dev)) return 0; @@ -3710,25 +4298,53 @@ static int bcmgenet_suspend(struct device *d) if (!device_may_wakeup(d)) phy_suspend(dev->phydev); + /* Preserve filter state and disable filtering */ + priv->hfb_en[0] = bcmgenet_hfb_reg_readl(priv, HFB_CTRL); + offset = HFB_FLT_ENABLE_V3PLUS; + priv->hfb_en[1] = bcmgenet_hfb_reg_readl(priv, offset); + priv->hfb_en[2] = bcmgenet_hfb_reg_readl(priv, offset + sizeof(u32)); + bcmgenet_hfb_reg_writel(priv, 0, HFB_CTRL); + + return 0; +} + +static int bcmgenet_suspend_noirq(struct device *d) +{ + struct net_device *dev = dev_get_drvdata(d); + struct bcmgenet_priv *priv = netdev_priv(dev); + int ret = 0; + + if (!netif_running(dev)) + return 0; + /* Prepare the device for Wake-on-LAN and switch to the slow clock */ - if (device_may_wakeup(d) && priv->wolopts) { + if (device_may_wakeup(d) && priv->wolopts) ret = bcmgenet_power_down(priv, GENET_POWER_WOL_MAGIC); - clk_prepare_enable(priv->clk_wol); - } else if (priv->internal_phy) { + else if (priv->internal_phy) ret = bcmgenet_power_down(priv, GENET_POWER_PASSIVE); - } + + /* Let the framework handle resumption and leave the clocks on */ + if (ret) + return ret; /* Turn off the clocks */ clk_disable_unprepare(priv->clk); - if (ret) - bcmgenet_resume(d); - - return ret; + return 0; } +#else +#define bcmgenet_suspend NULL +#define bcmgenet_suspend_noirq NULL +#define bcmgenet_resume NULL +#define bcmgenet_resume_noirq NULL #endif /* CONFIG_PM_SLEEP */ -static SIMPLE_DEV_PM_OPS(bcmgenet_pm_ops, bcmgenet_suspend, bcmgenet_resume); +static const struct dev_pm_ops bcmgenet_pm_ops = { + .suspend = bcmgenet_suspend, + .suspend_noirq = bcmgenet_suspend_noirq, + .resume = bcmgenet_resume, + .resume_noirq = bcmgenet_resume_noirq, +}; static const struct acpi_device_id genet_acpi_match[] = { { "BCM6E4E", (kernel_ulong_t)&bcm2711_plat_data }, @@ -3744,7 +4360,7 @@ static struct platform_driver bcmgenet_driver = { .name = "bcmgenet", .of_match_table = bcmgenet_match, .pm = &bcmgenet_pm_ops, - .acpi_match_table = ACPI_PTR(genet_acpi_match), + .acpi_match_table = genet_acpi_match, }, }; module_platform_driver(bcmgenet_driver); diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.h b/drivers/net/ethernet/broadcom/genet/bcmgenet.h index daf8fb2c39b6..a12cb59298f4 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet.h +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0-only */ /* - * Copyright (c) 2014-2017 Broadcom + * Copyright (c) 2014-2020 Broadcom */ #ifndef __BCMGENET_H__ @@ -14,6 +14,7 @@ #include <linux/if_vlan.h> #include <linux/phy.h> #include <linux/dim.h> +#include <linux/ethtool.h> /* total number of Buffer Descriptors, same for Rx/Tx */ #define TOTAL_DESC 256 @@ -31,6 +32,7 @@ #define DMA_MAX_BURST_LENGTH 0x10 /* misc. configuration */ +#define MAX_NUM_OF_FS_RULES 16 #define CLEAR_ALL_HFB 0xFF #define DMA_FC_THRESH_HI (TOTAL_DESC >> 4) #define DMA_FC_THRESH_LO 5 @@ -310,6 +312,8 @@ struct bcmgenet_mib_counters { #define UMAC_IRQ_HFB_SM (1 << 10) #define UMAC_IRQ_HFB_MM (1 << 11) #define UMAC_IRQ_MPD_R (1 << 12) +#define UMAC_IRQ_WAKE_EVENT (UMAC_IRQ_HFB_SM | UMAC_IRQ_HFB_MM | \ + UMAC_IRQ_MPD_R) #define UMAC_IRQ_RXDMA_MBDONE (1 << 13) #define UMAC_IRQ_RXDMA_PDONE (1 << 14) #define UMAC_IRQ_RXDMA_BDONE (1 << 15) @@ -608,6 +612,18 @@ struct bcmgenet_rx_ring { struct bcmgenet_priv *priv; }; +enum bcmgenet_rxnfc_state { + BCMGENET_RXNFC_STATE_UNUSED = 0, + BCMGENET_RXNFC_STATE_DISABLED, + BCMGENET_RXNFC_STATE_ENABLED +}; + +struct bcmgenet_rxnfc_rule { + struct list_head list; + struct ethtool_rx_flow_spec fs; + enum bcmgenet_rxnfc_state state; +}; + /* device context */ struct bcmgenet_priv { void __iomem *base; @@ -626,6 +642,8 @@ struct bcmgenet_priv { struct enet_cb *rx_cbs; unsigned int num_rx_bds; unsigned int rx_buf_len; + struct bcmgenet_rxnfc_rule rxnfc_rules[MAX_NUM_OF_FS_RULES]; + struct list_head rxnfc_list; struct bcmgenet_rx_ring rx_rings[DESC_INDEX + 1]; @@ -676,6 +694,9 @@ struct bcmgenet_priv { /* WOL */ struct clk *clk_wol; u32 wolopts; + u8 sopass[SOPASS_MAX]; + bool wol_active; + u32 hfb_en[3]; struct bcmgenet_mib_counters mib; diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c b/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c index c9a43695b182..4ea6a26b04f7 100644 --- a/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c +++ b/drivers/net/ethernet/broadcom/genet/bcmgenet_wol.c @@ -2,7 +2,7 @@ /* * Broadcom GENET (Gigabit Ethernet) Wake-on-LAN support * - * Copyright (c) 2014-2017 Broadcom + * Copyright (c) 2014-2020 Broadcom */ #define pr_fmt(fmt) "bcmgenet_wol: " fmt @@ -41,18 +41,13 @@ void bcmgenet_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { struct bcmgenet_priv *priv = netdev_priv(dev); - u32 reg; - wol->supported = WAKE_MAGIC | WAKE_MAGICSECURE; + wol->supported = WAKE_MAGIC | WAKE_MAGICSECURE | WAKE_FILTER; wol->wolopts = priv->wolopts; memset(wol->sopass, 0, sizeof(wol->sopass)); - if (wol->wolopts & WAKE_MAGICSECURE) { - reg = bcmgenet_umac_readl(priv, UMAC_MPD_PW_MS); - put_unaligned_be16(reg, &wol->sopass[0]); - reg = bcmgenet_umac_readl(priv, UMAC_MPD_PW_LS); - put_unaligned_be32(reg, &wol->sopass[2]); - } + if (wol->wolopts & WAKE_MAGICSECURE) + memcpy(wol->sopass, priv->sopass, sizeof(priv->sopass)); } /* ethtool function - set WOL (Wake on LAN) settings. @@ -62,25 +57,15 @@ int bcmgenet_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { struct bcmgenet_priv *priv = netdev_priv(dev); struct device *kdev = &priv->pdev->dev; - u32 reg; if (!device_can_wakeup(kdev)) return -ENOTSUPP; - if (wol->wolopts & ~(WAKE_MAGIC | WAKE_MAGICSECURE)) + if (wol->wolopts & ~(WAKE_MAGIC | WAKE_MAGICSECURE | WAKE_FILTER)) return -EINVAL; - reg = bcmgenet_umac_readl(priv, UMAC_MPD_CTRL); - if (wol->wolopts & WAKE_MAGICSECURE) { - bcmgenet_umac_writel(priv, get_unaligned_be16(&wol->sopass[0]), - UMAC_MPD_PW_MS); - bcmgenet_umac_writel(priv, get_unaligned_be32(&wol->sopass[2]), - UMAC_MPD_PW_LS); - reg |= MPD_PW_EN; - } else { - reg &= ~MPD_PW_EN; - } - bcmgenet_umac_writel(priv, reg, UMAC_MPD_CTRL); + if (wol->wolopts & WAKE_MAGICSECURE) + memcpy(priv->sopass, wol->sopass, sizeof(priv->sopass)); /* Flag the device and relevant IRQ as wakeup capable */ if (wol->wolopts) { @@ -120,12 +105,21 @@ static int bcmgenet_poll_wol_status(struct bcmgenet_priv *priv) return retries; } +static void bcmgenet_set_mpd_password(struct bcmgenet_priv *priv) +{ + bcmgenet_umac_writel(priv, get_unaligned_be16(&priv->sopass[0]), + UMAC_MPD_PW_MS); + bcmgenet_umac_writel(priv, get_unaligned_be32(&priv->sopass[2]), + UMAC_MPD_PW_LS); +} + int bcmgenet_wol_power_down_cfg(struct bcmgenet_priv *priv, enum bcmgenet_power_mode mode) { struct net_device *dev = priv->dev; + struct bcmgenet_rxnfc_rule *rule; + u32 reg, hfb_ctrl_reg, hfb_enable = 0; int retries = 0; - u32 reg; if (mode != GENET_POWER_WOL_MAGIC) { netif_err(priv, wol, dev, "unsupported mode: %d\n", mode); @@ -142,22 +136,48 @@ int bcmgenet_wol_power_down_cfg(struct bcmgenet_priv *priv, bcmgenet_umac_writel(priv, reg, UMAC_CMD); mdelay(10); - reg = bcmgenet_umac_readl(priv, UMAC_MPD_CTRL); - reg |= MPD_EN; - bcmgenet_umac_writel(priv, reg, UMAC_MPD_CTRL); + if (priv->wolopts & (WAKE_MAGIC | WAKE_MAGICSECURE)) { + reg = bcmgenet_umac_readl(priv, UMAC_MPD_CTRL); + reg |= MPD_EN; + if (priv->wolopts & WAKE_MAGICSECURE) { + bcmgenet_set_mpd_password(priv); + reg |= MPD_PW_EN; + } + bcmgenet_umac_writel(priv, reg, UMAC_MPD_CTRL); + } + + hfb_ctrl_reg = bcmgenet_hfb_reg_readl(priv, HFB_CTRL); + if (priv->wolopts & WAKE_FILTER) { + list_for_each_entry(rule, &priv->rxnfc_list, list) + if (rule->fs.ring_cookie == RX_CLS_FLOW_WAKE) + hfb_enable |= (1 << rule->fs.location); + reg = (hfb_ctrl_reg & ~RBUF_HFB_EN) | RBUF_ACPI_EN; + bcmgenet_hfb_reg_writel(priv, reg, HFB_CTRL); + } /* Do not leave UniMAC in MPD mode only */ retries = bcmgenet_poll_wol_status(priv); if (retries < 0) { reg = bcmgenet_umac_readl(priv, UMAC_MPD_CTRL); - reg &= ~MPD_EN; + reg &= ~(MPD_EN | MPD_PW_EN); bcmgenet_umac_writel(priv, reg, UMAC_MPD_CTRL); + bcmgenet_hfb_reg_writel(priv, hfb_ctrl_reg, HFB_CTRL); return retries; } netif_dbg(priv, wol, dev, "MPD WOL-ready status set after %d msec\n", retries); + clk_prepare_enable(priv->clk_wol); + priv->wol_active = 1; + + if (hfb_enable) { + bcmgenet_hfb_reg_writel(priv, hfb_enable, + HFB_FLT_ENABLE_V3PLUS + 4); + hfb_ctrl_reg = RBUF_HFB_EN | RBUF_ACPI_EN; + bcmgenet_hfb_reg_writel(priv, hfb_ctrl_reg, HFB_CTRL); + } + /* Enable CRC forward */ reg = bcmgenet_umac_readl(priv, UMAC_CMD); priv->crc_fwd_en = 1; @@ -173,6 +193,12 @@ int bcmgenet_wol_power_down_cfg(struct bcmgenet_priv *priv, bcmgenet_ext_writel(priv, reg, EXT_EXT_PWR_MGMT); } + reg = UMAC_IRQ_MPD_R; + if (hfb_enable) + reg |= UMAC_IRQ_HFB_SM | UMAC_IRQ_HFB_MM; + + bcmgenet_intrl2_0_writel(priv, reg, INTRL2_CPU_MASK_CLEAR); + return 0; } @@ -186,12 +212,22 @@ void bcmgenet_wol_power_up_cfg(struct bcmgenet_priv *priv, return; } + if (!priv->wol_active) + return; /* failed to suspend so skip the rest */ + + priv->wol_active = 0; + clk_disable_unprepare(priv->clk_wol); + + /* Disable Magic Packet Detection */ reg = bcmgenet_umac_readl(priv, UMAC_MPD_CTRL); - if (!(reg & MPD_EN)) - return; /* already powered up so skip the rest */ - reg &= ~MPD_EN; + reg &= ~(MPD_EN | MPD_PW_EN); bcmgenet_umac_writel(priv, reg, UMAC_MPD_CTRL); + /* Disable WAKE_FILTER Detection */ + reg = bcmgenet_hfb_reg_readl(priv, HFB_CTRL); + reg &= ~(RBUF_HFB_EN | RBUF_ACPI_EN); + bcmgenet_hfb_reg_writel(priv, reg, HFB_CTRL); + /* Disable CRC Forward */ reg = bcmgenet_umac_readl(priv, UMAC_CMD); reg &= ~CMD_CRC_FWD; diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index ff98a82b7bc4..7a3b22b35238 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -10797,17 +10797,15 @@ static int tg3_init_hw(struct tg3 *tp, bool reset_phy) #ifdef CONFIG_TIGON3_HWMON static void tg3_sd_scan_scratchpad(struct tg3 *tp, struct tg3_ocir *ocir) { + u32 off, len = TG3_OCIR_LEN; int i; - for (i = 0; i < TG3_SD_NUM_RECS; i++, ocir++) { - u32 off = i * TG3_OCIR_LEN, len = TG3_OCIR_LEN; - + for (i = 0, off = 0; i < TG3_SD_NUM_RECS; i++, ocir++, off += len) { tg3_ape_scratchpad_read(tp, (u32 *) ocir, off, len); - off += len; if (ocir->signature != TG3_OCIR_SIG_MAGIC || !(ocir->version_flags & TG3_OCIR_FLAG_ACTIVE)) - memset(ocir, 0, TG3_OCIR_LEN); + memset(ocir, 0, len); } } diff --git a/drivers/net/ethernet/cavium/liquidio/lio_core.c b/drivers/net/ethernet/cavium/liquidio/lio_core.c index d7e805749a5b..e40c64b79f66 100644 --- a/drivers/net/ethernet/cavium/liquidio/lio_core.c +++ b/drivers/net/ethernet/cavium/liquidio/lio_core.c @@ -782,7 +782,6 @@ static int liquidio_napi_poll(struct napi_struct *napi, int budget) if ((work_done < budget && tx_done) || (iq && iq->pkt_in_done >= MAX_REG_CNT) || (droq->pkt_count >= MAX_REG_CNT)) { - tx_done = 1; napi_complete_done(napi, work_done); octeon_enable_irq(droq->oct_dev, droq->q_no); diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_device.h b/drivers/net/ethernet/cavium/liquidio/octeon_device.h index 3d01d3602d8f..fb380b4f3e02 100644 --- a/drivers/net/ethernet/cavium/liquidio/octeon_device.h +++ b/drivers/net/ethernet/cavium/liquidio/octeon_device.h @@ -712,18 +712,6 @@ struct octeon_device *lio_get_device(u32 octeon_id); */ int lio_get_device_id(void *dev); -static inline u16 OCTEON_MAJOR_REV(struct octeon_device *oct) -{ - u16 rev = (oct->rev_id & 0xC) >> 2; - - return (rev == 0) ? 1 : rev; -} - -static inline u16 OCTEON_MINOR_REV(struct octeon_device *oct) -{ - return oct->rev_id & 0x3; -} - /** Read windowed register. * @param oct - pointer to the Octeon device. * @param addr - Address of the register to read. diff --git a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c index 9d868403d86c..cbaa1924afbe 100644 --- a/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c +++ b/drivers/net/ethernet/cavium/octeon/octeon_mgmt.c @@ -234,6 +234,11 @@ static void octeon_mgmt_rx_fill_ring(struct net_device *netdev) /* Put it in the ring. */ p->rx_ring[p->rx_next_fill] = re.d64; + /* Make sure there is no reorder of filling the ring and ringing + * the bell + */ + wmb(); + dma_sync_single_for_device(p->dev, p->rx_ring_handle, ring_size_to_bytes(OCTEON_MGMT_RX_RING_SIZE), DMA_BIDIRECTIONAL); diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c index b4b33368698f..2ba0ce115e63 100644 --- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c +++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c @@ -552,6 +552,7 @@ static inline bool nicvf_xdp_rx(struct nicvf *nic, struct bpf_prog *prog, xdp_set_data_meta_invalid(&xdp); xdp.data_end = xdp.data + len; xdp.rxq = &rq->xdp_rxq; + xdp.frame_sz = RCV_FRAG_LEN + XDP_PACKET_HEADROOM; orig_data = xdp.data; rcu_read_lock(); diff --git a/drivers/net/ethernet/chelsio/Kconfig b/drivers/net/ethernet/chelsio/Kconfig index 9909bfda167e..82cdfa51ce37 100644 --- a/drivers/net/ethernet/chelsio/Kconfig +++ b/drivers/net/ethernet/chelsio/Kconfig @@ -26,7 +26,7 @@ config CHELSIO_T1 This driver supports Chelsio gigabit and 10-gigabit Ethernet cards. More information about adapter features and performance tuning is in - <file:Documentation/networking/device_drivers/chelsio/cxgb.txt>. + <file:Documentation/networking/device_drivers/chelsio/cxgb.rst>. For general information about Chelsio and our products, visit our website at <http://www.chelsio.com>. diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h index e46a14f44a6f..cf69c6edcfec 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4.h @@ -60,6 +60,7 @@ #define CH_WARN(adap, fmt, ...) dev_warn(adap->pdev_dev, fmt, ## __VA_ARGS__) extern struct list_head adapter_list; +extern struct list_head uld_list; extern struct mutex uld_mutex; /* Suspend an Ethernet Tx queue with fewer available descriptors than this. @@ -466,8 +467,6 @@ static inline struct mbox_cmd *mbox_cmd_log_entry(struct mbox_cmd_log *log, return &((struct mbox_cmd *)&(log)[1])[entry_idx]; } -#include "t4fw_api.h" - #define FW_VERSION(chip) ( \ FW_HDR_FW_VER_MAJOR_G(chip##FW_VERSION_MAJOR) | \ FW_HDR_FW_VER_MINOR_G(chip##FW_VERSION_MINOR) | \ @@ -824,6 +823,13 @@ struct sge_uld_txq_info { u16 ntxq; /* # of egress uld queues */ }; +/* struct to maintain ULD list to reallocate ULD resources on hotplug */ +struct cxgb4_uld_list { + struct cxgb4_uld_info uld_info; + struct list_head list_node; + enum cxgb4_uld uld_type; +}; + enum sge_eosw_state { CXGB4_EO_STATE_CLOSED = 0, /* Not ready to accept traffic */ CXGB4_EO_STATE_FLOWC_OPEN_SEND, /* Send FLOWC open request */ @@ -1093,6 +1099,7 @@ struct adapter { /* TC u32 offload */ struct cxgb4_tc_u32_table *tc_u32; + struct chcr_ktls chcr_ktls; struct chcr_stats_debug chcr_stats; /* TC flower offload */ @@ -1127,19 +1134,20 @@ struct adapter { * programmed with various parameters. */ struct ch_sched_params { - s8 type; /* packet or flow */ + u8 type; /* packet or flow */ union { struct { - s8 level; /* scheduler hierarchy level */ - s8 mode; /* per-class or per-flow */ - s8 rateunit; /* bit or packet rate */ - s8 ratemode; /* %port relative or kbps absolute */ - s8 channel; /* scheduler channel [0..N] */ - s8 class; /* scheduler class [0..N] */ - s32 minrate; /* minimum rate */ - s32 maxrate; /* maximum rate */ - s16 weight; /* percent weight */ - s16 pktsize; /* average packet size */ + u8 level; /* scheduler hierarchy level */ + u8 mode; /* per-class or per-flow */ + u8 rateunit; /* bit or packet rate */ + u8 ratemode; /* %port relative or kbps absolute */ + u8 channel; /* scheduler channel [0..N] */ + u8 class; /* scheduler class [0..N] */ + u32 minrate; /* minimum rate */ + u32 maxrate; /* maximum rate */ + u16 weight; /* percent weight */ + u16 pktsize; /* average packet size */ + u16 burstsize; /* burst buffer size */ } params; } u; }; @@ -1954,9 +1962,10 @@ int t4_sge_ctxt_rd(struct adapter *adap, unsigned int mbox, unsigned int cid, enum ctxt_type ctype, u32 *data); int t4_sge_ctxt_rd_bd(struct adapter *adap, unsigned int cid, enum ctxt_type ctype, u32 *data); -int t4_sched_params(struct adapter *adapter, int type, int level, int mode, - int rateunit, int ratemode, int channel, int class, - int minrate, int maxrate, int weight, int pktsize); +int t4_sched_params(struct adapter *adapter, u8 type, u8 level, u8 mode, + u8 rateunit, u8 ratemode, u8 channel, u8 class, + u32 minrate, u32 maxrate, u16 weight, u16 pktsize, + u16 burstsize); void t4_sge_decode_idma_state(struct adapter *adapter, int state); void t4_idma_monitor_init(struct adapter *adapter, struct sge_idma_monitor_state *idma); @@ -2052,4 +2061,7 @@ int cxgb_open(struct net_device *dev); int cxgb_close(struct net_device *dev); void cxgb4_enable_rx(struct adapter *adap, struct sge_rspq *q); void cxgb4_quiesce_rx(struct sge_rspq *q); +#ifdef CONFIG_CHELSIO_TLS_DEVICE +int cxgb4_set_ktls_feature(struct adapter *adap, bool enable); +#endif #endif /* __CXGB4_H__ */ diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c index ebed99f3d4cf..41315712deb8 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_debugfs.c @@ -49,6 +49,7 @@ #include "cudbg_lib_common.h" #include "cudbg_entity.h" #include "cudbg_lib.h" +#include "cxgb4_tc_mqprio.h" /* generic seq_file support for showing a table of size rows x width. */ static void *seq_tab_get_idx(struct seq_tab *tb, loff_t pos) @@ -1812,12 +1813,8 @@ static int mps_tcam_show(struct seq_file *seq, void *v) /* Inner header lookup */ if (lookup_type && (lookup_type != DATALKPTYPE_M)) { seq_printf(seq, - "%3u %02x:%02x:%02x:%02x:%02x:%02x " - "%012llx %06x %06x - - %3c" - " 'I' %4x " - "%3c %#x%4u%4d", idx, addr[0], - addr[1], addr[2], addr[3], - addr[4], addr[5], + "%3u %pM %012llx %06x %06x - - %3c 'I' %4x %3c %#x%4u%4d", + idx, addr, (unsigned long long)mask, vniy, (vnix | vniy), dip_hit ? 'Y' : 'N', @@ -1829,10 +1826,8 @@ static int mps_tcam_show(struct seq_file *seq, void *v) T6_VF_G(cls_lo) : -1); } else { seq_printf(seq, - "%3u %02x:%02x:%02x:%02x:%02x:%02x " - "%012llx - - ", - idx, addr[0], addr[1], addr[2], - addr[3], addr[4], addr[5], + "%3u %pM %012llx - - ", + idx, addr, (unsigned long long)mask); if (vlan_vld) @@ -1850,10 +1845,8 @@ static int mps_tcam_show(struct seq_file *seq, void *v) T6_VF_G(cls_lo) : -1); } } else - seq_printf(seq, "%3u %02x:%02x:%02x:%02x:%02x:%02x " - "%012llx%3c %#x%4u%4d", - idx, addr[0], addr[1], addr[2], addr[3], - addr[4], addr[5], (unsigned long long)mask, + seq_printf(seq, "%3u %pM %012llx%3c %#x%4u%4d", + idx, addr, (unsigned long long)mask, (cls_lo & SRAM_VLD_F) ? 'Y' : 'N', PORTMAP_G(cls_hi), PF_G(cls_lo), @@ -2657,32 +2650,19 @@ static int sge_qinfo_uld_ciq_entries(const struct adapter *adap, int uld) static int sge_qinfo_show(struct seq_file *seq, void *v) { - int eth_entries, ctrl_entries, eo_entries = 0; + int eth_entries, ctrl_entries, eohw_entries = 0, eosw_entries = 0; int uld_rxq_entries[CXGB4_ULD_MAX] = { 0 }; int uld_ciq_entries[CXGB4_ULD_MAX] = { 0 }; int uld_txq_entries[CXGB4_TX_MAX] = { 0 }; const struct sge_uld_txq_info *utxq_info; const struct sge_uld_rxq_info *urxq_info; + struct cxgb4_tc_port_mqprio *port_mqprio; struct adapter *adap = seq->private; - int i, n, r = (uintptr_t)v - 1; + int i, j, n, r = (uintptr_t)v - 1; struct sge *s = &adap->sge; eth_entries = DIV_ROUND_UP(adap->sge.ethqsets, 4); ctrl_entries = DIV_ROUND_UP(MAX_CTRL_QUEUES, 4); - if (adap->sge.eohw_txq) - eo_entries = DIV_ROUND_UP(adap->sge.eoqsets, 4); - - mutex_lock(&uld_mutex); - if (s->uld_txq_info) - for (i = 0; i < ARRAY_SIZE(uld_txq_entries); i++) - uld_txq_entries[i] = sge_qinfo_uld_txq_entries(adap, i); - - if (s->uld_rxq_info) { - for (i = 0; i < ARRAY_SIZE(uld_rxq_entries); i++) { - uld_rxq_entries[i] = sge_qinfo_uld_rxq_entries(adap, i); - uld_ciq_entries[i] = sge_qinfo_uld_ciq_entries(adap, i); - } - } if (r) seq_putc(seq, '\n'); @@ -2759,11 +2739,21 @@ do { \ RL("FLLow:", fl.low); RL("FLStarving:", fl.starving); - goto unlock; + goto out; } r -= eth_entries; - if (r < eo_entries) { + if (!adap->tc_mqprio) + goto skip_mqprio; + + mutex_lock(&adap->tc_mqprio->mqprio_mutex); + if (!refcount_read(&adap->tc_mqprio->refcnt)) { + mutex_unlock(&adap->tc_mqprio->mqprio_mutex); + goto skip_mqprio; + } + + eohw_entries = DIV_ROUND_UP(adap->sge.eoqsets, 4); + if (r < eohw_entries) { int base_qset = r * 4; const struct sge_ofld_rxq *rx = &s->eohw_rxq[base_qset]; const struct sge_eohw_txq *tx = &s->eohw_txq[base_qset]; @@ -2808,10 +2798,71 @@ do { \ RL("FLLow:", fl.low); RL("FLStarving:", fl.starving); - goto unlock; + mutex_unlock(&adap->tc_mqprio->mqprio_mutex); + goto out; + } + + r -= eohw_entries; + for (j = 0; j < adap->params.nports; j++) { + int entries; + u8 tc; + + port_mqprio = &adap->tc_mqprio->port_mqprio[j]; + entries = 0; + for (tc = 0; tc < port_mqprio->mqprio.qopt.num_tc; tc++) + entries += port_mqprio->mqprio.qopt.count[tc]; + + if (!entries) + continue; + + eosw_entries = DIV_ROUND_UP(entries, 4); + if (r < eosw_entries) { + const struct sge_eosw_txq *tx; + + n = min(4, entries - 4 * r); + tx = &port_mqprio->eosw_txq[4 * r]; + + S("QType:", "EOSW-TXQ"); + S("Interface:", + adap->port[j] ? adap->port[j]->name : "N/A"); + T("EOTID:", hwtid); + T("HWQID:", hwqid); + T("State:", state); + T("Size:", ndesc); + T("In-Use:", inuse); + T("Credits:", cred); + T("Compl:", ncompl); + T("Last-Compl:", last_compl); + T("PIDX:", pidx); + T("Last-PIDX:", last_pidx); + T("CIDX:", cidx); + T("Last-CIDX:", last_cidx); + T("FLOWC-IDX:", flowc_idx); + + mutex_unlock(&adap->tc_mqprio->mqprio_mutex); + goto out; + } + + r -= eosw_entries; + } + mutex_unlock(&adap->tc_mqprio->mqprio_mutex); + +skip_mqprio: + if (!is_uld(adap)) + goto skip_uld; + + mutex_lock(&uld_mutex); + if (s->uld_txq_info) + for (i = 0; i < ARRAY_SIZE(uld_txq_entries); i++) + uld_txq_entries[i] = sge_qinfo_uld_txq_entries(adap, i); + + if (s->uld_rxq_info) { + for (i = 0; i < ARRAY_SIZE(uld_rxq_entries); i++) { + uld_rxq_entries[i] = sge_qinfo_uld_rxq_entries(adap, i); + uld_ciq_entries[i] = sge_qinfo_uld_ciq_entries(adap, i); + } } - r -= eo_entries; if (r < uld_txq_entries[CXGB4_TX_OFLD]) { const struct sge_uld_txq *tx; @@ -2994,6 +3045,9 @@ do { \ } r -= uld_txq_entries[CXGB4_TX_CRYPTO]; + mutex_unlock(&uld_mutex); + +skip_uld: if (r < ctrl_entries) { const struct sge_ctrl_txq *tx = &s->ctrlq[r * 4]; @@ -3008,7 +3062,7 @@ do { \ TL("TxQFull:", q.stops); TL("TxQRestarts:", q.restarts); - goto unlock; + goto out; } r -= ctrl_entries; @@ -3026,11 +3080,9 @@ do { \ seq_printf(seq, "%-12s %16u\n", "Intr pktcnt:", s->counter_val[evtq->pktcnt_idx]); - goto unlock; + goto out; } -unlock: - mutex_unlock(&uld_mutex); #undef R #undef RL #undef T @@ -3039,13 +3091,38 @@ unlock: #undef R3 #undef T3 #undef S3 +out: + return 0; + +unlock: + mutex_unlock(&uld_mutex); return 0; } static int sge_queue_entries(const struct adapter *adap) { - int tot_uld_entries = 0; - int i; + int i, tot_uld_entries = 0, eohw_entries = 0, eosw_entries = 0; + + if (adap->tc_mqprio) { + struct cxgb4_tc_port_mqprio *port_mqprio; + u8 tc; + + mutex_lock(&adap->tc_mqprio->mqprio_mutex); + if (adap->sge.eohw_txq) + eohw_entries = DIV_ROUND_UP(adap->sge.eoqsets, 4); + + for (i = 0; i < adap->params.nports; i++) { + u32 entries = 0; + + port_mqprio = &adap->tc_mqprio->port_mqprio[i]; + for (tc = 0; tc < port_mqprio->mqprio.qopt.num_tc; tc++) + entries += port_mqprio->mqprio.qopt.count[tc]; + + if (entries) + eosw_entries += DIV_ROUND_UP(entries, 4); + } + mutex_unlock(&adap->tc_mqprio->mqprio_mutex); + } if (!is_uld(adap)) goto lld_only; @@ -3062,8 +3139,7 @@ static int sge_queue_entries(const struct adapter *adap) lld_only: return DIV_ROUND_UP(adap->sge.ethqsets, 4) + - (adap->sge.eohw_txq ? DIV_ROUND_UP(adap->sge.eoqsets, 4) : 0) + - tot_uld_entries + + eohw_entries + eosw_entries + tot_uld_entries + DIV_ROUND_UP(MAX_CTRL_QUEUES, 4) + 1; } @@ -3244,6 +3320,10 @@ static int tid_info_show(struct seq_file *seq, void *v) if (t->nhpftids) seq_printf(seq, "HPFTID range: %u..%u\n", t->hpftid_base, t->hpftid_base + t->nhpftids - 1); + if (t->neotids) + seq_printf(seq, "EOTID range: %u..%u, in use: %u\n", + t->eotid_base, t->eotid_base + t->neotids - 1, + atomic_read(&t->eotids_in_use)); if (t->ntids) seq_printf(seq, "HW TID usage: %u IP users, %u IPv6 users\n", t4_read_reg(adap, LE_DB_ACT_CNT_IPV4_A), @@ -3411,6 +3491,8 @@ static int chcr_stats_show(struct seq_file *seq, void *v) atomic_read(&adap->chcr_stats.tls_key)); #ifdef CONFIG_CHELSIO_TLS_DEVICE seq_puts(seq, "\nChelsio KTLS Crypto Accelerator Stats\n"); + seq_printf(seq, "Tx TLS offload refcount: %20u\n", + refcount_read(&adap->chcr_ktls.ktls_refcount)); seq_printf(seq, "Tx HW offload contexts added: %20llu\n", atomic64_read(&adap->chcr_stats.ktls_tx_ctx)); seq_printf(seq, "Tx connection created: %20llu\n", diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index a70018f067aa..854b1717a70d 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -66,6 +66,9 @@ #include <linux/crash_dump.h> #include <net/udp_tunnel.h> #include <net/xfrm.h> +#if defined(CONFIG_CHELSIO_TLS_DEVICE) +#include <net/tls.h> +#endif #include "cxgb4.h" #include "cxgb4_filter.h" @@ -180,6 +183,7 @@ static struct dentry *cxgb4_debugfs_root; LIST_HEAD(adapter_list); DEFINE_MUTEX(uld_mutex); +LIST_HEAD(uld_list); static int cfg_queues(struct adapter *adap); @@ -1579,6 +1583,7 @@ static int tid_init(struct tid_info *t) atomic_set(&t->tids_in_use, 0); atomic_set(&t->conns_in_use, 0); atomic_set(&t->hash_tids_in_use, 0); + atomic_set(&t->eotids_in_use, 0); /* Setup the free list for atid_tab and clear the stid bitmap. */ if (natids) { @@ -3021,7 +3026,7 @@ static int cxgb4_mgmt_set_vf_rate(struct net_device *dev, int vf, SCHED_CLASS_RATEUNIT_BITS, SCHED_CLASS_RATEMODE_ABS, pi->tx_chan, class_id, 0, - max_tx_rate * 1000, 0, pktsize); + max_tx_rate * 1000, 0, pktsize, 0); if (ret) { dev_err(adap->pdev_dev, "Err %d for Traffic Class config\n", ret); @@ -6062,6 +6067,79 @@ static int cxgb4_iov_configure(struct pci_dev *pdev, int num_vfs) } #endif /* CONFIG_PCI_IOV */ +#if defined(CONFIG_CHELSIO_TLS_DEVICE) + +static int cxgb4_ktls_dev_add(struct net_device *netdev, struct sock *sk, + enum tls_offload_ctx_dir direction, + struct tls_crypto_info *crypto_info, + u32 tcp_sn) +{ + struct adapter *adap = netdev2adap(netdev); + int ret = 0; + + mutex_lock(&uld_mutex); + if (!adap->uld[CXGB4_ULD_CRYPTO].handle) { + dev_err(adap->pdev_dev, "chcr driver is not loaded\n"); + ret = -EOPNOTSUPP; + goto out_unlock; + } + + if (!adap->uld[CXGB4_ULD_CRYPTO].tlsdev_ops) { + dev_err(adap->pdev_dev, + "chcr driver has no registered tlsdev_ops()\n"); + ret = -EOPNOTSUPP; + goto out_unlock; + } + + ret = cxgb4_set_ktls_feature(adap, FW_PARAMS_PARAM_DEV_KTLS_HW_ENABLE); + if (ret) + goto out_unlock; + + ret = adap->uld[CXGB4_ULD_CRYPTO].tlsdev_ops->tls_dev_add(netdev, sk, + direction, + crypto_info, + tcp_sn); + /* if there is a failure, clear the refcount */ + if (ret) + cxgb4_set_ktls_feature(adap, + FW_PARAMS_PARAM_DEV_KTLS_HW_DISABLE); +out_unlock: + mutex_unlock(&uld_mutex); + return ret; +} + +static void cxgb4_ktls_dev_del(struct net_device *netdev, + struct tls_context *tls_ctx, + enum tls_offload_ctx_dir direction) +{ + struct adapter *adap = netdev2adap(netdev); + + mutex_lock(&uld_mutex); + if (!adap->uld[CXGB4_ULD_CRYPTO].handle) { + dev_err(adap->pdev_dev, "chcr driver is not loaded\n"); + goto out_unlock; + } + + if (!adap->uld[CXGB4_ULD_CRYPTO].tlsdev_ops) { + dev_err(adap->pdev_dev, + "chcr driver has no registered tlsdev_ops\n"); + goto out_unlock; + } + + adap->uld[CXGB4_ULD_CRYPTO].tlsdev_ops->tls_dev_del(netdev, tls_ctx, + direction); + cxgb4_set_ktls_feature(adap, FW_PARAMS_PARAM_DEV_KTLS_HW_DISABLE); + +out_unlock: + mutex_unlock(&uld_mutex); +} + +static const struct tlsdev_ops cxgb4_ktls_ops = { + .tls_dev_add = cxgb4_ktls_dev_add, + .tls_dev_del = cxgb4_ktls_dev_del, +}; +#endif /* CONFIG_CHELSIO_TLS_DEVICE */ + static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) { struct net_device *netdev; @@ -6311,7 +6389,14 @@ static int init_one(struct pci_dev *pdev, const struct pci_device_id *ent) netdev->hw_features |= NETIF_F_HIGHDMA; netdev->features |= netdev->hw_features; netdev->vlan_features = netdev->features & VLAN_FEAT; - +#if defined(CONFIG_CHELSIO_TLS_DEVICE) + if (pi->adapter->params.crypto & FW_CAPS_CONFIG_TLS_HW) { + netdev->hw_features |= NETIF_F_HW_TLS_TX; + netdev->tlsdev_ops = &cxgb4_ktls_ops; + /* initialize the refcount */ + refcount_set(&pi->adapter->chcr_ktls.ktls_refcount, 0); + } +#endif netdev->priv_flags |= IFF_UNICAST_FLT; /* MTU range: 81 - 9600 */ @@ -6518,11 +6603,8 @@ fw_attach_fail: /* PCIe EEH recovery on powerpc platforms needs fundamental reset */ pdev->needs_freset = 1; - if (is_uld(adapter)) { - mutex_lock(&uld_mutex); - list_add_tail(&adapter->list_node, &adapter_list); - mutex_unlock(&uld_mutex); - } + if (is_uld(adapter)) + cxgb4_uld_enable(adapter); if (!is_t4(adapter->params.chip)) cxgb4_ptp_init(adapter); diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c index e6af4906d674..ae7123a9de8e 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c @@ -342,6 +342,13 @@ static int cxgb4_mqprio_alloc_tc(struct net_device *dev, p.u.params.minrate = div_u64(mqprio->min_rate[i] * 8, 1000); p.u.params.maxrate = div_u64(mqprio->max_rate[i] * 8, 1000); + /* Request larger burst buffer for smaller MTU, so + * that hardware can work on more data per burst + * cycle. + */ + if (dev->mtu <= ETH_DATA_LEN) + p.u.params.burstsize = 8 * dev->mtu; + e = cxgb4_sched_class_alloc(dev, &p); if (!e) { ret = -ENOMEM; @@ -567,6 +574,7 @@ static void cxgb4_mqprio_disable_offload(struct net_device *dev) int cxgb4_setup_tc_mqprio(struct net_device *dev, struct tc_mqprio_qopt_offload *mqprio) { + struct adapter *adap = netdev2adap(dev); bool needs_bring_up = false; int ret; @@ -574,6 +582,8 @@ int cxgb4_setup_tc_mqprio(struct net_device *dev, if (ret) return ret; + mutex_lock(&adap->tc_mqprio->mqprio_mutex); + /* To configure tc params, the current allocated EOTIDs must * be freed up. However, they can't be freed up if there's * traffic running on the interface. So, ensure interface is @@ -609,6 +619,7 @@ out: if (needs_bring_up) cxgb_open(dev); + mutex_unlock(&adap->tc_mqprio->mqprio_mutex); return ret; } @@ -621,6 +632,7 @@ void cxgb4_mqprio_stop_offload(struct adapter *adap) if (!adap->tc_mqprio || !adap->tc_mqprio->port_mqprio) return; + mutex_lock(&adap->tc_mqprio->mqprio_mutex); for_each_port(adap, i) { dev = adap->port[i]; if (!dev) @@ -632,6 +644,7 @@ void cxgb4_mqprio_stop_offload(struct adapter *adap) cxgb4_mqprio_disable_offload(dev); } + mutex_unlock(&adap->tc_mqprio->mqprio_mutex); } int cxgb4_init_tc_mqprio(struct adapter *adap) @@ -653,6 +666,8 @@ int cxgb4_init_tc_mqprio(struct adapter *adap) goto out_free_mqprio; } + mutex_init(&tc_mqprio->mqprio_mutex); + tc_mqprio->port_mqprio = tc_port_mqprio; for (i = 0; i < adap->params.nports; i++) { port_mqprio = &tc_mqprio->port_mqprio[i]; @@ -687,6 +702,7 @@ void cxgb4_cleanup_tc_mqprio(struct adapter *adap) u8 i; if (adap->tc_mqprio) { + mutex_lock(&adap->tc_mqprio->mqprio_mutex); if (adap->tc_mqprio->port_mqprio) { for (i = 0; i < adap->params.nports; i++) { struct net_device *dev = adap->port[i]; @@ -698,6 +714,7 @@ void cxgb4_cleanup_tc_mqprio(struct adapter *adap) } kfree(adap->tc_mqprio->port_mqprio); } + mutex_unlock(&adap->tc_mqprio->mqprio_mutex); kfree(adap->tc_mqprio); } } diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.h index ff8794132b22..be96f1dc0372 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.h @@ -33,6 +33,7 @@ struct cxgb4_tc_port_mqprio { struct cxgb4_tc_mqprio { refcount_t refcnt; /* Refcount for adapter-wide resources */ + struct mutex mqprio_mutex; /* Lock for accessing MQPRIO info */ struct cxgb4_tc_port_mqprio *port_mqprio; /* Per port MQPRIO info */ }; diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c index e65b52375dd8..0307e9c69a47 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.c @@ -174,13 +174,14 @@ static int setup_sge_queues_uld(struct adapter *adap, unsigned int uld_type, bool lro) { struct sge_uld_rxq_info *rxq_info = adap->sge.uld_rxq_info[uld_type]; - int i, ret = 0; + int i, ret; - ret = !(!alloc_uld_rxqs(adap, rxq_info, lro)); + ret = alloc_uld_rxqs(adap, rxq_info, lro); + if (ret) + return ret; /* Tell uP to route control queue completions to rdma rspq */ - if (adap->flags & CXGB4_FULL_INIT_DONE && - !ret && uld_type == CXGB4_ULD_RDMA) { + if (adap->flags & CXGB4_FULL_INIT_DONE && uld_type == CXGB4_ULD_RDMA) { struct sge *s = &adap->sge; unsigned int cmplqid; u32 param, cmdop; @@ -662,25 +663,129 @@ static int uld_attach(struct adapter *adap, unsigned int uld) return 0; } +static bool cxgb4_uld_in_use(struct adapter *adap) +{ + const struct tid_info *t = &adap->tids; + + return (atomic_read(&t->conns_in_use) || t->stids_in_use); +} + #ifdef CONFIG_CHELSIO_TLS_DEVICE /* cxgb4_set_ktls_feature: request FW to enable/disable ktls settings. * @adap: adapter info * @enable: 1 to enable / 0 to disable ktls settings. */ -static void cxgb4_set_ktls_feature(struct adapter *adap, bool enable) +int cxgb4_set_ktls_feature(struct adapter *adap, bool enable) { - u32 params = (FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | - FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_KTLS_TX_HW) | - FW_PARAMS_PARAM_Y_V(enable)); int ret = 0; + u32 params = + FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) | + FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_KTLS_HW) | + FW_PARAMS_PARAM_Y_V(enable) | + FW_PARAMS_PARAM_Z_V(FW_PARAMS_PARAM_DEV_KTLS_HW_USER_ENABLE); + + if (enable) { + if (!refcount_read(&adap->chcr_ktls.ktls_refcount)) { + /* At this moment if ULD connection are up means, other + * ULD is/are already active, return failure. + */ + if (cxgb4_uld_in_use(adap)) { + dev_warn(adap->pdev_dev, + "ULD connections (tid/stid) active. Can't enable kTLS\n"); + return -EINVAL; + } + ret = t4_set_params(adap, adap->mbox, adap->pf, + 0, 1, ¶ms, ¶ms); + if (ret) + return ret; + refcount_set(&adap->chcr_ktls.ktls_refcount, 1); + pr_info("kTLS has been enabled. Restrictions placed on ULD support\n"); + } else { + /* ktls settings already up, just increment refcount. */ + refcount_inc(&adap->chcr_ktls.ktls_refcount); + } + } else { + /* return failure if refcount is already 0. */ + if (!refcount_read(&adap->chcr_ktls.ktls_refcount)) + return -EINVAL; + /* decrement refcount and test, if 0, disable ktls feature, + * else return command success. + */ + if (refcount_dec_and_test(&adap->chcr_ktls.ktls_refcount)) { + ret = t4_set_params(adap, adap->mbox, adap->pf, + 0, 1, ¶ms, ¶ms); + if (ret) + return ret; + pr_info("kTLS is disabled. Restrictions on ULD support removed\n"); + } + } - ret = t4_set_params(adap, adap->mbox, adap->pf, 0, 1, ¶ms, ¶ms); - /* if fw returns failure, clear the ktls flag */ - if (ret) - adap->params.crypto &= ~ULP_CRYPTO_KTLS_INLINE; + return ret; } #endif +static void cxgb4_uld_alloc_resources(struct adapter *adap, + enum cxgb4_uld type, + const struct cxgb4_uld_info *p) +{ + int ret = 0; + + if ((type == CXGB4_ULD_CRYPTO && !is_pci_uld(adap)) || + (type != CXGB4_ULD_CRYPTO && !is_offload(adap))) + return; + if (type == CXGB4_ULD_ISCSIT && is_t4(adap->params.chip)) + return; + ret = cfg_queues_uld(adap, type, p); + if (ret) + goto out; + ret = setup_sge_queues_uld(adap, type, p->lro); + if (ret) + goto free_queues; + if (adap->flags & CXGB4_USING_MSIX) { + ret = request_msix_queue_irqs_uld(adap, type); + if (ret) + goto free_rxq; + } + if (adap->flags & CXGB4_FULL_INIT_DONE) + enable_rx_uld(adap, type); + if (adap->uld[type].add) + goto free_irq; + ret = setup_sge_txq_uld(adap, type, p); + if (ret) + goto free_irq; + adap->uld[type] = *p; + ret = uld_attach(adap, type); + if (ret) + goto free_txq; + return; +free_txq: + release_sge_txq_uld(adap, type); +free_irq: + if (adap->flags & CXGB4_FULL_INIT_DONE) + quiesce_rx_uld(adap, type); + if (adap->flags & CXGB4_USING_MSIX) + free_msix_queue_irqs_uld(adap, type); +free_rxq: + free_sge_queues_uld(adap, type); +free_queues: + free_queues_uld(adap, type); +out: + dev_warn(adap->pdev_dev, + "ULD registration failed for uld type %d\n", type); +} + +void cxgb4_uld_enable(struct adapter *adap) +{ + struct cxgb4_uld_list *uld_entry; + + mutex_lock(&uld_mutex); + list_add_tail(&adap->list_node, &adapter_list); + list_for_each_entry(uld_entry, &uld_list, list_node) + cxgb4_uld_alloc_resources(adap, uld_entry->uld_type, + &uld_entry->uld_info); + mutex_unlock(&uld_mutex); +} + /* cxgb4_register_uld - register an upper-layer driver * @type: the ULD type * @p: the ULD methods @@ -691,63 +796,23 @@ static void cxgb4_set_ktls_feature(struct adapter *adap, bool enable) void cxgb4_register_uld(enum cxgb4_uld type, const struct cxgb4_uld_info *p) { + struct cxgb4_uld_list *uld_entry; struct adapter *adap; - int ret = 0; if (type >= CXGB4_ULD_MAX) return; + uld_entry = kzalloc(sizeof(*uld_entry), GFP_KERNEL); + if (!uld_entry) + return; + + memcpy(&uld_entry->uld_info, p, sizeof(struct cxgb4_uld_info)); mutex_lock(&uld_mutex); - list_for_each_entry(adap, &adapter_list, list_node) { - if ((type == CXGB4_ULD_CRYPTO && !is_pci_uld(adap)) || - (type != CXGB4_ULD_CRYPTO && !is_offload(adap))) - continue; - if (type == CXGB4_ULD_ISCSIT && is_t4(adap->params.chip)) - continue; - ret = cfg_queues_uld(adap, type, p); - if (ret) - goto out; - ret = setup_sge_queues_uld(adap, type, p->lro); - if (ret) - goto free_queues; - if (adap->flags & CXGB4_USING_MSIX) { - ret = request_msix_queue_irqs_uld(adap, type); - if (ret) - goto free_rxq; - } - if (adap->flags & CXGB4_FULL_INIT_DONE) - enable_rx_uld(adap, type); -#ifdef CONFIG_CHELSIO_TLS_DEVICE - /* send mbox to enable ktls related settings. */ - if (type == CXGB4_ULD_CRYPTO && - (adap->params.crypto & FW_CAPS_CONFIG_TX_TLS_HW)) - cxgb4_set_ktls_feature(adap, 1); -#endif - if (adap->uld[type].add) - goto free_irq; - ret = setup_sge_txq_uld(adap, type, p); - if (ret) - goto free_irq; - adap->uld[type] = *p; - ret = uld_attach(adap, type); - if (ret) - goto free_txq; - continue; -free_txq: - release_sge_txq_uld(adap, type); -free_irq: - if (adap->flags & CXGB4_FULL_INIT_DONE) - quiesce_rx_uld(adap, type); - if (adap->flags & CXGB4_USING_MSIX) - free_msix_queue_irqs_uld(adap, type); -free_rxq: - free_sge_queues_uld(adap, type); -free_queues: - free_queues_uld(adap, type); -out: - dev_warn(adap->pdev_dev, - "ULD registration failed for uld type %d\n", type); - } + list_for_each_entry(adap, &adapter_list, list_node) + cxgb4_uld_alloc_resources(adap, type, p); + + uld_entry->uld_type = type; + list_add_tail(&uld_entry->list_node, &uld_list); mutex_unlock(&uld_mutex); return; } @@ -761,6 +826,7 @@ EXPORT_SYMBOL(cxgb4_register_uld); */ int cxgb4_unregister_uld(enum cxgb4_uld type) { + struct cxgb4_uld_list *uld_entry, *tmp; struct adapter *adap; if (type >= CXGB4_ULD_MAX) @@ -775,13 +841,13 @@ int cxgb4_unregister_uld(enum cxgb4_uld type) continue; cxgb4_shutdown_uld_adapter(adap, type); + } -#ifdef CONFIG_CHELSIO_TLS_DEVICE - /* send mbox to disable ktls related settings. */ - if (type == CXGB4_ULD_CRYPTO && - (adap->params.crypto & FW_CAPS_CONFIG_TX_TLS_HW)) - cxgb4_set_ktls_feature(adap, 0); -#endif + list_for_each_entry_safe(uld_entry, tmp, &uld_list, list_node) { + if (uld_entry->uld_type == type) { + list_del(&uld_entry->list_node); + kfree(uld_entry); + } } mutex_unlock(&uld_mutex); diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h index be831317520a..dbce99b209d6 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h @@ -147,6 +147,9 @@ struct tid_info { /* TIDs in the HASH */ atomic_t hash_tids_in_use; atomic_t conns_in_use; + /* ETHOFLD TIDs used for rate limiting */ + atomic_t eotids_in_use; + /* lock for setting/clearing filter bitmap */ spinlock_t ftid_lock; @@ -221,12 +224,14 @@ static inline void cxgb4_alloc_eotid(struct tid_info *t, u32 eotid, void *data) { set_bit(eotid, t->eotid_bmap); t->eotid_tab[eotid].data = data; + atomic_inc(&t->eotids_in_use); } static inline void cxgb4_free_eotid(struct tid_info *t, u32 eotid) { clear_bit(eotid, t->eotid_bmap); t->eotid_tab[eotid].data = NULL; + atomic_dec(&t->eotids_in_use); } int cxgb4_alloc_atid(struct tid_info *t, void *data); @@ -263,6 +268,10 @@ struct filter_ctx { u32 tid; /* to store tid */ }; +struct chcr_ktls { + refcount_t ktls_refcount; +}; + struct ch_filter_specification; int cxgb4_get_free_ftid(struct net_device *dev, u8 family, bool hash_en, @@ -322,6 +331,7 @@ enum cxgb4_control { CXGB4_CONTROL_DB_DROP, }; +struct adapter; struct pci_dev; struct l2t_data; struct net_device; @@ -458,8 +468,12 @@ struct cxgb4_uld_info { struct napi_struct *napi); void (*lro_flush)(struct t4_lro_mgr *); int (*tx_handler)(struct sk_buff *skb, struct net_device *dev); +#if IS_ENABLED(CONFIG_TLS_DEVICE) + const struct tlsdev_ops *tlsdev_ops; +#endif }; +void cxgb4_uld_enable(struct adapter *adap); void cxgb4_register_uld(enum cxgb4_uld type, const struct cxgb4_uld_info *p); int cxgb4_unregister_uld(enum cxgb4_uld type); int cxgb4_ofld_send(struct net_device *dev, struct sk_buff *skb); diff --git a/drivers/net/ethernet/chelsio/cxgb4/sched.c b/drivers/net/ethernet/chelsio/cxgb4/sched.c index cebe1412d960..fde93c50cfec 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sched.c +++ b/drivers/net/ethernet/chelsio/cxgb4/sched.c @@ -57,7 +57,8 @@ static int t4_sched_class_fw_cmd(struct port_info *pi, p->u.params.ratemode, p->u.params.channel, e->idx, p->u.params.minrate, p->u.params.maxrate, - p->u.params.weight, p->u.params.pktsize); + p->u.params.weight, p->u.params.pktsize, + p->u.params.burstsize); break; default: err = -ENOTSUPP; diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c index 6516c45864b3..1359158652b7 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/sge.c +++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c @@ -2091,10 +2091,9 @@ static inline u8 ethofld_calc_tx_flits(struct adapter *adap, return flits + nsgl; } -static inline void *write_eo_wr(struct adapter *adap, - struct sge_eosw_txq *eosw_txq, - struct sk_buff *skb, struct fw_eth_tx_eo_wr *wr, - u32 hdr_len, u32 wrlen) +static void *write_eo_wr(struct adapter *adap, struct sge_eosw_txq *eosw_txq, + struct sk_buff *skb, struct fw_eth_tx_eo_wr *wr, + u32 hdr_len, u32 wrlen) { const struct skb_shared_info *ssi = skb_shinfo(skb); struct cpl_tx_pkt_core *cpl; @@ -2113,7 +2112,8 @@ static inline void *write_eo_wr(struct adapter *adap, immd_len += hdr_len; if (!eosw_txq->ncompl || - eosw_txq->last_compl >= adap->params.ofldq_wr_cred / 2) { + (eosw_txq->last_compl + wrlen16) >= + (adap->params.ofldq_wr_cred / 2)) { compl = true; eosw_txq->ncompl++; eosw_txq->last_compl = 0; @@ -2153,8 +2153,8 @@ static inline void *write_eo_wr(struct adapter *adap, return cpl; } -static void ethofld_hard_xmit(struct net_device *dev, - struct sge_eosw_txq *eosw_txq) +static int ethofld_hard_xmit(struct net_device *dev, + struct sge_eosw_txq *eosw_txq) { struct port_info *pi = netdev2pinfo(dev); struct adapter *adap = netdev2adap(dev); @@ -2167,8 +2167,8 @@ static void ethofld_hard_xmit(struct net_device *dev, bool skip_eotx_wr = false; struct tx_sw_desc *d; struct sk_buff *skb; + int left, ret = 0; u8 flits, ndesc; - int left; eohw_txq = &adap->sge.eohw_txq[eosw_txq->hwqid]; spin_lock(&eohw_txq->lock); @@ -2198,11 +2198,19 @@ static void ethofld_hard_xmit(struct net_device *dev, wrlen = flits * 8; wrlen16 = DIV_ROUND_UP(wrlen, 16); - /* If there are no CPL credits, then wait for credits - * to come back and retry again + left = txq_avail(&eohw_txq->q) - ndesc; + + /* If there are no descriptors left in hardware queues or no + * CPL credits left in software queues, then wait for them + * to come back and retry again. Note that we always request + * for credits update via interrupt for every half credits + * consumed. So, the interrupt will eventually restore the + * credits and invoke the Tx path again. */ - if (unlikely(wrlen16 > eosw_txq->cred)) + if (unlikely(left < 0 || wrlen16 > eosw_txq->cred)) { + ret = -ENOMEM; goto out_unlock; + } if (unlikely(skip_eotx_wr)) { start = (u64 *)wr; @@ -2231,7 +2239,8 @@ write_wr_headers: sgl = (u64 *)inline_tx_skb_header(skb, &eohw_txq->q, (void *)start, hdr_len); if (data_len) { - if (unlikely(cxgb4_map_skb(adap->pdev_dev, skb, d->addr))) { + ret = cxgb4_map_skb(adap->pdev_dev, skb, d->addr); + if (unlikely(ret)) { memset(d->addr, 0, sizeof(d->addr)); eohw_txq->mapping_err++; goto out_unlock; @@ -2277,12 +2286,13 @@ write_wr_headers: out_unlock: spin_unlock(&eohw_txq->lock); + return ret; } static void ethofld_xmit(struct net_device *dev, struct sge_eosw_txq *eosw_txq) { struct sk_buff *skb; - int pktcount; + int pktcount, ret; switch (eosw_txq->state) { case CXGB4_EO_STATE_ACTIVE: @@ -2307,7 +2317,9 @@ static void ethofld_xmit(struct net_device *dev, struct sge_eosw_txq *eosw_txq) continue; } - ethofld_hard_xmit(dev, eosw_txq); + ret = ethofld_hard_xmit(dev, eosw_txq); + if (ret) + break; } } diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c index 2a3480fc1d91..1c8068c02728 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c +++ b/drivers/net/ethernet/chelsio/cxgb4/t4_hw.c @@ -10361,9 +10361,10 @@ int t4_sge_ctxt_rd_bd(struct adapter *adap, unsigned int cid, return ret; } -int t4_sched_params(struct adapter *adapter, int type, int level, int mode, - int rateunit, int ratemode, int channel, int class, - int minrate, int maxrate, int weight, int pktsize) +int t4_sched_params(struct adapter *adapter, u8 type, u8 level, u8 mode, + u8 rateunit, u8 ratemode, u8 channel, u8 class, + u32 minrate, u32 maxrate, u16 weight, u16 pktsize, + u16 burstsize) { struct fw_sched_cmd cmd; @@ -10385,6 +10386,7 @@ int t4_sched_params(struct adapter *adapter, int type, int level, int mode, cmd.u.params.max = cpu_to_be32(maxrate); cmd.u.params.weight = cpu_to_be16(weight); cmd.u.params.pktsize = cpu_to_be16(pktsize); + cmd.u.params.burstsize = cpu_to_be16(burstsize); return t4_wr_mbox_meat(adapter, adapter->mbox, &cmd, sizeof(cmd), NULL, 1); diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h index 68fe734b9b37..0a326c054707 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h +++ b/drivers/net/ethernet/chelsio/cxgb4/t4fw_api.h @@ -1205,7 +1205,7 @@ enum fw_caps_config_crypto { FW_CAPS_CONFIG_CRYPTO_LOOKASIDE = 0x00000001, FW_CAPS_CONFIG_TLS_INLINE = 0x00000002, FW_CAPS_CONFIG_IPSEC_INLINE = 0x00000004, - FW_CAPS_CONFIG_TX_TLS_HW = 0x00000008, + FW_CAPS_CONFIG_TLS_HW = 0x00000008, }; enum fw_caps_config_fcoe { @@ -1329,7 +1329,7 @@ enum fw_params_param_dev { FW_PARAMS_PARAM_DEV_DBQ_TIMERTICK = 0x2A, FW_PARAMS_PARAM_DEV_NUM_TM_CLASS = 0x2B, FW_PARAMS_PARAM_DEV_FILTER = 0x2E, - FW_PARAMS_PARAM_DEV_KTLS_TX_HW = 0x31, + FW_PARAMS_PARAM_DEV_KTLS_HW = 0x31, }; /* @@ -1412,6 +1412,12 @@ enum fw_params_param_dmaq { FW_PARAMS_PARAM_DMAQ_CONM_CTXT = 0x20, }; +enum fw_params_param_dev_ktls_hw { + FW_PARAMS_PARAM_DEV_KTLS_HW_DISABLE = 0x00, + FW_PARAMS_PARAM_DEV_KTLS_HW_ENABLE = 0x01, + FW_PARAMS_PARAM_DEV_KTLS_HW_USER_ENABLE = 0x01, +}; + enum fw_params_param_dev_phyfw { FW_PARAMS_PARAM_DEV_PHYFW_DOWNLOAD = 0x00, FW_PARAMS_PARAM_DEV_PHYFW_VERSION = 0x01, diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c index 9cc3541a7e1c..cec865a97464 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c @@ -2480,7 +2480,7 @@ static int setup_debugfs(struct adapter *adapter) for (i = 0; i < ARRAY_SIZE(debugfs_files); i++) debugfs_create_file(debugfs_files[i].name, debugfs_files[i].mode, - adapter->debugfs_root, (void *)adapter, + adapter->debugfs_root, adapter, debugfs_files[i].fops); return 0; diff --git a/drivers/net/ethernet/cirrus/Kconfig b/drivers/net/ethernet/cirrus/Kconfig index 48f3198381bc..8d845f5ee0c5 100644 --- a/drivers/net/ethernet/cirrus/Kconfig +++ b/drivers/net/ethernet/cirrus/Kconfig @@ -24,7 +24,7 @@ config CS89x0 ---help--- Support for CS89x0 chipset based Ethernet cards. If you have a network (Ethernet) card of this type, say Y and read the file - <file:Documentation/networking/device_drivers/cirrus/cs89x0.txt>. + <file:Documentation/networking/device_drivers/cirrus/cs89x0.rst>. To compile this driver as a module, choose M here. The module will be called cs89x0. diff --git a/drivers/net/ethernet/cortina/gemini.c b/drivers/net/ethernet/cortina/gemini.c index 5bff5c2be88b..8d13ea370db1 100644 --- a/drivers/net/ethernet/cortina/gemini.c +++ b/drivers/net/ethernet/cortina/gemini.c @@ -1224,7 +1224,8 @@ map_error: return -ENOMEM; } -static int gmac_start_xmit(struct sk_buff *skb, struct net_device *netdev) +static netdev_tx_t gmac_start_xmit(struct sk_buff *skb, + struct net_device *netdev) { struct gemini_ethernet_port *port = netdev_priv(netdev); unsigned short m = (1 << port->txq_order) - 1; diff --git a/drivers/net/ethernet/dec/tulip/Kconfig b/drivers/net/ethernet/dec/tulip/Kconfig index 8ce6888ea722..177f36f4b89d 100644 --- a/drivers/net/ethernet/dec/tulip/Kconfig +++ b/drivers/net/ethernet/dec/tulip/Kconfig @@ -114,7 +114,7 @@ config DE4X5 These include the DE425, DE434, DE435, DE450 and DE500 models. If you have a network card of this type, say Y. More specific information is contained in - <file:Documentation/networking/device_drivers/dec/de4x5.txt>. + <file:Documentation/networking/device_drivers/dec/de4x5.rst>. To compile this driver as a module, choose M here. The module will be called de4x5. @@ -138,7 +138,7 @@ config DM9102 This driver is for DM9102(A)/DM9132/DM9801 compatible PCI cards from Davicom (<http://www.davicom.com.tw/>). If you have such a network (Ethernet) card, say Y. Some information is contained in the file - <file:Documentation/networking/device_drivers/dec/dmfe.txt>. + <file:Documentation/networking/device_drivers/dec/dmfe.rst>. To compile this driver as a module, choose M here. The module will be called dmfe. diff --git a/drivers/net/ethernet/dec/tulip/de4x5.c b/drivers/net/ethernet/dec/tulip/de4x5.c index f16853c3c851..0ccd9994ad45 100644 --- a/drivers/net/ethernet/dec/tulip/de4x5.c +++ b/drivers/net/ethernet/dec/tulip/de4x5.c @@ -951,7 +951,7 @@ static void reset_init_sia(struct net_device *dev, s32 sicr, s32 strr, s32 si static int test_ans(struct net_device *dev, s32 irqs, s32 irq_mask, s32 msec); static int test_tp(struct net_device *dev, s32 msec); static int EISA_signature(char *name, struct device *device); -static int PCI_signature(char *name, struct de4x5_private *lp); +static void PCI_signature(char *name, struct de4x5_private *lp); static void DevicePresent(struct net_device *dev, u_long iobase); static void enet_addr_rst(u_long aprom_addr); static int de4x5_bad_srom(struct de4x5_private *lp); @@ -3902,14 +3902,14 @@ EISA_signature(char *name, struct device *device) /* ** Look for a particular board name in the PCI configuration space */ -static int +static void PCI_signature(char *name, struct de4x5_private *lp) { - int i, status = 0, siglen = ARRAY_SIZE(de4x5_signatures); + int i, siglen = ARRAY_SIZE(de4x5_signatures); if (lp->chipset == DC21040) { strcpy(name, "DE434/5"); - return status; + return; } else { /* Search for a DEC name in the SROM */ int tmp = *((char *)&lp->srom + 19) * 3; strncpy(name, (char *)&lp->srom + 26 + tmp, 8); @@ -3935,8 +3935,6 @@ PCI_signature(char *name, struct de4x5_private *lp) } else if ((lp->chipset & ~0x00ff) == DC2114x) { lp->useSROM = true; } - - return status; } /* diff --git a/drivers/net/ethernet/dlink/dl2k.c b/drivers/net/ethernet/dlink/dl2k.c index 643090555cc7..5143722c4419 100644 --- a/drivers/net/ethernet/dlink/dl2k.c +++ b/drivers/net/ethernet/dlink/dl2k.c @@ -1869,7 +1869,7 @@ Compile command: gcc -D__KERNEL__ -DMODULE -I/usr/src/linux/include -Wall -Wstrict-prototypes -O2 -c dl2k.c -Read Documentation/networking/device_drivers/dlink/dl2k.txt for details. +Read Documentation/networking/device_drivers/dlink/dl2k.rst for details. */ diff --git a/drivers/net/ethernet/dnet.c b/drivers/net/ethernet/dnet.c index 057a508dd6e2..db98274501a0 100644 --- a/drivers/net/ethernet/dnet.c +++ b/drivers/net/ethernet/dnet.c @@ -776,8 +776,7 @@ static int dnet_probe(struct platform_device *pdev) spin_lock_init(&bp->lock); - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - bp->regs = devm_ioremap_resource(&pdev->dev, res); + bp->regs = devm_platform_get_and_ioremap_resource(pdev, 0, &res); if (IS_ERR(bp->regs)) { err = PTR_ERR(bp->regs); goto err_out_free_dev; diff --git a/drivers/net/ethernet/faraday/ftmac100.c b/drivers/net/ethernet/faraday/ftmac100.c index 32cf54f0e35b..473b337b2e3b 100644 --- a/drivers/net/ethernet/faraday/ftmac100.c +++ b/drivers/net/ethernet/faraday/ftmac100.c @@ -1057,9 +1057,6 @@ static int ftmac100_probe(struct platform_device *pdev) struct ftmac100 *priv; int err; - if (!pdev) - return -ENODEV; - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); if (!res) return -ENXIO; diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c index 2cd1f8efdfa3..2972244e6eb0 100644 --- a/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_eth.c @@ -2107,7 +2107,7 @@ workaround: /* Workaround for DPAA_A050385 requires data start to be aligned */ start = PTR_ALIGN(new_skb->data, DPAA_A050385_ALIGN); - if (start - new_skb->data != 0) + if (start - new_skb->data) skb_reserve(new_skb, start - new_skb->data); skb_put(new_skb, skb->len); @@ -2914,7 +2914,7 @@ static int dpaa_eth_probe(struct platform_device *pdev) } /* Do this here, so we can be verbose early */ - SET_NETDEV_DEV(net_dev, dev); + SET_NETDEV_DEV(net_dev, dev->parent); dev_set_drvdata(dev, net_dev); priv = netdev_priv(net_dev); diff --git a/drivers/net/ethernet/freescale/dpaa2/Kconfig b/drivers/net/ethernet/freescale/dpaa2/Kconfig index c6fb8e4021ac..feea797cde02 100644 --- a/drivers/net/ethernet/freescale/dpaa2/Kconfig +++ b/drivers/net/ethernet/freescale/dpaa2/Kconfig @@ -9,6 +9,16 @@ config FSL_DPAA2_ETH The driver manages network objects discovered on the Freescale MC bus. +if FSL_DPAA2_ETH +config FSL_DPAA2_ETH_DCB + bool "Data Center Bridging (DCB) Support" + default n + depends on DCB + help + Enable Priority-Based Flow Control (PFC) support for DPAA2 Ethernet + devices. +endif + config FSL_DPAA2_PTP_CLOCK tristate "Freescale DPAA2 PTP Clock" depends on FSL_DPAA2_ETH && PTP_1588_CLOCK_QORIQ diff --git a/drivers/net/ethernet/freescale/dpaa2/Makefile b/drivers/net/ethernet/freescale/dpaa2/Makefile index 69184ca3b7b9..6e7f33c956bf 100644 --- a/drivers/net/ethernet/freescale/dpaa2/Makefile +++ b/drivers/net/ethernet/freescale/dpaa2/Makefile @@ -7,6 +7,7 @@ obj-$(CONFIG_FSL_DPAA2_ETH) += fsl-dpaa2-eth.o obj-$(CONFIG_FSL_DPAA2_PTP_CLOCK) += fsl-dpaa2-ptp.o fsl-dpaa2-eth-objs := dpaa2-eth.o dpaa2-ethtool.o dpni.o dpaa2-mac.o dpmac.o +fsl-dpaa2-eth-${CONFIG_FSL_DPAA2_ETH_DCB} += dpaa2-eth-dcb.o fsl-dpaa2-eth-${CONFIG_DEBUG_FS} += dpaa2-eth-debugfs.o fsl-dpaa2-ptp-objs := dpaa2-ptp.o dprtc.o diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-dcb.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-dcb.c new file mode 100644 index 000000000000..83dee575c2fa --- /dev/null +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-dcb.c @@ -0,0 +1,150 @@ +// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) +/* Copyright 2020 NXP */ + +#include "dpaa2-eth.h" + +static int dpaa2_eth_dcbnl_ieee_getpfc(struct net_device *net_dev, + struct ieee_pfc *pfc) +{ + struct dpaa2_eth_priv *priv = netdev_priv(net_dev); + + if (!(priv->link_state.options & DPNI_LINK_OPT_PFC_PAUSE)) + return 0; + + memcpy(pfc, &priv->pfc, sizeof(priv->pfc)); + pfc->pfc_cap = dpaa2_eth_tc_count(priv); + + return 0; +} + +static inline bool is_prio_enabled(u8 pfc_en, u8 tc) +{ + return !!(pfc_en & (1 << tc)); +} + +static int set_pfc_cn(struct dpaa2_eth_priv *priv, u8 pfc_en) +{ + struct dpni_congestion_notification_cfg cfg = {0}; + int i, err; + + cfg.notification_mode = DPNI_CONG_OPT_FLOW_CONTROL; + cfg.units = DPNI_CONGESTION_UNIT_FRAMES; + cfg.message_iova = 0ULL; + cfg.message_ctx = 0ULL; + + for (i = 0; i < dpaa2_eth_tc_count(priv); i++) { + if (is_prio_enabled(pfc_en, i)) { + cfg.threshold_entry = DPAA2_ETH_CN_THRESH_ENTRY(priv); + cfg.threshold_exit = DPAA2_ETH_CN_THRESH_EXIT(priv); + } else { + /* For priorities not set in the pfc_en mask, we leave + * the congestion thresholds at zero, which effectively + * disables generation of PFC frames for them + */ + cfg.threshold_entry = 0; + cfg.threshold_exit = 0; + } + + err = dpni_set_congestion_notification(priv->mc_io, 0, + priv->mc_token, + DPNI_QUEUE_RX, i, &cfg); + if (err) { + netdev_err(priv->net_dev, + "dpni_set_congestion_notification failed\n"); + return err; + } + } + + return 0; +} + +static int dpaa2_eth_dcbnl_ieee_setpfc(struct net_device *net_dev, + struct ieee_pfc *pfc) +{ + struct dpaa2_eth_priv *priv = netdev_priv(net_dev); + struct dpni_link_cfg link_cfg = {0}; + bool tx_pause; + int err; + + if (pfc->mbc || pfc->delay) + return -EOPNOTSUPP; + + /* If same PFC enabled mask, nothing to do */ + if (priv->pfc.pfc_en == pfc->pfc_en) + return 0; + + /* We allow PFC configuration even if it won't have any effect until + * general pause frames are enabled + */ + tx_pause = dpaa2_eth_tx_pause_enabled(priv->link_state.options); + if (!dpaa2_eth_rx_pause_enabled(priv->link_state.options) || !tx_pause) + netdev_warn(net_dev, "Pause support must be enabled in order for PFC to work!\n"); + + link_cfg.rate = priv->link_state.rate; + link_cfg.options = priv->link_state.options; + if (pfc->pfc_en) + link_cfg.options |= DPNI_LINK_OPT_PFC_PAUSE; + else + link_cfg.options &= ~DPNI_LINK_OPT_PFC_PAUSE; + err = dpni_set_link_cfg(priv->mc_io, 0, priv->mc_token, &link_cfg); + if (err) { + netdev_err(net_dev, "dpni_set_link_cfg failed\n"); + return err; + } + + /* Configure congestion notifications for the enabled priorities */ + err = set_pfc_cn(priv, pfc->pfc_en); + if (err) + return err; + + memcpy(&priv->pfc, pfc, sizeof(priv->pfc)); + priv->pfc_enabled = !!pfc->pfc_en; + + dpaa2_eth_set_rx_taildrop(priv, tx_pause, priv->pfc_enabled); + + return 0; +} + +static u8 dpaa2_eth_dcbnl_getdcbx(struct net_device *net_dev) +{ + struct dpaa2_eth_priv *priv = netdev_priv(net_dev); + + return priv->dcbx_mode; +} + +static u8 dpaa2_eth_dcbnl_setdcbx(struct net_device *net_dev, u8 mode) +{ + struct dpaa2_eth_priv *priv = netdev_priv(net_dev); + + return (mode != (priv->dcbx_mode)) ? 1 : 0; +} + +static u8 dpaa2_eth_dcbnl_getcap(struct net_device *net_dev, int capid, u8 *cap) +{ + struct dpaa2_eth_priv *priv = netdev_priv(net_dev); + + switch (capid) { + case DCB_CAP_ATTR_PFC: + *cap = true; + break; + case DCB_CAP_ATTR_PFC_TCS: + *cap = 1 << (dpaa2_eth_tc_count(priv) - 1); + break; + case DCB_CAP_ATTR_DCBX: + *cap = priv->dcbx_mode; + break; + default: + *cap = false; + break; + } + + return 0; +} + +const struct dcbnl_rtnl_ops dpaa2_eth_dcbnl_ops = { + .ieee_getpfc = dpaa2_eth_dcbnl_ieee_getpfc, + .ieee_setpfc = dpaa2_eth_dcbnl_ieee_setpfc, + .getdcbx = dpaa2_eth_dcbnl_getdcbx, + .setdcbx = dpaa2_eth_dcbnl_setdcbx, + .getcap = dpaa2_eth_dcbnl_getcap, +}; diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.c index a9afe46b837f..c453a23045c1 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth-debugfs.c @@ -81,8 +81,8 @@ static int dpaa2_dbg_fqs_show(struct seq_file *file, void *offset) int i, err; seq_printf(file, "FQ stats for %s:\n", priv->net_dev->name); - seq_printf(file, "%s%16s%16s%16s%16s\n", - "VFQID", "CPU", "Type", "Frames", "Pending frames"); + seq_printf(file, "%s%16s%16s%16s%16s%16s\n", + "VFQID", "CPU", "TC", "Type", "Frames", "Pending frames"); for (i = 0; i < priv->num_fqs; i++) { fq = &priv->fq[i]; @@ -90,9 +90,10 @@ static int dpaa2_dbg_fqs_show(struct seq_file *file, void *offset) if (err) fcnt = 0; - seq_printf(file, "%5d%16d%16s%16llu%16u\n", + seq_printf(file, "%5d%16d%16d%16s%16llu%16u\n", fq->fqid, fq->target_cpu, + fq->tc, fq_type_to_str(fq), fq->stats.frames, fcnt); @@ -127,16 +128,19 @@ static int dpaa2_dbg_ch_show(struct seq_file *file, void *offset) int i; seq_printf(file, "Channel stats for %s:\n", priv->net_dev->name); - seq_printf(file, "%s%16s%16s%16s%16s\n", - "CHID", "CPU", "Deq busy", "CDANs", "Buf count"); + seq_printf(file, "%s%16s%16s%16s%16s%16s%16s\n", + "CHID", "CPU", "Deq busy", "Frames", "CDANs", + "Avg Frm/CDAN", "Buf count"); for (i = 0; i < priv->num_channels; i++) { ch = priv->channel[i]; - seq_printf(file, "%4d%16d%16llu%16llu%16d\n", + seq_printf(file, "%4d%16d%16llu%16llu%16llu%16llu%16d\n", ch->ch_id, ch->nctx.desired_cpu, ch->stats.dequeue_portal_busy, + ch->stats.frames, ch->stats.cdan, + div64_u64(ch->stats.frames, ch->stats.cdan), ch->buf_count); } diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c index d97c320a2dc0..8fb48de5d18c 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) /* Copyright 2014-2016 Freescale Semiconductor Inc. - * Copyright 2016-2019 NXP + * Copyright 2016-2020 NXP */ #include <linux/init.h> #include <linux/module.h> @@ -244,13 +244,72 @@ static void xdp_release_buf(struct dpaa2_eth_priv *priv, ch->xdp.drop_cnt = 0; } -static int xdp_enqueue(struct dpaa2_eth_priv *priv, struct dpaa2_fd *fd, - void *buf_start, u16 queue_id) +static int dpaa2_eth_xdp_flush(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_fq *fq, + struct dpaa2_eth_xdp_fds *xdp_fds) +{ + int total_enqueued = 0, retries = 0, enqueued; + struct dpaa2_eth_drv_stats *percpu_extras; + int num_fds, err, max_retries; + struct dpaa2_fd *fds; + + percpu_extras = this_cpu_ptr(priv->percpu_extras); + + /* try to enqueue all the FDs until the max number of retries is hit */ + fds = xdp_fds->fds; + num_fds = xdp_fds->num; + max_retries = num_fds * DPAA2_ETH_ENQUEUE_RETRIES; + while (total_enqueued < num_fds && retries < max_retries) { + err = priv->enqueue(priv, fq, &fds[total_enqueued], + 0, num_fds - total_enqueued, &enqueued); + if (err == -EBUSY) { + percpu_extras->tx_portal_busy += ++retries; + continue; + } + total_enqueued += enqueued; + } + xdp_fds->num = 0; + + return total_enqueued; +} + +static void xdp_tx_flush(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_channel *ch, + struct dpaa2_eth_fq *fq) +{ + struct rtnl_link_stats64 *percpu_stats; + struct dpaa2_fd *fds; + int enqueued, i; + + percpu_stats = this_cpu_ptr(priv->percpu_stats); + + // enqueue the array of XDP_TX frames + enqueued = dpaa2_eth_xdp_flush(priv, fq, &fq->xdp_tx_fds); + + /* update statistics */ + percpu_stats->tx_packets += enqueued; + fds = fq->xdp_tx_fds.fds; + for (i = 0; i < enqueued; i++) { + percpu_stats->tx_bytes += dpaa2_fd_get_len(&fds[i]); + ch->stats.xdp_tx++; + } + for (i = enqueued; i < fq->xdp_tx_fds.num; i++) { + xdp_release_buf(priv, ch, dpaa2_fd_get_addr(&fds[i])); + percpu_stats->tx_errors++; + ch->stats.xdp_tx_err++; + } + fq->xdp_tx_fds.num = 0; +} + +static void xdp_enqueue(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_channel *ch, + struct dpaa2_fd *fd, + void *buf_start, u16 queue_id) { - struct dpaa2_eth_fq *fq; struct dpaa2_faead *faead; + struct dpaa2_fd *dest_fd; + struct dpaa2_eth_fq *fq; u32 ctrl, frc; - int i, err; /* Mark the egress frame hardware annotation area as valid */ frc = dpaa2_fd_get_frc(fd); @@ -267,13 +326,13 @@ static int xdp_enqueue(struct dpaa2_eth_priv *priv, struct dpaa2_fd *fd, faead->conf_fqid = 0; fq = &priv->fq[queue_id]; - for (i = 0; i < DPAA2_ETH_ENQUEUE_RETRIES; i++) { - err = priv->enqueue(priv, fq, fd, 0); - if (err != -EBUSY) - break; - } + dest_fd = &fq->xdp_tx_fds.fds[fq->xdp_tx_fds.num++]; + memcpy(dest_fd, fd, sizeof(*dest_fd)); - return err; + if (fq->xdp_tx_fds.num < DEV_MAP_BULK_SIZE) + return; + + xdp_tx_flush(priv, ch, fq); } static u32 run_xdp(struct dpaa2_eth_priv *priv, @@ -282,14 +341,11 @@ static u32 run_xdp(struct dpaa2_eth_priv *priv, struct dpaa2_fd *fd, void *vaddr) { dma_addr_t addr = dpaa2_fd_get_addr(fd); - struct rtnl_link_stats64 *percpu_stats; struct bpf_prog *xdp_prog; struct xdp_buff xdp; u32 xdp_act = XDP_PASS; int err; - percpu_stats = this_cpu_ptr(priv->percpu_stats); - rcu_read_lock(); xdp_prog = READ_ONCE(ch->xdp.prog); @@ -302,6 +358,9 @@ static u32 run_xdp(struct dpaa2_eth_priv *priv, xdp_set_data_meta_invalid(&xdp); xdp.rxq = &ch->xdp_rxq; + xdp.frame_sz = DPAA2_ETH_RX_BUF_RAW_SIZE - + (dpaa2_fd_get_offset(fd) - XDP_PACKET_HEADROOM); + xdp_act = bpf_prog_run_xdp(xdp_prog, &xdp); /* xdp.data pointer may have changed */ @@ -312,16 +371,7 @@ static u32 run_xdp(struct dpaa2_eth_priv *priv, case XDP_PASS: break; case XDP_TX: - err = xdp_enqueue(priv, fd, vaddr, rx_fq->flowid); - if (err) { - xdp_release_buf(priv, ch, addr); - percpu_stats->tx_errors++; - ch->stats.xdp_tx_err++; - } else { - percpu_stats->tx_packets++; - percpu_stats->tx_bytes += dpaa2_fd_get_len(fd); - ch->stats.xdp_tx++; - } + xdp_enqueue(priv, ch, fd, vaddr, rx_fq->flowid); break; default: bpf_warn_invalid_xdp_action(xdp_act); @@ -337,7 +387,11 @@ static u32 run_xdp(struct dpaa2_eth_priv *priv, dma_unmap_page(priv->net_dev->dev.parent, addr, priv->rx_buf_size, DMA_BIDIRECTIONAL); ch->buf_count--; + + /* Allow redirect use of full headroom */ xdp.data_hard_start = vaddr; + xdp.frame_sz = DPAA2_ETH_RX_BUF_RAW_SIZE; + err = xdp_do_redirect(priv->net_dev, &xdp, xdp_prog); if (unlikely(err)) ch->stats.xdp_drop++; @@ -493,6 +547,7 @@ static int consume_frames(struct dpaa2_eth_channel *ch, return 0; fq->stats.frames += cleaned; + ch->stats.frames += cleaned; /* A dequeue operation only pulls frames from a single queue * into the store. Return the frame queue as an out param. @@ -847,7 +902,7 @@ static netdev_tx_t dpaa2_eth_tx(struct sk_buff *skb, struct net_device *net_dev) * the Tx confirmation callback for this frame */ for (i = 0; i < DPAA2_ETH_ENQUEUE_RETRIES; i++) { - err = priv->enqueue(priv, fq, &fd, prio); + err = priv->enqueue(priv, fq, &fd, prio, 1, NULL); if (err != -EBUSY) break; } @@ -1138,6 +1193,7 @@ static int dpaa2_eth_poll(struct napi_struct *napi, int budget) int store_cleaned, work_done; struct list_head rx_list; int retries = 0; + u16 flowid; int err; ch = container_of(napi, struct dpaa2_eth_channel, napi); @@ -1160,6 +1216,7 @@ static int dpaa2_eth_poll(struct napi_struct *napi, int budget) break; if (fq->type == DPAA2_RX_FQ) { rx_cleaned += store_cleaned; + flowid = fq->flowid; } else { txconf_cleaned += store_cleaned; /* We have a single Tx conf FQ on this channel */ @@ -1202,6 +1259,8 @@ out: if (ch->xdp.res & XDP_REDIRECT) xdp_do_flush_map(); + else if (rx_cleaned && ch->xdp.res & XDP_TX) + xdp_tx_flush(priv, ch, &priv->fq[flowid]); return work_done; } @@ -1228,31 +1287,67 @@ static void disable_ch_napi(struct dpaa2_eth_priv *priv) } } -static void dpaa2_eth_set_rx_taildrop(struct dpaa2_eth_priv *priv, bool enable) +void dpaa2_eth_set_rx_taildrop(struct dpaa2_eth_priv *priv, + bool tx_pause, bool pfc) { struct dpni_taildrop td = {0}; + struct dpaa2_eth_fq *fq; int i, err; - if (priv->rx_td_enabled == enable) - return; + /* FQ taildrop: threshold is in bytes, per frame queue. Enabled if + * flow control is disabled (as it might interfere with either the + * buffer pool depletion trigger for pause frames or with the group + * congestion trigger for PFC frames) + */ + td.enable = !tx_pause; + if (priv->rx_fqtd_enabled == td.enable) + goto set_cgtd; - td.enable = enable; - td.threshold = DPAA2_ETH_TAILDROP_THRESH; + td.threshold = DPAA2_ETH_FQ_TAILDROP_THRESH; + td.units = DPNI_CONGESTION_UNIT_BYTES; for (i = 0; i < priv->num_fqs; i++) { - if (priv->fq[i].type != DPAA2_RX_FQ) + fq = &priv->fq[i]; + if (fq->type != DPAA2_RX_FQ) continue; err = dpni_set_taildrop(priv->mc_io, 0, priv->mc_token, - DPNI_CP_QUEUE, DPNI_QUEUE_RX, 0, - priv->fq[i].flowid, &td); + DPNI_CP_QUEUE, DPNI_QUEUE_RX, + fq->tc, fq->flowid, &td); if (err) { netdev_err(priv->net_dev, - "dpni_set_taildrop() failed\n"); - break; + "dpni_set_taildrop(FQ) failed\n"); + return; + } + } + + priv->rx_fqtd_enabled = td.enable; + +set_cgtd: + /* Congestion group taildrop: threshold is in frames, per group + * of FQs belonging to the same traffic class + * Enabled if general Tx pause disabled or if PFCs are enabled + * (congestion group threhsold for PFC generation is lower than the + * CG taildrop threshold, so it won't interfere with it; we also + * want frames in non-PFC enabled traffic classes to be kept in check) + */ + td.enable = !tx_pause || (tx_pause && pfc); + if (priv->rx_cgtd_enabled == td.enable) + return; + + td.threshold = DPAA2_ETH_CG_TAILDROP_THRESH(priv); + td.units = DPNI_CONGESTION_UNIT_FRAMES; + for (i = 0; i < dpaa2_eth_tc_count(priv); i++) { + err = dpni_set_taildrop(priv->mc_io, 0, priv->mc_token, + DPNI_CP_GROUP, DPNI_QUEUE_RX, + i, 0, &td); + if (err) { + netdev_err(priv->net_dev, + "dpni_set_taildrop(CG) failed\n"); + return; } } - priv->rx_td_enabled = enable; + priv->rx_cgtd_enabled = td.enable; } static int link_state_update(struct dpaa2_eth_priv *priv) @@ -1272,9 +1367,8 @@ static int link_state_update(struct dpaa2_eth_priv *priv) * Rx FQ taildrop configuration as well. We configure taildrop * only when pause frame generation is disabled. */ - tx_pause = !!(state.options & DPNI_LINK_OPT_PAUSE) ^ - !!(state.options & DPNI_LINK_OPT_ASYM_PAUSE); - dpaa2_eth_set_rx_taildrop(priv, !tx_pause); + tx_pause = dpaa2_eth_tx_pause_enabled(state.options); + dpaa2_eth_set_rx_taildrop(priv, tx_pause, priv->pfc_enabled); /* When we manage the MAC/PHY using phylink there is no need * to manually update the netif_carrier. @@ -1880,20 +1974,16 @@ static int dpaa2_eth_xdp(struct net_device *dev, struct netdev_bpf *xdp) return 0; } -static int dpaa2_eth_xdp_xmit_frame(struct net_device *net_dev, - struct xdp_frame *xdpf) +static int dpaa2_eth_xdp_create_fd(struct net_device *net_dev, + struct xdp_frame *xdpf, + struct dpaa2_fd *fd) { struct dpaa2_eth_priv *priv = netdev_priv(net_dev); struct device *dev = net_dev->dev.parent; - struct rtnl_link_stats64 *percpu_stats; - struct dpaa2_eth_drv_stats *percpu_extras; unsigned int needed_headroom; struct dpaa2_eth_swa *swa; - struct dpaa2_eth_fq *fq; - struct dpaa2_fd fd; void *buffer_start, *aligned_start; dma_addr_t addr; - int err, i; /* We require a minimum headroom to be able to transmit the frame. * Otherwise return an error and let the original net_device handle it @@ -1902,11 +1992,8 @@ static int dpaa2_eth_xdp_xmit_frame(struct net_device *net_dev, if (xdpf->headroom < needed_headroom) return -EINVAL; - percpu_stats = this_cpu_ptr(priv->percpu_stats); - percpu_extras = this_cpu_ptr(priv->percpu_extras); - /* Setup the FD fields */ - memset(&fd, 0, sizeof(fd)); + memset(fd, 0, sizeof(*fd)); /* Align FD address, if possible */ buffer_start = xdpf->data - needed_headroom; @@ -1924,32 +2011,14 @@ static int dpaa2_eth_xdp_xmit_frame(struct net_device *net_dev, addr = dma_map_single(dev, buffer_start, swa->xdp.dma_size, DMA_BIDIRECTIONAL); - if (unlikely(dma_mapping_error(dev, addr))) { - percpu_stats->tx_dropped++; + if (unlikely(dma_mapping_error(dev, addr))) return -ENOMEM; - } - - dpaa2_fd_set_addr(&fd, addr); - dpaa2_fd_set_offset(&fd, xdpf->data - buffer_start); - dpaa2_fd_set_len(&fd, xdpf->len); - dpaa2_fd_set_format(&fd, dpaa2_fd_single); - dpaa2_fd_set_ctrl(&fd, FD_CTRL_PTA); - - fq = &priv->fq[smp_processor_id() % dpaa2_eth_queue_count(priv)]; - for (i = 0; i < DPAA2_ETH_ENQUEUE_RETRIES; i++) { - err = priv->enqueue(priv, fq, &fd, 0); - if (err != -EBUSY) - break; - } - percpu_extras->tx_portal_busy += i; - if (unlikely(err < 0)) { - percpu_stats->tx_errors++; - /* let the Rx device handle the cleanup */ - return err; - } - percpu_stats->tx_packets++; - percpu_stats->tx_bytes += dpaa2_fd_get_len(&fd); + dpaa2_fd_set_addr(fd, addr); + dpaa2_fd_set_offset(fd, xdpf->data - buffer_start); + dpaa2_fd_set_len(fd, xdpf->len); + dpaa2_fd_set_format(fd, dpaa2_fd_single); + dpaa2_fd_set_ctrl(fd, FD_CTRL_PTA); return 0; } @@ -1957,8 +2026,12 @@ static int dpaa2_eth_xdp_xmit_frame(struct net_device *net_dev, static int dpaa2_eth_xdp_xmit(struct net_device *net_dev, int n, struct xdp_frame **frames, u32 flags) { - int drops = 0; - int i, err; + struct dpaa2_eth_priv *priv = netdev_priv(net_dev); + struct dpaa2_eth_xdp_fds *xdp_redirect_fds; + struct rtnl_link_stats64 *percpu_stats; + struct dpaa2_eth_fq *fq; + struct dpaa2_fd *fds; + int enqueued, i, err; if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK)) return -EINVAL; @@ -1966,17 +2039,31 @@ static int dpaa2_eth_xdp_xmit(struct net_device *net_dev, int n, if (!netif_running(net_dev)) return -ENETDOWN; - for (i = 0; i < n; i++) { - struct xdp_frame *xdpf = frames[i]; + fq = &priv->fq[smp_processor_id()]; + xdp_redirect_fds = &fq->xdp_redirect_fds; + fds = xdp_redirect_fds->fds; - err = dpaa2_eth_xdp_xmit_frame(net_dev, xdpf); - if (err) { - xdp_return_frame_rx_napi(xdpf); - drops++; - } + percpu_stats = this_cpu_ptr(priv->percpu_stats); + + /* create a FD for each xdp_frame in the list received */ + for (i = 0; i < n; i++) { + err = dpaa2_eth_xdp_create_fd(net_dev, frames[i], &fds[i]); + if (err) + break; } + xdp_redirect_fds->num = i; + + /* enqueue all the frame descriptors */ + enqueued = dpaa2_eth_xdp_flush(priv, fq, xdp_redirect_fds); + + /* update statistics */ + percpu_stats->tx_packets += enqueued; + for (i = 0; i < enqueued; i++) + percpu_stats->tx_bytes += dpaa2_fd_get_len(&fds[i]); + for (i = enqueued; i < n; i++) + xdp_return_frame_rx_napi(frames[i]); - return n - drops; + return enqueued; } static int update_xps(struct dpaa2_eth_priv *priv) @@ -2018,7 +2105,7 @@ static int dpaa2_eth_setup_tc(struct net_device *net_dev, int i; if (type != TC_SETUP_QDISC_MQPRIO) - return -EINVAL; + return -EOPNOTSUPP; mqprio->hw = TC_MQPRIO_HW_OFFLOAD_TCS; num_queues = dpaa2_eth_queue_count(priv); @@ -2030,7 +2117,7 @@ static int dpaa2_eth_setup_tc(struct net_device *net_dev, if (num_tc > dpaa2_eth_tc_count(priv)) { netdev_err(net_dev, "Max %d traffic classes supported\n", dpaa2_eth_tc_count(priv)); - return -EINVAL; + return -EOPNOTSUPP; } if (!num_tc) { @@ -2355,7 +2442,7 @@ static void set_fq_affinity(struct dpaa2_eth_priv *priv) static void setup_fqs(struct dpaa2_eth_priv *priv) { - int i; + int i, j; /* We have one TxConf FQ per Tx flow. * The number of Tx and Rx queues is the same. @@ -2367,10 +2454,13 @@ static void setup_fqs(struct dpaa2_eth_priv *priv) priv->fq[priv->num_fqs++].flowid = (u16)i; } - for (i = 0; i < dpaa2_eth_queue_count(priv); i++) { - priv->fq[priv->num_fqs].type = DPAA2_RX_FQ; - priv->fq[priv->num_fqs].consume = dpaa2_eth_rx; - priv->fq[priv->num_fqs++].flowid = (u16)i; + for (j = 0; j < dpaa2_eth_tc_count(priv); j++) { + for (i = 0; i < dpaa2_eth_queue_count(priv); i++) { + priv->fq[priv->num_fqs].type = DPAA2_RX_FQ; + priv->fq[priv->num_fqs].consume = dpaa2_eth_rx; + priv->fq[priv->num_fqs].tc = (u8)j; + priv->fq[priv->num_fqs++].flowid = (u16)i; + } } /* For each FQ, decide on which core to process incoming frames */ @@ -2528,19 +2618,38 @@ static int set_buffer_layout(struct dpaa2_eth_priv *priv) static inline int dpaa2_eth_enqueue_qd(struct dpaa2_eth_priv *priv, struct dpaa2_eth_fq *fq, - struct dpaa2_fd *fd, u8 prio) + struct dpaa2_fd *fd, u8 prio, + u32 num_frames __always_unused, + int *frames_enqueued) { - return dpaa2_io_service_enqueue_qd(fq->channel->dpio, - priv->tx_qdid, prio, - fq->tx_qdbin, fd); + int err; + + err = dpaa2_io_service_enqueue_qd(fq->channel->dpio, + priv->tx_qdid, prio, + fq->tx_qdbin, fd); + if (!err && frames_enqueued) + *frames_enqueued = 1; + return err; } -static inline int dpaa2_eth_enqueue_fq(struct dpaa2_eth_priv *priv, - struct dpaa2_eth_fq *fq, - struct dpaa2_fd *fd, u8 prio) +static inline int dpaa2_eth_enqueue_fq_multiple(struct dpaa2_eth_priv *priv, + struct dpaa2_eth_fq *fq, + struct dpaa2_fd *fd, + u8 prio, u32 num_frames, + int *frames_enqueued) { - return dpaa2_io_service_enqueue_fq(fq->channel->dpio, - fq->tx_fqid[prio], fd); + int err; + + err = dpaa2_io_service_enqueue_multiple_fq(fq->channel->dpio, + fq->tx_fqid[prio], + fd, num_frames); + + if (err == 0) + return -EBUSY; + + if (frames_enqueued) + *frames_enqueued = err; + return 0; } static void set_enqueue_mode(struct dpaa2_eth_priv *priv) @@ -2549,7 +2658,7 @@ static void set_enqueue_mode(struct dpaa2_eth_priv *priv) DPNI_ENQUEUE_FQID_VER_MINOR) < 0) priv->enqueue = dpaa2_eth_enqueue_qd; else - priv->enqueue = dpaa2_eth_enqueue_fq; + priv->enqueue = dpaa2_eth_enqueue_fq_multiple; } static int set_pause(struct dpaa2_eth_priv *priv) @@ -2610,7 +2719,7 @@ static void update_tx_fqids(struct dpaa2_eth_priv *priv) } } - priv->enqueue = dpaa2_eth_enqueue_fq; + priv->enqueue = dpaa2_eth_enqueue_fq_multiple; return; @@ -2620,6 +2729,118 @@ out_err: priv->enqueue = dpaa2_eth_enqueue_qd; } +/* Configure ingress classification based on VLAN PCP */ +static int set_vlan_qos(struct dpaa2_eth_priv *priv) +{ + struct device *dev = priv->net_dev->dev.parent; + struct dpkg_profile_cfg kg_cfg = {0}; + struct dpni_qos_tbl_cfg qos_cfg = {0}; + struct dpni_rule_cfg key_params; + void *dma_mem, *key, *mask; + u8 key_size = 2; /* VLAN TCI field */ + int i, pcp, err; + + /* VLAN-based classification only makes sense if we have multiple + * traffic classes. + * Also, we need to extract just the 3-bit PCP field from the VLAN + * header and we can only do that by using a mask + */ + if (dpaa2_eth_tc_count(priv) == 1 || !dpaa2_eth_fs_mask_enabled(priv)) { + dev_dbg(dev, "VLAN-based QoS classification not supported\n"); + return -EOPNOTSUPP; + } + + dma_mem = kzalloc(DPAA2_CLASSIFIER_DMA_SIZE, GFP_KERNEL); + if (!dma_mem) + return -ENOMEM; + + kg_cfg.num_extracts = 1; + kg_cfg.extracts[0].type = DPKG_EXTRACT_FROM_HDR; + kg_cfg.extracts[0].extract.from_hdr.prot = NET_PROT_VLAN; + kg_cfg.extracts[0].extract.from_hdr.type = DPKG_FULL_FIELD; + kg_cfg.extracts[0].extract.from_hdr.field = NH_FLD_VLAN_TCI; + + err = dpni_prepare_key_cfg(&kg_cfg, dma_mem); + if (err) { + dev_err(dev, "dpni_prepare_key_cfg failed\n"); + goto out_free_tbl; + } + + /* set QoS table */ + qos_cfg.default_tc = 0; + qos_cfg.discard_on_miss = 0; + qos_cfg.key_cfg_iova = dma_map_single(dev, dma_mem, + DPAA2_CLASSIFIER_DMA_SIZE, + DMA_TO_DEVICE); + if (dma_mapping_error(dev, qos_cfg.key_cfg_iova)) { + dev_err(dev, "QoS table DMA mapping failed\n"); + err = -ENOMEM; + goto out_free_tbl; + } + + err = dpni_set_qos_table(priv->mc_io, 0, priv->mc_token, &qos_cfg); + if (err) { + dev_err(dev, "dpni_set_qos_table failed\n"); + goto out_unmap_tbl; + } + + /* Add QoS table entries */ + key = kzalloc(key_size * 2, GFP_KERNEL); + if (!key) { + err = -ENOMEM; + goto out_unmap_tbl; + } + mask = key + key_size; + *(__be16 *)mask = cpu_to_be16(VLAN_PRIO_MASK); + + key_params.key_iova = dma_map_single(dev, key, key_size * 2, + DMA_TO_DEVICE); + if (dma_mapping_error(dev, key_params.key_iova)) { + dev_err(dev, "Qos table entry DMA mapping failed\n"); + err = -ENOMEM; + goto out_free_key; + } + + key_params.mask_iova = key_params.key_iova + key_size; + key_params.key_size = key_size; + + /* We add rules for PCP-based distribution starting with highest + * priority (VLAN PCP = 7). If this DPNI doesn't have enough traffic + * classes to accommodate all priority levels, the lowest ones end up + * on TC 0 which was configured as default + */ + for (i = dpaa2_eth_tc_count(priv) - 1, pcp = 7; i >= 0; i--, pcp--) { + *(__be16 *)key = cpu_to_be16(pcp << VLAN_PRIO_SHIFT); + dma_sync_single_for_device(dev, key_params.key_iova, + key_size * 2, DMA_TO_DEVICE); + + err = dpni_add_qos_entry(priv->mc_io, 0, priv->mc_token, + &key_params, i, i); + if (err) { + dev_err(dev, "dpni_add_qos_entry failed\n"); + dpni_clear_qos_table(priv->mc_io, 0, priv->mc_token); + goto out_unmap_key; + } + } + + priv->vlan_cls_enabled = true; + + /* Table and key memory is not persistent, clean everything up after + * configuration is finished + */ +out_unmap_key: + dma_unmap_single(dev, key_params.key_iova, key_size * 2, DMA_TO_DEVICE); +out_free_key: + kfree(key); +out_unmap_tbl: + dma_unmap_single(dev, qos_cfg.key_cfg_iova, DPAA2_CLASSIFIER_DMA_SIZE, + DMA_TO_DEVICE); +out_free_tbl: + kfree(dma_mem); + + return err; +} + /* Configure the DPNI object this interface is associated with */ static int setup_dpni(struct fsl_mc_device *ls_dev) { @@ -2682,10 +2903,16 @@ static int setup_dpni(struct fsl_mc_device *ls_dev) goto close; } + err = set_vlan_qos(priv); + if (err && err != -EOPNOTSUPP) + goto close; + priv->cls_rules = devm_kzalloc(dev, sizeof(struct dpaa2_eth_cls_rule) * dpaa2_eth_fs_count(priv), GFP_KERNEL); - if (!priv->cls_rules) + if (!priv->cls_rules) { + err = -ENOMEM; goto close; + } return 0; @@ -2716,7 +2943,7 @@ static int setup_rx_flow(struct dpaa2_eth_priv *priv, int err; err = dpni_get_queue(priv->mc_io, 0, priv->mc_token, - DPNI_QUEUE_RX, 0, fq->flowid, &queue, &qid); + DPNI_QUEUE_RX, fq->tc, fq->flowid, &queue, &qid); if (err) { dev_err(dev, "dpni_get_queue(RX) failed\n"); return err; @@ -2729,7 +2956,7 @@ static int setup_rx_flow(struct dpaa2_eth_priv *priv, queue.destination.priority = 1; queue.user_context = (u64)(uintptr_t)fq; err = dpni_set_queue(priv->mc_io, 0, priv->mc_token, - DPNI_QUEUE_RX, 0, fq->flowid, + DPNI_QUEUE_RX, fq->tc, fq->flowid, DPNI_QUEUE_OPT_USER_CTX | DPNI_QUEUE_OPT_DEST, &queue); if (err) { @@ -2738,6 +2965,10 @@ static int setup_rx_flow(struct dpaa2_eth_priv *priv, } /* xdp_rxq setup */ + /* only once for each channel */ + if (fq->tc > 0) + return 0; + err = xdp_rxq_info_reg(&fq->channel->xdp_rxq, priv->net_dev, fq->flowid); if (err) { @@ -2875,7 +3106,7 @@ static int config_legacy_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key) { struct device *dev = priv->net_dev->dev.parent; struct dpni_rx_tc_dist_cfg dist_cfg; - int err; + int i, err = 0; memset(&dist_cfg, 0, sizeof(dist_cfg)); @@ -2883,9 +3114,14 @@ static int config_legacy_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key) dist_cfg.dist_size = dpaa2_eth_queue_count(priv); dist_cfg.dist_mode = DPNI_DIST_MODE_HASH; - err = dpni_set_rx_tc_dist(priv->mc_io, 0, priv->mc_token, 0, &dist_cfg); - if (err) - dev_err(dev, "dpni_set_rx_tc_dist failed\n"); + for (i = 0; i < dpaa2_eth_tc_count(priv); i++) { + err = dpni_set_rx_tc_dist(priv->mc_io, 0, priv->mc_token, + i, &dist_cfg); + if (err) { + dev_err(dev, "dpni_set_rx_tc_dist failed\n"); + break; + } + } return err; } @@ -2895,7 +3131,7 @@ static int config_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key) { struct device *dev = priv->net_dev->dev.parent; struct dpni_rx_dist_cfg dist_cfg; - int err; + int i, err = 0; memset(&dist_cfg, 0, sizeof(dist_cfg)); @@ -2903,9 +3139,15 @@ static int config_hash_key(struct dpaa2_eth_priv *priv, dma_addr_t key) dist_cfg.dist_size = dpaa2_eth_queue_count(priv); dist_cfg.enable = 1; - err = dpni_set_rx_hash_dist(priv->mc_io, 0, priv->mc_token, &dist_cfg); - if (err) - dev_err(dev, "dpni_set_rx_hash_dist failed\n"); + for (i = 0; i < dpaa2_eth_tc_count(priv); i++) { + dist_cfg.tc = i; + err = dpni_set_rx_hash_dist(priv->mc_io, 0, priv->mc_token, + &dist_cfg); + if (err) { + dev_err(dev, "dpni_set_rx_hash_dist failed\n"); + break; + } + } return err; } @@ -2915,7 +3157,7 @@ static int config_cls_key(struct dpaa2_eth_priv *priv, dma_addr_t key) { struct device *dev = priv->net_dev->dev.parent; struct dpni_rx_dist_cfg dist_cfg; - int err; + int i, err = 0; memset(&dist_cfg, 0, sizeof(dist_cfg)); @@ -2923,9 +3165,15 @@ static int config_cls_key(struct dpaa2_eth_priv *priv, dma_addr_t key) dist_cfg.dist_size = dpaa2_eth_queue_count(priv); dist_cfg.enable = 1; - err = dpni_set_rx_fs_dist(priv->mc_io, 0, priv->mc_token, &dist_cfg); - if (err) - dev_err(dev, "dpni_set_rx_fs_dist failed\n"); + for (i = 0; i < dpaa2_eth_tc_count(priv); i++) { + dist_cfg.tc = i; + err = dpni_set_rx_fs_dist(priv->mc_io, 0, priv->mc_token, + &dist_cfg); + if (err) { + dev_err(dev, "dpni_set_rx_fs_dist failed\n"); + break; + } + } return err; } @@ -3611,6 +3859,15 @@ static int dpaa2_eth_probe(struct fsl_mc_device *dpni_dev) if (err) goto err_alloc_rings; +#ifdef CONFIG_FSL_DPAA2_ETH_DCB + if (dpaa2_eth_has_pause_support(priv) && priv->vlan_cls_enabled) { + priv->dcbx_mode = DCB_CAP_DCBX_HOST | DCB_CAP_DCBX_VER_IEEE; + net_dev->dcbnl_ops = &dpaa2_eth_dcbnl_ops; + } else { + dev_dbg(dev, "PFC not supported\n"); + } +#endif + err = setup_irqs(dpni_dev); if (err) { netdev_warn(net_dev, "Failed to set link interrupt, fall back to polling\n"); diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h index 13242bf5b427..2d7ada0f0dbd 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-eth.h @@ -1,11 +1,12 @@ /* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */ /* Copyright 2014-2016 Freescale Semiconductor Inc. - * Copyright 2016 NXP + * Copyright 2016-2020 NXP */ #ifndef __DPAA2_ETH_H #define __DPAA2_ETH_H +#include <linux/dcbnl.h> #include <linux/netdevice.h> #include <linux/if_vlan.h> #include <linux/fsl/mc.h> @@ -36,27 +37,46 @@ /* Convert L3 MTU to L2 MFL */ #define DPAA2_ETH_L2_MAX_FRM(mtu) ((mtu) + VLAN_ETH_HLEN) -/* Set the taildrop threshold (in bytes) to allow the enqueue of several jumbo - * frames in the Rx queues (length of the current frame is not - * taken into account when making the taildrop decision) +/* Set the taildrop threshold (in bytes) to allow the enqueue of a large + * enough number of jumbo frames in the Rx queues (length of the current + * frame is not taken into account when making the taildrop decision) */ -#define DPAA2_ETH_TAILDROP_THRESH (64 * 1024) +#define DPAA2_ETH_FQ_TAILDROP_THRESH (1024 * 1024) /* Maximum number of Tx confirmation frames to be processed * in a single NAPI call */ #define DPAA2_ETH_TXCONF_PER_NAPI 256 -/* Buffer quota per queue. Must be large enough such that for minimum sized - * frames taildrop kicks in before the bpool gets depleted, so we compute - * how many 64B frames fit inside the taildrop threshold and add a margin - * to accommodate the buffer refill delay. +/* Buffer qouta per channel. We want to keep in check number of ingress frames + * in flight: for small sized frames, congestion group taildrop may kick in + * first; for large sizes, Rx FQ taildrop threshold will ensure only a + * reasonable number of frames will be pending at any given time. + * Ingress frame drop due to buffer pool depletion should be a corner case only */ -#define DPAA2_ETH_MAX_FRAMES_PER_QUEUE (DPAA2_ETH_TAILDROP_THRESH / 64) -#define DPAA2_ETH_NUM_BUFS (DPAA2_ETH_MAX_FRAMES_PER_QUEUE + 256) +#define DPAA2_ETH_NUM_BUFS 1280 #define DPAA2_ETH_REFILL_THRESH \ (DPAA2_ETH_NUM_BUFS - DPAA2_ETH_BUFS_PER_CMD) +/* Congestion group taildrop threshold: number of frames allowed to accumulate + * at any moment in a group of Rx queues belonging to the same traffic class. + * Choose value such that we don't risk depleting the buffer pool before the + * taildrop kicks in + */ +#define DPAA2_ETH_CG_TAILDROP_THRESH(priv) \ + (1024 * dpaa2_eth_queue_count(priv) / dpaa2_eth_tc_count(priv)) + +/* Congestion group notification threshold: when this many frames accumulate + * on the Rx queues belonging to the same TC, the MAC is instructed to send + * PFC frames for that TC. + * When number of pending frames drops below exit threshold transmission of + * PFC frames is stopped. + */ +#define DPAA2_ETH_CN_THRESH_ENTRY(priv) \ + (DPAA2_ETH_CG_TAILDROP_THRESH(priv) / 2) +#define DPAA2_ETH_CN_THRESH_EXIT(priv) \ + (DPAA2_ETH_CN_THRESH_ENTRY(priv) * 3 / 4) + /* Maximum number of buffers that can be acquired/released through a single * QBMan command */ @@ -288,11 +308,15 @@ struct dpaa2_eth_ch_stats { __u64 xdp_tx; __u64 xdp_tx_err; __u64 xdp_redirect; + /* Must be last, does not show up in ethtool stats */ + __u64 frames; }; /* Maximum number of queues associated with a DPNI */ #define DPAA2_ETH_MAX_TCS 8 -#define DPAA2_ETH_MAX_RX_QUEUES 16 +#define DPAA2_ETH_MAX_RX_QUEUES_PER_TC 16 +#define DPAA2_ETH_MAX_RX_QUEUES \ + (DPAA2_ETH_MAX_RX_QUEUES_PER_TC * DPAA2_ETH_MAX_TCS) #define DPAA2_ETH_MAX_TX_QUEUES 16 #define DPAA2_ETH_MAX_QUEUES (DPAA2_ETH_MAX_RX_QUEUES + \ DPAA2_ETH_MAX_TX_QUEUES) @@ -308,6 +332,11 @@ enum dpaa2_eth_fq_type { struct dpaa2_eth_priv; +struct dpaa2_eth_xdp_fds { + struct dpaa2_fd fds[DEV_MAP_BULK_SIZE]; + ssize_t num; +}; + struct dpaa2_eth_fq { u32 fqid; u32 tx_qdbin; @@ -325,6 +354,9 @@ struct dpaa2_eth_fq { const struct dpaa2_fd *fd, struct dpaa2_eth_fq *fq); struct dpaa2_eth_fq_stats stats; + + struct dpaa2_eth_xdp_fds xdp_redirect_fds; + struct dpaa2_eth_xdp_fds xdp_tx_fds; }; struct dpaa2_eth_ch_xdp { @@ -371,7 +403,9 @@ struct dpaa2_eth_priv { struct dpaa2_eth_fq fq[DPAA2_ETH_MAX_QUEUES]; int (*enqueue)(struct dpaa2_eth_priv *priv, struct dpaa2_eth_fq *fq, - struct dpaa2_fd *fd, u8 prio); + struct dpaa2_fd *fd, u8 prio, + u32 num_frames, + int *frames_enqueued); u8 num_channels; struct dpaa2_eth_channel *channel[DPAA2_ETH_MAX_DPCONS]; @@ -402,7 +436,8 @@ struct dpaa2_eth_priv { struct dpaa2_eth_drv_stats __percpu *percpu_extras; u16 mc_token; - u8 rx_td_enabled; + u8 rx_fqtd_enabled; + u8 rx_cgtd_enabled; struct dpni_link_state link_state; bool do_link_poll; @@ -413,6 +448,12 @@ struct dpaa2_eth_priv { u64 rx_cls_fields; struct dpaa2_eth_cls_rule *cls_rules; u8 rx_cls_enabled; + u8 vlan_cls_enabled; + u8 pfc_enabled; +#ifdef CONFIG_FSL_DPAA2_ETH_DCB + u8 dcbx_mode; + struct ieee_pfc pfc; +#endif struct bpf_prog *xdp_prog; #ifdef CONFIG_DEBUG_FS struct dpaa2_debugfs dbg; @@ -495,6 +536,17 @@ enum dpaa2_eth_rx_dist { (dpaa2_eth_cmp_dpni_ver((priv), DPNI_PAUSE_VER_MAJOR, \ DPNI_PAUSE_VER_MINOR) >= 0) +static inline bool dpaa2_eth_tx_pause_enabled(u64 link_options) +{ + return !!(link_options & DPNI_LINK_OPT_PAUSE) ^ + !!(link_options & DPNI_LINK_OPT_ASYM_PAUSE); +} + +static inline bool dpaa2_eth_rx_pause_enabled(u64 link_options) +{ + return !!(link_options & DPNI_LINK_OPT_PAUSE); +} + static inline unsigned int dpaa2_eth_needed_headroom(struct dpaa2_eth_priv *priv, struct sk_buff *skb) @@ -534,4 +586,9 @@ int dpaa2_eth_cls_key_size(u64 key); int dpaa2_eth_cls_fld_off(int prot, int field); void dpaa2_eth_cls_trim_rule(void *key_mem, u64 fields); +void dpaa2_eth_set_rx_taildrop(struct dpaa2_eth_priv *priv, + bool tx_pause, bool pfc); + +extern const struct dcbnl_rtnl_ops dpaa2_eth_dcbnl_ops; + #endif /* __DPAA2_H */ diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c index b7141fdc279e..e88269fe3de7 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c @@ -130,9 +130,8 @@ static void dpaa2_eth_get_pauseparam(struct net_device *net_dev, return; } - pause->rx_pause = !!(link_options & DPNI_LINK_OPT_PAUSE); - pause->tx_pause = pause->rx_pause ^ - !!(link_options & DPNI_LINK_OPT_ASYM_PAUSE); + pause->rx_pause = dpaa2_eth_rx_pause_enabled(link_options); + pause->tx_pause = dpaa2_eth_tx_pause_enabled(link_options); pause->autoneg = AUTONEG_DISABLE; } @@ -277,7 +276,7 @@ static void dpaa2_eth_get_ethtool_stats(struct net_device *net_dev, /* Per-channel stats */ for (k = 0; k < priv->num_channels; k++) { ch_stats = &priv->channel[k]->stats; - for (j = 0; j < sizeof(*ch_stats) / sizeof(__u64); j++) + for (j = 0; j < sizeof(*ch_stats) / sizeof(__u64) - 1; j++) *((__u64 *)data + i + j) += *((__u64 *)ch_stats + j); } i += j; @@ -547,7 +546,7 @@ static int do_cls_rule(struct net_device *net_dev, dma_addr_t key_iova; u64 fields = 0; void *key_buf; - int err; + int i, err; if (fs->ring_cookie != RX_CLS_FLOW_DISC && fs->ring_cookie >= dpaa2_eth_queue_count(priv)) @@ -607,11 +606,18 @@ static int do_cls_rule(struct net_device *net_dev, fs_act.options |= DPNI_FS_OPT_DISCARD; else fs_act.flow_id = fs->ring_cookie; - err = dpni_add_fs_entry(priv->mc_io, 0, priv->mc_token, 0, - fs->location, &rule_cfg, &fs_act); - } else { - err = dpni_remove_fs_entry(priv->mc_io, 0, priv->mc_token, 0, - &rule_cfg); + } + for (i = 0; i < dpaa2_eth_tc_count(priv); i++) { + if (add) + err = dpni_add_fs_entry(priv->mc_io, 0, priv->mc_token, + i, fs->location, &rule_cfg, + &fs_act); + else + err = dpni_remove_fs_entry(priv->mc_io, 0, + priv->mc_token, i, + &rule_cfg); + if (err) + break; } dma_unmap_single(dev, key_iova, rule_cfg.key_size * 2, DMA_TO_DEVICE); diff --git a/drivers/net/ethernet/freescale/dpaa2/dpni-cmd.h b/drivers/net/ethernet/freescale/dpaa2/dpni-cmd.h index d9b6918807af..fd069f67be9b 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpni-cmd.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpni-cmd.h @@ -59,6 +59,10 @@ #define DPNI_CMDID_SET_RX_TC_DIST DPNI_CMD(0x235) +#define DPNI_CMDID_SET_QOS_TBL DPNI_CMD(0x240) +#define DPNI_CMDID_ADD_QOS_ENT DPNI_CMD(0x241) +#define DPNI_CMDID_REMOVE_QOS_ENT DPNI_CMD(0x242) +#define DPNI_CMDID_CLR_QOS_TBL DPNI_CMD(0x243) #define DPNI_CMDID_ADD_FS_ENT DPNI_CMD(0x244) #define DPNI_CMDID_REMOVE_FS_ENT DPNI_CMD(0x245) #define DPNI_CMDID_CLR_FS_ENT DPNI_CMD(0x246) @@ -567,4 +571,59 @@ struct dpni_cmd_remove_fs_entry { __le64 mask_iova; }; +#define DPNI_DISCARD_ON_MISS_SHIFT 0 +#define DPNI_DISCARD_ON_MISS_SIZE 1 + +struct dpni_cmd_set_qos_table { + __le32 pad; + u8 default_tc; + /* only the LSB */ + u8 discard_on_miss; + __le16 pad1[21]; + __le64 key_cfg_iova; +}; + +struct dpni_cmd_add_qos_entry { + __le16 pad; + u8 tc_id; + u8 key_size; + __le16 index; + __le16 pad1; + __le64 key_iova; + __le64 mask_iova; +}; + +struct dpni_cmd_remove_qos_entry { + u8 pad[3]; + u8 key_size; + __le32 pad1; + __le64 key_iova; + __le64 mask_iova; +}; + +#define DPNI_DEST_TYPE_SHIFT 0 +#define DPNI_DEST_TYPE_SIZE 4 +#define DPNI_CONG_UNITS_SHIFT 4 +#define DPNI_CONG_UNITS_SIZE 2 + +struct dpni_cmd_set_congestion_notification { + /* cmd word 0 */ + u8 qtype; + u8 tc; + u8 pad[6]; + /* cmd word 1 */ + __le32 dest_id; + __le16 notification_mode; + u8 dest_priority; + /* from LSB: dest_type: 4 units:2 */ + u8 type_units; + /* cmd word 2 */ + __le64 message_iova; + /* cmd word 3 */ + __le64 message_ctx; + /* cmd word 4 */ + __le32 threshold_entry; + __le32 threshold_exit; +}; + #endif /* _FSL_DPNI_CMD_H */ diff --git a/drivers/net/ethernet/freescale/dpaa2/dpni.c b/drivers/net/ethernet/freescale/dpaa2/dpni.c index dd54e6953aeb..6b479ba66465 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpni.c +++ b/drivers/net/ethernet/freescale/dpaa2/dpni.c @@ -1355,6 +1355,52 @@ int dpni_set_rx_tc_dist(struct fsl_mc_io *mc_io, } /** + * dpni_set_congestion_notification() - Set traffic class congestion + * notification configuration + * @mc_io: Pointer to MC portal's I/O object + * @cmd_flags: Command flags; one or more of 'MC_CMD_FLAG_' + * @token: Token of DPNI object + * @qtype: Type of queue - Rx, Tx and Tx confirm types are supported + * @tc_id: Traffic class selection (0-7) + * @cfg: Congestion notification configuration + * + * Return: '0' on Success; error code otherwise. + */ +int dpni_set_congestion_notification( + struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token, + enum dpni_queue_type qtype, + u8 tc_id, + const struct dpni_congestion_notification_cfg *cfg) +{ + struct dpni_cmd_set_congestion_notification *cmd_params; + struct fsl_mc_command cmd = { 0 }; + + /* prepare command */ + cmd.header = + mc_encode_cmd_header(DPNI_CMDID_SET_CONGESTION_NOTIFICATION, + cmd_flags, + token); + cmd_params = (struct dpni_cmd_set_congestion_notification *)cmd.params; + cmd_params->qtype = qtype; + cmd_params->tc = tc_id; + cmd_params->dest_id = cpu_to_le32(cfg->dest_cfg.dest_id); + cmd_params->notification_mode = cpu_to_le16(cfg->notification_mode); + cmd_params->dest_priority = cfg->dest_cfg.priority; + dpni_set_field(cmd_params->type_units, DEST_TYPE, + cfg->dest_cfg.dest_type); + dpni_set_field(cmd_params->type_units, CONG_UNITS, cfg->units); + cmd_params->message_iova = cpu_to_le64(cfg->message_iova); + cmd_params->message_ctx = cpu_to_le64(cfg->message_ctx); + cmd_params->threshold_entry = cpu_to_le32(cfg->threshold_entry); + cmd_params->threshold_exit = cpu_to_le32(cfg->threshold_exit); + + /* send command to mc*/ + return mc_send_command(mc_io, &cmd); +} + +/** * dpni_set_queue() - Set queue parameters * @mc_io: Pointer to MC portal's I/O object * @cmd_flags: Command flags; one or more of 'MC_CMD_FLAG_' @@ -1786,3 +1832,134 @@ int dpni_remove_fs_entry(struct fsl_mc_io *mc_io, /* send command to mc*/ return mc_send_command(mc_io, &cmd); } + +/** + * dpni_set_qos_table() - Set QoS mapping table + * @mc_io: Pointer to MC portal's I/O object + * @cmd_flags: Command flags; one or more of 'MC_CMD_FLAG_' + * @token: Token of DPNI object + * @cfg: QoS table configuration + * + * This function and all QoS-related functions require that + *'max_tcs > 1' was set at DPNI creation. + * + * warning: Before calling this function, call dpkg_prepare_key_cfg() to + * prepare the key_cfg_iova parameter + * + * Return: '0' on Success; Error code otherwise. + */ +int dpni_set_qos_table(struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token, + const struct dpni_qos_tbl_cfg *cfg) +{ + struct dpni_cmd_set_qos_table *cmd_params; + struct fsl_mc_command cmd = { 0 }; + + /* prepare command */ + cmd.header = mc_encode_cmd_header(DPNI_CMDID_SET_QOS_TBL, + cmd_flags, + token); + cmd_params = (struct dpni_cmd_set_qos_table *)cmd.params; + cmd_params->default_tc = cfg->default_tc; + cmd_params->key_cfg_iova = cpu_to_le64(cfg->key_cfg_iova); + dpni_set_field(cmd_params->discard_on_miss, DISCARD_ON_MISS, + cfg->discard_on_miss); + + /* send command to mc*/ + return mc_send_command(mc_io, &cmd); +} + +/** + * dpni_add_qos_entry() - Add QoS mapping entry (to select a traffic class) + * @mc_io: Pointer to MC portal's I/O object + * @cmd_flags: Command flags; one or more of 'MC_CMD_FLAG_' + * @token: Token of DPNI object + * @cfg: QoS rule to add + * @tc_id: Traffic class selection (0-7) + * @index: Location in the QoS table where to insert the entry. + * Only relevant if MASKING is enabled for QoS classification on + * this DPNI, it is ignored for exact match. + * + * Return: '0' on Success; Error code otherwise. + */ +int dpni_add_qos_entry(struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token, + const struct dpni_rule_cfg *cfg, + u8 tc_id, + u16 index) +{ + struct dpni_cmd_add_qos_entry *cmd_params; + struct fsl_mc_command cmd = { 0 }; + + /* prepare command */ + cmd.header = mc_encode_cmd_header(DPNI_CMDID_ADD_QOS_ENT, + cmd_flags, + token); + cmd_params = (struct dpni_cmd_add_qos_entry *)cmd.params; + cmd_params->tc_id = tc_id; + cmd_params->key_size = cfg->key_size; + cmd_params->index = cpu_to_le16(index); + cmd_params->key_iova = cpu_to_le64(cfg->key_iova); + cmd_params->mask_iova = cpu_to_le64(cfg->mask_iova); + + /* send command to mc*/ + return mc_send_command(mc_io, &cmd); +} + +/** + * dpni_remove_qos_entry() - Remove QoS mapping entry + * @mc_io: Pointer to MC portal's I/O object + * @cmd_flags: Command flags; one or more of 'MC_CMD_FLAG_' + * @token: Token of DPNI object + * @cfg: QoS rule to remove + * + * Return: '0' on Success; Error code otherwise. + */ +int dpni_remove_qos_entry(struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token, + const struct dpni_rule_cfg *cfg) +{ + struct dpni_cmd_remove_qos_entry *cmd_params; + struct fsl_mc_command cmd = { 0 }; + + /* prepare command */ + cmd.header = mc_encode_cmd_header(DPNI_CMDID_REMOVE_QOS_ENT, + cmd_flags, + token); + cmd_params = (struct dpni_cmd_remove_qos_entry *)cmd.params; + cmd_params->key_size = cfg->key_size; + cmd_params->key_iova = cpu_to_le64(cfg->key_iova); + cmd_params->mask_iova = cpu_to_le64(cfg->mask_iova); + + /* send command to mc*/ + return mc_send_command(mc_io, &cmd); +} + +/** + * dpni_clear_qos_table() - Clear all QoS mapping entries + * @mc_io: Pointer to MC portal's I/O object + * @cmd_flags: Command flags; one or more of 'MC_CMD_FLAG_' + * @token: Token of DPNI object + * + * Following this function call, all frames are directed to + * the default traffic class (0) + * + * Return: '0' on Success; Error code otherwise. + */ +int dpni_clear_qos_table(struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token) +{ + struct fsl_mc_command cmd = { 0 }; + + /* prepare command */ + cmd.header = mc_encode_cmd_header(DPNI_CMDID_CLR_QOS_TBL, + cmd_flags, + token); + + /* send command to mc*/ + return mc_send_command(mc_io, &cmd); +} diff --git a/drivers/net/ethernet/freescale/dpaa2/dpni.h b/drivers/net/ethernet/freescale/dpaa2/dpni.h index ee0711d06b3a..e874d8084142 100644 --- a/drivers/net/ethernet/freescale/dpaa2/dpni.h +++ b/drivers/net/ethernet/freescale/dpaa2/dpni.h @@ -514,6 +514,11 @@ int dpni_get_statistics(struct fsl_mc_io *mc_io, #define DPNI_LINK_OPT_ASYM_PAUSE 0x0000000000000008ULL /** + * Enable priority flow control pause frames + */ +#define DPNI_LINK_OPT_PFC_PAUSE 0x0000000000000010ULL + +/** * struct - Structure representing DPNI link configuration * @rate: Rate * @options: Mask of available options; use 'DPNI_LINK_OPT_<X>' values @@ -716,6 +721,26 @@ int dpni_set_rx_hash_dist(struct fsl_mc_io *mc_io, const struct dpni_rx_dist_cfg *cfg); /** + * struct dpni_qos_tbl_cfg - Structure representing QOS table configuration + * @key_cfg_iova: I/O virtual address of 256 bytes DMA-able memory filled with + * key extractions to be used as the QoS criteria by calling + * dpkg_prepare_key_cfg() + * @discard_on_miss: Set to '1' to discard frames in case of no match (miss); + * '0' to use the 'default_tc' in such cases + * @default_tc: Used in case of no-match and 'discard_on_miss'= 0 + */ +struct dpni_qos_tbl_cfg { + u64 key_cfg_iova; + int discard_on_miss; + u8 default_tc; +}; + +int dpni_set_qos_table(struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token, + const struct dpni_qos_tbl_cfg *cfg); + +/** * enum dpni_dest - DPNI destination types * @DPNI_DEST_NONE: Unassigned destination; The queue is set in parked mode and * does not generate FQDAN notifications; user is expected to @@ -858,6 +883,62 @@ enum dpni_congestion_point { }; /** + * struct dpni_dest_cfg - Structure representing DPNI destination parameters + * @dest_type: Destination type + * @dest_id: Either DPIO ID or DPCON ID, depending on the destination type + * @priority: Priority selection within the DPIO or DPCON channel; valid + * values are 0-1 or 0-7, depending on the number of priorities + * in that channel; not relevant for 'DPNI_DEST_NONE' option + */ +struct dpni_dest_cfg { + enum dpni_dest dest_type; + int dest_id; + u8 priority; +}; + +/* DPNI congestion options */ + +/** + * This congestion will trigger flow control or priority flow control. + * This will have effect only if flow control is enabled with + * dpni_set_link_cfg(). + */ +#define DPNI_CONG_OPT_FLOW_CONTROL 0x00000040 + +/** + * struct dpni_congestion_notification_cfg - congestion notification + * configuration + * @units: Units type + * @threshold_entry: Above this threshold we enter a congestion state. + * set it to '0' to disable it + * @threshold_exit: Below this threshold we exit the congestion state. + * @message_ctx: The context that will be part of the CSCN message + * @message_iova: I/O virtual address (must be in DMA-able memory), + * must be 16B aligned; valid only if 'DPNI_CONG_OPT_WRITE_MEM_<X>' + * is contained in 'options' + * @dest_cfg: CSCN can be send to either DPIO or DPCON WQ channel + * @notification_mode: Mask of available options; use 'DPNI_CONG_OPT_<X>' values + */ + +struct dpni_congestion_notification_cfg { + enum dpni_congestion_unit units; + u32 threshold_entry; + u32 threshold_exit; + u64 message_ctx; + u64 message_iova; + struct dpni_dest_cfg dest_cfg; + u16 notification_mode; +}; + +int dpni_set_congestion_notification( + struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token, + enum dpni_queue_type qtype, + u8 tc_id, + const struct dpni_congestion_notification_cfg *cfg); + +/** * struct dpni_taildrop - Structure representing the taildrop * @enable: Indicates whether the taildrop is active or not. * @units: Indicates the unit of THRESHOLD. Queue taildrop only supports @@ -961,6 +1042,22 @@ int dpni_remove_fs_entry(struct fsl_mc_io *mc_io, u8 tc_id, const struct dpni_rule_cfg *cfg); +int dpni_add_qos_entry(struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token, + const struct dpni_rule_cfg *cfg, + u8 tc_id, + u16 index); + +int dpni_remove_qos_entry(struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token, + const struct dpni_rule_cfg *cfg); + +int dpni_clear_qos_table(struct fsl_mc_io *mc_io, + u32 cmd_flags, + u16 token); + int dpni_get_api_version(struct fsl_mc_io *mc_io, u32 cmd_flags, u16 *major_ver, diff --git a/drivers/net/ethernet/freescale/enetc/enetc.c b/drivers/net/ethernet/freescale/enetc/enetc.c index ccf2611f4a20..298c55786fd9 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.c +++ b/drivers/net/ethernet/freescale/enetc/enetc.c @@ -756,6 +756,9 @@ void enetc_get_si_caps(struct enetc_si *si) if (val & ENETC_SIPCAPR0_QBV) si->hw_features |= ENETC_SI_F_QBV; + + if (val & ENETC_SIPCAPR0_PSFP) + si->hw_features |= ENETC_SI_F_PSFP; } static int enetc_dma_alloc_bdr(struct enetc_bdr *r, size_t bd_size) @@ -1518,6 +1521,8 @@ int enetc_setup_tc(struct net_device *ndev, enum tc_setup_type type, return enetc_setup_tc_cbs(ndev, type_data); case TC_SETUP_QDISC_ETF: return enetc_setup_tc_txtime(ndev, type_data); + case TC_SETUP_BLOCK: + return enetc_setup_tc_psfp(ndev, type_data); default: return -EOPNOTSUPP; } @@ -1567,15 +1572,42 @@ static int enetc_set_rss(struct net_device *ndev, int en) return 0; } +static int enetc_set_psfp(struct net_device *ndev, int en) +{ + struct enetc_ndev_priv *priv = netdev_priv(ndev); + int err; + + if (en) { + err = enetc_psfp_enable(priv); + if (err) + return err; + + priv->active_offloads |= ENETC_F_QCI; + return 0; + } + + err = enetc_psfp_disable(priv); + if (err) + return err; + + priv->active_offloads &= ~ENETC_F_QCI; + + return 0; +} + int enetc_set_features(struct net_device *ndev, netdev_features_t features) { netdev_features_t changed = ndev->features ^ features; + int err = 0; if (changed & NETIF_F_RXHASH) enetc_set_rss(ndev, !!(features & NETIF_F_RXHASH)); - return 0; + if (changed & NETIF_F_HW_TC) + err = enetc_set_psfp(ndev, !!(features & NETIF_F_HW_TC)); + + return err; } #ifdef CONFIG_FSL_ENETC_PTP_CLOCK diff --git a/drivers/net/ethernet/freescale/enetc/enetc.h b/drivers/net/ethernet/freescale/enetc/enetc.h index 56c43f35b633..b705464f6882 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc.h +++ b/drivers/net/ethernet/freescale/enetc/enetc.h @@ -151,6 +151,7 @@ enum enetc_errata { }; #define ENETC_SI_F_QBV BIT(0) +#define ENETC_SI_F_PSFP BIT(1) /* PCI IEP device data */ struct enetc_si { @@ -203,12 +204,20 @@ struct enetc_cls_rule { }; #define ENETC_MAX_BDR_INT 2 /* fixed to max # of available cpus */ +struct psfp_cap { + u32 max_streamid; + u32 max_psfp_filter; + u32 max_psfp_gate; + u32 max_psfp_gatelist; + u32 max_psfp_meter; +}; /* TODO: more hardware offloads */ enum enetc_active_offloads { ENETC_F_RX_TSTAMP = BIT(0), ENETC_F_TX_TSTAMP = BIT(1), ENETC_F_QBV = BIT(2), + ENETC_F_QCI = BIT(3), }; struct enetc_ndev_priv { @@ -231,6 +240,8 @@ struct enetc_ndev_priv { struct enetc_cls_rule *cls_rules; + struct psfp_cap psfp_cap; + struct device_node *phy_node; phy_interface_t if_mode; }; @@ -289,9 +300,84 @@ int enetc_setup_tc_taprio(struct net_device *ndev, void *type_data); void enetc_sched_speed_set(struct net_device *ndev); int enetc_setup_tc_cbs(struct net_device *ndev, void *type_data); int enetc_setup_tc_txtime(struct net_device *ndev, void *type_data); +int enetc_setup_tc_block_cb(enum tc_setup_type type, void *type_data, + void *cb_priv); +int enetc_setup_tc_psfp(struct net_device *ndev, void *type_data); +int enetc_psfp_init(struct enetc_ndev_priv *priv); +int enetc_psfp_clean(struct enetc_ndev_priv *priv); + +static inline void enetc_get_max_cap(struct enetc_ndev_priv *priv) +{ + u32 reg; + + reg = enetc_port_rd(&priv->si->hw, ENETC_PSIDCAPR); + priv->psfp_cap.max_streamid = reg & ENETC_PSIDCAPR_MSK; + /* Port stream filter capability */ + reg = enetc_port_rd(&priv->si->hw, ENETC_PSFCAPR); + priv->psfp_cap.max_psfp_filter = reg & ENETC_PSFCAPR_MSK; + /* Port stream gate capability */ + reg = enetc_port_rd(&priv->si->hw, ENETC_PSGCAPR); + priv->psfp_cap.max_psfp_gate = (reg & ENETC_PSGCAPR_SGIT_MSK); + priv->psfp_cap.max_psfp_gatelist = (reg & ENETC_PSGCAPR_GCL_MSK) >> 16; + /* Port flow meter capability */ + reg = enetc_port_rd(&priv->si->hw, ENETC_PFMCAPR); + priv->psfp_cap.max_psfp_meter = reg & ENETC_PFMCAPR_MSK; +} + +static inline int enetc_psfp_enable(struct enetc_ndev_priv *priv) +{ + struct enetc_hw *hw = &priv->si->hw; + int err; + + enetc_get_max_cap(priv); + + err = enetc_psfp_init(priv); + if (err) + return err; + + enetc_wr(hw, ENETC_PPSFPMR, enetc_rd(hw, ENETC_PPSFPMR) | + ENETC_PPSFPMR_PSFPEN | ENETC_PPSFPMR_VS | + ENETC_PPSFPMR_PVC | ENETC_PPSFPMR_PVZC); + + return 0; +} + +static inline int enetc_psfp_disable(struct enetc_ndev_priv *priv) +{ + struct enetc_hw *hw = &priv->si->hw; + int err; + + err = enetc_psfp_clean(priv); + if (err) + return err; + + enetc_wr(hw, ENETC_PPSFPMR, enetc_rd(hw, ENETC_PPSFPMR) & + ~ENETC_PPSFPMR_PSFPEN & ~ENETC_PPSFPMR_VS & + ~ENETC_PPSFPMR_PVC & ~ENETC_PPSFPMR_PVZC); + + memset(&priv->psfp_cap, 0, sizeof(struct psfp_cap)); + + return 0; +} + #else #define enetc_setup_tc_taprio(ndev, type_data) -EOPNOTSUPP #define enetc_sched_speed_set(ndev) (void)0 #define enetc_setup_tc_cbs(ndev, type_data) -EOPNOTSUPP #define enetc_setup_tc_txtime(ndev, type_data) -EOPNOTSUPP +#define enetc_setup_tc_psfp(ndev, type_data) -EOPNOTSUPP +#define enetc_setup_tc_block_cb NULL + +#define enetc_get_max_cap(p) \ + memset(&((p)->psfp_cap), 0, sizeof(struct psfp_cap)) + +static inline int enetc_psfp_enable(struct enetc_ndev_priv *priv) +{ + return 0; +} + +static inline int enetc_psfp_disable(struct enetc_ndev_priv *priv) +{ + return 0; +} #endif diff --git a/drivers/net/ethernet/freescale/enetc/enetc_hw.h b/drivers/net/ethernet/freescale/enetc/enetc_hw.h index 2a6523136947..6314051bc6c1 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_hw.h +++ b/drivers/net/ethernet/freescale/enetc/enetc_hw.h @@ -19,6 +19,7 @@ #define ENETC_SICTR1 0x1c #define ENETC_SIPCAPR0 0x20 #define ENETC_SIPCAPR0_QBV BIT(4) +#define ENETC_SIPCAPR0_PSFP BIT(9) #define ENETC_SIPCAPR0_RSS BIT(8) #define ENETC_SIPCAPR1 0x24 #define ENETC_SITGTGR 0x30 @@ -228,6 +229,15 @@ enum enetc_bdr_type {TX, RX}; #define ENETC_PM0_IFM_RLP (BIT(5) | BIT(11)) #define ENETC_PM0_IFM_RGAUTO (BIT(15) | ENETC_PMO_IFM_RG | BIT(1)) #define ENETC_PM0_IFM_XGMII BIT(12) +#define ENETC_PSIDCAPR 0x1b08 +#define ENETC_PSIDCAPR_MSK GENMASK(15, 0) +#define ENETC_PSFCAPR 0x1b18 +#define ENETC_PSFCAPR_MSK GENMASK(15, 0) +#define ENETC_PSGCAPR 0x1b28 +#define ENETC_PSGCAPR_GCL_MSK GENMASK(18, 16) +#define ENETC_PSGCAPR_SGIT_MSK GENMASK(15, 0) +#define ENETC_PFMCAPR 0x1b38 +#define ENETC_PFMCAPR_MSK GENMASK(15, 0) /* MAC counters */ #define ENETC_PM0_REOCT 0x8100 @@ -557,6 +567,9 @@ enum bdcr_cmd_class { BDCR_CMD_RFS, BDCR_CMD_PORT_GCL, BDCR_CMD_RECV_CLASSIFIER, + BDCR_CMD_STREAM_IDENTIFY, + BDCR_CMD_STREAM_FILTER, + BDCR_CMD_STREAM_GCL, __BDCR_CMD_MAX_LEN, BDCR_CMD_MAX_LEN = __BDCR_CMD_MAX_LEN - 1, }; @@ -588,13 +601,152 @@ struct tgs_gcl_data { struct gce entry[]; }; +/* class 7, command 0, Stream Identity Entry Configuration */ +struct streamid_conf { + __le32 stream_handle; /* init gate value */ + __le32 iports; + u8 id_type; + u8 oui[3]; + u8 res[3]; + u8 en; +}; + +#define ENETC_CBDR_SID_VID_MASK 0xfff +#define ENETC_CBDR_SID_VIDM BIT(12) +#define ENETC_CBDR_SID_TG_MASK 0xc000 +/* streamid_conf address point to this data space */ +struct streamid_data { + union { + u8 dmac[6]; + u8 smac[6]; + }; + u16 vid_vidm_tg; +}; + +#define ENETC_CBDR_SFI_PRI_MASK 0x7 +#define ENETC_CBDR_SFI_PRIM BIT(3) +#define ENETC_CBDR_SFI_BLOV BIT(4) +#define ENETC_CBDR_SFI_BLEN BIT(5) +#define ENETC_CBDR_SFI_MSDUEN BIT(6) +#define ENETC_CBDR_SFI_FMITEN BIT(7) +#define ENETC_CBDR_SFI_ENABLE BIT(7) +/* class 8, command 0, Stream Filter Instance, Short Format */ +struct sfi_conf { + __le32 stream_handle; + u8 multi; + u8 res[2]; + u8 sthm; + /* Max Service Data Unit or Flow Meter Instance Table index. + * Depending on the value of FLT this represents either Max + * Service Data Unit (max frame size) allowed by the filter + * entry or is an index into the Flow Meter Instance table + * index identifying the policer which will be used to police + * it. + */ + __le16 fm_inst_table_index; + __le16 msdu; + __le16 sg_inst_table_index; + u8 res1[2]; + __le32 input_ports; + u8 res2[3]; + u8 en; +}; + +/* class 8, command 2 stream Filter Instance status query short format + * command no need structure define + * Stream Filter Instance Query Statistics Response data + */ +struct sfi_counter_data { + u32 matchl; + u32 matchh; + u32 msdu_dropl; + u32 msdu_droph; + u32 stream_gate_dropl; + u32 stream_gate_droph; + u32 flow_meter_dropl; + u32 flow_meter_droph; +}; + +#define ENETC_CBDR_SGI_OIPV_MASK 0x7 +#define ENETC_CBDR_SGI_OIPV_EN BIT(3) +#define ENETC_CBDR_SGI_CGTST BIT(6) +#define ENETC_CBDR_SGI_OGTST BIT(7) +#define ENETC_CBDR_SGI_CFG_CHG BIT(1) +#define ENETC_CBDR_SGI_CFG_PND BIT(2) +#define ENETC_CBDR_SGI_OEX BIT(4) +#define ENETC_CBDR_SGI_OEXEN BIT(5) +#define ENETC_CBDR_SGI_IRX BIT(6) +#define ENETC_CBDR_SGI_IRXEN BIT(7) +#define ENETC_CBDR_SGI_ACLLEN_MASK 0x3 +#define ENETC_CBDR_SGI_OCLLEN_MASK 0xc +#define ENETC_CBDR_SGI_EN BIT(7) +/* class 9, command 0, Stream Gate Instance Table, Short Format + * class 9, command 2, Stream Gate Instance Table entry query write back + * Short Format + */ +struct sgi_table { + u8 res[8]; + u8 oipv; + u8 res0[2]; + u8 ocgtst; + u8 res1[7]; + u8 gset; + u8 oacl_len; + u8 res2[2]; + u8 en; +}; + +#define ENETC_CBDR_SGI_AIPV_MASK 0x7 +#define ENETC_CBDR_SGI_AIPV_EN BIT(3) +#define ENETC_CBDR_SGI_AGTST BIT(7) + +/* class 9, command 1, Stream Gate Control List, Long Format */ +struct sgcl_conf { + u8 aipv; + u8 res[2]; + u8 agtst; + u8 res1[4]; + union { + struct { + u8 res2[4]; + u8 acl_len; + u8 res3[3]; + }; + u8 cct[8]; /* Config change time */ + }; +}; + +#define ENETC_CBDR_SGL_IOMEN BIT(0) +#define ENETC_CBDR_SGL_IPVEN BIT(3) +#define ENETC_CBDR_SGL_GTST BIT(4) +#define ENETC_CBDR_SGL_IPV_MASK 0xe +/* Stream Gate Control List Entry */ +struct sgce { + u32 interval; + u8 msdu[3]; + u8 multi; +}; + +/* stream control list class 9 , cmd 1 data buffer */ +struct sgcl_data { + u32 btl; + u32 bth; + u32 ct; + u32 cte; + struct sgce sgcl[0]; +}; + struct enetc_cbd { union{ + struct sfi_conf sfi_conf; + struct sgi_table sgi_table; struct { __le32 addr[2]; union { __le32 opt[4]; struct tgs_gcl_conf gcl_conf; + struct streamid_conf sid_set; + struct sgcl_conf sgcl_conf; }; }; /* Long format */ __le32 data[6]; @@ -621,3 +773,10 @@ struct enetc_cbd { /* Port time specific departure */ #define ENETC_PTCTSDR(n) (0x1210 + 4 * (n)) #define ENETC_TSDE BIT(31) + +/* PSFP setting */ +#define ENETC_PPSFPMR 0x11b00 +#define ENETC_PPSFPMR_PSFPEN BIT(0) +#define ENETC_PPSFPMR_VS BIT(1) +#define ENETC_PPSFPMR_PVC BIT(2) +#define ENETC_PPSFPMR_PVZC BIT(3) diff --git a/drivers/net/ethernet/freescale/enetc/enetc_pf.c b/drivers/net/ethernet/freescale/enetc/enetc_pf.c index 85e2b741df41..824d211ec00f 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_pf.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_pf.c @@ -50,21 +50,6 @@ static void enetc_set_vlan_promisc(struct enetc_hw *hw, char si_map) enetc_port_wr(hw, ENETC_PSIPVMR, ENETC_PSIPVMR_SET_VP(si_map) | val); } -static bool enetc_si_vlan_promisc_is_on(struct enetc_pf *pf, int si_idx) -{ - return pf->vlan_promisc_simap & BIT(si_idx); -} - -static bool enetc_vlan_filter_is_on(struct enetc_pf *pf) -{ - int i; - - for_each_set_bit(i, pf->active_vlans, VLAN_N_VID) - return true; - - return false; -} - static void enetc_enable_si_vlan_promisc(struct enetc_pf *pf, int si_idx) { pf->vlan_promisc_simap |= BIT(si_idx); @@ -204,6 +189,7 @@ static void enetc_pf_set_rx_mode(struct net_device *ndev) { struct enetc_ndev_priv *priv = netdev_priv(ndev); struct enetc_pf *pf = enetc_si_priv(priv->si); + char vlan_promisc_simap = pf->vlan_promisc_simap; struct enetc_hw *hw = &priv->si->hw; bool uprom = false, mprom = false; struct enetc_mac_filter *filter; @@ -216,16 +202,16 @@ static void enetc_pf_set_rx_mode(struct net_device *ndev) psipmr = ENETC_PSIPMR_SET_UP(0) | ENETC_PSIPMR_SET_MP(0); uprom = true; mprom = true; - /* enable VLAN promisc mode for SI0 */ - if (!enetc_si_vlan_promisc_is_on(pf, 0)) - enetc_enable_si_vlan_promisc(pf, 0); - + /* Enable VLAN promiscuous mode for SI0 (PF) */ + vlan_promisc_simap |= BIT(0); } else if (ndev->flags & IFF_ALLMULTI) { /* enable multi cast promisc mode for SI0 (PF) */ psipmr = ENETC_PSIPMR_SET_MP(0); mprom = true; } + enetc_set_vlan_promisc(&pf->si->hw, vlan_promisc_simap); + /* first 2 filter entries belong to PF */ if (!uprom) { /* Update unicast filters */ @@ -306,9 +292,6 @@ static int enetc_vlan_rx_add_vid(struct net_device *ndev, __be16 prot, u16 vid) struct enetc_pf *pf = enetc_si_priv(priv->si); int idx; - if (enetc_si_vlan_promisc_is_on(pf, 0)) - enetc_disable_si_vlan_promisc(pf, 0); - __set_bit(vid, pf->active_vlans); idx = enetc_vid_hash_idx(vid); @@ -326,9 +309,6 @@ static int enetc_vlan_rx_del_vid(struct net_device *ndev, __be16 prot, u16 vid) __clear_bit(vid, pf->active_vlans); enetc_sync_vlan_ht_filter(pf, true); - if (!enetc_vlan_filter_is_on(pf)) - enetc_enable_si_vlan_promisc(pf, 0); - return 0; } @@ -677,6 +657,15 @@ static int enetc_pf_set_features(struct net_device *ndev, enetc_enable_txvlan(&priv->si->hw, 0, !!(features & NETIF_F_HW_VLAN_CTAG_TX)); + if (changed & NETIF_F_HW_VLAN_CTAG_FILTER) { + struct enetc_pf *pf = enetc_si_priv(priv->si); + + if (!!(features & NETIF_F_HW_VLAN_CTAG_FILTER)) + enetc_disable_si_vlan_promisc(pf, 0); + else + enetc_enable_si_vlan_promisc(pf, 0); + } + if (changed & NETIF_F_LOOPBACK) enetc_set_loopback(ndev, !!(features & NETIF_F_LOOPBACK)); @@ -719,12 +708,11 @@ static void enetc_pf_netdev_setup(struct enetc_si *si, struct net_device *ndev, ndev->hw_features = NETIF_F_SG | NETIF_F_RXCSUM | NETIF_F_HW_CSUM | NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX | - NETIF_F_LOOPBACK; + NETIF_F_HW_VLAN_CTAG_FILTER | NETIF_F_LOOPBACK; ndev->features = NETIF_F_HIGHDMA | NETIF_F_SG | NETIF_F_RXCSUM | NETIF_F_HW_CSUM | NETIF_F_HW_VLAN_CTAG_TX | - NETIF_F_HW_VLAN_CTAG_RX | - NETIF_F_HW_VLAN_CTAG_FILTER; + NETIF_F_HW_VLAN_CTAG_RX; if (si->num_rss) ndev->hw_features |= NETIF_F_RXHASH; @@ -739,6 +727,12 @@ static void enetc_pf_netdev_setup(struct enetc_si *si, struct net_device *ndev, if (si->hw_features & ENETC_SI_F_QBV) priv->active_offloads |= ENETC_F_QBV; + if (si->hw_features & ENETC_SI_F_PSFP && !enetc_psfp_enable(priv)) { + priv->active_offloads |= ENETC_F_QCI; + ndev->features |= NETIF_F_HW_TC; + ndev->hw_features |= NETIF_F_HW_TC; + } + /* pick up primary MAC address from SI */ enetc_get_primary_mac_addr(&si->hw, ndev->dev_addr); } diff --git a/drivers/net/ethernet/freescale/enetc/enetc_qos.c b/drivers/net/ethernet/freescale/enetc/enetc_qos.c index 0c6bf3a55a9a..fd3df19eaa32 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_qos.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_qos.c @@ -5,6 +5,9 @@ #include <net/pkt_sched.h> #include <linux/math64.h> +#include <linux/refcount.h> +#include <net/pkt_cls.h> +#include <net/tc_act/tc_gate.h> static u16 enetc_get_max_gcl_len(struct enetc_hw *hw) { @@ -331,3 +334,1103 @@ int enetc_setup_tc_txtime(struct net_device *ndev, void *type_data) return 0; } + +enum streamid_type { + STREAMID_TYPE_RESERVED = 0, + STREAMID_TYPE_NULL, + STREAMID_TYPE_SMAC, +}; + +enum streamid_vlan_tagged { + STREAMID_VLAN_RESERVED = 0, + STREAMID_VLAN_TAGGED, + STREAMID_VLAN_UNTAGGED, + STREAMID_VLAN_ALL, +}; + +#define ENETC_PSFP_WILDCARD -1 +#define HANDLE_OFFSET 100 + +enum forward_type { + FILTER_ACTION_TYPE_PSFP = BIT(0), + FILTER_ACTION_TYPE_ACL = BIT(1), + FILTER_ACTION_TYPE_BOTH = GENMASK(1, 0), +}; + +/* This is for limit output type for input actions */ +struct actions_fwd { + u64 actions; + u64 keys; /* include the must needed keys */ + enum forward_type output; +}; + +struct psfp_streamfilter_counters { + u64 matching_frames_count; + u64 passing_frames_count; + u64 not_passing_frames_count; + u64 passing_sdu_count; + u64 not_passing_sdu_count; + u64 red_frames_count; +}; + +struct enetc_streamid { + u32 index; + union { + u8 src_mac[6]; + u8 dst_mac[6]; + }; + u8 filtertype; + u16 vid; + u8 tagged; + s32 handle; +}; + +struct enetc_psfp_filter { + u32 index; + s32 handle; + s8 prio; + u32 gate_id; + s32 meter_id; + refcount_t refcount; + struct hlist_node node; +}; + +struct enetc_psfp_gate { + u32 index; + s8 init_ipv; + u64 basetime; + u64 cycletime; + u64 cycletimext; + u32 num_entries; + refcount_t refcount; + struct hlist_node node; + struct action_gate_entry entries[0]; +}; + +struct enetc_stream_filter { + struct enetc_streamid sid; + u32 sfi_index; + u32 sgi_index; + struct flow_stats stats; + struct hlist_node node; +}; + +struct enetc_psfp { + unsigned long dev_bitmap; + unsigned long *psfp_sfi_bitmap; + struct hlist_head stream_list; + struct hlist_head psfp_filter_list; + struct hlist_head psfp_gate_list; + spinlock_t psfp_lock; /* spinlock for the struct enetc_psfp r/w */ +}; + +static struct actions_fwd enetc_act_fwd[] = { + { + BIT(FLOW_ACTION_GATE), + BIT(FLOW_DISSECTOR_KEY_ETH_ADDRS), + FILTER_ACTION_TYPE_PSFP + }, + /* example for ACL actions */ + { + BIT(FLOW_ACTION_DROP), + 0, + FILTER_ACTION_TYPE_ACL + } +}; + +static struct enetc_psfp epsfp = { + .psfp_sfi_bitmap = NULL, +}; + +static LIST_HEAD(enetc_block_cb_list); + +static inline int enetc_get_port(struct enetc_ndev_priv *priv) +{ + return priv->si->pdev->devfn & 0x7; +} + +/* Stream Identity Entry Set Descriptor */ +static int enetc_streamid_hw_set(struct enetc_ndev_priv *priv, + struct enetc_streamid *sid, + u8 enable) +{ + struct enetc_cbd cbd = {.cmd = 0}; + struct streamid_data *si_data; + struct streamid_conf *si_conf; + u16 data_size; + dma_addr_t dma; + int err; + + if (sid->index >= priv->psfp_cap.max_streamid) + return -EINVAL; + + if (sid->filtertype != STREAMID_TYPE_NULL && + sid->filtertype != STREAMID_TYPE_SMAC) + return -EOPNOTSUPP; + + /* Disable operation before enable */ + cbd.index = cpu_to_le16((u16)sid->index); + cbd.cls = BDCR_CMD_STREAM_IDENTIFY; + cbd.status_flags = 0; + + data_size = sizeof(struct streamid_data); + si_data = kzalloc(data_size, __GFP_DMA | GFP_KERNEL); + cbd.length = cpu_to_le16(data_size); + + dma = dma_map_single(&priv->si->pdev->dev, si_data, + data_size, DMA_FROM_DEVICE); + if (dma_mapping_error(&priv->si->pdev->dev, dma)) { + netdev_err(priv->si->ndev, "DMA mapping failed!\n"); + kfree(si_data); + return -ENOMEM; + } + + cbd.addr[0] = lower_32_bits(dma); + cbd.addr[1] = upper_32_bits(dma); + memset(si_data->dmac, 0xff, ETH_ALEN); + si_data->vid_vidm_tg = + cpu_to_le16(ENETC_CBDR_SID_VID_MASK + + ((0x3 << 14) | ENETC_CBDR_SID_VIDM)); + + si_conf = &cbd.sid_set; + /* Only one port supported for one entry, set itself */ + si_conf->iports = 1 << enetc_get_port(priv); + si_conf->id_type = 1; + si_conf->oui[2] = 0x0; + si_conf->oui[1] = 0x80; + si_conf->oui[0] = 0xC2; + + err = enetc_send_cmd(priv->si, &cbd); + if (err) + return -EINVAL; + + if (!enable) { + kfree(si_data); + return 0; + } + + /* Enable the entry overwrite again incase space flushed by hardware */ + memset(&cbd, 0, sizeof(cbd)); + + cbd.index = cpu_to_le16((u16)sid->index); + cbd.cmd = 0; + cbd.cls = BDCR_CMD_STREAM_IDENTIFY; + cbd.status_flags = 0; + + si_conf->en = 0x80; + si_conf->stream_handle = cpu_to_le32(sid->handle); + si_conf->iports = 1 << enetc_get_port(priv); + si_conf->id_type = sid->filtertype; + si_conf->oui[2] = 0x0; + si_conf->oui[1] = 0x80; + si_conf->oui[0] = 0xC2; + + memset(si_data, 0, data_size); + + cbd.length = cpu_to_le16(data_size); + + cbd.addr[0] = lower_32_bits(dma); + cbd.addr[1] = upper_32_bits(dma); + + /* VIDM default to be 1. + * VID Match. If set (b1) then the VID must match, otherwise + * any VID is considered a match. VIDM setting is only used + * when TG is set to b01. + */ + if (si_conf->id_type == STREAMID_TYPE_NULL) { + ether_addr_copy(si_data->dmac, sid->dst_mac); + si_data->vid_vidm_tg = + cpu_to_le16((sid->vid & ENETC_CBDR_SID_VID_MASK) + + ((((u16)(sid->tagged) & 0x3) << 14) + | ENETC_CBDR_SID_VIDM)); + } else if (si_conf->id_type == STREAMID_TYPE_SMAC) { + ether_addr_copy(si_data->smac, sid->src_mac); + si_data->vid_vidm_tg = + cpu_to_le16((sid->vid & ENETC_CBDR_SID_VID_MASK) + + ((((u16)(sid->tagged) & 0x3) << 14) + | ENETC_CBDR_SID_VIDM)); + } + + err = enetc_send_cmd(priv->si, &cbd); + kfree(si_data); + + return err; +} + +/* Stream Filter Instance Set Descriptor */ +static int enetc_streamfilter_hw_set(struct enetc_ndev_priv *priv, + struct enetc_psfp_filter *sfi, + u8 enable) +{ + struct enetc_cbd cbd = {.cmd = 0}; + struct sfi_conf *sfi_config; + + cbd.index = cpu_to_le16(sfi->index); + cbd.cls = BDCR_CMD_STREAM_FILTER; + cbd.status_flags = 0x80; + cbd.length = cpu_to_le16(1); + + sfi_config = &cbd.sfi_conf; + if (!enable) + goto exit; + + sfi_config->en = 0x80; + + if (sfi->handle >= 0) { + sfi_config->stream_handle = + cpu_to_le32(sfi->handle); + sfi_config->sthm |= 0x80; + } + + sfi_config->sg_inst_table_index = cpu_to_le16(sfi->gate_id); + sfi_config->input_ports = 1 << enetc_get_port(priv); + + /* The priority value which may be matched against the + * frame’s priority value to determine a match for this entry. + */ + if (sfi->prio >= 0) + sfi_config->multi |= (sfi->prio & 0x7) | 0x8; + + /* Filter Type. Identifies the contents of the MSDU/FM_INST_INDEX + * field as being either an MSDU value or an index into the Flow + * Meter Instance table. + * TODO: no limit max sdu + */ + + if (sfi->meter_id >= 0) { + sfi_config->fm_inst_table_index = cpu_to_le16(sfi->meter_id); + sfi_config->multi |= 0x80; + } + +exit: + return enetc_send_cmd(priv->si, &cbd); +} + +static int enetc_streamcounter_hw_get(struct enetc_ndev_priv *priv, + u32 index, + struct psfp_streamfilter_counters *cnt) +{ + struct enetc_cbd cbd = { .cmd = 2 }; + struct sfi_counter_data *data_buf; + dma_addr_t dma; + u16 data_size; + int err; + + cbd.index = cpu_to_le16((u16)index); + cbd.cmd = 2; + cbd.cls = BDCR_CMD_STREAM_FILTER; + cbd.status_flags = 0; + + data_size = sizeof(struct sfi_counter_data); + data_buf = kzalloc(data_size, __GFP_DMA | GFP_KERNEL); + if (!data_buf) + return -ENOMEM; + + dma = dma_map_single(&priv->si->pdev->dev, data_buf, + data_size, DMA_FROM_DEVICE); + if (dma_mapping_error(&priv->si->pdev->dev, dma)) { + netdev_err(priv->si->ndev, "DMA mapping failed!\n"); + err = -ENOMEM; + goto exit; + } + cbd.addr[0] = lower_32_bits(dma); + cbd.addr[1] = upper_32_bits(dma); + + cbd.length = cpu_to_le16(data_size); + + err = enetc_send_cmd(priv->si, &cbd); + if (err) + goto exit; + + cnt->matching_frames_count = + ((u64)le32_to_cpu(data_buf->matchh) << 32) + + data_buf->matchl; + + cnt->not_passing_sdu_count = + ((u64)le32_to_cpu(data_buf->msdu_droph) << 32) + + data_buf->msdu_dropl; + + cnt->passing_sdu_count = cnt->matching_frames_count + - cnt->not_passing_sdu_count; + + cnt->not_passing_frames_count = + ((u64)le32_to_cpu(data_buf->stream_gate_droph) << 32) + + le32_to_cpu(data_buf->stream_gate_dropl); + + cnt->passing_frames_count = cnt->matching_frames_count + - cnt->not_passing_sdu_count + - cnt->not_passing_frames_count; + + cnt->red_frames_count = + ((u64)le32_to_cpu(data_buf->flow_meter_droph) << 32) + + le32_to_cpu(data_buf->flow_meter_dropl); + +exit: + kfree(data_buf); + return err; +} + +static u64 get_ptp_now(struct enetc_hw *hw) +{ + u64 now_lo, now_hi, now; + + now_lo = enetc_rd(hw, ENETC_SICTR0); + now_hi = enetc_rd(hw, ENETC_SICTR1); + now = now_lo | now_hi << 32; + + return now; +} + +static int get_start_ns(u64 now, u64 cycle, u64 *start) +{ + u64 n; + + if (!cycle) + return -EFAULT; + + n = div64_u64(now, cycle); + + *start = (n + 1) * cycle; + + return 0; +} + +/* Stream Gate Instance Set Descriptor */ +static int enetc_streamgate_hw_set(struct enetc_ndev_priv *priv, + struct enetc_psfp_gate *sgi, + u8 enable) +{ + struct enetc_cbd cbd = { .cmd = 0 }; + struct sgi_table *sgi_config; + struct sgcl_conf *sgcl_config; + struct sgcl_data *sgcl_data; + struct sgce *sgce; + dma_addr_t dma; + u16 data_size; + int err, i; + u64 now; + + cbd.index = cpu_to_le16(sgi->index); + cbd.cmd = 0; + cbd.cls = BDCR_CMD_STREAM_GCL; + cbd.status_flags = 0x80; + + /* disable */ + if (!enable) + return enetc_send_cmd(priv->si, &cbd); + + if (!sgi->num_entries) + return 0; + + if (sgi->num_entries > priv->psfp_cap.max_psfp_gatelist || + !sgi->cycletime) + return -EINVAL; + + /* enable */ + sgi_config = &cbd.sgi_table; + + /* Keep open before gate list start */ + sgi_config->ocgtst = 0x80; + + sgi_config->oipv = (sgi->init_ipv < 0) ? + 0x0 : ((sgi->init_ipv & 0x7) | 0x8); + + sgi_config->en = 0x80; + + /* Basic config */ + err = enetc_send_cmd(priv->si, &cbd); + if (err) + return -EINVAL; + + memset(&cbd, 0, sizeof(cbd)); + + cbd.index = cpu_to_le16(sgi->index); + cbd.cmd = 1; + cbd.cls = BDCR_CMD_STREAM_GCL; + cbd.status_flags = 0; + + sgcl_config = &cbd.sgcl_conf; + + sgcl_config->acl_len = (sgi->num_entries - 1) & 0x3; + + data_size = struct_size(sgcl_data, sgcl, sgi->num_entries); + + sgcl_data = kzalloc(data_size, __GFP_DMA | GFP_KERNEL); + if (!sgcl_data) + return -ENOMEM; + + cbd.length = cpu_to_le16(data_size); + + dma = dma_map_single(&priv->si->pdev->dev, + sgcl_data, data_size, + DMA_FROM_DEVICE); + if (dma_mapping_error(&priv->si->pdev->dev, dma)) { + netdev_err(priv->si->ndev, "DMA mapping failed!\n"); + kfree(sgcl_data); + return -ENOMEM; + } + + cbd.addr[0] = lower_32_bits(dma); + cbd.addr[1] = upper_32_bits(dma); + + sgce = &sgcl_data->sgcl[0]; + + sgcl_config->agtst = 0x80; + + sgcl_data->ct = cpu_to_le32(sgi->cycletime); + sgcl_data->cte = cpu_to_le32(sgi->cycletimext); + + if (sgi->init_ipv >= 0) + sgcl_config->aipv = (sgi->init_ipv & 0x7) | 0x8; + + for (i = 0; i < sgi->num_entries; i++) { + struct action_gate_entry *from = &sgi->entries[i]; + struct sgce *to = &sgce[i]; + + if (from->gate_state) + to->multi |= 0x10; + + if (from->ipv >= 0) + to->multi |= ((from->ipv & 0x7) << 5) | 0x08; + + if (from->maxoctets >= 0) { + to->multi |= 0x01; + to->msdu[0] = from->maxoctets & 0xFF; + to->msdu[1] = (from->maxoctets >> 8) & 0xFF; + to->msdu[2] = (from->maxoctets >> 16) & 0xFF; + } + + to->interval = cpu_to_le32(from->interval); + } + + /* If basetime is less than now, calculate start time */ + now = get_ptp_now(&priv->si->hw); + + if (sgi->basetime < now) { + u64 start; + + err = get_start_ns(now, sgi->cycletime, &start); + if (err) + goto exit; + sgcl_data->btl = cpu_to_le32(lower_32_bits(start)); + sgcl_data->bth = cpu_to_le32(upper_32_bits(start)); + } else { + u32 hi, lo; + + hi = upper_32_bits(sgi->basetime); + lo = lower_32_bits(sgi->basetime); + sgcl_data->bth = cpu_to_le32(hi); + sgcl_data->btl = cpu_to_le32(lo); + } + + err = enetc_send_cmd(priv->si, &cbd); + +exit: + kfree(sgcl_data); + + return err; +} + +static struct enetc_stream_filter *enetc_get_stream_by_index(u32 index) +{ + struct enetc_stream_filter *f; + + hlist_for_each_entry(f, &epsfp.stream_list, node) + if (f->sid.index == index) + return f; + + return NULL; +} + +static struct enetc_psfp_gate *enetc_get_gate_by_index(u32 index) +{ + struct enetc_psfp_gate *g; + + hlist_for_each_entry(g, &epsfp.psfp_gate_list, node) + if (g->index == index) + return g; + + return NULL; +} + +static struct enetc_psfp_filter *enetc_get_filter_by_index(u32 index) +{ + struct enetc_psfp_filter *s; + + hlist_for_each_entry(s, &epsfp.psfp_filter_list, node) + if (s->index == index) + return s; + + return NULL; +} + +static struct enetc_psfp_filter + *enetc_psfp_check_sfi(struct enetc_psfp_filter *sfi) +{ + struct enetc_psfp_filter *s; + + hlist_for_each_entry(s, &epsfp.psfp_filter_list, node) + if (s->gate_id == sfi->gate_id && + s->prio == sfi->prio && + s->meter_id == sfi->meter_id) + return s; + + return NULL; +} + +static int enetc_get_free_index(struct enetc_ndev_priv *priv) +{ + u32 max_size = priv->psfp_cap.max_psfp_filter; + unsigned long index; + + index = find_first_zero_bit(epsfp.psfp_sfi_bitmap, max_size); + if (index == max_size) + return -1; + + return index; +} + +static void stream_filter_unref(struct enetc_ndev_priv *priv, u32 index) +{ + struct enetc_psfp_filter *sfi; + u8 z; + + sfi = enetc_get_filter_by_index(index); + WARN_ON(!sfi); + z = refcount_dec_and_test(&sfi->refcount); + + if (z) { + enetc_streamfilter_hw_set(priv, sfi, false); + hlist_del(&sfi->node); + kfree(sfi); + clear_bit(index, epsfp.psfp_sfi_bitmap); + } +} + +static void stream_gate_unref(struct enetc_ndev_priv *priv, u32 index) +{ + struct enetc_psfp_gate *sgi; + u8 z; + + sgi = enetc_get_gate_by_index(index); + WARN_ON(!sgi); + z = refcount_dec_and_test(&sgi->refcount); + if (z) { + enetc_streamgate_hw_set(priv, sgi, false); + hlist_del(&sgi->node); + kfree(sgi); + } +} + +static void remove_one_chain(struct enetc_ndev_priv *priv, + struct enetc_stream_filter *filter) +{ + stream_gate_unref(priv, filter->sgi_index); + stream_filter_unref(priv, filter->sfi_index); + + hlist_del(&filter->node); + kfree(filter); +} + +static int enetc_psfp_hw_set(struct enetc_ndev_priv *priv, + struct enetc_streamid *sid, + struct enetc_psfp_filter *sfi, + struct enetc_psfp_gate *sgi) +{ + int err; + + err = enetc_streamid_hw_set(priv, sid, true); + if (err) + return err; + + if (sfi) { + err = enetc_streamfilter_hw_set(priv, sfi, true); + if (err) + goto revert_sid; + } + + err = enetc_streamgate_hw_set(priv, sgi, true); + if (err) + goto revert_sfi; + + return 0; + +revert_sfi: + if (sfi) + enetc_streamfilter_hw_set(priv, sfi, false); +revert_sid: + enetc_streamid_hw_set(priv, sid, false); + return err; +} + +static struct actions_fwd *enetc_check_flow_actions(u64 acts, + unsigned int inputkeys) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(enetc_act_fwd); i++) + if (acts == enetc_act_fwd[i].actions && + inputkeys & enetc_act_fwd[i].keys) + return &enetc_act_fwd[i]; + + return NULL; +} + +static int enetc_psfp_parse_clsflower(struct enetc_ndev_priv *priv, + struct flow_cls_offload *f) +{ + struct flow_rule *rule = flow_cls_offload_flow_rule(f); + struct netlink_ext_ack *extack = f->common.extack; + struct enetc_stream_filter *filter, *old_filter; + struct enetc_psfp_filter *sfi, *old_sfi; + struct enetc_psfp_gate *sgi, *old_sgi; + struct flow_action_entry *entry; + struct action_gate_entry *e; + u8 sfi_overwrite = 0; + int entries_size; + int i, err; + + if (f->common.chain_index >= priv->psfp_cap.max_streamid) { + NL_SET_ERR_MSG_MOD(extack, "No Stream identify resource!"); + return -ENOSPC; + } + + flow_action_for_each(i, entry, &rule->action) + if (entry->id == FLOW_ACTION_GATE) + break; + + if (entry->id != FLOW_ACTION_GATE) + return -EINVAL; + + filter = kzalloc(sizeof(*filter), GFP_KERNEL); + if (!filter) + return -ENOMEM; + + filter->sid.index = f->common.chain_index; + + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ETH_ADDRS)) { + struct flow_match_eth_addrs match; + + flow_rule_match_eth_addrs(rule, &match); + + if (!is_zero_ether_addr(match.mask->dst) && + !is_zero_ether_addr(match.mask->src)) { + NL_SET_ERR_MSG_MOD(extack, + "Cannot match on both source and destination MAC"); + err = EINVAL; + goto free_filter; + } + + if (!is_zero_ether_addr(match.mask->dst)) { + if (!is_broadcast_ether_addr(match.mask->dst)) { + NL_SET_ERR_MSG_MOD(extack, + "Masked matching on destination MAC not supported"); + err = EINVAL; + goto free_filter; + } + ether_addr_copy(filter->sid.dst_mac, match.key->dst); + filter->sid.filtertype = STREAMID_TYPE_NULL; + } + + if (!is_zero_ether_addr(match.mask->src)) { + if (!is_broadcast_ether_addr(match.mask->src)) { + NL_SET_ERR_MSG_MOD(extack, + "Masked matching on source MAC not supported"); + err = EINVAL; + goto free_filter; + } + ether_addr_copy(filter->sid.src_mac, match.key->src); + filter->sid.filtertype = STREAMID_TYPE_SMAC; + } + } else { + NL_SET_ERR_MSG_MOD(extack, "Unsupported, must include ETH_ADDRS"); + err = EINVAL; + goto free_filter; + } + + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) { + struct flow_match_vlan match; + + flow_rule_match_vlan(rule, &match); + if (match.mask->vlan_priority) { + if (match.mask->vlan_priority != + (VLAN_PRIO_MASK >> VLAN_PRIO_SHIFT)) { + NL_SET_ERR_MSG_MOD(extack, "Only full mask is supported for VLAN priority"); + err = -EINVAL; + goto free_filter; + } + } + + if (match.mask->vlan_id) { + if (match.mask->vlan_id != VLAN_VID_MASK) { + NL_SET_ERR_MSG_MOD(extack, "Only full mask is supported for VLAN id"); + err = -EINVAL; + goto free_filter; + } + + filter->sid.vid = match.key->vlan_id; + if (!filter->sid.vid) + filter->sid.tagged = STREAMID_VLAN_UNTAGGED; + else + filter->sid.tagged = STREAMID_VLAN_TAGGED; + } + } else { + filter->sid.tagged = STREAMID_VLAN_ALL; + } + + /* parsing gate action */ + if (entry->gate.index >= priv->psfp_cap.max_psfp_gate) { + NL_SET_ERR_MSG_MOD(extack, "No Stream Gate resource!"); + err = -ENOSPC; + goto free_filter; + } + + if (entry->gate.num_entries >= priv->psfp_cap.max_psfp_gatelist) { + NL_SET_ERR_MSG_MOD(extack, "No Stream Gate resource!"); + err = -ENOSPC; + goto free_filter; + } + + entries_size = struct_size(sgi, entries, entry->gate.num_entries); + sgi = kzalloc(entries_size, GFP_KERNEL); + if (!sgi) { + err = -ENOMEM; + goto free_filter; + } + + refcount_set(&sgi->refcount, 1); + sgi->index = entry->gate.index; + sgi->init_ipv = entry->gate.prio; + sgi->basetime = entry->gate.basetime; + sgi->cycletime = entry->gate.cycletime; + sgi->num_entries = entry->gate.num_entries; + + e = sgi->entries; + for (i = 0; i < entry->gate.num_entries; i++) { + e[i].gate_state = entry->gate.entries[i].gate_state; + e[i].interval = entry->gate.entries[i].interval; + e[i].ipv = entry->gate.entries[i].ipv; + e[i].maxoctets = entry->gate.entries[i].maxoctets; + } + + filter->sgi_index = sgi->index; + + sfi = kzalloc(sizeof(*sfi), GFP_KERNEL); + if (!sfi) { + err = -ENOMEM; + goto free_gate; + } + + refcount_set(&sfi->refcount, 1); + sfi->gate_id = sgi->index; + + /* flow meter not support yet */ + sfi->meter_id = ENETC_PSFP_WILDCARD; + + /* prio ref the filter prio */ + if (f->common.prio && f->common.prio <= BIT(3)) + sfi->prio = f->common.prio - 1; + else + sfi->prio = ENETC_PSFP_WILDCARD; + + old_sfi = enetc_psfp_check_sfi(sfi); + if (!old_sfi) { + int index; + + index = enetc_get_free_index(priv); + if (sfi->handle < 0) { + NL_SET_ERR_MSG_MOD(extack, "No Stream Filter resource!"); + err = -ENOSPC; + goto free_sfi; + } + + sfi->index = index; + sfi->handle = index + HANDLE_OFFSET; + /* Update the stream filter handle also */ + filter->sid.handle = sfi->handle; + filter->sfi_index = sfi->index; + sfi_overwrite = 0; + } else { + filter->sfi_index = old_sfi->index; + filter->sid.handle = old_sfi->handle; + sfi_overwrite = 1; + } + + err = enetc_psfp_hw_set(priv, &filter->sid, + sfi_overwrite ? NULL : sfi, sgi); + if (err) + goto free_sfi; + + spin_lock(&epsfp.psfp_lock); + /* Remove the old node if exist and update with a new node */ + old_sgi = enetc_get_gate_by_index(filter->sgi_index); + if (old_sgi) { + refcount_set(&sgi->refcount, + refcount_read(&old_sgi->refcount) + 1); + hlist_del(&old_sgi->node); + kfree(old_sgi); + } + + hlist_add_head(&sgi->node, &epsfp.psfp_gate_list); + + if (!old_sfi) { + hlist_add_head(&sfi->node, &epsfp.psfp_filter_list); + set_bit(sfi->index, epsfp.psfp_sfi_bitmap); + } else { + kfree(sfi); + refcount_inc(&old_sfi->refcount); + } + + old_filter = enetc_get_stream_by_index(filter->sid.index); + if (old_filter) + remove_one_chain(priv, old_filter); + + filter->stats.lastused = jiffies; + hlist_add_head(&filter->node, &epsfp.stream_list); + + spin_unlock(&epsfp.psfp_lock); + + return 0; + +free_sfi: + kfree(sfi); +free_gate: + kfree(sgi); +free_filter: + kfree(filter); + + return err; +} + +static int enetc_config_clsflower(struct enetc_ndev_priv *priv, + struct flow_cls_offload *cls_flower) +{ + struct flow_rule *rule = flow_cls_offload_flow_rule(cls_flower); + struct netlink_ext_ack *extack = cls_flower->common.extack; + struct flow_dissector *dissector = rule->match.dissector; + struct flow_action *action = &rule->action; + struct flow_action_entry *entry; + struct actions_fwd *fwd; + u64 actions = 0; + int i, err; + + if (!flow_action_has_entries(action)) { + NL_SET_ERR_MSG_MOD(extack, "At least one action is needed"); + return -EINVAL; + } + + flow_action_for_each(i, entry, action) + actions |= BIT(entry->id); + + fwd = enetc_check_flow_actions(actions, dissector->used_keys); + if (!fwd) { + NL_SET_ERR_MSG_MOD(extack, "Unsupported filter type!"); + return -EOPNOTSUPP; + } + + if (fwd->output & FILTER_ACTION_TYPE_PSFP) { + err = enetc_psfp_parse_clsflower(priv, cls_flower); + if (err) { + NL_SET_ERR_MSG_MOD(extack, "Invalid PSFP inputs"); + return err; + } + } else { + NL_SET_ERR_MSG_MOD(extack, "Unsupported actions"); + return -EOPNOTSUPP; + } + + return 0; +} + +static int enetc_psfp_destroy_clsflower(struct enetc_ndev_priv *priv, + struct flow_cls_offload *f) +{ + struct enetc_stream_filter *filter; + struct netlink_ext_ack *extack = f->common.extack; + int err; + + if (f->common.chain_index >= priv->psfp_cap.max_streamid) { + NL_SET_ERR_MSG_MOD(extack, "No Stream identify resource!"); + return -ENOSPC; + } + + filter = enetc_get_stream_by_index(f->common.chain_index); + if (!filter) + return -EINVAL; + + err = enetc_streamid_hw_set(priv, &filter->sid, false); + if (err) + return err; + + remove_one_chain(priv, filter); + + return 0; +} + +static int enetc_destroy_clsflower(struct enetc_ndev_priv *priv, + struct flow_cls_offload *f) +{ + return enetc_psfp_destroy_clsflower(priv, f); +} + +static int enetc_psfp_get_stats(struct enetc_ndev_priv *priv, + struct flow_cls_offload *f) +{ + struct psfp_streamfilter_counters counters = {}; + struct enetc_stream_filter *filter; + struct flow_stats stats = {}; + int err; + + filter = enetc_get_stream_by_index(f->common.chain_index); + if (!filter) + return -EINVAL; + + err = enetc_streamcounter_hw_get(priv, filter->sfi_index, &counters); + if (err) + return -EINVAL; + + spin_lock(&epsfp.psfp_lock); + stats.pkts = counters.matching_frames_count - filter->stats.pkts; + stats.lastused = filter->stats.lastused; + filter->stats.pkts += stats.pkts; + spin_unlock(&epsfp.psfp_lock); + + flow_stats_update(&f->stats, 0x0, stats.pkts, stats.lastused, + FLOW_ACTION_HW_STATS_DELAYED); + + return 0; +} + +static int enetc_setup_tc_cls_flower(struct enetc_ndev_priv *priv, + struct flow_cls_offload *cls_flower) +{ + switch (cls_flower->command) { + case FLOW_CLS_REPLACE: + return enetc_config_clsflower(priv, cls_flower); + case FLOW_CLS_DESTROY: + return enetc_destroy_clsflower(priv, cls_flower); + case FLOW_CLS_STATS: + return enetc_psfp_get_stats(priv, cls_flower); + default: + return -EOPNOTSUPP; + } +} + +static inline void clean_psfp_sfi_bitmap(void) +{ + bitmap_free(epsfp.psfp_sfi_bitmap); + epsfp.psfp_sfi_bitmap = NULL; +} + +static void clean_stream_list(void) +{ + struct enetc_stream_filter *s; + struct hlist_node *tmp; + + hlist_for_each_entry_safe(s, tmp, &epsfp.stream_list, node) { + hlist_del(&s->node); + kfree(s); + } +} + +static void clean_sfi_list(void) +{ + struct enetc_psfp_filter *sfi; + struct hlist_node *tmp; + + hlist_for_each_entry_safe(sfi, tmp, &epsfp.psfp_filter_list, node) { + hlist_del(&sfi->node); + kfree(sfi); + } +} + +static void clean_sgi_list(void) +{ + struct enetc_psfp_gate *sgi; + struct hlist_node *tmp; + + hlist_for_each_entry_safe(sgi, tmp, &epsfp.psfp_gate_list, node) { + hlist_del(&sgi->node); + kfree(sgi); + } +} + +static void clean_psfp_all(void) +{ + /* Disable all list nodes and free all memory */ + clean_sfi_list(); + clean_sgi_list(); + clean_stream_list(); + epsfp.dev_bitmap = 0; + clean_psfp_sfi_bitmap(); +} + +int enetc_setup_tc_block_cb(enum tc_setup_type type, void *type_data, + void *cb_priv) +{ + struct net_device *ndev = cb_priv; + + if (!tc_can_offload(ndev)) + return -EOPNOTSUPP; + + switch (type) { + case TC_SETUP_CLSFLOWER: + return enetc_setup_tc_cls_flower(netdev_priv(ndev), type_data); + default: + return -EOPNOTSUPP; + } +} + +int enetc_psfp_init(struct enetc_ndev_priv *priv) +{ + if (epsfp.psfp_sfi_bitmap) + return 0; + + epsfp.psfp_sfi_bitmap = bitmap_zalloc(priv->psfp_cap.max_psfp_filter, + GFP_KERNEL); + if (!epsfp.psfp_sfi_bitmap) + return -ENOMEM; + + spin_lock_init(&epsfp.psfp_lock); + + if (list_empty(&enetc_block_cb_list)) + epsfp.dev_bitmap = 0; + + return 0; +} + +int enetc_psfp_clean(struct enetc_ndev_priv *priv) +{ + if (!list_empty(&enetc_block_cb_list)) + return -EBUSY; + + clean_psfp_all(); + + return 0; +} + +int enetc_setup_tc_psfp(struct net_device *ndev, void *type_data) +{ + struct enetc_ndev_priv *priv = netdev_priv(ndev); + struct flow_block_offload *f = type_data; + int err; + + err = flow_block_cb_setup_simple(f, &enetc_block_cb_list, + enetc_setup_tc_block_cb, + ndev, ndev, true); + if (err) + return err; + + switch (f->command) { + case FLOW_BLOCK_BIND: + set_bit(enetc_get_port(priv), &epsfp.dev_bitmap); + break; + case FLOW_BLOCK_UNBIND: + clear_bit(enetc_get_port(priv), &epsfp.dev_bitmap); + if (!epsfp.dev_bitmap) + clean_psfp_all(); + break; + } + + return 0; +} diff --git a/drivers/net/ethernet/freescale/fec.h b/drivers/net/ethernet/freescale/fec.h index e74dd1f86bba..a6cdd5b61921 100644 --- a/drivers/net/ethernet/freescale/fec.h +++ b/drivers/net/ethernet/freescale/fec.h @@ -376,8 +376,7 @@ struct bufdesc_ex { #define FEC_ENET_TS_AVAIL ((uint)0x00010000) #define FEC_ENET_TS_TIMER ((uint)0x00008000) -#define FEC_DEFAULT_IMASK (FEC_ENET_TXF | FEC_ENET_RXF | FEC_ENET_MII) -#define FEC_NAPI_IMASK FEC_ENET_MII +#define FEC_DEFAULT_IMASK (FEC_ENET_TXF | FEC_ENET_RXF) #define FEC_RX_DISABLED_IMASK (FEC_DEFAULT_IMASK & (~FEC_ENET_RXF)) /* ENET interrupt coalescing macro define */ @@ -543,7 +542,6 @@ struct fec_enet_private { int link; int full_duplex; int speed; - struct completion mdio_done; int irq[FEC_IRQ_NUM]; bool bufdesc_ex; int pause_flag; diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index dc6f8763a5d4..2d0d313ee7c5 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -88,8 +88,6 @@ static void fec_enet_itr_coal_init(struct net_device *ndev); struct fec_devinfo { u32 quirks; - u8 stop_gpr_reg; - u8 stop_gpr_bit; }; static const struct fec_devinfo fec_imx25_info = { @@ -112,8 +110,6 @@ static const struct fec_devinfo fec_imx6q_info = { FEC_QUIRK_HAS_BUFDESC_EX | FEC_QUIRK_HAS_CSUM | FEC_QUIRK_HAS_VLAN | FEC_QUIRK_ERR006358 | FEC_QUIRK_HAS_RACC, - .stop_gpr_reg = 0x34, - .stop_gpr_bit = 27, }; static const struct fec_devinfo fec_mvf600_info = { @@ -976,8 +972,8 @@ fec_restart(struct net_device *ndev) writel((__force u32)cpu_to_be32(temp_mac[1]), fep->hwp + FEC_ADDR_HIGH); - /* Clear any outstanding interrupt. */ - writel(0xffffffff, fep->hwp + FEC_IEVENT); + /* Clear any outstanding interrupt, except MDIO. */ + writel((0xffffffff & ~FEC_ENET_MII), fep->hwp + FEC_IEVENT); fec_enet_bd_init(ndev); @@ -1123,7 +1119,7 @@ fec_restart(struct net_device *ndev) if (fep->link) writel(FEC_DEFAULT_IMASK, fep->hwp + FEC_IMASK); else - writel(FEC_ENET_MII, fep->hwp + FEC_IMASK); + writel(0, fep->hwp + FEC_IMASK); /* Init the interrupt coalescing */ fec_enet_itr_coal_init(ndev); @@ -1652,6 +1648,10 @@ fec_enet_interrupt(int irq, void *dev_id) irqreturn_t ret = IRQ_NONE; int_events = readl(fep->hwp + FEC_IEVENT); + + /* Don't clear MDIO events, we poll for those */ + int_events &= ~FEC_ENET_MII; + writel(int_events, fep->hwp + FEC_IEVENT); fec_enet_collect_events(fep, int_events); @@ -1659,16 +1659,12 @@ fec_enet_interrupt(int irq, void *dev_id) ret = IRQ_HANDLED; if (napi_schedule_prep(&fep->napi)) { - /* Disable the NAPI interrupts */ - writel(FEC_NAPI_IMASK, fep->hwp + FEC_IMASK); + /* Disable interrupts */ + writel(0, fep->hwp + FEC_IMASK); __napi_schedule(&fep->napi); } } - if (int_events & FEC_ENET_MII) { - ret = IRQ_HANDLED; - complete(&fep->mdio_done); - } return ret; } @@ -1818,11 +1814,24 @@ static void fec_enet_adjust_link(struct net_device *ndev) phy_print_status(phy_dev); } +static int fec_enet_mdio_wait(struct fec_enet_private *fep) +{ + uint ievent; + int ret; + + ret = readl_poll_timeout_atomic(fep->hwp + FEC_IEVENT, ievent, + ievent & FEC_ENET_MII, 2, 30000); + + if (!ret) + writel(FEC_ENET_MII, fep->hwp + FEC_IEVENT); + + return ret; +} + static int fec_enet_mdio_read(struct mii_bus *bus, int mii_id, int regnum) { struct fec_enet_private *fep = bus->priv; struct device *dev = &fep->pdev->dev; - unsigned long time_left; int ret = 0, frame_start, frame_addr, frame_op; bool is_c45 = !!(regnum & MII_ADDR_C45); @@ -1830,8 +1839,6 @@ static int fec_enet_mdio_read(struct mii_bus *bus, int mii_id, int regnum) if (ret < 0) return ret; - reinit_completion(&fep->mdio_done); - if (is_c45) { frame_start = FEC_MMFR_ST_C45; @@ -1843,11 +1850,9 @@ static int fec_enet_mdio_read(struct mii_bus *bus, int mii_id, int regnum) fep->hwp + FEC_MII_DATA); /* wait for end of transfer */ - time_left = wait_for_completion_timeout(&fep->mdio_done, - usecs_to_jiffies(FEC_MII_TIMEOUT)); - if (time_left == 0) { + ret = fec_enet_mdio_wait(fep); + if (ret) { netdev_err(fep->netdev, "MDIO address write timeout\n"); - ret = -ETIMEDOUT; goto out; } @@ -1866,11 +1871,9 @@ static int fec_enet_mdio_read(struct mii_bus *bus, int mii_id, int regnum) FEC_MMFR_TA, fep->hwp + FEC_MII_DATA); /* wait for end of transfer */ - time_left = wait_for_completion_timeout(&fep->mdio_done, - usecs_to_jiffies(FEC_MII_TIMEOUT)); - if (time_left == 0) { + ret = fec_enet_mdio_wait(fep); + if (ret) { netdev_err(fep->netdev, "MDIO read timeout\n"); - ret = -ETIMEDOUT; goto out; } @@ -1888,7 +1891,6 @@ static int fec_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, { struct fec_enet_private *fep = bus->priv; struct device *dev = &fep->pdev->dev; - unsigned long time_left; int ret, frame_start, frame_addr; bool is_c45 = !!(regnum & MII_ADDR_C45); @@ -1898,8 +1900,6 @@ static int fec_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, else ret = 0; - reinit_completion(&fep->mdio_done); - if (is_c45) { frame_start = FEC_MMFR_ST_C45; @@ -1911,11 +1911,9 @@ static int fec_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, fep->hwp + FEC_MII_DATA); /* wait for end of transfer */ - time_left = wait_for_completion_timeout(&fep->mdio_done, - usecs_to_jiffies(FEC_MII_TIMEOUT)); - if (time_left == 0) { + ret = fec_enet_mdio_wait(fep); + if (ret) { netdev_err(fep->netdev, "MDIO address write timeout\n"); - ret = -ETIMEDOUT; goto out; } } else { @@ -1931,12 +1929,9 @@ static int fec_enet_mdio_write(struct mii_bus *bus, int mii_id, int regnum, fep->hwp + FEC_MII_DATA); /* wait for end of transfer */ - time_left = wait_for_completion_timeout(&fep->mdio_done, - usecs_to_jiffies(FEC_MII_TIMEOUT)); - if (time_left == 0) { + ret = fec_enet_mdio_wait(fep); + if (ret) netdev_err(fep->netdev, "MDIO write timeout\n"); - ret = -ETIMEDOUT; - } out: pm_runtime_mark_last_busy(dev); @@ -1986,8 +1981,12 @@ static int fec_enet_clk_enable(struct net_device *ndev, bool enable) return 0; failed_clk_ref: - if (fep->clk_ref) - clk_disable_unprepare(fep->clk_ref); + if (fep->clk_ptp) { + mutex_lock(&fep->ptp_clk_mutex); + clk_disable_unprepare(fep->clk_ptp); + fep->ptp_clk_on = false; + mutex_unlock(&fep->ptp_clk_mutex); + } failed_clk_ptp: if (fep->clk_enet_out) clk_disable_unprepare(fep->clk_enet_out); @@ -2065,9 +2064,11 @@ static int fec_enet_mii_init(struct platform_device *pdev) static struct mii_bus *fec0_mii_bus; struct net_device *ndev = platform_get_drvdata(pdev); struct fec_enet_private *fep = netdev_priv(ndev); + bool suppress_preamble = false; struct device_node *node; int err = -ENXIO; u32 mii_speed, holdtime; + u32 bus_freq; /* * The i.MX28 dual fec interfaces are not equal. @@ -2095,15 +2096,23 @@ static int fec_enet_mii_init(struct platform_device *pdev) return -ENOENT; } + bus_freq = 2500000; /* 2.5MHz by default */ + node = of_get_child_by_name(pdev->dev.of_node, "mdio"); + if (node) { + of_property_read_u32(node, "clock-frequency", &bus_freq); + suppress_preamble = of_property_read_bool(node, + "suppress-preamble"); + } + /* - * Set MII speed to 2.5 MHz (= clk_get_rate() / 2 * phy_speed) + * Set MII speed (= clk_get_rate() / 2 * phy_speed) * * The formula for FEC MDC is 'ref_freq / (MII_SPEED x 2)' while * for ENET-MAC is 'ref_freq / ((MII_SPEED + 1) x 2)'. The i.MX28 * Reference Manual has an error on this, and gets fixed on i.MX6Q * document. */ - mii_speed = DIV_ROUND_UP(clk_get_rate(fep->clk_ipg), 5000000); + mii_speed = DIV_ROUND_UP(clk_get_rate(fep->clk_ipg), bus_freq * 2); if (fep->quirks & FEC_QUIRK_ENET_MAC) mii_speed--; if (mii_speed > 63) { @@ -2130,8 +2139,24 @@ static int fec_enet_mii_init(struct platform_device *pdev) fep->phy_speed = mii_speed << 1 | holdtime << 8; + if (suppress_preamble) + fep->phy_speed |= BIT(7); + + /* Clear MMFR to avoid to generate MII event by writing MSCR. + * MII event generation condition: + * - writing MSCR: + * - mmfr[31:0]_not_zero & mscr[7:0]_is_zero & + * mscr_reg_data_in[7:0] != 0 + * - writing MMFR: + * - mscr[7:0]_not_zero + */ + writel(0, fep->hwp + FEC_MII_DATA); + writel(fep->phy_speed, fep->hwp + FEC_MII_SPEED); + /* Clear any pending transaction complete indication */ + writel(FEC_ENET_MII, fep->hwp + FEC_IEVENT); + fep->mii_bus = mdiobus_alloc(); if (fep->mii_bus == NULL) { err = -ENOMEM; @@ -2146,7 +2171,6 @@ static int fec_enet_mii_init(struct platform_device *pdev) fep->mii_bus->priv = fep; fep->mii_bus->parent = &pdev->dev; - node = of_get_child_by_name(pdev->dev.of_node, "mdio"); err = of_mdiobus_register(fep->mii_bus, node); of_node_put(node); if (err) @@ -3452,19 +3476,23 @@ static int fec_enet_get_irq_cnt(struct platform_device *pdev) } static int fec_enet_init_stop_mode(struct fec_enet_private *fep, - struct fec_devinfo *dev_info, struct device_node *np) { struct device_node *gpr_np; + u32 out_val[3]; int ret = 0; - if (!dev_info) - return 0; - - gpr_np = of_parse_phandle(np, "gpr", 0); + gpr_np = of_parse_phandle(np, "fsl,stop-mode", 0); if (!gpr_np) return 0; + ret = of_property_read_u32_array(np, "fsl,stop-mode", out_val, + ARRAY_SIZE(out_val)); + if (ret) { + dev_dbg(&fep->pdev->dev, "no stop mode property\n"); + return ret; + } + fep->stop_gpr.gpr = syscon_node_to_regmap(gpr_np); if (IS_ERR(fep->stop_gpr.gpr)) { dev_err(&fep->pdev->dev, "could not find gpr regmap\n"); @@ -3473,8 +3501,8 @@ static int fec_enet_init_stop_mode(struct fec_enet_private *fep, goto out; } - fep->stop_gpr.reg = dev_info->stop_gpr_reg; - fep->stop_gpr.bit = dev_info->stop_gpr_bit; + fep->stop_gpr.reg = out_val[1]; + fep->stop_gpr.bit = out_val[2]; out: of_node_put(gpr_np); @@ -3551,7 +3579,7 @@ fec_probe(struct platform_device *pdev) if (of_get_property(np, "fsl,magic-packet", NULL)) fep->wol_flag |= FEC_WOL_HAS_MAGIC_PACKET; - ret = fec_enet_init_stop_mode(fep, dev_info, np); + ret = fec_enet_init_stop_mode(fep, np); if (ret) goto failed_stop_mode; @@ -3674,7 +3702,6 @@ fec_probe(struct platform_device *pdev) fep->irq[i] = irq; } - init_completion(&fep->mdio_done); ret = fec_enet_mii_init(pdev); if (ret) goto failed_mii_init; diff --git a/drivers/net/ethernet/freescale/ucc_geth.c b/drivers/net/ethernet/freescale/ucc_geth.c index 6e5f6dd169b5..552e7554a9f8 100644 --- a/drivers/net/ethernet/freescale/ucc_geth.c +++ b/drivers/net/ethernet/freescale/ucc_geth.c @@ -42,6 +42,7 @@ #include <soc/fsl/qe/ucc.h> #include <soc/fsl/qe/ucc_fast.h> #include <asm/machdep.h> +#include <net/sch_generic.h> #include "ucc_geth.h" @@ -1548,11 +1549,8 @@ static int ugeth_disable(struct ucc_geth_private *ugeth, enum comm_dir mode) static void ugeth_quiesce(struct ucc_geth_private *ugeth) { - /* Prevent any further xmits, plus detach the device. */ - netif_device_detach(ugeth->ndev); - - /* Wait for any current xmits to finish. */ - netif_tx_disable(ugeth->ndev); + /* Prevent any further xmits */ + netif_tx_stop_all_queues(ugeth->ndev); /* Disable the interrupt to avoid NAPI rescheduling. */ disable_irq(ugeth->ug_info->uf_info.irq); @@ -1565,7 +1563,10 @@ static void ugeth_activate(struct ucc_geth_private *ugeth) { napi_enable(&ugeth->napi); enable_irq(ugeth->ug_info->uf_info.irq); - netif_device_attach(ugeth->ndev); + + /* allow to xmit again */ + netif_tx_wake_all_queues(ugeth->ndev); + __netdev_watchdog_up(ugeth->ndev); } /* Called every time the controller might need to be made diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c index 8aace2de0cc9..9a907947ba19 100644 --- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c +++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_mac.c @@ -697,9 +697,9 @@ hns_mac_register_phydev(struct mii_bus *mdio, struct hns_mac_cb *mac_cb, return rc; if (!strcmp(phy_type, phy_modes(PHY_INTERFACE_MODE_XGMII))) - is_c45 = 1; + is_c45 = true; else if (!strcmp(phy_type, phy_modes(PHY_INTERFACE_MODE_SGMII))) - is_c45 = 0; + is_c45 = false; else return -ENODATA; diff --git a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h index 948e67ef30fd..1ffe8fac702d 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h +++ b/drivers/net/ethernet/hisilicon/hns3/hclge_mbx.h @@ -24,7 +24,7 @@ enum HCLGE_MBX_OPCODE { HCLGE_MBX_GET_RETA, /* (VF -> PF) get RETA */ HCLGE_MBX_GET_RSS_KEY, /* (VF -> PF) get RSS key */ HCLGE_MBX_GET_MAC_ADDR, /* (VF -> PF) get MAC addr */ - HCLGE_MBX_PF_VF_RESP, /* (PF -> VF) generate respone to VF */ + HCLGE_MBX_PF_VF_RESP, /* (PF -> VF) generate response to VF */ HCLGE_MBX_GET_BDNUM, /* (VF -> PF) get BD num */ HCLGE_MBX_GET_BUFSIZE, /* (VF -> PF) get buffer size */ HCLGE_MBX_GET_STREAMID, /* (VF -> PF) get stream id */ @@ -45,6 +45,7 @@ enum HCLGE_MBX_OPCODE { HCLGE_MBX_GET_MEDIA_TYPE, /* (VF -> PF) get media type */ HCLGE_MBX_PUSH_PROMISC_INFO, /* (PF -> VF) push vf promisc info */ HCLGE_MBX_VF_UNINIT, /* (VF -> PF) vf is unintializing */ + HCLGE_MBX_HANDLE_VF_TBL, /* (VF -> PF) store/clear hw table */ HCLGE_MBX_GET_VF_FLR_STATUS = 200, /* (M7 -> PF) get vf flr status */ HCLGE_MBX_PUSH_LINK_STATUS, /* (M7 -> PF) get port link status */ @@ -70,6 +71,10 @@ enum hclge_mbx_vlan_cfg_subcode { HCLGE_MBX_GET_PORT_BASE_VLAN_STATE, /* get port based vlan state */ }; +enum hclge_mbx_tbl_cfg_subcode { + HCLGE_MBX_VPORT_LIST_CLEAR, +}; + #define HCLGE_MBX_MAX_MSG_SIZE 14 #define HCLGE_MBX_MAX_RESP_DATA_SIZE 8U #define HCLGE_MBX_MAX_RING_CHAIN_PARAM_NUM 4 diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h index 5587605d6deb..d041cac9a487 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h +++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h @@ -145,7 +145,6 @@ enum hnae3_reset_notify_type { HNAE3_DOWN_CLIENT, HNAE3_INIT_CLIENT, HNAE3_UNINIT_CLIENT, - HNAE3_RESTORE_CLIENT, }; enum hnae3_hw_error_type { @@ -233,7 +232,6 @@ struct hnae3_ae_dev { struct list_head node; u32 flag; unsigned long hw_err_reset_req; - enum hnae3_reset_type reset_type; void *priv; }; @@ -270,6 +268,8 @@ struct hnae3_ae_dev { * Set loopback * set_promisc_mode * Set promisc mode + * request_update_promisc_mode + * request to hclge(vf) to update promisc mode * set_mtu() * set mtu * get_pauseparam() @@ -354,8 +354,6 @@ struct hnae3_ae_dev { * Set vlan filter config of Ports * set_vf_vlan_filter() * Set vlan filter config of vf - * restore_vlan_table() - * Restore vlan filter entries after reset * enable_hw_strip_rxvtag() * Enable/disable hardware strip vlan tag of packets received * set_gro_en @@ -375,6 +373,8 @@ struct hnae3_ae_dev { * Set the max tx rate of specified vf. * set_vf_mac * Configure the default MAC for specified VF + * get_module_eeprom + * Get the optical module eeprom info. */ struct hnae3_ae_ops { int (*init_ae_dev)(struct hnae3_ae_dev *ae_dev); @@ -408,6 +408,7 @@ struct hnae3_ae_ops { int (*set_promisc_mode)(struct hnae3_handle *handle, bool en_uc_pmc, bool en_mc_pmc); + void (*request_update_promisc_mode)(struct hnae3_handle *handle); int (*set_mtu)(struct hnae3_handle *handle, int new_mtu); void (*get_pauseparam)(struct hnae3_handle *handle, @@ -525,7 +526,6 @@ struct hnae3_ae_ops { struct ethtool_rxnfc *cmd); int (*get_fd_all_rules)(struct hnae3_handle *handle, struct ethtool_rxnfc *cmd, u32 *rule_locs); - int (*restore_fd_rules)(struct hnae3_handle *handle); void (*enable_fd)(struct hnae3_handle *handle, bool enable); int (*add_arfs_entry)(struct hnae3_handle *handle, u16 queue_id, u16 flow_id, struct flow_keys *fkeys); @@ -539,7 +539,6 @@ struct hnae3_ae_ops { void (*set_timer_task)(struct hnae3_handle *handle, bool enable); int (*mac_connect_phy)(struct hnae3_handle *handle); void (*mac_disconnect_phy)(struct hnae3_handle *handle); - void (*restore_vlan_table)(struct hnae3_handle *handle); int (*get_vf_config)(struct hnae3_handle *handle, int vf, struct ifla_vf_info *ivf); int (*set_vf_link_state)(struct hnae3_handle *handle, int vf, @@ -550,6 +549,9 @@ struct hnae3_ae_ops { int (*set_vf_rate)(struct hnae3_handle *handle, int vf, int min_tx_rate, int max_tx_rate, bool force); int (*set_vf_mac)(struct hnae3_handle *handle, int vf, u8 *p); + int (*get_module_eeprom)(struct hnae3_handle *handle, u32 offset, + u32 len, u8 *data); + bool (*get_cmdq_stat)(struct hnae3_handle *handle); }; struct hnae3_dcb_ops { @@ -619,16 +621,6 @@ struct hnae3_roce_private_info { unsigned long state; }; -struct hnae3_unic_private_info { - struct net_device *netdev; - u16 rx_buf_len; - u16 num_tx_desc; - u16 num_rx_desc; - - u16 num_tqps; /* total number of tqps in this handle */ - struct hnae3_queue **tqp; /* array base of all TQPs of this instance */ -}; - #define HNAE3_SUPPORT_APP_LOOPBACK BIT(0) #define HNAE3_SUPPORT_PHY_LOOPBACK BIT(1) #define HNAE3_SUPPORT_SERDES_SERIAL_LOOPBACK BIT(2) @@ -654,7 +646,6 @@ struct hnae3_handle { union { struct net_device *netdev; /* first member */ struct hnae3_knic_private_info kinfo; - struct hnae3_unic_private_info uinfo; struct hnae3_roce_private_info rinfo; }; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c index e1d88095a77e..fe7fb565da19 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c @@ -262,6 +262,8 @@ static void hns3_dbg_help(struct hnae3_handle *h) dev_info(&h->pdev->dev, "dump mac tnl status\n"); dev_info(&h->pdev->dev, "dump loopback\n"); dev_info(&h->pdev->dev, "dump qs shaper [qs id]\n"); + dev_info(&h->pdev->dev, "dump uc mac list <func id>\n"); + dev_info(&h->pdev->dev, "dump mc mac list <func id>\n"); memset(printf_buf, 0, HNS3_DBG_BUF_LEN); strncat(printf_buf, "dump reg [[bios common] [ssu <port_id>]", @@ -270,7 +272,7 @@ static void hns3_dbg_help(struct hnae3_handle *h) " [igu egu <port_id>] [rpu <tc_queue_num>]", HNS3_DBG_BUF_LEN - strlen(printf_buf) - 1); strncat(printf_buf + strlen(printf_buf), - " [rtc] [ppp] [rcb] [tqp <queue_num>]]\n", + " [rtc] [ppp] [rcb] [tqp <queue_num>] [mac]]\n", HNS3_DBG_BUF_LEN - strlen(printf_buf) - 1); dev_info(&h->pdev->dev, "%s", printf_buf); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c index da98fd7c8eca..b14f2abc2425 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c @@ -15,7 +15,6 @@ #include <linux/aer.h> #include <linux/skbuff.h> #include <linux/sctp.h> -#include <linux/vermagic.h> #include <net/gre.h> #include <net/ip6_checksum.h> #include <net/pkt_cls.h> @@ -41,10 +40,8 @@ } while (0) static void hns3_clear_all_ring(struct hnae3_handle *h, bool force); -static void hns3_remove_hw_addr(struct net_device *netdev); static const char hns3_driver_name[] = "hns3"; -const char hns3_driver_version[] = VERMAGIC_STRING; static const char hns3_driver_string[] = "Hisilicon Ethernet Network Driver for Hip08 Family"; static const char hns3_copyright[] = "Copyright (c) 2017 Huawei Corporation."; @@ -550,6 +547,13 @@ static int hns3_nic_uc_unsync(struct net_device *netdev, { struct hnae3_handle *h = hns3_get_handle(netdev); + /* need ignore the request of removing device address, because + * we store the device address and other addresses of uc list + * in the function's mac filter list. + */ + if (ether_addr_equal(addr, netdev->dev_addr)) + return 0; + if (h->ae_algo->ops->rm_uc_addr) return h->ae_algo->ops->rm_uc_addr(h, addr); @@ -597,34 +601,25 @@ static void hns3_nic_set_rx_mode(struct net_device *netdev) { struct hnae3_handle *h = hns3_get_handle(netdev); u8 new_flags; - int ret; new_flags = hns3_get_netdev_flags(netdev); - ret = __dev_uc_sync(netdev, hns3_nic_uc_sync, hns3_nic_uc_unsync); - if (ret) { - netdev_err(netdev, "sync uc address fail\n"); - if (ret == -ENOSPC) - new_flags |= HNAE3_OVERFLOW_UPE; - } - - if (netdev->flags & IFF_MULTICAST) { - ret = __dev_mc_sync(netdev, hns3_nic_mc_sync, - hns3_nic_mc_unsync); - if (ret) { - netdev_err(netdev, "sync mc address fail\n"); - if (ret == -ENOSPC) - new_flags |= HNAE3_OVERFLOW_MPE; - } - } + __dev_uc_sync(netdev, hns3_nic_uc_sync, hns3_nic_uc_unsync); + __dev_mc_sync(netdev, hns3_nic_mc_sync, hns3_nic_mc_unsync); /* User mode Promisc mode enable and vlan filtering is disabled to - * let all packets in. MAC-VLAN Table overflow Promisc enabled and - * vlan fitering is enabled + * let all packets in. */ - hns3_enable_vlan_filter(netdev, new_flags & HNAE3_VLAN_FLTR); h->netdev_flags = new_flags; - hns3_update_promisc_mode(netdev, new_flags); + hns3_request_update_promisc_mode(h); +} + +void hns3_request_update_promisc_mode(struct hnae3_handle *handle) +{ + const struct hnae3_ae_ops *ops = handle->ae_algo->ops; + + if (ops->request_update_promisc_mode) + ops->request_update_promisc_mode(handle); } int hns3_update_promisc_mode(struct net_device *netdev, u8 promisc_flags) @@ -1450,9 +1445,6 @@ netdev_tx_t hns3_nic_net_xmit(struct sk_buff *skb, struct net_device *netdev) bd_num += ret; - if (!skb_has_frag_list(skb)) - goto out; - skb_walk_frags(skb, frag_skb) { ret = hns3_fill_skb_to_desc(ring, frag_skb, DESC_TYPE_FRAGLIST_SKB); @@ -1461,7 +1453,7 @@ netdev_tx_t hns3_nic_net_xmit(struct sk_buff *skb, struct net_device *netdev) bd_num += ret; } -out: + pre_ntu = ring->next_to_use ? (ring->next_to_use - 1) : (ring->desc_num - 1); ring->desc[pre_ntu].tx.bdtp_fe_sc_vld_ra_ri |= @@ -1552,12 +1544,6 @@ static int hns3_nic_set_features(struct net_device *netdev, return ret; } - if ((changed & NETIF_F_HW_VLAN_CTAG_FILTER) && - h->ae_algo->ops->enable_vlan_filter) { - enable = !!(features & NETIF_F_HW_VLAN_CTAG_FILTER); - h->ae_algo->ops->enable_vlan_filter(h, enable); - } - if ((changed & NETIF_F_HW_VLAN_CTAG_RX) && h->ae_algo->ops->enable_hw_strip_rxvtag) { enable = !!(features & NETIF_F_HW_VLAN_CTAG_RX); @@ -2107,7 +2093,6 @@ static int hns3_probe(struct pci_dev *pdev, const struct pci_device_id *ent) ae_dev->pdev = pdev; ae_dev->flag = ent->driver_data; - ae_dev->reset_type = HNAE3_NONE_RESET; hns3_get_dev_capability(pdev, ae_dev); pci_set_drvdata(pdev, ae_dev); @@ -3909,9 +3894,11 @@ static int hns3_init_mac_addr(struct net_device *netdev) eth_hw_addr_random(netdev); dev_warn(priv->dev, "using random MAC address %pM\n", netdev->dev_addr); - } else { + } else if (!ether_addr_equal(netdev->dev_addr, mac_addr_temp)) { ether_addr_copy(netdev->dev_addr, mac_addr_temp); ether_addr_copy(netdev->perm_addr, mac_addr_temp); + } else { + return 0; } if (h->ae_algo->ops->set_mac_addr) @@ -3939,17 +3926,6 @@ static void hns3_uninit_phy(struct net_device *netdev) h->ae_algo->ops->mac_disconnect_phy(h); } -static int hns3_restore_fd_rules(struct net_device *netdev) -{ - struct hnae3_handle *h = hns3_get_handle(netdev); - int ret = 0; - - if (h->ae_algo->ops->restore_fd_rules) - ret = h->ae_algo->ops->restore_fd_rules(h); - - return ret; -} - static void hns3_del_all_fd_rules(struct net_device *netdev, bool clear_list) { struct hnae3_handle *h = hns3_get_handle(netdev); @@ -4121,8 +4097,6 @@ static void hns3_client_uninit(struct hnae3_handle *handle, bool reset) struct hns3_nic_priv *priv = netdev_priv(netdev); int ret; - hns3_remove_hw_addr(netdev); - if (netdev->reg_state != NETREG_UNINITIALIZED) unregister_netdev(netdev); @@ -4193,56 +4167,6 @@ static int hns3_client_setup_tc(struct hnae3_handle *handle, u8 tc) return hns3_nic_set_real_num_queue(ndev); } -static int hns3_recover_hw_addr(struct net_device *ndev) -{ - struct netdev_hw_addr_list *list; - struct netdev_hw_addr *ha, *tmp; - int ret = 0; - - netif_addr_lock_bh(ndev); - /* go through and sync uc_addr entries to the device */ - list = &ndev->uc; - list_for_each_entry_safe(ha, tmp, &list->list, list) { - ret = hns3_nic_uc_sync(ndev, ha->addr); - if (ret) - goto out; - } - - /* go through and sync mc_addr entries to the device */ - list = &ndev->mc; - list_for_each_entry_safe(ha, tmp, &list->list, list) { - ret = hns3_nic_mc_sync(ndev, ha->addr); - if (ret) - goto out; - } - -out: - netif_addr_unlock_bh(ndev); - return ret; -} - -static void hns3_remove_hw_addr(struct net_device *netdev) -{ - struct netdev_hw_addr_list *list; - struct netdev_hw_addr *ha, *tmp; - - hns3_nic_uc_unsync(netdev, netdev->dev_addr); - - netif_addr_lock_bh(netdev); - /* go through and unsync uc_addr entries to the device */ - list = &netdev->uc; - list_for_each_entry_safe(ha, tmp, &list->list, list) - hns3_nic_uc_unsync(netdev, ha->addr); - - /* go through and unsync mc_addr entries to the device */ - list = &netdev->mc; - list_for_each_entry_safe(ha, tmp, &list->list, list) - if (ha->refcount > 1) - hns3_nic_mc_unsync(netdev, ha->addr); - - netif_addr_unlock_bh(netdev); -} - static void hns3_clear_tx_ring(struct hns3_enet_ring *ring) { while (ring->next_to_clean != ring->next_to_use) { @@ -4401,7 +4325,6 @@ static void hns3_restore_coal(struct hns3_nic_priv *priv) static int hns3_reset_notify_down_enet(struct hnae3_handle *handle) { - struct hnae3_ae_dev *ae_dev = pci_get_drvdata(handle->pdev); struct hnae3_knic_private_info *kinfo = &handle->kinfo; struct net_device *ndev = kinfo->netdev; struct hns3_nic_priv *priv = netdev_priv(ndev); @@ -4409,15 +4332,6 @@ static int hns3_reset_notify_down_enet(struct hnae3_handle *handle) if (test_and_set_bit(HNS3_NIC_STATE_RESETTING, &priv->state)) return 0; - /* it is cumbersome for hardware to pick-and-choose entries for deletion - * from table space. Hence, for function reset software intervention is - * required to delete the entries - */ - if (hns3_dev_ongoing_func_reset(ae_dev)) { - hns3_remove_hw_addr(ndev); - hns3_del_all_fd_rules(ndev, false); - } - if (!netif_running(ndev)) return 0; @@ -4484,6 +4398,9 @@ static int hns3_reset_notify_init_enet(struct hnae3_handle *handle) goto err_init_irq_fail; } + if (!hns3_is_phys_func(handle->pdev)) + hns3_init_mac_addr(netdev); + ret = hns3_client_start(handle); if (ret) { dev_err(priv->dev, "hns3_client_start fail! ret=%d\n", ret); @@ -4509,33 +4426,6 @@ err_put_ring: return ret; } -static int hns3_reset_notify_restore_enet(struct hnae3_handle *handle) -{ - struct net_device *netdev = handle->kinfo.netdev; - bool vlan_filter_enable; - int ret; - - ret = hns3_init_mac_addr(netdev); - if (ret) - return ret; - - ret = hns3_recover_hw_addr(netdev); - if (ret) - return ret; - - ret = hns3_update_promisc_mode(netdev, handle->netdev_flags); - if (ret) - return ret; - - vlan_filter_enable = netdev->flags & IFF_PROMISC ? false : true; - hns3_enable_vlan_filter(netdev, vlan_filter_enable); - - if (handle->ae_algo->ops->restore_vlan_table) - handle->ae_algo->ops->restore_vlan_table(handle); - - return hns3_restore_fd_rules(netdev); -} - static int hns3_reset_notify_uninit_enet(struct hnae3_handle *handle) { struct net_device *netdev = handle->kinfo.netdev; @@ -4585,9 +4475,6 @@ static int hns3_reset_notify(struct hnae3_handle *handle, case HNAE3_UNINIT_CLIENT: ret = hns3_reset_notify_uninit_enet(handle); break; - case HNAE3_RESTORE_CLIENT: - ret = hns3_reset_notify_restore_enet(handle); - break; default: break; } @@ -4765,4 +4652,3 @@ MODULE_DESCRIPTION("HNS3: Hisilicon Ethernet Driver"); MODULE_AUTHOR("Huawei Tech. Co., Ltd."); MODULE_LICENSE("GPL"); MODULE_ALIAS("pci:hns-nic"); -MODULE_VERSION(HNS3_MOD_VERSION); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h index abefd7a179f7..66cd4395f781 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.h @@ -8,10 +8,6 @@ #include "hnae3.h" -#define HNS3_MOD_VERSION "1.0" - -extern const char hns3_driver_version[]; - enum hns3_nic_state { HNS3_NIC_STATE_TESTING, HNS3_NIC_STATE_RESETTING, @@ -50,23 +46,6 @@ enum hns3_nic_state { #define HNS3_RING_CFG_VF_NUM_REG 0x00080 #define HNS3_RING_ASID_REG 0x0008C #define HNS3_RING_EN_REG 0x00090 -#define HNS3_RING_T0_BE_RST 0x00094 -#define HNS3_RING_COULD_BE_RST 0x00098 -#define HNS3_RING_WRR_WEIGHT_REG 0x0009c - -#define HNS3_RING_INTMSK_RXWL_REG 0x000A0 -#define HNS3_RING_INTSTS_RX_RING_REG 0x000A4 -#define HNS3_RX_RING_INT_STS_REG 0x000A8 -#define HNS3_RING_INTMSK_TXWL_REG 0x000AC -#define HNS3_RING_INTSTS_TX_RING_REG 0x000B0 -#define HNS3_TX_RING_INT_STS_REG 0x000B4 -#define HNS3_RING_INTMSK_RX_OVERTIME_REG 0x000B8 -#define HNS3_RING_INTSTS_RX_OVERTIME_REG 0x000BC -#define HNS3_RING_INTMSK_TX_OVERTIME_REG 0x000C4 -#define HNS3_RING_INTSTS_TX_OVERTIME_REG 0x000C8 - -#define HNS3_RING_MB_CTRL_REG 0x00100 -#define HNS3_RING_MB_DATA_BASE_REG 0x00200 #define HNS3_TX_REG_OFFSET 0x40 @@ -490,21 +469,8 @@ struct hns3_enet_tqp_vector { unsigned long last_jiffies; } ____cacheline_internodealigned_in_smp; -enum hns3_udp_tnl_type { - HNS3_UDP_TNL_VXLAN, - HNS3_UDP_TNL_GENEVE, - HNS3_UDP_TNL_MAX, -}; - -struct hns3_udp_tunnel { - u16 dst_port; - int used; -}; - struct hns3_nic_priv { struct hnae3_handle *ae_handle; - u32 enet_ver; - u32 port_id; struct net_device *netdev; struct device *dev; @@ -516,19 +482,10 @@ struct hns3_nic_priv { struct hns3_enet_tqp_vector *tqp_vector; u16 vector_num; - /* The most recently read link state */ - int link; u64 tx_timeout_count; unsigned long state; - struct timer_list service_timer; - - struct work_struct service_task; - - struct notifier_block notifier_block; - /* Vxlan/Geneve information */ - struct hns3_udp_tunnel udp_tnl[HNS3_UDP_TNL_MAX]; struct hns3_enet_coalesce tx_coal; struct hns3_enet_coalesce rx_coal; }; @@ -580,15 +537,6 @@ static inline void hns3_write_reg(void __iomem *base, u32 reg, u32 value) writel(value, reg_addr + reg); } -static inline bool hns3_dev_ongoing_func_reset(struct hnae3_ae_dev *ae_dev) -{ - return (ae_dev && (ae_dev->reset_type == HNAE3_FUNC_RESET || - ae_dev->reset_type == HNAE3_FLR_RESET || - ae_dev->reset_type == HNAE3_VF_FUNC_RESET || - ae_dev->reset_type == HNAE3_VF_FULL_RESET || - ae_dev->reset_type == HNAE3_VF_PF_FUNC_RESET)); -} - #define hns3_read_dev(a, reg) \ hns3_read_reg((a)->io_base, (reg)) @@ -662,6 +610,7 @@ void hns3_set_vector_coalesce_rl(struct hns3_enet_tqp_vector *tqp_vector, void hns3_enable_vlan_filter(struct net_device *netdev, bool enable); int hns3_update_promisc_mode(struct net_device *netdev, u8 promisc_flags); +void hns3_request_update_promisc_mode(struct hnae3_handle *handle); #ifdef CONFIG_HNS3_DCB void hns3_dcbnl_setup(struct hnae3_handle *handle); diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c index 28b81f24afa1..6b1545f982aa 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c @@ -4,6 +4,7 @@ #include <linux/etherdevice.h> #include <linux/string.h> #include <linux/phy.h> +#include <linux/sfp.h> #include "hns3_enet.h" @@ -12,6 +13,11 @@ struct hns3_stats { int stats_offset; }; +struct hns3_sfp_type { + u8 type; + u8 ext_type; +}; + /* tqp related stats */ #define HNS3_TQP_STAT(_string, _member) { \ .stats_string = _string, \ @@ -99,7 +105,7 @@ static int hns3_lp_setup(struct net_device *ndev, enum hnae3_loop loop, bool en) h->ae_algo->ops->set_promisc_mode(h, true, true); } else { /* recover promisc mode before loopback test */ - hns3_update_promisc_mode(ndev, h->netdev_flags); + hns3_request_update_promisc_mode(h); vlan_filter_enable = ndev->flags & IFF_PROMISC ? false : true; hns3_enable_vlan_filter(ndev, vlan_filter_enable); } @@ -546,10 +552,6 @@ static void hns3_get_drvinfo(struct net_device *netdev, return; } - strncpy(drvinfo->version, hns3_driver_version, - sizeof(drvinfo->version)); - drvinfo->version[sizeof(drvinfo->version) - 1] = '\0'; - strncpy(drvinfo->driver, h->pdev->driver->name, sizeof(drvinfo->driver)); drvinfo->driver[sizeof(drvinfo->driver) - 1] = '\0'; @@ -771,8 +773,13 @@ static int hns3_set_link_ksettings(struct net_device *netdev, cmd->base.autoneg, cmd->base.speed, cmd->base.duplex); /* Only support ksettings_set for netdev with phy attached for now */ - if (netdev->phydev) + if (netdev->phydev) { + if (cmd->base.speed == SPEED_1000 && + cmd->base.autoneg == AUTONEG_DISABLE) + return -EINVAL; + return phy_ethtool_ksettings_set(netdev->phydev, cmd); + } if (handle->pdev->revision == 0x20) return -EOPNOTSUPP; @@ -1390,6 +1397,73 @@ static int hns3_set_fecparam(struct net_device *netdev, return ops->set_fec(handle, fec_mode); } +static int hns3_get_module_info(struct net_device *netdev, + struct ethtool_modinfo *modinfo) +{ +#define HNS3_SFF_8636_V1_3 0x03 + + struct hnae3_handle *handle = hns3_get_handle(netdev); + const struct hnae3_ae_ops *ops = handle->ae_algo->ops; + struct hns3_sfp_type sfp_type; + int ret; + + if (handle->pdev->revision == 0x20 || !ops->get_module_eeprom) + return -EOPNOTSUPP; + + memset(&sfp_type, 0, sizeof(sfp_type)); + ret = ops->get_module_eeprom(handle, 0, sizeof(sfp_type) / sizeof(u8), + (u8 *)&sfp_type); + if (ret) + return ret; + + switch (sfp_type.type) { + case SFF8024_ID_SFP: + modinfo->type = ETH_MODULE_SFF_8472; + modinfo->eeprom_len = ETH_MODULE_SFF_8472_LEN; + break; + case SFF8024_ID_QSFP_8438: + modinfo->type = ETH_MODULE_SFF_8436; + modinfo->eeprom_len = ETH_MODULE_SFF_8436_MAX_LEN; + break; + case SFF8024_ID_QSFP_8436_8636: + if (sfp_type.ext_type < HNS3_SFF_8636_V1_3) { + modinfo->type = ETH_MODULE_SFF_8436; + modinfo->eeprom_len = ETH_MODULE_SFF_8436_MAX_LEN; + } else { + modinfo->type = ETH_MODULE_SFF_8636; + modinfo->eeprom_len = ETH_MODULE_SFF_8636_MAX_LEN; + } + break; + case SFF8024_ID_QSFP28_8636: + modinfo->type = ETH_MODULE_SFF_8636; + modinfo->eeprom_len = ETH_MODULE_SFF_8636_MAX_LEN; + break; + default: + netdev_err(netdev, "Optical module unknown: %#x\n", + sfp_type.type); + return -EINVAL; + } + + return 0; +} + +static int hns3_get_module_eeprom(struct net_device *netdev, + struct ethtool_eeprom *ee, u8 *data) +{ + struct hnae3_handle *handle = hns3_get_handle(netdev); + const struct hnae3_ae_ops *ops = handle->ae_algo->ops; + + if (handle->pdev->revision == 0x20 || !ops->get_module_eeprom) + return -EOPNOTSUPP; + + if (!ee->len) + return -EINVAL; + + memset(data, 0, ee->len); + + return ops->get_module_eeprom(handle, ee->offset, ee->len, data); +} + #define HNS3_ETHTOOL_COALESCE (ETHTOOL_COALESCE_USECS | \ ETHTOOL_COALESCE_USE_ADAPTIVE | \ ETHTOOL_COALESCE_RX_USECS_HIGH | \ @@ -1453,6 +1527,8 @@ static const struct ethtool_ops hns3_ethtool_ops = { .set_msglevel = hns3_set_msglevel, .get_fecparam = hns3_get_fecparam, .set_fecparam = hns3_set_fecparam, + .get_module_info = hns3_get_module_info, + .get_module_eeprom = hns3_get_module_eeprom, }; void hns3_ethtool_set_ops(struct net_device *netdev) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/Makefile b/drivers/net/ethernet/hisilicon/hns3/hns3pf/Makefile index 0fb61d440d3b..6c28c8f6292c 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/Makefile +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/Makefile @@ -4,6 +4,7 @@ # ccflags-y := -I $(srctree)/drivers/net/ethernet/hisilicon/hns3 +ccflags-y += -I $(srctree)/$(src) obj-$(CONFIG_HNS3_HCLGE) += hclge.o hclge-objs = hclge_main.o hclge_cmd.o hclge_mdio.o hclge_tm.o hclge_mbx.o hclge_err.o hclge_debugfs.o diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c index 7f509eff562e..1d6c328bd9fb 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.c @@ -11,8 +11,6 @@ #include "hnae3.h" #include "hclge_main.h" -#define hclge_is_csq(ring) ((ring)->flag & HCLGE_TYPE_CSQ) - #define cmq_ring_to_dev(ring) (&(ring)->dev->pdev->dev) static int hclge_ring_space(struct hclge_cmq_ring *ring) @@ -426,6 +424,9 @@ int hclge_cmd_init(struct hclge_dev *hdev) * reset may happen when lower level reset is being processed. */ if ((hclge_is_reset_pending(hdev))) { + dev_err(&hdev->pdev->dev, + "failed to init cmd since reset %#lx pending\n", + hdev->reset_pending); ret = -EBUSY; goto err_cmd_init; } diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h index 96498d9b4754..463f29151ef0 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_cmd.h @@ -184,11 +184,11 @@ enum hclge_opcode_type { /* TQP commands */ HCLGE_OPC_CFG_TX_QUEUE = 0x0B01, HCLGE_OPC_QUERY_TX_POINTER = 0x0B02, - HCLGE_OPC_QUERY_TX_STATUS = 0x0B03, + HCLGE_OPC_QUERY_TX_STATS = 0x0B03, HCLGE_OPC_TQP_TX_QUEUE_TC = 0x0B04, HCLGE_OPC_CFG_RX_QUEUE = 0x0B11, HCLGE_OPC_QUERY_RX_POINTER = 0x0B12, - HCLGE_OPC_QUERY_RX_STATUS = 0x0B13, + HCLGE_OPC_QUERY_RX_STATS = 0x0B13, HCLGE_OPC_STASH_RX_QUEUE_LRO = 0x0B16, HCLGE_OPC_CFG_RX_QUEUE_LRO = 0x0B17, HCLGE_OPC_CFG_COM_TQP_QUEUE = 0x0B20, @@ -270,6 +270,8 @@ enum hclge_opcode_type { HCLGE_OPC_M7_COMPAT_CFG = 0x701A, /* SFP command */ + HCLGE_OPC_GET_SFP_EEPROM = 0x7100, + HCLGE_OPC_GET_SFP_EXIST = 0x7101, HCLGE_OPC_GET_SFP_INFO = 0x7104, /* Error INT commands */ @@ -733,31 +735,6 @@ struct hclge_mac_mgr_tbl_entry_cmd { u8 rsv3[2]; }; -struct hclge_mac_vlan_add_cmd { - __le16 flags; - __le16 mac_addr_hi16; - __le32 mac_addr_lo32; - __le32 mac_addr_msk_hi32; - __le16 mac_addr_msk_lo16; - __le16 vlan_tag; - __le16 ingress_port; - __le16 egress_port; - u8 rsv[4]; -}; - -#define HNS3_MAC_VLAN_CFG_FLAG_BIT 0 -struct hclge_mac_vlan_remove_cmd { - __le16 flags; - __le16 mac_addr_hi16; - __le32 mac_addr_lo32; - __le32 mac_addr_msk_hi32; - __le16 mac_addr_msk_lo16; - __le16 vlan_tag; - __le16 ingress_port; - __le16 egress_port; - u8 rsv[4]; -}; - struct hclge_vlan_filter_ctrl_cmd { u8 vlan_type; u8 vlan_fe; @@ -907,8 +884,8 @@ struct hclge_cfg_tso_status_cmd { #define HCLGE_GRO_EN_B 0 struct hclge_cfg_gro_status_cmd { - __le16 gro_en; - u8 rsv[22]; + u8 gro_en; + u8 rsv[23]; }; #define HCLGE_TSO_MSS_MIN 256 @@ -1079,6 +1056,19 @@ struct hclge_firmware_compat_cmd { u8 rsv[20]; }; +#define HCLGE_SFP_INFO_CMD_NUM 6 +#define HCLGE_SFP_INFO_BD0_LEN 20 +#define HCLGE_SFP_INFO_BDX_LEN 24 +#define HCLGE_SFP_INFO_MAX_LEN \ + (HCLGE_SFP_INFO_BD0_LEN + \ + (HCLGE_SFP_INFO_CMD_NUM - 1) * HCLGE_SFP_INFO_BDX_LEN) + +struct hclge_sfp_info_bd0_cmd { + __le16 offset; + __le16 read_len; + u8 data[HCLGE_SFP_INFO_BD0_LEN]; +}; + int hclge_cmd_init(struct hclge_dev *hdev); static inline void hclge_write_reg(void __iomem *base, u32 reg, u32 value) { diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c index 17228288d4df..26f6f068b01d 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_debugfs.c @@ -143,7 +143,7 @@ static void hclge_dbg_dump_reg_common(struct hclge_dev *hdev, return; } - buf_len = sizeof(struct hclge_desc) * bd_num; + buf_len = sizeof(struct hclge_desc) * bd_num; desc_src = kzalloc(buf_len, GFP_KERNEL); if (!desc_src) return; @@ -173,6 +173,114 @@ static void hclge_dbg_dump_reg_common(struct hclge_dev *hdev, kfree(desc_src); } +static void hclge_dbg_dump_mac_enable_status(struct hclge_dev *hdev) +{ + struct hclge_config_mac_mode_cmd *req; + struct hclge_desc desc; + u32 loop_en; + int ret; + + hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_MAC_MODE, true); + + ret = hclge_cmd_send(&hdev->hw, &desc, 1); + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to dump mac enable status, ret = %d\n", ret); + return; + } + + req = (struct hclge_config_mac_mode_cmd *)desc.data; + loop_en = le32_to_cpu(req->txrx_pad_fcs_loop_en); + + dev_info(&hdev->pdev->dev, "config_mac_trans_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_TX_EN_B)); + dev_info(&hdev->pdev->dev, "config_mac_rcv_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_RX_EN_B)); + dev_info(&hdev->pdev->dev, "config_pad_trans_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_PAD_TX_B)); + dev_info(&hdev->pdev->dev, "config_pad_rcv_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_PAD_RX_B)); + dev_info(&hdev->pdev->dev, "config_1588_trans_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_1588_TX_B)); + dev_info(&hdev->pdev->dev, "config_1588_rcv_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_1588_RX_B)); + dev_info(&hdev->pdev->dev, "config_mac_app_loop_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_APP_LP_B)); + dev_info(&hdev->pdev->dev, "config_mac_line_loop_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_LINE_LP_B)); + dev_info(&hdev->pdev->dev, "config_mac_fcs_tx_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_FCS_TX_B)); + dev_info(&hdev->pdev->dev, "config_mac_rx_oversize_truncate_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_RX_OVERSIZE_TRUNCATE_B)); + dev_info(&hdev->pdev->dev, "config_mac_rx_fcs_strip_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_RX_FCS_STRIP_B)); + dev_info(&hdev->pdev->dev, "config_mac_rx_fcs_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_RX_FCS_B)); + dev_info(&hdev->pdev->dev, "config_mac_tx_under_min_err_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_TX_UNDER_MIN_ERR_B)); + dev_info(&hdev->pdev->dev, "config_mac_tx_oversize_truncate_en: %#x\n", + hnae3_get_bit(loop_en, HCLGE_MAC_TX_OVERSIZE_TRUNCATE_B)); +} + +static void hclge_dbg_dump_mac_frame_size(struct hclge_dev *hdev) +{ + struct hclge_config_max_frm_size_cmd *req; + struct hclge_desc desc; + int ret; + + hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_MAX_FRM_SIZE, true); + + ret = hclge_cmd_send(&hdev->hw, &desc, 1); + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to dump mac frame size, ret = %d\n", ret); + return; + } + + req = (struct hclge_config_max_frm_size_cmd *)desc.data; + + dev_info(&hdev->pdev->dev, "max_frame_size: %u\n", + le16_to_cpu(req->max_frm_size)); + dev_info(&hdev->pdev->dev, "min_frame_size: %u\n", req->min_frm_size); +} + +static void hclge_dbg_dump_mac_speed_duplex(struct hclge_dev *hdev) +{ +#define HCLGE_MAC_SPEED_SHIFT 0 +#define HCLGE_MAC_SPEED_MASK GENMASK(5, 0) +#define HCLGE_MAC_DUPLEX_SHIFT 7 + + struct hclge_config_mac_speed_dup_cmd *req; + struct hclge_desc desc; + int ret; + + hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_CONFIG_SPEED_DUP, true); + + ret = hclge_cmd_send(&hdev->hw, &desc, 1); + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to dump mac speed duplex, ret = %d\n", ret); + return; + } + + req = (struct hclge_config_mac_speed_dup_cmd *)desc.data; + + dev_info(&hdev->pdev->dev, "speed: %#lx\n", + hnae3_get_field(req->speed_dup, HCLGE_MAC_SPEED_MASK, + HCLGE_MAC_SPEED_SHIFT)); + dev_info(&hdev->pdev->dev, "duplex: %#x\n", + hnae3_get_bit(req->speed_dup, HCLGE_MAC_DUPLEX_SHIFT)); +} + +static void hclge_dbg_dump_mac(struct hclge_dev *hdev) +{ + hclge_dbg_dump_mac_enable_status(hdev); + + hclge_dbg_dump_mac_frame_size(hdev); + + hclge_dbg_dump_mac_speed_duplex(hdev); +} + static void hclge_dbg_dump_dcb(struct hclge_dev *hdev, const char *cmd_buf) { struct device *dev = &hdev->pdev->dev; @@ -304,6 +412,11 @@ static void hclge_dbg_dump_reg_cmd(struct hclge_dev *hdev, const char *cmd_buf) } } + if (strncmp(cmd_buf, "mac", strlen("mac")) == 0) { + hclge_dbg_dump_mac(hdev); + has_dump = true; + } + if (strncmp(cmd_buf, "dcb", 3) == 0) { hclge_dbg_dump_dcb(hdev, &cmd_buf[sizeof("dcb")]); has_dump = true; @@ -578,7 +691,7 @@ static void hclge_dbg_dump_tm_map(struct hclge_dev *hdev, enum hclge_opcode_type cmd; struct hclge_desc desc; int queue_id, group_id; - u32 qset_maping[32]; + u32 qset_mapping[32]; int tc_id, qset_id; int pri_id, ret; u32 i; @@ -633,7 +746,7 @@ static void hclge_dbg_dump_tm_map(struct hclge_dev *hdev, if (ret) goto err_tm_map_cmd_send; - qset_maping[group_id] = + qset_mapping[group_id] = le32_to_cpu(bp_to_qs_map_cmd->qs_bit_map); } @@ -643,11 +756,11 @@ static void hclge_dbg_dump_tm_map(struct hclge_dev *hdev, for (group_id = 0; group_id < 4; group_id++) { dev_info(&hdev->pdev->dev, "%04d | %08x:%08x:%08x:%08x:%08x:%08x:%08x:%08x\n", - group_id * 256, qset_maping[(u32)(i + 7)], - qset_maping[(u32)(i + 6)], qset_maping[(u32)(i + 5)], - qset_maping[(u32)(i + 4)], qset_maping[(u32)(i + 3)], - qset_maping[(u32)(i + 2)], qset_maping[(u32)(i + 1)], - qset_maping[i]); + group_id * 256, qset_mapping[(u32)(i + 7)], + qset_mapping[(u32)(i + 6)], qset_mapping[(u32)(i + 5)], + qset_mapping[(u32)(i + 4)], qset_mapping[(u32)(i + 3)], + qset_mapping[(u32)(i + 2)], qset_mapping[(u32)(i + 1)], + qset_mapping[i]); i += 8; } @@ -1145,6 +1258,7 @@ static void hclge_dbg_dump_ncl_config(struct hclge_dev *hdev, { #define HCLGE_MAX_NCL_CONFIG_OFFSET 4096 #define HCLGE_NCL_CONFIG_LENGTH_IN_EACH_CMD (20 + 24 * 4) +#define HCLGE_NCL_CONFIG_PARAM_NUM 2 struct hclge_desc desc[HCLGE_CMD_NCL_CONFIG_BD_NUM]; int bd_num = HCLGE_CMD_NCL_CONFIG_BD_NUM; @@ -1154,13 +1268,17 @@ static void hclge_dbg_dump_ncl_config(struct hclge_dev *hdev, int ret; ret = sscanf(cmd_buf, "%x %x", &offset, &length); - if (ret != 2 || offset >= HCLGE_MAX_NCL_CONFIG_OFFSET || - length > HCLGE_MAX_NCL_CONFIG_OFFSET - offset) { - dev_err(&hdev->pdev->dev, "Invalid offset or length.\n"); + if (ret != HCLGE_NCL_CONFIG_PARAM_NUM) { + dev_err(&hdev->pdev->dev, + "Too few parameters, num = %d.\n", ret); return; } - if (offset < 0 || length <= 0) { - dev_err(&hdev->pdev->dev, "Non-positive offset or length.\n"); + + if (offset < 0 || offset >= HCLGE_MAX_NCL_CONFIG_OFFSET || + length <= 0 || length > HCLGE_MAX_NCL_CONFIG_OFFSET - offset) { + dev_err(&hdev->pdev->dev, + "Invalid input, offset = %d, length = %d.\n", + offset, length); return; } @@ -1328,6 +1446,49 @@ static void hclge_dbg_dump_qs_shaper(struct hclge_dev *hdev, hclge_dbg_dump_qs_shaper_single(hdev, qsid); } +static int hclge_dbg_dump_mac_list(struct hclge_dev *hdev, const char *cmd_buf, + bool is_unicast) +{ + struct hclge_mac_node *mac_node, *tmp; + struct hclge_vport *vport; + struct list_head *list; + u32 func_id; + int ret; + + ret = kstrtouint(cmd_buf, 0, &func_id); + if (ret < 0) { + dev_err(&hdev->pdev->dev, + "dump mac list: bad command string, ret = %d\n", ret); + return -EINVAL; + } + + if (func_id >= hdev->num_alloc_vport) { + dev_err(&hdev->pdev->dev, + "function id(%u) is out of range(0-%u)\n", func_id, + hdev->num_alloc_vport - 1); + return -EINVAL; + } + + vport = &hdev->vport[func_id]; + + list = is_unicast ? &vport->uc_mac_list : &vport->mc_mac_list; + + dev_info(&hdev->pdev->dev, "vport %u %s mac list:\n", + func_id, is_unicast ? "uc" : "mc"); + dev_info(&hdev->pdev->dev, "mac address state\n"); + + spin_lock_bh(&vport->mac_list_lock); + + list_for_each_entry_safe(mac_node, tmp, list, node) { + dev_info(&hdev->pdev->dev, "%pM %d\n", + mac_node->mac_addr, mac_node->state); + } + + spin_unlock_bh(&vport->mac_list_lock); + + return 0; +} + int hclge_dbg_run_cmd(struct hnae3_handle *handle, const char *cmd_buf) { #define DUMP_REG "dump reg" @@ -1372,6 +1533,14 @@ int hclge_dbg_run_cmd(struct hnae3_handle *handle, const char *cmd_buf) } else if (strncmp(cmd_buf, "dump qs shaper", 14) == 0) { hclge_dbg_dump_qs_shaper(hdev, &cmd_buf[sizeof("dump qs shaper")]); + } else if (strncmp(cmd_buf, "dump uc mac list", 16) == 0) { + hclge_dbg_dump_mac_list(hdev, + &cmd_buf[sizeof("dump uc mac list")], + true); + } else if (strncmp(cmd_buf, "dump mc mac list", 16) == 0) { + hclge_dbg_dump_mac_list(hdev, + &cmd_buf[sizeof("dump mc mac list")], + false); } else { dev_info(&hdev->pdev->dev, "unknown command\n"); return -EINVAL; diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h index 876fd81ad2f1..608fe26fc3fe 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.h @@ -16,7 +16,6 @@ #define HCLGE_RAS_REG_NFE_MASK 0xFF00 #define HCLGE_RAS_REG_ROCEE_ERR_MASK 0x3000000 -#define HCLGE_VECTOR0_PF_OTHER_INT_STS_REG 0x20800 #define HCLGE_VECTOR0_REG_MSIX_MASK 0x1FF00 #define HCLGE_IMP_TCM_ECC_ERR_INT_EN 0xFFFF0000 diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c index a758f9ae32be..96bfad52630d 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c @@ -62,14 +62,16 @@ static int hclge_init_vlan_config(struct hclge_dev *hdev); static void hclge_sync_vlan_filter(struct hclge_dev *hdev); static int hclge_reset_ae_dev(struct hnae3_ae_dev *ae_dev); static bool hclge_get_hw_reset_stat(struct hnae3_handle *handle); -static int hclge_set_umv_space(struct hclge_dev *hdev, u16 space_size, - u16 *allocated_size, bool is_alloc); static void hclge_rfs_filter_expire(struct hclge_dev *hdev); static void hclge_clear_arfs_rules(struct hnae3_handle *handle); static enum hnae3_reset_type hclge_get_reset_level(struct hnae3_ae_dev *ae_dev, unsigned long *addr); static int hclge_set_default_loopback(struct hclge_dev *hdev); +static void hclge_sync_mac_table(struct hclge_dev *hdev); +static void hclge_restore_hw_table(struct hclge_dev *hdev); +static void hclge_sync_promisc_mode(struct hclge_dev *hdev); + static struct hnae3_ae_algo ae_algo; static struct workqueue_struct *hclge_wq; @@ -550,7 +552,7 @@ static int hclge_tqps_update_stats(struct hnae3_handle *handle) queue = handle->kinfo.tqp[i]; tqp = container_of(queue, struct hclge_tqp, q); /* command : HCLGE_OPC_QUERY_IGU_STAT */ - hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_QUERY_RX_STATUS, + hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_QUERY_RX_STATS, true); desc[0].data[0] = cpu_to_le32((tqp->index & 0x1ff)); @@ -570,7 +572,7 @@ static int hclge_tqps_update_stats(struct hnae3_handle *handle) tqp = container_of(queue, struct hclge_tqp, q); /* command : HCLGE_OPC_QUERY_IGU_STAT */ hclge_cmd_setup_basic_desc(&desc[0], - HCLGE_OPC_QUERY_TX_STATUS, + HCLGE_OPC_QUERY_TX_STATS, true); desc[0].data[0] = cpu_to_le32((tqp->index & 0x1ff)); @@ -1361,10 +1363,8 @@ static int hclge_configure(struct hclge_dev *hdev) int ret; ret = hclge_get_cfg(hdev, &cfg); - if (ret) { - dev_err(&hdev->pdev->dev, "get mac mode error %d.\n", ret); + if (ret) return ret; - } hdev->num_vmdq_vport = cfg.vmdq_vport_num; hdev->base_tqp_pid = 0; @@ -1387,7 +1387,8 @@ static int hclge_configure(struct hclge_dev *hdev) ret = hclge_parse_speed(cfg.default_speed, &hdev->hw.mac.speed); if (ret) { - dev_err(&hdev->pdev->dev, "Get wrong speed ret=%d.\n", ret); + dev_err(&hdev->pdev->dev, "failed to parse speed %u, ret = %d\n", + cfg.default_speed, ret); return ret; } @@ -1429,26 +1430,17 @@ static int hclge_configure(struct hclge_dev *hdev) return ret; } -static int hclge_config_tso(struct hclge_dev *hdev, unsigned int tso_mss_min, - unsigned int tso_mss_max) +static int hclge_config_tso(struct hclge_dev *hdev, u16 tso_mss_min, + u16 tso_mss_max) { struct hclge_cfg_tso_status_cmd *req; struct hclge_desc desc; - u16 tso_mss; hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_TSO_GENERIC_CONFIG, false); req = (struct hclge_cfg_tso_status_cmd *)desc.data; - - tso_mss = 0; - hnae3_set_field(tso_mss, HCLGE_TSO_MSS_MIN_M, - HCLGE_TSO_MSS_MIN_S, tso_mss_min); - req->tso_mss_min = cpu_to_le16(tso_mss); - - tso_mss = 0; - hnae3_set_field(tso_mss, HCLGE_TSO_MSS_MIN_M, - HCLGE_TSO_MSS_MIN_S, tso_mss_max); - req->tso_mss_max = cpu_to_le16(tso_mss); + req->tso_mss_min = cpu_to_le16(tso_mss_min); + req->tso_mss_max = cpu_to_le16(tso_mss_max); return hclge_cmd_send(&hdev->hw, &desc, 1); } @@ -1465,7 +1457,7 @@ static int hclge_config_gro(struct hclge_dev *hdev, bool en) hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_GRO_GENERIC_CONFIG, false); req = (struct hclge_cfg_gro_status_cmd *)desc.data; - req->gro_en = cpu_to_le16(en ? 1 : 0); + req->gro_en = en ? 1 : 0; ret = hclge_cmd_send(&hdev->hw, &desc, 1); if (ret) @@ -1687,6 +1679,7 @@ static int hclge_alloc_vport(struct hclge_dev *hdev) INIT_LIST_HEAD(&vport->vlan_list); INIT_LIST_HEAD(&vport->uc_mac_list); INIT_LIST_HEAD(&vport->mc_mac_list); + spin_lock_init(&vport->mac_list_lock); if (i == 0) ret = hclge_vport_setup(vport, tqp_main_vport); @@ -2965,13 +2958,11 @@ static int hclge_set_vf_link_state(struct hnae3_handle *handle, int vf, static u32 hclge_check_event_cause(struct hclge_dev *hdev, u32 *clearval) { - u32 rst_src_reg, cmdq_src_reg, msix_src_reg; + u32 cmdq_src_reg, msix_src_reg; /* fetch the events from their corresponding regs */ - rst_src_reg = hclge_read_dev(&hdev->hw, HCLGE_MISC_VECTOR_INT_STS); cmdq_src_reg = hclge_read_dev(&hdev->hw, HCLGE_VECTOR0_CMDQ_SRC_REG); - msix_src_reg = hclge_read_dev(&hdev->hw, - HCLGE_VECTOR0_PF_OTHER_INT_STS_REG); + msix_src_reg = hclge_read_dev(&hdev->hw, HCLGE_MISC_VECTOR_INT_STS); /* Assumption: If by any chance reset and mailbox events are reported * together then we will only process reset event in this go and will @@ -2981,7 +2972,7 @@ static u32 hclge_check_event_cause(struct hclge_dev *hdev, u32 *clearval) * * check for vector0 reset event sources */ - if (BIT(HCLGE_VECTOR0_IMPRESET_INT_B) & rst_src_reg) { + if (BIT(HCLGE_VECTOR0_IMPRESET_INT_B) & msix_src_reg) { dev_info(&hdev->pdev->dev, "IMP reset interrupt\n"); set_bit(HNAE3_IMP_RESET, &hdev->reset_pending); set_bit(HCLGE_STATE_CMD_DISABLE, &hdev->state); @@ -2990,7 +2981,7 @@ static u32 hclge_check_event_cause(struct hclge_dev *hdev, u32 *clearval) return HCLGE_VECTOR0_EVENT_RST; } - if (BIT(HCLGE_VECTOR0_GLOBALRESET_INT_B) & rst_src_reg) { + if (BIT(HCLGE_VECTOR0_GLOBALRESET_INT_B) & msix_src_reg) { dev_info(&hdev->pdev->dev, "global reset interrupt\n"); set_bit(HCLGE_STATE_CMD_DISABLE, &hdev->state); set_bit(HNAE3_GLOBAL_RESET, &hdev->reset_pending); @@ -3480,7 +3471,7 @@ static enum hnae3_reset_type hclge_get_reset_level(struct hnae3_ae_dev *ae_dev, /* first, resolve any unknown reset type to the known type(s) */ if (test_bit(HNAE3_UNKNOWN_RESET, addr)) { u32 msix_sts_reg = hclge_read_dev(&hdev->hw, - HCLGE_VECTOR0_PF_OTHER_INT_STS_REG); + HCLGE_MISC_VECTOR_INT_STS); /* we will intentionally ignore any errors from this function * as we will end up in *some* reset request in any case */ @@ -3729,22 +3720,13 @@ static int hclge_reset_stack(struct hclge_dev *hdev) if (ret) return ret; - ret = hclge_notify_client(hdev, HNAE3_INIT_CLIENT); - if (ret) - return ret; - - return hclge_notify_client(hdev, HNAE3_RESTORE_CLIENT); + return hclge_notify_client(hdev, HNAE3_INIT_CLIENT); } static int hclge_reset_prepare(struct hclge_dev *hdev) { - struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); int ret; - /* Initialize ae_dev reset status as well, in case enet layer wants to - * know if device is undergoing reset - */ - ae_dev->reset_type = hdev->reset_type; hdev->rst_stats.reset_cnt++; /* perform reset of the stack & ae device for a client */ ret = hclge_notify_roce_client(hdev, HNAE3_DOWN_CLIENT); @@ -3780,11 +3762,6 @@ static int hclge_reset_rebuild(struct hclge_dev *hdev) hclge_clear_reset_cause(hdev); - ret = hclge_reset_prepare_up(hdev); - if (ret) - return ret; - - ret = hclge_notify_roce_client(hdev, HNAE3_INIT_CLIENT); /* ignore RoCE notify error if it fails HCLGE_RESET_MAX_FAIL_CNT - 1 * times @@ -3793,6 +3770,10 @@ static int hclge_reset_rebuild(struct hclge_dev *hdev) hdev->rst_stats.reset_fail_cnt < HCLGE_RESET_MAX_FAIL_CNT - 1) return ret; + ret = hclge_reset_prepare_up(hdev); + if (ret) + return ret; + rtnl_lock(); ret = hclge_notify_client(hdev, HNAE3_UP_CLIENT); rtnl_unlock(); @@ -3806,7 +3787,6 @@ static int hclge_reset_rebuild(struct hclge_dev *hdev) hdev->last_reset_time = jiffies; hdev->rst_stats.reset_fail_cnt = 0; hdev->rst_stats.reset_done_cnt++; - ae_dev->reset_type = HNAE3_NONE_RESET; clear_bit(HCLGE_STATE_RST_FAIL, &hdev->state); /* if default_reset_request has a higher level reset request, @@ -3973,6 +3953,8 @@ static void hclge_periodic_service_task(struct hclge_dev *hdev) * updated when it is triggered by mbx. */ hclge_update_link_status(hdev); + hclge_sync_mac_table(hdev); + hclge_sync_promisc_mode(hdev); if (time_is_after_jiffies(hdev->last_serv_processed + HZ)) { delta = jiffies - hdev->last_serv_processed; @@ -4722,7 +4704,8 @@ static int hclge_cmd_set_promisc_mode(struct hclge_dev *hdev, ret = hclge_cmd_send(&hdev->hw, &desc, 1); if (ret) dev_err(&hdev->pdev->dev, - "Set promisc mode fail, status is %d.\n", ret); + "failed to set vport %d promisc mode, ret = %d.\n", + param->vf_id, ret); return ret; } @@ -4772,6 +4755,14 @@ static int hclge_set_promisc_mode(struct hnae3_handle *handle, bool en_uc_pmc, en_bc_pmc); } +static void hclge_request_update_promisc_mode(struct hnae3_handle *handle) +{ + struct hclge_vport *vport = hclge_get_vport(handle); + struct hclge_dev *hdev = vport->back; + + set_bit(HCLGE_STATE_PROMISC_CHANGED, &hdev->state); +} + static int hclge_get_fd_mode(struct hclge_dev *hdev, u8 *fd_mode) { struct hclge_get_fd_mode_cmd *req; @@ -4822,7 +4813,8 @@ static int hclge_get_fd_allocation(struct hclge_dev *hdev, return ret; } -static int hclge_set_fd_key_config(struct hclge_dev *hdev, int stage_num) +static int hclge_set_fd_key_config(struct hclge_dev *hdev, + enum HCLGE_FD_STAGE stage_num) { struct hclge_set_fd_key_config_cmd *req; struct hclge_fd_key_cfg *stage; @@ -4876,9 +4868,6 @@ static int hclge_init_fd_config(struct hclge_dev *hdev) return -EOPNOTSUPP; } - hdev->fd_cfg.proto_support = - TCP_V4_FLOW | UDP_V4_FLOW | SCTP_V4_FLOW | TCP_V6_FLOW | - UDP_V6_FLOW | SCTP_V6_FLOW | IPV4_USER_FLOW | IPV6_USER_FLOW; key_cfg = &hdev->fd_cfg.key_cfg[HCLGE_FD_STAGE_1]; key_cfg->key_sel = HCLGE_FD_KEY_BASE_ON_TUPLE, key_cfg->inner_sipv6_word_en = LOW_2_WORDS; @@ -4892,11 +4881,9 @@ static int hclge_init_fd_config(struct hclge_dev *hdev) BIT(INNER_SRC_PORT) | BIT(INNER_DST_PORT); /* If use max 400bit key, we can support tuples for ether type */ - if (hdev->fd_cfg.max_key_length == MAX_KEY_LENGTH) { - hdev->fd_cfg.proto_support |= ETHER_FLOW; + if (hdev->fd_cfg.fd_mode == HCLGE_FD_MODE_DEPTH_2K_WIDTH_400B_STAGE_1) key_cfg->tuple_active |= BIT(INNER_DST_MAC) | BIT(INNER_SRC_MAC); - } /* roce_type is used to filter roce frames * dst_vport is used to specify the rule @@ -5006,8 +4993,6 @@ static bool hclge_fd_convert_tuple(u32 tuple_bit, u8 *key_x, u8 *key_y, return true; switch (tuple_bit) { - case 0: - return false; case BIT(INNER_DST_MAC): for (i = 0; i < ETH_ALEN; i++) { calc_x(key_x[ETH_ALEN - 1 - i], rule->tuples.dst_mac[i], @@ -5165,9 +5150,10 @@ static int hclge_config_key(struct hclge_dev *hdev, u8 stage, struct hclge_fd_key_cfg *key_cfg = &hdev->fd_cfg.key_cfg[stage]; u8 key_x[MAX_KEY_BYTES], key_y[MAX_KEY_BYTES]; u8 *cur_key_x, *cur_key_y; - unsigned int i; - int ret, tuple_size; u8 meta_data_region; + u8 tuple_size; + int ret; + u32 i; memset(key_x, 0, sizeof(key_x)); memset(key_y, 0, sizeof(key_y)); @@ -5244,172 +5230,255 @@ static int hclge_config_action(struct hclge_dev *hdev, u8 stage, return hclge_fd_ad_config(hdev, stage, ad_data.ad_id, &ad_data); } -static int hclge_fd_check_spec(struct hclge_dev *hdev, - struct ethtool_rx_flow_spec *fs, u32 *unused) +static int hclge_fd_check_tcpip4_tuple(struct ethtool_tcpip4_spec *spec, + u32 *unused_tuple) { - struct ethtool_tcpip4_spec *tcp_ip4_spec; - struct ethtool_usrip4_spec *usr_ip4_spec; - struct ethtool_tcpip6_spec *tcp_ip6_spec; - struct ethtool_usrip6_spec *usr_ip6_spec; - struct ethhdr *ether_spec; - - if (fs->location >= hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1]) + if (!spec || !unused_tuple) return -EINVAL; - if (!(fs->flow_type & hdev->fd_cfg.proto_support)) - return -EOPNOTSUPP; - - if ((fs->flow_type & FLOW_EXT) && - (fs->h_ext.data[0] != 0 || fs->h_ext.data[1] != 0)) { - dev_err(&hdev->pdev->dev, "user-def bytes are not supported\n"); - return -EOPNOTSUPP; - } + *unused_tuple |= BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC); - switch (fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT)) { - case SCTP_V4_FLOW: - case TCP_V4_FLOW: - case UDP_V4_FLOW: - tcp_ip4_spec = &fs->h_u.tcp_ip4_spec; - *unused |= BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC); + if (!spec->ip4src) + *unused_tuple |= BIT(INNER_SRC_IP); - if (!tcp_ip4_spec->ip4src) - *unused |= BIT(INNER_SRC_IP); + if (!spec->ip4dst) + *unused_tuple |= BIT(INNER_DST_IP); - if (!tcp_ip4_spec->ip4dst) - *unused |= BIT(INNER_DST_IP); + if (!spec->psrc) + *unused_tuple |= BIT(INNER_SRC_PORT); - if (!tcp_ip4_spec->psrc) - *unused |= BIT(INNER_SRC_PORT); + if (!spec->pdst) + *unused_tuple |= BIT(INNER_DST_PORT); - if (!tcp_ip4_spec->pdst) - *unused |= BIT(INNER_DST_PORT); + if (!spec->tos) + *unused_tuple |= BIT(INNER_IP_TOS); - if (!tcp_ip4_spec->tos) - *unused |= BIT(INNER_IP_TOS); + return 0; +} - break; - case IP_USER_FLOW: - usr_ip4_spec = &fs->h_u.usr_ip4_spec; - *unused |= BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC) | - BIT(INNER_SRC_PORT) | BIT(INNER_DST_PORT); +static int hclge_fd_check_ip4_tuple(struct ethtool_usrip4_spec *spec, + u32 *unused_tuple) +{ + if (!spec || !unused_tuple) + return -EINVAL; - if (!usr_ip4_spec->ip4src) - *unused |= BIT(INNER_SRC_IP); + *unused_tuple |= BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC) | + BIT(INNER_SRC_PORT) | BIT(INNER_DST_PORT); - if (!usr_ip4_spec->ip4dst) - *unused |= BIT(INNER_DST_IP); + if (!spec->ip4src) + *unused_tuple |= BIT(INNER_SRC_IP); - if (!usr_ip4_spec->tos) - *unused |= BIT(INNER_IP_TOS); + if (!spec->ip4dst) + *unused_tuple |= BIT(INNER_DST_IP); - if (!usr_ip4_spec->proto) - *unused |= BIT(INNER_IP_PROTO); + if (!spec->tos) + *unused_tuple |= BIT(INNER_IP_TOS); - if (usr_ip4_spec->l4_4_bytes) - return -EOPNOTSUPP; + if (!spec->proto) + *unused_tuple |= BIT(INNER_IP_PROTO); - if (usr_ip4_spec->ip_ver != ETH_RX_NFC_IP4) - return -EOPNOTSUPP; + if (spec->l4_4_bytes) + return -EOPNOTSUPP; - break; - case SCTP_V6_FLOW: - case TCP_V6_FLOW: - case UDP_V6_FLOW: - tcp_ip6_spec = &fs->h_u.tcp_ip6_spec; - *unused |= BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC) | - BIT(INNER_IP_TOS); + if (spec->ip_ver != ETH_RX_NFC_IP4) + return -EOPNOTSUPP; - /* check whether src/dst ip address used */ - if (!tcp_ip6_spec->ip6src[0] && !tcp_ip6_spec->ip6src[1] && - !tcp_ip6_spec->ip6src[2] && !tcp_ip6_spec->ip6src[3]) - *unused |= BIT(INNER_SRC_IP); + return 0; +} - if (!tcp_ip6_spec->ip6dst[0] && !tcp_ip6_spec->ip6dst[1] && - !tcp_ip6_spec->ip6dst[2] && !tcp_ip6_spec->ip6dst[3]) - *unused |= BIT(INNER_DST_IP); +static int hclge_fd_check_tcpip6_tuple(struct ethtool_tcpip6_spec *spec, + u32 *unused_tuple) +{ + if (!spec || !unused_tuple) + return -EINVAL; - if (!tcp_ip6_spec->psrc) - *unused |= BIT(INNER_SRC_PORT); + *unused_tuple |= BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC) | + BIT(INNER_IP_TOS); - if (!tcp_ip6_spec->pdst) - *unused |= BIT(INNER_DST_PORT); + /* check whether src/dst ip address used */ + if (!spec->ip6src[0] && !spec->ip6src[1] && + !spec->ip6src[2] && !spec->ip6src[3]) + *unused_tuple |= BIT(INNER_SRC_IP); - if (tcp_ip6_spec->tclass) - return -EOPNOTSUPP; + if (!spec->ip6dst[0] && !spec->ip6dst[1] && + !spec->ip6dst[2] && !spec->ip6dst[3]) + *unused_tuple |= BIT(INNER_DST_IP); - break; - case IPV6_USER_FLOW: - usr_ip6_spec = &fs->h_u.usr_ip6_spec; - *unused |= BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC) | - BIT(INNER_IP_TOS) | BIT(INNER_SRC_PORT) | - BIT(INNER_DST_PORT); + if (!spec->psrc) + *unused_tuple |= BIT(INNER_SRC_PORT); - /* check whether src/dst ip address used */ - if (!usr_ip6_spec->ip6src[0] && !usr_ip6_spec->ip6src[1] && - !usr_ip6_spec->ip6src[2] && !usr_ip6_spec->ip6src[3]) - *unused |= BIT(INNER_SRC_IP); + if (!spec->pdst) + *unused_tuple |= BIT(INNER_DST_PORT); - if (!usr_ip6_spec->ip6dst[0] && !usr_ip6_spec->ip6dst[1] && - !usr_ip6_spec->ip6dst[2] && !usr_ip6_spec->ip6dst[3]) - *unused |= BIT(INNER_DST_IP); + if (spec->tclass) + return -EOPNOTSUPP; - if (!usr_ip6_spec->l4_proto) - *unused |= BIT(INNER_IP_PROTO); + return 0; +} - if (usr_ip6_spec->tclass) - return -EOPNOTSUPP; +static int hclge_fd_check_ip6_tuple(struct ethtool_usrip6_spec *spec, + u32 *unused_tuple) +{ + if (!spec || !unused_tuple) + return -EINVAL; - if (usr_ip6_spec->l4_4_bytes) - return -EOPNOTSUPP; + *unused_tuple |= BIT(INNER_SRC_MAC) | BIT(INNER_DST_MAC) | + BIT(INNER_IP_TOS) | BIT(INNER_SRC_PORT) | BIT(INNER_DST_PORT); - break; - case ETHER_FLOW: - ether_spec = &fs->h_u.ether_spec; - *unused |= BIT(INNER_SRC_IP) | BIT(INNER_DST_IP) | - BIT(INNER_SRC_PORT) | BIT(INNER_DST_PORT) | - BIT(INNER_IP_TOS) | BIT(INNER_IP_PROTO); + /* check whether src/dst ip address used */ + if (!spec->ip6src[0] && !spec->ip6src[1] && + !spec->ip6src[2] && !spec->ip6src[3]) + *unused_tuple |= BIT(INNER_SRC_IP); - if (is_zero_ether_addr(ether_spec->h_source)) - *unused |= BIT(INNER_SRC_MAC); + if (!spec->ip6dst[0] && !spec->ip6dst[1] && + !spec->ip6dst[2] && !spec->ip6dst[3]) + *unused_tuple |= BIT(INNER_DST_IP); - if (is_zero_ether_addr(ether_spec->h_dest)) - *unused |= BIT(INNER_DST_MAC); + if (!spec->l4_proto) + *unused_tuple |= BIT(INNER_IP_PROTO); - if (!ether_spec->h_proto) - *unused |= BIT(INNER_ETH_TYPE); + if (spec->tclass) + return -EOPNOTSUPP; - break; - default: + if (spec->l4_4_bytes) return -EOPNOTSUPP; - } - if ((fs->flow_type & FLOW_EXT)) { - if (fs->h_ext.vlan_etype) + return 0; +} + +static int hclge_fd_check_ether_tuple(struct ethhdr *spec, u32 *unused_tuple) +{ + if (!spec || !unused_tuple) + return -EINVAL; + + *unused_tuple |= BIT(INNER_SRC_IP) | BIT(INNER_DST_IP) | + BIT(INNER_SRC_PORT) | BIT(INNER_DST_PORT) | + BIT(INNER_IP_TOS) | BIT(INNER_IP_PROTO); + + if (is_zero_ether_addr(spec->h_source)) + *unused_tuple |= BIT(INNER_SRC_MAC); + + if (is_zero_ether_addr(spec->h_dest)) + *unused_tuple |= BIT(INNER_DST_MAC); + + if (!spec->h_proto) + *unused_tuple |= BIT(INNER_ETH_TYPE); + + return 0; +} + +static int hclge_fd_check_ext_tuple(struct hclge_dev *hdev, + struct ethtool_rx_flow_spec *fs, + u32 *unused_tuple) +{ + if (fs->flow_type & FLOW_EXT) { + if (fs->h_ext.vlan_etype) { + dev_err(&hdev->pdev->dev, "vlan-etype is not supported!\n"); return -EOPNOTSUPP; + } + if (!fs->h_ext.vlan_tci) - *unused |= BIT(INNER_VLAN_TAG_FST); + *unused_tuple |= BIT(INNER_VLAN_TAG_FST); - if (fs->m_ext.vlan_tci) { - if (be16_to_cpu(fs->h_ext.vlan_tci) >= VLAN_N_VID) - return -EINVAL; + if (fs->m_ext.vlan_tci && + be16_to_cpu(fs->h_ext.vlan_tci) >= VLAN_N_VID) { + dev_err(&hdev->pdev->dev, + "failed to config vlan_tci, invalid vlan_tci: %u, max is %u.\n", + ntohs(fs->h_ext.vlan_tci), VLAN_N_VID - 1); + return -EINVAL; } } else { - *unused |= BIT(INNER_VLAN_TAG_FST); + *unused_tuple |= BIT(INNER_VLAN_TAG_FST); } if (fs->flow_type & FLOW_MAC_EXT) { - if (!(hdev->fd_cfg.proto_support & ETHER_FLOW)) + if (hdev->fd_cfg.fd_mode != + HCLGE_FD_MODE_DEPTH_2K_WIDTH_400B_STAGE_1) { + dev_err(&hdev->pdev->dev, + "FLOW_MAC_EXT is not supported in current fd mode!\n"); return -EOPNOTSUPP; + } if (is_zero_ether_addr(fs->h_ext.h_dest)) - *unused |= BIT(INNER_DST_MAC); + *unused_tuple |= BIT(INNER_DST_MAC); else - *unused &= ~(BIT(INNER_DST_MAC)); + *unused_tuple &= ~BIT(INNER_DST_MAC); } return 0; } +static int hclge_fd_check_spec(struct hclge_dev *hdev, + struct ethtool_rx_flow_spec *fs, + u32 *unused_tuple) +{ + u32 flow_type; + int ret; + + if (fs->location >= hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1]) { + dev_err(&hdev->pdev->dev, + "failed to config fd rules, invalid rule location: %u, max is %u\n.", + fs->location, + hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1] - 1); + return -EINVAL; + } + + if ((fs->flow_type & FLOW_EXT) && + (fs->h_ext.data[0] != 0 || fs->h_ext.data[1] != 0)) { + dev_err(&hdev->pdev->dev, "user-def bytes are not supported\n"); + return -EOPNOTSUPP; + } + + flow_type = fs->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT); + switch (flow_type) { + case SCTP_V4_FLOW: + case TCP_V4_FLOW: + case UDP_V4_FLOW: + ret = hclge_fd_check_tcpip4_tuple(&fs->h_u.tcp_ip4_spec, + unused_tuple); + break; + case IP_USER_FLOW: + ret = hclge_fd_check_ip4_tuple(&fs->h_u.usr_ip4_spec, + unused_tuple); + break; + case SCTP_V6_FLOW: + case TCP_V6_FLOW: + case UDP_V6_FLOW: + ret = hclge_fd_check_tcpip6_tuple(&fs->h_u.tcp_ip6_spec, + unused_tuple); + break; + case IPV6_USER_FLOW: + ret = hclge_fd_check_ip6_tuple(&fs->h_u.usr_ip6_spec, + unused_tuple); + break; + case ETHER_FLOW: + if (hdev->fd_cfg.fd_mode != + HCLGE_FD_MODE_DEPTH_2K_WIDTH_400B_STAGE_1) { + dev_err(&hdev->pdev->dev, + "ETHER_FLOW is not supported in current fd mode!\n"); + return -EOPNOTSUPP; + } + + ret = hclge_fd_check_ether_tuple(&fs->h_u.ether_spec, + unused_tuple); + break; + default: + dev_err(&hdev->pdev->dev, + "unsupported protocol type, protocol type = %#x\n", + flow_type); + return -EOPNOTSUPP; + } + + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to check flow union tuple, ret = %d\n", + ret); + return ret; + } + + return hclge_fd_check_ext_tuple(hdev, fs, unused_tuple); +} + static bool hclge_fd_rule_exist(struct hclge_dev *hdev, u16 location) { struct hclge_fd_rule *rule = NULL; @@ -5618,7 +5687,7 @@ static int hclge_fd_get_tuple(struct hclge_dev *hdev, break; } - if ((fs->flow_type & FLOW_EXT)) { + if (fs->flow_type & FLOW_EXT) { rule->tuples.vlan_tag1 = be16_to_cpu(fs->h_ext.vlan_tci); rule->tuples_mask.vlan_tag1 = be16_to_cpu(fs->m_ext.vlan_tci); } @@ -5673,22 +5742,23 @@ static int hclge_add_fd_entry(struct hnae3_handle *handle, u8 action; int ret; - if (!hnae3_dev_fd_supported(hdev)) + if (!hnae3_dev_fd_supported(hdev)) { + dev_err(&hdev->pdev->dev, + "flow table director is not supported\n"); return -EOPNOTSUPP; + } if (!hdev->fd_en) { - dev_warn(&hdev->pdev->dev, - "Please enable flow director first\n"); + dev_err(&hdev->pdev->dev, + "please enable flow director first\n"); return -EOPNOTSUPP; } fs = (struct ethtool_rx_flow_spec *)&cmd->fs; ret = hclge_fd_check_spec(hdev, fs, &unused); - if (ret) { - dev_err(&hdev->pdev->dev, "Check fd spec failed\n"); + if (ret) return ret; - } if (fs->ring_cookie == RX_CLS_FLOW_DISC) { action = HCLGE_FD_ACTION_DROP_PACKET; @@ -5729,7 +5799,6 @@ static int hclge_add_fd_entry(struct hnae3_handle *handle, } rule->flow_type = fs->flow_type; - rule->location = fs->location; rule->unused_tuple = unused; rule->vf_id = dst_vport_id; @@ -5877,6 +5946,149 @@ static int hclge_get_fd_rule_cnt(struct hnae3_handle *handle, return 0; } +static void hclge_fd_get_tcpip4_info(struct hclge_fd_rule *rule, + struct ethtool_tcpip4_spec *spec, + struct ethtool_tcpip4_spec *spec_mask) +{ + spec->ip4src = cpu_to_be32(rule->tuples.src_ip[IPV4_INDEX]); + spec_mask->ip4src = rule->unused_tuple & BIT(INNER_SRC_IP) ? + 0 : cpu_to_be32(rule->tuples_mask.src_ip[IPV4_INDEX]); + + spec->ip4dst = cpu_to_be32(rule->tuples.dst_ip[IPV4_INDEX]); + spec_mask->ip4dst = rule->unused_tuple & BIT(INNER_DST_IP) ? + 0 : cpu_to_be32(rule->tuples_mask.dst_ip[IPV4_INDEX]); + + spec->psrc = cpu_to_be16(rule->tuples.src_port); + spec_mask->psrc = rule->unused_tuple & BIT(INNER_SRC_PORT) ? + 0 : cpu_to_be16(rule->tuples_mask.src_port); + + spec->pdst = cpu_to_be16(rule->tuples.dst_port); + spec_mask->pdst = rule->unused_tuple & BIT(INNER_DST_PORT) ? + 0 : cpu_to_be16(rule->tuples_mask.dst_port); + + spec->tos = rule->tuples.ip_tos; + spec_mask->tos = rule->unused_tuple & BIT(INNER_IP_TOS) ? + 0 : rule->tuples_mask.ip_tos; +} + +static void hclge_fd_get_ip4_info(struct hclge_fd_rule *rule, + struct ethtool_usrip4_spec *spec, + struct ethtool_usrip4_spec *spec_mask) +{ + spec->ip4src = cpu_to_be32(rule->tuples.src_ip[IPV4_INDEX]); + spec_mask->ip4src = rule->unused_tuple & BIT(INNER_SRC_IP) ? + 0 : cpu_to_be32(rule->tuples_mask.src_ip[IPV4_INDEX]); + + spec->ip4dst = cpu_to_be32(rule->tuples.dst_ip[IPV4_INDEX]); + spec_mask->ip4dst = rule->unused_tuple & BIT(INNER_DST_IP) ? + 0 : cpu_to_be32(rule->tuples_mask.dst_ip[IPV4_INDEX]); + + spec->tos = rule->tuples.ip_tos; + spec_mask->tos = rule->unused_tuple & BIT(INNER_IP_TOS) ? + 0 : rule->tuples_mask.ip_tos; + + spec->proto = rule->tuples.ip_proto; + spec_mask->proto = rule->unused_tuple & BIT(INNER_IP_PROTO) ? + 0 : rule->tuples_mask.ip_proto; + + spec->ip_ver = ETH_RX_NFC_IP4; +} + +static void hclge_fd_get_tcpip6_info(struct hclge_fd_rule *rule, + struct ethtool_tcpip6_spec *spec, + struct ethtool_tcpip6_spec *spec_mask) +{ + cpu_to_be32_array(spec->ip6src, + rule->tuples.src_ip, IPV6_SIZE); + cpu_to_be32_array(spec->ip6dst, + rule->tuples.dst_ip, IPV6_SIZE); + if (rule->unused_tuple & BIT(INNER_SRC_IP)) + memset(spec_mask->ip6src, 0, sizeof(spec_mask->ip6src)); + else + cpu_to_be32_array(spec_mask->ip6src, rule->tuples_mask.src_ip, + IPV6_SIZE); + + if (rule->unused_tuple & BIT(INNER_DST_IP)) + memset(spec_mask->ip6dst, 0, sizeof(spec_mask->ip6dst)); + else + cpu_to_be32_array(spec_mask->ip6dst, rule->tuples_mask.dst_ip, + IPV6_SIZE); + + spec->psrc = cpu_to_be16(rule->tuples.src_port); + spec_mask->psrc = rule->unused_tuple & BIT(INNER_SRC_PORT) ? + 0 : cpu_to_be16(rule->tuples_mask.src_port); + + spec->pdst = cpu_to_be16(rule->tuples.dst_port); + spec_mask->pdst = rule->unused_tuple & BIT(INNER_DST_PORT) ? + 0 : cpu_to_be16(rule->tuples_mask.dst_port); +} + +static void hclge_fd_get_ip6_info(struct hclge_fd_rule *rule, + struct ethtool_usrip6_spec *spec, + struct ethtool_usrip6_spec *spec_mask) +{ + cpu_to_be32_array(spec->ip6src, rule->tuples.src_ip, IPV6_SIZE); + cpu_to_be32_array(spec->ip6dst, rule->tuples.dst_ip, IPV6_SIZE); + if (rule->unused_tuple & BIT(INNER_SRC_IP)) + memset(spec_mask->ip6src, 0, sizeof(spec_mask->ip6src)); + else + cpu_to_be32_array(spec_mask->ip6src, + rule->tuples_mask.src_ip, IPV6_SIZE); + + if (rule->unused_tuple & BIT(INNER_DST_IP)) + memset(spec_mask->ip6dst, 0, sizeof(spec_mask->ip6dst)); + else + cpu_to_be32_array(spec_mask->ip6dst, + rule->tuples_mask.dst_ip, IPV6_SIZE); + + spec->l4_proto = rule->tuples.ip_proto; + spec_mask->l4_proto = rule->unused_tuple & BIT(INNER_IP_PROTO) ? + 0 : rule->tuples_mask.ip_proto; +} + +static void hclge_fd_get_ether_info(struct hclge_fd_rule *rule, + struct ethhdr *spec, + struct ethhdr *spec_mask) +{ + ether_addr_copy(spec->h_source, rule->tuples.src_mac); + ether_addr_copy(spec->h_dest, rule->tuples.dst_mac); + + if (rule->unused_tuple & BIT(INNER_SRC_MAC)) + eth_zero_addr(spec_mask->h_source); + else + ether_addr_copy(spec_mask->h_source, rule->tuples_mask.src_mac); + + if (rule->unused_tuple & BIT(INNER_DST_MAC)) + eth_zero_addr(spec_mask->h_dest); + else + ether_addr_copy(spec_mask->h_dest, rule->tuples_mask.dst_mac); + + spec->h_proto = cpu_to_be16(rule->tuples.ether_proto); + spec_mask->h_proto = rule->unused_tuple & BIT(INNER_ETH_TYPE) ? + 0 : cpu_to_be16(rule->tuples_mask.ether_proto); +} + +static void hclge_fd_get_ext_info(struct ethtool_rx_flow_spec *fs, + struct hclge_fd_rule *rule) +{ + if (fs->flow_type & FLOW_EXT) { + fs->h_ext.vlan_tci = cpu_to_be16(rule->tuples.vlan_tag1); + fs->m_ext.vlan_tci = + rule->unused_tuple & BIT(INNER_VLAN_TAG_FST) ? + cpu_to_be16(VLAN_VID_MASK) : + cpu_to_be16(rule->tuples_mask.vlan_tag1); + } + + if (fs->flow_type & FLOW_MAC_EXT) { + ether_addr_copy(fs->h_ext.h_dest, rule->tuples.dst_mac); + if (rule->unused_tuple & BIT(INNER_DST_MAC)) + eth_zero_addr(fs->m_u.ether_spec.h_dest); + else + ether_addr_copy(fs->m_u.ether_spec.h_dest, + rule->tuples_mask.dst_mac); + } +} + static int hclge_get_fd_rule_info(struct hnae3_handle *handle, struct ethtool_rxnfc *cmd) { @@ -5909,162 +6121,34 @@ static int hclge_get_fd_rule_info(struct hnae3_handle *handle, case SCTP_V4_FLOW: case TCP_V4_FLOW: case UDP_V4_FLOW: - fs->h_u.tcp_ip4_spec.ip4src = - cpu_to_be32(rule->tuples.src_ip[IPV4_INDEX]); - fs->m_u.tcp_ip4_spec.ip4src = - rule->unused_tuple & BIT(INNER_SRC_IP) ? - 0 : cpu_to_be32(rule->tuples_mask.src_ip[IPV4_INDEX]); - - fs->h_u.tcp_ip4_spec.ip4dst = - cpu_to_be32(rule->tuples.dst_ip[IPV4_INDEX]); - fs->m_u.tcp_ip4_spec.ip4dst = - rule->unused_tuple & BIT(INNER_DST_IP) ? - 0 : cpu_to_be32(rule->tuples_mask.dst_ip[IPV4_INDEX]); - - fs->h_u.tcp_ip4_spec.psrc = cpu_to_be16(rule->tuples.src_port); - fs->m_u.tcp_ip4_spec.psrc = - rule->unused_tuple & BIT(INNER_SRC_PORT) ? - 0 : cpu_to_be16(rule->tuples_mask.src_port); - - fs->h_u.tcp_ip4_spec.pdst = cpu_to_be16(rule->tuples.dst_port); - fs->m_u.tcp_ip4_spec.pdst = - rule->unused_tuple & BIT(INNER_DST_PORT) ? - 0 : cpu_to_be16(rule->tuples_mask.dst_port); - - fs->h_u.tcp_ip4_spec.tos = rule->tuples.ip_tos; - fs->m_u.tcp_ip4_spec.tos = - rule->unused_tuple & BIT(INNER_IP_TOS) ? - 0 : rule->tuples_mask.ip_tos; - + hclge_fd_get_tcpip4_info(rule, &fs->h_u.tcp_ip4_spec, + &fs->m_u.tcp_ip4_spec); break; case IP_USER_FLOW: - fs->h_u.usr_ip4_spec.ip4src = - cpu_to_be32(rule->tuples.src_ip[IPV4_INDEX]); - fs->m_u.tcp_ip4_spec.ip4src = - rule->unused_tuple & BIT(INNER_SRC_IP) ? - 0 : cpu_to_be32(rule->tuples_mask.src_ip[IPV4_INDEX]); - - fs->h_u.usr_ip4_spec.ip4dst = - cpu_to_be32(rule->tuples.dst_ip[IPV4_INDEX]); - fs->m_u.usr_ip4_spec.ip4dst = - rule->unused_tuple & BIT(INNER_DST_IP) ? - 0 : cpu_to_be32(rule->tuples_mask.dst_ip[IPV4_INDEX]); - - fs->h_u.usr_ip4_spec.tos = rule->tuples.ip_tos; - fs->m_u.usr_ip4_spec.tos = - rule->unused_tuple & BIT(INNER_IP_TOS) ? - 0 : rule->tuples_mask.ip_tos; - - fs->h_u.usr_ip4_spec.proto = rule->tuples.ip_proto; - fs->m_u.usr_ip4_spec.proto = - rule->unused_tuple & BIT(INNER_IP_PROTO) ? - 0 : rule->tuples_mask.ip_proto; - - fs->h_u.usr_ip4_spec.ip_ver = ETH_RX_NFC_IP4; - + hclge_fd_get_ip4_info(rule, &fs->h_u.usr_ip4_spec, + &fs->m_u.usr_ip4_spec); break; case SCTP_V6_FLOW: case TCP_V6_FLOW: case UDP_V6_FLOW: - cpu_to_be32_array(fs->h_u.tcp_ip6_spec.ip6src, - rule->tuples.src_ip, IPV6_SIZE); - if (rule->unused_tuple & BIT(INNER_SRC_IP)) - memset(fs->m_u.tcp_ip6_spec.ip6src, 0, - sizeof(int) * IPV6_SIZE); - else - cpu_to_be32_array(fs->m_u.tcp_ip6_spec.ip6src, - rule->tuples_mask.src_ip, IPV6_SIZE); - - cpu_to_be32_array(fs->h_u.tcp_ip6_spec.ip6dst, - rule->tuples.dst_ip, IPV6_SIZE); - if (rule->unused_tuple & BIT(INNER_DST_IP)) - memset(fs->m_u.tcp_ip6_spec.ip6dst, 0, - sizeof(int) * IPV6_SIZE); - else - cpu_to_be32_array(fs->m_u.tcp_ip6_spec.ip6dst, - rule->tuples_mask.dst_ip, IPV6_SIZE); - - fs->h_u.tcp_ip6_spec.psrc = cpu_to_be16(rule->tuples.src_port); - fs->m_u.tcp_ip6_spec.psrc = - rule->unused_tuple & BIT(INNER_SRC_PORT) ? - 0 : cpu_to_be16(rule->tuples_mask.src_port); - - fs->h_u.tcp_ip6_spec.pdst = cpu_to_be16(rule->tuples.dst_port); - fs->m_u.tcp_ip6_spec.pdst = - rule->unused_tuple & BIT(INNER_DST_PORT) ? - 0 : cpu_to_be16(rule->tuples_mask.dst_port); - + hclge_fd_get_tcpip6_info(rule, &fs->h_u.tcp_ip6_spec, + &fs->m_u.tcp_ip6_spec); break; case IPV6_USER_FLOW: - cpu_to_be32_array(fs->h_u.usr_ip6_spec.ip6src, - rule->tuples.src_ip, IPV6_SIZE); - if (rule->unused_tuple & BIT(INNER_SRC_IP)) - memset(fs->m_u.usr_ip6_spec.ip6src, 0, - sizeof(int) * IPV6_SIZE); - else - cpu_to_be32_array(fs->m_u.usr_ip6_spec.ip6src, - rule->tuples_mask.src_ip, IPV6_SIZE); - - cpu_to_be32_array(fs->h_u.usr_ip6_spec.ip6dst, - rule->tuples.dst_ip, IPV6_SIZE); - if (rule->unused_tuple & BIT(INNER_DST_IP)) - memset(fs->m_u.usr_ip6_spec.ip6dst, 0, - sizeof(int) * IPV6_SIZE); - else - cpu_to_be32_array(fs->m_u.usr_ip6_spec.ip6dst, - rule->tuples_mask.dst_ip, IPV6_SIZE); - - fs->h_u.usr_ip6_spec.l4_proto = rule->tuples.ip_proto; - fs->m_u.usr_ip6_spec.l4_proto = - rule->unused_tuple & BIT(INNER_IP_PROTO) ? - 0 : rule->tuples_mask.ip_proto; - - break; - case ETHER_FLOW: - ether_addr_copy(fs->h_u.ether_spec.h_source, - rule->tuples.src_mac); - if (rule->unused_tuple & BIT(INNER_SRC_MAC)) - eth_zero_addr(fs->m_u.ether_spec.h_source); - else - ether_addr_copy(fs->m_u.ether_spec.h_source, - rule->tuples_mask.src_mac); - - ether_addr_copy(fs->h_u.ether_spec.h_dest, - rule->tuples.dst_mac); - if (rule->unused_tuple & BIT(INNER_DST_MAC)) - eth_zero_addr(fs->m_u.ether_spec.h_dest); - else - ether_addr_copy(fs->m_u.ether_spec.h_dest, - rule->tuples_mask.dst_mac); - - fs->h_u.ether_spec.h_proto = - cpu_to_be16(rule->tuples.ether_proto); - fs->m_u.ether_spec.h_proto = - rule->unused_tuple & BIT(INNER_ETH_TYPE) ? - 0 : cpu_to_be16(rule->tuples_mask.ether_proto); - + hclge_fd_get_ip6_info(rule, &fs->h_u.usr_ip6_spec, + &fs->m_u.usr_ip6_spec); break; + /* The flow type of fd rule has been checked before adding in to rule + * list. As other flow types have been handled, it must be ETHER_FLOW + * for the default case + */ default: - spin_unlock_bh(&hdev->fd_rule_lock); - return -EOPNOTSUPP; - } - - if (fs->flow_type & FLOW_EXT) { - fs->h_ext.vlan_tci = cpu_to_be16(rule->tuples.vlan_tag1); - fs->m_ext.vlan_tci = - rule->unused_tuple & BIT(INNER_VLAN_TAG_FST) ? - cpu_to_be16(VLAN_VID_MASK) : - cpu_to_be16(rule->tuples_mask.vlan_tag1); + hclge_fd_get_ether_info(rule, &fs->h_u.ether_spec, + &fs->m_u.ether_spec); + break; } - if (fs->flow_type & FLOW_MAC_EXT) { - ether_addr_copy(fs->h_ext.h_dest, rule->tuples.dst_mac); - if (rule->unused_tuple & BIT(INNER_DST_MAC)) - eth_zero_addr(fs->m_u.ether_spec.h_dest); - else - ether_addr_copy(fs->m_u.ether_spec.h_dest, - rule->tuples_mask.dst_mac); - } + hclge_fd_get_ext_info(fs, rule); if (rule->action == HCLGE_FD_ACTION_DROP_PACKET) { fs->ring_cookie = RX_CLS_FLOW_DISC; @@ -6202,7 +6286,6 @@ static int hclge_add_fd_entry_by_arfs(struct hnae3_handle *handle, u16 queue_id, */ if (hdev->fd_active_type == HCLGE_FD_EP_ACTIVE) { spin_unlock_bh(&hdev->fd_rule_lock); - return -EOPNOTSUPP; } @@ -6216,14 +6299,12 @@ static int hclge_add_fd_entry_by_arfs(struct hnae3_handle *handle, u16 queue_id, bit_id = find_first_zero_bit(hdev->fd_bmap, MAX_FD_FILTER_NUM); if (bit_id >= hdev->fd_cfg.rule_num[HCLGE_FD_STAGE_1]) { spin_unlock_bh(&hdev->fd_rule_lock); - return -ENOSPC; } rule = kzalloc(sizeof(*rule), GFP_ATOMIC); if (!rule) { spin_unlock_bh(&hdev->fd_rule_lock); - return -ENOMEM; } @@ -6310,6 +6391,14 @@ static bool hclge_get_hw_reset_stat(struct hnae3_handle *handle) hclge_read_dev(&hdev->hw, HCLGE_FUN_RST_ING); } +static bool hclge_get_cmdq_stat(struct hnae3_handle *handle) +{ + struct hclge_vport *vport = hclge_get_vport(handle); + struct hclge_dev *hdev = vport->back; + + return test_bit(HCLGE_STATE_CMD_DISABLE, &hdev->state); +} + static bool hclge_ae_dev_resetting(struct hnae3_handle *handle) { struct hclge_vport *vport = hclge_get_vport(handle); @@ -6486,8 +6575,6 @@ static int hclge_set_app_loopback(struct hclge_dev *hdev, bool en) /* 2 Then setup the loopback flag */ loop_en = le32_to_cpu(req->txrx_pad_fcs_loop_en); hnae3_set_bit(loop_en, HCLGE_MAC_APP_LP_B, en ? 1 : 0); - hnae3_set_bit(loop_en, HCLGE_MAC_TX_EN_B, en ? 1 : 0); - hnae3_set_bit(loop_en, HCLGE_MAC_RX_EN_B, en ? 1 : 0); req->txrx_pad_fcs_loop_en = cpu_to_le32(loop_en); @@ -6834,8 +6921,22 @@ static void hclge_ae_stop(struct hnae3_handle *handle) int hclge_vport_start(struct hclge_vport *vport) { + struct hclge_dev *hdev = vport->back; + set_bit(HCLGE_VPORT_STATE_ALIVE, &vport->state); vport->last_active_jiffies = jiffies; + + if (test_bit(vport->vport_id, hdev->vport_config_block)) { + if (vport->vport_id) { + hclge_restore_mac_table_common(vport); + hclge_restore_vport_vlan_table(vport); + } else { + hclge_restore_hw_table(hdev); + } + } + + clear_bit(vport->vport_id, hdev->vport_config_block); + return 0; } @@ -6872,17 +6973,11 @@ static int hclge_get_mac_vlan_cmd_status(struct hclge_vport *vport, } if (op == HCLGE_MAC_VLAN_ADD) { - if ((!resp_code) || (resp_code == 1)) { + if (!resp_code || resp_code == 1) return 0; - } else if (resp_code == HCLGE_ADD_UC_OVERFLOW) { - dev_err(&hdev->pdev->dev, - "add mac addr failed for uc_overflow.\n"); - return -ENOSPC; - } else if (resp_code == HCLGE_ADD_MC_OVERFLOW) { - dev_err(&hdev->pdev->dev, - "add mac addr failed for mc_overflow.\n"); + else if (resp_code == HCLGE_ADD_UC_OVERFLOW || + resp_code == HCLGE_ADD_MC_OVERFLOW) return -ENOSPC; - } dev_err(&hdev->pdev->dev, "add mac addr failed for undefined, code=%u.\n", @@ -7106,52 +7201,8 @@ static int hclge_add_mac_vlan_tbl(struct hclge_vport *vport, return cfg_status; } -static int hclge_init_umv_space(struct hclge_dev *hdev) -{ - u16 allocated_size = 0; - int ret; - - ret = hclge_set_umv_space(hdev, hdev->wanted_umv_size, &allocated_size, - true); - if (ret) - return ret; - - if (allocated_size < hdev->wanted_umv_size) - dev_warn(&hdev->pdev->dev, - "Alloc umv space failed, want %u, get %u\n", - hdev->wanted_umv_size, allocated_size); - - mutex_init(&hdev->umv_mutex); - hdev->max_umv_size = allocated_size; - /* divide max_umv_size by (hdev->num_req_vfs + 2), in order to - * preserve some unicast mac vlan table entries shared by pf - * and its vfs. - */ - hdev->priv_umv_size = hdev->max_umv_size / (hdev->num_req_vfs + 2); - hdev->share_umv_size = hdev->priv_umv_size + - hdev->max_umv_size % (hdev->num_req_vfs + 2); - - return 0; -} - -static int hclge_uninit_umv_space(struct hclge_dev *hdev) -{ - int ret; - - if (hdev->max_umv_size > 0) { - ret = hclge_set_umv_space(hdev, hdev->max_umv_size, NULL, - false); - if (ret) - return ret; - hdev->max_umv_size = 0; - } - mutex_destroy(&hdev->umv_mutex); - - return 0; -} - static int hclge_set_umv_space(struct hclge_dev *hdev, u16 space_size, - u16 *allocated_size, bool is_alloc) + u16 *allocated_size) { struct hclge_umv_spc_alc_cmd *req; struct hclge_desc desc; @@ -7159,21 +7210,39 @@ static int hclge_set_umv_space(struct hclge_dev *hdev, u16 space_size, req = (struct hclge_umv_spc_alc_cmd *)desc.data; hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_MAC_VLAN_ALLOCATE, false); - if (!is_alloc) - hnae3_set_bit(req->allocate, HCLGE_UMV_SPC_ALC_B, 1); req->space_size = cpu_to_le32(space_size); ret = hclge_cmd_send(&hdev->hw, &desc, 1); if (ret) { - dev_err(&hdev->pdev->dev, - "%s umv space failed for cmd_send, ret =%d\n", - is_alloc ? "allocate" : "free", ret); + dev_err(&hdev->pdev->dev, "failed to set umv space, ret = %d\n", + ret); return ret; } - if (is_alloc && allocated_size) - *allocated_size = le32_to_cpu(desc.data[1]); + *allocated_size = le32_to_cpu(desc.data[1]); + + return 0; +} + +static int hclge_init_umv_space(struct hclge_dev *hdev) +{ + u16 allocated_size = 0; + int ret; + + ret = hclge_set_umv_space(hdev, hdev->wanted_umv_size, &allocated_size); + if (ret) + return ret; + + if (allocated_size < hdev->wanted_umv_size) + dev_warn(&hdev->pdev->dev, + "failed to alloc umv space, want %u, get %u\n", + hdev->wanted_umv_size, allocated_size); + + hdev->max_umv_size = allocated_size; + hdev->priv_umv_size = hdev->max_umv_size / (hdev->num_alloc_vport + 1); + hdev->share_umv_size = hdev->priv_umv_size + + hdev->max_umv_size % (hdev->num_alloc_vport + 1); return 0; } @@ -7188,21 +7257,25 @@ static void hclge_reset_umv_space(struct hclge_dev *hdev) vport->used_umv_num = 0; } - mutex_lock(&hdev->umv_mutex); + mutex_lock(&hdev->vport_lock); hdev->share_umv_size = hdev->priv_umv_size + - hdev->max_umv_size % (hdev->num_req_vfs + 2); - mutex_unlock(&hdev->umv_mutex); + hdev->max_umv_size % (hdev->num_alloc_vport + 1); + mutex_unlock(&hdev->vport_lock); } -static bool hclge_is_umv_space_full(struct hclge_vport *vport) +static bool hclge_is_umv_space_full(struct hclge_vport *vport, bool need_lock) { struct hclge_dev *hdev = vport->back; bool is_full; - mutex_lock(&hdev->umv_mutex); + if (need_lock) + mutex_lock(&hdev->vport_lock); + is_full = (vport->used_umv_num >= hdev->priv_umv_size && hdev->share_umv_size == 0); - mutex_unlock(&hdev->umv_mutex); + + if (need_lock) + mutex_unlock(&hdev->vport_lock); return is_full; } @@ -7211,7 +7284,6 @@ static void hclge_update_umv_space(struct hclge_vport *vport, bool is_free) { struct hclge_dev *hdev = vport->back; - mutex_lock(&hdev->umv_mutex); if (is_free) { if (vport->used_umv_num > hdev->priv_umv_size) hdev->share_umv_size++; @@ -7224,7 +7296,99 @@ static void hclge_update_umv_space(struct hclge_vport *vport, bool is_free) hdev->share_umv_size--; vport->used_umv_num++; } - mutex_unlock(&hdev->umv_mutex); +} + +static struct hclge_mac_node *hclge_find_mac_node(struct list_head *list, + const u8 *mac_addr) +{ + struct hclge_mac_node *mac_node, *tmp; + + list_for_each_entry_safe(mac_node, tmp, list, node) + if (ether_addr_equal(mac_addr, mac_node->mac_addr)) + return mac_node; + + return NULL; +} + +static void hclge_update_mac_node(struct hclge_mac_node *mac_node, + enum HCLGE_MAC_NODE_STATE state) +{ + switch (state) { + /* from set_rx_mode or tmp_add_list */ + case HCLGE_MAC_TO_ADD: + if (mac_node->state == HCLGE_MAC_TO_DEL) + mac_node->state = HCLGE_MAC_ACTIVE; + break; + /* only from set_rx_mode */ + case HCLGE_MAC_TO_DEL: + if (mac_node->state == HCLGE_MAC_TO_ADD) { + list_del(&mac_node->node); + kfree(mac_node); + } else { + mac_node->state = HCLGE_MAC_TO_DEL; + } + break; + /* only from tmp_add_list, the mac_node->state won't be + * ACTIVE. + */ + case HCLGE_MAC_ACTIVE: + if (mac_node->state == HCLGE_MAC_TO_ADD) + mac_node->state = HCLGE_MAC_ACTIVE; + + break; + } +} + +int hclge_update_mac_list(struct hclge_vport *vport, + enum HCLGE_MAC_NODE_STATE state, + enum HCLGE_MAC_ADDR_TYPE mac_type, + const unsigned char *addr) +{ + struct hclge_dev *hdev = vport->back; + struct hclge_mac_node *mac_node; + struct list_head *list; + + list = (mac_type == HCLGE_MAC_ADDR_UC) ? + &vport->uc_mac_list : &vport->mc_mac_list; + + spin_lock_bh(&vport->mac_list_lock); + + /* if the mac addr is already in the mac list, no need to add a new + * one into it, just check the mac addr state, convert it to a new + * new state, or just remove it, or do nothing. + */ + mac_node = hclge_find_mac_node(list, addr); + if (mac_node) { + hclge_update_mac_node(mac_node, state); + spin_unlock_bh(&vport->mac_list_lock); + set_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE, &vport->state); + return 0; + } + + /* if this address is never added, unnecessary to delete */ + if (state == HCLGE_MAC_TO_DEL) { + spin_unlock_bh(&vport->mac_list_lock); + dev_err(&hdev->pdev->dev, + "failed to delete address %pM from mac list\n", + addr); + return -ENOENT; + } + + mac_node = kzalloc(sizeof(*mac_node), GFP_ATOMIC); + if (!mac_node) { + spin_unlock_bh(&vport->mac_list_lock); + return -ENOMEM; + } + + set_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE, &vport->state); + + mac_node->state = state; + ether_addr_copy(mac_node->mac_addr, addr); + list_add_tail(&mac_node->node, list); + + spin_unlock_bh(&vport->mac_list_lock); + + return 0; } static int hclge_add_uc_addr(struct hnae3_handle *handle, @@ -7232,7 +7396,8 @@ static int hclge_add_uc_addr(struct hnae3_handle *handle, { struct hclge_vport *vport = hclge_get_vport(handle); - return hclge_add_uc_addr_common(vport, addr); + return hclge_update_mac_list(vport, HCLGE_MAC_TO_ADD, HCLGE_MAC_ADDR_UC, + addr); } int hclge_add_uc_addr_common(struct hclge_vport *vport, @@ -7271,15 +7436,19 @@ int hclge_add_uc_addr_common(struct hclge_vport *vport, */ ret = hclge_lookup_mac_vlan_tbl(vport, &req, &desc, false); if (ret == -ENOENT) { - if (!hclge_is_umv_space_full(vport)) { + mutex_lock(&hdev->vport_lock); + if (!hclge_is_umv_space_full(vport, false)) { ret = hclge_add_mac_vlan_tbl(vport, &req, NULL); if (!ret) hclge_update_umv_space(vport, false); + mutex_unlock(&hdev->vport_lock); return ret; } + mutex_unlock(&hdev->vport_lock); - dev_err(&hdev->pdev->dev, "UC MAC table full(%u)\n", - hdev->priv_umv_size); + if (!(vport->overflow_promisc_flags & HNAE3_OVERFLOW_UPE)) + dev_err(&hdev->pdev->dev, "UC MAC table full(%u)\n", + hdev->priv_umv_size); return -ENOSPC; } @@ -7303,7 +7472,8 @@ static int hclge_rm_uc_addr(struct hnae3_handle *handle, { struct hclge_vport *vport = hclge_get_vport(handle); - return hclge_rm_uc_addr_common(vport, addr); + return hclge_update_mac_list(vport, HCLGE_MAC_TO_DEL, HCLGE_MAC_ADDR_UC, + addr); } int hclge_rm_uc_addr_common(struct hclge_vport *vport, @@ -7326,8 +7496,13 @@ int hclge_rm_uc_addr_common(struct hclge_vport *vport, hnae3_set_bit(req.entry_type, HCLGE_MAC_VLAN_BIT0_EN_B, 0); hclge_prepare_mac_addr(&req, addr, false); ret = hclge_remove_mac_vlan_tbl(vport, &req); - if (!ret) + if (!ret) { + mutex_lock(&hdev->vport_lock); hclge_update_umv_space(vport, true); + mutex_unlock(&hdev->vport_lock); + } else if (ret == -ENOENT) { + ret = 0; + } return ret; } @@ -7337,7 +7512,8 @@ static int hclge_add_mc_addr(struct hnae3_handle *handle, { struct hclge_vport *vport = hclge_get_vport(handle); - return hclge_add_mc_addr_common(vport, addr); + return hclge_update_mac_list(vport, HCLGE_MAC_TO_ADD, HCLGE_MAC_ADDR_MC, + addr); } int hclge_add_mc_addr_common(struct hclge_vport *vport, @@ -7369,7 +7545,9 @@ int hclge_add_mc_addr_common(struct hclge_vport *vport, return status; status = hclge_add_mac_vlan_tbl(vport, &req, desc); - if (status == -ENOSPC) + /* if already overflow, not to print each time */ + if (status == -ENOSPC && + !(vport->overflow_promisc_flags & HNAE3_OVERFLOW_MPE)) dev_err(&hdev->pdev->dev, "mc mac vlan table is full\n"); return status; @@ -7380,7 +7558,8 @@ static int hclge_rm_mc_addr(struct hnae3_handle *handle, { struct hclge_vport *vport = hclge_get_vport(handle); - return hclge_rm_mc_addr_common(vport, addr); + return hclge_update_mac_list(vport, HCLGE_MAC_TO_DEL, HCLGE_MAC_ADDR_MC, + addr); } int hclge_rm_mc_addr_common(struct hclge_vport *vport, @@ -7415,111 +7594,354 @@ int hclge_rm_mc_addr_common(struct hclge_vport *vport, /* Not all the vfid is zero, update the vfid */ status = hclge_add_mac_vlan_tbl(vport, &req, desc); - } else { - /* Maybe this mac address is in mta table, but it cannot be - * deleted here because an entry of mta represents an address - * range rather than a specific address. the delete action to - * all entries will take effect in update_mta_status called by - * hns3_nic_set_rx_mode. - */ + } else if (status == -ENOENT) { status = 0; } return status; } -void hclge_add_vport_mac_table(struct hclge_vport *vport, const u8 *mac_addr, - enum HCLGE_MAC_ADDR_TYPE mac_type) +static void hclge_sync_vport_mac_list(struct hclge_vport *vport, + struct list_head *list, + int (*sync)(struct hclge_vport *, + const unsigned char *)) { - struct hclge_vport_mac_addr_cfg *mac_cfg; - struct list_head *list; + struct hclge_mac_node *mac_node, *tmp; + int ret; - if (!vport->vport_id) - return; + list_for_each_entry_safe(mac_node, tmp, list, node) { + ret = sync(vport, mac_node->mac_addr); + if (!ret) { + mac_node->state = HCLGE_MAC_ACTIVE; + } else { + set_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE, + &vport->state); + break; + } + } +} - mac_cfg = kzalloc(sizeof(*mac_cfg), GFP_KERNEL); - if (!mac_cfg) - return; +static void hclge_unsync_vport_mac_list(struct hclge_vport *vport, + struct list_head *list, + int (*unsync)(struct hclge_vport *, + const unsigned char *)) +{ + struct hclge_mac_node *mac_node, *tmp; + int ret; + + list_for_each_entry_safe(mac_node, tmp, list, node) { + ret = unsync(vport, mac_node->mac_addr); + if (!ret || ret == -ENOENT) { + list_del(&mac_node->node); + kfree(mac_node); + } else { + set_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE, + &vport->state); + break; + } + } +} - mac_cfg->hd_tbl_status = true; - memcpy(mac_cfg->mac_addr, mac_addr, ETH_ALEN); +static bool hclge_sync_from_add_list(struct list_head *add_list, + struct list_head *mac_list) +{ + struct hclge_mac_node *mac_node, *tmp, *new_node; + bool all_added = true; - list = (mac_type == HCLGE_MAC_ADDR_UC) ? - &vport->uc_mac_list : &vport->mc_mac_list; + list_for_each_entry_safe(mac_node, tmp, add_list, node) { + if (mac_node->state == HCLGE_MAC_TO_ADD) + all_added = false; - list_add_tail(&mac_cfg->node, list); + /* if the mac address from tmp_add_list is not in the + * uc/mc_mac_list, it means have received a TO_DEL request + * during the time window of adding the mac address into mac + * table. if mac_node state is ACTIVE, then change it to TO_DEL, + * then it will be removed at next time. else it must be TO_ADD, + * this address hasn't been added into mac table, + * so just remove the mac node. + */ + new_node = hclge_find_mac_node(mac_list, mac_node->mac_addr); + if (new_node) { + hclge_update_mac_node(new_node, mac_node->state); + list_del(&mac_node->node); + kfree(mac_node); + } else if (mac_node->state == HCLGE_MAC_ACTIVE) { + mac_node->state = HCLGE_MAC_TO_DEL; + list_del(&mac_node->node); + list_add_tail(&mac_node->node, mac_list); + } else { + list_del(&mac_node->node); + kfree(mac_node); + } + } + + return all_added; } -void hclge_rm_vport_mac_table(struct hclge_vport *vport, const u8 *mac_addr, - bool is_write_tbl, - enum HCLGE_MAC_ADDR_TYPE mac_type) +static void hclge_sync_from_del_list(struct list_head *del_list, + struct list_head *mac_list) { - struct hclge_vport_mac_addr_cfg *mac_cfg, *tmp; - struct list_head *list; - bool uc_flag, mc_flag; + struct hclge_mac_node *mac_node, *tmp, *new_node; - list = (mac_type == HCLGE_MAC_ADDR_UC) ? - &vport->uc_mac_list : &vport->mc_mac_list; + list_for_each_entry_safe(mac_node, tmp, del_list, node) { + new_node = hclge_find_mac_node(mac_list, mac_node->mac_addr); + if (new_node) { + /* If the mac addr exists in the mac list, it means + * received a new TO_ADD request during the time window + * of configuring the mac address. For the mac node + * state is TO_ADD, and the address is already in the + * in the hardware(due to delete fail), so we just need + * to change the mac node state to ACTIVE. + */ + new_node->state = HCLGE_MAC_ACTIVE; + list_del(&mac_node->node); + kfree(mac_node); + } else { + list_del(&mac_node->node); + list_add_tail(&mac_node->node, mac_list); + } + } +} - uc_flag = is_write_tbl && mac_type == HCLGE_MAC_ADDR_UC; - mc_flag = is_write_tbl && mac_type == HCLGE_MAC_ADDR_MC; +static void hclge_update_overflow_flags(struct hclge_vport *vport, + enum HCLGE_MAC_ADDR_TYPE mac_type, + bool is_all_added) +{ + if (mac_type == HCLGE_MAC_ADDR_UC) { + if (is_all_added) + vport->overflow_promisc_flags &= ~HNAE3_OVERFLOW_UPE; + else + vport->overflow_promisc_flags |= HNAE3_OVERFLOW_UPE; + } else { + if (is_all_added) + vport->overflow_promisc_flags &= ~HNAE3_OVERFLOW_MPE; + else + vport->overflow_promisc_flags |= HNAE3_OVERFLOW_MPE; + } +} - list_for_each_entry_safe(mac_cfg, tmp, list, node) { - if (ether_addr_equal(mac_cfg->mac_addr, mac_addr)) { - if (uc_flag && mac_cfg->hd_tbl_status) - hclge_rm_uc_addr_common(vport, mac_addr); +static void hclge_sync_vport_mac_table(struct hclge_vport *vport, + enum HCLGE_MAC_ADDR_TYPE mac_type) +{ + struct hclge_mac_node *mac_node, *tmp, *new_node; + struct list_head tmp_add_list, tmp_del_list; + struct list_head *list; + bool all_added; - if (mc_flag && mac_cfg->hd_tbl_status) - hclge_rm_mc_addr_common(vport, mac_addr); + INIT_LIST_HEAD(&tmp_add_list); + INIT_LIST_HEAD(&tmp_del_list); - list_del(&mac_cfg->node); - kfree(mac_cfg); + /* move the mac addr to the tmp_add_list and tmp_del_list, then + * we can add/delete these mac addr outside the spin lock + */ + list = (mac_type == HCLGE_MAC_ADDR_UC) ? + &vport->uc_mac_list : &vport->mc_mac_list; + + spin_lock_bh(&vport->mac_list_lock); + + list_for_each_entry_safe(mac_node, tmp, list, node) { + switch (mac_node->state) { + case HCLGE_MAC_TO_DEL: + list_del(&mac_node->node); + list_add_tail(&mac_node->node, &tmp_del_list); + break; + case HCLGE_MAC_TO_ADD: + new_node = kzalloc(sizeof(*new_node), GFP_ATOMIC); + if (!new_node) + goto stop_traverse; + ether_addr_copy(new_node->mac_addr, mac_node->mac_addr); + new_node->state = mac_node->state; + list_add_tail(&new_node->node, &tmp_add_list); + break; + default: break; } } + +stop_traverse: + spin_unlock_bh(&vport->mac_list_lock); + + /* delete first, in order to get max mac table space for adding */ + if (mac_type == HCLGE_MAC_ADDR_UC) { + hclge_unsync_vport_mac_list(vport, &tmp_del_list, + hclge_rm_uc_addr_common); + hclge_sync_vport_mac_list(vport, &tmp_add_list, + hclge_add_uc_addr_common); + } else { + hclge_unsync_vport_mac_list(vport, &tmp_del_list, + hclge_rm_mc_addr_common); + hclge_sync_vport_mac_list(vport, &tmp_add_list, + hclge_add_mc_addr_common); + } + + /* if some mac addresses were added/deleted fail, move back to the + * mac_list, and retry at next time. + */ + spin_lock_bh(&vport->mac_list_lock); + + hclge_sync_from_del_list(&tmp_del_list, list); + all_added = hclge_sync_from_add_list(&tmp_add_list, list); + + spin_unlock_bh(&vport->mac_list_lock); + + hclge_update_overflow_flags(vport, mac_type, all_added); +} + +static bool hclge_need_sync_mac_table(struct hclge_vport *vport) +{ + struct hclge_dev *hdev = vport->back; + + if (test_bit(vport->vport_id, hdev->vport_config_block)) + return false; + + if (test_and_clear_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE, &vport->state)) + return true; + + return false; +} + +static void hclge_sync_mac_table(struct hclge_dev *hdev) +{ + int i; + + for (i = 0; i < hdev->num_alloc_vport; i++) { + struct hclge_vport *vport = &hdev->vport[i]; + + if (!hclge_need_sync_mac_table(vport)) + continue; + + hclge_sync_vport_mac_table(vport, HCLGE_MAC_ADDR_UC); + hclge_sync_vport_mac_table(vport, HCLGE_MAC_ADDR_MC); + } } void hclge_rm_vport_all_mac_table(struct hclge_vport *vport, bool is_del_list, enum HCLGE_MAC_ADDR_TYPE mac_type) { - struct hclge_vport_mac_addr_cfg *mac_cfg, *tmp; - struct list_head *list; + int (*unsync)(struct hclge_vport *vport, const unsigned char *addr); + struct hclge_mac_node *mac_cfg, *tmp; + struct hclge_dev *hdev = vport->back; + struct list_head tmp_del_list, *list; + int ret; - list = (mac_type == HCLGE_MAC_ADDR_UC) ? - &vport->uc_mac_list : &vport->mc_mac_list; + if (mac_type == HCLGE_MAC_ADDR_UC) { + list = &vport->uc_mac_list; + unsync = hclge_rm_uc_addr_common; + } else { + list = &vport->mc_mac_list; + unsync = hclge_rm_mc_addr_common; + } - list_for_each_entry_safe(mac_cfg, tmp, list, node) { - if (mac_type == HCLGE_MAC_ADDR_UC && mac_cfg->hd_tbl_status) - hclge_rm_uc_addr_common(vport, mac_cfg->mac_addr); + INIT_LIST_HEAD(&tmp_del_list); - if (mac_type == HCLGE_MAC_ADDR_MC && mac_cfg->hd_tbl_status) - hclge_rm_mc_addr_common(vport, mac_cfg->mac_addr); + if (!is_del_list) + set_bit(vport->vport_id, hdev->vport_config_block); - mac_cfg->hd_tbl_status = false; - if (is_del_list) { + spin_lock_bh(&vport->mac_list_lock); + + list_for_each_entry_safe(mac_cfg, tmp, list, node) { + switch (mac_cfg->state) { + case HCLGE_MAC_TO_DEL: + case HCLGE_MAC_ACTIVE: list_del(&mac_cfg->node); - kfree(mac_cfg); + list_add_tail(&mac_cfg->node, &tmp_del_list); + break; + case HCLGE_MAC_TO_ADD: + if (is_del_list) { + list_del(&mac_cfg->node); + kfree(mac_cfg); + } + break; + } + } + + spin_unlock_bh(&vport->mac_list_lock); + + list_for_each_entry_safe(mac_cfg, tmp, &tmp_del_list, node) { + ret = unsync(vport, mac_cfg->mac_addr); + if (!ret || ret == -ENOENT) { + /* clear all mac addr from hardware, but remain these + * mac addr in the mac list, and restore them after + * vf reset finished. + */ + if (!is_del_list && + mac_cfg->state == HCLGE_MAC_ACTIVE) { + mac_cfg->state = HCLGE_MAC_TO_ADD; + } else { + list_del(&mac_cfg->node); + kfree(mac_cfg); + } + } else if (is_del_list) { + mac_cfg->state = HCLGE_MAC_TO_DEL; + } + } + + spin_lock_bh(&vport->mac_list_lock); + + hclge_sync_from_del_list(&tmp_del_list, list); + + spin_unlock_bh(&vport->mac_list_lock); +} + +/* remove all mac address when uninitailize */ +static void hclge_uninit_vport_mac_list(struct hclge_vport *vport, + enum HCLGE_MAC_ADDR_TYPE mac_type) +{ + struct hclge_mac_node *mac_node, *tmp; + struct hclge_dev *hdev = vport->back; + struct list_head tmp_del_list, *list; + + INIT_LIST_HEAD(&tmp_del_list); + + list = (mac_type == HCLGE_MAC_ADDR_UC) ? + &vport->uc_mac_list : &vport->mc_mac_list; + + spin_lock_bh(&vport->mac_list_lock); + + list_for_each_entry_safe(mac_node, tmp, list, node) { + switch (mac_node->state) { + case HCLGE_MAC_TO_DEL: + case HCLGE_MAC_ACTIVE: + list_del(&mac_node->node); + list_add_tail(&mac_node->node, &tmp_del_list); + break; + case HCLGE_MAC_TO_ADD: + list_del(&mac_node->node); + kfree(mac_node); + break; } } + + spin_unlock_bh(&vport->mac_list_lock); + + if (mac_type == HCLGE_MAC_ADDR_UC) + hclge_unsync_vport_mac_list(vport, &tmp_del_list, + hclge_rm_uc_addr_common); + else + hclge_unsync_vport_mac_list(vport, &tmp_del_list, + hclge_rm_mc_addr_common); + + if (!list_empty(&tmp_del_list)) + dev_warn(&hdev->pdev->dev, + "uninit %s mac list for vport %u not completely.\n", + mac_type == HCLGE_MAC_ADDR_UC ? "uc" : "mc", + vport->vport_id); + + list_for_each_entry_safe(mac_node, tmp, &tmp_del_list, node) { + list_del(&mac_node->node); + kfree(mac_node); + } } -void hclge_uninit_vport_mac_table(struct hclge_dev *hdev) +static void hclge_uninit_mac_table(struct hclge_dev *hdev) { - struct hclge_vport_mac_addr_cfg *mac, *tmp; struct hclge_vport *vport; int i; for (i = 0; i < hdev->num_alloc_vport; i++) { vport = &hdev->vport[i]; - list_for_each_entry_safe(mac, tmp, &vport->uc_mac_list, node) { - list_del(&mac->node); - kfree(mac); - } - - list_for_each_entry_safe(mac, tmp, &vport->mc_mac_list, node) { - list_del(&mac->node); - kfree(mac); - } + hclge_uninit_vport_mac_list(vport, HCLGE_MAC_ADDR_UC); + hclge_uninit_vport_mac_list(vport, HCLGE_MAC_ADDR_MC); } } @@ -7683,12 +8105,57 @@ static void hclge_get_mac_addr(struct hnae3_handle *handle, u8 *p) ether_addr_copy(p, hdev->hw.mac.mac_addr); } +int hclge_update_mac_node_for_dev_addr(struct hclge_vport *vport, + const u8 *old_addr, const u8 *new_addr) +{ + struct list_head *list = &vport->uc_mac_list; + struct hclge_mac_node *old_node, *new_node; + + new_node = hclge_find_mac_node(list, new_addr); + if (!new_node) { + new_node = kzalloc(sizeof(*new_node), GFP_ATOMIC); + if (!new_node) + return -ENOMEM; + + new_node->state = HCLGE_MAC_TO_ADD; + ether_addr_copy(new_node->mac_addr, new_addr); + list_add(&new_node->node, list); + } else { + if (new_node->state == HCLGE_MAC_TO_DEL) + new_node->state = HCLGE_MAC_ACTIVE; + + /* make sure the new addr is in the list head, avoid dev + * addr may be not re-added into mac table for the umv space + * limitation after global/imp reset which will clear mac + * table by hardware. + */ + list_move(&new_node->node, list); + } + + if (old_addr && !ether_addr_equal(old_addr, new_addr)) { + old_node = hclge_find_mac_node(list, old_addr); + if (old_node) { + if (old_node->state == HCLGE_MAC_TO_ADD) { + list_del(&old_node->node); + kfree(old_node); + } else { + old_node->state = HCLGE_MAC_TO_DEL; + } + } + } + + set_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE, &vport->state); + + return 0; +} + static int hclge_set_mac_addr(struct hnae3_handle *handle, void *p, bool is_first) { const unsigned char *new_addr = (const unsigned char *)p; struct hclge_vport *vport = hclge_get_vport(handle); struct hclge_dev *hdev = vport->back; + unsigned char *old_addr = NULL; int ret; /* mac addr check */ @@ -7696,39 +8163,42 @@ static int hclge_set_mac_addr(struct hnae3_handle *handle, void *p, is_broadcast_ether_addr(new_addr) || is_multicast_ether_addr(new_addr)) { dev_err(&hdev->pdev->dev, - "Change uc mac err! invalid mac:%pM.\n", + "change uc mac err! invalid mac: %pM.\n", new_addr); return -EINVAL; } - if ((!is_first || is_kdump_kernel()) && - hclge_rm_uc_addr(handle, hdev->hw.mac.mac_addr)) - dev_warn(&hdev->pdev->dev, - "remove old uc mac address fail.\n"); - - ret = hclge_add_uc_addr(handle, new_addr); + ret = hclge_pause_addr_cfg(hdev, new_addr); if (ret) { dev_err(&hdev->pdev->dev, - "add uc mac address fail, ret =%d.\n", + "failed to configure mac pause address, ret = %d\n", ret); - - if (!is_first && - hclge_add_uc_addr(handle, hdev->hw.mac.mac_addr)) - dev_err(&hdev->pdev->dev, - "restore uc mac address fail.\n"); - - return -EIO; + return ret; } - ret = hclge_pause_addr_cfg(hdev, new_addr); + if (!is_first) + old_addr = hdev->hw.mac.mac_addr; + + spin_lock_bh(&vport->mac_list_lock); + ret = hclge_update_mac_node_for_dev_addr(vport, old_addr, new_addr); if (ret) { dev_err(&hdev->pdev->dev, - "configure mac pause address fail, ret =%d.\n", - ret); - return -EIO; - } + "failed to change the mac addr:%pM, ret = %d\n", + new_addr, ret); + spin_unlock_bh(&vport->mac_list_lock); + if (!is_first) + hclge_pause_addr_cfg(hdev, old_addr); + + return ret; + } + /* we must update dev addr with spin lock protect, preventing dev addr + * being removed by set_rx_mode path. + */ ether_addr_copy(hdev->hw.mac.mac_addr, new_addr); + spin_unlock_bh(&vport->mac_list_lock); + + hclge_task_schedule(hdev, 0); return 0; } @@ -8308,42 +8778,80 @@ void hclge_uninit_vport_vlan_table(struct hclge_dev *hdev) } } -static void hclge_restore_vlan_table(struct hnae3_handle *handle) +void hclge_restore_vport_vlan_table(struct hclge_vport *vport) { - struct hclge_vport *vport = hclge_get_vport(handle); struct hclge_vport_vlan_cfg *vlan, *tmp; struct hclge_dev *hdev = vport->back; u16 vlan_proto; - u16 state, vlan_id; - int i; + u16 vlan_id; + u16 state; + int ret; - for (i = 0; i < hdev->num_alloc_vport; i++) { - vport = &hdev->vport[i]; - vlan_proto = vport->port_base_vlan_cfg.vlan_info.vlan_proto; - vlan_id = vport->port_base_vlan_cfg.vlan_info.vlan_tag; - state = vport->port_base_vlan_cfg.state; + vlan_proto = vport->port_base_vlan_cfg.vlan_info.vlan_proto; + vlan_id = vport->port_base_vlan_cfg.vlan_info.vlan_tag; + state = vport->port_base_vlan_cfg.state; - if (state != HNAE3_PORT_BASE_VLAN_DISABLE) { - hclge_set_vlan_filter_hw(hdev, htons(vlan_proto), - vport->vport_id, vlan_id, - false); - continue; - } + if (state != HNAE3_PORT_BASE_VLAN_DISABLE) { + clear_bit(vport->vport_id, hdev->vlan_table[vlan_id]); + hclge_set_vlan_filter_hw(hdev, htons(vlan_proto), + vport->vport_id, vlan_id, + false); + return; + } - list_for_each_entry_safe(vlan, tmp, &vport->vlan_list, node) { - int ret; + list_for_each_entry_safe(vlan, tmp, &vport->vlan_list, node) { + ret = hclge_set_vlan_filter_hw(hdev, htons(ETH_P_8021Q), + vport->vport_id, + vlan->vlan_id, false); + if (ret) + break; + vlan->hd_tbl_status = true; + } +} - if (!vlan->hd_tbl_status) - continue; - ret = hclge_set_vlan_filter_hw(hdev, htons(ETH_P_8021Q), - vport->vport_id, - vlan->vlan_id, false); - if (ret) - break; +/* For global reset and imp reset, hardware will clear the mac table, + * so we change the mac address state from ACTIVE to TO_ADD, then they + * can be restored in the service task after reset complete. Furtherly, + * the mac addresses with state TO_DEL or DEL_FAIL are unnecessary to + * be restored after reset, so just remove these mac nodes from mac_list. + */ +static void hclge_mac_node_convert_for_reset(struct list_head *list) +{ + struct hclge_mac_node *mac_node, *tmp; + + list_for_each_entry_safe(mac_node, tmp, list, node) { + if (mac_node->state == HCLGE_MAC_ACTIVE) { + mac_node->state = HCLGE_MAC_TO_ADD; + } else if (mac_node->state == HCLGE_MAC_TO_DEL) { + list_del(&mac_node->node); + kfree(mac_node); } } } +void hclge_restore_mac_table_common(struct hclge_vport *vport) +{ + spin_lock_bh(&vport->mac_list_lock); + + hclge_mac_node_convert_for_reset(&vport->uc_mac_list); + hclge_mac_node_convert_for_reset(&vport->mc_mac_list); + set_bit(HCLGE_VPORT_STATE_MAC_TBL_CHANGE, &vport->state); + + spin_unlock_bh(&vport->mac_list_lock); +} + +static void hclge_restore_hw_table(struct hclge_dev *hdev) +{ + struct hclge_vport *vport = &hdev->vport[0]; + struct hnae3_handle *handle = &vport->nic; + + hclge_restore_mac_table_common(vport); + hclge_restore_vport_vlan_table(vport); + set_bit(HCLGE_STATE_PROMISC_CHANGED, &hdev->state); + + hclge_restore_fd_entries(handle); +} + int hclge_en_hw_strip_rxvtag(struct hnae3_handle *handle, bool enable) { struct hclge_vport *vport = hclge_get_vport(handle); @@ -9412,10 +9920,8 @@ static int hclge_init_ae_dev(struct hnae3_ae_dev *ae_dev) int ret; hdev = devm_kzalloc(&pdev->dev, sizeof(*hdev), GFP_KERNEL); - if (!hdev) { - ret = -ENOMEM; - goto out; - } + if (!hdev) + return -ENOMEM; hdev->pdev = pdev; hdev->ae_dev = ae_dev; @@ -9594,6 +10100,7 @@ err_pci_uninit: pci_release_regions(pdev); pci_disable_device(pdev); out: + mutex_destroy(&hdev->vport_lock); return ret; } @@ -9658,7 +10165,7 @@ static int hclge_set_vf_spoofchk(struct hnae3_handle *handle, int vf, dev_warn(&hdev->pdev->dev, "vf %d vlan table is full, enable spoof check may cause its packet send fail\n", vf); - else if (enable && hclge_is_umv_space_full(vport)) + else if (enable && hclge_is_umv_space_full(vport, true)) dev_warn(&hdev->pdev->dev, "vf %d mac table is full, enable spoof check may cause its packet send fail\n", vf); @@ -9835,8 +10342,16 @@ static int hclge_reset_ae_dev(struct hnae3_ae_dev *ae_dev) set_bit(HCLGE_STATE_DOWN, &hdev->state); hclge_stats_clear(hdev); - memset(hdev->vlan_table, 0, sizeof(hdev->vlan_table)); - memset(hdev->vf_vlan_full, 0, sizeof(hdev->vf_vlan_full)); + /* NOTE: pf reset needn't to clear or restore pf and vf table entry. + * so here should not clean table in memory. + */ + if (hdev->reset_type == HNAE3_IMP_RESET || + hdev->reset_type == HNAE3_GLOBAL_RESET) { + memset(hdev->vlan_table, 0, sizeof(hdev->vlan_table)); + memset(hdev->vf_vlan_full, 0, sizeof(hdev->vf_vlan_full)); + bitmap_set(hdev->vport_config_block, 0, hdev->num_alloc_vport); + hclge_reset_umv_space(hdev); + } ret = hclge_cmd_init(hdev); if (ret) { @@ -9850,8 +10365,6 @@ static int hclge_reset_ae_dev(struct hnae3_ae_dev *ae_dev) return ret; } - hclge_reset_umv_space(hdev); - ret = hclge_mac_init(hdev); if (ret) { dev_err(&pdev->dev, "Mac init error, ret = %d\n", ret); @@ -9947,12 +10460,11 @@ static void hclge_uninit_ae_dev(struct hnae3_ae_dev *ae_dev) hclge_clear_vf_vlan(hdev); hclge_misc_affinity_teardown(hdev); hclge_state_uninit(hdev); + hclge_uninit_mac_table(hdev); if (mac->phydev) mdiobus_unregister(mac->mdio_bus); - hclge_uninit_umv_space(hdev); - /* Disable MISC vector(vector0) */ hclge_enable_vector(&hdev->misc_vector, false); synchronize_irq(hdev->misc_vector.vector_irq); @@ -9966,7 +10478,6 @@ static void hclge_uninit_ae_dev(struct hnae3_ae_dev *ae_dev) hclge_misc_irq_uninit(hdev); hclge_pci_uninit(hdev); mutex_destroy(&hdev->vport_lock); - hclge_uninit_vport_mac_table(hdev); hclge_uninit_vport_vlan_table(hdev); ae_dev->priv = NULL; } @@ -10213,16 +10724,19 @@ static int hclge_get_64_bit_regs(struct hclge_dev *hdev, u32 regs_num, int hclge_query_bd_num_cmd_send(struct hclge_dev *hdev, struct hclge_desc *desc) { - /*prepare 4 commands to query DFX BD number*/ - hclge_cmd_setup_basic_desc(&desc[0], HCLGE_OPC_DFX_BD_NUM, true); - desc[0].flag |= cpu_to_le16(HCLGE_CMD_FLAG_NEXT); - hclge_cmd_setup_basic_desc(&desc[1], HCLGE_OPC_DFX_BD_NUM, true); - desc[1].flag |= cpu_to_le16(HCLGE_CMD_FLAG_NEXT); - hclge_cmd_setup_basic_desc(&desc[2], HCLGE_OPC_DFX_BD_NUM, true); - desc[2].flag |= cpu_to_le16(HCLGE_CMD_FLAG_NEXT); - hclge_cmd_setup_basic_desc(&desc[3], HCLGE_OPC_DFX_BD_NUM, true); + int i; - return hclge_cmd_send(&hdev->hw, desc, 4); + /* initialize command BD except the last one */ + for (i = 0; i < HCLGE_GET_DFX_REG_TYPE_CNT - 1; i++) { + hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_DFX_BD_NUM, + true); + desc[i].flag |= cpu_to_le16(HCLGE_CMD_FLAG_NEXT); + } + + /* initialize the last command BD */ + hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_DFX_BD_NUM, true); + + return hclge_cmd_send(&hdev->hw, desc, HCLGE_GET_DFX_REG_TYPE_CNT); } static int hclge_get_dfx_reg_bd_num(struct hclge_dev *hdev, @@ -10576,6 +11090,131 @@ static int hclge_gro_en(struct hnae3_handle *handle, bool enable) return hclge_config_gro(hdev, enable); } +static void hclge_sync_promisc_mode(struct hclge_dev *hdev) +{ + struct hclge_vport *vport = &hdev->vport[0]; + struct hnae3_handle *handle = &vport->nic; + u8 tmp_flags = 0; + int ret; + + if (vport->last_promisc_flags != vport->overflow_promisc_flags) { + set_bit(HCLGE_STATE_PROMISC_CHANGED, &hdev->state); + vport->last_promisc_flags = vport->overflow_promisc_flags; + } + + if (test_bit(HCLGE_STATE_PROMISC_CHANGED, &hdev->state)) { + tmp_flags = handle->netdev_flags | vport->last_promisc_flags; + ret = hclge_set_promisc_mode(handle, tmp_flags & HNAE3_UPE, + tmp_flags & HNAE3_MPE); + if (!ret) { + clear_bit(HCLGE_STATE_PROMISC_CHANGED, &hdev->state); + hclge_enable_vlan_filter(handle, + tmp_flags & HNAE3_VLAN_FLTR); + } + } +} + +static bool hclge_module_existed(struct hclge_dev *hdev) +{ + struct hclge_desc desc; + u32 existed; + int ret; + + hclge_cmd_setup_basic_desc(&desc, HCLGE_OPC_GET_SFP_EXIST, true); + ret = hclge_cmd_send(&hdev->hw, &desc, 1); + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to get SFP exist state, ret = %d\n", ret); + return false; + } + + existed = le32_to_cpu(desc.data[0]); + + return existed != 0; +} + +/* need 6 bds(total 140 bytes) in one reading + * return the number of bytes actually read, 0 means read failed. + */ +static u16 hclge_get_sfp_eeprom_info(struct hclge_dev *hdev, u32 offset, + u32 len, u8 *data) +{ + struct hclge_desc desc[HCLGE_SFP_INFO_CMD_NUM]; + struct hclge_sfp_info_bd0_cmd *sfp_info_bd0; + u16 read_len; + u16 copy_len; + int ret; + int i; + + /* setup all 6 bds to read module eeprom info. */ + for (i = 0; i < HCLGE_SFP_INFO_CMD_NUM; i++) { + hclge_cmd_setup_basic_desc(&desc[i], HCLGE_OPC_GET_SFP_EEPROM, + true); + + /* bd0~bd4 need next flag */ + if (i < HCLGE_SFP_INFO_CMD_NUM - 1) + desc[i].flag |= cpu_to_le16(HCLGE_CMD_FLAG_NEXT); + } + + /* setup bd0, this bd contains offset and read length. */ + sfp_info_bd0 = (struct hclge_sfp_info_bd0_cmd *)desc[0].data; + sfp_info_bd0->offset = cpu_to_le16((u16)offset); + read_len = min_t(u16, len, HCLGE_SFP_INFO_MAX_LEN); + sfp_info_bd0->read_len = cpu_to_le16(read_len); + + ret = hclge_cmd_send(&hdev->hw, desc, i); + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to get SFP eeprom info, ret = %d\n", ret); + return 0; + } + + /* copy sfp info from bd0 to out buffer. */ + copy_len = min_t(u16, len, HCLGE_SFP_INFO_BD0_LEN); + memcpy(data, sfp_info_bd0->data, copy_len); + read_len = copy_len; + + /* copy sfp info from bd1~bd5 to out buffer if needed. */ + for (i = 1; i < HCLGE_SFP_INFO_CMD_NUM; i++) { + if (read_len >= len) + return read_len; + + copy_len = min_t(u16, len - read_len, HCLGE_SFP_INFO_BDX_LEN); + memcpy(data + read_len, desc[i].data, copy_len); + read_len += copy_len; + } + + return read_len; +} + +static int hclge_get_module_eeprom(struct hnae3_handle *handle, u32 offset, + u32 len, u8 *data) +{ + struct hclge_vport *vport = hclge_get_vport(handle); + struct hclge_dev *hdev = vport->back; + u32 read_len = 0; + u16 data_len; + + if (hdev->hw.mac.media_type != HNAE3_MEDIA_TYPE_FIBER) + return -EOPNOTSUPP; + + if (!hclge_module_existed(hdev)) + return -ENXIO; + + while (read_len < len) { + data_len = hclge_get_sfp_eeprom_info(hdev, + offset + read_len, + len - read_len, + data + read_len); + if (!data_len) + return -EIO; + + read_len += data_len; + } + + return 0; +} + static const struct hnae3_ae_ops hclge_ops = { .init_ae_dev = hclge_init_ae_dev, .uninit_ae_dev = hclge_uninit_ae_dev, @@ -10588,6 +11227,7 @@ static const struct hnae3_ae_ops hclge_ops = { .get_vector = hclge_get_vector, .put_vector = hclge_put_vector, .set_promisc_mode = hclge_set_promisc_mode, + .request_update_promisc_mode = hclge_request_update_promisc_mode, .set_loopback = hclge_set_loopback, .start = hclge_ae_start, .stop = hclge_ae_stop, @@ -10649,7 +11289,6 @@ static const struct hnae3_ae_ops hclge_ops = { .get_fd_rule_cnt = hclge_get_fd_rule_cnt, .get_fd_rule_info = hclge_get_fd_rule_info, .get_fd_all_rules = hclge_get_all_rules, - .restore_fd_rules = hclge_restore_fd_entries, .enable_fd = hclge_enable_fd, .add_arfs_entry = hclge_add_fd_entry_by_arfs, .dbg_run_cmd = hclge_dbg_run_cmd, @@ -10662,13 +11301,14 @@ static const struct hnae3_ae_ops hclge_ops = { .set_timer_task = hclge_set_timer_task, .mac_connect_phy = hclge_mac_connect_phy, .mac_disconnect_phy = hclge_mac_disconnect_phy, - .restore_vlan_table = hclge_restore_vlan_table, .get_vf_config = hclge_get_vf_config, .set_vf_link_state = hclge_set_vf_link_state, .set_vf_spoofchk = hclge_set_vf_spoofchk, .set_vf_trust = hclge_set_vf_trust, .set_vf_rate = hclge_set_vf_rate, .set_vf_mac = hclge_set_vf_mac, + .get_module_eeprom = hclge_get_module_eeprom, + .get_cmdq_stat = hclge_get_cmdq_stat, }; static struct hnae3_ae_algo ae_algo = { diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h index 71df23d5f1b4..46e6e0fef3ba 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.h @@ -217,6 +217,7 @@ enum HCLGE_DEV_STATE { HCLGE_STATE_STATISTICS_UPDATING, HCLGE_STATE_CMD_DISABLE, HCLGE_STATE_LINK_UPDATING, + HCLGE_STATE_PROMISC_CHANGED, HCLGE_STATE_RST_FAIL, HCLGE_STATE_MAX }; @@ -580,7 +581,6 @@ struct hclge_fd_key_cfg { struct hclge_fd_cfg { u8 fd_mode; u16 max_key_length; /* use bit as unit */ - u32 proto_support; u32 rule_num[MAX_STAGE_NUM]; /* rule entry number */ u16 cnt_num[MAX_STAGE_NUM]; /* rule hit counter number */ struct hclge_fd_key_cfg key_cfg[MAX_STAGE_NUM]; @@ -631,9 +631,15 @@ struct hclge_fd_ad_data { u16 rule_id; }; -struct hclge_vport_mac_addr_cfg { +enum HCLGE_MAC_NODE_STATE { + HCLGE_MAC_TO_ADD, + HCLGE_MAC_TO_DEL, + HCLGE_MAC_ACTIVE +}; + +struct hclge_mac_node { struct list_head node; - int hd_tbl_status; + enum HCLGE_MAC_NODE_STATE state; u8 mac_addr[ETH_ALEN]; }; @@ -765,12 +771,6 @@ struct hclge_dev { u16 num_roce_msi; /* Num of roce vectors for this PF */ int roce_base_vector; - u16 pending_udp_bitmap; - - u16 rx_itr_default; - u16 tx_itr_default; - - u16 adminq_work_limit; /* Num of admin receive queue desc to process */ unsigned long service_timer_period; unsigned long service_timer_previous; struct timer_list reset_timer; @@ -806,6 +806,8 @@ struct hclge_dev { unsigned long vlan_table[VLAN_N_VID][BITS_TO_LONGS(HCLGE_VPORT_NUM)]; unsigned long vf_vlan_full[BITS_TO_LONGS(HCLGE_VPORT_NUM)]; + unsigned long vport_config_block[BITS_TO_LONGS(HCLGE_VPORT_NUM)]; + struct hclge_fd_cfg fd_cfg; struct hlist_head fd_rule_list; spinlock_t fd_rule_lock; /* protect fd_rule_list and fd_bmap */ @@ -823,7 +825,6 @@ struct hclge_dev { u16 priv_umv_size; /* unicast mac vlan space shared by PF and its VFs */ u16 share_umv_size; - struct mutex umv_mutex; /* protect share_umv_size */ DECLARE_KFIFO(mac_tnl_log, struct hclge_mac_tnl_stats, HCLGE_MAC_TNL_LOG_SIZE); @@ -867,6 +868,7 @@ struct hclge_rss_tuple_cfg { enum HCLGE_VPORT_STATE { HCLGE_VPORT_STATE_ALIVE, + HCLGE_VPORT_STATE_MAC_TBL_CHANGE, HCLGE_VPORT_STATE_MAX }; @@ -923,6 +925,10 @@ struct hclge_vport { u32 mps; /* Max packet size */ struct hclge_vf_info vf_info; + u8 overflow_promisc_flags; + u8 last_promisc_flags; + + spinlock_t mac_list_lock; /* protect mac address need to add/detele */ struct list_head uc_mac_list; /* Store VF unicast table */ struct list_head mc_mac_list; /* Store VF multicast table */ struct list_head vlan_list; /* Store VF vlan table */ @@ -978,16 +984,18 @@ int hclge_dbg_run_cmd(struct hnae3_handle *handle, const char *cmd_buf); u16 hclge_covert_handle_qid_global(struct hnae3_handle *handle, u16 queue_id); int hclge_notify_client(struct hclge_dev *hdev, enum hnae3_reset_notify_type type); -void hclge_add_vport_mac_table(struct hclge_vport *vport, const u8 *mac_addr, - enum HCLGE_MAC_ADDR_TYPE mac_type); -void hclge_rm_vport_mac_table(struct hclge_vport *vport, const u8 *mac_addr, - bool is_write_tbl, - enum HCLGE_MAC_ADDR_TYPE mac_type); +int hclge_update_mac_list(struct hclge_vport *vport, + enum HCLGE_MAC_NODE_STATE state, + enum HCLGE_MAC_ADDR_TYPE mac_type, + const unsigned char *addr); +int hclge_update_mac_node_for_dev_addr(struct hclge_vport *vport, + const u8 *old_addr, const u8 *new_addr); void hclge_rm_vport_all_mac_table(struct hclge_vport *vport, bool is_del_list, enum HCLGE_MAC_ADDR_TYPE mac_type); -void hclge_uninit_vport_mac_table(struct hclge_dev *hdev); void hclge_rm_vport_all_vlan_table(struct hclge_vport *vport, bool is_del_list); void hclge_uninit_vport_vlan_table(struct hclge_dev *hdev); +void hclge_restore_mac_table_common(struct hclge_vport *vport); +void hclge_restore_vport_vlan_table(struct hclge_vport *vport); int hclge_update_port_base_vlan_cfg(struct hclge_vport *vport, u16 state, struct hclge_vlan_info *vlan_info); int hclge_push_vf_port_base_vlan_info(struct hclge_vport *vport, u8 vfid, diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c index 7f24fcb4f96a..0874ae47cb03 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mbx.c @@ -5,6 +5,9 @@ #include "hclge_mbx.h" #include "hnae3.h" +#define CREATE_TRACE_POINTS +#include "hclge_trace.h" + static u16 hclge_errno_to_resp(int errno) { return abs(errno); @@ -90,6 +93,8 @@ static int hclge_send_mbx_msg(struct hclge_vport *vport, u8 *msg, u16 msg_len, memcpy(&resp_pf_to_vf->msg.vf_mbx_msg_code, msg, msg_len); + trace_hclge_pf_mbx_send(hdev, resp_pf_to_vf); + status = hclge_cmd_send(&hdev->hw, &desc, 1); if (status) dev_err(&hdev->pdev->dev, @@ -270,26 +275,17 @@ static int hclge_set_vf_uc_mac_addr(struct hclge_vport *vport, if (!is_valid_ether_addr(mac_addr)) return -EINVAL; - hclge_rm_uc_addr_common(vport, old_addr); - status = hclge_add_uc_addr_common(vport, mac_addr); - if (status) { - hclge_add_uc_addr_common(vport, old_addr); - } else { - hclge_rm_vport_mac_table(vport, mac_addr, - false, HCLGE_MAC_ADDR_UC); - hclge_add_vport_mac_table(vport, mac_addr, - HCLGE_MAC_ADDR_UC); - } + spin_lock_bh(&vport->mac_list_lock); + status = hclge_update_mac_node_for_dev_addr(vport, old_addr, + mac_addr); + spin_unlock_bh(&vport->mac_list_lock); + hclge_task_schedule(hdev, 0); } else if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_UC_ADD) { - status = hclge_add_uc_addr_common(vport, mac_addr); - if (!status) - hclge_add_vport_mac_table(vport, mac_addr, - HCLGE_MAC_ADDR_UC); + status = hclge_update_mac_list(vport, HCLGE_MAC_TO_ADD, + HCLGE_MAC_ADDR_UC, mac_addr); } else if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_UC_REMOVE) { - status = hclge_rm_uc_addr_common(vport, mac_addr); - if (!status) - hclge_rm_vport_mac_table(vport, mac_addr, - false, HCLGE_MAC_ADDR_UC); + status = hclge_update_mac_list(vport, HCLGE_MAC_TO_DEL, + HCLGE_MAC_ADDR_UC, mac_addr); } else { dev_err(&hdev->pdev->dev, "failed to set unicast mac addr, unknown subcode %u\n", @@ -305,18 +301,13 @@ static int hclge_set_vf_mc_mac_addr(struct hclge_vport *vport, { const u8 *mac_addr = (const u8 *)(mbx_req->msg.data); struct hclge_dev *hdev = vport->back; - int status; if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_MC_ADD) { - status = hclge_add_mc_addr_common(vport, mac_addr); - if (!status) - hclge_add_vport_mac_table(vport, mac_addr, - HCLGE_MAC_ADDR_MC); + hclge_update_mac_list(vport, HCLGE_MAC_TO_ADD, + HCLGE_MAC_ADDR_MC, mac_addr); } else if (mbx_req->msg.subcode == HCLGE_MBX_MAC_VLAN_MC_REMOVE) { - status = hclge_rm_mc_addr_common(vport, mac_addr); - if (!status) - hclge_rm_vport_mac_table(vport, mac_addr, - false, HCLGE_MAC_ADDR_MC); + hclge_update_mac_list(vport, HCLGE_MAC_TO_DEL, + HCLGE_MAC_ADDR_MC, mac_addr); } else { dev_err(&hdev->pdev->dev, "failed to set mcast mac addr, unknown subcode %u\n", @@ -324,7 +315,7 @@ static int hclge_set_vf_mc_mac_addr(struct hclge_vport *vport, return -EIO; } - return status; + return 0; } int hclge_push_vf_port_base_vlan_info(struct hclge_vport *vport, u8 vfid, @@ -638,6 +629,23 @@ static void hclge_handle_ncsi_error(struct hclge_dev *hdev) ae_dev->ops->reset_event(hdev->pdev, NULL); } +static void hclge_handle_vf_tbl(struct hclge_vport *vport, + struct hclge_mbx_vf_to_pf_cmd *mbx_req) +{ + struct hclge_dev *hdev = vport->back; + struct hclge_vf_vlan_cfg *msg_cmd; + + msg_cmd = (struct hclge_vf_vlan_cfg *)&mbx_req->msg; + if (msg_cmd->subcode == HCLGE_MBX_VPORT_LIST_CLEAR) { + hclge_rm_vport_all_mac_table(vport, true, HCLGE_MAC_ADDR_UC); + hclge_rm_vport_all_mac_table(vport, true, HCLGE_MAC_ADDR_MC); + hclge_rm_vport_all_vlan_table(vport, true); + } else { + dev_warn(&hdev->pdev->dev, "Invalid cmd(%u)\n", + msg_cmd->subcode); + } +} + void hclge_mbx_handler(struct hclge_dev *hdev) { struct hclge_cmq_ring *crq = &hdev->hw.cmq.crq; @@ -645,6 +653,7 @@ void hclge_mbx_handler(struct hclge_dev *hdev) struct hclge_mbx_vf_to_pf_cmd *req; struct hclge_vport *vport; struct hclge_desc *desc; + bool is_del = false; unsigned int flag; int ret = 0; @@ -674,6 +683,8 @@ void hclge_mbx_handler(struct hclge_dev *hdev) vport = &hdev->vport[req->mbx_src_vfid]; + trace_hclge_pf_mbx_get(hdev, req); + switch (req->msg.code) { case HCLGE_MBX_MAP_RING_TO_VECTOR: ret = hclge_map_unmap_ring_to_vf_vector(vport, true, @@ -731,7 +742,7 @@ void hclge_mbx_handler(struct hclge_dev *hdev) ret = hclge_get_link_info(vport, req); if (ret) dev_err(&hdev->pdev->dev, - "PF fail(%d) to get link stat for VF\n", + "failed to inform link stat to VF, ret = %d\n", ret); break; case HCLGE_MBX_QUEUE_RESET: @@ -760,11 +771,12 @@ void hclge_mbx_handler(struct hclge_dev *hdev) break; case HCLGE_MBX_GET_VF_FLR_STATUS: case HCLGE_MBX_VF_UNINIT: - hclge_rm_vport_all_mac_table(vport, true, + is_del = req->msg.code == HCLGE_MBX_VF_UNINIT; + hclge_rm_vport_all_mac_table(vport, is_del, HCLGE_MAC_ADDR_UC); - hclge_rm_vport_all_mac_table(vport, true, + hclge_rm_vport_all_mac_table(vport, is_del, HCLGE_MAC_ADDR_MC); - hclge_rm_vport_all_vlan_table(vport, true); + hclge_rm_vport_all_vlan_table(vport, is_del); break; case HCLGE_MBX_GET_MEDIA_TYPE: hclge_get_vf_media_type(vport, &resp_msg); @@ -778,6 +790,9 @@ void hclge_mbx_handler(struct hclge_dev *hdev) case HCLGE_MBX_NCSI_ERROR: hclge_handle_ncsi_error(hdev); break; + case HCLGE_MBX_HANDLE_VF_TBL: + hclge_handle_vf_tbl(vport, req); + break; default: dev_err(&hdev->pdev->dev, "un-supported mailbox message, code = %u\n", diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c index 696c5ae922e3..e89820702540 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_mdio.c @@ -155,7 +155,7 @@ int hclge_mac_mdio_config(struct hclge_dev *hdev) ret = mdiobus_register(mdio_bus); if (ret) { dev_err(mdio_bus->parent, - "Failed to register MDIO bus ret = %#x\n", ret); + "failed to register MDIO bus, ret = %d\n", ret); return ret; } diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h new file mode 100644 index 000000000000..5b0b71bd6120 --- /dev/null +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_trace.h @@ -0,0 +1,87 @@ +/* SPDX-License-Identifier: GPL-2.0+ */ +/* Copyright (c) 2018-2020 Hisilicon Limited. */ + +/* This must be outside ifdef _HCLGE_TRACE_H */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM hns3 + +#if !defined(_HCLGE_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ) +#define _HCLGE_TRACE_H_ + +#include <linux/tracepoint.h> + +#define PF_GET_MBX_LEN (sizeof(struct hclge_mbx_vf_to_pf_cmd) / sizeof(u32)) +#define PF_SEND_MBX_LEN (sizeof(struct hclge_mbx_pf_to_vf_cmd) / sizeof(u32)) + +TRACE_EVENT(hclge_pf_mbx_get, + TP_PROTO( + struct hclge_dev *hdev, + struct hclge_mbx_vf_to_pf_cmd *req), + TP_ARGS(hdev, req), + + TP_STRUCT__entry( + __field(u8, vfid) + __field(u8, code) + __field(u8, subcode) + __string(pciname, pci_name(hdev->pdev)) + __string(devname, &hdev->vport[0].nic.kinfo.netdev->name) + __array(u32, mbx_data, PF_GET_MBX_LEN) + ), + + TP_fast_assign( + __entry->vfid = req->mbx_src_vfid; + __entry->code = req->msg.code; + __entry->subcode = req->msg.subcode; + __assign_str(pciname, pci_name(hdev->pdev)); + __assign_str(devname, &hdev->vport[0].nic.kinfo.netdev->name); + memcpy(__entry->mbx_data, req, + sizeof(struct hclge_mbx_vf_to_pf_cmd)); + ), + + TP_printk( + "%s %s vfid:%u code:%u subcode:%u data:%s", + __get_str(pciname), __get_str(devname), __entry->vfid, + __entry->code, __entry->subcode, + __print_array(__entry->mbx_data, PF_GET_MBX_LEN, sizeof(u32)) + ) +); + +TRACE_EVENT(hclge_pf_mbx_send, + TP_PROTO( + struct hclge_dev *hdev, + struct hclge_mbx_pf_to_vf_cmd *req), + TP_ARGS(hdev, req), + + TP_STRUCT__entry( + __field(u8, vfid) + __field(u16, code) + __string(pciname, pci_name(hdev->pdev)) + __string(devname, &hdev->vport[0].nic.kinfo.netdev->name) + __array(u32, mbx_data, PF_SEND_MBX_LEN) + ), + + TP_fast_assign( + __entry->vfid = req->dest_vfid; + __entry->code = req->msg.code; + __assign_str(pciname, pci_name(hdev->pdev)); + __assign_str(devname, &hdev->vport[0].nic.kinfo.netdev->name); + memcpy(__entry->mbx_data, req, + sizeof(struct hclge_mbx_pf_to_vf_cmd)); + ), + + TP_printk( + "%s %s vfid:%u code:%u data:%s", + __get_str(pciname), __get_str(devname), __entry->vfid, + __entry->code, + __print_array(__entry->mbx_data, PF_SEND_MBX_LEN, sizeof(u32)) + ) +); + +#endif /* _HCLGE_TRACE_H_ */ + +/* This must be outside ifdef _HCLGE_TRACE_H */ +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE hclge_trace +#include <trace/define_trace.h> diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/Makefile b/drivers/net/ethernet/hisilicon/hns3/hns3vf/Makefile index 53804d95ea90..2c26ea607a53 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/Makefile +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/Makefile @@ -4,6 +4,7 @@ # ccflags-y := -I $(srctree)/drivers/net/ethernet/hisilicon/hns3 +ccflags-y += -I $(srctree)/$(src) obj-$(CONFIG_HNS3_HCLGEVF) += hclgevf.o hclgevf-objs = hclgevf_main.o hclgevf_cmd.o hclgevf_mbx.o diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c index f38d236ebf4f..fec65239a3c8 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.c @@ -11,9 +11,6 @@ #include "hclgevf_main.h" #include "hnae3.h" -#define hclgevf_is_csq(ring) ((ring)->flag & HCLGEVF_TYPE_CSQ) -#define hclgevf_ring_to_dma_dir(ring) (hclgevf_is_csq(ring) ? \ - DMA_TO_DEVICE : DMA_FROM_DEVICE) #define cmq_ring_to_dev(ring) (&(ring)->dev->pdev->dev) static int hclgevf_ring_space(struct hclgevf_cmq_ring *ring) diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.h index f830eef02e5c..40d6e602ab51 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_cmd.h @@ -161,8 +161,8 @@ struct hclgevf_query_res_cmd { #define HCLGEVF_GRO_EN_B 0 struct hclgevf_cfg_gro_status_cmd { - __le16 gro_en; - u8 rsv[22]; + u8 gro_en; + u8 rsv[23]; }; #define HCLGEVF_RSS_DEFAULT_OUTPORT_B 4 diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c index e02d427131ee..1b9578d0bd80 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.c @@ -46,7 +46,7 @@ static const u32 cmdq_reg_addr_list[] = {HCLGEVF_CMDQ_TX_ADDR_L_REG, HCLGEVF_CMDQ_RX_TAIL_REG, HCLGEVF_CMDQ_RX_HEAD_REG, HCLGEVF_VECTOR0_CMDQ_SRC_REG, - HCLGEVF_CMDQ_INTR_STS_REG, + HCLGEVF_VECTOR0_CMDQ_STATE_REG, HCLGEVF_CMDQ_INTR_EN_REG, HCLGEVF_CMDQ_INTR_GEN_REG}; @@ -669,8 +669,8 @@ static int hclgevf_set_rss_tc_mode(struct hclgevf_dev *hdev, u16 rss_size) u16 tc_size[HCLGEVF_MAX_TC_NUM]; struct hclgevf_desc desc; u16 roundup_size; - int status; unsigned int i; + int status; req = (struct hclgevf_rss_tc_mode_cmd *)desc.data; @@ -1143,7 +1143,6 @@ static int hclgevf_cmd_set_promisc_mode(struct hclgevf_dev *hdev, send_msg.en_mc = en_mc_pmc ? 1 : 0; ret = hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); - if (ret) dev_err(&hdev->pdev->dev, "Set promisc mode fail, status is %d.\n", ret); @@ -1164,6 +1163,27 @@ static int hclgevf_set_promisc_mode(struct hnae3_handle *handle, bool en_uc_pmc, en_bc_pmc); } +static void hclgevf_request_update_promisc_mode(struct hnae3_handle *handle) +{ + struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); + + set_bit(HCLGEVF_STATE_PROMISC_CHANGED, &hdev->state); +} + +static void hclgevf_sync_promisc_mode(struct hclgevf_dev *hdev) +{ + struct hnae3_handle *handle = &hdev->nic; + bool en_uc_pmc = handle->netdev_flags & HNAE3_UPE; + bool en_mc_pmc = handle->netdev_flags & HNAE3_MPE; + int ret; + + if (test_bit(HCLGEVF_STATE_PROMISC_CHANGED, &hdev->state)) { + ret = hclgevf_set_promisc_mode(handle, en_uc_pmc, en_mc_pmc); + if (!ret) + clear_bit(HCLGEVF_STATE_PROMISC_CHANGED, &hdev->state); + } +} + static int hclgevf_tqp_enable(struct hclgevf_dev *hdev, unsigned int tqp_id, int stream_id, bool enable) { @@ -1245,10 +1265,12 @@ static int hclgevf_set_mac_addr(struct hnae3_handle *handle, void *p, int status; hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_UNICAST, 0); - send_msg.subcode = is_first ? HCLGE_MBX_MAC_VLAN_UC_ADD : - HCLGE_MBX_MAC_VLAN_UC_MODIFY; + send_msg.subcode = HCLGE_MBX_MAC_VLAN_UC_MODIFY; ether_addr_copy(send_msg.data, new_mac_addr); - ether_addr_copy(&send_msg.data[ETH_ALEN], old_mac_addr); + if (is_first && !hdev->has_pf_mac) + eth_zero_addr(&send_msg.data[ETH_ALEN]); + else + ether_addr_copy(&send_msg.data[ETH_ALEN], old_mac_addr); status = hclgevf_send_mbx_msg(hdev, &send_msg, true, NULL, 0); if (!status) ether_addr_copy(hdev->hw.mac.mac_addr, new_mac_addr); @@ -1256,54 +1278,302 @@ static int hclgevf_set_mac_addr(struct hnae3_handle *handle, void *p, return status; } -static int hclgevf_add_uc_addr(struct hnae3_handle *handle, - const unsigned char *addr) +static struct hclgevf_mac_addr_node * +hclgevf_find_mac_node(struct list_head *list, const u8 *mac_addr) +{ + struct hclgevf_mac_addr_node *mac_node, *tmp; + + list_for_each_entry_safe(mac_node, tmp, list, node) + if (ether_addr_equal(mac_addr, mac_node->mac_addr)) + return mac_node; + + return NULL; +} + +static void hclgevf_update_mac_node(struct hclgevf_mac_addr_node *mac_node, + enum HCLGEVF_MAC_NODE_STATE state) +{ + switch (state) { + /* from set_rx_mode or tmp_add_list */ + case HCLGEVF_MAC_TO_ADD: + if (mac_node->state == HCLGEVF_MAC_TO_DEL) + mac_node->state = HCLGEVF_MAC_ACTIVE; + break; + /* only from set_rx_mode */ + case HCLGEVF_MAC_TO_DEL: + if (mac_node->state == HCLGEVF_MAC_TO_ADD) { + list_del(&mac_node->node); + kfree(mac_node); + } else { + mac_node->state = HCLGEVF_MAC_TO_DEL; + } + break; + /* only from tmp_add_list, the mac_node->state won't be + * HCLGEVF_MAC_ACTIVE + */ + case HCLGEVF_MAC_ACTIVE: + if (mac_node->state == HCLGEVF_MAC_TO_ADD) + mac_node->state = HCLGEVF_MAC_ACTIVE; + break; + } +} + +static int hclgevf_update_mac_list(struct hnae3_handle *handle, + enum HCLGEVF_MAC_NODE_STATE state, + enum HCLGEVF_MAC_ADDR_TYPE mac_type, + const unsigned char *addr) { struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); - struct hclge_vf_to_pf_msg send_msg; + struct hclgevf_mac_addr_node *mac_node; + struct list_head *list; - hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_UNICAST, - HCLGE_MBX_MAC_VLAN_UC_ADD); - ether_addr_copy(send_msg.data, addr); - return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); + list = (mac_type == HCLGEVF_MAC_ADDR_UC) ? + &hdev->mac_table.uc_mac_list : &hdev->mac_table.mc_mac_list; + + spin_lock_bh(&hdev->mac_table.mac_list_lock); + + /* if the mac addr is already in the mac list, no need to add a new + * one into it, just check the mac addr state, convert it to a new + * new state, or just remove it, or do nothing. + */ + mac_node = hclgevf_find_mac_node(list, addr); + if (mac_node) { + hclgevf_update_mac_node(mac_node, state); + spin_unlock_bh(&hdev->mac_table.mac_list_lock); + return 0; + } + /* if this address is never added, unnecessary to delete */ + if (state == HCLGEVF_MAC_TO_DEL) { + spin_unlock_bh(&hdev->mac_table.mac_list_lock); + return -ENOENT; + } + + mac_node = kzalloc(sizeof(*mac_node), GFP_ATOMIC); + if (!mac_node) { + spin_unlock_bh(&hdev->mac_table.mac_list_lock); + return -ENOMEM; + } + + mac_node->state = state; + ether_addr_copy(mac_node->mac_addr, addr); + list_add_tail(&mac_node->node, list); + + spin_unlock_bh(&hdev->mac_table.mac_list_lock); + return 0; +} + +static int hclgevf_add_uc_addr(struct hnae3_handle *handle, + const unsigned char *addr) +{ + return hclgevf_update_mac_list(handle, HCLGEVF_MAC_TO_ADD, + HCLGEVF_MAC_ADDR_UC, addr); } static int hclgevf_rm_uc_addr(struct hnae3_handle *handle, const unsigned char *addr) { - struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); - struct hclge_vf_to_pf_msg send_msg; - - hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_UNICAST, - HCLGE_MBX_MAC_VLAN_UC_REMOVE); - ether_addr_copy(send_msg.data, addr); - return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); + return hclgevf_update_mac_list(handle, HCLGEVF_MAC_TO_DEL, + HCLGEVF_MAC_ADDR_UC, addr); } static int hclgevf_add_mc_addr(struct hnae3_handle *handle, const unsigned char *addr) { - struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); - struct hclge_vf_to_pf_msg send_msg; - - hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_MULTICAST, - HCLGE_MBX_MAC_VLAN_MC_ADD); - ether_addr_copy(send_msg.data, addr); - return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); + return hclgevf_update_mac_list(handle, HCLGEVF_MAC_TO_ADD, + HCLGEVF_MAC_ADDR_MC, addr); } static int hclgevf_rm_mc_addr(struct hnae3_handle *handle, const unsigned char *addr) { - struct hclgevf_dev *hdev = hclgevf_ae_get_hdev(handle); + return hclgevf_update_mac_list(handle, HCLGEVF_MAC_TO_DEL, + HCLGEVF_MAC_ADDR_MC, addr); +} + +static int hclgevf_add_del_mac_addr(struct hclgevf_dev *hdev, + struct hclgevf_mac_addr_node *mac_node, + enum HCLGEVF_MAC_ADDR_TYPE mac_type) +{ struct hclge_vf_to_pf_msg send_msg; + u8 code, subcode; + + if (mac_type == HCLGEVF_MAC_ADDR_UC) { + code = HCLGE_MBX_SET_UNICAST; + if (mac_node->state == HCLGEVF_MAC_TO_ADD) + subcode = HCLGE_MBX_MAC_VLAN_UC_ADD; + else + subcode = HCLGE_MBX_MAC_VLAN_UC_REMOVE; + } else { + code = HCLGE_MBX_SET_MULTICAST; + if (mac_node->state == HCLGEVF_MAC_TO_ADD) + subcode = HCLGE_MBX_MAC_VLAN_MC_ADD; + else + subcode = HCLGE_MBX_MAC_VLAN_MC_REMOVE; + } - hclgevf_build_send_msg(&send_msg, HCLGE_MBX_SET_MULTICAST, - HCLGE_MBX_MAC_VLAN_MC_REMOVE); - ether_addr_copy(send_msg.data, addr); + hclgevf_build_send_msg(&send_msg, code, subcode); + ether_addr_copy(send_msg.data, mac_node->mac_addr); return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); } +static void hclgevf_config_mac_list(struct hclgevf_dev *hdev, + struct list_head *list, + enum HCLGEVF_MAC_ADDR_TYPE mac_type) +{ + struct hclgevf_mac_addr_node *mac_node, *tmp; + int ret; + + list_for_each_entry_safe(mac_node, tmp, list, node) { + ret = hclgevf_add_del_mac_addr(hdev, mac_node, mac_type); + if (ret) { + dev_err(&hdev->pdev->dev, + "failed to configure mac %pM, state = %d, ret = %d\n", + mac_node->mac_addr, mac_node->state, ret); + return; + } + if (mac_node->state == HCLGEVF_MAC_TO_ADD) { + mac_node->state = HCLGEVF_MAC_ACTIVE; + } else { + list_del(&mac_node->node); + kfree(mac_node); + } + } +} + +static void hclgevf_sync_from_add_list(struct list_head *add_list, + struct list_head *mac_list) +{ + struct hclgevf_mac_addr_node *mac_node, *tmp, *new_node; + + list_for_each_entry_safe(mac_node, tmp, add_list, node) { + /* if the mac address from tmp_add_list is not in the + * uc/mc_mac_list, it means have received a TO_DEL request + * during the time window of sending mac config request to PF + * If mac_node state is ACTIVE, then change its state to TO_DEL, + * then it will be removed at next time. If is TO_ADD, it means + * send TO_ADD request failed, so just remove the mac node. + */ + new_node = hclgevf_find_mac_node(mac_list, mac_node->mac_addr); + if (new_node) { + hclgevf_update_mac_node(new_node, mac_node->state); + list_del(&mac_node->node); + kfree(mac_node); + } else if (mac_node->state == HCLGEVF_MAC_ACTIVE) { + mac_node->state = HCLGEVF_MAC_TO_DEL; + list_del(&mac_node->node); + list_add_tail(&mac_node->node, mac_list); + } else { + list_del(&mac_node->node); + kfree(mac_node); + } + } +} + +static void hclgevf_sync_from_del_list(struct list_head *del_list, + struct list_head *mac_list) +{ + struct hclgevf_mac_addr_node *mac_node, *tmp, *new_node; + + list_for_each_entry_safe(mac_node, tmp, del_list, node) { + new_node = hclgevf_find_mac_node(mac_list, mac_node->mac_addr); + if (new_node) { + /* If the mac addr is exist in the mac list, it means + * received a new request TO_ADD during the time window + * of sending mac addr configurrequest to PF, so just + * change the mac state to ACTIVE. + */ + new_node->state = HCLGEVF_MAC_ACTIVE; + list_del(&mac_node->node); + kfree(mac_node); + } else { + list_del(&mac_node->node); + list_add_tail(&mac_node->node, mac_list); + } + } +} + +static void hclgevf_clear_list(struct list_head *list) +{ + struct hclgevf_mac_addr_node *mac_node, *tmp; + + list_for_each_entry_safe(mac_node, tmp, list, node) { + list_del(&mac_node->node); + kfree(mac_node); + } +} + +static void hclgevf_sync_mac_list(struct hclgevf_dev *hdev, + enum HCLGEVF_MAC_ADDR_TYPE mac_type) +{ + struct hclgevf_mac_addr_node *mac_node, *tmp, *new_node; + struct list_head tmp_add_list, tmp_del_list; + struct list_head *list; + + INIT_LIST_HEAD(&tmp_add_list); + INIT_LIST_HEAD(&tmp_del_list); + + /* move the mac addr to the tmp_add_list and tmp_del_list, then + * we can add/delete these mac addr outside the spin lock + */ + list = (mac_type == HCLGEVF_MAC_ADDR_UC) ? + &hdev->mac_table.uc_mac_list : &hdev->mac_table.mc_mac_list; + + spin_lock_bh(&hdev->mac_table.mac_list_lock); + + list_for_each_entry_safe(mac_node, tmp, list, node) { + switch (mac_node->state) { + case HCLGEVF_MAC_TO_DEL: + list_del(&mac_node->node); + list_add_tail(&mac_node->node, &tmp_del_list); + break; + case HCLGEVF_MAC_TO_ADD: + new_node = kzalloc(sizeof(*new_node), GFP_ATOMIC); + if (!new_node) + goto stop_traverse; + + ether_addr_copy(new_node->mac_addr, mac_node->mac_addr); + new_node->state = mac_node->state; + list_add_tail(&new_node->node, &tmp_add_list); + break; + default: + break; + } + } + +stop_traverse: + spin_unlock_bh(&hdev->mac_table.mac_list_lock); + + /* delete first, in order to get max mac table space for adding */ + hclgevf_config_mac_list(hdev, &tmp_del_list, mac_type); + hclgevf_config_mac_list(hdev, &tmp_add_list, mac_type); + + /* if some mac addresses were added/deleted fail, move back to the + * mac_list, and retry at next time. + */ + spin_lock_bh(&hdev->mac_table.mac_list_lock); + + hclgevf_sync_from_del_list(&tmp_del_list, list); + hclgevf_sync_from_add_list(&tmp_add_list, list); + + spin_unlock_bh(&hdev->mac_table.mac_list_lock); +} + +static void hclgevf_sync_mac_table(struct hclgevf_dev *hdev) +{ + hclgevf_sync_mac_list(hdev, HCLGEVF_MAC_ADDR_UC); + hclgevf_sync_mac_list(hdev, HCLGEVF_MAC_ADDR_MC); +} + +static void hclgevf_uninit_mac_list(struct hclgevf_dev *hdev) +{ + spin_lock_bh(&hdev->mac_table.mac_list_lock); + + hclgevf_clear_list(&hdev->mac_table.uc_mac_list); + hclgevf_clear_list(&hdev->mac_table.mc_mac_list); + + spin_unlock_bh(&hdev->mac_table.mac_list_lock); +} + static int hclgevf_set_vlan_filter(struct hnae3_handle *handle, __be16 proto, u16 vlan_id, bool is_kill) @@ -1506,10 +1776,6 @@ static int hclgevf_reset_stack(struct hclgevf_dev *hdev) if (ret) return ret; - ret = hclgevf_notify_client(hdev, HNAE3_RESTORE_CLIENT); - if (ret) - return ret; - /* clear handshake status with IMP */ hclgevf_reset_handshake(hdev, false); @@ -1559,7 +1825,7 @@ static void hclgevf_dump_rst_info(struct hclgevf_dev *hdev) dev_info(&hdev->pdev->dev, "vector0 interrupt enable status: 0x%x\n", hclgevf_read_dev(&hdev->hw, HCLGEVF_MISC_VECTOR_REG_BASE)); dev_info(&hdev->pdev->dev, "vector0 interrupt status: 0x%x\n", - hclgevf_read_dev(&hdev->hw, HCLGEVF_VECTOR0_CMDQ_STAT_REG)); + hclgevf_read_dev(&hdev->hw, HCLGEVF_VECTOR0_CMDQ_STATE_REG)); dev_info(&hdev->pdev->dev, "handshake status: 0x%x\n", hclgevf_read_dev(&hdev->hw, HCLGEVF_CMDQ_TX_DEPTH_REG)); dev_info(&hdev->pdev->dev, "function reset status: 0x%x\n", @@ -1589,13 +1855,8 @@ static void hclgevf_reset_err_handle(struct hclgevf_dev *hdev) static int hclgevf_reset_prepare(struct hclgevf_dev *hdev) { - struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); int ret; - /* Initialize ae_dev reset status as well, in case enet layer wants to - * know if device is undergoing reset - */ - ae_dev->reset_type = hdev->reset_type; hdev->rst_stats.rst_cnt++; rtnl_lock(); @@ -1610,7 +1871,6 @@ static int hclgevf_reset_prepare(struct hclgevf_dev *hdev) static int hclgevf_reset_rebuild(struct hclgevf_dev *hdev) { - struct hnae3_ae_dev *ae_dev = pci_get_drvdata(hdev->pdev); int ret; hdev->rst_stats.hw_rst_done_cnt++; @@ -1625,7 +1885,6 @@ static int hclgevf_reset_rebuild(struct hclgevf_dev *hdev) } hdev->last_reset_time = jiffies; - ae_dev->reset_type = HNAE3_NONE_RESET; hdev->rst_stats.rst_done_cnt++; hdev->rst_stats.rst_fail_cnt = 0; clear_bit(HCLGEVF_STATE_RST_FAIL, &hdev->state); @@ -1951,6 +2210,10 @@ static void hclgevf_periodic_service_task(struct hclgevf_dev *hdev) hclgevf_sync_vlan_filter(hdev); + hclgevf_sync_mac_table(hdev); + + hclgevf_sync_promisc_mode(hdev); + hdev->last_serv_processed = jiffies; out: @@ -1986,7 +2249,7 @@ static enum hclgevf_evt_cause hclgevf_check_evt_cause(struct hclgevf_dev *hdev, /* fetch the events from their corresponding regs */ cmdq_stat_reg = hclgevf_read_dev(&hdev->hw, - HCLGEVF_VECTOR0_CMDQ_STAT_REG); + HCLGEVF_VECTOR0_CMDQ_STATE_REG); if (BIT(HCLGEVF_VECTOR0_RST_INT_B) & cmdq_stat_reg) { rst_ing_reg = hclgevf_read_dev(&hdev->hw, HCLGEVF_RST_ING); @@ -2139,7 +2402,7 @@ static int hclgevf_config_gro(struct hclgevf_dev *hdev, bool en) false); req = (struct hclgevf_cfg_gro_status_cmd *)desc.data; - req->gro_en = cpu_to_le16(en ? 1 : 0); + req->gro_en = en ? 1 : 0; ret = hclgevf_cmd_send(&hdev->hw, &desc, 1); if (ret) @@ -2313,6 +2576,10 @@ static void hclgevf_state_init(struct hclgevf_dev *hdev) mutex_init(&hdev->mbx_resp.mbx_mutex); sema_init(&hdev->reset_sem, 1); + spin_lock_init(&hdev->mac_table.mac_list_lock); + INIT_LIST_HEAD(&hdev->mac_table.uc_mac_list); + INIT_LIST_HEAD(&hdev->mac_table.mc_mac_list); + /* bring the device down */ set_bit(HCLGEVF_STATE_DOWN, &hdev->state); } @@ -2445,6 +2712,7 @@ static int hclgevf_init_nic_client_instance(struct hnae3_ae_dev *ae_dev, struct hnae3_client *client) { struct hclgevf_dev *hdev = ae_dev->priv; + int rst_cnt = hdev->rst_stats.rst_cnt; int ret; ret = client->ops->init_instance(&hdev->nic); @@ -2452,6 +2720,14 @@ static int hclgevf_init_nic_client_instance(struct hnae3_ae_dev *ae_dev, return ret; set_bit(HCLGEVF_STATE_NIC_REGISTERED, &hdev->state); + if (test_bit(HCLGEVF_STATE_RST_HANDLING, &hdev->state) || + rst_cnt != hdev->rst_stats.rst_cnt) { + clear_bit(HCLGEVF_STATE_NIC_REGISTERED, &hdev->state); + + client->ops->uninit_instance(&hdev->nic, 0); + return -EBUSY; + } + hnae3_set_client_init_flag(client, ae_dev, 1); if (netif_msg_drv(&hdev->nic)) @@ -2695,6 +2971,15 @@ static int hclgevf_pci_reset(struct hclgevf_dev *hdev) return ret; } +static int hclgevf_clear_vport_list(struct hclgevf_dev *hdev) +{ + struct hclge_vf_to_pf_msg send_msg; + + hclgevf_build_send_msg(&send_msg, HCLGE_MBX_HANDLE_VF_TBL, + HCLGE_MBX_VPORT_LIST_CLEAR); + return hclgevf_send_mbx_msg(hdev, &send_msg, false, NULL, 0); +} + static int hclgevf_reset_hdev(struct hclgevf_dev *hdev) { struct pci_dev *pdev = hdev->pdev; @@ -2730,6 +3015,8 @@ static int hclgevf_reset_hdev(struct hclgevf_dev *hdev) return ret; } + set_bit(HCLGEVF_STATE_PROMISC_CHANGED, &hdev->state); + dev_info(&hdev->pdev->dev, "Reset done\n"); return 0; @@ -2802,6 +3089,15 @@ static int hclgevf_init_hdev(struct hclgevf_dev *hdev) goto err_config; } + /* ensure vf tbl list as empty before init*/ + ret = hclgevf_clear_vport_list(hdev); + if (ret) { + dev_err(&pdev->dev, + "failed to clear tbl list configuration, ret = %d.\n", + ret); + goto err_config; + } + ret = hclgevf_init_vlan_config(hdev); if (ret) { dev_err(&hdev->pdev->dev, @@ -2846,6 +3142,7 @@ static void hclgevf_uninit_hdev(struct hclgevf_dev *hdev) hclgevf_pci_uninit(hdev); hclgevf_cmd_uninit(hdev); + hclgevf_uninit_mac_list(hdev); } static int hclgevf_init_ae_dev(struct hnae3_ae_dev *ae_dev) @@ -3213,6 +3510,7 @@ static const struct hnae3_ae_ops hclgevf_ops = { .set_timer_task = hclgevf_set_timer_task, .get_link_mode = hclgevf_get_link_mode, .set_promisc_mode = hclgevf_set_promisc_mode, + .request_update_promisc_mode = hclgevf_request_update_promisc_mode, }; static struct hnae3_ae_algo ae_algovf = { diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h index 3b88d866facc..c1fac8920ae3 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_main.h @@ -42,8 +42,6 @@ #define HCLGEVF_CMDQ_RX_DEPTH_REG 0x27020 #define HCLGEVF_CMDQ_RX_TAIL_REG 0x27024 #define HCLGEVF_CMDQ_RX_HEAD_REG 0x27028 -#define HCLGEVF_CMDQ_INTR_SRC_REG 0x27100 -#define HCLGEVF_CMDQ_INTR_STS_REG 0x27104 #define HCLGEVF_CMDQ_INTR_EN_REG 0x27108 #define HCLGEVF_CMDQ_INTR_GEN_REG 0x2710C @@ -88,7 +86,7 @@ /* Vector0 interrupt CMDQ event source register(RW) */ #define HCLGEVF_VECTOR0_CMDQ_SRC_REG 0x27100 /* Vector0 interrupt CMDQ event status register(RO) */ -#define HCLGEVF_VECTOR0_CMDQ_STAT_REG 0x27104 +#define HCLGEVF_VECTOR0_CMDQ_STATE_REG 0x27104 /* CMDQ register bits for RX event(=MBX event) */ #define HCLGEVF_VECTOR0_RX_CMDQ_INT_B 1 /* RST register bits for RESET event */ @@ -148,6 +146,7 @@ enum hclgevf_states { HCLGEVF_STATE_MBX_HANDLING, HCLGEVF_STATE_CMD_DISABLE, HCLGEVF_STATE_LINK_UPDATING, + HCLGEVF_STATE_PROMISC_CHANGED, HCLGEVF_STATE_RST_FAIL, }; @@ -234,6 +233,29 @@ struct hclgevf_rst_stats { u32 rst_fail_cnt; /* the number of VF reset fail */ }; +enum HCLGEVF_MAC_ADDR_TYPE { + HCLGEVF_MAC_ADDR_UC, + HCLGEVF_MAC_ADDR_MC +}; + +enum HCLGEVF_MAC_NODE_STATE { + HCLGEVF_MAC_TO_ADD, + HCLGEVF_MAC_TO_DEL, + HCLGEVF_MAC_ACTIVE +}; + +struct hclgevf_mac_addr_node { + struct list_head node; + enum HCLGEVF_MAC_NODE_STATE state; + u8 mac_addr[ETH_ALEN]; +}; + +struct hclgevf_mac_table_cfg { + spinlock_t mac_list_lock; /* protect mac address need to add/detele */ + struct list_head uc_mac_list; + struct list_head mc_mac_list; +}; + struct hclgevf_dev { struct pci_dev *pdev; struct hnae3_ae_dev *ae_dev; @@ -256,7 +278,7 @@ struct hclgevf_dev { struct semaphore reset_sem; /* protect reset process */ u32 fw_version; - u16 num_tqps; /* num task queue pairs of this PF */ + u16 num_tqps; /* num task queue pairs of this VF */ u16 alloc_rss_size; /* allocated RSS task queue */ u16 rss_size_max; /* HW defined max RSS task queue */ @@ -282,6 +304,8 @@ struct hclgevf_dev { unsigned long vlan_del_fail_bmap[BITS_TO_LONGS(VLAN_N_VID)]; + struct hclgevf_mac_table_cfg mac_table; + bool mbx_event_pending; struct hclgevf_mbx_resp_status mbx_resp; /* mailbox response */ struct hclgevf_mbx_arq_ring arq; /* mailbox async rx queue */ diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c index 9b8154955f91..5b2dcd97c107 100644 --- a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_mbx.c @@ -5,6 +5,9 @@ #include "hclgevf_main.h" #include "hnae3.h" +#define CREATE_TRACE_POINTS +#include "hclgevf_trace.h" + static int hclgevf_resp_to_errno(u16 resp_code) { return resp_code ? -resp_code : 0; @@ -106,6 +109,8 @@ int hclgevf_send_mbx_msg(struct hclgevf_dev *hdev, memcpy(&req->msg, send_msg, sizeof(struct hclge_vf_to_pf_msg)); + trace_hclge_vf_mbx_send(hdev, req); + /* synchronous send */ if (need_resp) { mutex_lock(&hdev->mbx_resp.mbx_mutex); @@ -179,6 +184,8 @@ void hclgevf_mbx_handler(struct hclgevf_dev *hdev) continue; } + trace_hclge_vf_mbx_get(hdev, req); + /* synchronous messages are time critical and need preferential * treatment. Therefore, we need to acknowledge all the sync * responses as quickly as possible so that waiting tasks do not diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h new file mode 100644 index 000000000000..e4bfb6191fef --- /dev/null +++ b/drivers/net/ethernet/hisilicon/hns3/hns3vf/hclgevf_trace.h @@ -0,0 +1,87 @@ +/* SPDX-License-Identifier: GPL-2.0+ */ +/* Copyright (c) 2018-2019 Hisilicon Limited. */ + +/* This must be outside ifdef _HCLGEVF_TRACE_H */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM hns3 + +#if !defined(_HCLGEVF_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ) +#define _HCLGEVF_TRACE_H_ + +#include <linux/tracepoint.h> + +#define VF_GET_MBX_LEN (sizeof(struct hclge_mbx_pf_to_vf_cmd) / sizeof(u32)) +#define VF_SEND_MBX_LEN (sizeof(struct hclge_mbx_vf_to_pf_cmd) / sizeof(u32)) + +TRACE_EVENT(hclge_vf_mbx_get, + TP_PROTO( + struct hclgevf_dev *hdev, + struct hclge_mbx_pf_to_vf_cmd *req), + TP_ARGS(hdev, req), + + TP_STRUCT__entry( + __field(u8, vfid) + __field(u16, code) + __string(pciname, pci_name(hdev->pdev)) + __string(devname, &hdev->nic.kinfo.netdev->name) + __array(u32, mbx_data, VF_GET_MBX_LEN) + ), + + TP_fast_assign( + __entry->vfid = req->dest_vfid; + __entry->code = req->msg.code; + __assign_str(pciname, pci_name(hdev->pdev)); + __assign_str(devname, &hdev->nic.kinfo.netdev->name); + memcpy(__entry->mbx_data, req, + sizeof(struct hclge_mbx_pf_to_vf_cmd)); + ), + + TP_printk( + "%s %s vfid:%u code:%u data:%s", + __get_str(pciname), __get_str(devname), __entry->vfid, + __entry->code, + __print_array(__entry->mbx_data, VF_GET_MBX_LEN, sizeof(u32)) + ) +); + +TRACE_EVENT(hclge_vf_mbx_send, + TP_PROTO( + struct hclgevf_dev *hdev, + struct hclge_mbx_vf_to_pf_cmd *req), + TP_ARGS(hdev, req), + + TP_STRUCT__entry( + __field(u8, vfid) + __field(u8, code) + __field(u8, subcode) + __string(pciname, pci_name(hdev->pdev)) + __string(devname, &hdev->nic.kinfo.netdev->name) + __array(u32, mbx_data, VF_SEND_MBX_LEN) + ), + + TP_fast_assign( + __entry->vfid = req->mbx_src_vfid; + __entry->code = req->msg.code; + __entry->subcode = req->msg.subcode; + __assign_str(pciname, pci_name(hdev->pdev)); + __assign_str(devname, &hdev->nic.kinfo.netdev->name); + memcpy(__entry->mbx_data, req, + sizeof(struct hclge_mbx_vf_to_pf_cmd)); + ), + + TP_printk( + "%s %s vfid:%u code:%u subcode:%u data:%s", + __get_str(pciname), __get_str(devname), __entry->vfid, + __entry->code, __entry->subcode, + __print_array(__entry->mbx_data, VF_SEND_MBX_LEN, sizeof(u32)) + ) +); + +#endif /* _HCLGEVF_TRACE_H_ */ + +/* This must be outside ifdef _HCLGEVF_TRACE_H */ +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE hclgevf_trace +#include <trace/define_trace.h> diff --git a/drivers/net/ethernet/huawei/hinic/Makefile b/drivers/net/ethernet/huawei/hinic/Makefile index fe88ab88cacc..32a011ca44c3 100644 --- a/drivers/net/ethernet/huawei/hinic/Makefile +++ b/drivers/net/ethernet/huawei/hinic/Makefile @@ -4,4 +4,4 @@ obj-$(CONFIG_HINIC) += hinic.o hinic-y := hinic_main.o hinic_tx.o hinic_rx.o hinic_port.o hinic_hw_dev.o \ hinic_hw_io.o hinic_hw_qp.o hinic_hw_cmdq.o hinic_hw_wq.o \ hinic_hw_mgmt.o hinic_hw_api_cmd.o hinic_hw_eqs.o hinic_hw_if.o \ - hinic_common.o hinic_ethtool.o + hinic_common.o hinic_ethtool.o hinic_hw_mbox.o hinic_sriov.o diff --git a/drivers/net/ethernet/huawei/hinic/hinic_dev.h b/drivers/net/ethernet/huawei/hinic/hinic_dev.h index a209b14160cc..48b40be3e84d 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_dev.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_dev.h @@ -16,6 +16,7 @@ #include "hinic_hw_dev.h" #include "hinic_tx.h" #include "hinic_rx.h" +#include "hinic_sriov.h" #define HINIC_DRV_NAME "hinic" @@ -23,6 +24,7 @@ enum hinic_flags { HINIC_LINK_UP = BIT(0), HINIC_INTF_UP = BIT(1), HINIC_RSS_ENABLE = BIT(2), + HINIC_LINK_DOWN = BIT(3), }; struct hinic_rx_mode_work { @@ -67,6 +69,8 @@ struct hinic_dev { struct hinic_txq *txqs; struct hinic_rxq *rxqs; + u16 sq_depth; + u16 rq_depth; struct hinic_txq_stats tx_stats; struct hinic_rxq_stats rx_stats; @@ -78,6 +82,7 @@ struct hinic_dev { struct hinic_rss_type rss_type; u8 *rss_hkey_user; s32 *rss_indir_user; + struct hinic_sriov_info sriov_info; }; #endif diff --git a/drivers/net/ethernet/huawei/hinic/hinic_ethtool.c b/drivers/net/ethernet/huawei/hinic/hinic_ethtool.c index 966aea949c0b..efb02e03e7da 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_ethtool.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_ethtool.c @@ -33,6 +33,99 @@ #include "hinic_rx.h" #include "hinic_dev.h" +#define SET_LINK_STR_MAX_LEN 128 + +#define GET_SUPPORTED_MODE 0 +#define GET_ADVERTISED_MODE 1 + +#define ETHTOOL_ADD_SUPPORTED_SPEED_LINK_MODE(ecmd, mode) \ + ((ecmd)->supported |= \ + (1UL << hw_to_ethtool_link_mode_table[mode].link_mode_bit)) +#define ETHTOOL_ADD_ADVERTISED_SPEED_LINK_MODE(ecmd, mode) \ + ((ecmd)->advertising |= \ + (1UL << hw_to_ethtool_link_mode_table[mode].link_mode_bit)) +#define ETHTOOL_ADD_SUPPORTED_LINK_MODE(ecmd, mode) \ + ((ecmd)->supported |= SUPPORTED_##mode) +#define ETHTOOL_ADD_ADVERTISED_LINK_MODE(ecmd, mode) \ + ((ecmd)->advertising |= ADVERTISED_##mode) + +struct hw2ethtool_link_mode { + enum ethtool_link_mode_bit_indices link_mode_bit; + u32 speed; + enum hinic_link_mode hw_link_mode; +}; + +struct cmd_link_settings { + u64 supported; + u64 advertising; + + u32 speed; + u8 duplex; + u8 port; + u8 autoneg; +}; + +static u32 hw_to_ethtool_speed[LINK_SPEED_LEVELS] = { + SPEED_10, SPEED_100, + SPEED_1000, SPEED_10000, + SPEED_25000, SPEED_40000, + SPEED_100000 +}; + +static struct hw2ethtool_link_mode + hw_to_ethtool_link_mode_table[HINIC_LINK_MODE_NUMBERS] = { + { + .link_mode_bit = ETHTOOL_LINK_MODE_10000baseKR_Full_BIT, + .speed = SPEED_10000, + .hw_link_mode = HINIC_10GE_BASE_KR, + }, + { + .link_mode_bit = ETHTOOL_LINK_MODE_40000baseKR4_Full_BIT, + .speed = SPEED_40000, + .hw_link_mode = HINIC_40GE_BASE_KR4, + }, + { + .link_mode_bit = ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT, + .speed = SPEED_40000, + .hw_link_mode = HINIC_40GE_BASE_CR4, + }, + { + .link_mode_bit = ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT, + .speed = SPEED_100000, + .hw_link_mode = HINIC_100GE_BASE_KR4, + }, + { + .link_mode_bit = ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT, + .speed = SPEED_100000, + .hw_link_mode = HINIC_100GE_BASE_CR4, + }, + { + .link_mode_bit = ETHTOOL_LINK_MODE_25000baseKR_Full_BIT, + .speed = SPEED_25000, + .hw_link_mode = HINIC_25GE_BASE_KR_S, + }, + { + .link_mode_bit = ETHTOOL_LINK_MODE_25000baseCR_Full_BIT, + .speed = SPEED_25000, + .hw_link_mode = HINIC_25GE_BASE_CR_S, + }, + { + .link_mode_bit = ETHTOOL_LINK_MODE_25000baseKR_Full_BIT, + .speed = SPEED_25000, + .hw_link_mode = HINIC_25GE_BASE_KR, + }, + { + .link_mode_bit = ETHTOOL_LINK_MODE_25000baseCR_Full_BIT, + .speed = SPEED_25000, + .hw_link_mode = HINIC_25GE_BASE_CR, + }, + { + .link_mode_bit = ETHTOOL_LINK_MODE_1000baseKX_Full_BIT, + .speed = SPEED_1000, + .hw_link_mode = HINIC_GE_BASE_KX, + }, +}; + static void set_link_speed(struct ethtool_link_ksettings *link_ksettings, enum hinic_speed speed) { @@ -71,18 +164,91 @@ static void set_link_speed(struct ethtool_link_ksettings *link_ksettings, } } +static int hinic_get_link_mode_index(enum hinic_link_mode link_mode) +{ + int i = 0; + + for (i = 0; i < HINIC_LINK_MODE_NUMBERS; i++) { + if (link_mode == hw_to_ethtool_link_mode_table[i].hw_link_mode) + break; + } + + return i; +} + +static void hinic_add_ethtool_link_mode(struct cmd_link_settings *link_settings, + enum hinic_link_mode hw_link_mode, + u32 name) +{ + enum hinic_link_mode link_mode; + int idx = 0; + + for (link_mode = 0; link_mode < HINIC_LINK_MODE_NUMBERS; link_mode++) { + if (hw_link_mode & ((u32)1 << link_mode)) { + idx = hinic_get_link_mode_index(link_mode); + if (idx >= HINIC_LINK_MODE_NUMBERS) + continue; + + if (name == GET_SUPPORTED_MODE) + ETHTOOL_ADD_SUPPORTED_SPEED_LINK_MODE + (link_settings, idx); + else + ETHTOOL_ADD_ADVERTISED_SPEED_LINK_MODE + (link_settings, idx); + } + } +} + +static void hinic_link_port_type(struct cmd_link_settings *link_settings, + enum hinic_port_type port_type) +{ + switch (port_type) { + case HINIC_PORT_ELEC: + case HINIC_PORT_TP: + ETHTOOL_ADD_SUPPORTED_LINK_MODE(link_settings, TP); + ETHTOOL_ADD_ADVERTISED_LINK_MODE(link_settings, TP); + link_settings->port = PORT_TP; + break; + + case HINIC_PORT_AOC: + case HINIC_PORT_FIBRE: + ETHTOOL_ADD_SUPPORTED_LINK_MODE(link_settings, FIBRE); + ETHTOOL_ADD_ADVERTISED_LINK_MODE(link_settings, FIBRE); + link_settings->port = PORT_FIBRE; + break; + + case HINIC_PORT_COPPER: + ETHTOOL_ADD_SUPPORTED_LINK_MODE(link_settings, FIBRE); + ETHTOOL_ADD_ADVERTISED_LINK_MODE(link_settings, FIBRE); + link_settings->port = PORT_DA; + break; + + case HINIC_PORT_BACKPLANE: + ETHTOOL_ADD_SUPPORTED_LINK_MODE(link_settings, Backplane); + ETHTOOL_ADD_ADVERTISED_LINK_MODE(link_settings, Backplane); + link_settings->port = PORT_NONE; + break; + + default: + link_settings->port = PORT_OTHER; + break; + } +} + static int hinic_get_link_ksettings(struct net_device *netdev, struct ethtool_link_ksettings *link_ksettings) { struct hinic_dev *nic_dev = netdev_priv(netdev); + struct hinic_link_mode_cmd link_mode = { 0 }; + struct hinic_pause_config pause_info = { 0 }; + struct cmd_link_settings settings = { 0 }; enum hinic_port_link_state link_state; struct hinic_port_cap port_cap; int err; + ethtool_link_ksettings_zero_link_mode(link_ksettings, supported); ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising); - ethtool_link_ksettings_add_link_mode(link_ksettings, supported, - Autoneg); link_ksettings->base.speed = SPEED_UNKNOWN; link_ksettings->base.autoneg = AUTONEG_DISABLE; @@ -92,14 +258,19 @@ static int hinic_get_link_ksettings(struct net_device *netdev, if (err) return err; + hinic_link_port_type(&settings, port_cap.port_type); + link_ksettings->base.port = settings.port; + err = hinic_port_link_state(nic_dev, &link_state); if (err) return err; - if (link_state != HINIC_LINK_STATE_UP) - return err; - - set_link_speed(link_ksettings, port_cap.speed); + if (link_state == HINIC_LINK_STATE_UP) { + set_link_speed(link_ksettings, port_cap.speed); + link_ksettings->base.duplex = + (port_cap.duplex == HINIC_DUPLEX_FULL) ? + DUPLEX_FULL : DUPLEX_HALF; + } if (!!(port_cap.autoneg_cap & HINIC_AUTONEG_SUPPORTED)) ethtool_link_ksettings_add_link_mode(link_ksettings, @@ -108,11 +279,243 @@ static int hinic_get_link_ksettings(struct net_device *netdev, if (port_cap.autoneg_state == HINIC_AUTONEG_ACTIVE) link_ksettings->base.autoneg = AUTONEG_ENABLE; - link_ksettings->base.duplex = (port_cap.duplex == HINIC_DUPLEX_FULL) ? - DUPLEX_FULL : DUPLEX_HALF; + err = hinic_get_link_mode(nic_dev->hwdev, &link_mode); + if (err || link_mode.supported == HINIC_SUPPORTED_UNKNOWN || + link_mode.advertised == HINIC_SUPPORTED_UNKNOWN) + return -EIO; + + hinic_add_ethtool_link_mode(&settings, link_mode.supported, + GET_SUPPORTED_MODE); + hinic_add_ethtool_link_mode(&settings, link_mode.advertised, + GET_ADVERTISED_MODE); + + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) { + err = hinic_get_hw_pause_info(nic_dev->hwdev, &pause_info); + if (err) + return err; + ETHTOOL_ADD_SUPPORTED_LINK_MODE(&settings, Pause); + if (pause_info.rx_pause && pause_info.tx_pause) { + ETHTOOL_ADD_ADVERTISED_LINK_MODE(&settings, Pause); + } else if (pause_info.tx_pause) { + ETHTOOL_ADD_ADVERTISED_LINK_MODE(&settings, Asym_Pause); + } else if (pause_info.rx_pause) { + ETHTOOL_ADD_ADVERTISED_LINK_MODE(&settings, Pause); + ETHTOOL_ADD_ADVERTISED_LINK_MODE(&settings, Asym_Pause); + } + } + + bitmap_copy(link_ksettings->link_modes.supported, + (unsigned long *)&settings.supported, + __ETHTOOL_LINK_MODE_MASK_NBITS); + bitmap_copy(link_ksettings->link_modes.advertising, + (unsigned long *)&settings.advertising, + __ETHTOOL_LINK_MODE_MASK_NBITS); + return 0; } +static int hinic_ethtool_to_hw_speed_level(u32 speed) +{ + int i; + + for (i = 0; i < LINK_SPEED_LEVELS; i++) { + if (hw_to_ethtool_speed[i] == speed) + break; + } + + return i; +} + +static bool hinic_is_support_speed(enum hinic_link_mode supported_link, + u32 speed) +{ + enum hinic_link_mode link_mode; + int idx; + + for (link_mode = 0; link_mode < HINIC_LINK_MODE_NUMBERS; link_mode++) { + if (!(supported_link & ((u32)1 << link_mode))) + continue; + + idx = hinic_get_link_mode_index(link_mode); + if (idx >= HINIC_LINK_MODE_NUMBERS) + continue; + + if (hw_to_ethtool_link_mode_table[idx].speed == speed) + return true; + } + + return false; +} + +static bool hinic_is_speed_legal(struct hinic_dev *nic_dev, u32 speed) +{ + struct hinic_link_mode_cmd link_mode = { 0 }; + struct net_device *netdev = nic_dev->netdev; + enum nic_speed_level speed_level = 0; + int err; + + err = hinic_get_link_mode(nic_dev->hwdev, &link_mode); + if (err) + return false; + + if (link_mode.supported == HINIC_SUPPORTED_UNKNOWN || + link_mode.advertised == HINIC_SUPPORTED_UNKNOWN) + return false; + + speed_level = hinic_ethtool_to_hw_speed_level(speed); + if (speed_level >= LINK_SPEED_LEVELS || + !hinic_is_support_speed(link_mode.supported, speed)) { + netif_err(nic_dev, drv, netdev, + "Unsupported speed: %d\n", speed); + return false; + } + + return true; +} + +static int get_link_settings_type(struct hinic_dev *nic_dev, + u8 autoneg, u32 speed, u32 *set_settings) +{ + struct hinic_port_cap port_cap = { 0 }; + int err; + + err = hinic_port_get_cap(nic_dev, &port_cap); + if (err) + return err; + + /* always set autonegotiation */ + if (port_cap.autoneg_cap) + *set_settings |= HILINK_LINK_SET_AUTONEG; + + if (autoneg == AUTONEG_ENABLE) { + if (!port_cap.autoneg_cap) { + netif_err(nic_dev, drv, nic_dev->netdev, "Not support autoneg\n"); + return -EOPNOTSUPP; + } + } else if (speed != (u32)SPEED_UNKNOWN) { + /* set speed only when autoneg is disabled */ + if (!hinic_is_speed_legal(nic_dev, speed)) + return -EINVAL; + *set_settings |= HILINK_LINK_SET_SPEED; + } else { + netif_err(nic_dev, drv, nic_dev->netdev, "Need to set speed when autoneg is off\n"); + return -EOPNOTSUPP; + } + + return 0; +} + +static int set_link_settings_separate_cmd(struct hinic_dev *nic_dev, + u32 set_settings, u8 autoneg, + u32 speed) +{ + enum nic_speed_level speed_level = 0; + int err = 0; + + if (set_settings & HILINK_LINK_SET_AUTONEG) { + err = hinic_set_autoneg(nic_dev->hwdev, + (autoneg == AUTONEG_ENABLE)); + if (err) + netif_err(nic_dev, drv, nic_dev->netdev, "%s autoneg failed\n", + (autoneg == AUTONEG_ENABLE) ? + "Enable" : "Disable"); + else + netif_info(nic_dev, drv, nic_dev->netdev, "%s autoneg successfully\n", + (autoneg == AUTONEG_ENABLE) ? + "Enable" : "Disable"); + } + + if (!err && (set_settings & HILINK_LINK_SET_SPEED)) { + speed_level = hinic_ethtool_to_hw_speed_level(speed); + err = hinic_set_speed(nic_dev->hwdev, speed_level); + if (err) + netif_err(nic_dev, drv, nic_dev->netdev, "Set speed %d failed\n", + speed); + else + netif_info(nic_dev, drv, nic_dev->netdev, "Set speed %d successfully\n", + speed); + } + + return err; +} + +static int hinic_set_settings_to_hw(struct hinic_dev *nic_dev, + u32 set_settings, u8 autoneg, u32 speed) +{ + struct hinic_link_ksettings_info settings = {0}; + char set_link_str[SET_LINK_STR_MAX_LEN] = {0}; + struct net_device *netdev = nic_dev->netdev; + enum nic_speed_level speed_level = 0; + int err; + + err = snprintf(set_link_str, SET_LINK_STR_MAX_LEN, "%s", + (set_settings & HILINK_LINK_SET_AUTONEG) ? + (autoneg ? "autong enable " : "autong disable ") : ""); + if (err < 0 || err >= SET_LINK_STR_MAX_LEN) { + netif_err(nic_dev, drv, netdev, "Failed to snprintf link state, function return(%d) and dest_len(%d)\n", + err, SET_LINK_STR_MAX_LEN); + return -EFAULT; + } + + if (set_settings & HILINK_LINK_SET_SPEED) { + speed_level = hinic_ethtool_to_hw_speed_level(speed); + err = snprintf(set_link_str, SET_LINK_STR_MAX_LEN, + "%sspeed %d ", set_link_str, speed); + if (err <= 0 || err >= SET_LINK_STR_MAX_LEN) { + netif_err(nic_dev, drv, netdev, "Failed to snprintf link speed, function return(%d) and dest_len(%d)\n", + err, SET_LINK_STR_MAX_LEN); + return -EFAULT; + } + } + + settings.func_id = HINIC_HWIF_FUNC_IDX(nic_dev->hwdev->hwif); + settings.valid_bitmap = set_settings; + settings.autoneg = autoneg; + settings.speed = speed_level; + + err = hinic_set_link_settings(nic_dev->hwdev, &settings); + if (err != HINIC_MGMT_CMD_UNSUPPORTED) { + if (err) + netif_err(nic_dev, drv, netdev, "Set %s failed\n", + set_link_str); + else + netif_info(nic_dev, drv, netdev, "Set %s successfully\n", + set_link_str); + + return err; + } + + return set_link_settings_separate_cmd(nic_dev, set_settings, autoneg, + speed); +} + +static int set_link_settings(struct net_device *netdev, u8 autoneg, u32 speed) +{ + struct hinic_dev *nic_dev = netdev_priv(netdev); + u32 set_settings = 0; + int err; + + err = get_link_settings_type(nic_dev, autoneg, speed, &set_settings); + if (err) + return err; + + if (set_settings) + err = hinic_set_settings_to_hw(nic_dev, set_settings, + autoneg, speed); + else + netif_info(nic_dev, drv, netdev, "Nothing changed, exit without setting anything\n"); + + return err; +} + +static int hinic_set_link_ksettings(struct net_device *netdev, const struct + ethtool_link_ksettings *link_settings) +{ + /* only support to set autoneg and speed */ + return set_link_settings(netdev, link_settings->base.autoneg, + link_settings->base.speed); +} + static void hinic_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *info) { @@ -135,26 +538,118 @@ static void hinic_get_drvinfo(struct net_device *netdev, static void hinic_get_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring) { - ring->rx_max_pending = HINIC_RQ_DEPTH; - ring->tx_max_pending = HINIC_SQ_DEPTH; - ring->rx_pending = HINIC_RQ_DEPTH; - ring->tx_pending = HINIC_SQ_DEPTH; + struct hinic_dev *nic_dev = netdev_priv(netdev); + + ring->rx_max_pending = HINIC_MAX_QUEUE_DEPTH; + ring->tx_max_pending = HINIC_MAX_QUEUE_DEPTH; + ring->rx_pending = nic_dev->rq_depth; + ring->tx_pending = nic_dev->sq_depth; +} + +static int check_ringparam_valid(struct hinic_dev *nic_dev, + struct ethtool_ringparam *ring) +{ + if (ring->rx_jumbo_pending || ring->rx_mini_pending) { + netif_err(nic_dev, drv, nic_dev->netdev, + "Unsupported rx_jumbo_pending/rx_mini_pending\n"); + return -EINVAL; + } + + if (ring->tx_pending > HINIC_MAX_QUEUE_DEPTH || + ring->tx_pending < HINIC_MIN_QUEUE_DEPTH || + ring->rx_pending > HINIC_MAX_QUEUE_DEPTH || + ring->rx_pending < HINIC_MIN_QUEUE_DEPTH) { + netif_err(nic_dev, drv, nic_dev->netdev, + "Queue depth out of range [%d-%d]\n", + HINIC_MIN_QUEUE_DEPTH, HINIC_MAX_QUEUE_DEPTH); + return -EINVAL; + } + + return 0; } +static int hinic_set_ringparam(struct net_device *netdev, + struct ethtool_ringparam *ring) +{ + struct hinic_dev *nic_dev = netdev_priv(netdev); + u16 new_sq_depth, new_rq_depth; + int err; + + err = check_ringparam_valid(nic_dev, ring); + if (err) + return err; + + new_sq_depth = (u16)(1U << (u16)ilog2(ring->tx_pending)); + new_rq_depth = (u16)(1U << (u16)ilog2(ring->rx_pending)); + + if (new_sq_depth == nic_dev->sq_depth && + new_rq_depth == nic_dev->rq_depth) + return 0; + + netif_info(nic_dev, drv, netdev, + "Change Tx/Rx ring depth from %d/%d to %d/%d\n", + nic_dev->sq_depth, nic_dev->rq_depth, + new_sq_depth, new_rq_depth); + + nic_dev->sq_depth = new_sq_depth; + nic_dev->rq_depth = new_rq_depth; + + if (netif_running(netdev)) { + netif_info(nic_dev, drv, netdev, "Restarting netdev\n"); + err = hinic_close(netdev); + if (err) { + netif_err(nic_dev, drv, netdev, + "Failed to close netdev\n"); + return -EFAULT; + } + + err = hinic_open(netdev); + if (err) { + netif_err(nic_dev, drv, netdev, + "Failed to open netdev\n"); + return -EFAULT; + } + } + + return 0; +} static void hinic_get_channels(struct net_device *netdev, struct ethtool_channels *channels) { struct hinic_dev *nic_dev = netdev_priv(netdev); struct hinic_hwdev *hwdev = nic_dev->hwdev; - channels->max_rx = hwdev->nic_cap.max_qps; - channels->max_tx = hwdev->nic_cap.max_qps; - channels->max_other = 0; - channels->max_combined = 0; - channels->rx_count = hinic_hwdev_num_qps(hwdev); - channels->tx_count = hinic_hwdev_num_qps(hwdev); - channels->other_count = 0; - channels->combined_count = 0; + channels->max_combined = nic_dev->max_qps; + channels->combined_count = hinic_hwdev_num_qps(hwdev); +} + +static int hinic_set_channels(struct net_device *netdev, + struct ethtool_channels *channels) +{ + struct hinic_dev *nic_dev = netdev_priv(netdev); + unsigned int count = channels->combined_count; + int err; + + netif_info(nic_dev, drv, netdev, "Set max combined queue number from %d to %d\n", + hinic_hwdev_num_qps(nic_dev->hwdev), count); + + if (netif_running(netdev)) { + netif_info(nic_dev, drv, netdev, "Restarting netdev\n"); + hinic_close(netdev); + + nic_dev->hwdev->nic_cap.num_qps = count; + + err = hinic_open(netdev); + if (err) { + netif_err(nic_dev, drv, netdev, + "Failed to open netdev\n"); + return -EFAULT; + } + } else { + nic_dev->hwdev->nic_cap.num_qps = count; + } + + return 0; } static int hinic_get_rss_hash_opts(struct hinic_dev *nic_dev, @@ -741,10 +1236,13 @@ static void hinic_get_strings(struct net_device *netdev, static const struct ethtool_ops hinic_ethtool_ops = { .get_link_ksettings = hinic_get_link_ksettings, + .set_link_ksettings = hinic_set_link_ksettings, .get_drvinfo = hinic_get_drvinfo, .get_link = ethtool_op_get_link, .get_ringparam = hinic_get_ringparam, + .set_ringparam = hinic_set_ringparam, .get_channels = hinic_get_channels, + .set_channels = hinic_set_channels, .get_rxnfc = hinic_get_rxnfc, .set_rxnfc = hinic_set_rxnfc, .get_rxfh_key_size = hinic_get_rxfh_key_size, diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c index 5f2d57d1b2d3..cb5b6e5f787f 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c @@ -64,7 +64,7 @@ #define CMDQ_WQE_SIZE 64 #define CMDQ_DEPTH SZ_4K -#define CMDQ_WQ_PAGE_SIZE SZ_4K +#define CMDQ_WQ_PAGE_SIZE SZ_256K #define WQE_LCMD_SIZE 64 #define WQE_SCMD_SIZE 64 @@ -705,7 +705,7 @@ static void cmdq_init_queue_ctxt(struct hinic_cmdq_ctxt *cmdq_ctxt, /* The data in the HW is in Big Endian Format */ wq_first_page_paddr = be64_to_cpu(*wq->block_vaddr); - pfn = CMDQ_PFN(wq_first_page_paddr, wq->wq_page_size); + pfn = CMDQ_PFN(wq_first_page_paddr, SZ_4K); ctxt_info->curr_wqe_page_pfn = HINIC_CMDQ_CTXT_PAGE_INFO_SET(pfn, CURR_WQE_PAGE_PFN) | @@ -714,16 +714,19 @@ static void cmdq_init_queue_ctxt(struct hinic_cmdq_ctxt *cmdq_ctxt, HINIC_CMDQ_CTXT_PAGE_INFO_SET(1, CEQ_EN) | HINIC_CMDQ_CTXT_PAGE_INFO_SET(cmdq->wrapped, WRAPPED); - /* block PFN - Read Modify Write */ - cmdq_first_block_paddr = cmdq_pages->page_paddr; + if (wq->num_q_pages != 1) { + /* block PFN - Read Modify Write */ + cmdq_first_block_paddr = cmdq_pages->page_paddr; - pfn = CMDQ_PFN(cmdq_first_block_paddr, wq->wq_page_size); + pfn = CMDQ_PFN(cmdq_first_block_paddr, wq->wq_page_size); + } ctxt_info->wq_block_pfn = HINIC_CMDQ_CTXT_BLOCK_INFO_SET(pfn, WQ_BLOCK_PFN) | HINIC_CMDQ_CTXT_BLOCK_INFO_SET(atomic_read(&wq->cons_idx), CI); cmdq_ctxt->func_idx = HINIC_HWIF_FUNC_IDX(cmdqs->hwif); + cmdq_ctxt->ppf_idx = HINIC_HWIF_PPF_IDX(cmdqs->hwif); cmdq_ctxt->cmdq_type = cmdq->cmdq_type; } @@ -795,11 +798,6 @@ static int init_cmdqs_ctxt(struct hinic_hwdev *hwdev, size_t cmdq_ctxts_size; int err; - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "Unsupported PCI function type\n"); - return -EINVAL; - } - cmdq_ctxts_size = HINIC_MAX_CMDQ_TYPES * sizeof(*cmdq_ctxts); cmdq_ctxts = devm_kzalloc(&pdev->dev, cmdq_ctxts_size, GFP_KERNEL); if (!cmdq_ctxts) @@ -851,6 +849,25 @@ err_init_cmdq: return err; } +static int hinic_set_cmdq_depth(struct hinic_hwdev *hwdev, u16 cmdq_depth) +{ + struct hinic_cmd_hw_ioctxt hw_ioctxt = { 0 }; + struct hinic_pfhwdev *pfhwdev; + + pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev); + + hw_ioctxt.func_idx = HINIC_HWIF_FUNC_IDX(hwdev->hwif); + hw_ioctxt.ppf_idx = HINIC_HWIF_PPF_IDX(hwdev->hwif); + + hw_ioctxt.set_cmdq_depth = HW_IOCTXT_SET_CMDQ_DEPTH_ENABLE; + hw_ioctxt.cmdq_depth = (u8)ilog2(cmdq_depth); + + return hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM, + HINIC_COMM_CMD_HWCTXT_SET, + &hw_ioctxt, sizeof(hw_ioctxt), NULL, + NULL, HINIC_MGMT_MSG_SYNC); +} + /** * hinic_init_cmdqs - init all cmdqs * @cmdqs: cmdqs to init @@ -901,8 +918,18 @@ int hinic_init_cmdqs(struct hinic_cmdqs *cmdqs, struct hinic_hwif *hwif, hinic_ceq_register_cb(&func_to_io->ceqs, HINIC_CEQ_CMDQ, cmdqs, cmdq_ceq_handler); + + err = hinic_set_cmdq_depth(hwdev, CMDQ_DEPTH); + if (err) { + dev_err(&hwif->pdev->dev, "Failed to set cmdq depth\n"); + goto err_set_cmdq_depth; + } + return 0; +err_set_cmdq_depth: + hinic_ceq_unregister_cb(&func_to_io->ceqs, HINIC_CEQ_CMDQ); + err_cmdq_ctxt: hinic_wqs_cmdq_free(&cmdqs->cmdq_pages, cmdqs->saved_wqs, HINIC_MAX_CMDQ_TYPES); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.h index 7a434b653faa..3e4b0aef9fe6 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.h @@ -122,7 +122,7 @@ struct hinic_cmdq_ctxt { u16 func_idx; u8 cmdq_type; - u8 rsvd1[1]; + u8 ppf_idx; u8 rsvd2[4]; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_csr.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_csr.h index cdec1d0a3962..7e84e4e33fff 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_csr.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_csr.h @@ -10,7 +10,7 @@ /* HW interface registers */ #define HINIC_CSR_FUNC_ATTR0_ADDR 0x0 #define HINIC_CSR_FUNC_ATTR1_ADDR 0x4 - +#define HINIC_CSR_FUNC_ATTR2_ADDR 0x8 #define HINIC_CSR_FUNC_ATTR4_ADDR 0x10 #define HINIC_CSR_FUNC_ATTR5_ADDR 0x14 diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c index c7c75b772a86..0245da02efbb 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c @@ -15,7 +15,9 @@ #include <linux/jiffies.h> #include <linux/log2.h> #include <linux/err.h> +#include <linux/netdevice.h> +#include "hinic_sriov.h" #include "hinic_hw_if.h" #include "hinic_hw_eqs.h" #include "hinic_hw_mgmt.h" @@ -42,24 +44,6 @@ enum io_status { IO_RUNNING = 1, }; -enum hw_ioctxt_set_cmdq_depth { - HW_IOCTXT_SET_CMDQ_DEPTH_DEFAULT, -}; - -/* HW struct */ -struct hinic_dev_cap { - u8 status; - u8 version; - u8 rsvd0[6]; - - u8 rsvd1[5]; - u8 intr_type; - u8 rsvd2[66]; - u16 max_sqs; - u16 max_rqs; - u8 rsvd3[208]; -}; - /** * get_capability - convert device capabilities to NIC capabilities * @hwdev: the HW device to set and convert device capabilities for @@ -67,16 +51,13 @@ struct hinic_dev_cap { * * Return 0 - Success, negative - Failure **/ -static int get_capability(struct hinic_hwdev *hwdev, - struct hinic_dev_cap *dev_cap) +static int parse_capability(struct hinic_hwdev *hwdev, + struct hinic_dev_cap *dev_cap) { struct hinic_cap *nic_cap = &hwdev->nic_cap; int num_aeqs, num_ceqs, num_irqs; - if (!HINIC_IS_PF(hwdev->hwif) && !HINIC_IS_PPF(hwdev->hwif)) - return -EINVAL; - - if (dev_cap->intr_type != INTR_MSIX_TYPE) + if (!HINIC_IS_VF(hwdev->hwif) && dev_cap->intr_type != INTR_MSIX_TYPE) return -EFAULT; num_aeqs = HINIC_HWIF_NUM_AEQS(hwdev->hwif); @@ -89,13 +70,19 @@ static int get_capability(struct hinic_hwdev *hwdev, if (nic_cap->num_qps > HINIC_Q_CTXT_MAX) nic_cap->num_qps = HINIC_Q_CTXT_MAX; - nic_cap->max_qps = dev_cap->max_sqs + 1; - if (nic_cap->max_qps != (dev_cap->max_rqs + 1)) - return -EFAULT; + if (!HINIC_IS_VF(hwdev->hwif)) + nic_cap->max_qps = dev_cap->max_sqs + 1; + else + nic_cap->max_qps = dev_cap->max_sqs; if (nic_cap->num_qps > nic_cap->max_qps) nic_cap->num_qps = nic_cap->max_qps; + if (!HINIC_IS_VF(hwdev->hwif)) { + nic_cap->max_vf = dev_cap->max_vf; + nic_cap->max_vf_qps = dev_cap->max_vf_sqs + 1; + } + return 0; } @@ -105,27 +92,26 @@ static int get_capability(struct hinic_hwdev *hwdev, * * Return 0 - Success, negative - Failure **/ -static int get_cap_from_fw(struct hinic_pfhwdev *pfhwdev) +static int get_capability(struct hinic_pfhwdev *pfhwdev) { struct hinic_hwdev *hwdev = &pfhwdev->hwdev; struct hinic_hwif *hwif = hwdev->hwif; struct pci_dev *pdev = hwif->pdev; struct hinic_dev_cap dev_cap; - u16 in_len, out_len; + u16 out_len; int err; - in_len = 0; out_len = sizeof(dev_cap); err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_CFGM, - HINIC_CFG_NIC_CAP, &dev_cap, in_len, &dev_cap, - &out_len, HINIC_MGMT_MSG_SYNC); + HINIC_CFG_NIC_CAP, &dev_cap, sizeof(dev_cap), + &dev_cap, &out_len, HINIC_MGMT_MSG_SYNC); if (err) { dev_err(&pdev->dev, "Failed to get capability from FW\n"); return err; } - return get_capability(hwdev, &dev_cap); + return parse_capability(hwdev, &dev_cap); } /** @@ -144,15 +130,14 @@ static int get_dev_cap(struct hinic_hwdev *hwdev) switch (HINIC_FUNC_TYPE(hwif)) { case HINIC_PPF: case HINIC_PF: + case HINIC_VF: pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev); - - err = get_cap_from_fw(pfhwdev); + err = get_capability(pfhwdev); if (err) { - dev_err(&pdev->dev, "Failed to get capability from FW\n"); + dev_err(&pdev->dev, "Failed to get capability\n"); return err; } break; - default: dev_err(&pdev->dev, "Unsupported PCI Function type\n"); return -EINVAL; @@ -225,15 +210,8 @@ static void disable_msix(struct hinic_hwdev *hwdev) int hinic_port_msg_cmd(struct hinic_hwdev *hwdev, enum hinic_port_cmd cmd, void *buf_in, u16 in_size, void *buf_out, u16 *out_size) { - struct hinic_hwif *hwif = hwdev->hwif; - struct pci_dev *pdev = hwif->pdev; struct hinic_pfhwdev *pfhwdev; - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "unsupported PCI Function type\n"); - return -EINVAL; - } - pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev); return hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_L2NIC, cmd, @@ -241,6 +219,19 @@ int hinic_port_msg_cmd(struct hinic_hwdev *hwdev, enum hinic_port_cmd cmd, HINIC_MGMT_MSG_SYNC); } +int hinic_hilink_msg_cmd(struct hinic_hwdev *hwdev, enum hinic_hilink_cmd cmd, + void *buf_in, u16 in_size, void *buf_out, + u16 *out_size) +{ + struct hinic_pfhwdev *pfhwdev; + + pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev); + + return hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_HILINK, cmd, + buf_in, in_size, buf_out, out_size, + HINIC_MGMT_MSG_SYNC); +} + /** * init_fw_ctxt- Init Firmware tables before network mgmt and io operations * @hwdev: the NIC HW device @@ -252,14 +243,9 @@ static int init_fw_ctxt(struct hinic_hwdev *hwdev) struct hinic_hwif *hwif = hwdev->hwif; struct pci_dev *pdev = hwif->pdev; struct hinic_cmd_fw_ctxt fw_ctxt; - u16 out_size; + u16 out_size = sizeof(fw_ctxt); int err; - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "Unsupported PCI Function type\n"); - return -EINVAL; - } - fw_ctxt.func_idx = HINIC_HWIF_FUNC_IDX(hwif); fw_ctxt.rx_buf_sz = HINIC_RX_BUF_SZ; @@ -283,19 +269,13 @@ static int init_fw_ctxt(struct hinic_hwdev *hwdev) * * Return 0 - Success, negative - Failure **/ -static int set_hw_ioctxt(struct hinic_hwdev *hwdev, unsigned int rq_depth, - unsigned int sq_depth) +static int set_hw_ioctxt(struct hinic_hwdev *hwdev, unsigned int sq_depth, + unsigned int rq_depth) { struct hinic_hwif *hwif = hwdev->hwif; struct hinic_cmd_hw_ioctxt hw_ioctxt; - struct pci_dev *pdev = hwif->pdev; struct hinic_pfhwdev *pfhwdev; - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "Unsupported PCI Function type\n"); - return -EINVAL; - } - hw_ioctxt.func_idx = HINIC_HWIF_FUNC_IDX(hwif); hw_ioctxt.ppf_idx = HINIC_HWIF_PPF_IDX(hwif); @@ -374,11 +354,6 @@ static int clear_io_resources(struct hinic_hwdev *hwdev) struct hinic_pfhwdev *pfhwdev; int err; - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "Unsupported PCI Function type\n"); - return -EINVAL; - } - /* sleep 100ms to wait for firmware stopping I/O */ msleep(100); @@ -410,14 +385,8 @@ static int set_resources_state(struct hinic_hwdev *hwdev, { struct hinic_cmd_set_res_state res_state; struct hinic_hwif *hwif = hwdev->hwif; - struct pci_dev *pdev = hwif->pdev; struct hinic_pfhwdev *pfhwdev; - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "Unsupported PCI Function type\n"); - return -EINVAL; - } - res_state.func_idx = HINIC_HWIF_FUNC_IDX(hwif); res_state.state = state; @@ -441,8 +410,8 @@ static int get_base_qpn(struct hinic_hwdev *hwdev, u16 *base_qpn) { struct hinic_cmd_base_qpn cmd_base_qpn; struct hinic_hwif *hwif = hwdev->hwif; + u16 out_size = sizeof(cmd_base_qpn); struct pci_dev *pdev = hwif->pdev; - u16 out_size; int err; cmd_base_qpn.func_idx = HINIC_HWIF_FUNC_IDX(hwif); @@ -466,7 +435,7 @@ static int get_base_qpn(struct hinic_hwdev *hwdev, u16 *base_qpn) * * Return 0 - Success, negative - Failure **/ -int hinic_hwdev_ifup(struct hinic_hwdev *hwdev) +int hinic_hwdev_ifup(struct hinic_hwdev *hwdev, u16 sq_depth, u16 rq_depth) { struct hinic_func_to_io *func_to_io = &hwdev->func_to_io; struct hinic_cap *nic_cap = &hwdev->nic_cap; @@ -488,6 +457,9 @@ int hinic_hwdev_ifup(struct hinic_hwdev *hwdev) num_ceqs = HINIC_HWIF_NUM_CEQS(hwif); ceq_msix_entries = &hwdev->msix_entries[num_aeqs]; + func_to_io->hwdev = hwdev; + func_to_io->sq_depth = sq_depth; + func_to_io->rq_depth = rq_depth; err = hinic_io_init(func_to_io, hwif, nic_cap->max_qps, num_ceqs, ceq_msix_entries); @@ -513,7 +485,7 @@ int hinic_hwdev_ifup(struct hinic_hwdev *hwdev) hinic_db_state_set(hwif, HINIC_DB_ENABLE); } - err = set_hw_ioctxt(hwdev, HINIC_SQ_DEPTH, HINIC_RQ_DEPTH); + err = set_hw_ioctxt(hwdev, sq_depth, rq_depth); if (err) { dev_err(&pdev->dev, "Failed to set HW IO ctxt\n"); goto err_hw_ioctxt; @@ -558,17 +530,10 @@ void hinic_hwdev_cb_register(struct hinic_hwdev *hwdev, u16 in_size, void *buf_out, u16 *out_size)) { - struct hinic_hwif *hwif = hwdev->hwif; - struct pci_dev *pdev = hwif->pdev; struct hinic_pfhwdev *pfhwdev; struct hinic_nic_cb *nic_cb; u8 cmd_cb; - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "unsupported PCI Function type\n"); - return; - } - pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev); cmd_cb = cmd - HINIC_MGMT_MSG_CMD_BASE; @@ -588,15 +553,12 @@ void hinic_hwdev_cb_unregister(struct hinic_hwdev *hwdev, enum hinic_mgmt_msg_cmd cmd) { struct hinic_hwif *hwif = hwdev->hwif; - struct pci_dev *pdev = hwif->pdev; struct hinic_pfhwdev *pfhwdev; struct hinic_nic_cb *nic_cb; u8 cmd_cb; - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "unsupported PCI Function type\n"); + if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) return; - } pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev); @@ -676,10 +638,23 @@ static int init_pfhwdev(struct hinic_pfhwdev *pfhwdev) return err; } - hinic_register_mgmt_msg_cb(&pfhwdev->pf_to_mgmt, HINIC_MOD_L2NIC, - pfhwdev, nic_mgmt_msg_handler); + err = hinic_func_to_func_init(hwdev); + if (err) { + dev_err(&hwif->pdev->dev, "Failed to init mailbox\n"); + hinic_pf_to_mgmt_free(&pfhwdev->pf_to_mgmt); + return err; + } + + if (!HINIC_IS_VF(hwif)) + hinic_register_mgmt_msg_cb(&pfhwdev->pf_to_mgmt, + HINIC_MOD_L2NIC, pfhwdev, + nic_mgmt_msg_handler); + else + hinic_register_vf_mbox_cb(hwdev, HINIC_MOD_L2NIC, + nic_mgmt_msg_handler); hinic_set_pf_action(hwif, HINIC_PF_MGMT_ACTIVE); + return 0; } @@ -693,11 +668,43 @@ static void free_pfhwdev(struct hinic_pfhwdev *pfhwdev) hinic_set_pf_action(hwdev->hwif, HINIC_PF_MGMT_INIT); - hinic_unregister_mgmt_msg_cb(&pfhwdev->pf_to_mgmt, HINIC_MOD_L2NIC); + if (!HINIC_IS_VF(hwdev->hwif)) + hinic_unregister_mgmt_msg_cb(&pfhwdev->pf_to_mgmt, + HINIC_MOD_L2NIC); + else + hinic_unregister_vf_mbox_cb(hwdev, HINIC_MOD_L2NIC); + + hinic_func_to_func_free(hwdev); hinic_pf_to_mgmt_free(&pfhwdev->pf_to_mgmt); } +static int hinic_l2nic_reset(struct hinic_hwdev *hwdev) +{ + struct hinic_cmd_l2nic_reset l2nic_reset = {0}; + u16 out_size = sizeof(l2nic_reset); + struct hinic_pfhwdev *pfhwdev; + int err; + + pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev); + + l2nic_reset.func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif); + /* 0 represents standard l2nic reset flow */ + l2nic_reset.reset_flag = 0; + + err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM, + HINIC_COMM_CMD_L2NIC_RESET, &l2nic_reset, + sizeof(l2nic_reset), &l2nic_reset, + &out_size, HINIC_MGMT_MSG_SYNC); + if (err || !out_size || l2nic_reset.status) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to reset L2NIC resources, err: %d, status: 0x%x, out_size: 0x%x\n", + err, l2nic_reset.status, out_size); + return -EIO; + } + + return 0; +} + /** * hinic_init_hwdev - Initialize the NIC HW * @pdev: the NIC pci device @@ -723,12 +730,6 @@ struct hinic_hwdev *hinic_init_hwdev(struct pci_dev *pdev) return ERR_PTR(err); } - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "Unsupported PCI Function type\n"); - err = -EFAULT; - goto err_func_type; - } - pfhwdev = devm_kzalloc(&pdev->dev, sizeof(*pfhwdev), GFP_KERNEL); if (!pfhwdev) { err = -ENOMEM; @@ -766,12 +767,22 @@ struct hinic_hwdev *hinic_init_hwdev(struct pci_dev *pdev) goto err_init_pfhwdev; } + err = hinic_l2nic_reset(hwdev); + if (err) + goto err_l2nic_reset; + err = get_dev_cap(hwdev); if (err) { dev_err(&pdev->dev, "Failed to get device capabilities\n"); goto err_dev_cap; } + err = hinic_vf_func_init(hwdev); + if (err) { + dev_err(&pdev->dev, "Failed to init nic mbox\n"); + goto err_vf_func_init; + } + err = init_fw_ctxt(hwdev); if (err) { dev_err(&pdev->dev, "Failed to init function table\n"); @@ -788,6 +799,9 @@ struct hinic_hwdev *hinic_init_hwdev(struct pci_dev *pdev) err_resources_state: err_init_fw_ctxt: + hinic_vf_func_free(hwdev); +err_vf_func_init: +err_l2nic_reset: err_dev_cap: free_pfhwdev(pfhwdev); @@ -799,7 +813,6 @@ err_aeqs_init: err_init_msix: err_pfhwdev_alloc: -err_func_type: hinic_free_hwif(hwif); return ERR_PTR(err); } @@ -930,15 +943,9 @@ int hinic_hwdev_hw_ci_addr_set(struct hinic_hwdev *hwdev, struct hinic_sq *sq, { struct hinic_qp *qp = container_of(sq, struct hinic_qp, sq); struct hinic_hwif *hwif = hwdev->hwif; - struct pci_dev *pdev = hwif->pdev; struct hinic_pfhwdev *pfhwdev; struct hinic_cmd_hw_ci hw_ci; - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "Unsupported PCI Function type\n"); - return -EINVAL; - } - hw_ci.dma_attr_off = 0; hw_ci.pending_limit = pending_limit; hw_ci.coalesc_timer = coalesc_timer; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h index 66fd2340d447..71ea7e46dbbc 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_dev.h @@ -16,18 +16,33 @@ #include "hinic_hw_mgmt.h" #include "hinic_hw_qp.h" #include "hinic_hw_io.h" +#include "hinic_hw_mbox.h" #define HINIC_MAX_QPS 32 #define HINIC_MGMT_NUM_MSG_CMD (HINIC_MGMT_MSG_CMD_MAX - \ HINIC_MGMT_MSG_CMD_BASE) +#define HINIC_PF_SET_VF_ALREADY 0x4 +#define HINIC_MGMT_STATUS_EXIST 0x6 +#define HINIC_MGMT_CMD_UNSUPPORTED 0xFF + struct hinic_cap { u16 max_qps; u16 num_qps; + u8 max_vf; + u16 max_vf_qps; +}; + +enum hw_ioctxt_set_cmdq_depth { + HW_IOCTXT_SET_CMDQ_DEPTH_DEFAULT, + HW_IOCTXT_SET_CMDQ_DEPTH_ENABLE, }; enum hinic_port_cmd { + HINIC_PORT_CMD_VF_REGISTER = 0x0, + HINIC_PORT_CMD_VF_UNREGISTER = 0x1, + HINIC_PORT_CMD_CHANGE_MTU = 2, HINIC_PORT_CMD_ADD_VLAN = 3, @@ -39,6 +54,9 @@ enum hinic_port_cmd { HINIC_PORT_CMD_SET_RX_MODE = 12, + HINIC_PORT_CMD_GET_PAUSE_INFO = 20, + HINIC_PORT_CMD_SET_PAUSE_INFO = 21, + HINIC_PORT_CMD_GET_LINK_STATE = 24, HINIC_PORT_CMD_SET_LRO = 25, @@ -77,19 +95,45 @@ enum hinic_port_cmd { HINIC_PORT_CMD_FWCTXT_INIT = 69, + HINIC_PORT_CMD_ENABLE_SPOOFCHK = 78, + HINIC_PORT_CMD_GET_MGMT_VERSION = 88, HINIC_PORT_CMD_SET_FUNC_STATE = 93, HINIC_PORT_CMD_GET_GLOBAL_QPN = 102, + HINIC_PORT_CMD_SET_VF_RATE = 105, + + HINIC_PORT_CMD_SET_VF_VLAN = 106, + + HINIC_PORT_CMD_CLR_VF_VLAN, + HINIC_PORT_CMD_SET_TSO = 112, HINIC_PORT_CMD_SET_RQ_IQ_MAP = 115, + HINIC_PORT_CMD_LINK_STATUS_REPORT = 160, + + HINIC_PORT_CMD_UPDATE_MAC = 164, + HINIC_PORT_CMD_GET_CAP = 170, + HINIC_PORT_CMD_GET_LINK_MODE = 217, + + HINIC_PORT_CMD_SET_SPEED = 218, + + HINIC_PORT_CMD_SET_AUTONEG = 219, + HINIC_PORT_CMD_SET_LRO_TIMER = 244, + + HINIC_PORT_CMD_SET_VF_MAX_MIN_RATE = 249, +}; + +/* cmd of mgmt CPU message for HILINK module */ +enum hinic_hilink_cmd { + HINIC_HILINK_CMD_GET_LINK_INFO = 0x3, + HINIC_HILINK_CMD_SET_LINK_SETTINGS = 0x8, }; enum hinic_ucode_cmd { @@ -191,6 +235,17 @@ struct hinic_cmd_set_res_state { u32 rsvd2; }; +struct hinic_ceq_ctrl_reg { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 q_id; + u32 ctrl0; + u32 ctrl1; +}; + struct hinic_cmd_base_qpn { u8 status; u8 version; @@ -219,12 +274,22 @@ struct hinic_cmd_hw_ci { u64 ci_addr; }; +struct hinic_cmd_l2nic_reset { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 reset_flag; +}; + struct hinic_hwdev { struct hinic_hwif *hwif; struct msix_entry *msix_entries; struct hinic_aeqs aeqs; struct hinic_func_to_io func_to_io; + struct hinic_mbox_func_to_func *func_to_func; struct hinic_cap nic_cap; }; @@ -246,6 +311,25 @@ struct hinic_pfhwdev { struct hinic_nic_cb nic_cb[HINIC_MGMT_NUM_MSG_CMD]; }; +struct hinic_dev_cap { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 rsvd1[5]; + u8 intr_type; + u8 max_cos_id; + u8 er_id; + u8 port_id; + u8 max_vf; + u8 rsvd2[62]; + u16 max_sqs; + u16 max_rqs; + u16 max_vf_sqs; + u16 max_vf_rqs; + u8 rsvd3[204]; +}; + void hinic_hwdev_cb_register(struct hinic_hwdev *hwdev, enum hinic_mgmt_msg_cmd cmd, void *handle, void (*handler)(void *handle, void *buf_in, @@ -259,7 +343,11 @@ int hinic_port_msg_cmd(struct hinic_hwdev *hwdev, enum hinic_port_cmd cmd, void *buf_in, u16 in_size, void *buf_out, u16 *out_size); -int hinic_hwdev_ifup(struct hinic_hwdev *hwdev); +int hinic_hilink_msg_cmd(struct hinic_hwdev *hwdev, enum hinic_hilink_cmd cmd, + void *buf_in, u16 in_size, void *buf_out, + u16 *out_size); + +int hinic_hwdev_ifup(struct hinic_hwdev *hwdev, u16 sq_depth, u16 rq_depth); void hinic_hwdev_ifdown(struct hinic_hwdev *hwdev); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c index c0b6bcb067cd..397936cac304 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.c @@ -17,6 +17,7 @@ #include <asm/byteorder.h> #include <asm/barrier.h> +#include "hinic_hw_dev.h" #include "hinic_hw_csr.h" #include "hinic_hw_if.h" #include "hinic_hw_eqs.h" @@ -416,11 +417,11 @@ static irqreturn_t ceq_interrupt(int irq, void *data) return IRQ_HANDLED; } -static void set_ctrl0(struct hinic_eq *eq) +static u32 get_ctrl0_val(struct hinic_eq *eq, u32 addr) { struct msix_entry *msix_entry = &eq->msix_entry; enum hinic_eq_type type = eq->type; - u32 addr, val, ctrl0; + u32 val, ctrl0; if (type == HINIC_AEQ) { /* RMW Ctrl0 */ @@ -440,9 +441,7 @@ static void set_ctrl0(struct hinic_eq *eq) HINIC_AEQ_CTRL_0_SET(EQ_INT_MODE_ARMED, INT_MODE); val |= ctrl0; - - hinic_hwif_write_reg(eq->hwif, addr, val); - } else if (type == HINIC_CEQ) { + } else { /* RMW Ctrl0 */ addr = HINIC_CSR_CEQ_CTRL_0_ADDR(eq->q_id); @@ -462,16 +461,28 @@ static void set_ctrl0(struct hinic_eq *eq) HINIC_CEQ_CTRL_0_SET(EQ_INT_MODE_ARMED, INTR_MODE); val |= ctrl0; - - hinic_hwif_write_reg(eq->hwif, addr, val); } + return val; } -static void set_ctrl1(struct hinic_eq *eq) +static void set_ctrl0(struct hinic_eq *eq) { + u32 val, addr; + + if (eq->type == HINIC_AEQ) + addr = HINIC_CSR_AEQ_CTRL_0_ADDR(eq->q_id); + else + addr = HINIC_CSR_CEQ_CTRL_0_ADDR(eq->q_id); + + val = get_ctrl0_val(eq, addr); + + hinic_hwif_write_reg(eq->hwif, addr, val); +} + +static u32 get_ctrl1_val(struct hinic_eq *eq, u32 addr) +{ + u32 page_size_val, elem_size, val, ctrl1; enum hinic_eq_type type = eq->type; - u32 page_size_val, elem_size; - u32 addr, val, ctrl1; if (type == HINIC_AEQ) { /* RMW Ctrl1 */ @@ -491,9 +502,7 @@ static void set_ctrl1(struct hinic_eq *eq) HINIC_AEQ_CTRL_1_SET(page_size_val, PAGE_SIZE); val |= ctrl1; - - hinic_hwif_write_reg(eq->hwif, addr, val); - } else if (type == HINIC_CEQ) { + } else { /* RMW Ctrl1 */ addr = HINIC_CSR_CEQ_CTRL_1_ADDR(eq->q_id); @@ -508,19 +517,70 @@ static void set_ctrl1(struct hinic_eq *eq) HINIC_CEQ_CTRL_1_SET(page_size_val, PAGE_SIZE); val |= ctrl1; + } + return val; +} - hinic_hwif_write_reg(eq->hwif, addr, val); +static void set_ctrl1(struct hinic_eq *eq) +{ + u32 addr, val; + + if (eq->type == HINIC_AEQ) + addr = HINIC_CSR_AEQ_CTRL_1_ADDR(eq->q_id); + else + addr = HINIC_CSR_CEQ_CTRL_1_ADDR(eq->q_id); + + val = get_ctrl1_val(eq, addr); + + hinic_hwif_write_reg(eq->hwif, addr, val); +} + +static int set_ceq_ctrl_reg(struct hinic_eq *eq) +{ + struct hinic_ceq_ctrl_reg ceq_ctrl = {0}; + struct hinic_hwdev *hwdev = eq->hwdev; + u16 out_size = sizeof(ceq_ctrl); + u16 in_size = sizeof(ceq_ctrl); + struct hinic_pfhwdev *pfhwdev; + u32 addr; + int err; + + pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev); + + addr = HINIC_CSR_CEQ_CTRL_0_ADDR(eq->q_id); + ceq_ctrl.ctrl0 = get_ctrl0_val(eq, addr); + addr = HINIC_CSR_CEQ_CTRL_1_ADDR(eq->q_id); + ceq_ctrl.ctrl1 = get_ctrl1_val(eq, addr); + + ceq_ctrl.func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif); + ceq_ctrl.q_id = eq->q_id; + + err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM, + HINIC_COMM_CMD_CEQ_CTRL_REG_WR_BY_UP, + &ceq_ctrl, in_size, + &ceq_ctrl, &out_size, HINIC_MGMT_MSG_SYNC); + if (err || !out_size || ceq_ctrl.status) { + dev_err(&hwdev->hwif->pdev->dev, + "Failed to set ceq %d ctrl reg, err: %d status: 0x%x, out_size: 0x%x\n", + eq->q_id, err, ceq_ctrl.status, out_size); + return -EFAULT; } + + return 0; } /** * set_eq_ctrls - setting eq's ctrl registers * @eq: the Event Queue for setting **/ -static void set_eq_ctrls(struct hinic_eq *eq) +static int set_eq_ctrls(struct hinic_eq *eq) { + if (HINIC_IS_VF(eq->hwif) && eq->type == HINIC_CEQ) + return set_ceq_ctrl_reg(eq); + set_ctrl0(eq); set_ctrl1(eq); + return 0; } /** @@ -703,7 +763,12 @@ static int init_eq(struct hinic_eq *eq, struct hinic_hwif *hwif, return -EINVAL; } - set_eq_ctrls(eq); + err = set_eq_ctrls(eq); + if (err) { + dev_err(&pdev->dev, "Failed to set eq ctrls\n"); + return err; + } + eq_update_ci(eq, EQ_ARMED); err = alloc_eq_pages(eq); @@ -859,6 +924,7 @@ int hinic_ceqs_init(struct hinic_ceqs *ceqs, struct hinic_hwif *hwif, ceqs->num_ceqs = num_ceqs; for (q_id = 0; q_id < num_ceqs; q_id++) { + ceqs->ceq[q_id].hwdev = ceqs->hwdev; err = init_eq(&ceqs->ceq[q_id], hwif, HINIC_CEQ, q_id, q_len, page_size, msix_entries[q_id]); if (err) { diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.h index d35f2068ee0c..74b9ff90640c 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_eqs.h @@ -143,8 +143,9 @@ enum hinic_eq_type { }; enum hinic_aeq_type { + HINIC_MBX_FROM_FUNC = 1, HINIC_MSG_FROM_MGMT_CPU = 2, - + HINIC_MBX_SEND_RSLT = 5, HINIC_MAX_AEQ_EVENTS, }; @@ -171,7 +172,7 @@ struct hinic_eq_work { struct hinic_eq { struct hinic_hwif *hwif; - + struct hinic_hwdev *hwdev; enum hinic_eq_type type; int q_id; u32 q_len; @@ -219,7 +220,7 @@ struct hinic_ceq_cb { struct hinic_ceqs { struct hinic_hwif *hwif; - + struct hinic_hwdev *hwdev; struct hinic_eq ceq[HINIC_MAX_CEQS]; int num_ceqs; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_if.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_if.c index 07bbfbf68577..cf127d896ba6 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_if.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_if.c @@ -10,6 +10,7 @@ #include <linux/io.h> #include <linux/types.h> #include <linux/bitops.h> +#include <linux/delay.h> #include "hinic_hw_csr.h" #include "hinic_hw_if.h" @@ -18,6 +19,8 @@ #define VALID_MSIX_IDX(attr, msix_index) ((msix_index) < (attr)->num_irqs) +#define WAIT_HWIF_READY_TIMEOUT 10000 + /** * hinic_msix_attr_set - set message attribute for msix entry * @hwif: the HW interface of a pci function device @@ -115,8 +118,12 @@ int hinic_msix_attr_cnt_clear(struct hinic_hwif *hwif, u16 msix_index) **/ void hinic_set_pf_action(struct hinic_hwif *hwif, enum hinic_pf_action action) { - u32 attr5 = hinic_hwif_read_reg(hwif, HINIC_CSR_FUNC_ATTR5_ADDR); + u32 attr5; + + if (HINIC_IS_VF(hwif)) + return; + attr5 = hinic_hwif_read_reg(hwif, HINIC_CSR_FUNC_ATTR5_ADDR); attr5 = HINIC_FA5_CLEAR(attr5, PF_ACTION); attr5 |= HINIC_FA5_SET(action, PF_ACTION); @@ -183,27 +190,47 @@ void hinic_set_msix_state(struct hinic_hwif *hwif, u16 msix_idx, **/ static int hwif_ready(struct hinic_hwif *hwif) { - struct pci_dev *pdev = hwif->pdev; u32 addr, attr1; addr = HINIC_CSR_FUNC_ATTR1_ADDR; attr1 = hinic_hwif_read_reg(hwif, addr); - if (!HINIC_FA1_GET(attr1, INIT_STATUS)) { - dev_err(&pdev->dev, "hwif status is not ready\n"); - return -EFAULT; + if (!HINIC_FA1_GET(attr1, MGMT_INIT_STATUS)) + return -EBUSY; + + if (HINIC_IS_VF(hwif)) { + if (!HINIC_FA1_GET(attr1, PF_INIT_STATUS)) + return -EBUSY; } return 0; } +static int wait_hwif_ready(struct hinic_hwif *hwif) +{ + unsigned long timeout = 0; + + do { + if (!hwif_ready(hwif)) + return 0; + + usleep_range(999, 1000); + timeout++; + } while (timeout <= WAIT_HWIF_READY_TIMEOUT); + + dev_err(&hwif->pdev->dev, "Wait for hwif timeout\n"); + + return -EBUSY; +} + /** * set_hwif_attr - set the attributes in the relevant members in hwif * @hwif: the HW interface of a pci function device * @attr0: the first attribute that was read from the hw * @attr1: the second attribute that was read from the hw **/ -static void set_hwif_attr(struct hinic_hwif *hwif, u32 attr0, u32 attr1) +static void set_hwif_attr(struct hinic_hwif *hwif, u32 attr0, u32 attr1, + u32 attr2) { hwif->attr.func_idx = HINIC_FA0_GET(attr0, FUNC_IDX); hwif->attr.pf_idx = HINIC_FA0_GET(attr0, PF_IDX); @@ -214,6 +241,8 @@ static void set_hwif_attr(struct hinic_hwif *hwif, u32 attr0, u32 attr1) hwif->attr.num_ceqs = BIT(HINIC_FA1_GET(attr1, CEQS_PER_FUNC)); hwif->attr.num_irqs = BIT(HINIC_FA1_GET(attr1, IRQS_PER_FUNC)); hwif->attr.num_dma_attr = BIT(HINIC_FA1_GET(attr1, DMA_ATTR_PER_FUNC)); + hwif->attr.global_vf_id_of_pf = HINIC_FA2_GET(attr2, + GLOBAL_VF_ID_OF_PF); } /** @@ -222,7 +251,7 @@ static void set_hwif_attr(struct hinic_hwif *hwif, u32 attr0, u32 attr1) **/ static void read_hwif_attr(struct hinic_hwif *hwif) { - u32 addr, attr0, attr1; + u32 addr, attr0, attr1, attr2; addr = HINIC_CSR_FUNC_ATTR0_ADDR; attr0 = hinic_hwif_read_reg(hwif, addr); @@ -230,7 +259,10 @@ static void read_hwif_attr(struct hinic_hwif *hwif) addr = HINIC_CSR_FUNC_ATTR1_ADDR; attr1 = hinic_hwif_read_reg(hwif, addr); - set_hwif_attr(hwif, attr0, attr1); + addr = HINIC_CSR_FUNC_ATTR2_ADDR; + attr2 = hinic_hwif_read_reg(hwif, addr); + + set_hwif_attr(hwif, attr0, attr1, attr2); } /** @@ -309,6 +341,34 @@ static void dma_attr_init(struct hinic_hwif *hwif) HINIC_PCIE_SNOOP, HINIC_PCIE_TPH_DISABLE); } +u16 hinic_glb_pf_vf_offset(struct hinic_hwif *hwif) +{ + if (!hwif) + return 0; + + return hwif->attr.global_vf_id_of_pf; +} + +u16 hinic_global_func_id_hw(struct hinic_hwif *hwif) +{ + u32 addr, attr0; + + addr = HINIC_CSR_FUNC_ATTR0_ADDR; + attr0 = hinic_hwif_read_reg(hwif, addr); + + return HINIC_FA0_GET(attr0, FUNC_IDX); +} + +u16 hinic_pf_id_of_vf_hw(struct hinic_hwif *hwif) +{ + u32 addr, attr0; + + addr = HINIC_CSR_FUNC_ATTR0_ADDR; + attr0 = hinic_hwif_read_reg(hwif, addr); + + return HINIC_FA0_GET(attr0, PF_IDX); +} + /** * hinic_init_hwif - initialize the hw interface * @hwif: the HW interface of a pci function device @@ -335,7 +395,7 @@ int hinic_init_hwif(struct hinic_hwif *hwif, struct pci_dev *pdev) goto err_map_intr_bar; } - err = hwif_ready(hwif); + err = wait_hwif_ready(hwif); if (err) { dev_err(&pdev->dev, "HW interface is not ready\n"); goto err_hwif_ready; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_if.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_if.h index c7bb9ceca72c..0872e035faa1 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_if.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_if.h @@ -35,6 +35,7 @@ #define HINIC_FA0_FUNC_IDX_SHIFT 0 #define HINIC_FA0_PF_IDX_SHIFT 10 #define HINIC_FA0_PCI_INTF_IDX_SHIFT 14 +#define HINIC_FA0_VF_IN_PF_SHIFT 16 /* reserved members - off 16 */ #define HINIC_FA0_FUNC_TYPE_SHIFT 24 @@ -42,6 +43,7 @@ #define HINIC_FA0_PF_IDX_MASK 0xF #define HINIC_FA0_PCI_INTF_IDX_MASK 0x3 #define HINIC_FA0_FUNC_TYPE_MASK 0x1 +#define HINIC_FA0_VF_IN_PF_MASK 0xFF #define HINIC_FA0_GET(val, member) \ (((val) >> HINIC_FA0_##member##_SHIFT) & HINIC_FA0_##member##_MASK) @@ -53,17 +55,25 @@ #define HINIC_FA1_IRQS_PER_FUNC_SHIFT 20 #define HINIC_FA1_DMA_ATTR_PER_FUNC_SHIFT 24 /* reserved members - off 27 */ -#define HINIC_FA1_INIT_STATUS_SHIFT 30 +#define HINIC_FA1_MGMT_INIT_STATUS_SHIFT 30 +#define HINIC_FA1_PF_INIT_STATUS_SHIFT 31 #define HINIC_FA1_AEQS_PER_FUNC_MASK 0x3 #define HINIC_FA1_CEQS_PER_FUNC_MASK 0x7 #define HINIC_FA1_IRQS_PER_FUNC_MASK 0xF #define HINIC_FA1_DMA_ATTR_PER_FUNC_MASK 0x7 -#define HINIC_FA1_INIT_STATUS_MASK 0x1 +#define HINIC_FA1_MGMT_INIT_STATUS_MASK 0x1 +#define HINIC_FA1_PF_INIT_STATUS_MASK 0x1 #define HINIC_FA1_GET(val, member) \ (((val) >> HINIC_FA1_##member##_SHIFT) & HINIC_FA1_##member##_MASK) +#define HINIC_FA2_GLOBAL_VF_ID_OF_PF_SHIFT 16 +#define HINIC_FA2_GLOBAL_VF_ID_OF_PF_MASK 0x3FF + +#define HINIC_FA2_GET(val, member) \ + (((val) >> HINIC_FA2_##member##_SHIFT) & HINIC_FA2_##member##_MASK) + #define HINIC_FA4_OUTBOUND_STATE_SHIFT 0 #define HINIC_FA4_DB_STATE_SHIFT 1 @@ -140,6 +150,7 @@ #define HINIC_HWIF_PPF_IDX(hwif) ((hwif)->attr.ppf_idx) #define HINIC_FUNC_TYPE(hwif) ((hwif)->attr.func_type) +#define HINIC_IS_VF(hwif) (HINIC_FUNC_TYPE(hwif) == HINIC_VF) #define HINIC_IS_PF(hwif) (HINIC_FUNC_TYPE(hwif) == HINIC_PF) #define HINIC_IS_PPF(hwif) (HINIC_FUNC_TYPE(hwif) == HINIC_PPF) @@ -173,6 +184,7 @@ enum hinic_pcie_tph { enum hinic_func_type { HINIC_PF = 0, + HINIC_VF = 1, HINIC_PPF = 2, }; @@ -180,7 +192,7 @@ enum hinic_mod_type { HINIC_MOD_COMM = 0, /* HW communication module */ HINIC_MOD_L2NIC = 1, /* L2NIC module */ HINIC_MOD_CFGM = 7, /* Configuration module */ - + HINIC_MOD_HILINK = 14, /* Hilink module */ HINIC_MOD_MAX = 15 }; @@ -223,6 +235,8 @@ struct hinic_func_attr { u8 num_ceqs; u8 num_dma_attr; + + u16 global_vf_id_of_pf; }; struct hinic_hwif { @@ -271,6 +285,12 @@ enum hinic_db_state hinic_db_state_get(struct hinic_hwif *hwif); void hinic_db_state_set(struct hinic_hwif *hwif, enum hinic_db_state db_state); +u16 hinic_glb_pf_vf_offset(struct hinic_hwif *hwif); + +u16 hinic_global_func_id_hw(struct hinic_hwif *hwif); + +u16 hinic_pf_id_of_vf_hw(struct hinic_hwif *hwif); + int hinic_init_hwif(struct hinic_hwif *hwif, struct pci_dev *pdev); void hinic_free_hwif(struct hinic_hwif *hwif); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_io.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_io.c index d66f86fa3f46..3e3fa742e476 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_io.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_io.c @@ -15,6 +15,7 @@ #include <linux/io.h> #include <linux/err.h> +#include "hinic_hw_dev.h" #include "hinic_hw_if.h" #include "hinic_hw_eqs.h" #include "hinic_hw_wqe.h" @@ -34,6 +35,8 @@ #define DB_IDX(db, db_base) \ (((unsigned long)(db) - (unsigned long)(db_base)) / HINIC_DB_PAGE_SIZE) +#define HINIC_PAGE_SIZE_HW(pg_size) ((u8)ilog2((u32)((pg_size) >> 12))) + enum io_cmd { IO_CMD_MODIFY_QUEUE_CTXT = 0, IO_CMD_CLEAN_QUEUE_CTXT, @@ -279,7 +282,7 @@ static int init_qp(struct hinic_func_to_io *func_to_io, err = hinic_wq_allocate(&func_to_io->wqs, &func_to_io->sq_wq[q_id], HINIC_SQ_WQEBB_SIZE, HINIC_SQ_PAGE_SIZE, - HINIC_SQ_DEPTH, HINIC_SQ_WQE_MAX_SIZE); + func_to_io->sq_depth, HINIC_SQ_WQE_MAX_SIZE); if (err) { dev_err(&pdev->dev, "Failed to allocate WQ for SQ\n"); return err; @@ -287,7 +290,7 @@ static int init_qp(struct hinic_func_to_io *func_to_io, err = hinic_wq_allocate(&func_to_io->wqs, &func_to_io->rq_wq[q_id], HINIC_RQ_WQEBB_SIZE, HINIC_RQ_PAGE_SIZE, - HINIC_RQ_DEPTH, HINIC_RQ_WQE_SIZE); + func_to_io->rq_depth, HINIC_RQ_WQE_SIZE); if (err) { dev_err(&pdev->dev, "Failed to allocate WQ for RQ\n"); goto err_rq_alloc; @@ -484,6 +487,33 @@ void hinic_io_destroy_qps(struct hinic_func_to_io *func_to_io, int num_qps) devm_kfree(&pdev->dev, func_to_io->qps); } +int hinic_set_wq_page_size(struct hinic_hwdev *hwdev, u16 func_idx, + u32 page_size) +{ + struct hinic_wq_page_size page_size_info = {0}; + u16 out_size = sizeof(page_size_info); + struct hinic_pfhwdev *pfhwdev; + int err; + + pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev); + + page_size_info.func_idx = func_idx; + page_size_info.ppf_idx = HINIC_HWIF_PPF_IDX(hwdev->hwif); + page_size_info.page_size = HINIC_PAGE_SIZE_HW(page_size); + + err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM, + HINIC_COMM_CMD_PAGESIZE_SET, &page_size_info, + sizeof(page_size_info), &page_size_info, + &out_size, HINIC_MGMT_MSG_SYNC); + if (err || !out_size || page_size_info.status) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to set wq page size, err: %d, status: 0x%x, out_size: 0x%0x\n", + err, page_size_info.status, out_size); + return -EFAULT; + } + + return 0; +} + /** * hinic_io_init - Initialize the IO components * @func_to_io: func to io channel that holds the IO components @@ -506,6 +536,7 @@ int hinic_io_init(struct hinic_func_to_io *func_to_io, func_to_io->hwif = hwif; func_to_io->qps = NULL; func_to_io->max_qps = max_qps; + func_to_io->ceqs.hwdev = func_to_io->hwdev; err = hinic_ceqs_init(&func_to_io->ceqs, hwif, num_ceqs, HINIC_DEFAULT_CEQ_LEN, HINIC_EQ_PAGE_SIZE, @@ -541,6 +572,14 @@ int hinic_io_init(struct hinic_func_to_io *func_to_io, func_to_io->cmdq_db_area[cmdq] = db_area; } + err = hinic_set_wq_page_size(func_to_io->hwdev, + HINIC_HWIF_FUNC_IDX(hwif), + HINIC_DEFAULT_WQ_PAGE_SIZE); + if (err) { + dev_err(&func_to_io->hwif->pdev->dev, "Failed to set wq page size\n"); + goto init_wq_pg_size_err; + } + err = hinic_init_cmdqs(&func_to_io->cmdqs, hwif, func_to_io->cmdq_db_area); if (err) { @@ -551,6 +590,11 @@ int hinic_io_init(struct hinic_func_to_io *func_to_io, return 0; err_init_cmdqs: + if (!HINIC_IS_VF(func_to_io->hwif)) + hinic_set_wq_page_size(func_to_io->hwdev, + HINIC_HWIF_FUNC_IDX(hwif), + HINIC_HW_WQ_PAGE_SIZE); +init_wq_pg_size_err: err_db_area: for (type = HINIC_CMDQ_SYNC; type < cmdq; type++) return_db_area(func_to_io, func_to_io->cmdq_db_area[type]); @@ -575,6 +619,11 @@ void hinic_io_free(struct hinic_func_to_io *func_to_io) hinic_free_cmdqs(&func_to_io->cmdqs); + if (!HINIC_IS_VF(func_to_io->hwif)) + hinic_set_wq_page_size(func_to_io->hwdev, + HINIC_HWIF_FUNC_IDX(func_to_io->hwif), + HINIC_HW_WQ_PAGE_SIZE); + for (cmdq = HINIC_CMDQ_SYNC; cmdq < HINIC_MAX_CMDQ_TYPES; cmdq++) return_db_area(func_to_io, func_to_io->cmdq_db_area[cmdq]); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_io.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_io.h index cac2b722e7dc..214f162f7579 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_io.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_io.h @@ -20,6 +20,8 @@ #define HINIC_DB_PAGE_SIZE SZ_4K #define HINIC_DB_SIZE SZ_4M +#define HINIC_HW_WQ_PAGE_SIZE SZ_4K +#define HINIC_DEFAULT_WQ_PAGE_SIZE SZ_256K #define HINIC_DB_MAX_AREAS (HINIC_DB_SIZE / HINIC_DB_PAGE_SIZE) @@ -47,7 +49,7 @@ struct hinic_free_db_area { struct hinic_func_to_io { struct hinic_hwif *hwif; - + struct hinic_hwdev *hwdev; struct hinic_ceqs ceqs; struct hinic_wqs wqs; @@ -58,6 +60,9 @@ struct hinic_func_to_io { struct hinic_qp *qps; u16 max_qps; + u16 sq_depth; + u16 rq_depth; + void __iomem **sq_db; void __iomem *db_base; @@ -69,8 +74,27 @@ struct hinic_func_to_io { void __iomem *cmdq_db_area[HINIC_MAX_CMDQ_TYPES]; struct hinic_cmdqs cmdqs; + + u16 max_vfs; + struct vf_data_storage *vf_infos; + u8 link_status; +}; + +struct hinic_wq_page_size { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_idx; + u8 ppf_idx; + u8 page_size; + + u32 rsvd1; }; +int hinic_set_wq_page_size(struct hinic_hwdev *hwdev, u16 func_idx, + u32 page_size); + int hinic_io_create_qps(struct hinic_func_to_io *func_to_io, u16 base_qpn, int num_qps, struct msix_entry *sq_msix_entries, diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_mbox.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_mbox.c new file mode 100644 index 000000000000..bc2f87e6cb5d --- /dev/null +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_mbox.c @@ -0,0 +1,1210 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Huawei HiNIC PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + */ +#include <linux/pci.h> +#include <linux/delay.h> +#include <linux/types.h> +#include <linux/completion.h> +#include <linux/semaphore.h> +#include <linux/spinlock.h> +#include <linux/workqueue.h> + +#include "hinic_hw_if.h" +#include "hinic_hw_mgmt.h" +#include "hinic_hw_csr.h" +#include "hinic_hw_dev.h" +#include "hinic_hw_mbox.h" + +#define HINIC_MBOX_INT_DST_FUNC_SHIFT 0 +#define HINIC_MBOX_INT_DST_AEQN_SHIFT 10 +#define HINIC_MBOX_INT_SRC_RESP_AEQN_SHIFT 12 +#define HINIC_MBOX_INT_STAT_DMA_SHIFT 14 +/* The size of data to be sended (unit of 4 bytes) */ +#define HINIC_MBOX_INT_TX_SIZE_SHIFT 20 +/* SO_RO(strong order, relax order) */ +#define HINIC_MBOX_INT_STAT_DMA_SO_RO_SHIFT 25 +#define HINIC_MBOX_INT_WB_EN_SHIFT 28 + +#define HINIC_MBOX_INT_DST_FUNC_MASK 0x3FF +#define HINIC_MBOX_INT_DST_AEQN_MASK 0x3 +#define HINIC_MBOX_INT_SRC_RESP_AEQN_MASK 0x3 +#define HINIC_MBOX_INT_STAT_DMA_MASK 0x3F +#define HINIC_MBOX_INT_TX_SIZE_MASK 0x1F +#define HINIC_MBOX_INT_STAT_DMA_SO_RO_MASK 0x3 +#define HINIC_MBOX_INT_WB_EN_MASK 0x1 + +#define HINIC_MBOX_INT_SET(val, field) \ + (((val) & HINIC_MBOX_INT_##field##_MASK) << \ + HINIC_MBOX_INT_##field##_SHIFT) + +enum hinic_mbox_tx_status { + TX_NOT_DONE = 1, +}; + +#define HINIC_MBOX_CTRL_TRIGGER_AEQE_SHIFT 0 + +/* specifies the issue request for the message data. + * 0 - Tx request is done; + * 1 - Tx request is in process. + */ +#define HINIC_MBOX_CTRL_TX_STATUS_SHIFT 1 + +#define HINIC_MBOX_CTRL_TRIGGER_AEQE_MASK 0x1 +#define HINIC_MBOX_CTRL_TX_STATUS_MASK 0x1 + +#define HINIC_MBOX_CTRL_SET(val, field) \ + (((val) & HINIC_MBOX_CTRL_##field##_MASK) << \ + HINIC_MBOX_CTRL_##field##_SHIFT) + +#define HINIC_MBOX_HEADER_MSG_LEN_SHIFT 0 +#define HINIC_MBOX_HEADER_MODULE_SHIFT 11 +#define HINIC_MBOX_HEADER_SEG_LEN_SHIFT 16 +#define HINIC_MBOX_HEADER_NO_ACK_SHIFT 22 +#define HINIC_MBOX_HEADER_SEQID_SHIFT 24 +#define HINIC_MBOX_HEADER_LAST_SHIFT 30 + +/* specifies the mailbox message direction + * 0 - send + * 1 - receive + */ +#define HINIC_MBOX_HEADER_DIRECTION_SHIFT 31 +#define HINIC_MBOX_HEADER_CMD_SHIFT 32 +#define HINIC_MBOX_HEADER_MSG_ID_SHIFT 40 +#define HINIC_MBOX_HEADER_STATUS_SHIFT 48 +#define HINIC_MBOX_HEADER_SRC_GLB_FUNC_IDX_SHIFT 54 + +#define HINIC_MBOX_HEADER_MSG_LEN_MASK 0x7FF +#define HINIC_MBOX_HEADER_MODULE_MASK 0x1F +#define HINIC_MBOX_HEADER_SEG_LEN_MASK 0x3F +#define HINIC_MBOX_HEADER_NO_ACK_MASK 0x1 +#define HINIC_MBOX_HEADER_SEQID_MASK 0x3F +#define HINIC_MBOX_HEADER_LAST_MASK 0x1 +#define HINIC_MBOX_HEADER_DIRECTION_MASK 0x1 +#define HINIC_MBOX_HEADER_CMD_MASK 0xFF +#define HINIC_MBOX_HEADER_MSG_ID_MASK 0xFF +#define HINIC_MBOX_HEADER_STATUS_MASK 0x3F +#define HINIC_MBOX_HEADER_SRC_GLB_FUNC_IDX_MASK 0x3FF + +#define HINIC_MBOX_HEADER_GET(val, field) \ + (((val) >> HINIC_MBOX_HEADER_##field##_SHIFT) & \ + HINIC_MBOX_HEADER_##field##_MASK) +#define HINIC_MBOX_HEADER_SET(val, field) \ + ((u64)((val) & HINIC_MBOX_HEADER_##field##_MASK) << \ + HINIC_MBOX_HEADER_##field##_SHIFT) + +#define MBOX_SEGLEN_MASK \ + HINIC_MBOX_HEADER_SET(HINIC_MBOX_HEADER_SEG_LEN_MASK, SEG_LEN) + +#define HINIC_MBOX_SEG_LEN 48 +#define HINIC_MBOX_COMP_TIME 8000U +#define MBOX_MSG_POLLING_TIMEOUT 8000 + +#define HINIC_MBOX_DATA_SIZE 2040 + +#define MBOX_MAX_BUF_SZ 2048UL +#define MBOX_HEADER_SZ 8 + +#define MBOX_INFO_SZ 4 + +/* MBOX size is 64B, 8B for mbox_header, 4B reserved */ +#define MBOX_SEG_LEN 48 +#define MBOX_SEG_LEN_ALIGN 4 +#define MBOX_WB_STATUS_LEN 16UL + +/* mbox write back status is 16B, only first 4B is used */ +#define MBOX_WB_STATUS_ERRCODE_MASK 0xFFFF +#define MBOX_WB_STATUS_MASK 0xFF +#define MBOX_WB_ERROR_CODE_MASK 0xFF00 +#define MBOX_WB_STATUS_FINISHED_SUCCESS 0xFF +#define MBOX_WB_STATUS_FINISHED_WITH_ERR 0xFE +#define MBOX_WB_STATUS_NOT_FINISHED 0x00 + +#define MBOX_STATUS_FINISHED(wb) \ + (((wb) & MBOX_WB_STATUS_MASK) != MBOX_WB_STATUS_NOT_FINISHED) +#define MBOX_STATUS_SUCCESS(wb) \ + (((wb) & MBOX_WB_STATUS_MASK) == MBOX_WB_STATUS_FINISHED_SUCCESS) +#define MBOX_STATUS_ERRCODE(wb) \ + ((wb) & MBOX_WB_ERROR_CODE_MASK) + +#define SEQ_ID_START_VAL 0 +#define SEQ_ID_MAX_VAL 42 + +#define DST_AEQ_IDX_DEFAULT_VAL 0 +#define SRC_AEQ_IDX_DEFAULT_VAL 0 +#define NO_DMA_ATTRIBUTE_VAL 0 + +#define HINIC_MGMT_RSP_AEQN 0 +#define HINIC_MBOX_RSP_AEQN 2 +#define HINIC_MBOX_RECV_AEQN 0 + +#define MBOX_MSG_NO_DATA_LEN 1 + +#define MBOX_BODY_FROM_HDR(header) ((u8 *)(header) + MBOX_HEADER_SZ) +#define MBOX_AREA(hwif) \ + ((hwif)->cfg_regs_bar + HINIC_FUNC_CSR_MAILBOX_DATA_OFF) + +#define IS_PF_OR_PPF_SRC(src_func_idx) ((src_func_idx) < HINIC_MAX_PF_FUNCS) + +#define MBOX_RESPONSE_ERROR 0x1 +#define MBOX_MSG_ID_MASK 0xFF +#define MBOX_MSG_ID(func_to_func) ((func_to_func)->send_msg_id) +#define MBOX_MSG_ID_INC(func_to_func_mbox) (MBOX_MSG_ID(func_to_func_mbox) = \ + (MBOX_MSG_ID(func_to_func_mbox) + 1) & MBOX_MSG_ID_MASK) + +#define FUNC_ID_OFF_SET_8B 8 +#define FUNC_ID_OFF_SET_10B 10 + +/* max message counter wait to process for one function */ +#define HINIC_MAX_MSG_CNT_TO_PROCESS 10 + +#define HINIC_QUEUE_MIN_DEPTH 6 +#define HINIC_QUEUE_MAX_DEPTH 12 +#define HINIC_MAX_RX_BUFFER_SIZE 15 + +enum hinic_hwif_direction_type { + HINIC_HWIF_DIRECT_SEND = 0, + HINIC_HWIF_RESPONSE = 1, +}; + +enum mbox_send_mod { + MBOX_SEND_MSG_INT, +}; + +enum mbox_seg_type { + NOT_LAST_SEG, + LAST_SEG, +}; + +enum mbox_ordering_type { + STRONG_ORDER, +}; + +enum mbox_write_back_type { + WRITE_BACK = 1, +}; + +enum mbox_aeq_trig_type { + NOT_TRIGGER, + TRIGGER, +}; + +/** + * hinic_register_pf_mbox_cb - register mbox callback for pf + * @hwdev: the pointer to hw device + * @mod: specific mod that the callback will handle + * @callback: callback function + * Return: 0 - success, negative - failure + */ +int hinic_register_pf_mbox_cb(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod, + hinic_pf_mbox_cb callback) +{ + struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func; + + if (mod >= HINIC_MOD_MAX) + return -EFAULT; + + func_to_func->pf_mbox_cb[mod] = callback; + + set_bit(HINIC_PF_MBOX_CB_REG, &func_to_func->pf_mbox_cb_state[mod]); + + return 0; +} + +/** + * hinic_register_vf_mbox_cb - register mbox callback for vf + * @hwdev: the pointer to hw device + * @mod: specific mod that the callback will handle + * @callback: callback function + * Return: 0 - success, negative - failure + */ +int hinic_register_vf_mbox_cb(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod, + hinic_vf_mbox_cb callback) +{ + struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func; + + if (mod >= HINIC_MOD_MAX) + return -EFAULT; + + func_to_func->vf_mbox_cb[mod] = callback; + + set_bit(HINIC_VF_MBOX_CB_REG, &func_to_func->vf_mbox_cb_state[mod]); + + return 0; +} + +/** + * hinic_unregister_pf_mbox_cb - unregister the mbox callback for pf + * @hwdev: the pointer to hw device + * @mod: specific mod that the callback will handle + */ +void hinic_unregister_pf_mbox_cb(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod) +{ + struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func; + + clear_bit(HINIC_PF_MBOX_CB_REG, &func_to_func->pf_mbox_cb_state[mod]); + + while (test_bit(HINIC_PF_MBOX_CB_RUNNING, + &func_to_func->pf_mbox_cb_state[mod])) + usleep_range(900, 1000); + + func_to_func->pf_mbox_cb[mod] = NULL; +} + +/** + * hinic_unregister_vf_mbox_cb - unregister the mbox callback for vf + * @hwdev: the pointer to hw device + * @mod: specific mod that the callback will handle + */ +void hinic_unregister_vf_mbox_cb(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod) +{ + struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func; + + clear_bit(HINIC_VF_MBOX_CB_REG, &func_to_func->vf_mbox_cb_state[mod]); + + while (test_bit(HINIC_VF_MBOX_CB_RUNNING, + &func_to_func->vf_mbox_cb_state[mod])) + usleep_range(900, 1000); + + func_to_func->vf_mbox_cb[mod] = NULL; +} + +static int recv_vf_mbox_handler(struct hinic_mbox_func_to_func *func_to_func, + struct hinic_recv_mbox *recv_mbox, + void *buf_out, u16 *out_size) +{ + hinic_vf_mbox_cb cb; + int ret = 0; + + if (recv_mbox->mod >= HINIC_MOD_MAX) { + dev_err(&func_to_func->hwif->pdev->dev, "Receive illegal mbox message, mod = %d\n", + recv_mbox->mod); + return -EINVAL; + } + + set_bit(HINIC_VF_MBOX_CB_RUNNING, + &func_to_func->vf_mbox_cb_state[recv_mbox->mod]); + + cb = func_to_func->vf_mbox_cb[recv_mbox->mod]; + if (cb && test_bit(HINIC_VF_MBOX_CB_REG, + &func_to_func->vf_mbox_cb_state[recv_mbox->mod])) { + cb(func_to_func->hwdev, recv_mbox->cmd, recv_mbox->mbox, + recv_mbox->mbox_len, buf_out, out_size); + } else { + dev_err(&func_to_func->hwif->pdev->dev, "VF mbox cb is not registered\n"); + ret = -EINVAL; + } + + clear_bit(HINIC_VF_MBOX_CB_RUNNING, + &func_to_func->vf_mbox_cb_state[recv_mbox->mod]); + + return ret; +} + +static int +recv_pf_from_vf_mbox_handler(struct hinic_mbox_func_to_func *func_to_func, + struct hinic_recv_mbox *recv_mbox, + u16 src_func_idx, void *buf_out, + u16 *out_size) +{ + hinic_pf_mbox_cb cb; + u16 vf_id = 0; + int ret; + + if (recv_mbox->mod >= HINIC_MOD_MAX) { + dev_err(&func_to_func->hwif->pdev->dev, "Receive illegal mbox message, mod = %d\n", + recv_mbox->mod); + return -EINVAL; + } + + set_bit(HINIC_PF_MBOX_CB_RUNNING, + &func_to_func->pf_mbox_cb_state[recv_mbox->mod]); + + cb = func_to_func->pf_mbox_cb[recv_mbox->mod]; + if (cb && test_bit(HINIC_PF_MBOX_CB_REG, + &func_to_func->pf_mbox_cb_state[recv_mbox->mod])) { + vf_id = src_func_idx - + hinic_glb_pf_vf_offset(func_to_func->hwif); + ret = cb(func_to_func->hwdev, vf_id, recv_mbox->cmd, + recv_mbox->mbox, recv_mbox->mbox_len, + buf_out, out_size); + } else { + dev_err(&func_to_func->hwif->pdev->dev, "PF mbox mod(0x%x) cb is not registered\n", + recv_mbox->mod); + ret = -EINVAL; + } + + clear_bit(HINIC_PF_MBOX_CB_RUNNING, + &func_to_func->pf_mbox_cb_state[recv_mbox->mod]); + + return ret; +} + +static bool check_mbox_seq_id_and_seg_len(struct hinic_recv_mbox *recv_mbox, + u8 seq_id, u8 seg_len) +{ + if (seq_id > SEQ_ID_MAX_VAL || seg_len > MBOX_SEG_LEN) + return false; + + if (seq_id == 0) { + recv_mbox->seq_id = seq_id; + } else { + if (seq_id != recv_mbox->seq_id + 1) + return false; + + recv_mbox->seq_id = seq_id; + } + + return true; +} + +static void resp_mbox_handler(struct hinic_mbox_func_to_func *func_to_func, + struct hinic_recv_mbox *recv_mbox) +{ + spin_lock(&func_to_func->mbox_lock); + if (recv_mbox->msg_info.msg_id == func_to_func->send_msg_id && + func_to_func->event_flag == EVENT_START) + complete(&recv_mbox->recv_done); + else + dev_err(&func_to_func->hwif->pdev->dev, + "Mbox response timeout, current send msg id(0x%x), recv msg id(0x%x), status(0x%x)\n", + func_to_func->send_msg_id, recv_mbox->msg_info.msg_id, + recv_mbox->msg_info.status); + spin_unlock(&func_to_func->mbox_lock); +} + +static void recv_func_mbox_handler(struct hinic_mbox_func_to_func *func_to_func, + struct hinic_recv_mbox *recv_mbox, + u16 src_func_idx); + +static void recv_func_mbox_work_handler(struct work_struct *work) +{ + struct hinic_mbox_work *mbox_work = + container_of(work, struct hinic_mbox_work, work); + struct hinic_recv_mbox *recv_mbox; + + recv_func_mbox_handler(mbox_work->func_to_func, mbox_work->recv_mbox, + mbox_work->src_func_idx); + + recv_mbox = + &mbox_work->func_to_func->mbox_send[mbox_work->src_func_idx]; + + atomic_dec(&recv_mbox->msg_cnt); + + kfree(mbox_work); +} + +static void recv_mbox_handler(struct hinic_mbox_func_to_func *func_to_func, + void *header, struct hinic_recv_mbox *recv_mbox) +{ + void *mbox_body = MBOX_BODY_FROM_HDR(header); + struct hinic_recv_mbox *rcv_mbox_temp = NULL; + u64 mbox_header = *((u64 *)header); + struct hinic_mbox_work *mbox_work; + u8 seq_id, seg_len; + u16 src_func_idx; + int pos; + + seq_id = HINIC_MBOX_HEADER_GET(mbox_header, SEQID); + seg_len = HINIC_MBOX_HEADER_GET(mbox_header, SEG_LEN); + src_func_idx = HINIC_MBOX_HEADER_GET(mbox_header, SRC_GLB_FUNC_IDX); + + if (!check_mbox_seq_id_and_seg_len(recv_mbox, seq_id, seg_len)) { + dev_err(&func_to_func->hwif->pdev->dev, + "Mailbox sequence and segment check fail, src func id: 0x%x, front id: 0x%x, current id: 0x%x, seg len: 0x%x\n", + src_func_idx, recv_mbox->seq_id, seq_id, seg_len); + recv_mbox->seq_id = SEQ_ID_MAX_VAL; + return; + } + + pos = seq_id * MBOX_SEG_LEN; + memcpy((u8 *)recv_mbox->mbox + pos, mbox_body, + HINIC_MBOX_HEADER_GET(mbox_header, SEG_LEN)); + + if (!HINIC_MBOX_HEADER_GET(mbox_header, LAST)) + return; + + recv_mbox->cmd = HINIC_MBOX_HEADER_GET(mbox_header, CMD); + recv_mbox->mod = HINIC_MBOX_HEADER_GET(mbox_header, MODULE); + recv_mbox->mbox_len = HINIC_MBOX_HEADER_GET(mbox_header, MSG_LEN); + recv_mbox->ack_type = HINIC_MBOX_HEADER_GET(mbox_header, NO_ACK); + recv_mbox->msg_info.msg_id = HINIC_MBOX_HEADER_GET(mbox_header, MSG_ID); + recv_mbox->msg_info.status = HINIC_MBOX_HEADER_GET(mbox_header, STATUS); + recv_mbox->seq_id = SEQ_ID_MAX_VAL; + + if (HINIC_MBOX_HEADER_GET(mbox_header, DIRECTION) == + HINIC_HWIF_RESPONSE) { + resp_mbox_handler(func_to_func, recv_mbox); + return; + } + + if (atomic_read(&recv_mbox->msg_cnt) > HINIC_MAX_MSG_CNT_TO_PROCESS) { + dev_warn(&func_to_func->hwif->pdev->dev, + "This function(%u) have %d message wait to process,can't add to work queue\n", + src_func_idx, atomic_read(&recv_mbox->msg_cnt)); + return; + } + + rcv_mbox_temp = kmemdup(recv_mbox, sizeof(*rcv_mbox_temp), GFP_KERNEL); + if (!rcv_mbox_temp) + return; + + rcv_mbox_temp->mbox = kmemdup(recv_mbox->mbox, MBOX_MAX_BUF_SZ, + GFP_KERNEL); + if (!rcv_mbox_temp->mbox) + goto err_alloc_rcv_mbox_msg; + + rcv_mbox_temp->buf_out = kzalloc(MBOX_MAX_BUF_SZ, GFP_KERNEL); + if (!rcv_mbox_temp->buf_out) + goto err_alloc_rcv_mbox_buf; + + mbox_work = kzalloc(sizeof(*mbox_work), GFP_KERNEL); + if (!mbox_work) + goto err_alloc_mbox_work; + + mbox_work->func_to_func = func_to_func; + mbox_work->recv_mbox = rcv_mbox_temp; + mbox_work->src_func_idx = src_func_idx; + + atomic_inc(&recv_mbox->msg_cnt); + INIT_WORK(&mbox_work->work, recv_func_mbox_work_handler); + queue_work(func_to_func->workq, &mbox_work->work); + + return; + +err_alloc_mbox_work: + kfree(rcv_mbox_temp->buf_out); + +err_alloc_rcv_mbox_buf: + kfree(rcv_mbox_temp->mbox); + +err_alloc_rcv_mbox_msg: + kfree(rcv_mbox_temp); +} + +void hinic_mbox_func_aeqe_handler(void *handle, void *header, u8 size) +{ + struct hinic_mbox_func_to_func *func_to_func; + u64 mbox_header = *((u64 *)header); + struct hinic_recv_mbox *recv_mbox; + u64 src, dir; + + func_to_func = ((struct hinic_hwdev *)handle)->func_to_func; + + dir = HINIC_MBOX_HEADER_GET(mbox_header, DIRECTION); + src = HINIC_MBOX_HEADER_GET(mbox_header, SRC_GLB_FUNC_IDX); + + if (src >= HINIC_MAX_FUNCTIONS) { + dev_err(&func_to_func->hwif->pdev->dev, + "Mailbox source function id:%u is invalid\n", (u32)src); + return; + } + + recv_mbox = (dir == HINIC_HWIF_DIRECT_SEND) ? + &func_to_func->mbox_send[src] : + &func_to_func->mbox_resp[src]; + + recv_mbox_handler(func_to_func, (u64 *)header, recv_mbox); +} + +void hinic_mbox_self_aeqe_handler(void *handle, void *header, u8 size) +{ + struct hinic_mbox_func_to_func *func_to_func; + struct hinic_send_mbox *send_mbox; + + func_to_func = ((struct hinic_hwdev *)handle)->func_to_func; + send_mbox = &func_to_func->send_mbox; + + complete(&send_mbox->send_done); +} + +static void clear_mbox_status(struct hinic_send_mbox *mbox) +{ + *mbox->wb_status = 0; + + /* clear mailbox write back status */ + wmb(); +} + +static void mbox_copy_header(struct hinic_hwdev *hwdev, + struct hinic_send_mbox *mbox, u64 *header) +{ + u32 i, idx_max = MBOX_HEADER_SZ / sizeof(u32); + u32 *data = (u32 *)header; + + for (i = 0; i < idx_max; i++) + __raw_writel(*(data + i), mbox->data + i * sizeof(u32)); +} + +static void mbox_copy_send_data(struct hinic_hwdev *hwdev, + struct hinic_send_mbox *mbox, void *seg, + u16 seg_len) +{ + u8 mbox_max_buf[MBOX_SEG_LEN] = {0}; + u32 data_len, chk_sz = sizeof(u32); + u32 *data = seg; + u32 i, idx_max; + + /* The mbox message should be aligned in 4 bytes. */ + if (seg_len % chk_sz) { + memcpy(mbox_max_buf, seg, seg_len); + data = (u32 *)mbox_max_buf; + } + + data_len = seg_len; + idx_max = ALIGN(data_len, chk_sz) / chk_sz; + + for (i = 0; i < idx_max; i++) + __raw_writel(*(data + i), + mbox->data + MBOX_HEADER_SZ + i * sizeof(u32)); +} + +static void write_mbox_msg_attr(struct hinic_mbox_func_to_func *func_to_func, + u16 dst_func, u16 dst_aeqn, u16 seg_len, + int poll) +{ + u16 rsp_aeq = (dst_aeqn == 0) ? 0 : HINIC_MBOX_RSP_AEQN; + u32 mbox_int, mbox_ctrl; + + mbox_int = HINIC_MBOX_INT_SET(dst_func, DST_FUNC) | + HINIC_MBOX_INT_SET(dst_aeqn, DST_AEQN) | + HINIC_MBOX_INT_SET(rsp_aeq, SRC_RESP_AEQN) | + HINIC_MBOX_INT_SET(NO_DMA_ATTRIBUTE_VAL, STAT_DMA) | + HINIC_MBOX_INT_SET(ALIGN(MBOX_SEG_LEN + MBOX_HEADER_SZ + + MBOX_INFO_SZ, MBOX_SEG_LEN_ALIGN) >> 2, + TX_SIZE) | + HINIC_MBOX_INT_SET(STRONG_ORDER, STAT_DMA_SO_RO) | + HINIC_MBOX_INT_SET(WRITE_BACK, WB_EN); + + hinic_hwif_write_reg(func_to_func->hwif, + HINIC_FUNC_CSR_MAILBOX_INT_OFFSET_OFF, mbox_int); + + wmb(); /* writing the mbox int attributes */ + mbox_ctrl = HINIC_MBOX_CTRL_SET(TX_NOT_DONE, TX_STATUS); + + if (poll) + mbox_ctrl |= HINIC_MBOX_CTRL_SET(NOT_TRIGGER, TRIGGER_AEQE); + else + mbox_ctrl |= HINIC_MBOX_CTRL_SET(TRIGGER, TRIGGER_AEQE); + + hinic_hwif_write_reg(func_to_func->hwif, + HINIC_FUNC_CSR_MAILBOX_CONTROL_OFF, mbox_ctrl); +} + +static void dump_mox_reg(struct hinic_hwdev *hwdev) +{ + u32 val; + + val = hinic_hwif_read_reg(hwdev->hwif, + HINIC_FUNC_CSR_MAILBOX_CONTROL_OFF); + dev_err(&hwdev->hwif->pdev->dev, "Mailbox control reg: 0x%x\n", val); + + val = hinic_hwif_read_reg(hwdev->hwif, + HINIC_FUNC_CSR_MAILBOX_INT_OFFSET_OFF); + dev_err(&hwdev->hwif->pdev->dev, "Mailbox interrupt offset: 0x%x\n", + val); +} + +static u16 get_mbox_status(struct hinic_send_mbox *mbox) +{ + /* write back is 16B, but only use first 4B */ + u64 wb_val = be64_to_cpu(*mbox->wb_status); + + rmb(); /* verify reading before check */ + + return (u16)(wb_val & MBOX_WB_STATUS_ERRCODE_MASK); +} + +static int +wait_for_mbox_seg_completion(struct hinic_mbox_func_to_func *func_to_func, + int poll, u16 *wb_status) +{ + struct hinic_send_mbox *send_mbox = &func_to_func->send_mbox; + struct hinic_hwdev *hwdev = func_to_func->hwdev; + struct completion *done = &send_mbox->send_done; + u32 cnt = 0; + unsigned long jif; + + if (poll) { + while (cnt < MBOX_MSG_POLLING_TIMEOUT) { + *wb_status = get_mbox_status(send_mbox); + if (MBOX_STATUS_FINISHED(*wb_status)) + break; + + usleep_range(900, 1000); + cnt++; + } + + if (cnt == MBOX_MSG_POLLING_TIMEOUT) { + dev_err(&hwdev->hwif->pdev->dev, "Send mailbox segment timeout, wb status: 0x%x\n", + *wb_status); + dump_mox_reg(hwdev); + return -ETIMEDOUT; + } + } else { + jif = msecs_to_jiffies(HINIC_MBOX_COMP_TIME); + if (!wait_for_completion_timeout(done, jif)) { + dev_err(&hwdev->hwif->pdev->dev, "Send mailbox segment timeout\n"); + dump_mox_reg(hwdev); + return -ETIMEDOUT; + } + + *wb_status = get_mbox_status(send_mbox); + } + + return 0; +} + +static int send_mbox_seg(struct hinic_mbox_func_to_func *func_to_func, + u64 header, u16 dst_func, void *seg, u16 seg_len, + int poll, void *msg_info) +{ + struct hinic_send_mbox *send_mbox = &func_to_func->send_mbox; + u16 seq_dir = HINIC_MBOX_HEADER_GET(header, DIRECTION); + struct hinic_hwdev *hwdev = func_to_func->hwdev; + struct completion *done = &send_mbox->send_done; + u8 num_aeqs = hwdev->hwif->attr.num_aeqs; + u16 dst_aeqn, wb_status = 0, errcode; + + if (num_aeqs >= 4) + dst_aeqn = (seq_dir == HINIC_HWIF_DIRECT_SEND) ? + HINIC_MBOX_RECV_AEQN : HINIC_MBOX_RSP_AEQN; + else + dst_aeqn = 0; + + if (!poll) + init_completion(done); + + clear_mbox_status(send_mbox); + + mbox_copy_header(hwdev, send_mbox, &header); + + mbox_copy_send_data(hwdev, send_mbox, seg, seg_len); + + write_mbox_msg_attr(func_to_func, dst_func, dst_aeqn, seg_len, poll); + + wmb(); /* writing the mbox msg attributes */ + + if (wait_for_mbox_seg_completion(func_to_func, poll, &wb_status)) + return -ETIMEDOUT; + + if (!MBOX_STATUS_SUCCESS(wb_status)) { + dev_err(&hwdev->hwif->pdev->dev, "Send mailbox segment to function %d error, wb status: 0x%x\n", + dst_func, wb_status); + errcode = MBOX_STATUS_ERRCODE(wb_status); + return errcode ? errcode : -EFAULT; + } + + return 0; +} + +static int send_mbox_to_func(struct hinic_mbox_func_to_func *func_to_func, + enum hinic_mod_type mod, u16 cmd, void *msg, + u16 msg_len, u16 dst_func, + enum hinic_hwif_direction_type direction, + enum hinic_mbox_ack_type ack_type, + struct mbox_msg_info *msg_info) +{ + struct hinic_hwdev *hwdev = func_to_func->hwdev; + u16 seg_len = MBOX_SEG_LEN; + u8 *msg_seg = (u8 *)msg; + u16 left = msg_len; + u32 seq_id = 0; + u64 header = 0; + int err = 0; + + down(&func_to_func->msg_send_sem); + + header = HINIC_MBOX_HEADER_SET(msg_len, MSG_LEN) | + HINIC_MBOX_HEADER_SET(mod, MODULE) | + HINIC_MBOX_HEADER_SET(seg_len, SEG_LEN) | + HINIC_MBOX_HEADER_SET(ack_type, NO_ACK) | + HINIC_MBOX_HEADER_SET(SEQ_ID_START_VAL, SEQID) | + HINIC_MBOX_HEADER_SET(NOT_LAST_SEG, LAST) | + HINIC_MBOX_HEADER_SET(direction, DIRECTION) | + HINIC_MBOX_HEADER_SET(cmd, CMD) | + /* The vf's offset to it's associated pf */ + HINIC_MBOX_HEADER_SET(msg_info->msg_id, MSG_ID) | + HINIC_MBOX_HEADER_SET(msg_info->status, STATUS) | + HINIC_MBOX_HEADER_SET(hinic_global_func_id_hw(hwdev->hwif), + SRC_GLB_FUNC_IDX); + + while (!(HINIC_MBOX_HEADER_GET(header, LAST))) { + if (left <= HINIC_MBOX_SEG_LEN) { + header &= ~MBOX_SEGLEN_MASK; + header |= HINIC_MBOX_HEADER_SET(left, SEG_LEN); + header |= HINIC_MBOX_HEADER_SET(LAST_SEG, LAST); + + seg_len = left; + } + + err = send_mbox_seg(func_to_func, header, dst_func, msg_seg, + seg_len, MBOX_SEND_MSG_INT, msg_info); + if (err) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to send mbox seg, seq_id=0x%llx\n", + HINIC_MBOX_HEADER_GET(header, SEQID)); + goto err_send_mbox_seg; + } + + left -= HINIC_MBOX_SEG_LEN; + msg_seg += HINIC_MBOX_SEG_LEN; + + seq_id++; + header &= ~(HINIC_MBOX_HEADER_SET(HINIC_MBOX_HEADER_SEQID_MASK, + SEQID)); + header |= HINIC_MBOX_HEADER_SET(seq_id, SEQID); + } + +err_send_mbox_seg: + up(&func_to_func->msg_send_sem); + + return err; +} + +static void +response_for_recv_func_mbox(struct hinic_mbox_func_to_func *func_to_func, + struct hinic_recv_mbox *recv_mbox, int err, + u16 out_size, u16 src_func_idx) +{ + struct mbox_msg_info msg_info = {0}; + + if (recv_mbox->ack_type == MBOX_ACK) { + msg_info.msg_id = recv_mbox->msg_info.msg_id; + if (err == HINIC_MBOX_PF_BUSY_ACTIVE_FW) + msg_info.status = HINIC_MBOX_PF_BUSY_ACTIVE_FW; + else if (err == HINIC_MBOX_VF_CMD_ERROR) + msg_info.status = HINIC_MBOX_VF_CMD_ERROR; + else if (err) + msg_info.status = HINIC_MBOX_PF_SEND_ERR; + + /* if no data needs to response, set out_size to 1 */ + if (!out_size || err) + out_size = MBOX_MSG_NO_DATA_LEN; + + send_mbox_to_func(func_to_func, recv_mbox->mod, recv_mbox->cmd, + recv_mbox->buf_out, out_size, src_func_idx, + HINIC_HWIF_RESPONSE, MBOX_ACK, + &msg_info); + } +} + +static void recv_func_mbox_handler(struct hinic_mbox_func_to_func *func_to_func, + struct hinic_recv_mbox *recv_mbox, + u16 src_func_idx) +{ + void *buf_out = recv_mbox->buf_out; + u16 out_size = MBOX_MAX_BUF_SZ; + int err = 0; + + if (HINIC_IS_VF(func_to_func->hwif)) { + err = recv_vf_mbox_handler(func_to_func, recv_mbox, buf_out, + &out_size); + } else { + if (IS_PF_OR_PPF_SRC(src_func_idx)) + dev_warn(&func_to_func->hwif->pdev->dev, + "Unsupported pf2pf mbox msg\n"); + else + err = recv_pf_from_vf_mbox_handler(func_to_func, + recv_mbox, + src_func_idx, + buf_out, &out_size); + } + + response_for_recv_func_mbox(func_to_func, recv_mbox, err, out_size, + src_func_idx); + kfree(recv_mbox->buf_out); + kfree(recv_mbox->mbox); + kfree(recv_mbox); +} + +static void set_mbox_to_func_event(struct hinic_mbox_func_to_func *func_to_func, + enum mbox_event_state event_flag) +{ + spin_lock(&func_to_func->mbox_lock); + func_to_func->event_flag = event_flag; + spin_unlock(&func_to_func->mbox_lock); +} + +static int mbox_resp_info_handler(struct hinic_mbox_func_to_func *func_to_func, + struct hinic_recv_mbox *mbox_for_resp, + enum hinic_mod_type mod, u16 cmd, + void *buf_out, u16 *out_size) +{ + int err; + + if (mbox_for_resp->msg_info.status) { + err = mbox_for_resp->msg_info.status; + if (err != HINIC_MBOX_PF_BUSY_ACTIVE_FW) + dev_err(&func_to_func->hwif->pdev->dev, "Mbox response error(0x%x)\n", + mbox_for_resp->msg_info.status); + return err; + } + + if (buf_out && out_size) { + if (*out_size < mbox_for_resp->mbox_len) { + dev_err(&func_to_func->hwif->pdev->dev, + "Invalid response mbox message length: %d for mod %d cmd %d, should less than: %d\n", + mbox_for_resp->mbox_len, mod, cmd, *out_size); + return -EFAULT; + } + + if (mbox_for_resp->mbox_len) + memcpy(buf_out, mbox_for_resp->mbox, + mbox_for_resp->mbox_len); + + *out_size = mbox_for_resp->mbox_len; + } + + return 0; +} + +int hinic_mbox_to_func(struct hinic_mbox_func_to_func *func_to_func, + enum hinic_mod_type mod, u16 cmd, u16 dst_func, + void *buf_in, u16 in_size, void *buf_out, + u16 *out_size, u32 timeout) +{ + struct hinic_recv_mbox *mbox_for_resp; + struct mbox_msg_info msg_info = {0}; + unsigned long timeo; + int err; + + mbox_for_resp = &func_to_func->mbox_resp[dst_func]; + + down(&func_to_func->mbox_send_sem); + + init_completion(&mbox_for_resp->recv_done); + + msg_info.msg_id = MBOX_MSG_ID_INC(func_to_func); + + set_mbox_to_func_event(func_to_func, EVENT_START); + + err = send_mbox_to_func(func_to_func, mod, cmd, buf_in, in_size, + dst_func, HINIC_HWIF_DIRECT_SEND, MBOX_ACK, + &msg_info); + if (err) { + dev_err(&func_to_func->hwif->pdev->dev, "Send mailbox failed, msg_id: %d\n", + msg_info.msg_id); + set_mbox_to_func_event(func_to_func, EVENT_FAIL); + goto err_send_mbox; + } + + timeo = msecs_to_jiffies(timeout ? timeout : HINIC_MBOX_COMP_TIME); + if (!wait_for_completion_timeout(&mbox_for_resp->recv_done, timeo)) { + set_mbox_to_func_event(func_to_func, EVENT_TIMEOUT); + dev_err(&func_to_func->hwif->pdev->dev, + "Send mbox msg timeout, msg_id: %d\n", msg_info.msg_id); + err = -ETIMEDOUT; + goto err_send_mbox; + } + + set_mbox_to_func_event(func_to_func, EVENT_END); + + err = mbox_resp_info_handler(func_to_func, mbox_for_resp, mod, cmd, + buf_out, out_size); + +err_send_mbox: + up(&func_to_func->mbox_send_sem); + + return err; +} + +static int mbox_func_params_valid(struct hinic_mbox_func_to_func *func_to_func, + void *buf_in, u16 in_size) +{ + if (in_size > HINIC_MBOX_DATA_SIZE) { + dev_err(&func_to_func->hwif->pdev->dev, + "Mbox msg len(%d) exceed limit(%d)\n", + in_size, HINIC_MBOX_DATA_SIZE); + return -EINVAL; + } + + return 0; +} + +int hinic_mbox_to_pf(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod, u8 cmd, void *buf_in, + u16 in_size, void *buf_out, u16 *out_size, u32 timeout) +{ + struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func; + int err = mbox_func_params_valid(func_to_func, buf_in, in_size); + + if (err) + return err; + + if (!HINIC_IS_VF(hwdev->hwif)) { + dev_err(&hwdev->hwif->pdev->dev, "Params error, func_type: %d\n", + HINIC_FUNC_TYPE(hwdev->hwif)); + return -EINVAL; + } + + return hinic_mbox_to_func(func_to_func, mod, cmd, + hinic_pf_id_of_vf_hw(hwdev->hwif), buf_in, + in_size, buf_out, out_size, timeout); +} + +int hinic_mbox_to_vf(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod, u16 vf_id, u8 cmd, void *buf_in, + u16 in_size, void *buf_out, u16 *out_size, u32 timeout) +{ + struct hinic_mbox_func_to_func *func_to_func; + u16 dst_func_idx; + int err; + + if (!hwdev) + return -EINVAL; + + func_to_func = hwdev->func_to_func; + err = mbox_func_params_valid(func_to_func, buf_in, in_size); + if (err) + return err; + + if (HINIC_IS_VF(hwdev->hwif)) { + dev_err(&hwdev->hwif->pdev->dev, "Params error, func_type: %d\n", + HINIC_FUNC_TYPE(hwdev->hwif)); + return -EINVAL; + } + + if (!vf_id) { + dev_err(&hwdev->hwif->pdev->dev, + "VF id(%d) error!\n", vf_id); + return -EINVAL; + } + + /* vf_offset_to_pf + vf_id is the vf's global function id of vf in + * this pf + */ + dst_func_idx = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id; + + return hinic_mbox_to_func(func_to_func, mod, cmd, dst_func_idx, buf_in, + in_size, buf_out, out_size, timeout); +} + +static int init_mbox_info(struct hinic_recv_mbox *mbox_info) +{ + int err; + + mbox_info->seq_id = SEQ_ID_MAX_VAL; + + mbox_info->mbox = kzalloc(MBOX_MAX_BUF_SZ, GFP_KERNEL); + if (!mbox_info->mbox) + return -ENOMEM; + + mbox_info->buf_out = kzalloc(MBOX_MAX_BUF_SZ, GFP_KERNEL); + if (!mbox_info->buf_out) { + err = -ENOMEM; + goto err_alloc_buf_out; + } + + atomic_set(&mbox_info->msg_cnt, 0); + + return 0; + +err_alloc_buf_out: + kfree(mbox_info->mbox); + + return err; +} + +static void clean_mbox_info(struct hinic_recv_mbox *mbox_info) +{ + kfree(mbox_info->buf_out); + kfree(mbox_info->mbox); +} + +static int alloc_mbox_info(struct hinic_hwdev *hwdev, + struct hinic_recv_mbox *mbox_info) +{ + u16 func_idx, i; + int err; + + for (func_idx = 0; func_idx < HINIC_MAX_FUNCTIONS; func_idx++) { + err = init_mbox_info(&mbox_info[func_idx]); + if (err) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to init function %d mbox info\n", + func_idx); + goto err_init_mbox_info; + } + } + + return 0; + +err_init_mbox_info: + for (i = 0; i < func_idx; i++) + clean_mbox_info(&mbox_info[i]); + + return err; +} + +static void free_mbox_info(struct hinic_recv_mbox *mbox_info) +{ + u16 func_idx; + + for (func_idx = 0; func_idx < HINIC_MAX_FUNCTIONS; func_idx++) + clean_mbox_info(&mbox_info[func_idx]); +} + +static void prepare_send_mbox(struct hinic_mbox_func_to_func *func_to_func) +{ + struct hinic_send_mbox *send_mbox = &func_to_func->send_mbox; + + send_mbox->data = MBOX_AREA(func_to_func->hwif); +} + +static int alloc_mbox_wb_status(struct hinic_mbox_func_to_func *func_to_func) +{ + struct hinic_send_mbox *send_mbox = &func_to_func->send_mbox; + struct hinic_hwdev *hwdev = func_to_func->hwdev; + u32 addr_h, addr_l; + + send_mbox->wb_vaddr = dma_alloc_coherent(&hwdev->hwif->pdev->dev, + MBOX_WB_STATUS_LEN, + &send_mbox->wb_paddr, + GFP_KERNEL); + if (!send_mbox->wb_vaddr) + return -ENOMEM; + + send_mbox->wb_status = send_mbox->wb_vaddr; + + addr_h = upper_32_bits(send_mbox->wb_paddr); + addr_l = lower_32_bits(send_mbox->wb_paddr); + + hinic_hwif_write_reg(hwdev->hwif, HINIC_FUNC_CSR_MAILBOX_RESULT_H_OFF, + addr_h); + hinic_hwif_write_reg(hwdev->hwif, HINIC_FUNC_CSR_MAILBOX_RESULT_L_OFF, + addr_l); + + return 0; +} + +static void free_mbox_wb_status(struct hinic_mbox_func_to_func *func_to_func) +{ + struct hinic_send_mbox *send_mbox = &func_to_func->send_mbox; + struct hinic_hwdev *hwdev = func_to_func->hwdev; + + hinic_hwif_write_reg(hwdev->hwif, HINIC_FUNC_CSR_MAILBOX_RESULT_H_OFF, + 0); + hinic_hwif_write_reg(hwdev->hwif, HINIC_FUNC_CSR_MAILBOX_RESULT_L_OFF, + 0); + + dma_free_coherent(&hwdev->hwif->pdev->dev, MBOX_WB_STATUS_LEN, + send_mbox->wb_vaddr, + send_mbox->wb_paddr); +} + +static int comm_pf_mbox_handler(void *handle, u16 vf_id, u8 cmd, void *buf_in, + u16 in_size, void *buf_out, u16 *out_size) +{ + struct hinic_hwdev *hwdev = handle; + struct hinic_pfhwdev *pfhwdev; + int err = 0; + + pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev); + + if (cmd == HINIC_COMM_CMD_START_FLR) { + *out_size = 0; + } else { + err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_COMM, + cmd, buf_in, in_size, buf_out, out_size, + HINIC_MGMT_MSG_SYNC); + if (err && err != HINIC_MBOX_PF_BUSY_ACTIVE_FW) + dev_err(&hwdev->hwif->pdev->dev, + "PF mbox common callback handler err: %d\n", + err); + } + + return err; +} + +int hinic_func_to_func_init(struct hinic_hwdev *hwdev) +{ + struct hinic_mbox_func_to_func *func_to_func; + struct hinic_pfhwdev *pfhwdev; + int err; + + pfhwdev = container_of(hwdev, struct hinic_pfhwdev, hwdev); + func_to_func = kzalloc(sizeof(*func_to_func), GFP_KERNEL); + if (!func_to_func) + return -ENOMEM; + + hwdev->func_to_func = func_to_func; + func_to_func->hwdev = hwdev; + func_to_func->hwif = hwdev->hwif; + sema_init(&func_to_func->mbox_send_sem, 1); + sema_init(&func_to_func->msg_send_sem, 1); + spin_lock_init(&func_to_func->mbox_lock); + func_to_func->workq = create_singlethread_workqueue(HINIC_MBOX_WQ_NAME); + if (!func_to_func->workq) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to initialize MBOX workqueue\n"); + err = -ENOMEM; + goto err_create_mbox_workq; + } + + err = alloc_mbox_info(hwdev, func_to_func->mbox_send); + if (err) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to alloc mem for mbox_active\n"); + goto err_alloc_mbox_for_send; + } + + err = alloc_mbox_info(hwdev, func_to_func->mbox_resp); + if (err) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to alloc mem for mbox_passive\n"); + goto err_alloc_mbox_for_resp; + } + + err = alloc_mbox_wb_status(func_to_func); + if (err) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to alloc mbox write back status\n"); + goto err_alloc_wb_status; + } + + prepare_send_mbox(func_to_func); + + hinic_aeq_register_hw_cb(&hwdev->aeqs, HINIC_MBX_FROM_FUNC, + &pfhwdev->hwdev, hinic_mbox_func_aeqe_handler); + hinic_aeq_register_hw_cb(&hwdev->aeqs, HINIC_MBX_SEND_RSLT, + &pfhwdev->hwdev, hinic_mbox_self_aeqe_handler); + + if (!HINIC_IS_VF(hwdev->hwif)) + hinic_register_pf_mbox_cb(hwdev, HINIC_MOD_COMM, + comm_pf_mbox_handler); + + return 0; + +err_alloc_wb_status: + free_mbox_info(func_to_func->mbox_resp); + +err_alloc_mbox_for_resp: + free_mbox_info(func_to_func->mbox_send); + +err_alloc_mbox_for_send: + destroy_workqueue(func_to_func->workq); + +err_create_mbox_workq: + kfree(func_to_func); + + return err; +} + +void hinic_func_to_func_free(struct hinic_hwdev *hwdev) +{ + struct hinic_mbox_func_to_func *func_to_func = hwdev->func_to_func; + + hinic_aeq_unregister_hw_cb(&hwdev->aeqs, HINIC_MBX_FROM_FUNC); + hinic_aeq_unregister_hw_cb(&hwdev->aeqs, HINIC_MBX_SEND_RSLT); + + hinic_unregister_pf_mbox_cb(hwdev, HINIC_MOD_COMM); + /* destroy workqueue before free related mbox resources in case of + * illegal resource access + */ + destroy_workqueue(func_to_func->workq); + + free_mbox_wb_status(func_to_func); + free_mbox_info(func_to_func->mbox_resp); + free_mbox_info(func_to_func->mbox_send); + + kfree(func_to_func); +} diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_mbox.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_mbox.h new file mode 100644 index 000000000000..7b18559bfe80 --- /dev/null +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_mbox.h @@ -0,0 +1,154 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Huawei HiNIC PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + */ + +#ifndef HINIC_MBOX_H_ +#define HINIC_MBOX_H_ + +#define HINIC_MBOX_PF_SEND_ERR 0x1 +#define HINIC_MBOX_PF_BUSY_ACTIVE_FW 0x2 +#define HINIC_MBOX_VF_CMD_ERROR 0x3 + +#define HINIC_MAX_FUNCTIONS 512 + +#define HINIC_MAX_PF_FUNCS 16 + +#define HINIC_MBOX_WQ_NAME "hinic_mbox" + +#define HINIC_FUNC_CSR_MAILBOX_DATA_OFF 0x80 +#define HINIC_FUNC_CSR_MAILBOX_CONTROL_OFF 0x0100 +#define HINIC_FUNC_CSR_MAILBOX_INT_OFFSET_OFF 0x0104 +#define HINIC_FUNC_CSR_MAILBOX_RESULT_H_OFF 0x0108 +#define HINIC_FUNC_CSR_MAILBOX_RESULT_L_OFF 0x010C + +enum hinic_mbox_ack_type { + MBOX_ACK, + MBOX_NO_ACK, +}; + +struct mbox_msg_info { + u8 msg_id; + u8 status; +}; + +struct hinic_recv_mbox { + struct completion recv_done; + void *mbox; + u8 cmd; + enum hinic_mod_type mod; + u16 mbox_len; + void *buf_out; + enum hinic_mbox_ack_type ack_type; + struct mbox_msg_info msg_info; + u8 seq_id; + atomic_t msg_cnt; +}; + +struct hinic_send_mbox { + struct completion send_done; + u8 *data; + + u64 *wb_status; + void *wb_vaddr; + dma_addr_t wb_paddr; +}; + +typedef void (*hinic_vf_mbox_cb)(void *handle, u8 cmd, void *buf_in, + u16 in_size, void *buf_out, u16 *out_size); +typedef int (*hinic_pf_mbox_cb)(void *handle, u16 vf_id, u8 cmd, void *buf_in, + u16 in_size, void *buf_out, u16 *out_size); + +enum mbox_event_state { + EVENT_START = 0, + EVENT_FAIL, + EVENT_TIMEOUT, + EVENT_END, +}; + +enum hinic_mbox_cb_state { + HINIC_VF_MBOX_CB_REG = 0, + HINIC_VF_MBOX_CB_RUNNING, + HINIC_PF_MBOX_CB_REG, + HINIC_PF_MBOX_CB_RUNNING, + HINIC_PPF_MBOX_CB_REG, + HINIC_PPF_MBOX_CB_RUNNING, + HINIC_PPF_TO_PF_MBOX_CB_REG, + HINIC_PPF_TO_PF_MBOX_CB_RUNNIG, +}; + +struct hinic_mbox_func_to_func { + struct hinic_hwdev *hwdev; + struct hinic_hwif *hwif; + + struct semaphore mbox_send_sem; + struct semaphore msg_send_sem; + struct hinic_send_mbox send_mbox; + + struct workqueue_struct *workq; + + struct hinic_recv_mbox mbox_resp[HINIC_MAX_FUNCTIONS]; + struct hinic_recv_mbox mbox_send[HINIC_MAX_FUNCTIONS]; + + hinic_vf_mbox_cb vf_mbox_cb[HINIC_MOD_MAX]; + hinic_pf_mbox_cb pf_mbox_cb[HINIC_MOD_MAX]; + unsigned long pf_mbox_cb_state[HINIC_MOD_MAX]; + unsigned long vf_mbox_cb_state[HINIC_MOD_MAX]; + + u8 send_msg_id; + enum mbox_event_state event_flag; + + /* lock for mbox event flag */ + spinlock_t mbox_lock; +}; + +struct hinic_mbox_work { + struct work_struct work; + u16 src_func_idx; + struct hinic_mbox_func_to_func *func_to_func; + struct hinic_recv_mbox *recv_mbox; +}; + +struct vf_cmd_msg_handle { + u8 cmd; + int (*cmd_msg_handler)(void *hwdev, u16 vf_id, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size); +}; + +int hinic_register_pf_mbox_cb(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod, + hinic_pf_mbox_cb callback); + +int hinic_register_vf_mbox_cb(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod, + hinic_vf_mbox_cb callback); + +void hinic_unregister_pf_mbox_cb(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod); + +void hinic_unregister_vf_mbox_cb(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod); + +void hinic_mbox_func_aeqe_handler(void *handle, void *header, u8 size); + +void hinic_mbox_self_aeqe_handler(void *handle, void *header, u8 size); + +int hinic_func_to_func_init(struct hinic_hwdev *hwdev); + +void hinic_func_to_func_free(struct hinic_hwdev *hwdev); + +int hinic_mbox_to_pf(struct hinic_hwdev *hwdev, enum hinic_mod_type mod, + u8 cmd, void *buf_in, u16 in_size, void *buf_out, + u16 *out_size, u32 timeout); + +int hinic_mbox_to_func(struct hinic_mbox_func_to_func *func_to_func, + enum hinic_mod_type mod, u16 cmd, u16 dst_func, + void *buf_in, u16 in_size, void *buf_out, + u16 *out_size, u32 timeout); + +int hinic_mbox_to_vf(struct hinic_hwdev *hwdev, + enum hinic_mod_type mod, u16 vf_id, u8 cmd, void *buf_in, + u16 in_size, void *buf_out, u16 *out_size, u32 timeout); + +#endif diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c index 992908e6eebf..c33eb1147055 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.c @@ -361,7 +361,11 @@ int hinic_msg_to_mgmt(struct hinic_pf_to_mgmt *pf_to_mgmt, if (cmd == HINIC_PORT_CMD_SET_FUNC_STATE) timeout = SET_FUNC_PORT_MGMT_TIMEOUT; - return msg_to_mgmt_sync(pf_to_mgmt, mod, cmd, buf_in, in_size, + if (HINIC_IS_VF(hwif)) + return hinic_mbox_to_pf(pf_to_mgmt->hwdev, mod, cmd, buf_in, + in_size, buf_out, out_size, 0); + else + return msg_to_mgmt_sync(pf_to_mgmt, mod, cmd, buf_in, in_size, buf_out, out_size, MGMT_DIRECT_SEND, MSG_NOT_RESP, timeout); } @@ -398,8 +402,8 @@ static void mgmt_recv_msg_handler(struct hinic_pf_to_mgmt *pf_to_mgmt, recv_msg->msg, recv_msg->msg_len, buf_out, &out_size); else - dev_err(&pdev->dev, "No MGMT msg handler, mod = %d\n", - recv_msg->mod); + dev_err(&pdev->dev, "No MGMT msg handler, mod: %d, cmd: %d\n", + recv_msg->mod, recv_msg->cmd); mgmt_cb->state &= ~HINIC_MGMT_CB_RUNNING; @@ -561,6 +565,10 @@ int hinic_pf_to_mgmt_init(struct hinic_pf_to_mgmt *pf_to_mgmt, int err; pf_to_mgmt->hwif = hwif; + pf_to_mgmt->hwdev = hwdev; + + if (HINIC_IS_VF(hwif)) + return 0; sema_init(&pf_to_mgmt->sync_msg_lock, 1); pf_to_mgmt->sync_msg_id = 0; @@ -592,6 +600,9 @@ void hinic_pf_to_mgmt_free(struct hinic_pf_to_mgmt *pf_to_mgmt) struct hinic_pfhwdev *pfhwdev = mgmt_to_pfhwdev(pf_to_mgmt); struct hinic_hwdev *hwdev = &pfhwdev->hwdev; + if (HINIC_IS_VF(hwdev->hwif)) + return; + hinic_aeq_unregister_hw_cb(&hwdev->aeqs, HINIC_MSG_FROM_MGMT_CPU); hinic_api_cmd_free(pf_to_mgmt->cmd_chain); } diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.h index 182fba17b643..c2b142c08b0e 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_mgmt.h @@ -60,7 +60,9 @@ enum hinic_cfg_cmd { }; enum hinic_comm_cmd { + HINIC_COMM_CMD_START_FLR = 0x1, HINIC_COMM_CMD_IO_STATUS_GET = 0x3, + HINIC_COMM_CMD_DMA_ATTR_SET = 0x4, HINIC_COMM_CMD_CMDQ_CTXT_SET = 0x10, HINIC_COMM_CMD_CMDQ_CTXT_GET = 0x11, @@ -74,7 +76,13 @@ enum hinic_comm_cmd { HINIC_COMM_CMD_IO_RES_CLEAR = 0x29, - HINIC_COMM_CMD_MAX = 0x32, + HINIC_COMM_CMD_CEQ_CTRL_REG_WR_BY_UP = 0x33, + + HINIC_COMM_CMD_L2NIC_RESET = 0x4b, + + HINIC_COMM_CMD_PAGESIZE_SET = 0x50, + + HINIC_COMM_CMD_MAX = 0x51, }; enum hinic_mgmt_cb_state { @@ -107,7 +115,7 @@ struct hinic_mgmt_cb { struct hinic_pf_to_mgmt { struct hinic_hwif *hwif; - + struct hinic_hwdev *hwdev; struct semaphore sync_msg_lock; u16 sync_msg_id; u8 *sync_msg_buf; diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.c index be364b7a7019..fcf7bfe4aa47 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.c @@ -108,7 +108,12 @@ void hinic_sq_prepare_ctxt(struct hinic_sq_ctxt *sq_ctxt, wq_page_pfn_hi = upper_32_bits(wq_page_pfn); wq_page_pfn_lo = lower_32_bits(wq_page_pfn); - wq_block_pfn = HINIC_WQ_BLOCK_PFN(wq->block_paddr); + /* If only one page, use 0-level CLA */ + if (wq->num_q_pages == 1) + wq_block_pfn = HINIC_WQ_BLOCK_PFN(wq_page_addr); + else + wq_block_pfn = HINIC_WQ_BLOCK_PFN(wq->block_paddr); + wq_block_pfn_hi = upper_32_bits(wq_block_pfn); wq_block_pfn_lo = lower_32_bits(wq_block_pfn); @@ -638,6 +643,7 @@ void hinic_sq_write_db(struct hinic_sq *sq, u16 prod_idx, unsigned int wqe_size, /* increment prod_idx to the next */ prod_idx += ALIGN(wqe_size, wq->wqebb_size) / wq->wqebb_size; + prod_idx = SQ_MASKED_IDX(sq, prod_idx); wmb(); /* Write all before the doorbell */ diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h index 79091e131418..ca3e2d060284 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_qp.h @@ -38,12 +38,15 @@ #define HINIC_SQ_WQEBB_SIZE 64 #define HINIC_RQ_WQEBB_SIZE 32 -#define HINIC_SQ_PAGE_SIZE SZ_4K -#define HINIC_RQ_PAGE_SIZE SZ_4K +#define HINIC_SQ_PAGE_SIZE SZ_256K +#define HINIC_RQ_PAGE_SIZE SZ_256K #define HINIC_SQ_DEPTH SZ_4K #define HINIC_RQ_DEPTH SZ_4K +#define HINIC_MAX_QUEUE_DEPTH SZ_4K +#define HINIC_MIN_QUEUE_DEPTH 128 + /* In any change to HINIC_RX_BUF_SZ, HINIC_RX_BUF_SZ_IDX must be changed */ #define HINIC_RX_BUF_SZ 2048 #define HINIC_RX_BUF_SZ_IDX HINIC_RX_BUF_SZ_2048_IDX diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.c b/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.c index 03363216ff59..5dc3743f8091 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.c @@ -503,7 +503,7 @@ err_alloc_wq_pages: * Return 0 - Success, negative - Failure **/ int hinic_wq_allocate(struct hinic_wqs *wqs, struct hinic_wq *wq, - u16 wqebb_size, u16 wq_page_size, u16 q_depth, + u16 wqebb_size, u32 wq_page_size, u16 q_depth, u16 max_wqe_size) { struct hinic_hwif *hwif = wqs->hwif; @@ -600,7 +600,7 @@ void hinic_wq_free(struct hinic_wqs *wqs, struct hinic_wq *wq) **/ int hinic_wqs_cmdq_alloc(struct hinic_cmdq_pages *cmdq_pages, struct hinic_wq *wq, struct hinic_hwif *hwif, - int cmdq_blocks, u16 wqebb_size, u16 wq_page_size, + int cmdq_blocks, u16 wqebb_size, u32 wq_page_size, u16 q_depth, u16 max_wqe_size) { struct pci_dev *pdev = hwif->pdev; @@ -768,7 +768,10 @@ struct hinic_hw_wqe *hinic_get_wqe(struct hinic_wq *wq, unsigned int wqe_size, *prod_idx = curr_prod_idx; - if (curr_pg != end_pg) { + /* If we only have one page, still need to get shadown wqe when + * wqe rolling-over page + */ + if (curr_pg != end_pg || MASKED_WQE_IDX(wq, end_prod_idx) < *prod_idx) { void *shadow_addr = &wq->shadow_wqe[curr_pg * wq->max_wqe_size]; copy_wqe_to_shadow(wq, shadow_addr, num_wqebbs, *prod_idx); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.h b/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.h index 811eef744140..b06f8c0255de 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_hw_wq.h @@ -26,7 +26,7 @@ struct hinic_wq { int block_idx; u16 wqebb_size; - u16 wq_page_size; + u32 wq_page_size; u16 q_depth; u16 max_wqe_size; u16 num_wqebbs_per_page; @@ -76,7 +76,7 @@ struct hinic_cmdq_pages { int hinic_wqs_cmdq_alloc(struct hinic_cmdq_pages *cmdq_pages, struct hinic_wq *wq, struct hinic_hwif *hwif, - int cmdq_blocks, u16 wqebb_size, u16 wq_page_size, + int cmdq_blocks, u16 wqebb_size, u32 wq_page_size, u16 q_depth, u16 max_wqe_size); void hinic_wqs_cmdq_free(struct hinic_cmdq_pages *cmdq_pages, @@ -88,7 +88,7 @@ int hinic_wqs_alloc(struct hinic_wqs *wqs, int num_wqs, void hinic_wqs_free(struct hinic_wqs *wqs); int hinic_wq_allocate(struct hinic_wqs *wqs, struct hinic_wq *wq, - u16 wqebb_size, u16 wq_page_size, u16 q_depth, + u16 wqebb_size, u32 wq_page_size, u16 q_depth, u16 max_wqe_size); void hinic_wq_free(struct hinic_wqs *wqs, struct hinic_wq *wq); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_main.c b/drivers/net/ethernet/huawei/hinic/hinic_main.c index 63b92f6cc856..e9e6f4c9309a 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_main.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_main.c @@ -29,6 +29,7 @@ #include "hinic_tx.h" #include "hinic_rx.h" #include "hinic_dev.h" +#include "hinic_sriov.h" MODULE_AUTHOR("Huawei Technologies CO., Ltd"); MODULE_DESCRIPTION("Huawei Intelligent NIC driver"); @@ -46,6 +47,7 @@ MODULE_PARM_DESC(rx_weight, "Number Rx packets for NAPI budget (default=64)"); #define HINIC_DEV_ID_DUAL_PORT_100GE 0x0200 #define HINIC_DEV_ID_DUAL_PORT_100GE_MEZZ 0x0205 #define HINIC_DEV_ID_QUAD_PORT_25GE_MEZZ 0x0210 +#define HINIC_DEV_ID_VF 0x375e #define HINIC_WQ_NAME "hinic_dev" @@ -65,6 +67,8 @@ MODULE_PARM_DESC(rx_weight, "Number Rx packets for NAPI budget (default=64)"); #define rx_mode_work_to_nic_dev(rx_mode_work) \ container_of(rx_mode_work, struct hinic_dev, rx_mode_work) +#define HINIC_WAIT_SRIOV_CFG_TIMEOUT 15000 + static int change_mac_addr(struct net_device *netdev, const u8 *addr); static int set_features(struct hinic_dev *nic_dev, @@ -322,7 +326,6 @@ static void hinic_enable_rss(struct hinic_dev *nic_dev) int i, node, err = 0; u16 num_cpus = 0; - nic_dev->max_qps = hinic_hwdev_max_num_qps(hwdev); if (nic_dev->max_qps <= 1) { nic_dev->flags &= ~HINIC_RSS_ENABLE; nic_dev->rss_limit = nic_dev->max_qps; @@ -368,14 +371,15 @@ static void hinic_enable_rss(struct hinic_dev *nic_dev) netif_err(nic_dev, drv, netdev, "Failed to init rss\n"); } -static int hinic_open(struct net_device *netdev) +int hinic_open(struct net_device *netdev) { struct hinic_dev *nic_dev = netdev_priv(netdev); enum hinic_port_link_state link_state; int err, ret; if (!(nic_dev->flags & HINIC_INTF_UP)) { - err = hinic_hwdev_ifup(nic_dev->hwdev); + err = hinic_hwdev_ifup(nic_dev->hwdev, nic_dev->sq_depth, + nic_dev->rq_depth); if (err) { netif_err(nic_dev, drv, netdev, "Failed - HW interface up\n"); @@ -423,9 +427,6 @@ static int hinic_open(struct net_device *netdev) goto err_func_port_state; } - /* Wait up to 3 sec between port enable to link state */ - msleep(3000); - down(&nic_dev->mgmt_lock); err = hinic_port_link_state(nic_dev, &link_state); @@ -434,6 +435,9 @@ static int hinic_open(struct net_device *netdev) goto err_port_link; } + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) + hinic_notify_all_vfs_link_changed(nic_dev->hwdev, link_state); + if (link_state == HINIC_LINK_STATE_UP) nic_dev->flags |= HINIC_LINK_UP; @@ -479,7 +483,7 @@ err_create_txqs: return err; } -static int hinic_close(struct net_device *netdev) +int hinic_close(struct net_device *netdev) { struct hinic_dev *nic_dev = netdev_priv(netdev); unsigned int flags; @@ -496,6 +500,9 @@ static int hinic_close(struct net_device *netdev) up(&nic_dev->mgmt_lock); + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) + hinic_notify_all_vfs_link_changed(nic_dev->hwdev, 0); + hinic_port_set_state(nic_dev, HINIC_PORT_DISABLE); hinic_port_set_func_state(nic_dev, HINIC_FUNC_PORT_DISABLE); @@ -673,7 +680,7 @@ static int hinic_vlan_rx_add_vid(struct net_device *netdev, } err = hinic_port_add_mac(nic_dev, netdev->dev_addr, vid); - if (err) { + if (err && err != HINIC_PF_SET_VF_ALREADY) { netif_err(nic_dev, drv, netdev, "Failed to set mac\n"); goto err_add_mac; } @@ -725,8 +732,6 @@ static void set_rx_mode(struct work_struct *work) struct hinic_rx_mode_work *rx_mode_work = work_to_rx_mode_work(work); struct hinic_dev *nic_dev = rx_mode_work_to_nic_dev(rx_mode_work); - netif_info(nic_dev, drv, nic_dev->netdev, "set rx mode work\n"); - hinic_port_set_rx_mode(nic_dev, rx_mode_work->rx_mode); __dev_uc_sync(nic_dev->netdev, add_mac_addr, remove_mac_addr); @@ -745,10 +750,12 @@ static void hinic_set_rx_mode(struct net_device *netdev) HINIC_RX_MODE_MC | HINIC_RX_MODE_BC; - if (netdev->flags & IFF_PROMISC) - rx_mode |= HINIC_RX_MODE_PROMISC; - else if (netdev->flags & IFF_ALLMULTI) + if (netdev->flags & IFF_PROMISC) { + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) + rx_mode |= HINIC_RX_MODE_PROMISC; + } else if (netdev->flags & IFF_ALLMULTI) { rx_mode |= HINIC_RX_MODE_MC_ALL; + } rx_mode_work->rx_mode = rx_mode; @@ -758,8 +765,26 @@ static void hinic_set_rx_mode(struct net_device *netdev) static void hinic_tx_timeout(struct net_device *netdev, unsigned int txqueue) { struct hinic_dev *nic_dev = netdev_priv(netdev); + u16 sw_pi, hw_ci, sw_ci; + struct hinic_sq *sq; + u16 num_sqs, q_id; + + num_sqs = hinic_hwdev_num_qps(nic_dev->hwdev); netif_err(nic_dev, drv, netdev, "Tx timeout\n"); + + for (q_id = 0; q_id < num_sqs; q_id++) { + if (!netif_xmit_stopped(netdev_get_tx_queue(netdev, q_id))) + continue; + + sq = hinic_hwdev_get_sq(nic_dev->hwdev, q_id); + sw_pi = atomic_read(&sq->wq->prod_idx) & sq->wq->mask; + hw_ci = be16_to_cpu(*(u16 *)(sq->hw_ci_addr)) & sq->wq->mask; + sw_ci = atomic_read(&sq->wq->cons_idx) & sq->wq->mask; + netif_err(nic_dev, drv, netdev, "Txq%d: sw_pi: %d, hw_ci: %d, sw_ci: %d, napi->state: 0x%lx\n", + q_id, sw_pi, hw_ci, sw_ci, + nic_dev->txqs[q_id].napi.state); + } } static void hinic_get_stats64(struct net_device *netdev, @@ -825,6 +850,29 @@ static const struct net_device_ops hinic_netdev_ops = { .ndo_get_stats64 = hinic_get_stats64, .ndo_fix_features = hinic_fix_features, .ndo_set_features = hinic_set_features, + .ndo_set_vf_mac = hinic_ndo_set_vf_mac, + .ndo_set_vf_vlan = hinic_ndo_set_vf_vlan, + .ndo_get_vf_config = hinic_ndo_get_vf_config, + .ndo_set_vf_trust = hinic_ndo_set_vf_trust, + .ndo_set_vf_rate = hinic_ndo_set_vf_bw, + .ndo_set_vf_spoofchk = hinic_ndo_set_vf_spoofchk, + .ndo_set_vf_link_state = hinic_ndo_set_vf_link_state, +}; + +static const struct net_device_ops hinicvf_netdev_ops = { + .ndo_open = hinic_open, + .ndo_stop = hinic_close, + .ndo_change_mtu = hinic_change_mtu, + .ndo_set_mac_address = hinic_set_mac_addr, + .ndo_validate_addr = eth_validate_addr, + .ndo_vlan_rx_add_vid = hinic_vlan_rx_add_vid, + .ndo_vlan_rx_kill_vid = hinic_vlan_rx_kill_vid, + .ndo_set_rx_mode = hinic_set_rx_mode, + .ndo_start_xmit = hinic_xmit_frame, + .ndo_tx_timeout = hinic_tx_timeout, + .ndo_get_stats64 = hinic_get_stats64, + .ndo_fix_features = hinic_fix_features, + .ndo_set_features = hinic_set_features, }; static void netdev_features_init(struct net_device *netdev) @@ -884,6 +932,10 @@ static void link_status_event_handler(void *handle, void *buf_in, u16 in_size, netif_info(nic_dev, drv, nic_dev->netdev, "HINIC_Link is DOWN\n"); } + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) + hinic_notify_all_vfs_link_changed(nic_dev->hwdev, + link_status->link); + ret_link_status = buf_out; ret_link_status->status = 0; @@ -957,7 +1009,12 @@ static int nic_dev_init(struct pci_dev *pdev) } hinic_set_ethtool_ops(netdev); - netdev->netdev_ops = &hinic_netdev_ops; + + if (!HINIC_IS_VF(hwdev->hwif)) + netdev->netdev_ops = &hinic_netdev_ops; + else + netdev->netdev_ops = &hinicvf_netdev_ops; + netdev->max_mtu = ETH_MAX_MTU; nic_dev = netdev_priv(netdev); @@ -969,6 +1026,11 @@ static int nic_dev_init(struct pci_dev *pdev) nic_dev->rxqs = NULL; nic_dev->tx_weight = tx_weight; nic_dev->rx_weight = rx_weight; + nic_dev->sq_depth = HINIC_SQ_DEPTH; + nic_dev->rq_depth = HINIC_RQ_DEPTH; + nic_dev->sriov_info.hwdev = hwdev; + nic_dev->sriov_info.pdev = pdev; + nic_dev->max_qps = num_qps; sema_init(&nic_dev->mgmt_lock, 1); @@ -995,11 +1057,25 @@ static int nic_dev_init(struct pci_dev *pdev) pci_set_drvdata(pdev, netdev); err = hinic_port_get_mac(nic_dev, netdev->dev_addr); - if (err) - dev_warn(&pdev->dev, "Failed to get mac address\n"); + if (err) { + dev_err(&pdev->dev, "Failed to get mac address\n"); + goto err_get_mac; + } + + if (!is_valid_ether_addr(netdev->dev_addr)) { + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) { + dev_err(&pdev->dev, "Invalid MAC address\n"); + err = -EIO; + goto err_add_mac; + } + + dev_info(&pdev->dev, "Invalid MAC address %pM, using random\n", + netdev->dev_addr); + eth_hw_addr_random(netdev); + } err = hinic_port_add_mac(nic_dev, netdev->dev_addr, 0); - if (err) { + if (err && err != HINIC_PF_SET_VF_ALREADY) { dev_err(&pdev->dev, "Failed to add mac\n"); goto err_add_mac; } @@ -1041,6 +1117,7 @@ err_set_features: cancel_work_sync(&rx_mode_work->work); err_set_mtu: +err_get_mac: err_add_mac: pci_set_drvdata(pdev, NULL); destroy_workqueue(nic_dev->workq); @@ -1114,14 +1191,41 @@ err_pci_regions: return err; } +#define HINIC_WAIT_SRIOV_CFG_TIMEOUT 15000 + +static void wait_sriov_cfg_complete(struct hinic_dev *nic_dev) +{ + struct hinic_sriov_info *sriov_info = &nic_dev->sriov_info; + u32 loop_cnt = 0; + + set_bit(HINIC_FUNC_REMOVE, &sriov_info->state); + usleep_range(9900, 10000); + + while (loop_cnt < HINIC_WAIT_SRIOV_CFG_TIMEOUT) { + if (!test_bit(HINIC_SRIOV_ENABLE, &sriov_info->state) && + !test_bit(HINIC_SRIOV_DISABLE, &sriov_info->state)) + return; + + usleep_range(9900, 10000); + loop_cnt++; + } +} + static void hinic_remove(struct pci_dev *pdev) { struct net_device *netdev = pci_get_drvdata(pdev); struct hinic_dev *nic_dev = netdev_priv(netdev); struct hinic_rx_mode_work *rx_mode_work; + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) { + wait_sriov_cfg_complete(nic_dev); + hinic_pci_sriov_disable(pdev); + } + unregister_netdev(netdev); + hinic_port_del_mac(nic_dev, netdev->dev_addr, 0); + hinic_hwdev_cb_unregister(nic_dev->hwdev, HINIC_MGMT_MSG_CMD_LINK_STATUS); @@ -1132,6 +1236,8 @@ static void hinic_remove(struct pci_dev *pdev) destroy_workqueue(nic_dev->workq); + hinic_vf_func_free(nic_dev->hwdev); + hinic_free_hwdev(nic_dev->hwdev); free_netdev(netdev); @@ -1152,6 +1258,7 @@ static const struct pci_device_id hinic_pci_table[] = { { PCI_VDEVICE(HUAWEI, HINIC_DEV_ID_DUAL_PORT_100GE), 0}, { PCI_VDEVICE(HUAWEI, HINIC_DEV_ID_DUAL_PORT_100GE_MEZZ), 0}, { PCI_VDEVICE(HUAWEI, HINIC_DEV_ID_QUAD_PORT_25GE_MEZZ), 0}, + { PCI_VDEVICE(HUAWEI, HINIC_DEV_ID_VF), 0}, { 0, 0} }; MODULE_DEVICE_TABLE(pci, hinic_pci_table); @@ -1162,6 +1269,7 @@ static struct pci_driver hinic_driver = { .probe = hinic_probe, .remove = hinic_remove, .shutdown = hinic_shutdown, + .sriov_configure = hinic_pci_sriov_configure, }; module_pci_driver(hinic_driver); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_port.c b/drivers/net/ethernet/huawei/hinic/hinic_port.c index 1e389a004e50..175c0ee00038 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_port.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_port.c @@ -37,20 +37,14 @@ enum mac_op { static int change_mac(struct hinic_dev *nic_dev, const u8 *addr, u16 vlan_id, enum mac_op op) { - struct net_device *netdev = nic_dev->netdev; struct hinic_hwdev *hwdev = nic_dev->hwdev; struct hinic_port_mac_cmd port_mac_cmd; struct hinic_hwif *hwif = hwdev->hwif; + u16 out_size = sizeof(port_mac_cmd); struct pci_dev *pdev = hwif->pdev; enum hinic_port_cmd cmd; - u16 out_size; int err; - if (vlan_id >= VLAN_N_VID) { - netif_err(nic_dev, drv, netdev, "Invalid VLAN number\n"); - return -EINVAL; - } - if (op == MAC_SET) cmd = HINIC_PORT_CMD_SET_MAC; else @@ -63,12 +57,25 @@ static int change_mac(struct hinic_dev *nic_dev, const u8 *addr, err = hinic_port_msg_cmd(hwdev, cmd, &port_mac_cmd, sizeof(port_mac_cmd), &port_mac_cmd, &out_size); - if (err || (out_size != sizeof(port_mac_cmd)) || port_mac_cmd.status) { + if (err || out_size != sizeof(port_mac_cmd) || + (port_mac_cmd.status && + port_mac_cmd.status != HINIC_PF_SET_VF_ALREADY && + port_mac_cmd.status != HINIC_MGMT_STATUS_EXIST)) { dev_err(&pdev->dev, "Failed to change MAC, ret = %d\n", port_mac_cmd.status); return -EFAULT; } + if (port_mac_cmd.status == HINIC_PF_SET_VF_ALREADY) { + dev_warn(&pdev->dev, "PF has already set VF mac, ignore %s operation\n", + (op == MAC_SET) ? "set" : "del"); + return HINIC_PF_SET_VF_ALREADY; + } + + if (cmd == HINIC_PORT_CMD_SET_MAC && port_mac_cmd.status == + HINIC_MGMT_STATUS_EXIST) + dev_warn(&pdev->dev, "MAC is repeated, ignore set operation\n"); + return 0; } @@ -112,8 +119,8 @@ int hinic_port_get_mac(struct hinic_dev *nic_dev, u8 *addr) struct hinic_hwdev *hwdev = nic_dev->hwdev; struct hinic_port_mac_cmd port_mac_cmd; struct hinic_hwif *hwif = hwdev->hwif; + u16 out_size = sizeof(port_mac_cmd); struct pci_dev *pdev = hwif->pdev; - u16 out_size; int err; port_mac_cmd.func_idx = HINIC_HWIF_FUNC_IDX(hwif); @@ -144,9 +151,9 @@ int hinic_port_set_mtu(struct hinic_dev *nic_dev, int new_mtu) struct hinic_hwdev *hwdev = nic_dev->hwdev; struct hinic_port_mtu_cmd port_mtu_cmd; struct hinic_hwif *hwif = hwdev->hwif; + u16 out_size = sizeof(port_mtu_cmd); struct pci_dev *pdev = hwif->pdev; int err, max_frame; - u16 out_size; if (new_mtu < HINIC_MIN_MTU_SIZE) { netif_err(nic_dev, drv, netdev, "mtu < MIN MTU size"); @@ -248,14 +255,9 @@ int hinic_port_link_state(struct hinic_dev *nic_dev, struct hinic_hwif *hwif = hwdev->hwif; struct hinic_port_link_cmd link_cmd; struct pci_dev *pdev = hwif->pdev; - u16 out_size; + u16 out_size = sizeof(link_cmd); int err; - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "unsupported PCI Function type\n"); - return -EINVAL; - } - link_cmd.func_idx = HINIC_HWIF_FUNC_IDX(hwif); err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_LINK_STATE, @@ -284,13 +286,11 @@ int hinic_port_set_state(struct hinic_dev *nic_dev, enum hinic_port_state state) struct hinic_port_state_cmd port_state; struct hinic_hwif *hwif = hwdev->hwif; struct pci_dev *pdev = hwif->pdev; - u16 out_size; + u16 out_size = sizeof(port_state); int err; - if (!HINIC_IS_PF(hwif) && !HINIC_IS_PPF(hwif)) { - dev_err(&pdev->dev, "unsupported PCI Function type\n"); - return -EINVAL; - } + if (HINIC_IS_VF(hwdev->hwif)) + return 0; port_state.state = state; @@ -320,7 +320,7 @@ int hinic_port_set_func_state(struct hinic_dev *nic_dev, struct hinic_hwdev *hwdev = nic_dev->hwdev; struct hinic_hwif *hwif = hwdev->hwif; struct pci_dev *pdev = hwif->pdev; - u16 out_size; + u16 out_size = sizeof(func_state); int err; func_state.func_idx = HINIC_HWIF_FUNC_IDX(hwif); @@ -351,7 +351,7 @@ int hinic_port_get_cap(struct hinic_dev *nic_dev, struct hinic_hwdev *hwdev = nic_dev->hwdev; struct hinic_hwif *hwif = hwdev->hwif; struct pci_dev *pdev = hwif->pdev; - u16 out_size; + u16 out_size = sizeof(*port_cap); int err; port_cap->func_idx = HINIC_HWIF_FUNC_IDX(hwif); @@ -382,7 +382,7 @@ int hinic_port_set_tso(struct hinic_dev *nic_dev, enum hinic_tso_state state) struct hinic_hwif *hwif = hwdev->hwif; struct hinic_tso_config tso_cfg = {0}; struct pci_dev *pdev = hwif->pdev; - u16 out_size; + u16 out_size = sizeof(tso_cfg); int err; tso_cfg.func_id = HINIC_HWIF_FUNC_IDX(hwif); @@ -405,9 +405,9 @@ int hinic_set_rx_csum_offload(struct hinic_dev *nic_dev, u32 en) { struct hinic_checksum_offload rx_csum_cfg = {0}; struct hinic_hwdev *hwdev = nic_dev->hwdev; + u16 out_size = sizeof(rx_csum_cfg); struct hinic_hwif *hwif; struct pci_dev *pdev; - u16 out_size; int err; if (!hwdev) @@ -443,6 +443,7 @@ int hinic_set_rx_vlan_offload(struct hinic_dev *nic_dev, u8 en) if (!hwdev) return -EINVAL; + out_size = sizeof(vlan_cfg); hwif = hwdev->hwif; pdev = hwif->pdev; vlan_cfg.func_id = HINIC_HWIF_FUNC_IDX(hwif); @@ -465,14 +466,14 @@ int hinic_set_max_qnum(struct hinic_dev *nic_dev, u8 num_rqs) { struct hinic_hwdev *hwdev = nic_dev->hwdev; struct hinic_hwif *hwif = hwdev->hwif; - struct pci_dev *pdev = hwif->pdev; struct hinic_rq_num rq_num = { 0 }; + struct pci_dev *pdev = hwif->pdev; u16 out_size = sizeof(rq_num); int err; rq_num.func_id = HINIC_HWIF_FUNC_IDX(hwif); rq_num.num_rqs = num_rqs; - rq_num.rq_depth = ilog2(HINIC_SQ_DEPTH); + rq_num.rq_depth = ilog2(nic_dev->rq_depth); err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_RQ_IQ_MAP, &rq_num, sizeof(rq_num), @@ -491,8 +492,8 @@ static int hinic_set_rx_lro(struct hinic_dev *nic_dev, u8 ipv4_en, u8 ipv6_en, u8 max_wqe_num) { struct hinic_hwdev *hwdev = nic_dev->hwdev; - struct hinic_hwif *hwif = hwdev->hwif; struct hinic_lro_config lro_cfg = { 0 }; + struct hinic_hwif *hwif = hwdev->hwif; struct pci_dev *pdev = hwif->pdev; u16 out_size = sizeof(lro_cfg); int err; @@ -568,6 +569,9 @@ int hinic_set_rx_lro_state(struct hinic_dev *nic_dev, u8 lro_en, if (err) return err; + if (HINIC_IS_VF(nic_dev->hwdev->hwif)) + return 0; + err = hinic_set_rx_lro_timer(nic_dev, lro_timer); if (err) return err; @@ -741,9 +745,9 @@ int hinic_get_rss_type(struct hinic_dev *nic_dev, u32 tmpl_idx, { struct hinic_rss_context_table ctx_tbl = { 0 }; struct hinic_hwdev *hwdev = nic_dev->hwdev; + u16 out_size = sizeof(ctx_tbl); struct hinic_hwif *hwif; struct pci_dev *pdev; - u16 out_size = sizeof(ctx_tbl); int err; if (!hwdev || !rss_type) @@ -784,7 +788,7 @@ int hinic_rss_set_template_tbl(struct hinic_dev *nic_dev, u32 template_id, struct hinic_hwif *hwif = hwdev->hwif; struct hinic_rss_key rss_key = { 0 }; struct pci_dev *pdev = hwif->pdev; - u16 out_size; + u16 out_size = sizeof(rss_key); int err; rss_key.func_id = HINIC_HWIF_FUNC_IDX(hwif); @@ -809,9 +813,9 @@ int hinic_rss_get_template_tbl(struct hinic_dev *nic_dev, u32 tmpl_idx, { struct hinic_rss_template_key temp_key = { 0 }; struct hinic_hwdev *hwdev = nic_dev->hwdev; + u16 out_size = sizeof(temp_key); struct hinic_hwif *hwif; struct pci_dev *pdev; - u16 out_size = sizeof(temp_key); int err; if (!hwdev || !temp) @@ -844,7 +848,7 @@ int hinic_rss_set_hash_engine(struct hinic_dev *nic_dev, u8 template_id, struct hinic_hwdev *hwdev = nic_dev->hwdev; struct hinic_hwif *hwif = hwdev->hwif; struct pci_dev *pdev = hwif->pdev; - u16 out_size; + u16 out_size = sizeof(rss_engine); int err; rss_engine.func_id = HINIC_HWIF_FUNC_IDX(hwif); @@ -868,9 +872,9 @@ int hinic_rss_get_hash_engine(struct hinic_dev *nic_dev, u8 tmpl_idx, u8 *type) { struct hinic_rss_engine_type hash_type = { 0 }; struct hinic_hwdev *hwdev = nic_dev->hwdev; + u16 out_size = sizeof(hash_type); struct hinic_hwif *hwif; struct pci_dev *pdev; - u16 out_size = sizeof(hash_type); int err; if (!hwdev || !type) @@ -901,7 +905,7 @@ int hinic_rss_cfg(struct hinic_dev *nic_dev, u8 rss_en, u8 template_id) struct hinic_rss_config rss_cfg = { 0 }; struct hinic_hwif *hwif = hwdev->hwif; struct pci_dev *pdev = hwif->pdev; - u16 out_size; + u16 out_size = sizeof(rss_cfg); int err; rss_cfg.func_id = HINIC_HWIF_FUNC_IDX(hwif); @@ -927,8 +931,8 @@ int hinic_rss_template_alloc(struct hinic_dev *nic_dev, u8 *tmpl_idx) struct hinic_rss_template_mgmt template_mgmt = { 0 }; struct hinic_hwdev *hwdev = nic_dev->hwdev; struct hinic_hwif *hwif = hwdev->hwif; + u16 out_size = sizeof(template_mgmt); struct pci_dev *pdev = hwif->pdev; - u16 out_size; int err; template_mgmt.func_id = HINIC_HWIF_FUNC_IDX(hwif); @@ -953,8 +957,8 @@ int hinic_rss_template_free(struct hinic_dev *nic_dev, u8 tmpl_idx) struct hinic_rss_template_mgmt template_mgmt = { 0 }; struct hinic_hwdev *hwdev = nic_dev->hwdev; struct hinic_hwif *hwif = hwdev->hwif; + u16 out_size = sizeof(template_mgmt); struct pci_dev *pdev = hwif->pdev; - u16 out_size; int err; template_mgmt.func_id = HINIC_HWIF_FUNC_IDX(hwif); @@ -1043,9 +1047,9 @@ int hinic_get_mgmt_version(struct hinic_dev *nic_dev, u8 *mgmt_ver) { struct hinic_hwdev *hwdev = nic_dev->hwdev; struct hinic_version_info up_ver = {0}; + u16 out_size = sizeof(up_ver); struct hinic_hwif *hwif; struct pci_dev *pdev; - u16 out_size; int err; if (!hwdev) @@ -1068,3 +1072,132 @@ int hinic_get_mgmt_version(struct hinic_dev *nic_dev, u8 *mgmt_ver) return 0; } + +int hinic_get_link_mode(struct hinic_hwdev *hwdev, + struct hinic_link_mode_cmd *link_mode) +{ + u16 out_size; + int err; + + if (!hwdev || !link_mode) + return -EINVAL; + + out_size = sizeof(*link_mode); + + err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_LINK_MODE, + link_mode, sizeof(*link_mode), + link_mode, &out_size); + if (err || !out_size || link_mode->status) { + dev_err(&hwdev->hwif->pdev->dev, + "Failed to get link mode, err: %d, status: 0x%x, out size: 0x%x\n", + err, link_mode->status, out_size); + return -EIO; + } + + return 0; +} + +int hinic_set_autoneg(struct hinic_hwdev *hwdev, bool enable) +{ + struct hinic_set_autoneg_cmd autoneg = {0}; + u16 out_size = sizeof(autoneg); + int err; + + if (!hwdev) + return -EINVAL; + + autoneg.func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif); + autoneg.enable = enable; + + err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_AUTONEG, + &autoneg, sizeof(autoneg), + &autoneg, &out_size); + if (err || !out_size || autoneg.status) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to %s autoneg, err: %d, status: 0x%x, out size: 0x%x\n", + enable ? "enable" : "disable", err, autoneg.status, + out_size); + return -EIO; + } + + return 0; +} + +int hinic_set_speed(struct hinic_hwdev *hwdev, enum nic_speed_level speed) +{ + struct hinic_speed_cmd speed_info = {0}; + u16 out_size = sizeof(speed_info); + int err; + + if (!hwdev) + return -EINVAL; + + speed_info.func_id = HINIC_HWIF_FUNC_IDX(hwdev->hwif); + speed_info.speed = speed; + + err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_SPEED, + &speed_info, sizeof(speed_info), + &speed_info, &out_size); + if (err || !out_size || speed_info.status) { + dev_err(&hwdev->hwif->pdev->dev, + "Failed to set speed, err: %d, status: 0x%x, out size: 0x%x\n", + err, speed_info.status, out_size); + return -EIO; + } + + return 0; +} + +int hinic_set_link_settings(struct hinic_hwdev *hwdev, + struct hinic_link_ksettings_info *info) +{ + u16 out_size = sizeof(*info); + int err; + + err = hinic_hilink_msg_cmd(hwdev, HINIC_HILINK_CMD_SET_LINK_SETTINGS, + info, sizeof(*info), info, &out_size); + if ((info->status != HINIC_MGMT_CMD_UNSUPPORTED && + info->status) || err || !out_size) { + dev_err(&hwdev->hwif->pdev->dev, + "Failed to set link settings, err: %d, status: 0x%x, out size: 0x%x\n", + err, info->status, out_size); + return -EFAULT; + } + + return info->status; +} + +int hinic_get_hw_pause_info(struct hinic_hwdev *hwdev, + struct hinic_pause_config *pause_info) +{ + u16 out_size = sizeof(*pause_info); + int err; + + err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_GET_PAUSE_INFO, + pause_info, sizeof(*pause_info), + pause_info, &out_size); + if (err || !out_size || pause_info->status) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to get pause info, err: %d, status: 0x%x, out size: 0x%x\n", + err, pause_info->status, out_size); + return -EIO; + } + + return 0; +} + +int hinic_set_hw_pause_info(struct hinic_hwdev *hwdev, + struct hinic_pause_config *pause_info) +{ + u16 out_size = sizeof(*pause_info); + int err; + + err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_PAUSE_INFO, + pause_info, sizeof(*pause_info), + pause_info, &out_size); + if (err || !out_size || pause_info->status) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to set pause info, err: %d, status: 0x%x, out size: 0x%x\n", + err, pause_info->status, out_size); + return -EIO; + } + + return 0; +} diff --git a/drivers/net/ethernet/huawei/hinic/hinic_port.h b/drivers/net/ethernet/huawei/hinic/hinic_port.h index 44772fd47fc1..661c6322dc15 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_port.h +++ b/drivers/net/ethernet/huawei/hinic/hinic_port.h @@ -79,6 +79,42 @@ enum hinic_speed { HINIC_SPEED_UNKNOWN = 0xFF, }; +enum hinic_link_mode { + HINIC_10GE_BASE_KR = 0, + HINIC_40GE_BASE_KR4 = 1, + HINIC_40GE_BASE_CR4 = 2, + HINIC_100GE_BASE_KR4 = 3, + HINIC_100GE_BASE_CR4 = 4, + HINIC_25GE_BASE_KR_S = 5, + HINIC_25GE_BASE_CR_S = 6, + HINIC_25GE_BASE_KR = 7, + HINIC_25GE_BASE_CR = 8, + HINIC_GE_BASE_KX = 9, + HINIC_LINK_MODE_NUMBERS, + + HINIC_SUPPORTED_UNKNOWN = 0xFFFF, +}; + +enum hinic_port_type { + HINIC_PORT_TP, /* BASET */ + HINIC_PORT_AUI, + HINIC_PORT_MII, + HINIC_PORT_FIBRE, /* OPTICAL */ + HINIC_PORT_BNC, + HINIC_PORT_ELEC, + HINIC_PORT_COPPER, /* PORT_DA */ + HINIC_PORT_AOC, + HINIC_PORT_BACKPLANE, + HINIC_PORT_NONE = 0xEF, + HINIC_PORT_OTHER = 0xFF, +}; + +enum hinic_valid_link_settings { + HILINK_LINK_SET_SPEED = 0x1, + HILINK_LINK_SET_AUTONEG = 0x2, + HILINK_LINK_SET_FEC = 0x4, +}; + enum hinic_tso_state { HINIC_TSO_DISABLE = 0, HINIC_TSO_ENABLE = 1, @@ -148,9 +184,9 @@ struct hinic_port_link_status { u8 version; u8 rsvd0[6]; - u16 rsvd1; + u16 func_id; u8 link; - u8 rsvd2; + u8 port_id; }; struct hinic_port_func_state_cmd { @@ -179,6 +215,50 @@ struct hinic_port_cap { u8 rsvd2[3]; }; +struct hinic_link_mode_cmd { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 rsvd1; + u16 supported; /* 0xFFFF represents invalid value */ + u16 advertised; +}; + +struct hinic_speed_cmd { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 speed; +}; + +struct hinic_set_autoneg_cmd { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 enable; /* 1: enable , 0: disable */ +}; + +struct hinic_link_ksettings_info { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 rsvd1; + + u32 valid_bitmap; + u32 speed; /* enum nic_speed_level */ + u8 autoneg; /* 0 - off; 1 - on */ + u8 fec; /* 0 - RSFEC; 1 - BASEFEC; 2 - NOFEC */ + u8 rsvd2[18]; /* reserved for duplex, port, etc. */ +}; + struct hinic_tso_config { u8 status; u8 version; @@ -506,6 +586,61 @@ struct hinic_cmd_vport_stats { struct hinic_vport_stats stats; }; +struct hinic_tx_rate_cfg_max_min { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 rsvd1; + u32 min_rate; + u32 max_rate; + u8 rsvd2[8]; +}; + +struct hinic_tx_rate_cfg { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 rsvd1; + u32 tx_rate; +}; + +enum nic_speed_level { + LINK_SPEED_10MB = 0, + LINK_SPEED_100MB, + LINK_SPEED_1GB, + LINK_SPEED_10GB, + LINK_SPEED_25GB, + LINK_SPEED_40GB, + LINK_SPEED_100GB, + LINK_SPEED_LEVELS, +}; + +struct hinic_spoofchk_set { + u8 status; + u8 version; + u8 rsvd0[6]; + + u8 state; + u8 rsvd1; + u16 func_id; +}; + +struct hinic_pause_config { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 rsvd1; + u32 auto_neg; + u32 rx_pause; + u32 tx_pause; +}; + int hinic_port_add_mac(struct hinic_dev *nic_dev, const u8 *addr, u16 vlan_id); @@ -585,4 +720,24 @@ int hinic_set_rx_vlan_offload(struct hinic_dev *nic_dev, u8 en); int hinic_get_mgmt_version(struct hinic_dev *nic_dev, u8 *mgmt_ver); +int hinic_set_link_settings(struct hinic_hwdev *hwdev, + struct hinic_link_ksettings_info *info); + +int hinic_get_link_mode(struct hinic_hwdev *hwdev, + struct hinic_link_mode_cmd *link_mode); + +int hinic_set_autoneg(struct hinic_hwdev *hwdev, bool enable); + +int hinic_set_speed(struct hinic_hwdev *hwdev, enum nic_speed_level speed); + +int hinic_get_hw_pause_info(struct hinic_hwdev *hwdev, + struct hinic_pause_config *pause_info); + +int hinic_set_hw_pause_info(struct hinic_hwdev *hwdev, + struct hinic_pause_config *pause_info); + +int hinic_open(struct net_device *netdev); + +int hinic_close(struct net_device *netdev); + #endif diff --git a/drivers/net/ethernet/huawei/hinic/hinic_rx.c b/drivers/net/ethernet/huawei/hinic/hinic_rx.c index 815649e37cb1..af20d0dd6de7 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_rx.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_rx.c @@ -432,9 +432,11 @@ static int rx_poll(struct napi_struct *napi, int budget) return budget; napi_complete(napi); - hinic_hwdev_set_msix_state(nic_dev->hwdev, - rq->msix_entry, - HINIC_MSIX_ENABLE); + + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) + hinic_hwdev_set_msix_state(nic_dev->hwdev, + rq->msix_entry, + HINIC_MSIX_ENABLE); return pkts; } @@ -461,9 +463,10 @@ static irqreturn_t rx_irq(int irq, void *data) /* Disable the interrupt until napi will be completed */ nic_dev = netdev_priv(rxq->netdev); - hinic_hwdev_set_msix_state(nic_dev->hwdev, - rq->msix_entry, - HINIC_MSIX_DISABLE); + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) + hinic_hwdev_set_msix_state(nic_dev->hwdev, + rq->msix_entry, + HINIC_MSIX_DISABLE); nic_dev = netdev_priv(rxq->netdev); hinic_hwdev_msix_cnt_set(nic_dev->hwdev, rq->msix_entry); diff --git a/drivers/net/ethernet/huawei/hinic/hinic_sriov.c b/drivers/net/ethernet/huawei/hinic/hinic_sriov.c new file mode 100644 index 000000000000..efab2dd2c889 --- /dev/null +++ b/drivers/net/ethernet/huawei/hinic/hinic_sriov.c @@ -0,0 +1,1294 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Huawei HiNIC PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + */ + +#include <linux/pci.h> +#include <linux/if_vlan.h> +#include <linux/interrupt.h> +#include <linux/etherdevice.h> +#include <linux/netdevice.h> + +#include "hinic_hw_dev.h" +#include "hinic_dev.h" +#include "hinic_hw_mbox.h" +#include "hinic_hw_cmdq.h" +#include "hinic_port.h" +#include "hinic_sriov.h" + +static unsigned char set_vf_link_state; +module_param(set_vf_link_state, byte, 0444); +MODULE_PARM_DESC(set_vf_link_state, "Set vf link state, 0 represents link auto, 1 represents link always up, 2 represents link always down. - default is 0."); + +#define HINIC_VLAN_PRIORITY_SHIFT 13 +#define HINIC_ADD_VLAN_IN_MAC 0x8000 +#define HINIC_TX_RATE_TABLE_FULL 12 + +static int hinic_set_mac(struct hinic_hwdev *hwdev, const u8 *mac_addr, + u16 vlan_id, u16 func_id) +{ + struct hinic_port_mac_cmd mac_info = {0}; + u16 out_size = sizeof(mac_info); + int err; + + mac_info.func_idx = func_id; + mac_info.vlan_id = vlan_id; + memcpy(mac_info.mac, mac_addr, ETH_ALEN); + + err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_MAC, &mac_info, + sizeof(mac_info), &mac_info, &out_size); + if (err || out_size != sizeof(mac_info) || + (mac_info.status && mac_info.status != HINIC_PF_SET_VF_ALREADY && + mac_info.status != HINIC_MGMT_STATUS_EXIST)) { + dev_err(&hwdev->func_to_io.hwif->pdev->dev, "Failed to change MAC, ret = %d\n", + mac_info.status); + return -EFAULT; + } + + return 0; +} + +static void hinic_notify_vf_link_status(struct hinic_hwdev *hwdev, u16 vf_id, + u8 link_status) +{ + struct vf_data_storage *vf_infos = hwdev->func_to_io.vf_infos; + struct hinic_port_link_status link = {0}; + u16 out_size = sizeof(link); + int err; + + if (vf_infos[HW_VF_ID_TO_OS(vf_id)].registered) { + link.link = link_status; + link.func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id; + err = hinic_mbox_to_vf(hwdev, HINIC_MOD_L2NIC, + vf_id, HINIC_PORT_CMD_LINK_STATUS_REPORT, + &link, sizeof(link), + &link, &out_size, 0); + if (err || !out_size || link.status) + dev_err(&hwdev->hwif->pdev->dev, + "Send link change event to VF %d failed, err: %d, status: 0x%x, out_size: 0x%x\n", + HW_VF_ID_TO_OS(vf_id), err, + link.status, out_size); + } +} + +/* send link change event mbox msg to active vfs under the pf */ +void hinic_notify_all_vfs_link_changed(struct hinic_hwdev *hwdev, + u8 link_status) +{ + struct hinic_func_to_io *nic_io = &hwdev->func_to_io; + u16 i; + + nic_io->link_status = link_status; + for (i = 1; i <= nic_io->max_vfs; i++) { + if (!nic_io->vf_infos[HW_VF_ID_TO_OS(i)].link_forced) + hinic_notify_vf_link_status(hwdev, i, link_status); + } +} + +static u16 hinic_vf_info_vlanprio(struct hinic_hwdev *hwdev, int vf_id) +{ + struct hinic_func_to_io *nic_io = &hwdev->func_to_io; + u16 pf_vlan, vlanprio; + u8 pf_qos; + + pf_vlan = nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_vlan; + pf_qos = nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_qos; + vlanprio = pf_vlan | pf_qos << HINIC_VLAN_PRIORITY_SHIFT; + + return vlanprio; +} + +static int hinic_set_vf_vlan(struct hinic_hwdev *hwdev, bool add, u16 vid, + u8 qos, int vf_id) +{ + struct hinic_vf_vlan_config vf_vlan = {0}; + u16 out_size = sizeof(vf_vlan); + int err; + u8 cmd; + + /* VLAN 0 is a special case, don't allow it to be removed */ + if (!vid && !add) + return 0; + + vf_vlan.func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id; + vf_vlan.vlan_id = vid; + vf_vlan.qos = qos; + + if (add) + cmd = HINIC_PORT_CMD_SET_VF_VLAN; + else + cmd = HINIC_PORT_CMD_CLR_VF_VLAN; + + err = hinic_port_msg_cmd(hwdev, cmd, &vf_vlan, + sizeof(vf_vlan), &vf_vlan, &out_size); + if (err || !out_size || vf_vlan.status) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to set VF %d vlan, err: %d, status: 0x%x, out size: 0x%x\n", + HW_VF_ID_TO_OS(vf_id), err, vf_vlan.status, out_size); + return -EFAULT; + } + + return 0; +} + +static int hinic_set_vf_tx_rate_max_min(struct hinic_hwdev *hwdev, u16 vf_id, + u32 max_rate, u32 min_rate) +{ + struct hinic_func_to_io *nic_io = &hwdev->func_to_io; + struct hinic_tx_rate_cfg_max_min rate_cfg = {0}; + u16 out_size = sizeof(rate_cfg); + int err; + + rate_cfg.func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id; + rate_cfg.max_rate = max_rate; + rate_cfg.min_rate = min_rate; + err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_VF_MAX_MIN_RATE, + &rate_cfg, sizeof(rate_cfg), &rate_cfg, + &out_size); + if ((rate_cfg.status != HINIC_MGMT_CMD_UNSUPPORTED && + rate_cfg.status) || err || !out_size) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to set VF(%d) max rate(%d), min rate(%d), err: %d, status: 0x%x, out size: 0x%x\n", + HW_VF_ID_TO_OS(vf_id), max_rate, min_rate, err, + rate_cfg.status, out_size); + return -EIO; + } + + if (!rate_cfg.status) { + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].max_rate = max_rate; + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].min_rate = min_rate; + } + + return rate_cfg.status; +} + +static int hinic_set_vf_rate_limit(struct hinic_hwdev *hwdev, u16 vf_id, + u32 tx_rate) +{ + struct hinic_func_to_io *nic_io = &hwdev->func_to_io; + struct hinic_tx_rate_cfg rate_cfg = {0}; + u16 out_size = sizeof(rate_cfg); + int err; + + rate_cfg.func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id; + rate_cfg.tx_rate = tx_rate; + err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_SET_VF_RATE, + &rate_cfg, sizeof(rate_cfg), &rate_cfg, + &out_size); + if (err || !out_size || rate_cfg.status) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to set VF(%d) rate(%d), err: %d, status: 0x%x, out size: 0x%x\n", + HW_VF_ID_TO_OS(vf_id), tx_rate, err, rate_cfg.status, + out_size); + if (rate_cfg.status) + return rate_cfg.status; + + return -EIO; + } + + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].max_rate = tx_rate; + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].min_rate = 0; + + return 0; +} + +static int hinic_set_vf_tx_rate(struct hinic_hwdev *hwdev, u16 vf_id, + u32 max_rate, u32 min_rate) +{ + int err; + + err = hinic_set_vf_tx_rate_max_min(hwdev, vf_id, max_rate, min_rate); + if (err != HINIC_MGMT_CMD_UNSUPPORTED) + return err; + + if (min_rate) { + dev_err(&hwdev->hwif->pdev->dev, "Current firmware doesn't support to set min tx rate\n"); + return -EOPNOTSUPP; + } + + dev_info(&hwdev->hwif->pdev->dev, "Current firmware doesn't support to set min tx rate, force min_tx_rate = max_tx_rate\n"); + + return hinic_set_vf_rate_limit(hwdev, vf_id, max_rate); +} + +static int hinic_init_vf_config(struct hinic_hwdev *hwdev, u16 vf_id) +{ + struct vf_data_storage *vf_info; + u16 func_id, vlan_id; + int err = 0; + + vf_info = hwdev->func_to_io.vf_infos + HW_VF_ID_TO_OS(vf_id); + if (vf_info->pf_set_mac) { + func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id; + + vlan_id = 0; + + err = hinic_set_mac(hwdev, vf_info->vf_mac_addr, vlan_id, + func_id); + if (err) { + dev_err(&hwdev->func_to_io.hwif->pdev->dev, "Failed to set VF %d MAC\n", + HW_VF_ID_TO_OS(vf_id)); + return err; + } + } + + if (hinic_vf_info_vlanprio(hwdev, vf_id)) { + err = hinic_set_vf_vlan(hwdev, true, vf_info->pf_vlan, + vf_info->pf_qos, vf_id); + if (err) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to add VF %d VLAN_QOS\n", + HW_VF_ID_TO_OS(vf_id)); + return err; + } + } + + if (vf_info->max_rate) { + err = hinic_set_vf_tx_rate(hwdev, vf_id, vf_info->max_rate, + vf_info->min_rate); + if (err) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to set VF %d max rate: %d, min rate: %d\n", + HW_VF_ID_TO_OS(vf_id), vf_info->max_rate, + vf_info->min_rate); + return err; + } + } + + return 0; +} + +static int hinic_register_vf_msg_handler(void *hwdev, u16 vf_id, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hinic_register_vf *register_info = buf_out; + struct hinic_hwdev *hw_dev = hwdev; + struct hinic_func_to_io *nic_io; + int err; + + nic_io = &hw_dev->func_to_io; + if (vf_id > nic_io->max_vfs) { + dev_err(&hw_dev->hwif->pdev->dev, "Register VF id %d exceed limit[0-%d]\n", + HW_VF_ID_TO_OS(vf_id), HW_VF_ID_TO_OS(nic_io->max_vfs)); + register_info->status = EFAULT; + return -EFAULT; + } + + *out_size = sizeof(*register_info); + err = hinic_init_vf_config(hw_dev, vf_id); + if (err) { + register_info->status = EFAULT; + return err; + } + + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].registered = true; + + return 0; +} + +static int hinic_unregister_vf_msg_handler(void *hwdev, u16 vf_id, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hinic_hwdev *hw_dev = hwdev; + struct hinic_func_to_io *nic_io; + + nic_io = &hw_dev->func_to_io; + *out_size = 0; + if (vf_id > nic_io->max_vfs) + return 0; + + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].registered = false; + + return 0; +} + +static int hinic_change_vf_mtu_msg_handler(void *hwdev, u16 vf_id, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hinic_hwdev *hw_dev = hwdev; + int err; + + err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_CHANGE_MTU, buf_in, + in_size, buf_out, out_size); + if (err) { + dev_err(&hw_dev->hwif->pdev->dev, "Failed to set VF %u mtu\n", + vf_id); + return err; + } + + return 0; +} + +static int hinic_get_vf_mac_msg_handler(void *hwdev, u16 vf_id, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hinic_port_mac_cmd *mac_info = buf_out; + struct hinic_hwdev *dev = hwdev; + struct hinic_func_to_io *nic_io; + struct vf_data_storage *vf_info; + + nic_io = &dev->func_to_io; + vf_info = nic_io->vf_infos + HW_VF_ID_TO_OS(vf_id); + + memcpy(mac_info->mac, vf_info->vf_mac_addr, ETH_ALEN); + mac_info->status = 0; + *out_size = sizeof(*mac_info); + + return 0; +} + +static int hinic_set_vf_mac_msg_handler(void *hwdev, u16 vf_id, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hinic_port_mac_cmd *mac_out = buf_out; + struct hinic_port_mac_cmd *mac_in = buf_in; + struct hinic_hwdev *hw_dev = hwdev; + struct hinic_func_to_io *nic_io; + struct vf_data_storage *vf_info; + int err; + + nic_io = &hw_dev->func_to_io; + vf_info = nic_io->vf_infos + HW_VF_ID_TO_OS(vf_id); + if (vf_info->pf_set_mac && !(vf_info->trust) && + is_valid_ether_addr(mac_in->mac)) { + dev_warn(&hw_dev->hwif->pdev->dev, "PF has already set VF %d MAC address\n", + HW_VF_ID_TO_OS(vf_id)); + mac_out->status = HINIC_PF_SET_VF_ALREADY; + *out_size = sizeof(*mac_out); + return 0; + } + + err = hinic_port_msg_cmd(hw_dev, HINIC_PORT_CMD_SET_MAC, buf_in, + in_size, buf_out, out_size); + if ((err && err != HINIC_MBOX_PF_BUSY_ACTIVE_FW) || !(*out_size)) { + dev_err(&hw_dev->hwif->pdev->dev, + "Failed to set VF %d MAC address, err: %d, status: 0x%x, out size: 0x%x\n", + HW_VF_ID_TO_OS(vf_id), err, mac_out->status, *out_size); + return -EFAULT; + } + + return err; +} + +static int hinic_del_vf_mac_msg_handler(void *hwdev, u16 vf_id, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hinic_port_mac_cmd *mac_out = buf_out; + struct hinic_port_mac_cmd *mac_in = buf_in; + struct hinic_hwdev *hw_dev = hwdev; + struct hinic_func_to_io *nic_io; + struct vf_data_storage *vf_info; + int err; + + nic_io = &hw_dev->func_to_io; + vf_info = nic_io->vf_infos + HW_VF_ID_TO_OS(vf_id); + if (vf_info->pf_set_mac && is_valid_ether_addr(mac_in->mac) && + !memcmp(vf_info->vf_mac_addr, mac_in->mac, ETH_ALEN)) { + dev_warn(&hw_dev->hwif->pdev->dev, "PF has already set VF mac.\n"); + mac_out->status = HINIC_PF_SET_VF_ALREADY; + *out_size = sizeof(*mac_out); + return 0; + } + + err = hinic_port_msg_cmd(hw_dev, HINIC_PORT_CMD_DEL_MAC, buf_in, + in_size, buf_out, out_size); + if ((err && err != HINIC_MBOX_PF_BUSY_ACTIVE_FW) || !(*out_size)) { + dev_err(&hw_dev->hwif->pdev->dev, "Failed to delete VF %d MAC, err: %d, status: 0x%x, out size: 0x%x\n", + HW_VF_ID_TO_OS(vf_id), err, mac_out->status, *out_size); + return -EFAULT; + } + + return err; +} + +static int hinic_get_vf_link_status_msg_handler(void *hwdev, u16 vf_id, + void *buf_in, u16 in_size, + void *buf_out, u16 *out_size) +{ + struct hinic_port_link_cmd *get_link = buf_out; + struct hinic_hwdev *hw_dev = hwdev; + struct vf_data_storage *vf_infos; + struct hinic_func_to_io *nic_io; + bool link_forced, link_up; + + nic_io = &hw_dev->func_to_io; + vf_infos = nic_io->vf_infos; + link_forced = vf_infos[HW_VF_ID_TO_OS(vf_id)].link_forced; + link_up = vf_infos[HW_VF_ID_TO_OS(vf_id)].link_up; + + if (link_forced) + get_link->state = link_up ? + HINIC_LINK_STATE_UP : HINIC_LINK_STATE_DOWN; + else + get_link->state = nic_io->link_status; + + get_link->status = 0; + *out_size = sizeof(*get_link); + + return 0; +} + +static struct vf_cmd_msg_handle nic_vf_cmd_msg_handler[] = { + {HINIC_PORT_CMD_VF_REGISTER, hinic_register_vf_msg_handler}, + {HINIC_PORT_CMD_VF_UNREGISTER, hinic_unregister_vf_msg_handler}, + {HINIC_PORT_CMD_CHANGE_MTU, hinic_change_vf_mtu_msg_handler}, + {HINIC_PORT_CMD_GET_MAC, hinic_get_vf_mac_msg_handler}, + {HINIC_PORT_CMD_SET_MAC, hinic_set_vf_mac_msg_handler}, + {HINIC_PORT_CMD_DEL_MAC, hinic_del_vf_mac_msg_handler}, + {HINIC_PORT_CMD_GET_LINK_STATE, hinic_get_vf_link_status_msg_handler}, +}; + +#define CHECK_IPSU_15BIT 0X8000 + +static +struct hinic_sriov_info *hinic_get_sriov_info_by_pcidev(struct pci_dev *pdev) +{ + struct net_device *netdev = pci_get_drvdata(pdev); + struct hinic_dev *nic_dev = netdev_priv(netdev); + + return &nic_dev->sriov_info; +} + +static int hinic_check_mac_info(u8 status, u16 vlan_id) +{ + if ((status && status != HINIC_MGMT_STATUS_EXIST && + status != HINIC_PF_SET_VF_ALREADY) || + (vlan_id & CHECK_IPSU_15BIT && + status == HINIC_MGMT_STATUS_EXIST)) + return -EINVAL; + + return 0; +} + +#define HINIC_VLAN_ID_MASK 0x7FFF + +static int hinic_update_mac(struct hinic_hwdev *hwdev, u8 *old_mac, + u8 *new_mac, u16 vlan_id, u16 func_id) +{ + struct hinic_port_mac_update mac_info = {0}; + u16 out_size = sizeof(mac_info); + int err; + + if (!hwdev || !old_mac || !new_mac) + return -EINVAL; + + if ((vlan_id & HINIC_VLAN_ID_MASK) >= VLAN_N_VID) { + dev_err(&hwdev->hwif->pdev->dev, "Invalid VLAN number: %d\n", + (vlan_id & HINIC_VLAN_ID_MASK)); + return -EINVAL; + } + + mac_info.func_id = func_id; + mac_info.vlan_id = vlan_id; + memcpy(mac_info.old_mac, old_mac, ETH_ALEN); + memcpy(mac_info.new_mac, new_mac, ETH_ALEN); + + err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_UPDATE_MAC, &mac_info, + sizeof(mac_info), &mac_info, &out_size); + + if (err || !out_size || + hinic_check_mac_info(mac_info.status, mac_info.vlan_id)) { + dev_err(&hwdev->hwif->pdev->dev, + "Failed to update MAC, err: %d, status: 0x%x, out size: 0x%x\n", + err, mac_info.status, out_size); + return -EINVAL; + } + + if (mac_info.status == HINIC_PF_SET_VF_ALREADY) { + dev_warn(&hwdev->hwif->pdev->dev, + "PF has already set VF MAC. Ignore update operation\n"); + return HINIC_PF_SET_VF_ALREADY; + } + + if (mac_info.status == HINIC_MGMT_STATUS_EXIST) + dev_warn(&hwdev->hwif->pdev->dev, "MAC is repeated. Ignore update operation\n"); + + return 0; +} + +static void hinic_get_vf_config(struct hinic_hwdev *hwdev, u16 vf_id, + struct ifla_vf_info *ivi) +{ + struct vf_data_storage *vfinfo; + + vfinfo = hwdev->func_to_io.vf_infos + HW_VF_ID_TO_OS(vf_id); + + ivi->vf = HW_VF_ID_TO_OS(vf_id); + memcpy(ivi->mac, vfinfo->vf_mac_addr, ETH_ALEN); + ivi->vlan = vfinfo->pf_vlan; + ivi->qos = vfinfo->pf_qos; + ivi->spoofchk = vfinfo->spoofchk; + ivi->trusted = vfinfo->trust; + ivi->max_tx_rate = vfinfo->max_rate; + ivi->min_tx_rate = vfinfo->min_rate; + + if (!vfinfo->link_forced) + ivi->linkstate = IFLA_VF_LINK_STATE_AUTO; + else if (vfinfo->link_up) + ivi->linkstate = IFLA_VF_LINK_STATE_ENABLE; + else + ivi->linkstate = IFLA_VF_LINK_STATE_DISABLE; +} + +int hinic_ndo_get_vf_config(struct net_device *netdev, + int vf, struct ifla_vf_info *ivi) +{ + struct hinic_dev *nic_dev = netdev_priv(netdev); + struct hinic_sriov_info *sriov_info; + + sriov_info = &nic_dev->sriov_info; + if (vf >= sriov_info->num_vfs) + return -EINVAL; + + hinic_get_vf_config(sriov_info->hwdev, OS_VF_ID_TO_HW(vf), ivi); + + return 0; +} + +static int hinic_set_vf_mac(struct hinic_hwdev *hwdev, int vf, + unsigned char *mac_addr) +{ + struct hinic_func_to_io *nic_io = &hwdev->func_to_io; + struct vf_data_storage *vf_info; + u16 func_id; + int err; + + vf_info = nic_io->vf_infos + HW_VF_ID_TO_OS(vf); + + /* duplicate request, so just return success */ + if (vf_info->pf_set_mac && + !memcmp(vf_info->vf_mac_addr, mac_addr, ETH_ALEN)) + return 0; + + vf_info->pf_set_mac = true; + + func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf; + err = hinic_update_mac(hwdev, vf_info->vf_mac_addr, + mac_addr, 0, func_id); + if (err) { + vf_info->pf_set_mac = false; + return err; + } + + memcpy(vf_info->vf_mac_addr, mac_addr, ETH_ALEN); + + return 0; +} + +int hinic_ndo_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) +{ + struct hinic_dev *nic_dev = netdev_priv(netdev); + struct hinic_sriov_info *sriov_info; + int err; + + sriov_info = &nic_dev->sriov_info; + if (!is_valid_ether_addr(mac) || vf >= sriov_info->num_vfs) + return -EINVAL; + + err = hinic_set_vf_mac(sriov_info->hwdev, OS_VF_ID_TO_HW(vf), mac); + if (err) + return err; + + netif_info(nic_dev, drv, netdev, "Setting MAC %pM on VF %d\n", mac, vf); + netif_info(nic_dev, drv, netdev, "Reload the VF driver to make this change effective."); + + return 0; +} + +static int hinic_add_vf_vlan(struct hinic_hwdev *hwdev, int vf_id, + u16 vlan, u8 qos) +{ + struct hinic_func_to_io *nic_io = &hwdev->func_to_io; + int err; + + err = hinic_set_vf_vlan(hwdev, true, vlan, qos, vf_id); + if (err) + return err; + + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_vlan = vlan; + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_qos = qos; + + dev_info(&hwdev->hwif->pdev->dev, "Setting VLAN %d, QOS 0x%x on VF %d\n", + vlan, qos, HW_VF_ID_TO_OS(vf_id)); + return 0; +} + +static int hinic_kill_vf_vlan(struct hinic_hwdev *hwdev, int vf_id) +{ + struct hinic_func_to_io *nic_io = &hwdev->func_to_io; + int err; + + err = hinic_set_vf_vlan(hwdev, false, + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_vlan, + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_qos, + vf_id); + if (err) + return err; + + dev_info(&hwdev->hwif->pdev->dev, "Remove VLAN %d on VF %d\n", + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_vlan, + HW_VF_ID_TO_OS(vf_id)); + + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_vlan = 0; + nic_io->vf_infos[HW_VF_ID_TO_OS(vf_id)].pf_qos = 0; + + return 0; +} + +static int hinic_update_mac_vlan(struct hinic_dev *nic_dev, u16 old_vlan, + u16 new_vlan, int vf_id) +{ + struct vf_data_storage *vf_info; + u16 vlan_id; + int err; + + if (!nic_dev || old_vlan >= VLAN_N_VID || new_vlan >= VLAN_N_VID) + return -EINVAL; + + vf_info = nic_dev->hwdev->func_to_io.vf_infos + HW_VF_ID_TO_OS(vf_id); + if (!vf_info->pf_set_mac) + return 0; + + vlan_id = old_vlan; + if (vlan_id) + vlan_id |= HINIC_ADD_VLAN_IN_MAC; + + err = hinic_port_del_mac(nic_dev, vf_info->vf_mac_addr, vlan_id); + if (err) { + dev_err(&nic_dev->hwdev->hwif->pdev->dev, "Failed to delete VF %d MAC %pM vlan %d\n", + HW_VF_ID_TO_OS(vf_id), vf_info->vf_mac_addr, old_vlan); + return err; + } + + vlan_id = new_vlan; + if (vlan_id) + vlan_id |= HINIC_ADD_VLAN_IN_MAC; + + err = hinic_port_add_mac(nic_dev, vf_info->vf_mac_addr, vlan_id); + if (err) { + dev_err(&nic_dev->hwdev->hwif->pdev->dev, "Failed to add VF %d MAC %pM vlan %d\n", + HW_VF_ID_TO_OS(vf_id), vf_info->vf_mac_addr, new_vlan); + goto out; + } + + return 0; + +out: + vlan_id = old_vlan; + if (vlan_id) + vlan_id |= HINIC_ADD_VLAN_IN_MAC; + hinic_port_add_mac(nic_dev, vf_info->vf_mac_addr, vlan_id); + + return err; +} + +static int set_hw_vf_vlan(struct hinic_dev *nic_dev, + u16 cur_vlanprio, int vf, u16 vlan, u8 qos) +{ + u16 old_vlan = cur_vlanprio & VLAN_VID_MASK; + int err = 0; + + if (vlan || qos) { + if (cur_vlanprio) { + err = hinic_kill_vf_vlan(nic_dev->hwdev, + OS_VF_ID_TO_HW(vf)); + if (err) { + dev_err(&nic_dev->sriov_info.pdev->dev, "Failed to delete vf %d old vlan %d\n", + vf, old_vlan); + goto out; + } + } + err = hinic_add_vf_vlan(nic_dev->hwdev, + OS_VF_ID_TO_HW(vf), vlan, qos); + if (err) { + dev_err(&nic_dev->sriov_info.pdev->dev, "Failed to add vf %d new vlan %d\n", + vf, vlan); + goto out; + } + } else { + err = hinic_kill_vf_vlan(nic_dev->hwdev, OS_VF_ID_TO_HW(vf)); + if (err) { + dev_err(&nic_dev->sriov_info.pdev->dev, "Failed to delete vf %d vlan %d\n", + vf, old_vlan); + goto out; + } + } + + err = hinic_update_mac_vlan(nic_dev, old_vlan, vlan, + OS_VF_ID_TO_HW(vf)); + +out: + return err; +} + +int hinic_ndo_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos, + __be16 vlan_proto) +{ + struct hinic_dev *nic_dev = netdev_priv(netdev); + struct hinic_sriov_info *sriov_info; + u16 vlanprio, cur_vlanprio; + + sriov_info = &nic_dev->sriov_info; + if (vf >= sriov_info->num_vfs || vlan > 4095 || qos > 7) + return -EINVAL; + if (vlan_proto != htons(ETH_P_8021Q)) + return -EPROTONOSUPPORT; + vlanprio = vlan | qos << HINIC_VLAN_PRIORITY_SHIFT; + cur_vlanprio = hinic_vf_info_vlanprio(nic_dev->hwdev, + OS_VF_ID_TO_HW(vf)); + /* duplicate request, so just return success */ + if (vlanprio == cur_vlanprio) + return 0; + + return set_hw_vf_vlan(nic_dev, cur_vlanprio, vf, vlan, qos); +} + +static int hinic_set_vf_trust(struct hinic_hwdev *hwdev, u16 vf_id, + bool trust) +{ + struct vf_data_storage *vf_infos; + struct hinic_func_to_io *nic_io; + + if (!hwdev) + return -EINVAL; + + nic_io = &hwdev->func_to_io; + vf_infos = nic_io->vf_infos; + vf_infos[vf_id].trust = trust; + + return 0; +} + +int hinic_ndo_set_vf_trust(struct net_device *netdev, int vf, bool setting) +{ + struct hinic_dev *adapter = netdev_priv(netdev); + struct hinic_sriov_info *sriov_info; + struct hinic_func_to_io *nic_io; + bool cur_trust; + int err; + + sriov_info = &adapter->sriov_info; + nic_io = &adapter->hwdev->func_to_io; + + if (vf >= sriov_info->num_vfs) + return -EINVAL; + + cur_trust = nic_io->vf_infos[vf].trust; + /* same request, so just return success */ + if ((setting && cur_trust) || (!setting && !cur_trust)) + return 0; + + err = hinic_set_vf_trust(adapter->hwdev, vf, setting); + if (!err) + dev_info(&sriov_info->pdev->dev, "Set VF %d trusted %s succeed\n", + vf, setting ? "on" : "off"); + else + dev_err(&sriov_info->pdev->dev, "Failed set VF %d trusted %s\n", + vf, setting ? "on" : "off"); + + return err; +} + +int hinic_ndo_set_vf_bw(struct net_device *netdev, + int vf, int min_tx_rate, int max_tx_rate) +{ + u32 speeds[] = {SPEED_10, SPEED_100, SPEED_1000, SPEED_10000, + SPEED_25000, SPEED_40000, SPEED_100000}; + struct hinic_dev *nic_dev = netdev_priv(netdev); + struct hinic_port_cap port_cap = { 0 }; + enum hinic_port_link_state link_state; + int err; + + if (vf >= nic_dev->sriov_info.num_vfs) { + netif_err(nic_dev, drv, netdev, "VF number must be less than %d\n", + nic_dev->sriov_info.num_vfs); + return -EINVAL; + } + + if (max_tx_rate < min_tx_rate) { + netif_err(nic_dev, drv, netdev, "Max rate %d must be greater than or equal to min rate %d\n", + max_tx_rate, min_tx_rate); + return -EINVAL; + } + + err = hinic_port_link_state(nic_dev, &link_state); + if (err) { + netif_err(nic_dev, drv, netdev, + "Get link status failed when setting vf tx rate\n"); + return -EIO; + } + + if (link_state == HINIC_LINK_STATE_DOWN) { + netif_err(nic_dev, drv, netdev, + "Link status must be up when setting vf tx rate\n"); + return -EPERM; + } + + err = hinic_port_get_cap(nic_dev, &port_cap); + if (err || port_cap.speed > LINK_SPEED_100GB) + return -EIO; + + /* rate limit cannot be less than 0 and greater than link speed */ + if (max_tx_rate < 0 || max_tx_rate > speeds[port_cap.speed]) { + netif_err(nic_dev, drv, netdev, "Max tx rate must be in [0 - %d]\n", + speeds[port_cap.speed]); + return -EINVAL; + } + + err = hinic_set_vf_tx_rate(nic_dev->hwdev, OS_VF_ID_TO_HW(vf), + max_tx_rate, min_tx_rate); + if (err) { + netif_err(nic_dev, drv, netdev, + "Unable to set VF %d max rate %d min rate %d%s\n", + vf, max_tx_rate, min_tx_rate, + err == HINIC_TX_RATE_TABLE_FULL ? + ", tx rate profile is full" : ""); + return -EIO; + } + + netif_info(nic_dev, drv, netdev, + "Set VF %d max tx rate %d min tx rate %d successfully\n", + vf, max_tx_rate, min_tx_rate); + + return 0; +} + +static int hinic_set_vf_spoofchk(struct hinic_hwdev *hwdev, u16 vf_id, + bool spoofchk) +{ + struct hinic_spoofchk_set spoofchk_cfg = {0}; + struct vf_data_storage *vf_infos = NULL; + u16 out_size = sizeof(spoofchk_cfg); + int err; + + if (!hwdev) + return -EINVAL; + + vf_infos = hwdev->func_to_io.vf_infos; + + spoofchk_cfg.func_id = hinic_glb_pf_vf_offset(hwdev->hwif) + vf_id; + spoofchk_cfg.state = spoofchk ? 1 : 0; + err = hinic_port_msg_cmd(hwdev, HINIC_PORT_CMD_ENABLE_SPOOFCHK, + &spoofchk_cfg, sizeof(spoofchk_cfg), + &spoofchk_cfg, &out_size); + if (spoofchk_cfg.status == HINIC_MGMT_CMD_UNSUPPORTED) { + err = HINIC_MGMT_CMD_UNSUPPORTED; + } else if (err || !out_size || spoofchk_cfg.status) { + dev_err(&hwdev->hwif->pdev->dev, "Failed to set VF(%d) spoofchk, err: %d, status: 0x%x, out size: 0x%x\n", + HW_VF_ID_TO_OS(vf_id), err, spoofchk_cfg.status, + out_size); + err = -EIO; + } + + vf_infos[HW_VF_ID_TO_OS(vf_id)].spoofchk = spoofchk; + + return err; +} + +int hinic_ndo_set_vf_spoofchk(struct net_device *netdev, int vf, bool setting) +{ + struct hinic_dev *nic_dev = netdev_priv(netdev); + struct hinic_sriov_info *sriov_info; + bool cur_spoofchk; + int err; + + sriov_info = &nic_dev->sriov_info; + if (vf >= sriov_info->num_vfs) + return -EINVAL; + + cur_spoofchk = nic_dev->hwdev->func_to_io.vf_infos[vf].spoofchk; + + /* same request, so just return success */ + if ((setting && cur_spoofchk) || (!setting && !cur_spoofchk)) + return 0; + + err = hinic_set_vf_spoofchk(sriov_info->hwdev, + OS_VF_ID_TO_HW(vf), setting); + + if (!err) { + netif_info(nic_dev, drv, netdev, "Set VF %d spoofchk %s successfully\n", + vf, setting ? "on" : "off"); + } else if (err == HINIC_MGMT_CMD_UNSUPPORTED) { + netif_err(nic_dev, drv, netdev, + "Current firmware doesn't support to set vf spoofchk, need to upgrade latest firmware version\n"); + err = -EOPNOTSUPP; + } + + return err; +} + +static int hinic_set_vf_link_state(struct hinic_hwdev *hwdev, u16 vf_id, + int link) +{ + struct hinic_func_to_io *nic_io = &hwdev->func_to_io; + struct vf_data_storage *vf_infos = nic_io->vf_infos; + u8 link_status = 0; + + switch (link) { + case HINIC_IFLA_VF_LINK_STATE_AUTO: + vf_infos[HW_VF_ID_TO_OS(vf_id)].link_forced = false; + vf_infos[HW_VF_ID_TO_OS(vf_id)].link_up = nic_io->link_status ? + true : false; + link_status = nic_io->link_status; + break; + case HINIC_IFLA_VF_LINK_STATE_ENABLE: + vf_infos[HW_VF_ID_TO_OS(vf_id)].link_forced = true; + vf_infos[HW_VF_ID_TO_OS(vf_id)].link_up = true; + link_status = HINIC_LINK_UP; + break; + case HINIC_IFLA_VF_LINK_STATE_DISABLE: + vf_infos[HW_VF_ID_TO_OS(vf_id)].link_forced = true; + vf_infos[HW_VF_ID_TO_OS(vf_id)].link_up = false; + link_status = HINIC_LINK_DOWN; + break; + default: + return -EINVAL; + } + + /* Notify the VF of its new link state */ + hinic_notify_vf_link_status(hwdev, vf_id, link_status); + + return 0; +} + +int hinic_ndo_set_vf_link_state(struct net_device *netdev, int vf_id, int link) +{ + struct hinic_dev *nic_dev = netdev_priv(netdev); + struct hinic_sriov_info *sriov_info; + + sriov_info = &nic_dev->sriov_info; + + if (vf_id >= sriov_info->num_vfs) { + netif_err(nic_dev, drv, netdev, + "Invalid VF Identifier %d\n", vf_id); + return -EINVAL; + } + + return hinic_set_vf_link_state(sriov_info->hwdev, + OS_VF_ID_TO_HW(vf_id), link); +} + +/* pf receive message from vf */ +static int nic_pf_mbox_handler(void *hwdev, u16 vf_id, u8 cmd, void *buf_in, + u16 in_size, void *buf_out, u16 *out_size) +{ + struct vf_cmd_msg_handle *vf_msg_handle; + struct hinic_hwdev *dev = hwdev; + struct hinic_func_to_io *nic_io; + struct hinic_pfhwdev *pfhwdev; + int err = 0; + u32 i; + + if (!hwdev) + return -EFAULT; + + pfhwdev = container_of(dev, struct hinic_pfhwdev, hwdev); + nic_io = &dev->func_to_io; + for (i = 0; i < ARRAY_SIZE(nic_vf_cmd_msg_handler); i++) { + vf_msg_handle = &nic_vf_cmd_msg_handler[i]; + if (cmd == vf_msg_handle->cmd && + vf_msg_handle->cmd_msg_handler) { + err = vf_msg_handle->cmd_msg_handler(hwdev, vf_id, + buf_in, in_size, + buf_out, + out_size); + break; + } + } + if (i == ARRAY_SIZE(nic_vf_cmd_msg_handler)) + err = hinic_msg_to_mgmt(&pfhwdev->pf_to_mgmt, HINIC_MOD_L2NIC, + cmd, buf_in, in_size, buf_out, + out_size, HINIC_MGMT_MSG_SYNC); + + if (err && err != HINIC_MBOX_PF_BUSY_ACTIVE_FW) + dev_err(&nic_io->hwif->pdev->dev, "PF receive VF L2NIC cmd: %d process error, err:%d\n", + cmd, err); + return err; +} + +static int cfg_mbx_pf_proc_vf_msg(void *hwdev, u16 vf_id, u8 cmd, void *buf_in, + u16 in_size, void *buf_out, u16 *out_size) +{ + struct hinic_dev_cap *dev_cap = buf_out; + struct hinic_hwdev *dev = hwdev; + struct hinic_cap *cap; + + cap = &dev->nic_cap; + memset(dev_cap, 0, sizeof(*dev_cap)); + + dev_cap->max_vf = cap->max_vf; + dev_cap->max_sqs = cap->max_vf_qps; + dev_cap->max_rqs = cap->max_vf_qps; + + *out_size = sizeof(*dev_cap); + + return 0; +} + +static int hinic_init_vf_infos(struct hinic_func_to_io *nic_io, u16 vf_id) +{ + struct vf_data_storage *vf_infos = nic_io->vf_infos; + + if (set_vf_link_state > HINIC_IFLA_VF_LINK_STATE_DISABLE) { + dev_warn(&nic_io->hwif->pdev->dev, "Module Parameter set_vf_link_state value %d is out of range, resetting to %d\n", + set_vf_link_state, HINIC_IFLA_VF_LINK_STATE_AUTO); + set_vf_link_state = HINIC_IFLA_VF_LINK_STATE_AUTO; + } + + switch (set_vf_link_state) { + case HINIC_IFLA_VF_LINK_STATE_AUTO: + vf_infos[vf_id].link_forced = false; + break; + case HINIC_IFLA_VF_LINK_STATE_ENABLE: + vf_infos[vf_id].link_forced = true; + vf_infos[vf_id].link_up = true; + break; + case HINIC_IFLA_VF_LINK_STATE_DISABLE: + vf_infos[vf_id].link_forced = true; + vf_infos[vf_id].link_up = false; + break; + default: + dev_err(&nic_io->hwif->pdev->dev, "Invalid input parameter set_vf_link_state: %d\n", + set_vf_link_state); + return -EINVAL; + } + + return 0; +} + +static void hinic_clear_vf_infos(struct hinic_dev *nic_dev, u16 vf_id) +{ + struct vf_data_storage *vf_infos; + u16 func_id; + + func_id = hinic_glb_pf_vf_offset(nic_dev->hwdev->hwif) + vf_id; + vf_infos = nic_dev->hwdev->func_to_io.vf_infos + HW_VF_ID_TO_OS(vf_id); + if (vf_infos->pf_set_mac) + hinic_port_del_mac(nic_dev, vf_infos->vf_mac_addr, 0); + + if (hinic_vf_info_vlanprio(nic_dev->hwdev, vf_id)) + hinic_kill_vf_vlan(nic_dev->hwdev, vf_id); + + if (vf_infos->max_rate) + hinic_set_vf_tx_rate(nic_dev->hwdev, vf_id, 0, 0); + + if (vf_infos->spoofchk) + hinic_set_vf_spoofchk(nic_dev->hwdev, vf_id, false); + + if (vf_infos->trust) + hinic_set_vf_trust(nic_dev->hwdev, vf_id, false); + + memset(vf_infos, 0, sizeof(*vf_infos)); + /* set vf_infos to default */ + hinic_init_vf_infos(&nic_dev->hwdev->func_to_io, HW_VF_ID_TO_OS(vf_id)); +} + +static int hinic_deinit_vf_hw(struct hinic_sriov_info *sriov_info, + u16 start_vf_id, u16 end_vf_id) +{ + struct hinic_dev *nic_dev; + u16 func_idx, idx; + + nic_dev = container_of(sriov_info, struct hinic_dev, sriov_info); + + for (idx = start_vf_id; idx <= end_vf_id; idx++) { + func_idx = hinic_glb_pf_vf_offset(nic_dev->hwdev->hwif) + idx; + hinic_set_wq_page_size(nic_dev->hwdev, func_idx, + HINIC_HW_WQ_PAGE_SIZE); + hinic_clear_vf_infos(nic_dev, idx); + } + + return 0; +} + +int hinic_vf_func_init(struct hinic_hwdev *hwdev) +{ + struct hinic_register_vf register_info = {0}; + u16 out_size = sizeof(register_info); + struct hinic_func_to_io *nic_io; + int err = 0; + u32 size, i; + + nic_io = &hwdev->func_to_io; + + if (HINIC_IS_VF(hwdev->hwif)) { + err = hinic_mbox_to_pf(hwdev, HINIC_MOD_L2NIC, + HINIC_PORT_CMD_VF_REGISTER, + ®ister_info, sizeof(register_info), + ®ister_info, &out_size, 0); + if (err || register_info.status || !out_size) { + dev_err(&hwdev->hwif->pdev->dev, + "Failed to register VF, err: %d, status: 0x%x, out size: 0x%x\n", + err, register_info.status, out_size); + hinic_unregister_vf_mbox_cb(hwdev, HINIC_MOD_L2NIC); + return -EIO; + } + } else { + err = hinic_register_pf_mbox_cb(hwdev, HINIC_MOD_CFGM, + cfg_mbx_pf_proc_vf_msg); + if (err) { + dev_err(&hwdev->hwif->pdev->dev, + "Register PF mailbox callback failed\n"); + return err; + } + nic_io->max_vfs = hwdev->nic_cap.max_vf; + size = sizeof(*nic_io->vf_infos) * nic_io->max_vfs; + if (size != 0) { + nic_io->vf_infos = kzalloc(size, GFP_KERNEL); + if (!nic_io->vf_infos) { + err = -ENOMEM; + goto out_free_nic_io; + } + + for (i = 0; i < nic_io->max_vfs; i++) { + err = hinic_init_vf_infos(nic_io, i); + if (err) + goto err_init_vf_infos; + } + + err = hinic_register_pf_mbox_cb(hwdev, HINIC_MOD_L2NIC, + nic_pf_mbox_handler); + if (err) + goto err_register_pf_mbox_cb; + } + } + + return 0; + +err_register_pf_mbox_cb: +err_init_vf_infos: + kfree(nic_io->vf_infos); +out_free_nic_io: + return err; +} + +void hinic_vf_func_free(struct hinic_hwdev *hwdev) +{ + struct hinic_register_vf unregister = {0}; + u16 out_size = sizeof(unregister); + int err; + + if (HINIC_IS_VF(hwdev->hwif)) { + err = hinic_mbox_to_pf(hwdev, HINIC_MOD_L2NIC, + HINIC_PORT_CMD_VF_UNREGISTER, + &unregister, sizeof(unregister), + &unregister, &out_size, 0); + if (err || !out_size || unregister.status) + dev_err(&hwdev->hwif->pdev->dev, "Failed to unregister VF, err: %d, status: 0x%x, out_size: 0x%x\n", + err, unregister.status, out_size); + } else { + if (hwdev->func_to_io.vf_infos) { + hinic_unregister_pf_mbox_cb(hwdev, HINIC_MOD_L2NIC); + kfree(hwdev->func_to_io.vf_infos); + } + } +} + +static int hinic_init_vf_hw(struct hinic_hwdev *hwdev, u16 start_vf_id, + u16 end_vf_id) +{ + u16 i, func_idx; + int err; + + /* vf use 256K as default wq page size, and can't change it */ + for (i = start_vf_id; i <= end_vf_id; i++) { + func_idx = hinic_glb_pf_vf_offset(hwdev->hwif) + i; + err = hinic_set_wq_page_size(hwdev, func_idx, + HINIC_DEFAULT_WQ_PAGE_SIZE); + if (err) + return err; + } + + return 0; +} + +int hinic_pci_sriov_disable(struct pci_dev *pdev) +{ + struct hinic_sriov_info *sriov_info; + u16 tmp_vfs; + + sriov_info = hinic_get_sriov_info_by_pcidev(pdev); + /* if SR-IOV is already disabled then nothing will be done */ + if (!sriov_info->sriov_enabled) + return 0; + + set_bit(HINIC_SRIOV_DISABLE, &sriov_info->state); + + /* If our VFs are assigned we cannot shut down SR-IOV + * without causing issues, so just leave the hardware + * available but disabled + */ + if (pci_vfs_assigned(sriov_info->pdev)) { + clear_bit(HINIC_SRIOV_DISABLE, &sriov_info->state); + dev_warn(&pdev->dev, "Unloading driver while VFs are assigned - VFs will not be deallocated\n"); + return -EPERM; + } + sriov_info->sriov_enabled = false; + + /* disable iov and allow time for transactions to clear */ + pci_disable_sriov(sriov_info->pdev); + + tmp_vfs = (u16)sriov_info->num_vfs; + sriov_info->num_vfs = 0; + hinic_deinit_vf_hw(sriov_info, OS_VF_ID_TO_HW(0), + OS_VF_ID_TO_HW(tmp_vfs - 1)); + + clear_bit(HINIC_SRIOV_DISABLE, &sriov_info->state); + + return 0; +} + +int hinic_pci_sriov_enable(struct pci_dev *pdev, int num_vfs) +{ + struct hinic_sriov_info *sriov_info; + int err; + + sriov_info = hinic_get_sriov_info_by_pcidev(pdev); + + if (test_and_set_bit(HINIC_SRIOV_ENABLE, &sriov_info->state)) { + dev_err(&pdev->dev, + "SR-IOV enable in process, please wait, num_vfs %d\n", + num_vfs); + return -EPERM; + } + + err = hinic_init_vf_hw(sriov_info->hwdev, OS_VF_ID_TO_HW(0), + OS_VF_ID_TO_HW((u16)num_vfs - 1)); + if (err) { + dev_err(&sriov_info->pdev->dev, + "Failed to init vf in hardware before enable sriov, error %d\n", + err); + clear_bit(HINIC_SRIOV_ENABLE, &sriov_info->state); + return err; + } + + err = pci_enable_sriov(sriov_info->pdev, num_vfs); + if (err) { + dev_err(&pdev->dev, + "Failed to enable SR-IOV, error %d\n", err); + clear_bit(HINIC_SRIOV_ENABLE, &sriov_info->state); + return err; + } + + sriov_info->sriov_enabled = true; + sriov_info->num_vfs = num_vfs; + clear_bit(HINIC_SRIOV_ENABLE, &sriov_info->state); + + return num_vfs; +} + +int hinic_pci_sriov_configure(struct pci_dev *dev, int num_vfs) +{ + struct hinic_sriov_info *sriov_info; + + sriov_info = hinic_get_sriov_info_by_pcidev(dev); + + if (test_bit(HINIC_FUNC_REMOVE, &sriov_info->state)) + return -EBUSY; + + if (!num_vfs) + return hinic_pci_sriov_disable(dev); + else + return hinic_pci_sriov_enable(dev, num_vfs); +} diff --git a/drivers/net/ethernet/huawei/hinic/hinic_sriov.h b/drivers/net/ethernet/huawei/hinic/hinic_sriov.h new file mode 100644 index 000000000000..ba627a362f9a --- /dev/null +++ b/drivers/net/ethernet/huawei/hinic/hinic_sriov.h @@ -0,0 +1,109 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Huawei HiNIC PCI Express Linux driver + * Copyright(c) 2017 Huawei Technologies Co., Ltd + */ + +#ifndef HINIC_SRIOV_H +#define HINIC_SRIOV_H + +#include "hinic_hw_dev.h" + +#define OS_VF_ID_TO_HW(os_vf_id) ((os_vf_id) + 1) +#define HW_VF_ID_TO_OS(hw_vf_id) ((hw_vf_id) - 1) + +enum hinic_sriov_state { + HINIC_SRIOV_DISABLE, + HINIC_SRIOV_ENABLE, + HINIC_FUNC_REMOVE, +}; + +enum { + HINIC_IFLA_VF_LINK_STATE_AUTO, /* link state of the uplink */ + HINIC_IFLA_VF_LINK_STATE_ENABLE, /* link always up */ + HINIC_IFLA_VF_LINK_STATE_DISABLE, /* link always down */ +}; + +struct hinic_sriov_info { + struct pci_dev *pdev; + struct hinic_hwdev *hwdev; + bool sriov_enabled; + unsigned int num_vfs; + unsigned long state; +}; + +struct vf_data_storage { + u8 vf_mac_addr[ETH_ALEN]; + bool registered; + bool pf_set_mac; + u16 pf_vlan; + u8 pf_qos; + u32 max_rate; + u32 min_rate; + + bool link_forced; + bool link_up; /* only valid if VF link is forced */ + bool spoofchk; + bool trust; +}; + +struct hinic_register_vf { + u8 status; + u8 version; + u8 rsvd0[6]; +}; + +struct hinic_port_mac_update { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 vlan_id; + u16 rsvd1; + u8 old_mac[ETH_ALEN]; + u16 rsvd2; + u8 new_mac[ETH_ALEN]; +}; + +struct hinic_vf_vlan_config { + u8 status; + u8 version; + u8 rsvd0[6]; + + u16 func_id; + u16 vlan_id; + u8 qos; + u8 rsvd1[7]; +}; + +int hinic_ndo_set_vf_mac(struct net_device *netdev, int vf, u8 *mac); + +int hinic_ndo_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos, + __be16 vlan_proto); + +int hinic_ndo_get_vf_config(struct net_device *netdev, + int vf, struct ifla_vf_info *ivi); + +int hinic_ndo_set_vf_trust(struct net_device *netdev, int vf, bool setting); + +int hinic_ndo_set_vf_bw(struct net_device *netdev, + int vf, int min_tx_rate, int max_tx_rate); + +int hinic_ndo_set_vf_spoofchk(struct net_device *netdev, int vf, bool setting); + +int hinic_ndo_set_vf_link_state(struct net_device *netdev, int vf_id, int link); + +void hinic_notify_all_vfs_link_changed(struct hinic_hwdev *hwdev, + u8 link_status); + +int hinic_pci_sriov_disable(struct pci_dev *dev); + +int hinic_pci_sriov_enable(struct pci_dev *dev, int num_vfs); + +int hinic_vf_func_init(struct hinic_hwdev *hwdev); + +void hinic_vf_func_free(struct hinic_hwdev *hwdev); + +int hinic_pci_sriov_configure(struct pci_dev *dev, int num_vfs); + +#endif diff --git a/drivers/net/ethernet/huawei/hinic/hinic_tx.c b/drivers/net/ethernet/huawei/hinic/hinic_tx.c index 365016450bdb..4c66a0bc1b28 100644 --- a/drivers/net/ethernet/huawei/hinic/hinic_tx.c +++ b/drivers/net/ethernet/huawei/hinic/hinic_tx.c @@ -673,9 +673,11 @@ static int free_tx_poll(struct napi_struct *napi, int budget) if (pkts < budget) { napi_complete(napi); - hinic_hwdev_set_msix_state(nic_dev->hwdev, - sq->msix_entry, - HINIC_MSIX_ENABLE); + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) + hinic_hwdev_set_msix_state(nic_dev->hwdev, + sq->msix_entry, + HINIC_MSIX_ENABLE); + return pkts; } @@ -701,10 +703,11 @@ static irqreturn_t tx_irq(int irq, void *data) nic_dev = netdev_priv(txq->netdev); - /* Disable the interrupt until napi will be completed */ - hinic_hwdev_set_msix_state(nic_dev->hwdev, - txq->sq->msix_entry, - HINIC_MSIX_DISABLE); + if (!HINIC_IS_VF(nic_dev->hwdev->hwif)) + /* Disable the interrupt until napi will be completed */ + hinic_hwdev_set_msix_state(nic_dev->hwdev, + txq->sq->msix_entry, + HINIC_MSIX_DISABLE); hinic_hwdev_msix_cnt_set(nic_dev->hwdev, txq->sq->msix_entry); diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 3de549c6c693..197dc5b2c090 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -4678,12 +4678,10 @@ static void ibmvnic_handle_crq(union ibmvnic_crq *crq, dev_err(dev, "Error %ld in VERSION_EXCHG_RSP\n", rc); break; } - dev_info(dev, "Partner protocol version is %d\n", - crq->version_exchange_rsp.version); - if (be16_to_cpu(crq->version_exchange_rsp.version) < - ibmvnic_version) - ibmvnic_version = + ibmvnic_version = be16_to_cpu(crq->version_exchange_rsp.version); + dev_info(dev, "Partner protocol version is %d\n", + ibmvnic_version); send_cap_queries(adapter); break; case QUERY_CAPABILITY_RSP: diff --git a/drivers/net/ethernet/intel/e1000/e1000_hw.c b/drivers/net/ethernet/intel/e1000/e1000_hw.c index 48428d6a00be..623e516a9630 100644 --- a/drivers/net/ethernet/intel/e1000/e1000_hw.c +++ b/drivers/net/ethernet/intel/e1000/e1000_hw.c @@ -3960,7 +3960,7 @@ static s32 e1000_do_read_eeprom(struct e1000_hw *hw, u16 offset, u16 words, * @hw: Struct containing variables accessed by shared code * * Reads the first 64 16 bit words of the EEPROM and sums the values read. - * If the the sum of the 64 16 bit words is 0xBABA, the EEPROM's checksum is + * If the sum of the 64 16 bit words is 0xBABA, the EEPROM's checksum is * valid. */ s32 e1000_validate_eeprom_checksum(struct e1000_hw *hw) diff --git a/drivers/net/ethernet/intel/e1000/e1000_main.c b/drivers/net/ethernet/intel/e1000/e1000_main.c index 0d51cbc88028..d9fa4600f745 100644 --- a/drivers/net/ethernet/intel/e1000/e1000_main.c +++ b/drivers/net/ethernet/intel/e1000/e1000_main.c @@ -542,8 +542,13 @@ void e1000_reinit_locked(struct e1000_adapter *adapter) WARN_ON(in_interrupt()); while (test_and_set_bit(__E1000_RESETTING, &adapter->flags)) msleep(1); - e1000_down(adapter); - e1000_up(adapter); + + /* only run the task if not already down */ + if (!test_bit(__E1000_DOWN, &adapter->flags)) { + e1000_down(adapter); + e1000_up(adapter); + } + clear_bit(__E1000_RESETTING, &adapter->flags); } @@ -1433,10 +1438,15 @@ int e1000_close(struct net_device *netdev) struct e1000_hw *hw = &adapter->hw; int count = E1000_CHECK_RESET_COUNT; - while (test_bit(__E1000_RESETTING, &adapter->flags) && count--) + while (test_and_set_bit(__E1000_RESETTING, &adapter->flags) && count--) usleep_range(10000, 20000); - WARN_ON(test_bit(__E1000_RESETTING, &adapter->flags)); + WARN_ON(count < 0); + + /* signal that we're down so that the reset task will no longer run */ + set_bit(__E1000_DOWN, &adapter->flags); + clear_bit(__E1000_RESETTING, &adapter->flags); + e1000_down(adapter); e1000_power_down_phy(adapter); e1000_free_irq(adapter); @@ -3136,8 +3146,9 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb, hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb); if (skb->data_len && hdr_len == len) { switch (hw->mac_type) { + case e1000_82544: { unsigned int pull_size; - case e1000_82544: + /* Make sure we have room to chop off 4 bytes, * and that the end alignment will work out to * this hardware's requirements @@ -3158,6 +3169,7 @@ static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb, } len = skb_headlen(skb); break; + } default: /* do nothing */ break; diff --git a/drivers/net/ethernet/intel/e1000e/e1000.h b/drivers/net/ethernet/intel/e1000e/e1000.h index 37a2314d3e6b..944abd5eae11 100644 --- a/drivers/net/ethernet/intel/e1000e/e1000.h +++ b/drivers/net/ethernet/intel/e1000e/e1000.h @@ -576,7 +576,6 @@ static inline u32 __er32(struct e1000_hw *hw, unsigned long reg) #define er32(reg) __er32(hw, E1000_##reg) -s32 __ew32_prepare(struct e1000_hw *hw); void __ew32(struct e1000_hw *hw, unsigned long reg, u32 val); #define ew32(reg, val) __ew32(hw, E1000_##reg, (val)) diff --git a/drivers/net/ethernet/intel/e1000e/ich8lan.c b/drivers/net/ethernet/intel/e1000e/ich8lan.c index 735bf25952fc..f999cca37a8a 100644 --- a/drivers/net/ethernet/intel/e1000e/ich8lan.c +++ b/drivers/net/ethernet/intel/e1000e/ich8lan.c @@ -300,7 +300,11 @@ static s32 e1000_init_phy_workarounds_pchlan(struct e1000_hw *hw) * so forcibly disable it. */ hw->dev_spec.ich8lan.ulp_state = e1000_ulp_state_unknown; - e1000_disable_ulp_lpt_lp(hw, true); + ret_val = e1000_disable_ulp_lpt_lp(hw, true); + if (ret_val) { + e_warn("Failed to disable ULP\n"); + goto out; + } ret_val = hw->phy.ops.acquire(hw); if (ret_val) { diff --git a/drivers/net/ethernet/intel/e1000e/netdev.c b/drivers/net/ethernet/intel/e1000e/netdev.c index 177c6da80c57..a279f4fa9962 100644 --- a/drivers/net/ethernet/intel/e1000e/netdev.c +++ b/drivers/net/ethernet/intel/e1000e/netdev.c @@ -107,6 +107,45 @@ static const struct e1000_reg_info e1000_reg_info_tbl[] = { {0, NULL} }; +struct e1000e_me_supported { + u16 device_id; /* supported device ID */ +}; + +static const struct e1000e_me_supported me_supported[] = { + {E1000_DEV_ID_PCH_LPT_I217_LM}, + {E1000_DEV_ID_PCH_LPTLP_I218_LM}, + {E1000_DEV_ID_PCH_I218_LM2}, + {E1000_DEV_ID_PCH_I218_LM3}, + {E1000_DEV_ID_PCH_SPT_I219_LM}, + {E1000_DEV_ID_PCH_SPT_I219_LM2}, + {E1000_DEV_ID_PCH_LBG_I219_LM3}, + {E1000_DEV_ID_PCH_SPT_I219_LM4}, + {E1000_DEV_ID_PCH_SPT_I219_LM5}, + {E1000_DEV_ID_PCH_CNP_I219_LM6}, + {E1000_DEV_ID_PCH_CNP_I219_LM7}, + {E1000_DEV_ID_PCH_ICP_I219_LM8}, + {E1000_DEV_ID_PCH_ICP_I219_LM9}, + {E1000_DEV_ID_PCH_CMP_I219_LM10}, + {E1000_DEV_ID_PCH_CMP_I219_LM11}, + {E1000_DEV_ID_PCH_CMP_I219_LM12}, + {E1000_DEV_ID_PCH_TGP_I219_LM13}, + {E1000_DEV_ID_PCH_TGP_I219_LM14}, + {E1000_DEV_ID_PCH_TGP_I219_LM15}, + {0} +}; + +static bool e1000e_check_me(u16 device_id) +{ + struct e1000e_me_supported *id; + + for (id = (struct e1000e_me_supported *)me_supported; + id->device_id; id++) + if (device_id == id->device_id) + return true; + + return false; +} + /** * __ew32_prepare - prepare to write to MAC CSR register on certain parts * @hw: pointer to the HW structure @@ -119,14 +158,12 @@ static const struct e1000_reg_info e1000_reg_info_tbl[] = { * has bit 24 set while ME is accessing MAC CSR registers, wait if it is set * and try again a number of times. **/ -s32 __ew32_prepare(struct e1000_hw *hw) +static void __ew32_prepare(struct e1000_hw *hw) { s32 i = E1000_ICH_FWSM_PCIM2PCI_COUNT; while ((er32(FWSM) & E1000_ICH_FWSM_PCIM2PCI) && --i) udelay(50); - - return i; } void __ew32(struct e1000_hw *hw, unsigned long reg, u32 val) @@ -607,11 +644,11 @@ static void e1000e_update_rdt_wa(struct e1000_ring *rx_ring, unsigned int i) { struct e1000_adapter *adapter = rx_ring->adapter; struct e1000_hw *hw = &adapter->hw; - s32 ret_val = __ew32_prepare(hw); + __ew32_prepare(hw); writel(i, rx_ring->tail); - if (unlikely(!ret_val && (i != readl(rx_ring->tail)))) { + if (unlikely(i != readl(rx_ring->tail))) { u32 rctl = er32(RCTL); ew32(RCTL, rctl & ~E1000_RCTL_EN); @@ -624,11 +661,11 @@ static void e1000e_update_tdt_wa(struct e1000_ring *tx_ring, unsigned int i) { struct e1000_adapter *adapter = tx_ring->adapter; struct e1000_hw *hw = &adapter->hw; - s32 ret_val = __ew32_prepare(hw); + __ew32_prepare(hw); writel(i, tx_ring->tail); - if (unlikely(!ret_val && (i != readl(tx_ring->tail)))) { + if (unlikely(i != readl(tx_ring->tail))) { u32 tctl = er32(TCTL); ew32(TCTL, tctl & ~E1000_TCTL_EN); @@ -5294,6 +5331,10 @@ static void e1000_watchdog_task(struct work_struct *work) /* oops */ break; } + if (hw->mac.type == e1000_pch_spt) { + netdev->features &= ~NETIF_F_TSO; + netdev->features &= ~NETIF_F_TSO6; + } } /* enable transmits in the hardware, need to do this @@ -6404,6 +6445,31 @@ static void e1000e_s0ix_entry_flow(struct e1000_adapter *adapter) mac_data |= BIT(3); ew32(CTRL_EXT, mac_data); + /* Disable disconnected cable conditioning for Power Gating */ + mac_data = er32(DPGFR); + mac_data |= BIT(2); + ew32(DPGFR, mac_data); + + /* Don't wake from dynamic Power Gating with clock request */ + mac_data = er32(FEXTNVM12); + mac_data |= BIT(12); + ew32(FEXTNVM12, mac_data); + + /* Ungate PGCB clock */ + mac_data = er32(FEXTNVM9); + mac_data |= BIT(28); + ew32(FEXTNVM9, mac_data); + + /* Enable K1 off to enable mPHY Power Gating */ + mac_data = er32(FEXTNVM6); + mac_data |= BIT(31); + ew32(FEXTNVM12, mac_data); + + /* Enable mPHY power gating for any link and speed */ + mac_data = er32(FEXTNVM8); + mac_data |= BIT(9); + ew32(FEXTNVM8, mac_data); + /* Enable the Dynamic Clock Gating in the DMA and MAC */ mac_data = er32(CTRL_EXT); mac_data |= E1000_CTRL_EXT_DMA_DYN_CLK_EN; @@ -6433,6 +6499,35 @@ static void e1000e_s0ix_exit_flow(struct e1000_adapter *adapter) mac_data |= BIT(0); ew32(FEXTNVM7, mac_data); + /* Disable mPHY power gating for any link and speed */ + mac_data = er32(FEXTNVM8); + mac_data &= ~BIT(9); + ew32(FEXTNVM8, mac_data); + + /* Disable K1 off */ + mac_data = er32(FEXTNVM6); + mac_data &= ~BIT(31); + ew32(FEXTNVM12, mac_data); + + /* Disable Ungate PGCB clock */ + mac_data = er32(FEXTNVM9); + mac_data &= ~BIT(28); + ew32(FEXTNVM9, mac_data); + + /* Cancel not waking from dynamic + * Power Gating with clock request + */ + mac_data = er32(FEXTNVM12); + mac_data &= ~BIT(12); + ew32(FEXTNVM12, mac_data); + + /* Cancel disable disconnected cable conditioning + * for Power Gating + */ + mac_data = er32(DPGFR); + mac_data &= ~BIT(2); + ew32(DPGFR, mac_data); + /* Disable Dynamic Power Gating */ mac_data = er32(CTRL_EXT); mac_data &= 0xFFFFFFF7; @@ -6858,7 +6953,8 @@ static int e1000e_pm_suspend(struct device *dev) e1000e_pm_thaw(dev); /* Introduce S0ix implementation */ - if (hw->mac.type >= e1000_pch_cnp) + if (hw->mac.type >= e1000_pch_cnp && + !e1000e_check_me(hw->adapter->pdev->device)) e1000e_s0ix_entry_flow(adapter); return rc; @@ -6873,7 +6969,8 @@ static int e1000e_pm_resume(struct device *dev) int rc; /* Introduce S0ix implementation */ - if (hw->mac.type >= e1000_pch_cnp) + if (hw->mac.type >= e1000_pch_cnp && + !e1000e_check_me(hw->adapter->pdev->device)) e1000e_s0ix_exit_flow(adapter); rc = __e1000_resume(pdev); @@ -7549,7 +7646,7 @@ static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent) e1000_print_device_info(adapter); - dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NEVER_SKIP); + dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); if (pci_dev_run_wake(pdev) && hw->mac.type < e1000_pch_cnp) pm_runtime_put_noidle(&pdev->dev); diff --git a/drivers/net/ethernet/intel/e1000e/regs.h b/drivers/net/ethernet/intel/e1000e/regs.h index df59fd1d660c..8165ba2619a4 100644 --- a/drivers/net/ethernet/intel/e1000e/regs.h +++ b/drivers/net/ethernet/intel/e1000e/regs.h @@ -21,9 +21,12 @@ #define E1000_FEXTNVM5 0x00014 /* Future Extended NVM 5 - RW */ #define E1000_FEXTNVM6 0x00010 /* Future Extended NVM 6 - RW */ #define E1000_FEXTNVM7 0x000E4 /* Future Extended NVM 7 - RW */ +#define E1000_FEXTNVM8 0x5BB0 /* Future Extended NVM 8 - RW */ #define E1000_FEXTNVM9 0x5BB4 /* Future Extended NVM 9 - RW */ #define E1000_FEXTNVM11 0x5BBC /* Future Extended NVM 11 - RW */ +#define E1000_FEXTNVM12 0x5BC0 /* Future Extended NVM 12 - RW */ #define E1000_PCIEANACFG 0x00F18 /* PCIE Analog Config */ +#define E1000_DPGFR 0x00FAC /* Dynamic Power Gate Force Control Register */ #define E1000_FCT 0x00030 /* Flow Control Type - RW */ #define E1000_VET 0x00038 /* VLAN Ether Type - RW */ #define E1000_ICR 0x000C0 /* Interrupt Cause Read - R/clr */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq.c b/drivers/net/ethernet/intel/i40e/i40e_adminq.c index 37514a75f928..6a089848c857 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_adminq.c +++ b/drivers/net/ethernet/intel/i40e/i40e_adminq.c @@ -694,10 +694,8 @@ init_adminq_exit: * i40e_shutdown_adminq - shutdown routine for the Admin Queue * @hw: pointer to the hardware structure **/ -i40e_status i40e_shutdown_adminq(struct i40e_hw *hw) +void i40e_shutdown_adminq(struct i40e_hw *hw) { - i40e_status ret_code = 0; - if (i40e_check_asq_alive(hw)) i40e_aq_queue_shutdown(hw, true); @@ -706,8 +704,6 @@ i40e_status i40e_shutdown_adminq(struct i40e_hw *hw) if (hw->nvm_buff.va) i40e_free_virt_mem(hw, &hw->nvm_buff); - - return ret_code; } /** diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 2a037ec244b9..5d807c8004f8 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -11,7 +11,7 @@ #include "i40e_diag.h" #include "i40e_xsk.h" #include <net/udp_tunnel.h> -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> /* All i40e tracepoints are defined by the include below, which * must be included exactly once across the whole kernel with * CREATE_TRACE_POINTS defined @@ -3260,26 +3260,31 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring) if (ring->vsi->type == I40E_VSI_MAIN) xdp_rxq_info_unreg_mem_model(&ring->xdp_rxq); + kfree(ring->rx_bi); ring->xsk_umem = i40e_xsk_umem(ring); if (ring->xsk_umem) { - ring->rx_buf_len = ring->xsk_umem->chunk_size_nohr - - XDP_PACKET_HEADROOM; + ret = i40e_alloc_rx_bi_zc(ring); + if (ret) + return ret; + ring->rx_buf_len = xsk_umem_get_rx_frame_size(ring->xsk_umem); /* For AF_XDP ZC, we disallow packets to span on * multiple buffers, thus letting us skip that * handling in the fast-path. */ chain_len = 1; - ring->zca.free = i40e_zca_free; ret = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq, - MEM_TYPE_ZERO_COPY, - &ring->zca); + MEM_TYPE_XSK_BUFF_POOL, + NULL); if (ret) return ret; dev_info(&vsi->back->pdev->dev, - "Registered XDP mem model MEM_TYPE_ZERO_COPY on Rx ring %d\n", + "Registered XDP mem model MEM_TYPE_XSK_BUFF_POOL on Rx ring %d\n", ring->queue_index); } else { + ret = i40e_alloc_rx_bi(ring); + if (ret) + return ret; ring->rx_buf_len = vsi->rx_buf_len; if (ring->vsi->type == I40E_VSI_MAIN) { ret = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq, @@ -3344,9 +3349,12 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring) ring->tail = hw->hw_addr + I40E_QRX_TAIL(pf_q); writel(0, ring->tail); - ok = ring->xsk_umem ? - i40e_alloc_rx_buffers_zc(ring, I40E_DESC_UNUSED(ring)) : - !i40e_alloc_rx_buffers(ring, I40E_DESC_UNUSED(ring)); + if (ring->xsk_umem) { + xsk_buff_set_rxq_info(ring->xsk_umem, &ring->xdp_rxq); + ok = i40e_alloc_rx_buffers_zc(ring, I40E_DESC_UNUSED(ring)); + } else { + ok = !i40e_alloc_rx_buffers(ring, I40E_DESC_UNUSED(ring)); + } if (!ok) { /* Log this in case the user has forgotten to give the kernel * any buffers, even later in the application. @@ -14478,29 +14486,29 @@ static void i40e_print_features(struct i40e_pf *pf) i = snprintf(buf, INFO_STRING_LEN, "Features: PF-id[%d]", hw->pf_id); #ifdef CONFIG_PCI_IOV - i += snprintf(&buf[i], REMAIN(i), " VFs: %d", pf->num_req_vfs); + i += scnprintf(&buf[i], REMAIN(i), " VFs: %d", pf->num_req_vfs); #endif - i += snprintf(&buf[i], REMAIN(i), " VSIs: %d QP: %d", + i += scnprintf(&buf[i], REMAIN(i), " VSIs: %d QP: %d", pf->hw.func_caps.num_vsis, pf->vsi[pf->lan_vsi]->num_queue_pairs); if (pf->flags & I40E_FLAG_RSS_ENABLED) - i += snprintf(&buf[i], REMAIN(i), " RSS"); + i += scnprintf(&buf[i], REMAIN(i), " RSS"); if (pf->flags & I40E_FLAG_FD_ATR_ENABLED) - i += snprintf(&buf[i], REMAIN(i), " FD_ATR"); + i += scnprintf(&buf[i], REMAIN(i), " FD_ATR"); if (pf->flags & I40E_FLAG_FD_SB_ENABLED) { - i += snprintf(&buf[i], REMAIN(i), " FD_SB"); - i += snprintf(&buf[i], REMAIN(i), " NTUPLE"); + i += scnprintf(&buf[i], REMAIN(i), " FD_SB"); + i += scnprintf(&buf[i], REMAIN(i), " NTUPLE"); } if (pf->flags & I40E_FLAG_DCB_CAPABLE) - i += snprintf(&buf[i], REMAIN(i), " DCB"); - i += snprintf(&buf[i], REMAIN(i), " VxLAN"); - i += snprintf(&buf[i], REMAIN(i), " Geneve"); + i += scnprintf(&buf[i], REMAIN(i), " DCB"); + i += scnprintf(&buf[i], REMAIN(i), " VxLAN"); + i += scnprintf(&buf[i], REMAIN(i), " Geneve"); if (pf->flags & I40E_FLAG_PTP) - i += snprintf(&buf[i], REMAIN(i), " PTP"); + i += scnprintf(&buf[i], REMAIN(i), " PTP"); if (pf->flags & I40E_FLAG_VEB_MODE_ENABLED) - i += snprintf(&buf[i], REMAIN(i), " VEB"); + i += scnprintf(&buf[i], REMAIN(i), " VEB"); else - i += snprintf(&buf[i], REMAIN(i), " VEPA"); + i += scnprintf(&buf[i], REMAIN(i), " VEPA"); dev_info(&pf->pdev->dev, "%s\n", buf); kfree(buf); diff --git a/drivers/net/ethernet/intel/i40e/i40e_prototype.h b/drivers/net/ethernet/intel/i40e/i40e_prototype.h index bbb478f09093..5c1378641b3b 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_prototype.h +++ b/drivers/net/ethernet/intel/i40e/i40e_prototype.h @@ -17,7 +17,7 @@ /* adminq functions */ i40e_status i40e_init_adminq(struct i40e_hw *hw); -i40e_status i40e_shutdown_adminq(struct i40e_hw *hw); +void i40e_shutdown_adminq(struct i40e_hw *hw); void i40e_adminq_init_ring_data(struct i40e_hw *hw); i40e_status i40e_clean_arq_element(struct i40e_hw *hw, struct i40e_arq_event_info *e, diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c index b8496037ef7f..f9555c847f73 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c @@ -521,28 +521,29 @@ int i40e_add_del_fdir(struct i40e_vsi *vsi, /** * i40e_fd_handle_status - check the Programming Status for FD * @rx_ring: the Rx ring for this descriptor - * @rx_desc: the Rx descriptor for programming Status, not a packet descriptor. + * @qword0_raw: qword0 + * @qword1: qword1 after le_to_cpu * @prog_id: the id originally used for programming * * This is used to verify if the FD programming or invalidation * requested by SW to the HW is successful or not and take actions accordingly. **/ -void i40e_fd_handle_status(struct i40e_ring *rx_ring, - union i40e_rx_desc *rx_desc, u8 prog_id) +static void i40e_fd_handle_status(struct i40e_ring *rx_ring, u64 qword0_raw, + u64 qword1, u8 prog_id) { struct i40e_pf *pf = rx_ring->vsi->back; struct pci_dev *pdev = pf->pdev; + struct i40e_32b_rx_wb_qw0 *qw0; u32 fcnt_prog, fcnt_avail; u32 error; - u64 qw; - qw = le64_to_cpu(rx_desc->wb.qword1.status_error_len); - error = (qw & I40E_RX_PROG_STATUS_DESC_QW1_ERROR_MASK) >> + qw0 = (struct i40e_32b_rx_wb_qw0 *)&qword0_raw; + error = (qword1 & I40E_RX_PROG_STATUS_DESC_QW1_ERROR_MASK) >> I40E_RX_PROG_STATUS_DESC_QW1_ERROR_SHIFT; if (error == BIT(I40E_RX_PROG_STATUS_DESC_FD_TBL_FULL_SHIFT)) { - pf->fd_inv = le32_to_cpu(rx_desc->wb.qword0.hi_dword.fd_id); - if ((rx_desc->wb.qword0.hi_dword.fd_id != 0) || + pf->fd_inv = le32_to_cpu(qw0->hi_dword.fd_id); + if (qw0->hi_dword.fd_id != 0 || (I40E_DEBUG_FD & pf->hw.debug_mask)) dev_warn(&pdev->dev, "ntuple filter loc = %d, could not be added\n", pf->fd_inv); @@ -560,7 +561,7 @@ void i40e_fd_handle_status(struct i40e_ring *rx_ring, /* store the current atr filter count */ pf->fd_atr_cnt = i40e_get_current_atr_cnt(pf); - if ((rx_desc->wb.qword0.hi_dword.fd_id == 0) && + if (qw0->hi_dword.fd_id == 0 && test_bit(__I40E_FD_SB_AUTO_DISABLED, pf->state)) { /* These set_bit() calls aren't atomic with the * test_bit() here, but worse case we potentially @@ -589,7 +590,7 @@ void i40e_fd_handle_status(struct i40e_ring *rx_ring, } else if (error == BIT(I40E_RX_PROG_STATUS_DESC_NO_FD_ENTRY_SHIFT)) { if (I40E_DEBUG_FD & pf->hw.debug_mask) dev_info(&pdev->dev, "ntuple filter fd_id = %d, could not be removed\n", - rx_desc->wb.qword0.hi_dword.fd_id); + qw0->hi_dword.fd_id); } } @@ -1195,6 +1196,11 @@ clear_counts: rc->total_packets = 0; } +static struct i40e_rx_buffer *i40e_rx_bi(struct i40e_ring *rx_ring, u32 idx) +{ + return &rx_ring->rx_bi[idx]; +} + /** * i40e_reuse_rx_page - page flip buffer and store it back on the ring * @rx_ring: rx descriptor ring to store buffers on @@ -1208,7 +1214,7 @@ static void i40e_reuse_rx_page(struct i40e_ring *rx_ring, struct i40e_rx_buffer *new_buff; u16 nta = rx_ring->next_to_alloc; - new_buff = &rx_ring->rx_bi[nta]; + new_buff = i40e_rx_bi(rx_ring, nta); /* update, and store next to alloc */ nta++; @@ -1227,29 +1233,10 @@ static void i40e_reuse_rx_page(struct i40e_ring *rx_ring, } /** - * i40e_rx_is_programming_status - check for programming status descriptor - * @qw: qword representing status_error_len in CPU ordering - * - * The value of in the descriptor length field indicate if this - * is a programming status descriptor for flow director or FCoE - * by the value of I40E_RX_PROG_STATUS_DESC_LENGTH, otherwise - * it is a packet descriptor. - **/ -static inline bool i40e_rx_is_programming_status(u64 qw) -{ - /* The Rx filter programming status and SPH bit occupy the same - * spot in the descriptor. Since we don't support packet split we - * can just reuse the bit as an indication that this is a - * programming status descriptor. - */ - return qw & I40E_RXD_QW1_LENGTH_SPH_MASK; -} - -/** - * i40e_clean_programming_status - try clean the programming status descriptor + * i40e_clean_programming_status - clean the programming status descriptor * @rx_ring: the rx ring that has this descriptor - * @rx_desc: the rx descriptor written back by HW - * @qw: qword representing status_error_len in CPU ordering + * @qword0_raw: qword0 + * @qword1: qword1 representing status_error_len in CPU ordering * * Flow director should handle FD_FILTER_STATUS to check its filter programming * status being successful or not and take actions accordingly. FCoE should @@ -1257,34 +1244,16 @@ static inline bool i40e_rx_is_programming_status(u64 qw) * * Returns an i40e_rx_buffer to reuse if the cleanup occurred, otherwise NULL. **/ -struct i40e_rx_buffer *i40e_clean_programming_status( - struct i40e_ring *rx_ring, - union i40e_rx_desc *rx_desc, - u64 qw) +void i40e_clean_programming_status(struct i40e_ring *rx_ring, u64 qword0_raw, + u64 qword1) { - struct i40e_rx_buffer *rx_buffer; - u32 ntc; u8 id; - if (!i40e_rx_is_programming_status(qw)) - return NULL; - - ntc = rx_ring->next_to_clean; - - /* fetch, update, and store next to clean */ - rx_buffer = &rx_ring->rx_bi[ntc++]; - ntc = (ntc < rx_ring->count) ? ntc : 0; - rx_ring->next_to_clean = ntc; - - prefetch(I40E_RX_DESC(rx_ring, ntc)); - - id = (qw & I40E_RX_PROG_STATUS_DESC_QW1_PROGID_MASK) >> + id = (qword1 & I40E_RX_PROG_STATUS_DESC_QW1_PROGID_MASK) >> I40E_RX_PROG_STATUS_DESC_QW1_PROGID_SHIFT; if (id == I40E_RX_PROG_STATUS_DESC_FD_FILTER_STATUS) - i40e_fd_handle_status(rx_ring, rx_desc, id); - - return rx_buffer; + i40e_fd_handle_status(rx_ring, qword0_raw, qword1, id); } /** @@ -1336,13 +1305,25 @@ err: return -ENOMEM; } +int i40e_alloc_rx_bi(struct i40e_ring *rx_ring) +{ + unsigned long sz = sizeof(*rx_ring->rx_bi) * rx_ring->count; + + rx_ring->rx_bi = kzalloc(sz, GFP_KERNEL); + return rx_ring->rx_bi ? 0 : -ENOMEM; +} + +static void i40e_clear_rx_bi(struct i40e_ring *rx_ring) +{ + memset(rx_ring->rx_bi, 0, sizeof(*rx_ring->rx_bi) * rx_ring->count); +} + /** * i40e_clean_rx_ring - Free Rx buffers * @rx_ring: ring to be cleaned **/ void i40e_clean_rx_ring(struct i40e_ring *rx_ring) { - unsigned long bi_size; u16 i; /* ring already cleared, nothing to do */ @@ -1361,7 +1342,7 @@ void i40e_clean_rx_ring(struct i40e_ring *rx_ring) /* Free all the Rx ring sk_buffs */ for (i = 0; i < rx_ring->count; i++) { - struct i40e_rx_buffer *rx_bi = &rx_ring->rx_bi[i]; + struct i40e_rx_buffer *rx_bi = i40e_rx_bi(rx_ring, i); if (!rx_bi->page) continue; @@ -1388,8 +1369,10 @@ void i40e_clean_rx_ring(struct i40e_ring *rx_ring) } skip_free: - bi_size = sizeof(struct i40e_rx_buffer) * rx_ring->count; - memset(rx_ring->rx_bi, 0, bi_size); + if (rx_ring->xsk_umem) + i40e_clear_rx_bi_zc(rx_ring); + else + i40e_clear_rx_bi(rx_ring); /* Zero out the descriptor ring */ memset(rx_ring->desc, 0, rx_ring->size); @@ -1430,15 +1413,7 @@ void i40e_free_rx_resources(struct i40e_ring *rx_ring) int i40e_setup_rx_descriptors(struct i40e_ring *rx_ring) { struct device *dev = rx_ring->dev; - int err = -ENOMEM; - int bi_size; - - /* warn if we are about to overwrite the pointer */ - WARN_ON(rx_ring->rx_bi); - bi_size = sizeof(struct i40e_rx_buffer) * rx_ring->count; - rx_ring->rx_bi = kzalloc(bi_size, GFP_KERNEL); - if (!rx_ring->rx_bi) - goto err; + int err; u64_stats_init(&rx_ring->syncp); @@ -1451,7 +1426,7 @@ int i40e_setup_rx_descriptors(struct i40e_ring *rx_ring) if (!rx_ring->desc) { dev_info(dev, "Unable to allocate memory for the Rx descriptor ring, size=%d\n", rx_ring->size); - goto err; + return -ENOMEM; } rx_ring->next_to_alloc = 0; @@ -1463,16 +1438,12 @@ int i40e_setup_rx_descriptors(struct i40e_ring *rx_ring) err = xdp_rxq_info_reg(&rx_ring->xdp_rxq, rx_ring->netdev, rx_ring->queue_index); if (err < 0) - goto err; + return err; } rx_ring->xdp_prog = rx_ring->vsi->xdp_prog; return 0; -err: - kfree(rx_ring->rx_bi); - rx_ring->rx_bi = NULL; - return err; } /** @@ -1507,6 +1478,22 @@ static inline unsigned int i40e_rx_offset(struct i40e_ring *rx_ring) return ring_uses_build_skb(rx_ring) ? I40E_SKB_PAD : 0; } +static unsigned int i40e_rx_frame_truesize(struct i40e_ring *rx_ring, + unsigned int size) +{ + unsigned int truesize; + +#if (PAGE_SIZE < 8192) + truesize = i40e_rx_pg_size(rx_ring) / 2; /* Must be power-of-2 */ +#else + truesize = i40e_rx_offset(rx_ring) ? + SKB_DATA_ALIGN(size + i40e_rx_offset(rx_ring)) + + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) : + SKB_DATA_ALIGN(size); +#endif + return truesize; +} + /** * i40e_alloc_mapped_page - recycle or make a new page * @rx_ring: ring to use @@ -1576,7 +1563,7 @@ bool i40e_alloc_rx_buffers(struct i40e_ring *rx_ring, u16 cleaned_count) return false; rx_desc = I40E_RX_DESC(rx_ring, ntu); - bi = &rx_ring->rx_bi[ntu]; + bi = i40e_rx_bi(rx_ring, ntu); do { if (!i40e_alloc_mapped_page(rx_ring, bi)) @@ -1598,7 +1585,7 @@ bool i40e_alloc_rx_buffers(struct i40e_ring *rx_ring, u16 cleaned_count) ntu++; if (unlikely(ntu == rx_ring->count)) { rx_desc = I40E_RX_DESC(rx_ring, 0); - bi = rx_ring->rx_bi; + bi = i40e_rx_bi(rx_ring, 0); ntu = 0; } @@ -1965,7 +1952,7 @@ static struct i40e_rx_buffer *i40e_get_rx_buffer(struct i40e_ring *rx_ring, { struct i40e_rx_buffer *rx_buffer; - rx_buffer = &rx_ring->rx_bi[rx_ring->next_to_clean]; + rx_buffer = i40e_rx_bi(rx_ring, rx_ring->next_to_clean); prefetchw(rx_buffer->page); /* we are reusing so sync this buffer for CPU use */ @@ -2180,7 +2167,7 @@ static int i40e_xmit_xdp_ring(struct xdp_frame *xdpf, int i40e_xmit_xdp_tx_ring(struct xdp_buff *xdp, struct i40e_ring *xdp_ring) { - struct xdp_frame *xdpf = convert_to_xdp_frame(xdp); + struct xdp_frame *xdpf = xdp_convert_buff_to_frame(xdp); if (unlikely(!xdpf)) return I40E_XDP_CONSUMED; @@ -2246,13 +2233,11 @@ static void i40e_rx_buffer_flip(struct i40e_ring *rx_ring, struct i40e_rx_buffer *rx_buffer, unsigned int size) { -#if (PAGE_SIZE < 8192) - unsigned int truesize = i40e_rx_pg_size(rx_ring) / 2; + unsigned int truesize = i40e_rx_frame_truesize(rx_ring, size); +#if (PAGE_SIZE < 8192) rx_buffer->page_offset ^= truesize; #else - unsigned int truesize = SKB_DATA_ALIGN(i40e_rx_offset(rx_ring) + size); - rx_buffer->page_offset += truesize; #endif } @@ -2335,6 +2320,9 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget) bool failure = false; struct xdp_buff xdp; +#if (PAGE_SIZE < 8192) + xdp.frame_sz = i40e_rx_frame_truesize(rx_ring, 0); +#endif xdp.rxq = &rx_ring->xdp_rxq; while (likely(total_rx_packets < (unsigned int)budget)) { @@ -2365,9 +2353,12 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget) */ dma_rmb(); - rx_buffer = i40e_clean_programming_status(rx_ring, rx_desc, - qword); - if (unlikely(rx_buffer)) { + if (i40e_rx_is_programming_status(qword)) { + i40e_clean_programming_status(rx_ring, + rx_desc->raw.qword[0], + qword); + rx_buffer = i40e_rx_bi(rx_ring, rx_ring->next_to_clean); + i40e_inc_ntc(rx_ring); i40e_reuse_rx_page(rx_ring, rx_buffer); cleaned_count++; continue; @@ -2389,7 +2380,10 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget) xdp.data_hard_start = xdp.data - i40e_rx_offset(rx_ring); xdp.data_end = xdp.data + size; - +#if (PAGE_SIZE > 4096) + /* At larger PAGE_SIZE, frame_sz depend on len size */ + xdp.frame_sz = i40e_rx_frame_truesize(rx_ring, size); +#endif skb = i40e_run_xdp(rx_ring, &xdp); } diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h index 36d37f31a287..5c255977fd58 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h @@ -296,17 +296,9 @@ struct i40e_tx_buffer { struct i40e_rx_buffer { dma_addr_t dma; - union { - struct { - struct page *page; - __u32 page_offset; - __u16 pagecnt_bias; - }; - struct { - void *addr; - u64 handle; - }; - }; + struct page *page; + __u32 page_offset; + __u16 pagecnt_bias; }; struct i40e_queue_stats { @@ -358,6 +350,7 @@ struct i40e_ring { union { struct i40e_tx_buffer *tx_bi; struct i40e_rx_buffer *rx_bi; + struct xdp_buff **rx_bi_zc; }; DECLARE_BITMAP(state, __I40E_RING_STATE_NBITS); u16 queue_index; /* Queue number of ring */ @@ -419,7 +412,6 @@ struct i40e_ring { struct i40e_channel *ch; struct xdp_rxq_info xdp_rxq; struct xdp_umem *xsk_umem; - struct zero_copy_allocator zca; /* ZC allocator anchor */ } ____cacheline_internodealigned_in_smp; static inline bool ring_uses_build_skb(struct i40e_ring *ring) @@ -495,6 +487,7 @@ int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size); bool __i40e_chk_linearize(struct sk_buff *skb); int i40e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames, u32 flags); +int i40e_alloc_rx_bi(struct i40e_ring *rx_ring); /** * i40e_get_head - Retrieve head from head writeback diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h b/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h index 8af0e99c6c0d..667c4dc4b39f 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h +++ b/drivers/net/ethernet/intel/i40e/i40e_txrx_common.h @@ -4,13 +4,9 @@ #ifndef I40E_TXRX_COMMON_ #define I40E_TXRX_COMMON_ -void i40e_fd_handle_status(struct i40e_ring *rx_ring, - union i40e_rx_desc *rx_desc, u8 prog_id); int i40e_xmit_xdp_tx_ring(struct xdp_buff *xdp, struct i40e_ring *xdp_ring); -struct i40e_rx_buffer *i40e_clean_programming_status( - struct i40e_ring *rx_ring, - union i40e_rx_desc *rx_desc, - u64 qw); +void i40e_clean_programming_status(struct i40e_ring *rx_ring, u64 qword0_raw, + u64 qword1); void i40e_process_skb_fields(struct i40e_ring *rx_ring, union i40e_rx_desc *rx_desc, struct sk_buff *skb); void i40e_xdp_ring_update_tail(struct i40e_ring *xdp_ring); @@ -84,6 +80,38 @@ static inline void i40e_arm_wb(struct i40e_ring *tx_ring, } } +/** + * i40e_rx_is_programming_status - check for programming status descriptor + * @qword1: qword1 representing status_error_len in CPU ordering + * + * The value of in the descriptor length field indicate if this + * is a programming status descriptor for flow director or FCoE + * by the value of I40E_RX_PROG_STATUS_DESC_LENGTH, otherwise + * it is a packet descriptor. + **/ +static inline bool i40e_rx_is_programming_status(u64 qword1) +{ + /* The Rx filter programming status and SPH bit occupy the same + * spot in the descriptor. Since we don't support packet split we + * can just reuse the bit as an indication that this is a + * programming status descriptor. + */ + return qword1 & I40E_RXD_QW1_LENGTH_SPH_MASK; +} + +/** + * i40e_inc_ntc: Advance the next_to_clean index + * @rx_ring: Rx ring + **/ +static inline void i40e_inc_ntc(struct i40e_ring *rx_ring) +{ + u32 ntc = rx_ring->next_to_clean + 1; + + ntc = (ntc < rx_ring->count) ? ntc : 0; + rx_ring->next_to_clean = ntc; + prefetch(I40E_RX_DESC(rx_ring, ntc)); +} + void i40e_xsk_clean_rx_ring(struct i40e_ring *rx_ring); void i40e_xsk_clean_tx_ring(struct i40e_ring *tx_ring); bool i40e_xsk_any_rx_ring_enabled(struct i40e_vsi *vsi); diff --git a/drivers/net/ethernet/intel/i40e/i40e_type.h b/drivers/net/ethernet/intel/i40e/i40e_type.h index 6ea2867ff60f..63e098f7cb63 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_type.h +++ b/drivers/net/ethernet/intel/i40e/i40e_type.h @@ -689,7 +689,7 @@ union i40e_32byte_rx_desc { __le64 rsvd2; } read; struct { - struct { + struct i40e_32b_rx_wb_qw0 { struct { union { __le16 mirroring_status; @@ -727,6 +727,9 @@ union i40e_32byte_rx_desc { } hi_dword; } qword3; } wb; /* writeback */ + struct { + u64 qword[4]; + } raw; }; enum i40e_rx_desc_status_bits { diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.c b/drivers/net/ethernet/intel/i40e/i40e_xsk.c index 0b7d29192b2c..7276580cbe64 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_xsk.c +++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.c @@ -2,68 +2,30 @@ /* Copyright(c) 2018 Intel Corporation. */ #include <linux/bpf_trace.h> -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> #include <net/xdp.h> #include "i40e.h" #include "i40e_txrx_common.h" #include "i40e_xsk.h" -/** - * i40e_xsk_umem_dma_map - DMA maps all UMEM memory for the netdev - * @vsi: Current VSI - * @umem: UMEM to DMA map - * - * Returns 0 on success, <0 on failure - **/ -static int i40e_xsk_umem_dma_map(struct i40e_vsi *vsi, struct xdp_umem *umem) +int i40e_alloc_rx_bi_zc(struct i40e_ring *rx_ring) { - struct i40e_pf *pf = vsi->back; - struct device *dev; - unsigned int i, j; - dma_addr_t dma; - - dev = &pf->pdev->dev; - for (i = 0; i < umem->npgs; i++) { - dma = dma_map_page_attrs(dev, umem->pgs[i], 0, PAGE_SIZE, - DMA_BIDIRECTIONAL, I40E_RX_DMA_ATTR); - if (dma_mapping_error(dev, dma)) - goto out_unmap; + unsigned long sz = sizeof(*rx_ring->rx_bi_zc) * rx_ring->count; - umem->pages[i].dma = dma; - } - - return 0; - -out_unmap: - for (j = 0; j < i; j++) { - dma_unmap_page_attrs(dev, umem->pages[i].dma, PAGE_SIZE, - DMA_BIDIRECTIONAL, I40E_RX_DMA_ATTR); - umem->pages[i].dma = 0; - } - - return -1; + rx_ring->rx_bi_zc = kzalloc(sz, GFP_KERNEL); + return rx_ring->rx_bi_zc ? 0 : -ENOMEM; } -/** - * i40e_xsk_umem_dma_unmap - DMA unmaps all UMEM memory for the netdev - * @vsi: Current VSI - * @umem: UMEM to DMA map - **/ -static void i40e_xsk_umem_dma_unmap(struct i40e_vsi *vsi, struct xdp_umem *umem) +void i40e_clear_rx_bi_zc(struct i40e_ring *rx_ring) { - struct i40e_pf *pf = vsi->back; - struct device *dev; - unsigned int i; - - dev = &pf->pdev->dev; - - for (i = 0; i < umem->npgs; i++) { - dma_unmap_page_attrs(dev, umem->pages[i].dma, PAGE_SIZE, - DMA_BIDIRECTIONAL, I40E_RX_DMA_ATTR); + memset(rx_ring->rx_bi_zc, 0, + sizeof(*rx_ring->rx_bi_zc) * rx_ring->count); +} - umem->pages[i].dma = 0; - } +static struct xdp_buff **i40e_rx_bi(struct i40e_ring *rx_ring, u32 idx) +{ + return &rx_ring->rx_bi_zc[idx]; } /** @@ -78,7 +40,6 @@ static int i40e_xsk_umem_enable(struct i40e_vsi *vsi, struct xdp_umem *umem, u16 qid) { struct net_device *netdev = vsi->netdev; - struct xdp_umem_fq_reuse *reuseq; bool if_running; int err; @@ -92,13 +53,7 @@ static int i40e_xsk_umem_enable(struct i40e_vsi *vsi, struct xdp_umem *umem, qid >= netdev->real_num_tx_queues) return -EINVAL; - reuseq = xsk_reuseq_prepare(vsi->rx_rings[0]->count); - if (!reuseq) - return -ENOMEM; - - xsk_reuseq_free(xsk_reuseq_swap(umem, reuseq)); - - err = i40e_xsk_umem_dma_map(vsi, umem); + err = xsk_buff_dma_map(umem, &vsi->back->pdev->dev, I40E_RX_DMA_ATTR); if (err) return err; @@ -151,7 +106,7 @@ static int i40e_xsk_umem_disable(struct i40e_vsi *vsi, u16 qid) } clear_bit(qid, vsi->af_xdp_zc_qps); - i40e_xsk_umem_dma_unmap(vsi, umem); + xsk_buff_dma_unmap(umem, I40E_RX_DMA_ATTR); if (if_running) { err = i40e_queue_pair_enable(vsi, qid); @@ -184,17 +139,13 @@ int i40e_xsk_umem_setup(struct i40e_vsi *vsi, struct xdp_umem *umem, * @rx_ring: Rx ring * @xdp: xdp_buff used as input to the XDP program * - * This function enables or disables a UMEM to a certain ring. - * * Returns any of I40E_XDP_{PASS, CONSUMED, TX, REDIR} **/ static int i40e_run_xdp_zc(struct i40e_ring *rx_ring, struct xdp_buff *xdp) { - struct xdp_umem *umem = rx_ring->xsk_umem; int err, result = I40E_XDP_PASS; struct i40e_ring *xdp_ring; struct bpf_prog *xdp_prog; - u64 offset; u32 act; rcu_read_lock(); @@ -203,9 +154,6 @@ static int i40e_run_xdp_zc(struct i40e_ring *rx_ring, struct xdp_buff *xdp) */ xdp_prog = READ_ONCE(rx_ring->xdp_prog); act = bpf_prog_run_xdp(xdp_prog, xdp); - offset = xdp->data - xdp->data_hard_start; - - xdp->handle = xsk_umem_adjust_offset(umem, xdp->handle, offset); switch (act) { case XDP_PASS: @@ -232,107 +180,26 @@ static int i40e_run_xdp_zc(struct i40e_ring *rx_ring, struct xdp_buff *xdp) return result; } -/** - * i40e_alloc_buffer_zc - Allocates an i40e_rx_buffer - * @rx_ring: Rx ring - * @bi: Rx buffer to populate - * - * This function allocates an Rx buffer. The buffer can come from fill - * queue, or via the recycle queue (next_to_alloc). - * - * Returns true for a successful allocation, false otherwise - **/ -static bool i40e_alloc_buffer_zc(struct i40e_ring *rx_ring, - struct i40e_rx_buffer *bi) -{ - struct xdp_umem *umem = rx_ring->xsk_umem; - void *addr = bi->addr; - u64 handle, hr; - - if (addr) { - rx_ring->rx_stats.page_reuse_count++; - return true; - } - - if (!xsk_umem_peek_addr(umem, &handle)) { - rx_ring->rx_stats.alloc_page_failed++; - return false; - } - - hr = umem->headroom + XDP_PACKET_HEADROOM; - - bi->dma = xdp_umem_get_dma(umem, handle); - bi->dma += hr; - - bi->addr = xdp_umem_get_data(umem, handle); - bi->addr += hr; - - bi->handle = xsk_umem_adjust_offset(umem, handle, umem->headroom); - - xsk_umem_release_addr(umem); - return true; -} - -/** - * i40e_alloc_buffer_slow_zc - Allocates an i40e_rx_buffer - * @rx_ring: Rx ring - * @bi: Rx buffer to populate - * - * This function allocates an Rx buffer. The buffer can come from fill - * queue, or via the reuse queue. - * - * Returns true for a successful allocation, false otherwise - **/ -static bool i40e_alloc_buffer_slow_zc(struct i40e_ring *rx_ring, - struct i40e_rx_buffer *bi) -{ - struct xdp_umem *umem = rx_ring->xsk_umem; - u64 handle, hr; - - if (!xsk_umem_peek_addr_rq(umem, &handle)) { - rx_ring->rx_stats.alloc_page_failed++; - return false; - } - - handle &= rx_ring->xsk_umem->chunk_mask; - - hr = umem->headroom + XDP_PACKET_HEADROOM; - - bi->dma = xdp_umem_get_dma(umem, handle); - bi->dma += hr; - - bi->addr = xdp_umem_get_data(umem, handle); - bi->addr += hr; - - bi->handle = xsk_umem_adjust_offset(umem, handle, umem->headroom); - - xsk_umem_release_addr_rq(umem); - return true; -} - -static __always_inline bool -__i40e_alloc_rx_buffers_zc(struct i40e_ring *rx_ring, u16 count, - bool alloc(struct i40e_ring *rx_ring, - struct i40e_rx_buffer *bi)) +bool i40e_alloc_rx_buffers_zc(struct i40e_ring *rx_ring, u16 count) { u16 ntu = rx_ring->next_to_use; union i40e_rx_desc *rx_desc; - struct i40e_rx_buffer *bi; + struct xdp_buff **bi, *xdp; + dma_addr_t dma; bool ok = true; rx_desc = I40E_RX_DESC(rx_ring, ntu); - bi = &rx_ring->rx_bi[ntu]; + bi = i40e_rx_bi(rx_ring, ntu); do { - if (!alloc(rx_ring, bi)) { + xdp = xsk_buff_alloc(rx_ring->xsk_umem); + if (!xdp) { ok = false; goto no_buffers; } - - dma_sync_single_range_for_device(rx_ring->dev, bi->dma, 0, - rx_ring->rx_buf_len, - DMA_BIDIRECTIONAL); - - rx_desc->read.pkt_addr = cpu_to_le64(bi->dma); + *bi = xdp; + dma = xsk_buff_xdp_get_dma(xdp); + rx_desc->read.pkt_addr = cpu_to_le64(dma); + rx_desc->read.hdr_addr = 0; rx_desc++; bi++; @@ -340,11 +207,10 @@ __i40e_alloc_rx_buffers_zc(struct i40e_ring *rx_ring, u16 count, if (unlikely(ntu == rx_ring->count)) { rx_desc = I40E_RX_DESC(rx_ring, 0); - bi = rx_ring->rx_bi; + bi = i40e_rx_bi(rx_ring, 0); ntu = 0; } - rx_desc->wb.qword1.status_error_len = 0; count--; } while (count); @@ -356,127 +222,8 @@ no_buffers: } /** - * i40e_alloc_rx_buffers_zc - Allocates a number of Rx buffers - * @rx_ring: Rx ring - * @count: The number of buffers to allocate - * - * This function allocates a number of Rx buffers from the reuse queue - * or fill ring and places them on the Rx ring. - * - * Returns true for a successful allocation, false otherwise - **/ -bool i40e_alloc_rx_buffers_zc(struct i40e_ring *rx_ring, u16 count) -{ - return __i40e_alloc_rx_buffers_zc(rx_ring, count, - i40e_alloc_buffer_slow_zc); -} - -/** - * i40e_alloc_rx_buffers_fast_zc - Allocates a number of Rx buffers - * @rx_ring: Rx ring - * @count: The number of buffers to allocate - * - * This function allocates a number of Rx buffers from the fill ring - * or the internal recycle mechanism and places them on the Rx ring. - * - * Returns true for a successful allocation, false otherwise - **/ -static bool i40e_alloc_rx_buffers_fast_zc(struct i40e_ring *rx_ring, u16 count) -{ - return __i40e_alloc_rx_buffers_zc(rx_ring, count, - i40e_alloc_buffer_zc); -} - -/** - * i40e_get_rx_buffer_zc - Return the current Rx buffer - * @rx_ring: Rx ring - * @size: The size of the rx buffer (read from descriptor) - * - * This function returns the current, received Rx buffer, and also - * does DMA synchronization. the Rx ring. - * - * Returns the received Rx buffer - **/ -static struct i40e_rx_buffer *i40e_get_rx_buffer_zc(struct i40e_ring *rx_ring, - const unsigned int size) -{ - struct i40e_rx_buffer *bi; - - bi = &rx_ring->rx_bi[rx_ring->next_to_clean]; - - /* we are reusing so sync this buffer for CPU use */ - dma_sync_single_range_for_cpu(rx_ring->dev, - bi->dma, 0, - size, - DMA_BIDIRECTIONAL); - - return bi; -} - -/** - * i40e_reuse_rx_buffer_zc - Recycle an Rx buffer - * @rx_ring: Rx ring - * @old_bi: The Rx buffer to recycle - * - * This function recycles a finished Rx buffer, and places it on the - * recycle queue (next_to_alloc). - **/ -static void i40e_reuse_rx_buffer_zc(struct i40e_ring *rx_ring, - struct i40e_rx_buffer *old_bi) -{ - struct i40e_rx_buffer *new_bi = &rx_ring->rx_bi[rx_ring->next_to_alloc]; - u16 nta = rx_ring->next_to_alloc; - - /* update, and store next to alloc */ - nta++; - rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0; - - /* transfer page from old buffer to new buffer */ - new_bi->dma = old_bi->dma; - new_bi->addr = old_bi->addr; - new_bi->handle = old_bi->handle; - - old_bi->addr = NULL; -} - -/** - * i40e_zca_free - Free callback for MEM_TYPE_ZERO_COPY allocations - * @alloc: Zero-copy allocator - * @handle: Buffer handle - **/ -void i40e_zca_free(struct zero_copy_allocator *alloc, unsigned long handle) -{ - struct i40e_rx_buffer *bi; - struct i40e_ring *rx_ring; - u64 hr, mask; - u16 nta; - - rx_ring = container_of(alloc, struct i40e_ring, zca); - hr = rx_ring->xsk_umem->headroom + XDP_PACKET_HEADROOM; - mask = rx_ring->xsk_umem->chunk_mask; - - nta = rx_ring->next_to_alloc; - bi = &rx_ring->rx_bi[nta]; - - nta++; - rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0; - - handle &= mask; - - bi->dma = xdp_umem_get_dma(rx_ring->xsk_umem, handle); - bi->dma += hr; - - bi->addr = xdp_umem_get_data(rx_ring->xsk_umem, handle); - bi->addr += hr; - - bi->handle = xsk_umem_adjust_offset(rx_ring->xsk_umem, (u64)handle, - rx_ring->xsk_umem->headroom); -} - -/** - * i40e_construct_skb_zc - Create skbufff from zero-copy Rx buffer + * i40e_construct_skb_zc - Create skbuff from zero-copy Rx buffer * @rx_ring: Rx ring - * @bi: Rx buffer * @xdp: xdp_buff * * This functions allocates a new skb from a zero-copy Rx buffer. @@ -484,7 +231,6 @@ void i40e_zca_free(struct zero_copy_allocator *alloc, unsigned long handle) * Returns the skb, or NULL on failure. **/ static struct sk_buff *i40e_construct_skb_zc(struct i40e_ring *rx_ring, - struct i40e_rx_buffer *bi, struct xdp_buff *xdp) { unsigned int metasize = xdp->data - xdp->data_meta; @@ -503,24 +249,11 @@ static struct sk_buff *i40e_construct_skb_zc(struct i40e_ring *rx_ring, if (metasize) skb_metadata_set(skb, metasize); - i40e_reuse_rx_buffer_zc(rx_ring, bi); + xsk_buff_free(xdp); return skb; } /** - * i40e_inc_ntc: Advance the next_to_clean index - * @rx_ring: Rx ring - **/ -static void i40e_inc_ntc(struct i40e_ring *rx_ring) -{ - u32 ntc = rx_ring->next_to_clean + 1; - - ntc = (ntc < rx_ring->count) ? ntc : 0; - rx_ring->next_to_clean = ntc; - prefetch(I40E_RX_DESC(rx_ring, ntc)); -} - -/** * i40e_clean_rx_irq_zc - Consumes Rx packets from the hardware ring * @rx_ring: Rx ring * @budget: NAPI budget @@ -534,20 +267,17 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget) unsigned int xdp_res, xdp_xmit = 0; bool failure = false; struct sk_buff *skb; - struct xdp_buff xdp; - - xdp.rxq = &rx_ring->xdp_rxq; while (likely(total_rx_packets < (unsigned int)budget)) { - struct i40e_rx_buffer *bi; union i40e_rx_desc *rx_desc; + struct xdp_buff **bi; unsigned int size; u64 qword; if (cleaned_count >= I40E_RX_BUFFER_WRITE) { failure = failure || - !i40e_alloc_rx_buffers_fast_zc(rx_ring, - cleaned_count); + !i40e_alloc_rx_buffers_zc(rx_ring, + cleaned_count); cleaned_count = 0; } @@ -560,35 +290,36 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget) */ dma_rmb(); - bi = i40e_clean_programming_status(rx_ring, rx_desc, - qword); - if (unlikely(bi)) { - i40e_reuse_rx_buffer_zc(rx_ring, bi); + if (i40e_rx_is_programming_status(qword)) { + i40e_clean_programming_status(rx_ring, + rx_desc->raw.qword[0], + qword); + bi = i40e_rx_bi(rx_ring, rx_ring->next_to_clean); + xsk_buff_free(*bi); + *bi = NULL; cleaned_count++; + i40e_inc_ntc(rx_ring); continue; } + bi = i40e_rx_bi(rx_ring, rx_ring->next_to_clean); size = (qword & I40E_RXD_QW1_LENGTH_PBUF_MASK) >> I40E_RXD_QW1_LENGTH_PBUF_SHIFT; if (!size) break; - bi = i40e_get_rx_buffer_zc(rx_ring, size); - xdp.data = bi->addr; - xdp.data_meta = xdp.data; - xdp.data_hard_start = xdp.data - XDP_PACKET_HEADROOM; - xdp.data_end = xdp.data + size; - xdp.handle = bi->handle; + bi = i40e_rx_bi(rx_ring, rx_ring->next_to_clean); + (*bi)->data_end = (*bi)->data + size; + xsk_buff_dma_sync_for_cpu(*bi); - xdp_res = i40e_run_xdp_zc(rx_ring, &xdp); + xdp_res = i40e_run_xdp_zc(rx_ring, *bi); if (xdp_res) { - if (xdp_res & (I40E_XDP_TX | I40E_XDP_REDIR)) { + if (xdp_res & (I40E_XDP_TX | I40E_XDP_REDIR)) xdp_xmit |= xdp_res; - bi->addr = NULL; - } else { - i40e_reuse_rx_buffer_zc(rx_ring, bi); - } + else + xsk_buff_free(*bi); + *bi = NULL; total_rx_bytes += size; total_rx_packets++; @@ -604,7 +335,8 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget) * BIT(I40E_RXD_QW1_ERROR_SHIFT). This is due to that * SBP is *not* set in PRT_SBPVSI (default not set). */ - skb = i40e_construct_skb_zc(rx_ring, bi, &xdp); + skb = i40e_construct_skb_zc(rx_ring, *bi); + *bi = NULL; if (!skb) { rx_ring->rx_stats.alloc_buff_failed++; break; @@ -662,10 +394,9 @@ static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget) if (!xsk_umem_consume_tx(xdp_ring->xsk_umem, &desc)) break; - dma = xdp_umem_get_dma(xdp_ring->xsk_umem, desc.addr); - - dma_sync_single_for_device(xdp_ring->dev, dma, desc.len, - DMA_BIDIRECTIONAL); + dma = xsk_buff_raw_get_dma(xdp_ring->xsk_umem, desc.addr); + xsk_buff_raw_dma_sync_for_device(xdp_ring->xsk_umem, dma, + desc.len); tx_bi = &xdp_ring->tx_bi[xdp_ring->next_to_use]; tx_bi->bytecount = desc.len; @@ -824,13 +555,13 @@ void i40e_xsk_clean_rx_ring(struct i40e_ring *rx_ring) u16 i; for (i = 0; i < rx_ring->count; i++) { - struct i40e_rx_buffer *rx_bi = &rx_ring->rx_bi[i]; + struct xdp_buff *rx_bi = *i40e_rx_bi(rx_ring, i); - if (!rx_bi->addr) + if (!rx_bi) continue; - xsk_umem_fq_reuse(rx_ring->xsk_umem, rx_bi->handle); - rx_bi->addr = NULL; + xsk_buff_free(rx_bi); + rx_bi = NULL; } } diff --git a/drivers/net/ethernet/intel/i40e/i40e_xsk.h b/drivers/net/ethernet/intel/i40e/i40e_xsk.h index 9ed59c14eb55..ea919a7d60ec 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_xsk.h +++ b/drivers/net/ethernet/intel/i40e/i40e_xsk.h @@ -12,12 +12,13 @@ int i40e_queue_pair_disable(struct i40e_vsi *vsi, int queue_pair); int i40e_queue_pair_enable(struct i40e_vsi *vsi, int queue_pair); int i40e_xsk_umem_setup(struct i40e_vsi *vsi, struct xdp_umem *umem, u16 qid); -void i40e_zca_free(struct zero_copy_allocator *alloc, unsigned long handle); bool i40e_alloc_rx_buffers_zc(struct i40e_ring *rx_ring, u16 cleaned_count); int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget); bool i40e_clean_xdp_tx_irq(struct i40e_vsi *vsi, struct i40e_ring *tx_ring, int napi_budget); int i40e_xsk_wakeup(struct net_device *dev, u32 queue_id, u32 flags); +int i40e_alloc_rx_bi_zc(struct i40e_ring *rx_ring); +void i40e_clear_rx_bi_zc(struct i40e_ring *rx_ring); #endif /* _I40E_XSK_H_ */ diff --git a/drivers/net/ethernet/intel/ice/Makefile b/drivers/net/ethernet/intel/ice/Makefile index 29c6c6743450..980bbcc64b4b 100644 --- a/drivers/net/ethernet/intel/ice/Makefile +++ b/drivers/net/ethernet/intel/ice/Makefile @@ -17,10 +17,14 @@ ice-y := ice_main.o \ ice_lib.o \ ice_txrx_lib.o \ ice_txrx.o \ + ice_fltr.o \ + ice_fdir.o \ + ice_ethtool_fdir.o \ ice_flex_pipe.o \ ice_flow.o \ ice_devlink.o \ ice_ethtool.o ice-$(CONFIG_PCI_IOV) += ice_virtchnl_pf.o ice_sriov.o ice-$(CONFIG_DCB) += ice_dcb.o ice_dcb_nl.o ice_dcb_lib.o +ice-$(CONFIG_RFS_ACCEL) += ice_arfs.o ice-$(CONFIG_XDP_SOCKETS) += ice_xsk.o diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 5c11448bfbb3..5792ee616b5c 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -34,9 +34,14 @@ #include <linux/ctype.h> #include <linux/bpf.h> #include <linux/avf/virtchnl.h> +#include <linux/cpu_rmap.h> #include <net/devlink.h> #include <net/ipv6.h> #include <net/xdp_sock.h> +#include <net/geneve.h> +#include <net/gre.h> +#include <net/udp_tunnel.h> +#include <net/vxlan.h> #include "ice_devids.h" #include "ice_type.h" #include "ice_txrx.h" @@ -46,7 +51,9 @@ #include "ice_sched.h" #include "ice_virtchnl_pf.h" #include "ice_sriov.h" +#include "ice_fdir.h" #include "ice_xsk.h" +#include "ice_arfs.h" extern const char ice_drv_ver[]; #define ICE_BAR0 0 @@ -62,6 +69,7 @@ extern const char ice_drv_ver[]; #define ICE_AQ_LEN 64 #define ICE_MBXSQ_LEN 64 #define ICE_MIN_MSIX 2 +#define ICE_FDIR_MSIX 1 #define ICE_NO_VSI 0xffff #define ICE_VSI_MAP_CONTIG 0 #define ICE_VSI_MAP_SCATTER 1 @@ -90,6 +98,7 @@ extern const char ice_drv_ver[]; #define ICE_TX_DESC(R, i) (&(((struct ice_tx_desc *)((R)->desc))[i])) #define ICE_RX_DESC(R, i) (&(((union ice_32b_rx_flex_desc *)((R)->desc))[i])) #define ICE_TX_CTX_DESC(R, i) (&(((struct ice_tx_ctx_desc *)((R)->desc))[i])) +#define ICE_TX_FDIRDESC(R, i) (&(((struct ice_fltr_desc *)((R)->desc))[i])) /* Macro for each VSI in a PF */ #define ice_for_each_vsi(pf, i) \ @@ -210,6 +219,7 @@ enum ice_state { __ICE_CFG_BUSY, __ICE_SERVICE_SCHED, __ICE_SERVICE_DIS, + __ICE_FD_FLUSH_REQ, __ICE_OICR_INTR_DIS, /* Global OICR interrupt disabled */ __ICE_MDD_VF_PRINT_PENDING, /* set when MDD event handle */ __ICE_VF_RESETS_DISABLED, /* disable resets during ice_remove */ @@ -244,8 +254,8 @@ struct ice_vsi { u32 tx_busy; u32 rx_buf_failed; u32 rx_page_failed; - int num_q_vectors; - int base_vector; /* IRQ base for OS reserved vectors */ + u16 num_q_vectors; + u16 base_vector; /* IRQ base for OS reserved vectors */ enum ice_vsi_type type; u16 vsi_num; /* HW (absolute) index of this VSI */ u16 idx; /* software index in pf->vsi[] */ @@ -253,6 +263,8 @@ struct ice_vsi { s16 vf_id; /* VF ID for SR-IOV VSIs */ u16 ethtype; /* Ethernet protocol for pause frame */ + u16 num_gfltr; + u16 num_bfltr; /* RSS config */ u16 rss_table_size; /* HW RSS table size */ @@ -261,6 +273,14 @@ struct ice_vsi { u8 *rss_lut_user; /* User configured lookup table entries */ u8 rss_lut_type; /* used to configure Get/Set RSS LUT AQ call */ + /* aRFS members only allocated for the PF VSI */ +#define ICE_MAX_ARFS_LIST 1024 +#define ICE_ARFS_LST_MASK (ICE_MAX_ARFS_LIST - 1) + struct hlist_head *arfs_fltr_list; + struct ice_arfs_active_fltr_cntrs *arfs_fltr_cntrs; + spinlock_t arfs_lock; /* protects aRFS hash table and filter state */ + atomic_t *arfs_last_fltr_id; + u16 max_frame; u16 rx_buf_len; @@ -335,12 +355,14 @@ enum ice_pf_flags { ICE_FLAG_SRIOV_CAPABLE, ICE_FLAG_DCB_CAPABLE, ICE_FLAG_DCB_ENA, + ICE_FLAG_FD_ENA, ICE_FLAG_ADV_FEATURES, ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA, ICE_FLAG_NO_MEDIA, ICE_FLAG_FW_LLDP_AGENT, ICE_FLAG_ETHTOOL_CTXT, /* set when ethtool holds RTNL lock */ ICE_FLAG_LEGACY_RX, + ICE_FLAG_VF_TRUE_PROMISC_ENA, ICE_FLAG_MDD_AUTO_RESET_VF, ICE_PF_FLAGS_NBITS /* must be last */ }; @@ -362,11 +384,13 @@ struct ice_pf { */ u16 sriov_base_vector; + u16 ctrl_vsi_idx; /* control VSI index in pf->vsi array */ + struct ice_vsi **vsi; /* VSIs created by the driver */ struct ice_sw *first_sw; /* first switch created by firmware */ /* Virtchnl/SR-IOV config info */ struct ice_vf *vf; - int num_alloc_vfs; /* actual number of VFs allocated */ + u16 num_alloc_vfs; /* actual number of VFs allocated */ u16 num_vfs_supported; /* num VFs supported for this PF */ u16 num_qps_per_vf; u16 num_msix_per_vf; @@ -385,11 +409,11 @@ struct ice_pf { struct mutex tc_mutex; /* lock to protect TC changes */ u32 msg_enable; u32 hw_csum_rx_error; - u32 oicr_idx; /* Other interrupt cause MSIX vector index */ - u32 num_avail_sw_msix; /* remaining MSIX SW vectors left unclaimed */ + u16 oicr_idx; /* Other interrupt cause MSIX vector index */ + u16 num_avail_sw_msix; /* remaining MSIX SW vectors left unclaimed */ u16 max_pf_txqs; /* Total Tx queues PF wide */ u16 max_pf_rxqs; /* Total Rx queues PF wide */ - u32 num_lan_msix; /* Total MSIX vectors for base driver */ + u16 num_lan_msix; /* Total MSIX vectors for base driver */ u16 num_lan_tx; /* num LAN Tx queues setup */ u16 num_lan_rx; /* num LAN Rx queues setup */ u16 next_vsi; /* Next free slot in pf->vsi[] - 0-based! */ @@ -500,8 +524,27 @@ static inline struct ice_vsi *ice_get_main_vsi(struct ice_pf *pf) return NULL; } +/** + * ice_get_ctrl_vsi - Get the control VSI + * @pf: PF instance + */ +static inline struct ice_vsi *ice_get_ctrl_vsi(struct ice_pf *pf) +{ + /* if pf->ctrl_vsi_idx is ICE_NO_VSI, control VSI was not set up */ + if (!pf->vsi || pf->ctrl_vsi_idx == ICE_NO_VSI) + return NULL; + + return pf->vsi[pf->ctrl_vsi_idx]; +} + +#define ICE_FD_STAT_CTR_BLOCK_COUNT 256 +#define ICE_FD_STAT_PF_IDX(base_idx) \ + ((base_idx) * ICE_FD_STAT_CTR_BLOCK_COUNT) +#define ICE_FD_SB_STAT_IDX(base_idx) ICE_FD_STAT_PF_IDX(base_idx) + int ice_vsi_setup_tx_rings(struct ice_vsi *vsi); int ice_vsi_setup_rx_rings(struct ice_vsi *vsi); +int ice_vsi_open_ctrl(struct ice_vsi *vsi); void ice_set_ethtool_ops(struct net_device *netdev); void ice_set_ethtool_safe_mode_ops(struct net_device *netdev); u16 ice_get_avail_txq_count(struct ice_pf *pf); @@ -523,7 +566,24 @@ int ice_get_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size); void ice_fill_rss_lut(u8 *lut, u16 rss_table_size, u16 rss_size); int ice_schedule_reset(struct ice_pf *pf, enum ice_reset_req reset); void ice_print_link_msg(struct ice_vsi *vsi, bool isup); +const char *ice_stat_str(enum ice_status stat_err); +const char *ice_aq_str(enum ice_aq_err aq_err); +int +ice_fdir_write_fltr(struct ice_pf *pf, struct ice_fdir_fltr *input, bool add, + bool is_tun); +void ice_vsi_manage_fdir(struct ice_vsi *vsi, bool ena); +int ice_add_fdir_ethtool(struct ice_vsi *vsi, struct ethtool_rxnfc *cmd); +int ice_del_fdir_ethtool(struct ice_vsi *vsi, struct ethtool_rxnfc *cmd); +int ice_get_ethtool_fdir_entry(struct ice_hw *hw, struct ethtool_rxnfc *cmd); +int +ice_get_fdir_fltr_ids(struct ice_hw *hw, struct ethtool_rxnfc *cmd, + u32 *rule_locs); +void ice_fdir_release_flows(struct ice_hw *hw); +void ice_fdir_replay_flows(struct ice_hw *hw); +void ice_fdir_replay_fltrs(struct ice_pf *pf); +int ice_fdir_create_dflt_rules(struct ice_pf *pf); int ice_open(struct net_device *netdev); int ice_stop(struct net_device *netdev); +void ice_service_task_schedule(struct ice_pf *pf); #endif /* _ICE_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h index 2381b4014ed6..92f82f2a8af4 100644 --- a/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h +++ b/drivers/net/ethernet/intel/ice/ice_adminq_cmd.h @@ -107,6 +107,7 @@ struct ice_aqc_list_caps_elem { #define ICE_AQC_CAPS_RXQS 0x0041 #define ICE_AQC_CAPS_TXQS 0x0042 #define ICE_AQC_CAPS_MSIX 0x0043 +#define ICE_AQC_CAPS_FD 0x0045 #define ICE_AQC_CAPS_MAX_MTU 0x0047 u8 major_ver; @@ -155,13 +156,11 @@ struct ice_aqc_manage_mac_write { #define ICE_AQC_MAN_MAC_WR_MC_MAG_EN BIT(0) #define ICE_AQC_MAN_MAC_WR_WOL_LAA_PFR_KEEP BIT(1) #define ICE_AQC_MAN_MAC_WR_S 6 -#define ICE_AQC_MAN_MAC_WR_M (3 << ICE_AQC_MAN_MAC_WR_S) +#define ICE_AQC_MAN_MAC_WR_M ICE_M(3, ICE_AQC_MAN_MAC_WR_S) #define ICE_AQC_MAN_MAC_UPDATE_LAA 0 -#define ICE_AQC_MAN_MAC_UPDATE_LAA_WOL (BIT(0) << ICE_AQC_MAN_MAC_WR_S) - /* High 16 bits of MAC address in big endian order */ - __be16 sah; - /* Low 32 bits of MAC address in big endian order */ - __be32 sal; +#define ICE_AQC_MAN_MAC_UPDATE_LAA_WOL BIT(ICE_AQC_MAN_MAC_WR_S) + /* byte stream in network order */ + u8 mac_addr[ETH_ALEN]; __le32 addr_high; __le32 addr_low; }; @@ -232,6 +231,11 @@ struct ice_aqc_get_sw_cfg_resp { */ #define ICE_AQC_RES_TYPE_VSI_LIST_REP 0x03 #define ICE_AQC_RES_TYPE_VSI_LIST_PRUNE 0x04 +#define ICE_AQC_RES_TYPE_FDIR_COUNTER_BLOCK 0x21 +#define ICE_AQC_RES_TYPE_FDIR_GUARANTEED_ENTRIES 0x22 +#define ICE_AQC_RES_TYPE_FDIR_SHARED_ENTRIES 0x23 +#define ICE_AQC_RES_TYPE_FD_PROF_BLDR_PROFID 0x58 +#define ICE_AQC_RES_TYPE_FD_PROF_BLDR_TCAM 0x59 #define ICE_AQC_RES_TYPE_HASH_PROF_BLDR_PROFID 0x60 #define ICE_AQC_RES_TYPE_HASH_PROF_BLDR_TCAM 0x61 @@ -240,6 +244,9 @@ struct ice_aqc_get_sw_cfg_resp { #define ICE_AQC_RES_TYPE_FLAG_DEDICATED 0x00 +#define ICE_AQC_RES_TYPE_S 0 +#define ICE_AQC_RES_TYPE_M (0x07F << ICE_AQC_RES_TYPE_S) + /* Allocate Resources command (indirect 0x0208) * Free Resources command (indirect 0x0209) */ @@ -541,7 +548,7 @@ struct ice_sw_rule_lkup_rx_tx { #define ICE_SINGLE_ACT_OTHER_ACTS 0x3 #define ICE_SINGLE_OTHER_ACT_IDENTIFIER_S 17 #define ICE_SINGLE_OTHER_ACT_IDENTIFIER_M \ - (0x3 << \ ICE_SINGLE_OTHER_ACT_IDENTIFIER_S) + (0x3 << ICE_SINGLE_OTHER_ACT_IDENTIFIER_S) /* Bit 17:18 - Defines other actions */ /* Other action = 0 - Mirror VSI */ @@ -967,7 +974,7 @@ struct ice_aqc_get_phy_caps { #define ICE_PHY_TYPE_HIGH_100G_CAUI2 BIT_ULL(2) #define ICE_PHY_TYPE_HIGH_100G_AUI2_AOC_ACC BIT_ULL(3) #define ICE_PHY_TYPE_HIGH_100G_AUI2 BIT_ULL(4) -#define ICE_PHY_TYPE_HIGH_MAX_INDEX 19 +#define ICE_PHY_TYPE_HIGH_MAX_INDEX 5 struct ice_aqc_get_phy_caps_data { __le64 phy_type_low; /* Use values from ICE_PHY_TYPE_LOW_* */ @@ -1059,6 +1066,25 @@ struct ice_aqc_set_phy_cfg_data { u8 rsvd1; }; +/* Set MAC Config command data structure (direct 0x0603) */ +struct ice_aqc_set_mac_cfg { + __le16 max_frame_size; + u8 params; +#define ICE_AQ_SET_MAC_PACE_S 3 +#define ICE_AQ_SET_MAC_PACE_M (0xF << ICE_AQ_SET_MAC_PACE_S) +#define ICE_AQ_SET_MAC_PACE_TYPE_M BIT(7) +#define ICE_AQ_SET_MAC_PACE_TYPE_RATE 0 +#define ICE_AQ_SET_MAC_PACE_TYPE_FIXED ICE_AQ_SET_MAC_PACE_TYPE_M + u8 tx_tmr_priority; + __le16 tx_tmr_value; + __le16 fc_refresh_threshold; + u8 drop_opts; +#define ICE_AQ_SET_MAC_AUTO_DROP_MASK BIT(0) +#define ICE_AQ_SET_MAC_AUTO_DROP_NONE 0 +#define ICE_AQ_SET_MAC_AUTO_DROP_BLOCKING_PKTS BIT(0) + u8 reserved[7]; +}; + /* Restart AN command data structure (direct 0x0605) * Also used for response, with only the lport_num field present. */ @@ -1264,6 +1290,33 @@ struct ice_aqc_nvm_checksum { u8 rsvd2[12]; }; +/* The result of netlist NVM read comes in a TLV format. The actual data + * (netlist header) starts from word offset 1 (byte 2). The FW strips + * out the type field from the TLV header so all the netlist fields + * should adjust their offset value by 1 word (2 bytes) in order to map + * their correct location. + */ +#define ICE_AQC_NVM_LINK_TOPO_NETLIST_MOD_ID 0x11B +#define ICE_AQC_NVM_LINK_TOPO_NETLIST_LEN_OFFSET 1 +#define ICE_AQC_NVM_LINK_TOPO_NETLIST_LEN 2 /* In bytes */ +#define ICE_AQC_NVM_NETLIST_NODE_COUNT_OFFSET 2 +#define ICE_AQC_NVM_NETLIST_NODE_COUNT_LEN 2 /* In bytes */ +#define ICE_AQC_NVM_NETLIST_NODE_COUNT_M ICE_M(0x3FF, 0) +#define ICE_AQC_NVM_NETLIST_ID_BLK_START_OFFSET 5 +#define ICE_AQC_NVM_NETLIST_ID_BLK_LEN 0x30 /* In words */ + +/* netlist ID block field offsets (word offsets) */ +#define ICE_AQC_NVM_NETLIST_ID_BLK_MAJOR_VER_LOW 2 +#define ICE_AQC_NVM_NETLIST_ID_BLK_MAJOR_VER_HIGH 3 +#define ICE_AQC_NVM_NETLIST_ID_BLK_MINOR_VER_LOW 4 +#define ICE_AQC_NVM_NETLIST_ID_BLK_MINOR_VER_HIGH 5 +#define ICE_AQC_NVM_NETLIST_ID_BLK_TYPE_LOW 6 +#define ICE_AQC_NVM_NETLIST_ID_BLK_TYPE_HIGH 7 +#define ICE_AQC_NVM_NETLIST_ID_BLK_REV_LOW 8 +#define ICE_AQC_NVM_NETLIST_ID_BLK_REV_HIGH 9 +#define ICE_AQC_NVM_NETLIST_ID_BLK_SHA_HASH 0xA +#define ICE_AQC_NVM_NETLIST_ID_BLK_CUST_VER 0x2F + /** * Send to PF command (indirect 0x0801) ID is only used by PF * @@ -1648,10 +1701,12 @@ struct ice_pkg_ver { }; #define ICE_PKG_NAME_SIZE 32 +#define ICE_SEG_NAME_SIZE 28 struct ice_aqc_get_pkg_info { struct ice_pkg_ver ver; - char name[ICE_PKG_NAME_SIZE]; + char name[ICE_SEG_NAME_SIZE]; + __le32 track_id; u8 is_in_nvm; u8 is_active; u8 is_active_at_boot; @@ -1738,6 +1793,7 @@ struct ice_aq_desc { struct ice_aqc_download_pkg download_pkg; struct ice_aqc_set_mac_lb set_mac_lb; struct ice_aqc_alloc_free_res_cmd sw_res_ctrl; + struct ice_aqc_set_mac_cfg set_mac_cfg; struct ice_aqc_set_event_mask set_event_mask; struct ice_aqc_get_link_status get_link_status; struct ice_aqc_event_lan_overflow lan_overflow; @@ -1770,6 +1826,7 @@ enum ice_aq_err { ICE_AQ_RC_EINVAL = 14, /* Invalid argument */ ICE_AQ_RC_ENOSPC = 16, /* No space left or allocation failure */ ICE_AQ_RC_ENOSYS = 17, /* Function not implemented */ + ICE_AQ_RC_EMODE = 21, /* Op not allowed in current dev mode */ ICE_AQ_RC_ENOSEC = 24, /* Missing security manifest */ ICE_AQ_RC_EBADSIG = 25, /* Bad RSA signature */ ICE_AQ_RC_ESVN = 26, /* SVN number prohibits this package */ @@ -1834,6 +1891,7 @@ enum ice_adminq_opc { /* PHY commands */ ice_aqc_opc_get_phy_caps = 0x0600, ice_aqc_opc_set_phy_cfg = 0x0601, + ice_aqc_opc_set_mac_cfg = 0x0603, ice_aqc_opc_restart_an = 0x0605, ice_aqc_opc_get_link_status = 0x0607, ice_aqc_opc_set_event_mask = 0x0613, diff --git a/drivers/net/ethernet/intel/ice/ice_arfs.c b/drivers/net/ethernet/intel/ice/ice_arfs.c new file mode 100644 index 000000000000..6560acd76c94 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_arfs.c @@ -0,0 +1,663 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (C) 2018-2020, Intel Corporation. */ + +#include "ice.h" + +/** + * ice_is_arfs_active - helper to check is aRFS is active + * @vsi: VSI to check + */ +static bool ice_is_arfs_active(struct ice_vsi *vsi) +{ + return !!vsi->arfs_fltr_list; +} + +/** + * ice_is_arfs_using_perfect_flow - check if aRFS has active perfect filters + * @hw: pointer to the HW structure + * @flow_type: flow type as Flow Director understands it + * + * Flow Director will query this function to see if aRFS is currently using + * the specified flow_type for perfect (4-tuple) filters. + */ +bool +ice_is_arfs_using_perfect_flow(struct ice_hw *hw, enum ice_fltr_ptype flow_type) +{ + struct ice_arfs_active_fltr_cntrs *arfs_fltr_cntrs; + struct ice_pf *pf = hw->back; + struct ice_vsi *vsi; + + vsi = ice_get_main_vsi(pf); + if (!vsi) + return false; + + arfs_fltr_cntrs = vsi->arfs_fltr_cntrs; + + /* active counters can be updated by multiple CPUs */ + smp_mb__before_atomic(); + switch (flow_type) { + case ICE_FLTR_PTYPE_NONF_IPV4_UDP: + return atomic_read(&arfs_fltr_cntrs->active_udpv4_cnt) > 0; + case ICE_FLTR_PTYPE_NONF_IPV6_UDP: + return atomic_read(&arfs_fltr_cntrs->active_udpv6_cnt) > 0; + case ICE_FLTR_PTYPE_NONF_IPV4_TCP: + return atomic_read(&arfs_fltr_cntrs->active_tcpv4_cnt) > 0; + case ICE_FLTR_PTYPE_NONF_IPV6_TCP: + return atomic_read(&arfs_fltr_cntrs->active_tcpv6_cnt) > 0; + default: + return false; + } +} + +/** + * ice_arfs_update_active_fltr_cntrs - update active filter counters for aRFS + * @vsi: VSI that aRFS is active on + * @entry: aRFS entry used to change counters + * @add: true to increment counter, false to decrement + */ +static void +ice_arfs_update_active_fltr_cntrs(struct ice_vsi *vsi, + struct ice_arfs_entry *entry, bool add) +{ + struct ice_arfs_active_fltr_cntrs *fltr_cntrs = vsi->arfs_fltr_cntrs; + + switch (entry->fltr_info.flow_type) { + case ICE_FLTR_PTYPE_NONF_IPV4_TCP: + if (add) + atomic_inc(&fltr_cntrs->active_tcpv4_cnt); + else + atomic_dec(&fltr_cntrs->active_tcpv4_cnt); + break; + case ICE_FLTR_PTYPE_NONF_IPV6_TCP: + if (add) + atomic_inc(&fltr_cntrs->active_tcpv6_cnt); + else + atomic_dec(&fltr_cntrs->active_tcpv6_cnt); + break; + case ICE_FLTR_PTYPE_NONF_IPV4_UDP: + if (add) + atomic_inc(&fltr_cntrs->active_udpv4_cnt); + else + atomic_dec(&fltr_cntrs->active_udpv4_cnt); + break; + case ICE_FLTR_PTYPE_NONF_IPV6_UDP: + if (add) + atomic_inc(&fltr_cntrs->active_udpv6_cnt); + else + atomic_dec(&fltr_cntrs->active_udpv6_cnt); + break; + default: + dev_err(ice_pf_to_dev(vsi->back), "aRFS: Failed to update filter counters, invalid filter type %d\n", + entry->fltr_info.flow_type); + } +} + +/** + * ice_arfs_del_flow_rules - delete the rules passed in from HW + * @vsi: VSI for the flow rules that need to be deleted + * @del_list_head: head of the list of ice_arfs_entry(s) for rule deletion + * + * Loop through the delete list passed in and remove the rules from HW. After + * each rule is deleted, disconnect and free the ice_arfs_entry because it is no + * longer being referenced by the aRFS hash table. + */ +static void +ice_arfs_del_flow_rules(struct ice_vsi *vsi, struct hlist_head *del_list_head) +{ + struct ice_arfs_entry *e; + struct hlist_node *n; + struct device *dev; + + dev = ice_pf_to_dev(vsi->back); + + hlist_for_each_entry_safe(e, n, del_list_head, list_entry) { + int result; + + result = ice_fdir_write_fltr(vsi->back, &e->fltr_info, false, + false); + if (!result) + ice_arfs_update_active_fltr_cntrs(vsi, e, false); + else + dev_dbg(dev, "Unable to delete aRFS entry, err %d fltr_state %d fltr_id %d flow_id %d Q %d\n", + result, e->fltr_state, e->fltr_info.fltr_id, + e->flow_id, e->fltr_info.q_index); + + /* The aRFS hash table is no longer referencing this entry */ + hlist_del(&e->list_entry); + devm_kfree(dev, e); + } +} + +/** + * ice_arfs_add_flow_rules - add the rules passed in from HW + * @vsi: VSI for the flow rules that need to be added + * @add_list_head: head of the list of ice_arfs_entry_ptr(s) for rule addition + * + * Loop through the add list passed in and remove the rules from HW. After each + * rule is added, disconnect and free the ice_arfs_entry_ptr node. Don't free + * the ice_arfs_entry(s) because they are still being referenced in the aRFS + * hash table. + */ +static void +ice_arfs_add_flow_rules(struct ice_vsi *vsi, struct hlist_head *add_list_head) +{ + struct ice_arfs_entry_ptr *ep; + struct hlist_node *n; + struct device *dev; + + dev = ice_pf_to_dev(vsi->back); + + hlist_for_each_entry_safe(ep, n, add_list_head, list_entry) { + int result; + + result = ice_fdir_write_fltr(vsi->back, + &ep->arfs_entry->fltr_info, true, + false); + if (!result) + ice_arfs_update_active_fltr_cntrs(vsi, ep->arfs_entry, + true); + else + dev_dbg(dev, "Unable to add aRFS entry, err %d fltr_state %d fltr_id %d flow_id %d Q %d\n", + result, ep->arfs_entry->fltr_state, + ep->arfs_entry->fltr_info.fltr_id, + ep->arfs_entry->flow_id, + ep->arfs_entry->fltr_info.q_index); + + hlist_del(&ep->list_entry); + devm_kfree(dev, ep); + } +} + +/** + * ice_arfs_is_flow_expired - check if the aRFS entry has expired + * @vsi: VSI containing the aRFS entry + * @arfs_entry: aRFS entry that's being checked for expiration + * + * Return true if the flow has expired, else false. This function should be used + * to determine whether or not an aRFS entry should be removed from the hardware + * and software structures. + */ +static bool +ice_arfs_is_flow_expired(struct ice_vsi *vsi, struct ice_arfs_entry *arfs_entry) +{ +#define ICE_ARFS_TIME_DELTA_EXPIRATION msecs_to_jiffies(5000) + if (rps_may_expire_flow(vsi->netdev, arfs_entry->fltr_info.q_index, + arfs_entry->flow_id, + arfs_entry->fltr_info.fltr_id)) + return true; + + /* expiration timer only used for UDP filters */ + if (arfs_entry->fltr_info.flow_type != ICE_FLTR_PTYPE_NONF_IPV4_UDP && + arfs_entry->fltr_info.flow_type != ICE_FLTR_PTYPE_NONF_IPV6_UDP) + return false; + + return time_in_range64(arfs_entry->time_activated + + ICE_ARFS_TIME_DELTA_EXPIRATION, + arfs_entry->time_activated, get_jiffies_64()); +} + +/** + * ice_arfs_update_flow_rules - add/delete aRFS rules in HW + * @vsi: the VSI to be forwarded to + * @idx: index into the table of aRFS filter lists. Obtained from skb->hash + * @add_list: list to populate with filters to be added to Flow Director + * @del_list: list to populate with filters to be deleted from Flow Director + * + * Iterate over the hlist at the index given in the aRFS hash table and + * determine if there are any aRFS entries that need to be either added or + * deleted in the HW. If the aRFS entry is marked as ICE_ARFS_INACTIVE the + * filter needs to be added to HW, else if it's marked as ICE_ARFS_ACTIVE and + * the flow has expired delete the filter from HW. The caller of this function + * is expected to add/delete rules on the add_list/del_list respectively. + */ +static void +ice_arfs_update_flow_rules(struct ice_vsi *vsi, u16 idx, + struct hlist_head *add_list, + struct hlist_head *del_list) +{ + struct ice_arfs_entry *e; + struct hlist_node *n; + struct device *dev; + + dev = ice_pf_to_dev(vsi->back); + + /* go through the aRFS hlist at this idx and check for needed updates */ + hlist_for_each_entry_safe(e, n, &vsi->arfs_fltr_list[idx], list_entry) + /* check if filter needs to be added to HW */ + if (e->fltr_state == ICE_ARFS_INACTIVE) { + enum ice_fltr_ptype flow_type = e->fltr_info.flow_type; + struct ice_arfs_entry_ptr *ep = + devm_kzalloc(dev, sizeof(*ep), GFP_ATOMIC); + + if (!ep) + continue; + INIT_HLIST_NODE(&ep->list_entry); + /* reference aRFS entry to add HW filter */ + ep->arfs_entry = e; + hlist_add_head(&ep->list_entry, add_list); + e->fltr_state = ICE_ARFS_ACTIVE; + /* expiration timer only used for UDP flows */ + if (flow_type == ICE_FLTR_PTYPE_NONF_IPV4_UDP || + flow_type == ICE_FLTR_PTYPE_NONF_IPV6_UDP) + e->time_activated = get_jiffies_64(); + } else if (e->fltr_state == ICE_ARFS_ACTIVE) { + /* check if filter needs to be removed from HW */ + if (ice_arfs_is_flow_expired(vsi, e)) { + /* remove aRFS entry from hash table for delete + * and to prevent referencing it the next time + * through this hlist index + */ + hlist_del(&e->list_entry); + e->fltr_state = ICE_ARFS_TODEL; + /* save reference to aRFS entry for delete */ + hlist_add_head(&e->list_entry, del_list); + } + } +} + +/** + * ice_sync_arfs_fltrs - update all aRFS filters + * @pf: board private structure + */ +void ice_sync_arfs_fltrs(struct ice_pf *pf) +{ + HLIST_HEAD(tmp_del_list); + HLIST_HEAD(tmp_add_list); + struct ice_vsi *pf_vsi; + unsigned int i; + + pf_vsi = ice_get_main_vsi(pf); + if (!pf_vsi) + return; + + if (!ice_is_arfs_active(pf_vsi)) + return; + + spin_lock_bh(&pf_vsi->arfs_lock); + /* Once we process aRFS for the PF VSI get out */ + for (i = 0; i < ICE_MAX_ARFS_LIST; i++) + ice_arfs_update_flow_rules(pf_vsi, i, &tmp_add_list, + &tmp_del_list); + spin_unlock_bh(&pf_vsi->arfs_lock); + + /* use list of ice_arfs_entry(s) for delete */ + ice_arfs_del_flow_rules(pf_vsi, &tmp_del_list); + + /* use list of ice_arfs_entry_ptr(s) for add */ + ice_arfs_add_flow_rules(pf_vsi, &tmp_add_list); +} + +/** + * ice_arfs_build_entry - builds an aRFS entry based on input + * @vsi: destination VSI for this flow + * @fk: flow dissector keys for creating the tuple + * @rxq_idx: Rx queue to steer this flow to + * @flow_id: passed down from the stack and saved for flow expiration + * + * returns an aRFS entry on success and NULL on failure + */ +static struct ice_arfs_entry * +ice_arfs_build_entry(struct ice_vsi *vsi, const struct flow_keys *fk, + u16 rxq_idx, u32 flow_id) +{ + struct ice_arfs_entry *arfs_entry; + struct ice_fdir_fltr *fltr_info; + u8 ip_proto; + + arfs_entry = devm_kzalloc(ice_pf_to_dev(vsi->back), + sizeof(*arfs_entry), + GFP_ATOMIC | __GFP_NOWARN); + if (!arfs_entry) + return NULL; + + fltr_info = &arfs_entry->fltr_info; + fltr_info->q_index = rxq_idx; + fltr_info->dest_ctl = ICE_FLTR_PRGM_DESC_DEST_DIRECT_PKT_QINDEX; + fltr_info->dest_vsi = vsi->idx; + ip_proto = fk->basic.ip_proto; + + if (fk->basic.n_proto == htons(ETH_P_IP)) { + fltr_info->ip.v4.proto = ip_proto; + fltr_info->flow_type = (ip_proto == IPPROTO_TCP) ? + ICE_FLTR_PTYPE_NONF_IPV4_TCP : + ICE_FLTR_PTYPE_NONF_IPV4_UDP; + fltr_info->ip.v4.src_ip = fk->addrs.v4addrs.src; + fltr_info->ip.v4.dst_ip = fk->addrs.v4addrs.dst; + fltr_info->ip.v4.src_port = fk->ports.src; + fltr_info->ip.v4.dst_port = fk->ports.dst; + } else { /* ETH_P_IPV6 */ + fltr_info->ip.v6.proto = ip_proto; + fltr_info->flow_type = (ip_proto == IPPROTO_TCP) ? + ICE_FLTR_PTYPE_NONF_IPV6_TCP : + ICE_FLTR_PTYPE_NONF_IPV6_UDP; + memcpy(&fltr_info->ip.v6.src_ip, &fk->addrs.v6addrs.src, + sizeof(struct in6_addr)); + memcpy(&fltr_info->ip.v6.dst_ip, &fk->addrs.v6addrs.dst, + sizeof(struct in6_addr)); + fltr_info->ip.v6.src_port = fk->ports.src; + fltr_info->ip.v6.dst_port = fk->ports.dst; + } + + arfs_entry->flow_id = flow_id; + fltr_info->fltr_id = + atomic_inc_return(vsi->arfs_last_fltr_id) % RPS_NO_FILTER; + + return arfs_entry; +} + +/** + * ice_arfs_is_perfect_flow_set - Check to see if perfect flow is set + * @hw: pointer to HW structure + * @l3_proto: ETH_P_IP or ETH_P_IPV6 in network order + * @l4_proto: IPPROTO_UDP or IPPROTO_TCP + * + * We only support perfect (4-tuple) filters for aRFS. This function allows aRFS + * to check if perfect (4-tuple) flow rules are currently in place by Flow + * Director. + */ +static bool +ice_arfs_is_perfect_flow_set(struct ice_hw *hw, __be16 l3_proto, u8 l4_proto) +{ + unsigned long *perfect_fltr = hw->fdir_perfect_fltr; + + /* advanced Flow Director disabled, perfect filters always supported */ + if (!perfect_fltr) + return true; + + if (l3_proto == htons(ETH_P_IP) && l4_proto == IPPROTO_UDP) + return test_bit(ICE_FLTR_PTYPE_NONF_IPV4_UDP, perfect_fltr); + else if (l3_proto == htons(ETH_P_IP) && l4_proto == IPPROTO_TCP) + return test_bit(ICE_FLTR_PTYPE_NONF_IPV4_TCP, perfect_fltr); + else if (l3_proto == htons(ETH_P_IPV6) && l4_proto == IPPROTO_UDP) + return test_bit(ICE_FLTR_PTYPE_NONF_IPV6_UDP, perfect_fltr); + else if (l3_proto == htons(ETH_P_IPV6) && l4_proto == IPPROTO_TCP) + return test_bit(ICE_FLTR_PTYPE_NONF_IPV6_TCP, perfect_fltr); + + return false; +} + +/** + * ice_rx_flow_steer - steer the Rx flow to where application is being run + * @netdev: ptr to the netdev being adjusted + * @skb: buffer with required header information + * @rxq_idx: queue to which the flow needs to move + * @flow_id: flow identifier provided by the netdev + * + * Based on the skb, rxq_idx, and flow_id passed in add/update an entry in the + * aRFS hash table. Iterate over one of the hlists in the aRFS hash table and + * if the flow_id already exists in the hash table but the rxq_idx has changed + * mark the entry as ICE_ARFS_INACTIVE so it can get updated in HW, else + * if the entry is marked as ICE_ARFS_TODEL delete it from the aRFS hash table. + * If neither of the previous conditions are true then add a new entry in the + * aRFS hash table, which gets set to ICE_ARFS_INACTIVE by default so it can be + * added to HW. + */ +int +ice_rx_flow_steer(struct net_device *netdev, const struct sk_buff *skb, + u16 rxq_idx, u32 flow_id) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_arfs_entry *arfs_entry; + struct ice_vsi *vsi = np->vsi; + struct flow_keys fk; + struct ice_pf *pf; + __be16 n_proto; + u8 ip_proto; + u16 idx; + int ret; + + /* failed to allocate memory for aRFS so don't crash */ + if (unlikely(!vsi->arfs_fltr_list)) + return -ENODEV; + + pf = vsi->back; + + if (skb->encapsulation) + return -EPROTONOSUPPORT; + + if (!skb_flow_dissect_flow_keys(skb, &fk, 0)) + return -EPROTONOSUPPORT; + + n_proto = fk.basic.n_proto; + /* Support only IPV4 and IPV6 */ + if ((n_proto == htons(ETH_P_IP) && !ip_is_fragment(ip_hdr(skb))) || + n_proto == htons(ETH_P_IPV6)) + ip_proto = fk.basic.ip_proto; + else + return -EPROTONOSUPPORT; + + /* Support only TCP and UDP */ + if (ip_proto != IPPROTO_TCP && ip_proto != IPPROTO_UDP) + return -EPROTONOSUPPORT; + + /* only support 4-tuple filters for aRFS */ + if (!ice_arfs_is_perfect_flow_set(&pf->hw, n_proto, ip_proto)) + return -EOPNOTSUPP; + + /* choose the aRFS list bucket based on skb hash */ + idx = skb_get_hash_raw(skb) & ICE_ARFS_LST_MASK; + /* search for entry in the bucket */ + spin_lock_bh(&vsi->arfs_lock); + hlist_for_each_entry(arfs_entry, &vsi->arfs_fltr_list[idx], + list_entry) { + struct ice_fdir_fltr *fltr_info; + + /* keep searching for the already existing arfs_entry flow */ + if (arfs_entry->flow_id != flow_id) + continue; + + fltr_info = &arfs_entry->fltr_info; + ret = fltr_info->fltr_id; + + if (fltr_info->q_index == rxq_idx || + arfs_entry->fltr_state != ICE_ARFS_ACTIVE) + goto out; + + /* update the queue to forward to on an already existing flow */ + fltr_info->q_index = rxq_idx; + arfs_entry->fltr_state = ICE_ARFS_INACTIVE; + ice_arfs_update_active_fltr_cntrs(vsi, arfs_entry, false); + goto out_schedule_service_task; + } + + arfs_entry = ice_arfs_build_entry(vsi, &fk, rxq_idx, flow_id); + if (!arfs_entry) { + ret = -ENOMEM; + goto out; + } + + ret = arfs_entry->fltr_info.fltr_id; + INIT_HLIST_NODE(&arfs_entry->list_entry); + hlist_add_head(&arfs_entry->list_entry, &vsi->arfs_fltr_list[idx]); +out_schedule_service_task: + ice_service_task_schedule(pf); +out: + spin_unlock_bh(&vsi->arfs_lock); + return ret; +} + +/** + * ice_init_arfs_cntrs - initialize aRFS counter values + * @vsi: VSI that aRFS counters need to be initialized on + */ +static int ice_init_arfs_cntrs(struct ice_vsi *vsi) +{ + if (!vsi || vsi->type != ICE_VSI_PF) + return -EINVAL; + + vsi->arfs_fltr_cntrs = kzalloc(sizeof(*vsi->arfs_fltr_cntrs), + GFP_KERNEL); + if (!vsi->arfs_fltr_cntrs) + return -ENOMEM; + + vsi->arfs_last_fltr_id = kzalloc(sizeof(*vsi->arfs_last_fltr_id), + GFP_KERNEL); + if (!vsi->arfs_last_fltr_id) { + kfree(vsi->arfs_fltr_cntrs); + vsi->arfs_fltr_cntrs = NULL; + return -ENOMEM; + } + + return 0; +} + +/** + * ice_init_arfs - initialize aRFS resources + * @vsi: the VSI to be forwarded to + */ +void ice_init_arfs(struct ice_vsi *vsi) +{ + struct hlist_head *arfs_fltr_list; + unsigned int i; + + if (!vsi || vsi->type != ICE_VSI_PF) + return; + + arfs_fltr_list = kzalloc(sizeof(*arfs_fltr_list) * ICE_MAX_ARFS_LIST, + GFP_KERNEL); + if (!arfs_fltr_list) + return; + + if (ice_init_arfs_cntrs(vsi)) + goto free_arfs_fltr_list; + + for (i = 0; i < ICE_MAX_ARFS_LIST; i++) + INIT_HLIST_HEAD(&arfs_fltr_list[i]); + + spin_lock_init(&vsi->arfs_lock); + + vsi->arfs_fltr_list = arfs_fltr_list; + + return; + +free_arfs_fltr_list: + kfree(arfs_fltr_list); +} + +/** + * ice_clear_arfs - clear the aRFS hash table and any memory used for aRFS + * @vsi: the VSI to be forwarded to + */ +void ice_clear_arfs(struct ice_vsi *vsi) +{ + struct device *dev; + unsigned int i; + + if (!vsi || vsi->type != ICE_VSI_PF || !vsi->back || + !vsi->arfs_fltr_list) + return; + + dev = ice_pf_to_dev(vsi->back); + for (i = 0; i < ICE_MAX_ARFS_LIST; i++) { + struct ice_arfs_entry *r; + struct hlist_node *n; + + spin_lock_bh(&vsi->arfs_lock); + hlist_for_each_entry_safe(r, n, &vsi->arfs_fltr_list[i], + list_entry) { + hlist_del(&r->list_entry); + devm_kfree(dev, r); + } + spin_unlock_bh(&vsi->arfs_lock); + } + + kfree(vsi->arfs_fltr_list); + vsi->arfs_fltr_list = NULL; + kfree(vsi->arfs_last_fltr_id); + vsi->arfs_last_fltr_id = NULL; + kfree(vsi->arfs_fltr_cntrs); + vsi->arfs_fltr_cntrs = NULL; +} + +/** + * ice_free_cpu_rx_rmap - free setup CPU reverse map + * @vsi: the VSI to be forwarded to + */ +void ice_free_cpu_rx_rmap(struct ice_vsi *vsi) +{ + struct net_device *netdev; + + if (!vsi || vsi->type != ICE_VSI_PF || !vsi->arfs_fltr_list) + return; + + netdev = vsi->netdev; + if (!netdev || !netdev->rx_cpu_rmap || + netdev->reg_state != NETREG_REGISTERED) + return; + + free_irq_cpu_rmap(netdev->rx_cpu_rmap); + netdev->rx_cpu_rmap = NULL; +} + +/** + * ice_set_cpu_rx_rmap - setup CPU reverse map for each queue + * @vsi: the VSI to be forwarded to + */ +int ice_set_cpu_rx_rmap(struct ice_vsi *vsi) +{ + struct net_device *netdev; + struct ice_pf *pf; + int base_idx, i; + + if (!vsi || vsi->type != ICE_VSI_PF) + return -EINVAL; + + pf = vsi->back; + netdev = vsi->netdev; + if (!pf || !netdev || !vsi->num_q_vectors || + vsi->netdev->reg_state != NETREG_REGISTERED) + return -EINVAL; + + netdev_dbg(netdev, "Setup CPU RMAP: vsi type 0x%x, ifname %s, q_vectors %d\n", + vsi->type, netdev->name, vsi->num_q_vectors); + + netdev->rx_cpu_rmap = alloc_irq_cpu_rmap(vsi->num_q_vectors); + if (unlikely(!netdev->rx_cpu_rmap)) + return -EINVAL; + + base_idx = vsi->base_vector; + for (i = 0; i < vsi->num_q_vectors; i++) + if (irq_cpu_rmap_add(netdev->rx_cpu_rmap, + pf->msix_entries[base_idx + i].vector)) { + ice_free_cpu_rx_rmap(vsi); + return -EINVAL; + } + + return 0; +} + +/** + * ice_remove_arfs - remove/clear all aRFS resources + * @pf: device private structure + */ +void ice_remove_arfs(struct ice_pf *pf) +{ + struct ice_vsi *pf_vsi; + + pf_vsi = ice_get_main_vsi(pf); + if (!pf_vsi) + return; + + ice_free_cpu_rx_rmap(pf_vsi); + ice_clear_arfs(pf_vsi); +} + +/** + * ice_rebuild_arfs - remove/clear all aRFS resources and rebuild after reset + * @pf: device private structure + */ +void ice_rebuild_arfs(struct ice_pf *pf) +{ + struct ice_vsi *pf_vsi; + + pf_vsi = ice_get_main_vsi(pf); + if (!pf_vsi) + return; + + ice_remove_arfs(pf); + if (ice_set_cpu_rx_rmap(pf_vsi)) { + dev_err(ice_pf_to_dev(pf), "Failed to rebuild aRFS\n"); + return; + } + ice_init_arfs(pf_vsi); +} diff --git a/drivers/net/ethernet/intel/ice/ice_arfs.h b/drivers/net/ethernet/intel/ice/ice_arfs.h new file mode 100644 index 000000000000..f39cd16403ed --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_arfs.h @@ -0,0 +1,82 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (C) 2018-2020, Intel Corporation. */ + +#ifndef _ICE_ARFS_H_ +#define _ICE_ARFS_H_ +enum ice_arfs_fltr_state { + ICE_ARFS_INACTIVE, + ICE_ARFS_ACTIVE, + ICE_ARFS_TODEL, +}; + +struct ice_arfs_entry { + struct ice_fdir_fltr fltr_info; + struct hlist_node list_entry; + u64 time_activated; /* only valid for UDP flows */ + u32 flow_id; + /* fltr_state = 0 - ICE_ARFS_INACTIVE: + * filter needs to be updated or programmed in HW. + * fltr_state = 1 - ICE_ARFS_ACTIVE: + * filter is active and programmed in HW. + * fltr_state = 2 - ICE_ARFS_TODEL: + * filter has been deleted from HW and needs to be removed from + * the aRFS hash table. + */ + u8 fltr_state; +}; + +struct ice_arfs_entry_ptr { + struct ice_arfs_entry *arfs_entry; + struct hlist_node list_entry; +}; + +struct ice_arfs_active_fltr_cntrs { + atomic_t active_tcpv4_cnt; + atomic_t active_tcpv6_cnt; + atomic_t active_udpv4_cnt; + atomic_t active_udpv6_cnt; +}; + +#ifdef CONFIG_RFS_ACCEL +int +ice_rx_flow_steer(struct net_device *netdev, const struct sk_buff *skb, + u16 rxq_idx, u32 flow_id); +void ice_clear_arfs(struct ice_vsi *vsi); +void ice_free_cpu_rx_rmap(struct ice_vsi *vsi); +void ice_init_arfs(struct ice_vsi *vsi); +void ice_sync_arfs_fltrs(struct ice_pf *pf); +int ice_set_cpu_rx_rmap(struct ice_vsi *vsi); +void ice_remove_arfs(struct ice_pf *pf); +void ice_rebuild_arfs(struct ice_pf *pf); +bool +ice_is_arfs_using_perfect_flow(struct ice_hw *hw, + enum ice_fltr_ptype flow_type); +#else +#define ice_sync_arfs_fltrs(pf) do {} while (0) +#define ice_init_arfs(vsi) do {} while (0) +#define ice_clear_arfs(vsi) do {} while (0) +#define ice_remove_arfs(pf) do {} while (0) +#define ice_free_cpu_rx_rmap(vsi) do {} while (0) +#define ice_rebuild_arfs(pf) do {} while (0) + +static inline int ice_set_cpu_rx_rmap(struct ice_vsi __always_unused *vsi) +{ + return 0; +} + +static inline int +ice_rx_flow_steer(struct net_device __always_unused *netdev, + const struct sk_buff __always_unused *skb, + u16 __always_unused rxq_idx, u32 __always_unused flow_id) +{ + return -EOPNOTSUPP; +} + +static inline bool +ice_is_arfs_using_perfect_flow(struct ice_hw __always_unused *hw, + enum ice_fltr_ptype __always_unused flow_type) +{ + return false; +} +#endif /* CONFIG_RFS_ACCEL */ +#endif /* _ICE_ARFS_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c index a19cd6f5436b..d620d26d42ed 100644 --- a/drivers/net/ethernet/intel/ice/ice_base.c +++ b/drivers/net/ethernet/intel/ice/ice_base.c @@ -1,7 +1,9 @@ // SPDX-License-Identifier: GPL-2.0 /* Copyright (c) 2019, Intel Corporation. */ +#include <net/xdp_sock_drv.h> #include "ice_base.h" +#include "ice_lib.h" #include "ice_dcb_lib.h" /** @@ -12,7 +14,7 @@ */ static int __ice_vsi_get_qs_contig(struct ice_qs_cfg *qs_cfg) { - int offset, i; + unsigned int offset, i; mutex_lock(qs_cfg->qs_mutex); offset = bitmap_find_next_zero_area(qs_cfg->pf_map, qs_cfg->pf_map_size, @@ -24,7 +26,7 @@ static int __ice_vsi_get_qs_contig(struct ice_qs_cfg *qs_cfg) bitmap_set(qs_cfg->pf_map, offset, qs_cfg->q_count); for (i = 0; i < qs_cfg->q_count; i++) - qs_cfg->vsi_map[i + qs_cfg->vsi_map_offset] = i + offset; + qs_cfg->vsi_map[i + qs_cfg->vsi_map_offset] = (u16)(i + offset); mutex_unlock(qs_cfg->qs_mutex); return 0; @@ -38,7 +40,7 @@ static int __ice_vsi_get_qs_contig(struct ice_qs_cfg *qs_cfg) */ static int __ice_vsi_get_qs_sc(struct ice_qs_cfg *qs_cfg) { - int i, index = 0; + unsigned int i, index = 0; mutex_lock(qs_cfg->qs_mutex); for (i = 0; i < qs_cfg->q_count; i++) { @@ -47,7 +49,7 @@ static int __ice_vsi_get_qs_sc(struct ice_qs_cfg *qs_cfg) if (index >= qs_cfg->pf_map_size) goto err_scatter; set_bit(index, qs_cfg->pf_map); - qs_cfg->vsi_map[i + qs_cfg->vsi_map_offset] = index; + qs_cfg->vsi_map[i + qs_cfg->vsi_map_offset] = (u16)index; } mutex_unlock(qs_cfg->qs_mutex); @@ -96,7 +98,7 @@ static int ice_pf_rxq_wait(struct ice_pf *pf, int pf_q, bool ena) * We allocate one q_vector and set default value for ITR setting associated * with this q_vector. If allocation fails we return -ENOMEM. */ -static int ice_vsi_alloc_q_vector(struct ice_vsi *vsi, int v_idx) +static int ice_vsi_alloc_q_vector(struct ice_vsi *vsi, u16 v_idx) { struct ice_pf *pf = vsi->back; struct ice_q_vector *q_vector; @@ -246,6 +248,7 @@ ice_setup_tx_ctx(struct ice_ring *ring, struct ice_tlan_ctx *tlan_ctx, u16 pf_q) */ switch (vsi->type) { case ICE_VSI_LB: + case ICE_VSI_CTRL: case ICE_VSI_PF: tlan_ctx->vmvf_type = ICE_TLAN_CTX_VMVF_TYPE_PF; break; @@ -279,12 +282,13 @@ ice_setup_tx_ctx(struct ice_ring *ring, struct ice_tlan_ctx *tlan_ctx, u16 pf_q) */ int ice_setup_rx_ctx(struct ice_ring *ring) { + struct device *dev = ice_pf_to_dev(ring->vsi->back); int chain_len = ICE_MAX_CHAINED_RX_BUFS; + u16 num_bufs = ICE_DESC_UNUSED(ring); struct ice_vsi *vsi = ring->vsi; u32 rxdid = ICE_RXDID_FLEX_NIC; struct ice_rlan_ctx rlan_ctx; struct ice_hw *hw; - u32 regval; u16 pf_q; int err; @@ -308,24 +312,23 @@ int ice_setup_rx_ctx(struct ice_ring *ring) if (ring->xsk_umem) { xdp_rxq_info_unreg_mem_model(&ring->xdp_rxq); - ring->rx_buf_len = ring->xsk_umem->chunk_size_nohr - - XDP_PACKET_HEADROOM; + ring->rx_buf_len = + xsk_umem_get_rx_frame_size(ring->xsk_umem); /* For AF_XDP ZC, we disallow packets to span on * multiple buffers, thus letting us skip that * handling in the fast-path. */ chain_len = 1; - ring->zca.free = ice_zca_free; err = xdp_rxq_info_reg_mem_model(&ring->xdp_rxq, - MEM_TYPE_ZERO_COPY, - &ring->zca); + MEM_TYPE_XSK_BUFF_POOL, + NULL); if (err) return err; + xsk_buff_set_rxq_info(ring->xsk_umem, &ring->xdp_rxq); - dev_info(ice_pf_to_dev(vsi->back), "Registered XDP mem model MEM_TYPE_ZERO_COPY on Rx ring %d\n", + dev_info(dev, "Registered XDP mem model MEM_TYPE_XSK_BUFF_POOL on Rx ring %d\n", ring->q_index); } else { - ring->zca.free = NULL; if (!xdp_rxq_info_is_reg(&ring->xdp_rxq)) /* coverity[check_return] */ xdp_rxq_info_reg(&ring->xdp_rxq, @@ -376,38 +379,27 @@ int ice_setup_rx_ctx(struct ice_ring *ring) /* Max packet size for this queue - must not be set to a larger value * than 5 x DBUF */ - rlan_ctx.rxmax = min_t(u16, vsi->max_frame, + rlan_ctx.rxmax = min_t(u32, vsi->max_frame, chain_len * ring->rx_buf_len); /* Rx queue threshold in units of 64 */ rlan_ctx.lrxqthresh = 1; - /* Enable Flexible Descriptors in the queue context which - * allows this driver to select a specific receive descriptor format - */ - regval = rd32(hw, QRXFLXP_CNTXT(pf_q)); - if (vsi->type != ICE_VSI_VF) { - regval |= (rxdid << QRXFLXP_CNTXT_RXDID_IDX_S) & - QRXFLXP_CNTXT_RXDID_IDX_M; - - /* increasing context priority to pick up profile ID; - * default is 0x01; setting to 0x03 to ensure profile - * is programming if prev context is of same priority - */ - regval |= (0x03 << QRXFLXP_CNTXT_RXDID_PRIO_S) & - QRXFLXP_CNTXT_RXDID_PRIO_M; - - } else { - regval &= ~(QRXFLXP_CNTXT_RXDID_IDX_M | - QRXFLXP_CNTXT_RXDID_PRIO_M | - QRXFLXP_CNTXT_TS_M); - } - wr32(hw, QRXFLXP_CNTXT(pf_q), regval); + /* Enable Flexible Descriptors in the queue context which + * allows this driver to select a specific receive descriptor format + * increasing context priority to pick up profile ID; default is 0x01; + * setting to 0x03 to ensure profile is programming if prev context is + * of same priority + */ + if (vsi->type != ICE_VSI_VF) + ice_write_qrxflxp_cntxt(hw, pf_q, rxdid, 0x3); + else + ice_write_qrxflxp_cntxt(hw, pf_q, ICE_RXDID_LEGACY_1, 0x3); /* Absolute queue number out of 2K needs to be passed */ err = ice_write_rxq_ctx(hw, &rlan_ctx, pf_q); if (err) { - dev_err(ice_pf_to_dev(vsi->back), "Failed to set LAN Rx queue context for absolute Rx queue %d error: %d\n", + dev_err(dev, "Failed to set LAN Rx queue context for absolute Rx queue %d error: %d\n", pf_q, err); return -EIO; } @@ -425,13 +417,23 @@ int ice_setup_rx_ctx(struct ice_ring *ring) ring->tail = hw->hw_addr + QRX_TAIL(pf_q); writel(0, ring->tail); - err = ring->xsk_umem ? - ice_alloc_rx_bufs_slow_zc(ring, ICE_DESC_UNUSED(ring)) : - ice_alloc_rx_bufs(ring, ICE_DESC_UNUSED(ring)); - if (err) - dev_info(ice_pf_to_dev(vsi->back), "Failed allocate some buffers on %sRx ring %d (pf_q %d)\n", - ring->xsk_umem ? "UMEM enabled " : "", - ring->q_index, pf_q); + if (ring->xsk_umem) { + if (!xsk_buff_can_alloc(ring->xsk_umem, num_bufs)) { + dev_warn(dev, "UMEM does not provide enough addresses to fill %d buffers on Rx ring %d\n", + num_bufs, ring->q_index); + dev_warn(dev, "Change Rx ring/fill queue size to avoid performance issues\n"); + + return 0; + } + + err = ice_alloc_rx_bufs_zc(ring, num_bufs); + if (err) + dev_info(dev, "Failed to allocate some buffers on UMEM enabled Rx ring %d (pf_q %d)\n", + ring->q_index, pf_q); + return 0; + } + + ice_alloc_rx_bufs(ring, num_bufs); return 0; } @@ -453,7 +455,7 @@ int __ice_vsi_get_qs(struct ice_qs_cfg *qs_cfg) if (ret) { /* contig failed, so try with scatter approach */ qs_cfg->mapping_mode = ICE_VSI_MAP_SCATTER; - qs_cfg->q_count = min_t(u16, qs_cfg->q_count, + qs_cfg->q_count = min_t(unsigned int, qs_cfg->q_count, qs_cfg->scatter_count); ret = __ice_vsi_get_qs_sc(qs_cfg); } @@ -526,7 +528,8 @@ int ice_vsi_wait_one_rx_ring(struct ice_vsi *vsi, bool ena, u16 rxq_idx) int ice_vsi_alloc_q_vectors(struct ice_vsi *vsi) { struct device *dev = ice_pf_to_dev(vsi->back); - int v_idx, err; + u16 v_idx; + int err; if (vsi->q_vectors[0]) { dev_dbg(dev, "VSI %d has existing q_vectors\n", vsi->vsi_num); @@ -562,7 +565,7 @@ err_out: void ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi) { int q_vectors = vsi->num_q_vectors; - int tx_rings_rem, rx_rings_rem; + u16 tx_rings_rem, rx_rings_rem; int v_id; /* initially assigning remaining rings count to VSIs num queue value */ @@ -571,10 +574,12 @@ void ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi) for (v_id = 0; v_id < q_vectors; v_id++) { struct ice_q_vector *q_vector = vsi->q_vectors[v_id]; - int tx_rings_per_v, rx_rings_per_v, q_id, q_base; + u8 tx_rings_per_v, rx_rings_per_v; + u16 q_id, q_base; /* Tx rings mapping to vector */ - tx_rings_per_v = DIV_ROUND_UP(tx_rings_rem, q_vectors - v_id); + tx_rings_per_v = (u8)DIV_ROUND_UP(tx_rings_rem, + q_vectors - v_id); q_vector->num_ring_tx = tx_rings_per_v; q_vector->tx.ring = NULL; q_vector->tx.itr_idx = ICE_TX_ITR; @@ -590,7 +595,8 @@ void ice_vsi_map_rings_to_vectors(struct ice_vsi *vsi) tx_rings_rem -= tx_rings_per_v; /* Rx rings mapping to vector */ - rx_rings_per_v = DIV_ROUND_UP(rx_rings_rem, q_vectors - v_id); + rx_rings_per_v = (u8)DIV_ROUND_UP(rx_rings_rem, + q_vectors - v_id); q_vector->num_ring_rx = rx_rings_per_v; q_vector->rx.ring = NULL; q_vector->rx.itr_idx = ICE_RX_ITR; @@ -633,6 +639,7 @@ ice_vsi_cfg_txq(struct ice_vsi *vsi, struct ice_ring *ring, struct ice_aqc_add_txqs_perq *txq; struct ice_pf *pf = vsi->back; u8 buf_len = sizeof(*qg_buf); + struct ice_hw *hw = &pf->hw; enum ice_status status; u16 pf_q; u8 tc; @@ -641,13 +648,13 @@ ice_vsi_cfg_txq(struct ice_vsi *vsi, struct ice_ring *ring, ice_setup_tx_ctx(ring, &tlan_ctx, pf_q); /* copy context contents into the qg_buf */ qg_buf->txqs[0].txq_id = cpu_to_le16(pf_q); - ice_set_ctx((u8 *)&tlan_ctx, qg_buf->txqs[0].txq_ctx, + ice_set_ctx(hw, (u8 *)&tlan_ctx, qg_buf->txqs[0].txq_ctx, ice_tlan_ctx_info); /* init queue specific tail reg. It is referred as * transmit comm scheduler queue doorbell. */ - ring->tail = pf->hw.hw_addr + QTX_COMM_DBELL(pf_q); + ring->tail = hw->hw_addr + QTX_COMM_DBELL(pf_q); if (IS_ENABLED(CONFIG_DCB)) tc = ring->dcb_tc; @@ -662,8 +669,8 @@ ice_vsi_cfg_txq(struct ice_vsi *vsi, struct ice_ring *ring, status = ice_ena_vsi_txq(vsi->port_info, vsi->idx, tc, ring->q_handle, 1, qg_buf, buf_len, NULL); if (status) { - dev_err(ice_pf_to_dev(pf), "Failed to set LAN Tx queue context, error: %d\n", - status); + dev_err(ice_pf_to_dev(pf), "Failed to set LAN Tx queue context, error: %s\n", + ice_stat_str(status)); return -ENODEV; } @@ -832,8 +839,8 @@ ice_vsi_stop_tx_ring(struct ice_vsi *vsi, enum ice_disq_rst_src rst_src, } else if (status == ICE_ERR_DOES_NOT_EXIST) { dev_dbg(ice_pf_to_dev(vsi->back), "LAN Tx queues do not exist, nothing to disable\n"); } else if (status) { - dev_err(ice_pf_to_dev(vsi->back), "Failed to disable LAN Tx queues, error: %d\n", - status); + dev_err(ice_pf_to_dev(vsi->back), "Failed to disable LAN Tx queues, error: %s\n", + ice_stat_str(status)); return -ENODEV; } diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c index 2c0d8fd3d5cd..bce0e1281168 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.c +++ b/drivers/net/ethernet/intel/ice/ice_common.c @@ -316,12 +316,78 @@ ice_aq_get_link_info(struct ice_port_info *pi, bool ena_lse, } /** + * ice_fill_tx_timer_and_fc_thresh + * @hw: pointer to the HW struct + * @cmd: pointer to MAC cfg structure + * + * Add Tx timer and FC refresh threshold info to Set MAC Config AQ command + * descriptor + */ +static void +ice_fill_tx_timer_and_fc_thresh(struct ice_hw *hw, + struct ice_aqc_set_mac_cfg *cmd) +{ + u16 fc_thres_val, tx_timer_val; + u32 val; + + /* We read back the transmit timer and FC threshold value of + * LFC. Thus, we will use index = + * PRTMAC_HSEC_CTL_TX_PAUSE_QUANTA_MAX_INDEX. + * + * Also, because we are operating on transmit timer and FC + * threshold of LFC, we don't turn on any bit in tx_tmr_priority + */ +#define IDX_OF_LFC PRTMAC_HSEC_CTL_TX_PAUSE_QUANTA_MAX_INDEX + + /* Retrieve the transmit timer */ + val = rd32(hw, PRTMAC_HSEC_CTL_TX_PAUSE_QUANTA(IDX_OF_LFC)); + tx_timer_val = val & + PRTMAC_HSEC_CTL_TX_PAUSE_QUANTA_HSEC_CTL_TX_PAUSE_QUANTA_M; + cmd->tx_tmr_value = cpu_to_le16(tx_timer_val); + + /* Retrieve the FC threshold */ + val = rd32(hw, PRTMAC_HSEC_CTL_TX_PAUSE_REFRESH_TIMER(IDX_OF_LFC)); + fc_thres_val = val & PRTMAC_HSEC_CTL_TX_PAUSE_REFRESH_TIMER_M; + + cmd->fc_refresh_threshold = cpu_to_le16(fc_thres_val); +} + +/** + * ice_aq_set_mac_cfg + * @hw: pointer to the HW struct + * @max_frame_size: Maximum Frame Size to be supported + * @cd: pointer to command details structure or NULL + * + * Set MAC configuration (0x0603) + */ +enum ice_status +ice_aq_set_mac_cfg(struct ice_hw *hw, u16 max_frame_size, struct ice_sq_cd *cd) +{ + struct ice_aqc_set_mac_cfg *cmd; + struct ice_aq_desc desc; + + cmd = &desc.params.set_mac_cfg; + + if (max_frame_size == 0) + return ICE_ERR_PARAM; + + ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_set_mac_cfg); + + cmd->max_frame_size = cpu_to_le16(max_frame_size); + + ice_fill_tx_timer_and_fc_thresh(hw, cmd); + + return ice_aq_send_cmd(hw, &desc, NULL, 0, cd); +} + +/** * ice_init_fltr_mgmt_struct - initializes filter management list and locks * @hw: pointer to the HW struct */ static enum ice_status ice_init_fltr_mgmt_struct(struct ice_hw *hw) { struct ice_switch_info *sw; + enum ice_status status; hw->switch_info = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*hw->switch_info), GFP_KERNEL); @@ -332,7 +398,12 @@ static enum ice_status ice_init_fltr_mgmt_struct(struct ice_hw *hw) INIT_LIST_HEAD(&sw->vsi_list_map_head); - return ice_init_def_sw_recp(hw); + status = ice_init_def_sw_recp(hw); + if (status) { + devm_kfree(ice_hw_to_dev(hw), hw->switch_info); + return status; + } + return 0; } /** @@ -653,6 +724,10 @@ enum ice_status ice_init_hw(struct ice_hw *hw) if (status) goto err_unroll_cqinit; + /* Set bit to enable Flow Director filters */ + wr32(hw, PFQF_FD_ENA, PFQF_FD_ENA_FD_ENA_M); + INIT_LIST_HEAD(&hw->fdir_list_head); + ice_clear_pxe_mode(hw); status = ice_init_nvm(hw); @@ -743,9 +818,18 @@ enum ice_status ice_init_hw(struct ice_hw *hw) if (status) goto err_unroll_fltr_mgmt_struct; + /* enable jumbo frame support at MAC level */ + status = ice_aq_set_mac_cfg(hw, ICE_AQ_SET_MAC_FRAME_SIZE_MAX, NULL); + if (status) + goto err_unroll_fltr_mgmt_struct; + /* Obtain counter base index which would be used by flow director */ + status = ice_alloc_fd_res_cntr(hw, &hw->fd_ctr_base); + if (status) + goto err_unroll_fltr_mgmt_struct; status = ice_init_hw_tbls(hw); if (status) goto err_unroll_fltr_mgmt_struct; + mutex_init(&hw->tnl_lock); return 0; err_unroll_fltr_mgmt_struct: @@ -769,12 +853,14 @@ err_unroll_cqinit: */ void ice_deinit_hw(struct ice_hw *hw) { + ice_free_fd_res_cntr(hw, hw->fd_ctr_base); ice_cleanup_fltr_mgmt_struct(hw); ice_sched_cleanup_all(hw); ice_sched_clear_agg(hw); ice_free_seg(hw); ice_free_hw_tbls(hw); + mutex_destroy(&hw->tnl_lock); if (hw->port_info) { devm_kfree(ice_hw_to_dev(hw), hw->port_info); @@ -878,7 +964,12 @@ static enum ice_status ice_pf_reset(struct ice_hw *hw) wr32(hw, PFGEN_CTRL, (reg | PFGEN_CTRL_PFSWR_M)); - for (cnt = 0; cnt < ICE_PF_RESET_WAIT_COUNT; cnt++) { + /* Wait for the PFR to complete. The wait time is the global config lock + * timeout plus the PFR timeout which will account for a possible reset + * that is occurring during a download package operation. + */ + for (cnt = 0; cnt < ICE_GLOBAL_CFG_LOCK_TIMEOUT + + ICE_PF_RESET_WAIT_COUNT; cnt++) { reg = rd32(hw, PFGEN_CTRL); if (!(reg & PFGEN_CTRL_PFSWR_M)) break; @@ -1012,7 +1103,7 @@ ice_write_rxq_ctx(struct ice_hw *hw, struct ice_rlan_ctx *rlan_ctx, rlan_ctx->prefena = 1; - ice_set_ctx((u8 *)rlan_ctx, ctx_buf, ice_rlan_ctx_info); + ice_set_ctx(hw, (u8 *)rlan_ctx, ctx_buf, ice_rlan_ctx_info); return ice_copy_rxq_ctx_to_hw(hw, ctx_buf, rxq_index); } @@ -1678,6 +1769,33 @@ ice_parse_caps(struct ice_hw *hw, void *buf, u32 cap_count, "%s: msix_vector_first_id = %d\n", prefix, caps->msix_vector_first_id); break; + case ICE_AQC_CAPS_FD: + if (dev_p) { + dev_p->num_flow_director_fltr = number; + ice_debug(hw, ICE_DBG_INIT, + "%s: num_flow_director_fltr = %d\n", + prefix, + dev_p->num_flow_director_fltr); + } + if (func_p) { + u32 reg_val, val; + + reg_val = rd32(hw, GLQF_FD_SIZE); + val = (reg_val & GLQF_FD_SIZE_FD_GSIZE_M) >> + GLQF_FD_SIZE_FD_GSIZE_S; + func_p->fd_fltr_guar = + ice_get_num_per_func(hw, val); + val = (reg_val & GLQF_FD_SIZE_FD_BSIZE_M) >> + GLQF_FD_SIZE_FD_BSIZE_S; + func_p->fd_fltr_best_effort = val; + ice_debug(hw, ICE_DBG_INIT, + "%s: fd_fltr_guar = %d\n", + prefix, func_p->fd_fltr_guar); + ice_debug(hw, ICE_DBG_INIT, + "%s: fd_fltr_best_effort = %d\n", + prefix, func_p->fd_fltr_best_effort); + } + break; case ICE_AQC_CAPS_MAX_MTU: caps->max_mtu = number; ice_debug(hw, ICE_DBG_INIT, "%s: max_mtu = %d\n", @@ -1887,10 +2005,7 @@ ice_aq_manage_mac_write(struct ice_hw *hw, const u8 *mac_addr, u8 flags, ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_manage_mac_write); cmd->flags = flags; - - /* Prep values for flags, sah, sal */ - cmd->sah = htons(*((const u16 *)mac_addr)); - cmd->sal = htonl(*((const u32 *)(mac_addr + 2))); + ether_addr_copy(cmd->mac_addr, mac_addr); return ice_aq_send_cmd(hw, &desc, NULL, 0, cd); } @@ -2117,6 +2232,7 @@ ice_aq_set_phy_cfg(struct ice_hw *hw, u8 lport, struct ice_aqc_set_phy_cfg_data *cfg, struct ice_sq_cd *cd) { struct ice_aq_desc desc; + enum ice_status status; if (!cfg) return ICE_ERR_PARAM; @@ -2145,7 +2261,11 @@ ice_aq_set_phy_cfg(struct ice_hw *hw, u8 lport, ice_debug(hw, ICE_DBG_LINK, "eeer_value = 0x%x\n", cfg->eeer_value); ice_debug(hw, ICE_DBG_LINK, "link_fec_opt = 0x%x\n", cfg->link_fec_opt); - return ice_aq_send_cmd(hw, &desc, cfg, sizeof(*cfg), cd); + status = ice_aq_send_cmd(hw, &desc, cfg, sizeof(*cfg), cd); + if (hw->adminq.sq_last_status == ICE_AQ_RC_EMODE) + status = 0; + + return status; } /** @@ -3089,12 +3209,14 @@ ice_write_qword(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info) /** * ice_set_ctx - set context bits in packed structure + * @hw: pointer to the hardware structure * @src_ctx: pointer to a generic non-packed context structure * @dest_ctx: pointer to memory for the packed structure * @ce_info: a description of the structure to be transformed */ enum ice_status -ice_set_ctx(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info) +ice_set_ctx(struct ice_hw *hw, u8 *src_ctx, u8 *dest_ctx, + const struct ice_ctx_ele *ce_info) { int f; @@ -3103,6 +3225,12 @@ ice_set_ctx(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info) * using the correct size so that we are correct regardless * of the endianness of the machine. */ + if (ce_info[f].width > (ce_info[f].size_of * BITS_PER_BYTE)) { + ice_debug(hw, ICE_DBG_QCTX, + "Field %d width of %d bits larger than size of %d byte(s) ... skipping write\n", + f, ce_info[f].width, ce_info[f].size_of); + continue; + } switch (ce_info[f].size_of) { case sizeof(u8): ice_write_byte(src_ctx, dest_ctx, &ce_info[f]); diff --git a/drivers/net/ethernet/intel/ice/ice_common.h b/drivers/net/ethernet/intel/ice/ice_common.h index 8104f3d64d96..9b9e50d2398b 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.h +++ b/drivers/net/ethernet/intel/ice/ice_common.h @@ -70,7 +70,8 @@ enum ice_status ice_aq_q_shutdown(struct ice_hw *hw, bool unloading); void ice_fill_dflt_direct_cmd_desc(struct ice_aq_desc *desc, u16 opcode); extern const struct ice_ctx_ele ice_tlan_ctx_info[]; enum ice_status -ice_set_ctx(u8 *src_ctx, u8 *dest_ctx, const struct ice_ctx_ele *ce_info); +ice_set_ctx(struct ice_hw *hw, u8 *src_ctx, u8 *dest_ctx, + const struct ice_ctx_ele *ce_info); extern struct mutex ice_global_cfg_lock_sw; @@ -108,6 +109,8 @@ enum ice_status ice_aq_set_link_restart_an(struct ice_port_info *pi, bool ena_link, struct ice_sq_cd *cd); enum ice_status +ice_aq_set_mac_cfg(struct ice_hw *hw, u16 max_frame_size, struct ice_sq_cd *cd); +enum ice_status ice_aq_get_link_info(struct ice_port_info *pi, bool ena_lse, struct ice_link_status *link, struct ice_sq_cd *cd); enum ice_status diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.c b/drivers/net/ethernet/intel/ice/ice_controlq.c index dd946866d7b8..1e18021aa073 100644 --- a/drivers/net/ethernet/intel/ice/ice_controlq.c +++ b/drivers/net/ethernet/intel/ice/ice_controlq.c @@ -12,6 +12,7 @@ do { \ (qinfo)->sq.bal = prefix##_ATQBAL; \ (qinfo)->sq.len_mask = prefix##_ATQLEN_ATQLEN_M; \ (qinfo)->sq.len_ena_mask = prefix##_ATQLEN_ATQENABLE_M; \ + (qinfo)->sq.len_crit_mask = prefix##_ATQLEN_ATQCRIT_M; \ (qinfo)->sq.head_mask = prefix##_ATQH_ATQH_M; \ (qinfo)->rq.head = prefix##_ARQH; \ (qinfo)->rq.tail = prefix##_ARQT; \ @@ -20,6 +21,7 @@ do { \ (qinfo)->rq.bal = prefix##_ARQBAL; \ (qinfo)->rq.len_mask = prefix##_ARQLEN_ARQLEN_M; \ (qinfo)->rq.len_ena_mask = prefix##_ARQLEN_ARQENABLE_M; \ + (qinfo)->rq.len_crit_mask = prefix##_ARQLEN_ARQCRIT_M; \ (qinfo)->rq.head_mask = prefix##_ARQH_ARQH_M; \ } while (0) @@ -199,7 +201,9 @@ unwind_alloc_rq_bufs: cq->rq.r.rq_bi[i].pa = 0; cq->rq.r.rq_bi[i].size = 0; } + cq->rq.r.rq_bi = NULL; devm_kfree(ice_hw_to_dev(hw), cq->rq.dma_head); + cq->rq.dma_head = NULL; return ICE_ERR_NO_MEMORY; } @@ -245,7 +249,9 @@ unwind_alloc_sq_bufs: cq->sq.r.sq_bi[i].pa = 0; cq->sq.r.sq_bi[i].size = 0; } + cq->sq.r.sq_bi = NULL; devm_kfree(ice_hw_to_dev(hw), cq->sq.dma_head); + cq->sq.dma_head = NULL; return ICE_ERR_NO_MEMORY; } @@ -304,6 +310,28 @@ ice_cfg_rq_regs(struct ice_hw *hw, struct ice_ctl_q_info *cq) return 0; } +#define ICE_FREE_CQ_BUFS(hw, qi, ring) \ +do { \ + int i; \ + /* free descriptors */ \ + if ((qi)->ring.r.ring##_bi) \ + for (i = 0; i < (qi)->num_##ring##_entries; i++) \ + if ((qi)->ring.r.ring##_bi[i].pa) { \ + dmam_free_coherent(ice_hw_to_dev(hw), \ + (qi)->ring.r.ring##_bi[i].size, \ + (qi)->ring.r.ring##_bi[i].va, \ + (qi)->ring.r.ring##_bi[i].pa); \ + (qi)->ring.r.ring##_bi[i].va = NULL;\ + (qi)->ring.r.ring##_bi[i].pa = 0;\ + (qi)->ring.r.ring##_bi[i].size = 0;\ + } \ + /* free the buffer info list */ \ + if ((qi)->ring.cmd_buf) \ + devm_kfree(ice_hw_to_dev(hw), (qi)->ring.cmd_buf); \ + /* free DMA head */ \ + devm_kfree(ice_hw_to_dev(hw), (qi)->ring.dma_head); \ +} while (0) + /** * ice_init_sq - main initialization routine for Control ATQ * @hw: pointer to the hardware structure @@ -357,6 +385,7 @@ static enum ice_status ice_init_sq(struct ice_hw *hw, struct ice_ctl_q_info *cq) goto init_ctrlq_exit; init_ctrlq_free_rings: + ICE_FREE_CQ_BUFS(hw, cq, sq); ice_free_cq_ring(hw, &cq->sq); init_ctrlq_exit: @@ -416,33 +445,13 @@ static enum ice_status ice_init_rq(struct ice_hw *hw, struct ice_ctl_q_info *cq) goto init_ctrlq_exit; init_ctrlq_free_rings: + ICE_FREE_CQ_BUFS(hw, cq, rq); ice_free_cq_ring(hw, &cq->rq); init_ctrlq_exit: return ret_code; } -#define ICE_FREE_CQ_BUFS(hw, qi, ring) \ -do { \ - int i; \ - /* free descriptors */ \ - for (i = 0; i < (qi)->num_##ring##_entries; i++) \ - if ((qi)->ring.r.ring##_bi[i].pa) { \ - dmam_free_coherent(ice_hw_to_dev(hw), \ - (qi)->ring.r.ring##_bi[i].size,\ - (qi)->ring.r.ring##_bi[i].va,\ - (qi)->ring.r.ring##_bi[i].pa);\ - (qi)->ring.r.ring##_bi[i].va = NULL; \ - (qi)->ring.r.ring##_bi[i].pa = 0; \ - (qi)->ring.r.ring##_bi[i].size = 0; \ - } \ - /* free the buffer info list */ \ - if ((qi)->ring.cmd_buf) \ - devm_kfree(ice_hw_to_dev(hw), (qi)->ring.cmd_buf); \ - /* free DMA head */ \ - devm_kfree(ice_hw_to_dev(hw), (qi)->ring.dma_head); \ -} while (0) - /** * ice_shutdown_sq - shutdown the Control ATQ * @hw: pointer to the hardware structure @@ -635,6 +644,50 @@ init_ctrlq_free_sq: } /** + * ice_shutdown_ctrlq - shutdown routine for any control queue + * @hw: pointer to the hardware structure + * @q_type: specific Control queue type + * + * NOTE: this function does not destroy the control queue locks. + */ +static void ice_shutdown_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type) +{ + struct ice_ctl_q_info *cq; + + switch (q_type) { + case ICE_CTL_Q_ADMIN: + cq = &hw->adminq; + if (ice_check_sq_alive(hw, cq)) + ice_aq_q_shutdown(hw, true); + break; + case ICE_CTL_Q_MAILBOX: + cq = &hw->mailboxq; + break; + default: + return; + } + + ice_shutdown_sq(hw, cq); + ice_shutdown_rq(hw, cq); +} + +/** + * ice_shutdown_all_ctrlq - shutdown routine for all control queues + * @hw: pointer to the hardware structure + * + * NOTE: this function does not destroy the control queue locks. The driver + * may call this at runtime to shutdown and later restart control queues, such + * as in response to a reset event. + */ +void ice_shutdown_all_ctrlq(struct ice_hw *hw) +{ + /* Shutdown FW admin queue */ + ice_shutdown_ctrlq(hw, ICE_CTL_Q_ADMIN); + /* Shutdown PF-VF Mailbox */ + ice_shutdown_ctrlq(hw, ICE_CTL_Q_MAILBOX); +} + +/** * ice_init_all_ctrlq - main initialization routine for all control queues * @hw: pointer to the hardware structure * @@ -649,17 +702,27 @@ init_ctrlq_free_sq: */ enum ice_status ice_init_all_ctrlq(struct ice_hw *hw) { - enum ice_status ret_code; + enum ice_status status; + u32 retry = 0; /* Init FW admin queue */ - ret_code = ice_init_ctrlq(hw, ICE_CTL_Q_ADMIN); - if (ret_code) - return ret_code; + do { + status = ice_init_ctrlq(hw, ICE_CTL_Q_ADMIN); + if (status) + return status; - ret_code = ice_init_check_adminq(hw); - if (ret_code) - return ret_code; + status = ice_init_check_adminq(hw); + if (status != ICE_ERR_AQ_FW_CRITICAL) + break; + + ice_debug(hw, ICE_DBG_AQ_MSG, + "Retry Admin Queue init due to FW critical error\n"); + ice_shutdown_ctrlq(hw, ICE_CTL_Q_ADMIN); + msleep(ICE_CTL_Q_ADMIN_INIT_MSEC); + } while (retry++ < ICE_CTL_Q_ADMIN_INIT_TIMEOUT); + if (status) + return status; /* Init Mailbox queue */ return ice_init_ctrlq(hw, ICE_CTL_Q_MAILBOX); } @@ -701,57 +764,12 @@ enum ice_status ice_create_all_ctrlq(struct ice_hw *hw) } /** - * ice_shutdown_ctrlq - shutdown routine for any control queue - * @hw: pointer to the hardware structure - * @q_type: specific Control queue type - * - * NOTE: this function does not destroy the control queue locks. - */ -static void ice_shutdown_ctrlq(struct ice_hw *hw, enum ice_ctl_q q_type) -{ - struct ice_ctl_q_info *cq; - - switch (q_type) { - case ICE_CTL_Q_ADMIN: - cq = &hw->adminq; - if (ice_check_sq_alive(hw, cq)) - ice_aq_q_shutdown(hw, true); - break; - case ICE_CTL_Q_MAILBOX: - cq = &hw->mailboxq; - break; - default: - return; - } - - ice_shutdown_sq(hw, cq); - ice_shutdown_rq(hw, cq); -} - -/** - * ice_shutdown_all_ctrlq - shutdown routine for all control queues - * @hw: pointer to the hardware structure - * - * NOTE: this function does not destroy the control queue locks. The driver - * may call this at runtime to shutdown and later restart control queues, such - * as in response to a reset event. - */ -void ice_shutdown_all_ctrlq(struct ice_hw *hw) -{ - /* Shutdown FW admin queue */ - ice_shutdown_ctrlq(hw, ICE_CTL_Q_ADMIN); - /* Shutdown PF-VF Mailbox */ - ice_shutdown_ctrlq(hw, ICE_CTL_Q_MAILBOX); -} - -/** * ice_destroy_ctrlq_locks - Destroy locks for a control queue * @cq: pointer to the control queue * * Destroys the send and receive queue locks for a given control queue. */ -static void -ice_destroy_ctrlq_locks(struct ice_ctl_q_info *cq) +static void ice_destroy_ctrlq_locks(struct ice_ctl_q_info *cq) { mutex_destroy(&cq->sq_lock); mutex_destroy(&cq->rq_lock); @@ -1042,9 +1060,15 @@ ice_sq_send_cmd(struct ice_hw *hw, struct ice_ctl_q_info *cq, /* update the error if time out occurred */ if (!cmd_completed) { - ice_debug(hw, ICE_DBG_AQ_MSG, - "Control Send Queue Writeback timeout.\n"); - status = ICE_ERR_AQ_TIMEOUT; + if (rd32(hw, cq->rq.len) & cq->rq.len_crit_mask || + rd32(hw, cq->sq.len) & cq->sq.len_crit_mask) { + ice_debug(hw, ICE_DBG_AQ_MSG, "Critical FW error.\n"); + status = ICE_ERR_AQ_FW_CRITICAL; + } else { + ice_debug(hw, ICE_DBG_AQ_MSG, + "Control Send Queue Writeback timeout.\n"); + status = ICE_ERR_AQ_TIMEOUT; + } } sq_send_command_error: @@ -1128,7 +1152,7 @@ ice_clean_rq_elem(struct ice_hw *hw, struct ice_ctl_q_info *cq, } memcpy(&e->desc, desc, sizeof(e->desc)); datalen = le16_to_cpu(desc->datalen); - e->msg_len = min(datalen, e->buf_len); + e->msg_len = min_t(u16, datalen, e->buf_len); if (e->msg_buf && e->msg_len) memcpy(e->msg_buf, cq->rq.r.rq_bi[desc_idx].va, e->msg_len); diff --git a/drivers/net/ethernet/intel/ice/ice_controlq.h b/drivers/net/ethernet/intel/ice/ice_controlq.h index bf0ebe6149e8..faaa08e8171b 100644 --- a/drivers/net/ethernet/intel/ice/ice_controlq.h +++ b/drivers/net/ethernet/intel/ice/ice_controlq.h @@ -34,6 +34,8 @@ enum ice_ctl_q { /* Control Queue timeout settings - max delay 250ms */ #define ICE_CTL_Q_SQ_CMD_TIMEOUT 2500 /* Count 2500 times */ #define ICE_CTL_Q_SQ_CMD_USEC 100 /* Check every 100usec */ +#define ICE_CTL_Q_ADMIN_INIT_TIMEOUT 10 /* Count 10 times */ +#define ICE_CTL_Q_ADMIN_INIT_MSEC 100 /* Check every 100msec */ struct ice_ctl_q_ring { void *dma_head; /* Virtual address to DMA head */ @@ -59,6 +61,7 @@ struct ice_ctl_q_ring { u32 bal; u32 len_mask; u32 len_ena_mask; + u32 len_crit_mask; u32 head_mask; }; diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_lib.c b/drivers/net/ethernet/intel/ice/ice_dcb_lib.c index 7bea09363b42..979af197f8a3 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_dcb_lib.c @@ -63,6 +63,64 @@ u8 ice_dcb_get_ena_tc(struct ice_dcbx_cfg *dcbcfg) } /** + * ice_is_pfc_causing_hung_q + * @pf: pointer to PF structure + * @txqueue: Tx queue which is supposedly hung queue + * + * find if PFC is causing the hung queue, if yes return true else false + */ +bool ice_is_pfc_causing_hung_q(struct ice_pf *pf, unsigned int txqueue) +{ + u8 num_tcs = 0, i, tc, up_mapped_tc, up_in_tc = 0; + u64 ref_prio_xoff[ICE_MAX_UP]; + struct ice_vsi *vsi; + u32 up2tc; + + vsi = ice_get_main_vsi(pf); + if (!vsi) + return false; + + ice_for_each_traffic_class(i) + if (vsi->tc_cfg.ena_tc & BIT(i)) + num_tcs++; + + /* first find out the TC to which the hung queue belongs to */ + for (tc = 0; tc < num_tcs - 1; tc++) + if (ice_find_q_in_range(vsi->tc_cfg.tc_info[tc].qoffset, + vsi->tc_cfg.tc_info[tc + 1].qoffset, + txqueue)) + break; + + /* Build a bit map of all UPs associated to the suspect hung queue TC, + * so that we check for its counter increment. + */ + up2tc = rd32(&pf->hw, PRTDCB_TUP2TC); + for (i = 0; i < ICE_MAX_UP; i++) { + up_mapped_tc = (up2tc >> (i * 3)) & 0x7; + if (up_mapped_tc == tc) + up_in_tc |= BIT(i); + } + + /* Now that we figured out that hung queue is PFC enabled, still the + * Tx timeout can be legitimate. So to make sure Tx timeout is + * absolutely caused by PFC storm, check if the counters are + * incrementing. + */ + for (i = 0; i < ICE_MAX_UP; i++) + if (up_in_tc & BIT(i)) + ref_prio_xoff[i] = pf->stats.priority_xoff_rx[i]; + + ice_update_dcb_stats(pf); + + for (i = 0; i < ICE_MAX_UP; i++) + if (up_in_tc & BIT(i)) + if (pf->stats.priority_xoff_rx[i] > ref_prio_xoff[i]) + return true; + + return false; +} + +/** * ice_dcb_get_mode - gets the DCB mode * @port_info: pointer to port info structure * @host: if set it's HOST if not it's MANAGED @@ -526,16 +584,21 @@ static int ice_dcb_sw_dflt_cfg(struct ice_pf *pf, bool ets_willing, bool locked) */ static bool ice_dcb_tc_contig(u8 *prio_table) { - u8 max_tc = 0; + bool found_empty = false; + u8 used_tc = 0; int i; - for (i = 0; i < CEE_DCBX_MAX_PRIO; i++) { - u8 cur_tc = prio_table[i]; + /* Create a bitmap of used TCs */ + for (i = 0; i < CEE_DCBX_MAX_PRIO; i++) + used_tc |= BIT(prio_table[i]); - if (cur_tc > max_tc) - return false; - else if (cur_tc == max_tc) - max_tc++; + for (i = 0; i < CEE_DCBX_MAX_PRIO; i++) { + if (used_tc & BIT(i)) { + if (found_empty) + return false; + } else { + found_empty = true; + } } return true; @@ -728,39 +791,31 @@ void ice_update_dcb_stats(struct ice_pf *pf) * ice_tx_prepare_vlan_flags_dcb - prepare VLAN tagging for DCB * @tx_ring: ring to send buffer on * @first: pointer to struct ice_tx_buf + * + * This should not be called if the outer VLAN is software offloaded as the VLAN + * tag will already be configured with the correct ID and priority bits */ -int +void ice_tx_prepare_vlan_flags_dcb(struct ice_ring *tx_ring, struct ice_tx_buf *first) { struct sk_buff *skb = first->skb; if (!test_bit(ICE_FLAG_DCB_ENA, tx_ring->vsi->back->flags)) - return 0; + return; /* Insert 802.1p priority into VLAN header */ - if ((first->tx_flags & (ICE_TX_FLAGS_HW_VLAN | ICE_TX_FLAGS_SW_VLAN)) || + if ((first->tx_flags & ICE_TX_FLAGS_HW_VLAN) || skb->priority != TC_PRIO_CONTROL) { first->tx_flags &= ~ICE_TX_FLAGS_VLAN_PR_M; /* Mask the lower 3 bits to set the 802.1p priority */ first->tx_flags |= (skb->priority & 0x7) << ICE_TX_FLAGS_VLAN_PR_S; - if (first->tx_flags & ICE_TX_FLAGS_SW_VLAN) { - struct vlan_ethhdr *vhdr; - int rc; - - rc = skb_cow_head(skb, 0); - if (rc < 0) - return rc; - vhdr = (struct vlan_ethhdr *)skb->data; - vhdr->h_vlan_TCI = htons(first->tx_flags >> - ICE_TX_FLAGS_VLAN_S); - } else { - first->tx_flags |= ICE_TX_FLAGS_HW_VLAN; - } + /* if this is not already set it means a VLAN 0 + priority needs + * to be offloaded + */ + first->tx_flags |= ICE_TX_FLAGS_HW_VLAN; } - - return 0; } /** diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_lib.h b/drivers/net/ethernet/intel/ice/ice_dcb_lib.h index 37680e815b02..323238669572 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_dcb_lib.h @@ -17,6 +17,8 @@ void ice_dcb_rebuild(struct ice_pf *pf); u8 ice_dcb_get_ena_tc(struct ice_dcbx_cfg *dcbcfg); u8 ice_dcb_get_num_tc(struct ice_dcbx_cfg *dcbcfg); +void ice_vsi_set_dcb_tc_cfg(struct ice_vsi *vsi); +bool ice_is_pfc_causing_hung_q(struct ice_pf *pf, unsigned int txqueue); u8 ice_dcb_get_tc(struct ice_vsi *vsi, int queue_index); int ice_pf_dcb_cfg(struct ice_pf *pf, struct ice_dcbx_cfg *new_cfg, bool locked); @@ -25,13 +27,27 @@ void ice_pf_dcb_recfg(struct ice_pf *pf); void ice_vsi_cfg_dcb_rings(struct ice_vsi *vsi); int ice_init_pf_dcb(struct ice_pf *pf, bool locked); void ice_update_dcb_stats(struct ice_pf *pf); -int +void ice_tx_prepare_vlan_flags_dcb(struct ice_ring *tx_ring, struct ice_tx_buf *first); void ice_dcb_process_lldp_set_mib_change(struct ice_pf *pf, struct ice_rq_event_info *event); void ice_vsi_cfg_netdev_tc(struct ice_vsi *vsi, u8 ena_tc); + +/** + * ice_find_q_in_range + * @low: start of queue range for a TC i.e. offset of TC + * @high: start of queue for next TC + * @tx_q: hung_queue/tx_queue + * + * finds if queue 'tx_q' falls between the two offsets of any given TC + */ +static inline bool ice_find_q_in_range(u16 low, u16 high, unsigned int tx_q) +{ + return (tx_q >= low) && (tx_q < high); +} + static inline void ice_set_cgd_num(struct ice_tlan_ctx *tlan_ctx, struct ice_ring *ring) { @@ -79,6 +95,13 @@ ice_tx_prepare_vlan_flags_dcb(struct ice_ring __always_unused *tx_ring, return 0; } +static inline bool +ice_is_pfc_causing_hung_q(struct ice_pf __always_unused *pf, + unsigned int __always_unused txqueue) +{ + return false; +} + #define ice_update_dcb_stats(pf) do {} while (0) #define ice_pf_dcb_recfg(pf) do {} while (0) #define ice_vsi_cfg_dcb_rings(vsi) do {} while (0) diff --git a/drivers/net/ethernet/intel/ice/ice_dcb_nl.c b/drivers/net/ethernet/intel/ice/ice_dcb_nl.c index c4c12414083a..87f91b750d59 100644 --- a/drivers/net/ethernet/intel/ice/ice_dcb_nl.c +++ b/drivers/net/ethernet/intel/ice/ice_dcb_nl.c @@ -7,8 +7,6 @@ #include "ice_dcb_nl.h" #include <net/dcbnl.h> -#define ICE_APP_PROT_ID_ROCE 0x8915 - /** * ice_dcbnl_devreset - perform enough of a ifdown/ifup to sync DCBNL info * @netdev: device associated with interface that needs reset @@ -671,7 +669,7 @@ static bool ice_dcbnl_find_app(struct ice_dcbx_cfg *cfg, struct ice_dcb_app_priority_table *app) { - int i; + unsigned int i; for (i = 0; i < cfg->numapps; i++) { if (app->selector == cfg->app[i].selector && @@ -746,7 +744,8 @@ static int ice_dcbnl_delapp(struct net_device *netdev, struct dcb_app *app) { struct ice_pf *pf = ice_netdev_to_pf(netdev); struct ice_dcbx_cfg *old_cfg, *new_cfg; - int i, j, ret = 0; + unsigned int i, j; + int ret = 0; if (pf->dcbx_cap & DCB_CAP_DCBX_LLD_MANAGED) return -EINVAL; @@ -869,7 +868,7 @@ void ice_dcbnl_set_all(struct ice_vsi *vsi) struct ice_port_info *pi; struct dcb_app sapp; struct ice_pf *pf; - int i; + unsigned int i; if (!netdev) return; @@ -941,7 +940,7 @@ ice_dcbnl_flush_apps(struct ice_pf *pf, struct ice_dcbx_cfg *old_cfg, struct ice_dcbx_cfg *new_cfg) { struct ice_vsi *main_vsi = ice_get_main_vsi(pf); - int i; + unsigned int i; if (!main_vsi) return; diff --git a/drivers/net/ethernet/intel/ice/ice_devlink.c b/drivers/net/ethernet/intel/ice/ice_devlink.c index c6833944b90a..a73d06e06b5d 100644 --- a/drivers/net/ethernet/intel/ice/ice_devlink.c +++ b/drivers/net/ethernet/intel/ice/ice_devlink.c @@ -105,6 +105,27 @@ static int ice_info_ddp_pkg_version(struct ice_pf *pf, char *buf, size_t len) return 0; } +static int ice_info_netlist_ver(struct ice_pf *pf, char *buf, size_t len) +{ + struct ice_netlist_ver_info *netlist = &pf->hw.netlist_ver; + + /* The netlist version fields are BCD formatted */ + snprintf(buf, len, "%x.%x.%x-%x.%x.%x", netlist->major, netlist->minor, + netlist->type >> 16, netlist->type & 0xFFFF, netlist->rev, + netlist->cust_ver); + + return 0; +} + +static int ice_info_netlist_build(struct ice_pf *pf, char *buf, size_t len) +{ + struct ice_netlist_ver_info *netlist = &pf->hw.netlist_ver; + + snprintf(buf, len, "0x%08x", netlist->hash); + + return 0; +} + #define fixed(key, getter) { ICE_VERSION_FIXED, key, getter } #define running(key, getter) { ICE_VERSION_RUNNING, key, getter } @@ -128,6 +149,8 @@ static const struct ice_devlink_version { running(DEVLINK_INFO_VERSION_GENERIC_FW_BUNDLE_ID, ice_info_eetrack), running("fw.app.name", ice_info_ddp_pkg_name), running(DEVLINK_INFO_VERSION_GENERIC_FW_APP, ice_info_ddp_pkg_version), + running("fw.netlist", ice_info_netlist_ver), + running("fw.netlist.build", ice_info_netlist_build), }; /** diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index 593fb37bd59e..68c38004a088 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -5,6 +5,7 @@ #include "ice.h" #include "ice_flow.h" +#include "ice_fltr.h" #include "ice_lib.h" #include "ice_dcb_lib.h" @@ -129,6 +130,8 @@ static const struct ice_stats ice_gstrings_pf_stats[] = { ICE_PF_STAT("illegal_bytes.nic", stats.illegal_bytes), ICE_PF_STAT("mac_local_faults.nic", stats.mac_local_faults), ICE_PF_STAT("mac_remote_faults.nic", stats.mac_remote_faults), + ICE_PF_STAT("fdir_sb_match.nic", stats.fd_sb_match), + ICE_PF_STAT("fdir_sb_status.nic", stats.fd_sb_status), }; static const u32 ice_regs_dump_list[] = { @@ -139,9 +142,6 @@ static const u32 ice_regs_dump_list[] = { QINT_RQCTL(0), PFINT_OICR_ENA, QRX_ITR(0), - PF0INT_ITR_0(0), - PF0INT_ITR_1(0), - PF0INT_ITR_2(0), }; struct ice_priv_flag { @@ -157,6 +157,8 @@ struct ice_priv_flag { static const struct ice_priv_flag ice_gstrings_priv_flags[] = { ICE_PRIV_FLAG("link-down-on-close", ICE_FLAG_LINK_DOWN_ON_CLOSE_ENA), ICE_PRIV_FLAG("fw-lldp-agent", ICE_FLAG_FW_LLDP_AGENT), + ICE_PRIV_FLAG("vf-true-promisc-support", + ICE_FLAG_VF_TRUE_PROMISC_ENA), ICE_PRIV_FLAG("mdd-auto-reset-vf", ICE_FLAG_MDD_AUTO_RESET_VF), ICE_PRIV_FLAG("legacy-rx", ICE_FLAG_LEGACY_RX), }; @@ -203,7 +205,7 @@ ice_get_regs(struct net_device *netdev, struct ethtool_regs *regs, void *p) struct ice_pf *pf = np->vsi->back; struct ice_hw *hw = &pf->hw; u32 *regs_buf = (u32 *)p; - int i; + unsigned int i; regs->version = 1; @@ -273,8 +275,9 @@ ice_get_eeprom(struct net_device *netdev, struct ethtool_eeprom *eeprom, status = ice_acquire_nvm(hw, ICE_RES_READ); if (status) { - dev_err(dev, "ice_acquire_nvm failed, err %d aq_err %d\n", - status, hw->adminq.sq_last_status); + dev_err(dev, "ice_acquire_nvm failed, err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); ret = -EIO; goto out; } @@ -282,8 +285,9 @@ ice_get_eeprom(struct net_device *netdev, struct ethtool_eeprom *eeprom, status = ice_read_flat_nvm(hw, eeprom->offset, &eeprom->len, buf, false); if (status) { - dev_err(dev, "ice_read_flat_nvm failed, err %d aq_err %d\n", - status, hw->adminq.sq_last_status); + dev_err(dev, "ice_read_flat_nvm failed, err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); ret = -EIO; goto release; } @@ -304,7 +308,7 @@ out: */ static bool ice_active_vfs(struct ice_pf *pf) { - int i; + unsigned int i; ice_for_each_vf(pf, i) { struct ice_vf *vf = &pf->vf[i]; @@ -332,7 +336,8 @@ static u64 ice_link_test(struct net_device *netdev) netdev_info(netdev, "link test\n"); status = ice_get_link_status(np->vsi->port_info, &link_up); if (status) { - netdev_err(netdev, "link query error, status = %d\n", status); + netdev_err(netdev, "link query error, status = %s\n", + ice_stat_str(status)); return 1; } @@ -373,7 +378,7 @@ static int ice_reg_pattern_test(struct ice_hw *hw, u32 reg, u32 mask) 0x00000000, 0xFFFFFFFF }; u32 val, orig_val; - int i; + unsigned int i; orig_val = rd32(hw, reg); for (i = 0; i < ARRAY_SIZE(patterns); ++i) { @@ -426,7 +431,7 @@ static u64 ice_reg_test(struct net_device *netdev) GLINT_ITR(2, 1) - GLINT_ITR(2, 0)}, {GLINT_CTL, 0xffff0001, 1, 0} }; - int i; + unsigned int i; netdev_dbg(netdev, "Register test\n"); for (i = 0; i < ARRAY_SIZE(ice_reg_list); ++i) { @@ -671,7 +676,6 @@ static u64 ice_loopback_test(struct net_device *netdev) struct ice_ring *tx_ring, *rx_ring; u8 broadcast[ETH_ALEN], ret = 0; int num_frames, valid_frames; - LIST_HEAD(tmp_list); struct device *dev; u8 *tx_frame; int i; @@ -707,16 +711,11 @@ static u64 ice_loopback_test(struct net_device *netdev) /* Test VSI needs to receive broadcast packets */ eth_broadcast_addr(broadcast); - if (ice_add_mac_to_list(test_vsi, &tmp_list, broadcast)) { + if (ice_fltr_add_mac(test_vsi, broadcast, ICE_FWD_TO_VSI)) { ret = 5; goto lbtest_mac_dis; } - if (ice_add_mac(&pf->hw, &tmp_list)) { - ret = 6; - goto free_mac_list; - } - if (ice_lbtest_create_frame(pf, &tx_frame, ICE_LB_FRAME_SIZE)) { ret = 7; goto remove_mac_filters; @@ -739,10 +738,8 @@ static u64 ice_loopback_test(struct net_device *netdev) lbtest_free_frame: devm_kfree(dev, tx_frame); remove_mac_filters: - if (ice_remove_mac(&pf->hw, &tmp_list)) + if (ice_fltr_remove_mac(test_vsi, broadcast, ICE_FWD_TO_VSI)) netdev_err(netdev, "Could not remove MAC filter for the test VSI\n"); -free_mac_list: - ice_free_fltr_list(dev, &tmp_list); lbtest_mac_dis: /* Disable MAC loopback after the test is completed. */ if (ice_aq_set_mac_loopback(&pf->hw, false, NULL)) @@ -1158,8 +1155,9 @@ static int ice_nway_reset(struct net_device *netdev) status = ice_aq_set_link_restart_an(pi, false, NULL); if (status) { - netdev_info(netdev, "link restart failed, err %d aq_err %d\n", - status, pi->hw->adminq.sq_last_status); + netdev_info(netdev, "link restart failed, err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(pi->hw->adminq.sq_last_status)); return -EIO; } @@ -1308,6 +1306,16 @@ static int ice_set_priv_flags(struct net_device *netdev, u32 flags) ice_down(vsi); ice_up(vsi); } + /* don't allow modification of this flag when a single VF is in + * promiscuous mode because it's not supported + */ + if (test_bit(ICE_FLAG_VF_TRUE_PROMISC_ENA, change_flags) && + ice_is_any_vf_in_promisc(pf)) { + dev_err(dev, "Changing vf-true-promisc-support flag while VF(s) are in promiscuous mode not supported\n"); + /* toggle bit back to previous state */ + change_bit(ICE_FLAG_VF_TRUE_PROMISC_ENA, pf->flags); + ret = -EAGAIN; + } clear_bit(ICE_FLAG_ETHTOOL_CTXT, pf->flags); return ret; } @@ -2450,8 +2458,8 @@ ice_set_rss_hash_opt(struct ice_vsi *vsi, struct ethtool_rxnfc *nfc) status = ice_add_rss_cfg(&pf->hw, vsi->idx, hashed_flds, hdrs); if (status) { - dev_dbg(dev, "ice_add_rss_cfg failed, vsi num = %d, error = %d\n", - vsi->vsi_num, status); + dev_dbg(dev, "ice_add_rss_cfg failed, vsi num = %d, error = %s\n", + vsi->vsi_num, ice_stat_str(status)); return -EINVAL; } @@ -2526,6 +2534,10 @@ static int ice_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd) struct ice_vsi *vsi = np->vsi; switch (cmd->cmd) { + case ETHTOOL_SRXCLSRLINS: + return ice_add_fdir_ethtool(vsi, cmd); + case ETHTOOL_SRXCLSRLDEL: + return ice_del_fdir_ethtool(vsi, cmd); case ETHTOOL_SRXFH: return ice_set_rss_hash_opt(vsi, cmd); default: @@ -2549,12 +2561,27 @@ ice_get_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd, struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; int ret = -EOPNOTSUPP; + struct ice_hw *hw; + + hw = &vsi->back->hw; switch (cmd->cmd) { case ETHTOOL_GRXRINGS: cmd->data = vsi->rss_size; ret = 0; break; + case ETHTOOL_GRXCLSRLCNT: + cmd->rule_cnt = hw->fdir_active_fltr; + /* report total rule count */ + cmd->data = ice_get_fdir_cnt_all(hw); + ret = 0; + break; + case ETHTOOL_GRXCLSRULE: + ret = ice_get_ethtool_fdir_entry(hw, cmd); + break; + case ETHTOOL_GRXCLSRLALL: + ret = ice_get_fdir_fltr_ids(hw, cmd, (u32 *)rule_locs); + break; case ETHTOOL_GRXFH: ice_get_rss_hash_opt(vsi, cmd); ret = 0; @@ -2593,7 +2620,7 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring) struct ice_vsi *vsi = np->vsi; struct ice_pf *pf = vsi->back; int i, timeout = 50, err = 0; - u32 new_rx_cnt, new_tx_cnt; + u16 new_rx_cnt, new_tx_cnt; if (ring->tx_pending > ICE_MAX_NUM_DESC || ring->tx_pending < ICE_MIN_NUM_DESC || @@ -2645,8 +2672,8 @@ ice_set_ringparam(struct net_device *netdev, struct ethtool_ringparam *ring) if (ice_is_xdp_ena_vsi(vsi)) for (i = 0; i < vsi->num_xdp_txq; i++) vsi->xdp_rings[i]->count = new_tx_cnt; - vsi->num_tx_desc = new_tx_cnt; - vsi->num_rx_desc = new_rx_cnt; + vsi->num_tx_desc = (u16)new_tx_cnt; + vsi->num_rx_desc = (u16)new_rx_cnt; netdev_dbg(netdev, "Link is down, descriptor count change happens when link is brought up\n"); goto done; } @@ -2952,31 +2979,22 @@ ice_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause) status = ice_set_fc(pi, &aq_failures, link_up); if (aq_failures & ICE_SET_FC_AQ_FAIL_GET) { - netdev_info(netdev, "Set fc failed on the get_phy_capabilities call with err %d aq_err %d\n", - status, hw->adminq.sq_last_status); + netdev_info(netdev, "Set fc failed on the get_phy_capabilities call with err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); err = -EAGAIN; } else if (aq_failures & ICE_SET_FC_AQ_FAIL_SET) { - netdev_info(netdev, "Set fc failed on the set_phy_config call with err %d aq_err %d\n", - status, hw->adminq.sq_last_status); + netdev_info(netdev, "Set fc failed on the set_phy_config call with err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); err = -EAGAIN; } else if (aq_failures & ICE_SET_FC_AQ_FAIL_UPDATE) { - netdev_info(netdev, "Set fc failed on the get_link_info call with err %d aq_err %d\n", - status, hw->adminq.sq_last_status); + netdev_info(netdev, "Set fc failed on the get_link_info call with err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); err = -EAGAIN; } - if (!test_bit(__ICE_DOWN, pf->state)) { - /* Give it a little more time to try to come back. If still - * down, restart autoneg link or reinitialize the interface. - */ - msleep(75); - if (!test_bit(__ICE_DOWN, pf->state)) - return ice_nway_reset(netdev); - - ice_down(vsi); - ice_up(vsi); - } - return err; } @@ -3171,10 +3189,6 @@ ice_get_channels(struct net_device *dev, struct ethtool_channels *ch) struct ice_vsi *vsi = np->vsi; struct ice_pf *pf = vsi->back; - /* check to see if VSI is active */ - if (test_bit(__ICE_DOWN, vsi->state)) - return; - /* report maximum channels */ ch->max_rx = ice_get_max_rxq(pf); ch->max_tx = ice_get_max_txq(pf); @@ -3184,6 +3198,10 @@ ice_get_channels(struct net_device *dev, struct ethtool_channels *ch) ch->combined_count = ice_get_combined_cnt(vsi); ch->rx_count = vsi->num_rxq - ch->combined_count; ch->tx_count = vsi->num_txq - ch->combined_count; + + /* report other queues */ + ch->other_count = test_bit(ICE_FLAG_FD_ENA, pf->flags) ? 1 : 0; + ch->max_other = ch->other_count; } /** @@ -3227,8 +3245,9 @@ static int ice_vsi_set_dflt_rss_lut(struct ice_vsi *vsi, int req_rss_size) status = ice_aq_set_rss_lut(hw, vsi->idx, vsi->rss_lut_type, lut, vsi->rss_table_size); if (status) { - dev_err(dev, "Cannot set RSS lut, err %d aq_err %d\n", - status, hw->adminq.rq_last_status); + dev_err(dev, "Cannot set RSS lut, err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); err = -EIO; } @@ -3255,9 +3274,14 @@ static int ice_set_channels(struct net_device *dev, struct ethtool_channels *ch) return -EOPNOTSUPP; } /* do not support changing other_count */ - if (ch->other_count) + if (ch->other_count != (test_bit(ICE_FLAG_FD_ENA, pf->flags) ? 1U : 0U)) return -EINVAL; + if (test_bit(ICE_FLAG_FD_ENA, pf->flags) && pf->hw.fdir_active_fltr) { + netdev_err(dev, "Cannot set channels when Flow Director filters are active\n"); + return -EOPNOTSUPP; + } + curr_combined = ice_get_combined_cnt(vsi); /* these checks are for cases where user didn't specify a particular @@ -3731,10 +3755,10 @@ ice_get_module_eeprom(struct net_device *netdev, struct ice_hw *hw = &pf->hw; enum ice_status status; bool is_sfp = false; + unsigned int i; u16 offset = 0; u8 value = 0; u8 page = 0; - int i; status = ice_aq_sff_eeprom(hw, 0, addr, offset, page, 0, &value, 1, 0, NULL); diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c b/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c new file mode 100644 index 000000000000..d7430ce6af26 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_ethtool_fdir.c @@ -0,0 +1,1697 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (C) 2018-2020, Intel Corporation. */ + +/* flow director ethtool support for ice */ + +#include "ice.h" +#include "ice_lib.h" +#include "ice_flow.h" + +static struct in6_addr full_ipv6_addr_mask = { + .in6_u = { + .u6_addr8 = { + 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, + 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, + } + } +}; + +static struct in6_addr zero_ipv6_addr_mask = { + .in6_u = { + .u6_addr8 = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + } + } +}; + +/* calls to ice_flow_add_prof require the number of segments in the array + * for segs_cnt. In this code that is one more than the index. + */ +#define TNL_SEG_CNT(_TNL_) ((_TNL_) + 1) + +/** + * ice_fltr_to_ethtool_flow - convert filter type values to ethtool + * flow type values + * @flow: filter type to be converted + * + * Returns the corresponding ethtool flow type. + */ +static int ice_fltr_to_ethtool_flow(enum ice_fltr_ptype flow) +{ + switch (flow) { + case ICE_FLTR_PTYPE_NONF_IPV4_TCP: + return TCP_V4_FLOW; + case ICE_FLTR_PTYPE_NONF_IPV4_UDP: + return UDP_V4_FLOW; + case ICE_FLTR_PTYPE_NONF_IPV4_SCTP: + return SCTP_V4_FLOW; + case ICE_FLTR_PTYPE_NONF_IPV4_OTHER: + return IPV4_USER_FLOW; + case ICE_FLTR_PTYPE_NONF_IPV6_TCP: + return TCP_V6_FLOW; + case ICE_FLTR_PTYPE_NONF_IPV6_UDP: + return UDP_V6_FLOW; + case ICE_FLTR_PTYPE_NONF_IPV6_SCTP: + return SCTP_V6_FLOW; + case ICE_FLTR_PTYPE_NONF_IPV6_OTHER: + return IPV6_USER_FLOW; + default: + /* 0 is undefined ethtool flow */ + return 0; + } +} + +/** + * ice_ethtool_flow_to_fltr - convert ethtool flow type to filter enum + * @eth: Ethtool flow type to be converted + * + * Returns flow enum + */ +static enum ice_fltr_ptype ice_ethtool_flow_to_fltr(int eth) +{ + switch (eth) { + case TCP_V4_FLOW: + return ICE_FLTR_PTYPE_NONF_IPV4_TCP; + case UDP_V4_FLOW: + return ICE_FLTR_PTYPE_NONF_IPV4_UDP; + case SCTP_V4_FLOW: + return ICE_FLTR_PTYPE_NONF_IPV4_SCTP; + case IPV4_USER_FLOW: + return ICE_FLTR_PTYPE_NONF_IPV4_OTHER; + case TCP_V6_FLOW: + return ICE_FLTR_PTYPE_NONF_IPV6_TCP; + case UDP_V6_FLOW: + return ICE_FLTR_PTYPE_NONF_IPV6_UDP; + case SCTP_V6_FLOW: + return ICE_FLTR_PTYPE_NONF_IPV6_SCTP; + case IPV6_USER_FLOW: + return ICE_FLTR_PTYPE_NONF_IPV6_OTHER; + default: + return ICE_FLTR_PTYPE_NONF_NONE; + } +} + +/** + * ice_is_mask_valid - check mask field set + * @mask: full mask to check + * @field: field for which mask should be valid + * + * If the mask is fully set return true. If it is not valid for field return + * false. + */ +static bool ice_is_mask_valid(u64 mask, u64 field) +{ + return (mask & field) == field; +} + +/** + * ice_get_ethtool_fdir_entry - fill ethtool structure with fdir filter data + * @hw: hardware structure that contains filter list + * @cmd: ethtool command data structure to receive the filter data + * + * Returns 0 on success and -EINVAL on failure + */ +int ice_get_ethtool_fdir_entry(struct ice_hw *hw, struct ethtool_rxnfc *cmd) +{ + struct ethtool_rx_flow_spec *fsp; + struct ice_fdir_fltr *rule; + int ret = 0; + u16 idx; + + fsp = (struct ethtool_rx_flow_spec *)&cmd->fs; + + mutex_lock(&hw->fdir_fltr_lock); + + rule = ice_fdir_find_fltr_by_idx(hw, fsp->location); + + if (!rule || fsp->location != rule->fltr_id) { + ret = -EINVAL; + goto release_lock; + } + + fsp->flow_type = ice_fltr_to_ethtool_flow(rule->flow_type); + + memset(&fsp->m_u, 0, sizeof(fsp->m_u)); + memset(&fsp->m_ext, 0, sizeof(fsp->m_ext)); + + switch (fsp->flow_type) { + case IPV4_USER_FLOW: + fsp->h_u.usr_ip4_spec.ip_ver = ETH_RX_NFC_IP4; + fsp->h_u.usr_ip4_spec.proto = 0; + fsp->h_u.usr_ip4_spec.l4_4_bytes = rule->ip.v4.l4_header; + fsp->h_u.usr_ip4_spec.tos = rule->ip.v4.tos; + fsp->h_u.usr_ip4_spec.ip4src = rule->ip.v4.src_ip; + fsp->h_u.usr_ip4_spec.ip4dst = rule->ip.v4.dst_ip; + fsp->m_u.usr_ip4_spec.ip4src = rule->mask.v4.src_ip; + fsp->m_u.usr_ip4_spec.ip4dst = rule->mask.v4.dst_ip; + fsp->m_u.usr_ip4_spec.ip_ver = 0xFF; + fsp->m_u.usr_ip4_spec.proto = 0; + fsp->m_u.usr_ip4_spec.l4_4_bytes = rule->mask.v4.l4_header; + fsp->m_u.usr_ip4_spec.tos = rule->mask.v4.tos; + break; + case TCP_V4_FLOW: + case UDP_V4_FLOW: + case SCTP_V4_FLOW: + fsp->h_u.tcp_ip4_spec.psrc = rule->ip.v4.src_port; + fsp->h_u.tcp_ip4_spec.pdst = rule->ip.v4.dst_port; + fsp->h_u.tcp_ip4_spec.ip4src = rule->ip.v4.src_ip; + fsp->h_u.tcp_ip4_spec.ip4dst = rule->ip.v4.dst_ip; + fsp->m_u.tcp_ip4_spec.psrc = rule->mask.v4.src_port; + fsp->m_u.tcp_ip4_spec.pdst = rule->mask.v4.dst_port; + fsp->m_u.tcp_ip4_spec.ip4src = rule->mask.v4.src_ip; + fsp->m_u.tcp_ip4_spec.ip4dst = rule->mask.v4.dst_ip; + break; + case IPV6_USER_FLOW: + fsp->h_u.usr_ip6_spec.l4_4_bytes = rule->ip.v6.l4_header; + fsp->h_u.usr_ip6_spec.tclass = rule->ip.v6.tc; + fsp->h_u.usr_ip6_spec.l4_proto = rule->ip.v6.proto; + memcpy(fsp->h_u.tcp_ip6_spec.ip6src, rule->ip.v6.src_ip, + sizeof(struct in6_addr)); + memcpy(fsp->h_u.tcp_ip6_spec.ip6dst, rule->ip.v6.dst_ip, + sizeof(struct in6_addr)); + memcpy(fsp->m_u.tcp_ip6_spec.ip6src, rule->mask.v6.src_ip, + sizeof(struct in6_addr)); + memcpy(fsp->m_u.tcp_ip6_spec.ip6dst, rule->mask.v6.dst_ip, + sizeof(struct in6_addr)); + fsp->m_u.usr_ip6_spec.l4_4_bytes = rule->mask.v6.l4_header; + fsp->m_u.usr_ip6_spec.tclass = rule->mask.v6.tc; + fsp->m_u.usr_ip6_spec.l4_proto = rule->mask.v6.proto; + break; + case TCP_V6_FLOW: + case UDP_V6_FLOW: + case SCTP_V6_FLOW: + memcpy(fsp->h_u.tcp_ip6_spec.ip6src, rule->ip.v6.src_ip, + sizeof(struct in6_addr)); + memcpy(fsp->h_u.tcp_ip6_spec.ip6dst, rule->ip.v6.dst_ip, + sizeof(struct in6_addr)); + fsp->h_u.tcp_ip6_spec.psrc = rule->ip.v6.src_port; + fsp->h_u.tcp_ip6_spec.pdst = rule->ip.v6.dst_port; + memcpy(fsp->m_u.tcp_ip6_spec.ip6src, + rule->mask.v6.src_ip, + sizeof(struct in6_addr)); + memcpy(fsp->m_u.tcp_ip6_spec.ip6dst, + rule->mask.v6.dst_ip, + sizeof(struct in6_addr)); + fsp->m_u.tcp_ip6_spec.psrc = rule->mask.v6.src_port; + fsp->m_u.tcp_ip6_spec.pdst = rule->mask.v6.dst_port; + fsp->h_u.tcp_ip6_spec.tclass = rule->ip.v6.tc; + fsp->m_u.tcp_ip6_spec.tclass = rule->mask.v6.tc; + break; + default: + break; + } + + if (rule->dest_ctl == ICE_FLTR_PRGM_DESC_DEST_DROP_PKT) + fsp->ring_cookie = RX_CLS_FLOW_DISC; + else + fsp->ring_cookie = rule->q_index; + + idx = ice_ethtool_flow_to_fltr(fsp->flow_type); + if (idx == ICE_FLTR_PTYPE_NONF_NONE) { + dev_err(ice_hw_to_dev(hw), "Missing input index for flow_type %d\n", + rule->flow_type); + ret = -EINVAL; + } + +release_lock: + mutex_unlock(&hw->fdir_fltr_lock); + return ret; +} + +/** + * ice_get_fdir_fltr_ids - fill buffer with filter IDs of active filters + * @hw: hardware structure containing the filter list + * @cmd: ethtool command data structure + * @rule_locs: ethtool array passed in from OS to receive filter IDs + * + * Returns 0 as expected for success by ethtool + */ +int +ice_get_fdir_fltr_ids(struct ice_hw *hw, struct ethtool_rxnfc *cmd, + u32 *rule_locs) +{ + struct ice_fdir_fltr *f_rule; + unsigned int cnt = 0; + int val = 0; + + /* report total rule count */ + cmd->data = ice_get_fdir_cnt_all(hw); + + mutex_lock(&hw->fdir_fltr_lock); + + list_for_each_entry(f_rule, &hw->fdir_list_head, fltr_node) { + if (cnt == cmd->rule_cnt) { + val = -EMSGSIZE; + goto release_lock; + } + rule_locs[cnt] = f_rule->fltr_id; + cnt++; + } + +release_lock: + mutex_unlock(&hw->fdir_fltr_lock); + if (!val) + cmd->rule_cnt = cnt; + return val; +} + +/** + * ice_fdir_get_hw_prof - return the ice_fd_hw_proc associated with a flow + * @hw: hardware structure containing the filter list + * @blk: hardware block + * @flow: FDir flow type to release + */ +static struct ice_fd_hw_prof * +ice_fdir_get_hw_prof(struct ice_hw *hw, enum ice_block blk, int flow) +{ + if (blk == ICE_BLK_FD && hw->fdir_prof) + return hw->fdir_prof[flow]; + + return NULL; +} + +/** + * ice_fdir_erase_flow_from_hw - remove a flow from the HW profile tables + * @hw: hardware structure containing the filter list + * @blk: hardware block + * @flow: FDir flow type to release + */ +static void +ice_fdir_erase_flow_from_hw(struct ice_hw *hw, enum ice_block blk, int flow) +{ + struct ice_fd_hw_prof *prof = ice_fdir_get_hw_prof(hw, blk, flow); + int tun; + + if (!prof) + return; + + for (tun = 0; tun < ICE_FD_HW_SEG_MAX; tun++) { + u64 prof_id; + int j; + + prof_id = flow + tun * ICE_FLTR_PTYPE_MAX; + for (j = 0; j < prof->cnt; j++) { + u16 vsi_num; + + if (!prof->entry_h[j][tun] || !prof->vsi_h[j]) + continue; + vsi_num = ice_get_hw_vsi_num(hw, prof->vsi_h[j]); + ice_rem_prof_id_flow(hw, blk, vsi_num, prof_id); + ice_flow_rem_entry(hw, blk, prof->entry_h[j][tun]); + prof->entry_h[j][tun] = 0; + } + ice_flow_rem_prof(hw, blk, prof_id); + } +} + +/** + * ice_fdir_rem_flow - release the ice_flow structures for a filter type + * @hw: hardware structure containing the filter list + * @blk: hardware block + * @flow_type: FDir flow type to release + */ +static void +ice_fdir_rem_flow(struct ice_hw *hw, enum ice_block blk, + enum ice_fltr_ptype flow_type) +{ + int flow = (int)flow_type & ~FLOW_EXT; + struct ice_fd_hw_prof *prof; + int tun, i; + + prof = ice_fdir_get_hw_prof(hw, blk, flow); + if (!prof) + return; + + ice_fdir_erase_flow_from_hw(hw, blk, flow); + for (i = 0; i < prof->cnt; i++) + prof->vsi_h[i] = 0; + for (tun = 0; tun < ICE_FD_HW_SEG_MAX; tun++) { + if (!prof->fdir_seg[tun]) + continue; + devm_kfree(ice_hw_to_dev(hw), prof->fdir_seg[tun]); + prof->fdir_seg[tun] = NULL; + } + prof->cnt = 0; +} + +/** + * ice_fdir_release_flows - release all flows in use for later replay + * @hw: pointer to HW instance + */ +void ice_fdir_release_flows(struct ice_hw *hw) +{ + int flow; + + /* release Flow Director HW table entries */ + for (flow = 0; flow < ICE_FLTR_PTYPE_MAX; flow++) + ice_fdir_erase_flow_from_hw(hw, ICE_BLK_FD, flow); +} + +/** + * ice_fdir_replay_flows - replay HW Flow Director filter info + * @hw: pointer to HW instance + */ +void ice_fdir_replay_flows(struct ice_hw *hw) +{ + int flow; + + for (flow = 0; flow < ICE_FLTR_PTYPE_MAX; flow++) { + int tun; + + if (!hw->fdir_prof[flow] || !hw->fdir_prof[flow]->cnt) + continue; + for (tun = 0; tun < ICE_FD_HW_SEG_MAX; tun++) { + struct ice_flow_prof *hw_prof; + struct ice_fd_hw_prof *prof; + u64 prof_id; + int j; + + prof = hw->fdir_prof[flow]; + prof_id = flow + tun * ICE_FLTR_PTYPE_MAX; + ice_flow_add_prof(hw, ICE_BLK_FD, ICE_FLOW_RX, prof_id, + prof->fdir_seg[tun], TNL_SEG_CNT(tun), + &hw_prof); + for (j = 0; j < prof->cnt; j++) { + enum ice_flow_priority prio; + u64 entry_h = 0; + int err; + + prio = ICE_FLOW_PRIO_NORMAL; + err = ice_flow_add_entry(hw, ICE_BLK_FD, + prof_id, + prof->vsi_h[0], + prof->vsi_h[j], + prio, prof->fdir_seg, + &entry_h); + if (err) { + dev_err(ice_hw_to_dev(hw), "Could not replay Flow Director, flow type %d\n", + flow); + continue; + } + prof->entry_h[j][tun] = entry_h; + } + } + } +} + +/** + * ice_parse_rx_flow_user_data - deconstruct user-defined data + * @fsp: pointer to ethtool Rx flow specification + * @data: pointer to userdef data structure for storage + * + * Returns 0 on success, negative error value on failure + */ +static int +ice_parse_rx_flow_user_data(struct ethtool_rx_flow_spec *fsp, + struct ice_rx_flow_userdef *data) +{ + u64 value, mask; + + memset(data, 0, sizeof(*data)); + if (!(fsp->flow_type & FLOW_EXT)) + return 0; + + value = be64_to_cpu(*((__force __be64 *)fsp->h_ext.data)); + mask = be64_to_cpu(*((__force __be64 *)fsp->m_ext.data)); + if (!mask) + return 0; + +#define ICE_USERDEF_FLEX_WORD_M GENMASK_ULL(15, 0) +#define ICE_USERDEF_FLEX_OFFS_S 16 +#define ICE_USERDEF_FLEX_OFFS_M GENMASK_ULL(31, ICE_USERDEF_FLEX_OFFS_S) +#define ICE_USERDEF_FLEX_FLTR_M GENMASK_ULL(31, 0) + + /* 0x1fe is the maximum value for offsets stored in the internal + * filtering tables. + */ +#define ICE_USERDEF_FLEX_MAX_OFFS_VAL 0x1fe + + if (!ice_is_mask_valid(mask, ICE_USERDEF_FLEX_FLTR_M) || + value > ICE_USERDEF_FLEX_FLTR_M) + return -EINVAL; + + data->flex_word = value & ICE_USERDEF_FLEX_WORD_M; + data->flex_offset = (value & ICE_USERDEF_FLEX_OFFS_M) >> + ICE_USERDEF_FLEX_OFFS_S; + if (data->flex_offset > ICE_USERDEF_FLEX_MAX_OFFS_VAL) + return -EINVAL; + + data->flex_fltr = true; + + return 0; +} + +/** + * ice_fdir_num_avail_fltr - return the number of unused flow director filters + * @hw: pointer to hardware structure + * @vsi: software VSI structure + * + * There are 2 filter pools: guaranteed and best effort(shared). Each VSI can + * use filters from either pool. The guaranteed pool is divided between VSIs. + * The best effort filter pool is common to all VSIs and is a device shared + * resource pool. The number of filters available to this VSI is the sum of + * the VSIs guaranteed filter pool and the global available best effort + * filter pool. + * + * Returns the number of available flow director filters to this VSI + */ +static int ice_fdir_num_avail_fltr(struct ice_hw *hw, struct ice_vsi *vsi) +{ + u16 vsi_num = ice_get_hw_vsi_num(hw, vsi->idx); + u16 num_guar; + u16 num_be; + + /* total guaranteed filters assigned to this VSI */ + num_guar = vsi->num_gfltr; + + /* minus the guaranteed filters programed by this VSI */ + num_guar -= (rd32(hw, VSIQF_FD_CNT(vsi_num)) & + VSIQF_FD_CNT_FD_GCNT_M) >> VSIQF_FD_CNT_FD_GCNT_S; + + /* total global best effort filters */ + num_be = hw->func_caps.fd_fltr_best_effort; + + /* minus the global best effort filters programmed */ + num_be -= (rd32(hw, GLQF_FD_CNT) & GLQF_FD_CNT_FD_BCNT_M) >> + GLQF_FD_CNT_FD_BCNT_S; + + return num_guar + num_be; +} + +/** + * ice_fdir_alloc_flow_prof - allocate FDir flow profile structure(s) + * @hw: HW structure containing the FDir flow profile structure(s) + * @flow: flow type to allocate the flow profile for + * + * Allocate the fdir_prof and fdir_prof[flow] if not already created. Return 0 + * on success and negative on error. + */ +static int +ice_fdir_alloc_flow_prof(struct ice_hw *hw, enum ice_fltr_ptype flow) +{ + if (!hw) + return -EINVAL; + + if (!hw->fdir_prof) { + hw->fdir_prof = devm_kcalloc(ice_hw_to_dev(hw), + ICE_FLTR_PTYPE_MAX, + sizeof(*hw->fdir_prof), + GFP_KERNEL); + if (!hw->fdir_prof) + return -ENOMEM; + } + + if (!hw->fdir_prof[flow]) { + hw->fdir_prof[flow] = devm_kzalloc(ice_hw_to_dev(hw), + sizeof(**hw->fdir_prof), + GFP_KERNEL); + if (!hw->fdir_prof[flow]) + return -ENOMEM; + } + + return 0; +} + +/** + * ice_fdir_set_hw_fltr_rule - Configure HW tables to generate a FDir rule + * @pf: pointer to the PF structure + * @seg: protocol header description pointer + * @flow: filter enum + * @tun: FDir segment to program + */ +static int +ice_fdir_set_hw_fltr_rule(struct ice_pf *pf, struct ice_flow_seg_info *seg, + enum ice_fltr_ptype flow, enum ice_fd_hw_seg tun) +{ + struct device *dev = ice_pf_to_dev(pf); + struct ice_vsi *main_vsi, *ctrl_vsi; + struct ice_flow_seg_info *old_seg; + struct ice_flow_prof *prof = NULL; + struct ice_fd_hw_prof *hw_prof; + struct ice_hw *hw = &pf->hw; + enum ice_status status; + u64 entry1_h = 0; + u64 entry2_h = 0; + u64 prof_id; + int err; + + main_vsi = ice_get_main_vsi(pf); + if (!main_vsi) + return -EINVAL; + + ctrl_vsi = ice_get_ctrl_vsi(pf); + if (!ctrl_vsi) + return -EINVAL; + + err = ice_fdir_alloc_flow_prof(hw, flow); + if (err) + return err; + + hw_prof = hw->fdir_prof[flow]; + old_seg = hw_prof->fdir_seg[tun]; + if (old_seg) { + /* This flow_type already has a changed input set. + * If it matches the requested input set then we are + * done. Or, if it's different then it's an error. + */ + if (!memcmp(old_seg, seg, sizeof(*seg))) + return -EEXIST; + + /* if there are FDir filters using this flow, + * then return error. + */ + if (hw->fdir_fltr_cnt[flow]) { + dev_err(dev, "Failed to add filter. Flow director filters on each port must have the same input set.\n"); + return -EINVAL; + } + + if (ice_is_arfs_using_perfect_flow(hw, flow)) { + dev_err(dev, "aRFS using perfect flow type %d, cannot change input set\n", + flow); + return -EINVAL; + } + + /* remove HW filter definition */ + ice_fdir_rem_flow(hw, ICE_BLK_FD, flow); + } + + /* Adding a profile, but there is only one header supported. + * That is the final parameters are 1 header (segment), no + * actions (NULL) and zero actions 0. + */ + prof_id = flow + tun * ICE_FLTR_PTYPE_MAX; + status = ice_flow_add_prof(hw, ICE_BLK_FD, ICE_FLOW_RX, prof_id, seg, + TNL_SEG_CNT(tun), &prof); + if (status) + return ice_status_to_errno(status); + status = ice_flow_add_entry(hw, ICE_BLK_FD, prof_id, main_vsi->idx, + main_vsi->idx, ICE_FLOW_PRIO_NORMAL, + seg, &entry1_h); + if (status) { + err = ice_status_to_errno(status); + goto err_prof; + } + status = ice_flow_add_entry(hw, ICE_BLK_FD, prof_id, main_vsi->idx, + ctrl_vsi->idx, ICE_FLOW_PRIO_NORMAL, + seg, &entry2_h); + if (status) { + err = ice_status_to_errno(status); + goto err_entry; + } + + hw_prof->fdir_seg[tun] = seg; + hw_prof->entry_h[0][tun] = entry1_h; + hw_prof->entry_h[1][tun] = entry2_h; + hw_prof->vsi_h[0] = main_vsi->idx; + hw_prof->vsi_h[1] = ctrl_vsi->idx; + if (!hw_prof->cnt) + hw_prof->cnt = 2; + + return 0; + +err_entry: + ice_rem_prof_id_flow(hw, ICE_BLK_FD, + ice_get_hw_vsi_num(hw, main_vsi->idx), prof_id); + ice_flow_rem_entry(hw, ICE_BLK_FD, entry1_h); +err_prof: + ice_flow_rem_prof(hw, ICE_BLK_FD, prof_id); + dev_err(dev, "Failed to add filter. Flow director filters on each port must have the same input set.\n"); + + return err; +} + +/** + * ice_set_init_fdir_seg + * @seg: flow segment for programming + * @l3_proto: ICE_FLOW_SEG_HDR_IPV4 or ICE_FLOW_SEG_HDR_IPV6 + * @l4_proto: ICE_FLOW_SEG_HDR_TCP or ICE_FLOW_SEG_HDR_UDP + * + * Set the configuration for perfect filters to the provided flow segment for + * programming the HW filter. This is to be called only when initializing + * filters as this function it assumes no filters exist. + */ +static int +ice_set_init_fdir_seg(struct ice_flow_seg_info *seg, + enum ice_flow_seg_hdr l3_proto, + enum ice_flow_seg_hdr l4_proto) +{ + enum ice_flow_field src_addr, dst_addr, src_port, dst_port; + + if (!seg) + return -EINVAL; + + if (l3_proto == ICE_FLOW_SEG_HDR_IPV4) { + src_addr = ICE_FLOW_FIELD_IDX_IPV4_SA; + dst_addr = ICE_FLOW_FIELD_IDX_IPV4_DA; + } else if (l3_proto == ICE_FLOW_SEG_HDR_IPV6) { + src_addr = ICE_FLOW_FIELD_IDX_IPV6_SA; + dst_addr = ICE_FLOW_FIELD_IDX_IPV6_DA; + } else { + return -EINVAL; + } + + if (l4_proto == ICE_FLOW_SEG_HDR_TCP) { + src_port = ICE_FLOW_FIELD_IDX_TCP_SRC_PORT; + dst_port = ICE_FLOW_FIELD_IDX_TCP_DST_PORT; + } else if (l4_proto == ICE_FLOW_SEG_HDR_UDP) { + src_port = ICE_FLOW_FIELD_IDX_UDP_SRC_PORT; + dst_port = ICE_FLOW_FIELD_IDX_UDP_DST_PORT; + } else { + return -EINVAL; + } + + ICE_FLOW_SET_HDRS(seg, l3_proto | l4_proto); + + /* IP source address */ + ice_flow_set_fld(seg, src_addr, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, false); + + /* IP destination address */ + ice_flow_set_fld(seg, dst_addr, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, false); + + /* Layer 4 source port */ + ice_flow_set_fld(seg, src_port, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, false); + + /* Layer 4 destination port */ + ice_flow_set_fld(seg, dst_port, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, false); + + return 0; +} + +/** + * ice_create_init_fdir_rule + * @pf: PF structure + * @flow: filter enum + * + * Return error value or 0 on success. + */ +static int +ice_create_init_fdir_rule(struct ice_pf *pf, enum ice_fltr_ptype flow) +{ + struct ice_flow_seg_info *seg, *tun_seg; + struct device *dev = ice_pf_to_dev(pf); + struct ice_hw *hw = &pf->hw; + int ret; + + /* if there is already a filter rule for kind return -EINVAL */ + if (hw->fdir_prof && hw->fdir_prof[flow] && + hw->fdir_prof[flow]->fdir_seg[0]) + return -EINVAL; + + seg = devm_kzalloc(dev, sizeof(*seg), GFP_KERNEL); + if (!seg) + return -ENOMEM; + + tun_seg = devm_kzalloc(dev, sizeof(*seg) * ICE_FD_HW_SEG_MAX, + GFP_KERNEL); + if (!tun_seg) { + devm_kfree(dev, seg); + return -ENOMEM; + } + + if (flow == ICE_FLTR_PTYPE_NONF_IPV4_TCP) + ret = ice_set_init_fdir_seg(seg, ICE_FLOW_SEG_HDR_IPV4, + ICE_FLOW_SEG_HDR_TCP); + else if (flow == ICE_FLTR_PTYPE_NONF_IPV4_UDP) + ret = ice_set_init_fdir_seg(seg, ICE_FLOW_SEG_HDR_IPV4, + ICE_FLOW_SEG_HDR_UDP); + else if (flow == ICE_FLTR_PTYPE_NONF_IPV6_TCP) + ret = ice_set_init_fdir_seg(seg, ICE_FLOW_SEG_HDR_IPV6, + ICE_FLOW_SEG_HDR_TCP); + else if (flow == ICE_FLTR_PTYPE_NONF_IPV6_UDP) + ret = ice_set_init_fdir_seg(seg, ICE_FLOW_SEG_HDR_IPV6, + ICE_FLOW_SEG_HDR_UDP); + else + ret = -EINVAL; + if (ret) + goto err_exit; + + /* add filter for outer headers */ + ret = ice_fdir_set_hw_fltr_rule(pf, seg, flow, ICE_FD_HW_SEG_NON_TUN); + if (ret) + /* could not write filter, free memory */ + goto err_exit; + + /* make tunneled filter HW entries if possible */ + memcpy(&tun_seg[1], seg, sizeof(*seg)); + ret = ice_fdir_set_hw_fltr_rule(pf, tun_seg, flow, ICE_FD_HW_SEG_TUN); + if (ret) + /* could not write tunnel filter, but outer header filter + * exists + */ + devm_kfree(dev, tun_seg); + + set_bit(flow, hw->fdir_perfect_fltr); + return ret; +err_exit: + devm_kfree(dev, tun_seg); + devm_kfree(dev, seg); + + return -EOPNOTSUPP; +} + +/** + * ice_set_fdir_ip4_seg + * @seg: flow segment for programming + * @tcp_ip4_spec: mask data from ethtool + * @l4_proto: Layer 4 protocol to program + * @perfect_fltr: only valid on success; returns true if perfect filter, + * false if not + * + * Set the mask data into the flow segment to be used to program HW + * table based on provided L4 protocol for IPv4 + */ +static int +ice_set_fdir_ip4_seg(struct ice_flow_seg_info *seg, + struct ethtool_tcpip4_spec *tcp_ip4_spec, + enum ice_flow_seg_hdr l4_proto, bool *perfect_fltr) +{ + enum ice_flow_field src_port, dst_port; + + /* make sure we don't have any empty rule */ + if (!tcp_ip4_spec->psrc && !tcp_ip4_spec->ip4src && + !tcp_ip4_spec->pdst && !tcp_ip4_spec->ip4dst) + return -EINVAL; + + /* filtering on TOS not supported */ + if (tcp_ip4_spec->tos) + return -EOPNOTSUPP; + + if (l4_proto == ICE_FLOW_SEG_HDR_TCP) { + src_port = ICE_FLOW_FIELD_IDX_TCP_SRC_PORT; + dst_port = ICE_FLOW_FIELD_IDX_TCP_DST_PORT; + } else if (l4_proto == ICE_FLOW_SEG_HDR_UDP) { + src_port = ICE_FLOW_FIELD_IDX_UDP_SRC_PORT; + dst_port = ICE_FLOW_FIELD_IDX_UDP_DST_PORT; + } else if (l4_proto == ICE_FLOW_SEG_HDR_SCTP) { + src_port = ICE_FLOW_FIELD_IDX_SCTP_SRC_PORT; + dst_port = ICE_FLOW_FIELD_IDX_SCTP_DST_PORT; + } else { + return -EOPNOTSUPP; + } + + *perfect_fltr = true; + ICE_FLOW_SET_HDRS(seg, ICE_FLOW_SEG_HDR_IPV4 | l4_proto); + + /* IP source address */ + if (tcp_ip4_spec->ip4src == htonl(0xFFFFFFFF)) + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_IPV4_SA, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + else if (!tcp_ip4_spec->ip4src) + *perfect_fltr = false; + else + return -EOPNOTSUPP; + + /* IP destination address */ + if (tcp_ip4_spec->ip4dst == htonl(0xFFFFFFFF)) + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_IPV4_DA, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + else if (!tcp_ip4_spec->ip4dst) + *perfect_fltr = false; + else + return -EOPNOTSUPP; + + /* Layer 4 source port */ + if (tcp_ip4_spec->psrc == htons(0xFFFF)) + ice_flow_set_fld(seg, src_port, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + false); + else if (!tcp_ip4_spec->psrc) + *perfect_fltr = false; + else + return -EOPNOTSUPP; + + /* Layer 4 destination port */ + if (tcp_ip4_spec->pdst == htons(0xFFFF)) + ice_flow_set_fld(seg, dst_port, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + false); + else if (!tcp_ip4_spec->pdst) + *perfect_fltr = false; + else + return -EOPNOTSUPP; + + return 0; +} + +/** + * ice_set_fdir_ip4_usr_seg + * @seg: flow segment for programming + * @usr_ip4_spec: ethtool userdef packet offset + * @perfect_fltr: only valid on success; returns true if perfect filter, + * false if not + * + * Set the offset data into the flow segment to be used to program HW + * table for IPv4 + */ +static int +ice_set_fdir_ip4_usr_seg(struct ice_flow_seg_info *seg, + struct ethtool_usrip4_spec *usr_ip4_spec, + bool *perfect_fltr) +{ + /* first 4 bytes of Layer 4 header */ + if (usr_ip4_spec->l4_4_bytes) + return -EINVAL; + if (usr_ip4_spec->tos) + return -EINVAL; + if (usr_ip4_spec->ip_ver) + return -EINVAL; + /* Filtering on Layer 4 protocol not supported */ + if (usr_ip4_spec->proto) + return -EOPNOTSUPP; + /* empty rules are not valid */ + if (!usr_ip4_spec->ip4src && !usr_ip4_spec->ip4dst) + return -EINVAL; + + *perfect_fltr = true; + ICE_FLOW_SET_HDRS(seg, ICE_FLOW_SEG_HDR_IPV4); + + /* IP source address */ + if (usr_ip4_spec->ip4src == htonl(0xFFFFFFFF)) + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_IPV4_SA, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + else if (!usr_ip4_spec->ip4src) + *perfect_fltr = false; + else + return -EOPNOTSUPP; + + /* IP destination address */ + if (usr_ip4_spec->ip4dst == htonl(0xFFFFFFFF)) + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_IPV4_DA, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + else if (!usr_ip4_spec->ip4dst) + *perfect_fltr = false; + else + return -EOPNOTSUPP; + + return 0; +} + +/** + * ice_set_fdir_ip6_seg + * @seg: flow segment for programming + * @tcp_ip6_spec: mask data from ethtool + * @l4_proto: Layer 4 protocol to program + * @perfect_fltr: only valid on success; returns true if perfect filter, + * false if not + * + * Set the mask data into the flow segment to be used to program HW + * table based on provided L4 protocol for IPv6 + */ +static int +ice_set_fdir_ip6_seg(struct ice_flow_seg_info *seg, + struct ethtool_tcpip6_spec *tcp_ip6_spec, + enum ice_flow_seg_hdr l4_proto, bool *perfect_fltr) +{ + enum ice_flow_field src_port, dst_port; + + /* make sure we don't have any empty rule */ + if (!memcmp(tcp_ip6_spec->ip6src, &zero_ipv6_addr_mask, + sizeof(struct in6_addr)) && + !memcmp(tcp_ip6_spec->ip6dst, &zero_ipv6_addr_mask, + sizeof(struct in6_addr)) && + !tcp_ip6_spec->psrc && !tcp_ip6_spec->pdst) + return -EINVAL; + + /* filtering on TC not supported */ + if (tcp_ip6_spec->tclass) + return -EOPNOTSUPP; + + if (l4_proto == ICE_FLOW_SEG_HDR_TCP) { + src_port = ICE_FLOW_FIELD_IDX_TCP_SRC_PORT; + dst_port = ICE_FLOW_FIELD_IDX_TCP_DST_PORT; + } else if (l4_proto == ICE_FLOW_SEG_HDR_UDP) { + src_port = ICE_FLOW_FIELD_IDX_UDP_SRC_PORT; + dst_port = ICE_FLOW_FIELD_IDX_UDP_DST_PORT; + } else if (l4_proto == ICE_FLOW_SEG_HDR_SCTP) { + src_port = ICE_FLOW_FIELD_IDX_SCTP_SRC_PORT; + dst_port = ICE_FLOW_FIELD_IDX_SCTP_DST_PORT; + } else { + return -EINVAL; + } + + *perfect_fltr = true; + ICE_FLOW_SET_HDRS(seg, ICE_FLOW_SEG_HDR_IPV6 | l4_proto); + + if (!memcmp(tcp_ip6_spec->ip6src, &full_ipv6_addr_mask, + sizeof(struct in6_addr))) + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_IPV6_SA, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + else if (!memcmp(tcp_ip6_spec->ip6src, &zero_ipv6_addr_mask, + sizeof(struct in6_addr))) + *perfect_fltr = false; + else + return -EOPNOTSUPP; + + if (!memcmp(tcp_ip6_spec->ip6dst, &full_ipv6_addr_mask, + sizeof(struct in6_addr))) + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_IPV6_DA, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + else if (!memcmp(tcp_ip6_spec->ip6dst, &zero_ipv6_addr_mask, + sizeof(struct in6_addr))) + *perfect_fltr = false; + else + return -EOPNOTSUPP; + + /* Layer 4 source port */ + if (tcp_ip6_spec->psrc == htons(0xFFFF)) + ice_flow_set_fld(seg, src_port, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + false); + else if (!tcp_ip6_spec->psrc) + *perfect_fltr = false; + else + return -EOPNOTSUPP; + + /* Layer 4 destination port */ + if (tcp_ip6_spec->pdst == htons(0xFFFF)) + ice_flow_set_fld(seg, dst_port, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + false); + else if (!tcp_ip6_spec->pdst) + *perfect_fltr = false; + else + return -EOPNOTSUPP; + + return 0; +} + +/** + * ice_set_fdir_ip6_usr_seg + * @seg: flow segment for programming + * @usr_ip6_spec: ethtool userdef packet offset + * @perfect_fltr: only valid on success; returns true if perfect filter, + * false if not + * + * Set the offset data into the flow segment to be used to program HW + * table for IPv6 + */ +static int +ice_set_fdir_ip6_usr_seg(struct ice_flow_seg_info *seg, + struct ethtool_usrip6_spec *usr_ip6_spec, + bool *perfect_fltr) +{ + /* filtering on Layer 4 bytes not supported */ + if (usr_ip6_spec->l4_4_bytes) + return -EOPNOTSUPP; + /* filtering on TC not supported */ + if (usr_ip6_spec->tclass) + return -EOPNOTSUPP; + /* filtering on Layer 4 protocol not supported */ + if (usr_ip6_spec->l4_proto) + return -EOPNOTSUPP; + /* empty rules are not valid */ + if (!memcmp(usr_ip6_spec->ip6src, &zero_ipv6_addr_mask, + sizeof(struct in6_addr)) && + !memcmp(usr_ip6_spec->ip6dst, &zero_ipv6_addr_mask, + sizeof(struct in6_addr))) + return -EINVAL; + + *perfect_fltr = true; + ICE_FLOW_SET_HDRS(seg, ICE_FLOW_SEG_HDR_IPV6); + + if (!memcmp(usr_ip6_spec->ip6src, &full_ipv6_addr_mask, + sizeof(struct in6_addr))) + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_IPV6_SA, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + else if (!memcmp(usr_ip6_spec->ip6src, &zero_ipv6_addr_mask, + sizeof(struct in6_addr))) + *perfect_fltr = false; + else + return -EOPNOTSUPP; + + if (!memcmp(usr_ip6_spec->ip6dst, &full_ipv6_addr_mask, + sizeof(struct in6_addr))) + ice_flow_set_fld(seg, ICE_FLOW_FIELD_IDX_IPV6_DA, + ICE_FLOW_FLD_OFF_INVAL, ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL, false); + else if (!memcmp(usr_ip6_spec->ip6dst, &zero_ipv6_addr_mask, + sizeof(struct in6_addr))) + *perfect_fltr = false; + else + return -EOPNOTSUPP; + + return 0; +} + +/** + * ice_cfg_fdir_xtrct_seq - Configure extraction sequence for the given filter + * @pf: PF structure + * @fsp: pointer to ethtool Rx flow specification + * @user: user defined data from flow specification + * + * Returns 0 on success. + */ +static int +ice_cfg_fdir_xtrct_seq(struct ice_pf *pf, struct ethtool_rx_flow_spec *fsp, + struct ice_rx_flow_userdef *user) +{ + struct ice_flow_seg_info *seg, *tun_seg; + struct device *dev = ice_pf_to_dev(pf); + enum ice_fltr_ptype fltr_idx; + struct ice_hw *hw = &pf->hw; + bool perfect_filter; + int ret; + + seg = devm_kzalloc(dev, sizeof(*seg), GFP_KERNEL); + if (!seg) + return -ENOMEM; + + tun_seg = devm_kzalloc(dev, sizeof(*seg) * ICE_FD_HW_SEG_MAX, + GFP_KERNEL); + if (!tun_seg) { + devm_kfree(dev, seg); + return -ENOMEM; + } + + switch (fsp->flow_type & ~FLOW_EXT) { + case TCP_V4_FLOW: + ret = ice_set_fdir_ip4_seg(seg, &fsp->m_u.tcp_ip4_spec, + ICE_FLOW_SEG_HDR_TCP, + &perfect_filter); + break; + case UDP_V4_FLOW: + ret = ice_set_fdir_ip4_seg(seg, &fsp->m_u.tcp_ip4_spec, + ICE_FLOW_SEG_HDR_UDP, + &perfect_filter); + break; + case SCTP_V4_FLOW: + ret = ice_set_fdir_ip4_seg(seg, &fsp->m_u.tcp_ip4_spec, + ICE_FLOW_SEG_HDR_SCTP, + &perfect_filter); + break; + case IPV4_USER_FLOW: + ret = ice_set_fdir_ip4_usr_seg(seg, &fsp->m_u.usr_ip4_spec, + &perfect_filter); + break; + case TCP_V6_FLOW: + ret = ice_set_fdir_ip6_seg(seg, &fsp->m_u.tcp_ip6_spec, + ICE_FLOW_SEG_HDR_TCP, + &perfect_filter); + break; + case UDP_V6_FLOW: + ret = ice_set_fdir_ip6_seg(seg, &fsp->m_u.tcp_ip6_spec, + ICE_FLOW_SEG_HDR_UDP, + &perfect_filter); + break; + case SCTP_V6_FLOW: + ret = ice_set_fdir_ip6_seg(seg, &fsp->m_u.tcp_ip6_spec, + ICE_FLOW_SEG_HDR_SCTP, + &perfect_filter); + break; + case IPV6_USER_FLOW: + ret = ice_set_fdir_ip6_usr_seg(seg, &fsp->m_u.usr_ip6_spec, + &perfect_filter); + break; + default: + ret = -EINVAL; + } + if (ret) + goto err_exit; + + /* tunnel segments are shifted up one. */ + memcpy(&tun_seg[1], seg, sizeof(*seg)); + + if (user && user->flex_fltr) { + perfect_filter = false; + ice_flow_add_fld_raw(seg, user->flex_offset, + ICE_FLTR_PRGM_FLEX_WORD_SIZE, + ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL); + ice_flow_add_fld_raw(&tun_seg[1], user->flex_offset, + ICE_FLTR_PRGM_FLEX_WORD_SIZE, + ICE_FLOW_FLD_OFF_INVAL, + ICE_FLOW_FLD_OFF_INVAL); + } + + /* add filter for outer headers */ + fltr_idx = ice_ethtool_flow_to_fltr(fsp->flow_type & ~FLOW_EXT); + ret = ice_fdir_set_hw_fltr_rule(pf, seg, fltr_idx, + ICE_FD_HW_SEG_NON_TUN); + if (ret == -EEXIST) + /* Rule already exists, free memory and continue */ + devm_kfree(dev, seg); + else if (ret) + /* could not write filter, free memory */ + goto err_exit; + + /* make tunneled filter HW entries if possible */ + memcpy(&tun_seg[1], seg, sizeof(*seg)); + ret = ice_fdir_set_hw_fltr_rule(pf, tun_seg, fltr_idx, + ICE_FD_HW_SEG_TUN); + if (ret == -EEXIST) { + /* Rule already exists, free memory and count as success */ + devm_kfree(dev, tun_seg); + ret = 0; + } else if (ret) { + /* could not write tunnel filter, but outer filter exists */ + devm_kfree(dev, tun_seg); + } + + if (perfect_filter) + set_bit(fltr_idx, hw->fdir_perfect_fltr); + else + clear_bit(fltr_idx, hw->fdir_perfect_fltr); + + return ret; + +err_exit: + devm_kfree(dev, tun_seg); + devm_kfree(dev, seg); + + return -EOPNOTSUPP; +} + +/** + * ice_fdir_write_fltr - send a flow director filter to the hardware + * @pf: PF data structure + * @input: filter structure + * @add: true adds filter and false removed filter + * @is_tun: true adds inner filter on tunnel and false outer headers + * + * returns 0 on success and negative value on error + */ +int +ice_fdir_write_fltr(struct ice_pf *pf, struct ice_fdir_fltr *input, bool add, + bool is_tun) +{ + struct device *dev = ice_pf_to_dev(pf); + struct ice_hw *hw = &pf->hw; + struct ice_fltr_desc desc; + struct ice_vsi *ctrl_vsi; + enum ice_status status; + u8 *pkt, *frag_pkt; + bool has_frag; + int err; + + ctrl_vsi = ice_get_ctrl_vsi(pf); + if (!ctrl_vsi) + return -EINVAL; + + pkt = devm_kzalloc(dev, ICE_FDIR_MAX_RAW_PKT_SIZE, GFP_KERNEL); + if (!pkt) + return -ENOMEM; + frag_pkt = devm_kzalloc(dev, ICE_FDIR_MAX_RAW_PKT_SIZE, GFP_KERNEL); + if (!frag_pkt) { + err = -ENOMEM; + goto err_free; + } + + ice_fdir_get_prgm_desc(hw, input, &desc, add); + status = ice_fdir_get_gen_prgm_pkt(hw, input, pkt, false, is_tun); + if (status) { + err = ice_status_to_errno(status); + goto err_free_all; + } + err = ice_prgm_fdir_fltr(ctrl_vsi, &desc, pkt); + if (err) + goto err_free_all; + + /* repeat for fragment packet */ + has_frag = ice_fdir_has_frag(input->flow_type); + if (has_frag) { + /* does not return error */ + ice_fdir_get_prgm_desc(hw, input, &desc, add); + status = ice_fdir_get_gen_prgm_pkt(hw, input, frag_pkt, true, + is_tun); + if (status) { + err = ice_status_to_errno(status); + goto err_frag; + } + err = ice_prgm_fdir_fltr(ctrl_vsi, &desc, frag_pkt); + if (err) + goto err_frag; + } else { + devm_kfree(dev, frag_pkt); + } + + return 0; + +err_free_all: + devm_kfree(dev, frag_pkt); +err_free: + devm_kfree(dev, pkt); + return err; + +err_frag: + devm_kfree(dev, frag_pkt); + return err; +} + +/** + * ice_fdir_write_all_fltr - send a flow director filter to the hardware + * @pf: PF data structure + * @input: filter structure + * @add: true adds filter and false removed filter + * + * returns 0 on success and negative value on error + */ +static int +ice_fdir_write_all_fltr(struct ice_pf *pf, struct ice_fdir_fltr *input, + bool add) +{ + u16 port_num; + int tun; + + for (tun = 0; tun < ICE_FD_HW_SEG_MAX; tun++) { + bool is_tun = tun == ICE_FD_HW_SEG_TUN; + int err; + + if (is_tun && !ice_get_open_tunnel_port(&pf->hw, TNL_ALL, + &port_num)) + continue; + err = ice_fdir_write_fltr(pf, input, add, is_tun); + if (err) + return err; + } + return 0; +} + +/** + * ice_fdir_replay_fltrs - replay filters from the HW filter list + * @pf: board private structure + */ +void ice_fdir_replay_fltrs(struct ice_pf *pf) +{ + struct ice_fdir_fltr *f_rule; + struct ice_hw *hw = &pf->hw; + + list_for_each_entry(f_rule, &hw->fdir_list_head, fltr_node) { + int err = ice_fdir_write_all_fltr(pf, f_rule, true); + + if (err) + dev_dbg(ice_pf_to_dev(pf), "Flow Director error %d, could not reprogram filter %d\n", + err, f_rule->fltr_id); + } +} + +/** + * ice_fdir_create_dflt_rules - create default perfect filters + * @pf: PF data structure + * + * Returns 0 for success or error. + */ +int ice_fdir_create_dflt_rules(struct ice_pf *pf) +{ + int err; + + /* Create perfect TCP and UDP rules in hardware. */ + err = ice_create_init_fdir_rule(pf, ICE_FLTR_PTYPE_NONF_IPV4_TCP); + if (err) + return err; + + err = ice_create_init_fdir_rule(pf, ICE_FLTR_PTYPE_NONF_IPV4_UDP); + if (err) + return err; + + err = ice_create_init_fdir_rule(pf, ICE_FLTR_PTYPE_NONF_IPV6_TCP); + if (err) + return err; + + err = ice_create_init_fdir_rule(pf, ICE_FLTR_PTYPE_NONF_IPV6_UDP); + + return err; +} + +/** + * ice_vsi_manage_fdir - turn on/off flow director + * @vsi: the VSI being changed + * @ena: boolean value indicating if this is an enable or disable request + */ +void ice_vsi_manage_fdir(struct ice_vsi *vsi, bool ena) +{ + struct ice_fdir_fltr *f_rule, *tmp; + struct ice_pf *pf = vsi->back; + struct ice_hw *hw = &pf->hw; + enum ice_fltr_ptype flow; + + if (ena) { + set_bit(ICE_FLAG_FD_ENA, pf->flags); + ice_fdir_create_dflt_rules(pf); + return; + } + + mutex_lock(&hw->fdir_fltr_lock); + if (!test_and_clear_bit(ICE_FLAG_FD_ENA, pf->flags)) + goto release_lock; + list_for_each_entry_safe(f_rule, tmp, &hw->fdir_list_head, fltr_node) { + /* ignore return value */ + ice_fdir_write_all_fltr(pf, f_rule, false); + ice_fdir_update_cntrs(hw, f_rule->flow_type, false); + list_del(&f_rule->fltr_node); + devm_kfree(ice_hw_to_dev(hw), f_rule); + } + + if (hw->fdir_prof) + for (flow = ICE_FLTR_PTYPE_NONF_NONE; flow < ICE_FLTR_PTYPE_MAX; + flow++) + if (hw->fdir_prof[flow]) + ice_fdir_rem_flow(hw, ICE_BLK_FD, flow); + +release_lock: + mutex_unlock(&hw->fdir_fltr_lock); +} + +/** + * ice_fdir_do_rem_flow - delete flow and possibly add perfect flow + * @pf: PF structure + * @flow_type: FDir flow type to release + */ +static void +ice_fdir_do_rem_flow(struct ice_pf *pf, enum ice_fltr_ptype flow_type) +{ + struct ice_hw *hw = &pf->hw; + bool need_perfect = false; + + if (flow_type == ICE_FLTR_PTYPE_NONF_IPV4_TCP || + flow_type == ICE_FLTR_PTYPE_NONF_IPV4_UDP || + flow_type == ICE_FLTR_PTYPE_NONF_IPV6_TCP || + flow_type == ICE_FLTR_PTYPE_NONF_IPV6_UDP) + need_perfect = true; + + if (need_perfect && test_bit(flow_type, hw->fdir_perfect_fltr)) + return; + + ice_fdir_rem_flow(hw, ICE_BLK_FD, flow_type); + if (need_perfect) + ice_create_init_fdir_rule(pf, flow_type); +} + +/** + * ice_fdir_update_list_entry - add or delete a filter from the filter list + * @pf: PF structure + * @input: filter structure + * @fltr_idx: ethtool index of filter to modify + * + * returns 0 on success and negative on errors + */ +static int +ice_fdir_update_list_entry(struct ice_pf *pf, struct ice_fdir_fltr *input, + int fltr_idx) +{ + struct ice_fdir_fltr *old_fltr; + struct ice_hw *hw = &pf->hw; + int err = -ENOENT; + + /* Do not update filters during reset */ + if (ice_is_reset_in_progress(pf->state)) + return -EBUSY; + + old_fltr = ice_fdir_find_fltr_by_idx(hw, fltr_idx); + if (old_fltr) { + err = ice_fdir_write_all_fltr(pf, old_fltr, false); + if (err) + return err; + ice_fdir_update_cntrs(hw, old_fltr->flow_type, false); + if (!input && !hw->fdir_fltr_cnt[old_fltr->flow_type]) + /* we just deleted the last filter of flow_type so we + * should also delete the HW filter info. + */ + ice_fdir_do_rem_flow(pf, old_fltr->flow_type); + list_del(&old_fltr->fltr_node); + devm_kfree(ice_hw_to_dev(hw), old_fltr); + } + if (!input) + return err; + ice_fdir_list_add_fltr(hw, input); + ice_fdir_update_cntrs(hw, input->flow_type, true); + return 0; +} + +/** + * ice_del_fdir_ethtool - delete Flow Director filter + * @vsi: pointer to target VSI + * @cmd: command to add or delete Flow Director filter + * + * Returns 0 on success and negative values for failure + */ +int ice_del_fdir_ethtool(struct ice_vsi *vsi, struct ethtool_rxnfc *cmd) +{ + struct ethtool_rx_flow_spec *fsp = + (struct ethtool_rx_flow_spec *)&cmd->fs; + struct ice_pf *pf = vsi->back; + struct ice_hw *hw = &pf->hw; + int val; + + if (!test_bit(ICE_FLAG_FD_ENA, pf->flags)) + return -EOPNOTSUPP; + + /* Do not delete filters during reset */ + if (ice_is_reset_in_progress(pf->state)) { + dev_err(ice_pf_to_dev(pf), "Device is resetting - deleting Flow Director filters not supported during reset\n"); + return -EBUSY; + } + + if (test_bit(__ICE_FD_FLUSH_REQ, pf->state)) + return -EBUSY; + + mutex_lock(&hw->fdir_fltr_lock); + val = ice_fdir_update_list_entry(pf, NULL, fsp->location); + mutex_unlock(&hw->fdir_fltr_lock); + + return val; +} + +/** + * ice_set_fdir_input_set - Set the input set for Flow Director + * @vsi: pointer to target VSI + * @fsp: pointer to ethtool Rx flow specification + * @input: filter structure + */ +static int +ice_set_fdir_input_set(struct ice_vsi *vsi, struct ethtool_rx_flow_spec *fsp, + struct ice_fdir_fltr *input) +{ + u16 dest_vsi, q_index = 0; + struct ice_pf *pf; + struct ice_hw *hw; + int flow_type; + u8 dest_ctl; + + if (!vsi || !fsp || !input) + return -EINVAL; + + pf = vsi->back; + hw = &pf->hw; + + dest_vsi = vsi->idx; + if (fsp->ring_cookie == RX_CLS_FLOW_DISC) { + dest_ctl = ICE_FLTR_PRGM_DESC_DEST_DROP_PKT; + } else { + u32 ring = ethtool_get_flow_spec_ring(fsp->ring_cookie); + u8 vf = ethtool_get_flow_spec_ring_vf(fsp->ring_cookie); + + if (vf) { + dev_err(ice_pf_to_dev(pf), "Failed to add filter. Flow director filters are not supported on VF queues.\n"); + return -EINVAL; + } + + if (ring >= vsi->num_rxq) + return -EINVAL; + + dest_ctl = ICE_FLTR_PRGM_DESC_DEST_DIRECT_PKT_QINDEX; + q_index = ring; + } + + input->fltr_id = fsp->location; + input->q_index = q_index; + flow_type = fsp->flow_type & ~FLOW_EXT; + + input->dest_vsi = dest_vsi; + input->dest_ctl = dest_ctl; + input->fltr_status = ICE_FLTR_PRGM_DESC_FD_STATUS_FD_ID; + input->cnt_index = ICE_FD_SB_STAT_IDX(hw->fd_ctr_base); + input->flow_type = ice_ethtool_flow_to_fltr(flow_type); + + if (fsp->flow_type & FLOW_EXT) { + memcpy(input->ext_data.usr_def, fsp->h_ext.data, + sizeof(input->ext_data.usr_def)); + input->ext_data.vlan_type = fsp->h_ext.vlan_etype; + input->ext_data.vlan_tag = fsp->h_ext.vlan_tci; + memcpy(input->ext_mask.usr_def, fsp->m_ext.data, + sizeof(input->ext_mask.usr_def)); + input->ext_mask.vlan_type = fsp->m_ext.vlan_etype; + input->ext_mask.vlan_tag = fsp->m_ext.vlan_tci; + } + + switch (flow_type) { + case TCP_V4_FLOW: + case UDP_V4_FLOW: + case SCTP_V4_FLOW: + input->ip.v4.dst_port = fsp->h_u.tcp_ip4_spec.pdst; + input->ip.v4.src_port = fsp->h_u.tcp_ip4_spec.psrc; + input->ip.v4.dst_ip = fsp->h_u.tcp_ip4_spec.ip4dst; + input->ip.v4.src_ip = fsp->h_u.tcp_ip4_spec.ip4src; + input->mask.v4.dst_port = fsp->m_u.tcp_ip4_spec.pdst; + input->mask.v4.src_port = fsp->m_u.tcp_ip4_spec.psrc; + input->mask.v4.dst_ip = fsp->m_u.tcp_ip4_spec.ip4dst; + input->mask.v4.src_ip = fsp->m_u.tcp_ip4_spec.ip4src; + break; + case IPV4_USER_FLOW: + input->ip.v4.dst_ip = fsp->h_u.usr_ip4_spec.ip4dst; + input->ip.v4.src_ip = fsp->h_u.usr_ip4_spec.ip4src; + input->ip.v4.l4_header = fsp->h_u.usr_ip4_spec.l4_4_bytes; + input->ip.v4.proto = fsp->h_u.usr_ip4_spec.proto; + input->ip.v4.ip_ver = fsp->h_u.usr_ip4_spec.ip_ver; + input->ip.v4.tos = fsp->h_u.usr_ip4_spec.tos; + input->mask.v4.dst_ip = fsp->m_u.usr_ip4_spec.ip4dst; + input->mask.v4.src_ip = fsp->m_u.usr_ip4_spec.ip4src; + input->mask.v4.l4_header = fsp->m_u.usr_ip4_spec.l4_4_bytes; + input->mask.v4.proto = fsp->m_u.usr_ip4_spec.proto; + input->mask.v4.ip_ver = fsp->m_u.usr_ip4_spec.ip_ver; + input->mask.v4.tos = fsp->m_u.usr_ip4_spec.tos; + break; + case TCP_V6_FLOW: + case UDP_V6_FLOW: + case SCTP_V6_FLOW: + memcpy(input->ip.v6.dst_ip, fsp->h_u.usr_ip6_spec.ip6dst, + sizeof(struct in6_addr)); + memcpy(input->ip.v6.src_ip, fsp->h_u.usr_ip6_spec.ip6src, + sizeof(struct in6_addr)); + input->ip.v6.dst_port = fsp->h_u.tcp_ip6_spec.pdst; + input->ip.v6.src_port = fsp->h_u.tcp_ip6_spec.psrc; + input->ip.v6.tc = fsp->h_u.tcp_ip6_spec.tclass; + memcpy(input->mask.v6.dst_ip, fsp->m_u.tcp_ip6_spec.ip6dst, + sizeof(struct in6_addr)); + memcpy(input->mask.v6.src_ip, fsp->m_u.tcp_ip6_spec.ip6src, + sizeof(struct in6_addr)); + input->mask.v6.dst_port = fsp->m_u.tcp_ip6_spec.pdst; + input->mask.v6.src_port = fsp->m_u.tcp_ip6_spec.psrc; + input->mask.v6.tc = fsp->m_u.tcp_ip6_spec.tclass; + break; + case IPV6_USER_FLOW: + memcpy(input->ip.v6.dst_ip, fsp->h_u.usr_ip6_spec.ip6dst, + sizeof(struct in6_addr)); + memcpy(input->ip.v6.src_ip, fsp->h_u.usr_ip6_spec.ip6src, + sizeof(struct in6_addr)); + input->ip.v6.l4_header = fsp->h_u.usr_ip6_spec.l4_4_bytes; + input->ip.v6.tc = fsp->h_u.usr_ip6_spec.tclass; + input->ip.v6.proto = fsp->h_u.usr_ip6_spec.l4_proto; + memcpy(input->mask.v6.dst_ip, fsp->m_u.usr_ip6_spec.ip6dst, + sizeof(struct in6_addr)); + memcpy(input->mask.v6.src_ip, fsp->m_u.usr_ip6_spec.ip6src, + sizeof(struct in6_addr)); + input->mask.v6.l4_header = fsp->m_u.usr_ip6_spec.l4_4_bytes; + input->mask.v6.tc = fsp->m_u.usr_ip6_spec.tclass; + input->mask.v6.proto = fsp->m_u.usr_ip6_spec.l4_proto; + break; + default: + /* not doing un-parsed flow types */ + return -EINVAL; + } + + return 0; +} + +/** + * ice_add_fdir_ethtool - Add/Remove Flow Director filter + * @vsi: pointer to target VSI + * @cmd: command to add or delete Flow Director filter + * + * Returns 0 on success and negative values for failure + */ +int ice_add_fdir_ethtool(struct ice_vsi *vsi, struct ethtool_rxnfc *cmd) +{ + struct ice_rx_flow_userdef userdata; + struct ethtool_rx_flow_spec *fsp; + struct ice_fdir_fltr *input; + struct device *dev; + struct ice_pf *pf; + struct ice_hw *hw; + int fltrs_needed; + u16 tunnel_port; + int ret; + + if (!vsi) + return -EINVAL; + + pf = vsi->back; + hw = &pf->hw; + dev = ice_pf_to_dev(pf); + + if (!test_bit(ICE_FLAG_FD_ENA, pf->flags)) + return -EOPNOTSUPP; + + /* Do not program filters during reset */ + if (ice_is_reset_in_progress(pf->state)) { + dev_err(dev, "Device is resetting - adding Flow Director filters not supported during reset\n"); + return -EBUSY; + } + + fsp = (struct ethtool_rx_flow_spec *)&cmd->fs; + + if (ice_parse_rx_flow_user_data(fsp, &userdata)) + return -EINVAL; + + if (fsp->flow_type & FLOW_MAC_EXT) + return -EINVAL; + + ret = ice_cfg_fdir_xtrct_seq(pf, fsp, &userdata); + if (ret) + return ret; + + if (fsp->location >= ice_get_fdir_cnt_all(hw)) { + dev_err(dev, "Failed to add filter. The maximum number of flow director filters has been reached.\n"); + return -ENOSPC; + } + + /* return error if not an update and no available filters */ + fltrs_needed = ice_get_open_tunnel_port(hw, TNL_ALL, &tunnel_port) ? + 2 : 1; + if (!ice_fdir_find_fltr_by_idx(hw, fsp->location) && + ice_fdir_num_avail_fltr(hw, pf->vsi[vsi->idx]) < fltrs_needed) { + dev_err(dev, "Failed to add filter. The maximum number of flow director filters has been reached.\n"); + return -ENOSPC; + } + + input = devm_kzalloc(dev, sizeof(*input), GFP_KERNEL); + if (!input) + return -ENOMEM; + + ret = ice_set_fdir_input_set(vsi, fsp, input); + if (ret) + goto free_input; + + mutex_lock(&hw->fdir_fltr_lock); + if (ice_fdir_is_dup_fltr(hw, input)) { + ret = -EINVAL; + goto release_lock; + } + + if (userdata.flex_fltr) { + input->flex_fltr = true; + input->flex_word = cpu_to_be16(userdata.flex_word); + input->flex_offset = userdata.flex_offset; + } + + /* input struct is added to the HW filter list */ + ice_fdir_update_list_entry(pf, input, fsp->location); + + ret = ice_fdir_write_all_fltr(pf, input, true); + if (ret) + goto remove_sw_rule; + + goto release_lock; + +remove_sw_rule: + ice_fdir_update_cntrs(hw, input->flow_type, false); + list_del(&input->fltr_node); +release_lock: + mutex_unlock(&hw->fdir_fltr_lock); +free_input: + if (ret) + devm_kfree(dev, input); + + return ret; +} diff --git a/drivers/net/ethernet/intel/ice/ice_fdir.c b/drivers/net/ethernet/intel/ice/ice_fdir.c new file mode 100644 index 000000000000..6834df14332f --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_fdir.c @@ -0,0 +1,840 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (C) 2018-2020, Intel Corporation. */ + +#include "ice_common.h" + +/* These are training packet headers used to program flow director filters. */ +static const u8 ice_fdir_tcpv4_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, + 0x00, 0x28, 0x00, 0x01, 0x00, 0x00, 0x40, 0x06, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x50, 0x00, + 0x20, 0x00, 0x00, 0x00, 0x00, 0x00 +}; + +static const u8 ice_fdir_udpv4_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, + 0x00, 0x1C, 0x00, 0x00, 0x40, 0x00, 0x40, 0x11, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, +}; + +static const u8 ice_fdir_sctpv4_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, + 0x00, 0x20, 0x00, 0x00, 0x40, 0x00, 0x40, 0x84, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +}; + +static const u8 ice_fdir_ipv4_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, + 0x00, 0x14, 0x00, 0x00, 0x40, 0x00, 0x40, 0x10, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00 +}; + +static const u8 ice_fdir_tcpv6_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x86, 0xDD, 0x60, 0x00, + 0x00, 0x00, 0x00, 0x14, 0x06, 0x40, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x50, 0x00, 0x20, 0x00, 0x00, 0x00, + 0x00, 0x00, +}; + +static const u8 ice_fdir_udpv6_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x86, 0xDD, 0x60, 0x00, + 0x00, 0x00, 0x00, 0x08, 0x11, 0x40, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x08, 0x00, 0x00, +}; + +static const u8 ice_fdir_sctpv6_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x86, 0xDD, 0x60, 0x00, + 0x00, 0x00, 0x00, 0x0C, 0x84, 0x40, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, +}; + +static const u8 ice_fdir_ipv6_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x86, 0xDD, 0x60, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x3B, 0x40, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +}; + +static const u8 ice_fdir_tcp4_tun_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, + 0x00, 0x5a, 0x00, 0x00, 0x40, 0x00, 0x40, 0x11, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x04, 0x00, 0x00, 0x03, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, + 0x45, 0x00, 0x00, 0x28, 0x00, 0x00, 0x40, 0x00, + 0x40, 0x06, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x50, 0x00, 0x20, 0x00, 0x00, 0x00, 0x00, 0x00, +}; + +static const u8 ice_fdir_udp4_tun_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, + 0x00, 0x4e, 0x00, 0x00, 0x40, 0x00, 0x40, 0x11, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x04, 0x00, 0x00, 0x03, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, + 0x45, 0x00, 0x00, 0x1c, 0x00, 0x00, 0x40, 0x00, + 0x40, 0x11, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, +}; + +static const u8 ice_fdir_sctp4_tun_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, + 0x00, 0x52, 0x00, 0x00, 0x40, 0x00, 0x40, 0x11, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x04, 0x00, 0x00, 0x03, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, + 0x45, 0x00, 0x00, 0x20, 0x00, 0x01, 0x00, 0x00, + 0x40, 0x84, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +}; + +static const u8 ice_fdir_ip4_tun_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, + 0x00, 0x46, 0x00, 0x00, 0x40, 0x00, 0x40, 0x11, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x04, 0x00, 0x00, 0x03, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, + 0x45, 0x00, 0x00, 0x14, 0x00, 0x00, 0x00, 0x00, + 0x40, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, +}; + +static const u8 ice_fdir_tcp6_tun_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, + 0x00, 0x6e, 0x00, 0x00, 0x40, 0x00, 0x40, 0x11, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x04, 0x00, 0x00, 0x03, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x86, 0xdd, + 0x60, 0x00, 0x00, 0x00, 0x00, 0x14, 0x06, 0x40, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x50, 0x00, 0x20, 0x00, + 0x00, 0x00, 0x00, 0x00, +}; + +static const u8 ice_fdir_udp6_tun_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, + 0x00, 0x62, 0x00, 0x00, 0x40, 0x00, 0x40, 0x11, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x04, 0x00, 0x00, 0x03, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x86, 0xdd, + 0x60, 0x00, 0x00, 0x00, 0x00, 0x08, 0x11, 0x40, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +}; + +static const u8 ice_fdir_sctp6_tun_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, + 0x00, 0x66, 0x00, 0x00, 0x40, 0x00, 0x40, 0x11, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x04, 0x00, 0x00, 0x03, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x86, 0xdd, + 0x60, 0x00, 0x00, 0x00, 0x00, 0x0c, 0x84, 0x40, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, +}; + +static const u8 ice_fdir_ip6_tun_pkt[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x08, 0x00, 0x45, 0x00, + 0x00, 0x5a, 0x00, 0x00, 0x40, 0x00, 0x40, 0x11, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x04, 0x00, 0x00, 0x03, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x86, 0xdd, + 0x60, 0x00, 0x00, 0x00, 0x00, 0x00, 0x3b, 0x40, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, +}; + +/* Flow Director no-op training packet table */ +static const struct ice_fdir_base_pkt ice_fdir_pkt[] = { + { + ICE_FLTR_PTYPE_NONF_IPV4_TCP, + sizeof(ice_fdir_tcpv4_pkt), ice_fdir_tcpv4_pkt, + sizeof(ice_fdir_tcp4_tun_pkt), ice_fdir_tcp4_tun_pkt, + }, + { + ICE_FLTR_PTYPE_NONF_IPV4_UDP, + sizeof(ice_fdir_udpv4_pkt), ice_fdir_udpv4_pkt, + sizeof(ice_fdir_udp4_tun_pkt), ice_fdir_udp4_tun_pkt, + }, + { + ICE_FLTR_PTYPE_NONF_IPV4_SCTP, + sizeof(ice_fdir_sctpv4_pkt), ice_fdir_sctpv4_pkt, + sizeof(ice_fdir_sctp4_tun_pkt), ice_fdir_sctp4_tun_pkt, + }, + { + ICE_FLTR_PTYPE_NONF_IPV4_OTHER, + sizeof(ice_fdir_ipv4_pkt), ice_fdir_ipv4_pkt, + sizeof(ice_fdir_ip4_tun_pkt), ice_fdir_ip4_tun_pkt, + }, + { + ICE_FLTR_PTYPE_NONF_IPV6_TCP, + sizeof(ice_fdir_tcpv6_pkt), ice_fdir_tcpv6_pkt, + sizeof(ice_fdir_tcp6_tun_pkt), ice_fdir_tcp6_tun_pkt, + }, + { + ICE_FLTR_PTYPE_NONF_IPV6_UDP, + sizeof(ice_fdir_udpv6_pkt), ice_fdir_udpv6_pkt, + sizeof(ice_fdir_udp6_tun_pkt), ice_fdir_udp6_tun_pkt, + }, + { + ICE_FLTR_PTYPE_NONF_IPV6_SCTP, + sizeof(ice_fdir_sctpv6_pkt), ice_fdir_sctpv6_pkt, + sizeof(ice_fdir_sctp6_tun_pkt), ice_fdir_sctp6_tun_pkt, + }, + { + ICE_FLTR_PTYPE_NONF_IPV6_OTHER, + sizeof(ice_fdir_ipv6_pkt), ice_fdir_ipv6_pkt, + sizeof(ice_fdir_ip6_tun_pkt), ice_fdir_ip6_tun_pkt, + }, +}; + +#define ICE_FDIR_NUM_PKT ARRAY_SIZE(ice_fdir_pkt) + +/** + * ice_set_dflt_val_fd_desc + * @fd_fltr_ctx: pointer to fd filter descriptor + */ +static void ice_set_dflt_val_fd_desc(struct ice_fd_fltr_desc_ctx *fd_fltr_ctx) +{ + fd_fltr_ctx->comp_q = ICE_FXD_FLTR_QW0_COMP_Q_ZERO; + fd_fltr_ctx->comp_report = ICE_FXD_FLTR_QW0_COMP_REPORT_SW_FAIL; + fd_fltr_ctx->fd_space = ICE_FXD_FLTR_QW0_FD_SPACE_GUAR_BEST; + fd_fltr_ctx->cnt_ena = ICE_FXD_FLTR_QW0_STAT_ENA_PKTS; + fd_fltr_ctx->evict_ena = ICE_FXD_FLTR_QW0_EVICT_ENA_TRUE; + fd_fltr_ctx->toq = ICE_FXD_FLTR_QW0_TO_Q_EQUALS_QINDEX; + fd_fltr_ctx->toq_prio = ICE_FXD_FLTR_QW0_TO_Q_PRIO1; + fd_fltr_ctx->dpu_recipe = ICE_FXD_FLTR_QW0_DPU_RECIPE_DFLT; + fd_fltr_ctx->drop = ICE_FXD_FLTR_QW0_DROP_NO; + fd_fltr_ctx->flex_prio = ICE_FXD_FLTR_QW0_FLEX_PRI_NONE; + fd_fltr_ctx->flex_mdid = ICE_FXD_FLTR_QW0_FLEX_MDID0; + fd_fltr_ctx->flex_val = ICE_FXD_FLTR_QW0_FLEX_VAL0; + fd_fltr_ctx->dtype = ICE_TX_DESC_DTYPE_FLTR_PROG; + fd_fltr_ctx->desc_prof_prio = ICE_FXD_FLTR_QW1_PROF_PRIO_ZERO; + fd_fltr_ctx->desc_prof = ICE_FXD_FLTR_QW1_PROF_ZERO; + fd_fltr_ctx->swap = ICE_FXD_FLTR_QW1_SWAP_SET; + fd_fltr_ctx->fdid_prio = ICE_FXD_FLTR_QW1_FDID_PRI_ONE; + fd_fltr_ctx->fdid_mdid = ICE_FXD_FLTR_QW1_FDID_MDID_FD; + fd_fltr_ctx->fdid = ICE_FXD_FLTR_QW1_FDID_ZERO; +} + +/** + * ice_set_fd_desc_val + * @ctx: pointer to fd filter descriptor context + * @fdir_desc: populated with fd filter descriptor values + */ +static void +ice_set_fd_desc_val(struct ice_fd_fltr_desc_ctx *ctx, + struct ice_fltr_desc *fdir_desc) +{ + u64 qword; + + /* prep QW0 of FD filter programming desc */ + qword = ((u64)ctx->qindex << ICE_FXD_FLTR_QW0_QINDEX_S) & + ICE_FXD_FLTR_QW0_QINDEX_M; + qword |= ((u64)ctx->comp_q << ICE_FXD_FLTR_QW0_COMP_Q_S) & + ICE_FXD_FLTR_QW0_COMP_Q_M; + qword |= ((u64)ctx->comp_report << ICE_FXD_FLTR_QW0_COMP_REPORT_S) & + ICE_FXD_FLTR_QW0_COMP_REPORT_M; + qword |= ((u64)ctx->fd_space << ICE_FXD_FLTR_QW0_FD_SPACE_S) & + ICE_FXD_FLTR_QW0_FD_SPACE_M; + qword |= ((u64)ctx->cnt_index << ICE_FXD_FLTR_QW0_STAT_CNT_S) & + ICE_FXD_FLTR_QW0_STAT_CNT_M; + qword |= ((u64)ctx->cnt_ena << ICE_FXD_FLTR_QW0_STAT_ENA_S) & + ICE_FXD_FLTR_QW0_STAT_ENA_M; + qword |= ((u64)ctx->evict_ena << ICE_FXD_FLTR_QW0_EVICT_ENA_S) & + ICE_FXD_FLTR_QW0_EVICT_ENA_M; + qword |= ((u64)ctx->toq << ICE_FXD_FLTR_QW0_TO_Q_S) & + ICE_FXD_FLTR_QW0_TO_Q_M; + qword |= ((u64)ctx->toq_prio << ICE_FXD_FLTR_QW0_TO_Q_PRI_S) & + ICE_FXD_FLTR_QW0_TO_Q_PRI_M; + qword |= ((u64)ctx->dpu_recipe << ICE_FXD_FLTR_QW0_DPU_RECIPE_S) & + ICE_FXD_FLTR_QW0_DPU_RECIPE_M; + qword |= ((u64)ctx->drop << ICE_FXD_FLTR_QW0_DROP_S) & + ICE_FXD_FLTR_QW0_DROP_M; + qword |= ((u64)ctx->flex_prio << ICE_FXD_FLTR_QW0_FLEX_PRI_S) & + ICE_FXD_FLTR_QW0_FLEX_PRI_M; + qword |= ((u64)ctx->flex_mdid << ICE_FXD_FLTR_QW0_FLEX_MDID_S) & + ICE_FXD_FLTR_QW0_FLEX_MDID_M; + qword |= ((u64)ctx->flex_val << ICE_FXD_FLTR_QW0_FLEX_VAL_S) & + ICE_FXD_FLTR_QW0_FLEX_VAL_M; + fdir_desc->qidx_compq_space_stat = cpu_to_le64(qword); + + /* prep QW1 of FD filter programming desc */ + qword = ((u64)ctx->dtype << ICE_FXD_FLTR_QW1_DTYPE_S) & + ICE_FXD_FLTR_QW1_DTYPE_M; + qword |= ((u64)ctx->pcmd << ICE_FXD_FLTR_QW1_PCMD_S) & + ICE_FXD_FLTR_QW1_PCMD_M; + qword |= ((u64)ctx->desc_prof_prio << ICE_FXD_FLTR_QW1_PROF_PRI_S) & + ICE_FXD_FLTR_QW1_PROF_PRI_M; + qword |= ((u64)ctx->desc_prof << ICE_FXD_FLTR_QW1_PROF_S) & + ICE_FXD_FLTR_QW1_PROF_M; + qword |= ((u64)ctx->fd_vsi << ICE_FXD_FLTR_QW1_FD_VSI_S) & + ICE_FXD_FLTR_QW1_FD_VSI_M; + qword |= ((u64)ctx->swap << ICE_FXD_FLTR_QW1_SWAP_S) & + ICE_FXD_FLTR_QW1_SWAP_M; + qword |= ((u64)ctx->fdid_prio << ICE_FXD_FLTR_QW1_FDID_PRI_S) & + ICE_FXD_FLTR_QW1_FDID_PRI_M; + qword |= ((u64)ctx->fdid_mdid << ICE_FXD_FLTR_QW1_FDID_MDID_S) & + ICE_FXD_FLTR_QW1_FDID_MDID_M; + qword |= ((u64)ctx->fdid << ICE_FXD_FLTR_QW1_FDID_S) & + ICE_FXD_FLTR_QW1_FDID_M; + fdir_desc->dtype_cmd_vsi_fdid = cpu_to_le64(qword); +} + +/** + * ice_fdir_get_prgm_desc - set a fdir descriptor from a fdir filter struct + * @hw: pointer to the hardware structure + * @input: filter + * @fdesc: filter descriptor + * @add: if add is true, this is an add operation, false implies delete + */ +void +ice_fdir_get_prgm_desc(struct ice_hw *hw, struct ice_fdir_fltr *input, + struct ice_fltr_desc *fdesc, bool add) +{ + struct ice_fd_fltr_desc_ctx fdir_fltr_ctx = { 0 }; + + /* set default context info */ + ice_set_dflt_val_fd_desc(&fdir_fltr_ctx); + + /* change sideband filtering values */ + fdir_fltr_ctx.fdid = input->fltr_id; + if (input->dest_ctl == ICE_FLTR_PRGM_DESC_DEST_DROP_PKT) { + fdir_fltr_ctx.drop = ICE_FXD_FLTR_QW0_DROP_YES; + fdir_fltr_ctx.qindex = 0; + } else { + fdir_fltr_ctx.drop = ICE_FXD_FLTR_QW0_DROP_NO; + fdir_fltr_ctx.qindex = input->q_index; + } + fdir_fltr_ctx.cnt_ena = ICE_FXD_FLTR_QW0_STAT_ENA_PKTS; + fdir_fltr_ctx.cnt_index = input->cnt_index; + fdir_fltr_ctx.fd_vsi = ice_get_hw_vsi_num(hw, input->dest_vsi); + fdir_fltr_ctx.evict_ena = ICE_FXD_FLTR_QW0_EVICT_ENA_FALSE; + fdir_fltr_ctx.toq_prio = 3; + fdir_fltr_ctx.pcmd = add ? ICE_FXD_FLTR_QW1_PCMD_ADD : + ICE_FXD_FLTR_QW1_PCMD_REMOVE; + fdir_fltr_ctx.swap = ICE_FXD_FLTR_QW1_SWAP_NOT_SET; + fdir_fltr_ctx.comp_q = ICE_FXD_FLTR_QW0_COMP_Q_ZERO; + fdir_fltr_ctx.comp_report = ICE_FXD_FLTR_QW0_COMP_REPORT_SW_FAIL; + fdir_fltr_ctx.fdid_prio = 3; + fdir_fltr_ctx.desc_prof = 1; + fdir_fltr_ctx.desc_prof_prio = 3; + ice_set_fd_desc_val(&fdir_fltr_ctx, fdesc); +} + +/** + * ice_alloc_fd_res_cntr - obtain counter resource for FD type + * @hw: pointer to the hardware structure + * @cntr_id: returns counter index + */ +enum ice_status ice_alloc_fd_res_cntr(struct ice_hw *hw, u16 *cntr_id) +{ + return ice_alloc_res_cntr(hw, ICE_AQC_RES_TYPE_FDIR_COUNTER_BLOCK, + ICE_AQC_RES_TYPE_FLAG_DEDICATED, 1, cntr_id); +} + +/** + * ice_free_fd_res_cntr - Free counter resource for FD type + * @hw: pointer to the hardware structure + * @cntr_id: counter index to be freed + */ +enum ice_status ice_free_fd_res_cntr(struct ice_hw *hw, u16 cntr_id) +{ + return ice_free_res_cntr(hw, ICE_AQC_RES_TYPE_FDIR_COUNTER_BLOCK, + ICE_AQC_RES_TYPE_FLAG_DEDICATED, 1, cntr_id); +} + +/** + * ice_alloc_fd_guar_item - allocate resource for FD guaranteed entries + * @hw: pointer to the hardware structure + * @cntr_id: returns counter index + * @num_fltr: number of filter entries to be allocated + */ +enum ice_status +ice_alloc_fd_guar_item(struct ice_hw *hw, u16 *cntr_id, u16 num_fltr) +{ + return ice_alloc_res_cntr(hw, ICE_AQC_RES_TYPE_FDIR_GUARANTEED_ENTRIES, + ICE_AQC_RES_TYPE_FLAG_DEDICATED, num_fltr, + cntr_id); +} + +/** + * ice_alloc_fd_shrd_item - allocate resource for flow director shared entries + * @hw: pointer to the hardware structure + * @cntr_id: returns counter index + * @num_fltr: number of filter entries to be allocated + */ +enum ice_status +ice_alloc_fd_shrd_item(struct ice_hw *hw, u16 *cntr_id, u16 num_fltr) +{ + return ice_alloc_res_cntr(hw, ICE_AQC_RES_TYPE_FDIR_SHARED_ENTRIES, + ICE_AQC_RES_TYPE_FLAG_DEDICATED, num_fltr, + cntr_id); +} + +/** + * ice_get_fdir_cnt_all - get the number of Flow Director filters + * @hw: hardware data structure + * + * Returns the number of filters available on device + */ +int ice_get_fdir_cnt_all(struct ice_hw *hw) +{ + return hw->func_caps.fd_fltr_guar + hw->func_caps.fd_fltr_best_effort; +} + +/** + * ice_pkt_insert_ipv6_addr - insert a be32 IPv6 address into a memory buffer + * @pkt: packet buffer + * @offset: offset into buffer + * @addr: IPv6 address to convert and insert into pkt at offset + */ +static void ice_pkt_insert_ipv6_addr(u8 *pkt, int offset, __be32 *addr) +{ + int idx; + + for (idx = 0; idx < ICE_IPV6_ADDR_LEN_AS_U32; idx++) + memcpy(pkt + offset + idx * sizeof(*addr), &addr[idx], + sizeof(*addr)); +} + +/** + * ice_pkt_insert_u16 - insert a be16 value into a memory buffer + * @pkt: packet buffer + * @offset: offset into buffer + * @data: 16 bit value to convert and insert into pkt at offset + */ +static void ice_pkt_insert_u16(u8 *pkt, int offset, __be16 data) +{ + memcpy(pkt + offset, &data, sizeof(data)); +} + +/** + * ice_pkt_insert_u32 - insert a be32 value into a memory buffer + * @pkt: packet buffer + * @offset: offset into buffer + * @data: 32 bit value to convert and insert into pkt at offset + */ +static void ice_pkt_insert_u32(u8 *pkt, int offset, __be32 data) +{ + memcpy(pkt + offset, &data, sizeof(data)); +} + +/** + * ice_fdir_get_gen_prgm_pkt - generate a training packet + * @hw: pointer to the hardware structure + * @input: flow director filter data structure + * @pkt: pointer to return filter packet + * @frag: generate a fragment packet + * @tun: true implies generate a tunnel packet + */ +enum ice_status +ice_fdir_get_gen_prgm_pkt(struct ice_hw *hw, struct ice_fdir_fltr *input, + u8 *pkt, bool frag, bool tun) +{ + enum ice_fltr_ptype flow; + u16 tnl_port; + u8 *loc; + u16 idx; + + if (input->flow_type == ICE_FLTR_PTYPE_NONF_IPV4_OTHER) { + switch (input->ip.v4.proto) { + case IPPROTO_TCP: + flow = ICE_FLTR_PTYPE_NONF_IPV4_TCP; + break; + case IPPROTO_UDP: + flow = ICE_FLTR_PTYPE_NONF_IPV4_UDP; + break; + case IPPROTO_SCTP: + flow = ICE_FLTR_PTYPE_NONF_IPV4_SCTP; + break; + case IPPROTO_IP: + flow = ICE_FLTR_PTYPE_NONF_IPV4_OTHER; + break; + default: + return ICE_ERR_PARAM; + } + } else if (input->flow_type == ICE_FLTR_PTYPE_NONF_IPV6_OTHER) { + switch (input->ip.v6.proto) { + case IPPROTO_TCP: + flow = ICE_FLTR_PTYPE_NONF_IPV6_TCP; + break; + case IPPROTO_UDP: + flow = ICE_FLTR_PTYPE_NONF_IPV6_UDP; + break; + case IPPROTO_SCTP: + flow = ICE_FLTR_PTYPE_NONF_IPV6_SCTP; + break; + case IPPROTO_IP: + flow = ICE_FLTR_PTYPE_NONF_IPV6_OTHER; + break; + default: + return ICE_ERR_PARAM; + } + } else { + flow = input->flow_type; + } + + for (idx = 0; idx < ICE_FDIR_NUM_PKT; idx++) + if (ice_fdir_pkt[idx].flow == flow) + break; + if (idx == ICE_FDIR_NUM_PKT) + return ICE_ERR_PARAM; + if (!tun) { + memcpy(pkt, ice_fdir_pkt[idx].pkt, ice_fdir_pkt[idx].pkt_len); + loc = pkt; + } else { + if (!ice_get_open_tunnel_port(hw, TNL_ALL, &tnl_port)) + return ICE_ERR_DOES_NOT_EXIST; + if (!ice_fdir_pkt[idx].tun_pkt) + return ICE_ERR_PARAM; + memcpy(pkt, ice_fdir_pkt[idx].tun_pkt, + ice_fdir_pkt[idx].tun_pkt_len); + ice_pkt_insert_u16(pkt, ICE_IPV4_UDP_DST_PORT_OFFSET, + htons(tnl_port)); + loc = &pkt[ICE_FDIR_TUN_PKT_OFF]; + } + + /* Reverse the src and dst, since the HW expects them to be from Tx + * perspective. The input from user is from Rx filter perspective. + */ + switch (flow) { + case ICE_FLTR_PTYPE_NONF_IPV4_TCP: + ice_pkt_insert_u32(loc, ICE_IPV4_DST_ADDR_OFFSET, + input->ip.v4.src_ip); + ice_pkt_insert_u16(loc, ICE_IPV4_TCP_DST_PORT_OFFSET, + input->ip.v4.src_port); + ice_pkt_insert_u32(loc, ICE_IPV4_SRC_ADDR_OFFSET, + input->ip.v4.dst_ip); + ice_pkt_insert_u16(loc, ICE_IPV4_TCP_SRC_PORT_OFFSET, + input->ip.v4.dst_port); + if (frag) + loc[20] = ICE_FDIR_IPV4_PKT_FLAG_DF; + break; + case ICE_FLTR_PTYPE_NONF_IPV4_UDP: + ice_pkt_insert_u32(loc, ICE_IPV4_DST_ADDR_OFFSET, + input->ip.v4.src_ip); + ice_pkt_insert_u16(loc, ICE_IPV4_UDP_DST_PORT_OFFSET, + input->ip.v4.src_port); + ice_pkt_insert_u32(loc, ICE_IPV4_SRC_ADDR_OFFSET, + input->ip.v4.dst_ip); + ice_pkt_insert_u16(loc, ICE_IPV4_UDP_SRC_PORT_OFFSET, + input->ip.v4.dst_port); + break; + case ICE_FLTR_PTYPE_NONF_IPV4_SCTP: + ice_pkt_insert_u32(loc, ICE_IPV4_DST_ADDR_OFFSET, + input->ip.v4.src_ip); + ice_pkt_insert_u16(loc, ICE_IPV4_SCTP_DST_PORT_OFFSET, + input->ip.v4.src_port); + ice_pkt_insert_u32(loc, ICE_IPV4_SRC_ADDR_OFFSET, + input->ip.v4.dst_ip); + ice_pkt_insert_u16(loc, ICE_IPV4_SCTP_SRC_PORT_OFFSET, + input->ip.v4.dst_port); + break; + case ICE_FLTR_PTYPE_NONF_IPV4_OTHER: + ice_pkt_insert_u32(loc, ICE_IPV4_DST_ADDR_OFFSET, + input->ip.v4.src_ip); + ice_pkt_insert_u32(loc, ICE_IPV4_SRC_ADDR_OFFSET, + input->ip.v4.dst_ip); + ice_pkt_insert_u16(loc, ICE_IPV4_PROTO_OFFSET, 0); + break; + case ICE_FLTR_PTYPE_NONF_IPV6_TCP: + ice_pkt_insert_ipv6_addr(loc, ICE_IPV6_DST_ADDR_OFFSET, + input->ip.v6.src_ip); + ice_pkt_insert_ipv6_addr(loc, ICE_IPV6_SRC_ADDR_OFFSET, + input->ip.v6.dst_ip); + ice_pkt_insert_u16(loc, ICE_IPV6_TCP_DST_PORT_OFFSET, + input->ip.v6.src_port); + ice_pkt_insert_u16(loc, ICE_IPV6_TCP_SRC_PORT_OFFSET, + input->ip.v6.dst_port); + break; + case ICE_FLTR_PTYPE_NONF_IPV6_UDP: + ice_pkt_insert_ipv6_addr(loc, ICE_IPV6_DST_ADDR_OFFSET, + input->ip.v6.src_ip); + ice_pkt_insert_ipv6_addr(loc, ICE_IPV6_SRC_ADDR_OFFSET, + input->ip.v6.dst_ip); + ice_pkt_insert_u16(loc, ICE_IPV6_UDP_DST_PORT_OFFSET, + input->ip.v6.src_port); + ice_pkt_insert_u16(loc, ICE_IPV6_UDP_SRC_PORT_OFFSET, + input->ip.v6.dst_port); + break; + case ICE_FLTR_PTYPE_NONF_IPV6_SCTP: + ice_pkt_insert_ipv6_addr(loc, ICE_IPV6_DST_ADDR_OFFSET, + input->ip.v6.src_ip); + ice_pkt_insert_ipv6_addr(loc, ICE_IPV6_SRC_ADDR_OFFSET, + input->ip.v6.dst_ip); + ice_pkt_insert_u16(loc, ICE_IPV6_SCTP_DST_PORT_OFFSET, + input->ip.v6.src_port); + ice_pkt_insert_u16(loc, ICE_IPV6_SCTP_SRC_PORT_OFFSET, + input->ip.v6.dst_port); + break; + case ICE_FLTR_PTYPE_NONF_IPV6_OTHER: + ice_pkt_insert_ipv6_addr(loc, ICE_IPV6_DST_ADDR_OFFSET, + input->ip.v6.src_ip); + ice_pkt_insert_ipv6_addr(loc, ICE_IPV6_SRC_ADDR_OFFSET, + input->ip.v6.dst_ip); + break; + default: + return ICE_ERR_PARAM; + } + + if (input->flex_fltr) + ice_pkt_insert_u16(loc, input->flex_offset, input->flex_word); + + return 0; +} + +/** + * ice_fdir_has_frag - does flow type have 2 ptypes + * @flow: flow ptype + * + * returns true is there is a fragment packet for this ptype + */ +bool ice_fdir_has_frag(enum ice_fltr_ptype flow) +{ + if (flow == ICE_FLTR_PTYPE_NONF_IPV4_OTHER) + return true; + else + return false; +} + +/** + * ice_fdir_find_by_idx - find filter with idx + * @hw: pointer to hardware structure + * @fltr_idx: index to find. + * + * Returns pointer to filter if found or null + */ +struct ice_fdir_fltr * +ice_fdir_find_fltr_by_idx(struct ice_hw *hw, u32 fltr_idx) +{ + struct ice_fdir_fltr *rule; + + list_for_each_entry(rule, &hw->fdir_list_head, fltr_node) { + /* rule ID found in the list */ + if (fltr_idx == rule->fltr_id) + return rule; + if (fltr_idx < rule->fltr_id) + break; + } + return NULL; +} + +/** + * ice_fdir_list_add_fltr - add a new node to the flow director filter list + * @hw: hardware structure + * @fltr: filter node to add to structure + */ +void ice_fdir_list_add_fltr(struct ice_hw *hw, struct ice_fdir_fltr *fltr) +{ + struct ice_fdir_fltr *rule, *parent = NULL; + + list_for_each_entry(rule, &hw->fdir_list_head, fltr_node) { + /* rule ID found or pass its spot in the list */ + if (rule->fltr_id >= fltr->fltr_id) + break; + parent = rule; + } + + if (parent) + list_add(&fltr->fltr_node, &parent->fltr_node); + else + list_add(&fltr->fltr_node, &hw->fdir_list_head); +} + +/** + * ice_fdir_update_cntrs - increment / decrement filter counter + * @hw: pointer to hardware structure + * @flow: filter flow type + * @add: true implies filters added + */ +void +ice_fdir_update_cntrs(struct ice_hw *hw, enum ice_fltr_ptype flow, bool add) +{ + int incr; + + incr = add ? 1 : -1; + hw->fdir_active_fltr += incr; + + if (flow == ICE_FLTR_PTYPE_NONF_NONE || flow >= ICE_FLTR_PTYPE_MAX) + ice_debug(hw, ICE_DBG_SW, "Unknown filter type %d\n", flow); + else + hw->fdir_fltr_cnt[flow] += incr; +} + +/** + * ice_cmp_ipv6_addr - compare 2 IP v6 addresses + * @a: IP v6 address + * @b: IP v6 address + * + * Returns 0 on equal, returns non-0 if different + */ +static int ice_cmp_ipv6_addr(__be32 *a, __be32 *b) +{ + return memcmp(a, b, 4 * sizeof(__be32)); +} + +/** + * ice_fdir_comp_rules - compare 2 filters + * @a: a Flow Director filter data structure + * @b: a Flow Director filter data structure + * @v6: bool true if v6 filter + * + * Returns true if the filters match + */ +static bool +ice_fdir_comp_rules(struct ice_fdir_fltr *a, struct ice_fdir_fltr *b, bool v6) +{ + enum ice_fltr_ptype flow_type = a->flow_type; + + /* The calling function already checks that the two filters have the + * same flow_type. + */ + if (!v6) { + if (flow_type == ICE_FLTR_PTYPE_NONF_IPV4_TCP || + flow_type == ICE_FLTR_PTYPE_NONF_IPV4_UDP || + flow_type == ICE_FLTR_PTYPE_NONF_IPV4_SCTP) { + if (a->ip.v4.dst_ip == b->ip.v4.dst_ip && + a->ip.v4.src_ip == b->ip.v4.src_ip && + a->ip.v4.dst_port == b->ip.v4.dst_port && + a->ip.v4.src_port == b->ip.v4.src_port) + return true; + } else if (flow_type == ICE_FLTR_PTYPE_NONF_IPV4_OTHER) { + if (a->ip.v4.dst_ip == b->ip.v4.dst_ip && + a->ip.v4.src_ip == b->ip.v4.src_ip && + a->ip.v4.l4_header == b->ip.v4.l4_header && + a->ip.v4.proto == b->ip.v4.proto && + a->ip.v4.ip_ver == b->ip.v4.ip_ver && + a->ip.v4.tos == b->ip.v4.tos) + return true; + } + } else { + if (flow_type == ICE_FLTR_PTYPE_NONF_IPV6_UDP || + flow_type == ICE_FLTR_PTYPE_NONF_IPV6_TCP || + flow_type == ICE_FLTR_PTYPE_NONF_IPV6_SCTP) { + if (a->ip.v6.dst_port == b->ip.v6.dst_port && + a->ip.v6.src_port == b->ip.v6.src_port && + !ice_cmp_ipv6_addr(a->ip.v6.dst_ip, + b->ip.v6.dst_ip) && + !ice_cmp_ipv6_addr(a->ip.v6.src_ip, + b->ip.v6.src_ip)) + return true; + } else if (flow_type == ICE_FLTR_PTYPE_NONF_IPV6_OTHER) { + if (a->ip.v6.dst_port == b->ip.v6.dst_port && + a->ip.v6.src_port == b->ip.v6.src_port) + return true; + } + } + + return false; +} + +/** + * ice_fdir_is_dup_fltr - test if filter is already in list for PF + * @hw: hardware data structure + * @input: Flow Director filter data structure + * + * Returns true if the filter is found in the list + */ +bool ice_fdir_is_dup_fltr(struct ice_hw *hw, struct ice_fdir_fltr *input) +{ + struct ice_fdir_fltr *rule; + bool ret = false; + + list_for_each_entry(rule, &hw->fdir_list_head, fltr_node) { + enum ice_fltr_ptype flow_type; + + if (rule->flow_type != input->flow_type) + continue; + + flow_type = input->flow_type; + if (flow_type == ICE_FLTR_PTYPE_NONF_IPV4_TCP || + flow_type == ICE_FLTR_PTYPE_NONF_IPV4_UDP || + flow_type == ICE_FLTR_PTYPE_NONF_IPV4_SCTP || + flow_type == ICE_FLTR_PTYPE_NONF_IPV4_OTHER) + ret = ice_fdir_comp_rules(rule, input, false); + else + ret = ice_fdir_comp_rules(rule, input, true); + if (ret) { + if (rule->fltr_id == input->fltr_id && + rule->q_index != input->q_index) + ret = false; + else + break; + } + } + + return ret; +} diff --git a/drivers/net/ethernet/intel/ice/ice_fdir.h b/drivers/net/ethernet/intel/ice/ice_fdir.h new file mode 100644 index 000000000000..1c587766daab --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_fdir.h @@ -0,0 +1,166 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (C) 2018-2020, Intel Corporation. */ + +#ifndef _ICE_FDIR_H_ +#define _ICE_FDIR_H_ + +#define ICE_FDIR_TUN_PKT_OFF 50 +#define ICE_FDIR_MAX_RAW_PKT_SIZE (512 + ICE_FDIR_TUN_PKT_OFF) + +/* macros for offsets into packets for flow director programming */ +#define ICE_IPV4_SRC_ADDR_OFFSET 26 +#define ICE_IPV4_DST_ADDR_OFFSET 30 +#define ICE_IPV4_TCP_SRC_PORT_OFFSET 34 +#define ICE_IPV4_TCP_DST_PORT_OFFSET 36 +#define ICE_IPV4_UDP_SRC_PORT_OFFSET 34 +#define ICE_IPV4_UDP_DST_PORT_OFFSET 36 +#define ICE_IPV4_SCTP_SRC_PORT_OFFSET 34 +#define ICE_IPV4_SCTP_DST_PORT_OFFSET 36 +#define ICE_IPV4_PROTO_OFFSET 23 +#define ICE_IPV6_SRC_ADDR_OFFSET 22 +#define ICE_IPV6_DST_ADDR_OFFSET 38 +#define ICE_IPV6_TCP_SRC_PORT_OFFSET 54 +#define ICE_IPV6_TCP_DST_PORT_OFFSET 56 +#define ICE_IPV6_UDP_SRC_PORT_OFFSET 54 +#define ICE_IPV6_UDP_DST_PORT_OFFSET 56 +#define ICE_IPV6_SCTP_SRC_PORT_OFFSET 54 +#define ICE_IPV6_SCTP_DST_PORT_OFFSET 56 +/* IP v4 has 2 flag bits that enable fragment processing: DF and MF. DF + * requests that the packet not be fragmented. MF indicates that a packet has + * been fragmented. + */ +#define ICE_FDIR_IPV4_PKT_FLAG_DF 0x20 + +enum ice_fltr_prgm_desc_dest { + ICE_FLTR_PRGM_DESC_DEST_DROP_PKT, + ICE_FLTR_PRGM_DESC_DEST_DIRECT_PKT_QINDEX, +}; + +enum ice_fltr_prgm_desc_fd_status { + ICE_FLTR_PRGM_DESC_FD_STATUS_NONE, + ICE_FLTR_PRGM_DESC_FD_STATUS_FD_ID, +}; + +/* Flow Director (FD) Filter Programming descriptor */ +struct ice_fd_fltr_desc_ctx { + u32 fdid; + u16 qindex; + u16 cnt_index; + u16 fd_vsi; + u16 flex_val; + u8 comp_q; + u8 comp_report; + u8 fd_space; + u8 cnt_ena; + u8 evict_ena; + u8 toq; + u8 toq_prio; + u8 dpu_recipe; + u8 drop; + u8 flex_prio; + u8 flex_mdid; + u8 dtype; + u8 pcmd; + u8 desc_prof_prio; + u8 desc_prof; + u8 swap; + u8 fdid_prio; + u8 fdid_mdid; +}; + +#define ICE_FLTR_PRGM_FLEX_WORD_SIZE sizeof(__be16) + +struct ice_rx_flow_userdef { + u16 flex_word; + u16 flex_offset; + u16 flex_fltr; +}; + +struct ice_fdir_v4 { + __be32 dst_ip; + __be32 src_ip; + __be16 dst_port; + __be16 src_port; + __be32 l4_header; + __be32 sec_parm_idx; /* security parameter index */ + u8 tos; + u8 ip_ver; + u8 proto; +}; + +#define ICE_IPV6_ADDR_LEN_AS_U32 4 + +struct ice_fdir_v6 { + __be32 dst_ip[ICE_IPV6_ADDR_LEN_AS_U32]; + __be32 src_ip[ICE_IPV6_ADDR_LEN_AS_U32]; + __be16 dst_port; + __be16 src_port; + __be32 l4_header; /* next header */ + __be32 sec_parm_idx; /* security parameter index */ + u8 tc; + u8 proto; +}; + +struct ice_fdir_extra { + u8 dst_mac[ETH_ALEN]; /* dest MAC address */ + u32 usr_def[2]; /* user data */ + __be16 vlan_type; /* VLAN ethertype */ + __be16 vlan_tag; /* VLAN tag info */ +}; + +struct ice_fdir_fltr { + struct list_head fltr_node; + enum ice_fltr_ptype flow_type; + + union { + struct ice_fdir_v4 v4; + struct ice_fdir_v6 v6; + } ip, mask; + + struct ice_fdir_extra ext_data; + struct ice_fdir_extra ext_mask; + + /* flex byte filter data */ + __be16 flex_word; + u16 flex_offset; + u16 flex_fltr; + + /* filter control */ + u16 q_index; + u16 dest_vsi; + u8 dest_ctl; + u8 fltr_status; + u16 cnt_index; + u32 fltr_id; +}; + +/* Dummy packet filter definition structure */ +struct ice_fdir_base_pkt { + enum ice_fltr_ptype flow; + u16 pkt_len; + const u8 *pkt; + u16 tun_pkt_len; + const u8 *tun_pkt; +}; + +enum ice_status ice_alloc_fd_res_cntr(struct ice_hw *hw, u16 *cntr_id); +enum ice_status ice_free_fd_res_cntr(struct ice_hw *hw, u16 cntr_id); +enum ice_status +ice_alloc_fd_guar_item(struct ice_hw *hw, u16 *cntr_id, u16 num_fltr); +enum ice_status +ice_alloc_fd_shrd_item(struct ice_hw *hw, u16 *cntr_id, u16 num_fltr); +void +ice_fdir_get_prgm_desc(struct ice_hw *hw, struct ice_fdir_fltr *input, + struct ice_fltr_desc *fdesc, bool add); +enum ice_status +ice_fdir_get_gen_prgm_pkt(struct ice_hw *hw, struct ice_fdir_fltr *input, + u8 *pkt, bool frag, bool tun); +int ice_get_fdir_cnt_all(struct ice_hw *hw); +bool ice_fdir_is_dup_fltr(struct ice_hw *hw, struct ice_fdir_fltr *input); +bool ice_fdir_has_frag(enum ice_fltr_ptype flow); +struct ice_fdir_fltr * +ice_fdir_find_fltr_by_idx(struct ice_hw *hw, u32 fltr_idx); +void +ice_fdir_update_cntrs(struct ice_hw *hw, enum ice_fltr_ptype flow, bool add); +void ice_fdir_list_add_fltr(struct ice_hw *hw, struct ice_fdir_fltr *input); +#endif /* _ICE_FDIR_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c index 42bac3ec5526..4420fc02f7e7 100644 --- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.c +++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.c @@ -5,6 +5,15 @@ #include "ice_flex_pipe.h" #include "ice_flow.h" +/* To support tunneling entries by PF, the package will append the PF number to + * the label; for example TNL_VXLAN_PF0, TNL_VXLAN_PF1, TNL_VXLAN_PF2, etc. + */ +static const struct ice_tunnel_type_scan tnls[] = { + { TNL_VXLAN, "TNL_VXLAN_PF" }, + { TNL_GENEVE, "TNL_GENEVE_PF" }, + { TNL_LAST, "" } +}; + static const u32 ice_sect_lkup[ICE_BLK_COUNT][ICE_SECT_COUNT] = { /* SWITCH */ { @@ -239,6 +248,268 @@ ice_pkg_enum_section(struct ice_seg *ice_seg, struct ice_pkg_enum *state, return state->sect; } +/** + * ice_pkg_enum_entry + * @ice_seg: pointer to the ice segment (or NULL on subsequent calls) + * @state: pointer to the enum state + * @sect_type: section type to enumerate + * @offset: pointer to variable that receives the offset in the table (optional) + * @handler: function that handles access to the entries into the section type + * + * This function will enumerate all the entries in particular section type in + * the ice segment. The first call is made with the ice_seg parameter non-NULL; + * on subsequent calls, ice_seg is set to NULL which continues the enumeration. + * When the function returns a NULL pointer, then the end of the entries has + * been reached. + * + * Since each section may have a different header and entry size, the handler + * function is needed to determine the number and location entries in each + * section. + * + * The offset parameter is optional, but should be used for sections that + * contain an offset for each section table. For such cases, the section handler + * function must return the appropriate offset + index to give the absolution + * offset for each entry. For example, if the base for a section's header + * indicates a base offset of 10, and the index for the entry is 2, then + * section handler function should set the offset to 10 + 2 = 12. + */ +static void * +ice_pkg_enum_entry(struct ice_seg *ice_seg, struct ice_pkg_enum *state, + u32 sect_type, u32 *offset, + void *(*handler)(u32 sect_type, void *section, + u32 index, u32 *offset)) +{ + void *entry; + + if (ice_seg) { + if (!handler) + return NULL; + + if (!ice_pkg_enum_section(ice_seg, state, sect_type)) + return NULL; + + state->entry_idx = 0; + state->handler = handler; + } else { + state->entry_idx++; + } + + if (!state->handler) + return NULL; + + /* get entry */ + entry = state->handler(state->sect_type, state->sect, state->entry_idx, + offset); + if (!entry) { + /* end of a section, look for another section of this type */ + if (!ice_pkg_enum_section(NULL, state, 0)) + return NULL; + + state->entry_idx = 0; + entry = state->handler(state->sect_type, state->sect, + state->entry_idx, offset); + } + + return entry; +} + +/** + * ice_boost_tcam_handler + * @sect_type: section type + * @section: pointer to section + * @index: index of the boost TCAM entry to be returned + * @offset: pointer to receive absolute offset, always 0 for boost TCAM sections + * + * This is a callback function that can be passed to ice_pkg_enum_entry. + * Handles enumeration of individual boost TCAM entries. + */ +static void * +ice_boost_tcam_handler(u32 sect_type, void *section, u32 index, u32 *offset) +{ + struct ice_boost_tcam_section *boost; + + if (!section) + return NULL; + + if (sect_type != ICE_SID_RXPARSER_BOOST_TCAM) + return NULL; + + if (index > ICE_MAX_BST_TCAMS_IN_BUF) + return NULL; + + if (offset) + *offset = 0; + + boost = section; + if (index >= le16_to_cpu(boost->count)) + return NULL; + + return boost->tcam + index; +} + +/** + * ice_find_boost_entry + * @ice_seg: pointer to the ice segment (non-NULL) + * @addr: Boost TCAM address of entry to search for + * @entry: returns pointer to the entry + * + * Finds a particular Boost TCAM entry and returns a pointer to that entry + * if it is found. The ice_seg parameter must not be NULL since the first call + * to ice_pkg_enum_entry requires a pointer to an actual ice_segment structure. + */ +static enum ice_status +ice_find_boost_entry(struct ice_seg *ice_seg, u16 addr, + struct ice_boost_tcam_entry **entry) +{ + struct ice_boost_tcam_entry *tcam; + struct ice_pkg_enum state; + + memset(&state, 0, sizeof(state)); + + if (!ice_seg) + return ICE_ERR_PARAM; + + do { + tcam = ice_pkg_enum_entry(ice_seg, &state, + ICE_SID_RXPARSER_BOOST_TCAM, NULL, + ice_boost_tcam_handler); + if (tcam && le16_to_cpu(tcam->addr) == addr) { + *entry = tcam; + return 0; + } + + ice_seg = NULL; + } while (tcam); + + *entry = NULL; + return ICE_ERR_CFG; +} + +/** + * ice_label_enum_handler + * @sect_type: section type + * @section: pointer to section + * @index: index of the label entry to be returned + * @offset: pointer to receive absolute offset, always zero for label sections + * + * This is a callback function that can be passed to ice_pkg_enum_entry. + * Handles enumeration of individual label entries. + */ +static void * +ice_label_enum_handler(u32 __always_unused sect_type, void *section, u32 index, + u32 *offset) +{ + struct ice_label_section *labels; + + if (!section) + return NULL; + + if (index > ICE_MAX_LABELS_IN_BUF) + return NULL; + + if (offset) + *offset = 0; + + labels = section; + if (index >= le16_to_cpu(labels->count)) + return NULL; + + return labels->label + index; +} + +/** + * ice_enum_labels + * @ice_seg: pointer to the ice segment (NULL on subsequent calls) + * @type: the section type that will contain the label (0 on subsequent calls) + * @state: ice_pkg_enum structure that will hold the state of the enumeration + * @value: pointer to a value that will return the label's value if found + * + * Enumerates a list of labels in the package. The caller will call + * ice_enum_labels(ice_seg, type, ...) to start the enumeration, then call + * ice_enum_labels(NULL, 0, ...) to continue. When the function returns a NULL + * the end of the list has been reached. + */ +static char * +ice_enum_labels(struct ice_seg *ice_seg, u32 type, struct ice_pkg_enum *state, + u16 *value) +{ + struct ice_label *label; + + /* Check for valid label section on first call */ + if (type && !(type >= ICE_SID_LBL_FIRST && type <= ICE_SID_LBL_LAST)) + return NULL; + + label = ice_pkg_enum_entry(ice_seg, state, type, NULL, + ice_label_enum_handler); + if (!label) + return NULL; + + *value = le16_to_cpu(label->value); + return label->name; +} + +/** + * ice_init_pkg_hints + * @hw: pointer to the HW structure + * @ice_seg: pointer to the segment of the package scan (non-NULL) + * + * This function will scan the package and save off relevant information + * (hints or metadata) for driver use. The ice_seg parameter must not be NULL + * since the first call to ice_enum_labels requires a pointer to an actual + * ice_seg structure. + */ +static void ice_init_pkg_hints(struct ice_hw *hw, struct ice_seg *ice_seg) +{ + struct ice_pkg_enum state; + char *label_name; + u16 val; + int i; + + memset(&hw->tnl, 0, sizeof(hw->tnl)); + memset(&state, 0, sizeof(state)); + + if (!ice_seg) + return; + + label_name = ice_enum_labels(ice_seg, ICE_SID_LBL_RXPARSER_TMEM, &state, + &val); + + while (label_name && hw->tnl.count < ICE_TUNNEL_MAX_ENTRIES) { + for (i = 0; tnls[i].type != TNL_LAST; i++) { + size_t len = strlen(tnls[i].label_prefix); + + /* Look for matching label start, before continuing */ + if (strncmp(label_name, tnls[i].label_prefix, len)) + continue; + + /* Make sure this label matches our PF. Note that the PF + * character ('0' - '7') will be located where our + * prefix string's null terminator is located. + */ + if ((label_name[len] - '0') == hw->pf_id) { + hw->tnl.tbl[hw->tnl.count].type = tnls[i].type; + hw->tnl.tbl[hw->tnl.count].valid = false; + hw->tnl.tbl[hw->tnl.count].in_use = false; + hw->tnl.tbl[hw->tnl.count].marked = false; + hw->tnl.tbl[hw->tnl.count].boost_addr = val; + hw->tnl.tbl[hw->tnl.count].port = 0; + hw->tnl.count++; + break; + } + } + + label_name = ice_enum_labels(NULL, 0, &state, &val); + } + + /* Cache the appropriate boost TCAM entry pointers */ + for (i = 0; i < hw->tnl.count; i++) { + ice_find_boost_entry(ice_seg, hw->tnl.tbl[i].boost_addr, + &hw->tnl.tbl[i].boost_entry); + if (hw->tnl.tbl[i].boost_entry) + hw->tnl.tbl[i].valid = true; + } +} + /* Key creation */ #define ICE_DC_KEY 0x1 /* don't care */ @@ -593,8 +864,9 @@ ice_find_seg_in_pkg(struct ice_hw *hw, u32 seg_type, u32 i; ice_debug(hw, ICE_DBG_PKG, "Package format version: %d.%d.%d.%d\n", - pkg_hdr->format_ver.major, pkg_hdr->format_ver.minor, - pkg_hdr->format_ver.update, pkg_hdr->format_ver.draft); + pkg_hdr->pkg_format_ver.major, pkg_hdr->pkg_format_ver.minor, + pkg_hdr->pkg_format_ver.update, + pkg_hdr->pkg_format_ver.draft); /* Search all package segments for the requested segment type */ for (i = 0; i < le32_to_cpu(pkg_hdr->seg_count); i++) { @@ -764,13 +1036,15 @@ ice_download_pkg(struct ice_hw *hw, struct ice_seg *ice_seg) { struct ice_buf_table *ice_buf_tbl; - ice_debug(hw, ICE_DBG_PKG, "Segment version: %d.%d.%d.%d\n", - ice_seg->hdr.seg_ver.major, ice_seg->hdr.seg_ver.minor, - ice_seg->hdr.seg_ver.update, ice_seg->hdr.seg_ver.draft); + ice_debug(hw, ICE_DBG_PKG, "Segment format version: %d.%d.%d.%d\n", + ice_seg->hdr.seg_format_ver.major, + ice_seg->hdr.seg_format_ver.minor, + ice_seg->hdr.seg_format_ver.update, + ice_seg->hdr.seg_format_ver.draft); ice_debug(hw, ICE_DBG_PKG, "Seg: type 0x%X, size %d, name %s\n", le32_to_cpu(ice_seg->hdr.seg_type), - le32_to_cpu(ice_seg->hdr.seg_size), ice_seg->hdr.seg_name); + le32_to_cpu(ice_seg->hdr.seg_size), ice_seg->hdr.seg_id); ice_buf_tbl = ice_find_buf_table(ice_seg); @@ -815,14 +1089,16 @@ ice_init_pkg_info(struct ice_hw *hw, struct ice_pkg_hdr *pkg_hdr) seg_hdr = ice_find_seg_in_pkg(hw, SEGMENT_TYPE_ICE, pkg_hdr); if (seg_hdr) { - hw->ice_pkg_ver = seg_hdr->seg_ver; - memcpy(hw->ice_pkg_name, seg_hdr->seg_name, + hw->ice_pkg_ver = seg_hdr->seg_format_ver; + memcpy(hw->ice_pkg_name, seg_hdr->seg_id, sizeof(hw->ice_pkg_name)); - ice_debug(hw, ICE_DBG_PKG, "Ice Pkg: %d.%d.%d.%d, %s\n", - seg_hdr->seg_ver.major, seg_hdr->seg_ver.minor, - seg_hdr->seg_ver.update, seg_hdr->seg_ver.draft, - seg_hdr->seg_name); + ice_debug(hw, ICE_DBG_PKG, "Ice Seg: %d.%d.%d.%d, %s\n", + seg_hdr->seg_format_ver.major, + seg_hdr->seg_format_ver.minor, + seg_hdr->seg_format_ver.update, + seg_hdr->seg_format_ver.draft, + seg_hdr->seg_id); } else { ice_debug(hw, ICE_DBG_INIT, "Did not find ice segment in driver package\n"); @@ -863,9 +1139,11 @@ static enum ice_status ice_get_pkg_info(struct ice_hw *hw) if (pkg_info->pkg_info[i].is_active) { flags[place++] = 'A'; hw->active_pkg_ver = pkg_info->pkg_info[i].ver; + hw->active_track_id = + le32_to_cpu(pkg_info->pkg_info[i].track_id); memcpy(hw->active_pkg_name, pkg_info->pkg_info[i].name, - sizeof(hw->active_pkg_name)); + sizeof(pkg_info->pkg_info[i].name)); hw->active_pkg_in_nvm = pkg_info->pkg_info[i].is_in_nvm; } if (pkg_info->pkg_info[i].is_active_at_boot) @@ -905,10 +1183,10 @@ static enum ice_status ice_verify_pkg(struct ice_pkg_hdr *pkg, u32 len) if (len < sizeof(*pkg)) return ICE_ERR_BUF_TOO_SHORT; - if (pkg->format_ver.major != ICE_PKG_FMT_VER_MAJ || - pkg->format_ver.minor != ICE_PKG_FMT_VER_MNR || - pkg->format_ver.update != ICE_PKG_FMT_VER_UPD || - pkg->format_ver.draft != ICE_PKG_FMT_VER_DFT) + if (pkg->pkg_format_ver.major != ICE_PKG_FMT_VER_MAJ || + pkg->pkg_format_ver.minor != ICE_PKG_FMT_VER_MNR || + pkg->pkg_format_ver.update != ICE_PKG_FMT_VER_UPD || + pkg->pkg_format_ver.draft != ICE_PKG_FMT_VER_DFT) return ICE_ERR_CFG; /* pkg must have at least one segment */ @@ -990,6 +1268,68 @@ static enum ice_status ice_chk_pkg_version(struct ice_pkg_ver *pkg_ver) } /** + * ice_chk_pkg_compat + * @hw: pointer to the hardware structure + * @ospkg: pointer to the package hdr + * @seg: pointer to the package segment hdr + * + * This function checks the package version compatibility with driver and NVM + */ +static enum ice_status +ice_chk_pkg_compat(struct ice_hw *hw, struct ice_pkg_hdr *ospkg, + struct ice_seg **seg) +{ + struct ice_aqc_get_pkg_info_resp *pkg; + enum ice_status status; + u16 size; + u32 i; + + /* Check package version compatibility */ + status = ice_chk_pkg_version(&hw->pkg_ver); + if (status) { + ice_debug(hw, ICE_DBG_INIT, "Package version check failed.\n"); + return status; + } + + /* find ICE segment in given package */ + *seg = (struct ice_seg *)ice_find_seg_in_pkg(hw, SEGMENT_TYPE_ICE, + ospkg); + if (!*seg) { + ice_debug(hw, ICE_DBG_INIT, "no ice segment in package.\n"); + return ICE_ERR_CFG; + } + + /* Check if FW is compatible with the OS package */ + size = struct_size(pkg, pkg_info, ICE_PKG_CNT - 1); + pkg = kzalloc(size, GFP_KERNEL); + if (!pkg) + return ICE_ERR_NO_MEMORY; + + status = ice_aq_get_pkg_info_list(hw, pkg, size, NULL); + if (status) + goto fw_ddp_compat_free_alloc; + + for (i = 0; i < le32_to_cpu(pkg->count); i++) { + /* loop till we find the NVM package */ + if (!pkg->pkg_info[i].is_in_nvm) + continue; + if ((*seg)->hdr.seg_format_ver.major != + pkg->pkg_info[i].ver.major || + (*seg)->hdr.seg_format_ver.minor > + pkg->pkg_info[i].ver.minor) { + status = ICE_ERR_FW_DDP_MISMATCH; + ice_debug(hw, ICE_DBG_INIT, + "OS package is not compatible with NVM.\n"); + } + /* done processing NVM package so break */ + break; + } +fw_ddp_compat_free_alloc: + kfree(pkg); + return status; +} + +/** * ice_init_pkg - initialize/download package * @hw: pointer to the hardware structure * @buf: pointer to the package buffer @@ -1039,18 +1379,12 @@ enum ice_status ice_init_pkg(struct ice_hw *hw, u8 *buf, u32 len) /* before downloading the package, check package version for * compatibility with driver */ - status = ice_chk_pkg_version(&hw->pkg_ver); + status = ice_chk_pkg_compat(hw, pkg, &seg); if (status) return status; - /* find segment in given package */ - seg = (struct ice_seg *)ice_find_seg_in_pkg(hw, SEGMENT_TYPE_ICE, pkg); - if (!seg) { - ice_debug(hw, ICE_DBG_INIT, "no ice segment in package.\n"); - return ICE_ERR_CFG; - } - - /* download package */ + /* initialize package hints and then download package */ + ice_init_pkg_hints(hw, seg); status = ice_download_pkg(hw, seg); if (status == ICE_ERR_AQ_NO_WORK) { ice_debug(hw, ICE_DBG_INIT, @@ -1292,6 +1626,284 @@ static struct ice_buf *ice_pkg_buf(struct ice_buf_build *bld) return &bld->buf; } +/** + * ice_tunnel_port_in_use_hlpr - helper function to determine tunnel usage + * @hw: pointer to the HW structure + * @port: port to search for + * @index: optionally returns index + * + * Returns whether a port is already in use as a tunnel, and optionally its + * index + */ +static bool ice_tunnel_port_in_use_hlpr(struct ice_hw *hw, u16 port, u16 *index) +{ + u16 i; + + for (i = 0; i < hw->tnl.count && i < ICE_TUNNEL_MAX_ENTRIES; i++) + if (hw->tnl.tbl[i].in_use && hw->tnl.tbl[i].port == port) { + if (index) + *index = i; + return true; + } + + return false; +} + +/** + * ice_tunnel_port_in_use + * @hw: pointer to the HW structure + * @port: port to search for + * @index: optionally returns index + * + * Returns whether a port is already in use as a tunnel, and optionally its + * index + */ +bool ice_tunnel_port_in_use(struct ice_hw *hw, u16 port, u16 *index) +{ + bool res; + + mutex_lock(&hw->tnl_lock); + res = ice_tunnel_port_in_use_hlpr(hw, port, index); + mutex_unlock(&hw->tnl_lock); + + return res; +} + +/** + * ice_find_free_tunnel_entry + * @hw: pointer to the HW structure + * @type: tunnel type + * @index: optionally returns index + * + * Returns whether there is a free tunnel entry, and optionally its index + */ +static bool +ice_find_free_tunnel_entry(struct ice_hw *hw, enum ice_tunnel_type type, + u16 *index) +{ + u16 i; + + for (i = 0; i < hw->tnl.count && i < ICE_TUNNEL_MAX_ENTRIES; i++) + if (hw->tnl.tbl[i].valid && !hw->tnl.tbl[i].in_use && + hw->tnl.tbl[i].type == type) { + if (index) + *index = i; + return true; + } + + return false; +} + +/** + * ice_get_open_tunnel_port - retrieve an open tunnel port + * @hw: pointer to the HW structure + * @type: tunnel type (TNL_ALL will return any open port) + * @port: returns open port + */ +bool +ice_get_open_tunnel_port(struct ice_hw *hw, enum ice_tunnel_type type, + u16 *port) +{ + bool res = false; + u16 i; + + mutex_lock(&hw->tnl_lock); + + for (i = 0; i < hw->tnl.count && i < ICE_TUNNEL_MAX_ENTRIES; i++) + if (hw->tnl.tbl[i].valid && hw->tnl.tbl[i].in_use && + (type == TNL_ALL || hw->tnl.tbl[i].type == type)) { + *port = hw->tnl.tbl[i].port; + res = true; + break; + } + + mutex_unlock(&hw->tnl_lock); + + return res; +} + +/** + * ice_create_tunnel + * @hw: pointer to the HW structure + * @type: type of tunnel + * @port: port of tunnel to create + * + * Create a tunnel by updating the parse graph in the parser. We do that by + * creating a package buffer with the tunnel info and issuing an update package + * command. + */ +enum ice_status +ice_create_tunnel(struct ice_hw *hw, enum ice_tunnel_type type, u16 port) +{ + struct ice_boost_tcam_section *sect_rx, *sect_tx; + enum ice_status status = ICE_ERR_MAX_LIMIT; + struct ice_buf_build *bld; + u16 index; + + mutex_lock(&hw->tnl_lock); + + if (ice_tunnel_port_in_use_hlpr(hw, port, &index)) { + hw->tnl.tbl[index].ref++; + status = 0; + goto ice_create_tunnel_end; + } + + if (!ice_find_free_tunnel_entry(hw, type, &index)) { + status = ICE_ERR_OUT_OF_RANGE; + goto ice_create_tunnel_end; + } + + bld = ice_pkg_buf_alloc(hw); + if (!bld) { + status = ICE_ERR_NO_MEMORY; + goto ice_create_tunnel_end; + } + + /* allocate 2 sections, one for Rx parser, one for Tx parser */ + if (ice_pkg_buf_reserve_section(bld, 2)) + goto ice_create_tunnel_err; + + sect_rx = ice_pkg_buf_alloc_section(bld, ICE_SID_RXPARSER_BOOST_TCAM, + sizeof(*sect_rx)); + if (!sect_rx) + goto ice_create_tunnel_err; + sect_rx->count = cpu_to_le16(1); + + sect_tx = ice_pkg_buf_alloc_section(bld, ICE_SID_TXPARSER_BOOST_TCAM, + sizeof(*sect_tx)); + if (!sect_tx) + goto ice_create_tunnel_err; + sect_tx->count = cpu_to_le16(1); + + /* copy original boost entry to update package buffer */ + memcpy(sect_rx->tcam, hw->tnl.tbl[index].boost_entry, + sizeof(*sect_rx->tcam)); + + /* over-write the never-match dest port key bits with the encoded port + * bits + */ + ice_set_key((u8 *)§_rx->tcam[0].key, sizeof(sect_rx->tcam[0].key), + (u8 *)&port, NULL, NULL, NULL, + (u16)offsetof(struct ice_boost_key_value, hv_dst_port_key), + sizeof(sect_rx->tcam[0].key.key.hv_dst_port_key)); + + /* exact copy of entry to Tx section entry */ + memcpy(sect_tx->tcam, sect_rx->tcam, sizeof(*sect_tx->tcam)); + + status = ice_update_pkg(hw, ice_pkg_buf(bld), 1); + if (!status) { + hw->tnl.tbl[index].port = port; + hw->tnl.tbl[index].in_use = true; + hw->tnl.tbl[index].ref = 1; + } + +ice_create_tunnel_err: + ice_pkg_buf_free(hw, bld); + +ice_create_tunnel_end: + mutex_unlock(&hw->tnl_lock); + + return status; +} + +/** + * ice_destroy_tunnel + * @hw: pointer to the HW structure + * @port: port of tunnel to destroy (ignored if the all parameter is true) + * @all: flag that states to destroy all tunnels + * + * Destroys a tunnel or all tunnels by creating an update package buffer + * targeting the specific updates requested and then performing an update + * package. + */ +enum ice_status ice_destroy_tunnel(struct ice_hw *hw, u16 port, bool all) +{ + struct ice_boost_tcam_section *sect_rx, *sect_tx; + enum ice_status status = ICE_ERR_MAX_LIMIT; + struct ice_buf_build *bld; + u16 count = 0; + u16 index; + u16 size; + u16 i; + + mutex_lock(&hw->tnl_lock); + + if (!all && ice_tunnel_port_in_use_hlpr(hw, port, &index)) + if (hw->tnl.tbl[index].ref > 1) { + hw->tnl.tbl[index].ref--; + status = 0; + goto ice_destroy_tunnel_end; + } + + /* determine count */ + for (i = 0; i < hw->tnl.count && i < ICE_TUNNEL_MAX_ENTRIES; i++) + if (hw->tnl.tbl[i].valid && hw->tnl.tbl[i].in_use && + (all || hw->tnl.tbl[i].port == port)) + count++; + + if (!count) { + status = ICE_ERR_PARAM; + goto ice_destroy_tunnel_end; + } + + /* size of section - there is at least one entry */ + size = struct_size(sect_rx, tcam, count - 1); + + bld = ice_pkg_buf_alloc(hw); + if (!bld) { + status = ICE_ERR_NO_MEMORY; + goto ice_destroy_tunnel_end; + } + + /* allocate 2 sections, one for Rx parser, one for Tx parser */ + if (ice_pkg_buf_reserve_section(bld, 2)) + goto ice_destroy_tunnel_err; + + sect_rx = ice_pkg_buf_alloc_section(bld, ICE_SID_RXPARSER_BOOST_TCAM, + size); + if (!sect_rx) + goto ice_destroy_tunnel_err; + sect_rx->count = cpu_to_le16(1); + + sect_tx = ice_pkg_buf_alloc_section(bld, ICE_SID_TXPARSER_BOOST_TCAM, + size); + if (!sect_tx) + goto ice_destroy_tunnel_err; + sect_tx->count = cpu_to_le16(1); + + /* copy original boost entry to update package buffer, one copy to Rx + * section, another copy to the Tx section + */ + for (i = 0; i < hw->tnl.count && i < ICE_TUNNEL_MAX_ENTRIES; i++) + if (hw->tnl.tbl[i].valid && hw->tnl.tbl[i].in_use && + (all || hw->tnl.tbl[i].port == port)) { + memcpy(sect_rx->tcam + i, hw->tnl.tbl[i].boost_entry, + sizeof(*sect_rx->tcam)); + memcpy(sect_tx->tcam + i, hw->tnl.tbl[i].boost_entry, + sizeof(*sect_tx->tcam)); + hw->tnl.tbl[i].marked = true; + } + + status = ice_update_pkg(hw, ice_pkg_buf(bld), 1); + if (!status) + for (i = 0; i < hw->tnl.count && + i < ICE_TUNNEL_MAX_ENTRIES; i++) + if (hw->tnl.tbl[i].marked) { + hw->tnl.tbl[i].ref = 0; + hw->tnl.tbl[i].port = 0; + hw->tnl.tbl[i].in_use = false; + hw->tnl.tbl[i].marked = false; + } + +ice_destroy_tunnel_err: + ice_pkg_buf_free(hw, bld); + +ice_destroy_tunnel_end: + mutex_unlock(&hw->tnl_lock); + + return status; +} + /* PTG Management */ /** @@ -1807,9 +2419,16 @@ ice_find_prof_id(struct ice_hw *hw, enum ice_block blk, struct ice_fv_word *fv, u8 *prof_id) { struct ice_es *es = &hw->blk[blk].es; - u16 off, i; + u16 off; + u8 i; - for (i = 0; i < es->count; i++) { + /* For FD, we don't want to re-use a existed profile with the same + * field vector and mask. This will cause rule interference. + */ + if (blk == ICE_BLK_FD) + return ICE_ERR_DOES_NOT_EXIST; + + for (i = 0; i < (u8)es->count; i++) { off = i * es->fvw; if (memcmp(&es->t[off], fv, es->fvw * sizeof(*fv))) @@ -1830,6 +2449,9 @@ ice_find_prof_id(struct ice_hw *hw, enum ice_block blk, static bool ice_prof_id_rsrc_type(enum ice_block blk, u16 *rsrc_type) { switch (blk) { + case ICE_BLK_FD: + *rsrc_type = ICE_AQC_RES_TYPE_FD_PROF_BLDR_PROFID; + break; case ICE_BLK_RSS: *rsrc_type = ICE_AQC_RES_TYPE_HASH_PROF_BLDR_PROFID; break; @@ -1847,6 +2469,9 @@ static bool ice_prof_id_rsrc_type(enum ice_block blk, u16 *rsrc_type) static bool ice_tcam_ent_rsrc_type(enum ice_block blk, u16 *rsrc_type) { switch (blk) { + case ICE_BLK_FD: + *rsrc_type = ICE_AQC_RES_TYPE_FD_PROF_BLDR_TCAM; + break; case ICE_BLK_RSS: *rsrc_type = ICE_AQC_RES_TYPE_HASH_PROF_BLDR_TCAM; break; @@ -2290,6 +2915,12 @@ static void ice_free_flow_profs(struct ice_hw *hw, u8 blk_idx) mutex_lock(&hw->fl_profs_locks[blk_idx]); list_for_each_entry_safe(p, tmp, &hw->fl_profs[blk_idx], l_entry) { + struct ice_flow_entry *e, *t; + + list_for_each_entry_safe(e, t, &p->entries, l_entry) + ice_flow_rem_entry(hw, (enum ice_block)blk_idx, + ICE_FLOW_ENTRY_HNDL(e)); + list_del(&p->l_entry); devm_kfree(ice_hw_to_dev(hw), p); } @@ -2919,6 +3550,212 @@ error_tmp: } /** + * ice_update_fd_mask - set Flow Director Field Vector mask for a profile + * @hw: pointer to the HW struct + * @prof_id: profile ID + * @mask_sel: mask select + * + * This function enable any of the masks selected by the mask select parameter + * for the profile specified. + */ +static void ice_update_fd_mask(struct ice_hw *hw, u16 prof_id, u32 mask_sel) +{ + wr32(hw, GLQF_FDMASK_SEL(prof_id), mask_sel); + + ice_debug(hw, ICE_DBG_INIT, "fd mask(%d): %x = %x\n", prof_id, + GLQF_FDMASK_SEL(prof_id), mask_sel); +} + +struct ice_fd_src_dst_pair { + u8 prot_id; + u8 count; + u16 off; +}; + +static const struct ice_fd_src_dst_pair ice_fd_pairs[] = { + /* These are defined in pairs */ + { ICE_PROT_IPV4_OF_OR_S, 2, 12 }, + { ICE_PROT_IPV4_OF_OR_S, 2, 16 }, + + { ICE_PROT_IPV4_IL, 2, 12 }, + { ICE_PROT_IPV4_IL, 2, 16 }, + + { ICE_PROT_IPV6_OF_OR_S, 8, 8 }, + { ICE_PROT_IPV6_OF_OR_S, 8, 24 }, + + { ICE_PROT_IPV6_IL, 8, 8 }, + { ICE_PROT_IPV6_IL, 8, 24 }, + + { ICE_PROT_TCP_IL, 1, 0 }, + { ICE_PROT_TCP_IL, 1, 2 }, + + { ICE_PROT_UDP_OF, 1, 0 }, + { ICE_PROT_UDP_OF, 1, 2 }, + + { ICE_PROT_UDP_IL_OR_S, 1, 0 }, + { ICE_PROT_UDP_IL_OR_S, 1, 2 }, + + { ICE_PROT_SCTP_IL, 1, 0 }, + { ICE_PROT_SCTP_IL, 1, 2 } +}; + +#define ICE_FD_SRC_DST_PAIR_COUNT ARRAY_SIZE(ice_fd_pairs) + +/** + * ice_update_fd_swap - set register appropriately for a FD FV extraction + * @hw: pointer to the HW struct + * @prof_id: profile ID + * @es: extraction sequence (length of array is determined by the block) + */ +static enum ice_status +ice_update_fd_swap(struct ice_hw *hw, u16 prof_id, struct ice_fv_word *es) +{ + DECLARE_BITMAP(pair_list, ICE_FD_SRC_DST_PAIR_COUNT); + u8 pair_start[ICE_FD_SRC_DST_PAIR_COUNT] = { 0 }; +#define ICE_FD_FV_NOT_FOUND (-2) + s8 first_free = ICE_FD_FV_NOT_FOUND; + u8 used[ICE_MAX_FV_WORDS] = { 0 }; + s8 orig_free, si; + u32 mask_sel = 0; + u8 i, j, k; + + bitmap_zero(pair_list, ICE_FD_SRC_DST_PAIR_COUNT); + + /* This code assumes that the Flow Director field vectors are assigned + * from the end of the FV indexes working towards the zero index, that + * only complete fields will be included and will be consecutive, and + * that there are no gaps between valid indexes. + */ + + /* Determine swap fields present */ + for (i = 0; i < hw->blk[ICE_BLK_FD].es.fvw; i++) { + /* Find the first free entry, assuming right to left population. + * This is where we can start adding additional pairs if needed. + */ + if (first_free == ICE_FD_FV_NOT_FOUND && es[i].prot_id != + ICE_PROT_INVALID) + first_free = i - 1; + + for (j = 0; j < ICE_FD_SRC_DST_PAIR_COUNT; j++) + if (es[i].prot_id == ice_fd_pairs[j].prot_id && + es[i].off == ice_fd_pairs[j].off) { + set_bit(j, pair_list); + pair_start[j] = i; + } + } + + orig_free = first_free; + + /* determine missing swap fields that need to be added */ + for (i = 0; i < ICE_FD_SRC_DST_PAIR_COUNT; i += 2) { + u8 bit1 = test_bit(i + 1, pair_list); + u8 bit0 = test_bit(i, pair_list); + + if (bit0 ^ bit1) { + u8 index; + + /* add the appropriate 'paired' entry */ + if (!bit0) + index = i; + else + index = i + 1; + + /* check for room */ + if (first_free + 1 < (s8)ice_fd_pairs[index].count) + return ICE_ERR_MAX_LIMIT; + + /* place in extraction sequence */ + for (k = 0; k < ice_fd_pairs[index].count; k++) { + es[first_free - k].prot_id = + ice_fd_pairs[index].prot_id; + es[first_free - k].off = + ice_fd_pairs[index].off + (k * 2); + + if (k > first_free) + return ICE_ERR_OUT_OF_RANGE; + + /* keep track of non-relevant fields */ + mask_sel |= BIT(first_free - k); + } + + pair_start[index] = first_free; + first_free -= ice_fd_pairs[index].count; + } + } + + /* fill in the swap array */ + si = hw->blk[ICE_BLK_FD].es.fvw - 1; + while (si >= 0) { + u8 indexes_used = 1; + + /* assume flat at this index */ +#define ICE_SWAP_VALID 0x80 + used[si] = si | ICE_SWAP_VALID; + + if (orig_free == ICE_FD_FV_NOT_FOUND || si <= orig_free) { + si -= indexes_used; + continue; + } + + /* check for a swap location */ + for (j = 0; j < ICE_FD_SRC_DST_PAIR_COUNT; j++) + if (es[si].prot_id == ice_fd_pairs[j].prot_id && + es[si].off == ice_fd_pairs[j].off) { + u8 idx; + + /* determine the appropriate matching field */ + idx = j + ((j % 2) ? -1 : 1); + + indexes_used = ice_fd_pairs[idx].count; + for (k = 0; k < indexes_used; k++) { + used[si - k] = (pair_start[idx] - k) | + ICE_SWAP_VALID; + } + + break; + } + + si -= indexes_used; + } + + /* for each set of 4 swap and 4 inset indexes, write the appropriate + * register + */ + for (j = 0; j < hw->blk[ICE_BLK_FD].es.fvw / 4; j++) { + u32 raw_swap = 0; + u32 raw_in = 0; + + for (k = 0; k < 4; k++) { + u8 idx; + + idx = (j * 4) + k; + if (used[idx] && !(mask_sel & BIT(idx))) { + raw_swap |= used[idx] << (k * BITS_PER_BYTE); +#define ICE_INSET_DFLT 0x9f + raw_in |= ICE_INSET_DFLT << (k * BITS_PER_BYTE); + } + } + + /* write the appropriate swap register set */ + wr32(hw, GLQF_FDSWAP(prof_id, j), raw_swap); + + ice_debug(hw, ICE_DBG_INIT, "swap wr(%d, %d): %x = %08x\n", + prof_id, j, GLQF_FDSWAP(prof_id, j), raw_swap); + + /* write the appropriate inset register set */ + wr32(hw, GLQF_FDINSET(prof_id, j), raw_in); + + ice_debug(hw, ICE_DBG_INIT, "inset wr(%d, %d): %x = %08x\n", + prof_id, j, GLQF_FDINSET(prof_id, j), raw_in); + } + + /* initially clear the mask select for this profile */ + ice_update_fd_mask(hw, prof_id, 0); + + return 0; +} + +/** * ice_add_prof - add profile * @hw: pointer to the HW struct * @blk: hardware block @@ -2939,7 +3776,7 @@ ice_add_prof(struct ice_hw *hw, enum ice_block blk, u64 id, u8 ptypes[], DECLARE_BITMAP(ptgs_used, ICE_XLT1_CNT); struct ice_prof_map *prof; enum ice_status status; - u32 byte = 0; + u8 byte = 0; u8 prof_id; bitmap_zero(ptgs_used, ICE_XLT1_CNT); @@ -2953,6 +3790,18 @@ ice_add_prof(struct ice_hw *hw, enum ice_block blk, u64 id, u8 ptypes[], status = ice_alloc_prof_id(hw, blk, &prof_id); if (status) goto err_ice_add_prof; + if (blk == ICE_BLK_FD) { + /* For Flow Director block, the extraction sequence may + * need to be altered in the case where there are paired + * fields that have no match. This is necessary because + * for Flow Director, src and dest fields need to paired + * for filter programming and these values are swapped + * during Tx. + */ + status = ice_update_fd_swap(hw, prof_id, es); + if (status) + goto err_ice_add_prof; + } /* and write new es */ ice_write_es(hw, blk, prof_id, es); @@ -2962,8 +3811,10 @@ ice_add_prof(struct ice_hw *hw, enum ice_block blk, u64 id, u8 ptypes[], /* add profile info */ prof = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*prof), GFP_KERNEL); - if (!prof) + if (!prof) { + status = ICE_ERR_NO_MEMORY; goto err_ice_add_prof; + } prof->profile_cookie = id; prof->prof_id = prof_id; @@ -2972,7 +3823,7 @@ ice_add_prof(struct ice_hw *hw, enum ice_block blk, u64 id, u8 ptypes[], /* build list of ptgs */ while (bytes && prof->ptg_cnt < ICE_MAX_PTG_PER_PROFILE) { - u32 bit; + u8 bit; if (!ptypes[byte]) { bytes--; @@ -3006,7 +3857,7 @@ ice_add_prof(struct ice_hw *hw, enum ice_block blk, u64 id, u8 ptypes[], break; /* nothing left in byte, then exit */ - m = ~((1 << (bit + 1)) - 1); + m = ~(u8)((1 << (bit + 1)) - 1); if (!(ptypes[byte] & m)) break; } @@ -3703,8 +4554,10 @@ ice_add_prof_id_vsig(struct ice_hw *hw, enum ice_block blk, u16 vsig, u64 hdl, t->tcam[i].prof_id, t->tcam[i].ptg, vsig, 0, 0, vl_msk, dc_msk, nm_msk); - if (status) + if (status) { + devm_kfree(ice_hw_to_dev(hw), p); goto err_ice_add_prof_id_vsig; + } /* log change */ list_add(&p->list_entry, chg); diff --git a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h index c7b5e1a6ea2b..568ea519af51 100644 --- a/drivers/net/ethernet/intel/ice/ice_flex_pipe.h +++ b/drivers/net/ethernet/intel/ice/ice_flex_pipe.h @@ -18,6 +18,14 @@ #define ICE_PKG_CNT 4 +bool +ice_get_open_tunnel_port(struct ice_hw *hw, enum ice_tunnel_type type, + u16 *port); +enum ice_status +ice_create_tunnel(struct ice_hw *hw, enum ice_tunnel_type type, u16 port); +enum ice_status ice_destroy_tunnel(struct ice_hw *hw, u16 port, bool all); +bool ice_tunnel_port_in_use(struct ice_hw *hw, u16 port, u16 *index); + enum ice_status ice_add_prof(struct ice_hw *hw, enum ice_block blk, u64 id, u8 ptypes[], struct ice_fv_word *es); diff --git a/drivers/net/ethernet/intel/ice/ice_flex_type.h b/drivers/net/ethernet/intel/ice/ice_flex_type.h index 0fb3fe3ff3ea..a6f391eac8ff 100644 --- a/drivers/net/ethernet/intel/ice/ice_flex_type.h +++ b/drivers/net/ethernet/intel/ice/ice_flex_type.h @@ -20,7 +20,7 @@ struct ice_fv { /* Package and segment headers and tables */ struct ice_pkg_hdr { - struct ice_pkg_ver format_ver; + struct ice_pkg_ver pkg_format_ver; __le32 seg_count; __le32 seg_offset[1]; }; @@ -30,9 +30,9 @@ struct ice_generic_seg_hdr { #define SEGMENT_TYPE_METADATA 0x00000001 #define SEGMENT_TYPE_ICE 0x00000010 __le32 seg_type; - struct ice_pkg_ver seg_ver; + struct ice_pkg_ver seg_format_ver; __le32 seg_size; - char seg_name[ICE_PKG_NAME_SIZE]; + char seg_id[ICE_PKG_NAME_SIZE]; }; /* ice specific segment */ @@ -75,7 +75,7 @@ struct ice_buf_table { struct ice_global_metadata_seg { struct ice_generic_seg_hdr hdr; struct ice_pkg_ver pkg_ver; - __le32 track_id; + __le32 rsvd; char pkg_name[ICE_PKG_NAME_SIZE]; }; @@ -149,6 +149,7 @@ struct ice_buf_hdr { #define ICE_SID_CDID_REDIR_RSS 48 #define ICE_SID_RXPARSER_BOOST_TCAM 56 +#define ICE_SID_TXPARSER_BOOST_TCAM 66 #define ICE_SID_XLT0_PE 80 #define ICE_SID_XLT_KEY_BUILDER_PE 81 @@ -291,6 +292,38 @@ struct ice_pkg_enum { void *(*handler)(u32 sect_type, void *section, u32 index, u32 *offset); }; +/* Tunnel enabling */ + +enum ice_tunnel_type { + TNL_VXLAN = 0, + TNL_GENEVE, + TNL_LAST = 0xFF, + TNL_ALL = 0xFF, +}; + +struct ice_tunnel_type_scan { + enum ice_tunnel_type type; + const char *label_prefix; +}; + +struct ice_tunnel_entry { + enum ice_tunnel_type type; + u16 boost_addr; + u16 port; + u16 ref; + struct ice_boost_tcam_entry *boost_entry; + u8 valid; + u8 in_use; + u8 marked; +}; + +#define ICE_TUNNEL_MAX_ENTRIES 16 + +struct ice_tunnel_table { + struct ice_tunnel_entry tbl[ICE_TUNNEL_MAX_ENTRIES]; + u16 count; +}; + struct ice_pkg_es { __le16 count; __le16 offset; diff --git a/drivers/net/ethernet/intel/ice/ice_flow.c b/drivers/net/ethernet/intel/ice/ice_flow.c index 3de862a3c789..d74e5290677f 100644 --- a/drivers/net/ethernet/intel/ice/ice_flow.c +++ b/drivers/net/ethernet/intel/ice/ice_flow.c @@ -42,7 +42,10 @@ struct ice_flow_field_info ice_flds_info[ICE_FLOW_FIELD_IDX_MAX] = { ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_SCTP, 0, sizeof(__be16)), /* ICE_FLOW_FIELD_IDX_SCTP_DST_PORT */ ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_SCTP, 2, sizeof(__be16)), - + /* GRE */ + /* ICE_FLOW_FIELD_IDX_GRE_KEYID */ + ICE_FLOW_FLD_INFO(ICE_FLOW_SEG_HDR_GRE, 12, + sizeof_field(struct gre_full_hdr, key)), }; /* Bitmaps indicating relevant packet types for a particular protocol header @@ -134,6 +137,18 @@ static const u32 ice_ptypes_sctp_il[] = { 0x00000000, 0x00000000, 0x00000000, 0x00000000, }; +/* Packet types for packets with an Outermost/First GRE header */ +static const u32 ice_ptypes_gre_of[] = { + 0x00000000, 0xBFBF7800, 0x000001DF, 0xFEFDE000, + 0x0000017E, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, + 0x00000000, 0x00000000, 0x00000000, 0x00000000, +}; + /* Manage parameters and info. used during the creation of a flow profile */ struct ice_flow_prof_params { enum ice_block blk; @@ -178,6 +193,40 @@ ice_flow_val_hdrs(struct ice_flow_seg_info *segs, u8 segs_cnt) return 0; } +/* Sizes of fixed known protocol headers without header options */ +#define ICE_FLOW_PROT_HDR_SZ_MAC 14 +#define ICE_FLOW_PROT_HDR_SZ_IPV4 20 +#define ICE_FLOW_PROT_HDR_SZ_IPV6 40 +#define ICE_FLOW_PROT_HDR_SZ_TCP 20 +#define ICE_FLOW_PROT_HDR_SZ_UDP 8 +#define ICE_FLOW_PROT_HDR_SZ_SCTP 12 + +/** + * ice_flow_calc_seg_sz - calculates size of a packet segment based on headers + * @params: information about the flow to be processed + * @seg: index of packet segment whose header size is to be determined + */ +static u16 ice_flow_calc_seg_sz(struct ice_flow_prof_params *params, u8 seg) +{ + u16 sz = ICE_FLOW_PROT_HDR_SZ_MAC; + + /* L3 headers */ + if (params->prof->segs[seg].hdrs & ICE_FLOW_SEG_HDR_IPV4) + sz += ICE_FLOW_PROT_HDR_SZ_IPV4; + else if (params->prof->segs[seg].hdrs & ICE_FLOW_SEG_HDR_IPV6) + sz += ICE_FLOW_PROT_HDR_SZ_IPV6; + + /* L4 headers */ + if (params->prof->segs[seg].hdrs & ICE_FLOW_SEG_HDR_TCP) + sz += ICE_FLOW_PROT_HDR_SZ_TCP; + else if (params->prof->segs[seg].hdrs & ICE_FLOW_SEG_HDR_UDP) + sz += ICE_FLOW_PROT_HDR_SZ_UDP; + else if (params->prof->segs[seg].hdrs & ICE_FLOW_SEG_HDR_SCTP) + sz += ICE_FLOW_PROT_HDR_SZ_SCTP; + + return sz; +} + /** * ice_flow_proc_seg_hdrs - process protocol headers present in pkt segments * @params: information about the flow to be processed @@ -225,6 +274,12 @@ ice_flow_proc_seg_hdrs(struct ice_flow_prof_params *params) src = (const unsigned long *)ice_ptypes_sctp_il; bitmap_and(params->ptypes, params->ptypes, src, ICE_FLOW_PTYPE_MAX); + } else if (hdrs & ICE_FLOW_SEG_HDR_GRE) { + if (!i) { + src = (const unsigned long *)ice_ptypes_gre_of; + bitmap_and(params->ptypes, params->ptypes, + src, ICE_FLOW_PTYPE_MAX); + } } } @@ -275,6 +330,9 @@ ice_flow_xtract_fld(struct ice_hw *hw, struct ice_flow_prof_params *params, case ICE_FLOW_FIELD_IDX_SCTP_DST_PORT: prot_id = ICE_PROT_SCTP_IL; break; + case ICE_FLOW_FIELD_IDX_GRE_KEYID: + prot_id = ICE_PROT_GRE_OF; + break; default: return ICE_ERR_NOT_IMPL; } @@ -324,6 +382,81 @@ ice_flow_xtract_fld(struct ice_hw *hw, struct ice_flow_prof_params *params, } /** + * ice_flow_xtract_raws - Create extract sequence entries for raw bytes + * @hw: pointer to the HW struct + * @params: information about the flow to be processed + * @seg: index of packet segment whose raw fields are to be be extracted + */ +static enum ice_status +ice_flow_xtract_raws(struct ice_hw *hw, struct ice_flow_prof_params *params, + u8 seg) +{ + u16 fv_words; + u16 hdrs_sz; + u8 i; + + if (!params->prof->segs[seg].raws_cnt) + return 0; + + if (params->prof->segs[seg].raws_cnt > + ARRAY_SIZE(params->prof->segs[seg].raws)) + return ICE_ERR_MAX_LIMIT; + + /* Offsets within the segment headers are not supported */ + hdrs_sz = ice_flow_calc_seg_sz(params, seg); + if (!hdrs_sz) + return ICE_ERR_PARAM; + + fv_words = hw->blk[params->blk].es.fvw; + + for (i = 0; i < params->prof->segs[seg].raws_cnt; i++) { + struct ice_flow_seg_fld_raw *raw; + u16 off, cnt, j; + + raw = ¶ms->prof->segs[seg].raws[i]; + + /* Storing extraction information */ + raw->info.xtrct.prot_id = ICE_PROT_MAC_OF_OR_S; + raw->info.xtrct.off = (raw->off / ICE_FLOW_FV_EXTRACT_SZ) * + ICE_FLOW_FV_EXTRACT_SZ; + raw->info.xtrct.disp = (raw->off % ICE_FLOW_FV_EXTRACT_SZ) * + BITS_PER_BYTE; + raw->info.xtrct.idx = params->es_cnt; + + /* Determine the number of field vector entries this raw field + * consumes. + */ + cnt = DIV_ROUND_UP(raw->info.xtrct.disp + + (raw->info.src.last * BITS_PER_BYTE), + (ICE_FLOW_FV_EXTRACT_SZ * BITS_PER_BYTE)); + off = raw->info.xtrct.off; + for (j = 0; j < cnt; j++) { + u16 idx; + + /* Make sure the number of extraction sequence required + * does not exceed the block's capability + */ + if (params->es_cnt >= hw->blk[params->blk].es.count || + params->es_cnt >= ICE_MAX_FV_WORDS) + return ICE_ERR_MAX_LIMIT; + + /* some blocks require a reversed field vector layout */ + if (hw->blk[params->blk].es.reverse) + idx = fv_words - params->es_cnt - 1; + else + idx = params->es_cnt; + + params->es[idx].prot_id = raw->info.xtrct.prot_id; + params->es[idx].off = off; + params->es_cnt++; + off += ICE_FLOW_FV_EXTRACT_SZ; + } + } + + return 0; +} + +/** * ice_flow_create_xtrct_seq - Create an extraction sequence for given segments * @hw: pointer to the HW struct * @params: information about the flow to be processed @@ -349,6 +482,11 @@ ice_flow_create_xtrct_seq(struct ice_hw *hw, if (status) return status; } + + /* Process raw matching bytes */ + status = ice_flow_xtract_raws(hw, params, i); + if (status) + return status; } return status; @@ -373,10 +511,8 @@ ice_flow_proc_segs(struct ice_hw *hw, struct ice_flow_prof_params *params) return status; switch (params->blk) { + case ICE_BLK_FD: case ICE_BLK_RSS: - /* Only header information is provided for RSS configuration. - * No further processing is needed. - */ status = 0; break; default: @@ -458,6 +594,43 @@ ice_flow_find_prof_id(struct ice_hw *hw, enum ice_block blk, u64 prof_id) } /** + * ice_dealloc_flow_entry - Deallocate flow entry memory + * @hw: pointer to the HW struct + * @entry: flow entry to be removed + */ +static void +ice_dealloc_flow_entry(struct ice_hw *hw, struct ice_flow_entry *entry) +{ + if (!entry) + return; + + if (entry->entry) + devm_kfree(ice_hw_to_dev(hw), entry->entry); + + devm_kfree(ice_hw_to_dev(hw), entry); +} + +/** + * ice_flow_rem_entry_sync - Remove a flow entry + * @hw: pointer to the HW struct + * @blk: classification stage + * @entry: flow entry to be removed + */ +static enum ice_status +ice_flow_rem_entry_sync(struct ice_hw *hw, enum ice_block __always_unused blk, + struct ice_flow_entry *entry) +{ + if (!entry) + return ICE_ERR_BAD_PTR; + + list_del(&entry->l_entry); + + ice_dealloc_flow_entry(hw, entry); + + return 0; +} + +/** * ice_flow_add_prof_sync - Add a flow profile for packet segments and fields * @hw: pointer to the HW struct * @blk: classification stage @@ -544,6 +717,21 @@ ice_flow_rem_prof_sync(struct ice_hw *hw, enum ice_block blk, { enum ice_status status; + /* Remove all remaining flow entries before removing the flow profile */ + if (!list_empty(&prof->entries)) { + struct ice_flow_entry *e, *t; + + mutex_lock(&prof->entries_lock); + + list_for_each_entry_safe(e, t, &prof->entries, l_entry) { + status = ice_flow_rem_entry_sync(hw, blk, e); + if (status) + break; + } + + mutex_unlock(&prof->entries_lock); + } + /* Remove all hardware profiles associated with this flow profile */ status = ice_rem_prof(hw, blk, prof->id); if (!status) { @@ -629,7 +817,7 @@ ice_flow_disassoc_prof(struct ice_hw *hw, enum ice_block blk, * @segs_cnt: number of packet segments provided * @prof: stores the returned flow profile added */ -static enum ice_status +enum ice_status ice_flow_add_prof(struct ice_hw *hw, enum ice_block blk, enum ice_flow_dir dir, u64 prof_id, struct ice_flow_seg_info *segs, u8 segs_cnt, struct ice_flow_prof **prof) @@ -667,7 +855,7 @@ ice_flow_add_prof(struct ice_hw *hw, enum ice_block blk, enum ice_flow_dir dir, * @blk: the block for which the flow profile is to be removed * @prof_id: unique ID of the flow profile to be removed */ -static enum ice_status +enum ice_status ice_flow_rem_prof(struct ice_hw *hw, enum ice_block blk, u64 prof_id) { struct ice_flow_prof *prof; @@ -691,6 +879,113 @@ out: } /** + * ice_flow_add_entry - Add a flow entry + * @hw: pointer to the HW struct + * @blk: classification stage + * @prof_id: ID of the profile to add a new flow entry to + * @entry_id: unique ID to identify this flow entry + * @vsi_handle: software VSI handle for the flow entry + * @prio: priority of the flow entry + * @data: pointer to a data buffer containing flow entry's match values/masks + * @entry_h: pointer to buffer that receives the new flow entry's handle + */ +enum ice_status +ice_flow_add_entry(struct ice_hw *hw, enum ice_block blk, u64 prof_id, + u64 entry_id, u16 vsi_handle, enum ice_flow_priority prio, + void *data, u64 *entry_h) +{ + struct ice_flow_entry *e = NULL; + struct ice_flow_prof *prof; + enum ice_status status; + + /* No flow entry data is expected for RSS */ + if (!entry_h || (!data && blk != ICE_BLK_RSS)) + return ICE_ERR_BAD_PTR; + + if (!ice_is_vsi_valid(hw, vsi_handle)) + return ICE_ERR_PARAM; + + mutex_lock(&hw->fl_profs_locks[blk]); + + prof = ice_flow_find_prof_id(hw, blk, prof_id); + if (!prof) { + status = ICE_ERR_DOES_NOT_EXIST; + } else { + /* Allocate memory for the entry being added and associate + * the VSI to the found flow profile + */ + e = devm_kzalloc(ice_hw_to_dev(hw), sizeof(*e), GFP_KERNEL); + if (!e) + status = ICE_ERR_NO_MEMORY; + else + status = ice_flow_assoc_prof(hw, blk, prof, vsi_handle); + } + + mutex_unlock(&hw->fl_profs_locks[blk]); + if (status) + goto out; + + e->id = entry_id; + e->vsi_handle = vsi_handle; + e->prof = prof; + e->priority = prio; + + switch (blk) { + case ICE_BLK_FD: + case ICE_BLK_RSS: + break; + default: + status = ICE_ERR_NOT_IMPL; + goto out; + } + + mutex_lock(&prof->entries_lock); + list_add(&e->l_entry, &prof->entries); + mutex_unlock(&prof->entries_lock); + + *entry_h = ICE_FLOW_ENTRY_HNDL(e); + +out: + if (status && e) { + if (e->entry) + devm_kfree(ice_hw_to_dev(hw), e->entry); + devm_kfree(ice_hw_to_dev(hw), e); + } + + return status; +} + +/** + * ice_flow_rem_entry - Remove a flow entry + * @hw: pointer to the HW struct + * @blk: classification stage + * @entry_h: handle to the flow entry to be removed + */ +enum ice_status ice_flow_rem_entry(struct ice_hw *hw, enum ice_block blk, + u64 entry_h) +{ + struct ice_flow_entry *entry; + struct ice_flow_prof *prof; + enum ice_status status = 0; + + if (entry_h == ICE_FLOW_ENTRY_HANDLE_INVAL) + return ICE_ERR_PARAM; + + entry = ICE_FLOW_ENTRY_PTR(entry_h); + + /* Retain the pointer to the flow profile as the entry will be freed */ + prof = entry->prof; + + if (prof) { + mutex_lock(&prof->entries_lock); + status = ice_flow_rem_entry_sync(hw, blk, entry); + mutex_unlock(&prof->entries_lock); + } + + return status; +} + +/** * ice_flow_set_fld_ext - specifies locations of field from entry's input buffer * @seg: packet segment the field being set belongs to * @fld: field to be set @@ -752,7 +1047,7 @@ ice_flow_set_fld_ext(struct ice_flow_seg_info *seg, enum ice_flow_field fld, * create the content of a match entry. This function should only be used for * fixed-size data structures. */ -static void +void ice_flow_set_fld(struct ice_flow_seg_info *seg, enum ice_flow_field fld, u16 val_loc, u16 mask_loc, u16 last_loc, bool range) { @@ -762,6 +1057,42 @@ ice_flow_set_fld(struct ice_flow_seg_info *seg, enum ice_flow_field fld, ice_flow_set_fld_ext(seg, fld, t, val_loc, mask_loc, last_loc); } +/** + * ice_flow_add_fld_raw - sets locations of a raw field from entry's input buf + * @seg: packet segment the field being set belongs to + * @off: offset of the raw field from the beginning of the segment in bytes + * @len: length of the raw pattern to be matched + * @val_loc: location of the value to match from entry's input buffer + * @mask_loc: location of mask value from entry's input buffer + * + * This function specifies the offset of the raw field to be match from the + * beginning of the specified packet segment, and the locations, in the form of + * byte offsets from the start of the input buffer for a flow entry, from where + * the value to match and the mask value to be extracted. These locations are + * then stored in the flow profile. When adding flow entries to the associated + * flow profile, these locations can be used to quickly extract the values to + * create the content of a match entry. This function should only be used for + * fixed-size data structures. + */ +void +ice_flow_add_fld_raw(struct ice_flow_seg_info *seg, u16 off, u8 len, + u16 val_loc, u16 mask_loc) +{ + if (seg->raws_cnt < ICE_FLOW_SEG_RAW_FLD_MAX) { + seg->raws[seg->raws_cnt].off = off; + seg->raws[seg->raws_cnt].info.type = ICE_FLOW_FLD_TYPE_SIZE; + seg->raws[seg->raws_cnt].info.src.val = val_loc; + seg->raws[seg->raws_cnt].info.src.mask = mask_loc; + /* The "last" field is used to store the length of the field */ + seg->raws[seg->raws_cnt].info.src.last = len; + } + + /* Overflows of "raws" will be handled as an error condition later in + * the flow when this information is processed. + */ + seg->raws_cnt++; +} + #define ICE_FLOW_RSS_SEG_HDR_L3_MASKS \ (ICE_FLOW_SEG_HDR_IPV4 | ICE_FLOW_SEG_HDR_IPV6) @@ -945,6 +1276,7 @@ ice_add_rss_list(struct ice_hw *hw, u16 vsi_handle, struct ice_flow_prof *prof) #define ICE_FLOW_PROF_ENCAP_M (BIT_ULL(ICE_FLOW_PROF_ENCAP_S)) #define ICE_RSS_OUTER_HEADERS 1 +#define ICE_RSS_INNER_HEADERS 2 /* Flow profile ID format: * [0:31] - Packet match fields @@ -1085,6 +1417,9 @@ ice_add_rss_cfg(struct ice_hw *hw, u16 vsi_handle, u64 hashed_flds, mutex_lock(&hw->rss_locks); status = ice_add_rss_cfg_sync(hw, vsi_handle, hashed_flds, addl_hdrs, ICE_RSS_OUTER_HEADERS); + if (!status) + status = ice_add_rss_cfg_sync(hw, vsi_handle, hashed_flds, + addl_hdrs, ICE_RSS_INNER_HEADERS); mutex_unlock(&hw->rss_locks); return status; @@ -1238,6 +1573,12 @@ enum ice_status ice_replay_rss_cfg(struct ice_hw *hw, u16 vsi_handle) ICE_RSS_OUTER_HEADERS); if (status) break; + status = ice_add_rss_cfg_sync(hw, vsi_handle, + r->hashed_flds, + r->packet_hdr, + ICE_RSS_INNER_HEADERS); + if (status) + break; } } mutex_unlock(&hw->rss_locks); diff --git a/drivers/net/ethernet/intel/ice/ice_flow.h b/drivers/net/ethernet/intel/ice/ice_flow.h index 5558627bd5eb..3913da2116d2 100644 --- a/drivers/net/ethernet/intel/ice/ice_flow.h +++ b/drivers/net/ethernet/intel/ice/ice_flow.h @@ -43,6 +43,7 @@ enum ice_flow_seg_hdr { ICE_FLOW_SEG_HDR_TCP = 0x00000040, ICE_FLOW_SEG_HDR_UDP = 0x00000080, ICE_FLOW_SEG_HDR_SCTP = 0x00000100, + ICE_FLOW_SEG_HDR_GRE = 0x00000200, }; enum ice_flow_field { @@ -58,6 +59,8 @@ enum ice_flow_field { ICE_FLOW_FIELD_IDX_UDP_DST_PORT, ICE_FLOW_FIELD_IDX_SCTP_SRC_PORT, ICE_FLOW_FIELD_IDX_SCTP_DST_PORT, + /* GRE */ + ICE_FLOW_FIELD_IDX_GRE_KEYID, /* The total number of enums must not exceed 64 */ ICE_FLOW_FIELD_IDX_MAX }; @@ -125,6 +128,7 @@ enum ice_flow_priority { }; #define ICE_FLOW_SEG_MAX 2 +#define ICE_FLOW_SEG_RAW_FLD_MAX 2 #define ICE_FLOW_FV_EXTRACT_SZ 2 #define ICE_FLOW_SET_HDRS(seg, val) ((seg)->hdrs |= (u32)(val)) @@ -161,14 +165,38 @@ struct ice_flow_fld_info { struct ice_flow_seg_xtrct xtrct; }; +struct ice_flow_seg_fld_raw { + struct ice_flow_fld_info info; + u16 off; /* Offset from the start of the segment */ +}; + struct ice_flow_seg_info { u32 hdrs; /* Bitmask indicating protocol headers present */ u64 match; /* Bitmask indicating header fields to be matched */ u64 range; /* Bitmask indicating header fields matched as ranges */ struct ice_flow_fld_info fields[ICE_FLOW_FIELD_IDX_MAX]; + + u8 raws_cnt; /* Number of raw fields to be matched */ + struct ice_flow_seg_fld_raw raws[ICE_FLOW_SEG_RAW_FLD_MAX]; }; +/* This structure describes a flow entry, and is tracked only in this file */ +struct ice_flow_entry { + struct list_head l_entry; + + u64 id; + struct ice_flow_prof *prof; + /* Flow entry's content */ + void *entry; + enum ice_flow_priority priority; + u16 vsi_handle; + u16 entry_sz; +}; + +#define ICE_FLOW_ENTRY_HNDL(e) ((u64)e) +#define ICE_FLOW_ENTRY_PTR(h) ((struct ice_flow_entry *)(h)) + struct ice_flow_prof { struct list_head l_entry; @@ -194,7 +222,24 @@ struct ice_rss_cfg { u32 packet_hdr; }; -enum ice_status ice_flow_rem_entry(struct ice_hw *hw, u64 entry_h); +enum ice_status +ice_flow_add_prof(struct ice_hw *hw, enum ice_block blk, enum ice_flow_dir dir, + u64 prof_id, struct ice_flow_seg_info *segs, u8 segs_cnt, + struct ice_flow_prof **prof); +enum ice_status +ice_flow_rem_prof(struct ice_hw *hw, enum ice_block blk, u64 prof_id); +enum ice_status +ice_flow_add_entry(struct ice_hw *hw, enum ice_block blk, u64 prof_id, + u64 entry_id, u16 vsi, enum ice_flow_priority prio, + void *data, u64 *entry_h); +enum ice_status +ice_flow_rem_entry(struct ice_hw *hw, enum ice_block blk, u64 entry_h); +void +ice_flow_set_fld(struct ice_flow_seg_info *seg, enum ice_flow_field fld, + u16 val_loc, u16 mask_loc, u16 last_loc, bool range); +void +ice_flow_add_fld_raw(struct ice_flow_seg_info *seg, u16 off, u8 len, + u16 val_loc, u16 mask_loc); void ice_rem_vsi_rss_list(struct ice_hw *hw, u16 vsi_handle); enum ice_status ice_replay_rss_cfg(struct ice_hw *hw, u16 vsi_handle); enum ice_status diff --git a/drivers/net/ethernet/intel/ice/ice_fltr.c b/drivers/net/ethernet/intel/ice/ice_fltr.c new file mode 100644 index 000000000000..2418d4fff037 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_fltr.c @@ -0,0 +1,397 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (C) 2018-2020, Intel Corporation. */ + +#include "ice.h" +#include "ice_fltr.h" + +/** + * ice_fltr_free_list - free filter lists helper + * @dev: pointer to the device struct + * @h: pointer to the list head to be freed + * + * Helper function to free filter lists previously created using + * ice_fltr_add_mac_to_list + */ +void ice_fltr_free_list(struct device *dev, struct list_head *h) +{ + struct ice_fltr_list_entry *e, *tmp; + + list_for_each_entry_safe(e, tmp, h, list_entry) { + list_del(&e->list_entry); + devm_kfree(dev, e); + } +} + +/** + * ice_fltr_add_entry_to_list - allocate and add filter entry to list + * @dev: pointer to device needed by alloc function + * @info: filter info struct that gets added to the passed in list + * @list: pointer to the list which contains MAC filters entry + */ +static int +ice_fltr_add_entry_to_list(struct device *dev, struct ice_fltr_info *info, + struct list_head *list) +{ + struct ice_fltr_list_entry *entry; + + entry = devm_kzalloc(dev, sizeof(*entry), GFP_ATOMIC); + if (!entry) + return -ENOMEM; + + entry->fltr_info = *info; + + INIT_LIST_HEAD(&entry->list_entry); + list_add(&entry->list_entry, list); + + return 0; +} + +/** + * ice_fltr_add_mac_list - add list of MAC filters + * @vsi: pointer to VSI struct + * @list: list of filters + */ +enum ice_status +ice_fltr_add_mac_list(struct ice_vsi *vsi, struct list_head *list) +{ + return ice_add_mac(&vsi->back->hw, list); +} + +/** + * ice_fltr_remove_mac_list - remove list of MAC filters + * @vsi: pointer to VSI struct + * @list: list of filters + */ +enum ice_status +ice_fltr_remove_mac_list(struct ice_vsi *vsi, struct list_head *list) +{ + return ice_remove_mac(&vsi->back->hw, list); +} + +/** + * ice_fltr_add_vlan_list - add list of VLAN filters + * @vsi: pointer to VSI struct + * @list: list of filters + */ +static enum ice_status +ice_fltr_add_vlan_list(struct ice_vsi *vsi, struct list_head *list) +{ + return ice_add_vlan(&vsi->back->hw, list); +} + +/** + * ice_fltr_remove_vlan_list - remove list of VLAN filters + * @vsi: pointer to VSI struct + * @list: list of filters + */ +static enum ice_status +ice_fltr_remove_vlan_list(struct ice_vsi *vsi, struct list_head *list) +{ + return ice_remove_vlan(&vsi->back->hw, list); +} + +/** + * ice_fltr_add_eth_list - add list of ethertype filters + * @vsi: pointer to VSI struct + * @list: list of filters + */ +static enum ice_status +ice_fltr_add_eth_list(struct ice_vsi *vsi, struct list_head *list) +{ + return ice_add_eth_mac(&vsi->back->hw, list); +} + +/** + * ice_fltr_remove_eth_list - remove list of ethertype filters + * @vsi: pointer to VSI struct + * @list: list of filters + */ +static enum ice_status +ice_fltr_remove_eth_list(struct ice_vsi *vsi, struct list_head *list) +{ + return ice_remove_eth_mac(&vsi->back->hw, list); +} + +/** + * ice_fltr_remove_all - remove all filters associated with VSI + * @vsi: pointer to VSI struct + */ +void ice_fltr_remove_all(struct ice_vsi *vsi) +{ + ice_remove_vsi_fltr(&vsi->back->hw, vsi->idx); +} + +/** + * ice_fltr_add_mac_to_list - add MAC filter info to exsisting list + * @vsi: pointer to VSI struct + * @list: list to add filter info to + * @mac: MAC address to add + * @action: filter action + */ +int +ice_fltr_add_mac_to_list(struct ice_vsi *vsi, struct list_head *list, + const u8 *mac, enum ice_sw_fwd_act_type action) +{ + struct ice_fltr_info info = { 0 }; + + info.flag = ICE_FLTR_TX; + info.src_id = ICE_SRC_ID_VSI; + info.lkup_type = ICE_SW_LKUP_MAC; + info.fltr_act = action; + info.vsi_handle = vsi->idx; + + ether_addr_copy(info.l_data.mac.mac_addr, mac); + + return ice_fltr_add_entry_to_list(ice_pf_to_dev(vsi->back), &info, + list); +} + +/** + * ice_fltr_add_vlan_to_list - add VLAN filter info to exsisting list + * @vsi: pointer to VSI struct + * @list: list to add filter info to + * @vlan_id: VLAN ID to add + * @action: filter action + */ +static int +ice_fltr_add_vlan_to_list(struct ice_vsi *vsi, struct list_head *list, + u16 vlan_id, enum ice_sw_fwd_act_type action) +{ + struct ice_fltr_info info = { 0 }; + + info.flag = ICE_FLTR_TX; + info.src_id = ICE_SRC_ID_VSI; + info.lkup_type = ICE_SW_LKUP_VLAN; + info.fltr_act = action; + info.vsi_handle = vsi->idx; + info.l_data.vlan.vlan_id = vlan_id; + + return ice_fltr_add_entry_to_list(ice_pf_to_dev(vsi->back), &info, + list); +} + +/** + * ice_fltr_add_eth_to_list - add ethertype filter info to exsisting list + * @vsi: pointer to VSI struct + * @list: list to add filter info to + * @ethertype: ethertype of packet that matches filter + * @flag: filter direction, Tx or Rx + * @action: filter action + */ +static int +ice_fltr_add_eth_to_list(struct ice_vsi *vsi, struct list_head *list, + u16 ethertype, u16 flag, + enum ice_sw_fwd_act_type action) +{ + struct ice_fltr_info info = { 0 }; + + info.flag = flag; + info.lkup_type = ICE_SW_LKUP_ETHERTYPE; + info.fltr_act = action; + info.vsi_handle = vsi->idx; + info.l_data.ethertype_mac.ethertype = ethertype; + + if (flag == ICE_FLTR_TX) + info.src_id = ICE_SRC_ID_VSI; + else + info.src_id = ICE_SRC_ID_LPORT; + + return ice_fltr_add_entry_to_list(ice_pf_to_dev(vsi->back), &info, + list); +} + +/** + * ice_fltr_prepare_mac - add or remove MAC rule + * @vsi: pointer to VSI struct + * @mac: MAC address to add + * @action: action to be performed on filter match + * @mac_action: pointer to add or remove MAC function + */ +static enum ice_status +ice_fltr_prepare_mac(struct ice_vsi *vsi, const u8 *mac, + enum ice_sw_fwd_act_type action, + enum ice_status (*mac_action)(struct ice_vsi *, + struct list_head *)) +{ + enum ice_status result; + LIST_HEAD(tmp_list); + + if (ice_fltr_add_mac_to_list(vsi, &tmp_list, mac, action)) { + ice_fltr_free_list(ice_pf_to_dev(vsi->back), &tmp_list); + return ICE_ERR_NO_MEMORY; + } + + result = mac_action(vsi, &tmp_list); + ice_fltr_free_list(ice_pf_to_dev(vsi->back), &tmp_list); + return result; +} + +/** + * ice_fltr_prepare_mac_and_broadcast - add or remove MAC and broadcast filter + * @vsi: pointer to VSI struct + * @mac: MAC address to add + * @action: action to be performed on filter match + * @mac_action: pointer to add or remove MAC function + */ +static enum ice_status +ice_fltr_prepare_mac_and_broadcast(struct ice_vsi *vsi, const u8 *mac, + enum ice_sw_fwd_act_type action, + enum ice_status(*mac_action) + (struct ice_vsi *, struct list_head *)) +{ + u8 broadcast[ETH_ALEN]; + enum ice_status result; + LIST_HEAD(tmp_list); + + eth_broadcast_addr(broadcast); + if (ice_fltr_add_mac_to_list(vsi, &tmp_list, mac, action) || + ice_fltr_add_mac_to_list(vsi, &tmp_list, broadcast, action)) { + ice_fltr_free_list(ice_pf_to_dev(vsi->back), &tmp_list); + return ICE_ERR_NO_MEMORY; + } + + result = mac_action(vsi, &tmp_list); + ice_fltr_free_list(ice_pf_to_dev(vsi->back), &tmp_list); + return result; +} + +/** + * ice_fltr_prepare_vlan - add or remove VLAN filter + * @vsi: pointer to VSI struct + * @vlan_id: VLAN ID to add + * @action: action to be performed on filter match + * @vlan_action: pointer to add or remove VLAN function + */ +static enum ice_status +ice_fltr_prepare_vlan(struct ice_vsi *vsi, u16 vlan_id, + enum ice_sw_fwd_act_type action, + enum ice_status (*vlan_action)(struct ice_vsi *, + struct list_head *)) +{ + enum ice_status result; + LIST_HEAD(tmp_list); + + if (ice_fltr_add_vlan_to_list(vsi, &tmp_list, vlan_id, action)) + return ICE_ERR_NO_MEMORY; + + result = vlan_action(vsi, &tmp_list); + ice_fltr_free_list(ice_pf_to_dev(vsi->back), &tmp_list); + return result; +} + +/** + * ice_fltr_prepare_eth - add or remove ethertype filter + * @vsi: pointer to VSI struct + * @ethertype: ethertype of packet to be filtered + * @flag: direction of packet, Tx or Rx + * @action: action to be performed on filter match + * @eth_action: pointer to add or remove ethertype function + */ +static enum ice_status +ice_fltr_prepare_eth(struct ice_vsi *vsi, u16 ethertype, u16 flag, + enum ice_sw_fwd_act_type action, + enum ice_status (*eth_action)(struct ice_vsi *, + struct list_head *)) +{ + enum ice_status result; + LIST_HEAD(tmp_list); + + if (ice_fltr_add_eth_to_list(vsi, &tmp_list, ethertype, flag, action)) + return ICE_ERR_NO_MEMORY; + + result = eth_action(vsi, &tmp_list); + ice_fltr_free_list(ice_pf_to_dev(vsi->back), &tmp_list); + return result; +} + +/** + * ice_fltr_add_mac - add single MAC filter + * @vsi: pointer to VSI struct + * @mac: MAC to add + * @action: action to be performed on filter match + */ +enum ice_status ice_fltr_add_mac(struct ice_vsi *vsi, const u8 *mac, + enum ice_sw_fwd_act_type action) +{ + return ice_fltr_prepare_mac(vsi, mac, action, ice_fltr_add_mac_list); +} + +/** + * ice_fltr_add_mac_and_broadcast - add single MAC and broadcast + * @vsi: pointer to VSI struct + * @mac: MAC to add + * @action: action to be performed on filter match + */ +enum ice_status +ice_fltr_add_mac_and_broadcast(struct ice_vsi *vsi, const u8 *mac, + enum ice_sw_fwd_act_type action) +{ + return ice_fltr_prepare_mac_and_broadcast(vsi, mac, action, + ice_fltr_add_mac_list); +} + +/** + * ice_fltr_remove_mac - remove MAC filter + * @vsi: pointer to VSI struct + * @mac: filter MAC to remove + * @action: action to remove + */ +enum ice_status ice_fltr_remove_mac(struct ice_vsi *vsi, const u8 *mac, + enum ice_sw_fwd_act_type action) +{ + return ice_fltr_prepare_mac(vsi, mac, action, ice_fltr_remove_mac_list); +} + +/** + * ice_fltr_add_vlan - add single VLAN filter + * @vsi: pointer to VSI struct + * @vlan_id: VLAN ID to add + * @action: action to be performed on filter match + */ +enum ice_status ice_fltr_add_vlan(struct ice_vsi *vsi, u16 vlan_id, + enum ice_sw_fwd_act_type action) +{ + return ice_fltr_prepare_vlan(vsi, vlan_id, action, + ice_fltr_add_vlan_list); +} + +/** + * ice_fltr_remove_vlan - remove VLAN filter + * @vsi: pointer to VSI struct + * @vlan_id: filter VLAN to remove + * @action: action to remove + */ +enum ice_status ice_fltr_remove_vlan(struct ice_vsi *vsi, u16 vlan_id, + enum ice_sw_fwd_act_type action) +{ + return ice_fltr_prepare_vlan(vsi, vlan_id, action, + ice_fltr_remove_vlan_list); +} + +/** + * ice_fltr_add_eth - add specyfic ethertype filter + * @vsi: pointer to VSI struct + * @ethertype: ethertype of filter + * @flag: direction of packet to be filtered, Tx or Rx + * @action: action to be performed on filter match + */ +enum ice_status ice_fltr_add_eth(struct ice_vsi *vsi, u16 ethertype, u16 flag, + enum ice_sw_fwd_act_type action) +{ + return ice_fltr_prepare_eth(vsi, ethertype, flag, action, + ice_fltr_add_eth_list); +} + +/** + * ice_fltr_remove_eth - remove ethertype filter + * @vsi: pointer to VSI struct + * @ethertype: ethertype of filter + * @flag: direction of filter + * @action: action to remove + */ +enum ice_status ice_fltr_remove_eth(struct ice_vsi *vsi, u16 ethertype, + u16 flag, enum ice_sw_fwd_act_type action) +{ + return ice_fltr_prepare_eth(vsi, ethertype, flag, action, + ice_fltr_remove_eth_list); +} diff --git a/drivers/net/ethernet/intel/ice/ice_fltr.h b/drivers/net/ethernet/intel/ice/ice_fltr.h new file mode 100644 index 000000000000..361cb4da9b43 --- /dev/null +++ b/drivers/net/ethernet/intel/ice/ice_fltr.h @@ -0,0 +1,39 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (C) 2018-2020, Intel Corporation. */ + +#ifndef _ICE_FLTR_H_ +#define _ICE_FLTR_H_ + +void ice_fltr_free_list(struct device *dev, struct list_head *h); +enum ice_status +ice_fltr_add_mac_to_list(struct ice_vsi *vsi, struct list_head *list, + const u8 *mac, enum ice_sw_fwd_act_type action); +enum ice_status +ice_fltr_add_mac(struct ice_vsi *vsi, const u8 *mac, + enum ice_sw_fwd_act_type action); +enum ice_status +ice_fltr_add_mac_and_broadcast(struct ice_vsi *vsi, const u8 *mac, + enum ice_sw_fwd_act_type action); +enum ice_status +ice_fltr_add_mac_list(struct ice_vsi *vsi, struct list_head *list); +enum ice_status +ice_fltr_remove_mac(struct ice_vsi *vsi, const u8 *mac, + enum ice_sw_fwd_act_type action); +enum ice_status +ice_fltr_remove_mac_list(struct ice_vsi *vsi, struct list_head *list); + +enum ice_status +ice_fltr_add_vlan(struct ice_vsi *vsi, u16 vid, + enum ice_sw_fwd_act_type action); +enum ice_status +ice_fltr_remove_vlan(struct ice_vsi *vsi, u16 vid, + enum ice_sw_fwd_act_type action); + +enum ice_status +ice_fltr_add_eth(struct ice_vsi *vsi, u16 ethertype, u16 flag, + enum ice_sw_fwd_act_type action); +enum ice_status +ice_fltr_remove_eth(struct ice_vsi *vsi, u16 ethertype, u16 flag, + enum ice_sw_fwd_act_type action); +void ice_fltr_remove_all(struct ice_vsi *vsi); +#endif diff --git a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h index 1d37a9f02c1c..1086c9f778b4 100644 --- a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h +++ b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h @@ -6,9 +6,6 @@ #ifndef _ICE_HW_AUTOGEN_H_ #define _ICE_HW_AUTOGEN_H_ -#define PF0INT_ITR_0(_i) (0x03000004 + ((_i) * 4096)) -#define PF0INT_ITR_1(_i) (0x03000008 + ((_i) * 4096)) -#define PF0INT_ITR_2(_i) (0x0300000C + ((_i) * 4096)) #define QTX_COMM_DBELL(_DBQM) (0x002C0000 + ((_DBQM) * 4)) #define QTX_COMM_HEAD(_DBQM) (0x000E0000 + ((_DBQM) * 4)) #define QTX_COMM_HEAD_HEAD_S 0 @@ -42,6 +39,7 @@ #define PF_MBX_ARQH_ARQH_M ICE_M(0x3FF, 0) #define PF_MBX_ARQLEN 0x0022E480 #define PF_MBX_ARQLEN_ARQLEN_M ICE_M(0x3FF, 0) +#define PF_MBX_ARQLEN_ARQCRIT_M BIT(30) #define PF_MBX_ARQLEN_ARQENABLE_M BIT(31) #define PF_MBX_ARQT 0x0022E580 #define PF_MBX_ATQBAH 0x0022E180 @@ -50,6 +48,7 @@ #define PF_MBX_ATQH_ATQH_M ICE_M(0x3FF, 0) #define PF_MBX_ATQLEN 0x0022E200 #define PF_MBX_ATQLEN_ATQLEN_M ICE_M(0x3FF, 0) +#define PF_MBX_ATQLEN_ATQCRIT_M BIT(30) #define PF_MBX_ATQLEN_ATQENABLE_M BIT(31) #define PF_MBX_ATQT 0x0022E300 #define PRTDCB_GENC 0x00083000 @@ -58,6 +57,7 @@ #define PRTDCB_GENS 0x00083020 #define PRTDCB_GENS_DCBX_STATUS_S 0 #define PRTDCB_GENS_DCBX_STATUS_M ICE_M(0x7, 0) +#define PRTDCB_TUP2TC 0x001D26C0 /* Reset Source: CORER */ #define GL_PREEXT_L2_PMASK0(_i) (0x0020F0FC + ((_i) * 4)) #define GL_PREEXT_L2_PMASK1(_i) (0x0020F108 + ((_i) * 4)) #define GLFLXP_RXDID_FLX_WRD_0(_i) (0x0045c800 + ((_i) * 4)) @@ -218,6 +218,11 @@ #define VPLAN_TX_QBASE_VFNUMQ_M ICE_M(0xFF, 16) #define VPLAN_TXQ_MAPENA(_VF) (0x00073800 + ((_VF) * 4)) #define VPLAN_TXQ_MAPENA_TX_ENA_M BIT(0) +#define PRTMAC_HSEC_CTL_TX_PAUSE_QUANTA(_i) (0x001E36E0 + ((_i) * 32)) +#define PRTMAC_HSEC_CTL_TX_PAUSE_QUANTA_MAX_INDEX 8 +#define PRTMAC_HSEC_CTL_TX_PAUSE_QUANTA_HSEC_CTL_TX_PAUSE_QUANTA_M ICE_M(0xFFFF, 0) +#define PRTMAC_HSEC_CTL_TX_PAUSE_REFRESH_TIMER(_i) (0x001E3800 + ((_i) * 32)) +#define PRTMAC_HSEC_CTL_TX_PAUSE_REFRESH_TIMER_M ICE_M(0xFFFF, 0) #define GL_MDCK_TX_TDPU 0x00049348 #define GL_MDCK_TX_TDPU_RCU_ANTISPOOF_ITR_DIS_M BIT(1) #define GL_MDET_RX 0x00294C00 @@ -289,6 +294,20 @@ #define GL_PWR_MODE_CTL 0x000B820C #define GL_PWR_MODE_CTL_CAR_MAX_BW_S 30 #define GL_PWR_MODE_CTL_CAR_MAX_BW_M ICE_M(0x3, 30) +#define GLQF_FD_CNT 0x00460018 +#define GLQF_FD_CNT_FD_BCNT_S 16 +#define GLQF_FD_CNT_FD_BCNT_M ICE_M(0x7FFF, 16) +#define GLQF_FD_SIZE 0x00460010 +#define GLQF_FD_SIZE_FD_GSIZE_S 0 +#define GLQF_FD_SIZE_FD_GSIZE_M ICE_M(0x7FFF, 0) +#define GLQF_FD_SIZE_FD_BSIZE_S 16 +#define GLQF_FD_SIZE_FD_BSIZE_M ICE_M(0x7FFF, 16) +#define GLQF_FDINSET(_i, _j) (0x00412000 + ((_i) * 4 + (_j) * 512)) +#define GLQF_FDMASK_SEL(_i) (0x00410400 + ((_i) * 4)) +#define GLQF_FDSWAP(_i, _j) (0x00413000 + ((_i) * 4 + (_j) * 512)) +#define PFQF_FD_ENA 0x0043A000 +#define PFQF_FD_ENA_FD_ENA_M BIT(0) +#define PFQF_FD_SIZE 0x00460100 #define GLDCB_RTCTQ_RXQNUM_S 0 #define GLDCB_RTCTQ_RXQNUM_M ICE_M(0x7FF, 0) #define GLPRT_BPRCL(_i) (0x00381380 + ((_i) * 8)) @@ -332,6 +351,7 @@ #define GLPRT_TDOLD(_i) (0x00381280 + ((_i) * 8)) #define GLPRT_UPRCL(_i) (0x00381300 + ((_i) * 8)) #define GLPRT_UPTCL(_i) (0x003811C0 + ((_i) * 8)) +#define GLSTAT_FD_CNT0L(_i) (0x003A0000 + ((_i) * 8)) #define GLV_BPRCL(_i) (0x003B6000 + ((_i) * 8)) #define GLV_BPTCL(_i) (0x0030E000 + ((_i) * 8)) #define GLV_GORCL(_i) (0x003B0000 + ((_i) * 8)) @@ -342,6 +362,9 @@ #define GLV_TEPC(_VSI) (0x00312000 + ((_VSI) * 4)) #define GLV_UPRCL(_i) (0x003B2000 + ((_i) * 8)) #define GLV_UPTCL(_i) (0x0030A000 + ((_i) * 8)) +#define VSIQF_FD_CNT(_VSI) (0x00464000 + ((_VSI) * 4)) +#define VSIQF_FD_CNT_FD_GCNT_S 0 +#define VSIQF_FD_CNT_FD_GCNT_M ICE_M(0x3FFF, 0) #define VSIQF_HKEY_MAX_INDEX 12 #define VSIQF_HLUT_MAX_INDEX 15 #define VFINT_DYN_CTLN(_i) (0x00003800 + ((_i) * 4)) diff --git a/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h b/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h index 878e125d8b42..14dfbbc1b2cf 100644 --- a/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h +++ b/drivers/net/ethernet/intel/ice/ice_lan_tx_rx.h @@ -40,6 +40,104 @@ union ice_32byte_rx_desc { } wb; /* writeback */ }; +struct ice_fltr_desc { + __le64 qidx_compq_space_stat; + __le64 dtype_cmd_vsi_fdid; +}; + +#define ICE_FXD_FLTR_QW0_QINDEX_S 0 +#define ICE_FXD_FLTR_QW0_QINDEX_M (0x7FFULL << ICE_FXD_FLTR_QW0_QINDEX_S) +#define ICE_FXD_FLTR_QW0_COMP_Q_S 11 +#define ICE_FXD_FLTR_QW0_COMP_Q_M BIT_ULL(ICE_FXD_FLTR_QW0_COMP_Q_S) +#define ICE_FXD_FLTR_QW0_COMP_Q_ZERO 0x0ULL + +#define ICE_FXD_FLTR_QW0_COMP_REPORT_S 12 +#define ICE_FXD_FLTR_QW0_COMP_REPORT_M \ + (0x3ULL << ICE_FXD_FLTR_QW0_COMP_REPORT_S) +#define ICE_FXD_FLTR_QW0_COMP_REPORT_SW_FAIL 0x1ULL + +#define ICE_FXD_FLTR_QW0_FD_SPACE_S 14 +#define ICE_FXD_FLTR_QW0_FD_SPACE_M (0x3ULL << ICE_FXD_FLTR_QW0_FD_SPACE_S) +#define ICE_FXD_FLTR_QW0_FD_SPACE_GUAR_BEST 0x2ULL + +#define ICE_FXD_FLTR_QW0_STAT_CNT_S 16 +#define ICE_FXD_FLTR_QW0_STAT_CNT_M \ + (0x1FFFULL << ICE_FXD_FLTR_QW0_STAT_CNT_S) +#define ICE_FXD_FLTR_QW0_STAT_ENA_S 29 +#define ICE_FXD_FLTR_QW0_STAT_ENA_M (0x3ULL << ICE_FXD_FLTR_QW0_STAT_ENA_S) +#define ICE_FXD_FLTR_QW0_STAT_ENA_PKTS 0x1ULL + +#define ICE_FXD_FLTR_QW0_EVICT_ENA_S 31 +#define ICE_FXD_FLTR_QW0_EVICT_ENA_M BIT_ULL(ICE_FXD_FLTR_QW0_EVICT_ENA_S) +#define ICE_FXD_FLTR_QW0_EVICT_ENA_FALSE 0x0ULL +#define ICE_FXD_FLTR_QW0_EVICT_ENA_TRUE 0x1ULL + +#define ICE_FXD_FLTR_QW0_TO_Q_S 32 +#define ICE_FXD_FLTR_QW0_TO_Q_M (0x7ULL << ICE_FXD_FLTR_QW0_TO_Q_S) +#define ICE_FXD_FLTR_QW0_TO_Q_EQUALS_QINDEX 0x0ULL + +#define ICE_FXD_FLTR_QW0_TO_Q_PRI_S 35 +#define ICE_FXD_FLTR_QW0_TO_Q_PRI_M (0x7ULL << ICE_FXD_FLTR_QW0_TO_Q_PRI_S) +#define ICE_FXD_FLTR_QW0_TO_Q_PRIO1 0x1ULL + +#define ICE_FXD_FLTR_QW0_DPU_RECIPE_S 38 +#define ICE_FXD_FLTR_QW0_DPU_RECIPE_M \ + (0x3ULL << ICE_FXD_FLTR_QW0_DPU_RECIPE_S) +#define ICE_FXD_FLTR_QW0_DPU_RECIPE_DFLT 0x0ULL + +#define ICE_FXD_FLTR_QW0_DROP_S 40 +#define ICE_FXD_FLTR_QW0_DROP_M BIT_ULL(ICE_FXD_FLTR_QW0_DROP_S) +#define ICE_FXD_FLTR_QW0_DROP_NO 0x0ULL +#define ICE_FXD_FLTR_QW0_DROP_YES 0x1ULL + +#define ICE_FXD_FLTR_QW0_FLEX_PRI_S 41 +#define ICE_FXD_FLTR_QW0_FLEX_PRI_M (0x7ULL << ICE_FXD_FLTR_QW0_FLEX_PRI_S) +#define ICE_FXD_FLTR_QW0_FLEX_PRI_NONE 0x0ULL + +#define ICE_FXD_FLTR_QW0_FLEX_MDID_S 44 +#define ICE_FXD_FLTR_QW0_FLEX_MDID_M (0xFULL << ICE_FXD_FLTR_QW0_FLEX_MDID_S) +#define ICE_FXD_FLTR_QW0_FLEX_MDID0 0x0ULL + +#define ICE_FXD_FLTR_QW0_FLEX_VAL_S 48 +#define ICE_FXD_FLTR_QW0_FLEX_VAL_M \ + (0xFFFFULL << ICE_FXD_FLTR_QW0_FLEX_VAL_S) +#define ICE_FXD_FLTR_QW0_FLEX_VAL0 0x0ULL + +#define ICE_FXD_FLTR_QW1_DTYPE_S 0 +#define ICE_FXD_FLTR_QW1_DTYPE_M (0xFULL << ICE_FXD_FLTR_QW1_DTYPE_S) +#define ICE_FXD_FLTR_QW1_PCMD_S 4 +#define ICE_FXD_FLTR_QW1_PCMD_M BIT_ULL(ICE_FXD_FLTR_QW1_PCMD_S) +#define ICE_FXD_FLTR_QW1_PCMD_ADD 0x0ULL +#define ICE_FXD_FLTR_QW1_PCMD_REMOVE 0x1ULL + +#define ICE_FXD_FLTR_QW1_PROF_PRI_S 5 +#define ICE_FXD_FLTR_QW1_PROF_PRI_M (0x7ULL << ICE_FXD_FLTR_QW1_PROF_PRI_S) +#define ICE_FXD_FLTR_QW1_PROF_PRIO_ZERO 0x0ULL + +#define ICE_FXD_FLTR_QW1_PROF_S 8 +#define ICE_FXD_FLTR_QW1_PROF_M (0x3FULL << ICE_FXD_FLTR_QW1_PROF_S) +#define ICE_FXD_FLTR_QW1_PROF_ZERO 0x0ULL + +#define ICE_FXD_FLTR_QW1_FD_VSI_S 14 +#define ICE_FXD_FLTR_QW1_FD_VSI_M (0x3FFULL << ICE_FXD_FLTR_QW1_FD_VSI_S) +#define ICE_FXD_FLTR_QW1_SWAP_S 24 +#define ICE_FXD_FLTR_QW1_SWAP_M BIT_ULL(ICE_FXD_FLTR_QW1_SWAP_S) +#define ICE_FXD_FLTR_QW1_SWAP_NOT_SET 0x0ULL +#define ICE_FXD_FLTR_QW1_SWAP_SET 0x1ULL + +#define ICE_FXD_FLTR_QW1_FDID_PRI_S 25 +#define ICE_FXD_FLTR_QW1_FDID_PRI_M (0x7ULL << ICE_FXD_FLTR_QW1_FDID_PRI_S) +#define ICE_FXD_FLTR_QW1_FDID_PRI_ONE 0x1ULL + +#define ICE_FXD_FLTR_QW1_FDID_MDID_S 28 +#define ICE_FXD_FLTR_QW1_FDID_MDID_M (0xFULL << ICE_FXD_FLTR_QW1_FDID_MDID_S) +#define ICE_FXD_FLTR_QW1_FDID_MDID_FD 0x05ULL + +#define ICE_FXD_FLTR_QW1_FDID_S 32 +#define ICE_FXD_FLTR_QW1_FDID_M \ + (0xFFFFFFFFULL << ICE_FXD_FLTR_QW1_FDID_S) +#define ICE_FXD_FLTR_QW1_FDID_ZERO 0x0ULL + struct ice_rx_ptype_decoded { u32 ptype:10; u32 known:1; @@ -262,6 +360,12 @@ enum ice_rx_flex_desc_status_error_0_bits { ICE_RX_FLEX_DESC_STATUS0_LAST /* this entry must be last!!! */ }; +enum ice_rx_flex_desc_status_error_1_bits { + /* Note: These are predefined bit offsets */ + ICE_RX_FLEX_DESC_STATUS1_NAT_S = 4, + ICE_RX_FLEX_DESC_STATUS1_LAST /* this entry must be last!!! */ +}; + #define ICE_RXQ_CTX_SIZE_DWORDS 8 #define ICE_RXQ_CTX_SZ (ICE_RXQ_CTX_SIZE_DWORDS * sizeof(u32)) #define ICE_TX_CMPLTNQ_CTX_SIZE_DWORDS 22 @@ -340,6 +444,7 @@ struct ice_tx_desc { enum ice_tx_desc_dtype_value { ICE_TX_DESC_DTYPE_DATA = 0x0, ICE_TX_DESC_DTYPE_CTX = 0x1, + ICE_TX_DESC_DTYPE_FLTR_PROG = 0x8, /* DESC_DONE - HW has completed write-back of descriptor */ ICE_TX_DESC_DTYPE_DESC_DONE = 0xF, }; @@ -351,12 +456,14 @@ enum ice_tx_desc_cmd_bits { ICE_TX_DESC_CMD_EOP = 0x0001, ICE_TX_DESC_CMD_RS = 0x0002, ICE_TX_DESC_CMD_IL2TAG1 = 0x0008, + ICE_TX_DESC_CMD_DUMMY = 0x0010, ICE_TX_DESC_CMD_IIPT_IPV6 = 0x0020, ICE_TX_DESC_CMD_IIPT_IPV4 = 0x0040, ICE_TX_DESC_CMD_IIPT_IPV4_CSUM = 0x0060, ICE_TX_DESC_CMD_L4T_EOFT_TCP = 0x0100, ICE_TX_DESC_CMD_L4T_EOFT_SCTP = 0x0200, ICE_TX_DESC_CMD_L4T_EOFT_UDP = 0x0300, + ICE_TX_DESC_CMD_RE = 0x0400, }; #define ICE_TXD_QW1_OFFSET_S 16 @@ -413,6 +520,25 @@ enum ice_tx_ctx_desc_cmd_bits { ICE_TX_CTX_DESC_RESERVED = 0x40 }; +enum ice_tx_ctx_desc_eipt_offload { + ICE_TX_CTX_EIPT_NONE = 0x0, + ICE_TX_CTX_EIPT_IPV6 = 0x1, + ICE_TX_CTX_EIPT_IPV4_NO_CSUM = 0x2, + ICE_TX_CTX_EIPT_IPV4 = 0x3 +}; + +#define ICE_TXD_CTX_QW0_EIPLEN_S 2 + +#define ICE_TXD_CTX_QW0_L4TUNT_S 9 + +#define ICE_TXD_CTX_UDP_TUNNELING BIT_ULL(ICE_TXD_CTX_QW0_L4TUNT_S) +#define ICE_TXD_CTX_GRE_TUNNELING (0x2ULL << ICE_TXD_CTX_QW0_L4TUNT_S) + +#define ICE_TXD_CTX_QW0_NATLEN_S 12 + +#define ICE_TXD_CTX_QW0_L4T_CS_S 23 +#define ICE_TXD_CTX_QW0_L4T_CS_M BIT_ULL(ICE_TXD_CTX_QW0_L4T_CS_S) + #define ICE_LAN_TXQ_MAX_QGRPS 127 #define ICE_LAN_TXQ_MAX_QDIS 1023 @@ -455,7 +581,7 @@ struct ice_tlan_ctx { u8 drop_ena; u8 cache_prof_idx; u8 pkt_shaper_prof_idx; - u8 int_q_state; /* width not needed - internal do not write */ + u8 int_q_state; /* width not needed - internal - DO NOT WRITE!!! */ }; /* macro to make the table lines short */ diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index 2f256bf45efc..28b46cc9f5cb 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -5,6 +5,7 @@ #include "ice_base.h" #include "ice_flow.h" #include "ice_lib.h" +#include "ice_fltr.h" #include "ice_dcb_lib.h" /** @@ -18,6 +19,8 @@ const char *ice_vsi_type_str(enum ice_vsi_type vsi_type) return "ICE_VSI_PF"; case ICE_VSI_VF: return "ICE_VSI_VF"; + case ICE_VSI_CTRL: + return "ICE_VSI_CTRL"; case ICE_VSI_LB: return "ICE_VSI_LB"; default: @@ -37,7 +40,8 @@ const char *ice_vsi_type_str(enum ice_vsi_type vsi_type) */ static int ice_vsi_ctrl_all_rx_rings(struct ice_vsi *vsi, bool ena) { - int i, ret = 0; + int ret = 0; + u16 i; for (i = 0; i < vsi->num_rxq; i++) ice_vsi_ctrl_one_rx_ring(vsi, ena, i, false); @@ -121,6 +125,7 @@ static void ice_vsi_set_num_desc(struct ice_vsi *vsi) { switch (vsi->type) { case ICE_VSI_PF: + case ICE_VSI_CTRL: case ICE_VSI_LB: vsi->num_rx_desc = ICE_DFLT_NUM_RX_DESC; vsi->num_tx_desc = ICE_DFLT_NUM_TX_DESC; @@ -185,6 +190,11 @@ static void ice_vsi_set_num_qs(struct ice_vsi *vsi, u16 vf_id) */ vsi->num_q_vectors = pf->num_msix_per_vf - ICE_NONQ_VECS_VF; break; + case ICE_VSI_CTRL: + vsi->alloc_txq = 1; + vsi->alloc_rxq = 1; + vsi->num_q_vectors = 1; + break; case ICE_VSI_LB: vsi->alloc_txq = 1; vsi->alloc_rxq = 1; @@ -248,8 +258,8 @@ void ice_vsi_delete(struct ice_vsi *vsi) status = ice_free_vsi(&pf->hw, vsi->idx, ctxt, false, NULL); if (status) - dev_err(ice_pf_to_dev(pf), "Failed to delete VSI %i in FW - error: %d\n", - vsi->vsi_num, status); + dev_err(ice_pf_to_dev(pf), "Failed to delete VSI %i in FW - error: %s\n", + vsi->vsi_num, ice_stat_str(status)); kfree(ctxt); } @@ -320,7 +330,7 @@ int ice_vsi_clear(struct ice_vsi *vsi) /* updates the PF for this cleared VSI */ pf->vsi[vsi->idx] = NULL; - if (vsi->idx < pf->next_vsi) + if (vsi->idx < pf->next_vsi && vsi->type != ICE_VSI_CTRL) pf->next_vsi = vsi->idx; ice_vsi_free_arrays(vsi); @@ -331,6 +341,25 @@ int ice_vsi_clear(struct ice_vsi *vsi) } /** + * ice_msix_clean_ctrl_vsi - MSIX mode interrupt handler for ctrl VSI + * @irq: interrupt number + * @data: pointer to a q_vector + */ +static irqreturn_t ice_msix_clean_ctrl_vsi(int __always_unused irq, void *data) +{ + struct ice_q_vector *q_vector = (struct ice_q_vector *)data; + + if (!q_vector->tx.ring) + return IRQ_HANDLED; + +#define FDIR_RX_DESC_CLEAN_BUDGET 64 + ice_clean_rx_irq(q_vector->rx.ring, FDIR_RX_DESC_CLEAN_BUDGET); + ice_clean_ctrl_tx_irq(q_vector->tx.ring); + + return IRQ_HANDLED; +} + +/** * ice_msix_clean_rings - MSIX mode Interrupt Handler * @irq: interrupt number * @data: pointer to a q_vector @@ -381,8 +410,6 @@ ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type vsi_type, u16 vf_id) vsi->back = pf; set_bit(__ICE_DOWN, vsi->state); - vsi->idx = pf->next_vsi; - if (vsi_type == ICE_VSI_VF) ice_vsi_set_num_qs(vsi, vf_id); else @@ -396,6 +423,13 @@ ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type vsi_type, u16 vf_id) /* Setup default MSIX irq handler for VSI */ vsi->irq_handler = ice_msix_clean_rings; break; + case ICE_VSI_CTRL: + if (ice_vsi_alloc_arrays(vsi)) + goto err_rings; + + /* Setup ctrl VSI MSIX irq handler */ + vsi->irq_handler = ice_msix_clean_ctrl_vsi; + break; case ICE_VSI_VF: if (ice_vsi_alloc_arrays(vsi)) goto err_rings; @@ -409,12 +443,20 @@ ice_vsi_alloc(struct ice_pf *pf, enum ice_vsi_type vsi_type, u16 vf_id) goto unlock_pf; } - /* fill VSI slot in the PF struct */ - pf->vsi[pf->next_vsi] = vsi; + if (vsi->type == ICE_VSI_CTRL) { + /* Use the last VSI slot as the index for the control VSI */ + vsi->idx = pf->num_alloc_vsi - 1; + pf->ctrl_vsi_idx = vsi->idx; + pf->vsi[vsi->idx] = vsi; + } else { + /* fill slot and make note of the index */ + vsi->idx = pf->next_vsi; + pf->vsi[pf->next_vsi] = vsi; - /* prepare pf->next_vsi for next use */ - pf->next_vsi = ice_get_free_slot(pf->vsi, pf->num_alloc_vsi, - pf->next_vsi); + /* prepare pf->next_vsi for next use */ + pf->next_vsi = ice_get_free_slot(pf->vsi, pf->num_alloc_vsi, + pf->next_vsi); + } goto unlock_pf; err_rings: @@ -426,6 +468,48 @@ unlock_pf: } /** + * ice_alloc_fd_res - Allocate FD resource for a VSI + * @vsi: pointer to the ice_vsi + * + * This allocates the FD resources + * + * Returns 0 on success, -EPERM on no-op or -EIO on failure + */ +static int ice_alloc_fd_res(struct ice_vsi *vsi) +{ + struct ice_pf *pf = vsi->back; + u32 g_val, b_val; + + /* Flow Director filters are only allocated/assigned to the PF VSI which + * passes the traffic. The CTRL VSI is only used to add/delete filters + * so we don't allocate resources to it + */ + + /* FD filters from guaranteed pool per VSI */ + g_val = pf->hw.func_caps.fd_fltr_guar; + if (!g_val) + return -EPERM; + + /* FD filters from best effort pool */ + b_val = pf->hw.func_caps.fd_fltr_best_effort; + if (!b_val) + return -EPERM; + + if (vsi->type != ICE_VSI_PF) + return -EPERM; + + if (!test_bit(ICE_FLAG_FD_ENA, pf->flags)) + return -EPERM; + + vsi->num_gfltr = g_val / pf->num_alloc_vsi; + + /* each VSI gets same "best_effort" quota */ + vsi->num_bfltr = b_val; + + return 0; +} + +/** * ice_vsi_get_qs - Assign queues from PF to VSI * @vsi: the VSI to assign queues to * @@ -521,8 +605,8 @@ static void ice_vsi_clean_rss_flow_fld(struct ice_vsi *vsi) status = ice_rem_vsi_rss_cfg(&pf->hw, vsi->idx); if (status) - dev_dbg(ice_pf_to_dev(pf), "ice_rem_vsi_rss_cfg failed for vsi = %d, error = %d\n", - vsi->vsi_num, status); + dev_dbg(ice_pf_to_dev(pf), "ice_rem_vsi_rss_cfg failed for vsi = %d, error = %s\n", + vsi->vsi_num, ice_stat_str(status)); } /** @@ -565,8 +649,8 @@ static void ice_vsi_set_rss_params(struct ice_vsi *vsi) switch (vsi->type) { case ICE_VSI_PF: /* PF VSI will inherit RSS instance of PF */ - vsi->rss_table_size = cap->rss_table_size; - vsi->rss_size = min_t(int, num_online_cpus(), + vsi->rss_table_size = (u16)cap->rss_table_size; + vsi->rss_size = min_t(u16, num_online_cpus(), BIT(cap->rss_table_entry_width)); vsi->rss_lut_type = ICE_AQC_GSET_RSS_LUT_TABLE_TYPE_PF; break; @@ -581,8 +665,8 @@ static void ice_vsi_set_rss_params(struct ice_vsi *vsi) case ICE_VSI_LB: break; default: - dev_warn(ice_pf_to_dev(pf), "Unknown VSI type %d\n", - vsi->type); + dev_dbg(ice_pf_to_dev(pf), "Unsupported VSI type %s\n", + ice_vsi_type_str(vsi->type)); break; } } @@ -684,15 +768,15 @@ static void ice_vsi_setup_q_map(struct ice_vsi *vsi, struct ice_vsi_ctx *ctxt) max_rss = ICE_MAX_LG_RSS_QS; else max_rss = ICE_MAX_RSS_QS_PER_VF; - qcount_rx = min_t(int, rx_numq_tc, max_rss); + qcount_rx = min_t(u16, rx_numq_tc, max_rss); if (!vsi->req_rxq) - qcount_rx = min_t(int, qcount_rx, + qcount_rx = min_t(u16, qcount_rx, vsi->rss_size); } } /* find the (rounded up) power-of-2 of qcount */ - pow = order_base_2(qcount_rx); + pow = (u16)order_base_2(qcount_rx); ice_for_each_traffic_class(i) { if (!(vsi->tc_cfg.ena_tc & BIT(i))) { @@ -752,6 +836,51 @@ static void ice_vsi_setup_q_map(struct ice_vsi *vsi, struct ice_vsi_ctx *ctxt) } /** + * ice_set_fd_vsi_ctx - Set FD VSI context before adding a VSI + * @ctxt: the VSI context being set + * @vsi: the VSI being configured + */ +static void ice_set_fd_vsi_ctx(struct ice_vsi_ctx *ctxt, struct ice_vsi *vsi) +{ + u8 dflt_q_group, dflt_q_prio; + u16 dflt_q, report_q, val; + + if (vsi->type != ICE_VSI_PF && vsi->type != ICE_VSI_CTRL) + return; + + val = ICE_AQ_VSI_PROP_FLOW_DIR_VALID; + ctxt->info.valid_sections |= cpu_to_le16(val); + dflt_q = 0; + dflt_q_group = 0; + report_q = 0; + dflt_q_prio = 0; + + /* enable flow director filtering/programming */ + val = ICE_AQ_VSI_FD_ENABLE | ICE_AQ_VSI_FD_PROG_ENABLE; + ctxt->info.fd_options = cpu_to_le16(val); + /* max of allocated flow director filters */ + ctxt->info.max_fd_fltr_dedicated = + cpu_to_le16(vsi->num_gfltr); + /* max of shared flow director filters any VSI may program */ + ctxt->info.max_fd_fltr_shared = + cpu_to_le16(vsi->num_bfltr); + /* default queue index within the VSI of the default FD */ + val = ((dflt_q << ICE_AQ_VSI_FD_DEF_Q_S) & + ICE_AQ_VSI_FD_DEF_Q_M); + /* target queue or queue group to the FD filter */ + val |= ((dflt_q_group << ICE_AQ_VSI_FD_DEF_GRP_S) & + ICE_AQ_VSI_FD_DEF_GRP_M); + ctxt->info.fd_def_q = cpu_to_le16(val); + /* queue index on which FD filter completion is reported */ + val = ((report_q << ICE_AQ_VSI_FD_REPORT_Q_S) & + ICE_AQ_VSI_FD_REPORT_Q_M); + /* priority of the default qindex action */ + val |= ((dflt_q_prio << ICE_AQ_VSI_FD_DEF_PRIORITY_S) & + ICE_AQ_VSI_FD_DEF_PRIORITY_M); + ctxt->info.fd_report_opt = cpu_to_le16(val); +} + +/** * ice_set_rss_vsi_ctx - Set RSS VSI context before adding a VSI * @ctxt: the VSI context being set * @vsi: the VSI being configured @@ -776,13 +905,10 @@ static void ice_set_rss_vsi_ctx(struct ice_vsi_ctx *ctxt, struct ice_vsi *vsi) lut_type = ICE_AQ_VSI_Q_OPT_RSS_LUT_VSI; hash_type = ICE_AQ_VSI_Q_OPT_RSS_TPLZ; break; - case ICE_VSI_LB: + default: dev_dbg(dev, "Unsupported VSI type %s\n", ice_vsi_type_str(vsi->type)); return; - default: - dev_warn(dev, "Unknown VSI type %d\n", vsi->type); - return; } ctxt->info.q_opt_rss = ((lut_type << ICE_AQ_VSI_Q_OPT_RSS_LUT_S) & @@ -812,8 +938,8 @@ static int ice_vsi_init(struct ice_vsi *vsi, bool init_vsi) if (!ctxt) return -ENOMEM; - ctxt->info = vsi->info; switch (vsi->type) { + case ICE_VSI_CTRL: case ICE_VSI_LB: case ICE_VSI_PF: ctxt->flags = ICE_AQ_VSI_TYPE_PF; @@ -829,12 +955,15 @@ static int ice_vsi_init(struct ice_vsi *vsi, bool init_vsi) } ice_set_dflt_vsi_ctx(ctxt); + if (test_bit(ICE_FLAG_FD_ENA, pf->flags)) + ice_set_fd_vsi_ctx(ctxt, vsi); /* if the switch is in VEB mode, allow VSI loopback */ if (vsi->vsw->bridge_mode == BRIDGE_MODE_VEB) ctxt->info.sw_flags |= ICE_AQ_VSI_SW_FLAG_ALLOW_LB; /* Set LUT type and HASH type if RSS is enabled */ - if (test_bit(ICE_FLAG_RSS_ENA, pf->flags)) { + if (test_bit(ICE_FLAG_RSS_ENA, pf->flags) && + vsi->type != ICE_VSI_CTRL) { ice_set_rss_vsi_ctx(ctxt, vsi); /* if updating VSI context, make sure to set valid_section: * to indicate which section of VSI context being updated @@ -941,7 +1070,7 @@ int ice_free_res(struct ice_res_tracker *res, u16 index, u16 id) */ static int ice_search_res(struct ice_res_tracker *res, u16 needed, u16 id) { - int start = 0, end = 0; + u16 start = 0, end = 0; if (needed > res->end) return -ENOMEM; @@ -1024,6 +1153,7 @@ static int ice_vsi_setup_vector_base(struct ice_vsi *vsi) struct ice_pf *pf = vsi->back; struct device *dev; u16 num_q_vectors; + int base; dev = ice_pf_to_dev(pf); /* SRIOV doesn't grab irq_tracker entries for each VSI */ @@ -1038,14 +1168,15 @@ static int ice_vsi_setup_vector_base(struct ice_vsi *vsi) num_q_vectors = vsi->num_q_vectors; /* reserve slots from OS requested IRQs */ - vsi->base_vector = ice_get_res(pf, pf->irq_tracker, num_q_vectors, - vsi->idx); - if (vsi->base_vector < 0) { + base = ice_get_res(pf, pf->irq_tracker, num_q_vectors, vsi->idx); + + if (base < 0) { dev_err(dev, "%d MSI-X interrupts available. %s %d failed to get %d MSI-X vectors\n", ice_get_free_res_count(pf->irq_tracker), ice_vsi_type_str(vsi->type), vsi->idx, num_q_vectors); return -ENOENT; } + vsi->base_vector = (u16)base; pf->num_avail_sw_msix -= num_q_vectors; return 0; @@ -1085,7 +1216,7 @@ static int ice_vsi_alloc_rings(struct ice_vsi *vsi) { struct ice_pf *pf = vsi->back; struct device *dev; - int i; + u16 i; dev = ice_pf_to_dev(pf); /* Allocate Tx rings */ @@ -1178,7 +1309,7 @@ static int ice_vsi_cfg_rss_lut_key(struct ice_vsi *vsi) u8 *lut; dev = ice_pf_to_dev(pf); - vsi->rss_size = min_t(int, vsi->rss_size, vsi->num_rxq); + vsi->rss_size = min_t(u16, vsi->rss_size, vsi->num_rxq); lut = kzalloc(vsi->rss_table_size, GFP_KERNEL); if (!lut) @@ -1193,7 +1324,8 @@ static int ice_vsi_cfg_rss_lut_key(struct ice_vsi *vsi) vsi->rss_table_size); if (status) { - dev_err(dev, "set_rss_lut failed, error %d\n", status); + dev_err(dev, "set_rss_lut failed, error %s\n", + ice_stat_str(status)); err = -EIO; goto ice_vsi_cfg_rss_exit; } @@ -1215,7 +1347,8 @@ static int ice_vsi_cfg_rss_lut_key(struct ice_vsi *vsi) status = ice_aq_set_rss_key(&pf->hw, vsi->idx, key); if (status) { - dev_err(dev, "set_rss_key failed, error %d\n", status); + dev_err(dev, "set_rss_key failed, error %s\n", + ice_stat_str(status)); err = -EIO; } @@ -1248,8 +1381,8 @@ static void ice_vsi_set_vf_rss_flow_fld(struct ice_vsi *vsi) status = ice_add_avf_rss_cfg(&pf->hw, vsi->idx, ICE_DEFAULT_RSS_HENA); if (status) - dev_dbg(dev, "ice_add_avf_rss_cfg failed for vsi = %d, error = %d\n", - vsi->vsi_num, status); + dev_dbg(dev, "ice_add_avf_rss_cfg failed for vsi = %d, error = %s\n", + vsi->vsi_num, ice_stat_str(status)); } /** @@ -1281,91 +1414,57 @@ static void ice_vsi_set_rss_flow_fld(struct ice_vsi *vsi) status = ice_add_rss_cfg(hw, vsi_handle, ICE_FLOW_HASH_IPV4, ICE_FLOW_SEG_HDR_IPV4); if (status) - dev_dbg(dev, "ice_add_rss_cfg failed for ipv4 flow, vsi = %d, error = %d\n", - vsi_num, status); + dev_dbg(dev, "ice_add_rss_cfg failed for ipv4 flow, vsi = %d, error = %s\n", + vsi_num, ice_stat_str(status)); /* configure RSS for IPv6 with input set IPv6 src/dst */ status = ice_add_rss_cfg(hw, vsi_handle, ICE_FLOW_HASH_IPV6, ICE_FLOW_SEG_HDR_IPV6); if (status) - dev_dbg(dev, "ice_add_rss_cfg failed for ipv6 flow, vsi = %d, error = %d\n", - vsi_num, status); + dev_dbg(dev, "ice_add_rss_cfg failed for ipv6 flow, vsi = %d, error = %s\n", + vsi_num, ice_stat_str(status)); /* configure RSS for tcp4 with input set IP src/dst, TCP src/dst */ status = ice_add_rss_cfg(hw, vsi_handle, ICE_HASH_TCP_IPV4, ICE_FLOW_SEG_HDR_TCP | ICE_FLOW_SEG_HDR_IPV4); if (status) - dev_dbg(dev, "ice_add_rss_cfg failed for tcp4 flow, vsi = %d, error = %d\n", - vsi_num, status); + dev_dbg(dev, "ice_add_rss_cfg failed for tcp4 flow, vsi = %d, error = %s\n", + vsi_num, ice_stat_str(status)); /* configure RSS for udp4 with input set IP src/dst, UDP src/dst */ status = ice_add_rss_cfg(hw, vsi_handle, ICE_HASH_UDP_IPV4, ICE_FLOW_SEG_HDR_UDP | ICE_FLOW_SEG_HDR_IPV4); if (status) - dev_dbg(dev, "ice_add_rss_cfg failed for udp4 flow, vsi = %d, error = %d\n", - vsi_num, status); + dev_dbg(dev, "ice_add_rss_cfg failed for udp4 flow, vsi = %d, error = %s\n", + vsi_num, ice_stat_str(status)); /* configure RSS for sctp4 with input set IP src/dst */ status = ice_add_rss_cfg(hw, vsi_handle, ICE_FLOW_HASH_IPV4, ICE_FLOW_SEG_HDR_SCTP | ICE_FLOW_SEG_HDR_IPV4); if (status) - dev_dbg(dev, "ice_add_rss_cfg failed for sctp4 flow, vsi = %d, error = %d\n", - vsi_num, status); + dev_dbg(dev, "ice_add_rss_cfg failed for sctp4 flow, vsi = %d, error = %s\n", + vsi_num, ice_stat_str(status)); /* configure RSS for tcp6 with input set IPv6 src/dst, TCP src/dst */ status = ice_add_rss_cfg(hw, vsi_handle, ICE_HASH_TCP_IPV6, ICE_FLOW_SEG_HDR_TCP | ICE_FLOW_SEG_HDR_IPV6); if (status) - dev_dbg(dev, "ice_add_rss_cfg failed for tcp6 flow, vsi = %d, error = %d\n", - vsi_num, status); + dev_dbg(dev, "ice_add_rss_cfg failed for tcp6 flow, vsi = %d, error = %s\n", + vsi_num, ice_stat_str(status)); /* configure RSS for udp6 with input set IPv6 src/dst, UDP src/dst */ status = ice_add_rss_cfg(hw, vsi_handle, ICE_HASH_UDP_IPV6, ICE_FLOW_SEG_HDR_UDP | ICE_FLOW_SEG_HDR_IPV6); if (status) - dev_dbg(dev, "ice_add_rss_cfg failed for udp6 flow, vsi = %d, error = %d\n", - vsi_num, status); + dev_dbg(dev, "ice_add_rss_cfg failed for udp6 flow, vsi = %d, error = %s\n", + vsi_num, ice_stat_str(status)); /* configure RSS for sctp6 with input set IPv6 src/dst */ status = ice_add_rss_cfg(hw, vsi_handle, ICE_FLOW_HASH_IPV6, ICE_FLOW_SEG_HDR_SCTP | ICE_FLOW_SEG_HDR_IPV6); if (status) - dev_dbg(dev, "ice_add_rss_cfg failed for sctp6 flow, vsi = %d, error = %d\n", - vsi_num, status); -} - -/** - * ice_add_mac_to_list - Add a MAC address filter entry to the list - * @vsi: the VSI to be forwarded to - * @add_list: pointer to the list which contains MAC filter entries - * @macaddr: the MAC address to be added. - * - * Adds MAC address filter entry to the temp list - * - * Returns 0 on success or ENOMEM on failure. - */ -int -ice_add_mac_to_list(struct ice_vsi *vsi, struct list_head *add_list, - const u8 *macaddr) -{ - struct ice_fltr_list_entry *tmp; - struct ice_pf *pf = vsi->back; - - tmp = devm_kzalloc(ice_pf_to_dev(pf), sizeof(*tmp), GFP_ATOMIC); - if (!tmp) - return -ENOMEM; - - tmp->fltr_info.flag = ICE_FLTR_TX; - tmp->fltr_info.src_id = ICE_SRC_ID_VSI; - tmp->fltr_info.lkup_type = ICE_SW_LKUP_MAC; - tmp->fltr_info.fltr_act = ICE_FWD_TO_VSI; - tmp->fltr_info.vsi_handle = vsi->idx; - ether_addr_copy(tmp->fltr_info.l_data.mac.mac_addr, macaddr); - - INIT_LIST_HEAD(&tmp->list_entry); - list_add(&tmp->list_entry, add_list); - - return 0; + dev_dbg(dev, "ice_add_rss_cfg failed for sctp6 flow, vsi = %d, error = %s\n", + vsi_num, ice_stat_str(status)); } /** @@ -1415,54 +1514,21 @@ void ice_update_eth_stats(struct ice_vsi *vsi) } /** - * ice_free_fltr_list - free filter lists helper - * @dev: pointer to the device struct - * @h: pointer to the list head to be freed - * - * Helper function to free filter lists previously created using - * ice_add_mac_to_list - */ -void ice_free_fltr_list(struct device *dev, struct list_head *h) -{ - struct ice_fltr_list_entry *e, *tmp; - - list_for_each_entry_safe(e, tmp, h, list_entry) { - list_del(&e->list_entry); - devm_kfree(dev, e); - } -} - -/** * ice_vsi_add_vlan - Add VSI membership for given VLAN * @vsi: the VSI being configured * @vid: VLAN ID to be added + * @action: filter action to be performed on match */ -int ice_vsi_add_vlan(struct ice_vsi *vsi, u16 vid) +int +ice_vsi_add_vlan(struct ice_vsi *vsi, u16 vid, enum ice_sw_fwd_act_type action) { - struct ice_fltr_list_entry *tmp; struct ice_pf *pf = vsi->back; - LIST_HEAD(tmp_add_list); - enum ice_status status; struct device *dev; int err = 0; dev = ice_pf_to_dev(pf); - tmp = devm_kzalloc(dev, sizeof(*tmp), GFP_KERNEL); - if (!tmp) - return -ENOMEM; - tmp->fltr_info.lkup_type = ICE_SW_LKUP_VLAN; - tmp->fltr_info.fltr_act = ICE_FWD_TO_VSI; - tmp->fltr_info.flag = ICE_FLTR_TX; - tmp->fltr_info.src_id = ICE_SRC_ID_VSI; - tmp->fltr_info.vsi_handle = vsi->idx; - tmp->fltr_info.l_data.vlan.vlan_id = vid; - - INIT_LIST_HEAD(&tmp->list_entry); - list_add(&tmp->list_entry, &tmp_add_list); - - status = ice_add_vlan(&pf->hw, &tmp_add_list); - if (!status) { + if (!ice_fltr_add_vlan(vsi, vid, action)) { vsi->num_vlan++; } else { err = -ENODEV; @@ -1470,7 +1536,6 @@ int ice_vsi_add_vlan(struct ice_vsi *vsi, u16 vid) vsi->vsi_num); } - ice_free_fltr_list(dev, &tmp_add_list); return err; } @@ -1483,41 +1548,25 @@ int ice_vsi_add_vlan(struct ice_vsi *vsi, u16 vid) */ int ice_vsi_kill_vlan(struct ice_vsi *vsi, u16 vid) { - struct ice_fltr_list_entry *list; struct ice_pf *pf = vsi->back; - LIST_HEAD(tmp_add_list); enum ice_status status; struct device *dev; int err = 0; dev = ice_pf_to_dev(pf); - list = devm_kzalloc(dev, sizeof(*list), GFP_KERNEL); - if (!list) - return -ENOMEM; - - list->fltr_info.lkup_type = ICE_SW_LKUP_VLAN; - list->fltr_info.vsi_handle = vsi->idx; - list->fltr_info.fltr_act = ICE_FWD_TO_VSI; - list->fltr_info.l_data.vlan.vlan_id = vid; - list->fltr_info.flag = ICE_FLTR_TX; - list->fltr_info.src_id = ICE_SRC_ID_VSI; - INIT_LIST_HEAD(&list->list_entry); - list_add(&list->list_entry, &tmp_add_list); - - status = ice_remove_vlan(&pf->hw, &tmp_add_list); + status = ice_fltr_remove_vlan(vsi, vid, ICE_FWD_TO_VSI); if (!status) { vsi->num_vlan--; } else if (status == ICE_ERR_DOES_NOT_EXIST) { - dev_dbg(dev, "Failed to remove VLAN %d on VSI %i, it does not exist, status: %d\n", - vid, vsi->vsi_num, status); + dev_dbg(dev, "Failed to remove VLAN %d on VSI %i, it does not exist, status: %s\n", + vid, vsi->vsi_num, ice_stat_str(status)); } else { - dev_err(dev, "Error removing VLAN %d on vsi %i error: %d\n", - vid, vsi->vsi_num, status); + dev_err(dev, "Error removing VLAN %d on vsi %i error: %s\n", + vid, vsi->vsi_num, ice_stat_str(status)); err = -EIO; } - ice_free_fltr_list(dev, &tmp_add_list); return err; } @@ -1547,6 +1596,32 @@ void ice_vsi_cfg_frame_size(struct ice_vsi *vsi) } /** + * ice_write_qrxflxp_cntxt - write/configure QRXFLXP_CNTXT register + * @hw: HW pointer + * @pf_q: index of the Rx queue in the PF's queue space + * @rxdid: flexible descriptor RXDID + * @prio: priority for the RXDID for this queue + */ +void +ice_write_qrxflxp_cntxt(struct ice_hw *hw, u16 pf_q, u32 rxdid, u32 prio) +{ + int regval = rd32(hw, QRXFLXP_CNTXT(pf_q)); + + /* clear any previous values */ + regval &= ~(QRXFLXP_CNTXT_RXDID_IDX_M | + QRXFLXP_CNTXT_RXDID_PRIO_M | + QRXFLXP_CNTXT_TS_M); + + regval |= (rxdid << QRXFLXP_CNTXT_RXDID_IDX_S) & + QRXFLXP_CNTXT_RXDID_IDX_M; + + regval |= (prio << QRXFLXP_CNTXT_RXDID_PRIO_S) & + QRXFLXP_CNTXT_RXDID_PRIO_M; + + wr32(hw, QRXFLXP_CNTXT(pf_q), regval); +} + +/** * ice_vsi_cfg_rxqs - Configure the VSI for Rx * @vsi: the VSI being configured * @@ -1671,7 +1746,7 @@ void ice_vsi_cfg_msix(struct ice_vsi *vsi) { struct ice_pf *pf = vsi->back; struct ice_hw *hw = &pf->hw; - u32 txq = 0, rxq = 0; + u16 txq = 0, rxq = 0; int i, q; for (i = 0; i < vsi->num_q_vectors; i++) { @@ -1737,8 +1812,9 @@ int ice_vsi_manage_vlan_insertion(struct ice_vsi *vsi) status = ice_update_vsi(hw, vsi->idx, ctxt, NULL); if (status) { - dev_err(ice_pf_to_dev(vsi->back), "update VSI for VLAN insert failed, err %d aq_err %d\n", - status, hw->adminq.sq_last_status); + dev_err(ice_pf_to_dev(vsi->back), "update VSI for VLAN insert failed, err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); ret = -EIO; goto out; } @@ -1761,6 +1837,12 @@ int ice_vsi_manage_vlan_stripping(struct ice_vsi *vsi, bool ena) enum ice_status status; int ret = 0; + /* do not allow modifying VLAN stripping when a port VLAN is configured + * on this VSI + */ + if (vsi->info.pvid) + return 0; + ctxt = kzalloc(sizeof(*ctxt), GFP_KERNEL); if (!ctxt) return -ENOMEM; @@ -1783,8 +1865,9 @@ int ice_vsi_manage_vlan_stripping(struct ice_vsi *vsi, bool ena) status = ice_update_vsi(hw, vsi->idx, ctxt, NULL); if (status) { - dev_err(ice_pf_to_dev(vsi->back), "update VSI for VLAN strip failed, ena = %d err %d aq_err %d\n", - ena, status, hw->adminq.sq_last_status); + dev_err(ice_pf_to_dev(vsi->back), "update VSI for VLAN strip failed, ena = %d err %s aq_err %s\n", + ena, ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); ret = -EIO; goto out; } @@ -1922,9 +2005,10 @@ int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena, bool vlan_promisc) status = ice_update_vsi(&pf->hw, vsi->idx, ctxt, NULL); if (status) { - netdev_err(vsi->netdev, "%sabling VLAN pruning on VSI handle: %d, VSI HW ID: %d failed, err = %d, aq_err = %d\n", - ena ? "En" : "Dis", vsi->idx, vsi->vsi_num, status, - pf->hw.adminq.sq_last_status); + netdev_err(vsi->netdev, "%sabling VLAN pruning on VSI handle: %d, VSI HW ID: %d failed, err = %s, aq_err = %s\n", + ena ? "En" : "Dis", vsi->idx, vsi->vsi_num, + ice_stat_str(status), + ice_aq_str(pf->hw.adminq.sq_last_status)); goto err_out; } @@ -1991,47 +2075,6 @@ clear_reg_idx: } /** - * ice_vsi_add_rem_eth_mac - Program VSI ethertype based filter with rule - * @vsi: the VSI being configured - * @add_rule: boolean value to add or remove ethertype filter rule - */ -static void -ice_vsi_add_rem_eth_mac(struct ice_vsi *vsi, bool add_rule) -{ - struct ice_fltr_list_entry *list; - struct ice_pf *pf = vsi->back; - LIST_HEAD(tmp_add_list); - enum ice_status status; - struct device *dev; - - dev = ice_pf_to_dev(pf); - list = devm_kzalloc(dev, sizeof(*list), GFP_KERNEL); - if (!list) - return; - - list->fltr_info.lkup_type = ICE_SW_LKUP_ETHERTYPE; - list->fltr_info.fltr_act = ICE_DROP_PACKET; - list->fltr_info.flag = ICE_FLTR_TX; - list->fltr_info.src_id = ICE_SRC_ID_VSI; - list->fltr_info.vsi_handle = vsi->idx; - list->fltr_info.l_data.ethertype_mac.ethertype = vsi->ethtype; - - INIT_LIST_HEAD(&list->list_entry); - list_add(&list->list_entry, &tmp_add_list); - - if (add_rule) - status = ice_add_eth_mac(&pf->hw, &tmp_add_list); - else - status = ice_remove_eth_mac(&pf->hw, &tmp_add_list); - - if (status) - dev_err(dev, "Failure Adding or Removing Ethertype on VSI %i error: %d\n", - vsi->vsi_num, status); - - ice_free_fltr_list(dev, &tmp_add_list); -} - -/** * ice_cfg_sw_lldp - Config switch rules for LLDP packet handling * @vsi: the VSI being configured * @tx: bool to determine Tx or Rx rule @@ -2039,45 +2082,25 @@ ice_vsi_add_rem_eth_mac(struct ice_vsi *vsi, bool add_rule) */ void ice_cfg_sw_lldp(struct ice_vsi *vsi, bool tx, bool create) { - struct ice_fltr_list_entry *list; + enum ice_status (*eth_fltr)(struct ice_vsi *v, u16 type, u16 flag, + enum ice_sw_fwd_act_type act); struct ice_pf *pf = vsi->back; - LIST_HEAD(tmp_add_list); enum ice_status status; struct device *dev; dev = ice_pf_to_dev(pf); - list = devm_kzalloc(dev, sizeof(*list), GFP_KERNEL); - if (!list) - return; + eth_fltr = create ? ice_fltr_add_eth : ice_fltr_remove_eth; - list->fltr_info.lkup_type = ICE_SW_LKUP_ETHERTYPE; - list->fltr_info.vsi_handle = vsi->idx; - list->fltr_info.l_data.ethertype_mac.ethertype = ETH_P_LLDP; - - if (tx) { - list->fltr_info.fltr_act = ICE_DROP_PACKET; - list->fltr_info.flag = ICE_FLTR_TX; - list->fltr_info.src_id = ICE_SRC_ID_VSI; - } else { - list->fltr_info.fltr_act = ICE_FWD_TO_VSI; - list->fltr_info.flag = ICE_FLTR_RX; - list->fltr_info.src_id = ICE_SRC_ID_LPORT; - } - - INIT_LIST_HEAD(&list->list_entry); - list_add(&list->list_entry, &tmp_add_list); - - if (create) - status = ice_add_eth_mac(&pf->hw, &tmp_add_list); + if (tx) + status = eth_fltr(vsi, ETH_P_LLDP, ICE_FLTR_TX, + ICE_DROP_PACKET); else - status = ice_remove_eth_mac(&pf->hw, &tmp_add_list); + status = eth_fltr(vsi, ETH_P_LLDP, ICE_FLTR_RX, ICE_FWD_TO_VSI); if (status) - dev_err(dev, "Fail %s %s LLDP rule on VSI %i error: %d\n", + dev_err(dev, "Fail %s %s LLDP rule on VSI %i error: %s\n", create ? "adding" : "removing", tx ? "TX" : "RX", - vsi->vsi_num, status); - - ice_free_fltr_list(dev, &tmp_add_list); + vsi->vsi_num, ice_stat_str(status)); } /** @@ -2122,10 +2145,12 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, if (vsi->type == ICE_VSI_VF) vsi->vf_id = vf_id; + ice_alloc_fd_res(vsi); + if (ice_vsi_get_qs(vsi)) { dev_err(dev, "Failed to allocate queues. vsi->idx = %d\n", vsi->idx); - goto unroll_get_qs; + goto unroll_vsi_alloc; } /* set RSS capabilities */ @@ -2140,6 +2165,7 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, goto unroll_get_qs; switch (vsi->type) { + case ICE_VSI_CTRL: case ICE_VSI_PF: ret = ice_vsi_alloc_q_vectors(vsi); if (ret) @@ -2164,20 +2190,23 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, * so this handles those cases (i.e. adding the PF to a bridge * without the 8021q module loaded). */ - ret = ice_vsi_add_vlan(vsi, 0); + ret = ice_vsi_add_vlan(vsi, 0, ICE_FWD_TO_VSI); if (ret) goto unroll_clear_rings; ice_vsi_map_rings_to_vectors(vsi); - /* Do not exit if configuring RSS had an issue, at least - * receive traffic on first queue. Hence no need to capture - * return value - */ - if (test_bit(ICE_FLAG_RSS_ENA, pf->flags)) { - ice_vsi_cfg_rss_lut_key(vsi); - ice_vsi_set_rss_flow_fld(vsi); - } + /* ICE_VSI_CTRL does not need RSS so skip RSS processing */ + if (vsi->type != ICE_VSI_CTRL) + /* Do not exit if configuring RSS had an issue, at + * least receive traffic on first queue. Hence no + * need to capture return value + */ + if (test_bit(ICE_FLAG_RSS_ENA, pf->flags)) { + ice_vsi_cfg_rss_lut_key(vsi); + ice_vsi_set_rss_flow_fld(vsi); + } + ice_init_arfs(vsi); break; case ICE_VSI_VF: /* VF driver will take care of creating netdev for this type and @@ -2223,8 +2252,8 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, status = ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc, max_txqs); if (status) { - dev_err(dev, "VSI %d failed lan queue config, error %d\n", - vsi->vsi_num, status); + dev_err(dev, "VSI %d failed lan queue config, error %s\n", + vsi->vsi_num, ice_stat_str(status)); goto unroll_vector_base; } @@ -2239,9 +2268,8 @@ ice_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, */ if (!ice_is_safe_mode(pf)) if (vsi->type == ICE_VSI_PF) { - ice_vsi_add_rem_eth_mac(vsi, true); - - /* Tx LLDP packets */ + ice_fltr_add_eth(vsi, ETH_P_PAUSE, ICE_FLTR_TX, + ICE_DROP_PACKET); ice_cfg_sw_lldp(vsi, true, true); } @@ -2259,6 +2287,7 @@ unroll_vsi_init: ice_vsi_delete(vsi); unroll_get_qs: ice_vsi_put_qs(vsi); +unroll_vsi_alloc: ice_vsi_clear(vsi); return NULL; @@ -2411,6 +2440,8 @@ int ice_ena_vsi(struct ice_vsi *vsi, bool locked) if (!locked) rtnl_unlock(); } + } else if (vsi->type == ICE_VSI_CTRL) { + err = ice_vsi_open_ctrl(vsi); } return err; @@ -2440,6 +2471,8 @@ void ice_dis_vsi(struct ice_vsi *vsi, bool locked) } else { ice_vsi_close(vsi); } + } else if (vsi->type == ICE_VSI_CTRL) { + ice_vsi_close(vsi); } } @@ -2558,7 +2591,8 @@ int ice_vsi_release(struct ice_vsi *vsi) if (!ice_is_safe_mode(pf)) { if (vsi->type == ICE_VSI_PF) { - ice_vsi_add_rem_eth_mac(vsi, false); + ice_fltr_remove_eth(vsi, ETH_P_PAUSE, ICE_FLTR_TX, + ICE_DROP_PACKET); ice_cfg_sw_lldp(vsi, true, false); /* The Rx rule will only exist to remove if the LLDP FW * engine is currently stopped @@ -2568,7 +2602,7 @@ int ice_vsi_release(struct ice_vsi *vsi) } } - ice_remove_vsi_fltr(&pf->hw, vsi->idx); + ice_fltr_remove_all(vsi); ice_rm_vsi_lan_cfg(vsi->port_info, vsi->idx); ice_vsi_delete(vsi); ice_vsi_free_q_vectors(vsi); @@ -2673,15 +2707,13 @@ ice_vsi_rebuild_set_coalesce(struct ice_vsi *vsi, ice_vsi_rebuild_update_coalesce(vsi->q_vectors[i], &coalesce[i]); - for (; i < vsi->num_q_vectors; i++) { - struct ice_coalesce_stored coalesce_dflt = { - .itr_tx = ICE_DFLT_TX_ITR, - .itr_rx = ICE_DFLT_RX_ITR, - .intrl = 0 - }; + /* number of q_vectors increased, so assume coalesce settings were + * changed globally (i.e. ethtool -C eth0 instead of per-queue) and use + * the previous settings from q_vector 0 for all of the new q_vectors + */ + for (; i < vsi->num_q_vectors; i++) ice_vsi_rebuild_update_coalesce(vsi->q_vectors[i], - &coalesce_dflt); - } + &coalesce[0]); } /** @@ -2746,6 +2778,8 @@ int ice_vsi_rebuild(struct ice_vsi *vsi, bool init_vsi) goto err_vsi; ice_vsi_get_qs(vsi); + + ice_alloc_fd_res(vsi); ice_vsi_set_tc_cfg(vsi); /* Initialize VSI struct elements and create VSI in FW */ @@ -2754,6 +2788,7 @@ int ice_vsi_rebuild(struct ice_vsi *vsi, bool init_vsi) goto err_vsi; switch (vsi->type) { + case ICE_VSI_CTRL: case ICE_VSI_PF: ret = ice_vsi_alloc_q_vectors(vsi); if (ret) @@ -2773,17 +2808,19 @@ int ice_vsi_rebuild(struct ice_vsi *vsi, bool init_vsi) ice_vsi_map_rings_to_vectors(vsi); if (ice_is_xdp_ena_vsi(vsi)) { - vsi->num_xdp_txq = vsi->alloc_txq; + vsi->num_xdp_txq = vsi->alloc_rxq; ret = ice_prepare_xdp_rings(vsi, vsi->xdp_prog); if (ret) goto err_vectors; } - /* Do not exit if configuring RSS had an issue, at least - * receive traffic on first queue. Hence no need to capture - * return value - */ - if (test_bit(ICE_FLAG_RSS_ENA, pf->flags)) - ice_vsi_cfg_rss_lut_key(vsi); + /* ICE_VSI_CTRL does not need RSS so skip RSS processing */ + if (vsi->type != ICE_VSI_CTRL) + /* Do not exit if configuring RSS had an issue, at + * least receive traffic on first queue. Hence no + * need to capture return value + */ + if (test_bit(ICE_FLAG_RSS_ENA, pf->flags)) + ice_vsi_cfg_rss_lut_key(vsi); break; case ICE_VSI_VF: ret = ice_vsi_alloc_q_vectors(vsi); @@ -2814,8 +2851,8 @@ int ice_vsi_rebuild(struct ice_vsi *vsi, bool init_vsi) status = ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc, max_txqs); if (status) { - dev_err(ice_pf_to_dev(pf), "VSI %d failed lan queue config, error %d\n", - vsi->vsi_num, status); + dev_err(ice_pf_to_dev(pf), "VSI %d failed lan queue config, error %s\n", + vsi->vsi_num, ice_stat_str(status)); if (init_vsi) { ret = -EIO; goto err_vectors; @@ -2924,8 +2961,8 @@ int ice_vsi_cfg_tc(struct ice_vsi *vsi, u8 ena_tc) max_txqs); if (status) { - dev_err(dev, "VSI %d failed TC config, error %d\n", - vsi->vsi_num, status); + dev_err(dev, "VSI %d failed TC config, error %s\n", + vsi->vsi_num, ice_stat_str(status)); ret = -EIO; goto out; } @@ -2985,33 +3022,27 @@ void ice_update_rx_ring_stats(struct ice_ring *rx_ring, u64 pkts, u64 bytes) } /** - * ice_vsi_cfg_mac_fltr - Add or remove a MAC address filter for a VSI - * @vsi: the VSI being configured MAC filter - * @macaddr: the MAC address to be added. - * @set: Add or delete a MAC filter - * - * Adds or removes MAC address filter entry for VF VSI + * ice_status_to_errno - convert from enum ice_status to Linux errno + * @err: ice_status value to convert */ -enum ice_status -ice_vsi_cfg_mac_fltr(struct ice_vsi *vsi, const u8 *macaddr, bool set) +int ice_status_to_errno(enum ice_status err) { - LIST_HEAD(tmp_add_list); - enum ice_status status; - - /* Update MAC filter list to be added or removed for a VSI */ - if (ice_add_mac_to_list(vsi, &tmp_add_list, macaddr)) { - status = ICE_ERR_NO_MEMORY; - goto cfg_mac_fltr_exit; + switch (err) { + case ICE_SUCCESS: + return 0; + case ICE_ERR_DOES_NOT_EXIST: + return -ENOENT; + case ICE_ERR_OUT_OF_RANGE: + return -ENOTTY; + case ICE_ERR_PARAM: + return -EINVAL; + case ICE_ERR_NO_MEMORY: + return -ENOMEM; + case ICE_ERR_MAX_LIMIT: + return -EAGAIN; + default: + return -EINVAL; } - - if (set) - status = ice_add_mac(&vsi->back->hw, &tmp_add_list); - else - status = ice_remove_mac(&vsi->back->hw, &tmp_add_list); - -cfg_mac_fltr_exit: - ice_free_fltr_list(ice_pf_to_dev(vsi->back), &tmp_add_list); - return status; } /** @@ -3079,8 +3110,8 @@ int ice_set_dflt_vsi(struct ice_sw *sw, struct ice_vsi *vsi) status = ice_cfg_dflt_vsi(&vsi->back->hw, vsi->idx, true, ICE_FLTR_RX); if (status) { - dev_err(dev, "Failed to set VSI %d as the default forwarding VSI, error %d\n", - vsi->vsi_num, status); + dev_err(dev, "Failed to set VSI %d as the default forwarding VSI, error %s\n", + vsi->vsi_num, ice_stat_str(status)); return -EIO; } @@ -3118,8 +3149,8 @@ int ice_clear_dflt_vsi(struct ice_sw *sw) status = ice_cfg_dflt_vsi(&dflt_vsi->back->hw, dflt_vsi->idx, false, ICE_FLTR_RX); if (status) { - dev_err(dev, "Failed to clear the default forwarding VSI %d, error %d\n", - dflt_vsi->vsi_num, status); + dev_err(dev, "Failed to clear the default forwarding VSI %d, error %s\n", + dflt_vsi->vsi_num, ice_stat_str(status)); return -EIO; } diff --git a/drivers/net/ethernet/intel/ice/ice_lib.h b/drivers/net/ethernet/intel/ice/ice_lib.h index 04ca00799364..d80e6afa4511 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_lib.h @@ -8,12 +8,6 @@ const char *ice_vsi_type_str(enum ice_vsi_type vsi_type); -int -ice_add_mac_to_list(struct ice_vsi *vsi, struct list_head *add_list, - const u8 *macaddr); - -void ice_free_fltr_list(struct device *dev, struct list_head *h); - void ice_update_eth_stats(struct ice_vsi *vsi); int ice_vsi_cfg_rxqs(struct ice_vsi *vsi); @@ -22,7 +16,8 @@ int ice_vsi_cfg_lan_txqs(struct ice_vsi *vsi); void ice_vsi_cfg_msix(struct ice_vsi *vsi); -int ice_vsi_add_vlan(struct ice_vsi *vsi, u16 vid); +int +ice_vsi_add_vlan(struct ice_vsi *vsi, u16 vid, enum ice_sw_fwd_act_type action); int ice_vsi_kill_vlan(struct ice_vsi *vsi, u16 vid); @@ -79,6 +74,9 @@ int ice_vsi_rebuild(struct ice_vsi *vsi, bool init_vsi); bool ice_is_reset_in_progress(unsigned long *state); +void +ice_write_qrxflxp_cntxt(struct ice_hw *hw, u16 pf_q, u32 rxdid, u32 prio); + void ice_vsi_put_qs(struct ice_vsi *vsi); void ice_vsi_dis_irq(struct ice_vsi *vsi); @@ -97,6 +95,8 @@ void ice_update_rx_ring_stats(struct ice_ring *ring, u64 pkts, u64 bytes); void ice_vsi_cfg_frame_size(struct ice_vsi *vsi); +int ice_status_to_errno(enum ice_status err); + u32 ice_intrl_usec_to_reg(u8 intrl, u8 gran); enum ice_status diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index 5b190c257124..082825e3cb39 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -8,6 +8,7 @@ #include "ice.h" #include "ice_base.h" #include "ice_lib.h" +#include "ice_fltr.h" #include "ice_dcb_lib.h" #include "ice_dcb_nl.h" #include "ice_devlink.h" @@ -133,38 +134,24 @@ static void ice_check_for_hang_subtask(struct ice_pf *pf) static int ice_init_mac_fltr(struct ice_pf *pf) { enum ice_status status; - u8 broadcast[ETH_ALEN]; struct ice_vsi *vsi; + u8 *perm_addr; vsi = ice_get_main_vsi(pf); if (!vsi) return -EINVAL; - /* To add a MAC filter, first add the MAC to a list and then - * pass the list to ice_add_mac. - */ - - /* Add a unicast MAC filter so the VSI can get its packets */ - status = ice_vsi_cfg_mac_fltr(vsi, vsi->port_info->mac.perm_addr, true); - if (status) - goto unregister; - - /* VSI needs to receive broadcast traffic, so add the broadcast - * MAC address to the list as well. - */ - eth_broadcast_addr(broadcast); - status = ice_vsi_cfg_mac_fltr(vsi, broadcast, true); - if (status) - goto unregister; + perm_addr = vsi->port_info->mac.perm_addr; + status = ice_fltr_add_mac_and_broadcast(vsi, perm_addr, ICE_FWD_TO_VSI); + if (!status) + return 0; - return 0; -unregister: /* We aren't useful with no MAC filters, so unregister if we * had an error */ - if (status && vsi->netdev->reg_state == NETREG_REGISTERED) { - dev_err(ice_pf_to_dev(pf), "Could not add MAC filters error %d. Unregistering device\n", - status); + if (vsi->netdev->reg_state == NETREG_REGISTERED) { + dev_err(ice_pf_to_dev(pf), "Could not add MAC filters error %s. Unregistering device\n", + ice_stat_str(status)); unregister_netdev(vsi->netdev); free_netdev(vsi->netdev); vsi->netdev = NULL; @@ -188,7 +175,8 @@ static int ice_add_mac_to_sync_list(struct net_device *netdev, const u8 *addr) struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; - if (ice_add_mac_to_list(vsi, &vsi->tmp_sync_list, addr)) + if (ice_fltr_add_mac_to_list(vsi, &vsi->tmp_sync_list, addr, + ICE_FWD_TO_VSI)) return -EINVAL; return 0; @@ -209,7 +197,8 @@ static int ice_add_mac_to_unsync_list(struct net_device *netdev, const u8 *addr) struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; - if (ice_add_mac_to_list(vsi, &vsi->tmp_unsync_list, addr)) + if (ice_fltr_add_mac_to_list(vsi, &vsi->tmp_unsync_list, addr, + ICE_FWD_TO_VSI)) return -EINVAL; return 0; @@ -307,8 +296,8 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) } /* Remove MAC addresses in the unsync list */ - status = ice_remove_mac(hw, &vsi->tmp_unsync_list); - ice_free_fltr_list(dev, &vsi->tmp_unsync_list); + status = ice_fltr_remove_mac_list(vsi, &vsi->tmp_unsync_list); + ice_fltr_free_list(dev, &vsi->tmp_unsync_list); if (status) { netdev_err(netdev, "Failed to delete MAC filters\n"); /* if we failed because of alloc failures, just bail */ @@ -319,8 +308,8 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) } /* Add MAC addresses in the sync list */ - status = ice_add_mac(hw, &vsi->tmp_sync_list); - ice_free_fltr_list(dev, &vsi->tmp_sync_list); + status = ice_fltr_add_mac_list(vsi, &vsi->tmp_sync_list); + ice_fltr_free_list(dev, &vsi->tmp_sync_list); /* If filter is added successfully or already exists, do not go into * 'if' condition and report it as error. Instead continue processing * rest of the function. @@ -357,7 +346,8 @@ static int ice_vsi_sync_fltr(struct ice_vsi *vsi) vsi->current_netdev_flags &= ~IFF_ALLMULTI; goto out_promisc; } - } else if (!(vsi->current_netdev_flags & IFF_ALLMULTI)) { + } else { + /* !(vsi->current_netdev_flags & IFF_ALLMULTI) */ if (vsi->vlan_ena) promisc_m = ICE_MCAST_VLAN_PROMISC_BITS; else @@ -462,7 +452,7 @@ static void ice_prepare_for_reset(struct ice_pf *pf) { struct ice_hw *hw = &pf->hw; - int i; + unsigned int i; /* already prepared for reset */ if (test_bit(__ICE_PREPARED_FOR_RESET, pf->state)) @@ -1017,8 +1007,8 @@ static int __ice_clean_ctrlq(struct ice_pf *pf, enum ice_ctl_q q_type) if (ret == ICE_ERR_AQ_NO_WORK) break; if (ret) { - dev_err(dev, "%s Receive Queue event error %d\n", qtype, - ret); + dev_err(dev, "%s Receive Queue event error %s\n", qtype, + ice_stat_str(ret)); break; } @@ -1123,7 +1113,7 @@ static void ice_clean_mailboxq_subtask(struct ice_pf *pf) * * If not already scheduled, this puts the task into the work queue. */ -static void ice_service_task_schedule(struct ice_pf *pf) +void ice_service_task_schedule(struct ice_pf *pf) { if (!test_bit(__ICE_SERVICE_DIS, pf->state) && !test_and_set_bit(__ICE_SERVICE_SCHED, pf->state) && @@ -1198,8 +1188,8 @@ static void ice_handle_mdd_event(struct ice_pf *pf) { struct device *dev = ice_pf_to_dev(pf); struct ice_hw *hw = &pf->hw; + unsigned int i; u32 reg; - int i; if (!test_and_clear_bit(__ICE_MDD_EVENT_PENDING, pf->state)) { /* Since the VF MDD event logging is rate limited, check if @@ -1332,8 +1322,13 @@ static void ice_handle_mdd_event(struct ice_pf *pf) * PF can be configured to reset the VF through ethtool * private flag mdd-auto-reset-vf. */ - if (test_bit(ICE_FLAG_MDD_AUTO_RESET_VF, pf->flags)) + if (test_bit(ICE_FLAG_MDD_AUTO_RESET_VF, pf->flags)) { + /* VF MDD event counters will be cleared by + * reset, so print the event prior to reset. + */ + ice_print_vf_rx_mdd_event(vf); ice_reset_vf(&pf->vf[i], false); + } } } @@ -1493,7 +1488,7 @@ static void ice_service_task(struct work_struct *work) ice_process_vflr_event(pf); ice_clean_mailboxq_subtask(pf); - + ice_sync_arfs_fltrs(pf); /* Clear __ICE_SERVICE_SCHED flag to allow scheduling next event */ ice_service_task_complete(pf); @@ -1652,9 +1647,14 @@ static int ice_vsi_req_irq_msix(struct ice_vsi *vsi, char *basename) } /* register for affinity change notifications */ - q_vector->affinity_notify.notify = ice_irq_affinity_notify; - q_vector->affinity_notify.release = ice_irq_affinity_release; - irq_set_affinity_notifier(irq_num, &q_vector->affinity_notify); + if (!IS_ENABLED(CONFIG_RFS_ACCEL)) { + struct irq_affinity_notify *affinity_notify; + + affinity_notify = &q_vector->affinity_notify; + affinity_notify->notify = ice_irq_affinity_notify; + affinity_notify->release = ice_irq_affinity_release; + irq_set_affinity_notifier(irq_num, affinity_notify); + } /* assign the mask for this irq */ irq_set_affinity_hint(irq_num, &q_vector->affinity_mask); @@ -1666,8 +1666,9 @@ static int ice_vsi_req_irq_msix(struct ice_vsi *vsi, char *basename) free_q_irqs: while (vector) { vector--; - irq_num = pf->msix_entries[base + vector].vector, - irq_set_affinity_notifier(irq_num, NULL); + irq_num = pf->msix_entries[base + vector].vector; + if (!IS_ENABLED(CONFIG_RFS_ACCEL)) + irq_set_affinity_notifier(irq_num, NULL); irq_set_affinity_hint(irq_num, NULL); devm_free_irq(dev, irq_num, &vsi->q_vectors[vector]); } @@ -1809,8 +1810,8 @@ int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog) status = ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc, max_txqs); if (status) { - dev_err(dev, "Failed VSI LAN queue config for XDP, error:%d\n", - status); + dev_err(dev, "Failed VSI LAN queue config for XDP, error: %s\n", + ice_stat_str(status)); goto clear_xdp_rings; } ice_vsi_assign_bpf_prog(vsi, prog); @@ -1898,6 +1899,9 @@ free_qmap: for (i = 0; i < vsi->tc_cfg.numtc; i++) max_txqs[i] = vsi->num_txq; + /* change number of XDP Tx queues to 0 */ + vsi->num_xdp_txq = 0; + return ice_cfg_vsi_lan(vsi->port_info, vsi->idx, vsi->tc_cfg.ena_tc, max_txqs); } @@ -1931,7 +1935,7 @@ ice_xdp_setup_prog(struct ice_vsi *vsi, struct bpf_prog *prog, } if (!ice_is_xdp_ena_vsi(vsi) && prog) { - vsi->num_xdp_txq = vsi->alloc_txq; + vsi->num_xdp_txq = vsi->alloc_rxq; xdp_ring_err = ice_prepare_xdp_rings(vsi, prog); if (xdp_ring_err) NL_SET_ERR_MSG_MOD(extack, "Setting up XDP Tx resources failed"); @@ -2137,10 +2141,8 @@ static irqreturn_t ice_misc_intr(int __always_unused irq, void *data) } ret = IRQ_HANDLED; - if (!test_bit(__ICE_DOWN, pf->state)) { - ice_service_task_schedule(pf); - ice_irq_dynamic_ena(hw, NULL, NULL); - } + ice_service_task_schedule(pf); + ice_irq_dynamic_ena(hw, NULL, NULL); return ret; } @@ -2247,7 +2249,7 @@ static int ice_req_irq_msix_misc(struct ice_pf *pf) return oicr_idx; pf->num_avail_sw_msix -= 1; - pf->oicr_idx = oicr_idx; + pf->oicr_idx = (u16)oicr_idx; err = devm_request_irq(dev, pf->msix_entries[pf->oicr_idx].vector, ice_misc_intr, 0, pf->int_name, pf); @@ -2331,6 +2333,7 @@ static void ice_set_netdev_features(struct net_device *netdev) dflt_features = NETIF_F_SG | NETIF_F_HIGHDMA | + NETIF_F_NTUPLE | NETIF_F_RXHASH; csumo_features = NETIF_F_RXCSUM | @@ -2342,13 +2345,27 @@ static void ice_set_netdev_features(struct net_device *netdev) NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX; - tso_features = NETIF_F_TSO | + tso_features = NETIF_F_TSO | + NETIF_F_TSO_ECN | + NETIF_F_TSO6 | + NETIF_F_GSO_GRE | + NETIF_F_GSO_UDP_TUNNEL | + NETIF_F_GSO_GRE_CSUM | + NETIF_F_GSO_UDP_TUNNEL_CSUM | + NETIF_F_GSO_PARTIAL | + NETIF_F_GSO_IPXIP4 | + NETIF_F_GSO_IPXIP6 | NETIF_F_GSO_UDP_L4; + netdev->gso_partial_features |= NETIF_F_GSO_UDP_TUNNEL_CSUM | + NETIF_F_GSO_GRE_CSUM; /* set features that user can change */ netdev->hw_features = dflt_features | csumo_features | vlano_features | tso_features; + /* add support for HW_CSUM on packets with MPLS header */ + netdev->mpls_features = NETIF_F_HW_CSUM; + /* enable features */ netdev->features |= netdev->hw_features; /* encap and VLAN devices inherit default, csumo and tso features */ @@ -2411,7 +2428,7 @@ static int ice_cfg_netdev(struct ice_vsi *vsi) err = register_netdev(vsi->netdev); if (err) - goto err_destroy_devlink_port; + goto err_free_netdev; devlink_port_type_eth_set(&pf->devlink_port, vsi->netdev); @@ -2422,9 +2439,11 @@ static int ice_cfg_netdev(struct ice_vsi *vsi) return 0; +err_free_netdev: + free_netdev(vsi->netdev); + vsi->netdev = NULL; err_destroy_devlink_port: ice_devlink_destroy_port(pf); - return err; } @@ -2457,6 +2476,20 @@ ice_pf_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi) } /** + * ice_ctrl_vsi_setup - Set up a control VSI + * @pf: board private structure + * @pi: pointer to the port_info instance + * + * Returns pointer to the successfully allocated VSI software struct + * on success, otherwise returns NULL on failure. + */ +static struct ice_vsi * +ice_ctrl_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi) +{ + return ice_vsi_setup(pf, pi, ICE_VSI_CTRL, ICE_INVAL_VFID); +} + +/** * ice_lb_vsi_setup - Set up a loopback VSI * @pf: board private structure * @pi: pointer to the port_info instance @@ -2509,7 +2542,7 @@ ice_vlan_rx_add_vid(struct net_device *netdev, __always_unused __be16 proto, /* Add a switch rule for this VLAN ID so its corresponding VLAN tagged * packets aren't pruned by the device's internal switch on Rx */ - ret = ice_vsi_add_vlan(vsi, vid); + ret = ice_vsi_add_vlan(vsi, vid, ICE_FWD_TO_VSI); if (!ret) { vsi->vlan_ena = true; set_bit(ICE_VSI_FLAG_VLAN_FLTR_CHANGED, vsi->flags); @@ -2594,12 +2627,22 @@ static int ice_setup_pf_sw(struct ice_pf *pf) */ ice_napi_add(vsi); + status = ice_set_cpu_rx_rmap(vsi); + if (status) { + dev_err(ice_pf_to_dev(pf), "Failed to set CPU Rx map VSI %d error %d\n", + vsi->vsi_num, status); + status = -EINVAL; + goto unroll_napi_add; + } status = ice_init_mac_fltr(pf); if (status) - goto unroll_napi_add; + goto free_cpu_rx_map; return status; +free_cpu_rx_map: + ice_free_cpu_rx_rmap(vsi); + unroll_napi_add: if (vsi) { ice_napi_del(vsi); @@ -2630,7 +2673,8 @@ unroll_vsi_setup: static u16 ice_get_avail_q_count(unsigned long *pf_qmap, struct mutex *lock, u16 size) { - u16 count = 0, bit; + unsigned long bit; + u16 count = 0; mutex_lock(lock); for_each_clear_bit(bit, pf_qmap, size) @@ -2703,6 +2747,23 @@ static void ice_set_pf_caps(struct ice_pf *pf) if (func_caps->common_cap.rss_table_size) set_bit(ICE_FLAG_RSS_ENA, pf->flags); + clear_bit(ICE_FLAG_FD_ENA, pf->flags); + if (func_caps->fd_fltr_guar > 0 || func_caps->fd_fltr_best_effort > 0) { + u16 unused; + + /* ctrl_vsi_idx will be set to a valid value when flow director + * is setup by ice_init_fdir + */ + pf->ctrl_vsi_idx = ICE_NO_VSI; + set_bit(ICE_FLAG_FD_ENA, pf->flags); + /* force guaranteed filter pool for PF */ + ice_alloc_fd_guar_item(&pf->hw, &unused, + func_caps->fd_fltr_guar); + /* force shared filter pool for PF */ + ice_alloc_fd_shrd_item(&pf->hw, &unused, + func_caps->fd_fltr_best_effort); + } + pf->max_pf_txqs = func_caps->common_cap.num_txq; pf->max_pf_rxqs = func_caps->common_cap.num_rxq; } @@ -2769,6 +2830,15 @@ static int ice_ena_msix_range(struct ice_pf *pf) v_budget += needed; v_left -= needed; + /* reserve one vector for flow director */ + if (test_bit(ICE_FLAG_FD_ENA, pf->flags)) { + needed = ICE_FDIR_MSIX; + if (v_left < needed) + goto no_hw_vecs_left_err; + v_budget += needed; + v_left -= needed; + } + pf->msix_entries = devm_kcalloc(dev, v_budget, sizeof(*pf->msix_entries), GFP_KERNEL); @@ -2793,8 +2863,10 @@ static int ice_ena_msix_range(struct ice_pf *pf) if (v_actual < v_budget) { dev_warn(dev, "not enough OS MSI-X vectors. requested = %d, obtained = %d\n", v_budget, v_actual); -/* 2 vectors for LAN (traffic + OICR) */ +/* 2 vectors each for LAN and RDMA (traffic + OICR), one for flow director */ #define ICE_MIN_LAN_VECS 2 +#define ICE_MIN_RDMA_VECS 2 +#define ICE_MIN_VECS (ICE_MIN_LAN_VECS + ICE_MIN_RDMA_VECS + 1) if (v_actual < ICE_MIN_LAN_VECS) { /* error if we can't get minimum vectors */ @@ -2869,8 +2941,8 @@ static int ice_init_interrupt_scheme(struct ice_pf *pf) } /* populate SW interrupts pool with number of OS granted IRQs. */ - pf->num_avail_sw_msix = vectors; - pf->irq_tracker->num_entries = vectors; + pf->num_avail_sw_msix = (u16)vectors; + pf->irq_tracker->num_entries = (u16)vectors; pf->irq_tracker->end = pf->irq_tracker->num_entries; return 0; @@ -2902,9 +2974,9 @@ int ice_vsi_recfg_qs(struct ice_vsi *vsi, int new_rx, int new_tx) } if (new_tx) - vsi->req_txq = new_tx; + vsi->req_txq = (u16)new_tx; if (new_rx) - vsi->req_rxq = new_rx; + vsi->req_rxq = (u16)new_rx; /* set for the next time the netdev is started */ if (!netif_running(vsi->netdev)) { @@ -2985,6 +3057,9 @@ ice_log_pkg_init(struct ice_hw *hw, enum ice_status *status) *status = ICE_ERR_NOT_SUPPORTED; } break; + case ICE_ERR_FW_DDP_MISMATCH: + dev_err(dev, "The firmware loaded on the device is not compatible with the DDP package. Please update the device's NVM. Entering safe mode.\n"); + break; case ICE_ERR_BUF_TOO_SHORT: case ICE_ERR_CFG: dev_err(dev, "The DDP package file is invalid. Entering Safe Mode.\n"); @@ -3013,6 +3088,9 @@ ice_log_pkg_init(struct ice_hw *hw, enum ice_status *status) case ICE_AQ_RC_EBADMAN: case ICE_AQ_RC_EBADBUF: dev_err(dev, "An error occurred on the device while loading the DDP package. The device will be reset.\n"); + /* poll for reset to complete */ + if (ice_check_reset(hw)) + dev_err(dev, "Error resetting device. Please reload the driver\n"); return; default: break; @@ -3100,6 +3178,53 @@ static enum ice_status ice_send_version(struct ice_pf *pf) } /** + * ice_init_fdir - Initialize flow director VSI and configuration + * @pf: pointer to the PF instance + * + * returns 0 on success, negative on error + */ +static int ice_init_fdir(struct ice_pf *pf) +{ + struct device *dev = ice_pf_to_dev(pf); + struct ice_vsi *ctrl_vsi; + int err; + + /* Side Band Flow Director needs to have a control VSI. + * Allocate it and store it in the PF. + */ + ctrl_vsi = ice_ctrl_vsi_setup(pf, pf->hw.port_info); + if (!ctrl_vsi) { + dev_dbg(dev, "could not create control VSI\n"); + return -ENOMEM; + } + + err = ice_vsi_open_ctrl(ctrl_vsi); + if (err) { + dev_dbg(dev, "could not open control VSI\n"); + goto err_vsi_open; + } + + mutex_init(&pf->hw.fdir_fltr_lock); + + err = ice_fdir_create_dflt_rules(pf); + if (err) + goto err_fdir_rule; + + return 0; + +err_fdir_rule: + ice_fdir_release_flows(&pf->hw); + ice_vsi_close(ctrl_vsi); +err_vsi_open: + ice_vsi_release(ctrl_vsi); + if (pf->ctrl_vsi_idx != ICE_NO_VSI) { + pf->vsi[pf->ctrl_vsi_idx] = NULL; + pf->ctrl_vsi_idx = ICE_NO_VSI; + } + return err; +} + +/** * ice_get_opt_fw_name - return optional firmware file name or NULL * @pf: pointer to the PF instance */ @@ -3123,7 +3248,7 @@ static char *ice_get_opt_fw_name(struct ice_pf *pf) if (!opt_fw_filename) return NULL; - snprintf(opt_fw_filename, NAME_MAX, "%sice-%016llX.pkg", + snprintf(opt_fw_filename, NAME_MAX, "%sice-%016llx.pkg", ICE_DDP_PKG_PATH, dsn); return opt_fw_filename; @@ -3295,12 +3420,9 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) if (err) { dev_err(dev, "ice_init_interrupt_scheme failed: %d\n", err); err = -EIO; - goto err_init_interrupt_unroll; + goto err_init_vsi_unroll; } - /* Driver is mostly up */ - clear_bit(__ICE_DOWN, pf->state); - /* In case of MSIX we are going to setup the misc vector right here * to handle admin queue events etc. In case of legacy and MSI * the misc functionality and queue processing is combined in @@ -3356,12 +3478,16 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) ice_verify_cacheline_size(pf); - /* If no DDP driven features have to be setup, return here */ + /* If no DDP driven features have to be setup, we are done with probe */ if (ice_is_safe_mode(pf)) - return 0; + goto probe_done; /* initialize DDP driven features */ + /* Note: Flow director init failure is non-fatal to load */ + if (ice_init_fdir(pf)) + dev_err(dev, "could not initialize flow director\n"); + /* Note: DCB init failure is non-fatal to load */ if (ice_init_pf_dcb(pf, false)) { clear_bit(ICE_FLAG_DCB_CAPABLE, pf->flags); @@ -3373,6 +3499,9 @@ ice_probe(struct pci_dev *pdev, const struct pci_device_id __always_unused *ent) /* print PCI link speed and width */ pcie_print_link_status(pf->pdev); +probe_done: + /* ready to go, so clear down state bit */ + clear_bit(__ICE_DOWN, pf->state); return 0; err_alloc_sw_unroll: @@ -3384,6 +3513,7 @@ err_msix_misc_unroll: ice_free_irq_msix_misc(pf); err_init_interrupt_unroll: ice_clear_interrupt_scheme(pf); +err_init_vsi_unroll: devm_kfree(dev, pf->vsi); err_init_pf_unroll: ice_deinit_pf(pf); @@ -3421,6 +3551,9 @@ static void ice_remove(struct pci_dev *pdev) set_bit(__ICE_DOWN, pf->state); ice_service_task_stop(pf); + mutex_destroy(&(&pf->hw)->fdir_fltr_lock); + if (!ice_is_safe_mode(pf)) + ice_remove_arfs(pf); ice_devlink_destroy_port(pf); ice_vsi_release_all(pf); ice_free_irq_msix_misc(pf); @@ -3705,25 +3838,24 @@ static int ice_set_mac_address(struct net_device *netdev, void *pi) return -EBUSY; } - /* When we change the MAC address we also have to change the MAC address - * based filter rules that were created previously for the old MAC - * address. So first, we remove the old filter rule using ice_remove_mac - * and then create a new filter rule using ice_add_mac via - * ice_vsi_cfg_mac_fltr function call for both add and/or remove - * filters. - */ - status = ice_vsi_cfg_mac_fltr(vsi, netdev->dev_addr, false); - if (status) { + /* Clean up old MAC filter. Not an error if old filter doesn't exist */ + status = ice_fltr_remove_mac(vsi, netdev->dev_addr, ICE_FWD_TO_VSI); + if (status && status != ICE_ERR_DOES_NOT_EXIST) { err = -EADDRNOTAVAIL; goto err_update_filters; } - status = ice_vsi_cfg_mac_fltr(vsi, mac, true); - if (status) { - err = -EADDRNOTAVAIL; - goto err_update_filters; + /* Add filter for new MAC. If filter exists, just return success */ + status = ice_fltr_add_mac(vsi, mac, ICE_FWD_TO_VSI); + if (status == ICE_ERR_ALREADY_EXISTS) { + netdev_dbg(netdev, "filter for MAC %pM already exists\n", mac); + return 0; } + /* error if the new filter addition failed */ + if (status) + err = -EADDRNOTAVAIL; + err_update_filters: if (err) { netdev_err(netdev, "can't set MAC %pM. filter update failed\n", @@ -3740,8 +3872,8 @@ err_update_filters: flags = ICE_AQC_MAN_MAC_UPDATE_LAA_WOL; status = ice_aq_manage_mac_write(hw, mac, flags, NULL); if (status) { - netdev_err(netdev, "can't set MAC %pM. write to firmware failed error %d\n", - mac, status); + netdev_err(netdev, "can't set MAC %pM. write to firmware failed error %s\n", + mac, ice_stat_str(status)); } return 0; } @@ -3805,8 +3937,8 @@ ice_set_tx_maxrate(struct net_device *netdev, int queue_index, u32 maxrate) status = ice_cfg_q_bw_lmt(vsi->port_info, vsi->idx, tc, q_handle, ICE_MAX_BW, maxrate * 1000); if (status) { - netdev_err(netdev, "Unable to set Tx max rate, error %d\n", - status); + netdev_err(netdev, "Unable to set Tx max rate, error %s\n", + ice_stat_str(status)); return -EIO; } @@ -3938,6 +4070,16 @@ ice_set_features(struct net_device *netdev, netdev_features_t features) (netdev->features & NETIF_F_HW_VLAN_CTAG_FILTER)) ret = ice_cfg_vlan_pruning(vsi, false, false); + if ((features & NETIF_F_NTUPLE) && + !(netdev->features & NETIF_F_NTUPLE)) { + ice_vsi_manage_fdir(vsi, true); + ice_init_arfs(vsi); + } else if (!(features & NETIF_F_NTUPLE) && + (netdev->features & NETIF_F_NTUPLE)) { + ice_vsi_manage_fdir(vsi, false); + ice_clear_arfs(vsi); + } + return ret; } @@ -4084,6 +4226,33 @@ ice_fetch_u64_stats_per_ring(struct ice_ring *ring, u64 *pkts, u64 *bytes) } /** + * ice_update_vsi_tx_ring_stats - Update VSI Tx ring stats counters + * @vsi: the VSI to be updated + * @rings: rings to work on + * @count: number of rings + */ +static void +ice_update_vsi_tx_ring_stats(struct ice_vsi *vsi, struct ice_ring **rings, + u16 count) +{ + struct rtnl_link_stats64 *vsi_stats = &vsi->net_stats; + u16 i; + + for (i = 0; i < count; i++) { + struct ice_ring *ring; + u64 pkts, bytes; + + ring = READ_ONCE(rings[i]); + ice_fetch_u64_stats_per_ring(ring, &pkts, &bytes); + vsi_stats->tx_packets += pkts; + vsi_stats->tx_bytes += bytes; + vsi->tx_restart += ring->tx_stats.restart_q; + vsi->tx_busy += ring->tx_stats.tx_busy; + vsi->tx_linearize += ring->tx_stats.tx_linearize; + } +} + +/** * ice_update_vsi_ring_stats - Update VSI stats counters * @vsi: the VSI to be updated */ @@ -4110,15 +4279,7 @@ static void ice_update_vsi_ring_stats(struct ice_vsi *vsi) rcu_read_lock(); /* update Tx rings counters */ - ice_for_each_txq(vsi, i) { - ring = READ_ONCE(vsi->tx_rings[i]); - ice_fetch_u64_stats_per_ring(ring, &pkts, &bytes); - vsi_stats->tx_packets += pkts; - vsi_stats->tx_bytes += bytes; - vsi->tx_restart += ring->tx_stats.restart_q; - vsi->tx_busy += ring->tx_stats.tx_busy; - vsi->tx_linearize += ring->tx_stats.tx_linearize; - } + ice_update_vsi_tx_ring_stats(vsi, vsi->tx_rings, vsi->num_txq); /* update Rx rings counters */ ice_for_each_rxq(vsi, i) { @@ -4130,6 +4291,11 @@ static void ice_update_vsi_ring_stats(struct ice_vsi *vsi) vsi->rx_page_failed += ring->rx_stats.alloc_page_failed; } + /* update XDP Tx rings counters */ + if (ice_is_xdp_ena_vsi(vsi)) + ice_update_vsi_tx_ring_stats(vsi, vsi->xdp_rings, + vsi->num_xdp_txq); + rcu_read_unlock(); } @@ -4162,7 +4328,13 @@ void ice_update_vsi_stats(struct ice_vsi *vsi) if (vsi->type == ICE_VSI_PF) { cur_ns->rx_crc_errors = pf->stats.crc_errors; cur_ns->rx_errors = pf->stats.crc_errors + - pf->stats.illegal_bytes; + pf->stats.illegal_bytes + + pf->stats.rx_len_errors + + pf->stats.rx_undersize + + pf->hw_csum_rx_error + + pf->stats.rx_jabber + + pf->stats.rx_fragments + + pf->stats.rx_oversize; cur_ns->rx_length_errors = pf->stats.rx_len_errors; /* record drops from the port level */ cur_ns->rx_missed_errors = pf->stats.eth.rx_discards; @@ -4177,6 +4349,7 @@ void ice_update_pf_stats(struct ice_pf *pf) { struct ice_hw_port_stats *prev_ps, *cur_ps; struct ice_hw *hw = &pf->hw; + u16 fd_ctr_base; u8 port; port = hw->port_info->lport; @@ -4265,6 +4438,12 @@ void ice_update_pf_stats(struct ice_pf *pf) ice_stat_update40(hw, GLPRT_PTC9522L(port), pf->stat_prev_loaded, &prev_ps->tx_size_big, &cur_ps->tx_size_big); + fd_ctr_base = hw->fd_ctr_base; + + ice_stat_update40(hw, + GLSTAT_FD_CNT0L(ICE_FD_SB_STAT_IDX(fd_ctr_base)), + pf->stat_prev_loaded, &prev_ps->fd_sb_match, + &cur_ps->fd_sb_match); ice_stat_update32(hw, GLPRT_LXONRXC(port), pf->stat_prev_loaded, &prev_ps->link_xon_rx, &cur_ps->link_xon_rx); @@ -4308,6 +4487,8 @@ void ice_update_pf_stats(struct ice_pf *pf) ice_stat_update32(hw, GLPRT_RJC(port), pf->stat_prev_loaded, &prev_ps->rx_jabber, &cur_ps->rx_jabber); + cur_ps->fd_sb_status = test_bit(ICE_FLAG_FD_ENA, pf->flags) ? 1 : 0; + pf->stat_prev_loaded = true; } @@ -4493,6 +4674,62 @@ int ice_vsi_setup_rx_rings(struct ice_vsi *vsi) } /** + * ice_vsi_open_ctrl - open control VSI for use + * @vsi: the VSI to open + * + * Initialization of the Control VSI + * + * Returns 0 on success, negative value on error + */ +int ice_vsi_open_ctrl(struct ice_vsi *vsi) +{ + char int_name[ICE_INT_NAME_STR_LEN]; + struct ice_pf *pf = vsi->back; + struct device *dev; + int err; + + dev = ice_pf_to_dev(pf); + /* allocate descriptors */ + err = ice_vsi_setup_tx_rings(vsi); + if (err) + goto err_setup_tx; + + err = ice_vsi_setup_rx_rings(vsi); + if (err) + goto err_setup_rx; + + err = ice_vsi_cfg(vsi); + if (err) + goto err_setup_rx; + + snprintf(int_name, sizeof(int_name) - 1, "%s-%s:ctrl", + dev_driver_string(dev), dev_name(dev)); + err = ice_vsi_req_irq_msix(vsi, int_name); + if (err) + goto err_setup_rx; + + ice_vsi_cfg_msix(vsi); + + err = ice_vsi_start_all_rx_rings(vsi); + if (err) + goto err_up_complete; + + clear_bit(__ICE_DOWN, vsi->state); + ice_vsi_ena_irq(vsi); + + return 0; + +err_up_complete: + ice_down(vsi); +err_setup_rx: + ice_vsi_free_rx_rings(vsi); +err_setup_tx: + ice_vsi_free_tx_rings(vsi); + + return err; +} + +/** * ice_vsi_open - Called when a network interface is made active * @vsi: the VSI to open * @@ -4604,8 +4841,9 @@ static int ice_vsi_rebuild_by_type(struct ice_pf *pf, enum ice_vsi_type type) /* replay filters for the VSI */ status = ice_replay_vsi(&pf->hw, vsi->idx); if (status) { - dev_err(dev, "replay VSI failed, status %d, VSI index %d, type %s\n", - status, vsi->idx, ice_vsi_type_str(type)); + dev_err(dev, "replay VSI failed, status %s, VSI index %d, type %s\n", + ice_stat_str(status), vsi->idx, + ice_vsi_type_str(type)); return -EIO; } @@ -4659,6 +4897,11 @@ static void ice_update_pf_netdev_link(struct ice_pf *pf) * ice_rebuild - rebuild after reset * @pf: PF to rebuild * @reset_type: type of reset + * + * Do not rebuild VF VSI in this flow because that is already handled via + * ice_reset_all_vfs(). This is because requirements for resetting a VF after a + * PFR/CORER/GLOBER/etc. are different than the normal flow. Also, we don't want + * to reset/rebuild all the VF VSI twice. */ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) { @@ -4674,7 +4917,8 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) ret = ice_init_all_ctrlq(hw); if (ret) { - dev_err(dev, "control queues init failed %d\n", ret); + dev_err(dev, "control queues init failed %s\n", + ice_stat_str(ret)); goto err_init_ctrlq; } @@ -4690,7 +4934,8 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) ret = ice_clear_pf_cfg(hw); if (ret) { - dev_err(dev, "clear PF configuration failed %d\n", ret); + dev_err(dev, "clear PF configuration failed %s\n", + ice_stat_str(ret)); goto err_init_ctrlq; } @@ -4704,7 +4949,13 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) ret = ice_get_caps(hw); if (ret) { - dev_err(dev, "ice_get_caps failed %d\n", ret); + dev_err(dev, "ice_get_caps failed %s\n", ice_stat_str(ret)); + goto err_init_ctrlq; + } + + ret = ice_aq_set_mac_cfg(hw, ICE_AQ_SET_MAC_FRAME_SIZE_MAX, NULL); + if (ret) { + dev_err(dev, "set_mac_cfg failed %s\n", ice_stat_str(ret)); goto err_init_ctrlq; } @@ -4723,6 +4974,21 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) goto err_sched_init_port; } + if (test_bit(ICE_FLAG_FD_ENA, pf->flags)) { + wr32(hw, PFQF_FD_ENA, PFQF_FD_ENA_FD_ENA_M); + if (!rd32(hw, PFQF_FD_SIZE)) { + u16 unused, guar, b_effort; + + guar = hw->func_caps.fd_fltr_guar; + b_effort = hw->func_caps.fd_fltr_best_effort; + + /* force guaranteed filter pool for PF */ + ice_alloc_fd_guar_item(hw, &unused, guar); + /* force shared filter pool for PF */ + ice_alloc_fd_shrd_item(hw, &unused, b_effort); + } + } + if (test_bit(ICE_FLAG_DCB_ENA, pf->flags)) ice_dcb_rebuild(pf); @@ -4733,12 +4999,22 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) goto err_vsi_rebuild; } - if (test_bit(ICE_FLAG_SRIOV_ENA, pf->flags)) { - err = ice_vsi_rebuild_by_type(pf, ICE_VSI_VF); + /* If Flow Director is active */ + if (test_bit(ICE_FLAG_FD_ENA, pf->flags)) { + err = ice_vsi_rebuild_by_type(pf, ICE_VSI_CTRL); if (err) { - dev_err(dev, "VF VSI rebuild failed: %d\n", err); + dev_err(dev, "control VSI rebuild failed: %d\n", err); goto err_vsi_rebuild; } + + /* replay HW Flow Director recipes */ + if (hw->fdir_prof) + ice_fdir_replay_flows(hw); + + /* replay Flow Director filters */ + ice_fdir_replay_fltrs(pf); + + ice_rebuild_arfs(pf); } ice_update_pf_netdev_link(pf); @@ -4746,8 +5022,8 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type) /* tell the firmware we are up */ ret = ice_send_version(pf); if (ret) { - dev_err(dev, "Rebuild failed due to error sending driver version: %d\n", - ret); + dev_err(dev, "Rebuild failed due to error sending driver version: %s\n", + ice_stat_str(ret)); goto err_vsi_rebuild; } @@ -4795,7 +5071,7 @@ static int ice_change_mtu(struct net_device *netdev, int new_mtu) struct ice_pf *pf = vsi->back; u8 count = 0; - if (new_mtu == netdev->mtu) { + if (new_mtu == (int)netdev->mtu) { netdev_warn(netdev, "MTU is already %u\n", netdev->mtu); return 0; } @@ -4810,11 +5086,11 @@ static int ice_change_mtu(struct net_device *netdev, int new_mtu) } } - if (new_mtu < netdev->min_mtu) { + if (new_mtu < (int)netdev->min_mtu) { netdev_err(netdev, "new MTU invalid. min_mtu is %d\n", netdev->min_mtu); return -EINVAL; - } else if (new_mtu > netdev->max_mtu) { + } else if (new_mtu > (int)netdev->max_mtu) { netdev_err(netdev, "new MTU invalid. max_mtu is %d\n", netdev->min_mtu); return -EINVAL; @@ -4835,7 +5111,7 @@ static int ice_change_mtu(struct net_device *netdev, int new_mtu) return -EBUSY; } - netdev->mtu = new_mtu; + netdev->mtu = (unsigned int)new_mtu; /* if VSI is up, bring it down and then back up */ if (!test_and_set_bit(__ICE_DOWN, vsi->state)) { @@ -4859,6 +5135,118 @@ static int ice_change_mtu(struct net_device *netdev, int new_mtu) } /** + * ice_aq_str - convert AQ err code to a string + * @aq_err: the AQ error code to convert + */ +const char *ice_aq_str(enum ice_aq_err aq_err) +{ + switch (aq_err) { + case ICE_AQ_RC_OK: + return "OK"; + case ICE_AQ_RC_EPERM: + return "ICE_AQ_RC_EPERM"; + case ICE_AQ_RC_ENOENT: + return "ICE_AQ_RC_ENOENT"; + case ICE_AQ_RC_ENOMEM: + return "ICE_AQ_RC_ENOMEM"; + case ICE_AQ_RC_EBUSY: + return "ICE_AQ_RC_EBUSY"; + case ICE_AQ_RC_EEXIST: + return "ICE_AQ_RC_EEXIST"; + case ICE_AQ_RC_EINVAL: + return "ICE_AQ_RC_EINVAL"; + case ICE_AQ_RC_ENOSPC: + return "ICE_AQ_RC_ENOSPC"; + case ICE_AQ_RC_ENOSYS: + return "ICE_AQ_RC_ENOSYS"; + case ICE_AQ_RC_EMODE: + return "ICE_AQ_RC_EMODE"; + case ICE_AQ_RC_ENOSEC: + return "ICE_AQ_RC_ENOSEC"; + case ICE_AQ_RC_EBADSIG: + return "ICE_AQ_RC_EBADSIG"; + case ICE_AQ_RC_ESVN: + return "ICE_AQ_RC_ESVN"; + case ICE_AQ_RC_EBADMAN: + return "ICE_AQ_RC_EBADMAN"; + case ICE_AQ_RC_EBADBUF: + return "ICE_AQ_RC_EBADBUF"; + } + + return "ICE_AQ_RC_UNKNOWN"; +} + +/** + * ice_stat_str - convert status err code to a string + * @stat_err: the status error code to convert + */ +const char *ice_stat_str(enum ice_status stat_err) +{ + switch (stat_err) { + case ICE_SUCCESS: + return "OK"; + case ICE_ERR_PARAM: + return "ICE_ERR_PARAM"; + case ICE_ERR_NOT_IMPL: + return "ICE_ERR_NOT_IMPL"; + case ICE_ERR_NOT_READY: + return "ICE_ERR_NOT_READY"; + case ICE_ERR_NOT_SUPPORTED: + return "ICE_ERR_NOT_SUPPORTED"; + case ICE_ERR_BAD_PTR: + return "ICE_ERR_BAD_PTR"; + case ICE_ERR_INVAL_SIZE: + return "ICE_ERR_INVAL_SIZE"; + case ICE_ERR_DEVICE_NOT_SUPPORTED: + return "ICE_ERR_DEVICE_NOT_SUPPORTED"; + case ICE_ERR_RESET_FAILED: + return "ICE_ERR_RESET_FAILED"; + case ICE_ERR_FW_API_VER: + return "ICE_ERR_FW_API_VER"; + case ICE_ERR_NO_MEMORY: + return "ICE_ERR_NO_MEMORY"; + case ICE_ERR_CFG: + return "ICE_ERR_CFG"; + case ICE_ERR_OUT_OF_RANGE: + return "ICE_ERR_OUT_OF_RANGE"; + case ICE_ERR_ALREADY_EXISTS: + return "ICE_ERR_ALREADY_EXISTS"; + case ICE_ERR_NVM_CHECKSUM: + return "ICE_ERR_NVM_CHECKSUM"; + case ICE_ERR_BUF_TOO_SHORT: + return "ICE_ERR_BUF_TOO_SHORT"; + case ICE_ERR_NVM_BLANK_MODE: + return "ICE_ERR_NVM_BLANK_MODE"; + case ICE_ERR_IN_USE: + return "ICE_ERR_IN_USE"; + case ICE_ERR_MAX_LIMIT: + return "ICE_ERR_MAX_LIMIT"; + case ICE_ERR_RESET_ONGOING: + return "ICE_ERR_RESET_ONGOING"; + case ICE_ERR_HW_TABLE: + return "ICE_ERR_HW_TABLE"; + case ICE_ERR_DOES_NOT_EXIST: + return "ICE_ERR_DOES_NOT_EXIST"; + case ICE_ERR_FW_DDP_MISMATCH: + return "ICE_ERR_FW_DDP_MISMATCH"; + case ICE_ERR_AQ_ERROR: + return "ICE_ERR_AQ_ERROR"; + case ICE_ERR_AQ_TIMEOUT: + return "ICE_ERR_AQ_TIMEOUT"; + case ICE_ERR_AQ_FULL: + return "ICE_ERR_AQ_FULL"; + case ICE_ERR_AQ_NO_WORK: + return "ICE_ERR_AQ_NO_WORK"; + case ICE_ERR_AQ_EMPTY: + return "ICE_ERR_AQ_EMPTY"; + case ICE_ERR_AQ_FW_CRITICAL: + return "ICE_ERR_AQ_FW_CRITICAL"; + } + + return "ICE_ERR_UNKNOWN"; +} + +/** * ice_set_rss - Set RSS keys and lut * @vsi: Pointer to VSI structure * @seed: RSS hash seed @@ -4882,8 +5270,9 @@ int ice_set_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size) status = ice_aq_set_rss_key(hw, vsi->idx, buf); if (status) { - dev_err(dev, "Cannot set RSS key, err %d aq_err %d\n", - status, hw->adminq.rq_last_status); + dev_err(dev, "Cannot set RSS key, err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); return -EIO; } } @@ -4892,8 +5281,9 @@ int ice_set_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size) status = ice_aq_set_rss_lut(hw, vsi->idx, vsi->rss_lut_type, lut, lut_size); if (status) { - dev_err(dev, "Cannot set RSS lut, err %d aq_err %d\n", - status, hw->adminq.rq_last_status); + dev_err(dev, "Cannot set RSS lut, err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); return -EIO; } } @@ -4924,8 +5314,9 @@ int ice_get_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size) status = ice_aq_get_rss_key(hw, vsi->idx, buf); if (status) { - dev_err(dev, "Cannot get RSS key, err %d aq_err %d\n", - status, hw->adminq.rq_last_status); + dev_err(dev, "Cannot get RSS key, err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); return -EIO; } } @@ -4934,8 +5325,9 @@ int ice_get_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size) status = ice_aq_get_rss_lut(hw, vsi->idx, vsi->rss_lut_type, lut, lut_size); if (status) { - dev_err(dev, "Cannot get RSS lut, err %d aq_err %d\n", - status, hw->adminq.rq_last_status); + dev_err(dev, "Cannot get RSS lut, err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); return -EIO; } } @@ -5002,8 +5394,9 @@ static int ice_vsi_update_bridge_mode(struct ice_vsi *vsi, u16 bmode) status = ice_update_vsi(hw, vsi->idx, ctxt, NULL); if (status) { - dev_err(ice_pf_to_dev(vsi->back), "update VSI for bridge mode failed, bmode = %d err %d aq_err %d\n", - bmode, status, hw->adminq.sq_last_status); + dev_err(ice_pf_to_dev(vsi->back), "update VSI for bridge mode failed, bmode = %d err %s aq_err %s\n", + bmode, ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); ret = -EIO; goto out; } @@ -5072,8 +5465,9 @@ ice_bridge_setlink(struct net_device *dev, struct nlmsghdr *nlh, */ status = ice_update_sw_rule_bridge_mode(hw); if (status) { - netdev_err(dev, "switch rule update failed, mode = %d err %d aq_err %d\n", - mode, status, hw->adminq.sq_last_status); + netdev_err(dev, "switch rule update failed, mode = %d err %s aq_err %s\n", + mode, ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); /* revert hw->evb_veb */ hw->evb_veb = (pf_sw->bridge_mode == BRIDGE_MODE_VEB); return -EIO; @@ -5100,6 +5494,16 @@ static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue) pf->tx_timeout_count++; + /* Check if PFC is enabled for the TC to which the queue belongs + * to. If yes then Tx timeout is not caused by a hung queue, no + * need to reset and rebuild + */ + if (ice_is_pfc_causing_hung_q(pf, txqueue)) { + dev_info(ice_pf_to_dev(pf), "Fake Tx hang detected on queue %u, timeout caused by PFC storm\n", + txqueue); + return; + } + /* now that we have an index, find the tx_ring struct */ for (i = 0; i < vsi->num_txq; i++) if (vsi->tx_rings[i] && vsi->tx_rings[i]->desc) @@ -5158,6 +5562,70 @@ static void ice_tx_timeout(struct net_device *netdev, unsigned int txqueue) } /** + * ice_udp_tunnel_add - Get notifications about UDP tunnel ports that come up + * @netdev: This physical port's netdev + * @ti: Tunnel endpoint information + */ +static void +ice_udp_tunnel_add(struct net_device *netdev, struct udp_tunnel_info *ti) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_vsi *vsi = np->vsi; + struct ice_pf *pf = vsi->back; + enum ice_tunnel_type tnl_type; + u16 port = ntohs(ti->port); + enum ice_status status; + + switch (ti->type) { + case UDP_TUNNEL_TYPE_VXLAN: + tnl_type = TNL_VXLAN; + break; + case UDP_TUNNEL_TYPE_GENEVE: + tnl_type = TNL_GENEVE; + break; + default: + netdev_err(netdev, "Unknown tunnel type\n"); + return; + } + + status = ice_create_tunnel(&pf->hw, tnl_type, port); + if (status == ICE_ERR_OUT_OF_RANGE) + netdev_info(netdev, "Max tunneled UDP ports reached, port %d not added\n", + port); + else if (status) + netdev_err(netdev, "Error adding UDP tunnel - %s\n", + ice_stat_str(status)); +} + +/** + * ice_udp_tunnel_del - Get notifications about UDP tunnel ports that go away + * @netdev: This physical port's netdev + * @ti: Tunnel endpoint information + */ +static void +ice_udp_tunnel_del(struct net_device *netdev, struct udp_tunnel_info *ti) +{ + struct ice_netdev_priv *np = netdev_priv(netdev); + struct ice_vsi *vsi = np->vsi; + struct ice_pf *pf = vsi->back; + u16 port = ntohs(ti->port); + enum ice_status status; + bool retval; + + retval = ice_tunnel_port_in_use(&pf->hw, port, NULL); + if (!retval) { + netdev_info(netdev, "port %d not found in UDP tunnels list\n", + port); + return; + } + + status = ice_destroy_tunnel(&pf->hw, port, false); + if (status) + netdev_err(netdev, "error deleting port %d from UDP tunnels list\n", + port); +} + +/** * ice_open - Called when a network interface becomes active * @netdev: network interface device structure * @@ -5173,14 +5641,20 @@ int ice_open(struct net_device *netdev) { struct ice_netdev_priv *np = netdev_priv(netdev); struct ice_vsi *vsi = np->vsi; + struct ice_pf *pf = vsi->back; struct ice_port_info *pi; int err; - if (test_bit(__ICE_NEEDS_RESTART, vsi->back->state)) { + if (test_bit(__ICE_NEEDS_RESTART, pf->state)) { netdev_err(netdev, "driver needs to be unloaded and reloaded\n"); return -EIO; } + if (test_bit(__ICE_DOWN, pf->state)) { + netdev_err(netdev, "device is not ready yet\n"); + return -EBUSY; + } + netif_carrier_off(netdev); pi = vsi->port_info; @@ -5213,6 +5687,10 @@ int ice_open(struct net_device *netdev) if (err) netdev_err(netdev, "Failed to open VSI 0x%04X on switch 0x%04X\n", vsi->vsi_num, vsi->vsw->sw_id); + + /* Update existing tunnels information */ + udp_tunnel_get_rx_info(netdev); + return err; } @@ -5263,21 +5741,21 @@ ice_features_check(struct sk_buff *skb, features &= ~NETIF_F_GSO_MASK; len = skb_network_header(skb) - skb->data; - if (len & ~(ICE_TXD_MACLEN_MAX)) + if (len > ICE_TXD_MACLEN_MAX || len & 0x1) goto out_rm_features; len = skb_transport_header(skb) - skb_network_header(skb); - if (len & ~(ICE_TXD_IPLEN_MAX)) + if (len > ICE_TXD_IPLEN_MAX || len & 0x1) goto out_rm_features; if (skb->encapsulation) { len = skb_inner_network_header(skb) - skb_transport_header(skb); - if (len & ~(ICE_TXD_L4LEN_MAX)) + if (len > ICE_TXD_L4LEN_MAX || len & 0x1) goto out_rm_features; len = skb_inner_transport_header(skb) - skb_inner_network_header(skb); - if (len & ~(ICE_TXD_IPLEN_MAX)) + if (len > ICE_TXD_IPLEN_MAX || len & 0x1) goto out_rm_features; } @@ -5322,8 +5800,13 @@ static const struct net_device_ops ice_netdev_ops = { .ndo_bridge_setlink = ice_bridge_setlink, .ndo_fdb_add = ice_fdb_add, .ndo_fdb_del = ice_fdb_del, +#ifdef CONFIG_RFS_ACCEL + .ndo_rx_flow_steer = ice_rx_flow_steer, +#endif .ndo_tx_timeout = ice_tx_timeout, .ndo_bpf = ice_xdp, .ndo_xdp_xmit = ice_xdp_xmit, .ndo_xsk_wakeup = ice_xsk_wakeup, + .ndo_udp_tunnel_add = ice_udp_tunnel_add, + .ndo_udp_tunnel_del = ice_udp_tunnel_del, }; diff --git a/drivers/net/ethernet/intel/ice/ice_nvm.c b/drivers/net/ethernet/intel/ice/ice_nvm.c index 8beb675d676b..b049c1c30c88 100644 --- a/drivers/net/ethernet/intel/ice/ice_nvm.c +++ b/drivers/net/ethernet/intel/ice/ice_nvm.c @@ -172,7 +172,8 @@ void ice_release_nvm(struct ice_hw *hw) * * Reads one 16 bit word from the Shadow RAM using the ice_read_sr_word_aq. */ -enum ice_status ice_read_sr_word(struct ice_hw *hw, u16 offset, u16 *data) +static enum ice_status +ice_read_sr_word(struct ice_hw *hw, u16 offset, u16 *data) { enum ice_status status; @@ -196,7 +197,7 @@ enum ice_status ice_read_sr_word(struct ice_hw *hw, u16 offset, u16 *data) * Area (PFA) and returns the TLV pointer and length. The caller can * use these to read the variable length TLV value. */ -enum ice_status +static enum ice_status ice_get_pfa_module_tlv(struct ice_hw *hw, u16 *module_tlv, u16 *module_tlv_len, u16 module_type) { @@ -367,6 +368,87 @@ static enum ice_status ice_get_orom_ver_info(struct ice_hw *hw) } /** + * ice_get_netlist_ver_info + * @hw: pointer to the HW struct + * + * Get the netlist version information + */ +static enum ice_status ice_get_netlist_ver_info(struct ice_hw *hw) +{ + struct ice_netlist_ver_info *ver = &hw->netlist_ver; + enum ice_status ret; + u32 id_blk_start; + __le16 raw_data; + u16 data, i; + u16 *buff; + + ret = ice_acquire_nvm(hw, ICE_RES_READ); + if (ret) + return ret; + buff = kcalloc(ICE_AQC_NVM_NETLIST_ID_BLK_LEN, sizeof(*buff), + GFP_KERNEL); + if (!buff) { + ret = ICE_ERR_NO_MEMORY; + goto exit_no_mem; + } + + /* read module length */ + ret = ice_aq_read_nvm(hw, ICE_AQC_NVM_LINK_TOPO_NETLIST_MOD_ID, + ICE_AQC_NVM_LINK_TOPO_NETLIST_LEN_OFFSET * 2, + ICE_AQC_NVM_LINK_TOPO_NETLIST_LEN, &raw_data, + false, false, NULL); + if (ret) + goto exit_error; + + data = le16_to_cpu(raw_data); + /* exit if length is = 0 */ + if (!data) + goto exit_error; + + /* read node count */ + ret = ice_aq_read_nvm(hw, ICE_AQC_NVM_LINK_TOPO_NETLIST_MOD_ID, + ICE_AQC_NVM_NETLIST_NODE_COUNT_OFFSET * 2, + ICE_AQC_NVM_NETLIST_NODE_COUNT_LEN, &raw_data, + false, false, NULL); + if (ret) + goto exit_error; + data = le16_to_cpu(raw_data) & ICE_AQC_NVM_NETLIST_NODE_COUNT_M; + + /* netlist ID block starts from offset 4 + node count * 2 */ + id_blk_start = ICE_AQC_NVM_NETLIST_ID_BLK_START_OFFSET + data * 2; + + /* read the entire netlist ID block */ + ret = ice_aq_read_nvm(hw, ICE_AQC_NVM_LINK_TOPO_NETLIST_MOD_ID, + id_blk_start * 2, + ICE_AQC_NVM_NETLIST_ID_BLK_LEN * 2, buff, false, + false, NULL); + if (ret) + goto exit_error; + + for (i = 0; i < ICE_AQC_NVM_NETLIST_ID_BLK_LEN; i++) + buff[i] = le16_to_cpu(((__force __le16 *)buff)[i]); + + ver->major = (buff[ICE_AQC_NVM_NETLIST_ID_BLK_MAJOR_VER_HIGH] << 16) | + buff[ICE_AQC_NVM_NETLIST_ID_BLK_MAJOR_VER_LOW]; + ver->minor = (buff[ICE_AQC_NVM_NETLIST_ID_BLK_MINOR_VER_HIGH] << 16) | + buff[ICE_AQC_NVM_NETLIST_ID_BLK_MINOR_VER_LOW]; + ver->type = (buff[ICE_AQC_NVM_NETLIST_ID_BLK_TYPE_HIGH] << 16) | + buff[ICE_AQC_NVM_NETLIST_ID_BLK_TYPE_LOW]; + ver->rev = (buff[ICE_AQC_NVM_NETLIST_ID_BLK_REV_HIGH] << 16) | + buff[ICE_AQC_NVM_NETLIST_ID_BLK_REV_LOW]; + ver->cust_ver = buff[ICE_AQC_NVM_NETLIST_ID_BLK_CUST_VER]; + /* Read the left most 4 bytes of SHA */ + ver->hash = buff[ICE_AQC_NVM_NETLIST_ID_BLK_SHA_HASH + 15] << 16 | + buff[ICE_AQC_NVM_NETLIST_ID_BLK_SHA_HASH + 14]; + +exit_error: + kfree(buff); +exit_no_mem: + ice_release_nvm(hw); + return ret; +} + +/** * ice_discover_flash_size - Discover the available flash size. * @hw: pointer to the HW struct * @@ -515,6 +597,11 @@ enum ice_status ice_init_nvm(struct ice_hw *hw) return status; } + /* read the netlist version information */ + status = ice_get_netlist_ver_info(hw); + if (status) + ice_debug(hw, ICE_DBG_INIT, "Failed to read netlist info.\n"); + return 0; } diff --git a/drivers/net/ethernet/intel/ice/ice_nvm.h b/drivers/net/ethernet/intel/ice/ice_nvm.h index 999f273ba6ad..165eda07b93d 100644 --- a/drivers/net/ethernet/intel/ice/ice_nvm.h +++ b/drivers/net/ethernet/intel/ice/ice_nvm.h @@ -11,10 +11,6 @@ enum ice_status ice_read_flat_nvm(struct ice_hw *hw, u32 offset, u32 *length, u8 *data, bool read_shadow_ram); enum ice_status -ice_get_pfa_module_tlv(struct ice_hw *hw, u16 *module_tlv, u16 *module_tlv_len, - u16 module_type); -enum ice_status ice_read_pba_string(struct ice_hw *hw, u8 *pba_num, u32 pba_num_size); enum ice_status ice_init_nvm(struct ice_hw *hw); -enum ice_status ice_read_sr_word(struct ice_hw *hw, u16 offset, u16 *data); #endif /* _ICE_NVM_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_protocol_type.h b/drivers/net/ethernet/intel/ice/ice_protocol_type.h index 71647566964e..7f4c1ec1eff2 100644 --- a/drivers/net/ethernet/intel/ice/ice_protocol_type.h +++ b/drivers/net/ethernet/intel/ice/ice_protocol_type.h @@ -12,12 +12,15 @@ */ enum ice_prot_id { ICE_PROT_ID_INVAL = 0, + ICE_PROT_MAC_OF_OR_S = 1, ICE_PROT_IPV4_OF_OR_S = 32, ICE_PROT_IPV4_IL = 33, ICE_PROT_IPV6_OF_OR_S = 40, ICE_PROT_IPV6_IL = 41, ICE_PROT_TCP_IL = 49, + ICE_PROT_UDP_OF = 52, ICE_PROT_UDP_IL_OR_S = 53, + ICE_PROT_GRE_OF = 64, ICE_PROT_SCTP_IL = 96, ICE_PROT_META_ID = 255, /* when offset == metadata */ ICE_PROT_INVALID = 255 /* when offset == ICE_FV_OFFSET_INVAL */ diff --git a/drivers/net/ethernet/intel/ice/ice_sched.c b/drivers/net/ethernet/intel/ice/ice_sched.c index eae707ddf8e8..0475134295e4 100644 --- a/drivers/net/ethernet/intel/ice/ice_sched.c +++ b/drivers/net/ethernet/intel/ice/ice_sched.c @@ -1714,8 +1714,7 @@ ice_sched_cfg_vsi(struct ice_port_info *pi, u16 vsi_handle, u8 tc, u16 maxqs, * This function removes single aggregator VSI info entry from * aggregator list. */ -static void -ice_sched_rm_agg_vsi_info(struct ice_port_info *pi, u16 vsi_handle) +static void ice_sched_rm_agg_vsi_info(struct ice_port_info *pi, u16 vsi_handle) { struct ice_sched_agg_info *agg_info; struct ice_sched_agg_info *atmp; @@ -1917,7 +1916,7 @@ ice_sched_update_elem(struct ice_hw *hw, struct ice_sched_node *node, */ static enum ice_status ice_sched_cfg_node_bw_alloc(struct ice_hw *hw, struct ice_sched_node *node, - enum ice_rl_type rl_type, u8 bw_alloc) + enum ice_rl_type rl_type, u16 bw_alloc) { struct ice_aqc_txsched_elem_data buf; struct ice_aqc_txsched_elem *data; @@ -1947,8 +1946,7 @@ ice_sched_cfg_node_bw_alloc(struct ice_hw *hw, struct ice_sched_node *node, * * Save or clear CIR bandwidth (BW) in the passed param bw_t_info. */ -static void -ice_set_clear_cir_bw(struct ice_bw_type_info *bw_t_info, u32 bw) +static void ice_set_clear_cir_bw(struct ice_bw_type_info *bw_t_info, u32 bw) { if (bw == ICE_SCHED_DFLT_BW) { clear_bit(ICE_BW_TYPE_CIR, bw_t_info->bw_t_bitmap); @@ -1967,8 +1965,7 @@ ice_set_clear_cir_bw(struct ice_bw_type_info *bw_t_info, u32 bw) * * Save or clear EIR bandwidth (BW) in the passed param bw_t_info. */ -static void -ice_set_clear_eir_bw(struct ice_bw_type_info *bw_t_info, u32 bw) +static void ice_set_clear_eir_bw(struct ice_bw_type_info *bw_t_info, u32 bw) { if (bw == ICE_SCHED_DFLT_BW) { clear_bit(ICE_BW_TYPE_EIR, bw_t_info->bw_t_bitmap); @@ -1993,8 +1990,7 @@ ice_set_clear_eir_bw(struct ice_bw_type_info *bw_t_info, u32 bw) * * Save or clear shared bandwidth (BW) in the passed param bw_t_info. */ -static void -ice_set_clear_shared_bw(struct ice_bw_type_info *bw_t_info, u32 bw) +static void ice_set_clear_shared_bw(struct ice_bw_type_info *bw_t_info, u32 bw) { if (bw == ICE_SCHED_DFLT_BW) { clear_bit(ICE_BW_TYPE_SHARED, bw_t_info->bw_t_bitmap); diff --git a/drivers/net/ethernet/intel/ice/ice_status.h b/drivers/net/ethernet/intel/ice/ice_status.h index a9a8bc3aca42..4028c6365172 100644 --- a/drivers/net/ethernet/intel/ice/ice_status.h +++ b/drivers/net/ethernet/intel/ice/ice_status.h @@ -27,6 +27,8 @@ enum ice_status { ICE_ERR_MAX_LIMIT = -17, ICE_ERR_RESET_ONGOING = -18, ICE_ERR_HW_TABLE = -19, + ICE_ERR_FW_DDP_MISMATCH = -20, + ICE_ERR_NVM_CHECKSUM = -51, ICE_ERR_BUF_TOO_SHORT = -52, ICE_ERR_NVM_BLANK_MODE = -53, @@ -35,6 +37,7 @@ enum ice_status { ICE_ERR_AQ_FULL = -102, ICE_ERR_AQ_NO_WORK = -103, ICE_ERR_AQ_EMPTY = -104, + ICE_ERR_AQ_FW_CRITICAL = -105, }; #endif /* _ICE_STATUS_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c index 51825a203e35..ff7d16ac693e 100644 --- a/drivers/net/ethernet/intel/ice/ice_switch.c +++ b/drivers/net/ethernet/intel/ice/ice_switch.c @@ -593,8 +593,8 @@ enum ice_status ice_get_initial_sw_cfg(struct ice_hw *hw) ICE_AQC_GET_SW_CONF_RESP_IS_VF) is_vf = true; - res_type = le16_to_cpu(ele->vsi_port_num) >> - ICE_AQC_GET_SW_CONF_RESP_TYPE_S; + res_type = (u8)(le16_to_cpu(ele->vsi_port_num) >> + ICE_AQC_GET_SW_CONF_RESP_TYPE_S); if (res_type == ICE_AQC_GET_SW_CONF_RESP_VSI) { /* FW VSI is not needed. Just continue. */ @@ -1612,18 +1612,17 @@ exit: * check for duplicates in this case, removing duplicates from a given * list should be taken care of in the caller of this function. */ -enum ice_status -ice_add_mac(struct ice_hw *hw, struct list_head *m_list) +enum ice_status ice_add_mac(struct ice_hw *hw, struct list_head *m_list) { struct ice_aqc_sw_rules_elem *s_rule, *r_iter; struct ice_fltr_list_entry *m_list_itr; struct list_head *rule_head; - u16 elem_sent, total_elem_left; + u16 total_elem_left, s_rule_size; struct ice_switch_info *sw; struct mutex *rule_lock; /* Lock to protect filter rule list */ enum ice_status status = 0; u16 num_unicast = 0; - u16 s_rule_size; + u8 elem_sent; if (!m_list || !hw) return ICE_ERR_PARAM; @@ -1707,8 +1706,8 @@ ice_add_mac(struct ice_hw *hw, struct list_head *m_list) total_elem_left -= elem_sent) { struct ice_aqc_sw_rules_elem *entry = r_iter; - elem_sent = min(total_elem_left, - (u16)(ICE_AQ_MAX_BUF_LEN / s_rule_size)); + elem_sent = min_t(u8, total_elem_left, + (ICE_AQ_MAX_BUF_LEN / s_rule_size)); status = ice_aq_sw_rules(hw, entry, elem_sent * s_rule_size, elem_sent, ice_aqc_opc_add_sw_rules, NULL); @@ -1914,8 +1913,7 @@ exit: * @hw: pointer to the hardware structure * @v_list: list of VLAN entries and forwarding information */ -enum ice_status -ice_add_vlan(struct ice_hw *hw, struct list_head *v_list) +enum ice_status ice_add_vlan(struct ice_hw *hw, struct list_head *v_list) { struct ice_fltr_list_entry *v_list_itr; @@ -2145,8 +2143,7 @@ ice_find_ucast_rule_entry(struct ice_hw *hw, u8 recp_id, * the entries passed into m_list were added previously. It will not attempt to * do a partial remove of entries that were found. */ -enum ice_status -ice_remove_mac(struct ice_hw *hw, struct list_head *m_list) +enum ice_status ice_remove_mac(struct ice_hw *hw, struct list_head *m_list) { struct ice_fltr_list_entry *list_itr, *tmp; struct mutex *rule_lock; /* Lock to protect filter rule list */ @@ -2678,6 +2675,81 @@ void ice_remove_vsi_fltr(struct ice_hw *hw, u16 vsi_handle) } /** + * ice_alloc_res_cntr - allocating resource counter + * @hw: pointer to the hardware structure + * @type: type of resource + * @alloc_shared: if set it is shared else dedicated + * @num_items: number of entries requested for FD resource type + * @counter_id: counter index returned by AQ call + */ +enum ice_status +ice_alloc_res_cntr(struct ice_hw *hw, u8 type, u8 alloc_shared, u16 num_items, + u16 *counter_id) +{ + struct ice_aqc_alloc_free_res_elem *buf; + enum ice_status status; + u16 buf_len; + + /* Allocate resource */ + buf_len = sizeof(*buf); + buf = kzalloc(buf_len, GFP_KERNEL); + if (!buf) + return ICE_ERR_NO_MEMORY; + + buf->num_elems = cpu_to_le16(num_items); + buf->res_type = cpu_to_le16(((type << ICE_AQC_RES_TYPE_S) & + ICE_AQC_RES_TYPE_M) | alloc_shared); + + status = ice_aq_alloc_free_res(hw, 1, buf, buf_len, + ice_aqc_opc_alloc_res, NULL); + if (status) + goto exit; + + *counter_id = le16_to_cpu(buf->elem[0].e.sw_resp); + +exit: + kfree(buf); + return status; +} + +/** + * ice_free_res_cntr - free resource counter + * @hw: pointer to the hardware structure + * @type: type of resource + * @alloc_shared: if set it is shared else dedicated + * @num_items: number of entries to be freed for FD resource type + * @counter_id: counter ID resource which needs to be freed + */ +enum ice_status +ice_free_res_cntr(struct ice_hw *hw, u8 type, u8 alloc_shared, u16 num_items, + u16 counter_id) +{ + struct ice_aqc_alloc_free_res_elem *buf; + enum ice_status status; + u16 buf_len; + + /* Free resource */ + buf_len = sizeof(*buf); + buf = kzalloc(buf_len, GFP_KERNEL); + if (!buf) + return ICE_ERR_NO_MEMORY; + + buf->num_elems = cpu_to_le16(num_items); + buf->res_type = cpu_to_le16(((type << ICE_AQC_RES_TYPE_S) & + ICE_AQC_RES_TYPE_M) | alloc_shared); + buf->elem[0].e.sw_resp = cpu_to_le16(counter_id); + + status = ice_aq_alloc_free_res(hw, 1, buf, buf_len, + ice_aqc_opc_free_res, NULL); + if (status) + ice_debug(hw, ICE_DBG_SW, + "counter resource could not be freed\n"); + + kfree(buf); + return status; +} + +/** * ice_replay_vsi_fltr - Replay filters for requested VSI * @hw: pointer to the hardware structure * @vsi_handle: driver VSI handle diff --git a/drivers/net/ethernet/intel/ice/ice_switch.h b/drivers/net/ethernet/intel/ice/ice_switch.h index fa14b9545dab..8b4f9d35c860 100644 --- a/drivers/net/ethernet/intel/ice/ice_switch.h +++ b/drivers/net/ethernet/intel/ice/ice_switch.h @@ -208,6 +208,13 @@ void ice_clear_all_vsi_ctx(struct ice_hw *hw); /* Switch config */ enum ice_status ice_get_initial_sw_cfg(struct ice_hw *hw); +enum ice_status +ice_alloc_res_cntr(struct ice_hw *hw, u8 type, u8 alloc_shared, u16 num_items, + u16 *counter_id); +enum ice_status +ice_free_res_cntr(struct ice_hw *hw, u8 type, u8 alloc_shared, u16 num_items, + u16 counter_id); + /* Switch/bridge related commands */ enum ice_status ice_update_sw_rule_bridge_mode(struct ice_hw *hw); enum ice_status ice_add_mac(struct ice_hw *hw, struct list_head *m_lst); diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c index f67e8362958c..abdb137c8bb7 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx.c @@ -15,6 +15,90 @@ #define ICE_RX_HDR_SIZE 256 +#define FDIR_DESC_RXDID 0x40 +#define ICE_FDIR_CLEAN_DELAY 10 + +/** + * ice_prgm_fdir_fltr - Program a Flow Director filter + * @vsi: VSI to send dummy packet + * @fdir_desc: flow director descriptor + * @raw_packet: allocated buffer for flow director + */ +int +ice_prgm_fdir_fltr(struct ice_vsi *vsi, struct ice_fltr_desc *fdir_desc, + u8 *raw_packet) +{ + struct ice_tx_buf *tx_buf, *first; + struct ice_fltr_desc *f_desc; + struct ice_tx_desc *tx_desc; + struct ice_ring *tx_ring; + struct device *dev; + dma_addr_t dma; + u32 td_cmd; + u16 i; + + /* VSI and Tx ring */ + if (!vsi) + return -ENOENT; + tx_ring = vsi->tx_rings[0]; + if (!tx_ring || !tx_ring->desc) + return -ENOENT; + dev = tx_ring->dev; + + /* we are using two descriptors to add/del a filter and we can wait */ + for (i = ICE_FDIR_CLEAN_DELAY; ICE_DESC_UNUSED(tx_ring) < 2; i--) { + if (!i) + return -EAGAIN; + msleep_interruptible(1); + } + + dma = dma_map_single(dev, raw_packet, ICE_FDIR_MAX_RAW_PKT_SIZE, + DMA_TO_DEVICE); + + if (dma_mapping_error(dev, dma)) + return -EINVAL; + + /* grab the next descriptor */ + i = tx_ring->next_to_use; + first = &tx_ring->tx_buf[i]; + f_desc = ICE_TX_FDIRDESC(tx_ring, i); + memcpy(f_desc, fdir_desc, sizeof(*f_desc)); + + i++; + i = (i < tx_ring->count) ? i : 0; + tx_desc = ICE_TX_DESC(tx_ring, i); + tx_buf = &tx_ring->tx_buf[i]; + + i++; + tx_ring->next_to_use = (i < tx_ring->count) ? i : 0; + + memset(tx_buf, 0, sizeof(*tx_buf)); + dma_unmap_len_set(tx_buf, len, ICE_FDIR_MAX_RAW_PKT_SIZE); + dma_unmap_addr_set(tx_buf, dma, dma); + + tx_desc->buf_addr = cpu_to_le64(dma); + td_cmd = ICE_TXD_LAST_DESC_CMD | ICE_TX_DESC_CMD_DUMMY | + ICE_TX_DESC_CMD_RE; + + tx_buf->tx_flags = ICE_TX_FLAGS_DUMMY_PKT; + tx_buf->raw_buf = raw_packet; + + tx_desc->cmd_type_offset_bsz = + ice_build_ctob(td_cmd, 0, ICE_FDIR_MAX_RAW_PKT_SIZE, 0); + + /* Force memory write to complete before letting h/w know + * there are new descriptors to fetch. + */ + wmb(); + + /* mark the data descriptor to be watched */ + first->next_to_watch = tx_desc; + + writel(tx_ring->next_to_use, tx_ring->tail); + + return 0; +} + /** * ice_unmap_and_free_tx_buf - Release a Tx buffer * @ring: the ring that owns the buffer @@ -24,7 +108,9 @@ static void ice_unmap_and_free_tx_buf(struct ice_ring *ring, struct ice_tx_buf *tx_buf) { if (tx_buf->skb) { - if (ice_ring_is_xdp(ring)) + if (tx_buf->tx_flags & ICE_TX_FLAGS_DUMMY_PKT) + devm_kfree(ring->dev, tx_buf->raw_buf); + else if (ice_ring_is_xdp(ring)) page_frag_free(tx_buf->raw_buf); else dev_kfree_skb_any(tx_buf->skb); @@ -423,6 +509,22 @@ static unsigned int ice_rx_offset(struct ice_ring *rx_ring) return 0; } +static unsigned int ice_rx_frame_truesize(struct ice_ring *rx_ring, + unsigned int size) +{ + unsigned int truesize; + +#if (PAGE_SIZE < 8192) + truesize = ice_rx_pg_size(rx_ring) / 2; /* Must be power-of-2 */ +#else + truesize = ice_rx_offset(rx_ring) ? + SKB_DATA_ALIGN(ice_rx_offset(rx_ring) + size) + + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) : + SKB_DATA_ALIGN(size); +#endif + return truesize; +} + /** * ice_run_xdp - Executes an XDP program on initialized xdp_buff * @rx_ring: Rx ring @@ -583,7 +685,8 @@ bool ice_alloc_rx_bufs(struct ice_ring *rx_ring, u16 cleaned_count) struct ice_rx_buf *bi; /* do nothing if no valid netdev defined */ - if (!rx_ring->netdev || !cleaned_count) + if ((!rx_ring->netdev && rx_ring->vsi->type != ICE_VSI_CTRL) || + !cleaned_count) return false; /* get the Rx descriptor and buffer based on next_to_use */ @@ -803,7 +906,7 @@ static struct sk_buff * ice_build_skb(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf, struct xdp_buff *xdp) { - unsigned int metasize = xdp->data - xdp->data_meta; + u8 metasize = xdp->data - xdp->data_meta; #if (PAGE_SIZE < 8192) unsigned int truesize = ice_rx_pg_size(rx_ring) / 2; #else @@ -918,7 +1021,7 @@ ice_construct_skb(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf, */ static void ice_put_rx_buf(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf) { - u32 ntc = rx_ring->next_to_clean + 1; + u16 ntc = rx_ring->next_to_clean + 1; /* fetch, update, and store next to clean */ ntc = (ntc < rx_ring->count) ? ntc : 0; @@ -981,7 +1084,7 @@ ice_is_non_eop(struct ice_ring *rx_ring, union ice_32b_rx_flex_desc *rx_desc, * * Returns amount of work completed */ -static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) +int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) { unsigned int total_rx_bytes = 0, total_rx_pkts = 0; u16 cleaned_count = ICE_DESC_UNUSED(rx_ring); @@ -991,6 +1094,10 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) bool failure; xdp.rxq = &rx_ring->xdp_rxq; + /* Frame size depend on rx_ring setup when PAGE_SIZE=4K */ +#if (PAGE_SIZE < 8192) + xdp.frame_sz = ice_rx_frame_truesize(rx_ring, 0); +#endif /* start the loop to process Rx packets bounded by 'budget' */ while (likely(total_rx_pkts < (unsigned int)budget)) { @@ -1020,6 +1127,12 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) */ dma_rmb(); + if (rx_desc->wb.rxdid == FDIR_DESC_RXDID || !rx_ring->netdev) { + ice_put_rx_buf(rx_ring, NULL); + cleaned_count++; + continue; + } + size = le16_to_cpu(rx_desc->wb.pkt_len) & ICE_RX_FLX_DESC_PKT_LEN_M; @@ -1038,6 +1151,10 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) xdp.data_hard_start = xdp.data - ice_rx_offset(rx_ring); xdp.data_meta = xdp.data; xdp.data_end = xdp.data + size; +#if (PAGE_SIZE > 4096) + /* At larger PAGE_SIZE, frame_sz depend on len size */ + xdp.frame_sz = ice_rx_frame_truesize(rx_ring, size); +#endif rcu_read_lock(); xdp_prog = READ_ONCE(rx_ring->xdp_prog); @@ -1051,16 +1168,8 @@ static int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget) if (!xdp_res) goto construct_skb; if (xdp_res & (ICE_XDP_TX | ICE_XDP_REDIR)) { - unsigned int truesize; - -#if (PAGE_SIZE < 8192) - truesize = ice_rx_pg_size(rx_ring) / 2; -#else - truesize = SKB_DATA_ALIGN(ice_rx_offset(rx_ring) + - size); -#endif xdp_xmit |= xdp_res; - ice_rx_buf_adjust_pg_offset(rx_buf, truesize); + ice_rx_buf_adjust_pg_offset(rx_buf, xdp.frame_sz); } else { rx_buf->pagecnt_bias++; } @@ -1528,7 +1637,7 @@ int ice_napi_poll(struct napi_struct *napi, int budget) * don't allow the budget to go below 1 because that would exit * polling early. */ - budget_per_ring = max(budget / q_vector->num_ring_rx, 1); + budget_per_ring = max_t(int, budget / q_vector->num_ring_rx, 1); else /* Max of 1 Rx ring in this q_vector so give it the budget */ budget_per_ring = budget; @@ -1664,7 +1773,8 @@ ice_tx_map(struct ice_ring *tx_ring, struct ice_tx_buf *first, */ while (unlikely(size > ICE_MAX_DATA_PER_TXD)) { tx_desc->cmd_type_offset_bsz = - build_ctob(td_cmd, td_offset, max_data, td_tag); + ice_build_ctob(td_cmd, td_offset, max_data, + td_tag); tx_desc++; i++; @@ -1684,8 +1794,8 @@ ice_tx_map(struct ice_ring *tx_ring, struct ice_tx_buf *first, if (likely(!data_len)) break; - tx_desc->cmd_type_offset_bsz = build_ctob(td_cmd, td_offset, - size, td_tag); + tx_desc->cmd_type_offset_bsz = ice_build_ctob(td_cmd, td_offset, + size, td_tag); tx_desc++; i++; @@ -1716,8 +1826,8 @@ ice_tx_map(struct ice_ring *tx_ring, struct ice_tx_buf *first, /* write last descriptor with RS and EOP bits */ td_cmd |= (u64)ICE_TXD_LAST_DESC_CMD; - tx_desc->cmd_type_offset_bsz = build_ctob(td_cmd, td_offset, size, - td_tag); + tx_desc->cmd_type_offset_bsz = + ice_build_ctob(td_cmd, td_offset, size, td_tag); /* Force memory writes to complete before letting h/w know there * are new descriptors to fetch. @@ -1791,12 +1901,94 @@ int ice_tx_csum(struct ice_tx_buf *first, struct ice_tx_offload_params *off) l2_len = ip.hdr - skb->data; offset = (l2_len / 2) << ICE_TX_DESC_LEN_MACLEN_S; - if (skb->encapsulation) - return -1; + protocol = vlan_get_protocol(skb); + + if (protocol == htons(ETH_P_IP)) + first->tx_flags |= ICE_TX_FLAGS_IPV4; + else if (protocol == htons(ETH_P_IPV6)) + first->tx_flags |= ICE_TX_FLAGS_IPV6; + + if (skb->encapsulation) { + bool gso_ena = false; + u32 tunnel = 0; + + /* define outer network header type */ + if (first->tx_flags & ICE_TX_FLAGS_IPV4) { + tunnel |= (first->tx_flags & ICE_TX_FLAGS_TSO) ? + ICE_TX_CTX_EIPT_IPV4 : + ICE_TX_CTX_EIPT_IPV4_NO_CSUM; + l4_proto = ip.v4->protocol; + } else if (first->tx_flags & ICE_TX_FLAGS_IPV6) { + tunnel |= ICE_TX_CTX_EIPT_IPV6; + exthdr = ip.hdr + sizeof(*ip.v6); + l4_proto = ip.v6->nexthdr; + if (l4.hdr != exthdr) + ipv6_skip_exthdr(skb, exthdr - skb->data, + &l4_proto, &frag_off); + } + + /* define outer transport */ + switch (l4_proto) { + case IPPROTO_UDP: + tunnel |= ICE_TXD_CTX_UDP_TUNNELING; + first->tx_flags |= ICE_TX_FLAGS_TUNNEL; + break; + case IPPROTO_GRE: + tunnel |= ICE_TXD_CTX_GRE_TUNNELING; + first->tx_flags |= ICE_TX_FLAGS_TUNNEL; + break; + case IPPROTO_IPIP: + case IPPROTO_IPV6: + first->tx_flags |= ICE_TX_FLAGS_TUNNEL; + l4.hdr = skb_inner_network_header(skb); + break; + default: + if (first->tx_flags & ICE_TX_FLAGS_TSO) + return -1; + + skb_checksum_help(skb); + return 0; + } + + /* compute outer L3 header size */ + tunnel |= ((l4.hdr - ip.hdr) / 4) << + ICE_TXD_CTX_QW0_EIPLEN_S; + + /* switch IP header pointer from outer to inner header */ + ip.hdr = skb_inner_network_header(skb); + + /* compute tunnel header size */ + tunnel |= ((ip.hdr - l4.hdr) / 2) << + ICE_TXD_CTX_QW0_NATLEN_S; + + gso_ena = skb_shinfo(skb)->gso_type & SKB_GSO_PARTIAL; + /* indicate if we need to offload outer UDP header */ + if ((first->tx_flags & ICE_TX_FLAGS_TSO) && !gso_ena && + (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM)) + tunnel |= ICE_TXD_CTX_QW0_L4T_CS_M; + + /* record tunnel offload values */ + off->cd_tunnel_params |= tunnel; + + /* set DTYP=1 to indicate that it's an Tx context descriptor + * in IPsec tunnel mode with Tx offloads in Quad word 1 + */ + off->cd_qw1 |= (u64)ICE_TX_DESC_DTYPE_CTX; + + /* switch L4 header pointer from outer to inner */ + l4.hdr = skb_inner_transport_header(skb); + l4_proto = 0; + + /* reset type as we transition from outer to inner headers */ + first->tx_flags &= ~(ICE_TX_FLAGS_IPV4 | ICE_TX_FLAGS_IPV6); + if (ip.v4->version == 4) + first->tx_flags |= ICE_TX_FLAGS_IPV4; + if (ip.v6->version == 6) + first->tx_flags |= ICE_TX_FLAGS_IPV6; + } /* Enable IP checksum offloads */ - protocol = vlan_get_protocol(skb); - if (protocol == htons(ETH_P_IP)) { + if (first->tx_flags & ICE_TX_FLAGS_IPV4) { l4_proto = ip.v4->protocol; /* the stack computes the IP header already, the only time we * need the hardware to recompute it is in the case of TSO. @@ -1806,7 +1998,7 @@ int ice_tx_csum(struct ice_tx_buf *first, struct ice_tx_offload_params *off) else cmd |= ICE_TX_DESC_CMD_IIPT_IPV4; - } else if (protocol == htons(ETH_P_IPV6)) { + } else if (first->tx_flags & ICE_TX_FLAGS_IPV6) { cmd |= ICE_TX_DESC_CMD_IIPT_IPV6; exthdr = ip.hdr + sizeof(*ip.v6); l4_proto = ip.v6->nexthdr; @@ -1861,49 +2053,25 @@ int ice_tx_csum(struct ice_tx_buf *first, struct ice_tx_offload_params *off) * * Checks the skb and set up correspondingly several generic transmit flags * related to VLAN tagging for the HW, such as VLAN, DCB, etc. - * - * Returns error code indicate the frame should be dropped upon error and the - * otherwise returns 0 to indicate the flags has been set properly. */ -static int +static void ice_tx_prepare_vlan_flags(struct ice_ring *tx_ring, struct ice_tx_buf *first) { struct sk_buff *skb = first->skb; - __be16 protocol = skb->protocol; - - if (protocol == htons(ETH_P_8021Q) && - !(tx_ring->netdev->features & NETIF_F_HW_VLAN_CTAG_TX)) { - /* when HW VLAN acceleration is turned off by the user the - * stack sets the protocol to 8021q so that the driver - * can take any steps required to support the SW only - * VLAN handling. In our case the driver doesn't need - * to take any further steps so just set the protocol - * to the encapsulated ethertype. - */ - skb->protocol = vlan_get_protocol(skb); - return 0; - } - /* if we have a HW VLAN tag being added, default to the HW one */ + /* nothing left to do, software offloaded VLAN */ + if (!skb_vlan_tag_present(skb) && eth_type_vlan(skb->protocol)) + return; + + /* currently, we always assume 802.1Q for VLAN insertion as VLAN + * insertion for 802.1AD is not supported + */ if (skb_vlan_tag_present(skb)) { first->tx_flags |= skb_vlan_tag_get(skb) << ICE_TX_FLAGS_VLAN_S; first->tx_flags |= ICE_TX_FLAGS_HW_VLAN; - } else if (protocol == htons(ETH_P_8021Q)) { - struct vlan_hdr *vhdr, _vhdr; - - /* for SW VLAN, check the next protocol and store the tag */ - vhdr = (struct vlan_hdr *)skb_header_pointer(skb, ETH_HLEN, - sizeof(_vhdr), - &_vhdr); - if (!vhdr) - return -EINVAL; - - first->tx_flags |= ntohs(vhdr->h_vlan_TCI) << - ICE_TX_FLAGS_VLAN_S; - first->tx_flags |= ICE_TX_FLAGS_SW_VLAN; } - return ice_tx_prepare_vlan_flags_dcb(tx_ring, first); + ice_tx_prepare_vlan_flags_dcb(tx_ring, first); } /** @@ -1928,7 +2096,8 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off) unsigned char *hdr; } l4; u64 cd_mss, cd_tso_len; - u32 paylen, l4_start; + u32 paylen; + u8 l4_start; int err; if (skb->ip_summed != CHECKSUM_PARTIAL) @@ -1953,8 +2122,42 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off) ip.v6->payload_len = 0; } + if (skb_shinfo(skb)->gso_type & (SKB_GSO_GRE | + SKB_GSO_GRE_CSUM | + SKB_GSO_IPXIP4 | + SKB_GSO_IPXIP6 | + SKB_GSO_UDP_TUNNEL | + SKB_GSO_UDP_TUNNEL_CSUM)) { + if (!(skb_shinfo(skb)->gso_type & SKB_GSO_PARTIAL) && + (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_TUNNEL_CSUM)) { + l4.udp->len = 0; + + /* determine offset of outer transport header */ + l4_start = (u8)(l4.hdr - skb->data); + + /* remove payload length from outer checksum */ + paylen = skb->len - l4_start; + csum_replace_by_diff(&l4.udp->check, + (__force __wsum)htonl(paylen)); + } + + /* reset pointers to inner headers */ + + /* cppcheck-suppress unreadVariable */ + ip.hdr = skb_inner_network_header(skb); + l4.hdr = skb_inner_transport_header(skb); + + /* initialize inner IP header fields */ + if (ip.v4->version == 4) { + ip.v4->tot_len = 0; + ip.v4->check = 0; + } else { + ip.v6->payload_len = 0; + } + } + /* determine offset of transport header */ - l4_start = l4.hdr - skb->data; + l4_start = (u8)(l4.hdr - skb->data); /* remove payload length from checksum */ paylen = skb->len - l4_start; @@ -1963,12 +2166,12 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off) csum_replace_by_diff(&l4.udp->check, (__force __wsum)htonl(paylen)); /* compute length of UDP segmentation header */ - off->header_len = sizeof(l4.udp) + l4_start; + off->header_len = (u8)sizeof(l4.udp) + l4_start; } else { csum_replace_by_diff(&l4.tcp->check, (__force __wsum)htonl(paylen)); /* compute length of TCP segmentation header */ - off->header_len = (l4.tcp->doff * 4) + l4_start; + off->header_len = (u8)((l4.tcp->doff * 4) + l4_start); } /* update gso_segs and bytecount */ @@ -2176,8 +2379,7 @@ ice_xmit_frame_ring(struct sk_buff *skb, struct ice_ring *tx_ring) first->tx_flags = 0; /* prepare the VLAN tagging flags for Tx */ - if (ice_tx_prepare_vlan_flags(tx_ring, first)) - goto out_drop; + ice_tx_prepare_vlan_flags(tx_ring, first); /* set up TSO offload */ tso = ice_tso(first, &offload); @@ -2199,7 +2401,7 @@ ice_xmit_frame_ring(struct sk_buff *skb, struct ice_ring *tx_ring) if (offload.cd_qw1 & ICE_TX_DESC_DTYPE_CTX) { struct ice_tx_ctx_desc *cdesc; - int i = tx_ring->next_to_use; + u16 i = tx_ring->next_to_use; /* grab the next descriptor */ cdesc = ICE_TX_CTX_DESC(tx_ring, i); @@ -2244,3 +2446,86 @@ netdev_tx_t ice_start_xmit(struct sk_buff *skb, struct net_device *netdev) return ice_xmit_frame_ring(skb, tx_ring); } + +/** + * ice_clean_ctrl_tx_irq - interrupt handler for flow director Tx queue + * @tx_ring: tx_ring to clean + */ +void ice_clean_ctrl_tx_irq(struct ice_ring *tx_ring) +{ + struct ice_vsi *vsi = tx_ring->vsi; + s16 i = tx_ring->next_to_clean; + int budget = ICE_DFLT_IRQ_WORK; + struct ice_tx_desc *tx_desc; + struct ice_tx_buf *tx_buf; + + tx_buf = &tx_ring->tx_buf[i]; + tx_desc = ICE_TX_DESC(tx_ring, i); + i -= tx_ring->count; + + do { + struct ice_tx_desc *eop_desc = tx_buf->next_to_watch; + + /* if next_to_watch is not set then there is no pending work */ + if (!eop_desc) + break; + + /* prevent any other reads prior to eop_desc */ + smp_rmb(); + + /* if the descriptor isn't done, no work to do */ + if (!(eop_desc->cmd_type_offset_bsz & + cpu_to_le64(ICE_TX_DESC_DTYPE_DESC_DONE))) + break; + + /* clear next_to_watch to prevent false hangs */ + tx_buf->next_to_watch = NULL; + tx_desc->buf_addr = 0; + tx_desc->cmd_type_offset_bsz = 0; + + /* move past filter desc */ + tx_buf++; + tx_desc++; + i++; + if (unlikely(!i)) { + i -= tx_ring->count; + tx_buf = tx_ring->tx_buf; + tx_desc = ICE_TX_DESC(tx_ring, 0); + } + + /* unmap the data header */ + if (dma_unmap_len(tx_buf, len)) + dma_unmap_single(tx_ring->dev, + dma_unmap_addr(tx_buf, dma), + dma_unmap_len(tx_buf, len), + DMA_TO_DEVICE); + if (tx_buf->tx_flags & ICE_TX_FLAGS_DUMMY_PKT) + devm_kfree(tx_ring->dev, tx_buf->raw_buf); + + /* clear next_to_watch to prevent false hangs */ + tx_buf->raw_buf = NULL; + tx_buf->tx_flags = 0; + tx_buf->next_to_watch = NULL; + dma_unmap_len_set(tx_buf, len, 0); + tx_desc->buf_addr = 0; + tx_desc->cmd_type_offset_bsz = 0; + + /* move past eop_desc for start of next FD desc */ + tx_buf++; + tx_desc++; + i++; + if (unlikely(!i)) { + i -= tx_ring->count; + tx_buf = tx_ring->tx_buf; + tx_desc = ICE_TX_DESC(tx_ring, 0); + } + + budget--; + } while (likely(budget)); + + i += tx_ring->count; + tx_ring->next_to_clean = i; + + /* re-enable interrupt if needed */ + ice_irq_dynamic_ena(&vsi->back->hw, vsi, vsi->q_vectors[0]); +} diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h index 7ee00a128663..e70c4619edc3 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx.h @@ -38,7 +38,8 @@ */ #if (PAGE_SIZE < 8192) #define ICE_2K_TOO_SMALL_WITH_PADDING \ -((NET_SKB_PAD + ICE_RXBUF_1536) > SKB_WITH_OVERHEAD(ICE_RXBUF_2048)) + ((unsigned int)(NET_SKB_PAD + ICE_RXBUF_1536) > \ + SKB_WITH_OVERHEAD(ICE_RXBUF_2048)) /** * ice_compute_pad - compute the padding @@ -107,12 +108,19 @@ static inline int ice_skb_pad(void) #define DESC_NEEDED (MAX_SKB_FRAGS + ICE_DESCS_FOR_CTX_DESC + \ ICE_DESCS_PER_CACHE_LINE + ICE_DESCS_FOR_SKB_DATA_PTR) #define ICE_DESC_UNUSED(R) \ - ((((R)->next_to_clean > (R)->next_to_use) ? 0 : (R)->count) + \ - (R)->next_to_clean - (R)->next_to_use - 1) + (u16)((((R)->next_to_clean > (R)->next_to_use) ? 0 : (R)->count) + \ + (R)->next_to_clean - (R)->next_to_use - 1) #define ICE_TX_FLAGS_TSO BIT(0) #define ICE_TX_FLAGS_HW_VLAN BIT(1) #define ICE_TX_FLAGS_SW_VLAN BIT(2) +/* ICE_TX_FLAGS_DUMMY_PKT is used to mark dummy packets that should be + * freed instead of returned like skb packets. + */ +#define ICE_TX_FLAGS_DUMMY_PKT BIT(3) +#define ICE_TX_FLAGS_IPV4 BIT(5) +#define ICE_TX_FLAGS_IPV6 BIT(6) +#define ICE_TX_FLAGS_TUNNEL BIT(7) #define ICE_TX_FLAGS_VLAN_M 0xffff0000 #define ICE_TX_FLAGS_VLAN_PR_M 0xe0000000 #define ICE_TX_FLAGS_VLAN_PR_S 29 @@ -155,17 +163,16 @@ struct ice_tx_offload_params { }; struct ice_rx_buf { - struct sk_buff *skb; - dma_addr_t dma; union { struct { + struct sk_buff *skb; + dma_addr_t dma; struct page *page; unsigned int page_offset; u16 pagecnt_bias; }; struct { - void *addr; - u64 handle; + struct xdp_buff *xdp; }; }; }; @@ -289,7 +296,6 @@ struct ice_ring { struct rcu_head rcu; /* to avoid race on free */ struct bpf_prog *xdp_prog; struct xdp_umem *xsk_umem; - struct zero_copy_allocator zca; /* CL3 - 3rd cacheline starts here */ struct xdp_rxq_info xdp_rxq; /* CLX - the below items are only accessed infrequently and should be @@ -373,5 +379,9 @@ int ice_setup_rx_ring(struct ice_ring *rx_ring); void ice_free_tx_ring(struct ice_ring *tx_ring); void ice_free_rx_ring(struct ice_ring *rx_ring); int ice_napi_poll(struct napi_struct *napi, int budget); - +int +ice_prgm_fdir_fltr(struct ice_vsi *vsi, struct ice_fltr_desc *fdir_desc, + u8 *raw_packet); +int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget); +void ice_clean_ctrl_tx_irq(struct ice_ring *tx_ring); #endif /* _ICE_TXRX_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_txrx_lib.c b/drivers/net/ethernet/intel/ice/ice_txrx_lib.c index 6da048a6ca7c..02b12736ea80 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_txrx_lib.c @@ -8,7 +8,7 @@ * @rx_ring: ring to bump * @val: new head index */ -void ice_release_rx_desc(struct ice_ring *rx_ring, u32 val) +void ice_release_rx_desc(struct ice_ring *rx_ring, u16 val) { u16 prev_ntu = rx_ring->next_to_use & ~0x7; @@ -84,11 +84,11 @@ ice_rx_csum(struct ice_ring *ring, struct sk_buff *skb, union ice_32b_rx_flex_desc *rx_desc, u8 ptype) { struct ice_rx_ptype_decoded decoded; - u32 rx_error, rx_status; + u16 rx_status0, rx_status1; bool ipv4, ipv6; - rx_status = le16_to_cpu(rx_desc->wb.status_error0); - rx_error = rx_status; + rx_status0 = le16_to_cpu(rx_desc->wb.status_error0); + rx_status1 = le16_to_cpu(rx_desc->wb.status_error1); decoded = ice_decode_rx_desc_ptype(ptype); @@ -101,7 +101,7 @@ ice_rx_csum(struct ice_ring *ring, struct sk_buff *skb, return; /* check if HW has decoded the packet and checksum */ - if (!(rx_status & BIT(ICE_RX_FLEX_DESC_STATUS0_L3L4P_S))) + if (!(rx_status0 & BIT(ICE_RX_FLEX_DESC_STATUS0_L3L4P_S))) return; if (!(decoded.known && decoded.outer_ip)) @@ -112,19 +112,31 @@ ice_rx_csum(struct ice_ring *ring, struct sk_buff *skb, ipv6 = (decoded.outer_ip == ICE_RX_PTYPE_OUTER_IP) && (decoded.outer_ip_ver == ICE_RX_PTYPE_OUTER_IPV6); - if (ipv4 && (rx_error & (BIT(ICE_RX_FLEX_DESC_STATUS0_XSUM_IPE_S) | - BIT(ICE_RX_FLEX_DESC_STATUS0_XSUM_EIPE_S)))) + if (ipv4 && (rx_status0 & (BIT(ICE_RX_FLEX_DESC_STATUS0_XSUM_IPE_S) | + BIT(ICE_RX_FLEX_DESC_STATUS0_XSUM_EIPE_S)))) goto checksum_fail; - else if (ipv6 && (rx_status & - (BIT(ICE_RX_FLEX_DESC_STATUS0_IPV6EXADD_S)))) + + if (ipv6 && (rx_status0 & (BIT(ICE_RX_FLEX_DESC_STATUS0_IPV6EXADD_S)))) goto checksum_fail; /* check for L4 errors and handle packets that were not able to be * checksummed due to arrival speed */ - if (rx_error & BIT(ICE_RX_FLEX_DESC_STATUS0_XSUM_L4E_S)) + if (rx_status0 & BIT(ICE_RX_FLEX_DESC_STATUS0_XSUM_L4E_S)) + goto checksum_fail; + + /* check for outer UDP checksum error in tunneled packets */ + if ((rx_status1 & BIT(ICE_RX_FLEX_DESC_STATUS1_NAT_S)) && + (rx_status0 & BIT(ICE_RX_FLEX_DESC_STATUS0_XSUM_EUDPE_S))) goto checksum_fail; + /* If there is an outer header present that might contain a checksum + * we need to bump the checksum level by 1 to reflect the fact that + * we are indicating we validated the inner checksum. + */ + if (decoded.tunnel_type >= ICE_RX_PTYPE_TUNNEL_IP_GRENAT) + skb->csum_level = 1; + /* Only report checksum unnecessary for TCP, UDP, or SCTP */ switch (decoded.inner_prot) { case ICE_RX_PTYPE_INNER_PROT_TCP: @@ -215,8 +227,8 @@ int ice_xmit_xdp_ring(void *data, u16 size, struct ice_ring *xdp_ring) tx_desc = ICE_TX_DESC(xdp_ring, i); tx_desc->buf_addr = cpu_to_le64(dma); - tx_desc->cmd_type_offset_bsz = build_ctob(ICE_TXD_LAST_DESC_CMD, 0, - size, 0); + tx_desc->cmd_type_offset_bsz = ice_build_ctob(ICE_TXD_LAST_DESC_CMD, 0, + size, 0); /* Make certain all of the status bits have been updated * before next_to_watch is written. @@ -242,7 +254,7 @@ int ice_xmit_xdp_ring(void *data, u16 size, struct ice_ring *xdp_ring) */ int ice_xmit_xdp_buff(struct xdp_buff *xdp, struct ice_ring *xdp_ring) { - struct xdp_frame *xdpf = convert_to_xdp_frame(xdp); + struct xdp_frame *xdpf = xdp_convert_buff_to_frame(xdp); if (unlikely(!xdpf)) return ICE_XDP_CONSUMED; diff --git a/drivers/net/ethernet/intel/ice/ice_txrx_lib.h b/drivers/net/ethernet/intel/ice/ice_txrx_lib.h index ba9164dad9ae..58ff58f0f972 100644 --- a/drivers/net/ethernet/intel/ice/ice_txrx_lib.h +++ b/drivers/net/ethernet/intel/ice/ice_txrx_lib.h @@ -22,7 +22,7 @@ ice_test_staterr(union ice_32b_rx_flex_desc *rx_desc, const u16 stat_err_bits) } static inline __le64 -build_ctob(u64 td_cmd, u64 td_offset, unsigned int size, u64 td_tag) +ice_build_ctob(u64 td_cmd, u64 td_offset, unsigned int size, u64 td_tag) { return cpu_to_le64(ICE_TX_DESC_DTYPE_DATA | (td_cmd << ICE_TXD_QW1_CMD_S) | @@ -49,7 +49,7 @@ static inline void ice_xdp_ring_update_tail(struct ice_ring *xdp_ring) void ice_finalize_xdp_rx(struct ice_ring *rx_ring, unsigned int xdp_res); int ice_xmit_xdp_buff(struct xdp_buff *xdp, struct ice_ring *xdp_ring); int ice_xmit_xdp_ring(void *data, u16 size, struct ice_ring *xdp_ring); -void ice_release_rx_desc(struct ice_ring *rx_ring, u32 val); +void ice_release_rx_desc(struct ice_ring *rx_ring, u16 val); void ice_process_skb_fields(struct ice_ring *rx_ring, union ice_32b_rx_flex_desc *rx_desc, diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h index 4ce5f92fca4a..c1ad8622e65c 100644 --- a/drivers/net/ethernet/intel/ice/ice_type.h +++ b/drivers/net/ethernet/intel/ice/ice_type.h @@ -118,7 +118,8 @@ enum ice_media_type { enum ice_vsi_type { ICE_VSI_PF = 0, - ICE_VSI_VF, + ICE_VSI_VF = 1, + ICE_VSI_CTRL = 3, /* equates to ICE_VSI_PF with 1 queue pair */ ICE_VSI_LB = 6, }; @@ -161,6 +162,38 @@ struct ice_phy_info { u8 get_link_info; }; +/* protocol enumeration for filters */ +enum ice_fltr_ptype { + /* NONE - used for undef/error */ + ICE_FLTR_PTYPE_NONF_NONE = 0, + ICE_FLTR_PTYPE_NONF_IPV4_UDP, + ICE_FLTR_PTYPE_NONF_IPV4_TCP, + ICE_FLTR_PTYPE_NONF_IPV4_SCTP, + ICE_FLTR_PTYPE_NONF_IPV4_OTHER, + ICE_FLTR_PTYPE_FRAG_IPV4, + ICE_FLTR_PTYPE_NONF_IPV6_UDP, + ICE_FLTR_PTYPE_NONF_IPV6_TCP, + ICE_FLTR_PTYPE_NONF_IPV6_SCTP, + ICE_FLTR_PTYPE_NONF_IPV6_OTHER, + ICE_FLTR_PTYPE_MAX, +}; + +enum ice_fd_hw_seg { + ICE_FD_HW_SEG_NON_TUN = 0, + ICE_FD_HW_SEG_TUN, + ICE_FD_HW_SEG_MAX, +}; + +/* 2 VSI = 1 ICE_VSI_PF + 1 ICE_VSI_CTRL */ +#define ICE_MAX_FDIR_VSI_PER_FILTER 2 + +struct ice_fd_hw_prof { + struct ice_flow_seg_info *fdir_seg[ICE_FD_HW_SEG_MAX]; + int cnt; + u64 entry_h[ICE_MAX_FDIR_VSI_PER_FILTER][ICE_FD_HW_SEG_MAX]; + u16 vsi_h[ICE_MAX_FDIR_VSI_PER_FILTER]; +}; + /* Common HW capabilities for SW use */ struct ice_hw_common_caps { u32 valid_functions; @@ -197,6 +230,8 @@ struct ice_hw_func_caps { u32 num_allocd_vfs; /* Number of allocated VFs */ u32 vf_base_id; /* Logical ID of the first VF */ u32 guar_num_vsi; + u32 fd_fltr_guar; /* Number of filters guaranteed */ + u32 fd_fltr_best_effort; /* Number of best effort filters */ }; /* Device wide capabilities */ @@ -204,6 +239,7 @@ struct ice_hw_dev_caps { struct ice_hw_common_caps common_cap; u32 num_vfs_exposed; /* Total number of VFs exposed */ u32 num_vsi_allocd_to_host; /* Excluding EMP VSI */ + u32 num_flow_director_fltr; /* Number of FD filters available */ u32 num_funcs; }; @@ -259,6 +295,16 @@ struct ice_nvm_info { #define ICE_NVM_VER_LEN 32 +/* netlist version information */ +struct ice_netlist_ver_info { + u32 major; /* major high/low */ + u32 minor; /* minor high/low */ + u32 type; /* type high/low */ + u32 rev; /* revision high/low */ + u32 hash; /* SHA-1 hash word */ + u16 cust_ver; /* customer version */ +}; + /* Max number of port to queue branches w.r.t topology */ #define ICE_MAX_TRAFFIC_CLASS 8 #define ICE_TXSCHED_MAX_BRANCHES ICE_MAX_TRAFFIC_CLASS @@ -479,6 +525,8 @@ struct ice_hw { u64 debug_mask; /* bitmap for debug mask */ enum ice_mac_type mac_type; + u16 fd_ctr_base; /* FD counter base index */ + /* pci info */ u16 device_id; u16 vendor_id; @@ -491,8 +539,8 @@ struct ice_hw { u16 max_burst_size; /* driver sets this value */ /* Tx Scheduler values */ - u16 num_tx_sched_layers; - u16 num_tx_sched_phys_layers; + u8 num_tx_sched_layers; + u8 num_tx_sched_phys_layers; u8 flattened_layers; u8 max_cgds; u8 sw_entry_point_layer; @@ -506,6 +554,7 @@ struct ice_hw { struct ice_nvm_info nvm; struct ice_hw_dev_caps dev_caps; /* device capabilities */ struct ice_hw_func_caps func_caps; /* function capabilities */ + struct ice_netlist_ver_info netlist_ver; /* netlist version info */ struct ice_switch_info *switch_info; /* switch filter lists */ @@ -548,6 +597,7 @@ struct ice_hw { /* Active package version (currently active) */ struct ice_pkg_ver active_pkg_ver; + u32 active_track_id; u8 active_pkg_name[ICE_PKG_NAME_SIZE]; u8 active_pkg_in_nvm; @@ -568,10 +618,29 @@ struct ice_hw { u8 *pkg_copy; u32 pkg_size; + /* tunneling info */ + struct mutex tnl_lock; + struct ice_tunnel_table tnl; + /* HW block tables */ struct ice_blk_info blk[ICE_BLK_COUNT]; struct mutex fl_profs_locks[ICE_BLK_COUNT]; /* lock fltr profiles */ struct list_head fl_profs[ICE_BLK_COUNT]; + + /* Flow Director filter info */ + int fdir_active_fltr; + + struct mutex fdir_fltr_lock; /* protect Flow Director */ + struct list_head fdir_list_head; + + /* Book-keeping of side-band filter count per flow-type. + * This is used to detect and handle input set changes for + * respective flow-type. + */ + u16 fdir_fltr_cnt[ICE_FLTR_PTYPE_MAX]; + + struct ice_fd_hw_prof **fdir_prof; + DECLARE_BITMAP(fdir_perfect_fltr, ICE_FLTR_PTYPE_MAX); struct mutex rss_locks; /* protect RSS configuration */ struct list_head rss_list_head; }; @@ -592,6 +661,8 @@ struct ice_eth_stats { u64 tx_errors; /* tepc */ }; +#define ICE_MAX_UP 8 + /* Statistics collected by the MAC */ struct ice_hw_port_stats { /* eth stats collected by the port */ @@ -631,6 +702,9 @@ struct ice_hw_port_stats { u64 tx_size_1023; /* ptc1023 */ u64 tx_size_1522; /* ptc1522 */ u64 tx_size_big; /* ptc9522 */ + /* flow director stats */ + u32 fd_sb_status; + u64 fd_sb_match; }; /* Checksum and Shadow RAM pointers */ diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c index 15191a325918..16a2f2526ccc 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c @@ -4,16 +4,18 @@ #include "ice.h" #include "ice_base.h" #include "ice_lib.h" +#include "ice_fltr.h" /** * ice_validate_vf_id - helper to check if VF ID is valid * @pf: pointer to the PF structure * @vf_id: the ID of the VF to check */ -static int ice_validate_vf_id(struct ice_pf *pf, int vf_id) +static int ice_validate_vf_id(struct ice_pf *pf, u16 vf_id) { + /* vf_id range is only valid for 0-255, and should always be unsigned */ if (vf_id >= pf->num_alloc_vfs) { - dev_err(ice_pf_to_dev(pf), "Invalid VF ID: %d\n", vf_id); + dev_err(ice_pf_to_dev(pf), "Invalid VF ID: %u\n", vf_id); return -EINVAL; } return 0; @@ -27,7 +29,7 @@ static int ice_validate_vf_id(struct ice_pf *pf, int vf_id) static int ice_check_vf_init(struct ice_pf *pf, struct ice_vf *vf) { if (!test_bit(ICE_VF_STATE_INIT, vf->vf_states)) { - dev_err(ice_pf_to_dev(pf), "VF ID: %d in reset. Try again.\n", + dev_err(ice_pf_to_dev(pf), "VF ID: %u in reset. Try again.\n", vf->vf_id); return -EBUSY; } @@ -35,6 +37,37 @@ static int ice_check_vf_init(struct ice_pf *pf, struct ice_vf *vf) } /** + * ice_err_to_virt_err - translate errors for VF return code + * @ice_err: error return code + */ +static enum virtchnl_status_code ice_err_to_virt_err(enum ice_status ice_err) +{ + switch (ice_err) { + case ICE_SUCCESS: + return VIRTCHNL_STATUS_SUCCESS; + case ICE_ERR_BAD_PTR: + case ICE_ERR_INVAL_SIZE: + case ICE_ERR_DEVICE_NOT_SUPPORTED: + case ICE_ERR_PARAM: + case ICE_ERR_CFG: + return VIRTCHNL_STATUS_ERR_PARAM; + case ICE_ERR_NO_MEMORY: + return VIRTCHNL_STATUS_ERR_NO_MEMORY; + case ICE_ERR_NOT_READY: + case ICE_ERR_RESET_FAILED: + case ICE_ERR_FW_API_VER: + case ICE_ERR_AQ_ERROR: + case ICE_ERR_AQ_TIMEOUT: + case ICE_ERR_AQ_FULL: + case ICE_ERR_AQ_NO_WORK: + case ICE_ERR_AQ_EMPTY: + return VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR; + default: + return VIRTCHNL_STATUS_ERR_NOT_SUPPORTED; + } +} + +/** * ice_vc_vf_broadcast - Broadcast a message to all VFs on PF * @pf: pointer to the PF structure * @v_opcode: operation code @@ -47,7 +80,7 @@ ice_vc_vf_broadcast(struct ice_pf *pf, enum virtchnl_ops v_opcode, enum virtchnl_status_code v_retval, u8 *msg, u16 msglen) { struct ice_hw *hw = &pf->hw; - int i; + unsigned int i; ice_for_each_vf(pf, i) { struct ice_vf *vf = &pf->vf[i]; @@ -149,6 +182,26 @@ static void ice_vc_notify_vf_link_state(struct ice_vf *vf) } /** + * ice_vf_invalidate_vsi - invalidate vsi_idx/vsi_num to remove VSI access + * @vf: VF to remove access to VSI for + */ +static void ice_vf_invalidate_vsi(struct ice_vf *vf) +{ + vf->lan_vsi_idx = ICE_NO_VSI; + vf->lan_vsi_num = ICE_NO_VSI; +} + +/** + * ice_vf_vsi_release - invalidate the VF's VSI after freeing it + * @vf: invalidate this VF's VSI after freeing it + */ +static void ice_vf_vsi_release(struct ice_vf *vf) +{ + ice_vsi_release(vf->pf->vsi[vf->lan_vsi_idx]); + ice_vf_invalidate_vsi(vf); +} + +/** * ice_free_vf_res - Free a VF's resources * @vf: pointer to the VF info */ @@ -163,10 +216,8 @@ static void ice_free_vf_res(struct ice_vf *vf) clear_bit(ICE_VF_STATE_INIT, vf->vf_states); /* free VSI and disconnect it from the parent uplink */ - if (vf->lan_vsi_idx) { - ice_vsi_release(pf->vsi[vf->lan_vsi_idx]); - vf->lan_vsi_idx = 0; - vf->lan_vsi_num = 0; + if (vf->lan_vsi_idx != ICE_NO_VSI) { + ice_vf_vsi_release(vf); vf->num_mac = 0; } @@ -292,7 +343,7 @@ void ice_free_vfs(struct ice_pf *pf) { struct device *dev = ice_pf_to_dev(pf); struct ice_hw *hw = &pf->hw; - int tmp, i; + unsigned int tmp, i; if (!pf->vf) return; @@ -337,7 +388,7 @@ void ice_free_vfs(struct ice_pf *pf) * before this function ever gets called. */ if (!pci_vfs_assigned(pf->pdev)) { - int vf_id; + unsigned int vf_id; /* Acknowledge VFLR for all VFs. Without this, VFs will fail to * work correctly when SR-IOV gets re-enabled. @@ -368,9 +419,9 @@ static void ice_trigger_vf_reset(struct ice_vf *vf, bool is_vflr, bool is_pfr) { struct ice_pf *pf = vf->pf; u32 reg, reg_idx, bit_idx; + unsigned int vf_abs_id, i; struct device *dev; struct ice_hw *hw; - int vf_abs_id, i; dev = ice_pf_to_dev(pf); hw = &pf->hw; @@ -380,10 +431,7 @@ static void ice_trigger_vf_reset(struct ice_vf *vf, bool is_vflr, bool is_pfr) clear_bit(ICE_VF_STATE_ACTIVE, vf->vf_states); /* Disable VF's configuration API during reset. The flag is re-enabled - * in ice_alloc_vf_res(), when it's safe again to access VF's VSI. - * It's normally disabled in ice_free_vf_res(), but it's safer - * to do it earlier to give some time to finish to any VF config - * functions that may still be running at this point. + * when it's safe again to access VF's VSI. */ clear_bit(ICE_VF_STATE_INIT, vf->vf_states); @@ -418,7 +466,7 @@ static void ice_trigger_vf_reset(struct ice_vf *vf, bool is_vflr, bool is_pfr) if ((reg & VF_TRANS_PENDING_M) == 0) break; - dev_err(dev, "VF %d PCI transactions stuck\n", vf->vf_id); + dev_err(dev, "VF %u PCI transactions stuck\n", vf->vf_id); udelay(ICE_PCI_CIAD_WAIT_DELAY_US); } } @@ -460,8 +508,9 @@ static int ice_vsi_manage_pvid(struct ice_vsi *vsi, u16 pvid_info, bool enable) status = ice_update_vsi(hw, vsi->idx, ctxt, NULL); if (status) { - dev_info(ice_hw_to_dev(hw), "update VSI for port VLAN failed, err %d aq_err %d\n", - status, hw->adminq.sq_last_status); + dev_info(ice_hw_to_dev(hw), "update VSI for port VLAN failed, err %s aq_err %s\n", + ice_stat_str(status), + ice_aq_str(hw->adminq.sq_last_status)); ret = -EIO; goto out; } @@ -475,18 +524,39 @@ out: } /** + * ice_vf_get_port_info - Get the VF's port info structure + * @vf: VF used to get the port info structure for + */ +static struct ice_port_info *ice_vf_get_port_info(struct ice_vf *vf) +{ + return vf->pf->hw.port_info; +} + +/** * ice_vf_vsi_setup - Set up a VF VSI - * @pf: board private structure - * @pi: pointer to the port_info instance - * @vf_id: defines VF ID to which this VSI connects. + * @vf: VF to setup VSI for * * Returns pointer to the successfully allocated VSI struct on success, * otherwise returns NULL on failure. */ -static struct ice_vsi * -ice_vf_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi, u16 vf_id) +static struct ice_vsi *ice_vf_vsi_setup(struct ice_vf *vf) { - return ice_vsi_setup(pf, pi, ICE_VSI_VF, vf_id); + struct ice_port_info *pi = ice_vf_get_port_info(vf); + struct ice_pf *pf = vf->pf; + struct ice_vsi *vsi; + + vsi = ice_vsi_setup(pf, pi, ICE_VSI_VF, vf->vf_id); + + if (!vsi) { + dev_err(ice_pf_to_dev(pf), "Failed to create VF VSI\n"); + ice_vf_invalidate_vsi(vf); + return NULL; + } + + vf->lan_vsi_idx = vsi->idx; + vf->lan_vsi_num = vsi->vsi_num; + + return vsi; } /** @@ -507,170 +577,158 @@ static int ice_calc_vf_first_vector_idx(struct ice_pf *pf, struct ice_vf *vf) } /** - * ice_alloc_vsi_res - Setup VF VSI and its resources - * @vf: pointer to the VF structure + * ice_vf_rebuild_host_vlan_cfg - add VLAN 0 filter or rebuild the Port VLAN + * @vf: VF to add MAC filters for * - * Returns 0 on success, negative value on failure + * Called after a VF VSI has been re-added/rebuilt during reset. The PF driver + * always re-adds either a VLAN 0 or port VLAN based filter after reset. */ -static int ice_alloc_vsi_res(struct ice_vf *vf) +static int ice_vf_rebuild_host_vlan_cfg(struct ice_vf *vf) { - struct ice_pf *pf = vf->pf; - LIST_HEAD(tmp_add_list); - u8 broadcast[ETH_ALEN]; - struct ice_vsi *vsi; - struct device *dev; - int status = 0; + struct ice_vsi *vsi = vf->pf->vsi[vf->lan_vsi_idx]; + struct device *dev = ice_pf_to_dev(vf->pf); + u16 vlan_id = 0; + int err; - dev = ice_pf_to_dev(pf); - /* first vector index is the VFs OICR index */ - vf->first_vector_idx = ice_calc_vf_first_vector_idx(pf, vf); + if (vf->port_vlan_info) { + err = ice_vsi_manage_pvid(vsi, vf->port_vlan_info, true); + if (err) { + dev_err(dev, "failed to configure port VLAN via VSI parameters for VF %u, error %d\n", + vf->vf_id, err); + return err; + } - vsi = ice_vf_vsi_setup(pf, pf->hw.port_info, vf->vf_id); - if (!vsi) { - dev_err(dev, "Failed to create VF VSI\n"); - return -ENOMEM; + vlan_id = vf->port_vlan_info & VLAN_VID_MASK; } - vf->lan_vsi_idx = vsi->idx; - vf->lan_vsi_num = vsi->vsi_num; - - /* Check if port VLAN exist before, and restore it accordingly */ - if (vf->port_vlan_info) { - ice_vsi_manage_pvid(vsi, vf->port_vlan_info, true); - if (ice_vsi_add_vlan(vsi, vf->port_vlan_info & VLAN_VID_MASK)) - dev_warn(ice_pf_to_dev(pf), "Failed to add Port VLAN %d filter for VF %d\n", - vf->port_vlan_info & VLAN_VID_MASK, vf->vf_id); - } else { - /* set VLAN 0 filter by default when no port VLAN is - * enabled. If a port VLAN is enabled we don't want - * untagged broadcast/multicast traffic seen on the VF - * interface. - */ - if (ice_vsi_add_vlan(vsi, 0)) - dev_warn(ice_pf_to_dev(pf), "Failed to add VLAN 0 filter for VF %d, MDD events will trigger. Reset the VF, disable spoofchk, or enable 8021q module on the guest\n", - vf->vf_id); + /* vlan_id will either be 0 or the port VLAN number */ + err = ice_vsi_add_vlan(vsi, vlan_id, ICE_FWD_TO_VSI); + if (err) { + dev_err(dev, "failed to add %s VLAN %u filter for VF %u, error %d\n", + vf->port_vlan_info ? "port" : "", vlan_id, vf->vf_id, + err); + return err; } + return 0; +} + +/** + * ice_vf_rebuild_host_mac_cfg - add broadcast and the VF's perm_addr/LAA + * @vf: VF to add MAC filters for + * + * Called after a VF VSI has been re-added/rebuilt during reset. The PF driver + * always re-adds a broadcast filter and the VF's perm_addr/LAA after reset. + */ +static int ice_vf_rebuild_host_mac_cfg(struct ice_vf *vf) +{ + struct ice_vsi *vsi = vf->pf->vsi[vf->lan_vsi_idx]; + struct device *dev = ice_pf_to_dev(vf->pf); + enum ice_status status; + u8 broadcast[ETH_ALEN]; + eth_broadcast_addr(broadcast); + status = ice_fltr_add_mac(vsi, broadcast, ICE_FWD_TO_VSI); + if (status) { + dev_err(dev, "failed to add broadcast MAC filter for VF %u, error %s\n", + vf->vf_id, ice_stat_str(status)); + return ice_status_to_errno(status); + } - status = ice_add_mac_to_list(vsi, &tmp_add_list, broadcast); - if (status) - goto ice_alloc_vsi_res_exit; + vf->num_mac++; if (is_valid_ether_addr(vf->dflt_lan_addr.addr)) { - status = ice_add_mac_to_list(vsi, &tmp_add_list, - vf->dflt_lan_addr.addr); - if (status) - goto ice_alloc_vsi_res_exit; + status = ice_fltr_add_mac(vsi, vf->dflt_lan_addr.addr, + ICE_FWD_TO_VSI); + if (status) { + dev_err(dev, "failed to add default unicast MAC filter %pM for VF %u, error %s\n", + &vf->dflt_lan_addr.addr[0], vf->vf_id, + ice_stat_str(status)); + return ice_status_to_errno(status); + } + vf->num_mac++; } - status = ice_add_mac(&pf->hw, &tmp_add_list); - if (status) - dev_err(dev, "could not add mac filters error %d\n", status); - else - vf->num_mac = 1; - - /* Clear this bit after VF initialization since we shouldn't reclaim - * and reassign interrupts for synchronous or asynchronous VFR events. - * We don't want to reconfigure interrupts since AVF driver doesn't - * expect vector assignment to be changed unless there is a request for - * more vectors. - */ -ice_alloc_vsi_res_exit: - ice_free_fltr_list(dev, &tmp_add_list); - return status; + return 0; } /** - * ice_alloc_vf_res - Allocate VF resources - * @vf: pointer to the VF structure + * ice_vf_set_host_trust_cfg - set trust setting based on pre-reset value + * @vf: VF to configure trust setting for */ -static int ice_alloc_vf_res(struct ice_vf *vf) +static void ice_vf_set_host_trust_cfg(struct ice_vf *vf) { - struct ice_pf *pf = vf->pf; - int tx_rx_queue_left; - int status; - - /* Update number of VF queues, in case VF had requested for queue - * changes - */ - tx_rx_queue_left = min_t(int, ice_get_avail_txq_count(pf), - ice_get_avail_rxq_count(pf)); - tx_rx_queue_left += pf->num_qps_per_vf; - if (vf->num_req_qs && vf->num_req_qs <= tx_rx_queue_left && - vf->num_req_qs != vf->num_vf_qs) - vf->num_vf_qs = vf->num_req_qs; - - /* setup VF VSI and necessary resources */ - status = ice_alloc_vsi_res(vf); - if (status) - goto ice_alloc_vf_res_exit; - if (vf->trusted) set_bit(ICE_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps); else clear_bit(ICE_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps); - - /* VF is now completely initialized */ - set_bit(ICE_VF_STATE_INIT, vf->vf_states); - - return status; - -ice_alloc_vf_res_exit: - ice_free_vf_res(vf); - return status; } /** - * ice_ena_vf_mappings - * @vf: pointer to the VF structure + * ice_ena_vf_msix_mappings - enable VF MSIX mappings in hardware + * @vf: VF to enable MSIX mappings for * - * Enable VF vectors and queues allocation by writing the details into - * respective registers. + * Some of the registers need to be indexed/configured using hardware global + * device values and other registers need 0-based values, which represent PF + * based values. */ -static void ice_ena_vf_mappings(struct ice_vf *vf) +static void ice_ena_vf_msix_mappings(struct ice_vf *vf) { - int abs_vf_id, abs_first, abs_last; + int device_based_first_msix, device_based_last_msix; + int pf_based_first_msix, pf_based_last_msix, v; struct ice_pf *pf = vf->pf; - struct ice_vsi *vsi; - struct device *dev; - int first, last, v; + int device_based_vf_id; struct ice_hw *hw; u32 reg; - dev = ice_pf_to_dev(pf); hw = &pf->hw; - vsi = pf->vsi[vf->lan_vsi_idx]; - first = vf->first_vector_idx; - last = (first + pf->num_msix_per_vf) - 1; - abs_first = first + pf->hw.func_caps.common_cap.msix_vector_first_id; - abs_last = (abs_first + pf->num_msix_per_vf) - 1; - abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id; - - /* VF Vector allocation */ - reg = (((abs_first << VPINT_ALLOC_FIRST_S) & VPINT_ALLOC_FIRST_M) | - ((abs_last << VPINT_ALLOC_LAST_S) & VPINT_ALLOC_LAST_M) | - VPINT_ALLOC_VALID_M); + pf_based_first_msix = vf->first_vector_idx; + pf_based_last_msix = (pf_based_first_msix + pf->num_msix_per_vf) - 1; + + device_based_first_msix = pf_based_first_msix + + pf->hw.func_caps.common_cap.msix_vector_first_id; + device_based_last_msix = + (device_based_first_msix + pf->num_msix_per_vf) - 1; + device_based_vf_id = vf->vf_id + hw->func_caps.vf_base_id; + + reg = (((device_based_first_msix << VPINT_ALLOC_FIRST_S) & + VPINT_ALLOC_FIRST_M) | + ((device_based_last_msix << VPINT_ALLOC_LAST_S) & + VPINT_ALLOC_LAST_M) | VPINT_ALLOC_VALID_M); wr32(hw, VPINT_ALLOC(vf->vf_id), reg); - reg = (((abs_first << VPINT_ALLOC_PCI_FIRST_S) + reg = (((device_based_first_msix << VPINT_ALLOC_PCI_FIRST_S) & VPINT_ALLOC_PCI_FIRST_M) | - ((abs_last << VPINT_ALLOC_PCI_LAST_S) & VPINT_ALLOC_PCI_LAST_M) | - VPINT_ALLOC_PCI_VALID_M); + ((device_based_last_msix << VPINT_ALLOC_PCI_LAST_S) & + VPINT_ALLOC_PCI_LAST_M) | VPINT_ALLOC_PCI_VALID_M); wr32(hw, VPINT_ALLOC_PCI(vf->vf_id), reg); + /* map the interrupts to its functions */ - for (v = first; v <= last; v++) { - reg = (((abs_vf_id << GLINT_VECT2FUNC_VF_NUM_S) & + for (v = pf_based_first_msix; v <= pf_based_last_msix; v++) { + reg = (((device_based_vf_id << GLINT_VECT2FUNC_VF_NUM_S) & GLINT_VECT2FUNC_VF_NUM_M) | ((hw->pf_id << GLINT_VECT2FUNC_PF_NUM_S) & GLINT_VECT2FUNC_PF_NUM_M)); wr32(hw, GLINT_VECT2FUNC(v), reg); } - /* Map mailbox interrupt. We put an explicit 0 here to remind us that - * VF admin queue interrupts will go to VF MSI-X vector 0. - */ - wr32(hw, VPINT_MBX_CTL(abs_vf_id), VPINT_MBX_CTL_CAUSE_ENA_M | 0); + /* Map mailbox interrupt to VF MSI-X vector 0 */ + wr32(hw, VPINT_MBX_CTL(device_based_vf_id), VPINT_MBX_CTL_CAUSE_ENA_M); +} + +/** + * ice_ena_vf_q_mappings - enable Rx/Tx queue mappings for a VF + * @vf: VF to enable the mappings for + * @max_txq: max Tx queues allowed on the VF's VSI + * @max_rxq: max Rx queues allowed on the VF's VSI + */ +static void ice_ena_vf_q_mappings(struct ice_vf *vf, u16 max_txq, u16 max_rxq) +{ + struct ice_vsi *vsi = vf->pf->vsi[vf->lan_vsi_idx]; + struct device *dev = ice_pf_to_dev(vf->pf); + struct ice_hw *hw = &vf->pf->hw; + u32 reg; + /* set regardless of mapping mode */ wr32(hw, VPLAN_TXQ_MAPENA(vf->vf_id), VPLAN_TXQ_MAPENA_TX_ENA_M); @@ -682,7 +740,7 @@ static void ice_ena_vf_mappings(struct ice_vf *vf) */ reg = (((vsi->txq_map[0] << VPLAN_TX_QBASE_VFFIRSTQ_S) & VPLAN_TX_QBASE_VFFIRSTQ_M) | - (((vsi->alloc_txq - 1) << VPLAN_TX_QBASE_VFNUMQ_S) & + (((max_txq - 1) << VPLAN_TX_QBASE_VFNUMQ_S) & VPLAN_TX_QBASE_VFNUMQ_M)); wr32(hw, VPLAN_TX_QBASE(vf->vf_id), reg); } else { @@ -700,7 +758,7 @@ static void ice_ena_vf_mappings(struct ice_vf *vf) */ reg = (((vsi->rxq_map[0] << VPLAN_RX_QBASE_VFFIRSTQ_S) & VPLAN_RX_QBASE_VFFIRSTQ_M) | - (((vsi->alloc_txq - 1) << VPLAN_RX_QBASE_VFNUMQ_S) & + (((max_rxq - 1) << VPLAN_RX_QBASE_VFNUMQ_S) & VPLAN_RX_QBASE_VFNUMQ_M)); wr32(hw, VPLAN_RX_QBASE(vf->vf_id), reg); } else { @@ -709,6 +767,18 @@ static void ice_ena_vf_mappings(struct ice_vf *vf) } /** + * ice_ena_vf_mappings - enable VF MSIX and queue mapping + * @vf: pointer to the VF structure + */ +static void ice_ena_vf_mappings(struct ice_vf *vf) +{ + struct ice_vsi *vsi = vf->pf->vsi[vf->lan_vsi_idx]; + + ice_ena_vf_msix_mappings(vf); + ice_ena_vf_q_mappings(vf, vsi->alloc_txq, vsi->alloc_rxq); +} + +/** * ice_determine_res * @pf: pointer to the PF structure * @avail_res: available resources in the PF structure @@ -906,51 +976,18 @@ static int ice_set_per_vf_res(struct ice_pf *pf) } /** - * ice_cleanup_and_realloc_vf - Clean up VF and reallocate resources after reset - * @vf: pointer to the VF structure - * - * Cleanup a VF after the hardware reset is finished. Expects the caller to - * have verified whether the reset is finished properly, and ensure the - * minimum amount of wait time has passed. Reallocate VF resources back to make - * VF state active + * ice_clear_vf_reset_trigger - enable VF to access hardware + * @vf: VF to enabled hardware access for */ -static void ice_cleanup_and_realloc_vf(struct ice_vf *vf) +static void ice_clear_vf_reset_trigger(struct ice_vf *vf) { - struct ice_pf *pf = vf->pf; - struct ice_hw *hw; + struct ice_hw *hw = &vf->pf->hw; u32 reg; - hw = &pf->hw; - - /* PF software completes the flow by notifying VF that reset flow is - * completed. This is done by enabling hardware by clearing the reset - * bit in the VPGEN_VFRTRIG reg and setting VFR_STATE in the VFGEN_RSTAT - * register to VFR completed (done at the end of this function) - * By doing this we allow HW to access VF memory at any point. If we - * did it any sooner, HW could access memory while it was being freed - * in ice_free_vf_res(), causing an IOMMU fault. - * - * On the other hand, this needs to be done ASAP, because the VF driver - * is waiting for this to happen and may report a timeout. It's - * harmless, but it gets logged into Guest OS kernel log, so best avoid - * it. - */ reg = rd32(hw, VPGEN_VFRTRIG(vf->vf_id)); reg &= ~VPGEN_VFRTRIG_VFSWR_M; wr32(hw, VPGEN_VFRTRIG(vf->vf_id), reg); - - /* reallocate VF resources to finish resetting the VSI state */ - if (!ice_alloc_vf_res(vf)) { - ice_ena_vf_mappings(vf); - set_bit(ICE_VF_STATE_ACTIVE, vf->vf_states); - clear_bit(ICE_VF_STATE_DIS, vf->vf_states); - } - - /* Tell the VF driver the reset is done. This needs to be done only - * after VF has been fully initialized, because the VF driver may - * request resources immediately after setting this flag. - */ - wr32(hw, VFGEN_RSTAT(vf->vf_id), VIRTCHNL_VFR_VFACTIVE); + ice_flush(hw); } /** @@ -994,44 +1031,124 @@ ice_vf_set_vsi_promisc(struct ice_vf *vf, struct ice_vsi *vsi, u8 promisc_m, return status; } +static void ice_vf_clear_counters(struct ice_vf *vf) +{ + struct ice_vsi *vsi = vf->pf->vsi[vf->lan_vsi_idx]; + + vf->num_mac = 0; + vsi->num_vlan = 0; + memset(&vf->mdd_tx_events, 0, sizeof(vf->mdd_tx_events)); + memset(&vf->mdd_rx_events, 0, sizeof(vf->mdd_rx_events)); +} + /** - * ice_config_res_vfs - Finalize allocation of VFs resources in one go - * @pf: pointer to the PF structure + * ice_vf_pre_vsi_rebuild - tasks to be done prior to VSI rebuild + * @vf: VF to perform pre VSI rebuild tasks * - * This function is being called as last part of resetting all VFs, or when - * configuring VFs for the first time, where there is no resource to be freed - * Returns true if resources were properly allocated for all VFs, and false - * otherwise. + * These tasks are items that don't need to be amortized since they are most + * likely called in a for loop with all VF(s) in the reset_all_vfs() case. */ -static bool ice_config_res_vfs(struct ice_pf *pf) +static void ice_vf_pre_vsi_rebuild(struct ice_vf *vf) { - struct device *dev = ice_pf_to_dev(pf); - struct ice_hw *hw = &pf->hw; - int v; + ice_vf_clear_counters(vf); + ice_clear_vf_reset_trigger(vf); +} - if (ice_set_per_vf_res(pf)) { - dev_err(dev, "Cannot allocate VF resources, try with fewer number of VFs\n"); - return false; - } +/** + * ice_vf_rebuild_host_cfg - host admin configuration is persistent across reset + * @vf: VF to rebuild host configuration on + */ +static void ice_vf_rebuild_host_cfg(struct ice_vf *vf) +{ + struct device *dev = ice_pf_to_dev(vf->pf); - /* rearm global interrupts */ - if (test_and_clear_bit(__ICE_OICR_INTR_DIS, pf->state)) - ice_irq_dynamic_ena(hw, NULL, NULL); + ice_vf_set_host_trust_cfg(vf); - /* Finish resetting each VF and allocate resources */ - ice_for_each_vf(pf, v) { - struct ice_vf *vf = &pf->vf[v]; + if (ice_vf_rebuild_host_mac_cfg(vf)) + dev_err(dev, "failed to rebuild default MAC configuration for VF %d\n", + vf->vf_id); - vf->num_vf_qs = pf->num_qps_per_vf; - dev_dbg(dev, "VF-id %d has %d queues configured\n", vf->vf_id, - vf->num_vf_qs); - ice_cleanup_and_realloc_vf(vf); + if (ice_vf_rebuild_host_vlan_cfg(vf)) + dev_err(dev, "failed to rebuild VLAN configuration for VF %u\n", + vf->vf_id); +} + +/** + * ice_vf_rebuild_vsi_with_release - release and setup the VF's VSI + * @vf: VF to release and setup the VSI for + * + * This is only called when a single VF is being reset (i.e. VFR, VFLR, host VF + * configuration change, etc.). + */ +static int ice_vf_rebuild_vsi_with_release(struct ice_vf *vf) +{ + ice_vf_vsi_release(vf); + if (!ice_vf_vsi_setup(vf)) + return -ENOMEM; + + return 0; +} + +/** + * ice_vf_rebuild_vsi - rebuild the VF's VSI + * @vf: VF to rebuild the VSI for + * + * This is only called when all VF(s) are being reset (i.e. PCIe Reset on the + * host, PFR, CORER, etc.). + */ +static int ice_vf_rebuild_vsi(struct ice_vf *vf) +{ + struct ice_pf *pf = vf->pf; + struct ice_vsi *vsi; + + vsi = pf->vsi[vf->lan_vsi_idx]; + + if (ice_vsi_rebuild(vsi, true)) { + dev_err(ice_pf_to_dev(pf), "failed to rebuild VF %d VSI\n", + vf->vf_id); + return -EIO; } + /* vsi->idx will remain the same in this case so don't update + * vf->lan_vsi_idx + */ + vsi->vsi_num = ice_get_hw_vsi_num(&pf->hw, vsi->idx); + vf->lan_vsi_num = vsi->vsi_num; - ice_flush(hw); - clear_bit(__ICE_VF_DIS, pf->state); + return 0; +} - return true; +/** + * ice_vf_set_initialized - VF is ready for VIRTCHNL communication + * @vf: VF to set in initialized state + * + * After this function the VF will be ready to receive/handle the + * VIRTCHNL_OP_GET_VF_RESOURCES message + */ +static void ice_vf_set_initialized(struct ice_vf *vf) +{ + ice_set_vf_state_qs_dis(vf); + clear_bit(ICE_VF_STATE_MC_PROMISC, vf->vf_states); + clear_bit(ICE_VF_STATE_UC_PROMISC, vf->vf_states); + clear_bit(ICE_VF_STATE_DIS, vf->vf_states); + set_bit(ICE_VF_STATE_INIT, vf->vf_states); +} + +/** + * ice_vf_post_vsi_rebuild - tasks to do after the VF's VSI have been rebuilt + * @vf: VF to perform tasks on + */ +static void ice_vf_post_vsi_rebuild(struct ice_vf *vf) +{ + struct ice_pf *pf = vf->pf; + struct ice_hw *hw; + + hw = &pf->hw; + + ice_vf_rebuild_host_cfg(vf); + + ice_vf_set_initialized(vf); + ice_ena_vf_mappings(vf); + wr32(hw, VFGEN_RSTAT(vf->vf_id), VIRTCHNL_VFR_VFACTIVE); } /** @@ -1065,17 +1182,6 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) ice_for_each_vf(pf, v) ice_trigger_vf_reset(&pf->vf[v], is_vflr, true); - ice_for_each_vf(pf, v) { - struct ice_vsi *vsi; - - vf = &pf->vf[v]; - vsi = pf->vsi[vf->lan_vsi_idx]; - if (test_bit(ICE_VF_STATE_QS_ENA, vf->vf_states)) - ice_dis_vf_qs(vf); - ice_dis_vsi_txq(vsi->port_info, vsi->idx, 0, 0, NULL, NULL, - NULL, ICE_VF_RESET, vf->vf_id, NULL); - } - /* HW requires some time to make sure it can flush the FIFO for a VF * when it resets it. Poll the VPGEN_VFRSTAT register for each VF in * sequence to make sure that it has completed. We'll keep track of @@ -1112,21 +1218,13 @@ bool ice_reset_all_vfs(struct ice_pf *pf, bool is_vflr) ice_for_each_vf(pf, v) { vf = &pf->vf[v]; - ice_free_vf_res(vf); - - /* Free VF queues as well, and reallocate later. - * If a given VF has different number of queues - * configured, the request for update will come - * via mailbox communication. - */ - vf->num_vf_qs = 0; + ice_vf_pre_vsi_rebuild(vf); + ice_vf_rebuild_vsi(vf); + ice_vf_post_vsi_rebuild(vf); } - if (ice_sriov_free_msix_res(pf)) - dev_err(dev, "Failed to free MSIX resources used by SR-IOV\n"); - - if (!ice_config_res_vfs(pf)) - return false; + ice_flush(hw); + clear_bit(__ICE_VF_DIS, pf->state); return true; } @@ -1238,12 +1336,9 @@ bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) dev_err(dev, "disabling promiscuous mode failed\n"); } - /* free VF resources to begin resetting the VSI state */ - ice_free_vf_res(vf); - - ice_cleanup_and_realloc_vf(vf); - - ice_flush(hw); + ice_vf_pre_vsi_rebuild(vf); + ice_vf_rebuild_vsi_with_release(vf); + ice_vf_post_vsi_rebuild(vf); return true; } @@ -1311,16 +1406,144 @@ static void ice_vc_notify_vf_reset(struct ice_vf *vf) } /** - * ice_alloc_vfs - Allocate and set up VFs resources + * ice_init_vf_vsi_res - initialize/setup VF VSI resources + * @vf: VF to initialize/setup the VSI for + * + * This function creates a VSI for the VF, adds a VLAN 0 filter, and sets up the + * VF VSI's broadcast filter and is only used during initial VF creation. + */ +static int ice_init_vf_vsi_res(struct ice_vf *vf) +{ + struct ice_pf *pf = vf->pf; + u8 broadcast[ETH_ALEN]; + enum ice_status status; + struct ice_vsi *vsi; + struct device *dev; + int err; + + vf->first_vector_idx = ice_calc_vf_first_vector_idx(pf, vf); + + dev = ice_pf_to_dev(pf); + vsi = ice_vf_vsi_setup(vf); + if (!vsi) + return -ENOMEM; + + err = ice_vsi_add_vlan(vsi, 0, ICE_FWD_TO_VSI); + if (err) { + dev_warn(dev, "Failed to add VLAN 0 filter for VF %d\n", + vf->vf_id); + goto release_vsi; + } + + eth_broadcast_addr(broadcast); + status = ice_fltr_add_mac(vsi, broadcast, ICE_FWD_TO_VSI); + if (status) { + dev_err(dev, "Failed to add broadcast MAC filter for VF %d, status %s\n", + vf->vf_id, ice_stat_str(status)); + err = ice_status_to_errno(status); + goto release_vsi; + } + + vf->num_mac = 1; + + return 0; + +release_vsi: + ice_vf_vsi_release(vf); + return err; +} + +/** + * ice_start_vfs - start VFs so they are ready to be used by SR-IOV + * @pf: PF the VFs are associated with + */ +static int ice_start_vfs(struct ice_pf *pf) +{ + struct ice_hw *hw = &pf->hw; + int retval, i; + + ice_for_each_vf(pf, i) { + struct ice_vf *vf = &pf->vf[i]; + + ice_clear_vf_reset_trigger(vf); + + retval = ice_init_vf_vsi_res(vf); + if (retval) { + dev_err(ice_pf_to_dev(pf), "Failed to initialize VSI resources for VF %d, error %d\n", + vf->vf_id, retval); + goto teardown; + } + + set_bit(ICE_VF_STATE_INIT, vf->vf_states); + ice_ena_vf_mappings(vf); + wr32(hw, VFGEN_RSTAT(vf->vf_id), VIRTCHNL_VFR_VFACTIVE); + } + + ice_flush(hw); + return 0; + +teardown: + for (i = i - 1; i >= 0; i--) { + struct ice_vf *vf = &pf->vf[i]; + + ice_dis_vf_mappings(vf); + ice_vf_vsi_release(vf); + } + + return retval; +} + +/** + * ice_set_dflt_settings - set VF defaults during initialization/creation + * @pf: PF holding reference to all VFs for default configuration + */ +static void ice_set_dflt_settings_vfs(struct ice_pf *pf) +{ + int i; + + ice_for_each_vf(pf, i) { + struct ice_vf *vf = &pf->vf[i]; + + vf->pf = pf; + vf->vf_id = i; + vf->vf_sw_id = pf->first_sw; + /* assign default capabilities */ + set_bit(ICE_VIRTCHNL_VF_CAP_L2, &vf->vf_caps); + vf->spoofchk = true; + vf->num_vf_qs = pf->num_qps_per_vf; + } +} + +/** + * ice_alloc_vfs - allocate num_vfs in the PF structure + * @pf: PF to store the allocated VFs in + * @num_vfs: number of VFs to allocate + */ +static int ice_alloc_vfs(struct ice_pf *pf, int num_vfs) +{ + struct ice_vf *vfs; + + vfs = devm_kcalloc(ice_pf_to_dev(pf), num_vfs, sizeof(*vfs), + GFP_KERNEL); + if (!vfs) + return -ENOMEM; + + pf->vf = vfs; + pf->num_alloc_vfs = num_vfs; + + return 0; +} + +/** + * ice_ena_vfs - enable VFs so they are ready to be used * @pf: pointer to the PF structure - * @num_alloc_vfs: number of VFs to allocate + * @num_vfs: number of VFs to enable */ -static int ice_alloc_vfs(struct ice_pf *pf, u16 num_alloc_vfs) +static int ice_ena_vfs(struct ice_pf *pf, u16 num_vfs) { struct device *dev = ice_pf_to_dev(pf); struct ice_hw *hw = &pf->hw; - struct ice_vf *vfs; - int i, ret; + int ret; /* Disable global interrupt 0 so we don't try to handle the VFLR. */ wr32(hw, GLINT_DYN_CTL(pf->oicr_idx), @@ -1328,43 +1551,37 @@ static int ice_alloc_vfs(struct ice_pf *pf, u16 num_alloc_vfs) set_bit(__ICE_OICR_INTR_DIS, pf->state); ice_flush(hw); - ret = pci_enable_sriov(pf->pdev, num_alloc_vfs); + ret = pci_enable_sriov(pf->pdev, num_vfs); if (ret) { pf->num_alloc_vfs = 0; goto err_unroll_intr; } - /* allocate memory */ - vfs = devm_kcalloc(dev, num_alloc_vfs, sizeof(*vfs), GFP_KERNEL); - if (!vfs) { - ret = -ENOMEM; - goto err_pci_disable_sriov; - } - pf->vf = vfs; - pf->num_alloc_vfs = num_alloc_vfs; - /* apply default profile */ - ice_for_each_vf(pf, i) { - vfs[i].pf = pf; - vfs[i].vf_sw_id = pf->first_sw; - vfs[i].vf_id = i; + ret = ice_alloc_vfs(pf, num_vfs); + if (ret) + goto err_pci_disable_sriov; - /* assign default capabilities */ - set_bit(ICE_VIRTCHNL_VF_CAP_L2, &vfs[i].vf_caps); - vfs[i].spoofchk = true; + if (ice_set_per_vf_res(pf)) { + dev_err(dev, "Not enough resources for %d VFs, try with fewer number of VFs\n", + num_vfs); + ret = -ENOSPC; + goto err_unroll_sriov; } - /* VF resources get allocated with initialization */ - if (!ice_config_res_vfs(pf)) { - ret = -EIO; + ice_set_dflt_settings_vfs(pf); + + if (ice_start_vfs(pf)) { + dev_err(dev, "Failed to start VF(s)\n"); + ret = -EAGAIN; goto err_unroll_sriov; } - return ret; + clear_bit(__ICE_VF_DIS, pf->state); + return 0; err_unroll_sriov: + devm_kfree(dev, pf->vf); pf->vf = NULL; - devm_kfree(dev, vfs); - vfs = NULL; pf->num_alloc_vfs = 0; err_pci_disable_sriov: pci_disable_sriov(pf->pdev); @@ -1404,6 +1621,8 @@ static bool ice_pf_state_is_nominal(struct ice_pf *pf) * ice_pci_sriov_ena - Enable or change number of VFs * @pf: pointer to the PF structure * @num_vfs: number of VFs to allocate + * + * Returns 0 on success and negative on failure */ static int ice_pci_sriov_ena(struct ice_pf *pf, int num_vfs) { @@ -1411,20 +1630,10 @@ static int ice_pci_sriov_ena(struct ice_pf *pf, int num_vfs) struct device *dev = ice_pf_to_dev(pf); int err; - if (!ice_pf_state_is_nominal(pf)) { - dev_err(dev, "Cannot enable SR-IOV, device not ready\n"); - return -EBUSY; - } - - if (!test_bit(ICE_FLAG_SRIOV_CAPABLE, pf->flags)) { - dev_err(dev, "This device is not capable of SR-IOV\n"); - return -EOPNOTSUPP; - } - if (pre_existing_vfs && pre_existing_vfs != num_vfs) ice_free_vfs(pf); else if (pre_existing_vfs && pre_existing_vfs == num_vfs) - return num_vfs; + return 0; if (num_vfs > pf->num_vfs_supported) { dev_err(dev, "Can't enable %d VFs, max VFs supported is %d\n", @@ -1432,45 +1641,77 @@ static int ice_pci_sriov_ena(struct ice_pf *pf, int num_vfs) return -EOPNOTSUPP; } - dev_info(dev, "Allocating %d VFs\n", num_vfs); - err = ice_alloc_vfs(pf, num_vfs); + dev_info(dev, "Enabling %d VFs\n", num_vfs); + err = ice_ena_vfs(pf, num_vfs); if (err) { dev_err(dev, "Failed to enable SR-IOV: %d\n", err); return err; } set_bit(ICE_FLAG_SRIOV_ENA, pf->flags); - return num_vfs; + return 0; +} + +/** + * ice_check_sriov_allowed - check if SR-IOV is allowed based on various checks + * @pf: PF to enabled SR-IOV on + */ +static int ice_check_sriov_allowed(struct ice_pf *pf) +{ + struct device *dev = ice_pf_to_dev(pf); + + if (!test_bit(ICE_FLAG_SRIOV_CAPABLE, pf->flags)) { + dev_err(dev, "This device is not capable of SR-IOV\n"); + return -EOPNOTSUPP; + } + + if (ice_is_safe_mode(pf)) { + dev_err(dev, "SR-IOV cannot be configured - Device is in Safe Mode\n"); + return -EOPNOTSUPP; + } + + if (!ice_pf_state_is_nominal(pf)) { + dev_err(dev, "Cannot enable SR-IOV, device not ready\n"); + return -EBUSY; + } + + return 0; } /** * ice_sriov_configure - Enable or change number of VFs via sysfs * @pdev: pointer to a pci_dev structure - * @num_vfs: number of VFs to allocate + * @num_vfs: number of VFs to allocate or 0 to free VFs * - * This function is called when the user updates the number of VFs in sysfs. + * This function is called when the user updates the number of VFs in sysfs. On + * success return whatever num_vfs was set to by the caller. Return negative on + * failure. */ int ice_sriov_configure(struct pci_dev *pdev, int num_vfs) { struct ice_pf *pf = pci_get_drvdata(pdev); struct device *dev = ice_pf_to_dev(pf); + int err; - if (ice_is_safe_mode(pf)) { - dev_err(dev, "SR-IOV cannot be configured - Device is in Safe Mode\n"); - return -EOPNOTSUPP; - } + err = ice_check_sriov_allowed(pf); + if (err) + return err; - if (num_vfs) - return ice_pci_sriov_ena(pf, num_vfs); + if (!num_vfs) { + if (!pci_vfs_assigned(pdev)) { + ice_free_vfs(pf); + return 0; + } - if (!pci_vfs_assigned(pdev)) { - ice_free_vfs(pf); - } else { dev_err(dev, "can't free VFs because some are assigned to VMs.\n"); return -EBUSY; } - return 0; + err = ice_pci_sriov_ena(pf, num_vfs); + if (err) + return err; + + return num_vfs; } /** @@ -1483,7 +1724,7 @@ int ice_sriov_configure(struct pci_dev *pdev, int num_vfs) void ice_process_vflr_event(struct ice_pf *pf) { struct ice_hw *hw = &pf->hw; - int vf_id; + unsigned int vf_id; u32 reg; if (!test_and_clear_bit(__ICE_VFLR_EVENT_PENDING, pf->state) || @@ -1524,7 +1765,7 @@ static void ice_vc_reset_vf(struct ice_vf *vf) */ static struct ice_vf *ice_get_vf_from_pfq(struct ice_pf *pf, u16 pfq) { - int vf_id; + unsigned int vf_id; ice_for_each_vf(pf, vf_id) { struct ice_vf *vf = &pf->vf[vf_id]; @@ -1628,8 +1869,9 @@ ice_vc_send_msg_to_vf(struct ice_vf *vf, u32 v_opcode, aq_ret = ice_aq_send_msg_to_vf(&pf->hw, vf->vf_id, v_opcode, v_retval, msg, msglen, NULL); if (aq_ret && pf->hw.mailboxq.sq_last_status != ICE_AQ_RC_ENOSYS) { - dev_info(dev, "Unable to send the message to VF %d ret %d aq_err %d\n", - vf->vf_id, aq_ret, pf->hw.mailboxq.sq_last_status); + dev_info(dev, "Unable to send the message to VF %d ret %s aq_err %s\n", + vf->vf_id, ice_stat_str(aq_ret), + ice_aq_str(pf->hw.mailboxq.sq_last_status)); return -EIO; } @@ -1772,7 +2014,7 @@ err: */ static void ice_vc_reset_vf_msg(struct ice_vf *vf) { - if (test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) + if (test_bit(ICE_VF_STATE_INIT, vf->vf_states)) ice_reset_vf(vf, false); } @@ -2044,8 +2286,9 @@ int ice_set_vf_spoofchk(struct net_device *netdev, int vf_id, bool ena) status = ice_update_vsi(&pf->hw, vf_vsi->idx, ctx, NULL); if (status) { - dev_err(dev, "Failed to %sable spoofchk on VF %d VSI %d\n error %d\n", - ena ? "en" : "dis", vf->vf_id, vf_vsi->vsi_num, status); + dev_err(dev, "Failed to %sable spoofchk on VF %d VSI %d\n error %s\n", + ena ? "en" : "dis", vf->vf_id, vf_vsi->vsi_num, + ice_stat_str(status)); ret = -EIO; goto out; } @@ -2060,6 +2303,174 @@ out: } /** + * ice_is_any_vf_in_promisc - check if any VF(s) are in promiscuous mode + * @pf: PF structure for accessing VF(s) + * + * Return false if no VF(s) are in unicast and/or multicast promiscuous mode, + * else return true + */ +bool ice_is_any_vf_in_promisc(struct ice_pf *pf) +{ + int vf_idx; + + ice_for_each_vf(pf, vf_idx) { + struct ice_vf *vf = &pf->vf[vf_idx]; + + /* found a VF that has promiscuous mode configured */ + if (test_bit(ICE_VF_STATE_UC_PROMISC, vf->vf_states) || + test_bit(ICE_VF_STATE_MC_PROMISC, vf->vf_states)) + return true; + } + + return false; +} + +/** + * ice_vc_cfg_promiscuous_mode_msg + * @vf: pointer to the VF info + * @msg: pointer to the msg buffer + * + * called from the VF to configure VF VSIs promiscuous mode + */ +static int ice_vc_cfg_promiscuous_mode_msg(struct ice_vf *vf, u8 *msg) +{ + enum virtchnl_status_code v_ret = VIRTCHNL_STATUS_SUCCESS; + struct virtchnl_promisc_info *info = + (struct virtchnl_promisc_info *)msg; + struct ice_pf *pf = vf->pf; + struct ice_vsi *vsi; + struct device *dev; + bool rm_promisc; + int ret = 0; + + if (!test_bit(ICE_VF_STATE_ACTIVE, vf->vf_states)) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + if (!ice_vc_isvalid_vsi_id(vf, info->vsi_id)) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + vsi = pf->vsi[vf->lan_vsi_idx]; + if (!vsi) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + dev = ice_pf_to_dev(pf); + if (!test_bit(ICE_VIRTCHNL_VF_CAP_PRIVILEGE, &vf->vf_caps)) { + dev_err(dev, "Unprivileged VF %d is attempting to configure promiscuous mode\n", + vf->vf_id); + /* Leave v_ret alone, lie to the VF on purpose. */ + goto error_param; + } + + rm_promisc = !(info->flags & FLAG_VF_UNICAST_PROMISC) && + !(info->flags & FLAG_VF_MULTICAST_PROMISC); + + if (vsi->num_vlan || vf->port_vlan_info) { + struct ice_vsi *pf_vsi = ice_get_main_vsi(pf); + struct net_device *pf_netdev; + + if (!pf_vsi) { + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + + pf_netdev = pf_vsi->netdev; + + ret = ice_set_vf_spoofchk(pf_netdev, vf->vf_id, rm_promisc); + if (ret) { + dev_err(dev, "Failed to update spoofchk to %s for VF %d VSI %d when setting promiscuous mode\n", + rm_promisc ? "ON" : "OFF", vf->vf_id, + vsi->vsi_num); + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + } + + ret = ice_cfg_vlan_pruning(vsi, true, !rm_promisc); + if (ret) { + dev_err(dev, "Failed to configure VLAN pruning in promiscuous mode\n"); + v_ret = VIRTCHNL_STATUS_ERR_PARAM; + goto error_param; + } + } + + if (!test_bit(ICE_FLAG_VF_TRUE_PROMISC_ENA, pf->flags)) { + bool set_dflt_vsi = !!(info->flags & FLAG_VF_UNICAST_PROMISC); + + if (set_dflt_vsi && !ice_is_dflt_vsi_in_use(pf->first_sw)) + /* only attempt to set the default forwarding VSI if + * it's not currently set + */ + ret = ice_set_dflt_vsi(pf->first_sw, vsi); + else if (!set_dflt_vsi && + ice_is_vsi_dflt_vsi(pf->first_sw, vsi)) + /* only attempt to free the default forwarding VSI if we + * are the owner + */ + ret = ice_clear_dflt_vsi(pf->first_sw); + + if (ret) { + dev_err(dev, "%sable VF %d as the default VSI failed, error %d\n", + set_dflt_vsi ? "en" : "dis", vf->vf_id, ret); + v_ret = VIRTCHNL_STATUS_ERR_ADMIN_QUEUE_ERROR; + goto error_param; + } + } else { + enum ice_status status; + u8 promisc_m; + + if (info->flags & FLAG_VF_UNICAST_PROMISC) { + if (vf->port_vlan_info || vsi->num_vlan) + promisc_m = ICE_UCAST_VLAN_PROMISC_BITS; + else + promisc_m = ICE_UCAST_PROMISC_BITS; + } else if (info->flags & FLAG_VF_MULTICAST_PROMISC) { + if (vf->port_vlan_info || vsi->num_vlan) + promisc_m = ICE_MCAST_VLAN_PROMISC_BITS; + else + promisc_m = ICE_MCAST_PROMISC_BITS; + } else { + if (vf->port_vlan_info || vsi->num_vlan) + promisc_m = ICE_UCAST_VLAN_PROMISC_BITS; + else + promisc_m = ICE_UCAST_PROMISC_BITS; + } + + /* Configure multicast/unicast with or without VLAN promiscuous + * mode + */ + status = ice_vf_set_vsi_promisc(vf, vsi, promisc_m, rm_promisc); + if (status) { + dev_err(dev, "%sable Tx/Rx filter promiscuous mode on VF-%d failed, error: %s\n", + rm_promisc ? "dis" : "en", vf->vf_id, + ice_stat_str(status)); + v_ret = ice_err_to_virt_err(status); + goto error_param; + } else { + dev_dbg(dev, "%sable Tx/Rx filter promiscuous mode on VF-%d succeeded\n", + rm_promisc ? "dis" : "en", vf->vf_id); + } + } + + if (info->flags & FLAG_VF_MULTICAST_PROMISC) + set_bit(ICE_VF_STATE_MC_PROMISC, vf->vf_states); + else + clear_bit(ICE_VF_STATE_MC_PROMISC, vf->vf_states); + + if (info->flags & FLAG_VF_UNICAST_PROMISC) + set_bit(ICE_VF_STATE_UC_PROMISC, vf->vf_states); + else + clear_bit(ICE_VF_STATE_UC_PROMISC, vf->vf_states); + +error_param: + return ice_vc_send_msg_to_vf(vf, VIRTCHNL_OP_CONFIG_PROMISCUOUS_MODE, + v_ret, NULL, 0); +} + +/** * ice_vc_get_stats_msg * @vf: pointer to the VF info * @msg: pointer to the msg buffer @@ -2118,6 +2529,52 @@ static bool ice_vc_validate_vqs_bitmaps(struct virtchnl_queue_select *vqs) } /** + * ice_vf_ena_txq_interrupt - enable Tx queue interrupt via QINT_TQCTL + * @vsi: VSI of the VF to configure + * @q_idx: VF queue index used to determine the queue in the PF's space + */ +static void ice_vf_ena_txq_interrupt(struct ice_vsi *vsi, u32 q_idx) +{ + struct ice_hw *hw = &vsi->back->hw; + u32 pfq = vsi->txq_map[q_idx]; + u32 reg; + + reg = rd32(hw, QINT_TQCTL(pfq)); + + /* MSI-X index 0 in the VF's space is always for the OICR, which means + * this is most likely a poll mode VF driver, so don't enable an + * interrupt that was never configured via VIRTCHNL_OP_CONFIG_IRQ_MAP + */ + if (!(reg & QINT_TQCTL_MSIX_INDX_M)) + return; + + wr32(hw, QINT_TQCTL(pfq), reg | QINT_TQCTL_CAUSE_ENA_M); +} + +/** + * ice_vf_ena_rxq_interrupt - enable Tx queue interrupt via QINT_RQCTL + * @vsi: VSI of the VF to configure + * @q_idx: VF queue index used to determine the queue in the PF's space + */ +static void ice_vf_ena_rxq_interrupt(struct ice_vsi *vsi, u32 q_idx) +{ + struct ice_hw *hw = &vsi->back->hw; + u32 pfq = vsi->rxq_map[q_idx]; + u32 reg; + + reg = rd32(hw, QINT_RQCTL(pfq)); + + /* MSI-X index 0 in the VF's space is always for the OICR, which means + * this is most likely a poll mode VF driver, so don't enable an + * interrupt that was never configured via VIRTCHNL_OP_CONFIG_IRQ_MAP + */ + if (!(reg & QINT_RQCTL_MSIX_INDX_M)) + return; + + wr32(hw, QINT_RQCTL(pfq), reg | QINT_RQCTL_CAUSE_ENA_M); +} + +/** * ice_vc_ena_qs_msg * @vf: pointer to the VF info * @msg: pointer to the msg buffer @@ -2177,6 +2634,7 @@ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg) goto error_param; } + ice_vf_ena_rxq_interrupt(vsi, vf_q_id); set_bit(vf_q_id, vf->rxq_ena); } @@ -2192,6 +2650,7 @@ static int ice_vc_ena_qs_msg(struct ice_vf *vf, u8 *msg) if (test_bit(vf_q_id, vf->txq_ena)) continue; + ice_vf_ena_txq_interrupt(vsi, vf_q_id); set_bit(vf_q_id, vf->txq_ena); } @@ -2604,20 +3063,22 @@ ice_vc_add_mac_addr(struct ice_vf *vf, struct ice_vsi *vsi, u8 *mac_addr) return -EPERM; } - status = ice_vsi_cfg_mac_fltr(vsi, mac_addr, true); + status = ice_fltr_add_mac(vsi, mac_addr, ICE_FWD_TO_VSI); if (status == ICE_ERR_ALREADY_EXISTS) { dev_err(dev, "MAC %pM already exists for VF %d\n", mac_addr, vf->vf_id); return -EEXIST; } else if (status) { - dev_err(dev, "Failed to add MAC %pM for VF %d\n, error %d\n", - mac_addr, vf->vf_id, status); + dev_err(dev, "Failed to add MAC %pM for VF %d\n, error %s\n", + mac_addr, vf->vf_id, ice_stat_str(status)); return -EIO; } - /* only set dflt_lan_addr once */ - if (is_zero_ether_addr(vf->dflt_lan_addr.addr) && - is_unicast_ether_addr(mac_addr)) + /* Set the default LAN address to the latest unicast MAC address added + * by the VF. The default LAN address is reported by the PF via + * ndo_get_vf_config. + */ + if (is_unicast_ether_addr(mac_addr)) ether_addr_copy(vf->dflt_lan_addr.addr, mac_addr); vf->num_mac++; @@ -2641,14 +3102,14 @@ ice_vc_del_mac_addr(struct ice_vf *vf, struct ice_vsi *vsi, u8 *mac_addr) ether_addr_equal(mac_addr, vf->dflt_lan_addr.addr)) return 0; - status = ice_vsi_cfg_mac_fltr(vsi, mac_addr, false); + status = ice_fltr_remove_mac(vsi, mac_addr, ICE_FWD_TO_VSI); if (status == ICE_ERR_DOES_NOT_EXIST) { dev_err(dev, "MAC %pM does not exist for VF %d\n", mac_addr, vf->vf_id); return -ENOENT; } else if (status) { - dev_err(dev, "Failed to delete MAC %pM for VF %d, error %d\n", - mac_addr, vf->vf_id, status); + dev_err(dev, "Failed to delete MAC %pM for VF %d, error %s\n", + mac_addr, vf->vf_id, ice_stat_str(status)); return -EIO; } @@ -2834,7 +3295,6 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, __be16 vlan_proto) { struct ice_pf *pf = ice_netdev_to_pf(netdev); - struct ice_vsi *vsi; struct device *dev; struct ice_vf *vf; u16 vlanprio; @@ -2856,8 +3316,6 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, } vf = &pf->vf[vf_id]; - vsi = pf->vsi[vf->lan_vsi_idx]; - ret = ice_check_vf_ready_for_cfg(vf); if (ret) return ret; @@ -2870,44 +3328,15 @@ ice_set_vf_port_vlan(struct net_device *netdev, int vf_id, u16 vlan_id, u8 qos, return 0; } - if (vlan_id || qos) { - /* remove VLAN 0 filter set by default when transitioning from - * no port VLAN to a port VLAN. No change to old port VLAN on - * failure. - */ - ret = ice_vsi_kill_vlan(vsi, 0); - if (ret) - return ret; - ret = ice_vsi_manage_pvid(vsi, vlanprio, true); - if (ret) - return ret; - } else { - /* add VLAN 0 filter back when transitioning from port VLAN to - * no port VLAN. No change to old port VLAN on failure. - */ - ret = ice_vsi_add_vlan(vsi, 0); - if (ret) - return ret; - ret = ice_vsi_manage_pvid(vsi, 0, false); - if (ret) - return ret; - } + vf->port_vlan_info = vlanprio; - if (vlan_id) { + if (vf->port_vlan_info) dev_info(dev, "Setting VLAN %d, QoS 0x%x on VF %d\n", vlan_id, qos, vf_id); + else + dev_info(dev, "Clearing port VLAN on VF %d\n", vf_id); - /* add VLAN filter for the port VLAN */ - ret = ice_vsi_add_vlan(vsi, vlan_id); - if (ret) - return ret; - } - /* remove old port VLAN filter with valid VLAN ID or QoS fields */ - if (vf->port_vlan_info) - ice_vsi_kill_vlan(vsi, vf->port_vlan_info & VLAN_VID_MASK); - - /* keep port VLAN information persistent on resets */ - vf->port_vlan_info = le16_to_cpu(vsi->info.pvid); + ice_vc_reset_vf(vf); return 0; } @@ -2992,8 +3421,9 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v) goto error_param; } - if (test_bit(ICE_VF_STATE_UC_PROMISC, vf->vf_states) || - test_bit(ICE_VF_STATE_MC_PROMISC, vf->vf_states)) + if ((test_bit(ICE_VF_STATE_UC_PROMISC, vf->vf_states) || + test_bit(ICE_VF_STATE_MC_PROMISC, vf->vf_states)) && + test_bit(ICE_FLAG_VF_TRUE_PROMISC_ENA, pf->flags)) vlan_promisc = true; if (add_v) { @@ -3018,7 +3448,7 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v) if (!vid) continue; - status = ice_vsi_add_vlan(vsi, vid); + status = ice_vsi_add_vlan(vsi, vid, ICE_FWD_TO_VSI); if (status) { v_ret = VIRTCHNL_STATUS_ERR_PARAM; goto error_param; @@ -3317,6 +3747,9 @@ error_handler: case VIRTCHNL_OP_GET_STATS: err = ice_vc_get_stats_msg(vf, msg); break; + case VIRTCHNL_OP_CONFIG_PROMISCUOUS_MODE: + err = ice_vc_cfg_promiscuous_mode_msg(vf, msg); + break; case VIRTCHNL_OP_ADD_VLAN: err = ice_vc_add_vlan_msg(vf, msg); break; @@ -3390,6 +3823,39 @@ ice_get_vf_cfg(struct net_device *netdev, int vf_id, struct ifla_vf_info *ivi) } /** + * ice_unicast_mac_exists - check if the unicast MAC exists on the PF's switch + * @pf: PF used to reference the switch's rules + * @umac: unicast MAC to compare against existing switch rules + * + * Return true on the first/any match, else return false + */ +static bool ice_unicast_mac_exists(struct ice_pf *pf, u8 *umac) +{ + struct ice_sw_recipe *mac_recipe_list = + &pf->hw.switch_info->recp_list[ICE_SW_LKUP_MAC]; + struct ice_fltr_mgmt_list_entry *list_itr; + struct list_head *rule_head; + struct mutex *rule_lock; /* protect MAC filter list access */ + + rule_head = &mac_recipe_list->filt_rules; + rule_lock = &mac_recipe_list->filt_rule_lock; + + mutex_lock(rule_lock); + list_for_each_entry(list_itr, rule_head, list_entry) { + u8 *existing_mac = &list_itr->fltr_info.l_data.mac.mac_addr[0]; + + if (ether_addr_equal(existing_mac, umac)) { + mutex_unlock(rule_lock); + return true; + } + } + + mutex_unlock(rule_lock); + + return false; +} + +/** * ice_set_vf_mac * @netdev: network interface device structure * @vf_id: VF identifier @@ -3406,25 +3872,41 @@ int ice_set_vf_mac(struct net_device *netdev, int vf_id, u8 *mac) if (ice_validate_vf_id(pf, vf_id)) return -EINVAL; - if (is_zero_ether_addr(mac) || is_multicast_ether_addr(mac)) { + if (is_multicast_ether_addr(mac)) { netdev_err(netdev, "%pM not a valid unicast address\n", mac); return -EINVAL; } vf = &pf->vf[vf_id]; + /* nothing left to do, unicast MAC already set */ + if (ether_addr_equal(vf->dflt_lan_addr.addr, mac)) + return 0; + ret = ice_check_vf_ready_for_cfg(vf); if (ret) return ret; - /* copy MAC into dflt_lan_addr and trigger a VF reset. The reset - * flow will use the updated dflt_lan_addr and add a MAC filter - * using ice_add_mac. Also set pf_set_mac to indicate that the PF has - * set the MAC address for this VF. + if (ice_unicast_mac_exists(pf, mac)) { + netdev_err(netdev, "Unicast MAC %pM already exists on this PF. Preventing setting VF %u unicast MAC address to %pM\n", + mac, vf_id, mac); + return -EINVAL; + } + + /* VF is notified of its new MAC via the PF's response to the + * VIRTCHNL_OP_GET_VF_RESOURCES message after the VF has been reset */ ether_addr_copy(vf->dflt_lan_addr.addr, mac); - vf->pf_set_mac = true; - netdev_info(netdev, "MAC on VF %d set to %pM. VF driver will be reinitialized\n", - vf_id, mac); + if (is_zero_ether_addr(mac)) { + /* VF will send VIRTCHNL_OP_ADD_ETH_ADDR message with its MAC */ + vf->pf_set_mac = false; + netdev_info(netdev, "Removing MAC on VF %d. VF driver will be reinitialized\n", + vf->vf_id); + } else { + /* PF will add MAC rule for the VF */ + vf->pf_set_mac = true; + netdev_info(netdev, "Setting MAC %pM on VF %d. VF driver will be reinitialized\n", + mac, vf_id); + } ice_vc_reset_vf(vf); return 0; @@ -3554,6 +4036,24 @@ int ice_get_vf_stats(struct net_device *netdev, int vf_id, } /** + * ice_print_vf_rx_mdd_event - print VF Rx malicious driver detect event + * @vf: pointer to the VF structure + */ +void ice_print_vf_rx_mdd_event(struct ice_vf *vf) +{ + struct ice_pf *pf = vf->pf; + struct device *dev; + + dev = ice_pf_to_dev(pf); + + dev_info(dev, "%d Rx Malicious Driver Detection events detected on PF %d VF %d MAC %pM. mdd-auto-reset-vfs=%s\n", + vf->mdd_rx_events.count, pf->hw.pf_id, vf->vf_id, + vf->dflt_lan_addr.addr, + test_bit(ICE_FLAG_MDD_AUTO_RESET_VF, pf->flags) + ? "on" : "off"); +} + +/** * ice_print_vfs_mdd_event - print VFs malicious driver detect event * @pf: pointer to the PF structure * @@ -3582,12 +4082,7 @@ void ice_print_vfs_mdd_events(struct ice_pf *pf) if (vf->mdd_rx_events.count != vf->mdd_rx_events.last_printed) { vf->mdd_rx_events.last_printed = vf->mdd_rx_events.count; - - dev_info(dev, "%d Rx Malicious Driver Detection events detected on PF %d VF %d MAC %pM. mdd-auto-reset-vfs=%s\n", - vf->mdd_rx_events.count, hw->pf_id, i, - vf->dflt_lan_addr.addr, - test_bit(ICE_FLAG_MDD_AUTO_RESET_VF, pf->flags) - ? "on" : "off"); + ice_print_vf_rx_mdd_event(vf); } /* only print Tx MDD event message if there are new events */ diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h index 3f9464269bd2..67aa9110fdd1 100644 --- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h +++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.h @@ -7,7 +7,10 @@ /* Restrict number of MAC Addr and VLAN that non-trusted VF can programmed */ #define ICE_MAX_VLAN_PER_VF 8 -#define ICE_MAX_MACADDR_PER_VF 12 +/* MAC filters: 1 is reserved for the VF's default/perm_addr/LAA MAC, 1 for + * broadcast, and 16 for additional unicast/multicast filters + */ +#define ICE_MAX_MACADDR_PER_VF 18 /* Malicious Driver Detection */ #define ICE_DFLT_NUM_INVAL_MSGS_ALLOWED 10 @@ -64,7 +67,7 @@ struct ice_mdd_vf_events { struct ice_vf { struct ice_pf *pf; - s16 vf_id; /* VF ID in the PF space */ + u16 vf_id; /* VF ID in the PF space */ u16 lan_vsi_idx; /* index into PF struct */ /* first vector index of this VF in the PF space */ int first_vector_idx; @@ -128,9 +131,11 @@ void ice_set_vf_state_qs_dis(struct ice_vf *vf); int ice_get_vf_stats(struct net_device *netdev, int vf_id, struct ifla_vf_stats *vf_stats); +bool ice_is_any_vf_in_promisc(struct ice_pf *pf); void ice_vf_lan_overflow_event(struct ice_pf *pf, struct ice_rq_event_info *event); void ice_print_vfs_mdd_events(struct ice_pf *pf); +void ice_print_vf_rx_mdd_event(struct ice_vf *vf); #else /* CONFIG_PCI_IOV */ #define ice_process_vflr_event(pf) do {} while (0) #define ice_free_vfs(pf) do {} while (0) @@ -140,6 +145,7 @@ void ice_print_vfs_mdd_events(struct ice_pf *pf); #define ice_set_vf_state_qs_dis(vf) do {} while (0) #define ice_vf_lan_overflow_event(pf, event) do {} while (0) #define ice_print_vfs_mdd_events(pf) do {} while (0) +#define ice_print_vf_rx_mdd_event(vf) do {} while (0) static inline bool ice_reset_all_vfs(struct ice_pf __always_unused *pf, @@ -219,5 +225,10 @@ ice_get_vf_stats(struct net_device __always_unused *netdev, { return -EOPNOTSUPP; } + +static inline bool ice_is_any_vf_in_promisc(struct ice_pf __always_unused *pf) +{ + return false; +} #endif /* CONFIG_PCI_IOV */ #endif /* _ICE_VIRTCHNL_PF_H_ */ diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.c b/drivers/net/ethernet/intel/ice/ice_xsk.c index 8279db15e870..b6f928c9e9c9 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.c +++ b/drivers/net/ethernet/intel/ice/ice_xsk.c @@ -2,7 +2,7 @@ /* Copyright (c) 2019, Intel Corporation. */ #include <linux/bpf_trace.h> -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> #include <net/xdp.h> #include "ice.h" #include "ice_base.h" @@ -280,28 +280,6 @@ static int ice_xsk_alloc_umems(struct ice_vsi *vsi) } /** - * ice_xsk_add_umem - add a UMEM region for XDP sockets - * @vsi: VSI to which the UMEM will be added - * @umem: pointer to a requested UMEM region - * @qid: queue ID - * - * Returns 0 on success, negative on error - */ -static int ice_xsk_add_umem(struct ice_vsi *vsi, struct xdp_umem *umem, u16 qid) -{ - int err; - - err = ice_xsk_alloc_umems(vsi); - if (err) - return err; - - vsi->xsk_umems[qid] = umem; - vsi->num_xsk_umems_used++; - - return 0; -} - -/** * ice_xsk_remove_umem - Remove an UMEM for a certain ring/qid * @vsi: VSI from which the VSI will be removed * @qid: Ring/qid associated with the UMEM @@ -318,65 +296,6 @@ static void ice_xsk_remove_umem(struct ice_vsi *vsi, u16 qid) } } -/** - * ice_xsk_umem_dma_map - DMA map UMEM region for XDP sockets - * @vsi: VSI to map the UMEM region - * @umem: UMEM to map - * - * Returns 0 on success, negative on error - */ -static int ice_xsk_umem_dma_map(struct ice_vsi *vsi, struct xdp_umem *umem) -{ - struct ice_pf *pf = vsi->back; - struct device *dev; - unsigned int i; - - dev = ice_pf_to_dev(pf); - for (i = 0; i < umem->npgs; i++) { - dma_addr_t dma = dma_map_page_attrs(dev, umem->pgs[i], 0, - PAGE_SIZE, - DMA_BIDIRECTIONAL, - ICE_RX_DMA_ATTR); - if (dma_mapping_error(dev, dma)) { - dev_dbg(dev, "XSK UMEM DMA mapping error on page num %d\n", - i); - goto out_unmap; - } - - umem->pages[i].dma = dma; - } - - return 0; - -out_unmap: - for (; i > 0; i--) { - dma_unmap_page_attrs(dev, umem->pages[i].dma, PAGE_SIZE, - DMA_BIDIRECTIONAL, ICE_RX_DMA_ATTR); - umem->pages[i].dma = 0; - } - - return -EFAULT; -} - -/** - * ice_xsk_umem_dma_unmap - DMA unmap UMEM region for XDP sockets - * @vsi: VSI from which the UMEM will be unmapped - * @umem: UMEM to unmap - */ -static void ice_xsk_umem_dma_unmap(struct ice_vsi *vsi, struct xdp_umem *umem) -{ - struct ice_pf *pf = vsi->back; - struct device *dev; - unsigned int i; - - dev = ice_pf_to_dev(pf); - for (i = 0; i < umem->npgs; i++) { - dma_unmap_page_attrs(dev, umem->pages[i].dma, PAGE_SIZE, - DMA_BIDIRECTIONAL, ICE_RX_DMA_ATTR); - - umem->pages[i].dma = 0; - } -} /** * ice_xsk_umem_disable - disable a UMEM region @@ -391,7 +310,7 @@ static int ice_xsk_umem_disable(struct ice_vsi *vsi, u16 qid) !vsi->xsk_umems[qid]) return -EINVAL; - ice_xsk_umem_dma_unmap(vsi, vsi->xsk_umems[qid]); + xsk_buff_dma_unmap(vsi->xsk_umems[qid], ICE_RX_DMA_ATTR); ice_xsk_remove_umem(vsi, qid); return 0; @@ -408,7 +327,6 @@ static int ice_xsk_umem_disable(struct ice_vsi *vsi, u16 qid) static int ice_xsk_umem_enable(struct ice_vsi *vsi, struct xdp_umem *umem, u16 qid) { - struct xdp_umem_fq_reuse *reuseq; int err; if (vsi->type != ICE_VSI_PF) @@ -419,20 +337,18 @@ ice_xsk_umem_enable(struct ice_vsi *vsi, struct xdp_umem *umem, u16 qid) if (qid >= vsi->num_xsk_umems) return -EINVAL; + err = ice_xsk_alloc_umems(vsi); + if (err) + return err; + if (vsi->xsk_umems && vsi->xsk_umems[qid]) return -EBUSY; - reuseq = xsk_reuseq_prepare(vsi->rx_rings[0]->count); - if (!reuseq) - return -ENOMEM; - - xsk_reuseq_free(xsk_reuseq_swap(umem, reuseq)); - - err = ice_xsk_umem_dma_map(vsi, umem); - if (err) - return err; + vsi->xsk_umems[qid] = umem; + vsi->num_xsk_umems_used++; - err = ice_xsk_add_umem(vsi, umem, qid); + err = xsk_buff_dma_map(vsi->xsk_umems[qid], ice_pf_to_dev(vsi->back), + ICE_RX_DMA_ATTR); if (err) return err; @@ -484,137 +400,22 @@ xsk_umem_if_up: } /** - * ice_zca_free - Callback for MEM_TYPE_ZERO_COPY allocations - * @zca: zero-cpoy allocator - * @handle: Buffer handle - */ -void ice_zca_free(struct zero_copy_allocator *zca, unsigned long handle) -{ - struct ice_rx_buf *rx_buf; - struct ice_ring *rx_ring; - struct xdp_umem *umem; - u64 hr, mask; - u16 nta; - - rx_ring = container_of(zca, struct ice_ring, zca); - umem = rx_ring->xsk_umem; - hr = umem->headroom + XDP_PACKET_HEADROOM; - - mask = umem->chunk_mask; - - nta = rx_ring->next_to_alloc; - rx_buf = &rx_ring->rx_buf[nta]; - - nta++; - rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0; - - handle &= mask; - - rx_buf->dma = xdp_umem_get_dma(umem, handle); - rx_buf->dma += hr; - - rx_buf->addr = xdp_umem_get_data(umem, handle); - rx_buf->addr += hr; - - rx_buf->handle = (u64)handle + umem->headroom; -} - -/** - * ice_alloc_buf_fast_zc - Retrieve buffer address from XDP umem - * @rx_ring: ring with an xdp_umem bound to it - * @rx_buf: buffer to which xsk page address will be assigned - * - * This function allocates an Rx buffer in the hot path. - * The buffer can come from fill queue or recycle queue. - * - * Returns true if an assignment was successful, false if not. - */ -static __always_inline bool -ice_alloc_buf_fast_zc(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf) -{ - struct xdp_umem *umem = rx_ring->xsk_umem; - void *addr = rx_buf->addr; - u64 handle, hr; - - if (addr) { - rx_ring->rx_stats.page_reuse_count++; - return true; - } - - if (!xsk_umem_peek_addr(umem, &handle)) { - rx_ring->rx_stats.alloc_page_failed++; - return false; - } - - hr = umem->headroom + XDP_PACKET_HEADROOM; - - rx_buf->dma = xdp_umem_get_dma(umem, handle); - rx_buf->dma += hr; - - rx_buf->addr = xdp_umem_get_data(umem, handle); - rx_buf->addr += hr; - - rx_buf->handle = handle + umem->headroom; - - xsk_umem_release_addr(umem); - return true; -} - -/** - * ice_alloc_buf_slow_zc - Retrieve buffer address from XDP umem - * @rx_ring: ring with an xdp_umem bound to it - * @rx_buf: buffer to which xsk page address will be assigned - * - * This function allocates an Rx buffer in the slow path. - * The buffer can come from fill queue or recycle queue. - * - * Returns true if an assignment was successful, false if not. - */ -static __always_inline bool -ice_alloc_buf_slow_zc(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf) -{ - struct xdp_umem *umem = rx_ring->xsk_umem; - u64 handle, headroom; - - if (!xsk_umem_peek_addr_rq(umem, &handle)) { - rx_ring->rx_stats.alloc_page_failed++; - return false; - } - - handle &= umem->chunk_mask; - headroom = umem->headroom + XDP_PACKET_HEADROOM; - - rx_buf->dma = xdp_umem_get_dma(umem, handle); - rx_buf->dma += headroom; - - rx_buf->addr = xdp_umem_get_data(umem, handle); - rx_buf->addr += headroom; - - rx_buf->handle = handle + umem->headroom; - - xsk_umem_release_addr_rq(umem); - return true; -} - -/** * ice_alloc_rx_bufs_zc - allocate a number of Rx buffers * @rx_ring: Rx ring * @count: The number of buffers to allocate - * @alloc: the function pointer to call for allocation * * This function allocates a number of Rx buffers from the fill ring * or the internal recycle mechanism and places them on the Rx ring. * * Returns false if all allocations were successful, true if any fail. */ -static bool -ice_alloc_rx_bufs_zc(struct ice_ring *rx_ring, int count, - bool (*alloc)(struct ice_ring *, struct ice_rx_buf *)) +bool ice_alloc_rx_bufs_zc(struct ice_ring *rx_ring, u16 count) { union ice_32b_rx_flex_desc *rx_desc; u16 ntu = rx_ring->next_to_use; struct ice_rx_buf *rx_buf; bool ret = false; + dma_addr_t dma; if (!count) return false; @@ -623,16 +424,14 @@ ice_alloc_rx_bufs_zc(struct ice_ring *rx_ring, int count, rx_buf = &rx_ring->rx_buf[ntu]; do { - if (!alloc(rx_ring, rx_buf)) { + rx_buf->xdp = xsk_buff_alloc(rx_ring->xsk_umem); + if (!rx_buf->xdp) { ret = true; break; } - dma_sync_single_range_for_device(rx_ring->dev, rx_buf->dma, 0, - rx_ring->rx_buf_len, - DMA_BIDIRECTIONAL); - - rx_desc->read.pkt_addr = cpu_to_le64(rx_buf->dma); + dma = xsk_buff_xdp_get_dma(rx_buf->xdp); + rx_desc->read.pkt_addr = cpu_to_le64(dma); rx_desc->wb.status_error0 = 0; rx_desc++; @@ -653,32 +452,6 @@ ice_alloc_rx_bufs_zc(struct ice_ring *rx_ring, int count, } /** - * ice_alloc_rx_bufs_fast_zc - allocate zero copy bufs in the hot path - * @rx_ring: Rx ring - * @count: number of bufs to allocate - * - * Returns false on success, true on failure. - */ -static bool ice_alloc_rx_bufs_fast_zc(struct ice_ring *rx_ring, u16 count) -{ - return ice_alloc_rx_bufs_zc(rx_ring, count, - ice_alloc_buf_fast_zc); -} - -/** - * ice_alloc_rx_bufs_slow_zc - allocate zero copy bufs in the slow path - * @rx_ring: Rx ring - * @count: number of bufs to allocate - * - * Returns false on success, true on failure. - */ -bool ice_alloc_rx_bufs_slow_zc(struct ice_ring *rx_ring, u16 count) -{ - return ice_alloc_rx_bufs_zc(rx_ring, count, - ice_alloc_buf_slow_zc); -} - -/** * ice_bump_ntc - Bump the next_to_clean counter of an Rx ring * @rx_ring: Rx ring */ @@ -692,76 +465,21 @@ static void ice_bump_ntc(struct ice_ring *rx_ring) } /** - * ice_get_rx_buf_zc - Fetch the current Rx buffer - * @rx_ring: Rx ring - * @size: size of a buffer - * - * This function returns the current, received Rx buffer and does - * DMA synchronization. - * - * Returns a pointer to the received Rx buffer. - */ -static struct ice_rx_buf *ice_get_rx_buf_zc(struct ice_ring *rx_ring, int size) -{ - struct ice_rx_buf *rx_buf; - - rx_buf = &rx_ring->rx_buf[rx_ring->next_to_clean]; - - dma_sync_single_range_for_cpu(rx_ring->dev, rx_buf->dma, 0, - size, DMA_BIDIRECTIONAL); - - return rx_buf; -} - -/** - * ice_reuse_rx_buf_zc - reuse an Rx buffer - * @rx_ring: Rx ring - * @old_buf: The buffer to recycle - * - * This function recycles a finished Rx buffer, and places it on the recycle - * queue (next_to_alloc). - */ -static void -ice_reuse_rx_buf_zc(struct ice_ring *rx_ring, struct ice_rx_buf *old_buf) -{ - unsigned long mask = (unsigned long)rx_ring->xsk_umem->chunk_mask; - u64 hr = rx_ring->xsk_umem->headroom + XDP_PACKET_HEADROOM; - u16 nta = rx_ring->next_to_alloc; - struct ice_rx_buf *new_buf; - - new_buf = &rx_ring->rx_buf[nta++]; - rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0; - - new_buf->dma = old_buf->dma & mask; - new_buf->dma += hr; - - new_buf->addr = (void *)((unsigned long)old_buf->addr & mask); - new_buf->addr += hr; - - new_buf->handle = old_buf->handle & mask; - new_buf->handle += rx_ring->xsk_umem->headroom; - - old_buf->addr = NULL; -} - -/** * ice_construct_skb_zc - Create an sk_buff from zero-copy buffer * @rx_ring: Rx ring * @rx_buf: zero-copy Rx buffer - * @xdp: XDP buffer * * This function allocates a new skb from a zero-copy Rx buffer. * * Returns the skb on success, NULL on failure. */ static struct sk_buff * -ice_construct_skb_zc(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf, - struct xdp_buff *xdp) +ice_construct_skb_zc(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf) { - unsigned int metasize = xdp->data - xdp->data_meta; - unsigned int datasize = xdp->data_end - xdp->data; - unsigned int datasize_hard = xdp->data_end - - xdp->data_hard_start; + unsigned int metasize = rx_buf->xdp->data - rx_buf->xdp->data_meta; + unsigned int datasize = rx_buf->xdp->data_end - rx_buf->xdp->data; + unsigned int datasize_hard = rx_buf->xdp->data_end - + rx_buf->xdp->data_hard_start; struct sk_buff *skb; skb = __napi_alloc_skb(&rx_ring->q_vector->napi, datasize_hard, @@ -769,13 +487,13 @@ ice_construct_skb_zc(struct ice_ring *rx_ring, struct ice_rx_buf *rx_buf, if (unlikely(!skb)) return NULL; - skb_reserve(skb, xdp->data - xdp->data_hard_start); - memcpy(__skb_put(skb, datasize), xdp->data, datasize); + skb_reserve(skb, rx_buf->xdp->data - rx_buf->xdp->data_hard_start); + memcpy(__skb_put(skb, datasize), rx_buf->xdp->data, datasize); if (metasize) skb_metadata_set(skb, metasize); - ice_reuse_rx_buf_zc(rx_ring, rx_buf); - + xsk_buff_free(rx_buf->xdp); + rx_buf->xdp = NULL; return skb; } @@ -802,7 +520,6 @@ ice_run_xdp_zc(struct ice_ring *rx_ring, struct xdp_buff *xdp) } act = bpf_prog_run_xdp(xdp_prog, xdp); - xdp->handle += xdp->data - xdp->data_hard_start; switch (act) { case XDP_PASS: break; @@ -842,9 +559,6 @@ int ice_clean_rx_irq_zc(struct ice_ring *rx_ring, int budget) u16 cleaned_count = ICE_DESC_UNUSED(rx_ring); unsigned int xdp_xmit = 0; bool failure = false; - struct xdp_buff xdp; - - xdp.rxq = &rx_ring->xdp_rxq; while (likely(total_rx_packets < (unsigned int)budget)) { union ice_32b_rx_flex_desc *rx_desc; @@ -856,8 +570,8 @@ int ice_clean_rx_irq_zc(struct ice_ring *rx_ring, int budget) u8 rx_ptype; if (cleaned_count >= ICE_RX_BUF_WRITE) { - failure |= ice_alloc_rx_bufs_fast_zc(rx_ring, - cleaned_count); + failure |= ice_alloc_rx_bufs_zc(rx_ring, + cleaned_count); cleaned_count = 0; } @@ -878,25 +592,19 @@ int ice_clean_rx_irq_zc(struct ice_ring *rx_ring, int budget) if (!size) break; - rx_buf = ice_get_rx_buf_zc(rx_ring, size); - if (!rx_buf->addr) - break; - xdp.data = rx_buf->addr; - xdp.data_meta = xdp.data; - xdp.data_hard_start = xdp.data - XDP_PACKET_HEADROOM; - xdp.data_end = xdp.data + size; - xdp.handle = rx_buf->handle; + rx_buf = &rx_ring->rx_buf[rx_ring->next_to_clean]; + rx_buf->xdp->data_end = rx_buf->xdp->data + size; + xsk_buff_dma_sync_for_cpu(rx_buf->xdp); - xdp_res = ice_run_xdp_zc(rx_ring, &xdp); + xdp_res = ice_run_xdp_zc(rx_ring, rx_buf->xdp); if (xdp_res) { - if (xdp_res & (ICE_XDP_TX | ICE_XDP_REDIR)) { + if (xdp_res & (ICE_XDP_TX | ICE_XDP_REDIR)) xdp_xmit |= xdp_res; - rx_buf->addr = NULL; - } else { - ice_reuse_rx_buf_zc(rx_ring, rx_buf); - } + else + xsk_buff_free(rx_buf->xdp); + rx_buf->xdp = NULL; total_rx_bytes += size; total_rx_packets++; cleaned_count++; @@ -906,7 +614,7 @@ int ice_clean_rx_irq_zc(struct ice_ring *rx_ring, int budget) } /* XDP_PASS path */ - skb = ice_construct_skb_zc(rx_ring, rx_buf, &xdp); + skb = ice_construct_skb_zc(rx_ring, rx_buf); if (!skb) { rx_ring->rx_stats.alloc_buf_failed++; break; @@ -977,17 +685,16 @@ static bool ice_xmit_zc(struct ice_ring *xdp_ring, int budget) if (!xsk_umem_consume_tx(xdp_ring->xsk_umem, &desc)) break; - dma = xdp_umem_get_dma(xdp_ring->xsk_umem, desc.addr); - - dma_sync_single_for_device(xdp_ring->dev, dma, desc.len, - DMA_BIDIRECTIONAL); + dma = xsk_buff_raw_get_dma(xdp_ring->xsk_umem, desc.addr); + xsk_buff_raw_dma_sync_for_device(xdp_ring->xsk_umem, dma, + desc.len); tx_buf->bytecount = desc.len; tx_desc = ICE_TX_DESC(xdp_ring, xdp_ring->next_to_use); tx_desc->buf_addr = cpu_to_le64(dma); - tx_desc->cmd_type_offset_bsz = build_ctob(ICE_TXD_LAST_DESC_CMD, - 0, desc.len, 0); + tx_desc->cmd_type_offset_bsz = + ice_build_ctob(ICE_TXD_LAST_DESC_CMD, 0, desc.len, 0); xdp_ring->next_to_use++; if (xdp_ring->next_to_use == xdp_ring->count) @@ -1163,11 +870,10 @@ void ice_xsk_clean_rx_ring(struct ice_ring *rx_ring) for (i = 0; i < rx_ring->count; i++) { struct ice_rx_buf *rx_buf = &rx_ring->rx_buf[i]; - if (!rx_buf->addr) + if (!rx_buf->xdp) continue; - xsk_umem_fq_reuse(rx_ring->xsk_umem, rx_buf->handle); - rx_buf->addr = NULL; + rx_buf->xdp = NULL; } } diff --git a/drivers/net/ethernet/intel/ice/ice_xsk.h b/drivers/net/ethernet/intel/ice/ice_xsk.h index 8a4ba7c6d549..fc1a06b4df36 100644 --- a/drivers/net/ethernet/intel/ice/ice_xsk.h +++ b/drivers/net/ethernet/intel/ice/ice_xsk.h @@ -10,11 +10,10 @@ struct ice_vsi; #ifdef CONFIG_XDP_SOCKETS int ice_xsk_umem_setup(struct ice_vsi *vsi, struct xdp_umem *umem, u16 qid); -void ice_zca_free(struct zero_copy_allocator *zca, unsigned long handle); int ice_clean_rx_irq_zc(struct ice_ring *rx_ring, int budget); bool ice_clean_tx_irq_zc(struct ice_ring *xdp_ring, int budget); int ice_xsk_wakeup(struct net_device *netdev, u32 queue_id, u32 flags); -bool ice_alloc_rx_bufs_slow_zc(struct ice_ring *rx_ring, u16 count); +bool ice_alloc_rx_bufs_zc(struct ice_ring *rx_ring, u16 count); bool ice_xsk_any_rx_ring_ena(struct ice_vsi *vsi); void ice_xsk_clean_rx_ring(struct ice_ring *rx_ring); void ice_xsk_clean_xdp_ring(struct ice_ring *xdp_ring); @@ -27,12 +26,6 @@ ice_xsk_umem_setup(struct ice_vsi __always_unused *vsi, return -EOPNOTSUPP; } -static inline void -ice_zca_free(struct zero_copy_allocator __always_unused *zca, - unsigned long __always_unused handle) -{ -} - static inline int ice_clean_rx_irq_zc(struct ice_ring __always_unused *rx_ring, int __always_unused budget) @@ -48,8 +41,8 @@ ice_clean_tx_irq_zc(struct ice_ring __always_unused *xdp_ring, } static inline bool -ice_alloc_rx_bufs_slow_zc(struct ice_ring __always_unused *rx_ring, - u16 __always_unused count) +ice_alloc_rx_bufs_zc(struct ice_ring __always_unused *rx_ring, + u16 __always_unused count) { return false; } diff --git a/drivers/net/ethernet/intel/igb/e1000_mac.c b/drivers/net/ethernet/intel/igb/e1000_mac.c index 79ee0a747260..3254737c07a3 100644 --- a/drivers/net/ethernet/intel/igb/e1000_mac.c +++ b/drivers/net/ethernet/intel/igb/e1000_mac.c @@ -12,7 +12,7 @@ #include "igb.h" static s32 igb_set_default_fc(struct e1000_hw *hw); -static s32 igb_set_fc_watermarks(struct e1000_hw *hw); +static void igb_set_fc_watermarks(struct e1000_hw *hw); /** * igb_get_bus_info_pcie - Get PCIe bus information @@ -687,7 +687,7 @@ s32 igb_setup_link(struct e1000_hw *hw) wr32(E1000_FCTTV, hw->fc.pause_time); - ret_val = igb_set_fc_watermarks(hw); + igb_set_fc_watermarks(hw); out: @@ -723,9 +723,8 @@ void igb_config_collision_dist(struct e1000_hw *hw) * flow control XON frame transmission is enabled, then set XON frame * tansmission as well. **/ -static s32 igb_set_fc_watermarks(struct e1000_hw *hw) +static void igb_set_fc_watermarks(struct e1000_hw *hw) { - s32 ret_val = 0; u32 fcrtl = 0, fcrth = 0; /* Set the flow control receive threshold registers. Normally, @@ -747,8 +746,6 @@ static s32 igb_set_fc_watermarks(struct e1000_hw *hw) } wr32(E1000_FCRTL, fcrtl); wr32(E1000_FCRTH, fcrth); - - return ret_val; } /** diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c index 39d3b76a6f5d..2cd003c5ad43 100644 --- a/drivers/net/ethernet/intel/igb/igb_ethtool.c +++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c @@ -143,7 +143,8 @@ static int igb_get_link_ksettings(struct net_device *netdev, u32 speed; u32 supported, advertising; - status = rd32(E1000_STATUS); + status = pm_runtime_suspended(&adapter->pdev->dev) ? + 0 : rd32(E1000_STATUS); if (hw->phy.media_type == e1000_media_type_copper) { supported = (SUPPORTED_10baseT_Half | diff --git a/drivers/net/ethernet/intel/igb/igb_main.c b/drivers/net/ethernet/intel/igb/igb_main.c index b46bff8fe056..8bb3db2cbd41 100644 --- a/drivers/net/ethernet/intel/igb/igb_main.c +++ b/drivers/net/ethernet/intel/igb/igb_main.c @@ -3445,7 +3445,7 @@ static int igb_probe(struct pci_dev *pdev, const struct pci_device_id *ent) } } - dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NEVER_SKIP); + dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); pm_runtime_put_noidle(&pdev->dev); return 0; diff --git a/drivers/net/ethernet/intel/igc/Makefile b/drivers/net/ethernet/intel/igc/Makefile index e3c164c12e10..1c3051db9085 100644 --- a/drivers/net/ethernet/intel/igc/Makefile +++ b/drivers/net/ethernet/intel/igc/Makefile @@ -8,4 +8,4 @@ obj-$(CONFIG_IGC) += igc.o igc-objs := igc_main.o igc_mac.o igc_i225.o igc_base.o igc_nvm.o igc_phy.o \ -igc_ethtool.o igc_ptp.o igc_dump.o +igc_diag.o igc_ethtool.o igc_ptp.o igc_dump.o igc_tsn.o diff --git a/drivers/net/ethernet/intel/igc/igc.h b/drivers/net/ethernet/intel/igc/igc.h index a1f845a2aa80..5dbc5a156626 100644 --- a/drivers/net/ethernet/intel/igc/igc.h +++ b/drivers/net/ethernet/intel/igc/igc.h @@ -16,14 +16,210 @@ #include "igc_hw.h" -/* forward declaration */ -void igc_set_ethtool_ops(struct net_device *); +void igc_ethtool_set_ops(struct net_device *); -struct igc_adapter; -struct igc_ring; +/* Transmit and receive queues */ +#define IGC_MAX_RX_QUEUES 4 +#define IGC_MAX_TX_QUEUES 4 + +#define MAX_Q_VECTORS 8 +#define MAX_STD_JUMBO_FRAME_SIZE 9216 + +#define MAX_ETYPE_FILTER 8 +#define IGC_RETA_SIZE 128 + +enum igc_mac_filter_type { + IGC_MAC_FILTER_TYPE_DST = 0, + IGC_MAC_FILTER_TYPE_SRC +}; + +struct igc_tx_queue_stats { + u64 packets; + u64 bytes; + u64 restart_queue; + u64 restart_queue2; +}; + +struct igc_rx_queue_stats { + u64 packets; + u64 bytes; + u64 drops; + u64 csum_err; + u64 alloc_failed; +}; + +struct igc_rx_packet_stats { + u64 ipv4_packets; /* IPv4 headers processed */ + u64 ipv4e_packets; /* IPv4E headers with extensions processed */ + u64 ipv6_packets; /* IPv6 headers processed */ + u64 ipv6e_packets; /* IPv6E headers with extensions processed */ + u64 tcp_packets; /* TCP headers processed */ + u64 udp_packets; /* UDP headers processed */ + u64 sctp_packets; /* SCTP headers processed */ + u64 nfs_packets; /* NFS headers processe */ + u64 other_packets; +}; + +struct igc_ring_container { + struct igc_ring *ring; /* pointer to linked list of rings */ + unsigned int total_bytes; /* total bytes processed this int */ + unsigned int total_packets; /* total packets processed this int */ + u16 work_limit; /* total work allowed per interrupt */ + u8 count; /* total number of rings in vector */ + u8 itr; /* current ITR setting for ring */ +}; + +struct igc_ring { + struct igc_q_vector *q_vector; /* backlink to q_vector */ + struct net_device *netdev; /* back pointer to net_device */ + struct device *dev; /* device for dma mapping */ + union { /* array of buffer info structs */ + struct igc_tx_buffer *tx_buffer_info; + struct igc_rx_buffer *rx_buffer_info; + }; + void *desc; /* descriptor ring memory */ + unsigned long flags; /* ring specific flags */ + void __iomem *tail; /* pointer to ring tail register */ + dma_addr_t dma; /* phys address of the ring */ + unsigned int size; /* length of desc. ring in bytes */ + + u16 count; /* number of desc. in the ring */ + u8 queue_index; /* logical index of the ring*/ + u8 reg_idx; /* physical index of the ring */ + bool launchtime_enable; /* true if LaunchTime is enabled */ + + u32 start_time; + u32 end_time; + + /* everything past this point are written often */ + u16 next_to_clean; + u16 next_to_use; + u16 next_to_alloc; + + union { + /* TX */ + struct { + struct igc_tx_queue_stats tx_stats; + struct u64_stats_sync tx_syncp; + struct u64_stats_sync tx_syncp2; + }; + /* RX */ + struct { + struct igc_rx_queue_stats rx_stats; + struct igc_rx_packet_stats pkt_stats; + struct u64_stats_sync rx_syncp; + struct sk_buff *skb; + }; + }; +} ____cacheline_internodealigned_in_smp; + +/* Board specific private data structure */ +struct igc_adapter { + struct net_device *netdev; + + unsigned long state; + unsigned int flags; + unsigned int num_q_vectors; + + struct msix_entry *msix_entries; + + /* TX */ + u16 tx_work_limit; + u32 tx_timeout_count; + int num_tx_queues; + struct igc_ring *tx_ring[IGC_MAX_TX_QUEUES]; + + /* RX */ + int num_rx_queues; + struct igc_ring *rx_ring[IGC_MAX_RX_QUEUES]; + + struct timer_list watchdog_timer; + struct timer_list dma_err_timer; + struct timer_list phy_info_timer; + + u32 wol; + u32 en_mng_pt; + u16 link_speed; + u16 link_duplex; + + u8 port_num; + + u8 __iomem *io_addr; + /* Interrupt Throttle Rate */ + u32 rx_itr_setting; + u32 tx_itr_setting; + + struct work_struct reset_task; + struct work_struct watchdog_task; + struct work_struct dma_err_task; + bool fc_autoneg; + + u8 tx_timeout_factor; + + int msg_enable; + u32 max_frame_size; + u32 min_frame_size; + + ktime_t base_time; + ktime_t cycle_time; + + /* OS defined structs */ + struct pci_dev *pdev; + /* lock for statistics */ + spinlock_t stats64_lock; + struct rtnl_link_stats64 stats64; + + /* structs defined in igc_hw.h */ + struct igc_hw hw; + struct igc_hw_stats stats; + + struct igc_q_vector *q_vector[MAX_Q_VECTORS]; + u32 eims_enable_mask; + u32 eims_other; + + u16 tx_ring_count; + u16 rx_ring_count; + + u32 tx_hwtstamp_timeouts; + u32 tx_hwtstamp_skipped; + u32 rx_hwtstamp_cleared; + + u32 rss_queues; + u32 rss_indir_tbl_init; + + /* Any access to elements in nfc_rule_list is protected by the + * nfc_rule_lock. + */ + struct mutex nfc_rule_lock; + struct list_head nfc_rule_list; + unsigned int nfc_rule_count; + + u8 rss_indir_tbl[IGC_RETA_SIZE]; + + unsigned long link_check_timeout; + struct igc_info ei; + + u32 test_icr; + + struct ptp_clock *ptp_clock; + struct ptp_clock_info ptp_caps; + struct work_struct ptp_tx_work; + struct sk_buff *ptp_tx_skb; + struct hwtstamp_config tstamp_config; + unsigned long ptp_tx_start; + unsigned long last_rx_ptp_check; + unsigned long last_rx_timestamp; + unsigned int ptp_flags; + /* System time value lock */ + spinlock_t tmreg_lock; + struct cyclecounter cc; + struct timecounter tc; +}; void igc_up(struct igc_adapter *adapter); void igc_down(struct igc_adapter *adapter); +int igc_open(struct net_device *netdev); +int igc_close(struct net_device *netdev); int igc_setup_tx_resources(struct igc_ring *ring); int igc_setup_rx_resources(struct igc_ring *ring); void igc_free_tx_resources(struct igc_ring *ring); @@ -36,10 +232,6 @@ void igc_write_rss_indir_tbl(struct igc_adapter *adapter); bool igc_has_link(struct igc_adapter *adapter); void igc_reset(struct igc_adapter *adapter); int igc_set_spd_dplx(struct igc_adapter *adapter, u32 spd, u8 dplx); -int igc_add_mac_steering_filter(struct igc_adapter *adapter, - const u8 *addr, u8 queue, u8 flags); -int igc_del_mac_steering_filter(struct igc_adapter *adapter, - const u8 *addr, u8 queue, u8 flags); void igc_update_stats(struct igc_adapter *adapter); /* igc_dump declarations */ @@ -50,14 +242,10 @@ extern char igc_driver_name[]; extern char igc_driver_version[]; #define IGC_REGS_LEN 740 -#define IGC_RETA_SIZE 128 /* flags controlling PTP/1588 function */ #define IGC_PTP_ENABLED BIT(0) -/* Interrupt defines */ -#define IGC_START_ITR 648 /* ~6000 ints/sec */ - /* Flags definitions */ #define IGC_FLAG_HAS_MSI BIT(0) #define IGC_FLAG_QUEUE_PAIRS BIT(3) @@ -70,6 +258,7 @@ extern char igc_driver_version[]; #define IGC_FLAG_HAS_MSIX BIT(13) #define IGC_FLAG_VLAN_PROMISC BIT(15) #define IGC_FLAG_RX_LEGACY BIT(16) +#define IGC_FLAG_TSN_QBV_ENABLED BIT(17) #define IGC_FLAG_RSS_FIELD_IPV4_UDP BIT(6) #define IGC_FLAG_RSS_FIELD_IPV6_UDP BIT(7) @@ -78,6 +267,7 @@ extern char igc_driver_version[]; #define IGC_MRQC_RSS_FIELD_IPV4_UDP 0x00400000 #define IGC_MRQC_RSS_FIELD_IPV6_UDP 0x00800000 +/* Interrupt defines */ #define IGC_START_ITR 648 /* ~6000 ints/sec */ #define IGC_4K_ITR 980 #define IGC_20K_ITR 196 @@ -99,13 +289,6 @@ extern char igc_driver_version[]; #define IGC_MIN_RXD 80 #define IGC_MAX_RXD 4096 -/* Transmit and receive queues */ -#define IGC_MAX_RX_QUEUES 4 -#define IGC_MAX_TX_QUEUES 4 - -#define MAX_Q_VECTORS 8 -#define MAX_STD_JUMBO_FRAME_SIZE 9216 - /* Supported Rx Buffer Sizes */ #define IGC_RXBUFFER_256 256 #define IGC_RXBUFFER_2048 2048 @@ -232,83 +415,6 @@ struct igc_rx_buffer { __u16 pagecnt_bias; }; -struct igc_tx_queue_stats { - u64 packets; - u64 bytes; - u64 restart_queue; - u64 restart_queue2; -}; - -struct igc_rx_queue_stats { - u64 packets; - u64 bytes; - u64 drops; - u64 csum_err; - u64 alloc_failed; -}; - -struct igc_rx_packet_stats { - u64 ipv4_packets; /* IPv4 headers processed */ - u64 ipv4e_packets; /* IPv4E headers with extensions processed */ - u64 ipv6_packets; /* IPv6 headers processed */ - u64 ipv6e_packets; /* IPv6E headers with extensions processed */ - u64 tcp_packets; /* TCP headers processed */ - u64 udp_packets; /* UDP headers processed */ - u64 sctp_packets; /* SCTP headers processed */ - u64 nfs_packets; /* NFS headers processe */ - u64 other_packets; -}; - -struct igc_ring_container { - struct igc_ring *ring; /* pointer to linked list of rings */ - unsigned int total_bytes; /* total bytes processed this int */ - unsigned int total_packets; /* total packets processed this int */ - u16 work_limit; /* total work allowed per interrupt */ - u8 count; /* total number of rings in vector */ - u8 itr; /* current ITR setting for ring */ -}; - -struct igc_ring { - struct igc_q_vector *q_vector; /* backlink to q_vector */ - struct net_device *netdev; /* back pointer to net_device */ - struct device *dev; /* device for dma mapping */ - union { /* array of buffer info structs */ - struct igc_tx_buffer *tx_buffer_info; - struct igc_rx_buffer *rx_buffer_info; - }; - void *desc; /* descriptor ring memory */ - unsigned long flags; /* ring specific flags */ - void __iomem *tail; /* pointer to ring tail register */ - dma_addr_t dma; /* phys address of the ring */ - unsigned int size; /* length of desc. ring in bytes */ - - u16 count; /* number of desc. in the ring */ - u8 queue_index; /* logical index of the ring*/ - u8 reg_idx; /* physical index of the ring */ - bool launchtime_enable; /* true if LaunchTime is enabled */ - - /* everything past this point are written often */ - u16 next_to_clean; - u16 next_to_use; - u16 next_to_alloc; - - union { - /* TX */ - struct { - struct igc_tx_queue_stats tx_stats; - struct u64_stats_sync tx_syncp; - struct u64_stats_sync tx_syncp2; - }; - /* RX */ - struct { - struct igc_rx_queue_stats rx_stats; - struct igc_rx_packet_stats pkt_stats; - struct u64_stats_sync rx_syncp; - struct sk_buff *skb; - }; - }; -} ____cacheline_internodealigned_in_smp; - struct igc_q_vector { struct igc_adapter *adapter; /* backlink */ void __iomem *itr_register; @@ -329,8 +435,6 @@ struct igc_q_vector { struct igc_ring ring[] ____cacheline_internodealigned_in_smp; }; -#define MAX_ETYPE_FILTER (4 - 1) - enum igc_filter_match_flags { IGC_FILTER_FLAG_ETHER_TYPE = 0x1, IGC_FILTER_FLAG_VLAN_TCI = 0x2, @@ -338,143 +442,25 @@ enum igc_filter_match_flags { IGC_FILTER_FLAG_DST_MAC_ADDR = 0x8, }; -/* RX network flow classification data structure */ -struct igc_nfc_input { - /* Byte layout in order, all values with MSB first: - * match_flags - 1 byte - * etype - 2 bytes - * vlan_tci - 2 bytes - */ +struct igc_nfc_filter { u8 match_flags; - __be16 etype; - __be16 vlan_tci; + u16 etype; + u16 vlan_tci; u8 src_addr[ETH_ALEN]; u8 dst_addr[ETH_ALEN]; }; -struct igc_nfc_filter { - struct hlist_node nfc_node; - struct igc_nfc_input filter; - unsigned long cookie; - u16 etype_reg_index; - u16 sw_idx; +struct igc_nfc_rule { + struct list_head list; + struct igc_nfc_filter filter; + u32 location; u16 action; }; -struct igc_mac_addr { - u8 addr[ETH_ALEN]; - u8 queue; - u8 state; /* bitmask */ -}; - -#define IGC_MAC_STATE_DEFAULT 0x1 -#define IGC_MAC_STATE_IN_USE 0x2 -#define IGC_MAC_STATE_SRC_ADDR 0x4 -#define IGC_MAC_STATE_QUEUE_STEERING 0x8 - -#define IGC_MAX_RXNFC_FILTERS 16 - -/* Board specific private data structure */ -struct igc_adapter { - struct net_device *netdev; - - unsigned long state; - unsigned int flags; - unsigned int num_q_vectors; - - struct msix_entry *msix_entries; - - /* TX */ - u16 tx_work_limit; - u32 tx_timeout_count; - int num_tx_queues; - struct igc_ring *tx_ring[IGC_MAX_TX_QUEUES]; - - /* RX */ - int num_rx_queues; - struct igc_ring *rx_ring[IGC_MAX_RX_QUEUES]; - - struct timer_list watchdog_timer; - struct timer_list dma_err_timer; - struct timer_list phy_info_timer; - - u32 wol; - u32 en_mng_pt; - u16 link_speed; - u16 link_duplex; - - u8 port_num; - - u8 __iomem *io_addr; - /* Interrupt Throttle Rate */ - u32 rx_itr_setting; - u32 tx_itr_setting; - - struct work_struct reset_task; - struct work_struct watchdog_task; - struct work_struct dma_err_task; - bool fc_autoneg; - - u8 tx_timeout_factor; - - int msg_enable; - u32 max_frame_size; - u32 min_frame_size; - - /* OS defined structs */ - struct pci_dev *pdev; - /* lock for statistics */ - spinlock_t stats64_lock; - struct rtnl_link_stats64 stats64; - - /* structs defined in igc_hw.h */ - struct igc_hw hw; - struct igc_hw_stats stats; - - struct igc_q_vector *q_vector[MAX_Q_VECTORS]; - u32 eims_enable_mask; - u32 eims_other; - - u16 tx_ring_count; - u16 rx_ring_count; - - u32 tx_hwtstamp_timeouts; - u32 tx_hwtstamp_skipped; - u32 rx_hwtstamp_cleared; - - u32 rss_queues; - u32 rss_indir_tbl_init; - - /* RX network flow classification support */ - struct hlist_head nfc_filter_list; - struct hlist_head cls_flower_list; - unsigned int nfc_filter_count; - - /* lock for RX network flow classification filter */ - spinlock_t nfc_lock; - bool etype_bitmap[MAX_ETYPE_FILTER]; - - struct igc_mac_addr *mac_table; - - u8 rss_indir_tbl[IGC_RETA_SIZE]; - - unsigned long link_check_timeout; - struct igc_info ei; - - struct ptp_clock *ptp_clock; - struct ptp_clock_info ptp_caps; - struct work_struct ptp_tx_work; - struct sk_buff *ptp_tx_skb; - struct hwtstamp_config tstamp_config; - unsigned long ptp_tx_start; - unsigned long last_rx_ptp_check; - unsigned long last_rx_timestamp; - unsigned int ptp_flags; - /* System time value lock */ - spinlock_t tmreg_lock; - struct cyclecounter cc; - struct timecounter tc; -}; +/* IGC supports a total of 32 NFC rules: 16 MAC address based,, 8 VLAN priority + * based, and 8 ethertype based. + */ +#define IGC_MAX_RXNFC_RULES 32 /* igc_desc_unused - calculate if we have unused descriptors */ static inline u16 igc_desc_unused(const struct igc_ring *ring) @@ -550,12 +536,11 @@ static inline s32 igc_read_phy_reg(struct igc_hw *hw, u32 offset, u16 *data) return 0; } -/* forward declaration */ void igc_reinit_locked(struct igc_adapter *); -int igc_add_filter(struct igc_adapter *adapter, - struct igc_nfc_filter *input); -int igc_erase_filter(struct igc_adapter *adapter, - struct igc_nfc_filter *input); +struct igc_nfc_rule *igc_get_nfc_rule(struct igc_adapter *adapter, + u32 location); +int igc_add_nfc_rule(struct igc_adapter *adapter, struct igc_nfc_rule *rule); +void igc_del_nfc_rule(struct igc_adapter *adapter, struct igc_nfc_rule *rule); void igc_ptp_init(struct igc_adapter *adapter); void igc_ptp_reset(struct igc_adapter *adapter); diff --git a/drivers/net/ethernet/intel/igc/igc_base.c b/drivers/net/ethernet/intel/igc/igc_base.c index 5a506440560a..cc5a6cf531c7 100644 --- a/drivers/net/ethernet/intel/igc/igc_base.c +++ b/drivers/net/ethernet/intel/igc/igc_base.c @@ -26,7 +26,7 @@ static s32 igc_reset_hw_base(struct igc_hw *hw) */ ret_val = igc_disable_pcie_master(hw); if (ret_val) - hw_dbg("PCI-E Master disable polling has failed.\n"); + hw_dbg("PCI-E Master disable polling has failed\n"); hw_dbg("Masking off all interrupts\n"); wr32(IGC_IMC, 0xffffffff); @@ -177,7 +177,7 @@ static s32 igc_init_phy_params_base(struct igc_hw *hw) */ ret_val = hw->phy.ops.reset(hw); if (ret_val) { - hw_dbg("Error resetting the PHY.\n"); + hw_dbg("Error resetting the PHY\n"); goto out; } @@ -212,6 +212,9 @@ static s32 igc_get_invariants_base(struct igc_hw *hw) case IGC_DEV_ID_I225_I: case IGC_DEV_ID_I220_V: case IGC_DEV_ID_I225_K: + case IGC_DEV_ID_I225_K2: + case IGC_DEV_ID_I225_LMVP: + case IGC_DEV_ID_I225_IT: case IGC_DEV_ID_I225_BLANK_NVM: mac->type = igc_i225; break; @@ -364,7 +367,7 @@ void igc_rx_fifo_flush_base(struct igc_hw *hw) } if (ms_wait == 10) - pr_debug("Queue disable timed out after 10ms\n"); + hw_dbg("Queue disable timed out after 10ms\n"); /* Clear RLPML, RCTL.SBP, RFCTL.LEF, and set RCTL.LPE so that all * incoming packets are rejected. Set enable and wait 2ms so that diff --git a/drivers/net/ethernet/intel/igc/igc_defines.h b/drivers/net/ethernet/intel/igc/igc_defines.h index 4ddccccf42cc..186deb1d9375 100644 --- a/drivers/net/ethernet/intel/igc/igc_defines.h +++ b/drivers/net/ethernet/intel/igc/igc_defines.h @@ -44,13 +44,9 @@ /* Wake Up Packet Memory stores the first 128 bytes of the wake up packet */ #define IGC_WUPM_BYTES 128 -/* Physical Func Reset Done Indication */ -#define IGC_CTRL_EXT_LINK_MODE_MASK 0x00C00000 - /* Loop limit on how long we wait for auto-negotiation to complete */ #define COPPER_LINK_UP_LIMIT 10 #define PHY_AUTO_NEG_LIMIT 45 -#define PHY_FORCE_LIMIT 20 /* Number of 100 microseconds we wait for PCI Express master disable */ #define MASTER_DISABLE_TIMEOUT 800 @@ -66,8 +62,14 @@ * (RAR[15]) for our directed address used by controllers with * manageability enabled, allowing us room for 15 multicast addresses. */ +#define IGC_RAH_RAH_MASK 0x0000FFFF +#define IGC_RAH_ASEL_MASK 0x00030000 +#define IGC_RAH_ASEL_SRC_ADDR BIT(16) +#define IGC_RAH_QSEL_MASK 0x000C0000 +#define IGC_RAH_QSEL_SHIFT 18 +#define IGC_RAH_QSEL_ENABLE BIT(28) #define IGC_RAH_AV 0x80000000 /* Receive descriptor valid */ -#define IGC_RAH_POOL_1 0x00040000 + #define IGC_RAL_MAC_ADDR_LEN 4 #define IGC_RAH_MAC_ADDR_LEN 2 @@ -94,8 +96,6 @@ #define IGC_CTRL_RFCE 0x08000000 /* Receive Flow Control enable */ #define IGC_CTRL_TFCE 0x10000000 /* Transmit flow control enable */ -#define IGC_CONNSW_AUTOSENSE_EN 0x1 - /* As per the EAS the maximum supported size is 9.5KB (9728 bytes) */ #define MAX_JUMBO_FRAME_SIZE 0x2600 @@ -166,11 +166,6 @@ /* For checksumming, the sum of all words in the NVM should equal 0xBABA. */ #define NVM_SUM 0xBABA - -#define NVM_PBA_OFFSET_0 8 -#define NVM_PBA_OFFSET_1 9 -#define NVM_RESERVED_WORD 0xFFFF -#define NVM_PBA_PTR_GUARD 0xFAFA #define NVM_WORD_SIZE_BASE_SHIFT 6 /* Collision related configuration parameters */ @@ -252,7 +247,6 @@ /* Interrupt Cause Set */ #define IGC_ICS_LSC IGC_ICR_LSC /* Link Status Change */ #define IGC_ICS_RXDMT0 IGC_ICR_RXDMT0 /* rx desc min. threshold */ -#define IGC_ICS_DRSTA IGC_ICR_DRSTA /* Device Reset Aserted */ #define IGC_ICR_DOUTSYNC 0x10000000 /* NIC DMA out of sync */ #define IGC_EITR_CNT_IGNR 0x80000000 /* Don't reset counters on write */ @@ -271,21 +265,13 @@ #define IGC_TXD_POPTS_IXSM 0x01 /* Insert IP checksum */ #define IGC_TXD_POPTS_TXSM 0x02 /* Insert TCP/UDP checksum */ #define IGC_TXD_CMD_EOP 0x01000000 /* End of Packet */ -#define IGC_TXD_CMD_IFCS 0x02000000 /* Insert FCS (Ethernet CRC) */ #define IGC_TXD_CMD_IC 0x04000000 /* Insert Checksum */ -#define IGC_TXD_CMD_RS 0x08000000 /* Report Status */ -#define IGC_TXD_CMD_RPS 0x10000000 /* Report Packet Sent */ #define IGC_TXD_CMD_DEXT 0x20000000 /* Desc extension (0 = legacy) */ #define IGC_TXD_CMD_VLE 0x40000000 /* Add VLAN tag */ -#define IGC_TXD_CMD_IDE 0x80000000 /* Enable Tidv register */ #define IGC_TXD_STAT_DD 0x00000001 /* Descriptor Done */ -#define IGC_TXD_STAT_EC 0x00000002 /* Excess Collisions */ -#define IGC_TXD_STAT_LC 0x00000004 /* Late Collisions */ -#define IGC_TXD_STAT_TU 0x00000008 /* Transmit underrun */ #define IGC_TXD_CMD_TCP 0x01000000 /* TCP packet */ #define IGC_TXD_CMD_IP 0x02000000 /* IP packet */ #define IGC_TXD_CMD_TSE 0x04000000 /* TCP Seg enable */ -#define IGC_TXD_STAT_TC 0x00000004 /* Tx Underrun */ #define IGC_TXD_EXTCMD_TSTAMP 0x00000010 /* IEEE1588 Timestamp packet */ /* IPSec Encrypt Enable */ @@ -377,6 +363,11 @@ #define I225_TXPBSIZE_DEFAULT 0x04000014 /* TXPBSIZE default */ #define IGC_RXPBS_CFG_TS_EN 0x80000000 /* Timestamp in Rx buffer */ +#define IGC_TXPBSIZE_TSN 0x04145145 /* 5k bytes buffer for each queue */ + +#define IGC_DTXMXPKTSZ_TSN 0x19 /* 1600 bytes of max TX DMA packet size */ +#define IGC_DTXMXPKTSZ_DEFAULT 0x98 /* 9728-byte Jumbo frames */ + /* Time Sync Interrupt Causes */ #define IGC_TSICR_SYS_WRAP BIT(0) /* SYSTIM Wrap around. */ #define IGC_TSICR_TXTS BIT(1) /* Transmit Timestamp. */ @@ -387,9 +378,6 @@ #define IGC_TSICR_INTERRUPTS IGC_TSICR_TXTS -/* PTP Queue Filter */ -#define IGC_ETQF_1588 BIT(30) - #define IGC_FTQF_VF_BP 0x00008000 #define IGC_FTQF_1588_TIME_STAMP 0x08000000 #define IGC_FTQF_MASK 0xF0000000 @@ -431,6 +419,14 @@ #define IGC_TSYNCTXCTL_START_SYNC 0x80000000 /* initiate sync */ #define IGC_TSYNCTXCTL_TXSYNSIG 0x00000020 /* Sample TX tstamp in PHY sop */ +/* Transmit Scheduling */ +#define IGC_TQAVCTRL_TRANSMIT_MODE_TSN 0x00000001 +#define IGC_TQAVCTRL_ENHANCED_QAV 0x00000008 + +#define IGC_TXQCTL_QUEUE_MODE_LAUNCHT 0x00000001 +#define IGC_TXQCTL_STRICT_CYCLE 0x00000002 +#define IGC_TXQCTL_STRICT_END 0x00000004 + /* Receive Checksum Control */ #define IGC_RXCSUM_CRCOFL 0x00000800 /* CRC32 offload enable */ #define IGC_RXCSUM_PCSD 0x00002000 /* packet checksum disabled */ @@ -497,16 +493,15 @@ #define IGC_MDIC_READY 0x10000000 #define IGC_MDIC_INT_EN 0x20000000 #define IGC_MDIC_ERROR 0x40000000 -#define IGC_MDIC_DEST 0x80000000 #define IGC_N0_QUEUE -1 #define IGC_MAX_MAC_HDR_LEN 127 #define IGC_MAX_NETWORK_HDR_LEN 511 -#define IGC_VLAPQF_QUEUE_SEL(_n, q_idx) ((q_idx) << ((_n) * 4)) -#define IGC_VLAPQF_P_VALID(_n) (0x1 << (3 + (_n) * 4)) -#define IGC_VLAPQF_QUEUE_MASK 0x03 +#define IGC_VLANPQF_QSEL(_n, q_idx) ((q_idx) << ((_n) * 4)) +#define IGC_VLANPQF_VALID(_n) (0x1 << (3 + (_n) * 4)) +#define IGC_VLANPQF_QUEUE_MASK 0x03 #define IGC_ADVTXD_MACLEN_SHIFT 9 /* Adv ctxt desc mac len shift */ #define IGC_ADVTXD_TUCMD_IPV4 0x00000400 /* IP Packet Type:1=IPv4 */ diff --git a/drivers/net/ethernet/intel/igc/igc_diag.c b/drivers/net/ethernet/intel/igc/igc_diag.c new file mode 100644 index 000000000000..cc621970c0cd --- /dev/null +++ b/drivers/net/ethernet/intel/igc/igc_diag.c @@ -0,0 +1,186 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2020 Intel Corporation */ + +#include "igc.h" +#include "igc_diag.h" + +static struct igc_reg_test reg_test[] = { + { IGC_FCAL, 1, PATTERN_TEST, 0xFFFFFFFF, 0xFFFFFFFF }, + { IGC_FCAH, 1, PATTERN_TEST, 0x0000FFFF, 0xFFFFFFFF }, + { IGC_FCT, 1, PATTERN_TEST, 0x0000FFFF, 0xFFFFFFFF }, + { IGC_RDBAH(0), 4, PATTERN_TEST, 0xFFFFFFFF, 0xFFFFFFFF }, + { IGC_RDBAL(0), 4, PATTERN_TEST, 0xFFFFFF80, 0xFFFFFF80 }, + { IGC_RDLEN(0), 4, PATTERN_TEST, 0x000FFF80, 0x000FFFFF }, + { IGC_RDT(0), 4, PATTERN_TEST, 0x0000FFFF, 0x0000FFFF }, + { IGC_FCRTH, 1, PATTERN_TEST, 0x0003FFF0, 0x0003FFF0 }, + { IGC_FCTTV, 1, PATTERN_TEST, 0x0000FFFF, 0x0000FFFF }, + { IGC_TIPG, 1, PATTERN_TEST, 0x3FFFFFFF, 0x3FFFFFFF }, + { IGC_TDBAH(0), 4, PATTERN_TEST, 0xFFFFFFFF, 0xFFFFFFFF }, + { IGC_TDBAL(0), 4, PATTERN_TEST, 0xFFFFFF80, 0xFFFFFF80 }, + { IGC_TDLEN(0), 4, PATTERN_TEST, 0x000FFF80, 0x000FFFFF }, + { IGC_TDT(0), 4, PATTERN_TEST, 0x0000FFFF, 0x0000FFFF }, + { IGC_RCTL, 1, SET_READ_TEST, 0xFFFFFFFF, 0x00000000 }, + { IGC_RCTL, 1, SET_READ_TEST, 0x04CFB2FE, 0x003FFFFB }, + { IGC_RCTL, 1, SET_READ_TEST, 0x04CFB2FE, 0xFFFFFFFF }, + { IGC_TCTL, 1, SET_READ_TEST, 0xFFFFFFFF, 0x00000000 }, + { IGC_RA, 16, TABLE64_TEST_LO, + 0xFFFFFFFF, 0xFFFFFFFF }, + { IGC_RA, 16, TABLE64_TEST_HI, + 0x900FFFFF, 0xFFFFFFFF }, + { IGC_MTA, 128, TABLE32_TEST, + 0xFFFFFFFF, 0xFFFFFFFF }, + { 0, 0, 0, 0} +}; + +static bool reg_pattern_test(struct igc_adapter *adapter, u64 *data, int reg, + u32 mask, u32 write) +{ + struct igc_hw *hw = &adapter->hw; + u32 pat, val, before; + static const u32 test_pattern[] = { + 0x5A5A5A5A, 0xA5A5A5A5, 0x00000000, 0xFFFFFFFF + }; + + for (pat = 0; pat < ARRAY_SIZE(test_pattern); pat++) { + before = rd32(reg); + wr32(reg, test_pattern[pat] & write); + val = rd32(reg); + if (val != (test_pattern[pat] & write & mask)) { + netdev_err(adapter->netdev, + "pattern test reg %04X failed: got 0x%08X expected 0x%08X", + reg, val, test_pattern[pat] & write & mask); + *data = reg; + wr32(reg, before); + return false; + } + wr32(reg, before); + } + return true; +} + +static bool reg_set_and_check(struct igc_adapter *adapter, u64 *data, int reg, + u32 mask, u32 write) +{ + struct igc_hw *hw = &adapter->hw; + u32 val, before; + + before = rd32(reg); + wr32(reg, write & mask); + val = rd32(reg); + if ((write & mask) != (val & mask)) { + netdev_err(adapter->netdev, + "set/check reg %04X test failed: got 0x%08X expected 0x%08X", + reg, (val & mask), (write & mask)); + *data = reg; + wr32(reg, before); + return false; + } + wr32(reg, before); + return true; +} + +bool igc_reg_test(struct igc_adapter *adapter, u64 *data) +{ + struct igc_reg_test *test = reg_test; + struct igc_hw *hw = &adapter->hw; + u32 value, before, after; + u32 i, toggle, b = false; + + /* Because the status register is such a special case, + * we handle it separately from the rest of the register + * tests. Some bits are read-only, some toggle, and some + * are writeable. + */ + toggle = 0x6800D3; + before = rd32(IGC_STATUS); + value = before & toggle; + wr32(IGC_STATUS, toggle); + after = rd32(IGC_STATUS) & toggle; + if (value != after) { + netdev_err(adapter->netdev, + "failed STATUS register test got: 0x%08X expected: 0x%08X", + after, value); + *data = 1; + return false; + } + /* restore previous status */ + wr32(IGC_STATUS, before); + + /* Perform the remainder of the register test, looping through + * the test table until we either fail or reach the null entry. + */ + while (test->reg) { + for (i = 0; i < test->array_len; i++) { + switch (test->test_type) { + case PATTERN_TEST: + b = reg_pattern_test(adapter, data, + test->reg + (i * 0x40), + test->mask, + test->write); + break; + case SET_READ_TEST: + b = reg_set_and_check(adapter, data, + test->reg + (i * 0x40), + test->mask, + test->write); + break; + case TABLE64_TEST_LO: + b = reg_pattern_test(adapter, data, + test->reg + (i * 8), + test->mask, + test->write); + break; + case TABLE64_TEST_HI: + b = reg_pattern_test(adapter, data, + test->reg + 4 + (i * 8), + test->mask, + test->write); + break; + case TABLE32_TEST: + b = reg_pattern_test(adapter, data, + test->reg + (i * 4), + test->mask, + test->write); + break; + } + if (!b) + return false; + } + test++; + } + *data = 0; + return true; +} + +bool igc_eeprom_test(struct igc_adapter *adapter, u64 *data) +{ + struct igc_hw *hw = &adapter->hw; + + *data = 0; + + if (hw->nvm.ops.validate(hw) != IGC_SUCCESS) { + *data = 1; + return false; + } + + return true; +} + +bool igc_link_test(struct igc_adapter *adapter, u64 *data) +{ + bool link_up; + + *data = 0; + + /* add delay to give enough time for autonegotioation to finish */ + if (adapter->hw.mac.autoneg) + ssleep(5); + + link_up = igc_has_link(adapter); + if (!link_up) { + *data = 1; + return false; + } + + return true; +} diff --git a/drivers/net/ethernet/intel/igc/igc_diag.h b/drivers/net/ethernet/intel/igc/igc_diag.h new file mode 100644 index 000000000000..600658e33bec --- /dev/null +++ b/drivers/net/ethernet/intel/igc/igc_diag.h @@ -0,0 +1,30 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2020 Intel Corporation */ + +bool igc_reg_test(struct igc_adapter *adapter, u64 *data); +bool igc_eeprom_test(struct igc_adapter *adapter, u64 *data); +bool igc_link_test(struct igc_adapter *adapter, u64 *data); + +struct igc_reg_test { + u16 reg; + u8 array_len; + u8 test_type; + u32 mask; + u32 write; +}; + +/* In the hardware, registers are laid out either singly, in arrays + * spaced 0x40 bytes apart, or in contiguous tables. We assume + * most tests take place on arrays or single registers (handled + * as a single-element array) and special-case the tables. + * Table tests are always pattern tests. + * + * We also make provision for some required setup steps by specifying + * registers to be written without any read-back testing. + */ + +#define PATTERN_TEST 1 +#define SET_READ_TEST 2 +#define TABLE32_TEST 3 +#define TABLE64_TEST_LO 4 +#define TABLE64_TEST_HI 5 diff --git a/drivers/net/ethernet/intel/igc/igc_dump.c b/drivers/net/ethernet/intel/igc/igc_dump.c index 657ab50ae296..4b9ec7d0b727 100644 --- a/drivers/net/ethernet/intel/igc/igc_dump.c +++ b/drivers/net/ethernet/intel/igc/igc_dump.c @@ -35,10 +35,6 @@ static const struct igc_reg_info igc_reg_info_tbl[] = { {IGC_TDH(0), "TDH"}, {IGC_TDT(0), "TDT"}, {IGC_TXDCTL(0), "TXDCTL"}, - {IGC_TDFH, "TDFH"}, - {IGC_TDFT, "TDFT"}, - {IGC_TDFHS, "TDFHS"}, - {IGC_TDFPC, "TDFPC"}, /* List Terminator */ {} @@ -47,6 +43,7 @@ static const struct igc_reg_info igc_reg_info_tbl[] = { /* igc_regdump - register printout routine */ static void igc_regdump(struct igc_hw *hw, struct igc_reg_info *reginfo) { + struct net_device *dev = igc_get_hw_dev(hw); int n = 0; char rname[16]; u32 regs[8]; @@ -101,13 +98,14 @@ static void igc_regdump(struct igc_hw *hw, struct igc_reg_info *reginfo) regs[n] = rd32(IGC_TXDCTL(n)); break; default: - pr_info("%-15s %08x\n", reginfo->name, rd32(reginfo->ofs)); + netdev_info(dev, "%-15s %08x\n", reginfo->name, + rd32(reginfo->ofs)); return; } snprintf(rname, 16, "%s%s", reginfo->name, "[0-3]"); - pr_info("%-15s %08x %08x %08x %08x\n", rname, regs[0], regs[1], - regs[2], regs[3]); + netdev_info(dev, "%-15s %08x %08x %08x %08x\n", rname, regs[0], regs[1], + regs[2], regs[3]); } /* igc_rings_dump - Tx-rings and Rx-rings */ @@ -125,39 +123,34 @@ void igc_rings_dump(struct igc_adapter *adapter) if (!netif_msg_hw(adapter)) return; - /* Print netdevice Info */ - if (netdev) { - dev_info(&adapter->pdev->dev, "Net device Info\n"); - pr_info("Device Name state trans_start\n"); - pr_info("%-15s %016lX %016lX\n", netdev->name, - netdev->state, dev_trans_start(netdev)); - } + netdev_info(netdev, "Device info: state %016lX trans_start %016lX\n", + netdev->state, dev_trans_start(netdev)); /* Print TX Ring Summary */ - if (!netdev || !netif_running(netdev)) + if (!netif_running(netdev)) goto exit; - dev_info(&adapter->pdev->dev, "TX Rings Summary\n"); - pr_info("Queue [NTU] [NTC] [bi(ntc)->dma ] leng ntw timestamp\n"); + netdev_info(netdev, "TX Rings Summary\n"); + netdev_info(netdev, "Queue [NTU] [NTC] [bi(ntc)->dma ] leng ntw timestamp\n"); for (n = 0; n < adapter->num_tx_queues; n++) { struct igc_tx_buffer *buffer_info; tx_ring = adapter->tx_ring[n]; buffer_info = &tx_ring->tx_buffer_info[tx_ring->next_to_clean]; - pr_info(" %5d %5X %5X %016llX %04X %p %016llX\n", - n, tx_ring->next_to_use, tx_ring->next_to_clean, - (u64)dma_unmap_addr(buffer_info, dma), - dma_unmap_len(buffer_info, len), - buffer_info->next_to_watch, - (u64)buffer_info->time_stamp); + netdev_info(netdev, "%5d %5X %5X %016llX %04X %p %016llX\n", + n, tx_ring->next_to_use, tx_ring->next_to_clean, + (u64)dma_unmap_addr(buffer_info, dma), + dma_unmap_len(buffer_info, len), + buffer_info->next_to_watch, + (u64)buffer_info->time_stamp); } /* Print TX Rings */ if (!netif_msg_tx_done(adapter)) goto rx_ring_summary; - dev_info(&adapter->pdev->dev, "TX Rings Dump\n"); + netdev_info(netdev, "TX Rings Dump\n"); /* Transmit Descriptor Formats * @@ -172,10 +165,11 @@ void igc_rings_dump(struct igc_adapter *adapter) for (n = 0; n < adapter->num_tx_queues; n++) { tx_ring = adapter->tx_ring[n]; - pr_info("------------------------------------\n"); - pr_info("TX QUEUE INDEX = %d\n", tx_ring->queue_index); - pr_info("------------------------------------\n"); - pr_info("T [desc] [address 63:0 ] [PlPOCIStDDM Ln] [bi->dma ] leng ntw timestamp bi->skb\n"); + netdev_info(netdev, "------------------------------------\n"); + netdev_info(netdev, "TX QUEUE INDEX = %d\n", + tx_ring->queue_index); + netdev_info(netdev, "------------------------------------\n"); + netdev_info(netdev, "T [desc] [address 63:0 ] [PlPOCIStDDM Ln] [bi->dma ] leng ntw timestamp bi->skb\n"); for (i = 0; tx_ring->desc && (i < tx_ring->count); i++) { const char *next_desc; @@ -194,14 +188,14 @@ void igc_rings_dump(struct igc_adapter *adapter) else next_desc = ""; - pr_info("T [0x%03X] %016llX %016llX %016llX %04X %p %016llX %p%s\n", - i, le64_to_cpu(u0->a), - le64_to_cpu(u0->b), - (u64)dma_unmap_addr(buffer_info, dma), - dma_unmap_len(buffer_info, len), - buffer_info->next_to_watch, - (u64)buffer_info->time_stamp, - buffer_info->skb, next_desc); + netdev_info(netdev, "T [0x%03X] %016llX %016llX %016llX %04X %p %016llX %p%s\n", + i, le64_to_cpu(u0->a), + le64_to_cpu(u0->b), + (u64)dma_unmap_addr(buffer_info, dma), + dma_unmap_len(buffer_info, len), + buffer_info->next_to_watch, + (u64)buffer_info->time_stamp, + buffer_info->skb, next_desc); if (netif_msg_pktdata(adapter) && buffer_info->skb) print_hex_dump(KERN_INFO, "", @@ -214,19 +208,19 @@ void igc_rings_dump(struct igc_adapter *adapter) /* Print RX Rings Summary */ rx_ring_summary: - dev_info(&adapter->pdev->dev, "RX Rings Summary\n"); - pr_info("Queue [NTU] [NTC]\n"); + netdev_info(netdev, "RX Rings Summary\n"); + netdev_info(netdev, "Queue [NTU] [NTC]\n"); for (n = 0; n < adapter->num_rx_queues; n++) { rx_ring = adapter->rx_ring[n]; - pr_info(" %5d %5X %5X\n", - n, rx_ring->next_to_use, rx_ring->next_to_clean); + netdev_info(netdev, "%5d %5X %5X\n", n, rx_ring->next_to_use, + rx_ring->next_to_clean); } /* Print RX Rings */ if (!netif_msg_rx_status(adapter)) goto exit; - dev_info(&adapter->pdev->dev, "RX Rings Dump\n"); + netdev_info(netdev, "RX Rings Dump\n"); /* Advanced Receive Descriptor (Read) Format * 63 1 0 @@ -251,11 +245,12 @@ rx_ring_summary: for (n = 0; n < adapter->num_rx_queues; n++) { rx_ring = adapter->rx_ring[n]; - pr_info("------------------------------------\n"); - pr_info("RX QUEUE INDEX = %d\n", rx_ring->queue_index); - pr_info("------------------------------------\n"); - pr_info("R [desc] [ PktBuf A0] [ HeadBuf DD] [bi->dma ] [bi->skb] <-- Adv Rx Read format\n"); - pr_info("RWB[desc] [PcsmIpSHl PtRs] [vl er S cks ln] ---------------- [bi->skb] <-- Adv Rx Write-Back format\n"); + netdev_info(netdev, "------------------------------------\n"); + netdev_info(netdev, "RX QUEUE INDEX = %d\n", + rx_ring->queue_index); + netdev_info(netdev, "------------------------------------\n"); + netdev_info(netdev, "R [desc] [ PktBuf A0] [ HeadBuf DD] [bi->dma ] [bi->skb] <-- Adv Rx Read format\n"); + netdev_info(netdev, "RWB[desc] [PcsmIpSHl PtRs] [vl er S cks ln] ---------------- [bi->skb] <-- Adv Rx Write-Back format\n"); for (i = 0; i < rx_ring->count; i++) { const char *next_desc; @@ -275,18 +270,18 @@ rx_ring_summary: if (staterr & IGC_RXD_STAT_DD) { /* Descriptor Done */ - pr_info("%s[0x%03X] %016llX %016llX ---------------- %s\n", - "RWB", i, - le64_to_cpu(u0->a), - le64_to_cpu(u0->b), - next_desc); + netdev_info(netdev, "%s[0x%03X] %016llX %016llX ---------------- %s\n", + "RWB", i, + le64_to_cpu(u0->a), + le64_to_cpu(u0->b), + next_desc); } else { - pr_info("%s[0x%03X] %016llX %016llX %016llX %s\n", - "R ", i, - le64_to_cpu(u0->a), - le64_to_cpu(u0->b), - (u64)buffer_info->dma, - next_desc); + netdev_info(netdev, "%s[0x%03X] %016llX %016llX %016llX %s\n", + "R ", i, + le64_to_cpu(u0->a), + le64_to_cpu(u0->b), + (u64)buffer_info->dma, + next_desc); if (netif_msg_pktdata(adapter) && buffer_info->dma && buffer_info->page) { @@ -314,8 +309,8 @@ void igc_regs_dump(struct igc_adapter *adapter) struct igc_reg_info *reginfo; /* Print Registers */ - dev_info(&adapter->pdev->dev, "Register Dump\n"); - pr_info(" Register Name Value\n"); + netdev_info(adapter->netdev, "Register Dump\n"); + netdev_info(adapter->netdev, "Register Name Value\n"); for (reginfo = (struct igc_reg_info *)igc_reg_info_tbl; reginfo->name; reginfo++) { igc_regdump(hw, reginfo); diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c b/drivers/net/ethernet/intel/igc/igc_ethtool.c index f530fc29b074..a938ec8db681 100644 --- a/drivers/net/ethernet/intel/igc/igc_ethtool.c +++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c @@ -6,6 +6,7 @@ #include <linux/pm_runtime.h> #include "igc.h" +#include "igc_diag.h" /* forward declaration */ struct igc_stats { @@ -123,8 +124,8 @@ static const char igc_priv_flags_strings[][ETH_GSTRING_LEN] = { #define IGC_PRIV_FLAGS_STR_LEN ARRAY_SIZE(igc_priv_flags_strings) -static void igc_get_drvinfo(struct net_device *netdev, - struct ethtool_drvinfo *drvinfo) +static void igc_ethtool_get_drvinfo(struct net_device *netdev, + struct ethtool_drvinfo *drvinfo) { struct igc_adapter *adapter = netdev_priv(netdev); @@ -138,13 +139,13 @@ static void igc_get_drvinfo(struct net_device *netdev, drvinfo->n_priv_flags = IGC_PRIV_FLAGS_STR_LEN; } -static int igc_get_regs_len(struct net_device *netdev) +static int igc_ethtool_get_regs_len(struct net_device *netdev) { return IGC_REGS_LEN * sizeof(u32); } -static void igc_get_regs(struct net_device *netdev, - struct ethtool_regs *regs, void *p) +static void igc_ethtool_get_regs(struct net_device *netdev, + struct ethtool_regs *regs, void *p) { struct igc_adapter *adapter = netdev_priv(netdev); struct igc_hw *hw = &adapter->hw; @@ -153,7 +154,7 @@ static void igc_get_regs(struct net_device *netdev, memset(p, 0, IGC_REGS_LEN * sizeof(u32)); - regs->version = (1u << 24) | (hw->revision_id << 16) | hw->device_id; + regs->version = (2u << 24) | (hw->revision_id << 16) | hw->device_id; /* General Registers */ regs_buff[0] = rd32(IGC_CTRL); @@ -306,9 +307,24 @@ static void igc_get_regs(struct net_device *netdev, regs_buff[164 + i] = rd32(IGC_TDT(i)); for (i = 0; i < 4; i++) regs_buff[168 + i] = rd32(IGC_TXDCTL(i)); + + /* XXX: Due to a bug few lines above, RAL and RAH registers are + * overwritten. To preserve the ABI, we write these registers again in + * regs_buff. + */ + for (i = 0; i < 16; i++) + regs_buff[172 + i] = rd32(IGC_RAL(i)); + for (i = 0; i < 16; i++) + regs_buff[188 + i] = rd32(IGC_RAH(i)); + + regs_buff[204] = rd32(IGC_VLANPQF); + + for (i = 0; i < 8; i++) + regs_buff[205 + i] = rd32(IGC_ETQF(i)); } -static void igc_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) +static void igc_ethtool_get_wol(struct net_device *netdev, + struct ethtool_wolinfo *wol) { struct igc_adapter *adapter = netdev_priv(netdev); @@ -339,7 +355,8 @@ static void igc_get_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) wol->wolopts |= WAKE_PHY; } -static int igc_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) +static int igc_ethtool_set_wol(struct net_device *netdev, + struct ethtool_wolinfo *wol) { struct igc_adapter *adapter = netdev_priv(netdev); @@ -367,21 +384,21 @@ static int igc_set_wol(struct net_device *netdev, struct ethtool_wolinfo *wol) return 0; } -static u32 igc_get_msglevel(struct net_device *netdev) +static u32 igc_ethtool_get_msglevel(struct net_device *netdev) { struct igc_adapter *adapter = netdev_priv(netdev); return adapter->msg_enable; } -static void igc_set_msglevel(struct net_device *netdev, u32 data) +static void igc_ethtool_set_msglevel(struct net_device *netdev, u32 data) { struct igc_adapter *adapter = netdev_priv(netdev); adapter->msg_enable = data; } -static int igc_nway_reset(struct net_device *netdev) +static int igc_ethtool_nway_reset(struct net_device *netdev) { struct igc_adapter *adapter = netdev_priv(netdev); @@ -390,7 +407,7 @@ static int igc_nway_reset(struct net_device *netdev) return 0; } -static u32 igc_get_link(struct net_device *netdev) +static u32 igc_ethtool_get_link(struct net_device *netdev) { struct igc_adapter *adapter = netdev_priv(netdev); struct igc_mac_info *mac = &adapter->hw.mac; @@ -407,15 +424,15 @@ static u32 igc_get_link(struct net_device *netdev) return igc_has_link(adapter); } -static int igc_get_eeprom_len(struct net_device *netdev) +static int igc_ethtool_get_eeprom_len(struct net_device *netdev) { struct igc_adapter *adapter = netdev_priv(netdev); return adapter->hw.nvm.word_size * 2; } -static int igc_get_eeprom(struct net_device *netdev, - struct ethtool_eeprom *eeprom, u8 *bytes) +static int igc_ethtool_get_eeprom(struct net_device *netdev, + struct ethtool_eeprom *eeprom, u8 *bytes) { struct igc_adapter *adapter = netdev_priv(netdev); struct igc_hw *hw = &adapter->hw; @@ -461,8 +478,8 @@ static int igc_get_eeprom(struct net_device *netdev, return ret_val; } -static int igc_set_eeprom(struct net_device *netdev, - struct ethtool_eeprom *eeprom, u8 *bytes) +static int igc_ethtool_set_eeprom(struct net_device *netdev, + struct ethtool_eeprom *eeprom, u8 *bytes) { struct igc_adapter *adapter = netdev_priv(netdev); struct igc_hw *hw = &adapter->hw; @@ -529,8 +546,8 @@ static int igc_set_eeprom(struct net_device *netdev, return ret_val; } -static void igc_get_ringparam(struct net_device *netdev, - struct ethtool_ringparam *ring) +static void igc_ethtool_get_ringparam(struct net_device *netdev, + struct ethtool_ringparam *ring) { struct igc_adapter *adapter = netdev_priv(netdev); @@ -540,8 +557,8 @@ static void igc_get_ringparam(struct net_device *netdev, ring->tx_pending = adapter->tx_ring_count; } -static int igc_set_ringparam(struct net_device *netdev, - struct ethtool_ringparam *ring) +static int igc_ethtool_set_ringparam(struct net_device *netdev, + struct ethtool_ringparam *ring) { struct igc_adapter *adapter = netdev_priv(netdev); struct igc_ring *temp_ring; @@ -655,8 +672,8 @@ clear_reset: return err; } -static void igc_get_pauseparam(struct net_device *netdev, - struct ethtool_pauseparam *pause) +static void igc_ethtool_get_pauseparam(struct net_device *netdev, + struct ethtool_pauseparam *pause) { struct igc_adapter *adapter = netdev_priv(netdev); struct igc_hw *hw = &adapter->hw; @@ -674,8 +691,8 @@ static void igc_get_pauseparam(struct net_device *netdev, } } -static int igc_set_pauseparam(struct net_device *netdev, - struct ethtool_pauseparam *pause) +static int igc_ethtool_set_pauseparam(struct net_device *netdev, + struct ethtool_pauseparam *pause) { struct igc_adapter *adapter = netdev_priv(netdev); struct igc_hw *hw = &adapter->hw; @@ -714,7 +731,8 @@ static int igc_set_pauseparam(struct net_device *netdev, return retval; } -static void igc_get_strings(struct net_device *netdev, u32 stringset, u8 *data) +static void igc_ethtool_get_strings(struct net_device *netdev, u32 stringset, + u8 *data) { struct igc_adapter *adapter = netdev_priv(netdev); u8 *p = data; @@ -765,7 +783,7 @@ static void igc_get_strings(struct net_device *netdev, u32 stringset, u8 *data) } } -static int igc_get_sset_count(struct net_device *netdev, int sset) +static int igc_ethtool_get_sset_count(struct net_device *netdev, int sset) { switch (sset) { case ETH_SS_STATS: @@ -779,7 +797,7 @@ static int igc_get_sset_count(struct net_device *netdev, int sset) } } -static void igc_get_ethtool_stats(struct net_device *netdev, +static void igc_ethtool_get_stats(struct net_device *netdev, struct ethtool_stats *stats, u64 *data) { struct igc_adapter *adapter = netdev_priv(netdev); @@ -835,8 +853,8 @@ static void igc_get_ethtool_stats(struct net_device *netdev, spin_unlock(&adapter->stats64_lock); } -static int igc_get_coalesce(struct net_device *netdev, - struct ethtool_coalesce *ec) +static int igc_ethtool_get_coalesce(struct net_device *netdev, + struct ethtool_coalesce *ec) { struct igc_adapter *adapter = netdev_priv(netdev); @@ -855,8 +873,8 @@ static int igc_get_coalesce(struct net_device *netdev, return 0; } -static int igc_set_coalesce(struct net_device *netdev, - struct ethtool_coalesce *ec) +static int igc_ethtool_set_coalesce(struct net_device *netdev, + struct ethtool_coalesce *ec) { struct igc_adapter *adapter = netdev_priv(netdev); int i; @@ -913,81 +931,83 @@ static int igc_set_coalesce(struct net_device *netdev, } #define ETHER_TYPE_FULL_MASK ((__force __be16)~0) -static int igc_get_ethtool_nfc_entry(struct igc_adapter *adapter, - struct ethtool_rxnfc *cmd) +static int igc_ethtool_get_nfc_rule(struct igc_adapter *adapter, + struct ethtool_rxnfc *cmd) { struct ethtool_rx_flow_spec *fsp = &cmd->fs; - struct igc_nfc_filter *rule = NULL; + struct igc_nfc_rule *rule = NULL; - /* report total rule count */ - cmd->data = IGC_MAX_RXNFC_FILTERS; + cmd->data = IGC_MAX_RXNFC_RULES; - hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node) { - if (fsp->location <= rule->sw_idx) - break; + mutex_lock(&adapter->nfc_rule_lock); + + rule = igc_get_nfc_rule(adapter, fsp->location); + if (!rule) + goto out; + + fsp->flow_type = ETHER_FLOW; + fsp->ring_cookie = rule->action; + + if (rule->filter.match_flags & IGC_FILTER_FLAG_ETHER_TYPE) { + fsp->h_u.ether_spec.h_proto = htons(rule->filter.etype); + fsp->m_u.ether_spec.h_proto = ETHER_TYPE_FULL_MASK; } - if (!rule || fsp->location != rule->sw_idx) - return -EINVAL; + if (rule->filter.match_flags & IGC_FILTER_FLAG_VLAN_TCI) { + fsp->flow_type |= FLOW_EXT; + fsp->h_ext.vlan_tci = htons(rule->filter.vlan_tci); + fsp->m_ext.vlan_tci = htons(VLAN_PRIO_MASK); + } - if (rule->filter.match_flags) { - fsp->flow_type = ETHER_FLOW; - fsp->ring_cookie = rule->action; - if (rule->filter.match_flags & IGC_FILTER_FLAG_ETHER_TYPE) { - fsp->h_u.ether_spec.h_proto = rule->filter.etype; - fsp->m_u.ether_spec.h_proto = ETHER_TYPE_FULL_MASK; - } - if (rule->filter.match_flags & IGC_FILTER_FLAG_VLAN_TCI) { - fsp->flow_type |= FLOW_EXT; - fsp->h_ext.vlan_tci = rule->filter.vlan_tci; - fsp->m_ext.vlan_tci = htons(VLAN_PRIO_MASK); - } - if (rule->filter.match_flags & IGC_FILTER_FLAG_DST_MAC_ADDR) { - ether_addr_copy(fsp->h_u.ether_spec.h_dest, - rule->filter.dst_addr); - /* As we only support matching by the full - * mask, return the mask to userspace - */ - eth_broadcast_addr(fsp->m_u.ether_spec.h_dest); - } - if (rule->filter.match_flags & IGC_FILTER_FLAG_SRC_MAC_ADDR) { - ether_addr_copy(fsp->h_u.ether_spec.h_source, - rule->filter.src_addr); - /* As we only support matching by the full - * mask, return the mask to userspace - */ - eth_broadcast_addr(fsp->m_u.ether_spec.h_source); - } + if (rule->filter.match_flags & IGC_FILTER_FLAG_DST_MAC_ADDR) { + ether_addr_copy(fsp->h_u.ether_spec.h_dest, + rule->filter.dst_addr); + eth_broadcast_addr(fsp->m_u.ether_spec.h_dest); + } - return 0; + if (rule->filter.match_flags & IGC_FILTER_FLAG_SRC_MAC_ADDR) { + ether_addr_copy(fsp->h_u.ether_spec.h_source, + rule->filter.src_addr); + eth_broadcast_addr(fsp->m_u.ether_spec.h_source); } + + mutex_unlock(&adapter->nfc_rule_lock); + return 0; + +out: + mutex_unlock(&adapter->nfc_rule_lock); return -EINVAL; } -static int igc_get_ethtool_nfc_all(struct igc_adapter *adapter, - struct ethtool_rxnfc *cmd, - u32 *rule_locs) +static int igc_ethtool_get_nfc_rules(struct igc_adapter *adapter, + struct ethtool_rxnfc *cmd, + u32 *rule_locs) { - struct igc_nfc_filter *rule; + struct igc_nfc_rule *rule; int cnt = 0; - /* report total rule count */ - cmd->data = IGC_MAX_RXNFC_FILTERS; + cmd->data = IGC_MAX_RXNFC_RULES; + + mutex_lock(&adapter->nfc_rule_lock); - hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node) { - if (cnt == cmd->rule_cnt) + list_for_each_entry(rule, &adapter->nfc_rule_list, list) { + if (cnt == cmd->rule_cnt) { + mutex_unlock(&adapter->nfc_rule_lock); return -EMSGSIZE; - rule_locs[cnt] = rule->sw_idx; + } + rule_locs[cnt] = rule->location; cnt++; } + mutex_unlock(&adapter->nfc_rule_lock); + cmd->rule_cnt = cnt; return 0; } -static int igc_get_rss_hash_opts(struct igc_adapter *adapter, - struct ethtool_rxnfc *cmd) +static int igc_ethtool_get_rss_hash_opts(struct igc_adapter *adapter, + struct ethtool_rxnfc *cmd) { cmd->data = 0; @@ -1036,41 +1056,33 @@ static int igc_get_rss_hash_opts(struct igc_adapter *adapter, return 0; } -static int igc_get_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd, - u32 *rule_locs) +static int igc_ethtool_get_rxnfc(struct net_device *dev, + struct ethtool_rxnfc *cmd, u32 *rule_locs) { struct igc_adapter *adapter = netdev_priv(dev); - int ret = -EOPNOTSUPP; switch (cmd->cmd) { case ETHTOOL_GRXRINGS: cmd->data = adapter->num_rx_queues; - ret = 0; - break; + return 0; case ETHTOOL_GRXCLSRLCNT: - cmd->rule_cnt = adapter->nfc_filter_count; - ret = 0; - break; + cmd->rule_cnt = adapter->nfc_rule_count; + return 0; case ETHTOOL_GRXCLSRULE: - ret = igc_get_ethtool_nfc_entry(adapter, cmd); - break; + return igc_ethtool_get_nfc_rule(adapter, cmd); case ETHTOOL_GRXCLSRLALL: - ret = igc_get_ethtool_nfc_all(adapter, cmd, rule_locs); - break; + return igc_ethtool_get_nfc_rules(adapter, cmd, rule_locs); case ETHTOOL_GRXFH: - ret = igc_get_rss_hash_opts(adapter, cmd); - break; + return igc_ethtool_get_rss_hash_opts(adapter, cmd); default: - break; + return -EOPNOTSUPP; } - - return ret; } #define UDP_RSS_FLAGS (IGC_FLAG_RSS_FIELD_IPV4_UDP | \ IGC_FLAG_RSS_FIELD_IPV6_UDP) -static int igc_set_rss_hash_opt(struct igc_adapter *adapter, - struct ethtool_rxnfc *nfc) +static int igc_ethtool_set_rss_hash_opt(struct igc_adapter *adapter, + struct ethtool_rxnfc *nfc) { u32 flags = adapter->flags; @@ -1145,8 +1157,8 @@ static int igc_set_rss_hash_opt(struct igc_adapter *adapter, if ((flags & UDP_RSS_FLAGS) && !(adapter->flags & UDP_RSS_FLAGS)) - dev_err(&adapter->pdev->dev, - "enabling UDP RSS: fragmented packets may arrive out of order to the stack above\n"); + netdev_err(adapter->netdev, + "Enabling UDP RSS: fragmented packets may arrive out of order to the stack above\n"); adapter->flags = flags; @@ -1171,344 +1183,184 @@ static int igc_set_rss_hash_opt(struct igc_adapter *adapter, return 0; } -static int igc_rxnfc_write_etype_filter(struct igc_adapter *adapter, - struct igc_nfc_filter *input) +static void igc_ethtool_init_nfc_rule(struct igc_nfc_rule *rule, + const struct ethtool_rx_flow_spec *fsp) { - struct igc_hw *hw = &adapter->hw; - u8 i; - u32 etqf; - u16 etype; - - /* find an empty etype filter register */ - for (i = 0; i < MAX_ETYPE_FILTER; ++i) { - if (!adapter->etype_bitmap[i]) - break; - } - if (i == MAX_ETYPE_FILTER) { - dev_err(&adapter->pdev->dev, "ethtool -N: etype filters are all used.\n"); - return -EINVAL; - } + INIT_LIST_HEAD(&rule->list); - adapter->etype_bitmap[i] = true; + rule->action = fsp->ring_cookie; + rule->location = fsp->location; - etqf = rd32(IGC_ETQF(i)); - etype = ntohs(input->filter.etype & ETHER_TYPE_FULL_MASK); - - etqf |= IGC_ETQF_FILTER_ENABLE; - etqf &= ~IGC_ETQF_ETYPE_MASK; - etqf |= (etype & IGC_ETQF_ETYPE_MASK); - - etqf &= ~IGC_ETQF_QUEUE_MASK; - etqf |= ((input->action << IGC_ETQF_QUEUE_SHIFT) - & IGC_ETQF_QUEUE_MASK); - etqf |= IGC_ETQF_QUEUE_ENABLE; - - wr32(IGC_ETQF(i), etqf); - - input->etype_reg_index = i; - - return 0; -} - -static int igc_rxnfc_write_vlan_prio_filter(struct igc_adapter *adapter, - struct igc_nfc_filter *input) -{ - struct igc_hw *hw = &adapter->hw; - u8 vlan_priority; - u16 queue_index; - u32 vlapqf; - - vlapqf = rd32(IGC_VLAPQF); - vlan_priority = (ntohs(input->filter.vlan_tci) & VLAN_PRIO_MASK) - >> VLAN_PRIO_SHIFT; - queue_index = (vlapqf >> (vlan_priority * 4)) & IGC_VLAPQF_QUEUE_MASK; - - /* check whether this vlan prio is already set */ - if (vlapqf & IGC_VLAPQF_P_VALID(vlan_priority) && - queue_index != input->action) { - dev_err(&adapter->pdev->dev, "ethtool rxnfc set vlan prio filter failed.\n"); - return -EEXIST; - } - - vlapqf |= IGC_VLAPQF_P_VALID(vlan_priority); - vlapqf |= IGC_VLAPQF_QUEUE_SEL(vlan_priority, input->action); - - wr32(IGC_VLAPQF, vlapqf); - - return 0; -} - -int igc_add_filter(struct igc_adapter *adapter, struct igc_nfc_filter *input) -{ - struct igc_hw *hw = &adapter->hw; - int err = -EINVAL; - - if (hw->mac.type == igc_i225 && - !(input->filter.match_flags & ~IGC_FILTER_FLAG_SRC_MAC_ADDR)) { - dev_err(&adapter->pdev->dev, - "i225 doesn't support flow classification rules specifying only source addresses.\n"); - return -EOPNOTSUPP; + if ((fsp->flow_type & FLOW_EXT) && fsp->m_ext.vlan_tci) { + rule->filter.vlan_tci = ntohs(fsp->h_ext.vlan_tci); + rule->filter.match_flags |= IGC_FILTER_FLAG_VLAN_TCI; } - if (input->filter.match_flags & IGC_FILTER_FLAG_ETHER_TYPE) { - err = igc_rxnfc_write_etype_filter(adapter, input); - if (err) - return err; + if (fsp->m_u.ether_spec.h_proto == ETHER_TYPE_FULL_MASK) { + rule->filter.etype = ntohs(fsp->h_u.ether_spec.h_proto); + rule->filter.match_flags = IGC_FILTER_FLAG_ETHER_TYPE; } - if (input->filter.match_flags & IGC_FILTER_FLAG_DST_MAC_ADDR) { - err = igc_add_mac_steering_filter(adapter, - input->filter.dst_addr, - input->action, 0); - err = min_t(int, err, 0); - if (err) - return err; + /* Both source and destination address filters only support the full + * mask. + */ + if (is_broadcast_ether_addr(fsp->m_u.ether_spec.h_source)) { + rule->filter.match_flags |= IGC_FILTER_FLAG_SRC_MAC_ADDR; + ether_addr_copy(rule->filter.src_addr, + fsp->h_u.ether_spec.h_source); } - if (input->filter.match_flags & IGC_FILTER_FLAG_SRC_MAC_ADDR) { - err = igc_add_mac_steering_filter(adapter, - input->filter.src_addr, - input->action, - IGC_MAC_STATE_SRC_ADDR); - err = min_t(int, err, 0); - if (err) - return err; + if (is_broadcast_ether_addr(fsp->m_u.ether_spec.h_dest)) { + rule->filter.match_flags |= IGC_FILTER_FLAG_DST_MAC_ADDR; + ether_addr_copy(rule->filter.dst_addr, + fsp->h_u.ether_spec.h_dest); } - - if (input->filter.match_flags & IGC_FILTER_FLAG_VLAN_TCI) - err = igc_rxnfc_write_vlan_prio_filter(adapter, input); - - return err; -} - -static void igc_clear_etype_filter_regs(struct igc_adapter *adapter, - u16 reg_index) -{ - struct igc_hw *hw = &adapter->hw; - u32 etqf = rd32(IGC_ETQF(reg_index)); - - etqf &= ~IGC_ETQF_QUEUE_ENABLE; - etqf &= ~IGC_ETQF_QUEUE_MASK; - etqf &= ~IGC_ETQF_FILTER_ENABLE; - - wr32(IGC_ETQF(reg_index), etqf); - - adapter->etype_bitmap[reg_index] = false; -} - -static void igc_clear_vlan_prio_filter(struct igc_adapter *adapter, - u16 vlan_tci) -{ - struct igc_hw *hw = &adapter->hw; - u8 vlan_priority; - u32 vlapqf; - - vlan_priority = (vlan_tci & VLAN_PRIO_MASK) >> VLAN_PRIO_SHIFT; - - vlapqf = rd32(IGC_VLAPQF); - vlapqf &= ~IGC_VLAPQF_P_VALID(vlan_priority); - vlapqf &= ~IGC_VLAPQF_QUEUE_SEL(vlan_priority, - IGC_VLAPQF_QUEUE_MASK); - - wr32(IGC_VLAPQF, vlapqf); } -int igc_erase_filter(struct igc_adapter *adapter, struct igc_nfc_filter *input) +/** + * igc_ethtool_check_nfc_rule() - Check if NFC rule is valid + * @adapter: Pointer to adapter + * @rule: Rule under evaluation + * + * The driver doesn't support rules with multiple matches so if more than + * one bit in filter flags is set, @rule is considered invalid. + * + * Also, if there is already another rule with the same filter in a different + * location, @rule is considered invalid. + * + * Context: Expects adapter->nfc_rule_lock to be held by caller. + * + * Return: 0 in case of success, negative errno code otherwise. + */ +static int igc_ethtool_check_nfc_rule(struct igc_adapter *adapter, + struct igc_nfc_rule *rule) { - if (input->filter.match_flags & IGC_FILTER_FLAG_ETHER_TYPE) - igc_clear_etype_filter_regs(adapter, - input->etype_reg_index); - - if (input->filter.match_flags & IGC_FILTER_FLAG_VLAN_TCI) - igc_clear_vlan_prio_filter(adapter, - ntohs(input->filter.vlan_tci)); - - if (input->filter.match_flags & IGC_FILTER_FLAG_SRC_MAC_ADDR) - igc_del_mac_steering_filter(adapter, input->filter.src_addr, - input->action, - IGC_MAC_STATE_SRC_ADDR); - - if (input->filter.match_flags & IGC_FILTER_FLAG_DST_MAC_ADDR) - igc_del_mac_steering_filter(adapter, input->filter.dst_addr, - input->action, 0); - - return 0; -} - -static int igc_update_ethtool_nfc_entry(struct igc_adapter *adapter, - struct igc_nfc_filter *input, - u16 sw_idx) -{ - struct igc_nfc_filter *rule, *parent; - int err = -EINVAL; - - parent = NULL; - rule = NULL; + struct net_device *dev = adapter->netdev; + u8 flags = rule->filter.match_flags; + struct igc_nfc_rule *tmp; - hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node) { - /* hash found, or no matching entry */ - if (rule->sw_idx >= sw_idx) - break; - parent = rule; + if (!flags) { + netdev_dbg(dev, "Rule with no match\n"); + return -EINVAL; } - /* if there is an old rule occupying our place remove it */ - if (rule && rule->sw_idx == sw_idx) { - if (!input) - err = igc_erase_filter(adapter, rule); - - hlist_del(&rule->nfc_node); - kfree(rule); - adapter->nfc_filter_count--; + if (flags & (flags - 1)) { + netdev_dbg(dev, "Rule with multiple matches not supported\n"); + return -EOPNOTSUPP; } - /* If no input this was a delete, err should be 0 if a rule was - * successfully found and removed from the list else -EINVAL - */ - if (!input) - return err; - - /* initialize node */ - INIT_HLIST_NODE(&input->nfc_node); - - /* add filter to the list */ - if (parent) - hlist_add_behind(&input->nfc_node, &parent->nfc_node); - else - hlist_add_head(&input->nfc_node, &adapter->nfc_filter_list); - - /* update counts */ - adapter->nfc_filter_count++; + list_for_each_entry(tmp, &adapter->nfc_rule_list, list) { + if (!memcmp(&rule->filter, &tmp->filter, + sizeof(rule->filter)) && + tmp->location != rule->location) { + netdev_dbg(dev, "Rule already exists\n"); + return -EEXIST; + } + } return 0; } -static int igc_add_ethtool_nfc_entry(struct igc_adapter *adapter, - struct ethtool_rxnfc *cmd) +static int igc_ethtool_add_nfc_rule(struct igc_adapter *adapter, + struct ethtool_rxnfc *cmd) { struct net_device *netdev = adapter->netdev; struct ethtool_rx_flow_spec *fsp = (struct ethtool_rx_flow_spec *)&cmd->fs; - struct igc_nfc_filter *input, *rule; - int err = 0; + struct igc_nfc_rule *rule, *old_rule; + int err; - if (!(netdev->hw_features & NETIF_F_NTUPLE)) + if (!(netdev->hw_features & NETIF_F_NTUPLE)) { + netdev_dbg(netdev, "N-tuple filters disabled\n"); return -EOPNOTSUPP; + } - /* Don't allow programming if the action is a queue greater than - * the number of online Rx queues. - */ - if (fsp->ring_cookie == RX_CLS_FLOW_DISC || - fsp->ring_cookie >= adapter->num_rx_queues) { - dev_err(&adapter->pdev->dev, "ethtool -N: The specified action is invalid\n"); - return -EINVAL; + if ((fsp->flow_type & ~FLOW_EXT) != ETHER_FLOW) { + netdev_dbg(netdev, "Only ethernet flow type is supported\n"); + return -EOPNOTSUPP; } - /* Don't allow indexes to exist outside of available space */ - if (fsp->location >= IGC_MAX_RXNFC_FILTERS) { - dev_err(&adapter->pdev->dev, "Location out of range\n"); - return -EINVAL; + if ((fsp->flow_type & FLOW_EXT) && + fsp->m_ext.vlan_tci != htons(VLAN_PRIO_MASK)) { + netdev_dbg(netdev, "VLAN mask not supported\n"); + return -EOPNOTSUPP; } - if ((fsp->flow_type & ~FLOW_EXT) != ETHER_FLOW) + if (fsp->ring_cookie >= adapter->num_rx_queues) { + netdev_dbg(netdev, "Invalid action\n"); return -EINVAL; - - input = kzalloc(sizeof(*input), GFP_KERNEL); - if (!input) - return -ENOMEM; - - if (fsp->m_u.ether_spec.h_proto == ETHER_TYPE_FULL_MASK) { - input->filter.etype = fsp->h_u.ether_spec.h_proto; - input->filter.match_flags = IGC_FILTER_FLAG_ETHER_TYPE; } - /* Only support matching addresses by the full mask */ - if (is_broadcast_ether_addr(fsp->m_u.ether_spec.h_source)) { - input->filter.match_flags |= IGC_FILTER_FLAG_SRC_MAC_ADDR; - ether_addr_copy(input->filter.src_addr, - fsp->h_u.ether_spec.h_source); + if (fsp->location >= IGC_MAX_RXNFC_RULES) { + netdev_dbg(netdev, "Invalid location\n"); + return -EINVAL; } - /* Only support matching addresses by the full mask */ - if (is_broadcast_ether_addr(fsp->m_u.ether_spec.h_dest)) { - input->filter.match_flags |= IGC_FILTER_FLAG_DST_MAC_ADDR; - ether_addr_copy(input->filter.dst_addr, - fsp->h_u.ether_spec.h_dest); - } + rule = kzalloc(sizeof(*rule), GFP_KERNEL); + if (!rule) + return -ENOMEM; - if ((fsp->flow_type & FLOW_EXT) && fsp->m_ext.vlan_tci) { - if (fsp->m_ext.vlan_tci != htons(VLAN_PRIO_MASK)) { - err = -EINVAL; - goto err_out; - } - input->filter.vlan_tci = fsp->h_ext.vlan_tci; - input->filter.match_flags |= IGC_FILTER_FLAG_VLAN_TCI; - } + igc_ethtool_init_nfc_rule(rule, fsp); - input->action = fsp->ring_cookie; - input->sw_idx = fsp->location; + mutex_lock(&adapter->nfc_rule_lock); - spin_lock(&adapter->nfc_lock); + err = igc_ethtool_check_nfc_rule(adapter, rule); + if (err) + goto err; - hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node) { - if (!memcmp(&input->filter, &rule->filter, - sizeof(input->filter))) { - err = -EEXIST; - dev_err(&adapter->pdev->dev, - "ethtool: this filter is already set\n"); - goto err_out_w_lock; - } - } + old_rule = igc_get_nfc_rule(adapter, fsp->location); + if (old_rule) + igc_del_nfc_rule(adapter, old_rule); - err = igc_add_filter(adapter, input); + err = igc_add_nfc_rule(adapter, rule); if (err) - goto err_out_w_lock; - - igc_update_ethtool_nfc_entry(adapter, input, input->sw_idx); + goto err; - spin_unlock(&adapter->nfc_lock); + mutex_unlock(&adapter->nfc_rule_lock); return 0; -err_out_w_lock: - spin_unlock(&adapter->nfc_lock); -err_out: - kfree(input); +err: + mutex_unlock(&adapter->nfc_rule_lock); + kfree(rule); return err; } -static int igc_del_ethtool_nfc_entry(struct igc_adapter *adapter, - struct ethtool_rxnfc *cmd) +static int igc_ethtool_del_nfc_rule(struct igc_adapter *adapter, + struct ethtool_rxnfc *cmd) { struct ethtool_rx_flow_spec *fsp = (struct ethtool_rx_flow_spec *)&cmd->fs; - int err; + struct igc_nfc_rule *rule; - spin_lock(&adapter->nfc_lock); - err = igc_update_ethtool_nfc_entry(adapter, NULL, fsp->location); - spin_unlock(&adapter->nfc_lock); + mutex_lock(&adapter->nfc_rule_lock); - return err; + rule = igc_get_nfc_rule(adapter, fsp->location); + if (!rule) { + mutex_unlock(&adapter->nfc_rule_lock); + return -EINVAL; + } + + igc_del_nfc_rule(adapter, rule); + + mutex_unlock(&adapter->nfc_rule_lock); + return 0; } -static int igc_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd) +static int igc_ethtool_set_rxnfc(struct net_device *dev, + struct ethtool_rxnfc *cmd) { struct igc_adapter *adapter = netdev_priv(dev); - int ret = -EOPNOTSUPP; switch (cmd->cmd) { case ETHTOOL_SRXFH: - ret = igc_set_rss_hash_opt(adapter, cmd); - break; + return igc_ethtool_set_rss_hash_opt(adapter, cmd); case ETHTOOL_SRXCLSRLINS: - ret = igc_add_ethtool_nfc_entry(adapter, cmd); - break; + return igc_ethtool_add_nfc_rule(adapter, cmd); case ETHTOOL_SRXCLSRLDEL: - ret = igc_del_ethtool_nfc_entry(adapter, cmd); + return igc_ethtool_del_nfc_rule(adapter, cmd); default: - break; + return -EOPNOTSUPP; } - - return ret; } void igc_write_rss_indir_tbl(struct igc_adapter *adapter) @@ -1533,13 +1385,13 @@ void igc_write_rss_indir_tbl(struct igc_adapter *adapter) } } -static u32 igc_get_rxfh_indir_size(struct net_device *netdev) +static u32 igc_ethtool_get_rxfh_indir_size(struct net_device *netdev) { return IGC_RETA_SIZE; } -static int igc_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, - u8 *hfunc) +static int igc_ethtool_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, + u8 *hfunc) { struct igc_adapter *adapter = netdev_priv(netdev); int i; @@ -1554,8 +1406,8 @@ static int igc_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key, return 0; } -static int igc_set_rxfh(struct net_device *netdev, const u32 *indir, - const u8 *key, const u8 hfunc) +static int igc_ethtool_set_rxfh(struct net_device *netdev, const u32 *indir, + const u8 *key, const u8 hfunc) { struct igc_adapter *adapter = netdev_priv(netdev); u32 num_queues; @@ -1583,18 +1435,13 @@ static int igc_set_rxfh(struct net_device *netdev, const u32 *indir, return 0; } -static unsigned int igc_max_channels(struct igc_adapter *adapter) -{ - return igc_get_max_rss_queues(adapter); -} - -static void igc_get_channels(struct net_device *netdev, - struct ethtool_channels *ch) +static void igc_ethtool_get_channels(struct net_device *netdev, + struct ethtool_channels *ch) { struct igc_adapter *adapter = netdev_priv(netdev); /* Report maximum channels */ - ch->max_combined = igc_max_channels(adapter); + ch->max_combined = igc_get_max_rss_queues(adapter); /* Report info for other vector */ if (adapter->flags & IGC_FLAG_HAS_MSIX) { @@ -1605,8 +1452,8 @@ static void igc_get_channels(struct net_device *netdev, ch->combined_count = adapter->rss_queues; } -static int igc_set_channels(struct net_device *netdev, - struct ethtool_channels *ch) +static int igc_ethtool_set_channels(struct net_device *netdev, + struct ethtool_channels *ch) { struct igc_adapter *adapter = netdev_priv(netdev); unsigned int count = ch->combined_count; @@ -1621,7 +1468,7 @@ static int igc_set_channels(struct net_device *netdev, return -EINVAL; /* Verify the number of channels doesn't exceed hw limits */ - max_combined = igc_max_channels(adapter); + max_combined = igc_get_max_rss_queues(adapter); if (count > max_combined) return -EINVAL; @@ -1638,8 +1485,8 @@ static int igc_set_channels(struct net_device *netdev, return 0; } -static int igc_get_ts_info(struct net_device *dev, - struct ethtool_ts_info *info) +static int igc_ethtool_get_ts_info(struct net_device *dev, + struct ethtool_ts_info *info) { struct igc_adapter *adapter = netdev_priv(dev); @@ -1671,7 +1518,7 @@ static int igc_get_ts_info(struct net_device *dev, } } -static u32 igc_get_priv_flags(struct net_device *netdev) +static u32 igc_ethtool_get_priv_flags(struct net_device *netdev) { struct igc_adapter *adapter = netdev_priv(netdev); u32 priv_flags = 0; @@ -1682,7 +1529,7 @@ static u32 igc_get_priv_flags(struct net_device *netdev) return priv_flags; } -static int igc_set_priv_flags(struct net_device *netdev, u32 priv_flags) +static int igc_ethtool_set_priv_flags(struct net_device *netdev, u32 priv_flags) { struct igc_adapter *adapter = netdev_priv(netdev); unsigned int flags = adapter->flags; @@ -1717,8 +1564,8 @@ static void igc_ethtool_complete(struct net_device *netdev) pm_runtime_put(&adapter->pdev->dev); } -static int igc_get_link_ksettings(struct net_device *netdev, - struct ethtool_link_ksettings *cmd) +static int igc_ethtool_get_link_ksettings(struct net_device *netdev, + struct ethtool_link_ksettings *cmd) { struct igc_adapter *adapter = netdev_priv(netdev); struct igc_hw *hw = &adapter->hw; @@ -1824,10 +1671,12 @@ static int igc_get_link_ksettings(struct net_device *netdev, return 0; } -static int igc_set_link_ksettings(struct net_device *netdev, - const struct ethtool_link_ksettings *cmd) +static int +igc_ethtool_set_link_ksettings(struct net_device *netdev, + const struct ethtool_link_ksettings *cmd) { struct igc_adapter *adapter = netdev_priv(netdev); + struct net_device *dev = adapter->netdev; struct igc_hw *hw = &adapter->hw; u32 advertising; @@ -1835,8 +1684,7 @@ static int igc_set_link_ksettings(struct net_device *netdev, * cannot be changed */ if (igc_check_reset_block(hw)) { - dev_err(&adapter->pdev->dev, - "Cannot change link characteristics when reset is active.\n"); + netdev_err(dev, "Cannot change link characteristics when reset is active\n"); return -EINVAL; } @@ -1847,7 +1695,7 @@ static int igc_set_link_ksettings(struct net_device *netdev, if (cmd->base.eth_tp_mdix_ctrl) { if (cmd->base.eth_tp_mdix_ctrl != ETH_TP_MDI_AUTO && cmd->base.autoneg != AUTONEG_ENABLE) { - dev_err(&adapter->pdev->dev, "forcing MDI/MDI-X state is not supported when link speed and/or duplex are forced\n"); + netdev_err(dev, "Forcing MDI/MDI-X state is not supported when link speed and/or duplex are forced\n"); return -EINVAL; } } @@ -1864,9 +1712,7 @@ static int igc_set_link_ksettings(struct net_device *netdev, if (adapter->fc_autoneg) hw->fc.requested_mode = igc_fc_default; } else { - /* calling this overrides forced MDI setting */ - dev_info(&adapter->pdev->dev, - "Force mode currently not supported\n"); + netdev_info(dev, "Force mode currently not supported\n"); } /* MDI-X => 2; MDI => 1; Auto => 3 */ @@ -1893,46 +1739,105 @@ static int igc_set_link_ksettings(struct net_device *netdev, return 0; } +static void igc_ethtool_diag_test(struct net_device *netdev, + struct ethtool_test *eth_test, u64 *data) +{ + struct igc_adapter *adapter = netdev_priv(netdev); + bool if_running = netif_running(netdev); + + if (eth_test->flags == ETH_TEST_FL_OFFLINE) { + netdev_info(adapter->netdev, "Offline testing starting"); + set_bit(__IGC_TESTING, &adapter->state); + + /* Link test performed before hardware reset so autoneg doesn't + * interfere with test result + */ + if (!igc_link_test(adapter, &data[TEST_LINK])) + eth_test->flags |= ETH_TEST_FL_FAILED; + + if (if_running) + igc_close(netdev); + else + igc_reset(adapter); + + netdev_info(adapter->netdev, "Register testing starting"); + if (!igc_reg_test(adapter, &data[TEST_REG])) + eth_test->flags |= ETH_TEST_FL_FAILED; + + igc_reset(adapter); + + netdev_info(adapter->netdev, "EEPROM testing starting"); + if (!igc_eeprom_test(adapter, &data[TEST_EEP])) + eth_test->flags |= ETH_TEST_FL_FAILED; + + igc_reset(adapter); + + /* loopback and interrupt tests + * will be implemented in the future + */ + data[TEST_LOOP] = 0; + data[TEST_IRQ] = 0; + + clear_bit(__IGC_TESTING, &adapter->state); + if (if_running) + igc_open(netdev); + } else { + netdev_info(adapter->netdev, "Online testing starting"); + + /* register, eeprom, intr and loopback tests not run online */ + data[TEST_REG] = 0; + data[TEST_EEP] = 0; + data[TEST_IRQ] = 0; + data[TEST_LOOP] = 0; + + if (!igc_link_test(adapter, &data[TEST_LINK])) + eth_test->flags |= ETH_TEST_FL_FAILED; + } + + msleep_interruptible(4 * 1000); +} + static const struct ethtool_ops igc_ethtool_ops = { .supported_coalesce_params = ETHTOOL_COALESCE_USECS, - .get_drvinfo = igc_get_drvinfo, - .get_regs_len = igc_get_regs_len, - .get_regs = igc_get_regs, - .get_wol = igc_get_wol, - .set_wol = igc_set_wol, - .get_msglevel = igc_get_msglevel, - .set_msglevel = igc_set_msglevel, - .nway_reset = igc_nway_reset, - .get_link = igc_get_link, - .get_eeprom_len = igc_get_eeprom_len, - .get_eeprom = igc_get_eeprom, - .set_eeprom = igc_set_eeprom, - .get_ringparam = igc_get_ringparam, - .set_ringparam = igc_set_ringparam, - .get_pauseparam = igc_get_pauseparam, - .set_pauseparam = igc_set_pauseparam, - .get_strings = igc_get_strings, - .get_sset_count = igc_get_sset_count, - .get_ethtool_stats = igc_get_ethtool_stats, - .get_coalesce = igc_get_coalesce, - .set_coalesce = igc_set_coalesce, - .get_rxnfc = igc_get_rxnfc, - .set_rxnfc = igc_set_rxnfc, - .get_rxfh_indir_size = igc_get_rxfh_indir_size, - .get_rxfh = igc_get_rxfh, - .set_rxfh = igc_set_rxfh, - .get_ts_info = igc_get_ts_info, - .get_channels = igc_get_channels, - .set_channels = igc_set_channels, - .get_priv_flags = igc_get_priv_flags, - .set_priv_flags = igc_set_priv_flags, + .get_drvinfo = igc_ethtool_get_drvinfo, + .get_regs_len = igc_ethtool_get_regs_len, + .get_regs = igc_ethtool_get_regs, + .get_wol = igc_ethtool_get_wol, + .set_wol = igc_ethtool_set_wol, + .get_msglevel = igc_ethtool_get_msglevel, + .set_msglevel = igc_ethtool_set_msglevel, + .nway_reset = igc_ethtool_nway_reset, + .get_link = igc_ethtool_get_link, + .get_eeprom_len = igc_ethtool_get_eeprom_len, + .get_eeprom = igc_ethtool_get_eeprom, + .set_eeprom = igc_ethtool_set_eeprom, + .get_ringparam = igc_ethtool_get_ringparam, + .set_ringparam = igc_ethtool_set_ringparam, + .get_pauseparam = igc_ethtool_get_pauseparam, + .set_pauseparam = igc_ethtool_set_pauseparam, + .get_strings = igc_ethtool_get_strings, + .get_sset_count = igc_ethtool_get_sset_count, + .get_ethtool_stats = igc_ethtool_get_stats, + .get_coalesce = igc_ethtool_get_coalesce, + .set_coalesce = igc_ethtool_set_coalesce, + .get_rxnfc = igc_ethtool_get_rxnfc, + .set_rxnfc = igc_ethtool_set_rxnfc, + .get_rxfh_indir_size = igc_ethtool_get_rxfh_indir_size, + .get_rxfh = igc_ethtool_get_rxfh, + .set_rxfh = igc_ethtool_set_rxfh, + .get_ts_info = igc_ethtool_get_ts_info, + .get_channels = igc_ethtool_get_channels, + .set_channels = igc_ethtool_set_channels, + .get_priv_flags = igc_ethtool_get_priv_flags, + .set_priv_flags = igc_ethtool_set_priv_flags, .begin = igc_ethtool_begin, .complete = igc_ethtool_complete, - .get_link_ksettings = igc_get_link_ksettings, - .set_link_ksettings = igc_set_link_ksettings, + .get_link_ksettings = igc_ethtool_get_link_ksettings, + .set_link_ksettings = igc_ethtool_set_link_ksettings, + .self_test = igc_ethtool_diag_test, }; -void igc_set_ethtool_ops(struct net_device *netdev) +void igc_ethtool_set_ops(struct net_device *netdev) { netdev->ethtool_ops = &igc_ethtool_ops; } diff --git a/drivers/net/ethernet/intel/igc/igc_hw.h b/drivers/net/ethernet/intel/igc/igc_hw.h index 90ac0e0144d8..af34ae310327 100644 --- a/drivers/net/ethernet/intel/igc/igc_hw.h +++ b/drivers/net/ethernet/intel/igc/igc_hw.h @@ -21,6 +21,9 @@ #define IGC_DEV_ID_I225_I 0x15F8 #define IGC_DEV_ID_I220_V 0x15F7 #define IGC_DEV_ID_I225_K 0x3100 +#define IGC_DEV_ID_I225_K2 0x3101 +#define IGC_DEV_ID_I225_LMVP 0x5502 +#define IGC_DEV_ID_I225_IT 0x0D9F #define IGC_DEV_ID_I225_BLANK_NVM 0x15FD /* Function pointers for the MAC. */ diff --git a/drivers/net/ethernet/intel/igc/igc_mac.c b/drivers/net/ethernet/intel/igc/igc_mac.c index 12aa6b5fcb5d..410aeb01de5c 100644 --- a/drivers/net/ethernet/intel/igc/igc_mac.c +++ b/drivers/net/ethernet/intel/igc/igc_mac.c @@ -235,15 +235,14 @@ out: void igc_clear_hw_cntrs_base(struct igc_hw *hw) { rd32(IGC_CRCERRS); - rd32(IGC_SYMERRS); rd32(IGC_MPC); rd32(IGC_SCC); rd32(IGC_ECOL); rd32(IGC_MCC); rd32(IGC_LATECOL); rd32(IGC_COLC); + rd32(IGC_RERC); rd32(IGC_DC); - rd32(IGC_SEC); rd32(IGC_RLEC); rd32(IGC_XONRXC); rd32(IGC_XONTXC); @@ -288,7 +287,7 @@ void igc_clear_hw_cntrs_base(struct igc_hw *hw) rd32(IGC_ALGNERRC); rd32(IGC_RXERRC); rd32(IGC_TNCRS); - rd32(IGC_CEXTERR); + rd32(IGC_HTDPMC); rd32(IGC_TSCTC); rd32(IGC_TSCTFC); @@ -307,12 +306,8 @@ void igc_clear_hw_cntrs_base(struct igc_hw *hw) rd32(IGC_ICTXQMTC); rd32(IGC_ICRXDMTC); - rd32(IGC_CBTMPC); - rd32(IGC_HTDPMC); - rd32(IGC_CBRMPC); rd32(IGC_RPTHC); rd32(IGC_HGPTC); - rd32(IGC_HTCBDPC); rd32(IGC_HGORCL); rd32(IGC_HGORCH); rd32(IGC_HGOTCL); diff --git a/drivers/net/ethernet/intel/igc/igc_mac.h b/drivers/net/ethernet/intel/igc/igc_mac.h index 832cccec87cd..b5963f86defb 100644 --- a/drivers/net/ethernet/intel/igc/igc_mac.h +++ b/drivers/net/ethernet/intel/igc/igc_mac.h @@ -8,10 +8,6 @@ #include "igc_phy.h" #include "igc_defines.h" -#ifndef IGC_REMOVED -#define IGC_REMOVED(a) (0) -#endif /* IGC_REMOVED */ - /* forward declaration */ s32 igc_disable_pcie_master(struct igc_hw *hw); s32 igc_check_for_copper_link(struct igc_hw *hw); diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c index 69fa1ce1f927..6919c50e449a 100644 --- a/drivers/net/ethernet/intel/igc/igc_main.c +++ b/drivers/net/ethernet/intel/igc/igc_main.c @@ -9,11 +9,13 @@ #include <linux/udp.h> #include <linux/ip.h> #include <linux/pm_runtime.h> +#include <net/pkt_sched.h> #include <net/ipv6.h> #include "igc.h" #include "igc_hw.h" +#include "igc_tsn.h" #define DRV_VERSION "0.0.1-k" #define DRV_SUMMARY "Intel(R) 2.5G Ethernet Linux Driver" @@ -45,6 +47,9 @@ static const struct pci_device_id igc_pci_tbl[] = { { PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_I), board_base }, { PCI_VDEVICE(INTEL, IGC_DEV_ID_I220_V), board_base }, { PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_K), board_base }, + { PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_K2), board_base }, + { PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_LMVP), board_base }, + { PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_IT), board_base }, { PCI_VDEVICE(INTEL, IGC_DEV_ID_I225_BLANK_NVM), board_base }, /* required last entry */ {0, } @@ -71,7 +76,7 @@ static void igc_power_down_link(struct igc_adapter *adapter) void igc_reset(struct igc_adapter *adapter) { - struct pci_dev *pdev = adapter->pdev; + struct net_device *dev = adapter->netdev; struct igc_hw *hw = &adapter->hw; struct igc_fc_info *fc = &hw->fc; u32 pba, hwm; @@ -98,7 +103,7 @@ void igc_reset(struct igc_adapter *adapter) hw->mac.ops.reset_hw(hw); if (hw->mac.ops.init_hw(hw)) - dev_err(&pdev->dev, "Hardware Error\n"); + netdev_err(dev, "Error on hardware initialization\n"); if (!netif_running(adapter->netdev)) igc_power_down_link(adapter); @@ -106,6 +111,9 @@ void igc_reset(struct igc_adapter *adapter) /* Re-enable PTP, where applicable. */ igc_ptp_reset(adapter); + /* Re-enable TSN offloading, where applicable. */ + igc_tsn_offload_apply(adapter); + igc_get_phy_info(hw); } @@ -280,6 +288,7 @@ static void igc_clean_all_tx_rings(struct igc_adapter *adapter) */ int igc_setup_tx_resources(struct igc_ring *tx_ring) { + struct net_device *ndev = tx_ring->netdev; struct device *dev = tx_ring->dev; int size = 0; @@ -305,8 +314,7 @@ int igc_setup_tx_resources(struct igc_ring *tx_ring) err: vfree(tx_ring->tx_buffer_info); - dev_err(dev, - "Unable to allocate memory for the transmit descriptor ring\n"); + netdev_err(ndev, "Unable to allocate memory for Tx descriptor ring\n"); return -ENOMEM; } @@ -318,14 +326,13 @@ err: */ static int igc_setup_all_tx_resources(struct igc_adapter *adapter) { - struct pci_dev *pdev = adapter->pdev; + struct net_device *dev = adapter->netdev; int i, err = 0; for (i = 0; i < adapter->num_tx_queues; i++) { err = igc_setup_tx_resources(adapter->tx_ring[i]); if (err) { - dev_err(&pdev->dev, - "Allocation for Tx Queue %u failed\n", i); + netdev_err(dev, "Error on Tx queue %u setup\n", i); for (i--; i >= 0; i--) igc_free_tx_resources(adapter->tx_ring[i]); break; @@ -436,6 +443,7 @@ static void igc_free_all_rx_resources(struct igc_adapter *adapter) */ int igc_setup_rx_resources(struct igc_ring *rx_ring) { + struct net_device *ndev = rx_ring->netdev; struct device *dev = rx_ring->dev; int size, desc_len; @@ -465,8 +473,7 @@ int igc_setup_rx_resources(struct igc_ring *rx_ring) err: vfree(rx_ring->rx_buffer_info); rx_ring->rx_buffer_info = NULL; - dev_err(dev, - "Unable to allocate memory for the receive descriptor ring\n"); + netdev_err(ndev, "Unable to allocate memory for Rx descriptor ring\n"); return -ENOMEM; } @@ -479,14 +486,13 @@ err: */ static int igc_setup_all_rx_resources(struct igc_adapter *adapter) { - struct pci_dev *pdev = adapter->pdev; + struct net_device *dev = adapter->netdev; int i, err = 0; for (i = 0; i < adapter->num_rx_queues; i++) { err = igc_setup_rx_resources(adapter->rx_ring[i]); if (err) { - dev_err(&pdev->dev, - "Allocation for Rx Queue %u failed\n", i); + netdev_err(dev, "Error on Rx queue %u setup\n", i); for (i--; i >= 0; i--) igc_free_rx_resources(adapter->rx_ring[i]); break; @@ -757,48 +763,76 @@ static void igc_setup_tctl(struct igc_adapter *adapter) } /** - * igc_rar_set_index - Sync RAL[index] and RAH[index] registers with MAC table - * @adapter: address of board private structure - * @index: Index of the RAR entry which need to be synced with MAC table + * igc_set_mac_filter_hw() - Set MAC address filter in hardware + * @adapter: Pointer to adapter where the filter should be set + * @index: Filter index + * @type: MAC address filter type (source or destination) + * @addr: MAC address + * @queue: If non-negative, queue assignment feature is enabled and frames + * matching the filter are enqueued onto 'queue'. Otherwise, queue + * assignment is disabled. */ -static void igc_rar_set_index(struct igc_adapter *adapter, u32 index) +static void igc_set_mac_filter_hw(struct igc_adapter *adapter, int index, + enum igc_mac_filter_type type, + const u8 *addr, int queue) { - u8 *addr = adapter->mac_table[index].addr; + struct net_device *dev = adapter->netdev; struct igc_hw *hw = &adapter->hw; - u32 rar_low, rar_high; + u32 ral, rah; - /* HW expects these to be in network order when they are plugged - * into the registers which are little endian. In order to guarantee - * that ordering we need to do an leXX_to_cpup here in order to be - * ready for the byteswap that occurs with writel - */ - rar_low = le32_to_cpup((__le32 *)(addr)); - rar_high = le16_to_cpup((__le16 *)(addr + 4)); + if (WARN_ON(index >= hw->mac.rar_entry_count)) + return; - /* Indicate to hardware the Address is Valid. */ - if (adapter->mac_table[index].state & IGC_MAC_STATE_IN_USE) { - if (is_valid_ether_addr(addr)) - rar_high |= IGC_RAH_AV; + ral = le32_to_cpup((__le32 *)(addr)); + rah = le16_to_cpup((__le16 *)(addr + 4)); - rar_high |= IGC_RAH_POOL_1 << - adapter->mac_table[index].queue; + if (type == IGC_MAC_FILTER_TYPE_SRC) { + rah &= ~IGC_RAH_ASEL_MASK; + rah |= IGC_RAH_ASEL_SRC_ADDR; } - wr32(IGC_RAL(index), rar_low); - wrfl(); - wr32(IGC_RAH(index), rar_high); - wrfl(); + if (queue >= 0) { + rah &= ~IGC_RAH_QSEL_MASK; + rah |= (queue << IGC_RAH_QSEL_SHIFT); + rah |= IGC_RAH_QSEL_ENABLE; + } + + rah |= IGC_RAH_AV; + + wr32(IGC_RAL(index), ral); + wr32(IGC_RAH(index), rah); + + netdev_dbg(dev, "MAC address filter set in HW: index %d", index); +} + +/** + * igc_clear_mac_filter_hw() - Clear MAC address filter in hardware + * @adapter: Pointer to adapter where the filter should be cleared + * @index: Filter index + */ +static void igc_clear_mac_filter_hw(struct igc_adapter *adapter, int index) +{ + struct net_device *dev = adapter->netdev; + struct igc_hw *hw = &adapter->hw; + + if (WARN_ON(index >= hw->mac.rar_entry_count)) + return; + + wr32(IGC_RAL(index), 0); + wr32(IGC_RAH(index), 0); + + netdev_dbg(dev, "MAC address filter cleared in HW: index %d", index); } /* Set default MAC address for the PF in the first RAR entry */ static void igc_set_default_mac_filter(struct igc_adapter *adapter) { - struct igc_mac_addr *mac_table = &adapter->mac_table[0]; + struct net_device *dev = adapter->netdev; + u8 *addr = adapter->hw.mac.addr; - ether_addr_copy(mac_table->addr, adapter->hw.mac.addr); - mac_table->state = IGC_MAC_STATE_DEFAULT | IGC_MAC_STATE_IN_USE; + netdev_dbg(dev, "Set default MAC address filter: address %pM", addr); - igc_rar_set_index(adapter, 0); + igc_set_mac_filter_hw(adapter, 0, IGC_MAC_FILTER_TYPE_DST, addr, -1); } /** @@ -864,6 +898,23 @@ static int igc_write_mc_addr_list(struct net_device *netdev) return netdev_mc_count(netdev); } +static __le32 igc_tx_launchtime(struct igc_adapter *adapter, ktime_t txtime) +{ + ktime_t cycle_time = adapter->cycle_time; + ktime_t base_time = adapter->base_time; + u32 launchtime; + + /* FIXME: when using ETF together with taprio, we may have a + * case where 'delta' is larger than the cycle_time, this may + * cause problems if we don't read the current value of + * IGC_BASET, as the value writen into the launchtime + * descriptor field may be misinterpreted. + */ + div_s64_rem(ktime_sub_ns(txtime, base_time), cycle_time, &launchtime); + + return cpu_to_le32(launchtime); +} + static void igc_tx_ctxtdesc(struct igc_ring *tx_ring, struct igc_tx_buffer *first, u32 vlan_macip_lens, u32 type_tucmd, @@ -871,7 +922,6 @@ static void igc_tx_ctxtdesc(struct igc_ring *tx_ring, { struct igc_adv_tx_context_desc *context_desc; u16 i = tx_ring->next_to_use; - struct timespec64 ts; context_desc = IGC_TX_CTXTDESC(tx_ring, i); @@ -893,9 +943,12 @@ static void igc_tx_ctxtdesc(struct igc_ring *tx_ring, * should have been handled by the upper layers. */ if (tx_ring->launchtime_enable) { - ts = ktime_to_timespec64(first->skb->tstamp); + struct igc_adapter *adapter = netdev_priv(tx_ring->netdev); + ktime_t txtime = first->skb->tstamp; + first->skb->tstamp = ktime_set(0, 0); - context_desc->launch_time = cpu_to_le32(ts.tv_nsec / 32); + context_desc->launch_time = igc_tx_launchtime(adapter, + txtime); } else { context_desc->launch_time = 0; } @@ -1143,7 +1196,7 @@ static int igc_tx_map(struct igc_ring *tx_ring, return 0; dma_error: - dev_err(tx_ring->dev, "TX DMA map failed\n"); + netdev_err(tx_ring->netdev, "TX DMA map failed\n"); tx_buffer = &tx_ring->tx_buffer_info[i]; /* clear dma mappings for failed tx_buffer_info map */ @@ -1406,8 +1459,8 @@ static void igc_rx_checksum(struct igc_ring *ring, IGC_RXD_STAT_UDPCS)) skb->ip_summed = CHECKSUM_UNNECESSARY; - dev_dbg(ring->dev, "cksum success: bits %08X\n", - le32_to_cpu(rx_desc->wb.upper.status_error)); + netdev_dbg(ring->netdev, "cksum success: bits %08X\n", + le32_to_cpu(rx_desc->wb.upper.status_error)); } static inline void igc_rx_hash(struct igc_ring *ring, @@ -2069,27 +2122,27 @@ static bool igc_clean_tx_irq(struct igc_q_vector *q_vector, int napi_budget) (adapter->tx_timeout_factor * HZ)) && !(rd32(IGC_STATUS) & IGC_STATUS_TXOFF)) { /* detected Tx unit hang */ - dev_err(tx_ring->dev, - "Detected Tx Unit Hang\n" - " Tx Queue <%d>\n" - " TDH <%x>\n" - " TDT <%x>\n" - " next_to_use <%x>\n" - " next_to_clean <%x>\n" - "buffer_info[next_to_clean]\n" - " time_stamp <%lx>\n" - " next_to_watch <%p>\n" - " jiffies <%lx>\n" - " desc.status <%x>\n", - tx_ring->queue_index, - rd32(IGC_TDH(tx_ring->reg_idx)), - readl(tx_ring->tail), - tx_ring->next_to_use, - tx_ring->next_to_clean, - tx_buffer->time_stamp, - tx_buffer->next_to_watch, - jiffies, - tx_buffer->next_to_watch->wb.status); + netdev_err(tx_ring->netdev, + "Detected Tx Unit Hang\n" + " Tx Queue <%d>\n" + " TDH <%x>\n" + " TDT <%x>\n" + " next_to_use <%x>\n" + " next_to_clean <%x>\n" + "buffer_info[next_to_clean]\n" + " time_stamp <%lx>\n" + " next_to_watch <%p>\n" + " jiffies <%lx>\n" + " desc.status <%x>\n", + tx_ring->queue_index, + rd32(IGC_TDH(tx_ring->reg_idx)), + readl(tx_ring->tail), + tx_ring->next_to_use, + tx_ring->next_to_clean, + tx_buffer->time_stamp, + tx_buffer->next_to_watch, + jiffies, + tx_buffer->next_to_watch->wb.status); netif_stop_subqueue(tx_ring->netdev, tx_ring->queue_index); @@ -2121,140 +2174,435 @@ static bool igc_clean_tx_irq(struct igc_q_vector *q_vector, int napi_budget) return !!budget; } -static void igc_nfc_filter_restore(struct igc_adapter *adapter) +static int igc_find_mac_filter(struct igc_adapter *adapter, + enum igc_mac_filter_type type, const u8 *addr) { - struct igc_nfc_filter *rule; + struct igc_hw *hw = &adapter->hw; + int max_entries = hw->mac.rar_entry_count; + u32 ral, rah; + int i; - spin_lock(&adapter->nfc_lock); + for (i = 0; i < max_entries; i++) { + ral = rd32(IGC_RAL(i)); + rah = rd32(IGC_RAH(i)); - hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node) - igc_add_filter(adapter, rule); + if (!(rah & IGC_RAH_AV)) + continue; + if (!!(rah & IGC_RAH_ASEL_SRC_ADDR) != type) + continue; + if ((rah & IGC_RAH_RAH_MASK) != + le16_to_cpup((__le16 *)(addr + 4))) + continue; + if (ral != le32_to_cpup((__le32 *)(addr))) + continue; + + return i; + } - spin_unlock(&adapter->nfc_lock); + return -1; } -/* If the filter to be added and an already existing filter express - * the same address and address type, it should be possible to only - * override the other configurations, for example the queue to steer - * traffic. - */ -static bool igc_mac_entry_can_be_used(const struct igc_mac_addr *entry, - const u8 *addr, const u8 flags) +static int igc_get_avail_mac_filter_slot(struct igc_adapter *adapter) { - if (!(entry->state & IGC_MAC_STATE_IN_USE)) - return true; + struct igc_hw *hw = &adapter->hw; + int max_entries = hw->mac.rar_entry_count; + u32 rah; + int i; - if ((entry->state & IGC_MAC_STATE_SRC_ADDR) != - (flags & IGC_MAC_STATE_SRC_ADDR)) - return false; + for (i = 0; i < max_entries; i++) { + rah = rd32(IGC_RAH(i)); - if (!ether_addr_equal(addr, entry->addr)) - return false; + if (!(rah & IGC_RAH_AV)) + return i; + } - return true; + return -1; } -/* Add a MAC filter for 'addr' directing matching traffic to 'queue', - * 'flags' is used to indicate what kind of match is made, match is by - * default for the destination address, if matching by source address - * is desired the flag IGC_MAC_STATE_SRC_ADDR can be used. +/** + * igc_add_mac_filter() - Add MAC address filter + * @adapter: Pointer to adapter where the filter should be added + * @type: MAC address filter type (source or destination) + * @addr: MAC address + * @queue: If non-negative, queue assignment feature is enabled and frames + * matching the filter are enqueued onto 'queue'. Otherwise, queue + * assignment is disabled. + * + * Return: 0 in case of success, negative errno code otherwise. */ static int igc_add_mac_filter(struct igc_adapter *adapter, - const u8 *addr, const u8 queue) + enum igc_mac_filter_type type, const u8 *addr, + int queue) +{ + struct net_device *dev = adapter->netdev; + int index; + + index = igc_find_mac_filter(adapter, type, addr); + if (index >= 0) + goto update_filter; + + index = igc_get_avail_mac_filter_slot(adapter); + if (index < 0) + return -ENOSPC; + + netdev_dbg(dev, "Add MAC address filter: index %d type %s address %pM queue %d\n", + index, type == IGC_MAC_FILTER_TYPE_DST ? "dst" : "src", + addr, queue); + +update_filter: + igc_set_mac_filter_hw(adapter, index, type, addr, queue); + return 0; +} + +/** + * igc_del_mac_filter() - Delete MAC address filter + * @adapter: Pointer to adapter where the filter should be deleted from + * @type: MAC address filter type (source or destination) + * @addr: MAC address + */ +static void igc_del_mac_filter(struct igc_adapter *adapter, + enum igc_mac_filter_type type, const u8 *addr) { + struct net_device *dev = adapter->netdev; + int index; + + index = igc_find_mac_filter(adapter, type, addr); + if (index < 0) + return; + + if (index == 0) { + /* If this is the default filter, we don't actually delete it. + * We just reset to its default value i.e. disable queue + * assignment. + */ + netdev_dbg(dev, "Disable default MAC filter queue assignment"); + + igc_set_mac_filter_hw(adapter, 0, type, addr, -1); + } else { + netdev_dbg(dev, "Delete MAC address filter: index %d type %s address %pM\n", + index, + type == IGC_MAC_FILTER_TYPE_DST ? "dst" : "src", + addr); + + igc_clear_mac_filter_hw(adapter, index); + } +} + +/** + * igc_add_vlan_prio_filter() - Add VLAN priority filter + * @adapter: Pointer to adapter where the filter should be added + * @prio: VLAN priority value + * @queue: Queue number which matching frames are assigned to + * + * Return: 0 in case of success, negative errno code otherwise. + */ +static int igc_add_vlan_prio_filter(struct igc_adapter *adapter, int prio, + int queue) +{ + struct net_device *dev = adapter->netdev; struct igc_hw *hw = &adapter->hw; - int rar_entries = hw->mac.rar_entry_count; - int i; + u32 vlanpqf; - if (is_zero_ether_addr(addr)) - return -EINVAL; + vlanpqf = rd32(IGC_VLANPQF); - /* Search for the first empty entry in the MAC table. - * Do not touch entries at the end of the table reserved for the VF MAC - * addresses. - */ - for (i = 0; i < rar_entries; i++) { - if (!igc_mac_entry_can_be_used(&adapter->mac_table[i], - addr, 0)) - continue; + if (vlanpqf & IGC_VLANPQF_VALID(prio)) { + netdev_dbg(dev, "VLAN priority filter already in use\n"); + return -EEXIST; + } - ether_addr_copy(adapter->mac_table[i].addr, addr); - adapter->mac_table[i].queue = queue; - adapter->mac_table[i].state |= IGC_MAC_STATE_IN_USE; + vlanpqf |= IGC_VLANPQF_QSEL(prio, queue); + vlanpqf |= IGC_VLANPQF_VALID(prio); - igc_rar_set_index(adapter, i); - return i; + wr32(IGC_VLANPQF, vlanpqf); + + netdev_dbg(dev, "Add VLAN priority filter: prio %d queue %d\n", + prio, queue); + return 0; +} + +/** + * igc_del_vlan_prio_filter() - Delete VLAN priority filter + * @adapter: Pointer to adapter where the filter should be deleted from + * @prio: VLAN priority value + */ +static void igc_del_vlan_prio_filter(struct igc_adapter *adapter, int prio) +{ + struct igc_hw *hw = &adapter->hw; + u32 vlanpqf; + + vlanpqf = rd32(IGC_VLANPQF); + + vlanpqf &= ~IGC_VLANPQF_VALID(prio); + vlanpqf &= ~IGC_VLANPQF_QSEL(prio, IGC_VLANPQF_QUEUE_MASK); + + wr32(IGC_VLANPQF, vlanpqf); + + netdev_dbg(adapter->netdev, "Delete VLAN priority filter: prio %d\n", + prio); +} + +static int igc_get_avail_etype_filter_slot(struct igc_adapter *adapter) +{ + struct igc_hw *hw = &adapter->hw; + int i; + + for (i = 0; i < MAX_ETYPE_FILTER; i++) { + u32 etqf = rd32(IGC_ETQF(i)); + + if (!(etqf & IGC_ETQF_FILTER_ENABLE)) + return i; } - return -ENOSPC; + return -1; } -/* Remove a MAC filter for 'addr' directing matching traffic to - * 'queue', 'flags' is used to indicate what kind of match need to be - * removed, match is by default for the destination address, if - * matching by source address is to be removed the flag - * IGC_MAC_STATE_SRC_ADDR can be used. +/** + * igc_add_etype_filter() - Add ethertype filter + * @adapter: Pointer to adapter where the filter should be added + * @etype: Ethertype value + * @queue: If non-negative, queue assignment feature is enabled and frames + * matching the filter are enqueued onto 'queue'. Otherwise, queue + * assignment is disabled. + * + * Return: 0 in case of success, negative errno code otherwise. */ -static int igc_del_mac_filter(struct igc_adapter *adapter, - const u8 *addr, const u8 queue) +static int igc_add_etype_filter(struct igc_adapter *adapter, u16 etype, + int queue) +{ + struct igc_hw *hw = &adapter->hw; + int index; + u32 etqf; + + index = igc_get_avail_etype_filter_slot(adapter); + if (index < 0) + return -ENOSPC; + + etqf = rd32(IGC_ETQF(index)); + + etqf &= ~IGC_ETQF_ETYPE_MASK; + etqf |= etype; + + if (queue >= 0) { + etqf &= ~IGC_ETQF_QUEUE_MASK; + etqf |= (queue << IGC_ETQF_QUEUE_SHIFT); + etqf |= IGC_ETQF_QUEUE_ENABLE; + } + + etqf |= IGC_ETQF_FILTER_ENABLE; + + wr32(IGC_ETQF(index), etqf); + + netdev_dbg(adapter->netdev, "Add ethertype filter: etype %04x queue %d\n", + etype, queue); + return 0; +} + +static int igc_find_etype_filter(struct igc_adapter *adapter, u16 etype) { struct igc_hw *hw = &adapter->hw; - int rar_entries = hw->mac.rar_entry_count; int i; - if (is_zero_ether_addr(addr)) - return -EINVAL; + for (i = 0; i < MAX_ETYPE_FILTER; i++) { + u32 etqf = rd32(IGC_ETQF(i)); - /* Search for matching entry in the MAC table based on given address - * and queue. Do not touch entries at the end of the table reserved - * for the VF MAC addresses. - */ - for (i = 0; i < rar_entries; i++) { - if (!(adapter->mac_table[i].state & IGC_MAC_STATE_IN_USE)) - continue; - if (adapter->mac_table[i].state != 0) - continue; - if (adapter->mac_table[i].queue != queue) - continue; - if (!ether_addr_equal(adapter->mac_table[i].addr, addr)) - continue; + if ((etqf & IGC_ETQF_ETYPE_MASK) == etype) + return i; + } - /* When a filter for the default address is "deleted", - * we return it to its initial configuration - */ - if (adapter->mac_table[i].state & IGC_MAC_STATE_DEFAULT) { - adapter->mac_table[i].state = - IGC_MAC_STATE_DEFAULT | IGC_MAC_STATE_IN_USE; - adapter->mac_table[i].queue = 0; - } else { - adapter->mac_table[i].state = 0; - adapter->mac_table[i].queue = 0; - memset(adapter->mac_table[i].addr, 0, ETH_ALEN); - } + return -1; +} - igc_rar_set_index(adapter, i); - return 0; +/** + * igc_del_etype_filter() - Delete ethertype filter + * @adapter: Pointer to adapter where the filter should be deleted from + * @etype: Ethertype value + */ +static void igc_del_etype_filter(struct igc_adapter *adapter, u16 etype) +{ + struct igc_hw *hw = &adapter->hw; + int index; + + index = igc_find_etype_filter(adapter, etype); + if (index < 0) + return; + + wr32(IGC_ETQF(index), 0); + + netdev_dbg(adapter->netdev, "Delete ethertype filter: etype %04x\n", + etype); +} + +static int igc_enable_nfc_rule(struct igc_adapter *adapter, + const struct igc_nfc_rule *rule) +{ + int err; + + if (rule->filter.match_flags & IGC_FILTER_FLAG_ETHER_TYPE) { + err = igc_add_etype_filter(adapter, rule->filter.etype, + rule->action); + if (err) + return err; + } + + if (rule->filter.match_flags & IGC_FILTER_FLAG_SRC_MAC_ADDR) { + err = igc_add_mac_filter(adapter, IGC_MAC_FILTER_TYPE_SRC, + rule->filter.src_addr, rule->action); + if (err) + return err; + } + + if (rule->filter.match_flags & IGC_FILTER_FLAG_DST_MAC_ADDR) { + err = igc_add_mac_filter(adapter, IGC_MAC_FILTER_TYPE_DST, + rule->filter.dst_addr, rule->action); + if (err) + return err; + } + + if (rule->filter.match_flags & IGC_FILTER_FLAG_VLAN_TCI) { + int prio = (rule->filter.vlan_tci & VLAN_PRIO_MASK) >> + VLAN_PRIO_SHIFT; + + err = igc_add_vlan_prio_filter(adapter, prio, rule->action); + if (err) + return err; + } + + return 0; +} + +static void igc_disable_nfc_rule(struct igc_adapter *adapter, + const struct igc_nfc_rule *rule) +{ + if (rule->filter.match_flags & IGC_FILTER_FLAG_ETHER_TYPE) + igc_del_etype_filter(adapter, rule->filter.etype); + + if (rule->filter.match_flags & IGC_FILTER_FLAG_VLAN_TCI) { + int prio = (rule->filter.vlan_tci & VLAN_PRIO_MASK) >> + VLAN_PRIO_SHIFT; + + igc_del_vlan_prio_filter(adapter, prio); } - return -ENOENT; + if (rule->filter.match_flags & IGC_FILTER_FLAG_SRC_MAC_ADDR) + igc_del_mac_filter(adapter, IGC_MAC_FILTER_TYPE_SRC, + rule->filter.src_addr); + + if (rule->filter.match_flags & IGC_FILTER_FLAG_DST_MAC_ADDR) + igc_del_mac_filter(adapter, IGC_MAC_FILTER_TYPE_DST, + rule->filter.dst_addr); +} + +/** + * igc_get_nfc_rule() - Get NFC rule + * @adapter: Pointer to adapter + * @location: Rule location + * + * Context: Expects adapter->nfc_rule_lock to be held by caller. + * + * Return: Pointer to NFC rule at @location. If not found, NULL. + */ +struct igc_nfc_rule *igc_get_nfc_rule(struct igc_adapter *adapter, + u32 location) +{ + struct igc_nfc_rule *rule; + + list_for_each_entry(rule, &adapter->nfc_rule_list, list) { + if (rule->location == location) + return rule; + if (rule->location > location) + break; + } + + return NULL; +} + +/** + * igc_del_nfc_rule() - Delete NFC rule + * @adapter: Pointer to adapter + * @rule: Pointer to rule to be deleted + * + * Disable NFC rule in hardware and delete it from adapter. + * + * Context: Expects adapter->nfc_rule_lock to be held by caller. + */ +void igc_del_nfc_rule(struct igc_adapter *adapter, struct igc_nfc_rule *rule) +{ + igc_disable_nfc_rule(adapter, rule); + + list_del(&rule->list); + adapter->nfc_rule_count--; + + kfree(rule); +} + +static void igc_flush_nfc_rules(struct igc_adapter *adapter) +{ + struct igc_nfc_rule *rule, *tmp; + + mutex_lock(&adapter->nfc_rule_lock); + + list_for_each_entry_safe(rule, tmp, &adapter->nfc_rule_list, list) + igc_del_nfc_rule(adapter, rule); + + mutex_unlock(&adapter->nfc_rule_lock); +} + +/** + * igc_add_nfc_rule() - Add NFC rule + * @adapter: Pointer to adapter + * @rule: Pointer to rule to be added + * + * Enable NFC rule in hardware and add it to adapter. + * + * Context: Expects adapter->nfc_rule_lock to be held by caller. + * + * Return: 0 on success, negative errno on failure. + */ +int igc_add_nfc_rule(struct igc_adapter *adapter, struct igc_nfc_rule *rule) +{ + struct igc_nfc_rule *pred, *cur; + int err; + + err = igc_enable_nfc_rule(adapter, rule); + if (err) + return err; + + pred = NULL; + list_for_each_entry(cur, &adapter->nfc_rule_list, list) { + if (cur->location >= rule->location) + break; + pred = cur; + } + + list_add(&rule->list, pred ? &pred->list : &adapter->nfc_rule_list); + adapter->nfc_rule_count++; + return 0; +} + +static void igc_restore_nfc_rules(struct igc_adapter *adapter) +{ + struct igc_nfc_rule *rule; + + mutex_lock(&adapter->nfc_rule_lock); + + list_for_each_entry_reverse(rule, &adapter->nfc_rule_list, list) + igc_enable_nfc_rule(adapter, rule); + + mutex_unlock(&adapter->nfc_rule_lock); } static int igc_uc_sync(struct net_device *netdev, const unsigned char *addr) { struct igc_adapter *adapter = netdev_priv(netdev); - int ret; - - ret = igc_add_mac_filter(adapter, addr, adapter->num_rx_queues); - return min_t(int, ret, 0); + return igc_add_mac_filter(adapter, IGC_MAC_FILTER_TYPE_DST, addr, -1); } static int igc_uc_unsync(struct net_device *netdev, const unsigned char *addr) { struct igc_adapter *adapter = netdev_priv(netdev); - igc_del_mac_filter(adapter, addr, adapter->num_rx_queues); - + igc_del_mac_filter(adapter, IGC_MAC_FILTER_TYPE_DST, addr); return 0; } @@ -2325,7 +2673,9 @@ static void igc_configure(struct igc_adapter *adapter) igc_setup_mrqc(adapter); igc_setup_rctl(adapter); - igc_nfc_filter_restore(adapter); + igc_set_default_mac_filter(adapter); + igc_restore_nfc_rules(adapter); + igc_configure_tx(adapter); igc_configure_rx(adapter); @@ -2518,12 +2868,7 @@ void igc_set_flag_queue_pairs(struct igc_adapter *adapter, unsigned int igc_get_max_rss_queues(struct igc_adapter *adapter) { - unsigned int max_rss_queues; - - /* Determine the maximum number of RSS queues supported. */ - max_rss_queues = IGC_MAX_RX_QUEUES; - - return max_rss_queues; + return IGC_MAX_RX_QUEUES; } static void igc_init_queue_configuration(struct igc_adapter *adapter) @@ -3164,14 +3509,14 @@ err_out: */ static int igc_init_interrupt_scheme(struct igc_adapter *adapter, bool msix) { - struct pci_dev *pdev = adapter->pdev; + struct net_device *dev = adapter->netdev; int err = 0; igc_set_interrupt_capability(adapter, msix); err = igc_alloc_q_vectors(adapter); if (err) { - dev_err(&pdev->dev, "Unable to allocate memory for vectors\n"); + netdev_err(dev, "Unable to allocate memory for vectors\n"); goto err_alloc_q_vectors; } @@ -3198,8 +3543,6 @@ static int igc_sw_init(struct igc_adapter *adapter) struct pci_dev *pdev = adapter->pdev; struct igc_hw *hw = &adapter->hw; - int size = sizeof(struct igc_mac_addr) * hw->mac.rar_entry_count; - pci_read_config_word(pdev, PCI_COMMAND, &hw->bus.pci_cmd_word); /* set default ring sizes */ @@ -3218,20 +3561,19 @@ static int igc_sw_init(struct igc_adapter *adapter) VLAN_HLEN; adapter->min_frame_size = ETH_ZLEN + ETH_FCS_LEN; - spin_lock_init(&adapter->nfc_lock); + mutex_init(&adapter->nfc_rule_lock); + INIT_LIST_HEAD(&adapter->nfc_rule_list); + adapter->nfc_rule_count = 0; + spin_lock_init(&adapter->stats64_lock); /* Assume MSI-X interrupts, will be checked during IRQ allocation */ adapter->flags |= IGC_FLAG_HAS_MSIX; - adapter->mac_table = kzalloc(size, GFP_ATOMIC); - if (!adapter->mac_table) - return -ENOMEM; - igc_init_queue_configuration(adapter); /* This call may decrease the number of queues */ if (igc_init_interrupt_scheme(adapter, true)) { - dev_err(&pdev->dev, "Unable to allocate memory for queues\n"); + netdev_err(netdev, "Unable to allocate memory for queues\n"); return -ENOMEM; } @@ -3359,8 +3701,6 @@ void igc_update_stats(struct igc_adapter *adapter) adapter->stats.prc511 += rd32(IGC_PRC511); adapter->stats.prc1023 += rd32(IGC_PRC1023); adapter->stats.prc1522 += rd32(IGC_PRC1522); - adapter->stats.symerrs += rd32(IGC_SYMERRS); - adapter->stats.sec += rd32(IGC_SEC); mpc = rd32(IGC_MPC); adapter->stats.mpc += mpc; @@ -3399,6 +3739,7 @@ void igc_update_stats(struct igc_adapter *adapter) adapter->stats.tpt += rd32(IGC_TPT); adapter->stats.colc += rd32(IGC_COLC); + adapter->stats.colc += rd32(IGC_RERC); adapter->stats.algnerrc += rd32(IGC_ALGNERRC); @@ -3449,21 +3790,6 @@ void igc_update_stats(struct igc_adapter *adapter) adapter->stats.mgpdc += rd32(IGC_MGTPDC); } -static void igc_nfc_filter_exit(struct igc_adapter *adapter) -{ - struct igc_nfc_filter *rule; - - spin_lock(&adapter->nfc_lock); - - hlist_for_each_entry(rule, &adapter->nfc_filter_list, nfc_node) - igc_erase_filter(adapter, rule); - - hlist_for_each_entry(rule, &adapter->cls_flower_list, nfc_node) - igc_erase_filter(adapter, rule); - - spin_unlock(&adapter->nfc_lock); -} - /** * igc_down - Close the interface * @adapter: board private structure @@ -3482,8 +3808,6 @@ void igc_down(struct igc_adapter *adapter) wr32(IGC_RCTL, rctl & ~IGC_RCTL_EN); /* flush and sleep below */ - igc_nfc_filter_exit(adapter); - /* set trans_start so we don't get spurious watchdogs during reset */ netif_trans_update(netdev); @@ -3577,8 +3901,7 @@ static int igc_change_mtu(struct net_device *netdev, int new_mtu) if (netif_running(netdev)) igc_down(adapter); - netdev_dbg(netdev, "changing MTU from %d to %d\n", - netdev->mtu, new_mtu); + netdev_dbg(netdev, "changing MTU from %d to %d\n", netdev->mtu, new_mtu); netdev->mtu = new_mtu; if (netif_running(netdev)) @@ -3633,20 +3956,8 @@ static int igc_set_features(struct net_device *netdev, if (!(changed & (NETIF_F_RXALL | NETIF_F_NTUPLE))) return 0; - if (!(features & NETIF_F_NTUPLE)) { - struct hlist_node *node2; - struct igc_nfc_filter *rule; - - spin_lock(&adapter->nfc_lock); - hlist_for_each_entry_safe(rule, node2, - &adapter->nfc_filter_list, nfc_node) { - igc_erase_filter(adapter, rule); - hlist_del(&rule->nfc_node); - kfree(rule); - } - spin_unlock(&adapter->nfc_lock); - adapter->nfc_filter_count = 0; - } + if (!(features & NETIF_F_NTUPLE)) + igc_flush_nfc_rules(adapter); netdev->features = features; @@ -3689,106 +4000,6 @@ igc_features_check(struct sk_buff *skb, struct net_device *dev, return features; } -/* Add a MAC filter for 'addr' directing matching traffic to 'queue', - * 'flags' is used to indicate what kind of match is made, match is by - * default for the destination address, if matching by source address - * is desired the flag IGC_MAC_STATE_SRC_ADDR can be used. - */ -static int igc_add_mac_filter_flags(struct igc_adapter *adapter, - const u8 *addr, const u8 queue, - const u8 flags) -{ - struct igc_hw *hw = &adapter->hw; - int rar_entries = hw->mac.rar_entry_count; - int i; - - if (is_zero_ether_addr(addr)) - return -EINVAL; - - /* Search for the first empty entry in the MAC table. - * Do not touch entries at the end of the table reserved for the VF MAC - * addresses. - */ - for (i = 0; i < rar_entries; i++) { - if (!igc_mac_entry_can_be_used(&adapter->mac_table[i], - addr, flags)) - continue; - - ether_addr_copy(adapter->mac_table[i].addr, addr); - adapter->mac_table[i].queue = queue; - adapter->mac_table[i].state |= IGC_MAC_STATE_IN_USE | flags; - - igc_rar_set_index(adapter, i); - return i; - } - - return -ENOSPC; -} - -int igc_add_mac_steering_filter(struct igc_adapter *adapter, - const u8 *addr, u8 queue, u8 flags) -{ - return igc_add_mac_filter_flags(adapter, addr, queue, - IGC_MAC_STATE_QUEUE_STEERING | flags); -} - -/* Remove a MAC filter for 'addr' directing matching traffic to - * 'queue', 'flags' is used to indicate what kind of match need to be - * removed, match is by default for the destination address, if - * matching by source address is to be removed the flag - * IGC_MAC_STATE_SRC_ADDR can be used. - */ -static int igc_del_mac_filter_flags(struct igc_adapter *adapter, - const u8 *addr, const u8 queue, - const u8 flags) -{ - struct igc_hw *hw = &adapter->hw; - int rar_entries = hw->mac.rar_entry_count; - int i; - - if (is_zero_ether_addr(addr)) - return -EINVAL; - - /* Search for matching entry in the MAC table based on given address - * and queue. Do not touch entries at the end of the table reserved - * for the VF MAC addresses. - */ - for (i = 0; i < rar_entries; i++) { - if (!(adapter->mac_table[i].state & IGC_MAC_STATE_IN_USE)) - continue; - if ((adapter->mac_table[i].state & flags) != flags) - continue; - if (adapter->mac_table[i].queue != queue) - continue; - if (!ether_addr_equal(adapter->mac_table[i].addr, addr)) - continue; - - /* When a filter for the default address is "deleted", - * we return it to its initial configuration - */ - if (adapter->mac_table[i].state & IGC_MAC_STATE_DEFAULT) { - adapter->mac_table[i].state = - IGC_MAC_STATE_DEFAULT | IGC_MAC_STATE_IN_USE; - } else { - adapter->mac_table[i].state = 0; - adapter->mac_table[i].queue = 0; - memset(adapter->mac_table[i].addr, 0, ETH_ALEN); - } - - igc_rar_set_index(adapter, i); - return 0; - } - - return -ENOENT; -} - -int igc_del_mac_steering_filter(struct igc_adapter *adapter, - const u8 *addr, u8 queue, u8 flags) -{ - return igc_del_mac_filter_flags(adapter, addr, queue, - IGC_MAC_STATE_QUEUE_STEERING | flags); -} - static void igc_tsync_interrupt(struct igc_adapter *adapter) { struct igc_hw *hw = &adapter->hw; @@ -4009,7 +4220,6 @@ static void igc_watchdog_task(struct work_struct *work) struct igc_hw *hw = &adapter->hw; struct igc_phy_info *phy = &hw->phy; u16 phy_data, retry_count = 20; - u32 connsw; u32 link; int i; @@ -4022,14 +4232,6 @@ static void igc_watchdog_task(struct work_struct *work) link = false; } - /* Force link down if we have fiber to swap to */ - if (adapter->flags & IGC_FLAG_MAS_ENABLE) { - if (hw->phy.media_type == igc_media_type_copper) { - connsw = rd32(IGC_CONNSW); - if (!(connsw & IGC_CONNSW_AUTOSENSE_EN)) - link = 0; - } - } if (link) { /* Cancel scheduled suspend requests. */ pm_runtime_resume(netdev->dev.parent); @@ -4044,8 +4246,7 @@ static void igc_watchdog_task(struct work_struct *work) ctrl = rd32(IGC_CTRL); /* Link status message must follow this format */ netdev_info(netdev, - "igc: %s NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n", - netdev->name, + "NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n", adapter->link_speed, adapter->link_duplex == FULL_DUPLEX ? "Full" : "Half", @@ -4083,10 +4284,10 @@ retry_read_status: retry_count--; goto retry_read_status; } else if (!retry_count) { - dev_err(&adapter->pdev->dev, "exceed max 2 second\n"); + netdev_err(netdev, "exceed max 2 second\n"); } } else { - dev_err(&adapter->pdev->dev, "read 1000Base-T Status Reg\n"); + netdev_err(netdev, "read 1000Base-T Status Reg\n"); } no_wait: netif_carrier_on(netdev); @@ -4102,8 +4303,7 @@ no_wait: adapter->link_duplex = 0; /* Links status message must follow this format */ - netdev_info(netdev, "igc: %s NIC Link is Down\n", - netdev->name); + netdev_info(netdev, "NIC Link is Down\n"); netif_carrier_off(netdev); /* link state has changed, schedule phy info update */ @@ -4321,8 +4521,7 @@ static int igc_request_irq(struct igc_adapter *adapter) netdev->name, adapter); if (err) - dev_err(&pdev->dev, "Error %d getting interrupt\n", - err); + netdev_err(netdev, "Error %d getting interrupt\n", err); request_done: return err; @@ -4424,7 +4623,7 @@ err_setup_tx: return err; } -static int igc_open(struct net_device *netdev) +int igc_open(struct net_device *netdev) { return __igc_open(netdev, false); } @@ -4466,7 +4665,7 @@ static int __igc_close(struct net_device *netdev, bool suspending) return 0; } -static int igc_close(struct net_device *netdev) +int igc_close(struct net_device *netdev) { if (netif_device_present(netdev) || netdev->dismantle) return __igc_close(netdev, false); @@ -4491,6 +4690,158 @@ static int igc_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd) } } +static int igc_save_launchtime_params(struct igc_adapter *adapter, int queue, + bool enable) +{ + struct igc_ring *ring; + int i; + + if (queue < 0 || queue >= adapter->num_tx_queues) + return -EINVAL; + + ring = adapter->tx_ring[queue]; + ring->launchtime_enable = enable; + + if (adapter->base_time) + return 0; + + adapter->cycle_time = NSEC_PER_SEC; + + for (i = 0; i < adapter->num_tx_queues; i++) { + ring = adapter->tx_ring[i]; + ring->start_time = 0; + ring->end_time = NSEC_PER_SEC; + } + + return 0; +} + +static bool validate_schedule(const struct tc_taprio_qopt_offload *qopt) +{ + int queue_uses[IGC_MAX_TX_QUEUES] = { }; + size_t n; + + if (qopt->cycle_time_extension) + return false; + + for (n = 0; n < qopt->num_entries; n++) { + const struct tc_taprio_sched_entry *e; + int i; + + e = &qopt->entries[n]; + + /* i225 only supports "global" frame preemption + * settings. + */ + if (e->command != TC_TAPRIO_CMD_SET_GATES) + return false; + + for (i = 0; i < IGC_MAX_TX_QUEUES; i++) { + if (e->gate_mask & BIT(i)) + queue_uses[i]++; + + if (queue_uses[i] > 1) + return false; + } + } + + return true; +} + +static int igc_tsn_enable_launchtime(struct igc_adapter *adapter, + struct tc_etf_qopt_offload *qopt) +{ + struct igc_hw *hw = &adapter->hw; + int err; + + if (hw->mac.type != igc_i225) + return -EOPNOTSUPP; + + err = igc_save_launchtime_params(adapter, qopt->queue, qopt->enable); + if (err) + return err; + + return igc_tsn_offload_apply(adapter); +} + +static int igc_save_qbv_schedule(struct igc_adapter *adapter, + struct tc_taprio_qopt_offload *qopt) +{ + u32 start_time = 0, end_time = 0; + size_t n; + + if (!qopt->enable) { + adapter->base_time = 0; + return 0; + } + + if (adapter->base_time) + return -EALREADY; + + if (!validate_schedule(qopt)) + return -EINVAL; + + adapter->cycle_time = qopt->cycle_time; + adapter->base_time = qopt->base_time; + + /* FIXME: be a little smarter about cases when the gate for a + * queue stays open for more than one entry. + */ + for (n = 0; n < qopt->num_entries; n++) { + struct tc_taprio_sched_entry *e = &qopt->entries[n]; + int i; + + end_time += e->interval; + + for (i = 0; i < IGC_MAX_TX_QUEUES; i++) { + struct igc_ring *ring = adapter->tx_ring[i]; + + if (!(e->gate_mask & BIT(i))) + continue; + + ring->start_time = start_time; + ring->end_time = end_time; + } + + start_time += e->interval; + } + + return 0; +} + +static int igc_tsn_enable_qbv_scheduling(struct igc_adapter *adapter, + struct tc_taprio_qopt_offload *qopt) +{ + struct igc_hw *hw = &adapter->hw; + int err; + + if (hw->mac.type != igc_i225) + return -EOPNOTSUPP; + + err = igc_save_qbv_schedule(adapter, qopt); + if (err) + return err; + + return igc_tsn_offload_apply(adapter); +} + +static int igc_setup_tc(struct net_device *dev, enum tc_setup_type type, + void *type_data) +{ + struct igc_adapter *adapter = netdev_priv(dev); + + switch (type) { + case TC_SETUP_QDISC_TAPRIO: + return igc_tsn_enable_qbv_scheduling(adapter, type_data); + + case TC_SETUP_QDISC_ETF: + return igc_tsn_enable_launchtime(adapter, type_data); + + default: + return -EOPNOTSUPP; + } +} + static const struct net_device_ops igc_netdev_ops = { .ndo_open = igc_open, .ndo_stop = igc_close, @@ -4503,6 +4854,7 @@ static const struct net_device_ops igc_netdev_ops = { .ndo_set_features = igc_set_features, .ndo_features_check = igc_features_check, .ndo_do_ioctl = igc_ioctl, + .ndo_setup_tc = igc_setup_tc, }; /* PCIe configuration access */ @@ -4550,9 +4902,6 @@ u32 igc_rd32(struct igc_hw *hw, u32 reg) u8 __iomem *hw_addr = READ_ONCE(hw->hw_addr); u32 value = 0; - if (IGC_REMOVED(hw_addr)) - return ~value; - value = readl(&hw_addr[reg]); /* reads should not return all F's */ @@ -4571,7 +4920,6 @@ u32 igc_rd32(struct igc_hw *hw, u32 reg) int igc_set_spd_dplx(struct igc_adapter *adapter, u32 spd, u8 dplx) { - struct pci_dev *pdev = adapter->pdev; struct igc_mac_info *mac = &adapter->hw.mac; mac->autoneg = 0; @@ -4616,7 +4964,7 @@ int igc_set_spd_dplx(struct igc_adapter *adapter, u32 spd, u8 dplx) return 0; err_inval: - dev_err(&pdev->dev, "Unsupported Speed/Duplex configuration\n"); + netdev_err(adapter->netdev, "Unsupported Speed/Duplex configuration\n"); return -EINVAL; } @@ -4697,7 +5045,7 @@ static int igc_probe(struct pci_dev *pdev, hw->hw_addr = adapter->io_addr; netdev->netdev_ops = &igc_netdev_ops; - igc_set_ethtool_ops(netdev); + igc_ethtool_set_ops(netdev); netdev->watchdog_timeo = 5 * HZ; netdev->mem_start = pci_resource_start(pdev, 0); @@ -4723,9 +5071,21 @@ static int igc_probe(struct pci_dev *pdev, netdev->features |= NETIF_F_SG; netdev->features |= NETIF_F_TSO; netdev->features |= NETIF_F_TSO6; + netdev->features |= NETIF_F_TSO_ECN; netdev->features |= NETIF_F_RXCSUM; netdev->features |= NETIF_F_HW_CSUM; netdev->features |= NETIF_F_SCTP_CRC; + netdev->features |= NETIF_F_HW_TC; + +#define IGC_GSO_PARTIAL_FEATURES (NETIF_F_GSO_GRE | \ + NETIF_F_GSO_GRE_CSUM | \ + NETIF_F_GSO_IPXIP4 | \ + NETIF_F_GSO_IPXIP6 | \ + NETIF_F_GSO_UDP_TUNNEL | \ + NETIF_F_GSO_UDP_TUNNEL_CSUM) + + netdev->gso_partial_features = IGC_GSO_PARTIAL_FEATURES; + netdev->features |= NETIF_F_GSO_PARTIAL | IGC_GSO_PARTIAL_FEATURES; /* setup the private structure */ err = igc_sw_init(adapter); @@ -4750,8 +5110,7 @@ static int igc_probe(struct pci_dev *pdev, if (igc_get_flash_presence_i225(hw)) { if (hw->nvm.ops.validate(hw) < 0) { - dev_err(&pdev->dev, - "The NVM Checksum Is Not Valid\n"); + dev_err(&pdev->dev, "The NVM Checksum Is Not Valid\n"); err = -EIO; goto err_eeprom; } @@ -4825,7 +5184,7 @@ static int igc_probe(struct pci_dev *pdev, pcie_print_link_status(pdev); netdev_info(netdev, "MAC: %pM\n", netdev->dev_addr); - dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NEVER_SKIP); + dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NO_DIRECT_COMPLETE); pm_runtime_put_noidle(&pdev->dev); @@ -4865,6 +5224,8 @@ static void igc_remove(struct pci_dev *pdev) pm_runtime_get_noresume(&pdev->dev); + igc_flush_nfc_rules(adapter); + igc_ptp_stop(adapter); set_bit(__IGC_DOWN, &adapter->state); @@ -4885,7 +5246,6 @@ static void igc_remove(struct pci_dev *pdev) pci_iounmap(pdev, adapter->io_addr); pci_release_mem_regions(pdev); - kfree(adapter->mac_table); free_netdev(netdev); pci_disable_pcie_error_reporting(pdev); @@ -5014,8 +5374,7 @@ static int __maybe_unused igc_resume(struct device *dev) return -ENODEV; err = pci_enable_device_mem(pdev); if (err) { - dev_err(&pdev->dev, - "igc: Cannot enable PCI device from suspend\n"); + netdev_err(netdev, "Cannot enable PCI device from suspend\n"); return err; } pci_set_master(pdev); @@ -5024,7 +5383,7 @@ static int __maybe_unused igc_resume(struct device *dev) pci_enable_wake(pdev, PCI_D3cold, 0); if (igc_init_interrupt_scheme(adapter, true)) { - dev_err(&pdev->dev, "Unable to allocate memory for queues\n"); + netdev_err(netdev, "Unable to allocate memory for queues\n"); return -ENOMEM; } @@ -5128,8 +5487,7 @@ static pci_ers_result_t igc_io_slot_reset(struct pci_dev *pdev) pci_ers_result_t result; if (pci_enable_device_mem(pdev)) { - dev_err(&pdev->dev, - "Could not re-enable PCI device after reset.\n"); + netdev_err(netdev, "Could not re-enable PCI device after reset\n"); result = PCI_ERS_RESULT_DISCONNECT; } else { pci_set_master(pdev); @@ -5168,7 +5526,7 @@ static void igc_io_resume(struct pci_dev *pdev) rtnl_lock(); if (netif_running(netdev)) { if (igc_open(netdev)) { - dev_err(&pdev->dev, "igc_open failed after reset\n"); + netdev_err(netdev, "igc_open failed after reset\n"); return; } } @@ -5215,7 +5573,6 @@ static struct pci_driver igc_driver = { int igc_reinit_queues(struct igc_adapter *adapter) { struct net_device *netdev = adapter->netdev; - struct pci_dev *pdev = adapter->pdev; int err = 0; if (netif_running(netdev)) @@ -5224,7 +5581,7 @@ int igc_reinit_queues(struct igc_adapter *adapter) igc_reset_interrupt_capability(adapter); if (igc_init_interrupt_scheme(adapter, true)) { - dev_err(&pdev->dev, "Unable to allocate memory for queues\n"); + netdev_err(netdev, "Unable to allocate memory for queues\n"); return -ENOMEM; } diff --git a/drivers/net/ethernet/intel/igc/igc_ptp.c b/drivers/net/ethernet/intel/igc/igc_ptp.c index f99c514ad0f4..0d746f8588c8 100644 --- a/drivers/net/ethernet/intel/igc/igc_ptp.c +++ b/drivers/net/ethernet/intel/igc/igc_ptp.c @@ -305,7 +305,6 @@ static int igc_ptp_set_timestamp_mode(struct igc_adapter *adapter, struct igc_hw *hw = &adapter->hw; u32 tsync_rx_cfg = 0; bool is_l4 = false; - bool is_l2 = false; u32 regval; /* reserved for future extensions */ @@ -346,7 +345,6 @@ static int igc_ptp_set_timestamp_mode(struct igc_adapter *adapter, case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ: tsync_rx_ctl |= IGC_TSYNCRXCTL_TYPE_EVENT_V2; config->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT; - is_l2 = true; is_l4 = true; break; case HWTSTAMP_FILTER_PTP_V1_L4_EVENT: @@ -370,7 +368,6 @@ static int igc_ptp_set_timestamp_mode(struct igc_adapter *adapter, tsync_rx_ctl |= IGC_TSYNCRXCTL_TYPE_ALL; tsync_rx_ctl |= IGC_TSYNCRXCTL_RXSYNSIG; config->rx_filter = HWTSTAMP_FILTER_ALL; - is_l2 = true; is_l4 = true; if (hw->mac.type == igc_i225) { @@ -405,15 +402,6 @@ static int igc_ptp_set_timestamp_mode(struct igc_adapter *adapter, /* define which PTP packets are time stamped */ wr32(IGC_TSYNCRXCFG, tsync_rx_cfg); - /* define ethertype filter for timestamped packets */ - if (is_l2) - wr32(IGC_ETQF(3), - (IGC_ETQF_FILTER_ENABLE | /* enable filter */ - IGC_ETQF_1588 | /* enable timestamping */ - ETH_P_1588)); /* 1588 eth protocol type */ - else - wr32(IGC_ETQF(3), 0); - /* L4 Queue Filter[3]: filter by destination port and protocol */ if (is_l4) { u32 ftqf = (IPPROTO_UDP /* UDP */ @@ -466,7 +454,7 @@ void igc_ptp_tx_hang(struct igc_adapter *adapter) * interrupt */ rd32(IGC_TXSTMPH); - dev_warn(&adapter->pdev->dev, "clearing Tx timestamp hang\n"); + netdev_warn(adapter->netdev, "Clearing Tx timestamp hang\n"); } } @@ -529,7 +517,7 @@ static void igc_ptp_tx_work(struct work_struct *work) * interrupt */ rd32(IGC_TXSTMPH); - dev_warn(&adapter->pdev->dev, "clearing Tx timestamp hang\n"); + netdev_warn(adapter->netdev, "Clearing Tx timestamp hang\n"); return; } @@ -626,10 +614,9 @@ void igc_ptp_init(struct igc_adapter *adapter) &adapter->pdev->dev); if (IS_ERR(adapter->ptp_clock)) { adapter->ptp_clock = NULL; - dev_err(&adapter->pdev->dev, "ptp_clock_register failed\n"); + netdev_err(netdev, "ptp_clock_register failed\n"); } else if (adapter->ptp_clock) { - dev_info(&adapter->pdev->dev, "added PHC on %s\n", - adapter->netdev->name); + netdev_info(netdev, "PHC added\n"); adapter->ptp_flags |= IGC_PTP_ENABLED; } } @@ -666,8 +653,7 @@ void igc_ptp_stop(struct igc_adapter *adapter) if (adapter->ptp_clock) { ptp_clock_unregister(adapter->ptp_clock); - dev_info(&adapter->pdev->dev, "removed PHC on %s\n", - adapter->netdev->name); + netdev_info(adapter->netdev, "PHC removed\n"); adapter->ptp_flags &= ~IGC_PTP_ENABLED; } } diff --git a/drivers/net/ethernet/intel/igc/igc_regs.h b/drivers/net/ethernet/intel/igc/igc_regs.h index d4af53a80f11..232e82dec62e 100644 --- a/drivers/net/ethernet/intel/igc/igc_regs.h +++ b/drivers/net/ethernet/intel/igc/igc_regs.h @@ -17,11 +17,6 @@ /* Internal Packet Buffer Size Registers */ #define IGC_RXPBS 0x02404 /* Rx Packet Buffer Size - RW */ #define IGC_TXPBS 0x03404 /* Tx Packet Buffer Size - RW */ -#define IGC_TDFH 0x03410 /* Tx Data FIFO Head - RW */ -#define IGC_TDFT 0x03418 /* Tx Data FIFO Tail - RW */ -#define IGC_TDFHS 0x03420 /* Tx Data FIFO Head Saved - RW */ -#define IGC_TDFTS 0x03428 /* Tx Data FIFO Tail Saved - RW */ -#define IGC_TDFPC 0x03430 /* Tx Data FIFO Packet Count - RW */ /* NVM Register Descriptions */ #define IGC_EERD 0x12014 /* EEprom mode read - RW */ @@ -35,10 +30,6 @@ #define IGC_FCRTL 0x02160 /* FC Receive Threshold Low - RW */ #define IGC_FCRTH 0x02168 /* FC Receive Threshold High - RW */ #define IGC_FCRTV 0x02460 /* FC Refresh Timer Value - RW */ -#define IGC_FCSTS 0x02464 /* FC Status - RO */ - -/* PCIe Register Description */ -#define IGC_GCR 0x05B00 /* PCIe control- RW */ /* Semaphore registers */ #define IGC_SW_FW_SYNC 0x05B5C /* SW-FW Synchronization - RW */ @@ -49,6 +40,7 @@ #define IGC_FACTPS 0x05B30 /* Interrupt Register Description */ +#define IGC_EICR 0x01580 /* Ext. Interrupt Cause read - W0 */ #define IGC_EICS 0x01520 /* Ext. Interrupt Cause Set - W0 */ #define IGC_EIMS 0x01524 /* Ext. Interrupt Mask Set/Read - RW */ #define IGC_EIMC 0x01528 /* Ext. Interrupt Mask Clear - WO */ @@ -76,13 +68,6 @@ #define IGC_ICRXDMTC 0x04120 /* Rx Descriptor Min Threshold Count */ #define IGC_ICRXOC 0x04124 /* Receiver Overrun Count */ -#define IGC_CBTMPC 0x0402C /* Circuit Breaker TX Packet Count */ -#define IGC_HTDPMC 0x0403C /* Host Transmit Discarded Packets */ -#define IGC_CBRMPC 0x040FC /* Circuit Breaker RX Packet Count */ -#define IGC_RPTHC 0x04104 /* Rx Packets To Host */ -#define IGC_HGPTC 0x04118 /* Host Good Packets TX Count */ -#define IGC_HTCBDPC 0x04124 /* Host TX Circ.Breaker Drop Count */ - /* MSI-X Table Register Descriptions */ #define IGC_PBACL 0x05B68 /* MSIx PBA Clear - R/W 1 to clear */ @@ -119,10 +104,11 @@ #define IGC_RLPML 0x05004 /* Rx Long Packet Max Length */ #define IGC_RFCTL 0x05008 /* Receive Filter Control*/ #define IGC_MTA 0x05200 /* Multicast Table Array - RW Array */ +#define IGC_RA 0x05400 /* Receive Address - RW Array */ #define IGC_UTA 0x0A000 /* Unicast Table Array - RW */ #define IGC_RAL(_n) (0x05400 + ((_n) * 0x08)) #define IGC_RAH(_n) (0x05404 + ((_n) * 0x08)) -#define IGC_VLAPQF 0x055B0 /* VLAN Priority Queue Filter VLAPQF */ +#define IGC_VLANPQF 0x055B0 /* VLAN Priority Queue Filter - RW */ /* Transmit Register Descriptions */ #define IGC_TCTL 0x00400 /* Tx Control - RW */ @@ -138,13 +124,9 @@ #define IGC_MMDAC 13 /* MMD Access Control */ #define IGC_MMDAAD 14 /* MMD Access Address/Data */ -/* Good transmitted packets counter registers */ -#define IGC_PQGPTC(_n) (0x010014 + (0x100 * (_n))) - /* Statistics Register Descriptions */ #define IGC_CRCERRS 0x04000 /* CRC Error Count - R/clr */ #define IGC_ALGNERRC 0x04004 /* Alignment Error Count - R/clr */ -#define IGC_SYMERRS 0x04008 /* Symbol Error Count - R/clr */ #define IGC_RXERRC 0x0400C /* Receive Error Count - R/clr */ #define IGC_MPC 0x04010 /* Missed Packet Count - R/clr */ #define IGC_SCC 0x04014 /* Single Collision Count - R/clr */ @@ -152,10 +134,10 @@ #define IGC_MCC 0x0401C /* Multiple Collision Count - R/clr */ #define IGC_LATECOL 0x04020 /* Late Collision Count - R/clr */ #define IGC_COLC 0x04028 /* Collision Count - R/clr */ +#define IGC_RERC 0x0402C /* Receive Error Count - R/clr */ #define IGC_DC 0x04030 /* Defer Count - R/clr */ #define IGC_TNCRS 0x04034 /* Tx-No CRS - R/clr */ -#define IGC_SEC 0x04038 /* Sequence Error Count - R/clr */ -#define IGC_CEXTERR 0x0403C /* Carrier Extension Error Count - R/clr */ +#define IGC_HTDPMC 0x0403C /* Host Transmit Discarded by MAC - R/clr */ #define IGC_RLEC 0x04040 /* Receive Length Error Count - R/clr */ #define IGC_XONRXC 0x04048 /* XON Rx Count - R/clr */ #define IGC_XONTXC 0x0404C /* XON Tx Count - R/clr */ @@ -213,7 +195,6 @@ #define IGC_HGOTCL 0x04130 /* Host Good Octets Transmit Count Low */ #define IGC_HGOTCH 0x04134 /* Host Good Octets Transmit Count High */ #define IGC_LENERRS 0x04138 /* Length Errors Count */ -#define IGC_HRMPC 0x0A018 /* Header Redirection Missed Packet Count */ /* Time sync registers */ #define IGC_TSICR 0x0B66C /* Time Sync Interrupt Cause */ @@ -229,7 +210,17 @@ #define IGC_FTQF(_n) (0x059E0 + (4 * (_n))) /* 5-tuple Queue Fltr */ -#define IGC_RXPBS 0x02404 /* Rx Packet Buffer Size - RW */ +/* Transmit Scheduling Registers */ +#define IGC_TQAVCTRL 0x3570 +#define IGC_TXQCTL(_n) (0x3344 + 0x4 * (_n)) +#define IGC_BASET_L 0x3314 +#define IGC_BASET_H 0x3318 +#define IGC_QBVCYCLET 0x331C +#define IGC_QBVCYCLET_S 0x3320 + +#define IGC_STQT(_n) (0x3324 + 0x4 * (_n)) +#define IGC_ENDQT(_n) (0x3334 + 0x4 * (_n)) +#define IGC_DTXMXPKTSZ 0x355C /* System Time Registers */ #define IGC_SYSTIML 0x0B600 /* System time register Low - RO */ @@ -265,8 +256,7 @@ u32 igc_rd32(struct igc_hw *hw, u32 reg); #define wr32(reg, val) \ do { \ u8 __iomem *hw_addr = READ_ONCE((hw)->hw_addr); \ - if (!IGC_REMOVED(hw_addr)) \ - writel((val), &hw_addr[(reg)]); \ + writel((val), &hw_addr[(reg)]); \ } while (0) #define rd32(reg) (igc_rd32(hw, reg)) diff --git a/drivers/net/ethernet/intel/igc/igc_tsn.c b/drivers/net/ethernet/intel/igc/igc_tsn.c new file mode 100644 index 000000000000..174103c4bea6 --- /dev/null +++ b/drivers/net/ethernet/intel/igc/igc_tsn.c @@ -0,0 +1,157 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2019 Intel Corporation */ + +#include "igc.h" +#include "igc_tsn.h" + +static bool is_any_launchtime(struct igc_adapter *adapter) +{ + int i; + + for (i = 0; i < adapter->num_tx_queues; i++) { + struct igc_ring *ring = adapter->tx_ring[i]; + + if (ring->launchtime_enable) + return true; + } + + return false; +} + +/* Returns the TSN specific registers to their default values after + * TSN offloading is disabled. + */ +static int igc_tsn_disable_offload(struct igc_adapter *adapter) +{ + struct igc_hw *hw = &adapter->hw; + u32 tqavctrl; + int i; + + if (!(adapter->flags & IGC_FLAG_TSN_QBV_ENABLED)) + return 0; + + adapter->cycle_time = 0; + + wr32(IGC_TXPBS, I225_TXPBSIZE_DEFAULT); + wr32(IGC_DTXMXPKTSZ, IGC_DTXMXPKTSZ_DEFAULT); + + tqavctrl = rd32(IGC_TQAVCTRL); + tqavctrl &= ~(IGC_TQAVCTRL_TRANSMIT_MODE_TSN | + IGC_TQAVCTRL_ENHANCED_QAV); + wr32(IGC_TQAVCTRL, tqavctrl); + + for (i = 0; i < adapter->num_tx_queues; i++) { + struct igc_ring *ring = adapter->tx_ring[i]; + + ring->start_time = 0; + ring->end_time = 0; + ring->launchtime_enable = false; + + wr32(IGC_TXQCTL(i), 0); + wr32(IGC_STQT(i), 0); + wr32(IGC_ENDQT(i), NSEC_PER_SEC); + } + + wr32(IGC_QBVCYCLET_S, NSEC_PER_SEC); + wr32(IGC_QBVCYCLET, NSEC_PER_SEC); + + adapter->flags &= ~IGC_FLAG_TSN_QBV_ENABLED; + + return 0; +} + +static int igc_tsn_enable_offload(struct igc_adapter *adapter) +{ + struct igc_hw *hw = &adapter->hw; + u32 tqavctrl, baset_l, baset_h; + u32 sec, nsec, cycle; + ktime_t base_time, systim; + int i; + + if (adapter->flags & IGC_FLAG_TSN_QBV_ENABLED) + return 0; + + cycle = adapter->cycle_time; + base_time = adapter->base_time; + + wr32(IGC_TSAUXC, 0); + wr32(IGC_DTXMXPKTSZ, IGC_DTXMXPKTSZ_TSN); + wr32(IGC_TXPBS, IGC_TXPBSIZE_TSN); + + tqavctrl = rd32(IGC_TQAVCTRL); + tqavctrl |= IGC_TQAVCTRL_TRANSMIT_MODE_TSN | IGC_TQAVCTRL_ENHANCED_QAV; + wr32(IGC_TQAVCTRL, tqavctrl); + + wr32(IGC_QBVCYCLET_S, cycle); + wr32(IGC_QBVCYCLET, cycle); + + for (i = 0; i < adapter->num_tx_queues; i++) { + struct igc_ring *ring = adapter->tx_ring[i]; + u32 txqctl = 0; + + wr32(IGC_STQT(i), ring->start_time); + wr32(IGC_ENDQT(i), ring->end_time); + + if (adapter->base_time) { + /* If we have a base_time we are in "taprio" + * mode and we need to be strict about the + * cycles: only transmit a packet if it can be + * completed during that cycle. + */ + txqctl |= IGC_TXQCTL_STRICT_CYCLE | + IGC_TXQCTL_STRICT_END; + } + + if (ring->launchtime_enable) + txqctl |= IGC_TXQCTL_QUEUE_MODE_LAUNCHT; + + wr32(IGC_TXQCTL(i), txqctl); + } + + nsec = rd32(IGC_SYSTIML); + sec = rd32(IGC_SYSTIMH); + + systim = ktime_set(sec, nsec); + + if (ktime_compare(systim, base_time) > 0) { + s64 n; + + n = div64_s64(ktime_sub_ns(systim, base_time), cycle); + base_time = ktime_add_ns(base_time, (n + 1) * cycle); + } + + baset_h = div_s64_rem(base_time, NSEC_PER_SEC, &baset_l); + + wr32(IGC_BASET_H, baset_h); + wr32(IGC_BASET_L, baset_l); + + adapter->flags |= IGC_FLAG_TSN_QBV_ENABLED; + + return 0; +} + +int igc_tsn_offload_apply(struct igc_adapter *adapter) +{ + bool is_any_enabled = adapter->base_time || is_any_launchtime(adapter); + + if (!(adapter->flags & IGC_FLAG_TSN_QBV_ENABLED) && !is_any_enabled) + return 0; + + if (!is_any_enabled) { + int err = igc_tsn_disable_offload(adapter); + + if (err < 0) + return err; + + /* The BASET registers aren't cleared when writing + * into them, force a reset if the interface is + * running. + */ + if (netif_running(adapter->netdev)) + schedule_work(&adapter->reset_task); + + return 0; + } + + return igc_tsn_enable_offload(adapter); +} diff --git a/drivers/net/ethernet/intel/igc/igc_tsn.h b/drivers/net/ethernet/intel/igc/igc_tsn.h new file mode 100644 index 000000000000..f76bc86ddccd --- /dev/null +++ b/drivers/net/ethernet/intel/igc/igc_tsn.h @@ -0,0 +1,9 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (c) 2020 Intel Corporation */ + +#ifndef _IGC_TSN_H_ +#define _IGC_TSN_H_ + +int igc_tsn_offload_apply(struct igc_adapter *adapter); + +#endif /* _IGC_BASE_H */ diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe.h b/drivers/net/ethernet/intel/ixgbe/ixgbe.h index 2833e4f041ce..5ddfc83a1e46 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe.h @@ -224,17 +224,17 @@ struct ixgbe_tx_buffer { }; struct ixgbe_rx_buffer { - struct sk_buff *skb; - dma_addr_t dma; union { struct { + struct sk_buff *skb; + dma_addr_t dma; struct page *page; __u32 page_offset; __u16 pagecnt_bias; }; struct { - void *addr; - u64 handle; + bool discard; + struct xdp_buff *xdp; }; }; }; @@ -351,7 +351,6 @@ struct ixgbe_ring { }; struct xdp_rxq_info xdp_rxq; struct xdp_umem *xsk_umem; - struct zero_copy_allocator zca; /* ZC allocator anchor */ u16 ring_idx; /* {rx,tx,xdp}_ring back reference idx */ u16 rx_buf_len; } ____cacheline_internodealigned_in_smp; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c index 0bd1294ba517..17357a12cbdc 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_common.c @@ -64,8 +64,7 @@ bool ixgbe_device_supports_autoneg_fc(struct ixgbe_hw *hw) hw->mac.ops.check_link(hw, &speed, &link_up, false); /* if link is down, assume supported */ if (link_up) - supported = speed == IXGBE_LINK_SPEED_1GB_FULL ? - true : false; + supported = speed == IXGBE_LINK_SPEED_1GB_FULL; else supported = true; } @@ -2243,7 +2242,7 @@ s32 ixgbe_fc_enable_generic(struct ixgbe_hw *hw) } /* Configure pause time (2 TCs per register) */ - reg = hw->fc.pause_time * 0x00010001; + reg = hw->fc.pause_time * 0x00010001U; for (i = 0; i < (MAX_TRAFFIC_CLASS / 2); i++) IXGBE_WRITE_REG(hw, IXGBE_FCTTV(i), reg); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c index 718931d951bc..f162b8b8f345 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c @@ -35,7 +35,7 @@ #include <net/tc_act/tc_mirred.h> #include <net/vxlan.h> #include <net/mpls.h> -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> #include <net/xfrm.h> #include "ixgbe.h" @@ -2215,7 +2215,7 @@ static struct sk_buff *ixgbe_run_xdp(struct ixgbe_adapter *adapter, case XDP_PASS: break; case XDP_TX: - xdpf = convert_to_xdp_frame(xdp); + xdpf = xdp_convert_buff_to_frame(xdp); if (unlikely(!xdpf)) { result = IXGBE_XDP_CONSUMED; break; @@ -2244,19 +2244,30 @@ xdp_out: return ERR_PTR(-result); } +static unsigned int ixgbe_rx_frame_truesize(struct ixgbe_ring *rx_ring, + unsigned int size) +{ + unsigned int truesize; + +#if (PAGE_SIZE < 8192) + truesize = ixgbe_rx_pg_size(rx_ring) / 2; /* Must be power-of-2 */ +#else + truesize = ring_uses_build_skb(rx_ring) ? + SKB_DATA_ALIGN(IXGBE_SKB_PAD + size) + + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) : + SKB_DATA_ALIGN(size); +#endif + return truesize; +} + static void ixgbe_rx_buffer_flip(struct ixgbe_ring *rx_ring, struct ixgbe_rx_buffer *rx_buffer, unsigned int size) { + unsigned int truesize = ixgbe_rx_frame_truesize(rx_ring, size); #if (PAGE_SIZE < 8192) - unsigned int truesize = ixgbe_rx_pg_size(rx_ring) / 2; - rx_buffer->page_offset ^= truesize; #else - unsigned int truesize = ring_uses_build_skb(rx_ring) ? - SKB_DATA_ALIGN(IXGBE_SKB_PAD + size) : - SKB_DATA_ALIGN(size); - rx_buffer->page_offset += truesize; #endif } @@ -2290,6 +2301,11 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, xdp.rxq = &rx_ring->xdp_rxq; + /* Frame size depend on rx_ring setup when PAGE_SIZE=4K */ +#if (PAGE_SIZE < 8192) + xdp.frame_sz = ixgbe_rx_frame_truesize(rx_ring, 0); +#endif + while (likely(total_rx_packets < budget)) { union ixgbe_adv_rx_desc *rx_desc; struct ixgbe_rx_buffer *rx_buffer; @@ -2323,7 +2339,10 @@ static int ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector, xdp.data_hard_start = xdp.data - ixgbe_rx_offset(rx_ring); xdp.data_end = xdp.data + size; - +#if (PAGE_SIZE > 4096) + /* At larger PAGE_SIZE, frame_sz depend on len size */ + xdp.frame_sz = ixgbe_rx_frame_truesize(rx_ring, size); +#endif skb = ixgbe_run_xdp(adapter, rx_ring, &xdp); } @@ -2954,35 +2973,6 @@ static inline void ixgbe_irq_enable_queues(struct ixgbe_adapter *adapter, /* skip the flush */ } -static inline void ixgbe_irq_disable_queues(struct ixgbe_adapter *adapter, - u64 qmask) -{ - u32 mask; - struct ixgbe_hw *hw = &adapter->hw; - - switch (hw->mac.type) { - case ixgbe_mac_82598EB: - mask = (IXGBE_EIMS_RTX_QUEUE & qmask); - IXGBE_WRITE_REG(hw, IXGBE_EIMC, mask); - break; - case ixgbe_mac_82599EB: - case ixgbe_mac_X540: - case ixgbe_mac_X550: - case ixgbe_mac_X550EM_x: - case ixgbe_mac_x550em_a: - mask = (qmask & 0xFFFFFFFF); - if (mask) - IXGBE_WRITE_REG(hw, IXGBE_EIMC_EX(0), mask); - mask = (qmask >> 32); - if (mask) - IXGBE_WRITE_REG(hw, IXGBE_EIMC_EX(1), mask); - break; - default: - break; - } - /* skip the flush */ -} - /** * ixgbe_irq_enable - Enable default interrupt generation settings * @adapter: board private structure @@ -3726,8 +3716,7 @@ static void ixgbe_configure_srrctl(struct ixgbe_adapter *adapter, /* configure the packet buffer length */ if (rx_ring->xsk_umem) { - u32 xsk_buf_len = rx_ring->xsk_umem->chunk_size_nohr - - XDP_PACKET_HEADROOM; + u32 xsk_buf_len = xsk_umem_get_rx_frame_size(rx_ring->xsk_umem); /* If the MAC support setting RXDCTL.RLPML, the * SRRCTL[n].BSIZEPKT is set to PAGE_SIZE and @@ -4074,11 +4063,10 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter, xdp_rxq_info_unreg_mem_model(&ring->xdp_rxq); ring->xsk_umem = ixgbe_xsk_umem(adapter, ring); if (ring->xsk_umem) { - ring->zca.free = ixgbe_zca_free; WARN_ON(xdp_rxq_info_reg_mem_model(&ring->xdp_rxq, - MEM_TYPE_ZERO_COPY, - &ring->zca)); - + MEM_TYPE_XSK_BUFF_POOL, + NULL)); + xsk_buff_set_rxq_info(ring->xsk_umem, &ring->xdp_rxq); } else { WARN_ON(xdp_rxq_info_reg_mem_model(&ring->xdp_rxq, MEM_TYPE_PAGE_SHARED, NULL)); @@ -4134,8 +4122,7 @@ void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter, } if (ring->xsk_umem && hw->mac.type != ixgbe_mac_82599EB) { - u32 xsk_buf_len = ring->xsk_umem->chunk_size_nohr - - XDP_PACKET_HEADROOM; + u32 xsk_buf_len = xsk_umem_get_rx_frame_size(ring->xsk_umem); rxdctl &= ~(IXGBE_RXDCTL_RLPMLMASK | IXGBE_RXDCTL_RLPML_EN); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c index 537dfff585e0..d05a5690e66b 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_sriov.c @@ -102,7 +102,7 @@ static int __ixgbe_enable_sriov(struct ixgbe_adapter *adapter, * indirection table and RSS hash key with PF therefore * we want to disable the querying by default. */ - adapter->vfinfo[i].rss_query_enabled = 0; + adapter->vfinfo[i].rss_query_enabled = false; /* Untrust all VFs */ adapter->vfinfo[i].trusted = false; diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_txrx_common.h b/drivers/net/ethernet/intel/ixgbe/ixgbe_txrx_common.h index 6d01700b46bc..7887ae4aaf4f 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_txrx_common.h +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_txrx_common.h @@ -35,7 +35,7 @@ int ixgbe_xsk_umem_setup(struct ixgbe_adapter *adapter, struct xdp_umem *umem, void ixgbe_zca_free(struct zero_copy_allocator *alloc, unsigned long handle); -void ixgbe_alloc_rx_buffers_zc(struct ixgbe_ring *rx_ring, u16 cleaned_count); +bool ixgbe_alloc_rx_buffers_zc(struct ixgbe_ring *rx_ring, u16 cleaned_count); int ixgbe_clean_rx_irq_zc(struct ixgbe_q_vector *q_vector, struct ixgbe_ring *rx_ring, const int budget); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c index 74b540ebb3dc..be9d2a8da515 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c @@ -2,7 +2,7 @@ /* Copyright(c) 2018 Intel Corporation. */ #include <linux/bpf_trace.h> -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> #include <net/xdp.h> #include "ixgbe.h" @@ -20,54 +20,11 @@ struct xdp_umem *ixgbe_xsk_umem(struct ixgbe_adapter *adapter, return xdp_get_umem_from_qid(adapter->netdev, qid); } -static int ixgbe_xsk_umem_dma_map(struct ixgbe_adapter *adapter, - struct xdp_umem *umem) -{ - struct device *dev = &adapter->pdev->dev; - unsigned int i, j; - dma_addr_t dma; - - for (i = 0; i < umem->npgs; i++) { - dma = dma_map_page_attrs(dev, umem->pgs[i], 0, PAGE_SIZE, - DMA_BIDIRECTIONAL, IXGBE_RX_DMA_ATTR); - if (dma_mapping_error(dev, dma)) - goto out_unmap; - - umem->pages[i].dma = dma; - } - - return 0; - -out_unmap: - for (j = 0; j < i; j++) { - dma_unmap_page_attrs(dev, umem->pages[i].dma, PAGE_SIZE, - DMA_BIDIRECTIONAL, IXGBE_RX_DMA_ATTR); - umem->pages[i].dma = 0; - } - - return -1; -} - -static void ixgbe_xsk_umem_dma_unmap(struct ixgbe_adapter *adapter, - struct xdp_umem *umem) -{ - struct device *dev = &adapter->pdev->dev; - unsigned int i; - - for (i = 0; i < umem->npgs; i++) { - dma_unmap_page_attrs(dev, umem->pages[i].dma, PAGE_SIZE, - DMA_BIDIRECTIONAL, IXGBE_RX_DMA_ATTR); - - umem->pages[i].dma = 0; - } -} - static int ixgbe_xsk_umem_enable(struct ixgbe_adapter *adapter, struct xdp_umem *umem, u16 qid) { struct net_device *netdev = adapter->netdev; - struct xdp_umem_fq_reuse *reuseq; bool if_running; int err; @@ -78,13 +35,7 @@ static int ixgbe_xsk_umem_enable(struct ixgbe_adapter *adapter, qid >= netdev->real_num_tx_queues) return -EINVAL; - reuseq = xsk_reuseq_prepare(adapter->rx_ring[0]->count); - if (!reuseq) - return -ENOMEM; - - xsk_reuseq_free(xsk_reuseq_swap(umem, reuseq)); - - err = ixgbe_xsk_umem_dma_map(adapter, umem); + err = xsk_buff_dma_map(umem, &adapter->pdev->dev, IXGBE_RX_DMA_ATTR); if (err) return err; @@ -124,7 +75,7 @@ static int ixgbe_xsk_umem_disable(struct ixgbe_adapter *adapter, u16 qid) ixgbe_txrx_ring_disable(adapter, qid); clear_bit(qid, adapter->af_xdp_zc_qps); - ixgbe_xsk_umem_dma_unmap(adapter, umem); + xsk_buff_dma_unmap(umem, IXGBE_RX_DMA_ATTR); if (if_running) ixgbe_txrx_ring_enable(adapter, qid); @@ -143,25 +94,20 @@ static int ixgbe_run_xdp_zc(struct ixgbe_adapter *adapter, struct ixgbe_ring *rx_ring, struct xdp_buff *xdp) { - struct xdp_umem *umem = rx_ring->xsk_umem; int err, result = IXGBE_XDP_PASS; struct bpf_prog *xdp_prog; struct xdp_frame *xdpf; - u64 offset; u32 act; rcu_read_lock(); xdp_prog = READ_ONCE(rx_ring->xdp_prog); act = bpf_prog_run_xdp(xdp_prog, xdp); - offset = xdp->data - xdp->data_hard_start; - - xdp->handle = xsk_umem_adjust_offset(umem, xdp->handle, offset); switch (act) { case XDP_PASS: break; case XDP_TX: - xdpf = convert_to_xdp_frame(xdp); + xdpf = xdp_convert_buff_to_frame(xdp); if (unlikely(!xdpf)) { result = IXGBE_XDP_CONSUMED; break; @@ -186,140 +132,16 @@ static int ixgbe_run_xdp_zc(struct ixgbe_adapter *adapter, return result; } -static struct -ixgbe_rx_buffer *ixgbe_get_rx_buffer_zc(struct ixgbe_ring *rx_ring, - unsigned int size) -{ - struct ixgbe_rx_buffer *bi; - - bi = &rx_ring->rx_buffer_info[rx_ring->next_to_clean]; - - /* we are reusing so sync this buffer for CPU use */ - dma_sync_single_range_for_cpu(rx_ring->dev, - bi->dma, 0, - size, - DMA_BIDIRECTIONAL); - - return bi; -} - -static void ixgbe_reuse_rx_buffer_zc(struct ixgbe_ring *rx_ring, - struct ixgbe_rx_buffer *obi) -{ - u16 nta = rx_ring->next_to_alloc; - struct ixgbe_rx_buffer *nbi; - - nbi = &rx_ring->rx_buffer_info[rx_ring->next_to_alloc]; - /* update, and store next to alloc */ - nta++; - rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0; - - /* transfer page from old buffer to new buffer */ - nbi->dma = obi->dma; - nbi->addr = obi->addr; - nbi->handle = obi->handle; - - obi->addr = NULL; - obi->skb = NULL; -} - -void ixgbe_zca_free(struct zero_copy_allocator *alloc, unsigned long handle) -{ - struct ixgbe_rx_buffer *bi; - struct ixgbe_ring *rx_ring; - u64 hr, mask; - u16 nta; - - rx_ring = container_of(alloc, struct ixgbe_ring, zca); - hr = rx_ring->xsk_umem->headroom + XDP_PACKET_HEADROOM; - mask = rx_ring->xsk_umem->chunk_mask; - - nta = rx_ring->next_to_alloc; - bi = rx_ring->rx_buffer_info; - - nta++; - rx_ring->next_to_alloc = (nta < rx_ring->count) ? nta : 0; - - handle &= mask; - - bi->dma = xdp_umem_get_dma(rx_ring->xsk_umem, handle); - bi->dma += hr; - - bi->addr = xdp_umem_get_data(rx_ring->xsk_umem, handle); - bi->addr += hr; - - bi->handle = xsk_umem_adjust_offset(rx_ring->xsk_umem, (u64)handle, - rx_ring->xsk_umem->headroom); -} - -static bool ixgbe_alloc_buffer_zc(struct ixgbe_ring *rx_ring, - struct ixgbe_rx_buffer *bi) -{ - struct xdp_umem *umem = rx_ring->xsk_umem; - void *addr = bi->addr; - u64 handle, hr; - - if (addr) - return true; - - if (!xsk_umem_peek_addr(umem, &handle)) { - rx_ring->rx_stats.alloc_rx_page_failed++; - return false; - } - - hr = umem->headroom + XDP_PACKET_HEADROOM; - - bi->dma = xdp_umem_get_dma(umem, handle); - bi->dma += hr; - - bi->addr = xdp_umem_get_data(umem, handle); - bi->addr += hr; - - bi->handle = xsk_umem_adjust_offset(umem, handle, umem->headroom); - - xsk_umem_release_addr(umem); - return true; -} - -static bool ixgbe_alloc_buffer_slow_zc(struct ixgbe_ring *rx_ring, - struct ixgbe_rx_buffer *bi) -{ - struct xdp_umem *umem = rx_ring->xsk_umem; - u64 handle, hr; - - if (!xsk_umem_peek_addr_rq(umem, &handle)) { - rx_ring->rx_stats.alloc_rx_page_failed++; - return false; - } - - handle &= rx_ring->xsk_umem->chunk_mask; - - hr = umem->headroom + XDP_PACKET_HEADROOM; - - bi->dma = xdp_umem_get_dma(umem, handle); - bi->dma += hr; - - bi->addr = xdp_umem_get_data(umem, handle); - bi->addr += hr; - - bi->handle = xsk_umem_adjust_offset(umem, handle, umem->headroom); - - xsk_umem_release_addr_rq(umem); - return true; -} - -static __always_inline bool -__ixgbe_alloc_rx_buffers_zc(struct ixgbe_ring *rx_ring, u16 cleaned_count, - bool alloc(struct ixgbe_ring *rx_ring, - struct ixgbe_rx_buffer *bi)) +bool ixgbe_alloc_rx_buffers_zc(struct ixgbe_ring *rx_ring, u16 count) { union ixgbe_adv_rx_desc *rx_desc; struct ixgbe_rx_buffer *bi; u16 i = rx_ring->next_to_use; + dma_addr_t dma; bool ok = true; /* nothing to do */ - if (!cleaned_count) + if (!count) return true; rx_desc = IXGBE_RX_DESC(rx_ring, i); @@ -327,21 +149,18 @@ __ixgbe_alloc_rx_buffers_zc(struct ixgbe_ring *rx_ring, u16 cleaned_count, i -= rx_ring->count; do { - if (!alloc(rx_ring, bi)) { + bi->xdp = xsk_buff_alloc(rx_ring->xsk_umem); + if (!bi->xdp) { ok = false; break; } - /* sync the buffer for use by the device */ - dma_sync_single_range_for_device(rx_ring->dev, bi->dma, - bi->page_offset, - rx_ring->rx_buf_len, - DMA_BIDIRECTIONAL); + dma = xsk_buff_xdp_get_dma(bi->xdp); /* Refresh the desc even if buffer_addrs didn't change * because each write-back erases this info. */ - rx_desc->read.pkt_addr = cpu_to_le64(bi->dma); + rx_desc->read.pkt_addr = cpu_to_le64(dma); rx_desc++; bi++; @@ -355,17 +174,14 @@ __ixgbe_alloc_rx_buffers_zc(struct ixgbe_ring *rx_ring, u16 cleaned_count, /* clear the length for the next_to_use descriptor */ rx_desc->wb.upper.length = 0; - cleaned_count--; - } while (cleaned_count); + count--; + } while (count); i += rx_ring->count; if (rx_ring->next_to_use != i) { rx_ring->next_to_use = i; - /* update next to alloc since we have filled the ring */ - rx_ring->next_to_alloc = i; - /* Force memory writes to complete before letting h/w * know there are new descriptors to fetch. (Only * applicable for weak-ordered memory model archs, @@ -378,40 +194,27 @@ __ixgbe_alloc_rx_buffers_zc(struct ixgbe_ring *rx_ring, u16 cleaned_count, return ok; } -void ixgbe_alloc_rx_buffers_zc(struct ixgbe_ring *rx_ring, u16 count) -{ - __ixgbe_alloc_rx_buffers_zc(rx_ring, count, - ixgbe_alloc_buffer_slow_zc); -} - -static bool ixgbe_alloc_rx_buffers_fast_zc(struct ixgbe_ring *rx_ring, - u16 count) -{ - return __ixgbe_alloc_rx_buffers_zc(rx_ring, count, - ixgbe_alloc_buffer_zc); -} - static struct sk_buff *ixgbe_construct_skb_zc(struct ixgbe_ring *rx_ring, - struct ixgbe_rx_buffer *bi, - struct xdp_buff *xdp) + struct ixgbe_rx_buffer *bi) { - unsigned int metasize = xdp->data - xdp->data_meta; - unsigned int datasize = xdp->data_end - xdp->data; + unsigned int metasize = bi->xdp->data - bi->xdp->data_meta; + unsigned int datasize = bi->xdp->data_end - bi->xdp->data; struct sk_buff *skb; /* allocate a skb to store the frags */ skb = __napi_alloc_skb(&rx_ring->q_vector->napi, - xdp->data_end - xdp->data_hard_start, + bi->xdp->data_end - bi->xdp->data_hard_start, GFP_ATOMIC | __GFP_NOWARN); if (unlikely(!skb)) return NULL; - skb_reserve(skb, xdp->data - xdp->data_hard_start); - memcpy(__skb_put(skb, datasize), xdp->data, datasize); + skb_reserve(skb, bi->xdp->data - bi->xdp->data_hard_start); + memcpy(__skb_put(skb, datasize), bi->xdp->data, datasize); if (metasize) skb_metadata_set(skb, metasize); - ixgbe_reuse_rx_buffer_zc(rx_ring, bi); + xsk_buff_free(bi->xdp); + bi->xdp = NULL; return skb; } @@ -434,9 +237,6 @@ int ixgbe_clean_rx_irq_zc(struct ixgbe_q_vector *q_vector, unsigned int xdp_res, xdp_xmit = 0; bool failure = false; struct sk_buff *skb; - struct xdp_buff xdp; - - xdp.rxq = &rx_ring->xdp_rxq; while (likely(total_rx_packets < budget)) { union ixgbe_adv_rx_desc *rx_desc; @@ -446,8 +246,8 @@ int ixgbe_clean_rx_irq_zc(struct ixgbe_q_vector *q_vector, /* return some buffers to hardware, one at a time is too slow */ if (cleaned_count >= IXGBE_RX_BUFFER_WRITE) { failure = failure || - !ixgbe_alloc_rx_buffers_fast_zc(rx_ring, - cleaned_count); + !ixgbe_alloc_rx_buffers_zc(rx_ring, + cleaned_count); cleaned_count = 0; } @@ -462,42 +262,40 @@ int ixgbe_clean_rx_irq_zc(struct ixgbe_q_vector *q_vector, */ dma_rmb(); - bi = ixgbe_get_rx_buffer_zc(rx_ring, size); + bi = &rx_ring->rx_buffer_info[rx_ring->next_to_clean]; if (unlikely(!ixgbe_test_staterr(rx_desc, IXGBE_RXD_STAT_EOP))) { struct ixgbe_rx_buffer *next_bi; - ixgbe_reuse_rx_buffer_zc(rx_ring, bi); + xsk_buff_free(bi->xdp); + bi->xdp = NULL; ixgbe_inc_ntc(rx_ring); next_bi = &rx_ring->rx_buffer_info[rx_ring->next_to_clean]; - next_bi->skb = ERR_PTR(-EINVAL); + next_bi->discard = true; continue; } - if (unlikely(bi->skb)) { - ixgbe_reuse_rx_buffer_zc(rx_ring, bi); + if (unlikely(bi->discard)) { + xsk_buff_free(bi->xdp); + bi->xdp = NULL; + bi->discard = false; ixgbe_inc_ntc(rx_ring); continue; } - xdp.data = bi->addr; - xdp.data_meta = xdp.data; - xdp.data_hard_start = xdp.data - XDP_PACKET_HEADROOM; - xdp.data_end = xdp.data + size; - xdp.handle = bi->handle; - - xdp_res = ixgbe_run_xdp_zc(adapter, rx_ring, &xdp); + bi->xdp->data_end = bi->xdp->data + size; + xsk_buff_dma_sync_for_cpu(bi->xdp); + xdp_res = ixgbe_run_xdp_zc(adapter, rx_ring, bi->xdp); if (xdp_res) { - if (xdp_res & (IXGBE_XDP_TX | IXGBE_XDP_REDIR)) { + if (xdp_res & (IXGBE_XDP_TX | IXGBE_XDP_REDIR)) xdp_xmit |= xdp_res; - bi->addr = NULL; - bi->skb = NULL; - } else { - ixgbe_reuse_rx_buffer_zc(rx_ring, bi); - } + else + xsk_buff_free(bi->xdp); + + bi->xdp = NULL; total_rx_packets++; total_rx_bytes += size; @@ -507,7 +305,7 @@ int ixgbe_clean_rx_irq_zc(struct ixgbe_q_vector *q_vector, } /* XDP_PASS path */ - skb = ixgbe_construct_skb_zc(rx_ring, bi, &xdp); + skb = ixgbe_construct_skb_zc(rx_ring, bi); if (!skb) { rx_ring->rx_stats.alloc_rx_buff_failed++; break; @@ -559,17 +357,17 @@ int ixgbe_clean_rx_irq_zc(struct ixgbe_q_vector *q_vector, void ixgbe_xsk_clean_rx_ring(struct ixgbe_ring *rx_ring) { - u16 i = rx_ring->next_to_clean; - struct ixgbe_rx_buffer *bi = &rx_ring->rx_buffer_info[i]; + struct ixgbe_rx_buffer *bi; + u16 i; - while (i != rx_ring->next_to_alloc) { - xsk_umem_fq_reuse(rx_ring->xsk_umem, bi->handle); - i++; - bi++; - if (i == rx_ring->count) { - i = 0; - bi = rx_ring->rx_buffer_info; - } + for (i = 0; i < rx_ring->count; i++) { + bi = &rx_ring->rx_buffer_info[i]; + + if (!bi->xdp) + continue; + + xsk_buff_free(bi->xdp); + bi->xdp = NULL; } } @@ -592,10 +390,9 @@ static bool ixgbe_xmit_zc(struct ixgbe_ring *xdp_ring, unsigned int budget) if (!xsk_umem_consume_tx(xdp_ring->xsk_umem, &desc)) break; - dma = xdp_umem_get_dma(xdp_ring->xsk_umem, desc.addr); - - dma_sync_single_for_device(xdp_ring->dev, dma, desc.len, - DMA_BIDIRECTIONAL); + dma = xsk_buff_raw_get_dma(xdp_ring->xsk_umem, desc.addr); + xsk_buff_raw_dma_sync_for_device(xdp_ring->xsk_umem, dma, + desc.len); tx_bi = &xdp_ring->tx_buffer_info[xdp_ring->next_to_use]; tx_bi->bytecount = desc.len; diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c index 4622c4ea2e46..a39e2cb384dd 100644 --- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c @@ -1095,19 +1095,31 @@ xdp_out: return ERR_PTR(-result); } +static unsigned int ixgbevf_rx_frame_truesize(struct ixgbevf_ring *rx_ring, + unsigned int size) +{ + unsigned int truesize; + +#if (PAGE_SIZE < 8192) + truesize = ixgbevf_rx_pg_size(rx_ring) / 2; /* Must be power-of-2 */ +#else + truesize = ring_uses_build_skb(rx_ring) ? + SKB_DATA_ALIGN(IXGBEVF_SKB_PAD + size) + + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) : + SKB_DATA_ALIGN(size); +#endif + return truesize; +} + static void ixgbevf_rx_buffer_flip(struct ixgbevf_ring *rx_ring, struct ixgbevf_rx_buffer *rx_buffer, unsigned int size) { -#if (PAGE_SIZE < 8192) - unsigned int truesize = ixgbevf_rx_pg_size(rx_ring) / 2; + unsigned int truesize = ixgbevf_rx_frame_truesize(rx_ring, size); +#if (PAGE_SIZE < 8192) rx_buffer->page_offset ^= truesize; #else - unsigned int truesize = ring_uses_build_skb(rx_ring) ? - SKB_DATA_ALIGN(IXGBEVF_SKB_PAD + size) : - SKB_DATA_ALIGN(size); - rx_buffer->page_offset += truesize; #endif } @@ -1125,6 +1137,11 @@ static int ixgbevf_clean_rx_irq(struct ixgbevf_q_vector *q_vector, xdp.rxq = &rx_ring->xdp_rxq; + /* Frame size depend on rx_ring setup when PAGE_SIZE=4K */ +#if (PAGE_SIZE < 8192) + xdp.frame_sz = ixgbevf_rx_frame_truesize(rx_ring, 0); +#endif + while (likely(total_rx_packets < budget)) { struct ixgbevf_rx_buffer *rx_buffer; union ixgbe_adv_rx_desc *rx_desc; @@ -1157,7 +1174,10 @@ static int ixgbevf_clean_rx_irq(struct ixgbevf_q_vector *q_vector, xdp.data_hard_start = xdp.data - ixgbevf_rx_offset(rx_ring); xdp.data_end = xdp.data + size; - +#if (PAGE_SIZE > 4096) + /* At larger PAGE_SIZE, frame_sz depend on len size */ + xdp.frame_sz = ixgbevf_rx_frame_truesize(rx_ring, size); +#endif skb = ixgbevf_run_xdp(adapter, rx_ring, &xdp); } diff --git a/drivers/net/ethernet/lantiq_xrx200.c b/drivers/net/ethernet/lantiq_xrx200.c index 900affbdcc0e..1645e4e7ebdb 100644 --- a/drivers/net/ethernet/lantiq_xrx200.c +++ b/drivers/net/ethernet/lantiq_xrx200.c @@ -276,7 +276,8 @@ static int xrx200_tx_housekeeping(struct napi_struct *napi, int budget) return pkts; } -static int xrx200_start_xmit(struct sk_buff *skb, struct net_device *net_dev) +static netdev_tx_t xrx200_start_xmit(struct sk_buff *skb, + struct net_device *net_dev) { struct xrx200_priv *priv = netdev_priv(net_dev); struct xrx200_chan *ch = &priv->chan_tx; diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c index 81d24481b22c..4d4b6243318a 100644 --- a/drivers/net/ethernet/marvell/mv643xx_eth.c +++ b/drivers/net/ethernet/marvell/mv643xx_eth.c @@ -666,11 +666,6 @@ static inline unsigned int has_tiny_unaligned_frags(struct sk_buff *skb) return 0; } -static inline __be16 sum16_as_be(__sum16 sum) -{ - return (__force __be16)sum; -} - static int skb_tx_csum(struct mv643xx_eth_private *mp, struct sk_buff *skb, u16 *l4i_chk, u32 *command, int length) { diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c index 51889770958d..011cd26953d9 100644 --- a/drivers/net/ethernet/marvell/mvneta.c +++ b/drivers/net/ethernet/marvell/mvneta.c @@ -324,7 +324,8 @@ ETH_HLEN + ETH_FCS_LEN, \ cache_line_size()) -#define MVNETA_SKB_HEADROOM max(XDP_PACKET_HEADROOM, NET_SKB_PAD) +/* Driver assumes that the last 3 bits are 0 */ +#define MVNETA_SKB_HEADROOM (max(XDP_PACKET_HEADROOM, NET_SKB_PAD) & ~0x7) #define MVNETA_SKB_PAD (SKB_DATA_ALIGN(sizeof(struct skb_shared_info) + \ MVNETA_SKB_HEADROOM)) #define MVNETA_SKB_SIZE(len) (SKB_DATA_ALIGN(len) + MVNETA_SKB_PAD) @@ -2072,7 +2073,7 @@ mvneta_xdp_xmit_back(struct mvneta_port *pp, struct xdp_buff *xdp) int cpu; u32 ret; - xdpf = convert_to_xdp_frame(xdp); + xdpf = xdp_convert_buff_to_frame(xdp); if (unlikely(!xdpf)) return MVNETA_XDP_DROPPED; @@ -2148,12 +2149,17 @@ mvneta_run_xdp(struct mvneta_port *pp, struct mvneta_rx_queue *rxq, struct bpf_prog *prog, struct xdp_buff *xdp, struct mvneta_stats *stats) { - unsigned int len; + unsigned int len, sync; + struct page *page; u32 ret, act; len = xdp->data_end - xdp->data_hard_start - pp->rx_offset_correction; act = bpf_prog_run_xdp(prog, xdp); + /* Due xdp_adjust_tail: DMA sync for_device cover max len CPU touch */ + sync = xdp->data_end - xdp->data_hard_start - pp->rx_offset_correction; + sync = max(sync, len); + switch (act) { case XDP_PASS: stats->xdp_pass++; @@ -2164,9 +2170,8 @@ mvneta_run_xdp(struct mvneta_port *pp, struct mvneta_rx_queue *rxq, err = xdp_do_redirect(pp->dev, xdp, prog); if (unlikely(err)) { ret = MVNETA_XDP_DROPPED; - page_pool_put_page(rxq->page_pool, - virt_to_head_page(xdp->data), len, - true); + page = virt_to_head_page(xdp->data); + page_pool_put_page(rxq->page_pool, page, sync, true); } else { ret = MVNETA_XDP_REDIR; stats->xdp_redirect++; @@ -2175,10 +2180,10 @@ mvneta_run_xdp(struct mvneta_port *pp, struct mvneta_rx_queue *rxq, } case XDP_TX: ret = mvneta_xdp_xmit_back(pp, xdp); - if (ret != MVNETA_XDP_TX) - page_pool_put_page(rxq->page_pool, - virt_to_head_page(xdp->data), len, - true); + if (ret != MVNETA_XDP_TX) { + page = virt_to_head_page(xdp->data); + page_pool_put_page(rxq->page_pool, page, sync, true); + } break; default: bpf_warn_invalid_xdp_action(act); @@ -2187,8 +2192,8 @@ mvneta_run_xdp(struct mvneta_port *pp, struct mvneta_rx_queue *rxq, trace_xdp_exception(pp->dev, prog, act); /* fall through */ case XDP_DROP: - page_pool_put_page(rxq->page_pool, - virt_to_head_page(xdp->data), len, true); + page = virt_to_head_page(xdp->data); + page_pool_put_page(rxq->page_pool, page, sync, true); ret = MVNETA_XDP_DROPPED; stats->xdp_drop++; break; @@ -2320,6 +2325,7 @@ static int mvneta_rx_swbm(struct napi_struct *napi, rcu_read_lock(); xdp_prog = READ_ONCE(pp->xdp_prog); xdp_buf.rxq = &rxq->xdp_rxq; + xdp_buf.frame_sz = PAGE_SIZE; /* Fairness NAPI loop */ while (rx_proc < budget && rx_proc < rx_todo) { diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_cls.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_cls.c index 7352244c5e68..d4a4e241333d 100644 --- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_cls.c +++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_cls.c @@ -1070,7 +1070,7 @@ void mvpp2_cls_oversize_rxq_set(struct mvpp2_port *port) (port->first_rxq >> MVPP2_CLS_OVERSIZE_RXQ_LOW_BITS)); val = mvpp2_read(port->priv, MVPP2_CLS_SWFWD_PCTRL_REG); - val |= MVPP2_CLS_SWFWD_PCTRL_MASK(port->id); + val &= ~MVPP2_CLS_SWFWD_PCTRL_MASK(port->id); mvpp2_write(port->priv, MVPP2_CLS_SWFWD_PCTRL_REG, val); } diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c index f1d2dea90a8c..5975521a4c86 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c @@ -379,40 +379,35 @@ void otx2_config_irq_coalescing(struct otx2_nic *pfvf, int qidx) (pfvf->hw.cq_ecount_wait - 1)); } -dma_addr_t otx2_alloc_rbuf(struct otx2_nic *pfvf, struct otx2_pool *pool, - gfp_t gfp) +dma_addr_t __otx2_alloc_rbuf(struct otx2_nic *pfvf, struct otx2_pool *pool) { dma_addr_t iova; + u8 *buf; - /* Check if request can be accommodated in previous allocated page */ - if (pool->page && ((pool->page_offset + pool->rbsize) <= - (PAGE_SIZE << pool->rbpage_order))) { - pool->pageref++; - goto ret; - } - - otx2_get_page(pool); - - /* Allocate a new page */ - pool->page = alloc_pages(gfp | __GFP_COMP | __GFP_NOWARN, - pool->rbpage_order); - if (unlikely(!pool->page)) + buf = napi_alloc_frag(pool->rbsize); + if (unlikely(!buf)) return -ENOMEM; - pool->page_offset = 0; -ret: - iova = (u64)otx2_dma_map_page(pfvf, pool->page, pool->page_offset, - pool->rbsize, DMA_FROM_DEVICE); - if (!iova) { - if (!pool->page_offset) - __free_pages(pool->page, pool->rbpage_order); - pool->page = NULL; + iova = dma_map_single_attrs(pfvf->dev, buf, pool->rbsize, + DMA_FROM_DEVICE, DMA_ATTR_SKIP_CPU_SYNC); + if (unlikely(dma_mapping_error(pfvf->dev, iova))) { + page_frag_free(buf); return -ENOMEM; } - pool->page_offset += pool->rbsize; + return iova; } +static dma_addr_t otx2_alloc_rbuf(struct otx2_nic *pfvf, struct otx2_pool *pool) +{ + dma_addr_t addr; + + local_bh_disable(); + addr = __otx2_alloc_rbuf(pfvf, pool); + local_bh_enable(); + return addr; +} + void otx2_tx_timeout(struct net_device *netdev, unsigned int txq) { struct otx2_nic *pfvf = netdev_priv(netdev); @@ -805,7 +800,7 @@ static void otx2_pool_refill_task(struct work_struct *work) free_ptrs = cq->pool_ptrs; while (cq->pool_ptrs) { - bufptr = otx2_alloc_rbuf(pfvf, rbpool, GFP_KERNEL); + bufptr = otx2_alloc_rbuf(pfvf, rbpool); if (bufptr <= 0) { /* Schedule a WQ if we fails to free atleast half of the * pointers else enable napi for this RQ. @@ -1064,7 +1059,6 @@ static int otx2_pool_init(struct otx2_nic *pfvf, u16 pool_id, return err; pool->rbsize = buf_size; - pool->rbpage_order = get_order(buf_size); /* Initialize this pool's context via AF */ aq = otx2_mbox_alloc_msg_npa_aq_enq(&pfvf->mbox); @@ -1152,13 +1146,12 @@ int otx2_sq_aura_pool_init(struct otx2_nic *pfvf) return -ENOMEM; for (ptr = 0; ptr < num_sqbs; ptr++) { - bufptr = otx2_alloc_rbuf(pfvf, pool, GFP_KERNEL); + bufptr = otx2_alloc_rbuf(pfvf, pool); if (bufptr <= 0) return bufptr; otx2_aura_freeptr(pfvf, pool_id, bufptr); sq->sqb_ptrs[sq->sqb_count++] = (u64)bufptr; } - otx2_get_page(pool); } return 0; @@ -1204,13 +1197,12 @@ int otx2_rq_aura_pool_init(struct otx2_nic *pfvf) for (pool_id = 0; pool_id < hw->rqpool_cnt; pool_id++) { pool = &pfvf->qset.pool[pool_id]; for (ptr = 0; ptr < num_ptrs; ptr++) { - bufptr = otx2_alloc_rbuf(pfvf, pool, GFP_KERNEL); + bufptr = otx2_alloc_rbuf(pfvf, pool); if (bufptr <= 0) return bufptr; otx2_aura_freeptr(pfvf, pool_id, bufptr + OTX2_HEAD_ROOM); } - otx2_get_page(pool); } return 0; diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h index 018c283a0ac4..2fa29889522e 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h @@ -309,7 +309,7 @@ static inline void __iomem *otx2_get_regaddr(struct otx2_nic *nic, u64 offset) default: blkaddr = BLKADDR_RVUM; break; - }; + } offset &= ~(RVU_FUNC_BLKADDR_MASK << RVU_FUNC_BLKADDR_SHIFT); offset |= (blkaddr << RVU_FUNC_BLKADDR_SHIFT); @@ -434,18 +434,6 @@ static inline void otx2_aura_freeptr(struct otx2_nic *pfvf, otx2_get_regaddr(pfvf, NPA_LF_AURA_OP_FREE0)); } -/* Update page ref count */ -static inline void otx2_get_page(struct otx2_pool *pool) -{ - if (!pool->page) - return; - - if (pool->pageref) - page_ref_add(pool->page, pool->pageref); - pool->pageref = 0; - pool->page = NULL; -} - static inline int otx2_get_pool_idx(struct otx2_nic *pfvf, int type, int idx) { if (type == AURA_NIX_SQ) @@ -589,8 +577,7 @@ int otx2_txschq_config(struct otx2_nic *pfvf, int lvl); int otx2_txsch_alloc(struct otx2_nic *pfvf); int otx2_txschq_stop(struct otx2_nic *pfvf); void otx2_sqb_flush(struct otx2_nic *pfvf); -dma_addr_t otx2_alloc_rbuf(struct otx2_nic *pfvf, struct otx2_pool *pool, - gfp_t gfp); +dma_addr_t __otx2_alloc_rbuf(struct otx2_nic *pfvf, struct otx2_pool *pool); int otx2_rxtx_enable(struct otx2_nic *pfvf, bool enable); void otx2_ctx_disable(struct mbox *mbox, int type, bool npa); int otx2_nix_config_bp(struct otx2_nic *pfvf, bool enable); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c index 411e5ea1031e..64786568af0d 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c @@ -1856,13 +1856,17 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id) num_vec = pci_msix_vec_count(pdev); hw->irq_name = devm_kmalloc_array(&hw->pdev->dev, num_vec, NAME_SIZE, GFP_KERNEL); - if (!hw->irq_name) + if (!hw->irq_name) { + err = -ENOMEM; goto err_free_netdev; + } hw->affinity_mask = devm_kcalloc(&hw->pdev->dev, num_vec, sizeof(cpumask_var_t), GFP_KERNEL); - if (!hw->affinity_mask) + if (!hw->affinity_mask) { + err = -ENOMEM; goto err_free_netdev; + } /* Map CSRs */ pf->reg_base = pcim_iomap(pdev, PCI_CFG_REG_BAR_NUM, 0); diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c index 45abe0cd0e7b..b04f5429d72d 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c @@ -286,7 +286,7 @@ static int otx2_rx_napi_handler(struct otx2_nic *pfvf, /* Refill pool with new buffers */ while (cq->pool_ptrs) { - bufptr = otx2_alloc_rbuf(pfvf, cq->rbpool, GFP_ATOMIC); + bufptr = __otx2_alloc_rbuf(pfvf, cq->rbpool); if (unlikely(bufptr <= 0)) { struct refill_work *work; struct delayed_work *dwork; @@ -304,7 +304,6 @@ static int otx2_rx_napi_handler(struct otx2_nic *pfvf, otx2_aura_freeptr(pfvf, cq->cq_idx, bufptr + OTX2_HEAD_ROOM); cq->pool_ptrs--; } - otx2_get_page(cq->rbpool); return processed_cqe; } diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h index 4ab32d3adb78..da97f2d4416f 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.h @@ -113,11 +113,7 @@ struct otx2_cq_poll { struct otx2_pool { struct qmem *stack; struct qmem *fc_addr; - u8 rbpage_order; u16 rbsize; - u32 page_offset; - u16 pageref; - struct page *page; }; struct otx2_cq_queue { diff --git a/drivers/net/ethernet/marvell/pxa168_eth.c b/drivers/net/ethernet/marvell/pxa168_eth.c index 7a0d785b826c..17243bb5ba91 100644 --- a/drivers/net/ethernet/marvell/pxa168_eth.c +++ b/drivers/net/ethernet/marvell/pxa168_eth.c @@ -1418,7 +1418,7 @@ static int pxa168_eth_probe(struct platform_device *pdev) pep->base = devm_platform_ioremap_resource(pdev, 0); if (IS_ERR(pep->base)) { - err = -ENOMEM; + err = PTR_ERR(pep->base); goto err_netdev; } diff --git a/drivers/net/ethernet/mediatek/Kconfig b/drivers/net/ethernet/mediatek/Kconfig index 4968352ba188..500c15e7ea4a 100644 --- a/drivers/net/ethernet/mediatek/Kconfig +++ b/drivers/net/ethernet/mediatek/Kconfig @@ -1,6 +1,6 @@ # SPDX-License-Identifier: GPL-2.0-only config NET_VENDOR_MEDIATEK - bool "MediaTek ethernet driver" + bool "MediaTek devices" depends on ARCH_MEDIATEK || SOC_MT7621 || SOC_MT7620 ---help--- If you have a Mediatek SoC with ethernet, say Y. @@ -14,4 +14,11 @@ config NET_MEDIATEK_SOC This driver supports the gigabit ethernet MACs in the MediaTek SoC family. +config NET_MEDIATEK_STAR_EMAC + tristate "MediaTek STAR Ethernet MAC support" + select PHYLIB + help + This driver supports the ethernet MAC IP first used on + MediaTek MT85** SoCs. + endif #NET_VENDOR_MEDIATEK diff --git a/drivers/net/ethernet/mediatek/Makefile b/drivers/net/ethernet/mediatek/Makefile index 2d8362f9341b..3a777b4a6cd3 100644 --- a/drivers/net/ethernet/mediatek/Makefile +++ b/drivers/net/ethernet/mediatek/Makefile @@ -3,5 +3,6 @@ # Makefile for the Mediatek SoCs built-in ethernet macs # -obj-$(CONFIG_NET_MEDIATEK_SOC) += mtk_eth.o +obj-$(CONFIG_NET_MEDIATEK_SOC) += mtk_eth.o mtk_eth-y := mtk_eth_soc.o mtk_sgmii.o mtk_eth_path.o +obj-$(CONFIG_NET_MEDIATEK_STAR_EMAC) += mtk_star_emac.o diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 09047109d0da..f6a1f8666f95 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -65,7 +65,7 @@ u32 mtk_r32(struct mtk_eth *eth, unsigned reg) return __raw_readl(eth->base + reg); } -u32 mtk_m32(struct mtk_eth *eth, u32 mask, u32 set, unsigned reg) +static u32 mtk_m32(struct mtk_eth *eth, u32 mask, u32 set, unsigned reg) { u32 val; @@ -1122,7 +1122,7 @@ static void mtk_stop_queue(struct mtk_eth *eth) } } -static int mtk_start_xmit(struct sk_buff *skb, struct net_device *dev) +static netdev_tx_t mtk_start_xmit(struct sk_buff *skb, struct net_device *dev) { struct mtk_mac *mac = netdev_priv(dev); struct mtk_eth *eth = mac->hw; diff --git a/drivers/net/ethernet/mediatek/mtk_star_emac.c b/drivers/net/ethernet/mediatek/mtk_star_emac.c new file mode 100644 index 000000000000..f1ace4fec19f --- /dev/null +++ b/drivers/net/ethernet/mediatek/mtk_star_emac.c @@ -0,0 +1,1651 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2020 MediaTek Corporation + * Copyright (c) 2020 BayLibre SAS + * + * Author: Bartosz Golaszewski <bgolaszewski@baylibre.com> + */ + +#include <linux/bits.h> +#include <linux/clk.h> +#include <linux/compiler.h> +#include <linux/dma-mapping.h> +#include <linux/etherdevice.h> +#include <linux/kernel.h> +#include <linux/mfd/syscon.h> +#include <linux/mii.h> +#include <linux/module.h> +#include <linux/netdevice.h> +#include <linux/of.h> +#include <linux/of_mdio.h> +#include <linux/of_net.h> +#include <linux/platform_device.h> +#include <linux/pm.h> +#include <linux/regmap.h> +#include <linux/skbuff.h> +#include <linux/spinlock.h> +#include <linux/workqueue.h> + +#define MTK_STAR_DRVNAME "mtk_star_emac" + +#define MTK_STAR_WAIT_TIMEOUT 300 +#define MTK_STAR_MAX_FRAME_SIZE 1514 +#define MTK_STAR_SKB_ALIGNMENT 16 +#define MTK_STAR_NAPI_WEIGHT 64 +#define MTK_STAR_HASHTABLE_MC_LIMIT 256 +#define MTK_STAR_HASHTABLE_SIZE_MAX 512 + +/* Normally we'd use NET_IP_ALIGN but on arm64 its value is 0 and it doesn't + * work for this controller. + */ +#define MTK_STAR_IP_ALIGN 2 + +static const char *const mtk_star_clk_names[] = { "core", "reg", "trans" }; +#define MTK_STAR_NCLKS ARRAY_SIZE(mtk_star_clk_names) + +/* PHY Control Register 0 */ +#define MTK_STAR_REG_PHY_CTRL0 0x0000 +#define MTK_STAR_BIT_PHY_CTRL0_WTCMD BIT(13) +#define MTK_STAR_BIT_PHY_CTRL0_RDCMD BIT(14) +#define MTK_STAR_BIT_PHY_CTRL0_RWOK BIT(15) +#define MTK_STAR_MSK_PHY_CTRL0_PREG GENMASK(12, 8) +#define MTK_STAR_OFF_PHY_CTRL0_PREG 8 +#define MTK_STAR_MSK_PHY_CTRL0_RWDATA GENMASK(31, 16) +#define MTK_STAR_OFF_PHY_CTRL0_RWDATA 16 + +/* PHY Control Register 1 */ +#define MTK_STAR_REG_PHY_CTRL1 0x0004 +#define MTK_STAR_BIT_PHY_CTRL1_LINK_ST BIT(0) +#define MTK_STAR_BIT_PHY_CTRL1_AN_EN BIT(8) +#define MTK_STAR_OFF_PHY_CTRL1_FORCE_SPD 9 +#define MTK_STAR_VAL_PHY_CTRL1_FORCE_SPD_10M 0x00 +#define MTK_STAR_VAL_PHY_CTRL1_FORCE_SPD_100M 0x01 +#define MTK_STAR_VAL_PHY_CTRL1_FORCE_SPD_1000M 0x02 +#define MTK_STAR_BIT_PHY_CTRL1_FORCE_DPX BIT(11) +#define MTK_STAR_BIT_PHY_CTRL1_FORCE_FC_RX BIT(12) +#define MTK_STAR_BIT_PHY_CTRL1_FORCE_FC_TX BIT(13) + +/* MAC Configuration Register */ +#define MTK_STAR_REG_MAC_CFG 0x0008 +#define MTK_STAR_OFF_MAC_CFG_IPG 10 +#define MTK_STAR_VAL_MAC_CFG_IPG_96BIT GENMASK(4, 0) +#define MTK_STAR_BIT_MAC_CFG_MAXLEN_1522 BIT(16) +#define MTK_STAR_BIT_MAC_CFG_AUTO_PAD BIT(19) +#define MTK_STAR_BIT_MAC_CFG_CRC_STRIP BIT(20) +#define MTK_STAR_BIT_MAC_CFG_VLAN_STRIP BIT(22) +#define MTK_STAR_BIT_MAC_CFG_NIC_PD BIT(31) + +/* Flow-Control Configuration Register */ +#define MTK_STAR_REG_FC_CFG 0x000c +#define MTK_STAR_BIT_FC_CFG_BP_EN BIT(7) +#define MTK_STAR_BIT_FC_CFG_UC_PAUSE_DIR BIT(8) +#define MTK_STAR_OFF_FC_CFG_SEND_PAUSE_TH 16 +#define MTK_STAR_MSK_FC_CFG_SEND_PAUSE_TH GENMASK(27, 16) +#define MTK_STAR_VAL_FC_CFG_SEND_PAUSE_TH_2K 0x800 + +/* ARL Configuration Register */ +#define MTK_STAR_REG_ARL_CFG 0x0010 +#define MTK_STAR_BIT_ARL_CFG_HASH_ALG BIT(0) +#define MTK_STAR_BIT_ARL_CFG_MISC_MODE BIT(4) + +/* MAC High and Low Bytes Registers */ +#define MTK_STAR_REG_MY_MAC_H 0x0014 +#define MTK_STAR_REG_MY_MAC_L 0x0018 + +/* Hash Table Control Register */ +#define MTK_STAR_REG_HASH_CTRL 0x001c +#define MTK_STAR_MSK_HASH_CTRL_HASH_BIT_ADDR GENMASK(8, 0) +#define MTK_STAR_BIT_HASH_CTRL_HASH_BIT_DATA BIT(12) +#define MTK_STAR_BIT_HASH_CTRL_ACC_CMD BIT(13) +#define MTK_STAR_BIT_HASH_CTRL_CMD_START BIT(14) +#define MTK_STAR_BIT_HASH_CTRL_BIST_OK BIT(16) +#define MTK_STAR_BIT_HASH_CTRL_BIST_DONE BIT(17) +#define MTK_STAR_BIT_HASH_CTRL_BIST_EN BIT(31) + +/* TX DMA Control Register */ +#define MTK_STAR_REG_TX_DMA_CTRL 0x0034 +#define MTK_STAR_BIT_TX_DMA_CTRL_START BIT(0) +#define MTK_STAR_BIT_TX_DMA_CTRL_STOP BIT(1) +#define MTK_STAR_BIT_TX_DMA_CTRL_RESUME BIT(2) + +/* RX DMA Control Register */ +#define MTK_STAR_REG_RX_DMA_CTRL 0x0038 +#define MTK_STAR_BIT_RX_DMA_CTRL_START BIT(0) +#define MTK_STAR_BIT_RX_DMA_CTRL_STOP BIT(1) +#define MTK_STAR_BIT_RX_DMA_CTRL_RESUME BIT(2) + +/* DMA Address Registers */ +#define MTK_STAR_REG_TX_DPTR 0x003c +#define MTK_STAR_REG_RX_DPTR 0x0040 +#define MTK_STAR_REG_TX_BASE_ADDR 0x0044 +#define MTK_STAR_REG_RX_BASE_ADDR 0x0048 + +/* Interrupt Status Register */ +#define MTK_STAR_REG_INT_STS 0x0050 +#define MTK_STAR_REG_INT_STS_PORT_STS_CHG BIT(2) +#define MTK_STAR_REG_INT_STS_MIB_CNT_TH BIT(3) +#define MTK_STAR_BIT_INT_STS_FNRC BIT(6) +#define MTK_STAR_BIT_INT_STS_TNTC BIT(8) + +/* Interrupt Mask Register */ +#define MTK_STAR_REG_INT_MASK 0x0054 +#define MTK_STAR_BIT_INT_MASK_FNRC BIT(6) + +/* Misc. Config Register */ +#define MTK_STAR_REG_TEST1 0x005c +#define MTK_STAR_BIT_TEST1_RST_HASH_MBIST BIT(31) + +/* Extended Configuration Register */ +#define MTK_STAR_REG_EXT_CFG 0x0060 +#define MTK_STAR_OFF_EXT_CFG_SND_PAUSE_RLS 16 +#define MTK_STAR_MSK_EXT_CFG_SND_PAUSE_RLS GENMASK(26, 16) +#define MTK_STAR_VAL_EXT_CFG_SND_PAUSE_RLS_1K 0x400 + +/* EthSys Configuration Register */ +#define MTK_STAR_REG_SYS_CONF 0x0094 +#define MTK_STAR_BIT_MII_PAD_OUT_ENABLE BIT(0) +#define MTK_STAR_BIT_EXT_MDC_MODE BIT(1) +#define MTK_STAR_BIT_SWC_MII_MODE BIT(2) + +/* MAC Clock Configuration Register */ +#define MTK_STAR_REG_MAC_CLK_CONF 0x00ac +#define MTK_STAR_MSK_MAC_CLK_CONF GENMASK(7, 0) +#define MTK_STAR_BIT_CLK_DIV_10 0x0a + +/* Counter registers. */ +#define MTK_STAR_REG_C_RXOKPKT 0x0100 +#define MTK_STAR_REG_C_RXOKBYTE 0x0104 +#define MTK_STAR_REG_C_RXRUNT 0x0108 +#define MTK_STAR_REG_C_RXLONG 0x010c +#define MTK_STAR_REG_C_RXDROP 0x0110 +#define MTK_STAR_REG_C_RXCRC 0x0114 +#define MTK_STAR_REG_C_RXARLDROP 0x0118 +#define MTK_STAR_REG_C_RXVLANDROP 0x011c +#define MTK_STAR_REG_C_RXCSERR 0x0120 +#define MTK_STAR_REG_C_RXPAUSE 0x0124 +#define MTK_STAR_REG_C_TXOKPKT 0x0128 +#define MTK_STAR_REG_C_TXOKBYTE 0x012c +#define MTK_STAR_REG_C_TXPAUSECOL 0x0130 +#define MTK_STAR_REG_C_TXRTY 0x0134 +#define MTK_STAR_REG_C_TXSKIP 0x0138 +#define MTK_STAR_REG_C_TX_ARP 0x013c +#define MTK_STAR_REG_C_RX_RERR 0x01d8 +#define MTK_STAR_REG_C_RX_UNI 0x01dc +#define MTK_STAR_REG_C_RX_MULTI 0x01e0 +#define MTK_STAR_REG_C_RX_BROAD 0x01e4 +#define MTK_STAR_REG_C_RX_ALIGNERR 0x01e8 +#define MTK_STAR_REG_C_TX_UNI 0x01ec +#define MTK_STAR_REG_C_TX_MULTI 0x01f0 +#define MTK_STAR_REG_C_TX_BROAD 0x01f4 +#define MTK_STAR_REG_C_TX_TIMEOUT 0x01f8 +#define MTK_STAR_REG_C_TX_LATECOL 0x01fc +#define MTK_STAR_REG_C_RX_LENGTHERR 0x0214 +#define MTK_STAR_REG_C_RX_TWIST 0x0218 + +/* Ethernet CFG Control */ +#define MTK_PERICFG_REG_NIC_CFG_CON 0x03c4 +#define MTK_PERICFG_MSK_NIC_CFG_CON_CFG_MII GENMASK(3, 0) +#define MTK_PERICFG_BIT_NIC_CFG_CON_RMII BIT(0) + +/* Represents the actual structure of descriptors used by the MAC. We can + * reuse the same structure for both TX and RX - the layout is the same, only + * the flags differ slightly. + */ +struct mtk_star_ring_desc { + /* Contains both the status flags as well as packet length. */ + u32 status; + u32 data_ptr; + u32 vtag; + u32 reserved; +}; + +#define MTK_STAR_DESC_MSK_LEN GENMASK(15, 0) +#define MTK_STAR_DESC_BIT_RX_CRCE BIT(24) +#define MTK_STAR_DESC_BIT_RX_OSIZE BIT(25) +#define MTK_STAR_DESC_BIT_INT BIT(27) +#define MTK_STAR_DESC_BIT_LS BIT(28) +#define MTK_STAR_DESC_BIT_FS BIT(29) +#define MTK_STAR_DESC_BIT_EOR BIT(30) +#define MTK_STAR_DESC_BIT_COWN BIT(31) + +/* Helper structure for storing data read from/written to descriptors in order + * to limit reads from/writes to DMA memory. + */ +struct mtk_star_ring_desc_data { + unsigned int len; + unsigned int flags; + dma_addr_t dma_addr; + struct sk_buff *skb; +}; + +#define MTK_STAR_RING_NUM_DESCS 128 +#define MTK_STAR_NUM_TX_DESCS MTK_STAR_RING_NUM_DESCS +#define MTK_STAR_NUM_RX_DESCS MTK_STAR_RING_NUM_DESCS +#define MTK_STAR_NUM_DESCS_TOTAL (MTK_STAR_RING_NUM_DESCS * 2) +#define MTK_STAR_DMA_SIZE \ + (MTK_STAR_NUM_DESCS_TOTAL * sizeof(struct mtk_star_ring_desc)) + +struct mtk_star_ring { + struct mtk_star_ring_desc *descs; + struct sk_buff *skbs[MTK_STAR_RING_NUM_DESCS]; + dma_addr_t dma_addrs[MTK_STAR_RING_NUM_DESCS]; + unsigned int head; + unsigned int tail; +}; + +struct mtk_star_priv { + struct net_device *ndev; + + struct regmap *regs; + struct regmap *pericfg; + + struct clk_bulk_data clks[MTK_STAR_NCLKS]; + + void *ring_base; + struct mtk_star_ring_desc *descs_base; + dma_addr_t dma_addr; + struct mtk_star_ring tx_ring; + struct mtk_star_ring rx_ring; + + struct mii_bus *mii; + struct napi_struct napi; + + struct device_node *phy_node; + phy_interface_t phy_intf; + struct phy_device *phydev; + unsigned int link; + int speed; + int duplex; + int pause; + + /* Protects against concurrent descriptor access. */ + spinlock_t lock; + + struct rtnl_link_stats64 stats; + struct work_struct stats_work; +}; + +static struct device *mtk_star_get_dev(struct mtk_star_priv *priv) +{ + return priv->ndev->dev.parent; +} + +static const struct regmap_config mtk_star_regmap_config = { + .reg_bits = 32, + .val_bits = 32, + .reg_stride = 4, + .disable_locking = true, +}; + +static void mtk_star_ring_init(struct mtk_star_ring *ring, + struct mtk_star_ring_desc *descs) +{ + memset(ring, 0, sizeof(*ring)); + ring->descs = descs; + ring->head = 0; + ring->tail = 0; +} + +static int mtk_star_ring_pop_tail(struct mtk_star_ring *ring, + struct mtk_star_ring_desc_data *desc_data) +{ + struct mtk_star_ring_desc *desc = &ring->descs[ring->tail]; + unsigned int status; + + status = READ_ONCE(desc->status); + dma_rmb(); /* Make sure we read the status bits before checking it. */ + + if (!(status & MTK_STAR_DESC_BIT_COWN)) + return -1; + + desc_data->len = status & MTK_STAR_DESC_MSK_LEN; + desc_data->flags = status & ~MTK_STAR_DESC_MSK_LEN; + desc_data->dma_addr = ring->dma_addrs[ring->tail]; + desc_data->skb = ring->skbs[ring->tail]; + + ring->dma_addrs[ring->tail] = 0; + ring->skbs[ring->tail] = NULL; + + status &= MTK_STAR_DESC_BIT_COWN | MTK_STAR_DESC_BIT_EOR; + + WRITE_ONCE(desc->data_ptr, 0); + WRITE_ONCE(desc->status, status); + + ring->tail = (ring->tail + 1) % MTK_STAR_RING_NUM_DESCS; + + return 0; +} + +static void mtk_star_ring_push_head(struct mtk_star_ring *ring, + struct mtk_star_ring_desc_data *desc_data, + unsigned int flags) +{ + struct mtk_star_ring_desc *desc = &ring->descs[ring->head]; + unsigned int status; + + status = READ_ONCE(desc->status); + + ring->skbs[ring->head] = desc_data->skb; + ring->dma_addrs[ring->head] = desc_data->dma_addr; + + status |= desc_data->len; + if (flags) + status |= flags; + + WRITE_ONCE(desc->data_ptr, desc_data->dma_addr); + WRITE_ONCE(desc->status, status); + status &= ~MTK_STAR_DESC_BIT_COWN; + /* Flush previous modifications before ownership change. */ + dma_wmb(); + WRITE_ONCE(desc->status, status); + + ring->head = (ring->head + 1) % MTK_STAR_RING_NUM_DESCS; +} + +static void +mtk_star_ring_push_head_rx(struct mtk_star_ring *ring, + struct mtk_star_ring_desc_data *desc_data) +{ + mtk_star_ring_push_head(ring, desc_data, 0); +} + +static void +mtk_star_ring_push_head_tx(struct mtk_star_ring *ring, + struct mtk_star_ring_desc_data *desc_data) +{ + static const unsigned int flags = MTK_STAR_DESC_BIT_FS | + MTK_STAR_DESC_BIT_LS | + MTK_STAR_DESC_BIT_INT; + + mtk_star_ring_push_head(ring, desc_data, flags); +} + +static unsigned int mtk_star_ring_num_used_descs(struct mtk_star_ring *ring) +{ + return abs(ring->head - ring->tail); +} + +static bool mtk_star_ring_full(struct mtk_star_ring *ring) +{ + return mtk_star_ring_num_used_descs(ring) == MTK_STAR_RING_NUM_DESCS; +} + +static bool mtk_star_ring_descs_available(struct mtk_star_ring *ring) +{ + return mtk_star_ring_num_used_descs(ring) > 0; +} + +static dma_addr_t mtk_star_dma_map_rx(struct mtk_star_priv *priv, + struct sk_buff *skb) +{ + struct device *dev = mtk_star_get_dev(priv); + + /* Data pointer for the RX DMA descriptor must be aligned to 4N + 2. */ + return dma_map_single(dev, skb_tail_pointer(skb) - 2, + skb_tailroom(skb), DMA_FROM_DEVICE); +} + +static void mtk_star_dma_unmap_rx(struct mtk_star_priv *priv, + struct mtk_star_ring_desc_data *desc_data) +{ + struct device *dev = mtk_star_get_dev(priv); + + dma_unmap_single(dev, desc_data->dma_addr, + skb_tailroom(desc_data->skb), DMA_FROM_DEVICE); +} + +static dma_addr_t mtk_star_dma_map_tx(struct mtk_star_priv *priv, + struct sk_buff *skb) +{ + struct device *dev = mtk_star_get_dev(priv); + + return dma_map_single(dev, skb->data, skb_headlen(skb), DMA_TO_DEVICE); +} + +static void mtk_star_dma_unmap_tx(struct mtk_star_priv *priv, + struct mtk_star_ring_desc_data *desc_data) +{ + struct device *dev = mtk_star_get_dev(priv); + + return dma_unmap_single(dev, desc_data->dma_addr, + skb_headlen(desc_data->skb), DMA_TO_DEVICE); +} + +static void mtk_star_nic_disable_pd(struct mtk_star_priv *priv) +{ + regmap_clear_bits(priv->regs, MTK_STAR_REG_MAC_CFG, + MTK_STAR_BIT_MAC_CFG_NIC_PD); +} + +/* Unmask the three interrupts we care about, mask all others. */ +static void mtk_star_intr_enable(struct mtk_star_priv *priv) +{ + unsigned int val = MTK_STAR_BIT_INT_STS_TNTC | + MTK_STAR_BIT_INT_STS_FNRC | + MTK_STAR_REG_INT_STS_MIB_CNT_TH; + + regmap_write(priv->regs, MTK_STAR_REG_INT_MASK, ~val); +} + +static void mtk_star_intr_disable(struct mtk_star_priv *priv) +{ + regmap_write(priv->regs, MTK_STAR_REG_INT_MASK, ~0); +} + +static void mtk_star_intr_enable_tx(struct mtk_star_priv *priv) +{ + regmap_clear_bits(priv->regs, MTK_STAR_REG_INT_MASK, + MTK_STAR_BIT_INT_STS_TNTC); +} + +static void mtk_star_intr_enable_rx(struct mtk_star_priv *priv) +{ + regmap_clear_bits(priv->regs, MTK_STAR_REG_INT_MASK, + MTK_STAR_BIT_INT_STS_FNRC); +} + +static void mtk_star_intr_enable_stats(struct mtk_star_priv *priv) +{ + regmap_clear_bits(priv->regs, MTK_STAR_REG_INT_MASK, + MTK_STAR_REG_INT_STS_MIB_CNT_TH); +} + +static void mtk_star_intr_disable_tx(struct mtk_star_priv *priv) +{ + regmap_set_bits(priv->regs, MTK_STAR_REG_INT_MASK, + MTK_STAR_BIT_INT_STS_TNTC); +} + +static void mtk_star_intr_disable_rx(struct mtk_star_priv *priv) +{ + regmap_set_bits(priv->regs, MTK_STAR_REG_INT_MASK, + MTK_STAR_BIT_INT_STS_FNRC); +} + +static void mtk_star_intr_disable_stats(struct mtk_star_priv *priv) +{ + regmap_set_bits(priv->regs, MTK_STAR_REG_INT_MASK, + MTK_STAR_REG_INT_STS_MIB_CNT_TH); +} + +static unsigned int mtk_star_intr_read(struct mtk_star_priv *priv) +{ + unsigned int val; + + regmap_read(priv->regs, MTK_STAR_REG_INT_STS, &val); + + return val; +} + +static unsigned int mtk_star_intr_ack_all(struct mtk_star_priv *priv) +{ + unsigned int val; + + val = mtk_star_intr_read(priv); + regmap_write(priv->regs, MTK_STAR_REG_INT_STS, val); + + return val; +} + +static void mtk_star_dma_init(struct mtk_star_priv *priv) +{ + struct mtk_star_ring_desc *desc; + unsigned int val; + int i; + + priv->descs_base = (struct mtk_star_ring_desc *)priv->ring_base; + + for (i = 0; i < MTK_STAR_NUM_DESCS_TOTAL; i++) { + desc = &priv->descs_base[i]; + + memset(desc, 0, sizeof(*desc)); + desc->status = MTK_STAR_DESC_BIT_COWN; + if ((i == MTK_STAR_NUM_TX_DESCS - 1) || + (i == MTK_STAR_NUM_DESCS_TOTAL - 1)) + desc->status |= MTK_STAR_DESC_BIT_EOR; + } + + mtk_star_ring_init(&priv->tx_ring, priv->descs_base); + mtk_star_ring_init(&priv->rx_ring, + priv->descs_base + MTK_STAR_NUM_TX_DESCS); + + /* Set DMA pointers. */ + val = (unsigned int)priv->dma_addr; + regmap_write(priv->regs, MTK_STAR_REG_TX_BASE_ADDR, val); + regmap_write(priv->regs, MTK_STAR_REG_TX_DPTR, val); + + val += sizeof(struct mtk_star_ring_desc) * MTK_STAR_NUM_TX_DESCS; + regmap_write(priv->regs, MTK_STAR_REG_RX_BASE_ADDR, val); + regmap_write(priv->regs, MTK_STAR_REG_RX_DPTR, val); +} + +static void mtk_star_dma_start(struct mtk_star_priv *priv) +{ + regmap_set_bits(priv->regs, MTK_STAR_REG_TX_DMA_CTRL, + MTK_STAR_BIT_TX_DMA_CTRL_START); + regmap_set_bits(priv->regs, MTK_STAR_REG_RX_DMA_CTRL, + MTK_STAR_BIT_RX_DMA_CTRL_START); +} + +static void mtk_star_dma_stop(struct mtk_star_priv *priv) +{ + regmap_write(priv->regs, MTK_STAR_REG_TX_DMA_CTRL, + MTK_STAR_BIT_TX_DMA_CTRL_STOP); + regmap_write(priv->regs, MTK_STAR_REG_RX_DMA_CTRL, + MTK_STAR_BIT_RX_DMA_CTRL_STOP); +} + +static void mtk_star_dma_disable(struct mtk_star_priv *priv) +{ + int i; + + mtk_star_dma_stop(priv); + + /* Take back all descriptors. */ + for (i = 0; i < MTK_STAR_NUM_DESCS_TOTAL; i++) + priv->descs_base[i].status |= MTK_STAR_DESC_BIT_COWN; +} + +static void mtk_star_dma_resume_rx(struct mtk_star_priv *priv) +{ + regmap_set_bits(priv->regs, MTK_STAR_REG_RX_DMA_CTRL, + MTK_STAR_BIT_RX_DMA_CTRL_RESUME); +} + +static void mtk_star_dma_resume_tx(struct mtk_star_priv *priv) +{ + regmap_set_bits(priv->regs, MTK_STAR_REG_TX_DMA_CTRL, + MTK_STAR_BIT_TX_DMA_CTRL_RESUME); +} + +static void mtk_star_set_mac_addr(struct net_device *ndev) +{ + struct mtk_star_priv *priv = netdev_priv(ndev); + u8 *mac_addr = ndev->dev_addr; + unsigned int high, low; + + high = mac_addr[0] << 8 | mac_addr[1] << 0; + low = mac_addr[2] << 24 | mac_addr[3] << 16 | + mac_addr[4] << 8 | mac_addr[5]; + + regmap_write(priv->regs, MTK_STAR_REG_MY_MAC_H, high); + regmap_write(priv->regs, MTK_STAR_REG_MY_MAC_L, low); +} + +static void mtk_star_reset_counters(struct mtk_star_priv *priv) +{ + static const unsigned int counter_regs[] = { + MTK_STAR_REG_C_RXOKPKT, + MTK_STAR_REG_C_RXOKBYTE, + MTK_STAR_REG_C_RXRUNT, + MTK_STAR_REG_C_RXLONG, + MTK_STAR_REG_C_RXDROP, + MTK_STAR_REG_C_RXCRC, + MTK_STAR_REG_C_RXARLDROP, + MTK_STAR_REG_C_RXVLANDROP, + MTK_STAR_REG_C_RXCSERR, + MTK_STAR_REG_C_RXPAUSE, + MTK_STAR_REG_C_TXOKPKT, + MTK_STAR_REG_C_TXOKBYTE, + MTK_STAR_REG_C_TXPAUSECOL, + MTK_STAR_REG_C_TXRTY, + MTK_STAR_REG_C_TXSKIP, + MTK_STAR_REG_C_TX_ARP, + MTK_STAR_REG_C_RX_RERR, + MTK_STAR_REG_C_RX_UNI, + MTK_STAR_REG_C_RX_MULTI, + MTK_STAR_REG_C_RX_BROAD, + MTK_STAR_REG_C_RX_ALIGNERR, + MTK_STAR_REG_C_TX_UNI, + MTK_STAR_REG_C_TX_MULTI, + MTK_STAR_REG_C_TX_BROAD, + MTK_STAR_REG_C_TX_TIMEOUT, + MTK_STAR_REG_C_TX_LATECOL, + MTK_STAR_REG_C_RX_LENGTHERR, + MTK_STAR_REG_C_RX_TWIST, + }; + + unsigned int i, val; + + for (i = 0; i < ARRAY_SIZE(counter_regs); i++) + regmap_read(priv->regs, counter_regs[i], &val); +} + +static void mtk_star_update_stat(struct mtk_star_priv *priv, + unsigned int reg, u64 *stat) +{ + unsigned int val; + + regmap_read(priv->regs, reg, &val); + *stat += val; +} + +/* Try to get as many stats as possible from the internal registers instead + * of tracking them ourselves. + */ +static void mtk_star_update_stats(struct mtk_star_priv *priv) +{ + struct rtnl_link_stats64 *stats = &priv->stats; + + /* OK packets and bytes. */ + mtk_star_update_stat(priv, MTK_STAR_REG_C_RXOKPKT, &stats->rx_packets); + mtk_star_update_stat(priv, MTK_STAR_REG_C_TXOKPKT, &stats->tx_packets); + mtk_star_update_stat(priv, MTK_STAR_REG_C_RXOKBYTE, &stats->rx_bytes); + mtk_star_update_stat(priv, MTK_STAR_REG_C_TXOKBYTE, &stats->tx_bytes); + + /* RX & TX multicast. */ + mtk_star_update_stat(priv, MTK_STAR_REG_C_RX_MULTI, &stats->multicast); + mtk_star_update_stat(priv, MTK_STAR_REG_C_TX_MULTI, &stats->multicast); + + /* Collisions. */ + mtk_star_update_stat(priv, MTK_STAR_REG_C_TXPAUSECOL, + &stats->collisions); + mtk_star_update_stat(priv, MTK_STAR_REG_C_TX_LATECOL, + &stats->collisions); + mtk_star_update_stat(priv, MTK_STAR_REG_C_RXRUNT, &stats->collisions); + + /* RX Errors. */ + mtk_star_update_stat(priv, MTK_STAR_REG_C_RX_LENGTHERR, + &stats->rx_length_errors); + mtk_star_update_stat(priv, MTK_STAR_REG_C_RXLONG, + &stats->rx_over_errors); + mtk_star_update_stat(priv, MTK_STAR_REG_C_RXCRC, &stats->rx_crc_errors); + mtk_star_update_stat(priv, MTK_STAR_REG_C_RX_ALIGNERR, + &stats->rx_frame_errors); + mtk_star_update_stat(priv, MTK_STAR_REG_C_RXDROP, + &stats->rx_fifo_errors); + /* Sum of the general RX error counter + all of the above. */ + mtk_star_update_stat(priv, MTK_STAR_REG_C_RX_RERR, &stats->rx_errors); + stats->rx_errors += stats->rx_length_errors; + stats->rx_errors += stats->rx_over_errors; + stats->rx_errors += stats->rx_crc_errors; + stats->rx_errors += stats->rx_frame_errors; + stats->rx_errors += stats->rx_fifo_errors; +} + +/* This runs in process context and parallel TX and RX paths executing in + * napi context may result in losing some stats data but this should happen + * seldom enough to be acceptable. + */ +static void mtk_star_update_stats_work(struct work_struct *work) +{ + struct mtk_star_priv *priv = container_of(work, struct mtk_star_priv, + stats_work); + + mtk_star_update_stats(priv); + mtk_star_reset_counters(priv); + mtk_star_intr_enable_stats(priv); +} + +static struct sk_buff *mtk_star_alloc_skb(struct net_device *ndev) +{ + uintptr_t tail, offset; + struct sk_buff *skb; + + skb = dev_alloc_skb(MTK_STAR_MAX_FRAME_SIZE); + if (!skb) + return NULL; + + /* Align to 16 bytes. */ + tail = (uintptr_t)skb_tail_pointer(skb); + if (tail & (MTK_STAR_SKB_ALIGNMENT - 1)) { + offset = tail & (MTK_STAR_SKB_ALIGNMENT - 1); + skb_reserve(skb, MTK_STAR_SKB_ALIGNMENT - offset); + } + + /* Ensure 16-byte alignment of the skb pointer: eth_type_trans() will + * extract the Ethernet header (14 bytes) so we need two more bytes. + */ + skb_reserve(skb, MTK_STAR_IP_ALIGN); + + return skb; +} + +static int mtk_star_prepare_rx_skbs(struct net_device *ndev) +{ + struct mtk_star_priv *priv = netdev_priv(ndev); + struct mtk_star_ring *ring = &priv->rx_ring; + struct device *dev = mtk_star_get_dev(priv); + struct mtk_star_ring_desc *desc; + struct sk_buff *skb; + dma_addr_t dma_addr; + int i; + + for (i = 0; i < MTK_STAR_NUM_RX_DESCS; i++) { + skb = mtk_star_alloc_skb(ndev); + if (!skb) + return -ENOMEM; + + dma_addr = mtk_star_dma_map_rx(priv, skb); + if (dma_mapping_error(dev, dma_addr)) { + dev_kfree_skb(skb); + return -ENOMEM; + } + + desc = &ring->descs[i]; + desc->data_ptr = dma_addr; + desc->status |= skb_tailroom(skb) & MTK_STAR_DESC_MSK_LEN; + desc->status &= ~MTK_STAR_DESC_BIT_COWN; + ring->skbs[i] = skb; + ring->dma_addrs[i] = dma_addr; + } + + return 0; +} + +static void +mtk_star_ring_free_skbs(struct mtk_star_priv *priv, struct mtk_star_ring *ring, + void (*unmap_func)(struct mtk_star_priv *, + struct mtk_star_ring_desc_data *)) +{ + struct mtk_star_ring_desc_data desc_data; + int i; + + for (i = 0; i < MTK_STAR_RING_NUM_DESCS; i++) { + if (!ring->dma_addrs[i]) + continue; + + desc_data.dma_addr = ring->dma_addrs[i]; + desc_data.skb = ring->skbs[i]; + + unmap_func(priv, &desc_data); + dev_kfree_skb(desc_data.skb); + } +} + +static void mtk_star_free_rx_skbs(struct mtk_star_priv *priv) +{ + struct mtk_star_ring *ring = &priv->rx_ring; + + mtk_star_ring_free_skbs(priv, ring, mtk_star_dma_unmap_rx); +} + +static void mtk_star_free_tx_skbs(struct mtk_star_priv *priv) +{ + struct mtk_star_ring *ring = &priv->tx_ring; + + mtk_star_ring_free_skbs(priv, ring, mtk_star_dma_unmap_tx); +} + +/* All processing for TX and RX happens in the napi poll callback. */ +static irqreturn_t mtk_star_handle_irq(int irq, void *data) +{ + struct mtk_star_priv *priv; + struct net_device *ndev; + bool need_napi = false; + unsigned int status; + + ndev = data; + priv = netdev_priv(ndev); + + if (netif_running(ndev)) { + status = mtk_star_intr_read(priv); + + if (status & MTK_STAR_BIT_INT_STS_TNTC) { + mtk_star_intr_disable_tx(priv); + need_napi = true; + } + + if (status & MTK_STAR_BIT_INT_STS_FNRC) { + mtk_star_intr_disable_rx(priv); + need_napi = true; + } + + if (need_napi) + napi_schedule(&priv->napi); + + /* One of the counters reached 0x8000000 - update stats and + * reset all counters. + */ + if (unlikely(status & MTK_STAR_REG_INT_STS_MIB_CNT_TH)) { + mtk_star_intr_disable_stats(priv); + schedule_work(&priv->stats_work); + } + + mtk_star_intr_ack_all(priv); + } + + return IRQ_HANDLED; +} + +/* Wait for the completion of any previous command - CMD_START bit must be + * cleared by hardware. + */ +static int mtk_star_hash_wait_cmd_start(struct mtk_star_priv *priv) +{ + unsigned int val; + + return regmap_read_poll_timeout_atomic(priv->regs, + MTK_STAR_REG_HASH_CTRL, val, + !(val & MTK_STAR_BIT_HASH_CTRL_CMD_START), + 10, MTK_STAR_WAIT_TIMEOUT); +} + +static int mtk_star_hash_wait_ok(struct mtk_star_priv *priv) +{ + unsigned int val; + int ret; + + /* Wait for BIST_DONE bit. */ + ret = regmap_read_poll_timeout_atomic(priv->regs, + MTK_STAR_REG_HASH_CTRL, val, + val & MTK_STAR_BIT_HASH_CTRL_BIST_DONE, + 10, MTK_STAR_WAIT_TIMEOUT); + if (ret) + return ret; + + /* Check the BIST_OK bit. */ + if (!regmap_test_bits(priv->regs, MTK_STAR_REG_HASH_CTRL, + MTK_STAR_BIT_HASH_CTRL_BIST_OK)) + return -EIO; + + return 0; +} + +static int mtk_star_set_hashbit(struct mtk_star_priv *priv, + unsigned int hash_addr) +{ + unsigned int val; + int ret; + + ret = mtk_star_hash_wait_cmd_start(priv); + if (ret) + return ret; + + val = hash_addr & MTK_STAR_MSK_HASH_CTRL_HASH_BIT_ADDR; + val |= MTK_STAR_BIT_HASH_CTRL_ACC_CMD; + val |= MTK_STAR_BIT_HASH_CTRL_CMD_START; + val |= MTK_STAR_BIT_HASH_CTRL_BIST_EN; + val |= MTK_STAR_BIT_HASH_CTRL_HASH_BIT_DATA; + regmap_write(priv->regs, MTK_STAR_REG_HASH_CTRL, val); + + return mtk_star_hash_wait_ok(priv); +} + +static int mtk_star_reset_hash_table(struct mtk_star_priv *priv) +{ + int ret; + + ret = mtk_star_hash_wait_cmd_start(priv); + if (ret) + return ret; + + regmap_set_bits(priv->regs, MTK_STAR_REG_HASH_CTRL, + MTK_STAR_BIT_HASH_CTRL_BIST_EN); + regmap_set_bits(priv->regs, MTK_STAR_REG_TEST1, + MTK_STAR_BIT_TEST1_RST_HASH_MBIST); + + return mtk_star_hash_wait_ok(priv); +} + +static void mtk_star_phy_config(struct mtk_star_priv *priv) +{ + unsigned int val; + + if (priv->speed == SPEED_1000) + val = MTK_STAR_VAL_PHY_CTRL1_FORCE_SPD_1000M; + else if (priv->speed == SPEED_100) + val = MTK_STAR_VAL_PHY_CTRL1_FORCE_SPD_100M; + else + val = MTK_STAR_VAL_PHY_CTRL1_FORCE_SPD_10M; + val <<= MTK_STAR_OFF_PHY_CTRL1_FORCE_SPD; + + val |= MTK_STAR_BIT_PHY_CTRL1_AN_EN; + val |= MTK_STAR_BIT_PHY_CTRL1_FORCE_FC_RX; + val |= MTK_STAR_BIT_PHY_CTRL1_FORCE_FC_TX; + /* Only full-duplex supported for now. */ + val |= MTK_STAR_BIT_PHY_CTRL1_FORCE_DPX; + + regmap_write(priv->regs, MTK_STAR_REG_PHY_CTRL1, val); + + if (priv->pause) { + val = MTK_STAR_VAL_FC_CFG_SEND_PAUSE_TH_2K; + val <<= MTK_STAR_OFF_FC_CFG_SEND_PAUSE_TH; + val |= MTK_STAR_BIT_FC_CFG_UC_PAUSE_DIR; + } else { + val = 0; + } + + regmap_update_bits(priv->regs, MTK_STAR_REG_FC_CFG, + MTK_STAR_MSK_FC_CFG_SEND_PAUSE_TH | + MTK_STAR_BIT_FC_CFG_UC_PAUSE_DIR, val); + + if (priv->pause) { + val = MTK_STAR_VAL_EXT_CFG_SND_PAUSE_RLS_1K; + val <<= MTK_STAR_OFF_EXT_CFG_SND_PAUSE_RLS; + } else { + val = 0; + } + + regmap_update_bits(priv->regs, MTK_STAR_REG_EXT_CFG, + MTK_STAR_MSK_EXT_CFG_SND_PAUSE_RLS, val); +} + +static void mtk_star_adjust_link(struct net_device *ndev) +{ + struct mtk_star_priv *priv = netdev_priv(ndev); + struct phy_device *phydev = priv->phydev; + bool new_state = false; + + if (phydev->link) { + if (!priv->link) { + priv->link = phydev->link; + new_state = true; + } + + if (priv->speed != phydev->speed) { + priv->speed = phydev->speed; + new_state = true; + } + + if (priv->pause != phydev->pause) { + priv->pause = phydev->pause; + new_state = true; + } + } else { + if (priv->link) { + priv->link = phydev->link; + new_state = true; + } + } + + if (new_state) { + if (phydev->link) + mtk_star_phy_config(priv); + + phy_print_status(ndev->phydev); + } +} + +static void mtk_star_init_config(struct mtk_star_priv *priv) +{ + unsigned int val; + + val = (MTK_STAR_BIT_MII_PAD_OUT_ENABLE | + MTK_STAR_BIT_EXT_MDC_MODE | + MTK_STAR_BIT_SWC_MII_MODE); + + regmap_write(priv->regs, MTK_STAR_REG_SYS_CONF, val); + regmap_update_bits(priv->regs, MTK_STAR_REG_MAC_CLK_CONF, + MTK_STAR_MSK_MAC_CLK_CONF, + MTK_STAR_BIT_CLK_DIV_10); +} + +static void mtk_star_set_mode_rmii(struct mtk_star_priv *priv) +{ + regmap_update_bits(priv->pericfg, MTK_PERICFG_REG_NIC_CFG_CON, + MTK_PERICFG_MSK_NIC_CFG_CON_CFG_MII, + MTK_PERICFG_BIT_NIC_CFG_CON_RMII); +} + +static int mtk_star_enable(struct net_device *ndev) +{ + struct mtk_star_priv *priv = netdev_priv(ndev); + unsigned int val; + int ret; + + mtk_star_nic_disable_pd(priv); + mtk_star_intr_disable(priv); + mtk_star_dma_stop(priv); + + mtk_star_set_mac_addr(ndev); + + /* Configure the MAC */ + val = MTK_STAR_VAL_MAC_CFG_IPG_96BIT; + val <<= MTK_STAR_OFF_MAC_CFG_IPG; + val |= MTK_STAR_BIT_MAC_CFG_MAXLEN_1522; + val |= MTK_STAR_BIT_MAC_CFG_AUTO_PAD; + val |= MTK_STAR_BIT_MAC_CFG_CRC_STRIP; + regmap_write(priv->regs, MTK_STAR_REG_MAC_CFG, val); + + /* Enable Hash Table BIST and reset it */ + ret = mtk_star_reset_hash_table(priv); + if (ret) + return ret; + + /* Setup the hashing algorithm */ + regmap_clear_bits(priv->regs, MTK_STAR_REG_ARL_CFG, + MTK_STAR_BIT_ARL_CFG_HASH_ALG | + MTK_STAR_BIT_ARL_CFG_MISC_MODE); + + /* Don't strip VLAN tags */ + regmap_clear_bits(priv->regs, MTK_STAR_REG_MAC_CFG, + MTK_STAR_BIT_MAC_CFG_VLAN_STRIP); + + /* Setup DMA */ + mtk_star_dma_init(priv); + + ret = mtk_star_prepare_rx_skbs(ndev); + if (ret) + goto err_out; + + /* Request the interrupt */ + ret = request_irq(ndev->irq, mtk_star_handle_irq, + IRQF_TRIGGER_FALLING, ndev->name, ndev); + if (ret) + goto err_free_skbs; + + napi_enable(&priv->napi); + + mtk_star_intr_ack_all(priv); + mtk_star_intr_enable(priv); + + /* Connect to and start PHY */ + priv->phydev = of_phy_connect(ndev, priv->phy_node, + mtk_star_adjust_link, 0, priv->phy_intf); + if (!priv->phydev) { + netdev_err(ndev, "failed to connect to PHY\n"); + goto err_free_irq; + } + + mtk_star_dma_start(priv); + phy_start(priv->phydev); + netif_start_queue(ndev); + + return 0; + +err_free_irq: + free_irq(ndev->irq, ndev); +err_free_skbs: + mtk_star_free_rx_skbs(priv); +err_out: + return ret; +} + +static void mtk_star_disable(struct net_device *ndev) +{ + struct mtk_star_priv *priv = netdev_priv(ndev); + + netif_stop_queue(ndev); + napi_disable(&priv->napi); + mtk_star_intr_disable(priv); + mtk_star_dma_disable(priv); + mtk_star_intr_ack_all(priv); + phy_stop(priv->phydev); + phy_disconnect(priv->phydev); + free_irq(ndev->irq, ndev); + mtk_star_free_rx_skbs(priv); + mtk_star_free_tx_skbs(priv); +} + +static int mtk_star_netdev_open(struct net_device *ndev) +{ + return mtk_star_enable(ndev); +} + +static int mtk_star_netdev_stop(struct net_device *ndev) +{ + mtk_star_disable(ndev); + + return 0; +} + +static int mtk_star_netdev_ioctl(struct net_device *ndev, + struct ifreq *req, int cmd) +{ + if (!netif_running(ndev)) + return -EINVAL; + + return phy_mii_ioctl(ndev->phydev, req, cmd); +} + +static int mtk_star_netdev_start_xmit(struct sk_buff *skb, + struct net_device *ndev) +{ + struct mtk_star_priv *priv = netdev_priv(ndev); + struct mtk_star_ring *ring = &priv->tx_ring; + struct device *dev = mtk_star_get_dev(priv); + struct mtk_star_ring_desc_data desc_data; + + desc_data.dma_addr = mtk_star_dma_map_tx(priv, skb); + if (dma_mapping_error(dev, desc_data.dma_addr)) + goto err_drop_packet; + + desc_data.skb = skb; + desc_data.len = skb->len; + + spin_lock_bh(&priv->lock); + + mtk_star_ring_push_head_tx(ring, &desc_data); + + netdev_sent_queue(ndev, skb->len); + + if (mtk_star_ring_full(ring)) + netif_stop_queue(ndev); + + spin_unlock_bh(&priv->lock); + + mtk_star_dma_resume_tx(priv); + + return NETDEV_TX_OK; + +err_drop_packet: + dev_kfree_skb(skb); + ndev->stats.tx_dropped++; + return NETDEV_TX_BUSY; +} + +/* Returns the number of bytes sent or a negative number on the first + * descriptor owned by DMA. + */ +static int mtk_star_tx_complete_one(struct mtk_star_priv *priv) +{ + struct mtk_star_ring *ring = &priv->tx_ring; + struct mtk_star_ring_desc_data desc_data; + int ret; + + ret = mtk_star_ring_pop_tail(ring, &desc_data); + if (ret) + return ret; + + mtk_star_dma_unmap_tx(priv, &desc_data); + ret = desc_data.skb->len; + dev_kfree_skb_irq(desc_data.skb); + + return ret; +} + +static void mtk_star_tx_complete_all(struct mtk_star_priv *priv) +{ + struct mtk_star_ring *ring = &priv->tx_ring; + struct net_device *ndev = priv->ndev; + int ret, pkts_compl, bytes_compl; + bool wake = false; + + spin_lock(&priv->lock); + + for (pkts_compl = 0, bytes_compl = 0;; + pkts_compl++, bytes_compl += ret, wake = true) { + if (!mtk_star_ring_descs_available(ring)) + break; + + ret = mtk_star_tx_complete_one(priv); + if (ret < 0) + break; + } + + netdev_completed_queue(ndev, pkts_compl, bytes_compl); + + if (wake && netif_queue_stopped(ndev)) + netif_wake_queue(ndev); + + mtk_star_intr_enable_tx(priv); + + spin_unlock(&priv->lock); +} + +static void mtk_star_netdev_get_stats64(struct net_device *ndev, + struct rtnl_link_stats64 *stats) +{ + struct mtk_star_priv *priv = netdev_priv(ndev); + + mtk_star_update_stats(priv); + + memcpy(stats, &priv->stats, sizeof(*stats)); +} + +static void mtk_star_set_rx_mode(struct net_device *ndev) +{ + struct mtk_star_priv *priv = netdev_priv(ndev); + struct netdev_hw_addr *hw_addr; + unsigned int hash_addr, i; + int ret; + + if (ndev->flags & IFF_PROMISC) { + regmap_set_bits(priv->regs, MTK_STAR_REG_ARL_CFG, + MTK_STAR_BIT_ARL_CFG_MISC_MODE); + } else if (netdev_mc_count(ndev) > MTK_STAR_HASHTABLE_MC_LIMIT || + ndev->flags & IFF_ALLMULTI) { + for (i = 0; i < MTK_STAR_HASHTABLE_SIZE_MAX; i++) { + ret = mtk_star_set_hashbit(priv, i); + if (ret) + goto hash_fail; + } + } else { + /* Clear previous settings. */ + ret = mtk_star_reset_hash_table(priv); + if (ret) + goto hash_fail; + + netdev_for_each_mc_addr(hw_addr, ndev) { + hash_addr = (hw_addr->addr[0] & 0x01) << 8; + hash_addr += hw_addr->addr[5]; + ret = mtk_star_set_hashbit(priv, hash_addr); + if (ret) + goto hash_fail; + } + } + + return; + +hash_fail: + if (ret == -ETIMEDOUT) + netdev_err(ndev, "setting hash bit timed out\n"); + else + /* Should be -EIO */ + netdev_err(ndev, "unable to set hash bit"); +} + +static const struct net_device_ops mtk_star_netdev_ops = { + .ndo_open = mtk_star_netdev_open, + .ndo_stop = mtk_star_netdev_stop, + .ndo_start_xmit = mtk_star_netdev_start_xmit, + .ndo_get_stats64 = mtk_star_netdev_get_stats64, + .ndo_set_rx_mode = mtk_star_set_rx_mode, + .ndo_do_ioctl = mtk_star_netdev_ioctl, + .ndo_set_mac_address = eth_mac_addr, + .ndo_validate_addr = eth_validate_addr, +}; + +static void mtk_star_get_drvinfo(struct net_device *dev, + struct ethtool_drvinfo *info) +{ + strlcpy(info->driver, MTK_STAR_DRVNAME, sizeof(info->driver)); +} + +/* TODO Add ethtool stats. */ +static const struct ethtool_ops mtk_star_ethtool_ops = { + .get_drvinfo = mtk_star_get_drvinfo, + .get_link = ethtool_op_get_link, + .get_link_ksettings = phy_ethtool_get_link_ksettings, + .set_link_ksettings = phy_ethtool_set_link_ksettings, +}; + +static int mtk_star_receive_packet(struct mtk_star_priv *priv) +{ + struct mtk_star_ring *ring = &priv->rx_ring; + struct device *dev = mtk_star_get_dev(priv); + struct mtk_star_ring_desc_data desc_data; + struct net_device *ndev = priv->ndev; + struct sk_buff *curr_skb, *new_skb; + dma_addr_t new_dma_addr; + int ret; + + spin_lock(&priv->lock); + ret = mtk_star_ring_pop_tail(ring, &desc_data); + spin_unlock(&priv->lock); + if (ret) + return -1; + + curr_skb = desc_data.skb; + + if ((desc_data.flags & MTK_STAR_DESC_BIT_RX_CRCE) || + (desc_data.flags & MTK_STAR_DESC_BIT_RX_OSIZE)) { + /* Error packet -> drop and reuse skb. */ + new_skb = curr_skb; + goto push_new_skb; + } + + /* Prepare new skb before receiving the current one. Reuse the current + * skb if we fail at any point. + */ + new_skb = mtk_star_alloc_skb(ndev); + if (!new_skb) { + ndev->stats.rx_dropped++; + new_skb = curr_skb; + goto push_new_skb; + } + + new_dma_addr = mtk_star_dma_map_rx(priv, new_skb); + if (dma_mapping_error(dev, new_dma_addr)) { + ndev->stats.rx_dropped++; + dev_kfree_skb(new_skb); + new_skb = curr_skb; + netdev_err(ndev, "DMA mapping error of RX descriptor\n"); + goto push_new_skb; + } + + desc_data.dma_addr = new_dma_addr; + + /* We can't fail anymore at this point: it's safe to unmap the skb. */ + mtk_star_dma_unmap_rx(priv, &desc_data); + + skb_put(desc_data.skb, desc_data.len); + desc_data.skb->ip_summed = CHECKSUM_NONE; + desc_data.skb->protocol = eth_type_trans(desc_data.skb, ndev); + desc_data.skb->dev = ndev; + netif_receive_skb(desc_data.skb); + +push_new_skb: + desc_data.len = skb_tailroom(new_skb); + desc_data.skb = new_skb; + + spin_lock(&priv->lock); + mtk_star_ring_push_head_rx(ring, &desc_data); + spin_unlock(&priv->lock); + + return 0; +} + +static int mtk_star_process_rx(struct mtk_star_priv *priv, int budget) +{ + int received, ret; + + for (received = 0, ret = 0; received < budget && ret == 0; received++) + ret = mtk_star_receive_packet(priv); + + mtk_star_dma_resume_rx(priv); + + return received; +} + +static int mtk_star_poll(struct napi_struct *napi, int budget) +{ + struct mtk_star_priv *priv; + int received = 0; + + priv = container_of(napi, struct mtk_star_priv, napi); + + /* Clean-up all TX descriptors. */ + mtk_star_tx_complete_all(priv); + /* Receive up to $budget packets. */ + received = mtk_star_process_rx(priv, budget); + + if (received < budget) { + napi_complete_done(napi, received); + mtk_star_intr_enable_rx(priv); + } + + return received; +} + +static void mtk_star_mdio_rwok_clear(struct mtk_star_priv *priv) +{ + regmap_write(priv->regs, MTK_STAR_REG_PHY_CTRL0, + MTK_STAR_BIT_PHY_CTRL0_RWOK); +} + +static int mtk_star_mdio_rwok_wait(struct mtk_star_priv *priv) +{ + unsigned int val; + + return regmap_read_poll_timeout(priv->regs, MTK_STAR_REG_PHY_CTRL0, + val, val & MTK_STAR_BIT_PHY_CTRL0_RWOK, + 10, MTK_STAR_WAIT_TIMEOUT); +} + +static int mtk_star_mdio_read(struct mii_bus *mii, int phy_id, int regnum) +{ + struct mtk_star_priv *priv = mii->priv; + unsigned int val, data; + int ret; + + if (regnum & MII_ADDR_C45) + return -EOPNOTSUPP; + + mtk_star_mdio_rwok_clear(priv); + + val = (regnum << MTK_STAR_OFF_PHY_CTRL0_PREG); + val &= MTK_STAR_MSK_PHY_CTRL0_PREG; + val |= MTK_STAR_BIT_PHY_CTRL0_RDCMD; + + regmap_write(priv->regs, MTK_STAR_REG_PHY_CTRL0, val); + + ret = mtk_star_mdio_rwok_wait(priv); + if (ret) + return ret; + + regmap_read(priv->regs, MTK_STAR_REG_PHY_CTRL0, &data); + + data &= MTK_STAR_MSK_PHY_CTRL0_RWDATA; + data >>= MTK_STAR_OFF_PHY_CTRL0_RWDATA; + + return data; +} + +static int mtk_star_mdio_write(struct mii_bus *mii, int phy_id, + int regnum, u16 data) +{ + struct mtk_star_priv *priv = mii->priv; + unsigned int val; + + if (regnum & MII_ADDR_C45) + return -EOPNOTSUPP; + + mtk_star_mdio_rwok_clear(priv); + + val = data; + val <<= MTK_STAR_OFF_PHY_CTRL0_RWDATA; + val &= MTK_STAR_MSK_PHY_CTRL0_RWDATA; + regnum <<= MTK_STAR_OFF_PHY_CTRL0_PREG; + regnum &= MTK_STAR_MSK_PHY_CTRL0_PREG; + val |= regnum; + val |= MTK_STAR_BIT_PHY_CTRL0_WTCMD; + + regmap_write(priv->regs, MTK_STAR_REG_PHY_CTRL0, val); + + return mtk_star_mdio_rwok_wait(priv); +} + +static int mtk_star_mdio_init(struct net_device *ndev) +{ + struct mtk_star_priv *priv = netdev_priv(ndev); + struct device *dev = mtk_star_get_dev(priv); + struct device_node *of_node, *mdio_node; + int ret; + + of_node = dev->of_node; + + mdio_node = of_get_child_by_name(of_node, "mdio"); + if (!mdio_node) + return -ENODEV; + + if (!of_device_is_available(mdio_node)) { + ret = -ENODEV; + goto out_put_node; + } + + priv->mii = devm_mdiobus_alloc(dev); + if (!priv->mii) { + ret = -ENOMEM; + goto out_put_node; + } + + snprintf(priv->mii->id, MII_BUS_ID_SIZE, "%s", dev_name(dev)); + priv->mii->name = "mtk-mac-mdio"; + priv->mii->parent = dev; + priv->mii->read = mtk_star_mdio_read; + priv->mii->write = mtk_star_mdio_write; + priv->mii->priv = priv; + + ret = of_mdiobus_register(priv->mii, mdio_node); + +out_put_node: + of_node_put(mdio_node); + return ret; +} + +static __maybe_unused int mtk_star_suspend(struct device *dev) +{ + struct mtk_star_priv *priv; + struct net_device *ndev; + + ndev = dev_get_drvdata(dev); + priv = netdev_priv(ndev); + + if (netif_running(ndev)) + mtk_star_disable(ndev); + + clk_bulk_disable_unprepare(MTK_STAR_NCLKS, priv->clks); + + return 0; +} + +static __maybe_unused int mtk_star_resume(struct device *dev) +{ + struct mtk_star_priv *priv; + struct net_device *ndev; + int ret; + + ndev = dev_get_drvdata(dev); + priv = netdev_priv(ndev); + + ret = clk_bulk_prepare_enable(MTK_STAR_NCLKS, priv->clks); + if (ret) + return ret; + + if (netif_running(ndev)) { + ret = mtk_star_enable(ndev); + if (ret) + clk_bulk_disable_unprepare(MTK_STAR_NCLKS, priv->clks); + } + + return ret; +} + +static void mtk_star_clk_disable_unprepare(void *data) +{ + struct mtk_star_priv *priv = data; + + clk_bulk_disable_unprepare(MTK_STAR_NCLKS, priv->clks); +} + +static void mtk_star_mdiobus_unregister(void *data) +{ + struct mtk_star_priv *priv = data; + + mdiobus_unregister(priv->mii); +} + +static int mtk_star_probe(struct platform_device *pdev) +{ + struct device_node *of_node; + struct mtk_star_priv *priv; + struct net_device *ndev; + struct device *dev; + void __iomem *base; + int ret, i; + + dev = &pdev->dev; + of_node = dev->of_node; + + ndev = devm_alloc_etherdev(dev, sizeof(*priv)); + if (!ndev) + return -ENOMEM; + + priv = netdev_priv(ndev); + priv->ndev = ndev; + SET_NETDEV_DEV(ndev, dev); + platform_set_drvdata(pdev, ndev); + + ndev->min_mtu = ETH_ZLEN; + ndev->max_mtu = MTK_STAR_MAX_FRAME_SIZE; + + spin_lock_init(&priv->lock); + INIT_WORK(&priv->stats_work, mtk_star_update_stats_work); + + base = devm_platform_ioremap_resource(pdev, 0); + if (IS_ERR(base)) + return PTR_ERR(base); + + /* We won't be checking the return values of regmap read & write + * functions. They can only fail for mmio if there's a clock attached + * to regmap which is not the case here. + */ + priv->regs = devm_regmap_init_mmio(dev, base, + &mtk_star_regmap_config); + if (IS_ERR(priv->regs)) + return PTR_ERR(priv->regs); + + priv->pericfg = syscon_regmap_lookup_by_phandle(of_node, + "mediatek,pericfg"); + if (IS_ERR(priv->pericfg)) { + dev_err(dev, "Failed to lookup the PERICFG syscon\n"); + return PTR_ERR(priv->pericfg); + } + + ndev->irq = platform_get_irq(pdev, 0); + if (ndev->irq < 0) + return ndev->irq; + + for (i = 0; i < MTK_STAR_NCLKS; i++) + priv->clks[i].id = mtk_star_clk_names[i]; + ret = devm_clk_bulk_get(dev, MTK_STAR_NCLKS, priv->clks); + if (ret) + return ret; + + ret = clk_bulk_prepare_enable(MTK_STAR_NCLKS, priv->clks); + if (ret) + return ret; + + ret = devm_add_action_or_reset(dev, + mtk_star_clk_disable_unprepare, priv); + if (ret) + return ret; + + ret = of_get_phy_mode(of_node, &priv->phy_intf); + if (ret) { + return ret; + } else if (priv->phy_intf != PHY_INTERFACE_MODE_RMII) { + dev_err(dev, "unsupported phy mode: %s\n", + phy_modes(priv->phy_intf)); + return -EINVAL; + } + + priv->phy_node = of_parse_phandle(of_node, "phy-handle", 0); + if (!priv->phy_node) { + dev_err(dev, "failed to retrieve the phy handle from device tree\n"); + return -ENODEV; + } + + mtk_star_set_mode_rmii(priv); + + ret = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(32)); + if (ret) { + dev_err(dev, "unsupported DMA mask\n"); + return ret; + } + + priv->ring_base = dmam_alloc_coherent(dev, MTK_STAR_DMA_SIZE, + &priv->dma_addr, + GFP_KERNEL | GFP_DMA); + if (!priv->ring_base) + return -ENOMEM; + + mtk_star_nic_disable_pd(priv); + mtk_star_init_config(priv); + + ret = mtk_star_mdio_init(ndev); + if (ret) + return ret; + + ret = devm_add_action_or_reset(dev, mtk_star_mdiobus_unregister, priv); + if (ret) + return ret; + + ret = eth_platform_get_mac_address(dev, ndev->dev_addr); + if (ret || !is_valid_ether_addr(ndev->dev_addr)) + eth_hw_addr_random(ndev); + + ndev->netdev_ops = &mtk_star_netdev_ops; + ndev->ethtool_ops = &mtk_star_ethtool_ops; + + netif_napi_add(ndev, &priv->napi, mtk_star_poll, MTK_STAR_NAPI_WEIGHT); + + return devm_register_netdev(dev, ndev); +} + +static const struct of_device_id mtk_star_of_match[] = { + { .compatible = "mediatek,mt8516-eth", }, + { .compatible = "mediatek,mt8518-eth", }, + { .compatible = "mediatek,mt8175-eth", }, + { } +}; +MODULE_DEVICE_TABLE(of, mtk_star_of_match); + +static SIMPLE_DEV_PM_OPS(mtk_star_pm_ops, + mtk_star_suspend, mtk_star_resume); + +static struct platform_driver mtk_star_driver = { + .driver = { + .name = MTK_STAR_DRVNAME, + .pm = &mtk_star_pm_ops, + .of_match_table = of_match_ptr(mtk_star_of_match), + }, + .probe = mtk_star_probe, +}; +module_platform_driver(mtk_star_driver); + +MODULE_AUTHOR("Bartosz Golaszewski <bgolaszewski@baylibre.com>"); +MODULE_DESCRIPTION("Mediatek STAR Ethernet MAC Driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/ethernet/mellanox/mlx4/crdump.c b/drivers/net/ethernet/mellanox/mlx4/crdump.c index 73eae80e1cb7..ac5468b77488 100644 --- a/drivers/net/ethernet/mellanox/mlx4/crdump.c +++ b/drivers/net/ethernet/mellanox/mlx4/crdump.c @@ -197,6 +197,7 @@ int mlx4_crdump_collect(struct mlx4_dev *dev) err = devlink_region_snapshot_id_get(devlink, &id); if (err) { mlx4_err(dev, "crdump: devlink get snapshot id err %d\n", err); + iounmap(cr_space); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index 8a5ea2543670..b816154bc79a 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -1235,7 +1235,6 @@ static int mlx4_en_get_rxfh(struct net_device *dev, u32 *ring_index, u8 *key, struct mlx4_en_priv *priv = netdev_priv(dev); u32 n = mlx4_en_get_rxfh_indir_size(dev); u32 i, rss_rings; - int err = 0; rss_rings = priv->prof->rss_rings ?: n; rss_rings = rounddown_pow_of_two(rss_rings); @@ -1249,7 +1248,7 @@ static int mlx4_en_get_rxfh(struct net_device *dev, u32 *ring_index, u8 *key, memcpy(key, priv->rss_key, MLX4_EN_RSS_KEY_SIZE); if (hfunc) *hfunc = priv->rss_hash_fn; - return err; + return 0; } static int mlx4_en_set_rxfh(struct net_device *dev, const u32 *ring_index, @@ -1393,7 +1392,6 @@ static int mlx4_en_ethtool_add_mac_rule(struct ethtool_rxnfc *cmd, struct mlx4_spec_list *spec_l2, unsigned char *mac) { - int err = 0; __be64 mac_msk = cpu_to_be64(MLX4_MAC_MASK << 16); spec_l2->id = MLX4_NET_TRANS_RULE_ID_ETH; @@ -1408,7 +1406,7 @@ static int mlx4_en_ethtool_add_mac_rule(struct ethtool_rxnfc *cmd, list_add_tail(&spec_l2->list, rule_list_h); - return err; + return 0; } static int mlx4_en_ethtool_add_mac_rule_by_ipv4(struct mlx4_en_priv *priv, diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c index 43dcbd8214c6..5bd3cd37d50f 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c @@ -51,7 +51,8 @@ #include "en_port.h" #define MLX4_EN_MAX_XDP_MTU ((int)(PAGE_SIZE - ETH_HLEN - (2 * VLAN_HLEN) - \ - XDP_PACKET_HEADROOM)) + XDP_PACKET_HEADROOM - \ + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)))) int mlx4_en_setup_tc(struct net_device *dev, u8 up) { diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index db3552f2d087..8a10285b0e10 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -683,6 +683,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud rcu_read_lock(); xdp_prog = rcu_dereference(ring->xdp_prog); xdp.rxq = &ring->xdp_rxq; + xdp.frame_sz = priv->frag_info[0].frag_stride; doorbell_pending = 0; /* We assume a 1:1 mapping between CQEs and Rx descriptors, so Rx @@ -946,7 +947,7 @@ int mlx4_en_poll_rx_cq(struct napi_struct *napi, int budget) xdp_tx_cq = priv->tx_cq[TX_XDP][cq->ring]; if (xdp_tx_cq->xdp_busy) { clean_complete = mlx4_en_process_tx_cq(dev, xdp_tx_cq, - budget); + budget) < budget; xdp_tx_cq->xdp_busy = !clean_complete; } } diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c index a30edb436f4a..9dff7b086c9f 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c @@ -392,8 +392,8 @@ int mlx4_en_free_tx_buf(struct net_device *dev, struct mlx4_en_tx_ring *ring) return cnt; } -bool mlx4_en_process_tx_cq(struct net_device *dev, - struct mlx4_en_cq *cq, int napi_budget) +int mlx4_en_process_tx_cq(struct net_device *dev, + struct mlx4_en_cq *cq, int napi_budget) { struct mlx4_en_priv *priv = netdev_priv(dev); struct mlx4_cq *mcq = &cq->mcq; @@ -415,7 +415,7 @@ bool mlx4_en_process_tx_cq(struct net_device *dev, u32 ring_cons; if (unlikely(!priv->port_up)) - return true; + return 0; netdev_txq_bql_complete_prefetchw(ring->tx_queue); @@ -492,7 +492,7 @@ bool mlx4_en_process_tx_cq(struct net_device *dev, WRITE_ONCE(ring->cons, ring_cons + txbbs_skipped); if (cq->type == TX_XDP) - return done < budget; + return done; netdev_tx_completed_queue(ring->tx_queue, packets, bytes); @@ -504,7 +504,7 @@ bool mlx4_en_process_tx_cq(struct net_device *dev, ring->wake_queue++; } - return done < budget; + return done; } void mlx4_en_tx_irq(struct mlx4_cq *mcq) @@ -524,14 +524,14 @@ int mlx4_en_poll_tx_cq(struct napi_struct *napi, int budget) struct mlx4_en_cq *cq = container_of(napi, struct mlx4_en_cq, napi); struct net_device *dev = cq->dev; struct mlx4_en_priv *priv = netdev_priv(dev); - bool clean_complete; + int work_done; - clean_complete = mlx4_en_process_tx_cq(dev, cq, budget); - if (!clean_complete) + work_done = mlx4_en_process_tx_cq(dev, cq, budget); + if (work_done >= budget) return budget; - napi_complete(napi); - mlx4_en_arm_cq(priv, cq); + if (napi_complete_done(napi, work_done)) + mlx4_en_arm_cq(priv, cq); return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx4/fw.c b/drivers/net/ethernet/mellanox/mlx4/fw.c index 6e501af0e532..f6ff9620a137 100644 --- a/drivers/net/ethernet/mellanox/mlx4/fw.c +++ b/drivers/net/ethernet/mellanox/mlx4/fw.c @@ -2734,7 +2734,7 @@ void mlx4_opreq_action(struct work_struct *work) if (err) { mlx4_err(dev, "Failed to retrieve required operation: %d\n", err); - return; + goto out; } MLX4_GET(modifier, outbox, GET_OP_REQ_MODIFIER_OFFSET); MLX4_GET(token, outbox, GET_OP_REQ_TOKEN_OFFSET); diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index 630f15977f09..9f5603612960 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -737,8 +737,8 @@ int mlx4_en_process_rx_cq(struct net_device *dev, int budget); int mlx4_en_poll_rx_cq(struct napi_struct *napi, int budget); int mlx4_en_poll_tx_cq(struct napi_struct *napi, int budget); -bool mlx4_en_process_tx_cq(struct net_device *dev, - struct mlx4_en_cq *cq, int napi_budget); +int mlx4_en_process_tx_cq(struct net_device *dev, + struct mlx4_en_cq *cq, int napi_budget); u32 mlx4_en_free_tx_desc(struct mlx4_en_priv *priv, struct mlx4_en_tx_ring *ring, int index, u64 timestamp, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig index 7d69a3061f17..b6ffd1622cfd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig +++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig @@ -78,9 +78,24 @@ config MLX5_ESWITCH Legacy SRIOV mode (L2 mac vlan steering based). Switchdev mode (eswitch offloads). +config MLX5_CLS_ACT + bool "MLX5 TC classifier action support" + depends on MLX5_ESWITCH && NET_CLS_ACT + default y + help + mlx5 ConnectX offloads support for TC classifier action (NET_CLS_ACT), + works in both native NIC mode and Switchdev SRIOV mode. + Actions get attached to a Hardware offloaded classifiers and are + invoked after a successful classification. Actions are used to + overwrite the classification result, instantly drop or redirect and/or + reformat packets in wire speeds without involving the host cpu. + + If set to N, TC offloads in both NIC and switchdev modes will be disabled. + If unsure, set to Y + config MLX5_TC_CT bool "MLX5 TC connection tracking offload support" - depends on MLX5_CORE_EN && NET_SWITCHDEV && NF_FLOW_TABLE && NET_ACT_CT && NET_TC_SKB_EXT + depends on MLX5_CLS_ACT && NF_FLOW_TABLE && NET_ACT_CT && NET_TC_SKB_EXT default y help Say Y here if you want to support offloading connection tracking rules diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile index d3c7dbd7f1d5..b61e47bc16e8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile +++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile @@ -33,17 +33,24 @@ mlx5_core-$(CONFIG_MLX5_CORE_EN) += en_main.o en_common.o en_fs.o en_ethtool.o \ mlx5_core-$(CONFIG_MLX5_EN_ARFS) += en_arfs.o mlx5_core-$(CONFIG_MLX5_EN_RXNFC) += en_fs_ethtool.o mlx5_core-$(CONFIG_MLX5_CORE_EN_DCB) += en_dcbnl.o en/port_buffer.o -mlx5_core-$(CONFIG_MLX5_ESWITCH) += en_rep.o en_tc.o en/tc_tun.o lib/port_tun.o lag_mp.o \ - lib/geneve.o en/mapping.o en/tc_tun_vxlan.o en/tc_tun_gre.o \ - en/tc_tun_geneve.o diag/en_tc_tracepoint.o mlx5_core-$(CONFIG_PCI_HYPERV_INTERFACE) += en/hv_vhca_stats.o +mlx5_core-$(CONFIG_MLX5_ESWITCH) += lag_mp.o lib/geneve.o lib/port_tun.o \ + en_rep.o en/rep/bond.o +mlx5_core-$(CONFIG_MLX5_CLS_ACT) += en_tc.o en/rep/tc.o en/rep/neigh.o \ + en/mapping.o esw/chains.o en/tc_tun.o \ + en/tc_tun_vxlan.o en/tc_tun_gre.o en/tc_tun_geneve.o \ + en/tc_tun_mplsoudp.o diag/en_tc_tracepoint.o mlx5_core-$(CONFIG_MLX5_TC_CT) += en/tc_ct.o # # Core extra # mlx5_core-$(CONFIG_MLX5_ESWITCH) += eswitch.o eswitch_offloads.o eswitch_offloads_termtbl.o \ - ecpf.o rdma.o esw/chains.o + ecpf.o rdma.o +mlx5_core-$(CONFIG_MLX5_ESWITCH) += esw/acl/helper.o \ + esw/acl/egress_lgcy.o esw/acl/egress_ofld.o \ + esw/acl/ingress_lgcy.o esw/acl/ingress_ofld.o + mlx5_core-$(CONFIG_MLX5_MPFS) += lib/mpfs.o mlx5_core-$(CONFIG_VXLAN) += lib/vxlan.o mlx5_core-$(CONFIG_PTP_1588_CLOCK) += lib/clock.o diff --git a/drivers/net/ethernet/mellanox/mlx5/core/accel/accel.h b/drivers/net/ethernet/mellanox/mlx5/core/accel/accel.h index c13260467750..82b185121edb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/accel/accel.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/accel/accel.h @@ -5,7 +5,6 @@ #include <linux/skbuff.h> #include <linux/netdevice.h> -#include "en.h" static inline bool is_metadata_hdr_valid(struct sk_buff *skb) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.c index eddc34e4a762..8a4985d8cbfe 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.c @@ -58,12 +58,21 @@ int mlx5_accel_ipsec_counters_read(struct mlx5_core_dev *mdev, u64 *counters, void *mlx5_accel_esp_create_hw_context(struct mlx5_core_dev *mdev, struct mlx5_accel_esp_xfrm *xfrm, - const __be32 saddr[4], - const __be32 daddr[4], - const __be32 spi, bool is_ipv6) + u32 *sa_handle) { - return mlx5_fpga_ipsec_create_sa_ctx(mdev, xfrm, saddr, daddr, - spi, is_ipv6); + __be32 saddr[4] = {}, daddr[4] = {}; + + if (!xfrm->attrs.is_ipv6) { + saddr[3] = xfrm->attrs.saddr.a4; + daddr[3] = xfrm->attrs.daddr.a4; + } else { + memcpy(saddr, xfrm->attrs.saddr.a6, sizeof(saddr)); + memcpy(daddr, xfrm->attrs.daddr.a6, sizeof(daddr)); + } + + return mlx5_fpga_ipsec_create_sa_ctx(mdev, xfrm, saddr, + daddr, xfrm->attrs.spi, + xfrm->attrs.is_ipv6, sa_handle); } void mlx5_accel_esp_free_hw_context(void *context) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.h index 530e428d46ab..e89747674712 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.h @@ -48,9 +48,7 @@ int mlx5_accel_ipsec_counters_read(struct mlx5_core_dev *mdev, u64 *counters, void *mlx5_accel_esp_create_hw_context(struct mlx5_core_dev *mdev, struct mlx5_accel_esp_xfrm *xfrm, - const __be32 saddr[4], - const __be32 daddr[4], - const __be32 spi, bool is_ipv6); + u32 *sa_handle); void mlx5_accel_esp_free_hw_context(void *context); int mlx5_accel_ipsec_init(struct mlx5_core_dev *mdev); @@ -64,9 +62,7 @@ void mlx5_accel_ipsec_cleanup(struct mlx5_core_dev *mdev); static inline void * mlx5_accel_esp_create_hw_context(struct mlx5_core_dev *mdev, struct mlx5_accel_esp_xfrm *xfrm, - const __be32 saddr[4], - const __be32 daddr[4], - const __be32 spi, bool is_ipv6) + u32 *sa_handle) { return NULL; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.c b/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.c index cab708af3422..cbf3d76c05a8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.c @@ -56,8 +56,8 @@ void mlx5_accel_tls_del_flow(struct mlx5_core_dev *mdev, u32 swid, mlx5_fpga_tls_del_flow(mdev, swid, GFP_KERNEL, direction_sx); } -int mlx5_accel_tls_resync_rx(struct mlx5_core_dev *mdev, u32 handle, u32 seq, - u64 rcd_sn) +int mlx5_accel_tls_resync_rx(struct mlx5_core_dev *mdev, __be32 handle, + u32 seq, __be64 rcd_sn) { return mlx5_fpga_tls_resync_rx(mdev, handle, seq, rcd_sn); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.h b/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.h index e09bc3858d57..aefea467f7b3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.h @@ -109,8 +109,8 @@ int mlx5_accel_tls_add_flow(struct mlx5_core_dev *mdev, void *flow, bool direction_sx); void mlx5_accel_tls_del_flow(struct mlx5_core_dev *mdev, u32 swid, bool direction_sx); -int mlx5_accel_tls_resync_rx(struct mlx5_core_dev *mdev, u32 handle, u32 seq, - u64 rcd_sn); +int mlx5_accel_tls_resync_rx(struct mlx5_core_dev *mdev, __be32 handle, + u32 seq, __be64 rcd_sn); bool mlx5_accel_is_tls_device(struct mlx5_core_dev *mdev); u32 mlx5_accel_tls_device_caps(struct mlx5_core_dev *mdev); int mlx5_accel_tls_init(struct mlx5_core_dev *mdev); @@ -125,8 +125,8 @@ mlx5_accel_tls_add_flow(struct mlx5_core_dev *mdev, void *flow, bool direction_sx) { return -ENOTSUPP; } static inline void mlx5_accel_tls_del_flow(struct mlx5_core_dev *mdev, u32 swid, bool direction_sx) { } -static inline int mlx5_accel_tls_resync_rx(struct mlx5_core_dev *mdev, u32 handle, - u32 seq, u64 rcd_sn) { return 0; } +static inline int mlx5_accel_tls_resync_rx(struct mlx5_core_dev *mdev, __be32 handle, + u32 seq, __be64 rcd_sn) { return 0; } static inline bool mlx5_accel_is_tls_device(struct mlx5_core_dev *mdev) { return mlx5_accel_is_ktls_device(mdev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index 4a26456117ed..1d91a0d0ab1d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -848,6 +848,14 @@ static void free_msg(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *msg); static void mlx5_free_cmd_msg(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *msg); +static bool opcode_allowed(struct mlx5_cmd *cmd, u16 opcode) +{ + if (cmd->allowed_opcode == CMD_ALLOWED_OPCODE_ALL) + return true; + + return cmd->allowed_opcode == opcode; +} + static void cmd_work_handler(struct work_struct *work) { struct mlx5_cmd_work_ent *ent = container_of(work, struct mlx5_cmd_work_ent, work); @@ -861,6 +869,7 @@ static void cmd_work_handler(struct work_struct *work) int alloc_ret; int cmd_mode; + complete(&ent->handling); sem = ent->page_queue ? &cmd->pages_sem : &cmd->sem; down(sem); if (!ent->page_queue) { @@ -913,7 +922,9 @@ static void cmd_work_handler(struct work_struct *work) /* Skip sending command to fw if internal error */ if (pci_channel_offline(dev->pdev) || - dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) { + dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR || + cmd->state != MLX5_CMDIF_STATE_UP || + !opcode_allowed(&dev->cmd, ent->op)) { u8 status = 0; u32 drv_synd; @@ -978,6 +989,11 @@ static int wait_func(struct mlx5_core_dev *dev, struct mlx5_cmd_work_ent *ent) struct mlx5_cmd *cmd = &dev->cmd; int err; + if (!wait_for_completion_timeout(&ent->handling, timeout) && + cancel_work_sync(&ent->work)) { + ent->ret = -ECANCELED; + goto out_err; + } if (cmd->mode == CMD_MODE_POLLING || ent->polling) { wait_for_completion(&ent->done); } else if (!wait_for_completion_timeout(&ent->done, timeout)) { @@ -985,12 +1001,17 @@ static int wait_func(struct mlx5_core_dev *dev, struct mlx5_cmd_work_ent *ent) mlx5_cmd_comp_handler(dev, 1UL << ent->idx, true); } +out_err: err = ent->ret; if (err == -ETIMEDOUT) { mlx5_core_warn(dev, "%s(0x%x) timeout. Will cause a leak of a command resource\n", mlx5_command_str(msg_to_opcode(ent->in)), msg_to_opcode(ent->in)); + } else if (err == -ECANCELED) { + mlx5_core_warn(dev, "%s(0x%x) canceled on out of queue timeout.\n", + mlx5_command_str(msg_to_opcode(ent->in)), + msg_to_opcode(ent->in)); } mlx5_core_dbg(dev, "err %d, delivery status %s(%d)\n", err, deliv_status_to_str(ent->status), ent->status); @@ -1026,6 +1047,7 @@ static int mlx5_cmd_invoke(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *in, ent->token = token; ent->polling = force_polling; + init_completion(&ent->handling); if (!callback) init_completion(&ent->done); @@ -1045,10 +1067,12 @@ static int mlx5_cmd_invoke(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *in, err = wait_func(dev, ent); if (err == -ETIMEDOUT) goto out; + if (err == -ECANCELED) + goto out_free; ds = ent->ts2 - ent->ts1; op = MLX5_GET(mbox_in, in->first.data, opcode); - if (op < ARRAY_SIZE(cmd->stats)) { + if (op < MLX5_CMD_OP_MAX) { stats = &cmd->stats[op]; spin_lock_irq(&stats->lock); stats->sum += ds; @@ -1391,6 +1415,22 @@ static void create_debugfs_files(struct mlx5_core_dev *dev) mlx5_cmdif_debugfs_init(dev); } +void mlx5_cmd_allowed_opcode(struct mlx5_core_dev *dev, u16 opcode) +{ + struct mlx5_cmd *cmd = &dev->cmd; + int i; + + for (i = 0; i < cmd->max_reg_cmds; i++) + down(&cmd->sem); + down(&cmd->pages_sem); + + cmd->allowed_opcode = opcode; + + up(&cmd->pages_sem); + for (i = 0; i < cmd->max_reg_cmds; i++) + up(&cmd->sem); +} + static void mlx5_cmd_change_mod(struct mlx5_core_dev *dev, int mode) { struct mlx5_cmd *cmd = &dev->cmd; @@ -1511,7 +1551,7 @@ static void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, u64 vec, bool force if (ent->callback) { ds = ent->ts2 - ent->ts1; - if (ent->op < ARRAY_SIZE(cmd->stats)) { + if (ent->op < MLX5_CMD_OP_MAX) { stats = &cmd->stats[ent->op]; spin_lock_irqsave(&stats->lock, flags); stats->sum += ds; @@ -1667,12 +1707,14 @@ static int cmd_exec(struct mlx5_core_dev *dev, void *in, int in_size, void *out, int err; u8 status = 0; u32 drv_synd; + u16 opcode; u8 token; + opcode = MLX5_GET(mbox_in, in, opcode); if (pci_channel_offline(dev->pdev) || - dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR) { - u16 opcode = MLX5_GET(mbox_in, in, opcode); - + dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR || + dev->cmd.state != MLX5_CMDIF_STATE_UP || + !opcode_allowed(&dev->cmd, opcode)) { err = mlx5_internal_err_ret_value(dev, opcode, &drv_synd, &status); MLX5_SET(mbox_out, out, status, status); MLX5_SET(mbox_out, out, syndrome, drv_synd); @@ -1918,10 +1960,16 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) return -EINVAL; } - cmd->pool = dma_pool_create("mlx5_cmd", dev->device, size, align, 0); - if (!cmd->pool) + cmd->stats = kvzalloc(MLX5_CMD_OP_MAX * sizeof(*cmd->stats), GFP_KERNEL); + if (!cmd->stats) return -ENOMEM; + cmd->pool = dma_pool_create("mlx5_cmd", dev->device, size, align, 0); + if (!cmd->pool) { + err = -ENOMEM; + goto dma_pool_err; + } + err = alloc_cmd_page(dev, cmd); if (err) goto err_free_pool; @@ -1942,6 +1990,7 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) goto err_free_page; } + cmd->state = MLX5_CMDIF_STATE_DOWN; cmd->checksum_disabled = 1; cmd->max_reg_cmds = (1 << cmd->log_sz) - 1; cmd->bitmask = (1UL << cmd->max_reg_cmds) - 1; @@ -1956,7 +2005,7 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) spin_lock_init(&cmd->alloc_lock); spin_lock_init(&cmd->token_lock); - for (i = 0; i < ARRAY_SIZE(cmd->stats); i++) + for (i = 0; i < MLX5_CMD_OP_MAX; i++) spin_lock_init(&cmd->stats[i].lock); sema_init(&cmd->sem, cmd->max_reg_cmds); @@ -1979,6 +2028,7 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) mlx5_core_dbg(dev, "descriptor at dma 0x%llx\n", (unsigned long long)(cmd->dma)); cmd->mode = CMD_MODE_POLLING; + cmd->allowed_opcode = CMD_ALLOWED_OPCODE_ALL; create_msg_cache(dev); @@ -2002,7 +2052,8 @@ err_free_page: err_free_pool: dma_pool_destroy(cmd->pool); - +dma_pool_err: + kvfree(cmd->stats); return err; } EXPORT_SYMBOL(mlx5_cmd_init); @@ -2016,5 +2067,13 @@ void mlx5_cmd_cleanup(struct mlx5_core_dev *dev) destroy_msg_cache(dev); free_cmd_page(dev, cmd); dma_pool_destroy(cmd->pool); + kvfree(cmd->stats); } EXPORT_SYMBOL(mlx5_cmd_cleanup); + +void mlx5_cmd_set_state(struct mlx5_core_dev *dev, + enum mlx5_cmdif_state cmdif_state) +{ + dev->cmd.state = cmdif_state; +} +EXPORT_SYMBOL(mlx5_cmd_set_state); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c b/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c index 6409090b3ec5..07c8d9811bc8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/debugfs.c @@ -171,7 +171,7 @@ void mlx5_cmdif_debugfs_init(struct mlx5_core_dev *dev) cmd = &dev->priv.cmdif_debugfs; *cmd = debugfs_create_dir("commands", dev->priv.dbg_root); - for (i = 0; i < ARRAY_SIZE(dev->cmd.stats); i++) { + for (i = 0; i < MLX5_CMD_OP_MAX; i++) { stats = &dev->cmd.stats[i]; namep = mlx5_command_str(i); if (strcmp(namep, "unknown command opcode")) { @@ -202,18 +202,23 @@ void mlx5_cq_debugfs_cleanup(struct mlx5_core_dev *dev) static u64 qp_read_field(struct mlx5_core_dev *dev, struct mlx5_core_qp *qp, int index, int *is_str) { - u32 out[MLX5_ST_SZ_BYTES(query_qp_out)] = {}; + int outlen = MLX5_ST_SZ_BYTES(query_qp_out); u32 in[MLX5_ST_SZ_DW(query_qp_in)] = {}; u64 param = 0; + u32 *out; int state; u32 *qpc; int err; + out = kzalloc(outlen, GFP_KERNEL); + if (!out) + return 0; + MLX5_SET(query_qp_in, in, opcode, MLX5_CMD_OP_QUERY_QP); MLX5_SET(query_qp_in, in, qpn, qp->qpn); err = mlx5_cmd_exec_inout(dev, query_qp, in, out); if (err) - return 0; + goto out; *is_str = 0; @@ -269,7 +274,8 @@ static u64 qp_read_field(struct mlx5_core_dev *dev, struct mlx5_core_qp *qp, param = MLX5_GET(qpc, qpc, remote_qpn); break; } - +out: + kfree(out); return param; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.c index 8ecac81a385d..a700f3c86899 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.c @@ -76,58 +76,59 @@ static void print_lyr_2_4_hdrs(struct trace_seq *p, .v = MLX5_GET(fte_match_set_lyr_2_4, value, dmac_47_16) << 16 | MLX5_GET(fte_match_set_lyr_2_4, value, dmac_15_0)}; MASK_VAL_L2(u16, ethertype, ethertype); + MASK_VAL_L2(u8, ip_version, ip_version); PRINT_MASKED_VALP(smac, u8 *, p, "%pM"); PRINT_MASKED_VALP(dmac, u8 *, p, "%pM"); PRINT_MASKED_VAL(ethertype, p, "%04x"); - if (ethertype.m == 0xffff) { - if (ethertype.v == ETH_P_IP) { + if ((ethertype.m == 0xffff && ethertype.v == ETH_P_IP) || + (ip_version.m == 0xf && ip_version.v == 4)) { #define MASK_VAL_L2_BE(type, name, fld) \ MASK_VAL_BE(type, fte_match_set_lyr_2_4, name, mask, value, fld) - MASK_VAL_L2_BE(u32, src_ipv4, - src_ipv4_src_ipv6.ipv4_layout.ipv4); - MASK_VAL_L2_BE(u32, dst_ipv4, - dst_ipv4_dst_ipv6.ipv4_layout.ipv4); + MASK_VAL_L2_BE(u32, src_ipv4, + src_ipv4_src_ipv6.ipv4_layout.ipv4); + MASK_VAL_L2_BE(u32, dst_ipv4, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4); - PRINT_MASKED_VALP(src_ipv4, typeof(&src_ipv4.v), p, - "%pI4"); - PRINT_MASKED_VALP(dst_ipv4, typeof(&dst_ipv4.v), p, - "%pI4"); - } else if (ethertype.v == ETH_P_IPV6) { - static const struct in6_addr full_ones = { - .in6_u.u6_addr32 = {__constant_htonl(0xffffffff), - __constant_htonl(0xffffffff), - __constant_htonl(0xffffffff), - __constant_htonl(0xffffffff)}, - }; - DECLARE_MASK_VAL(struct in6_addr, src_ipv6); - DECLARE_MASK_VAL(struct in6_addr, dst_ipv6); + PRINT_MASKED_VALP(src_ipv4, typeof(&src_ipv4.v), p, + "%pI4"); + PRINT_MASKED_VALP(dst_ipv4, typeof(&dst_ipv4.v), p, + "%pI4"); + } else if ((ethertype.m == 0xffff && ethertype.v == ETH_P_IPV6) || + (ip_version.m == 0xf && ip_version.v == 6)) { + static const struct in6_addr full_ones = { + .in6_u.u6_addr32 = {__constant_htonl(0xffffffff), + __constant_htonl(0xffffffff), + __constant_htonl(0xffffffff), + __constant_htonl(0xffffffff)}, + }; + DECLARE_MASK_VAL(struct in6_addr, src_ipv6); + DECLARE_MASK_VAL(struct in6_addr, dst_ipv6); - memcpy(src_ipv6.m.in6_u.u6_addr8, - MLX5_ADDR_OF(fte_match_set_lyr_2_4, mask, - src_ipv4_src_ipv6.ipv6_layout.ipv6), - sizeof(src_ipv6.m)); - memcpy(dst_ipv6.m.in6_u.u6_addr8, - MLX5_ADDR_OF(fte_match_set_lyr_2_4, mask, - dst_ipv4_dst_ipv6.ipv6_layout.ipv6), - sizeof(dst_ipv6.m)); - memcpy(src_ipv6.v.in6_u.u6_addr8, - MLX5_ADDR_OF(fte_match_set_lyr_2_4, value, - src_ipv4_src_ipv6.ipv6_layout.ipv6), - sizeof(src_ipv6.v)); - memcpy(dst_ipv6.v.in6_u.u6_addr8, - MLX5_ADDR_OF(fte_match_set_lyr_2_4, value, - dst_ipv4_dst_ipv6.ipv6_layout.ipv6), - sizeof(dst_ipv6.v)); + memcpy(src_ipv6.m.in6_u.u6_addr8, + MLX5_ADDR_OF(fte_match_set_lyr_2_4, mask, + src_ipv4_src_ipv6.ipv6_layout.ipv6), + sizeof(src_ipv6.m)); + memcpy(dst_ipv6.m.in6_u.u6_addr8, + MLX5_ADDR_OF(fte_match_set_lyr_2_4, mask, + dst_ipv4_dst_ipv6.ipv6_layout.ipv6), + sizeof(dst_ipv6.m)); + memcpy(src_ipv6.v.in6_u.u6_addr8, + MLX5_ADDR_OF(fte_match_set_lyr_2_4, value, + src_ipv4_src_ipv6.ipv6_layout.ipv6), + sizeof(src_ipv6.v)); + memcpy(dst_ipv6.v.in6_u.u6_addr8, + MLX5_ADDR_OF(fte_match_set_lyr_2_4, value, + dst_ipv4_dst_ipv6.ipv6_layout.ipv6), + sizeof(dst_ipv6.v)); - if (!memcmp(&src_ipv6.m, &full_ones, sizeof(full_ones))) - trace_seq_printf(p, "src_ipv6=%pI6 ", - &src_ipv6.v); - if (!memcmp(&dst_ipv6.m, &full_ones, sizeof(full_ones))) - trace_seq_printf(p, "dst_ipv6=%pI6 ", - &dst_ipv6.v); - } + if (!memcmp(&src_ipv6.m, &full_ones, sizeof(full_ones))) + trace_seq_printf(p, "src_ipv6=%pI6 ", + &src_ipv6.v); + if (!memcmp(&dst_ipv6.m, &full_ones, sizeof(full_ones))) + trace_seq_printf(p, "dst_ipv6=%pI6 ", + &dst_ipv6.v); } #define PRINT_MASKED_VAL_L2(type, name, fld, p, format) {\ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c index 5ce6ebbc7f10..a7551274be58 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c @@ -684,7 +684,7 @@ static void mlx5_fw_tracer_handle_traces(struct work_struct *work) get_block_timestamp(tracer, &tmp_trace_block[TRACES_PER_BLOCK - 1]); while (block_timestamp > tracer->last_timestamp) { - /* Check block override if its not the first block */ + /* Check block override if it's not the first block */ if (!tracer->last_timestamp) { u64 *ts_event; /* To avoid block override be the HW in case of buffer diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index e8508c74eaa8..842db20493df 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -36,7 +36,6 @@ #include <linux/etherdevice.h> #include <linux/timecounter.h> #include <linux/net_tstamp.h> -#include <linux/ptp_clock_kernel.h> #include <linux/crash_dump.h> #include <linux/mlx5/driver.h> #include <linux/mlx5/qp.h> @@ -53,6 +52,7 @@ #include "wq.h" #include "mlx5_core.h" #include "en_stats.h" +#include "en/dcbnl.h" #include "en/fs.h" #include "lib/hv_vhca.h" @@ -69,8 +69,6 @@ struct page_pool; #define MLX5E_HW2SW_MTU(params, hwmtu) ((hwmtu) - ((params)->hard_mtu)) #define MLX5E_SW2HW_MTU(params, swmtu) ((swmtu) + ((params)->hard_mtu)) -#define MLX5E_MAX_PRIORITY 8 -#define MLX5E_MAX_DSCP 64 #define MLX5E_MAX_NUM_TC 8 #define MLX5_RX_HEADROOM NET_SKB_PAD @@ -243,10 +241,6 @@ enum mlx5e_priv_flag { #define MLX5E_GET_PFLAG(params, pflag) (!!((params)->pflags & (BIT(pflag)))) -#ifdef CONFIG_MLX5_CORE_EN_DCB -#define MLX5E_MAX_BW_ALLOC 100 /* Max percentage of BW allocation */ -#endif - struct mlx5e_params { u8 log_sq_size; u8 rq_wq_type; @@ -271,42 +265,6 @@ struct mlx5e_params { int hard_mtu; }; -#ifdef CONFIG_MLX5_CORE_EN_DCB -struct mlx5e_cee_config { - /* bw pct for priority group */ - u8 pg_bw_pct[CEE_DCBX_MAX_PGS]; - u8 prio_to_pg_map[CEE_DCBX_MAX_PRIO]; - bool pfc_setting[CEE_DCBX_MAX_PRIO]; - bool pfc_enable; -}; - -enum { - MLX5_DCB_CHG_RESET, - MLX5_DCB_NO_CHG, - MLX5_DCB_CHG_NO_RESET, -}; - -struct mlx5e_dcbx { - enum mlx5_dcbx_oper_mode mode; - struct mlx5e_cee_config cee_cfg; /* pending configuration */ - u8 dscp_app_cnt; - - /* The only setting that cannot be read from FW */ - u8 tc_tsa[IEEE_8021QAZ_MAX_TCS]; - u8 cap; - - /* Buffer configuration */ - bool manual_buffer; - u32 cable_len; - u32 xoff; -}; - -struct mlx5e_dcbx_dp { - u8 dscp2prio[MLX5E_MAX_DSCP]; - u8 trust_state; -}; -#endif - enum { MLX5E_RQ_STATE_ENABLED, MLX5E_RQ_STATE_RECOVERING, @@ -339,16 +297,6 @@ struct mlx5e_cq_decomp { u16 wqe_counter; } ____cacheline_aligned_in_smp; -struct mlx5e_tx_wqe_info { - struct sk_buff *skb; - u32 num_bytes; - u8 num_wqebbs; - u8 num_dma; -#ifdef CONFIG_MLX5_EN_TLS - struct page *resync_dump_frag_page; -#endif -}; - enum mlx5e_dma_map_type { MLX5E_DMA_MAP_SINGLE, MLX5E_DMA_MAP_PAGE @@ -370,18 +318,6 @@ enum { MLX5E_SQ_STATE_PENDING_XSK_TX, }; -struct mlx5e_sq_wqe_info { - u8 opcode; - u8 num_wqebbs; - - /* Auxiliary data for different opcodes. */ - union { - struct { - struct mlx5e_rq *rq; - } umr; - }; -}; - struct mlx5e_txqsq { /* data path */ @@ -429,10 +365,7 @@ struct mlx5e_dma_info { dma_addr_t addr; union { struct page *page; - struct { - u64 handle; - void *data; - } xsk; + struct xdp_buff *xsk; }; }; @@ -484,11 +417,6 @@ struct mlx5e_xdp_info_fifo { u32 mask; }; -struct mlx5e_xdp_wqe_info { - u8 num_wqebbs; - u8 num_pkts; -}; - struct mlx5e_xdp_mpwqe { /* Current MPWQE session */ struct mlx5e_tx_wqe *wqe; @@ -552,7 +480,7 @@ struct mlx5e_icosq { /* write@xmit, read@completion */ struct { - struct mlx5e_sq_wqe_info *ico_wqe; + struct mlx5e_icosq_wqe_info *wqe_info; } db; /* read only */ @@ -650,8 +578,8 @@ struct mlx5e_rq { } mpwqe; }; struct { - u16 umem_headroom; u16 headroom; + u32 frame0_sz; u8 map_dir; /* dma map direction */ } buff; @@ -682,7 +610,6 @@ struct mlx5e_rq { struct page_pool *page_pool; /* AF_XDP zero-copy */ - struct zero_copy_allocator zca; struct xdp_umem *umem; struct work_struct recover_work; @@ -919,8 +846,8 @@ void mlx5e_build_ptys2ethtool_map(void); u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb, struct net_device *sb_dev); netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev); -netdev_tx_t mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, - struct mlx5e_tx_wqe *wqe, u16 pi, bool xmit_more); +void mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5e_tx_wqe *wqe, u16 pi, bool xmit_more); void mlx5e_trigger_irq(struct mlx5e_icosq *sq); void mlx5e_completion_event(struct mlx5_core_cq *mcq, struct mlx5_eqe *eqe); @@ -1068,10 +995,12 @@ void mlx5e_deactivate_priv_channels(struct mlx5e_priv *priv); void mlx5e_build_default_indir_rqt(u32 *indirection_rqt, int len, int num_channels); -void mlx5e_set_tx_cq_mode_params(struct mlx5e_params *params, - u8 cq_period_mode); -void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, - u8 cq_period_mode); + +void mlx5e_reset_tx_moderation(struct mlx5e_params *params, u8 cq_period_mode); +void mlx5e_reset_rx_moderation(struct mlx5e_params *params, u8 cq_period_mode); +void mlx5e_set_tx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode); +void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode); + void mlx5e_set_rq_type(struct mlx5_core_dev *mdev, struct mlx5e_params *params); void mlx5e_init_rq_type_params(struct mlx5_core_dev *mdev, struct mlx5e_params *params); @@ -1095,13 +1024,6 @@ static inline bool mlx5_tx_swp_supported(struct mlx5_core_dev *mdev) } extern const struct ethtool_ops mlx5e_ethtool_ops; -#ifdef CONFIG_MLX5_CORE_EN_DCB -extern const struct dcbnl_rtnl_ops mlx5e_dcbnl_ops; -int mlx5e_dcbnl_ieee_setets_core(struct mlx5e_priv *priv, struct ieee_ets *ets); -void mlx5e_dcbnl_initialize(struct mlx5e_priv *priv); -void mlx5e_dcbnl_init_app(struct mlx5e_priv *priv); -void mlx5e_dcbnl_delete_app(struct mlx5e_priv *priv); -#endif int mlx5e_create_tir(struct mlx5_core_dev *mdev, struct mlx5e_tir *tir, u32 *in); @@ -1109,7 +1031,8 @@ void mlx5e_destroy_tir(struct mlx5_core_dev *mdev, struct mlx5e_tir *tir); int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev); void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev); -int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb); +int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb, + bool enable_mc_lb); /* common netdev helpers */ void mlx5e_create_q_counters(struct mlx5e_priv *priv); @@ -1121,7 +1044,7 @@ void mlx5e_close_drop_rq(struct mlx5e_rq *drop_rq); int mlx5e_create_indirect_rqt(struct mlx5e_priv *priv); int mlx5e_create_indirect_tirs(struct mlx5e_priv *priv, bool inner_ttc); -void mlx5e_destroy_indirect_tirs(struct mlx5e_priv *priv, bool inner_ttc); +void mlx5e_destroy_indirect_tirs(struct mlx5e_priv *priv); int mlx5e_create_direct_rqts(struct mlx5e_priv *priv, struct mlx5e_tir *tirs); void mlx5e_destroy_direct_rqts(struct mlx5e_priv *priv, struct mlx5e_tir *tirs); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/dcbnl.h b/drivers/net/ethernet/mellanox/mlx5/core/en/dcbnl.h new file mode 100644 index 000000000000..7be6b2d36b60 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/dcbnl.h @@ -0,0 +1,54 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* Copyright (c) 2020 Mellanox Technologies. */ + +#ifndef __MLX5E_DCBNL_H__ +#define __MLX5E_DCBNL_H__ + +#ifdef CONFIG_MLX5_CORE_EN_DCB + +#define MLX5E_MAX_PRIORITY (8) + +struct mlx5e_cee_config { + /* bw pct for priority group */ + u8 pg_bw_pct[CEE_DCBX_MAX_PGS]; + u8 prio_to_pg_map[CEE_DCBX_MAX_PRIO]; + bool pfc_setting[CEE_DCBX_MAX_PRIO]; + bool pfc_enable; +}; + +struct mlx5e_dcbx { + enum mlx5_dcbx_oper_mode mode; + struct mlx5e_cee_config cee_cfg; /* pending configuration */ + u8 dscp_app_cnt; + + /* The only setting that cannot be read from FW */ + u8 tc_tsa[IEEE_8021QAZ_MAX_TCS]; + u8 cap; + + /* Buffer configuration */ + bool manual_buffer; + u32 cable_len; + u32 xoff; +}; + +#define MLX5E_MAX_DSCP (64) + +struct mlx5e_dcbx_dp { + u8 dscp2prio[MLX5E_MAX_DSCP]; + u8 trust_state; +}; + +void mlx5e_dcbnl_build_netdev(struct net_device *netdev); +void mlx5e_dcbnl_build_rep_netdev(struct net_device *netdev); +void mlx5e_dcbnl_initialize(struct mlx5e_priv *priv); +void mlx5e_dcbnl_init_app(struct mlx5e_priv *priv); +void mlx5e_dcbnl_delete_app(struct mlx5e_priv *priv); +#else +static inline void mlx5e_dcbnl_build_netdev(struct net_device *netdev) {} +static inline void mlx5e_dcbnl_build_rep_netdev(struct net_device *netdev) {} +static inline void mlx5e_dcbnl_initialize(struct mlx5e_priv *priv) {} +static inline void mlx5e_dcbnl_init_app(struct mlx5e_priv *priv) {} +static inline void mlx5e_dcbnl_delete_app(struct mlx5e_priv *priv) {} +#endif + +#endif /* __MLX5E_DCBNL_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c index eb2e1f2138e4..38e4f19d69f8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/params.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/params.c @@ -12,15 +12,16 @@ static inline bool mlx5e_rx_is_xdp(struct mlx5e_params *params, u16 mlx5e_get_linear_rq_headroom(struct mlx5e_params *params, struct mlx5e_xsk_param *xsk) { - u16 headroom = NET_IP_ALIGN; + u16 headroom; - if (mlx5e_rx_is_xdp(params, xsk)) { + if (xsk) + return xsk->headroom; + + headroom = NET_IP_ALIGN; + if (mlx5e_rx_is_xdp(params, xsk)) headroom += XDP_PACKET_HEADROOM; - if (xsk) - headroom += xsk->headroom; - } else { + else headroom += MLX5_RX_HEADROOM; - } return headroom; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c index 2c4a670c8ffd..2a8950b3056f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c @@ -369,17 +369,19 @@ enum mlx5e_fec_supported_link_mode { *_policy = MLX5_GET(pplm_reg, _buf, fec_override_admin_##link); \ } while (0) -#define MLX5E_FEC_OVERRIDE_ADMIN_50G_POLICY(buf, policy, write, link) \ - do { \ - u16 *__policy = &(policy); \ - bool _write = (write); \ - \ - if (_write && *__policy) \ - *__policy = find_first_bit((u_long *)__policy, \ - sizeof(u16) * BITS_PER_BYTE);\ - MLX5E_FEC_OVERRIDE_ADMIN_POLICY(buf, *__policy, _write, link); \ - if (!_write && *__policy) \ - *__policy = 1 << *__policy; \ +#define MLX5E_FEC_OVERRIDE_ADMIN_50G_POLICY(buf, policy, write, link) \ + do { \ + unsigned long policy_long; \ + u16 *__policy = &(policy); \ + bool _write = (write); \ + \ + policy_long = *__policy; \ + if (_write && *__policy) \ + *__policy = find_first_bit(&policy_long, \ + sizeof(policy_long) * BITS_PER_BYTE);\ + MLX5E_FEC_OVERRIDE_ADMIN_POLICY(buf, *__policy, _write, link); \ + if (!_write && *__policy) \ + *__policy = 1 << *__policy; \ } while (0) /* get/set FEC admin field for a given speed */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bond.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bond.c new file mode 100644 index 000000000000..bdb71332cbf2 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/bond.c @@ -0,0 +1,350 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2020 Mellanox Technologies Inc. All rights reserved. */ + +#include <linux/netdevice.h> +#include <linux/list.h> +#include <net/lag.h> + +#include "mlx5_core.h" +#include "eswitch.h" +#include "esw/acl/ofld.h" +#include "en_rep.h" + +struct mlx5e_rep_bond { + struct notifier_block nb; + struct netdev_net_notifier nn; + struct list_head metadata_list; +}; + +struct mlx5e_rep_bond_slave_entry { + struct list_head list; + struct net_device *netdev; +}; + +struct mlx5e_rep_bond_metadata { + struct list_head list; /* link to global list of rep_bond_metadata */ + struct mlx5_eswitch *esw; + /* private of uplink holding rep bond metadata list */ + struct net_device *lag_dev; + u32 metadata_reg_c_0; + + struct list_head slaves_list; /* slaves list */ + int slaves; +}; + +static struct mlx5e_rep_bond_metadata * +mlx5e_lookup_rep_bond_metadata(struct mlx5_rep_uplink_priv *uplink_priv, + const struct net_device *lag_dev) +{ + struct mlx5e_rep_bond_metadata *found = NULL; + struct mlx5e_rep_bond_metadata *cur; + + list_for_each_entry(cur, &uplink_priv->bond->metadata_list, list) { + if (cur->lag_dev == lag_dev) { + found = cur; + break; + } + } + + return found; +} + +static struct mlx5e_rep_bond_slave_entry * +mlx5e_lookup_rep_bond_slave_entry(struct mlx5e_rep_bond_metadata *mdata, + const struct net_device *netdev) +{ + struct mlx5e_rep_bond_slave_entry *found = NULL; + struct mlx5e_rep_bond_slave_entry *cur; + + list_for_each_entry(cur, &mdata->slaves_list, list) { + if (cur->netdev == netdev) { + found = cur; + break; + } + } + + return found; +} + +static void mlx5e_rep_bond_metadata_release(struct mlx5e_rep_bond_metadata *mdata) +{ + netdev_dbg(mdata->lag_dev, "destroy rep_bond_metadata(%d)\n", + mdata->metadata_reg_c_0); + list_del(&mdata->list); + mlx5_esw_match_metadata_free(mdata->esw, mdata->metadata_reg_c_0); + WARN_ON(!list_empty(&mdata->slaves_list)); + kfree(mdata); +} + +/* This must be called under rtnl_lock */ +int mlx5e_rep_bond_enslave(struct mlx5_eswitch *esw, struct net_device *netdev, + struct net_device *lag_dev) +{ + struct mlx5e_rep_bond_slave_entry *s_entry; + struct mlx5e_rep_bond_metadata *mdata; + struct mlx5e_rep_priv *rpriv; + struct mlx5e_priv *priv; + int err; + + ASSERT_RTNL(); + + rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); + mdata = mlx5e_lookup_rep_bond_metadata(&rpriv->uplink_priv, lag_dev); + if (!mdata) { + /* First netdev becomes slave, no metadata presents the lag_dev. Create one */ + mdata = kzalloc(sizeof(*mdata), GFP_KERNEL); + if (!mdata) + return -ENOMEM; + + mdata->lag_dev = lag_dev; + mdata->esw = esw; + INIT_LIST_HEAD(&mdata->slaves_list); + mdata->metadata_reg_c_0 = mlx5_esw_match_metadata_alloc(esw); + if (!mdata->metadata_reg_c_0) { + kfree(mdata); + return -ENOSPC; + } + list_add(&mdata->list, &rpriv->uplink_priv.bond->metadata_list); + + netdev_dbg(lag_dev, "create rep_bond_metadata(%d)\n", + mdata->metadata_reg_c_0); + } + + s_entry = kzalloc(sizeof(*s_entry), GFP_KERNEL); + if (!s_entry) { + err = -ENOMEM; + goto entry_alloc_err; + } + + s_entry->netdev = netdev; + priv = netdev_priv(netdev); + rpriv = priv->ppriv; + + err = mlx5_esw_acl_ingress_vport_bond_update(esw, rpriv->rep->vport, + mdata->metadata_reg_c_0); + if (err) + goto ingress_err; + + mdata->slaves++; + list_add_tail(&s_entry->list, &mdata->slaves_list); + netdev_dbg(netdev, "enslave rep vport(%d) lag_dev(%s) metadata(0x%x)\n", + rpriv->rep->vport, lag_dev->name, mdata->metadata_reg_c_0); + + return 0; + +ingress_err: + kfree(s_entry); +entry_alloc_err: + if (!mdata->slaves) + mlx5e_rep_bond_metadata_release(mdata); + return err; +} + +/* This must be called under rtnl_lock */ +void mlx5e_rep_bond_unslave(struct mlx5_eswitch *esw, + const struct net_device *netdev, + const struct net_device *lag_dev) +{ + struct mlx5e_rep_bond_slave_entry *s_entry; + struct mlx5e_rep_bond_metadata *mdata; + struct mlx5e_rep_priv *rpriv; + struct mlx5e_priv *priv; + + ASSERT_RTNL(); + + rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); + mdata = mlx5e_lookup_rep_bond_metadata(&rpriv->uplink_priv, lag_dev); + if (!mdata) + return; + + s_entry = mlx5e_lookup_rep_bond_slave_entry(mdata, netdev); + if (!s_entry) + return; + + priv = netdev_priv(netdev); + rpriv = priv->ppriv; + + /* Reset bond_metadata to zero first then reset all ingress/egress + * acls and rx rules of unslave representor's vport + */ + mlx5_esw_acl_ingress_vport_bond_update(esw, rpriv->rep->vport, 0); + mlx5_esw_acl_egress_vport_unbond(esw, rpriv->rep->vport); + mlx5e_rep_bond_update(priv, false); + + list_del(&s_entry->list); + + netdev_dbg(netdev, "unslave rep vport(%d) lag_dev(%s) metadata(0x%x)\n", + rpriv->rep->vport, lag_dev->name, mdata->metadata_reg_c_0); + + if (--mdata->slaves == 0) + mlx5e_rep_bond_metadata_release(mdata); + kfree(s_entry); +} + +static bool mlx5e_rep_is_lag_netdev(struct net_device *netdev) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5e_rep_priv *rpriv = priv->ppriv; + + /* A given netdev is not a representor or not a slave of LAG configuration */ + if (!mlx5e_eswitch_rep(netdev) || !bond_slave_get_rtnl(netdev)) + return false; + + /* Egress acl forward to vport is supported only non-uplink representor */ + return rpriv->rep->vport != MLX5_VPORT_UPLINK; +} + +static void mlx5e_rep_changelowerstate_event(struct net_device *netdev, void *ptr) +{ + struct netdev_notifier_changelowerstate_info *info; + struct netdev_lag_lower_state_info *lag_info; + struct mlx5e_rep_priv *rpriv; + struct net_device *lag_dev; + struct mlx5e_priv *priv; + struct list_head *iter; + struct net_device *dev; + u16 acl_vport_num; + u16 fwd_vport_num; + int err; + + if (!mlx5e_rep_is_lag_netdev(netdev)) + return; + + info = ptr; + lag_info = info->lower_state_info; + /* This is not an event of a representor becoming active slave */ + if (!lag_info->tx_enabled) + return; + + priv = netdev_priv(netdev); + rpriv = priv->ppriv; + fwd_vport_num = rpriv->rep->vport; + lag_dev = netdev_master_upper_dev_get(netdev); + + netdev_dbg(netdev, "lag_dev(%s)'s slave vport(%d) is txable(%d)\n", + lag_dev->name, fwd_vport_num, net_lag_port_dev_txable(netdev)); + + /* Point everyone's egress acl to the vport of the active representor */ + netdev_for_each_lower_dev(lag_dev, dev, iter) { + priv = netdev_priv(dev); + rpriv = priv->ppriv; + acl_vport_num = rpriv->rep->vport; + if (acl_vport_num != fwd_vport_num) { + /* Only single rx_rule for unique bond_metadata should be + * present, delete it if it's saved as passive vport's + * rx_rule with destination as passive vport's root_ft + */ + mlx5e_rep_bond_update(priv, true); + err = mlx5_esw_acl_egress_vport_bond(priv->mdev->priv.eswitch, + fwd_vport_num, + acl_vport_num); + if (err) + netdev_warn(dev, + "configure slave vport(%d) egress fwd, err(%d)", + acl_vport_num, err); + } + } + + /* Insert new rx_rule for unique bond_metadata, save it as active vport's + * rx_rule with new destination as active vport's root_ft + */ + err = mlx5e_rep_bond_update(netdev_priv(netdev), false); + if (err) + netdev_warn(netdev, "configure active slave vport(%d) rx_rule, err(%d)", + fwd_vport_num, err); +} + +static void mlx5e_rep_changeupper_event(struct net_device *netdev, void *ptr) +{ + struct netdev_notifier_changeupper_info *info = ptr; + struct mlx5e_rep_priv *rpriv; + struct net_device *lag_dev; + struct mlx5e_priv *priv; + + if (!mlx5e_rep_is_lag_netdev(netdev)) + return; + + priv = netdev_priv(netdev); + rpriv = priv->ppriv; + lag_dev = info->upper_dev; + + netdev_dbg(netdev, "%sslave vport(%d) lag(%s)\n", + info->linking ? "en" : "un", rpriv->rep->vport, lag_dev->name); + + if (info->linking) + mlx5e_rep_bond_enslave(priv->mdev->priv.eswitch, netdev, lag_dev); + else + mlx5e_rep_bond_unslave(priv->mdev->priv.eswitch, netdev, lag_dev); +} + +/* Bond device of representors and netdev events are used here in specific way + * to support eswitch vports bonding and to perform failover of eswitch vport + * by modifying the vport's egress acl of lower dev representors. Thus this + * also change the traditional behavior of lower dev under bond device. + * All non-representor netdevs or representors of other vendors as lower dev + * of bond device are not supported. + */ +static int mlx5e_rep_esw_bond_netevent(struct notifier_block *nb, + unsigned long event, void *ptr) +{ + struct net_device *netdev = netdev_notifier_info_to_dev(ptr); + + switch (event) { + case NETDEV_CHANGELOWERSTATE: + mlx5e_rep_changelowerstate_event(netdev, ptr); + break; + case NETDEV_CHANGEUPPER: + mlx5e_rep_changeupper_event(netdev, ptr); + break; + } + return NOTIFY_DONE; +} + +/* If HW support eswitch vports bonding, register a specific notifier to + * handle it when two or more representors are bonded + */ +int mlx5e_rep_bond_init(struct mlx5e_rep_priv *rpriv) +{ + struct mlx5_rep_uplink_priv *uplink_priv = &rpriv->uplink_priv; + struct net_device *netdev = rpriv->netdev; + struct mlx5e_priv *priv; + int ret = 0; + + priv = netdev_priv(netdev); + if (!mlx5_esw_acl_egress_fwd2vport_supported(priv->mdev->priv.eswitch)) + goto out; + + uplink_priv->bond = kvzalloc(sizeof(*uplink_priv->bond), GFP_KERNEL); + if (!uplink_priv->bond) { + ret = -ENOMEM; + goto out; + } + + INIT_LIST_HEAD(&uplink_priv->bond->metadata_list); + uplink_priv->bond->nb.notifier_call = mlx5e_rep_esw_bond_netevent; + ret = register_netdevice_notifier_dev_net(netdev, + &uplink_priv->bond->nb, + &uplink_priv->bond->nn); + if (ret) { + netdev_err(netdev, "register bonding netevent notifier, err(%d)\n", ret); + kvfree(uplink_priv->bond); + uplink_priv->bond = NULL; + } + +out: + return ret; +} + +void mlx5e_rep_bond_cleanup(struct mlx5e_rep_priv *rpriv) +{ + struct mlx5e_priv *priv = netdev_priv(rpriv->netdev); + + if (!mlx5_esw_acl_egress_fwd2vport_supported(priv->mdev->priv.eswitch) || + !rpriv->uplink_priv.bond) + return; + + unregister_netdevice_notifier_dev_net(rpriv->netdev, + &rpriv->uplink_priv.bond->nb, + &rpriv->uplink_priv.bond->nn); + kvfree(rpriv->uplink_priv.bond); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/neigh.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/neigh.c new file mode 100644 index 000000000000..baa162432e75 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/neigh.c @@ -0,0 +1,368 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2020 Mellanox Technologies. */ + +#include <linux/refcount.h> +#include <linux/list.h> +#include <linux/rculist.h> +#include <linux/rtnetlink.h> +#include <linux/workqueue.h> +#include <linux/rwlock.h> +#include <linux/spinlock.h> +#include <linux/notifier.h> +#include <net/netevent.h> +#include "neigh.h" +#include "tc.h" +#include "en_rep.h" +#include "fs_core.h" +#include "diag/en_rep_tracepoint.h" + +static unsigned long mlx5e_rep_ipv6_interval(void) +{ + if (IS_ENABLED(CONFIG_IPV6) && ipv6_stub->nd_tbl) + return NEIGH_VAR(&ipv6_stub->nd_tbl->parms, DELAY_PROBE_TIME); + + return ~0UL; +} + +static void mlx5e_rep_neigh_update_init_interval(struct mlx5e_rep_priv *rpriv) +{ + unsigned long ipv4_interval = NEIGH_VAR(&arp_tbl.parms, DELAY_PROBE_TIME); + unsigned long ipv6_interval = mlx5e_rep_ipv6_interval(); + struct net_device *netdev = rpriv->netdev; + struct mlx5e_priv *priv = netdev_priv(netdev); + + rpriv->neigh_update.min_interval = min_t(unsigned long, ipv6_interval, ipv4_interval); + mlx5_fc_update_sampling_interval(priv->mdev, rpriv->neigh_update.min_interval); +} + +void mlx5e_rep_queue_neigh_stats_work(struct mlx5e_priv *priv) +{ + struct mlx5e_rep_priv *rpriv = priv->ppriv; + struct mlx5e_neigh_update_table *neigh_update = &rpriv->neigh_update; + + mlx5_fc_queue_stats_work(priv->mdev, + &neigh_update->neigh_stats_work, + neigh_update->min_interval); +} + +static bool mlx5e_rep_neigh_entry_hold(struct mlx5e_neigh_hash_entry *nhe) +{ + return refcount_inc_not_zero(&nhe->refcnt); +} + +static void mlx5e_rep_neigh_entry_remove(struct mlx5e_neigh_hash_entry *nhe); + +void mlx5e_rep_neigh_entry_release(struct mlx5e_neigh_hash_entry *nhe) +{ + if (refcount_dec_and_test(&nhe->refcnt)) { + mlx5e_rep_neigh_entry_remove(nhe); + kfree_rcu(nhe, rcu); + } +} + +static struct mlx5e_neigh_hash_entry * +mlx5e_get_next_nhe(struct mlx5e_rep_priv *rpriv, + struct mlx5e_neigh_hash_entry *nhe) +{ + struct mlx5e_neigh_hash_entry *next = NULL; + + rcu_read_lock(); + + for (next = nhe ? + list_next_or_null_rcu(&rpriv->neigh_update.neigh_list, + &nhe->neigh_list, + struct mlx5e_neigh_hash_entry, + neigh_list) : + list_first_or_null_rcu(&rpriv->neigh_update.neigh_list, + struct mlx5e_neigh_hash_entry, + neigh_list); + next; + next = list_next_or_null_rcu(&rpriv->neigh_update.neigh_list, + &next->neigh_list, + struct mlx5e_neigh_hash_entry, + neigh_list)) + if (mlx5e_rep_neigh_entry_hold(next)) + break; + + rcu_read_unlock(); + + if (nhe) + mlx5e_rep_neigh_entry_release(nhe); + + return next; +} + +static void mlx5e_rep_neigh_stats_work(struct work_struct *work) +{ + struct mlx5e_rep_priv *rpriv = container_of(work, struct mlx5e_rep_priv, + neigh_update.neigh_stats_work.work); + struct net_device *netdev = rpriv->netdev; + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5e_neigh_hash_entry *nhe = NULL; + + rtnl_lock(); + if (!list_empty(&rpriv->neigh_update.neigh_list)) + mlx5e_rep_queue_neigh_stats_work(priv); + + while ((nhe = mlx5e_get_next_nhe(rpriv, nhe)) != NULL) + mlx5e_tc_update_neigh_used_value(nhe); + + rtnl_unlock(); +} + +static void mlx5e_rep_neigh_update(struct work_struct *work) +{ + struct mlx5e_neigh_hash_entry *nhe = + container_of(work, struct mlx5e_neigh_hash_entry, neigh_update_work); + struct neighbour *n = nhe->n; + struct mlx5e_encap_entry *e; + unsigned char ha[ETH_ALEN]; + struct mlx5e_priv *priv; + bool neigh_connected; + u8 nud_state, dead; + + rtnl_lock(); + + /* If these parameters are changed after we release the lock, + * we'll receive another event letting us know about it. + * We use this lock to avoid inconsistency between the neigh validity + * and it's hw address. + */ + read_lock_bh(&n->lock); + memcpy(ha, n->ha, ETH_ALEN); + nud_state = n->nud_state; + dead = n->dead; + read_unlock_bh(&n->lock); + + neigh_connected = (nud_state & NUD_VALID) && !dead; + + trace_mlx5e_rep_neigh_update(nhe, ha, neigh_connected); + + list_for_each_entry(e, &nhe->encap_list, encap_list) { + if (!mlx5e_encap_take(e)) + continue; + + priv = netdev_priv(e->out_dev); + mlx5e_rep_update_flows(priv, e, neigh_connected, ha); + mlx5e_encap_put(priv, e); + } + mlx5e_rep_neigh_entry_release(nhe); + rtnl_unlock(); + neigh_release(n); +} + +static void mlx5e_rep_queue_neigh_update_work(struct mlx5e_priv *priv, + struct mlx5e_neigh_hash_entry *nhe, + struct neighbour *n) +{ + /* Take a reference to ensure the neighbour and mlx5 encap + * entry won't be destructed until we drop the reference in + * delayed work. + */ + neigh_hold(n); + + /* This assignment is valid as long as the the neigh reference + * is taken + */ + nhe->n = n; + + if (!queue_work(priv->wq, &nhe->neigh_update_work)) { + mlx5e_rep_neigh_entry_release(nhe); + neigh_release(n); + } +} + +static int mlx5e_rep_netevent_event(struct notifier_block *nb, + unsigned long event, void *ptr) +{ + struct mlx5e_rep_priv *rpriv = container_of(nb, struct mlx5e_rep_priv, + neigh_update.netevent_nb); + struct mlx5e_neigh_update_table *neigh_update = &rpriv->neigh_update; + struct net_device *netdev = rpriv->netdev; + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5e_neigh_hash_entry *nhe = NULL; + struct mlx5e_neigh m_neigh = {}; + struct neigh_parms *p; + struct neighbour *n; + bool found = false; + + switch (event) { + case NETEVENT_NEIGH_UPDATE: + n = ptr; +#if IS_ENABLED(CONFIG_IPV6) + if (n->tbl != ipv6_stub->nd_tbl && n->tbl != &arp_tbl) +#else + if (n->tbl != &arp_tbl) +#endif + return NOTIFY_DONE; + + m_neigh.dev = n->dev; + m_neigh.family = n->ops->family; + memcpy(&m_neigh.dst_ip, n->primary_key, n->tbl->key_len); + + rcu_read_lock(); + nhe = mlx5e_rep_neigh_entry_lookup(priv, &m_neigh); + rcu_read_unlock(); + if (!nhe) + return NOTIFY_DONE; + + mlx5e_rep_queue_neigh_update_work(priv, nhe, n); + break; + + case NETEVENT_DELAY_PROBE_TIME_UPDATE: + p = ptr; + + /* We check the device is present since we don't care about + * changes in the default table, we only care about changes + * done per device delay prob time parameter. + */ +#if IS_ENABLED(CONFIG_IPV6) + if (!p->dev || (p->tbl != ipv6_stub->nd_tbl && p->tbl != &arp_tbl)) +#else + if (!p->dev || p->tbl != &arp_tbl) +#endif + return NOTIFY_DONE; + + rcu_read_lock(); + list_for_each_entry_rcu(nhe, &neigh_update->neigh_list, + neigh_list) { + if (p->dev == nhe->m_neigh.dev) { + found = true; + break; + } + } + rcu_read_unlock(); + if (!found) + return NOTIFY_DONE; + + neigh_update->min_interval = min_t(unsigned long, + NEIGH_VAR(p, DELAY_PROBE_TIME), + neigh_update->min_interval); + mlx5_fc_update_sampling_interval(priv->mdev, + neigh_update->min_interval); + break; + } + return NOTIFY_DONE; +} + +static const struct rhashtable_params mlx5e_neigh_ht_params = { + .head_offset = offsetof(struct mlx5e_neigh_hash_entry, rhash_node), + .key_offset = offsetof(struct mlx5e_neigh_hash_entry, m_neigh), + .key_len = sizeof(struct mlx5e_neigh), + .automatic_shrinking = true, +}; + +int mlx5e_rep_neigh_init(struct mlx5e_rep_priv *rpriv) +{ + struct mlx5e_neigh_update_table *neigh_update = &rpriv->neigh_update; + int err; + + err = rhashtable_init(&neigh_update->neigh_ht, &mlx5e_neigh_ht_params); + if (err) + return err; + + INIT_LIST_HEAD(&neigh_update->neigh_list); + mutex_init(&neigh_update->encap_lock); + INIT_DELAYED_WORK(&neigh_update->neigh_stats_work, + mlx5e_rep_neigh_stats_work); + mlx5e_rep_neigh_update_init_interval(rpriv); + + rpriv->neigh_update.netevent_nb.notifier_call = mlx5e_rep_netevent_event; + err = register_netevent_notifier(&rpriv->neigh_update.netevent_nb); + if (err) + goto out_err; + return 0; + +out_err: + rhashtable_destroy(&neigh_update->neigh_ht); + return err; +} + +void mlx5e_rep_neigh_cleanup(struct mlx5e_rep_priv *rpriv) +{ + struct mlx5e_neigh_update_table *neigh_update = &rpriv->neigh_update; + struct mlx5e_priv *priv = netdev_priv(rpriv->netdev); + + unregister_netevent_notifier(&neigh_update->netevent_nb); + + flush_workqueue(priv->wq); /* flush neigh update works */ + + cancel_delayed_work_sync(&rpriv->neigh_update.neigh_stats_work); + + mutex_destroy(&neigh_update->encap_lock); + rhashtable_destroy(&neigh_update->neigh_ht); +} + +static int mlx5e_rep_neigh_entry_insert(struct mlx5e_priv *priv, + struct mlx5e_neigh_hash_entry *nhe) +{ + struct mlx5e_rep_priv *rpriv = priv->ppriv; + int err; + + err = rhashtable_insert_fast(&rpriv->neigh_update.neigh_ht, + &nhe->rhash_node, + mlx5e_neigh_ht_params); + if (err) + return err; + + list_add_rcu(&nhe->neigh_list, &rpriv->neigh_update.neigh_list); + + return err; +} + +static void mlx5e_rep_neigh_entry_remove(struct mlx5e_neigh_hash_entry *nhe) +{ + struct mlx5e_rep_priv *rpriv = nhe->priv->ppriv; + + mutex_lock(&rpriv->neigh_update.encap_lock); + + list_del_rcu(&nhe->neigh_list); + + rhashtable_remove_fast(&rpriv->neigh_update.neigh_ht, + &nhe->rhash_node, + mlx5e_neigh_ht_params); + mutex_unlock(&rpriv->neigh_update.encap_lock); +} + +/* This function must only be called under the representor's encap_lock or + * inside rcu read lock section. + */ +struct mlx5e_neigh_hash_entry * +mlx5e_rep_neigh_entry_lookup(struct mlx5e_priv *priv, + struct mlx5e_neigh *m_neigh) +{ + struct mlx5e_rep_priv *rpriv = priv->ppriv; + struct mlx5e_neigh_update_table *neigh_update = &rpriv->neigh_update; + struct mlx5e_neigh_hash_entry *nhe; + + nhe = rhashtable_lookup_fast(&neigh_update->neigh_ht, m_neigh, + mlx5e_neigh_ht_params); + return nhe && mlx5e_rep_neigh_entry_hold(nhe) ? nhe : NULL; +} + +int mlx5e_rep_neigh_entry_create(struct mlx5e_priv *priv, + struct mlx5e_encap_entry *e, + struct mlx5e_neigh_hash_entry **nhe) +{ + int err; + + *nhe = kzalloc(sizeof(**nhe), GFP_KERNEL); + if (!*nhe) + return -ENOMEM; + + (*nhe)->priv = priv; + memcpy(&(*nhe)->m_neigh, &e->m_neigh, sizeof(e->m_neigh)); + INIT_WORK(&(*nhe)->neigh_update_work, mlx5e_rep_neigh_update); + spin_lock_init(&(*nhe)->encap_list_lock); + INIT_LIST_HEAD(&(*nhe)->encap_list); + refcount_set(&(*nhe)->refcnt, 1); + + err = mlx5e_rep_neigh_entry_insert(priv, *nhe); + if (err) + goto out_free; + return 0; + +out_free: + kfree(*nhe); + return err; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/neigh.h b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/neigh.h new file mode 100644 index 000000000000..32b239189c95 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/neigh.h @@ -0,0 +1,34 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* Copyright (c) 2020 Mellanox Technologies. */ + +#ifndef __MLX5_EN_REP_NEIGH__ +#define __MLX5_EN_REP_NEIGH__ + +#include "en.h" +#include "en_rep.h" + +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) + +int mlx5e_rep_neigh_init(struct mlx5e_rep_priv *rpriv); +void mlx5e_rep_neigh_cleanup(struct mlx5e_rep_priv *rpriv); + +struct mlx5e_neigh_hash_entry * +mlx5e_rep_neigh_entry_lookup(struct mlx5e_priv *priv, + struct mlx5e_neigh *m_neigh); +int mlx5e_rep_neigh_entry_create(struct mlx5e_priv *priv, + struct mlx5e_encap_entry *e, + struct mlx5e_neigh_hash_entry **nhe); +void mlx5e_rep_neigh_entry_release(struct mlx5e_neigh_hash_entry *nhe); + +void mlx5e_rep_queue_neigh_stats_work(struct mlx5e_priv *priv); + +#else /* CONFIG_MLX5_CLS_ACT */ + +static inline int +mlx5e_rep_neigh_init(struct mlx5e_rep_priv *rpriv) { return 0; } +static inline void +mlx5e_rep_neigh_cleanup(struct mlx5e_rep_priv *rpriv) {} + +#endif /* CONFIG_MLX5_CLS_ACT */ + +#endif /* __MLX5_EN_REP_NEIGH__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c new file mode 100644 index 000000000000..80713123de5c --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.c @@ -0,0 +1,646 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2020 Mellanox Technologies. */ + +#include <net/dst_metadata.h> +#include <linux/netdevice.h> +#include <linux/list.h> +#include <linux/rculist.h> +#include <linux/rtnetlink.h> +#include <linux/workqueue.h> +#include <linux/spinlock.h> +#include "tc.h" +#include "neigh.h" +#include "en_rep.h" +#include "eswitch.h" +#include "esw/chains.h" +#include "en/tc_ct.h" +#include "en/mapping.h" +#include "en/tc_tun.h" +#include "lib/port_tun.h" + +struct mlx5e_rep_indr_block_priv { + struct net_device *netdev; + struct mlx5e_rep_priv *rpriv; + + struct list_head list; +}; + +int mlx5e_rep_encap_entry_attach(struct mlx5e_priv *priv, + struct mlx5e_encap_entry *e) +{ + struct mlx5e_rep_priv *rpriv = priv->ppriv; + struct mlx5_rep_uplink_priv *uplink_priv = &rpriv->uplink_priv; + struct mlx5_tun_entropy *tun_entropy = &uplink_priv->tun_entropy; + struct mlx5e_neigh_hash_entry *nhe; + int err; + + err = mlx5_tun_entropy_refcount_inc(tun_entropy, e->reformat_type); + if (err) + return err; + + mutex_lock(&rpriv->neigh_update.encap_lock); + nhe = mlx5e_rep_neigh_entry_lookup(priv, &e->m_neigh); + if (!nhe) { + err = mlx5e_rep_neigh_entry_create(priv, e, &nhe); + if (err) { + mutex_unlock(&rpriv->neigh_update.encap_lock); + mlx5_tun_entropy_refcount_dec(tun_entropy, + e->reformat_type); + return err; + } + } + + e->nhe = nhe; + spin_lock(&nhe->encap_list_lock); + list_add_rcu(&e->encap_list, &nhe->encap_list); + spin_unlock(&nhe->encap_list_lock); + + mutex_unlock(&rpriv->neigh_update.encap_lock); + + return 0; +} + +void mlx5e_rep_encap_entry_detach(struct mlx5e_priv *priv, + struct mlx5e_encap_entry *e) +{ + struct mlx5e_rep_priv *rpriv = priv->ppriv; + struct mlx5_rep_uplink_priv *uplink_priv = &rpriv->uplink_priv; + struct mlx5_tun_entropy *tun_entropy = &uplink_priv->tun_entropy; + + if (!e->nhe) + return; + + spin_lock(&e->nhe->encap_list_lock); + list_del_rcu(&e->encap_list); + spin_unlock(&e->nhe->encap_list_lock); + + mlx5e_rep_neigh_entry_release(e->nhe); + e->nhe = NULL; + mlx5_tun_entropy_refcount_dec(tun_entropy, e->reformat_type); +} + +void mlx5e_rep_update_flows(struct mlx5e_priv *priv, + struct mlx5e_encap_entry *e, + bool neigh_connected, + unsigned char ha[ETH_ALEN]) +{ + struct ethhdr *eth = (struct ethhdr *)e->encap_header; + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + bool encap_connected; + LIST_HEAD(flow_list); + + ASSERT_RTNL(); + + /* wait for encap to be fully initialized */ + wait_for_completion(&e->res_ready); + + mutex_lock(&esw->offloads.encap_tbl_lock); + encap_connected = !!(e->flags & MLX5_ENCAP_ENTRY_VALID); + if (e->compl_result < 0 || (encap_connected == neigh_connected && + ether_addr_equal(e->h_dest, ha))) + goto unlock; + + mlx5e_take_all_encap_flows(e, &flow_list); + + if ((e->flags & MLX5_ENCAP_ENTRY_VALID) && + (!neigh_connected || !ether_addr_equal(e->h_dest, ha))) + mlx5e_tc_encap_flows_del(priv, e, &flow_list); + + if (neigh_connected && !(e->flags & MLX5_ENCAP_ENTRY_VALID)) { + ether_addr_copy(e->h_dest, ha); + ether_addr_copy(eth->h_dest, ha); + /* Update the encap source mac, in case that we delete + * the flows when encap source mac changed. + */ + ether_addr_copy(eth->h_source, e->route_dev->dev_addr); + + mlx5e_tc_encap_flows_add(priv, e, &flow_list); + } +unlock: + mutex_unlock(&esw->offloads.encap_tbl_lock); + mlx5e_put_encap_flow_list(priv, &flow_list); +} + +static int +mlx5e_rep_setup_tc_cls_flower(struct mlx5e_priv *priv, + struct flow_cls_offload *cls_flower, int flags) +{ + switch (cls_flower->command) { + case FLOW_CLS_REPLACE: + return mlx5e_configure_flower(priv->netdev, priv, cls_flower, + flags); + case FLOW_CLS_DESTROY: + return mlx5e_delete_flower(priv->netdev, priv, cls_flower, + flags); + case FLOW_CLS_STATS: + return mlx5e_stats_flower(priv->netdev, priv, cls_flower, + flags); + default: + return -EOPNOTSUPP; + } +} + +static +int mlx5e_rep_setup_tc_cls_matchall(struct mlx5e_priv *priv, + struct tc_cls_matchall_offload *ma) +{ + switch (ma->command) { + case TC_CLSMATCHALL_REPLACE: + return mlx5e_tc_configure_matchall(priv, ma); + case TC_CLSMATCHALL_DESTROY: + return mlx5e_tc_delete_matchall(priv, ma); + case TC_CLSMATCHALL_STATS: + mlx5e_tc_stats_matchall(priv, ma); + return 0; + default: + return -EOPNOTSUPP; + } +} + +static int mlx5e_rep_setup_tc_cb(enum tc_setup_type type, void *type_data, + void *cb_priv) +{ + unsigned long flags = MLX5_TC_FLAG(INGRESS) | MLX5_TC_FLAG(ESW_OFFLOAD); + struct mlx5e_priv *priv = cb_priv; + + switch (type) { + case TC_SETUP_CLSFLOWER: + return mlx5e_rep_setup_tc_cls_flower(priv, type_data, flags); + case TC_SETUP_CLSMATCHALL: + return mlx5e_rep_setup_tc_cls_matchall(priv, type_data); + default: + return -EOPNOTSUPP; + } +} + +static int mlx5e_rep_setup_ft_cb(enum tc_setup_type type, void *type_data, + void *cb_priv) +{ + struct flow_cls_offload tmp, *f = type_data; + struct mlx5e_priv *priv = cb_priv; + struct mlx5_eswitch *esw; + unsigned long flags; + int err; + + flags = MLX5_TC_FLAG(INGRESS) | + MLX5_TC_FLAG(ESW_OFFLOAD) | + MLX5_TC_FLAG(FT_OFFLOAD); + esw = priv->mdev->priv.eswitch; + + switch (type) { + case TC_SETUP_CLSFLOWER: + memcpy(&tmp, f, sizeof(*f)); + + if (!mlx5_esw_chains_prios_supported(esw)) + return -EOPNOTSUPP; + + /* Re-use tc offload path by moving the ft flow to the + * reserved ft chain. + * + * FT offload can use prio range [0, INT_MAX], so we normalize + * it to range [1, mlx5_esw_chains_get_prio_range(esw)] + * as with tc, where prio 0 isn't supported. + * + * We only support chain 0 of FT offload. + */ + if (tmp.common.prio >= mlx5_esw_chains_get_prio_range(esw)) + return -EOPNOTSUPP; + if (tmp.common.chain_index != 0) + return -EOPNOTSUPP; + + tmp.common.chain_index = mlx5_esw_chains_get_ft_chain(esw); + tmp.common.prio++; + err = mlx5e_rep_setup_tc_cls_flower(priv, &tmp, flags); + memcpy(&f->stats, &tmp.stats, sizeof(f->stats)); + return err; + default: + return -EOPNOTSUPP; + } +} + +static LIST_HEAD(mlx5e_rep_block_tc_cb_list); +static LIST_HEAD(mlx5e_rep_block_ft_cb_list); +int mlx5e_rep_setup_tc(struct net_device *dev, enum tc_setup_type type, + void *type_data) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + struct flow_block_offload *f = type_data; + + f->unlocked_driver_cb = true; + + switch (type) { + case TC_SETUP_BLOCK: + return flow_block_cb_setup_simple(type_data, + &mlx5e_rep_block_tc_cb_list, + mlx5e_rep_setup_tc_cb, + priv, priv, true); + case TC_SETUP_FT: + return flow_block_cb_setup_simple(type_data, + &mlx5e_rep_block_ft_cb_list, + mlx5e_rep_setup_ft_cb, + priv, priv, true); + default: + return -EOPNOTSUPP; + } +} + +int mlx5e_rep_tc_init(struct mlx5e_rep_priv *rpriv) +{ + struct mlx5_rep_uplink_priv *uplink_priv = &rpriv->uplink_priv; + int err; + + mutex_init(&uplink_priv->unready_flows_lock); + INIT_LIST_HEAD(&uplink_priv->unready_flows); + + /* init shared tc flow table */ + err = mlx5e_tc_esw_init(&uplink_priv->tc_ht); + return err; +} + +void mlx5e_rep_tc_cleanup(struct mlx5e_rep_priv *rpriv) +{ + /* delete shared tc flow table */ + mlx5e_tc_esw_cleanup(&rpriv->uplink_priv.tc_ht); + mutex_destroy(&rpriv->uplink_priv.unready_flows_lock); +} + +void mlx5e_rep_tc_enable(struct mlx5e_priv *priv) +{ + struct mlx5e_rep_priv *rpriv = priv->ppriv; + + INIT_WORK(&rpriv->uplink_priv.reoffload_flows_work, + mlx5e_tc_reoffload_flows_work); +} + +void mlx5e_rep_tc_disable(struct mlx5e_priv *priv) +{ + struct mlx5e_rep_priv *rpriv = priv->ppriv; + + cancel_work_sync(&rpriv->uplink_priv.reoffload_flows_work); +} + +int mlx5e_rep_tc_event_port_affinity(struct mlx5e_priv *priv) +{ + struct mlx5e_rep_priv *rpriv = priv->ppriv; + + queue_work(priv->wq, &rpriv->uplink_priv.reoffload_flows_work); + + return NOTIFY_OK; +} + +static struct mlx5e_rep_indr_block_priv * +mlx5e_rep_indr_block_priv_lookup(struct mlx5e_rep_priv *rpriv, + struct net_device *netdev) +{ + struct mlx5e_rep_indr_block_priv *cb_priv; + + /* All callback list access should be protected by RTNL. */ + ASSERT_RTNL(); + + list_for_each_entry(cb_priv, + &rpriv->uplink_priv.tc_indr_block_priv_list, + list) + if (cb_priv->netdev == netdev) + return cb_priv; + + return NULL; +} + +static int +mlx5e_rep_indr_offload(struct net_device *netdev, + struct flow_cls_offload *flower, + struct mlx5e_rep_indr_block_priv *indr_priv, + unsigned long flags) +{ + struct mlx5e_priv *priv = netdev_priv(indr_priv->rpriv->netdev); + int err = 0; + + switch (flower->command) { + case FLOW_CLS_REPLACE: + err = mlx5e_configure_flower(netdev, priv, flower, flags); + break; + case FLOW_CLS_DESTROY: + err = mlx5e_delete_flower(netdev, priv, flower, flags); + break; + case FLOW_CLS_STATS: + err = mlx5e_stats_flower(netdev, priv, flower, flags); + break; + default: + err = -EOPNOTSUPP; + } + + return err; +} + +static int mlx5e_rep_indr_setup_tc_cb(enum tc_setup_type type, + void *type_data, void *indr_priv) +{ + unsigned long flags = MLX5_TC_FLAG(EGRESS) | MLX5_TC_FLAG(ESW_OFFLOAD); + struct mlx5e_rep_indr_block_priv *priv = indr_priv; + + switch (type) { + case TC_SETUP_CLSFLOWER: + return mlx5e_rep_indr_offload(priv->netdev, type_data, priv, + flags); + default: + return -EOPNOTSUPP; + } +} + +static int mlx5e_rep_indr_setup_ft_cb(enum tc_setup_type type, + void *type_data, void *indr_priv) +{ + struct mlx5e_rep_indr_block_priv *priv = indr_priv; + struct flow_cls_offload *f = type_data; + struct flow_cls_offload tmp; + struct mlx5e_priv *mpriv; + struct mlx5_eswitch *esw; + unsigned long flags; + int err; + + mpriv = netdev_priv(priv->rpriv->netdev); + esw = mpriv->mdev->priv.eswitch; + + flags = MLX5_TC_FLAG(EGRESS) | + MLX5_TC_FLAG(ESW_OFFLOAD) | + MLX5_TC_FLAG(FT_OFFLOAD); + + switch (type) { + case TC_SETUP_CLSFLOWER: + memcpy(&tmp, f, sizeof(*f)); + + /* Re-use tc offload path by moving the ft flow to the + * reserved ft chain. + * + * FT offload can use prio range [0, INT_MAX], so we normalize + * it to range [1, mlx5_esw_chains_get_prio_range(esw)] + * as with tc, where prio 0 isn't supported. + * + * We only support chain 0 of FT offload. + */ + if (!mlx5_esw_chains_prios_supported(esw) || + tmp.common.prio >= mlx5_esw_chains_get_prio_range(esw) || + tmp.common.chain_index) + return -EOPNOTSUPP; + + tmp.common.chain_index = mlx5_esw_chains_get_ft_chain(esw); + tmp.common.prio++; + err = mlx5e_rep_indr_offload(priv->netdev, &tmp, priv, flags); + memcpy(&f->stats, &tmp.stats, sizeof(f->stats)); + return err; + default: + return -EOPNOTSUPP; + } +} + +static void mlx5e_rep_indr_block_unbind(void *cb_priv) +{ + struct mlx5e_rep_indr_block_priv *indr_priv = cb_priv; + + list_del(&indr_priv->list); + kfree(indr_priv); +} + +static LIST_HEAD(mlx5e_block_cb_list); + +static int +mlx5e_rep_indr_setup_block(struct net_device *netdev, + struct mlx5e_rep_priv *rpriv, + struct flow_block_offload *f, + flow_setup_cb_t *setup_cb) +{ + struct mlx5e_priv *priv = netdev_priv(rpriv->netdev); + struct mlx5e_rep_indr_block_priv *indr_priv; + struct flow_block_cb *block_cb; + + if (!mlx5e_tc_tun_device_to_offload(priv, netdev) && + !(is_vlan_dev(netdev) && vlan_dev_real_dev(netdev) == rpriv->netdev)) + return -EOPNOTSUPP; + + if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS) + return -EOPNOTSUPP; + + f->unlocked_driver_cb = true; + f->driver_block_list = &mlx5e_block_cb_list; + + switch (f->command) { + case FLOW_BLOCK_BIND: + indr_priv = mlx5e_rep_indr_block_priv_lookup(rpriv, netdev); + if (indr_priv) + return -EEXIST; + + indr_priv = kmalloc(sizeof(*indr_priv), GFP_KERNEL); + if (!indr_priv) + return -ENOMEM; + + indr_priv->netdev = netdev; + indr_priv->rpriv = rpriv; + list_add(&indr_priv->list, + &rpriv->uplink_priv.tc_indr_block_priv_list); + + block_cb = flow_block_cb_alloc(setup_cb, indr_priv, indr_priv, + mlx5e_rep_indr_block_unbind); + if (IS_ERR(block_cb)) { + list_del(&indr_priv->list); + kfree(indr_priv); + return PTR_ERR(block_cb); + } + flow_block_cb_add(block_cb, f); + list_add_tail(&block_cb->driver_list, &mlx5e_block_cb_list); + + return 0; + case FLOW_BLOCK_UNBIND: + indr_priv = mlx5e_rep_indr_block_priv_lookup(rpriv, netdev); + if (!indr_priv) + return -ENOENT; + + block_cb = flow_block_cb_lookup(f->block, setup_cb, indr_priv); + if (!block_cb) + return -ENOENT; + + flow_block_cb_remove(block_cb, f); + list_del(&block_cb->driver_list); + return 0; + default: + return -EOPNOTSUPP; + } + return 0; +} + +static +int mlx5e_rep_indr_setup_cb(struct net_device *netdev, void *cb_priv, + enum tc_setup_type type, void *type_data) +{ + switch (type) { + case TC_SETUP_BLOCK: + return mlx5e_rep_indr_setup_block(netdev, cb_priv, type_data, + mlx5e_rep_indr_setup_tc_cb); + case TC_SETUP_FT: + return mlx5e_rep_indr_setup_block(netdev, cb_priv, type_data, + mlx5e_rep_indr_setup_ft_cb); + default: + return -EOPNOTSUPP; + } +} + +int mlx5e_rep_tc_netdevice_event_register(struct mlx5e_rep_priv *rpriv) +{ + struct mlx5_rep_uplink_priv *uplink_priv = &rpriv->uplink_priv; + + /* init indirect block notifications */ + INIT_LIST_HEAD(&uplink_priv->tc_indr_block_priv_list); + + return flow_indr_dev_register(mlx5e_rep_indr_setup_cb, rpriv); +} + +void mlx5e_rep_tc_netdevice_event_unregister(struct mlx5e_rep_priv *rpriv) +{ + flow_indr_dev_unregister(mlx5e_rep_indr_setup_cb, rpriv, + mlx5e_rep_indr_setup_tc_cb); +} + +#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) +static bool mlx5e_restore_tunnel(struct mlx5e_priv *priv, struct sk_buff *skb, + struct mlx5e_tc_update_priv *tc_priv, + u32 tunnel_id) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct tunnel_match_enc_opts enc_opts = {}; + struct mlx5_rep_uplink_priv *uplink_priv; + struct mlx5e_rep_priv *uplink_rpriv; + struct metadata_dst *tun_dst; + struct tunnel_match_key key; + u32 tun_id, enc_opts_id; + struct net_device *dev; + int err; + + enc_opts_id = tunnel_id & ENC_OPTS_BITS_MASK; + tun_id = tunnel_id >> ENC_OPTS_BITS; + + if (!tun_id) + return true; + + uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); + uplink_priv = &uplink_rpriv->uplink_priv; + + err = mapping_find(uplink_priv->tunnel_mapping, tun_id, &key); + if (err) { + WARN_ON_ONCE(true); + netdev_dbg(priv->netdev, + "Couldn't find tunnel for tun_id: %d, err: %d\n", + tun_id, err); + return false; + } + + if (enc_opts_id) { + err = mapping_find(uplink_priv->tunnel_enc_opts_mapping, + enc_opts_id, &enc_opts); + if (err) { + netdev_dbg(priv->netdev, + "Couldn't find tunnel (opts) for tun_id: %d, err: %d\n", + enc_opts_id, err); + return false; + } + } + + tun_dst = tun_rx_dst(enc_opts.key.len); + if (!tun_dst) { + WARN_ON_ONCE(true); + return false; + } + + ip_tunnel_key_init(&tun_dst->u.tun_info.key, + key.enc_ipv4.src, key.enc_ipv4.dst, + key.enc_ip.tos, key.enc_ip.ttl, + 0, /* label */ + key.enc_tp.src, key.enc_tp.dst, + key32_to_tunnel_id(key.enc_key_id.keyid), + TUNNEL_KEY); + + if (enc_opts.key.len) + ip_tunnel_info_opts_set(&tun_dst->u.tun_info, + enc_opts.key.data, + enc_opts.key.len, + enc_opts.key.dst_opt_type); + + skb_dst_set(skb, (struct dst_entry *)tun_dst); + dev = dev_get_by_index(&init_net, key.filter_ifindex); + if (!dev) { + netdev_dbg(priv->netdev, + "Couldn't find tunnel device with ifindex: %d\n", + key.filter_ifindex); + return false; + } + + /* Set tun_dev so we do dev_put() after datapath */ + tc_priv->tun_dev = dev; + + skb->dev = dev; + + return true; +} +#endif /* CONFIG_NET_TC_SKB_EXT */ + +bool mlx5e_rep_tc_update_skb(struct mlx5_cqe64 *cqe, + struct sk_buff *skb, + struct mlx5e_tc_update_priv *tc_priv) +{ +#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) + u32 chain = 0, reg_c0, reg_c1, tunnel_id, tuple_id; + struct mlx5_rep_uplink_priv *uplink_priv; + struct mlx5e_rep_priv *uplink_rpriv; + struct tc_skb_ext *tc_skb_ext; + struct mlx5_eswitch *esw; + struct mlx5e_priv *priv; + int tunnel_moffset; + int err; + + reg_c0 = (be32_to_cpu(cqe->sop_drop_qpn) & MLX5E_TC_FLOW_ID_MASK); + if (reg_c0 == MLX5_FS_DEFAULT_FLOW_TAG) + reg_c0 = 0; + reg_c1 = be32_to_cpu(cqe->ft_metadata); + + if (!reg_c0) + return true; + + priv = netdev_priv(skb->dev); + esw = priv->mdev->priv.eswitch; + + err = mlx5_eswitch_get_chain_for_tag(esw, reg_c0, &chain); + if (err) { + netdev_dbg(priv->netdev, + "Couldn't find chain for chain tag: %d, err: %d\n", + reg_c0, err); + return false; + } + + if (chain) { + tc_skb_ext = skb_ext_add(skb, TC_SKB_EXT); + if (!tc_skb_ext) { + WARN_ON(1); + return false; + } + + tc_skb_ext->chain = chain; + + tuple_id = reg_c1 & TUPLE_ID_MAX; + + uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); + uplink_priv = &uplink_rpriv->uplink_priv; + if (!mlx5e_tc_ct_restore_flow(uplink_priv, skb, tuple_id)) + return false; + } + + tunnel_moffset = mlx5e_tc_attr_to_reg_mappings[TUNNEL_TO_REG].moffset; + tunnel_id = reg_c1 >> (8 * tunnel_moffset); + return mlx5e_restore_tunnel(priv, skb, tc_priv, tunnel_id); +#endif /* CONFIG_NET_TC_SKB_EXT */ + + return true; +} + +void mlx5_rep_tc_post_napi_receive(struct mlx5e_tc_update_priv *tc_priv) +{ + if (tc_priv->tun_dev) + dev_put(tc_priv->tun_dev); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.h b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.h new file mode 100644 index 000000000000..fdf9702c2d7d --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/rep/tc.h @@ -0,0 +1,77 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* Copyright (c) 2020 Mellanox Technologies. */ + +#ifndef __MLX5_EN_REP_TC_H__ +#define __MLX5_EN_REP_TC_H__ + +#include <linux/skbuff.h> +#include "en_tc.h" +#include "en_rep.h" + +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) + +int mlx5e_rep_tc_init(struct mlx5e_rep_priv *rpriv); +void mlx5e_rep_tc_cleanup(struct mlx5e_rep_priv *rpriv); + +int mlx5e_rep_tc_netdevice_event_register(struct mlx5e_rep_priv *rpriv); +void mlx5e_rep_tc_netdevice_event_unregister(struct mlx5e_rep_priv *rpriv); + +void mlx5e_rep_tc_enable(struct mlx5e_priv *priv); +void mlx5e_rep_tc_disable(struct mlx5e_priv *priv); + +int mlx5e_rep_tc_event_port_affinity(struct mlx5e_priv *priv); + +void mlx5e_rep_update_flows(struct mlx5e_priv *priv, + struct mlx5e_encap_entry *e, + bool neigh_connected, + unsigned char ha[ETH_ALEN]); + +int mlx5e_rep_encap_entry_attach(struct mlx5e_priv *priv, + struct mlx5e_encap_entry *e); +void mlx5e_rep_encap_entry_detach(struct mlx5e_priv *priv, + struct mlx5e_encap_entry *e); + +int mlx5e_rep_setup_tc(struct net_device *dev, enum tc_setup_type type, + void *type_data); + +bool mlx5e_rep_tc_update_skb(struct mlx5_cqe64 *cqe, + struct sk_buff *skb, + struct mlx5e_tc_update_priv *tc_priv); +void mlx5_rep_tc_post_napi_receive(struct mlx5e_tc_update_priv *tc_priv); + +#else /* CONFIG_MLX5_CLS_ACT */ + +struct mlx5e_rep_priv; +static inline int +mlx5e_rep_tc_init(struct mlx5e_rep_priv *rpriv) { return 0; } +static inline void +mlx5e_rep_tc_cleanup(struct mlx5e_rep_priv *rpriv) {} + +static inline int +mlx5e_rep_tc_netdevice_event_register(struct mlx5e_rep_priv *rpriv) { return 0; } +static inline void +mlx5e_rep_tc_netdevice_event_unregister(struct mlx5e_rep_priv *rpriv) {} + +static inline void +mlx5e_rep_tc_enable(struct mlx5e_priv *priv) {} +static inline void +mlx5e_rep_tc_disable(struct mlx5e_priv *priv) {} + +static inline int +mlx5e_rep_tc_event_port_affinity(struct mlx5e_priv *priv) { return NOTIFY_DONE; } + +static inline int +mlx5e_rep_setup_tc(struct net_device *dev, enum tc_setup_type type, + void *type_data) { return -EOPNOTSUPP; } + +struct mlx5e_tc_update_priv; +static inline bool +mlx5e_rep_tc_update_skb(struct mlx5_cqe64 *cqe, + struct sk_buff *skb, + struct mlx5e_tc_update_priv *tc_priv) { return true; } +static inline void +mlx5_rep_tc_post_napi_receive(struct mlx5e_tc_update_priv *tc_priv) {} + +#endif /* CONFIG_MLX5_CLS_ACT */ + +#endif /* __MLX5_EN_REP_TC_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c index 1079558d292a..afc19dca1f5f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c @@ -24,6 +24,7 @@ #define MLX5_CT_ZONE_MASK GENMASK(MLX5_CT_ZONE_BITS - 1, 0) #define MLX5_CT_STATE_ESTABLISHED_BIT BIT(1) #define MLX5_CT_STATE_TRK_BIT BIT(2) +#define MLX5_CT_STATE_NAT_BIT BIT(3) #define MLX5_FTE_ID_BITS (mlx5e_tc_attr_to_reg_mappings[FTEID_TO_REG].mlen * 8) #define MLX5_FTE_ID_MAX GENMASK(MLX5_FTE_ID_BITS - 1, 0) @@ -61,6 +62,15 @@ struct mlx5_ct_zone_rule { bool nat; }; +struct mlx5_tc_ct_pre { + struct mlx5_flow_table *fdb; + struct mlx5_flow_group *flow_grp; + struct mlx5_flow_group *miss_grp; + struct mlx5_flow_handle *flow_rule; + struct mlx5_flow_handle *miss_rule; + struct mlx5_modify_hdr *modify_hdr; +}; + struct mlx5_ct_ft { struct rhash_head node; u16 zone; @@ -68,14 +78,14 @@ struct mlx5_ct_ft { struct nf_flowtable *nf_ft; struct mlx5_tc_ct_priv *ct_priv; struct rhashtable ct_entries_ht; + struct mlx5_tc_ct_pre pre_ct; + struct mlx5_tc_ct_pre pre_ct_nat; }; struct mlx5_ct_entry { u16 zone; struct rhash_head node; - struct flow_rule *flow_rule; struct mlx5_fc *counter; - unsigned long lastuse; unsigned long cookie; unsigned long restore_cookie; struct mlx5_ct_zone_rule zone_rules[2]; @@ -109,7 +119,7 @@ mlx5_tc_ct_get_ct_priv(struct mlx5e_priv *priv) } static int -mlx5_tc_ct_set_tuple_match(struct mlx5_flow_spec *spec, +mlx5_tc_ct_set_tuple_match(struct mlx5e_priv *priv, struct mlx5_flow_spec *spec, struct flow_rule *rule) { void *headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, @@ -124,10 +134,8 @@ mlx5_tc_ct_set_tuple_match(struct mlx5_flow_spec *spec, flow_rule_match_basic(rule, &match); - MLX5_SET(fte_match_set_lyr_2_4, headers_c, ethertype, - ntohs(match.mask->n_proto)); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, - ntohs(match.key->n_proto)); + mlx5e_tc_set_ethertype(priv->mdev, &match, true, headers_c, + headers_v); MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_protocol, match.mask->ip_proto); MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_protocol, @@ -428,6 +436,7 @@ mlx5_tc_ct_entry_create_mod_hdr(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_eswitch *esw = ct_priv->esw; struct mlx5_modify_hdr *mod_hdr; struct flow_action_entry *meta; + u16 ct_state = 0; int err; meta = mlx5_tc_ct_get_ct_metadata_action(flow_rule); @@ -446,11 +455,13 @@ mlx5_tc_ct_entry_create_mod_hdr(struct mlx5_tc_ct_priv *ct_priv, &mod_acts); if (err) goto err_mapping; + + ct_state |= MLX5_CT_STATE_NAT_BIT; } + ct_state |= MLX5_CT_STATE_ESTABLISHED_BIT | MLX5_CT_STATE_TRK_BIT; err = mlx5_tc_ct_entry_set_registers(ct_priv, &mod_acts, - (MLX5_CT_STATE_ESTABLISHED_BIT | - MLX5_CT_STATE_TRK_BIT), + ct_state, meta->ct_metadata.mark, meta->ct_metadata.labels[0], tupleid); @@ -520,7 +531,7 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv, attr->counter = entry->counter; attr->flags |= MLX5_ESW_ATTR_FLAG_NO_IN_PORT; - mlx5_tc_ct_set_tuple_match(spec, flow_rule); + mlx5_tc_ct_set_tuple_match(netdev_priv(ct_priv->netdev), spec, flow_rule); mlx5e_tc_match_to_reg_match(spec, ZONE_TO_REG, entry->zone & MLX5_CT_ZONE_MASK, MLX5_CT_ZONE_MASK); @@ -603,7 +614,6 @@ mlx5_tc_ct_block_flow_offload_add(struct mlx5_ct_ft *ft, return -ENOMEM; entry->zone = ft->zone; - entry->flow_rule = flow_rule; entry->cookie = flow->cookie; entry->restore_cookie = meta_action->ct_metadata.cookie; @@ -687,7 +697,7 @@ mlx5_tc_ct_block_flow_offload(enum tc_setup_type type, void *type_data, return mlx5_tc_ct_block_flow_offload_stats(ft, f); default: break; - }; + } return -EOPNOTSUPP; } @@ -699,6 +709,7 @@ mlx5_tc_ct_parse_match(struct mlx5e_priv *priv, struct netlink_ext_ack *extack) { struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); + struct flow_rule *rule = flow_cls_offload_flow_rule(f); struct flow_dissector_key_ct *mask, *key; bool trk, est, untrk, unest, new; u32 ctstate = 0, ctstate_mask = 0; @@ -706,7 +717,7 @@ mlx5_tc_ct_parse_match(struct mlx5e_priv *priv, u16 ct_state, ct_state_mask; struct flow_match_ct match; - if (!flow_rule_match_key(f->rule, FLOW_DISSECTOR_KEY_CT)) + if (!flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_CT)) return 0; if (!ct_priv) { @@ -715,7 +726,7 @@ mlx5_tc_ct_parse_match(struct mlx5e_priv *priv, return -EOPNOTSUPP; } - flow_rule_match_ct(f->rule, &match); + flow_rule_match_ct(rule, &match); key = match.key; mask = match.mask; @@ -794,6 +805,238 @@ mlx5_tc_ct_parse_action(struct mlx5e_priv *priv, return 0; } +static int tc_ct_pre_ct_add_rules(struct mlx5_ct_ft *ct_ft, + struct mlx5_tc_ct_pre *pre_ct, + bool nat) +{ + struct mlx5_tc_ct_priv *ct_priv = ct_ft->ct_priv; + struct mlx5e_tc_mod_hdr_acts pre_mod_acts = {}; + struct mlx5_core_dev *dev = ct_priv->esw->dev; + struct mlx5_flow_table *fdb = pre_ct->fdb; + struct mlx5_flow_destination dest = {}; + struct mlx5_flow_act flow_act = {}; + struct mlx5_modify_hdr *mod_hdr; + struct mlx5_flow_handle *rule; + struct mlx5_flow_spec *spec; + u32 ctstate; + u16 zone; + int err; + + spec = kvzalloc(sizeof(*spec), GFP_KERNEL); + if (!spec) + return -ENOMEM; + + zone = ct_ft->zone & MLX5_CT_ZONE_MASK; + err = mlx5e_tc_match_to_reg_set(dev, &pre_mod_acts, ZONE_TO_REG, zone); + if (err) { + ct_dbg("Failed to set zone register mapping"); + goto err_mapping; + } + + mod_hdr = mlx5_modify_header_alloc(dev, + MLX5_FLOW_NAMESPACE_FDB, + pre_mod_acts.num_actions, + pre_mod_acts.actions); + + if (IS_ERR(mod_hdr)) { + err = PTR_ERR(mod_hdr); + ct_dbg("Failed to create pre ct mod hdr"); + goto err_mapping; + } + pre_ct->modify_hdr = mod_hdr; + + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST | + MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; + flow_act.flags |= FLOW_ACT_IGNORE_FLOW_LEVEL; + flow_act.modify_hdr = mod_hdr; + dest.type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE; + + /* add flow rule */ + mlx5e_tc_match_to_reg_match(spec, ZONE_TO_REG, + zone, MLX5_CT_ZONE_MASK); + ctstate = MLX5_CT_STATE_TRK_BIT; + if (nat) + ctstate |= MLX5_CT_STATE_NAT_BIT; + mlx5e_tc_match_to_reg_match(spec, CTSTATE_TO_REG, ctstate, ctstate); + + dest.ft = ct_priv->post_ct; + rule = mlx5_add_flow_rules(fdb, spec, &flow_act, &dest, 1); + if (IS_ERR(rule)) { + err = PTR_ERR(rule); + ct_dbg("Failed to add pre ct flow rule zone %d", zone); + goto err_flow_rule; + } + pre_ct->flow_rule = rule; + + /* add miss rule */ + memset(spec, 0, sizeof(*spec)); + dest.ft = nat ? ct_priv->ct_nat : ct_priv->ct; + rule = mlx5_add_flow_rules(fdb, spec, &flow_act, &dest, 1); + if (IS_ERR(rule)) { + err = PTR_ERR(rule); + ct_dbg("Failed to add pre ct miss rule zone %d", zone); + goto err_miss_rule; + } + pre_ct->miss_rule = rule; + + dealloc_mod_hdr_actions(&pre_mod_acts); + kvfree(spec); + return 0; + +err_miss_rule: + mlx5_del_flow_rules(pre_ct->flow_rule); +err_flow_rule: + mlx5_modify_header_dealloc(dev, pre_ct->modify_hdr); +err_mapping: + dealloc_mod_hdr_actions(&pre_mod_acts); + kvfree(spec); + return err; +} + +static void +tc_ct_pre_ct_del_rules(struct mlx5_ct_ft *ct_ft, + struct mlx5_tc_ct_pre *pre_ct) +{ + struct mlx5_tc_ct_priv *ct_priv = ct_ft->ct_priv; + struct mlx5_core_dev *dev = ct_priv->esw->dev; + + mlx5_del_flow_rules(pre_ct->flow_rule); + mlx5_del_flow_rules(pre_ct->miss_rule); + mlx5_modify_header_dealloc(dev, pre_ct->modify_hdr); +} + +static int +mlx5_tc_ct_alloc_pre_ct(struct mlx5_ct_ft *ct_ft, + struct mlx5_tc_ct_pre *pre_ct, + bool nat) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_tc_ct_priv *ct_priv = ct_ft->ct_priv; + struct mlx5_core_dev *dev = ct_priv->esw->dev; + struct mlx5_flow_table_attr ft_attr = {}; + struct mlx5_flow_namespace *ns; + struct mlx5_flow_table *ft; + struct mlx5_flow_group *g; + u32 metadata_reg_c_2_mask; + u32 *flow_group_in; + void *misc; + int err; + + ns = mlx5_get_flow_namespace(dev, MLX5_FLOW_NAMESPACE_FDB); + if (!ns) { + err = -EOPNOTSUPP; + ct_dbg("Failed to get FDB flow namespace"); + return err; + } + + flow_group_in = kvzalloc(inlen, GFP_KERNEL); + if (!flow_group_in) + return -ENOMEM; + + ft_attr.flags = MLX5_FLOW_TABLE_UNMANAGED; + ft_attr.prio = FDB_TC_OFFLOAD; + ft_attr.max_fte = 2; + ft_attr.level = 1; + ft = mlx5_create_flow_table(ns, &ft_attr); + if (IS_ERR(ft)) { + err = PTR_ERR(ft); + ct_dbg("Failed to create pre ct table"); + goto out_free; + } + pre_ct->fdb = ft; + + /* create flow group */ + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 0); + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, + MLX5_MATCH_MISC_PARAMETERS_2); + + misc = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, + match_criteria.misc_parameters_2); + + metadata_reg_c_2_mask = MLX5_CT_ZONE_MASK; + metadata_reg_c_2_mask |= (MLX5_CT_STATE_TRK_BIT << 16); + if (nat) + metadata_reg_c_2_mask |= (MLX5_CT_STATE_NAT_BIT << 16); + + MLX5_SET(fte_match_set_misc2, misc, metadata_reg_c_2, + metadata_reg_c_2_mask); + + g = mlx5_create_flow_group(ft, flow_group_in); + if (IS_ERR(g)) { + err = PTR_ERR(g); + ct_dbg("Failed to create pre ct group"); + goto err_flow_grp; + } + pre_ct->flow_grp = g; + + /* create miss group */ + memset(flow_group_in, 0, inlen); + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 1); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 1); + g = mlx5_create_flow_group(ft, flow_group_in); + if (IS_ERR(g)) { + err = PTR_ERR(g); + ct_dbg("Failed to create pre ct miss group"); + goto err_miss_grp; + } + pre_ct->miss_grp = g; + + err = tc_ct_pre_ct_add_rules(ct_ft, pre_ct, nat); + if (err) + goto err_add_rules; + + kvfree(flow_group_in); + return 0; + +err_add_rules: + mlx5_destroy_flow_group(pre_ct->miss_grp); +err_miss_grp: + mlx5_destroy_flow_group(pre_ct->flow_grp); +err_flow_grp: + mlx5_destroy_flow_table(ft); +out_free: + kvfree(flow_group_in); + return err; +} + +static void +mlx5_tc_ct_free_pre_ct(struct mlx5_ct_ft *ct_ft, + struct mlx5_tc_ct_pre *pre_ct) +{ + tc_ct_pre_ct_del_rules(ct_ft, pre_ct); + mlx5_destroy_flow_group(pre_ct->miss_grp); + mlx5_destroy_flow_group(pre_ct->flow_grp); + mlx5_destroy_flow_table(pre_ct->fdb); +} + +static int +mlx5_tc_ct_alloc_pre_ct_tables(struct mlx5_ct_ft *ft) +{ + int err; + + err = mlx5_tc_ct_alloc_pre_ct(ft, &ft->pre_ct, false); + if (err) + return err; + + err = mlx5_tc_ct_alloc_pre_ct(ft, &ft->pre_ct_nat, true); + if (err) + goto err_pre_ct_nat; + + return 0; + +err_pre_ct_nat: + mlx5_tc_ct_free_pre_ct(ft, &ft->pre_ct); + return err; +} + +static void +mlx5_tc_ct_free_pre_ct_tables(struct mlx5_ct_ft *ft) +{ + mlx5_tc_ct_free_pre_ct(ft, &ft->pre_ct_nat); + mlx5_tc_ct_free_pre_ct(ft, &ft->pre_ct); +} + static struct mlx5_ct_ft * mlx5_tc_ct_add_ft_cb(struct mlx5_tc_ct_priv *ct_priv, u16 zone, struct nf_flowtable *nf_ft) @@ -816,6 +1059,10 @@ mlx5_tc_ct_add_ft_cb(struct mlx5_tc_ct_priv *ct_priv, u16 zone, ft->ct_priv = ct_priv; refcount_set(&ft->refcount, 1); + err = mlx5_tc_ct_alloc_pre_ct_tables(ft); + if (err) + goto err_alloc_pre_ct; + err = rhashtable_init(&ft->ct_entries_ht, &cts_ht_params); if (err) goto err_init; @@ -837,6 +1084,8 @@ err_add_cb: err_insert: rhashtable_destroy(&ft->ct_entries_ht); err_init: + mlx5_tc_ct_free_pre_ct_tables(ft); +err_alloc_pre_ct: kfree(ft); return ERR_PTR(err); } @@ -862,21 +1111,40 @@ mlx5_tc_ct_del_ft_cb(struct mlx5_tc_ct_priv *ct_priv, struct mlx5_ct_ft *ft) rhashtable_free_and_destroy(&ft->ct_entries_ht, mlx5_tc_ct_flush_ft_entry, ct_priv); + mlx5_tc_ct_free_pre_ct_tables(ft); kfree(ft); } /* We translate the tc filter with CT action to the following HW model: * - * +-------------------+ +--------------------+ +--------------+ - * + pre_ct (tc chain) +----->+ CT (nat or no nat) +--->+ post_ct +-----> - * + original match + | + tuple + zone match + | + fte_id match + | - * +-------------------+ | +--------------------+ | +--------------+ | - * v v v - * set chain miss mapping set mark original - * set fte_id set label filter - * set zone set established actions - * set tunnel_id do nat (if needed) - * do decap + * +---------------------+ + * + fdb prio (tc chain) + + * + original match + + * +---------------------+ + * | set chain miss mapping + * | set fte_id + * | set tunnel_id + * | do decap + * v + * +---------------------+ + * + pre_ct/pre_ct_nat + if matches +---------------------+ + * + zone+nat match +---------------->+ post_ct (see below) + + * +---------------------+ set zone +---------------------+ + * | set zone + * v + * +--------------------+ + * + CT (nat or no nat) + + * + tuple + zone match + + * +--------------------+ + * | set mark + * | set label + * | set established + * | do nat (if needed) + * v + * +--------------+ + * + post_ct + original filter actions + * + fte_id match +------------------------> + * +--------------+ */ static int __mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, @@ -891,7 +1159,7 @@ __mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, struct mlx5_flow_spec *post_ct_spec = NULL; struct mlx5_eswitch *esw = ct_priv->esw; struct mlx5_esw_flow_attr *pre_ct_attr; - struct mlx5_modify_hdr *mod_hdr; + struct mlx5_modify_hdr *mod_hdr; struct mlx5_flow_handle *rule; struct mlx5_ct_flow *ct_flow; int chain_mapping = 0, err; @@ -954,14 +1222,6 @@ __mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, goto err_mapping; } - err = mlx5e_tc_match_to_reg_set(esw->dev, &pre_mod_acts, ZONE_TO_REG, - attr->ct_attr.zone & - MLX5_CT_ZONE_MASK); - if (err) { - ct_dbg("Failed to set zone register mapping"); - goto err_mapping; - } - err = mlx5e_tc_match_to_reg_set(esw->dev, &pre_mod_acts, FTEID_TO_REG, fte_id); if (err) { @@ -1021,7 +1281,7 @@ __mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, /* Change original rule point to ct table */ pre_ct_attr->dest_chain = 0; - pre_ct_attr->dest_ft = nat ? ct_priv->ct_nat : ct_priv->ct; + pre_ct_attr->dest_ft = nat ? ft->pre_ct_nat.fdb : ft->pre_ct.fdb; ct_flow->pre_ct_rule = mlx5_eswitch_add_offloaded_rule(esw, orig_spec, pre_ct_attr); @@ -1131,7 +1391,7 @@ mlx5_tc_ct_flow_offload(struct mlx5e_priv *priv, { bool clear_action = attr->ct_attr.ct_action & TCA_CT_ACT_CLEAR; struct mlx5_tc_ct_priv *ct_priv = mlx5_tc_ct_get_ct_priv(priv); - struct mlx5_flow_handle *rule; + struct mlx5_flow_handle *rule = ERR_PTR(-EINVAL); int err; if (!ct_priv) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h index 091d305b633e..626f6c04882e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.h @@ -130,7 +130,9 @@ mlx5_tc_ct_parse_match(struct mlx5e_priv *priv, struct flow_cls_offload *f, struct netlink_ext_ack *extack) { - if (!flow_rule_match_key(f->rule, FLOW_DISSECTOR_KEY_CT)) + struct flow_rule *rule = flow_cls_offload_flow_rule(f); + + if (!flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_CT)) return 0; NL_SET_ERR_MSG_MOD(extack, "mlx5 tc ct offload isn't enabled."); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c index b45c3f46570b..7cce85faa16f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c @@ -4,8 +4,11 @@ #include <net/vxlan.h> #include <net/gre.h> #include <net/geneve.h> +#include <net/bareudp.h> #include "en/tc_tun.h" #include "en_tc.h" +#include "rep/tc.h" +#include "rep/neigh.h" struct mlx5e_tc_tunnel *mlx5e_get_tc_tun(struct net_device *tunnel_dev) { @@ -16,6 +19,8 @@ struct mlx5e_tc_tunnel *mlx5e_get_tc_tun(struct net_device *tunnel_dev) else if (netif_is_gretap(tunnel_dev) || netif_is_ip6gretap(tunnel_dev)) return &gre_tunnel; + else if (netif_is_bareudp(tunnel_dev)) + return &mplsoudp_tunnel; else return NULL; } @@ -96,9 +101,8 @@ static int mlx5e_route_lookup_ipv4(struct mlx5e_priv *priv, } rt = ip_route_output_key(dev_net(mirred_dev), fl4); - ret = PTR_ERR_OR_ZERO(rt); - if (ret) - return ret; + if (IS_ERR(rt)) + return PTR_ERR(rt); if (mlx5_lag_is_multipath(mdev) && rt->rt_gw_family != AF_INET) { ip_rt_put(rt); @@ -508,6 +512,13 @@ int mlx5e_tc_tun_parse(struct net_device *filter_dev, } if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_CONTROL)) { + struct flow_dissector_key_basic key_basic = {}; + struct flow_dissector_key_basic mask_basic = { + .n_proto = htons(0xFFFF), + }; + struct flow_match_basic match_basic = { + .key = &key_basic, .mask = &mask_basic, + }; struct flow_match_control match; u16 addr_type; @@ -533,10 +544,9 @@ int mlx5e_tc_tun_parse(struct net_device *filter_dev, dst_ipv4_dst_ipv6.ipv4_layout.ipv4, ntohl(match.key->dst)); - MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, - ethertype); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, - ETH_P_IP); + key_basic.n_proto = htons(ETH_P_IP); + mlx5e_tc_set_ethertype(priv->mdev, &match_basic, true, + headers_c, headers_v); } else if (addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) { struct flow_match_ipv6_addrs match; @@ -559,10 +569,9 @@ int mlx5e_tc_tun_parse(struct net_device *filter_dev, &match.key->dst, MLX5_FLD_SZ_BYTES(ipv6_layout, ipv6)); - MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, - ethertype); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, - ETH_P_IPV6); + key_basic.n_proto = htons(ETH_P_IPV6); + mlx5e_tc_set_ethertype(priv->mdev, &match_basic, true, + headers_c, headers_v); } } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h index 1630f0ec3ad7..704359df6095 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.h @@ -16,6 +16,7 @@ enum { MLX5E_TC_TUNNEL_TYPE_VXLAN, MLX5E_TC_TUNNEL_TYPE_GENEVE, MLX5E_TC_TUNNEL_TYPE_GRETAP, + MLX5E_TC_TUNNEL_TYPE_MPLSOUDP, }; struct mlx5e_tc_tunnel { @@ -46,6 +47,7 @@ struct mlx5e_tc_tunnel { extern struct mlx5e_tc_tunnel vxlan_tunnel; extern struct mlx5e_tc_tunnel geneve_tunnel; extern struct mlx5e_tc_tunnel gre_tunnel; +extern struct mlx5e_tc_tunnel mplsoudp_tunnel; struct mlx5e_tc_tunnel *mlx5e_get_tc_tun(struct net_device *tunnel_dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_mplsoudp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_mplsoudp.c new file mode 100644 index 000000000000..1f9526244222 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_mplsoudp.c @@ -0,0 +1,134 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2018 Mellanox Technologies. */ + +#include <net/bareudp.h> +#include <net/mpls.h> +#include "en/tc_tun.h" + +static bool can_offload(struct mlx5e_priv *priv) +{ + return MLX5_CAP_ESW_FLOWTABLE_FDB(priv->mdev, reformat_l3_tunnel_to_l2); +} + +static int calc_hlen(struct mlx5e_encap_entry *e) +{ + return sizeof(struct udphdr) + MPLS_HLEN; +} + +static int init_encap_attr(struct net_device *tunnel_dev, + struct mlx5e_priv *priv, + struct mlx5e_encap_entry *e, + struct netlink_ext_ack *extack) +{ + e->tunnel = &mplsoudp_tunnel; + e->reformat_type = MLX5_REFORMAT_TYPE_L2_TO_L3_TUNNEL; + return 0; +} + +static int generate_ip_tun_hdr(char buf[], + __u8 *ip_proto, + struct mlx5e_encap_entry *r) +{ + const struct ip_tunnel_key *tun_key = &r->tun_info->key; + struct udphdr *udp = (struct udphdr *)(buf); + struct mpls_shim_hdr *mpls; + u32 tun_id; + + tun_id = be32_to_cpu(tunnel_id_to_key32(tun_key->tun_id)); + mpls = (struct mpls_shim_hdr *)(udp + 1); + *ip_proto = IPPROTO_UDP; + + udp->dest = tun_key->tp_dst; + *mpls = mpls_entry_encode(tun_id, tun_key->ttl, tun_key->tos, true); + + return 0; +} + +static int parse_udp_ports(struct mlx5e_priv *priv, + struct mlx5_flow_spec *spec, + struct flow_cls_offload *f, + void *headers_c, + void *headers_v) +{ + return mlx5e_tc_tun_parse_udp_ports(priv, spec, f, headers_c, headers_v); +} + +static int parse_tunnel(struct mlx5e_priv *priv, + struct mlx5_flow_spec *spec, + struct flow_cls_offload *f, + void *headers_c, + void *headers_v) +{ + struct flow_rule *rule = flow_cls_offload_flow_rule(f); + struct flow_match_enc_keyid enc_keyid; + struct flow_match_mpls match; + void *misc2_c; + void *misc2_v; + + misc2_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + misc_parameters_2); + misc2_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, + misc_parameters_2); + + if (!flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_MPLS)) + return 0; + + if (!flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_KEYID)) + return 0; + + flow_rule_match_enc_keyid(rule, &enc_keyid); + + if (!enc_keyid.mask->keyid) + return 0; + + if (!(MLX5_CAP_GEN(priv->mdev, flex_parser_protocols) & + MLX5_FLEX_PROTO_CW_MPLS_UDP)) + return -EOPNOTSUPP; + + flow_rule_match_mpls(rule, &match); + + /* Only support matching the first LSE */ + if (match.mask->used_lses != 1) + return -EOPNOTSUPP; + + MLX5_SET(fte_match_set_misc2, misc2_c, + outer_first_mpls_over_udp.mpls_label, + match.mask->ls[0].mpls_label); + MLX5_SET(fte_match_set_misc2, misc2_v, + outer_first_mpls_over_udp.mpls_label, + match.key->ls[0].mpls_label); + + MLX5_SET(fte_match_set_misc2, misc2_c, + outer_first_mpls_over_udp.mpls_exp, + match.mask->ls[0].mpls_tc); + MLX5_SET(fte_match_set_misc2, misc2_v, + outer_first_mpls_over_udp.mpls_exp, match.key->ls[0].mpls_tc); + + MLX5_SET(fte_match_set_misc2, misc2_c, + outer_first_mpls_over_udp.mpls_s_bos, + match.mask->ls[0].mpls_bos); + MLX5_SET(fte_match_set_misc2, misc2_v, + outer_first_mpls_over_udp.mpls_s_bos, + match.key->ls[0].mpls_bos); + + MLX5_SET(fte_match_set_misc2, misc2_c, + outer_first_mpls_over_udp.mpls_ttl, + match.mask->ls[0].mpls_ttl); + MLX5_SET(fte_match_set_misc2, misc2_v, + outer_first_mpls_over_udp.mpls_ttl, + match.key->ls[0].mpls_ttl); + spec->match_criteria_enable |= MLX5_MATCH_MISC_PARAMETERS_2; + + return 0; +} + +struct mlx5e_tc_tunnel mplsoudp_tunnel = { + .tunnel_type = MLX5E_TC_TUNNEL_TYPE_MPLSOUDP, + .match_level = MLX5_MATCH_L4, + .can_offload = can_offload, + .calc_hlen = calc_hlen, + .init_encap_attr = init_encap_attr, + .generate_ip_tun_hdr = generate_ip_tun_hdr, + .parse_udp_ports = parse_udp_ports, + .parse_tunnel = parse_tunnel, +}; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h index f07b1399744e..bfd3e1161bc6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h @@ -6,46 +6,32 @@ #include "en.h" -#define MLX5E_SQ_NOPS_ROOM (MLX5_SEND_WQE_MAX_WQEBBS - 1) -#define MLX5E_SQ_STOP_ROOM (MLX5_SEND_WQE_MAX_WQEBBS +\ - MLX5E_SQ_NOPS_ROOM) - -#ifndef CONFIG_MLX5_EN_TLS -#define MLX5E_SQ_TLS_ROOM (0) -#else -/* TLS offload requires additional stop_room for: - * - a resync SKB. - * kTLS offload requires fixed additional stop_room for: - * - a static params WQE, and a progress params WQE. - * The additional MTU-depending room for the resync DUMP WQEs - * will be calculated and added in runtime. - */ -#define MLX5E_SQ_TLS_ROOM \ - (MLX5_SEND_WQE_MAX_WQEBBS + \ - MLX5E_KTLS_STATIC_WQEBBS + MLX5E_KTLS_PROGRESS_WQEBBS) -#endif - #define INL_HDR_START_SZ (sizeof(((struct mlx5_wqe_eth_seg *)NULL)->inline_hdr.start)) +enum mlx5e_icosq_wqe_type { + MLX5E_ICOSQ_WQE_NOP, + MLX5E_ICOSQ_WQE_UMR_RX, +}; + static inline bool mlx5e_wqc_has_room_for(struct mlx5_wq_cyc *wq, u16 cc, u16 pc, u16 n) { return (mlx5_wq_cyc_ctr2ix(wq, cc - pc) >= n) || (cc == pc); } -static inline void * -mlx5e_sq_fetch_wqe(struct mlx5e_txqsq *sq, size_t size, u16 *pi) +static inline void *mlx5e_fetch_wqe(struct mlx5_wq_cyc *wq, u16 pi, size_t wqe_size) { - struct mlx5_wq_cyc *wq = &sq->wq; void *wqe; - *pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - wqe = mlx5_wq_cyc_get_wqe(wq, *pi); - memset(wqe, 0, size); + wqe = mlx5_wq_cyc_get_wqe(wq, pi); + memset(wqe, 0, wqe_size); return wqe; } +#define MLX5E_TX_FETCH_WQE(sq, pi) \ + ((struct mlx5e_tx_wqe *)mlx5e_fetch_wqe(&(sq)->wq, pi, sizeof(struct mlx5e_tx_wqe))) + static inline struct mlx5e_tx_wqe * mlx5e_post_nop(struct mlx5_wq_cyc *wq, u32 sqn, u16 *pc) { @@ -81,6 +67,84 @@ mlx5e_post_nop_fence(struct mlx5_wq_cyc *wq, u32 sqn, u16 *pc) return wqe; } +struct mlx5e_tx_wqe_info { + struct sk_buff *skb; + u32 num_bytes; + u8 num_wqebbs; + u8 num_dma; +#ifdef CONFIG_MLX5_EN_TLS + struct page *resync_dump_frag_page; +#endif +}; + +static inline u16 mlx5e_txqsq_get_next_pi(struct mlx5e_txqsq *sq, u16 size) +{ + struct mlx5_wq_cyc *wq = &sq->wq; + u16 pi, contig_wqebbs; + + pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); + contig_wqebbs = mlx5_wq_cyc_get_contig_wqebbs(wq, pi); + if (unlikely(contig_wqebbs < size)) { + struct mlx5e_tx_wqe_info *wi, *edge_wi; + + wi = &sq->db.wqe_info[pi]; + edge_wi = wi + contig_wqebbs; + + /* Fill SQ frag edge with NOPs to avoid WQE wrapping two pages. */ + for (; wi < edge_wi; wi++) { + *wi = (struct mlx5e_tx_wqe_info) { + .num_wqebbs = 1, + }; + mlx5e_post_nop(wq, sq->sqn, &sq->pc); + } + sq->stats->nop += contig_wqebbs; + + pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); + } + + return pi; +} + +struct mlx5e_icosq_wqe_info { + u8 wqe_type; + u8 num_wqebbs; + + /* Auxiliary data for different wqe types. */ + union { + struct { + struct mlx5e_rq *rq; + } umr; + }; +}; + +static inline u16 mlx5e_icosq_get_next_pi(struct mlx5e_icosq *sq, u16 size) +{ + struct mlx5_wq_cyc *wq = &sq->wq; + u16 pi, contig_wqebbs; + + pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); + contig_wqebbs = mlx5_wq_cyc_get_contig_wqebbs(wq, pi); + if (unlikely(contig_wqebbs < size)) { + struct mlx5e_icosq_wqe_info *wi, *edge_wi; + + wi = &sq->db.wqe_info[pi]; + edge_wi = wi + contig_wqebbs; + + /* Fill SQ frag edge with NOPs to avoid WQE wrapping two pages. */ + for (; wi < edge_wi; wi++) { + *wi = (struct mlx5e_icosq_wqe_info) { + .wqe_type = MLX5E_ICOSQ_WQE_NOP, + .num_wqebbs = 1, + }; + mlx5e_post_nop(wq, sq->sqn, &sq->pc); + } + + pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); + } + + return pi; +} + static inline void mlx5e_fill_sq_frag_edge(struct mlx5e_txqsq *sq, struct mlx5_wq_cyc *wq, u16 pi, u16 nnops) @@ -102,7 +166,7 @@ static inline void mlx5e_notify_hw(struct mlx5_wq_cyc *wq, u16 pc, void __iomem *uar_map, struct mlx5_wqe_ctrl_seg *ctrl) { - ctrl->fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE; + ctrl->fm_ce_se |= MLX5_WQE_CTRL_CQ_UPDATE; /* ensure wqe is visible to device before updating doorbell record */ dma_wmb(); @@ -189,6 +253,22 @@ static inline void mlx5e_rqwq_reset(struct mlx5e_rq *rq) } } +static inline void mlx5e_dump_error_cqe(struct mlx5e_cq *cq, u32 sqn, + struct mlx5_err_cqe *err_cqe) +{ + struct mlx5_cqwq *wq = &cq->wq; + u32 ci; + + ci = mlx5_cqwq_ctr2ix(wq, wq->cc - 1); + + netdev_err(cq->channel->netdev, + "Error cqe on cqn 0x%x, ci 0x%x, sqn 0x%x, opcode 0x%x, syndrome 0x%x, vendor syndrome 0x%x\n", + cq->mcq.cqn, ci, sqn, + get_cqe_opcode((struct mlx5_cqe64 *)err_cqe), + err_cqe->syndrome, err_cqe->vendor_err_synd); + mlx5_dump_err_cqe(cq->mdev, err_cqe); +} + /* SW parser related functions */ struct mlx5e_swp_spec { @@ -232,4 +312,25 @@ mlx5e_set_eseg_swp(struct sk_buff *skb, struct mlx5_wqe_eth_seg *eseg, } } +static inline u16 mlx5e_stop_room_for_wqe(u16 wqe_size) +{ + BUILD_BUG_ON(PAGE_SIZE / MLX5_SEND_WQE_BB < MLX5_SEND_WQE_MAX_WQEBBS); + + /* A WQE must not cross the page boundary, hence two conditions: + * 1. Its size must not exceed the page size. + * 2. If the WQE size is X, and the space remaining in a page is less + * than X, this space needs to be padded with NOPs. So, one WQE of + * size X may require up to X-1 WQEBBs of padding, which makes the + * stop room of X-1 + X. + * WQE size is also limited by the hardware limit. + */ + + if (__builtin_constant_p(wqe_size)) + BUILD_BUG_ON(wqe_size > MLX5_SEND_WQE_MAX_WQEBBS); + else + WARN_ON_ONCE(wqe_size > MLX5_SEND_WQE_MAX_WQEBBS); + + return wqe_size * 2 - 1; +} + #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c index f049e0ac308a..c9d308e91965 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c @@ -31,7 +31,7 @@ */ #include <linux/bpf_trace.h> -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> #include "en/xdp.h" #include "en/params.h" @@ -64,14 +64,14 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq, struct xdp_frame *xdpf; dma_addr_t dma_addr; - xdpf = convert_to_xdp_frame(xdp); + xdpf = xdp_convert_buff_to_frame(xdp); if (unlikely(!xdpf)) return false; xdptxd.data = xdpf->data; xdptxd.len = xdpf->len; - if (xdp->rxq->mem.type == MEM_TYPE_ZERO_COPY) { + if (xdp->rxq->mem.type == MEM_TYPE_XSK_BUFF_POOL) { /* The xdp_buff was in the UMEM and was copied into a newly * allocated page. The UMEM page was returned via the ZCA, and * this new page has to be mapped at this point and has to be @@ -97,10 +97,10 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq, xdpi.frame.xdpf = xdpf; xdpi.frame.dma_addr = dma_addr; } else { - /* Driver assumes that convert_to_xdp_frame returns an xdp_frame - * that points to the same memory region as the original - * xdp_buff. It allows to map the memory only once and to use - * the DMA_BIDIRECTIONAL mode. + /* Driver assumes that xdp_convert_buff_to_frame returns + * an xdp_frame that points to the same memory region as + * the original xdp_buff. It allows to map the memory only + * once and to use the DMA_BIDIRECTIONAL mode. */ xdpi.mode = MLX5E_XDP_XMIT_MODE_PAGE; @@ -119,49 +119,33 @@ mlx5e_xmit_xdp_buff(struct mlx5e_xdpsq *sq, struct mlx5e_rq *rq, /* returns true if packet was consumed by xdp */ bool mlx5e_xdp_handle(struct mlx5e_rq *rq, struct mlx5e_dma_info *di, - void *va, u16 *rx_headroom, u32 *len, bool xsk) + u32 *len, struct xdp_buff *xdp) { struct bpf_prog *prog = READ_ONCE(rq->xdp_prog); - struct xdp_umem *umem = rq->umem; - struct xdp_buff xdp; u32 act; int err; if (!prog) return false; - xdp.data = va + *rx_headroom; - xdp_set_data_meta_invalid(&xdp); - xdp.data_end = xdp.data + *len; - xdp.data_hard_start = va; - if (xsk) - xdp.handle = di->xsk.handle; - xdp.rxq = &rq->xdp_rxq; - - act = bpf_prog_run_xdp(prog, &xdp); - if (xsk) { - u64 off = xdp.data - xdp.data_hard_start; - - xdp.handle = xsk_umem_adjust_offset(umem, xdp.handle, off); - } + act = bpf_prog_run_xdp(prog, xdp); switch (act) { case XDP_PASS: - *rx_headroom = xdp.data - xdp.data_hard_start; - *len = xdp.data_end - xdp.data; + *len = xdp->data_end - xdp->data; return false; case XDP_TX: - if (unlikely(!mlx5e_xmit_xdp_buff(rq->xdpsq, rq, di, &xdp))) + if (unlikely(!mlx5e_xmit_xdp_buff(rq->xdpsq, rq, di, xdp))) goto xdp_abort; __set_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags); /* non-atomic */ return true; case XDP_REDIRECT: /* When XDP enabled then page-refcnt==1 here */ - err = xdp_do_redirect(rq->netdev, &xdp, prog); + err = xdp_do_redirect(rq->netdev, xdp, prog); if (unlikely(err)) goto xdp_abort; __set_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags); __set_bit(MLX5E_RQ_FLAG_XDP_REDIRECT, rq->flags); - if (!xsk) + if (xdp->rxq->mem.type != MEM_TYPE_XSK_BUFF_POOL) mlx5e_page_dma_unmap(rq, di); rq->stats->xdp_redirect++; return true; @@ -178,20 +162,43 @@ xdp_abort: } } -static void mlx5e_xdp_mpwqe_session_start(struct mlx5e_xdpsq *sq) +static u16 mlx5e_xdpsq_get_next_pi(struct mlx5e_xdpsq *sq, u16 size) { - struct mlx5e_xdp_mpwqe *session = &sq->mpwqe; - struct mlx5e_xdpsq_stats *stats = sq->stats; struct mlx5_wq_cyc *wq = &sq->wq; u16 pi, contig_wqebbs; pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); contig_wqebbs = mlx5_wq_cyc_get_contig_wqebbs(wq, pi); + if (unlikely(contig_wqebbs < size)) { + struct mlx5e_xdp_wqe_info *wi, *edge_wi; + + wi = &sq->db.wqe_info[pi]; + edge_wi = wi + contig_wqebbs; + + /* Fill SQ frag edge with NOPs to avoid WQE wrapping two pages. */ + for (; wi < edge_wi; wi++) { + *wi = (struct mlx5e_xdp_wqe_info) { + .num_wqebbs = 1, + .num_pkts = 0, + }; + mlx5e_post_nop(wq, sq->sqn, &sq->pc); + } + sq->stats->nops += contig_wqebbs; - if (unlikely(contig_wqebbs < MLX5_SEND_WQE_MAX_WQEBBS)) - mlx5e_fill_xdpsq_frag_edge(sq, wq, pi, contig_wqebbs); + pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); + } + + return pi; +} - session->wqe = mlx5e_xdpsq_fetch_wqe(sq, &pi); +static void mlx5e_xdp_mpwqe_session_start(struct mlx5e_xdpsq *sq) +{ + struct mlx5e_xdp_mpwqe *session = &sq->mpwqe; + struct mlx5e_xdpsq_stats *stats = sq->stats; + u16 pi; + + pi = mlx5e_xdpsq_get_next_pi(sq, MLX5_SEND_WQE_MAX_WQEBBS); + session->wqe = MLX5E_TX_FETCH_WQE(sq, pi); prefetchw(session->wqe->data); session->ds_count = MLX5E_XDP_TX_EMPTY_DS_COUNT; @@ -233,8 +240,10 @@ enum { static int mlx5e_xmit_xdp_frame_check_mpwqe(struct mlx5e_xdpsq *sq) { if (unlikely(!sq->mpwqe.wqe)) { + const u16 stop_room = mlx5e_stop_room_for_wqe(MLX5_SEND_WQE_MAX_WQEBBS); + if (unlikely(!mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, - MLX5E_XDPSQ_STOP_ROOM))) { + stop_room))) { /* SQ is full, ring doorbell */ mlx5e_xmit_xdp_doorbell(sq); sq->stats->full++; @@ -408,22 +417,15 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq) i = 0; do { - u16 wqe_counter; + struct mlx5e_xdp_wqe_info *wi; + u16 wqe_counter, ci; bool last_wqe; mlx5_cqwq_pop(&cq->wq); wqe_counter = be16_to_cpu(cqe->wqe_counter); - if (unlikely(get_cqe_opcode(cqe) != MLX5_CQE_REQ)) - netdev_WARN_ONCE(sq->channel->netdev, - "Bad OP in XDPSQ CQE: 0x%x\n", - get_cqe_opcode(cqe)); - do { - struct mlx5e_xdp_wqe_info *wi; - u16 ci; - last_wqe = (sqcc == wqe_counter); ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sqcc); wi = &sq->db.wqe_info[ci]; @@ -432,6 +434,15 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq) mlx5e_free_xdpsq_desc(sq, wi, &xsk_frames, true); } while (!last_wqe); + + if (unlikely(get_cqe_opcode(cqe) != MLX5_CQE_REQ)) { + netdev_WARN_ONCE(sq->channel->netdev, + "Bad OP in XDPSQ CQE: 0x%x\n", + get_cqe_opcode(cqe)); + mlx5e_dump_error_cqe(&sq->cq, sq->sqn, + (struct mlx5_err_cqe *)cqe); + mlx5_wq_cyc_wqe_dump(&sq->wq, ci, wi->num_wqebbs); + } } while ((++i < MLX5E_TX_CQ_POLL_BUDGET) && (cqe = mlx5_cqwq_get_cqe(&cq->wq))); if (xsk_frames) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h index d7587f40ecae..ca48c293151b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.h @@ -40,8 +40,6 @@ (sizeof(struct mlx5e_tx_wqe) / MLX5_SEND_WQE_DS) #define MLX5E_XDP_TX_DS_COUNT (MLX5E_XDP_TX_EMPTY_DS_COUNT + 1 /* SG DS */) -#define MLX5E_XDPSQ_STOP_ROOM (MLX5E_SQ_STOP_ROOM) - #define MLX5E_XDP_INLINE_WQE_SZ_THRSD (256 - sizeof(struct mlx5_wqe_inline_seg)) #define MLX5E_XDP_INLINE_WQE_MAX_DS_CNT \ DIV_ROUND_UP(MLX5E_XDP_INLINE_WQE_SZ_THRSD, MLX5_SEND_WQE_DS) @@ -63,7 +61,7 @@ struct mlx5e_xsk_param; int mlx5e_xdp_max_mtu(struct mlx5e_params *params, struct mlx5e_xsk_param *xsk); bool mlx5e_xdp_handle(struct mlx5e_rq *rq, struct mlx5e_dma_info *di, - void *va, u16 *rx_headroom, u32 *len, bool xsk); + u32 *len, struct xdp_buff *xdp); void mlx5e_xdp_mpwqe_complete(struct mlx5e_xdpsq *sq); bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq); void mlx5e_free_xdpsq_descs(struct mlx5e_xdpsq *sq); @@ -137,22 +135,10 @@ mlx5e_xdp_no_room_for_inline_pkt(struct mlx5e_xdp_mpwqe *session) session->ds_count + MLX5E_XDP_INLINE_WQE_MAX_DS_CNT > MLX5E_XDP_MPW_MAX_NUM_DS; } -static inline void -mlx5e_fill_xdpsq_frag_edge(struct mlx5e_xdpsq *sq, struct mlx5_wq_cyc *wq, - u16 pi, u16 nnops) -{ - struct mlx5e_xdp_wqe_info *edge_wi, *wi = &sq->db.wqe_info[pi]; - - edge_wi = wi + nnops; - /* fill sq frag edge with nops to avoid wqe wrapping two pages */ - for (; wi < edge_wi; wi++) { - wi->num_wqebbs = 1; - wi->num_pkts = 0; - mlx5e_post_nop(wq, sq->sqn, &sq->pc); - } - - sq->stats->nops += nnops; -} +struct mlx5e_xdp_wqe_info { + u8 num_wqebbs; + u8 num_pkts; +}; static inline void mlx5e_xdp_mpwqe_add_dseg(struct mlx5e_xdpsq *sq, @@ -186,19 +172,6 @@ mlx5e_xdp_mpwqe_add_dseg(struct mlx5e_xdpsq *sq, session->ds_count++; } -static inline struct mlx5e_tx_wqe * -mlx5e_xdpsq_fetch_wqe(struct mlx5e_xdpsq *sq, u16 *pi) -{ - struct mlx5_wq_cyc *wq = &sq->wq; - struct mlx5e_tx_wqe *wqe; - - *pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - wqe = mlx5_wq_cyc_get_wqe(wq, *pi); - memset(wqe, 0, sizeof(*wqe)); - - return wqe; -} - static inline void mlx5e_xdpi_fifo_push(struct mlx5e_xdp_info_fifo *fifo, struct mlx5e_xdp_info *xi) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c index 62fc8a128a8d..a33a1f762c70 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.c @@ -3,71 +3,10 @@ #include "rx.h" #include "en/xdp.h" -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> /* RX data path */ -bool mlx5e_xsk_pages_enough_umem(struct mlx5e_rq *rq, int count) -{ - /* Check in advance that we have enough frames, instead of allocating - * one-by-one, failing and moving frames to the Reuse Ring. - */ - return xsk_umem_has_addrs_rq(rq->umem, count); -} - -int mlx5e_xsk_page_alloc_umem(struct mlx5e_rq *rq, - struct mlx5e_dma_info *dma_info) -{ - struct xdp_umem *umem = rq->umem; - u64 handle; - - if (!xsk_umem_peek_addr_rq(umem, &handle)) - return -ENOMEM; - - dma_info->xsk.handle = xsk_umem_adjust_offset(umem, handle, - rq->buff.umem_headroom); - dma_info->xsk.data = xdp_umem_get_data(umem, dma_info->xsk.handle); - - /* No need to add headroom to the DMA address. In striding RQ case, we - * just provide pages for UMR, and headroom is counted at the setup - * stage when creating a WQE. In non-striding RQ case, headroom is - * accounted in mlx5e_alloc_rx_wqe. - */ - dma_info->addr = xdp_umem_get_dma(umem, handle); - - xsk_umem_release_addr_rq(umem); - - dma_sync_single_for_device(rq->pdev, dma_info->addr, PAGE_SIZE, - DMA_BIDIRECTIONAL); - - return 0; -} - -static inline void mlx5e_xsk_recycle_frame(struct mlx5e_rq *rq, u64 handle) -{ - xsk_umem_fq_reuse(rq->umem, handle & rq->umem->chunk_mask); -} - -/* XSKRQ uses pages from UMEM, they must not be released. They are returned to - * the userspace if possible, and if not, this function is called to reuse them - * in the driver. - */ -void mlx5e_xsk_page_release(struct mlx5e_rq *rq, - struct mlx5e_dma_info *dma_info) -{ - mlx5e_xsk_recycle_frame(rq, dma_info->xsk.handle); -} - -/* Return a frame back to the hardware to fill in again. It is used by XDP when - * the XDP program returns XDP_TX or XDP_REDIRECT not to an XSKMAP. - */ -void mlx5e_xsk_zca_free(struct zero_copy_allocator *zca, unsigned long handle) -{ - struct mlx5e_rq *rq = container_of(zca, struct mlx5e_rq, zca); - - mlx5e_xsk_recycle_frame(rq, handle); -} - static struct sk_buff *mlx5e_xsk_construct_skb(struct mlx5e_rq *rq, void *data, u32 cqe_bcnt) { @@ -90,11 +29,8 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, u32 head_offset, u32 page_idx) { - struct mlx5e_dma_info *di = &wi->umr.dma_info[page_idx]; - u16 rx_headroom = rq->buff.headroom - rq->buff.umem_headroom; + struct xdp_buff *xdp = wi->umr.dma_info[page_idx].xsk; u32 cqe_bcnt32 = cqe_bcnt; - void *va, *data; - u32 frag_size; bool consumed; /* Check packet size. Note LRO doesn't use linear SKB */ @@ -103,22 +39,20 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, return NULL; } - /* head_offset is not used in this function, because di->xsk.data and - * di->addr point directly to the necessary place. Furthermore, in the - * current implementation, UMR pages are mapped to XSK frames, so + /* head_offset is not used in this function, because xdp->data and the + * DMA address point directly to the necessary place. Furthermore, in + * the current implementation, UMR pages are mapped to XSK frames, so * head_offset should always be 0. */ WARN_ON_ONCE(head_offset); - va = di->xsk.data; - data = va + rx_headroom; - frag_size = rq->buff.headroom + cqe_bcnt32; - - dma_sync_single_for_cpu(rq->pdev, di->addr, frag_size, DMA_BIDIRECTIONAL); - prefetch(data); + xdp->data_end = xdp->data + cqe_bcnt32; + xdp_set_data_meta_invalid(xdp); + xsk_buff_dma_sync_for_cpu(xdp); + prefetch(xdp->data); rcu_read_lock(); - consumed = mlx5e_xdp_handle(rq, di, va, &rx_headroom, &cqe_bcnt32, true); + consumed = mlx5e_xdp_handle(rq, NULL, &cqe_bcnt32, xdp); rcu_read_unlock(); /* Possible flows: @@ -145,7 +79,7 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, /* XDP_PASS: copy the data from the UMEM to a new SKB and reuse the * frame. On SKB allocation failure, NULL is returned. */ - return mlx5e_xsk_construct_skb(rq, data, cqe_bcnt32); + return mlx5e_xsk_construct_skb(rq, xdp->data, cqe_bcnt32); } struct sk_buff *mlx5e_xsk_skb_from_cqe_linear(struct mlx5e_rq *rq, @@ -153,25 +87,20 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi, u32 cqe_bcnt) { - struct mlx5e_dma_info *di = wi->di; - u16 rx_headroom = rq->buff.headroom - rq->buff.umem_headroom; - void *va, *data; + struct xdp_buff *xdp = wi->di->xsk; bool consumed; - u32 frag_size; - /* wi->offset is not used in this function, because di->xsk.data and - * di->addr point directly to the necessary place. Furthermore, in the - * current implementation, one page = one packet = one frame, so + /* wi->offset is not used in this function, because xdp->data and the + * DMA address point directly to the necessary place. Furthermore, the + * XSK allocator allocates frames per packet, instead of pages, so * wi->offset should always be 0. */ WARN_ON_ONCE(wi->offset); - va = di->xsk.data; - data = va + rx_headroom; - frag_size = rq->buff.headroom + cqe_bcnt; - - dma_sync_single_for_cpu(rq->pdev, di->addr, frag_size, DMA_BIDIRECTIONAL); - prefetch(data); + xdp->data_end = xdp->data + cqe_bcnt; + xdp_set_data_meta_invalid(xdp); + xsk_buff_dma_sync_for_cpu(xdp); + prefetch(xdp->data); if (unlikely(get_cqe_opcode(cqe) != MLX5_CQE_RESP_SEND)) { rq->stats->wqe_err++; @@ -179,7 +108,7 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_linear(struct mlx5e_rq *rq, } rcu_read_lock(); - consumed = mlx5e_xdp_handle(rq, di, va, &rx_headroom, &cqe_bcnt, true); + consumed = mlx5e_xdp_handle(rq, NULL, &cqe_bcnt, xdp); rcu_read_unlock(); if (likely(consumed)) @@ -189,5 +118,5 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_linear(struct mlx5e_rq *rq, * will be handled by mlx5e_put_rx_frag. * On SKB allocation failure, NULL is returned. */ - return mlx5e_xsk_construct_skb(rq, data, cqe_bcnt); + return mlx5e_xsk_construct_skb(rq, xdp->data, cqe_bcnt); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.h index cab0e93497ae..d147b2f13b54 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/rx.h @@ -5,16 +5,10 @@ #define __MLX5_EN_XSK_RX_H__ #include "en.h" -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> /* RX data path */ -bool mlx5e_xsk_pages_enough_umem(struct mlx5e_rq *rq, int count); -int mlx5e_xsk_page_alloc_umem(struct mlx5e_rq *rq, - struct mlx5e_dma_info *dma_info); -void mlx5e_xsk_page_release(struct mlx5e_rq *rq, - struct mlx5e_dma_info *dma_info); -void mlx5e_xsk_zca_free(struct zero_copy_allocator *zca, unsigned long handle); struct sk_buff *mlx5e_xsk_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, u16 cqe_bcnt, @@ -25,6 +19,23 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5e_wqe_frag_info *wi, u32 cqe_bcnt); +static inline int mlx5e_xsk_page_alloc_umem(struct mlx5e_rq *rq, + struct mlx5e_dma_info *dma_info) +{ + dma_info->xsk = xsk_buff_alloc(rq->umem); + if (!dma_info->xsk) + return -ENOMEM; + + /* Store the DMA address without headroom. In striding RQ case, we just + * provide pages for UMR, and headroom is counted at the setup stage + * when creating a WQE. In non-striding RQ case, headroom is accounted + * in mlx5e_alloc_rx_wqe. + */ + dma_info->addr = xsk_buff_xdp_get_frame_dma(dma_info->xsk); + + return 0; +} + static inline bool mlx5e_xsk_update_rx_wakeup(struct mlx5e_rq *rq, bool alloc_err) { if (!xsk_umem_uses_need_wakeup(rq->umem)) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c index 3bcdb5b2fc20..83dce9cdb8c2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.c @@ -5,7 +5,7 @@ #include "umem.h" #include "en/xdp.h" #include "en/params.h" -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> int mlx5e_xsk_wakeup(struct net_device *dev, u32 qid, u32 flags) { @@ -92,12 +92,11 @@ bool mlx5e_xsk_tx(struct mlx5e_xdpsq *sq, unsigned int budget) break; } - xdptxd.dma_addr = xdp_umem_get_dma(umem, desc.addr); - xdptxd.data = xdp_umem_get_data(umem, desc.addr); + xdptxd.dma_addr = xsk_buff_raw_get_dma(umem, desc.addr); + xdptxd.data = xsk_buff_raw_get_data(umem, desc.addr); xdptxd.len = desc.len; - dma_sync_single_for_device(sq->pdev, xdptxd.dma_addr, - xdptxd.len, DMA_BIDIRECTIONAL); + xsk_buff_raw_dma_sync_for_device(umem, xdptxd.dma_addr, xdptxd.len); if (unlikely(!sq->xmit_xdp_frame(sq, &xdptxd, &xdpi, check_result))) { if (sq->mpwqe.wqe) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.h index 79b487d89757..39fa0a705856 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/tx.h @@ -5,7 +5,7 @@ #define __MLX5_EN_XSK_TX_H__ #include "en.h" -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> /* TX data path */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/umem.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/umem.c index 4baaa5788320..7b17fcd0a56d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/umem.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xsk/umem.c @@ -1,7 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB /* Copyright (c) 2019 Mellanox Technologies. */ -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> #include "umem.h" #include "setup.h" #include "en/params.h" @@ -10,40 +10,14 @@ static int mlx5e_xsk_map_umem(struct mlx5e_priv *priv, struct xdp_umem *umem) { struct device *dev = priv->mdev->device; - u32 i; - for (i = 0; i < umem->npgs; i++) { - dma_addr_t dma = dma_map_page(dev, umem->pgs[i], 0, PAGE_SIZE, - DMA_BIDIRECTIONAL); - - if (unlikely(dma_mapping_error(dev, dma))) - goto err_unmap; - umem->pages[i].dma = dma; - } - - return 0; - -err_unmap: - while (i--) { - dma_unmap_page(dev, umem->pages[i].dma, PAGE_SIZE, - DMA_BIDIRECTIONAL); - umem->pages[i].dma = 0; - } - - return -ENOMEM; + return xsk_buff_dma_map(umem, dev, 0); } static void mlx5e_xsk_unmap_umem(struct mlx5e_priv *priv, struct xdp_umem *umem) { - struct device *dev = priv->mdev->device; - u32 i; - - for (i = 0; i < umem->npgs; i++) { - dma_unmap_page(dev, umem->pages[i].dma, PAGE_SIZE, - DMA_BIDIRECTIONAL); - umem->pages[i].dma = 0; - } + return xsk_buff_dma_unmap(umem, 0); } static int mlx5e_xsk_get_umems(struct mlx5e_xsk *xsk) @@ -90,13 +64,14 @@ static void mlx5e_xsk_remove_umem(struct mlx5e_xsk *xsk, u16 ix) static bool mlx5e_xsk_is_umem_sane(struct xdp_umem *umem) { - return umem->headroom <= 0xffff && umem->chunk_size_nohr <= 0xffff; + return xsk_umem_get_headroom(umem) <= 0xffff && + xsk_umem_get_chunk_size(umem) <= 0xffff; } void mlx5e_build_xsk_param(struct xdp_umem *umem, struct mlx5e_xsk_param *xsk) { - xsk->headroom = umem->headroom; - xsk->chunk_size = umem->chunk_size_nohr + umem->headroom; + xsk->headroom = xsk_umem_get_headroom(umem); + xsk->chunk_size = xsk_umem_get_chunk_size(umem); } static int mlx5e_xsk_enable_locked(struct mlx5e_priv *priv, @@ -241,18 +216,6 @@ int mlx5e_xsk_setup_umem(struct net_device *dev, struct xdp_umem *umem, u16 qid) mlx5e_xsk_disable_umem(priv, ix); } -int mlx5e_xsk_resize_reuseq(struct xdp_umem *umem, u32 nentries) -{ - struct xdp_umem_fq_reuse *reuseq; - - reuseq = xsk_reuseq_prepare(nentries); - if (unlikely(!reuseq)) - return -ENOMEM; - xsk_reuseq_free(xsk_reuseq_swap(umem, reuseq)); - - return 0; -} - u16 mlx5e_xsk_first_unused_channel(struct mlx5e_params *params, struct mlx5e_xsk *xsk) { u16 res = xsk->refcnt ? params->num_channels : 0; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h index 3022463f2284..fac145dcf2ce 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h @@ -42,6 +42,8 @@ #include "en/txrx.h" #if IS_ENABLED(CONFIG_GENEVE) +#include <net/geneve.h> + static inline bool mlx5_geneve_tx_allowed(struct mlx5_core_dev *mdev) { return mlx5_tx_swp_supported(mdev); @@ -100,33 +102,49 @@ mlx5e_udp_gso_handle_tx_skb(struct sk_buff *skb) udp_hdr(skb)->len = htons(payload_len); } -static inline struct sk_buff * -mlx5e_accel_handle_tx(struct sk_buff *skb, - struct mlx5e_txqsq *sq, - struct net_device *dev, - struct mlx5e_tx_wqe **wqe, - u16 *pi) +struct mlx5e_accel_tx_state { +#ifdef CONFIG_MLX5_EN_TLS + struct mlx5e_accel_tx_tls_state tls; +#endif +}; + +static inline bool mlx5e_accel_tx_begin(struct net_device *dev, + struct mlx5e_txqsq *sq, + struct sk_buff *skb, + struct mlx5e_accel_tx_state *state) { + if (skb_is_gso(skb) && skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) + mlx5e_udp_gso_handle_tx_skb(skb); + #ifdef CONFIG_MLX5_EN_TLS if (test_bit(MLX5E_SQ_STATE_TLS, &sq->state)) { - skb = mlx5e_tls_handle_tx_skb(dev, sq, skb, wqe, pi); - if (unlikely(!skb)) - return NULL; + /* May send SKBs and WQEs. */ + if (unlikely(!mlx5e_tls_handle_tx_skb(dev, sq, skb, &state->tls))) + return false; } #endif + return true; +} + +static inline bool mlx5e_accel_tx_finish(struct mlx5e_priv *priv, + struct mlx5e_txqsq *sq, + struct sk_buff *skb, + struct mlx5e_tx_wqe *wqe, + struct mlx5e_accel_tx_state *state) +{ +#ifdef CONFIG_MLX5_EN_TLS + mlx5e_tls_handle_tx_wqe(sq, &wqe->ctrl, &state->tls); +#endif + #ifdef CONFIG_MLX5_EN_IPSEC if (test_bit(MLX5E_SQ_STATE_IPSEC, &sq->state)) { - skb = mlx5e_ipsec_handle_tx_skb(dev, *wqe, skb); - if (unlikely(!skb)) - return NULL; + if (unlikely(!mlx5e_ipsec_handle_tx_skb(priv, &wqe->eth, skb))) + return false; } #endif - if (skb_is_gso(skb) && skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) - mlx5e_udp_gso_handle_tx_skb(skb); - - return skb; + return true; } #endif /* __MLX5E_EN_ACCEL_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c index 29626c6c9c25..92eb3bad4acd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c @@ -75,18 +75,23 @@ struct xfrm_state *mlx5e_ipsec_sadb_rx_lookup(struct mlx5e_ipsec *ipsec, return ret; } -static int mlx5e_ipsec_sadb_rx_add(struct mlx5e_ipsec_sa_entry *sa_entry) +static int mlx5e_ipsec_sadb_rx_add(struct mlx5e_ipsec_sa_entry *sa_entry, + unsigned int handle) { struct mlx5e_ipsec *ipsec = sa_entry->ipsec; + struct mlx5e_ipsec_sa_entry *_sa_entry; unsigned long flags; - int ret; - ret = ida_simple_get(&ipsec->halloc, 1, 0, GFP_KERNEL); - if (ret < 0) - return ret; + rcu_read_lock(); + hash_for_each_possible_rcu(ipsec->sadb_rx, _sa_entry, hlist, handle) + if (_sa_entry->handle == handle) { + rcu_read_unlock(); + return -EEXIST; + } + rcu_read_unlock(); spin_lock_irqsave(&ipsec->sadb_rx_lock, flags); - sa_entry->handle = ret; + sa_entry->handle = handle; hash_add_rcu(ipsec->sadb_rx, &sa_entry->hlist, sa_entry->handle); spin_unlock_irqrestore(&ipsec->sadb_rx_lock, flags); @@ -103,15 +108,6 @@ static void mlx5e_ipsec_sadb_rx_del(struct mlx5e_ipsec_sa_entry *sa_entry) spin_unlock_irqrestore(&ipsec->sadb_rx_lock, flags); } -static void mlx5e_ipsec_sadb_rx_free(struct mlx5e_ipsec_sa_entry *sa_entry) -{ - struct mlx5e_ipsec *ipsec = sa_entry->ipsec; - - /* xfrm already doing sync rcu between del and free callbacks */ - - ida_simple_remove(&ipsec->halloc, sa_entry->handle); -} - static bool mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry) { struct xfrm_replay_state_esn *replay_esn; @@ -199,6 +195,14 @@ mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry, attrs->flags |= (x->props.mode == XFRM_MODE_TRANSPORT) ? MLX5_ACCEL_ESP_FLAGS_TRANSPORT : MLX5_ACCEL_ESP_FLAGS_TUNNEL; + + /* spi */ + attrs->spi = x->id.spi; + + /* source , destination ips */ + memcpy(&attrs->saddr, x->props.saddr.a6, sizeof(attrs->saddr)); + memcpy(&attrs->daddr, x->id.daddr.a6, sizeof(attrs->daddr)); + attrs->is_ipv6 = (x->props.family != AF_INET); } static inline int mlx5e_xfrm_validate_state(struct xfrm_state *x) @@ -284,8 +288,7 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x) struct net_device *netdev = x->xso.dev; struct mlx5_accel_esp_xfrm_attrs attrs; struct mlx5e_priv *priv; - __be32 saddr[4] = {0}, daddr[4] = {0}, spi; - bool is_ipv6 = false; + unsigned int sa_handle; int err; priv = netdev_priv(netdev); @@ -303,20 +306,6 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x) sa_entry->x = x; sa_entry->ipsec = priv->ipsec; - /* Add the SA to handle processed incoming packets before the add SA - * completion was received - */ - if (x->xso.flags & XFRM_OFFLOAD_INBOUND) { - err = mlx5e_ipsec_sadb_rx_add(sa_entry); - if (err) { - netdev_info(netdev, "Failed adding to SADB_RX: %d\n", err); - goto err_entry; - } - } else { - sa_entry->set_iv_op = (x->props.flags & XFRM_STATE_ESN) ? - mlx5e_ipsec_set_iv_esn : mlx5e_ipsec_set_iv; - } - /* check esn */ mlx5e_ipsec_update_esn_state(sa_entry); @@ -327,41 +316,38 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x) MLX5_ACCEL_XFRM_FLAG_REQUIRE_METADATA); if (IS_ERR(sa_entry->xfrm)) { err = PTR_ERR(sa_entry->xfrm); - goto err_sadb_rx; + goto err_sa_entry; } /* create hw context */ - if (x->props.family == AF_INET) { - saddr[3] = x->props.saddr.a4; - daddr[3] = x->id.daddr.a4; - } else { - memcpy(saddr, x->props.saddr.a6, sizeof(saddr)); - memcpy(daddr, x->id.daddr.a6, sizeof(daddr)); - is_ipv6 = true; - } - spi = x->id.spi; sa_entry->hw_context = mlx5_accel_esp_create_hw_context(priv->mdev, sa_entry->xfrm, - saddr, daddr, spi, - is_ipv6); + &sa_handle); if (IS_ERR(sa_entry->hw_context)) { err = PTR_ERR(sa_entry->hw_context); goto err_xfrm; } + if (x->xso.flags & XFRM_OFFLOAD_INBOUND) { + err = mlx5e_ipsec_sadb_rx_add(sa_entry, sa_handle); + if (err) + goto err_hw_ctx; + } else { + sa_entry->set_iv_op = (x->props.flags & XFRM_STATE_ESN) ? + mlx5e_ipsec_set_iv_esn : mlx5e_ipsec_set_iv; + } + x->xso.offload_handle = (unsigned long)sa_entry; goto out; +err_hw_ctx: + mlx5_accel_esp_free_hw_context(sa_entry->hw_context); err_xfrm: mlx5_accel_esp_destroy_xfrm(sa_entry->xfrm); -err_sadb_rx: - if (x->xso.flags & XFRM_OFFLOAD_INBOUND) { - mlx5e_ipsec_sadb_rx_del(sa_entry); - mlx5e_ipsec_sadb_rx_free(sa_entry); - } -err_entry: +err_sa_entry: kfree(sa_entry); + out: return err; } @@ -390,9 +376,6 @@ static void mlx5e_xfrm_free_state(struct xfrm_state *x) mlx5_accel_esp_destroy_xfrm(sa_entry->xfrm); } - if (x->xso.flags & XFRM_OFFLOAD_INBOUND) - mlx5e_ipsec_sadb_rx_free(sa_entry); - kfree(sa_entry); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h index 93bf10e6508c..c85151a1e008 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h @@ -109,11 +109,6 @@ int mlx5e_ipsec_init(struct mlx5e_priv *priv); void mlx5e_ipsec_cleanup(struct mlx5e_priv *priv); void mlx5e_ipsec_build_netdev(struct mlx5e_priv *priv); -int mlx5e_ipsec_get_count(struct mlx5e_priv *priv); -int mlx5e_ipsec_get_strings(struct mlx5e_priv *priv, uint8_t *data); -void mlx5e_ipsec_update_stats(struct mlx5e_priv *priv); -int mlx5e_ipsec_get_stats(struct mlx5e_priv *priv, u64 *data); - struct xfrm_state *mlx5e_ipsec_sadb_rx_lookup(struct mlx5e_ipsec *dev, unsigned int handle); @@ -136,26 +131,6 @@ static inline void mlx5e_ipsec_build_netdev(struct mlx5e_priv *priv) { } -static inline int mlx5e_ipsec_get_count(struct mlx5e_priv *priv) -{ - return 0; -} - -static inline int mlx5e_ipsec_get_strings(struct mlx5e_priv *priv, - uint8_t *data) -{ - return 0; -} - -static inline void mlx5e_ipsec_update_stats(struct mlx5e_priv *priv) -{ -} - -static inline int mlx5e_ipsec_get_stats(struct mlx5e_priv *priv, u64 *data) -{ - return 0; -} - #endif #endif /* __MLX5E_IPSEC_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c index 0dd17514caae..824b87ac8f9e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c @@ -233,11 +233,10 @@ static void mlx5e_ipsec_set_metadata(struct sk_buff *skb, ntohs(mdata->content.tx.seq)); } -struct sk_buff *mlx5e_ipsec_handle_tx_skb(struct net_device *netdev, - struct mlx5e_tx_wqe *wqe, - struct sk_buff *skb) +bool mlx5e_ipsec_handle_tx_skb(struct mlx5e_priv *priv, + struct mlx5_wqe_eth_seg *eseg, + struct sk_buff *skb) { - struct mlx5e_priv *priv = netdev_priv(netdev); struct xfrm_offload *xo = xfrm_offload(skb); struct mlx5e_ipsec_metadata *mdata; struct mlx5e_ipsec_sa_entry *sa_entry; @@ -245,7 +244,7 @@ struct sk_buff *mlx5e_ipsec_handle_tx_skb(struct net_device *netdev, struct sec_path *sp; if (!xo) - return skb; + return true; sp = skb_sec_path(skb); if (unlikely(sp->len != 1)) { @@ -276,16 +275,16 @@ struct sk_buff *mlx5e_ipsec_handle_tx_skb(struct net_device *netdev, atomic64_inc(&priv->ipsec->sw_stats.ipsec_tx_drop_metadata); goto drop; } - mlx5e_ipsec_set_swp(skb, &wqe->eth, x->props.mode, xo); + mlx5e_ipsec_set_swp(skb, eseg, x->props.mode, xo); sa_entry = (struct mlx5e_ipsec_sa_entry *)x->xso.offload_handle; sa_entry->set_iv_op(skb, x, xo); mlx5e_ipsec_set_metadata(skb, mdata, xo); - return skb; + return true; drop: kfree_skb(skb); - return NULL; + return false; } static inline struct xfrm_state * diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h index db84500b024f..ba02643586a5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h @@ -52,9 +52,9 @@ void mlx5e_ipsec_set_iv_esn(struct sk_buff *skb, struct xfrm_state *x, struct xfrm_offload *xo); void mlx5e_ipsec_set_iv(struct sk_buff *skb, struct xfrm_state *x, struct xfrm_offload *xo); -struct sk_buff *mlx5e_ipsec_handle_tx_skb(struct net_device *netdev, - struct mlx5e_tx_wqe *wqe, - struct sk_buff *skb); +bool mlx5e_ipsec_handle_tx_skb(struct mlx5e_priv *priv, + struct mlx5_wqe_eth_seg *eseg, + struct sk_buff *skb); #endif /* CONFIG_MLX5_EN_IPSEC */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_stats.c index 6fea59223dc4..6c5c54bcd9be 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_stats.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_stats.c @@ -38,6 +38,7 @@ #include "accel/ipsec.h" #include "fpga/sdk.h" #include "en_accel/ipsec.h" +#include "fpga/ipsec.h" static const struct counter_desc mlx5e_ipsec_hw_stats_desc[] = { { MLX5E_DECLARE_STAT(struct mlx5e_ipsec_stats, ipsec_dec_in_packets) }, @@ -73,61 +74,74 @@ static const struct counter_desc mlx5e_ipsec_sw_stats_desc[] = { #define NUM_IPSEC_HW_COUNTERS ARRAY_SIZE(mlx5e_ipsec_hw_stats_desc) #define NUM_IPSEC_SW_COUNTERS ARRAY_SIZE(mlx5e_ipsec_sw_stats_desc) -#define NUM_IPSEC_COUNTERS (NUM_IPSEC_HW_COUNTERS + NUM_IPSEC_SW_COUNTERS) - -int mlx5e_ipsec_get_count(struct mlx5e_priv *priv) +static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(ipsec_sw) { - if (!priv->ipsec) - return 0; - - return NUM_IPSEC_COUNTERS; + return NUM_IPSEC_SW_COUNTERS; } -int mlx5e_ipsec_get_strings(struct mlx5e_priv *priv, uint8_t *data) -{ - unsigned int i, idx = 0; +static inline MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(ipsec_sw) {} - if (!priv->ipsec) - return 0; +static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(ipsec_sw) +{ + unsigned int i; - for (i = 0; i < NUM_IPSEC_HW_COUNTERS; i++) - strcpy(data + (idx++) * ETH_GSTRING_LEN, - mlx5e_ipsec_hw_stats_desc[i].format); + if (priv->ipsec) + for (i = 0; i < NUM_IPSEC_SW_COUNTERS; i++) + strcpy(data + (idx++) * ETH_GSTRING_LEN, + mlx5e_ipsec_sw_stats_desc[i].format); + return idx; +} - for (i = 0; i < NUM_IPSEC_SW_COUNTERS; i++) - strcpy(data + (idx++) * ETH_GSTRING_LEN, - mlx5e_ipsec_sw_stats_desc[i].format); +static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(ipsec_sw) +{ + int i; - return NUM_IPSEC_COUNTERS; + if (priv->ipsec) + for (i = 0; i < NUM_IPSEC_SW_COUNTERS; i++) + data[idx++] = MLX5E_READ_CTR_ATOMIC64(&priv->ipsec->sw_stats, + mlx5e_ipsec_sw_stats_desc, i); + return idx; } -void mlx5e_ipsec_update_stats(struct mlx5e_priv *priv) +static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(ipsec_hw) { - int ret; + return (mlx5_fpga_ipsec_device_caps(priv->mdev)) ? NUM_IPSEC_HW_COUNTERS : 0; +} - if (!priv->ipsec) - return; +static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(ipsec_hw) +{ + int ret = 0; - ret = mlx5_accel_ipsec_counters_read(priv->mdev, (u64 *)&priv->ipsec->stats, - NUM_IPSEC_HW_COUNTERS); + if (priv->ipsec) + ret = mlx5_accel_ipsec_counters_read(priv->mdev, (u64 *)&priv->ipsec->stats, + NUM_IPSEC_HW_COUNTERS); if (ret) memset(&priv->ipsec->stats, 0, sizeof(priv->ipsec->stats)); } -int mlx5e_ipsec_get_stats(struct mlx5e_priv *priv, u64 *data) +static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(ipsec_hw) { - int i, idx = 0; - - if (!priv->ipsec) - return 0; + unsigned int i; - for (i = 0; i < NUM_IPSEC_HW_COUNTERS; i++) - data[idx++] = MLX5E_READ_CTR64_CPU(&priv->ipsec->stats, - mlx5e_ipsec_hw_stats_desc, i); + if (priv->ipsec && mlx5_fpga_ipsec_device_caps(priv->mdev)) + for (i = 0; i < NUM_IPSEC_HW_COUNTERS; i++) + strcpy(data + (idx++) * ETH_GSTRING_LEN, + mlx5e_ipsec_hw_stats_desc[i].format); - for (i = 0; i < NUM_IPSEC_SW_COUNTERS; i++) - data[idx++] = MLX5E_READ_CTR_ATOMIC64(&priv->ipsec->sw_stats, - mlx5e_ipsec_sw_stats_desc, i); + return idx; +} - return NUM_IPSEC_COUNTERS; +static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(ipsec_hw) +{ + int i; + + if (priv->ipsec && mlx5_fpga_ipsec_device_caps(priv->mdev)) + for (i = 0; i < NUM_IPSEC_HW_COUNTERS; i++) + data[idx++] = MLX5E_READ_CTR64_CPU(&priv->ipsec->stats, + mlx5e_ipsec_hw_stats_desc, + i); + return idx; } + +MLX5E_DEFINE_STATS_GRP(ipsec_sw, 0); +MLX5E_DEFINE_STATS_GRP(ipsec_hw, 0); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.c index 46725cd743a3..452fcf59c36b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.c @@ -4,6 +4,19 @@ #include "en.h" #include "en_accel/ktls.h" +u16 mlx5e_ktls_get_stop_room(struct mlx5e_txqsq *sq) +{ + u16 num_dumps, stop_room = 0; + + num_dumps = mlx5e_ktls_dumps_num_wqes(sq, MAX_SKB_FRAGS, TLS_MAX_PAYLOAD_SIZE); + + stop_room += mlx5e_stop_room_for_wqe(MLX5E_KTLS_STATIC_WQEBBS); + stop_room += mlx5e_stop_room_for_wqe(MLX5E_KTLS_PROGRESS_WQEBBS); + stop_room += num_dumps * mlx5e_stop_room_for_wqe(MLX5E_KTLS_DUMP_WQEBBS); + + return stop_room; +} + static int mlx5e_ktls_create_tis(struct mlx5_core_dev *mdev, u32 *tisn) { u32 in[MLX5_ST_SZ_DW(create_tis_in)] = {}; @@ -69,8 +82,8 @@ static void mlx5e_ktls_del(struct net_device *netdev, struct mlx5e_ktls_offload_context_tx *tx_priv = mlx5e_get_ktls_tx_priv_ctx(tls_ctx); - mlx5_ktls_destroy_key(priv->mdev, tx_priv->key_id); mlx5e_destroy_tis(priv->mdev, tx_priv->tisn); + mlx5_ktls_destroy_key(priv->mdev, tx_priv->key_id); kvfree(tx_priv); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.h index 63116be6b1d6..c6180892cfcb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls.h @@ -9,6 +9,7 @@ #ifdef CONFIG_MLX5_EN_TLS #include <net/tls.h> #include "accel/tls.h" +#include "en_accel/tls_rxtx.h" #define MLX5E_KTLS_STATIC_UMR_WQE_SZ \ (offsetof(struct mlx5e_umr_wqe, tls_static_params_ctx) + \ @@ -27,6 +28,14 @@ struct mlx5e_dump_wqe { struct mlx5_wqe_data_seg data; }; +#define MLX5E_TLS_FETCH_UMR_WQE(sq, pi) \ + ((struct mlx5e_umr_wqe *)mlx5e_fetch_wqe(&(sq)->wq, pi, MLX5E_KTLS_STATIC_UMR_WQE_SZ)) +#define MLX5E_TLS_FETCH_PROGRESS_WQE(sq, pi) \ + ((struct mlx5e_tx_wqe *)mlx5e_fetch_wqe(&(sq)->wq, pi, MLX5E_KTLS_PROGRESS_WQE_SZ)) +#define MLX5E_TLS_FETCH_DUMP_WQE(sq, pi) \ + ((struct mlx5e_dump_wqe *)mlx5e_fetch_wqe(&(sq)->wq, pi, \ + sizeof(struct mlx5e_dump_wqe))) + #define MLX5E_KTLS_DUMP_WQEBBS \ (DIV_ROUND_UP(sizeof(struct mlx5e_dump_wqe), MLX5_SEND_WQE_BB)) @@ -87,22 +96,22 @@ mlx5e_get_ktls_tx_priv_ctx(struct tls_context *tls_ctx) void mlx5e_ktls_build_netdev(struct mlx5e_priv *priv); void mlx5e_ktls_tx_offload_set_pending(struct mlx5e_ktls_offload_context_tx *priv_tx); -struct sk_buff *mlx5e_ktls_handle_tx_skb(struct net_device *netdev, - struct mlx5e_txqsq *sq, - struct sk_buff *skb, - struct mlx5e_tx_wqe **wqe, u16 *pi); +bool mlx5e_ktls_handle_tx_skb(struct tls_context *tls_ctx, struct mlx5e_txqsq *sq, + struct sk_buff *skb, int datalen, + struct mlx5e_accel_tx_tls_state *state); void mlx5e_ktls_tx_handle_resync_dump_comp(struct mlx5e_txqsq *sq, struct mlx5e_tx_wqe_info *wi, u32 *dma_fifo_cc); +u16 mlx5e_ktls_get_stop_room(struct mlx5e_txqsq *sq); + static inline u8 -mlx5e_ktls_dumps_num_wqebbs(struct mlx5e_txqsq *sq, unsigned int nfrags, - unsigned int sync_len) +mlx5e_ktls_dumps_num_wqes(struct mlx5e_txqsq *sq, unsigned int nfrags, + unsigned int sync_len) { /* Given the MTU and sync_len, calculates an upper bound for the - * number of WQEBBs needed for the TX resync DUMP WQEs of a record. + * number of DUMP WQEs needed for the TX resync of a record. */ - return MLX5E_KTLS_DUMP_WQEBBS * - (nfrags + DIV_ROUND_UP(sync_len, sq->hw_mtu)); + return nfrags + DIV_ROUND_UP(sync_len, sq->hw_mtu); } #else @@ -114,7 +123,6 @@ static inline void mlx5e_ktls_tx_handle_resync_dump_comp(struct mlx5e_txqsq *sq, struct mlx5e_tx_wqe_info *wi, u32 *dma_fifo_cc) {} - #endif #endif /* __MLX5E_TLS_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c index 52a56622034a..3cd78d9503c1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c @@ -108,10 +108,11 @@ static void tx_fill_wi(struct mlx5e_txqsq *sq, { struct mlx5e_tx_wqe_info *wi = &sq->db.wqe_info[pi]; - memset(wi, 0, sizeof(*wi)); - wi->num_wqebbs = num_wqebbs; - wi->num_bytes = num_bytes; - wi->resync_dump_frag_page = page; + *wi = (struct mlx5e_tx_wqe_info) { + .num_wqebbs = num_wqebbs, + .num_bytes = num_bytes, + .resync_dump_frag_page = page, + }; } void mlx5e_ktls_tx_offload_set_pending(struct mlx5e_ktls_offload_context_tx *priv_tx) @@ -134,13 +135,14 @@ post_static_params(struct mlx5e_txqsq *sq, struct mlx5e_ktls_offload_context_tx *priv_tx, bool fence) { + u16 pi, num_wqebbs = MLX5E_KTLS_STATIC_WQEBBS; struct mlx5e_umr_wqe *umr_wqe; - u16 pi; - umr_wqe = mlx5e_sq_fetch_wqe(sq, MLX5E_KTLS_STATIC_UMR_WQE_SZ, &pi); + pi = mlx5e_txqsq_get_next_pi(sq, num_wqebbs); + umr_wqe = MLX5E_TLS_FETCH_UMR_WQE(sq, pi); build_static_params(umr_wqe, sq->pc, sq->sqn, priv_tx, fence); - tx_fill_wi(sq, pi, MLX5E_KTLS_STATIC_WQEBBS, 0, NULL); - sq->pc += MLX5E_KTLS_STATIC_WQEBBS; + tx_fill_wi(sq, pi, num_wqebbs, 0, NULL); + sq->pc += num_wqebbs; } static void @@ -148,13 +150,14 @@ post_progress_params(struct mlx5e_txqsq *sq, struct mlx5e_ktls_offload_context_tx *priv_tx, bool fence) { + u16 pi, num_wqebbs = MLX5E_KTLS_PROGRESS_WQEBBS; struct mlx5e_tx_wqe *wqe; - u16 pi; - wqe = mlx5e_sq_fetch_wqe(sq, MLX5E_KTLS_PROGRESS_WQE_SZ, &pi); + pi = mlx5e_txqsq_get_next_pi(sq, num_wqebbs); + wqe = MLX5E_TLS_FETCH_PROGRESS_WQE(sq, pi); build_progress_params(wqe, sq->pc, sq->sqn, priv_tx, fence); - tx_fill_wi(sq, pi, MLX5E_KTLS_PROGRESS_WQEBBS, 0, NULL); - sq->pc += MLX5E_KTLS_PROGRESS_WQEBBS; + tx_fill_wi(sq, pi, num_wqebbs, 0, NULL); + sq->pc += num_wqebbs; } static void @@ -163,14 +166,6 @@ mlx5e_ktls_tx_post_param_wqes(struct mlx5e_txqsq *sq, bool skip_static_post, bool fence_first_post) { bool progress_fence = skip_static_post || !fence_first_post; - struct mlx5_wq_cyc *wq = &sq->wq; - u16 contig_wqebbs_room, pi; - - pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - contig_wqebbs_room = mlx5_wq_cyc_get_contig_wqebbs(wq, pi); - if (unlikely(contig_wqebbs_room < - MLX5E_KTLS_STATIC_WQEBBS + MLX5E_KTLS_PROGRESS_WQEBBS)) - mlx5e_fill_sq_frag_edge(sq, wq, pi, contig_wqebbs_room); if (!skip_static_post) post_static_params(sq, priv_tx, fence_first_post); @@ -278,7 +273,9 @@ tx_post_resync_dump(struct mlx5e_txqsq *sq, skb_frag_t *frag, u32 tisn, bool fir int fsz; u16 pi; - wqe = mlx5e_sq_fetch_wqe(sq, sizeof(*wqe), &pi); + BUILD_BUG_ON(MLX5E_KTLS_DUMP_WQEBBS != 1); + pi = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->pc); + wqe = MLX5E_TLS_FETCH_DUMP_WQE(sq, pi); ds_cnt = sizeof(*wqe) / MLX5_SEND_WQE_DS; @@ -343,11 +340,8 @@ mlx5e_ktls_tx_handle_ooo(struct mlx5e_ktls_offload_context_tx *priv_tx, u32 seq) { struct mlx5e_sq_stats *stats = sq->stats; - struct mlx5_wq_cyc *wq = &sq->wq; enum mlx5e_ktls_sync_retval ret; struct tx_sync_info info = {}; - u16 contig_wqebbs_room, pi; - u8 num_wqebbs; int i = 0; ret = tx_sync_info_get(priv_tx, seq, datalen, &info); @@ -376,13 +370,6 @@ mlx5e_ktls_tx_handle_ooo(struct mlx5e_ktls_offload_context_tx *priv_tx, return MLX5E_KTLS_SYNC_DONE; } - num_wqebbs = mlx5e_ktls_dumps_num_wqebbs(sq, info.nr_frags, info.sync_len); - pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - contig_wqebbs_room = mlx5_wq_cyc_get_contig_wqebbs(wq, pi); - - if (unlikely(contig_wqebbs_room < num_wqebbs)) - mlx5e_fill_sq_frag_edge(sq, wq, pi, contig_wqebbs_room); - for (; i < info.nr_frags; i++) { unsigned int orig_fsz, frag_offset = 0, n = 0; skb_frag_t *f = &info.frags[i]; @@ -422,34 +409,18 @@ err_out: return MLX5E_KTLS_SYNC_FAIL; } -struct sk_buff *mlx5e_ktls_handle_tx_skb(struct net_device *netdev, - struct mlx5e_txqsq *sq, - struct sk_buff *skb, - struct mlx5e_tx_wqe **wqe, u16 *pi) +bool mlx5e_ktls_handle_tx_skb(struct tls_context *tls_ctx, struct mlx5e_txqsq *sq, + struct sk_buff *skb, int datalen, + struct mlx5e_accel_tx_tls_state *state) { struct mlx5e_ktls_offload_context_tx *priv_tx; struct mlx5e_sq_stats *stats = sq->stats; - struct mlx5_wqe_ctrl_seg *cseg; - struct tls_context *tls_ctx; - int datalen; u32 seq; - if (!skb->sk || !tls_is_sk_tx_device_offloaded(skb->sk)) - goto out; - - datalen = skb->len - (skb_transport_offset(skb) + tcp_hdrlen(skb)); - if (!datalen) - goto out; - - tls_ctx = tls_get_ctx(skb->sk); - if (WARN_ON_ONCE(tls_ctx->netdev != netdev)) - goto err_out; - priv_tx = mlx5e_get_ktls_tx_priv_ctx(tls_ctx); if (unlikely(mlx5e_ktls_tx_offload_test_and_clear_pending(priv_tx))) { mlx5e_ktls_tx_post_param_wqes(sq, priv_tx, false, false); - *wqe = mlx5e_sq_fetch_wqe(sq, sizeof(**wqe), pi); stats->tls_ctx++; } @@ -460,30 +431,28 @@ struct sk_buff *mlx5e_ktls_handle_tx_skb(struct net_device *netdev, switch (ret) { case MLX5E_KTLS_SYNC_DONE: - *wqe = mlx5e_sq_fetch_wqe(sq, sizeof(**wqe), pi); break; case MLX5E_KTLS_SYNC_SKIP_NO_DATA: if (likely(!skb->decrypted)) goto out; WARN_ON_ONCE(1); /* fall-through */ - default: /* MLX5E_KTLS_SYNC_FAIL */ + case MLX5E_KTLS_SYNC_FAIL: goto err_out; } } priv_tx->expected_seq = seq + datalen; - cseg = &(*wqe)->ctrl; - cseg->tisn = cpu_to_be32(priv_tx->tisn << 8); + state->tls_tisn = priv_tx->tisn; stats->tls_encrypted_packets += skb_is_gso(skb) ? skb_shinfo(skb)->gso_segs : 1; stats->tls_encrypted_bytes += datalen; out: - return skb; + return true; err_out: dev_kfree_skb_any(skb); - return NULL; + return false; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.c index fba561ffe1d4..1fbb5a90cb38 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.c @@ -167,7 +167,7 @@ static int mlx5e_tls_resync(struct net_device *netdev, struct sock *sk, struct tls_context *tls_ctx = tls_get_ctx(sk); struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5e_tls_offload_context_rx *rx_ctx; - u64 rcd_sn = *(u64 *)rcd_sn_data; + __be64 rcd_sn = *(__be64 *)rcd_sn_data; if (WARN_ON_ONCE(direction != TLS_OFFLOAD_CTX_DIR_RX)) return -EINVAL; @@ -240,3 +240,17 @@ void mlx5e_tls_cleanup(struct mlx5e_priv *priv) kfree(tls); priv->tls = NULL; } + +u16 mlx5e_tls_get_stop_room(struct mlx5e_txqsq *sq) +{ + struct mlx5_core_dev *mdev = sq->channel->mdev; + + if (!mlx5_accel_is_tls_device(mdev)) + return 0; + + if (MLX5_CAP_GEN(mdev, tls_tx)) + return mlx5e_ktls_get_stop_room(sq); + + /* Resync SKB. */ + return mlx5e_stop_room_for_wqe(MLX5_SEND_WQE_MAX_WQEBBS); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.h index 9015f3f7792d..9219bdb2786e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.h @@ -94,6 +94,8 @@ int mlx5e_tls_get_count(struct mlx5e_priv *priv); int mlx5e_tls_get_strings(struct mlx5e_priv *priv, uint8_t *data); int mlx5e_tls_get_stats(struct mlx5e_priv *priv, u64 *data); +u16 mlx5e_tls_get_stop_room(struct mlx5e_txqsq *sq); + #else static inline void mlx5e_tls_build_netdev(struct mlx5e_priv *priv) @@ -108,6 +110,11 @@ static inline int mlx5e_tls_get_count(struct mlx5e_priv *priv) { return 0; } static inline int mlx5e_tls_get_strings(struct mlx5e_priv *priv, uint8_t *data) { return 0; } static inline int mlx5e_tls_get_stats(struct mlx5e_priv *priv, u64 *data) { return 0; } +static inline u16 mlx5e_tls_get_stop_room(struct mlx5e_txqsq *sq) +{ + return 0; +} + #endif #endif /* __MLX5E_TLS_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c index ef1ed15a53b4..05454a843b28 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c @@ -184,18 +184,17 @@ static void mlx5e_tls_complete_sync_skb(struct sk_buff *skb, nskb->queue_mapping = skb->queue_mapping; } -static struct sk_buff * -mlx5e_tls_handle_ooo(struct mlx5e_tls_offload_context_tx *context, - struct mlx5e_txqsq *sq, struct sk_buff *skb, - struct mlx5e_tx_wqe **wqe, - u16 *pi, - struct mlx5e_tls *tls) +static bool mlx5e_tls_handle_ooo(struct mlx5e_tls_offload_context_tx *context, + struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5e_tls *tls) { u32 tcp_seq = ntohl(tcp_hdr(skb)->seq); + struct mlx5e_tx_wqe *wqe; struct sync_info info; struct sk_buff *nskb; int linear_len = 0; int headln; + u16 pi; int i; sq->stats->tls_ooo++; @@ -217,7 +216,7 @@ mlx5e_tls_handle_ooo(struct mlx5e_tls_offload_context_tx *context, if (likely(payload <= -info.sync_len)) /* SKB payload doesn't require offload */ - return skb; + return true; atomic64_inc(&tls->sw_stats.tx_tls_drop_bypass_required); goto err_out; @@ -247,20 +246,19 @@ mlx5e_tls_handle_ooo(struct mlx5e_tls_offload_context_tx *context, sq->stats->tls_resync_bytes += nskb->len; mlx5e_tls_complete_sync_skb(skb, nskb, tcp_seq, headln, cpu_to_be64(info.rcd_sn)); - mlx5e_sq_xmit(sq, nskb, *wqe, *pi, true); - *wqe = mlx5e_sq_fetch_wqe(sq, sizeof(**wqe), pi); - return skb; + pi = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->pc); + wqe = MLX5E_TX_FETCH_WQE(sq, pi); + mlx5e_sq_xmit(sq, nskb, wqe, pi, true); + + return true; err_out: dev_kfree_skb_any(skb); - return NULL; + return false; } -struct sk_buff *mlx5e_tls_handle_tx_skb(struct net_device *netdev, - struct mlx5e_txqsq *sq, - struct sk_buff *skb, - struct mlx5e_tx_wqe **wqe, - u16 *pi) +bool mlx5e_tls_handle_tx_skb(struct net_device *netdev, struct mlx5e_txqsq *sq, + struct sk_buff *skb, struct mlx5e_accel_tx_tls_state *state) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5e_tls_offload_context_tx *context; @@ -269,41 +267,45 @@ struct sk_buff *mlx5e_tls_handle_tx_skb(struct net_device *netdev, int datalen; u32 skb_seq; - if (MLX5_CAP_GEN(sq->channel->mdev, tls_tx)) { - skb = mlx5e_ktls_handle_tx_skb(netdev, sq, skb, wqe, pi); - goto out; - } - if (!skb->sk || !tls_is_sk_tx_device_offloaded(skb->sk)) - goto out; + return true; datalen = skb->len - (skb_transport_offset(skb) + tcp_hdrlen(skb)); if (!datalen) - goto out; + return true; tls_ctx = tls_get_ctx(skb->sk); - if (unlikely(tls_ctx->netdev != netdev)) - goto out; + if (WARN_ON_ONCE(tls_ctx->netdev != netdev)) + goto err_out; + + if (MLX5_CAP_GEN(sq->channel->mdev, tls_tx)) + return mlx5e_ktls_handle_tx_skb(tls_ctx, sq, skb, datalen, state); skb_seq = ntohl(tcp_hdr(skb)->seq); context = mlx5e_get_tls_tx_context(tls_ctx); expected_seq = context->expected_seq; - if (unlikely(expected_seq != skb_seq)) { - skb = mlx5e_tls_handle_ooo(context, sq, skb, wqe, pi, priv->tls); - goto out; - } + if (unlikely(expected_seq != skb_seq)) + return mlx5e_tls_handle_ooo(context, sq, skb, priv->tls); if (unlikely(mlx5e_tls_add_metadata(skb, context->swid))) { atomic64_inc(&priv->tls->sw_stats.tx_tls_drop_metadata); dev_kfree_skb_any(skb); - skb = NULL; - goto out; + return false; } context->expected_seq = skb_seq + datalen; -out: - return skb; + return true; + +err_out: + dev_kfree_skb_any(skb); + return false; +} + +void mlx5e_tls_handle_tx_wqe(struct mlx5e_txqsq *sq, struct mlx5_wqe_ctrl_seg *cseg, + struct mlx5e_accel_tx_tls_state *state) +{ + cseg->tisn = cpu_to_be32(state->tls_tisn << 8); } static int tls_update_resync_sn(struct net_device *netdev, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.h index 90bc1f2384c8..a50d0394df0a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.h @@ -40,11 +40,14 @@ #include "en.h" #include "en/txrx.h" -struct sk_buff *mlx5e_tls_handle_tx_skb(struct net_device *netdev, - struct mlx5e_txqsq *sq, - struct sk_buff *skb, - struct mlx5e_tx_wqe **wqe, - u16 *pi); +struct mlx5e_accel_tx_tls_state { + u32 tls_tisn; +}; + +bool mlx5e_tls_handle_tx_skb(struct net_device *netdev, struct mlx5e_txqsq *sq, + struct sk_buff *skb, struct mlx5e_accel_tx_tls_state *state); +void mlx5e_tls_handle_tx_wqe(struct mlx5e_txqsq *sq, struct mlx5_wqe_ctrl_seg *cseg, + struct mlx5e_accel_tx_tls_state *state); void mlx5e_tls_handle_rx_skb(struct net_device *netdev, struct sk_buff *skb, u32 *cqe_bcnt); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c index af3228b3f303..1e42c7ae621b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c @@ -141,10 +141,12 @@ void mlx5e_destroy_mdev_resources(struct mlx5_core_dev *mdev) memset(res, 0, sizeof(*res)); } -int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb) +int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb, + bool enable_mc_lb) { struct mlx5_core_dev *mdev = priv->mdev; struct mlx5e_tir *tir; + u8 lb_flags = 0; int err = 0; u32 tirn = 0; int inlen; @@ -158,8 +160,13 @@ int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb) } if (enable_uc_lb) - MLX5_SET(modify_tir_in, in, ctx.self_lb_block, - MLX5_TIRC_SELF_LB_BLOCK_BLOCK_UNICAST); + lb_flags = MLX5_TIRC_SELF_LB_BLOCK_BLOCK_UNICAST; + + if (enable_mc_lb) + lb_flags |= MLX5_TIRC_SELF_LB_BLOCK_BLOCK_MULTICAST; + + if (lb_flags) + MLX5_SET(modify_tir_in, in, ctx.self_lb_block, lb_flags); MLX5_SET(modify_tir_in, in, bitmask.self_lb_en, 1); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c index 47874d34156b..bc102d094bbd 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_dcbnl.c @@ -35,6 +35,8 @@ #include "en/port.h" #include "en/port_buffer.h" +#define MLX5E_MAX_BW_ALLOC 100 /* Max percentage of BW allocation */ + #define MLX5E_100MB (100000) #define MLX5E_1GB (1000000) @@ -49,6 +51,12 @@ enum { MLX5E_LOWEST_PRIO_GROUP = 0, }; +enum { + MLX5_DCB_CHG_RESET, + MLX5_DCB_NO_CHG, + MLX5_DCB_CHG_NO_RESET, +}; + #define MLX5_DSCP_SUPPORTED(mdev) (MLX5_CAP_GEN(mdev, qcam_reg) && \ MLX5_CAP_QCAM_REG(mdev, qpts) && \ MLX5_CAP_QCAM_REG(mdev, qpdpm)) @@ -238,7 +246,7 @@ static void mlx5e_build_tc_tx_bw(struct ieee_ets *ets, u8 *tc_tx_bw, * Report both group #0 and #1 as ETS type. * All the tcs in group #0 will be reported with 0% BW. */ -int mlx5e_dcbnl_ieee_setets_core(struct mlx5e_priv *priv, struct ieee_ets *ets) +static int mlx5e_dcbnl_ieee_setets_core(struct mlx5e_priv *priv, struct ieee_ets *ets) { struct mlx5_core_dev *mdev = priv->mdev; u8 tc_tx_bw[IEEE_8021QAZ_MAX_TCS]; @@ -977,7 +985,7 @@ static int mlx5e_dcbnl_setbuffer(struct net_device *dev, return err; } -const struct dcbnl_rtnl_ops mlx5e_dcbnl_ops = { +static const struct dcbnl_rtnl_ops mlx5e_dcbnl_ops = { .ieee_getets = mlx5e_dcbnl_ieee_getets, .ieee_setets = mlx5e_dcbnl_ieee_setets, .ieee_getmaxrate = mlx5e_dcbnl_ieee_getmaxrate, @@ -1009,6 +1017,24 @@ const struct dcbnl_rtnl_ops mlx5e_dcbnl_ops = { .setpfcstate = mlx5e_dcbnl_setpfcstate, }; +void mlx5e_dcbnl_build_netdev(struct net_device *netdev) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + + if (MLX5_CAP_GEN(mdev, vport_group_manager) && MLX5_CAP_GEN(mdev, qos)) + netdev->dcbnl_ops = &mlx5e_dcbnl_ops; +} + +void mlx5e_dcbnl_build_rep_netdev(struct net_device *netdev) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + + if (MLX5_CAP_GEN(mdev, qos)) + netdev->dcbnl_ops = &mlx5e_dcbnl_ops; +} + static void mlx5e_dcbnl_query_dcbx_mode(struct mlx5e_priv *priv, enum mlx5_dcbx_oper_mode *mode) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index de8250820b06..3ef2525e8de9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -432,7 +432,7 @@ int mlx5e_ethtool_set_channels(struct mlx5e_priv *priv, if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { *cur_params = new_channels.params; - mlx5e_num_channels_changed(priv); + err = mlx5e_num_channels_changed(priv); goto out; } @@ -527,8 +527,8 @@ int mlx5e_ethtool_set_coalesce(struct mlx5e_priv *priv, struct dim_cq_moder *rx_moder, *tx_moder; struct mlx5_core_dev *mdev = priv->mdev; struct mlx5e_channels new_channels = {}; + bool reset_rx, reset_tx; int err = 0; - bool reset; if (!MLX5_CAP_GEN(mdev, cq_moderation)) return -EOPNOTSUPP; @@ -566,15 +566,28 @@ int mlx5e_ethtool_set_coalesce(struct mlx5e_priv *priv, } /* we are opened */ - reset = (!!coal->use_adaptive_rx_coalesce != priv->channels.params.rx_dim_enabled) || - (!!coal->use_adaptive_tx_coalesce != priv->channels.params.tx_dim_enabled); + reset_rx = !!coal->use_adaptive_rx_coalesce != priv->channels.params.rx_dim_enabled; + reset_tx = !!coal->use_adaptive_tx_coalesce != priv->channels.params.tx_dim_enabled; - if (!reset) { + if (!reset_rx && !reset_tx) { mlx5e_set_priv_channels_coalesce(priv, coal); priv->channels.params = new_channels.params; goto out; } + if (reset_rx) { + u8 mode = MLX5E_GET_PFLAG(&new_channels.params, + MLX5E_PFLAG_RX_CQE_BASED_MODER); + + mlx5e_reset_rx_moderation(&new_channels.params, mode); + } + if (reset_tx) { + u8 mode = MLX5E_GET_PFLAG(&new_channels.params, + MLX5E_PFLAG_TX_CQE_BASED_MODER); + + mlx5e_reset_tx_moderation(&new_channels.params, mode); + } + err = mlx5e_safe_switch_channels(priv, &new_channels, NULL, NULL); out: @@ -665,11 +678,12 @@ static const u32 pplm_fec_2_ethtool_linkmodes[] = { static int get_fec_supported_advertised(struct mlx5_core_dev *dev, struct ethtool_link_ksettings *link_ksettings) { - u_long active_fec = 0; + unsigned long active_fec_long; + u32 active_fec; u32 bitn; int err; - err = mlx5e_get_fec_mode(dev, (u32 *)&active_fec, NULL); + err = mlx5e_get_fec_mode(dev, &active_fec, NULL); if (err) return (err == -EOPNOTSUPP) ? 0 : err; @@ -682,10 +696,11 @@ static int get_fec_supported_advertised(struct mlx5_core_dev *dev, MLX5E_ADVERTISE_SUPPORTED_FEC(MLX5E_FEC_LLRS_272_257_1, ETHTOOL_LINK_MODE_FEC_LLRS_BIT); + active_fec_long = active_fec; /* active fec is a bit set, find out which bit is set and * advertise the corresponding ethtool bit */ - bitn = find_first_bit(&active_fec, sizeof(u32) * BITS_PER_BYTE); + bitn = find_first_bit(&active_fec_long, sizeof(active_fec_long) * BITS_PER_BYTE); if (bitn < ARRAY_SIZE(pplm_fec_2_ethtool_linkmodes)) __set_bit(pplm_fec_2_ethtool_linkmodes[bitn], link_ksettings->link_modes.advertising); @@ -1517,8 +1532,8 @@ static int mlx5e_get_fecparam(struct net_device *netdev, { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5_core_dev *mdev = priv->mdev; - u16 fec_configured = 0; - u32 fec_active = 0; + u16 fec_configured; + u32 fec_active; int err; err = mlx5e_get_fec_mode(mdev, &fec_active, &fec_configured); @@ -1526,14 +1541,14 @@ static int mlx5e_get_fecparam(struct net_device *netdev, if (err) return err; - fecparam->active_fec = pplm2ethtool_fec((u_long)fec_active, - sizeof(u32) * BITS_PER_BYTE); + fecparam->active_fec = pplm2ethtool_fec((unsigned long)fec_active, + sizeof(unsigned long) * BITS_PER_BYTE); if (!fecparam->active_fec) return -EOPNOTSUPP; - fecparam->fec = pplm2ethtool_fec((u_long)fec_configured, - sizeof(u16) * BITS_PER_BYTE); + fecparam->fec = pplm2ethtool_fec((unsigned long)fec_configured, + sizeof(unsigned long) * BITS_PER_BYTE); return 0; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index a942c06e9ba4..a836a02a2116 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -38,7 +38,7 @@ #include <linux/bpf.h> #include <linux/if_bridge.h> #include <net/page_pool.h> -#include <net/xdp_sock.h> +#include <net/xdp_sock_drv.h> #include "eswitch.h" #include "en.h" #include "en/txrx.h" @@ -66,7 +66,6 @@ #include "en/devlink.h" #include "lib/mlx5.h" - bool mlx5e_check_fragmented_striding_rq_cap(struct mlx5_core_dev *mdev) { bool striding_rq_umr = MLX5_CAP_GEN(mdev, striding_rq) && @@ -233,7 +232,7 @@ static inline void mlx5e_build_umr_wqe(struct mlx5e_rq *rq, cseg->qpn_ds = cpu_to_be32((sq->sqn << MLX5_WQE_CTRL_QPN_SHIFT) | ds_cnt); cseg->fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE; - cseg->imm = rq->mkey_be; + cseg->umr_mkey = rq->mkey_be; ucseg->flags = MLX5_UMR_TRANSLATION_OFFSET_EN | MLX5_UMR_INLINE; ucseg->xlt_octowords = @@ -374,7 +373,6 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, struct mlx5_core_dev *mdev = c->mdev; void *rqc = rqp->rqc; void *rqc_wq = MLX5_ADDR_OF(rqc, rqc, wq); - u32 num_xsk_frames = 0; u32 rq_xdp_ix; u32 pool_size; int wq_sz; @@ -414,7 +412,6 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, rq->buff.map_dir = rq->xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE; rq->buff.headroom = mlx5e_get_rq_headroom(mdev, params, xsk); - rq->buff.umem_headroom = xsk ? xsk->headroom : 0; pool_size = 1 << params->log_rq_mtu_frames; switch (rq->wq_type) { @@ -428,10 +425,6 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, wq_sz = mlx5_wq_ll_get_size(&rq->mpwqe.wq); - if (xsk) - num_xsk_frames = wq_sz << - mlx5e_mpwqe_get_log_num_strides(mdev, params, xsk); - pool_size = MLX5_MPWRQ_PAGES_PER_WQE << mlx5e_mpwqe_get_log_rq_size(params, xsk); @@ -462,6 +455,8 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, rq->mpwqe.num_strides = BIT(mlx5e_mpwqe_get_log_num_strides(mdev, params, xsk)); + rq->buff.frame0_sz = (1 << rq->mpwqe.log_stride_sz); + err = mlx5e_create_rq_umr_mkey(mdev, rq); if (err) goto err_rq_wq_destroy; @@ -481,10 +476,9 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, wq_sz = mlx5_wq_cyc_get_size(&rq->wqe.wq); - if (xsk) - num_xsk_frames = wq_sz << rq->wqe.info.log_num_frags; - rq->wqe.info = rqp->frags_info; + rq->buff.frame0_sz = rq->wqe.info.arr[0].frag_stride; + rq->wqe.frags = kvzalloc_node(array_size(sizeof(*rq->wqe.frags), (wq_sz << rq->wqe.info.log_num_frags)), @@ -522,17 +516,9 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, } if (xsk) { - err = mlx5e_xsk_resize_reuseq(umem, num_xsk_frames); - if (unlikely(err)) { - mlx5_core_err(mdev, "Unable to allocate the Reuse Ring for %u frames\n", - num_xsk_frames); - goto err_free; - } - - rq->zca.free = mlx5e_xsk_zca_free; err = xdp_rxq_info_reg_mem_model(&rq->xdp_rxq, - MEM_TYPE_ZERO_COPY, - &rq->zca); + MEM_TYPE_XSK_BUFF_POOL, NULL); + xsk_buff_set_rxq_info(rq->umem, &rq->xdp_rxq); } else { /* Create a page_pool and register it with rxq */ pp_params.order = 0; @@ -1027,17 +1013,17 @@ static void mlx5e_free_xdpsq(struct mlx5e_xdpsq *sq) static void mlx5e_free_icosq_db(struct mlx5e_icosq *sq) { - kvfree(sq->db.ico_wqe); + kvfree(sq->db.wqe_info); } static int mlx5e_alloc_icosq_db(struct mlx5e_icosq *sq, int numa) { int wq_sz = mlx5_wq_cyc_get_size(&sq->wq); + size_t size; - sq->db.ico_wqe = kvzalloc_node(array_size(wq_sz, - sizeof(*sq->db.ico_wqe)), - GFP_KERNEL, numa); - if (!sq->db.ico_wqe) + size = array_size(wq_sz, sizeof(*sq->db.wqe_info)); + sq->db.wqe_info = kvzalloc_node(size, GFP_KERNEL, numa); + if (!sq->db.wqe_info) return -ENOMEM; return 0; @@ -1116,6 +1102,22 @@ static int mlx5e_alloc_txqsq_db(struct mlx5e_txqsq *sq, int numa) return 0; } +static int mlx5e_calc_sq_stop_room(struct mlx5e_txqsq *sq, u8 log_sq_size) +{ + int sq_size = 1 << log_sq_size; + + sq->stop_room = mlx5e_tls_get_stop_room(sq); + sq->stop_room += mlx5e_stop_room_for_wqe(MLX5_SEND_WQE_MAX_WQEBBS); + + if (WARN_ON(sq->stop_room >= sq_size)) { + netdev_err(sq->channel->netdev, "Stop room %hu is bigger than the SQ size %d\n", + sq->stop_room, sq_size); + return -ENOSPC; + } + + return 0; +} + static void mlx5e_tx_err_cqe_work(struct work_struct *recover_work); static int mlx5e_alloc_txqsq(struct mlx5e_channel *c, int txq_ix, @@ -1140,20 +1142,16 @@ static int mlx5e_alloc_txqsq(struct mlx5e_channel *c, sq->min_inline_mode = params->tx_min_inline_mode; sq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); sq->stats = &c->priv->channel_stats[c->ix].sq[tc]; - sq->stop_room = MLX5E_SQ_STOP_ROOM; INIT_WORK(&sq->recover_work, mlx5e_tx_err_cqe_work); if (!MLX5_CAP_ETH(mdev, wqe_vlan_insert)) set_bit(MLX5E_SQ_STATE_VLAN_NEED_L2_INLINE, &sq->state); if (MLX5_IPSEC_DEV(c->priv->mdev)) set_bit(MLX5E_SQ_STATE_IPSEC, &sq->state); -#ifdef CONFIG_MLX5_EN_TLS - if (mlx5_accel_is_tls_device(c->priv->mdev)) { + if (mlx5_accel_is_tls_device(c->priv->mdev)) set_bit(MLX5E_SQ_STATE_TLS, &sq->state); - sq->stop_room += MLX5E_SQ_TLS_ROOM + - mlx5e_ktls_dumps_num_wqebbs(sq, MAX_SKB_FRAGS, - TLS_MAX_PAYLOAD_SIZE); - } -#endif + err = mlx5e_calc_sq_stop_room(sq, params->log_sq_size); + if (err) + return err; param->wq.db_numa_node = cpu_to_node(c->cpu); err = mlx5_wq_cyc_create(mdev, ¶m->wq, sqc_wq, wq, &sq->wq_ctrl); @@ -1364,13 +1362,12 @@ static void mlx5e_deactivate_txqsq(struct mlx5e_txqsq *sq) /* last doorbell out, godspeed .. */ if (mlx5e_wqc_has_room_for(wq, sq->cc, sq->pc, 1)) { u16 pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - struct mlx5e_tx_wqe_info *wi; struct mlx5e_tx_wqe *nop; - wi = &sq->db.wqe_info[pi]; + sq->db.wqe_info[pi] = (struct mlx5e_tx_wqe_info) { + .num_wqebbs = 1, + }; - memset(wi, 0, sizeof(*wi)); - wi->num_wqebbs = 1; nop = mlx5e_post_nop(wq, sq->sqn, &sq->pc); mlx5e_notify_hw(wq, sq->pc, sq->uar_map, &nop->ctrl); } @@ -1482,20 +1479,21 @@ int mlx5e_open_xdpsq(struct mlx5e_channel *c, struct mlx5e_params *params, /* Pre initialize fixed WQE fields */ for (i = 0; i < mlx5_wq_cyc_get_size(&sq->wq); i++) { - struct mlx5e_xdp_wqe_info *wi = &sq->db.wqe_info[i]; struct mlx5e_tx_wqe *wqe = mlx5_wq_cyc_get_wqe(&sq->wq, i); struct mlx5_wqe_ctrl_seg *cseg = &wqe->ctrl; struct mlx5_wqe_eth_seg *eseg = &wqe->eth; struct mlx5_wqe_data_seg *dseg; + sq->db.wqe_info[i] = (struct mlx5e_xdp_wqe_info) { + .num_wqebbs = 1, + .num_pkts = 1, + }; + cseg->qpn_ds = cpu_to_be32((sq->sqn << 8) | ds_cnt); eseg->inline_hdr.sz = cpu_to_be16(inline_hdr_sz); dseg = (struct mlx5_wqe_data_seg *)cseg + (ds_cnt - 1); dseg->lkey = sq->mkey_be; - - wi->num_wqebbs = 1; - wi->num_pkts = 1; } } @@ -2717,7 +2715,8 @@ void mlx5e_modify_tirs_hash(struct mlx5e_priv *priv, void *in) mlx5_core_modify_tir(mdev, priv->indir_tir[tt].tirn, in); } - if (!mlx5e_tunnel_inner_ft_supported(priv->mdev)) + /* Verify inner tirs resources allocated */ + if (!priv->inner_indir_tir[0].tirn) return; for (tt = 0; tt < MLX5E_NUM_INDIR_TIRS; tt++) { @@ -2836,11 +2835,8 @@ void mlx5e_set_netdev_mtu_boundaries(struct mlx5e_priv *priv) ETH_MAX_MTU); } -static void mlx5e_netdev_set_tcs(struct net_device *netdev) +static void mlx5e_netdev_set_tcs(struct net_device *netdev, u16 nch, u8 ntc) { - struct mlx5e_priv *priv = netdev_priv(netdev); - int nch = priv->channels.params.num_channels; - int ntc = priv->channels.params.num_tc; int tc; netdev_reset_tc(netdev); @@ -2857,15 +2853,47 @@ static void mlx5e_netdev_set_tcs(struct net_device *netdev) netdev_set_tc_queue(netdev, tc, nch, 0); } -static void mlx5e_update_netdev_queues(struct mlx5e_priv *priv, u16 count) +static int mlx5e_update_netdev_queues(struct mlx5e_priv *priv) { - int num_txqs = count * priv->channels.params.num_tc; - int num_rxqs = count * priv->profile->rq_groups; struct net_device *netdev = priv->netdev; + int num_txqs, num_rxqs, nch, ntc; + int old_num_txqs, old_ntc; + int err; + + old_num_txqs = netdev->real_num_tx_queues; + old_ntc = netdev->num_tc; + + nch = priv->channels.params.num_channels; + ntc = priv->channels.params.num_tc; + num_txqs = nch * ntc; + num_rxqs = nch * priv->profile->rq_groups; + + mlx5e_netdev_set_tcs(netdev, nch, ntc); + + err = netif_set_real_num_tx_queues(netdev, num_txqs); + if (err) { + netdev_warn(netdev, "netif_set_real_num_tx_queues failed, %d\n", err); + goto err_tcs; + } + err = netif_set_real_num_rx_queues(netdev, num_rxqs); + if (err) { + netdev_warn(netdev, "netif_set_real_num_rx_queues failed, %d\n", err); + goto err_txqs; + } + + return 0; + +err_txqs: + /* netif_set_real_num_rx_queues could fail only when nch increased. Only + * one of nch and ntc is changed in this function. That means, the call + * to netif_set_real_num_tx_queues below should not fail, because it + * decreases the number of TX queues. + */ + WARN_ON_ONCE(netif_set_real_num_tx_queues(netdev, old_num_txqs)); - mlx5e_netdev_set_tcs(netdev); - netif_set_real_num_tx_queues(netdev, num_txqs); - netif_set_real_num_rx_queues(netdev, num_rxqs); +err_tcs: + mlx5e_netdev_set_tcs(netdev, old_num_txqs / old_ntc, old_ntc); + return err; } static void mlx5e_set_default_xps_cpumasks(struct mlx5e_priv *priv, @@ -2892,8 +2920,12 @@ static void mlx5e_set_default_xps_cpumasks(struct mlx5e_priv *priv, int mlx5e_num_channels_changed(struct mlx5e_priv *priv) { u16 count = priv->channels.params.num_channels; + int err; + + err = mlx5e_update_netdev_queues(priv); + if (err) + return err; - mlx5e_update_netdev_queues(priv, count); mlx5e_set_default_xps_cpumasks(priv, &priv->channels.params); if (!netif_is_rxfh_configured(priv->netdev)) @@ -3405,14 +3437,15 @@ out: return err; } -void mlx5e_destroy_indirect_tirs(struct mlx5e_priv *priv, bool inner_ttc) +void mlx5e_destroy_indirect_tirs(struct mlx5e_priv *priv) { int i; for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) mlx5e_destroy_tir(priv->mdev, &priv->indir_tir[i]); - if (!inner_ttc || !mlx5e_tunnel_inner_ft_supported(priv->mdev)) + /* Verify inner tirs resources allocated */ + if (!priv->inner_indir_tir[0].tirn) return; for (i = 0; i < MLX5E_NUM_INDIR_TIRS; i++) @@ -3489,41 +3522,6 @@ out: return err; } -#ifdef CONFIG_MLX5_ESWITCH -static int mlx5e_setup_tc_cls_flower(struct mlx5e_priv *priv, - struct flow_cls_offload *cls_flower, - unsigned long flags) -{ - switch (cls_flower->command) { - case FLOW_CLS_REPLACE: - return mlx5e_configure_flower(priv->netdev, priv, cls_flower, - flags); - case FLOW_CLS_DESTROY: - return mlx5e_delete_flower(priv->netdev, priv, cls_flower, - flags); - case FLOW_CLS_STATS: - return mlx5e_stats_flower(priv->netdev, priv, cls_flower, - flags); - default: - return -EOPNOTSUPP; - } -} - -static int mlx5e_setup_tc_block_cb(enum tc_setup_type type, void *type_data, - void *cb_priv) -{ - unsigned long flags = MLX5_TC_FLAG(INGRESS) | MLX5_TC_FLAG(NIC_OFFLOAD); - struct mlx5e_priv *priv = cb_priv; - - switch (type) { - case TC_SETUP_CLSFLOWER: - return mlx5e_setup_tc_cls_flower(priv, type_data, flags); - default: - return -EOPNOTSUPP; - } -} -#endif - static LIST_HEAD(mlx5e_block_cb_list); static int mlx5e_setup_tc(struct net_device *dev, enum tc_setup_type type, @@ -3532,7 +3530,6 @@ static int mlx5e_setup_tc(struct net_device *dev, enum tc_setup_type type, struct mlx5e_priv *priv = netdev_priv(dev); switch (type) { -#ifdef CONFIG_MLX5_ESWITCH case TC_SETUP_BLOCK: { struct flow_block_offload *f = type_data; @@ -3542,7 +3539,6 @@ static int mlx5e_setup_tc(struct net_device *dev, enum tc_setup_type type, mlx5e_setup_tc_block_cb, priv, priv, true); } -#endif case TC_SETUP_QDISC_MQPRIO: return mlx5e_setup_tc_mqprio(priv, type_data); default: @@ -3715,7 +3711,7 @@ static int set_feature_cvlan_filter(struct net_device *netdev, bool enable) return 0; } -#ifdef CONFIG_MLX5_ESWITCH +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) static int set_feature_tc_num_filters(struct net_device *netdev, bool enable) { struct mlx5e_priv *priv = netdev_priv(netdev); @@ -3826,7 +3822,7 @@ int mlx5e_set_features(struct net_device *netdev, netdev_features_t features) err |= MLX5E_HANDLE_FEATURE(NETIF_F_LRO, set_feature_lro); err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_VLAN_CTAG_FILTER, set_feature_cvlan_filter); -#ifdef CONFIG_MLX5_ESWITCH +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_TC, set_feature_tc_num_filters); #endif err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXALL, set_feature_rx_all); @@ -4711,7 +4707,7 @@ static u8 mlx5_to_net_dim_cq_period_mode(u8 cq_period_mode) DIM_CQ_PERIOD_MODE_START_FROM_EQE; } -void mlx5e_set_tx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) +void mlx5e_reset_tx_moderation(struct mlx5e_params *params, u8 cq_period_mode) { if (params->tx_dim_enabled) { u8 dim_period_mode = mlx5_to_net_dim_cq_period_mode(cq_period_mode); @@ -4720,13 +4716,9 @@ void mlx5e_set_tx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) } else { params->tx_cq_moderation = mlx5e_get_def_tx_moderation(cq_period_mode); } - - MLX5E_SET_PFLAG(params, MLX5E_PFLAG_TX_CQE_BASED_MODER, - params->tx_cq_moderation.cq_period_mode == - MLX5_CQ_PERIOD_MODE_START_FROM_CQE); } -void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) +void mlx5e_reset_rx_moderation(struct mlx5e_params *params, u8 cq_period_mode) { if (params->rx_dim_enabled) { u8 dim_period_mode = mlx5_to_net_dim_cq_period_mode(cq_period_mode); @@ -4735,7 +4727,19 @@ void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) } else { params->rx_cq_moderation = mlx5e_get_def_rx_moderation(cq_period_mode); } +} + +void mlx5e_set_tx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) +{ + mlx5e_reset_tx_moderation(params, cq_period_mode); + MLX5E_SET_PFLAG(params, MLX5E_PFLAG_TX_CQE_BASED_MODER, + params->tx_cq_moderation.cq_period_mode == + MLX5_CQ_PERIOD_MODE_START_FROM_CQE); +} +void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) +{ + mlx5e_reset_rx_moderation(params, cq_period_mode); MLX5E_SET_PFLAG(params, MLX5E_PFLAG_RX_CQE_BASED_MODER, params->rx_cq_moderation.cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE); @@ -4876,10 +4880,7 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) netdev->netdev_ops = &mlx5e_netdev_ops; -#ifdef CONFIG_MLX5_CORE_EN_DCB - if (MLX5_CAP_GEN(mdev, vport_group_manager) && MLX5_CAP_GEN(mdev, qos)) - netdev->dcbnl_ops = &mlx5e_dcbnl_ops; -#endif + mlx5e_dcbnl_build_netdev(netdev); netdev->watchdog_timeo = 15 * HZ; @@ -5131,7 +5132,7 @@ err_destroy_xsk_rqts: err_destroy_direct_tirs: mlx5e_destroy_direct_tirs(priv, priv->direct_tir); err_destroy_indirect_tirs: - mlx5e_destroy_indirect_tirs(priv, true); + mlx5e_destroy_indirect_tirs(priv); err_destroy_direct_rqts: mlx5e_destroy_direct_rqts(priv, priv->direct_tir); err_destroy_indirect_rqts: @@ -5150,7 +5151,7 @@ static void mlx5e_cleanup_nic_rx(struct mlx5e_priv *priv) mlx5e_destroy_direct_tirs(priv, priv->xsk_tir); mlx5e_destroy_direct_rqts(priv, priv->xsk_tir); mlx5e_destroy_direct_tirs(priv, priv->direct_tir); - mlx5e_destroy_indirect_tirs(priv, true); + mlx5e_destroy_indirect_tirs(priv); mlx5e_destroy_direct_rqts(priv, priv->direct_tir); mlx5e_destroy_rqt(priv, &priv->indir_rqt); mlx5e_close_drop_rq(&priv->drop_rq); @@ -5167,9 +5168,7 @@ static int mlx5e_init_nic_tx(struct mlx5e_priv *priv) return err; } -#ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_dcbnl_initialize(priv); -#endif return 0; } @@ -5196,9 +5195,7 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv) mlx5e_hv_vhca_stats_create(priv); if (netdev->reg_state != NETREG_REGISTERED) return; -#ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_dcbnl_init_app(priv); -#endif queue_work(priv->wq, &priv->set_rx_mode_work); @@ -5213,10 +5210,8 @@ static void mlx5e_nic_disable(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; -#ifdef CONFIG_MLX5_CORE_EN_DCB if (priv->netdev->reg_state == NETREG_REGISTERED) mlx5e_dcbnl_delete_app(priv); -#endif rtnl_lock(); if (netif_running(priv->netdev)) @@ -5236,7 +5231,7 @@ static void mlx5e_nic_disable(struct mlx5e_priv *priv) int mlx5e_update_nic_rx(struct mlx5e_priv *priv) { - return mlx5e_refresh_tirs(priv, false); + return mlx5e_refresh_tirs(priv, false, false); } static const struct mlx5e_profile mlx5e_nic_profile = { @@ -5371,9 +5366,11 @@ int mlx5e_attach_netdev(struct mlx5e_priv *priv) */ if (take_rtnl) rtnl_lock(); - mlx5e_num_channels_changed(priv); + err = mlx5e_num_channels_changed(priv); if (take_rtnl) rtnl_unlock(); + if (err) + goto out; err = profile->init_tx(priv); if (err) @@ -5511,9 +5508,7 @@ static void *mlx5e_add(struct mlx5_core_dev *mdev) mlx5e_devlink_port_type_eth_set(priv); -#ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_dcbnl_init_app(priv); -#endif return priv; err_devlink_port_unregister: @@ -5536,9 +5531,7 @@ static void mlx5e_remove(struct mlx5_core_dev *mdev, void *vpriv) } #endif priv = vpriv; -#ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_dcbnl_delete_app(priv); -#endif unregister_netdev(priv->netdev); mlx5e_devlink_port_unregister(priv); mlx5e_detach(mdev, vpriv); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index f372e94948fd..006807e04eda 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -35,7 +35,6 @@ #include <net/switchdev.h> #include <net/pkt_cls.h> #include <net/act_api.h> -#include <net/netevent.h> #include <net/arp.h> #include <net/devlink.h> #include <net/ipv6_stubs.h> @@ -45,9 +44,9 @@ #include "en.h" #include "en_rep.h" #include "en_tc.h" -#include "en/tc_tun.h" +#include "en/rep/tc.h" +#include "en/rep/neigh.h" #include "fs_core.h" -#include "lib/port_tun.h" #include "lib/mlx5.h" #define CREATE_TRACE_POINTS #include "diag/en_rep_tracepoint.h" @@ -58,16 +57,6 @@ static const char mlx5e_rep_driver_name[] = "mlx5e_rep"; -struct mlx5e_rep_indr_block_priv { - struct net_device *netdev; - struct mlx5e_rep_priv *rpriv; - - struct list_head list; -}; - -static void mlx5e_rep_indr_unregister_block(struct mlx5e_rep_priv *rpriv, - struct net_device *netdev); - static void mlx5e_rep_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo) { @@ -485,706 +474,6 @@ void mlx5e_remove_sqs_fwd_rules(struct mlx5e_priv *priv) mlx5e_sqs2vport_stop(esw, rep); } -static unsigned long mlx5e_rep_ipv6_interval(void) -{ - if (IS_ENABLED(CONFIG_IPV6) && ipv6_stub->nd_tbl) - return NEIGH_VAR(&ipv6_stub->nd_tbl->parms, DELAY_PROBE_TIME); - - return ~0UL; -} - -static void mlx5e_rep_neigh_update_init_interval(struct mlx5e_rep_priv *rpriv) -{ - unsigned long ipv4_interval = NEIGH_VAR(&arp_tbl.parms, DELAY_PROBE_TIME); - unsigned long ipv6_interval = mlx5e_rep_ipv6_interval(); - struct net_device *netdev = rpriv->netdev; - struct mlx5e_priv *priv = netdev_priv(netdev); - - rpriv->neigh_update.min_interval = min_t(unsigned long, ipv6_interval, ipv4_interval); - mlx5_fc_update_sampling_interval(priv->mdev, rpriv->neigh_update.min_interval); -} - -void mlx5e_rep_queue_neigh_stats_work(struct mlx5e_priv *priv) -{ - struct mlx5e_rep_priv *rpriv = priv->ppriv; - struct mlx5e_neigh_update_table *neigh_update = &rpriv->neigh_update; - - mlx5_fc_queue_stats_work(priv->mdev, - &neigh_update->neigh_stats_work, - neigh_update->min_interval); -} - -static bool mlx5e_rep_neigh_entry_hold(struct mlx5e_neigh_hash_entry *nhe) -{ - return refcount_inc_not_zero(&nhe->refcnt); -} - -static void mlx5e_rep_neigh_entry_remove(struct mlx5e_neigh_hash_entry *nhe); - -static void mlx5e_rep_neigh_entry_release(struct mlx5e_neigh_hash_entry *nhe) -{ - if (refcount_dec_and_test(&nhe->refcnt)) { - mlx5e_rep_neigh_entry_remove(nhe); - kfree_rcu(nhe, rcu); - } -} - -static struct mlx5e_neigh_hash_entry * -mlx5e_get_next_nhe(struct mlx5e_rep_priv *rpriv, - struct mlx5e_neigh_hash_entry *nhe) -{ - struct mlx5e_neigh_hash_entry *next = NULL; - - rcu_read_lock(); - - for (next = nhe ? - list_next_or_null_rcu(&rpriv->neigh_update.neigh_list, - &nhe->neigh_list, - struct mlx5e_neigh_hash_entry, - neigh_list) : - list_first_or_null_rcu(&rpriv->neigh_update.neigh_list, - struct mlx5e_neigh_hash_entry, - neigh_list); - next; - next = list_next_or_null_rcu(&rpriv->neigh_update.neigh_list, - &next->neigh_list, - struct mlx5e_neigh_hash_entry, - neigh_list)) - if (mlx5e_rep_neigh_entry_hold(next)) - break; - - rcu_read_unlock(); - - if (nhe) - mlx5e_rep_neigh_entry_release(nhe); - - return next; -} - -static void mlx5e_rep_neigh_stats_work(struct work_struct *work) -{ - struct mlx5e_rep_priv *rpriv = container_of(work, struct mlx5e_rep_priv, - neigh_update.neigh_stats_work.work); - struct net_device *netdev = rpriv->netdev; - struct mlx5e_priv *priv = netdev_priv(netdev); - struct mlx5e_neigh_hash_entry *nhe = NULL; - - rtnl_lock(); - if (!list_empty(&rpriv->neigh_update.neigh_list)) - mlx5e_rep_queue_neigh_stats_work(priv); - - while ((nhe = mlx5e_get_next_nhe(rpriv, nhe)) != NULL) - mlx5e_tc_update_neigh_used_value(nhe); - - rtnl_unlock(); -} - -static void mlx5e_rep_update_flows(struct mlx5e_priv *priv, - struct mlx5e_encap_entry *e, - bool neigh_connected, - unsigned char ha[ETH_ALEN]) -{ - struct ethhdr *eth = (struct ethhdr *)e->encap_header; - struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - bool encap_connected; - LIST_HEAD(flow_list); - - ASSERT_RTNL(); - - /* wait for encap to be fully initialized */ - wait_for_completion(&e->res_ready); - - mutex_lock(&esw->offloads.encap_tbl_lock); - encap_connected = !!(e->flags & MLX5_ENCAP_ENTRY_VALID); - if (e->compl_result < 0 || (encap_connected == neigh_connected && - ether_addr_equal(e->h_dest, ha))) - goto unlock; - - mlx5e_take_all_encap_flows(e, &flow_list); - - if ((e->flags & MLX5_ENCAP_ENTRY_VALID) && - (!neigh_connected || !ether_addr_equal(e->h_dest, ha))) - mlx5e_tc_encap_flows_del(priv, e, &flow_list); - - if (neigh_connected && !(e->flags & MLX5_ENCAP_ENTRY_VALID)) { - ether_addr_copy(e->h_dest, ha); - ether_addr_copy(eth->h_dest, ha); - /* Update the encap source mac, in case that we delete - * the flows when encap source mac changed. - */ - ether_addr_copy(eth->h_source, e->route_dev->dev_addr); - - mlx5e_tc_encap_flows_add(priv, e, &flow_list); - } -unlock: - mutex_unlock(&esw->offloads.encap_tbl_lock); - mlx5e_put_encap_flow_list(priv, &flow_list); -} - -static void mlx5e_rep_neigh_update(struct work_struct *work) -{ - struct mlx5e_neigh_hash_entry *nhe = - container_of(work, struct mlx5e_neigh_hash_entry, neigh_update_work); - struct neighbour *n = nhe->n; - struct mlx5e_encap_entry *e; - unsigned char ha[ETH_ALEN]; - struct mlx5e_priv *priv; - bool neigh_connected; - u8 nud_state, dead; - - rtnl_lock(); - - /* If these parameters are changed after we release the lock, - * we'll receive another event letting us know about it. - * We use this lock to avoid inconsistency between the neigh validity - * and it's hw address. - */ - read_lock_bh(&n->lock); - memcpy(ha, n->ha, ETH_ALEN); - nud_state = n->nud_state; - dead = n->dead; - read_unlock_bh(&n->lock); - - neigh_connected = (nud_state & NUD_VALID) && !dead; - - trace_mlx5e_rep_neigh_update(nhe, ha, neigh_connected); - - list_for_each_entry(e, &nhe->encap_list, encap_list) { - if (!mlx5e_encap_take(e)) - continue; - - priv = netdev_priv(e->out_dev); - mlx5e_rep_update_flows(priv, e, neigh_connected, ha); - mlx5e_encap_put(priv, e); - } - mlx5e_rep_neigh_entry_release(nhe); - rtnl_unlock(); - neigh_release(n); -} - -static struct mlx5e_rep_indr_block_priv * -mlx5e_rep_indr_block_priv_lookup(struct mlx5e_rep_priv *rpriv, - struct net_device *netdev) -{ - struct mlx5e_rep_indr_block_priv *cb_priv; - - /* All callback list access should be protected by RTNL. */ - ASSERT_RTNL(); - - list_for_each_entry(cb_priv, - &rpriv->uplink_priv.tc_indr_block_priv_list, - list) - if (cb_priv->netdev == netdev) - return cb_priv; - - return NULL; -} - -static void mlx5e_rep_indr_clean_block_privs(struct mlx5e_rep_priv *rpriv) -{ - struct mlx5e_rep_indr_block_priv *cb_priv, *temp; - struct list_head *head = &rpriv->uplink_priv.tc_indr_block_priv_list; - - list_for_each_entry_safe(cb_priv, temp, head, list) { - mlx5e_rep_indr_unregister_block(rpriv, cb_priv->netdev); - kfree(cb_priv); - } -} - -static int -mlx5e_rep_indr_offload(struct net_device *netdev, - struct flow_cls_offload *flower, - struct mlx5e_rep_indr_block_priv *indr_priv, - unsigned long flags) -{ - struct mlx5e_priv *priv = netdev_priv(indr_priv->rpriv->netdev); - int err = 0; - - switch (flower->command) { - case FLOW_CLS_REPLACE: - err = mlx5e_configure_flower(netdev, priv, flower, flags); - break; - case FLOW_CLS_DESTROY: - err = mlx5e_delete_flower(netdev, priv, flower, flags); - break; - case FLOW_CLS_STATS: - err = mlx5e_stats_flower(netdev, priv, flower, flags); - break; - default: - err = -EOPNOTSUPP; - } - - return err; -} - -static int mlx5e_rep_indr_setup_tc_cb(enum tc_setup_type type, - void *type_data, void *indr_priv) -{ - unsigned long flags = MLX5_TC_FLAG(EGRESS) | MLX5_TC_FLAG(ESW_OFFLOAD); - struct mlx5e_rep_indr_block_priv *priv = indr_priv; - - switch (type) { - case TC_SETUP_CLSFLOWER: - return mlx5e_rep_indr_offload(priv->netdev, type_data, priv, - flags); - default: - return -EOPNOTSUPP; - } -} - -static int mlx5e_rep_indr_setup_ft_cb(enum tc_setup_type type, - void *type_data, void *indr_priv) -{ - struct mlx5e_rep_indr_block_priv *priv = indr_priv; - struct flow_cls_offload *f = type_data; - struct flow_cls_offload tmp; - struct mlx5e_priv *mpriv; - struct mlx5_eswitch *esw; - unsigned long flags; - int err; - - mpriv = netdev_priv(priv->rpriv->netdev); - esw = mpriv->mdev->priv.eswitch; - - flags = MLX5_TC_FLAG(EGRESS) | - MLX5_TC_FLAG(ESW_OFFLOAD) | - MLX5_TC_FLAG(FT_OFFLOAD); - - switch (type) { - case TC_SETUP_CLSFLOWER: - memcpy(&tmp, f, sizeof(*f)); - - /* Re-use tc offload path by moving the ft flow to the - * reserved ft chain. - * - * FT offload can use prio range [0, INT_MAX], so we normalize - * it to range [1, mlx5_esw_chains_get_prio_range(esw)] - * as with tc, where prio 0 isn't supported. - * - * We only support chain 0 of FT offload. - */ - if (!mlx5_esw_chains_prios_supported(esw) || - tmp.common.prio >= mlx5_esw_chains_get_prio_range(esw) || - tmp.common.chain_index) - return -EOPNOTSUPP; - - tmp.common.chain_index = mlx5_esw_chains_get_ft_chain(esw); - tmp.common.prio++; - err = mlx5e_rep_indr_offload(priv->netdev, &tmp, priv, flags); - memcpy(&f->stats, &tmp.stats, sizeof(f->stats)); - return err; - default: - return -EOPNOTSUPP; - } -} - -static void mlx5e_rep_indr_block_unbind(void *cb_priv) -{ - struct mlx5e_rep_indr_block_priv *indr_priv = cb_priv; - - list_del(&indr_priv->list); - kfree(indr_priv); -} - -static LIST_HEAD(mlx5e_block_cb_list); - -static int -mlx5e_rep_indr_setup_block(struct net_device *netdev, - struct mlx5e_rep_priv *rpriv, - struct flow_block_offload *f, - flow_setup_cb_t *setup_cb) -{ - struct mlx5e_rep_indr_block_priv *indr_priv; - struct flow_block_cb *block_cb; - - if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS) - return -EOPNOTSUPP; - - f->unlocked_driver_cb = true; - f->driver_block_list = &mlx5e_block_cb_list; - - switch (f->command) { - case FLOW_BLOCK_BIND: - indr_priv = mlx5e_rep_indr_block_priv_lookup(rpriv, netdev); - if (indr_priv) - return -EEXIST; - - indr_priv = kmalloc(sizeof(*indr_priv), GFP_KERNEL); - if (!indr_priv) - return -ENOMEM; - - indr_priv->netdev = netdev; - indr_priv->rpriv = rpriv; - list_add(&indr_priv->list, - &rpriv->uplink_priv.tc_indr_block_priv_list); - - block_cb = flow_block_cb_alloc(setup_cb, indr_priv, indr_priv, - mlx5e_rep_indr_block_unbind); - if (IS_ERR(block_cb)) { - list_del(&indr_priv->list); - kfree(indr_priv); - return PTR_ERR(block_cb); - } - flow_block_cb_add(block_cb, f); - list_add_tail(&block_cb->driver_list, &mlx5e_block_cb_list); - - return 0; - case FLOW_BLOCK_UNBIND: - indr_priv = mlx5e_rep_indr_block_priv_lookup(rpriv, netdev); - if (!indr_priv) - return -ENOENT; - - block_cb = flow_block_cb_lookup(f->block, setup_cb, indr_priv); - if (!block_cb) - return -ENOENT; - - flow_block_cb_remove(block_cb, f); - list_del(&block_cb->driver_list); - return 0; - default: - return -EOPNOTSUPP; - } - return 0; -} - -static -int mlx5e_rep_indr_setup_cb(struct net_device *netdev, void *cb_priv, - enum tc_setup_type type, void *type_data) -{ - switch (type) { - case TC_SETUP_BLOCK: - return mlx5e_rep_indr_setup_block(netdev, cb_priv, type_data, - mlx5e_rep_indr_setup_tc_cb); - case TC_SETUP_FT: - return mlx5e_rep_indr_setup_block(netdev, cb_priv, type_data, - mlx5e_rep_indr_setup_ft_cb); - default: - return -EOPNOTSUPP; - } -} - -static int mlx5e_rep_indr_register_block(struct mlx5e_rep_priv *rpriv, - struct net_device *netdev) -{ - int err; - - err = __flow_indr_block_cb_register(netdev, rpriv, - mlx5e_rep_indr_setup_cb, - rpriv); - if (err) { - struct mlx5e_priv *priv = netdev_priv(rpriv->netdev); - - mlx5_core_err(priv->mdev, "Failed to register remote block notifier for %s err=%d\n", - netdev_name(netdev), err); - } - return err; -} - -static void mlx5e_rep_indr_unregister_block(struct mlx5e_rep_priv *rpriv, - struct net_device *netdev) -{ - __flow_indr_block_cb_unregister(netdev, mlx5e_rep_indr_setup_cb, - rpriv); -} - -static int mlx5e_nic_rep_netdevice_event(struct notifier_block *nb, - unsigned long event, void *ptr) -{ - struct mlx5e_rep_priv *rpriv = container_of(nb, struct mlx5e_rep_priv, - uplink_priv.netdevice_nb); - struct mlx5e_priv *priv = netdev_priv(rpriv->netdev); - struct net_device *netdev = netdev_notifier_info_to_dev(ptr); - - if (!mlx5e_tc_tun_device_to_offload(priv, netdev) && - !(is_vlan_dev(netdev) && vlan_dev_real_dev(netdev) == rpriv->netdev)) - return NOTIFY_OK; - - switch (event) { - case NETDEV_REGISTER: - mlx5e_rep_indr_register_block(rpriv, netdev); - break; - case NETDEV_UNREGISTER: - mlx5e_rep_indr_unregister_block(rpriv, netdev); - break; - } - return NOTIFY_OK; -} - -static void -mlx5e_rep_queue_neigh_update_work(struct mlx5e_priv *priv, - struct mlx5e_neigh_hash_entry *nhe, - struct neighbour *n) -{ - /* Take a reference to ensure the neighbour and mlx5 encap - * entry won't be destructed until we drop the reference in - * delayed work. - */ - neigh_hold(n); - - /* This assignment is valid as long as the the neigh reference - * is taken - */ - nhe->n = n; - - if (!queue_work(priv->wq, &nhe->neigh_update_work)) { - mlx5e_rep_neigh_entry_release(nhe); - neigh_release(n); - } -} - -static struct mlx5e_neigh_hash_entry * -mlx5e_rep_neigh_entry_lookup(struct mlx5e_priv *priv, - struct mlx5e_neigh *m_neigh); - -static int mlx5e_rep_netevent_event(struct notifier_block *nb, - unsigned long event, void *ptr) -{ - struct mlx5e_rep_priv *rpriv = container_of(nb, struct mlx5e_rep_priv, - neigh_update.netevent_nb); - struct mlx5e_neigh_update_table *neigh_update = &rpriv->neigh_update; - struct net_device *netdev = rpriv->netdev; - struct mlx5e_priv *priv = netdev_priv(netdev); - struct mlx5e_neigh_hash_entry *nhe = NULL; - struct mlx5e_neigh m_neigh = {}; - struct neigh_parms *p; - struct neighbour *n; - bool found = false; - - switch (event) { - case NETEVENT_NEIGH_UPDATE: - n = ptr; -#if IS_ENABLED(CONFIG_IPV6) - if (n->tbl != ipv6_stub->nd_tbl && n->tbl != &arp_tbl) -#else - if (n->tbl != &arp_tbl) -#endif - return NOTIFY_DONE; - - m_neigh.dev = n->dev; - m_neigh.family = n->ops->family; - memcpy(&m_neigh.dst_ip, n->primary_key, n->tbl->key_len); - - rcu_read_lock(); - nhe = mlx5e_rep_neigh_entry_lookup(priv, &m_neigh); - rcu_read_unlock(); - if (!nhe) - return NOTIFY_DONE; - - mlx5e_rep_queue_neigh_update_work(priv, nhe, n); - break; - - case NETEVENT_DELAY_PROBE_TIME_UPDATE: - p = ptr; - - /* We check the device is present since we don't care about - * changes in the default table, we only care about changes - * done per device delay prob time parameter. - */ -#if IS_ENABLED(CONFIG_IPV6) - if (!p->dev || (p->tbl != ipv6_stub->nd_tbl && p->tbl != &arp_tbl)) -#else - if (!p->dev || p->tbl != &arp_tbl) -#endif - return NOTIFY_DONE; - - rcu_read_lock(); - list_for_each_entry_rcu(nhe, &neigh_update->neigh_list, - neigh_list) { - if (p->dev == nhe->m_neigh.dev) { - found = true; - break; - } - } - rcu_read_unlock(); - if (!found) - return NOTIFY_DONE; - - neigh_update->min_interval = min_t(unsigned long, - NEIGH_VAR(p, DELAY_PROBE_TIME), - neigh_update->min_interval); - mlx5_fc_update_sampling_interval(priv->mdev, - neigh_update->min_interval); - break; - } - return NOTIFY_DONE; -} - -static const struct rhashtable_params mlx5e_neigh_ht_params = { - .head_offset = offsetof(struct mlx5e_neigh_hash_entry, rhash_node), - .key_offset = offsetof(struct mlx5e_neigh_hash_entry, m_neigh), - .key_len = sizeof(struct mlx5e_neigh), - .automatic_shrinking = true, -}; - -static int mlx5e_rep_neigh_init(struct mlx5e_rep_priv *rpriv) -{ - struct mlx5e_neigh_update_table *neigh_update = &rpriv->neigh_update; - int err; - - err = rhashtable_init(&neigh_update->neigh_ht, &mlx5e_neigh_ht_params); - if (err) - return err; - - INIT_LIST_HEAD(&neigh_update->neigh_list); - mutex_init(&neigh_update->encap_lock); - INIT_DELAYED_WORK(&neigh_update->neigh_stats_work, - mlx5e_rep_neigh_stats_work); - mlx5e_rep_neigh_update_init_interval(rpriv); - - rpriv->neigh_update.netevent_nb.notifier_call = mlx5e_rep_netevent_event; - err = register_netevent_notifier(&rpriv->neigh_update.netevent_nb); - if (err) - goto out_err; - return 0; - -out_err: - rhashtable_destroy(&neigh_update->neigh_ht); - return err; -} - -static void mlx5e_rep_neigh_cleanup(struct mlx5e_rep_priv *rpriv) -{ - struct mlx5e_neigh_update_table *neigh_update = &rpriv->neigh_update; - struct mlx5e_priv *priv = netdev_priv(rpriv->netdev); - - unregister_netevent_notifier(&neigh_update->netevent_nb); - - flush_workqueue(priv->wq); /* flush neigh update works */ - - cancel_delayed_work_sync(&rpriv->neigh_update.neigh_stats_work); - - mutex_destroy(&neigh_update->encap_lock); - rhashtable_destroy(&neigh_update->neigh_ht); -} - -static int mlx5e_rep_neigh_entry_insert(struct mlx5e_priv *priv, - struct mlx5e_neigh_hash_entry *nhe) -{ - struct mlx5e_rep_priv *rpriv = priv->ppriv; - int err; - - err = rhashtable_insert_fast(&rpriv->neigh_update.neigh_ht, - &nhe->rhash_node, - mlx5e_neigh_ht_params); - if (err) - return err; - - list_add_rcu(&nhe->neigh_list, &rpriv->neigh_update.neigh_list); - - return err; -} - -static void mlx5e_rep_neigh_entry_remove(struct mlx5e_neigh_hash_entry *nhe) -{ - struct mlx5e_rep_priv *rpriv = nhe->priv->ppriv; - - mutex_lock(&rpriv->neigh_update.encap_lock); - - list_del_rcu(&nhe->neigh_list); - - rhashtable_remove_fast(&rpriv->neigh_update.neigh_ht, - &nhe->rhash_node, - mlx5e_neigh_ht_params); - mutex_unlock(&rpriv->neigh_update.encap_lock); -} - -/* This function must only be called under the representor's encap_lock or - * inside rcu read lock section. - */ -static struct mlx5e_neigh_hash_entry * -mlx5e_rep_neigh_entry_lookup(struct mlx5e_priv *priv, - struct mlx5e_neigh *m_neigh) -{ - struct mlx5e_rep_priv *rpriv = priv->ppriv; - struct mlx5e_neigh_update_table *neigh_update = &rpriv->neigh_update; - struct mlx5e_neigh_hash_entry *nhe; - - nhe = rhashtable_lookup_fast(&neigh_update->neigh_ht, m_neigh, - mlx5e_neigh_ht_params); - return nhe && mlx5e_rep_neigh_entry_hold(nhe) ? nhe : NULL; -} - -static int mlx5e_rep_neigh_entry_create(struct mlx5e_priv *priv, - struct mlx5e_encap_entry *e, - struct mlx5e_neigh_hash_entry **nhe) -{ - int err; - - *nhe = kzalloc(sizeof(**nhe), GFP_KERNEL); - if (!*nhe) - return -ENOMEM; - - (*nhe)->priv = priv; - memcpy(&(*nhe)->m_neigh, &e->m_neigh, sizeof(e->m_neigh)); - INIT_WORK(&(*nhe)->neigh_update_work, mlx5e_rep_neigh_update); - spin_lock_init(&(*nhe)->encap_list_lock); - INIT_LIST_HEAD(&(*nhe)->encap_list); - refcount_set(&(*nhe)->refcnt, 1); - - err = mlx5e_rep_neigh_entry_insert(priv, *nhe); - if (err) - goto out_free; - return 0; - -out_free: - kfree(*nhe); - return err; -} - -int mlx5e_rep_encap_entry_attach(struct mlx5e_priv *priv, - struct mlx5e_encap_entry *e) -{ - struct mlx5e_rep_priv *rpriv = priv->ppriv; - struct mlx5_rep_uplink_priv *uplink_priv = &rpriv->uplink_priv; - struct mlx5_tun_entropy *tun_entropy = &uplink_priv->tun_entropy; - struct mlx5e_neigh_hash_entry *nhe; - int err; - - err = mlx5_tun_entropy_refcount_inc(tun_entropy, e->reformat_type); - if (err) - return err; - - mutex_lock(&rpriv->neigh_update.encap_lock); - nhe = mlx5e_rep_neigh_entry_lookup(priv, &e->m_neigh); - if (!nhe) { - err = mlx5e_rep_neigh_entry_create(priv, e, &nhe); - if (err) { - mutex_unlock(&rpriv->neigh_update.encap_lock); - mlx5_tun_entropy_refcount_dec(tun_entropy, - e->reformat_type); - return err; - } - } - - e->nhe = nhe; - spin_lock(&nhe->encap_list_lock); - list_add_rcu(&e->encap_list, &nhe->encap_list); - spin_unlock(&nhe->encap_list_lock); - - mutex_unlock(&rpriv->neigh_update.encap_lock); - - return 0; -} - -void mlx5e_rep_encap_entry_detach(struct mlx5e_priv *priv, - struct mlx5e_encap_entry *e) -{ - struct mlx5e_rep_priv *rpriv = priv->ppriv; - struct mlx5_rep_uplink_priv *uplink_priv = &rpriv->uplink_priv; - struct mlx5_tun_entropy *tun_entropy = &uplink_priv->tun_entropy; - - if (!e->nhe) - return; - - spin_lock(&e->nhe->encap_list_lock); - list_del_rcu(&e->encap_list); - spin_unlock(&e->nhe->encap_list_lock); - - mlx5e_rep_neigh_entry_release(e->nhe); - e->nhe = NULL; - mlx5_tun_entropy_refcount_dec(tun_entropy, e->reformat_type); -} - static int mlx5e_rep_open(struct net_device *dev) { struct mlx5e_priv *priv = netdev_priv(dev); @@ -1225,129 +514,6 @@ static int mlx5e_rep_close(struct net_device *dev) return ret; } -static int -mlx5e_rep_setup_tc_cls_flower(struct mlx5e_priv *priv, - struct flow_cls_offload *cls_flower, int flags) -{ - switch (cls_flower->command) { - case FLOW_CLS_REPLACE: - return mlx5e_configure_flower(priv->netdev, priv, cls_flower, - flags); - case FLOW_CLS_DESTROY: - return mlx5e_delete_flower(priv->netdev, priv, cls_flower, - flags); - case FLOW_CLS_STATS: - return mlx5e_stats_flower(priv->netdev, priv, cls_flower, - flags); - default: - return -EOPNOTSUPP; - } -} - -static -int mlx5e_rep_setup_tc_cls_matchall(struct mlx5e_priv *priv, - struct tc_cls_matchall_offload *ma) -{ - switch (ma->command) { - case TC_CLSMATCHALL_REPLACE: - return mlx5e_tc_configure_matchall(priv, ma); - case TC_CLSMATCHALL_DESTROY: - return mlx5e_tc_delete_matchall(priv, ma); - case TC_CLSMATCHALL_STATS: - mlx5e_tc_stats_matchall(priv, ma); - return 0; - default: - return -EOPNOTSUPP; - } -} - -static int mlx5e_rep_setup_tc_cb(enum tc_setup_type type, void *type_data, - void *cb_priv) -{ - unsigned long flags = MLX5_TC_FLAG(INGRESS) | MLX5_TC_FLAG(ESW_OFFLOAD); - struct mlx5e_priv *priv = cb_priv; - - switch (type) { - case TC_SETUP_CLSFLOWER: - return mlx5e_rep_setup_tc_cls_flower(priv, type_data, flags); - case TC_SETUP_CLSMATCHALL: - return mlx5e_rep_setup_tc_cls_matchall(priv, type_data); - default: - return -EOPNOTSUPP; - } -} - -static int mlx5e_rep_setup_ft_cb(enum tc_setup_type type, void *type_data, - void *cb_priv) -{ - struct flow_cls_offload tmp, *f = type_data; - struct mlx5e_priv *priv = cb_priv; - struct mlx5_eswitch *esw; - unsigned long flags; - int err; - - flags = MLX5_TC_FLAG(INGRESS) | - MLX5_TC_FLAG(ESW_OFFLOAD) | - MLX5_TC_FLAG(FT_OFFLOAD); - esw = priv->mdev->priv.eswitch; - - switch (type) { - case TC_SETUP_CLSFLOWER: - memcpy(&tmp, f, sizeof(*f)); - - if (!mlx5_esw_chains_prios_supported(esw)) - return -EOPNOTSUPP; - - /* Re-use tc offload path by moving the ft flow to the - * reserved ft chain. - * - * FT offload can use prio range [0, INT_MAX], so we normalize - * it to range [1, mlx5_esw_chains_get_prio_range(esw)] - * as with tc, where prio 0 isn't supported. - * - * We only support chain 0 of FT offload. - */ - if (tmp.common.prio >= mlx5_esw_chains_get_prio_range(esw)) - return -EOPNOTSUPP; - if (tmp.common.chain_index != 0) - return -EOPNOTSUPP; - - tmp.common.chain_index = mlx5_esw_chains_get_ft_chain(esw); - tmp.common.prio++; - err = mlx5e_rep_setup_tc_cls_flower(priv, &tmp, flags); - memcpy(&f->stats, &tmp.stats, sizeof(f->stats)); - return err; - default: - return -EOPNOTSUPP; - } -} - -static LIST_HEAD(mlx5e_rep_block_tc_cb_list); -static LIST_HEAD(mlx5e_rep_block_ft_cb_list); -static int mlx5e_rep_setup_tc(struct net_device *dev, enum tc_setup_type type, - void *type_data) -{ - struct mlx5e_priv *priv = netdev_priv(dev); - struct flow_block_offload *f = type_data; - - f->unlocked_driver_cb = true; - - switch (type) { - case TC_SETUP_BLOCK: - return flow_block_cb_setup_simple(type_data, - &mlx5e_rep_block_tc_cb_list, - mlx5e_rep_setup_tc_cb, - priv, priv, true); - case TC_SETUP_FT: - return flow_block_cb_setup_simple(type_data, - &mlx5e_rep_block_ft_cb_list, - mlx5e_rep_setup_ft_cb, - priv, priv, true); - default: - return -EOPNOTSUPP; - } -} - bool mlx5e_is_uplink_rep(struct mlx5e_priv *priv) { struct mlx5e_rep_priv *rpriv = priv->ppriv; @@ -1484,13 +650,9 @@ bool mlx5e_eswitch_uplink_rep(struct net_device *netdev) return netdev->netdev_ops == &mlx5e_netdev_ops_uplink_rep; } -bool mlx5e_eswitch_rep(struct net_device *netdev) +bool mlx5e_eswitch_vf_rep(struct net_device *netdev) { - if (netdev->netdev_ops == &mlx5e_netdev_ops_rep || - netdev->netdev_ops == &mlx5e_netdev_ops_uplink_rep) - return true; - - return false; + return netdev->netdev_ops == &mlx5e_netdev_ops_rep; } static void mlx5e_build_rep_params(struct net_device *netdev) @@ -1544,10 +706,7 @@ static void mlx5e_build_rep_netdev(struct net_device *netdev) /* we want a persistent mac for the uplink rep */ mlx5_query_mac_address(mdev, netdev->dev_addr); netdev->ethtool_ops = &mlx5e_uplink_rep_ethtool_ops; -#ifdef CONFIG_MLX5_CORE_EN_DCB - if (MLX5_CAP_GEN(mdev, qos)) - netdev->dcbnl_ops = &mlx5e_dcbnl_ops; -#endif + mlx5e_dcbnl_build_rep_netdev(netdev); } else { netdev->netdev_ops = &mlx5e_netdev_ops_rep; eth_hw_addr_random(netdev); @@ -1695,6 +854,24 @@ static int mlx5e_create_rep_vport_rx_rule(struct mlx5e_priv *priv) return 0; } +static void rep_vport_rx_rule_destroy(struct mlx5e_priv *priv) +{ + struct mlx5e_rep_priv *rpriv = priv->ppriv; + + if (!rpriv->vport_rx_rule) + return; + + mlx5_del_flow_rules(rpriv->vport_rx_rule); + rpriv->vport_rx_rule = NULL; +} + +int mlx5e_rep_bond_update(struct mlx5e_priv *priv, bool cleanup) +{ + rep_vport_rx_rule_destroy(priv); + + return cleanup ? 0 : mlx5e_create_rep_vport_rx_rule(priv); +} + static int mlx5e_init_rep_rx(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; @@ -1747,7 +924,7 @@ err_destroy_ttc_table: err_destroy_direct_tirs: mlx5e_destroy_direct_tirs(priv, priv->direct_tir); err_destroy_indirect_tirs: - mlx5e_destroy_indirect_tirs(priv, false); + mlx5e_destroy_indirect_tirs(priv); err_destroy_direct_rqts: mlx5e_destroy_direct_rqts(priv, priv->direct_tir); err_destroy_indirect_rqts: @@ -1759,13 +936,11 @@ err_close_drop_rq: static void mlx5e_cleanup_rep_rx(struct mlx5e_priv *priv) { - struct mlx5e_rep_priv *rpriv = priv->ppriv; - - mlx5_del_flow_rules(rpriv->vport_rx_rule); + rep_vport_rx_rule_destroy(priv); mlx5e_destroy_rep_root_ft(priv); mlx5e_destroy_ttc_table(priv, &priv->fs.ttc); mlx5e_destroy_direct_tirs(priv, priv->direct_tir); - mlx5e_destroy_indirect_tirs(priv, false); + mlx5e_destroy_indirect_tirs(priv); mlx5e_destroy_direct_rqts(priv, priv->direct_tir); mlx5e_destroy_rqt(priv, &priv->indir_rqt); mlx5e_close_drop_rq(&priv->drop_rq); @@ -1794,31 +969,25 @@ static int mlx5e_init_uplink_rep_tx(struct mlx5e_rep_priv *rpriv) priv = netdev_priv(netdev); uplink_priv = &rpriv->uplink_priv; - mutex_init(&uplink_priv->unready_flows_lock); - INIT_LIST_HEAD(&uplink_priv->unready_flows); - - /* init shared tc flow table */ - err = mlx5e_tc_esw_init(&uplink_priv->tc_ht); + err = mlx5e_rep_tc_init(rpriv); if (err) return err; mlx5_init_port_tun_entropy(&uplink_priv->tun_entropy, priv->mdev); - /* init indirect block notifications */ - INIT_LIST_HEAD(&uplink_priv->tc_indr_block_priv_list); - uplink_priv->netdevice_nb.notifier_call = mlx5e_nic_rep_netdevice_event; - err = register_netdevice_notifier_dev_net(rpriv->netdev, - &uplink_priv->netdevice_nb, - &uplink_priv->netdevice_nn); + mlx5e_rep_bond_init(rpriv); + err = mlx5e_rep_tc_netdevice_event_register(rpriv); if (err) { - mlx5_core_err(priv->mdev, "Failed to register netdev notifier\n"); - goto tc_esw_cleanup; + mlx5_core_err(priv->mdev, "Failed to register netdev notifier, err: %d\n", + err); + goto err_event_reg; } return 0; -tc_esw_cleanup: - mlx5e_tc_esw_cleanup(&uplink_priv->tc_ht); +err_event_reg: + mlx5e_rep_bond_cleanup(rpriv); + mlx5e_rep_tc_cleanup(rpriv); return err; } @@ -1848,17 +1017,9 @@ destroy_tises: static void mlx5e_cleanup_uplink_rep_tx(struct mlx5e_rep_priv *rpriv) { - struct mlx5_rep_uplink_priv *uplink_priv = &rpriv->uplink_priv; - - /* clean indirect TC block notifications */ - unregister_netdevice_notifier_dev_net(rpriv->netdev, - &uplink_priv->netdevice_nb, - &uplink_priv->netdevice_nn); - mlx5e_rep_indr_clean_block_privs(rpriv); - - /* delete shared tc flow table */ - mlx5e_tc_esw_cleanup(&rpriv->uplink_priv.tc_ht); - mutex_destroy(&rpriv->uplink_priv.unready_flows_lock); + mlx5e_rep_tc_netdevice_event_unregister(rpriv); + mlx5e_rep_bond_cleanup(rpriv); + mlx5e_rep_tc_cleanup(rpriv); } static void mlx5e_cleanup_rep_tx(struct mlx5e_priv *priv) @@ -1900,13 +1061,8 @@ static int uplink_rep_async_event(struct notifier_block *nb, unsigned long event return NOTIFY_OK; } - if (event == MLX5_DEV_EVENT_PORT_AFFINITY) { - struct mlx5e_rep_priv *rpriv = priv->ppriv; - - queue_work(priv->wq, &rpriv->uplink_priv.reoffload_flows_work); - - return NOTIFY_OK; - } + if (event == MLX5_DEV_EVENT_PORT_AFFINITY) + return mlx5e_rep_tc_event_port_affinity(priv); return NOTIFY_DONE; } @@ -1915,7 +1071,6 @@ static void mlx5e_uplink_rep_enable(struct mlx5e_priv *priv) { struct net_device *netdev = priv->netdev; struct mlx5_core_dev *mdev = priv->mdev; - struct mlx5e_rep_priv *rpriv = priv->ppriv; u16 max_mtu; netdev->min_mtu = ETH_MIN_MTU; @@ -1923,28 +1078,22 @@ static void mlx5e_uplink_rep_enable(struct mlx5e_priv *priv) netdev->max_mtu = MLX5E_HW2SW_MTU(&priv->channels.params, max_mtu); mlx5e_set_dev_port_mtu(priv); - INIT_WORK(&rpriv->uplink_priv.reoffload_flows_work, - mlx5e_tc_reoffload_flows_work); + mlx5e_rep_tc_enable(priv); mlx5_lag_add(mdev, netdev); priv->events_nb.notifier_call = uplink_rep_async_event; mlx5_notifier_register(mdev, &priv->events_nb); -#ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_dcbnl_initialize(priv); mlx5e_dcbnl_init_app(priv); -#endif } static void mlx5e_uplink_rep_disable(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; - struct mlx5e_rep_priv *rpriv = priv->ppriv; -#ifdef CONFIG_MLX5_CORE_EN_DCB mlx5e_dcbnl_delete_app(priv); -#endif mlx5_notifier_unregister(mdev, &priv->events_nb); - cancel_work_sync(&rpriv->uplink_priv.reoffload_flows_work); + mlx5e_rep_tc_disable(priv); mlx5_lag_remove(mdev); } @@ -2051,26 +1200,22 @@ static int register_devlink_port(struct mlx5_core_dev *dev, return 0; mlx5e_rep_get_port_parent_id(rpriv->netdev, &ppid); + dl_port_index = vport_to_devlink_port_index(dev, rep->vport); pfnum = PCI_FUNC(dev->pdev->devfn); - if (rep->vport == MLX5_VPORT_UPLINK) { + if (rep->vport == MLX5_VPORT_UPLINK) devlink_port_attrs_set(&rpriv->dl_port, DEVLINK_PORT_FLAVOUR_PHYSICAL, pfnum, false, 0, &ppid.id[0], ppid.id_len); - dl_port_index = vport_to_devlink_port_index(dev, rep->vport); - } else if (rep->vport == MLX5_VPORT_PF) { + else if (rep->vport == MLX5_VPORT_PF) devlink_port_attrs_pci_pf_set(&rpriv->dl_port, &ppid.id[0], ppid.id_len, pfnum); - dl_port_index = rep->vport; - } else if (mlx5_eswitch_is_vf_vport(dev->priv.eswitch, - rpriv->rep->vport)) { + else if (mlx5_eswitch_is_vf_vport(dev->priv.eswitch, rpriv->rep->vport)) devlink_port_attrs_pci_vf_set(&rpriv->dl_port, &ppid.id[0], ppid.id_len, pfnum, rep->vport - 1); - dl_port_index = vport_to_devlink_port_index(dev, rep->vport); - } return devlink_port_register(devlink, &rpriv->dl_port, dl_port_index); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h index 6a2337900420..1d5669801484 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.h @@ -56,6 +56,7 @@ struct mlx5e_neigh_update_table { }; struct mlx5_tc_ct_priv; +struct mlx5e_rep_bond; struct mlx5_rep_uplink_priv { /* Filters DB - instantiated by the uplink representor and shared by * the uplink's VFs @@ -68,13 +69,8 @@ struct mlx5_rep_uplink_priv { * tc_indr_block_cb_priv_list is used to lookup indirect callback * private data * - * netdevice_nb is the netdev events notifier - used to register - * tunnel devices for block events - * */ struct list_head tc_indr_block_priv_list; - struct notifier_block netdevice_nb; - struct netdev_net_notifier netdevice_nn; struct mlx5_tun_entropy tun_entropy; @@ -89,6 +85,9 @@ struct mlx5_rep_uplink_priv { struct mapping_ctx *tunnel_enc_opts_mapping; struct mlx5_tc_ct_priv *ct_priv; + + /* support eswitch vports bonding */ + struct mlx5e_rep_bond *bond; }; struct mlx5e_rep_priv { @@ -158,6 +157,22 @@ struct mlx5e_neigh_hash_entry { enum { /* set when the encap entry is successfully offloaded into HW */ MLX5_ENCAP_ENTRY_VALID = BIT(0), + MLX5_REFORMAT_DECAP = BIT(1), +}; + +struct mlx5e_decap_key { + struct ethhdr key; +}; + +struct mlx5e_decap_entry { + struct mlx5e_decap_key key; + struct list_head flows; + struct hlist_node hlist; + refcount_t refcnt; + struct completion res_ready; + int compl_result; + struct mlx5_pkt_reformat *pkt_reformat; + struct rcu_head rcu; }; struct mlx5e_encap_entry { @@ -195,6 +210,15 @@ struct mlx5e_rep_sq { void mlx5e_rep_register_vport_reps(struct mlx5_core_dev *mdev); void mlx5e_rep_unregister_vport_reps(struct mlx5_core_dev *mdev); +int mlx5e_rep_bond_init(struct mlx5e_rep_priv *rpriv); +void mlx5e_rep_bond_cleanup(struct mlx5e_rep_priv *rpriv); +int mlx5e_rep_bond_enslave(struct mlx5_eswitch *esw, struct net_device *netdev, + struct net_device *lag_dev); +void mlx5e_rep_bond_unslave(struct mlx5_eswitch *esw, + const struct net_device *netdev, + const struct net_device *lag_dev); +int mlx5e_rep_bond_update(struct mlx5e_priv *priv, bool cleanup); + bool mlx5e_is_uplink_rep(struct mlx5e_priv *priv); int mlx5e_add_sqs_fwd_rules(struct mlx5e_priv *priv); void mlx5e_remove_sqs_fwd_rules(struct mlx5e_priv *priv); @@ -203,15 +227,15 @@ void mlx5e_handle_rx_cqe_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe); void mlx5e_handle_rx_cqe_mpwrq_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe); -int mlx5e_rep_encap_entry_attach(struct mlx5e_priv *priv, - struct mlx5e_encap_entry *e); -void mlx5e_rep_encap_entry_detach(struct mlx5e_priv *priv, - struct mlx5e_encap_entry *e); - void mlx5e_rep_queue_neigh_stats_work(struct mlx5e_priv *priv); -bool mlx5e_eswitch_rep(struct net_device *netdev); +bool mlx5e_eswitch_vf_rep(struct net_device *netdev); bool mlx5e_eswitch_uplink_rep(struct net_device *netdev); +static inline bool mlx5e_eswitch_rep(struct net_device *netdev) +{ + return mlx5e_eswitch_vf_rep(netdev) || + mlx5e_eswitch_uplink_rep(netdev); +} #else /* CONFIG_MLX5_ESWITCH */ static inline bool mlx5e_is_uplink_rep(struct mlx5e_priv *priv) { return false; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index e2beb89c1832..dbb1c6323967 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -42,6 +42,7 @@ #include "en_tc.h" #include "eswitch.h" #include "en_rep.h" +#include "en/rep/tc.h" #include "ipoib/ipoib.h" #include "en_accel/ipsec_rxtx.h" #include "en_accel/tls_rxtx.h" @@ -300,7 +301,7 @@ static inline void mlx5e_page_release(struct mlx5e_rq *rq, * put into the Reuse Ring, because there is no way to return * the page to the userspace when the interface goes down. */ - mlx5e_xsk_page_release(rq, dma_info); + xsk_buff_free(dma_info->xsk); else mlx5e_page_release_dynamic(rq, dma_info, recycle); } @@ -385,7 +386,11 @@ static int mlx5e_alloc_rx_wqes(struct mlx5e_rq *rq, u16 ix, u8 wqe_bulk) if (rq->umem) { int pages_desired = wqe_bulk << rq->wqe.info.log_num_frags; - if (unlikely(!mlx5e_xsk_pages_enough_umem(rq, pages_desired))) + /* Check in advance that we have enough frames, instead of + * allocating one-by-one, failing and moving frames to the + * Reuse Ring. + */ + if (unlikely(!xsk_buff_can_alloc(rq->umem, pages_desired))) return -ENOMEM; } @@ -468,22 +473,6 @@ static void mlx5e_post_rx_mpwqe(struct mlx5e_rq *rq, u8 n) mlx5_wq_ll_update_db_record(wq); } -static inline void mlx5e_fill_icosq_frag_edge(struct mlx5e_icosq *sq, - struct mlx5_wq_cyc *wq, - u16 pi, u16 nnops) -{ - struct mlx5e_sq_wqe_info *edge_wi, *wi = &sq->db.ico_wqe[pi]; - - edge_wi = wi + nnops; - - /* fill sq frag edge with nops to avoid wqe wrapping two pages */ - for (; wi < edge_wi; wi++) { - wi->opcode = MLX5_OPCODE_NOP; - wi->num_wqebbs = 1; - mlx5e_post_nop(wq, sq->sqn, &sq->pc); - } -} - static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix) { struct mlx5e_mpw_info *wi = &rq->mpwqe.info[ix]; @@ -492,23 +481,20 @@ static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix) struct mlx5_wq_cyc *wq = &sq->wq; struct mlx5e_umr_wqe *umr_wqe; u16 xlt_offset = ix << (MLX5E_LOG_ALIGNED_MPWQE_PPW - 1); - u16 pi, contig_wqebbs_room; + u16 pi; int err; int i; + /* Check in advance that we have enough frames, instead of allocating + * one-by-one, failing and moving frames to the Reuse Ring. + */ if (rq->umem && - unlikely(!mlx5e_xsk_pages_enough_umem(rq, MLX5_MPWRQ_PAGES_PER_WQE))) { + unlikely(!xsk_buff_can_alloc(rq->umem, MLX5_MPWRQ_PAGES_PER_WQE))) { err = -ENOMEM; goto err; } - pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - contig_wqebbs_room = mlx5_wq_cyc_get_contig_wqebbs(wq, pi); - if (unlikely(contig_wqebbs_room < MLX5E_UMR_WQEBBS)) { - mlx5e_fill_icosq_frag_edge(sq, wq, pi, contig_wqebbs_room); - pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - } - + pi = mlx5e_icosq_get_next_pi(sq, MLX5E_UMR_WQEBBS); umr_wqe = mlx5_wq_cyc_get_wqe(wq, pi); memcpy(umr_wqe, &rq->mpwqe.umr_wqe, offsetof(struct mlx5e_umr_wqe, inline_mtts)); @@ -527,9 +513,12 @@ static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix) MLX5_OPCODE_UMR); umr_wqe->uctrl.xlt_offset = cpu_to_be16(xlt_offset); - sq->db.ico_wqe[pi].opcode = MLX5_OPCODE_UMR; - sq->db.ico_wqe[pi].num_wqebbs = MLX5E_UMR_WQEBBS; - sq->db.ico_wqe[pi].umr.rq = rq; + sq->db.wqe_info[pi] = (struct mlx5e_icosq_wqe_info) { + .wqe_type = MLX5E_ICOSQ_WQE_UMR_RX, + .num_wqebbs = MLX5E_UMR_WQEBBS, + .umr.rq = rq, + }; + sq->pc += MLX5E_UMR_WQEBBS; sq->doorbell_cseg = &umr_wqe->ctrl; @@ -618,33 +607,38 @@ int mlx5e_poll_ico_cq(struct mlx5e_cq *cq) wqe_counter = be16_to_cpu(cqe->wqe_counter); do { - struct mlx5e_sq_wqe_info *wi; + struct mlx5e_icosq_wqe_info *wi; u16 ci; last_wqe = (sqcc == wqe_counter); ci = mlx5_wq_cyc_ctr2ix(&sq->wq, sqcc); - wi = &sq->db.ico_wqe[ci]; + wi = &sq->db.wqe_info[ci]; sqcc += wi->num_wqebbs; if (last_wqe && unlikely(get_cqe_opcode(cqe) != MLX5_CQE_REQ)) { netdev_WARN_ONCE(cq->channel->netdev, "Bad OP in ICOSQ CQE: 0x%x\n", get_cqe_opcode(cqe)); + mlx5e_dump_error_cqe(&sq->cq, sq->sqn, + (struct mlx5_err_cqe *)cqe); if (!test_and_set_bit(MLX5E_SQ_STATE_RECOVERING, &sq->state)) queue_work(cq->channel->priv->wq, &sq->recover_work); break; } - if (likely(wi->opcode == MLX5_OPCODE_UMR)) + switch (wi->wqe_type) { + case MLX5E_ICOSQ_WQE_UMR_RX: wi->umr.rq->mpwqe.umr_completed++; - else if (unlikely(wi->opcode != MLX5_OPCODE_NOP)) + break; + case MLX5E_ICOSQ_WQE_NOP: + break; + default: netdev_WARN_ONCE(cq->channel->netdev, - "Bad OPCODE in ICOSQ WQE info: 0x%x\n", - wi->opcode); - + "Bad WQE type in ICOSQ WQE info: 0x%x\n", + wi->wqe_type); + } } while (!last_wqe); - } while ((++i < MLX5E_TX_CQ_POLL_BUDGET) && (cqe = mlx5_cqwq_get_cqe(&cq->wq))); sq->cc = sqcc; @@ -1058,12 +1052,24 @@ struct sk_buff *mlx5e_build_linear_skb(struct mlx5e_rq *rq, void *va, return skb; } +static void mlx5e_fill_xdp_buff(struct mlx5e_rq *rq, void *va, u16 headroom, + u32 len, struct xdp_buff *xdp) +{ + xdp->data_hard_start = va; + xdp->data = va + headroom; + xdp_set_data_meta_invalid(xdp); + xdp->data_end = xdp->data + len; + xdp->rxq = &rq->xdp_rxq; + xdp->frame_sz = rq->buff.frame0_sz; +} + struct sk_buff * mlx5e_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe, struct mlx5e_wqe_frag_info *wi, u32 cqe_bcnt) { struct mlx5e_dma_info *di = wi->di; u16 rx_headroom = rq->buff.headroom; + struct xdp_buff xdp; struct sk_buff *skb; void *va, *data; bool consumed; @@ -1079,11 +1085,14 @@ mlx5e_skb_from_cqe_linear(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe, prefetch(data); rcu_read_lock(); - consumed = mlx5e_xdp_handle(rq, di, va, &rx_headroom, &cqe_bcnt, false); + mlx5e_fill_xdp_buff(rq, va, rx_headroom, cqe_bcnt, &xdp); + consumed = mlx5e_xdp_handle(rq, di, &cqe_bcnt, &xdp); rcu_read_unlock(); if (consumed) return NULL; /* page/packet was consumed by XDP */ + rx_headroom = xdp.data - xdp.data_hard_start; + frag_size = MLX5_SKB_FRAG_SZ(rx_headroom + cqe_bcnt); skb = mlx5e_build_linear_skb(rq, va, frag_size, rx_headroom, cqe_bcnt); if (unlikely(!skb)) return NULL; @@ -1229,12 +1238,12 @@ void mlx5e_handle_rx_cqe_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe) if (rep->vlan && skb_vlan_tag_present(skb)) skb_vlan_pop(skb); - if (!mlx5e_tc_rep_update_skb(cqe, skb, &tc_priv)) + if (!mlx5e_rep_tc_update_skb(cqe, skb, &tc_priv)) goto free_wqe; napi_gro_receive(rq->cq.napi, skb); - mlx5_tc_rep_post_napi_receive(&tc_priv); + mlx5_rep_tc_post_napi_receive(&tc_priv); free_wqe: mlx5e_free_rx_wqe(rq, wi, true); @@ -1285,12 +1294,12 @@ void mlx5e_handle_rx_cqe_mpwrq_rep(struct mlx5e_rq *rq, mlx5e_complete_rx_cqe(rq, cqe, cqe_bcnt, skb); - if (!mlx5e_tc_rep_update_skb(cqe, skb, &tc_priv)) + if (!mlx5e_rep_tc_update_skb(cqe, skb, &tc_priv)) goto mpwrq_cqe_out; napi_gro_receive(rq->cq.napi, skb); - mlx5_tc_rep_post_napi_receive(&tc_priv); + mlx5_rep_tc_post_napi_receive(&tc_priv); mpwrq_cqe_out: if (likely(wi->consumed_strides < rq->mpwqe.num_strides)) @@ -1356,6 +1365,7 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, struct mlx5e_dma_info *di = &wi->umr.dma_info[page_idx]; u16 rx_headroom = rq->buff.headroom; u32 cqe_bcnt32 = cqe_bcnt; + struct xdp_buff xdp; struct sk_buff *skb; void *va, *data; u32 frag_size; @@ -1377,7 +1387,8 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, prefetch(data); rcu_read_lock(); - consumed = mlx5e_xdp_handle(rq, di, va, &rx_headroom, &cqe_bcnt32, false); + mlx5e_fill_xdp_buff(rq, va, rx_headroom, cqe_bcnt32, &xdp); + consumed = mlx5e_xdp_handle(rq, di, &cqe_bcnt32, &xdp); rcu_read_unlock(); if (consumed) { if (__test_and_clear_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags)) @@ -1385,6 +1396,8 @@ mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, return NULL; /* page/packet was consumed by XDP */ } + rx_headroom = xdp.data - xdp.data_hard_start; + frag_size = MLX5_SKB_FRAG_SZ(rx_headroom + cqe_bcnt32); skb = mlx5e_build_linear_skb(rq, va, frag_size, rx_headroom, cqe_bcnt32); if (unlikely(!skb)) return NULL; @@ -1501,6 +1514,7 @@ out: #ifdef CONFIG_MLX5_CORE_IPOIB +#define MLX5_IB_GRH_SGID_OFFSET 8 #define MLX5_IB_GRH_DGID_OFFSET 24 #define MLX5_GID_SIZE 16 @@ -1514,6 +1528,7 @@ static inline void mlx5i_complete_rx_cqe(struct mlx5e_rq *rq, struct net_device *netdev; struct mlx5e_priv *priv; char *pseudo_header; + u32 flags_rqpn; u32 qpn; u8 *dgid; u8 g; @@ -1535,7 +1550,8 @@ static inline void mlx5i_complete_rx_cqe(struct mlx5e_rq *rq, tstamp = &priv->tstamp; stats = &priv->channel_stats[rq->ix].rq; - g = (be32_to_cpu(cqe->flags_rqpn) >> 28) & 3; + flags_rqpn = be32_to_cpu(cqe->flags_rqpn); + g = (flags_rqpn >> 28) & 3; dgid = skb->data + MLX5_IB_GRH_DGID_OFFSET; if ((!g) || dgid[0] != 0xff) skb->pkt_type = PACKET_HOST; @@ -1544,9 +1560,15 @@ static inline void mlx5i_complete_rx_cqe(struct mlx5e_rq *rq, else skb->pkt_type = PACKET_MULTICAST; - /* TODO: IB/ipoib: Allow mcast packets from other VFs - * 68996a6e760e5c74654723eeb57bf65628ae87f4 + /* Drop packets that this interface sent, ie multicast packets + * that the HCA has replicated. */ + if (g && (qpn == (flags_rqpn & 0xffffff)) && + (memcmp(netdev->dev_addr + 4, skb->data + MLX5_IB_GRH_SGID_OFFSET, + MLX5_GID_SIZE) == 0)) { + skb->dev = NULL; + return; + } skb_pull(skb, MLX5_IB_GRH_BYTES); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c index bbff8d8ded76..46790216ce86 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_selftest.c @@ -234,7 +234,7 @@ static int mlx5e_test_loopback_setup(struct mlx5e_priv *priv, return err; } - err = mlx5e_refresh_tirs(priv, true); + err = mlx5e_refresh_tirs(priv, true, false); if (err) goto out; @@ -263,7 +263,7 @@ static void mlx5e_test_loopback_cleanup(struct mlx5e_priv *priv, mlx5_nic_vport_update_local_lb(priv->mdev, false); dev_remove_pack(&lbtp->pt); - mlx5e_refresh_tirs(priv, false); + mlx5e_refresh_tirs(priv, false, false); } #define MLX5E_LB_VERIFY_TIMEOUT (msecs_to_jiffies(200)) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c index e91a8b22eba6..f009fe09e99b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c @@ -32,8 +32,8 @@ #include "lib/mlx5.h" #include "en.h" -#include "en_accel/ipsec.h" #include "en_accel/tls.h" +#include "en_accel/en_accel.h" static unsigned int stats_grps_num(struct mlx5e_priv *priv) { @@ -1428,27 +1428,6 @@ static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(pme) static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(pme) { return; } -static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(ipsec) -{ - return mlx5e_ipsec_get_count(priv); -} - -static MLX5E_DECLARE_STATS_GRP_OP_FILL_STRS(ipsec) -{ - return idx + mlx5e_ipsec_get_strings(priv, - data + idx * ETH_GSTRING_LEN); -} - -static MLX5E_DECLARE_STATS_GRP_OP_FILL_STATS(ipsec) -{ - return idx + mlx5e_ipsec_get_stats(priv, data + idx); -} - -static MLX5E_DECLARE_STATS_GRP_OP_UPDATE_STATS(ipsec) -{ - mlx5e_ipsec_update_stats(priv); -} - static MLX5E_DECLARE_STATS_GRP_OP_NUM_STATS(tls) { return mlx5e_tls_get_count(priv); @@ -1718,7 +1697,6 @@ MLX5E_DEFINE_STATS_GRP(pme, 0); MLX5E_DEFINE_STATS_GRP(channels, 0); MLX5E_DEFINE_STATS_GRP(per_port_buff_congest, 0); MLX5E_DEFINE_STATS_GRP(eth_ext, 0); -static MLX5E_DEFINE_STATS_GRP(ipsec, 0); static MLX5E_DEFINE_STATS_GRP(tls, 0); /* The stats groups order is opposite to the update_stats() order calls */ @@ -1735,7 +1713,10 @@ mlx5e_stats_grp_t mlx5e_nic_stats_grps[] = { &MLX5E_STATS_GRP(pcie), &MLX5E_STATS_GRP(per_prio), &MLX5E_STATS_GRP(pme), - &MLX5E_STATS_GRP(ipsec), +#ifdef CONFIG_MLX5_EN_IPSEC + &MLX5E_STATS_GRP(ipsec_sw), + &MLX5E_STATS_GRP(ipsec_hw), +#endif &MLX5E_STATS_GRP(tls), &MLX5E_STATS_GRP(channels), &MLX5E_STATS_GRP(per_port_buff_congest), diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h index 092b39ffa32a..2b83ba990714 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h @@ -390,5 +390,7 @@ extern MLX5E_DECLARE_STATS_GRP(per_prio); extern MLX5E_DECLARE_STATS_GRP(pme); extern MLX5E_DECLARE_STATS_GRP(channels); extern MLX5E_DECLARE_STATS_GRP(per_port_buff_congest); +extern MLX5E_DECLARE_STATS_GRP(ipsec_hw); +extern MLX5E_DECLARE_STATS_GRP(ipsec_sw); #endif /* __MLX5_EN_STATS_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index b4a8a6e309e4..7fc84f58e28a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -31,6 +31,7 @@ */ #include <net/flow_dissector.h> +#include <net/flow_offload.h> #include <net/sch_generic.h> #include <net/pkt_cls.h> #include <net/tc_act/tc_gact.h> @@ -45,10 +46,15 @@ #include <net/tc_act/tc_tunnel_key.h> #include <net/tc_act/tc_pedit.h> #include <net/tc_act/tc_csum.h> +#include <net/tc_act/tc_mpls.h> #include <net/arp.h> #include <net/ipv6_stubs.h> +#include <net/bareudp.h> +#include <net/bonding.h> #include "en.h" #include "en_rep.h" +#include "en/rep/tc.h" +#include "en/rep/neigh.h" #include "en_tc.h" #include "eswitch.h" #include "esw/chains.h" @@ -89,6 +95,7 @@ enum { MLX5E_TC_FLOW_FLAG_NOT_READY = MLX5E_TC_FLOW_BASE + 5, MLX5E_TC_FLOW_FLAG_DELETED = MLX5E_TC_FLOW_BASE + 6, MLX5E_TC_FLOW_FLAG_CT = MLX5E_TC_FLOW_BASE + 7, + MLX5E_TC_FLOW_FLAG_L3_TO_L2_DECAP = MLX5E_TC_FLOW_BASE + 8, }; #define MLX5E_TC_MAX_SPLITS 1 @@ -122,6 +129,11 @@ struct mlx5e_tc_flow { u64 cookie; unsigned long flags; struct mlx5_flow_handle *rule[MLX5E_TC_MAX_SPLITS + 1]; + + /* flows sharing the same reformat object - currently mpls decap */ + struct list_head l3_to_l2_reformat; + struct mlx5e_decap_entry *decap_reformat; + /* Flow can be associated with multiple encap IDs. * The number of encaps is bounded by the number of supported * destinations. @@ -134,6 +146,7 @@ struct mlx5e_tc_flow { struct list_head hairpin; /* flows sharing the same hairpin */ struct list_head peer; /* flows with peer flow */ struct list_head unready; /* flows not ready to be offloaded (e.g due to missing route) */ + struct net_device *orig_dev; /* netdev adding flow first */ int tmp_efi_index; struct list_head tmp_list; /* temporary flow list used by neigh update */ refcount_t refcnt; @@ -153,35 +166,12 @@ struct mlx5e_tc_flow_parse_attr { struct mlx5_flow_spec spec; struct mlx5e_tc_mod_hdr_acts mod_hdr_acts; int mirred_ifindex[MLX5_MAX_FLOW_FWD_VPORTS]; + struct ethhdr eth; }; #define MLX5E_TC_TABLE_NUM_GROUPS 4 #define MLX5E_TC_TABLE_MAX_GROUP_SIZE BIT(16) -struct tunnel_match_key { - struct flow_dissector_key_control enc_control; - struct flow_dissector_key_keyid enc_key_id; - struct flow_dissector_key_ports enc_tp; - struct flow_dissector_key_ip enc_ip; - union { - struct flow_dissector_key_ipv4_addrs enc_ipv4; - struct flow_dissector_key_ipv6_addrs enc_ipv6; - }; - - int filter_ifindex; -}; - -/* Tunnel_id mapping is TUNNEL_INFO_BITS + ENC_OPTS_BITS. - * Upper TUNNEL_INFO_BITS for general tunnel info. - * Lower ENC_OPTS_BITS bits for enc_opts. - */ -#define TUNNEL_INFO_BITS 6 -#define TUNNEL_INFO_BITS_MASK GENMASK(TUNNEL_INFO_BITS - 1, 0) -#define ENC_OPTS_BITS 2 -#define ENC_OPTS_BITS_MASK GENMASK(ENC_OPTS_BITS - 1, 0) -#define TUNNEL_ID_BITS (TUNNEL_INFO_BITS + ENC_OPTS_BITS) -#define TUNNEL_ID_MASK GENMASK(TUNNEL_ID_BITS - 1, 0) - struct mlx5e_tc_attr_to_reg_mapping mlx5e_tc_attr_to_reg_mappings[] = { [CHAIN_TO_REG] = { .mfield = MLX5_ACTION_IN_FIELD_METADATA_REG_C_0, @@ -220,8 +210,8 @@ mlx5e_tc_match_to_reg_match(struct mlx5_flow_spec *spec, fmask = headers_c + soffset; fval = headers_v + soffset; - mask = cpu_to_be32(mask) >> (32 - (match_len * 8)); - data = cpu_to_be32(data) >> (32 - (match_len * 8)); + mask = (__force u32)(cpu_to_be32(mask)) >> (32 - (match_len * 8)); + data = (__force u32)(cpu_to_be32(data)) >> (32 - (match_len * 8)); memcpy(fmask, &mask, match_len); memcpy(fval, &data, match_len); @@ -1092,7 +1082,7 @@ mlx5e_tc_add_nic_flow(struct mlx5e_priv *priv, if (IS_ERR(priv->fs.tc.t)) { mutex_unlock(&priv->fs.tc.t_lock); NL_SET_ERR_MSG_MOD(extack, - "Failed to create tc offload table\n"); + "Failed to create tc offload table"); netdev_err(priv->netdev, "Failed to create tc offload table\n"); return PTR_ERR(priv->fs.tc.t); @@ -1144,6 +1134,11 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv, struct netlink_ext_ack *extack, struct net_device **encap_dev, bool *encap_valid); +static int mlx5e_attach_decap(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct netlink_ext_ack *extack); +static void mlx5e_detach_decap(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow); static struct mlx5_flow_handle * mlx5e_tc_offload_fdb_rules(struct mlx5_eswitch *esw, @@ -1319,6 +1314,12 @@ mlx5e_tc_add_fdb_flow(struct mlx5e_priv *priv, return -EOPNOTSUPP; } + if (flow_flag_test(flow, L3_TO_L2_DECAP)) { + err = mlx5e_attach_decap(priv, flow, extack); + if (err) + return err; + } + for (out_index = 0; out_index < MLX5_MAX_FLOW_FWD_VPORTS; out_index++) { int mirred_ifindex; @@ -1428,6 +1429,9 @@ static void mlx5e_tc_del_fdb_flow(struct mlx5e_priv *priv, if (attr->action & MLX5_FLOW_CONTEXT_ACTION_COUNT) mlx5_fc_destroy(attr->counter_dev, attr->counter); + + if (flow_flag_test(flow, L3_TO_L2_DECAP)) + mlx5e_detach_decap(priv, flow); } void mlx5e_tc_encap_flows_add(struct mlx5e_priv *priv, @@ -1704,6 +1708,17 @@ static void mlx5e_encap_dealloc(struct mlx5e_priv *priv, struct mlx5e_encap_entr kfree_rcu(e, rcu); } +static void mlx5e_decap_dealloc(struct mlx5e_priv *priv, + struct mlx5e_decap_entry *d) +{ + WARN_ON(!list_empty(&d->flows)); + + if (!d->compl_result) + mlx5_packet_reformat_dealloc(priv->mdev, d->pkt_reformat); + + kfree_rcu(d, rcu); +} + void mlx5e_encap_put(struct mlx5e_priv *priv, struct mlx5e_encap_entry *e) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; @@ -1716,6 +1731,18 @@ void mlx5e_encap_put(struct mlx5e_priv *priv, struct mlx5e_encap_entry *e) mlx5e_encap_dealloc(priv, e); } +static void mlx5e_decap_put(struct mlx5e_priv *priv, struct mlx5e_decap_entry *d) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + + if (!refcount_dec_and_mutex_lock(&d->refcnt, &esw->offloads.decap_tbl_lock)) + return; + hash_del_rcu(&d->hlist); + mutex_unlock(&esw->offloads.decap_tbl_lock); + + mlx5e_decap_dealloc(priv, d); +} + static void mlx5e_detach_encap(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow, int out_index) { @@ -1739,6 +1766,29 @@ static void mlx5e_detach_encap(struct mlx5e_priv *priv, mlx5e_encap_dealloc(priv, e); } +static void mlx5e_detach_decap(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5e_decap_entry *d = flow->decap_reformat; + + if (!d) + return; + + mutex_lock(&esw->offloads.decap_tbl_lock); + list_del(&flow->l3_to_l2_reformat); + flow->decap_reformat = NULL; + + if (!refcount_dec_and_test(&d->refcnt)) { + mutex_unlock(&esw->offloads.decap_tbl_lock); + return; + } + hash_del_rcu(&d->hlist); + mutex_unlock(&esw->offloads.decap_tbl_lock); + + mlx5e_decap_dealloc(priv, d); +} + static void __mlx5e_tc_del_fdb_peer_flow(struct mlx5e_tc_flow *flow) { struct mlx5_eswitch *esw = flow->priv->mdev->priv.eswitch; @@ -1823,10 +1873,8 @@ enc_opts_is_dont_care_or_full_match(struct mlx5e_priv *priv, memchr_inv(opt->opt_data, 0, opt->length * 4)) { *dont_care = false; - if (opt->opt_class != U16_MAX || - opt->type != U8_MAX || - memchr_inv(opt->opt_data, 0xFF, - opt->length * 4)) { + if (opt->opt_class != htons(U16_MAX) || + opt->type != U8_MAX) { NL_SET_ERR_MSG(extack, "Partial match of tunnel options in chain > 0 isn't supported"); netdev_warn(priv->netdev, @@ -1863,6 +1911,7 @@ static int mlx5e_get_flow_tunnel_id(struct mlx5e_priv *priv, struct mlx5_esw_flow_attr *attr = flow->esw_attr; struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts; struct flow_match_enc_opts enc_opts_match; + struct tunnel_match_enc_opts tun_enc_opts; struct mlx5_rep_uplink_priv *uplink_priv; struct mlx5e_rep_priv *uplink_rpriv; struct tunnel_match_key tunnel_key; @@ -1905,8 +1954,14 @@ static int mlx5e_get_flow_tunnel_id(struct mlx5e_priv *priv, goto err_enc_opts; if (!enc_opts_is_dont_care) { + memset(&tun_enc_opts, 0, sizeof(tun_enc_opts)); + memcpy(&tun_enc_opts.key, enc_opts_match.key, + sizeof(*enc_opts_match.key)); + memcpy(&tun_enc_opts.mask, enc_opts_match.mask, + sizeof(*enc_opts_match.mask)); + err = mapping_add(uplink_priv->tunnel_enc_opts_mapping, - enc_opts_match.key, &enc_opts_id); + &tun_enc_opts, &enc_opts_id); if (err) goto err_enc_opts; } @@ -1965,6 +2020,32 @@ u32 mlx5e_tc_get_flow_tun_id(struct mlx5e_tc_flow *flow) return flow->tunnel_id; } +void mlx5e_tc_set_ethertype(struct mlx5_core_dev *mdev, + struct flow_match_basic *match, bool outer, + void *headers_c, void *headers_v) +{ + bool ip_version_cap; + + ip_version_cap = outer ? + MLX5_CAP_FLOWTABLE_NIC_RX(mdev, + ft_field_support.outer_ip_version) : + MLX5_CAP_FLOWTABLE_NIC_RX(mdev, + ft_field_support.inner_ip_version); + + if (ip_version_cap && match->mask->n_proto == htons(0xFFFF) && + (match->key->n_proto == htons(ETH_P_IP) || + match->key->n_proto == htons(ETH_P_IPV6))) { + MLX5_SET_TO_ONES(fte_match_set_lyr_2_4, headers_c, ip_version); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_version, + match->key->n_proto == htons(ETH_P_IP) ? 4 : 6); + } else { + MLX5_SET(fte_match_set_lyr_2_4, headers_c, ethertype, + ntohs(match->mask->n_proto)); + MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, + ntohs(match->key->n_proto)); + } +} + static int parse_tunnel_attr(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow, struct mlx5_flow_spec *spec, @@ -2005,7 +2086,11 @@ static int parse_tunnel_attr(struct mlx5e_priv *priv, return err; } - flow->esw_attr->action |= MLX5_FLOW_CONTEXT_ACTION_DECAP; + /* With mpls over udp we decapsulate using packet reformat + * object + */ + if (!netif_is_bareudp(filter_dev)) + flow->esw_attr->action |= MLX5_FLOW_CONTEXT_ACTION_DECAP; } if (!needs_mapping && !sets_mapping) @@ -2068,7 +2153,7 @@ static int mlx5e_flower_parse_meta(struct net_device *filter_dev, flow_rule_match_meta(rule, &match); if (match.mask->ingress_ifindex != 0xFFFFFFFF) { NL_SET_ERR_MSG_MOD(extack, "Unsupported ingress ifindex mask"); - return -EINVAL; + return -EOPNOTSUPP; } ingress_dev = __dev_get_by_index(dev_net(filter_dev), @@ -2076,18 +2161,32 @@ static int mlx5e_flower_parse_meta(struct net_device *filter_dev, if (!ingress_dev) { NL_SET_ERR_MSG_MOD(extack, "Can't find the ingress port to match on"); - return -EINVAL; + return -ENOENT; } if (ingress_dev != filter_dev) { NL_SET_ERR_MSG_MOD(extack, "Can't match on the ingress filter port"); - return -EINVAL; + return -EOPNOTSUPP; } return 0; } +static bool skip_key_basic(struct net_device *filter_dev, + struct flow_cls_offload *f) +{ + /* When doing mpls over udp decap, the user needs to provide + * MPLS_UC as the protocol in order to be able to match on mpls + * label fields. However, the actual ethertype is IP so we want to + * avoid matching on this, otherwise we'll fail the match. + */ + if (netif_is_bareudp(filter_dev) && f->common.chain_index == 0) + return true; + + return false; +} + static int __parse_cls_flower(struct mlx5e_priv *priv, struct mlx5e_tc_flow *flow, struct mlx5_flow_spec *spec, @@ -2132,7 +2231,8 @@ static int __parse_cls_flower(struct mlx5e_priv *priv, BIT(FLOW_DISSECTOR_KEY_IP) | BIT(FLOW_DISSECTOR_KEY_CT) | BIT(FLOW_DISSECTOR_KEY_ENC_IP) | - BIT(FLOW_DISSECTOR_KEY_ENC_OPTS))) { + BIT(FLOW_DISSECTOR_KEY_ENC_OPTS) | + BIT(FLOW_DISSECTOR_KEY_MPLS))) { NL_SET_ERR_MSG_MOD(extack, "Unsupported key"); netdev_warn(priv->netdev, "Unsupported key used: 0x%x\n", dissector->used_keys); @@ -2162,14 +2262,14 @@ static int __parse_cls_flower(struct mlx5e_priv *priv, if (err) return err; - if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) { + if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC) && + !skip_key_basic(filter_dev, f)) { struct flow_match_basic match; flow_rule_match_basic(rule, &match); - MLX5_SET(fte_match_set_lyr_2_4, headers_c, ethertype, - ntohs(match.mask->n_proto)); - MLX5_SET(fte_match_set_lyr_2_4, headers_v, ethertype, - ntohs(match.key->n_proto)); + mlx5e_tc_set_ethertype(priv->mdev, &match, + match_level == outer_match_level, + headers_c, headers_v); if (match.mask->n_proto) *match_level = MLX5_MATCH_L2; @@ -2715,10 +2815,10 @@ static int offload_pedit_fields(struct mlx5e_priv *priv, continue; if (f->field_bsize == 32) { - mask_be32 = (__be32)mask; + mask_be32 = (__force __be32)(mask); mask = (__force unsigned long)cpu_to_le32(be32_to_cpu(mask_be32)); } else if (f->field_bsize == 16) { - mask_be32 = (__be32)mask; + mask_be32 = (__force __be32)(mask); mask_be16 = *(__be16 *)&mask_be32; mask = (__force unsigned long)cpu_to_le16(be16_to_cpu(mask_be16)); } @@ -2827,10 +2927,12 @@ void dealloc_mod_hdr_actions(struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts) static const struct pedit_headers zero_masks = {}; -static int parse_tc_pedit_action(struct mlx5e_priv *priv, - const struct flow_action_entry *act, int namespace, - struct pedit_headers_action *hdrs, - struct netlink_ext_ack *extack) +static int +parse_pedit_to_modify_hdr(struct mlx5e_priv *priv, + const struct flow_action_entry *act, int namespace, + struct mlx5e_tc_flow_parse_attr *parse_attr, + struct pedit_headers_action *hdrs, + struct netlink_ext_ack *extack) { u8 cmd = (act->id == FLOW_ACTION_MANGLE) ? 0 : 1; int err = -EOPNOTSUPP; @@ -2866,6 +2968,46 @@ out_err: return err; } +static int +parse_pedit_to_reformat(struct mlx5e_priv *priv, + const struct flow_action_entry *act, + struct mlx5e_tc_flow_parse_attr *parse_attr, + struct netlink_ext_ack *extack) +{ + u32 mask, val, offset; + u32 *p; + + if (act->id != FLOW_ACTION_MANGLE) + return -EOPNOTSUPP; + + if (act->mangle.htype != FLOW_ACT_MANGLE_HDR_TYPE_ETH) { + NL_SET_ERR_MSG_MOD(extack, "Only Ethernet modification is supported"); + return -EOPNOTSUPP; + } + + mask = ~act->mangle.mask; + val = act->mangle.val; + offset = act->mangle.offset; + p = (u32 *)&parse_attr->eth; + *(p + (offset >> 2)) |= (val & mask); + + return 0; +} + +static int parse_tc_pedit_action(struct mlx5e_priv *priv, + const struct flow_action_entry *act, int namespace, + struct mlx5e_tc_flow_parse_attr *parse_attr, + struct pedit_headers_action *hdrs, + struct mlx5e_tc_flow *flow, + struct netlink_ext_ack *extack) +{ + if (flow && flow_flag_test(flow, L3_TO_L2_DECAP)) + return parse_pedit_to_reformat(priv, act, parse_attr, extack); + + return parse_pedit_to_modify_hdr(priv, act, namespace, + parse_attr, hdrs, extack); +} + static int alloc_tc_pedit_action(struct mlx5e_priv *priv, int namespace, struct mlx5e_tc_flow_parse_attr *parse_attr, struct pedit_headers_action *hdrs, @@ -3003,16 +3145,19 @@ static bool modify_header_match_supported(struct mlx5_flow_spec *spec, { const struct flow_action_entry *act; bool modify_ip_header; + void *headers_c; void *headers_v; u16 ethertype; u8 ip_proto; int i, err; + headers_c = get_match_headers_criteria(actions, spec); headers_v = get_match_headers_value(actions, spec); ethertype = MLX5_GET(fte_match_set_lyr_2_4, headers_v, ethertype); /* for non-IP we only re-write MACs, so we're okay */ - if (ethertype != ETH_P_IP && ethertype != ETH_P_IPV6) + if (MLX5_GET(fte_match_set_lyr_2_4, headers_c, ip_version) == 0 && + ethertype != ETH_P_IP && ethertype != ETH_P_IPV6) goto out_ok; modify_ip_header = false; @@ -3073,6 +3218,11 @@ static bool actions_match_supported(struct mlx5e_priv *priv, return true; } +static bool same_port_devs(struct mlx5e_priv *priv, struct mlx5e_priv *peer_priv) +{ + return priv->mdev == peer_priv->mdev; +} + static bool same_hw_devs(struct mlx5e_priv *priv, struct mlx5e_priv *peer_priv) { struct mlx5_core_dev *fmdev, *pmdev; @@ -3124,7 +3274,7 @@ static int add_vlan_rewrite_action(struct mlx5e_priv *priv, int namespace, return -EOPNOTSUPP; } - err = parse_tc_pedit_action(priv, &pedit_act, namespace, hdrs, NULL); + err = parse_tc_pedit_action(priv, &pedit_act, namespace, parse_attr, hdrs, NULL, extack); *action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; return err; @@ -3190,7 +3340,7 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv, case FLOW_ACTION_MANGLE: case FLOW_ACTION_ADD: err = parse_tc_pedit_action(priv, act, MLX5_FLOW_NAMESPACE_KERNEL, - hdrs, extack); + parse_attr, hdrs, NULL, extack); if (err) return err; @@ -3284,14 +3434,24 @@ static inline int cmp_encap_info(struct encap_key *a, a->tc_tunnel->tunnel_type != b->tc_tunnel->tunnel_type; } +static inline int cmp_decap_info(struct mlx5e_decap_key *a, + struct mlx5e_decap_key *b) +{ + return memcmp(&a->key, &b->key, sizeof(b->key)); +} + static inline int hash_encap_info(struct encap_key *key) { return jhash(key->ip_tun_key, sizeof(*key->ip_tun_key), key->tc_tunnel->tunnel_type); } +static inline int hash_decap_info(struct mlx5e_decap_key *key) +{ + return jhash(&key->key, sizeof(key->key), 0); +} -static bool is_merged_eswitch_dev(struct mlx5e_priv *priv, +static bool is_merged_eswitch_vfs(struct mlx5e_priv *priv, struct net_device *peer_netdev) { struct mlx5e_priv *peer_priv; @@ -3299,18 +3459,21 @@ static bool is_merged_eswitch_dev(struct mlx5e_priv *priv, peer_priv = netdev_priv(peer_netdev); return (MLX5_CAP_ESW(priv->mdev, merged_eswitch) && - mlx5e_eswitch_rep(priv->netdev) && - mlx5e_eswitch_rep(peer_netdev) && + mlx5e_eswitch_vf_rep(priv->netdev) && + mlx5e_eswitch_vf_rep(peer_netdev) && same_hw_devs(priv, peer_priv)); } - - bool mlx5e_encap_take(struct mlx5e_encap_entry *e) { return refcount_inc_not_zero(&e->refcnt); } +static bool mlx5e_decap_take(struct mlx5e_decap_entry *e) +{ + return refcount_inc_not_zero(&e->refcnt); +} + static struct mlx5e_encap_entry * mlx5e_encap_get(struct mlx5e_priv *priv, struct encap_key *key, uintptr_t hash_key) @@ -3331,6 +3494,24 @@ mlx5e_encap_get(struct mlx5e_priv *priv, struct encap_key *key, return NULL; } +static struct mlx5e_decap_entry * +mlx5e_decap_get(struct mlx5e_priv *priv, struct mlx5e_decap_key *key, + uintptr_t hash_key) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5e_decap_key r_key; + struct mlx5e_decap_entry *e; + + hash_for_each_possible_rcu(esw->offloads.decap_tbl, e, + hlist, hash_key) { + r_key = e->key; + if (!cmp_decap_info(&r_key, key) && + mlx5e_decap_take(e)) + return e; + } + return NULL; +} + static struct ip_tunnel_info *dup_tun_info(const struct ip_tunnel_info *tun_info) { size_t tun_size = sizeof(*tun_info) + tun_info->options_len; @@ -3476,6 +3657,84 @@ out_err_init: return err; } +static int mlx5e_attach_decap(struct mlx5e_priv *priv, + struct mlx5e_tc_flow *flow, + struct netlink_ext_ack *extack) +{ + struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; + struct mlx5_esw_flow_attr *attr = flow->esw_attr; + struct mlx5e_tc_flow_parse_attr *parse_attr; + struct mlx5e_decap_entry *d; + struct mlx5e_decap_key key; + uintptr_t hash_key; + int err = 0; + + parse_attr = attr->parse_attr; + if (sizeof(parse_attr->eth) > MLX5_CAP_ESW(priv->mdev, max_encap_header_size)) { + NL_SET_ERR_MSG_MOD(extack, + "encap header larger than max supported"); + return -EOPNOTSUPP; + } + + key.key = parse_attr->eth; + hash_key = hash_decap_info(&key); + mutex_lock(&esw->offloads.decap_tbl_lock); + d = mlx5e_decap_get(priv, &key, hash_key); + if (d) { + mutex_unlock(&esw->offloads.decap_tbl_lock); + wait_for_completion(&d->res_ready); + mutex_lock(&esw->offloads.decap_tbl_lock); + if (d->compl_result) { + err = -EREMOTEIO; + goto out_free; + } + goto found; + } + + d = kzalloc(sizeof(*d), GFP_KERNEL); + if (!d) { + err = -ENOMEM; + goto out_err; + } + + d->key = key; + refcount_set(&d->refcnt, 1); + init_completion(&d->res_ready); + INIT_LIST_HEAD(&d->flows); + hash_add_rcu(esw->offloads.decap_tbl, &d->hlist, hash_key); + mutex_unlock(&esw->offloads.decap_tbl_lock); + + d->pkt_reformat = mlx5_packet_reformat_alloc(priv->mdev, + MLX5_REFORMAT_TYPE_L3_TUNNEL_TO_L2, + sizeof(parse_attr->eth), + &parse_attr->eth, + MLX5_FLOW_NAMESPACE_FDB); + if (IS_ERR(d->pkt_reformat)) { + err = PTR_ERR(d->pkt_reformat); + d->compl_result = err; + } + mutex_lock(&esw->offloads.decap_tbl_lock); + complete_all(&d->res_ready); + if (err) + goto out_free; + +found: + flow->decap_reformat = d; + attr->decap_pkt_reformat = d->pkt_reformat; + list_add(&flow->l3_to_l2_reformat, &d->flows); + mutex_unlock(&esw->offloads.decap_tbl_lock); + return 0; + +out_free: + mutex_unlock(&esw->offloads.decap_tbl_lock); + mlx5e_decap_put(priv, d); + return err; + +out_err: + mutex_unlock(&esw->offloads.decap_tbl_lock); + return err; +} + static int parse_tc_vlan_action(struct mlx5e_priv *priv, const struct flow_action_entry *act, struct mlx5_esw_flow_attr *attr, @@ -3529,6 +3788,28 @@ static int parse_tc_vlan_action(struct mlx5e_priv *priv, return 0; } +static struct net_device *get_fdb_out_dev(struct net_device *uplink_dev, + struct net_device *out_dev) +{ + struct net_device *fdb_out_dev = out_dev; + struct net_device *uplink_upper; + + rcu_read_lock(); + uplink_upper = netdev_master_upper_dev_get_rcu(uplink_dev); + if (uplink_upper && netif_is_lag_master(uplink_upper) && + uplink_upper == out_dev) { + fdb_out_dev = uplink_dev; + } else if (netif_is_lag_master(out_dev)) { + fdb_out_dev = bond_option_active_slave_get_rcu(netdev_priv(out_dev)); + if (fdb_out_dev && + (!mlx5e_eswitch_rep(fdb_out_dev) || + !netdev_port_same_parent_id(fdb_out_dev, uplink_dev))) + fdb_out_dev = NULL; + } + rcu_read_unlock(); + return fdb_out_dev; +} + static int add_vlan_push_action(struct mlx5e_priv *priv, struct mlx5_esw_flow_attr *attr, struct net_device **out_dev, @@ -3575,14 +3856,37 @@ static int add_vlan_pop_action(struct mlx5e_priv *priv, return err; } +static bool same_hw_reps(struct mlx5e_priv *priv, + struct net_device *peer_netdev) +{ + struct mlx5e_priv *peer_priv; + + peer_priv = netdev_priv(peer_netdev); + + return mlx5e_eswitch_rep(priv->netdev) && + mlx5e_eswitch_rep(peer_netdev) && + same_hw_devs(priv, peer_priv); +} + +static bool is_lag_dev(struct mlx5e_priv *priv, + struct net_device *peer_netdev) +{ + return ((mlx5_lag_is_sriov(priv->mdev) || + mlx5_lag_is_multipath(priv->mdev)) && + same_hw_reps(priv, peer_netdev)); +} + bool mlx5e_is_valid_eswitch_fwd_dev(struct mlx5e_priv *priv, struct net_device *out_dev) { - if (is_merged_eswitch_dev(priv, out_dev)) + if (is_merged_eswitch_vfs(priv, out_dev)) + return true; + + if (is_lag_dev(priv, out_dev)) return true; return mlx5e_eswitch_rep(out_dev) && - same_hw_devs(priv, netdev_priv(out_dev)); + same_port_devs(priv, netdev_priv(out_dev)); } static bool is_duplicated_output_device(struct net_device *dev, @@ -3687,7 +3991,8 @@ static int verify_uplink_forwarding(struct mlx5e_priv *priv, static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct flow_action *flow_action, struct mlx5e_tc_flow *flow, - struct netlink_ext_ack *extack) + struct netlink_ext_ack *extack, + struct net_device *filter_dev) { struct pedit_headers_action hdrs[2] = {}; struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; @@ -3701,6 +4006,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, bool encap = false, decap = false; u32 action = attr->action; int err, i, if_count = 0; + bool mpls_push = false; if (!flow_action_has_entries(flow_action)) return -EINVAL; @@ -3715,15 +4021,48 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, action |= MLX5_FLOW_CONTEXT_ACTION_DROP | MLX5_FLOW_CONTEXT_ACTION_COUNT; break; + case FLOW_ACTION_MPLS_PUSH: + if (!MLX5_CAP_ESW_FLOWTABLE_FDB(priv->mdev, + reformat_l2_to_l3_tunnel) || + act->mpls_push.proto != htons(ETH_P_MPLS_UC)) { + NL_SET_ERR_MSG_MOD(extack, + "mpls push is supported only for mpls_uc protocol"); + return -EOPNOTSUPP; + } + mpls_push = true; + break; + case FLOW_ACTION_MPLS_POP: + /* we only support mpls pop if it is the first action + * and the filter net device is bareudp. Subsequent + * actions can be pedit and the last can be mirred + * egress redirect. + */ + if (i) { + NL_SET_ERR_MSG_MOD(extack, + "mpls pop supported only as first action"); + return -EOPNOTSUPP; + } + if (!netif_is_bareudp(filter_dev)) { + NL_SET_ERR_MSG_MOD(extack, + "mpls pop supported only on bareudp devices"); + return -EOPNOTSUPP; + } + + parse_attr->eth.h_proto = act->mpls_pop.proto; + action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT; + flow_flag_set(flow, L3_TO_L2_DECAP); + break; case FLOW_ACTION_MANGLE: case FLOW_ACTION_ADD: err = parse_tc_pedit_action(priv, act, MLX5_FLOW_NAMESPACE_FDB, - hdrs, extack); + parse_attr, hdrs, flow, extack); if (err) return err; - action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; - attr->split_count = attr->out_count; + if (!flow_flag_test(flow, L3_TO_L2_DECAP)) { + action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; + attr->split_count = attr->out_count; + } break; case FLOW_ACTION_CSUM: if (csum_offload_supported(priv, action, @@ -3745,6 +4084,12 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, return -EINVAL; } + if (mpls_push && !netif_is_bareudp(out_dev)) { + NL_SET_ERR_MSG_MOD(extack, + "mpls is supported only through a bareudp device"); + return -EOPNOTSUPP; + } + if (ft_flow && out_dev == priv->netdev) { /* Ignore forward to self rules generated * by adding both mlx5 devs to the flow table @@ -3780,7 +4125,6 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, } else if (netdev_port_same_parent_id(priv->netdev, out_dev)) { struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; struct net_device *uplink_dev = mlx5_eswitch_uplink_get_proto_dev(esw, REP_ETH); - struct net_device *uplink_upper; if (is_duplicated_output_device(priv->netdev, out_dev, @@ -3792,14 +4136,9 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, ifindexes[if_count] = out_dev->ifindex; if_count++; - rcu_read_lock(); - uplink_upper = - netdev_master_upper_dev_get_rcu(uplink_dev); - if (uplink_upper && - netif_is_lag_master(uplink_upper) && - uplink_upper == out_dev) - out_dev = uplink_dev; - rcu_read_unlock(); + out_dev = get_fdb_out_dev(uplink_dev, out_dev); + if (!out_dev) + return -ENODEV; if (is_vlan_dev(out_dev)) { err = add_vlan_push_action(priv, attr, @@ -3823,10 +4162,6 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, if (!mlx5e_is_valid_eswitch_fwd_dev(priv, out_dev)) { NL_SET_ERR_MSG_MOD(extack, "devices are not on same switch HW, can't offload forwarding"); - netdev_warn(priv->netdev, - "devices %s %s not on same switch HW, can't offload forwarding\n", - priv->netdev->name, - out_dev->name); return -EOPNOTSUPP; } @@ -4075,6 +4410,7 @@ mlx5e_alloc_flow(struct mlx5e_priv *priv, int attr_size, INIT_LIST_HEAD(&flow->encaps[out_index].list); INIT_LIST_HEAD(&flow->mod_hdr); INIT_LIST_HEAD(&flow->hairpin); + INIT_LIST_HEAD(&flow->l3_to_l2_reformat); refcount_set(&flow->refcnt, 1); init_completion(&flow->init_done); @@ -4144,7 +4480,7 @@ __mlx5e_add_fdb_flow(struct mlx5e_priv *priv, if (err) goto err_free; - err = parse_tc_fdb_actions(priv, &rule->action, flow, extack); + err = parse_tc_fdb_actions(priv, &rule->action, flow, extack, filter_dev); if (err) goto err_free; @@ -4330,11 +4666,21 @@ mlx5e_tc_add_flow(struct mlx5e_priv *priv, return err; } +static bool is_flow_rule_duplicate_allowed(struct net_device *dev, + struct mlx5e_rep_priv *rpriv) +{ + /* Offloaded flow rule is allowed to duplicate on non-uplink representor + * sharing tc block with other slaves of a lag device. + */ + return netif_is_lag_port(dev) && rpriv->rep->vport != MLX5_VPORT_UPLINK; +} + int mlx5e_configure_flower(struct net_device *dev, struct mlx5e_priv *priv, struct flow_cls_offload *f, unsigned long flags) { struct netlink_ext_ack *extack = f->common.extack; struct rhashtable *tc_ht = get_tc_ht(priv, flags); + struct mlx5e_rep_priv *rpriv = priv->ppriv; struct mlx5e_tc_flow *flow; int err = 0; @@ -4342,6 +4688,12 @@ int mlx5e_configure_flower(struct net_device *dev, struct mlx5e_priv *priv, flow = rhashtable_lookup(tc_ht, &f->cookie, tc_ht_params); rcu_read_unlock(); if (flow) { + /* Same flow rule offloaded to non-uplink representor sharing tc block, + * just return 0. + */ + if (is_flow_rule_duplicate_allowed(dev, rpriv) && flow->orig_dev != dev) + goto out; + NL_SET_ERR_MSG_MOD(extack, "flow cookie already exists, ignoring"); netdev_warn_once(priv->netdev, @@ -4356,6 +4708,12 @@ int mlx5e_configure_flower(struct net_device *dev, struct mlx5e_priv *priv, if (err) goto out; + /* Flow rule offloaded to non-uplink representor sharing tc block, + * set the flow's owner dev. + */ + if (is_flow_rule_duplicate_allowed(dev, rpriv)) + flow->orig_dev = dev; + err = rhashtable_lookup_insert_fast(tc_ht, &flow->node, tc_ht_params); if (err) goto err_free; @@ -4588,7 +4946,7 @@ void mlx5e_tc_stats_matchall(struct mlx5e_priv *priv, dpkts = cur_stats.rx_packets - rpriv->prev_vf_vport_stats.rx_packets; dbytes = cur_stats.rx_bytes - rpriv->prev_vf_vport_stats.rx_bytes; rpriv->prev_vf_vport_stats = cur_stats; - flow_stats_update(&ma->stats, dpkts, dbytes, jiffies, + flow_stats_update(&ma->stats, dbytes, dpkts, jiffies, FLOW_ACTION_HW_STATS_DELAYED); } @@ -4707,7 +5065,7 @@ void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv) int mlx5e_tc_esw_init(struct rhashtable *tc_ht) { - const size_t sz_enc_opts = sizeof(struct flow_dissector_key_enc_opts); + const size_t sz_enc_opts = sizeof(struct tunnel_match_enc_opts); struct mlx5_rep_uplink_priv *uplink_priv; struct mlx5e_rep_priv *priv; struct mapping_ctx *mapping; @@ -4796,146 +5154,35 @@ void mlx5e_tc_reoffload_flows_work(struct work_struct *work) mutex_unlock(&rpriv->unready_flows_lock); } -#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) -static bool mlx5e_restore_tunnel(struct mlx5e_priv *priv, struct sk_buff *skb, - struct mlx5e_tc_update_priv *tc_priv, - u32 tunnel_id) -{ - struct mlx5_eswitch *esw = priv->mdev->priv.eswitch; - struct flow_dissector_key_enc_opts enc_opts = {}; - struct mlx5_rep_uplink_priv *uplink_priv; - struct mlx5e_rep_priv *uplink_rpriv; - struct metadata_dst *tun_dst; - struct tunnel_match_key key; - u32 tun_id, enc_opts_id; - struct net_device *dev; - int err; - - enc_opts_id = tunnel_id & ENC_OPTS_BITS_MASK; - tun_id = tunnel_id >> ENC_OPTS_BITS; - - if (!tun_id) - return true; - - uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); - uplink_priv = &uplink_rpriv->uplink_priv; - - err = mapping_find(uplink_priv->tunnel_mapping, tun_id, &key); - if (err) { - WARN_ON_ONCE(true); - netdev_dbg(priv->netdev, - "Couldn't find tunnel for tun_id: %d, err: %d\n", - tun_id, err); - return false; - } - - if (enc_opts_id) { - err = mapping_find(uplink_priv->tunnel_enc_opts_mapping, - enc_opts_id, &enc_opts); - if (err) { - netdev_dbg(priv->netdev, - "Couldn't find tunnel (opts) for tun_id: %d, err: %d\n", - enc_opts_id, err); - return false; - } - } - - tun_dst = tun_rx_dst(enc_opts.len); - if (!tun_dst) { - WARN_ON_ONCE(true); - return false; - } - - ip_tunnel_key_init(&tun_dst->u.tun_info.key, - key.enc_ipv4.src, key.enc_ipv4.dst, - key.enc_ip.tos, key.enc_ip.ttl, - 0, /* label */ - key.enc_tp.src, key.enc_tp.dst, - key32_to_tunnel_id(key.enc_key_id.keyid), - TUNNEL_KEY); - - if (enc_opts.len) - ip_tunnel_info_opts_set(&tun_dst->u.tun_info, enc_opts.data, - enc_opts.len, enc_opts.dst_opt_type); - - skb_dst_set(skb, (struct dst_entry *)tun_dst); - dev = dev_get_by_index(&init_net, key.filter_ifindex); - if (!dev) { - netdev_dbg(priv->netdev, - "Couldn't find tunnel device with ifindex: %d\n", - key.filter_ifindex); - return false; +static int mlx5e_setup_tc_cls_flower(struct mlx5e_priv *priv, + struct flow_cls_offload *cls_flower, + unsigned long flags) +{ + switch (cls_flower->command) { + case FLOW_CLS_REPLACE: + return mlx5e_configure_flower(priv->netdev, priv, cls_flower, + flags); + case FLOW_CLS_DESTROY: + return mlx5e_delete_flower(priv->netdev, priv, cls_flower, + flags); + case FLOW_CLS_STATS: + return mlx5e_stats_flower(priv->netdev, priv, cls_flower, + flags); + default: + return -EOPNOTSUPP; } - - /* Set tun_dev so we do dev_put() after datapath */ - tc_priv->tun_dev = dev; - - skb->dev = dev; - - return true; } -#endif /* CONFIG_NET_TC_SKB_EXT */ -bool mlx5e_tc_rep_update_skb(struct mlx5_cqe64 *cqe, - struct sk_buff *skb, - struct mlx5e_tc_update_priv *tc_priv) +int mlx5e_setup_tc_block_cb(enum tc_setup_type type, void *type_data, + void *cb_priv) { -#if IS_ENABLED(CONFIG_NET_TC_SKB_EXT) - u32 chain = 0, reg_c0, reg_c1, tunnel_id, tuple_id; - struct mlx5_rep_uplink_priv *uplink_priv; - struct mlx5e_rep_priv *uplink_rpriv; - struct tc_skb_ext *tc_skb_ext; - struct mlx5_eswitch *esw; - struct mlx5e_priv *priv; - int tunnel_moffset; - int err; - - reg_c0 = (be32_to_cpu(cqe->sop_drop_qpn) & MLX5E_TC_FLOW_ID_MASK); - if (reg_c0 == MLX5_FS_DEFAULT_FLOW_TAG) - reg_c0 = 0; - reg_c1 = be32_to_cpu(cqe->ft_metadata); - - if (!reg_c0) - return true; - - priv = netdev_priv(skb->dev); - esw = priv->mdev->priv.eswitch; - - err = mlx5_eswitch_get_chain_for_tag(esw, reg_c0, &chain); - if (err) { - netdev_dbg(priv->netdev, - "Couldn't find chain for chain tag: %d, err: %d\n", - reg_c0, err); - return false; - } - - if (chain) { - tc_skb_ext = skb_ext_add(skb, TC_SKB_EXT); - if (!tc_skb_ext) { - WARN_ON(1); - return false; - } + unsigned long flags = MLX5_TC_FLAG(INGRESS) | MLX5_TC_FLAG(NIC_OFFLOAD); + struct mlx5e_priv *priv = cb_priv; - tc_skb_ext->chain = chain; - - tuple_id = reg_c1 & TUPLE_ID_MAX; - - uplink_rpriv = mlx5_eswitch_get_uplink_priv(esw, REP_ETH); - uplink_priv = &uplink_rpriv->uplink_priv; - if (!mlx5e_tc_ct_restore_flow(uplink_priv, skb, tuple_id)) - return false; + switch (type) { + case TC_SETUP_CLSFLOWER: + return mlx5e_setup_tc_cls_flower(priv, type_data, flags); + default: + return -EOPNOTSUPP; } - - tunnel_moffset = mlx5e_tc_attr_to_reg_mappings[TUNNEL_TO_REG].moffset; - tunnel_id = reg_c1 >> (8 * tunnel_moffset); - return mlx5e_restore_tunnel(priv, skb, tc_priv, tunnel_id); -#endif /* CONFIG_NET_TC_SKB_EXT */ - - return true; -} - -void mlx5_tc_rep_post_napi_receive(struct mlx5e_tc_update_priv *tc_priv) -{ - if (tc_priv->tun_dev) - dev_put(tc_priv->tun_dev); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h index abdcfa4c4e0e..5c330b0cae21 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.h @@ -34,11 +34,41 @@ #define __MLX5_EN_TC_H__ #include <net/pkt_cls.h> +#include "en.h" #define MLX5E_TC_FLOW_ID_MASK 0x0000ffff #ifdef CONFIG_MLX5_ESWITCH +struct tunnel_match_key { + struct flow_dissector_key_control enc_control; + struct flow_dissector_key_keyid enc_key_id; + struct flow_dissector_key_ports enc_tp; + struct flow_dissector_key_ip enc_ip; + union { + struct flow_dissector_key_ipv4_addrs enc_ipv4; + struct flow_dissector_key_ipv6_addrs enc_ipv6; + }; + + int filter_ifindex; +}; + +struct tunnel_match_enc_opts { + struct flow_dissector_key_enc_opts key; + struct flow_dissector_key_enc_opts mask; +}; + +/* Tunnel_id mapping is TUNNEL_INFO_BITS + ENC_OPTS_BITS. + * Upper TUNNEL_INFO_BITS for general tunnel info. + * Lower ENC_OPTS_BITS bits for enc_opts. + */ +#define TUNNEL_INFO_BITS 6 +#define TUNNEL_INFO_BITS_MASK GENMASK(TUNNEL_INFO_BITS - 1, 0) +#define ENC_OPTS_BITS 2 +#define ENC_OPTS_BITS_MASK GENMASK(ENC_OPTS_BITS - 1, 0) +#define TUNNEL_ID_BITS (TUNNEL_INFO_BITS + ENC_OPTS_BITS) +#define TUNNEL_ID_MASK GENMASK(TUNNEL_ID_BITS - 1, 0) + enum { MLX5E_TC_FLAG_INGRESS_BIT, MLX5E_TC_FLAG_EGRESS_BIT, @@ -50,9 +80,6 @@ enum { #define MLX5_TC_FLAG(flag) BIT(MLX5E_TC_FLAG_##flag##_BIT) -int mlx5e_tc_nic_init(struct mlx5e_priv *priv); -void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv); - int mlx5e_tc_esw_init(struct rhashtable *tc_ht); void mlx5e_tc_esw_cleanup(struct rhashtable *tc_ht); @@ -119,11 +146,6 @@ struct mlx5e_tc_update_priv { struct net_device *tun_dev; }; -bool mlx5e_tc_rep_update_skb(struct mlx5_cqe64 *cqe, struct sk_buff *skb, - struct mlx5e_tc_update_priv *tc_priv); - -void mlx5_tc_rep_post_napi_receive(struct mlx5e_tc_update_priv *tc_priv); - struct mlx5e_tc_mod_hdr_acts { int num_actions; int max_actions; @@ -148,6 +170,26 @@ void dealloc_mod_hdr_actions(struct mlx5e_tc_mod_hdr_acts *mod_hdr_acts); struct mlx5e_tc_flow; u32 mlx5e_tc_get_flow_tun_id(struct mlx5e_tc_flow *flow); +void mlx5e_tc_set_ethertype(struct mlx5_core_dev *mdev, + struct flow_match_basic *match, bool outer, + void *headers_c, void *headers_v); + +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) + +int mlx5e_tc_nic_init(struct mlx5e_priv *priv); +void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv); + +int mlx5e_setup_tc_block_cb(enum tc_setup_type type, void *type_data, + void *cb_priv); + +#else /* CONFIG_MLX5_CLS_ACT */ +static inline int mlx5e_tc_nic_init(struct mlx5e_priv *priv) { return 0; } +static inline void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv) {} +static inline int +mlx5e_setup_tc_block_cb(enum tc_setup_type type, void *type_data, void *cb_priv) +{ return -EOPNOTSUPP; } +#endif /* CONFIG_MLX5_CLS_ACT */ + #else /* CONFIG_MLX5_ESWITCH */ static inline int mlx5e_tc_nic_init(struct mlx5e_priv *priv) { return 0; } static inline void mlx5e_tc_nic_cleanup(struct mlx5e_priv *priv) {} @@ -156,6 +198,10 @@ static inline int mlx5e_tc_num_filters(struct mlx5e_priv *priv, { return 0; } + +static inline int +mlx5e_setup_tc_block_cb(enum tc_setup_type type, void *type_data, void *cb_priv) +{ return -EOPNOTSUPP; } #endif #endif /* __MLX5_EN_TC_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index fd6b2a1898c5..6d406063aca4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -265,8 +265,8 @@ mlx5e_txwqe_complete(struct mlx5e_txqsq *sq, struct sk_buff *skb, mlx5e_notify_hw(wq, sq->pc, sq->uar_map, cseg); } -netdev_tx_t mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, - struct mlx5e_tx_wqe *wqe, u16 pi, bool xmit_more) +void mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5e_tx_wqe *wqe, u16 pi, bool xmit_more) { struct mlx5_wq_cyc *wq = &sq->wq; struct mlx5_wqe_ctrl_seg *cseg; @@ -324,7 +324,8 @@ netdev_tx_t mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, struct mlx5_wqe_ctrl_seg cur_ctrl = wqe->ctrl; #endif mlx5e_fill_sq_frag_edge(sq, wq, pi, contig_wqebbs_room); - wqe = mlx5e_sq_fetch_wqe(sq, sizeof(*wqe), &pi); + pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); + wqe = MLX5E_TX_FETCH_WQE(sq, pi); #ifdef CONFIG_MLX5_EN_IPSEC wqe->eth = cur_eth; #endif @@ -372,47 +373,38 @@ netdev_tx_t mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, mlx5e_txwqe_complete(sq, skb, opcode, ds_cnt, num_wqebbs, num_bytes, num_dma, wi, cseg, xmit_more); - return NETDEV_TX_OK; + return; err_drop: stats->dropped++; dev_kfree_skb_any(skb); - - return NETDEV_TX_OK; } netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev) { struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5e_accel_tx_state accel = {}; struct mlx5e_tx_wqe *wqe; struct mlx5e_txqsq *sq; u16 pi; sq = priv->txq2sq[skb_get_queue_mapping(skb)]; - wqe = mlx5e_sq_fetch_wqe(sq, sizeof(*wqe), &pi); - /* might send skbs and update wqe and pi */ - skb = mlx5e_accel_handle_tx(skb, sq, dev, &wqe, &pi); - if (unlikely(!skb)) - return NETDEV_TX_OK; + /* May send SKBs and WQEs. */ + if (unlikely(!mlx5e_accel_tx_begin(dev, sq, skb, &accel))) + goto out; - return mlx5e_sq_xmit(sq, skb, wqe, pi, netdev_xmit_more()); -} + pi = mlx5_wq_cyc_ctr2ix(&sq->wq, sq->pc); + wqe = MLX5E_TX_FETCH_WQE(sq, pi); -static void mlx5e_dump_error_cqe(struct mlx5e_txqsq *sq, - struct mlx5_err_cqe *err_cqe) -{ - struct mlx5_cqwq *wq = &sq->cq.wq; - u32 ci; + /* May update the WQE, but may not post other WQEs. */ + if (unlikely(!mlx5e_accel_tx_finish(priv, sq, skb, wqe, &accel))) + goto out; - ci = mlx5_cqwq_ctr2ix(wq, wq->cc - 1); + mlx5e_sq_xmit(sq, skb, wqe, pi, netdev_xmit_more()); - netdev_err(sq->channel->netdev, - "Error cqe on cqn 0x%x, ci 0x%x, sqn 0x%x, opcode 0x%x, syndrome 0x%x, vendor syndrome 0x%x\n", - sq->cq.mcq.cqn, ci, sq->sqn, - get_cqe_opcode((struct mlx5_cqe64 *)err_cqe), - err_cqe->syndrome, err_cqe->vendor_err_synd); - mlx5_dump_err_cqe(sq->cq.mdev, err_cqe); +out: + return NETDEV_TX_OK; } bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) @@ -501,7 +493,7 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) if (unlikely(get_cqe_opcode(cqe) == MLX5_CQE_REQ_ERR)) { if (!test_and_set_bit(MLX5E_SQ_STATE_RECOVERING, &sq->state)) { - mlx5e_dump_error_cqe(sq, + mlx5e_dump_error_cqe(&sq->cq, sq->sqn, (struct mlx5_err_cqe *)cqe); mlx5_wq_cyc_wqe_dump(&sq->wq, ci, wi->num_wqebbs); queue_work(cq->channel->priv->wq, @@ -537,10 +529,9 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq) { struct mlx5e_tx_wqe_info *wi; + u32 dma_fifo_cc, nbytes = 0; + u16 ci, sqcc, npkts = 0; struct sk_buff *skb; - u32 dma_fifo_cc; - u16 sqcc; - u16 ci; int i; sqcc = sq->cc; @@ -565,11 +556,15 @@ void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq) } dev_kfree_skb_any(skb); + npkts++; + nbytes += wi->num_bytes; sqcc += wi->num_wqebbs; } sq->dma_fifo_cc = dma_fifo_cc; sq->cc = sqcc; + + netdev_tx_completed_queue(sq->txq, npkts, nbytes); } #ifdef CONFIG_MLX5_CORE_IPOIB @@ -582,11 +577,9 @@ mlx5i_txwqe_build_datagram(struct mlx5_av *av, u32 dqpn, u32 dqkey, dseg->av.key.qkey.qkey = cpu_to_be32(dqkey); } -netdev_tx_t mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, - struct mlx5_av *av, u32 dqpn, u32 dqkey, - bool xmit_more) +void mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5_av *av, u32 dqpn, u32 dqkey, bool xmit_more) { - struct mlx5_wq_cyc *wq = &sq->wq; struct mlx5i_tx_wqe *wqe; struct mlx5_wqe_datagram_seg *datagram; @@ -596,9 +589,9 @@ netdev_tx_t mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, struct mlx5e_tx_wqe_info *wi; struct mlx5e_sq_stats *stats = sq->stats; - u16 headlen, ihs, pi, contig_wqebbs_room; u16 ds_cnt, ds_cnt_inl = 0; u8 num_wqebbs, opcode; + u16 headlen, ihs, pi; u32 num_bytes; int num_dma; __be16 mss; @@ -634,14 +627,8 @@ netdev_tx_t mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, } num_wqebbs = DIV_ROUND_UP(ds_cnt, MLX5_SEND_WQEBB_NUM_DS); - pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - contig_wqebbs_room = mlx5_wq_cyc_get_contig_wqebbs(wq, pi); - if (unlikely(contig_wqebbs_room < num_wqebbs)) { - mlx5e_fill_sq_frag_edge(sq, wq, pi, contig_wqebbs_room); - pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - } - - mlx5i_sq_fetch_wqe(sq, &wqe, pi); + pi = mlx5e_txqsq_get_next_pi(sq, num_wqebbs); + wqe = MLX5I_SQ_FETCH_WQE(sq, pi); /* fill wqe */ wi = &sq->db.wqe_info[pi]; @@ -669,12 +656,10 @@ netdev_tx_t mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, mlx5e_txwqe_complete(sq, skb, opcode, ds_cnt, num_wqebbs, num_bytes, num_dma, wi, cseg, xmit_more); - return NETDEV_TX_OK; + return; err_drop: stats->dropped++; dev_kfree_skb_any(skb); - - return NETDEV_TX_OK; } #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c index acb20215a33b..8480278f2ee2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c @@ -78,8 +78,11 @@ void mlx5e_trigger_irq(struct mlx5e_icosq *sq) struct mlx5e_tx_wqe *nopwqe; u16 pi = mlx5_wq_cyc_ctr2ix(wq, sq->pc); - sq->db.ico_wqe[pi].opcode = MLX5_OPCODE_NOP; - sq->db.ico_wqe[pi].num_wqebbs = 1; + sq->db.wqe_info[pi] = (struct mlx5e_icosq_wqe_info) { + .wqe_type = MLX5E_ICOSQ_WQE_NOP, + .num_wqebbs = 1, + }; + nopwqe = mlx5e_post_nop(wq, sq->sqn, &sq->pc); mlx5e_notify_hw(wq, sq->pc, sq->uar_map, &nopwqe->ctrl); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c index 4d974b5405b5..31ef9f8420c8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c @@ -609,11 +609,13 @@ static int create_async_eqs(struct mlx5_core_dev *dev) .nent = MLX5_NUM_CMD_EQE, .mask[0] = 1ull << MLX5_EVENT_TYPE_CMD, }; + mlx5_cmd_allowed_opcode(dev, MLX5_CMD_OP_CREATE_EQ); err = setup_async_eq(dev, &table->cmd_eq, ¶m, "cmd"); if (err) goto err1; mlx5_cmd_use_events(dev); + mlx5_cmd_allowed_opcode(dev, CMD_ALLOWED_OPCODE_ALL); param = (struct mlx5_eq_param) { .irq_index = 0, @@ -643,6 +645,7 @@ err2: mlx5_cmd_use_polling(dev); cleanup_async_eq(dev, &table->cmd_eq, "cmd"); err1: + mlx5_cmd_allowed_opcode(dev, CMD_ALLOWED_OPCODE_ALL); mlx5_eq_notifier_unregister(dev, &table->cq_err_nb); return err; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_lgcy.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_lgcy.c new file mode 100644 index 000000000000..d46f8b225ebe --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_lgcy.c @@ -0,0 +1,170 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2020 Mellanox Technologies Inc. All rights reserved. */ + +#include "mlx5_core.h" +#include "eswitch.h" +#include "helper.h" +#include "lgcy.h" + +static void esw_acl_egress_lgcy_rules_destroy(struct mlx5_vport *vport) +{ + esw_acl_egress_vlan_destroy(vport); + if (!IS_ERR_OR_NULL(vport->egress.legacy.drop_rule)) { + mlx5_del_flow_rules(vport->egress.legacy.drop_rule); + vport->egress.legacy.drop_rule = NULL; + } +} + +static int esw_acl_egress_lgcy_groups_create(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_core_dev *dev = esw->dev; + struct mlx5_flow_group *drop_grp; + u32 *flow_group_in; + int err = 0; + + err = esw_acl_egress_vlan_grp_create(esw, vport); + if (err) + return err; + + flow_group_in = kvzalloc(inlen, GFP_KERNEL); + if (!flow_group_in) { + err = -ENOMEM; + goto alloc_err; + } + + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 1); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 1); + drop_grp = mlx5_create_flow_group(vport->egress.acl, flow_group_in); + if (IS_ERR(drop_grp)) { + err = PTR_ERR(drop_grp); + esw_warn(dev, "Failed to create E-Switch vport[%d] egress drop flow group, err(%d)\n", + vport->vport, err); + goto drop_grp_err; + } + + vport->egress.legacy.drop_grp = drop_grp; + kvfree(flow_group_in); + return 0; + +drop_grp_err: + kvfree(flow_group_in); +alloc_err: + esw_acl_egress_vlan_grp_destroy(vport); + return err; +} + +static void esw_acl_egress_lgcy_groups_destroy(struct mlx5_vport *vport) +{ + if (!IS_ERR_OR_NULL(vport->egress.legacy.drop_grp)) { + mlx5_destroy_flow_group(vport->egress.legacy.drop_grp); + vport->egress.legacy.drop_grp = NULL; + } + esw_acl_egress_vlan_grp_destroy(vport); +} + +int esw_acl_egress_lgcy_setup(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + struct mlx5_flow_destination drop_ctr_dst = {}; + struct mlx5_flow_destination *dst = NULL; + struct mlx5_fc *drop_counter = NULL; + struct mlx5_flow_act flow_act = {}; + /* The egress acl table contains 2 rules: + * 1)Allow traffic with vlan_tag=vst_vlan_id + * 2)Drop all other traffic. + */ + int table_size = 2; + int dest_num = 0; + int err = 0; + + if (MLX5_CAP_ESW_EGRESS_ACL(esw->dev, flow_counter)) { + drop_counter = mlx5_fc_create(esw->dev, false); + if (IS_ERR(drop_counter)) + esw_warn(esw->dev, + "vport[%d] configure egress drop rule counter err(%ld)\n", + vport->vport, PTR_ERR(drop_counter)); + vport->egress.legacy.drop_counter = drop_counter; + } + + esw_acl_egress_lgcy_rules_destroy(vport); + + if (!vport->info.vlan && !vport->info.qos) { + esw_acl_egress_lgcy_cleanup(esw, vport); + return 0; + } + + if (!IS_ERR_OR_NULL(vport->egress.acl)) + return 0; + + vport->egress.acl = esw_acl_table_create(esw, vport->vport, + MLX5_FLOW_NAMESPACE_ESW_EGRESS, + table_size); + if (IS_ERR_OR_NULL(vport->egress.acl)) { + err = PTR_ERR(vport->egress.acl); + vport->egress.acl = NULL; + goto out; + } + + err = esw_acl_egress_lgcy_groups_create(esw, vport); + if (err) + goto out; + + esw_debug(esw->dev, + "vport[%d] configure egress rules, vlan(%d) qos(%d)\n", + vport->vport, vport->info.vlan, vport->info.qos); + + /* Allowed vlan rule */ + err = esw_egress_acl_vlan_create(esw, vport, NULL, vport->info.vlan, + MLX5_FLOW_CONTEXT_ACTION_ALLOW); + if (err) + goto out; + + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP; + + /* Attach egress drop flow counter */ + if (!IS_ERR_OR_NULL(drop_counter)) { + flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_COUNT; + drop_ctr_dst.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + drop_ctr_dst.counter_id = mlx5_fc_id(drop_counter); + dst = &drop_ctr_dst; + dest_num++; + } + vport->egress.legacy.drop_rule = + mlx5_add_flow_rules(vport->egress.acl, NULL, + &flow_act, dst, dest_num); + if (IS_ERR(vport->egress.legacy.drop_rule)) { + err = PTR_ERR(vport->egress.legacy.drop_rule); + esw_warn(esw->dev, + "vport[%d] configure egress drop rule failed, err(%d)\n", + vport->vport, err); + vport->egress.legacy.drop_rule = NULL; + goto out; + } + + return err; + +out: + esw_acl_egress_lgcy_cleanup(esw, vport); + return err; +} + +void esw_acl_egress_lgcy_cleanup(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + if (IS_ERR_OR_NULL(vport->egress.acl)) + goto clean_drop_counter; + + esw_debug(esw->dev, "Destroy vport[%d] E-Switch egress ACL\n", vport->vport); + + esw_acl_egress_lgcy_rules_destroy(vport); + esw_acl_egress_lgcy_groups_destroy(vport); + esw_acl_egress_table_destroy(vport); + +clean_drop_counter: + if (!IS_ERR_OR_NULL(vport->egress.legacy.drop_counter)) { + mlx5_fc_destroy(esw->dev, vport->egress.legacy.drop_counter); + vport->egress.legacy.drop_counter = NULL; + } +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_ofld.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_ofld.c new file mode 100644 index 000000000000..07b2acd7e6b3 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/egress_ofld.c @@ -0,0 +1,235 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2020 Mellanox Technologies Inc. All rights reserved. */ + +#include "mlx5_core.h" +#include "eswitch.h" +#include "helper.h" +#include "ofld.h" + +static void esw_acl_egress_ofld_fwd2vport_destroy(struct mlx5_vport *vport) +{ + if (!vport->egress.offloads.fwd_rule) + return; + + mlx5_del_flow_rules(vport->egress.offloads.fwd_rule); + vport->egress.offloads.fwd_rule = NULL; +} + +static int esw_acl_egress_ofld_fwd2vport_create(struct mlx5_eswitch *esw, + struct mlx5_vport *vport, + struct mlx5_flow_destination *fwd_dest) +{ + struct mlx5_flow_act flow_act = {}; + int err = 0; + + esw_debug(esw->dev, "vport(%d) configure egress acl rule fwd2vport(%d)\n", + vport->vport, fwd_dest->vport.num); + + /* Delete the old egress forward-to-vport rule if any */ + esw_acl_egress_ofld_fwd2vport_destroy(vport); + + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; + + vport->egress.offloads.fwd_rule = + mlx5_add_flow_rules(vport->egress.acl, NULL, + &flow_act, fwd_dest, 1); + if (IS_ERR(vport->egress.offloads.fwd_rule)) { + err = PTR_ERR(vport->egress.offloads.fwd_rule); + esw_warn(esw->dev, + "vport(%d) failed to add fwd2vport acl rule err(%d)\n", + vport->vport, err); + vport->egress.offloads.fwd_rule = NULL; + } + + return err; +} + +static int esw_acl_egress_ofld_rules_create(struct mlx5_eswitch *esw, + struct mlx5_vport *vport, + struct mlx5_flow_destination *fwd_dest) +{ + int err = 0; + int action; + + if (MLX5_CAP_GEN(esw->dev, prio_tag_required)) { + /* For prio tag mode, there is only 1 FTEs: + * 1) prio tag packets - pop the prio tag VLAN, allow + * Unmatched traffic is allowed by default + */ + esw_debug(esw->dev, + "vport[%d] configure prio tag egress rules\n", vport->vport); + + action = MLX5_FLOW_CONTEXT_ACTION_VLAN_POP; + action |= fwd_dest ? MLX5_FLOW_CONTEXT_ACTION_FWD_DEST : + MLX5_FLOW_CONTEXT_ACTION_ALLOW; + + /* prio tag vlan rule - pop it so vport receives untagged packets */ + err = esw_egress_acl_vlan_create(esw, vport, fwd_dest, 0, action); + if (err) + goto prio_err; + } + + if (fwd_dest) { + err = esw_acl_egress_ofld_fwd2vport_create(esw, vport, fwd_dest); + if (err) + goto fwd_err; + } + + return 0; + +fwd_err: + esw_acl_egress_vlan_destroy(vport); +prio_err: + return err; +} + +static void esw_acl_egress_ofld_rules_destroy(struct mlx5_vport *vport) +{ + esw_acl_egress_vlan_destroy(vport); + esw_acl_egress_ofld_fwd2vport_destroy(vport); +} + +static int esw_acl_egress_ofld_groups_create(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_flow_group *fwd_grp; + u32 *flow_group_in; + u32 flow_index = 0; + int ret = 0; + + if (MLX5_CAP_GEN(esw->dev, prio_tag_required)) { + ret = esw_acl_egress_vlan_grp_create(esw, vport); + if (ret) + return ret; + + flow_index++; + } + + if (!mlx5_esw_acl_egress_fwd2vport_supported(esw)) + goto out; + + flow_group_in = kvzalloc(inlen, GFP_KERNEL); + if (!flow_group_in) { + ret = -ENOMEM; + goto fwd_grp_err; + } + + /* This group holds 1 FTE to forward all packets to other vport + * when bond vports is supported. + */ + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, flow_index); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, flow_index); + fwd_grp = mlx5_create_flow_group(vport->egress.acl, flow_group_in); + if (IS_ERR(fwd_grp)) { + ret = PTR_ERR(fwd_grp); + esw_warn(esw->dev, + "Failed to create vport[%d] egress fwd2vport flow group, err(%d)\n", + vport->vport, ret); + kvfree(flow_group_in); + goto fwd_grp_err; + } + vport->egress.offloads.fwd_grp = fwd_grp; + kvfree(flow_group_in); + return 0; + +fwd_grp_err: + esw_acl_egress_vlan_grp_destroy(vport); +out: + return ret; +} + +static void esw_acl_egress_ofld_groups_destroy(struct mlx5_vport *vport) +{ + if (!IS_ERR_OR_NULL(vport->egress.offloads.fwd_grp)) { + mlx5_destroy_flow_group(vport->egress.offloads.fwd_grp); + vport->egress.offloads.fwd_grp = NULL; + } + esw_acl_egress_vlan_grp_destroy(vport); +} + +int esw_acl_egress_ofld_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport) +{ + int table_size = 0; + int err; + + if (!mlx5_esw_acl_egress_fwd2vport_supported(esw) && + !MLX5_CAP_GEN(esw->dev, prio_tag_required)) + return 0; + + esw_acl_egress_ofld_rules_destroy(vport); + + if (mlx5_esw_acl_egress_fwd2vport_supported(esw)) + table_size++; + if (MLX5_CAP_GEN(esw->dev, prio_tag_required)) + table_size++; + vport->egress.acl = esw_acl_table_create(esw, vport->vport, + MLX5_FLOW_NAMESPACE_ESW_EGRESS, table_size); + if (IS_ERR_OR_NULL(vport->egress.acl)) { + err = PTR_ERR(vport->egress.acl); + vport->egress.acl = NULL; + return err; + } + + err = esw_acl_egress_ofld_groups_create(esw, vport); + if (err) + goto group_err; + + esw_debug(esw->dev, "vport[%d] configure egress rules\n", vport->vport); + + err = esw_acl_egress_ofld_rules_create(esw, vport, NULL); + if (err) + goto rules_err; + + return 0; + +rules_err: + esw_acl_egress_ofld_groups_destroy(vport); +group_err: + esw_acl_egress_table_destroy(vport); + return err; +} + +void esw_acl_egress_ofld_cleanup(struct mlx5_vport *vport) +{ + esw_acl_egress_ofld_rules_destroy(vport); + esw_acl_egress_ofld_groups_destroy(vport); + esw_acl_egress_table_destroy(vport); +} + +int mlx5_esw_acl_egress_vport_bond(struct mlx5_eswitch *esw, u16 active_vport_num, + u16 passive_vport_num) +{ + struct mlx5_vport *passive_vport = mlx5_eswitch_get_vport(esw, passive_vport_num); + struct mlx5_vport *active_vport = mlx5_eswitch_get_vport(esw, active_vport_num); + struct mlx5_flow_destination fwd_dest = {}; + + if (IS_ERR(active_vport)) + return PTR_ERR(active_vport); + if (IS_ERR(passive_vport)) + return PTR_ERR(passive_vport); + + /* Cleanup and recreate rules WITHOUT fwd2vport of active vport */ + esw_acl_egress_ofld_rules_destroy(active_vport); + esw_acl_egress_ofld_rules_create(esw, active_vport, NULL); + + /* Cleanup and recreate all rules + fwd2vport rule of passive vport to forward */ + esw_acl_egress_ofld_rules_destroy(passive_vport); + fwd_dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; + fwd_dest.vport.num = active_vport_num; + fwd_dest.vport.vhca_id = MLX5_CAP_GEN(esw->dev, vhca_id); + fwd_dest.vport.flags = MLX5_FLOW_DEST_VPORT_VHCA_ID; + + return esw_acl_egress_ofld_rules_create(esw, passive_vport, &fwd_dest); +} + +int mlx5_esw_acl_egress_vport_unbond(struct mlx5_eswitch *esw, u16 vport_num) +{ + struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num); + + if (IS_ERR(vport)) + return PTR_ERR(vport); + + esw_acl_egress_ofld_rules_destroy(vport); + return esw_acl_egress_ofld_rules_create(esw, vport, NULL); +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/helper.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/helper.c new file mode 100644 index 000000000000..22f4c1c28006 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/helper.c @@ -0,0 +1,160 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2020 Mellanox Technologies Inc. All rights reserved. */ + +#include "mlx5_core.h" +#include "eswitch.h" +#include "helper.h" + +struct mlx5_flow_table * +esw_acl_table_create(struct mlx5_eswitch *esw, u16 vport_num, int ns, int size) +{ + struct mlx5_core_dev *dev = esw->dev; + struct mlx5_flow_namespace *root_ns; + struct mlx5_flow_table *acl; + int acl_supported; + int vport_index; + int err; + + acl_supported = (ns == MLX5_FLOW_NAMESPACE_ESW_INGRESS) ? + MLX5_CAP_ESW_INGRESS_ACL(dev, ft_support) : + MLX5_CAP_ESW_EGRESS_ACL(dev, ft_support); + + if (!acl_supported) + return ERR_PTR(-EOPNOTSUPP); + + esw_debug(dev, "Create vport[%d] %s ACL table\n", vport_num, + ns == MLX5_FLOW_NAMESPACE_ESW_INGRESS ? "ingress" : "egress"); + + vport_index = mlx5_eswitch_vport_num_to_index(esw, vport_num); + root_ns = mlx5_get_flow_vport_acl_namespace(dev, ns, vport_index); + if (!root_ns) { + esw_warn(dev, "Failed to get E-Switch root namespace for vport (%d)\n", + vport_num); + return ERR_PTR(-EOPNOTSUPP); + } + + acl = mlx5_create_vport_flow_table(root_ns, 0, size, 0, vport_num); + if (IS_ERR(acl)) { + err = PTR_ERR(acl); + esw_warn(dev, "vport[%d] create %s ACL table, err(%d)\n", vport_num, + ns == MLX5_FLOW_NAMESPACE_ESW_INGRESS ? "ingress" : "egress", err); + } + return acl; +} + +int esw_egress_acl_vlan_create(struct mlx5_eswitch *esw, + struct mlx5_vport *vport, + struct mlx5_flow_destination *fwd_dest, + u16 vlan_id, u32 flow_action) +{ + struct mlx5_flow_act flow_act = {}; + struct mlx5_flow_spec *spec; + int err = 0; + + if (vport->egress.allowed_vlan) + return -EEXIST; + + spec = kvzalloc(sizeof(*spec), GFP_KERNEL); + if (!spec) + return -ENOMEM; + + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.cvlan_tag); + MLX5_SET_TO_ONES(fte_match_param, spec->match_value, outer_headers.cvlan_tag); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.first_vid); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.first_vid, vlan_id); + + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + flow_act.action = flow_action; + vport->egress.allowed_vlan = + mlx5_add_flow_rules(vport->egress.acl, spec, + &flow_act, fwd_dest, 0); + if (IS_ERR(vport->egress.allowed_vlan)) { + err = PTR_ERR(vport->egress.allowed_vlan); + esw_warn(esw->dev, + "vport[%d] configure egress vlan rule failed, err(%d)\n", + vport->vport, err); + vport->egress.allowed_vlan = NULL; + } + + kvfree(spec); + return err; +} + +void esw_acl_egress_vlan_destroy(struct mlx5_vport *vport) +{ + if (!IS_ERR_OR_NULL(vport->egress.allowed_vlan)) { + mlx5_del_flow_rules(vport->egress.allowed_vlan); + vport->egress.allowed_vlan = NULL; + } +} + +int esw_acl_egress_vlan_grp_create(struct mlx5_eswitch *esw, struct mlx5_vport *vport) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_flow_group *vlan_grp; + void *match_criteria; + u32 *flow_group_in; + int ret = 0; + + flow_group_in = kvzalloc(inlen, GFP_KERNEL); + if (!flow_group_in) + return -ENOMEM; + + MLX5_SET(create_flow_group_in, flow_group_in, + match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); + match_criteria = MLX5_ADDR_OF(create_flow_group_in, + flow_group_in, match_criteria); + MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.cvlan_tag); + MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.first_vid); + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 0); + + vlan_grp = mlx5_create_flow_group(vport->egress.acl, flow_group_in); + if (IS_ERR(vlan_grp)) { + ret = PTR_ERR(vlan_grp); + esw_warn(esw->dev, + "Failed to create E-Switch vport[%d] egress pop vlans flow group, err(%d)\n", + vport->vport, ret); + goto out; + } + vport->egress.vlan_grp = vlan_grp; + +out: + kvfree(flow_group_in); + return ret; +} + +void esw_acl_egress_vlan_grp_destroy(struct mlx5_vport *vport) +{ + if (!IS_ERR_OR_NULL(vport->egress.vlan_grp)) { + mlx5_destroy_flow_group(vport->egress.vlan_grp); + vport->egress.vlan_grp = NULL; + } +} + +void esw_acl_egress_table_destroy(struct mlx5_vport *vport) +{ + if (IS_ERR_OR_NULL(vport->egress.acl)) + return; + + mlx5_destroy_flow_table(vport->egress.acl); + vport->egress.acl = NULL; +} + +void esw_acl_ingress_table_destroy(struct mlx5_vport *vport) +{ + if (!vport->ingress.acl) + return; + + mlx5_destroy_flow_table(vport->ingress.acl); + vport->ingress.acl = NULL; +} + +void esw_acl_ingress_allow_rule_destroy(struct mlx5_vport *vport) +{ + if (!vport->ingress.allow_rule) + return; + + mlx5_del_flow_rules(vport->ingress.allow_rule); + vport->ingress.allow_rule = NULL; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/helper.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/helper.h new file mode 100644 index 000000000000..8dc4cab66a71 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/helper.h @@ -0,0 +1,26 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* Copyright (c) 2020 Mellanox Technologies Inc. All rights reserved. */ + +#ifndef __MLX5_ESWITCH_ACL_HELPER_H__ +#define __MLX5_ESWITCH_ACL_HELPER_H__ + +#include "eswitch.h" + +/* General acl helper functions */ +struct mlx5_flow_table * +esw_acl_table_create(struct mlx5_eswitch *esw, u16 vport_num, int ns, int size); + +/* Egress acl helper functions */ +void esw_acl_egress_table_destroy(struct mlx5_vport *vport); +int esw_egress_acl_vlan_create(struct mlx5_eswitch *esw, struct mlx5_vport *vport, + struct mlx5_flow_destination *fwd_dest, + u16 vlan_id, u32 flow_action); +void esw_acl_egress_vlan_destroy(struct mlx5_vport *vport); +int esw_acl_egress_vlan_grp_create(struct mlx5_eswitch *esw, struct mlx5_vport *vport); +void esw_acl_egress_vlan_grp_destroy(struct mlx5_vport *vport); + +/* Ingress acl helper functions */ +void esw_acl_ingress_table_destroy(struct mlx5_vport *vport); +void esw_acl_ingress_allow_rule_destroy(struct mlx5_vport *vport); + +#endif /* __MLX5_ESWITCH_ACL_HELPER_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ingress_lgcy.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ingress_lgcy.c new file mode 100644 index 000000000000..9bda4fe2eafa --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ingress_lgcy.c @@ -0,0 +1,279 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2020 Mellanox Technologies Inc. All rights reserved. */ + +#include "mlx5_core.h" +#include "eswitch.h" +#include "helper.h" +#include "lgcy.h" + +static void esw_acl_ingress_lgcy_rules_destroy(struct mlx5_vport *vport) +{ + if (vport->ingress.legacy.drop_rule) { + mlx5_del_flow_rules(vport->ingress.legacy.drop_rule); + vport->ingress.legacy.drop_rule = NULL; + } + esw_acl_ingress_allow_rule_destroy(vport); +} + +static int esw_acl_ingress_lgcy_groups_create(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_core_dev *dev = esw->dev; + struct mlx5_flow_group *g; + void *match_criteria; + u32 *flow_group_in; + int err; + + flow_group_in = kvzalloc(inlen, GFP_KERNEL); + if (!flow_group_in) + return -ENOMEM; + + match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria); + + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, + MLX5_MATCH_OUTER_HEADERS); + MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.cvlan_tag); + MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.smac_47_16); + MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.smac_15_0); + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 0); + + g = mlx5_create_flow_group(vport->ingress.acl, flow_group_in); + if (IS_ERR(g)) { + err = PTR_ERR(g); + esw_warn(dev, "vport[%d] ingress create untagged spoofchk flow group, err(%d)\n", + vport->vport, err); + goto spoof_err; + } + vport->ingress.legacy.allow_untagged_spoofchk_grp = g; + + memset(flow_group_in, 0, inlen); + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, + MLX5_MATCH_OUTER_HEADERS); + MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.cvlan_tag); + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 1); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 1); + + g = mlx5_create_flow_group(vport->ingress.acl, flow_group_in); + if (IS_ERR(g)) { + err = PTR_ERR(g); + esw_warn(dev, "vport[%d] ingress create untagged flow group, err(%d)\n", + vport->vport, err); + goto untagged_err; + } + vport->ingress.legacy.allow_untagged_only_grp = g; + + memset(flow_group_in, 0, inlen); + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, + MLX5_MATCH_OUTER_HEADERS); + MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.smac_47_16); + MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.smac_15_0); + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 2); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 2); + + g = mlx5_create_flow_group(vport->ingress.acl, flow_group_in); + if (IS_ERR(g)) { + err = PTR_ERR(g); + esw_warn(dev, "vport[%d] ingress create spoofchk flow group, err(%d)\n", + vport->vport, err); + goto allow_spoof_err; + } + vport->ingress.legacy.allow_spoofchk_only_grp = g; + + memset(flow_group_in, 0, inlen); + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 3); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 3); + + g = mlx5_create_flow_group(vport->ingress.acl, flow_group_in); + if (IS_ERR(g)) { + err = PTR_ERR(g); + esw_warn(dev, "vport[%d] ingress create drop flow group, err(%d)\n", + vport->vport, err); + goto drop_err; + } + vport->ingress.legacy.drop_grp = g; + kvfree(flow_group_in); + return 0; + +drop_err: + if (!IS_ERR_OR_NULL(vport->ingress.legacy.allow_spoofchk_only_grp)) { + mlx5_destroy_flow_group(vport->ingress.legacy.allow_spoofchk_only_grp); + vport->ingress.legacy.allow_spoofchk_only_grp = NULL; + } +allow_spoof_err: + if (!IS_ERR_OR_NULL(vport->ingress.legacy.allow_untagged_only_grp)) { + mlx5_destroy_flow_group(vport->ingress.legacy.allow_untagged_only_grp); + vport->ingress.legacy.allow_untagged_only_grp = NULL; + } +untagged_err: + if (!IS_ERR_OR_NULL(vport->ingress.legacy.allow_untagged_spoofchk_grp)) { + mlx5_destroy_flow_group(vport->ingress.legacy.allow_untagged_spoofchk_grp); + vport->ingress.legacy.allow_untagged_spoofchk_grp = NULL; + } +spoof_err: + kvfree(flow_group_in); + return err; +} + +static void esw_acl_ingress_lgcy_groups_destroy(struct mlx5_vport *vport) +{ + if (vport->ingress.legacy.allow_spoofchk_only_grp) { + mlx5_destroy_flow_group(vport->ingress.legacy.allow_spoofchk_only_grp); + vport->ingress.legacy.allow_spoofchk_only_grp = NULL; + } + if (vport->ingress.legacy.allow_untagged_only_grp) { + mlx5_destroy_flow_group(vport->ingress.legacy.allow_untagged_only_grp); + vport->ingress.legacy.allow_untagged_only_grp = NULL; + } + if (vport->ingress.legacy.allow_untagged_spoofchk_grp) { + mlx5_destroy_flow_group(vport->ingress.legacy.allow_untagged_spoofchk_grp); + vport->ingress.legacy.allow_untagged_spoofchk_grp = NULL; + } + if (vport->ingress.legacy.drop_grp) { + mlx5_destroy_flow_group(vport->ingress.legacy.drop_grp); + vport->ingress.legacy.drop_grp = NULL; + } +} + +int esw_acl_ingress_lgcy_setup(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + struct mlx5_flow_destination drop_ctr_dst = {}; + struct mlx5_flow_destination *dst = NULL; + struct mlx5_flow_act flow_act = {}; + struct mlx5_flow_spec *spec = NULL; + struct mlx5_fc *counter = NULL; + /* The ingress acl table contains 4 groups + * (2 active rules at the same time - + * 1 allow rule from one of the first 3 groups. + * 1 drop rule from the last group): + * 1)Allow untagged traffic with smac=original mac. + * 2)Allow untagged traffic. + * 3)Allow traffic with smac=original mac. + * 4)Drop all other traffic. + */ + int table_size = 4; + int dest_num = 0; + int err = 0; + u8 *smac_v; + + esw_acl_ingress_lgcy_rules_destroy(vport); + + if (MLX5_CAP_ESW_INGRESS_ACL(esw->dev, flow_counter)) { + counter = mlx5_fc_create(esw->dev, false); + if (IS_ERR(counter)) + esw_warn(esw->dev, + "vport[%d] configure ingress drop rule counter failed\n", + vport->vport); + vport->ingress.legacy.drop_counter = counter; + } + + if (!vport->info.vlan && !vport->info.qos && !vport->info.spoofchk) { + esw_acl_ingress_lgcy_cleanup(esw, vport); + return 0; + } + + if (!vport->ingress.acl) { + vport->ingress.acl = esw_acl_table_create(esw, vport->vport, + MLX5_FLOW_NAMESPACE_ESW_INGRESS, + table_size); + if (IS_ERR_OR_NULL(vport->ingress.acl)) { + err = PTR_ERR(vport->ingress.acl); + vport->ingress.acl = NULL; + return err; + } + + err = esw_acl_ingress_lgcy_groups_create(esw, vport); + if (err) + goto out; + } + + esw_debug(esw->dev, + "vport[%d] configure ingress rules, vlan(%d) qos(%d)\n", + vport->vport, vport->info.vlan, vport->info.qos); + + spec = kvzalloc(sizeof(*spec), GFP_KERNEL); + if (!spec) { + err = -ENOMEM; + goto out; + } + + if (vport->info.vlan || vport->info.qos) + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, + outer_headers.cvlan_tag); + + if (vport->info.spoofchk) { + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, + outer_headers.smac_47_16); + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, + outer_headers.smac_15_0); + smac_v = MLX5_ADDR_OF(fte_match_param, + spec->match_value, + outer_headers.smac_47_16); + ether_addr_copy(smac_v, vport->info.mac); + } + + /* Create ingress allow rule */ + memset(spec, 0, sizeof(*spec)); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_ALLOW; + vport->ingress.allow_rule = mlx5_add_flow_rules(vport->ingress.acl, spec, + &flow_act, NULL, 0); + if (IS_ERR(vport->ingress.allow_rule)) { + err = PTR_ERR(vport->ingress.allow_rule); + esw_warn(esw->dev, + "vport[%d] configure ingress allow rule, err(%d)\n", + vport->vport, err); + vport->ingress.allow_rule = NULL; + goto out; + } + + memset(&flow_act, 0, sizeof(flow_act)); + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP; + /* Attach drop flow counter */ + if (counter) { + flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_COUNT; + drop_ctr_dst.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; + drop_ctr_dst.counter_id = mlx5_fc_id(counter); + dst = &drop_ctr_dst; + dest_num++; + } + vport->ingress.legacy.drop_rule = + mlx5_add_flow_rules(vport->ingress.acl, NULL, + &flow_act, dst, dest_num); + if (IS_ERR(vport->ingress.legacy.drop_rule)) { + err = PTR_ERR(vport->ingress.legacy.drop_rule); + esw_warn(esw->dev, + "vport[%d] configure ingress drop rule, err(%d)\n", + vport->vport, err); + vport->ingress.legacy.drop_rule = NULL; + goto out; + } + kvfree(spec); + return 0; + +out: + esw_acl_ingress_lgcy_cleanup(esw, vport); + kvfree(spec); + return err; +} + +void esw_acl_ingress_lgcy_cleanup(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + if (IS_ERR_OR_NULL(vport->ingress.acl)) + goto clean_drop_counter; + + esw_debug(esw->dev, "Destroy vport[%d] E-Switch ingress ACL\n", vport->vport); + + esw_acl_ingress_lgcy_rules_destroy(vport); + esw_acl_ingress_lgcy_groups_destroy(vport); + esw_acl_ingress_table_destroy(vport); + +clean_drop_counter: + if (!IS_ERR_OR_NULL(vport->ingress.legacy.drop_counter)) { + mlx5_fc_destroy(esw->dev, vport->ingress.legacy.drop_counter); + vport->ingress.legacy.drop_counter = NULL; + } +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ingress_ofld.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ingress_ofld.c new file mode 100644 index 000000000000..4e55d7225a26 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ingress_ofld.c @@ -0,0 +1,322 @@ +// SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB +/* Copyright (c) 2020 Mellanox Technologies Inc. All rights reserved. */ + +#include "mlx5_core.h" +#include "eswitch.h" +#include "helper.h" +#include "ofld.h" + +static bool +esw_acl_ingress_prio_tag_enabled(const struct mlx5_eswitch *esw, + const struct mlx5_vport *vport) +{ + return (MLX5_CAP_GEN(esw->dev, prio_tag_required) && + mlx5_eswitch_is_vf_vport(esw, vport->vport)); +} + +static int esw_acl_ingress_prio_tag_create(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + struct mlx5_flow_act flow_act = {}; + struct mlx5_flow_spec *spec; + int err = 0; + + /* For prio tag mode, there is only 1 FTEs: + * 1) Untagged packets - push prio tag VLAN and modify metadata if + * required, allow + * Unmatched traffic is allowed by default + */ + spec = kvzalloc(sizeof(*spec), GFP_KERNEL); + if (!spec) + return -ENOMEM; + + /* Untagged packets - push prio tag VLAN, allow */ + MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.cvlan_tag); + MLX5_SET(fte_match_param, spec->match_value, outer_headers.cvlan_tag, 0); + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH | + MLX5_FLOW_CONTEXT_ACTION_ALLOW; + flow_act.vlan[0].ethtype = ETH_P_8021Q; + flow_act.vlan[0].vid = 0; + flow_act.vlan[0].prio = 0; + + if (vport->ingress.offloads.modify_metadata_rule) { + flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; + flow_act.modify_hdr = vport->ingress.offloads.modify_metadata; + } + + vport->ingress.allow_rule = mlx5_add_flow_rules(vport->ingress.acl, spec, + &flow_act, NULL, 0); + if (IS_ERR(vport->ingress.allow_rule)) { + err = PTR_ERR(vport->ingress.allow_rule); + esw_warn(esw->dev, + "vport[%d] configure ingress untagged allow rule, err(%d)\n", + vport->vport, err); + vport->ingress.allow_rule = NULL; + } + + kvfree(spec); + return err; +} + +static int esw_acl_ingress_mod_metadata_create(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + u8 action[MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)] = {}; + struct mlx5_flow_act flow_act = {}; + int err = 0; + u32 key; + + key = mlx5_eswitch_get_vport_metadata_for_match(esw, vport->vport); + key >>= ESW_SOURCE_PORT_METADATA_OFFSET; + + MLX5_SET(set_action_in, action, action_type, MLX5_ACTION_TYPE_SET); + MLX5_SET(set_action_in, action, field, + MLX5_ACTION_IN_FIELD_METADATA_REG_C_0); + MLX5_SET(set_action_in, action, data, key); + MLX5_SET(set_action_in, action, offset, + ESW_SOURCE_PORT_METADATA_OFFSET); + MLX5_SET(set_action_in, action, length, + ESW_SOURCE_PORT_METADATA_BITS); + + vport->ingress.offloads.modify_metadata = + mlx5_modify_header_alloc(esw->dev, MLX5_FLOW_NAMESPACE_ESW_INGRESS, + 1, action); + if (IS_ERR(vport->ingress.offloads.modify_metadata)) { + err = PTR_ERR(vport->ingress.offloads.modify_metadata); + esw_warn(esw->dev, + "failed to alloc modify header for vport %d ingress acl (%d)\n", + vport->vport, err); + return err; + } + + flow_act.action = MLX5_FLOW_CONTEXT_ACTION_MOD_HDR | MLX5_FLOW_CONTEXT_ACTION_ALLOW; + flow_act.modify_hdr = vport->ingress.offloads.modify_metadata; + vport->ingress.offloads.modify_metadata_rule = + mlx5_add_flow_rules(vport->ingress.acl, + NULL, &flow_act, NULL, 0); + if (IS_ERR(vport->ingress.offloads.modify_metadata_rule)) { + err = PTR_ERR(vport->ingress.offloads.modify_metadata_rule); + esw_warn(esw->dev, + "failed to add setting metadata rule for vport %d ingress acl, err(%d)\n", + vport->vport, err); + mlx5_modify_header_dealloc(esw->dev, vport->ingress.offloads.modify_metadata); + vport->ingress.offloads.modify_metadata_rule = NULL; + } + return err; +} + +static void esw_acl_ingress_mod_metadata_destroy(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + if (!vport->ingress.offloads.modify_metadata_rule) + return; + + mlx5_del_flow_rules(vport->ingress.offloads.modify_metadata_rule); + mlx5_modify_header_dealloc(esw->dev, vport->ingress.offloads.modify_metadata); + vport->ingress.offloads.modify_metadata_rule = NULL; +} + +static int esw_acl_ingress_ofld_rules_create(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + int err; + + if (mlx5_eswitch_vport_match_metadata_enabled(esw)) { + err = esw_acl_ingress_mod_metadata_create(esw, vport); + if (err) { + esw_warn(esw->dev, + "vport(%d) create ingress modify metadata, err(%d)\n", + vport->vport, err); + return err; + } + } + + if (esw_acl_ingress_prio_tag_enabled(esw, vport)) { + err = esw_acl_ingress_prio_tag_create(esw, vport); + if (err) { + esw_warn(esw->dev, + "vport(%d) create ingress prio tag rule, err(%d)\n", + vport->vport, err); + goto prio_tag_err; + } + } + + return 0; + +prio_tag_err: + esw_acl_ingress_mod_metadata_destroy(esw, vport); + return err; +} + +static void esw_acl_ingress_ofld_rules_destroy(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + esw_acl_ingress_allow_rule_destroy(vport); + esw_acl_ingress_mod_metadata_destroy(esw, vport); +} + +static int esw_acl_ingress_ofld_groups_create(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); + struct mlx5_flow_group *g; + void *match_criteria; + u32 *flow_group_in; + u32 flow_index = 0; + int ret = 0; + + flow_group_in = kvzalloc(inlen, GFP_KERNEL); + if (!flow_group_in) + return -ENOMEM; + + if (esw_acl_ingress_prio_tag_enabled(esw, vport)) { + /* This group is to hold FTE to match untagged packets when prio_tag + * is enabled. + */ + match_criteria = MLX5_ADDR_OF(create_flow_group_in, + flow_group_in, match_criteria); + MLX5_SET(create_flow_group_in, flow_group_in, + match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); + MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.cvlan_tag); + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, flow_index); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, flow_index); + + g = mlx5_create_flow_group(vport->ingress.acl, flow_group_in); + if (IS_ERR(g)) { + ret = PTR_ERR(g); + esw_warn(esw->dev, "vport[%d] ingress create untagged flow group, err(%d)\n", + vport->vport, ret); + goto prio_tag_err; + } + vport->ingress.offloads.metadata_prio_tag_grp = g; + flow_index++; + } + + if (mlx5_eswitch_vport_match_metadata_enabled(esw)) { + /* This group holds an FTE with no match to add metadata for + * tagged packets if prio-tag is enabled, or for all untagged + * traffic in case prio-tag is disabled. + */ + memset(flow_group_in, 0, inlen); + MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, flow_index); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, flow_index); + + g = mlx5_create_flow_group(vport->ingress.acl, flow_group_in); + if (IS_ERR(g)) { + ret = PTR_ERR(g); + esw_warn(esw->dev, "vport[%d] ingress create drop flow group, err(%d)\n", + vport->vport, ret); + goto metadata_err; + } + vport->ingress.offloads.metadata_allmatch_grp = g; + } + + kvfree(flow_group_in); + return 0; + +metadata_err: + if (!IS_ERR_OR_NULL(vport->ingress.offloads.metadata_prio_tag_grp)) { + mlx5_destroy_flow_group(vport->ingress.offloads.metadata_prio_tag_grp); + vport->ingress.offloads.metadata_prio_tag_grp = NULL; + } +prio_tag_err: + kvfree(flow_group_in); + return ret; +} + +static void esw_acl_ingress_ofld_groups_destroy(struct mlx5_vport *vport) +{ + if (vport->ingress.offloads.metadata_allmatch_grp) { + mlx5_destroy_flow_group(vport->ingress.offloads.metadata_allmatch_grp); + vport->ingress.offloads.metadata_allmatch_grp = NULL; + } + + if (vport->ingress.offloads.metadata_prio_tag_grp) { + mlx5_destroy_flow_group(vport->ingress.offloads.metadata_prio_tag_grp); + vport->ingress.offloads.metadata_prio_tag_grp = NULL; + } +} + +int esw_acl_ingress_ofld_setup(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + int num_ftes = 0; + int err; + + if (!mlx5_eswitch_vport_match_metadata_enabled(esw) && + !esw_acl_ingress_prio_tag_enabled(esw, vport)) + return 0; + + esw_acl_ingress_allow_rule_destroy(vport); + + if (mlx5_eswitch_vport_match_metadata_enabled(esw)) + num_ftes++; + if (esw_acl_ingress_prio_tag_enabled(esw, vport)) + num_ftes++; + + vport->ingress.acl = esw_acl_table_create(esw, vport->vport, + MLX5_FLOW_NAMESPACE_ESW_INGRESS, + num_ftes); + if (IS_ERR_OR_NULL(vport->ingress.acl)) { + err = PTR_ERR(vport->ingress.acl); + vport->ingress.acl = NULL; + return err; + } + + err = esw_acl_ingress_ofld_groups_create(esw, vport); + if (err) + goto group_err; + + esw_debug(esw->dev, + "vport[%d] configure ingress rules\n", vport->vport); + + err = esw_acl_ingress_ofld_rules_create(esw, vport); + if (err) + goto rules_err; + + return 0; + +rules_err: + esw_acl_ingress_ofld_groups_destroy(vport); +group_err: + esw_acl_ingress_table_destroy(vport); + return err; +} + +void esw_acl_ingress_ofld_cleanup(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + esw_acl_ingress_ofld_rules_destroy(esw, vport); + esw_acl_ingress_ofld_groups_destroy(vport); + esw_acl_ingress_table_destroy(vport); +} + +/* Caller must hold rtnl_lock */ +int mlx5_esw_acl_ingress_vport_bond_update(struct mlx5_eswitch *esw, u16 vport_num, + u32 metadata) +{ + struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num); + int err; + + if (WARN_ON_ONCE(IS_ERR(vport))) { + esw_warn(esw->dev, "vport(%d) invalid!\n", vport_num); + err = PTR_ERR(vport); + goto out; + } + + esw_acl_ingress_ofld_rules_destroy(esw, vport); + + vport->metadata = metadata ? metadata : vport->default_metadata; + + /* Recreate ingress acl rules with vport->metadata */ + err = esw_acl_ingress_ofld_rules_create(esw, vport); + if (err) + goto out; + + return 0; + +out: + vport->metadata = vport->default_metadata; + return err; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/lgcy.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/lgcy.h new file mode 100644 index 000000000000..44c152da3d83 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/lgcy.h @@ -0,0 +1,17 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* Copyright (c) 2020 Mellanox Technologies Inc. All rights reserved. */ + +#ifndef __MLX5_ESWITCH_ACL_LGCY_H__ +#define __MLX5_ESWITCH_ACL_LGCY_H__ + +#include "eswitch.h" + +/* Eswitch acl egress external APIs */ +int esw_acl_egress_lgcy_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport); +void esw_acl_egress_lgcy_cleanup(struct mlx5_eswitch *esw, struct mlx5_vport *vport); + +/* Eswitch acl ingress external APIs */ +int esw_acl_ingress_lgcy_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport); +void esw_acl_ingress_lgcy_cleanup(struct mlx5_eswitch *esw, struct mlx5_vport *vport); + +#endif /* __MLX5_ESWITCH_ACL_LGCY_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ofld.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ofld.h new file mode 100644 index 000000000000..c57869b93d60 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/acl/ofld.h @@ -0,0 +1,29 @@ +/* SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB */ +/* Copyright (c) 2020 Mellanox Technologies Inc. All rights reserved. */ + +#ifndef __MLX5_ESWITCH_ACL_OFLD_H__ +#define __MLX5_ESWITCH_ACL_OFLD_H__ + +#include "eswitch.h" + +/* Eswitch acl egress external APIs */ +int esw_acl_egress_ofld_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport); +void esw_acl_egress_ofld_cleanup(struct mlx5_vport *vport); +int mlx5_esw_acl_egress_vport_bond(struct mlx5_eswitch *esw, u16 active_vport_num, + u16 passive_vport_num); +int mlx5_esw_acl_egress_vport_unbond(struct mlx5_eswitch *esw, u16 vport_num); + +static inline bool mlx5_esw_acl_egress_fwd2vport_supported(struct mlx5_eswitch *esw) +{ + return esw && esw->mode == MLX5_ESWITCH_OFFLOADS && + mlx5_eswitch_vport_match_metadata_enabled(esw) && + MLX5_CAP_ESW_FLOWTABLE(esw->dev, egress_acl_forward_to_vport); +} + +/* Eswitch acl ingress external APIs */ +int esw_acl_ingress_ofld_setup(struct mlx5_eswitch *esw, struct mlx5_vport *vport); +void esw_acl_ingress_ofld_cleanup(struct mlx5_eswitch *esw, struct mlx5_vport *vport); +int mlx5_esw_acl_ingress_vport_bond_update(struct mlx5_eswitch *esw, u16 vport_num, + u32 metadata); + +#endif /* __MLX5_ESWITCH_ACL_OFLD_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.h b/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.h index f8c4239846ea..7679ac359e31 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/chains.h @@ -6,6 +6,8 @@ #include "eswitch.h" +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) + bool mlx5_esw_chains_prios_supported(struct mlx5_eswitch *esw); bool @@ -46,4 +48,21 @@ void mlx5_esw_chains_destroy(struct mlx5_eswitch *esw); int mlx5_eswitch_get_chain_for_tag(struct mlx5_eswitch *esw, u32 tag, u32 *chain); +#else /* CONFIG_MLX5_CLS_ACT */ + +static inline struct mlx5_flow_table * +mlx5_esw_chains_get_table(struct mlx5_eswitch *esw, u32 chain, u32 prio, + u32 level) { return ERR_PTR(-EOPNOTSUPP); } +static inline void +mlx5_esw_chains_put_table(struct mlx5_eswitch *esw, u32 chain, u32 prio, + u32 level) {} + +static inline struct mlx5_flow_table * +mlx5_esw_chains_get_tc_end_ft(struct mlx5_eswitch *esw) { return ERR_PTR(-EOPNOTSUPP); } + +static inline int mlx5_esw_chains_create(struct mlx5_eswitch *esw) { return 0; } +static inline void mlx5_esw_chains_destroy(struct mlx5_eswitch *esw) {} + +#endif /* CONFIG_MLX5_CLS_ACT */ + #endif /* __ML5_ESW_CHAINS_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index c5eb4e7754a9..1116ab9bea6c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -35,6 +35,7 @@ #include <linux/mlx5/mlx5_ifc.h> #include <linux/mlx5/vport.h> #include <linux/mlx5/fs.h> +#include "esw/acl/lgcy.h" #include "mlx5_core.h" #include "lib/eq.h" #include "eswitch.h" @@ -936,512 +937,6 @@ static void esw_vport_change_handler(struct work_struct *work) mutex_unlock(&esw->state_lock); } -int esw_vport_enable_egress_acl(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); - struct mlx5_flow_group *vlan_grp = NULL; - struct mlx5_flow_group *drop_grp = NULL; - struct mlx5_core_dev *dev = esw->dev; - struct mlx5_flow_namespace *root_ns; - struct mlx5_flow_table *acl; - void *match_criteria; - u32 *flow_group_in; - /* The egress acl table contains 2 rules: - * 1)Allow traffic with vlan_tag=vst_vlan_id - * 2)Drop all other traffic. - */ - int table_size = 2; - int err = 0; - - if (!MLX5_CAP_ESW_EGRESS_ACL(dev, ft_support)) - return -EOPNOTSUPP; - - if (!IS_ERR_OR_NULL(vport->egress.acl)) - return 0; - - esw_debug(dev, "Create vport[%d] egress ACL log_max_size(%d)\n", - vport->vport, MLX5_CAP_ESW_EGRESS_ACL(dev, log_max_ft_size)); - - root_ns = mlx5_get_flow_vport_acl_namespace(dev, MLX5_FLOW_NAMESPACE_ESW_EGRESS, - mlx5_eswitch_vport_num_to_index(esw, vport->vport)); - if (!root_ns) { - esw_warn(dev, "Failed to get E-Switch egress flow namespace for vport (%d)\n", vport->vport); - return -EOPNOTSUPP; - } - - flow_group_in = kvzalloc(inlen, GFP_KERNEL); - if (!flow_group_in) - return -ENOMEM; - - acl = mlx5_create_vport_flow_table(root_ns, 0, table_size, 0, vport->vport); - if (IS_ERR(acl)) { - err = PTR_ERR(acl); - esw_warn(dev, "Failed to create E-Switch vport[%d] egress flow Table, err(%d)\n", - vport->vport, err); - goto out; - } - - MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); - match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.cvlan_tag); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.first_vid); - MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); - MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 0); - - vlan_grp = mlx5_create_flow_group(acl, flow_group_in); - if (IS_ERR(vlan_grp)) { - err = PTR_ERR(vlan_grp); - esw_warn(dev, "Failed to create E-Switch vport[%d] egress allowed vlans flow group, err(%d)\n", - vport->vport, err); - goto out; - } - - memset(flow_group_in, 0, inlen); - MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 1); - MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 1); - drop_grp = mlx5_create_flow_group(acl, flow_group_in); - if (IS_ERR(drop_grp)) { - err = PTR_ERR(drop_grp); - esw_warn(dev, "Failed to create E-Switch vport[%d] egress drop flow group, err(%d)\n", - vport->vport, err); - goto out; - } - - vport->egress.acl = acl; - vport->egress.drop_grp = drop_grp; - vport->egress.allowed_vlans_grp = vlan_grp; -out: - kvfree(flow_group_in); - if (err && !IS_ERR_OR_NULL(vlan_grp)) - mlx5_destroy_flow_group(vlan_grp); - if (err && !IS_ERR_OR_NULL(acl)) - mlx5_destroy_flow_table(acl); - return err; -} - -void esw_vport_cleanup_egress_rules(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - if (!IS_ERR_OR_NULL(vport->egress.allowed_vlan)) { - mlx5_del_flow_rules(vport->egress.allowed_vlan); - vport->egress.allowed_vlan = NULL; - } - - if (!IS_ERR_OR_NULL(vport->egress.legacy.drop_rule)) { - mlx5_del_flow_rules(vport->egress.legacy.drop_rule); - vport->egress.legacy.drop_rule = NULL; - } -} - -void esw_vport_disable_egress_acl(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - if (IS_ERR_OR_NULL(vport->egress.acl)) - return; - - esw_debug(esw->dev, "Destroy vport[%d] E-Switch egress ACL\n", vport->vport); - - esw_vport_cleanup_egress_rules(esw, vport); - mlx5_destroy_flow_group(vport->egress.allowed_vlans_grp); - mlx5_destroy_flow_group(vport->egress.drop_grp); - mlx5_destroy_flow_table(vport->egress.acl); - vport->egress.allowed_vlans_grp = NULL; - vport->egress.drop_grp = NULL; - vport->egress.acl = NULL; -} - -static int -esw_vport_create_legacy_ingress_acl_groups(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); - struct mlx5_core_dev *dev = esw->dev; - struct mlx5_flow_group *g; - void *match_criteria; - u32 *flow_group_in; - int err; - - flow_group_in = kvzalloc(inlen, GFP_KERNEL); - if (!flow_group_in) - return -ENOMEM; - - match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, match_criteria); - - MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.cvlan_tag); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.smac_47_16); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.smac_15_0); - MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); - MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 0); - - g = mlx5_create_flow_group(vport->ingress.acl, flow_group_in); - if (IS_ERR(g)) { - err = PTR_ERR(g); - esw_warn(dev, "vport[%d] ingress create untagged spoofchk flow group, err(%d)\n", - vport->vport, err); - goto spoof_err; - } - vport->ingress.legacy.allow_untagged_spoofchk_grp = g; - - memset(flow_group_in, 0, inlen); - MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.cvlan_tag); - MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 1); - MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 1); - - g = mlx5_create_flow_group(vport->ingress.acl, flow_group_in); - if (IS_ERR(g)) { - err = PTR_ERR(g); - esw_warn(dev, "vport[%d] ingress create untagged flow group, err(%d)\n", - vport->vport, err); - goto untagged_err; - } - vport->ingress.legacy.allow_untagged_only_grp = g; - - memset(flow_group_in, 0, inlen); - MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.smac_47_16); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.smac_15_0); - MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 2); - MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 2); - - g = mlx5_create_flow_group(vport->ingress.acl, flow_group_in); - if (IS_ERR(g)) { - err = PTR_ERR(g); - esw_warn(dev, "vport[%d] ingress create spoofchk flow group, err(%d)\n", - vport->vport, err); - goto allow_spoof_err; - } - vport->ingress.legacy.allow_spoofchk_only_grp = g; - - memset(flow_group_in, 0, inlen); - MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 3); - MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, 3); - - g = mlx5_create_flow_group(vport->ingress.acl, flow_group_in); - if (IS_ERR(g)) { - err = PTR_ERR(g); - esw_warn(dev, "vport[%d] ingress create drop flow group, err(%d)\n", - vport->vport, err); - goto drop_err; - } - vport->ingress.legacy.drop_grp = g; - kvfree(flow_group_in); - return 0; - -drop_err: - if (!IS_ERR_OR_NULL(vport->ingress.legacy.allow_spoofchk_only_grp)) { - mlx5_destroy_flow_group(vport->ingress.legacy.allow_spoofchk_only_grp); - vport->ingress.legacy.allow_spoofchk_only_grp = NULL; - } -allow_spoof_err: - if (!IS_ERR_OR_NULL(vport->ingress.legacy.allow_untagged_only_grp)) { - mlx5_destroy_flow_group(vport->ingress.legacy.allow_untagged_only_grp); - vport->ingress.legacy.allow_untagged_only_grp = NULL; - } -untagged_err: - if (!IS_ERR_OR_NULL(vport->ingress.legacy.allow_untagged_spoofchk_grp)) { - mlx5_destroy_flow_group(vport->ingress.legacy.allow_untagged_spoofchk_grp); - vport->ingress.legacy.allow_untagged_spoofchk_grp = NULL; - } -spoof_err: - kvfree(flow_group_in); - return err; -} - -int esw_vport_create_ingress_acl_table(struct mlx5_eswitch *esw, - struct mlx5_vport *vport, int table_size) -{ - struct mlx5_core_dev *dev = esw->dev; - struct mlx5_flow_namespace *root_ns; - struct mlx5_flow_table *acl; - int vport_index; - int err; - - if (!MLX5_CAP_ESW_INGRESS_ACL(dev, ft_support)) - return -EOPNOTSUPP; - - esw_debug(dev, "Create vport[%d] ingress ACL log_max_size(%d)\n", - vport->vport, MLX5_CAP_ESW_INGRESS_ACL(dev, log_max_ft_size)); - - vport_index = mlx5_eswitch_vport_num_to_index(esw, vport->vport); - root_ns = mlx5_get_flow_vport_acl_namespace(dev, MLX5_FLOW_NAMESPACE_ESW_INGRESS, - vport_index); - if (!root_ns) { - esw_warn(dev, "Failed to get E-Switch ingress flow namespace for vport (%d)\n", - vport->vport); - return -EOPNOTSUPP; - } - - acl = mlx5_create_vport_flow_table(root_ns, 0, table_size, 0, vport->vport); - if (IS_ERR(acl)) { - err = PTR_ERR(acl); - esw_warn(dev, "vport[%d] ingress create flow Table, err(%d)\n", - vport->vport, err); - return err; - } - vport->ingress.acl = acl; - return 0; -} - -void esw_vport_destroy_ingress_acl_table(struct mlx5_vport *vport) -{ - if (!vport->ingress.acl) - return; - - mlx5_destroy_flow_table(vport->ingress.acl); - vport->ingress.acl = NULL; -} - -void esw_vport_cleanup_ingress_rules(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - if (vport->ingress.legacy.drop_rule) { - mlx5_del_flow_rules(vport->ingress.legacy.drop_rule); - vport->ingress.legacy.drop_rule = NULL; - } - - if (vport->ingress.allow_rule) { - mlx5_del_flow_rules(vport->ingress.allow_rule); - vport->ingress.allow_rule = NULL; - } -} - -static void esw_vport_disable_legacy_ingress_acl(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - if (!vport->ingress.acl) - return; - - esw_debug(esw->dev, "Destroy vport[%d] E-Switch ingress ACL\n", vport->vport); - - esw_vport_cleanup_ingress_rules(esw, vport); - if (vport->ingress.legacy.allow_spoofchk_only_grp) { - mlx5_destroy_flow_group(vport->ingress.legacy.allow_spoofchk_only_grp); - vport->ingress.legacy.allow_spoofchk_only_grp = NULL; - } - if (vport->ingress.legacy.allow_untagged_only_grp) { - mlx5_destroy_flow_group(vport->ingress.legacy.allow_untagged_only_grp); - vport->ingress.legacy.allow_untagged_only_grp = NULL; - } - if (vport->ingress.legacy.allow_untagged_spoofchk_grp) { - mlx5_destroy_flow_group(vport->ingress.legacy.allow_untagged_spoofchk_grp); - vport->ingress.legacy.allow_untagged_spoofchk_grp = NULL; - } - if (vport->ingress.legacy.drop_grp) { - mlx5_destroy_flow_group(vport->ingress.legacy.drop_grp); - vport->ingress.legacy.drop_grp = NULL; - } - esw_vport_destroy_ingress_acl_table(vport); -} - -static int esw_vport_ingress_config(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - struct mlx5_fc *counter = vport->ingress.legacy.drop_counter; - struct mlx5_flow_destination drop_ctr_dst = {0}; - struct mlx5_flow_destination *dst = NULL; - struct mlx5_flow_act flow_act = {0}; - struct mlx5_flow_spec *spec = NULL; - int dest_num = 0; - int err = 0; - u8 *smac_v; - - /* The ingress acl table contains 4 groups - * (2 active rules at the same time - - * 1 allow rule from one of the first 3 groups. - * 1 drop rule from the last group): - * 1)Allow untagged traffic with smac=original mac. - * 2)Allow untagged traffic. - * 3)Allow traffic with smac=original mac. - * 4)Drop all other traffic. - */ - int table_size = 4; - - esw_vport_cleanup_ingress_rules(esw, vport); - - if (!vport->info.vlan && !vport->info.qos && !vport->info.spoofchk) { - esw_vport_disable_legacy_ingress_acl(esw, vport); - return 0; - } - - if (!vport->ingress.acl) { - err = esw_vport_create_ingress_acl_table(esw, vport, table_size); - if (err) { - esw_warn(esw->dev, - "vport[%d] enable ingress acl err (%d)\n", - err, vport->vport); - return err; - } - - err = esw_vport_create_legacy_ingress_acl_groups(esw, vport); - if (err) - goto out; - } - - esw_debug(esw->dev, - "vport[%d] configure ingress rules, vlan(%d) qos(%d)\n", - vport->vport, vport->info.vlan, vport->info.qos); - - spec = kvzalloc(sizeof(*spec), GFP_KERNEL); - if (!spec) { - err = -ENOMEM; - goto out; - } - - if (vport->info.vlan || vport->info.qos) - MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.cvlan_tag); - - if (vport->info.spoofchk) { - MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.smac_47_16); - MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.smac_15_0); - smac_v = MLX5_ADDR_OF(fte_match_param, - spec->match_value, - outer_headers.smac_47_16); - ether_addr_copy(smac_v, vport->info.mac); - } - - spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - flow_act.action = MLX5_FLOW_CONTEXT_ACTION_ALLOW; - vport->ingress.allow_rule = - mlx5_add_flow_rules(vport->ingress.acl, spec, - &flow_act, NULL, 0); - if (IS_ERR(vport->ingress.allow_rule)) { - err = PTR_ERR(vport->ingress.allow_rule); - esw_warn(esw->dev, - "vport[%d] configure ingress allow rule, err(%d)\n", - vport->vport, err); - vport->ingress.allow_rule = NULL; - goto out; - } - - flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP; - - /* Attach drop flow counter */ - if (counter) { - flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_COUNT; - drop_ctr_dst.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; - drop_ctr_dst.counter_id = mlx5_fc_id(counter); - dst = &drop_ctr_dst; - dest_num++; - } - vport->ingress.legacy.drop_rule = - mlx5_add_flow_rules(vport->ingress.acl, NULL, - &flow_act, dst, dest_num); - if (IS_ERR(vport->ingress.legacy.drop_rule)) { - err = PTR_ERR(vport->ingress.legacy.drop_rule); - esw_warn(esw->dev, - "vport[%d] configure ingress drop rule, err(%d)\n", - vport->vport, err); - vport->ingress.legacy.drop_rule = NULL; - goto out; - } - kvfree(spec); - return 0; - -out: - esw_vport_disable_legacy_ingress_acl(esw, vport); - kvfree(spec); - return err; -} - -int mlx5_esw_create_vport_egress_acl_vlan(struct mlx5_eswitch *esw, - struct mlx5_vport *vport, - u16 vlan_id, u32 flow_action) -{ - struct mlx5_flow_act flow_act = {}; - struct mlx5_flow_spec *spec; - int err = 0; - - if (vport->egress.allowed_vlan) - return -EEXIST; - - spec = kvzalloc(sizeof(*spec), GFP_KERNEL); - if (!spec) - return -ENOMEM; - - MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.cvlan_tag); - MLX5_SET_TO_ONES(fte_match_param, spec->match_value, outer_headers.cvlan_tag); - MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.first_vid); - MLX5_SET(fte_match_param, spec->match_value, outer_headers.first_vid, vlan_id); - - spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - flow_act.action = flow_action; - vport->egress.allowed_vlan = - mlx5_add_flow_rules(vport->egress.acl, spec, - &flow_act, NULL, 0); - if (IS_ERR(vport->egress.allowed_vlan)) { - err = PTR_ERR(vport->egress.allowed_vlan); - esw_warn(esw->dev, - "vport[%d] configure egress vlan rule failed, err(%d)\n", - vport->vport, err); - vport->egress.allowed_vlan = NULL; - } - - kvfree(spec); - return err; -} - -static int esw_vport_egress_config(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - struct mlx5_fc *counter = vport->egress.legacy.drop_counter; - struct mlx5_flow_destination drop_ctr_dst = {0}; - struct mlx5_flow_destination *dst = NULL; - struct mlx5_flow_act flow_act = {0}; - int dest_num = 0; - int err = 0; - - esw_vport_cleanup_egress_rules(esw, vport); - - if (!vport->info.vlan && !vport->info.qos) { - esw_vport_disable_egress_acl(esw, vport); - return 0; - } - - err = esw_vport_enable_egress_acl(esw, vport); - if (err) { - mlx5_core_warn(esw->dev, - "failed to enable egress acl (%d) on vport[%d]\n", - err, vport->vport); - return err; - } - - esw_debug(esw->dev, - "vport[%d] configure egress rules, vlan(%d) qos(%d)\n", - vport->vport, vport->info.vlan, vport->info.qos); - - /* Allowed vlan rule */ - err = mlx5_esw_create_vport_egress_acl_vlan(esw, vport, vport->info.vlan, - MLX5_FLOW_CONTEXT_ACTION_ALLOW); - if (err) - return err; - - flow_act.action = MLX5_FLOW_CONTEXT_ACTION_DROP; - - /* Attach egress drop flow counter */ - if (counter) { - flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_COUNT; - drop_ctr_dst.type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; - drop_ctr_dst.counter_id = mlx5_fc_id(counter); - dst = &drop_ctr_dst; - dest_num++; - } - vport->egress.legacy.drop_rule = - mlx5_add_flow_rules(vport->egress.acl, NULL, - &flow_act, dst, dest_num); - if (IS_ERR(vport->egress.legacy.drop_rule)) { - err = PTR_ERR(vport->egress.legacy.drop_rule); - esw_warn(esw->dev, - "vport[%d] configure egress drop rule failed, err(%d)\n", - vport->vport, err); - vport->egress.legacy.drop_rule = NULL; - } - - return err; -} - static bool element_type_supported(struct mlx5_eswitch *esw, int type) { const struct mlx5_core_dev *dev = esw->dev; @@ -1653,44 +1148,19 @@ static int esw_vport_create_legacy_acl_tables(struct mlx5_eswitch *esw, if (mlx5_esw_is_manager_vport(esw, vport->vport)) return 0; - if (MLX5_CAP_ESW_INGRESS_ACL(esw->dev, flow_counter)) { - vport->ingress.legacy.drop_counter = mlx5_fc_create(esw->dev, false); - if (IS_ERR(vport->ingress.legacy.drop_counter)) { - esw_warn(esw->dev, - "vport[%d] configure ingress drop rule counter failed\n", - vport->vport); - vport->ingress.legacy.drop_counter = NULL; - } - } - - ret = esw_vport_ingress_config(esw, vport); + ret = esw_acl_ingress_lgcy_setup(esw, vport); if (ret) goto ingress_err; - if (MLX5_CAP_ESW_EGRESS_ACL(esw->dev, flow_counter)) { - vport->egress.legacy.drop_counter = mlx5_fc_create(esw->dev, false); - if (IS_ERR(vport->egress.legacy.drop_counter)) { - esw_warn(esw->dev, - "vport[%d] configure egress drop rule counter failed\n", - vport->vport); - vport->egress.legacy.drop_counter = NULL; - } - } - - ret = esw_vport_egress_config(esw, vport); + ret = esw_acl_egress_lgcy_setup(esw, vport); if (ret) goto egress_err; return 0; egress_err: - esw_vport_disable_legacy_ingress_acl(esw, vport); - mlx5_fc_destroy(esw->dev, vport->egress.legacy.drop_counter); - vport->egress.legacy.drop_counter = NULL; - + esw_acl_ingress_lgcy_cleanup(esw, vport); ingress_err: - mlx5_fc_destroy(esw->dev, vport->ingress.legacy.drop_counter); - vport->ingress.legacy.drop_counter = NULL; return ret; } @@ -1710,13 +1180,8 @@ static void esw_vport_destroy_legacy_acl_tables(struct mlx5_eswitch *esw, if (mlx5_esw_is_manager_vport(esw, vport->vport)) return; - esw_vport_disable_egress_acl(esw, vport); - mlx5_fc_destroy(esw->dev, vport->egress.legacy.drop_counter); - vport->egress.legacy.drop_counter = NULL; - - esw_vport_disable_legacy_ingress_acl(esw, vport); - mlx5_fc_destroy(esw->dev, vport->ingress.legacy.drop_counter); - vport->ingress.legacy.drop_counter = NULL; + esw_acl_egress_lgcy_cleanup(esw, vport); + esw_acl_ingress_lgcy_cleanup(esw, vport); } static void esw_vport_cleanup_acl(struct mlx5_eswitch *esw, @@ -2262,7 +1727,10 @@ int mlx5_eswitch_init(struct mlx5_core_dev *dev) hash_init(esw->offloads.encap_tbl); mutex_init(&esw->offloads.mod_hdr.lock); hash_init(esw->offloads.mod_hdr.hlist); + mutex_init(&esw->offloads.decap_tbl_lock); + hash_init(esw->offloads.decap_tbl); atomic64_set(&esw->offloads.num_flows, 0); + ida_init(&esw->offloads.vport_metadata_ida); mutex_init(&esw->state_lock); mutex_init(&esw->mode_lock); @@ -2301,8 +1769,10 @@ void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw) esw_offloads_cleanup_reps(esw); mutex_destroy(&esw->mode_lock); mutex_destroy(&esw->state_lock); + ida_destroy(&esw->offloads.vport_metadata_ida); mutex_destroy(&esw->offloads.mod_hdr.lock); mutex_destroy(&esw->offloads.encap_tbl_lock); + mutex_destroy(&esw->offloads.decap_tbl_lock); kfree(esw->vports); kfree(esw); } @@ -2345,7 +1815,7 @@ int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw, ether_addr_copy(evport->info.mac, mac); evport->info.node_guid = node_guid; if (evport->enabled && esw->mode == MLX5_ESWITCH_LEGACY) - err = esw_vport_ingress_config(esw, evport); + err = esw_acl_ingress_lgcy_setup(esw, evport); unlock: mutex_unlock(&esw->state_lock); @@ -2427,10 +1897,10 @@ int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw, evport->info.vlan = vlan; evport->info.qos = qos; if (evport->enabled && esw->mode == MLX5_ESWITCH_LEGACY) { - err = esw_vport_ingress_config(esw, evport); + err = esw_acl_ingress_lgcy_setup(esw, evport); if (err) return err; - err = esw_vport_egress_config(esw, evport); + err = esw_acl_egress_lgcy_setup(esw, evport); } return err; @@ -2472,7 +1942,7 @@ int mlx5_eswitch_set_vport_spoofchk(struct mlx5_eswitch *esw, "Spoofchk in set while MAC is invalid, vport(%d)\n", evport->vport); if (evport->enabled && esw->mode == MLX5_ESWITCH_LEGACY) - err = esw_vport_ingress_config(esw, evport); + err = esw_acl_ingress_lgcy_setup(esw, evport); if (err) evport->info.spoofchk = pschk; mutex_unlock(&esw->state_lock); @@ -2731,7 +2201,7 @@ static int mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev, if (!vport->enabled) goto unlock; - if (vport->egress.legacy.drop_counter) + if (!IS_ERR_OR_NULL(vport->egress.legacy.drop_counter)) mlx5_fc_query(dev, vport->egress.legacy.drop_counter, &stats->rx_dropped, &bytes); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 4a1c6c78bb14..a5175e98c0b3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -99,13 +99,19 @@ struct vport_ingress { struct vport_egress { struct mlx5_flow_table *acl; - struct mlx5_flow_group *allowed_vlans_grp; - struct mlx5_flow_group *drop_grp; struct mlx5_flow_handle *allowed_vlan; - struct { - struct mlx5_flow_handle *drop_rule; - struct mlx5_fc *drop_counter; - } legacy; + struct mlx5_flow_group *vlan_grp; + union { + struct { + struct mlx5_flow_group *drop_grp; + struct mlx5_flow_handle *drop_rule; + struct mlx5_fc *drop_counter; + } legacy; + struct { + struct mlx5_flow_group *fwd_grp; + struct mlx5_flow_handle *fwd_rule; + } offloads; + }; }; struct mlx5_vport_drop_stats { @@ -143,6 +149,8 @@ struct mlx5_vport { struct vport_ingress ingress; struct vport_egress egress; + u32 default_metadata; + u32 metadata; struct mlx5_vport_info info; @@ -209,6 +217,8 @@ struct mlx5_esw_offload { struct mutex peer_mutex; struct mutex encap_tbl_lock; /* protects encap_tbl */ DECLARE_HASHTABLE(encap_tbl, 8); + struct mutex decap_tbl_lock; /* protects decap_tbl */ + DECLARE_HASHTABLE(decap_tbl, 8); struct mod_hdr_tbl mod_hdr; DECLARE_HASHTABLE(termtbl_tbl, 8); struct mutex termtbl_mutex; /* protects termtbl hash */ @@ -216,6 +226,7 @@ struct mlx5_esw_offload { u8 inline_mode; atomic64_t num_flows; enum devlink_eswitch_encap_mode encap; + struct ida vport_metadata_ida; }; /* E-Switch MC FDB table hash node */ @@ -283,18 +294,10 @@ void esw_offloads_disable(struct mlx5_eswitch *esw); int esw_offloads_enable(struct mlx5_eswitch *esw); void esw_offloads_cleanup_reps(struct mlx5_eswitch *esw); int esw_offloads_init_reps(struct mlx5_eswitch *esw); -void esw_vport_cleanup_ingress_rules(struct mlx5_eswitch *esw, - struct mlx5_vport *vport); -int esw_vport_create_ingress_acl_table(struct mlx5_eswitch *esw, - struct mlx5_vport *vport, - int table_size); -void esw_vport_destroy_ingress_acl_table(struct mlx5_vport *vport); -void esw_vport_cleanup_egress_rules(struct mlx5_eswitch *esw, - struct mlx5_vport *vport); -int esw_vport_enable_egress_acl(struct mlx5_eswitch *esw, - struct mlx5_vport *vport); -void esw_vport_disable_egress_acl(struct mlx5_eswitch *esw, - struct mlx5_vport *vport); + +u32 mlx5_esw_match_metadata_alloc(struct mlx5_eswitch *esw); +void mlx5_esw_match_metadata_free(struct mlx5_eswitch *esw, u32 metadata); + int mlx5_esw_modify_vport_rate(struct mlx5_eswitch *esw, u16 vport_num, u32 rate_mbps); @@ -432,6 +435,7 @@ struct mlx5_esw_flow_attr { struct mlx5_flow_table *fdb; struct mlx5_flow_table *dest_ft; struct mlx5_ct_attr ct_attr; + struct mlx5_pkt_reformat *decap_pkt_reformat; struct mlx5e_tc_flow_parse_attr *parse_attr; }; @@ -455,10 +459,6 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw, int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw, u16 vport, u16 vlan, u8 qos, u8 set_flags); -int mlx5_esw_create_vport_egress_acl_vlan(struct mlx5_eswitch *esw, - struct mlx5_vport *vport, - u16 vlan_id, u32 flow_action); - static inline bool mlx5_esw_qos_enabled(struct mlx5_eswitch *esw) { return esw->qos.enabled; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 57ac2ef52e80..060354bb211a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -31,12 +31,14 @@ */ #include <linux/etherdevice.h> +#include <linux/idr.h> #include <linux/mlx5/driver.h> #include <linux/mlx5/mlx5_ifc.h> #include <linux/mlx5/vport.h> #include <linux/mlx5/fs.h> #include "mlx5_core.h" #include "eswitch.h" +#include "esw/acl/ofld.h" #include "esw/chains.h" #include "rdma.h" #include "en.h" @@ -234,13 +236,6 @@ static struct mlx5_eswitch_rep *mlx5_eswitch_get_rep(struct mlx5_eswitch *esw, return &esw->offloads.vport_reps[idx]; } -static bool -esw_check_ingress_prio_tag_enabled(const struct mlx5_eswitch *esw, - const struct mlx5_vport *vport) -{ - return (MLX5_CAP_GEN(esw->dev, prio_tag_required) && - mlx5_eswitch_is_vf_vport(esw, vport->vport)); -} static void mlx5_eswitch_set_rule_source_port(struct mlx5_eswitch *esw, @@ -366,6 +361,10 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, } } } + + if (attr->decap_pkt_reformat) + flow_act.pkt_reformat = attr->decap_pkt_reformat; + if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_COUNT) { dest[i].type = MLX5_FLOW_DESTINATION_TYPE_COUNTER; dest[i].counter_id = mlx5_fc_id(attr->counter); @@ -1727,7 +1726,9 @@ static int mlx5_esw_offloads_pair(struct mlx5_eswitch *esw, static void mlx5_esw_offloads_unpair(struct mlx5_eswitch *esw) { +#if IS_ENABLED(CONFIG_MLX5_CLS_ACT) mlx5e_tc_clean_fdb_peer_flows(esw); +#endif esw_del_fdb_peer_miss_rules(esw); } @@ -1845,279 +1846,6 @@ static void esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) mlx5_devcom_unregister_component(devcom, MLX5_DEVCOM_ESW_OFFLOADS); } -static int esw_vport_ingress_prio_tag_config(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - struct mlx5_flow_act flow_act = {0}; - struct mlx5_flow_spec *spec; - int err = 0; - - /* For prio tag mode, there is only 1 FTEs: - * 1) Untagged packets - push prio tag VLAN and modify metadata if - * required, allow - * Unmatched traffic is allowed by default - */ - spec = kvzalloc(sizeof(*spec), GFP_KERNEL); - if (!spec) - return -ENOMEM; - - /* Untagged packets - push prio tag VLAN, allow */ - MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.cvlan_tag); - MLX5_SET(fte_match_param, spec->match_value, outer_headers.cvlan_tag, 0); - spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; - flow_act.action = MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH | - MLX5_FLOW_CONTEXT_ACTION_ALLOW; - flow_act.vlan[0].ethtype = ETH_P_8021Q; - flow_act.vlan[0].vid = 0; - flow_act.vlan[0].prio = 0; - - if (vport->ingress.offloads.modify_metadata_rule) { - flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_MOD_HDR; - flow_act.modify_hdr = vport->ingress.offloads.modify_metadata; - } - - vport->ingress.allow_rule = - mlx5_add_flow_rules(vport->ingress.acl, spec, - &flow_act, NULL, 0); - if (IS_ERR(vport->ingress.allow_rule)) { - err = PTR_ERR(vport->ingress.allow_rule); - esw_warn(esw->dev, - "vport[%d] configure ingress untagged allow rule, err(%d)\n", - vport->vport, err); - vport->ingress.allow_rule = NULL; - } - - kvfree(spec); - return err; -} - -static int esw_vport_add_ingress_acl_modify_metadata(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - u8 action[MLX5_UN_SZ_BYTES(set_add_copy_action_in_auto)] = {}; - struct mlx5_flow_act flow_act = {}; - int err = 0; - u32 key; - - key = mlx5_eswitch_get_vport_metadata_for_match(esw, vport->vport); - key >>= ESW_SOURCE_PORT_METADATA_OFFSET; - - MLX5_SET(set_action_in, action, action_type, MLX5_ACTION_TYPE_SET); - MLX5_SET(set_action_in, action, field, - MLX5_ACTION_IN_FIELD_METADATA_REG_C_0); - MLX5_SET(set_action_in, action, data, key); - MLX5_SET(set_action_in, action, offset, - ESW_SOURCE_PORT_METADATA_OFFSET); - MLX5_SET(set_action_in, action, length, - ESW_SOURCE_PORT_METADATA_BITS); - - vport->ingress.offloads.modify_metadata = - mlx5_modify_header_alloc(esw->dev, MLX5_FLOW_NAMESPACE_ESW_INGRESS, - 1, action); - if (IS_ERR(vport->ingress.offloads.modify_metadata)) { - err = PTR_ERR(vport->ingress.offloads.modify_metadata); - esw_warn(esw->dev, - "failed to alloc modify header for vport %d ingress acl (%d)\n", - vport->vport, err); - return err; - } - - flow_act.action = MLX5_FLOW_CONTEXT_ACTION_MOD_HDR | MLX5_FLOW_CONTEXT_ACTION_ALLOW; - flow_act.modify_hdr = vport->ingress.offloads.modify_metadata; - vport->ingress.offloads.modify_metadata_rule = - mlx5_add_flow_rules(vport->ingress.acl, - NULL, &flow_act, NULL, 0); - if (IS_ERR(vport->ingress.offloads.modify_metadata_rule)) { - err = PTR_ERR(vport->ingress.offloads.modify_metadata_rule); - esw_warn(esw->dev, - "failed to add setting metadata rule for vport %d ingress acl, err(%d)\n", - vport->vport, err); - mlx5_modify_header_dealloc(esw->dev, vport->ingress.offloads.modify_metadata); - vport->ingress.offloads.modify_metadata_rule = NULL; - } - return err; -} - -static void esw_vport_del_ingress_acl_modify_metadata(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - if (vport->ingress.offloads.modify_metadata_rule) { - mlx5_del_flow_rules(vport->ingress.offloads.modify_metadata_rule); - mlx5_modify_header_dealloc(esw->dev, vport->ingress.offloads.modify_metadata); - - vport->ingress.offloads.modify_metadata_rule = NULL; - } -} - -static int esw_vport_create_ingress_acl_group(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); - struct mlx5_flow_group *g; - void *match_criteria; - u32 *flow_group_in; - u32 flow_index = 0; - int ret = 0; - - flow_group_in = kvzalloc(inlen, GFP_KERNEL); - if (!flow_group_in) - return -ENOMEM; - - if (esw_check_ingress_prio_tag_enabled(esw, vport)) { - /* This group is to hold FTE to match untagged packets when prio_tag - * is enabled. - */ - memset(flow_group_in, 0, inlen); - - match_criteria = MLX5_ADDR_OF(create_flow_group_in, - flow_group_in, match_criteria); - MLX5_SET(create_flow_group_in, flow_group_in, - match_criteria_enable, MLX5_MATCH_OUTER_HEADERS); - MLX5_SET_TO_ONES(fte_match_param, match_criteria, outer_headers.cvlan_tag); - MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, flow_index); - MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, flow_index); - - g = mlx5_create_flow_group(vport->ingress.acl, flow_group_in); - if (IS_ERR(g)) { - ret = PTR_ERR(g); - esw_warn(esw->dev, "vport[%d] ingress create untagged flow group, err(%d)\n", - vport->vport, ret); - goto prio_tag_err; - } - vport->ingress.offloads.metadata_prio_tag_grp = g; - flow_index++; - } - - if (mlx5_eswitch_vport_match_metadata_enabled(esw)) { - /* This group holds an FTE with no matches for add metadata for - * tagged packets, if prio-tag is enabled (as a fallthrough), - * or all traffic in case prio-tag is disabled. - */ - memset(flow_group_in, 0, inlen); - MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, flow_index); - MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, flow_index); - - g = mlx5_create_flow_group(vport->ingress.acl, flow_group_in); - if (IS_ERR(g)) { - ret = PTR_ERR(g); - esw_warn(esw->dev, "vport[%d] ingress create drop flow group, err(%d)\n", - vport->vport, ret); - goto metadata_err; - } - vport->ingress.offloads.metadata_allmatch_grp = g; - } - - kvfree(flow_group_in); - return 0; - -metadata_err: - if (!IS_ERR_OR_NULL(vport->ingress.offloads.metadata_prio_tag_grp)) { - mlx5_destroy_flow_group(vport->ingress.offloads.metadata_prio_tag_grp); - vport->ingress.offloads.metadata_prio_tag_grp = NULL; - } -prio_tag_err: - kvfree(flow_group_in); - return ret; -} - -static void esw_vport_destroy_ingress_acl_group(struct mlx5_vport *vport) -{ - if (vport->ingress.offloads.metadata_allmatch_grp) { - mlx5_destroy_flow_group(vport->ingress.offloads.metadata_allmatch_grp); - vport->ingress.offloads.metadata_allmatch_grp = NULL; - } - - if (vport->ingress.offloads.metadata_prio_tag_grp) { - mlx5_destroy_flow_group(vport->ingress.offloads.metadata_prio_tag_grp); - vport->ingress.offloads.metadata_prio_tag_grp = NULL; - } -} - -static int esw_vport_ingress_config(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - int num_ftes = 0; - int err; - - if (!mlx5_eswitch_vport_match_metadata_enabled(esw) && - !esw_check_ingress_prio_tag_enabled(esw, vport)) - return 0; - - esw_vport_cleanup_ingress_rules(esw, vport); - - if (mlx5_eswitch_vport_match_metadata_enabled(esw)) - num_ftes++; - if (esw_check_ingress_prio_tag_enabled(esw, vport)) - num_ftes++; - - err = esw_vport_create_ingress_acl_table(esw, vport, num_ftes); - if (err) { - esw_warn(esw->dev, - "failed to enable ingress acl (%d) on vport[%d]\n", - err, vport->vport); - return err; - } - - err = esw_vport_create_ingress_acl_group(esw, vport); - if (err) - goto group_err; - - esw_debug(esw->dev, - "vport[%d] configure ingress rules\n", vport->vport); - - if (mlx5_eswitch_vport_match_metadata_enabled(esw)) { - err = esw_vport_add_ingress_acl_modify_metadata(esw, vport); - if (err) - goto metadata_err; - } - - if (esw_check_ingress_prio_tag_enabled(esw, vport)) { - err = esw_vport_ingress_prio_tag_config(esw, vport); - if (err) - goto prio_tag_err; - } - return 0; - -prio_tag_err: - esw_vport_del_ingress_acl_modify_metadata(esw, vport); -metadata_err: - esw_vport_destroy_ingress_acl_group(vport); -group_err: - esw_vport_destroy_ingress_acl_table(vport); - return err; -} - -static int esw_vport_egress_config(struct mlx5_eswitch *esw, - struct mlx5_vport *vport) -{ - int err; - - if (!MLX5_CAP_GEN(esw->dev, prio_tag_required)) - return 0; - - esw_vport_cleanup_egress_rules(esw, vport); - - err = esw_vport_enable_egress_acl(esw, vport); - if (err) - return err; - - /* For prio tag mode, there is only 1 FTEs: - * 1) prio tag packets - pop the prio tag VLAN, allow - * Unmatched traffic is allowed by default - */ - esw_debug(esw->dev, - "vport[%d] configure prio tag egress rules\n", vport->vport); - - /* prio tag vlan rule - pop it so VF receives untagged packets */ - err = mlx5_esw_create_vport_egress_acl_vlan(esw, vport, 0, - MLX5_FLOW_CONTEXT_ACTION_VLAN_POP | - MLX5_FLOW_CONTEXT_ACTION_ALLOW); - if (err) - esw_vport_disable_egress_acl(esw, vport); - - return err; -} - static bool esw_check_vport_match_metadata_supported(const struct mlx5_eswitch *esw) { @@ -2150,25 +1878,83 @@ static bool esw_use_vport_metadata(const struct mlx5_eswitch *esw) esw_check_vport_match_metadata_supported(esw); } +u32 mlx5_esw_match_metadata_alloc(struct mlx5_eswitch *esw) +{ + u32 num_vports = GENMASK(ESW_VPORT_BITS - 1, 0) - 1; + u32 vhca_id_mask = GENMASK(ESW_VHCA_ID_BITS - 1, 0); + u32 vhca_id = MLX5_CAP_GEN(esw->dev, vhca_id); + u32 start; + u32 end; + int id; + + /* Make sure the vhca_id fits the ESW_VHCA_ID_BITS */ + WARN_ON_ONCE(vhca_id >= BIT(ESW_VHCA_ID_BITS)); + + /* Trim vhca_id to ESW_VHCA_ID_BITS */ + vhca_id &= vhca_id_mask; + + start = (vhca_id << ESW_VPORT_BITS); + end = start + num_vports; + if (!vhca_id) + start += 1; /* zero is reserved/invalid metadata */ + id = ida_alloc_range(&esw->offloads.vport_metadata_ida, start, end, GFP_KERNEL); + + return (id < 0) ? 0 : id; +} + +void mlx5_esw_match_metadata_free(struct mlx5_eswitch *esw, u32 metadata) +{ + ida_free(&esw->offloads.vport_metadata_ida, metadata); +} + +static int esw_offloads_vport_metadata_setup(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + if (vport->vport == MLX5_VPORT_UPLINK) + return 0; + + vport->default_metadata = mlx5_esw_match_metadata_alloc(esw); + vport->metadata = vport->default_metadata; + return vport->metadata ? 0 : -ENOSPC; +} + +static void esw_offloads_vport_metadata_cleanup(struct mlx5_eswitch *esw, + struct mlx5_vport *vport) +{ + if (vport->vport == MLX5_VPORT_UPLINK || !vport->default_metadata) + return; + + WARN_ON(vport->metadata != vport->default_metadata); + mlx5_esw_match_metadata_free(esw, vport->default_metadata); +} + int esw_vport_create_offloads_acl_tables(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { int err; - err = esw_vport_ingress_config(esw, vport); + err = esw_offloads_vport_metadata_setup(esw, vport); if (err) - return err; + goto metadata_err; + + err = esw_acl_ingress_ofld_setup(esw, vport); + if (err) + goto ingress_err; if (mlx5_eswitch_is_vf_vport(esw, vport->vport)) { - err = esw_vport_egress_config(esw, vport); - if (err) { - esw_vport_cleanup_ingress_rules(esw, vport); - esw_vport_del_ingress_acl_modify_metadata(esw, vport); - esw_vport_destroy_ingress_acl_group(vport); - esw_vport_destroy_ingress_acl_table(vport); - } + err = esw_acl_egress_ofld_setup(esw, vport); + if (err) + goto egress_err; } + + return 0; + +egress_err: + esw_acl_ingress_ofld_cleanup(esw, vport); +ingress_err: + esw_offloads_vport_metadata_cleanup(esw, vport); +metadata_err: return err; } @@ -2176,11 +1962,9 @@ void esw_vport_destroy_offloads_acl_tables(struct mlx5_eswitch *esw, struct mlx5_vport *vport) { - esw_vport_disable_egress_acl(esw, vport); - esw_vport_cleanup_ingress_rules(esw, vport); - esw_vport_del_ingress_acl_modify_metadata(esw, vport); - esw_vport_destroy_ingress_acl_group(vport); - esw_vport_destroy_ingress_acl_table(vport); + esw_acl_egress_ofld_cleanup(vport); + esw_acl_ingress_ofld_cleanup(esw, vport); + esw_offloads_vport_metadata_cleanup(esw, vport); } static int esw_create_uplink_offloads_acl_tables(struct mlx5_eswitch *esw) @@ -2846,38 +2630,11 @@ EXPORT_SYMBOL(mlx5_eswitch_vport_match_metadata_enabled); u32 mlx5_eswitch_get_vport_metadata_for_match(struct mlx5_eswitch *esw, u16 vport_num) { - u32 vport_num_mask = GENMASK(ESW_VPORT_BITS - 1, 0); - u32 vhca_id_mask = GENMASK(ESW_VHCA_ID_BITS - 1, 0); - u32 vhca_id = MLX5_CAP_GEN(esw->dev, vhca_id); - u32 val; + struct mlx5_vport *vport = mlx5_eswitch_get_vport(esw, vport_num); - /* Make sure the vhca_id fits the ESW_VHCA_ID_BITS */ - WARN_ON_ONCE(vhca_id >= BIT(ESW_VHCA_ID_BITS)); - - /* Trim vhca_id to ESW_VHCA_ID_BITS */ - vhca_id &= vhca_id_mask; - - /* Make sure pf and ecpf map to end of ESW_VPORT_BITS range so they - * don't overlap with VF numbers, and themselves, after trimming. - */ - WARN_ON_ONCE((MLX5_VPORT_UPLINK & vport_num_mask) < - vport_num_mask - 1); - WARN_ON_ONCE((MLX5_VPORT_ECPF & vport_num_mask) < - vport_num_mask - 1); - WARN_ON_ONCE((MLX5_VPORT_UPLINK & vport_num_mask) == - (MLX5_VPORT_ECPF & vport_num_mask)); - - /* Make sure that the VF vport_num fits ESW_VPORT_BITS and don't - * overlap with pf and ecpf. - */ - if (vport_num != MLX5_VPORT_UPLINK && - vport_num != MLX5_VPORT_ECPF) - WARN_ON_ONCE(vport_num >= vport_num_mask - 1); - - /* We can now trim vport_num to ESW_VPORT_BITS */ - vport_num &= vport_num_mask; + if (WARN_ON_ONCE(IS_ERR(vport))) + return 0; - val = (vhca_id << ESW_VPORT_BITS) | vport_num; - return val << (32 - ESW_SOURCE_PORT_METADATA_BITS); + return vport->metadata << (32 - ESW_SOURCE_PORT_METADATA_BITS); } EXPORT_SYMBOL(mlx5_eswitch_get_vport_metadata_for_match); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/events.c b/drivers/net/ethernet/mellanox/mlx5/core/events.c index 8bcf3426b9c6..3ce17c3d7a00 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/events.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/events.c @@ -346,8 +346,10 @@ int mlx5_events_init(struct mlx5_core_dev *dev) events->dev = dev; dev->priv.events = events; events->wq = create_singlethread_workqueue("mlx5_events"); - if (!events->wq) + if (!events->wq) { + kfree(events); return -ENOMEM; + } INIT_WORK(&events->pcie_core_work, mlx5_pcie_event); return 0; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c index b794888fa3ba..b463787d6ca1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c @@ -65,6 +65,7 @@ struct mlx5_fpga_esp_xfrm; struct mlx5_fpga_ipsec_sa_ctx { struct rhash_head hash; struct mlx5_ifc_fpga_ipsec_sa hw_sa; + u32 sa_handle; struct mlx5_core_dev *dev; struct mlx5_fpga_esp_xfrm *fpga_xfrm; }; @@ -119,6 +120,8 @@ struct mlx5_fpga_ipsec { */ struct rb_root rules_rb; struct mutex rules_rb_lock; /* rules lock */ + + struct ida halloc; }; static bool mlx5_fpga_is_ipsec_device(struct mlx5_core_dev *mdev) @@ -602,7 +605,7 @@ static bool mlx5_is_fpga_ipsec_rule(struct mlx5_core_dev *dev, const u32 *match_c, const u32 *match_v) { - u32 ipsec_dev_caps = mlx5_accel_ipsec_device_caps(dev); + u32 ipsec_dev_caps = mlx5_fpga_ipsec_device_caps(dev); bool ipv6_flow; ipv6_flow = mlx5_fs_is_outer_ipv6_flow(dev, match_c, match_v); @@ -666,7 +669,8 @@ void *mlx5_fpga_ipsec_create_sa_ctx(struct mlx5_core_dev *mdev, struct mlx5_accel_esp_xfrm *accel_xfrm, const __be32 saddr[4], const __be32 daddr[4], - const __be32 spi, bool is_ipv6) + const __be32 spi, bool is_ipv6, + u32 *sa_handle) { struct mlx5_fpga_ipsec_sa_ctx *sa_ctx; struct mlx5_fpga_esp_xfrm *fpga_xfrm = @@ -704,6 +708,17 @@ void *mlx5_fpga_ipsec_create_sa_ctx(struct mlx5_core_dev *mdev, goto exists; } + if (accel_xfrm->attrs.action == MLX5_ACCEL_ESP_ACTION_DECRYPT) { + err = ida_simple_get(&fipsec->halloc, 1, 0, GFP_KERNEL); + if (err < 0) { + context = ERR_PTR(err); + goto exists; + } + + sa_ctx->sa_handle = err; + if (sa_handle) + *sa_handle = sa_ctx->sa_handle; + } /* This is unbounded fpga_xfrm, try to add to hash */ mutex_lock(&fipsec->sa_hash_lock); @@ -744,7 +759,8 @@ delete_hash: rhash_sa)); unlock_hash: mutex_unlock(&fipsec->sa_hash_lock); - + if (accel_xfrm->attrs.action == MLX5_ACCEL_ESP_ACTION_DECRYPT) + ida_simple_remove(&fipsec->halloc, sa_ctx->sa_handle); exists: mutex_unlock(&fpga_xfrm->lock); kfree(sa_ctx); @@ -816,7 +832,7 @@ mlx5_fpga_ipsec_fs_create_sa_ctx(struct mlx5_core_dev *mdev, /* create */ return mlx5_fpga_ipsec_create_sa_ctx(mdev, accel_xfrm, saddr, daddr, - spi, is_ipv6); + spi, is_ipv6, NULL); } static void @@ -836,6 +852,10 @@ mlx5_fpga_ipsec_release_sa_ctx(struct mlx5_fpga_ipsec_sa_ctx *sa_ctx) return; } + if (sa_ctx->fpga_xfrm->accel_xfrm.attrs.action & + MLX5_ACCEL_ESP_ACTION_DECRYPT) + ida_simple_remove(&fipsec->halloc, sa_ctx->sa_handle); + mutex_lock(&fipsec->sa_hash_lock); WARN_ON(rhashtable_remove_fast(&fipsec->sa_hash, &sa_ctx->hash, rhash_sa)); @@ -1299,6 +1319,8 @@ int mlx5_fpga_ipsec_init(struct mlx5_core_dev *mdev) goto err_destroy_hash; } + ida_init(&fdev->ipsec->halloc); + return 0; err_destroy_hash: @@ -1331,6 +1353,7 @@ void mlx5_fpga_ipsec_cleanup(struct mlx5_core_dev *mdev) if (!mlx5_fpga_is_ipsec_device(mdev)) return; + ida_destroy(&fdev->ipsec->halloc); destroy_rules_rb(&fdev->ipsec->rules_rb); rhashtable_destroy(&fdev->ipsec->sa_hash); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.h index 382985e65b48..9ba637f0f0f2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.h @@ -37,6 +37,7 @@ #include "accel/ipsec.h" #include "fs_cmd.h" +#ifdef CONFIG_MLX5_FPGA_IPSEC u32 mlx5_fpga_ipsec_device_caps(struct mlx5_core_dev *mdev); unsigned int mlx5_fpga_ipsec_counters_count(struct mlx5_core_dev *mdev); int mlx5_fpga_ipsec_counters_read(struct mlx5_core_dev *mdev, u64 *counters, @@ -46,7 +47,8 @@ void *mlx5_fpga_ipsec_create_sa_ctx(struct mlx5_core_dev *mdev, struct mlx5_accel_esp_xfrm *accel_xfrm, const __be32 saddr[4], const __be32 daddr[4], - const __be32 spi, bool is_ipv6); + const __be32 spi, bool is_ipv6, + u32 *sa_handle); void mlx5_fpga_ipsec_delete_sa_ctx(void *context); int mlx5_fpga_ipsec_init(struct mlx5_core_dev *mdev); @@ -63,5 +65,17 @@ int mlx5_fpga_esp_modify_xfrm(struct mlx5_accel_esp_xfrm *xfrm, const struct mlx5_flow_cmds * mlx5_fs_cmd_get_default_ipsec_fpga_cmds(enum fs_flow_table_type type); +#else +static inline u32 mlx5_fpga_ipsec_device_caps(struct mlx5_core_dev *mdev) +{ + return 0; +} -#endif /* __MLX5_FPGA_SADB_H__ */ +static inline const struct mlx5_flow_cmds * +mlx5_fs_cmd_get_default_ipsec_fpga_cmds(enum fs_flow_table_type type) +{ + return mlx5_fs_cmd_get_default(type); +} + +#endif /* CONFIG_MLX5_FPGA_IPSEC */ +#endif /* __MLX5_FPGA_IPSEC_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c index 22a2ef111514..29b7339ebfa3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c @@ -194,8 +194,8 @@ static void mlx5_fpga_tls_flow_to_cmd(void *flow, void *cmd) MLX5_GET(tls_flow, flow, direction_sx)); } -int mlx5_fpga_tls_resync_rx(struct mlx5_core_dev *mdev, u32 handle, u32 seq, - u64 rcd_sn) +int mlx5_fpga_tls_resync_rx(struct mlx5_core_dev *mdev, __be32 handle, + u32 seq, __be64 rcd_sn) { struct mlx5_fpga_dma_buf *buf; int size = sizeof(*buf) + MLX5_TLS_COMMAND_SIZE; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.h b/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.h index 3b2e37bf76fe..5714cf391d1b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.h @@ -68,7 +68,7 @@ static inline u32 mlx5_fpga_tls_device_caps(struct mlx5_core_dev *mdev) return mdev->fpga->tls->caps; } -int mlx5_fpga_tls_resync_rx(struct mlx5_core_dev *mdev, u32 handle, u32 seq, - u64 rcd_sn); +int mlx5_fpga_tls_resync_rx(struct mlx5_core_dev *mdev, __be32 handle, + u32 seq, __be64 rcd_sn); #endif /* __MLX5_FPGA_TLS_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 41aa1fa0c69e..13e2fb79c21a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -344,17 +344,12 @@ static void tree_put_node(struct fs_node *node, bool locked) if (node->del_hw_func) node->del_hw_func(node); if (parent_node) { - /* Only root namespace doesn't have parent and we just - * need to free its node. - */ down_write_ref_node(parent_node, locked); list_del_init(&node->list); - if (node->del_sw_func) - node->del_sw_func(node); - up_write_ref_node(parent_node, locked); - } else { - kfree(node); } + node->del_sw_func(node); + if (parent_node) + up_write_ref_node(parent_node, locked); node = NULL; } if (!node && parent_node) @@ -474,8 +469,10 @@ static void del_sw_flow_table(struct fs_node *node) fs_get_obj(ft, node); rhltable_destroy(&ft->fgs_hash); - fs_get_obj(prio, ft->node.parent); - prio->num_ft--; + if (ft->node.parent) { + fs_get_obj(prio, ft->node.parent); + prio->num_ft--; + } kfree(ft); } @@ -1796,11 +1793,13 @@ skip_search: list_for_each_entry(iter, match_head, list) { g = iter->g; - if (!g->node.active) - continue; - nested_down_write_ref_node(&g->node, FS_LOCK_PARENT); + if (!g->node.active) { + up_write_ref_node(&g->node, false); + continue; + } + err = insert_fte(g, fte); if (err) { up_write_ref_node(&g->node, false); @@ -2403,6 +2402,17 @@ static int init_root_tree(struct mlx5_flow_steering *steering, return 0; } +static void del_sw_root_ns(struct fs_node *node) +{ + struct mlx5_flow_root_namespace *root_ns; + struct mlx5_flow_namespace *ns; + + fs_get_obj(ns, node); + root_ns = container_of(ns, struct mlx5_flow_root_namespace, ns); + mutex_destroy(&root_ns->chain_lock); + kfree(node); +} + static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_flow_steering *steering, enum fs_flow_table_type table_type) @@ -2411,7 +2421,7 @@ static struct mlx5_flow_root_namespace struct mlx5_flow_root_namespace *root_ns; struct mlx5_flow_namespace *ns; - if (mlx5_accel_ipsec_device_caps(steering->dev) & MLX5_ACCEL_IPSEC_CAP_DEVICE && + if (mlx5_fpga_ipsec_device_caps(steering->dev) & MLX5_ACCEL_IPSEC_CAP_DEVICE && (table_type == FS_FT_NIC_RX || table_type == FS_FT_NIC_TX)) cmds = mlx5_fs_cmd_get_default_ipsec_fpga_cmds(table_type); @@ -2429,7 +2439,7 @@ static struct mlx5_flow_root_namespace ns = &root_ns->ns; fs_init_namespace(ns); mutex_init(&root_ns->chain_lock); - tree_init_node(&ns->node, NULL, NULL); + tree_init_node(&ns->node, NULL, del_sw_root_ns); tree_add_node(&ns->node, NULL); return root_ns; @@ -2995,7 +3005,8 @@ int mlx5_init_fs(struct mlx5_core_dev *dev) goto err; } - if (MLX5_IPSEC_DEV(dev) || MLX5_CAP_FLOWTABLE_NIC_TX(dev, ft_support)) { + if (mlx5_fpga_ipsec_device_caps(steering->dev) & MLX5_ACCEL_IPSEC_CAP_DEVICE || + MLX5_CAP_FLOWTABLE_NIC_TX(dev, ft_support)) { err = init_egress_root_ns(steering); if (err) goto err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c index 068578be00f1..690b822c6152 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c @@ -262,6 +262,11 @@ void mlx5i_destroy_underlay_qp(struct mlx5_core_dev *mdev, u32 qpn) mlx5_cmd_exec_in(mdev, destroy_qp, in); } +int mlx5i_update_nic_rx(struct mlx5e_priv *priv) +{ + return mlx5e_refresh_tirs(priv, true, true); +} + int mlx5i_create_tis(struct mlx5_core_dev *mdev, u32 underlay_qpn, u32 *tisn) { u32 in[MLX5_ST_SZ_DW(create_tis_in)] = {}; @@ -402,7 +407,7 @@ static int mlx5i_init_rx(struct mlx5e_priv *priv) err_destroy_direct_tirs: mlx5e_destroy_direct_tirs(priv, priv->direct_tir); err_destroy_indirect_tirs: - mlx5e_destroy_indirect_tirs(priv, true); + mlx5e_destroy_indirect_tirs(priv); err_destroy_direct_rqts: mlx5e_destroy_direct_rqts(priv, priv->direct_tir); err_destroy_indirect_rqts: @@ -418,7 +423,7 @@ static void mlx5i_cleanup_rx(struct mlx5e_priv *priv) { mlx5i_destroy_flow_steering(priv); mlx5e_destroy_direct_tirs(priv, priv->direct_tir); - mlx5e_destroy_indirect_tirs(priv, true); + mlx5e_destroy_indirect_tirs(priv); mlx5e_destroy_direct_rqts(priv, priv->direct_tir); mlx5e_destroy_rqt(priv, &priv->indir_rqt); mlx5e_close_drop_rq(&priv->drop_rq); @@ -456,7 +461,7 @@ static const struct mlx5e_profile mlx5i_nic_profile = { .cleanup_rx = mlx5i_cleanup_rx, .enable = NULL, /* mlx5i_enable */ .disable = NULL, /* mlx5i_disable */ - .update_rx = mlx5e_update_nic_rx, + .update_rx = mlx5i_update_nic_rx, .update_stats = NULL, /* mlx5i_update_stats */ .update_carrier = NULL, /* no HW update in IB link */ .rx_handlers.handle_rx_cqe = mlx5i_handle_rx_cqe, @@ -663,7 +668,9 @@ static int mlx5i_xmit(struct net_device *dev, struct sk_buff *skb, struct mlx5_ib_ah *mah = to_mah(address); struct mlx5i_priv *ipriv = epriv->ppriv; - return mlx5i_sq_xmit(sq, skb, &mah->av, dqpn, ipriv->qkey, netdev_xmit_more()); + mlx5i_sq_xmit(sq, skb, &mah->av, dqpn, ipriv->qkey, netdev_xmit_more()); + + return NETDEV_TX_OK; } static void mlx5i_set_pkey_index(struct net_device *netdev, int id) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.h b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.h index 3483ba642cfe..79071a15c4ca 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.h @@ -92,6 +92,8 @@ int mlx5i_init(struct mlx5_core_dev *mdev, void *ppriv); void mlx5i_cleanup(struct mlx5e_priv *priv); +int mlx5i_update_nic_rx(struct mlx5e_priv *priv); + /* Get child interface nic profile */ const struct mlx5e_profile *mlx5i_pkey_get_profile(void); @@ -110,19 +112,11 @@ struct mlx5i_tx_wqe { struct mlx5_wqe_data_seg data[]; }; -static inline void mlx5i_sq_fetch_wqe(struct mlx5e_txqsq *sq, - struct mlx5i_tx_wqe **wqe, - u16 pi) -{ - struct mlx5_wq_cyc *wq = &sq->wq; - - *wqe = mlx5_wq_cyc_get_wqe(wq, pi); - memset(*wqe, 0, sizeof(**wqe)); -} +#define MLX5I_SQ_FETCH_WQE(sq, pi) \ + ((struct mlx5i_tx_wqe *)mlx5e_fetch_wqe(&(sq)->wq, pi, sizeof(struct mlx5i_tx_wqe))) -netdev_tx_t mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, - struct mlx5_av *av, u32 dqpn, u32 dqkey, - bool xmit_more); +void mlx5i_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb, + struct mlx5_av *av, u32 dqpn, u32 dqkey, bool xmit_more); void mlx5i_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe); void mlx5i_get_stats(struct net_device *dev, struct rtnl_link_stats64 *stats); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c index b9af37ad40bf..f70367018862 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c @@ -347,7 +347,7 @@ static const struct mlx5e_profile mlx5i_pkey_nic_profile = { .cleanup_rx = mlx5i_pkey_cleanup_rx, .enable = NULL, .disable = NULL, - .update_rx = mlx5e_update_nic_rx, + .update_rx = mlx5i_update_nic_rx, .update_stats = NULL, .rx_handlers.handle_rx_cqe = mlx5i_handle_rx_cqe, .rx_handlers.handle_rx_cqe_mpwqe = NULL, /* Not supported */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c index 43f97601b500..ef0706d15a5b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c @@ -32,6 +32,7 @@ #include <linux/clocksource.h> #include <linux/highmem.h> +#include <linux/ptp_clock_kernel.h> #include <rdma/mlx5-abi.h> #include "lib/eq.h" #include "en.h" @@ -66,6 +67,26 @@ enum { MLX5_MTPPS_FS_ENH_OUT_PER_ADJ = BIT(0x7), }; +static u64 mlx5_read_internal_timer(struct mlx5_core_dev *dev, + struct ptp_system_timestamp *sts) +{ + u32 timer_h, timer_h1, timer_l; + + timer_h = ioread32be(&dev->iseg->internal_timer_h); + ptp_read_system_prets(sts); + timer_l = ioread32be(&dev->iseg->internal_timer_l); + ptp_read_system_postts(sts); + timer_h1 = ioread32be(&dev->iseg->internal_timer_h); + if (timer_h != timer_h1) { + /* wrap around */ + ptp_read_system_prets(sts); + timer_l = ioread32be(&dev->iseg->internal_timer_l); + ptp_read_system_postts(sts); + } + + return (u64)timer_l | (u64)timer_h1 << 32; +} + static u64 read_internal_timer(const struct cyclecounter *cc) { struct mlx5_clock *clock = container_of(cc, struct mlx5_clock, cycles); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h index 4be4d2d36218..4aaca7400fb2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h @@ -27,7 +27,6 @@ struct mlx5_eq { __be32 __iomem *doorbell; u32 cons_index; struct mlx5_frag_buf buf; - int size; unsigned int vecidx; unsigned int irqn; u8 eqn; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/port_tun.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/port_tun.c index 8809a65ecefb..e042e0924079 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/lib/port_tun.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/port_tun.c @@ -144,11 +144,11 @@ static int mlx5_set_entropy(struct mlx5_tun_entropy *tun_entropy, int mlx5_tun_entropy_refcount_inc(struct mlx5_tun_entropy *tun_entropy, int reformat_type) { - /* the default is error for unknown (non VXLAN/GRE tunnel types) */ int err = -EOPNOTSUPP; mutex_lock(&tun_entropy->lock); - if (reformat_type == MLX5_REFORMAT_TYPE_L2_TO_VXLAN && + if ((reformat_type == MLX5_REFORMAT_TYPE_L2_TO_VXLAN || + reformat_type == MLX5_REFORMAT_TYPE_L2_TO_L3_TUNNEL) && tun_entropy->enabled) { /* in case entropy calculation is enabled for all tunneling * types, it is ok for VXLAN, so approve. diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 8a375e3ed5c1..df46b1fce3a7 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -554,6 +554,9 @@ static int handle_hca_cap(struct mlx5_core_dev *dev, void *set_ctx) num_vhca_ports, MLX5_CAP_GEN_MAX(dev, num_vhca_ports)); + if (MLX5_CAP_GEN_MAX(dev, release_all_pages)) + MLX5_SET(cmd_hca_cap, set_hca_cap, release_all_pages, 1); + return set_caps(dev, set_ctx, MLX5_SET_HCA_CAP_OP_MOD_GENERAL_DEVICE); } @@ -674,26 +677,6 @@ int mlx5_core_disable_hca(struct mlx5_core_dev *dev, u16 func_id) return mlx5_cmd_exec_in(dev, disable_hca, in); } -u64 mlx5_read_internal_timer(struct mlx5_core_dev *dev, - struct ptp_system_timestamp *sts) -{ - u32 timer_h, timer_h1, timer_l; - - timer_h = ioread32be(&dev->iseg->internal_timer_h); - ptp_read_system_prets(sts); - timer_l = ioread32be(&dev->iseg->internal_timer_l); - ptp_read_system_postts(sts); - timer_h1 = ioread32be(&dev->iseg->internal_timer_h); - if (timer_h != timer_h1) { - /* wrap around */ - ptp_read_system_prets(sts); - timer_l = ioread32be(&dev->iseg->internal_timer_l); - ptp_read_system_postts(sts); - } - - return (u64)timer_l | (u64)timer_h1 << 32; -} - static int mlx5_core_set_issi(struct mlx5_core_dev *dev) { u32 query_out[MLX5_ST_SZ_DW(query_issi_out)] = {}; @@ -789,7 +772,7 @@ static int mlx5_pci_init(struct mlx5_core_dev *dev, struct pci_dev *pdev, } mlx5_pci_vsc_init(dev); - + dev->caps.embedded_cpu = mlx5_read_embedded_cpu(dev); return 0; err_clr_master: @@ -968,6 +951,8 @@ static int mlx5_function_setup(struct mlx5_core_dev *dev, bool boot) goto err_cmd_cleanup; } + mlx5_cmd_set_state(dev, MLX5_CMDIF_STATE_UP); + err = mlx5_core_enable_hca(dev, 0); if (err) { mlx5_core_err(dev, "enable hca failed\n"); @@ -1029,6 +1014,7 @@ reclaim_boot_pages: err_disable_hca: mlx5_core_disable_hca(dev, 0); err_cmd_cleanup: + mlx5_cmd_set_state(dev, MLX5_CMDIF_STATE_DOWN); mlx5_cmd_cleanup(dev); return err; @@ -1046,6 +1032,7 @@ static int mlx5_function_teardown(struct mlx5_core_dev *dev, bool boot) } mlx5_reclaim_startup_pages(dev); mlx5_core_disable_hca(dev, 0); + mlx5_cmd_set_state(dev, MLX5_CMDIF_STATE_DOWN); mlx5_cmd_cleanup(dev); return 0; @@ -1183,7 +1170,6 @@ int mlx5_load_one(struct mlx5_core_dev *dev, bool boot) { int err = 0; - dev->caps.embedded_cpu = mlx5_read_embedded_cpu(dev); mutex_lock(&dev->intf_state_mutex); if (test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state)) { mlx5_core_warn(dev, "interface is up, NOP\n"); @@ -1194,7 +1180,7 @@ int mlx5_load_one(struct mlx5_core_dev *dev, bool boot) err = mlx5_function_setup(dev, boot); if (err) - goto out; + goto err_function; if (boot) { err = mlx5_init_once(dev); @@ -1220,10 +1206,9 @@ int mlx5_load_one(struct mlx5_core_dev *dev, bool boot) mlx5_register_device(dev); set_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state); -out: - mutex_unlock(&dev->intf_state_mutex); - return err; + mutex_unlock(&dev->intf_state_mutex); + return 0; err_devlink_reg: mlx5_unload(dev); @@ -1232,18 +1217,17 @@ err_load: mlx5_cleanup_once(dev); function_teardown: mlx5_function_teardown(dev, boot); +err_function: dev->state = MLX5_DEVICE_STATE_INTERNAL_ERROR; +out: mutex_unlock(&dev->intf_state_mutex); - return err; } void mlx5_unload_one(struct mlx5_core_dev *dev, bool cleanup) { - if (cleanup) { + if (cleanup) mlx5_unregister_device(dev); - mlx5_drain_health_wq(dev); - } mutex_lock(&dev->intf_state_mutex); if (!test_bit(MLX5_INTERFACE_STATE_UP, &dev->intf_state)) { @@ -1293,7 +1277,7 @@ static int mlx5_mdev_init(struct mlx5_core_dev *dev, int profile_idx) mlx5_debugfs_root); if (!priv->dbg_root) { dev_err(dev->device, "mlx5_core: error, Cannot create debugfs dir, aborting\n"); - return -ENOMEM; + goto err_dbg_root; } err = mlx5_health_init(dev); @@ -1310,15 +1294,27 @@ err_pagealloc_init: mlx5_health_cleanup(dev); err_health_init: debugfs_remove(dev->priv.dbg_root); - +err_dbg_root: + mutex_destroy(&priv->pgdir_mutex); + mutex_destroy(&priv->alloc_mutex); + mutex_destroy(&priv->bfregs.wc_head.lock); + mutex_destroy(&priv->bfregs.reg_head.lock); + mutex_destroy(&dev->intf_state_mutex); return err; } static void mlx5_mdev_uninit(struct mlx5_core_dev *dev) { + struct mlx5_priv *priv = &dev->priv; + mlx5_pagealloc_cleanup(dev); mlx5_health_cleanup(dev); debugfs_remove_recursive(dev->priv.dbg_root); + mutex_destroy(&priv->pgdir_mutex); + mutex_destroy(&priv->alloc_mutex); + mutex_destroy(&priv->bfregs.wc_head.lock); + mutex_destroy(&priv->bfregs.reg_head.lock); + mutex_destroy(&dev->intf_state_mutex); } #define MLX5_IB_MOD "mlx5_ib" @@ -1386,6 +1382,7 @@ static void remove_one(struct pci_dev *pdev) mlx5_crdump_disable(dev); mlx5_devlink_unregister(devlink); + mlx5_drain_health_wq(dev); mlx5_unload_one(dev, true); mlx5_pci_close(dev); mlx5_mdev_uninit(dev); @@ -1547,6 +1544,22 @@ static void shutdown(struct pci_dev *pdev) mlx5_pci_disable_device(dev); } +static int mlx5_suspend(struct pci_dev *pdev, pm_message_t state) +{ + struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + + mlx5_unload_one(dev, false); + + return 0; +} + +static int mlx5_resume(struct pci_dev *pdev) +{ + struct mlx5_core_dev *dev = pci_get_drvdata(pdev); + + return mlx5_load_one(dev, false); +} + static const struct pci_device_id mlx5_core_pci_table[] = { { PCI_VDEVICE(MELLANOX, PCI_DEVICE_ID_MELLANOX_CONNECTIB) }, { PCI_VDEVICE(MELLANOX, 0x1012), MLX5_PCI_DEV_IS_VF}, /* Connect-IB VF */ @@ -1590,6 +1603,8 @@ static struct pci_driver mlx5_core_driver = { .id_table = mlx5_core_pci_table, .probe = init_one, .remove = remove_one, + .suspend = mlx5_suspend, + .resume = mlx5_resume, .shutdown = shutdown, .err_handler = &mlx5_err_handler, .sriov_configure = mlx5_core_sriov_configure, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index a8fb43a85d1d..fc1649dac11b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -38,7 +38,6 @@ #include <linux/sched.h> #include <linux/if_link.h> #include <linux/firmware.h> -#include <linux/ptp_clock_kernel.h> #include <linux/mlx5/cq.h> #include <linux/mlx5/fs.h> #include <linux/mlx5/driver.h> @@ -141,8 +140,6 @@ int mlx5_modify_scheduling_element_cmd(struct mlx5_core_dev *dev, u8 hierarchy, int mlx5_destroy_scheduling_element_cmd(struct mlx5_core_dev *dev, u8 hierarchy, u32 element_id); int mlx5_wait_for_pages(struct mlx5_core_dev *dev, int *pages); -u64 mlx5_read_internal_timer(struct mlx5_core_dev *dev, - struct ptp_system_timestamp *sts); void mlx5_cmd_trigger_completions(struct mlx5_core_dev *dev); void mlx5_cmd_flush(struct mlx5_core_dev *dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c index 3d6f617abb7d..5ddd18639a1e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c @@ -50,6 +50,7 @@ struct mlx5_pages_req { u8 ec_function; s32 npages; struct work_struct work; + u8 release_all; }; struct fw_page { @@ -155,15 +156,21 @@ static int mlx5_cmd_query_pages(struct mlx5_core_dev *dev, u16 *func_id, return err; } -static int alloc_4k(struct mlx5_core_dev *dev, u64 *addr) +static int alloc_4k(struct mlx5_core_dev *dev, u64 *addr, u16 func_id) { - struct fw_page *fp; + struct fw_page *fp = NULL; + struct fw_page *iter; unsigned n; - if (list_empty(&dev->priv.free_list)) + list_for_each_entry(iter, &dev->priv.free_list, list) { + if (iter->func_id != func_id) + continue; + fp = iter; + } + + if (list_empty(&dev->priv.free_list) || !fp) return -ENOMEM; - fp = list_entry(dev->priv.free_list.next, struct fw_page, list); n = find_first_bit(&fp->bitmask, 8 * sizeof(fp->bitmask)); if (n >= MLX5_NUM_4K_IN_PAGE) { mlx5_core_warn(dev, "alloc 4k bug\n"); @@ -181,6 +188,18 @@ static int alloc_4k(struct mlx5_core_dev *dev, u64 *addr) #define MLX5_U64_4K_PAGE_MASK ((~(u64)0U) << PAGE_SHIFT) +static void free_fwp(struct mlx5_core_dev *dev, struct fw_page *fwp, + bool in_free_list) +{ + rb_erase(&fwp->rb_node, &dev->priv.page_root); + if (in_free_list) + list_del(&fwp->list); + dma_unmap_page(dev->device, fwp->addr & MLX5_U64_4K_PAGE_MASK, + PAGE_SIZE, DMA_BIDIRECTIONAL); + __free_page(fwp->page); + kfree(fwp); +} + static void free_4k(struct mlx5_core_dev *dev, u64 addr) { struct fw_page *fwp; @@ -188,24 +207,16 @@ static void free_4k(struct mlx5_core_dev *dev, u64 addr) fwp = find_fw_page(dev, addr & MLX5_U64_4K_PAGE_MASK); if (!fwp) { - mlx5_core_warn(dev, "page not found\n"); + mlx5_core_warn_rl(dev, "page not found\n"); return; } - n = (addr & ~MLX5_U64_4K_PAGE_MASK) >> MLX5_ADAPTER_PAGE_SHIFT; fwp->free_count++; set_bit(n, &fwp->bitmask); - if (fwp->free_count == MLX5_NUM_4K_IN_PAGE) { - rb_erase(&fwp->rb_node, &dev->priv.page_root); - if (fwp->free_count != 1) - list_del(&fwp->list); - dma_unmap_page(dev->device, addr & MLX5_U64_4K_PAGE_MASK, - PAGE_SIZE, DMA_BIDIRECTIONAL); - __free_page(fwp->page); - kfree(fwp); - } else if (fwp->free_count == 1) { + if (fwp->free_count == MLX5_NUM_4K_IN_PAGE) + free_fwp(dev, fwp, fwp->free_count != 1); + else if (fwp->free_count == 1) list_add(&fwp->list, &dev->priv.free_list); - } } static int alloc_system_page(struct mlx5_core_dev *dev, u16 func_id) @@ -290,7 +301,7 @@ static int give_pages(struct mlx5_core_dev *dev, u16 func_id, int npages, for (i = 0; i < npages; i++) { retry: - err = alloc_4k(dev, &addr); + err = alloc_4k(dev, &addr, func_id); if (err) { if (err == -ENOMEM) err = alloc_system_page(dev, func_id); @@ -337,6 +348,33 @@ out_free: return err; } +static void release_all_pages(struct mlx5_core_dev *dev, u32 func_id, + bool ec_function) +{ + struct rb_node *p; + int npages = 0; + + p = rb_first(&dev->priv.page_root); + while (p) { + struct fw_page *fwp = rb_entry(p, struct fw_page, rb_node); + + p = rb_next(p); + if (fwp->func_id != func_id) + continue; + npages += (MLX5_NUM_4K_IN_PAGE - fwp->free_count); + free_fwp(dev, fwp, fwp->free_count); + } + + dev->priv.fw_pages -= npages; + if (func_id) + dev->priv.vfs_pages -= npages; + else if (mlx5_core_is_ecpf(dev) && !ec_function) + dev->priv.peer_pf_pages -= npages; + + mlx5_core_dbg(dev, "npages %d, ec_function %d, func_id 0x%x\n", + npages, ec_function, func_id); +} + static int reclaim_pages_cmd(struct mlx5_core_dev *dev, u32 *in, int in_size, u32 *out, int out_size) { @@ -430,7 +468,9 @@ static void pages_work_handler(struct work_struct *work) struct mlx5_core_dev *dev = req->dev; int err = 0; - if (req->npages < 0) + if (req->release_all) + release_all_pages(dev, req->func_id, req->ec_function); + else if (req->npages < 0) err = reclaim_pages(dev, req->func_id, -1 * req->npages, NULL, req->ec_function); else if (req->npages > 0) @@ -445,6 +485,7 @@ static void pages_work_handler(struct work_struct *work) enum { EC_FUNCTION_MASK = 0x8000, + RELEASE_ALL_PAGES_MASK = 0x4000, }; static int req_pages_handler(struct notifier_block *nb, @@ -455,6 +496,7 @@ static int req_pages_handler(struct notifier_block *nb, struct mlx5_priv *priv; struct mlx5_eqe *eqe; bool ec_function; + bool release_all; u16 func_id; s32 npages; @@ -465,8 +507,10 @@ static int req_pages_handler(struct notifier_block *nb, func_id = be16_to_cpu(eqe->data.req_pages.func_id); npages = be32_to_cpu(eqe->data.req_pages.num_pages); ec_function = be16_to_cpu(eqe->data.req_pages.ec_function) & EC_FUNCTION_MASK; - mlx5_core_dbg(dev, "page request for func 0x%x, npages %d\n", - func_id, npages); + release_all = be16_to_cpu(eqe->data.req_pages.ec_function) & + RELEASE_ALL_PAGES_MASK; + mlx5_core_dbg(dev, "page request for func 0x%x, npages %d, release_all %d\n", + func_id, npages, release_all); req = kzalloc(sizeof(*req), GFP_ATOMIC); if (!req) { mlx5_core_warn(dev, "failed to allocate pages request\n"); @@ -477,6 +521,7 @@ static int req_pages_handler(struct notifier_block *nb, req->func_id = func_id; req->npages = npages; req->ec_function = ec_function; + req->release_all = release_all; INIT_WORK(&req->work, pages_work_handler); queue_work(dev->priv.pg_wq, &req->work); return NOTIFY_OK; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c index 554811de4c9d..df1363a34a42 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c @@ -1662,7 +1662,7 @@ dr_action_modify_check_field_limitation(struct mlx5dr_action *action, } static bool -dr_action_modify_check_is_ttl_modify(const u64 *sw_action) +dr_action_modify_check_is_ttl_modify(const void *sw_action) { u16 sw_field = MLX5_GET(set_action_in, sw_action, field); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c index 48b6358b6845..890767a2a7cb 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c @@ -297,7 +297,8 @@ mlx5dr_domain_create(struct mlx5_core_dev *mdev, enum mlx5dr_domain_type type) dmn->mdev = mdev; dmn->type = type; refcount_set(&dmn->refcount, 1); - mutex_init(&dmn->mutex); + mutex_init(&dmn->info.rx.mutex); + mutex_init(&dmn->info.tx.mutex); if (dr_domain_caps_init(mdev, dmn)) { mlx5dr_err(dmn, "Failed init domain, no caps\n"); @@ -345,9 +346,9 @@ int mlx5dr_domain_sync(struct mlx5dr_domain *dmn, u32 flags) int ret = 0; if (flags & MLX5DR_DOMAIN_SYNC_FLAGS_SW) { - mutex_lock(&dmn->mutex); + mlx5dr_domain_lock(dmn); ret = mlx5dr_send_ring_force_drain(dmn); - mutex_unlock(&dmn->mutex); + mlx5dr_domain_unlock(dmn); if (ret) { mlx5dr_err(dmn, "Force drain failed flags: %d, ret: %d\n", flags, ret); @@ -371,7 +372,8 @@ int mlx5dr_domain_destroy(struct mlx5dr_domain *dmn) dr_domain_uninit_cache(dmn); dr_domain_uninit_resources(dmn); dr_domain_caps_uninit(dmn); - mutex_destroy(&dmn->mutex); + mutex_destroy(&dmn->info.tx.mutex); + mutex_destroy(&dmn->info.rx.mutex); kfree(dmn); return 0; } @@ -379,7 +381,7 @@ int mlx5dr_domain_destroy(struct mlx5dr_domain *dmn) void mlx5dr_domain_set_peer(struct mlx5dr_domain *dmn, struct mlx5dr_domain *peer_dmn) { - mutex_lock(&dmn->mutex); + mlx5dr_domain_lock(dmn); if (dmn->peer_dmn) refcount_dec(&dmn->peer_dmn->refcount); @@ -389,5 +391,5 @@ void mlx5dr_domain_set_peer(struct mlx5dr_domain *dmn, if (dmn->peer_dmn) refcount_inc(&dmn->peer_dmn->refcount); - mutex_unlock(&dmn->mutex); + mlx5dr_domain_unlock(dmn); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c index a95938874798..31abcbb95ca2 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_matcher.c @@ -690,7 +690,7 @@ mlx5dr_matcher_create(struct mlx5dr_table *tbl, refcount_set(&matcher->refcount, 1); INIT_LIST_HEAD(&matcher->matcher_list); - mutex_lock(&tbl->dmn->mutex); + mlx5dr_domain_lock(tbl->dmn); ret = dr_matcher_init(matcher, mask); if (ret) @@ -700,14 +700,14 @@ mlx5dr_matcher_create(struct mlx5dr_table *tbl, if (ret) goto matcher_uninit; - mutex_unlock(&tbl->dmn->mutex); + mlx5dr_domain_unlock(tbl->dmn); return matcher; matcher_uninit: dr_matcher_uninit(matcher); free_matcher: - mutex_unlock(&tbl->dmn->mutex); + mlx5dr_domain_unlock(tbl->dmn); kfree(matcher); dec_ref: refcount_dec(&tbl->refcount); @@ -791,13 +791,13 @@ int mlx5dr_matcher_destroy(struct mlx5dr_matcher *matcher) if (refcount_read(&matcher->refcount) > 1) return -EBUSY; - mutex_lock(&tbl->dmn->mutex); + mlx5dr_domain_lock(tbl->dmn); dr_matcher_remove_from_tbl(matcher); dr_matcher_uninit(matcher); refcount_dec(&matcher->tbl->refcount); - mutex_unlock(&tbl->dmn->mutex); + mlx5dr_domain_unlock(tbl->dmn); kfree(matcher); return 0; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c index cce3ee7a6614..cd708dcc2e3a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_rule.c @@ -938,7 +938,10 @@ static bool dr_rule_verify(struct mlx5dr_matcher *matcher, static int dr_rule_destroy_rule_nic(struct mlx5dr_rule *rule, struct mlx5dr_rule_rx_tx *nic_rule) { + mlx5dr_domain_nic_lock(nic_rule->nic_matcher->nic_tbl->nic_dmn); dr_rule_clean_rule_members(rule, nic_rule); + mlx5dr_domain_nic_unlock(nic_rule->nic_matcher->nic_tbl->nic_dmn); + return 0; } @@ -1039,18 +1042,18 @@ dr_rule_create_rule_nic(struct mlx5dr_rule *rule, if (dr_rule_skip(dmn->type, nic_dmn->ste_type, &matcher->mask, param)) return 0; + hw_ste_arr = kzalloc(DR_RULE_MAX_STE_CHAIN * DR_STE_SIZE, GFP_KERNEL); + if (!hw_ste_arr) + return -ENOMEM; + + mlx5dr_domain_nic_lock(nic_dmn); + ret = mlx5dr_matcher_select_builders(matcher, nic_matcher, dr_rule_get_ipv(¶m->outer), dr_rule_get_ipv(¶m->inner)); if (ret) - goto out_err; - - hw_ste_arr = kzalloc(DR_RULE_MAX_STE_CHAIN * DR_STE_SIZE, GFP_KERNEL); - if (!hw_ste_arr) { - ret = -ENOMEM; - goto out_err; - } + goto free_hw_ste; /* Set the tag values inside the ste array */ ret = mlx5dr_ste_build_ste_arr(matcher, nic_matcher, param, hw_ste_arr); @@ -1115,6 +1118,8 @@ dr_rule_create_rule_nic(struct mlx5dr_rule *rule, if (htbl) mlx5dr_htbl_put(htbl); + mlx5dr_domain_nic_unlock(nic_dmn); + kfree(hw_ste_arr); return 0; @@ -1129,8 +1134,8 @@ free_rule: kfree(ste_info); } free_hw_ste: + mlx5dr_domain_nic_unlock(nic_dmn); kfree(hw_ste_arr); -out_err: return ret; } @@ -1232,31 +1237,23 @@ struct mlx5dr_rule *mlx5dr_rule_create(struct mlx5dr_matcher *matcher, { struct mlx5dr_rule *rule; - mutex_lock(&matcher->tbl->dmn->mutex); refcount_inc(&matcher->refcount); rule = dr_rule_create_rule(matcher, value, num_actions, actions); if (!rule) refcount_dec(&matcher->refcount); - mutex_unlock(&matcher->tbl->dmn->mutex); - return rule; } int mlx5dr_rule_destroy(struct mlx5dr_rule *rule) { struct mlx5dr_matcher *matcher = rule->matcher; - struct mlx5dr_table *tbl = rule->matcher->tbl; int ret; - mutex_lock(&tbl->dmn->mutex); - ret = dr_rule_destroy_rule(rule); - - mutex_unlock(&tbl->dmn->mutex); - if (!ret) refcount_dec(&matcher->refcount); + return ret; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c index b8d97d44be7b..f421013b0b54 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c @@ -357,9 +357,11 @@ static int dr_postsend_icm_data(struct mlx5dr_domain *dmn, u32 buff_offset; int ret; + spin_lock(&send_ring->lock); + ret = dr_handle_pending_wc(dmn, send_ring); if (ret) - return ret; + goto out_unlock; if (send_info->write.length > dmn->info.max_inline_size) { buff_offset = (send_ring->tx_head & @@ -377,7 +379,9 @@ static int dr_postsend_icm_data(struct mlx5dr_domain *dmn, dr_fill_data_segs(send_ring, send_info); dr_post_send(send_ring->qp, send_info); - return 0; +out_unlock: + spin_unlock(&send_ring->lock); + return ret; } static int dr_get_tbl_copy_details(struct mlx5dr_domain *dmn, @@ -563,9 +567,7 @@ int mlx5dr_send_postsend_action(struct mlx5dr_domain *dmn, send_info.remote_addr = action->rewrite.chunk->mr_addr; send_info.rkey = action->rewrite.chunk->rkey; - mutex_lock(&dmn->mutex); ret = dr_postsend_icm_data(dmn, &send_info); - mutex_unlock(&dmn->mutex); return ret; } @@ -886,6 +888,7 @@ int mlx5dr_send_ring_alloc(struct mlx5dr_domain *dmn) init_attr.pdn = dmn->pdn; init_attr.uar = dmn->uar; init_attr.max_send_wr = QUEUE_SIZE; + spin_lock_init(&dmn->send_ring->lock); dmn->send_ring->qp = dr_create_rc_qp(dmn->mdev, &init_attr); if (!dmn->send_ring->qp) { @@ -990,7 +993,9 @@ int mlx5dr_send_ring_force_drain(struct mlx5dr_domain *dmn) return ret; } + spin_lock(&send_ring->lock); ret = dr_handle_pending_wc(dmn, send_ring); + spin_unlock(&send_ring->lock); return ret; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c index 78c884911ceb..00c2f598f034 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c @@ -112,7 +112,7 @@ static u32 dr_ste_crc32_calc(const void *input_data, size_t length) { u32 crc = crc32(0, input_data, length); - return htonl(crc); + return (__force u32)htonl(crc); } u32 mlx5dr_ste_calc_hash_index(u8 *hw_ste_p, struct mlx5dr_ste_htbl *htbl) @@ -869,7 +869,7 @@ static void dr_ste_copy_mask_misc(char *mask, struct mlx5dr_match_misc *spec) static void dr_ste_copy_mask_spec(char *mask, struct mlx5dr_match_spec *spec) { - u32 raw_ip[4]; + __be32 raw_ip[4]; spec->smac_47_16 = MLX5_GET(fte_match_set_lyr_2_4, mask, smac_47_16); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c index c2fe48d7b75a..b599b6beb5b9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_table.c @@ -14,7 +14,7 @@ int mlx5dr_table_set_miss_action(struct mlx5dr_table *tbl, if (action && action->action_type != DR_ACTION_TYP_FT) return -EOPNOTSUPP; - mutex_lock(&tbl->dmn->mutex); + mlx5dr_domain_lock(tbl->dmn); if (!list_empty(&tbl->matcher_list)) last_matcher = list_last_entry(&tbl->matcher_list, @@ -78,7 +78,7 @@ int mlx5dr_table_set_miss_action(struct mlx5dr_table *tbl, refcount_inc(&action->refcount); out: - mutex_unlock(&tbl->dmn->mutex); + mlx5dr_domain_unlock(tbl->dmn); return ret; } @@ -95,7 +95,7 @@ static void dr_table_uninit_fdb(struct mlx5dr_table *tbl) static void dr_table_uninit(struct mlx5dr_table *tbl) { - mutex_lock(&tbl->dmn->mutex); + mlx5dr_domain_lock(tbl->dmn); switch (tbl->dmn->type) { case MLX5DR_DOMAIN_TYPE_NIC_RX: @@ -112,7 +112,7 @@ static void dr_table_uninit(struct mlx5dr_table *tbl) break; } - mutex_unlock(&tbl->dmn->mutex); + mlx5dr_domain_unlock(tbl->dmn); } static int dr_table_init_nic(struct mlx5dr_domain *dmn, @@ -177,7 +177,7 @@ static int dr_table_init(struct mlx5dr_table *tbl) INIT_LIST_HEAD(&tbl->matcher_list); - mutex_lock(&tbl->dmn->mutex); + mlx5dr_domain_lock(tbl->dmn); switch (tbl->dmn->type) { case MLX5DR_DOMAIN_TYPE_NIC_RX: @@ -201,7 +201,7 @@ static int dr_table_init(struct mlx5dr_table *tbl) break; } - mutex_unlock(&tbl->dmn->mutex); + mlx5dr_domain_unlock(tbl->dmn); return ret; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h index 71fa01ce348a..0883956c58c0 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h @@ -635,6 +635,7 @@ struct mlx5dr_domain_rx_tx { u64 drop_icm_addr; u64 default_icm_addr; enum mlx5dr_ste_entry_type ste_type; + struct mutex mutex; /* protect rx/tx domain */ }; struct mlx5dr_domain_info { @@ -659,7 +660,6 @@ struct mlx5dr_domain { struct mlx5_uars_page *uar; enum mlx5dr_domain_type type; refcount_t refcount; - struct mutex mutex; /* protect domain */ struct mlx5dr_icm_pool *ste_icm_pool; struct mlx5dr_icm_pool *action_icm_pool; struct mlx5dr_send_ring *send_ring; @@ -813,6 +813,28 @@ struct mlx5dr_icm_chunk { struct list_head *miss_list; }; +static inline void mlx5dr_domain_nic_lock(struct mlx5dr_domain_rx_tx *nic_dmn) +{ + mutex_lock(&nic_dmn->mutex); +} + +static inline void mlx5dr_domain_nic_unlock(struct mlx5dr_domain_rx_tx *nic_dmn) +{ + mutex_unlock(&nic_dmn->mutex); +} + +static inline void mlx5dr_domain_lock(struct mlx5dr_domain *dmn) +{ + mlx5dr_domain_nic_lock(&dmn->info.rx); + mlx5dr_domain_nic_lock(&dmn->info.tx); +} + +static inline void mlx5dr_domain_unlock(struct mlx5dr_domain *dmn) +{ + mlx5dr_domain_nic_unlock(&dmn->info.tx); + mlx5dr_domain_nic_unlock(&dmn->info.rx); +} + static inline int mlx5dr_matcher_supp_flex_parser_icmp_v4(struct mlx5dr_cmd_caps *caps) { @@ -1042,6 +1064,7 @@ struct mlx5dr_send_ring { struct ib_wc wc[MAX_SEND_CQE]; u8 sync_buff[MIN_READ_SYNC]; struct mlx5dr_mr *sync_mr; + spinlock_t lock; /* Protect the data path of the send ring */ }; int mlx5dr_send_ring_alloc(struct mlx5dr_domain *dmn); diff --git a/drivers/net/ethernet/mellanox/mlxfw/mlxfw_fsm.c b/drivers/net/ethernet/mellanox/mlxfw/mlxfw_fsm.c index 046a0cb82ed8..7a04c626a2aa 100644 --- a/drivers/net/ethernet/mellanox/mlxfw/mlxfw_fsm.c +++ b/drivers/net/ethernet/mellanox/mlxfw/mlxfw_fsm.c @@ -76,7 +76,7 @@ static int mlxfw_fsm_state_err(struct mlxfw_dev *mlxfw_dev, case MLXFW_FSM_STATE_ERR_MAX: MLXFW_ERR_MSG(mlxfw_dev, extack, "unknown error", err); break; - }; + } return mlxfw_fsm_state_errno[fsm_state_err]; }; @@ -159,7 +159,7 @@ mlxfw_fsm_reactivate_err(struct mlxfw_dev *mlxfw_dev, case MLXFW_FSM_REACTIVATE_STATUS_MAX: MLXFW_REACT_ERR("unexpected error", err); break; - }; + } return -EREMOTEIO; }; diff --git a/drivers/net/ethernet/mellanox/mlxsw/Makefile b/drivers/net/ethernet/mellanox/mlxsw/Makefile index 0e86a581d45b..4aeabb35c943 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/Makefile +++ b/drivers/net/ethernet/mellanox/mlxsw/Makefile @@ -21,6 +21,7 @@ mlxsw_spectrum-objs := spectrum.o spectrum_buffers.o \ spectrum_acl_atcam.o spectrum_acl_erp.o \ spectrum1_acl_tcam.o spectrum2_acl_tcam.o \ spectrum_acl_bloom_filter.o spectrum_acl.o \ + spectrum_flow.o spectrum_matchall.o \ spectrum_flower.o spectrum_cnt.o \ spectrum_fid.o spectrum_ipip.o \ spectrum_acl_flex_actions.o \ diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h index 9b39b8e70519..fcb88d4271bf 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/reg.h +++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h @@ -3203,7 +3203,7 @@ MLXSW_ITEM32_INDEXED(reg, iedr, rec_type, MLXSW_REG_IEDR_BASE_LEN, 24, 8, * Size of entries do be deleted. The unit is 1 entry, regardless of entry type. * Access: OP */ -MLXSW_ITEM32_INDEXED(reg, iedr, rec_size, MLXSW_REG_IEDR_BASE_LEN, 0, 11, +MLXSW_ITEM32_INDEXED(reg, iedr, rec_size, MLXSW_REG_IEDR_BASE_LEN, 0, 13, MLXSW_REG_IEDR_REC_LEN, 0x00, false); /* reg_iedr_rec_index_start @@ -5526,40 +5526,37 @@ MLXSW_ITEM32(reg, htgt, type, 0x00, 8, 4); enum mlxsw_reg_htgt_trap_group { MLXSW_REG_HTGT_TRAP_GROUP_EMAD, - MLXSW_REG_HTGT_TRAP_GROUP_SX2_RX, - MLXSW_REG_HTGT_TRAP_GROUP_SX2_CTRL, MLXSW_REG_HTGT_TRAP_GROUP_SP_STP, MLXSW_REG_HTGT_TRAP_GROUP_SP_LACP, MLXSW_REG_HTGT_TRAP_GROUP_SP_LLDP, - MLXSW_REG_HTGT_TRAP_GROUP_SP_IGMP, + MLXSW_REG_HTGT_TRAP_GROUP_SP_MC_SNOOPING, MLXSW_REG_HTGT_TRAP_GROUP_SP_BGP, MLXSW_REG_HTGT_TRAP_GROUP_SP_OSPF, MLXSW_REG_HTGT_TRAP_GROUP_SP_PIM, MLXSW_REG_HTGT_TRAP_GROUP_SP_MULTICAST, - MLXSW_REG_HTGT_TRAP_GROUP_SP_ARP, + MLXSW_REG_HTGT_TRAP_GROUP_SP_NEIGH_DISCOVERY, MLXSW_REG_HTGT_TRAP_GROUP_SP_ROUTER_EXP, - MLXSW_REG_HTGT_TRAP_GROUP_SP_REMOTE_ROUTE, MLXSW_REG_HTGT_TRAP_GROUP_SP_IP2ME, MLXSW_REG_HTGT_TRAP_GROUP_SP_DHCP, MLXSW_REG_HTGT_TRAP_GROUP_SP_EVENT, - MLXSW_REG_HTGT_TRAP_GROUP_SP_IPV6_MLD, - MLXSW_REG_HTGT_TRAP_GROUP_SP_IPV6_ND, + MLXSW_REG_HTGT_TRAP_GROUP_SP_IPV6, MLXSW_REG_HTGT_TRAP_GROUP_SP_LBERROR, MLXSW_REG_HTGT_TRAP_GROUP_SP_PTP0, MLXSW_REG_HTGT_TRAP_GROUP_SP_PTP1, MLXSW_REG_HTGT_TRAP_GROUP_SP_VRRP, - - __MLXSW_REG_HTGT_TRAP_GROUP_MAX, - MLXSW_REG_HTGT_TRAP_GROUP_MAX = __MLXSW_REG_HTGT_TRAP_GROUP_MAX - 1 -}; - -enum mlxsw_reg_htgt_discard_trap_group { - MLXSW_REG_HTGT_DISCARD_TRAP_GROUP_BASE = MLXSW_REG_HTGT_TRAP_GROUP_MAX, + MLXSW_REG_HTGT_TRAP_GROUP_SP_PKT_SAMPLE, + MLXSW_REG_HTGT_TRAP_GROUP_SP_FLOW_LOGGING, + MLXSW_REG_HTGT_TRAP_GROUP_SP_FID_MISS, + MLXSW_REG_HTGT_TRAP_GROUP_SP_BFD, MLXSW_REG_HTGT_TRAP_GROUP_SP_DUMMY, MLXSW_REG_HTGT_TRAP_GROUP_SP_L2_DISCARDS, MLXSW_REG_HTGT_TRAP_GROUP_SP_L3_DISCARDS, + MLXSW_REG_HTGT_TRAP_GROUP_SP_L3_EXCEPTIONS, MLXSW_REG_HTGT_TRAP_GROUP_SP_TUNNEL_DISCARDS, MLXSW_REG_HTGT_TRAP_GROUP_SP_ACL_DISCARDS, + + __MLXSW_REG_HTGT_TRAP_GROUP_MAX, + MLXSW_REG_HTGT_TRAP_GROUP_MAX = __MLXSW_REG_HTGT_TRAP_GROUP_MAX - 1 }; /* reg_htgt_trap_group diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c index 24ca8d5bc564..5ffa32b75e5f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c @@ -25,9 +25,7 @@ #include <linux/log2.h> #include <net/switchdev.h> #include <net/pkt_cls.h> -#include <net/tc_act/tc_mirred.h> #include <net/netevent.h> -#include <net/tc_act/tc_sample.h> #include <net/addrconf.h> #include "spectrum.h" @@ -582,16 +580,6 @@ static int mlxsw_sp_base_mac_get(struct mlxsw_sp *mlxsw_sp) return 0; } -static int mlxsw_sp_port_sample_set(struct mlxsw_sp_port *mlxsw_sp_port, - bool enable, u32 rate) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - char mpsc_pl[MLXSW_REG_MPSC_LEN]; - - mlxsw_reg_mpsc_pack(mpsc_pl, mlxsw_sp_port->local_port, enable, rate); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mpsc), mpsc_pl); -} - static int mlxsw_sp_port_admin_status_set(struct mlxsw_sp_port *mlxsw_sp_port, bool is_up) { @@ -1362,412 +1350,6 @@ static int mlxsw_sp_port_kill_vid(struct net_device *dev, return 0; } -static struct mlxsw_sp_port_mall_tc_entry * -mlxsw_sp_port_mall_tc_entry_find(struct mlxsw_sp_port *port, - unsigned long cookie) { - struct mlxsw_sp_port_mall_tc_entry *mall_tc_entry; - - list_for_each_entry(mall_tc_entry, &port->mall_tc_list, list) - if (mall_tc_entry->cookie == cookie) - return mall_tc_entry; - - return NULL; -} - -static int -mlxsw_sp_port_add_cls_matchall_mirror(struct mlxsw_sp_port *mlxsw_sp_port, - struct mlxsw_sp_port_mall_mirror_tc_entry *mirror, - const struct flow_action_entry *act, - bool ingress) -{ - enum mlxsw_sp_span_type span_type; - - if (!act->dev) { - netdev_err(mlxsw_sp_port->dev, "Could not find requested device\n"); - return -EINVAL; - } - - mirror->ingress = ingress; - span_type = ingress ? MLXSW_SP_SPAN_INGRESS : MLXSW_SP_SPAN_EGRESS; - return mlxsw_sp_span_mirror_add(mlxsw_sp_port, act->dev, span_type, - true, &mirror->span_id); -} - -static void -mlxsw_sp_port_del_cls_matchall_mirror(struct mlxsw_sp_port *mlxsw_sp_port, - struct mlxsw_sp_port_mall_mirror_tc_entry *mirror) -{ - enum mlxsw_sp_span_type span_type; - - span_type = mirror->ingress ? - MLXSW_SP_SPAN_INGRESS : MLXSW_SP_SPAN_EGRESS; - mlxsw_sp_span_mirror_del(mlxsw_sp_port, mirror->span_id, - span_type, true); -} - -static int -mlxsw_sp_port_add_cls_matchall_sample(struct mlxsw_sp_port *mlxsw_sp_port, - struct tc_cls_matchall_offload *cls, - const struct flow_action_entry *act, - bool ingress) -{ - int err; - - if (!mlxsw_sp_port->sample) - return -EOPNOTSUPP; - if (rtnl_dereference(mlxsw_sp_port->sample->psample_group)) { - netdev_err(mlxsw_sp_port->dev, "sample already active\n"); - return -EEXIST; - } - if (act->sample.rate > MLXSW_REG_MPSC_RATE_MAX) { - netdev_err(mlxsw_sp_port->dev, "sample rate not supported\n"); - return -EOPNOTSUPP; - } - - rcu_assign_pointer(mlxsw_sp_port->sample->psample_group, - act->sample.psample_group); - mlxsw_sp_port->sample->truncate = act->sample.truncate; - mlxsw_sp_port->sample->trunc_size = act->sample.trunc_size; - mlxsw_sp_port->sample->rate = act->sample.rate; - - err = mlxsw_sp_port_sample_set(mlxsw_sp_port, true, act->sample.rate); - if (err) - goto err_port_sample_set; - return 0; - -err_port_sample_set: - RCU_INIT_POINTER(mlxsw_sp_port->sample->psample_group, NULL); - return err; -} - -static void -mlxsw_sp_port_del_cls_matchall_sample(struct mlxsw_sp_port *mlxsw_sp_port) -{ - if (!mlxsw_sp_port->sample) - return; - - mlxsw_sp_port_sample_set(mlxsw_sp_port, false, 1); - RCU_INIT_POINTER(mlxsw_sp_port->sample->psample_group, NULL); -} - -static int mlxsw_sp_port_add_cls_matchall(struct mlxsw_sp_port *mlxsw_sp_port, - struct tc_cls_matchall_offload *f, - bool ingress) -{ - struct mlxsw_sp_port_mall_tc_entry *mall_tc_entry; - __be16 protocol = f->common.protocol; - struct flow_action_entry *act; - int err; - - if (!flow_offload_has_one_action(&f->rule->action)) { - netdev_err(mlxsw_sp_port->dev, "only singular actions are supported\n"); - return -EOPNOTSUPP; - } - - mall_tc_entry = kzalloc(sizeof(*mall_tc_entry), GFP_KERNEL); - if (!mall_tc_entry) - return -ENOMEM; - mall_tc_entry->cookie = f->cookie; - - act = &f->rule->action.entries[0]; - - if (act->id == FLOW_ACTION_MIRRED && protocol == htons(ETH_P_ALL)) { - struct mlxsw_sp_port_mall_mirror_tc_entry *mirror; - - mall_tc_entry->type = MLXSW_SP_PORT_MALL_MIRROR; - mirror = &mall_tc_entry->mirror; - err = mlxsw_sp_port_add_cls_matchall_mirror(mlxsw_sp_port, - mirror, act, - ingress); - } else if (act->id == FLOW_ACTION_SAMPLE && - protocol == htons(ETH_P_ALL)) { - mall_tc_entry->type = MLXSW_SP_PORT_MALL_SAMPLE; - err = mlxsw_sp_port_add_cls_matchall_sample(mlxsw_sp_port, f, - act, ingress); - } else { - err = -EOPNOTSUPP; - } - - if (err) - goto err_add_action; - - list_add_tail(&mall_tc_entry->list, &mlxsw_sp_port->mall_tc_list); - return 0; - -err_add_action: - kfree(mall_tc_entry); - return err; -} - -static void mlxsw_sp_port_del_cls_matchall(struct mlxsw_sp_port *mlxsw_sp_port, - struct tc_cls_matchall_offload *f) -{ - struct mlxsw_sp_port_mall_tc_entry *mall_tc_entry; - - mall_tc_entry = mlxsw_sp_port_mall_tc_entry_find(mlxsw_sp_port, - f->cookie); - if (!mall_tc_entry) { - netdev_dbg(mlxsw_sp_port->dev, "tc entry not found on port\n"); - return; - } - list_del(&mall_tc_entry->list); - - switch (mall_tc_entry->type) { - case MLXSW_SP_PORT_MALL_MIRROR: - mlxsw_sp_port_del_cls_matchall_mirror(mlxsw_sp_port, - &mall_tc_entry->mirror); - break; - case MLXSW_SP_PORT_MALL_SAMPLE: - mlxsw_sp_port_del_cls_matchall_sample(mlxsw_sp_port); - break; - default: - WARN_ON(1); - } - - kfree(mall_tc_entry); -} - -static int mlxsw_sp_setup_tc_cls_matchall(struct mlxsw_sp_port *mlxsw_sp_port, - struct tc_cls_matchall_offload *f, - bool ingress) -{ - switch (f->command) { - case TC_CLSMATCHALL_REPLACE: - return mlxsw_sp_port_add_cls_matchall(mlxsw_sp_port, f, - ingress); - case TC_CLSMATCHALL_DESTROY: - mlxsw_sp_port_del_cls_matchall(mlxsw_sp_port, f); - return 0; - default: - return -EOPNOTSUPP; - } -} - -static int -mlxsw_sp_setup_tc_cls_flower(struct mlxsw_sp_acl_block *acl_block, - struct flow_cls_offload *f) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_acl_block_mlxsw_sp(acl_block); - - switch (f->command) { - case FLOW_CLS_REPLACE: - return mlxsw_sp_flower_replace(mlxsw_sp, acl_block, f); - case FLOW_CLS_DESTROY: - mlxsw_sp_flower_destroy(mlxsw_sp, acl_block, f); - return 0; - case FLOW_CLS_STATS: - return mlxsw_sp_flower_stats(mlxsw_sp, acl_block, f); - case FLOW_CLS_TMPLT_CREATE: - return mlxsw_sp_flower_tmplt_create(mlxsw_sp, acl_block, f); - case FLOW_CLS_TMPLT_DESTROY: - mlxsw_sp_flower_tmplt_destroy(mlxsw_sp, acl_block, f); - return 0; - default: - return -EOPNOTSUPP; - } -} - -static int mlxsw_sp_setup_tc_block_cb_matchall(enum tc_setup_type type, - void *type_data, - void *cb_priv, bool ingress) -{ - struct mlxsw_sp_port *mlxsw_sp_port = cb_priv; - - switch (type) { - case TC_SETUP_CLSMATCHALL: - if (!tc_cls_can_offload_and_chain0(mlxsw_sp_port->dev, - type_data)) - return -EOPNOTSUPP; - - return mlxsw_sp_setup_tc_cls_matchall(mlxsw_sp_port, type_data, - ingress); - case TC_SETUP_CLSFLOWER: - return 0; - default: - return -EOPNOTSUPP; - } -} - -static int mlxsw_sp_setup_tc_block_cb_matchall_ig(enum tc_setup_type type, - void *type_data, - void *cb_priv) -{ - return mlxsw_sp_setup_tc_block_cb_matchall(type, type_data, - cb_priv, true); -} - -static int mlxsw_sp_setup_tc_block_cb_matchall_eg(enum tc_setup_type type, - void *type_data, - void *cb_priv) -{ - return mlxsw_sp_setup_tc_block_cb_matchall(type, type_data, - cb_priv, false); -} - -static int mlxsw_sp_setup_tc_block_cb_flower(enum tc_setup_type type, - void *type_data, void *cb_priv) -{ - struct mlxsw_sp_acl_block *acl_block = cb_priv; - - switch (type) { - case TC_SETUP_CLSMATCHALL: - return 0; - case TC_SETUP_CLSFLOWER: - if (mlxsw_sp_acl_block_disabled(acl_block)) - return -EOPNOTSUPP; - - return mlxsw_sp_setup_tc_cls_flower(acl_block, type_data); - default: - return -EOPNOTSUPP; - } -} - -static void mlxsw_sp_tc_block_flower_release(void *cb_priv) -{ - struct mlxsw_sp_acl_block *acl_block = cb_priv; - - mlxsw_sp_acl_block_destroy(acl_block); -} - -static LIST_HEAD(mlxsw_sp_block_cb_list); - -static int -mlxsw_sp_setup_tc_block_flower_bind(struct mlxsw_sp_port *mlxsw_sp_port, - struct flow_block_offload *f, bool ingress) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - struct mlxsw_sp_acl_block *acl_block; - struct flow_block_cb *block_cb; - bool register_block = false; - int err; - - block_cb = flow_block_cb_lookup(f->block, - mlxsw_sp_setup_tc_block_cb_flower, - mlxsw_sp); - if (!block_cb) { - acl_block = mlxsw_sp_acl_block_create(mlxsw_sp, f->net); - if (!acl_block) - return -ENOMEM; - block_cb = flow_block_cb_alloc(mlxsw_sp_setup_tc_block_cb_flower, - mlxsw_sp, acl_block, - mlxsw_sp_tc_block_flower_release); - if (IS_ERR(block_cb)) { - mlxsw_sp_acl_block_destroy(acl_block); - err = PTR_ERR(block_cb); - goto err_cb_register; - } - register_block = true; - } else { - acl_block = flow_block_cb_priv(block_cb); - } - flow_block_cb_incref(block_cb); - err = mlxsw_sp_acl_block_bind(mlxsw_sp, acl_block, - mlxsw_sp_port, ingress, f->extack); - if (err) - goto err_block_bind; - - if (ingress) - mlxsw_sp_port->ing_acl_block = acl_block; - else - mlxsw_sp_port->eg_acl_block = acl_block; - - if (register_block) { - flow_block_cb_add(block_cb, f); - list_add_tail(&block_cb->driver_list, &mlxsw_sp_block_cb_list); - } - - return 0; - -err_block_bind: - if (!flow_block_cb_decref(block_cb)) - flow_block_cb_free(block_cb); -err_cb_register: - return err; -} - -static void -mlxsw_sp_setup_tc_block_flower_unbind(struct mlxsw_sp_port *mlxsw_sp_port, - struct flow_block_offload *f, bool ingress) -{ - struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; - struct mlxsw_sp_acl_block *acl_block; - struct flow_block_cb *block_cb; - int err; - - block_cb = flow_block_cb_lookup(f->block, - mlxsw_sp_setup_tc_block_cb_flower, - mlxsw_sp); - if (!block_cb) - return; - - if (ingress) - mlxsw_sp_port->ing_acl_block = NULL; - else - mlxsw_sp_port->eg_acl_block = NULL; - - acl_block = flow_block_cb_priv(block_cb); - err = mlxsw_sp_acl_block_unbind(mlxsw_sp, acl_block, - mlxsw_sp_port, ingress); - if (!err && !flow_block_cb_decref(block_cb)) { - flow_block_cb_remove(block_cb, f); - list_del(&block_cb->driver_list); - } -} - -static int mlxsw_sp_setup_tc_block(struct mlxsw_sp_port *mlxsw_sp_port, - struct flow_block_offload *f) -{ - struct flow_block_cb *block_cb; - flow_setup_cb_t *cb; - bool ingress; - int err; - - if (f->binder_type == FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS) { - cb = mlxsw_sp_setup_tc_block_cb_matchall_ig; - ingress = true; - } else if (f->binder_type == FLOW_BLOCK_BINDER_TYPE_CLSACT_EGRESS) { - cb = mlxsw_sp_setup_tc_block_cb_matchall_eg; - ingress = false; - } else { - return -EOPNOTSUPP; - } - - f->driver_block_list = &mlxsw_sp_block_cb_list; - - switch (f->command) { - case FLOW_BLOCK_BIND: - if (flow_block_cb_is_busy(cb, mlxsw_sp_port, - &mlxsw_sp_block_cb_list)) - return -EBUSY; - - block_cb = flow_block_cb_alloc(cb, mlxsw_sp_port, - mlxsw_sp_port, NULL); - if (IS_ERR(block_cb)) - return PTR_ERR(block_cb); - err = mlxsw_sp_setup_tc_block_flower_bind(mlxsw_sp_port, f, - ingress); - if (err) { - flow_block_cb_free(block_cb); - return err; - } - flow_block_cb_add(block_cb, f); - list_add_tail(&block_cb->driver_list, &mlxsw_sp_block_cb_list); - return 0; - case FLOW_BLOCK_UNBIND: - mlxsw_sp_setup_tc_block_flower_unbind(mlxsw_sp_port, - f, ingress); - block_cb = flow_block_cb_lookup(f->block, cb, mlxsw_sp_port); - if (!block_cb) - return -ENOENT; - - flow_block_cb_remove(block_cb, f); - list_del(&block_cb->driver_list); - return 0; - default: - return -EOPNOTSUPP; - } -} - static int mlxsw_sp_setup_tc(struct net_device *dev, enum tc_setup_type type, void *type_data) { @@ -1791,23 +1373,21 @@ static int mlxsw_sp_setup_tc(struct net_device *dev, enum tc_setup_type type, } } - static int mlxsw_sp_feature_hw_tc(struct net_device *dev, bool enable) { struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev); if (!enable) { - if (mlxsw_sp_acl_block_rule_count(mlxsw_sp_port->ing_acl_block) || - mlxsw_sp_acl_block_rule_count(mlxsw_sp_port->eg_acl_block) || - !list_empty(&mlxsw_sp_port->mall_tc_list)) { + if (mlxsw_sp_flow_block_rule_count(mlxsw_sp_port->ing_flow_block) || + mlxsw_sp_flow_block_rule_count(mlxsw_sp_port->eg_flow_block)) { netdev_err(dev, "Active offloaded tc filters, can't turn hw_tc_offload off\n"); return -EINVAL; } - mlxsw_sp_acl_block_disable_inc(mlxsw_sp_port->ing_acl_block); - mlxsw_sp_acl_block_disable_inc(mlxsw_sp_port->eg_acl_block); + mlxsw_sp_flow_block_disable_inc(mlxsw_sp_port->ing_flow_block); + mlxsw_sp_flow_block_disable_inc(mlxsw_sp_port->eg_flow_block); } else { - mlxsw_sp_acl_block_disable_dec(mlxsw_sp_port->ing_acl_block); - mlxsw_sp_acl_block_disable_dec(mlxsw_sp_port->eg_acl_block); + mlxsw_sp_flow_block_disable_dec(mlxsw_sp_port->ing_flow_block); + mlxsw_sp_flow_block_disable_dec(mlxsw_sp_port->eg_flow_block); } return 0; } @@ -3695,7 +3275,6 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, mlxsw_sp_port->mapping = *port_mapping; mlxsw_sp_port->link.autoneg = 1; INIT_LIST_HEAD(&mlxsw_sp_port->vlans_list); - INIT_LIST_HEAD(&mlxsw_sp_port->mall_tc_list); mlxsw_sp_port->pcpu_stats = netdev_alloc_pcpu_stats(struct mlxsw_sp_port_pcpu_stats); @@ -3704,13 +3283,6 @@ static int mlxsw_sp_port_create(struct mlxsw_sp *mlxsw_sp, u8 local_port, goto err_alloc_stats; } - mlxsw_sp_port->sample = kzalloc(sizeof(*mlxsw_sp_port->sample), - GFP_KERNEL); - if (!mlxsw_sp_port->sample) { - err = -ENOMEM; - goto err_alloc_sample; - } - INIT_DELAYED_WORK(&mlxsw_sp_port->periodic_hw_stats.update_dw, &update_stats_cache); @@ -3897,8 +3469,6 @@ err_dev_addr_init: err_port_swid_set: mlxsw_sp_port_module_unmap(mlxsw_sp_port); err_port_module_map: - kfree(mlxsw_sp_port->sample); -err_alloc_sample: free_percpu(mlxsw_sp_port->pcpu_stats); err_alloc_stats: free_netdev(dev); @@ -3926,7 +3496,6 @@ static void mlxsw_sp_port_remove(struct mlxsw_sp *mlxsw_sp, u8 local_port) mlxsw_sp_port_tc_mc_mode_set(mlxsw_sp_port, false); mlxsw_sp_port_swid_set(mlxsw_sp_port, MLXSW_PORT_SWID_DISABLED_PORT); mlxsw_sp_port_module_unmap(mlxsw_sp_port); - kfree(mlxsw_sp_port->sample); free_percpu(mlxsw_sp_port->pcpu_stats); WARN_ON_ONCE(!list_empty(&mlxsw_sp_port->vlans_list)); free_netdev(mlxsw_sp_port->dev); @@ -3986,6 +3555,7 @@ static void mlxsw_sp_ports_remove(struct mlxsw_sp *mlxsw_sp) mlxsw_sp_port_remove(mlxsw_sp, i); mlxsw_sp_cpu_port_remove(mlxsw_sp); kfree(mlxsw_sp->ports); + mlxsw_sp->ports = NULL; } static int mlxsw_sp_ports_create(struct mlxsw_sp *mlxsw_sp) @@ -4022,6 +3592,7 @@ err_port_create: mlxsw_sp_cpu_port_remove(mlxsw_sp); err_cpu_port_create: kfree(mlxsw_sp->ports); + mlxsw_sp->ports = NULL; return err; } @@ -4143,6 +3714,14 @@ static int mlxsw_sp_local_ports_offset(struct mlxsw_core *mlxsw_core, return mlxsw_core_res_get(mlxsw_core, local_ports_in_x_res_id); } +static struct mlxsw_sp_port * +mlxsw_sp_port_get_by_local_port(struct mlxsw_sp *mlxsw_sp, u8 local_port) +{ + if (mlxsw_sp->ports && mlxsw_sp->ports[local_port]) + return mlxsw_sp->ports[local_port]; + return NULL; +} + static int mlxsw_sp_port_split(struct mlxsw_core *mlxsw_core, u8 local_port, unsigned int count, struct netlink_ext_ack *extack) @@ -4156,7 +3735,7 @@ static int mlxsw_sp_port_split(struct mlxsw_core *mlxsw_core, u8 local_port, int i; int err; - mlxsw_sp_port = mlxsw_sp->ports[local_port]; + mlxsw_sp_port = mlxsw_sp_port_get_by_local_port(mlxsw_sp, local_port); if (!mlxsw_sp_port) { dev_err(mlxsw_sp->bus_info->dev, "Port number \"%d\" does not exist\n", local_port); @@ -4251,7 +3830,7 @@ static int mlxsw_sp_port_unsplit(struct mlxsw_core *mlxsw_core, u8 local_port, int offset; int i; - mlxsw_sp_port = mlxsw_sp->ports[local_port]; + mlxsw_sp_port = mlxsw_sp_port_get_by_local_port(mlxsw_sp, local_port); if (!mlxsw_sp_port) { dev_err(mlxsw_sp->bus_info->dev, "Port number \"%d\" does not exist\n", local_port); @@ -4408,12 +3987,17 @@ static void mlxsw_sp_rx_listener_l3_mark_func(struct sk_buff *skb, return mlxsw_sp_rx_listener_no_mark_func(skb, local_port, priv); } -static void mlxsw_sp_rx_listener_sample_func(struct sk_buff *skb, u8 local_port, - void *priv) +void mlxsw_sp_ptp_receive(struct mlxsw_sp *mlxsw_sp, struct sk_buff *skb, + u8 local_port) +{ + mlxsw_sp->ptp_ops->receive(mlxsw_sp, skb, local_port); +} + +void mlxsw_sp_sample_receive(struct mlxsw_sp *mlxsw_sp, struct sk_buff *skb, + u8 local_port) { - struct mlxsw_sp *mlxsw_sp = priv; struct mlxsw_sp_port *mlxsw_sp_port = mlxsw_sp->ports[local_port]; - struct psample_group *psample_group; + struct mlxsw_sp_port_sample *sample; u32 size; if (unlikely(!mlxsw_sp_port)) { @@ -4421,36 +4005,20 @@ static void mlxsw_sp_rx_listener_sample_func(struct sk_buff *skb, u8 local_port, local_port); goto out; } - if (unlikely(!mlxsw_sp_port->sample)) { - dev_warn_ratelimited(mlxsw_sp->bus_info->dev, "Port %d: sample skb received on unsupported port\n", - local_port); - goto out; - } - - size = mlxsw_sp_port->sample->truncate ? - mlxsw_sp_port->sample->trunc_size : skb->len; rcu_read_lock(); - psample_group = rcu_dereference(mlxsw_sp_port->sample->psample_group); - if (!psample_group) + sample = rcu_dereference(mlxsw_sp_port->sample); + if (!sample) goto out_unlock; - psample_sample_packet(psample_group, skb, size, - mlxsw_sp_port->dev->ifindex, 0, - mlxsw_sp_port->sample->rate); + size = sample->truncate ? sample->trunc_size : skb->len; + psample_sample_packet(sample->psample_group, skb, size, + mlxsw_sp_port->dev->ifindex, 0, sample->rate); out_unlock: rcu_read_unlock(); out: consume_skb(skb); } -static void mlxsw_sp_rx_listener_ptp(struct sk_buff *skb, u8 local_port, - void *priv) -{ - struct mlxsw_sp *mlxsw_sp = priv; - - mlxsw_sp->ptp_ops->receive(mlxsw_sp, skb, local_port); -} - #define MLXSW_SP_RXL_NO_MARK(_trap_id, _action, _trap_group, _is_ctrl) \ MLXSW_RXL(mlxsw_sp_rx_listener_no_mark_func, _trap_id, _action, \ _is_ctrl, SP_##_trap_group, DISCARD) @@ -4470,58 +4038,13 @@ static const struct mlxsw_listener mlxsw_sp_listener[] = { /* Events */ MLXSW_SP_EVENTL(mlxsw_sp_pude_event_func, PUDE), /* L2 traps */ - MLXSW_SP_RXL_NO_MARK(STP, TRAP_TO_CPU, STP, true), - MLXSW_SP_RXL_NO_MARK(LACP, TRAP_TO_CPU, LACP, true), - MLXSW_RXL(mlxsw_sp_rx_listener_ptp, LLDP, TRAP_TO_CPU, - false, SP_LLDP, DISCARD), - MLXSW_SP_RXL_MARK(DHCP, MIRROR_TO_CPU, DHCP, false), - MLXSW_SP_RXL_MARK(IGMP_QUERY, MIRROR_TO_CPU, IGMP, false), - MLXSW_SP_RXL_NO_MARK(IGMP_V1_REPORT, TRAP_TO_CPU, IGMP, false), - MLXSW_SP_RXL_NO_MARK(IGMP_V2_REPORT, TRAP_TO_CPU, IGMP, false), - MLXSW_SP_RXL_NO_MARK(IGMP_V2_LEAVE, TRAP_TO_CPU, IGMP, false), - MLXSW_SP_RXL_NO_MARK(IGMP_V3_REPORT, TRAP_TO_CPU, IGMP, false), - MLXSW_SP_RXL_MARK(ARPBC, MIRROR_TO_CPU, ARP, false), - MLXSW_SP_RXL_MARK(ARPUC, MIRROR_TO_CPU, ARP, false), - MLXSW_SP_RXL_NO_MARK(FID_MISS, TRAP_TO_CPU, IP2ME, false), - MLXSW_SP_RXL_MARK(IPV6_MLDV12_LISTENER_QUERY, MIRROR_TO_CPU, IPV6_MLD, - false), - MLXSW_SP_RXL_NO_MARK(IPV6_MLDV1_LISTENER_REPORT, TRAP_TO_CPU, IPV6_MLD, - false), - MLXSW_SP_RXL_NO_MARK(IPV6_MLDV1_LISTENER_DONE, TRAP_TO_CPU, IPV6_MLD, - false), - MLXSW_SP_RXL_NO_MARK(IPV6_MLDV2_LISTENER_REPORT, TRAP_TO_CPU, IPV6_MLD, - false), + MLXSW_SP_RXL_NO_MARK(FID_MISS, TRAP_TO_CPU, FID_MISS, false), /* L3 traps */ - MLXSW_SP_RXL_L3_MARK(LBERROR, MIRROR_TO_CPU, LBERROR, false), - MLXSW_SP_RXL_MARK(IP2ME, TRAP_TO_CPU, IP2ME, false), MLXSW_SP_RXL_MARK(IPV6_UNSPECIFIED_ADDRESS, TRAP_TO_CPU, ROUTER_EXP, false), - MLXSW_SP_RXL_MARK(IPV6_LINK_LOCAL_DEST, TRAP_TO_CPU, ROUTER_EXP, false), MLXSW_SP_RXL_MARK(IPV6_LINK_LOCAL_SRC, TRAP_TO_CPU, ROUTER_EXP, false), - MLXSW_SP_RXL_MARK(IPV6_ALL_NODES_LINK, TRAP_TO_CPU, ROUTER_EXP, false), - MLXSW_SP_RXL_MARK(IPV6_ALL_ROUTERS_LINK, TRAP_TO_CPU, ROUTER_EXP, - false), - MLXSW_SP_RXL_MARK(IPV4_OSPF, TRAP_TO_CPU, OSPF, false), - MLXSW_SP_RXL_MARK(IPV6_OSPF, TRAP_TO_CPU, OSPF, false), - MLXSW_SP_RXL_MARK(IPV6_DHCP, TRAP_TO_CPU, DHCP, false), - MLXSW_SP_RXL_MARK(RTR_INGRESS0, TRAP_TO_CPU, REMOTE_ROUTE, false), - MLXSW_SP_RXL_MARK(IPV4_BGP, TRAP_TO_CPU, BGP, false), - MLXSW_SP_RXL_MARK(IPV6_BGP, TRAP_TO_CPU, BGP, false), - MLXSW_SP_RXL_MARK(L3_IPV6_ROUTER_SOLICITATION, TRAP_TO_CPU, IPV6_ND, - false), - MLXSW_SP_RXL_MARK(L3_IPV6_ROUTER_ADVERTISMENT, TRAP_TO_CPU, IPV6_ND, - false), - MLXSW_SP_RXL_MARK(L3_IPV6_NEIGHBOR_SOLICITATION, TRAP_TO_CPU, IPV6_ND, - false), - MLXSW_SP_RXL_MARK(L3_IPV6_NEIGHBOR_ADVERTISMENT, TRAP_TO_CPU, IPV6_ND, - false), - MLXSW_SP_RXL_MARK(L3_IPV6_REDIRECTION, TRAP_TO_CPU, IPV6_ND, false), MLXSW_SP_RXL_MARK(IPV6_MC_LINK_LOCAL_DEST, TRAP_TO_CPU, ROUTER_EXP, false), - MLXSW_SP_RXL_MARK(ROUTER_ALERT_IPV4, TRAP_TO_CPU, ROUTER_EXP, false), - MLXSW_SP_RXL_MARK(ROUTER_ALERT_IPV6, TRAP_TO_CPU, ROUTER_EXP, false), - MLXSW_SP_RXL_MARK(IPV4_VRRP, TRAP_TO_CPU, VRRP, false), - MLXSW_SP_RXL_MARK(IPV6_VRRP, TRAP_TO_CPU, VRRP, false), MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_SIP_CLASS_E, FORWARD, ROUTER_EXP, false), MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_MC_DMAC, FORWARD, @@ -4530,23 +4053,11 @@ static const struct mlxsw_listener mlxsw_sp_listener[] = { ROUTER_EXP, false), MLXSW_SP_RXL_NO_MARK(DISCARD_ING_ROUTER_DIP_LINK_LOCAL, FORWARD, ROUTER_EXP, false), - /* PKT Sample trap */ - MLXSW_RXL(mlxsw_sp_rx_listener_sample_func, PKT_SAMPLE, MIRROR_TO_CPU, - false, SP_IP2ME, DISCARD), - /* ACL trap */ - MLXSW_SP_RXL_NO_MARK(ACL0, TRAP_TO_CPU, IP2ME, false), /* Multicast Router Traps */ - MLXSW_SP_RXL_MARK(IPV4_PIM, TRAP_TO_CPU, PIM, false), - MLXSW_SP_RXL_MARK(IPV6_PIM, TRAP_TO_CPU, PIM, false), MLXSW_SP_RXL_MARK(ACL1, TRAP_TO_CPU, MULTICAST, false), MLXSW_SP_RXL_L3_MARK(ACL2, TRAP_TO_CPU, MULTICAST, false), /* NVE traps */ - MLXSW_SP_RXL_MARK(NVE_ENCAP_ARP, TRAP_TO_CPU, ARP, false), - MLXSW_SP_RXL_NO_MARK(NVE_DECAP_ARP, TRAP_TO_CPU, ARP, false), - /* PTP traps */ - MLXSW_RXL(mlxsw_sp_rx_listener_ptp, PTP0, TRAP_TO_CPU, - false, SP_PTP0, DISCARD), - MLXSW_SP_RXL_NO_MARK(PTP1, TRAP_TO_CPU, PTP1, false), + MLXSW_SP_RXL_MARK(NVE_ENCAP_ARP, TRAP_TO_CPU, NEIGH_DISCOVERY, false), }; static const struct mlxsw_listener mlxsw_sp1_listener[] = { @@ -4575,46 +4086,12 @@ static int mlxsw_sp_cpu_policers_set(struct mlxsw_core *mlxsw_core) for (i = 0; i < max_cpu_policers; i++) { is_bytes = false; switch (i) { - case MLXSW_REG_HTGT_TRAP_GROUP_SP_STP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_LACP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_LLDP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_OSPF: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_PIM: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_LBERROR: - rate = 128; - burst_size = 7; - break; - case MLXSW_REG_HTGT_TRAP_GROUP_SP_IGMP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_IPV6_MLD: - rate = 16 * 1024; - burst_size = 10; - break; - case MLXSW_REG_HTGT_TRAP_GROUP_SP_BGP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_ARP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_DHCP: case MLXSW_REG_HTGT_TRAP_GROUP_SP_ROUTER_EXP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_REMOTE_ROUTE: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_IPV6_ND: case MLXSW_REG_HTGT_TRAP_GROUP_SP_MULTICAST: + case MLXSW_REG_HTGT_TRAP_GROUP_SP_FID_MISS: rate = 1024; burst_size = 7; break; - case MLXSW_REG_HTGT_TRAP_GROUP_SP_IP2ME: - rate = 1024; - burst_size = 7; - break; - case MLXSW_REG_HTGT_TRAP_GROUP_SP_PTP0: - rate = 24 * 1024; - burst_size = 12; - break; - case MLXSW_REG_HTGT_TRAP_GROUP_SP_PTP1: - rate = 19 * 1024; - burst_size = 12; - break; - case MLXSW_REG_HTGT_TRAP_GROUP_SP_VRRP: - rate = 360; - burst_size = 7; - break; default: continue; } @@ -4649,43 +4126,12 @@ static int mlxsw_sp_trap_groups_set(struct mlxsw_core *mlxsw_core) for (i = 0; i < max_trap_groups; i++) { policer_id = i; switch (i) { - case MLXSW_REG_HTGT_TRAP_GROUP_SP_STP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_LACP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_LLDP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_OSPF: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_PIM: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_PTP0: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_VRRP: - priority = 5; - tc = 5; - break; - case MLXSW_REG_HTGT_TRAP_GROUP_SP_BGP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_DHCP: - priority = 4; - tc = 4; - break; - case MLXSW_REG_HTGT_TRAP_GROUP_SP_IGMP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_IP2ME: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_IPV6_MLD: - priority = 3; - tc = 3; - break; - case MLXSW_REG_HTGT_TRAP_GROUP_SP_ARP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_IPV6_ND: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_PTP1: - priority = 2; - tc = 2; - break; case MLXSW_REG_HTGT_TRAP_GROUP_SP_ROUTER_EXP: - case MLXSW_REG_HTGT_TRAP_GROUP_SP_REMOTE_ROUTE: case MLXSW_REG_HTGT_TRAP_GROUP_SP_MULTICAST: + case MLXSW_REG_HTGT_TRAP_GROUP_SP_FID_MISS: priority = 1; tc = 1; break; - case MLXSW_REG_HTGT_TRAP_GROUP_SP_LBERROR: - priority = 0; - tc = 1; - break; case MLXSW_REG_HTGT_TRAP_GROUP_SP_EVENT: priority = MLXSW_REG_HTGT_DEFAULT_PRIORITY; tc = MLXSW_REG_HTGT_DEFAULT_TC; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h index ca56e72cb4b7..6f96ca50c9ba 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h @@ -109,25 +109,6 @@ struct mlxsw_sp_mid { unsigned long *ports_in_mid; /* bits array */ }; -enum mlxsw_sp_port_mall_action_type { - MLXSW_SP_PORT_MALL_MIRROR, - MLXSW_SP_PORT_MALL_SAMPLE, -}; - -struct mlxsw_sp_port_mall_mirror_tc_entry { - int span_id; - bool ingress; -}; - -struct mlxsw_sp_port_mall_tc_entry { - struct list_head list; - unsigned long cookie; - enum mlxsw_sp_port_mall_action_type type; - union { - struct mlxsw_sp_port_mall_mirror_tc_entry mirror; - }; -}; - struct mlxsw_sp_sb; struct mlxsw_sp_bridge; struct mlxsw_sp_router; @@ -211,7 +192,7 @@ struct mlxsw_sp_port_pcpu_stats { }; struct mlxsw_sp_port_sample { - struct psample_group __rcu *psample_group; + struct psample_group *psample_group; u32 trunc_size; u32 rate; bool truncate; @@ -274,21 +255,19 @@ struct mlxsw_sp_port { * the same localport can have * different mapping. */ - /* TC handles */ - struct list_head mall_tc_list; struct { #define MLXSW_HW_STATS_UPDATE_TIME HZ struct rtnl_link_stats64 stats; struct mlxsw_sp_port_xstats xstats; struct delayed_work update_dw; } periodic_hw_stats; - struct mlxsw_sp_port_sample *sample; + struct mlxsw_sp_port_sample __rcu *sample; struct list_head vlans_list; struct mlxsw_sp_port_vlan *default_vlan; struct mlxsw_sp_qdisc_state *qdisc; unsigned acl_rule_count; - struct mlxsw_sp_acl_block *ing_acl_block; - struct mlxsw_sp_acl_block *eg_acl_block; + struct mlxsw_sp_flow_block *ing_flow_block; + struct mlxsw_sp_flow_block *eg_flow_block; struct { struct delayed_work shaper_dw; struct hwtstamp_config hwtstamp_config; @@ -472,6 +451,10 @@ extern struct notifier_block mlxsw_sp_switchdev_notifier; /* spectrum.c */ void mlxsw_sp_rx_listener_no_mark_func(struct sk_buff *skb, u8 local_port, void *priv); +void mlxsw_sp_ptp_receive(struct mlxsw_sp *mlxsw_sp, struct sk_buff *skb, + u8 local_port); +void mlxsw_sp_sample_receive(struct mlxsw_sp *mlxsw_sp, struct sk_buff *skb, + u8 local_port); int mlxsw_sp_port_speed_get(struct mlxsw_sp_port *mlxsw_sp_port, u32 *speed); int mlxsw_sp_port_ets_set(struct mlxsw_sp_port *mlxsw_sp_port, enum mlxsw_reg_qeec_hr hr, u8 index, u8 next_index, @@ -654,17 +637,14 @@ struct mlxsw_sp_acl_rule_info { unsigned int counter_index; }; -struct mlxsw_sp_acl_block; -struct mlxsw_sp_acl_ruleset; - -/* spectrum_acl.c */ -enum mlxsw_sp_acl_profile { - MLXSW_SP_ACL_PROFILE_FLOWER, - MLXSW_SP_ACL_PROFILE_MR, -}; - -struct mlxsw_sp_acl_block { +/* spectrum_flow.c */ +struct mlxsw_sp_flow_block { struct list_head binding_list; + struct { + struct list_head list; + unsigned int min_prio; + unsigned int max_prio; + } mall; struct mlxsw_sp_acl_ruleset *ruleset_zero; struct mlxsw_sp *mlxsw_sp; unsigned int rule_count; @@ -676,40 +656,100 @@ struct mlxsw_sp_acl_block { struct net *net; }; +struct mlxsw_sp_flow_block_binding { + struct list_head list; + struct net_device *dev; + struct mlxsw_sp_port *mlxsw_sp_port; + bool ingress; +}; + +static inline struct mlxsw_sp * +mlxsw_sp_flow_block_mlxsw_sp(struct mlxsw_sp_flow_block *block) +{ + return block->mlxsw_sp; +} + +static inline unsigned int +mlxsw_sp_flow_block_rule_count(const struct mlxsw_sp_flow_block *block) +{ + return block ? block->rule_count : 0; +} + +static inline void +mlxsw_sp_flow_block_disable_inc(struct mlxsw_sp_flow_block *block) +{ + if (block) + block->disable_count++; +} + +static inline void +mlxsw_sp_flow_block_disable_dec(struct mlxsw_sp_flow_block *block) +{ + if (block) + block->disable_count--; +} + +static inline bool +mlxsw_sp_flow_block_disabled(const struct mlxsw_sp_flow_block *block) +{ + return block->disable_count; +} + +static inline bool +mlxsw_sp_flow_block_is_egress_bound(const struct mlxsw_sp_flow_block *block) +{ + return block->egress_binding_count; +} + +static inline bool +mlxsw_sp_flow_block_is_ingress_bound(const struct mlxsw_sp_flow_block *block) +{ + return block->ingress_binding_count; +} + +static inline bool +mlxsw_sp_flow_block_is_mixed_bound(const struct mlxsw_sp_flow_block *block) +{ + return block->ingress_binding_count && block->egress_binding_count; +} + +struct mlxsw_sp_flow_block *mlxsw_sp_flow_block_create(struct mlxsw_sp *mlxsw_sp, + struct net *net); +void mlxsw_sp_flow_block_destroy(struct mlxsw_sp_flow_block *block); +int mlxsw_sp_setup_tc_block(struct mlxsw_sp_port *mlxsw_sp_port, + struct flow_block_offload *f); + +/* spectrum_acl.c */ +struct mlxsw_sp_acl_ruleset; + +enum mlxsw_sp_acl_profile { + MLXSW_SP_ACL_PROFILE_FLOWER, + MLXSW_SP_ACL_PROFILE_MR, +}; + struct mlxsw_afk *mlxsw_sp_acl_afk(struct mlxsw_sp_acl *acl); -struct mlxsw_sp *mlxsw_sp_acl_block_mlxsw_sp(struct mlxsw_sp_acl_block *block); -unsigned int -mlxsw_sp_acl_block_rule_count(const struct mlxsw_sp_acl_block *block); -void mlxsw_sp_acl_block_disable_inc(struct mlxsw_sp_acl_block *block); -void mlxsw_sp_acl_block_disable_dec(struct mlxsw_sp_acl_block *block); -bool mlxsw_sp_acl_block_disabled(const struct mlxsw_sp_acl_block *block); -struct mlxsw_sp_acl_block *mlxsw_sp_acl_block_create(struct mlxsw_sp *mlxsw_sp, - struct net *net); -void mlxsw_sp_acl_block_destroy(struct mlxsw_sp_acl_block *block); -int mlxsw_sp_acl_block_bind(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, - struct mlxsw_sp_port *mlxsw_sp_port, - bool ingress, - struct netlink_ext_ack *extack); -int mlxsw_sp_acl_block_unbind(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, - struct mlxsw_sp_port *mlxsw_sp_port, - bool ingress); -bool mlxsw_sp_acl_block_is_egress_bound(const struct mlxsw_sp_acl_block *block); -bool mlxsw_sp_acl_block_is_ingress_bound(const struct mlxsw_sp_acl_block *block); -bool mlxsw_sp_acl_block_is_mixed_bound(const struct mlxsw_sp_acl_block *block); + +int mlxsw_sp_acl_ruleset_bind(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_flow_block *block, + struct mlxsw_sp_flow_block_binding *binding); +void mlxsw_sp_acl_ruleset_unbind(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_flow_block *block, + struct mlxsw_sp_flow_block_binding *binding); struct mlxsw_sp_acl_ruleset * mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, u32 chain_index, + struct mlxsw_sp_flow_block *block, u32 chain_index, enum mlxsw_sp_acl_profile profile); struct mlxsw_sp_acl_ruleset * mlxsw_sp_acl_ruleset_get(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, u32 chain_index, + struct mlxsw_sp_flow_block *block, u32 chain_index, enum mlxsw_sp_acl_profile profile, struct mlxsw_afk_element_usage *tmplt_elusage); void mlxsw_sp_acl_ruleset_put(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_ruleset *ruleset); u16 mlxsw_sp_acl_ruleset_group_id(struct mlxsw_sp_acl_ruleset *ruleset); +void mlxsw_sp_acl_ruleset_prio_get(struct mlxsw_sp_acl_ruleset *ruleset, + unsigned int *p_min_prio, + unsigned int *p_max_prio); struct mlxsw_sp_acl_rule_info * mlxsw_sp_acl_rulei_create(struct mlxsw_sp_acl *acl, @@ -736,7 +776,7 @@ int mlxsw_sp_acl_rulei_act_drop(struct mlxsw_sp_acl_rule_info *rulei, int mlxsw_sp_acl_rulei_act_trap(struct mlxsw_sp_acl_rule_info *rulei); int mlxsw_sp_acl_rulei_act_mirror(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_rule_info *rulei, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct net_device *out_dev, struct netlink_ext_ack *extack); int mlxsw_sp_acl_rulei_act_fwd(struct mlxsw_sp *mlxsw_sp, @@ -857,22 +897,39 @@ extern const struct mlxsw_afa_ops mlxsw_sp2_act_afa_ops; extern const struct mlxsw_afk_ops mlxsw_sp1_afk_ops; extern const struct mlxsw_afk_ops mlxsw_sp2_afk_ops; +/* spectrum_matchall.c */ +int mlxsw_sp_mall_replace(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_flow_block *block, + struct tc_cls_matchall_offload *f); +void mlxsw_sp_mall_destroy(struct mlxsw_sp_flow_block *block, + struct tc_cls_matchall_offload *f); +int mlxsw_sp_mall_port_bind(struct mlxsw_sp_flow_block *block, + struct mlxsw_sp_port *mlxsw_sp_port); +void mlxsw_sp_mall_port_unbind(struct mlxsw_sp_flow_block *block, + struct mlxsw_sp_port *mlxsw_sp_port); +int mlxsw_sp_mall_prio_get(struct mlxsw_sp_flow_block *block, u32 chain_index, + unsigned int *p_min_prio, unsigned int *p_max_prio); + /* spectrum_flower.c */ int mlxsw_sp_flower_replace(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct flow_cls_offload *f); void mlxsw_sp_flower_destroy(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct flow_cls_offload *f); int mlxsw_sp_flower_stats(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct flow_cls_offload *f); int mlxsw_sp_flower_tmplt_create(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct flow_cls_offload *f); void mlxsw_sp_flower_tmplt_destroy(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct flow_cls_offload *f); +int mlxsw_sp_flower_prio_get(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_flow_block *block, + u32 chain_index, unsigned int *p_min_prio, + unsigned int *p_max_prio); /* spectrum_qdisc.c */ int mlxsw_sp_tc_qdisc_init(struct mlxsw_sp_port *mlxsw_sp_port); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum2_mr_tcam.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum2_mr_tcam.c index e31ec75ac035..a11d911302f1 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum2_mr_tcam.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum2_mr_tcam.c @@ -9,7 +9,7 @@ struct mlxsw_sp2_mr_tcam { struct mlxsw_sp *mlxsw_sp; - struct mlxsw_sp_acl_block *acl_block; + struct mlxsw_sp_flow_block *flow_block; struct mlxsw_sp_acl_ruleset *ruleset4; struct mlxsw_sp_acl_ruleset *ruleset6; }; @@ -61,7 +61,7 @@ static int mlxsw_sp2_mr_tcam_ipv4_init(struct mlxsw_sp2_mr_tcam *mr_tcam) mlxsw_sp2_mr_tcam_usage_ipv4, ARRAY_SIZE(mlxsw_sp2_mr_tcam_usage_ipv4)); mr_tcam->ruleset4 = mlxsw_sp_acl_ruleset_get(mr_tcam->mlxsw_sp, - mr_tcam->acl_block, + mr_tcam->flow_block, MLXSW_SP_L3_PROTO_IPV4, MLXSW_SP_ACL_PROFILE_MR, &elusage); @@ -111,7 +111,7 @@ static int mlxsw_sp2_mr_tcam_ipv6_init(struct mlxsw_sp2_mr_tcam *mr_tcam) mlxsw_sp2_mr_tcam_usage_ipv6, ARRAY_SIZE(mlxsw_sp2_mr_tcam_usage_ipv6)); mr_tcam->ruleset6 = mlxsw_sp_acl_ruleset_get(mr_tcam->mlxsw_sp, - mr_tcam->acl_block, + mr_tcam->flow_block, MLXSW_SP_L3_PROTO_IPV6, MLXSW_SP_ACL_PROFILE_MR, &elusage); @@ -289,8 +289,8 @@ static int mlxsw_sp2_mr_tcam_init(struct mlxsw_sp *mlxsw_sp, void *priv) int err; mr_tcam->mlxsw_sp = mlxsw_sp; - mr_tcam->acl_block = mlxsw_sp_acl_block_create(mlxsw_sp, NULL); - if (!mr_tcam->acl_block) + mr_tcam->flow_block = mlxsw_sp_flow_block_create(mlxsw_sp, NULL); + if (!mr_tcam->flow_block) return -ENOMEM; err = mlxsw_sp2_mr_tcam_ipv4_init(mr_tcam); @@ -306,7 +306,7 @@ static int mlxsw_sp2_mr_tcam_init(struct mlxsw_sp *mlxsw_sp, void *priv) err_ipv6_init: mlxsw_sp2_mr_tcam_ipv4_fini(mr_tcam); err_ipv4_init: - mlxsw_sp_acl_block_destroy(mr_tcam->acl_block); + mlxsw_sp_flow_block_destroy(mr_tcam->flow_block); return err; } @@ -316,7 +316,7 @@ static void mlxsw_sp2_mr_tcam_fini(void *priv) mlxsw_sp2_mr_tcam_ipv6_fini(mr_tcam); mlxsw_sp2_mr_tcam_ipv4_fini(mr_tcam); - mlxsw_sp_acl_block_destroy(mr_tcam->acl_block); + mlxsw_sp_flow_block_destroy(mr_tcam->flow_block); } const struct mlxsw_sp_mr_tcam_ops mlxsw_sp2_mr_tcam_ops = { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c index 01cff711bbd2..47da9ee0045d 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c @@ -40,15 +40,8 @@ struct mlxsw_afk *mlxsw_sp_acl_afk(struct mlxsw_sp_acl *acl) return acl->afk; } -struct mlxsw_sp_acl_block_binding { - struct list_head list; - struct net_device *dev; - struct mlxsw_sp_port *mlxsw_sp_port; - bool ingress; -}; - struct mlxsw_sp_acl_ruleset_ht_key { - struct mlxsw_sp_acl_block *block; + struct mlxsw_sp_flow_block *block; u32 chain_index; const struct mlxsw_sp_acl_profile_ops *ops; }; @@ -58,6 +51,8 @@ struct mlxsw_sp_acl_ruleset { struct mlxsw_sp_acl_ruleset_ht_key ht_key; struct rhashtable rule_ht; unsigned int ref_count; + unsigned int min_prio; + unsigned int max_prio; unsigned long priv[]; /* priv has to be always the last item */ }; @@ -94,49 +89,6 @@ struct mlxsw_sp_fid *mlxsw_sp_acl_dummy_fid(struct mlxsw_sp *mlxsw_sp) return mlxsw_sp->acl->dummy_fid; } -struct mlxsw_sp *mlxsw_sp_acl_block_mlxsw_sp(struct mlxsw_sp_acl_block *block) -{ - return block->mlxsw_sp; -} - -unsigned int -mlxsw_sp_acl_block_rule_count(const struct mlxsw_sp_acl_block *block) -{ - return block ? block->rule_count : 0; -} - -void mlxsw_sp_acl_block_disable_inc(struct mlxsw_sp_acl_block *block) -{ - if (block) - block->disable_count++; -} - -void mlxsw_sp_acl_block_disable_dec(struct mlxsw_sp_acl_block *block) -{ - if (block) - block->disable_count--; -} - -bool mlxsw_sp_acl_block_disabled(const struct mlxsw_sp_acl_block *block) -{ - return block->disable_count; -} - -bool mlxsw_sp_acl_block_is_egress_bound(const struct mlxsw_sp_acl_block *block) -{ - return block->egress_binding_count; -} - -bool mlxsw_sp_acl_block_is_ingress_bound(const struct mlxsw_sp_acl_block *block) -{ - return block->ingress_binding_count; -} - -bool mlxsw_sp_acl_block_is_mixed_bound(const struct mlxsw_sp_acl_block *block) -{ - return block->ingress_binding_count && block->egress_binding_count; -} - static bool mlxsw_sp_acl_ruleset_is_singular(const struct mlxsw_sp_acl_ruleset *ruleset) { @@ -144,10 +96,9 @@ mlxsw_sp_acl_ruleset_is_singular(const struct mlxsw_sp_acl_ruleset *ruleset) return ruleset->ref_count == 2; } -static int -mlxsw_sp_acl_ruleset_bind(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, - struct mlxsw_sp_acl_block_binding *binding) +int mlxsw_sp_acl_ruleset_bind(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_flow_block *block, + struct mlxsw_sp_flow_block_binding *binding) { struct mlxsw_sp_acl_ruleset *ruleset = block->ruleset_zero; const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops; @@ -156,10 +107,9 @@ mlxsw_sp_acl_ruleset_bind(struct mlxsw_sp *mlxsw_sp, binding->mlxsw_sp_port, binding->ingress); } -static void -mlxsw_sp_acl_ruleset_unbind(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, - struct mlxsw_sp_acl_block_binding *binding) +void mlxsw_sp_acl_ruleset_unbind(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_flow_block *block, + struct mlxsw_sp_flow_block_binding *binding) { struct mlxsw_sp_acl_ruleset *ruleset = block->ruleset_zero; const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops; @@ -168,18 +118,12 @@ mlxsw_sp_acl_ruleset_unbind(struct mlxsw_sp *mlxsw_sp, binding->mlxsw_sp_port, binding->ingress); } -static bool -mlxsw_sp_acl_ruleset_block_bound(const struct mlxsw_sp_acl_block *block) -{ - return block->ruleset_zero; -} - static int mlxsw_sp_acl_ruleset_block_bind(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_ruleset *ruleset, - struct mlxsw_sp_acl_block *block) + struct mlxsw_sp_flow_block *block) { - struct mlxsw_sp_acl_block_binding *binding; + struct mlxsw_sp_flow_block_binding *binding; int err; block->ruleset_zero = ruleset; @@ -202,122 +146,18 @@ rollback: static void mlxsw_sp_acl_ruleset_block_unbind(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_ruleset *ruleset, - struct mlxsw_sp_acl_block *block) + struct mlxsw_sp_flow_block *block) { - struct mlxsw_sp_acl_block_binding *binding; + struct mlxsw_sp_flow_block_binding *binding; list_for_each_entry(binding, &block->binding_list, list) mlxsw_sp_acl_ruleset_unbind(mlxsw_sp, block, binding); block->ruleset_zero = NULL; } -struct mlxsw_sp_acl_block *mlxsw_sp_acl_block_create(struct mlxsw_sp *mlxsw_sp, - struct net *net) -{ - struct mlxsw_sp_acl_block *block; - - block = kzalloc(sizeof(*block), GFP_KERNEL); - if (!block) - return NULL; - INIT_LIST_HEAD(&block->binding_list); - block->mlxsw_sp = mlxsw_sp; - block->net = net; - return block; -} - -void mlxsw_sp_acl_block_destroy(struct mlxsw_sp_acl_block *block) -{ - WARN_ON(!list_empty(&block->binding_list)); - kfree(block); -} - -static struct mlxsw_sp_acl_block_binding * -mlxsw_sp_acl_block_lookup(struct mlxsw_sp_acl_block *block, - struct mlxsw_sp_port *mlxsw_sp_port, bool ingress) -{ - struct mlxsw_sp_acl_block_binding *binding; - - list_for_each_entry(binding, &block->binding_list, list) - if (binding->mlxsw_sp_port == mlxsw_sp_port && - binding->ingress == ingress) - return binding; - return NULL; -} - -int mlxsw_sp_acl_block_bind(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, - struct mlxsw_sp_port *mlxsw_sp_port, - bool ingress, - struct netlink_ext_ack *extack) -{ - struct mlxsw_sp_acl_block_binding *binding; - int err; - - if (WARN_ON(mlxsw_sp_acl_block_lookup(block, mlxsw_sp_port, ingress))) - return -EEXIST; - - if (ingress && block->ingress_blocker_rule_count) { - NL_SET_ERR_MSG_MOD(extack, "Block cannot be bound to ingress because it contains unsupported rules"); - return -EOPNOTSUPP; - } - - if (!ingress && block->egress_blocker_rule_count) { - NL_SET_ERR_MSG_MOD(extack, "Block cannot be bound to egress because it contains unsupported rules"); - return -EOPNOTSUPP; - } - - binding = kzalloc(sizeof(*binding), GFP_KERNEL); - if (!binding) - return -ENOMEM; - binding->mlxsw_sp_port = mlxsw_sp_port; - binding->ingress = ingress; - - if (mlxsw_sp_acl_ruleset_block_bound(block)) { - err = mlxsw_sp_acl_ruleset_bind(mlxsw_sp, block, binding); - if (err) - goto err_ruleset_bind; - } - - if (ingress) - block->ingress_binding_count++; - else - block->egress_binding_count++; - list_add(&binding->list, &block->binding_list); - return 0; - -err_ruleset_bind: - kfree(binding); - return err; -} - -int mlxsw_sp_acl_block_unbind(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, - struct mlxsw_sp_port *mlxsw_sp_port, - bool ingress) -{ - struct mlxsw_sp_acl_block_binding *binding; - - binding = mlxsw_sp_acl_block_lookup(block, mlxsw_sp_port, ingress); - if (!binding) - return -ENOENT; - - list_del(&binding->list); - - if (ingress) - block->ingress_binding_count--; - else - block->egress_binding_count--; - - if (mlxsw_sp_acl_ruleset_block_bound(block)) - mlxsw_sp_acl_ruleset_unbind(mlxsw_sp, block, binding); - - kfree(binding); - return 0; -} - static struct mlxsw_sp_acl_ruleset * mlxsw_sp_acl_ruleset_create(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, u32 chain_index, + struct mlxsw_sp_flow_block *block, u32 chain_index, const struct mlxsw_sp_acl_profile_ops *ops, struct mlxsw_afk_element_usage *tmplt_elusage) { @@ -340,7 +180,8 @@ mlxsw_sp_acl_ruleset_create(struct mlxsw_sp *mlxsw_sp, goto err_rhashtable_init; err = ops->ruleset_add(mlxsw_sp, &acl->tcam, ruleset->priv, - tmplt_elusage); + tmplt_elusage, &ruleset->min_prio, + &ruleset->max_prio); if (err) goto err_ops_ruleset_add; @@ -388,7 +229,7 @@ static void mlxsw_sp_acl_ruleset_ref_dec(struct mlxsw_sp *mlxsw_sp, static struct mlxsw_sp_acl_ruleset * __mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp_acl *acl, - struct mlxsw_sp_acl_block *block, u32 chain_index, + struct mlxsw_sp_flow_block *block, u32 chain_index, const struct mlxsw_sp_acl_profile_ops *ops) { struct mlxsw_sp_acl_ruleset_ht_key ht_key; @@ -403,7 +244,7 @@ __mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp_acl *acl, struct mlxsw_sp_acl_ruleset * mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, u32 chain_index, + struct mlxsw_sp_flow_block *block, u32 chain_index, enum mlxsw_sp_acl_profile profile) { const struct mlxsw_sp_acl_profile_ops *ops; @@ -421,7 +262,7 @@ mlxsw_sp_acl_ruleset_lookup(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_ruleset * mlxsw_sp_acl_ruleset_get(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, u32 chain_index, + struct mlxsw_sp_flow_block *block, u32 chain_index, enum mlxsw_sp_acl_profile profile, struct mlxsw_afk_element_usage *tmplt_elusage) { @@ -455,6 +296,14 @@ u16 mlxsw_sp_acl_ruleset_group_id(struct mlxsw_sp_acl_ruleset *ruleset) return ops->ruleset_group_id(ruleset->priv); } +void mlxsw_sp_acl_ruleset_prio_get(struct mlxsw_sp_acl_ruleset *ruleset, + unsigned int *p_min_prio, + unsigned int *p_max_prio) +{ + *p_min_prio = ruleset->min_prio; + *p_max_prio = ruleset->max_prio; +} + struct mlxsw_sp_acl_rule_info * mlxsw_sp_acl_rulei_create(struct mlxsw_sp_acl *acl, struct mlxsw_afa_block *afa_block) @@ -584,11 +433,11 @@ int mlxsw_sp_acl_rulei_act_fwd(struct mlxsw_sp *mlxsw_sp, int mlxsw_sp_acl_rulei_act_mirror(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_rule_info *rulei, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct net_device *out_dev, struct netlink_ext_ack *extack) { - struct mlxsw_sp_acl_block_binding *binding; + struct mlxsw_sp_flow_block_binding *binding; struct mlxsw_sp_port *in_port; if (!list_is_singular(&block->binding_list)) { @@ -596,7 +445,7 @@ int mlxsw_sp_acl_rulei_act_mirror(struct mlxsw_sp *mlxsw_sp, return -EOPNOTSUPP; } binding = list_first_entry(&block->binding_list, - struct mlxsw_sp_acl_block_binding, list); + struct mlxsw_sp_flow_block_binding, list); in_port = binding->mlxsw_sp_port; return mlxsw_afa_block_append_mirror(rulei->act_block, @@ -818,7 +667,7 @@ int mlxsw_sp_acl_rule_add(struct mlxsw_sp *mlxsw_sp, { struct mlxsw_sp_acl_ruleset *ruleset = rule->ruleset; const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops; - struct mlxsw_sp_acl_block *block = ruleset->ht_key.block; + struct mlxsw_sp_flow_block *block = ruleset->ht_key.block; int err; err = ops->rule_add(mlxsw_sp, ruleset->priv, rule->priv, rule->rulei); @@ -862,18 +711,17 @@ void mlxsw_sp_acl_rule_del(struct mlxsw_sp *mlxsw_sp, { struct mlxsw_sp_acl_ruleset *ruleset = rule->ruleset; const struct mlxsw_sp_acl_profile_ops *ops = ruleset->ht_key.ops; - struct mlxsw_sp_acl_block *block = ruleset->ht_key.block; + struct mlxsw_sp_flow_block *block = ruleset->ht_key.block; block->egress_blocker_rule_count -= rule->rulei->egress_bind_blocker; block->ingress_blocker_rule_count -= rule->rulei->ingress_bind_blocker; - ruleset->ht_key.block->rule_count--; + block->rule_count--; mutex_lock(&mlxsw_sp->acl->rules_lock); list_del(&rule->list); mutex_unlock(&mlxsw_sp->acl->rules_lock); if (!ruleset->ht_key.chain_index && mlxsw_sp_acl_ruleset_is_singular(ruleset)) - mlxsw_sp_acl_ruleset_block_unbind(mlxsw_sp, ruleset, - ruleset->ht_key.block); + mlxsw_sp_acl_ruleset_block_unbind(mlxsw_sp, ruleset, block); rhashtable_remove_fast(&ruleset->rule_ht, &rule->ht_node, mlxsw_sp_acl_rule_ht_params); ops->rule_del(mlxsw_sp, rule->priv); diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_actions.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_actions.c index e47d1d286e93..73d56012654b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_actions.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_flex_actions.c @@ -136,28 +136,35 @@ mlxsw_sp_act_mirror_add(void *priv, u8 local_in_port, const struct net_device *out_dev, bool ingress, int *p_span_id) { - struct mlxsw_sp_port *in_port; + struct mlxsw_sp_port *mlxsw_sp_port; struct mlxsw_sp *mlxsw_sp = priv; - enum mlxsw_sp_span_type type; + int err; - type = ingress ? MLXSW_SP_SPAN_INGRESS : MLXSW_SP_SPAN_EGRESS; - in_port = mlxsw_sp->ports[local_in_port]; + err = mlxsw_sp_span_agent_get(mlxsw_sp, out_dev, p_span_id); + if (err) + return err; + + mlxsw_sp_port = mlxsw_sp->ports[local_in_port]; + err = mlxsw_sp_span_analyzed_port_get(mlxsw_sp_port, ingress); + if (err) + goto err_analyzed_port_get; - return mlxsw_sp_span_mirror_add(in_port, out_dev, type, - false, p_span_id); + return 0; + +err_analyzed_port_get: + mlxsw_sp_span_agent_put(mlxsw_sp, *p_span_id); + return err; } static void mlxsw_sp_act_mirror_del(void *priv, u8 local_in_port, int span_id, bool ingress) { + struct mlxsw_sp_port *mlxsw_sp_port; struct mlxsw_sp *mlxsw_sp = priv; - struct mlxsw_sp_port *in_port; - enum mlxsw_sp_span_type type; - - type = ingress ? MLXSW_SP_SPAN_INGRESS : MLXSW_SP_SPAN_EGRESS; - in_port = mlxsw_sp->ports[local_in_port]; - mlxsw_sp_span_mirror_del(in_port, span_id, type, false); + mlxsw_sp_port = mlxsw_sp->ports[local_in_port]; + mlxsw_sp_span_analyzed_port_put(mlxsw_sp_port, ingress); + mlxsw_sp_span_agent_put(mlxsw_sp, span_id); } const struct mlxsw_afa_ops mlxsw_sp1_act_afa_ops = { diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c index a6e30e020b5c..5c020403342f 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c @@ -179,6 +179,8 @@ struct mlxsw_sp_acl_tcam_vgroup { bool tmplt_elusage_set; struct mlxsw_afk_element_usage tmplt_elusage; bool vregion_rehash_enabled; + unsigned int *p_min_prio; + unsigned int *p_max_prio; }; struct mlxsw_sp_acl_tcam_rehash_ctx { @@ -316,13 +318,17 @@ mlxsw_sp_acl_tcam_vgroup_add(struct mlxsw_sp *mlxsw_sp, const struct mlxsw_sp_acl_tcam_pattern *patterns, unsigned int patterns_count, struct mlxsw_afk_element_usage *tmplt_elusage, - bool vregion_rehash_enabled) + bool vregion_rehash_enabled, + unsigned int *p_min_prio, + unsigned int *p_max_prio) { int err; vgroup->patterns = patterns; vgroup->patterns_count = patterns_count; vgroup->vregion_rehash_enabled = vregion_rehash_enabled; + vgroup->p_min_prio = p_min_prio; + vgroup->p_max_prio = p_max_prio; if (tmplt_elusage) { vgroup->tmplt_elusage_set = true; @@ -416,6 +422,21 @@ mlxsw_sp_acl_tcam_vregion_max_prio(struct mlxsw_sp_acl_tcam_vregion *vregion) return vchunk->priority; } +static void +mlxsw_sp_acl_tcam_vgroup_prio_update(struct mlxsw_sp_acl_tcam_vgroup *vgroup) +{ + struct mlxsw_sp_acl_tcam_vregion *vregion; + + if (list_empty(&vgroup->vregion_list)) + return; + vregion = list_first_entry(&vgroup->vregion_list, + typeof(*vregion), list); + *vgroup->p_min_prio = mlxsw_sp_acl_tcam_vregion_prio(vregion); + vregion = list_last_entry(&vgroup->vregion_list, + typeof(*vregion), list); + *vgroup->p_max_prio = mlxsw_sp_acl_tcam_vregion_max_prio(vregion); +} + static int mlxsw_sp_acl_tcam_group_region_attach(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_tcam_group *group, @@ -1035,6 +1056,7 @@ mlxsw_sp_acl_tcam_vchunk_create(struct mlxsw_sp *mlxsw_sp, } list_add_tail(&vchunk->list, pos); mutex_unlock(&vregion->lock); + mlxsw_sp_acl_tcam_vgroup_prio_update(vgroup); return vchunk; @@ -1066,6 +1088,7 @@ mlxsw_sp_acl_tcam_vchunk_destroy(struct mlxsw_sp *mlxsw_sp, mlxsw_sp_acl_tcam_vchunk_ht_params); mlxsw_sp_acl_tcam_vregion_put(mlxsw_sp, vchunk->vregion); kfree(vchunk); + mlxsw_sp_acl_tcam_vgroup_prio_update(vgroup); } static struct mlxsw_sp_acl_tcam_vchunk * @@ -1582,14 +1605,17 @@ static int mlxsw_sp_acl_tcam_flower_ruleset_add(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_tcam *tcam, void *ruleset_priv, - struct mlxsw_afk_element_usage *tmplt_elusage) + struct mlxsw_afk_element_usage *tmplt_elusage, + unsigned int *p_min_prio, + unsigned int *p_max_prio) { struct mlxsw_sp_acl_tcam_flower_ruleset *ruleset = ruleset_priv; return mlxsw_sp_acl_tcam_vgroup_add(mlxsw_sp, tcam, &ruleset->vgroup, mlxsw_sp_acl_tcam_patterns, MLXSW_SP_ACL_TCAM_PATTERNS_COUNT, - tmplt_elusage, true); + tmplt_elusage, true, + p_min_prio, p_max_prio); } static void @@ -1698,7 +1724,9 @@ static int mlxsw_sp_acl_tcam_mr_ruleset_add(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_tcam *tcam, void *ruleset_priv, - struct mlxsw_afk_element_usage *tmplt_elusage) + struct mlxsw_afk_element_usage *tmplt_elusage, + unsigned int *p_min_prio, + unsigned int *p_max_prio) { struct mlxsw_sp_acl_tcam_mr_ruleset *ruleset = ruleset_priv; int err; @@ -1706,7 +1734,8 @@ mlxsw_sp_acl_tcam_mr_ruleset_add(struct mlxsw_sp *mlxsw_sp, err = mlxsw_sp_acl_tcam_vgroup_add(mlxsw_sp, tcam, &ruleset->vgroup, mlxsw_sp_acl_tcam_patterns, MLXSW_SP_ACL_TCAM_PATTERNS_COUNT, - tmplt_elusage, false); + tmplt_elusage, false, + p_min_prio, p_max_prio); if (err) return err; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h index 96437992b102..a41df10ade9b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h @@ -42,7 +42,8 @@ struct mlxsw_sp_acl_profile_ops { size_t ruleset_priv_size; int (*ruleset_add)(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_tcam *tcam, void *ruleset_priv, - struct mlxsw_afk_element_usage *tmplt_elusage); + struct mlxsw_afk_element_usage *tmplt_elusage, + unsigned int *p_min_prio, unsigned int *p_max_prio); void (*ruleset_del)(struct mlxsw_sp *mlxsw_sp, void *ruleset_priv); int (*ruleset_bind)(struct mlxsw_sp *mlxsw_sp, void *ruleset_priv, struct mlxsw_sp_port *mlxsw_sp_port, diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c index 968f0902e4fe..21bfb2f6a6f0 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_buffers.c @@ -614,7 +614,7 @@ static const struct mlxsw_sp_sb_cm mlxsw_sp2_sb_cms_egress[] = { #define MLXSW_SP_CPU_PORT_SB_CM MLXSW_SP_SB_CM(0, 0, MLXSW_SP_SB_POOL_EGR_CPU) static const struct mlxsw_sp_sb_cm mlxsw_sp_cpu_port_sb_cms[] = { - MLXSW_SP_CPU_PORT_SB_CM, + MLXSW_SP_SB_CM(1000, 8, MLXSW_SP_SB_POOL_EGR_CPU), MLXSW_SP_SB_CM(1000, 8, MLXSW_SP_SB_POOL_EGR_CPU), MLXSW_SP_SB_CM(1000, 8, MLXSW_SP_SB_POOL_EGR_CPU), MLXSW_SP_SB_CM(1000, 8, MLXSW_SP_SB_POOL_EGR_CPU), diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_flow.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_flow.c new file mode 100644 index 000000000000..47b66f347ff1 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_flow.c @@ -0,0 +1,305 @@ +// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 +/* Copyright (c) 2017-2020 Mellanox Technologies. All rights reserved */ + +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/errno.h> +#include <linux/list.h> +#include <net/net_namespace.h> + +#include "spectrum.h" + +struct mlxsw_sp_flow_block * +mlxsw_sp_flow_block_create(struct mlxsw_sp *mlxsw_sp, struct net *net) +{ + struct mlxsw_sp_flow_block *block; + + block = kzalloc(sizeof(*block), GFP_KERNEL); + if (!block) + return NULL; + INIT_LIST_HEAD(&block->binding_list); + INIT_LIST_HEAD(&block->mall.list); + block->mlxsw_sp = mlxsw_sp; + block->net = net; + return block; +} + +void mlxsw_sp_flow_block_destroy(struct mlxsw_sp_flow_block *block) +{ + WARN_ON(!list_empty(&block->binding_list)); + kfree(block); +} + +static struct mlxsw_sp_flow_block_binding * +mlxsw_sp_flow_block_lookup(struct mlxsw_sp_flow_block *block, + struct mlxsw_sp_port *mlxsw_sp_port, bool ingress) +{ + struct mlxsw_sp_flow_block_binding *binding; + + list_for_each_entry(binding, &block->binding_list, list) + if (binding->mlxsw_sp_port == mlxsw_sp_port && + binding->ingress == ingress) + return binding; + return NULL; +} + +static bool +mlxsw_sp_flow_block_ruleset_bound(const struct mlxsw_sp_flow_block *block) +{ + return block->ruleset_zero; +} + +static int mlxsw_sp_flow_block_bind(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_flow_block *block, + struct mlxsw_sp_port *mlxsw_sp_port, + bool ingress, + struct netlink_ext_ack *extack) +{ + struct mlxsw_sp_flow_block_binding *binding; + int err; + + if (WARN_ON(mlxsw_sp_flow_block_lookup(block, mlxsw_sp_port, ingress))) + return -EEXIST; + + if (ingress && block->ingress_blocker_rule_count) { + NL_SET_ERR_MSG_MOD(extack, "Block cannot be bound to ingress because it contains unsupported rules"); + return -EOPNOTSUPP; + } + + if (!ingress && block->egress_blocker_rule_count) { + NL_SET_ERR_MSG_MOD(extack, "Block cannot be bound to egress because it contains unsupported rules"); + return -EOPNOTSUPP; + } + + err = mlxsw_sp_mall_port_bind(block, mlxsw_sp_port); + if (err) + return err; + + binding = kzalloc(sizeof(*binding), GFP_KERNEL); + if (!binding) { + err = -ENOMEM; + goto err_binding_alloc; + } + binding->mlxsw_sp_port = mlxsw_sp_port; + binding->ingress = ingress; + + if (mlxsw_sp_flow_block_ruleset_bound(block)) { + err = mlxsw_sp_acl_ruleset_bind(mlxsw_sp, block, binding); + if (err) + goto err_ruleset_bind; + } + + if (ingress) + block->ingress_binding_count++; + else + block->egress_binding_count++; + list_add(&binding->list, &block->binding_list); + return 0; + +err_ruleset_bind: + kfree(binding); +err_binding_alloc: + mlxsw_sp_mall_port_unbind(block, mlxsw_sp_port); + + return err; +} + +static int mlxsw_sp_flow_block_unbind(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_flow_block *block, + struct mlxsw_sp_port *mlxsw_sp_port, + bool ingress) +{ + struct mlxsw_sp_flow_block_binding *binding; + + binding = mlxsw_sp_flow_block_lookup(block, mlxsw_sp_port, ingress); + if (!binding) + return -ENOENT; + + list_del(&binding->list); + + if (ingress) + block->ingress_binding_count--; + else + block->egress_binding_count--; + + if (mlxsw_sp_flow_block_ruleset_bound(block)) + mlxsw_sp_acl_ruleset_unbind(mlxsw_sp, block, binding); + + kfree(binding); + + mlxsw_sp_mall_port_unbind(block, mlxsw_sp_port); + + return 0; +} + +static int mlxsw_sp_flow_block_mall_cb(struct mlxsw_sp_flow_block *flow_block, + struct tc_cls_matchall_offload *f) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_flow_block_mlxsw_sp(flow_block); + + switch (f->command) { + case TC_CLSMATCHALL_REPLACE: + return mlxsw_sp_mall_replace(mlxsw_sp, flow_block, f); + case TC_CLSMATCHALL_DESTROY: + mlxsw_sp_mall_destroy(flow_block, f); + return 0; + default: + return -EOPNOTSUPP; + } +} + +static int mlxsw_sp_flow_block_flower_cb(struct mlxsw_sp_flow_block *flow_block, + struct flow_cls_offload *f) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_flow_block_mlxsw_sp(flow_block); + + switch (f->command) { + case FLOW_CLS_REPLACE: + return mlxsw_sp_flower_replace(mlxsw_sp, flow_block, f); + case FLOW_CLS_DESTROY: + mlxsw_sp_flower_destroy(mlxsw_sp, flow_block, f); + return 0; + case FLOW_CLS_STATS: + return mlxsw_sp_flower_stats(mlxsw_sp, flow_block, f); + case FLOW_CLS_TMPLT_CREATE: + return mlxsw_sp_flower_tmplt_create(mlxsw_sp, flow_block, f); + case FLOW_CLS_TMPLT_DESTROY: + mlxsw_sp_flower_tmplt_destroy(mlxsw_sp, flow_block, f); + return 0; + default: + return -EOPNOTSUPP; + } +} + +static int mlxsw_sp_flow_block_cb(enum tc_setup_type type, + void *type_data, void *cb_priv) +{ + struct mlxsw_sp_flow_block *flow_block = cb_priv; + + if (mlxsw_sp_flow_block_disabled(flow_block)) + return -EOPNOTSUPP; + + switch (type) { + case TC_SETUP_CLSMATCHALL: + return mlxsw_sp_flow_block_mall_cb(flow_block, type_data); + case TC_SETUP_CLSFLOWER: + return mlxsw_sp_flow_block_flower_cb(flow_block, type_data); + default: + return -EOPNOTSUPP; + } +} + +static void mlxsw_sp_tc_block_release(void *cb_priv) +{ + struct mlxsw_sp_flow_block *flow_block = cb_priv; + + mlxsw_sp_flow_block_destroy(flow_block); +} + +static LIST_HEAD(mlxsw_sp_block_cb_list); + +static int mlxsw_sp_setup_tc_block_bind(struct mlxsw_sp_port *mlxsw_sp_port, + struct flow_block_offload *f, + bool ingress) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_flow_block *flow_block; + struct flow_block_cb *block_cb; + bool register_block = false; + int err; + + block_cb = flow_block_cb_lookup(f->block, mlxsw_sp_flow_block_cb, + mlxsw_sp); + if (!block_cb) { + flow_block = mlxsw_sp_flow_block_create(mlxsw_sp, f->net); + if (!flow_block) + return -ENOMEM; + block_cb = flow_block_cb_alloc(mlxsw_sp_flow_block_cb, + mlxsw_sp, flow_block, + mlxsw_sp_tc_block_release); + if (IS_ERR(block_cb)) { + mlxsw_sp_flow_block_destroy(flow_block); + err = PTR_ERR(block_cb); + goto err_cb_register; + } + register_block = true; + } else { + flow_block = flow_block_cb_priv(block_cb); + } + flow_block_cb_incref(block_cb); + err = mlxsw_sp_flow_block_bind(mlxsw_sp, flow_block, + mlxsw_sp_port, ingress, f->extack); + if (err) + goto err_block_bind; + + if (ingress) + mlxsw_sp_port->ing_flow_block = flow_block; + else + mlxsw_sp_port->eg_flow_block = flow_block; + + if (register_block) { + flow_block_cb_add(block_cb, f); + list_add_tail(&block_cb->driver_list, &mlxsw_sp_block_cb_list); + } + + return 0; + +err_block_bind: + if (!flow_block_cb_decref(block_cb)) + flow_block_cb_free(block_cb); +err_cb_register: + return err; +} + +static void mlxsw_sp_setup_tc_block_unbind(struct mlxsw_sp_port *mlxsw_sp_port, + struct flow_block_offload *f, + bool ingress) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_flow_block *flow_block; + struct flow_block_cb *block_cb; + int err; + + block_cb = flow_block_cb_lookup(f->block, mlxsw_sp_flow_block_cb, + mlxsw_sp); + if (!block_cb) + return; + + if (ingress) + mlxsw_sp_port->ing_flow_block = NULL; + else + mlxsw_sp_port->eg_flow_block = NULL; + + flow_block = flow_block_cb_priv(block_cb); + err = mlxsw_sp_flow_block_unbind(mlxsw_sp, flow_block, + mlxsw_sp_port, ingress); + if (!err && !flow_block_cb_decref(block_cb)) { + flow_block_cb_remove(block_cb, f); + list_del(&block_cb->driver_list); + } +} + +int mlxsw_sp_setup_tc_block(struct mlxsw_sp_port *mlxsw_sp_port, + struct flow_block_offload *f) +{ + bool ingress; + + if (f->binder_type == FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS) + ingress = true; + else if (f->binder_type == FLOW_BLOCK_BINDER_TYPE_CLSACT_EGRESS) + ingress = false; + else + return -EOPNOTSUPP; + + f->driver_block_list = &mlxsw_sp_block_cb_list; + + switch (f->command) { + case FLOW_BLOCK_BIND: + return mlxsw_sp_setup_tc_block_bind(mlxsw_sp_port, f, ingress); + case FLOW_BLOCK_UNBIND: + mlxsw_sp_setup_tc_block_unbind(mlxsw_sp_port, f, ingress); + return 0; + default: + return -EOPNOTSUPP; + } +} diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c index 890b078851c9..51e1b3930c56 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_flower.c @@ -15,7 +15,7 @@ #include "core_acl_flex_keys.h" static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct mlxsw_sp_acl_rule_info *rulei, struct flow_action *flow_action, struct netlink_ext_ack *extack) @@ -30,14 +30,14 @@ static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp, return -EOPNOTSUPP; act = flow_action_first_entry_get(flow_action); - if (act->hw_stats == FLOW_ACTION_HW_STATS_ANY || - act->hw_stats == FLOW_ACTION_HW_STATS_IMMEDIATE) { + if (act->hw_stats & FLOW_ACTION_HW_STATS_DISABLED) { + /* Nothing to do */ + } else if (act->hw_stats & FLOW_ACTION_HW_STATS_IMMEDIATE) { /* Count action is inserted first */ err = mlxsw_sp_acl_rulei_act_count(mlxsw_sp, rulei, extack); if (err) return err; - } else if (act->hw_stats != FLOW_ACTION_HW_STATS_DISABLED && - act->hw_stats != FLOW_ACTION_HW_STATS_DONT_CARE) { + } else { NL_SET_ERR_MSG_MOD(extack, "Unsupported action HW stats type"); return -EOPNOTSUPP; } @@ -54,11 +54,11 @@ static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp, case FLOW_ACTION_DROP: { bool ingress; - if (mlxsw_sp_acl_block_is_mixed_bound(block)) { + if (mlxsw_sp_flow_block_is_mixed_bound(block)) { NL_SET_ERR_MSG_MOD(extack, "Drop action is not supported when block is bound to ingress and egress"); return -EOPNOTSUPP; } - ingress = mlxsw_sp_acl_block_is_ingress_bound(block); + ingress = mlxsw_sp_flow_block_is_ingress_bound(block); err = mlxsw_sp_acl_rulei_act_drop(rulei, ingress, act->cookie, extack); if (err) { @@ -107,7 +107,7 @@ static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_fid *fid; u16 fid_index; - if (mlxsw_sp_acl_block_is_egress_bound(block)) { + if (mlxsw_sp_flow_block_is_egress_bound(block)) { NL_SET_ERR_MSG_MOD(extack, "Redirect action is not supported on egress"); return -EOPNOTSUPP; } @@ -191,7 +191,7 @@ static int mlxsw_sp_flower_parse_actions(struct mlxsw_sp *mlxsw_sp, static int mlxsw_sp_flower_parse_meta(struct mlxsw_sp_acl_rule_info *rulei, struct flow_cls_offload *f, - struct mlxsw_sp_acl_block *block) + struct mlxsw_sp_flow_block *block) { struct flow_rule *rule = flow_cls_offload_flow_rule(f); struct mlxsw_sp_port *mlxsw_sp_port; @@ -372,7 +372,7 @@ static int mlxsw_sp_flower_parse_ip(struct mlxsw_sp *mlxsw_sp, } static int mlxsw_sp_flower_parse(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct mlxsw_sp_acl_rule_info *rulei, struct flow_cls_offload *f) { @@ -461,7 +461,7 @@ static int mlxsw_sp_flower_parse(struct mlxsw_sp *mlxsw_sp, struct flow_match_vlan match; flow_rule_match_vlan(rule, &match); - if (mlxsw_sp_acl_block_is_egress_bound(block)) { + if (mlxsw_sp_flow_block_is_egress_bound(block)) { NL_SET_ERR_MSG_MOD(f->common.extack, "vlan_id key is not supported on egress"); return -EOPNOTSUPP; } @@ -505,8 +505,36 @@ static int mlxsw_sp_flower_parse(struct mlxsw_sp *mlxsw_sp, f->common.extack); } +static int mlxsw_sp_flower_mall_prio_check(struct mlxsw_sp_flow_block *block, + struct flow_cls_offload *f) +{ + bool ingress = mlxsw_sp_flow_block_is_ingress_bound(block); + unsigned int mall_min_prio; + unsigned int mall_max_prio; + int err; + + err = mlxsw_sp_mall_prio_get(block, f->common.chain_index, + &mall_min_prio, &mall_max_prio); + if (err) { + if (err == -ENOENT) + /* No matchall filters installed on this chain. */ + return 0; + NL_SET_ERR_MSG(f->common.extack, "Failed to get matchall priorities"); + return err; + } + if (ingress && f->common.prio <= mall_min_prio) { + NL_SET_ERR_MSG(f->common.extack, "Failed to add in front of existing matchall rules"); + return -EOPNOTSUPP; + } + if (!ingress && f->common.prio >= mall_max_prio) { + NL_SET_ERR_MSG(f->common.extack, "Failed to add behind of existing matchall rules"); + return -EOPNOTSUPP; + } + return 0; +} + int mlxsw_sp_flower_replace(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct flow_cls_offload *f) { struct mlxsw_sp_acl_rule_info *rulei; @@ -514,6 +542,10 @@ int mlxsw_sp_flower_replace(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_acl_rule *rule; int err; + err = mlxsw_sp_flower_mall_prio_check(block, f); + if (err) + return err; + ruleset = mlxsw_sp_acl_ruleset_get(mlxsw_sp, block, f->common.chain_index, MLXSW_SP_ACL_PROFILE_FLOWER, NULL); @@ -553,7 +585,7 @@ err_rule_create: } void mlxsw_sp_flower_destroy(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct flow_cls_offload *f) { struct mlxsw_sp_acl_ruleset *ruleset; @@ -575,7 +607,7 @@ void mlxsw_sp_flower_destroy(struct mlxsw_sp *mlxsw_sp, } int mlxsw_sp_flower_stats(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct flow_cls_offload *f) { enum flow_action_hw_stats used_hw_stats = FLOW_ACTION_HW_STATS_DISABLED; @@ -612,7 +644,7 @@ err_rule_get_stats: } int mlxsw_sp_flower_tmplt_create(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct flow_cls_offload *f) { struct mlxsw_sp_acl_ruleset *ruleset; @@ -633,7 +665,7 @@ int mlxsw_sp_flower_tmplt_create(struct mlxsw_sp *mlxsw_sp, } void mlxsw_sp_flower_tmplt_destroy(struct mlxsw_sp *mlxsw_sp, - struct mlxsw_sp_acl_block *block, + struct mlxsw_sp_flow_block *block, struct flow_cls_offload *f) { struct mlxsw_sp_acl_ruleset *ruleset; @@ -647,3 +679,23 @@ void mlxsw_sp_flower_tmplt_destroy(struct mlxsw_sp *mlxsw_sp, mlxsw_sp_acl_ruleset_put(mlxsw_sp, ruleset); mlxsw_sp_acl_ruleset_put(mlxsw_sp, ruleset); } + +int mlxsw_sp_flower_prio_get(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_flow_block *block, + u32 chain_index, unsigned int *p_min_prio, + unsigned int *p_max_prio) +{ + struct mlxsw_sp_acl_ruleset *ruleset; + + ruleset = mlxsw_sp_acl_ruleset_lookup(mlxsw_sp, block, + chain_index, + MLXSW_SP_ACL_PROFILE_FLOWER); + if (IS_ERR(ruleset)) + /* In case there are no flower rules, the caller + * receives -ENOENT to indicate there is no need + * to check the priorities. + */ + return PTR_ERR(ruleset); + mlxsw_sp_acl_ruleset_prio_get(ruleset, p_min_prio, p_max_prio); + return 0; +} diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_matchall.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_matchall.c new file mode 100644 index 000000000000..f1a44a8eda55 --- /dev/null +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_matchall.c @@ -0,0 +1,378 @@ +// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0 +/* Copyright (c) 2017-2020 Mellanox Technologies. All rights reserved */ + +#include <linux/kernel.h> +#include <linux/errno.h> +#include <linux/netdevice.h> +#include <net/flow_offload.h> + +#include "spectrum.h" +#include "spectrum_span.h" +#include "reg.h" + +enum mlxsw_sp_mall_action_type { + MLXSW_SP_MALL_ACTION_TYPE_MIRROR, + MLXSW_SP_MALL_ACTION_TYPE_SAMPLE, +}; + +struct mlxsw_sp_mall_mirror_entry { + const struct net_device *to_dev; + int span_id; +}; + +struct mlxsw_sp_mall_entry { + struct list_head list; + unsigned long cookie; + unsigned int priority; + enum mlxsw_sp_mall_action_type type; + bool ingress; + union { + struct mlxsw_sp_mall_mirror_entry mirror; + struct mlxsw_sp_port_sample sample; + }; + struct rcu_head rcu; +}; + +static struct mlxsw_sp_mall_entry * +mlxsw_sp_mall_entry_find(struct mlxsw_sp_flow_block *block, unsigned long cookie) +{ + struct mlxsw_sp_mall_entry *mall_entry; + + list_for_each_entry(mall_entry, &block->mall.list, list) + if (mall_entry->cookie == cookie) + return mall_entry; + + return NULL; +} + +static int +mlxsw_sp_mall_port_mirror_add(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_mall_entry *mall_entry) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_span_trigger_parms parms; + enum mlxsw_sp_span_trigger trigger; + int err; + + if (!mall_entry->mirror.to_dev) { + netdev_err(mlxsw_sp_port->dev, "Could not find requested device\n"); + return -EINVAL; + } + + err = mlxsw_sp_span_agent_get(mlxsw_sp, mall_entry->mirror.to_dev, + &mall_entry->mirror.span_id); + if (err) + return err; + + err = mlxsw_sp_span_analyzed_port_get(mlxsw_sp_port, + mall_entry->ingress); + if (err) + goto err_analyzed_port_get; + + trigger = mall_entry->ingress ? MLXSW_SP_SPAN_TRIGGER_INGRESS : + MLXSW_SP_SPAN_TRIGGER_EGRESS; + parms.span_id = mall_entry->mirror.span_id; + err = mlxsw_sp_span_agent_bind(mlxsw_sp, trigger, mlxsw_sp_port, + &parms); + if (err) + goto err_agent_bind; + + return 0; + +err_agent_bind: + mlxsw_sp_span_analyzed_port_put(mlxsw_sp_port, mall_entry->ingress); +err_analyzed_port_get: + mlxsw_sp_span_agent_put(mlxsw_sp, mall_entry->mirror.span_id); + return err; +} + +static void +mlxsw_sp_mall_port_mirror_del(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_mall_entry *mall_entry) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_span_trigger_parms parms; + enum mlxsw_sp_span_trigger trigger; + + trigger = mall_entry->ingress ? MLXSW_SP_SPAN_TRIGGER_INGRESS : + MLXSW_SP_SPAN_TRIGGER_EGRESS; + parms.span_id = mall_entry->mirror.span_id; + mlxsw_sp_span_agent_unbind(mlxsw_sp, trigger, mlxsw_sp_port, &parms); + mlxsw_sp_span_analyzed_port_put(mlxsw_sp_port, mall_entry->ingress); + mlxsw_sp_span_agent_put(mlxsw_sp, mall_entry->mirror.span_id); +} + +static int mlxsw_sp_mall_port_sample_set(struct mlxsw_sp_port *mlxsw_sp_port, + bool enable, u32 rate) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + char mpsc_pl[MLXSW_REG_MPSC_LEN]; + + mlxsw_reg_mpsc_pack(mpsc_pl, mlxsw_sp_port->local_port, enable, rate); + return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mpsc), mpsc_pl); +} + +static int +mlxsw_sp_mall_port_sample_add(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_mall_entry *mall_entry) +{ + int err; + + if (rtnl_dereference(mlxsw_sp_port->sample)) { + netdev_err(mlxsw_sp_port->dev, "sample already active\n"); + return -EEXIST; + } + rcu_assign_pointer(mlxsw_sp_port->sample, &mall_entry->sample); + + err = mlxsw_sp_mall_port_sample_set(mlxsw_sp_port, true, + mall_entry->sample.rate); + if (err) + goto err_port_sample_set; + return 0; + +err_port_sample_set: + RCU_INIT_POINTER(mlxsw_sp_port->sample, NULL); + return err; +} + +static void +mlxsw_sp_mall_port_sample_del(struct mlxsw_sp_port *mlxsw_sp_port) +{ + if (!mlxsw_sp_port->sample) + return; + + mlxsw_sp_mall_port_sample_set(mlxsw_sp_port, false, 1); + RCU_INIT_POINTER(mlxsw_sp_port->sample, NULL); +} + +static int +mlxsw_sp_mall_port_rule_add(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_mall_entry *mall_entry) +{ + switch (mall_entry->type) { + case MLXSW_SP_MALL_ACTION_TYPE_MIRROR: + return mlxsw_sp_mall_port_mirror_add(mlxsw_sp_port, mall_entry); + case MLXSW_SP_MALL_ACTION_TYPE_SAMPLE: + return mlxsw_sp_mall_port_sample_add(mlxsw_sp_port, mall_entry); + default: + WARN_ON(1); + return -EINVAL; + } +} + +static void +mlxsw_sp_mall_port_rule_del(struct mlxsw_sp_port *mlxsw_sp_port, + struct mlxsw_sp_mall_entry *mall_entry) +{ + switch (mall_entry->type) { + case MLXSW_SP_MALL_ACTION_TYPE_MIRROR: + mlxsw_sp_mall_port_mirror_del(mlxsw_sp_port, mall_entry); + break; + case MLXSW_SP_MALL_ACTION_TYPE_SAMPLE: + mlxsw_sp_mall_port_sample_del(mlxsw_sp_port); + break; + default: + WARN_ON(1); + } +} + +static void mlxsw_sp_mall_prio_update(struct mlxsw_sp_flow_block *block) +{ + struct mlxsw_sp_mall_entry *mall_entry; + + if (list_empty(&block->mall.list)) + return; + block->mall.min_prio = UINT_MAX; + block->mall.max_prio = 0; + list_for_each_entry(mall_entry, &block->mall.list, list) { + if (mall_entry->priority < block->mall.min_prio) + block->mall.min_prio = mall_entry->priority; + if (mall_entry->priority > block->mall.max_prio) + block->mall.max_prio = mall_entry->priority; + } +} + +int mlxsw_sp_mall_replace(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_flow_block *block, + struct tc_cls_matchall_offload *f) +{ + struct mlxsw_sp_flow_block_binding *binding; + struct mlxsw_sp_mall_entry *mall_entry; + __be16 protocol = f->common.protocol; + struct flow_action_entry *act; + unsigned int flower_min_prio; + unsigned int flower_max_prio; + bool flower_prio_valid; + int err; + + if (!flow_offload_has_one_action(&f->rule->action)) { + NL_SET_ERR_MSG(f->common.extack, "Only singular actions are supported"); + return -EOPNOTSUPP; + } + + if (f->common.chain_index) { + NL_SET_ERR_MSG(f->common.extack, "Only chain 0 is supported"); + return -EOPNOTSUPP; + } + + if (mlxsw_sp_flow_block_is_mixed_bound(block)) { + NL_SET_ERR_MSG(f->common.extack, "Only not mixed bound blocks are supported"); + return -EOPNOTSUPP; + } + + err = mlxsw_sp_flower_prio_get(mlxsw_sp, block, f->common.chain_index, + &flower_min_prio, &flower_max_prio); + if (err) { + if (err != -ENOENT) { + NL_SET_ERR_MSG(f->common.extack, "Failed to get flower priorities"); + return err; + } + flower_prio_valid = false; + /* No flower filters are installed in specified chain. */ + } else { + flower_prio_valid = true; + } + + mall_entry = kzalloc(sizeof(*mall_entry), GFP_KERNEL); + if (!mall_entry) + return -ENOMEM; + mall_entry->cookie = f->cookie; + mall_entry->priority = f->common.prio; + mall_entry->ingress = mlxsw_sp_flow_block_is_ingress_bound(block); + + act = &f->rule->action.entries[0]; + + if (act->id == FLOW_ACTION_MIRRED && protocol == htons(ETH_P_ALL)) { + if (flower_prio_valid && mall_entry->ingress && + mall_entry->priority >= flower_min_prio) { + NL_SET_ERR_MSG(f->common.extack, "Failed to add behind existing flower rules"); + err = -EOPNOTSUPP; + goto errout; + } + if (flower_prio_valid && !mall_entry->ingress && + mall_entry->priority <= flower_max_prio) { + NL_SET_ERR_MSG(f->common.extack, "Failed to add in front of existing flower rules"); + err = -EOPNOTSUPP; + goto errout; + } + mall_entry->type = MLXSW_SP_MALL_ACTION_TYPE_MIRROR; + mall_entry->mirror.to_dev = act->dev; + } else if (act->id == FLOW_ACTION_SAMPLE && + protocol == htons(ETH_P_ALL)) { + if (!mall_entry->ingress) { + NL_SET_ERR_MSG(f->common.extack, "Sample is not supported on egress"); + err = -EOPNOTSUPP; + goto errout; + } + if (flower_prio_valid && + mall_entry->priority >= flower_min_prio) { + NL_SET_ERR_MSG(f->common.extack, "Failed to add behind existing flower rules"); + err = -EOPNOTSUPP; + goto errout; + } + if (act->sample.rate > MLXSW_REG_MPSC_RATE_MAX) { + NL_SET_ERR_MSG(f->common.extack, "Sample rate not supported"); + err = -EOPNOTSUPP; + goto errout; + } + mall_entry->type = MLXSW_SP_MALL_ACTION_TYPE_SAMPLE; + mall_entry->sample.psample_group = act->sample.psample_group; + mall_entry->sample.truncate = act->sample.truncate; + mall_entry->sample.trunc_size = act->sample.trunc_size; + mall_entry->sample.rate = act->sample.rate; + } else { + err = -EOPNOTSUPP; + goto errout; + } + + list_for_each_entry(binding, &block->binding_list, list) { + err = mlxsw_sp_mall_port_rule_add(binding->mlxsw_sp_port, + mall_entry); + if (err) + goto rollback; + } + + block->rule_count++; + if (mall_entry->ingress) + block->egress_blocker_rule_count++; + else + block->ingress_blocker_rule_count++; + list_add_tail(&mall_entry->list, &block->mall.list); + mlxsw_sp_mall_prio_update(block); + return 0; + +rollback: + list_for_each_entry_continue_reverse(binding, &block->binding_list, + list) + mlxsw_sp_mall_port_rule_del(binding->mlxsw_sp_port, mall_entry); +errout: + kfree(mall_entry); + return err; +} + +void mlxsw_sp_mall_destroy(struct mlxsw_sp_flow_block *block, + struct tc_cls_matchall_offload *f) +{ + struct mlxsw_sp_flow_block_binding *binding; + struct mlxsw_sp_mall_entry *mall_entry; + + mall_entry = mlxsw_sp_mall_entry_find(block, f->cookie); + if (!mall_entry) { + NL_SET_ERR_MSG(f->common.extack, "Entry not found"); + return; + } + + list_del(&mall_entry->list); + if (mall_entry->ingress) + block->egress_blocker_rule_count--; + else + block->ingress_blocker_rule_count--; + block->rule_count--; + list_for_each_entry(binding, &block->binding_list, list) + mlxsw_sp_mall_port_rule_del(binding->mlxsw_sp_port, mall_entry); + kfree_rcu(mall_entry, rcu); /* sample RX packets may be in-flight */ + mlxsw_sp_mall_prio_update(block); +} + +int mlxsw_sp_mall_port_bind(struct mlxsw_sp_flow_block *block, + struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct mlxsw_sp_mall_entry *mall_entry; + int err; + + list_for_each_entry(mall_entry, &block->mall.list, list) { + err = mlxsw_sp_mall_port_rule_add(mlxsw_sp_port, mall_entry); + if (err) + goto rollback; + } + return 0; + +rollback: + list_for_each_entry_continue_reverse(mall_entry, &block->mall.list, + list) + mlxsw_sp_mall_port_rule_del(mlxsw_sp_port, mall_entry); + return err; +} + +void mlxsw_sp_mall_port_unbind(struct mlxsw_sp_flow_block *block, + struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct mlxsw_sp_mall_entry *mall_entry; + + list_for_each_entry(mall_entry, &block->mall.list, list) + mlxsw_sp_mall_port_rule_del(mlxsw_sp_port, mall_entry); +} + +int mlxsw_sp_mall_prio_get(struct mlxsw_sp_flow_block *block, u32 chain_index, + unsigned int *p_min_prio, unsigned int *p_max_prio) +{ + if (chain_index || list_empty(&block->mall.list)) + /* In case there are no matchall rules, the caller + * receives -ENOENT to indicate there is no need + * to check the priorities. + */ + return -ENOENT; + *p_min_prio = block->mall.min_prio; + *p_max_prio = block->mall.max_prio; + return 0; +} diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c index d5bca1be3ef5..770de0222e7b 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c @@ -2999,6 +2999,7 @@ static u32 mlxsw_sp_nexthop_group_hash_obj(const void *data, u32 len, u32 seed) for (i = 0; i < nh_grp->count; i++) { nh = &nh_grp->nexthops[i]; val ^= jhash(&nh->ifindex, sizeof(nh->ifindex), seed); + val ^= jhash(&nh->gw_addr, sizeof(nh->gw_addr), seed); } return jhash(&val, sizeof(val), seed); default: @@ -3012,11 +3013,14 @@ mlxsw_sp_nexthop6_group_hash(struct mlxsw_sp_fib6_entry *fib6_entry, u32 seed) { unsigned int val = fib6_entry->nrt6; struct mlxsw_sp_rt6 *mlxsw_sp_rt6; - struct net_device *dev; list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) { - dev = mlxsw_sp_rt6->rt->fib6_nh->fib_nh_dev; + struct fib6_nh *fib6_nh = mlxsw_sp_rt6->rt->fib6_nh; + struct net_device *dev = fib6_nh->fib_nh_dev; + struct in6_addr *gw = &fib6_nh->fib_nh_gw6; + val ^= jhash(&dev->ifindex, sizeof(dev->ifindex), seed); + val ^= jhash(gw, sizeof(*gw), seed); } return jhash(&val, sizeof(val), seed); @@ -4999,9 +5003,11 @@ static bool mlxsw_sp_fib6_rt_should_ignore(const struct fib6_info *rt) { /* Packets with link-local destination IP arriving to the router * are trapped to the CPU, so no need to program specific routes - * for them. + * for them. Only allow prefix routes (usually one fe80::/64) so + * that packets are trapped for the right reason. */ - if (ipv6_addr_type(&rt->fib6_dst.addr) & IPV6_ADDR_LINKLOCAL) + if ((ipv6_addr_type(&rt->fib6_dst.addr) & IPV6_ADDR_LINKLOCAL) && + (rt->fib6_flags & (RTF_LOCAL | RTF_ANYCAST))) return true; /* Multicast routes aren't supported, so ignore them. Neighbour @@ -7568,7 +7574,7 @@ static struct mlxsw_sp_fid * mlxsw_sp_rif_vlan_fid_get(struct mlxsw_sp_rif *rif, struct netlink_ext_ack *extack) { - struct net_device *br_dev = rif->dev; + struct net_device *br_dev; u16 vid; int err; diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c index 9fb2e9d93929..304eb8c3d8bd 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.c @@ -3,6 +3,8 @@ #include <linux/if_bridge.h> #include <linux/list.h> +#include <linux/mutex.h> +#include <linux/refcount.h> #include <linux/rtnetlink.h> #include <linux/workqueue.h> #include <net/arp.h> @@ -19,9 +21,27 @@ struct mlxsw_sp_span { struct work_struct work; struct mlxsw_sp *mlxsw_sp; + struct list_head analyzed_ports_list; + struct mutex analyzed_ports_lock; /* Protects analyzed_ports_list */ + struct list_head trigger_entries_list; atomic_t active_entries_count; int entries_count; - struct mlxsw_sp_span_entry entries[0]; + struct mlxsw_sp_span_entry entries[]; +}; + +struct mlxsw_sp_span_analyzed_port { + struct list_head list; /* Member of analyzed_ports_list */ + refcount_t ref_count; + u8 local_port; + bool ingress; +}; + +struct mlxsw_sp_span_trigger_entry { + struct list_head list; /* Member of trigger_entries_list */ + refcount_t ref_count; + u8 local_port; + enum mlxsw_sp_span_trigger trigger; + struct mlxsw_sp_span_trigger_parms parms; }; static void mlxsw_sp_span_respin_work(struct work_struct *work); @@ -48,15 +68,14 @@ int mlxsw_sp_span_init(struct mlxsw_sp *mlxsw_sp) return -ENOMEM; span->entries_count = entries_count; atomic_set(&span->active_entries_count, 0); + mutex_init(&span->analyzed_ports_lock); + INIT_LIST_HEAD(&span->analyzed_ports_list); + INIT_LIST_HEAD(&span->trigger_entries_list); span->mlxsw_sp = mlxsw_sp; mlxsw_sp->span = span; - for (i = 0; i < mlxsw_sp->span->entries_count; i++) { - struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; - - INIT_LIST_HEAD(&curr->bound_ports_list); - curr->id = i; - } + for (i = 0; i < mlxsw_sp->span->entries_count; i++) + mlxsw_sp->span->entries[i].id = i; devlink_resource_occ_get_register(devlink, MLXSW_SP_RESOURCE_SPAN, mlxsw_sp_span_occ_get, mlxsw_sp); @@ -68,16 +87,13 @@ int mlxsw_sp_span_init(struct mlxsw_sp *mlxsw_sp) void mlxsw_sp_span_fini(struct mlxsw_sp *mlxsw_sp) { struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); - int i; cancel_work_sync(&mlxsw_sp->span->work); devlink_resource_occ_get_unregister(devlink, MLXSW_SP_RESOURCE_SPAN); - for (i = 0; i < mlxsw_sp->span->entries_count; i++) { - struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; - - WARN_ON_ONCE(!list_empty(&curr->bound_ports_list)); - } + WARN_ON_ONCE(!list_empty(&mlxsw_sp->span->trigger_entries_list)); + WARN_ON_ONCE(!list_empty(&mlxsw_sp->span->analyzed_ports_list)); + mutex_destroy(&mlxsw_sp->span->analyzed_ports_lock); kfree(mlxsw_sp->span); } @@ -130,7 +146,7 @@ mlxsw_sp_span_entry_phys_deconfigure(struct mlxsw_sp_span_entry *span_entry) static const struct mlxsw_sp_span_entry_ops mlxsw_sp_span_entry_ops_phys = { .can_handle = mlxsw_sp_port_dev_check, - .parms = mlxsw_sp_span_entry_phys_parms, + .parms_set = mlxsw_sp_span_entry_phys_parms, .configure = mlxsw_sp_span_entry_phys_configure, .deconfigure = mlxsw_sp_span_entry_phys_deconfigure, }; @@ -418,7 +434,7 @@ mlxsw_sp_span_entry_gretap4_deconfigure(struct mlxsw_sp_span_entry *span_entry) static const struct mlxsw_sp_span_entry_ops mlxsw_sp_span_entry_ops_gretap4 = { .can_handle = netif_is_gretap, - .parms = mlxsw_sp_span_entry_gretap4_parms, + .parms_set = mlxsw_sp_span_entry_gretap4_parms, .configure = mlxsw_sp_span_entry_gretap4_configure, .deconfigure = mlxsw_sp_span_entry_gretap4_deconfigure, }; @@ -519,7 +535,7 @@ mlxsw_sp_span_entry_gretap6_deconfigure(struct mlxsw_sp_span_entry *span_entry) static const struct mlxsw_sp_span_entry_ops mlxsw_sp_span_entry_ops_gretap6 = { .can_handle = netif_is_ip6gretap, - .parms = mlxsw_sp_span_entry_gretap6_parms, + .parms_set = mlxsw_sp_span_entry_gretap6_parms, .configure = mlxsw_sp_span_entry_gretap6_configure, .deconfigure = mlxsw_sp_span_entry_gretap6_deconfigure, }; @@ -575,7 +591,7 @@ mlxsw_sp_span_entry_vlan_deconfigure(struct mlxsw_sp_span_entry *span_entry) static const struct mlxsw_sp_span_entry_ops mlxsw_sp_span_entry_ops_vlan = { .can_handle = mlxsw_sp_span_vlan_can_handle, - .parms = mlxsw_sp_span_entry_vlan_parms, + .parms_set = mlxsw_sp_span_entry_vlan_parms, .configure = mlxsw_sp_span_entry_vlan_configure, .deconfigure = mlxsw_sp_span_entry_vlan_deconfigure, }; @@ -612,7 +628,7 @@ mlxsw_sp_span_entry_nop_deconfigure(struct mlxsw_sp_span_entry *span_entry) } static const struct mlxsw_sp_span_entry_ops mlxsw_sp_span_entry_ops_nop = { - .parms = mlxsw_sp_span_entry_nop_parms, + .parms_set = mlxsw_sp_span_entry_nop_parms, .configure = mlxsw_sp_span_entry_nop_configure, .deconfigure = mlxsw_sp_span_entry_nop_deconfigure, }; @@ -622,18 +638,27 @@ mlxsw_sp_span_entry_configure(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_span_entry *span_entry, struct mlxsw_sp_span_parms sparms) { - if (sparms.dest_port) { - if (sparms.dest_port->mlxsw_sp != mlxsw_sp) { - netdev_err(span_entry->to_dev, "Cannot mirror to %s, which belongs to a different mlxsw instance", - sparms.dest_port->dev->name); - sparms.dest_port = NULL; - } else if (span_entry->ops->configure(span_entry, sparms)) { - netdev_err(span_entry->to_dev, "Failed to offload mirror to %s", - sparms.dest_port->dev->name); - sparms.dest_port = NULL; - } + int err; + + if (!sparms.dest_port) + goto set_parms; + + if (sparms.dest_port->mlxsw_sp != mlxsw_sp) { + netdev_err(span_entry->to_dev, "Cannot mirror to %s, which belongs to a different mlxsw instance", + sparms.dest_port->dev->name); + sparms.dest_port = NULL; + goto set_parms; + } + + err = span_entry->ops->configure(span_entry, sparms); + if (err) { + netdev_err(span_entry->to_dev, "Failed to offload mirror to %s", + sparms.dest_port->dev->name); + sparms.dest_port = NULL; + goto set_parms; } +set_parms: span_entry->parms = sparms; } @@ -655,7 +680,7 @@ mlxsw_sp_span_entry_create(struct mlxsw_sp *mlxsw_sp, /* find a free entry to use */ for (i = 0; i < mlxsw_sp->span->entries_count; i++) { - if (!mlxsw_sp->span->entries[i].ref_count) { + if (!refcount_read(&mlxsw_sp->span->entries[i].ref_count)) { span_entry = &mlxsw_sp->span->entries[i]; break; } @@ -665,7 +690,7 @@ mlxsw_sp_span_entry_create(struct mlxsw_sp *mlxsw_sp, atomic_inc(&mlxsw_sp->span->active_entries_count); span_entry->ops = ops; - span_entry->ref_count = 1; + refcount_set(&span_entry->ref_count, 1); span_entry->to_dev = to_dev; mlxsw_sp_span_entry_configure(mlxsw_sp, span_entry, sparms); @@ -688,7 +713,7 @@ mlxsw_sp_span_entry_find_by_port(struct mlxsw_sp *mlxsw_sp, for (i = 0; i < mlxsw_sp->span->entries_count; i++) { struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; - if (curr->ref_count && curr->to_dev == to_dev) + if (refcount_read(&curr->ref_count) && curr->to_dev == to_dev) return curr; } return NULL; @@ -709,7 +734,7 @@ mlxsw_sp_span_entry_find_by_id(struct mlxsw_sp *mlxsw_sp, int span_id) for (i = 0; i < mlxsw_sp->span->entries_count; i++) { struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; - if (curr->ref_count && curr->id == span_id) + if (refcount_read(&curr->ref_count) && curr->id == span_id) return curr; } return NULL; @@ -726,7 +751,7 @@ mlxsw_sp_span_entry_get(struct mlxsw_sp *mlxsw_sp, span_entry = mlxsw_sp_span_entry_find_by_port(mlxsw_sp, to_dev); if (span_entry) { /* Already exists, just take a reference */ - span_entry->ref_count++; + refcount_inc(&span_entry->ref_count); return span_entry; } @@ -736,32 +761,13 @@ mlxsw_sp_span_entry_get(struct mlxsw_sp *mlxsw_sp, static int mlxsw_sp_span_entry_put(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_span_entry *span_entry) { - WARN_ON(!span_entry->ref_count); - if (--span_entry->ref_count == 0) + if (refcount_dec_and_test(&span_entry->ref_count)) mlxsw_sp_span_entry_destroy(mlxsw_sp, span_entry); return 0; } -static bool mlxsw_sp_span_is_egress_mirror(struct mlxsw_sp_port *port) -{ - struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; - struct mlxsw_sp_span_inspected_port *p; - int i; - - for (i = 0; i < mlxsw_sp->span->entries_count; i++) { - struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; - - list_for_each_entry(p, &curr->bound_ports_list, list) - if (p->local_port == port->local_port && - p->type == MLXSW_SP_SPAN_EGRESS) - return true; - } - - return false; -} - static int -mlxsw_sp_span_port_buffsize_update(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu) +mlxsw_sp_span_port_buffer_update(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu) { struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; char sbib_pl[MLXSW_REG_SBIB_LEN]; @@ -780,20 +786,54 @@ mlxsw_sp_span_port_buffsize_update(struct mlxsw_sp_port *mlxsw_sp_port, u16 mtu) return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); } +static void mlxsw_sp_span_port_buffer_disable(struct mlxsw_sp *mlxsw_sp, + u8 local_port) +{ + char sbib_pl[MLXSW_REG_SBIB_LEN]; + + mlxsw_reg_sbib_pack(sbib_pl, local_port, 0); + mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); +} + +static struct mlxsw_sp_span_analyzed_port * +mlxsw_sp_span_analyzed_port_find(struct mlxsw_sp_span *span, u8 local_port, + bool ingress) +{ + struct mlxsw_sp_span_analyzed_port *analyzed_port; + + list_for_each_entry(analyzed_port, &span->analyzed_ports_list, list) { + if (analyzed_port->local_port == local_port && + analyzed_port->ingress == ingress) + return analyzed_port; + } + + return NULL; +} + int mlxsw_sp_span_port_mtu_update(struct mlxsw_sp_port *port, u16 mtu) { + struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; + int err = 0; + /* If port is egress mirrored, the shared buffer size should be * updated according to the mtu value */ - if (mlxsw_sp_span_is_egress_mirror(port)) - return mlxsw_sp_span_port_buffsize_update(port, mtu); - return 0; + mutex_lock(&mlxsw_sp->span->analyzed_ports_lock); + + if (mlxsw_sp_span_analyzed_port_find(mlxsw_sp->span, port->local_port, + false)) + err = mlxsw_sp_span_port_buffer_update(port, mtu); + + mutex_unlock(&mlxsw_sp->span->analyzed_ports_lock); + + return err; } void mlxsw_sp_span_speed_update_work(struct work_struct *work) { struct delayed_work *dwork = to_delayed_work(work); struct mlxsw_sp_port *mlxsw_sp_port; + struct mlxsw_sp *mlxsw_sp; mlxsw_sp_port = container_of(dwork, struct mlxsw_sp_port, span.speed_update_dw); @@ -801,237 +841,368 @@ void mlxsw_sp_span_speed_update_work(struct work_struct *work) /* If port is egress mirrored, the shared buffer size should be * updated according to the speed value. */ - if (mlxsw_sp_span_is_egress_mirror(mlxsw_sp_port)) - mlxsw_sp_span_port_buffsize_update(mlxsw_sp_port, - mlxsw_sp_port->dev->mtu); + mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + mutex_lock(&mlxsw_sp->span->analyzed_ports_lock); + + if (mlxsw_sp_span_analyzed_port_find(mlxsw_sp->span, + mlxsw_sp_port->local_port, false)) + mlxsw_sp_span_port_buffer_update(mlxsw_sp_port, + mlxsw_sp_port->dev->mtu); + + mutex_unlock(&mlxsw_sp->span->analyzed_ports_lock); } -static struct mlxsw_sp_span_inspected_port * -mlxsw_sp_span_entry_bound_port_find(struct mlxsw_sp_span_entry *span_entry, - enum mlxsw_sp_span_type type, - struct mlxsw_sp_port *port, - bool bind) +static const struct mlxsw_sp_span_entry_ops * +mlxsw_sp_span_entry_ops(struct mlxsw_sp *mlxsw_sp, + const struct net_device *to_dev) { - struct mlxsw_sp_span_inspected_port *p; + size_t i; + + for (i = 0; i < ARRAY_SIZE(mlxsw_sp_span_entry_types); ++i) + if (mlxsw_sp_span_entry_types[i]->can_handle(to_dev)) + return mlxsw_sp_span_entry_types[i]; - list_for_each_entry(p, &span_entry->bound_ports_list, list) - if (type == p->type && - port->local_port == p->local_port && - bind == p->bound) - return p; return NULL; } -static int -mlxsw_sp_span_inspected_port_bind(struct mlxsw_sp_port *port, - struct mlxsw_sp_span_entry *span_entry, - enum mlxsw_sp_span_type type, - bool bind) +static void mlxsw_sp_span_respin_work(struct work_struct *work) { - struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; - char mpar_pl[MLXSW_REG_MPAR_LEN]; - int pa_id = span_entry->id; + struct mlxsw_sp_span *span; + struct mlxsw_sp *mlxsw_sp; + int i, err; + + span = container_of(work, struct mlxsw_sp_span, work); + mlxsw_sp = span->mlxsw_sp; + + rtnl_lock(); + for (i = 0; i < mlxsw_sp->span->entries_count; i++) { + struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; + struct mlxsw_sp_span_parms sparms = {NULL}; + + if (!refcount_read(&curr->ref_count)) + continue; - /* bind the port to the SPAN entry */ - mlxsw_reg_mpar_pack(mpar_pl, port->local_port, - (enum mlxsw_reg_mpar_i_e)type, bind, pa_id); - return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(mpar), mpar_pl); + err = curr->ops->parms_set(curr->to_dev, &sparms); + if (err) + continue; + + if (memcmp(&sparms, &curr->parms, sizeof(sparms))) { + mlxsw_sp_span_entry_deconfigure(curr); + mlxsw_sp_span_entry_configure(mlxsw_sp, curr, sparms); + } + } + rtnl_unlock(); } -static int -mlxsw_sp_span_inspected_port_add(struct mlxsw_sp_port *port, - struct mlxsw_sp_span_entry *span_entry, - enum mlxsw_sp_span_type type, - bool bind) +void mlxsw_sp_span_respin(struct mlxsw_sp *mlxsw_sp) { - struct mlxsw_sp_span_inspected_port *inspected_port; - struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; - char sbib_pl[MLXSW_REG_SBIB_LEN]; - int i; + if (atomic_read(&mlxsw_sp->span->active_entries_count) == 0) + return; + mlxsw_core_schedule_work(&mlxsw_sp->span->work); +} + +int mlxsw_sp_span_agent_get(struct mlxsw_sp *mlxsw_sp, + const struct net_device *to_dev, int *p_span_id) +{ + const struct mlxsw_sp_span_entry_ops *ops; + struct mlxsw_sp_span_entry *span_entry; + struct mlxsw_sp_span_parms sparms; int err; - /* A given (source port, direction) can only be bound to one analyzer, - * so if a binding is requested, check for conflicts. - */ - if (bind) - for (i = 0; i < mlxsw_sp->span->entries_count; i++) { - struct mlxsw_sp_span_entry *curr = - &mlxsw_sp->span->entries[i]; - - if (mlxsw_sp_span_entry_bound_port_find(curr, type, - port, bind)) - return -EEXIST; - } + ASSERT_RTNL(); - /* if it is an egress SPAN, bind a shared buffer to it */ - if (type == MLXSW_SP_SPAN_EGRESS) { - err = mlxsw_sp_span_port_buffsize_update(port, port->dev->mtu); - if (err) - return err; + ops = mlxsw_sp_span_entry_ops(mlxsw_sp, to_dev); + if (!ops) { + dev_err(mlxsw_sp->bus_info->dev, "Cannot mirror to requested destination\n"); + return -EOPNOTSUPP; } - if (bind) { - err = mlxsw_sp_span_inspected_port_bind(port, span_entry, type, - true); - if (err) - goto err_port_bind; - } + memset(&sparms, 0, sizeof(sparms)); + err = ops->parms_set(to_dev, &sparms); + if (err) + return err; - inspected_port = kzalloc(sizeof(*inspected_port), GFP_KERNEL); - if (!inspected_port) { - err = -ENOMEM; - goto err_inspected_port_alloc; - } - inspected_port->local_port = port->local_port; - inspected_port->type = type; - inspected_port->bound = bind; - list_add_tail(&inspected_port->list, &span_entry->bound_ports_list); + span_entry = mlxsw_sp_span_entry_get(mlxsw_sp, to_dev, ops, sparms); + if (!span_entry) + return -ENOBUFS; + + *p_span_id = span_entry->id; return 0; +} -err_inspected_port_alloc: - if (bind) - mlxsw_sp_span_inspected_port_bind(port, span_entry, type, - false); -err_port_bind: - if (type == MLXSW_SP_SPAN_EGRESS) { - mlxsw_reg_sbib_pack(sbib_pl, port->local_port, 0); - mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); +void mlxsw_sp_span_agent_put(struct mlxsw_sp *mlxsw_sp, int span_id) +{ + struct mlxsw_sp_span_entry *span_entry; + + ASSERT_RTNL(); + + span_entry = mlxsw_sp_span_entry_find_by_id(mlxsw_sp, span_id); + if (WARN_ON_ONCE(!span_entry)) + return; + + mlxsw_sp_span_entry_put(mlxsw_sp, span_entry); +} + +static struct mlxsw_sp_span_analyzed_port * +mlxsw_sp_span_analyzed_port_create(struct mlxsw_sp_span *span, + struct mlxsw_sp_port *mlxsw_sp_port, + bool ingress) +{ + struct mlxsw_sp_span_analyzed_port *analyzed_port; + int err; + + analyzed_port = kzalloc(sizeof(*analyzed_port), GFP_KERNEL); + if (!analyzed_port) + return ERR_PTR(-ENOMEM); + + refcount_set(&analyzed_port->ref_count, 1); + analyzed_port->local_port = mlxsw_sp_port->local_port; + analyzed_port->ingress = ingress; + list_add_tail(&analyzed_port->list, &span->analyzed_ports_list); + + /* An egress mirror buffer should be allocated on the egress port which + * does the mirroring. + */ + if (!ingress) { + u16 mtu = mlxsw_sp_port->dev->mtu; + + err = mlxsw_sp_span_port_buffer_update(mlxsw_sp_port, mtu); + if (err) + goto err_buffer_update; } - return err; + + return analyzed_port; + +err_buffer_update: + list_del(&analyzed_port->list); + kfree(analyzed_port); + return ERR_PTR(err); } static void -mlxsw_sp_span_inspected_port_del(struct mlxsw_sp_port *port, - struct mlxsw_sp_span_entry *span_entry, - enum mlxsw_sp_span_type type, - bool bind) +mlxsw_sp_span_analyzed_port_destroy(struct mlxsw_sp_span *span, + struct mlxsw_sp_span_analyzed_port * + analyzed_port) { - struct mlxsw_sp_span_inspected_port *inspected_port; - struct mlxsw_sp *mlxsw_sp = port->mlxsw_sp; - char sbib_pl[MLXSW_REG_SBIB_LEN]; + struct mlxsw_sp *mlxsw_sp = span->mlxsw_sp; - inspected_port = mlxsw_sp_span_entry_bound_port_find(span_entry, type, - port, bind); - if (!inspected_port) - return; + /* Remove egress mirror buffer now that port is no longer analyzed + * at egress. + */ + if (!analyzed_port->ingress) + mlxsw_sp_span_port_buffer_disable(mlxsw_sp, + analyzed_port->local_port); + + list_del(&analyzed_port->list); + kfree(analyzed_port); +} + +int mlxsw_sp_span_analyzed_port_get(struct mlxsw_sp_port *mlxsw_sp_port, + bool ingress) +{ + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_span_analyzed_port *analyzed_port; + u8 local_port = mlxsw_sp_port->local_port; + int err = 0; + + mutex_lock(&mlxsw_sp->span->analyzed_ports_lock); - if (bind) - mlxsw_sp_span_inspected_port_bind(port, span_entry, type, - false); - /* remove the SBIB buffer if it was egress SPAN */ - if (type == MLXSW_SP_SPAN_EGRESS) { - mlxsw_reg_sbib_pack(sbib_pl, port->local_port, 0); - mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(sbib), sbib_pl); + analyzed_port = mlxsw_sp_span_analyzed_port_find(mlxsw_sp->span, + local_port, ingress); + if (analyzed_port) { + refcount_inc(&analyzed_port->ref_count); + goto out_unlock; } - mlxsw_sp_span_entry_put(mlxsw_sp, span_entry); + analyzed_port = mlxsw_sp_span_analyzed_port_create(mlxsw_sp->span, + mlxsw_sp_port, + ingress); + if (IS_ERR(analyzed_port)) + err = PTR_ERR(analyzed_port); - list_del(&inspected_port->list); - kfree(inspected_port); +out_unlock: + mutex_unlock(&mlxsw_sp->span->analyzed_ports_lock); + return err; } -static const struct mlxsw_sp_span_entry_ops * -mlxsw_sp_span_entry_ops(struct mlxsw_sp *mlxsw_sp, - const struct net_device *to_dev) +void mlxsw_sp_span_analyzed_port_put(struct mlxsw_sp_port *mlxsw_sp_port, + bool ingress) { - size_t i; + struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp; + struct mlxsw_sp_span_analyzed_port *analyzed_port; + u8 local_port = mlxsw_sp_port->local_port; - for (i = 0; i < ARRAY_SIZE(mlxsw_sp_span_entry_types); ++i) - if (mlxsw_sp_span_entry_types[i]->can_handle(to_dev)) - return mlxsw_sp_span_entry_types[i]; + mutex_lock(&mlxsw_sp->span->analyzed_ports_lock); - return NULL; + analyzed_port = mlxsw_sp_span_analyzed_port_find(mlxsw_sp->span, + local_port, ingress); + if (WARN_ON_ONCE(!analyzed_port)) + goto out_unlock; + + if (!refcount_dec_and_test(&analyzed_port->ref_count)) + goto out_unlock; + + mlxsw_sp_span_analyzed_port_destroy(mlxsw_sp->span, analyzed_port); + +out_unlock: + mutex_unlock(&mlxsw_sp->span->analyzed_ports_lock); } -int mlxsw_sp_span_mirror_add(struct mlxsw_sp_port *from, - const struct net_device *to_dev, - enum mlxsw_sp_span_type type, bool bind, - int *p_span_id) +static int +__mlxsw_sp_span_trigger_entry_bind(struct mlxsw_sp_span *span, + struct mlxsw_sp_span_trigger_entry * + trigger_entry, bool enable) { - struct mlxsw_sp *mlxsw_sp = from->mlxsw_sp; - const struct mlxsw_sp_span_entry_ops *ops; - struct mlxsw_sp_span_parms sparms = {NULL}; - struct mlxsw_sp_span_entry *span_entry; - int err; - - ops = mlxsw_sp_span_entry_ops(mlxsw_sp, to_dev); - if (!ops) { - netdev_err(to_dev, "Cannot mirror to %s", to_dev->name); - return -EOPNOTSUPP; + char mpar_pl[MLXSW_REG_MPAR_LEN]; + enum mlxsw_reg_mpar_i_e i_e; + + switch (trigger_entry->trigger) { + case MLXSW_SP_SPAN_TRIGGER_INGRESS: + i_e = MLXSW_REG_MPAR_TYPE_INGRESS; + break; + case MLXSW_SP_SPAN_TRIGGER_EGRESS: + i_e = MLXSW_REG_MPAR_TYPE_EGRESS; + break; + default: + WARN_ON_ONCE(1); + return -EINVAL; } - err = ops->parms(to_dev, &sparms); - if (err) - return err; + mlxsw_reg_mpar_pack(mpar_pl, trigger_entry->local_port, i_e, enable, + trigger_entry->parms.span_id); + return mlxsw_reg_write(span->mlxsw_sp->core, MLXSW_REG(mpar), mpar_pl); +} - span_entry = mlxsw_sp_span_entry_get(mlxsw_sp, to_dev, ops, sparms); - if (!span_entry) - return -ENOBUFS; +static int +mlxsw_sp_span_trigger_entry_bind(struct mlxsw_sp_span *span, + struct mlxsw_sp_span_trigger_entry * + trigger_entry) +{ + return __mlxsw_sp_span_trigger_entry_bind(span, trigger_entry, true); +} - netdev_dbg(from->dev, "Adding inspected port to SPAN entry %d\n", - span_entry->id); +static void +mlxsw_sp_span_trigger_entry_unbind(struct mlxsw_sp_span *span, + struct mlxsw_sp_span_trigger_entry * + trigger_entry) +{ + __mlxsw_sp_span_trigger_entry_bind(span, trigger_entry, false); +} - err = mlxsw_sp_span_inspected_port_add(from, span_entry, type, bind); +static struct mlxsw_sp_span_trigger_entry * +mlxsw_sp_span_trigger_entry_create(struct mlxsw_sp_span *span, + enum mlxsw_sp_span_trigger trigger, + struct mlxsw_sp_port *mlxsw_sp_port, + const struct mlxsw_sp_span_trigger_parms + *parms) +{ + struct mlxsw_sp_span_trigger_entry *trigger_entry; + int err; + + trigger_entry = kzalloc(sizeof(*trigger_entry), GFP_KERNEL); + if (!trigger_entry) + return ERR_PTR(-ENOMEM); + + refcount_set(&trigger_entry->ref_count, 1); + trigger_entry->local_port = mlxsw_sp_port->local_port; + trigger_entry->trigger = trigger; + memcpy(&trigger_entry->parms, parms, sizeof(trigger_entry->parms)); + list_add_tail(&trigger_entry->list, &span->trigger_entries_list); + + err = mlxsw_sp_span_trigger_entry_bind(span, trigger_entry); if (err) - goto err_port_bind; + goto err_trigger_entry_bind; - *p_span_id = span_entry->id; - return 0; + return trigger_entry; -err_port_bind: - mlxsw_sp_span_entry_put(mlxsw_sp, span_entry); - return err; +err_trigger_entry_bind: + list_del(&trigger_entry->list); + kfree(trigger_entry); + return ERR_PTR(err); } -void mlxsw_sp_span_mirror_del(struct mlxsw_sp_port *from, int span_id, - enum mlxsw_sp_span_type type, bool bind) +static void +mlxsw_sp_span_trigger_entry_destroy(struct mlxsw_sp_span *span, + struct mlxsw_sp_span_trigger_entry * + trigger_entry) { - struct mlxsw_sp_span_entry *span_entry; + mlxsw_sp_span_trigger_entry_unbind(span, trigger_entry); + list_del(&trigger_entry->list); + kfree(trigger_entry); +} - span_entry = mlxsw_sp_span_entry_find_by_id(from->mlxsw_sp, span_id); - if (!span_entry) { - netdev_err(from->dev, "no span entry found\n"); - return; +static struct mlxsw_sp_span_trigger_entry * +mlxsw_sp_span_trigger_entry_find(struct mlxsw_sp_span *span, + enum mlxsw_sp_span_trigger trigger, + struct mlxsw_sp_port *mlxsw_sp_port) +{ + struct mlxsw_sp_span_trigger_entry *trigger_entry; + + list_for_each_entry(trigger_entry, &span->trigger_entries_list, list) { + if (trigger_entry->trigger == trigger && + trigger_entry->local_port == mlxsw_sp_port->local_port) + return trigger_entry; } - netdev_dbg(from->dev, "removing inspected port from SPAN entry %d\n", - span_entry->id); - mlxsw_sp_span_inspected_port_del(from, span_entry, type, bind); + return NULL; } -static void mlxsw_sp_span_respin_work(struct work_struct *work) +int mlxsw_sp_span_agent_bind(struct mlxsw_sp *mlxsw_sp, + enum mlxsw_sp_span_trigger trigger, + struct mlxsw_sp_port *mlxsw_sp_port, + const struct mlxsw_sp_span_trigger_parms *parms) { - struct mlxsw_sp_span *span; - struct mlxsw_sp *mlxsw_sp; - int i, err; + struct mlxsw_sp_span_trigger_entry *trigger_entry; + int err = 0; - span = container_of(work, struct mlxsw_sp_span, work); - mlxsw_sp = span->mlxsw_sp; + ASSERT_RTNL(); - rtnl_lock(); - for (i = 0; i < mlxsw_sp->span->entries_count; i++) { - struct mlxsw_sp_span_entry *curr = &mlxsw_sp->span->entries[i]; - struct mlxsw_sp_span_parms sparms = {NULL}; + if (!mlxsw_sp_span_entry_find_by_id(mlxsw_sp, parms->span_id)) + return -EINVAL; - if (!curr->ref_count) - continue; + trigger_entry = mlxsw_sp_span_trigger_entry_find(mlxsw_sp->span, + trigger, + mlxsw_sp_port); + if (trigger_entry) { + if (trigger_entry->parms.span_id != parms->span_id) + return -EINVAL; + refcount_inc(&trigger_entry->ref_count); + goto out; + } - err = curr->ops->parms(curr->to_dev, &sparms); - if (err) - continue; + trigger_entry = mlxsw_sp_span_trigger_entry_create(mlxsw_sp->span, + trigger, + mlxsw_sp_port, + parms); + if (IS_ERR(trigger_entry)) + err = PTR_ERR(trigger_entry); - if (memcmp(&sparms, &curr->parms, sizeof(sparms))) { - mlxsw_sp_span_entry_deconfigure(curr); - mlxsw_sp_span_entry_configure(mlxsw_sp, curr, sparms); - } - } - rtnl_unlock(); +out: + return err; } -void mlxsw_sp_span_respin(struct mlxsw_sp *mlxsw_sp) +void mlxsw_sp_span_agent_unbind(struct mlxsw_sp *mlxsw_sp, + enum mlxsw_sp_span_trigger trigger, + struct mlxsw_sp_port *mlxsw_sp_port, + const struct mlxsw_sp_span_trigger_parms *parms) { - if (atomic_read(&mlxsw_sp->span->active_entries_count) == 0) + struct mlxsw_sp_span_trigger_entry *trigger_entry; + + ASSERT_RTNL(); + + if (WARN_ON_ONCE(!mlxsw_sp_span_entry_find_by_id(mlxsw_sp, + parms->span_id))) return; - mlxsw_core_schedule_work(&mlxsw_sp->span->work); + + trigger_entry = mlxsw_sp_span_trigger_entry_find(mlxsw_sp->span, + trigger, + mlxsw_sp_port); + if (WARN_ON_ONCE(!trigger_entry)) + return; + + if (!refcount_dec_and_test(&trigger_entry->ref_count)) + return; + + mlxsw_sp_span_trigger_entry_destroy(mlxsw_sp->span, trigger_entry); } diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h index 59724335525f..9f6dd2d0f4e6 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_span.h @@ -6,26 +6,13 @@ #include <linux/types.h> #include <linux/if_ether.h> +#include <linux/refcount.h> #include "spectrum_router.h" struct mlxsw_sp; struct mlxsw_sp_port; -enum mlxsw_sp_span_type { - MLXSW_SP_SPAN_EGRESS, - MLXSW_SP_SPAN_INGRESS -}; - -struct mlxsw_sp_span_inspected_port { - struct list_head list; - enum mlxsw_sp_span_type type; - u8 local_port; - - /* Whether this is a directly bound mirror (port-to-port) or an ACL. */ - bool bound; -}; - struct mlxsw_sp_span_parms { struct mlxsw_sp_port *dest_port; /* NULL for unoffloaded SPAN. */ unsigned int ttl; @@ -36,21 +23,29 @@ struct mlxsw_sp_span_parms { u16 vid; }; +enum mlxsw_sp_span_trigger { + MLXSW_SP_SPAN_TRIGGER_INGRESS, + MLXSW_SP_SPAN_TRIGGER_EGRESS, +}; + +struct mlxsw_sp_span_trigger_parms { + int span_id; +}; + struct mlxsw_sp_span_entry_ops; struct mlxsw_sp_span_entry { const struct net_device *to_dev; const struct mlxsw_sp_span_entry_ops *ops; struct mlxsw_sp_span_parms parms; - struct list_head bound_ports_list; - int ref_count; + refcount_t ref_count; int id; }; struct mlxsw_sp_span_entry_ops { bool (*can_handle)(const struct net_device *to_dev); - int (*parms)(const struct net_device *to_dev, - struct mlxsw_sp_span_parms *sparmsp); + int (*parms_set)(const struct net_device *to_dev, + struct mlxsw_sp_span_parms *sparmsp); int (*configure)(struct mlxsw_sp_span_entry *span_entry, struct mlxsw_sp_span_parms sparms); void (*deconfigure)(struct mlxsw_sp_span_entry *span_entry); @@ -60,12 +55,6 @@ int mlxsw_sp_span_init(struct mlxsw_sp *mlxsw_sp); void mlxsw_sp_span_fini(struct mlxsw_sp *mlxsw_sp); void mlxsw_sp_span_respin(struct mlxsw_sp *mlxsw_sp); -int mlxsw_sp_span_mirror_add(struct mlxsw_sp_port *from, - const struct net_device *to_dev, - enum mlxsw_sp_span_type type, - bool bind, int *p_span_id); -void mlxsw_sp_span_mirror_del(struct mlxsw_sp_port *from, int span_id, - enum mlxsw_sp_span_type type, bool bind); struct mlxsw_sp_span_entry * mlxsw_sp_span_entry_find_by_port(struct mlxsw_sp *mlxsw_sp, const struct net_device *to_dev); @@ -76,4 +65,21 @@ void mlxsw_sp_span_entry_invalidate(struct mlxsw_sp *mlxsw_sp, int mlxsw_sp_span_port_mtu_update(struct mlxsw_sp_port *port, u16 mtu); void mlxsw_sp_span_speed_update_work(struct work_struct *work); +int mlxsw_sp_span_agent_get(struct mlxsw_sp *mlxsw_sp, + const struct net_device *to_dev, int *p_span_id); +void mlxsw_sp_span_agent_put(struct mlxsw_sp *mlxsw_sp, int span_id); +int mlxsw_sp_span_analyzed_port_get(struct mlxsw_sp_port *mlxsw_sp_port, + bool ingress); +void mlxsw_sp_span_analyzed_port_put(struct mlxsw_sp_port *mlxsw_sp_port, + bool ingress); +int mlxsw_sp_span_agent_bind(struct mlxsw_sp *mlxsw_sp, + enum mlxsw_sp_span_trigger trigger, + struct mlxsw_sp_port *mlxsw_sp_port, + const struct mlxsw_sp_span_trigger_parms *parms); +void +mlxsw_sp_span_agent_unbind(struct mlxsw_sp *mlxsw_sp, + enum mlxsw_sp_span_trigger trigger, + struct mlxsw_sp_port *mlxsw_sp_port, + const struct mlxsw_sp_span_trigger_parms *parms); + #endif diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c index fbf714d027d8..157a42c63066 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.c @@ -12,6 +12,24 @@ #include "spectrum.h" #include "spectrum_trap.h" +struct mlxsw_sp_trap_policer_item { + struct devlink_trap_policer policer; + u16 hw_id; +}; + +struct mlxsw_sp_trap_group_item { + struct devlink_trap_group group; + u16 hw_group_id; + u8 priority; +}; + +#define MLXSW_SP_TRAP_LISTENERS_MAX 3 + +struct mlxsw_sp_trap_item { + struct devlink_trap trap; + struct mlxsw_listener listeners_arr[MLXSW_SP_TRAP_LISTENERS_MAX]; +}; + /* All driver-specific traps must be documented in * Documentation/networking/devlink/mlxsw.rst */ @@ -107,8 +125,8 @@ static void mlxsw_sp_rx_acl_drop_listener(struct sk_buff *skb, u8 local_port, consume_skb(skb); } -static void mlxsw_sp_rx_exception_listener(struct sk_buff *skb, u8 local_port, - void *trap_ctx) +static int __mlxsw_sp_rx_no_mark_listener(struct sk_buff *skb, u8 local_port, + void *trap_ctx) { struct devlink_port *in_devlink_port; struct mlxsw_sp_port *mlxsw_sp_port; @@ -121,7 +139,7 @@ static void mlxsw_sp_rx_exception_listener(struct sk_buff *skb, u8 local_port, err = mlxsw_sp_rx_listener(mlxsw_sp, skb, local_port, mlxsw_sp_port); if (err) - return; + return err; devlink = priv_to_devlink(mlxsw_sp->core); in_devlink_port = mlxsw_core_port_devlink_port_get(mlxsw_sp->core, @@ -129,10 +147,71 @@ static void mlxsw_sp_rx_exception_listener(struct sk_buff *skb, u8 local_port, skb_push(skb, ETH_HLEN); devlink_trap_report(devlink, skb, trap_ctx, in_devlink_port, NULL); skb_pull(skb, ETH_HLEN); - skb->offload_fwd_mark = 1; + + return 0; +} + +static void mlxsw_sp_rx_no_mark_listener(struct sk_buff *skb, u8 local_port, + void *trap_ctx) +{ + int err; + + err = __mlxsw_sp_rx_no_mark_listener(skb, local_port, trap_ctx); + if (err) + return; + netif_receive_skb(skb); } +static void mlxsw_sp_rx_mark_listener(struct sk_buff *skb, u8 local_port, + void *trap_ctx) +{ + skb->offload_fwd_mark = 1; + mlxsw_sp_rx_no_mark_listener(skb, local_port, trap_ctx); +} + +static void mlxsw_sp_rx_l3_mark_listener(struct sk_buff *skb, u8 local_port, + void *trap_ctx) +{ + skb->offload_l3_fwd_mark = 1; + skb->offload_fwd_mark = 1; + mlxsw_sp_rx_no_mark_listener(skb, local_port, trap_ctx); +} + +static void mlxsw_sp_rx_ptp_listener(struct sk_buff *skb, u8 local_port, + void *trap_ctx) +{ + struct mlxsw_sp *mlxsw_sp = devlink_trap_ctx_priv(trap_ctx); + int err; + + err = __mlxsw_sp_rx_no_mark_listener(skb, local_port, trap_ctx); + if (err) + return; + + /* The PTP handler expects skb->data to point to the start of the + * Ethernet header. + */ + skb_push(skb, ETH_HLEN); + mlxsw_sp_ptp_receive(mlxsw_sp, skb, local_port); +} + +static void mlxsw_sp_rx_sample_listener(struct sk_buff *skb, u8 local_port, + void *trap_ctx) +{ + struct mlxsw_sp *mlxsw_sp = devlink_trap_ctx_priv(trap_ctx); + int err; + + err = __mlxsw_sp_rx_no_mark_listener(skb, local_port, trap_ctx); + if (err) + return; + + /* The sample handler expects skb->data to point to the start of the + * Ethernet header. + */ + skb_push(skb, ETH_HLEN); + mlxsw_sp_sample_receive(mlxsw_sp, skb, local_port); +} + #define MLXSW_SP_TRAP_DROP(_id, _group_id) \ DEVLINK_TRAP_GENERIC(DROP, DROP, _id, \ DEVLINK_TRAP_GROUP_GENERIC_ID_##_group_id, \ @@ -154,6 +233,11 @@ static void mlxsw_sp_rx_exception_listener(struct sk_buff *skb, u8 local_port, DEVLINK_TRAP_GROUP_GENERIC_ID_##_group_id, \ MLXSW_SP_TRAP_METADATA) +#define MLXSW_SP_TRAP_CONTROL(_id, _group_id, _action) \ + DEVLINK_TRAP_GENERIC(CONTROL, _action, _id, \ + DEVLINK_TRAP_GROUP_GENERIC_ID_##_group_id, \ + MLXSW_SP_TRAP_METADATA) + #define MLXSW_SP_RXL_DISCARD(_id, _group_id) \ MLXSW_RXL_DIS(mlxsw_sp_rx_drop_listener, DISCARD_##_id, \ TRAP_EXCEPTION_TO_CPU, false, SP_##_group_id, \ @@ -165,9 +249,21 @@ static void mlxsw_sp_rx_exception_listener(struct sk_buff *skb, u8 local_port, SET_FW_DEFAULT, SP_##_dis_group_id) #define MLXSW_SP_RXL_EXCEPTION(_id, _group_id, _action) \ - MLXSW_RXL(mlxsw_sp_rx_exception_listener, _id, \ + MLXSW_RXL(mlxsw_sp_rx_mark_listener, _id, \ _action, false, SP_##_group_id, SET_FW_DEFAULT) +#define MLXSW_SP_RXL_NO_MARK(_id, _group_id, _action, _is_ctrl) \ + MLXSW_RXL(mlxsw_sp_rx_no_mark_listener, _id, _action, \ + _is_ctrl, SP_##_group_id, DISCARD) + +#define MLXSW_SP_RXL_MARK(_id, _group_id, _action, _is_ctrl) \ + MLXSW_RXL(mlxsw_sp_rx_mark_listener, _id, _action, _is_ctrl, \ + SP_##_group_id, DISCARD) + +#define MLXSW_SP_RXL_L3_MARK(_id, _group_id, _action, _is_ctrl) \ + MLXSW_RXL(mlxsw_sp_rx_l3_mark_listener, _id, _action, _is_ctrl, \ + SP_##_group_id, DISCARD) + #define MLXSW_SP_TRAP_POLICER(_id, _rate, _burst) \ DEVLINK_TRAP_POLICER(_id, _rate, _burst, \ MLXSW_REG_QPCR_HIGHEST_CIR, \ @@ -176,149 +272,753 @@ static void mlxsw_sp_rx_exception_listener(struct sk_buff *skb, u8 local_port, 1 << MLXSW_REG_QPCR_LOWEST_CBS) /* Ordered by policer identifier */ -static const struct devlink_trap_policer mlxsw_sp_trap_policers_arr[] = { - MLXSW_SP_TRAP_POLICER(1, 10 * 1024, 128), +static const struct mlxsw_sp_trap_policer_item +mlxsw_sp_trap_policer_items_arr[] = { + { + .policer = MLXSW_SP_TRAP_POLICER(1, 10 * 1024, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(2, 128, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(3, 128, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(4, 128, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(5, 16 * 1024, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(6, 128, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(7, 1024, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(8, 20 * 1024, 1024), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(9, 128, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(10, 1024, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(11, 360, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(12, 128, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(13, 128, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(14, 1024, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(15, 1024, 128), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(16, 24 * 1024, 4096), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(17, 19 * 1024, 4096), + }, + { + .policer = MLXSW_SP_TRAP_POLICER(18, 1024, 128), + }, }; -static const struct devlink_trap_group mlxsw_sp_trap_groups_arr[] = { - DEVLINK_TRAP_GROUP_GENERIC(L2_DROPS, 1), - DEVLINK_TRAP_GROUP_GENERIC(L3_DROPS, 1), - DEVLINK_TRAP_GROUP_GENERIC(TUNNEL_DROPS, 1), - DEVLINK_TRAP_GROUP_GENERIC(ACL_DROPS, 1), +static const struct mlxsw_sp_trap_group_item mlxsw_sp_trap_group_items_arr[] = { + { + .group = DEVLINK_TRAP_GROUP_GENERIC(L2_DROPS, 1), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_L2_DISCARDS, + .priority = 0, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(L3_DROPS, 1), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_L3_DISCARDS, + .priority = 0, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(L3_EXCEPTIONS, 1), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_L3_EXCEPTIONS, + .priority = 2, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(TUNNEL_DROPS, 1), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_TUNNEL_DISCARDS, + .priority = 0, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(ACL_DROPS, 1), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_ACL_DISCARDS, + .priority = 0, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(STP, 2), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_STP, + .priority = 5, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(LACP, 3), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_LACP, + .priority = 5, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(LLDP, 4), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_LLDP, + .priority = 5, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(MC_SNOOPING, 5), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_MC_SNOOPING, + .priority = 3, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(DHCP, 6), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_DHCP, + .priority = 2, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(NEIGH_DISCOVERY, 7), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_NEIGH_DISCOVERY, + .priority = 2, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(BFD, 8), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_BFD, + .priority = 5, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(OSPF, 9), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_OSPF, + .priority = 5, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(BGP, 10), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_BGP, + .priority = 4, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(VRRP, 11), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_VRRP, + .priority = 5, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(PIM, 12), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_PIM, + .priority = 5, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(UC_LB, 13), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_LBERROR, + .priority = 0, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(LOCAL_DELIVERY, 14), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_IP2ME, + .priority = 2, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(IPV6, 15), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_IPV6, + .priority = 2, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(PTP_EVENT, 16), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_PTP0, + .priority = 5, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(PTP_GENERAL, 17), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_PTP1, + .priority = 2, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(ACL_SAMPLE, 0), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_PKT_SAMPLE, + .priority = 0, + }, + { + .group = DEVLINK_TRAP_GROUP_GENERIC(ACL_TRAP, 18), + .hw_group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_FLOW_LOGGING, + .priority = 4, + }, }; -static const struct devlink_trap mlxsw_sp_traps_arr[] = { - MLXSW_SP_TRAP_DROP(SMAC_MC, L2_DROPS), - MLXSW_SP_TRAP_DROP(VLAN_TAG_MISMATCH, L2_DROPS), - MLXSW_SP_TRAP_DROP(INGRESS_VLAN_FILTER, L2_DROPS), - MLXSW_SP_TRAP_DROP(INGRESS_STP_FILTER, L2_DROPS), - MLXSW_SP_TRAP_DROP(EMPTY_TX_LIST, L2_DROPS), - MLXSW_SP_TRAP_DROP(PORT_LOOPBACK_FILTER, L2_DROPS), - MLXSW_SP_TRAP_DROP(BLACKHOLE_ROUTE, L3_DROPS), - MLXSW_SP_TRAP_DROP(NON_IP_PACKET, L3_DROPS), - MLXSW_SP_TRAP_DROP(UC_DIP_MC_DMAC, L3_DROPS), - MLXSW_SP_TRAP_DROP(DIP_LB, L3_DROPS), - MLXSW_SP_TRAP_DROP(SIP_MC, L3_DROPS), - MLXSW_SP_TRAP_DROP(SIP_LB, L3_DROPS), - MLXSW_SP_TRAP_DROP(CORRUPTED_IP_HDR, L3_DROPS), - MLXSW_SP_TRAP_DROP(IPV4_SIP_BC, L3_DROPS), - MLXSW_SP_TRAP_DROP(IPV6_MC_DIP_RESERVED_SCOPE, L3_DROPS), - MLXSW_SP_TRAP_DROP(IPV6_MC_DIP_INTERFACE_LOCAL_SCOPE, L3_DROPS), - MLXSW_SP_TRAP_EXCEPTION(MTU_ERROR, L3_DROPS), - MLXSW_SP_TRAP_EXCEPTION(TTL_ERROR, L3_DROPS), - MLXSW_SP_TRAP_EXCEPTION(RPF, L3_DROPS), - MLXSW_SP_TRAP_EXCEPTION(REJECT_ROUTE, L3_DROPS), - MLXSW_SP_TRAP_EXCEPTION(UNRESOLVED_NEIGH, L3_DROPS), - MLXSW_SP_TRAP_EXCEPTION(IPV4_LPM_UNICAST_MISS, L3_DROPS), - MLXSW_SP_TRAP_EXCEPTION(IPV6_LPM_UNICAST_MISS, L3_DROPS), - MLXSW_SP_TRAP_DRIVER_DROP(IRIF_DISABLED, L3_DROPS), - MLXSW_SP_TRAP_DRIVER_DROP(ERIF_DISABLED, L3_DROPS), - MLXSW_SP_TRAP_DROP(NON_ROUTABLE, L3_DROPS), - MLXSW_SP_TRAP_EXCEPTION(DECAP_ERROR, TUNNEL_DROPS), - MLXSW_SP_TRAP_DROP(OVERLAY_SMAC_MC, TUNNEL_DROPS), - MLXSW_SP_TRAP_DROP_EXT(INGRESS_FLOW_ACTION_DROP, ACL_DROPS, - DEVLINK_TRAP_METADATA_TYPE_F_FA_COOKIE), - MLXSW_SP_TRAP_DROP_EXT(EGRESS_FLOW_ACTION_DROP, ACL_DROPS, - DEVLINK_TRAP_METADATA_TYPE_F_FA_COOKIE), +static const struct mlxsw_sp_trap_item mlxsw_sp_trap_items_arr[] = { + { + .trap = MLXSW_SP_TRAP_DROP(SMAC_MC, L2_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ING_PACKET_SMAC_MC, L2_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(VLAN_TAG_MISMATCH, L2_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ING_SWITCH_VTAG_ALLOW, + L2_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(INGRESS_VLAN_FILTER, L2_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ING_SWITCH_VLAN, L2_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(INGRESS_STP_FILTER, L2_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ING_SWITCH_STP, L2_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(EMPTY_TX_LIST, L2_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(LOOKUP_SWITCH_UC, L2_DISCARDS), + MLXSW_SP_RXL_DISCARD(LOOKUP_SWITCH_MC_NULL, L2_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(PORT_LOOPBACK_FILTER, L2_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(LOOKUP_SWITCH_LB, L2_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(BLACKHOLE_ROUTE, L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ROUTER2, L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(NON_IP_PACKET, L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ING_ROUTER_NON_IP_PACKET, + L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(UC_DIP_MC_DMAC, L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ING_ROUTER_UC_DIP_MC_DMAC, + L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(DIP_LB, L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ING_ROUTER_DIP_LB, L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(SIP_MC, L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ING_ROUTER_SIP_MC, L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(SIP_LB, L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ING_ROUTER_SIP_LB, L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(CORRUPTED_IP_HDR, L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ING_ROUTER_CORRUPTED_IP_HDR, + L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(IPV4_SIP_BC, L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ING_ROUTER_IPV4_SIP_BC, + L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(IPV6_MC_DIP_RESERVED_SCOPE, + L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(IPV6_MC_DIP_RESERVED_SCOPE, + L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(IPV6_MC_DIP_INTERFACE_LOCAL_SCOPE, + L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(IPV6_MC_DIP_INTERFACE_LOCAL_SCOPE, + L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_EXCEPTION(MTU_ERROR, L3_EXCEPTIONS), + .listeners_arr = { + MLXSW_SP_RXL_EXCEPTION(MTUERROR, L3_EXCEPTIONS, + TRAP_TO_CPU), + }, + }, + { + .trap = MLXSW_SP_TRAP_EXCEPTION(TTL_ERROR, L3_EXCEPTIONS), + .listeners_arr = { + MLXSW_SP_RXL_EXCEPTION(TTLERROR, L3_EXCEPTIONS, + TRAP_TO_CPU), + }, + }, + { + .trap = MLXSW_SP_TRAP_EXCEPTION(RPF, L3_EXCEPTIONS), + .listeners_arr = { + MLXSW_SP_RXL_EXCEPTION(RPF, L3_EXCEPTIONS, TRAP_TO_CPU), + }, + }, + { + .trap = MLXSW_SP_TRAP_EXCEPTION(REJECT_ROUTE, L3_EXCEPTIONS), + .listeners_arr = { + MLXSW_SP_RXL_EXCEPTION(RTR_INGRESS1, L3_EXCEPTIONS, + TRAP_TO_CPU), + }, + }, + { + .trap = MLXSW_SP_TRAP_EXCEPTION(UNRESOLVED_NEIGH, + L3_EXCEPTIONS), + .listeners_arr = { + MLXSW_SP_RXL_EXCEPTION(HOST_MISS_IPV4, L3_EXCEPTIONS, + TRAP_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(HOST_MISS_IPV6, L3_EXCEPTIONS, + TRAP_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER3, L3_EXCEPTIONS, + TRAP_EXCEPTION_TO_CPU), + }, + }, + { + .trap = MLXSW_SP_TRAP_EXCEPTION(IPV4_LPM_UNICAST_MISS, + L3_EXCEPTIONS), + .listeners_arr = { + MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER_LPM4, + L3_EXCEPTIONS, + TRAP_EXCEPTION_TO_CPU), + }, + }, + { + .trap = MLXSW_SP_TRAP_EXCEPTION(IPV6_LPM_UNICAST_MISS, + L3_EXCEPTIONS), + .listeners_arr = { + MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER_LPM6, + L3_EXCEPTIONS, + TRAP_EXCEPTION_TO_CPU), + }, + }, + { + .trap = MLXSW_SP_TRAP_DRIVER_DROP(IRIF_DISABLED, L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ROUTER_IRIF_EN, L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DRIVER_DROP(ERIF_DISABLED, L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(ROUTER_ERIF_EN, L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(NON_ROUTABLE, L3_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(NON_ROUTABLE, L3_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_EXCEPTION(DECAP_ERROR, TUNNEL_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_EXCEPTION(DECAP_ECN0, TUNNEL_DISCARDS, + TRAP_EXCEPTION_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(IPIP_DECAP_ERROR, + TUNNEL_DISCARDS, + TRAP_EXCEPTION_TO_CPU), + MLXSW_SP_RXL_EXCEPTION(DISCARD_DEC_PKT, TUNNEL_DISCARDS, + TRAP_EXCEPTION_TO_CPU), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP(OVERLAY_SMAC_MC, TUNNEL_DROPS), + .listeners_arr = { + MLXSW_SP_RXL_DISCARD(OVERLAY_SMAC_MC, TUNNEL_DISCARDS), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP_EXT(INGRESS_FLOW_ACTION_DROP, + ACL_DROPS, + DEVLINK_TRAP_METADATA_TYPE_F_FA_COOKIE), + .listeners_arr = { + MLXSW_SP_RXL_ACL_DISCARD(INGRESS_ACL, ACL_DISCARDS, + DUMMY), + }, + }, + { + .trap = MLXSW_SP_TRAP_DROP_EXT(EGRESS_FLOW_ACTION_DROP, + ACL_DROPS, + DEVLINK_TRAP_METADATA_TYPE_F_FA_COOKIE), + .listeners_arr = { + MLXSW_SP_RXL_ACL_DISCARD(EGRESS_ACL, ACL_DISCARDS, + DUMMY), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(STP, STP, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_NO_MARK(STP, STP, TRAP_TO_CPU, true), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(LACP, LACP, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_NO_MARK(LACP, LACP, TRAP_TO_CPU, true), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(LLDP, LLDP, TRAP), + .listeners_arr = { + MLXSW_RXL(mlxsw_sp_rx_ptp_listener, LLDP, TRAP_TO_CPU, + false, SP_LLDP, DISCARD), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IGMP_QUERY, MC_SNOOPING, MIRROR), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IGMP_QUERY, MC_SNOOPING, + MIRROR_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IGMP_V1_REPORT, MC_SNOOPING, + TRAP), + .listeners_arr = { + MLXSW_SP_RXL_NO_MARK(IGMP_V1_REPORT, MC_SNOOPING, + TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IGMP_V2_REPORT, MC_SNOOPING, + TRAP), + .listeners_arr = { + MLXSW_SP_RXL_NO_MARK(IGMP_V2_REPORT, MC_SNOOPING, + TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IGMP_V3_REPORT, MC_SNOOPING, + TRAP), + .listeners_arr = { + MLXSW_SP_RXL_NO_MARK(IGMP_V3_REPORT, MC_SNOOPING, + TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IGMP_V2_LEAVE, MC_SNOOPING, + TRAP), + .listeners_arr = { + MLXSW_SP_RXL_NO_MARK(IGMP_V2_LEAVE, MC_SNOOPING, + TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(MLD_QUERY, MC_SNOOPING, MIRROR), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV6_MLDV12_LISTENER_QUERY, + MC_SNOOPING, MIRROR_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(MLD_V1_REPORT, MC_SNOOPING, + TRAP), + .listeners_arr = { + MLXSW_SP_RXL_NO_MARK(IPV6_MLDV1_LISTENER_REPORT, + MC_SNOOPING, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(MLD_V2_REPORT, MC_SNOOPING, + TRAP), + .listeners_arr = { + MLXSW_SP_RXL_NO_MARK(IPV6_MLDV2_LISTENER_REPORT, + MC_SNOOPING, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(MLD_V1_DONE, MC_SNOOPING, + TRAP), + .listeners_arr = { + MLXSW_SP_RXL_NO_MARK(IPV6_MLDV1_LISTENER_DONE, + MC_SNOOPING, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV4_DHCP, DHCP, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV4_DHCP, DHCP, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_DHCP, DHCP, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV6_DHCP, DHCP, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(ARP_REQUEST, NEIGH_DISCOVERY, + MIRROR), + .listeners_arr = { + MLXSW_SP_RXL_MARK(ARPBC, NEIGH_DISCOVERY, MIRROR_TO_CPU, + false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(ARP_RESPONSE, NEIGH_DISCOVERY, + MIRROR), + .listeners_arr = { + MLXSW_SP_RXL_MARK(ARPUC, NEIGH_DISCOVERY, MIRROR_TO_CPU, + false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(ARP_OVERLAY, NEIGH_DISCOVERY, + TRAP), + .listeners_arr = { + MLXSW_SP_RXL_NO_MARK(NVE_DECAP_ARP, NEIGH_DISCOVERY, + TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_NEIGH_SOLICIT, + NEIGH_DISCOVERY, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(L3_IPV6_NEIGHBOR_SOLICITATION, + NEIGH_DISCOVERY, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_NEIGH_ADVERT, + NEIGH_DISCOVERY, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(L3_IPV6_NEIGHBOR_ADVERTISEMENT, + NEIGH_DISCOVERY, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV4_BFD, BFD, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV4_BFD, BFD, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_BFD, BFD, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV6_BFD, BFD, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV4_OSPF, OSPF, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV4_OSPF, OSPF, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_OSPF, OSPF, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV6_OSPF, OSPF, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV4_BGP, BGP, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV4_BGP, BGP, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_BGP, BGP, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV6_BGP, BGP, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV4_VRRP, VRRP, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV4_VRRP, VRRP, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_VRRP, VRRP, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV6_VRRP, VRRP, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV4_PIM, PIM, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV4_PIM, PIM, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_PIM, PIM, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV6_PIM, PIM, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(UC_LB, UC_LB, MIRROR), + .listeners_arr = { + MLXSW_SP_RXL_L3_MARK(LBERROR, LBERROR, MIRROR_TO_CPU, + false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(LOCAL_ROUTE, LOCAL_DELIVERY, + TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IP2ME, IP2ME, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(EXTERNAL_ROUTE, LOCAL_DELIVERY, + TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(RTR_INGRESS0, IP2ME, TRAP_TO_CPU, + false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_UC_DIP_LINK_LOCAL_SCOPE, + LOCAL_DELIVERY, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV6_LINK_LOCAL_DEST, IP2ME, + TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV4_ROUTER_ALERT, LOCAL_DELIVERY, + TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(ROUTER_ALERT_IPV4, IP2ME, TRAP_TO_CPU, + false), + }, + }, + { + /* IPV6_ROUTER_ALERT is defined in uAPI as 22, but it is not + * used in this file, so undefine it. + */ + #undef IPV6_ROUTER_ALERT + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_ROUTER_ALERT, LOCAL_DELIVERY, + TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(ROUTER_ALERT_IPV6, IP2ME, TRAP_TO_CPU, + false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_DIP_ALL_NODES, IPV6, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV6_ALL_NODES_LINK, IPV6, + TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_DIP_ALL_ROUTERS, IPV6, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(IPV6_ALL_ROUTERS_LINK, IPV6, + TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_ROUTER_SOLICIT, IPV6, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(L3_IPV6_ROUTER_SOLICITATION, IPV6, + TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_ROUTER_ADVERT, IPV6, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(L3_IPV6_ROUTER_ADVERTISEMENT, IPV6, + TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(IPV6_REDIRECT, IPV6, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_MARK(L3_IPV6_REDIRECTION, IPV6, + TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(PTP_EVENT, PTP_EVENT, TRAP), + .listeners_arr = { + MLXSW_RXL(mlxsw_sp_rx_ptp_listener, PTP0, TRAP_TO_CPU, + false, SP_PTP0, DISCARD), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(PTP_GENERAL, PTP_GENERAL, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_NO_MARK(PTP1, PTP1, TRAP_TO_CPU, false), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(FLOW_ACTION_SAMPLE, ACL_SAMPLE, + MIRROR), + .listeners_arr = { + MLXSW_RXL(mlxsw_sp_rx_sample_listener, PKT_SAMPLE, + MIRROR_TO_CPU, false, SP_PKT_SAMPLE, DISCARD), + }, + }, + { + .trap = MLXSW_SP_TRAP_CONTROL(FLOW_ACTION_TRAP, ACL_TRAP, TRAP), + .listeners_arr = { + MLXSW_SP_RXL_NO_MARK(ACL0, FLOW_LOGGING, TRAP_TO_CPU, + false), + }, + }, }; -static const struct mlxsw_listener mlxsw_sp_listeners_arr[] = { - MLXSW_SP_RXL_DISCARD(ING_PACKET_SMAC_MC, L2_DISCARDS), - MLXSW_SP_RXL_DISCARD(ING_SWITCH_VTAG_ALLOW, L2_DISCARDS), - MLXSW_SP_RXL_DISCARD(ING_SWITCH_VLAN, L2_DISCARDS), - MLXSW_SP_RXL_DISCARD(ING_SWITCH_STP, L2_DISCARDS), - MLXSW_SP_RXL_DISCARD(LOOKUP_SWITCH_UC, L2_DISCARDS), - MLXSW_SP_RXL_DISCARD(LOOKUP_SWITCH_MC_NULL, L2_DISCARDS), - MLXSW_SP_RXL_DISCARD(LOOKUP_SWITCH_LB, L2_DISCARDS), - MLXSW_SP_RXL_DISCARD(ROUTER2, L3_DISCARDS), - MLXSW_SP_RXL_DISCARD(ING_ROUTER_NON_IP_PACKET, L3_DISCARDS), - MLXSW_SP_RXL_DISCARD(ING_ROUTER_UC_DIP_MC_DMAC, L3_DISCARDS), - MLXSW_SP_RXL_DISCARD(ING_ROUTER_DIP_LB, L3_DISCARDS), - MLXSW_SP_RXL_DISCARD(ING_ROUTER_SIP_MC, L3_DISCARDS), - MLXSW_SP_RXL_DISCARD(ING_ROUTER_SIP_LB, L3_DISCARDS), - MLXSW_SP_RXL_DISCARD(ING_ROUTER_CORRUPTED_IP_HDR, L3_DISCARDS), - MLXSW_SP_RXL_DISCARD(ING_ROUTER_IPV4_SIP_BC, L3_DISCARDS), - MLXSW_SP_RXL_DISCARD(IPV6_MC_DIP_RESERVED_SCOPE, L3_DISCARDS), - MLXSW_SP_RXL_DISCARD(IPV6_MC_DIP_INTERFACE_LOCAL_SCOPE, L3_DISCARDS), - MLXSW_SP_RXL_EXCEPTION(MTUERROR, L3_DISCARDS, TRAP_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(TTLERROR, L3_DISCARDS, TRAP_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(RPF, L3_DISCARDS, TRAP_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(RTR_INGRESS1, L3_DISCARDS, TRAP_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(HOST_MISS_IPV4, L3_DISCARDS, TRAP_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(HOST_MISS_IPV6, L3_DISCARDS, TRAP_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER3, L3_DISCARDS, - TRAP_EXCEPTION_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER_LPM4, L3_DISCARDS, - TRAP_EXCEPTION_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(DISCARD_ROUTER_LPM6, L3_DISCARDS, - TRAP_EXCEPTION_TO_CPU), - MLXSW_SP_RXL_DISCARD(ROUTER_IRIF_EN, L3_DISCARDS), - MLXSW_SP_RXL_DISCARD(ROUTER_ERIF_EN, L3_DISCARDS), - MLXSW_SP_RXL_DISCARD(NON_ROUTABLE, L3_DISCARDS), - MLXSW_SP_RXL_EXCEPTION(DECAP_ECN0, TUNNEL_DISCARDS, - TRAP_EXCEPTION_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(IPIP_DECAP_ERROR, TUNNEL_DISCARDS, - TRAP_EXCEPTION_TO_CPU), - MLXSW_SP_RXL_EXCEPTION(DISCARD_DEC_PKT, TUNNEL_DISCARDS, - TRAP_EXCEPTION_TO_CPU), - MLXSW_SP_RXL_DISCARD(OVERLAY_SMAC_MC, TUNNEL_DISCARDS), - MLXSW_SP_RXL_ACL_DISCARD(INGRESS_ACL, ACL_DISCARDS, DUMMY), - MLXSW_SP_RXL_ACL_DISCARD(EGRESS_ACL, ACL_DISCARDS, DUMMY), -}; +static struct mlxsw_sp_trap_policer_item * +mlxsw_sp_trap_policer_item_lookup(struct mlxsw_sp *mlxsw_sp, u32 id) +{ + struct mlxsw_sp_trap *trap = mlxsw_sp->trap; + int i; -/* Mapping between hardware trap and devlink trap. Multiple hardware traps can - * be mapped to the same devlink trap. Order is according to - * 'mlxsw_sp_listeners_arr'. - */ -static const u16 mlxsw_sp_listener_devlink_map[] = { - DEVLINK_TRAP_GENERIC_ID_SMAC_MC, - DEVLINK_TRAP_GENERIC_ID_VLAN_TAG_MISMATCH, - DEVLINK_TRAP_GENERIC_ID_INGRESS_VLAN_FILTER, - DEVLINK_TRAP_GENERIC_ID_INGRESS_STP_FILTER, - DEVLINK_TRAP_GENERIC_ID_EMPTY_TX_LIST, - DEVLINK_TRAP_GENERIC_ID_EMPTY_TX_LIST, - DEVLINK_TRAP_GENERIC_ID_PORT_LOOPBACK_FILTER, - DEVLINK_TRAP_GENERIC_ID_BLACKHOLE_ROUTE, - DEVLINK_TRAP_GENERIC_ID_NON_IP_PACKET, - DEVLINK_TRAP_GENERIC_ID_UC_DIP_MC_DMAC, - DEVLINK_TRAP_GENERIC_ID_DIP_LB, - DEVLINK_TRAP_GENERIC_ID_SIP_MC, - DEVLINK_TRAP_GENERIC_ID_SIP_LB, - DEVLINK_TRAP_GENERIC_ID_CORRUPTED_IP_HDR, - DEVLINK_TRAP_GENERIC_ID_IPV4_SIP_BC, - DEVLINK_TRAP_GENERIC_ID_IPV6_MC_DIP_RESERVED_SCOPE, - DEVLINK_TRAP_GENERIC_ID_IPV6_MC_DIP_INTERFACE_LOCAL_SCOPE, - DEVLINK_TRAP_GENERIC_ID_MTU_ERROR, - DEVLINK_TRAP_GENERIC_ID_TTL_ERROR, - DEVLINK_TRAP_GENERIC_ID_RPF, - DEVLINK_TRAP_GENERIC_ID_REJECT_ROUTE, - DEVLINK_TRAP_GENERIC_ID_UNRESOLVED_NEIGH, - DEVLINK_TRAP_GENERIC_ID_UNRESOLVED_NEIGH, - DEVLINK_TRAP_GENERIC_ID_UNRESOLVED_NEIGH, - DEVLINK_TRAP_GENERIC_ID_IPV4_LPM_UNICAST_MISS, - DEVLINK_TRAP_GENERIC_ID_IPV6_LPM_UNICAST_MISS, - DEVLINK_MLXSW_TRAP_ID_IRIF_DISABLED, - DEVLINK_MLXSW_TRAP_ID_ERIF_DISABLED, - DEVLINK_TRAP_GENERIC_ID_NON_ROUTABLE, - DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR, - DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR, - DEVLINK_TRAP_GENERIC_ID_DECAP_ERROR, - DEVLINK_TRAP_GENERIC_ID_OVERLAY_SMAC_MC, - DEVLINK_TRAP_GENERIC_ID_INGRESS_FLOW_ACTION_DROP, - DEVLINK_TRAP_GENERIC_ID_EGRESS_FLOW_ACTION_DROP, -}; + for (i = 0; i < trap->policers_count; i++) { + if (trap->policer_items_arr[i].policer.id == id) + return &trap->policer_items_arr[i]; + } -#define MLXSW_SP_THIN_POLICER_ID (MLXSW_REG_HTGT_TRAP_GROUP_MAX + 1) + return NULL; +} -static struct mlxsw_sp_trap_policer_item * -mlxsw_sp_trap_policer_item_lookup(struct mlxsw_sp *mlxsw_sp, u32 id) +static struct mlxsw_sp_trap_group_item * +mlxsw_sp_trap_group_item_lookup(struct mlxsw_sp *mlxsw_sp, u16 id) +{ + struct mlxsw_sp_trap *trap = mlxsw_sp->trap; + int i; + + for (i = 0; i < trap->groups_count; i++) { + if (trap->group_items_arr[i].group.id == id) + return &trap->group_items_arr[i]; + } + + return NULL; +} + +static struct mlxsw_sp_trap_item * +mlxsw_sp_trap_item_lookup(struct mlxsw_sp *mlxsw_sp, u16 id) { - struct mlxsw_sp_trap_policer_item *policer_item; struct mlxsw_sp_trap *trap = mlxsw_sp->trap; + int i; - list_for_each_entry(policer_item, &trap->policer_item_list, list) { - if (policer_item->id == id) - return policer_item; + for (i = 0; i < trap->traps_count; i++) { + if (trap->trap_items_arr[i].trap.id == id) + return &trap->trap_items_arr[i]; } return NULL; @@ -326,14 +1026,21 @@ mlxsw_sp_trap_policer_item_lookup(struct mlxsw_sp *mlxsw_sp, u32 id) static int mlxsw_sp_trap_cpu_policers_set(struct mlxsw_sp *mlxsw_sp) { + struct mlxsw_sp_trap *trap = mlxsw_sp->trap; char qpcr_pl[MLXSW_REG_QPCR_LEN]; + u16 hw_id; /* The purpose of "thin" policer is to drop as many packets * as possible. The dummy group is using it. */ - __set_bit(MLXSW_SP_THIN_POLICER_ID, mlxsw_sp->trap->policers_usage); - mlxsw_reg_qpcr_pack(qpcr_pl, MLXSW_SP_THIN_POLICER_ID, - MLXSW_REG_QPCR_IR_UNITS_M, false, 1, 4); + hw_id = find_first_zero_bit(trap->policers_usage, trap->max_policers); + if (WARN_ON(hw_id == trap->max_policers)) + return -ENOBUFS; + + __set_bit(hw_id, trap->policers_usage); + trap->thin_policer_hw_id = hw_id; + mlxsw_reg_qpcr_pack(qpcr_pl, hw_id, MLXSW_REG_QPCR_IR_UNITS_M, + false, 1, 4); return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(qpcr), qpcr_pl); } @@ -342,82 +1049,214 @@ static int mlxsw_sp_trap_dummy_group_init(struct mlxsw_sp *mlxsw_sp) char htgt_pl[MLXSW_REG_HTGT_LEN]; mlxsw_reg_htgt_pack(htgt_pl, MLXSW_REG_HTGT_TRAP_GROUP_SP_DUMMY, - MLXSW_SP_THIN_POLICER_ID, 0, 1); + mlxsw_sp->trap->thin_policer_hw_id, 0, 1); return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(htgt), htgt_pl); } -static int mlxsw_sp_trap_policers_init(struct mlxsw_sp *mlxsw_sp) +static int mlxsw_sp_trap_policer_items_arr_init(struct mlxsw_sp *mlxsw_sp) { - struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); + size_t elem_size = sizeof(struct mlxsw_sp_trap_policer_item); + u64 arr_size = ARRAY_SIZE(mlxsw_sp_trap_policer_items_arr); struct mlxsw_sp_trap *trap = mlxsw_sp->trap; u64 free_policers = 0; - u32 last_id = 0; - int err, i; + u32 last_id; + int i; for_each_clear_bit(i, trap->policers_usage, trap->max_policers) free_policers++; - if (ARRAY_SIZE(mlxsw_sp_trap_policers_arr) > free_policers) { + if (arr_size > free_policers) { dev_err(mlxsw_sp->bus_info->dev, "Exceeded number of supported packet trap policers\n"); return -ENOBUFS; } - trap->policers_arr = kcalloc(free_policers, - sizeof(struct devlink_trap_policer), - GFP_KERNEL); - if (!trap->policers_arr) + trap->policer_items_arr = kcalloc(free_policers, elem_size, GFP_KERNEL); + if (!trap->policer_items_arr) return -ENOMEM; trap->policers_count = free_policers; - for (i = 0; i < free_policers; i++) { - const struct devlink_trap_policer *policer; - - if (i < ARRAY_SIZE(mlxsw_sp_trap_policers_arr)) { - policer = &mlxsw_sp_trap_policers_arr[i]; - trap->policers_arr[i] = *policer; - last_id = policer->id; - } else { - /* Use parameters set for first policer and override - * relevant ones. - */ - policer = &mlxsw_sp_trap_policers_arr[0]; - trap->policers_arr[i] = *policer; - trap->policers_arr[i].id = ++last_id; - trap->policers_arr[i].init_rate = 1; - trap->policers_arr[i].init_burst = 16; - } + /* Initialize policer items array with pre-defined policers. */ + memcpy(trap->policer_items_arr, mlxsw_sp_trap_policer_items_arr, + elem_size * arr_size); + + /* Initialize policer items array with the rest of the available + * policers. + */ + last_id = mlxsw_sp_trap_policer_items_arr[arr_size - 1].policer.id; + for (i = arr_size; i < trap->policers_count; i++) { + const struct mlxsw_sp_trap_policer_item *policer_item; + + /* Use parameters set for first policer and override + * relevant ones. + */ + policer_item = &mlxsw_sp_trap_policer_items_arr[0]; + trap->policer_items_arr[i] = *policer_item; + trap->policer_items_arr[i].policer.id = ++last_id; + trap->policer_items_arr[i].policer.init_rate = 1; + trap->policer_items_arr[i].policer.init_burst = 16; } - INIT_LIST_HEAD(&trap->policer_item_list); + return 0; +} + +static void mlxsw_sp_trap_policer_items_arr_fini(struct mlxsw_sp *mlxsw_sp) +{ + kfree(mlxsw_sp->trap->policer_items_arr); +} + +static int mlxsw_sp_trap_policers_init(struct mlxsw_sp *mlxsw_sp) +{ + struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); + const struct mlxsw_sp_trap_policer_item *policer_item; + struct mlxsw_sp_trap *trap = mlxsw_sp->trap; + int err, i; - err = devlink_trap_policers_register(devlink, trap->policers_arr, - trap->policers_count); + err = mlxsw_sp_trap_policer_items_arr_init(mlxsw_sp); if (err) - goto err_trap_policers_register; + return err; + + for (i = 0; i < trap->policers_count; i++) { + policer_item = &trap->policer_items_arr[i]; + err = devlink_trap_policers_register(devlink, + &policer_item->policer, 1); + if (err) + goto err_trap_policer_register; + } return 0; -err_trap_policers_register: - kfree(trap->policers_arr); +err_trap_policer_register: + for (i--; i >= 0; i--) { + policer_item = &trap->policer_items_arr[i]; + devlink_trap_policers_unregister(devlink, + &policer_item->policer, 1); + } + mlxsw_sp_trap_policer_items_arr_fini(mlxsw_sp); return err; } static void mlxsw_sp_trap_policers_fini(struct mlxsw_sp *mlxsw_sp) { struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); + const struct mlxsw_sp_trap_policer_item *policer_item; struct mlxsw_sp_trap *trap = mlxsw_sp->trap; + int i; - devlink_trap_policers_unregister(devlink, trap->policers_arr, - trap->policers_count); - WARN_ON(!list_empty(&trap->policer_item_list)); - kfree(trap->policers_arr); + for (i = trap->policers_count - 1; i >= 0; i--) { + policer_item = &trap->policer_items_arr[i]; + devlink_trap_policers_unregister(devlink, + &policer_item->policer, 1); + } + mlxsw_sp_trap_policer_items_arr_fini(mlxsw_sp); } -int mlxsw_sp_devlink_traps_init(struct mlxsw_sp *mlxsw_sp) +static int mlxsw_sp_trap_groups_init(struct mlxsw_sp *mlxsw_sp) { - size_t groups_count = ARRAY_SIZE(mlxsw_sp_trap_groups_arr); struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); + const struct mlxsw_sp_trap_group_item *group_item; + struct mlxsw_sp_trap *trap = mlxsw_sp->trap; + int err, i; + + trap->group_items_arr = kmemdup(mlxsw_sp_trap_group_items_arr, + sizeof(mlxsw_sp_trap_group_items_arr), + GFP_KERNEL); + if (!trap->group_items_arr) + return -ENOMEM; + + trap->groups_count = ARRAY_SIZE(mlxsw_sp_trap_group_items_arr); + + for (i = 0; i < trap->groups_count; i++) { + group_item = &trap->group_items_arr[i]; + err = devlink_trap_groups_register(devlink, &group_item->group, + 1); + if (err) + goto err_trap_group_register; + } + + return 0; + +err_trap_group_register: + for (i--; i >= 0; i--) { + group_item = &trap->group_items_arr[i]; + devlink_trap_groups_unregister(devlink, &group_item->group, 1); + } + kfree(trap->group_items_arr); + return err; +} + +static void mlxsw_sp_trap_groups_fini(struct mlxsw_sp *mlxsw_sp) +{ + struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); + struct mlxsw_sp_trap *trap = mlxsw_sp->trap; + int i; + + for (i = trap->groups_count - 1; i >= 0; i--) { + const struct mlxsw_sp_trap_group_item *group_item; + + group_item = &trap->group_items_arr[i]; + devlink_trap_groups_unregister(devlink, &group_item->group, 1); + } + kfree(trap->group_items_arr); +} + +static bool +mlxsw_sp_trap_listener_is_valid(const struct mlxsw_listener *listener) +{ + return listener->trap_id != 0; +} + +static int mlxsw_sp_traps_init(struct mlxsw_sp *mlxsw_sp) +{ + struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); + struct mlxsw_sp_trap *trap = mlxsw_sp->trap; + const struct mlxsw_sp_trap_item *trap_item; + int err, i; + + trap->trap_items_arr = kmemdup(mlxsw_sp_trap_items_arr, + sizeof(mlxsw_sp_trap_items_arr), + GFP_KERNEL); + if (!trap->trap_items_arr) + return -ENOMEM; + + trap->traps_count = ARRAY_SIZE(mlxsw_sp_trap_items_arr); + + for (i = 0; i < trap->traps_count; i++) { + trap_item = &trap->trap_items_arr[i]; + err = devlink_traps_register(devlink, &trap_item->trap, 1, + mlxsw_sp); + if (err) + goto err_trap_register; + } + + return 0; + +err_trap_register: + for (i--; i >= 0; i--) { + trap_item = &trap->trap_items_arr[i]; + devlink_traps_unregister(devlink, &trap_item->trap, 1); + } + kfree(trap->trap_items_arr); + return err; +} + +static void mlxsw_sp_traps_fini(struct mlxsw_sp *mlxsw_sp) +{ + struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); + struct mlxsw_sp_trap *trap = mlxsw_sp->trap; + int i; + + for (i = trap->traps_count - 1; i >= 0; i--) { + const struct mlxsw_sp_trap_item *trap_item; + + trap_item = &trap->trap_items_arr[i]; + devlink_traps_unregister(devlink, &trap_item->trap, 1); + } + kfree(trap->trap_items_arr); +} + +int mlxsw_sp_devlink_traps_init(struct mlxsw_sp *mlxsw_sp) +{ int err; err = mlxsw_sp_trap_cpu_policers_set(mlxsw_sp); @@ -428,59 +1267,52 @@ int mlxsw_sp_devlink_traps_init(struct mlxsw_sp *mlxsw_sp) if (err) return err; - if (WARN_ON(ARRAY_SIZE(mlxsw_sp_listener_devlink_map) != - ARRAY_SIZE(mlxsw_sp_listeners_arr))) - return -EINVAL; - err = mlxsw_sp_trap_policers_init(mlxsw_sp); if (err) return err; - err = devlink_trap_groups_register(devlink, mlxsw_sp_trap_groups_arr, - groups_count); + err = mlxsw_sp_trap_groups_init(mlxsw_sp); if (err) - goto err_trap_groups_register; + goto err_trap_groups_init; - err = devlink_traps_register(devlink, mlxsw_sp_traps_arr, - ARRAY_SIZE(mlxsw_sp_traps_arr), mlxsw_sp); + err = mlxsw_sp_traps_init(mlxsw_sp); if (err) - goto err_traps_register; + goto err_traps_init; return 0; -err_traps_register: - devlink_trap_groups_unregister(devlink, mlxsw_sp_trap_groups_arr, - groups_count); -err_trap_groups_register: +err_traps_init: + mlxsw_sp_trap_groups_fini(mlxsw_sp); +err_trap_groups_init: mlxsw_sp_trap_policers_fini(mlxsw_sp); return err; } void mlxsw_sp_devlink_traps_fini(struct mlxsw_sp *mlxsw_sp) { - size_t groups_count = ARRAY_SIZE(mlxsw_sp_trap_groups_arr); - struct devlink *devlink = priv_to_devlink(mlxsw_sp->core); - - devlink_traps_unregister(devlink, mlxsw_sp_traps_arr, - ARRAY_SIZE(mlxsw_sp_traps_arr)); - devlink_trap_groups_unregister(devlink, mlxsw_sp_trap_groups_arr, - groups_count); + mlxsw_sp_traps_fini(mlxsw_sp); + mlxsw_sp_trap_groups_fini(mlxsw_sp); mlxsw_sp_trap_policers_fini(mlxsw_sp); } int mlxsw_sp_trap_init(struct mlxsw_core *mlxsw_core, const struct devlink_trap *trap, void *trap_ctx) { + struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); + const struct mlxsw_sp_trap_item *trap_item; int i; - for (i = 0; i < ARRAY_SIZE(mlxsw_sp_listener_devlink_map); i++) { + trap_item = mlxsw_sp_trap_item_lookup(mlxsw_sp, trap->id); + if (WARN_ON(!trap_item)) + return -EINVAL; + + for (i = 0; i < MLXSW_SP_TRAP_LISTENERS_MAX; i++) { const struct mlxsw_listener *listener; int err; - if (mlxsw_sp_listener_devlink_map[i] != trap->id) + listener = &trap_item->listeners_arr[i]; + if (!mlxsw_sp_trap_listener_is_valid(listener)) continue; - listener = &mlxsw_sp_listeners_arr[i]; - err = mlxsw_core_trap_register(mlxsw_core, listener, trap_ctx); if (err) return err; @@ -492,15 +1324,20 @@ int mlxsw_sp_trap_init(struct mlxsw_core *mlxsw_core, void mlxsw_sp_trap_fini(struct mlxsw_core *mlxsw_core, const struct devlink_trap *trap, void *trap_ctx) { + struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); + const struct mlxsw_sp_trap_item *trap_item; int i; - for (i = 0; i < ARRAY_SIZE(mlxsw_sp_listener_devlink_map); i++) { + trap_item = mlxsw_sp_trap_item_lookup(mlxsw_sp, trap->id); + if (WARN_ON(!trap_item)) + return; + + for (i = MLXSW_SP_TRAP_LISTENERS_MAX - 1; i >= 0; i--) { const struct mlxsw_listener *listener; - if (mlxsw_sp_listener_devlink_map[i] != trap->id) + listener = &trap_item->listeners_arr[i]; + if (!mlxsw_sp_trap_listener_is_valid(listener)) continue; - listener = &mlxsw_sp_listeners_arr[i]; - mlxsw_core_trap_unregister(mlxsw_core, listener, trap_ctx); } } @@ -509,16 +1346,23 @@ int mlxsw_sp_trap_action_set(struct mlxsw_core *mlxsw_core, const struct devlink_trap *trap, enum devlink_trap_action action) { + struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); + const struct mlxsw_sp_trap_item *trap_item; int i; - for (i = 0; i < ARRAY_SIZE(mlxsw_sp_listener_devlink_map); i++) { + trap_item = mlxsw_sp_trap_item_lookup(mlxsw_sp, trap->id); + if (WARN_ON(!trap_item)) + return -EINVAL; + + for (i = 0; i < MLXSW_SP_TRAP_LISTENERS_MAX; i++) { const struct mlxsw_listener *listener; bool enabled; int err; - if (mlxsw_sp_listener_devlink_map[i] != trap->id) + listener = &trap_item->listeners_arr[i]; + if (!mlxsw_sp_trap_listener_is_valid(listener)) continue; - listener = &mlxsw_sp_listeners_arr[i]; + switch (action) { case DEVLINK_TRAP_ACTION_DROP: enabled = false; @@ -544,33 +1388,12 @@ __mlxsw_sp_trap_group_init(struct mlxsw_core *mlxsw_core, { struct mlxsw_sp *mlxsw_sp = mlxsw_core_driver_priv(mlxsw_core); u16 hw_policer_id = MLXSW_REG_HTGT_INVALID_POLICER; + const struct mlxsw_sp_trap_group_item *group_item; char htgt_pl[MLXSW_REG_HTGT_LEN]; - u8 priority, tc, group_id; - - switch (group->id) { - case DEVLINK_TRAP_GROUP_GENERIC_ID_L2_DROPS: - group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_L2_DISCARDS; - priority = 0; - tc = 1; - break; - case DEVLINK_TRAP_GROUP_GENERIC_ID_L3_DROPS: - group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_L3_DISCARDS; - priority = 0; - tc = 1; - break; - case DEVLINK_TRAP_GROUP_GENERIC_ID_TUNNEL_DROPS: - group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_TUNNEL_DISCARDS; - priority = 0; - tc = 1; - break; - case DEVLINK_TRAP_GROUP_GENERIC_ID_ACL_DROPS: - group_id = MLXSW_REG_HTGT_TRAP_GROUP_SP_ACL_DISCARDS; - priority = 0; - tc = 1; - break; - default: + + group_item = mlxsw_sp_trap_group_item_lookup(mlxsw_sp, group->id); + if (WARN_ON(!group_item)) return -EINVAL; - } if (policer_id) { struct mlxsw_sp_trap_policer_item *policer_item; @@ -582,7 +1405,8 @@ __mlxsw_sp_trap_group_init(struct mlxsw_core *mlxsw_core, hw_policer_id = policer_item->hw_id; } - mlxsw_reg_htgt_pack(htgt_pl, group_id, hw_policer_id, priority, tc); + mlxsw_reg_htgt_pack(htgt_pl, group_item->hw_group_id, hw_policer_id, + group_item->priority, group_item->priority); return mlxsw_reg_write(mlxsw_core, MLXSW_REG(htgt), htgt_pl); } @@ -602,10 +1426,10 @@ int mlxsw_sp_trap_group_set(struct mlxsw_core *mlxsw_core, return __mlxsw_sp_trap_group_init(mlxsw_core, group, policer_id); } -static struct mlxsw_sp_trap_policer_item * -mlxsw_sp_trap_policer_item_init(struct mlxsw_sp *mlxsw_sp, u32 id) +static int +mlxsw_sp_trap_policer_item_init(struct mlxsw_sp *mlxsw_sp, + struct mlxsw_sp_trap_policer_item *policer_item) { - struct mlxsw_sp_trap_policer_item *policer_item; struct mlxsw_sp_trap *trap = mlxsw_sp->trap; u16 hw_id; @@ -615,27 +1439,19 @@ mlxsw_sp_trap_policer_item_init(struct mlxsw_sp *mlxsw_sp, u32 id) */ hw_id = find_first_zero_bit(trap->policers_usage, trap->max_policers); if (WARN_ON(hw_id == trap->max_policers)) - return ERR_PTR(-ENOBUFS); - - policer_item = kzalloc(sizeof(*policer_item), GFP_KERNEL); - if (!policer_item) - return ERR_PTR(-ENOMEM); + return -ENOBUFS; __set_bit(hw_id, trap->policers_usage); policer_item->hw_id = hw_id; - policer_item->id = id; - list_add_tail(&policer_item->list, &trap->policer_item_list); - return policer_item; + return 0; } static void mlxsw_sp_trap_policer_item_fini(struct mlxsw_sp *mlxsw_sp, struct mlxsw_sp_trap_policer_item *policer_item) { - list_del(&policer_item->list); __clear_bit(policer_item->hw_id, mlxsw_sp->trap->policers_usage); - kfree(policer_item); } static int mlxsw_sp_trap_policer_bs(u64 burst, u8 *p_burst_size, @@ -678,9 +1494,13 @@ int mlxsw_sp_trap_policer_init(struct mlxsw_core *mlxsw_core, struct mlxsw_sp_trap_policer_item *policer_item; int err; - policer_item = mlxsw_sp_trap_policer_item_init(mlxsw_sp, policer->id); - if (IS_ERR(policer_item)) - return PTR_ERR(policer_item); + policer_item = mlxsw_sp_trap_policer_item_lookup(mlxsw_sp, policer->id); + if (WARN_ON(!policer_item)) + return -EINVAL; + + err = mlxsw_sp_trap_policer_item_init(mlxsw_sp, policer_item); + if (err) + return err; err = __mlxsw_sp_trap_policer_set(mlxsw_sp, policer_item->hw_id, policer->init_rate, diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.h index 8c54897ba173..13ac412f4d53 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.h +++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_trap.h @@ -8,17 +8,19 @@ #include <net/devlink.h> struct mlxsw_sp_trap { - struct devlink_trap_policer *policers_arr; /* Registered policers */ + struct mlxsw_sp_trap_policer_item *policer_items_arr; u64 policers_count; /* Number of registered policers */ - struct list_head policer_item_list; + + struct mlxsw_sp_trap_group_item *group_items_arr; + u64 groups_count; /* Number of registered groups */ + + struct mlxsw_sp_trap_item *trap_items_arr; + u64 traps_count; /* Number of registered traps */ + + u16 thin_policer_hw_id; + u64 max_policers; unsigned long policers_usage[]; /* Usage bitmap */ }; -struct mlxsw_sp_trap_policer_item { - u16 hw_id; - u32 id; - struct list_head list; /* Member of policer_item_list */ -}; - #endif diff --git a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c index 90535820b559..b438f5576e18 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/switchx2.c +++ b/drivers/net/ethernet/mellanox/mlxsw/switchx2.c @@ -1259,6 +1259,7 @@ static void mlxsw_sx_ports_remove(struct mlxsw_sx *mlxsw_sx) if (mlxsw_sx_port_created(mlxsw_sx, i)) mlxsw_sx_port_remove(mlxsw_sx, i); kfree(mlxsw_sx->ports); + mlxsw_sx->ports = NULL; } static int mlxsw_sx_ports_create(struct mlxsw_sx *mlxsw_sx) @@ -1293,6 +1294,7 @@ err_port_module_info_get: if (mlxsw_sx_port_created(mlxsw_sx, i)) mlxsw_sx_port_remove(mlxsw_sx, i); kfree(mlxsw_sx->ports); + mlxsw_sx->ports = NULL; return err; } @@ -1376,6 +1378,12 @@ static int mlxsw_sx_port_type_set(struct mlxsw_core *mlxsw_core, u8 local_port, u8 module, width; int err; + if (!mlxsw_sx->ports || !mlxsw_sx->ports[local_port]) { + dev_err(mlxsw_sx->bus_info->dev, "Port number \"%d\" does not exist\n", + local_port); + return -EINVAL; + } + if (new_type == DEVLINK_PORT_TYPE_AUTO) return -EOPNOTSUPP; @@ -1396,6 +1404,11 @@ err_port_module_info_get: return err; } +enum { + MLXSW_REG_HTGT_TRAP_GROUP_SX2_RX = 1, + MLXSW_REG_HTGT_TRAP_GROUP_SX2_CTRL = 2, +}; + #define MLXSW_SX_RXL(_trap_id) \ MLXSW_RXL(mlxsw_sx_rx_listener_func, _trap_id, TRAP_TO_CPU, \ false, SX2_RX, FORWARD) diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h index eaa521b7561b..28e60697d14e 100644 --- a/drivers/net/ethernet/mellanox/mlxsw/trap.h +++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h @@ -55,16 +55,19 @@ enum { MLXSW_TRAP_ID_IPV4_BGP = 0x88, MLXSW_TRAP_ID_IPV6_BGP = 0x89, MLXSW_TRAP_ID_L3_IPV6_ROUTER_SOLICITATION = 0x8A, - MLXSW_TRAP_ID_L3_IPV6_ROUTER_ADVERTISMENT = 0x8B, + MLXSW_TRAP_ID_L3_IPV6_ROUTER_ADVERTISEMENT = 0x8B, MLXSW_TRAP_ID_L3_IPV6_NEIGHBOR_SOLICITATION = 0x8C, - MLXSW_TRAP_ID_L3_IPV6_NEIGHBOR_ADVERTISMENT = 0x8D, + MLXSW_TRAP_ID_L3_IPV6_NEIGHBOR_ADVERTISEMENT = 0x8D, MLXSW_TRAP_ID_L3_IPV6_REDIRECTION = 0x8E, + MLXSW_TRAP_ID_IPV4_DHCP = 0x8F, MLXSW_TRAP_ID_HOST_MISS_IPV4 = 0x90, MLXSW_TRAP_ID_IPV6_MC_LINK_LOCAL_DEST = 0x91, MLXSW_TRAP_ID_HOST_MISS_IPV6 = 0x92, MLXSW_TRAP_ID_IPIP_DECAP_ERROR = 0xB1, MLXSW_TRAP_ID_NVE_DECAP_ARP = 0xB8, MLXSW_TRAP_ID_NVE_ENCAP_ARP = 0xBD, + MLXSW_TRAP_ID_IPV4_BFD = 0xD0, + MLXSW_TRAP_ID_IPV6_BFD = 0xD1, MLXSW_TRAP_ID_ROUTER_ALERT_IPV4 = 0xD6, MLXSW_TRAP_ID_ROUTER_ALERT_IPV6 = 0xD7, MLXSW_TRAP_ID_DISCARD_NON_ROUTABLE = 0x11A, diff --git a/drivers/net/ethernet/micrel/Kconfig b/drivers/net/ethernet/micrel/Kconfig index b9c4d48e28e4..09f35209d43d 100644 --- a/drivers/net/ethernet/micrel/Kconfig +++ b/drivers/net/ethernet/micrel/Kconfig @@ -38,6 +38,8 @@ config KS8851_MLL tristate "Micrel KS8851 MLL" depends on HAS_IOMEM select MII + select CRC32 + select EEPROM_93CX6 ---help--- This platform driver is for Micrel KS8851 Address/data bus multiplexed network chip. diff --git a/drivers/net/ethernet/micrel/Makefile b/drivers/net/ethernet/micrel/Makefile index 6d8ac5527aef..5cc00d22c708 100644 --- a/drivers/net/ethernet/micrel/Makefile +++ b/drivers/net/ethernet/micrel/Makefile @@ -5,5 +5,7 @@ obj-$(CONFIG_KS8842) += ks8842.o obj-$(CONFIG_KS8851) += ks8851.o +ks8851-objs = ks8851_common.o ks8851_spi.o obj-$(CONFIG_KS8851_MLL) += ks8851_mll.o +ks8851_mll-objs = ks8851_common.o ks8851_par.o obj-$(CONFIG_KSZ884X_PCI) += ksz884x.o diff --git a/drivers/net/ethernet/micrel/ks8851.h b/drivers/net/ethernet/micrel/ks8851.h index 8f834aef8e32..2b319e451121 100644 --- a/drivers/net/ethernet/micrel/ks8851.h +++ b/drivers/net/ethernet/micrel/ks8851.h @@ -7,6 +7,11 @@ * KS8851 register definitions */ +#ifndef __KS8851_H__ +#define __KS8851_H__ + +#include <linux/eeprom_93cx6.h> + #define KS_CCR 0x08 #define CCR_LE (1 << 10) /* KSZ8851-16MLL */ #define CCR_EEPROM (1 << 9) @@ -19,7 +24,7 @@ #define CCR_32PIN (1 << 0) /* KSZ8851SNL */ /* MAC address registers */ -#define KS_MAR(_m) (0x15 - (_m)) +#define KS_MAR(_m) (0x14 - (_m)) #define KS_MARL 0x10 #define KS_MARM 0x12 #define KS_MARH 0x14 @@ -300,3 +305,147 @@ #define TXFR_TXIC (1 << 15) #define TXFR_TXFID_MASK (0x3f << 0) #define TXFR_TXFID_SHIFT (0) + +/** + * struct ks8851_rxctrl - KS8851 driver rx control + * @mchash: Multicast hash-table data. + * @rxcr1: KS_RXCR1 register setting + * @rxcr2: KS_RXCR2 register setting + * + * Representation of the settings needs to control the receive filtering + * such as the multicast hash-filter and the receive register settings. This + * is used to make the job of working out if the receive settings change and + * then issuing the new settings to the worker that will send the necessary + * commands. + */ +struct ks8851_rxctrl { + u16 mchash[4]; + u16 rxcr1; + u16 rxcr2; +}; + +/** + * union ks8851_tx_hdr - tx header data + * @txb: The header as bytes + * @txw: The header as 16bit, little-endian words + * + * A dual representation of the tx header data to allow + * access to individual bytes, and to allow 16bit accesses + * with 16bit alignment. + */ +union ks8851_tx_hdr { + u8 txb[6]; + __le16 txw[3]; +}; + +/** + * struct ks8851_net - KS8851 driver private data + * @netdev: The network device we're bound to + * @statelock: Lock on this structure for tx list. + * @mii: The MII state information for the mii calls. + * @rxctrl: RX settings for @rxctrl_work. + * @rxctrl_work: Work queue for updating RX mode and multicast lists + * @txq: Queue of packets for transmission. + * @txh: Space for generating packet TX header in DMA-able data + * @rxd: Space for receiving SPI data, in DMA-able space. + * @txd: Space for transmitting SPI data, in DMA-able space. + * @msg_enable: The message flags controlling driver output (see ethtool). + * @fid: Incrementing frame id tag. + * @rc_ier: Cached copy of KS_IER. + * @rc_ccr: Cached copy of KS_CCR. + * @rc_rxqcr: Cached copy of KS_RXQCR. + * @eeprom: 93CX6 EEPROM state for accessing on-board EEPROM. + * @vdd_reg: Optional regulator supplying the chip + * @vdd_io: Optional digital power supply for IO + * @gpio: Optional reset_n gpio + * @lock: Bus access lock callback + * @unlock: Bus access unlock callback + * @rdreg16: 16bit register read callback + * @wrreg16: 16bit register write callback + * @rdfifo: FIFO read callback + * @wrfifo: FIFO write callback + * @start_xmit: start_xmit() implementation callback + * @rx_skb: rx_skb() implementation callback + * @flush_tx_work: flush_tx_work() implementation callback + * + * The @statelock is used to protect information in the structure which may + * need to be accessed via several sources, such as the network driver layer + * or one of the work queues. + * + * We align the buffers we may use for rx/tx to ensure that if the SPI driver + * wants to DMA map them, it will not have any problems with data the driver + * modifies. + */ +struct ks8851_net { + struct net_device *netdev; + spinlock_t statelock; + + union ks8851_tx_hdr txh ____cacheline_aligned; + u8 rxd[8]; + u8 txd[8]; + + u32 msg_enable ____cacheline_aligned; + u16 tx_space; + u8 fid; + + u16 rc_ier; + u16 rc_rxqcr; + u16 rc_ccr; + + struct mii_if_info mii; + struct ks8851_rxctrl rxctrl; + + struct work_struct rxctrl_work; + + struct sk_buff_head txq; + + struct eeprom_93cx6 eeprom; + struct regulator *vdd_reg; + struct regulator *vdd_io; + int gpio; + + void (*lock)(struct ks8851_net *ks, + unsigned long *flags); + void (*unlock)(struct ks8851_net *ks, + unsigned long *flags); + unsigned int (*rdreg16)(struct ks8851_net *ks, + unsigned int reg); + void (*wrreg16)(struct ks8851_net *ks, + unsigned int reg, unsigned int val); + void (*rdfifo)(struct ks8851_net *ks, u8 *buff, + unsigned int len); + void (*wrfifo)(struct ks8851_net *ks, + struct sk_buff *txp, bool irq); + netdev_tx_t (*start_xmit)(struct sk_buff *skb, + struct net_device *dev); + void (*rx_skb)(struct ks8851_net *ks, + struct sk_buff *skb); + void (*flush_tx_work)(struct ks8851_net *ks); +}; + +int ks8851_probe_common(struct net_device *netdev, struct device *dev, + int msg_en); +int ks8851_remove_common(struct device *dev); +int ks8851_suspend(struct device *dev); +int ks8851_resume(struct device *dev); + +static __maybe_unused SIMPLE_DEV_PM_OPS(ks8851_pm_ops, + ks8851_suspend, ks8851_resume); + +/** + * ks8851_done_tx - update and then free skbuff after transmitting + * @ks: The device state + * @txb: The buffer transmitted + */ +static void __maybe_unused ks8851_done_tx(struct ks8851_net *ks, + struct sk_buff *txb) +{ + struct net_device *dev = ks->netdev; + + dev->stats.tx_bytes += txb->len; + dev->stats.tx_packets++; + + dev_kfree_skb(txb); +} + +#endif /* __KS8851_H__ */ diff --git a/drivers/net/ethernet/micrel/ks8851.c b/drivers/net/ethernet/micrel/ks8851_common.c index 33305c9c5a62..d65872172229 100644 --- a/drivers/net/ethernet/micrel/ks8851.c +++ b/drivers/net/ethernet/micrel/ks8851_common.c @@ -19,10 +19,8 @@ #include <linux/cache.h> #include <linux/crc32.h> #include <linux/mii.h> -#include <linux/eeprom_93cx6.h> #include <linux/regulator/consumer.h> -#include <linux/spi/spi.h> #include <linux/gpio.h> #include <linux/of_gpio.h> #include <linux/of_net.h> @@ -30,255 +28,41 @@ #include "ks8851.h" /** - * struct ks8851_rxctrl - KS8851 driver rx control - * @mchash: Multicast hash-table data. - * @rxcr1: KS_RXCR1 register setting - * @rxcr2: KS_RXCR2 register setting - * - * Representation of the settings needs to control the receive filtering - * such as the multicast hash-filter and the receive register settings. This - * is used to make the job of working out if the receive settings change and - * then issuing the new settings to the worker that will send the necessary - * commands. - */ -struct ks8851_rxctrl { - u16 mchash[4]; - u16 rxcr1; - u16 rxcr2; -}; - -/** - * union ks8851_tx_hdr - tx header data - * @txb: The header as bytes - * @txw: The header as 16bit, little-endian words - * - * A dual representation of the tx header data to allow - * access to individual bytes, and to allow 16bit accesses - * with 16bit alignment. - */ -union ks8851_tx_hdr { - u8 txb[6]; - __le16 txw[3]; -}; - -/** - * struct ks8851_net - KS8851 driver private data - * @netdev: The network device we're bound to - * @spidev: The spi device we're bound to. - * @lock: Lock to ensure that the device is not accessed when busy. - * @statelock: Lock on this structure for tx list. - * @mii: The MII state information for the mii calls. - * @rxctrl: RX settings for @rxctrl_work. - * @tx_work: Work queue for tx packets - * @rxctrl_work: Work queue for updating RX mode and multicast lists - * @txq: Queue of packets for transmission. - * @spi_msg1: pre-setup SPI transfer with one message, @spi_xfer1. - * @spi_msg2: pre-setup SPI transfer with two messages, @spi_xfer2. - * @txh: Space for generating packet TX header in DMA-able data - * @rxd: Space for receiving SPI data, in DMA-able space. - * @txd: Space for transmitting SPI data, in DMA-able space. - * @msg_enable: The message flags controlling driver output (see ethtool). - * @fid: Incrementing frame id tag. - * @rc_ier: Cached copy of KS_IER. - * @rc_ccr: Cached copy of KS_CCR. - * @rc_rxqcr: Cached copy of KS_RXQCR. - * @eeprom: 93CX6 EEPROM state for accessing on-board EEPROM. - * @vdd_reg: Optional regulator supplying the chip - * @vdd_io: Optional digital power supply for IO - * @gpio: Optional reset_n gpio - * - * The @lock ensures that the chip is protected when certain operations are - * in progress. When the read or write packet transfer is in progress, most - * of the chip registers are not ccessible until the transfer is finished and - * the DMA has been de-asserted. - * - * The @statelock is used to protect information in the structure which may - * need to be accessed via several sources, such as the network driver layer - * or one of the work queues. - * - * We align the buffers we may use for rx/tx to ensure that if the SPI driver - * wants to DMA map them, it will not have any problems with data the driver - * modifies. - */ -struct ks8851_net { - struct net_device *netdev; - struct spi_device *spidev; - struct mutex lock; - spinlock_t statelock; - - union ks8851_tx_hdr txh ____cacheline_aligned; - u8 rxd[8]; - u8 txd[8]; - - u32 msg_enable ____cacheline_aligned; - u16 tx_space; - u8 fid; - - u16 rc_ier; - u16 rc_rxqcr; - u16 rc_ccr; - - struct mii_if_info mii; - struct ks8851_rxctrl rxctrl; - - struct work_struct tx_work; - struct work_struct rxctrl_work; - - struct sk_buff_head txq; - - struct spi_message spi_msg1; - struct spi_message spi_msg2; - struct spi_transfer spi_xfer1; - struct spi_transfer spi_xfer2[2]; - - struct eeprom_93cx6 eeprom; - struct regulator *vdd_reg; - struct regulator *vdd_io; - int gpio; -}; - -static int msg_enable; - -/* SPI frame opcodes */ -#define KS_SPIOP_RD (0x00) -#define KS_SPIOP_WR (0x40) -#define KS_SPIOP_RXFIFO (0x80) -#define KS_SPIOP_TXFIFO (0xC0) - -/* shift for byte-enable data */ -#define BYTE_EN(_x) ((_x) << 2) - -/* turn register number and byte-enable mask into data for start of packet */ -#define MK_OP(_byteen, _reg) (BYTE_EN(_byteen) | (_reg) << (8+2) | (_reg) >> 6) - -/* SPI register read/write calls. + * ks8851_lock - register access lock + * @ks: The chip state + * @flags: Spinlock flags * - * All these calls issue SPI transactions to access the chip's registers. They - * all require that the necessary lock is held to prevent accesses when the - * chip is busy transferring packet data (RX/TX FIFO accesses). + * Claim chip register access lock */ +static void ks8851_lock(struct ks8851_net *ks, unsigned long *flags) +{ + ks->lock(ks, flags); +} /** - * ks8851_wrreg16 - write 16bit register value to chip + * ks8851_unlock - register access unlock * @ks: The chip state - * @reg: The register address - * @val: The value to write + * @flags: Spinlock flags * - * Issue a write to put the value @val into the register specified in @reg. + * Release chip register access lock */ -static void ks8851_wrreg16(struct ks8851_net *ks, unsigned reg, unsigned val) +static void ks8851_unlock(struct ks8851_net *ks, unsigned long *flags) { - struct spi_transfer *xfer = &ks->spi_xfer1; - struct spi_message *msg = &ks->spi_msg1; - __le16 txb[2]; - int ret; - - txb[0] = cpu_to_le16(MK_OP(reg & 2 ? 0xC : 0x03, reg) | KS_SPIOP_WR); - txb[1] = cpu_to_le16(val); - - xfer->tx_buf = txb; - xfer->rx_buf = NULL; - xfer->len = 4; - - ret = spi_sync(ks->spidev, msg); - if (ret < 0) - netdev_err(ks->netdev, "spi_sync() failed\n"); + ks->unlock(ks, flags); } /** - * ks8851_wrreg8 - write 8bit register value to chip + * ks8851_wrreg16 - write 16bit register value to chip * @ks: The chip state * @reg: The register address * @val: The value to write * * Issue a write to put the value @val into the register specified in @reg. */ -static void ks8851_wrreg8(struct ks8851_net *ks, unsigned reg, unsigned val) -{ - struct spi_transfer *xfer = &ks->spi_xfer1; - struct spi_message *msg = &ks->spi_msg1; - __le16 txb[2]; - int ret; - int bit; - - bit = 1 << (reg & 3); - - txb[0] = cpu_to_le16(MK_OP(bit, reg) | KS_SPIOP_WR); - txb[1] = val; - - xfer->tx_buf = txb; - xfer->rx_buf = NULL; - xfer->len = 3; - - ret = spi_sync(ks->spidev, msg); - if (ret < 0) - netdev_err(ks->netdev, "spi_sync() failed\n"); -} - -/** - * ks8851_rdreg - issue read register command and return the data - * @ks: The device state - * @op: The register address and byte enables in message format. - * @rxb: The RX buffer to return the result into - * @rxl: The length of data expected. - * - * This is the low level read call that issues the necessary spi message(s) - * to read data from the register specified in @op. - */ -static void ks8851_rdreg(struct ks8851_net *ks, unsigned op, - u8 *rxb, unsigned rxl) -{ - struct spi_transfer *xfer; - struct spi_message *msg; - __le16 *txb = (__le16 *)ks->txd; - u8 *trx = ks->rxd; - int ret; - - txb[0] = cpu_to_le16(op | KS_SPIOP_RD); - - if (ks->spidev->master->flags & SPI_MASTER_HALF_DUPLEX) { - msg = &ks->spi_msg2; - xfer = ks->spi_xfer2; - - xfer->tx_buf = txb; - xfer->rx_buf = NULL; - xfer->len = 2; - - xfer++; - xfer->tx_buf = NULL; - xfer->rx_buf = trx; - xfer->len = rxl; - } else { - msg = &ks->spi_msg1; - xfer = &ks->spi_xfer1; - - xfer->tx_buf = txb; - xfer->rx_buf = trx; - xfer->len = rxl + 2; - } - - ret = spi_sync(ks->spidev, msg); - if (ret < 0) - netdev_err(ks->netdev, "read: spi_sync() failed\n"); - else if (ks->spidev->master->flags & SPI_MASTER_HALF_DUPLEX) - memcpy(rxb, trx, rxl); - else - memcpy(rxb, trx + 2, rxl); -} - -/** - * ks8851_rdreg8 - read 8 bit register from device - * @ks: The chip information - * @reg: The register address - * - * Read a 8bit register from the chip, returning the result -*/ -static unsigned ks8851_rdreg8(struct ks8851_net *ks, unsigned reg) +static void ks8851_wrreg16(struct ks8851_net *ks, unsigned int reg, + unsigned int val) { - u8 rxb[1]; - - ks8851_rdreg(ks, MK_OP(1 << (reg & 3), reg), rxb, 1); - return rxb[0]; + ks->wrreg16(ks, reg, val); } /** @@ -287,32 +71,11 @@ static unsigned ks8851_rdreg8(struct ks8851_net *ks, unsigned reg) * @reg: The register address * * Read a 16bit register from the chip, returning the result -*/ -static unsigned ks8851_rdreg16(struct ks8851_net *ks, unsigned reg) -{ - __le16 rx = 0; - - ks8851_rdreg(ks, MK_OP(reg & 2 ? 0xC : 0x3, reg), (u8 *)&rx, 2); - return le16_to_cpu(rx); -} - -/** - * ks8851_rdreg32 - read 32 bit register from device - * @ks: The chip information - * @reg: The register address - * - * Read a 32bit register from the chip. - * - * Note, this read requires the address be aligned to 4 bytes. -*/ -static unsigned ks8851_rdreg32(struct ks8851_net *ks, unsigned reg) + */ +static unsigned int ks8851_rdreg16(struct ks8851_net *ks, + unsigned int reg) { - __le32 rx = 0; - - WARN_ON(reg & 3); - - ks8851_rdreg(ks, MK_OP(0xf, reg), (u8 *)&rx, 4); - return le32_to_cpu(rx); + return ks->rdreg16(ks, reg); } /** @@ -368,21 +131,27 @@ static void ks8851_set_powermode(struct ks8851_net *ks, unsigned pwrmode) static int ks8851_write_mac_addr(struct net_device *dev) { struct ks8851_net *ks = netdev_priv(dev); + unsigned long flags; + u16 val; int i; - mutex_lock(&ks->lock); + ks8851_lock(ks, &flags); /* * Wake up chip in case it was powered off when stopped; otherwise, * the first write to the MAC address does not take effect. */ ks8851_set_powermode(ks, PMECR_PM_NORMAL); - for (i = 0; i < ETH_ALEN; i++) - ks8851_wrreg8(ks, KS_MAR(i), dev->dev_addr[i]); + + for (i = 0; i < ETH_ALEN; i += 2) { + val = (dev->dev_addr[i] << 8) | dev->dev_addr[i + 1]; + ks8851_wrreg16(ks, KS_MAR(i), val); + } + if (!netif_running(dev)) ks8851_set_powermode(ks, PMECR_PM_SOFTDOWN); - mutex_unlock(&ks->lock); + ks8851_unlock(ks, &flags); return 0; } @@ -396,19 +165,25 @@ static int ks8851_write_mac_addr(struct net_device *dev) static void ks8851_read_mac_addr(struct net_device *dev) { struct ks8851_net *ks = netdev_priv(dev); + unsigned long flags; + u16 reg; int i; - mutex_lock(&ks->lock); + ks8851_lock(ks, &flags); - for (i = 0; i < ETH_ALEN; i++) - dev->dev_addr[i] = ks8851_rdreg8(ks, KS_MAR(i)); + for (i = 0; i < ETH_ALEN; i += 2) { + reg = ks8851_rdreg16(ks, KS_MAR(i)); + dev->dev_addr[i] = reg >> 8; + dev->dev_addr[i + 1] = reg & 0xff; + } - mutex_unlock(&ks->lock); + ks8851_unlock(ks, &flags); } /** * ks8851_init_mac - initialise the mac address * @ks: The device structure + * @np: The device node pointer * * Get or create the initial mac address for the device and then set that * into the station address register. A mac address supplied in the device @@ -416,12 +191,12 @@ static void ks8851_read_mac_addr(struct net_device *dev) * we try that. If no valid mac address is found we use eth_random_addr() * to create a new one. */ -static void ks8851_init_mac(struct ks8851_net *ks) +static void ks8851_init_mac(struct ks8851_net *ks, struct device_node *np) { struct net_device *dev = ks->netdev; const u8 *mac_addr; - mac_addr = of_get_mac_address(ks->spidev->dev.of_node); + mac_addr = of_get_mac_address(np); if (!IS_ERR(mac_addr)) { ether_addr_copy(dev->dev_addr, mac_addr); ks8851_write_mac_addr(dev); @@ -442,48 +217,12 @@ static void ks8851_init_mac(struct ks8851_net *ks) } /** - * ks8851_rdfifo - read data from the receive fifo - * @ks: The device state. - * @buff: The buffer address - * @len: The length of the data to read - * - * Issue an RXQ FIFO read command and read the @len amount of data from - * the FIFO into the buffer specified by @buff. - */ -static void ks8851_rdfifo(struct ks8851_net *ks, u8 *buff, unsigned len) -{ - struct spi_transfer *xfer = ks->spi_xfer2; - struct spi_message *msg = &ks->spi_msg2; - u8 txb[1]; - int ret; - - netif_dbg(ks, rx_status, ks->netdev, - "%s: %d@%p\n", __func__, len, buff); - - /* set the operation we're issuing */ - txb[0] = KS_SPIOP_RXFIFO; - - xfer->tx_buf = txb; - xfer->rx_buf = NULL; - xfer->len = 1; - - xfer++; - xfer->rx_buf = buff; - xfer->tx_buf = NULL; - xfer->len = len; - - ret = spi_sync(ks->spidev, msg); - if (ret < 0) - netdev_err(ks->netdev, "%s: spi_sync() failed\n", __func__); -} - -/** * ks8851_dbg_dumpkkt - dump initial packet contents to debug * @ks: The device state * @rxpkt: The data for the received packet * * Dump the initial data from the packet to dev_dbg(). -*/ + */ static void ks8851_dbg_dumpkkt(struct ks8851_net *ks, u8 *rxpkt) { netdev_dbg(ks->netdev, @@ -494,6 +233,16 @@ static void ks8851_dbg_dumpkkt(struct ks8851_net *ks, u8 *rxpkt) } /** + * ks8851_rx_skb - receive skbuff + * @ks: The device state. + * @skb: The skbuff + */ +static void ks8851_rx_skb(struct ks8851_net *ks, struct sk_buff *skb) +{ + ks->rx_skb(ks, skb); +} + +/** * ks8851_rx_pkts - receive packets from the host * @ks: The device information. * @@ -507,10 +256,9 @@ static void ks8851_rx_pkts(struct ks8851_net *ks) unsigned rxfc; unsigned rxlen; unsigned rxstat; - u32 rxh; u8 *rxpkt; - rxfc = ks8851_rdreg8(ks, KS_RXFC); + rxfc = (ks8851_rdreg16(ks, KS_RXFCTR) >> 8) & 0xff; netif_dbg(ks, rx_status, ks->netdev, "%s: %d packets\n", __func__, rxfc); @@ -526,9 +274,8 @@ static void ks8851_rx_pkts(struct ks8851_net *ks) */ for (; rxfc != 0; rxfc--) { - rxh = ks8851_rdreg32(ks, KS_RXFHSR); - rxstat = rxh & 0xffff; - rxlen = (rxh >> 16) & 0xfff; + rxstat = ks8851_rdreg16(ks, KS_RXFHSR); + rxlen = ks8851_rdreg16(ks, KS_RXFHBCR) & RXFHBCR_CNT_MASK; netif_dbg(ks, rx_status, ks->netdev, "rx: stat 0x%04x, len 0x%04x\n", rxstat, rxlen); @@ -557,13 +304,13 @@ static void ks8851_rx_pkts(struct ks8851_net *ks) rxpkt = skb_put(skb, rxlen) - 8; - ks8851_rdfifo(ks, rxpkt, rxalign + 8); + ks->rdfifo(ks, rxpkt, rxalign + 8); if (netif_msg_pktdata(ks)) ks8851_dbg_dumpkkt(ks, rxpkt); skb->protocol = eth_type_trans(skb, ks->netdev); - netif_rx_ni(skb); + ks8851_rx_skb(ks, skb); ks->netdev->stats.rx_packets++; ks->netdev->stats.rx_bytes += rxlen; @@ -590,10 +337,11 @@ static void ks8851_rx_pkts(struct ks8851_net *ks) static irqreturn_t ks8851_irq(int irq, void *_ks) { struct ks8851_net *ks = _ks; - unsigned status; unsigned handled = 0; + unsigned long flags; + unsigned int status; - mutex_lock(&ks->lock); + ks8851_lock(ks, &flags); status = ks8851_rdreg16(ks, KS_ISR); @@ -631,7 +379,7 @@ static irqreturn_t ks8851_irq(int irq, void *_ks) handled |= IRQ_RXI; if (status & IRQ_SPIBEI) { - dev_err(&ks->spidev->dev, "%s: spi bus error\n", __func__); + netdev_err(ks->netdev, "%s: spi bus error\n", __func__); handled |= IRQ_SPIBEI; } @@ -662,7 +410,7 @@ static irqreturn_t ks8851_irq(int irq, void *_ks) ks8851_wrreg16(ks, KS_RXCR1, rxc->rxcr1); } - mutex_unlock(&ks->lock); + ks8851_unlock(ks, &flags); if (status & IRQ_LCI) mii_check_link(&ks->mii); @@ -674,108 +422,13 @@ static irqreturn_t ks8851_irq(int irq, void *_ks) } /** - * calc_txlen - calculate size of message to send packet - * @len: Length of data - * - * Returns the size of the TXFIFO message needed to send - * this packet. - */ -static inline unsigned calc_txlen(unsigned len) -{ - return ALIGN(len + 4, 4); -} - -/** - * ks8851_wrpkt - write packet to TX FIFO - * @ks: The device state. - * @txp: The sk_buff to transmit. - * @irq: IRQ on completion of the packet. - * - * Send the @txp to the chip. This means creating the relevant packet header - * specifying the length of the packet and the other information the chip - * needs, such as IRQ on completion. Send the header and the packet data to - * the device. - */ -static void ks8851_wrpkt(struct ks8851_net *ks, struct sk_buff *txp, bool irq) -{ - struct spi_transfer *xfer = ks->spi_xfer2; - struct spi_message *msg = &ks->spi_msg2; - unsigned fid = 0; - int ret; - - netif_dbg(ks, tx_queued, ks->netdev, "%s: skb %p, %d@%p, irq %d\n", - __func__, txp, txp->len, txp->data, irq); - - fid = ks->fid++; - fid &= TXFR_TXFID_MASK; - - if (irq) - fid |= TXFR_TXIC; /* irq on completion */ - - /* start header at txb[1] to align txw entries */ - ks->txh.txb[1] = KS_SPIOP_TXFIFO; - ks->txh.txw[1] = cpu_to_le16(fid); - ks->txh.txw[2] = cpu_to_le16(txp->len); - - xfer->tx_buf = &ks->txh.txb[1]; - xfer->rx_buf = NULL; - xfer->len = 5; - - xfer++; - xfer->tx_buf = txp->data; - xfer->rx_buf = NULL; - xfer->len = ALIGN(txp->len, 4); - - ret = spi_sync(ks->spidev, msg); - if (ret < 0) - netdev_err(ks->netdev, "%s: spi_sync() failed\n", __func__); -} - -/** - * ks8851_done_tx - update and then free skbuff after transmitting + * ks8851_flush_tx_work - flush outstanding TX work * @ks: The device state - * @txb: The buffer transmitted - */ -static void ks8851_done_tx(struct ks8851_net *ks, struct sk_buff *txb) -{ - struct net_device *dev = ks->netdev; - - dev->stats.tx_bytes += txb->len; - dev->stats.tx_packets++; - - dev_kfree_skb(txb); -} - -/** - * ks8851_tx_work - process tx packet(s) - * @work: The work strucutre what was scheduled. - * - * This is called when a number of packets have been scheduled for - * transmission and need to be sent to the device. */ -static void ks8851_tx_work(struct work_struct *work) +static void ks8851_flush_tx_work(struct ks8851_net *ks) { - struct ks8851_net *ks = container_of(work, struct ks8851_net, tx_work); - struct sk_buff *txb; - bool last = skb_queue_empty(&ks->txq); - - mutex_lock(&ks->lock); - - while (!last) { - txb = skb_dequeue(&ks->txq); - last = skb_queue_empty(&ks->txq); - - if (txb != NULL) { - ks8851_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr | RXQCR_SDA); - ks8851_wrpkt(ks, txb, last); - ks8851_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr); - ks8851_wrreg16(ks, KS_TXQCR, TXQCR_METFE); - - ks8851_done_tx(ks, txb); - } - } - - mutex_unlock(&ks->lock); + if (ks->flush_tx_work) + ks->flush_tx_work(ks); } /** @@ -788,6 +441,7 @@ static void ks8851_tx_work(struct work_struct *work) static int ks8851_net_open(struct net_device *dev) { struct ks8851_net *ks = netdev_priv(dev); + unsigned long flags; int ret; ret = request_threaded_irq(dev->irq, NULL, ks8851_irq, @@ -800,7 +454,7 @@ static int ks8851_net_open(struct net_device *dev) /* lock the card, even if we may not actually be doing anything * else at the moment */ - mutex_lock(&ks->lock); + ks8851_lock(ks, &flags); netif_dbg(ks, ifup, ks->netdev, "opening\n"); @@ -844,23 +498,14 @@ static int ks8851_net_open(struct net_device *dev) ks8851_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr); /* clear then enable interrupts */ - -#define STD_IRQ (IRQ_LCI | /* Link Change */ \ - IRQ_TXI | /* TX done */ \ - IRQ_RXI | /* RX done */ \ - IRQ_SPIBEI | /* SPI bus error */ \ - IRQ_TXPSI | /* TX process stop */ \ - IRQ_RXPSI) /* RX process stop */ - - ks->rc_ier = STD_IRQ; - ks8851_wrreg16(ks, KS_ISR, STD_IRQ); - ks8851_wrreg16(ks, KS_IER, STD_IRQ); + ks8851_wrreg16(ks, KS_ISR, ks->rc_ier); + ks8851_wrreg16(ks, KS_IER, ks->rc_ier); netif_start_queue(ks->netdev); netif_dbg(ks, ifup, ks->netdev, "network device up\n"); - mutex_unlock(&ks->lock); + ks8851_unlock(ks, &flags); mii_check_link(&ks->mii); return 0; } @@ -876,22 +521,23 @@ static int ks8851_net_open(struct net_device *dev) static int ks8851_net_stop(struct net_device *dev) { struct ks8851_net *ks = netdev_priv(dev); + unsigned long flags; netif_info(ks, ifdown, dev, "shutting down\n"); netif_stop_queue(dev); - mutex_lock(&ks->lock); + ks8851_lock(ks, &flags); /* turn off the IRQs and ack any outstanding */ ks8851_wrreg16(ks, KS_IER, 0x0000); ks8851_wrreg16(ks, KS_ISR, 0xffff); - mutex_unlock(&ks->lock); + ks8851_unlock(ks, &flags); /* stop any outstanding work */ - flush_work(&ks->tx_work); + ks8851_flush_tx_work(ks); flush_work(&ks->rxctrl_work); - mutex_lock(&ks->lock); + ks8851_lock(ks, &flags); /* shutdown RX process */ ks8851_wrreg16(ks, KS_RXCR1, 0x0000); @@ -900,7 +546,7 @@ static int ks8851_net_stop(struct net_device *dev) /* set powermode to soft power down to save power */ ks8851_set_powermode(ks, PMECR_PM_SOFTDOWN); - mutex_unlock(&ks->lock); + ks8851_unlock(ks, &flags); /* ensure any queued tx buffers are dumped */ while (!skb_queue_empty(&ks->txq)) { @@ -934,26 +580,8 @@ static netdev_tx_t ks8851_start_xmit(struct sk_buff *skb, struct net_device *dev) { struct ks8851_net *ks = netdev_priv(dev); - unsigned needed = calc_txlen(skb->len); - netdev_tx_t ret = NETDEV_TX_OK; - - netif_dbg(ks, tx_queued, ks->netdev, - "%s: skb %p, %d@%p\n", __func__, skb, skb->len, skb->data); - spin_lock(&ks->statelock); - - if (needed > ks->tx_space) { - netif_stop_queue(dev); - ret = NETDEV_TX_BUSY; - } else { - ks->tx_space -= needed; - skb_queue_tail(&ks->txq, skb); - } - - spin_unlock(&ks->statelock); - schedule_work(&ks->tx_work); - - return ret; + return ks->start_xmit(skb, dev); } /** @@ -972,13 +600,14 @@ static netdev_tx_t ks8851_start_xmit(struct sk_buff *skb, static void ks8851_rxctrl_work(struct work_struct *work) { struct ks8851_net *ks = container_of(work, struct ks8851_net, rxctrl_work); + unsigned long flags; - mutex_lock(&ks->lock); + ks8851_lock(ks, &flags); /* need to shutdown RXQ before modifying filter parameters */ ks8851_wrreg16(ks, KS_RXCR1, 0x00); - mutex_unlock(&ks->lock); + ks8851_unlock(ks, &flags); } static void ks8851_set_rx_mode(struct net_device *dev) @@ -1160,11 +789,6 @@ static void ks8851_eeprom_regwrite(struct eeprom_93cx6 *ee) */ static int ks8851_eeprom_claim(struct ks8851_net *ks) { - if (!(ks->rc_ccr & CCR_EEPROM)) - return -ENOENT; - - mutex_lock(&ks->lock); - /* start with clock low, cs high */ ks8851_wrreg16(ks, KS_EEPCR, EEPCR_EESA | EEPCR_EECS); return 0; @@ -1181,7 +805,6 @@ static void ks8851_eeprom_release(struct ks8851_net *ks) unsigned val = ks8851_rdreg16(ks, KS_EEPCR); ks8851_wrreg16(ks, KS_EEPCR, val & ~EEPCR_EESA); - mutex_unlock(&ks->lock); } #define KS_EEPROM_MAGIC (0x00008851) @@ -1191,6 +814,7 @@ static int ks8851_set_eeprom(struct net_device *dev, { struct ks8851_net *ks = netdev_priv(dev); int offset = ee->offset; + unsigned long flags; int len = ee->len; u16 tmp; @@ -1201,9 +825,13 @@ static int ks8851_set_eeprom(struct net_device *dev, if (ee->magic != KS_EEPROM_MAGIC) return -EINVAL; - if (ks8851_eeprom_claim(ks)) + if (!(ks->rc_ccr & CCR_EEPROM)) return -ENOENT; + ks8851_lock(ks, &flags); + + ks8851_eeprom_claim(ks); + eeprom_93cx6_wren(&ks->eeprom, true); /* ethtool currently only supports writing bytes, which means @@ -1223,6 +851,7 @@ static int ks8851_set_eeprom(struct net_device *dev, eeprom_93cx6_wren(&ks->eeprom, false); ks8851_eeprom_release(ks); + ks8851_unlock(ks, &flags); return 0; } @@ -1232,19 +861,25 @@ static int ks8851_get_eeprom(struct net_device *dev, { struct ks8851_net *ks = netdev_priv(dev); int offset = ee->offset; + unsigned long flags; int len = ee->len; /* must be 2 byte aligned */ if (len & 1 || offset & 1) return -EINVAL; - if (ks8851_eeprom_claim(ks)) + if (!(ks->rc_ccr & CCR_EEPROM)) return -ENOENT; + ks8851_lock(ks, &flags); + + ks8851_eeprom_claim(ks); + ee->magic = KS_EEPROM_MAGIC; eeprom_93cx6_multiread(&ks->eeprom, offset/2, (__le16 *)data, len/2); ks8851_eeprom_release(ks); + ks8851_unlock(ks, &flags); return 0; } @@ -1318,6 +953,7 @@ static int ks8851_phy_reg(int reg) static int ks8851_phy_read(struct net_device *dev, int phy_addr, int reg) { struct ks8851_net *ks = netdev_priv(dev); + unsigned long flags; int ksreg; int result; @@ -1325,9 +961,9 @@ static int ks8851_phy_read(struct net_device *dev, int phy_addr, int reg) if (!ksreg) return 0x0; /* no error return allowed, so use zero */ - mutex_lock(&ks->lock); + ks8851_lock(ks, &flags); result = ks8851_rdreg16(ks, ksreg); - mutex_unlock(&ks->lock); + ks8851_unlock(ks, &flags); return result; } @@ -1336,13 +972,14 @@ static void ks8851_phy_write(struct net_device *dev, int phy, int reg, int value) { struct ks8851_net *ks = netdev_priv(dev); + unsigned long flags; int ksreg; ksreg = ks8851_phy_reg(reg); if (ksreg) { - mutex_lock(&ks->lock); + ks8851_lock(ks, &flags); ks8851_wrreg16(ks, ksreg, value); - mutex_unlock(&ks->lock); + ks8851_unlock(ks, &flags); } } @@ -1382,7 +1019,7 @@ static int ks8851_read_selftest(struct ks8851_net *ks) #ifdef CONFIG_PM_SLEEP -static int ks8851_suspend(struct device *dev) +int ks8851_suspend(struct device *dev) { struct ks8851_net *ks = dev_get_drvdata(dev); struct net_device *netdev = ks->netdev; @@ -1395,7 +1032,7 @@ static int ks8851_suspend(struct device *dev) return 0; } -static int ks8851_resume(struct device *dev) +int ks8851_resume(struct device *dev) { struct ks8851_net *ks = dev_get_drvdata(dev); struct net_device *netdev = ks->netdev; @@ -1409,46 +1046,32 @@ static int ks8851_resume(struct device *dev) } #endif -static SIMPLE_DEV_PM_OPS(ks8851_pm_ops, ks8851_suspend, ks8851_resume); - -static int ks8851_probe(struct spi_device *spi) +int ks8851_probe_common(struct net_device *netdev, struct device *dev, + int msg_en) { - struct net_device *ndev; - struct ks8851_net *ks; - int ret; + struct ks8851_net *ks = netdev_priv(netdev); unsigned cider; int gpio; + int ret; - ndev = alloc_etherdev(sizeof(struct ks8851_net)); - if (!ndev) - return -ENOMEM; - - spi->bits_per_word = 8; - - ks = netdev_priv(ndev); - - ks->netdev = ndev; - ks->spidev = spi; + ks->netdev = netdev; ks->tx_space = 6144; - gpio = of_get_named_gpio_flags(spi->dev.of_node, "reset-gpios", - 0, NULL); - if (gpio == -EPROBE_DEFER) { - ret = gpio; - goto err_gpio; - } + gpio = of_get_named_gpio_flags(dev->of_node, "reset-gpios", 0, NULL); + if (gpio == -EPROBE_DEFER) + return gpio; ks->gpio = gpio; if (gpio_is_valid(gpio)) { - ret = devm_gpio_request_one(&spi->dev, gpio, + ret = devm_gpio_request_one(dev, gpio, GPIOF_OUT_INIT_LOW, "ks8851_rst_n"); if (ret) { - dev_err(&spi->dev, "reset gpio request failed\n"); - goto err_gpio; + dev_err(dev, "reset gpio request failed\n"); + return ret; } } - ks->vdd_io = devm_regulator_get(&spi->dev, "vdd-io"); + ks->vdd_io = devm_regulator_get(dev, "vdd-io"); if (IS_ERR(ks->vdd_io)) { ret = PTR_ERR(ks->vdd_io); goto err_reg_io; @@ -1456,12 +1079,11 @@ static int ks8851_probe(struct spi_device *spi) ret = regulator_enable(ks->vdd_io); if (ret) { - dev_err(&spi->dev, "regulator vdd_io enable fail: %d\n", - ret); + dev_err(dev, "regulator vdd_io enable fail: %d\n", ret); goto err_reg_io; } - ks->vdd_reg = devm_regulator_get(&spi->dev, "vdd"); + ks->vdd_reg = devm_regulator_get(dev, "vdd"); if (IS_ERR(ks->vdd_reg)) { ret = PTR_ERR(ks->vdd_reg); goto err_reg; @@ -1469,8 +1091,7 @@ static int ks8851_probe(struct spi_device *spi) ret = regulator_enable(ks->vdd_reg); if (ret) { - dev_err(&spi->dev, "regulator vdd enable fail: %d\n", - ret); + dev_err(dev, "regulator vdd enable fail: %d\n", ret); goto err_reg; } @@ -1479,54 +1100,41 @@ static int ks8851_probe(struct spi_device *spi) gpio_set_value(gpio, 1); } - mutex_init(&ks->lock); spin_lock_init(&ks->statelock); - INIT_WORK(&ks->tx_work, ks8851_tx_work); INIT_WORK(&ks->rxctrl_work, ks8851_rxctrl_work); - /* initialise pre-made spi transfer messages */ - - spi_message_init(&ks->spi_msg1); - spi_message_add_tail(&ks->spi_xfer1, &ks->spi_msg1); - - spi_message_init(&ks->spi_msg2); - spi_message_add_tail(&ks->spi_xfer2[0], &ks->spi_msg2); - spi_message_add_tail(&ks->spi_xfer2[1], &ks->spi_msg2); - /* setup EEPROM state */ - ks->eeprom.data = ks; ks->eeprom.width = PCI_EEPROM_WIDTH_93C46; ks->eeprom.register_read = ks8851_eeprom_regread; ks->eeprom.register_write = ks8851_eeprom_regwrite; /* setup mii state */ - ks->mii.dev = ndev; + ks->mii.dev = netdev; ks->mii.phy_id = 1, ks->mii.phy_id_mask = 1; ks->mii.reg_num_mask = 0xf; ks->mii.mdio_read = ks8851_phy_read; ks->mii.mdio_write = ks8851_phy_write; - dev_info(&spi->dev, "message enable is %d\n", msg_enable); + dev_info(dev, "message enable is %d\n", msg_en); /* set the default message enable */ - ks->msg_enable = netif_msg_init(msg_enable, (NETIF_MSG_DRV | - NETIF_MSG_PROBE | - NETIF_MSG_LINK)); + ks->msg_enable = netif_msg_init(msg_en, NETIF_MSG_DRV | + NETIF_MSG_PROBE | + NETIF_MSG_LINK); skb_queue_head_init(&ks->txq); - ndev->ethtool_ops = &ks8851_ethtool_ops; - SET_NETDEV_DEV(ndev, &spi->dev); + netdev->ethtool_ops = &ks8851_ethtool_ops; + SET_NETDEV_DEV(netdev, dev); - spi_set_drvdata(spi, ks); + dev_set_drvdata(dev, ks); netif_carrier_off(ks->netdev); - ndev->if_port = IF_PORT_100BASET; - ndev->netdev_ops = &ks8851_netdev_ops; - ndev->irq = spi->irq; + netdev->if_port = IF_PORT_100BASET; + netdev->netdev_ops = &ks8851_netdev_ops; /* issue a global soft reset to reset the device. */ ks8851_soft_reset(ks, GRR_GSR); @@ -1534,7 +1142,7 @@ static int ks8851_probe(struct spi_device *spi) /* simple check for a valid chip being connected to the bus */ cider = ks8851_rdreg16(ks, KS_CIDER); if ((cider & ~CIDER_REV_MASK) != CIDER_ID) { - dev_err(&spi->dev, "failed to read device ID\n"); + dev_err(dev, "failed to read device ID\n"); ret = -ENODEV; goto err_id; } @@ -1543,16 +1151,16 @@ static int ks8851_probe(struct spi_device *spi) ks->rc_ccr = ks8851_rdreg16(ks, KS_CCR); ks8851_read_selftest(ks); - ks8851_init_mac(ks); + ks8851_init_mac(ks, dev->of_node); - ret = register_netdev(ndev); + ret = register_netdev(netdev); if (ret) { - dev_err(&spi->dev, "failed to register network device\n"); + dev_err(dev, "failed to register network device\n"); goto err_netdev; } - netdev_info(ndev, "revision %d, MAC %pM, IRQ %d, %s EEPROM\n", - CIDER_REV_GET(cider), ndev->dev_addr, ndev->irq, + netdev_info(netdev, "revision %d, MAC %pM, IRQ %d, %s EEPROM\n", + CIDER_REV_GET(cider), netdev->dev_addr, netdev->irq, ks->rc_ccr & CCR_EEPROM ? "has" : "no"); return 0; @@ -1565,49 +1173,21 @@ err_id: err_reg: regulator_disable(ks->vdd_io); err_reg_io: -err_gpio: - free_netdev(ndev); return ret; } -static int ks8851_remove(struct spi_device *spi) +int ks8851_remove_common(struct device *dev) { - struct ks8851_net *priv = spi_get_drvdata(spi); + struct ks8851_net *priv = dev_get_drvdata(dev); if (netif_msg_drv(priv)) - dev_info(&spi->dev, "remove\n"); + dev_info(dev, "remove\n"); unregister_netdev(priv->netdev); if (gpio_is_valid(priv->gpio)) gpio_set_value(priv->gpio, 0); regulator_disable(priv->vdd_reg); regulator_disable(priv->vdd_io); - free_netdev(priv->netdev); return 0; } - -static const struct of_device_id ks8851_match_table[] = { - { .compatible = "micrel,ks8851" }, - { } -}; -MODULE_DEVICE_TABLE(of, ks8851_match_table); - -static struct spi_driver ks8851_driver = { - .driver = { - .name = "ks8851", - .of_match_table = ks8851_match_table, - .pm = &ks8851_pm_ops, - }, - .probe = ks8851_probe, - .remove = ks8851_remove, -}; -module_spi_driver(ks8851_driver); - -MODULE_DESCRIPTION("KS8851 Network driver"); -MODULE_AUTHOR("Ben Dooks <ben@simtec.co.uk>"); -MODULE_LICENSE("GPL"); - -module_param_named(message, msg_enable, int, 0); -MODULE_PARM_DESC(message, "Message verbosity level (0=none, 31=all)"); -MODULE_ALIAS("spi:ks8851"); diff --git a/drivers/net/ethernet/micrel/ks8851_mll.c b/drivers/net/ethernet/micrel/ks8851_mll.c deleted file mode 100644 index 45cc840d8e2e..000000000000 --- a/drivers/net/ethernet/micrel/ks8851_mll.c +++ /dev/null @@ -1,1393 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/** - * drivers/net/ethernet/micrel/ks8851_mll.c - * Copyright (c) 2009 Micrel Inc. - */ - -/* Supports: - * KS8851 16bit MLL chip from Micrel Inc. - */ - -#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt - -#include <linux/interrupt.h> -#include <linux/module.h> -#include <linux/kernel.h> -#include <linux/netdevice.h> -#include <linux/etherdevice.h> -#include <linux/ethtool.h> -#include <linux/cache.h> -#include <linux/crc32.h> -#include <linux/crc32poly.h> -#include <linux/mii.h> -#include <linux/platform_device.h> -#include <linux/delay.h> -#include <linux/slab.h> -#include <linux/ks8851_mll.h> -#include <linux/of.h> -#include <linux/of_device.h> -#include <linux/of_net.h> - -#include "ks8851.h" - -#define DRV_NAME "ks8851_mll" - -static u8 KS_DEFAULT_MAC_ADDRESS[] = { 0x00, 0x10, 0xA1, 0x86, 0x95, 0x11 }; -#define MAX_RECV_FRAMES 255 -#define MAX_BUF_SIZE 2048 -#define TX_BUF_SIZE 2000 -#define RX_BUF_SIZE 2000 - -#define RXCR1_FILTER_MASK (RXCR1_RXINVF | RXCR1_RXAE | \ - RXCR1_RXMAFMA | RXCR1_RXPAFMA) -#define RXQCR_CMD_CNTL (RXQCR_RXFCTE|RXQCR_ADRFE) - -#define ENUM_BUS_NONE 0 -#define ENUM_BUS_8BIT 1 -#define ENUM_BUS_16BIT 2 -#define ENUM_BUS_32BIT 3 - -#define MAX_MCAST_LST 32 -#define HW_MCAST_SIZE 8 - -/** - * union ks_tx_hdr - tx header data - * @txb: The header as bytes - * @txw: The header as 16bit, little-endian words - * - * A dual representation of the tx header data to allow - * access to individual bytes, and to allow 16bit accesses - * with 16bit alignment. - */ -union ks_tx_hdr { - u8 txb[4]; - __le16 txw[2]; -}; - -/** - * struct ks_net - KS8851 driver private data - * @net_device : The network device we're bound to - * @hw_addr : start address of data register. - * @hw_addr_cmd : start address of command register. - * @txh : temporaly buffer to save status/length. - * @lock : Lock to ensure that the device is not accessed when busy. - * @pdev : Pointer to platform device. - * @mii : The MII state information for the mii calls. - * @frame_head_info : frame header information for multi-pkt rx. - * @statelock : Lock on this structure for tx list. - * @msg_enable : The message flags controlling driver output (see ethtool). - * @frame_cnt : number of frames received. - * @bus_width : i/o bus width. - * @rc_rxqcr : Cached copy of KS_RXQCR. - * @rc_txcr : Cached copy of KS_TXCR. - * @rc_ier : Cached copy of KS_IER. - * @sharedbus : Multipex(addr and data bus) mode indicator. - * @cmd_reg_cache : command register cached. - * @cmd_reg_cache_int : command register cached. Used in the irq handler. - * @promiscuous : promiscuous mode indicator. - * @all_mcast : mutlicast indicator. - * @mcast_lst_size : size of multicast list. - * @mcast_lst : multicast list. - * @mcast_bits : multicast enabed. - * @mac_addr : MAC address assigned to this device. - * @fid : frame id. - * @extra_byte : number of extra byte prepended rx pkt. - * @enabled : indicator this device works. - * - * The @lock ensures that the chip is protected when certain operations are - * in progress. When the read or write packet transfer is in progress, most - * of the chip registers are not accessible until the transfer is finished and - * the DMA has been de-asserted. - * - * The @statelock is used to protect information in the structure which may - * need to be accessed via several sources, such as the network driver layer - * or one of the work queues. - * - */ - -/* Receive multiplex framer header info */ -struct type_frame_head { - u16 sts; /* Frame status */ - u16 len; /* Byte count */ -}; - -struct ks_net { - struct net_device *netdev; - void __iomem *hw_addr; - void __iomem *hw_addr_cmd; - union ks_tx_hdr txh ____cacheline_aligned; - struct mutex lock; /* spinlock to be interrupt safe */ - struct platform_device *pdev; - struct mii_if_info mii; - struct type_frame_head *frame_head_info; - spinlock_t statelock; - u32 msg_enable; - u32 frame_cnt; - int bus_width; - - u16 rc_rxqcr; - u16 rc_txcr; - u16 rc_ier; - u16 sharedbus; - u16 cmd_reg_cache; - u16 cmd_reg_cache_int; - u16 promiscuous; - u16 all_mcast; - u16 mcast_lst_size; - u8 mcast_lst[MAX_MCAST_LST][ETH_ALEN]; - u8 mcast_bits[HW_MCAST_SIZE]; - u8 mac_addr[6]; - u8 fid; - u8 extra_byte; - u8 enabled; -}; - -static int msg_enable; - -#define BE3 0x8000 /* Byte Enable 3 */ -#define BE2 0x4000 /* Byte Enable 2 */ -#define BE1 0x2000 /* Byte Enable 1 */ -#define BE0 0x1000 /* Byte Enable 0 */ - -/* register read/write calls. - * - * All these calls issue transactions to access the chip's registers. They - * all require that the necessary lock is held to prevent accesses when the - * chip is busy transferring packet data (RX/TX FIFO accesses). - */ - -/** - * ks_check_endian - Check whether endianness of the bus is correct - * @ks : The chip information - * - * The KS8851-16MLL EESK pin allows selecting the endianness of the 16bit - * bus. To maintain optimum performance, the bus endianness should be set - * such that it matches the endianness of the CPU. - */ - -static int ks_check_endian(struct ks_net *ks) -{ - u16 cider; - - /* - * Read CIDER register first, however read it the "wrong" way around. - * If the endian strap on the KS8851-16MLL in incorrect and the chip - * is operating in different endianness than the CPU, then the meaning - * of BE[3:0] byte-enable bits is also swapped such that: - * BE[3,2,1,0] becomes BE[1,0,3,2] - * - * Luckily for us, the byte-enable bits are the top four MSbits of - * the address register and the CIDER register is at offset 0xc0. - * Hence, by reading address 0xc0c0, which is not impacted by endian - * swapping, we assert either BE[3:2] or BE[1:0] while reading the - * CIDER register. - * - * If the bus configuration is correct, reading 0xc0c0 asserts - * BE[3:2] and this read returns 0x0000, because to read register - * with bottom two LSbits of address set to 0, BE[1:0] must be - * asserted. - * - * If the bus configuration is NOT correct, reading 0xc0c0 asserts - * BE[1:0] and this read returns non-zero 0x8872 value. - */ - iowrite16(BE3 | BE2 | KS_CIDER, ks->hw_addr_cmd); - cider = ioread16(ks->hw_addr); - if (!cider) - return 0; - - netdev_err(ks->netdev, "incorrect EESK endian strap setting\n"); - - return -EINVAL; -} - -/** - * ks_rdreg16 - read 16 bit register from device - * @ks : The chip information - * @offset: The register address - * - * Read a 16bit register from the chip, returning the result - */ - -static u16 ks_rdreg16(struct ks_net *ks, int offset) -{ - ks->cmd_reg_cache = (u16)offset | ((BE1 | BE0) << (offset & 0x02)); - iowrite16(ks->cmd_reg_cache, ks->hw_addr_cmd); - return ioread16(ks->hw_addr); -} - -/** - * ks_wrreg16 - write 16bit register value to chip - * @ks: The chip information - * @offset: The register address - * @value: The value to write - * - */ - -static void ks_wrreg16(struct ks_net *ks, int offset, u16 value) -{ - ks->cmd_reg_cache = (u16)offset | ((BE1 | BE0) << (offset & 0x02)); - iowrite16(ks->cmd_reg_cache, ks->hw_addr_cmd); - iowrite16(value, ks->hw_addr); -} - -/** - * ks_inblk - read a block of data from QMU. This is called after sudo DMA mode enabled. - * @ks: The chip state - * @wptr: buffer address to save data - * @len: length in byte to read - * - */ -static inline void ks_inblk(struct ks_net *ks, u16 *wptr, u32 len) -{ - len >>= 1; - while (len--) - *wptr++ = (u16)ioread16(ks->hw_addr); -} - -/** - * ks_outblk - write data to QMU. This is called after sudo DMA mode enabled. - * @ks: The chip information - * @wptr: buffer address - * @len: length in byte to write - * - */ -static inline void ks_outblk(struct ks_net *ks, u16 *wptr, u32 len) -{ - len >>= 1; - while (len--) - iowrite16(*wptr++, ks->hw_addr); -} - -static void ks_disable_int(struct ks_net *ks) -{ - ks_wrreg16(ks, KS_IER, 0x0000); -} /* ks_disable_int */ - -static void ks_enable_int(struct ks_net *ks) -{ - ks_wrreg16(ks, KS_IER, ks->rc_ier); -} /* ks_enable_int */ - -/** - * ks_tx_fifo_space - return the available hardware buffer size. - * @ks: The chip information - * - */ -static inline u16 ks_tx_fifo_space(struct ks_net *ks) -{ - return ks_rdreg16(ks, KS_TXMIR) & 0x1fff; -} - -/** - * ks_save_cmd_reg - save the command register from the cache. - * @ks: The chip information - * - */ -static inline void ks_save_cmd_reg(struct ks_net *ks) -{ - /*ks8851 MLL has a bug to read back the command register. - * So rely on software to save the content of command register. - */ - ks->cmd_reg_cache_int = ks->cmd_reg_cache; -} - -/** - * ks_restore_cmd_reg - restore the command register from the cache and - * write to hardware register. - * @ks: The chip information - * - */ -static inline void ks_restore_cmd_reg(struct ks_net *ks) -{ - ks->cmd_reg_cache = ks->cmd_reg_cache_int; - iowrite16(ks->cmd_reg_cache, ks->hw_addr_cmd); -} - -/** - * ks_set_powermode - set power mode of the device - * @ks: The chip information - * @pwrmode: The power mode value to write to KS_PMECR. - * - * Change the power mode of the chip. - */ -static void ks_set_powermode(struct ks_net *ks, unsigned pwrmode) -{ - unsigned pmecr; - - netif_dbg(ks, hw, ks->netdev, "setting power mode %d\n", pwrmode); - - ks_rdreg16(ks, KS_GRR); - pmecr = ks_rdreg16(ks, KS_PMECR); - pmecr &= ~PMECR_PM_MASK; - pmecr |= pwrmode; - - ks_wrreg16(ks, KS_PMECR, pmecr); -} - -/** - * ks_read_config - read chip configuration of bus width. - * @ks: The chip information - * - */ -static void ks_read_config(struct ks_net *ks) -{ - u16 reg_data = 0; - - /* Regardless of bus width, 8 bit read should always work.*/ - reg_data = ks_rdreg16(ks, KS_CCR); - - /* addr/data bus are multiplexed */ - ks->sharedbus = (reg_data & CCR_SHARED) == CCR_SHARED; - - /* There are garbage data when reading data from QMU, - depending on bus-width. - */ - - if (reg_data & CCR_8BIT) { - ks->bus_width = ENUM_BUS_8BIT; - ks->extra_byte = 1; - } else if (reg_data & CCR_16BIT) { - ks->bus_width = ENUM_BUS_16BIT; - ks->extra_byte = 2; - } else { - ks->bus_width = ENUM_BUS_32BIT; - ks->extra_byte = 4; - } -} - -/** - * ks_soft_reset - issue one of the soft reset to the device - * @ks: The device state. - * @op: The bit(s) to set in the GRR - * - * Issue the relevant soft-reset command to the device's GRR register - * specified by @op. - * - * Note, the delays are in there as a caution to ensure that the reset - * has time to take effect and then complete. Since the datasheet does - * not currently specify the exact sequence, we have chosen something - * that seems to work with our device. - */ -static void ks_soft_reset(struct ks_net *ks, unsigned op) -{ - /* Disable interrupt first */ - ks_wrreg16(ks, KS_IER, 0x0000); - ks_wrreg16(ks, KS_GRR, op); - mdelay(10); /* wait a short time to effect reset */ - ks_wrreg16(ks, KS_GRR, 0); - mdelay(1); /* wait for condition to clear */ -} - - -static void ks_enable_qmu(struct ks_net *ks) -{ - u16 w; - - w = ks_rdreg16(ks, KS_TXCR); - /* Enables QMU Transmit (TXCR). */ - ks_wrreg16(ks, KS_TXCR, w | TXCR_TXE); - - /* - * RX Frame Count Threshold Enable and Auto-Dequeue RXQ Frame - * Enable - */ - - w = ks_rdreg16(ks, KS_RXQCR); - ks_wrreg16(ks, KS_RXQCR, w | RXQCR_RXFCTE); - - /* Enables QMU Receive (RXCR1). */ - w = ks_rdreg16(ks, KS_RXCR1); - ks_wrreg16(ks, KS_RXCR1, w | RXCR1_RXE); - ks->enabled = true; -} /* ks_enable_qmu */ - -static void ks_disable_qmu(struct ks_net *ks) -{ - u16 w; - - w = ks_rdreg16(ks, KS_TXCR); - - /* Disables QMU Transmit (TXCR). */ - w &= ~TXCR_TXE; - ks_wrreg16(ks, KS_TXCR, w); - - /* Disables QMU Receive (RXCR1). */ - w = ks_rdreg16(ks, KS_RXCR1); - w &= ~RXCR1_RXE ; - ks_wrreg16(ks, KS_RXCR1, w); - - ks->enabled = false; - -} /* ks_disable_qmu */ - -/** - * ks_read_qmu - read 1 pkt data from the QMU. - * @ks: The chip information - * @buf: buffer address to save 1 pkt - * @len: Pkt length - * Here is the sequence to read 1 pkt: - * 1. set sudo DMA mode - * 2. read prepend data - * 3. read pkt data - * 4. reset sudo DMA Mode - */ -static inline void ks_read_qmu(struct ks_net *ks, u16 *buf, u32 len) -{ - u32 r = ks->extra_byte & 0x1 ; - u32 w = ks->extra_byte - r; - - /* 1. set sudo DMA mode */ - ks_wrreg16(ks, KS_RXFDPR, RXFDPR_RXFPAI); - ks_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr | RXQCR_SDA); - - /* 2. read prepend data */ - /** - * read 4 + extra bytes and discard them. - * extra bytes for dummy, 2 for status, 2 for len - */ - - /* use likely(r) for 8 bit access for performance */ - if (unlikely(r)) - ioread8(ks->hw_addr); - ks_inblk(ks, buf, w + 2 + 2); - - /* 3. read pkt data */ - ks_inblk(ks, buf, ALIGN(len, 4)); - - /* 4. reset sudo DMA Mode */ - ks_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr); -} - -/** - * ks_rcv - read multiple pkts data from the QMU. - * @ks: The chip information - * @netdev: The network device being opened. - * - * Read all of header information before reading pkt content. - * It is not allowed only port of pkts in QMU after issuing - * interrupt ack. - */ -static void ks_rcv(struct ks_net *ks, struct net_device *netdev) -{ - u32 i; - struct type_frame_head *frame_hdr = ks->frame_head_info; - struct sk_buff *skb; - - ks->frame_cnt = ks_rdreg16(ks, KS_RXFCTR) >> 8; - - /* read all header information */ - for (i = 0; i < ks->frame_cnt; i++) { - /* Checking Received packet status */ - frame_hdr->sts = ks_rdreg16(ks, KS_RXFHSR); - /* Get packet len from hardware */ - frame_hdr->len = ks_rdreg16(ks, KS_RXFHBCR); - frame_hdr++; - } - - frame_hdr = ks->frame_head_info; - while (ks->frame_cnt--) { - if (unlikely(!(frame_hdr->sts & RXFSHR_RXFV) || - frame_hdr->len >= RX_BUF_SIZE || - frame_hdr->len <= 0)) { - - /* discard an invalid packet */ - ks_wrreg16(ks, KS_RXQCR, (ks->rc_rxqcr | RXQCR_RRXEF)); - netdev->stats.rx_dropped++; - if (!(frame_hdr->sts & RXFSHR_RXFV)) - netdev->stats.rx_frame_errors++; - else - netdev->stats.rx_length_errors++; - frame_hdr++; - continue; - } - - skb = netdev_alloc_skb(netdev, frame_hdr->len + 16); - if (likely(skb)) { - skb_reserve(skb, 2); - /* read data block including CRC 4 bytes */ - ks_read_qmu(ks, (u16 *)skb->data, frame_hdr->len); - skb_put(skb, frame_hdr->len - 4); - skb->protocol = eth_type_trans(skb, netdev); - netif_rx(skb); - /* exclude CRC size */ - netdev->stats.rx_bytes += frame_hdr->len - 4; - netdev->stats.rx_packets++; - } else { - ks_wrreg16(ks, KS_RXQCR, (ks->rc_rxqcr | RXQCR_RRXEF)); - netdev->stats.rx_dropped++; - } - frame_hdr++; - } -} - -/** - * ks_update_link_status - link status update. - * @netdev: The network device being opened. - * @ks: The chip information - * - */ - -static void ks_update_link_status(struct net_device *netdev, struct ks_net *ks) -{ - /* check the status of the link */ - u32 link_up_status; - if (ks_rdreg16(ks, KS_P1SR) & P1SR_LINK_GOOD) { - netif_carrier_on(netdev); - link_up_status = true; - } else { - netif_carrier_off(netdev); - link_up_status = false; - } - netif_dbg(ks, link, ks->netdev, - "%s: %s\n", __func__, link_up_status ? "UP" : "DOWN"); -} - -/** - * ks_irq - device interrupt handler - * @irq: Interrupt number passed from the IRQ handler. - * @pw: The private word passed to register_irq(), our struct ks_net. - * - * This is the handler invoked to find out what happened - * - * Read the interrupt status, work out what needs to be done and then clear - * any of the interrupts that are not needed. - */ - -static irqreturn_t ks_irq(int irq, void *pw) -{ - struct net_device *netdev = pw; - struct ks_net *ks = netdev_priv(netdev); - unsigned long flags; - u16 status; - - spin_lock_irqsave(&ks->statelock, flags); - /*this should be the first in IRQ handler */ - ks_save_cmd_reg(ks); - - status = ks_rdreg16(ks, KS_ISR); - if (unlikely(!status)) { - ks_restore_cmd_reg(ks); - spin_unlock_irqrestore(&ks->statelock, flags); - return IRQ_NONE; - } - - ks_wrreg16(ks, KS_ISR, status); - - if (likely(status & IRQ_RXI)) - ks_rcv(ks, netdev); - - if (unlikely(status & IRQ_LCI)) - ks_update_link_status(netdev, ks); - - if (unlikely(status & IRQ_TXI)) - netif_wake_queue(netdev); - - if (unlikely(status & IRQ_LDI)) { - - u16 pmecr = ks_rdreg16(ks, KS_PMECR); - pmecr &= ~PMECR_WKEVT_MASK; - ks_wrreg16(ks, KS_PMECR, pmecr | PMECR_WKEVT_LINK); - } - - if (unlikely(status & IRQ_RXOI)) - ks->netdev->stats.rx_over_errors++; - /* this should be the last in IRQ handler*/ - ks_restore_cmd_reg(ks); - spin_unlock_irqrestore(&ks->statelock, flags); - return IRQ_HANDLED; -} - - -/** - * ks_net_open - open network device - * @netdev: The network device being opened. - * - * Called when the network device is marked active, such as a user executing - * 'ifconfig up' on the device. - */ -static int ks_net_open(struct net_device *netdev) -{ - struct ks_net *ks = netdev_priv(netdev); - int err; - -#define KS_INT_FLAGS IRQF_TRIGGER_LOW - /* lock the card, even if we may not actually do anything - * else at the moment. - */ - - netif_dbg(ks, ifup, ks->netdev, "%s - entry\n", __func__); - - /* reset the HW */ - err = request_irq(netdev->irq, ks_irq, KS_INT_FLAGS, DRV_NAME, netdev); - - if (err) { - pr_err("Failed to request IRQ: %d: %d\n", netdev->irq, err); - return err; - } - - /* wake up powermode to normal mode */ - ks_set_powermode(ks, PMECR_PM_NORMAL); - mdelay(1); /* wait for normal mode to take effect */ - - ks_wrreg16(ks, KS_ISR, 0xffff); - ks_enable_int(ks); - ks_enable_qmu(ks); - netif_start_queue(ks->netdev); - - netif_dbg(ks, ifup, ks->netdev, "network device up\n"); - - return 0; -} - -/** - * ks_net_stop - close network device - * @netdev: The device being closed. - * - * Called to close down a network device which has been active. Cancell any - * work, shutdown the RX and TX process and then place the chip into a low - * power state whilst it is not being used. - */ -static int ks_net_stop(struct net_device *netdev) -{ - struct ks_net *ks = netdev_priv(netdev); - - netif_info(ks, ifdown, netdev, "shutting down\n"); - - netif_stop_queue(netdev); - - mutex_lock(&ks->lock); - - /* turn off the IRQs and ack any outstanding */ - ks_wrreg16(ks, KS_IER, 0x0000); - ks_wrreg16(ks, KS_ISR, 0xffff); - - /* shutdown RX/TX QMU */ - ks_disable_qmu(ks); - ks_disable_int(ks); - - /* set powermode to soft power down to save power */ - ks_set_powermode(ks, PMECR_PM_SOFTDOWN); - free_irq(netdev->irq, netdev); - mutex_unlock(&ks->lock); - return 0; -} - - -/** - * ks_write_qmu - write 1 pkt data to the QMU. - * @ks: The chip information - * @pdata: buffer address to save 1 pkt - * @len: Pkt length in byte - * Here is the sequence to write 1 pkt: - * 1. set sudo DMA mode - * 2. write status/length - * 3. write pkt data - * 4. reset sudo DMA Mode - * 5. reset sudo DMA mode - * 6. Wait until pkt is out - */ -static void ks_write_qmu(struct ks_net *ks, u8 *pdata, u16 len) -{ - /* start header at txb[0] to align txw entries */ - ks->txh.txw[0] = 0; - ks->txh.txw[1] = cpu_to_le16(len); - - /* 1. set sudo-DMA mode */ - ks_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr | RXQCR_SDA); - /* 2. write status/lenth info */ - ks_outblk(ks, ks->txh.txw, 4); - /* 3. write pkt data */ - ks_outblk(ks, (u16 *)pdata, ALIGN(len, 4)); - /* 4. reset sudo-DMA mode */ - ks_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr); - /* 5. Enqueue Tx(move the pkt from TX buffer into TXQ) */ - ks_wrreg16(ks, KS_TXQCR, TXQCR_METFE); - /* 6. wait until TXQCR_METFE is auto-cleared */ - while (ks_rdreg16(ks, KS_TXQCR) & TXQCR_METFE) - ; -} - -/** - * ks_start_xmit - transmit packet - * @skb : The buffer to transmit - * @netdev : The device used to transmit the packet. - * - * Called by the network layer to transmit the @skb. - * spin_lock_irqsave is required because tx and rx should be mutual exclusive. - * So while tx is in-progress, prevent IRQ interrupt from happenning. - */ -static netdev_tx_t ks_start_xmit(struct sk_buff *skb, struct net_device *netdev) -{ - netdev_tx_t retv = NETDEV_TX_OK; - struct ks_net *ks = netdev_priv(netdev); - unsigned long flags; - - spin_lock_irqsave(&ks->statelock, flags); - - /* Extra space are required: - * 4 byte for alignment, 4 for status/length, 4 for CRC - */ - - if (likely(ks_tx_fifo_space(ks) >= skb->len + 12)) { - ks_write_qmu(ks, skb->data, skb->len); - /* add tx statistics */ - netdev->stats.tx_bytes += skb->len; - netdev->stats.tx_packets++; - dev_kfree_skb(skb); - } else - retv = NETDEV_TX_BUSY; - spin_unlock_irqrestore(&ks->statelock, flags); - return retv; -} - -/** - * ks_start_rx - ready to serve pkts - * @ks : The chip information - * - */ -static void ks_start_rx(struct ks_net *ks) -{ - u16 cntl; - - /* Enables QMU Receive (RXCR1). */ - cntl = ks_rdreg16(ks, KS_RXCR1); - cntl |= RXCR1_RXE ; - ks_wrreg16(ks, KS_RXCR1, cntl); -} /* ks_start_rx */ - -/** - * ks_stop_rx - stop to serve pkts - * @ks : The chip information - * - */ -static void ks_stop_rx(struct ks_net *ks) -{ - u16 cntl; - - /* Disables QMU Receive (RXCR1). */ - cntl = ks_rdreg16(ks, KS_RXCR1); - cntl &= ~RXCR1_RXE ; - ks_wrreg16(ks, KS_RXCR1, cntl); - -} /* ks_stop_rx */ - -static unsigned long const ethernet_polynomial = CRC32_POLY_BE; - -static unsigned long ether_gen_crc(int length, u8 *data) -{ - long crc = -1; - while (--length >= 0) { - u8 current_octet = *data++; - int bit; - - for (bit = 0; bit < 8; bit++, current_octet >>= 1) { - crc = (crc << 1) ^ - ((crc < 0) ^ (current_octet & 1) ? - ethernet_polynomial : 0); - } - } - return (unsigned long)crc; -} /* ether_gen_crc */ - -/** -* ks_set_grpaddr - set multicast information -* @ks : The chip information -*/ - -static void ks_set_grpaddr(struct ks_net *ks) -{ - u8 i; - u32 index, position, value; - - memset(ks->mcast_bits, 0, sizeof(u8) * HW_MCAST_SIZE); - - for (i = 0; i < ks->mcast_lst_size; i++) { - position = (ether_gen_crc(6, ks->mcast_lst[i]) >> 26) & 0x3f; - index = position >> 3; - value = 1 << (position & 7); - ks->mcast_bits[index] |= (u8)value; - } - - for (i = 0; i < HW_MCAST_SIZE; i++) { - if (i & 1) { - ks_wrreg16(ks, (u16)((KS_MAHTR0 + i) & ~1), - (ks->mcast_bits[i] << 8) | - ks->mcast_bits[i - 1]); - } - } -} /* ks_set_grpaddr */ - -/** -* ks_clear_mcast - clear multicast information -* -* @ks : The chip information -* This routine removes all mcast addresses set in the hardware. -*/ - -static void ks_clear_mcast(struct ks_net *ks) -{ - u16 i, mcast_size; - for (i = 0; i < HW_MCAST_SIZE; i++) - ks->mcast_bits[i] = 0; - - mcast_size = HW_MCAST_SIZE >> 2; - for (i = 0; i < mcast_size; i++) - ks_wrreg16(ks, KS_MAHTR0 + (2*i), 0); -} - -static void ks_set_promis(struct ks_net *ks, u16 promiscuous_mode) -{ - u16 cntl; - ks->promiscuous = promiscuous_mode; - ks_stop_rx(ks); /* Stop receiving for reconfiguration */ - cntl = ks_rdreg16(ks, KS_RXCR1); - - cntl &= ~RXCR1_FILTER_MASK; - if (promiscuous_mode) - /* Enable Promiscuous mode */ - cntl |= RXCR1_RXAE | RXCR1_RXINVF; - else - /* Disable Promiscuous mode (default normal mode) */ - cntl |= RXCR1_RXPAFMA; - - ks_wrreg16(ks, KS_RXCR1, cntl); - - if (ks->enabled) - ks_start_rx(ks); - -} /* ks_set_promis */ - -static void ks_set_mcast(struct ks_net *ks, u16 mcast) -{ - u16 cntl; - - ks->all_mcast = mcast; - ks_stop_rx(ks); /* Stop receiving for reconfiguration */ - cntl = ks_rdreg16(ks, KS_RXCR1); - cntl &= ~RXCR1_FILTER_MASK; - if (mcast) - /* Enable "Perfect with Multicast address passed mode" */ - cntl |= (RXCR1_RXAE | RXCR1_RXMAFMA | RXCR1_RXPAFMA); - else - /** - * Disable "Perfect with Multicast address passed - * mode" (normal mode). - */ - cntl |= RXCR1_RXPAFMA; - - ks_wrreg16(ks, KS_RXCR1, cntl); - - if (ks->enabled) - ks_start_rx(ks); -} /* ks_set_mcast */ - -static void ks_set_rx_mode(struct net_device *netdev) -{ - struct ks_net *ks = netdev_priv(netdev); - struct netdev_hw_addr *ha; - - /* Turn on/off promiscuous mode. */ - if ((netdev->flags & IFF_PROMISC) == IFF_PROMISC) - ks_set_promis(ks, - (u16)((netdev->flags & IFF_PROMISC) == IFF_PROMISC)); - /* Turn on/off all mcast mode. */ - else if ((netdev->flags & IFF_ALLMULTI) == IFF_ALLMULTI) - ks_set_mcast(ks, - (u16)((netdev->flags & IFF_ALLMULTI) == IFF_ALLMULTI)); - else - ks_set_promis(ks, false); - - if ((netdev->flags & IFF_MULTICAST) && netdev_mc_count(netdev)) { - if (netdev_mc_count(netdev) <= MAX_MCAST_LST) { - int i = 0; - - netdev_for_each_mc_addr(ha, netdev) { - if (i >= MAX_MCAST_LST) - break; - memcpy(ks->mcast_lst[i++], ha->addr, ETH_ALEN); - } - ks->mcast_lst_size = (u8)i; - ks_set_grpaddr(ks); - } else { - /** - * List too big to support so - * turn on all mcast mode. - */ - ks->mcast_lst_size = MAX_MCAST_LST; - ks_set_mcast(ks, true); - } - } else { - ks->mcast_lst_size = 0; - ks_clear_mcast(ks); - } -} /* ks_set_rx_mode */ - -static void ks_set_mac(struct ks_net *ks, u8 *data) -{ - u16 *pw = (u16 *)data; - u16 w, u; - - ks_stop_rx(ks); /* Stop receiving for reconfiguration */ - - u = *pw++; - w = ((u & 0xFF) << 8) | ((u >> 8) & 0xFF); - ks_wrreg16(ks, KS_MARH, w); - - u = *pw++; - w = ((u & 0xFF) << 8) | ((u >> 8) & 0xFF); - ks_wrreg16(ks, KS_MARM, w); - - u = *pw; - w = ((u & 0xFF) << 8) | ((u >> 8) & 0xFF); - ks_wrreg16(ks, KS_MARL, w); - - memcpy(ks->mac_addr, data, ETH_ALEN); - - if (ks->enabled) - ks_start_rx(ks); -} - -static int ks_set_mac_address(struct net_device *netdev, void *paddr) -{ - struct ks_net *ks = netdev_priv(netdev); - struct sockaddr *addr = paddr; - u8 *da; - - memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len); - - da = (u8 *)netdev->dev_addr; - - ks_set_mac(ks, da); - return 0; -} - -static int ks_net_ioctl(struct net_device *netdev, struct ifreq *req, int cmd) -{ - struct ks_net *ks = netdev_priv(netdev); - - if (!netif_running(netdev)) - return -EINVAL; - - return generic_mii_ioctl(&ks->mii, if_mii(req), cmd, NULL); -} - -static const struct net_device_ops ks_netdev_ops = { - .ndo_open = ks_net_open, - .ndo_stop = ks_net_stop, - .ndo_do_ioctl = ks_net_ioctl, - .ndo_start_xmit = ks_start_xmit, - .ndo_set_mac_address = ks_set_mac_address, - .ndo_set_rx_mode = ks_set_rx_mode, - .ndo_validate_addr = eth_validate_addr, -}; - -/* ethtool support */ - -static void ks_get_drvinfo(struct net_device *netdev, - struct ethtool_drvinfo *di) -{ - strlcpy(di->driver, DRV_NAME, sizeof(di->driver)); - strlcpy(di->version, "1.00", sizeof(di->version)); - strlcpy(di->bus_info, dev_name(netdev->dev.parent), - sizeof(di->bus_info)); -} - -static u32 ks_get_msglevel(struct net_device *netdev) -{ - struct ks_net *ks = netdev_priv(netdev); - return ks->msg_enable; -} - -static void ks_set_msglevel(struct net_device *netdev, u32 to) -{ - struct ks_net *ks = netdev_priv(netdev); - ks->msg_enable = to; -} - -static int ks_get_link_ksettings(struct net_device *netdev, - struct ethtool_link_ksettings *cmd) -{ - struct ks_net *ks = netdev_priv(netdev); - - mii_ethtool_get_link_ksettings(&ks->mii, cmd); - - return 0; -} - -static int ks_set_link_ksettings(struct net_device *netdev, - const struct ethtool_link_ksettings *cmd) -{ - struct ks_net *ks = netdev_priv(netdev); - return mii_ethtool_set_link_ksettings(&ks->mii, cmd); -} - -static u32 ks_get_link(struct net_device *netdev) -{ - struct ks_net *ks = netdev_priv(netdev); - return mii_link_ok(&ks->mii); -} - -static int ks_nway_reset(struct net_device *netdev) -{ - struct ks_net *ks = netdev_priv(netdev); - return mii_nway_restart(&ks->mii); -} - -static const struct ethtool_ops ks_ethtool_ops = { - .get_drvinfo = ks_get_drvinfo, - .get_msglevel = ks_get_msglevel, - .set_msglevel = ks_set_msglevel, - .get_link = ks_get_link, - .nway_reset = ks_nway_reset, - .get_link_ksettings = ks_get_link_ksettings, - .set_link_ksettings = ks_set_link_ksettings, -}; - -/* MII interface controls */ - -/** - * ks_phy_reg - convert MII register into a KS8851 register - * @reg: MII register number. - * - * Return the KS8851 register number for the corresponding MII PHY register - * if possible. Return zero if the MII register has no direct mapping to the - * KS8851 register set. - */ -static int ks_phy_reg(int reg) -{ - switch (reg) { - case MII_BMCR: - return KS_P1MBCR; - case MII_BMSR: - return KS_P1MBSR; - case MII_PHYSID1: - return KS_PHY1ILR; - case MII_PHYSID2: - return KS_PHY1IHR; - case MII_ADVERTISE: - return KS_P1ANAR; - case MII_LPA: - return KS_P1ANLPR; - } - - return 0x0; -} - -/** - * ks_phy_read - MII interface PHY register read. - * @netdev: The network device the PHY is on. - * @phy_addr: Address of PHY (ignored as we only have one) - * @reg: The register to read. - * - * This call reads data from the PHY register specified in @reg. Since the - * device does not support all the MII registers, the non-existent values - * are always returned as zero. - * - * We return zero for unsupported registers as the MII code does not check - * the value returned for any error status, and simply returns it to the - * caller. The mii-tool that the driver was tested with takes any -ve error - * as real PHY capabilities, thus displaying incorrect data to the user. - */ -static int ks_phy_read(struct net_device *netdev, int phy_addr, int reg) -{ - struct ks_net *ks = netdev_priv(netdev); - int ksreg; - int result; - - ksreg = ks_phy_reg(reg); - if (!ksreg) - return 0x0; /* no error return allowed, so use zero */ - - mutex_lock(&ks->lock); - result = ks_rdreg16(ks, ksreg); - mutex_unlock(&ks->lock); - - return result; -} - -static void ks_phy_write(struct net_device *netdev, - int phy, int reg, int value) -{ - struct ks_net *ks = netdev_priv(netdev); - int ksreg; - - ksreg = ks_phy_reg(reg); - if (ksreg) { - mutex_lock(&ks->lock); - ks_wrreg16(ks, ksreg, value); - mutex_unlock(&ks->lock); - } -} - -/** - * ks_read_selftest - read the selftest memory info. - * @ks: The device state - * - * Read and check the TX/RX memory selftest information. - */ -static int ks_read_selftest(struct ks_net *ks) -{ - unsigned both_done = MBIR_TXMBF | MBIR_RXMBF; - int ret = 0; - unsigned rd; - - rd = ks_rdreg16(ks, KS_MBIR); - - if ((rd & both_done) != both_done) { - netdev_warn(ks->netdev, "Memory selftest not finished\n"); - return 0; - } - - if (rd & MBIR_TXMBFA) { - netdev_err(ks->netdev, "TX memory selftest fails\n"); - ret |= 1; - } - - if (rd & MBIR_RXMBFA) { - netdev_err(ks->netdev, "RX memory selftest fails\n"); - ret |= 2; - } - - netdev_info(ks->netdev, "the selftest passes\n"); - return ret; -} - -static void ks_setup(struct ks_net *ks) -{ - u16 w; - - /** - * Configure QMU Transmit - */ - - /* Setup Transmit Frame Data Pointer Auto-Increment (TXFDPR) */ - ks_wrreg16(ks, KS_TXFDPR, TXFDPR_TXFPAI); - - /* Setup Receive Frame Data Pointer Auto-Increment */ - ks_wrreg16(ks, KS_RXFDPR, RXFDPR_RXFPAI); - - /* Setup Receive Frame Threshold - 1 frame (RXFCTFC) */ - ks_wrreg16(ks, KS_RXFCTR, 1 & RXFCTR_RXFCT_MASK); - - /* Setup RxQ Command Control (RXQCR) */ - ks->rc_rxqcr = RXQCR_CMD_CNTL; - ks_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr); - - /** - * set the force mode to half duplex, default is full duplex - * because if the auto-negotiation fails, most switch uses - * half-duplex. - */ - - w = ks_rdreg16(ks, KS_P1MBCR); - w &= ~BMCR_FULLDPLX; - ks_wrreg16(ks, KS_P1MBCR, w); - - w = TXCR_TXFCE | TXCR_TXPE | TXCR_TXCRC | TXCR_TCGIP; - ks_wrreg16(ks, KS_TXCR, w); - - w = RXCR1_RXFCE | RXCR1_RXBE | RXCR1_RXUE | RXCR1_RXME | RXCR1_RXIPFCC; - - if (ks->promiscuous) /* bPromiscuous */ - w |= (RXCR1_RXAE | RXCR1_RXINVF); - else if (ks->all_mcast) /* Multicast address passed mode */ - w |= (RXCR1_RXAE | RXCR1_RXMAFMA | RXCR1_RXPAFMA); - else /* Normal mode */ - w |= RXCR1_RXPAFMA; - - ks_wrreg16(ks, KS_RXCR1, w); -} /*ks_setup */ - - -static void ks_setup_int(struct ks_net *ks) -{ - ks->rc_ier = 0x00; - /* Clear the interrupts status of the hardware. */ - ks_wrreg16(ks, KS_ISR, 0xffff); - - /* Enables the interrupts of the hardware. */ - ks->rc_ier = (IRQ_LCI | IRQ_TXI | IRQ_RXI); -} /* ks_setup_int */ - -static int ks_hw_init(struct ks_net *ks) -{ -#define MHEADER_SIZE (sizeof(struct type_frame_head) * MAX_RECV_FRAMES) - ks->promiscuous = 0; - ks->all_mcast = 0; - ks->mcast_lst_size = 0; - - ks->frame_head_info = devm_kmalloc(&ks->pdev->dev, MHEADER_SIZE, - GFP_KERNEL); - if (!ks->frame_head_info) - return false; - - ks_set_mac(ks, KS_DEFAULT_MAC_ADDRESS); - return true; -} - -#if defined(CONFIG_OF) -static const struct of_device_id ks8851_ml_dt_ids[] = { - { .compatible = "micrel,ks8851-mll" }, - { /* sentinel */ } -}; -MODULE_DEVICE_TABLE(of, ks8851_ml_dt_ids); -#endif - -static int ks8851_probe(struct platform_device *pdev) -{ - int err; - struct net_device *netdev; - struct ks_net *ks; - u16 id, data; - const char *mac; - - netdev = alloc_etherdev(sizeof(struct ks_net)); - if (!netdev) - return -ENOMEM; - - SET_NETDEV_DEV(netdev, &pdev->dev); - - ks = netdev_priv(netdev); - ks->netdev = netdev; - - ks->hw_addr = devm_platform_ioremap_resource(pdev, 0); - if (IS_ERR(ks->hw_addr)) { - err = PTR_ERR(ks->hw_addr); - goto err_free; - } - - ks->hw_addr_cmd = devm_platform_ioremap_resource(pdev, 1); - if (IS_ERR(ks->hw_addr_cmd)) { - err = PTR_ERR(ks->hw_addr_cmd); - goto err_free; - } - - err = ks_check_endian(ks); - if (err) - goto err_free; - - netdev->irq = platform_get_irq(pdev, 0); - - if ((int)netdev->irq < 0) { - err = netdev->irq; - goto err_free; - } - - ks->pdev = pdev; - - mutex_init(&ks->lock); - spin_lock_init(&ks->statelock); - - netdev->netdev_ops = &ks_netdev_ops; - netdev->ethtool_ops = &ks_ethtool_ops; - - /* setup mii state */ - ks->mii.dev = netdev; - ks->mii.phy_id = 1, - ks->mii.phy_id_mask = 1; - ks->mii.reg_num_mask = 0xf; - ks->mii.mdio_read = ks_phy_read; - ks->mii.mdio_write = ks_phy_write; - - netdev_info(netdev, "message enable is %d\n", msg_enable); - /* set the default message enable */ - ks->msg_enable = netif_msg_init(msg_enable, (NETIF_MSG_DRV | - NETIF_MSG_PROBE | - NETIF_MSG_LINK)); - ks_read_config(ks); - - /* simple check for a valid chip being connected to the bus */ - if ((ks_rdreg16(ks, KS_CIDER) & ~CIDER_REV_MASK) != CIDER_ID) { - netdev_err(netdev, "failed to read device ID\n"); - err = -ENODEV; - goto err_free; - } - - if (ks_read_selftest(ks)) { - netdev_err(netdev, "failed to read device ID\n"); - err = -ENODEV; - goto err_free; - } - - err = register_netdev(netdev); - if (err) - goto err_free; - - platform_set_drvdata(pdev, netdev); - - ks_soft_reset(ks, GRR_GSR); - ks_hw_init(ks); - ks_disable_qmu(ks); - ks_setup(ks); - ks_setup_int(ks); - - data = ks_rdreg16(ks, KS_OBCR); - ks_wrreg16(ks, KS_OBCR, data | OBCR_ODS_16mA); - - /* overwriting the default MAC address */ - if (pdev->dev.of_node) { - mac = of_get_mac_address(pdev->dev.of_node); - if (!IS_ERR(mac)) - ether_addr_copy(ks->mac_addr, mac); - } else { - struct ks8851_mll_platform_data *pdata; - - pdata = dev_get_platdata(&pdev->dev); - if (!pdata) { - netdev_err(netdev, "No platform data\n"); - err = -ENODEV; - goto err_pdata; - } - memcpy(ks->mac_addr, pdata->mac_addr, ETH_ALEN); - } - if (!is_valid_ether_addr(ks->mac_addr)) { - /* Use random MAC address if none passed */ - eth_random_addr(ks->mac_addr); - netdev_info(netdev, "Using random mac address\n"); - } - netdev_info(netdev, "Mac address is: %pM\n", ks->mac_addr); - - memcpy(netdev->dev_addr, ks->mac_addr, ETH_ALEN); - - ks_set_mac(ks, netdev->dev_addr); - - id = ks_rdreg16(ks, KS_CIDER); - - netdev_info(netdev, "Found chip, family: 0x%x, id: 0x%x, rev: 0x%x\n", - (id >> 8) & 0xff, (id >> 4) & 0xf, (id >> 1) & 0x7); - return 0; - -err_pdata: - unregister_netdev(netdev); -err_free: - free_netdev(netdev); - return err; -} - -static int ks8851_remove(struct platform_device *pdev) -{ - struct net_device *netdev = platform_get_drvdata(pdev); - - unregister_netdev(netdev); - free_netdev(netdev); - return 0; - -} - -static struct platform_driver ks8851_platform_driver = { - .driver = { - .name = DRV_NAME, - .of_match_table = of_match_ptr(ks8851_ml_dt_ids), - }, - .probe = ks8851_probe, - .remove = ks8851_remove, -}; - -module_platform_driver(ks8851_platform_driver); - -MODULE_DESCRIPTION("KS8851 MLL Network driver"); -MODULE_AUTHOR("David Choi <david.choi@micrel.com>"); -MODULE_LICENSE("GPL"); -module_param_named(message, msg_enable, int, 0); -MODULE_PARM_DESC(message, "Message verbosity level (0=none, 31=all)"); - diff --git a/drivers/net/ethernet/micrel/ks8851_par.c b/drivers/net/ethernet/micrel/ks8851_par.c new file mode 100644 index 000000000000..3bab0cb2b1a5 --- /dev/null +++ b/drivers/net/ethernet/micrel/ks8851_par.c @@ -0,0 +1,357 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* drivers/net/ethernet/micrel/ks8851.c + * + * Copyright 2009 Simtec Electronics + * http://www.simtec.co.uk/ + * Ben Dooks <ben@simtec.co.uk> + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#define DEBUG + +#include <linux/interrupt.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> +#include <linux/ethtool.h> +#include <linux/iopoll.h> +#include <linux/mii.h> + +#include <linux/platform_device.h> +#include <linux/of_net.h> + +#include "ks8851.h" + +static int msg_enable; + +#define BE3 0x8000 /* Byte Enable 3 */ +#define BE2 0x4000 /* Byte Enable 2 */ +#define BE1 0x2000 /* Byte Enable 1 */ +#define BE0 0x1000 /* Byte Enable 0 */ + +/** + * struct ks8851_net_par - KS8851 Parallel driver private data + * @ks8851: KS8851 driver common private data + * @lock: Lock to ensure that the device is not accessed when busy. + * @hw_addr : start address of data register. + * @hw_addr_cmd : start address of command register. + * @cmd_reg_cache : command register cached. + * + * The @lock ensures that the chip is protected when certain operations are + * in progress. When the read or write packet transfer is in progress, most + * of the chip registers are not accessible until the transfer is finished + * and the DMA has been de-asserted. + */ +struct ks8851_net_par { + struct ks8851_net ks8851; + spinlock_t lock; + void __iomem *hw_addr; + void __iomem *hw_addr_cmd; + u16 cmd_reg_cache; +}; + +#define to_ks8851_par(ks) container_of((ks), struct ks8851_net_par, ks8851) + +/** + * ks8851_lock_par - register access lock + * @ks: The chip state + * @flags: Spinlock flags + * + * Claim chip register access lock + */ +static void ks8851_lock_par(struct ks8851_net *ks, unsigned long *flags) +{ + struct ks8851_net_par *ksp = to_ks8851_par(ks); + + spin_lock_irqsave(&ksp->lock, *flags); +} + +/** + * ks8851_unlock_par - register access unlock + * @ks: The chip state + * @flags: Spinlock flags + * + * Release chip register access lock + */ +static void ks8851_unlock_par(struct ks8851_net *ks, unsigned long *flags) +{ + struct ks8851_net_par *ksp = to_ks8851_par(ks); + + spin_unlock_irqrestore(&ksp->lock, *flags); +} + +/** + * ks_check_endian - Check whether endianness of the bus is correct + * @ks : The chip information + * + * The KS8851-16MLL EESK pin allows selecting the endianness of the 16bit + * bus. To maintain optimum performance, the bus endianness should be set + * such that it matches the endianness of the CPU. + */ +static int ks_check_endian(struct ks8851_net *ks) +{ + struct ks8851_net_par *ksp = to_ks8851_par(ks); + u16 cider; + + /* + * Read CIDER register first, however read it the "wrong" way around. + * If the endian strap on the KS8851-16MLL in incorrect and the chip + * is operating in different endianness than the CPU, then the meaning + * of BE[3:0] byte-enable bits is also swapped such that: + * BE[3,2,1,0] becomes BE[1,0,3,2] + * + * Luckily for us, the byte-enable bits are the top four MSbits of + * the address register and the CIDER register is at offset 0xc0. + * Hence, by reading address 0xc0c0, which is not impacted by endian + * swapping, we assert either BE[3:2] or BE[1:0] while reading the + * CIDER register. + * + * If the bus configuration is correct, reading 0xc0c0 asserts + * BE[3:2] and this read returns 0x0000, because to read register + * with bottom two LSbits of address set to 0, BE[1:0] must be + * asserted. + * + * If the bus configuration is NOT correct, reading 0xc0c0 asserts + * BE[1:0] and this read returns non-zero 0x8872 value. + */ + iowrite16(BE3 | BE2 | KS_CIDER, ksp->hw_addr_cmd); + cider = ioread16(ksp->hw_addr); + if (!cider) + return 0; + + netdev_err(ks->netdev, "incorrect EESK endian strap setting\n"); + + return -EINVAL; +} + +/** + * ks8851_wrreg16_par - write 16bit register value to chip + * @ks: The chip state + * @reg: The register address + * @val: The value to write + * + * Issue a write to put the value @val into the register specified in @reg. + */ +static void ks8851_wrreg16_par(struct ks8851_net *ks, unsigned int reg, + unsigned int val) +{ + struct ks8851_net_par *ksp = to_ks8851_par(ks); + + ksp->cmd_reg_cache = (u16)reg | ((BE1 | BE0) << (reg & 0x02)); + iowrite16(ksp->cmd_reg_cache, ksp->hw_addr_cmd); + iowrite16(val, ksp->hw_addr); +} + +/** + * ks8851_rdreg16_par - read 16 bit register from chip + * @ks: The chip information + * @reg: The register address + * + * Read a 16bit register from the chip, returning the result + */ +static unsigned int ks8851_rdreg16_par(struct ks8851_net *ks, unsigned int reg) +{ + struct ks8851_net_par *ksp = to_ks8851_par(ks); + + ksp->cmd_reg_cache = (u16)reg | ((BE1 | BE0) << (reg & 0x02)); + iowrite16(ksp->cmd_reg_cache, ksp->hw_addr_cmd); + return ioread16(ksp->hw_addr); +} + +/** + * ks8851_rdfifo_par - read data from the receive fifo + * @ks: The device state. + * @buff: The buffer address + * @len: The length of the data to read + * + * Issue an RXQ FIFO read command and read the @len amount of data from + * the FIFO into the buffer specified by @buff. + */ +static void ks8851_rdfifo_par(struct ks8851_net *ks, u8 *buff, unsigned int len) +{ + struct ks8851_net_par *ksp = to_ks8851_par(ks); + + netif_dbg(ks, rx_status, ks->netdev, + "%s: %d@%p\n", __func__, len, buff); + + ioread16_rep(ksp->hw_addr, (u16 *)buff + 1, len / 2); +} + +/** + * ks8851_wrfifo_par - write packet to TX FIFO + * @ks: The device state. + * @txp: The sk_buff to transmit. + * @irq: IRQ on completion of the packet. + * + * Send the @txp to the chip. This means creating the relevant packet header + * specifying the length of the packet and the other information the chip + * needs, such as IRQ on completion. Send the header and the packet data to + * the device. + */ +static void ks8851_wrfifo_par(struct ks8851_net *ks, struct sk_buff *txp, + bool irq) +{ + struct ks8851_net_par *ksp = to_ks8851_par(ks); + unsigned int len = ALIGN(txp->len, 4); + unsigned int fid = 0; + + netif_dbg(ks, tx_queued, ks->netdev, "%s: skb %p, %d@%p, irq %d\n", + __func__, txp, txp->len, txp->data, irq); + + fid = ks->fid++; + fid &= TXFR_TXFID_MASK; + + if (irq) + fid |= TXFR_TXIC; /* irq on completion */ + + iowrite16(fid, ksp->hw_addr); + iowrite16(txp->len, ksp->hw_addr); + + iowrite16_rep(ksp->hw_addr, txp->data, len / 2); +} + +/** + * ks8851_rx_skb_par - receive skbuff + * @ks: The device state. + * @skb: The skbuff + */ +static void ks8851_rx_skb_par(struct ks8851_net *ks, struct sk_buff *skb) +{ + netif_rx(skb); +} + +static unsigned int ks8851_rdreg16_par_txqcr(struct ks8851_net *ks) +{ + return ks8851_rdreg16_par(ks, KS_TXQCR); +} + +/** + * ks8851_start_xmit_par - transmit packet + * @skb: The buffer to transmit + * @dev: The device used to transmit the packet. + * + * Called by the network layer to transmit the @skb. Queue the packet for + * the device and schedule the necessary work to transmit the packet when + * it is free. + * + * We do this to firstly avoid sleeping with the network device locked, + * and secondly so we can round up more than one packet to transmit which + * means we can try and avoid generating too many transmit done interrupts. + */ +static netdev_tx_t ks8851_start_xmit_par(struct sk_buff *skb, + struct net_device *dev) +{ + struct ks8851_net *ks = netdev_priv(dev); + netdev_tx_t ret = NETDEV_TX_OK; + unsigned long flags; + unsigned int txqcr; + u16 txmir; + int err; + + netif_dbg(ks, tx_queued, ks->netdev, + "%s: skb %p, %d@%p\n", __func__, skb, skb->len, skb->data); + + ks8851_lock_par(ks, &flags); + + txmir = ks8851_rdreg16_par(ks, KS_TXMIR) & 0x1fff; + + if (likely(txmir >= skb->len + 12)) { + ks8851_wrreg16_par(ks, KS_RXQCR, ks->rc_rxqcr | RXQCR_SDA); + ks8851_wrfifo_par(ks, skb, false); + ks8851_wrreg16_par(ks, KS_RXQCR, ks->rc_rxqcr); + ks8851_wrreg16_par(ks, KS_TXQCR, TXQCR_METFE); + + err = readx_poll_timeout_atomic(ks8851_rdreg16_par_txqcr, ks, + txqcr, !(txqcr & TXQCR_METFE), + 5, 1000000); + if (err) + ret = NETDEV_TX_BUSY; + + ks8851_done_tx(ks, skb); + } else { + ret = NETDEV_TX_BUSY; + } + + ks8851_unlock_par(ks, &flags); + + return ret; +} + +static int ks8851_probe_par(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct ks8851_net_par *ksp; + struct net_device *netdev; + struct ks8851_net *ks; + int ret; + + netdev = devm_alloc_etherdev(dev, sizeof(struct ks8851_net_par)); + if (!netdev) + return -ENOMEM; + + ks = netdev_priv(netdev); + + ks->lock = ks8851_lock_par; + ks->unlock = ks8851_unlock_par; + ks->rdreg16 = ks8851_rdreg16_par; + ks->wrreg16 = ks8851_wrreg16_par; + ks->rdfifo = ks8851_rdfifo_par; + ks->wrfifo = ks8851_wrfifo_par; + ks->start_xmit = ks8851_start_xmit_par; + ks->rx_skb = ks8851_rx_skb_par; + +#define STD_IRQ (IRQ_LCI | /* Link Change */ \ + IRQ_RXI | /* RX done */ \ + IRQ_RXPSI) /* RX process stop */ + ks->rc_ier = STD_IRQ; + + ksp = to_ks8851_par(ks); + spin_lock_init(&ksp->lock); + + ksp->hw_addr = devm_platform_ioremap_resource(pdev, 0); + if (IS_ERR(ksp->hw_addr)) + return PTR_ERR(ksp->hw_addr); + + ksp->hw_addr_cmd = devm_platform_ioremap_resource(pdev, 1); + if (IS_ERR(ksp->hw_addr_cmd)) + return PTR_ERR(ksp->hw_addr_cmd); + + ret = ks_check_endian(ks); + if (ret) + return ret; + + netdev->irq = platform_get_irq(pdev, 0); + + return ks8851_probe_common(netdev, dev, msg_enable); +} + +static int ks8851_remove_par(struct platform_device *pdev) +{ + return ks8851_remove_common(&pdev->dev); +} + +static const struct of_device_id ks8851_match_table[] = { + { .compatible = "micrel,ks8851-mll" }, + { } +}; +MODULE_DEVICE_TABLE(of, ks8851_match_table); + +static struct platform_driver ks8851_driver = { + .driver = { + .name = "ks8851", + .of_match_table = ks8851_match_table, + .pm = &ks8851_pm_ops, + }, + .probe = ks8851_probe_par, + .remove = ks8851_remove_par, +}; +module_platform_driver(ks8851_driver); + +MODULE_DESCRIPTION("KS8851 Network driver"); +MODULE_AUTHOR("Ben Dooks <ben@simtec.co.uk>"); +MODULE_LICENSE("GPL"); + +module_param_named(message, msg_enable, int, 0); +MODULE_PARM_DESC(message, "Message verbosity level (0=none, 31=all)"); diff --git a/drivers/net/ethernet/micrel/ks8851_spi.c b/drivers/net/ethernet/micrel/ks8851_spi.c new file mode 100644 index 000000000000..4ec7f1615977 --- /dev/null +++ b/drivers/net/ethernet/micrel/ks8851_spi.c @@ -0,0 +1,485 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* drivers/net/ethernet/micrel/ks8851.c + * + * Copyright 2009 Simtec Electronics + * http://www.simtec.co.uk/ + * Ben Dooks <ben@simtec.co.uk> + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#define DEBUG + +#include <linux/interrupt.h> +#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> +#include <linux/ethtool.h> +#include <linux/cache.h> +#include <linux/crc32.h> +#include <linux/mii.h> +#include <linux/regulator/consumer.h> + +#include <linux/spi/spi.h> +#include <linux/gpio.h> +#include <linux/of_gpio.h> +#include <linux/of_net.h> + +#include "ks8851.h" + +static int msg_enable; + +/** + * struct ks8851_net_spi - KS8851 SPI driver private data + * @lock: Lock to ensure that the device is not accessed when busy. + * @tx_work: Work queue for tx packets + * @ks8851: KS8851 driver common private data + * @spidev: The spi device we're bound to. + * @spi_msg1: pre-setup SPI transfer with one message, @spi_xfer1. + * @spi_msg2: pre-setup SPI transfer with two messages, @spi_xfer2. + * @spi_xfer1: @spi_msg1 SPI transfer structure + * @spi_xfer2: @spi_msg2 SPI transfer structure + * + * The @lock ensures that the chip is protected when certain operations are + * in progress. When the read or write packet transfer is in progress, most + * of the chip registers are not ccessible until the transfer is finished and + * the DMA has been de-asserted. + */ +struct ks8851_net_spi { + struct ks8851_net ks8851; + struct mutex lock; + struct work_struct tx_work; + struct spi_device *spidev; + struct spi_message spi_msg1; + struct spi_message spi_msg2; + struct spi_transfer spi_xfer1; + struct spi_transfer spi_xfer2[2]; +}; + +#define to_ks8851_spi(ks) container_of((ks), struct ks8851_net_spi, ks8851) + +/* SPI frame opcodes */ +#define KS_SPIOP_RD 0x00 +#define KS_SPIOP_WR 0x40 +#define KS_SPIOP_RXFIFO 0x80 +#define KS_SPIOP_TXFIFO 0xC0 + +/* shift for byte-enable data */ +#define BYTE_EN(_x) ((_x) << 2) + +/* turn register number and byte-enable mask into data for start of packet */ +#define MK_OP(_byteen, _reg) \ + (BYTE_EN(_byteen) | (_reg) << (8 + 2) | (_reg) >> 6) + +/** + * ks8851_lock_spi - register access lock + * @ks: The chip state + * @flags: Spinlock flags + * + * Claim chip register access lock + */ +static void ks8851_lock_spi(struct ks8851_net *ks, unsigned long *flags) +{ + struct ks8851_net_spi *kss = to_ks8851_spi(ks); + + mutex_lock(&kss->lock); +} + +/** + * ks8851_unlock_spi - register access unlock + * @ks: The chip state + * @flags: Spinlock flags + * + * Release chip register access lock + */ +static void ks8851_unlock_spi(struct ks8851_net *ks, unsigned long *flags) +{ + struct ks8851_net_spi *kss = to_ks8851_spi(ks); + + mutex_unlock(&kss->lock); +} + +/* SPI register read/write calls. + * + * All these calls issue SPI transactions to access the chip's registers. They + * all require that the necessary lock is held to prevent accesses when the + * chip is busy transferring packet data (RX/TX FIFO accesses). + */ + +/** + * ks8851_wrreg16_spi - write 16bit register value to chip via SPI + * @ks: The chip state + * @reg: The register address + * @val: The value to write + * + * Issue a write to put the value @val into the register specified in @reg. + */ +static void ks8851_wrreg16_spi(struct ks8851_net *ks, unsigned int reg, + unsigned int val) +{ + struct ks8851_net_spi *kss = to_ks8851_spi(ks); + struct spi_transfer *xfer = &kss->spi_xfer1; + struct spi_message *msg = &kss->spi_msg1; + __le16 txb[2]; + int ret; + + txb[0] = cpu_to_le16(MK_OP(reg & 2 ? 0xC : 0x03, reg) | KS_SPIOP_WR); + txb[1] = cpu_to_le16(val); + + xfer->tx_buf = txb; + xfer->rx_buf = NULL; + xfer->len = 4; + + ret = spi_sync(kss->spidev, msg); + if (ret < 0) + netdev_err(ks->netdev, "spi_sync() failed\n"); +} + +/** + * ks8851_rdreg - issue read register command and return the data + * @ks: The device state + * @op: The register address and byte enables in message format. + * @rxb: The RX buffer to return the result into + * @rxl: The length of data expected. + * + * This is the low level read call that issues the necessary spi message(s) + * to read data from the register specified in @op. + */ +static void ks8851_rdreg(struct ks8851_net *ks, unsigned int op, + u8 *rxb, unsigned int rxl) +{ + struct ks8851_net_spi *kss = to_ks8851_spi(ks); + struct spi_transfer *xfer; + struct spi_message *msg; + __le16 *txb = (__le16 *)ks->txd; + u8 *trx = ks->rxd; + int ret; + + txb[0] = cpu_to_le16(op | KS_SPIOP_RD); + + if (kss->spidev->master->flags & SPI_MASTER_HALF_DUPLEX) { + msg = &kss->spi_msg2; + xfer = kss->spi_xfer2; + + xfer->tx_buf = txb; + xfer->rx_buf = NULL; + xfer->len = 2; + + xfer++; + xfer->tx_buf = NULL; + xfer->rx_buf = trx; + xfer->len = rxl; + } else { + msg = &kss->spi_msg1; + xfer = &kss->spi_xfer1; + + xfer->tx_buf = txb; + xfer->rx_buf = trx; + xfer->len = rxl + 2; + } + + ret = spi_sync(kss->spidev, msg); + if (ret < 0) + netdev_err(ks->netdev, "read: spi_sync() failed\n"); + else if (kss->spidev->master->flags & SPI_MASTER_HALF_DUPLEX) + memcpy(rxb, trx, rxl); + else + memcpy(rxb, trx + 2, rxl); +} + +/** + * ks8851_rdreg16_spi - read 16 bit register from device via SPI + * @ks: The chip information + * @reg: The register address + * + * Read a 16bit register from the chip, returning the result + */ +static unsigned int ks8851_rdreg16_spi(struct ks8851_net *ks, unsigned int reg) +{ + __le16 rx = 0; + + ks8851_rdreg(ks, MK_OP(reg & 2 ? 0xC : 0x3, reg), (u8 *)&rx, 2); + return le16_to_cpu(rx); +} + +/** + * ks8851_rdfifo_spi - read data from the receive fifo via SPI + * @ks: The device state. + * @buff: The buffer address + * @len: The length of the data to read + * + * Issue an RXQ FIFO read command and read the @len amount of data from + * the FIFO into the buffer specified by @buff. + */ +static void ks8851_rdfifo_spi(struct ks8851_net *ks, u8 *buff, unsigned int len) +{ + struct ks8851_net_spi *kss = to_ks8851_spi(ks); + struct spi_transfer *xfer = kss->spi_xfer2; + struct spi_message *msg = &kss->spi_msg2; + u8 txb[1]; + int ret; + + netif_dbg(ks, rx_status, ks->netdev, + "%s: %d@%p\n", __func__, len, buff); + + /* set the operation we're issuing */ + txb[0] = KS_SPIOP_RXFIFO; + + xfer->tx_buf = txb; + xfer->rx_buf = NULL; + xfer->len = 1; + + xfer++; + xfer->rx_buf = buff; + xfer->tx_buf = NULL; + xfer->len = len; + + ret = spi_sync(kss->spidev, msg); + if (ret < 0) + netdev_err(ks->netdev, "%s: spi_sync() failed\n", __func__); +} + +/** + * ks8851_wrfifo_spi - write packet to TX FIFO via SPI + * @ks: The device state. + * @txp: The sk_buff to transmit. + * @irq: IRQ on completion of the packet. + * + * Send the @txp to the chip. This means creating the relevant packet header + * specifying the length of the packet and the other information the chip + * needs, such as IRQ on completion. Send the header and the packet data to + * the device. + */ +static void ks8851_wrfifo_spi(struct ks8851_net *ks, struct sk_buff *txp, + bool irq) +{ + struct ks8851_net_spi *kss = to_ks8851_spi(ks); + struct spi_transfer *xfer = kss->spi_xfer2; + struct spi_message *msg = &kss->spi_msg2; + unsigned int fid = 0; + int ret; + + netif_dbg(ks, tx_queued, ks->netdev, "%s: skb %p, %d@%p, irq %d\n", + __func__, txp, txp->len, txp->data, irq); + + fid = ks->fid++; + fid &= TXFR_TXFID_MASK; + + if (irq) + fid |= TXFR_TXIC; /* irq on completion */ + + /* start header at txb[1] to align txw entries */ + ks->txh.txb[1] = KS_SPIOP_TXFIFO; + ks->txh.txw[1] = cpu_to_le16(fid); + ks->txh.txw[2] = cpu_to_le16(txp->len); + + xfer->tx_buf = &ks->txh.txb[1]; + xfer->rx_buf = NULL; + xfer->len = 5; + + xfer++; + xfer->tx_buf = txp->data; + xfer->rx_buf = NULL; + xfer->len = ALIGN(txp->len, 4); + + ret = spi_sync(kss->spidev, msg); + if (ret < 0) + netdev_err(ks->netdev, "%s: spi_sync() failed\n", __func__); +} + +/** + * ks8851_rx_skb_spi - receive skbuff + * @ks: The device state + * @skb: The skbuff + */ +static void ks8851_rx_skb_spi(struct ks8851_net *ks, struct sk_buff *skb) +{ + netif_rx_ni(skb); +} + +/** + * ks8851_tx_work - process tx packet(s) + * @work: The work strucutre what was scheduled. + * + * This is called when a number of packets have been scheduled for + * transmission and need to be sent to the device. + */ +static void ks8851_tx_work(struct work_struct *work) +{ + struct ks8851_net_spi *kss; + struct ks8851_net *ks; + unsigned long flags; + struct sk_buff *txb; + bool last; + + kss = container_of(work, struct ks8851_net_spi, tx_work); + ks = &kss->ks8851; + last = skb_queue_empty(&ks->txq); + + ks8851_lock_spi(ks, &flags); + + while (!last) { + txb = skb_dequeue(&ks->txq); + last = skb_queue_empty(&ks->txq); + + if (txb) { + ks8851_wrreg16_spi(ks, KS_RXQCR, + ks->rc_rxqcr | RXQCR_SDA); + ks8851_wrfifo_spi(ks, txb, last); + ks8851_wrreg16_spi(ks, KS_RXQCR, ks->rc_rxqcr); + ks8851_wrreg16_spi(ks, KS_TXQCR, TXQCR_METFE); + + ks8851_done_tx(ks, txb); + } + } + + ks8851_unlock_spi(ks, &flags); +} + +/** + * ks8851_flush_tx_work_spi - flush outstanding TX work + * @ks: The device state + */ +static void ks8851_flush_tx_work_spi(struct ks8851_net *ks) +{ + struct ks8851_net_spi *kss = to_ks8851_spi(ks); + + flush_work(&kss->tx_work); +} + +/** + * calc_txlen - calculate size of message to send packet + * @len: Length of data + * + * Returns the size of the TXFIFO message needed to send + * this packet. + */ +static unsigned int calc_txlen(unsigned int len) +{ + return ALIGN(len + 4, 4); +} + +/** + * ks8851_start_xmit_spi - transmit packet using SPI + * @skb: The buffer to transmit + * @dev: The device used to transmit the packet. + * + * Called by the network layer to transmit the @skb. Queue the packet for + * the device and schedule the necessary work to transmit the packet when + * it is free. + * + * We do this to firstly avoid sleeping with the network device locked, + * and secondly so we can round up more than one packet to transmit which + * means we can try and avoid generating too many transmit done interrupts. + */ +static netdev_tx_t ks8851_start_xmit_spi(struct sk_buff *skb, + struct net_device *dev) +{ + unsigned int needed = calc_txlen(skb->len); + struct ks8851_net *ks = netdev_priv(dev); + netdev_tx_t ret = NETDEV_TX_OK; + struct ks8851_net_spi *kss; + + kss = to_ks8851_spi(ks); + + netif_dbg(ks, tx_queued, ks->netdev, + "%s: skb %p, %d@%p\n", __func__, skb, skb->len, skb->data); + + spin_lock(&ks->statelock); + + if (needed > ks->tx_space) { + netif_stop_queue(dev); + ret = NETDEV_TX_BUSY; + } else { + ks->tx_space -= needed; + skb_queue_tail(&ks->txq, skb); + } + + spin_unlock(&ks->statelock); + schedule_work(&kss->tx_work); + + return ret; +} + +static int ks8851_probe_spi(struct spi_device *spi) +{ + struct device *dev = &spi->dev; + struct ks8851_net_spi *kss; + struct net_device *netdev; + struct ks8851_net *ks; + + netdev = devm_alloc_etherdev(dev, sizeof(struct ks8851_net_spi)); + if (!netdev) + return -ENOMEM; + + spi->bits_per_word = 8; + + ks = netdev_priv(netdev); + + ks->lock = ks8851_lock_spi; + ks->unlock = ks8851_unlock_spi; + ks->rdreg16 = ks8851_rdreg16_spi; + ks->wrreg16 = ks8851_wrreg16_spi; + ks->rdfifo = ks8851_rdfifo_spi; + ks->wrfifo = ks8851_wrfifo_spi; + ks->start_xmit = ks8851_start_xmit_spi; + ks->rx_skb = ks8851_rx_skb_spi; + ks->flush_tx_work = ks8851_flush_tx_work_spi; + +#define STD_IRQ (IRQ_LCI | /* Link Change */ \ + IRQ_TXI | /* TX done */ \ + IRQ_RXI | /* RX done */ \ + IRQ_SPIBEI | /* SPI bus error */ \ + IRQ_TXPSI | /* TX process stop */ \ + IRQ_RXPSI) /* RX process stop */ + ks->rc_ier = STD_IRQ; + + kss = to_ks8851_spi(ks); + + kss->spidev = spi; + mutex_init(&kss->lock); + INIT_WORK(&kss->tx_work, ks8851_tx_work); + + /* initialise pre-made spi transfer messages */ + spi_message_init(&kss->spi_msg1); + spi_message_add_tail(&kss->spi_xfer1, &kss->spi_msg1); + + spi_message_init(&kss->spi_msg2); + spi_message_add_tail(&kss->spi_xfer2[0], &kss->spi_msg2); + spi_message_add_tail(&kss->spi_xfer2[1], &kss->spi_msg2); + + netdev->irq = spi->irq; + + return ks8851_probe_common(netdev, dev, msg_enable); +} + +static int ks8851_remove_spi(struct spi_device *spi) +{ + return ks8851_remove_common(&spi->dev); +} + +static const struct of_device_id ks8851_match_table[] = { + { .compatible = "micrel,ks8851" }, + { } +}; +MODULE_DEVICE_TABLE(of, ks8851_match_table); + +static struct spi_driver ks8851_driver = { + .driver = { + .name = "ks8851", + .of_match_table = ks8851_match_table, + .pm = &ks8851_pm_ops, + }, + .probe = ks8851_probe_spi, + .remove = ks8851_remove_spi, +}; +module_spi_driver(ks8851_driver); + +MODULE_DESCRIPTION("KS8851 Network driver"); +MODULE_AUTHOR("Ben Dooks <ben@simtec.co.uk>"); +MODULE_LICENSE("GPL"); + +module_param_named(message, msg_enable, int, 0); +MODULE_PARM_DESC(message, "Message verbosity level (0=none, 31=all)"); +MODULE_ALIAS("spi:ks8851"); diff --git a/drivers/net/ethernet/microchip/encx24j600-regmap.c b/drivers/net/ethernet/microchip/encx24j600-regmap.c index 1f496fac7033..5bd7fb917b7a 100644 --- a/drivers/net/ethernet/microchip/encx24j600-regmap.c +++ b/drivers/net/ethernet/microchip/encx24j600-regmap.c @@ -17,11 +17,6 @@ #include "encx24j600_hw.h" -static inline bool is_bits_set(int value, int mask) -{ - return (value & mask) == mask; -} - static int encx24j600_switch_bank(struct encx24j600_context *ctx, int bank) { diff --git a/drivers/net/ethernet/microchip/encx24j600.c b/drivers/net/ethernet/microchip/encx24j600.c index b25a13da900a..2c0dcd7acf3f 100644 --- a/drivers/net/ethernet/microchip/encx24j600.c +++ b/drivers/net/ethernet/microchip/encx24j600.c @@ -604,9 +604,8 @@ static void encx24j600_set_rxfilter_mode(struct encx24j600_priv *priv) } } -static int encx24j600_hw_init(struct encx24j600_priv *priv) +static void encx24j600_hw_init(struct encx24j600_priv *priv) { - int ret = 0; u16 macon2; priv->hw_enabled = false; @@ -649,8 +648,6 @@ static int encx24j600_hw_init(struct encx24j600_priv *priv) if (netif_msg_hw(priv)) encx24j600_dump_config(priv, "Hw is initialized"); - - return ret; } static void encx24j600_hw_enable(struct encx24j600_priv *priv) @@ -1042,12 +1039,7 @@ static int encx24j600_spi_probe(struct spi_device *spi) } /* Initialize the device HW to the consistent state */ - if (encx24j600_hw_init(priv)) { - netif_err(priv, probe, ndev, - DRV_NAME ": HW initialization error\n"); - ret = -EIO; - goto out_free; - } + encx24j600_hw_init(priv); kthread_init_worker(&priv->kworker); kthread_init_work(&priv->tx_work, encx24j600_tx_proc); diff --git a/drivers/net/ethernet/microchip/lan743x_ethtool.c b/drivers/net/ethernet/microchip/lan743x_ethtool.c index 3a0b289d9771..c533d06fbe3a 100644 --- a/drivers/net/ethernet/microchip/lan743x_ethtool.c +++ b/drivers/net/ethernet/microchip/lan743x_ethtool.c @@ -2,11 +2,11 @@ /* Copyright (C) 2018 Microchip Technology Inc. */ #include <linux/netdevice.h> -#include "lan743x_main.h" -#include "lan743x_ethtool.h" #include <linux/net_tstamp.h> #include <linux/pci.h> #include <linux/phy.h> +#include "lan743x_main.h" +#include "lan743x_ethtool.h" /* eeprom */ #define LAN743X_EEPROM_MAGIC (0x74A5) diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c index a43140f7b5eb..36624e3c633b 100644 --- a/drivers/net/ethernet/microchip/lan743x_main.c +++ b/drivers/net/ethernet/microchip/lan743x_main.c @@ -8,7 +8,10 @@ #include <linux/crc32.h> #include <linux/microchipphy.h> #include <linux/net_tstamp.h> +#include <linux/of_mdio.h> +#include <linux/of_net.h> #include <linux/phy.h> +#include <linux/phy_fixed.h> #include <linux/rtnetlink.h> #include <linux/iopoll.h> #include <linux/crc16.h> @@ -798,9 +801,9 @@ static int lan743x_mac_init(struct lan743x_adapter *adapter) netdev = adapter->netdev; - /* setup auto duplex, and speed detection */ + /* disable auto duplex, and speed detection. Phylib does that */ data = lan743x_csr_read(adapter, MAC_CR); - data |= MAC_CR_ADD_ | MAC_CR_ASD_; + data &= ~(MAC_CR_ADD_ | MAC_CR_ASD_); data |= MAC_CR_CNTR_RST_; lan743x_csr_write(adapter, MAC_CR, data); @@ -946,6 +949,7 @@ static void lan743x_phy_link_status_change(struct net_device *netdev) { struct lan743x_adapter *adapter = netdev_priv(netdev); struct phy_device *phydev = netdev->phydev; + u32 data; phy_print_status(phydev); if (phydev->state == PHY_RUNNING) { @@ -953,6 +957,39 @@ static void lan743x_phy_link_status_change(struct net_device *netdev) int remote_advertisement = 0; int local_advertisement = 0; + data = lan743x_csr_read(adapter, MAC_CR); + + /* set interface mode */ + if (phy_interface_mode_is_rgmii(adapter->phy_mode)) + /* RGMII */ + data &= ~MAC_CR_MII_EN_; + else + /* GMII */ + data |= MAC_CR_MII_EN_; + + /* set duplex mode */ + if (phydev->duplex) + data |= MAC_CR_DPX_; + else + data &= ~MAC_CR_DPX_; + + /* set bus speed */ + switch (phydev->speed) { + case SPEED_10: + data &= ~MAC_CR_CFG_H_; + data &= ~MAC_CR_CFG_L_; + break; + case SPEED_100: + data &= ~MAC_CR_CFG_H_; + data |= MAC_CR_CFG_L_; + break; + case SPEED_1000: + data |= MAC_CR_CFG_H_; + data |= MAC_CR_CFG_L_; + break; + } + lan743x_csr_write(adapter, MAC_CR, data); + memset(&ksettings, 0, sizeof(ksettings)); phy_ethtool_get_link_ksettings(netdev, &ksettings); local_advertisement = @@ -980,20 +1017,44 @@ static void lan743x_phy_close(struct lan743x_adapter *adapter) static int lan743x_phy_open(struct lan743x_adapter *adapter) { struct lan743x_phy *phy = &adapter->phy; + struct device_node *phynode; struct phy_device *phydev; struct net_device *netdev; int ret = -EIO; netdev = adapter->netdev; - phydev = phy_find_first(adapter->mdiobus); - if (!phydev) - goto return_error; + phynode = of_node_get(adapter->pdev->dev.of_node); + adapter->phy_mode = PHY_INTERFACE_MODE_GMII; + + if (phynode) { + of_get_phy_mode(phynode, &adapter->phy_mode); + + if (of_phy_is_fixed_link(phynode)) { + ret = of_phy_register_fixed_link(phynode); + if (ret) { + netdev_err(netdev, + "cannot register fixed PHY\n"); + of_node_put(phynode); + goto return_error; + } + } + phydev = of_phy_connect(netdev, phynode, + lan743x_phy_link_status_change, 0, + adapter->phy_mode); + of_node_put(phynode); + if (!phydev) + goto return_error; + } else { + phydev = phy_find_first(adapter->mdiobus); + if (!phydev) + goto return_error; - ret = phy_connect_direct(netdev, phydev, - lan743x_phy_link_status_change, - PHY_INTERFACE_MODE_GMII); - if (ret) - goto return_error; + ret = phy_connect_direct(netdev, phydev, + lan743x_phy_link_status_change, + adapter->phy_mode); + if (ret) + goto return_error; + } /* MAC doesn't support 1000T Half */ phy_remove_link_mode(phydev, ETHTOOL_LINK_MODE_1000baseT_Half_BIT); diff --git a/drivers/net/ethernet/microchip/lan743x_main.h b/drivers/net/ethernet/microchip/lan743x_main.h index 3b02eeae5f45..c61a40411317 100644 --- a/drivers/net/ethernet/microchip/lan743x_main.h +++ b/drivers/net/ethernet/microchip/lan743x_main.h @@ -4,6 +4,7 @@ #ifndef _LAN743X_H #define _LAN743X_H +#include <linux/phy.h> #include "lan743x_ptp.h" #define DRIVER_AUTHOR "Bryan Whitehead <Bryan.Whitehead@microchip.com>" @@ -104,10 +105,14 @@ ((value << 0) & FCT_FLOW_CTL_ON_THRESHOLD_) #define MAC_CR (0x100) +#define MAC_CR_MII_EN_ BIT(19) #define MAC_CR_EEE_EN_ BIT(17) #define MAC_CR_ADD_ BIT(12) #define MAC_CR_ASD_ BIT(11) #define MAC_CR_CNTR_RST_ BIT(5) +#define MAC_CR_DPX_ BIT(3) +#define MAC_CR_CFG_H_ BIT(2) +#define MAC_CR_CFG_L_ BIT(1) #define MAC_CR_RST_ BIT(0) #define MAC_RX (0x104) @@ -698,6 +703,7 @@ struct lan743x_rx { struct lan743x_adapter { struct net_device *netdev; struct mii_bus *mdiobus; + phy_interface_t phy_mode; int msg_enable; #ifdef CONFIG_PM u32 wolopts; diff --git a/drivers/net/ethernet/microchip/lan743x_ptp.c b/drivers/net/ethernet/microchip/lan743x_ptp.c index 9399f6a98748..ab6d719d40f0 100644 --- a/drivers/net/ethernet/microchip/lan743x_ptp.c +++ b/drivers/net/ethernet/microchip/lan743x_ptp.c @@ -2,12 +2,12 @@ /* Copyright (C) 2018 Microchip Technology Inc. */ #include <linux/netdevice.h> -#include "lan743x_main.h" #include <linux/ptp_clock_kernel.h> #include <linux/module.h> #include <linux/pci.h> #include <linux/net_tstamp.h> +#include "lan743x_main.h" #include "lan743x_ptp.h" diff --git a/drivers/net/ethernet/moxa/moxart_ether.c b/drivers/net/ethernet/moxa/moxart_ether.c index f70bb81e1ed6..49fd843c4c8a 100644 --- a/drivers/net/ethernet/moxa/moxart_ether.c +++ b/drivers/net/ethernet/moxa/moxart_ether.c @@ -331,14 +331,15 @@ static irqreturn_t moxart_mac_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } -static int moxart_mac_start_xmit(struct sk_buff *skb, struct net_device *ndev) +static netdev_tx_t moxart_mac_start_xmit(struct sk_buff *skb, + struct net_device *ndev) { struct moxart_mac_priv_t *priv = netdev_priv(ndev); void *desc; unsigned int len; unsigned int tx_head; u32 txdes1; - int ret = NETDEV_TX_BUSY; + netdev_tx_t ret = NETDEV_TX_BUSY; spin_lock_irq(&priv->txlock); diff --git a/drivers/net/ethernet/mscc/Makefile b/drivers/net/ethernet/mscc/Makefile index 9a36c26095c8..91b33b55054e 100644 --- a/drivers/net/ethernet/mscc/Makefile +++ b/drivers/net/ethernet/mscc/Makefile @@ -1,5 +1,5 @@ # SPDX-License-Identifier: (GPL-2.0 OR MIT) obj-$(CONFIG_MSCC_OCELOT_SWITCH) += mscc_ocelot_common.o mscc_ocelot_common-y := ocelot.o ocelot_io.o -mscc_ocelot_common-y += ocelot_regs.o ocelot_tc.o ocelot_police.o ocelot_ace.o ocelot_flower.o +mscc_ocelot_common-y += ocelot_regs.o ocelot_tc.o ocelot_police.o ocelot_ace.o ocelot_flower.o ocelot_ptp.o obj-$(CONFIG_MSCC_OCELOT_SWITCH_OCELOT) += ocelot_board.o diff --git a/drivers/net/ethernet/mscc/ocelot.c b/drivers/net/ethernet/mscc/ocelot.c index 02350c3d9d01..9cfe1fd98c30 100644 --- a/drivers/net/ethernet/mscc/ocelot.c +++ b/drivers/net/ethernet/mscc/ocelot.c @@ -14,7 +14,6 @@ #include <linux/module.h> #include <linux/netdevice.h> #include <linux/phy.h> -#include <linux/ptp_clock_kernel.h> #include <linux/skbuff.h> #include <linux/iopoll.h> #include <net/arp.h> @@ -1205,18 +1204,19 @@ static int ocelot_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) struct ocelot *ocelot = priv->port.ocelot; int port = priv->chip_port; - /* The function is only used for PTP operations for now */ - if (!ocelot->ptp) - return -EOPNOTSUPP; - - switch (cmd) { - case SIOCSHWTSTAMP: - return ocelot_hwstamp_set(ocelot, port, ifr); - case SIOCGHWTSTAMP: - return ocelot_hwstamp_get(ocelot, port, ifr); - default: - return -EOPNOTSUPP; + /* If the attached PHY device isn't capable of timestamping operations, + * use our own (when possible). + */ + if (!phy_has_hwtstamp(dev->phydev) && ocelot->ptp) { + switch (cmd) { + case SIOCSHWTSTAMP: + return ocelot_hwstamp_set(ocelot, port, ifr); + case SIOCGHWTSTAMP: + return ocelot_hwstamp_get(ocelot, port, ifr); + } } + + return phy_mii_ioctl(dev->phydev, ifr, cmd); } static const struct net_device_ops ocelot_port_netdev_ops = { @@ -1348,6 +1348,12 @@ int ocelot_get_ts_info(struct ocelot *ocelot, int port, { info->phc_index = ocelot->ptp_clock ? ptp_clock_index(ocelot->ptp_clock) : -1; + if (info->phc_index == -1) { + info->so_timestamping |= SOF_TIMESTAMPING_TX_SOFTWARE | + SOF_TIMESTAMPING_RX_SOFTWARE | + SOF_TIMESTAMPING_SOFTWARE; + return 0; + } info->so_timestamping |= SOF_TIMESTAMPING_TX_SOFTWARE | SOF_TIMESTAMPING_RX_SOFTWARE | SOF_TIMESTAMPING_SOFTWARE | @@ -1467,7 +1473,7 @@ static void ocelot_port_attr_ageing_set(struct ocelot *ocelot, int port, unsigned long ageing_clock_t) { unsigned long ageing_jiffies = clock_t_to_jiffies(ageing_clock_t); - u32 ageing_time = jiffies_to_msecs(ageing_jiffies) / 1000; + u32 ageing_time = jiffies_to_msecs(ageing_jiffies); ocelot_set_ageing_time(ocelot, ageing_time); } @@ -1996,200 +2002,6 @@ struct notifier_block ocelot_switchdev_blocking_nb __read_mostly = { }; EXPORT_SYMBOL(ocelot_switchdev_blocking_nb); -int ocelot_ptp_gettime64(struct ptp_clock_info *ptp, struct timespec64 *ts) -{ - struct ocelot *ocelot = container_of(ptp, struct ocelot, ptp_info); - unsigned long flags; - time64_t s; - u32 val; - s64 ns; - - spin_lock_irqsave(&ocelot->ptp_clock_lock, flags); - - val = ocelot_read_rix(ocelot, PTP_PIN_CFG, TOD_ACC_PIN); - val &= ~(PTP_PIN_CFG_SYNC | PTP_PIN_CFG_ACTION_MASK | PTP_PIN_CFG_DOM); - val |= PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_SAVE); - ocelot_write_rix(ocelot, val, PTP_PIN_CFG, TOD_ACC_PIN); - - s = ocelot_read_rix(ocelot, PTP_PIN_TOD_SEC_MSB, TOD_ACC_PIN) & 0xffff; - s <<= 32; - s += ocelot_read_rix(ocelot, PTP_PIN_TOD_SEC_LSB, TOD_ACC_PIN); - ns = ocelot_read_rix(ocelot, PTP_PIN_TOD_NSEC, TOD_ACC_PIN); - - spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); - - /* Deal with negative values */ - if (ns >= 0x3ffffff0 && ns <= 0x3fffffff) { - s--; - ns &= 0xf; - ns += 999999984; - } - - set_normalized_timespec64(ts, s, ns); - return 0; -} -EXPORT_SYMBOL(ocelot_ptp_gettime64); - -static int ocelot_ptp_settime64(struct ptp_clock_info *ptp, - const struct timespec64 *ts) -{ - struct ocelot *ocelot = container_of(ptp, struct ocelot, ptp_info); - unsigned long flags; - u32 val; - - spin_lock_irqsave(&ocelot->ptp_clock_lock, flags); - - val = ocelot_read_rix(ocelot, PTP_PIN_CFG, TOD_ACC_PIN); - val &= ~(PTP_PIN_CFG_SYNC | PTP_PIN_CFG_ACTION_MASK | PTP_PIN_CFG_DOM); - val |= PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_IDLE); - - ocelot_write_rix(ocelot, val, PTP_PIN_CFG, TOD_ACC_PIN); - - ocelot_write_rix(ocelot, lower_32_bits(ts->tv_sec), PTP_PIN_TOD_SEC_LSB, - TOD_ACC_PIN); - ocelot_write_rix(ocelot, upper_32_bits(ts->tv_sec), PTP_PIN_TOD_SEC_MSB, - TOD_ACC_PIN); - ocelot_write_rix(ocelot, ts->tv_nsec, PTP_PIN_TOD_NSEC, TOD_ACC_PIN); - - val = ocelot_read_rix(ocelot, PTP_PIN_CFG, TOD_ACC_PIN); - val &= ~(PTP_PIN_CFG_SYNC | PTP_PIN_CFG_ACTION_MASK | PTP_PIN_CFG_DOM); - val |= PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_LOAD); - - ocelot_write_rix(ocelot, val, PTP_PIN_CFG, TOD_ACC_PIN); - - spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); - return 0; -} - -static int ocelot_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta) -{ - if (delta > -(NSEC_PER_SEC / 2) && delta < (NSEC_PER_SEC / 2)) { - struct ocelot *ocelot = container_of(ptp, struct ocelot, ptp_info); - unsigned long flags; - u32 val; - - spin_lock_irqsave(&ocelot->ptp_clock_lock, flags); - - val = ocelot_read_rix(ocelot, PTP_PIN_CFG, TOD_ACC_PIN); - val &= ~(PTP_PIN_CFG_SYNC | PTP_PIN_CFG_ACTION_MASK | PTP_PIN_CFG_DOM); - val |= PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_IDLE); - - ocelot_write_rix(ocelot, val, PTP_PIN_CFG, TOD_ACC_PIN); - - ocelot_write_rix(ocelot, 0, PTP_PIN_TOD_SEC_LSB, TOD_ACC_PIN); - ocelot_write_rix(ocelot, 0, PTP_PIN_TOD_SEC_MSB, TOD_ACC_PIN); - ocelot_write_rix(ocelot, delta, PTP_PIN_TOD_NSEC, TOD_ACC_PIN); - - val = ocelot_read_rix(ocelot, PTP_PIN_CFG, TOD_ACC_PIN); - val &= ~(PTP_PIN_CFG_SYNC | PTP_PIN_CFG_ACTION_MASK | PTP_PIN_CFG_DOM); - val |= PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_DELTA); - - ocelot_write_rix(ocelot, val, PTP_PIN_CFG, TOD_ACC_PIN); - - spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); - } else { - /* Fall back using ocelot_ptp_settime64 which is not exact. */ - struct timespec64 ts; - u64 now; - - ocelot_ptp_gettime64(ptp, &ts); - - now = ktime_to_ns(timespec64_to_ktime(ts)); - ts = ns_to_timespec64(now + delta); - - ocelot_ptp_settime64(ptp, &ts); - } - return 0; -} - -static int ocelot_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm) -{ - struct ocelot *ocelot = container_of(ptp, struct ocelot, ptp_info); - u32 unit = 0, direction = 0; - unsigned long flags; - u64 adj = 0; - - spin_lock_irqsave(&ocelot->ptp_clock_lock, flags); - - if (!scaled_ppm) - goto disable_adj; - - if (scaled_ppm < 0) { - direction = PTP_CFG_CLK_ADJ_CFG_DIR; - scaled_ppm = -scaled_ppm; - } - - adj = PSEC_PER_SEC << 16; - do_div(adj, scaled_ppm); - do_div(adj, 1000); - - /* If the adjustment value is too large, use ns instead */ - if (adj >= (1L << 30)) { - unit = PTP_CFG_CLK_ADJ_FREQ_NS; - do_div(adj, 1000); - } - - /* Still too big */ - if (adj >= (1L << 30)) - goto disable_adj; - - ocelot_write(ocelot, unit | adj, PTP_CLK_CFG_ADJ_FREQ); - ocelot_write(ocelot, PTP_CFG_CLK_ADJ_CFG_ENA | direction, - PTP_CLK_CFG_ADJ_CFG); - - spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); - return 0; - -disable_adj: - ocelot_write(ocelot, 0, PTP_CLK_CFG_ADJ_CFG); - - spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); - return 0; -} - -static struct ptp_clock_info ocelot_ptp_clock_info = { - .owner = THIS_MODULE, - .name = "ocelot ptp", - .max_adj = 0x7fffffff, - .n_alarm = 0, - .n_ext_ts = 0, - .n_per_out = 0, - .n_pins = 0, - .pps = 0, - .gettime64 = ocelot_ptp_gettime64, - .settime64 = ocelot_ptp_settime64, - .adjtime = ocelot_ptp_adjtime, - .adjfine = ocelot_ptp_adjfine, -}; - -static int ocelot_init_timestamp(struct ocelot *ocelot) -{ - struct ptp_clock *ptp_clock; - - ocelot->ptp_info = ocelot_ptp_clock_info; - ptp_clock = ptp_clock_register(&ocelot->ptp_info, ocelot->dev); - if (IS_ERR(ptp_clock)) - return PTR_ERR(ptp_clock); - /* Check if PHC support is missing at the configuration level */ - if (!ptp_clock) - return 0; - - ocelot->ptp_clock = ptp_clock; - - ocelot_write(ocelot, SYS_PTP_CFG_PTP_STAMP_WID(30), SYS_PTP_CFG); - ocelot_write(ocelot, 0xffffffff, ANA_TABLES_PTP_ID_LOW); - ocelot_write(ocelot, 0xffffffff, ANA_TABLES_PTP_ID_HIGH); - - ocelot_write(ocelot, PTP_CFG_MISC_PTP_EN, PTP_CFG_MISC); - - /* There is no device reconfiguration, PTP Rx stamping is always - * enabled. - */ - ocelot->hwtstamp_config.rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT; - - return 0; -} - /* Configure the maximum SDU (L2 payload) on RX to the value specified in @sdu. * The length of VLAN tags is accounted for automatically via DEV_MAC_TAGS_CFG. * In the special case that it's the NPI port that we're configuring, the @@ -2535,15 +2347,6 @@ int ocelot_init(struct ocelot *ocelot) queue_delayed_work(ocelot->stats_queue, &ocelot->stats_work, OCELOT_STATS_CHECK_DELAY); - if (ocelot->ptp) { - ret = ocelot_init_timestamp(ocelot); - if (ret) { - dev_err(ocelot->dev, - "Timestamp initialization failed\n"); - return ret; - } - } - return 0; } EXPORT_SYMBOL(ocelot_init); @@ -2556,8 +2359,6 @@ void ocelot_deinit(struct ocelot *ocelot) cancel_delayed_work(&ocelot->stats_work); destroy_workqueue(ocelot->stats_queue); mutex_destroy(&ocelot->stats_lock); - if (ocelot->ptp_clock) - ptp_clock_unregister(ocelot->ptp_clock); for (i = 0; i < ocelot->num_phys_ports; i++) { port = ocelot->ports[i]; diff --git a/drivers/net/ethernet/mscc/ocelot.h b/drivers/net/ethernet/mscc/ocelot.h index 641af929497f..f0a15aa187f2 100644 --- a/drivers/net/ethernet/mscc/ocelot.h +++ b/drivers/net/ethernet/mscc/ocelot.h @@ -15,18 +15,17 @@ #include <linux/phy.h> #include <linux/phy/phy.h> #include <linux/platform_device.h> -#include <linux/ptp_clock_kernel.h> #include <linux/regmap.h> #include <soc/mscc/ocelot_qsys.h> #include <soc/mscc/ocelot_sys.h> #include <soc/mscc/ocelot_dev.h> #include <soc/mscc/ocelot_ana.h> +#include <soc/mscc/ocelot_ptp.h> #include <soc/mscc/ocelot.h> #include "ocelot_rew.h" #include "ocelot_qs.h" #include "ocelot_tc.h" -#include "ocelot_ptp.h" #define OCELOT_BUFFER_CELL_SZ 60 diff --git a/drivers/net/ethernet/mscc/ocelot_ace.c b/drivers/net/ethernet/mscc/ocelot_ace.c index 3bd286044480..dfd82a3baab2 100644 --- a/drivers/net/ethernet/mscc/ocelot_ace.c +++ b/drivers/net/ethernet/mscc/ocelot_ace.c @@ -706,13 +706,124 @@ ocelot_ace_rule_get_rule_index(struct ocelot_acl_block *block, int index) return NULL; } +/* If @on=false, then SNAP, ARP, IP and OAM frames will not match on keys based + * on destination and source MAC addresses, but only on higher-level protocol + * information. The only frame types to match on keys containing MAC addresses + * in this case are non-SNAP, non-ARP, non-IP and non-OAM frames. + * + * If @on=true, then the above frame types (SNAP, ARP, IP and OAM) will match + * on MAC_ETYPE keys such as destination and source MAC on this ingress port. + * However the setting has the side effect of making these frames not matching + * on any _other_ keys than MAC_ETYPE ones. + */ +static void ocelot_match_all_as_mac_etype(struct ocelot *ocelot, int port, + bool on) +{ + u32 val = 0; + + if (on) + val = ANA_PORT_VCAP_S2_CFG_S2_SNAP_DIS(3) | + ANA_PORT_VCAP_S2_CFG_S2_ARP_DIS(3) | + ANA_PORT_VCAP_S2_CFG_S2_IP_TCPUDP_DIS(3) | + ANA_PORT_VCAP_S2_CFG_S2_IP_OTHER_DIS(3) | + ANA_PORT_VCAP_S2_CFG_S2_OAM_DIS(3); + + ocelot_rmw_gix(ocelot, val, + ANA_PORT_VCAP_S2_CFG_S2_SNAP_DIS_M | + ANA_PORT_VCAP_S2_CFG_S2_ARP_DIS_M | + ANA_PORT_VCAP_S2_CFG_S2_IP_TCPUDP_DIS_M | + ANA_PORT_VCAP_S2_CFG_S2_IP_OTHER_DIS_M | + ANA_PORT_VCAP_S2_CFG_S2_OAM_DIS_M, + ANA_PORT_VCAP_S2_CFG, port); +} + +static bool ocelot_ace_is_problematic_mac_etype(struct ocelot_ace_rule *ace) +{ + u16 proto, mask; + + if (ace->type != OCELOT_ACE_TYPE_ETYPE) + return false; + + proto = ntohs(*(u16 *)ace->frame.etype.etype.value); + mask = ntohs(*(u16 *)ace->frame.etype.etype.mask); + + /* ETH_P_ALL match, so all protocols below are included */ + if (mask == 0) + return true; + if (proto == ETH_P_ARP) + return true; + if (proto == ETH_P_IP) + return true; + if (proto == ETH_P_IPV6) + return true; + + return false; +} + +static bool ocelot_ace_is_problematic_non_mac_etype(struct ocelot_ace_rule *ace) +{ + if (ace->type == OCELOT_ACE_TYPE_SNAP) + return true; + if (ace->type == OCELOT_ACE_TYPE_ARP) + return true; + if (ace->type == OCELOT_ACE_TYPE_IPV4) + return true; + if (ace->type == OCELOT_ACE_TYPE_IPV6) + return true; + return false; +} + +static bool ocelot_exclusive_mac_etype_ace_rules(struct ocelot *ocelot, + struct ocelot_ace_rule *ace) +{ + struct ocelot_acl_block *block = &ocelot->acl_block; + struct ocelot_ace_rule *tmp; + unsigned long port; + int i; + + if (ocelot_ace_is_problematic_mac_etype(ace)) { + /* Search for any non-MAC_ETYPE rules on the port */ + for (i = 0; i < block->count; i++) { + tmp = ocelot_ace_rule_get_rule_index(block, i); + if (tmp->ingress_port_mask & ace->ingress_port_mask && + ocelot_ace_is_problematic_non_mac_etype(tmp)) + return false; + } + + for_each_set_bit(port, &ace->ingress_port_mask, + ocelot->num_phys_ports) + ocelot_match_all_as_mac_etype(ocelot, port, true); + } else if (ocelot_ace_is_problematic_non_mac_etype(ace)) { + /* Search for any MAC_ETYPE rules on the port */ + for (i = 0; i < block->count; i++) { + tmp = ocelot_ace_rule_get_rule_index(block, i); + if (tmp->ingress_port_mask & ace->ingress_port_mask && + ocelot_ace_is_problematic_mac_etype(tmp)) + return false; + } + + for_each_set_bit(port, &ace->ingress_port_mask, + ocelot->num_phys_ports) + ocelot_match_all_as_mac_etype(ocelot, port, false); + } + + return true; +} + int ocelot_ace_rule_offload_add(struct ocelot *ocelot, - struct ocelot_ace_rule *rule) + struct ocelot_ace_rule *rule, + struct netlink_ext_ack *extack) { struct ocelot_acl_block *block = &ocelot->acl_block; struct ocelot_ace_rule *ace; int i, index; + if (!ocelot_exclusive_mac_etype_ace_rules(ocelot, rule)) { + NL_SET_ERR_MSG_MOD(extack, + "Cannot mix MAC_ETYPE with non-MAC_ETYPE rules"); + return -EBUSY; + } + /* Add rule to the linked list */ ocelot_ace_rule_add(ocelot, block, rule); diff --git a/drivers/net/ethernet/mscc/ocelot_ace.h b/drivers/net/ethernet/mscc/ocelot_ace.h index 29d22c566786..099e177f2617 100644 --- a/drivers/net/ethernet/mscc/ocelot_ace.h +++ b/drivers/net/ethernet/mscc/ocelot_ace.h @@ -194,7 +194,7 @@ struct ocelot_ace_rule { enum ocelot_ace_action action; struct ocelot_ace_stats stats; - u16 ingress_port_mask; + unsigned long ingress_port_mask; enum ocelot_vcap_bit dmac_mc; enum ocelot_vcap_bit dmac_bc; @@ -215,7 +215,8 @@ struct ocelot_ace_rule { }; int ocelot_ace_rule_offload_add(struct ocelot *ocelot, - struct ocelot_ace_rule *rule); + struct ocelot_ace_rule *rule, + struct netlink_ext_ack *extack); int ocelot_ace_rule_offload_del(struct ocelot *ocelot, struct ocelot_ace_rule *rule); int ocelot_ace_rule_stats_update(struct ocelot *ocelot, diff --git a/drivers/net/ethernet/mscc/ocelot_board.c b/drivers/net/ethernet/mscc/ocelot_board.c index 0ac9fbf77a01..4a15d2ff8b70 100644 --- a/drivers/net/ethernet/mscc/ocelot_board.c +++ b/drivers/net/ethernet/mscc/ocelot_board.c @@ -189,7 +189,8 @@ static irqreturn_t ocelot_xtr_irq_handler(int irq, void *arg) skb->offload_fwd_mark = 1; skb->protocol = eth_type_trans(skb, dev); - netif_rx(skb); + if (!skb_defer_rx_timestamp(skb)) + netif_rx(skb); dev->stats.rx_bytes += len; dev->stats.rx_packets++; } while (ocelot_read(ocelot, QS_XTR_DATA_PRESENT) & BIT(grp)); @@ -366,6 +367,23 @@ static const struct vcap_props vsc7514_vcap_props[] = { }, }; +static struct ptp_clock_info ocelot_ptp_clock_info = { + .owner = THIS_MODULE, + .name = "ocelot ptp", + .max_adj = 0x7fffffff, + .n_alarm = 0, + .n_ext_ts = 0, + .n_per_out = OCELOT_PTP_PINS_NUM, + .n_pins = OCELOT_PTP_PINS_NUM, + .pps = 0, + .gettime64 = ocelot_ptp_gettime64, + .settime64 = ocelot_ptp_settime64, + .adjtime = ocelot_ptp_adjtime, + .adjfine = ocelot_ptp_adjfine, + .verify = ocelot_ptp_verify, + .enable = ocelot_ptp_enable, +}; + static int mscc_ocelot_probe(struct platform_device *pdev) { struct device_node *np = pdev->dev.of_node; @@ -469,6 +487,15 @@ static int mscc_ocelot_probe(struct platform_device *pdev) ocelot->vcap = vsc7514_vcap_props; ocelot_init(ocelot); + if (ocelot->ptp) { + err = ocelot_init_timestamp(ocelot, &ocelot_ptp_clock_info); + if (err) { + dev_err(ocelot->dev, + "Timestamp initialization failed\n"); + ocelot->ptp = 0; + } + } + /* No NPI port */ ocelot_configure_cpu(ocelot, -1, OCELOT_TAG_PREFIX_NONE, OCELOT_TAG_PREFIX_NONE); @@ -574,6 +601,7 @@ static int mscc_ocelot_remove(struct platform_device *pdev) { struct ocelot *ocelot = platform_get_drvdata(pdev); + ocelot_deinit_timestamp(ocelot); ocelot_deinit(ocelot); unregister_switchdev_blocking_notifier(&ocelot_switchdev_blocking_nb); unregister_switchdev_notifier(&ocelot_switchdev_nb); diff --git a/drivers/net/ethernet/mscc/ocelot_flower.c b/drivers/net/ethernet/mscc/ocelot_flower.c index 341923311fec..5ce172e22b43 100644 --- a/drivers/net/ethernet/mscc/ocelot_flower.c +++ b/drivers/net/ethernet/mscc/ocelot_flower.c @@ -51,6 +51,8 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, { struct flow_rule *rule = flow_cls_offload_flow_rule(f); struct flow_dissector *dissector = rule->match.dissector; + u16 proto = ntohs(f->common.protocol); + bool match_protocol = true; if (dissector->used_keys & ~(BIT(FLOW_DISSECTOR_KEY_CONTROL) | @@ -71,7 +73,6 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ETH_ADDRS)) { struct flow_match_eth_addrs match; - u16 proto = ntohs(f->common.protocol); /* The hw support mac matches only for MAC_ETYPE key, * therefore if other matches(port, tcp flags, etc) are added @@ -86,11 +87,6 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, BIT(FLOW_DISSECTOR_KEY_CONTROL))) return -EOPNOTSUPP; - if (proto == ETH_P_IP || - proto == ETH_P_IPV6 || - proto == ETH_P_ARP) - return -EOPNOTSUPP; - flow_rule_match_eth_addrs(rule, &match); ace->type = OCELOT_ACE_TYPE_ETYPE; ether_addr_copy(ace->frame.etype.dmac.value, @@ -114,6 +110,7 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, match.key->ip_proto; ace->frame.ipv4.proto.mask[0] = match.mask->ip_proto; + match_protocol = false; } if (ntohs(match.key->n_proto) == ETH_P_IPV6) { ace->type = OCELOT_ACE_TYPE_IPV6; @@ -121,11 +118,12 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, match.key->ip_proto; ace->frame.ipv6.proto.mask[0] = match.mask->ip_proto; + match_protocol = false; } } if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IPV4_ADDRS) && - ntohs(f->common.protocol) == ETH_P_IP) { + proto == ETH_P_IP) { struct flow_match_ipv4_addrs match; u8 *tmp; @@ -141,10 +139,11 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, tmp = &ace->frame.ipv4.dip.mask.addr[0]; memcpy(tmp, &match.mask->dst, 4); + match_protocol = false; } if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IPV6_ADDRS) && - ntohs(f->common.protocol) == ETH_P_IPV6) { + proto == ETH_P_IPV6) { return -EOPNOTSUPP; } @@ -156,6 +155,7 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, ace->frame.ipv4.sport.mask = ntohs(match.mask->src); ace->frame.ipv4.dport.value = ntohs(match.key->dst); ace->frame.ipv4.dport.mask = ntohs(match.mask->dst); + match_protocol = false; } if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) { @@ -167,9 +167,20 @@ static int ocelot_flower_parse(struct flow_cls_offload *f, ace->vlan.vid.mask = match.mask->vlan_id; ace->vlan.pcp.value[0] = match.key->vlan_priority; ace->vlan.pcp.mask[0] = match.mask->vlan_priority; + match_protocol = false; } finished_key_parsing: + if (match_protocol && proto != ETH_P_ALL) { + /* TODO: support SNAP, LLC etc */ + if (proto < ETH_P_802_3_MIN) + return -EOPNOTSUPP; + ace->type = OCELOT_ACE_TYPE_ETYPE; + *(u16 *)ace->frame.etype.etype.value = htons(proto); + *(u16 *)ace->frame.etype.etype.mask = 0xffff; + } + /* else, a rule of type OCELOT_ACE_TYPE_ANY is implicitly added */ + ace->prio = f->common.prio; ace->id = f->cookie; return ocelot_flower_parse_action(f, ace); @@ -205,7 +216,7 @@ int ocelot_cls_flower_replace(struct ocelot *ocelot, int port, return ret; } - return ocelot_ace_rule_offload_add(ocelot, ace); + return ocelot_ace_rule_offload_add(ocelot, ace, f->common.extack); } EXPORT_SYMBOL_GPL(ocelot_cls_flower_replace); diff --git a/drivers/net/ethernet/mscc/ocelot_ptp.c b/drivers/net/ethernet/mscc/ocelot_ptp.c new file mode 100644 index 000000000000..a3088a1676ed --- /dev/null +++ b/drivers/net/ethernet/mscc/ocelot_ptp.c @@ -0,0 +1,324 @@ +// SPDX-License-Identifier: (GPL-2.0 OR MIT) +/* Microsemi Ocelot PTP clock driver + * + * Copyright (c) 2017 Microsemi Corporation + * Copyright 2020 NXP + */ +#include <soc/mscc/ocelot_ptp.h> +#include <soc/mscc/ocelot_sys.h> +#include <soc/mscc/ocelot.h> + +int ocelot_ptp_gettime64(struct ptp_clock_info *ptp, struct timespec64 *ts) +{ + struct ocelot *ocelot = container_of(ptp, struct ocelot, ptp_info); + unsigned long flags; + time64_t s; + u32 val; + s64 ns; + + spin_lock_irqsave(&ocelot->ptp_clock_lock, flags); + + val = ocelot_read_rix(ocelot, PTP_PIN_CFG, TOD_ACC_PIN); + val &= ~(PTP_PIN_CFG_SYNC | PTP_PIN_CFG_ACTION_MASK | PTP_PIN_CFG_DOM); + val |= PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_SAVE); + ocelot_write_rix(ocelot, val, PTP_PIN_CFG, TOD_ACC_PIN); + + s = ocelot_read_rix(ocelot, PTP_PIN_TOD_SEC_MSB, TOD_ACC_PIN) & 0xffff; + s <<= 32; + s += ocelot_read_rix(ocelot, PTP_PIN_TOD_SEC_LSB, TOD_ACC_PIN); + ns = ocelot_read_rix(ocelot, PTP_PIN_TOD_NSEC, TOD_ACC_PIN); + + spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); + + /* Deal with negative values */ + if (ns >= 0x3ffffff0 && ns <= 0x3fffffff) { + s--; + ns &= 0xf; + ns += 999999984; + } + + set_normalized_timespec64(ts, s, ns); + return 0; +} +EXPORT_SYMBOL(ocelot_ptp_gettime64); + +int ocelot_ptp_settime64(struct ptp_clock_info *ptp, + const struct timespec64 *ts) +{ + struct ocelot *ocelot = container_of(ptp, struct ocelot, ptp_info); + unsigned long flags; + u32 val; + + spin_lock_irqsave(&ocelot->ptp_clock_lock, flags); + + val = ocelot_read_rix(ocelot, PTP_PIN_CFG, TOD_ACC_PIN); + val &= ~(PTP_PIN_CFG_SYNC | PTP_PIN_CFG_ACTION_MASK | PTP_PIN_CFG_DOM); + val |= PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_IDLE); + + ocelot_write_rix(ocelot, val, PTP_PIN_CFG, TOD_ACC_PIN); + + ocelot_write_rix(ocelot, lower_32_bits(ts->tv_sec), PTP_PIN_TOD_SEC_LSB, + TOD_ACC_PIN); + ocelot_write_rix(ocelot, upper_32_bits(ts->tv_sec), PTP_PIN_TOD_SEC_MSB, + TOD_ACC_PIN); + ocelot_write_rix(ocelot, ts->tv_nsec, PTP_PIN_TOD_NSEC, TOD_ACC_PIN); + + val = ocelot_read_rix(ocelot, PTP_PIN_CFG, TOD_ACC_PIN); + val &= ~(PTP_PIN_CFG_SYNC | PTP_PIN_CFG_ACTION_MASK | PTP_PIN_CFG_DOM); + val |= PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_LOAD); + + ocelot_write_rix(ocelot, val, PTP_PIN_CFG, TOD_ACC_PIN); + + spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); + return 0; +} +EXPORT_SYMBOL(ocelot_ptp_settime64); + +int ocelot_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta) +{ + if (delta > -(NSEC_PER_SEC / 2) && delta < (NSEC_PER_SEC / 2)) { + struct ocelot *ocelot = container_of(ptp, struct ocelot, + ptp_info); + unsigned long flags; + u32 val; + + spin_lock_irqsave(&ocelot->ptp_clock_lock, flags); + + val = ocelot_read_rix(ocelot, PTP_PIN_CFG, TOD_ACC_PIN); + val &= ~(PTP_PIN_CFG_SYNC | PTP_PIN_CFG_ACTION_MASK | + PTP_PIN_CFG_DOM); + val |= PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_IDLE); + + ocelot_write_rix(ocelot, val, PTP_PIN_CFG, TOD_ACC_PIN); + + ocelot_write_rix(ocelot, 0, PTP_PIN_TOD_SEC_LSB, TOD_ACC_PIN); + ocelot_write_rix(ocelot, 0, PTP_PIN_TOD_SEC_MSB, TOD_ACC_PIN); + ocelot_write_rix(ocelot, delta, PTP_PIN_TOD_NSEC, TOD_ACC_PIN); + + val = ocelot_read_rix(ocelot, PTP_PIN_CFG, TOD_ACC_PIN); + val &= ~(PTP_PIN_CFG_SYNC | PTP_PIN_CFG_ACTION_MASK | + PTP_PIN_CFG_DOM); + val |= PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_DELTA); + + ocelot_write_rix(ocelot, val, PTP_PIN_CFG, TOD_ACC_PIN); + + spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); + } else { + /* Fall back using ocelot_ptp_settime64 which is not exact. */ + struct timespec64 ts; + u64 now; + + ocelot_ptp_gettime64(ptp, &ts); + + now = ktime_to_ns(timespec64_to_ktime(ts)); + ts = ns_to_timespec64(now + delta); + + ocelot_ptp_settime64(ptp, &ts); + } + return 0; +} +EXPORT_SYMBOL(ocelot_ptp_adjtime); + +int ocelot_ptp_adjfine(struct ptp_clock_info *ptp, long scaled_ppm) +{ + struct ocelot *ocelot = container_of(ptp, struct ocelot, ptp_info); + u32 unit = 0, direction = 0; + unsigned long flags; + u64 adj = 0; + + spin_lock_irqsave(&ocelot->ptp_clock_lock, flags); + + if (!scaled_ppm) + goto disable_adj; + + if (scaled_ppm < 0) { + direction = PTP_CFG_CLK_ADJ_CFG_DIR; + scaled_ppm = -scaled_ppm; + } + + adj = PSEC_PER_SEC << 16; + do_div(adj, scaled_ppm); + do_div(adj, 1000); + + /* If the adjustment value is too large, use ns instead */ + if (adj >= (1L << 30)) { + unit = PTP_CFG_CLK_ADJ_FREQ_NS; + do_div(adj, 1000); + } + + /* Still too big */ + if (adj >= (1L << 30)) + goto disable_adj; + + ocelot_write(ocelot, unit | adj, PTP_CLK_CFG_ADJ_FREQ); + ocelot_write(ocelot, PTP_CFG_CLK_ADJ_CFG_ENA | direction, + PTP_CLK_CFG_ADJ_CFG); + + spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); + return 0; + +disable_adj: + ocelot_write(ocelot, 0, PTP_CLK_CFG_ADJ_CFG); + + spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); + return 0; +} +EXPORT_SYMBOL(ocelot_ptp_adjfine); + +int ocelot_ptp_verify(struct ptp_clock_info *ptp, unsigned int pin, + enum ptp_pin_function func, unsigned int chan) +{ + switch (func) { + case PTP_PF_NONE: + case PTP_PF_PEROUT: + break; + case PTP_PF_EXTTS: + case PTP_PF_PHYSYNC: + return -1; + } + return 0; +} +EXPORT_SYMBOL(ocelot_ptp_verify); + +int ocelot_ptp_enable(struct ptp_clock_info *ptp, + struct ptp_clock_request *rq, int on) +{ + struct ocelot *ocelot = container_of(ptp, struct ocelot, ptp_info); + struct timespec64 ts_start, ts_period; + enum ocelot_ptp_pins ptp_pin; + unsigned long flags; + bool pps = false; + int pin = -1; + u32 val; + s64 ns; + + switch (rq->type) { + case PTP_CLK_REQ_PEROUT: + /* Reject requests with unsupported flags */ + if (rq->perout.flags) + return -EOPNOTSUPP; + + pin = ptp_find_pin(ocelot->ptp_clock, PTP_PF_PEROUT, + rq->perout.index); + if (pin == 0) + ptp_pin = PTP_PIN_0; + else if (pin == 1) + ptp_pin = PTP_PIN_1; + else if (pin == 2) + ptp_pin = PTP_PIN_2; + else if (pin == 3) + ptp_pin = PTP_PIN_3; + else + return -EBUSY; + + ts_start.tv_sec = rq->perout.start.sec; + ts_start.tv_nsec = rq->perout.start.nsec; + ts_period.tv_sec = rq->perout.period.sec; + ts_period.tv_nsec = rq->perout.period.nsec; + + if (ts_period.tv_sec == 1 && ts_period.tv_nsec == 0) + pps = true; + + if (ts_start.tv_sec || (ts_start.tv_nsec && !pps)) { + dev_warn(ocelot->dev, + "Absolute start time not supported!\n"); + dev_warn(ocelot->dev, + "Accept nsec for PPS phase adjustment, otherwise start time should be 0 0.\n"); + return -EINVAL; + } + + /* Handle turning off */ + if (!on) { + spin_lock_irqsave(&ocelot->ptp_clock_lock, flags); + val = PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_IDLE); + ocelot_write_rix(ocelot, val, PTP_PIN_CFG, ptp_pin); + spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); + break; + } + + /* Handle PPS request */ + if (pps) { + spin_lock_irqsave(&ocelot->ptp_clock_lock, flags); + /* Pulse generated perout.start.nsec after TOD has + * increased seconds. + * Pulse width is set to 1us. + */ + ocelot_write_rix(ocelot, ts_start.tv_nsec, + PTP_PIN_WF_LOW_PERIOD, ptp_pin); + ocelot_write_rix(ocelot, 1000, + PTP_PIN_WF_HIGH_PERIOD, ptp_pin); + val = PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_CLOCK); + val |= PTP_PIN_CFG_SYNC; + ocelot_write_rix(ocelot, val, PTP_PIN_CFG, ptp_pin); + spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); + break; + } + + /* Handle periodic clock */ + ns = timespec64_to_ns(&ts_period); + ns = ns >> 1; + if (ns > 0x3fffffff || ns <= 0x6) + return -EINVAL; + + spin_lock_irqsave(&ocelot->ptp_clock_lock, flags); + ocelot_write_rix(ocelot, ns, PTP_PIN_WF_LOW_PERIOD, ptp_pin); + ocelot_write_rix(ocelot, ns, PTP_PIN_WF_HIGH_PERIOD, ptp_pin); + val = PTP_PIN_CFG_ACTION(PTP_PIN_ACTION_CLOCK); + ocelot_write_rix(ocelot, val, PTP_PIN_CFG, ptp_pin); + spin_unlock_irqrestore(&ocelot->ptp_clock_lock, flags); + break; + default: + return -EOPNOTSUPP; + } + return 0; +} +EXPORT_SYMBOL(ocelot_ptp_enable); + +int ocelot_init_timestamp(struct ocelot *ocelot, struct ptp_clock_info *info) +{ + struct ptp_clock *ptp_clock; + int i; + + ocelot->ptp_info = *info; + + for (i = 0; i < OCELOT_PTP_PINS_NUM; i++) { + struct ptp_pin_desc *p = &ocelot->ptp_pins[i]; + + snprintf(p->name, sizeof(p->name), "switch_1588_dat%d", i); + p->index = i; + p->func = PTP_PF_NONE; + } + + ocelot->ptp_info.pin_config = &ocelot->ptp_pins[0]; + + ptp_clock = ptp_clock_register(&ocelot->ptp_info, ocelot->dev); + if (IS_ERR(ptp_clock)) + return PTR_ERR(ptp_clock); + /* Check if PHC support is missing at the configuration level */ + if (!ptp_clock) + return 0; + + ocelot->ptp_clock = ptp_clock; + + ocelot_write(ocelot, SYS_PTP_CFG_PTP_STAMP_WID(30), SYS_PTP_CFG); + ocelot_write(ocelot, 0xffffffff, ANA_TABLES_PTP_ID_LOW); + ocelot_write(ocelot, 0xffffffff, ANA_TABLES_PTP_ID_HIGH); + + ocelot_write(ocelot, PTP_CFG_MISC_PTP_EN, PTP_CFG_MISC); + + /* There is no device reconfiguration, PTP Rx stamping is always + * enabled. + */ + ocelot->hwtstamp_config.rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT; + + return 0; +} +EXPORT_SYMBOL(ocelot_init_timestamp); + +int ocelot_deinit_timestamp(struct ocelot *ocelot) +{ + if (ocelot->ptp_clock) + ptp_clock_unregister(ocelot->ptp_clock); + return 0; +} +EXPORT_SYMBOL(ocelot_deinit_timestamp); diff --git a/drivers/net/ethernet/mscc/ocelot_ptp.h b/drivers/net/ethernet/mscc/ocelot_ptp.h deleted file mode 100644 index 9ede14a12573..000000000000 --- a/drivers/net/ethernet/mscc/ocelot_ptp.h +++ /dev/null @@ -1,41 +0,0 @@ -/* SPDX-License-Identifier: (GPL-2.0 OR MIT) */ -/* - * Microsemi Ocelot Switch driver - * - * License: Dual MIT/GPL - * Copyright (c) 2017 Microsemi Corporation - */ - -#ifndef _MSCC_OCELOT_PTP_H_ -#define _MSCC_OCELOT_PTP_H_ - -#define PTP_PIN_CFG_RSZ 0x20 -#define PTP_PIN_TOD_SEC_MSB_RSZ PTP_PIN_CFG_RSZ -#define PTP_PIN_TOD_SEC_LSB_RSZ PTP_PIN_CFG_RSZ -#define PTP_PIN_TOD_NSEC_RSZ PTP_PIN_CFG_RSZ - -#define PTP_PIN_CFG_DOM BIT(0) -#define PTP_PIN_CFG_SYNC BIT(2) -#define PTP_PIN_CFG_ACTION(x) ((x) << 3) -#define PTP_PIN_CFG_ACTION_MASK PTP_PIN_CFG_ACTION(0x7) - -enum { - PTP_PIN_ACTION_IDLE = 0, - PTP_PIN_ACTION_LOAD, - PTP_PIN_ACTION_SAVE, - PTP_PIN_ACTION_CLOCK, - PTP_PIN_ACTION_DELTA, - PTP_PIN_ACTION_NOSYNC, - PTP_PIN_ACTION_SYNC, -}; - -#define PTP_CFG_MISC_PTP_EN BIT(2) - -#define PSEC_PER_SEC 1000000000000LL - -#define PTP_CFG_CLK_ADJ_CFG_ENA BIT(0) -#define PTP_CFG_CLK_ADJ_CFG_DIR BIT(1) - -#define PTP_CFG_CLK_ADJ_FREQ_NS BIT(30) - -#endif diff --git a/drivers/net/ethernet/mscc/ocelot_regs.c b/drivers/net/ethernet/mscc/ocelot_regs.c index 7d4fd1b6adda..81d81ff75646 100644 --- a/drivers/net/ethernet/mscc/ocelot_regs.c +++ b/drivers/net/ethernet/mscc/ocelot_regs.c @@ -239,6 +239,8 @@ static const u32 ocelot_ptp_regmap[] = { REG(PTP_PIN_TOD_SEC_MSB, 0x000004), REG(PTP_PIN_TOD_SEC_LSB, 0x000008), REG(PTP_PIN_TOD_NSEC, 0x00000c), + REG(PTP_PIN_WF_HIGH_PERIOD, 0x000014), + REG(PTP_PIN_WF_LOW_PERIOD, 0x000018), REG(PTP_CFG_MISC, 0x0000a0), REG(PTP_CLK_CFG_ADJ_CFG, 0x0000a4), REG(PTP_CLK_CFG_ADJ_FREQ, 0x0000a8), diff --git a/drivers/net/ethernet/mscc/ocelot_tc.c b/drivers/net/ethernet/mscc/ocelot_tc.c index d326e231f0ad..b7baf7624e18 100644 --- a/drivers/net/ethernet/mscc/ocelot_tc.c +++ b/drivers/net/ethernet/mscc/ocelot_tc.c @@ -48,7 +48,7 @@ static int ocelot_setup_tc_cls_matchall(struct ocelot_port_private *priv, if (priv->tc.police_id && priv->tc.police_id != f->cookie) { NL_SET_ERR_MSG_MOD(extack, - "Only one policer per port is supported\n"); + "Only one policer per port is supported"); return -EEXIST; } @@ -59,7 +59,7 @@ static int ocelot_setup_tc_cls_matchall(struct ocelot_port_private *priv, err = ocelot_port_policer_add(ocelot, port, &pol); if (err) { - NL_SET_ERR_MSG_MOD(extack, "Could not add policer\n"); + NL_SET_ERR_MSG_MOD(extack, "Could not add policer"); return err; } @@ -73,7 +73,7 @@ static int ocelot_setup_tc_cls_matchall(struct ocelot_port_private *priv, err = ocelot_port_policer_del(ocelot, port); if (err) { NL_SET_ERR_MSG_MOD(extack, - "Could not delete policer\n"); + "Could not delete policer"); return err; } priv->tc.police_id = 0; diff --git a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c index 2616fd735aab..e1e1f4e3639e 100644 --- a/drivers/net/ethernet/myricom/myri10ge/myri10ge.c +++ b/drivers/net/ethernet/myricom/myri10ge/myri10ge.c @@ -1174,18 +1174,6 @@ myri10ge_submit_8rx(struct mcp_kreq_ether_recv __iomem * dst, mb(); } -static inline void myri10ge_vlan_ip_csum(struct sk_buff *skb, __wsum hw_csum) -{ - struct vlan_hdr *vh = (struct vlan_hdr *)(skb->data); - - if ((skb->protocol == htons(ETH_P_8021Q)) && - (vh->h_vlan_encapsulated_proto == htons(ETH_P_IP) || - vh->h_vlan_encapsulated_proto == htons(ETH_P_IPV6))) { - skb->csum = hw_csum; - skb->ip_summed = CHECKSUM_COMPLETE; - } -} - static void myri10ge_alloc_rx_pages(struct myri10ge_priv *mgp, struct myri10ge_rx_buf *rx, int bytes, int watchdog) diff --git a/drivers/net/ethernet/neterion/Kconfig b/drivers/net/ethernet/neterion/Kconfig index 5e630f3a0189..a82a37094579 100644 --- a/drivers/net/ethernet/neterion/Kconfig +++ b/drivers/net/ethernet/neterion/Kconfig @@ -27,7 +27,7 @@ config S2IO on its age. More specific information on configuring the driver is in - <file:Documentation/networking/device_drivers/neterion/s2io.txt>. + <file:Documentation/networking/device_drivers/neterion/s2io.rst>. To compile this driver as a module, choose M here. The module will be called s2io. @@ -42,7 +42,7 @@ config VXGE labeled as either one, depending on its age. More specific information on configuring the driver is in - <file:Documentation/networking/device_drivers/neterion/vxge.txt>. + <file:Documentation/networking/device_drivers/neterion/vxge.rst>. To compile this driver as a module, choose M here. The module will be called vxge. diff --git a/drivers/net/ethernet/netronome/nfp/flower/action.c b/drivers/net/ethernet/netronome/nfp/flower/action.c index 1c76e1592ca2..ff844e5cc41f 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/action.c +++ b/drivers/net/ethernet/netronome/nfp/flower/action.c @@ -209,7 +209,7 @@ nfp_fl_output(struct nfp_app *app, struct nfp_fl_output *output, NFP_FL_OUT_FLAGS_USE_TUN); output->port = cpu_to_be32(NFP_FL_PORT_TYPE_TUN | tun_type); } else if (netif_is_lag_master(out_dev) && - priv->flower_ext_feats & NFP_FL_FEATS_LAG) { + priv->flower_en_feats & NFP_FL_ENABLE_LAG) { int gid; output->flags = cpu_to_be16(tmp_flags); @@ -956,7 +956,7 @@ nfp_flower_output_action(struct nfp_app *app, *a_len += sizeof(struct nfp_fl_output); - if (priv->flower_ext_feats & NFP_FL_FEATS_LAG) { + if (priv->flower_en_feats & NFP_FL_ENABLE_LAG) { /* nfp_fl_pre_lag returns -err or size of prelag action added. * This will be 0 if it is not egressing to a lag dev. */ diff --git a/drivers/net/ethernet/netronome/nfp/flower/cmsg.c b/drivers/net/ethernet/netronome/nfp/flower/cmsg.c index a595ddb92bff..a050cb898782 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/cmsg.c +++ b/drivers/net/ethernet/netronome/nfp/flower/cmsg.c @@ -264,7 +264,7 @@ nfp_flower_cmsg_process_one_rx(struct nfp_app *app, struct sk_buff *skb) nfp_flower_cmsg_portmod_rx(app, skb); break; case NFP_FLOWER_CMSG_TYPE_MERGE_HINT: - if (app_priv->flower_ext_feats & NFP_FL_FEATS_FLOW_MERGE) { + if (app_priv->flower_en_feats & NFP_FL_ENABLE_FLOW_MERGE) { nfp_flower_cmsg_merge_hint_rx(app, skb); break; } @@ -285,7 +285,7 @@ nfp_flower_cmsg_process_one_rx(struct nfp_app *app, struct sk_buff *skb) nfp_flower_stats_rlim_reply(app, skb); break; case NFP_FLOWER_CMSG_TYPE_LAG_CONFIG: - if (app_priv->flower_ext_feats & NFP_FL_FEATS_LAG) { + if (app_priv->flower_en_feats & NFP_FL_ENABLE_LAG) { skb_stored = nfp_flower_lag_unprocessed_msg(app, skb); break; } diff --git a/drivers/net/ethernet/netronome/nfp/flower/main.c b/drivers/net/ethernet/netronome/nfp/flower/main.c index d8ad9346a26a..c39327677a7d 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/main.c +++ b/drivers/net/ethernet/netronome/nfp/flower/main.c @@ -665,6 +665,77 @@ err_clear_nn: return err; } +static void nfp_flower_wait_host_bit(struct nfp_app *app) +{ + unsigned long err_at; + u64 feat; + int err; + + /* Wait for HOST_ACK flag bit to propagate */ + err_at = jiffies + msecs_to_jiffies(100); + do { + feat = nfp_rtsym_read_le(app->pf->rtbl, + "_abi_flower_combined_features_global", + &err); + if (time_is_before_eq_jiffies(err_at)) { + nfp_warn(app->cpp, + "HOST_ACK bit not propagated in FW.\n"); + break; + } + usleep_range(1000, 2000); + } while (!err && !(feat & NFP_FL_FEATS_HOST_ACK)); + + if (err) + nfp_warn(app->cpp, + "Could not read global features entry from FW\n"); +} + +static int nfp_flower_sync_feature_bits(struct nfp_app *app) +{ + struct nfp_flower_priv *app_priv = app->priv; + int err; + + /* Tell the firmware of the host supported features. */ + err = nfp_rtsym_write_le(app->pf->rtbl, "_abi_flower_host_mask", + app_priv->flower_ext_feats | + NFP_FL_FEATS_HOST_ACK); + if (!err) + nfp_flower_wait_host_bit(app); + else if (err != -ENOENT) + return err; + + /* Tell the firmware that the driver supports lag. */ + err = nfp_rtsym_write_le(app->pf->rtbl, + "_abi_flower_balance_sync_enable", 1); + if (!err) { + app_priv->flower_en_feats |= NFP_FL_ENABLE_LAG; + nfp_flower_lag_init(&app_priv->nfp_lag); + } else if (err == -ENOENT) { + nfp_warn(app->cpp, "LAG not supported by FW.\n"); + } else { + return err; + } + + if (app_priv->flower_ext_feats & NFP_FL_FEATS_FLOW_MOD) { + /* Tell the firmware that the driver supports flow merging. */ + err = nfp_rtsym_write_le(app->pf->rtbl, + "_abi_flower_merge_hint_enable", 1); + if (!err) { + app_priv->flower_en_feats |= NFP_FL_ENABLE_FLOW_MERGE; + nfp_flower_internal_port_init(app_priv); + } else if (err == -ENOENT) { + nfp_warn(app->cpp, + "Flow merge not supported by FW.\n"); + } else { + return err; + } + } else { + nfp_warn(app->cpp, "Flow mod/merge not supported by FW.\n"); + } + + return 0; +} + static int nfp_flower_init(struct nfp_app *app) { u64 version, features, ctx_count, num_mems; @@ -753,35 +824,15 @@ static int nfp_flower_init(struct nfp_app *app) if (err) app_priv->flower_ext_feats = 0; else - app_priv->flower_ext_feats = features; + app_priv->flower_ext_feats = features & NFP_FL_FEATS_HOST; - /* Tell the firmware that the driver supports lag. */ - err = nfp_rtsym_write_le(app->pf->rtbl, - "_abi_flower_balance_sync_enable", 1); - if (!err) { - app_priv->flower_ext_feats |= NFP_FL_FEATS_LAG; - nfp_flower_lag_init(&app_priv->nfp_lag); - } else if (err == -ENOENT) { - nfp_warn(app->cpp, "LAG not supported by FW.\n"); - } else { - goto err_cleanup_metadata; - } + err = nfp_flower_sync_feature_bits(app); + if (err) + goto err_cleanup; - if (app_priv->flower_ext_feats & NFP_FL_FEATS_FLOW_MOD) { - /* Tell the firmware that the driver supports flow merging. */ - err = nfp_rtsym_write_le(app->pf->rtbl, - "_abi_flower_merge_hint_enable", 1); - if (!err) { - app_priv->flower_ext_feats |= NFP_FL_FEATS_FLOW_MERGE; - nfp_flower_internal_port_init(app_priv); - } else if (err == -ENOENT) { - nfp_warn(app->cpp, "Flow merge not supported by FW.\n"); - } else { - goto err_lag_clean; - } - } else { - nfp_warn(app->cpp, "Flow mod/merge not supported by FW.\n"); - } + err = flow_indr_dev_register(nfp_flower_indr_setup_tc_cb, app); + if (err) + goto err_cleanup; if (app_priv->flower_ext_feats & NFP_FL_FEATS_VF_RLIM) nfp_flower_qos_init(app); @@ -792,10 +843,9 @@ static int nfp_flower_init(struct nfp_app *app) return 0; -err_lag_clean: - if (app_priv->flower_ext_feats & NFP_FL_FEATS_LAG) +err_cleanup: + if (app_priv->flower_en_feats & NFP_FL_ENABLE_LAG) nfp_flower_lag_cleanup(&app_priv->nfp_lag); -err_cleanup_metadata: nfp_flower_metadata_cleanup(app); err_free_app_priv: vfree(app->priv); @@ -810,13 +860,16 @@ static void nfp_flower_clean(struct nfp_app *app) skb_queue_purge(&app_priv->cmsg_skbs_low); flush_work(&app_priv->cmsg_work); + flow_indr_dev_unregister(nfp_flower_indr_setup_tc_cb, app, + nfp_flower_setup_indr_block_cb); + if (app_priv->flower_ext_feats & NFP_FL_FEATS_VF_RLIM) nfp_flower_qos_cleanup(app); - if (app_priv->flower_ext_feats & NFP_FL_FEATS_LAG) + if (app_priv->flower_en_feats & NFP_FL_ENABLE_LAG) nfp_flower_lag_cleanup(&app_priv->nfp_lag); - if (app_priv->flower_ext_feats & NFP_FL_FEATS_FLOW_MERGE) + if (app_priv->flower_en_feats & NFP_FL_ENABLE_FLOW_MERGE) nfp_flower_internal_port_cleanup(app_priv); nfp_flower_metadata_cleanup(app); @@ -886,7 +939,7 @@ static int nfp_flower_start(struct nfp_app *app) struct nfp_flower_priv *app_priv = app->priv; int err; - if (app_priv->flower_ext_feats & NFP_FL_FEATS_LAG) { + if (app_priv->flower_en_feats & NFP_FL_ENABLE_LAG) { err = nfp_flower_lag_reset(&app_priv->nfp_lag); if (err) return err; @@ -907,16 +960,12 @@ nfp_flower_netdev_event(struct nfp_app *app, struct net_device *netdev, struct nfp_flower_priv *app_priv = app->priv; int ret; - if (app_priv->flower_ext_feats & NFP_FL_FEATS_LAG) { + if (app_priv->flower_en_feats & NFP_FL_ENABLE_LAG) { ret = nfp_flower_lag_netdev_event(app_priv, netdev, event, ptr); if (ret & NOTIFY_STOP_MASK) return ret; } - ret = nfp_flower_reg_indir_block_handler(app, netdev, event); - if (ret & NOTIFY_STOP_MASK) - return ret; - ret = nfp_flower_internal_port_event_handler(app, netdev, event); if (ret & NOTIFY_STOP_MASK) return ret; diff --git a/drivers/net/ethernet/netronome/nfp/flower/main.h b/drivers/net/ethernet/netronome/nfp/flower/main.h index d55d0d33bc45..6c3dc3baf387 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/main.h +++ b/drivers/net/ethernet/netronome/nfp/flower/main.h @@ -44,8 +44,20 @@ struct nfp_app; #define NFP_FL_FEATS_FLOW_MOD BIT(5) #define NFP_FL_FEATS_PRE_TUN_RULES BIT(6) #define NFP_FL_FEATS_IPV6_TUN BIT(7) -#define NFP_FL_FEATS_FLOW_MERGE BIT(30) -#define NFP_FL_FEATS_LAG BIT(31) +#define NFP_FL_FEATS_HOST_ACK BIT(31) + +#define NFP_FL_ENABLE_FLOW_MERGE BIT(0) +#define NFP_FL_ENABLE_LAG BIT(1) + +#define NFP_FL_FEATS_HOST \ + (NFP_FL_FEATS_GENEVE | \ + NFP_FL_NBI_MTU_SETTING | \ + NFP_FL_FEATS_GENEVE_OPT | \ + NFP_FL_FEATS_VLAN_PCP | \ + NFP_FL_FEATS_VF_RLIM | \ + NFP_FL_FEATS_FLOW_MOD | \ + NFP_FL_FEATS_PRE_TUN_RULES | \ + NFP_FL_FEATS_IPV6_TUN) struct nfp_fl_mask_id { struct circ_buf mask_id_free_list; @@ -145,6 +157,7 @@ struct nfp_fl_internal_ports { * @mask_id_seed: Seed used for mask hash table * @flower_version: HW version of flower * @flower_ext_feats: Bitmap of extra features the HW supports + * @flower_en_feats: Bitmap of features enabled by HW * @stats_ids: List of free stats ids * @mask_ids: List of free mask ids * @mask_table: Hash table used to store masks @@ -180,6 +193,7 @@ struct nfp_flower_priv { u32 mask_id_seed; u64 flower_version; u64 flower_ext_feats; + u8 flower_en_feats; struct nfp_fl_stats_id stats_ids; struct nfp_fl_mask_id mask_ids; DECLARE_HASHTABLE(mask_table, NFP_FLOWER_MASK_HASH_BITS); @@ -346,7 +360,7 @@ nfp_flower_internal_port_can_offload(struct nfp_app *app, { struct nfp_flower_priv *app_priv = app->priv; - if (!(app_priv->flower_ext_feats & NFP_FL_FEATS_FLOW_MERGE)) + if (!(app_priv->flower_en_feats & NFP_FL_ENABLE_FLOW_MERGE)) return false; if (!netdev->rtnl_link_ops) return false; @@ -444,9 +458,10 @@ void nfp_flower_qos_cleanup(struct nfp_app *app); int nfp_flower_setup_qos_offload(struct nfp_app *app, struct net_device *netdev, struct tc_cls_matchall_offload *flow); void nfp_flower_stats_rlim_reply(struct nfp_app *app, struct sk_buff *skb); -int nfp_flower_reg_indir_block_handler(struct nfp_app *app, - struct net_device *netdev, - unsigned long event); +int nfp_flower_indr_setup_tc_cb(struct net_device *netdev, void *cb_priv, + enum tc_setup_type type, void *type_data); +int nfp_flower_setup_indr_block_cb(enum tc_setup_type type, void *type_data, + void *cb_priv); void __nfp_flower_non_repr_priv_get(struct nfp_flower_non_repr_priv *non_repr_priv); diff --git a/drivers/net/ethernet/netronome/nfp/flower/match.c b/drivers/net/ethernet/netronome/nfp/flower/match.c index 546bc01d507d..f7f01e2e3dce 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/match.c +++ b/drivers/net/ethernet/netronome/nfp/flower/match.c @@ -74,9 +74,10 @@ nfp_flower_compile_port(struct nfp_flower_in_port *frame, u32 cmsg_port, return 0; } -static void +static int nfp_flower_compile_mac(struct nfp_flower_mac_mpls *ext, - struct nfp_flower_mac_mpls *msk, struct flow_rule *rule) + struct nfp_flower_mac_mpls *msk, struct flow_rule *rule, + struct netlink_ext_ack *extack) { memset(ext, 0, sizeof(struct nfp_flower_mac_mpls)); memset(msk, 0, sizeof(struct nfp_flower_mac_mpls)); @@ -97,14 +98,28 @@ nfp_flower_compile_mac(struct nfp_flower_mac_mpls *ext, u32 t_mpls; flow_rule_match_mpls(rule, &match); - t_mpls = FIELD_PREP(NFP_FLOWER_MASK_MPLS_LB, match.key->mpls_label) | - FIELD_PREP(NFP_FLOWER_MASK_MPLS_TC, match.key->mpls_tc) | - FIELD_PREP(NFP_FLOWER_MASK_MPLS_BOS, match.key->mpls_bos) | + + /* Only support matching the first LSE */ + if (match.mask->used_lses != 1) { + NL_SET_ERR_MSG_MOD(extack, + "unsupported offload: invalid LSE depth for MPLS match offload"); + return -EOPNOTSUPP; + } + + t_mpls = FIELD_PREP(NFP_FLOWER_MASK_MPLS_LB, + match.key->ls[0].mpls_label) | + FIELD_PREP(NFP_FLOWER_MASK_MPLS_TC, + match.key->ls[0].mpls_tc) | + FIELD_PREP(NFP_FLOWER_MASK_MPLS_BOS, + match.key->ls[0].mpls_bos) | NFP_FLOWER_MASK_MPLS_Q; ext->mpls_lse = cpu_to_be32(t_mpls); - t_mpls = FIELD_PREP(NFP_FLOWER_MASK_MPLS_LB, match.mask->mpls_label) | - FIELD_PREP(NFP_FLOWER_MASK_MPLS_TC, match.mask->mpls_tc) | - FIELD_PREP(NFP_FLOWER_MASK_MPLS_BOS, match.mask->mpls_bos) | + t_mpls = FIELD_PREP(NFP_FLOWER_MASK_MPLS_LB, + match.mask->ls[0].mpls_label) | + FIELD_PREP(NFP_FLOWER_MASK_MPLS_TC, + match.mask->ls[0].mpls_tc) | + FIELD_PREP(NFP_FLOWER_MASK_MPLS_BOS, + match.mask->ls[0].mpls_bos) | NFP_FLOWER_MASK_MPLS_Q; msk->mpls_lse = cpu_to_be32(t_mpls); } else if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_BASIC)) { @@ -121,6 +136,8 @@ nfp_flower_compile_mac(struct nfp_flower_mac_mpls *ext, msk->mpls_lse = cpu_to_be32(NFP_FLOWER_MASK_MPLS_Q); } } + + return 0; } static void @@ -461,9 +478,12 @@ int nfp_flower_compile_flow_match(struct nfp_app *app, msk += sizeof(struct nfp_flower_in_port); if (NFP_FLOWER_LAYER_MAC & key_ls->key_layer) { - nfp_flower_compile_mac((struct nfp_flower_mac_mpls *)ext, - (struct nfp_flower_mac_mpls *)msk, - rule); + err = nfp_flower_compile_mac((struct nfp_flower_mac_mpls *)ext, + (struct nfp_flower_mac_mpls *)msk, + rule, extack); + if (err) + return err; + ext += sizeof(struct nfp_flower_mac_mpls); msk += sizeof(struct nfp_flower_mac_mpls); } diff --git a/drivers/net/ethernet/netronome/nfp/flower/offload.c b/drivers/net/ethernet/netronome/nfp/flower/offload.c index c694dbc239d0..695d24b9dd92 100644 --- a/drivers/net/ethernet/netronome/nfp/flower/offload.c +++ b/drivers/net/ethernet/netronome/nfp/flower/offload.c @@ -1440,7 +1440,8 @@ __nfp_flower_update_merge_stats(struct nfp_app *app, ctx_id = be32_to_cpu(sub_flow->meta.host_ctx_id); priv->stats[ctx_id].pkts += pkts; priv->stats[ctx_id].bytes += bytes; - max_t(u64, priv->stats[ctx_id].used, used); + priv->stats[ctx_id].used = max_t(u64, used, + priv->stats[ctx_id].used); } } @@ -1618,8 +1619,8 @@ nfp_flower_indr_block_cb_priv_lookup(struct nfp_app *app, return NULL; } -static int nfp_flower_setup_indr_block_cb(enum tc_setup_type type, - void *type_data, void *cb_priv) +int nfp_flower_setup_indr_block_cb(enum tc_setup_type type, + void *type_data, void *cb_priv) { struct nfp_flower_indr_block_cb_priv *priv = cb_priv; struct flow_cls_offload *flower = type_data; @@ -1707,10 +1708,13 @@ nfp_flower_setup_indr_tc_block(struct net_device *netdev, struct nfp_app *app, return 0; } -static int +int nfp_flower_indr_setup_tc_cb(struct net_device *netdev, void *cb_priv, enum tc_setup_type type, void *type_data) { + if (!nfp_fl_is_netdev_to_offload(netdev)) + return -EOPNOTSUPP; + switch (type) { case TC_SETUP_BLOCK: return nfp_flower_setup_indr_tc_block(netdev, cb_priv, @@ -1719,29 +1723,3 @@ nfp_flower_indr_setup_tc_cb(struct net_device *netdev, void *cb_priv, return -EOPNOTSUPP; } } - -int nfp_flower_reg_indir_block_handler(struct nfp_app *app, - struct net_device *netdev, - unsigned long event) -{ - int err; - - if (!nfp_fl_is_netdev_to_offload(netdev)) - return NOTIFY_OK; - - if (event == NETDEV_REGISTER) { - err = __flow_indr_block_cb_register(netdev, app, - nfp_flower_indr_setup_tc_cb, - app); - if (err) - nfp_flower_cmsg_warn(app, - "Indirect block reg failed - %s\n", - netdev->name); - } else if (event == NETDEV_UNREGISTER) { - __flow_indr_block_cb_unregister(netdev, - nfp_flower_indr_setup_tc_cb, - app); - } - - return NOTIFY_OK; -} diff --git a/drivers/net/ethernet/netronome/nfp/nfp_main.c b/drivers/net/ethernet/netronome/nfp/nfp_main.c index 4d282fc56009..7ff2ccbd43b0 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_main.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_main.c @@ -14,7 +14,6 @@ #include <linux/mutex.h> #include <linux/pci.h> #include <linux/firmware.h> -#include <linux/vermagic.h> #include <linux/vmalloc.h> #include <net/devlink.h> @@ -31,7 +30,6 @@ #include "nfp_net.h" static const char nfp_driver_name[] = "nfp"; -const char nfp_driver_version[] = VERMAGIC_STRING; static const struct pci_device_id nfp_pci_device_ids[] = { { PCI_VENDOR_ID_NETRONOME, PCI_DEVICE_ID_NETRONOME_NFP6000, @@ -920,4 +918,3 @@ MODULE_FIRMWARE("netronome/nic_AMDA0099-0001_1x10_1x25.nffw"); MODULE_AUTHOR("Netronome Systems <oss-drivers@netronome.com>"); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("The Netronome Flow Processor (NFP) driver."); -MODULE_VERSION(UTS_RELEASE); diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c index 9bfb3b077bc1..0e0cc3d58bdc 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_common.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_common.c @@ -1741,10 +1741,15 @@ nfp_net_tx_xdp_buf(struct nfp_net_dp *dp, struct nfp_net_rx_ring *rx_ring, struct nfp_net_rx_buf *rxbuf, unsigned int dma_off, unsigned int pkt_len, bool *completed) { + unsigned int dma_map_sz = dp->fl_bufsz - NFP_NET_RX_BUF_NON_DATA; struct nfp_net_tx_buf *txbuf; struct nfp_net_tx_desc *txd; int wr_idx; + /* Reject if xdp_adjust_tail grow packet beyond DMA area */ + if (pkt_len + dma_off > dma_map_sz) + return false; + if (unlikely(nfp_net_tx_full(tx_ring, 1))) { if (!*completed) { nfp_net_xdp_complete(tx_ring); @@ -1817,6 +1822,7 @@ static int nfp_net_rx(struct nfp_net_rx_ring *rx_ring, int budget) rcu_read_lock(); xdp_prog = READ_ONCE(dp->xdp_prog); true_bufsz = xdp_prog ? PAGE_SIZE : dp->fl_bufsz; + xdp.frame_sz = PAGE_SIZE - NFP_NET_RX_BUF_HEADROOM; xdp.rxq = &rx_ring->xdp_rxq; tx_ring = r_vec->xdp_ring; diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c index 2779f1526d1e..6eb9fb9a1814 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_ethtool.c @@ -203,8 +203,6 @@ nfp_get_drvinfo(struct nfp_app *app, struct pci_dev *pdev, char nsp_version[ETHTOOL_FWVERS_LEN] = {}; strlcpy(drvinfo->driver, pdev->driver->name, sizeof(drvinfo->driver)); - strlcpy(drvinfo->version, nfp_driver_version, sizeof(drvinfo->version)); - nfp_net_get_nspinfo(app, nsp_version); snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), "%s %s %s %s", vnic_version, nsp_version, @@ -1440,8 +1438,7 @@ static int nfp_net_set_channels(struct net_device *netdev, unsigned int total_rx, total_tx; /* Reject unsupported */ - if (!channel->combined_count || - channel->other_count != NFP_NET_NON_Q_VECTORS || + if (channel->other_count != NFP_NET_NON_Q_VECTORS || (channel->rx_count && channel->tx_count)) return -EINVAL; diff --git a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c index 79d72c88bbef..b3cabc274121 100644 --- a/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c +++ b/drivers/net/ethernet/netronome/nfp/nfp_net_repr.c @@ -299,6 +299,20 @@ static void nfp_repr_clean(struct nfp_repr *repr) nfp_port_free(repr->port); } +static struct lock_class_key nfp_repr_netdev_xmit_lock_key; + +static void nfp_repr_set_lockdep_class_one(struct net_device *dev, + struct netdev_queue *txq, + void *_unused) +{ + lockdep_set_class(&txq->_xmit_lock, &nfp_repr_netdev_xmit_lock_key); +} + +static void nfp_repr_set_lockdep_class(struct net_device *dev) +{ + netdev_for_each_tx_queue(dev, nfp_repr_set_lockdep_class_one, NULL); +} + int nfp_repr_init(struct nfp_app *app, struct net_device *netdev, u32 cmsg_port_id, struct nfp_port *port, struct net_device *pf_netdev) @@ -308,6 +322,8 @@ int nfp_repr_init(struct nfp_app *app, struct net_device *netdev, u32 repr_cap = nn->tlv_caps.repr_cap; int err; + nfp_repr_set_lockdep_class(netdev); + repr->port = port; repr->dst = metadata_dst_alloc(0, METADATA_HW_PORT_MUX, GFP_KERNEL); if (!repr->dst) diff --git a/drivers/net/ethernet/ni/nixge.c b/drivers/net/ethernet/ni/nixge.c index 2fdd0753b3af..d2708a57f2ff 100644 --- a/drivers/net/ethernet/ni/nixge.c +++ b/drivers/net/ethernet/ni/nixge.c @@ -502,7 +502,8 @@ static int nixge_check_tx_bd_space(struct nixge_priv *priv, return 0; } -static int nixge_start_xmit(struct sk_buff *skb, struct net_device *ndev) +static netdev_tx_t nixge_start_xmit(struct sk_buff *skb, + struct net_device *ndev) { struct nixge_priv *priv = netdev_priv(ndev); struct nixge_hw_dma_bd *cur_p; diff --git a/drivers/net/ethernet/nxp/lpc_eth.c b/drivers/net/ethernet/nxp/lpc_eth.c index d20cf03a3ea0..d3cbb4215f5c 100644 --- a/drivers/net/ethernet/nxp/lpc_eth.c +++ b/drivers/net/ethernet/nxp/lpc_eth.c @@ -823,7 +823,8 @@ static int lpc_mii_init(struct netdata_local *pldat) if (err) goto err_out_unregister_bus; - if (lpc_mii_probe(pldat->ndev) != 0) + err = lpc_mii_probe(pldat->ndev); + if (err) goto err_out_unregister_bus; return 0; @@ -1029,7 +1030,8 @@ static int lpc_eth_close(struct net_device *ndev) return 0; } -static int lpc_eth_hard_start_xmit(struct sk_buff *skb, struct net_device *ndev) +static netdev_tx_t lpc_eth_hard_start_xmit(struct sk_buff *skb, + struct net_device *ndev) { struct netdata_local *pldat = netdev_priv(ndev); u32 len, txidx; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.c b/drivers/net/ethernet/pensando/ionic/ionic_dev.c index f4ae40ae1e53..d83eff0ae0ac 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_dev.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.c @@ -388,6 +388,19 @@ int ionic_set_vf_config(struct ionic *ionic, int vf, u8 attr, u8 *data) } /* LIF commands */ +void ionic_dev_cmd_queue_identify(struct ionic_dev *idev, + u16 lif_type, u8 qtype, u8 qver) +{ + union ionic_dev_cmd cmd = { + .q_identify.opcode = IONIC_CMD_Q_IDENTIFY, + .q_identify.lif_type = lif_type, + .q_identify.type = qtype, + .q_identify.ver = qver, + }; + + ionic_dev_cmd_go(idev, &cmd); +} + void ionic_dev_cmd_lif_identify(struct ionic_dev *idev, u8 type, u8 ver) { union ionic_dev_cmd cmd = { @@ -431,6 +444,7 @@ void ionic_dev_cmd_adminq_init(struct ionic_dev *idev, struct ionic_qcq *qcq, .q_init.opcode = IONIC_CMD_Q_INIT, .q_init.lif_index = cpu_to_le16(lif_index), .q_init.type = q->type, + .q_init.ver = qcq->q.lif->qtype_info[q->type].version, .q_init.index = cpu_to_le32(q->index), .q_init.flags = cpu_to_le16(IONIC_QINIT_F_IRQ | IONIC_QINIT_F_ENA), diff --git a/drivers/net/ethernet/pensando/ionic/ionic_dev.h b/drivers/net/ethernet/pensando/ionic/ionic_dev.h index 587398b01997..525434f10025 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_dev.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_dev.h @@ -12,7 +12,8 @@ #define IONIC_MIN_MTU ETH_MIN_MTU #define IONIC_MAX_MTU 9194 -#define IONIC_MAX_TXRX_DESC 16384 +#define IONIC_MAX_TX_DESC 8192 +#define IONIC_MAX_RX_DESC 16384 #define IONIC_MIN_TXRX_DESC 16 #define IONIC_DEF_TXRX_DESC 4096 #define IONIC_LIFS_MAX 1024 @@ -83,6 +84,8 @@ static_assert(sizeof(struct ionic_q_init_cmd) == 64); static_assert(sizeof(struct ionic_q_init_comp) == 16); static_assert(sizeof(struct ionic_q_control_cmd) == 64); static_assert(sizeof(ionic_q_control_comp) == 16); +static_assert(sizeof(struct ionic_q_identify_cmd) == 64); +static_assert(sizeof(struct ionic_q_identify_comp) == 16); static_assert(sizeof(struct ionic_rx_mode_set_cmd) == 64); static_assert(sizeof(ionic_rx_mode_set_comp) == 16); @@ -179,7 +182,7 @@ struct ionic_desc_info { void *cb_arg; }; -#define QUEUE_NAME_MAX_SZ 32 +#define IONIC_QUEUE_NAME_MAX_SZ 32 struct ionic_queue { u64 dbell_count; @@ -204,14 +207,14 @@ struct ionic_queue { unsigned int desc_size; unsigned int sg_desc_size; unsigned int pid; - char name[QUEUE_NAME_MAX_SZ]; + char name[IONIC_QUEUE_NAME_MAX_SZ]; }; -#define INTR_INDEX_NOT_ASSIGNED -1 -#define INTR_NAME_MAX_SZ 32 +#define IONIC_INTR_INDEX_NOT_ASSIGNED -1 +#define IONIC_INTR_NAME_MAX_SZ 32 struct ionic_intr_info { - char name[INTR_NAME_MAX_SZ]; + char name[IONIC_INTR_NAME_MAX_SZ]; unsigned int index; unsigned int vector; u64 rearm_count; @@ -283,6 +286,8 @@ void ionic_dev_cmd_port_fec(struct ionic_dev *idev, u8 fec_type); void ionic_dev_cmd_port_pause(struct ionic_dev *idev, u8 pause_type); int ionic_set_vf_config(struct ionic *ionic, int vf, u8 attr, u8 *data); +void ionic_dev_cmd_queue_identify(struct ionic_dev *idev, + u16 lif_type, u8 qtype, u8 qver); void ionic_dev_cmd_lif_identify(struct ionic_dev *idev, u8 type, u8 ver); void ionic_dev_cmd_lif_init(struct ionic_dev *idev, u16 lif_index, dma_addr_t addr); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c index 6996229facfd..f7e3ce3de04d 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c @@ -12,10 +12,11 @@ #include "ionic_stats.h" static const char ionic_priv_flags_strings[][ETH_GSTRING_LEN] = { -#define PRIV_F_SW_DBG_STATS BIT(0) +#define IONIC_PRIV_F_SW_DBG_STATS BIT(0) "sw-dbg-stats", }; -#define PRIV_FLAGS_COUNT ARRAY_SIZE(ionic_priv_flags_strings) + +#define IONIC_PRIV_FLAGS_COUNT ARRAY_SIZE(ionic_priv_flags_strings) static void ionic_get_stats_strings(struct ionic_lif *lif, u8 *buf) { @@ -58,7 +59,7 @@ static int ionic_get_sset_count(struct net_device *netdev, int sset) count = ionic_get_stats_count(lif); break; case ETH_SS_PRIV_FLAGS: - count = PRIV_FLAGS_COUNT; + count = IONIC_PRIV_FLAGS_COUNT; break; } return count; @@ -75,7 +76,7 @@ static void ionic_get_strings(struct net_device *netdev, break; case ETH_SS_PRIV_FLAGS: memcpy(buf, ionic_priv_flags_strings, - PRIV_FLAGS_COUNT * ETH_GSTRING_LEN); + IONIC_PRIV_FLAGS_COUNT * ETH_GSTRING_LEN); break; } } @@ -159,6 +160,8 @@ static int ionic_get_link_ksettings(struct net_device *netdev, ethtool_link_ksettings_add_link_mode(ks, supported, 100000baseSR4_Full); break; + case IONIC_XCVR_PID_QSFP_100G_CWDM4: + case IONIC_XCVR_PID_QSFP_100G_PSM4: case IONIC_XCVR_PID_QSFP_100G_LR4: ethtool_link_ksettings_add_link_mode(ks, supported, 100000baseLR4_ER4_Full); @@ -178,6 +181,7 @@ static int ionic_get_link_ksettings(struct net_device *netdev, break; case IONIC_XCVR_PID_SFP_25GBASE_SR: case IONIC_XCVR_PID_SFP_25GBASE_AOC: + case IONIC_XCVR_PID_SFP_25GBASE_ACC: ethtool_link_ksettings_add_link_mode(ks, supported, 25000baseSR_Full); break; @@ -458,9 +462,9 @@ static void ionic_get_ringparam(struct net_device *netdev, { struct ionic_lif *lif = netdev_priv(netdev); - ring->tx_max_pending = IONIC_MAX_TXRX_DESC; + ring->tx_max_pending = IONIC_MAX_TX_DESC; ring->tx_pending = lif->ntxq_descs; - ring->rx_max_pending = IONIC_MAX_TXRX_DESC; + ring->rx_max_pending = IONIC_MAX_RX_DESC; ring->rx_pending = lif->nrxq_descs; } @@ -554,7 +558,7 @@ static u32 ionic_get_priv_flags(struct net_device *netdev) u32 priv_flags = 0; if (test_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state)) - priv_flags |= PRIV_F_SW_DBG_STATS; + priv_flags |= IONIC_PRIV_F_SW_DBG_STATS; return priv_flags; } @@ -564,7 +568,7 @@ static int ionic_set_priv_flags(struct net_device *netdev, u32 priv_flags) struct ionic_lif *lif = netdev_priv(netdev); clear_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state); - if (priv_flags & PRIV_F_SW_DBG_STATS) + if (priv_flags & IONIC_PRIV_F_SW_DBG_STATS) set_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state); return 0; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_if.h b/drivers/net/ethernet/pensando/ionic/ionic_if.h index ceeb7629e7a0..7e22ba4ed915 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_if.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_if.h @@ -1,5 +1,5 @@ /* SPDX-License-Identifier: (GPL-2.0 OR Linux-OpenIB) OR BSD-2-Clause */ -/* Copyright (c) 2017-2019 Pensando Systems, Inc. All rights reserved. */ +/* Copyright (c) 2017-2020 Pensando Systems, Inc. All rights reserved. */ #ifndef _IONIC_IF_H_ #define _IONIC_IF_H_ @@ -9,7 +9,7 @@ #define IONIC_IFNAMSIZ 16 /** - * Commands + * enum ionic_cmd_opcode - Device commands */ enum ionic_cmd_opcode { IONIC_CMD_NOP = 0, @@ -40,6 +40,7 @@ enum ionic_cmd_opcode { IONIC_CMD_RX_FILTER_DEL = 32, /* Queue commands */ + IONIC_CMD_Q_IDENTIFY = 39, IONIC_CMD_Q_INIT = 40, IONIC_CMD_Q_CONTROL = 41, @@ -57,6 +58,7 @@ enum ionic_cmd_opcode { IONIC_CMD_QOS_CLASS_IDENTIFY = 240, IONIC_CMD_QOS_CLASS_INIT = 241, IONIC_CMD_QOS_CLASS_RESET = 242, + IONIC_CMD_QOS_CLASS_UPDATE = 243, /* Firmware commands */ IONIC_CMD_FW_DOWNLOAD = 254, @@ -64,7 +66,7 @@ enum ionic_cmd_opcode { }; /** - * Command Return codes + * enum ionic_status_code - Device command return codes */ enum ionic_status_code { IONIC_RC_SUCCESS = 0, /* Success */ @@ -97,6 +99,7 @@ enum ionic_notifyq_opcode { IONIC_EVENT_RESET = 2, IONIC_EVENT_HEARTBEAT = 3, IONIC_EVENT_LOG = 4, + IONIC_EVENT_XCVR = 5, }; /** @@ -114,12 +117,11 @@ struct ionic_admin_cmd { /** * struct ionic_admin_comp - General admin command completion format - * @status: The status of the command (enum status_code) - * @comp_index: The index in the descriptor ring for which this - * is the completion. - * @cmd_data: Command-specific bytes. - * @color: Color bit. (Always 0 for commands issued to the - * Device Cmd Registers.) + * @status: Status of the command (enum ionic_status_code) + * @comp_index: Index in the descriptor ring for which this is the completion + * @cmd_data: Command-specific bytes + * @color: Color bit (Always 0 for commands issued to the + * Device Cmd Registers) */ struct ionic_admin_comp { u8 status; @@ -146,7 +148,7 @@ struct ionic_nop_cmd { /** * struct ionic_nop_comp - NOP command completion - * @status: The status of the command (enum status_code) + * @status: Status of the command (enum ionic_status_code) */ struct ionic_nop_comp { u8 status; @@ -156,7 +158,7 @@ struct ionic_nop_comp { /** * struct ionic_dev_init_cmd - Device init command * @opcode: opcode - * @type: device type + * @type: Device type */ struct ionic_dev_init_cmd { u8 opcode; @@ -166,7 +168,7 @@ struct ionic_dev_init_cmd { /** * struct init_comp - Device init command completion - * @status: The status of the command (enum status_code) + * @status: Status of the command (enum ionic_status_code) */ struct ionic_dev_init_comp { u8 status; @@ -184,7 +186,7 @@ struct ionic_dev_reset_cmd { /** * struct reset_comp - Reset command completion - * @status: The status of the command (enum status_code) + * @status: Status of the command (enum ionic_status_code) */ struct ionic_dev_reset_comp { u8 status; @@ -205,8 +207,8 @@ struct ionic_dev_identify_cmd { }; /** - * struct dev_identify_comp - Driver/device identify command completion - * @status: The status of the command (enum status_code) + * struct ionic_dev_identify_comp - Driver/device identify command completion + * @status: Status of the command (enum ionic_status_code) * @ver: Version of identify returned by device */ struct ionic_dev_identify_comp { @@ -225,8 +227,8 @@ enum ionic_os_type { }; /** - * union drv_identity - driver identity information - * @os_type: OS type (see enum os_type) + * union ionic_drv_identity - driver identity information + * @os_type: OS type (see enum ionic_os_type) * @os_dist: OS distribution, numeric format * @os_dist_str: OS distribution, string format * @kernel_ver: Kernel version, numeric format @@ -242,26 +244,26 @@ union ionic_drv_identity { char kernel_ver_str[32]; char driver_ver_str[32]; }; - __le32 words[512]; + __le32 words[478]; }; /** - * union dev_identity - device identity information + * union ionic_dev_identity - device identity information * @version: Version of device identify * @type: Identify type (0 for now) * @nports: Number of ports provisioned * @nlifs: Number of LIFs provisioned * @nintrs: Number of interrupts provisioned * @ndbpgs_per_lif: Number of doorbell pages per LIF - * @intr_coal_mult: Interrupt coalescing multiplication factor. + * @intr_coal_mult: Interrupt coalescing multiplication factor * Scale user-supplied interrupt coalescing * value in usecs to device units using: * device units = usecs * mult / div - * @intr_coal_div: Interrupt coalescing division factor. + * @intr_coal_div: Interrupt coalescing division factor * Scale user-supplied interrupt coalescing * value in usecs to device units using: * device units = usecs * mult / div - * + * @eq_count: Number of shared event queues */ union ionic_dev_identity { struct { @@ -275,8 +277,9 @@ union ionic_dev_identity { __le32 ndbpgs_per_lif; __le32 intr_coal_mult; __le32 intr_coal_div; + __le32 eq_count; }; - __le32 words[512]; + __le32 words[478]; }; enum ionic_lif_type { @@ -286,10 +289,10 @@ enum ionic_lif_type { }; /** - * struct ionic_lif_identify_cmd - lif identify command + * struct ionic_lif_identify_cmd - LIF identify command * @opcode: opcode - * @type: lif type (enum lif_type) - * @ver: version of identify returned by device + * @type: LIF type (enum ionic_lif_type) + * @ver: Version of identify returned by device */ struct ionic_lif_identify_cmd { u8 opcode; @@ -299,9 +302,9 @@ struct ionic_lif_identify_cmd { }; /** - * struct ionic_lif_identify_comp - lif identify command completion - * @status: status of the command (enum status_code) - * @ver: version of identify returned by device + * struct ionic_lif_identify_comp - LIF identify command completion + * @status: Status of the command (enum ionic_status_code) + * @ver: Version of identify returned by device */ struct ionic_lif_identify_comp { u8 status; @@ -309,13 +312,24 @@ struct ionic_lif_identify_comp { u8 rsvd2[14]; }; +/** + * enum ionic_lif_capability - LIF capabilities + * @IONIC_LIF_CAP_ETH: LIF supports Ethernet + * @IONIC_LIF_CAP_RDMA: LIF support RDMA + */ enum ionic_lif_capability { IONIC_LIF_CAP_ETH = BIT(0), IONIC_LIF_CAP_RDMA = BIT(1), }; /** - * Logical Queue Types + * enum ionic_logical_qtype - Logical Queue Types + * @IONIC_QTYPE_ADMINQ: Administrative Queue + * @IONIC_QTYPE_NOTIFYQ: Notify Queue + * @IONIC_QTYPE_RXQ: Receive Queue + * @IONIC_QTYPE_TXQ: Transmit Queue + * @IONIC_QTYPE_EQ: Event Queue + * @IONIC_QTYPE_MAX: Max queue type supported */ enum ionic_logical_qtype { IONIC_QTYPE_ADMINQ = 0, @@ -327,10 +341,10 @@ enum ionic_logical_qtype { }; /** - * struct ionic_lif_logical_qtype - Descriptor of logical to hardware queue type. - * @qtype: Hardware Queue Type. - * @qid_count: Number of Queue IDs of the logical type. - * @qid_base: Minimum Queue ID of the logical type. + * struct ionic_lif_logical_qtype - Descriptor of logical to HW queue type + * @qtype: Hardware Queue Type + * @qid_count: Number of Queue IDs of the logical type + * @qid_base: Minimum Queue ID of the logical type */ struct ionic_lif_logical_qtype { u8 qtype; @@ -339,6 +353,12 @@ struct ionic_lif_logical_qtype { __le32 qid_base; }; +/** + * enum ionic_lif_state - LIF state + * @IONIC_LIF_DISABLE: LIF disabled + * @IONIC_LIF_ENABLE: LIF enabled + * @IONIC_LIF_HANG_RESET: LIF hung, being reset + */ enum ionic_lif_state { IONIC_LIF_DISABLE = 0, IONIC_LIF_ENABLE = 1, @@ -346,13 +366,13 @@ enum ionic_lif_state { }; /** - * LIF configuration - * @state: lif state (enum lif_state) - * @name: lif name - * @mtu: mtu - * @mac: station mac address - * @features: features (enum ionic_eth_hw_features) - * @queue_count: queue counts per queue-type + * union ionic_lif_config - LIF configuration + * @state: LIF state (enum ionic_lif_state) + * @name: LIF name + * @mtu: MTU + * @mac: Station MAC address + * @features: Features (enum ionic_eth_hw_features) + * @queue_count: Queue counts per queue-type */ union ionic_lif_config { struct { @@ -369,37 +389,36 @@ union ionic_lif_config { }; /** - * struct ionic_lif_identity - lif identity information (type-specific) + * struct ionic_lif_identity - LIF identity information (type-specific) * - * @capabilities LIF capabilities + * @capabilities: LIF capabilities * - * Ethernet: - * @version: Ethernet identify structure version. - * @features: Ethernet features supported on this lif type. - * @max_ucast_filters: Number of perfect unicast addresses supported. - * @max_mcast_filters: Number of perfect multicast addresses supported. - * @min_frame_size: Minimum size of frames to be sent - * @max_frame_size: Maximim size of frames to be sent - * @config: LIF config struct with features, mtu, mac, q counts + * @eth: Ethernet identify structure + * @version: Ethernet identify structure version + * @max_ucast_filters: Number of perfect unicast addresses supported + * @max_mcast_filters: Number of perfect multicast addresses supported + * @min_frame_size: Minimum size of frames to be sent + * @max_frame_size: Maximim size of frames to be sent + * @config: LIF config struct with features, mtu, mac, q counts * - * RDMA: - * @version: RDMA version of opcodes and queue descriptors. - * @qp_opcodes: Number of rdma queue pair opcodes supported. - * @admin_opcodes: Number of rdma admin opcodes supported. - * @npts_per_lif: Page table size per lif - * @nmrs_per_lif: Number of memory regions per lif - * @nahs_per_lif: Number of address handles per lif - * @max_stride: Max work request stride. - * @cl_stride: Cache line stride. - * @pte_stride: Page table entry stride. - * @rrq_stride: Remote RQ work request stride. - * @rsq_stride: Remote SQ work request stride. + * @rdma: RDMA identify structure + * @version: RDMA version of opcodes and queue descriptors + * @qp_opcodes: Number of RDMA queue pair opcodes supported + * @admin_opcodes: Number of RDMA admin opcodes supported + * @npts_per_lif: Page table size per LIF + * @nmrs_per_lif: Number of memory regions per LIF + * @nahs_per_lif: Number of address handles per LIF + * @max_stride: Max work request stride + * @cl_stride: Cache line stride + * @pte_stride: Page table entry stride + * @rrq_stride: Remote RQ work request stride + * @rsq_stride: Remote SQ work request stride * @dcqcn_profiles: Number of DCQCN profiles - * @aq_qtype: RDMA Admin Qtype. - * @sq_qtype: RDMA Send Qtype. - * @rq_qtype: RDMA Receive Qtype. - * @cq_qtype: RDMA Completion Qtype. - * @eq_qtype: RDMA Event Qtype. + * @aq_qtype: RDMA Admin Qtype + * @sq_qtype: RDMA Send Qtype + * @rq_qtype: RDMA Receive Qtype + * @cq_qtype: RDMA Completion Qtype + * @eq_qtype: RDMA Event Qtype */ union ionic_lif_identity { struct { @@ -439,15 +458,15 @@ union ionic_lif_identity { struct ionic_lif_logical_qtype eq_qtype; } __packed rdma; } __packed; - __le32 words[512]; + __le32 words[478]; }; /** * struct ionic_lif_init_cmd - LIF init command - * @opcode: opcode - * @type: LIF type (enum lif_type) + * @opcode: Opcode + * @type: LIF type (enum ionic_lif_type) * @index: LIF index - * @info_pa: destination address for lif info (struct ionic_lif_info) + * @info_pa: Destination address for LIF info (struct ionic_lif_info) */ struct ionic_lif_init_cmd { u8 opcode; @@ -460,7 +479,8 @@ struct ionic_lif_init_cmd { /** * struct ionic_lif_init_comp - LIF init command completion - * @status: The status of the command (enum status_code) + * @status: Status of the command (enum ionic_status_code) + * @hw_index: Hardware index of the initialized LIF */ struct ionic_lif_init_comp { u8 status; @@ -469,14 +489,74 @@ struct ionic_lif_init_comp { u8 rsvd2[12]; }; + /** + * struct ionic_q_identify_cmd - queue identify command + * @opcode: opcode + * @lif_type: LIF type (enum ionic_lif_type) + * @type: Logical queue type (enum ionic_logical_qtype) + * @ver: Highest queue type version that the driver supports + */ +struct ionic_q_identify_cmd { + u8 opcode; + u8 rsvd; + __le16 lif_type; + u8 type; + u8 ver; + u8 rsvd2[58]; +}; + +/** + * struct ionic_q_identify_comp - queue identify command completion + * @status: Status of the command (enum ionic_status_code) + * @comp_index: Index in the descriptor ring for which this is the completion + * @ver: Queue type version that can be used with FW + */ +struct ionic_q_identify_comp { + u8 status; + u8 rsvd; + __le16 comp_index; + u8 ver; + u8 rsvd2[11]; +}; + +/** + * union ionic_q_identity - queue identity information + * @version: Queue type version that can be used with FW + * @supported: Bitfield of queue versions, first bit = ver 0 + * @features: Queue features + * @desc_sz: Descriptor size + * @comp_sz: Completion descriptor size + * @sg_desc_sz: Scatter/Gather descriptor size + * @max_sg_elems: Maximum number of Scatter/Gather elements + * @sg_desc_stride: Number of Scatter/Gather elements per descriptor + */ +union ionic_q_identity { + struct { + u8 version; + u8 supported; + u8 rsvd[6]; +#define IONIC_QIDENT_F_CQ 0x01 /* queue has completion ring */ +#define IONIC_QIDENT_F_SG 0x02 /* queue has scatter/gather ring */ +#define IONIC_QIDENT_F_EQ 0x04 /* queue can use event queue */ +#define IONIC_QIDENT_F_CMB 0x08 /* queue is in cmb bar */ + __le64 features; + __le16 desc_sz; + __le16 comp_sz; + __le16 sg_desc_sz; + __le16 max_sg_elems; + __le16 sg_desc_stride; + }; + __le32 words[478]; +}; + /** * struct ionic_q_init_cmd - Queue init command * @opcode: opcode * @type: Logical queue type - * @ver: Queue version (defines opcode/descriptor scope) + * @ver: Queue type version * @lif_index: LIF index - * @index: (lif, qtype) relative admin queue index - * @intr_index: Interrupt control register index + * @index: (LIF, qtype) relative admin queue index + * @intr_index: Interrupt control register index, or Event queue index * @pid: Process ID * @flags: * IRQ: Interrupt requested on completion @@ -494,12 +574,11 @@ struct ionic_lif_init_comp { * descriptors. Values of ring_size <2 and >16 are * reserved. * EQ: Enable the Event Queue - * @cos: Class of service for this queue. + * @cos: Class of service for this queue * @ring_size: Queue ring size, encoded as a log2(size) * @ring_base: Queue ring base address * @cq_ring_base: Completion queue ring base address * @sg_ring_base: Scatter/Gather ring base address - * @eq_index: Event queue index */ struct ionic_q_init_cmd { u8 opcode; @@ -516,29 +595,27 @@ struct ionic_q_init_cmd { #define IONIC_QINIT_F_ENA 0x02 /* Enable the queue */ #define IONIC_QINIT_F_SG 0x04 /* Enable scatter/gather on the queue */ #define IONIC_QINIT_F_EQ 0x08 /* Enable event queue */ -#define IONIC_QINIT_F_DEBUG 0x80 /* Enable queue debugging */ +#define IONIC_QINIT_F_CMB 0x10 /* Enable cmb-based queue */ +#define IONIC_QINIT_F_DEBUG 0x80 /* Enable queue debugging */ u8 cos; u8 ring_size; __le64 ring_base; __le64 cq_ring_base; __le64 sg_ring_base; - __le32 eq_index; - u8 rsvd2[16]; + u8 rsvd2[20]; } __packed; /** * struct ionic_q_init_comp - Queue init command completion - * @status: The status of the command (enum status_code) - * @ver: Queue version (defines opcode/descriptor scope) - * @comp_index: The index in the descriptor ring for which this - * is the completion. + * @status: Status of the command (enum ionic_status_code) + * @comp_index: Index in the descriptor ring for which this is the completion * @hw_index: Hardware Queue ID * @hw_type: Hardware Queue type * @color: Color */ struct ionic_q_init_comp { u8 status; - u8 ver; + u8 rsvd; __le16 comp_index; __le32 hw_index; u8 hw_type; @@ -559,10 +636,9 @@ enum ionic_txq_desc_opcode { /** * struct ionic_txq_desc - Ethernet Tx queue descriptor format - * @opcode: Tx operation, see TXQ_DESC_OPCODE_*: + * @cmd: Tx operation, see IONIC_TXQ_DESC_OPCODE_*: * * IONIC_TXQ_DESC_OPCODE_CSUM_NONE: - * * Non-offload send. No segmentation, * fragmentation or checksum calc/insertion is * performed by device; packet is prepared @@ -570,7 +646,6 @@ enum ionic_txq_desc_opcode { * no further manipulation from device. * * IONIC_TXQ_DESC_OPCODE_CSUM_PARTIAL: - * * Offload 16-bit L4 checksum * calculation/insertion. The device will * calculate the L4 checksum value and @@ -579,14 +654,16 @@ enum ionic_txq_desc_opcode { * is calculated starting at @csum_start bytes * into the packet to the end of the packet. * The checksum insertion position is given - * in @csum_offset. This feature is only - * applicable to protocols such as TCP, UDP - * and ICMP where a standard (i.e. the - * 'IP-style' checksum) one's complement - * 16-bit checksum is used, using an IP - * pseudo-header to seed the calculation. - * Software will preload the L4 checksum - * field with the IP pseudo-header checksum. + * in @csum_offset, which is the offset from + * @csum_start to the checksum field in the L4 + * header. This feature is only applicable to + * protocols such as TCP, UDP and ICMP where a + * standard (i.e. the 'IP-style' checksum) + * one's complement 16-bit checksum is used, + * using an IP pseudo-header to seed the + * calculation. Software will preload the L4 + * checksum field with the IP pseudo-header + * checksum. * * For tunnel encapsulation, @csum_start and * @csum_offset refer to the inner L4 @@ -602,7 +679,6 @@ enum ionic_txq_desc_opcode { * for more info). * * IONIC_TXQ_DESC_OPCODE_CSUM_HW: - * * Offload 16-bit checksum computation to hardware. * If @csum_l3 is set then the packet's L3 checksum is * updated. Similarly, if @csum_l4 is set the the L4 @@ -610,7 +686,6 @@ enum ionic_txq_desc_opcode { * checksums are also updated. * * IONIC_TXQ_DESC_OPCODE_TSO: - * * Device preforms TCP segmentation offload * (TSO). @hdr_len is the number of bytes * to the end of TCP header (the offset to @@ -637,40 +712,41 @@ enum ionic_txq_desc_opcode { * clear CWR in remaining segments. * @flags: * vlan: - * Insert an L2 VLAN header using @vlan_tci. + * Insert an L2 VLAN header using @vlan_tci * encap: - * Calculate encap header checksum. + * Calculate encap header checksum * csum_l3: - * Compute L3 header checksum. + * Compute L3 header checksum * csum_l4: - * Compute L4 header checksum. + * Compute L4 header checksum * tso_sot: * TSO start * tso_eot: * TSO end * @num_sg_elems: Number of scatter-gather elements in SG * descriptor - * @addr: First data buffer's DMA address. - * (Subsequent data buffers are on txq_sg_desc). + * @addr: First data buffer's DMA address + * (Subsequent data buffers are on txq_sg_desc) * @len: First data buffer's length, in bytes * @vlan_tci: VLAN tag to insert in the packet (if requested * by @V-bit). Includes .1p and .1q tags * @hdr_len: Length of packet headers, including - * encapsulating outer header, if applicable. - * Valid for opcodes TXQ_DESC_OPCODE_CALC_CSUM and - * TXQ_DESC_OPCODE_TSO. Should be set to zero for + * encapsulating outer header, if applicable + * Valid for opcodes IONIC_TXQ_DESC_OPCODE_CALC_CSUM and + * IONIC_TXQ_DESC_OPCODE_TSO. Should be set to zero for * all other modes. For - * TXQ_DESC_OPCODE_CALC_CSUM, @hdr_len is length + * IONIC_TXQ_DESC_OPCODE_CALC_CSUM, @hdr_len is length * of headers up to inner-most L4 header. For - * TXQ_DESC_OPCODE_TSO, @hdr_len is up to + * IONIC_TXQ_DESC_OPCODE_TSO, @hdr_len is up to * inner-most L4 payload, so inclusive of * inner-most L4 header. - * @mss: Desired MSS value for TSO. Only applicable for - * TXQ_DESC_OPCODE_TSO. - * @csum_start: Offset into inner-most L3 header of checksum - * @csum_offset: Offset into inner-most L4 header of checksum + * @mss: Desired MSS value for TSO; only applicable for + * IONIC_TXQ_DESC_OPCODE_TSO + * @csum_start: Offset from packet to first byte checked in L4 checksum + * @csum_offset: Offset from csum_start to L4 checksum field */ - +struct ionic_txq_desc { + __le64 cmd; #define IONIC_TXQ_DESC_OPCODE_MASK 0xf #define IONIC_TXQ_DESC_OPCODE_SHIFT 4 #define IONIC_TXQ_DESC_FLAGS_MASK 0xf @@ -692,8 +768,6 @@ enum ionic_txq_desc_opcode { #define IONIC_TXQ_DESC_FLAG_TSO_SOT 0x4 #define IONIC_TXQ_DESC_FLAG_TSO_EOT 0x8 -struct ionic_txq_desc { - __le64 cmd; __le16 len; union { __le16 vlan_tci; @@ -733,28 +807,38 @@ static inline void decode_txq_desc_cmd(u64 cmd, u8 *opcode, u8 *flags, *addr = (cmd >> IONIC_TXQ_DESC_ADDR_SHIFT) & IONIC_TXQ_DESC_ADDR_MASK; }; -#define IONIC_TX_MAX_SG_ELEMS 8 -#define IONIC_RX_MAX_SG_ELEMS 8 - /** - * struct ionic_txq_sg_desc - Transmit scatter-gather (SG) list + * struct ionic_txq_sg_elem - Transmit scatter-gather (SG) descriptor element * @addr: DMA address of SG element data buffer * @len: Length of SG element data buffer, in bytes */ +struct ionic_txq_sg_elem { + __le64 addr; + __le16 len; + __le16 rsvd[3]; +}; + +/** + * struct ionic_txq_sg_desc - Transmit scatter-gather (SG) list + * @elems: Scatter-gather elements + */ struct ionic_txq_sg_desc { - struct ionic_txq_sg_elem { - __le64 addr; - __le16 len; - __le16 rsvd[3]; - } elems[IONIC_TX_MAX_SG_ELEMS]; +#define IONIC_TX_MAX_SG_ELEMS 8 +#define IONIC_TX_SG_DESC_STRIDE 8 + struct ionic_txq_sg_elem elems[IONIC_TX_MAX_SG_ELEMS]; +}; + +struct ionic_txq_sg_desc_v1 { +#define IONIC_TX_MAX_SG_ELEMS_V1 15 +#define IONIC_TX_SG_DESC_STRIDE_V1 16 + struct ionic_txq_sg_elem elems[IONIC_TX_SG_DESC_STRIDE_V1]; }; /** * struct ionic_txq_comp - Ethernet transmit queue completion descriptor - * @status: The status of the command (enum status_code) - * @comp_index: The index in the descriptor ring for which this - * is the completion. - * @color: Color bit. + * @status: Status of the command (enum ionic_status_code) + * @comp_index: Index in the descriptor ring for which this is the completion + * @color: Color bit */ struct ionic_txq_comp { u8 status; @@ -771,16 +855,15 @@ enum ionic_rxq_desc_opcode { /** * struct ionic_rxq_desc - Ethernet Rx queue descriptor format - * @opcode: Rx operation, see RXQ_DESC_OPCODE_*: - * - * RXQ_DESC_OPCODE_SIMPLE: + * @opcode: Rx operation, see IONIC_RXQ_DESC_OPCODE_*: * + * IONIC_RXQ_DESC_OPCODE_SIMPLE: * Receive full packet into data buffer * starting at @addr. Results of * receive, including actual bytes received, * are recorded in Rx completion descriptor. * - * @len: Data buffer's length, in bytes. + * @len: Data buffer's length, in bytes * @addr: Data buffer's DMA address */ struct ionic_rxq_desc { @@ -791,26 +874,33 @@ struct ionic_rxq_desc { }; /** - * struct ionic_rxq_sg_desc - Receive scatter-gather (SG) list + * struct ionic_rxq_sg_elem - Receive scatter-gather (SG) descriptor element * @addr: DMA address of SG element data buffer * @len: Length of SG element data buffer, in bytes */ +struct ionic_rxq_sg_elem { + __le64 addr; + __le16 len; + __le16 rsvd[3]; +}; + +/** + * struct ionic_rxq_sg_desc - Receive scatter-gather (SG) list + * @elems: Scatter-gather elements + */ struct ionic_rxq_sg_desc { - struct ionic_rxq_sg_elem { - __le64 addr; - __le16 len; - __le16 rsvd[3]; - } elems[IONIC_RX_MAX_SG_ELEMS]; +#define IONIC_RX_MAX_SG_ELEMS 8 +#define IONIC_RX_SG_DESC_STRIDE 8 + struct ionic_rxq_sg_elem elems[IONIC_RX_SG_DESC_STRIDE]; }; /** * struct ionic_rxq_comp - Ethernet receive queue completion descriptor - * @status: The status of the command (enum status_code) + * @status: Status of the command (enum ionic_status_code) * @num_sg_elems: Number of SG elements used by this descriptor - * @comp_index: The index in the descriptor ring for which this - * is the completion. + * @comp_index: Index in the descriptor ring for which this is the completion * @rss_hash: 32-bit RSS hash - * @csum: 16-bit sum of the packet's L2 payload. + * @csum: 16-bit sum of the packet's L2 payload * If the packet's L2 payload is odd length, an extra * zero-value byte is included in the @csum calculation but * not included in @len. @@ -818,33 +908,51 @@ struct ionic_rxq_sg_desc { * set. Includes .1p and .1q tags. * @len: Received packet length, in bytes. Excludes FCS. * @csum_calc L2 payload checksum is computed or not - * @csum_tcp_ok: The TCP checksum calculated by the device - * matched the checksum in the receive packet's - * TCP header - * @csum_tcp_bad: The TCP checksum calculated by the device did - * not match the checksum in the receive packet's - * TCP header. - * @csum_udp_ok: The UDP checksum calculated by the device - * matched the checksum in the receive packet's - * UDP header - * @csum_udp_bad: The UDP checksum calculated by the device did - * not match the checksum in the receive packet's - * UDP header. - * @csum_ip_ok: The IPv4 checksum calculated by the device - * matched the checksum in the receive packet's - * first IPv4 header. If the receive packet - * contains both a tunnel IPv4 header and a - * transport IPv4 header, the device validates the - * checksum for the both IPv4 headers. - * @csum_ip_bad: The IPv4 checksum calculated by the device did - * not match the checksum in the receive packet's - * first IPv4 header. If the receive packet - * contains both a tunnel IPv4 header and a - * transport IPv4 header, the device validates the - * checksum for both IP headers. - * @VLAN: VLAN header was stripped and placed in @vlan_tci. - * @pkt_type: Packet type - * @color: Color bit. + * @csum_flags: See IONIC_RXQ_COMP_CSUM_F_*: + * + * IONIC_RXQ_COMP_CSUM_F_TCP_OK: + * The TCP checksum calculated by the device + * matched the checksum in the receive packet's + * TCP header. + * + * IONIC_RXQ_COMP_CSUM_F_TCP_BAD: + * The TCP checksum calculated by the device did + * not match the checksum in the receive packet's + * TCP header. + * + * IONIC_RXQ_COMP_CSUM_F_UDP_OK: + * The UDP checksum calculated by the device + * matched the checksum in the receive packet's + * UDP header + * + * IONIC_RXQ_COMP_CSUM_F_UDP_BAD: + * The UDP checksum calculated by the device did + * not match the checksum in the receive packet's + * UDP header. + * + * IONIC_RXQ_COMP_CSUM_F_IP_OK: + * The IPv4 checksum calculated by the device + * matched the checksum in the receive packet's + * first IPv4 header. If the receive packet + * contains both a tunnel IPv4 header and a + * transport IPv4 header, the device validates the + * checksum for the both IPv4 headers. + * + * IONIC_RXQ_COMP_CSUM_F_IP_BAD: + * The IPv4 checksum calculated by the device did + * not match the checksum in the receive packet's + * first IPv4 header. If the receive packet + * contains both a tunnel IPv4 header and a + * transport IPv4 header, the device validates the + * checksum for both IP headers. + * + * IONIC_RXQ_COMP_CSUM_F_VLAN: + * The VLAN header was stripped and placed in @vlan_tci. + * + * IONIC_RXQ_COMP_CSUM_F_CALC: + * The checksum was calculated by the device. + * + * @pkt_type_color: Packet type and color bit; see IONIC_RXQ_COMP_PKT_TYPE_MASK */ struct ionic_rxq_comp { u8 status; @@ -891,8 +999,8 @@ enum ionic_eth_hw_features { IONIC_ETH_HW_TSO_ECN = BIT(10), IONIC_ETH_HW_TSO_GRE = BIT(11), IONIC_ETH_HW_TSO_GRE_CSUM = BIT(12), - IONIC_ETH_HW_TSO_IPXIP4 = BIT(13), - IONIC_ETH_HW_TSO_IPXIP6 = BIT(14), + IONIC_ETH_HW_TSO_IPXIP4 = BIT(13), + IONIC_ETH_HW_TSO_IPXIP6 = BIT(14), IONIC_ETH_HW_TSO_UDP = BIT(15), IONIC_ETH_HW_TSO_UDP_CSUM = BIT(16), }; @@ -923,7 +1031,10 @@ enum q_control_oper { }; /** - * Physical connection type + * enum ionic_phy_type - Physical connection type + * @IONIC_PHY_TYPE_NONE: No PHY installed + * @IONIC_PHY_TYPE_COPPER: Copper PHY + * @IONIC_PHY_TYPE_FIBER: Fiber PHY */ enum ionic_phy_type { IONIC_PHY_TYPE_NONE = 0, @@ -932,18 +1043,23 @@ enum ionic_phy_type { }; /** - * Transceiver status + * enum ionic_xcvr_state - Transceiver status + * @IONIC_XCVR_STATE_REMOVED: Transceiver removed + * @IONIC_XCVR_STATE_INSERTED: Transceiver inserted + * @IONIC_XCVR_STATE_PENDING: Transceiver pending + * @IONIC_XCVR_STATE_SPROM_READ: Transceiver data read + * @IONIC_XCVR_STATE_SPROM_READ_ERR: Transceiver data read error */ enum ionic_xcvr_state { IONIC_XCVR_STATE_REMOVED = 0, IONIC_XCVR_STATE_INSERTED = 1, IONIC_XCVR_STATE_PENDING = 2, IONIC_XCVR_STATE_SPROM_READ = 3, - IONIC_XCVR_STATE_SPROM_READ_ERR = 4, + IONIC_XCVR_STATE_SPROM_READ_ERR = 4, }; /** - * Supported link modes + * enum ionic_xcvr_pid - Supported link modes */ enum ionic_xcvr_pid { IONIC_XCVR_PID_UNKNOWN = 0, @@ -977,64 +1093,83 @@ enum ionic_xcvr_pid { IONIC_XCVR_PID_SFP_10GBASE_CU = 68, IONIC_XCVR_PID_QSFP_100G_CWDM4 = 69, IONIC_XCVR_PID_QSFP_100G_PSM4 = 70, + IONIC_XCVR_PID_SFP_25GBASE_ACC = 71, }; /** - * Port types + * enum ionic_port_type - Port types + * @IONIC_PORT_TYPE_NONE: Port type not configured + * @IONIC_PORT_TYPE_ETH: Port carries ethernet traffic (inband) + * @IONIC_PORT_TYPE_MGMT: Port carries mgmt traffic (out-of-band) */ enum ionic_port_type { - IONIC_PORT_TYPE_NONE = 0, /* port type not configured */ - IONIC_PORT_TYPE_ETH = 1, /* port carries ethernet traffic (inband) */ - IONIC_PORT_TYPE_MGMT = 2, /* port carries mgmt traffic (out-of-band) */ + IONIC_PORT_TYPE_NONE = 0, + IONIC_PORT_TYPE_ETH = 1, + IONIC_PORT_TYPE_MGMT = 2, }; /** - * Port config state + * enum ionic_port_admin_state - Port config state + * @IONIC_PORT_ADMIN_STATE_NONE: Port admin state not configured + * @IONIC_PORT_ADMIN_STATE_DOWN: Port admin disabled + * @IONIC_PORT_ADMIN_STATE_UP: Port admin enabled */ enum ionic_port_admin_state { - IONIC_PORT_ADMIN_STATE_NONE = 0, /* port admin state not configured */ - IONIC_PORT_ADMIN_STATE_DOWN = 1, /* port is admin disabled */ - IONIC_PORT_ADMIN_STATE_UP = 2, /* port is admin enabled */ + IONIC_PORT_ADMIN_STATE_NONE = 0, + IONIC_PORT_ADMIN_STATE_DOWN = 1, + IONIC_PORT_ADMIN_STATE_UP = 2, }; /** - * Port operational status + * enum ionic_port_oper_status - Port operational status + * @IONIC_PORT_OPER_STATUS_NONE: Port disabled + * @IONIC_PORT_OPER_STATUS_UP: Port link status up + * @IONIC_PORT_OPER_STATUS_DOWN: Port link status down */ enum ionic_port_oper_status { - IONIC_PORT_OPER_STATUS_NONE = 0, /* port is disabled */ - IONIC_PORT_OPER_STATUS_UP = 1, /* port is linked up */ - IONIC_PORT_OPER_STATUS_DOWN = 2, /* port link status is down */ + IONIC_PORT_OPER_STATUS_NONE = 0, + IONIC_PORT_OPER_STATUS_UP = 1, + IONIC_PORT_OPER_STATUS_DOWN = 2, }; /** - * Ethernet Forward error correction (fec) modes + * enum ionic_port_fec_type - Ethernet Forward error correction (FEC) modes + * @IONIC_PORT_FEC_TYPE_NONE: FEC Disabled + * @IONIC_PORT_FEC_TYPE_FC: FireCode FEC + * @IONIC_PORT_FEC_TYPE_RS: ReedSolomon FEC */ enum ionic_port_fec_type { - IONIC_PORT_FEC_TYPE_NONE = 0, /* Disabled */ - IONIC_PORT_FEC_TYPE_FC = 1, /* FireCode */ - IONIC_PORT_FEC_TYPE_RS = 2, /* ReedSolomon */ + IONIC_PORT_FEC_TYPE_NONE = 0, + IONIC_PORT_FEC_TYPE_FC = 1, + IONIC_PORT_FEC_TYPE_RS = 2, }; /** - * Ethernet pause (flow control) modes + * enum ionic_port_pause_type - Ethernet pause (flow control) modes + * @IONIC_PORT_PAUSE_TYPE_NONE: Disable Pause + * @IONIC_PORT_PAUSE_TYPE_LINK: Link level pause + * @IONIC_PORT_PAUSE_TYPE_PFC: Priority-Flow Control */ enum ionic_port_pause_type { - IONIC_PORT_PAUSE_TYPE_NONE = 0, /* Disable Pause */ - IONIC_PORT_PAUSE_TYPE_LINK = 1, /* Link level pause */ - IONIC_PORT_PAUSE_TYPE_PFC = 2, /* Priority-Flow control */ + IONIC_PORT_PAUSE_TYPE_NONE = 0, + IONIC_PORT_PAUSE_TYPE_LINK = 1, + IONIC_PORT_PAUSE_TYPE_PFC = 2, }; /** - * Loopback modes + * enum ionic_port_loopback_mode - Loopback modes + * @IONIC_PORT_LOOPBACK_MODE_NONE: Disable loopback + * @IONIC_PORT_LOOPBACK_MODE_MAC: MAC loopback + * @IONIC_PORT_LOOPBACK_MODE_PHY: PHY/SerDes loopback */ enum ionic_port_loopback_mode { - IONIC_PORT_LOOPBACK_MODE_NONE = 0, /* Disable loopback */ - IONIC_PORT_LOOPBACK_MODE_MAC = 1, /* MAC loopback */ - IONIC_PORT_LOOPBACK_MODE_PHY = 2, /* PHY/Serdes loopback */ + IONIC_PORT_LOOPBACK_MODE_NONE = 0, + IONIC_PORT_LOOPBACK_MODE_MAC = 1, + IONIC_PORT_LOOPBACK_MODE_PHY = 2, }; /** - * Transceiver Status information + * struct ionic_xcvr_status - Transceiver Status information * @state: Transceiver status (enum ionic_xcvr_state) * @phy: Physical connection type (enum ionic_phy_type) * @pid: Transceiver link mode (enum pid) @@ -1048,7 +1183,7 @@ struct ionic_xcvr_status { }; /** - * Port configuration + * union ionic_port_config - Port configuration * @speed: port speed (in Mbps) * @mtu: mtu * @state: port admin state (enum port_admin_state) @@ -1081,17 +1216,21 @@ union ionic_port_config { }; /** - * Port Status information + * struct ionic_port_status - Port Status information * @status: link status (enum ionic_port_oper_status) * @id: port id * @speed: link speed (in Mbps) + * @link_down_count: number of times link went from from up to down + * @fec_type: fec type (enum ionic_port_fec_type) * @xcvr: tranceiver status */ struct ionic_port_status { __le32 id; __le32 speed; u8 status; - u8 rsvd[51]; + __le16 link_down_count; + u8 fec_type; + u8 rsvd[48]; struct ionic_xcvr_status xcvr; } __packed; @@ -1110,7 +1249,7 @@ struct ionic_port_identify_cmd { /** * struct ionic_port_identify_comp - Port identify command completion - * @status: The status of the command (enum status_code) + * @status: Status of the command (enum ionic_status_code) * @ver: Version of identify returned by device */ struct ionic_port_identify_comp { @@ -1135,7 +1274,7 @@ struct ionic_port_init_cmd { /** * struct ionic_port_init_comp - Port initialization command completion - * @status: The status of the command (enum status_code) + * @status: Status of the command (enum ionic_status_code) */ struct ionic_port_init_comp { u8 status; @@ -1155,7 +1294,7 @@ struct ionic_port_reset_cmd { /** * struct ionic_port_reset_comp - Port reset command completion - * @status: The status of the command (enum status_code) + * @status: Status of the command (enum ionic_status_code) */ struct ionic_port_reset_comp { u8 status; @@ -1163,15 +1302,23 @@ struct ionic_port_reset_comp { }; /** - * enum stats_ctl_cmd - List of commands for stats control + * enum ionic_stats_ctl_cmd - List of commands for stats control + * @IONIC_STATS_CTL_RESET: Reset statistics */ enum ionic_stats_ctl_cmd { IONIC_STATS_CTL_RESET = 0, }; - /** * enum ionic_port_attr - List of device attributes + * @IONIC_PORT_ATTR_STATE: Port state attribute + * @IONIC_PORT_ATTR_SPEED: Port speed attribute + * @IONIC_PORT_ATTR_MTU: Port MTU attribute + * @IONIC_PORT_ATTR_AUTONEG: Port autonegotation attribute + * @IONIC_PORT_ATTR_FEC: Port FEC attribute + * @IONIC_PORT_ATTR_PAUSE: Port pause attribute + * @IONIC_PORT_ATTR_LOOPBACK: Port loopback attribute + * @IONIC_PORT_ATTR_STATS_CTRL: Port statistics control attribute */ enum ionic_port_attr { IONIC_PORT_ATTR_STATE = 0, @@ -1186,9 +1333,17 @@ enum ionic_port_attr { /** * struct ionic_port_setattr_cmd - Set port attributes on the NIC - * @opcode: Opcode - * @index: port index - * @attr: Attribute type (enum ionic_port_attr) + * @opcode: Opcode + * @index: Port index + * @attr: Attribute type (enum ionic_port_attr) + * @state: Port state + * @speed: Port speed + * @mtu: Port MTU + * @an_enable: Port autonegotiation setting + * @fec_type: Port FEC type setting + * @pause_type: Port pause type setting + * @loopback_mode: Port loopback mode + * @stats_ctl: Port stats setting */ struct ionic_port_setattr_cmd { u8 opcode; @@ -1203,14 +1358,14 @@ struct ionic_port_setattr_cmd { u8 fec_type; u8 pause_type; u8 loopback_mode; - u8 stats_ctl; + u8 stats_ctl; u8 rsvd2[60]; }; }; /** * struct ionic_port_setattr_comp - Port set attr command completion - * @status: The status of the command (enum status_code) + * @status: Status of the command (enum ionic_status_code) * @color: Color bit */ struct ionic_port_setattr_comp { @@ -1234,8 +1389,15 @@ struct ionic_port_getattr_cmd { /** * struct ionic_port_getattr_comp - Port get attr command completion - * @status: The status of the command (enum status_code) - * @color: Color bit + * @status: Status of the command (enum ionic_status_code) + * @state: Port state + * @speed: Port speed + * @mtu: Port MTU + * @an_enable: Port autonegotiation setting + * @fec_type: Port FEC type setting + * @pause_type: Port pause type setting + * @loopback_mode: Port loopback mode + * @color: Color bit */ struct ionic_port_getattr_comp { u8 status; @@ -1254,12 +1416,12 @@ struct ionic_port_getattr_comp { }; /** - * struct ionic_lif_status - Lif status register + * struct ionic_lif_status - LIF status register * @eid: most recent NotifyQ event id - * @port_num: port the lif is connected to + * @port_num: port the LIF is connected to * @link_status: port status (enum ionic_port_oper_status) * @link_speed: speed of link in Mbps - * @link_down_count: number of times link status changes + * @link_down_count: number of times link went from up to down */ struct ionic_lif_status { __le64 eid; @@ -1293,6 +1455,9 @@ enum ionic_dev_state { /** * enum ionic_dev_attr - List of device attributes + * @IONIC_DEV_ATTR_STATE: Device state attribute + * @IONIC_DEV_ATTR_NAME: Device name attribute + * @IONIC_DEV_ATTR_FEATURES: Device feature attributes */ enum ionic_dev_attr { IONIC_DEV_ATTR_STATE = 0, @@ -1322,7 +1487,7 @@ struct ionic_dev_setattr_cmd { /** * struct ionic_dev_setattr_comp - Device set attr command completion - * @status: The status of the command (enum status_code) + * @status: Status of the command (enum ionic_status_code) * @features: Device features * @color: Color bit */ @@ -1349,7 +1514,7 @@ struct ionic_dev_getattr_cmd { /** * struct ionic_dev_setattr_comp - Device set attr command completion - * @status: The status of the command (enum status_code) + * @status: Status of the command (enum ionic_status_code) * @features: Device features * @color: Color bit */ @@ -1379,6 +1544,13 @@ enum ionic_rss_hash_types { /** * enum ionic_lif_attr - List of LIF attributes + * @IONIC_LIF_ATTR_STATE: LIF state attribute + * @IONIC_LIF_ATTR_NAME: LIF name attribute + * @IONIC_LIF_ATTR_MTU: LIF MTU attribute + * @IONIC_LIF_ATTR_MAC: LIF MAC attribute + * @IONIC_LIF_ATTR_FEATURES: LIF features attribute + * @IONIC_LIF_ATTR_RSS: LIF RSS attribute + * @IONIC_LIF_ATTR_STATS_CTRL: LIF statistics control attribute */ enum ionic_lif_attr { IONIC_LIF_ATTR_STATE = 0, @@ -1393,18 +1565,18 @@ enum ionic_lif_attr { /** * struct ionic_lif_setattr_cmd - Set LIF attributes on the NIC * @opcode: Opcode - * @type: Attribute type (enum ionic_lif_attr) + * @attr: Attribute type (enum ionic_lif_attr) * @index: LIF index - * @state: lif state (enum lif_state) + * @state: LIF state (enum ionic_lif_state) * @name: The netdev name string, 0 terminated * @mtu: Mtu * @mac: Station mac * @features: Features (enum ionic_eth_hw_features) * @rss: RSS properties - * @types: The hash types to enable (see rss_hash_types). - * @key: The hash secret key. - * @addr: Address for the indirection table shared memory. - * @stats_ctl: stats control commands (enum stats_ctl_cmd) + * @types: The hash types to enable (see rss_hash_types) + * @key: The hash secret key + * @addr: Address for the indirection table shared memory + * @stats_ctl: stats control commands (enum ionic_stats_ctl_cmd) */ struct ionic_lif_setattr_cmd { u8 opcode; @@ -1422,16 +1594,15 @@ struct ionic_lif_setattr_cmd { u8 rsvd[6]; __le64 addr; } rss; - u8 stats_ctl; + u8 stats_ctl; u8 rsvd[60]; } __packed; }; /** * struct ionic_lif_setattr_comp - LIF set attr command completion - * @status: The status of the command (enum status_code) - * @comp_index: The index in the descriptor ring for which this - * is the completion. + * @status: Status of the command (enum ionic_status_code) + * @comp_index: Index in the descriptor ring for which this is the completion * @features: features (enum ionic_eth_hw_features) * @color: Color bit */ @@ -1461,10 +1632,9 @@ struct ionic_lif_getattr_cmd { /** * struct ionic_lif_getattr_comp - LIF get attr command completion - * @status: The status of the command (enum status_code) - * @comp_index: The index in the descriptor ring for which this - * is the completion. - * @state: lif state (enum lif_state) + * @status: Status of the command (enum ionic_status_code) + * @comp_index: Index in the descriptor ring for which this is the completion + * @state: LIF state (enum ionic_lif_state) * @name: The netdev name string, 0 terminated * @mtu: Mtu * @mac: Station mac @@ -1486,11 +1656,12 @@ struct ionic_lif_getattr_comp { }; enum ionic_rx_mode { - IONIC_RX_MODE_F_UNICAST = BIT(0), - IONIC_RX_MODE_F_MULTICAST = BIT(1), - IONIC_RX_MODE_F_BROADCAST = BIT(2), - IONIC_RX_MODE_F_PROMISC = BIT(3), - IONIC_RX_MODE_F_ALLMULTI = BIT(4), + IONIC_RX_MODE_F_UNICAST = BIT(0), + IONIC_RX_MODE_F_MULTICAST = BIT(1), + IONIC_RX_MODE_F_BROADCAST = BIT(2), + IONIC_RX_MODE_F_PROMISC = BIT(3), + IONIC_RX_MODE_F_ALLMULTI = BIT(4), + IONIC_RX_MODE_F_RDMA_SNIFFER = BIT(5), }; /** @@ -1498,11 +1669,12 @@ enum ionic_rx_mode { * @opcode: opcode * @lif_index: LIF index * @rx_mode: Rx mode flags: - * IONIC_RX_MODE_F_UNICAST: Accept known unicast packets. - * IONIC_RX_MODE_F_MULTICAST: Accept known multicast packets. - * IONIC_RX_MODE_F_BROADCAST: Accept broadcast packets. - * IONIC_RX_MODE_F_PROMISC: Accept any packets. - * IONIC_RX_MODE_F_ALLMULTI: Accept any multicast packets. + * IONIC_RX_MODE_F_UNICAST: Accept known unicast packets + * IONIC_RX_MODE_F_MULTICAST: Accept known multicast packets + * IONIC_RX_MODE_F_BROADCAST: Accept broadcast packets + * IONIC_RX_MODE_F_PROMISC: Accept any packets + * IONIC_RX_MODE_F_ALLMULTI: Accept any multicast packets + * IONIC_RX_MODE_F_RDMA_SNIFFER: Sniff RDMA packets */ struct ionic_rx_mode_set_cmd { u8 opcode; @@ -1526,9 +1698,14 @@ enum ionic_rx_filter_match_type { * @qtype: Queue type * @lif_index: LIF index * @qid: Queue ID - * @match: Rx filter match type. (See IONIC_RX_FILTER_MATCH_xxx) - * @vlan: VLAN ID - * @addr: MAC address (network-byte order) + * @match: Rx filter match type (see IONIC_RX_FILTER_MATCH_xxx) + * @vlan: VLAN filter + * @vlan: VLAN ID + * @mac: MAC filter + * @addr: MAC address (network-byte order) + * @mac_vlan: MACVLAN filter + * @vlan: VLAN ID + * @addr: MAC address (network-byte order) */ struct ionic_rx_filter_add_cmd { u8 opcode; @@ -1553,11 +1730,10 @@ struct ionic_rx_filter_add_cmd { /** * struct ionic_rx_filter_add_comp - Add LIF Rx filter command completion - * @status: The status of the command (enum status_code) - * @comp_index: The index in the descriptor ring for which this - * is the completion. + * @status: Status of the command (enum ionic_status_code) + * @comp_index: Index in the descriptor ring for which this is the completion * @filter_id: Filter ID - * @color: Color bit. + * @color: Color bit */ struct ionic_rx_filter_add_comp { u8 status; @@ -1584,63 +1760,6 @@ struct ionic_rx_filter_del_cmd { typedef struct ionic_admin_comp ionic_rx_filter_del_comp; -/** - * struct ionic_qos_identify_cmd - QoS identify command - * @opcode: opcode - * @ver: Highest version of identify supported by driver - * - */ -struct ionic_qos_identify_cmd { - u8 opcode; - u8 ver; - u8 rsvd[62]; -}; - -/** - * struct ionic_qos_identify_comp - QoS identify command completion - * @status: The status of the command (enum status_code) - * @ver: Version of identify returned by device - */ -struct ionic_qos_identify_comp { - u8 status; - u8 ver; - u8 rsvd[14]; -}; - -#define IONIC_QOS_CLASS_MAX 7 -#define IONIC_QOS_CLASS_NAME_SZ 32 -#define IONIC_QOS_DSCP_MAX_VALUES 64 - -/** - * enum ionic_qos_class - */ -enum ionic_qos_class { - IONIC_QOS_CLASS_DEFAULT = 0, - IONIC_QOS_CLASS_USER_DEFINED_1 = 1, - IONIC_QOS_CLASS_USER_DEFINED_2 = 2, - IONIC_QOS_CLASS_USER_DEFINED_3 = 3, - IONIC_QOS_CLASS_USER_DEFINED_4 = 4, - IONIC_QOS_CLASS_USER_DEFINED_5 = 5, - IONIC_QOS_CLASS_USER_DEFINED_6 = 6, -}; - -/** - * enum ionic_qos_class_type - Traffic classification criteria - */ -enum ionic_qos_class_type { - IONIC_QOS_CLASS_TYPE_NONE = 0, - IONIC_QOS_CLASS_TYPE_PCP = 1, /* Dot1Q pcp */ - IONIC_QOS_CLASS_TYPE_DSCP = 2, /* IP dscp */ -}; - -/** - * enum ionic_qos_sched_type - Qos class scheduling type - */ -enum ionic_qos_sched_type { - IONIC_QOS_SCHED_TYPE_STRICT = 0, /* Strict priority */ - IONIC_QOS_SCHED_TYPE_DWRR = 1, /* Deficit weighted round-robin */ -}; - enum ionic_vf_attr { IONIC_VF_ATTR_SPOOFCHK = 1, IONIC_VF_ATTR_TRUST = 2, @@ -1652,26 +1771,29 @@ enum ionic_vf_attr { }; /** - * VF link status + * enum ionic_vf_link_status - Virtual Function link status + * @IONIC_VF_LINK_STATUS_AUTO: Use link state of the uplink + * @IONIC_VF_LINK_STATUS_UP: Link always up + * @IONIC_VF_LINK_STATUS_DOWN: Link always down */ enum ionic_vf_link_status { - IONIC_VF_LINK_STATUS_AUTO = 0, /* link state of the uplink */ - IONIC_VF_LINK_STATUS_UP = 1, /* link is always up */ - IONIC_VF_LINK_STATUS_DOWN = 2, /* link is always down */ + IONIC_VF_LINK_STATUS_AUTO = 0, + IONIC_VF_LINK_STATUS_UP = 1, + IONIC_VF_LINK_STATUS_DOWN = 2, }; /** * struct ionic_vf_setattr_cmd - Set VF attributes on the NIC * @opcode: Opcode - * @index: VF index * @attr: Attribute type (enum ionic_vf_attr) - * macaddr mac address - * vlanid vlan ID - * maxrate max Tx rate in Mbps - * spoofchk enable address spoof checking - * trust enable VF trust - * linkstate set link up or down - * stats_pa set DMA address for VF stats + * @vf_index: VF index + * @macaddr: mac address + * @vlanid: vlan ID + * @maxrate: max Tx rate in Mbps + * @spoofchk: enable address spoof checking + * @trust: enable VF trust + * @linkstate: set link up or down + * @stats_pa: set DMA address for VF stats */ struct ionic_vf_setattr_cmd { u8 opcode; @@ -1701,8 +1823,8 @@ struct ionic_vf_setattr_comp { /** * struct ionic_vf_getattr_cmd - Get VF attributes from the NIC * @opcode: Opcode - * @index: VF index * @attr: Attribute type (enum ionic_vf_attr) + * @vf_index: VF index */ struct ionic_vf_getattr_cmd { u8 opcode; @@ -1729,19 +1851,85 @@ struct ionic_vf_getattr_comp { }; /** - * union ionic_qos_config - Qos configuration structure + * struct ionic_qos_identify_cmd - QoS identify command + * @opcode: opcode + * @ver: Highest version of identify supported by driver + * + */ +struct ionic_qos_identify_cmd { + u8 opcode; + u8 ver; + u8 rsvd[62]; +}; + +/** + * struct ionic_qos_identify_comp - QoS identify command completion + * @status: Status of the command (enum ionic_status_code) + * @ver: Version of identify returned by device + */ +struct ionic_qos_identify_comp { + u8 status; + u8 ver; + u8 rsvd[14]; +}; + +#define IONIC_QOS_TC_MAX 8 +/* Capri max supported, should be renamed. */ +#define IONIC_QOS_CLASS_MAX 7 +#define IONIC_QOS_PCP_MAX 8 +#define IONIC_QOS_CLASS_NAME_SZ 32 +#define IONIC_QOS_DSCP_MAX 64 +#define IONIC_QOS_ALL_PCP 0xFF + +/** + * enum ionic_qos_class + */ +enum ionic_qos_class { + IONIC_QOS_CLASS_DEFAULT = 0, + IONIC_QOS_CLASS_USER_DEFINED_1 = 1, + IONIC_QOS_CLASS_USER_DEFINED_2 = 2, + IONIC_QOS_CLASS_USER_DEFINED_3 = 3, + IONIC_QOS_CLASS_USER_DEFINED_4 = 4, + IONIC_QOS_CLASS_USER_DEFINED_5 = 5, + IONIC_QOS_CLASS_USER_DEFINED_6 = 6, +}; + +/** + * enum ionic_qos_class_type - Traffic classification criteria + * @IONIC_QOS_CLASS_TYPE_NONE: No QoS + * @IONIC_QOS_CLASS_TYPE_PCP: Dot1Q PCP + * @IONIC_QOS_CLASS_TYPE_DSCP: IP DSCP + */ +enum ionic_qos_class_type { + IONIC_QOS_CLASS_TYPE_NONE = 0, + IONIC_QOS_CLASS_TYPE_PCP = 1, + IONIC_QOS_CLASS_TYPE_DSCP = 2, +}; + +/** + * enum ionic_qos_sched_type - QoS class scheduling type + * @IONIC_QOS_SCHED_TYPE_STRICT: Strict priority + * @IONIC_QOS_SCHED_TYPE_DWRR: Deficit weighted round-robin + */ +enum ionic_qos_sched_type { + IONIC_QOS_SCHED_TYPE_STRICT = 0, + IONIC_QOS_SCHED_TYPE_DWRR = 1, +}; + +/** + * union ionic_qos_config - QoS configuration structure * @flags: Configuration flags * IONIC_QOS_CONFIG_F_ENABLE enable - * IONIC_QOS_CONFIG_F_DROP drop/nodrop + * IONIC_QOS_CONFIG_F_NO_DROP drop/nodrop * IONIC_QOS_CONFIG_F_RW_DOT1Q_PCP enable dot1q pcp rewrite * IONIC_QOS_CONFIG_F_RW_IP_DSCP enable ip dscp rewrite - * @sched_type: Qos class scheduling type (enum ionic_qos_sched_type) - * @class_type: Qos class type (enum ionic_qos_class_type) - * @pause_type: Qos pause type (enum ionic_qos_pause_type) - * @name: Qos class name + * @sched_type: QoS class scheduling type (enum ionic_qos_sched_type) + * @class_type: QoS class type (enum ionic_qos_class_type) + * @pause_type: QoS pause type (enum ionic_qos_pause_type) + * @name: QoS class name * @mtu: MTU of the class - * @pfc_dot1q_pcp: Pcp value for pause frames (valid iff F_NODROP) - * @dwrr_weight: Qos class scheduling weight + * @pfc_cos: Priority-Flow Control class of service + * @dwrr_weight: QoS class scheduling weight * @strict_rlmt: Rate limit for strict priority scheduling * @rw_dot1q_pcp: Rewrite dot1q pcp to this value (valid iff F_RW_DOT1Q_PCP) * @rw_ip_dscp: Rewrite ip dscp to this value (valid iff F_RW_IP_DSCP) @@ -1752,7 +1940,8 @@ struct ionic_vf_getattr_comp { union ionic_qos_config { struct { #define IONIC_QOS_CONFIG_F_ENABLE BIT(0) -#define IONIC_QOS_CONFIG_F_DROP BIT(1) +#define IONIC_QOS_CONFIG_F_NO_DROP BIT(1) +/* Used to rewrite PCP or DSCP value. */ #define IONIC_QOS_CONFIG_F_RW_DOT1Q_PCP BIT(2) #define IONIC_QOS_CONFIG_F_RW_IP_DSCP BIT(3) u8 flags; @@ -1769,6 +1958,7 @@ union ionic_qos_config { __le64 strict_rlmt; }; /* marking */ + /* Used to rewrite PCP or DSCP value. */ union { u8 rw_dot1q_pcp; u8 rw_ip_dscp; @@ -1778,7 +1968,7 @@ union ionic_qos_config { u8 dot1q_pcp; struct { u8 ndscp; - u8 ip_dscp[IONIC_QOS_DSCP_MAX_VALUES]; + u8 ip_dscp[IONIC_QOS_DSCP_MAX]; }; }; }; @@ -1797,15 +1987,15 @@ union ionic_qos_identity { u8 version; u8 type; u8 rsvd[62]; - union ionic_qos_config config[IONIC_QOS_CLASS_MAX]; + union ionic_qos_config config[IONIC_QOS_CLASS_MAX]; }; - __le32 words[512]; + __le32 words[478]; }; /** - * struct qos_init_cmd - QoS config init command + * struct ionic_qos_init_cmd - QoS config init command * @opcode: Opcode - * @group: Qos class id + * @group: QoS class id * @info_pa: destination address for qos info */ struct ionic_qos_init_cmd { @@ -1819,8 +2009,9 @@ struct ionic_qos_init_cmd { typedef struct ionic_admin_comp ionic_qos_init_comp; /** - * struct ionic_qos_reset_cmd - Qos config reset command + * struct ionic_qos_reset_cmd - QoS config reset command * @opcode: Opcode + * @group: QoS class id */ struct ionic_qos_reset_cmd { u8 opcode; @@ -1847,10 +2038,16 @@ struct ionic_fw_download_cmd { typedef struct ionic_admin_comp ionic_fw_download_comp; +/** + * enum ionic_fw_control_oper - FW control operations + * @IONIC_FW_RESET: Reset firmware + * @IONIC_FW_INSTALL: Install firmware + * @IONIC_FW_ACTIVATE: Activate firmware + */ enum ionic_fw_control_oper { - IONIC_FW_RESET = 0, /* Reset firmware */ - IONIC_FW_INSTALL = 1, /* Install firmware */ - IONIC_FW_ACTIVATE = 2, /* Activate firmware */ + IONIC_FW_RESET = 0, + IONIC_FW_INSTALL = 1, + IONIC_FW_ACTIVATE = 2, }; /** @@ -1869,8 +2066,10 @@ struct ionic_fw_control_cmd { /** * struct ionic_fw_control_comp - Firmware control copletion - * @opcode: opcode - * @slot: slot where the firmware was installed + * @status: Status of the command (enum ionic_status_code) + * @comp_index: Index in the descriptor ring for which this is the completion + * @slot: Slot where the firmware was installed + * @color: Color bit */ struct ionic_fw_control_comp { u8 status; @@ -1888,11 +2087,11 @@ struct ionic_fw_control_comp { /** * struct ionic_rdma_reset_cmd - Reset RDMA LIF cmd * @opcode: opcode - * @lif_index: lif index + * @lif_index: LIF index * - * There is no rdma specific dev command completion struct. Completion uses + * There is no RDMA specific dev command completion struct. Completion uses * the common struct ionic_admin_comp. Only the status is indicated. - * Nonzero status means the LIF does not support rdma. + * Nonzero status means the LIF does not support RDMA. **/ struct ionic_rdma_reset_cmd { u8 opcode; @@ -1904,30 +2103,29 @@ struct ionic_rdma_reset_cmd { /** * struct ionic_rdma_queue_cmd - Create RDMA Queue command * @opcode: opcode, 52, 53 - * @lif_index lif index - * @qid_ver: (qid | (rdma version << 24)) + * @lif_index: LIF index + * @qid_ver: (qid | (RDMA version << 24)) * @cid: intr, eq_id, or cq_id * @dbid: doorbell page id * @depth_log2: log base two of queue depth * @stride_log2: log base two of queue stride * @dma_addr: address of the queue memory - * @xxx_table_index: temporary, but should not need pgtbl for contig. queues. * - * The same command struct is used to create an rdma event queue, completion - * queue, or rdma admin queue. The cid is an interrupt number for an event + * The same command struct is used to create an RDMA event queue, completion + * queue, or RDMA admin queue. The cid is an interrupt number for an event * queue, an event queue id for a completion queue, or a completion queue id - * for an rdma admin queue. + * for an RDMA admin queue. * * The queue created via a dev command must be contiguous in dma space. * * The dev commands are intended only to be used during driver initialization, - * to create queues supporting the rdma admin queue. Other queues, and other - * types of rdma resources like memory regions, will be created and registered - * via the rdma admin queue, and will support a more complete interface + * to create queues supporting the RDMA admin queue. Other queues, and other + * types of RDMA resources like memory regions, will be created and registered + * via the RDMA admin queue, and will support a more complete interface * providing scatter gather lists for larger, scattered queue buffers and * memory registration. * - * There is no rdma specific dev command completion struct. Completion uses + * There is no RDMA specific dev command completion struct. Completion uses * the common struct ionic_admin_comp. Only the status is indicated. **/ struct ionic_rdma_queue_cmd { @@ -1940,8 +2138,7 @@ struct ionic_rdma_queue_cmd { u8 depth_log2; u8 stride_log2; __le64 dma_addr; - u8 rsvd2[36]; - __le32 xxx_table_index; + u8 rsvd2[40]; }; /****************************************************************** @@ -1949,7 +2146,7 @@ struct ionic_rdma_queue_cmd { ******************************************************************/ /** - * struct ionic_notifyq_event + * struct ionic_notifyq_event - Generic event reporting structure * @eid: event number * @ecode: event code * @data: unspecified data about the event @@ -1964,9 +2161,9 @@ struct ionic_notifyq_event { }; /** - * struct ionic_link_change_event + * struct ionic_link_change_event - Link change event notification * @eid: event number - * @ecode: event code = EVENT_OPCODE_LINK_CHANGE + * @ecode: event code = IONIC_EVENT_LINK_CHANGE * @link_status: link up or down, with error bits (enum port_status) * @link_speed: speed of the network link * @@ -1981,9 +2178,9 @@ struct ionic_link_change_event { }; /** - * struct ionic_reset_event + * struct ionic_reset_event - Reset event notification * @eid: event number - * @ecode: event code = EVENT_OPCODE_RESET + * @ecode: event code = IONIC_EVENT_RESET * @reset_code: reset type * @state: 0=pending, 1=complete, 2=error * @@ -1999,11 +2196,9 @@ struct ionic_reset_event { }; /** - * struct ionic_heartbeat_event + * struct ionic_heartbeat_event - Sent periodically by NIC to indicate health * @eid: event number - * @ecode: event code = EVENT_OPCODE_HEARTBEAT - * - * Sent periodically by the NIC to indicate continued health + * @ecode: event code = IONIC_EVENT_HEARTBEAT */ struct ionic_heartbeat_event { __le64 eid; @@ -2012,12 +2207,10 @@ struct ionic_heartbeat_event { }; /** - * struct ionic_log_event + * struct ionic_log_event - Sent to notify the driver of an internal error * @eid: event number - * @ecode: event code = EVENT_OPCODE_LOG + * @ecode: event code = IONIC_EVENT_LOG * @data: log data - * - * Sent to notify the driver of an internal error. */ struct ionic_log_event { __le64 eid; @@ -2026,7 +2219,18 @@ struct ionic_log_event { }; /** - * struct ionic_port_stats + * struct ionic_xcvr_event - Transceiver change event + * @eid: event number + * @ecode: event code = IONIC_EVENT_XCVR + */ +struct ionic_xcvr_event { + __le64 eid; + __le16 ecode; + u8 rsvd[54]; +}; + +/** + * struct ionic_port_stats - Port statistics structure */ struct ionic_port_stats { __le64 frames_rx_ok; @@ -2131,28 +2335,61 @@ struct ionic_mgmt_port_stats { __le64 frames_rx_multicast; __le64 frames_rx_broadcast; __le64 frames_rx_pause; - __le64 frames_rx_bad_length0; - __le64 frames_rx_undersized1; - __le64 frames_rx_oversized2; - __le64 frames_rx_fragments3; - __le64 frames_rx_jabber4; - __le64 frames_rx_64b5; - __le64 frames_rx_65b_127b6; - __le64 frames_rx_128b_255b7; - __le64 frames_rx_256b_511b8; - __le64 frames_rx_512b_1023b9; - __le64 frames_rx_1024b_1518b0; - __le64 frames_rx_gt_1518b1; - __le64 frames_rx_fifo_full2; - __le64 frames_tx_ok3; - __le64 frames_tx_all4; - __le64 frames_tx_bad5; - __le64 octets_tx_ok6; - __le64 octets_tx_total7; - __le64 frames_tx_unicast8; - __le64 frames_tx_multicast9; - __le64 frames_tx_broadcast0; - __le64 frames_tx_pause1; + __le64 frames_rx_bad_length; + __le64 frames_rx_undersized; + __le64 frames_rx_oversized; + __le64 frames_rx_fragments; + __le64 frames_rx_jabber; + __le64 frames_rx_64b; + __le64 frames_rx_65b_127b; + __le64 frames_rx_128b_255b; + __le64 frames_rx_256b_511b; + __le64 frames_rx_512b_1023b; + __le64 frames_rx_1024b_1518b; + __le64 frames_rx_gt_1518b; + __le64 frames_rx_fifo_full; + __le64 frames_tx_ok; + __le64 frames_tx_all; + __le64 frames_tx_bad; + __le64 octets_tx_ok; + __le64 octets_tx_total; + __le64 frames_tx_unicast; + __le64 frames_tx_multicast; + __le64 frames_tx_broadcast; + __le64 frames_tx_pause; +}; + +enum ionic_pb_buffer_drop_stats { + IONIC_BUFFER_INTRINSIC_DROP = 0, + IONIC_BUFFER_DISCARDED, + IONIC_BUFFER_ADMITTED, + IONIC_BUFFER_OUT_OF_CELLS_DROP, + IONIC_BUFFER_OUT_OF_CELLS_DROP_2, + IONIC_BUFFER_OUT_OF_CREDIT_DROP, + IONIC_BUFFER_TRUNCATION_DROP, + IONIC_BUFFER_PORT_DISABLED_DROP, + IONIC_BUFFER_COPY_TO_CPU_TAIL_DROP, + IONIC_BUFFER_SPAN_TAIL_DROP, + IONIC_BUFFER_MIN_SIZE_VIOLATION_DROP, + IONIC_BUFFER_ENQUEUE_ERROR_DROP, + IONIC_BUFFER_INVALID_PORT_DROP, + IONIC_BUFFER_INVALID_OUTPUT_QUEUE_DROP, + IONIC_BUFFER_DROP_MAX, +}; + +/** + * struct port_pb_stats - packet buffers system stats + * uses ionic_pb_buffer_drop_stats for drop_counts[] + */ +struct ionic_port_pb_stats { + __le64 sop_count_in; + __le64 eop_count_in; + __le64 sop_count_out; + __le64 eop_count_out; + __le64 drop_counts[IONIC_BUFFER_DROP_MAX]; + __le64 input_queue_buffer_occupancy[IONIC_QOS_TC_MAX]; + __le64 input_queue_port_monitor[IONIC_QOS_TC_MAX]; + __le64 output_queue_port_monitor[IONIC_QOS_TC_MAX]; }; /** @@ -2184,22 +2421,31 @@ union ionic_port_identity { u8 rsvd2[44]; union ionic_port_config config; }; - __le32 words[512]; + __le32 words[478]; }; /** * struct ionic_port_info - port info structure - * @port_status: port status - * @port_stats: port stats + * @config: Port configuration data + * @status: Port status data + * @stats: Port statistics data + * @mgmt_stats: Port management statistics data + * @port_pb_drop_stats: uplink pb drop stats */ struct ionic_port_info { union ionic_port_config config; struct ionic_port_status status; - struct ionic_port_stats stats; + union { + struct ionic_port_stats stats; + struct ionic_mgmt_port_stats mgmt_stats; + }; + /* room for pb_stats to start at 2k offset */ + u8 rsvd[760]; + struct ionic_port_pb_stats pb_stats; }; /** - * struct ionic_lif_stats + * struct ionic_lif_stats - LIF statistics structure */ struct ionic_lif_stats { /* RX */ @@ -2252,7 +2498,7 @@ struct ionic_lif_stats { __le64 tx_queue_error; __le64 tx_desc_fetch_error; __le64 tx_desc_data_error; - __le64 rsvd9; + __le64 tx_queue_empty; __le64 rsvd10; __le64 rsvd11; __le64 rsvd12; @@ -2353,7 +2599,10 @@ struct ionic_lif_stats { }; /** - * struct ionic_lif_info - lif info structure + * struct ionic_lif_info - LIF info structure + * @config: LIF configuration structure + * @status: LIF status structure + * @stats: LIF statistics structure */ struct ionic_lif_info { union ionic_lif_config config; @@ -2389,7 +2638,9 @@ union ionic_dev_cmd { struct ionic_qos_init_cmd qos_init; struct ionic_qos_reset_cmd qos_reset; + struct ionic_q_identify_cmd q_identify; struct ionic_q_init_cmd q_init; + struct ionic_q_control_cmd q_control; }; union ionic_dev_cmd_comp { @@ -2421,19 +2672,20 @@ union ionic_dev_cmd_comp { ionic_qos_init_comp qos_init; ionic_qos_reset_comp qos_reset; + struct ionic_q_identify_comp q_identify; struct ionic_q_init_comp q_init; }; /** - * union dev_info - Device info register format (read-only) - * @signature: Signature value of 0x44455649 ('DEVI'). - * @version: Current version of info. - * @asic_type: Asic type. - * @asic_rev: Asic revision. - * @fw_status: Firmware status. - * @fw_heartbeat: Firmware heartbeat counter. - * @serial_num: Serial number. - * @fw_version: Firmware version. + * union ionic_dev_info_regs - Device info register format (read-only) + * @signature: Signature value of 0x44455649 ('DEVI') + * @version: Current version of info + * @asic_type: Asic type + * @asic_rev: Asic revision + * @fw_status: Firmware status + * @fw_heartbeat: Firmware heartbeat counter + * @serial_num: Serial number + * @fw_version: Firmware version */ union ionic_dev_info_regs { #define IONIC_DEVINFO_FWVERS_BUFLEN 32 @@ -2454,10 +2706,10 @@ union ionic_dev_info_regs { /** * union ionic_dev_cmd_regs - Device command register format (read-write) - * @doorbell: Device Cmd Doorbell, write-only. + * @doorbell: Device Cmd Doorbell, write-only * Write a 1 to signal device to process cmd, * poll done for completion. - * @done: Done indicator, bit 0 == 1 when command is complete. + * @done: Done indicator, bit 0 == 1 when command is complete * @cmd: Opcode-specific command bytes * @comp: Opcode-specific response bytes * @data: Opcode-specific side-data @@ -2475,7 +2727,7 @@ union ionic_dev_cmd_regs { }; /** - * union ionic_dev_regs - Device register format in for bar 0 page 0 + * union ionic_dev_regs - Device register format for bar 0 page 0 * @info: Device info registers * @devcmd: Device command registers */ @@ -2490,6 +2742,7 @@ union ionic_dev_regs { union ionic_adminq_cmd { struct ionic_admin_cmd cmd; struct ionic_nop_cmd nop; + struct ionic_q_identify_cmd q_identify; struct ionic_q_init_cmd q_init; struct ionic_q_control_cmd q_control; struct ionic_lif_setattr_cmd lif_setattr; @@ -2506,6 +2759,7 @@ union ionic_adminq_cmd { union ionic_adminq_comp { struct ionic_admin_comp comp; struct ionic_nop_comp nop; + struct ionic_q_identify_comp q_identify; struct ionic_q_init_comp q_init; struct ionic_lif_setattr_comp lif_setattr; struct ionic_lif_getattr_comp lif_getattr; @@ -2531,14 +2785,14 @@ union ionic_adminq_comp { /** * struct ionic_doorbell - Doorbell register layout * @p_index: Producer index - * @ring: Selects the specific ring of the queue to update. + * @ring: Selects the specific ring of the queue to update * Type-specific meaning: - * ring=0: Default producer/consumer queue. + * ring=0: Default producer/consumer queue * ring=1: (CQ, EQ) Re-Arm queue. RDMA CQs * send events to EQs when armed. EQs send * interrupts when armed. - * @qid: The queue id selects the queue destination for the - * producer index and flags. + * @qid_lo: Queue destination for the producer index and flags (low bits) + * @qid_hi: Queue destination for the producer index and flags (high bits) */ struct ionic_doorbell { __le16 p_index; @@ -2571,6 +2825,7 @@ struct ionic_identity { union ionic_lif_identity lif; union ionic_port_identity port; union ionic_qos_identity qos; + union ionic_q_identity txq; }; #endif /* _IONIC_IF_H_ */ diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c index f8a9c1bcffc9..7321a92f8395 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c @@ -17,6 +17,16 @@ #include "ionic_ethtool.h" #include "ionic_debugfs.h" +/* queuetype support level */ +static const u8 ionic_qtype_versions[IONIC_QTYPE_MAX] = { + [IONIC_QTYPE_ADMINQ] = 0, /* 0 = Base version with CQ support */ + [IONIC_QTYPE_NOTIFYQ] = 0, /* 0 = Base version */ + [IONIC_QTYPE_RXQ] = 0, /* 0 = Base version with CQ+SG support */ + [IONIC_QTYPE_TXQ] = 1, /* 0 = Base version with CQ+SG support + * 1 = ... with Tx SG version 1 + */ +}; + static void ionic_lif_rx_mode(struct ionic_lif *lif, unsigned int rx_mode); static int ionic_lif_addr_add(struct ionic_lif *lif, const u8 *addr); static int ionic_lif_addr_del(struct ionic_lif *lif, const u8 *addr); @@ -27,6 +37,7 @@ static void ionic_lif_set_netdev_info(struct ionic_lif *lif); static int ionic_start_queues(struct ionic_lif *lif); static void ionic_stop_queues(struct ionic_lif *lif); +static void ionic_lif_queue_identify(struct ionic_lif *lif); static void ionic_lif_deferred_work(struct work_struct *work) { @@ -186,10 +197,10 @@ static int ionic_intr_alloc(struct ionic_lif *lif, struct ionic_intr_info *intr) return 0; } -static void ionic_intr_free(struct ionic_lif *lif, int index) +static void ionic_intr_free(struct ionic *ionic, int index) { - if (index != INTR_INDEX_NOT_ASSIGNED && index < lif->ionic->nintrs) - clear_bit(index, lif->ionic->intrs); + if (index != IONIC_INTR_INDEX_NOT_ASSIGNED && index < ionic->nintrs) + clear_bit(index, ionic->intrs); } static int ionic_qcq_enable(struct ionic_qcq *qcq) @@ -299,7 +310,7 @@ static void ionic_qcq_free(struct ionic_lif *lif, struct ionic_qcq *qcq) irq_set_affinity_hint(qcq->intr.vector, NULL); devm_free_irq(dev, qcq->intr.vector, &qcq->napi); qcq->intr.vector = 0; - ionic_intr_free(lif, qcq->intr.index); + ionic_intr_free(lif->ionic, qcq->intr.index); } devm_kfree(dev, qcq->cq.info); @@ -345,7 +356,7 @@ static void ionic_link_qcq_interrupts(struct ionic_qcq *src_qcq, struct ionic_qcq *n_qcq) { if (WARN_ON(n_qcq->flags & IONIC_QCQ_F_INTR)) { - ionic_intr_free(n_qcq->cq.lif, n_qcq->intr.index); + ionic_intr_free(n_qcq->cq.lif->ionic, n_qcq->intr.index); n_qcq->flags &= ~IONIC_QCQ_F_INTR; } @@ -444,7 +455,7 @@ static int ionic_qcq_alloc(struct ionic_lif *lif, unsigned int type, cpumask_set_cpu(new->intr.cpu, &new->intr.affinity_mask); } else { - new->intr.index = INTR_INDEX_NOT_ASSIGNED; + new->intr.index = IONIC_INTR_INDEX_NOT_ASSIGNED; } new->cq.info = devm_kzalloc(dev, sizeof(*new->cq.info) * num_descs, @@ -497,7 +508,7 @@ err_out_free_irq: devm_free_irq(dev, new->intr.vector, &new->napi); err_out_free_intr: if (flags & IONIC_QCQ_F_INTR) - ionic_intr_free(lif, new->intr.index); + ionic_intr_free(lif->ionic, new->intr.index); err_out: dev_err(dev, "qcq alloc of %s%d failed %d\n", name, index, err); return err; @@ -597,6 +608,7 @@ static int ionic_lif_txq_init(struct ionic_lif *lif, struct ionic_qcq *qcq) .opcode = IONIC_CMD_Q_INIT, .lif_index = cpu_to_le16(lif->index), .type = q->type, + .ver = lif->qtype_info[q->type].version, .index = cpu_to_le32(q->index), .flags = cpu_to_le16(IONIC_QINIT_F_IRQ | IONIC_QINIT_F_SG), @@ -614,6 +626,8 @@ static int ionic_lif_txq_init(struct ionic_lif *lif, struct ionic_qcq *qcq) dev_dbg(dev, "txq_init.index %d\n", ctx.cmd.q_init.index); dev_dbg(dev, "txq_init.ring_base 0x%llx\n", ctx.cmd.q_init.ring_base); dev_dbg(dev, "txq_init.ring_size %d\n", ctx.cmd.q_init.ring_size); + dev_dbg(dev, "txq_init.flags 0x%x\n", ctx.cmd.q_init.flags); + dev_dbg(dev, "txq_init.ver %d\n", ctx.cmd.q_init.ver); q->tail = q->info; q->head = q->tail; @@ -646,6 +660,7 @@ static int ionic_lif_rxq_init(struct ionic_lif *lif, struct ionic_qcq *qcq) .opcode = IONIC_CMD_Q_INIT, .lif_index = cpu_to_le16(lif->index), .type = q->type, + .ver = lif->qtype_info[q->type].version, .index = cpu_to_le32(q->index), .flags = cpu_to_le16(IONIC_QINIT_F_IRQ | IONIC_QINIT_F_SG), @@ -663,6 +678,8 @@ static int ionic_lif_rxq_init(struct ionic_lif *lif, struct ionic_qcq *qcq) dev_dbg(dev, "rxq_init.index %d\n", ctx.cmd.q_init.index); dev_dbg(dev, "rxq_init.ring_base 0x%llx\n", ctx.cmd.q_init.ring_base); dev_dbg(dev, "rxq_init.ring_size %d\n", ctx.cmd.q_init.ring_size); + dev_dbg(dev, "rxq_init.flags 0x%x\n", ctx.cmd.q_init.flags); + dev_dbg(dev, "rxq_init.ver %d\n", ctx.cmd.q_init.ver); q->tail = q->info; q->head = q->tail; @@ -726,7 +743,7 @@ static bool ionic_notifyq_service(struct ionic_cq *cq, } break; default: - netdev_warn(netdev, "Notifyq unknown event ecode=%d eid=%lld\n", + netdev_warn(netdev, "Notifyq event ecode=%d eid=%lld\n", comp->event.ecode, eid); break; } @@ -775,8 +792,8 @@ static int ionic_adminq_napi(struct napi_struct *napi, int budget) return max(n_work, a_work); } -static void ionic_get_stats64(struct net_device *netdev, - struct rtnl_link_stats64 *ns) +void ionic_get_stats64(struct net_device *netdev, + struct rtnl_link_stats64 *ns) { struct ionic_lif *lif = netdev_priv(netdev); struct ionic_lif_stats *ls; @@ -1509,17 +1526,25 @@ static void ionic_txrx_free(struct ionic_lif *lif) static int ionic_txrx_alloc(struct ionic_lif *lif) { + unsigned int sg_desc_sz; unsigned int flags; unsigned int i; int err = 0; + if (lif->qtype_info[IONIC_QTYPE_TXQ].version >= 1 && + lif->qtype_info[IONIC_QTYPE_TXQ].sg_desc_sz == + sizeof(struct ionic_txq_sg_desc_v1)) + sg_desc_sz = sizeof(struct ionic_txq_sg_desc_v1); + else + sg_desc_sz = sizeof(struct ionic_txq_sg_desc); + flags = IONIC_QCQ_F_TX_STATS | IONIC_QCQ_F_SG; for (i = 0; i < lif->nxqs; i++) { err = ionic_qcq_alloc(lif, IONIC_QTYPE_TXQ, i, "tx", flags, lif->ntxq_descs, sizeof(struct ionic_txq_desc), sizeof(struct ionic_txq_comp), - sizeof(struct ionic_txq_sg_desc), + sg_desc_sz, lif->kern_pid, &lif->txqcqs[i].qcq); if (err) goto err_out; @@ -1682,7 +1707,7 @@ int ionic_stop(struct net_device *netdev) { struct ionic_lif *lif = netdev_priv(netdev); - if (test_bit(IONIC_LIF_F_FW_RESET, lif->state)) + if (!netif_device_present(netdev)) return 0; ionic_stop_queues(lif); @@ -1699,6 +1724,9 @@ static int ionic_get_vf_config(struct net_device *netdev, struct ionic *ionic = lif->ionic; int ret = 0; + if (!netif_device_present(netdev)) + return -EBUSY; + down_read(&ionic->vf_op_lock); if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) { @@ -1726,6 +1754,9 @@ static int ionic_get_vf_stats(struct net_device *netdev, int vf, struct ionic_lif_stats *vs; int ret = 0; + if (!netif_device_present(netdev)) + return -EBUSY; + down_read(&ionic->vf_op_lock); if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) { @@ -1761,6 +1792,9 @@ static int ionic_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) if (!(is_zero_ether_addr(mac) || is_valid_ether_addr(mac))) return -EINVAL; + if (!netif_device_present(netdev)) + return -EBUSY; + down_write(&ionic->vf_op_lock); if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) { @@ -1792,6 +1826,9 @@ static int ionic_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, if (proto != htons(ETH_P_8021Q)) return -EPROTONOSUPPORT; + if (!netif_device_present(netdev)) + return -EBUSY; + down_write(&ionic->vf_op_lock); if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) { @@ -1818,6 +1855,9 @@ static int ionic_set_vf_rate(struct net_device *netdev, int vf, if (tx_min) return -EINVAL; + if (!netif_device_present(netdev)) + return -EBUSY; + down_write(&ionic->vf_op_lock); if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) { @@ -1840,6 +1880,9 @@ static int ionic_set_vf_spoofchk(struct net_device *netdev, int vf, bool set) u8 data = set; /* convert to u8 for config */ int ret; + if (!netif_device_present(netdev)) + return -EBUSY; + down_write(&ionic->vf_op_lock); if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) { @@ -1862,6 +1905,9 @@ static int ionic_set_vf_trust(struct net_device *netdev, int vf, bool set) u8 data = set; /* convert to u8 for config */ int ret; + if (!netif_device_present(netdev)) + return -EBUSY; + down_write(&ionic->vf_op_lock); if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) { @@ -1898,6 +1944,9 @@ static int ionic_set_vf_link_state(struct net_device *netdev, int vf, int set) return -EINVAL; } + if (!netif_device_present(netdev)) + return -EBUSY; + down_write(&ionic->vf_op_lock); if (vf >= pci_num_vf(ionic->pdev) || !ionic->vfs) { @@ -2065,9 +2114,17 @@ int ionic_lifs_alloc(struct ionic *ionic) /* only build the first lif, others are for later features */ set_bit(0, ionic->lifbits); + lif = ionic_lif_alloc(ionic, 0); + if (IS_ERR_OR_NULL(lif)) { + clear_bit(0, ionic->lifbits); + return -ENOMEM; + } - return PTR_ERR_OR_ZERO(lif); + lif->lif_type = IONIC_LIF_TYPE_CLASSIC; + ionic_lif_queue_identify(lif); + + return 0; } static void ionic_lif_reset(struct ionic_lif *lif) @@ -2292,6 +2349,7 @@ static int ionic_lif_notifyq_init(struct ionic_lif *lif) .opcode = IONIC_CMD_Q_INIT, .lif_index = cpu_to_le16(lif->index), .type = q->type, + .ver = lif->qtype_info[q->type].version, .index = cpu_to_le32(q->index), .flags = cpu_to_le16(IONIC_QINIT_F_IRQ | IONIC_QINIT_F_ENA), @@ -2578,6 +2636,80 @@ void ionic_lifs_unregister(struct ionic *ionic) unregister_netdev(ionic->master_lif->netdev); } +static void ionic_lif_queue_identify(struct ionic_lif *lif) +{ + struct ionic *ionic = lif->ionic; + union ionic_q_identity *q_ident; + struct ionic_dev *idev; + int qtype; + int err; + + idev = &lif->ionic->idev; + q_ident = (union ionic_q_identity *)&idev->dev_cmd_regs->data; + + for (qtype = 0; qtype < ARRAY_SIZE(ionic_qtype_versions); qtype++) { + struct ionic_qtype_info *qti = &lif->qtype_info[qtype]; + + /* filter out the ones we know about */ + switch (qtype) { + case IONIC_QTYPE_ADMINQ: + case IONIC_QTYPE_NOTIFYQ: + case IONIC_QTYPE_RXQ: + case IONIC_QTYPE_TXQ: + break; + default: + continue; + } + + memset(qti, 0, sizeof(*qti)); + + mutex_lock(&ionic->dev_cmd_lock); + ionic_dev_cmd_queue_identify(idev, lif->lif_type, qtype, + ionic_qtype_versions[qtype]); + err = ionic_dev_cmd_wait(ionic, DEVCMD_TIMEOUT); + if (!err) { + qti->version = q_ident->version; + qti->supported = q_ident->supported; + qti->features = le64_to_cpu(q_ident->features); + qti->desc_sz = le16_to_cpu(q_ident->desc_sz); + qti->comp_sz = le16_to_cpu(q_ident->comp_sz); + qti->sg_desc_sz = le16_to_cpu(q_ident->sg_desc_sz); + qti->max_sg_elems = le16_to_cpu(q_ident->max_sg_elems); + qti->sg_desc_stride = le16_to_cpu(q_ident->sg_desc_stride); + } + mutex_unlock(&ionic->dev_cmd_lock); + + if (err == -EINVAL) { + dev_err(ionic->dev, "qtype %d not supported\n", qtype); + continue; + } else if (err == -EIO) { + dev_err(ionic->dev, "q_ident failed, not supported on older FW\n"); + return; + } else if (err) { + dev_err(ionic->dev, "q_ident failed, qtype %d: %d\n", + qtype, err); + return; + } + + dev_dbg(ionic->dev, " qtype[%d].version = %d\n", + qtype, qti->version); + dev_dbg(ionic->dev, " qtype[%d].supported = 0x%02x\n", + qtype, qti->supported); + dev_dbg(ionic->dev, " qtype[%d].features = 0x%04llx\n", + qtype, qti->features); + dev_dbg(ionic->dev, " qtype[%d].desc_sz = %d\n", + qtype, qti->desc_sz); + dev_dbg(ionic->dev, " qtype[%d].comp_sz = %d\n", + qtype, qti->comp_sz); + dev_dbg(ionic->dev, " qtype[%d].sg_desc_sz = %d\n", + qtype, qti->sg_desc_sz); + dev_dbg(ionic->dev, " qtype[%d].max_sg_elems = %d\n", + qtype, qti->max_sg_elems); + dev_dbg(ionic->dev, " qtype[%d].sg_desc_stride = %d\n", + qtype, qti->sg_desc_stride); + } +} + int ionic_lif_identify(struct ionic *ionic, u8 lif_type, union ionic_lif_identity *lid) { diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.h b/drivers/net/ethernet/pensando/ionic/ionic_lif.h index 5d4ffda5c05f..c3428034a17b 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.h @@ -20,11 +20,13 @@ struct ionic_tx_stats { u64 bytes; u64 clean; u64 linearize; - u64 no_csum; + u64 csum_none; u64 csum; u64 crc32_csum; u64 tso; + u64 tso_bytes; u64 frags; + u64 vlan_inserted; u64 sg_cntr[IONIC_MAX_NUM_SG_CNTR]; }; @@ -38,6 +40,7 @@ struct ionic_rx_stats { u64 csum_error; u64 buffers_posted; u64 dropped; + u64 vlan_stripped; }; #define IONIC_QCQ_F_INITED BIT(0) @@ -114,11 +117,17 @@ struct ionic_lif_sw_stats { u64 rx_packets; u64 rx_bytes; u64 tx_tso; - u64 tx_no_csum; + u64 tx_tso_bytes; + u64 tx_csum_none; u64 tx_csum; u64 rx_csum_none; u64 rx_csum_complete; u64 rx_csum_error; + u64 hw_tx_dropped; + u64 hw_rx_dropped; + u64 hw_rx_over_errors; + u64 hw_rx_missed_errors; + u64 hw_tx_aborted_errors; }; enum ionic_lif_state_flags { @@ -133,6 +142,17 @@ enum ionic_lif_state_flags { IONIC_LIF_F_STATE_SIZE }; +struct ionic_qtype_info { + u8 version; + u8 supported; + u64 features; + u16 desc_sz; + u16 comp_sz; + u16 sg_desc_sz; + u16 max_sg_elems; + u16 sg_desc_stride; +}; + #define IONIC_LIF_NAME_MAX_SZ 32 struct ionic_lif { char name[IONIC_LIF_NAME_MAX_SZ]; @@ -161,11 +181,13 @@ struct ionic_lif { bool mc_overflow; unsigned int nmcast; bool uc_overflow; + u16 lif_type; unsigned int nucast; struct ionic_lif_info *info; dma_addr_t info_pa; u32 info_sz; + struct ionic_qtype_info qtype_info[IONIC_QTYPE_MAX]; u16 rss_types; u8 rss_hash_key[IONIC_RSS_HASH_KEY_SIZE]; @@ -227,6 +249,8 @@ static inline u32 ionic_coal_hw_to_usec(struct ionic *ionic, u32 units) } void ionic_link_status_check_request(struct ionic_lif *lif); +void ionic_get_stats64(struct net_device *netdev, + struct rtnl_link_stats64 *ns); void ionic_lif_deferred_enqueue(struct ionic_deferred *def, struct ionic_deferred_work *work); int ionic_lifs_alloc(struct ionic *ionic); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_main.c b/drivers/net/ethernet/pensando/ionic/ionic_main.c index 3344bc1f7671..df5b9bcc3aba 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_main.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_main.c @@ -6,7 +6,7 @@ #include <linux/module.h> #include <linux/netdevice.h> #include <linux/utsname.h> -#include <linux/vermagic.h> +#include <generated/utsrelease.h> #include "ionic.h" #include "ionic_bus.h" @@ -152,6 +152,8 @@ static const char *ionic_opcode_to_str(enum ionic_cmd_opcode opcode) return "IONIC_CMD_RX_FILTER_ADD"; case IONIC_CMD_RX_FILTER_DEL: return "IONIC_CMD_RX_FILTER_DEL"; + case IONIC_CMD_Q_IDENTIFY: + return "IONIC_CMD_Q_IDENTIFY"; case IONIC_CMD_Q_INIT: return "IONIC_CMD_Q_INIT"; case IONIC_CMD_Q_CONTROL: @@ -356,7 +358,7 @@ try_again: done = ionic_dev_cmd_done(idev); if (done) break; - msleep(20); + msleep(5); hb = ionic_heartbeat_check(ionic); } while (!done && !hb && time_before(jiffies, max_wait)); duration = jiffies - start_time; @@ -413,6 +415,7 @@ int ionic_setup(struct ionic *ionic) err = ionic_dev_setup(ionic); if (err) return err; + ionic_reset(ionic); return 0; } diff --git a/drivers/net/ethernet/pensando/ionic/ionic_stats.c b/drivers/net/ethernet/pensando/ionic/ionic_stats.c index 8f2a8fb029f1..2a1885da58a6 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_stats.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_stats.c @@ -15,11 +15,109 @@ static const struct ionic_stat_desc ionic_lif_stats_desc[] = { IONIC_LIF_STAT_DESC(rx_packets), IONIC_LIF_STAT_DESC(rx_bytes), IONIC_LIF_STAT_DESC(tx_tso), - IONIC_LIF_STAT_DESC(tx_no_csum), + IONIC_LIF_STAT_DESC(tx_tso_bytes), + IONIC_LIF_STAT_DESC(tx_csum_none), IONIC_LIF_STAT_DESC(tx_csum), IONIC_LIF_STAT_DESC(rx_csum_none), IONIC_LIF_STAT_DESC(rx_csum_complete), IONIC_LIF_STAT_DESC(rx_csum_error), + IONIC_LIF_STAT_DESC(hw_tx_dropped), + IONIC_LIF_STAT_DESC(hw_rx_dropped), + IONIC_LIF_STAT_DESC(hw_rx_over_errors), + IONIC_LIF_STAT_DESC(hw_rx_missed_errors), + IONIC_LIF_STAT_DESC(hw_tx_aborted_errors), +}; + +static const struct ionic_stat_desc ionic_port_stats_desc[] = { + IONIC_PORT_STAT_DESC(frames_rx_ok), + IONIC_PORT_STAT_DESC(frames_rx_all), + IONIC_PORT_STAT_DESC(frames_rx_bad_fcs), + IONIC_PORT_STAT_DESC(frames_rx_bad_all), + IONIC_PORT_STAT_DESC(octets_rx_ok), + IONIC_PORT_STAT_DESC(octets_rx_all), + IONIC_PORT_STAT_DESC(frames_rx_unicast), + IONIC_PORT_STAT_DESC(frames_rx_multicast), + IONIC_PORT_STAT_DESC(frames_rx_broadcast), + IONIC_PORT_STAT_DESC(frames_rx_pause), + IONIC_PORT_STAT_DESC(frames_rx_bad_length), + IONIC_PORT_STAT_DESC(frames_rx_undersized), + IONIC_PORT_STAT_DESC(frames_rx_oversized), + IONIC_PORT_STAT_DESC(frames_rx_fragments), + IONIC_PORT_STAT_DESC(frames_rx_jabber), + IONIC_PORT_STAT_DESC(frames_rx_pripause), + IONIC_PORT_STAT_DESC(frames_rx_stomped_crc), + IONIC_PORT_STAT_DESC(frames_rx_too_long), + IONIC_PORT_STAT_DESC(frames_rx_vlan_good), + IONIC_PORT_STAT_DESC(frames_rx_dropped), + IONIC_PORT_STAT_DESC(frames_rx_less_than_64b), + IONIC_PORT_STAT_DESC(frames_rx_64b), + IONIC_PORT_STAT_DESC(frames_rx_65b_127b), + IONIC_PORT_STAT_DESC(frames_rx_128b_255b), + IONIC_PORT_STAT_DESC(frames_rx_256b_511b), + IONIC_PORT_STAT_DESC(frames_rx_512b_1023b), + IONIC_PORT_STAT_DESC(frames_rx_1024b_1518b), + IONIC_PORT_STAT_DESC(frames_rx_1519b_2047b), + IONIC_PORT_STAT_DESC(frames_rx_2048b_4095b), + IONIC_PORT_STAT_DESC(frames_rx_4096b_8191b), + IONIC_PORT_STAT_DESC(frames_rx_8192b_9215b), + IONIC_PORT_STAT_DESC(frames_rx_other), + IONIC_PORT_STAT_DESC(frames_tx_ok), + IONIC_PORT_STAT_DESC(frames_tx_all), + IONIC_PORT_STAT_DESC(frames_tx_bad), + IONIC_PORT_STAT_DESC(octets_tx_ok), + IONIC_PORT_STAT_DESC(octets_tx_total), + IONIC_PORT_STAT_DESC(frames_tx_unicast), + IONIC_PORT_STAT_DESC(frames_tx_multicast), + IONIC_PORT_STAT_DESC(frames_tx_broadcast), + IONIC_PORT_STAT_DESC(frames_tx_pause), + IONIC_PORT_STAT_DESC(frames_tx_pripause), + IONIC_PORT_STAT_DESC(frames_tx_vlan), + IONIC_PORT_STAT_DESC(frames_tx_less_than_64b), + IONIC_PORT_STAT_DESC(frames_tx_64b), + IONIC_PORT_STAT_DESC(frames_tx_65b_127b), + IONIC_PORT_STAT_DESC(frames_tx_128b_255b), + IONIC_PORT_STAT_DESC(frames_tx_256b_511b), + IONIC_PORT_STAT_DESC(frames_tx_512b_1023b), + IONIC_PORT_STAT_DESC(frames_tx_1024b_1518b), + IONIC_PORT_STAT_DESC(frames_tx_1519b_2047b), + IONIC_PORT_STAT_DESC(frames_tx_2048b_4095b), + IONIC_PORT_STAT_DESC(frames_tx_4096b_8191b), + IONIC_PORT_STAT_DESC(frames_tx_8192b_9215b), + IONIC_PORT_STAT_DESC(frames_tx_other), + IONIC_PORT_STAT_DESC(frames_tx_pri_0), + IONIC_PORT_STAT_DESC(frames_tx_pri_1), + IONIC_PORT_STAT_DESC(frames_tx_pri_2), + IONIC_PORT_STAT_DESC(frames_tx_pri_3), + IONIC_PORT_STAT_DESC(frames_tx_pri_4), + IONIC_PORT_STAT_DESC(frames_tx_pri_5), + IONIC_PORT_STAT_DESC(frames_tx_pri_6), + IONIC_PORT_STAT_DESC(frames_tx_pri_7), + IONIC_PORT_STAT_DESC(frames_rx_pri_0), + IONIC_PORT_STAT_DESC(frames_rx_pri_1), + IONIC_PORT_STAT_DESC(frames_rx_pri_2), + IONIC_PORT_STAT_DESC(frames_rx_pri_3), + IONIC_PORT_STAT_DESC(frames_rx_pri_4), + IONIC_PORT_STAT_DESC(frames_rx_pri_5), + IONIC_PORT_STAT_DESC(frames_rx_pri_6), + IONIC_PORT_STAT_DESC(frames_rx_pri_7), + IONIC_PORT_STAT_DESC(tx_pripause_0_1us_count), + IONIC_PORT_STAT_DESC(tx_pripause_1_1us_count), + IONIC_PORT_STAT_DESC(tx_pripause_2_1us_count), + IONIC_PORT_STAT_DESC(tx_pripause_3_1us_count), + IONIC_PORT_STAT_DESC(tx_pripause_4_1us_count), + IONIC_PORT_STAT_DESC(tx_pripause_5_1us_count), + IONIC_PORT_STAT_DESC(tx_pripause_6_1us_count), + IONIC_PORT_STAT_DESC(tx_pripause_7_1us_count), + IONIC_PORT_STAT_DESC(rx_pripause_0_1us_count), + IONIC_PORT_STAT_DESC(rx_pripause_1_1us_count), + IONIC_PORT_STAT_DESC(rx_pripause_2_1us_count), + IONIC_PORT_STAT_DESC(rx_pripause_3_1us_count), + IONIC_PORT_STAT_DESC(rx_pripause_4_1us_count), + IONIC_PORT_STAT_DESC(rx_pripause_5_1us_count), + IONIC_PORT_STAT_DESC(rx_pripause_6_1us_count), + IONIC_PORT_STAT_DESC(rx_pripause_7_1us_count), + IONIC_PORT_STAT_DESC(rx_pause_1us_count), + IONIC_PORT_STAT_DESC(frames_tx_truncated), }; static const struct ionic_stat_desc ionic_tx_stats_desc[] = { @@ -29,6 +127,11 @@ static const struct ionic_stat_desc ionic_tx_stats_desc[] = { IONIC_TX_STAT_DESC(dma_map_err), IONIC_TX_STAT_DESC(linearize), IONIC_TX_STAT_DESC(frags), + IONIC_TX_STAT_DESC(tso), + IONIC_TX_STAT_DESC(tso_bytes), + IONIC_TX_STAT_DESC(csum_none), + IONIC_TX_STAT_DESC(csum), + IONIC_TX_STAT_DESC(vlan_inserted), }; static const struct ionic_stat_desc ionic_rx_stats_desc[] = { @@ -40,6 +143,7 @@ static const struct ionic_stat_desc ionic_rx_stats_desc[] = { IONIC_RX_STAT_DESC(csum_complete), IONIC_RX_STAT_DESC(csum_error), IONIC_RX_STAT_DESC(dropped), + IONIC_RX_STAT_DESC(vlan_stripped), }; static const struct ionic_stat_desc ionic_txq_stats_desc[] = { @@ -62,6 +166,7 @@ static const struct ionic_stat_desc ionic_dbg_napi_stats_desc[] = { }; #define IONIC_NUM_LIF_STATS ARRAY_SIZE(ionic_lif_stats_desc) +#define IONIC_NUM_PORT_STATS ARRAY_SIZE(ionic_port_stats_desc) #define IONIC_NUM_TX_STATS ARRAY_SIZE(ionic_tx_stats_desc) #define IONIC_NUM_RX_STATS ARRAY_SIZE(ionic_rx_stats_desc) #define IONIC_NUM_TX_Q_STATS ARRAY_SIZE(ionic_txq_stats_desc) @@ -76,6 +181,7 @@ static void ionic_get_lif_stats(struct ionic_lif *lif, { struct ionic_tx_stats *tstats; struct ionic_rx_stats *rstats; + struct rtnl_link_stats64 ns; struct ionic_qcq *txqcq; struct ionic_qcq *rxqcq; int q_num; @@ -89,7 +195,8 @@ static void ionic_get_lif_stats(struct ionic_lif *lif, stats->tx_packets += tstats->pkts; stats->tx_bytes += tstats->bytes; stats->tx_tso += tstats->tso; - stats->tx_no_csum += tstats->no_csum; + stats->tx_tso_bytes += tstats->tso_bytes; + stats->tx_csum_none += tstats->csum_none; stats->tx_csum += tstats->csum; } @@ -103,6 +210,13 @@ static void ionic_get_lif_stats(struct ionic_lif *lif, stats->rx_csum_error += rstats->csum_error; } } + + ionic_get_stats64(lif->netdev, &ns); + stats->hw_tx_dropped = ns.tx_dropped; + stats->hw_rx_dropped = ns.rx_dropped; + stats->hw_rx_over_errors = ns.rx_over_errors; + stats->hw_rx_missed_errors = ns.rx_missed_errors; + stats->hw_tx_aborted_errors = ns.tx_aborted_errors; } static u64 ionic_sw_stats_get_count(struct ionic_lif *lif) @@ -118,6 +232,9 @@ static u64 ionic_sw_stats_get_count(struct ionic_lif *lif) /* rx stats */ total += MAX_Q(lif) * IONIC_NUM_RX_STATS; + /* port stats */ + total += IONIC_NUM_PORT_STATS; + if (test_bit(IONIC_LIF_F_UP, lif->state) && test_bit(IONIC_LIF_F_SW_DEBUG_STATS, lif->state)) { /* tx debug stats */ @@ -144,6 +261,13 @@ static void ionic_sw_stats_get_strings(struct ionic_lif *lif, u8 **buf) snprintf(*buf, ETH_GSTRING_LEN, ionic_lif_stats_desc[i].name); *buf += ETH_GSTRING_LEN; } + + for (i = 0; i < IONIC_NUM_PORT_STATS; i++) { + snprintf(*buf, ETH_GSTRING_LEN, + ionic_port_stats_desc[i].name); + *buf += ETH_GSTRING_LEN; + } + for (q_num = 0; q_num < MAX_Q(lif); q_num++) { for (i = 0; i < IONIC_NUM_TX_STATS; i++) { snprintf(*buf, ETH_GSTRING_LEN, "tx_%d_%s", @@ -225,6 +349,7 @@ static void ionic_sw_stats_get_strings(struct ionic_lif *lif, u8 **buf) static void ionic_sw_stats_get_values(struct ionic_lif *lif, u64 **buf) { + struct ionic_port_stats *port_stats; struct ionic_lif_sw_stats lif_stats; struct ionic_qcq *txqcq, *rxqcq; struct ionic_tx_stats *txstats; @@ -238,6 +363,13 @@ static void ionic_sw_stats_get_values(struct ionic_lif *lif, u64 **buf) (*buf)++; } + port_stats = &lif->ionic->idev.port_info->stats; + for (i = 0; i < IONIC_NUM_PORT_STATS; i++) { + **buf = IONIC_READ_STAT_LE64(port_stats, + &ionic_port_stats_desc[i]); + (*buf)++; + } + for (q_num = 0; q_num < MAX_Q(lif); q_num++) { txstats = &lif_to_txstats(lif, q_num); diff --git a/drivers/net/ethernet/pensando/ionic/ionic_stats.h b/drivers/net/ethernet/pensando/ionic/ionic_stats.h index d2c1122a2c6e..3f543512616e 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_stats.h +++ b/drivers/net/ethernet/pensando/ionic/ionic_stats.h @@ -11,6 +11,9 @@ .offset = IONIC_STAT_TO_OFFSET(type, stat_name) \ } +#define IONIC_PORT_STAT_DESC(stat_name) \ + IONIC_STAT_DESC(struct ionic_port_stats, stat_name) + #define IONIC_LIF_STAT_DESC(stat_name) \ IONIC_STAT_DESC(struct ionic_lif_sw_stats, stat_name) @@ -45,6 +48,9 @@ extern const int ionic_num_stats_grps; #define IONIC_READ_STAT64(base_ptr, desc_ptr) \ (*((u64 *)(((u8 *)(base_ptr)) + (desc_ptr)->offset))) +#define IONIC_READ_STAT_LE64(base_ptr, desc_ptr) \ + __le64_to_cpu(*((u64 *)(((u8 *)(base_ptr)) + (desc_ptr)->offset))) + struct ionic_stat_desc { char name[ETH_GSTRING_LEN]; u64 offset; diff --git a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c index d233b6e77b1e..b7f900c11834 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_txrx.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_txrx.c @@ -10,8 +10,10 @@ #include "ionic_lif.h" #include "ionic_txrx.h" -static void ionic_rx_clean(struct ionic_queue *q, struct ionic_desc_info *desc_info, - struct ionic_cq_info *cq_info, void *cb_arg); +static void ionic_rx_clean(struct ionic_queue *q, + struct ionic_desc_info *desc_info, + struct ionic_cq_info *cq_info, + void *cb_arg); static inline void ionic_txq_post(struct ionic_queue *q, bool ring_dbell, ionic_desc_cb cb_func, void *cb_arg) @@ -140,8 +142,10 @@ static struct sk_buff *ionic_rx_copybreak(struct ionic_queue *q, return skb; } -static void ionic_rx_clean(struct ionic_queue *q, struct ionic_desc_info *desc_info, - struct ionic_cq_info *cq_info, void *cb_arg) +static void ionic_rx_clean(struct ionic_queue *q, + struct ionic_desc_info *desc_info, + struct ionic_cq_info *cq_info, + void *cb_arg) { struct ionic_rxq_comp *comp = cq_info->cq_desc; struct ionic_qcq *qcq = q_to_qcq(q); @@ -210,10 +214,11 @@ static void ionic_rx_clean(struct ionic_queue *q, struct ionic_desc_info *desc_i (comp->csum_flags & IONIC_RXQ_COMP_CSUM_F_IP_BAD))) stats->csum_error++; - if (likely(netdev->features & NETIF_F_HW_VLAN_CTAG_RX)) { - if (comp->csum_flags & IONIC_RXQ_COMP_CSUM_F_VLAN) - __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), - le16_to_cpu(comp->vlan_tci)); + if (likely(netdev->features & NETIF_F_HW_VLAN_CTAG_RX) && + (comp->csum_flags & IONIC_RXQ_COMP_CSUM_F_VLAN)) { + __vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), + le16_to_cpu(comp->vlan_tci)); + stats->vlan_stripped++; } if (le16_to_cpu(comp->len) <= q->lif->rx_copybreak) @@ -475,7 +480,8 @@ int ionic_rx_napi(struct napi_struct *napi, int budget) return work_done; } -static dma_addr_t ionic_tx_map_single(struct ionic_queue *q, void *data, size_t len) +static dma_addr_t ionic_tx_map_single(struct ionic_queue *q, + void *data, size_t len) { struct ionic_tx_stats *stats = q_to_tx_stats(q); struct device *dev = q->lif->ionic->dev; @@ -491,7 +497,8 @@ static dma_addr_t ionic_tx_map_single(struct ionic_queue *q, void *data, size_t return dma_addr; } -static dma_addr_t ionic_tx_map_frag(struct ionic_queue *q, const skb_frag_t *frag, +static dma_addr_t ionic_tx_map_frag(struct ionic_queue *q, + const skb_frag_t *frag, size_t offset, size_t len) { struct ionic_tx_stats *stats = q_to_tx_stats(q); @@ -507,8 +514,10 @@ static dma_addr_t ionic_tx_map_frag(struct ionic_queue *q, const skb_frag_t *fra return dma_addr; } -static void ionic_tx_clean(struct ionic_queue *q, struct ionic_desc_info *desc_info, - struct ionic_cq_info *cq_info, void *cb_arg) +static void ionic_tx_clean(struct ionic_queue *q, + struct ionic_desc_info *desc_info, + struct ionic_cq_info *cq_info, + void *cb_arg) { struct ionic_txq_sg_desc *sg_desc = desc_info->sg_desc; struct ionic_txq_sg_elem *elem = sg_desc->elems; @@ -852,6 +861,7 @@ static int ionic_tx_tso(struct ionic_queue *q, struct sk_buff *skb) stats->pkts += total_pkts; stats->bytes += total_bytes; stats->tso++; + stats->tso_bytes += total_bytes; return 0; @@ -890,9 +900,12 @@ static int ionic_tx_calc_csum(struct ionic_queue *q, struct sk_buff *skb) flags, skb_shinfo(skb)->nr_frags, dma_addr); desc->cmd = cpu_to_le64(cmd); desc->len = cpu_to_le16(skb_headlen(skb)); - desc->vlan_tci = cpu_to_le16(skb_vlan_tag_get(skb)); desc->csum_start = cpu_to_le16(skb_checksum_start_offset(skb)); desc->csum_offset = cpu_to_le16(skb->csum_offset); + if (has_vlan) { + desc->vlan_tci = cpu_to_le16(skb_vlan_tag_get(skb)); + stats->vlan_inserted++; + } if (skb->csum_not_inet) stats->crc32_csum++; @@ -927,9 +940,12 @@ static int ionic_tx_calc_no_csum(struct ionic_queue *q, struct sk_buff *skb) flags, skb_shinfo(skb)->nr_frags, dma_addr); desc->cmd = cpu_to_le64(cmd); desc->len = cpu_to_le16(skb_headlen(skb)); - desc->vlan_tci = cpu_to_le16(skb_vlan_tag_get(skb)); + if (has_vlan) { + desc->vlan_tci = cpu_to_le16(skb_vlan_tag_get(skb)); + stats->vlan_inserted++; + } - stats->no_csum++; + stats->csum_none++; return 0; } @@ -989,6 +1005,7 @@ static int ionic_tx(struct ionic_queue *q, struct sk_buff *skb) static int ionic_tx_descs_needed(struct ionic_queue *q, struct sk_buff *skb) { + int sg_elems = q->lif->qtype_info[IONIC_QTYPE_TXQ].max_sg_elems; struct ionic_tx_stats *stats = q_to_tx_stats(q); int err; @@ -997,7 +1014,7 @@ static int ionic_tx_descs_needed(struct ionic_queue *q, struct sk_buff *skb) return (skb->len / skb_shinfo(skb)->gso_size) + 1; /* If non-TSO, just need 1 desc and nr_frags sg elems */ - if (skb_shinfo(skb)->nr_frags <= IONIC_TX_MAX_SG_ELEMS) + if (skb_shinfo(skb)->nr_frags <= sg_elems) return 1; /* Too many frags, so linearize */ diff --git a/drivers/net/ethernet/qlogic/qed/qed.h b/drivers/net/ethernet/qlogic/qed/qed.h index fa41bf08a589..66ed39d6f357 100644 --- a/drivers/net/ethernet/qlogic/qed/qed.h +++ b/drivers/net/ethernet/qlogic/qed/qed.h @@ -740,12 +740,6 @@ struct qed_dbg_feature { u32 dumped_dwords; }; -struct qed_dbg_params { - struct qed_dbg_feature features[DBG_FEATURE_NUM]; - u8 engine_for_debug; - bool print_data; -}; - struct qed_dev { u32 dp_module; u8 dp_level; @@ -844,6 +838,9 @@ struct qed_dev { /* Recovery */ bool recov_in_prog; + /* Indicates whether should prevent attentions from being reasserted */ + bool attn_clr_en; + /* LLH info */ u8 ppfid_bitmap; struct qed_llh_info *p_llh_info; @@ -872,17 +869,18 @@ struct qed_dev { } protocol_ops; void *ops_cookie; - struct qed_dbg_params dbg_params; - #ifdef CONFIG_QED_LL2 struct qed_cb_ll2_info *ll2; u8 ll2_mac_address[ETH_ALEN]; #endif struct qed_dbg_feature dbg_features[DBG_FEATURE_NUM]; + u8 engine_for_debug; bool disable_ilt_dump; DECLARE_HASHTABLE(connections, 10); const struct firmware *firmware; + bool print_dbg_data; + u32 rdma_max_sge; u32 rdma_max_inline; u32 rdma_max_srq_sge; @@ -1020,6 +1018,8 @@ u32 qed_unzip_data(struct qed_hwfn *p_hwfn, u32 input_len, u8 *input_buf, u32 max_size, u8 *unzip_buf); void qed_schedule_recovery_handler(struct qed_hwfn *p_hwfn); +void qed_hw_error_occurred(struct qed_hwfn *p_hwfn, + enum qed_hw_err_type err_type); void qed_get_protocol_stats(struct qed_dev *cdev, enum qed_mcp_protocol_type type, union qed_mcp_protocol_stats *stats); diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.c b/drivers/net/ethernet/qlogic/qed/qed_cxt.c index 1a636bad717d..7b76667acaba 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_cxt.c +++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.c @@ -110,6 +110,7 @@ struct src_ent { ALIGNED_TYPE_SIZE(union conn_context, p_hwfn) #define SRQ_CXT_SIZE (sizeof(struct rdma_srq_context)) +#define XRC_SRQ_CXT_SIZE (sizeof(struct rdma_xrc_srq_context)) #define TYPE0_TASK_CXT_SIZE(p_hwfn) \ ALIGNED_TYPE_SIZE(union type0_task_context, p_hwfn) @@ -293,18 +294,40 @@ static struct qed_tid_seg *qed_cxt_tid_seg_info(struct qed_hwfn *p_hwfn, return NULL; } -static void qed_cxt_set_srq_count(struct qed_hwfn *p_hwfn, u32 num_srqs) +static void qed_cxt_set_srq_count(struct qed_hwfn *p_hwfn, + u32 num_srqs, u32 num_xrc_srqs) { struct qed_cxt_mngr *p_mgr = p_hwfn->p_cxt_mngr; p_mgr->srq_count = num_srqs; + p_mgr->xrc_srq_count = num_xrc_srqs; } -u32 qed_cxt_get_srq_count(struct qed_hwfn *p_hwfn) +u32 qed_cxt_get_ilt_page_size(struct qed_hwfn *p_hwfn, + enum ilt_clients ilt_client) +{ + struct qed_cxt_mngr *p_mngr = p_hwfn->p_cxt_mngr; + struct qed_ilt_client_cfg *p_cli = &p_mngr->clients[ilt_client]; + + return ILT_PAGE_IN_BYTES(p_cli->p_size.val); +} + +static u32 qed_cxt_xrc_srqs_per_page(struct qed_hwfn *p_hwfn) +{ + u32 page_size; + + page_size = qed_cxt_get_ilt_page_size(p_hwfn, ILT_CLI_TSDM); + return page_size / XRC_SRQ_CXT_SIZE; +} + +u32 qed_cxt_get_total_srq_count(struct qed_hwfn *p_hwfn) { struct qed_cxt_mngr *p_mgr = p_hwfn->p_cxt_mngr; + u32 total_srqs; + + total_srqs = p_mgr->srq_count + p_mgr->xrc_srq_count; - return p_mgr->srq_count; + return total_srqs; } /* set the iids count per protocol */ @@ -692,7 +715,7 @@ int qed_cxt_cfg_ilt_compute(struct qed_hwfn *p_hwfn, u32 *line_count) } /* TSDM (SRQ CONTEXT) */ - total = qed_cxt_get_srq_count(p_hwfn); + total = qed_cxt_get_total_srq_count(p_hwfn); if (total) { p_cli = qed_cxt_set_cli(&p_mngr->clients[ILT_CLI_TSDM]); @@ -1962,11 +1985,9 @@ static void qed_rdma_set_pf_params(struct qed_hwfn *p_hwfn, struct qed_rdma_pf_params *p_params, u32 num_tasks) { - u32 num_cons, num_qps, num_srqs; + u32 num_cons, num_qps; enum protocol_type proto; - num_srqs = min_t(u32, QED_RDMA_MAX_SRQS, p_params->num_srqs); - if (p_hwfn->mcp_info->func_info.protocol == QED_PCI_ETH_RDMA) { DP_NOTICE(p_hwfn, "Current day drivers don't support RoCE & iWARP simultaneously on the same PF. Default to RoCE-only\n"); @@ -1989,6 +2010,8 @@ static void qed_rdma_set_pf_params(struct qed_hwfn *p_hwfn, } if (num_cons && num_tasks) { + u32 num_srqs, num_xrc_srqs; + qed_cxt_set_proto_cid_count(p_hwfn, proto, num_cons, 0); /* Deliberatly passing ROCE for tasks id. This is because @@ -1997,7 +2020,13 @@ static void qed_rdma_set_pf_params(struct qed_hwfn *p_hwfn, qed_cxt_set_proto_tid_count(p_hwfn, PROTOCOLID_ROCE, QED_CXT_ROCE_TID_SEG, 1, num_tasks, false); - qed_cxt_set_srq_count(p_hwfn, num_srqs); + + num_srqs = min_t(u32, QED_RDMA_MAX_SRQS, p_params->num_srqs); + + /* XRC SRQs populate a single ILT page */ + num_xrc_srqs = qed_cxt_xrc_srqs_per_page(p_hwfn); + + qed_cxt_set_srq_count(p_hwfn, num_srqs, num_xrc_srqs); } else { DP_INFO(p_hwfn->cdev, "RDMA personality used without setting params!\n"); @@ -2163,10 +2192,17 @@ qed_cxt_dynamic_ilt_alloc(struct qed_hwfn *p_hwfn, p_blk = &p_cli->pf_blks[CDUC_BLK]; break; case QED_ELEM_SRQ: + /* The first ILT page is not used for regular SRQs. Skip it. */ + iid += p_hwfn->p_cxt_mngr->xrc_srq_count; p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_TSDM]; elem_size = SRQ_CXT_SIZE; p_blk = &p_cli->pf_blks[SRQ_BLK]; break; + case QED_ELEM_XRC_SRQ: + p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_TSDM]; + elem_size = XRC_SRQ_CXT_SIZE; + p_blk = &p_cli->pf_blks[SRQ_BLK]; + break; case QED_ELEM_TASK: p_cli = &p_hwfn->p_cxt_mngr->clients[ILT_CLI_CDUT]; elem_size = TYPE1_TASK_CXT_SIZE(p_hwfn); @@ -2386,8 +2422,12 @@ int qed_cxt_free_proto_ilt(struct qed_hwfn *p_hwfn, enum protocol_type proto) return rc; /* Free TSDM CXT */ - rc = qed_cxt_free_ilt_range(p_hwfn, QED_ELEM_SRQ, 0, - qed_cxt_get_srq_count(p_hwfn)); + rc = qed_cxt_free_ilt_range(p_hwfn, QED_ELEM_XRC_SRQ, 0, + p_hwfn->p_cxt_mngr->xrc_srq_count); + + rc = qed_cxt_free_ilt_range(p_hwfn, QED_ELEM_SRQ, + p_hwfn->p_cxt_mngr->xrc_srq_count, + p_hwfn->p_cxt_mngr->srq_count); return rc; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_cxt.h b/drivers/net/ethernet/qlogic/qed/qed_cxt.h index c4e815f6cabd..ce08ae8d8498 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_cxt.h +++ b/drivers/net/ethernet/qlogic/qed/qed_cxt.h @@ -82,7 +82,8 @@ int qed_cxt_get_tid_mem_info(struct qed_hwfn *p_hwfn, enum qed_cxt_elem_type { QED_ELEM_CXT, QED_ELEM_SRQ, - QED_ELEM_TASK + QED_ELEM_TASK, + QED_ELEM_XRC_SRQ, }; u32 qed_cxt_get_proto_cid_count(struct qed_hwfn *p_hwfn, @@ -235,7 +236,6 @@ u32 qed_cxt_get_proto_tid_count(struct qed_hwfn *p_hwfn, enum protocol_type type); u32 qed_cxt_get_proto_cid_start(struct qed_hwfn *p_hwfn, enum protocol_type type); -u32 qed_cxt_get_srq_count(struct qed_hwfn *p_hwfn); int qed_cxt_free_proto_ilt(struct qed_hwfn *p_hwfn, enum protocol_type proto); #define QED_CTX_WORKING_MEM 0 @@ -358,6 +358,7 @@ struct qed_cxt_mngr { /* total number of SRQ's for this hwfn */ u32 srq_count; + u32 xrc_srq_count; /* Maximal number of L2 steering filters */ u32 arfs_count; @@ -372,4 +373,9 @@ u16 qed_get_cdut_num_vf_init_pages(struct qed_hwfn *p_hwfn); u16 qed_get_cdut_num_pf_work_pages(struct qed_hwfn *p_hwfn); u16 qed_get_cdut_num_vf_work_pages(struct qed_hwfn *p_hwfn); +u32 qed_cxt_get_ilt_page_size(struct qed_hwfn *p_hwfn, + enum ilt_clients ilt_client); + +u32 qed_cxt_get_total_srq_count(struct qed_hwfn *p_hwfn); + #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_debug.c b/drivers/net/ethernet/qlogic/qed/qed_debug.c index f4eebaabb6d0..57a0dab88431 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_debug.c +++ b/drivers/net/ethernet/qlogic/qed/qed_debug.c @@ -7453,7 +7453,7 @@ static enum dbg_status format_feature(struct qed_hwfn *p_hwfn, enum qed_dbg_features feature_idx) { struct qed_dbg_feature *feature = - &p_hwfn->cdev->dbg_params.features[feature_idx]; + &p_hwfn->cdev->dbg_features[feature_idx]; u32 text_size_bytes, null_char_pos, i; enum dbg_status rc; char *text_buf; @@ -7502,7 +7502,7 @@ static enum dbg_status format_feature(struct qed_hwfn *p_hwfn, text_buf[i] = '\n'; /* Dump printable feature to log */ - if (p_hwfn->cdev->dbg_params.print_data) + if (p_hwfn->cdev->print_dbg_data) qed_dbg_print_feature(text_buf, text_size_bytes); /* Free the old dump_buf and point the dump_buf to the newly allocagted @@ -7523,7 +7523,7 @@ static enum dbg_status qed_dbg_dump(struct qed_hwfn *p_hwfn, enum qed_dbg_features feature_idx) { struct qed_dbg_feature *feature = - &p_hwfn->cdev->dbg_params.features[feature_idx]; + &p_hwfn->cdev->dbg_features[feature_idx]; u32 buf_size_dwords; enum dbg_status rc; @@ -7648,7 +7648,7 @@ static int qed_dbg_nvm_image(struct qed_dev *cdev, void *buffer, enum qed_nvm_images image_id) { struct qed_hwfn *p_hwfn = - &cdev->hwfns[cdev->dbg_params.engine_for_debug]; + &cdev->hwfns[cdev->engine_for_debug]; u32 len_rounded, i; __be32 val; int rc; @@ -7780,7 +7780,7 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer) { u8 cur_engine, omit_engine = 0, org_engine; struct qed_hwfn *p_hwfn = - &cdev->hwfns[cdev->dbg_params.engine_for_debug]; + &cdev->hwfns[cdev->engine_for_debug]; struct dbg_tools_data *dev_data = &p_hwfn->dbg_info; int grc_params[MAX_DBG_GRC_PARAMS], i; u32 offset = 0, feature_size; @@ -8000,7 +8000,7 @@ int qed_dbg_all_data(struct qed_dev *cdev, void *buffer) int qed_dbg_all_data_size(struct qed_dev *cdev) { struct qed_hwfn *p_hwfn = - &cdev->hwfns[cdev->dbg_params.engine_for_debug]; + &cdev->hwfns[cdev->engine_for_debug]; u32 regs_len = 0, image_len = 0, ilt_len = 0, total_ilt_len = 0; u8 cur_engine, org_engine; @@ -8059,9 +8059,9 @@ int qed_dbg_feature(struct qed_dev *cdev, void *buffer, enum qed_dbg_features feature, u32 *num_dumped_bytes) { struct qed_hwfn *p_hwfn = - &cdev->hwfns[cdev->dbg_params.engine_for_debug]; + &cdev->hwfns[cdev->engine_for_debug]; struct qed_dbg_feature *qed_feature = - &cdev->dbg_params.features[feature]; + &cdev->dbg_features[feature]; enum dbg_status dbg_rc; struct qed_ptt *p_ptt; int rc = 0; @@ -8084,7 +8084,7 @@ int qed_dbg_feature(struct qed_dev *cdev, void *buffer, DP_VERBOSE(cdev, QED_MSG_DEBUG, "copying debugfs feature to external buffer\n"); memcpy(buffer, qed_feature->dump_buf, qed_feature->buf_size); - *num_dumped_bytes = cdev->dbg_params.features[feature].dumped_dwords * + *num_dumped_bytes = cdev->dbg_features[feature].dumped_dwords * 4; out: @@ -8095,7 +8095,7 @@ out: int qed_dbg_feature_size(struct qed_dev *cdev, enum qed_dbg_features feature) { struct qed_hwfn *p_hwfn = - &cdev->hwfns[cdev->dbg_params.engine_for_debug]; + &cdev->hwfns[cdev->engine_for_debug]; struct qed_dbg_feature *qed_feature = &cdev->dbg_features[feature]; struct qed_ptt *p_ptt = qed_ptt_acquire(p_hwfn); u32 buf_size_dwords; @@ -8120,14 +8120,14 @@ int qed_dbg_feature_size(struct qed_dev *cdev, enum qed_dbg_features feature) u8 qed_get_debug_engine(struct qed_dev *cdev) { - return cdev->dbg_params.engine_for_debug; + return cdev->engine_for_debug; } void qed_set_debug_engine(struct qed_dev *cdev, int engine_number) { DP_VERBOSE(cdev, QED_MSG_DEBUG, "set debug engine to %d\n", engine_number); - cdev->dbg_params.engine_for_debug = engine_number; + cdev->engine_for_debug = engine_number; } void qed_dbg_pf_init(struct qed_dev *cdev) @@ -8146,7 +8146,7 @@ void qed_dbg_pf_init(struct qed_dev *cdev) } /* Set the hwfn to be 0 as default */ - cdev->dbg_params.engine_for_debug = 0; + cdev->engine_for_debug = 0; } void qed_dbg_pf_exit(struct qed_dev *cdev) diff --git a/drivers/net/ethernet/qlogic/qed/qed_dev.c b/drivers/net/ethernet/qlogic/qed/qed_dev.c index 38a65b984e47..1eebf30fa798 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_dev.c +++ b/drivers/net/ethernet/qlogic/qed/qed_dev.c @@ -1972,7 +1972,7 @@ static int qed_init_qm_sanity(struct qed_hwfn *p_hwfn) return 0; if (QED_IS_ROCE_PERSONALITY(p_hwfn)) { - p_hwfn->hw_info.multi_tc_roce_en = 0; + p_hwfn->hw_info.multi_tc_roce_en = false; DP_NOTICE(p_hwfn, "multi-tc roce was disabled to reduce requested amount of pqs\n"); if (qed_init_qm_get_num_pqs(p_hwfn) <= RESC_NUM(p_hwfn, QED_PQ)) @@ -2269,6 +2269,7 @@ int qed_resc_alloc(struct qed_dev *cdev) /* EQ */ n_eqes = qed_chain_get_capacity(&p_hwfn->p_spq->chain); if (QED_IS_RDMA_PERSONALITY(p_hwfn)) { + u32 n_srq = qed_cxt_get_total_srq_count(p_hwfn); enum protocol_type rdma_proto; if (QED_IS_ROCE_PERSONALITY(p_hwfn)) @@ -2279,7 +2280,10 @@ int qed_resc_alloc(struct qed_dev *cdev) num_cons = qed_cxt_get_proto_cid_count(p_hwfn, rdma_proto, NULL) * 2; - n_eqes += num_cons + 2 * MAX_NUM_VFS_BB; + /* EQ should be able to get events from all SRQ's + * at the same time + */ + n_eqes += num_cons + 2 * MAX_NUM_VFS_BB + n_srq; } else if (p_hwfn->hw_info.personality == QED_PCI_ISCSI) { num_cons = qed_cxt_get_proto_cid_count(p_hwfn, @@ -3085,7 +3089,9 @@ int qed_hw_init(struct qed_dev *cdev, struct qed_hw_init_params *p_params) rc = qed_final_cleanup(p_hwfn, p_hwfn->p_main_ptt, p_hwfn->rel_pf_id, false); if (rc) { - DP_NOTICE(p_hwfn, "Final cleanup failed\n"); + qed_hw_err_notify(p_hwfn, p_hwfn->p_main_ptt, + QED_HW_ERR_RAMROD_FAIL, + "Final cleanup failed\n"); goto load_err; } } @@ -4392,7 +4398,7 @@ qed_get_hw_info(struct qed_hwfn *p_hwfn, } if (QED_IS_ROCE_PERSONALITY(p_hwfn)) - p_hwfn->hw_info.multi_tc_roce_en = 1; + p_hwfn->hw_info.multi_tc_roce_en = true; p_hwfn->hw_info.num_hw_tc = NUM_PHYS_TCS_4PORT_K2; p_hwfn->hw_info.num_active_tc = 1; diff --git a/drivers/net/ethernet/qlogic/qed/qed_hsi.h b/drivers/net/ethernet/qlogic/qed/qed_hsi.h index 4597015b8bff..f00460d00cab 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hsi.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hsi.h @@ -12400,6 +12400,13 @@ struct load_rsp_stc { #define LOAD_RSP_FLAGS0_DRV_EXISTS (0x1 << 0) }; +struct mdump_retain_data_stc { + u32 valid; + u32 epoch; + u32 pf; + u32 status; +}; + union drv_union_data { u32 ver_str[MCP_DRV_VER_STR_SIZE_DWORD]; struct mcp_mac wol_mac; @@ -12488,10 +12495,14 @@ struct public_drv_mb { #define DRV_MSG_CODE_BIST_TEST 0x001e0000 #define DRV_MSG_CODE_SET_LED_MODE 0x00200000 #define DRV_MSG_CODE_RESOURCE_CMD 0x00230000 +/* Send crash dump commands with param[3:0] - opcode */ +#define DRV_MSG_CODE_MDUMP_CMD 0x00250000 #define DRV_MSG_CODE_GET_TLV_DONE 0x002f0000 #define DRV_MSG_CODE_GET_ENGINE_CONFIG 0x00370000 #define DRV_MSG_CODE_GET_PPFID_BITMAP 0x43000000 +#define DRV_MSG_CODE_DEBUG_DATA_SEND 0xc0040000 + #define RESOURCE_CMD_REQ_RESC_MASK 0x0000001F #define RESOURCE_CMD_REQ_RESC_SHIFT 0 #define RESOURCE_CMD_REQ_OPCODE_MASK 0x000000E0 @@ -12517,6 +12528,21 @@ struct public_drv_mb { #define RESOURCE_DUMP 0 +/* DRV_MSG_CODE_MDUMP_CMD parameters */ +#define MDUMP_DRV_PARAM_OPCODE_MASK 0x0000000f +#define DRV_MSG_CODE_MDUMP_ACK 0x01 +#define DRV_MSG_CODE_MDUMP_SET_VALUES 0x02 +#define DRV_MSG_CODE_MDUMP_TRIGGER 0x03 +#define DRV_MSG_CODE_MDUMP_GET_CONFIG 0x04 +#define DRV_MSG_CODE_MDUMP_SET_ENABLE 0x05 +#define DRV_MSG_CODE_MDUMP_CLEAR_LOGS 0x06 +#define DRV_MSG_CODE_MDUMP_GET_RETAIN 0x07 +#define DRV_MSG_CODE_MDUMP_CLR_RETAIN 0x08 + +#define DRV_MSG_CODE_HW_DUMP_TRIGGER 0x0a +#define DRV_MSG_CODE_MDUMP_GEN_MDUMP2 0x0b +#define DRV_MSG_CODE_MDUMP_FREE_MDUMP2 0x0c + #define DRV_MSG_CODE_GET_PF_RDMA_PROTOCOL 0x002b0000 #define DRV_MSG_CODE_OS_WOL 0x002e0000 @@ -12626,6 +12652,17 @@ struct public_drv_mb { #define DRV_MB_PARAM_FEATURE_SUPPORT_PORT_EEE 0x00000002 #define DRV_MB_PARAM_FEATURE_SUPPORT_FUNC_VLINK 0x00010000 +/* DRV_MSG_CODE_DEBUG_DATA_SEND parameters */ +#define DRV_MSG_CODE_DEBUG_DATA_SEND_SIZE_OFFSET 0 +#define DRV_MSG_CODE_DEBUG_DATA_SEND_SIZE_MASK 0xFF + +/* Driver attributes params */ +#define DRV_MB_PARAM_ATTRIBUTE_KEY_OFFSET 0 +#define DRV_MB_PARAM_ATTRIBUTE_KEY_MASK 0x00FFFFFF +#define DRV_MB_PARAM_ATTRIBUTE_CMD_OFFSET 24 +#define DRV_MB_PARAM_ATTRIBUTE_CMD_MASK 0xFF000000 + +#define DRV_MB_PARAM_NVM_CFG_OPTION_ID_OFFSET 0 #define DRV_MB_PARAM_NVM_CFG_OPTION_ID_SHIFT 0 #define DRV_MB_PARAM_NVM_CFG_OPTION_ID_MASK 0x0000FFFF #define DRV_MB_PARAM_NVM_CFG_OPTION_ALL_SHIFT 16 @@ -12678,6 +12715,14 @@ struct public_drv_mb { #define FW_MSG_CODE_DRV_CFG_PF_VFS_MSIX_DONE 0x00870000 #define FW_MSG_SEQ_NUMBER_MASK 0x0000ffff +#define FW_MSG_CODE_DEBUG_DATA_SEND_INV_ARG 0xb0070000 +#define FW_MSG_CODE_DEBUG_DATA_SEND_BUF_FULL 0xb0080000 +#define FW_MSG_CODE_DEBUG_DATA_SEND_NO_BUF 0xb0090000 +#define FW_MSG_CODE_DEBUG_NOT_ENABLED 0xb00a0000 +#define FW_MSG_CODE_DEBUG_DATA_SEND_OK 0xb00b0000 + +#define FW_MSG_CODE_MDUMP_INVALID_CMD 0x00030000 + u32 fw_mb_param; #define FW_MB_PARAM_RESOURCE_ALLOC_VERSION_MAJOR_MASK 0xFFFF0000 #define FW_MB_PARAM_RESOURCE_ALLOC_VERSION_MAJOR_SHIFT 16 @@ -12742,9 +12787,9 @@ enum MFW_DRV_MSG_TYPE { MFW_DRV_MSG_GET_FCOE_STATS, MFW_DRV_MSG_GET_ISCSI_STATS, MFW_DRV_MSG_GET_RDMA_STATS, - MFW_DRV_MSG_BW_UPDATE10, + MFW_DRV_MSG_FAILURE_DETECTED, MFW_DRV_MSG_TRANSCEIVER_STATE_CHANGE, - MFW_DRV_MSG_BW_UPDATE11, + MFW_DRV_MSG_CRITICAL_ERROR_OCCURRED, MFW_DRV_MSG_RESERVED, MFW_DRV_MSG_GET_TLV_REQ, MFW_DRV_MSG_OEM_CFG_UPDATE, diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.c b/drivers/net/ethernet/qlogic/qed/qed_hw.c index 4ab8cfaf63d1..5fa251489536 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hw.c +++ b/drivers/net/ethernet/qlogic/qed/qed_hw.c @@ -762,9 +762,10 @@ static int qed_dmae_execute_command(struct qed_hwfn *p_hwfn, dst_type, length_cur); if (qed_status) { - DP_NOTICE(p_hwfn, - "qed_dmae_execute_sub_operation Failed with error 0x%x. source_addr 0x%llx, destination addr 0x%llx, size_in_dwords 0x%x\n", - qed_status, src_addr, dst_addr, length_cur); + qed_hw_err_notify(p_hwfn, p_ptt, QED_HW_ERR_DMAE_FAIL, + "qed_dmae_execute_sub_operation Failed with error 0x%x. source_addr 0x%llx, destination addr 0x%llx, size_in_dwords 0x%x\n", + qed_status, src_addr, + dst_addr, length_cur); break; } } @@ -837,6 +838,41 @@ int qed_dmae_host2host(struct qed_hwfn *p_hwfn, return rc; } +void qed_hw_err_notify(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + enum qed_hw_err_type err_type, char *fmt, ...) +{ + char buf[QED_HW_ERR_MAX_STR_SIZE]; + va_list vl; + int len; + + if (fmt) { + va_start(vl, fmt); + len = vsnprintf(buf, QED_HW_ERR_MAX_STR_SIZE, fmt, vl); + va_end(vl); + + if (len > QED_HW_ERR_MAX_STR_SIZE - 1) + len = QED_HW_ERR_MAX_STR_SIZE - 1; + + DP_NOTICE(p_hwfn, "%s", buf); + } + + /* Fan failure cannot be masked by handling of another HW error */ + if (p_hwfn->cdev->recov_in_prog && + err_type != QED_HW_ERR_FAN_FAIL) { + DP_VERBOSE(p_hwfn, + NETIF_MSG_DRV, + "Recovery is in progress. Avoid notifying about HW error %d.\n", + err_type); + return; + } + + qed_hw_error_occurred(p_hwfn, err_type); + + if (fmt) + qed_mcp_send_raw_debug_data(p_hwfn, p_ptt, buf, len); +} + int qed_dmae_sanity(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, const char *phase) { diff --git a/drivers/net/ethernet/qlogic/qed/qed_hw.h b/drivers/net/ethernet/qlogic/qed/qed_hw.h index 505e94db939d..f5b109b04b66 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_hw.h +++ b/drivers/net/ethernet/qlogic/qed/qed_hw.h @@ -315,4 +315,19 @@ int qed_init_fw_data(struct qed_dev *cdev, int qed_dmae_sanity(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, const char *phase); +#define QED_HW_ERR_MAX_STR_SIZE 256 + +/** + * @brief qed_hw_err_notify - Notify upper layer driver and management FW + * about a HW error. + * + * @param p_hwfn + * @param p_ptt + * @param err_type + * @param fmt - debug data buffer to send to the MFW + * @param ... - buffer format args + */ +void qed_hw_err_notify(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + enum qed_hw_err_type err_type, char *fmt, ...); #endif diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.c b/drivers/net/ethernet/qlogic/qed/qed_int.c index 9f5113639eaf..b7b974f0ef21 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.c +++ b/drivers/net/ethernet/qlogic/qed/qed_int.c @@ -96,6 +96,7 @@ struct aeu_invert_reg_bit { #define ATTENTION_BB(value) (value << ATTENTION_BB_SHIFT) #define ATTENTION_BB_DIFFERENT BIT(23) +#define ATTENTION_CLEAR_ENABLE BIT(28) unsigned int flags; /* Callback to call if attention will be triggered */ @@ -363,6 +364,21 @@ static int qed_pglueb_rbc_attn_cb(struct qed_hwfn *p_hwfn) return qed_pglueb_rbc_attn_handler(p_hwfn, p_hwfn->p_dpc_ptt); } +static int qed_fw_assertion(struct qed_hwfn *p_hwfn) +{ + qed_hw_err_notify(p_hwfn, p_hwfn->p_dpc_ptt, QED_HW_ERR_FW_ASSERT, + "FW assertion!\n"); + + return -EINVAL; +} + +static int qed_general_attention_35(struct qed_hwfn *p_hwfn) +{ + DP_INFO(p_hwfn, "General attention 35!\n"); + + return 0; +} + #define QED_DORQ_ATTENTION_REASON_MASK (0xfffff) #define QED_DORQ_ATTENTION_OPAQUE_MASK (0xffff) #define QED_DORQ_ATTENTION_OPAQUE_SHIFT (0x0) @@ -605,13 +621,15 @@ static struct aeu_invert_reg aeu_descs[NUM_ATTN_REGS] = { { { /* After Invert 4 */ - {"General Attention 32", ATTENTION_SINGLE, - NULL, MAX_BLOCK_ID}, + {"General Attention 32", ATTENTION_SINGLE | + ATTENTION_CLEAR_ENABLE, qed_fw_assertion, + MAX_BLOCK_ID}, {"General Attention %d", (2 << ATTENTION_LENGTH_SHIFT) | (33 << ATTENTION_OFFSET_SHIFT), NULL, MAX_BLOCK_ID}, - {"General Attention 35", ATTENTION_SINGLE, - NULL, MAX_BLOCK_ID}, + {"General Attention 35", ATTENTION_SINGLE | + ATTENTION_CLEAR_ENABLE, qed_general_attention_35, + MAX_BLOCK_ID}, {"NWS Parity", ATTENTION_PAR | ATTENTION_BB_DIFFERENT | ATTENTION_BB(AEU_INVERT_REG_SPECIAL_CNIG_0), @@ -927,9 +945,12 @@ qed_int_deassertion_aeu_bit(struct qed_hwfn *p_hwfn, qed_int_attn_print(p_hwfn, p_aeu->block_index, ATTN_TYPE_INTERRUPT, !b_fatal); - - /* If the attention is benign, no need to prevent it */ - if (!rc) + /* Reach assertion if attention is fatal */ + if (b_fatal) + qed_hw_err_notify(p_hwfn, p_hwfn->p_dpc_ptt, QED_HW_ERR_HW_ATTN, + "`%s': Fatal attention\n", + p_bit_name); + else /* If the attention is benign, no need to prevent it */ goto out; /* Prevent this Attention from being asserted in the future */ @@ -2349,6 +2370,11 @@ void qed_int_disable_post_isr_release(struct qed_dev *cdev) cdev->hwfns[i].b_int_requested = false; } +void qed_int_attn_clr_enable(struct qed_dev *cdev, bool clr_enable) +{ + cdev->attn_clr_en = clr_enable; +} + int qed_int_set_timer_res(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u8 timer_res, u16 sb_id, bool tx) { diff --git a/drivers/net/ethernet/qlogic/qed/qed_int.h b/drivers/net/ethernet/qlogic/qed/qed_int.h index 9ad568d93ae6..e09db3386367 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_int.h +++ b/drivers/net/ethernet/qlogic/qed/qed_int.h @@ -191,6 +191,17 @@ void qed_int_get_num_sbs(struct qed_hwfn *p_hwfn, void qed_int_disable_post_isr_release(struct qed_dev *cdev); /** + * @brief qed_int_attn_clr_enable - sets whether the general behavior is + * preventing attentions from being reasserted, or following the + * attributes of the specific attention. + * + * @param cdev + * @param clr_enable + * + */ +void qed_int_attn_clr_enable(struct qed_dev *cdev, bool clr_enable); + +/** * @brief - Doorbell Recovery handler. * Run doorbell recovery in case of PF overflow (and flush DORQ if * needed). diff --git a/drivers/net/ethernet/qlogic/qed/qed_ll2.c b/drivers/net/ethernet/qlogic/qed/qed_ll2.c index 037e5978787e..4afd8572ada6 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_ll2.c +++ b/drivers/net/ethernet/qlogic/qed/qed_ll2.c @@ -2331,7 +2331,7 @@ static void qed_ll2_register_cb_ops(struct qed_dev *cdev, cdev->ll2->cb_cookie = cookie; } -struct qed_ll2_cbs ll2_cbs = { +static struct qed_ll2_cbs ll2_cbs = { .rx_comp_cb = &qed_ll2b_complete_rx_packet, .rx_release_cb = &qed_ll2b_release_rx_packet, .tx_comp_cb = &qed_ll2b_complete_tx_packet, diff --git a/drivers/net/ethernet/qlogic/qed/qed_main.c b/drivers/net/ethernet/qlogic/qed/qed_main.c index 96356e897c80..83e798d4eebb 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_main.c +++ b/drivers/net/ethernet/qlogic/qed/qed_main.c @@ -49,6 +49,7 @@ #include <linux/qed/qed_if.h> #include <linux/qed/qed_ll2_if.h> #include <net/devlink.h> +#include <linux/aer.h> #include "qed.h" #include "qed_sriov.h" @@ -129,6 +130,8 @@ static void qed_free_pci(struct qed_dev *cdev) { struct pci_dev *pdev = cdev->pdev; + pci_disable_pcie_error_reporting(pdev); + if (cdev->doorbells && cdev->db_size) iounmap(cdev->doorbells); if (cdev->regview) @@ -231,6 +234,12 @@ static int qed_init_pci(struct qed_dev *cdev, struct pci_dev *pdev) return -ENOMEM; } + /* AER (Advanced Error reporting) configuration */ + rc = pci_enable_pcie_error_reporting(pdev); + if (rc) + DP_VERBOSE(cdev, NETIF_MSG_DRV, + "Failed to configure PCIe AER [%d]\n", rc); + return 0; err2: @@ -2459,6 +2468,39 @@ void qed_schedule_recovery_handler(struct qed_hwfn *p_hwfn) ops->schedule_recovery_handler(cookie); } +char *qed_hw_err_type_descr[] = { + [QED_HW_ERR_FAN_FAIL] = "Fan Failure", + [QED_HW_ERR_MFW_RESP_FAIL] = "MFW Response Failure", + [QED_HW_ERR_HW_ATTN] = "HW Attention", + [QED_HW_ERR_DMAE_FAIL] = "DMAE Failure", + [QED_HW_ERR_RAMROD_FAIL] = "Ramrod Failure", + [QED_HW_ERR_FW_ASSERT] = "FW Assertion", + [QED_HW_ERR_LAST] = "Unknown", +}; + +void qed_hw_error_occurred(struct qed_hwfn *p_hwfn, + enum qed_hw_err_type err_type) +{ + struct qed_common_cb_ops *ops = p_hwfn->cdev->protocol_ops.common; + void *cookie = p_hwfn->cdev->ops_cookie; + char *err_str; + + if (err_type > QED_HW_ERR_LAST) + err_type = QED_HW_ERR_LAST; + err_str = qed_hw_err_type_descr[err_type]; + + DP_NOTICE(p_hwfn, "HW error occurred [%s]\n", err_str); + + /* Call the HW error handler of the protocol driver. + * If it is not available - perform a minimal handling of preventing + * HW attentions from being reasserted. + */ + if (ops && ops->schedule_hw_err_handler) + ops->schedule_hw_err_handler(cookie, err_type); + else + qed_int_attn_clr_enable(p_hwfn->cdev, true); +} + static int qed_set_coalesce(struct qed_dev *cdev, u16 rx_coal, u16 tx_coal, void *handle) { @@ -2680,6 +2722,7 @@ const struct qed_common_ops qed_common_ops_pass = { .set_led = &qed_set_led, .recovery_process = &qed_recovery_process, .recovery_prolog = &qed_recovery_prolog, + .attn_clr_enable = &qed_int_attn_clr_enable, .update_drv_state = &qed_update_drv_state, .update_mac = &qed_update_mac, .update_mtu = &qed_update_mtu, diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c index 280527cc0578..9624616806e7 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c @@ -575,6 +575,8 @@ _qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn, if (!QED_MB_FLAGS_IS_SET(p_mb_params, AVOID_BLOCK)) qed_mcp_cmd_set_blocking(p_hwfn, true); + qed_hw_err_notify(p_hwfn, p_ptt, + QED_HW_ERR_MFW_RESP_FAIL, NULL); return -EAGAIN; } @@ -1704,6 +1706,127 @@ static void qed_mcp_update_stag(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) &resp, ¶m); } +static void qed_mcp_handle_fan_failure(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt) +{ + /* A single notification should be sent to upper driver in CMT mode */ + if (p_hwfn != QED_LEADING_HWFN(p_hwfn->cdev)) + return; + + qed_hw_err_notify(p_hwfn, p_ptt, QED_HW_ERR_FAN_FAIL, + "Fan failure was detected on the network interface card and it's going to be shut down.\n"); +} + +struct qed_mdump_cmd_params { + u32 cmd; + void *p_data_src; + u8 data_src_size; + void *p_data_dst; + u8 data_dst_size; + u32 mcp_resp; +}; + +static int +qed_mcp_mdump_cmd(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct qed_mdump_cmd_params *p_mdump_cmd_params) +{ + struct qed_mcp_mb_params mb_params; + int rc; + + memset(&mb_params, 0, sizeof(mb_params)); + mb_params.cmd = DRV_MSG_CODE_MDUMP_CMD; + mb_params.param = p_mdump_cmd_params->cmd; + mb_params.p_data_src = p_mdump_cmd_params->p_data_src; + mb_params.data_src_size = p_mdump_cmd_params->data_src_size; + mb_params.p_data_dst = p_mdump_cmd_params->p_data_dst; + mb_params.data_dst_size = p_mdump_cmd_params->data_dst_size; + rc = qed_mcp_cmd_and_union(p_hwfn, p_ptt, &mb_params); + if (rc) + return rc; + + p_mdump_cmd_params->mcp_resp = mb_params.mcp_resp; + + if (p_mdump_cmd_params->mcp_resp == FW_MSG_CODE_MDUMP_INVALID_CMD) { + DP_INFO(p_hwfn, + "The mdump sub command is unsupported by the MFW [mdump_cmd 0x%x]\n", + p_mdump_cmd_params->cmd); + rc = -EOPNOTSUPP; + } else if (p_mdump_cmd_params->mcp_resp == FW_MSG_CODE_UNSUPPORTED) { + DP_INFO(p_hwfn, + "The mdump command is not supported by the MFW\n"); + rc = -EOPNOTSUPP; + } + + return rc; +} + +static int qed_mcp_mdump_ack(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) +{ + struct qed_mdump_cmd_params mdump_cmd_params; + + memset(&mdump_cmd_params, 0, sizeof(mdump_cmd_params)); + mdump_cmd_params.cmd = DRV_MSG_CODE_MDUMP_ACK; + + return qed_mcp_mdump_cmd(p_hwfn, p_ptt, &mdump_cmd_params); +} + +int +qed_mcp_mdump_get_retain(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct mdump_retain_data_stc *p_mdump_retain) +{ + struct qed_mdump_cmd_params mdump_cmd_params; + int rc; + + memset(&mdump_cmd_params, 0, sizeof(mdump_cmd_params)); + mdump_cmd_params.cmd = DRV_MSG_CODE_MDUMP_GET_RETAIN; + mdump_cmd_params.p_data_dst = p_mdump_retain; + mdump_cmd_params.data_dst_size = sizeof(*p_mdump_retain); + + rc = qed_mcp_mdump_cmd(p_hwfn, p_ptt, &mdump_cmd_params); + if (rc) + return rc; + + if (mdump_cmd_params.mcp_resp != FW_MSG_CODE_OK) { + DP_INFO(p_hwfn, + "Failed to get the mdump retained data [mcp_resp 0x%x]\n", + mdump_cmd_params.mcp_resp); + return -EINVAL; + } + + return 0; +} + +static void qed_mcp_handle_critical_error(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt) +{ + struct mdump_retain_data_stc mdump_retain; + int rc; + + /* In CMT mode - no need for more than a single acknowledgment to the + * MFW, and no more than a single notification to the upper driver. + */ + if (p_hwfn != QED_LEADING_HWFN(p_hwfn->cdev)) + return; + + rc = qed_mcp_mdump_get_retain(p_hwfn, p_ptt, &mdump_retain); + if (rc == 0 && mdump_retain.valid) + DP_NOTICE(p_hwfn, + "The MFW notified that a critical error occurred in the device [epoch 0x%08x, pf 0x%x, status 0x%08x]\n", + mdump_retain.epoch, + mdump_retain.pf, mdump_retain.status); + else + DP_NOTICE(p_hwfn, + "The MFW notified that a critical error occurred in the device\n"); + + DP_NOTICE(p_hwfn, + "Acknowledging the notification to not allow the MFW crash dump [driver debug data collection is preferable]\n"); + qed_mcp_mdump_ack(p_hwfn, p_ptt); + + qed_hw_err_notify(p_hwfn, p_ptt, QED_HW_ERR_HW_ATTN, NULL); +} + void qed_mcp_read_ufp_config(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt) { struct public_func shmem_info; @@ -1850,6 +1973,12 @@ int qed_mcp_handle_events(struct qed_hwfn *p_hwfn, case MFW_DRV_MSG_S_TAG_UPDATE: qed_mcp_update_stag(p_hwfn, p_ptt); break; + case MFW_DRV_MSG_FAILURE_DETECTED: + qed_mcp_handle_fan_failure(p_hwfn, p_ptt); + break; + case MFW_DRV_MSG_CRITICAL_ERROR_OCCURRED: + qed_mcp_handle_critical_error(p_hwfn, p_ptt); + break; case MFW_DRV_MSG_GET_TLV_REQ: qed_mfw_tlv_req(p_hwfn); break; @@ -3819,3 +3948,127 @@ int qed_mcp_nvm_set_cfg(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, DRV_MSG_CODE_SET_NVM_CFG_OPTION, mb_param, &resp, ¶m, len, (u32 *)p_buf); } + +#define QED_MCP_DBG_DATA_MAX_SIZE MCP_DRV_NVM_BUF_LEN +#define QED_MCP_DBG_DATA_MAX_HEADER_SIZE sizeof(u32) +#define QED_MCP_DBG_DATA_MAX_PAYLOAD_SIZE \ + (QED_MCP_DBG_DATA_MAX_SIZE - QED_MCP_DBG_DATA_MAX_HEADER_SIZE) + +static int +__qed_mcp_send_debug_data(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, u8 *p_buf, u8 size) +{ + struct qed_mcp_mb_params mb_params; + int rc; + + if (size > QED_MCP_DBG_DATA_MAX_SIZE) { + DP_ERR(p_hwfn, + "Debug data size is %d while it should not exceed %d\n", + size, QED_MCP_DBG_DATA_MAX_SIZE); + return -EINVAL; + } + + memset(&mb_params, 0, sizeof(mb_params)); + mb_params.cmd = DRV_MSG_CODE_DEBUG_DATA_SEND; + SET_MFW_FIELD(mb_params.param, DRV_MSG_CODE_DEBUG_DATA_SEND_SIZE, size); + mb_params.p_data_src = p_buf; + mb_params.data_src_size = size; + rc = qed_mcp_cmd_and_union(p_hwfn, p_ptt, &mb_params); + if (rc) + return rc; + + if (mb_params.mcp_resp == FW_MSG_CODE_UNSUPPORTED) { + DP_INFO(p_hwfn, + "The DEBUG_DATA_SEND command is unsupported by the MFW\n"); + return -EOPNOTSUPP; + } else if (mb_params.mcp_resp == (u32)FW_MSG_CODE_DEBUG_NOT_ENABLED) { + DP_INFO(p_hwfn, "The DEBUG_DATA_SEND command is not enabled\n"); + return -EBUSY; + } else if (mb_params.mcp_resp != (u32)FW_MSG_CODE_DEBUG_DATA_SEND_OK) { + DP_NOTICE(p_hwfn, + "Failed to send debug data to the MFW [resp 0x%08x]\n", + mb_params.mcp_resp); + return -EINVAL; + } + + return 0; +} + +enum qed_mcp_dbg_data_type { + QED_MCP_DBG_DATA_TYPE_RAW, +}; + +/* Header format: [31:28] PFID, [27:20] flags, [19:12] type, [11:0] S/N */ +#define QED_MCP_DBG_DATA_HDR_SN_OFFSET 0 +#define QED_MCP_DBG_DATA_HDR_SN_MASK 0x00000fff +#define QED_MCP_DBG_DATA_HDR_TYPE_OFFSET 12 +#define QED_MCP_DBG_DATA_HDR_TYPE_MASK 0x000ff000 +#define QED_MCP_DBG_DATA_HDR_FLAGS_OFFSET 20 +#define QED_MCP_DBG_DATA_HDR_FLAGS_MASK 0x0ff00000 +#define QED_MCP_DBG_DATA_HDR_PF_OFFSET 28 +#define QED_MCP_DBG_DATA_HDR_PF_MASK 0xf0000000 + +#define QED_MCP_DBG_DATA_HDR_FLAGS_FIRST 0x1 +#define QED_MCP_DBG_DATA_HDR_FLAGS_LAST 0x2 + +static int +qed_mcp_send_debug_data(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + enum qed_mcp_dbg_data_type type, u8 *p_buf, u32 size) +{ + u8 raw_data[QED_MCP_DBG_DATA_MAX_SIZE], *p_tmp_buf = p_buf; + u32 tmp_size = size, *p_header, *p_payload; + u8 flags = 0; + u16 seq; + int rc; + + p_header = (u32 *)raw_data; + p_payload = (u32 *)(raw_data + QED_MCP_DBG_DATA_MAX_HEADER_SIZE); + + seq = (u16)atomic_inc_return(&p_hwfn->mcp_info->dbg_data_seq); + + /* First chunk is marked as 'first' */ + flags |= QED_MCP_DBG_DATA_HDR_FLAGS_FIRST; + + *p_header = 0; + SET_MFW_FIELD(*p_header, QED_MCP_DBG_DATA_HDR_SN, seq); + SET_MFW_FIELD(*p_header, QED_MCP_DBG_DATA_HDR_TYPE, type); + SET_MFW_FIELD(*p_header, QED_MCP_DBG_DATA_HDR_FLAGS, flags); + SET_MFW_FIELD(*p_header, QED_MCP_DBG_DATA_HDR_PF, p_hwfn->abs_pf_id); + + while (tmp_size > QED_MCP_DBG_DATA_MAX_PAYLOAD_SIZE) { + memcpy(p_payload, p_tmp_buf, QED_MCP_DBG_DATA_MAX_PAYLOAD_SIZE); + rc = __qed_mcp_send_debug_data(p_hwfn, p_ptt, raw_data, + QED_MCP_DBG_DATA_MAX_SIZE); + if (rc) + return rc; + + /* Clear the 'first' marking after sending the first chunk */ + if (p_tmp_buf == p_buf) { + flags &= ~QED_MCP_DBG_DATA_HDR_FLAGS_FIRST; + SET_MFW_FIELD(*p_header, QED_MCP_DBG_DATA_HDR_FLAGS, + flags); + } + + p_tmp_buf += QED_MCP_DBG_DATA_MAX_PAYLOAD_SIZE; + tmp_size -= QED_MCP_DBG_DATA_MAX_PAYLOAD_SIZE; + } + + /* Last chunk is marked as 'last' */ + flags |= QED_MCP_DBG_DATA_HDR_FLAGS_LAST; + SET_MFW_FIELD(*p_header, QED_MCP_DBG_DATA_HDR_FLAGS, flags); + memcpy(p_payload, p_tmp_buf, tmp_size); + + /* Casting the left size to u8 is ok since at this point it is <= 32 */ + return __qed_mcp_send_debug_data(p_hwfn, p_ptt, raw_data, + (u8)(QED_MCP_DBG_DATA_MAX_HEADER_SIZE + + tmp_size)); +} + +int +qed_mcp_send_raw_debug_data(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, u8 *p_buf, u32 size) +{ + return qed_mcp_send_debug_data(p_hwfn, p_ptt, + QED_MCP_DBG_DATA_TYPE_RAW, p_buf, size); +} diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.h b/drivers/net/ethernet/qlogic/qed/qed_mcp.h index 9c4c2763de8d..5750b4c5ef63 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_mcp.h +++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.h @@ -685,6 +685,18 @@ int qed_mcp_bist_nvm_get_image_att(struct qed_hwfn *p_hwfn, */ int qed_mfw_process_tlv_req(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt); +/** + * @brief Send raw debug data to the MFW + * + * @param p_hwfn + * @param p_ptt + * @param p_buf - raw debug data buffer + * @param size - buffer size + */ +int +qed_mcp_send_raw_debug_data(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, u8 *p_buf, u32 size); + /* Using hwfn number (and not pf_num) is required since in CMT mode, * same pf_num may be used by two different hwfn * TODO - this shouldn't really be in .h file, but until all fields @@ -731,6 +743,9 @@ struct qed_mcp_info { /* Capabilties negotiated with the MFW */ u32 capabilities; + + /* S/N for debug data mailbox commands */ + atomic_t dbg_data_seq; }; struct qed_mcp_mb_params { @@ -1001,6 +1016,19 @@ int __qed_configure_pf_min_bandwidth(struct qed_hwfn *p_hwfn, int qed_mcp_mask_parities(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt, u32 mask_parities); +/* @brief - Gets the mdump retained data from the MFW. + * + * @param p_hwfn + * @param p_ptt + * @param p_mdump_retain + * + * @param return 0 upon success. + */ +int +qed_mcp_mdump_get_retain(struct qed_hwfn *p_hwfn, + struct qed_ptt *p_ptt, + struct mdump_retain_data_stc *p_mdump_retain); + /** * @brief - Sets the MFW's max value for the given resource * diff --git a/drivers/net/ethernet/qlogic/qed/qed_rdma.c b/drivers/net/ethernet/qlogic/qed/qed_rdma.c index 5dc18a4bdda4..19c0c8864da1 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_rdma.c +++ b/drivers/net/ethernet/qlogic/qed/qed_rdma.c @@ -212,13 +212,22 @@ static int qed_rdma_alloc(struct qed_hwfn *p_hwfn) goto free_rdma_port; } + /* Allocate bit map for XRC Domains */ + rc = qed_rdma_bmap_alloc(p_hwfn, &p_rdma_info->xrcd_map, + QED_RDMA_MAX_XRCDS, "XRCD"); + if (rc) { + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, + "Failed to allocate xrcd_map,rc = %d\n", rc); + goto free_pd_map; + } + /* Allocate DPI bitmap */ rc = qed_rdma_bmap_alloc(p_hwfn, &p_rdma_info->dpi_map, p_hwfn->dpi_count, "DPI"); if (rc) { DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "Failed to allocate DPI bitmap, rc = %d\n", rc); - goto free_pd_map; + goto free_xrcd_map; } /* Allocate bitmap for cq's. The maximum number of CQs is bound to @@ -271,14 +280,27 @@ static int qed_rdma_alloc(struct qed_hwfn *p_hwfn) goto free_cid_map; } + /* The first SRQ follows the last XRC SRQ. This means that the + * SRQ IDs start from an offset equals to max_xrc_srqs. + */ + p_rdma_info->srq_id_offset = p_hwfn->p_cxt_mngr->xrc_srq_count; + rc = qed_rdma_bmap_alloc(p_hwfn, + &p_rdma_info->xrc_srq_map, + p_hwfn->p_cxt_mngr->xrc_srq_count, "XRC SRQ"); + if (rc) { + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, + "Failed to allocate xrc srq bitmap, rc = %d\n", rc); + goto free_real_cid_map; + } + /* Allocate bitmap for srqs */ - p_rdma_info->num_srqs = qed_cxt_get_srq_count(p_hwfn); + p_rdma_info->num_srqs = p_hwfn->p_cxt_mngr->srq_count; rc = qed_rdma_bmap_alloc(p_hwfn, &p_rdma_info->srq_map, p_rdma_info->num_srqs, "SRQ"); if (rc) { DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "Failed to allocate srq bitmap, rc = %d\n", rc); - goto free_real_cid_map; + goto free_xrc_srq_map; } if (QED_IS_IWARP_PERSONALITY(p_hwfn)) @@ -292,6 +314,8 @@ static int qed_rdma_alloc(struct qed_hwfn *p_hwfn) free_srq_map: kfree(p_rdma_info->srq_map.bitmap); +free_xrc_srq_map: + kfree(p_rdma_info->xrc_srq_map.bitmap); free_real_cid_map: kfree(p_rdma_info->real_cid_map.bitmap); free_cid_map: @@ -304,6 +328,8 @@ free_cq_map: kfree(p_rdma_info->cq_map.bitmap); free_dpi_map: kfree(p_rdma_info->dpi_map.bitmap); +free_xrcd_map: + kfree(p_rdma_info->xrcd_map.bitmap); free_pd_map: kfree(p_rdma_info->pd_map.bitmap); free_rdma_port: @@ -377,6 +403,7 @@ static void qed_rdma_resc_free(struct qed_hwfn *p_hwfn) qed_rdma_bmap_free(p_hwfn, &p_hwfn->p_rdma_info->tid_map, 1); qed_rdma_bmap_free(p_hwfn, &p_hwfn->p_rdma_info->srq_map, 1); qed_rdma_bmap_free(p_hwfn, &p_hwfn->p_rdma_info->real_cid_map, 1); + qed_rdma_bmap_free(p_hwfn, &p_hwfn->p_rdma_info->xrc_srq_map, 1); kfree(p_rdma_info->port); kfree(p_rdma_info->dev); @@ -611,7 +638,10 @@ static int qed_rdma_start_fw(struct qed_hwfn *p_hwfn, p_params_header->cnq_start_offset = (u8)RESC_START(p_hwfn, QED_RDMA_CNQ_RAM); p_params_header->num_cnqs = params->desired_cnq; - + p_params_header->first_reg_srq_id = + cpu_to_le16(p_hwfn->p_rdma_info->srq_id_offset); + p_params_header->reg_srq_base_addr = + cpu_to_le32(qed_cxt_get_ilt_page_size(p_hwfn, ILT_CLI_TSDM)); if (params->cq_mode == QED_RDMA_CQ_MODE_16_BITS) p_params_header->cq_ring_mode = 1; else @@ -982,6 +1012,41 @@ static void qed_rdma_free_pd(void *rdma_cxt, u16 pd) spin_unlock_bh(&p_hwfn->p_rdma_info->lock); } +static int qed_rdma_alloc_xrcd(void *rdma_cxt, u16 *xrcd_id) +{ + struct qed_hwfn *p_hwfn = (struct qed_hwfn *)rdma_cxt; + u32 returned_id; + int rc; + + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "Alloc XRCD\n"); + + spin_lock_bh(&p_hwfn->p_rdma_info->lock); + rc = qed_rdma_bmap_alloc_id(p_hwfn, + &p_hwfn->p_rdma_info->xrcd_map, + &returned_id); + spin_unlock_bh(&p_hwfn->p_rdma_info->lock); + if (rc) { + DP_NOTICE(p_hwfn, "Failed in allocating xrcd id\n"); + return rc; + } + + *xrcd_id = (u16)returned_id; + + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "Alloc XRCD - done, rc = %d\n", rc); + return rc; +} + +static void qed_rdma_free_xrcd(void *rdma_cxt, u16 xrcd_id) +{ + struct qed_hwfn *p_hwfn = (struct qed_hwfn *)rdma_cxt; + + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "xrcd_id = %08x\n", xrcd_id); + + spin_lock_bh(&p_hwfn->p_rdma_info->lock); + qed_bmap_release_id(p_hwfn, &p_hwfn->p_rdma_info->xrcd_map, xrcd_id); + spin_unlock_bh(&p_hwfn->p_rdma_info->lock); +} + static enum qed_rdma_toggle_bit qed_rdma_toggle_bit_create_resize_cq(struct qed_hwfn *p_hwfn, u16 icid) { @@ -1305,11 +1370,14 @@ qed_rdma_create_qp(void *rdma_cxt, qp->resp_offloaded = false; qp->e2e_flow_control_en = qp->use_srq ? false : true; qp->stats_queue = in_params->stats_queue; + qp->qp_type = in_params->qp_type; + qp->xrcd_id = in_params->xrcd_id; if (QED_IS_IWARP_PERSONALITY(p_hwfn)) { rc = qed_iwarp_create_qp(p_hwfn, qp, out_params); qp->qpid = qp->icid; } else { + qp->edpm_mode = GET_FIELD(in_params->flags, QED_ROCE_EDPM_MODE); rc = qed_roce_alloc_cid(p_hwfn, &qp->icid); qp->qpid = ((0xFF << 16) | qp->icid); } @@ -1417,6 +1485,18 @@ static int qed_rdma_modify_qp(void *rdma_cxt, qp->cur_state); } + switch (qp->qp_type) { + case QED_RDMA_QP_TYPE_XRC_INI: + qp->has_req = 1; + break; + case QED_RDMA_QP_TYPE_XRC_TGT: + qp->has_resp = 1; + break; + default: + qp->has_req = 1; + qp->has_resp = 1; + } + if (QED_IS_IWARP_PERSONALITY(p_hwfn)) { enum qed_iwarp_qp_state new_state = qed_roce2iwarp_state(qp->cur_state); @@ -1656,6 +1736,15 @@ static void *qed_rdma_get_rdma_ctx(struct qed_dev *cdev) return QED_AFFIN_HWFN(cdev); } +static struct qed_bmap *qed_rdma_get_srq_bmap(struct qed_hwfn *p_hwfn, + bool is_xrc) +{ + if (is_xrc) + return &p_hwfn->p_rdma_info->xrc_srq_map; + + return &p_hwfn->p_rdma_info->srq_map; +} + static int qed_rdma_modify_srq(void *rdma_cxt, struct qed_rdma_modify_srq_in_params *in_params) { @@ -1685,8 +1774,8 @@ static int qed_rdma_modify_srq(void *rdma_cxt, if (rc) return rc; - DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "modified SRQ id = %x", - in_params->srq_id); + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "modified SRQ id = %x, is_xrc=%u\n", + in_params->srq_id, in_params->is_xrc); return rc; } @@ -1701,6 +1790,7 @@ qed_rdma_destroy_srq(void *rdma_cxt, struct qed_spq_entry *p_ent; struct qed_bmap *bmap; u16 opaque_fid; + u16 offset; int rc; opaque_fid = p_hwfn->hw_info.opaque_fid; @@ -1722,14 +1812,16 @@ qed_rdma_destroy_srq(void *rdma_cxt, if (rc) return rc; - bmap = &p_hwfn->p_rdma_info->srq_map; + bmap = qed_rdma_get_srq_bmap(p_hwfn, in_params->is_xrc); + offset = (in_params->is_xrc) ? 0 : p_hwfn->p_rdma_info->srq_id_offset; spin_lock_bh(&p_hwfn->p_rdma_info->lock); - qed_bmap_release_id(p_hwfn, bmap, in_params->srq_id); + qed_bmap_release_id(p_hwfn, bmap, in_params->srq_id - offset); spin_unlock_bh(&p_hwfn->p_rdma_info->lock); - DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "SRQ destroyed Id = %x", - in_params->srq_id); + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, + "XRC/SRQ destroyed Id = %x, is_xrc=%u\n", + in_params->srq_id, in_params->is_xrc); return rc; } @@ -1747,24 +1839,26 @@ qed_rdma_create_srq(void *rdma_cxt, u16 opaque_fid, srq_id; struct qed_bmap *bmap; u32 returned_id; + u16 offset; int rc; - bmap = &p_hwfn->p_rdma_info->srq_map; + bmap = qed_rdma_get_srq_bmap(p_hwfn, in_params->is_xrc); spin_lock_bh(&p_hwfn->p_rdma_info->lock); rc = qed_rdma_bmap_alloc_id(p_hwfn, bmap, &returned_id); spin_unlock_bh(&p_hwfn->p_rdma_info->lock); if (rc) { - DP_NOTICE(p_hwfn, "failed to allocate srq id\n"); + DP_NOTICE(p_hwfn, + "failed to allocate xrc/srq id (is_xrc=%u)\n", + in_params->is_xrc); return rc; } - elem_type = QED_ELEM_SRQ; + elem_type = (in_params->is_xrc) ? (QED_ELEM_XRC_SRQ) : (QED_ELEM_SRQ); rc = qed_cxt_dynamic_ilt_alloc(p_hwfn, elem_type, returned_id); if (rc) goto err; - /* returned id is no greater than u16 */ - srq_id = (u16)returned_id; + opaque_fid = p_hwfn->hw_info.opaque_fid; opaque_fid = p_hwfn->hw_info.opaque_fid; @@ -1781,20 +1875,34 @@ qed_rdma_create_srq(void *rdma_cxt, DMA_REGPAIR_LE(p_ramrod->pbl_base_addr, in_params->pbl_base_addr); p_ramrod->pages_in_srq_pbl = cpu_to_le16(in_params->num_pages); p_ramrod->pd_id = cpu_to_le16(in_params->pd_id); - p_ramrod->srq_id.srq_idx = cpu_to_le16(srq_id); p_ramrod->srq_id.opaque_fid = cpu_to_le16(opaque_fid); p_ramrod->page_size = cpu_to_le16(in_params->page_size); DMA_REGPAIR_LE(p_ramrod->producers_addr, in_params->prod_pair_addr); + offset = (in_params->is_xrc) ? 0 : p_hwfn->p_rdma_info->srq_id_offset; + srq_id = (u16)returned_id + offset; + p_ramrod->srq_id.srq_idx = cpu_to_le16(srq_id); + if (in_params->is_xrc) { + SET_FIELD(p_ramrod->flags, + RDMA_SRQ_CREATE_RAMROD_DATA_XRC_FLAG, 1); + SET_FIELD(p_ramrod->flags, + RDMA_SRQ_CREATE_RAMROD_DATA_RESERVED_KEY_EN, + in_params->reserved_key_en); + p_ramrod->xrc_srq_cq_cid = + cpu_to_le32((p_hwfn->hw_info.opaque_fid << 16) | + in_params->cq_cid); + p_ramrod->xrc_domain = cpu_to_le16(in_params->xrcd_id); + } rc = qed_spq_post(p_hwfn, p_ent, NULL); if (rc) goto err; out_params->srq_id = srq_id; - DP_VERBOSE(p_hwfn, QED_MSG_RDMA, - "SRQ created Id = %x\n", out_params->srq_id); - + DP_VERBOSE(p_hwfn, + QED_MSG_RDMA, + "XRC/SRQ created Id = %x (is_xrc=%u)\n", + out_params->srq_id, in_params->is_xrc); return rc; err: @@ -1960,6 +2068,8 @@ static const struct qed_rdma_ops qed_rdma_ops_pass = { .rdma_cnq_prod_update = &qed_rdma_cnq_prod_update, .rdma_alloc_pd = &qed_rdma_alloc_pd, .rdma_dealloc_pd = &qed_rdma_free_pd, + .rdma_alloc_xrcd = &qed_rdma_alloc_xrcd, + .rdma_dealloc_xrcd = &qed_rdma_free_xrcd, .rdma_create_cq = &qed_rdma_create_cq, .rdma_destroy_cq = &qed_rdma_destroy_cq, .rdma_create_qp = &qed_rdma_create_qp, diff --git a/drivers/net/ethernet/qlogic/qed/qed_rdma.h b/drivers/net/ethernet/qlogic/qed/qed_rdma.h index dfaa2f552627..1e69d5bb0a70 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_rdma.h +++ b/drivers/net/ethernet/qlogic/qed/qed_rdma.h @@ -62,6 +62,11 @@ #define QED_RDMA_MAX_CQE_32_BIT (0x7FFFFFFF - 1) #define QED_RDMA_MAX_CQE_16_BIT (0x7FFF - 1) +/* Up to 2^16 XRC Domains are supported, but the actual number of supported XRC + * SRQs is much smaller so there's no need to have that many domains. + */ +#define QED_RDMA_MAX_XRCDS (roundup_pow_of_two(RDMA_MAX_XRC_SRQS)) + enum qed_rdma_toggle_bit { QED_RDMA_TOGGLE_BIT_CLEAR = 0, QED_RDMA_TOGGLE_BIT_SET = 1 @@ -80,9 +85,11 @@ struct qed_rdma_info { struct qed_bmap cq_map; struct qed_bmap pd_map; + struct qed_bmap xrcd_map; struct qed_bmap tid_map; struct qed_bmap qp_map; struct qed_bmap srq_map; + struct qed_bmap xrc_srq_map; struct qed_bmap cid_map; struct qed_bmap tcp_cid_map; struct qed_bmap real_cid_map; @@ -110,6 +117,7 @@ struct qed_rdma_qp { u32 qpid; u16 icid; enum qed_roce_qp_state cur_state; + enum qed_rdma_qp_type qp_type; enum qed_iwarp_qp_state iwarp_state; bool use_srq; bool signal_all; @@ -152,18 +160,21 @@ struct qed_rdma_qp { dma_addr_t orq_phys_addr; u8 orq_num_pages; bool req_offloaded; + bool has_req; /* responder */ u8 max_rd_atomic_resp; u32 rq_psn; u16 rq_cq_id; u16 rq_num_pages; + u16 xrcd_id; dma_addr_t rq_pbl_ptr; void *irq; dma_addr_t irq_phys_addr; u8 irq_num_pages; bool resp_offloaded; u32 cq_prod; + bool has_resp; u8 remote_mac_addr[6]; u8 local_mac_addr[6]; @@ -171,8 +182,17 @@ struct qed_rdma_qp { void *shared_queue; dma_addr_t shared_queue_phys_addr; struct qed_iwarp_ep *ep; + u8 edpm_mode; }; +static inline bool qed_rdma_is_xrc_qp(struct qed_rdma_qp *qp) +{ + if (qp->qp_type == QED_RDMA_QP_TYPE_XRC_TGT || + qp->qp_type == QED_RDMA_QP_TYPE_XRC_INI) + return true; + + return false; +} #if IS_ENABLED(CONFIG_QED_RDMA) void qed_rdma_dpm_bar(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt); void qed_rdma_dpm_conf(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt); diff --git a/drivers/net/ethernet/qlogic/qed/qed_roce.c b/drivers/net/ethernet/qlogic/qed/qed_roce.c index 37e70562a964..4566815f7b87 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_roce.c +++ b/drivers/net/ethernet/qlogic/qed/qed_roce.c @@ -254,6 +254,9 @@ static int qed_roce_sp_create_responder(struct qed_hwfn *p_hwfn, int rc; u8 tc; + if (!qp->has_resp) + return 0; + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "icid = %08x\n", qp->icid); /* Allocate DMA-able memory for IRQ */ @@ -315,6 +318,10 @@ static int qed_roce_sp_create_responder(struct qed_hwfn *p_hwfn, ROCE_CREATE_QP_RESP_RAMROD_DATA_MIN_RNR_NAK_TIMER, qp->min_rnr_nak_timer); + SET_FIELD(p_ramrod->flags, + ROCE_CREATE_QP_RESP_RAMROD_DATA_XRC_FLAG, + qed_rdma_is_xrc_qp(qp)); + p_ramrod->max_ird = qp->max_rd_atomic_resp; p_ramrod->traffic_class = qp->traffic_class_tos; p_ramrod->hop_limit = qp->hop_limit_ttl; @@ -335,6 +342,7 @@ static int qed_roce_sp_create_responder(struct qed_hwfn *p_hwfn, p_ramrod->qp_handle_for_cqe.lo = cpu_to_le32(qp->qp_handle.lo); p_ramrod->cq_cid = cpu_to_le32((p_hwfn->hw_info.opaque_fid << 16) | qp->rq_cq_id); + p_ramrod->xrc_domain = cpu_to_le16(qp->xrcd_id); tc = qed_roce_get_qp_tc(p_hwfn, qp); regular_latency_queue = qed_get_cm_pq_idx_ofld_mtc(p_hwfn, tc); @@ -395,6 +403,9 @@ static int qed_roce_sp_create_requester(struct qed_hwfn *p_hwfn, int rc; u8 tc; + if (!qp->has_req) + return 0; + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "icid = %08x\n", qp->icid); /* Allocate DMA-able memory for ORQ */ @@ -444,6 +455,13 @@ static int qed_roce_sp_create_requester(struct qed_hwfn *p_hwfn, ROCE_CREATE_QP_REQ_RAMROD_DATA_RNR_NAK_CNT, qp->rnr_retry_cnt); + SET_FIELD(p_ramrod->flags, + ROCE_CREATE_QP_REQ_RAMROD_DATA_XRC_FLAG, + qed_rdma_is_xrc_qp(qp)); + + SET_FIELD(p_ramrod->flags2, + ROCE_CREATE_QP_REQ_RAMROD_DATA_EDPM_MODE, qp->edpm_mode); + p_ramrod->max_ord = qp->max_rd_atomic_req; p_ramrod->traffic_class = qp->traffic_class_tos; p_ramrod->hop_limit = qp->hop_limit_ttl; @@ -517,6 +535,9 @@ static int qed_roce_sp_modify_responder(struct qed_hwfn *p_hwfn, struct qed_spq_entry *p_ent; int rc; + if (!qp->has_resp) + return 0; + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "icid = %08x\n", qp->icid); if (move_to_err && !qp->resp_offloaded) @@ -611,6 +632,9 @@ static int qed_roce_sp_modify_requester(struct qed_hwfn *p_hwfn, struct qed_spq_entry *p_ent; int rc; + if (!qp->has_req) + return 0; + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "icid = %08x\n", qp->icid); if (move_to_err && !(qp->req_offloaded)) @@ -705,6 +729,11 @@ static int qed_roce_sp_destroy_qp_responder(struct qed_hwfn *p_hwfn, dma_addr_t ramrod_res_phys; int rc; + if (!qp->has_resp) { + *cq_prod = 0; + return 0; + } + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "icid = %08x\n", qp->icid); *cq_prod = qp->cq_prod; @@ -736,9 +765,9 @@ static int qed_roce_sp_destroy_qp_responder(struct qed_hwfn *p_hwfn, p_ramrod = &p_ent->ramrod.roce_destroy_qp_resp; - p_ramrod_res = (struct roce_destroy_qp_resp_output_params *) - dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, sizeof(*p_ramrod_res), - &ramrod_res_phys, GFP_KERNEL); + p_ramrod_res = dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, + sizeof(*p_ramrod_res), + &ramrod_res_phys, GFP_KERNEL); if (!p_ramrod_res) { rc = -ENOMEM; @@ -785,6 +814,9 @@ static int qed_roce_sp_destroy_qp_requester(struct qed_hwfn *p_hwfn, dma_addr_t ramrod_res_phys; int rc = -ENOMEM; + if (!qp->has_req) + return 0; + DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "icid = %08x\n", qp->icid); if (!qp->req_offloaded) @@ -872,10 +904,10 @@ int qed_roce_query_qp(struct qed_hwfn *p_hwfn, } /* Send a query responder ramrod to FW to get RQ-PSN and state */ - p_resp_ramrod_res = (struct roce_query_qp_resp_output_params *) - dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, - sizeof(*p_resp_ramrod_res), - &resp_ramrod_res_phys, GFP_KERNEL); + p_resp_ramrod_res = + dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, + sizeof(*p_resp_ramrod_res), + &resp_ramrod_res_phys, GFP_KERNEL); if (!p_resp_ramrod_res) { DP_NOTICE(p_hwfn, "qed query qp failed: cannot allocate memory (ramrod)\n"); @@ -920,8 +952,7 @@ int qed_roce_query_qp(struct qed_hwfn *p_hwfn, } /* Send a query requester ramrod to FW to get SQ-PSN and state */ - p_req_ramrod_res = (struct roce_query_qp_req_output_params *) - dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, + p_req_ramrod_res = dma_alloc_coherent(&p_hwfn->cdev->pdev->dev, sizeof(*p_req_ramrod_res), &req_ramrod_res_phys, GFP_KERNEL); diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c index f5f3c03b9dd2..790c28d696a0 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_spq.c +++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c @@ -160,12 +160,16 @@ static int qed_spq_block(struct qed_hwfn *p_hwfn, return 0; } err: - DP_NOTICE(p_hwfn, - "Ramrod is stuck [CID %08x cmd %02x protocol %02x echo %04x]\n", - le32_to_cpu(p_ent->elem.hdr.cid), - p_ent->elem.hdr.cmd_id, - p_ent->elem.hdr.protocol_id, - le16_to_cpu(p_ent->elem.hdr.echo)); + p_ptt = qed_ptt_acquire(p_hwfn); + if (!p_ptt) + return -EBUSY; + qed_hw_err_notify(p_hwfn, p_ptt, QED_HW_ERR_RAMROD_FAIL, + "Ramrod is stuck [CID %08x cmd %02x protocol %02x echo %04x]\n", + le32_to_cpu(p_ent->elem.hdr.cid), + p_ent->elem.hdr.cmd_id, + p_ent->elem.hdr.protocol_id, + le16_to_cpu(p_ent->elem.hdr.echo)); + qed_ptt_release(p_hwfn, p_ptt); return -EBUSY; } diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.h b/drivers/net/ethernet/qlogic/qed/qed_sriov.h index 368e88565783..aabeaf03135e 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.h +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.h @@ -32,6 +32,7 @@ #ifndef _QED_SRIOV_H #define _QED_SRIOV_H +#include <linux/crash_dump.h> #include <linux/types.h> #include "qed_vf.h" @@ -40,9 +41,12 @@ #define QED_VF_ARRAY_LENGTH (3) #ifdef CONFIG_QED_SRIOV -#define IS_VF(cdev) ((cdev)->b_is_vf) -#define IS_PF(cdev) (!((cdev)->b_is_vf)) -#define IS_PF_SRIOV(p_hwfn) (!!((p_hwfn)->cdev->p_iov_info)) +#define IS_VF(cdev) (is_kdump_kernel() ? \ + (0) : ((cdev)->b_is_vf)) +#define IS_PF(cdev) (is_kdump_kernel() ? \ + (1) : !((cdev)->b_is_vf)) +#define IS_PF_SRIOV(p_hwfn) (is_kdump_kernel() ? \ + (0) : !!((p_hwfn)->cdev->p_iov_info)) #else #define IS_VF(cdev) (0) #define IS_PF(cdev) (1) diff --git a/drivers/net/ethernet/qlogic/qede/qede.h b/drivers/net/ethernet/qlogic/qede/qede.h index 234c6f30effb..8857da1208d7 100644 --- a/drivers/net/ethernet/qlogic/qede/qede.h +++ b/drivers/net/ethernet/qlogic/qede/qede.h @@ -278,6 +278,14 @@ struct qede_dev { struct qede_rdma_dev rdma_info; struct bpf_prog *xdp_prog; + + unsigned long err_flags; +#define QEDE_ERR_IS_HANDLED 31 +#define QEDE_ERR_ATTN_CLR_EN 0 +#define QEDE_ERR_GET_DBG_INFO 1 +#define QEDE_ERR_IS_RECOVERABLE 2 +#define QEDE_ERR_WARN 3 + struct qede_dump_info dump_info; }; @@ -485,11 +493,15 @@ struct qede_fastpath { #define QEDE_SP_RECOVERY 0 #define QEDE_SP_RX_MODE 1 +#define QEDE_SP_RSVD1 2 +#define QEDE_SP_RSVD2 3 +#define QEDE_SP_HW_ERR 4 +#define QEDE_SP_ARFS_CONFIG 5 +#define QEDE_SP_AER 7 #ifdef CONFIG_RFS_ACCEL int qede_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb, u16 rxq_index, u32 flow_id); -#define QEDE_SP_ARFS_CONFIG 4 #define QEDE_SP_TASK_POLL_DELAY (5 * HZ) #endif @@ -521,7 +533,6 @@ u16 qede_select_queue(struct net_device *dev, struct sk_buff *skb, netdev_features_t qede_features_check(struct sk_buff *skb, struct net_device *dev, netdev_features_t features); -void qede_tx_log_print(struct qede_dev *edev, struct qede_fastpath *fp); int qede_alloc_rx_buffer(struct qede_rx_queue *rxq, bool allow_lazy); int qede_free_tx_pkt(struct qede_dev *edev, struct qede_tx_queue *txq, int *len); @@ -574,12 +585,14 @@ int qede_add_tc_flower_fltr(struct qede_dev *edev, __be16 proto, #define RX_RING_SIZE ((u16)BIT(RX_RING_SIZE_POW)) #define NUM_RX_BDS_MAX (RX_RING_SIZE - 1) #define NUM_RX_BDS_MIN 128 +#define NUM_RX_BDS_KDUMP_MIN 63 #define NUM_RX_BDS_DEF ((u16)BIT(10) - 1) #define TX_RING_SIZE_POW 13 #define TX_RING_SIZE ((u16)BIT(TX_RING_SIZE_POW)) #define NUM_TX_BDS_MAX (TX_RING_SIZE - 1) #define NUM_TX_BDS_MIN 128 +#define NUM_TX_BDS_KDUMP_MIN 63 #define NUM_TX_BDS_DEF NUM_TX_BDS_MAX #define QEDE_MIN_PKT_LEN 64 diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c index 812c7766e096..24cc68391ac4 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c +++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c @@ -190,12 +190,14 @@ static const struct { enum { QEDE_PRI_FLAG_CMT, QEDE_PRI_FLAG_SMART_AN_SUPPORT, /* MFW supports SmartAN */ + QEDE_PRI_FLAG_RECOVER_ON_ERROR, QEDE_PRI_FLAG_LEN, }; static const char qede_private_arr[QEDE_PRI_FLAG_LEN][ETH_GSTRING_LEN] = { "Coupled-Function", "SmartAN capable", + "Recover on error", }; enum qede_ethtool_tests { @@ -417,9 +419,30 @@ static u32 qede_get_priv_flags(struct net_device *dev) if (edev->dev_info.common.smart_an) flags |= BIT(QEDE_PRI_FLAG_SMART_AN_SUPPORT); + if (edev->err_flags & BIT(QEDE_ERR_IS_RECOVERABLE)) + flags |= BIT(QEDE_PRI_FLAG_RECOVER_ON_ERROR); + return flags; } +static int qede_set_priv_flags(struct net_device *dev, u32 flags) +{ + struct qede_dev *edev = netdev_priv(dev); + u32 cflags = qede_get_priv_flags(dev); + u32 dflags = flags ^ cflags; + + /* can only change RECOVER_ON_ERROR flag */ + if (dflags & ~BIT(QEDE_PRI_FLAG_RECOVER_ON_ERROR)) + return -EINVAL; + + if (flags & BIT(QEDE_PRI_FLAG_RECOVER_ON_ERROR)) + set_bit(QEDE_ERR_IS_RECOVERABLE, &edev->err_flags); + else + clear_bit(QEDE_ERR_IS_RECOVERABLE, &edev->err_flags); + + return 0; +} + struct qede_link_mode_mapping { u32 qed_link_mode; u32 ethtool_link_mode; @@ -2098,6 +2121,7 @@ static const struct ethtool_ops qede_ethtool_ops = { .set_phys_id = qede_set_phys_id, .get_ethtool_stats = qede_get_ethtool_stats, .get_priv_flags = qede_get_priv_flags, + .set_priv_flags = qede_set_priv_flags, .get_sset_count = qede_get_sset_count, .get_rxnfc = qede_get_rxnfc, .set_rxnfc = qede_set_rxnfc, diff --git a/drivers/net/ethernet/qlogic/qede/qede_fp.c b/drivers/net/ethernet/qlogic/qede/qede_fp.c index c6c20776b474..7598ebe0962a 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_fp.c +++ b/drivers/net/ethernet/qlogic/qede/qede_fp.c @@ -1066,6 +1066,7 @@ static bool qede_rx_xdp(struct qede_dev *edev, xdp_set_data_meta_invalid(&xdp); xdp.data_end = xdp.data + *len; xdp.rxq = &rxq->xdp_rxq; + xdp.frame_sz = rxq->rx_buf_seg_size; /* PAGE_SIZE when XDP enabled */ /* Queues always have a full reset currently, so for the time * being until there's atomic program replace just mark read diff --git a/drivers/net/ethernet/qlogic/qede/qede_main.c b/drivers/net/ethernet/qlogic/qede/qede_main.c index 34fa3917eb33..b2d154258b07 100644 --- a/drivers/net/ethernet/qlogic/qede/qede_main.c +++ b/drivers/net/ethernet/qlogic/qede/qede_main.c @@ -29,6 +29,7 @@ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. */ +#include <linux/crash_dump.h> #include <linux/module.h> #include <linux/pci.h> #include <linux/version.h> @@ -60,6 +61,7 @@ #include <net/ip6_checksum.h> #include <linux/bitops.h> #include <linux/vmalloc.h> +#include <linux/aer.h> #include "qede.h" #include "qede_ptp.h" @@ -124,6 +126,8 @@ static const struct pci_device_id qede_pci_tbl[] = { MODULE_DEVICE_TABLE(pci, qede_pci_tbl); static int qede_probe(struct pci_dev *pdev, const struct pci_device_id *id); +static pci_ers_result_t +qede_io_error_detected(struct pci_dev *pdev, pci_channel_state_t state); #define TX_TIMEOUT (5 * HZ) @@ -135,10 +139,12 @@ static void qede_shutdown(struct pci_dev *pdev); static void qede_link_update(void *dev, struct qed_link_output *link); static void qede_schedule_recovery_handler(void *dev); static void qede_recovery_handler(struct qede_dev *edev); +static void qede_schedule_hw_err_handler(void *dev, + enum qed_hw_err_type err_type); static void qede_get_eth_tlv_data(void *edev, void *data); static void qede_get_generic_tlv_data(void *edev, struct qed_generic_tlvs *data); - +static void qede_generic_hw_err_handler(struct qede_dev *edev); #ifdef CONFIG_QED_SRIOV static int qede_set_vf_vlan(struct net_device *ndev, int vf, u16 vlan, u8 qos, __be16 vlan_proto) @@ -203,6 +209,10 @@ static int qede_sriov_configure(struct pci_dev *pdev, int num_vfs_param) } #endif +static const struct pci_error_handlers qede_err_handler = { + .error_detected = qede_io_error_detected, +}; + static struct pci_driver qede_pci_driver = { .name = "qede", .id_table = qede_pci_tbl, @@ -212,6 +222,7 @@ static struct pci_driver qede_pci_driver = { #ifdef CONFIG_QED_SRIOV .sriov_configure = qede_sriov_configure, #endif + .err_handler = &qede_err_handler, }; static struct qed_eth_cb_ops qede_ll_ops = { @@ -221,6 +232,7 @@ static struct qed_eth_cb_ops qede_ll_ops = { #endif .link_update = qede_link_update, .schedule_recovery_handler = qede_schedule_recovery_handler, + .schedule_hw_err_handler = qede_schedule_hw_err_handler, .get_generic_tlv_data = qede_get_generic_tlv_data, .get_protocol_tlv_data = qede_get_eth_tlv_data, }, @@ -527,6 +539,51 @@ static int qede_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) return 0; } +static void qede_tx_log_print(struct qede_dev *edev, struct qede_tx_queue *txq) +{ + DP_NOTICE(edev, + "Txq[%d]: FW cons [host] %04x, SW cons %04x, SW prod %04x [Jiffies %lu]\n", + txq->index, le16_to_cpu(*txq->hw_cons_ptr), + qed_chain_get_cons_idx(&txq->tx_pbl), + qed_chain_get_prod_idx(&txq->tx_pbl), + jiffies); +} + +static void qede_tx_timeout(struct net_device *dev, unsigned int txqueue) +{ + struct qede_dev *edev = netdev_priv(dev); + struct qede_tx_queue *txq; + int cos; + + netif_carrier_off(dev); + DP_NOTICE(edev, "TX timeout on queue %u!\n", txqueue); + + if (!(edev->fp_array[txqueue].type & QEDE_FASTPATH_TX)) + return; + + for_each_cos_in_txq(edev, cos) { + txq = &edev->fp_array[txqueue].txq[cos]; + + if (qed_chain_get_cons_idx(&txq->tx_pbl) != + qed_chain_get_prod_idx(&txq->tx_pbl)) + qede_tx_log_print(edev, txq); + } + + if (IS_VF(edev)) + return; + + if (test_and_set_bit(QEDE_ERR_IS_HANDLED, &edev->err_flags) || + edev->state == QEDE_STATE_RECOVERY) { + DP_INFO(edev, + "Avoid handling a Tx timeout while another HW error is being handled\n"); + return; + } + + set_bit(QEDE_ERR_GET_DBG_INFO, &edev->err_flags); + set_bit(QEDE_SP_HW_ERR, &edev->sp_flags); + schedule_delayed_work(&edev->sp_task, 0); +} + static int qede_setup_tc(struct net_device *ndev, u8 num_tc) { struct qede_dev *edev = netdev_priv(ndev); @@ -614,6 +671,7 @@ static const struct net_device_ops qede_netdev_ops = { .ndo_validate_addr = eth_validate_addr, .ndo_change_mtu = qede_change_mtu, .ndo_do_ioctl = qede_ioctl, + .ndo_tx_timeout = qede_tx_timeout, #ifdef CONFIG_QED_SRIOV .ndo_set_vf_mac = qede_set_vf_mac, .ndo_set_vf_vlan = qede_set_vf_vlan, @@ -707,8 +765,14 @@ static struct qede_dev *qede_alloc_etherdev(struct qed_dev *cdev, edev->dp_module = dp_module; edev->dp_level = dp_level; edev->ops = qed_ops; - edev->q_num_rx_buffers = NUM_RX_BDS_DEF; - edev->q_num_tx_buffers = NUM_TX_BDS_DEF; + + if (is_kdump_kernel()) { + edev->q_num_rx_buffers = NUM_RX_BDS_KDUMP_MIN; + edev->q_num_tx_buffers = NUM_TX_BDS_KDUMP_MIN; + } else { + edev->q_num_rx_buffers = NUM_RX_BDS_DEF; + edev->q_num_tx_buffers = NUM_TX_BDS_DEF; + } DP_INFO(edev, "Allocated netdev with %d tx queues and %d rx queues\n", info->num_queues, info->num_queues); @@ -974,7 +1038,8 @@ static void qede_sp_task(struct work_struct *work) /* SRIOV must be disabled outside the lock to avoid a deadlock. * The recovery of the active VFs is currently not supported. */ - qede_sriov_configure(edev->pdev, 0); + if (pci_num_vf(edev->pdev)) + qede_sriov_configure(edev->pdev, 0); #endif qede_lock(edev); qede_recovery_handler(edev); @@ -993,7 +1058,20 @@ static void qede_sp_task(struct work_struct *work) qede_process_arfs_filters(edev, false); } #endif + if (test_and_clear_bit(QEDE_SP_HW_ERR, &edev->sp_flags)) + qede_generic_hw_err_handler(edev); __qede_unlock(edev); + + if (test_and_clear_bit(QEDE_SP_AER, &edev->sp_flags)) { +#ifdef CONFIG_QED_SRIOV + /* SRIOV must be disabled outside the lock to avoid a deadlock. + * The recovery of the active VFs is currently not supported. + */ + if (pci_num_vf(edev->pdev)) + qede_sriov_configure(edev->pdev, 0); +#endif + edev->ops->common->recovery_process(edev->cdev); + } } static void qede_update_pf_params(struct qed_dev *cdev) @@ -1187,7 +1265,7 @@ static int qede_probe(struct pci_dev *pdev, const struct pci_device_id *id) case QEDE_PRIVATE_VF: if (debug & QED_LOG_VERBOSE_MASK) dev_err(&pdev->dev, "Probing a VF\n"); - is_vf = true; + is_vf = is_kdump_kernel() ? false : true; break; default: if (debug & QED_LOG_VERBOSE_MASK) @@ -1398,7 +1476,7 @@ static int qede_alloc_mem_rxq(struct qede_dev *edev, struct qede_rx_queue *rxq) if (rxq->rx_buf_size + size > PAGE_SIZE) rxq->rx_buf_size = PAGE_SIZE - size; - /* Segment size to spilt a page in multiple equal parts , + /* Segment size to split a page in multiple equal parts, * unless XDP is used in which case we'd use the entire page. */ if (!edev->xdp_prog) { @@ -1694,7 +1772,7 @@ static void qede_init_fp(struct qede_dev *edev) txq->ndev_txq_id = ndev_tx_id; if (edev->dev_info.is_legacy) - txq->is_legacy = 1; + txq->is_legacy = true; txq->dev = &edev->pdev->dev; } @@ -2482,6 +2560,100 @@ err: qede_recovery_failed(edev); } +static void qede_atomic_hw_err_handler(struct qede_dev *edev) +{ + struct qed_dev *cdev = edev->cdev; + + DP_NOTICE(edev, + "Generic non-sleepable HW error handling started - err_flags 0x%lx\n", + edev->err_flags); + + /* Get a call trace of the flow that led to the error */ + WARN_ON(test_bit(QEDE_ERR_WARN, &edev->err_flags)); + + /* Prevent HW attentions from being reasserted */ + if (test_bit(QEDE_ERR_ATTN_CLR_EN, &edev->err_flags)) + edev->ops->common->attn_clr_enable(cdev, true); + + DP_NOTICE(edev, "Generic non-sleepable HW error handling is done\n"); +} + +static void qede_generic_hw_err_handler(struct qede_dev *edev) +{ + struct qed_dev *cdev = edev->cdev; + + DP_NOTICE(edev, + "Generic sleepable HW error handling started - err_flags 0x%lx\n", + edev->err_flags); + + /* Trigger a recovery process. + * This is placed in the sleep requiring section just to make + * sure it is the last one, and that all the other operations + * were completed. + */ + if (test_bit(QEDE_ERR_IS_RECOVERABLE, &edev->err_flags)) + edev->ops->common->recovery_process(cdev); + + clear_bit(QEDE_ERR_IS_HANDLED, &edev->err_flags); + + DP_NOTICE(edev, "Generic sleepable HW error handling is done\n"); +} + +static void qede_set_hw_err_flags(struct qede_dev *edev, + enum qed_hw_err_type err_type) +{ + unsigned long err_flags = 0; + + switch (err_type) { + case QED_HW_ERR_DMAE_FAIL: + set_bit(QEDE_ERR_WARN, &err_flags); + fallthrough; + case QED_HW_ERR_MFW_RESP_FAIL: + case QED_HW_ERR_HW_ATTN: + case QED_HW_ERR_RAMROD_FAIL: + case QED_HW_ERR_FW_ASSERT: + set_bit(QEDE_ERR_ATTN_CLR_EN, &err_flags); + set_bit(QEDE_ERR_GET_DBG_INFO, &err_flags); + break; + + default: + DP_NOTICE(edev, "Unexpected HW error [%d]\n", err_type); + break; + } + + edev->err_flags |= err_flags; +} + +static void qede_schedule_hw_err_handler(void *dev, + enum qed_hw_err_type err_type) +{ + struct qede_dev *edev = dev; + + /* Fan failure cannot be masked by handling of another HW error or by a + * concurrent recovery process. + */ + if ((test_and_set_bit(QEDE_ERR_IS_HANDLED, &edev->err_flags) || + edev->state == QEDE_STATE_RECOVERY) && + err_type != QED_HW_ERR_FAN_FAIL) { + DP_INFO(edev, + "Avoid scheduling an error handling while another HW error is being handled\n"); + return; + } + + if (err_type >= QED_HW_ERR_LAST) { + DP_NOTICE(edev, "Unknown HW error [%d]\n", err_type); + clear_bit(QEDE_ERR_IS_HANDLED, &edev->err_flags); + return; + } + + qede_set_hw_err_flags(edev, err_type); + qede_atomic_hw_err_handler(edev); + set_bit(QEDE_SP_HW_ERR, &edev->sp_flags); + schedule_delayed_work(&edev->sp_task, 0); + + DP_INFO(edev, "Scheduled a error handler [err_type %d]\n", err_type); +} + static bool qede_is_txq_full(struct qede_dev *edev, struct qede_tx_queue *txq) { struct netdev_queue *netdev_txq; @@ -2579,3 +2751,49 @@ static void qede_get_eth_tlv_data(void *dev, void *data) etlv->num_txqs_full_set = true; etlv->num_rxqs_full_set = true; } + +/** + * qede_io_error_detected - called when PCI error is detected + * @pdev: Pointer to PCI device + * @state: The current pci connection state + * + * This function is called after a PCI bus error affecting + * this device has been detected. + */ +static pci_ers_result_t +qede_io_error_detected(struct pci_dev *pdev, pci_channel_state_t state) +{ + struct net_device *dev = pci_get_drvdata(pdev); + struct qede_dev *edev = netdev_priv(dev); + + if (!edev) + return PCI_ERS_RESULT_NONE; + + DP_NOTICE(edev, "IO error detected [%d]\n", state); + + __qede_lock(edev); + if (edev->state == QEDE_STATE_RECOVERY) { + DP_NOTICE(edev, "Device already in the recovery state\n"); + __qede_unlock(edev); + return PCI_ERS_RESULT_NONE; + } + + /* PF handles the recovery of its VFs */ + if (IS_VF(edev)) { + DP_VERBOSE(edev, QED_MSG_IOV, + "VF recovery is handled by its PF\n"); + __qede_unlock(edev); + return PCI_ERS_RESULT_RECOVERED; + } + + /* Close OS Tx */ + netif_tx_disable(edev->ndev); + netif_carrier_off(edev->ndev); + + set_bit(QEDE_SP_AER, &edev->sp_flags); + schedule_delayed_work(&edev->sp_task, 0); + + __qede_unlock(edev); + + return PCI_ERS_RESULT_CAN_RECOVER; +} diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h index 134611aa2c9a..d838774af5a6 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h @@ -1880,12 +1880,6 @@ static inline void qlcnic_write_crb(struct qlcnic_adapter *adapter, char *buf, adapter->ahw->hw_ops->write_crb(adapter, buf, offset, size); } -static inline int qlcnic_hw_write_wx_2M(struct qlcnic_adapter *adapter, - ulong off, u32 data) -{ - return adapter->ahw->hw_ops->write_reg(adapter, off, data); -} - static inline int qlcnic_get_mac_address(struct qlcnic_adapter *adapter, u8 *mac, u8 function) { diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c index 2a533280b124..29b9c728a65e 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_83xx_hw.c @@ -3651,7 +3651,7 @@ int qlcnic_83xx_interrupt_test(struct net_device *netdev) ahw->diag_cnt = 0; ret = qlcnic_alloc_mbx_args(&cmd, adapter, QLCNIC_CMD_INTRPT_TEST); if (ret) - goto fail_diag_irq; + goto fail_mbx_args; if (adapter->flags & QLCNIC_MSIX_ENABLED) intrpt_id = ahw->intr_tbl[0].id; @@ -3681,6 +3681,8 @@ int qlcnic_83xx_interrupt_test(struct net_device *netdev) done: qlcnic_free_mbx_args(&cmd); + +fail_mbx_args: qlcnic_83xx_diag_free_res(netdev, drv_sds_rings); fail_diag_irq: diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c index f7c2f32237cb..7adbb03cb931 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_common.c @@ -1582,10 +1582,10 @@ void qlcnic_sriov_vf_set_multi(struct net_device *netdev) if (mode == VPORT_MISS_MODE_ACCEPT_ALL && !adapter->fdb_mac_learn) { qlcnic_alloc_lb_filters_mem(adapter); - adapter->drv_mac_learn = 1; + adapter->drv_mac_learn = true; adapter->rx_mac_learn = true; } else { - adapter->drv_mac_learn = 0; + adapter->drv_mac_learn = false; adapter->rx_mac_learn = false; } } diff --git a/drivers/net/ethernet/qualcomm/emac/emac-mac.c b/drivers/net/ethernet/qualcomm/emac/emac-mac.c index 251d4ac4af02..117188e3c7de 100644 --- a/drivers/net/ethernet/qualcomm/emac/emac-mac.c +++ b/drivers/net/ethernet/qualcomm/emac/emac-mac.c @@ -1431,8 +1431,9 @@ error: } /* Transmit the packet using specified transmit queue */ -int emac_mac_tx_buf_send(struct emac_adapter *adpt, struct emac_tx_queue *tx_q, - struct sk_buff *skb) +netdev_tx_t emac_mac_tx_buf_send(struct emac_adapter *adpt, + struct emac_tx_queue *tx_q, + struct sk_buff *skb) { struct emac_tpd tpd; u32 prod_idx; diff --git a/drivers/net/ethernet/qualcomm/emac/emac-mac.h b/drivers/net/ethernet/qualcomm/emac/emac-mac.h index ae08bdd9046c..920123eb8ace 100644 --- a/drivers/net/ethernet/qualcomm/emac/emac-mac.h +++ b/drivers/net/ethernet/qualcomm/emac/emac-mac.h @@ -227,8 +227,9 @@ void emac_mac_stop(struct emac_adapter *adpt); void emac_mac_mode_config(struct emac_adapter *adpt); void emac_mac_rx_process(struct emac_adapter *adpt, struct emac_rx_queue *rx_q, int *num_pkts, int max_pkts); -int emac_mac_tx_buf_send(struct emac_adapter *adpt, struct emac_tx_queue *tx_q, - struct sk_buff *skb); +netdev_tx_t emac_mac_tx_buf_send(struct emac_adapter *adpt, + struct emac_tx_queue *tx_q, + struct sk_buff *skb); void emac_mac_tx_process(struct emac_adapter *adpt, struct emac_tx_queue *tx_q); void emac_mac_rx_tx_ring_init_all(struct platform_device *pdev, struct emac_adapter *adpt); diff --git a/drivers/net/ethernet/qualcomm/emac/emac.c b/drivers/net/ethernet/qualcomm/emac/emac.c index 18b0c7a2d6dc..20b1b43a0e39 100644 --- a/drivers/net/ethernet/qualcomm/emac/emac.c +++ b/drivers/net/ethernet/qualcomm/emac/emac.c @@ -115,7 +115,8 @@ static int emac_napi_rtx(struct napi_struct *napi, int budget) } /* Transmit the packet */ -static int emac_start_xmit(struct sk_buff *skb, struct net_device *netdev) +static netdev_tx_t emac_start_xmit(struct sk_buff *skb, + struct net_device *netdev) { struct emac_adapter *adpt = netdev_priv(netdev); diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index 60d342f82fb3..e291e6ac40cb 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -2054,10 +2054,9 @@ static void cp_remove_one (struct pci_dev *pdev) free_netdev(dev); } -#ifdef CONFIG_PM -static int cp_suspend (struct pci_dev *pdev, pm_message_t state) +static int __maybe_unused cp_suspend(struct device *device) { - struct net_device *dev = pci_get_drvdata(pdev); + struct net_device *dev = dev_get_drvdata(device); struct cp_private *cp = netdev_priv(dev); unsigned long flags; @@ -2075,16 +2074,14 @@ static int cp_suspend (struct pci_dev *pdev, pm_message_t state) spin_unlock_irqrestore (&cp->lock, flags); - pci_save_state(pdev); - pci_enable_wake(pdev, pci_choose_state(pdev, state), cp->wol_enabled); - pci_set_power_state(pdev, pci_choose_state(pdev, state)); + device_set_wakeup_enable(device, cp->wol_enabled); return 0; } -static int cp_resume (struct pci_dev *pdev) +static int __maybe_unused cp_resume(struct device *device) { - struct net_device *dev = pci_get_drvdata (pdev); + struct net_device *dev = dev_get_drvdata(device); struct cp_private *cp = netdev_priv(dev); unsigned long flags; @@ -2093,10 +2090,6 @@ static int cp_resume (struct pci_dev *pdev) netif_device_attach (dev); - pci_set_power_state(pdev, PCI_D0); - pci_restore_state(pdev); - pci_enable_wake(pdev, PCI_D0, 0); - /* FIXME: sh*t may happen if the Rx ring buffer is depleted */ cp_init_rings_index (cp); cp_init_hw (cp); @@ -2111,7 +2104,6 @@ static int cp_resume (struct pci_dev *pdev) return 0; } -#endif /* CONFIG_PM */ static const struct pci_device_id cp_pci_tbl[] = { { PCI_DEVICE(PCI_VENDOR_ID_REALTEK, PCI_DEVICE_ID_REALTEK_8139), }, @@ -2120,15 +2112,14 @@ static const struct pci_device_id cp_pci_tbl[] = { }; MODULE_DEVICE_TABLE(pci, cp_pci_tbl); +static SIMPLE_DEV_PM_OPS(cp_pm_ops, cp_suspend, cp_resume); + static struct pci_driver cp_driver = { .name = DRV_NAME, .id_table = cp_pci_tbl, .probe = cp_init_one, .remove = cp_remove_one, -#ifdef CONFIG_PM - .resume = cp_resume, - .suspend = cp_suspend, -#endif + .driver.pm = &cp_pm_ops, }; module_pci_driver(cp_driver); diff --git a/drivers/net/ethernet/realtek/8139too.c b/drivers/net/ethernet/realtek/8139too.c index 5caeb8368eab..227139d42227 100644 --- a/drivers/net/ethernet/realtek/8139too.c +++ b/drivers/net/ethernet/realtek/8139too.c @@ -2603,17 +2603,13 @@ static void rtl8139_set_rx_mode (struct net_device *dev) spin_unlock_irqrestore (&tp->lock, flags); } -#ifdef CONFIG_PM - -static int rtl8139_suspend (struct pci_dev *pdev, pm_message_t state) +static int __maybe_unused rtl8139_suspend(struct device *device) { - struct net_device *dev = pci_get_drvdata (pdev); + struct net_device *dev = dev_get_drvdata(device); struct rtl8139_private *tp = netdev_priv(dev); void __iomem *ioaddr = tp->mmio_addr; unsigned long flags; - pci_save_state (pdev); - if (!netif_running (dev)) return 0; @@ -2631,38 +2627,30 @@ static int rtl8139_suspend (struct pci_dev *pdev, pm_message_t state) spin_unlock_irqrestore (&tp->lock, flags); - pci_set_power_state (pdev, PCI_D3hot); - return 0; } - -static int rtl8139_resume (struct pci_dev *pdev) +static int __maybe_unused rtl8139_resume(struct device *device) { - struct net_device *dev = pci_get_drvdata (pdev); + struct net_device *dev = dev_get_drvdata(device); - pci_restore_state (pdev); if (!netif_running (dev)) return 0; - pci_set_power_state (pdev, PCI_D0); + rtl8139_init_ring (dev); rtl8139_hw_start (dev); netif_device_attach (dev); return 0; } -#endif /* CONFIG_PM */ - +static SIMPLE_DEV_PM_OPS(rtl8139_pm_ops, rtl8139_suspend, rtl8139_resume); static struct pci_driver rtl8139_pci_driver = { .name = DRV_NAME, .id_table = rtl8139_pci_tbl, .probe = rtl8139_init_one, .remove = rtl8139_remove_one, -#ifdef CONFIG_PM - .suspend = rtl8139_suspend, - .resume = rtl8139_resume, -#endif /* CONFIG_PM */ + .driver.pm = &rtl8139_pm_ops, }; diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index 78e15cc00e0a..4d2ec9742cee 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -10,7 +10,6 @@ */ #include <linux/module.h> -#include <linux/moduleparam.h> #include <linux/pci.h> #include <linux/netdevice.h> #include <linux/etherdevice.h> @@ -19,7 +18,6 @@ #include <linux/ethtool.h> #include <linux/phy.h> #include <linux/if_vlan.h> -#include <linux/crc32.h> #include <linux/in.h> #include <linux/io.h> #include <linux/ip.h> @@ -27,6 +25,7 @@ #include <linux/interrupt.h> #include <linux/dma-mapping.h> #include <linux/pm_runtime.h> +#include <linux/bitfield.h> #include <linux/prefetch.h> #include <linux/ipv6.h> #include <net/ip6_checksum.h> @@ -58,9 +57,6 @@ #define FIRMWARE_8107E_2 "rtl_nic/rtl8107e-2.fw" #define FIRMWARE_8125A_3 "rtl_nic/rtl8125a-3.fw" -#define R8169_MSG_DEFAULT \ - (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_IFUP | NETIF_MSG_IFDOWN) - /* Maximum number of multicast addresses to filter (vs. Rx-all-multicast). The RTL chips use a 64 element hash table based on the Ethernet CRC. */ #define MC_FILTER_LIMIT 32 @@ -75,6 +71,8 @@ #define R8169_TX_RING_BYTES (NUM_TX_DESC * sizeof(struct TxDesc)) #define R8169_RX_RING_BYTES (NUM_RX_DESC * sizeof(struct RxDesc)) +#define OCP_STD_PHY_BASE 0xa400 + #define RTL_CFG_NO_GBIT 1 /* write/read MMIO register */ @@ -85,10 +83,10 @@ #define RTL_R16(tp, reg) readw(tp->mmio_addr + (reg)) #define RTL_R32(tp, reg) readl(tp->mmio_addr + (reg)) -#define JUMBO_4K (4*1024 - ETH_HLEN - 2) -#define JUMBO_6K (6*1024 - ETH_HLEN - 2) -#define JUMBO_7K (7*1024 - ETH_HLEN - 2) -#define JUMBO_9K (9*1024 - ETH_HLEN - 2) +#define JUMBO_4K (4 * SZ_1K - VLAN_ETH_HLEN - ETH_FCS_LEN) +#define JUMBO_6K (6 * SZ_1K - VLAN_ETH_HLEN - ETH_FCS_LEN) +#define JUMBO_7K (7 * SZ_1K - VLAN_ETH_HLEN - ETH_FCS_LEN) +#define JUMBO_9K (9 * SZ_1K - VLAN_ETH_HLEN - ETH_FCS_LEN) static const struct { const char *name; @@ -176,10 +174,6 @@ static const struct pci_device_id rtl8169_pci_tbl[] = { MODULE_DEVICE_TABLE(pci, rtl8169_pci_tbl); -static struct { - u32 msg_enable; -} debug = { -1 }; - enum rtl_registers { MAC0 = 0, /* Ethernet hardware address. */ MAC4 = 4, @@ -227,10 +221,13 @@ enum rtl_registers { CPlusCmd = 0xe0, IntrMitigate = 0xe2, -#define RTL_COALESCE_MASK 0x0f -#define RTL_COALESCE_SHIFT 4 -#define RTL_COALESCE_T_MAX (RTL_COALESCE_MASK) -#define RTL_COALESCE_FRAME_MAX (RTL_COALESCE_MASK << 2) +#define RTL_COALESCE_TX_USECS GENMASK(15, 12) +#define RTL_COALESCE_TX_FRAMES GENMASK(11, 8) +#define RTL_COALESCE_RX_USECS GENMASK(7, 4) +#define RTL_COALESCE_RX_FRAMES GENMASK(3, 0) + +#define RTL_COALESCE_T_MAX 0x0fU +#define RTL_COALESCE_FRAME_MAX (RTL_COALESCE_T_MAX * 4) RxDescAddrLow = 0xe4, RxDescAddrHigh = 0xe8, @@ -386,10 +383,12 @@ enum rtl_register_content { /* rx_mode_bits */ AcceptErr = 0x20, AcceptRunt = 0x10, +#define RX_CONFIG_ACCEPT_ERR_MASK 0x30 AcceptBroadcast = 0x08, AcceptMulticast = 0x04, AcceptMyPhys = 0x02, AcceptAllPhys = 0x01, +#define RX_CONFIG_ACCEPT_OK_MASK 0x0f #define RX_CONFIG_ACCEPT_MASK 0x3f /* TxConfigBits */ @@ -596,7 +595,6 @@ struct rtl8169_private { struct net_device *dev; struct phy_device *phydev; struct napi_struct napi; - u32 msg_enable; enum mac_version mac_version; u32 cur_rx; /* Index into the Rx descriptor buffer of next Rx pkt. */ u32 cur_tx; /* Index into the Tx descriptor buffer of next Rx pkt. */ @@ -638,8 +636,6 @@ typedef void (*rtl_generic_fct)(struct rtl8169_private *tp); MODULE_AUTHOR("Realtek and the Linux r8169 crew <netdev@vger.kernel.org>"); MODULE_DESCRIPTION("RealTek RTL-8169 Gigabit Ethernet driver"); -module_param_named(debug, debug.msg_enable, int, 0); -MODULE_PARM_DESC(debug, "Debug verbosity level (0=none, ..., 16=all)"); MODULE_SOFTDEP("pre: realtek"); MODULE_LICENSE("GPL"); MODULE_FIRMWARE(FIRMWARE_8168D_1); @@ -727,53 +723,35 @@ struct rtl_cond { const char *msg; }; -static void rtl_udelay(unsigned int d) -{ - udelay(d); -} - static bool rtl_loop_wait(struct rtl8169_private *tp, const struct rtl_cond *c, - void (*delay)(unsigned int), unsigned int d, int n, - bool high) + unsigned long usecs, int n, bool high) { int i; for (i = 0; i < n; i++) { if (c->check(tp) == high) return true; - delay(d); + fsleep(usecs); } - netif_err(tp, drv, tp->dev, "%s == %d (loop: %d, delay: %d).\n", - c->msg, !high, n, d); - return false; -} -static bool rtl_udelay_loop_wait_high(struct rtl8169_private *tp, - const struct rtl_cond *c, - unsigned int d, int n) -{ - return rtl_loop_wait(tp, c, rtl_udelay, d, n, true); -} - -static bool rtl_udelay_loop_wait_low(struct rtl8169_private *tp, - const struct rtl_cond *c, - unsigned int d, int n) -{ - return rtl_loop_wait(tp, c, rtl_udelay, d, n, false); + if (net_ratelimit()) + netdev_err(tp->dev, "%s == %d (loop: %d, delay: %lu).\n", + c->msg, !high, n, usecs); + return false; } -static bool rtl_msleep_loop_wait_high(struct rtl8169_private *tp, - const struct rtl_cond *c, - unsigned int d, int n) +static bool rtl_loop_wait_high(struct rtl8169_private *tp, + const struct rtl_cond *c, + unsigned long d, int n) { - return rtl_loop_wait(tp, c, msleep, d, n, true); + return rtl_loop_wait(tp, c, d, n, true); } -static bool rtl_msleep_loop_wait_low(struct rtl8169_private *tp, - const struct rtl_cond *c, - unsigned int d, int n) +static bool rtl_loop_wait_low(struct rtl8169_private *tp, + const struct rtl_cond *c, + unsigned long d, int n) { - return rtl_loop_wait(tp, c, msleep, d, n, false); + return rtl_loop_wait(tp, c, d, n, false); } #define DECLARE_RTL_COND(name) \ @@ -789,7 +767,8 @@ static bool name ## _check(struct rtl8169_private *tp) static bool rtl_ocp_reg_failure(struct rtl8169_private *tp, u32 reg) { if (reg & 0xffff0001) { - netif_err(tp, drv, tp->dev, "Invalid ocp reg %x!\n", reg); + if (net_ratelimit()) + netdev_err(tp->dev, "Invalid ocp reg %x!\n", reg); return true; } return false; @@ -807,7 +786,7 @@ static void r8168_phy_ocp_write(struct rtl8169_private *tp, u32 reg, u32 data) RTL_W32(tp, GPHY_OCP, OCPAR_FLAG | (reg << 15) | data); - rtl_udelay_loop_wait_low(tp, &rtl_ocp_gphy_cond, 25, 10); + rtl_loop_wait_low(tp, &rtl_ocp_gphy_cond, 25, 10); } static int r8168_phy_ocp_read(struct rtl8169_private *tp, u32 reg) @@ -817,7 +796,7 @@ static int r8168_phy_ocp_read(struct rtl8169_private *tp, u32 reg) RTL_W32(tp, GPHY_OCP, reg << 15); - return rtl_udelay_loop_wait_high(tp, &rtl_ocp_gphy_cond, 25, 10) ? + return rtl_loop_wait_high(tp, &rtl_ocp_gphy_cond, 25, 10) ? (RTL_R32(tp, GPHY_OCP) & 0xffff) : -ETIMEDOUT; } @@ -847,8 +826,6 @@ static void r8168_mac_ocp_modify(struct rtl8169_private *tp, u32 reg, u16 mask, r8168_mac_ocp_write(tp, reg, (data & ~mask) | set); } -#define OCP_STD_PHY_BASE 0xa400 - static void r8168g_mdio_write(struct rtl8169_private *tp, int reg, int value) { if (reg == 0x1f) { @@ -897,7 +874,7 @@ static void r8169_mdio_write(struct rtl8169_private *tp, int reg, int value) { RTL_W32(tp, PHYAR, 0x80000000 | (reg & 0x1f) << 16 | (value & 0xffff)); - rtl_udelay_loop_wait_low(tp, &rtl_phyar_cond, 25, 20); + rtl_loop_wait_low(tp, &rtl_phyar_cond, 25, 20); /* * According to hardware specs a 20us delay is required after write * complete indication, but before sending next command. @@ -911,7 +888,7 @@ static int r8169_mdio_read(struct rtl8169_private *tp, int reg) RTL_W32(tp, PHYAR, 0x0 | (reg & 0x1f) << 16); - value = rtl_udelay_loop_wait_high(tp, &rtl_phyar_cond, 25, 20) ? + value = rtl_loop_wait_high(tp, &rtl_phyar_cond, 25, 20) ? RTL_R32(tp, PHYAR) & 0xffff : -ETIMEDOUT; /* @@ -934,7 +911,7 @@ static void r8168dp_1_mdio_access(struct rtl8169_private *tp, int reg, u32 data) RTL_W32(tp, OCPAR, OCPAR_GPHY_WRITE_CMD); RTL_W32(tp, EPHY_RXER_NUM, 0); - rtl_udelay_loop_wait_low(tp, &rtl_ocpar_cond, 1000, 100); + rtl_loop_wait_low(tp, &rtl_ocpar_cond, 1000, 100); } static void r8168dp_1_mdio_write(struct rtl8169_private *tp, int reg, int value) @@ -951,7 +928,7 @@ static int r8168dp_1_mdio_read(struct rtl8169_private *tp, int reg) RTL_W32(tp, OCPAR, OCPAR_GPHY_READ_CMD); RTL_W32(tp, EPHY_RXER_NUM, 0); - return rtl_udelay_loop_wait_high(tp, &rtl_ocpar_cond, 1000, 100) ? + return rtl_loop_wait_high(tp, &rtl_ocpar_cond, 1000, 100) ? RTL_R32(tp, OCPDR) & OCPDR_DATA_MASK : -ETIMEDOUT; } @@ -1037,7 +1014,7 @@ static void rtl_ephy_write(struct rtl8169_private *tp, int reg_addr, int value) RTL_W32(tp, EPHYAR, EPHYAR_WRITE_CMD | (value & EPHYAR_DATA_MASK) | (reg_addr & EPHYAR_REG_MASK) << EPHYAR_REG_SHIFT); - rtl_udelay_loop_wait_low(tp, &rtl_ephyar_cond, 10, 100); + rtl_loop_wait_low(tp, &rtl_ephyar_cond, 10, 100); udelay(10); } @@ -1046,10 +1023,17 @@ static u16 rtl_ephy_read(struct rtl8169_private *tp, int reg_addr) { RTL_W32(tp, EPHYAR, (reg_addr & EPHYAR_REG_MASK) << EPHYAR_REG_SHIFT); - return rtl_udelay_loop_wait_high(tp, &rtl_ephyar_cond, 10, 100) ? + return rtl_loop_wait_high(tp, &rtl_ephyar_cond, 10, 100) ? RTL_R32(tp, EPHYAR) & EPHYAR_DATA_MASK : ~0; } +static void r8168fp_adjust_ocp_cmd(struct rtl8169_private *tp, u32 *cmd, int type) +{ + /* based on RTL8168FP_OOBMAC_BASE in vendor driver */ + if (tp->mac_version == RTL_GIGA_MAC_VER_52 && type == ERIAR_OOB) + *cmd |= 0x7f0 << 18; +} + DECLARE_RTL_COND(rtl_eriar_cond) { return RTL_R32(tp, ERIAR) & ERIAR_FLAG; @@ -1058,11 +1042,14 @@ DECLARE_RTL_COND(rtl_eriar_cond) static void _rtl_eri_write(struct rtl8169_private *tp, int addr, u32 mask, u32 val, int type) { + u32 cmd = ERIAR_WRITE_CMD | type | mask | addr; + BUG_ON((addr & 3) || (mask == 0)); RTL_W32(tp, ERIDR, val); - RTL_W32(tp, ERIAR, ERIAR_WRITE_CMD | type | mask | addr); + r8168fp_adjust_ocp_cmd(tp, &cmd, type); + RTL_W32(tp, ERIAR, cmd); - rtl_udelay_loop_wait_low(tp, &rtl_eriar_cond, 100, 100); + rtl_loop_wait_low(tp, &rtl_eriar_cond, 100, 100); } static void rtl_eri_write(struct rtl8169_private *tp, int addr, u32 mask, @@ -1073,9 +1060,12 @@ static void rtl_eri_write(struct rtl8169_private *tp, int addr, u32 mask, static u32 _rtl_eri_read(struct rtl8169_private *tp, int addr, int type) { - RTL_W32(tp, ERIAR, ERIAR_READ_CMD | type | ERIAR_MASK_1111 | addr); + u32 cmd = ERIAR_READ_CMD | type | ERIAR_MASK_1111 | addr; + + r8168fp_adjust_ocp_cmd(tp, &cmd, type); + RTL_W32(tp, ERIAR, cmd); - return rtl_udelay_loop_wait_high(tp, &rtl_eriar_cond, 100, 100) ? + return rtl_loop_wait_high(tp, &rtl_eriar_cond, 100, 100) ? RTL_R32(tp, ERIDR) : ~0; } @@ -1084,35 +1074,31 @@ static u32 rtl_eri_read(struct rtl8169_private *tp, int addr) return _rtl_eri_read(tp, addr, ERIAR_EXGMAC); } -static void rtl_w0w1_eri(struct rtl8169_private *tp, int addr, u32 mask, u32 p, - u32 m) +static void rtl_w0w1_eri(struct rtl8169_private *tp, int addr, u32 p, u32 m) { - u32 val; + u32 val = rtl_eri_read(tp, addr); - val = rtl_eri_read(tp, addr); - rtl_eri_write(tp, addr, mask, (val & ~m) | p); + rtl_eri_write(tp, addr, ERIAR_MASK_1111, (val & ~m) | p); } -static void rtl_eri_set_bits(struct rtl8169_private *tp, int addr, u32 mask, - u32 p) +static void rtl_eri_set_bits(struct rtl8169_private *tp, int addr, u32 p) { - rtl_w0w1_eri(tp, addr, mask, p, 0); + rtl_w0w1_eri(tp, addr, p, 0); } -static void rtl_eri_clear_bits(struct rtl8169_private *tp, int addr, u32 mask, - u32 m) +static void rtl_eri_clear_bits(struct rtl8169_private *tp, int addr, u32 m) { - rtl_w0w1_eri(tp, addr, mask, 0, m); + rtl_w0w1_eri(tp, addr, 0, m); } -static u32 r8168dp_ocp_read(struct rtl8169_private *tp, u8 mask, u16 reg) +static u32 r8168dp_ocp_read(struct rtl8169_private *tp, u16 reg) { - RTL_W32(tp, OCPAR, ((u32)mask & 0x0f) << 12 | (reg & 0x0fff)); - return rtl_udelay_loop_wait_high(tp, &rtl_ocpar_cond, 100, 20) ? + RTL_W32(tp, OCPAR, 0x0fu << 12 | (reg & 0x0fff)); + return rtl_loop_wait_high(tp, &rtl_ocpar_cond, 100, 20) ? RTL_R32(tp, OCPDR) : ~0; } -static u32 r8168ep_ocp_read(struct rtl8169_private *tp, u8 mask, u16 reg) +static u32 r8168ep_ocp_read(struct rtl8169_private *tp, u16 reg) { return _rtl_eri_read(tp, reg, ERIAR_OOB); } @@ -1122,7 +1108,7 @@ static void r8168dp_ocp_write(struct rtl8169_private *tp, u8 mask, u16 reg, { RTL_W32(tp, OCPDR, data); RTL_W32(tp, OCPAR, OCPAR_FLAG | ((u32)mask & 0x0f) << 12 | (reg & 0x0fff)); - rtl_udelay_loop_wait_low(tp, &rtl_ocpar_cond, 100, 20); + rtl_loop_wait_low(tp, &rtl_ocpar_cond, 100, 20); } static void r8168ep_ocp_write(struct rtl8169_private *tp, u8 mask, u16 reg, @@ -1154,12 +1140,12 @@ DECLARE_RTL_COND(rtl_dp_ocp_read_cond) reg = rtl8168_get_ocp_reg(tp); - return r8168dp_ocp_read(tp, 0x0f, reg) & 0x00000800; + return r8168dp_ocp_read(tp, reg) & 0x00000800; } DECLARE_RTL_COND(rtl_ep_ocp_read_cond) { - return r8168ep_ocp_read(tp, 0x0f, 0x124) & 0x00000001; + return r8168ep_ocp_read(tp, 0x124) & 0x00000001; } DECLARE_RTL_COND(rtl_ocp_tx_cond) @@ -1170,7 +1156,7 @@ DECLARE_RTL_COND(rtl_ocp_tx_cond) static void rtl8168ep_stop_cmac(struct rtl8169_private *tp) { RTL_W8(tp, IBCR2, RTL_R8(tp, IBCR2) & ~0x01); - rtl_msleep_loop_wait_high(tp, &rtl_ocp_tx_cond, 50, 2000); + rtl_loop_wait_high(tp, &rtl_ocp_tx_cond, 50000, 2000); RTL_W8(tp, IBISR0, RTL_R8(tp, IBISR0) | 0x20); RTL_W8(tp, IBCR0, RTL_R8(tp, IBCR0) & ~0x01); } @@ -1178,15 +1164,14 @@ static void rtl8168ep_stop_cmac(struct rtl8169_private *tp) static void rtl8168dp_driver_start(struct rtl8169_private *tp) { r8168dp_oob_notify(tp, OOB_CMD_DRIVER_START); - rtl_msleep_loop_wait_high(tp, &rtl_dp_ocp_read_cond, 10, 10); + rtl_loop_wait_high(tp, &rtl_dp_ocp_read_cond, 10000, 10); } static void rtl8168ep_driver_start(struct rtl8169_private *tp) { r8168ep_ocp_write(tp, 0x01, 0x180, OOB_CMD_DRIVER_START); - r8168ep_ocp_write(tp, 0x01, 0x30, - r8168ep_ocp_read(tp, 0x01, 0x30) | 0x01); - rtl_msleep_loop_wait_high(tp, &rtl_ep_ocp_read_cond, 10, 10); + r8168ep_ocp_write(tp, 0x01, 0x30, r8168ep_ocp_read(tp, 0x30) | 0x01); + rtl_loop_wait_high(tp, &rtl_ep_ocp_read_cond, 10000, 10); } static void rtl8168_driver_start(struct rtl8169_private *tp) @@ -1209,16 +1194,15 @@ static void rtl8168_driver_start(struct rtl8169_private *tp) static void rtl8168dp_driver_stop(struct rtl8169_private *tp) { r8168dp_oob_notify(tp, OOB_CMD_DRIVER_STOP); - rtl_msleep_loop_wait_low(tp, &rtl_dp_ocp_read_cond, 10, 10); + rtl_loop_wait_low(tp, &rtl_dp_ocp_read_cond, 10000, 10); } static void rtl8168ep_driver_stop(struct rtl8169_private *tp) { rtl8168ep_stop_cmac(tp); r8168ep_ocp_write(tp, 0x01, 0x180, OOB_CMD_DRIVER_STOP); - r8168ep_ocp_write(tp, 0x01, 0x30, - r8168ep_ocp_read(tp, 0x01, 0x30) | 0x01); - rtl_msleep_loop_wait_low(tp, &rtl_ep_ocp_read_cond, 10, 10); + r8168ep_ocp_write(tp, 0x01, 0x30, r8168ep_ocp_read(tp, 0x30) | 0x01); + rtl_loop_wait_low(tp, &rtl_ep_ocp_read_cond, 10000, 10); } static void rtl8168_driver_stop(struct rtl8169_private *tp) @@ -1242,12 +1226,12 @@ static bool r8168dp_check_dash(struct rtl8169_private *tp) { u16 reg = rtl8168_get_ocp_reg(tp); - return !!(r8168dp_ocp_read(tp, 0x0f, reg) & 0x00008000); + return !!(r8168dp_ocp_read(tp, reg) & 0x00008000); } static bool r8168ep_check_dash(struct rtl8169_private *tp) { - return !!(r8168ep_ocp_read(tp, 0x0f, 0x128) & 0x00000001); + return r8168ep_ocp_read(tp, 0x128) & 0x00000001; } static bool r8168_check_dash(struct rtl8169_private *tp) @@ -1266,8 +1250,8 @@ static bool r8168_check_dash(struct rtl8169_private *tp) static void rtl_reset_packet_filter(struct rtl8169_private *tp) { - rtl_eri_clear_bits(tp, 0xdc, ERIAR_MASK_0001, BIT(0)); - rtl_eri_set_bits(tp, 0xdc, ERIAR_MASK_0001, BIT(0)); + rtl_eri_clear_bits(tp, 0xdc, BIT(0)); + rtl_eri_set_bits(tp, 0xdc, BIT(0)); } DECLARE_RTL_COND(rtl_efusear_cond) @@ -1279,7 +1263,7 @@ u8 rtl8168d_efuse_read(struct rtl8169_private *tp, int reg_addr) { RTL_W32(tp, EFUSEAR, (reg_addr & EFUSEAR_REG_MASK) << EFUSEAR_REG_SHIFT); - return rtl_udelay_loop_wait_high(tp, &rtl_efusear_cond, 100, 300) ? + return rtl_loop_wait_high(tp, &rtl_efusear_cond, 100, 300) ? RTL_R32(tp, EFUSEAR) & EFUSEAR_DATA_MASK : ~0; } @@ -1394,11 +1378,9 @@ static void __rtl8169_set_wol(struct rtl8169_private *tp, u32 wolopts) if (rtl_is_8168evl_up(tp)) { tmp--; if (wolopts & WAKE_MAGIC) - rtl_eri_set_bits(tp, 0x0dc, ERIAR_MASK_0100, - MagicPacket_v2); + rtl_eri_set_bits(tp, 0x0dc, MagicPacket_v2); else - rtl_eri_clear_bits(tp, 0x0dc, ERIAR_MASK_0100, - MagicPacket_v2); + rtl_eri_clear_bits(tp, 0x0dc, MagicPacket_v2); } else if (rtl_is_8125(tp)) { tmp--; if (wolopts & WAKE_MAGIC) @@ -1423,7 +1405,7 @@ static void __rtl8169_set_wol(struct rtl8169_private *tp, u32 wolopts) break; case RTL_GIGA_MAC_VER_34: case RTL_GIGA_MAC_VER_37: - case RTL_GIGA_MAC_VER_39 ... RTL_GIGA_MAC_VER_52: + case RTL_GIGA_MAC_VER_39 ... RTL_GIGA_MAC_VER_61: options = RTL_R8(tp, Config2) & ~PME_SIGNAL; if (wolopts) options |= PME_SIGNAL; @@ -1497,19 +1479,15 @@ static netdev_features_t rtl8169_fix_features(struct net_device *dev, return features; } -static int rtl8169_set_features(struct net_device *dev, - netdev_features_t features) +static void rtl_set_rx_config_features(struct rtl8169_private *tp, + netdev_features_t features) { - struct rtl8169_private *tp = netdev_priv(dev); - u32 rx_config; + u32 rx_config = RTL_R32(tp, RxConfig); - rtl_lock_work(tp); - - rx_config = RTL_R32(tp, RxConfig); if (features & NETIF_F_RXALL) - rx_config |= (AcceptErr | AcceptRunt); + rx_config |= RX_CONFIG_ACCEPT_ERR_MASK; else - rx_config &= ~(AcceptErr | AcceptRunt); + rx_config &= ~RX_CONFIG_ACCEPT_ERR_MASK; if (rtl_is_8125(tp)) { if (features & NETIF_F_HW_VLAN_CTAG_RX) @@ -1519,6 +1497,16 @@ static int rtl8169_set_features(struct net_device *dev, } RTL_W32(tp, RxConfig, rx_config); +} + +static int rtl8169_set_features(struct net_device *dev, + netdev_features_t features) +{ + struct rtl8169_private *tp = netdev_priv(dev); + + rtl_lock_work(tp); + + rtl_set_rx_config_features(tp, features); if (features & NETIF_F_RXCSUM) tp->cp_cmd |= RxChkSum; @@ -1568,20 +1556,6 @@ static void rtl8169_get_regs(struct net_device *dev, struct ethtool_regs *regs, rtl_unlock_work(tp); } -static u32 rtl8169_get_msglevel(struct net_device *dev) -{ - struct rtl8169_private *tp = netdev_priv(dev); - - return tp->msg_enable; -} - -static void rtl8169_set_msglevel(struct net_device *dev, u32 value) -{ - struct rtl8169_private *tp = netdev_priv(dev); - - tp->msg_enable = value; -} - static const char rtl8169_gstrings[][ETH_GSTRING_LEN] = { "tx_packets", "rx_packets", @@ -1613,7 +1587,7 @@ DECLARE_RTL_COND(rtl_counters_cond) return RTL_R32(tp, CounterAddrLow) & (CounterReset | CounterDump); } -static bool rtl8169_do_counters(struct rtl8169_private *tp, u32 counter_cmd) +static void rtl8169_do_counters(struct rtl8169_private *tp, u32 counter_cmd) { dma_addr_t paddr = tp->counters_phys_addr; u32 cmd; @@ -1624,22 +1598,20 @@ static bool rtl8169_do_counters(struct rtl8169_private *tp, u32 counter_cmd) RTL_W32(tp, CounterAddrLow, cmd); RTL_W32(tp, CounterAddrLow, cmd | counter_cmd); - return rtl_udelay_loop_wait_low(tp, &rtl_counters_cond, 10, 1000); + rtl_loop_wait_low(tp, &rtl_counters_cond, 10, 1000); } -static bool rtl8169_reset_counters(struct rtl8169_private *tp) +static void rtl8169_reset_counters(struct rtl8169_private *tp) { /* * Versions prior to RTL_GIGA_MAC_VER_19 don't support resetting the * tally counters. */ - if (tp->mac_version < RTL_GIGA_MAC_VER_19) - return true; - - return rtl8169_do_counters(tp, CounterReset); + if (tp->mac_version >= RTL_GIGA_MAC_VER_19) + rtl8169_do_counters(tp, CounterReset); } -static bool rtl8169_update_counters(struct rtl8169_private *tp) +static void rtl8169_update_counters(struct rtl8169_private *tp) { u8 val = RTL_R8(tp, ChipCmd); @@ -1647,16 +1619,13 @@ static bool rtl8169_update_counters(struct rtl8169_private *tp) * Some chips are unable to dump tally counters when the receiver * is disabled. If 0xff chip may be in a PCI power-save state. */ - if (!(val & CmdRxEnb) || val == 0xff) - return true; - - return rtl8169_do_counters(tp, CounterDump); + if (val & CmdRxEnb && val != 0xff) + rtl8169_do_counters(tp, CounterDump); } -static bool rtl8169_init_counter_offsets(struct rtl8169_private *tp) +static void rtl8169_init_counter_offsets(struct rtl8169_private *tp) { struct rtl8169_counters *counters = tp->counters; - bool ret = false; /* * rtl8169_init_counter_offsets is called from rtl_open. On chip @@ -1674,22 +1643,16 @@ static bool rtl8169_init_counter_offsets(struct rtl8169_private *tp) */ if (tp->tc_offset.inited) - return true; - - /* If both, reset and update fail, propagate to caller. */ - if (rtl8169_reset_counters(tp)) - ret = true; + return; - if (rtl8169_update_counters(tp)) - ret = true; + rtl8169_reset_counters(tp); + rtl8169_update_counters(tp); tp->tc_offset.tx_errors = counters->tx_errors; tp->tc_offset.tx_multi_collision = counters->tx_multi_collision; tp->tc_offset.tx_aborted = counters->tx_aborted; tp->tc_offset.rx_missed = counters->rx_missed; tp->tc_offset.inited = true; - - return ret; } static void rtl8169_get_ethtool_stats(struct net_device *dev, @@ -1760,46 +1723,34 @@ static void rtl8169_get_strings(struct net_device *dev, u32 stringset, u8 *data) * 1 1 160us 81.92us 1.31ms */ -/* rx/tx scale factors for one particular CPlusCmd[0:1] value */ -struct rtl_coalesce_scale { - /* Rx / Tx */ - u32 nsecs[2]; -}; - /* rx/tx scale factors for all CPlusCmd[0:1] cases */ struct rtl_coalesce_info { u32 speed; - struct rtl_coalesce_scale scalev[4]; /* each CPlusCmd[0:1] case */ + u32 scale_nsecs[4]; }; -/* produce (r,t) pairs with each being in series of *1, *8, *8*2, *8*2*2 */ -#define rxtx_x1822(r, t) { \ - {{(r), (t)}}, \ - {{(r)*8, (t)*8}}, \ - {{(r)*8*2, (t)*8*2}}, \ - {{(r)*8*2*2, (t)*8*2*2}}, \ -} +/* produce array with base delay *1, *8, *8*2, *8*2*2 */ +#define COALESCE_DELAY(d) { (d), 8 * (d), 16 * (d), 32 * (d) } + static const struct rtl_coalesce_info rtl_coalesce_info_8169[] = { - /* speed delays: rx00 tx00 */ - { SPEED_10, rxtx_x1822(40960, 40960) }, - { SPEED_100, rxtx_x1822( 2560, 2560) }, - { SPEED_1000, rxtx_x1822( 320, 320) }, + { SPEED_10, COALESCE_DELAY(40960) }, + { SPEED_100, COALESCE_DELAY(2560) }, + { SPEED_1000, COALESCE_DELAY(320) }, { 0 }, }; static const struct rtl_coalesce_info rtl_coalesce_info_8168_8136[] = { - /* speed delays: rx00 tx00 */ - { SPEED_10, rxtx_x1822(40960, 40960) }, - { SPEED_100, rxtx_x1822( 2560, 2560) }, - { SPEED_1000, rxtx_x1822( 5000, 5000) }, + { SPEED_10, COALESCE_DELAY(40960) }, + { SPEED_100, COALESCE_DELAY(2560) }, + { SPEED_1000, COALESCE_DELAY(5000) }, { 0 }, }; -#undef rxtx_x1822 +#undef COALESCE_DELAY /* get rx/tx scale vector corresponding to current speed */ -static const struct rtl_coalesce_info *rtl_coalesce_info(struct net_device *dev) +static const struct rtl_coalesce_info * +rtl_coalesce_info(struct rtl8169_private *tp) { - struct rtl8169_private *tp = netdev_priv(dev); const struct rtl_coalesce_info *ci; if (tp->mac_version <= RTL_GIGA_MAC_VER_06) @@ -1819,16 +1770,8 @@ static int rtl_get_coalesce(struct net_device *dev, struct ethtool_coalesce *ec) { struct rtl8169_private *tp = netdev_priv(dev); const struct rtl_coalesce_info *ci; - const struct rtl_coalesce_scale *scale; - struct { - u32 *max_frames; - u32 *usecs; - } coal_settings [] = { - { &ec->rx_max_coalesced_frames, &ec->rx_coalesce_usecs }, - { &ec->tx_max_coalesced_frames, &ec->tx_coalesce_usecs } - }, *p = coal_settings; - int i; - u16 w; + u32 scale, c_us, c_fr; + u16 intrmit; if (rtl_is_8125(tp)) return -EOPNOTSUPP; @@ -1836,111 +1779,111 @@ static int rtl_get_coalesce(struct net_device *dev, struct ethtool_coalesce *ec) memset(ec, 0, sizeof(*ec)); /* get rx/tx scale corresponding to current speed and CPlusCmd[0:1] */ - ci = rtl_coalesce_info(dev); + ci = rtl_coalesce_info(tp); if (IS_ERR(ci)) return PTR_ERR(ci); - scale = &ci->scalev[tp->cp_cmd & INTT_MASK]; + scale = ci->scale_nsecs[tp->cp_cmd & INTT_MASK]; - /* read IntrMitigate and adjust according to scale */ - for (w = RTL_R16(tp, IntrMitigate); w; w >>= RTL_COALESCE_SHIFT, p++) { - *p->max_frames = (w & RTL_COALESCE_MASK) << 2; - w >>= RTL_COALESCE_SHIFT; - *p->usecs = w & RTL_COALESCE_MASK; - } + intrmit = RTL_R16(tp, IntrMitigate); - for (i = 0; i < 2; i++) { - p = coal_settings + i; - *p->usecs = (*p->usecs * scale->nsecs[i]) / 1000; + c_us = FIELD_GET(RTL_COALESCE_TX_USECS, intrmit); + ec->tx_coalesce_usecs = DIV_ROUND_UP(c_us * scale, 1000); - /* - * ethtool_coalesce says it is illegal to set both usecs and - * max_frames to 0. - */ - if (!*p->usecs && !*p->max_frames) - *p->max_frames = 1; - } + c_fr = FIELD_GET(RTL_COALESCE_TX_FRAMES, intrmit); + /* ethtool_coalesce states usecs and max_frames must not both be 0 */ + ec->tx_max_coalesced_frames = (c_us || c_fr) ? c_fr * 4 : 1; + + c_us = FIELD_GET(RTL_COALESCE_RX_USECS, intrmit); + ec->rx_coalesce_usecs = DIV_ROUND_UP(c_us * scale, 1000); + + c_fr = FIELD_GET(RTL_COALESCE_RX_FRAMES, intrmit); + ec->rx_max_coalesced_frames = (c_us || c_fr) ? c_fr * 4 : 1; return 0; } -/* choose appropriate scale factor and CPlusCmd[0:1] for (speed, nsec) */ -static const struct rtl_coalesce_scale *rtl_coalesce_choose_scale( - struct net_device *dev, u32 nsec, u16 *cp01) +/* choose appropriate scale factor and CPlusCmd[0:1] for (speed, usec) */ +static int rtl_coalesce_choose_scale(struct rtl8169_private *tp, u32 usec, + u16 *cp01) { const struct rtl_coalesce_info *ci; u16 i; - ci = rtl_coalesce_info(dev); + ci = rtl_coalesce_info(tp); if (IS_ERR(ci)) - return ERR_CAST(ci); + return PTR_ERR(ci); for (i = 0; i < 4; i++) { - u32 rxtx_maxscale = max(ci->scalev[i].nsecs[0], - ci->scalev[i].nsecs[1]); - if (nsec <= rxtx_maxscale * RTL_COALESCE_T_MAX) { + if (usec <= ci->scale_nsecs[i] * RTL_COALESCE_T_MAX / 1000U) { *cp01 = i; - return &ci->scalev[i]; + return ci->scale_nsecs[i]; } } - return ERR_PTR(-EINVAL); + return -ERANGE; } static int rtl_set_coalesce(struct net_device *dev, struct ethtool_coalesce *ec) { struct rtl8169_private *tp = netdev_priv(dev); - const struct rtl_coalesce_scale *scale; - struct { - u32 frames; - u32 usecs; - } coal_settings [] = { - { ec->rx_max_coalesced_frames, ec->rx_coalesce_usecs }, - { ec->tx_max_coalesced_frames, ec->tx_coalesce_usecs } - }, *p = coal_settings; - u16 w = 0, cp01; - int i; + u32 tx_fr = ec->tx_max_coalesced_frames; + u32 rx_fr = ec->rx_max_coalesced_frames; + u32 coal_usec_max, units; + u16 w = 0, cp01 = 0; + int scale; if (rtl_is_8125(tp)) return -EOPNOTSUPP; - scale = rtl_coalesce_choose_scale(dev, - max(p[0].usecs, p[1].usecs) * 1000, &cp01); - if (IS_ERR(scale)) - return PTR_ERR(scale); + if (rx_fr > RTL_COALESCE_FRAME_MAX || tx_fr > RTL_COALESCE_FRAME_MAX) + return -ERANGE; - for (i = 0; i < 2; i++, p++) { - u32 units; + coal_usec_max = max(ec->rx_coalesce_usecs, ec->tx_coalesce_usecs); + scale = rtl_coalesce_choose_scale(tp, coal_usec_max, &cp01); + if (scale < 0) + return scale; - /* - * accept max_frames=1 we returned in rtl_get_coalesce. - * accept it not only when usecs=0 because of e.g. the following scenario: - * - * - both rx_usecs=0 & rx_frames=0 in hardware (no delay on RX) - * - rtl_get_coalesce returns rx_usecs=0, rx_frames=1 - * - then user does `ethtool -C eth0 rx-usecs 100` - * - * since ethtool sends to kernel whole ethtool_coalesce - * settings, if we do not handle rx_usecs=!0, rx_frames=1 - * we'll reject it below in `frames % 4 != 0`. - */ - if (p->frames == 1) { - p->frames = 0; - } + /* Accept max_frames=1 we returned in rtl_get_coalesce. Accept it + * not only when usecs=0 because of e.g. the following scenario: + * + * - both rx_usecs=0 & rx_frames=0 in hardware (no delay on RX) + * - rtl_get_coalesce returns rx_usecs=0, rx_frames=1 + * - then user does `ethtool -C eth0 rx-usecs 100` + * + * Since ethtool sends to kernel whole ethtool_coalesce settings, + * if we want to ignore rx_frames then it has to be set to 0. + */ + if (rx_fr == 1) + rx_fr = 0; + if (tx_fr == 1) + tx_fr = 0; + + /* HW requires time limit to be set if frame limit is set */ + if ((tx_fr && !ec->tx_coalesce_usecs) || + (rx_fr && !ec->rx_coalesce_usecs)) + return -EINVAL; - units = p->usecs * 1000 / scale->nsecs[i]; - if (p->frames > RTL_COALESCE_FRAME_MAX || p->frames % 4) - return -EINVAL; + w |= FIELD_PREP(RTL_COALESCE_TX_FRAMES, DIV_ROUND_UP(tx_fr, 4)); + w |= FIELD_PREP(RTL_COALESCE_RX_FRAMES, DIV_ROUND_UP(rx_fr, 4)); - w <<= RTL_COALESCE_SHIFT; - w |= units; - w <<= RTL_COALESCE_SHIFT; - w |= p->frames >> 2; - } + units = DIV_ROUND_UP(ec->tx_coalesce_usecs * 1000U, scale); + w |= FIELD_PREP(RTL_COALESCE_TX_USECS, units); + units = DIV_ROUND_UP(ec->rx_coalesce_usecs * 1000U, scale); + w |= FIELD_PREP(RTL_COALESCE_RX_USECS, units); rtl_lock_work(tp); - RTL_W16(tp, IntrMitigate, swab16(w)); + RTL_W16(tp, IntrMitigate, w); + + /* Meaning of PktCntrDisable bit changed from RTL8168e-vl */ + if (rtl_is_8168evl_up(tp)) { + if (!rx_fr && !tx_fr) + /* disable packet counter */ + tp->cp_cmd |= PktCntrDisable; + else + tp->cp_cmd &= ~PktCntrDisable; + } tp->cp_cmd = (tp->cp_cmd & ~INTT_MASK) | cp01; RTL_W16(tp, CPlusCmd, tp->cp_cmd); @@ -1989,12 +1932,6 @@ static int rtl8169_set_eee(struct net_device *dev, struct ethtool_eee *data) goto out; } - if (dev->phydev->autoneg == AUTONEG_DISABLE || - dev->phydev->duplex != DUPLEX_FULL) { - ret = -EPROTONOSUPPORT; - goto out; - } - ret = phy_ethtool_set_eee(tp->phydev, data); if (!ret) @@ -2013,8 +1950,6 @@ static const struct ethtool_ops rtl8169_ethtool_ops = { .get_link = ethtool_op_get_link, .get_coalesce = rtl_get_coalesce, .set_coalesce = rtl_set_coalesce, - .get_msglevel = rtl8169_get_msglevel, - .set_msglevel = rtl8169_set_msglevel, .get_regs = rtl8169_get_regs, .get_wol = rtl8169_get_wol, .set_wol = rtl8169_set_wol, @@ -2189,7 +2124,7 @@ static void rtl8168_config_eee_mac(struct rtl8169_private *tp) if (tp->mac_version != RTL_GIGA_MAC_VER_38) RTL_W8(tp, EEE_LED, RTL_R8(tp, EEE_LED) & ~0x07); - rtl_eri_set_bits(tp, 0x1b0, ERIAR_MASK_1111, 0x0003); + rtl_eri_set_bits(tp, 0x1b0, 0x0003); } static void rtl8125_config_eee_mac(struct rtl8169_private *tp) @@ -2353,7 +2288,7 @@ static void rtl_pll_power_down(struct rtl8169_private *tp) case RTL_GIGA_MAC_VER_40: case RTL_GIGA_MAC_VER_41: case RTL_GIGA_MAC_VER_49: - rtl_eri_clear_bits(tp, 0x1a8, ERIAR_MASK_1111, 0xfc000000); + rtl_eri_clear_bits(tp, 0x1a8, 0xfc000000); RTL_W8(tp, PMCH, RTL_R8(tp, PMCH) & ~0x80); break; default: @@ -2386,15 +2321,13 @@ static void rtl_pll_power_up(struct rtl8169_private *tp) case RTL_GIGA_MAC_VER_41: case RTL_GIGA_MAC_VER_49: RTL_W8(tp, PMCH, RTL_R8(tp, PMCH) | 0xc0); - rtl_eri_set_bits(tp, 0x1a8, ERIAR_MASK_1111, 0xfc000000); + rtl_eri_set_bits(tp, 0x1a8, 0xfc000000); break; default: break; } phy_resume(tp->phydev); - /* give MAC/PHY some time to resume */ - msleep(20); } static void rtl_init_rxcfg(struct rtl8169_private *tp) @@ -2413,8 +2346,7 @@ static void rtl_init_rxcfg(struct rtl8169_private *tp) RTL_W32(tp, RxConfig, RX128_INT_EN | RX_MULTI_EN | RX_DMA_BURST | RX_EARLY_OFF); break; case RTL_GIGA_MAC_VER_60 ... RTL_GIGA_MAC_VER_61: - RTL_W32(tp, RxConfig, RX_FETCH_DFLT_8125 | RX_VLAN_8125 | - RX_DMA_BURST); + RTL_W32(tp, RxConfig, RX_FETCH_DFLT_8125 | RX_DMA_BURST); break; default: RTL_W32(tp, RxConfig, RX128_INT_EN | RX_DMA_BURST); @@ -2528,7 +2460,7 @@ static void rtl_hw_reset(struct rtl8169_private *tp) { RTL_W8(tp, ChipCmd, CmdReset); - rtl_udelay_loop_wait_low(tp, &rtl_chipcmd_cond, 100, 100); + rtl_loop_wait_low(tp, &rtl_chipcmd_cond, 100, 100); } static void rtl_request_firmware(struct rtl8169_private *tp) @@ -2540,10 +2472,8 @@ static void rtl_request_firmware(struct rtl8169_private *tp) return; rtl_fw = kzalloc(sizeof(*rtl_fw), GFP_KERNEL); - if (!rtl_fw) { - netif_warn(tp, ifup, tp->dev, "Unable to load firmware, out of memory\n"); + if (!rtl_fw) return; - } rtl_fw->phy_write = rtl_writephy; rtl_fw->phy_read = rtl_readphy; @@ -2573,31 +2503,31 @@ DECLARE_RTL_COND(rtl_txcfg_empty_cond) return RTL_R32(tp, TxConfig) & TXCFG_EMPTY; } -static void rtl8169_hw_reset(struct rtl8169_private *tp) +DECLARE_RTL_COND(rtl_rxtx_empty_cond) { - /* Disable interrupts */ - rtl8169_irq_mask_and_ack(tp); - - rtl_rx_close(tp); + return (RTL_R8(tp, MCU) & RXTX_EMPTY) == RXTX_EMPTY; +} +static void rtl_wait_txrx_fifo_empty(struct rtl8169_private *tp) +{ switch (tp->mac_version) { - case RTL_GIGA_MAC_VER_27: - case RTL_GIGA_MAC_VER_28: - case RTL_GIGA_MAC_VER_31: - rtl_udelay_loop_wait_low(tp, &rtl_npq_cond, 20, 42*42); - break; - case RTL_GIGA_MAC_VER_34 ... RTL_GIGA_MAC_VER_38: case RTL_GIGA_MAC_VER_40 ... RTL_GIGA_MAC_VER_52: - RTL_W8(tp, ChipCmd, RTL_R8(tp, ChipCmd) | StopReq); - rtl_udelay_loop_wait_high(tp, &rtl_txcfg_empty_cond, 100, 666); + rtl_loop_wait_high(tp, &rtl_txcfg_empty_cond, 100, 42); + rtl_loop_wait_high(tp, &rtl_rxtx_empty_cond, 100, 42); + break; + case RTL_GIGA_MAC_VER_60 ... RTL_GIGA_MAC_VER_61: + rtl_loop_wait_high(tp, &rtl_rxtx_empty_cond, 100, 42); break; default: - RTL_W8(tp, ChipCmd, RTL_R8(tp, ChipCmd) | StopReq); - udelay(100); break; } +} - rtl_hw_reset(tp); +static void rtl_enable_rxdvgate(struct rtl8169_private *tp) +{ + RTL_W32(tp, MISC, RTL_R32(tp, MISC) | RXDV_GATED_EN); + fsleep(2000); + rtl_wait_txrx_fifo_empty(tp); } static void rtl_set_tx_config_registers(struct rtl8169_private *tp) @@ -2630,7 +2560,7 @@ static void rtl_set_rx_tx_desc_registers(struct rtl8169_private *tp) RTL_W32(tp, RxDescAddrLow, ((u64) tp->RxPhyAddr) & DMA_BIT_MASK(32)); } -static void rtl8169_set_magic_reg(struct rtl8169_private *tp, unsigned mac_version) +static void rtl8169_set_magic_reg(struct rtl8169_private *tp) { u32 val; @@ -2656,8 +2586,6 @@ static void rtl_set_rx_mode(struct net_device *dev) u32 tmp; if (dev->flags & IFF_PROMISC) { - /* Unconditionally log net taps. */ - netif_notice(tp, link, dev, "Promiscuous mode enabled\n"); rx_mode |= AcceptAllPhys; } else if (netdev_mc_count(dev) > MC_FILTER_LIMIT || dev->flags & IFF_ALLMULTI || @@ -2670,7 +2598,7 @@ static void rtl_set_rx_mode(struct net_device *dev) mc_filter[1] = mc_filter[0] = 0; netdev_for_each_mc_addr(ha, dev) { - u32 bit_nr = ether_crc(ETH_ALEN, ha->addr) >> 26; + u32 bit_nr = eth_hw_addr_crc(ha) >> 26; mc_filter[bit_nr >> 5] |= BIT(bit_nr & 31); } @@ -2681,14 +2609,11 @@ static void rtl_set_rx_mode(struct net_device *dev) } } - if (dev->features & NETIF_F_RXALL) - rx_mode |= (AcceptErr | AcceptRunt); - RTL_W32(tp, MAR0 + 4, mc_filter[1]); RTL_W32(tp, MAR0 + 0, mc_filter[0]); tmp = RTL_R32(tp, RxConfig); - RTL_W32(tp, RxConfig, (tmp & ~RX_CONFIG_ACCEPT_MASK) | rx_mode); + RTL_W32(tp, RxConfig, (tmp & ~RX_CONFIG_ACCEPT_OK_MASK) | rx_mode); } DECLARE_RTL_COND(rtl_csiar_cond) @@ -2704,7 +2629,7 @@ static void rtl_csi_write(struct rtl8169_private *tp, int addr, int value) RTL_W32(tp, CSIAR, CSIAR_WRITE_CMD | (addr & CSIAR_ADDR_MASK) | CSIAR_BYTE_ENABLE | func << 16); - rtl_udelay_loop_wait_low(tp, &rtl_csiar_cond, 10, 100); + rtl_loop_wait_low(tp, &rtl_csiar_cond, 10, 100); } static u32 rtl_csi_read(struct rtl8169_private *tp, int addr) @@ -2714,7 +2639,7 @@ static u32 rtl_csi_read(struct rtl8169_private *tp, int addr) RTL_W32(tp, CSIAR, (addr & CSIAR_ADDR_MASK) | func << 16 | CSIAR_BYTE_ENABLE); - return rtl_udelay_loop_wait_high(tp, &rtl_csiar_cond, 10, 100) ? + return rtl_loop_wait_high(tp, &rtl_csiar_cond, 10, 100) ? RTL_R32(tp, CSIDR) : ~0; } @@ -2971,12 +2896,14 @@ static void rtl_hw_start_8168e_2(struct rtl8169_private *tp) rtl_ephy_init(tp, e_info_8168e_2); rtl_eri_write(tp, 0xc0, ERIAR_MASK_0011, 0x0000); - rtl_eri_write(tp, 0xb8, ERIAR_MASK_0011, 0x0000); + rtl_eri_write(tp, 0xb8, ERIAR_MASK_1111, 0x0000); rtl_set_fifo_size(tp, 0x10, 0x10, 0x02, 0x06); + rtl_eri_set_bits(tp, 0x0d4, 0x1f00); + rtl_eri_set_bits(tp, 0x1d0, BIT(1)); + rtl_reset_packet_filter(tp); + rtl_eri_set_bits(tp, 0x1b0, BIT(4)); rtl_eri_write(tp, 0xcc, ERIAR_MASK_1111, 0x00000050); rtl_eri_write(tp, 0xd0, ERIAR_MASK_1111, 0x07ff0060); - rtl_eri_set_bits(tp, 0x1b0, ERIAR_MASK_0001, BIT(4)); - rtl_w0w1_eri(tp, 0x0d4, ERIAR_MASK_0011, 0x0c00, 0xff00); rtl_disable_clock_request(tp); @@ -2996,11 +2923,11 @@ static void rtl_hw_start_8168f(struct rtl8169_private *tp) rtl_set_def_aspm_entry_latency(tp); rtl_eri_write(tp, 0xc0, ERIAR_MASK_0011, 0x0000); - rtl_eri_write(tp, 0xb8, ERIAR_MASK_0011, 0x0000); + rtl_eri_write(tp, 0xb8, ERIAR_MASK_1111, 0x0000); rtl_set_fifo_size(tp, 0x10, 0x10, 0x02, 0x06); rtl_reset_packet_filter(tp); - rtl_eri_set_bits(tp, 0x1b0, ERIAR_MASK_0001, BIT(4)); - rtl_eri_set_bits(tp, 0x1d0, ERIAR_MASK_0001, BIT(4)); + rtl_eri_set_bits(tp, 0x1b0, BIT(4)); + rtl_eri_set_bits(tp, 0x1d0, BIT(4) | BIT(1)); rtl_eri_write(tp, 0xcc, ERIAR_MASK_1111, 0x00000050); rtl_eri_write(tp, 0xd0, ERIAR_MASK_1111, 0x00000060); @@ -3029,7 +2956,7 @@ static void rtl_hw_start_8168f_1(struct rtl8169_private *tp) rtl_ephy_init(tp, e_info_8168f_1); - rtl_w0w1_eri(tp, 0x0d4, ERIAR_MASK_0011, 0x0c00, 0xff00); + rtl_eri_set_bits(tp, 0x0d4, 0x1f00); } static void rtl_hw_start_8411(struct rtl8169_private *tp) @@ -3047,7 +2974,7 @@ static void rtl_hw_start_8411(struct rtl8169_private *tp) rtl_ephy_init(tp, e_info_8168f_1); - rtl_eri_set_bits(tp, 0x0d4, ERIAR_MASK_0011, 0x0c00); + rtl_eri_set_bits(tp, 0x0d4, 0x0c00); } static void rtl_hw_start_8168g(struct rtl8169_private *tp) @@ -3064,11 +2991,12 @@ static void rtl_hw_start_8168g(struct rtl8169_private *tp) rtl_eri_write(tp, 0xc0, ERIAR_MASK_0011, 0x0000); rtl_eri_write(tp, 0xb8, ERIAR_MASK_0011, 0x0000); + rtl_eri_set_bits(tp, 0x0d4, 0x1f80); rtl8168_config_eee_mac(tp); - rtl_w0w1_eri(tp, 0x2fc, ERIAR_MASK_0001, 0x01, 0x06); - rtl_eri_clear_bits(tp, 0x1b0, ERIAR_MASK_0011, BIT(12)); + rtl_w0w1_eri(tp, 0x2fc, 0x01, 0x06); + rtl_eri_clear_bits(tp, 0x1b0, BIT(12)); rtl_pcie_state_l2l3_disable(tp); } @@ -3294,9 +3222,8 @@ static void rtl_hw_start_8168h_1(struct rtl8169_private *tp) rtl_reset_packet_filter(tp); - rtl_eri_set_bits(tp, 0xdc, ERIAR_MASK_1111, BIT(4)); - - rtl_eri_set_bits(tp, 0xd4, ERIAR_MASK_1111, 0x1f00); + rtl_eri_set_bits(tp, 0xd4, 0x1f00); + rtl_eri_set_bits(tp, 0xdc, 0x001c); rtl_eri_write(tp, 0x5f0, ERIAR_MASK_0011, 0x4f87); @@ -3312,7 +3239,7 @@ static void rtl_hw_start_8168h_1(struct rtl8169_private *tp) RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) & ~TX_10M_PS_EN); - rtl_eri_clear_bits(tp, 0x1b0, ERIAR_MASK_0011, BIT(12)); + rtl_eri_clear_bits(tp, 0x1b0, BIT(12)); rtl_pcie_state_l2l3_disable(tp); @@ -3349,7 +3276,7 @@ static void rtl_hw_start_8168ep(struct rtl8169_private *tp) rtl_reset_packet_filter(tp); - rtl_eri_set_bits(tp, 0xd4, ERIAR_MASK_1111, 0x1f80); + rtl_eri_set_bits(tp, 0xd4, 0x1f80); rtl_eri_write(tp, 0x5f0, ERIAR_MASK_0011, 0x4f87); @@ -3360,7 +3287,7 @@ static void rtl_hw_start_8168ep(struct rtl8169_private *tp) rtl8168_config_eee_mac(tp); - rtl_w0w1_eri(tp, 0x2fc, ERIAR_MASK_0001, 0x01, 0x06); + rtl_w0w1_eri(tp, 0x2fc, 0x01, 0x06); RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) & ~TX_10M_PS_EN); @@ -3452,7 +3379,7 @@ static void rtl_hw_start_8117(struct rtl8169_private *tp) rtl_reset_packet_filter(tp); - rtl_eri_set_bits(tp, 0xd4, ERIAR_MASK_1111, 0x1f90); + rtl_eri_set_bits(tp, 0xd4, 0x1f90); rtl_eri_write(tp, 0x5f0, ERIAR_MASK_0011, 0x4f87); @@ -3468,7 +3395,7 @@ static void rtl_hw_start_8117(struct rtl8169_private *tp) RTL_W8(tp, DLLPR, RTL_R8(tp, DLLPR) & ~TX_10M_PS_EN); - rtl_eri_clear_bits(tp, 0x1b0, ERIAR_MASK_0011, BIT(12)); + rtl_eri_clear_bits(tp, 0x1b0, BIT(12)); rtl_pcie_state_l2l3_disable(tp); @@ -3593,7 +3520,7 @@ static void rtl_hw_start_8402(struct rtl8169_private *tp) rtl_reset_packet_filter(tp); rtl_eri_write(tp, 0xc0, ERIAR_MASK_0011, 0x0000); rtl_eri_write(tp, 0xb8, ERIAR_MASK_0011, 0x0000); - rtl_w0w1_eri(tp, 0x0d4, ERIAR_MASK_0011, 0x0e00, 0xff00); + rtl_w0w1_eri(tp, 0x0d4, 0x0e00, 0xff00); /* disable EEE */ rtl_eri_write(tp, 0x1b0, ERIAR_MASK_0011, 0x0000); @@ -3669,7 +3596,7 @@ static void rtl_hw_start_8125_common(struct rtl8169_private *tp) r8168_mac_ocp_write(tp, 0xe098, 0xc302); - rtl_udelay_loop_wait_low(tp, &rtl_mac_ocp_e00e_cond, 1000, 10); + rtl_loop_wait_low(tp, &rtl_mac_ocp_e00e_cond, 1000, 10); rtl8125_config_eee_mac(tp); @@ -3836,7 +3763,7 @@ static void rtl_hw_start_8169(struct rtl8169_private *tp) RTL_W16(tp, CPlusCmd, tp->cp_cmd); - rtl8169_set_magic_reg(tp, tp->mac_version); + rtl8169_set_magic_reg(tp); /* disable interrupt coalescing */ RTL_W16(tp, IntrMitigate, 0x0000); @@ -3846,7 +3773,6 @@ static void rtl_hw_start(struct rtl8169_private *tp) { rtl_unlock_config_regs(tp); - tp->cp_cmd &= CPCMD_MASK; RTL_W16(tp, CPlusCmd, tp->cp_cmd); if (tp->mac_version <= RTL_GIGA_MAC_VER_06) @@ -3868,6 +3794,7 @@ static void rtl_hw_start(struct rtl8169_private *tp) RTL_W8(tp, ChipCmd, CmdTxEnb | CmdRxEnb); rtl_init_rxcfg(tp); rtl_set_tx_config_registers(tp); + rtl_set_rx_config_features(tp, tp->dev->features); rtl_set_rx_mode(tp->dev); rtl_irq_enable(tp); } @@ -3883,21 +3810,14 @@ static int rtl8169_change_mtu(struct net_device *dev, int new_mtu) return 0; } -static inline void rtl8169_make_unusable_by_asic(struct RxDesc *desc) -{ - desc->addr = cpu_to_le64(0x0badbadbadbadbadull); - desc->opts1 &= ~cpu_to_le32(DescOwn | RsvdMask); -} - -static inline void rtl8169_mark_to_asic(struct RxDesc *desc) +static void rtl8169_mark_to_asic(struct RxDesc *desc) { u32 eor = le32_to_cpu(desc->opts1) & RingEnd; desc->opts2 = 0; /* Force memory writes to complete before releasing descriptor */ dma_wmb(); - - desc->opts1 = cpu_to_le32(DescOwn | eor | R8169_RX_BUF_SIZE); + WRITE_ONCE(desc->opts1, cpu_to_le32(DescOwn | eor | R8169_RX_BUF_SIZE)); } static struct page *rtl8169_alloc_rx_data(struct rtl8169_private *tp, @@ -3914,8 +3834,7 @@ static struct page *rtl8169_alloc_rx_data(struct rtl8169_private *tp, mapping = dma_map_page(d, data, 0, R8169_RX_BUF_SIZE, DMA_FROM_DEVICE); if (unlikely(dma_mapping_error(d, mapping))) { - if (net_ratelimit()) - netif_err(tp, drv, tp->dev, "Failed to map RX DMA!\n"); + netdev_err(tp->dev, "Failed to map RX DMA!\n"); __free_pages(data, get_order(R8169_RX_BUF_SIZE)); return NULL; } @@ -3936,15 +3855,11 @@ static void rtl8169_rx_clear(struct rtl8169_private *tp) R8169_RX_BUF_SIZE, DMA_FROM_DEVICE); __free_pages(tp->Rx_databuff[i], get_order(R8169_RX_BUF_SIZE)); tp->Rx_databuff[i] = NULL; - rtl8169_make_unusable_by_asic(tp->RxDescArray + i); + tp->RxDescArray[i].addr = 0; + tp->RxDescArray[i].opts1 = 0; } } -static inline void rtl8169_mark_as_last_descriptor(struct RxDesc *desc) -{ - desc->opts1 |= cpu_to_le32(RingEnd); -} - static int rtl8169_rx_fill(struct rtl8169_private *tp) { unsigned int i; @@ -3960,7 +3875,8 @@ static int rtl8169_rx_fill(struct rtl8169_private *tp) tp->Rx_databuff[i] = data; } - rtl8169_mark_as_last_descriptor(tp->RxDescArray + NUM_RX_DESC - 1); + /* mark as last descriptor in the ring */ + tp->RxDescArray[NUM_RX_DESC - 1].opts1 |= cpu_to_le32(RingEnd); return 0; } @@ -4009,10 +3925,45 @@ static void rtl8169_tx_clear_range(struct rtl8169_private *tp, u32 start, static void rtl8169_tx_clear(struct rtl8169_private *tp) { rtl8169_tx_clear_range(tp, tp->dirty_tx, NUM_TX_DESC); - tp->cur_tx = tp->dirty_tx = 0; netdev_reset_queue(tp->dev); } +static void rtl8169_hw_reset(struct rtl8169_private *tp) +{ + /* Give a racing hard_start_xmit a few cycles to complete. */ + synchronize_rcu(); + + /* Disable interrupts */ + rtl8169_irq_mask_and_ack(tp); + + rtl_rx_close(tp); + + switch (tp->mac_version) { + case RTL_GIGA_MAC_VER_27: + case RTL_GIGA_MAC_VER_28: + case RTL_GIGA_MAC_VER_31: + rtl_loop_wait_low(tp, &rtl_npq_cond, 20, 2000); + break; + case RTL_GIGA_MAC_VER_34 ... RTL_GIGA_MAC_VER_38: + RTL_W8(tp, ChipCmd, RTL_R8(tp, ChipCmd) | StopReq); + rtl_loop_wait_high(tp, &rtl_txcfg_empty_cond, 100, 666); + break; + case RTL_GIGA_MAC_VER_40 ... RTL_GIGA_MAC_VER_61: + rtl_enable_rxdvgate(tp); + fsleep(2000); + break; + default: + RTL_W8(tp, ChipCmd, RTL_R8(tp, ChipCmd) | StopReq); + fsleep(100); + break; + } + + rtl_hw_reset(tp); + + rtl8169_tx_clear(tp); + rtl8169_init_ring_indexes(tp); +} + static void rtl_reset_work(struct rtl8169_private *tp) { struct net_device *dev = tp->dev; @@ -4020,16 +3971,12 @@ static void rtl_reset_work(struct rtl8169_private *tp) napi_disable(&tp->napi); netif_stop_queue(dev); - synchronize_rcu(); rtl8169_hw_reset(tp); for (i = 0; i < NUM_RX_DESC; i++) rtl8169_mark_to_asic(tp->RxDescArray + i); - rtl8169_tx_clear(tp); - rtl8169_init_ring_indexes(tp); - napi_enable(&tp->napi); rtl_hw_start(tp); netif_wake_queue(dev); @@ -4055,7 +4002,7 @@ static int rtl8169_tx_map(struct rtl8169_private *tp, const u32 *opts, u32 len, ret = dma_mapping_error(d, mapping); if (unlikely(ret)) { if (net_ratelimit()) - netif_err(tp, drv, tp->dev, "Failed to map TX data!\n"); + netdev_err(tp->dev, "Failed to map TX data!\n"); return ret; } @@ -4126,25 +4073,20 @@ static bool rtl8169_tso_csum_v2(struct rtl8169_private *tp, struct sk_buff *skb, u32 *opts) { u32 transport_offset = (u32)skb_transport_offset(skb); - u32 mss = skb_shinfo(skb)->gso_size; + struct skb_shared_info *shinfo = skb_shinfo(skb); + u32 mss = shinfo->gso_size; if (mss) { - switch (vlan_get_protocol(skb)) { - case htons(ETH_P_IP): + if (shinfo->gso_type & SKB_GSO_TCPV4) { opts[0] |= TD1_GTSENV4; - break; - - case htons(ETH_P_IPV6): + } else if (shinfo->gso_type & SKB_GSO_TCPV6) { if (skb_cow_head(skb, 0)) return false; tcp_v6_gso_csum_prep(skb); opts[0] |= TD1_GTSENV6; - break; - - default: + } else { WARN_ON_ONCE(1); - break; } opts[0] |= transport_offset << GTTCPHO_SHIFT; @@ -4226,7 +4168,8 @@ static netdev_tx_t rtl8169_start_xmit(struct sk_buff *skb, txd_first = tp->TxDescArray + entry; if (unlikely(!rtl_tx_slots_avail(tp, frags))) { - netif_err(tp, drv, dev, "BUG! Tx Ring full when queue awake!\n"); + if (net_ratelimit()) + netdev_err(dev, "BUG! Tx Ring full when queue awake!\n"); goto err_stop_0; } @@ -4264,8 +4207,8 @@ static netdev_tx_t rtl8169_start_xmit(struct sk_buff *skb, txd_first->opts1 |= cpu_to_le32(DescOwn | FirstFrag); - /* Force all memory writes to complete before notifying device */ - wmb(); + /* rtl_tx needs to see descriptor changes before updated tp->cur_tx */ + smp_wmb(); tp->cur_tx += frags + 1; @@ -4310,6 +4253,37 @@ err_stop_0: return NETDEV_TX_BUSY; } +static unsigned int rtl_last_frag_len(struct sk_buff *skb) +{ + struct skb_shared_info *info = skb_shinfo(skb); + unsigned int nr_frags = info->nr_frags; + + if (!nr_frags) + return UINT_MAX; + + return skb_frag_size(info->frags + nr_frags - 1); +} + +/* Workaround for hw issues with TSO on RTL8168evl */ +static netdev_features_t rtl8168evl_fix_tso(struct sk_buff *skb, + netdev_features_t features) +{ + /* IPv4 header has options field */ + if (vlan_get_protocol(skb) == htons(ETH_P_IP) && + ip_hdrlen(skb) > sizeof(struct iphdr)) + features &= ~NETIF_F_ALL_TSO; + + /* IPv4 TCP header has options field */ + else if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV4 && + tcp_hdrlen(skb) > sizeof(struct tcphdr)) + features &= ~NETIF_F_ALL_TSO; + + else if (rtl_last_frag_len(skb) <= 6) + features &= ~NETIF_F_ALL_TSO; + + return features; +} + static netdev_features_t rtl8169_features_check(struct sk_buff *skb, struct net_device *dev, netdev_features_t features) @@ -4318,6 +4292,9 @@ static netdev_features_t rtl8169_features_check(struct sk_buff *skb, struct rtl8169_private *tp = netdev_priv(dev); if (skb_is_gso(skb)) { + if (tp->mac_version == RTL_GIGA_MAC_VER_34) + features = rtl8168evl_fix_tso(skb, features); + if (transport_offset > GTTCPHO_MAX && rtl_chip_supports_csum_v2(tp)) features &= ~NETIF_F_ALL_TSO; @@ -4354,9 +4331,9 @@ static void rtl8169_pcierr_interrupt(struct net_device *dev) pci_status_errs = pci_status_get_and_clear_errors(pdev); - netif_err(tp, intr, dev, "PCI error (cmd = 0x%04x, status_errs = 0x%04x)\n", - pci_cmd, pci_status_errs); - + if (net_ratelimit()) + netdev_err(dev, "PCI error (cmd = 0x%04x, status_errs = 0x%04x)\n", + pci_cmd, pci_status_errs); /* * The recovery sequence below admits a very elaborated explanation: * - it seems to work; @@ -4452,15 +4429,17 @@ static inline void rtl8169_rx_csum(struct sk_buff *skb, u32 opts1) static int rtl_rx(struct net_device *dev, struct rtl8169_private *tp, u32 budget) { - unsigned int cur_rx, rx_left; - unsigned int count; + unsigned int cur_rx, rx_left, count; + struct device *d = tp_to_dev(tp); cur_rx = tp->cur_rx; for (rx_left = min(budget, NUM_RX_DESC); rx_left > 0; rx_left--, cur_rx++) { - unsigned int entry = cur_rx % NUM_RX_DESC; - const void *rx_buf = page_address(tp->Rx_databuff[entry]); + unsigned int pkt_size, entry = cur_rx % NUM_RX_DESC; struct RxDesc *desc = tp->RxDescArray + entry; + struct sk_buff *skb; + const void *rx_buf; + dma_addr_t addr; u32 status; status = le32_to_cpu(desc->opts1); @@ -4474,69 +4453,65 @@ static int rtl_rx(struct net_device *dev, struct rtl8169_private *tp, u32 budget dma_rmb(); if (unlikely(status & RxRES)) { - netif_info(tp, rx_err, dev, "Rx ERROR. status = %08x\n", - status); + if (net_ratelimit()) + netdev_warn(dev, "Rx ERROR. status = %08x\n", + status); dev->stats.rx_errors++; if (status & (RxRWT | RxRUNT)) dev->stats.rx_length_errors++; if (status & RxCRC) dev->stats.rx_crc_errors++; - if (status & (RxRUNT | RxCRC) && !(status & RxRWT) && - dev->features & NETIF_F_RXALL) { - goto process_pkt; - } - } else { - unsigned int pkt_size; - struct sk_buff *skb; - -process_pkt: - pkt_size = status & GENMASK(13, 0); - if (likely(!(dev->features & NETIF_F_RXFCS))) - pkt_size -= ETH_FCS_LEN; - /* - * The driver does not support incoming fragmented - * frames. They are seen as a symptom of over-mtu - * sized frames. - */ - if (unlikely(rtl8169_fragmented_frame(status))) { - dev->stats.rx_dropped++; - dev->stats.rx_length_errors++; - goto release_descriptor; - } - skb = napi_alloc_skb(&tp->napi, pkt_size); - if (unlikely(!skb)) { - dev->stats.rx_dropped++; + if (!(dev->features & NETIF_F_RXALL)) goto release_descriptor; - } + else if (status & RxRWT || !(status & (RxRUNT | RxCRC))) + goto release_descriptor; + } - dma_sync_single_for_cpu(tp_to_dev(tp), - le64_to_cpu(desc->addr), - pkt_size, DMA_FROM_DEVICE); - prefetch(rx_buf); - skb_copy_to_linear_data(skb, rx_buf, pkt_size); - skb->tail += pkt_size; - skb->len = pkt_size; + pkt_size = status & GENMASK(13, 0); + if (likely(!(dev->features & NETIF_F_RXFCS))) + pkt_size -= ETH_FCS_LEN; - dma_sync_single_for_device(tp_to_dev(tp), - le64_to_cpu(desc->addr), - pkt_size, DMA_FROM_DEVICE); + /* The driver does not support incoming fragmented frames. + * They are seen as a symptom of over-mtu sized frames. + */ + if (unlikely(rtl8169_fragmented_frame(status))) { + dev->stats.rx_dropped++; + dev->stats.rx_length_errors++; + goto release_descriptor; + } - rtl8169_rx_csum(skb, status); - skb->protocol = eth_type_trans(skb, dev); + skb = napi_alloc_skb(&tp->napi, pkt_size); + if (unlikely(!skb)) { + dev->stats.rx_dropped++; + goto release_descriptor; + } - rtl8169_rx_vlan_tag(desc, skb); + addr = le64_to_cpu(desc->addr); + rx_buf = page_address(tp->Rx_databuff[entry]); - if (skb->pkt_type == PACKET_MULTICAST) - dev->stats.multicast++; + dma_sync_single_for_cpu(d, addr, pkt_size, DMA_FROM_DEVICE); + prefetch(rx_buf); + skb_copy_to_linear_data(skb, rx_buf, pkt_size); + skb->tail += pkt_size; + skb->len = pkt_size; + dma_sync_single_for_device(d, addr, pkt_size, DMA_FROM_DEVICE); - napi_gro_receive(&tp->napi, skb); + rtl8169_rx_csum(skb, status); + skb->protocol = eth_type_trans(skb, dev); + + rtl8169_rx_vlan_tag(desc, skb); + + if (skb->pkt_type == PACKET_MULTICAST) + dev->stats.multicast++; + + napi_gro_receive(&tp->napi, skb); + + u64_stats_update_begin(&tp->rx_stats.syncp); + tp->rx_stats.packets++; + tp->rx_stats.bytes += pkt_size; + u64_stats_update_end(&tp->rx_stats.syncp); - u64_stats_update_begin(&tp->rx_stats.syncp); - tp->rx_stats.packets++; - tp->rx_stats.bytes += pkt_size; - u64_stats_update_end(&tp->rx_stats.syncp); - } release_descriptor: rtl8169_mark_to_asic(desc); } @@ -4652,25 +4627,21 @@ static int r8169_phy_connect(struct rtl8169_private *tp) return 0; } -static void rtl8169_down(struct net_device *dev) +static void rtl8169_down(struct rtl8169_private *tp) { - struct rtl8169_private *tp = netdev_priv(dev); + rtl_lock_work(tp); - phy_stop(tp->phydev); + /* Clear all task flags */ + bitmap_zero(tp->wk.flags, RTL_FLAG_MAX); + phy_stop(tp->phydev); napi_disable(&tp->napi); - netif_stop_queue(dev); rtl8169_hw_reset(tp); - /* Give a racing hard_start_xmit a few cycles to complete. */ - synchronize_rcu(); - - rtl8169_tx_clear(tp); - - rtl8169_rx_clear(tp); - rtl_pll_power_down(tp); + + rtl_unlock_work(tp); } static int rtl8169_close(struct net_device *dev) @@ -4683,12 +4654,9 @@ static int rtl8169_close(struct net_device *dev) /* Update counters before going down */ rtl8169_update_counters(tp); - rtl_lock_work(tp); - /* Clear all task flags */ - bitmap_zero(tp->wk.flags, RTL_FLAG_MAX); - - rtl8169_down(dev); - rtl_unlock_work(tp); + netif_stop_queue(dev); + rtl8169_down(tp); + rtl8169_rx_clear(tp); cancel_work_sync(&tp->wk.work); @@ -4766,8 +4734,7 @@ static int rtl_open(struct net_device *dev) rtl_hw_start(tp); - if (!rtl8169_init_counter_offsets(tp)) - netif_warn(tp, hw, dev, "counter reset/update failed\n"); + rtl8169_init_counter_offsets(tp); phy_start(tp->phydev); netif_start_queue(dev); @@ -4843,44 +4810,30 @@ rtl8169_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats) pm_runtime_put_noidle(&pdev->dev); } -static void rtl8169_net_suspend(struct net_device *dev) +static void rtl8169_net_suspend(struct rtl8169_private *tp) { - struct rtl8169_private *tp = netdev_priv(dev); - - if (!netif_running(dev)) + if (!netif_running(tp->dev)) return; - phy_stop(tp->phydev); - netif_device_detach(dev); - - rtl_lock_work(tp); - napi_disable(&tp->napi); - /* Clear all task flags */ - bitmap_zero(tp->wk.flags, RTL_FLAG_MAX); - - rtl_unlock_work(tp); - - rtl_pll_power_down(tp); + netif_device_detach(tp->dev); + rtl8169_down(tp); } #ifdef CONFIG_PM -static int rtl8169_suspend(struct device *device) +static int __maybe_unused rtl8169_suspend(struct device *device) { - struct net_device *dev = dev_get_drvdata(device); - struct rtl8169_private *tp = netdev_priv(dev); + struct rtl8169_private *tp = dev_get_drvdata(device); - rtl8169_net_suspend(dev); + rtl8169_net_suspend(tp); clk_disable_unprepare(tp->clk); return 0; } -static void __rtl8169_resume(struct net_device *dev) +static void __rtl8169_resume(struct rtl8169_private *tp) { - struct rtl8169_private *tp = netdev_priv(dev); - - netif_device_attach(dev); + netif_device_attach(tp->dev); rtl_pll_power_up(tp); rtl8169_init_phy(tp); @@ -4894,34 +4847,32 @@ static void __rtl8169_resume(struct net_device *dev) rtl_unlock_work(tp); } -static int rtl8169_resume(struct device *device) +static int __maybe_unused rtl8169_resume(struct device *device) { - struct net_device *dev = dev_get_drvdata(device); - struct rtl8169_private *tp = netdev_priv(dev); + struct rtl8169_private *tp = dev_get_drvdata(device); - rtl_rar_set(tp, dev->dev_addr); + rtl_rar_set(tp, tp->dev->dev_addr); clk_prepare_enable(tp->clk); - if (netif_running(dev)) - __rtl8169_resume(dev); + if (netif_running(tp->dev)) + __rtl8169_resume(tp); return 0; } static int rtl8169_runtime_suspend(struct device *device) { - struct net_device *dev = dev_get_drvdata(device); - struct rtl8169_private *tp = netdev_priv(dev); + struct rtl8169_private *tp = dev_get_drvdata(device); if (!tp->TxDescArray) return 0; rtl_lock_work(tp); - __rtl8169_set_wol(tp, WAKE_ANY); + __rtl8169_set_wol(tp, WAKE_PHY); rtl_unlock_work(tp); - rtl8169_net_suspend(dev); + rtl8169_net_suspend(tp); /* Update counters before going runtime suspend */ rtl8169_update_counters(tp); @@ -4931,10 +4882,9 @@ static int rtl8169_runtime_suspend(struct device *device) static int rtl8169_runtime_resume(struct device *device) { - struct net_device *dev = dev_get_drvdata(device); - struct rtl8169_private *tp = netdev_priv(dev); + struct rtl8169_private *tp = dev_get_drvdata(device); - rtl_rar_set(tp, dev->dev_addr); + rtl_rar_set(tp, tp->dev->dev_addr); if (!tp->TxDescArray) return 0; @@ -4943,40 +4893,28 @@ static int rtl8169_runtime_resume(struct device *device) __rtl8169_set_wol(tp, tp->saved_wolopts); rtl_unlock_work(tp); - __rtl8169_resume(dev); + __rtl8169_resume(tp); return 0; } static int rtl8169_runtime_idle(struct device *device) { - struct net_device *dev = dev_get_drvdata(device); + struct rtl8169_private *tp = dev_get_drvdata(device); - if (!netif_running(dev) || !netif_carrier_ok(dev)) + if (!netif_running(tp->dev) || !netif_carrier_ok(tp->dev)) pm_schedule_suspend(device, 10000); return -EBUSY; } static const struct dev_pm_ops rtl8169_pm_ops = { - .suspend = rtl8169_suspend, - .resume = rtl8169_resume, - .freeze = rtl8169_suspend, - .thaw = rtl8169_resume, - .poweroff = rtl8169_suspend, - .restore = rtl8169_resume, - .runtime_suspend = rtl8169_runtime_suspend, - .runtime_resume = rtl8169_runtime_resume, - .runtime_idle = rtl8169_runtime_idle, + SET_SYSTEM_SLEEP_PM_OPS(rtl8169_suspend, rtl8169_resume) + SET_RUNTIME_PM_OPS(rtl8169_runtime_suspend, rtl8169_runtime_resume, + rtl8169_runtime_idle) }; -#define RTL8169_PM_OPS (&rtl8169_pm_ops) - -#else /* !CONFIG_PM */ - -#define RTL8169_PM_OPS NULL - -#endif /* !CONFIG_PM */ +#endif /* CONFIG_PM */ static void rtl_wol_shutdown_quirk(struct rtl8169_private *tp) { @@ -4997,13 +4935,12 @@ static void rtl_wol_shutdown_quirk(struct rtl8169_private *tp) static void rtl_shutdown(struct pci_dev *pdev) { - struct net_device *dev = pci_get_drvdata(pdev); - struct rtl8169_private *tp = netdev_priv(dev); + struct rtl8169_private *tp = pci_get_drvdata(pdev); - rtl8169_net_suspend(dev); + rtl8169_net_suspend(tp); /* Restore original MAC address */ - rtl_rar_set(tp, dev->perm_addr); + rtl_rar_set(tp, tp->dev->perm_addr); rtl8169_hw_reset(tp); @@ -5020,24 +4957,20 @@ static void rtl_shutdown(struct pci_dev *pdev) static void rtl_remove_one(struct pci_dev *pdev) { - struct net_device *dev = pci_get_drvdata(pdev); - struct rtl8169_private *tp = netdev_priv(dev); + struct rtl8169_private *tp = pci_get_drvdata(pdev); - if (r8168_check_dash(tp)) - rtl8168_driver_stop(tp); + if (pci_dev_run_wake(pdev)) + pm_runtime_get_noresume(&pdev->dev); - netif_napi_del(&tp->napi); + unregister_netdev(tp->dev); - unregister_netdev(dev); - mdiobus_unregister(tp->phydev->mdio.bus); + if (r8168_check_dash(tp)) + rtl8168_driver_stop(tp); rtl_release_firmware(tp); - if (pci_dev_run_wake(pdev)) - pm_runtime_get_noresume(&pdev->dev); - /* restore original MAC address */ - rtl_rar_set(tp, dev->perm_addr); + rtl_rar_set(tp, tp->dev->perm_addr); } static const struct net_device_ops rtl_netdev_ops = { @@ -5119,9 +5052,9 @@ DECLARE_RTL_COND(rtl_link_list_ready_cond) return RTL_R8(tp, MCU) & LINK_LIST_RDY; } -DECLARE_RTL_COND(rtl_rxtx_empty_cond) +static void r8168g_wait_ll_share_fifo_ready(struct rtl8169_private *tp) { - return (RTL_R8(tp, MCU) & RXTX_EMPTY) == RXTX_EMPTY; + rtl_loop_wait_high(tp, &rtl_link_list_ready_cond, 100, 42); } static int r8169_mdio_read_reg(struct mii_bus *mii_bus, int phyaddr, int phyreg) @@ -5166,20 +5099,19 @@ static int r8169_mdio_register(struct rtl8169_private *tp) new_bus->read = r8169_mdio_read_reg; new_bus->write = r8169_mdio_write_reg; - ret = mdiobus_register(new_bus); + ret = devm_mdiobus_register(new_bus); if (ret) return ret; tp->phydev = mdiobus_get_phy(new_bus, 0); if (!tp->phydev) { - mdiobus_unregister(new_bus); return -ENODEV; } else if (!tp->phydev->drv) { /* Most chip versions fail with the genphy driver. * Therefore ensure that the dedicated PHY driver is loaded. */ - dev_err(&pdev->dev, "realtek.ko not loaded, maybe it needs to be added to initramfs?\n"); - mdiobus_unregister(new_bus); + dev_err(&pdev->dev, "no dedicated PHY driver found for PHY ID 0x%08x, maybe realtek.ko needs to be added to initramfs?\n", + tp->phydev->phy_id); return -EUNATCH; } @@ -5191,53 +5123,34 @@ static int r8169_mdio_register(struct rtl8169_private *tp) static void rtl_hw_init_8168g(struct rtl8169_private *tp) { - tp->ocp_base = OCP_STD_PHY_BASE; - - RTL_W32(tp, MISC, RTL_R32(tp, MISC) | RXDV_GATED_EN); - - if (!rtl_udelay_loop_wait_high(tp, &rtl_txcfg_empty_cond, 100, 42)) - return; - - if (!rtl_udelay_loop_wait_high(tp, &rtl_rxtx_empty_cond, 100, 42)) - return; + rtl_enable_rxdvgate(tp); RTL_W8(tp, ChipCmd, RTL_R8(tp, ChipCmd) & ~(CmdTxEnb | CmdRxEnb)); msleep(1); RTL_W8(tp, MCU, RTL_R8(tp, MCU) & ~NOW_IS_OOB); r8168_mac_ocp_modify(tp, 0xe8de, BIT(14), 0); - - if (!rtl_udelay_loop_wait_high(tp, &rtl_link_list_ready_cond, 100, 42)) - return; + r8168g_wait_ll_share_fifo_ready(tp); r8168_mac_ocp_modify(tp, 0xe8de, 0, BIT(15)); - - rtl_udelay_loop_wait_high(tp, &rtl_link_list_ready_cond, 100, 42); + r8168g_wait_ll_share_fifo_ready(tp); } static void rtl_hw_init_8125(struct rtl8169_private *tp) { - tp->ocp_base = OCP_STD_PHY_BASE; - - RTL_W32(tp, MISC, RTL_R32(tp, MISC) | RXDV_GATED_EN); - - if (!rtl_udelay_loop_wait_high(tp, &rtl_rxtx_empty_cond, 100, 42)) - return; + rtl_enable_rxdvgate(tp); RTL_W8(tp, ChipCmd, RTL_R8(tp, ChipCmd) & ~(CmdTxEnb | CmdRxEnb)); msleep(1); RTL_W8(tp, MCU, RTL_R8(tp, MCU) & ~NOW_IS_OOB); r8168_mac_ocp_modify(tp, 0xe8de, BIT(14), 0); - - if (!rtl_udelay_loop_wait_high(tp, &rtl_link_list_ready_cond, 100, 42)) - return; + r8168g_wait_ll_share_fifo_ready(tp); r8168_mac_ocp_write(tp, 0xc0aa, 0x07d0); r8168_mac_ocp_write(tp, 0xc0a6, 0x0150); r8168_mac_ocp_write(tp, 0xc01e, 0x5555); - - rtl_udelay_loop_wait_high(tp, &rtl_link_list_ready_cond, 100, 42); + r8168g_wait_ll_share_fifo_ready(tp); } static void rtl_hw_initialize(struct rtl8169_private *tp) @@ -5352,9 +5265,9 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) tp = netdev_priv(dev); tp->dev = dev; tp->pci_dev = pdev; - tp->msg_enable = netif_msg_init(debug.msg_enable, R8169_MSG_DEFAULT); tp->supports_gmii = ent->driver_data == RTL_CFG_NO_GBIT ? 0 : 1; tp->eee_adv = -1; + tp->ocp_base = OCP_STD_PHY_BASE; /* Get the *optional* external "ether_clk" used on some boards */ rc = rtl_get_ether_clk(tp); @@ -5410,7 +5323,7 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) tp->mac_version = chipset; - tp->cp_cmd = RTL_R16(tp, CPlusCmd); + tp->cp_cmd = RTL_R16(tp, CPlusCmd) & CPCMD_MASK; if (sizeof(dma_addr_t) > 4 && tp->mac_version >= RTL_GIGA_MAC_VER_18 && !dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64))) @@ -5445,14 +5358,9 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) dev->hw_features = NETIF_F_IP_CSUM | NETIF_F_RXCSUM | NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_CTAG_RX; - dev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO | - NETIF_F_HIGHDMA; + dev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO; dev->priv_flags |= IFF_LIVE_ADDR_CHANGE; - tp->cp_cmd |= RxChkSum; - /* RTL8125 uses register RxConfig for VLAN offloading config */ - if (!rtl_is_8125(tp)) - tp->cp_cmd |= RxVlan; /* * Pretend we are using VLANs; This bypasses a nasty bug where * Interrupts stop flowing on high load on 8110SCd controllers. @@ -5484,6 +5392,9 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) dev->hw_features |= NETIF_F_RXALL; dev->hw_features |= NETIF_F_RXFCS; + /* configure chip for default features */ + rtl8169_set_features(dev, dev->features); + jumbo_max = rtl_jumbo_max(tp); if (jumbo_max) dev->max_mtu = jumbo_max; @@ -5498,7 +5409,7 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (!tp->counters) return -ENOMEM; - pci_set_drvdata(pdev, dev); + pci_set_drvdata(pdev, tp); rc = r8169_mdio_register(tp); if (rc) @@ -5509,17 +5420,16 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) rc = register_netdev(dev); if (rc) - goto err_mdio_unregister; + return rc; - netif_info(tp, probe, dev, "%s, %pM, XID %03x, IRQ %d\n", - rtl_chip_infos[chipset].name, dev->dev_addr, xid, - pci_irq_vector(pdev, 0)); + netdev_info(dev, "%s, %pM, XID %03x, IRQ %d\n", + rtl_chip_infos[chipset].name, dev->dev_addr, xid, + pci_irq_vector(pdev, 0)); if (jumbo_max) - netif_info(tp, probe, dev, - "jumbo features [frames: %d bytes, tx checksumming: %s]\n", - jumbo_max, tp->mac_version <= RTL_GIGA_MAC_VER_06 ? - "ok" : "ko"); + netdev_info(dev, "jumbo features [frames: %d bytes, tx checksumming: %s]\n", + jumbo_max, tp->mac_version <= RTL_GIGA_MAC_VER_06 ? + "ok" : "ko"); if (r8168_check_dash(tp)) rtl8168_driver_start(tp); @@ -5528,10 +5438,6 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) pm_runtime_put_sync(&pdev->dev); return 0; - -err_mdio_unregister: - mdiobus_unregister(tp->phydev->mdio.bus); - return rc; } static struct pci_driver rtl8169_pci_driver = { @@ -5540,7 +5446,9 @@ static struct pci_driver rtl8169_pci_driver = { .probe = rtl_init_one, .remove = rtl_remove_one, .shutdown = rtl_shutdown, - .driver.pm = RTL8169_PM_OPS, +#ifdef CONFIG_PM + .driver.pm = &rtl8169_pm_ops, +#endif }; module_pci_driver(rtl8169_pci_driver); diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c index 067ad25553b9..a442bcf64b9c 100644 --- a/drivers/net/ethernet/renesas/ravb_main.c +++ b/drivers/net/ethernet/renesas/ravb_main.c @@ -1014,6 +1014,7 @@ static int ravb_phy_init(struct net_device *ndev) struct ravb_private *priv = netdev_priv(ndev); struct phy_device *phydev; struct device_node *pn; + phy_interface_t iface; int err; priv->link = 0; @@ -1032,8 +1033,13 @@ static int ravb_phy_init(struct net_device *ndev) } pn = of_node_get(np); } - phydev = of_phy_connect(ndev, pn, ravb_adjust_link, 0, - priv->phy_interface); + + iface = priv->phy_interface; + if (priv->chip_id != RCAR_GEN2 && phy_interface_mode_is_rgmii(iface)) { + /* ravb_set_delay_mode() takes care of internal delay mode */ + iface = PHY_INTERFACE_MODE_RGMII; + } + phydev = of_phy_connect(ndev, pn, ravb_adjust_link, 0, iface); of_node_put(pn); if (!phydev) { netdev_err(ndev, "failed to connect PHY\n"); diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c index 8ed73f44405d..f45331ed90b0 100644 --- a/drivers/net/ethernet/renesas/sh_eth.c +++ b/drivers/net/ethernet/renesas/sh_eth.c @@ -2472,7 +2472,8 @@ static void sh_eth_tx_timeout(struct net_device *ndev, unsigned int txqueue) } /* Packet transmit function */ -static int sh_eth_start_xmit(struct sk_buff *skb, struct net_device *ndev) +static netdev_tx_t sh_eth_start_xmit(struct sk_buff *skb, + struct net_device *ndev) { struct sh_eth_private *mdp = netdev_priv(ndev); struct sh_eth_txdesc *txdesc; diff --git a/drivers/net/ethernet/seeq/ether3.c b/drivers/net/ethernet/seeq/ether3.c index 128ee7cda1ed..65c98837ec45 100644 --- a/drivers/net/ethernet/seeq/ether3.c +++ b/drivers/net/ethernet/seeq/ether3.c @@ -610,12 +610,9 @@ static int ether3_rx(struct net_device *dev, unsigned int maxcnt) ether3_readbuffer(dev, addrs+2, 12); if (next_ptr < RX_START || next_ptr >= RX_END) { - int i; printk("%s: bad next pointer @%04X: ", dev->name, priv(dev)->rx_head); printk("%02X %02X %02X %02X ", next_ptr >> 8, next_ptr & 255, status & 255, status >> 8); - for (i = 2; i < 14; i++) - printk("%02X ", addrs[i]); - printk("\n"); + printk("%pM %pM\n", addrs + 2, addrs + 8); next_ptr = priv(dev)->rx_head; break; } diff --git a/drivers/net/ethernet/sfc/ef10.c b/drivers/net/ethernet/sfc/ef10.c index 3f16bd807c6e..4b0e3695a71a 100644 --- a/drivers/net/ethernet/sfc/ef10.c +++ b/drivers/net/ethernet/sfc/ef10.c @@ -553,7 +553,7 @@ static int efx_ef10_probe(struct efx_nic *efx) efx->rss_context.context_id = EFX_MCDI_RSS_CONTEXT_INVALID; - nic_data->vport_id = EVB_PORT_ID_ASSIGNED; + efx->vport_id = EVB_PORT_ID_ASSIGNED; /* In case we're recovering from a crash (kexec), we want to * cancel any outstanding request by the previous user of this @@ -1281,13 +1281,13 @@ static int efx_ef10_init_nic(struct efx_nic *efx) nic_data->must_check_datapath_caps = false; } - if (nic_data->must_realloc_vis) { + if (efx->must_realloc_vis) { /* We cannot let the number of VIs change now */ rc = efx_ef10_alloc_vis(efx, nic_data->n_allocated_vis, nic_data->n_allocated_vis); if (rc) return rc; - nic_data->must_realloc_vis = false; + efx->must_realloc_vis = false; } if (nic_data->must_restore_piobufs && nic_data->n_piobufs) { @@ -1326,16 +1326,15 @@ static void efx_ef10_table_reset_mc_allocations(struct efx_nic *efx) #endif /* All our allocations have been reset */ - nic_data->must_realloc_vis = true; - nic_data->must_restore_rss_contexts = true; - nic_data->must_restore_filters = true; + efx->must_realloc_vis = true; + efx_mcdi_filter_table_reset_mc_allocations(efx); nic_data->must_restore_piobufs = true; efx_ef10_forget_old_piobufs(efx); efx->rss_context.context_id = EFX_MCDI_RSS_CONTEXT_INVALID; /* Driver-created vswitches and vports must be re-created */ nic_data->must_probe_vswitching = true; - nic_data->vport_id = EVB_PORT_ID_ASSIGNED; + efx->vport_id = EVB_PORT_ID_ASSIGNED; #ifdef CONFIG_SFC_SRIOV if (nic_data->vf) for (i = 0; i < efx->vf_count; i++) @@ -1820,6 +1819,7 @@ static size_t efx_ef10_update_stats_pf(struct efx_nic *efx, u64 *full_stats, } static int efx_ef10_try_update_nic_stats_vf(struct efx_nic *efx) + __must_hold(&efx->stats_lock) { MCDI_DECLARE_BUF(inbuf, MC_CMD_MAC_STATS_IN_LEN); struct efx_ef10_nic_data *nic_data = efx->nic_data; @@ -2389,6 +2389,86 @@ static void efx_ef10_tx_write(struct efx_tx_queue *tx_queue) } } +static int efx_ef10_probe_multicast_chaining(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + unsigned int enabled, implemented; + bool want_workaround_26807; + int rc; + + rc = efx_mcdi_get_workarounds(efx, &implemented, &enabled); + if (rc == -ENOSYS) { + /* GET_WORKAROUNDS was implemented before this workaround, + * thus it must be unavailable in this firmware. + */ + nic_data->workaround_26807 = false; + return 0; + } + if (rc) + return rc; + want_workaround_26807 = + implemented & MC_CMD_GET_WORKAROUNDS_OUT_BUG26807; + nic_data->workaround_26807 = + !!(enabled & MC_CMD_GET_WORKAROUNDS_OUT_BUG26807); + + if (want_workaround_26807 && !nic_data->workaround_26807) { + unsigned int flags; + + rc = efx_mcdi_set_workaround(efx, + MC_CMD_WORKAROUND_BUG26807, + true, &flags); + if (!rc) { + if (flags & + 1 << MC_CMD_WORKAROUND_EXT_OUT_FLR_DONE_LBN) { + netif_info(efx, drv, efx->net_dev, + "other functions on NIC have been reset\n"); + + /* With MCFW v4.6.x and earlier, the + * boot count will have incremented, + * so re-read the warm_boot_count + * value now to ensure this function + * doesn't think it has changed next + * time it checks. + */ + rc = efx_ef10_get_warm_boot_count(efx); + if (rc >= 0) { + nic_data->warm_boot_count = rc; + rc = 0; + } + } + nic_data->workaround_26807 = true; + } else if (rc == -EPERM) { + rc = 0; + } + } + return rc; +} + +static int efx_ef10_filter_table_probe(struct efx_nic *efx) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + int rc = efx_ef10_probe_multicast_chaining(efx); + struct efx_mcdi_filter_vlan *vlan; + + if (rc) + return rc; + rc = efx_mcdi_filter_table_probe(efx, nic_data->workaround_26807); + + if (rc) + return rc; + + list_for_each_entry(vlan, &nic_data->vlan_list, list) { + rc = efx_mcdi_filter_add_vlan(efx, vlan->vid); + if (rc) + goto fail_add_vlan; + } + return 0; + +fail_add_vlan: + efx_mcdi_filter_table_remove(efx); + return rc; +} + /* This creates an entry in the RX descriptor queue */ static inline void efx_ef10_build_rx_desc(struct efx_rx_queue *rx_queue, unsigned int index) @@ -2464,75 +2544,14 @@ static int efx_ef10_ev_init(struct efx_channel *channel) { struct efx_nic *efx = channel->efx; struct efx_ef10_nic_data *nic_data; - unsigned int enabled, implemented; bool use_v2, cut_thru; - int rc; nic_data = efx->nic_data; use_v2 = nic_data->datapath_caps2 & 1 << MC_CMD_GET_CAPABILITIES_V2_OUT_INIT_EVQ_V2_LBN; cut_thru = !(nic_data->datapath_caps & 1 << MC_CMD_GET_CAPABILITIES_OUT_RX_BATCHING_LBN); - rc = efx_mcdi_ev_init(channel, cut_thru, use_v2); - - /* IRQ return is ignored */ - if (channel->channel || rc) - return rc; - - /* Successfully created event queue on channel 0 */ - rc = efx_mcdi_get_workarounds(efx, &implemented, &enabled); - if (rc == -ENOSYS) { - /* GET_WORKAROUNDS was implemented before this workaround, - * thus it must be unavailable in this firmware. - */ - nic_data->workaround_26807 = false; - rc = 0; - } else if (rc) { - goto fail; - } else { - nic_data->workaround_26807 = - !!(enabled & MC_CMD_GET_WORKAROUNDS_OUT_BUG26807); - - if (implemented & MC_CMD_GET_WORKAROUNDS_OUT_BUG26807 && - !nic_data->workaround_26807) { - unsigned int flags; - - rc = efx_mcdi_set_workaround(efx, - MC_CMD_WORKAROUND_BUG26807, - true, &flags); - - if (!rc) { - if (flags & - 1 << MC_CMD_WORKAROUND_EXT_OUT_FLR_DONE_LBN) { - netif_info(efx, drv, efx->net_dev, - "other functions on NIC have been reset\n"); - - /* With MCFW v4.6.x and earlier, the - * boot count will have incremented, - * so re-read the warm_boot_count - * value now to ensure this function - * doesn't think it has changed next - * time it checks. - */ - rc = efx_ef10_get_warm_boot_count(efx); - if (rc >= 0) { - nic_data->warm_boot_count = rc; - rc = 0; - } - } - nic_data->workaround_26807 = true; - } else if (rc == -EPERM) { - rc = 0; - } - } - } - - if (!rc) - return 0; - -fail: - efx_mcdi_ev_fini(channel); - return rc; + return efx_mcdi_ev_init(channel, cut_thru, use_v2); } static void efx_ef10_handle_rx_wrong_queue(struct efx_rx_queue *rx_queue, @@ -3100,16 +3119,15 @@ void efx_ef10_handle_drain_event(struct efx_nic *efx) static int efx_ef10_fini_dmaq(struct efx_nic *efx) { - struct efx_ef10_nic_data *nic_data = efx->nic_data; - struct efx_channel *channel; struct efx_tx_queue *tx_queue; struct efx_rx_queue *rx_queue; + struct efx_channel *channel; int pending; /* If the MC has just rebooted, the TX/RX queues will have already been * torn down, but efx->active_queues needs to be set to zero. */ - if (nic_data->must_realloc_vis) { + if (efx->must_realloc_vis) { atomic_set(&efx->active_queues, 0); return 0; } @@ -3158,22 +3176,22 @@ static int efx_ef10_vport_set_mac_address(struct efx_nic *efx) efx_mcdi_filter_table_remove(efx); up_write(&efx->filter_sem); - rc = efx_ef10_vadaptor_free(efx, nic_data->vport_id); + rc = efx_ef10_vadaptor_free(efx, efx->vport_id); if (rc) goto restore_filters; ether_addr_copy(mac_old, nic_data->vport_mac); - rc = efx_ef10_vport_del_mac(efx, nic_data->vport_id, + rc = efx_ef10_vport_del_mac(efx, efx->vport_id, nic_data->vport_mac); if (rc) goto restore_vadaptor; - rc = efx_ef10_vport_add_mac(efx, nic_data->vport_id, + rc = efx_ef10_vport_add_mac(efx, efx->vport_id, efx->net_dev->dev_addr); if (!rc) { ether_addr_copy(nic_data->vport_mac, efx->net_dev->dev_addr); } else { - rc2 = efx_ef10_vport_add_mac(efx, nic_data->vport_id, mac_old); + rc2 = efx_ef10_vport_add_mac(efx, efx->vport_id, mac_old); if (rc2) { /* Failed to add original MAC, so clear vport_mac */ eth_zero_addr(nic_data->vport_mac); @@ -3182,12 +3200,12 @@ static int efx_ef10_vport_set_mac_address(struct efx_nic *efx) } restore_vadaptor: - rc2 = efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); + rc2 = efx_ef10_vadaptor_alloc(efx, efx->vport_id); if (rc2) goto reset_nic; restore_filters: down_write(&efx->filter_sem); - rc2 = efx_mcdi_filter_table_probe(efx); + rc2 = efx_ef10_filter_table_probe(efx); up_write(&efx->filter_sem); if (rc2) goto reset_nic; @@ -3211,7 +3229,6 @@ reset_nic: static int efx_ef10_set_mac_address(struct efx_nic *efx) { MCDI_DECLARE_BUF(inbuf, MC_CMD_VADAPTOR_SET_MAC_IN_LEN); - struct efx_ef10_nic_data *nic_data = efx->nic_data; bool was_enabled = efx->port_enabled; int rc; @@ -3225,11 +3242,11 @@ static int efx_ef10_set_mac_address(struct efx_nic *efx) ether_addr_copy(MCDI_PTR(inbuf, VADAPTOR_SET_MAC_IN_MACADDR), efx->net_dev->dev_addr); MCDI_SET_DWORD(inbuf, VADAPTOR_SET_MAC_IN_UPSTREAM_PORT_ID, - nic_data->vport_id); + efx->vport_id); rc = efx_mcdi_rpc_quiet(efx, MC_CMD_VADAPTOR_SET_MAC, inbuf, sizeof(inbuf), NULL, 0, NULL); - efx_mcdi_filter_table_probe(efx); + efx_ef10_filter_table_probe(efx); up_write(&efx->filter_sem); mutex_unlock(&efx->mac_lock); @@ -3239,6 +3256,7 @@ static int efx_ef10_set_mac_address(struct efx_nic *efx) #ifdef CONFIG_SFC_SRIOV if (efx->pci_dev->is_virtfn && efx->pci_dev->physfn) { + struct efx_ef10_nic_data *nic_data = efx->nic_data; struct pci_dev *pci_dev_pf = efx->pci_dev->physfn; if (rc == -EPERM) { @@ -3961,6 +3979,35 @@ out_unlock: return rc; } +/* EF10 may have multiple datapath firmware variants within a + * single version. Report which variants are running. + */ +static size_t efx_ef10_print_additional_fwver(struct efx_nic *efx, char *buf, + size_t len) +{ + struct efx_ef10_nic_data *nic_data = efx->nic_data; + + return scnprintf(buf, len, " rx%x tx%x", + nic_data->rx_dpcpu_fw_id, + nic_data->tx_dpcpu_fw_id); +} + +static unsigned int ef10_check_caps(const struct efx_nic *efx, + u8 flag, + u32 offset) +{ + const struct efx_ef10_nic_data *nic_data = efx->nic_data; + + switch (offset) { + case(MC_CMD_GET_CAPABILITIES_V4_OUT_FLAGS1_OFST): + return nic_data->datapath_caps & BIT_ULL(flag); + case(MC_CMD_GET_CAPABILITIES_V4_OUT_FLAGS2_OFST): + return nic_data->datapath_caps2 & BIT_ULL(flag); + default: + return 0; + } +} + #define EF10_OFFLOAD_FEATURES \ (NETIF_F_IP_CSUM | \ NETIF_F_HW_VLAN_CTAG_FILTER | \ @@ -4027,7 +4074,7 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = { .ev_process = efx_ef10_ev_process, .ev_read_ack = efx_ef10_ev_read_ack, .ev_test_generate = efx_ef10_ev_test_generate, - .filter_table_probe = efx_mcdi_filter_table_probe, + .filter_table_probe = efx_ef10_filter_table_probe, .filter_table_restore = efx_mcdi_filter_table_restore, .filter_table_remove = efx_mcdi_filter_table_remove, .filter_update_rx_scatter = efx_mcdi_update_rx_scatter, @@ -4073,6 +4120,8 @@ const struct efx_nic_type efx_hunt_a0_vf_nic_type = { .hwtstamp_filters = 1 << HWTSTAMP_FILTER_NONE | 1 << HWTSTAMP_FILTER_ALL, .rx_hash_key_size = 40, + .check_caps = ef10_check_caps, + .print_additional_fwver = efx_ef10_print_additional_fwver, }; const struct efx_nic_type efx_hunt_a0_nic_type = { @@ -4139,7 +4188,7 @@ const struct efx_nic_type efx_hunt_a0_nic_type = { .ev_process = efx_ef10_ev_process, .ev_read_ack = efx_ef10_ev_read_ack, .ev_test_generate = efx_ef10_ev_test_generate, - .filter_table_probe = efx_mcdi_filter_table_probe, + .filter_table_probe = efx_ef10_filter_table_probe, .filter_table_restore = efx_mcdi_filter_table_restore, .filter_table_remove = efx_mcdi_filter_table_remove, .filter_update_rx_scatter = efx_mcdi_update_rx_scatter, @@ -4208,4 +4257,6 @@ const struct efx_nic_type efx_hunt_a0_nic_type = { .hwtstamp_filters = 1 << HWTSTAMP_FILTER_NONE | 1 << HWTSTAMP_FILTER_ALL, .rx_hash_key_size = 40, + .check_caps = ef10_check_caps, + .print_additional_fwver = efx_ef10_print_additional_fwver, }; diff --git a/drivers/net/ethernet/sfc/ef10_sriov.c b/drivers/net/ethernet/sfc/ef10_sriov.c index 4580b30caae1..21fa6c0e8873 100644 --- a/drivers/net/ethernet/sfc/ef10_sriov.c +++ b/drivers/net/ethernet/sfc/ef10_sriov.c @@ -232,15 +232,14 @@ fail: static int efx_ef10_vadaptor_alloc_set_features(struct efx_nic *efx) { - struct efx_ef10_nic_data *nic_data = efx->nic_data; u32 port_flags; int rc; - rc = efx_ef10_vadaptor_alloc(efx, nic_data->vport_id); + rc = efx_ef10_vadaptor_alloc(efx, efx->vport_id); if (rc) goto fail_vadaptor_alloc; - rc = efx_ef10_vadaptor_query(efx, nic_data->vport_id, + rc = efx_ef10_vadaptor_query(efx, efx->vport_id, &port_flags, NULL, NULL); if (rc) goto fail_vadaptor_query; @@ -281,11 +280,11 @@ int efx_ef10_vswitching_probe_pf(struct efx_nic *efx) rc = efx_ef10_vport_alloc(efx, EVB_PORT_ID_ASSIGNED, MC_CMD_VPORT_ALLOC_IN_VPORT_TYPE_NORMAL, - EFX_EF10_NO_VLAN, &nic_data->vport_id); + EFX_EF10_NO_VLAN, &efx->vport_id); if (rc) goto fail2; - rc = efx_ef10_vport_add_mac(efx, nic_data->vport_id, net_dev->dev_addr); + rc = efx_ef10_vport_add_mac(efx, efx->vport_id, net_dev->dev_addr); if (rc) goto fail3; ether_addr_copy(nic_data->vport_mac, net_dev->dev_addr); @@ -296,11 +295,11 @@ int efx_ef10_vswitching_probe_pf(struct efx_nic *efx) return 0; fail4: - efx_ef10_vport_del_mac(efx, nic_data->vport_id, nic_data->vport_mac); + efx_ef10_vport_del_mac(efx, efx->vport_id, nic_data->vport_mac); eth_zero_addr(nic_data->vport_mac); fail3: - efx_ef10_vport_free(efx, nic_data->vport_id); - nic_data->vport_id = EVB_PORT_ID_ASSIGNED; + efx_ef10_vport_free(efx, efx->vport_id); + efx->vport_id = EVB_PORT_ID_ASSIGNED; fail2: efx_ef10_vswitch_free(efx, EVB_PORT_ID_ASSIGNED); fail1: @@ -355,22 +354,22 @@ void efx_ef10_vswitching_remove_pf(struct efx_nic *efx) efx_ef10_sriov_free_vf_vswitching(efx); - efx_ef10_vadaptor_free(efx, nic_data->vport_id); + efx_ef10_vadaptor_free(efx, efx->vport_id); - if (nic_data->vport_id == EVB_PORT_ID_ASSIGNED) + if (efx->vport_id == EVB_PORT_ID_ASSIGNED) return; /* No vswitch was ever created */ if (!is_zero_ether_addr(nic_data->vport_mac)) { - efx_ef10_vport_del_mac(efx, nic_data->vport_id, + efx_ef10_vport_del_mac(efx, efx->vport_id, efx->net_dev->dev_addr); eth_zero_addr(nic_data->vport_mac); } - efx_ef10_vport_free(efx, nic_data->vport_id); - nic_data->vport_id = EVB_PORT_ID_ASSIGNED; + efx_ef10_vport_free(efx, efx->vport_id); + efx->vport_id = EVB_PORT_ID_ASSIGNED; /* Only free the vswitch if no VFs are assigned */ if (!pci_vfs_assigned(efx->pci_dev)) - efx_ef10_vswitch_free(efx, nic_data->vport_id); + efx_ef10_vswitch_free(efx, efx->vport_id); } void efx_ef10_vswitching_remove_vf(struct efx_nic *efx) diff --git a/drivers/net/ethernet/sfc/mcdi.c b/drivers/net/ethernet/sfc/mcdi.c index 15c731d04065..a8cc3881edce 100644 --- a/drivers/net/ethernet/sfc/mcdi.c +++ b/drivers/net/ethernet/sfc/mcdi.c @@ -1425,23 +1425,16 @@ void efx_mcdi_print_fwver(struct efx_nic *efx, char *buf, size_t len) le16_to_cpu(ver_words[2]), le16_to_cpu(ver_words[3])); - /* EF10 may have multiple datapath firmware variants within a - * single version. Report which variants are running. - */ - if (efx_nic_rev(efx) >= EFX_REV_HUNT_A0) { - struct efx_ef10_nic_data *nic_data = efx->nic_data; - - offset += scnprintf(buf + offset, len - offset, " rx%x tx%x", - nic_data->rx_dpcpu_fw_id, - nic_data->tx_dpcpu_fw_id); + if (efx->type->print_additional_fwver) + offset += efx->type->print_additional_fwver(efx, buf + offset, + len - offset); - /* It's theoretically possible for the string to exceed 31 - * characters, though in practice the first three version - * components are short enough that this doesn't happen. - */ - if (WARN_ON(offset >= len)) - buf[0] = 0; - } + /* It's theoretically possible for the string to exceed 31 + * characters, though in practice the first three version + * components are short enough that this doesn't happen. + */ + if (WARN_ON(offset >= len)) + buf[0] = 0; return; diff --git a/drivers/net/ethernet/sfc/mcdi.h b/drivers/net/ethernet/sfc/mcdi.h index 54a45010b576..b107e4c00285 100644 --- a/drivers/net/ethernet/sfc/mcdi.h +++ b/drivers/net/ethernet/sfc/mcdi.h @@ -326,6 +326,18 @@ void efx_mcdi_sensor_event(struct efx_nic *efx, efx_qword_t *ev); #define MCDI_EVENT_FIELD(_ev, _field) \ EFX_QWORD_FIELD(_ev, MCDI_EVENT_ ## _field) +#define MCDI_CAPABILITY(field) \ + MC_CMD_GET_CAPABILITIES_V4_OUT_ ## field ## _LBN + +#define MCDI_CAPABILITY_OFST(field) \ + MC_CMD_GET_CAPABILITIES_V4_OUT_ ## field ## _OFST + +/* field is FLAGS1 or FLAGS2 */ +#define efx_has_cap(efx, flag, field) \ + efx->type->check_caps(efx, \ + MCDI_CAPABILITY(flag), \ + MCDI_CAPABILITY_OFST(field)) + void efx_mcdi_print_fwver(struct efx_nic *efx, char *buf, size_t len); int efx_mcdi_get_board_cfg(struct efx_nic *efx, u8 *mac_address, u16 *fw_subtype_list, u32 *capabilities); diff --git a/drivers/net/ethernet/sfc/mcdi_filters.c b/drivers/net/ethernet/sfc/mcdi_filters.c index 4310ae5bd898..455a62814fb9 100644 --- a/drivers/net/ethernet/sfc/mcdi_filters.c +++ b/drivers/net/ethernet/sfc/mcdi_filters.c @@ -186,7 +186,6 @@ static void efx_mcdi_filter_push_prep(struct efx_nic *efx, struct efx_rss_context *ctx, bool replacing) { - struct efx_ef10_nic_data *nic_data = efx->nic_data; u32 flags = spec->flags; memset(inbuf, 0, MC_CMD_FILTER_OP_EXT_IN_LEN); @@ -211,7 +210,7 @@ static void efx_mcdi_filter_push_prep(struct efx_nic *efx, efx_mcdi_filter_push_prep_set_match_fields(efx, spec, inbuf); } - MCDI_SET_DWORD(inbuf, FILTER_OP_IN_PORT_ID, nic_data->vport_id); + MCDI_SET_DWORD(inbuf, FILTER_OP_IN_PORT_ID, efx->vport_id); MCDI_SET_DWORD(inbuf, FILTER_OP_IN_RX_DEST, spec->dmaq_id == EFX_FILTER_RX_DMAQ_ID_DROP ? MC_CMD_FILTER_OP_IN_RX_DEST_DROP : @@ -332,7 +331,6 @@ static s32 efx_mcdi_filter_insert_locked(struct efx_nic *efx, bool replace_equal) { DECLARE_BITMAP(mc_rem_map, EFX_EF10_FILTER_SEARCH_LIMIT); - struct efx_ef10_nic_data *nic_data = efx->nic_data; struct efx_mcdi_filter_table *table; struct efx_filter_spec *saved_spec; struct efx_rss_context *ctx = NULL; @@ -461,7 +459,7 @@ static s32 efx_mcdi_filter_insert_locked(struct efx_nic *efx, rc = efx_mcdi_filter_push(efx, spec, &table->entry[ins_index].handle, ctx, replacing); - if (rc == -EINVAL && nic_data->must_realloc_vis) + if (rc == -EINVAL && efx->must_realloc_vis) /* The MC rebooted under us, causing it to reject our filter * insertion as pointing to an invalid VI (spec->dmaq_id). */ @@ -813,7 +811,7 @@ static int efx_mcdi_filter_insert_def(struct efx_nic *efx, enum efx_encap_type encap_type, bool multicast, bool rollback) { - struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct efx_mcdi_filter_table *table = efx->filter_state; enum efx_filter_flags filter_flags; struct efx_filter_spec spec; u8 baddr[ETH_ALEN]; @@ -830,8 +828,7 @@ static int efx_mcdi_filter_insert_def(struct efx_nic *efx, efx_filter_set_uc_def(&spec); if (encap_type) { - if (nic_data->datapath_caps & - (1 << MC_CMD_GET_CAPABILITIES_OUT_VXLAN_NVGRE_LBN)) + if (efx_has_cap(efx, VXLAN_NVGRE, FLAGS1)) efx_filter_set_encap_type(&spec, encap_type); else /* @@ -899,7 +896,7 @@ static int efx_mcdi_filter_insert_def(struct efx_nic *efx, EFX_WARN_ON_PARANOID(*id != EFX_EF10_FILTER_ID_INVALID); *id = efx_mcdi_filter_get_unsafe_id(rc); - if (!nic_data->workaround_26807 && !encap_type) { + if (!table->mc_chaining && !encap_type) { /* Also need an Ethernet broadcast filter */ efx_filter_init_rx(&spec, EFX_FILTER_PRI_AUTO, filter_flags, 0); @@ -965,7 +962,6 @@ static void efx_mcdi_filter_vlan_sync_rx_mode(struct efx_nic *efx, struct efx_mcdi_filter_vlan *vlan) { struct efx_mcdi_filter_table *table = efx->filter_state; - struct efx_ef10_nic_data *nic_data = efx->nic_data; /* * Do not install unspecified VID if VLAN filtering is enabled. @@ -1012,11 +1008,10 @@ static void efx_mcdi_filter_vlan_sync_rx_mode(struct efx_nic *efx, * If changing promiscuous state with cascaded multicast filters, remove * old filters first, so that packets are dropped rather than duplicated */ - if (nic_data->workaround_26807 && - table->mc_promisc_last != table->mc_promisc) + if (table->mc_chaining && table->mc_promisc_last != table->mc_promisc) efx_mcdi_filter_remove_old(efx); if (table->mc_promisc) { - if (nic_data->workaround_26807) { + if (table->mc_chaining) { /* * If we failed to insert promiscuous filters, rollback * and fall back to individual multicast filters @@ -1051,7 +1046,7 @@ static void efx_mcdi_filter_vlan_sync_rx_mode(struct efx_nic *efx, */ if (efx_mcdi_filter_insert_addr_list(efx, vlan, true, true)) { /* Changing promisc state, so remove old filters */ - if (nic_data->workaround_26807) + if (table->mc_chaining) efx_mcdi_filter_remove_old(efx); if (efx_mcdi_filter_insert_def(efx, vlan, EFX_ENCAP_TYPE_NONE, @@ -1288,12 +1283,10 @@ efx_mcdi_filter_table_probe_matches(struct efx_nic *efx, return 0; } -int efx_mcdi_filter_table_probe(struct efx_nic *efx) +int efx_mcdi_filter_table_probe(struct efx_nic *efx, bool multicast_chaining) { - struct efx_ef10_nic_data *nic_data = efx->nic_data; struct net_device *net_dev = efx->net_dev; struct efx_mcdi_filter_table *table; - struct efx_mcdi_filter_vlan *vlan; int rc; if (!efx_rwsem_assert_write_locked(&efx->filter_sem)) @@ -1306,12 +1299,12 @@ int efx_mcdi_filter_table_probe(struct efx_nic *efx) if (!table) return -ENOMEM; + table->mc_chaining = multicast_chaining; table->rx_match_count = 0; rc = efx_mcdi_filter_table_probe_matches(efx, table, false); if (rc) goto fail; - if (nic_data->datapath_caps & - (1 << MC_CMD_GET_CAPABILITIES_OUT_VXLAN_NVGRE_LBN)) + if (efx_has_cap(efx, VXLAN_NVGRE, FLAGS1)) rc = efx_mcdi_filter_table_probe_matches(efx, table, true); if (rc) goto fail; @@ -1342,22 +1335,22 @@ int efx_mcdi_filter_table_probe(struct efx_nic *efx) efx->filter_state = table; - list_for_each_entry(vlan, &nic_data->vlan_list, list) { - rc = efx_mcdi_filter_add_vlan(efx, vlan->vid); - if (rc) - goto fail_add_vlan; - } - return 0; - -fail_add_vlan: - efx_mcdi_filter_cleanup_vlans(efx); - efx->filter_state = NULL; fail: kfree(table); return rc; } +void efx_mcdi_filter_table_reset_mc_allocations(struct efx_nic *efx) +{ + struct efx_mcdi_filter_table *table = efx->filter_state; + + if (table) { + table->must_restore_filters = true; + table->must_restore_rss_contexts = true; + } +} + /* * Caller must hold efx->filter_sem for read if race against * efx_mcdi_filter_table_remove() is possible @@ -1365,7 +1358,6 @@ fail: void efx_mcdi_filter_table_restore(struct efx_nic *efx) { struct efx_mcdi_filter_table *table = efx->filter_state; - struct efx_ef10_nic_data *nic_data = efx->nic_data; unsigned int invalid_filters = 0, failed = 0; struct efx_mcdi_filter_vlan *vlan; struct efx_filter_spec *spec; @@ -1377,10 +1369,7 @@ void efx_mcdi_filter_table_restore(struct efx_nic *efx) WARN_ON(!rwsem_is_locked(&efx->filter_sem)); - if (!nic_data->must_restore_filters) - return; - - if (!table) + if (!table || !table->must_restore_filters) return; down_write(&table->lock); @@ -1456,7 +1445,7 @@ not_restored: netif_err(efx, hw, efx->net_dev, "unable to restore %u filters\n", failed); else - nic_data->must_restore_filters = false; + table->must_restore_filters = false; } void efx_mcdi_filter_table_remove(struct efx_nic *efx) @@ -1921,7 +1910,6 @@ static int efx_mcdi_filter_alloc_rss_context(struct efx_nic *efx, bool exclusive { MCDI_DECLARE_BUF(inbuf, MC_CMD_RSS_CONTEXT_ALLOC_IN_LEN); MCDI_DECLARE_BUF(outbuf, MC_CMD_RSS_CONTEXT_ALLOC_OUT_LEN); - struct efx_ef10_nic_data *nic_data = efx->nic_data; size_t outlen; int rc; u32 alloc_type = exclusive ? @@ -1939,12 +1927,11 @@ static int efx_mcdi_filter_alloc_rss_context(struct efx_nic *efx, bool exclusive return 0; } - if (nic_data->datapath_caps & - 1 << MC_CMD_GET_CAPABILITIES_OUT_RX_RSS_LIMITED_LBN) + if (efx_has_cap(efx, RX_RSS_LIMITED, FLAGS1)) return -EOPNOTSUPP; MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_UPSTREAM_PORT_ID, - nic_data->vport_id); + efx->vport_id); MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_TYPE, alloc_type); MCDI_SET_DWORD(inbuf, RSS_CONTEXT_ALLOC_IN_NUM_QUEUES, rss_spread); @@ -1961,8 +1948,7 @@ static int efx_mcdi_filter_alloc_rss_context(struct efx_nic *efx, bool exclusive if (context_size) *context_size = rss_spread; - if (nic_data->datapath_caps & - 1 << MC_CMD_GET_CAPABILITIES_OUT_ADDITIONAL_RSS_MODES_LBN) + if (efx_has_cap(efx, ADDITIONAL_RSS_MODES, FLAGS1)) efx_mcdi_set_rss_context_flags(efx, ctx); return 0; @@ -2030,14 +2016,14 @@ void efx_mcdi_rx_free_indir_table(struct efx_nic *efx) static int efx_mcdi_filter_rx_push_shared_rss_config(struct efx_nic *efx, unsigned *context_size) { - struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct efx_mcdi_filter_table *table = efx->filter_state; int rc = efx_mcdi_filter_alloc_rss_context(efx, false, &efx->rss_context, context_size); if (rc != 0) return rc; - nic_data->rx_rss_context_exclusive = false; + table->rx_rss_context_exclusive = false; efx_set_default_rx_indir_table(efx, &efx->rss_context); return 0; } @@ -2046,12 +2032,12 @@ static int efx_mcdi_filter_rx_push_exclusive_rss_config(struct efx_nic *efx, const u32 *rx_indir_table, const u8 *key) { + struct efx_mcdi_filter_table *table = efx->filter_state; u32 old_rx_rss_context = efx->rss_context.context_id; - struct efx_ef10_nic_data *nic_data = efx->nic_data; int rc; if (efx->rss_context.context_id == EFX_MCDI_RSS_CONTEXT_INVALID || - !nic_data->rx_rss_context_exclusive) { + !table->rx_rss_context_exclusive) { rc = efx_mcdi_filter_alloc_rss_context(efx, true, &efx->rss_context, NULL); if (rc == -EOPNOTSUPP) @@ -2068,7 +2054,7 @@ static int efx_mcdi_filter_rx_push_exclusive_rss_config(struct efx_nic *efx, if (efx->rss_context.context_id != old_rx_rss_context && old_rx_rss_context != EFX_MCDI_RSS_CONTEXT_INVALID) WARN_ON(efx_mcdi_filter_free_rss_context(efx, old_rx_rss_context) != 0); - nic_data->rx_rss_context_exclusive = true; + table->rx_rss_context_exclusive = true; if (rx_indir_table != efx->rss_context.rx_indir_table) memcpy(efx->rss_context.rx_indir_table, rx_indir_table, sizeof(efx->rss_context.rx_indir_table)); @@ -2182,13 +2168,13 @@ int efx_mcdi_rx_pull_rss_config(struct efx_nic *efx) void efx_mcdi_rx_restore_rss_contexts(struct efx_nic *efx) { - struct efx_ef10_nic_data *nic_data = efx->nic_data; + struct efx_mcdi_filter_table *table = efx->filter_state; struct efx_rss_context *ctx; int rc; WARN_ON(!mutex_is_locked(&efx->rss_lock)); - if (!nic_data->must_restore_rss_contexts) + if (!table->must_restore_rss_contexts) return; list_for_each_entry(ctx, &efx->rss_context.list, list) { @@ -2204,7 +2190,7 @@ void efx_mcdi_rx_restore_rss_contexts(struct efx_nic *efx) "; RSS filters may fail to be applied\n", ctx->user_id, rc); } - nic_data->must_restore_rss_contexts = false; + table->must_restore_rss_contexts = false; } int efx_mcdi_pf_rx_push_rss_config(struct efx_nic *efx, bool user, diff --git a/drivers/net/ethernet/sfc/mcdi_filters.h b/drivers/net/ethernet/sfc/mcdi_filters.h index 1837f4f5d661..03a8bf74c733 100644 --- a/drivers/net/ethernet/sfc/mcdi_filters.h +++ b/drivers/net/ethernet/sfc/mcdi_filters.h @@ -55,6 +55,8 @@ struct efx_mcdi_filter_table { u32 rx_match_mcdi_flags[ MC_CMD_GET_PARSER_DISP_INFO_OUT_SUPPORTED_MATCHES_MAXNUM * 2]; unsigned int rx_match_count; + /* Our RSS context is exclusive (as opposed to shared) */ + bool rx_rss_context_exclusive; struct rw_semaphore lock; /* Protects entries */ struct { @@ -75,14 +77,27 @@ struct efx_mcdi_filter_table { /* Whether in multicast promiscuous mode when last changed */ bool mc_promisc_last; bool mc_overflow; /* Too many MC addrs; should always imply mc_promisc */ + /* RSS contexts have yet to be restored after MC reboot */ + bool must_restore_rss_contexts; + /* filters have yet to be restored after MC reboot */ + bool must_restore_filters; + /* Multicast filter chaining allows less-specific filters to receive + * multicast packets that matched more-specific filters. Early EF10 + * firmware didn't support this (SF bug 26807); if mc_chaining == false + * then we still subscribe the dev_mc_list even when mc_promisc to + * prevent another VI stealing the traffic. + */ + bool mc_chaining; bool vlan_filter; struct list_head vlan_list; }; -int efx_mcdi_filter_table_probe(struct efx_nic *efx); +int efx_mcdi_filter_table_probe(struct efx_nic *efx, bool multicast_chaining); void efx_mcdi_filter_table_remove(struct efx_nic *efx); void efx_mcdi_filter_table_restore(struct efx_nic *efx); +void efx_mcdi_filter_table_reset_mc_allocations(struct efx_nic *efx); + /* * The filter table(s) are managed by firmware and we have write-only * access. When removing filters we must identify them to the diff --git a/drivers/net/ethernet/sfc/mcdi_functions.c b/drivers/net/ethernet/sfc/mcdi_functions.c index dcfe78b0fa5a..962d8395d958 100644 --- a/drivers/net/ethernet/sfc/mcdi_functions.c +++ b/drivers/net/ethernet/sfc/mcdi_functions.c @@ -168,21 +168,18 @@ int efx_mcdi_tx_init(struct efx_tx_queue *tx_queue, bool tso_v2) size_t entries = tx_queue->txd.buf.len / EFX_BUF_SIZE; struct efx_channel *channel = tx_queue->channel; struct efx_nic *efx = tx_queue->efx; - struct efx_ef10_nic_data *nic_data; dma_addr_t dma_addr; size_t inlen; int rc, i; BUILD_BUG_ON(MC_CMD_INIT_TXQ_OUT_LEN != 0); - nic_data = efx->nic_data; - MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_SIZE, tx_queue->ptr_mask + 1); MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_TARGET_EVQ, channel->channel); MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_LABEL, tx_queue->queue); MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_INSTANCE, tx_queue->queue); MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_OWNER_ID, 0); - MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_PORT_ID, nic_data->vport_id); + MCDI_SET_DWORD(inbuf, INIT_TXQ_IN_PORT_ID, efx->vport_id); dma_addr = tx_queue->txd.buf.dma_addr; @@ -276,7 +273,6 @@ void efx_mcdi_rx_init(struct efx_rx_queue *rx_queue) struct efx_channel *channel = efx_rx_queue_channel(rx_queue); size_t entries = rx_queue->rxd.buf.len / EFX_BUF_SIZE; struct efx_nic *efx = rx_queue->efx; - struct efx_ef10_nic_data *nic_data = efx->nic_data; dma_addr_t dma_addr; size_t inlen; int rc; @@ -295,7 +291,7 @@ void efx_mcdi_rx_init(struct efx_rx_queue *rx_queue) INIT_RXQ_IN_FLAG_PREFIX, 1, INIT_RXQ_IN_FLAG_TIMESTAMP, 1); MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_OWNER_ID, 0); - MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_PORT_ID, nic_data->vport_id); + MCDI_SET_DWORD(inbuf, INIT_RXQ_IN_PORT_ID, efx->vport_id); dma_addr = rx_queue->rxd.buf.dma_addr; diff --git a/drivers/net/ethernet/sfc/mcdi_port.c b/drivers/net/ethernet/sfc/mcdi_port.c index ab5227b13ae6..b807871d8f69 100644 --- a/drivers/net/ethernet/sfc/mcdi_port.c +++ b/drivers/net/ethernet/sfc/mcdi_port.c @@ -722,11 +722,8 @@ static int efx_mcdi_mac_stats(struct efx_nic *efx, MAC_STATS_IN_PERIOD_MS, period); MCDI_SET_DWORD(inbuf, MAC_STATS_IN_DMA_LEN, dma_len); - if (efx_nic_rev(efx) >= EFX_REV_HUNT_A0) { - struct efx_ef10_nic_data *nic_data = efx->nic_data; - - MCDI_SET_DWORD(inbuf, MAC_STATS_IN_PORT_ID, nic_data->vport_id); - } + if (efx_nic_rev(efx) >= EFX_REV_HUNT_A0) + MCDI_SET_DWORD(inbuf, MAC_STATS_IN_PORT_ID, efx->vport_id); rc = efx_mcdi_rpc_quiet(efx, MC_CMD_MAC_STATS, inbuf, sizeof(inbuf), NULL, 0, NULL); diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h index b084e623b5f4..1afb58feb9ab 100644 --- a/drivers/net/ethernet/sfc/net_driver.h +++ b/drivers/net/ethernet/sfc/net_driver.h @@ -887,8 +887,10 @@ struct efx_async_filter_insertion { * @rss_context: Main RSS context. Its @list member is the head of the list of * RSS contexts created by user requests * @rss_lock: Protects custom RSS context software state in @rss_context.list + * @vport_id: The function's vport ID, only relevant for PFs * @int_error_count: Number of internal errors seen recently * @int_error_expire: Time at which error count will be expired + * @must_realloc_vis: Flag: VIs have yet to be reallocated after MC reboot * @irq_soft_enabled: Are IRQs soft-enabled? If not, IRQ handler will * acknowledge but do nothing else. * @irq_status: Interrupt status buffer @@ -1044,10 +1046,12 @@ struct efx_nic { bool rx_scatter; struct efx_rss_context rss_context; struct mutex rss_lock; + u32 vport_id; unsigned int_error_count; unsigned long int_error_expire; + bool must_realloc_vis; bool irq_soft_enabled; struct efx_buffer irq_status; unsigned irq_zero_count; @@ -1292,6 +1296,7 @@ struct efx_udp_tunnel { * @udp_tnl_add_port: Add a UDP tunnel port * @udp_tnl_has_port: Check if a port has been added as UDP tunnel * @udp_tnl_del_port: Remove a UDP tunnel port + * @print_additional_fwver: Dump NIC-specific additional FW version info * @revision: Hardware architecture revision * @txd_ptr_tbl_base: TX descriptor ring base address * @rxd_ptr_tbl_base: RX descriptor ring base address @@ -1352,6 +1357,9 @@ struct efx_nic_type { void (*get_wol)(struct efx_nic *efx, struct ethtool_wolinfo *wol); int (*set_wol)(struct efx_nic *efx, u32 type); void (*resume_wol)(struct efx_nic *efx); + unsigned int (*check_caps)(const struct efx_nic *efx, + u8 flag, + u32 offset); int (*test_chip)(struct efx_nic *efx, struct efx_self_tests *tests); int (*test_nvram)(struct efx_nic *efx); void (*mcdi_request)(struct efx_nic *efx, @@ -1462,6 +1470,8 @@ struct efx_nic_type { int (*udp_tnl_add_port)(struct efx_nic *efx, struct efx_udp_tunnel tnl); bool (*udp_tnl_has_port)(struct efx_nic *efx, __be16 port); int (*udp_tnl_del_port)(struct efx_nic *efx, struct efx_udp_tunnel tnl); + size_t (*print_additional_fwver)(struct efx_nic *efx, char *buf, + size_t len); int revision; unsigned int txd_ptr_tbl_base; diff --git a/drivers/net/ethernet/sfc/nic.h b/drivers/net/ethernet/sfc/nic.h index 6670fda8f35a..8f73c5d996eb 100644 --- a/drivers/net/ethernet/sfc/nic.h +++ b/drivers/net/ethernet/sfc/nic.h @@ -360,10 +360,6 @@ enum { * @warm_boot_count: Last seen MC warm boot count * @vi_base: Absolute index of first VI in this function * @n_allocated_vis: Number of VIs allocated to this function - * @must_realloc_vis: Flag: VIs have yet to be reallocated after MC reboot - * @must_restore_rss_contexts: Flag: RSS contexts have yet to be restored after - * MC reboot - * @must_restore_filters: Flag: filters have yet to be restored after MC reboot * @n_piobufs: Number of PIO buffers allocated to this function * @wc_membase: Base address of write-combining mapping of the memory BAR * @pio_write_base: Base address for writing PIO buffers @@ -372,7 +368,6 @@ enum { * @piobuf_size: size of a single PIO buffer * @must_restore_piobufs: Flag: PIO buffers have yet to be restored after MC * reboot - * @rx_rss_context_exclusive: Whether our RSS context is exclusive or shared * @stats: Hardware statistics * @workaround_35388: Flag: firmware supports workaround for bug 35388 * @workaround_26807: Flag: firmware supports workaround for bug 26807 @@ -385,7 +380,6 @@ enum { * %MC_CMD_GET_CAPABILITIES response) * @rx_dpcpu_fw_id: Firmware ID of the RxDPCPU * @tx_dpcpu_fw_id: Firmware ID of the TxDPCPU - * @vport_id: The function's vport ID, only relevant for PFs * @must_probe_vswitching: Flag: vswitching has yet to be setup after MC reboot * @pf_index: The number for this PF, or the parent PF if this is a VF #ifdef CONFIG_SFC_SRIOV @@ -404,16 +398,12 @@ struct efx_ef10_nic_data { u16 warm_boot_count; unsigned int vi_base; unsigned int n_allocated_vis; - bool must_realloc_vis; - bool must_restore_rss_contexts; - bool must_restore_filters; unsigned int n_piobufs; void __iomem *wc_membase, *pio_write_base; unsigned int pio_write_vi_base; unsigned int piobuf_handle[EF10_TX_PIOBUF_COUNT]; u16 piobuf_size; bool must_restore_piobufs; - bool rx_rss_context_exclusive; u64 stats[EF10_STAT_COUNT]; bool workaround_35388; bool workaround_26807; @@ -423,7 +413,6 @@ struct efx_ef10_nic_data { u32 datapath_caps2; unsigned int rx_dpcpu_fw_id; unsigned int tx_dpcpu_fw_id; - unsigned int vport_id; bool must_probe_vswitching; unsigned int pf_index; u8 port_id[ETH_ALEN]; diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c index 59b4f16896a8..04c7283d205e 100644 --- a/drivers/net/ethernet/sfc/ptp.c +++ b/drivers/net/ethernet/sfc/ptp.c @@ -352,12 +352,7 @@ static int efx_phc_enable(struct ptp_clock_info *ptp, bool efx_ptp_use_mac_tx_timestamps(struct efx_nic *efx) { - struct efx_ef10_nic_data *nic_data = efx->nic_data; - - return ((efx_nic_rev(efx) >= EFX_REV_HUNT_A0) && - (nic_data->datapath_caps2 & - (1 << MC_CMD_GET_CAPABILITIES_V2_OUT_TX_MAC_TIMESTAMPING_LBN) - )); + return efx_has_cap(efx, TX_MAC_TIMESTAMPING, FLAGS2); } /* PTP 'extra' channel is still a traffic channel, but we only create TX queues diff --git a/drivers/net/ethernet/sfc/rx.c b/drivers/net/ethernet/sfc/rx.c index 260352d97d9d..c01916cff507 100644 --- a/drivers/net/ethernet/sfc/rx.c +++ b/drivers/net/ethernet/sfc/rx.c @@ -308,6 +308,7 @@ static bool efx_do_xdp(struct efx_nic *efx, struct efx_channel *channel, xdp_set_data_meta_invalid(&xdp); xdp.data_end = xdp.data + rx_buf->len; xdp.rxq = &rx_queue->xdp_rxq_info; + xdp.frame_sz = efx->rx_page_buf_step; xdp_act = bpf_prog_run_xdp(xdp_prog, &xdp); rcu_read_unlock(); @@ -328,7 +329,7 @@ static bool efx_do_xdp(struct efx_nic *efx, struct efx_channel *channel, case XDP_TX: /* Buffer ownership passes to tx on success. */ - xdpf = convert_to_xdp_frame(&xdp); + xdpf = xdp_convert_buff_to_frame(&xdp); err = efx_xdp_tx_buffers(efx, 1, &xdpf, true); if (unlikely(err != 1)) { efx_free_rx_buffers(rx_queue, rx_buf, 1); diff --git a/drivers/net/ethernet/sfc/siena.c b/drivers/net/ethernet/sfc/siena.c index baa464161626..891e9fb6abec 100644 --- a/drivers/net/ethernet/sfc/siena.c +++ b/drivers/net/ethernet/sfc/siena.c @@ -948,6 +948,13 @@ fail: #endif /* CONFIG_SFC_MTD */ +static unsigned int siena_check_caps(const struct efx_nic *efx, + u8 flag, u32 offset) +{ + /* Siena did not support MC_CMD_GET_CAPABILITIES */ + return 0; +} + /************************************************************************** * * Revision-dependent attributes used by efx.c and nic.c @@ -1086,4 +1093,5 @@ const struct efx_nic_type siena_a0_nic_type = { 1 << HWTSTAMP_FILTER_PTP_V1_L4_EVENT | 1 << HWTSTAMP_FILTER_PTP_V2_L4_EVENT), .rx_hash_key_size = 16, + .check_caps = siena_check_caps, }; diff --git a/drivers/net/ethernet/sgi/ioc3-eth.c b/drivers/net/ethernet/sgi/ioc3-eth.c index 7305e8e86c51..6646eba9f57f 100644 --- a/drivers/net/ethernet/sgi/ioc3-eth.c +++ b/drivers/net/ethernet/sgi/ioc3-eth.c @@ -848,14 +848,14 @@ static int ioc3eth_probe(struct platform_device *pdev) ip = netdev_priv(dev); ip->dma_dev = pdev->dev.parent; ip->regs = devm_platform_ioremap_resource(pdev, 0); - if (!ip->regs) { - err = -ENOMEM; + if (IS_ERR(ip->regs)) { + err = PTR_ERR(ip->regs); goto out_free; } ip->ssram = devm_platform_ioremap_resource(pdev, 1); - if (!ip->ssram) { - err = -ENOMEM; + if (IS_ERR(ip->ssram)) { + err = PTR_ERR(ip->ssram); goto out_free; } diff --git a/drivers/net/ethernet/smsc/Kconfig b/drivers/net/ethernet/smsc/Kconfig index 9e1c3752b200..4d2d91ec8b41 100644 --- a/drivers/net/ethernet/smsc/Kconfig +++ b/drivers/net/ethernet/smsc/Kconfig @@ -28,7 +28,7 @@ config SMC9194 option if you have a DELL laptop with the docking station, or another SMC9192/9194 based chipset. Say Y if you want it compiled into the kernel, and read the file - <file:Documentation/networking/device_drivers/smsc/smc9.txt>. + <file:Documentation/networking/device_drivers/smsc/smc9.rst>. To compile this driver as a module, choose M here. The module will be called smc9194. @@ -44,7 +44,7 @@ config SMC91X This is a driver for SMC's 91x series of Ethernet chipsets, including the SMC91C94 and the SMC91C111. Say Y if you want it compiled into the kernel, and read the file - <file:Documentation/networking/device_drivers/smsc/smc9.txt>. + <file:Documentation/networking/device_drivers/smsc/smc9.rst>. This driver is also available as a module ( = code which can be inserted in and removed from the running kernel whenever you want). diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index 49a6a9167af4..fc168f85e7af 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -2493,20 +2493,20 @@ static int smsc911x_drv_probe(struct platform_device *pdev) retval = smsc911x_init(dev); if (retval < 0) - goto out_disable_resources; + goto out_init_fail; netif_carrier_off(dev); retval = smsc911x_mii_init(pdev, dev); if (retval) { SMSC_WARN(pdata, probe, "Error %i initialising mii", retval); - goto out_disable_resources; + goto out_init_fail; } retval = register_netdev(dev); if (retval) { SMSC_WARN(pdata, probe, "Error %i registering device", retval); - goto out_disable_resources; + goto out_init_fail; } else { SMSC_TRACE(pdata, probe, "Network interface: \"%s\"", dev->name); @@ -2547,9 +2547,10 @@ static int smsc911x_drv_probe(struct platform_device *pdev) return 0; -out_disable_resources: +out_init_fail: pm_runtime_put(&pdev->dev); pm_runtime_disable(&pdev->dev); +out_disable_resources: (void)smsc911x_disable_resources(pdev); out_enable_resources_fail: smsc911x_free_resources(pdev); diff --git a/drivers/net/ethernet/socionext/netsec.c b/drivers/net/ethernet/socionext/netsec.c index a5a0fb60193a..328bc38848bb 100644 --- a/drivers/net/ethernet/socionext/netsec.c +++ b/drivers/net/ethernet/socionext/netsec.c @@ -867,7 +867,7 @@ static u32 netsec_xdp_queue_one(struct netsec_priv *priv, static u32 netsec_xdp_xmit_back(struct netsec_priv *priv, struct xdp_buff *xdp) { struct netsec_desc_ring *tx_ring = &priv->desc_ring[NETSEC_RING_TX]; - struct xdp_frame *xdpf = convert_to_xdp_frame(xdp); + struct xdp_frame *xdpf = xdp_convert_buff_to_frame(xdp); u32 ret; if (unlikely(!xdpf)) @@ -884,23 +884,28 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, struct xdp_buff *xdp) { struct netsec_desc_ring *dring = &priv->desc_ring[NETSEC_RING_RX]; - unsigned int len = xdp->data_end - xdp->data; + unsigned int sync, len = xdp->data_end - xdp->data; u32 ret = NETSEC_XDP_PASS; + struct page *page; int err; u32 act; act = bpf_prog_run_xdp(prog, xdp); + /* Due xdp_adjust_tail: DMA sync for_device cover max len CPU touch */ + sync = xdp->data_end - xdp->data_hard_start - NETSEC_RXBUF_HEADROOM; + sync = max(sync, len); + switch (act) { case XDP_PASS: ret = NETSEC_XDP_PASS; break; case XDP_TX: ret = netsec_xdp_xmit_back(priv, xdp); - if (ret != NETSEC_XDP_TX) - page_pool_put_page(dring->page_pool, - virt_to_head_page(xdp->data), len, - true); + if (ret != NETSEC_XDP_TX) { + page = virt_to_head_page(xdp->data); + page_pool_put_page(dring->page_pool, page, sync, true); + } break; case XDP_REDIRECT: err = xdp_do_redirect(priv->ndev, xdp, prog); @@ -908,9 +913,8 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, ret = NETSEC_XDP_REDIR; } else { ret = NETSEC_XDP_CONSUMED; - page_pool_put_page(dring->page_pool, - virt_to_head_page(xdp->data), len, - true); + page = virt_to_head_page(xdp->data); + page_pool_put_page(dring->page_pool, page, sync, true); } break; default: @@ -921,8 +925,8 @@ static u32 netsec_run_xdp(struct netsec_priv *priv, struct bpf_prog *prog, /* fall through -- handle aborts by dropping packet */ case XDP_DROP: ret = NETSEC_XDP_CONSUMED; - page_pool_put_page(dring->page_pool, - virt_to_head_page(xdp->data), len, true); + page = virt_to_head_page(xdp->data); + page_pool_put_page(dring->page_pool, page, sync, true); break; } @@ -936,10 +940,14 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget) struct netsec_rx_pkt_info rx_info; enum dma_data_direction dma_dir; struct bpf_prog *xdp_prog; + struct xdp_buff xdp; u16 xdp_xmit = 0; u32 xdp_act = 0; int done = 0; + xdp.rxq = &dring->xdp_rxq; + xdp.frame_sz = PAGE_SIZE; + rcu_read_lock(); xdp_prog = READ_ONCE(priv->xdp_prog); dma_dir = page_pool_get_dma_dir(dring->page_pool); @@ -953,7 +961,6 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget) struct sk_buff *skb = NULL; u16 pkt_len, desc_len; dma_addr_t dma_handle; - struct xdp_buff xdp; void *buf_addr; if (de->attr & (1U << NETSEC_RX_PKT_OWN_FIELD)) { @@ -1002,7 +1009,6 @@ static int netsec_process_rx(struct netsec_priv *priv, int budget) xdp.data = desc->addr + NETSEC_RXBUF_HEADROOM; xdp_set_data_meta_invalid(&xdp); xdp.data_end = xdp.data + pkt_len; - xdp.rxq = &dring->xdp_rxq; if (xdp_prog) { xdp_result = netsec_run_xdp(priv, xdp_prog, &xdp); diff --git a/drivers/net/ethernet/socionext/sni_ave.c b/drivers/net/ethernet/socionext/sni_ave.c index 67ddf782d98a..f2638446b62e 100644 --- a/drivers/net/ethernet/socionext/sni_ave.c +++ b/drivers/net/ethernet/socionext/sni_ave.c @@ -1394,7 +1394,7 @@ static int ave_stop(struct net_device *ndev) return 0; } -static int ave_start_xmit(struct sk_buff *skb, struct net_device *ndev) +static netdev_tx_t ave_start_xmit(struct sk_buff *skb, struct net_device *ndev) { struct ave_private *priv = netdev_priv(ndev); u32 proc_idx, done_idx, ndesc, cmdsts; diff --git a/drivers/net/ethernet/stmicro/stmmac/Kconfig b/drivers/net/ethernet/stmicro/stmmac/Kconfig index b46f8d2ae6d7..36bd2e18f23b 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Kconfig +++ b/drivers/net/ethernet/stmicro/stmmac/Kconfig @@ -196,6 +196,19 @@ config DWMAC_SUN8I This selects Allwinner SoC glue layer support for the stmmac device driver. This driver is used for H3/A83T/A64 EMAC ethernet controller. + +config DWMAC_IMX8 + tristate "NXP IMX8 DWMAC support" + default ARCH_MXC + depends on OF && (ARCH_MXC || COMPILE_TEST) + select MFD_SYSCON + ---help--- + Support for ethernet controller on NXP i.MX8 SOCs. + + This selects NXP SoC glue layer support for the stmmac + device driver. This driver is used for i.MX8 series like + iMX8MP/iMX8DXL GMAC ethernet controller. + endif config DWMAC_INTEL diff --git a/drivers/net/ethernet/stmicro/stmmac/Makefile b/drivers/net/ethernet/stmicro/stmmac/Makefile index 5a6f265bc540..295615ab36a7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/Makefile +++ b/drivers/net/ethernet/stmicro/stmmac/Makefile @@ -27,9 +27,10 @@ obj-$(CONFIG_DWMAC_SUNXI) += dwmac-sunxi.o obj-$(CONFIG_DWMAC_SUN8I) += dwmac-sun8i.o obj-$(CONFIG_DWMAC_DWC_QOS_ETH) += dwmac-dwc-qos-eth.o obj-$(CONFIG_DWMAC_GENERIC) += dwmac-generic.o +obj-$(CONFIG_DWMAC_IMX8) += dwmac-imx.o stmmac-platform-objs:= stmmac_platform.o dwmac-altr-socfpga-objs := altr_tse_pcs.o dwmac-socfpga.o -obj-$(CONFIG_DWMAC_INTEL) += dwmac-intel.o -obj-$(CONFIG_STMMAC_PCI) += stmmac-pci.o +obj-$(CONFIG_STMMAC_PCI) += stmmac-pci.o +obj-$(CONFIG_DWMAC_INTEL) += dwmac-intel.o stmmac-pci-objs:= stmmac_pci.o diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 6208a68a331d..127f75862962 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -473,6 +473,7 @@ struct mac_device_info { unsigned int xlgmac; unsigned int num_vlan; u32 vlan_filter[32]; + unsigned int promisc; }; struct stmmac_rx_routing { diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c new file mode 100644 index 000000000000..5010af7dab4a --- /dev/null +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-imx.c @@ -0,0 +1,315 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * dwmac-imx.c - DWMAC Specific Glue layer for NXP imx8 + * + * Copyright 2020 NXP + * + */ + +#include <linux/clk.h> +#include <linux/gpio/consumer.h> +#include <linux/kernel.h> +#include <linux/mfd/syscon.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/of_net.h> +#include <linux/phy.h> +#include <linux/platform_device.h> +#include <linux/pm_wakeirq.h> +#include <linux/regmap.h> +#include <linux/slab.h> +#include <linux/stmmac.h> + +#include "stmmac_platform.h" + +#define GPR_ENET_QOS_INTF_MODE_MASK GENMASK(21, 16) +#define GPR_ENET_QOS_INTF_SEL_MII (0x0 << 16) +#define GPR_ENET_QOS_INTF_SEL_RMII (0x4 << 16) +#define GPR_ENET_QOS_INTF_SEL_RGMII (0x1 << 16) +#define GPR_ENET_QOS_CLK_GEN_EN (0x1 << 19) +#define GPR_ENET_QOS_CLK_TX_CLK_SEL (0x1 << 20) +#define GPR_ENET_QOS_RGMII_EN (0x1 << 21) + +struct imx_dwmac_ops { + u32 addr_width; + bool mac_rgmii_txclk_auto_adj; + + int (*set_intf_mode)(struct plat_stmmacenet_data *plat_dat); +}; + +struct imx_priv_data { + struct device *dev; + struct clk *clk_tx; + struct clk *clk_mem; + struct regmap *intf_regmap; + u32 intf_reg_off; + bool rmii_refclk_ext; + + const struct imx_dwmac_ops *ops; + struct plat_stmmacenet_data *plat_dat; +}; + +static int imx8mp_set_intf_mode(struct plat_stmmacenet_data *plat_dat) +{ + struct imx_priv_data *dwmac = plat_dat->bsp_priv; + int val; + + switch (plat_dat->interface) { + case PHY_INTERFACE_MODE_MII: + val = GPR_ENET_QOS_INTF_SEL_MII; + break; + case PHY_INTERFACE_MODE_RMII: + val = GPR_ENET_QOS_INTF_SEL_RMII; + val |= (dwmac->rmii_refclk_ext ? 0 : GPR_ENET_QOS_CLK_TX_CLK_SEL); + break; + case PHY_INTERFACE_MODE_RGMII: + case PHY_INTERFACE_MODE_RGMII_ID: + case PHY_INTERFACE_MODE_RGMII_RXID: + case PHY_INTERFACE_MODE_RGMII_TXID: + val = GPR_ENET_QOS_INTF_SEL_RGMII | + GPR_ENET_QOS_RGMII_EN; + break; + default: + pr_debug("imx dwmac doesn't support %d interface\n", + plat_dat->interface); + return -EINVAL; + } + + val |= GPR_ENET_QOS_CLK_GEN_EN; + return regmap_update_bits(dwmac->intf_regmap, dwmac->intf_reg_off, + GPR_ENET_QOS_INTF_MODE_MASK, val); +}; + +static int +imx8dxl_set_intf_mode(struct plat_stmmacenet_data *plat_dat) +{ + int ret = 0; + + /* TBD: depends on imx8dxl scu interfaces to be upstreamed */ + return ret; +} + +static int imx_dwmac_init(struct platform_device *pdev, void *priv) +{ + struct plat_stmmacenet_data *plat_dat; + struct imx_priv_data *dwmac = priv; + int ret; + + plat_dat = dwmac->plat_dat; + + ret = clk_prepare_enable(dwmac->clk_mem); + if (ret) { + dev_err(&pdev->dev, "mem clock enable failed\n"); + return ret; + } + + ret = clk_prepare_enable(dwmac->clk_tx); + if (ret) { + dev_err(&pdev->dev, "tx clock enable failed\n"); + goto clk_tx_en_failed; + } + + if (dwmac->ops->set_intf_mode) { + ret = dwmac->ops->set_intf_mode(plat_dat); + if (ret) + goto intf_mode_failed; + } + + return 0; + +intf_mode_failed: + clk_disable_unprepare(dwmac->clk_tx); +clk_tx_en_failed: + clk_disable_unprepare(dwmac->clk_mem); + return ret; +} + +static void imx_dwmac_exit(struct platform_device *pdev, void *priv) +{ + struct imx_priv_data *dwmac = priv; + + if (dwmac->clk_tx) + clk_disable_unprepare(dwmac->clk_tx); + clk_disable_unprepare(dwmac->clk_mem); +} + +static void imx_dwmac_fix_speed(void *priv, unsigned int speed) +{ + struct plat_stmmacenet_data *plat_dat; + struct imx_priv_data *dwmac = priv; + unsigned long rate; + int err; + + plat_dat = dwmac->plat_dat; + + if (dwmac->ops->mac_rgmii_txclk_auto_adj || + (plat_dat->interface == PHY_INTERFACE_MODE_RMII) || + (plat_dat->interface == PHY_INTERFACE_MODE_MII)) + return; + + switch (speed) { + case SPEED_1000: + rate = 125000000; + break; + case SPEED_100: + rate = 25000000; + break; + case SPEED_10: + rate = 2500000; + break; + default: + dev_err(dwmac->dev, "invalid speed %u\n", speed); + return; + } + + err = clk_set_rate(dwmac->clk_tx, rate); + if (err < 0) + dev_err(dwmac->dev, "failed to set tx rate %lu\n", rate); +} + +static int +imx_dwmac_parse_dt(struct imx_priv_data *dwmac, struct device *dev) +{ + struct device_node *np = dev->of_node; + int err = 0; + + if (of_get_property(np, "snps,rmii_refclk_ext", NULL)) + dwmac->rmii_refclk_ext = true; + + dwmac->clk_tx = devm_clk_get(dev, "tx"); + if (IS_ERR(dwmac->clk_tx)) { + dev_err(dev, "failed to get tx clock\n"); + return PTR_ERR(dwmac->clk_tx); + } + + dwmac->clk_mem = NULL; + if (of_machine_is_compatible("fsl,imx8dxl")) { + dwmac->clk_mem = devm_clk_get(dev, "mem"); + if (IS_ERR(dwmac->clk_mem)) { + dev_err(dev, "failed to get mem clock\n"); + return PTR_ERR(dwmac->clk_mem); + } + } + + if (of_machine_is_compatible("fsl,imx8mp")) { + /* Binding doc describes the propety: + is required by i.MX8MP. + is optinoal for i.MX8DXL. + */ + dwmac->intf_regmap = syscon_regmap_lookup_by_phandle(np, "intf_mode"); + if (IS_ERR(dwmac->intf_regmap)) + return PTR_ERR(dwmac->intf_regmap); + + err = of_property_read_u32_index(np, "intf_mode", 1, &dwmac->intf_reg_off); + if (err) { + dev_err(dev, "Can't get intf mode reg offset (%d)\n", err); + return err; + } + } + + return err; +} + +static int imx_dwmac_probe(struct platform_device *pdev) +{ + struct plat_stmmacenet_data *plat_dat; + struct stmmac_resources stmmac_res; + struct imx_priv_data *dwmac; + const struct imx_dwmac_ops *data; + int ret; + + ret = stmmac_get_platform_resources(pdev, &stmmac_res); + if (ret) + return ret; + + dwmac = devm_kzalloc(&pdev->dev, sizeof(*dwmac), GFP_KERNEL); + if (!dwmac) + return PTR_ERR(dwmac); + + plat_dat = stmmac_probe_config_dt(pdev, &stmmac_res.mac); + if (IS_ERR(plat_dat)) + return PTR_ERR(plat_dat); + + data = of_device_get_match_data(&pdev->dev); + if (!data) { + dev_err(&pdev->dev, "failed to get match data\n"); + ret = -EINVAL; + goto err_match_data; + } + + dwmac->ops = data; + dwmac->dev = &pdev->dev; + + ret = imx_dwmac_parse_dt(dwmac, &pdev->dev); + if (ret) { + dev_err(&pdev->dev, "failed to parse OF data\n"); + goto err_parse_dt; + } + + ret = dma_set_mask_and_coherent(&pdev->dev, + DMA_BIT_MASK(dwmac->ops->addr_width)); + if (ret) { + dev_err(&pdev->dev, "DMA mask set failed\n"); + goto err_dma_mask; + } + + plat_dat->init = imx_dwmac_init; + plat_dat->exit = imx_dwmac_exit; + plat_dat->fix_mac_speed = imx_dwmac_fix_speed; + plat_dat->bsp_priv = dwmac; + dwmac->plat_dat = plat_dat; + + ret = imx_dwmac_init(pdev, dwmac); + if (ret) + goto err_dwmac_init; + + ret = stmmac_dvr_probe(&pdev->dev, plat_dat, &stmmac_res); + if (ret) + goto err_drv_probe; + + return 0; + +err_dwmac_init: +err_drv_probe: + imx_dwmac_exit(pdev, plat_dat->bsp_priv); +err_dma_mask: +err_parse_dt: +err_match_data: + stmmac_remove_config_dt(pdev, plat_dat); + return ret; +} + +static struct imx_dwmac_ops imx8mp_dwmac_data = { + .addr_width = 34, + .mac_rgmii_txclk_auto_adj = false, + .set_intf_mode = imx8mp_set_intf_mode, +}; + +static struct imx_dwmac_ops imx8dxl_dwmac_data = { + .addr_width = 32, + .mac_rgmii_txclk_auto_adj = true, + .set_intf_mode = imx8dxl_set_intf_mode, +}; + +static const struct of_device_id imx_dwmac_match[] = { + { .compatible = "nxp,imx8mp-dwmac-eqos", .data = &imx8mp_dwmac_data }, + { .compatible = "nxp,imx8dxl-dwmac-eqos", .data = &imx8dxl_dwmac_data }, + { } +}; +MODULE_DEVICE_TABLE(of, imx_dwmac_match); + +static struct platform_driver imx_dwmac_driver = { + .probe = imx_dwmac_probe, + .remove = stmmac_pltfr_remove, + .driver = { + .name = "imx-dwmac", + .pm = &stmmac_pltfr_pm_ops, + .of_match_table = imx_dwmac_match, + }, +}; +module_platform_driver(imx_dwmac_driver); + +MODULE_AUTHOR("NXP"); +MODULE_DESCRIPTION("NXP imx8 DWMAC Specific Glue layer"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c index 2e4aaedb93f5..2ac9dfb3462c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-intel.c @@ -83,13 +83,9 @@ static int intel_serdes_powerup(struct net_device *ndev, void *priv_data) serdes_phy_addr = intel_priv->mdio_adhoc_addr; /* assert clk_req */ - data = mdiobus_read(priv->mii, serdes_phy_addr, - SERDES_GCR0); - + data = mdiobus_read(priv->mii, serdes_phy_addr, SERDES_GCR0); data |= SERDES_PLL_CLK; - - mdiobus_write(priv->mii, serdes_phy_addr, - SERDES_GCR0, data); + mdiobus_write(priv->mii, serdes_phy_addr, SERDES_GCR0, data); /* check for clk_ack assertion */ data = serdes_status_poll(priv, serdes_phy_addr, @@ -103,13 +99,9 @@ static int intel_serdes_powerup(struct net_device *ndev, void *priv_data) } /* assert lane reset */ - data = mdiobus_read(priv->mii, serdes_phy_addr, - SERDES_GCR0); - + data = mdiobus_read(priv->mii, serdes_phy_addr, SERDES_GCR0); data |= SERDES_RST; - - mdiobus_write(priv->mii, serdes_phy_addr, - SERDES_GCR0, data); + mdiobus_write(priv->mii, serdes_phy_addr, SERDES_GCR0, data); /* check for assert lane reset reflection */ data = serdes_status_poll(priv, serdes_phy_addr, @@ -123,14 +115,12 @@ static int intel_serdes_powerup(struct net_device *ndev, void *priv_data) } /* move power state to P0 */ - data = mdiobus_read(priv->mii, serdes_phy_addr, - SERDES_GCR0); + data = mdiobus_read(priv->mii, serdes_phy_addr, SERDES_GCR0); data &= ~SERDES_PWR_ST_MASK; data |= SERDES_PWR_ST_P0 << SERDES_PWR_ST_SHIFT; - mdiobus_write(priv->mii, serdes_phy_addr, - SERDES_GCR0, data); + mdiobus_write(priv->mii, serdes_phy_addr, SERDES_GCR0, data); /* Check for P0 state */ data = serdes_status_poll(priv, serdes_phy_addr, @@ -159,14 +149,12 @@ static void intel_serdes_powerdown(struct net_device *ndev, void *intel_data) serdes_phy_addr = intel_priv->mdio_adhoc_addr; /* move power state to P3 */ - data = mdiobus_read(priv->mii, serdes_phy_addr, - SERDES_GCR0); + data = mdiobus_read(priv->mii, serdes_phy_addr, SERDES_GCR0); data &= ~SERDES_PWR_ST_MASK; data |= SERDES_PWR_ST_P3 << SERDES_PWR_ST_SHIFT; - mdiobus_write(priv->mii, serdes_phy_addr, - SERDES_GCR0, data); + mdiobus_write(priv->mii, serdes_phy_addr, SERDES_GCR0, data); /* Check for P3 state */ data = serdes_status_poll(priv, serdes_phy_addr, @@ -180,13 +168,9 @@ static void intel_serdes_powerdown(struct net_device *ndev, void *intel_data) } /* de-assert clk_req */ - data = mdiobus_read(priv->mii, serdes_phy_addr, - SERDES_GCR0); - + data = mdiobus_read(priv->mii, serdes_phy_addr, SERDES_GCR0); data &= ~SERDES_PLL_CLK; - - mdiobus_write(priv->mii, serdes_phy_addr, - SERDES_GCR0, data); + mdiobus_write(priv->mii, serdes_phy_addr, SERDES_GCR0, data); /* check for clk_ack de-assert */ data = serdes_status_poll(priv, serdes_phy_addr, @@ -200,13 +184,9 @@ static void intel_serdes_powerdown(struct net_device *ndev, void *intel_data) } /* de-assert lane reset */ - data = mdiobus_read(priv->mii, serdes_phy_addr, - SERDES_GCR0); - + data = mdiobus_read(priv->mii, serdes_phy_addr, SERDES_GCR0); data &= ~SERDES_RST; - - mdiobus_write(priv->mii, serdes_phy_addr, - SERDES_GCR0, data); + mdiobus_write(priv->mii, serdes_phy_addr, SERDES_GCR0, data); /* check for de-assert lane reset reflection */ data = serdes_status_poll(priv, serdes_phy_addr, @@ -252,6 +232,7 @@ static void common_default_data(struct plat_stmmacenet_data *plat) static int intel_mgbe_common_data(struct pci_dev *pdev, struct plat_stmmacenet_data *plat) { + int ret; int i; plat->clk_csr = 5; @@ -324,7 +305,12 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, dev_warn(&pdev->dev, "Fail to register stmmac-clk\n"); plat->stmmac_clk = NULL; } - clk_prepare_enable(plat->stmmac_clk); + + ret = clk_prepare_enable(plat->stmmac_clk); + if (ret) { + clk_unregister_fixed_rate(plat->stmmac_clk); + return ret; + } /* Set default value for multicast hash bins */ plat->multicast_filter_bins = HASH_TABLE_SIZE; @@ -341,16 +327,11 @@ static int intel_mgbe_common_data(struct pci_dev *pdev, static int ehl_common_data(struct pci_dev *pdev, struct plat_stmmacenet_data *plat) { - int ret; - plat->rx_queues_to_use = 8; plat->tx_queues_to_use = 8; plat->clk_ptp_rate = 200000000; - ret = intel_mgbe_common_data(pdev, plat); - if (ret) - return ret; - return 0; + return intel_mgbe_common_data(pdev, plat); } static int ehl_sgmii_data(struct pci_dev *pdev, @@ -366,7 +347,7 @@ static int ehl_sgmii_data(struct pci_dev *pdev, return ehl_common_data(pdev, plat); } -static struct stmmac_pci_info ehl_sgmii1g_pci_info = { +static struct stmmac_pci_info ehl_sgmii1g_info = { .setup = ehl_sgmii_data, }; @@ -380,7 +361,7 @@ static int ehl_rgmii_data(struct pci_dev *pdev, return ehl_common_data(pdev, plat); } -static struct stmmac_pci_info ehl_rgmii1g_pci_info = { +static struct stmmac_pci_info ehl_rgmii1g_info = { .setup = ehl_rgmii_data, }; @@ -399,7 +380,7 @@ static int ehl_pse0_rgmii1g_data(struct pci_dev *pdev, return ehl_pse0_common_data(pdev, plat); } -static struct stmmac_pci_info ehl_pse0_rgmii1g_pci_info = { +static struct stmmac_pci_info ehl_pse0_rgmii1g_info = { .setup = ehl_pse0_rgmii1g_data, }; @@ -412,7 +393,7 @@ static int ehl_pse0_sgmii1g_data(struct pci_dev *pdev, return ehl_pse0_common_data(pdev, plat); } -static struct stmmac_pci_info ehl_pse0_sgmii1g_pci_info = { +static struct stmmac_pci_info ehl_pse0_sgmii1g_info = { .setup = ehl_pse0_sgmii1g_data, }; @@ -431,7 +412,7 @@ static int ehl_pse1_rgmii1g_data(struct pci_dev *pdev, return ehl_pse1_common_data(pdev, plat); } -static struct stmmac_pci_info ehl_pse1_rgmii1g_pci_info = { +static struct stmmac_pci_info ehl_pse1_rgmii1g_info = { .setup = ehl_pse1_rgmii1g_data, }; @@ -444,23 +425,18 @@ static int ehl_pse1_sgmii1g_data(struct pci_dev *pdev, return ehl_pse1_common_data(pdev, plat); } -static struct stmmac_pci_info ehl_pse1_sgmii1g_pci_info = { +static struct stmmac_pci_info ehl_pse1_sgmii1g_info = { .setup = ehl_pse1_sgmii1g_data, }; static int tgl_common_data(struct pci_dev *pdev, struct plat_stmmacenet_data *plat) { - int ret; - plat->rx_queues_to_use = 6; plat->tx_queues_to_use = 4; plat->clk_ptp_rate = 200000000; - ret = intel_mgbe_common_data(pdev, plat); - if (ret) - return ret; - return 0; + return intel_mgbe_common_data(pdev, plat); } static int tgl_sgmii_data(struct pci_dev *pdev, @@ -474,7 +450,7 @@ static int tgl_sgmii_data(struct pci_dev *pdev, return tgl_common_data(pdev, plat); } -static struct stmmac_pci_info tgl_sgmii1g_pci_info = { +static struct stmmac_pci_info tgl_sgmii1g_info = { .setup = tgl_sgmii_data, }; @@ -577,7 +553,7 @@ static int quark_default_data(struct pci_dev *pdev, return 0; } -static const struct stmmac_pci_info quark_pci_info = { +static const struct stmmac_pci_info quark_info = { .setup = quark_default_data, }; @@ -600,11 +576,9 @@ static int intel_eth_pci_probe(struct pci_dev *pdev, struct intel_priv_data *intel_priv; struct plat_stmmacenet_data *plat; struct stmmac_resources res; - int i; int ret; - intel_priv = devm_kzalloc(&pdev->dev, sizeof(*intel_priv), - GFP_KERNEL); + intel_priv = devm_kzalloc(&pdev->dev, sizeof(*intel_priv), GFP_KERNEL); if (!intel_priv) return -ENOMEM; @@ -631,15 +605,9 @@ static int intel_eth_pci_probe(struct pci_dev *pdev, return ret; } - /* Get the base address of device */ - for (i = 0; i < PCI_STD_NUM_BARS; i++) { - if (pci_resource_len(pdev, i) == 0) - continue; - ret = pcim_iomap_regions(pdev, BIT(i), pci_name(pdev)); - if (ret) - return ret; - break; - } + ret = pcim_iomap_regions(pdev, BIT(0), pci_name(pdev)); + if (ret) + return ret; pci_set_master(pdev); @@ -650,14 +618,23 @@ static int intel_eth_pci_probe(struct pci_dev *pdev, if (ret) return ret; - pci_enable_msi(pdev); + ret = pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_ALL_TYPES); + if (ret < 0) + return ret; memset(&res, 0, sizeof(res)); - res.addr = pcim_iomap_table(pdev)[i]; - res.wol_irq = pdev->irq; - res.irq = pdev->irq; + res.addr = pcim_iomap_table(pdev)[0]; + res.wol_irq = pci_irq_vector(pdev, 0); + res.irq = pci_irq_vector(pdev, 0); + + ret = stmmac_dvr_probe(&pdev->dev, plat, &res); + if (ret) { + pci_free_irq_vectors(pdev); + clk_disable_unprepare(plat->stmmac_clk); + clk_unregister_fixed_rate(plat->stmmac_clk); + } - return stmmac_dvr_probe(&pdev->dev, plat, &res); + return ret; } /** @@ -671,19 +648,15 @@ static void intel_eth_pci_remove(struct pci_dev *pdev) { struct net_device *ndev = dev_get_drvdata(&pdev->dev); struct stmmac_priv *priv = netdev_priv(ndev); - int i; stmmac_dvr_remove(&pdev->dev); - if (priv->plat->stmmac_clk) - clk_unregister_fixed_rate(priv->plat->stmmac_clk); + pci_free_irq_vectors(pdev); - for (i = 0; i < PCI_STD_NUM_BARS; i++) { - if (pci_resource_len(pdev, i) == 0) - continue; - pcim_iounmap_regions(pdev, BIT(i)); - break; - } + clk_disable_unprepare(priv->plat->stmmac_clk); + clk_unregister_fixed_rate(priv->plat->stmmac_clk); + + pcim_iounmap_regions(pdev, BIT(0)); pci_disable_device(pdev); } @@ -742,26 +715,19 @@ static SIMPLE_DEV_PM_OPS(intel_eth_pm_ops, intel_eth_pci_suspend, #define PCI_DEVICE_ID_INTEL_TGL_SGMII1G_ID 0xa0ac static const struct pci_device_id intel_eth_pci_id_table[] = { - { PCI_DEVICE_DATA(INTEL, QUARK_ID, &quark_pci_info) }, - { PCI_DEVICE_DATA(INTEL, EHL_RGMII1G_ID, &ehl_rgmii1g_pci_info) }, - { PCI_DEVICE_DATA(INTEL, EHL_SGMII1G_ID, &ehl_sgmii1g_pci_info) }, - { PCI_DEVICE_DATA(INTEL, EHL_SGMII2G5_ID, &ehl_sgmii1g_pci_info) }, - { PCI_DEVICE_DATA(INTEL, EHL_PSE0_RGMII1G_ID, - &ehl_pse0_rgmii1g_pci_info) }, - { PCI_DEVICE_DATA(INTEL, EHL_PSE0_SGMII1G_ID, - &ehl_pse0_sgmii1g_pci_info) }, - { PCI_DEVICE_DATA(INTEL, EHL_PSE0_SGMII2G5_ID, - &ehl_pse0_sgmii1g_pci_info) }, - { PCI_DEVICE_DATA(INTEL, EHL_PSE1_RGMII1G_ID, - &ehl_pse1_rgmii1g_pci_info) }, - { PCI_DEVICE_DATA(INTEL, EHL_PSE1_SGMII1G_ID, - &ehl_pse1_sgmii1g_pci_info) }, - { PCI_DEVICE_DATA(INTEL, EHL_PSE1_SGMII2G5_ID, - &ehl_pse1_sgmii1g_pci_info) }, - { PCI_DEVICE_DATA(INTEL, TGL_SGMII1G_ID, &tgl_sgmii1g_pci_info) }, + { PCI_DEVICE_DATA(INTEL, QUARK_ID, &quark_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_RGMII1G_ID, &ehl_rgmii1g_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_SGMII1G_ID, &ehl_sgmii1g_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_SGMII2G5_ID, &ehl_sgmii1g_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_PSE0_RGMII1G_ID, &ehl_pse0_rgmii1g_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_PSE0_SGMII1G_ID, &ehl_pse0_sgmii1g_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_PSE0_SGMII2G5_ID, &ehl_pse0_sgmii1g_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_PSE1_RGMII1G_ID, &ehl_pse1_rgmii1g_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_PSE1_SGMII1G_ID, &ehl_pse1_sgmii1g_info) }, + { PCI_DEVICE_DATA(INTEL, EHL_PSE1_SGMII2G5_ID, &ehl_pse1_sgmii1g_info) }, + { PCI_DEVICE_DATA(INTEL, TGL_SGMII1G_ID, &tgl_sgmii1g_info) }, {} }; - MODULE_DEVICE_TABLE(pci, intel_eth_pci_id_table); static struct pci_driver intel_eth_pci_driver = { diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c index 6ae13dc19510..02102c781a8c 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-ipq806x.c @@ -319,6 +319,19 @@ static int ipq806x_gmac_probe(struct platform_device *pdev) /* Enable PTP clock */ regmap_read(gmac->nss_common, NSS_COMMON_CLK_GATE, &val); val |= NSS_COMMON_CLK_GATE_PTP_EN(gmac->id); + switch (gmac->phy_mode) { + case PHY_INTERFACE_MODE_RGMII: + val |= NSS_COMMON_CLK_GATE_RGMII_RX_EN(gmac->id) | + NSS_COMMON_CLK_GATE_RGMII_TX_EN(gmac->id); + break; + case PHY_INTERFACE_MODE_SGMII: + val |= NSS_COMMON_CLK_GATE_GMII_RX_EN(gmac->id) | + NSS_COMMON_CLK_GATE_GMII_TX_EN(gmac->id); + break; + default: + /* We don't get here; the switch above will have errored out */ + unreachable(); + } regmap_write(gmac->nss_common, NSS_COMMON_CLK_GATE, val); if (gmac->phy_mode == PHY_INTERFACE_MODE_SGMII) { diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c index a3934ca6a043..234e8b6816ce 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-meson8b.c @@ -5,6 +5,7 @@ * Copyright (C) 2016 Martin Blumenstingl <martin.blumenstingl@googlemail.com> */ +#include <linux/bitfield.h> #include <linux/clk.h> #include <linux/clk-provider.h> #include <linux/device.h> @@ -32,7 +33,10 @@ #define PRG_ETH0_CLK_M250_SEL_SHIFT 4 #define PRG_ETH0_CLK_M250_SEL_MASK GENMASK(4, 4) -#define PRG_ETH0_TXDLY_SHIFT 5 +/* TX clock delay in ns = "8ns / 4 * tx_dly_val" (where 8ns are exactly one + * cycle of the 125MHz RGMII TX clock): + * 0ns = 0x0, 2ns = 0x1, 4ns = 0x2, 6ns = 0x3 + */ #define PRG_ETH0_TXDLY_MASK GENMASK(6, 5) /* divider for the result of m250_sel */ @@ -44,6 +48,27 @@ #define PRG_ETH0_INVERTED_RMII_CLK BIT(11) #define PRG_ETH0_TX_AND_PHY_REF_CLK BIT(12) +/* Bypass (= 0, the signal from the GPIO input directly connects to the + * internal sampling) or enable (= 1) the internal logic for RXEN and RXD[3:0] + * timing tuning. + */ +#define PRG_ETH0_ADJ_ENABLE BIT(13) +/* Controls whether the RXEN and RXD[3:0] signals should be aligned with the + * input RX rising/falling edge and sent to the Ethernet internals. This sets + * the automatically delay and skew automatically (internally). + */ +#define PRG_ETH0_ADJ_SETUP BIT(14) +/* An internal counter based on the "timing-adjustment" clock. The counter is + * cleared on both, the falling and rising edge of the RX_CLK. This selects the + * delay (= the counter value) when to start sampling RXEN and RXD[3:0]. + */ +#define PRG_ETH0_ADJ_DELAY GENMASK(19, 15) +/* Adjusts the skew between each bit of RXEN and RXD[3:0]. If a signal has a + * large input delay, the bit for that signal (RXEN = bit 0, RXD[3] = bit 1, + * ...) can be configured to be 1 to compensate for a delay of about 1ns. + */ +#define PRG_ETH0_ADJ_SKEW GENMASK(24, 20) + #define MUX_CLK_NUM_PARENTS 2 struct meson8b_dwmac; @@ -60,6 +85,8 @@ struct meson8b_dwmac { phy_interface_t phy_mode; struct clk *rgmii_tx_clk; u32 tx_delay_ns; + u32 rx_delay_ns; + struct clk *timing_adj_clk; }; struct meson8b_dwmac_clk_configs { @@ -240,30 +267,82 @@ static int meson_axg_set_phy_mode(struct meson8b_dwmac *dwmac) return 0; } +static int meson8b_devm_clk_prepare_enable(struct meson8b_dwmac *dwmac, + struct clk *clk) +{ + int ret; + + ret = clk_prepare_enable(clk); + if (ret) + return ret; + + devm_add_action_or_reset(dwmac->dev, + (void(*)(void *))clk_disable_unprepare, + dwmac->rgmii_tx_clk); + + return 0; +} + static int meson8b_init_prg_eth(struct meson8b_dwmac *dwmac) { + u32 tx_dly_config, rx_dly_config, delay_config; int ret; - u8 tx_dly_val = 0; + + tx_dly_config = FIELD_PREP(PRG_ETH0_TXDLY_MASK, + dwmac->tx_delay_ns >> 1); + + if (dwmac->rx_delay_ns == 2) + rx_dly_config = PRG_ETH0_ADJ_ENABLE | PRG_ETH0_ADJ_SETUP; + else + rx_dly_config = 0; switch (dwmac->phy_mode) { case PHY_INTERFACE_MODE_RGMII: + delay_config = tx_dly_config | rx_dly_config; + break; case PHY_INTERFACE_MODE_RGMII_RXID: - /* TX clock delay in ns = "8ns / 4 * tx_dly_val" (where - * 8ns are exactly one cycle of the 125MHz RGMII TX clock): - * 0ns = 0x0, 2ns = 0x1, 4ns = 0x2, 6ns = 0x3 - */ - tx_dly_val = dwmac->tx_delay_ns >> 1; - /* fall through */ - - case PHY_INTERFACE_MODE_RGMII_ID: + delay_config = tx_dly_config; + break; case PHY_INTERFACE_MODE_RGMII_TXID: + delay_config = rx_dly_config; + break; + case PHY_INTERFACE_MODE_RGMII_ID: + case PHY_INTERFACE_MODE_RMII: + delay_config = 0; + break; + default: + dev_err(dwmac->dev, "unsupported phy-mode %s\n", + phy_modes(dwmac->phy_mode)); + return -EINVAL; + }; + + if (rx_dly_config & PRG_ETH0_ADJ_ENABLE) { + if (!dwmac->timing_adj_clk) { + dev_err(dwmac->dev, + "The timing-adjustment clock is mandatory for the RX delay re-timing\n"); + return -EINVAL; + } + + /* The timing adjustment logic is driven by a separate clock */ + ret = meson8b_devm_clk_prepare_enable(dwmac, + dwmac->timing_adj_clk); + if (ret) { + dev_err(dwmac->dev, + "Failed to enable the timing-adjustment clock\n"); + return ret; + } + } + + meson8b_dwmac_mask_bits(dwmac, PRG_ETH0, PRG_ETH0_TXDLY_MASK | + PRG_ETH0_ADJ_ENABLE | PRG_ETH0_ADJ_SETUP | + PRG_ETH0_ADJ_DELAY | PRG_ETH0_ADJ_SKEW, + delay_config); + + if (phy_interface_mode_is_rgmii(dwmac->phy_mode)) { /* only relevant for RMII mode -> disable in RGMII mode */ meson8b_dwmac_mask_bits(dwmac, PRG_ETH0, PRG_ETH0_INVERTED_RMII_CLK, 0); - meson8b_dwmac_mask_bits(dwmac, PRG_ETH0, PRG_ETH0_TXDLY_MASK, - tx_dly_val << PRG_ETH0_TXDLY_SHIFT); - /* Configure the 125MHz RGMII TX clock, the IP block changes * the output automatically (= without us having to configure * a register) based on the line-speed (125MHz for Gbit speeds, @@ -276,34 +355,18 @@ static int meson8b_init_prg_eth(struct meson8b_dwmac *dwmac) return ret; } - ret = clk_prepare_enable(dwmac->rgmii_tx_clk); + ret = meson8b_devm_clk_prepare_enable(dwmac, + dwmac->rgmii_tx_clk); if (ret) { dev_err(dwmac->dev, "failed to enable the RGMII TX clock\n"); return ret; } - - devm_add_action_or_reset(dwmac->dev, - (void(*)(void *))clk_disable_unprepare, - dwmac->rgmii_tx_clk); - break; - - case PHY_INTERFACE_MODE_RMII: + } else { /* invert internal clk_rmii_i to generate 25/2.5 tx_rx_clk */ meson8b_dwmac_mask_bits(dwmac, PRG_ETH0, PRG_ETH0_INVERTED_RMII_CLK, PRG_ETH0_INVERTED_RMII_CLK); - - /* TX clock delay cannot be configured in RMII mode */ - meson8b_dwmac_mask_bits(dwmac, PRG_ETH0, PRG_ETH0_TXDLY_MASK, - 0); - - break; - - default: - dev_err(dwmac->dev, "unsupported phy-mode %s\n", - phy_modes(dwmac->phy_mode)); - return -EINVAL; } /* enable TX_CLK and PHY_REF_CLK generator */ @@ -358,6 +421,25 @@ static int meson8b_dwmac_probe(struct platform_device *pdev) &dwmac->tx_delay_ns)) dwmac->tx_delay_ns = 2; + /* use 0ns as fallback since this is what most boards actually use */ + if (of_property_read_u32(pdev->dev.of_node, "amlogic,rx-delay-ns", + &dwmac->rx_delay_ns)) + dwmac->rx_delay_ns = 0; + + if (dwmac->rx_delay_ns != 0 && dwmac->rx_delay_ns != 2) { + dev_err(&pdev->dev, + "The only allowed RX delays values are: 0ns, 2ns"); + ret = -EINVAL; + goto err_remove_config_dt; + } + + dwmac->timing_adj_clk = devm_clk_get_optional(dwmac->dev, + "timing-adjustment"); + if (IS_ERR(dwmac->timing_adj_clk)) { + ret = PTR_ERR(dwmac->timing_adj_clk); + goto err_remove_config_dt; + } + ret = meson8b_init_rgmii_tx_clk(dwmac); if (ret) goto err_remove_config_dt; diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c index b2dc99289687..5d4df4c5254e 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-stm32.c @@ -29,6 +29,11 @@ #define SYSCFG_PMCR_ETH_CLK_SEL BIT(16) #define SYSCFG_PMCR_ETH_REF_CLK_SEL BIT(17) +/* CLOCK feed to PHY*/ +#define ETH_CK_F_25M 25000000 +#define ETH_CK_F_50M 50000000 +#define ETH_CK_F_125M 125000000 + /* Ethernet PHY interface selection in register SYSCFG Configuration *------------------------------------------ * src |BIT(23)| BIT(22)| BIT(21)|BIT(20)| @@ -58,33 +63,20 @@ *| | | 25MHz | 50MHz | | * --------------------------------------------------------------------------- *| MII | - | eth-ck | n/a | n/a | - *| | | | | | + *| | | st,ext-phyclk | | | * --------------------------------------------------------------------------- *| GMII | - | eth-ck | n/a | n/a | - *| | | | | | + *| | | st,ext-phyclk | | | * --------------------------------------------------------------------------- - *| RGMII | - | eth-ck | n/a | eth-ck (no pin) | - *| | | | | st,eth-clk-sel | + *| RGMII | - | eth-ck | n/a | eth-ck | + *| | | st,ext-phyclk | | st,eth-clk-sel or| + *| | | | | st,ext-phyclk | * --------------------------------------------------------------------------- *| RMII | - | eth-ck | eth-ck | n/a | - *| | | | st,eth-ref-clk-sel | | + *| | | st,ext-phyclk | st,eth-ref-clk-sel | | + *| | | | or st,ext-phyclk | | * --------------------------------------------------------------------------- * - * BIT(17) : set this bit in RMII mode when you have PHY without crystal 50MHz - * BIT(16) : set this bit in GMII/RGMII PHY when you do not want use 125Mhz - * from PHY - *----------------------------------------------------- - * src | BIT(17) | BIT(16) | - *----------------------------------------------------- - * MII | n/a | n/a | - *----------------------------------------------------- - * GMII | n/a | st,eth-clk-sel | - *----------------------------------------------------- - * RGMII | n/a | st,eth-clk-sel | - *----------------------------------------------------- - * RMII | st,eth-ref-clk-sel | n/a | - *----------------------------------------------------- - * */ struct stm32_dwmac { @@ -93,6 +85,8 @@ struct stm32_dwmac { struct clk *clk_eth_ck; struct clk *clk_ethstp; struct clk *syscfg_clk; + int ext_phyclk; + int enable_eth_ck; int eth_clk_sel_reg; int eth_ref_clk_sel_reg; int irq_pwr_wakeup; @@ -155,14 +149,17 @@ static int stm32mp1_clk_prepare(struct stm32_dwmac *dwmac, bool prepare) ret = clk_prepare_enable(dwmac->syscfg_clk); if (ret) return ret; - ret = clk_prepare_enable(dwmac->clk_eth_ck); - if (ret) { - clk_disable_unprepare(dwmac->syscfg_clk); - return ret; + if (dwmac->enable_eth_ck) { + ret = clk_prepare_enable(dwmac->clk_eth_ck); + if (ret) { + clk_disable_unprepare(dwmac->syscfg_clk); + return ret; + } } } else { clk_disable_unprepare(dwmac->syscfg_clk); - clk_disable_unprepare(dwmac->clk_eth_ck); + if (dwmac->enable_eth_ck) + clk_disable_unprepare(dwmac->clk_eth_ck); } return ret; } @@ -170,24 +167,34 @@ static int stm32mp1_clk_prepare(struct stm32_dwmac *dwmac, bool prepare) static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat) { struct stm32_dwmac *dwmac = plat_dat->bsp_priv; - u32 reg = dwmac->mode_reg; + u32 reg = dwmac->mode_reg, clk_rate; int val; + clk_rate = clk_get_rate(dwmac->clk_eth_ck); + dwmac->enable_eth_ck = false; switch (plat_dat->interface) { case PHY_INTERFACE_MODE_MII: + if (clk_rate == ETH_CK_F_25M && dwmac->ext_phyclk) + dwmac->enable_eth_ck = true; val = SYSCFG_PMCR_ETH_SEL_MII; pr_debug("SYSCFG init : PHY_INTERFACE_MODE_MII\n"); break; case PHY_INTERFACE_MODE_GMII: val = SYSCFG_PMCR_ETH_SEL_GMII; - if (dwmac->eth_clk_sel_reg) + if (clk_rate == ETH_CK_F_25M && + (dwmac->eth_clk_sel_reg || dwmac->ext_phyclk)) { + dwmac->enable_eth_ck = true; val |= SYSCFG_PMCR_ETH_CLK_SEL; + } pr_debug("SYSCFG init : PHY_INTERFACE_MODE_GMII\n"); break; case PHY_INTERFACE_MODE_RMII: val = SYSCFG_PMCR_ETH_SEL_RMII; - if (dwmac->eth_ref_clk_sel_reg) + if ((clk_rate == ETH_CK_F_25M || clk_rate == ETH_CK_F_50M) && + (dwmac->eth_ref_clk_sel_reg || dwmac->ext_phyclk)) { + dwmac->enable_eth_ck = true; val |= SYSCFG_PMCR_ETH_REF_CLK_SEL; + } pr_debug("SYSCFG init : PHY_INTERFACE_MODE_RMII\n"); break; case PHY_INTERFACE_MODE_RGMII: @@ -195,8 +202,11 @@ static int stm32mp1_set_mode(struct plat_stmmacenet_data *plat_dat) case PHY_INTERFACE_MODE_RGMII_RXID: case PHY_INTERFACE_MODE_RGMII_TXID: val = SYSCFG_PMCR_ETH_SEL_RGMII; - if (dwmac->eth_clk_sel_reg) + if ((clk_rate == ETH_CK_F_25M || clk_rate == ETH_CK_F_125M) && + (dwmac->eth_clk_sel_reg || dwmac->ext_phyclk)) { + dwmac->enable_eth_ck = true; val |= SYSCFG_PMCR_ETH_CLK_SEL; + } pr_debug("SYSCFG init : PHY_INTERFACE_MODE_RGMII\n"); break; default: @@ -294,6 +304,9 @@ static int stm32mp1_parse_data(struct stm32_dwmac *dwmac, struct device_node *np = dev->of_node; int err = 0; + /* Ethernet PHY have no crystal */ + dwmac->ext_phyclk = of_property_read_bool(np, "st,ext-phyclk"); + /* Gigabit Ethernet 125MHz clock selection. */ dwmac->eth_clk_sel_reg = of_property_read_bool(np, "st,eth-clk-sel"); @@ -431,7 +444,8 @@ static int stm32mp1_suspend(struct stm32_dwmac *dwmac) clk_disable_unprepare(dwmac->clk_tx); clk_disable_unprepare(dwmac->syscfg_clk); - clk_disable_unprepare(dwmac->clk_eth_ck); + if (dwmac->enable_eth_ck) + clk_disable_unprepare(dwmac->clk_eth_ck); return ret; } diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h index 28cac28253b8..61f3249bd724 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4.h @@ -90,6 +90,7 @@ #define GMAC_VLAN_CSVL BIT(19) #define GMAC_VLAN_VLC GENMASK(17, 16) #define GMAC_VLAN_VLC_SHIFT 16 +#define GMAC_VLAN_VLHT GENMASK(15, 0) /* MAC VLAN Tag */ #define GMAC_VLAN_TAG_VID GENMASK(15, 0) diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c index 39692d15d80c..ecd834e0e121 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac4_core.c @@ -450,6 +450,12 @@ static int dwmac4_add_hw_vlan_rx_fltr(struct net_device *dev, if (vid > 4095) return -EINVAL; + if (hw->promisc) { + netdev_err(dev, + "Adding VLAN in promisc mode not supported\n"); + return -EPERM; + } + /* Single Rx VLAN Filter */ if (hw->num_vlan == 1) { /* For single VLAN filter, VID 0 means VLAN promiscuous */ @@ -499,6 +505,12 @@ static int dwmac4_del_hw_vlan_rx_fltr(struct net_device *dev, { int i, ret = 0; + if (hw->promisc) { + netdev_err(dev, + "Deleting VLAN in promisc mode not supported\n"); + return -EPERM; + } + /* Single Rx VLAN Filter */ if (hw->num_vlan == 1) { if ((hw->vlan_filter[0] & GMAC_VLAN_TAG_VID) == vid) { @@ -523,9 +535,45 @@ static int dwmac4_del_hw_vlan_rx_fltr(struct net_device *dev, return ret; } +static void dwmac4_vlan_promisc_enable(struct net_device *dev, + struct mac_device_info *hw) +{ + void __iomem *ioaddr = hw->pcsr; + u32 value; + u32 hash; + u32 val; + int i; + + /* Single Rx VLAN Filter */ + if (hw->num_vlan == 1) { + dwmac4_write_single_vlan(dev, 0); + return; + } + + /* Extended Rx VLAN Filter Enable */ + for (i = 0; i < hw->num_vlan; i++) { + if (hw->vlan_filter[i] & GMAC_VLAN_TAG_DATA_VEN) { + val = hw->vlan_filter[i] & ~GMAC_VLAN_TAG_DATA_VEN; + dwmac4_write_vlan_filter(dev, hw, i, val); + } + } + + hash = readl(ioaddr + GMAC_VLAN_HASH_TABLE); + if (hash & GMAC_VLAN_VLHT) { + value = readl(ioaddr + GMAC_VLAN_TAG); + if (value & GMAC_VLAN_VTHM) { + value &= ~GMAC_VLAN_VTHM; + writel(value, ioaddr + GMAC_VLAN_TAG); + } + } +} + static void dwmac4_restore_hw_vlan_rx_fltr(struct net_device *dev, struct mac_device_info *hw) { + void __iomem *ioaddr = hw->pcsr; + u32 value; + u32 hash; u32 val; int i; @@ -542,6 +590,13 @@ static void dwmac4_restore_hw_vlan_rx_fltr(struct net_device *dev, dwmac4_write_vlan_filter(dev, hw, i, val); } } + + hash = readl(ioaddr + GMAC_VLAN_HASH_TABLE); + if (hash & GMAC_VLAN_VLHT) { + value = readl(ioaddr + GMAC_VLAN_TAG); + value |= GMAC_VLAN_VTHM; + writel(value, ioaddr + GMAC_VLAN_TAG); + } } static void dwmac4_set_filter(struct mac_device_info *hw, @@ -624,6 +679,18 @@ static void dwmac4_set_filter(struct mac_device_info *hw, value |= GMAC_PACKET_FILTER_VTFE; writel(value, ioaddr + GMAC_PACKET_FILTER); + + if (dev->flags & IFF_PROMISC) { + if (!hw->promisc) { + hw->promisc = 1; + dwmac4_vlan_promisc_enable(dev, hw); + } + } else { + if (hw->promisc) { + hw->promisc = 0; + dwmac4_restore_hw_vlan_rx_fltr(dev, hw); + } + } } static void dwmac4_flow_ctrl(struct mac_device_info *hw, unsigned int duplex, diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index a999d6b33a64..73677c3b33b6 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -630,7 +630,8 @@ static int stmmac_hwtstamp_set(struct net_device *dev, struct ifreq *ifr) config.rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT; ptp_v2 = PTP_TCR_TSVER2ENA; snap_type_sel = PTP_TCR_SNAPTYPSEL_1; - ts_event_en = PTP_TCR_TSEVNTENA; + if (priv->synopsys_id != DWMAC_CORE_5_10) + ts_event_en = PTP_TCR_TSEVNTENA; ptp_over_ipv4_udp = PTP_TCR_TSIPV4ENA; ptp_over_ipv6_udp = PTP_TCR_TSIPV6ENA; ptp_over_ethernet = PTP_TCR_TSIPENA; @@ -3543,15 +3544,6 @@ static void stmmac_rx_vlan(struct net_device *dev, struct sk_buff *skb) } } - -static inline int stmmac_rx_threshold_count(struct stmmac_rx_queue *rx_q) -{ - if (rx_q->rx_zeroc_thresh < STMMAC_RX_THRESH) - return 0; - - return 1; -} - /** * stmmac_rx_refill - refill used skb preallocated buffers * @priv: driver private structure @@ -5190,8 +5182,6 @@ int stmmac_resume(struct device *dev) return ret; } - netif_device_attach(ndev); - mutex_lock(&priv->lock); stmmac_reset_queues_param(priv); @@ -5218,6 +5208,8 @@ int stmmac_resume(struct device *dev) phylink_mac_change(priv->phylink, true); + netif_device_attach(ndev); + return 0; } EXPORT_SYMBOL_GPL(stmmac_resume); diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c index 3fb21f7ac9fb..272cb47af9f2 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_pci.c @@ -217,15 +217,10 @@ static int stmmac_pci_probe(struct pci_dev *pdev, */ static void stmmac_pci_remove(struct pci_dev *pdev) { - struct net_device *ndev = dev_get_drvdata(&pdev->dev); - struct stmmac_priv *priv = netdev_priv(ndev); int i; stmmac_dvr_remove(&pdev->dev); - if (priv->plat->stmmac_clk) - clk_unregister_fixed_rate(priv->plat->stmmac_clk); - for (i = 0; i < PCI_STD_NUM_BARS; i++) { if (pci_resource_len(pdev, i) == 0) continue; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index bcda49dcf619..f32317fa75c8 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -507,7 +507,8 @@ stmmac_probe_config_dt(struct platform_device *pdev, const char **mac) if (of_device_is_compatible(np, "snps,dwmac-4.00") || of_device_is_compatible(np, "snps,dwmac-4.10a") || - of_device_is_compatible(np, "snps,dwmac-4.20a")) { + of_device_is_compatible(np, "snps,dwmac-4.20a") || + of_device_is_compatible(np, "snps,dwmac-5.10a")) { plat->has_gmac4 = 1; plat->has_gmac = 0; plat->pmt = 1; diff --git a/drivers/net/ethernet/sun/cassini.c b/drivers/net/ethernet/sun/cassini.c index e6d1aa882fa5..debd3c3fa6fb 100644 --- a/drivers/net/ethernet/sun/cassini.c +++ b/drivers/net/ethernet/sun/cassini.c @@ -237,12 +237,6 @@ static inline void cas_lock_tx(struct cas *cp) spin_lock_nested(&cp->tx_lock[i], i); } -static inline void cas_lock_all(struct cas *cp) -{ - spin_lock_irq(&cp->lock); - cas_lock_tx(cp); -} - /* WTZ: QA was finding deadlock problems with the previous * versions after long test runs with multiple cards per machine. * See if replacing cas_lock_all with safer versions helps. The @@ -266,12 +260,6 @@ static inline void cas_unlock_tx(struct cas *cp) spin_unlock(&cp->tx_lock[i - 1]); } -static inline void cas_unlock_all(struct cas *cp) -{ - cas_unlock_tx(cp); - spin_unlock_irq(&cp->lock); -} - #define cas_unlock_all_restore(cp, flags) \ do { \ struct cas *xxxcp = (cp); \ @@ -4963,7 +4951,7 @@ static int cas_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) cas_cacheline_size)) { dev_err(&pdev->dev, "Could not set PCI cache " "line size\n"); - goto err_write_cacheline; + goto err_out_free_res; } } #endif @@ -5059,7 +5047,7 @@ static int cas_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (cp->cas_flags & CAS_FLAG_SATURN) cas_saturn_firmware_init(cp); - cp->init_block = (struct cas_init_block *) + cp->init_block = pci_alloc_consistent(pdev, sizeof(struct cas_init_block), &cp->block_dvma); if (!cp->init_block) { @@ -5136,7 +5124,6 @@ err_out_iounmap: err_out_free_res: pci_release_regions(pdev); -err_write_cacheline: /* Try to restore it in case the error occurred after we * set it. */ diff --git a/drivers/net/ethernet/tehuti/tehuti.c b/drivers/net/ethernet/tehuti/tehuti.c index 40a2ce0ca808..e28727297563 100644 --- a/drivers/net/ethernet/tehuti/tehuti.c +++ b/drivers/net/ethernet/tehuti/tehuti.c @@ -1362,18 +1362,6 @@ static void print_rxfd(struct rxf_desc *rxfd) * As our benchmarks shows, it adds 1.5 Gbit/sec to NIS's throuput. */ -/************************************************************************* - * Tx DB * - *************************************************************************/ -static inline int bdx_tx_db_size(struct txdb *db) -{ - int taken = db->wptr - db->rptr; - if (taken < 0) - taken = db->size + 1 + taken; /* (size + 1) equals memsz */ - - return db->size - taken; -} - /** * __bdx_tx_db_ptr_next - helper function, increment read/write pointer + wrap * @db: tx data base diff --git a/drivers/net/ethernet/ti/Kconfig b/drivers/net/ethernet/ti/Kconfig index 62f809b67469..182d10f171f6 100644 --- a/drivers/net/ethernet/ti/Kconfig +++ b/drivers/net/ethernet/ti/Kconfig @@ -94,6 +94,7 @@ config TI_K3_AM65_CPSW_NUSS depends on ARCH_K3 && OF && TI_K3_UDMA_GLUE_LAYER select TI_DAVINCI_MDIO imply PHY_TI_GMII_SEL + depends on TI_K3_AM65_CPTS || !TI_K3_AM65_CPTS help This driver supports TI K3 AM654/J721E CPSW2G Ethernet SubSystem. The two-port Gigabit Ethernet MAC (MCU_CPSW0) subsystem provides @@ -104,6 +105,28 @@ config TI_K3_AM65_CPSW_NUSS To compile this driver as a module, choose M here: the module will be called ti-am65-cpsw-nuss. +config TI_K3_AM65_CPTS + tristate "TI K3 AM65x CPTS" + depends on ARCH_K3 && OF + depends on PTP_1588_CLOCK + help + Say y here to support the TI K3 AM65x CPTS with 1588 features such as + PTP hardware clock for each CPTS device and network packets + timestamping where applicable. + Depending on integration CPTS blocks enable compliance with + the IEEE 1588-2008 standard for a precision clock synchronization + protocol, Ethernet Enhanced Scheduled Traffic Operations (CPTS_ESTFn) + and PCIe Subsystem Precision Time Measurement (PTM). + +config TI_AM65_CPSW_TAS + bool "Enable TAS offload in AM65 CPSW" + depends on TI_K3_AM65_CPSW_NUSS && NET_SCH_TAPRIO && TI_K3_AM65_CPTS + help + Say y here to support Time Aware Shaper(TAS) offload in AM65 CPSW. + AM65 CPSW hardware supports Enhanced Scheduled Traffic (EST) + defined in IEEE 802.1Q 2018. The EST scheduler runs on CPTS and the + TAS/EST schedule is updated in the Fetch RAM memory of the CPSW. + config TI_KEYSTONE_NETCP tristate "TI Keystone NETCP Core Support" select TI_DAVINCI_MDIO @@ -133,7 +156,7 @@ config TLAN Devices currently supported by this driver are Compaq Netelligent, Compaq NetFlex and Olicom cards. Please read the file - <file:Documentation/networking/device_drivers/ti/tlan.txt> + <file:Documentation/networking/device_drivers/ti/tlan.rst> for more details. To compile this driver as a module, choose M here. The module diff --git a/drivers/net/ethernet/ti/Makefile b/drivers/net/ethernet/ti/Makefile index cb26a9d21869..6e779292545d 100644 --- a/drivers/net/ethernet/ti/Makefile +++ b/drivers/net/ethernet/ti/Makefile @@ -25,4 +25,5 @@ obj-$(CONFIG_TI_KEYSTONE_NETCP_ETHSS) += keystone_netcp_ethss.o keystone_netcp_ethss-y := netcp_ethss.o netcp_sgmii.o netcp_xgbepcsr.o cpsw_ale.o obj-$(CONFIG_TI_K3_AM65_CPSW_NUSS) += ti-am65-cpsw-nuss.o -ti-am65-cpsw-nuss-y := am65-cpsw-nuss.o cpsw_sl.o am65-cpsw-ethtool.o cpsw_ale.o k3-cppi-desc-pool.o +ti-am65-cpsw-nuss-y := am65-cpsw-nuss.o cpsw_sl.o am65-cpsw-ethtool.o cpsw_ale.o k3-cppi-desc-pool.o am65-cpsw-qos.o +obj-$(CONFIG_TI_K3_AM65_CPTS) += am65-cpts.o diff --git a/drivers/net/ethernet/ti/am65-cpsw-ethtool.c b/drivers/net/ethernet/ti/am65-cpsw-ethtool.c index c3502aa15ea0..8c4690f3ebcb 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-ethtool.c +++ b/drivers/net/ethernet/ti/am65-cpsw-ethtool.c @@ -12,6 +12,7 @@ #include "am65-cpsw-nuss.h" #include "cpsw_ale.h" +#include "am65-cpts.h" #define AM65_CPSW_REGDUMP_VER 0x1 @@ -694,6 +695,27 @@ static void am65_cpsw_get_ethtool_stats(struct net_device *ndev, hw_stats[i].offset); } +static int am65_cpsw_get_ethtool_ts_info(struct net_device *ndev, + struct ethtool_ts_info *info) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + + if (!IS_ENABLED(CONFIG_TI_K3_AM65_CPTS)) + return ethtool_op_get_ts_info(ndev, info); + + info->so_timestamping = + SOF_TIMESTAMPING_TX_HARDWARE | + SOF_TIMESTAMPING_TX_SOFTWARE | + SOF_TIMESTAMPING_RX_HARDWARE | + SOF_TIMESTAMPING_RX_SOFTWARE | + SOF_TIMESTAMPING_SOFTWARE | + SOF_TIMESTAMPING_RAW_HARDWARE; + info->phc_index = am65_cpts_phc_index(common->cpts); + info->tx_types = BIT(HWTSTAMP_TX_OFF) | BIT(HWTSTAMP_TX_ON); + info->rx_filters = BIT(HWTSTAMP_FILTER_NONE) | BIT(HWTSTAMP_FILTER_ALL); + return 0; +} + static u32 am65_cpsw_get_ethtool_priv_flags(struct net_device *ndev) { struct am65_cpsw_common *common = am65_ndev_to_common(ndev); @@ -708,9 +730,17 @@ static u32 am65_cpsw_get_ethtool_priv_flags(struct net_device *ndev) static int am65_cpsw_set_ethtool_priv_flags(struct net_device *ndev, u32 flags) { struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + int rrobin; + + rrobin = !!(flags & AM65_CPSW_PRIV_P0_RX_PTYPE_RROBIN); + + if (common->est_enabled && rrobin) { + netdev_err(ndev, + "p0-rx-ptype-rrobin flag conflicts with QOS\n"); + return -EINVAL; + } - common->pf_p0_rx_ptype_rrobin = - !!(flags & AM65_CPSW_PRIV_P0_RX_PTYPE_RROBIN); + common->pf_p0_rx_ptype_rrobin = rrobin; am65_cpsw_nuss_set_p0_ptype(common); return 0; @@ -730,7 +760,7 @@ const struct ethtool_ops am65_cpsw_ethtool_ops_slave = { .get_sset_count = am65_cpsw_get_sset_count, .get_strings = am65_cpsw_get_strings, .get_ethtool_stats = am65_cpsw_get_ethtool_stats, - .get_ts_info = ethtool_op_get_ts_info, + .get_ts_info = am65_cpsw_get_ethtool_ts_info, .get_priv_flags = am65_cpsw_get_ethtool_priv_flags, .set_priv_flags = am65_cpsw_set_ethtool_priv_flags, diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.c b/drivers/net/ethernet/ti/am65-cpsw-nuss.c index 2517ffba8178..87a4775ed53a 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.c +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.c @@ -30,18 +30,21 @@ #include "cpsw_sl.h" #include "am65-cpsw-nuss.h" #include "k3-cppi-desc-pool.h" +#include "am65-cpts.h" #define AM65_CPSW_SS_BASE 0x0 #define AM65_CPSW_SGMII_BASE 0x100 #define AM65_CPSW_XGMII_BASE 0x2100 #define AM65_CPSW_CPSW_NU_BASE 0x20000 #define AM65_CPSW_NU_PORTS_BASE 0x1000 +#define AM65_CPSW_NU_FRAM_BASE 0x12000 #define AM65_CPSW_NU_STATS_BASE 0x1a000 #define AM65_CPSW_NU_ALE_BASE 0x1e000 #define AM65_CPSW_NU_CPTS_BASE 0x1d000 #define AM65_CPSW_NU_PORTS_OFFSET 0x1000 #define AM65_CPSW_NU_STATS_PORT_OFFSET 0x200 +#define AM65_CPSW_NU_FRAM_PORT_OFFSET 0x200 #define AM65_CPSW_MAX_PORTS 8 @@ -187,9 +190,11 @@ void am65_cpsw_nuss_adjust_link(struct net_device *ndev) cpsw_ale_control_set(common->ale, port->port_id, ALE_PORT_STATE, ALE_PORT_STATE_FORWARD); + am65_cpsw_qos_link_up(ndev, phy->speed); netif_tx_wake_all_queues(ndev); } else { int tmo; + /* disable forwarding */ cpsw_ale_control_set(common->ale, port->port_id, ALE_PORT_STATE, ALE_PORT_STATE_DISABLE); @@ -203,6 +208,7 @@ void am65_cpsw_nuss_adjust_link(struct net_device *ndev) cpsw_sl_ctl_reset(port->slave.mac_sl); + am65_cpsw_qos_link_down(ndev); netif_tx_stop_all_queues(ndev); } @@ -668,6 +674,18 @@ static void am65_cpsw_nuss_rx_cleanup(void *data, dma_addr_t desc_dma) dev_kfree_skb_any(skb); } +static void am65_cpsw_nuss_rx_ts(struct sk_buff *skb, u32 *psdata) +{ + struct skb_shared_hwtstamps *ssh; + u64 ns; + + ns = ((u64)psdata[1] << 32) | psdata[0]; + + ssh = skb_hwtstamps(skb); + memset(ssh, 0, sizeof(*ssh)); + ssh->hwtstamp = ns_to_ktime(ns); +} + /* RX psdata[2] word format - checksum information */ #define AM65_CPSW_RX_PSD_CSUM_ADD GENMASK(15, 0) #define AM65_CPSW_RX_PSD_CSUM_ERR BIT(16) @@ -745,6 +763,9 @@ static int am65_cpsw_nuss_rx_packets(struct am65_cpsw_common *common, skb->dev = ndev; psdata = cppi5_hdesc_get_psdata(desc_rx); + /* add RX timestamp */ + if (port->rx_ts_enabled) + am65_cpsw_nuss_rx_ts(skb, psdata); csum_info = psdata[2]; dev_dbg(dev, "%s rx csum_info:%#x\n", __func__, csum_info); @@ -904,6 +925,8 @@ static int am65_cpsw_nuss_tx_compl_packets(struct am65_cpsw_common *common, ndev = skb->dev; + am65_cpts_tx_timestamp(common->cpts, skb); + ndev_priv = netdev_priv(ndev); stats = this_cpu_ptr(ndev_priv->stats); u64_stats_update_begin(&stats->syncp); @@ -995,6 +1018,10 @@ static netdev_tx_t am65_cpsw_nuss_ndo_slave_xmit(struct sk_buff *skb, /* padding enabled in hw */ pkt_len = skb_headlen(skb); + /* SKB TX timestamp */ + if (port->tx_ts_enabled) + am65_cpts_prep_tx_timestamp(common->cpts, skb); + q_idx = skb_get_queue_mapping(skb); dev_dbg(dev, "%s skb_queue:%d\n", __func__, q_idx); @@ -1158,6 +1185,111 @@ static int am65_cpsw_nuss_ndo_slave_set_mac_address(struct net_device *ndev, return 0; } +static int am65_cpsw_nuss_hwtstamp_set(struct net_device *ndev, + struct ifreq *ifr) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + u32 ts_ctrl, seq_id, ts_ctrl_ltype2, ts_vlan_ltype; + struct hwtstamp_config cfg; + + if (!IS_ENABLED(CONFIG_TI_K3_AM65_CPTS)) + return -EOPNOTSUPP; + + if (copy_from_user(&cfg, ifr->ifr_data, sizeof(cfg))) + return -EFAULT; + + /* TX HW timestamp */ + switch (cfg.tx_type) { + case HWTSTAMP_TX_OFF: + case HWTSTAMP_TX_ON: + break; + default: + return -ERANGE; + } + + switch (cfg.rx_filter) { + case HWTSTAMP_FILTER_NONE: + port->rx_ts_enabled = false; + break; + case HWTSTAMP_FILTER_ALL: + case HWTSTAMP_FILTER_SOME: + case HWTSTAMP_FILTER_PTP_V1_L4_EVENT: + case HWTSTAMP_FILTER_PTP_V1_L4_SYNC: + case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ: + case HWTSTAMP_FILTER_PTP_V2_L4_EVENT: + case HWTSTAMP_FILTER_PTP_V2_L4_SYNC: + case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ: + case HWTSTAMP_FILTER_PTP_V2_L2_EVENT: + case HWTSTAMP_FILTER_PTP_V2_L2_SYNC: + case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ: + case HWTSTAMP_FILTER_PTP_V2_EVENT: + case HWTSTAMP_FILTER_PTP_V2_SYNC: + case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ: + case HWTSTAMP_FILTER_NTP_ALL: + port->rx_ts_enabled = true; + cfg.rx_filter = HWTSTAMP_FILTER_ALL; + break; + default: + return -ERANGE; + } + + port->tx_ts_enabled = (cfg.tx_type == HWTSTAMP_TX_ON); + + /* cfg TX timestamp */ + seq_id = (AM65_CPSW_TS_SEQ_ID_OFFSET << + AM65_CPSW_PN_TS_SEQ_ID_OFFSET_SHIFT) | ETH_P_1588; + + ts_vlan_ltype = ETH_P_8021Q; + + ts_ctrl_ltype2 = ETH_P_1588 | + AM65_CPSW_PN_TS_CTL_LTYPE2_TS_107 | + AM65_CPSW_PN_TS_CTL_LTYPE2_TS_129 | + AM65_CPSW_PN_TS_CTL_LTYPE2_TS_130 | + AM65_CPSW_PN_TS_CTL_LTYPE2_TS_131 | + AM65_CPSW_PN_TS_CTL_LTYPE2_TS_132 | + AM65_CPSW_PN_TS_CTL_LTYPE2_TS_319 | + AM65_CPSW_PN_TS_CTL_LTYPE2_TS_320 | + AM65_CPSW_PN_TS_CTL_LTYPE2_TS_TTL_NONZERO; + + ts_ctrl = AM65_CPSW_TS_EVENT_MSG_TYPE_BITS << + AM65_CPSW_PN_TS_CTL_MSG_TYPE_EN_SHIFT; + + if (port->tx_ts_enabled) + ts_ctrl |= AM65_CPSW_TS_TX_ANX_ALL_EN | + AM65_CPSW_PN_TS_CTL_TX_VLAN_LT1_EN; + + writel(seq_id, port->port_base + AM65_CPSW_PORTN_REG_TS_SEQ_LTYPE_REG); + writel(ts_vlan_ltype, port->port_base + + AM65_CPSW_PORTN_REG_TS_VLAN_LTYPE_REG); + writel(ts_ctrl_ltype2, port->port_base + + AM65_CPSW_PORTN_REG_TS_CTL_LTYPE2); + writel(ts_ctrl, port->port_base + AM65_CPSW_PORTN_REG_TS_CTL); + + /* en/dis RX timestamp */ + am65_cpts_rx_enable(common->cpts, port->rx_ts_enabled); + + return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0; +} + +static int am65_cpsw_nuss_hwtstamp_get(struct net_device *ndev, + struct ifreq *ifr) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + struct hwtstamp_config cfg; + + if (!IS_ENABLED(CONFIG_TI_K3_AM65_CPTS)) + return -EOPNOTSUPP; + + cfg.flags = 0; + cfg.tx_type = port->tx_ts_enabled ? + HWTSTAMP_TX_ON : HWTSTAMP_TX_OFF; + cfg.rx_filter = port->rx_ts_enabled ? + HWTSTAMP_FILTER_ALL : HWTSTAMP_FILTER_NONE; + + return copy_to_user(ifr->ifr_data, &cfg, sizeof(cfg)) ? -EFAULT : 0; +} + static int am65_cpsw_nuss_ndo_slave_ioctl(struct net_device *ndev, struct ifreq *req, int cmd) { @@ -1166,6 +1298,13 @@ static int am65_cpsw_nuss_ndo_slave_ioctl(struct net_device *ndev, if (!netif_running(ndev)) return -EINVAL; + switch (cmd) { + case SIOCSHWTSTAMP: + return am65_cpsw_nuss_hwtstamp_set(ndev, req); + case SIOCGHWTSTAMP: + return am65_cpsw_nuss_hwtstamp_get(ndev, req); + } + if (!port->slave.phy) return -EOPNOTSUPP; @@ -1244,6 +1383,7 @@ static const struct net_device_ops am65_cpsw_nuss_netdev_ops_2g = { .ndo_vlan_rx_kill_vid = am65_cpsw_nuss_ndo_slave_kill_vid, .ndo_do_ioctl = am65_cpsw_nuss_ndo_slave_ioctl, .ndo_set_features = am65_cpsw_nuss_ndo_slave_set_features, + .ndo_setup_tc = am65_cpsw_qos_ndo_setup_tc, }; static void am65_cpsw_nuss_slave_disable_unused(struct am65_cpsw_port *port) @@ -1531,6 +1671,40 @@ static int am65_cpsw_am654_get_efuse_macid(struct device_node *of_node, return 0; } +static int am65_cpsw_init_cpts(struct am65_cpsw_common *common) +{ + struct device *dev = common->dev; + struct device_node *node; + struct am65_cpts *cpts; + void __iomem *reg_base; + + if (!IS_ENABLED(CONFIG_TI_K3_AM65_CPTS)) + return 0; + + node = of_get_child_by_name(dev->of_node, "cpts"); + if (!node) { + dev_err(dev, "%s cpts not found\n", __func__); + return -ENOENT; + } + + reg_base = common->cpsw_base + AM65_CPSW_NU_CPTS_BASE; + cpts = am65_cpts_create(dev, reg_base, node); + if (IS_ERR(cpts)) { + int ret = PTR_ERR(cpts); + + if (ret == -EOPNOTSUPP) { + dev_info(dev, "cpts disabled\n"); + return 0; + } + + dev_err(dev, "cpts create err %d\n", ret); + return ret; + } + common->cpts = cpts; + + return 0; +} + static int am65_cpsw_nuss_init_slave_ports(struct am65_cpsw_common *common) { struct device_node *node, *port_np; @@ -1571,6 +1745,9 @@ static int am65_cpsw_nuss_init_slave_ports(struct am65_cpsw_common *common) port->stat_base = common->cpsw_base + AM65_CPSW_NU_STATS_BASE + (AM65_CPSW_NU_STATS_PORT_OFFSET * port_id); port->name = of_get_property(port_np, "label", NULL); + port->fetch_ram_base = + common->cpsw_base + AM65_CPSW_NU_FRAM_BASE + + (AM65_CPSW_NU_FRAM_PORT_OFFSET * (port_id - 1)); port->disabled = !of_device_is_available(port_np); if (port->disabled) @@ -1863,10 +2040,21 @@ static int am65_cpsw_nuss_probe(struct platform_device *pdev) return ret; } - ret = of_platform_populate(dev->of_node, NULL, NULL, dev); - /* We do not want to force this, as in some cases may not have child */ - if (ret) - dev_warn(dev, "populating child nodes err:%d\n", ret); + node = of_get_child_by_name(dev->of_node, "mdio"); + if (!node) { + dev_warn(dev, "MDIO node not found\n"); + } else if (of_device_is_available(node)) { + struct platform_device *mdio_pdev; + + mdio_pdev = of_platform_device_create(node, NULL, dev); + if (!mdio_pdev) { + ret = -ENODEV; + goto err_pm_clear; + } + + common->mdio_dev = &mdio_pdev->dev; + } + of_node_put(node); am65_cpsw_nuss_get_ver(common); @@ -1895,11 +2083,16 @@ static int am65_cpsw_nuss_probe(struct platform_device *pdev) ale_params.nu_switch_ale = true; common->ale = cpsw_ale_create(&ale_params); - if (!common->ale) { + if (IS_ERR(common->ale)) { dev_err(dev, "error initializing ale engine\n"); + ret = PTR_ERR(common->ale); goto err_of_clear; } + ret = am65_cpsw_init_cpts(common); + if (ret) + goto err_of_clear; + /* init ports */ for (i = 0; i < common->port_num; i++) am65_cpsw_nuss_slave_disable_unused(&common->ports[i]); @@ -1918,7 +2111,8 @@ static int am65_cpsw_nuss_probe(struct platform_device *pdev) return 0; err_of_clear: - of_platform_depopulate(dev); + of_platform_device_destroy(common->mdio_dev, NULL); +err_pm_clear: pm_runtime_put_sync(dev); pm_runtime_disable(dev); return ret; @@ -1943,7 +2137,7 @@ static int am65_cpsw_nuss_remove(struct platform_device *pdev) */ am65_cpsw_nuss_cleanup_ndev(common); - of_platform_depopulate(dev); + of_platform_device_destroy(common->mdio_dev, NULL); pm_runtime_put_sync(&pdev->dev); pm_runtime_disable(&pdev->dev); diff --git a/drivers/net/ethernet/ti/am65-cpsw-nuss.h b/drivers/net/ethernet/ti/am65-cpsw-nuss.h index 41ae5b4c7931..9faf4fb1409b 100644 --- a/drivers/net/ethernet/ti/am65-cpsw-nuss.h +++ b/drivers/net/ethernet/ti/am65-cpsw-nuss.h @@ -9,6 +9,11 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/netdevice.h> +#include <linux/phy.h> +#include <linux/platform_device.h> +#include "am65-cpsw-qos.h" + +struct am65_cpts; #define HOST_PORT_NUM 0 @@ -35,8 +40,12 @@ struct am65_cpsw_port { u32 port_id; void __iomem *port_base; void __iomem *stat_base; + void __iomem *fetch_ram_base; bool disabled; struct am65_cpsw_slave_data slave; + bool tx_ts_enabled; + bool rx_ts_enabled; + struct am65_cpsw_qos qos; }; struct am65_cpsw_host { @@ -72,6 +81,7 @@ struct am65_cpsw_pdata { struct am65_cpsw_common { struct device *dev; + struct device *mdio_dev; const struct am65_cpsw_pdata *pdata; void __iomem *ss_base; @@ -96,8 +106,9 @@ struct am65_cpsw_common { u32 nuss_ver; u32 cpsw_ver; - bool pf_p0_rx_ptype_rrobin; + struct am65_cpts *cpts; + int est_enabled; }; struct am65_cpsw_ndev_stats { diff --git a/drivers/net/ethernet/ti/am65-cpsw-qos.c b/drivers/net/ethernet/ti/am65-cpsw-qos.c new file mode 100644 index 000000000000..32eac04468bb --- /dev/null +++ b/drivers/net/ethernet/ti/am65-cpsw-qos.c @@ -0,0 +1,626 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Texas Instruments K3 AM65 Ethernet QoS submodule + * Copyright (C) 2020 Texas Instruments Incorporated - http://www.ti.com/ + * + * quality of service module includes: + * Enhanced Scheduler Traffic (EST - P802.1Qbv/D2.2) + */ + +#include <linux/pm_runtime.h> +#include <linux/time.h> + +#include "am65-cpsw-nuss.h" +#include "am65-cpsw-qos.h" +#include "am65-cpts.h" + +#define AM65_CPSW_REG_CTL 0x004 +#define AM65_CPSW_PN_REG_CTL 0x004 +#define AM65_CPSW_PN_REG_FIFO_STATUS 0x050 +#define AM65_CPSW_PN_REG_EST_CTL 0x060 + +/* AM65_CPSW_REG_CTL register fields */ +#define AM65_CPSW_CTL_EST_EN BIT(18) + +/* AM65_CPSW_PN_REG_CTL register fields */ +#define AM65_CPSW_PN_CTL_EST_PORT_EN BIT(17) + +/* AM65_CPSW_PN_REG_EST_CTL register fields */ +#define AM65_CPSW_PN_EST_ONEBUF BIT(0) +#define AM65_CPSW_PN_EST_BUFSEL BIT(1) +#define AM65_CPSW_PN_EST_TS_EN BIT(2) +#define AM65_CPSW_PN_EST_TS_FIRST BIT(3) +#define AM65_CPSW_PN_EST_ONEPRI BIT(4) +#define AM65_CPSW_PN_EST_TS_PRI_MSK GENMASK(7, 5) + +/* AM65_CPSW_PN_REG_FIFO_STATUS register fields */ +#define AM65_CPSW_PN_FST_TX_PRI_ACTIVE_MSK GENMASK(7, 0) +#define AM65_CPSW_PN_FST_TX_E_MAC_ALLOW_MSK GENMASK(15, 8) +#define AM65_CPSW_PN_FST_EST_CNT_ERR BIT(16) +#define AM65_CPSW_PN_FST_EST_ADD_ERR BIT(17) +#define AM65_CPSW_PN_FST_EST_BUFACT BIT(18) + +/* EST FETCH COMMAND RAM */ +#define AM65_CPSW_FETCH_RAM_CMD_NUM 0x80 +#define AM65_CPSW_FETCH_CNT_MSK GENMASK(21, 8) +#define AM65_CPSW_FETCH_CNT_MAX (AM65_CPSW_FETCH_CNT_MSK >> 8) +#define AM65_CPSW_FETCH_CNT_OFFSET 8 +#define AM65_CPSW_FETCH_ALLOW_MSK GENMASK(7, 0) +#define AM65_CPSW_FETCH_ALLOW_MAX AM65_CPSW_FETCH_ALLOW_MSK + +enum timer_act { + TACT_PROG, /* need program timer */ + TACT_NEED_STOP, /* need stop first */ + TACT_SKIP_PROG, /* just buffer can be updated */ +}; + +static int am65_cpsw_port_est_enabled(struct am65_cpsw_port *port) +{ + return port->qos.est_oper || port->qos.est_admin; +} + +static void am65_cpsw_est_enable(struct am65_cpsw_common *common, int enable) +{ + u32 val; + + val = readl(common->cpsw_base + AM65_CPSW_REG_CTL); + + if (enable) + val |= AM65_CPSW_CTL_EST_EN; + else + val &= ~AM65_CPSW_CTL_EST_EN; + + writel(val, common->cpsw_base + AM65_CPSW_REG_CTL); + common->est_enabled = enable; +} + +static void am65_cpsw_port_est_enable(struct am65_cpsw_port *port, int enable) +{ + u32 val; + + val = readl(port->port_base + AM65_CPSW_PN_REG_CTL); + if (enable) + val |= AM65_CPSW_PN_CTL_EST_PORT_EN; + else + val &= ~AM65_CPSW_PN_CTL_EST_PORT_EN; + + writel(val, port->port_base + AM65_CPSW_PN_REG_CTL); +} + +/* target new EST RAM buffer, actual toggle happens after cycle completion */ +static void am65_cpsw_port_est_assign_buf_num(struct net_device *ndev, + int buf_num) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + u32 val; + + val = readl(port->port_base + AM65_CPSW_PN_REG_EST_CTL); + if (buf_num) + val |= AM65_CPSW_PN_EST_BUFSEL; + else + val &= ~AM65_CPSW_PN_EST_BUFSEL; + + writel(val, port->port_base + AM65_CPSW_PN_REG_EST_CTL); +} + +/* am65_cpsw_port_est_is_swapped() - Indicate if h/w is transitioned + * admin -> oper or not + * + * Return true if already transitioned. i.e oper is equal to admin and buf + * numbers match (est_oper->buf match with est_admin->buf). + * false if before transition. i.e oper is not equal to admin, (i.e a + * previous admin command is waiting to be transitioned to oper state + * and est_oper->buf not match with est_oper->buf). + */ +static int am65_cpsw_port_est_is_swapped(struct net_device *ndev, int *oper, + int *admin) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + u32 val; + + val = readl(port->port_base + AM65_CPSW_PN_REG_FIFO_STATUS); + *oper = !!(val & AM65_CPSW_PN_FST_EST_BUFACT); + + val = readl(port->port_base + AM65_CPSW_PN_REG_EST_CTL); + *admin = !!(val & AM65_CPSW_PN_EST_BUFSEL); + + return *admin == *oper; +} + +/* am65_cpsw_port_est_get_free_buf_num() - Get free buffer number for + * Admin to program the new schedule. + * + * Logic as follows:- + * If oper is same as admin, return the other buffer (!oper) as the admin + * buffer. If oper is not the same, driver let the current oper to continue + * as it is in the process of transitioning from admin -> oper. So keep the + * oper by selecting the same oper buffer by writing to EST_BUFSEL bit in + * EST CTL register. In the second iteration they will match and code returns. + * The actual buffer to write command is selected later before it is ready + * to update the schedule. + */ +static int am65_cpsw_port_est_get_free_buf_num(struct net_device *ndev) +{ + int oper, admin; + int roll = 2; + + while (roll--) { + if (am65_cpsw_port_est_is_swapped(ndev, &oper, &admin)) + return !oper; + + /* admin is not set, so hinder transition as it's not allowed + * to touch memory in-flight, by targeting same oper buf. + */ + am65_cpsw_port_est_assign_buf_num(ndev, oper); + + dev_info(&ndev->dev, + "Prev. EST admin cycle is in transit %d -> %d\n", + oper, admin); + } + + return admin; +} + +static void am65_cpsw_admin_to_oper(struct net_device *ndev) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + + if (port->qos.est_oper) + devm_kfree(&ndev->dev, port->qos.est_oper); + + port->qos.est_oper = port->qos.est_admin; + port->qos.est_admin = NULL; +} + +static void am65_cpsw_port_est_get_buf_num(struct net_device *ndev, + struct am65_cpsw_est *est_new) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + u32 val; + + val = readl(port->port_base + AM65_CPSW_PN_REG_EST_CTL); + val &= ~AM65_CPSW_PN_EST_ONEBUF; + writel(val, port->port_base + AM65_CPSW_PN_REG_EST_CTL); + + est_new->buf = am65_cpsw_port_est_get_free_buf_num(ndev); + + /* rolled buf num means changed buf while configuring */ + if (port->qos.est_oper && port->qos.est_admin && + est_new->buf == port->qos.est_oper->buf) + am65_cpsw_admin_to_oper(ndev); +} + +static void am65_cpsw_est_set(struct net_device *ndev, int enable) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + struct am65_cpsw_common *common = port->common; + int common_enable = 0; + int i; + + am65_cpsw_port_est_enable(port, enable); + + for (i = 0; i < common->port_num; i++) + common_enable |= am65_cpsw_port_est_enabled(&common->ports[i]); + + common_enable |= enable; + am65_cpsw_est_enable(common, common_enable); +} + +/* This update is supposed to be used in any routine before getting real state + * of admin -> oper transition, particularly it's supposed to be used in some + * generic routine for providing real state to Taprio Qdisc. + */ +static void am65_cpsw_est_update_state(struct net_device *ndev) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + int oper, admin; + + if (!port->qos.est_admin) + return; + + if (!am65_cpsw_port_est_is_swapped(ndev, &oper, &admin)) + return; + + am65_cpsw_admin_to_oper(ndev); +} + +/* Fetch command count it's number of bytes in Gigabit mode or nibbles in + * 10/100Mb mode. So, having speed and time in ns, recalculate ns to number of + * bytes/nibbles that can be sent while transmission on given speed. + */ +static int am65_est_cmd_ns_to_cnt(u64 ns, int link_speed) +{ + u64 temp; + + temp = ns * link_speed; + if (link_speed < SPEED_1000) + temp <<= 1; + + return DIV_ROUND_UP(temp, 8 * 1000); +} + +static void __iomem *am65_cpsw_est_set_sched_cmds(void __iomem *addr, + int fetch_cnt, + int fetch_allow) +{ + u32 prio_mask, cmd_fetch_cnt, cmd; + + do { + if (fetch_cnt > AM65_CPSW_FETCH_CNT_MAX) { + fetch_cnt -= AM65_CPSW_FETCH_CNT_MAX; + cmd_fetch_cnt = AM65_CPSW_FETCH_CNT_MAX; + } else { + cmd_fetch_cnt = fetch_cnt; + /* fetch count can't be less than 16? */ + if (cmd_fetch_cnt && cmd_fetch_cnt < 16) + cmd_fetch_cnt = 16; + + fetch_cnt = 0; + } + + prio_mask = fetch_allow & AM65_CPSW_FETCH_ALLOW_MSK; + cmd = (cmd_fetch_cnt << AM65_CPSW_FETCH_CNT_OFFSET) | prio_mask; + + writel(cmd, addr); + addr += 4; + } while (fetch_cnt); + + return addr; +} + +static int am65_cpsw_est_calc_cmd_num(struct net_device *ndev, + struct tc_taprio_qopt_offload *taprio, + int link_speed) +{ + int i, cmd_cnt, cmd_sum = 0; + u32 fetch_cnt; + + for (i = 0; i < taprio->num_entries; i++) { + if (taprio->entries[i].command != TC_TAPRIO_CMD_SET_GATES) { + dev_err(&ndev->dev, "Only SET command is supported"); + return -EINVAL; + } + + fetch_cnt = am65_est_cmd_ns_to_cnt(taprio->entries[i].interval, + link_speed); + + cmd_cnt = DIV_ROUND_UP(fetch_cnt, AM65_CPSW_FETCH_CNT_MAX); + if (!cmd_cnt) + cmd_cnt++; + + cmd_sum += cmd_cnt; + + if (!fetch_cnt) + break; + } + + return cmd_sum; +} + +static int am65_cpsw_est_check_scheds(struct net_device *ndev, + struct am65_cpsw_est *est_new) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + int cmd_num; + + cmd_num = am65_cpsw_est_calc_cmd_num(ndev, &est_new->taprio, + port->qos.link_speed); + if (cmd_num < 0) + return cmd_num; + + if (cmd_num > AM65_CPSW_FETCH_RAM_CMD_NUM / 2) { + dev_err(&ndev->dev, "No fetch RAM"); + return -ENOMEM; + } + + return 0; +} + +static void am65_cpsw_est_set_sched_list(struct net_device *ndev, + struct am65_cpsw_est *est_new) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + u32 fetch_cnt, fetch_allow, all_fetch_allow = 0; + void __iomem *ram_addr, *max_ram_addr; + struct tc_taprio_sched_entry *entry; + int i, ram_size; + + ram_addr = port->fetch_ram_base; + ram_size = AM65_CPSW_FETCH_RAM_CMD_NUM * 2; + ram_addr += est_new->buf * ram_size; + + max_ram_addr = ram_size + ram_addr; + for (i = 0; i < est_new->taprio.num_entries; i++) { + entry = &est_new->taprio.entries[i]; + + fetch_cnt = am65_est_cmd_ns_to_cnt(entry->interval, + port->qos.link_speed); + fetch_allow = entry->gate_mask; + if (fetch_allow > AM65_CPSW_FETCH_ALLOW_MAX) + dev_dbg(&ndev->dev, "fetch_allow > 8 bits: %d\n", + fetch_allow); + + ram_addr = am65_cpsw_est_set_sched_cmds(ram_addr, fetch_cnt, + fetch_allow); + + if (!fetch_cnt && i < est_new->taprio.num_entries - 1) { + dev_info(&ndev->dev, + "next scheds after %d have no impact", i + 1); + break; + } + + all_fetch_allow |= fetch_allow; + } + + /* end cmd, enabling non-timed queues for potential over cycle time */ + if (ram_addr < max_ram_addr) + writel(~all_fetch_allow & AM65_CPSW_FETCH_ALLOW_MSK, ram_addr); +} + +/** + * Enable ESTf periodic output, set cycle start time and interval. + */ +static int am65_cpsw_timer_set(struct net_device *ndev, + struct am65_cpsw_est *est_new) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + struct am65_cpsw_common *common = port->common; + struct am65_cpts *cpts = common->cpts; + struct am65_cpts_estf_cfg cfg; + + cfg.ns_period = est_new->taprio.cycle_time; + cfg.ns_start = est_new->taprio.base_time; + + return am65_cpts_estf_enable(cpts, port->port_id - 1, &cfg); +} + +static void am65_cpsw_timer_stop(struct net_device *ndev) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + struct am65_cpts *cpts = port->common->cpts; + + am65_cpts_estf_disable(cpts, port->port_id - 1); +} + +static enum timer_act am65_cpsw_timer_act(struct net_device *ndev, + struct am65_cpsw_est *est_new) +{ + struct tc_taprio_qopt_offload *taprio_oper, *taprio_new; + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + struct am65_cpts *cpts = port->common->cpts; + u64 cur_time; + s64 diff; + + if (!port->qos.est_oper) + return TACT_PROG; + + taprio_new = &est_new->taprio; + taprio_oper = &port->qos.est_oper->taprio; + + if (taprio_new->cycle_time != taprio_oper->cycle_time) + return TACT_NEED_STOP; + + /* in order to avoid timer reset get base_time form oper taprio */ + if (!taprio_new->base_time && taprio_oper) + taprio_new->base_time = taprio_oper->base_time; + + if (taprio_new->base_time == taprio_oper->base_time) + return TACT_SKIP_PROG; + + /* base times are cycle synchronized */ + diff = taprio_new->base_time - taprio_oper->base_time; + diff = diff < 0 ? -diff : diff; + if (diff % taprio_new->cycle_time) + return TACT_NEED_STOP; + + cur_time = am65_cpts_ns_gettime(cpts); + if (taprio_new->base_time <= cur_time + taprio_new->cycle_time) + return TACT_SKIP_PROG; + + /* TODO: Admin schedule at future time is not currently supported */ + return TACT_NEED_STOP; +} + +static void am65_cpsw_stop_est(struct net_device *ndev) +{ + am65_cpsw_est_set(ndev, 0); + am65_cpsw_timer_stop(ndev); +} + +static void am65_cpsw_purge_est(struct net_device *ndev) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + + am65_cpsw_stop_est(ndev); + + if (port->qos.est_admin) + devm_kfree(&ndev->dev, port->qos.est_admin); + + if (port->qos.est_oper) + devm_kfree(&ndev->dev, port->qos.est_oper); + + port->qos.est_oper = NULL; + port->qos.est_admin = NULL; +} + +static int am65_cpsw_configure_taprio(struct net_device *ndev, + struct am65_cpsw_est *est_new) +{ + struct am65_cpsw_common *common = am65_ndev_to_common(ndev); + struct am65_cpts *cpts = common->cpts; + int ret = 0, tact = TACT_PROG; + + am65_cpsw_est_update_state(ndev); + + if (!est_new->taprio.enable) { + am65_cpsw_stop_est(ndev); + return ret; + } + + ret = am65_cpsw_est_check_scheds(ndev, est_new); + if (ret < 0) + return ret; + + tact = am65_cpsw_timer_act(ndev, est_new); + if (tact == TACT_NEED_STOP) { + dev_err(&ndev->dev, + "Can't toggle estf timer, stop taprio first"); + return -EINVAL; + } + + if (tact == TACT_PROG) + am65_cpsw_timer_stop(ndev); + + if (!est_new->taprio.base_time) + est_new->taprio.base_time = am65_cpts_ns_gettime(cpts); + + am65_cpsw_port_est_get_buf_num(ndev, est_new); + am65_cpsw_est_set_sched_list(ndev, est_new); + am65_cpsw_port_est_assign_buf_num(ndev, est_new->buf); + + am65_cpsw_est_set(ndev, est_new->taprio.enable); + + if (tact == TACT_PROG) { + ret = am65_cpsw_timer_set(ndev, est_new); + if (ret) { + dev_err(&ndev->dev, "Failed to set cycle time"); + return ret; + } + } + + return ret; +} + +static void am65_cpsw_cp_taprio(struct tc_taprio_qopt_offload *from, + struct tc_taprio_qopt_offload *to) +{ + int i; + + *to = *from; + for (i = 0; i < from->num_entries; i++) + to->entries[i] = from->entries[i]; +} + +static int am65_cpsw_set_taprio(struct net_device *ndev, void *type_data) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + struct tc_taprio_qopt_offload *taprio = type_data; + struct am65_cpsw_est *est_new; + size_t size; + int ret = 0; + + if (taprio->cycle_time_extension) { + dev_err(&ndev->dev, "Failed to set cycle time extension"); + return -EOPNOTSUPP; + } + + size = sizeof(struct tc_taprio_sched_entry) * taprio->num_entries + + sizeof(struct am65_cpsw_est); + + est_new = devm_kzalloc(&ndev->dev, size, GFP_KERNEL); + if (!est_new) + return -ENOMEM; + + am65_cpsw_cp_taprio(taprio, &est_new->taprio); + ret = am65_cpsw_configure_taprio(ndev, est_new); + if (!ret) { + if (taprio->enable) { + if (port->qos.est_admin) + devm_kfree(&ndev->dev, port->qos.est_admin); + + port->qos.est_admin = est_new; + } else { + devm_kfree(&ndev->dev, est_new); + am65_cpsw_purge_est(ndev); + } + } else { + devm_kfree(&ndev->dev, est_new); + } + + return ret; +} + +static void am65_cpsw_est_link_up(struct net_device *ndev, int link_speed) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + ktime_t cur_time; + s64 delta; + + port->qos.link_speed = link_speed; + if (!am65_cpsw_port_est_enabled(port)) + return; + + if (port->qos.link_down_time) { + cur_time = ktime_get(); + delta = ktime_us_delta(cur_time, port->qos.link_down_time); + if (delta > USEC_PER_SEC) { + dev_err(&ndev->dev, + "Link has been lost too long, stopping TAS"); + goto purge_est; + } + } + + return; + +purge_est: + am65_cpsw_purge_est(ndev); +} + +static int am65_cpsw_setup_taprio(struct net_device *ndev, void *type_data) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + struct am65_cpsw_common *common = port->common; + + if (!IS_ENABLED(CONFIG_TI_AM65_CPSW_TAS)) + return -ENODEV; + + if (!netif_running(ndev)) { + dev_err(&ndev->dev, "interface is down, link speed unknown\n"); + return -ENETDOWN; + } + + if (common->pf_p0_rx_ptype_rrobin) { + dev_err(&ndev->dev, + "p0-rx-ptype-rrobin flag conflicts with taprio qdisc\n"); + return -EINVAL; + } + + if (port->qos.link_speed == SPEED_UNKNOWN) + return -ENOLINK; + + return am65_cpsw_set_taprio(ndev, type_data); +} + +int am65_cpsw_qos_ndo_setup_tc(struct net_device *ndev, enum tc_setup_type type, + void *type_data) +{ + switch (type) { + case TC_SETUP_QDISC_TAPRIO: + return am65_cpsw_setup_taprio(ndev, type_data); + default: + return -EOPNOTSUPP; + } +} + +void am65_cpsw_qos_link_up(struct net_device *ndev, int link_speed) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + + if (!IS_ENABLED(CONFIG_TI_AM65_CPSW_TAS)) + return; + + am65_cpsw_est_link_up(ndev, link_speed); + port->qos.link_down_time = 0; +} + +void am65_cpsw_qos_link_down(struct net_device *ndev) +{ + struct am65_cpsw_port *port = am65_ndev_to_port(ndev); + + if (!IS_ENABLED(CONFIG_TI_AM65_CPSW_TAS)) + return; + + if (!port->qos.link_down_time) + port->qos.link_down_time = ktime_get(); + + port->qos.link_speed = SPEED_UNKNOWN; +} diff --git a/drivers/net/ethernet/ti/am65-cpsw-qos.h b/drivers/net/ethernet/ti/am65-cpsw-qos.h new file mode 100644 index 000000000000..e8f1b6b59e93 --- /dev/null +++ b/drivers/net/ethernet/ti/am65-cpsw-qos.h @@ -0,0 +1,29 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Copyright (C) 2020 Texas Instruments Incorporated - http://www.ti.com/ + */ + +#ifndef AM65_CPSW_QOS_H_ +#define AM65_CPSW_QOS_H_ + +#include <linux/netdevice.h> +#include <net/pkt_sched.h> + +struct am65_cpsw_est { + int buf; + /* has to be the last one */ + struct tc_taprio_qopt_offload taprio; +}; + +struct am65_cpsw_qos { + struct am65_cpsw_est *est_admin; + struct am65_cpsw_est *est_oper; + ktime_t link_down_time; + int link_speed; +}; + +int am65_cpsw_qos_ndo_setup_tc(struct net_device *ndev, enum tc_setup_type type, + void *type_data); +void am65_cpsw_qos_link_up(struct net_device *ndev, int link_speed); +void am65_cpsw_qos_link_down(struct net_device *ndev); + +#endif /* AM65_CPSW_QOS_H_ */ diff --git a/drivers/net/ethernet/ti/am65-cpts.c b/drivers/net/ethernet/ti/am65-cpts.c new file mode 100644 index 000000000000..c59a289e428c --- /dev/null +++ b/drivers/net/ethernet/ti/am65-cpts.c @@ -0,0 +1,1086 @@ +// SPDX-License-Identifier: GPL-2.0 +/* TI K3 AM65x Common Platform Time Sync + * + * Copyright (C) 2020 Texas Instruments Incorporated - http://www.ti.com + * + */ + +#include <linux/clk.h> +#include <linux/clk-provider.h> +#include <linux/err.h> +#include <linux/if_vlan.h> +#include <linux/interrupt.h> +#include <linux/module.h> +#include <linux/netdevice.h> +#include <linux/net_tstamp.h> +#include <linux/of.h> +#include <linux/of_irq.h> +#include <linux/platform_device.h> +#include <linux/pm_runtime.h> +#include <linux/ptp_classify.h> +#include <linux/ptp_clock_kernel.h> + +#include "am65-cpts.h" + +struct am65_genf_regs { + u32 comp_lo; /* Comparison Low Value 0:31 */ + u32 comp_hi; /* Comparison High Value 32:63 */ + u32 control; /* control */ + u32 length; /* Length */ + u32 ppm_low; /* PPM Load Low Value 0:31 */ + u32 ppm_hi; /* PPM Load High Value 32:63 */ + u32 ts_nudge; /* Nudge value */ +} __aligned(32) __packed; + +#define AM65_CPTS_GENF_MAX_NUM 9 +#define AM65_CPTS_ESTF_MAX_NUM 8 + +struct am65_cpts_regs { + u32 idver; /* Identification and version */ + u32 control; /* Time sync control */ + u32 rftclk_sel; /* Reference Clock Select Register */ + u32 ts_push; /* Time stamp event push */ + u32 ts_load_val_lo; /* Time Stamp Load Low Value 0:31 */ + u32 ts_load_en; /* Time stamp load enable */ + u32 ts_comp_lo; /* Time Stamp Comparison Low Value 0:31 */ + u32 ts_comp_length; /* Time Stamp Comparison Length */ + u32 intstat_raw; /* Time sync interrupt status raw */ + u32 intstat_masked; /* Time sync interrupt status masked */ + u32 int_enable; /* Time sync interrupt enable */ + u32 ts_comp_nudge; /* Time Stamp Comparison Nudge Value */ + u32 event_pop; /* Event interrupt pop */ + u32 event_0; /* Event Time Stamp lo 0:31 */ + u32 event_1; /* Event Type Fields */ + u32 event_2; /* Event Type Fields domain */ + u32 event_3; /* Event Time Stamp hi 32:63 */ + u32 ts_load_val_hi; /* Time Stamp Load High Value 32:63 */ + u32 ts_comp_hi; /* Time Stamp Comparison High Value 32:63 */ + u32 ts_add_val; /* Time Stamp Add value */ + u32 ts_ppm_low; /* Time Stamp PPM Load Low Value 0:31 */ + u32 ts_ppm_hi; /* Time Stamp PPM Load High Value 32:63 */ + u32 ts_nudge; /* Time Stamp Nudge value */ + u32 reserv[33]; + struct am65_genf_regs genf[AM65_CPTS_GENF_MAX_NUM]; + struct am65_genf_regs estf[AM65_CPTS_ESTF_MAX_NUM]; +}; + +/* CONTROL_REG */ +#define AM65_CPTS_CONTROL_EN BIT(0) +#define AM65_CPTS_CONTROL_INT_TEST BIT(1) +#define AM65_CPTS_CONTROL_TS_COMP_POLARITY BIT(2) +#define AM65_CPTS_CONTROL_TSTAMP_EN BIT(3) +#define AM65_CPTS_CONTROL_SEQUENCE_EN BIT(4) +#define AM65_CPTS_CONTROL_64MODE BIT(5) +#define AM65_CPTS_CONTROL_TS_COMP_TOG BIT(6) +#define AM65_CPTS_CONTROL_TS_PPM_DIR BIT(7) +#define AM65_CPTS_CONTROL_HW1_TS_PUSH_EN BIT(8) +#define AM65_CPTS_CONTROL_HW2_TS_PUSH_EN BIT(9) +#define AM65_CPTS_CONTROL_HW3_TS_PUSH_EN BIT(10) +#define AM65_CPTS_CONTROL_HW4_TS_PUSH_EN BIT(11) +#define AM65_CPTS_CONTROL_HW5_TS_PUSH_EN BIT(12) +#define AM65_CPTS_CONTROL_HW6_TS_PUSH_EN BIT(13) +#define AM65_CPTS_CONTROL_HW7_TS_PUSH_EN BIT(14) +#define AM65_CPTS_CONTROL_HW8_TS_PUSH_EN BIT(15) +#define AM65_CPTS_CONTROL_HW1_TS_PUSH_OFFSET (8) + +#define AM65_CPTS_CONTROL_TS_SYNC_SEL_MASK (0xF) +#define AM65_CPTS_CONTROL_TS_SYNC_SEL_SHIFT (28) + +/* RFTCLK_SEL_REG */ +#define AM65_CPTS_RFTCLK_SEL_MASK (0x1F) + +/* TS_PUSH_REG */ +#define AM65_CPTS_TS_PUSH BIT(0) + +/* TS_LOAD_EN_REG */ +#define AM65_CPTS_TS_LOAD_EN BIT(0) + +/* INTSTAT_RAW_REG */ +#define AM65_CPTS_INTSTAT_RAW_TS_PEND BIT(0) + +/* INTSTAT_MASKED_REG */ +#define AM65_CPTS_INTSTAT_MASKED_TS_PEND BIT(0) + +/* INT_ENABLE_REG */ +#define AM65_CPTS_INT_ENABLE_TS_PEND_EN BIT(0) + +/* TS_COMP_NUDGE_REG */ +#define AM65_CPTS_TS_COMP_NUDGE_MASK (0xFF) + +/* EVENT_POP_REG */ +#define AM65_CPTS_EVENT_POP BIT(0) + +/* EVENT_1_REG */ +#define AM65_CPTS_EVENT_1_SEQUENCE_ID_MASK GENMASK(15, 0) + +#define AM65_CPTS_EVENT_1_MESSAGE_TYPE_MASK GENMASK(19, 16) +#define AM65_CPTS_EVENT_1_MESSAGE_TYPE_SHIFT (16) + +#define AM65_CPTS_EVENT_1_EVENT_TYPE_MASK GENMASK(23, 20) +#define AM65_CPTS_EVENT_1_EVENT_TYPE_SHIFT (20) + +#define AM65_CPTS_EVENT_1_PORT_NUMBER_MASK GENMASK(28, 24) +#define AM65_CPTS_EVENT_1_PORT_NUMBER_SHIFT (24) + +/* EVENT_2_REG */ +#define AM65_CPTS_EVENT_2_REG_DOMAIN_MASK (0xFF) +#define AM65_CPTS_EVENT_2_REG_DOMAIN_SHIFT (0) + +enum { + AM65_CPTS_EV_PUSH, /* Time Stamp Push Event */ + AM65_CPTS_EV_ROLL, /* Time Stamp Rollover Event */ + AM65_CPTS_EV_HALF, /* Time Stamp Half Rollover Event */ + AM65_CPTS_EV_HW, /* Hardware Time Stamp Push Event */ + AM65_CPTS_EV_RX, /* Ethernet Receive Event */ + AM65_CPTS_EV_TX, /* Ethernet Transmit Event */ + AM65_CPTS_EV_TS_COMP, /* Time Stamp Compare Event */ + AM65_CPTS_EV_HOST, /* Host Transmit Event */ +}; + +struct am65_cpts_event { + struct list_head list; + unsigned long tmo; + u32 event1; + u32 event2; + u64 timestamp; +}; + +#define AM65_CPTS_FIFO_DEPTH (16) +#define AM65_CPTS_MAX_EVENTS (32) +#define AM65_CPTS_EVENT_RX_TX_TIMEOUT (20) /* ms */ +#define AM65_CPTS_SKB_TX_WORK_TIMEOUT 1 /* jiffies */ +#define AM65_CPTS_MIN_PPM 0x400 + +struct am65_cpts { + struct device *dev; + struct am65_cpts_regs __iomem *reg; + struct ptp_clock_info ptp_info; + struct ptp_clock *ptp_clock; + int phc_index; + struct clk_hw *clk_mux_hw; + struct device_node *clk_mux_np; + struct clk *refclk; + u32 refclk_freq; + struct list_head events; + struct list_head pool; + struct am65_cpts_event pool_data[AM65_CPTS_MAX_EVENTS]; + spinlock_t lock; /* protects events lists*/ + u32 ext_ts_inputs; + u32 genf_num; + u32 ts_add_val; + int irq; + struct mutex ptp_clk_lock; /* PHC access sync */ + u64 timestamp; + u32 genf_enable; + u32 hw_ts_enable; + struct sk_buff_head txq; +}; + +struct am65_cpts_skb_cb_data { + unsigned long tmo; + u32 skb_mtype_seqid; +}; + +#define am65_cpts_write32(c, v, r) writel(v, &(c)->reg->r) +#define am65_cpts_read32(c, r) readl(&(c)->reg->r) + +static void am65_cpts_settime(struct am65_cpts *cpts, u64 start_tstamp) +{ + u32 val; + + val = upper_32_bits(start_tstamp); + am65_cpts_write32(cpts, val, ts_load_val_hi); + val = lower_32_bits(start_tstamp); + am65_cpts_write32(cpts, val, ts_load_val_lo); + + am65_cpts_write32(cpts, AM65_CPTS_TS_LOAD_EN, ts_load_en); +} + +static void am65_cpts_set_add_val(struct am65_cpts *cpts) +{ + /* select coefficient according to the rate */ + cpts->ts_add_val = (NSEC_PER_SEC / cpts->refclk_freq - 1) & 0x7; + + am65_cpts_write32(cpts, cpts->ts_add_val, ts_add_val); +} + +static void am65_cpts_disable(struct am65_cpts *cpts) +{ + am65_cpts_write32(cpts, 0, control); + am65_cpts_write32(cpts, 0, int_enable); +} + +static int am65_cpts_event_get_port(struct am65_cpts_event *event) +{ + return (event->event1 & AM65_CPTS_EVENT_1_PORT_NUMBER_MASK) >> + AM65_CPTS_EVENT_1_PORT_NUMBER_SHIFT; +} + +static int am65_cpts_event_get_type(struct am65_cpts_event *event) +{ + return (event->event1 & AM65_CPTS_EVENT_1_EVENT_TYPE_MASK) >> + AM65_CPTS_EVENT_1_EVENT_TYPE_SHIFT; +} + +static int am65_cpts_cpts_purge_events(struct am65_cpts *cpts) +{ + struct list_head *this, *next; + struct am65_cpts_event *event; + int removed = 0; + + list_for_each_safe(this, next, &cpts->events) { + event = list_entry(this, struct am65_cpts_event, list); + if (time_after(jiffies, event->tmo)) { + list_del_init(&event->list); + list_add(&event->list, &cpts->pool); + ++removed; + } + } + + if (removed) + dev_dbg(cpts->dev, "event pool cleaned up %d\n", removed); + return removed ? 0 : -1; +} + +static bool am65_cpts_fifo_pop_event(struct am65_cpts *cpts, + struct am65_cpts_event *event) +{ + u32 r = am65_cpts_read32(cpts, intstat_raw); + + if (r & AM65_CPTS_INTSTAT_RAW_TS_PEND) { + event->timestamp = am65_cpts_read32(cpts, event_0); + event->event1 = am65_cpts_read32(cpts, event_1); + event->event2 = am65_cpts_read32(cpts, event_2); + event->timestamp |= (u64)am65_cpts_read32(cpts, event_3) << 32; + am65_cpts_write32(cpts, AM65_CPTS_EVENT_POP, event_pop); + return false; + } + return true; +} + +static int am65_cpts_fifo_read(struct am65_cpts *cpts) +{ + struct ptp_clock_event pevent; + struct am65_cpts_event *event; + bool schedule = false; + int i, type, ret = 0; + unsigned long flags; + + spin_lock_irqsave(&cpts->lock, flags); + for (i = 0; i < AM65_CPTS_FIFO_DEPTH; i++) { + event = list_first_entry_or_null(&cpts->pool, + struct am65_cpts_event, list); + + if (!event) { + if (am65_cpts_cpts_purge_events(cpts)) { + dev_err(cpts->dev, "cpts: event pool empty\n"); + ret = -1; + goto out; + } + continue; + } + + if (am65_cpts_fifo_pop_event(cpts, event)) + break; + + type = am65_cpts_event_get_type(event); + switch (type) { + case AM65_CPTS_EV_PUSH: + cpts->timestamp = event->timestamp; + dev_dbg(cpts->dev, "AM65_CPTS_EV_PUSH t:%llu\n", + cpts->timestamp); + break; + case AM65_CPTS_EV_RX: + case AM65_CPTS_EV_TX: + event->tmo = jiffies + + msecs_to_jiffies(AM65_CPTS_EVENT_RX_TX_TIMEOUT); + + list_del_init(&event->list); + list_add_tail(&event->list, &cpts->events); + + dev_dbg(cpts->dev, + "AM65_CPTS_EV_TX e1:%08x e2:%08x t:%lld\n", + event->event1, event->event2, + event->timestamp); + schedule = true; + break; + case AM65_CPTS_EV_HW: + pevent.index = am65_cpts_event_get_port(event) - 1; + pevent.timestamp = event->timestamp; + pevent.type = PTP_CLOCK_EXTTS; + dev_dbg(cpts->dev, "AM65_CPTS_EV_HW p:%d t:%llu\n", + pevent.index, event->timestamp); + + ptp_clock_event(cpts->ptp_clock, &pevent); + break; + case AM65_CPTS_EV_HOST: + break; + case AM65_CPTS_EV_ROLL: + case AM65_CPTS_EV_HALF: + case AM65_CPTS_EV_TS_COMP: + dev_dbg(cpts->dev, + "AM65_CPTS_EVT: %d e1:%08x e2:%08x t:%lld\n", + type, + event->event1, event->event2, + event->timestamp); + break; + default: + dev_err(cpts->dev, "cpts: unknown event type\n"); + ret = -1; + goto out; + } + } + +out: + spin_unlock_irqrestore(&cpts->lock, flags); + + if (schedule) + ptp_schedule_worker(cpts->ptp_clock, 0); + + return ret; +} + +static u64 am65_cpts_gettime(struct am65_cpts *cpts, + struct ptp_system_timestamp *sts) +{ + unsigned long flags; + u64 val = 0; + + /* temporarily disable cpts interrupt to avoid intentional + * doubled read. Interrupt can be in-flight - it's Ok. + */ + am65_cpts_write32(cpts, 0, int_enable); + + /* use spin_lock_irqsave() here as it has to run very fast */ + spin_lock_irqsave(&cpts->lock, flags); + ptp_read_system_prets(sts); + am65_cpts_write32(cpts, AM65_CPTS_TS_PUSH, ts_push); + am65_cpts_read32(cpts, ts_push); + ptp_read_system_postts(sts); + spin_unlock_irqrestore(&cpts->lock, flags); + + am65_cpts_fifo_read(cpts); + + am65_cpts_write32(cpts, AM65_CPTS_INT_ENABLE_TS_PEND_EN, int_enable); + + val = cpts->timestamp; + + return val; +} + +static irqreturn_t am65_cpts_interrupt(int irq, void *dev_id) +{ + struct am65_cpts *cpts = dev_id; + + if (am65_cpts_fifo_read(cpts)) + dev_dbg(cpts->dev, "cpts: unable to obtain a time stamp\n"); + + return IRQ_HANDLED; +} + +/* PTP clock operations */ +static int am65_cpts_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb) +{ + struct am65_cpts *cpts = container_of(ptp, struct am65_cpts, ptp_info); + int neg_adj = 0; + u64 adj_period; + u32 val; + + if (ppb < 0) { + neg_adj = 1; + ppb = -ppb; + } + + /* base freq = 1GHz = 1 000 000 000 + * ppb_norm = ppb * base_freq / clock_freq; + * ppm_norm = ppb_norm / 1000 + * adj_period = 1 000 000 / ppm_norm + * adj_period = 1 000 000 000 / ppb_norm + * adj_period = 1 000 000 000 / (ppb * base_freq / clock_freq) + * adj_period = (1 000 000 000 * clock_freq) / (ppb * base_freq) + * adj_period = clock_freq / ppb + */ + adj_period = div_u64(cpts->refclk_freq, ppb); + + mutex_lock(&cpts->ptp_clk_lock); + + val = am65_cpts_read32(cpts, control); + if (neg_adj) + val |= AM65_CPTS_CONTROL_TS_PPM_DIR; + else + val &= ~AM65_CPTS_CONTROL_TS_PPM_DIR; + am65_cpts_write32(cpts, val, control); + + val = upper_32_bits(adj_period) & 0x3FF; + am65_cpts_write32(cpts, val, ts_ppm_hi); + val = lower_32_bits(adj_period); + am65_cpts_write32(cpts, val, ts_ppm_low); + + mutex_unlock(&cpts->ptp_clk_lock); + + return 0; +} + +static int am65_cpts_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta) +{ + struct am65_cpts *cpts = container_of(ptp, struct am65_cpts, ptp_info); + s64 ns; + + mutex_lock(&cpts->ptp_clk_lock); + ns = am65_cpts_gettime(cpts, NULL); + ns += delta; + am65_cpts_settime(cpts, ns); + mutex_unlock(&cpts->ptp_clk_lock); + + return 0; +} + +static int am65_cpts_ptp_gettimex(struct ptp_clock_info *ptp, + struct timespec64 *ts, + struct ptp_system_timestamp *sts) +{ + struct am65_cpts *cpts = container_of(ptp, struct am65_cpts, ptp_info); + u64 ns; + + mutex_lock(&cpts->ptp_clk_lock); + ns = am65_cpts_gettime(cpts, sts); + mutex_unlock(&cpts->ptp_clk_lock); + *ts = ns_to_timespec64(ns); + + return 0; +} + +u64 am65_cpts_ns_gettime(struct am65_cpts *cpts) +{ + u64 ns; + + /* reuse ptp_clk_lock as it serialize ts push */ + mutex_lock(&cpts->ptp_clk_lock); + ns = am65_cpts_gettime(cpts, NULL); + mutex_unlock(&cpts->ptp_clk_lock); + + return ns; +} +EXPORT_SYMBOL_GPL(am65_cpts_ns_gettime); + +static int am65_cpts_ptp_settime(struct ptp_clock_info *ptp, + const struct timespec64 *ts) +{ + struct am65_cpts *cpts = container_of(ptp, struct am65_cpts, ptp_info); + u64 ns; + + ns = timespec64_to_ns(ts); + mutex_lock(&cpts->ptp_clk_lock); + am65_cpts_settime(cpts, ns); + mutex_unlock(&cpts->ptp_clk_lock); + + return 0; +} + +static void am65_cpts_extts_enable_hw(struct am65_cpts *cpts, u32 index, int on) +{ + u32 v; + + v = am65_cpts_read32(cpts, control); + if (on) { + v |= BIT(AM65_CPTS_CONTROL_HW1_TS_PUSH_OFFSET + index); + cpts->hw_ts_enable |= BIT(index); + } else { + v &= ~BIT(AM65_CPTS_CONTROL_HW1_TS_PUSH_OFFSET + index); + cpts->hw_ts_enable &= ~BIT(index); + } + am65_cpts_write32(cpts, v, control); +} + +static int am65_cpts_extts_enable(struct am65_cpts *cpts, u32 index, int on) +{ + if (!!(cpts->hw_ts_enable & BIT(index)) == !!on) + return 0; + + mutex_lock(&cpts->ptp_clk_lock); + am65_cpts_extts_enable_hw(cpts, index, on); + mutex_unlock(&cpts->ptp_clk_lock); + + dev_dbg(cpts->dev, "%s: ExtTS:%u %s\n", + __func__, index, on ? "enabled" : "disabled"); + + return 0; +} + +int am65_cpts_estf_enable(struct am65_cpts *cpts, int idx, + struct am65_cpts_estf_cfg *cfg) +{ + u64 cycles; + u32 val; + + cycles = cfg->ns_period * cpts->refclk_freq; + cycles = DIV_ROUND_UP(cycles, NSEC_PER_SEC); + if (cycles > U32_MAX) + return -EINVAL; + + /* according to TRM should be zeroed */ + am65_cpts_write32(cpts, 0, estf[idx].length); + + val = upper_32_bits(cfg->ns_start); + am65_cpts_write32(cpts, val, estf[idx].comp_hi); + val = lower_32_bits(cfg->ns_start); + am65_cpts_write32(cpts, val, estf[idx].comp_lo); + val = lower_32_bits(cycles); + am65_cpts_write32(cpts, val, estf[idx].length); + + dev_dbg(cpts->dev, "%s: ESTF:%u enabled\n", __func__, idx); + + return 0; +} +EXPORT_SYMBOL_GPL(am65_cpts_estf_enable); + +void am65_cpts_estf_disable(struct am65_cpts *cpts, int idx) +{ + am65_cpts_write32(cpts, 0, estf[idx].length); + + dev_dbg(cpts->dev, "%s: ESTF:%u disabled\n", __func__, idx); +} +EXPORT_SYMBOL_GPL(am65_cpts_estf_disable); + +static void am65_cpts_perout_enable_hw(struct am65_cpts *cpts, + struct ptp_perout_request *req, int on) +{ + u64 ns_period, ns_start, cycles; + struct timespec64 ts; + u32 val; + + if (on) { + ts.tv_sec = req->period.sec; + ts.tv_nsec = req->period.nsec; + ns_period = timespec64_to_ns(&ts); + + cycles = (ns_period * cpts->refclk_freq) / NSEC_PER_SEC; + + ts.tv_sec = req->start.sec; + ts.tv_nsec = req->start.nsec; + ns_start = timespec64_to_ns(&ts); + + val = upper_32_bits(ns_start); + am65_cpts_write32(cpts, val, genf[req->index].comp_hi); + val = lower_32_bits(ns_start); + am65_cpts_write32(cpts, val, genf[req->index].comp_lo); + val = lower_32_bits(cycles); + am65_cpts_write32(cpts, val, genf[req->index].length); + + cpts->genf_enable |= BIT(req->index); + } else { + am65_cpts_write32(cpts, 0, genf[req->index].length); + + cpts->genf_enable &= ~BIT(req->index); + } +} + +static int am65_cpts_perout_enable(struct am65_cpts *cpts, + struct ptp_perout_request *req, int on) +{ + if (!!(cpts->genf_enable & BIT(req->index)) == !!on) + return 0; + + mutex_lock(&cpts->ptp_clk_lock); + am65_cpts_perout_enable_hw(cpts, req, on); + mutex_unlock(&cpts->ptp_clk_lock); + + dev_dbg(cpts->dev, "%s: GenF:%u %s\n", + __func__, req->index, on ? "enabled" : "disabled"); + + return 0; +} + +static int am65_cpts_ptp_enable(struct ptp_clock_info *ptp, + struct ptp_clock_request *rq, int on) +{ + struct am65_cpts *cpts = container_of(ptp, struct am65_cpts, ptp_info); + + switch (rq->type) { + case PTP_CLK_REQ_EXTTS: + return am65_cpts_extts_enable(cpts, rq->extts.index, on); + case PTP_CLK_REQ_PEROUT: + return am65_cpts_perout_enable(cpts, &rq->perout, on); + default: + break; + } + + return -EOPNOTSUPP; +} + +static long am65_cpts_ts_work(struct ptp_clock_info *ptp); + +static struct ptp_clock_info am65_ptp_info = { + .owner = THIS_MODULE, + .name = "CTPS timer", + .adjfreq = am65_cpts_ptp_adjfreq, + .adjtime = am65_cpts_ptp_adjtime, + .gettimex64 = am65_cpts_ptp_gettimex, + .settime64 = am65_cpts_ptp_settime, + .enable = am65_cpts_ptp_enable, + .do_aux_work = am65_cpts_ts_work, +}; + +static bool am65_cpts_match_tx_ts(struct am65_cpts *cpts, + struct am65_cpts_event *event) +{ + struct sk_buff_head txq_list; + struct sk_buff *skb, *tmp; + unsigned long flags; + bool found = false; + u32 mtype_seqid; + + mtype_seqid = event->event1 & + (AM65_CPTS_EVENT_1_MESSAGE_TYPE_MASK | + AM65_CPTS_EVENT_1_EVENT_TYPE_MASK | + AM65_CPTS_EVENT_1_SEQUENCE_ID_MASK); + + __skb_queue_head_init(&txq_list); + + spin_lock_irqsave(&cpts->txq.lock, flags); + skb_queue_splice_init(&cpts->txq, &txq_list); + spin_unlock_irqrestore(&cpts->txq.lock, flags); + + /* no need to grab txq.lock as access is always done under cpts->lock */ + skb_queue_walk_safe(&txq_list, skb, tmp) { + struct skb_shared_hwtstamps ssh; + struct am65_cpts_skb_cb_data *skb_cb = + (struct am65_cpts_skb_cb_data *)skb->cb; + + if (mtype_seqid == skb_cb->skb_mtype_seqid) { + u64 ns = event->timestamp; + + memset(&ssh, 0, sizeof(ssh)); + ssh.hwtstamp = ns_to_ktime(ns); + skb_tstamp_tx(skb, &ssh); + found = true; + __skb_unlink(skb, &txq_list); + dev_consume_skb_any(skb); + dev_dbg(cpts->dev, + "match tx timestamp mtype_seqid %08x\n", + mtype_seqid); + break; + } + + if (time_after(jiffies, skb_cb->tmo)) { + /* timeout any expired skbs over 100 ms */ + dev_dbg(cpts->dev, + "expiring tx timestamp mtype_seqid %08x\n", + mtype_seqid); + __skb_unlink(skb, &txq_list); + dev_consume_skb_any(skb); + } + } + + spin_lock_irqsave(&cpts->txq.lock, flags); + skb_queue_splice(&txq_list, &cpts->txq); + spin_unlock_irqrestore(&cpts->txq.lock, flags); + + return found; +} + +static void am65_cpts_find_ts(struct am65_cpts *cpts) +{ + struct am65_cpts_event *event; + struct list_head *this, *next; + LIST_HEAD(events_free); + unsigned long flags; + LIST_HEAD(events); + + spin_lock_irqsave(&cpts->lock, flags); + list_splice_init(&cpts->events, &events); + spin_unlock_irqrestore(&cpts->lock, flags); + + list_for_each_safe(this, next, &events) { + event = list_entry(this, struct am65_cpts_event, list); + if (am65_cpts_match_tx_ts(cpts, event) || + time_after(jiffies, event->tmo)) { + list_del_init(&event->list); + list_add(&event->list, &events_free); + } + } + + spin_lock_irqsave(&cpts->lock, flags); + list_splice_tail(&events, &cpts->events); + list_splice_tail(&events_free, &cpts->pool); + spin_unlock_irqrestore(&cpts->lock, flags); +} + +static long am65_cpts_ts_work(struct ptp_clock_info *ptp) +{ + struct am65_cpts *cpts = container_of(ptp, struct am65_cpts, ptp_info); + unsigned long flags; + long delay = -1; + + am65_cpts_find_ts(cpts); + + spin_lock_irqsave(&cpts->txq.lock, flags); + if (!skb_queue_empty(&cpts->txq)) + delay = AM65_CPTS_SKB_TX_WORK_TIMEOUT; + spin_unlock_irqrestore(&cpts->txq.lock, flags); + + return delay; +} + +/** + * am65_cpts_rx_enable - enable rx timestamping + * @cpts: cpts handle + * @skb: packet + * + * This functions enables rx packets timestamping. The CPTS can timestamp all + * rx packets. + */ +void am65_cpts_rx_enable(struct am65_cpts *cpts, bool en) +{ + u32 val; + + mutex_lock(&cpts->ptp_clk_lock); + val = am65_cpts_read32(cpts, control); + if (en) + val |= AM65_CPTS_CONTROL_TSTAMP_EN; + else + val &= ~AM65_CPTS_CONTROL_TSTAMP_EN; + am65_cpts_write32(cpts, val, control); + mutex_unlock(&cpts->ptp_clk_lock); +} +EXPORT_SYMBOL_GPL(am65_cpts_rx_enable); + +static int am65_skb_get_mtype_seqid(struct sk_buff *skb, u32 *mtype_seqid) +{ + unsigned int ptp_class = ptp_classify_raw(skb); + u8 *msgtype, *data = skb->data; + unsigned int offset = 0; + __be16 *seqid; + + if (ptp_class == PTP_CLASS_NONE) + return 0; + + if (ptp_class & PTP_CLASS_VLAN) + offset += VLAN_HLEN; + + switch (ptp_class & PTP_CLASS_PMASK) { + case PTP_CLASS_IPV4: + offset += ETH_HLEN + IPV4_HLEN(data + offset) + UDP_HLEN; + break; + case PTP_CLASS_IPV6: + offset += ETH_HLEN + IP6_HLEN + UDP_HLEN; + break; + case PTP_CLASS_L2: + offset += ETH_HLEN; + break; + default: + return 0; + } + + if (skb->len + ETH_HLEN < offset + OFF_PTP_SEQUENCE_ID + sizeof(*seqid)) + return 0; + + if (unlikely(ptp_class & PTP_CLASS_V1)) + msgtype = data + offset + OFF_PTP_CONTROL; + else + msgtype = data + offset; + + seqid = (__be16 *)(data + offset + OFF_PTP_SEQUENCE_ID); + *mtype_seqid = (*msgtype << AM65_CPTS_EVENT_1_MESSAGE_TYPE_SHIFT) & + AM65_CPTS_EVENT_1_MESSAGE_TYPE_MASK; + *mtype_seqid |= (ntohs(*seqid) & AM65_CPTS_EVENT_1_SEQUENCE_ID_MASK); + + return 1; +} + +/** + * am65_cpts_tx_timestamp - save tx packet for timestamping + * @cpts: cpts handle + * @skb: packet + * + * This functions saves tx packet for timestamping if packet can be timestamped. + * The future processing is done in from PTP auxiliary worker. + */ +void am65_cpts_tx_timestamp(struct am65_cpts *cpts, struct sk_buff *skb) +{ + struct am65_cpts_skb_cb_data *skb_cb = (void *)skb->cb; + + if (!(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) + return; + + /* add frame to queue for processing later. + * The periodic FIFO check will handle this. + */ + skb_get(skb); + /* get the timestamp for timeouts */ + skb_cb->tmo = jiffies + msecs_to_jiffies(100); + skb_queue_tail(&cpts->txq, skb); + ptp_schedule_worker(cpts->ptp_clock, 0); +} +EXPORT_SYMBOL_GPL(am65_cpts_tx_timestamp); + +/** + * am65_cpts_prep_tx_timestamp - check and prepare tx packet for timestamping + * @cpts: cpts handle + * @skb: packet + * + * This functions should be called from .xmit(). + * It checks if packet can be timestamped, fills internal cpts data + * in skb-cb and marks packet as SKBTX_IN_PROGRESS. + */ +void am65_cpts_prep_tx_timestamp(struct am65_cpts *cpts, struct sk_buff *skb) +{ + struct am65_cpts_skb_cb_data *skb_cb = (void *)skb->cb; + int ret; + + if (!(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP)) + return; + + ret = am65_skb_get_mtype_seqid(skb, &skb_cb->skb_mtype_seqid); + if (!ret) + return; + skb_cb->skb_mtype_seqid |= (AM65_CPTS_EV_TX << + AM65_CPTS_EVENT_1_EVENT_TYPE_SHIFT); + + skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; +} +EXPORT_SYMBOL_GPL(am65_cpts_prep_tx_timestamp); + +int am65_cpts_phc_index(struct am65_cpts *cpts) +{ + return cpts->phc_index; +} +EXPORT_SYMBOL_GPL(am65_cpts_phc_index); + +static void cpts_free_clk_mux(void *data) +{ + struct am65_cpts *cpts = data; + + of_clk_del_provider(cpts->clk_mux_np); + clk_hw_unregister_mux(cpts->clk_mux_hw); + of_node_put(cpts->clk_mux_np); +} + +static int cpts_of_mux_clk_setup(struct am65_cpts *cpts, + struct device_node *node) +{ + unsigned int num_parents; + const char **parent_names; + char *clk_mux_name; + void __iomem *reg; + int ret = -EINVAL; + + cpts->clk_mux_np = of_get_child_by_name(node, "refclk-mux"); + if (!cpts->clk_mux_np) + return 0; + + num_parents = of_clk_get_parent_count(cpts->clk_mux_np); + if (num_parents < 1) { + dev_err(cpts->dev, "mux-clock %pOF must have parents\n", + cpts->clk_mux_np); + goto mux_fail; + } + + parent_names = devm_kcalloc(cpts->dev, sizeof(char *), num_parents, + GFP_KERNEL); + if (!parent_names) { + ret = -ENOMEM; + goto mux_fail; + } + + of_clk_parent_fill(cpts->clk_mux_np, parent_names, num_parents); + + clk_mux_name = devm_kasprintf(cpts->dev, GFP_KERNEL, "%s.%pOFn", + dev_name(cpts->dev), cpts->clk_mux_np); + if (!clk_mux_name) { + ret = -ENOMEM; + goto mux_fail; + } + + reg = &cpts->reg->rftclk_sel; + /* dev must be NULL to avoid recursive incrementing + * of module refcnt + */ + cpts->clk_mux_hw = clk_hw_register_mux(NULL, clk_mux_name, + parent_names, num_parents, + 0, reg, 0, 5, 0, NULL); + if (IS_ERR(cpts->clk_mux_hw)) { + ret = PTR_ERR(cpts->clk_mux_hw); + goto mux_fail; + } + + ret = of_clk_add_hw_provider(cpts->clk_mux_np, of_clk_hw_simple_get, + cpts->clk_mux_hw); + if (ret) + goto clk_hw_register; + + ret = devm_add_action_or_reset(cpts->dev, cpts_free_clk_mux, cpts); + if (ret) + dev_err(cpts->dev, "failed to add clkmux reset action %d", ret); + + return ret; + +clk_hw_register: + clk_hw_unregister_mux(cpts->clk_mux_hw); +mux_fail: + of_node_put(cpts->clk_mux_np); + return ret; +} + +static int am65_cpts_of_parse(struct am65_cpts *cpts, struct device_node *node) +{ + u32 prop[2]; + + if (!of_property_read_u32(node, "ti,cpts-ext-ts-inputs", &prop[0])) + cpts->ext_ts_inputs = prop[0]; + + if (!of_property_read_u32(node, "ti,cpts-periodic-outputs", &prop[0])) + cpts->genf_num = prop[0]; + + return cpts_of_mux_clk_setup(cpts, node); +} + +static void am65_cpts_release(void *data) +{ + struct am65_cpts *cpts = data; + + ptp_clock_unregister(cpts->ptp_clock); + am65_cpts_disable(cpts); + clk_disable_unprepare(cpts->refclk); +} + +struct am65_cpts *am65_cpts_create(struct device *dev, void __iomem *regs, + struct device_node *node) +{ + struct am65_cpts *cpts; + int ret, i; + + cpts = devm_kzalloc(dev, sizeof(*cpts), GFP_KERNEL); + if (!cpts) + return ERR_PTR(-ENOMEM); + + cpts->dev = dev; + cpts->reg = (struct am65_cpts_regs __iomem *)regs; + + cpts->irq = of_irq_get_byname(node, "cpts"); + if (cpts->irq <= 0) { + ret = cpts->irq ?: -ENXIO; + if (ret != -EPROBE_DEFER) + dev_err(dev, "Failed to get IRQ number (err = %d)\n", + ret); + return ERR_PTR(ret); + } + + ret = am65_cpts_of_parse(cpts, node); + if (ret) + return ERR_PTR(ret); + + mutex_init(&cpts->ptp_clk_lock); + INIT_LIST_HEAD(&cpts->events); + INIT_LIST_HEAD(&cpts->pool); + spin_lock_init(&cpts->lock); + skb_queue_head_init(&cpts->txq); + + for (i = 0; i < AM65_CPTS_MAX_EVENTS; i++) + list_add(&cpts->pool_data[i].list, &cpts->pool); + + cpts->refclk = devm_get_clk_from_child(dev, node, "cpts"); + if (IS_ERR(cpts->refclk)) { + ret = PTR_ERR(cpts->refclk); + if (ret != -EPROBE_DEFER) + dev_err(dev, "Failed to get refclk %d\n", ret); + return ERR_PTR(ret); + } + + ret = clk_prepare_enable(cpts->refclk); + if (ret) { + dev_err(dev, "Failed to enable refclk %d\n", ret); + return ERR_PTR(ret); + } + + cpts->refclk_freq = clk_get_rate(cpts->refclk); + + am65_ptp_info.max_adj = cpts->refclk_freq / AM65_CPTS_MIN_PPM; + cpts->ptp_info = am65_ptp_info; + + if (cpts->ext_ts_inputs) + cpts->ptp_info.n_ext_ts = cpts->ext_ts_inputs; + if (cpts->genf_num) + cpts->ptp_info.n_per_out = cpts->genf_num; + + am65_cpts_set_add_val(cpts); + + am65_cpts_write32(cpts, AM65_CPTS_CONTROL_EN | AM65_CPTS_CONTROL_64MODE, + control); + am65_cpts_write32(cpts, AM65_CPTS_INT_ENABLE_TS_PEND_EN, int_enable); + + /* set time to the current system time */ + am65_cpts_settime(cpts, ktime_to_ns(ktime_get_real())); + + cpts->ptp_clock = ptp_clock_register(&cpts->ptp_info, cpts->dev); + if (IS_ERR_OR_NULL(cpts->ptp_clock)) { + dev_err(dev, "Failed to register ptp clk %ld\n", + PTR_ERR(cpts->ptp_clock)); + if (!cpts->ptp_clock) + ret = -ENODEV; + goto refclk_disable; + } + cpts->phc_index = ptp_clock_index(cpts->ptp_clock); + + ret = devm_add_action_or_reset(dev, am65_cpts_release, cpts); + if (ret) { + dev_err(dev, "failed to add ptpclk reset action %d", ret); + return ERR_PTR(ret); + } + + ret = devm_request_threaded_irq(dev, cpts->irq, NULL, + am65_cpts_interrupt, + IRQF_ONESHOT, dev_name(dev), cpts); + if (ret < 0) { + dev_err(cpts->dev, "error attaching irq %d\n", ret); + return ERR_PTR(ret); + } + + dev_info(dev, "CPTS ver 0x%08x, freq:%u, add_val:%u\n", + am65_cpts_read32(cpts, idver), + cpts->refclk_freq, cpts->ts_add_val); + + return cpts; + +refclk_disable: + clk_disable_unprepare(cpts->refclk); + return ERR_PTR(ret); +} +EXPORT_SYMBOL_GPL(am65_cpts_create); + +static int am65_cpts_probe(struct platform_device *pdev) +{ + struct device_node *node = pdev->dev.of_node; + struct device *dev = &pdev->dev; + struct am65_cpts *cpts; + struct resource *res; + void __iomem *base; + + res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "cpts"); + base = devm_ioremap_resource(dev, res); + if (IS_ERR(base)) + return PTR_ERR(base); + + cpts = am65_cpts_create(dev, base, node); + return PTR_ERR_OR_ZERO(cpts); +} + +static const struct of_device_id am65_cpts_of_match[] = { + { .compatible = "ti,am65-cpts", }, + { .compatible = "ti,j721e-cpts", }, + {}, +}; +MODULE_DEVICE_TABLE(of, am65_cpts_of_match); + +static struct platform_driver am65_cpts_driver = { + .probe = am65_cpts_probe, + .driver = { + .name = "am65-cpts", + .of_match_table = am65_cpts_of_match, + }, +}; +module_platform_driver(am65_cpts_driver); + +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Grygorii Strashko <grygorii.strashko@ti.com>"); +MODULE_DESCRIPTION("TI K3 AM65 CPTS driver"); diff --git a/drivers/net/ethernet/ti/am65-cpts.h b/drivers/net/ethernet/ti/am65-cpts.h new file mode 100644 index 000000000000..cf9fbc28fd03 --- /dev/null +++ b/drivers/net/ethernet/ti/am65-cpts.h @@ -0,0 +1,74 @@ +/* SPDX-License-Identifier: GPL-2.0+ */ +/* TI K3 AM65 CPTS driver interface + * + * Copyright (C) 2020 Texas Instruments Incorporated - http://www.ti.com + */ + +#ifndef K3_CPTS_H_ +#define K3_CPTS_H_ + +#include <linux/device.h> +#include <linux/of.h> + +struct am65_cpts; + +struct am65_cpts_estf_cfg { + u64 ns_period; + u64 ns_start; +}; + +#if IS_ENABLED(CONFIG_TI_K3_AM65_CPTS) +struct am65_cpts *am65_cpts_create(struct device *dev, void __iomem *regs, + struct device_node *node); +int am65_cpts_phc_index(struct am65_cpts *cpts); +void am65_cpts_tx_timestamp(struct am65_cpts *cpts, struct sk_buff *skb); +void am65_cpts_prep_tx_timestamp(struct am65_cpts *cpts, struct sk_buff *skb); +void am65_cpts_rx_enable(struct am65_cpts *cpts, bool en); +u64 am65_cpts_ns_gettime(struct am65_cpts *cpts); +int am65_cpts_estf_enable(struct am65_cpts *cpts, int idx, + struct am65_cpts_estf_cfg *cfg); +void am65_cpts_estf_disable(struct am65_cpts *cpts, int idx); +#else +static inline struct am65_cpts *am65_cpts_create(struct device *dev, + void __iomem *regs, + struct device_node *node) +{ + return ERR_PTR(-EOPNOTSUPP); +} + +static inline int am65_cpts_phc_index(struct am65_cpts *cpts) +{ + return -1; +} + +static inline void am65_cpts_tx_timestamp(struct am65_cpts *cpts, + struct sk_buff *skb) +{ +} + +static inline void am65_cpts_prep_tx_timestamp(struct am65_cpts *cpts, + struct sk_buff *skb) +{ +} + +static inline void am65_cpts_rx_enable(struct am65_cpts *cpts, bool en) +{ +} + +static inline s64 am65_cpts_ns_gettime(struct am65_cpts *cpts) +{ + return 0; +} + +static inline int am65_cpts_estf_enable(struct am65_cpts *cpts, int idx, + struct am65_cpts_estf_cfg *cfg) +{ + return 0; +} + +static inline void am65_cpts_estf_disable(struct am65_cpts *cpts, int idx) +{ +} +#endif + +#endif /* K3_CPTS_H_ */ diff --git a/drivers/net/ethernet/ti/cpmac.c b/drivers/net/ethernet/ti/cpmac.c index a530afe3ce12..c20715107075 100644 --- a/drivers/net/ethernet/ti/cpmac.c +++ b/drivers/net/ethernet/ti/cpmac.c @@ -532,7 +532,7 @@ fatal_error: } -static int cpmac_start_xmit(struct sk_buff *skb, struct net_device *dev) +static netdev_tx_t cpmac_start_xmit(struct sk_buff *skb, struct net_device *dev) { int queue; unsigned int len; diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c index c2c5bf87da01..9b17bbbe102f 100644 --- a/drivers/net/ethernet/ti/cpsw.c +++ b/drivers/net/ethernet/ti/cpsw.c @@ -406,6 +406,7 @@ static void cpsw_rx_handler(void *token, int len, int status) xdp.data_hard_start = pa; xdp.rxq = &priv->xdp_rxq[ch]; + xdp.frame_sz = PAGE_SIZE; port = priv->emac_port + cpsw->data.dual_emac; ret = cpsw_run_xdp(priv, ch, &xdp, page, port); @@ -1569,6 +1570,12 @@ static int cpsw_probe(struct platform_device *pdev) return irq; cpsw->irqs_table[1] = irq; + /* get misc irq*/ + irq = platform_get_irq(pdev, 3); + if (irq <= 0) + return irq; + cpsw->misc_irq = irq; + /* * This may be required here for child devices. */ @@ -1703,6 +1710,21 @@ static int cpsw_probe(struct platform_device *pdev) goto clean_unregister_netdev_ret; } + if (!cpsw->cpts) + goto skip_cpts; + + ret = devm_request_irq(&pdev->dev, cpsw->misc_irq, cpsw_misc_interrupt, + 0, dev_name(&pdev->dev), cpsw); + if (ret < 0) { + dev_err(dev, "error attaching misc irq (%d)\n", ret); + goto clean_unregister_netdev_ret; + } + + /* Enable misc CPTS evnt_pend IRQ */ + cpts_set_irqpoll(cpsw->cpts, false); + writel(0x10, &cpsw->wr_regs->misc_en); + +skip_cpts: cpsw_notice(priv, probe, "initialized device (regs %pa, irq %d, pool size %d)\n", &ss_res->start, cpsw->irqs_table[0], descs_pool_size); @@ -1753,11 +1775,15 @@ static int cpsw_suspend(struct device *dev) struct cpsw_common *cpsw = dev_get_drvdata(dev); int i; + rtnl_lock(); + for (i = 0; i < cpsw->data.slaves; i++) if (cpsw->slaves[i].ndev) if (netif_running(cpsw->slaves[i].ndev)) cpsw_ndo_stop(cpsw->slaves[i].ndev); + rtnl_unlock(); + /* Select sleep pin state */ pinctrl_pm_select_sleep_state(dev); diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c index 0374e6936091..8dc6be11b2ff 100644 --- a/drivers/net/ethernet/ti/cpsw_ale.c +++ b/drivers/net/ethernet/ti/cpsw_ale.c @@ -955,7 +955,7 @@ struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params) ale = devm_kzalloc(params->dev, sizeof(*ale), GFP_KERNEL); if (!ale) - return NULL; + return ERR_PTR(-ENOMEM); ale->p0_untag_vid_mask = devm_kmalloc_array(params->dev, BITS_TO_LONGS(VLAN_N_VID), diff --git a/drivers/net/ethernet/ti/cpsw_new.c b/drivers/net/ethernet/ti/cpsw_new.c index 9209e613257d..1247d35d42ef 100644 --- a/drivers/net/ethernet/ti/cpsw_new.c +++ b/drivers/net/ethernet/ti/cpsw_new.c @@ -348,6 +348,7 @@ static void cpsw_rx_handler(void *token, int len, int status) xdp.data_hard_start = pa; xdp.rxq = &priv->xdp_rxq[ch]; + xdp.frame_sz = PAGE_SIZE; ret = cpsw_run_xdp(priv, ch, &xdp, page, priv->emac_port); if (ret != CPSW_XDP_PASS) @@ -1228,7 +1229,7 @@ static int cpsw_probe_dt(struct cpsw_common *cpsw) data->active_slave = 0; data->channels = CPSW_MAX_QUEUES; data->ale_entries = CPSW_ALE_NUM_ENTRIES; - data->dual_emac = 1; + data->dual_emac = true; data->bd_ram_size = CPSW_BD_RAM_SIZE; data->mac_control = 0; @@ -1896,6 +1897,11 @@ static int cpsw_probe(struct platform_device *pdev) return irq; cpsw->irqs_table[1] = irq; + irq = platform_get_irq_byname(pdev, "misc"); + if (irq <= 0) + return irq; + cpsw->misc_irq = irq; + platform_set_drvdata(pdev, cpsw); /* This may be required here for child devices. */ pm_runtime_enable(dev); @@ -1916,7 +1922,7 @@ static int cpsw_probe(struct platform_device *pdev) soc = soc_device_match(cpsw_soc_devices); if (soc) - cpsw->quirk_irq = 1; + cpsw->quirk_irq = true; cpsw->rx_packet_max = rx_packet_max; cpsw->descs_pool_size = descs_pool_size; @@ -1975,6 +1981,21 @@ static int cpsw_probe(struct platform_device *pdev) goto clean_unregister_netdev; } + if (!cpsw->cpts) + goto skip_cpts; + + ret = devm_request_irq(dev, cpsw->misc_irq, cpsw_misc_interrupt, + 0, dev_name(&pdev->dev), cpsw); + if (ret < 0) { + dev_err(dev, "error attaching misc irq (%d)\n", ret); + goto clean_unregister_netdev; + } + + /* Enable misc CPTS evnt_pend IRQ */ + cpts_set_irqpoll(cpsw->cpts, false); + writel(0x10, &cpsw->wr_regs->misc_en); + +skip_cpts: ret = cpsw_register_notifiers(cpsw); if (ret) goto clean_unregister_netdev; diff --git a/drivers/net/ethernet/ti/cpsw_priv.c b/drivers/net/ethernet/ti/cpsw_priv.c index 97a058ca60ac..a399f3659346 100644 --- a/drivers/net/ethernet/ti/cpsw_priv.c +++ b/drivers/net/ethernet/ti/cpsw_priv.c @@ -28,6 +28,8 @@ #include "cpsw_sl.h" #include "davinci_cpdma.h" +#define CPTS_N_ETX_TS 4 + int (*cpsw_slave_index)(struct cpsw_common *cpsw, struct cpsw_priv *priv); void cpsw_intr_enable(struct cpsw_common *cpsw) @@ -112,6 +114,18 @@ irqreturn_t cpsw_rx_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } +irqreturn_t cpsw_misc_interrupt(int irq, void *dev_id) +{ + struct cpsw_common *cpsw = dev_id; + + writel(0, &cpsw->wr_regs->misc_en); + cpdma_ctlr_eoi(cpsw->dma, CPDMA_EOI_MISC); + cpts_misc_interrupt(cpsw->cpts); + writel(0x10, &cpsw->wr_regs->misc_en); + + return IRQ_HANDLED; +} + int cpsw_tx_mq_poll(struct napi_struct *napi_tx, int budget) { struct cpsw_common *cpsw = napi_to_cpsw(napi_tx); @@ -490,9 +504,9 @@ int cpsw_init_common(struct cpsw_common *cpsw, void __iomem *ss_regs, ale_params.ale_ports = CPSW_ALE_PORTS_NUM; cpsw->ale = cpsw_ale_create(&ale_params); - if (!cpsw->ale) { + if (IS_ERR(cpsw->ale)) { dev_err(dev, "error initializing ale engine\n"); - return -ENODEV; + return PTR_ERR(cpsw->ale); } dma_params.dev = dev; @@ -522,7 +536,8 @@ int cpsw_init_common(struct cpsw_common *cpsw, void __iomem *ss_regs, if (!cpts_node) cpts_node = cpsw->dev->of_node; - cpsw->cpts = cpts_create(cpsw->dev, cpts_regs, cpts_node); + cpsw->cpts = cpts_create(cpsw->dev, cpts_regs, cpts_node, + CPTS_N_ETX_TS); if (IS_ERR(cpsw->cpts)) { ret = PTR_ERR(cpsw->cpts); cpdma_ctlr_destroy(cpsw->dma); @@ -1340,7 +1355,7 @@ int cpsw_run_xdp(struct cpsw_priv *priv, int ch, struct xdp_buff *xdp, ret = CPSW_XDP_PASS; break; case XDP_TX: - xdpf = convert_to_xdp_frame(xdp); + xdpf = xdp_convert_buff_to_frame(xdp); if (unlikely(!xdpf)) goto drop; diff --git a/drivers/net/ethernet/ti/cpsw_priv.h b/drivers/net/ethernet/ti/cpsw_priv.h index b8d7b924ee3d..bf4e179b4ca4 100644 --- a/drivers/net/ethernet/ti/cpsw_priv.h +++ b/drivers/net/ethernet/ti/cpsw_priv.h @@ -350,6 +350,7 @@ struct cpsw_common { bool rx_irq_disabled; bool tx_irq_disabled; u32 irqs_table[IRQ_NUM]; + int misc_irq; struct cpts *cpts; struct devlink *devlink; int rx_ch_num, tx_ch_num; @@ -442,6 +443,7 @@ int cpsw_run_xdp(struct cpsw_priv *priv, int ch, struct xdp_buff *xdp, struct page *page, int port); irqreturn_t cpsw_tx_interrupt(int irq, void *dev_id); irqreturn_t cpsw_rx_interrupt(int irq, void *dev_id); +irqreturn_t cpsw_misc_interrupt(int irq, void *dev_id); int cpsw_tx_mq_poll(struct napi_struct *napi_tx, int budget); int cpsw_tx_poll(struct napi_struct *napi_tx, int budget); int cpsw_rx_mq_poll(struct napi_struct *napi_rx, int budget); diff --git a/drivers/net/ethernet/ti/cpts.c b/drivers/net/ethernet/ti/cpts.c index 729ce09dded9..7c55d395de2c 100644 --- a/drivers/net/ethernet/ti/cpts.c +++ b/drivers/net/ethernet/ti/cpts.c @@ -21,16 +21,21 @@ #include "cpts.h" #define CPTS_SKB_TX_WORK_TIMEOUT 1 /* jiffies */ +#define CPTS_SKB_RX_TX_TMO 100 /*ms */ +#define CPTS_EVENT_RX_TX_TIMEOUT (100) /* ms */ struct cpts_skb_cb_data { + u32 skb_mtype_seqid; unsigned long tmo; }; #define cpts_read32(c, r) readl_relaxed(&c->reg->r) #define cpts_write32(c, v, r) writel_relaxed(v, &c->reg->r) -static int cpts_match(struct sk_buff *skb, unsigned int ptp_class, - u16 ts_seqid, u8 ts_msgtype); +static int cpts_event_port(struct cpts_event *event) +{ + return (event->high >> PORT_NUMBER_SHIFT) & PORT_NUMBER_MASK; +} static int event_expired(struct cpts_event *event) { @@ -71,7 +76,7 @@ static int cpts_purge_events(struct cpts *cpts) } if (removed) - pr_debug("cpts: event pool cleaned up %d\n", removed); + dev_dbg(cpts->dev, "cpts: event pool cleaned up %d\n", removed); return removed ? 0 : -1; } @@ -94,132 +99,126 @@ static void cpts_purge_txq(struct cpts *cpts) dev_dbg(cpts->dev, "txq cleaned up %d\n", removed); } -static bool cpts_match_tx_ts(struct cpts *cpts, struct cpts_event *event) -{ - struct sk_buff *skb, *tmp; - u16 seqid; - u8 mtype; - bool found = false; - - mtype = (event->high >> MESSAGE_TYPE_SHIFT) & MESSAGE_TYPE_MASK; - seqid = (event->high >> SEQUENCE_ID_SHIFT) & SEQUENCE_ID_MASK; - - /* no need to grab txq.lock as access is always done under cpts->lock */ - skb_queue_walk_safe(&cpts->txq, skb, tmp) { - struct skb_shared_hwtstamps ssh; - unsigned int class = ptp_classify_raw(skb); - struct cpts_skb_cb_data *skb_cb = - (struct cpts_skb_cb_data *)skb->cb; - - if (cpts_match(skb, class, seqid, mtype)) { - u64 ns = timecounter_cyc2time(&cpts->tc, event->low); - - memset(&ssh, 0, sizeof(ssh)); - ssh.hwtstamp = ns_to_ktime(ns); - skb_tstamp_tx(skb, &ssh); - found = true; - __skb_unlink(skb, &cpts->txq); - dev_consume_skb_any(skb); - dev_dbg(cpts->dev, "match tx timestamp mtype %u seqid %04x\n", - mtype, seqid); - break; - } - - if (time_after(jiffies, skb_cb->tmo)) { - /* timeout any expired skbs over 1s */ - dev_dbg(cpts->dev, "expiring tx timestamp from txq\n"); - __skb_unlink(skb, &cpts->txq); - dev_consume_skb_any(skb); - } - } - - return found; -} - /* * Returns zero if matching event type was found. */ static int cpts_fifo_read(struct cpts *cpts, int match) { + struct ptp_clock_event pevent; + bool need_schedule = false; + struct cpts_event *event; + unsigned long flags; int i, type = -1; u32 hi, lo; - struct cpts_event *event; + + spin_lock_irqsave(&cpts->lock, flags); for (i = 0; i < CPTS_FIFO_DEPTH; i++) { if (cpts_fifo_pop(cpts, &hi, &lo)) break; if (list_empty(&cpts->pool) && cpts_purge_events(cpts)) { - pr_err("cpts: event pool empty\n"); - return -1; + dev_warn(cpts->dev, "cpts: event pool empty\n"); + break; } event = list_first_entry(&cpts->pool, struct cpts_event, list); - event->tmo = jiffies + 2; event->high = hi; event->low = lo; + event->timestamp = timecounter_cyc2time(&cpts->tc, event->low); type = event_type(event); + + dev_dbg(cpts->dev, "CPTS_EV: %d high:%08X low:%08x\n", + type, event->high, event->low); switch (type) { - case CPTS_EV_TX: - if (cpts_match_tx_ts(cpts, event)) { - /* if the new event matches an existing skb, - * then don't queue it - */ - break; - } - /* fall through */ case CPTS_EV_PUSH: + WRITE_ONCE(cpts->cur_timestamp, lo); + timecounter_read(&cpts->tc); + if (cpts->mult_new) { + cpts->cc.mult = cpts->mult_new; + cpts->mult_new = 0; + } + if (!cpts->irq_poll) + complete(&cpts->ts_push_complete); + break; + case CPTS_EV_TX: case CPTS_EV_RX: + event->tmo = jiffies + + msecs_to_jiffies(CPTS_EVENT_RX_TX_TIMEOUT); + list_del_init(&event->list); list_add_tail(&event->list, &cpts->events); + need_schedule = true; break; case CPTS_EV_ROLL: case CPTS_EV_HALF: + break; case CPTS_EV_HW: + pevent.timestamp = event->timestamp; + pevent.type = PTP_CLOCK_EXTTS; + pevent.index = cpts_event_port(event) - 1; + ptp_clock_event(cpts->clock, &pevent); break; default: - pr_err("cpts: unknown event type\n"); + dev_err(cpts->dev, "cpts: unknown event type\n"); break; } if (type == match) break; } + + spin_unlock_irqrestore(&cpts->lock, flags); + + if (!cpts->irq_poll && need_schedule) + ptp_schedule_worker(cpts->clock, 0); + return type == match ? 0 : -1; } +void cpts_misc_interrupt(struct cpts *cpts) +{ + cpts_fifo_read(cpts, -1); +} +EXPORT_SYMBOL_GPL(cpts_misc_interrupt); + static u64 cpts_systim_read(const struct cyclecounter *cc) { - u64 val = 0; - struct cpts_event *event; - struct list_head *this, *next; struct cpts *cpts = container_of(cc, struct cpts, cc); + return READ_ONCE(cpts->cur_timestamp); +} + +static void cpts_update_cur_time(struct cpts *cpts, int match, + struct ptp_system_timestamp *sts) +{ + unsigned long flags; + + reinit_completion(&cpts->ts_push_complete); + + /* use spin_lock_irqsave() here as it has to run very fast */ + spin_lock_irqsave(&cpts->lock, flags); + ptp_read_system_prets(sts); cpts_write32(cpts, TS_PUSH, ts_push); - if (cpts_fifo_read(cpts, CPTS_EV_PUSH)) - pr_err("cpts: unable to obtain a time stamp\n"); + cpts_read32(cpts, ts_push); + ptp_read_system_postts(sts); + spin_unlock_irqrestore(&cpts->lock, flags); - list_for_each_safe(this, next, &cpts->events) { - event = list_entry(this, struct cpts_event, list); - if (event_type(event) == CPTS_EV_PUSH) { - list_del_init(&event->list); - list_add(&event->list, &cpts->pool); - val = event->low; - break; - } - } + if (cpts->irq_poll && cpts_fifo_read(cpts, match) && match != -1) + dev_err(cpts->dev, "cpts: unable to obtain a time stamp\n"); - return val; + if (!cpts->irq_poll && + !wait_for_completion_timeout(&cpts->ts_push_complete, HZ)) + dev_err(cpts->dev, "cpts: obtain a time stamp timeout\n"); } /* PTP clock operations */ static int cpts_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb) { - u64 adj; - u32 diff, mult; - int neg_adj = 0; - unsigned long flags; struct cpts *cpts = container_of(ptp, struct cpts, info); + int neg_adj = 0; + u32 diff, mult; + u64 adj; if (ppb < 0) { neg_adj = 1; @@ -230,38 +229,40 @@ static int cpts_ptp_adjfreq(struct ptp_clock_info *ptp, s32 ppb) adj *= ppb; diff = div_u64(adj, 1000000000ULL); - spin_lock_irqsave(&cpts->lock, flags); + mutex_lock(&cpts->ptp_clk_mutex); - timecounter_read(&cpts->tc); + cpts->mult_new = neg_adj ? mult - diff : mult + diff; - cpts->cc.mult = neg_adj ? mult - diff : mult + diff; - - spin_unlock_irqrestore(&cpts->lock, flags); + cpts_update_cur_time(cpts, CPTS_EV_PUSH, NULL); + mutex_unlock(&cpts->ptp_clk_mutex); return 0; } static int cpts_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta) { - unsigned long flags; struct cpts *cpts = container_of(ptp, struct cpts, info); - spin_lock_irqsave(&cpts->lock, flags); + mutex_lock(&cpts->ptp_clk_mutex); timecounter_adjtime(&cpts->tc, delta); - spin_unlock_irqrestore(&cpts->lock, flags); + mutex_unlock(&cpts->ptp_clk_mutex); return 0; } -static int cpts_ptp_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts) +static int cpts_ptp_gettimeex(struct ptp_clock_info *ptp, + struct timespec64 *ts, + struct ptp_system_timestamp *sts) { - u64 ns; - unsigned long flags; struct cpts *cpts = container_of(ptp, struct cpts, info); + u64 ns; + + mutex_lock(&cpts->ptp_clk_mutex); + + cpts_update_cur_time(cpts, CPTS_EV_PUSH, sts); - spin_lock_irqsave(&cpts->lock, flags); ns = timecounter_read(&cpts->tc); - spin_unlock_irqrestore(&cpts->lock, flags); + mutex_unlock(&cpts->ptp_clk_mutex); *ts = ns_to_timespec64(ns); @@ -271,15 +272,38 @@ static int cpts_ptp_gettime(struct ptp_clock_info *ptp, struct timespec64 *ts) static int cpts_ptp_settime(struct ptp_clock_info *ptp, const struct timespec64 *ts) { - u64 ns; - unsigned long flags; struct cpts *cpts = container_of(ptp, struct cpts, info); + u64 ns; ns = timespec64_to_ns(ts); - spin_lock_irqsave(&cpts->lock, flags); + mutex_lock(&cpts->ptp_clk_mutex); timecounter_init(&cpts->tc, &cpts->cc, ns); - spin_unlock_irqrestore(&cpts->lock, flags); + mutex_unlock(&cpts->ptp_clk_mutex); + + return 0; +} + +static int cpts_extts_enable(struct cpts *cpts, u32 index, int on) +{ + u32 v; + + if (((cpts->hw_ts_enable & BIT(index)) >> index) == on) + return 0; + + mutex_lock(&cpts->ptp_clk_mutex); + + v = cpts_read32(cpts, control); + if (on) { + v |= BIT(8 + index); + cpts->hw_ts_enable |= BIT(index); + } else { + v &= ~BIT(8 + index); + cpts->hw_ts_enable &= ~BIT(index); + } + cpts_write32(cpts, v, control); + + mutex_unlock(&cpts->ptp_clk_mutex); return 0; } @@ -287,28 +311,120 @@ static int cpts_ptp_settime(struct ptp_clock_info *ptp, static int cpts_ptp_enable(struct ptp_clock_info *ptp, struct ptp_clock_request *rq, int on) { + struct cpts *cpts = container_of(ptp, struct cpts, info); + + switch (rq->type) { + case PTP_CLK_REQ_EXTTS: + return cpts_extts_enable(cpts, rq->extts.index, on); + default: + break; + } + return -EOPNOTSUPP; } +static bool cpts_match_tx_ts(struct cpts *cpts, struct cpts_event *event) +{ + struct sk_buff_head txq_list; + struct sk_buff *skb, *tmp; + unsigned long flags; + bool found = false; + u32 mtype_seqid; + + mtype_seqid = event->high & + ((MESSAGE_TYPE_MASK << MESSAGE_TYPE_SHIFT) | + (SEQUENCE_ID_MASK << SEQUENCE_ID_SHIFT) | + (EVENT_TYPE_MASK << EVENT_TYPE_SHIFT)); + + __skb_queue_head_init(&txq_list); + + spin_lock_irqsave(&cpts->txq.lock, flags); + skb_queue_splice_init(&cpts->txq, &txq_list); + spin_unlock_irqrestore(&cpts->txq.lock, flags); + + skb_queue_walk_safe(&txq_list, skb, tmp) { + struct skb_shared_hwtstamps ssh; + struct cpts_skb_cb_data *skb_cb = + (struct cpts_skb_cb_data *)skb->cb; + + if (mtype_seqid == skb_cb->skb_mtype_seqid) { + memset(&ssh, 0, sizeof(ssh)); + ssh.hwtstamp = ns_to_ktime(event->timestamp); + skb_tstamp_tx(skb, &ssh); + found = true; + __skb_unlink(skb, &txq_list); + dev_consume_skb_any(skb); + dev_dbg(cpts->dev, "match tx timestamp mtype_seqid %08x\n", + mtype_seqid); + break; + } + + if (time_after(jiffies, skb_cb->tmo)) { + /* timeout any expired skbs over 1s */ + dev_dbg(cpts->dev, "expiring tx timestamp from txq\n"); + __skb_unlink(skb, &txq_list); + dev_consume_skb_any(skb); + } + } + + spin_lock_irqsave(&cpts->txq.lock, flags); + skb_queue_splice(&txq_list, &cpts->txq); + spin_unlock_irqrestore(&cpts->txq.lock, flags); + + return found; +} + +static void cpts_process_events(struct cpts *cpts) +{ + struct list_head *this, *next; + struct cpts_event *event; + LIST_HEAD(events_free); + unsigned long flags; + LIST_HEAD(events); + + spin_lock_irqsave(&cpts->lock, flags); + list_splice_init(&cpts->events, &events); + spin_unlock_irqrestore(&cpts->lock, flags); + + list_for_each_safe(this, next, &events) { + event = list_entry(this, struct cpts_event, list); + if (cpts_match_tx_ts(cpts, event) || + time_after(jiffies, event->tmo)) { + list_del_init(&event->list); + list_add(&event->list, &events_free); + } + } + + spin_lock_irqsave(&cpts->lock, flags); + list_splice_tail(&events, &cpts->events); + list_splice_tail(&events_free, &cpts->pool); + spin_unlock_irqrestore(&cpts->lock, flags); +} + static long cpts_overflow_check(struct ptp_clock_info *ptp) { struct cpts *cpts = container_of(ptp, struct cpts, info); unsigned long delay = cpts->ov_check_period; - struct timespec64 ts; unsigned long flags; + u64 ns; - spin_lock_irqsave(&cpts->lock, flags); - ts = ns_to_timespec64(timecounter_read(&cpts->tc)); + mutex_lock(&cpts->ptp_clk_mutex); + + cpts_update_cur_time(cpts, -1, NULL); + ns = timecounter_read(&cpts->tc); + + cpts_process_events(cpts); + spin_lock_irqsave(&cpts->txq.lock, flags); if (!skb_queue_empty(&cpts->txq)) { cpts_purge_txq(cpts); if (!skb_queue_empty(&cpts->txq)) delay = CPTS_SKB_TX_WORK_TIMEOUT; } - spin_unlock_irqrestore(&cpts->lock, flags); + spin_unlock_irqrestore(&cpts->txq.lock, flags); - pr_debug("cpts overflow check at %lld.%09ld\n", - (long long)ts.tv_sec, ts.tv_nsec); + dev_dbg(cpts->dev, "cpts overflow check at %lld\n", ns); + mutex_unlock(&cpts->ptp_clk_mutex); return (long)delay; } @@ -321,18 +437,21 @@ static const struct ptp_clock_info cpts_info = { .pps = 0, .adjfreq = cpts_ptp_adjfreq, .adjtime = cpts_ptp_adjtime, - .gettime64 = cpts_ptp_gettime, + .gettimex64 = cpts_ptp_gettimeex, .settime64 = cpts_ptp_settime, .enable = cpts_ptp_enable, .do_aux_work = cpts_overflow_check, }; -static int cpts_match(struct sk_buff *skb, unsigned int ptp_class, - u16 ts_seqid, u8 ts_msgtype) +static int cpts_skb_get_mtype_seqid(struct sk_buff *skb, u32 *mtype_seqid) { - u16 *seqid; - unsigned int offset = 0; + unsigned int ptp_class = ptp_classify_raw(skb); u8 *msgtype, *data = skb->data; + unsigned int offset = 0; + u16 *seqid; + + if (ptp_class == PTP_CLASS_NONE) + return 0; if (ptp_class & PTP_CLASS_VLAN) offset += VLAN_HLEN; @@ -360,25 +479,23 @@ static int cpts_match(struct sk_buff *skb, unsigned int ptp_class, msgtype = data + offset; seqid = (u16 *)(data + offset + OFF_PTP_SEQUENCE_ID); + *mtype_seqid = (*msgtype & MESSAGE_TYPE_MASK) << MESSAGE_TYPE_SHIFT; + *mtype_seqid |= (ntohs(*seqid) & SEQUENCE_ID_MASK) << SEQUENCE_ID_SHIFT; - return (ts_msgtype == (*msgtype & 0xf) && ts_seqid == ntohs(*seqid)); + return 1; } -static u64 cpts_find_ts(struct cpts *cpts, struct sk_buff *skb, int ev_type) +static u64 cpts_find_ts(struct cpts *cpts, struct sk_buff *skb, + int ev_type, u32 skb_mtype_seqid) { - u64 ns = 0; - struct cpts_event *event; struct list_head *this, *next; - unsigned int class = ptp_classify_raw(skb); + struct cpts_event *event; unsigned long flags; - u16 seqid; - u8 mtype; - - if (class == PTP_CLASS_NONE) - return 0; + u32 mtype_seqid; + u64 ns = 0; - spin_lock_irqsave(&cpts->lock, flags); cpts_fifo_read(cpts, -1); + spin_lock_irqsave(&cpts->lock, flags); list_for_each_safe(this, next, &cpts->events) { event = list_entry(this, struct cpts_event, list); if (event_expired(event)) { @@ -386,29 +503,19 @@ static u64 cpts_find_ts(struct cpts *cpts, struct sk_buff *skb, int ev_type) list_add(&event->list, &cpts->pool); continue; } - mtype = (event->high >> MESSAGE_TYPE_SHIFT) & MESSAGE_TYPE_MASK; - seqid = (event->high >> SEQUENCE_ID_SHIFT) & SEQUENCE_ID_MASK; - if (ev_type == event_type(event) && - cpts_match(skb, class, seqid, mtype)) { - ns = timecounter_cyc2time(&cpts->tc, event->low); + + mtype_seqid = event->high & + ((MESSAGE_TYPE_MASK << MESSAGE_TYPE_SHIFT) | + (SEQUENCE_ID_MASK << SEQUENCE_ID_SHIFT) | + (EVENT_TYPE_MASK << EVENT_TYPE_SHIFT)); + + if (mtype_seqid == skb_mtype_seqid) { + ns = event->timestamp; list_del_init(&event->list); list_add(&event->list, &cpts->pool); break; } } - - if (ev_type == CPTS_EV_TX && !ns) { - struct cpts_skb_cb_data *skb_cb = - (struct cpts_skb_cb_data *)skb->cb; - /* Not found, add frame to queue for processing later. - * The periodic FIFO check will handle this. - */ - skb_get(skb); - /* get the timestamp for timeouts */ - skb_cb->tmo = jiffies + msecs_to_jiffies(100); - __skb_queue_tail(&cpts->txq, skb); - ptp_schedule_worker(cpts->clock, 0); - } spin_unlock_irqrestore(&cpts->lock, flags); return ns; @@ -416,10 +523,21 @@ static u64 cpts_find_ts(struct cpts *cpts, struct sk_buff *skb, int ev_type) void cpts_rx_timestamp(struct cpts *cpts, struct sk_buff *skb) { - u64 ns; + struct cpts_skb_cb_data *skb_cb = (struct cpts_skb_cb_data *)skb->cb; struct skb_shared_hwtstamps *ssh; + int ret; + u64 ns; - ns = cpts_find_ts(cpts, skb, CPTS_EV_RX); + ret = cpts_skb_get_mtype_seqid(skb, &skb_cb->skb_mtype_seqid); + if (!ret) + return; + + skb_cb->skb_mtype_seqid |= (CPTS_EV_RX << EVENT_TYPE_SHIFT); + + dev_dbg(cpts->dev, "%s mtype seqid %08x\n", + __func__, skb_cb->skb_mtype_seqid); + + ns = cpts_find_ts(cpts, skb, CPTS_EV_RX, skb_cb->skb_mtype_seqid); if (!ns) return; ssh = skb_hwtstamps(skb); @@ -430,17 +548,27 @@ EXPORT_SYMBOL_GPL(cpts_rx_timestamp); void cpts_tx_timestamp(struct cpts *cpts, struct sk_buff *skb) { - u64 ns; - struct skb_shared_hwtstamps ssh; + struct cpts_skb_cb_data *skb_cb = (struct cpts_skb_cb_data *)skb->cb; + int ret; if (!(skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS)) return; - ns = cpts_find_ts(cpts, skb, CPTS_EV_TX); - if (!ns) + + ret = cpts_skb_get_mtype_seqid(skb, &skb_cb->skb_mtype_seqid); + if (!ret) return; - memset(&ssh, 0, sizeof(ssh)); - ssh.hwtstamp = ns_to_ktime(ns); - skb_tstamp_tx(skb, &ssh); + + skb_cb->skb_mtype_seqid |= (CPTS_EV_TX << EVENT_TYPE_SHIFT); + + dev_dbg(cpts->dev, "%s mtype seqid %08x\n", + __func__, skb_cb->skb_mtype_seqid); + + /* Always defer TX TS processing to PTP worker */ + skb_get(skb); + /* get the timestamp for timeouts */ + skb_cb->tmo = jiffies + msecs_to_jiffies(CPTS_SKB_RX_TX_TMO); + skb_queue_tail(&cpts->txq, skb); + ptp_schedule_worker(cpts->clock, 0); } EXPORT_SYMBOL_GPL(cpts_tx_timestamp); @@ -632,7 +760,7 @@ of_error: } struct cpts *cpts_create(struct device *dev, void __iomem *regs, - struct device_node *node) + struct device_node *node, u32 n_ext_ts) { struct cpts *cpts; int ret; @@ -643,7 +771,10 @@ struct cpts *cpts_create(struct device *dev, void __iomem *regs, cpts->dev = dev; cpts->reg = (struct cpsw_cpts __iomem *)regs; + cpts->irq_poll = true; spin_lock_init(&cpts->lock); + mutex_init(&cpts->ptp_clk_mutex); + init_completion(&cpts->ts_push_complete); ret = cpts_of_parse(cpts, node); if (ret) @@ -668,6 +799,9 @@ struct cpts *cpts_create(struct device *dev, void __iomem *regs, cpts->cc.mask = CLOCKSOURCE_MASK(32); cpts->info = cpts_info; + if (n_ext_ts) + cpts->info.n_ext_ts = n_ext_ts; + cpts_calc_mult_shift(cpts); /* save cc.mult original value as it can be modified * by cpts_ptp_adjfreq(). diff --git a/drivers/net/ethernet/ti/cpts.h b/drivers/net/ethernet/ti/cpts.h index bb997c11ee15..07222f651d2e 100644 --- a/drivers/net/ethernet/ti/cpts.h +++ b/drivers/net/ethernet/ti/cpts.h @@ -94,6 +94,7 @@ struct cpts_event { unsigned long tmo; u32 high; u32 low; + u64 timestamp; }; struct cpts { @@ -103,7 +104,7 @@ struct cpts { int rx_enable; struct ptp_clock_info info; struct ptp_clock *clock; - spinlock_t lock; /* protects time registers */ + spinlock_t lock; /* protects fifo/events */ u32 cc_mult; /* for the nominal frequency */ struct cyclecounter cc; struct timecounter tc; @@ -114,6 +115,12 @@ struct cpts { struct cpts_event pool_data[CPTS_MAX_EVENTS]; unsigned long ov_check_period; struct sk_buff_head txq; + u64 cur_timestamp; + u32 mult_new; + struct mutex ptp_clk_mutex; /* sync PTP interface and worker */ + bool irq_poll; + struct completion ts_push_complete; + u32 hw_ts_enable; }; void cpts_rx_timestamp(struct cpts *cpts, struct sk_buff *skb); @@ -121,8 +128,9 @@ void cpts_tx_timestamp(struct cpts *cpts, struct sk_buff *skb); int cpts_register(struct cpts *cpts); void cpts_unregister(struct cpts *cpts); struct cpts *cpts_create(struct device *dev, void __iomem *regs, - struct device_node *node); + struct device_node *node, u32 n_ext_ts); void cpts_release(struct cpts *cpts); +void cpts_misc_interrupt(struct cpts *cpts); static inline bool cpts_can_timestamp(struct cpts *cpts, struct sk_buff *skb) { @@ -134,6 +142,11 @@ static inline bool cpts_can_timestamp(struct cpts *cpts, struct sk_buff *skb) return true; } +static inline void cpts_set_irqpoll(struct cpts *cpts, bool en) +{ + cpts->irq_poll = en; +} + #else struct cpts; @@ -146,7 +159,7 @@ static inline void cpts_tx_timestamp(struct cpts *cpts, struct sk_buff *skb) static inline struct cpts *cpts_create(struct device *dev, void __iomem *regs, - struct device_node *node) + struct device_node *node, u32 n_ext_ts) { return NULL; } @@ -169,6 +182,14 @@ static inline bool cpts_can_timestamp(struct cpts *cpts, struct sk_buff *skb) { return false; } + +static inline void cpts_misc_interrupt(struct cpts *cpts) +{ +} + +static inline void cpts_set_irqpoll(struct cpts *cpts, bool en) +{ +} #endif diff --git a/drivers/net/ethernet/ti/davinci_mdio.c b/drivers/net/ethernet/ti/davinci_mdio.c index 38b7f6d35759..702fdc393da0 100644 --- a/drivers/net/ethernet/ti/davinci_mdio.c +++ b/drivers/net/ethernet/ti/davinci_mdio.c @@ -397,6 +397,8 @@ static int davinci_mdio_probe(struct platform_device *pdev) data->dev = dev; res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!res) + return -EINVAL; data->regs = devm_ioremap(dev, res->start, resource_size(res)); if (!data->regs) return -ENOMEM; diff --git a/drivers/net/ethernet/ti/k3-cppi-desc-pool.c b/drivers/net/ethernet/ti/k3-cppi-desc-pool.c index ad7cfc1316ce..38cc12f9f133 100644 --- a/drivers/net/ethernet/ti/k3-cppi-desc-pool.c +++ b/drivers/net/ethernet/ti/k3-cppi-desc-pool.c @@ -64,8 +64,8 @@ k3_cppi_desc_pool_create_name(struct device *dev, size_t size, return ERR_PTR(-ENOMEM); pool->gen_pool = gen_pool_create(ilog2(pool->desc_size), -1); - if (IS_ERR(pool->gen_pool)) { - ret = PTR_ERR(pool->gen_pool); + if (!pool->gen_pool) { + ret = -ENOMEM; dev_err(pool->dev, "pool create failed %d\n", ret); kfree_const(pool_name); goto gen_pool_create_fail; diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c index fb36115e9c51..28093923a7fb 100644 --- a/drivers/net/ethernet/ti/netcp_ethss.c +++ b/drivers/net/ethernet/ti/netcp_ethss.c @@ -3704,9 +3704,9 @@ static int gbe_probe(struct netcp_device *netcp_device, struct device *dev, ale_params.nu_switch_ale = true; } gbe_dev->ale = cpsw_ale_create(&ale_params); - if (!gbe_dev->ale) { + if (IS_ERR(gbe_dev->ale)) { dev_err(gbe_dev->dev, "error initializing ale engine\n"); - ret = -ENODEV; + ret = PTR_ERR(gbe_dev->ale); goto free_sec_ports; } else { dev_dbg(gbe_dev->dev, "Created a gbe ale engine\n"); @@ -3716,7 +3716,8 @@ static int gbe_probe(struct netcp_device *netcp_device, struct device *dev, if (!cpts_node) cpts_node = of_node_get(node); - gbe_dev->cpts = cpts_create(gbe_dev->dev, gbe_dev->cpts_reg, cpts_node); + gbe_dev->cpts = cpts_create(gbe_dev->dev, gbe_dev->cpts_reg, + cpts_node, 0); of_node_put(cpts_node); if (IS_ENABLED(CONFIG_TI_CPTS) && IS_ERR(gbe_dev->cpts)) { ret = PTR_ERR(gbe_dev->cpts); diff --git a/drivers/net/ethernet/ti/tlan.c b/drivers/net/ethernet/ti/tlan.c index ad465202980a..857709828058 100644 --- a/drivers/net/ethernet/ti/tlan.c +++ b/drivers/net/ethernet/ti/tlan.c @@ -70,7 +70,7 @@ MODULE_DESCRIPTION("Driver for TI ThunderLAN based ethernet PCI adapters"); MODULE_LICENSE("GPL"); /* Turn on debugging. - * See Documentation/networking/device_drivers/ti/tlan.txt for details + * See Documentation/networking/device_drivers/ti/tlan.rst for details */ static int debug; module_param(debug, int, 0); diff --git a/drivers/net/ethernet/toshiba/ps3_gelic_net.c b/drivers/net/ethernet/toshiba/ps3_gelic_net.c index 070dd6fa9401..d9a5722f561b 100644 --- a/drivers/net/ethernet/toshiba/ps3_gelic_net.c +++ b/drivers/net/ethernet/toshiba/ps3_gelic_net.c @@ -382,8 +382,6 @@ static int gelic_descr_prepare_rx(struct gelic_card *card, descr->skb = dev_alloc_skb(bufsize + GELIC_NET_RXBUF_ALIGN - 1); if (!descr->skb) { descr->buf_addr = 0; /* tell DMAC don't touch memory */ - dev_info(ctodev(card), - "%s:allocate skb failed !!\n", __func__); return -ENOMEM; } descr->buf_size = cpu_to_be32(bufsize); @@ -1150,7 +1148,7 @@ static irqreturn_t gelic_card_interrupt(int irq, void *ptr) * gelic_net_poll_controller - artificial interrupt for netconsole etc. * @netdev: interface device structure * - * see Documentation/networking/netconsole.txt + * see Documentation/networking/netconsole.rst */ void gelic_net_poll_controller(struct net_device *netdev) { diff --git a/drivers/net/ethernet/toshiba/spider_net.c b/drivers/net/ethernet/toshiba/spider_net.c index 6576271642c1..3902b3aeb0c2 100644 --- a/drivers/net/ethernet/toshiba/spider_net.c +++ b/drivers/net/ethernet/toshiba/spider_net.c @@ -1615,7 +1615,7 @@ spider_net_interrupt(int irq, void *ptr) * spider_net_poll_controller - artificial interrupt for netconsole etc. * @netdev: interface device structure * - * see Documentation/networking/netconsole.txt + * see Documentation/networking/netconsole.rst */ static void spider_net_poll_controller(struct net_device *netdev) diff --git a/drivers/net/ethernet/via/Kconfig b/drivers/net/ethernet/via/Kconfig index a962097b58c6..6cff5f7d57c4 100644 --- a/drivers/net/ethernet/via/Kconfig +++ b/drivers/net/ethernet/via/Kconfig @@ -19,6 +19,7 @@ if NET_VENDOR_VIA config VIA_RHINE tristate "VIA Rhine support" depends on PCI || (OF_IRQ && GENERIC_PCI_IOMAP) + depends on PCI || ARCH_VT8500 || COMPILE_TEST depends on HAS_DMA select CRC32 select MII diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index 3e313e71ae36..929244064abd 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -1410,9 +1410,9 @@ static int temac_probe(struct platform_device *pdev) res = platform_get_resource(pdev, IORESOURCE_MEM, 0); lp->regs = devm_ioremap(&pdev->dev, res->start, resource_size(res)); - if (IS_ERR(lp->regs)) { + if (!lp->regs) { dev_err(&pdev->dev, "could not map TEMAC registers\n"); - return PTR_ERR(lp->regs); + return -ENOMEM; } /* Select register access functions with the specified @@ -1505,10 +1505,10 @@ static int temac_probe(struct platform_device *pdev) res = platform_get_resource(pdev, IORESOURCE_MEM, 1); lp->sdma_regs = devm_ioremap(&pdev->dev, res->start, resource_size(res)); - if (IS_ERR(lp->sdma_regs)) { + if (!lp->sdma_regs) { dev_err(&pdev->dev, "could not map DMA registers\n"); - return PTR_ERR(lp->sdma_regs); + return -ENOMEM; } if (pdata->dma_little_endian) { lp->dma_in = temac_dma_in32_le; |