diff options
92 files changed, 7093 insertions, 1029 deletions
diff --git a/Documentation/devicetree/bindings/net/dsa/ar9331.txt b/Documentation/devicetree/bindings/net/dsa/ar9331.txt deleted file mode 100644 index f824fdae0da2..000000000000 --- a/Documentation/devicetree/bindings/net/dsa/ar9331.txt +++ /dev/null @@ -1,147 +0,0 @@ -Atheros AR9331 built-in switch -============================= - -It is a switch built-in to Atheros AR9331 WiSoC and addressable over internal -MDIO bus. All PHYs are built-in as well. - -Required properties: - - - compatible: should be: "qca,ar9331-switch" - - reg: Address on the MII bus for the switch. - - resets : Must contain an entry for each entry in reset-names. - - reset-names : Must include the following entries: "switch" - - interrupt-parent: Phandle to the parent interrupt controller - - interrupts: IRQ line for the switch - - interrupt-controller: Indicates the switch is itself an interrupt - controller. This is used for the PHY interrupts. - - #interrupt-cells: must be 1 - - mdio: Container of PHY and devices on the switches MDIO bus. - -See Documentation/devicetree/bindings/net/dsa/dsa.txt for a list of additional -required and optional properties. -Examples: - -eth0: ethernet@19000000 { - compatible = "qca,ar9330-eth"; - reg = <0x19000000 0x200>; - interrupts = <4>; - - resets = <&rst 9>, <&rst 22>; - reset-names = "mac", "mdio"; - clocks = <&pll ATH79_CLK_AHB>, <&pll ATH79_CLK_AHB>; - clock-names = "eth", "mdio"; - - phy-mode = "mii"; - phy-handle = <&phy_port4>; -}; - -eth1: ethernet@1a000000 { - compatible = "qca,ar9330-eth"; - reg = <0x1a000000 0x200>; - interrupts = <5>; - resets = <&rst 13>, <&rst 23>; - reset-names = "mac", "mdio"; - clocks = <&pll ATH79_CLK_AHB>, <&pll ATH79_CLK_AHB>; - clock-names = "eth", "mdio"; - - phy-mode = "gmii"; - - fixed-link { - speed = <1000>; - full-duplex; - }; - - mdio { - #address-cells = <1>; - #size-cells = <0>; - - switch10: switch@10 { - #address-cells = <1>; - #size-cells = <0>; - - compatible = "qca,ar9331-switch"; - reg = <0x10>; - resets = <&rst 8>; - reset-names = "switch"; - - interrupt-parent = <&miscintc>; - interrupts = <12>; - - interrupt-controller; - #interrupt-cells = <1>; - - ports { - #address-cells = <1>; - #size-cells = <0>; - - switch_port0: port@0 { - reg = <0x0>; - ethernet = <ð1>; - - phy-mode = "gmii"; - - fixed-link { - speed = <1000>; - full-duplex; - }; - }; - - switch_port1: port@1 { - reg = <0x1>; - phy-handle = <&phy_port0>; - phy-mode = "internal"; - }; - - switch_port2: port@2 { - reg = <0x2>; - phy-handle = <&phy_port1>; - phy-mode = "internal"; - }; - - switch_port3: port@3 { - reg = <0x3>; - phy-handle = <&phy_port2>; - phy-mode = "internal"; - }; - - switch_port4: port@4 { - reg = <0x4>; - phy-handle = <&phy_port3>; - phy-mode = "internal"; - }; - }; - - mdio { - #address-cells = <1>; - #size-cells = <0>; - - interrupt-parent = <&switch10>; - - phy_port0: phy@0 { - reg = <0x0>; - interrupts = <0>; - }; - - phy_port1: phy@1 { - reg = <0x1>; - interrupts = <0>; - }; - - phy_port2: phy@2 { - reg = <0x2>; - interrupts = <0>; - }; - - phy_port3: phy@3 { - reg = <0x3>; - interrupts = <0>; - }; - - phy_port4: phy@4 { - reg = <0x4>; - interrupts = <0>; - }; - }; - }; - }; -}; diff --git a/Documentation/devicetree/bindings/net/dsa/qca,ar9331.yaml b/Documentation/devicetree/bindings/net/dsa/qca,ar9331.yaml new file mode 100644 index 000000000000..fd9ddc59d38c --- /dev/null +++ b/Documentation/devicetree/bindings/net/dsa/qca,ar9331.yaml @@ -0,0 +1,161 @@ +# SPDX-License-Identifier: GPL-2.0-only OR BSD-2-Clause +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/net/dsa/qca,ar9331.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Qualcomm Atheros AR9331 built-in switch + +maintainers: + - Oleksij Rempel <o.rempel@pengutronix.de> + +description: + Qualcomm Atheros AR9331 is a switch built-in to Atheros AR9331 WiSoC and + addressable over internal MDIO bus. All PHYs are built-in as well. + +properties: + compatible: + const: qca,ar9331-switch + + reg: + maxItems: 1 + + interrupts: + maxItems: 1 + + interrupt-controller: true + + '#interrupt-cells': + const: 1 + + mdio: + $ref: /schemas/net/mdio.yaml# + unevaluatedProperties: false + properties: + interrupt-parent: true + + patternProperties: + '(ethernet-)?phy@[0-4]+$': + type: object + unevaluatedProperties: false + + properties: + reg: true + interrupts: + maxItems: 1 + + resets: + maxItems: 1 + + reset-names: + items: + - const: switch + +required: + - compatible + - reg + - interrupts + - interrupt-controller + - '#interrupt-cells' + - mdio + - ports + - resets + - reset-names + +allOf: + - $ref: dsa.yaml#/$defs/ethernet-ports + +unevaluatedProperties: false + +examples: + - | + mdio { + #address-cells = <1>; + #size-cells = <0>; + + switch10: switch@10 { + compatible = "qca,ar9331-switch"; + reg = <0x10>; + + interrupt-parent = <&miscintc>; + interrupts = <12>; + interrupt-controller; + #interrupt-cells = <1>; + + resets = <&rst 8>; + reset-names = "switch"; + + ports { + #address-cells = <1>; + #size-cells = <0>; + + port@0 { + reg = <0x0>; + ethernet = <ð1>; + + phy-mode = "gmii"; + + fixed-link { + speed = <1000>; + full-duplex; + }; + }; + + port@1 { + reg = <0x1>; + phy-handle = <&phy_port0>; + phy-mode = "internal"; + }; + + port@2 { + reg = <0x2>; + phy-handle = <&phy_port1>; + phy-mode = "internal"; + }; + + port@3 { + reg = <0x3>; + phy-handle = <&phy_port2>; + phy-mode = "internal"; + }; + + port@4 { + reg = <0x4>; + phy-handle = <&phy_port3>; + phy-mode = "internal"; + }; + }; + + mdio { + #address-cells = <1>; + #size-cells = <0>; + + interrupt-parent = <&switch10>; + + phy_port0: ethernet-phy@0 { + reg = <0x0>; + interrupts = <0>; + }; + + phy_port1: ethernet-phy@1 { + reg = <0x1>; + interrupts = <0>; + }; + + phy_port2: ethernet-phy@2 { + reg = <0x2>; + interrupts = <0>; + }; + + phy_port3: ethernet-phy@3 { + reg = <0x3>; + interrupts = <0>; + }; + + phy_port4: ethernet-phy@4 { + reg = <0x4>; + interrupts = <0>; + }; + }; + }; + }; diff --git a/Documentation/devicetree/bindings/net/qcom,ethqos.yaml b/Documentation/devicetree/bindings/net/qcom,ethqos.yaml index 7bdb412a0185..69a337c7e345 100644 --- a/Documentation/devicetree/bindings/net/qcom,ethqos.yaml +++ b/Documentation/devicetree/bindings/net/qcom,ethqos.yaml @@ -37,12 +37,14 @@ properties: items: - description: Combined signal for various interrupt events - description: The interrupt that occurs when Rx exits the LPI state + - description: The interrupt that occurs when HW safety error triggered interrupt-names: minItems: 1 items: - const: macirq - - const: eth_lpi + - enum: [eth_lpi, sfty] + - const: sfty clocks: maxItems: 4 @@ -89,8 +91,9 @@ examples: <&gcc GCC_ETH_PTP_CLK>, <&gcc GCC_ETH_RGMII_CLK>; interrupts = <GIC_SPI 56 IRQ_TYPE_LEVEL_HIGH>, - <GIC_SPI 55 IRQ_TYPE_LEVEL_HIGH>; - interrupt-names = "macirq", "eth_lpi"; + <GIC_SPI 55 IRQ_TYPE_LEVEL_HIGH>, + <GIC_SPI 782 IRQ_TYPE_LEVEL_HIGH>; + interrupt-names = "macirq", "eth_lpi", "sfty"; rx-fifo-depth = <4096>; tx-fifo-depth = <4096>; diff --git a/Documentation/devicetree/bindings/net/snps,dwmac.yaml b/Documentation/devicetree/bindings/net/snps,dwmac.yaml index 90c4db178c67..6b0341a8e0ea 100644 --- a/Documentation/devicetree/bindings/net/snps,dwmac.yaml +++ b/Documentation/devicetree/bindings/net/snps,dwmac.yaml @@ -108,13 +108,15 @@ properties: - description: Combined signal for various interrupt events - description: The interrupt to manage the remote wake-up packet detection - description: The interrupt that occurs when Rx exits the LPI state + - description: The interrupt that occurs when HW safety error triggered interrupt-names: minItems: 1 items: - const: macirq - - enum: [eth_wake_irq, eth_lpi] - - const: eth_lpi + - enum: [eth_wake_irq, eth_lpi, sfty] + - enum: [eth_wake_irq, eth_lpi, sfty] + - enum: [eth_wake_irq, eth_lpi, sfty] clocks: minItems: 1 diff --git a/Documentation/networking/device_drivers/ethernet/index.rst b/Documentation/networking/device_drivers/ethernet/index.rst index 43de285b8a92..6932d8c043c2 100644 --- a/Documentation/networking/device_drivers/ethernet/index.rst +++ b/Documentation/networking/device_drivers/ethernet/index.rst @@ -42,6 +42,7 @@ Contents: intel/ice marvell/octeontx2 marvell/octeon_ep + marvell/octeon_ep_vf mellanox/mlx5/index microsoft/netvsc neterion/s2io diff --git a/Documentation/networking/device_drivers/ethernet/marvell/octeon_ep_vf.rst b/Documentation/networking/device_drivers/ethernet/marvell/octeon_ep_vf.rst new file mode 100644 index 000000000000..603133d0b92f --- /dev/null +++ b/Documentation/networking/device_drivers/ethernet/marvell/octeon_ep_vf.rst @@ -0,0 +1,24 @@ +.. SPDX-License-Identifier: GPL-2.0+ + +======================================================================= +Linux kernel networking driver for Marvell's Octeon PCI Endpoint NIC VF +======================================================================= + +Network driver for Marvell's Octeon PCI EndPoint NIC VF. +Copyright (c) 2020 Marvell International Ltd. + +Overview +======== +This driver implements networking functionality of Marvell's Octeon PCI +EndPoint NIC VF. + +Supported Devices +================= +Currently, this driver support following devices: + * Network controller: Cavium, Inc. Device b203 + * Network controller: Cavium, Inc. Device b403 + * Network controller: Cavium, Inc. Device b103 + * Network controller: Cavium, Inc. Device b903 + * Network controller: Cavium, Inc. Device ba03 + * Network controller: Cavium, Inc. Device bc03 + * Network controller: Cavium, Inc. Device bd03 diff --git a/MAINTAINERS b/MAINTAINERS index 1da02866febe..2b775f4369e0 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -13066,6 +13066,15 @@ L: netdev@vger.kernel.org S: Supported F: drivers/net/ethernet/marvell/octeon_ep +MARVELL OCTEON ENDPOINT VF DRIVER +M: Veerasenareddy Burru <vburru@marvell.com> +M: Sathesh Edara <sedara@marvell.com> +M: Shinas Rasheed <srasheed@marvell.com> +M: Satananda Burla <sburla@marvell.com> +L: netdev@vger.kernel.org +S: Supported +F: drivers/net/ethernet/marvell/octeon_ep_vf + MARVELL OCTEONTX2 PHYSICAL FUNCTION DRIVER M: Sunil Goutham <sgoutham@marvell.com> M: Geetha sowjanya <gakula@marvell.com> diff --git a/drivers/net/dsa/realtek/Kconfig b/drivers/net/dsa/realtek/Kconfig index 060165a85fb7..6989972eebc3 100644 --- a/drivers/net/dsa/realtek/Kconfig +++ b/drivers/net/dsa/realtek/Kconfig @@ -16,37 +16,29 @@ menuconfig NET_DSA_REALTEK if NET_DSA_REALTEK config NET_DSA_REALTEK_MDIO - tristate "Realtek MDIO interface driver" + bool "Realtek MDIO interface support" depends on OF - depends on NET_DSA_REALTEK_RTL8365MB || NET_DSA_REALTEK_RTL8366RB - depends on NET_DSA_REALTEK_RTL8365MB || !NET_DSA_REALTEK_RTL8365MB - depends on NET_DSA_REALTEK_RTL8366RB || !NET_DSA_REALTEK_RTL8366RB help Select to enable support for registering switches configured through MDIO. config NET_DSA_REALTEK_SMI - tristate "Realtek SMI interface driver" + bool "Realtek SMI interface support" depends on OF - depends on NET_DSA_REALTEK_RTL8365MB || NET_DSA_REALTEK_RTL8366RB - depends on NET_DSA_REALTEK_RTL8365MB || !NET_DSA_REALTEK_RTL8365MB - depends on NET_DSA_REALTEK_RTL8366RB || !NET_DSA_REALTEK_RTL8366RB help Select to enable support for registering switches connected through SMI. config NET_DSA_REALTEK_RTL8365MB - tristate "Realtek RTL8365MB switch subdriver" - imply NET_DSA_REALTEK_SMI - imply NET_DSA_REALTEK_MDIO + tristate "Realtek RTL8365MB switch driver" + depends on NET_DSA_REALTEK_SMI || NET_DSA_REALTEK_MDIO select NET_DSA_TAG_RTL8_4 help Select to enable support for Realtek RTL8365MB-VC and RTL8367S. config NET_DSA_REALTEK_RTL8366RB - tristate "Realtek RTL8366RB switch subdriver" - imply NET_DSA_REALTEK_SMI - imply NET_DSA_REALTEK_MDIO + tristate "Realtek RTL8366RB switch driver" + depends on NET_DSA_REALTEK_SMI || NET_DSA_REALTEK_MDIO select NET_DSA_TAG_RTL4_A help Select to enable support for Realtek RTL8366RB. diff --git a/drivers/net/dsa/realtek/Makefile b/drivers/net/dsa/realtek/Makefile index 0aab57252a7c..35491dc20d6d 100644 --- a/drivers/net/dsa/realtek/Makefile +++ b/drivers/net/dsa/realtek/Makefile @@ -1,6 +1,15 @@ # SPDX-License-Identifier: GPL-2.0 -obj-$(CONFIG_NET_DSA_REALTEK_MDIO) += realtek-mdio.o -obj-$(CONFIG_NET_DSA_REALTEK_SMI) += realtek-smi.o +obj-$(CONFIG_NET_DSA_REALTEK) += realtek_dsa.o +realtek_dsa-objs := rtl83xx.o + +ifdef CONFIG_NET_DSA_REALTEK_MDIO +realtek_dsa-objs += realtek-mdio.o +endif + +ifdef CONFIG_NET_DSA_REALTEK_SMI +realtek_dsa-objs += realtek-smi.o +endif + obj-$(CONFIG_NET_DSA_REALTEK_RTL8366RB) += rtl8366.o rtl8366-objs := rtl8366-core.o rtl8366rb.o obj-$(CONFIG_NET_DSA_REALTEK_RTL8365MB) += rtl8365mb.o diff --git a/drivers/net/dsa/realtek/realtek-mdio.c b/drivers/net/dsa/realtek/realtek-mdio.c index 292e6d087e8b..04b758e5a680 100644 --- a/drivers/net/dsa/realtek/realtek-mdio.c +++ b/drivers/net/dsa/realtek/realtek-mdio.c @@ -25,6 +25,8 @@ #include <linux/regmap.h> #include "realtek.h" +#include "realtek-mdio.h" +#include "rtl83xx.h" /* Read/write via mdiobus */ #define REALTEK_MDIO_CTRL0_REG 31 @@ -99,192 +101,87 @@ out_unlock: return ret; } -static void realtek_mdio_lock(void *ctx) -{ - struct realtek_priv *priv = ctx; - - mutex_lock(&priv->map_lock); -} - -static void realtek_mdio_unlock(void *ctx) -{ - struct realtek_priv *priv = ctx; - - mutex_unlock(&priv->map_lock); -} - -static const struct regmap_config realtek_mdio_regmap_config = { - .reg_bits = 10, /* A4..A0 R4..R0 */ - .val_bits = 16, - .reg_stride = 1, - /* PHY regs are at 0x8000 */ - .max_register = 0xffff, - .reg_format_endian = REGMAP_ENDIAN_BIG, +static const struct realtek_interface_info realtek_mdio_info = { .reg_read = realtek_mdio_read, .reg_write = realtek_mdio_write, - .cache_type = REGCACHE_NONE, - .lock = realtek_mdio_lock, - .unlock = realtek_mdio_unlock, }; -static const struct regmap_config realtek_mdio_nolock_regmap_config = { - .reg_bits = 10, /* A4..A0 R4..R0 */ - .val_bits = 16, - .reg_stride = 1, - /* PHY regs are at 0x8000 */ - .max_register = 0xffff, - .reg_format_endian = REGMAP_ENDIAN_BIG, - .reg_read = realtek_mdio_read, - .reg_write = realtek_mdio_write, - .cache_type = REGCACHE_NONE, - .disable_locking = true, -}; - -static int realtek_mdio_probe(struct mdio_device *mdiodev) +/** + * realtek_mdio_probe() - Probe a platform device for an MDIO-connected switch + * @mdiodev: mdio_device to probe on. + * + * This function should be used as the .probe in an mdio_driver. After + * calling the common probe function for both interfaces, it initializes the + * values specific for MDIO-connected devices. Finally, it calls a common + * function to register the DSA switch. + * + * Context: Can sleep. Takes and releases priv->map_lock. + * Return: Returns 0 on success, a negative error on failure. + */ +int realtek_mdio_probe(struct mdio_device *mdiodev) { - struct realtek_priv *priv; struct device *dev = &mdiodev->dev; - const struct realtek_variant *var; - struct regmap_config rc; - struct device_node *np; + struct realtek_priv *priv; int ret; - var = of_device_get_match_data(dev); - if (!var) - return -EINVAL; - - priv = devm_kzalloc(&mdiodev->dev, - size_add(sizeof(*priv), var->chip_data_sz), - GFP_KERNEL); - if (!priv) - return -ENOMEM; - - mutex_init(&priv->map_lock); + priv = rtl83xx_probe(dev, &realtek_mdio_info); + if (IS_ERR(priv)) + return PTR_ERR(priv); - rc = realtek_mdio_regmap_config; - rc.lock_arg = priv; - priv->map = devm_regmap_init(dev, NULL, priv, &rc); - if (IS_ERR(priv->map)) { - ret = PTR_ERR(priv->map); - dev_err(dev, "regmap init failed: %d\n", ret); - return ret; - } - - rc = realtek_mdio_nolock_regmap_config; - priv->map_nolock = devm_regmap_init(dev, NULL, priv, &rc); - if (IS_ERR(priv->map_nolock)) { - ret = PTR_ERR(priv->map_nolock); - dev_err(dev, "regmap init failed: %d\n", ret); - return ret; - } - - priv->mdio_addr = mdiodev->addr; priv->bus = mdiodev->bus; - priv->dev = &mdiodev->dev; - priv->chip_data = (void *)priv + sizeof(*priv); - - priv->clk_delay = var->clk_delay; - priv->cmd_read = var->cmd_read; - priv->cmd_write = var->cmd_write; - priv->ops = var->ops; - + priv->mdio_addr = mdiodev->addr; priv->write_reg_noack = realtek_mdio_write; - np = dev->of_node; - - dev_set_drvdata(dev, priv); - - /* TODO: if power is software controlled, set up any regulators here */ - priv->leds_disabled = of_property_read_bool(np, "realtek,disable-leds"); - - priv->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW); - if (IS_ERR(priv->reset)) { - dev_err(dev, "failed to get RESET GPIO\n"); - return PTR_ERR(priv->reset); - } - - if (priv->reset) { - gpiod_set_value(priv->reset, 1); - dev_dbg(dev, "asserted RESET\n"); - msleep(REALTEK_HW_STOP_DELAY); - gpiod_set_value(priv->reset, 0); - msleep(REALTEK_HW_START_DELAY); - dev_dbg(dev, "deasserted RESET\n"); - } - - ret = priv->ops->detect(priv); - if (ret) { - dev_err(dev, "unable to detect switch\n"); - return ret; - } - - priv->ds = devm_kzalloc(dev, sizeof(*priv->ds), GFP_KERNEL); - if (!priv->ds) - return -ENOMEM; - - priv->ds->dev = dev; - priv->ds->num_ports = priv->num_ports; - priv->ds->priv = priv; - priv->ds->ops = var->ds_ops_mdio; - - ret = dsa_register_switch(priv->ds); + ret = rtl83xx_register_switch(priv); if (ret) { - dev_err(priv->dev, "unable to register switch ret = %d\n", ret); + rtl83xx_remove(priv); return ret; } return 0; } +EXPORT_SYMBOL_NS_GPL(realtek_mdio_probe, REALTEK_DSA); -static void realtek_mdio_remove(struct mdio_device *mdiodev) +/** + * realtek_mdio_remove() - Remove the driver of an MDIO-connected switch + * @mdiodev: mdio_device to be removed. + * + * This function should be used as the .remove_new in an mdio_driver. First + * it unregisters the DSA switch and then it calls the common remove function. + * + * Context: Can sleep. + * Return: Nothing. + */ +void realtek_mdio_remove(struct mdio_device *mdiodev) { struct realtek_priv *priv = dev_get_drvdata(&mdiodev->dev); if (!priv) return; - dsa_unregister_switch(priv->ds); + rtl83xx_unregister_switch(priv); - /* leave the device reset asserted */ - if (priv->reset) - gpiod_set_value(priv->reset, 1); + rtl83xx_remove(priv); } +EXPORT_SYMBOL_NS_GPL(realtek_mdio_remove, REALTEK_DSA); -static void realtek_mdio_shutdown(struct mdio_device *mdiodev) +/** + * realtek_mdio_shutdown() - Shutdown the driver of a MDIO-connected switch + * @mdiodev: mdio_device shutting down. + * + * This function should be used as the .shutdown in a platform_driver. It calls + * the common shutdown function. + * + * Context: Can sleep. + * Return: Nothing. + */ +void realtek_mdio_shutdown(struct mdio_device *mdiodev) { struct realtek_priv *priv = dev_get_drvdata(&mdiodev->dev); if (!priv) return; - dsa_switch_shutdown(priv->ds); - - dev_set_drvdata(&mdiodev->dev, NULL); + rtl83xx_shutdown(priv); } - -static const struct of_device_id realtek_mdio_of_match[] = { -#if IS_ENABLED(CONFIG_NET_DSA_REALTEK_RTL8366RB) - { .compatible = "realtek,rtl8366rb", .data = &rtl8366rb_variant, }, -#endif -#if IS_ENABLED(CONFIG_NET_DSA_REALTEK_RTL8365MB) - { .compatible = "realtek,rtl8365mb", .data = &rtl8365mb_variant, }, -#endif - { /* sentinel */ }, -}; -MODULE_DEVICE_TABLE(of, realtek_mdio_of_match); - -static struct mdio_driver realtek_mdio_driver = { - .mdiodrv.driver = { - .name = "realtek-mdio", - .of_match_table = realtek_mdio_of_match, - }, - .probe = realtek_mdio_probe, - .remove = realtek_mdio_remove, - .shutdown = realtek_mdio_shutdown, -}; - -mdio_module_driver(realtek_mdio_driver); - -MODULE_AUTHOR("Luiz Angelo Daros de Luca <luizluca@gmail.com>"); -MODULE_DESCRIPTION("Driver for Realtek ethernet switch connected via MDIO interface"); -MODULE_LICENSE("GPL"); +EXPORT_SYMBOL_NS_GPL(realtek_mdio_shutdown, REALTEK_DSA); diff --git a/drivers/net/dsa/realtek/realtek-mdio.h b/drivers/net/dsa/realtek/realtek-mdio.h new file mode 100644 index 000000000000..ee70f6a5b8ff --- /dev/null +++ b/drivers/net/dsa/realtek/realtek-mdio.h @@ -0,0 +1,48 @@ +/* SPDX-License-Identifier: GPL-2.0+ */ + +#ifndef _REALTEK_MDIO_H +#define _REALTEK_MDIO_H + +#if IS_ENABLED(CONFIG_NET_DSA_REALTEK_MDIO) + +static inline int realtek_mdio_driver_register(struct mdio_driver *drv) +{ + return mdio_driver_register(drv); +} + +static inline void realtek_mdio_driver_unregister(struct mdio_driver *drv) +{ + mdio_driver_unregister(drv); +} + +int realtek_mdio_probe(struct mdio_device *mdiodev); +void realtek_mdio_remove(struct mdio_device *mdiodev); +void realtek_mdio_shutdown(struct mdio_device *mdiodev); + +#else /* IS_ENABLED(CONFIG_NET_DSA_REALTEK_MDIO) */ + +static inline int realtek_mdio_driver_register(struct mdio_driver *drv) +{ + return 0; +} + +static inline void realtek_mdio_driver_unregister(struct mdio_driver *drv) +{ +} + +static inline int realtek_mdio_probe(struct mdio_device *mdiodev) +{ + return -ENOENT; +} + +static inline void realtek_mdio_remove(struct mdio_device *mdiodev) +{ +} + +static inline void realtek_mdio_shutdown(struct mdio_device *mdiodev) +{ +} + +#endif /* IS_ENABLED(CONFIG_NET_DSA_REALTEK_MDIO) */ + +#endif /* _REALTEK_MDIO_H */ diff --git a/drivers/net/dsa/realtek/realtek-smi.c b/drivers/net/dsa/realtek/realtek-smi.c index 755546ed8db6..88590ae95a75 100644 --- a/drivers/net/dsa/realtek/realtek-smi.c +++ b/drivers/net/dsa/realtek/realtek-smi.c @@ -31,7 +31,6 @@ #include <linux/spinlock.h> #include <linux/skbuff.h> #include <linux/of.h> -#include <linux/of_mdio.h> #include <linux/delay.h> #include <linux/gpio/consumer.h> #include <linux/platform_device.h> @@ -40,12 +39,14 @@ #include <linux/if_bridge.h> #include "realtek.h" +#include "realtek-smi.h" +#include "rtl83xx.h" #define REALTEK_SMI_ACK_RETRY_COUNT 5 static inline void realtek_smi_clk_delay(struct realtek_priv *priv) { - ndelay(priv->clk_delay); + ndelay(priv->variant->clk_delay); } static void realtek_smi_start(struct realtek_priv *priv) @@ -208,7 +209,7 @@ static int realtek_smi_read_reg(struct realtek_priv *priv, u32 addr, u32 *data) realtek_smi_start(priv); /* Send READ command */ - ret = realtek_smi_write_byte(priv, priv->cmd_read); + ret = realtek_smi_write_byte(priv, priv->variant->cmd_read); if (ret) goto out; @@ -249,7 +250,7 @@ static int realtek_smi_write_reg(struct realtek_priv *priv, realtek_smi_start(priv); /* Send WRITE command */ - ret = realtek_smi_write_byte(priv, priv->cmd_write); + ret = realtek_smi_write_byte(priv, priv->variant->cmd_write); if (ret) goto out; @@ -310,258 +311,98 @@ static int realtek_smi_read(void *ctx, u32 reg, u32 *val) return realtek_smi_read_reg(priv, reg, val); } -static void realtek_smi_lock(void *ctx) -{ - struct realtek_priv *priv = ctx; - - mutex_lock(&priv->map_lock); -} - -static void realtek_smi_unlock(void *ctx) -{ - struct realtek_priv *priv = ctx; - - mutex_unlock(&priv->map_lock); -} - -static const struct regmap_config realtek_smi_regmap_config = { - .reg_bits = 10, /* A4..A0 R4..R0 */ - .val_bits = 16, - .reg_stride = 1, - /* PHY regs are at 0x8000 */ - .max_register = 0xffff, - .reg_format_endian = REGMAP_ENDIAN_BIG, +static const struct realtek_interface_info realtek_smi_info = { .reg_read = realtek_smi_read, .reg_write = realtek_smi_write, - .cache_type = REGCACHE_NONE, - .lock = realtek_smi_lock, - .unlock = realtek_smi_unlock, }; -static const struct regmap_config realtek_smi_nolock_regmap_config = { - .reg_bits = 10, /* A4..A0 R4..R0 */ - .val_bits = 16, - .reg_stride = 1, - /* PHY regs are at 0x8000 */ - .max_register = 0xffff, - .reg_format_endian = REGMAP_ENDIAN_BIG, - .reg_read = realtek_smi_read, - .reg_write = realtek_smi_write, - .cache_type = REGCACHE_NONE, - .disable_locking = true, -}; - -static int realtek_smi_mdio_read(struct mii_bus *bus, int addr, int regnum) -{ - struct realtek_priv *priv = bus->priv; - - return priv->ops->phy_read(priv, addr, regnum); -} - -static int realtek_smi_mdio_write(struct mii_bus *bus, int addr, int regnum, - u16 val) -{ - struct realtek_priv *priv = bus->priv; - - return priv->ops->phy_write(priv, addr, regnum, val); -} - -static int realtek_smi_setup_mdio(struct dsa_switch *ds) -{ - struct realtek_priv *priv = ds->priv; - struct device_node *mdio_np; - int ret; - - mdio_np = of_get_compatible_child(priv->dev->of_node, "realtek,smi-mdio"); - if (!mdio_np) { - dev_err(priv->dev, "no MDIO bus node\n"); - return -ENODEV; - } - - priv->user_mii_bus = devm_mdiobus_alloc(priv->dev); - if (!priv->user_mii_bus) { - ret = -ENOMEM; - goto err_put_node; - } - priv->user_mii_bus->priv = priv; - priv->user_mii_bus->name = "SMI user MII"; - priv->user_mii_bus->read = realtek_smi_mdio_read; - priv->user_mii_bus->write = realtek_smi_mdio_write; - snprintf(priv->user_mii_bus->id, MII_BUS_ID_SIZE, "SMI-%d", - ds->index); - priv->user_mii_bus->dev.of_node = mdio_np; - priv->user_mii_bus->parent = priv->dev; - ds->user_mii_bus = priv->user_mii_bus; - - ret = devm_of_mdiobus_register(priv->dev, priv->user_mii_bus, mdio_np); - if (ret) { - dev_err(priv->dev, "unable to register MDIO bus %s\n", - priv->user_mii_bus->id); - goto err_put_node; - } - - return 0; - -err_put_node: - of_node_put(mdio_np); - - return ret; -} - -static int realtek_smi_probe(struct platform_device *pdev) +/** + * realtek_smi_probe() - Probe a platform device for an SMI-connected switch + * @pdev: platform_device to probe on. + * + * This function should be used as the .probe in a platform_driver. After + * calling the common probe function for both interfaces, it initializes the + * values specific for SMI-connected devices. Finally, it calls a common + * function to register the DSA switch. + * + * Context: Can sleep. Takes and releases priv->map_lock. + * Return: Returns 0 on success, a negative error on failure. + */ +int realtek_smi_probe(struct platform_device *pdev) { - const struct realtek_variant *var; struct device *dev = &pdev->dev; struct realtek_priv *priv; - struct regmap_config rc; - struct device_node *np; int ret; - var = of_device_get_match_data(dev); - np = dev->of_node; - - priv = devm_kzalloc(dev, sizeof(*priv) + var->chip_data_sz, GFP_KERNEL); - if (!priv) - return -ENOMEM; - priv->chip_data = (void *)priv + sizeof(*priv); - - mutex_init(&priv->map_lock); - - rc = realtek_smi_regmap_config; - rc.lock_arg = priv; - priv->map = devm_regmap_init(dev, NULL, priv, &rc); - if (IS_ERR(priv->map)) { - ret = PTR_ERR(priv->map); - dev_err(dev, "regmap init failed: %d\n", ret); - return ret; - } - - rc = realtek_smi_nolock_regmap_config; - priv->map_nolock = devm_regmap_init(dev, NULL, priv, &rc); - if (IS_ERR(priv->map_nolock)) { - ret = PTR_ERR(priv->map_nolock); - dev_err(dev, "regmap init failed: %d\n", ret); - return ret; - } - - /* Link forward and backward */ - priv->dev = dev; - priv->clk_delay = var->clk_delay; - priv->cmd_read = var->cmd_read; - priv->cmd_write = var->cmd_write; - priv->ops = var->ops; - - priv->setup_interface = realtek_smi_setup_mdio; - priv->write_reg_noack = realtek_smi_write_reg_noack; - - dev_set_drvdata(dev, priv); - spin_lock_init(&priv->lock); - - /* TODO: if power is software controlled, set up any regulators here */ - - priv->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW); - if (IS_ERR(priv->reset)) { - dev_err(dev, "failed to get RESET GPIO\n"); - return PTR_ERR(priv->reset); - } - if (priv->reset) { - gpiod_set_value(priv->reset, 1); - dev_dbg(dev, "asserted RESET\n"); - msleep(REALTEK_HW_STOP_DELAY); - gpiod_set_value(priv->reset, 0); - msleep(REALTEK_HW_START_DELAY); - dev_dbg(dev, "deasserted RESET\n"); - } + priv = rtl83xx_probe(dev, &realtek_smi_info); + if (IS_ERR(priv)) + return PTR_ERR(priv); /* Fetch MDIO pins */ priv->mdc = devm_gpiod_get_optional(dev, "mdc", GPIOD_OUT_LOW); - if (IS_ERR(priv->mdc)) + if (IS_ERR(priv->mdc)) { + rtl83xx_remove(priv); return PTR_ERR(priv->mdc); + } + priv->mdio = devm_gpiod_get_optional(dev, "mdio", GPIOD_OUT_LOW); - if (IS_ERR(priv->mdio)) + if (IS_ERR(priv->mdio)) { + rtl83xx_remove(priv); return PTR_ERR(priv->mdio); - - priv->leds_disabled = of_property_read_bool(np, "realtek,disable-leds"); - - ret = priv->ops->detect(priv); - if (ret) { - dev_err(dev, "unable to detect switch\n"); - return ret; } - priv->ds = devm_kzalloc(dev, sizeof(*priv->ds), GFP_KERNEL); - if (!priv->ds) - return -ENOMEM; - - priv->ds->dev = dev; - priv->ds->num_ports = priv->num_ports; - priv->ds->priv = priv; + priv->write_reg_noack = realtek_smi_write_reg_noack; - priv->ds->ops = var->ds_ops_smi; - ret = dsa_register_switch(priv->ds); + ret = rtl83xx_register_switch(priv); if (ret) { - dev_err_probe(dev, ret, "unable to register switch\n"); + rtl83xx_remove(priv); return ret; } + return 0; } +EXPORT_SYMBOL_NS_GPL(realtek_smi_probe, REALTEK_DSA); -static void realtek_smi_remove(struct platform_device *pdev) +/** + * realtek_smi_remove() - Remove the driver of a SMI-connected switch + * @pdev: platform_device to be removed. + * + * This function should be used as the .remove_new in a platform_driver. First + * it unregisters the DSA switch and then it calls the common remove function. + * + * Context: Can sleep. + * Return: Nothing. + */ +void realtek_smi_remove(struct platform_device *pdev) { struct realtek_priv *priv = platform_get_drvdata(pdev); if (!priv) return; - dsa_unregister_switch(priv->ds); - if (priv->user_mii_bus) - of_node_put(priv->user_mii_bus->dev.of_node); + rtl83xx_unregister_switch(priv); - /* leave the device reset asserted */ - if (priv->reset) - gpiod_set_value(priv->reset, 1); + rtl83xx_remove(priv); } +EXPORT_SYMBOL_NS_GPL(realtek_smi_remove, REALTEK_DSA); -static void realtek_smi_shutdown(struct platform_device *pdev) +/** + * realtek_smi_shutdown() - Shutdown the driver of a SMI-connected switch + * @pdev: platform_device shutting down. + * + * This function should be used as the .shutdown in a platform_driver. It calls + * the common shutdown function. + * + * Context: Can sleep. + * Return: Nothing. + */ +void realtek_smi_shutdown(struct platform_device *pdev) { struct realtek_priv *priv = platform_get_drvdata(pdev); if (!priv) return; - dsa_switch_shutdown(priv->ds); - - platform_set_drvdata(pdev, NULL); + rtl83xx_shutdown(priv); } - -static const struct of_device_id realtek_smi_of_match[] = { -#if IS_ENABLED(CONFIG_NET_DSA_REALTEK_RTL8366RB) - { - .compatible = "realtek,rtl8366rb", - .data = &rtl8366rb_variant, - }, -#endif -#if IS_ENABLED(CONFIG_NET_DSA_REALTEK_RTL8365MB) - { - .compatible = "realtek,rtl8365mb", - .data = &rtl8365mb_variant, - }, -#endif - { /* sentinel */ }, -}; -MODULE_DEVICE_TABLE(of, realtek_smi_of_match); - -static struct platform_driver realtek_smi_driver = { - .driver = { - .name = "realtek-smi", - .of_match_table = realtek_smi_of_match, - }, - .probe = realtek_smi_probe, - .remove_new = realtek_smi_remove, - .shutdown = realtek_smi_shutdown, -}; -module_platform_driver(realtek_smi_driver); - -MODULE_AUTHOR("Linus Walleij <linus.walleij@linaro.org>"); -MODULE_DESCRIPTION("Driver for Realtek ethernet switch connected via SMI interface"); -MODULE_LICENSE("GPL"); +EXPORT_SYMBOL_NS_GPL(realtek_smi_shutdown, REALTEK_DSA); diff --git a/drivers/net/dsa/realtek/realtek-smi.h b/drivers/net/dsa/realtek/realtek-smi.h new file mode 100644 index 000000000000..ea49a2edd3c8 --- /dev/null +++ b/drivers/net/dsa/realtek/realtek-smi.h @@ -0,0 +1,48 @@ +/* SPDX-License-Identifier: GPL-2.0+ */ + +#ifndef _REALTEK_SMI_H +#define _REALTEK_SMI_H + +#if IS_ENABLED(CONFIG_NET_DSA_REALTEK_SMI) + +static inline int realtek_smi_driver_register(struct platform_driver *drv) +{ + return platform_driver_register(drv); +} + +static inline void realtek_smi_driver_unregister(struct platform_driver *drv) +{ + platform_driver_unregister(drv); +} + +int realtek_smi_probe(struct platform_device *pdev); +void realtek_smi_remove(struct platform_device *pdev); +void realtek_smi_shutdown(struct platform_device *pdev); + +#else /* IS_ENABLED(CONFIG_NET_DSA_REALTEK_SMI) */ + +static inline int realtek_smi_driver_register(struct platform_driver *drv) +{ + return 0; +} + +static inline void realtek_smi_driver_unregister(struct platform_driver *drv) +{ +} + +static inline int realtek_smi_probe(struct platform_device *pdev) +{ + return -ENOENT; +} + +static inline void realtek_smi_remove(struct platform_device *pdev) +{ +} + +static inline void realtek_smi_shutdown(struct platform_device *pdev) +{ +} + +#endif /* IS_ENABLED(CONFIG_NET_DSA_REALTEK_SMI) */ + +#endif /* _REALTEK_SMI_H */ diff --git a/drivers/net/dsa/realtek/realtek.h b/drivers/net/dsa/realtek/realtek.h index 790488e9c667..b80bfde1ad04 100644 --- a/drivers/net/dsa/realtek/realtek.h +++ b/drivers/net/dsa/realtek/realtek.h @@ -58,11 +58,10 @@ struct realtek_priv { struct mii_bus *bus; int mdio_addr; - unsigned int clk_delay; - u8 cmd_read; - u8 cmd_write; + const struct realtek_variant *variant; + spinlock_t lock; /* Locks around command writes */ - struct dsa_switch *ds; + struct dsa_switch ds; struct irq_domain *irqdomain; bool leds_disabled; @@ -73,7 +72,6 @@ struct realtek_priv { struct rtl8366_mib_counter *mib_counters; const struct realtek_ops *ops; - int (*setup_interface)(struct dsa_switch *ds); int (*write_reg_noack)(void *ctx, u32 addr, u32 data); int vlan_enabled; @@ -91,7 +89,6 @@ struct realtek_ops { int (*detect)(struct realtek_priv *priv); int (*reset_chip)(struct realtek_priv *priv); int (*setup)(struct realtek_priv *priv); - void (*cleanup)(struct realtek_priv *priv); int (*get_mib_counter)(struct realtek_priv *priv, int port, struct rtl8366_mib_counter *mib, @@ -116,8 +113,7 @@ struct realtek_ops { }; struct realtek_variant { - const struct dsa_switch_ops *ds_ops_smi; - const struct dsa_switch_ops *ds_ops_mdio; + const struct dsa_switch_ops *ds_ops; const struct realtek_ops *ops; unsigned int clk_delay; u8 cmd_read; diff --git a/drivers/net/dsa/realtek/rtl8365mb.c b/drivers/net/dsa/realtek/rtl8365mb.c index b072045eb154..12665a8a3412 100644 --- a/drivers/net/dsa/realtek/rtl8365mb.c +++ b/drivers/net/dsa/realtek/rtl8365mb.c @@ -101,6 +101,9 @@ #include <linux/if_vlan.h> #include "realtek.h" +#include "realtek-smi.h" +#include "realtek-mdio.h" +#include "rtl83xx.h" /* Family-specific data and limits */ #define RTL8365MB_PHYADDRMAX 7 @@ -206,10 +209,10 @@ #define RTL8365MB_EXT_PORT_MODE_100FX 13 /* External interface mode configuration registers 0~1 */ -#define RTL8365MB_DIGITAL_INTERFACE_SELECT_REG0 0x1305 /* EXT1 */ +#define RTL8365MB_DIGITAL_INTERFACE_SELECT_REG0 0x1305 /* EXT0,EXT1 */ #define RTL8365MB_DIGITAL_INTERFACE_SELECT_REG1 0x13C3 /* EXT2 */ #define RTL8365MB_DIGITAL_INTERFACE_SELECT_REG(_extint) \ - ((_extint) == 1 ? RTL8365MB_DIGITAL_INTERFACE_SELECT_REG0 : \ + ((_extint) <= 1 ? RTL8365MB_DIGITAL_INTERFACE_SELECT_REG0 : \ (_extint) == 2 ? RTL8365MB_DIGITAL_INTERFACE_SELECT_REG1 : \ 0x0) #define RTL8365MB_DIGITAL_INTERFACE_SELECT_MODE_MASK(_extint) \ @@ -689,7 +692,7 @@ static int rtl8365mb_phy_ocp_read(struct realtek_priv *priv, int phy, u32 val; int ret; - mutex_lock(&priv->map_lock); + rtl83xx_lock(priv); ret = rtl8365mb_phy_poll_busy(priv); if (ret) @@ -722,7 +725,7 @@ static int rtl8365mb_phy_ocp_read(struct realtek_priv *priv, int phy, *data = val & 0xFFFF; out: - mutex_unlock(&priv->map_lock); + rtl83xx_unlock(priv); return ret; } @@ -733,7 +736,7 @@ static int rtl8365mb_phy_ocp_write(struct realtek_priv *priv, int phy, u32 val; int ret; - mutex_lock(&priv->map_lock); + rtl83xx_lock(priv); ret = rtl8365mb_phy_poll_busy(priv); if (ret) @@ -764,7 +767,7 @@ static int rtl8365mb_phy_ocp_write(struct realtek_priv *priv, int phy, goto out; out: - mutex_unlock(&priv->map_lock); + rtl83xx_unlock(priv); return 0; } @@ -825,17 +828,6 @@ static int rtl8365mb_phy_write(struct realtek_priv *priv, int phy, int regnum, return 0; } -static int rtl8365mb_dsa_phy_read(struct dsa_switch *ds, int phy, int regnum) -{ - return rtl8365mb_phy_read(ds->priv, phy, regnum); -} - -static int rtl8365mb_dsa_phy_write(struct dsa_switch *ds, int phy, int regnum, - u16 val) -{ - return rtl8365mb_phy_write(ds->priv, phy, regnum, val); -} - static const struct rtl8365mb_extint * rtl8365mb_get_port_extint(struct realtek_priv *priv, int port) { @@ -878,6 +870,7 @@ static int rtl8365mb_ext_config_rgmii(struct realtek_priv *priv, int port, { const struct rtl8365mb_extint *extint = rtl8365mb_get_port_extint(priv, port); + struct dsa_switch *ds = &priv->ds; struct device_node *dn; struct dsa_port *dp; int tx_delay = 0; @@ -888,7 +881,7 @@ static int rtl8365mb_ext_config_rgmii(struct realtek_priv *priv, int port, if (!extint) return -ENODEV; - dp = dsa_to_port(priv->ds, port); + dp = dsa_to_port(ds, port); dn = dp->dn; /* Set the RGMII TX/RX delay @@ -1541,6 +1534,7 @@ static void rtl8365mb_get_stats64(struct dsa_switch *ds, int port, static void rtl8365mb_stats_setup(struct realtek_priv *priv) { struct rtl8365mb *mb = priv->chip_data; + struct dsa_switch *ds = &priv->ds; int i; /* Per-chip global mutex to protect MIB counter access, since doing @@ -1551,7 +1545,7 @@ static void rtl8365mb_stats_setup(struct realtek_priv *priv) for (i = 0; i < priv->num_ports; i++) { struct rtl8365mb_port *p = &mb->ports[i]; - if (dsa_is_unused_port(priv->ds, i)) + if (dsa_is_unused_port(ds, i)) continue; /* Per-port spinlock to protect the stats64 data */ @@ -1567,12 +1561,13 @@ static void rtl8365mb_stats_setup(struct realtek_priv *priv) static void rtl8365mb_stats_teardown(struct realtek_priv *priv) { struct rtl8365mb *mb = priv->chip_data; + struct dsa_switch *ds = &priv->ds; int i; for (i = 0; i < priv->num_ports; i++) { struct rtl8365mb_port *p = &mb->ports[i]; - if (dsa_is_unused_port(priv->ds, i)) + if (dsa_is_unused_port(ds, i)) continue; cancel_delayed_work_sync(&p->mib_work); @@ -1971,7 +1966,7 @@ static int rtl8365mb_setup(struct dsa_switch *ds) dev_info(priv->dev, "no interrupt support\n"); /* Configure CPU tagging */ - dsa_switch_for_each_cpu_port(cpu_dp, priv->ds) { + dsa_switch_for_each_cpu_port(cpu_dp, ds) { cpu->mask |= BIT(cpu_dp->index); if (cpu->trap_port == RTL8365MB_MAX_NUM_PORTS) @@ -1986,7 +1981,7 @@ static int rtl8365mb_setup(struct dsa_switch *ds) for (i = 0; i < priv->num_ports; i++) { struct rtl8365mb_port *p = &mb->ports[i]; - if (dsa_is_unused_port(priv->ds, i)) + if (dsa_is_unused_port(ds, i)) continue; /* Forward only to the CPU */ @@ -2003,7 +1998,7 @@ static int rtl8365mb_setup(struct dsa_switch *ds) * ports will still forward frames to the CPU despite being * administratively down by default. */ - rtl8365mb_port_stp_state_set(priv->ds, i, BR_STATE_DISABLED); + rtl8365mb_port_stp_state_set(ds, i, BR_STATE_DISABLED); /* Set up per-port private data */ p->priv = priv; @@ -2014,12 +2009,10 @@ static int rtl8365mb_setup(struct dsa_switch *ds) if (ret) goto out_teardown_irq; - if (priv->setup_interface) { - ret = priv->setup_interface(ds); - if (ret) { - dev_err(priv->dev, "could not set up MDIO bus\n"); - goto out_teardown_irq; - } + ret = rtl83xx_setup_user_mdio(ds); + if (ret) { + dev_err(priv->dev, "could not set up MDIO bus\n"); + goto out_teardown_irq; } /* Start statistics counter polling */ @@ -2113,7 +2106,7 @@ static int rtl8365mb_detect(struct realtek_priv *priv) return 0; } -static const struct dsa_switch_ops rtl8365mb_switch_ops_smi = { +static const struct dsa_switch_ops rtl8365mb_switch_ops = { .get_tag_protocol = rtl8365mb_get_tag_protocol, .change_tag_protocol = rtl8365mb_change_tag_protocol, .setup = rtl8365mb_setup, @@ -2134,29 +2127,6 @@ static const struct dsa_switch_ops rtl8365mb_switch_ops_smi = { .port_max_mtu = rtl8365mb_port_max_mtu, }; -static const struct dsa_switch_ops rtl8365mb_switch_ops_mdio = { - .get_tag_protocol = rtl8365mb_get_tag_protocol, - .change_tag_protocol = rtl8365mb_change_tag_protocol, - .setup = rtl8365mb_setup, - .teardown = rtl8365mb_teardown, - .phylink_get_caps = rtl8365mb_phylink_get_caps, - .phylink_mac_config = rtl8365mb_phylink_mac_config, - .phylink_mac_link_down = rtl8365mb_phylink_mac_link_down, - .phylink_mac_link_up = rtl8365mb_phylink_mac_link_up, - .phy_read = rtl8365mb_dsa_phy_read, - .phy_write = rtl8365mb_dsa_phy_write, - .port_stp_state_set = rtl8365mb_port_stp_state_set, - .get_strings = rtl8365mb_get_strings, - .get_ethtool_stats = rtl8365mb_get_ethtool_stats, - .get_sset_count = rtl8365mb_get_sset_count, - .get_eth_phy_stats = rtl8365mb_get_phy_stats, - .get_eth_mac_stats = rtl8365mb_get_mac_stats, - .get_eth_ctrl_stats = rtl8365mb_get_ctrl_stats, - .get_stats64 = rtl8365mb_get_stats64, - .port_change_mtu = rtl8365mb_port_change_mtu, - .port_max_mtu = rtl8365mb_port_max_mtu, -}; - static const struct realtek_ops rtl8365mb_ops = { .detect = rtl8365mb_detect, .phy_read = rtl8365mb_phy_read, @@ -2164,16 +2134,66 @@ static const struct realtek_ops rtl8365mb_ops = { }; const struct realtek_variant rtl8365mb_variant = { - .ds_ops_smi = &rtl8365mb_switch_ops_smi, - .ds_ops_mdio = &rtl8365mb_switch_ops_mdio, + .ds_ops = &rtl8365mb_switch_ops, .ops = &rtl8365mb_ops, .clk_delay = 10, .cmd_read = 0xb9, .cmd_write = 0xb8, .chip_data_sz = sizeof(struct rtl8365mb), }; -EXPORT_SYMBOL_GPL(rtl8365mb_variant); + +static const struct of_device_id rtl8365mb_of_match[] = { + { .compatible = "realtek,rtl8365mb", .data = &rtl8365mb_variant, }, + { /* sentinel */ }, +}; +MODULE_DEVICE_TABLE(of, rtl8365mb_of_match); + +static struct platform_driver rtl8365mb_smi_driver = { + .driver = { + .name = "rtl8365mb-smi", + .of_match_table = rtl8365mb_of_match, + }, + .probe = realtek_smi_probe, + .remove_new = realtek_smi_remove, + .shutdown = realtek_smi_shutdown, +}; + +static struct mdio_driver rtl8365mb_mdio_driver = { + .mdiodrv.driver = { + .name = "rtl8365mb-mdio", + .of_match_table = rtl8365mb_of_match, + }, + .probe = realtek_mdio_probe, + .remove = realtek_mdio_remove, + .shutdown = realtek_mdio_shutdown, +}; + +static int rtl8365mb_init(void) +{ + int ret; + + ret = realtek_mdio_driver_register(&rtl8365mb_mdio_driver); + if (ret) + return ret; + + ret = realtek_smi_driver_register(&rtl8365mb_smi_driver); + if (ret) { + realtek_mdio_driver_unregister(&rtl8365mb_mdio_driver); + return ret; + } + + return 0; +} +module_init(rtl8365mb_init); + +static void __exit rtl8365mb_exit(void) +{ + realtek_smi_driver_unregister(&rtl8365mb_smi_driver); + realtek_mdio_driver_unregister(&rtl8365mb_mdio_driver); +} +module_exit(rtl8365mb_exit); MODULE_AUTHOR("Alvin Å ipraga <alsi@bang-olufsen.dk>"); MODULE_DESCRIPTION("Driver for RTL8365MB-VC ethernet switch"); MODULE_LICENSE("GPL"); +MODULE_IMPORT_NS(REALTEK_DSA); diff --git a/drivers/net/dsa/realtek/rtl8366-core.c b/drivers/net/dsa/realtek/rtl8366-core.c index 59f98d2c8769..7c6520ba3a26 100644 --- a/drivers/net/dsa/realtek/rtl8366-core.c +++ b/drivers/net/dsa/realtek/rtl8366-core.c @@ -34,7 +34,7 @@ int rtl8366_mc_is_used(struct realtek_priv *priv, int mc_index, int *used) return 0; } -EXPORT_SYMBOL_GPL(rtl8366_mc_is_used); +EXPORT_SYMBOL_NS_GPL(rtl8366_mc_is_used, REALTEK_DSA); /** * rtl8366_obtain_mc() - retrieve or allocate a VLAN member configuration @@ -187,7 +187,7 @@ int rtl8366_set_vlan(struct realtek_priv *priv, int vid, u32 member, return ret; } -EXPORT_SYMBOL_GPL(rtl8366_set_vlan); +EXPORT_SYMBOL_NS_GPL(rtl8366_set_vlan, REALTEK_DSA); int rtl8366_set_pvid(struct realtek_priv *priv, unsigned int port, unsigned int vid) @@ -217,7 +217,7 @@ int rtl8366_set_pvid(struct realtek_priv *priv, unsigned int port, return 0; } -EXPORT_SYMBOL_GPL(rtl8366_set_pvid); +EXPORT_SYMBOL_NS_GPL(rtl8366_set_pvid, REALTEK_DSA); int rtl8366_enable_vlan4k(struct realtek_priv *priv, bool enable) { @@ -243,7 +243,7 @@ int rtl8366_enable_vlan4k(struct realtek_priv *priv, bool enable) priv->vlan4k_enabled = enable; return 0; } -EXPORT_SYMBOL_GPL(rtl8366_enable_vlan4k); +EXPORT_SYMBOL_NS_GPL(rtl8366_enable_vlan4k, REALTEK_DSA); int rtl8366_enable_vlan(struct realtek_priv *priv, bool enable) { @@ -265,7 +265,7 @@ int rtl8366_enable_vlan(struct realtek_priv *priv, bool enable) return ret; } -EXPORT_SYMBOL_GPL(rtl8366_enable_vlan); +EXPORT_SYMBOL_NS_GPL(rtl8366_enable_vlan, REALTEK_DSA); int rtl8366_reset_vlan(struct realtek_priv *priv) { @@ -290,7 +290,7 @@ int rtl8366_reset_vlan(struct realtek_priv *priv) return 0; } -EXPORT_SYMBOL_GPL(rtl8366_reset_vlan); +EXPORT_SYMBOL_NS_GPL(rtl8366_reset_vlan, REALTEK_DSA); int rtl8366_vlan_add(struct dsa_switch *ds, int port, const struct switchdev_obj_port_vlan *vlan, @@ -345,7 +345,7 @@ int rtl8366_vlan_add(struct dsa_switch *ds, int port, return 0; } -EXPORT_SYMBOL_GPL(rtl8366_vlan_add); +EXPORT_SYMBOL_NS_GPL(rtl8366_vlan_add, REALTEK_DSA); int rtl8366_vlan_del(struct dsa_switch *ds, int port, const struct switchdev_obj_port_vlan *vlan) @@ -389,7 +389,7 @@ int rtl8366_vlan_del(struct dsa_switch *ds, int port, return 0; } -EXPORT_SYMBOL_GPL(rtl8366_vlan_del); +EXPORT_SYMBOL_NS_GPL(rtl8366_vlan_del, REALTEK_DSA); void rtl8366_get_strings(struct dsa_switch *ds, int port, u32 stringset, uint8_t *data) @@ -403,7 +403,7 @@ void rtl8366_get_strings(struct dsa_switch *ds, int port, u32 stringset, for (i = 0; i < priv->num_mib_counters; i++) ethtool_puts(&data, priv->mib_counters[i].name); } -EXPORT_SYMBOL_GPL(rtl8366_get_strings); +EXPORT_SYMBOL_NS_GPL(rtl8366_get_strings, REALTEK_DSA); int rtl8366_get_sset_count(struct dsa_switch *ds, int port, int sset) { @@ -417,7 +417,7 @@ int rtl8366_get_sset_count(struct dsa_switch *ds, int port, int sset) return priv->num_mib_counters; } -EXPORT_SYMBOL_GPL(rtl8366_get_sset_count); +EXPORT_SYMBOL_NS_GPL(rtl8366_get_sset_count, REALTEK_DSA); void rtl8366_get_ethtool_stats(struct dsa_switch *ds, int port, uint64_t *data) { @@ -441,4 +441,4 @@ void rtl8366_get_ethtool_stats(struct dsa_switch *ds, int port, uint64_t *data) data[i] = mibvalue; } } -EXPORT_SYMBOL_GPL(rtl8366_get_ethtool_stats); +EXPORT_SYMBOL_NS_GPL(rtl8366_get_ethtool_stats, REALTEK_DSA); diff --git a/drivers/net/dsa/realtek/rtl8366rb.c b/drivers/net/dsa/realtek/rtl8366rb.c index e3b6a470ca67..e10ae94cf771 100644 --- a/drivers/net/dsa/realtek/rtl8366rb.c +++ b/drivers/net/dsa/realtek/rtl8366rb.c @@ -23,6 +23,9 @@ #include <linux/regmap.h> #include "realtek.h" +#include "realtek-smi.h" +#include "realtek-mdio.h" +#include "rtl83xx.h" #define RTL8366RB_PORT_NUM_CPU 5 #define RTL8366RB_NUM_PORTS 6 @@ -1030,12 +1033,10 @@ static int rtl8366rb_setup(struct dsa_switch *ds) if (ret) dev_info(priv->dev, "no interrupt support\n"); - if (priv->setup_interface) { - ret = priv->setup_interface(ds); - if (ret) { - dev_err(priv->dev, "could not set up MDIO bus\n"); - return -ENODEV; - } + ret = rtl83xx_setup_user_mdio(ds); + if (ret) { + dev_err(priv->dev, "could not set up MDIO bus\n"); + return -ENODEV; } return 0; @@ -1650,6 +1651,7 @@ static int rtl8366rb_get_mc_index(struct realtek_priv *priv, int port, int *val) static int rtl8366rb_set_mc_index(struct realtek_priv *priv, int port, int index) { + struct dsa_switch *ds = &priv->ds; struct rtl8366rb *rb; bool pvid_enabled; int ret; @@ -1674,7 +1676,7 @@ static int rtl8366rb_set_mc_index(struct realtek_priv *priv, int port, int index * not drop any untagged or C-tagged frames. Make sure to update the * filtering setting. */ - if (dsa_port_is_vlan_filtering(dsa_to_port(priv->ds, port))) + if (dsa_port_is_vlan_filtering(dsa_to_port(ds, port))) ret = rtl8366rb_drop_untagged(priv, port, !pvid_enabled); return ret; @@ -1718,7 +1720,7 @@ static int rtl8366rb_phy_read(struct realtek_priv *priv, int phy, int regnum) if (phy > RTL8366RB_PHY_NO_MAX) return -EINVAL; - mutex_lock(&priv->map_lock); + rtl83xx_lock(priv); ret = regmap_write(priv->map_nolock, RTL8366RB_PHY_ACCESS_CTRL_REG, RTL8366RB_PHY_CTRL_READ); @@ -1746,7 +1748,7 @@ static int rtl8366rb_phy_read(struct realtek_priv *priv, int phy, int regnum) phy, regnum, reg, val); out: - mutex_unlock(&priv->map_lock); + rtl83xx_unlock(priv); return ret; } @@ -1760,7 +1762,7 @@ static int rtl8366rb_phy_write(struct realtek_priv *priv, int phy, int regnum, if (phy > RTL8366RB_PHY_NO_MAX) return -EINVAL; - mutex_lock(&priv->map_lock); + rtl83xx_lock(priv); ret = regmap_write(priv->map_nolock, RTL8366RB_PHY_ACCESS_CTRL_REG, RTL8366RB_PHY_CTRL_WRITE); @@ -1777,22 +1779,11 @@ static int rtl8366rb_phy_write(struct realtek_priv *priv, int phy, int regnum, goto out; out: - mutex_unlock(&priv->map_lock); + rtl83xx_unlock(priv); return ret; } -static int rtl8366rb_dsa_phy_read(struct dsa_switch *ds, int phy, int regnum) -{ - return rtl8366rb_phy_read(ds->priv, phy, regnum); -} - -static int rtl8366rb_dsa_phy_write(struct dsa_switch *ds, int phy, int regnum, - u16 val) -{ - return rtl8366rb_phy_write(ds->priv, phy, regnum, val); -} - static int rtl8366rb_reset_chip(struct realtek_priv *priv) { int timeout = 10; @@ -1858,7 +1849,7 @@ static int rtl8366rb_detect(struct realtek_priv *priv) return 0; } -static const struct dsa_switch_ops rtl8366rb_switch_ops_smi = { +static const struct dsa_switch_ops rtl8366rb_switch_ops = { .get_tag_protocol = rtl8366_get_tag_protocol, .setup = rtl8366rb_setup, .phylink_get_caps = rtl8366rb_phylink_get_caps, @@ -1882,32 +1873,6 @@ static const struct dsa_switch_ops rtl8366rb_switch_ops_smi = { .port_max_mtu = rtl8366rb_max_mtu, }; -static const struct dsa_switch_ops rtl8366rb_switch_ops_mdio = { - .get_tag_protocol = rtl8366_get_tag_protocol, - .setup = rtl8366rb_setup, - .phy_read = rtl8366rb_dsa_phy_read, - .phy_write = rtl8366rb_dsa_phy_write, - .phylink_get_caps = rtl8366rb_phylink_get_caps, - .phylink_mac_link_up = rtl8366rb_mac_link_up, - .phylink_mac_link_down = rtl8366rb_mac_link_down, - .get_strings = rtl8366_get_strings, - .get_ethtool_stats = rtl8366_get_ethtool_stats, - .get_sset_count = rtl8366_get_sset_count, - .port_bridge_join = rtl8366rb_port_bridge_join, - .port_bridge_leave = rtl8366rb_port_bridge_leave, - .port_vlan_filtering = rtl8366rb_vlan_filtering, - .port_vlan_add = rtl8366_vlan_add, - .port_vlan_del = rtl8366_vlan_del, - .port_enable = rtl8366rb_port_enable, - .port_disable = rtl8366rb_port_disable, - .port_pre_bridge_flags = rtl8366rb_port_pre_bridge_flags, - .port_bridge_flags = rtl8366rb_port_bridge_flags, - .port_stp_state_set = rtl8366rb_port_stp_state_set, - .port_fast_age = rtl8366rb_port_fast_age, - .port_change_mtu = rtl8366rb_change_mtu, - .port_max_mtu = rtl8366rb_max_mtu, -}; - static const struct realtek_ops rtl8366rb_ops = { .detect = rtl8366rb_detect, .get_vlan_mc = rtl8366rb_get_vlan_mc, @@ -1925,16 +1890,66 @@ static const struct realtek_ops rtl8366rb_ops = { }; const struct realtek_variant rtl8366rb_variant = { - .ds_ops_smi = &rtl8366rb_switch_ops_smi, - .ds_ops_mdio = &rtl8366rb_switch_ops_mdio, + .ds_ops = &rtl8366rb_switch_ops, .ops = &rtl8366rb_ops, .clk_delay = 10, .cmd_read = 0xa9, .cmd_write = 0xa8, .chip_data_sz = sizeof(struct rtl8366rb), }; -EXPORT_SYMBOL_GPL(rtl8366rb_variant); + +static const struct of_device_id rtl8366rb_of_match[] = { + { .compatible = "realtek,rtl8366rb", .data = &rtl8366rb_variant, }, + { /* sentinel */ }, +}; +MODULE_DEVICE_TABLE(of, rtl8366rb_of_match); + +static struct platform_driver rtl8366rb_smi_driver = { + .driver = { + .name = "rtl8366rb-smi", + .of_match_table = rtl8366rb_of_match, + }, + .probe = realtek_smi_probe, + .remove_new = realtek_smi_remove, + .shutdown = realtek_smi_shutdown, +}; + +static struct mdio_driver rtl8366rb_mdio_driver = { + .mdiodrv.driver = { + .name = "rtl8366rb-mdio", + .of_match_table = rtl8366rb_of_match, + }, + .probe = realtek_mdio_probe, + .remove = realtek_mdio_remove, + .shutdown = realtek_mdio_shutdown, +}; + +static int rtl8366rb_init(void) +{ + int ret; + + ret = realtek_mdio_driver_register(&rtl8366rb_mdio_driver); + if (ret) + return ret; + + ret = realtek_smi_driver_register(&rtl8366rb_smi_driver); + if (ret) { + realtek_mdio_driver_unregister(&rtl8366rb_mdio_driver); + return ret; + } + + return 0; +} +module_init(rtl8366rb_init); + +static void __exit rtl8366rb_exit(void) +{ + realtek_smi_driver_unregister(&rtl8366rb_smi_driver); + realtek_mdio_driver_unregister(&rtl8366rb_mdio_driver); +} +module_exit(rtl8366rb_exit); MODULE_AUTHOR("Linus Walleij <linus.walleij@linaro.org>"); MODULE_DESCRIPTION("Driver for RTL8366RB ethernet switch"); MODULE_LICENSE("GPL"); +MODULE_IMPORT_NS(REALTEK_DSA); diff --git a/drivers/net/dsa/realtek/rtl83xx.c b/drivers/net/dsa/realtek/rtl83xx.c new file mode 100644 index 000000000000..801873754df2 --- /dev/null +++ b/drivers/net/dsa/realtek/rtl83xx.c @@ -0,0 +1,303 @@ +// SPDX-License-Identifier: GPL-2.0+ + +#include <linux/module.h> +#include <linux/regmap.h> +#include <linux/of_mdio.h> + +#include "realtek.h" +#include "rtl83xx.h" + +/** + * rtl83xx_lock() - Locks the mutex used by regmaps + * @ctx: realtek_priv pointer + * + * This function is passed to regmap to be used as the lock function. + * It is also used externally to block regmap before executing multiple + * operations that must happen in sequence (which will use + * realtek_priv.map_nolock instead). + * + * Context: Can sleep. Holds priv->map_lock lock. + * Return: nothing + */ +void rtl83xx_lock(void *ctx) +{ + struct realtek_priv *priv = ctx; + + mutex_lock(&priv->map_lock); +} +EXPORT_SYMBOL_NS_GPL(rtl83xx_lock, REALTEK_DSA); + +/** + * rtl83xx_unlock() - Unlocks the mutex used by regmaps + * @ctx: realtek_priv pointer + * + * This function unlocks the lock acquired by rtl83xx_lock. + * + * Context: Releases priv->map_lock lock. + * Return: nothing + */ +void rtl83xx_unlock(void *ctx) +{ + struct realtek_priv *priv = ctx; + + mutex_unlock(&priv->map_lock); +} +EXPORT_SYMBOL_NS_GPL(rtl83xx_unlock, REALTEK_DSA); + +static int rtl83xx_user_mdio_read(struct mii_bus *bus, int addr, int regnum) +{ + struct realtek_priv *priv = bus->priv; + + return priv->ops->phy_read(priv, addr, regnum); +} + +static int rtl83xx_user_mdio_write(struct mii_bus *bus, int addr, int regnum, + u16 val) +{ + struct realtek_priv *priv = bus->priv; + + return priv->ops->phy_write(priv, addr, regnum, val); +} + +/** + * rtl83xx_setup_user_mdio() - register the user mii bus driver + * @ds: DSA switch associated with this user_mii_bus + * + * Registers the MDIO bus for built-in Ethernet PHYs, and associates it with + * the mandatory 'mdio' child OF node of the switch. + * + * Context: Can sleep. + * Return: 0 on success, negative value for failure. + */ +int rtl83xx_setup_user_mdio(struct dsa_switch *ds) +{ + struct realtek_priv *priv = ds->priv; + struct device_node *mdio_np; + struct mii_bus *bus; + int ret = 0; + + mdio_np = of_get_child_by_name(priv->dev->of_node, "mdio"); + if (!mdio_np) { + dev_err(priv->dev, "no MDIO bus node\n"); + return -ENODEV; + } + + bus = devm_mdiobus_alloc(priv->dev); + if (!bus) { + ret = -ENOMEM; + goto err_put_node; + } + + bus->priv = priv; + bus->name = "Realtek user MII"; + bus->read = rtl83xx_user_mdio_read; + bus->write = rtl83xx_user_mdio_write; + snprintf(bus->id, MII_BUS_ID_SIZE, "%s:user_mii", dev_name(priv->dev)); + bus->parent = priv->dev; + + ret = devm_of_mdiobus_register(priv->dev, bus, mdio_np); + if (ret) { + dev_err(priv->dev, "unable to register MDIO bus %s\n", + bus->id); + goto err_put_node; + } + + priv->user_mii_bus = bus; + +err_put_node: + of_node_put(mdio_np); + + return ret; +} +EXPORT_SYMBOL_NS_GPL(rtl83xx_setup_user_mdio, REALTEK_DSA); + +/** + * rtl83xx_probe() - probe a Realtek switch + * @dev: the device being probed + * @interface_info: specific management interface info. + * + * This function initializes realtek_priv and reads data from the device tree + * node. The switch is hard resetted if a method is provided. + * + * Context: Can sleep. + * Return: Pointer to the realtek_priv or ERR_PTR() in case of failure. + * + * The realtek_priv pointer does not need to be freed as it is controlled by + * devres. + */ +struct realtek_priv * +rtl83xx_probe(struct device *dev, + const struct realtek_interface_info *interface_info) +{ + const struct realtek_variant *var; + struct realtek_priv *priv; + struct regmap_config rc = { + .reg_bits = 10, /* A4..A0 R4..R0 */ + .val_bits = 16, + .reg_stride = 1, + .max_register = 0xffff, + .reg_format_endian = REGMAP_ENDIAN_BIG, + .reg_read = interface_info->reg_read, + .reg_write = interface_info->reg_write, + .cache_type = REGCACHE_NONE, + .lock = rtl83xx_lock, + .unlock = rtl83xx_unlock, + }; + int ret; + + var = of_device_get_match_data(dev); + if (!var) + return ERR_PTR(-EINVAL); + + priv = devm_kzalloc(dev, size_add(sizeof(*priv), var->chip_data_sz), + GFP_KERNEL); + if (!priv) + return ERR_PTR(-ENOMEM); + + mutex_init(&priv->map_lock); + + rc.lock_arg = priv; + priv->map = devm_regmap_init(dev, NULL, priv, &rc); + if (IS_ERR(priv->map)) { + ret = PTR_ERR(priv->map); + dev_err(dev, "regmap init failed: %d\n", ret); + return ERR_PTR(ret); + } + + rc.disable_locking = true; + priv->map_nolock = devm_regmap_init(dev, NULL, priv, &rc); + if (IS_ERR(priv->map_nolock)) { + ret = PTR_ERR(priv->map_nolock); + dev_err(dev, "regmap init failed: %d\n", ret); + return ERR_PTR(ret); + } + + /* Link forward and backward */ + priv->dev = dev; + priv->variant = var; + priv->ops = var->ops; + priv->chip_data = (void *)priv + sizeof(*priv); + + spin_lock_init(&priv->lock); + + priv->leds_disabled = of_property_read_bool(dev->of_node, + "realtek,disable-leds"); + + /* TODO: if power is software controlled, set up any regulators here */ + priv->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_LOW); + if (IS_ERR(priv->reset)) { + dev_err(dev, "failed to get RESET GPIO\n"); + return ERR_CAST(priv->reset); + } + + dev_set_drvdata(dev, priv); + + if (priv->reset) { + gpiod_set_value(priv->reset, 1); + dev_dbg(dev, "asserted RESET\n"); + msleep(REALTEK_HW_STOP_DELAY); + gpiod_set_value(priv->reset, 0); + msleep(REALTEK_HW_START_DELAY); + dev_dbg(dev, "deasserted RESET\n"); + } + + return priv; +} +EXPORT_SYMBOL_NS_GPL(rtl83xx_probe, REALTEK_DSA); + +/** + * rtl83xx_register_switch() - detects and register a switch + * @priv: realtek_priv pointer + * + * This function first checks the switch chip ID and register a DSA + * switch. + * + * Context: Can sleep. Takes and releases priv->map_lock. + * Return: 0 on success, negative value for failure. + */ +int rtl83xx_register_switch(struct realtek_priv *priv) +{ + struct dsa_switch *ds = &priv->ds; + int ret; + + ret = priv->ops->detect(priv); + if (ret) { + dev_err_probe(priv->dev, ret, "unable to detect switch\n"); + return ret; + } + + ds->priv = priv; + ds->dev = priv->dev; + ds->ops = priv->variant->ds_ops; + ds->num_ports = priv->num_ports; + + ret = dsa_register_switch(ds); + if (ret) { + dev_err_probe(priv->dev, ret, "unable to register switch\n"); + return ret; + } + + return 0; +} +EXPORT_SYMBOL_NS_GPL(rtl83xx_register_switch, REALTEK_DSA); + +/** + * rtl83xx_unregister_switch() - unregister a switch + * @priv: realtek_priv pointer + * + * This function unregister a DSA switch. + * + * Context: Can sleep. + * Return: Nothing. + */ +void rtl83xx_unregister_switch(struct realtek_priv *priv) +{ + struct dsa_switch *ds = &priv->ds; + + dsa_unregister_switch(ds); +} +EXPORT_SYMBOL_NS_GPL(rtl83xx_unregister_switch, REALTEK_DSA); + +/** + * rtl83xx_shutdown() - shutdown a switch + * @priv: realtek_priv pointer + * + * This function shuts down the DSA switch and cleans the platform driver data, + * to prevent realtek_{smi,mdio}_remove() from running afterwards, which is + * possible if the parent bus implements its own .shutdown() as .remove(). + * + * Context: Can sleep. + * Return: Nothing. + */ +void rtl83xx_shutdown(struct realtek_priv *priv) +{ + struct dsa_switch *ds = &priv->ds; + + dsa_switch_shutdown(ds); + + dev_set_drvdata(priv->dev, NULL); +} +EXPORT_SYMBOL_NS_GPL(rtl83xx_shutdown, REALTEK_DSA); + +/** + * rtl83xx_remove() - Cleanup a realtek switch driver + * @priv: realtek_priv pointer + * + * If a method is provided, this function asserts the hard reset of the switch + * in order to avoid leaking traffic when the driver is gone. + * + * Context: Might sleep if priv->gdev->chip->can_sleep. + * Return: nothing + */ +void rtl83xx_remove(struct realtek_priv *priv) +{ + /* leave the device reset asserted */ + if (priv->reset) + gpiod_set_value(priv->reset, 1); +} +EXPORT_SYMBOL_NS_GPL(rtl83xx_remove, REALTEK_DSA); + +MODULE_AUTHOR("Luiz Angelo Daros de Luca <luizluca@gmail.com>"); +MODULE_AUTHOR("Linus Walleij <linus.walleij@linaro.org>"); +MODULE_DESCRIPTION("Realtek DSA switches common module"); +MODULE_LICENSE("GPL"); diff --git a/drivers/net/dsa/realtek/rtl83xx.h b/drivers/net/dsa/realtek/rtl83xx.h new file mode 100644 index 000000000000..0ddff33df6b0 --- /dev/null +++ b/drivers/net/dsa/realtek/rtl83xx.h @@ -0,0 +1,22 @@ +/* SPDX-License-Identifier: GPL-2.0+ */ + +#ifndef _RTL83XX_H +#define _RTL83XX_H + +struct realtek_interface_info { + int (*reg_read)(void *ctx, u32 reg, u32 *val); + int (*reg_write)(void *ctx, u32 reg, u32 val); +}; + +void rtl83xx_lock(void *ctx); +void rtl83xx_unlock(void *ctx); +int rtl83xx_setup_user_mdio(struct dsa_switch *ds); +struct realtek_priv * +rtl83xx_probe(struct device *dev, + const struct realtek_interface_info *interface_info); +int rtl83xx_register_switch(struct realtek_priv *priv); +void rtl83xx_unregister_switch(struct realtek_priv *priv); +void rtl83xx_shutdown(struct realtek_priv *priv); +void rtl83xx_remove(struct realtek_priv *priv); + +#endif /* _RTL83XX_H */ diff --git a/drivers/net/dummy.c b/drivers/net/dummy.c index 768454aa36d6..946bba0701a4 100644 --- a/drivers/net/dummy.c +++ b/drivers/net/dummy.c @@ -71,6 +71,7 @@ static int dummy_dev_init(struct net_device *dev) if (!dev->lstats) return -ENOMEM; + netdev_lockdep_set_classes(dev); return 0; } diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index 50f674031f72..7d0a2f5f3282 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -4616,7 +4616,7 @@ static int tg3_init_5401phy_dsp(struct tg3 *tp) static bool tg3_phy_eee_config_ok(struct tg3 *tp) { - struct ethtool_keee eee; + struct ethtool_keee eee = {}; if (!(tp->phy_flags & TG3_PHYFLG_EEE_CAP)) return true; diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index 42bdc01a304e..207f1f66c117 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -85,8 +85,6 @@ static int fec_enet_xdp_tx_xmit(struct fec_enet_private *fep, static const u16 fec_enet_vlan_pri_to_queue[8] = {0, 0, 1, 1, 1, 2, 2, 2}; -/* Pause frame feild and FIFO threshold */ -#define FEC_ENET_FCE (1 << 5) #define FEC_ENET_RSEM_V 0x84 #define FEC_ENET_RSFL_V 16 #define FEC_ENET_RAEM_V 0x8 @@ -240,8 +238,8 @@ MODULE_PARM_DESC(macaddr, "FEC Ethernet MAC address"); #define PKT_MINBUF_SIZE 64 /* FEC receive acceleration */ -#define FEC_RACC_IPDIS (1 << 1) -#define FEC_RACC_PRODIS (1 << 2) +#define FEC_RACC_IPDIS BIT(1) +#define FEC_RACC_PRODIS BIT(2) #define FEC_RACC_SHIFT16 BIT(7) #define FEC_RACC_OPTIONS (FEC_RACC_IPDIS | FEC_RACC_PRODIS) @@ -273,8 +271,23 @@ MODULE_PARM_DESC(macaddr, "FEC Ethernet MAC address"); #define FEC_MMFR_TA (2 << 16) #define FEC_MMFR_DATA(v) (v & 0xffff) /* FEC ECR bits definition */ -#define FEC_ECR_MAGICEN (1 << 2) -#define FEC_ECR_SLEEP (1 << 3) +#define FEC_ECR_RESET BIT(0) +#define FEC_ECR_ETHEREN BIT(1) +#define FEC_ECR_MAGICEN BIT(2) +#define FEC_ECR_SLEEP BIT(3) +#define FEC_ECR_EN1588 BIT(4) +#define FEC_ECR_BYTESWP BIT(8) +/* FEC RCR bits definition */ +#define FEC_RCR_LOOP BIT(0) +#define FEC_RCR_HALFDPX BIT(1) +#define FEC_RCR_MII BIT(2) +#define FEC_RCR_PROMISC BIT(3) +#define FEC_RCR_BC_REJ BIT(4) +#define FEC_RCR_FLOWCTL BIT(5) +#define FEC_RCR_RMII BIT(8) +#define FEC_RCR_10BASET BIT(9) +/* TX WMARK bits */ +#define FEC_TXWMRK_STRFWD BIT(8) #define FEC_MII_TIMEOUT 30000 /* us */ @@ -1062,7 +1075,7 @@ fec_restart(struct net_device *ndev) struct fec_enet_private *fep = netdev_priv(ndev); u32 temp_mac[2]; u32 rcntl = OPT_FRAME_SIZE | 0x04; - u32 ecntl = 0x2; /* ETHEREN */ + u32 ecntl = FEC_ECR_ETHEREN; /* Whack a reset. We should wait for this. * For i.MX6SX SOC, enet use AXI bus, we use disable MAC @@ -1137,18 +1150,18 @@ fec_restart(struct net_device *ndev) fep->phy_interface == PHY_INTERFACE_MODE_RGMII_TXID) rcntl |= (1 << 6); else if (fep->phy_interface == PHY_INTERFACE_MODE_RMII) - rcntl |= (1 << 8); + rcntl |= FEC_RCR_RMII; else - rcntl &= ~(1 << 8); + rcntl &= ~FEC_RCR_RMII; /* 1G, 100M or 10M */ if (ndev->phydev) { if (ndev->phydev->speed == SPEED_1000) ecntl |= (1 << 5); else if (ndev->phydev->speed == SPEED_100) - rcntl &= ~(1 << 9); + rcntl &= ~FEC_RCR_10BASET; else - rcntl |= (1 << 9); + rcntl |= FEC_RCR_10BASET; } } else { #ifdef FEC_MIIGSK_ENR @@ -1181,7 +1194,7 @@ fec_restart(struct net_device *ndev) if ((fep->pause_flag & FEC_PAUSE_FLAG_ENABLE) || ((fep->pause_flag & FEC_PAUSE_FLAG_AUTONEG) && ndev->phydev && ndev->phydev->pause)) { - rcntl |= FEC_ENET_FCE; + rcntl |= FEC_RCR_FLOWCTL; /* set FIFO threshold parameter to reduce overrun */ writel(FEC_ENET_RSEM_V, fep->hwp + FEC_R_FIFO_RSEM); @@ -1192,7 +1205,7 @@ fec_restart(struct net_device *ndev) /* OPD */ writel(FEC_ENET_OPD_V, fep->hwp + FEC_OPD); } else { - rcntl &= ~FEC_ENET_FCE; + rcntl &= ~FEC_RCR_FLOWCTL; } #endif /* !defined(CONFIG_M5272) */ @@ -1207,13 +1220,13 @@ fec_restart(struct net_device *ndev) if (fep->quirks & FEC_QUIRK_ENET_MAC) { /* enable ENET endian swap */ - ecntl |= (1 << 8); + ecntl |= FEC_ECR_BYTESWP; /* enable ENET store and forward mode */ - writel(1 << 8, fep->hwp + FEC_X_WMRK); + writel(FEC_TXWMRK_STRFWD, fep->hwp + FEC_X_WMRK); } if (fep->bufdesc_ex) - ecntl |= (1 << 4); + ecntl |= FEC_ECR_EN1588; if (fep->quirks & FEC_QUIRK_DELAYED_CLKS_SUPPORT && fep->rgmii_txc_dly) @@ -1312,7 +1325,7 @@ static void fec_stop(struct net_device *ndev) { struct fec_enet_private *fep = netdev_priv(ndev); - u32 rmii_mode = readl(fep->hwp + FEC_R_CNTRL) & (1 << 8); + u32 rmii_mode = readl(fep->hwp + FEC_R_CNTRL) & FEC_RCR_RMII; u32 val; /* We cannot expect a graceful transmit stop without link !!! */ @@ -1331,7 +1344,7 @@ fec_stop(struct net_device *ndev) if (fep->quirks & FEC_QUIRK_HAS_MULTI_QUEUES) { writel(0, fep->hwp + FEC_ECNTRL); } else { - writel(1, fep->hwp + FEC_ECNTRL); + writel(FEC_ECR_RESET, fep->hwp + FEC_ECNTRL); udelay(10); } } else { @@ -1345,12 +1358,11 @@ fec_stop(struct net_device *ndev) /* We have to keep ENET enabled to have MII interrupt stay working */ if (fep->quirks & FEC_QUIRK_ENET_MAC && !(fep->wol_flag & FEC_WOL_FLAG_SLEEP_ON)) { - writel(2, fep->hwp + FEC_ECNTRL); + writel(FEC_ECR_ETHEREN, fep->hwp + FEC_ECNTRL); writel(rmii_mode, fep->hwp + FEC_R_CNTRL); } } - static void fec_timeout(struct net_device *ndev, unsigned int txqueue) { diff --git a/drivers/net/ethernet/marvell/Kconfig b/drivers/net/ethernet/marvell/Kconfig index 884d64114bff..837295fecd17 100644 --- a/drivers/net/ethernet/marvell/Kconfig +++ b/drivers/net/ethernet/marvell/Kconfig @@ -180,6 +180,7 @@ config SKY2_DEBUG source "drivers/net/ethernet/marvell/octeontx2/Kconfig" source "drivers/net/ethernet/marvell/octeon_ep/Kconfig" +source "drivers/net/ethernet/marvell/octeon_ep_vf/Kconfig" source "drivers/net/ethernet/marvell/prestera/Kconfig" endif # NET_VENDOR_MARVELL diff --git a/drivers/net/ethernet/marvell/Makefile b/drivers/net/ethernet/marvell/Makefile index ceba4aa4f026..a399defe25fd 100644 --- a/drivers/net/ethernet/marvell/Makefile +++ b/drivers/net/ethernet/marvell/Makefile @@ -12,5 +12,6 @@ obj-$(CONFIG_PXA168_ETH) += pxa168_eth.o obj-$(CONFIG_SKGE) += skge.o obj-$(CONFIG_SKY2) += sky2.o obj-y += octeon_ep/ +obj-y += octeon_ep_vf/ obj-y += octeontx2/ obj-y += prestera/ diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/Kconfig b/drivers/net/ethernet/marvell/octeon_ep_vf/Kconfig new file mode 100644 index 000000000000..dbd1267bda0c --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/Kconfig @@ -0,0 +1,19 @@ +# SPDX-License-Identifier: GPL-2.0-only +# +# Marvell's Octeon PCI Endpoint NIC VF Driver Configuration +# + +config OCTEON_EP_VF + tristate "Marvell Octeon PCI Endpoint NIC VF Driver" + depends on 64BIT + depends on PCI + help + This driver supports networking functionality of Marvell's + Octeon PCI Endpoint NIC VF. + + To know the list of devices supported by this driver, refer + documentation in + <file:Documentation/networking/device_drivers/ethernet/marvell/octeon_ep_vf.rst>. + + To compile this drivers as a module, choose M here. Name of the + module is octeon_ep_vf. diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/Makefile b/drivers/net/ethernet/marvell/octeon_ep_vf/Makefile new file mode 100644 index 000000000000..4a5f9fcb0b40 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/Makefile @@ -0,0 +1,10 @@ +# SPDX-License-Identifier: GPL-2.0 +# +# Network driver for Marvell's Octeon PCI Endpoint NIC VF +# + +obj-$(CONFIG_OCTEON_EP_VF) += octeon_ep_vf.o + +octeon_ep_vf-y := octep_vf_main.o octep_vf_cn9k.o octep_vf_cnxk.o \ + octep_vf_tx.o octep_vf_rx.o octep_vf_mbox.o \ + octep_vf_ethtool.o diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_cn9k.c b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_cn9k.c new file mode 100644 index 000000000000..88937fce75f1 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_cn9k.c @@ -0,0 +1,489 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ + +#include <linux/pci.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> + +#include "octep_vf_config.h" +#include "octep_vf_main.h" +#include "octep_vf_regs_cn9k.h" + +/* Dump useful hardware IQ/OQ CSRs for debug purpose */ +static void cn93_vf_dump_q_regs(struct octep_vf_device *oct, int qno) +{ + struct device *dev = &oct->pdev->dev; + + dev_info(dev, "IQ-%d register dump\n", qno); + dev_info(dev, "R[%d]_IN_INSTR_DBELL[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_IN_INSTR_DBELL(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_INSTR_DBELL(qno))); + dev_info(dev, "R[%d]_IN_CONTROL[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_IN_CONTROL(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_CONTROL(qno))); + dev_info(dev, "R[%d]_IN_ENABLE[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_IN_ENABLE(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_ENABLE(qno))); + dev_info(dev, "R[%d]_IN_INSTR_BADDR[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_IN_INSTR_BADDR(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_INSTR_BADDR(qno))); + dev_info(dev, "R[%d]_IN_INSTR_RSIZE[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_IN_INSTR_RSIZE(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_INSTR_RSIZE(qno))); + dev_info(dev, "R[%d]_IN_CNTS[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_IN_CNTS(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_CNTS(qno))); + dev_info(dev, "R[%d]_IN_INT_LEVELS[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_IN_INT_LEVELS(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_INT_LEVELS(qno))); + dev_info(dev, "R[%d]_IN_PKT_CNT[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_IN_PKT_CNT(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_PKT_CNT(qno))); + dev_info(dev, "R[%d]_IN_BYTE_CNT[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_IN_BYTE_CNT(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_BYTE_CNT(qno))); + + dev_info(dev, "OQ-%d register dump\n", qno); + dev_info(dev, "R[%d]_OUT_SLIST_DBELL[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_OUT_SLIST_DBELL(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_SLIST_DBELL(qno))); + dev_info(dev, "R[%d]_OUT_CONTROL[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_OUT_CONTROL(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_CONTROL(qno))); + dev_info(dev, "R[%d]_OUT_ENABLE[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_OUT_ENABLE(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_ENABLE(qno))); + dev_info(dev, "R[%d]_OUT_SLIST_BADDR[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_OUT_SLIST_BADDR(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_SLIST_BADDR(qno))); + dev_info(dev, "R[%d]_OUT_SLIST_RSIZE[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_OUT_SLIST_RSIZE(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_SLIST_RSIZE(qno))); + dev_info(dev, "R[%d]_OUT_CNTS[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_OUT_CNTS(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_CNTS(qno))); + dev_info(dev, "R[%d]_OUT_INT_LEVELS[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_OUT_INT_LEVELS(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_INT_LEVELS(qno))); + dev_info(dev, "R[%d]_OUT_PKT_CNT[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_OUT_PKT_CNT(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_PKT_CNT(qno))); + dev_info(dev, "R[%d]_OUT_BYTE_CNT[0x%llx]: 0x%016llx\n", + qno, CN93_VF_SDP_R_OUT_BYTE_CNT(qno), + octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_BYTE_CNT(qno))); +} + +/* Reset Hardware Tx queue */ +static void cn93_vf_reset_iq(struct octep_vf_device *oct, int q_no) +{ + u64 val = ULL(0); + + dev_dbg(&oct->pdev->dev, "Reset VF IQ-%d\n", q_no); + + /* Disable the Tx/Instruction Ring */ + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_ENABLE(q_no), val); + + /* clear the Instruction Ring packet/byte counts and doorbell CSRs */ + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_INT_LEVELS(q_no), val); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_PKT_CNT(q_no), val); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_BYTE_CNT(q_no), val); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_INSTR_BADDR(q_no), val); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_INSTR_RSIZE(q_no), val); + + val = GENMASK_ULL(31, 0); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_INSTR_DBELL(q_no), val); + + val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_CNTS(q_no)); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_CNTS(q_no), + val & GENMASK_ULL(31, 0)); +} + +/* Reset Hardware Rx queue */ +static void cn93_vf_reset_oq(struct octep_vf_device *oct, int q_no) +{ + u64 val = ULL(0); + + /* Disable Output (Rx) Ring */ + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_ENABLE(q_no), val); + + /* Clear count CSRs */ + val = octep_vf_read_csr(oct, CN93_VF_SDP_R_OUT_CNTS(q_no)); + octep_vf_write_csr(oct, CN93_VF_SDP_R_OUT_CNTS(q_no), val); + + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_PKT_CNT(q_no), GENMASK_ULL(35, 0)); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_SLIST_DBELL(q_no), GENMASK_ULL(31, 0)); +} + +/* Reset all hardware Tx/Rx queues */ +static void octep_vf_reset_io_queues_cn93(struct octep_vf_device *oct) +{ + struct pci_dev *pdev = oct->pdev; + int q; + + dev_dbg(&pdev->dev, "Reset OCTEP_CN93 VF IO Queues\n"); + + for (q = 0; q < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); q++) { + cn93_vf_reset_iq(oct, q); + cn93_vf_reset_oq(oct, q); + } +} + +/* Initialize configuration limits and initial active config */ +static void octep_vf_init_config_cn93_vf(struct octep_vf_device *oct) +{ + struct octep_vf_config *conf = oct->conf; + u64 reg_val; + + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_CONTROL(0)); + conf->ring_cfg.max_io_rings = (reg_val >> CN93_VF_R_IN_CTL_RPVF_POS) & + CN93_VF_R_IN_CTL_RPVF_MASK; + conf->ring_cfg.active_io_rings = conf->ring_cfg.max_io_rings; + + conf->iq.num_descs = OCTEP_VF_IQ_MAX_DESCRIPTORS; + conf->iq.instr_type = OCTEP_VF_64BYTE_INSTR; + conf->iq.db_min = OCTEP_VF_DB_MIN; + conf->iq.intr_threshold = OCTEP_VF_IQ_INTR_THRESHOLD; + + conf->oq.num_descs = OCTEP_VF_OQ_MAX_DESCRIPTORS; + conf->oq.buf_size = OCTEP_VF_OQ_BUF_SIZE; + conf->oq.refill_threshold = OCTEP_VF_OQ_REFILL_THRESHOLD; + conf->oq.oq_intr_pkt = OCTEP_VF_OQ_INTR_PKT_THRESHOLD; + conf->oq.oq_intr_time = OCTEP_VF_OQ_INTR_TIME_THRESHOLD; + + conf->msix_cfg.ioq_msix = conf->ring_cfg.active_io_rings; +} + +/* Setup registers for a hardware Tx Queue */ +static void octep_vf_setup_iq_regs_cn93(struct octep_vf_device *oct, int iq_no) +{ + struct octep_vf_iq *iq = oct->iq[iq_no]; + u32 reset_instr_cnt; + u64 reg_val; + + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_CONTROL(iq_no)); + + /* wait for IDLE to set to 1 */ + if (!(reg_val & CN93_VF_R_IN_CTL_IDLE)) { + do { + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_CONTROL(iq_no)); + } while (!(reg_val & CN93_VF_R_IN_CTL_IDLE)); + } + reg_val |= CN93_VF_R_IN_CTL_RDSIZE; + reg_val |= CN93_VF_R_IN_CTL_IS_64B; + reg_val |= CN93_VF_R_IN_CTL_ESR; + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_CONTROL(iq_no), reg_val); + + /* Write the start of the input queue's ring and its size */ + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_INSTR_BADDR(iq_no), iq->desc_ring_dma); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_INSTR_RSIZE(iq_no), iq->max_count); + + /* Remember the doorbell & instruction count register addr for this queue */ + iq->doorbell_reg = oct->mmio.hw_addr + CN93_VF_SDP_R_IN_INSTR_DBELL(iq_no); + iq->inst_cnt_reg = oct->mmio.hw_addr + CN93_VF_SDP_R_IN_CNTS(iq_no); + iq->intr_lvl_reg = oct->mmio.hw_addr + CN93_VF_SDP_R_IN_INT_LEVELS(iq_no); + + /* Store the current instruction counter (used in flush_iq calculation) */ + reset_instr_cnt = readl(iq->inst_cnt_reg); + writel(reset_instr_cnt, iq->inst_cnt_reg); + + /* INTR_THRESHOLD is set to max(FFFFFFFF) to disable the INTR */ + reg_val = CFG_GET_IQ_INTR_THRESHOLD(oct->conf) & GENMASK_ULL(31, 0); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_INT_LEVELS(iq_no), reg_val); +} + +/* Setup registers for a hardware Rx Queue */ +static void octep_vf_setup_oq_regs_cn93(struct octep_vf_device *oct, int oq_no) +{ + struct octep_vf_oq *oq = oct->oq[oq_no]; + u32 time_threshold = 0; + u64 oq_ctl = ULL(0); + u64 reg_val; + + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_CONTROL(oq_no)); + + /* wait for IDLE to set to 1 */ + if (!(reg_val & CN93_VF_R_OUT_CTL_IDLE)) { + do { + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_CONTROL(oq_no)); + } while (!(reg_val & CN93_VF_R_OUT_CTL_IDLE)); + } + + reg_val &= ~(CN93_VF_R_OUT_CTL_IMODE); + reg_val &= ~(CN93_VF_R_OUT_CTL_ROR_P); + reg_val &= ~(CN93_VF_R_OUT_CTL_NSR_P); + reg_val &= ~(CN93_VF_R_OUT_CTL_ROR_I); + reg_val &= ~(CN93_VF_R_OUT_CTL_NSR_I); + reg_val &= ~(CN93_VF_R_OUT_CTL_ES_I); + reg_val &= ~(CN93_VF_R_OUT_CTL_ROR_D); + reg_val &= ~(CN93_VF_R_OUT_CTL_NSR_D); + reg_val &= ~(CN93_VF_R_OUT_CTL_ES_D); + reg_val |= (CN93_VF_R_OUT_CTL_ES_P); + + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_CONTROL(oq_no), reg_val); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_SLIST_BADDR(oq_no), oq->desc_ring_dma); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_SLIST_RSIZE(oq_no), oq->max_count); + + oq_ctl = octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_CONTROL(oq_no)); + oq_ctl &= ~GENMASK_ULL(22, 0); //clear the ISIZE and BSIZE (22-0) + oq_ctl |= (oq->buffer_size & GENMASK_ULL(15, 0)); //populate the BSIZE (15-0) + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_CONTROL(oq_no), oq_ctl); + + /* Get the mapped address of the pkt_sent and pkts_credit regs */ + oq->pkts_sent_reg = oct->mmio.hw_addr + CN93_VF_SDP_R_OUT_CNTS(oq_no); + oq->pkts_credit_reg = oct->mmio.hw_addr + CN93_VF_SDP_R_OUT_SLIST_DBELL(oq_no); + + time_threshold = CFG_GET_OQ_INTR_TIME(oct->conf); + reg_val = ((u64)time_threshold << 32) | CFG_GET_OQ_INTR_PKT(oct->conf); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_INT_LEVELS(oq_no), reg_val); +} + +/* Setup registers for a VF mailbox */ +static void octep_vf_setup_mbox_regs_cn93(struct octep_vf_device *oct, int q_no) +{ + struct octep_vf_mbox *mbox = oct->mbox; + + /* PF to VF DATA reg. VF reads from this reg */ + mbox->mbox_read_reg = oct->mmio.hw_addr + CN93_VF_SDP_R_MBOX_PF_VF_DATA(q_no); + + /* VF mbox interrupt reg */ + mbox->mbox_int_reg = oct->mmio.hw_addr + CN93_VF_SDP_R_MBOX_PF_VF_INT(q_no); + + /* VF to PF DATA reg. VF writes into this reg */ + mbox->mbox_write_reg = oct->mmio.hw_addr + CN93_VF_SDP_R_MBOX_VF_PF_DATA(q_no); +} + +/* Mailbox Interrupt handler */ +static void cn93_handle_vf_mbox_intr(struct octep_vf_device *oct) +{ + if (oct->mbox) + schedule_work(&oct->mbox->wk.work); + else + dev_err(&oct->pdev->dev, "cannot schedule work on invalid mbox\n"); +} + +/* Tx/Rx queue interrupt handler */ +static irqreturn_t octep_vf_ioq_intr_handler_cn93(void *data) +{ + struct octep_vf_ioq_vector *vector = data; + struct octep_vf_device *oct; + struct octep_vf_oq *oq; + u64 reg_val; + + oct = vector->octep_vf_dev; + oq = vector->oq; + /* Mailbox interrupt arrives along with interrupt of tx/rx ring pair 0 */ + if (oq->q_no == 0) { + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_MBOX_PF_VF_INT(0)); + if (reg_val & CN93_VF_SDP_R_MBOX_PF_VF_INT_STATUS) { + cn93_handle_vf_mbox_intr(oct); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_MBOX_PF_VF_INT(0), reg_val); + } + } + napi_schedule_irqoff(oq->napi); + return IRQ_HANDLED; +} + +/* Re-initialize Octeon hardware registers */ +static void octep_vf_reinit_regs_cn93(struct octep_vf_device *oct) +{ + u32 i; + + for (i = 0; i < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); i++) + oct->hw_ops.setup_iq_regs(oct, i); + + for (i = 0; i < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); i++) + oct->hw_ops.setup_oq_regs(oct, i); + + oct->hw_ops.enable_interrupts(oct); + oct->hw_ops.enable_io_queues(oct); + + for (i = 0; i < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); i++) + writel(oct->oq[i]->max_count, oct->oq[i]->pkts_credit_reg); +} + +/* Enable all interrupts */ +static void octep_vf_enable_interrupts_cn93(struct octep_vf_device *oct) +{ + int num_rings, q; + u64 reg_val; + + num_rings = CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); + for (q = 0; q < num_rings; q++) { + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_INT_LEVELS(q)); + reg_val |= BIT_ULL_MASK(62); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_INT_LEVELS(q), reg_val); + + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_INT_LEVELS(q)); + reg_val |= BIT_ULL_MASK(62); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_INT_LEVELS(q), reg_val); + } + /* Enable PF to VF mbox interrupt by setting 2nd bit*/ + octep_vf_write_csr64(oct, CN93_VF_SDP_R_MBOX_PF_VF_INT(0), + CN93_VF_SDP_R_MBOX_PF_VF_INT_ENAB); +} + +/* Disable all interrupts */ +static void octep_vf_disable_interrupts_cn93(struct octep_vf_device *oct) +{ + int num_rings, q; + u64 reg_val; + + /* Disable PF to VF mbox interrupt by setting 2nd bit*/ + if (oct->mbox) + octep_vf_write_csr64(oct, CN93_VF_SDP_R_MBOX_PF_VF_INT(0), 0x0); + + num_rings = CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); + for (q = 0; q < num_rings; q++) { + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_INT_LEVELS(q)); + reg_val &= ~BIT_ULL_MASK(62); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_INT_LEVELS(q), reg_val); + + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_INT_LEVELS(q)); + reg_val &= ~BIT_ULL_MASK(62); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_INT_LEVELS(q), reg_val); + } +} + +/* Get new Octeon Read Index: index of descriptor that Octeon reads next. */ +static u32 octep_vf_update_iq_read_index_cn93(struct octep_vf_iq *iq) +{ + u32 pkt_in_done = readl(iq->inst_cnt_reg); + u32 last_done, new_idx; + + last_done = pkt_in_done - iq->pkt_in_done; + iq->pkt_in_done = pkt_in_done; + + new_idx = (iq->octep_vf_read_index + last_done) % iq->max_count; + + return new_idx; +} + +/* Enable a hardware Tx Queue */ +static void octep_vf_enable_iq_cn93(struct octep_vf_device *oct, int iq_no) +{ + u64 loop = HZ; + u64 reg_val; + + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_INSTR_DBELL(iq_no), GENMASK_ULL(31, 0)); + + while (octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_INSTR_DBELL(iq_no)) && + loop--) { + schedule_timeout_interruptible(1); + } + + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_INT_LEVELS(iq_no)); + reg_val |= BIT_ULL_MASK(62); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_INT_LEVELS(iq_no), reg_val); + + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_ENABLE(iq_no)); + reg_val |= ULL(1); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_ENABLE(iq_no), reg_val); +} + +/* Enable a hardware Rx Queue */ +static void octep_vf_enable_oq_cn93(struct octep_vf_device *oct, int oq_no) +{ + u64 reg_val; + + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_INT_LEVELS(oq_no)); + reg_val |= BIT_ULL_MASK(62); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_INT_LEVELS(oq_no), reg_val); + + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_SLIST_DBELL(oq_no), GENMASK_ULL(31, 0)); + + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_ENABLE(oq_no)); + reg_val |= ULL(1); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_ENABLE(oq_no), reg_val); +} + +/* Enable all hardware Tx/Rx Queues assigned to VF */ +static void octep_vf_enable_io_queues_cn93(struct octep_vf_device *oct) +{ + u8 q; + + for (q = 0; q < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); q++) { + octep_vf_enable_iq_cn93(oct, q); + octep_vf_enable_oq_cn93(oct, q); + } +} + +/* Disable a hardware Tx Queue assigned to VF */ +static void octep_vf_disable_iq_cn93(struct octep_vf_device *oct, int iq_no) +{ + u64 reg_val; + + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_IN_ENABLE(iq_no)); + reg_val &= ~ULL(1); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_IN_ENABLE(iq_no), reg_val); +} + +/* Disable a hardware Rx Queue assigned to VF */ +static void octep_vf_disable_oq_cn93(struct octep_vf_device *oct, int oq_no) +{ + u64 reg_val; + + reg_val = octep_vf_read_csr64(oct, CN93_VF_SDP_R_OUT_ENABLE(oq_no)); + reg_val &= ~ULL(1); + octep_vf_write_csr64(oct, CN93_VF_SDP_R_OUT_ENABLE(oq_no), reg_val); +} + +/* Disable all hardware Tx/Rx Queues assigned to VF */ +static void octep_vf_disable_io_queues_cn93(struct octep_vf_device *oct) +{ + int q; + + for (q = 0; q < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); q++) { + octep_vf_disable_iq_cn93(oct, q); + octep_vf_disable_oq_cn93(oct, q); + } +} + +/* Dump hardware registers (including Tx/Rx queues) for debugging. */ +static void octep_vf_dump_registers_cn93(struct octep_vf_device *oct) +{ + u8 num_rings, q; + + num_rings = CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); + for (q = 0; q < num_rings; q++) + cn93_vf_dump_q_regs(oct, q); +} + +/** + * octep_vf_device_setup_cn93() - Setup Octeon device. + * + * @oct: Octeon device private data structure. + * + * - initialize hardware operations. + * - get target side pcie port number for the device. + * - set initial configuration and max limits. + */ +void octep_vf_device_setup_cn93(struct octep_vf_device *oct) +{ + oct->hw_ops.setup_iq_regs = octep_vf_setup_iq_regs_cn93; + oct->hw_ops.setup_oq_regs = octep_vf_setup_oq_regs_cn93; + oct->hw_ops.setup_mbox_regs = octep_vf_setup_mbox_regs_cn93; + + oct->hw_ops.ioq_intr_handler = octep_vf_ioq_intr_handler_cn93; + oct->hw_ops.reinit_regs = octep_vf_reinit_regs_cn93; + + oct->hw_ops.enable_interrupts = octep_vf_enable_interrupts_cn93; + oct->hw_ops.disable_interrupts = octep_vf_disable_interrupts_cn93; + + oct->hw_ops.update_iq_read_idx = octep_vf_update_iq_read_index_cn93; + + oct->hw_ops.enable_iq = octep_vf_enable_iq_cn93; + oct->hw_ops.enable_oq = octep_vf_enable_oq_cn93; + oct->hw_ops.enable_io_queues = octep_vf_enable_io_queues_cn93; + + oct->hw_ops.disable_iq = octep_vf_disable_iq_cn93; + oct->hw_ops.disable_oq = octep_vf_disable_oq_cn93; + oct->hw_ops.disable_io_queues = octep_vf_disable_io_queues_cn93; + oct->hw_ops.reset_io_queues = octep_vf_reset_io_queues_cn93; + + oct->hw_ops.dump_registers = octep_vf_dump_registers_cn93; + octep_vf_init_config_cn93_vf(oct); +} diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_cnxk.c b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_cnxk.c new file mode 100644 index 000000000000..1f79dfad42c6 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_cnxk.c @@ -0,0 +1,500 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ + +#include <linux/pci.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> + +#include "octep_vf_config.h" +#include "octep_vf_main.h" +#include "octep_vf_regs_cnxk.h" + +/* Dump useful hardware IQ/OQ CSRs for debug purpose */ +static void cnxk_vf_dump_q_regs(struct octep_vf_device *oct, int qno) +{ + struct device *dev = &oct->pdev->dev; + + dev_info(dev, "IQ-%d register dump\n", qno); + dev_info(dev, "R[%d]_IN_INSTR_DBELL[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_IN_INSTR_DBELL(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_INSTR_DBELL(qno))); + dev_info(dev, "R[%d]_IN_CONTROL[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_IN_CONTROL(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_CONTROL(qno))); + dev_info(dev, "R[%d]_IN_ENABLE[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_IN_ENABLE(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_ENABLE(qno))); + dev_info(dev, "R[%d]_IN_INSTR_BADDR[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_IN_INSTR_BADDR(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_INSTR_BADDR(qno))); + dev_info(dev, "R[%d]_IN_INSTR_RSIZE[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_IN_INSTR_RSIZE(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_INSTR_RSIZE(qno))); + dev_info(dev, "R[%d]_IN_CNTS[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_IN_CNTS(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_CNTS(qno))); + dev_info(dev, "R[%d]_IN_INT_LEVELS[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_IN_INT_LEVELS(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_INT_LEVELS(qno))); + dev_info(dev, "R[%d]_IN_PKT_CNT[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_IN_PKT_CNT(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_PKT_CNT(qno))); + dev_info(dev, "R[%d]_IN_BYTE_CNT[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_IN_BYTE_CNT(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_BYTE_CNT(qno))); + + dev_info(dev, "OQ-%d register dump\n", qno); + dev_info(dev, "R[%d]_OUT_SLIST_DBELL[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_OUT_SLIST_DBELL(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_SLIST_DBELL(qno))); + dev_info(dev, "R[%d]_OUT_CONTROL[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_OUT_CONTROL(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_CONTROL(qno))); + dev_info(dev, "R[%d]_OUT_ENABLE[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_OUT_ENABLE(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_ENABLE(qno))); + dev_info(dev, "R[%d]_OUT_SLIST_BADDR[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_OUT_SLIST_BADDR(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_SLIST_BADDR(qno))); + dev_info(dev, "R[%d]_OUT_SLIST_RSIZE[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_OUT_SLIST_RSIZE(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_SLIST_RSIZE(qno))); + dev_info(dev, "R[%d]_OUT_CNTS[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_OUT_CNTS(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_CNTS(qno))); + dev_info(dev, "R[%d]_OUT_INT_LEVELS[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_OUT_INT_LEVELS(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_INT_LEVELS(qno))); + dev_info(dev, "R[%d]_OUT_PKT_CNT[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_OUT_PKT_CNT(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_PKT_CNT(qno))); + dev_info(dev, "R[%d]_OUT_BYTE_CNT[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_OUT_BYTE_CNT(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_BYTE_CNT(qno))); + dev_info(dev, "R[%d]_ERR_TYPE[0x%llx]: 0x%016llx\n", + qno, CNXK_VF_SDP_R_ERR_TYPE(qno), + octep_vf_read_csr64(oct, CNXK_VF_SDP_R_ERR_TYPE(qno))); +} + +/* Reset Hardware Tx queue */ +static void cnxk_vf_reset_iq(struct octep_vf_device *oct, int q_no) +{ + u64 val = ULL(0); + + dev_dbg(&oct->pdev->dev, "Reset VF IQ-%d\n", q_no); + + /* Disable the Tx/Instruction Ring */ + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_ENABLE(q_no), val); + + /* clear the Instruction Ring packet/byte counts and doorbell CSRs */ + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_INT_LEVELS(q_no), val); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_PKT_CNT(q_no), val); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_BYTE_CNT(q_no), val); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_INSTR_BADDR(q_no), val); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_INSTR_RSIZE(q_no), val); + + val = GENMASK_ULL(31, 0); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_INSTR_DBELL(q_no), val); + + val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_CNTS(q_no)); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_CNTS(q_no), val & GENMASK_ULL(31, 0)); +} + +/* Reset Hardware Rx queue */ +static void cnxk_vf_reset_oq(struct octep_vf_device *oct, int q_no) +{ + u64 val = ULL(0); + + /* Disable Output (Rx) Ring */ + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_ENABLE(q_no), val); + + /* Clear count CSRs */ + val = octep_vf_read_csr(oct, CNXK_VF_SDP_R_OUT_CNTS(q_no)); + octep_vf_write_csr(oct, CNXK_VF_SDP_R_OUT_CNTS(q_no), val); + + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_PKT_CNT(q_no), GENMASK_ULL(35, 0)); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_SLIST_DBELL(q_no), GENMASK_ULL(31, 0)); +} + +/* Reset all hardware Tx/Rx queues */ +static void octep_vf_reset_io_queues_cnxk(struct octep_vf_device *oct) +{ + struct pci_dev *pdev = oct->pdev; + int q; + + dev_dbg(&pdev->dev, "Reset OCTEP_CNXK VF IO Queues\n"); + + for (q = 0; q < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); q++) { + cnxk_vf_reset_iq(oct, q); + cnxk_vf_reset_oq(oct, q); + } +} + +/* Initialize configuration limits and initial active config */ +static void octep_vf_init_config_cnxk_vf(struct octep_vf_device *oct) +{ + struct octep_vf_config *conf = oct->conf; + u64 reg_val; + + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_CONTROL(0)); + conf->ring_cfg.max_io_rings = (reg_val >> CNXK_VF_R_IN_CTL_RPVF_POS) & + CNXK_VF_R_IN_CTL_RPVF_MASK; + conf->ring_cfg.active_io_rings = conf->ring_cfg.max_io_rings; + + conf->iq.num_descs = OCTEP_VF_IQ_MAX_DESCRIPTORS; + conf->iq.instr_type = OCTEP_VF_64BYTE_INSTR; + conf->iq.db_min = OCTEP_VF_DB_MIN; + conf->iq.intr_threshold = OCTEP_VF_IQ_INTR_THRESHOLD; + + conf->oq.num_descs = OCTEP_VF_OQ_MAX_DESCRIPTORS; + conf->oq.buf_size = OCTEP_VF_OQ_BUF_SIZE; + conf->oq.refill_threshold = OCTEP_VF_OQ_REFILL_THRESHOLD; + conf->oq.oq_intr_pkt = OCTEP_VF_OQ_INTR_PKT_THRESHOLD; + conf->oq.oq_intr_time = OCTEP_VF_OQ_INTR_TIME_THRESHOLD; + conf->oq.wmark = OCTEP_VF_OQ_WMARK_MIN; + + conf->msix_cfg.ioq_msix = conf->ring_cfg.active_io_rings; +} + +/* Setup registers for a hardware Tx Queue */ +static void octep_vf_setup_iq_regs_cnxk(struct octep_vf_device *oct, int iq_no) +{ + struct octep_vf_iq *iq = oct->iq[iq_no]; + u32 reset_instr_cnt; + u64 reg_val; + + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_CONTROL(iq_no)); + + /* wait for IDLE to set to 1 */ + if (!(reg_val & CNXK_VF_R_IN_CTL_IDLE)) { + do { + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_CONTROL(iq_no)); + } while (!(reg_val & CNXK_VF_R_IN_CTL_IDLE)); + } + reg_val |= CNXK_VF_R_IN_CTL_RDSIZE; + reg_val |= CNXK_VF_R_IN_CTL_IS_64B; + reg_val |= CNXK_VF_R_IN_CTL_ESR; + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_CONTROL(iq_no), reg_val); + + /* Write the start of the input queue's ring and its size */ + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_INSTR_BADDR(iq_no), iq->desc_ring_dma); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_INSTR_RSIZE(iq_no), iq->max_count); + + /* Remember the doorbell & instruction count register addr for this queue */ + iq->doorbell_reg = oct->mmio.hw_addr + CNXK_VF_SDP_R_IN_INSTR_DBELL(iq_no); + iq->inst_cnt_reg = oct->mmio.hw_addr + CNXK_VF_SDP_R_IN_CNTS(iq_no); + iq->intr_lvl_reg = oct->mmio.hw_addr + CNXK_VF_SDP_R_IN_INT_LEVELS(iq_no); + + /* Store the current instruction counter (used in flush_iq calculation) */ + reset_instr_cnt = readl(iq->inst_cnt_reg); + writel(reset_instr_cnt, iq->inst_cnt_reg); + + /* INTR_THRESHOLD is set to max(FFFFFFFF) to disable the INTR */ + reg_val = CFG_GET_IQ_INTR_THRESHOLD(oct->conf) & GENMASK_ULL(31, 0); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_INT_LEVELS(iq_no), reg_val); +} + +/* Setup registers for a hardware Rx Queue */ +static void octep_vf_setup_oq_regs_cnxk(struct octep_vf_device *oct, int oq_no) +{ + struct octep_vf_oq *oq = oct->oq[oq_no]; + u32 time_threshold = 0; + u64 oq_ctl = ULL(0); + u64 reg_val; + + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_CONTROL(oq_no)); + + /* wait for IDLE to set to 1 */ + if (!(reg_val & CNXK_VF_R_OUT_CTL_IDLE)) { + do { + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_CONTROL(oq_no)); + } while (!(reg_val & CNXK_VF_R_OUT_CTL_IDLE)); + } + + reg_val &= ~(CNXK_VF_R_OUT_CTL_IMODE); + reg_val &= ~(CNXK_VF_R_OUT_CTL_ROR_P); + reg_val &= ~(CNXK_VF_R_OUT_CTL_NSR_P); + reg_val &= ~(CNXK_VF_R_OUT_CTL_ROR_I); + reg_val &= ~(CNXK_VF_R_OUT_CTL_NSR_I); + reg_val &= ~(CNXK_VF_R_OUT_CTL_ES_I); + reg_val &= ~(CNXK_VF_R_OUT_CTL_ROR_D); + reg_val &= ~(CNXK_VF_R_OUT_CTL_NSR_D); + reg_val &= ~(CNXK_VF_R_OUT_CTL_ES_D); + reg_val |= (CNXK_VF_R_OUT_CTL_ES_P); + + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_CONTROL(oq_no), reg_val); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_SLIST_BADDR(oq_no), oq->desc_ring_dma); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_SLIST_RSIZE(oq_no), oq->max_count); + + oq_ctl = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_CONTROL(oq_no)); + /* Clear the ISIZE and BSIZE (22-0) */ + oq_ctl &= ~GENMASK_ULL(22, 0); + /* Populate the BSIZE (15-0) */ + oq_ctl |= (oq->buffer_size & GENMASK_ULL(15, 0)); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_CONTROL(oq_no), oq_ctl); + + /* Get the mapped address of the pkt_sent and pkts_credit regs */ + oq->pkts_sent_reg = oct->mmio.hw_addr + CNXK_VF_SDP_R_OUT_CNTS(oq_no); + oq->pkts_credit_reg = oct->mmio.hw_addr + CNXK_VF_SDP_R_OUT_SLIST_DBELL(oq_no); + + time_threshold = CFG_GET_OQ_INTR_TIME(oct->conf); + reg_val = ((u64)time_threshold << 32) | CFG_GET_OQ_INTR_PKT(oct->conf); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_INT_LEVELS(oq_no), reg_val); + + /* set watermark for backpressure */ + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_WMARK(oq_no)); + reg_val &= ~GENMASK_ULL(31, 0); + reg_val |= CFG_GET_OQ_WMARK(oct->conf); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_WMARK(oq_no), reg_val); +} + +/* Setup registers for a VF mailbox */ +static void octep_vf_setup_mbox_regs_cnxk(struct octep_vf_device *oct, int q_no) +{ + struct octep_vf_mbox *mbox = oct->mbox; + + /* PF to VF DATA reg. VF reads from this reg */ + mbox->mbox_read_reg = oct->mmio.hw_addr + CNXK_VF_SDP_R_MBOX_PF_VF_DATA(q_no); + + /* VF mbox interrupt reg */ + mbox->mbox_int_reg = oct->mmio.hw_addr + CNXK_VF_SDP_R_MBOX_PF_VF_INT(q_no); + + /* VF to PF DATA reg. VF writes into this reg */ + mbox->mbox_write_reg = oct->mmio.hw_addr + CNXK_VF_SDP_R_MBOX_VF_PF_DATA(q_no); +} + +/* Mailbox Interrupt handler */ +static void cnxk_handle_vf_mbox_intr(struct octep_vf_device *oct) +{ + if (oct->mbox) + schedule_work(&oct->mbox->wk.work); + else + dev_err(&oct->pdev->dev, "cannot schedule work on invalid mbox\n"); +} + +/* Tx/Rx queue interrupt handler */ +static irqreturn_t octep_vf_ioq_intr_handler_cnxk(void *data) +{ + struct octep_vf_ioq_vector *vector = data; + struct octep_vf_device *oct; + struct octep_vf_oq *oq; + u64 reg_val; + + oct = vector->octep_vf_dev; + oq = vector->oq; + /* Mailbox interrupt arrives along with interrupt of tx/rx ring pair 0 */ + if (oq->q_no == 0) { + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_MBOX_PF_VF_INT(0)); + if (reg_val & CNXK_VF_SDP_R_MBOX_PF_VF_INT_STATUS) { + cnxk_handle_vf_mbox_intr(oct); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_MBOX_PF_VF_INT(0), reg_val); + } + } + napi_schedule_irqoff(oq->napi); + return IRQ_HANDLED; +} + +/* Re-initialize Octeon hardware registers */ +static void octep_vf_reinit_regs_cnxk(struct octep_vf_device *oct) +{ + u32 i; + + for (i = 0; i < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); i++) + oct->hw_ops.setup_iq_regs(oct, i); + + for (i = 0; i < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); i++) + oct->hw_ops.setup_oq_regs(oct, i); + + oct->hw_ops.enable_interrupts(oct); + oct->hw_ops.enable_io_queues(oct); + + for (i = 0; i < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); i++) + writel(oct->oq[i]->max_count, oct->oq[i]->pkts_credit_reg); +} + +/* Enable all interrupts */ +static void octep_vf_enable_interrupts_cnxk(struct octep_vf_device *oct) +{ + int num_rings, q; + u64 reg_val; + + num_rings = CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); + for (q = 0; q < num_rings; q++) { + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_INT_LEVELS(q)); + reg_val |= BIT_ULL_MASK(62); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_INT_LEVELS(q), reg_val); + + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_INT_LEVELS(q)); + reg_val |= BIT_ULL_MASK(62); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_INT_LEVELS(q), reg_val); + } + /* Enable PF to VF mbox interrupt by setting 2nd bit*/ + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_MBOX_PF_VF_INT(0), + CNXK_VF_SDP_R_MBOX_PF_VF_INT_ENAB); +} + +/* Disable all interrupts */ +static void octep_vf_disable_interrupts_cnxk(struct octep_vf_device *oct) +{ + int num_rings, q; + u64 reg_val; + + /* Disable PF to VF mbox interrupt by setting 2nd bit*/ + if (oct->mbox) + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_MBOX_PF_VF_INT(0), 0x0); + + num_rings = CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); + for (q = 0; q < num_rings; q++) { + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_INT_LEVELS(q)); + reg_val &= ~BIT_ULL_MASK(62); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_INT_LEVELS(q), reg_val); + + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_INT_LEVELS(q)); + reg_val &= ~BIT_ULL_MASK(62); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_INT_LEVELS(q), reg_val); + } +} + +/* Get new Octeon Read Index: index of descriptor that Octeon reads next. */ +static u32 octep_vf_update_iq_read_index_cnxk(struct octep_vf_iq *iq) +{ + u32 pkt_in_done = readl(iq->inst_cnt_reg); + u32 last_done, new_idx; + + last_done = pkt_in_done - iq->pkt_in_done; + iq->pkt_in_done = pkt_in_done; + + new_idx = (iq->octep_vf_read_index + last_done) % iq->max_count; + + return new_idx; +} + +/* Enable a hardware Tx Queue */ +static void octep_vf_enable_iq_cnxk(struct octep_vf_device *oct, int iq_no) +{ + u64 loop = HZ; + u64 reg_val; + + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_INSTR_DBELL(iq_no), GENMASK_ULL(31, 0)); + + while (octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_INSTR_DBELL(iq_no)) && + loop--) { + schedule_timeout_interruptible(1); + } + + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_INT_LEVELS(iq_no)); + reg_val |= BIT_ULL_MASK(62); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_INT_LEVELS(iq_no), reg_val); + + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_ENABLE(iq_no)); + reg_val |= ULL(1); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_ENABLE(iq_no), reg_val); +} + +/* Enable a hardware Rx Queue */ +static void octep_vf_enable_oq_cnxk(struct octep_vf_device *oct, int oq_no) +{ + u64 reg_val; + + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_INT_LEVELS(oq_no)); + reg_val |= BIT_ULL_MASK(62); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_INT_LEVELS(oq_no), reg_val); + + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_SLIST_DBELL(oq_no), GENMASK_ULL(31, 0)); + + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_ENABLE(oq_no)); + reg_val |= ULL(1); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_ENABLE(oq_no), reg_val); +} + +/* Enable all hardware Tx/Rx Queues assigned to VF */ +static void octep_vf_enable_io_queues_cnxk(struct octep_vf_device *oct) +{ + u8 q; + + for (q = 0; q < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); q++) { + octep_vf_enable_iq_cnxk(oct, q); + octep_vf_enable_oq_cnxk(oct, q); + } +} + +/* Disable a hardware Tx Queue assigned to VF */ +static void octep_vf_disable_iq_cnxk(struct octep_vf_device *oct, int iq_no) +{ + u64 reg_val; + + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_IN_ENABLE(iq_no)); + reg_val &= ~ULL(1); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_IN_ENABLE(iq_no), reg_val); +} + +/* Disable a hardware Rx Queue assigned to VF */ +static void octep_vf_disable_oq_cnxk(struct octep_vf_device *oct, int oq_no) +{ + u64 reg_val; + + reg_val = octep_vf_read_csr64(oct, CNXK_VF_SDP_R_OUT_ENABLE(oq_no)); + reg_val &= ~ULL(1); + octep_vf_write_csr64(oct, CNXK_VF_SDP_R_OUT_ENABLE(oq_no), reg_val); +} + +/* Disable all hardware Tx/Rx Queues assigned to VF */ +static void octep_vf_disable_io_queues_cnxk(struct octep_vf_device *oct) +{ + int q; + + for (q = 0; q < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); q++) { + octep_vf_disable_iq_cnxk(oct, q); + octep_vf_disable_oq_cnxk(oct, q); + } +} + +/* Dump hardware registers (including Tx/Rx queues) for debugging. */ +static void octep_vf_dump_registers_cnxk(struct octep_vf_device *oct) +{ + u8 num_rings, q; + + num_rings = CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); + for (q = 0; q < num_rings; q++) + cnxk_vf_dump_q_regs(oct, q); +} + +/** + * octep_vf_device_setup_cnxk() - Setup Octeon device. + * + * @oct: Octeon device private data structure. + * + * - initialize hardware operations. + * - get target side pcie port number for the device. + * - set initial configuration and max limits. + */ +void octep_vf_device_setup_cnxk(struct octep_vf_device *oct) +{ + oct->hw_ops.setup_iq_regs = octep_vf_setup_iq_regs_cnxk; + oct->hw_ops.setup_oq_regs = octep_vf_setup_oq_regs_cnxk; + oct->hw_ops.setup_mbox_regs = octep_vf_setup_mbox_regs_cnxk; + + oct->hw_ops.ioq_intr_handler = octep_vf_ioq_intr_handler_cnxk; + oct->hw_ops.reinit_regs = octep_vf_reinit_regs_cnxk; + + oct->hw_ops.enable_interrupts = octep_vf_enable_interrupts_cnxk; + oct->hw_ops.disable_interrupts = octep_vf_disable_interrupts_cnxk; + + oct->hw_ops.update_iq_read_idx = octep_vf_update_iq_read_index_cnxk; + + oct->hw_ops.enable_iq = octep_vf_enable_iq_cnxk; + oct->hw_ops.enable_oq = octep_vf_enable_oq_cnxk; + oct->hw_ops.enable_io_queues = octep_vf_enable_io_queues_cnxk; + + oct->hw_ops.disable_iq = octep_vf_disable_iq_cnxk; + oct->hw_ops.disable_oq = octep_vf_disable_oq_cnxk; + oct->hw_ops.disable_io_queues = octep_vf_disable_io_queues_cnxk; + oct->hw_ops.reset_io_queues = octep_vf_reset_io_queues_cnxk; + + oct->hw_ops.dump_registers = octep_vf_dump_registers_cnxk; + octep_vf_init_config_cnxk_vf(oct); +} diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_config.h b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_config.h new file mode 100644 index 000000000000..e03a647b0110 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_config.h @@ -0,0 +1,160 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ + +#ifndef _OCTEP_VF_CONFIG_H_ +#define _OCTEP_VF_CONFIG_H_ + +/* Tx instruction types by length */ +#define OCTEP_VF_32BYTE_INSTR 32 +#define OCTEP_VF_64BYTE_INSTR 64 + +/* Tx Queue: maximum descriptors per ring */ +#define OCTEP_VF_IQ_MAX_DESCRIPTORS 1024 +/* Minimum input (Tx) requests to be enqueued to ring doorbell */ +#define OCTEP_VF_DB_MIN 8 +/* Packet threshold for Tx queue interrupt */ +#define OCTEP_VF_IQ_INTR_THRESHOLD 0x0 + +/* Minimum watermark for backpressure */ +#define OCTEP_VF_OQ_WMARK_MIN 256 + +/* Rx Queue: maximum descriptors per ring */ +#define OCTEP_VF_OQ_MAX_DESCRIPTORS 1024 + +/* Rx buffer size: Use page size buffers. + * Build skb from allocated page buffer once the packet is received. + * When a gathered packet is received, make head page as skb head and + * page buffers in consecutive Rx descriptors as fragments. + */ +#define OCTEP_VF_OQ_BUF_SIZE (SKB_WITH_OVERHEAD(PAGE_SIZE)) +#define OCTEP_VF_OQ_PKTS_PER_INTR 128 +#define OCTEP_VF_OQ_REFILL_THRESHOLD (OCTEP_VF_OQ_MAX_DESCRIPTORS / 4) + +#define OCTEP_VF_OQ_INTR_PKT_THRESHOLD 1 +#define OCTEP_VF_OQ_INTR_TIME_THRESHOLD 10 + +#define OCTEP_VF_MSIX_NAME_SIZE (IFNAMSIZ + 32) + +/* Tx Queue wake threshold + * wakeup a stopped Tx queue if minimum 2 descriptors are available. + * Even a skb with fragments consume only one Tx queue descriptor entry. + */ +#define OCTEP_VF_WAKE_QUEUE_THRESHOLD 2 + +/* Minimum MTU supported by Octeon network interface */ +#define OCTEP_VF_MIN_MTU ETH_MIN_MTU +/* Maximum MTU supported by Octeon interface*/ +#define OCTEP_VF_MAX_MTU (10000 - (ETH_HLEN + ETH_FCS_LEN)) +/* Default MTU */ +#define OCTEP_VF_DEFAULT_MTU 1500 + +/* Macros to get octeon config params */ +#define CFG_GET_IQ_CFG(cfg) ((cfg)->iq) +#define CFG_GET_IQ_NUM_DESC(cfg) ((cfg)->iq.num_descs) +#define CFG_GET_IQ_INSTR_TYPE(cfg) ((cfg)->iq.instr_type) +#define CFG_GET_IQ_INSTR_SIZE(cfg) (64) +#define CFG_GET_IQ_DB_MIN(cfg) ((cfg)->iq.db_min) +#define CFG_GET_IQ_INTR_THRESHOLD(cfg) ((cfg)->iq.intr_threshold) + +#define CFG_GET_OQ_NUM_DESC(cfg) ((cfg)->oq.num_descs) +#define CFG_GET_OQ_BUF_SIZE(cfg) ((cfg)->oq.buf_size) +#define CFG_GET_OQ_REFILL_THRESHOLD(cfg) ((cfg)->oq.refill_threshold) +#define CFG_GET_OQ_INTR_PKT(cfg) ((cfg)->oq.oq_intr_pkt) +#define CFG_GET_OQ_INTR_TIME(cfg) ((cfg)->oq.oq_intr_time) +#define CFG_GET_OQ_WMARK(cfg) ((cfg)->oq.wmark) + +#define CFG_GET_PORTS_ACTIVE_IO_RINGS(cfg) ((cfg)->ring_cfg.active_io_rings) +#define CFG_GET_PORTS_MAX_IO_RINGS(cfg) ((cfg)->ring_cfg.max_io_rings) + +#define CFG_GET_CORE_TICS_PER_US(cfg) ((cfg)->core_cfg.core_tics_per_us) +#define CFG_GET_COPROC_TICS_PER_US(cfg) ((cfg)->core_cfg.coproc_tics_per_us) + +#define CFG_GET_IOQ_MSIX(cfg) ((cfg)->msix_cfg.ioq_msix) + +/* Hardware Tx Queue configuration. */ +struct octep_vf_iq_config { + /* Size of the Input queue (number of commands) */ + u16 num_descs; + + /* Command size - 32 or 64 bytes */ + u16 instr_type; + + /* Minimum number of commands pending to be posted to Octeon before driver + * hits the Input queue doorbell. + */ + u16 db_min; + + /* Trigger the IQ interrupt when processed cmd count reaches + * this level. + */ + u32 intr_threshold; +}; + +/* Hardware Rx Queue configuration. */ +struct octep_vf_oq_config { + /* Size of Output queue (number of descriptors) */ + u16 num_descs; + + /* Size of buffer in this Output queue. */ + u16 buf_size; + + /* The number of buffers that were consumed during packet processing + * by the driver on this Output queue before the driver attempts to + * replenish the descriptor ring with new buffers. + */ + u16 refill_threshold; + + /* Interrupt Coalescing (Packet Count). Octeon will interrupt the host + * only if it sent as many packets as specified by this field. + * The driver usually does not use packet count interrupt coalescing. + */ + u32 oq_intr_pkt; + + /* Interrupt Coalescing (Time Interval). Octeon will interrupt the host + * if at least one packet was sent in the time interval specified by + * this field. The driver uses time interval interrupt coalescing by + * default. The time is specified in microseconds. + */ + u32 oq_intr_time; + + /* Water mark for backpressure. + * Output queue sends backpressure signal to source when + * free buffer count falls below wmark. + */ + u32 wmark; +}; + +/* Tx/Rx configuration */ +struct octep_vf_ring_config { + /* Max number of IOQs */ + u16 max_io_rings; + + /* Number of active IOQs */ + u16 active_io_rings; +}; + +/* Octeon MSI-x config. */ +struct octep_vf_msix_config { + /* Number of IOQ interrupts */ + u16 ioq_msix; +}; + +/* Data Structure to hold configuration limits and active config */ +struct octep_vf_config { + /* Input Queue attributes. */ + struct octep_vf_iq_config iq; + + /* Output Queue attributes. */ + struct octep_vf_oq_config oq; + + /* MSI-X interrupt config */ + struct octep_vf_msix_config msix_cfg; + + /* NIC VF ring Configuration */ + struct octep_vf_ring_config ring_cfg; +}; +#endif /* _OCTEP_VF_CONFIG_H_ */ diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_ethtool.c b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_ethtool.c new file mode 100644 index 000000000000..a1979b45e355 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_ethtool.c @@ -0,0 +1,273 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ + +#include <linux/pci.h> +#include <linux/netdevice.h> +#include <linux/ethtool.h> + +#include "octep_vf_config.h" +#include "octep_vf_main.h" + +static const char octep_vf_gstrings_global_stats[][ETH_GSTRING_LEN] = { + "rx_alloc_errors", + "tx_busy_errors", + "tx_hw_pkts", + "tx_hw_octs", + "tx_hw_bcast", + "tx_hw_mcast", + "rx_hw_pkts", + "rx_hw_bytes", + "rx_hw_bcast", + "rx_dropped_bytes_fifo_full", +}; + +#define OCTEP_VF_GLOBAL_STATS_CNT (sizeof(octep_vf_gstrings_global_stats) / ETH_GSTRING_LEN) + +static const char octep_vf_gstrings_tx_q_stats[][ETH_GSTRING_LEN] = { + "tx_packets_posted[Q-%u]", + "tx_packets_completed[Q-%u]", + "tx_bytes[Q-%u]", + "tx_busy[Q-%u]", +}; + +#define OCTEP_VF_TX_Q_STATS_CNT (sizeof(octep_vf_gstrings_tx_q_stats) / ETH_GSTRING_LEN) + +static const char octep_vf_gstrings_rx_q_stats[][ETH_GSTRING_LEN] = { + "rx_packets[Q-%u]", + "rx_bytes[Q-%u]", + "rx_alloc_errors[Q-%u]", +}; + +#define OCTEP_VF_RX_Q_STATS_CNT (sizeof(octep_vf_gstrings_rx_q_stats) / ETH_GSTRING_LEN) + +static void octep_vf_get_drvinfo(struct net_device *netdev, + struct ethtool_drvinfo *info) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + + strscpy(info->driver, OCTEP_VF_DRV_NAME, sizeof(info->driver)); + strscpy(info->bus_info, pci_name(oct->pdev), sizeof(info->bus_info)); +} + +static void octep_vf_get_strings(struct net_device *netdev, + u32 stringset, u8 *data) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + u16 num_queues = CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); + char *strings = (char *)data; + int i, j; + + switch (stringset) { + case ETH_SS_STATS: + for (i = 0; i < OCTEP_VF_GLOBAL_STATS_CNT; i++) { + snprintf(strings, ETH_GSTRING_LEN, + octep_vf_gstrings_global_stats[i]); + strings += ETH_GSTRING_LEN; + } + + for (i = 0; i < num_queues; i++) { + for (j = 0; j < OCTEP_VF_TX_Q_STATS_CNT; j++) { + snprintf(strings, ETH_GSTRING_LEN, + octep_vf_gstrings_tx_q_stats[j], i); + strings += ETH_GSTRING_LEN; + } + } + + for (i = 0; i < num_queues; i++) { + for (j = 0; j < OCTEP_VF_RX_Q_STATS_CNT; j++) { + snprintf(strings, ETH_GSTRING_LEN, + octep_vf_gstrings_rx_q_stats[j], i); + strings += ETH_GSTRING_LEN; + } + } + break; + default: + break; + } +} + +static int octep_vf_get_sset_count(struct net_device *netdev, int sset) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + u16 num_queues = CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); + + switch (sset) { + case ETH_SS_STATS: + return OCTEP_VF_GLOBAL_STATS_CNT + (num_queues * + (OCTEP_VF_TX_Q_STATS_CNT + OCTEP_VF_RX_Q_STATS_CNT)); + break; + default: + return -EOPNOTSUPP; + } +} + +static void octep_vf_get_ethtool_stats(struct net_device *netdev, + struct ethtool_stats *stats, u64 *data) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + struct octep_vf_iface_tx_stats *iface_tx_stats; + struct octep_vf_iface_rx_stats *iface_rx_stats; + u64 rx_alloc_errors, tx_busy_errors; + int q, i; + + rx_alloc_errors = 0; + tx_busy_errors = 0; + + octep_vf_get_if_stats(oct); + iface_tx_stats = &oct->iface_tx_stats; + iface_rx_stats = &oct->iface_rx_stats; + + for (q = 0; q < oct->num_oqs; q++) { + struct octep_vf_iq *iq = oct->iq[q]; + struct octep_vf_oq *oq = oct->oq[q]; + + tx_busy_errors += iq->stats.tx_busy; + rx_alloc_errors += oq->stats.alloc_failures; + } + i = 0; + data[i++] = rx_alloc_errors; + data[i++] = tx_busy_errors; + data[i++] = iface_tx_stats->pkts; + data[i++] = iface_tx_stats->octs; + data[i++] = iface_tx_stats->bcst; + data[i++] = iface_tx_stats->mcst; + data[i++] = iface_rx_stats->pkts; + data[i++] = iface_rx_stats->octets; + data[i++] = iface_rx_stats->bcast_pkts; + data[i++] = iface_rx_stats->dropped_octets_fifo_full; + + /* Per Tx Queue stats */ + for (q = 0; q < oct->num_iqs; q++) { + struct octep_vf_iq *iq = oct->iq[q]; + + data[i++] = iq->stats.instr_posted; + data[i++] = iq->stats.instr_completed; + data[i++] = iq->stats.bytes_sent; + data[i++] = iq->stats.tx_busy; + } + + /* Per Rx Queue stats */ + for (q = 0; q < oct->num_oqs; q++) { + struct octep_vf_oq *oq = oct->oq[q]; + + data[i++] = oq->stats.packets; + data[i++] = oq->stats.bytes; + data[i++] = oq->stats.alloc_failures; + } +} + +#define OCTEP_VF_SET_ETHTOOL_LINK_MODES_BITMAP(octep_vf_speeds, ksettings, name) \ +{ \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_10GBASE_T)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 10000baseT_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_10GBASE_R)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 10000baseR_FEC); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_10GBASE_CR)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 10000baseCR_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_10GBASE_KR)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 10000baseKR_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_10GBASE_LR)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 10000baseLR_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_10GBASE_SR)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 10000baseSR_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_25GBASE_CR)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 25000baseCR_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_25GBASE_KR)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 25000baseKR_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_25GBASE_SR)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 25000baseSR_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_40GBASE_CR4)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 40000baseCR4_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_40GBASE_KR4)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 40000baseKR4_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_40GBASE_LR4)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 40000baseLR4_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_40GBASE_SR4)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 40000baseSR4_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_50GBASE_CR2)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 50000baseCR2_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_50GBASE_KR2)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 50000baseKR2_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_50GBASE_SR2)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 50000baseSR2_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_50GBASE_CR)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 50000baseCR_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_50GBASE_KR)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 50000baseKR_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_50GBASE_LR)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 50000baseLR_ER_FR_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_50GBASE_SR)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 50000baseSR_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_100GBASE_CR4)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 100000baseCR4_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_100GBASE_KR4)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 100000baseKR4_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_100GBASE_LR4)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 100000baseLR4_ER4_Full); \ + if ((octep_vf_speeds) & BIT(OCTEP_VF_LINK_MODE_100GBASE_SR4)) \ + ethtool_link_ksettings_add_link_mode(ksettings, name, 100000baseSR4_Full); \ +} + +static int octep_vf_get_link_ksettings(struct net_device *netdev, + struct ethtool_link_ksettings *cmd) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + struct octep_vf_iface_link_info *link_info; + u32 advertised_modes, supported_modes; + + ethtool_link_ksettings_zero_link_mode(cmd, supported); + ethtool_link_ksettings_zero_link_mode(cmd, advertising); + + octep_vf_get_link_info(oct); + + advertised_modes = oct->link_info.advertised_modes; + supported_modes = oct->link_info.supported_modes; + link_info = &oct->link_info; + + OCTEP_VF_SET_ETHTOOL_LINK_MODES_BITMAP(supported_modes, cmd, supported); + OCTEP_VF_SET_ETHTOOL_LINK_MODES_BITMAP(advertised_modes, cmd, advertising); + + if (link_info->autoneg) { + if (link_info->autoneg & OCTEP_VF_LINK_MODE_AUTONEG_SUPPORTED) + ethtool_link_ksettings_add_link_mode(cmd, supported, Autoneg); + if (link_info->autoneg & OCTEP_VF_LINK_MODE_AUTONEG_ADVERTISED) { + ethtool_link_ksettings_add_link_mode(cmd, advertising, Autoneg); + cmd->base.autoneg = AUTONEG_ENABLE; + } else { + cmd->base.autoneg = AUTONEG_DISABLE; + } + } else { + cmd->base.autoneg = AUTONEG_DISABLE; + } + + cmd->base.port = PORT_FIBRE; + ethtool_link_ksettings_add_link_mode(cmd, supported, FIBRE); + ethtool_link_ksettings_add_link_mode(cmd, advertising, FIBRE); + + if (netif_carrier_ok(netdev)) { + cmd->base.speed = link_info->speed; + cmd->base.duplex = DUPLEX_FULL; + } else { + cmd->base.speed = SPEED_UNKNOWN; + cmd->base.duplex = DUPLEX_UNKNOWN; + } + return 0; +} + +static const struct ethtool_ops octep_vf_ethtool_ops = { + .get_drvinfo = octep_vf_get_drvinfo, + .get_link = ethtool_op_get_link, + .get_strings = octep_vf_get_strings, + .get_sset_count = octep_vf_get_sset_count, + .get_ethtool_stats = octep_vf_get_ethtool_stats, + .get_link_ksettings = octep_vf_get_link_ksettings, +}; + +void octep_vf_set_ethtool_ops(struct net_device *netdev) +{ + netdev->ethtool_ops = &octep_vf_ethtool_ops; +} diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c new file mode 100644 index 000000000000..dd49d0b8b494 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.c @@ -0,0 +1,1231 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ + +#include <linux/types.h> +#include <linux/module.h> +#include <linux/pci.h> +#include <linux/aer.h> +#include <linux/netdevice.h> +#include <linux/etherdevice.h> +#include <linux/rtnetlink.h> +#include <linux/vmalloc.h> +#include <net/netdev_queues.h> + +#include "octep_vf_config.h" +#include "octep_vf_main.h" + +struct workqueue_struct *octep_vf_wq; + +/* Supported Devices */ +static const struct pci_device_id octep_vf_pci_id_tbl[] = { + {PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, OCTEP_PCI_DEVICE_ID_CN93_VF)}, + {PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, OCTEP_PCI_DEVICE_ID_CNF95N_VF)}, + {PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, OCTEP_PCI_DEVICE_ID_CN98_VF)}, + {PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, OCTEP_PCI_DEVICE_ID_CN10KA_VF)}, + {PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, OCTEP_PCI_DEVICE_ID_CNF10KA_VF)}, + {PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, OCTEP_PCI_DEVICE_ID_CNF10KB_VF)}, + {PCI_DEVICE(PCI_VENDOR_ID_CAVIUM, OCTEP_PCI_DEVICE_ID_CN10KB_VF)}, + {0, }, +}; +MODULE_DEVICE_TABLE(pci, octep_vf_pci_id_tbl); + +MODULE_AUTHOR("Veerasenareddy Burru <vburru@marvell.com>"); +MODULE_DESCRIPTION(OCTEP_VF_DRV_STRING); +MODULE_LICENSE("GPL"); + +/** + * octep_vf_alloc_ioq_vectors() - Allocate Tx/Rx Queue interrupt info. + * + * @oct: Octeon device private data structure. + * + * Allocate resources to hold per Tx/Rx queue interrupt info. + * This is the information passed to interrupt handler, from which napi poll + * is scheduled and includes quick access to private data of Tx/Rx queue + * corresponding to the interrupt being handled. + * + * Return: 0, on successful allocation of resources for all queue interrupts. + * -1, if failed to allocate any resource. + */ +static int octep_vf_alloc_ioq_vectors(struct octep_vf_device *oct) +{ + struct octep_vf_ioq_vector *ioq_vector; + int i; + + for (i = 0; i < oct->num_oqs; i++) { + oct->ioq_vector[i] = vzalloc(sizeof(*oct->ioq_vector[i])); + if (!oct->ioq_vector[i]) + goto free_ioq_vector; + + ioq_vector = oct->ioq_vector[i]; + ioq_vector->iq = oct->iq[i]; + ioq_vector->oq = oct->oq[i]; + ioq_vector->octep_vf_dev = oct; + } + + dev_info(&oct->pdev->dev, "Allocated %d IOQ vectors\n", oct->num_oqs); + return 0; + +free_ioq_vector: + while (i) { + i--; + vfree(oct->ioq_vector[i]); + oct->ioq_vector[i] = NULL; + } + return -1; +} + +/** + * octep_vf_free_ioq_vectors() - Free Tx/Rx Queue interrupt vector info. + * + * @oct: Octeon device private data structure. + */ +static void octep_vf_free_ioq_vectors(struct octep_vf_device *oct) +{ + int i; + + for (i = 0; i < oct->num_oqs; i++) { + if (oct->ioq_vector[i]) { + vfree(oct->ioq_vector[i]); + oct->ioq_vector[i] = NULL; + } + } + netdev_info(oct->netdev, "Freed IOQ Vectors\n"); +} + +/** + * octep_vf_enable_msix_range() - enable MSI-x interrupts. + * + * @oct: Octeon device private data structure. + * + * Allocate and enable all MSI-x interrupts (queue and non-queue interrupts) + * for the Octeon device. + * + * Return: 0, on successfully enabling all MSI-x interrupts. + * -1, if failed to enable any MSI-x interrupt. + */ +static int octep_vf_enable_msix_range(struct octep_vf_device *oct) +{ + int num_msix, msix_allocated; + int i; + + /* Generic interrupts apart from input/output queues */ + //num_msix = oct->num_oqs + CFG_GET_NON_IOQ_MSIX(oct->conf); + num_msix = oct->num_oqs; + oct->msix_entries = kcalloc(num_msix, sizeof(struct msix_entry), GFP_KERNEL); + if (!oct->msix_entries) + goto msix_alloc_err; + + for (i = 0; i < num_msix; i++) + oct->msix_entries[i].entry = i; + + msix_allocated = pci_enable_msix_range(oct->pdev, oct->msix_entries, + num_msix, num_msix); + if (msix_allocated != num_msix) { + dev_err(&oct->pdev->dev, + "Failed to enable %d msix irqs; got only %d\n", + num_msix, msix_allocated); + goto enable_msix_err; + } + oct->num_irqs = msix_allocated; + dev_info(&oct->pdev->dev, "MSI-X enabled successfully\n"); + + return 0; + +enable_msix_err: + if (msix_allocated > 0) + pci_disable_msix(oct->pdev); + kfree(oct->msix_entries); + oct->msix_entries = NULL; +msix_alloc_err: + return -1; +} + +/** + * octep_vf_disable_msix() - disable MSI-x interrupts. + * + * @oct: Octeon device private data structure. + * + * Disable MSI-x on the Octeon device. + */ +static void octep_vf_disable_msix(struct octep_vf_device *oct) +{ + pci_disable_msix(oct->pdev); + kfree(oct->msix_entries); + oct->msix_entries = NULL; + dev_info(&oct->pdev->dev, "Disabled MSI-X\n"); +} + +/** + * octep_vf_ioq_intr_handler() - handler for all Tx/Rx queue interrupts. + * + * @irq: Interrupt number. + * @data: interrupt data contains pointers to Tx/Rx queue private data + * and correspong NAPI context. + * + * this is common handler for all non-queue (generic) interrupts. + */ +static irqreturn_t octep_vf_ioq_intr_handler(int irq, void *data) +{ + struct octep_vf_ioq_vector *ioq_vector = data; + struct octep_vf_device *oct = ioq_vector->octep_vf_dev; + + return oct->hw_ops.ioq_intr_handler(ioq_vector); +} + +/** + * octep_vf_request_irqs() - Register interrupt handlers. + * + * @oct: Octeon device private data structure. + * + * Register handlers for all queue and non-queue interrupts. + * + * Return: 0, on successful registration of all interrupt handlers. + * -1, on any error. + */ +static int octep_vf_request_irqs(struct octep_vf_device *oct) +{ + struct net_device *netdev = oct->netdev; + struct octep_vf_ioq_vector *ioq_vector; + struct msix_entry *msix_entry; + int ret, i; + + /* Request IRQs for Tx/Rx queues */ + for (i = 0; i < oct->num_oqs; i++) { + ioq_vector = oct->ioq_vector[i]; + msix_entry = &oct->msix_entries[i]; + + snprintf(ioq_vector->name, sizeof(ioq_vector->name), + "%s-q%d", netdev->name, i); + ret = request_irq(msix_entry->vector, + octep_vf_ioq_intr_handler, 0, + ioq_vector->name, ioq_vector); + if (ret) { + netdev_err(netdev, + "request_irq failed for Q-%d; err=%d", + i, ret); + goto ioq_irq_err; + } + + cpumask_set_cpu(i % num_online_cpus(), + &ioq_vector->affinity_mask); + irq_set_affinity_hint(msix_entry->vector, + &ioq_vector->affinity_mask); + } + + return 0; +ioq_irq_err: + while (i) { + --i; + free_irq(oct->msix_entries[i].vector, oct); + } + return -1; +} + +/** + * octep_vf_free_irqs() - free all registered interrupts. + * + * @oct: Octeon device private data structure. + * + * Free all queue and non-queue interrupts of the Octeon device. + */ +static void octep_vf_free_irqs(struct octep_vf_device *oct) +{ + int i; + + for (i = 0; i < oct->num_irqs; i++) { + irq_set_affinity_hint(oct->msix_entries[i].vector, NULL); + free_irq(oct->msix_entries[i].vector, oct->ioq_vector[i]); + } + netdev_info(oct->netdev, "IRQs freed\n"); +} + +/** + * octep_vf_setup_irqs() - setup interrupts for the Octeon device. + * + * @oct: Octeon device private data structure. + * + * Allocate data structures to hold per interrupt information, allocate/enable + * MSI-x interrupt and register interrupt handlers. + * + * Return: 0, on successful allocation and registration of all interrupts. + * -1, on any error. + */ +static int octep_vf_setup_irqs(struct octep_vf_device *oct) +{ + if (octep_vf_alloc_ioq_vectors(oct)) + goto ioq_vector_err; + + if (octep_vf_enable_msix_range(oct)) + goto enable_msix_err; + + if (octep_vf_request_irqs(oct)) + goto request_irq_err; + + return 0; + +request_irq_err: + octep_vf_disable_msix(oct); +enable_msix_err: + octep_vf_free_ioq_vectors(oct); +ioq_vector_err: + return -1; +} + +/** + * octep_vf_clean_irqs() - free all interrupts and its resources. + * + * @oct: Octeon device private data structure. + */ +static void octep_vf_clean_irqs(struct octep_vf_device *oct) +{ + octep_vf_free_irqs(oct); + octep_vf_disable_msix(oct); + octep_vf_free_ioq_vectors(oct); +} + +/** + * octep_vf_enable_ioq_irq() - Enable MSI-x interrupt of a Tx/Rx queue. + * + * @iq: Octeon Tx queue data structure. + * @oq: Octeon Rx queue data structure. + */ +static void octep_vf_enable_ioq_irq(struct octep_vf_iq *iq, struct octep_vf_oq *oq) +{ + u32 pkts_pend = oq->pkts_pending; + + netdev_dbg(iq->netdev, "enabling intr for Q-%u\n", iq->q_no); + if (iq->pkts_processed) { + writel(iq->pkts_processed, iq->inst_cnt_reg); + iq->pkt_in_done -= iq->pkts_processed; + iq->pkts_processed = 0; + } + if (oq->last_pkt_count - pkts_pend) { + writel(oq->last_pkt_count - pkts_pend, oq->pkts_sent_reg); + oq->last_pkt_count = pkts_pend; + } + + /* Flush the previous wrties before writing to RESEND bit */ + smp_wmb(); + writeq(1UL << OCTEP_VF_OQ_INTR_RESEND_BIT, oq->pkts_sent_reg); + writeq(1UL << OCTEP_VF_IQ_INTR_RESEND_BIT, iq->inst_cnt_reg); +} + +/** + * octep_vf_napi_poll() - NAPI poll function for Tx/Rx. + * + * @napi: pointer to napi context. + * @budget: max number of packets to be processed in single invocation. + */ +static int octep_vf_napi_poll(struct napi_struct *napi, int budget) +{ + struct octep_vf_ioq_vector *ioq_vector = + container_of(napi, struct octep_vf_ioq_vector, napi); + u32 tx_pending, rx_done; + + tx_pending = octep_vf_iq_process_completions(ioq_vector->iq, 64); + rx_done = octep_vf_oq_process_rx(ioq_vector->oq, budget); + + /* need more polling if tx completion processing is still pending or + * processed at least 'budget' number of rx packets. + */ + if (tx_pending || rx_done >= budget) + return budget; + + if (likely(napi_complete_done(napi, rx_done))) + octep_vf_enable_ioq_irq(ioq_vector->iq, ioq_vector->oq); + + return rx_done; +} + +/** + * octep_vf_napi_add() - Add NAPI poll for all Tx/Rx queues. + * + * @oct: Octeon device private data structure. + */ +static void octep_vf_napi_add(struct octep_vf_device *oct) +{ + int i; + + for (i = 0; i < oct->num_oqs; i++) { + netdev_dbg(oct->netdev, "Adding NAPI on Q-%d\n", i); + netif_napi_add(oct->netdev, &oct->ioq_vector[i]->napi, octep_vf_napi_poll); + oct->oq[i]->napi = &oct->ioq_vector[i]->napi; + } +} + +/** + * octep_vf_napi_delete() - delete NAPI poll callback for all Tx/Rx queues. + * + * @oct: Octeon device private data structure. + */ +static void octep_vf_napi_delete(struct octep_vf_device *oct) +{ + int i; + + for (i = 0; i < oct->num_oqs; i++) { + netdev_dbg(oct->netdev, "Deleting NAPI on Q-%d\n", i); + netif_napi_del(&oct->ioq_vector[i]->napi); + oct->oq[i]->napi = NULL; + } +} + +/** + * octep_vf_napi_enable() - enable NAPI for all Tx/Rx queues. + * + * @oct: Octeon device private data structure. + */ +static void octep_vf_napi_enable(struct octep_vf_device *oct) +{ + int i; + + for (i = 0; i < oct->num_oqs; i++) { + netdev_dbg(oct->netdev, "Enabling NAPI on Q-%d\n", i); + napi_enable(&oct->ioq_vector[i]->napi); + } +} + +/** + * octep_vf_napi_disable() - disable NAPI for all Tx/Rx queues. + * + * @oct: Octeon device private data structure. + */ +static void octep_vf_napi_disable(struct octep_vf_device *oct) +{ + int i; + + for (i = 0; i < oct->num_oqs; i++) { + netdev_dbg(oct->netdev, "Disabling NAPI on Q-%d\n", i); + napi_disable(&oct->ioq_vector[i]->napi); + } +} + +static void octep_vf_link_up(struct net_device *netdev) +{ + netif_carrier_on(netdev); + netif_tx_start_all_queues(netdev); +} + +static void octep_vf_set_rx_state(struct octep_vf_device *oct, bool up) +{ + int err; + + err = octep_vf_mbox_set_rx_state(oct, up); + if (err) + netdev_err(oct->netdev, "Set Rx state to %d failed with err:%d\n", up, err); +} + +static int octep_vf_get_link_status(struct octep_vf_device *oct) +{ + int err; + + err = octep_vf_mbox_get_link_status(oct, &oct->link_info.oper_up); + if (err) + netdev_err(oct->netdev, "Get link status failed with err:%d\n", err); + return oct->link_info.oper_up; +} + +static void octep_vf_set_link_status(struct octep_vf_device *oct, bool up) +{ + int err; + + err = octep_vf_mbox_set_link_status(oct, up); + if (err) { + netdev_err(oct->netdev, "Set link status to %d failed with err:%d\n", up, err); + return; + } + oct->link_info.oper_up = up; +} + +/** + * octep_vf_open() - start the octeon network device. + * + * @netdev: pointer to kernel network device. + * + * setup Tx/Rx queues, interrupts and enable hardware operation of Tx/Rx queues + * and interrupts.. + * + * Return: 0, on successfully setting up device and bring it up. + * -1, on any error. + */ +static int octep_vf_open(struct net_device *netdev) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + int err, ret; + + netdev_info(netdev, "Starting netdev ...\n"); + netif_carrier_off(netdev); + + oct->hw_ops.reset_io_queues(oct); + + if (octep_vf_setup_iqs(oct)) + goto setup_iq_err; + if (octep_vf_setup_oqs(oct)) + goto setup_oq_err; + if (octep_vf_setup_irqs(oct)) + goto setup_irq_err; + + err = netif_set_real_num_tx_queues(netdev, oct->num_oqs); + if (err) + goto set_queues_err; + err = netif_set_real_num_rx_queues(netdev, oct->num_iqs); + if (err) + goto set_queues_err; + + octep_vf_napi_add(oct); + octep_vf_napi_enable(oct); + + oct->link_info.admin_up = 1; + octep_vf_set_rx_state(oct, true); + + ret = octep_vf_get_link_status(oct); + if (!ret) + octep_vf_set_link_status(oct, true); + + /* Enable the input and output queues for this Octeon device */ + oct->hw_ops.enable_io_queues(oct); + + /* Enable Octeon device interrupts */ + oct->hw_ops.enable_interrupts(oct); + + octep_vf_oq_dbell_init(oct); + + ret = octep_vf_get_link_status(oct); + if (ret) + octep_vf_link_up(netdev); + + return 0; + +set_queues_err: + octep_vf_napi_disable(oct); + octep_vf_napi_delete(oct); + octep_vf_clean_irqs(oct); +setup_irq_err: + octep_vf_free_oqs(oct); +setup_oq_err: + octep_vf_free_iqs(oct); +setup_iq_err: + return -1; +} + +/** + * octep_vf_stop() - stop the octeon network device. + * + * @netdev: pointer to kernel network device. + * + * stop the device Tx/Rx operations, bring down the link and + * free up all resources allocated for Tx/Rx queues and interrupts. + */ +static int octep_vf_stop(struct net_device *netdev) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + + netdev_info(netdev, "Stopping the device ...\n"); + + /* Stop Tx from stack */ + netif_carrier_off(netdev); + netif_tx_disable(netdev); + + octep_vf_set_link_status(oct, false); + octep_vf_set_rx_state(oct, false); + + oct->link_info.admin_up = 0; + oct->link_info.oper_up = 0; + + oct->hw_ops.disable_interrupts(oct); + octep_vf_napi_disable(oct); + octep_vf_napi_delete(oct); + + octep_vf_clean_irqs(oct); + octep_vf_clean_iqs(oct); + + oct->hw_ops.disable_io_queues(oct); + oct->hw_ops.reset_io_queues(oct); + octep_vf_free_oqs(oct); + octep_vf_free_iqs(oct); + netdev_info(netdev, "Device stopped !!\n"); + return 0; +} + +/** + * octep_vf_iq_full_check() - check if a Tx queue is full. + * + * @iq: Octeon Tx queue data structure. + * + * Return: 0, if the Tx queue is not full. + * 1, if the Tx queue is full. + */ +static int octep_vf_iq_full_check(struct octep_vf_iq *iq) +{ + int ret; + + ret = netif_subqueue_maybe_stop(iq->netdev, iq->q_no, IQ_INSTR_SPACE(iq), + OCTEP_VF_WAKE_QUEUE_THRESHOLD, + OCTEP_VF_WAKE_QUEUE_THRESHOLD); + switch (ret) { + case 0: /* Stopped the queue, since IQ is full */ + return 1; + case -1: /* + * Pending updates in write index from + * iq_process_completion in other cpus + * caused queues to get re-enabled after + * being stopped + */ + iq->stats.restart_cnt++; + fallthrough; + case 1: /* Queue left enabled, since IQ is not yet full*/ + return 0; + } + + return 1; +} + +/** + * octep_vf_start_xmit() - Enqueue packet to Octoen hardware Tx Queue. + * + * @skb: packet skbuff pointer. + * @netdev: kernel network device. + * + * Return: NETDEV_TX_BUSY, if Tx Queue is full. + * NETDEV_TX_OK, if successfully enqueued to hardware Tx queue. + */ +static netdev_tx_t octep_vf_start_xmit(struct sk_buff *skb, + struct net_device *netdev) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + netdev_features_t feat = netdev->features; + struct octep_vf_tx_sglist_desc *sglist; + struct octep_vf_tx_buffer *tx_buffer; + struct octep_vf_tx_desc_hw *hw_desc; + struct skb_shared_info *shinfo; + struct octep_vf_instr_hdr *ih; + struct octep_vf_iq *iq; + skb_frag_t *frag; + u16 nr_frags, si; + int xmit_more; + u16 q_no, wi; + + if (skb_put_padto(skb, ETH_ZLEN)) + return NETDEV_TX_OK; + + q_no = skb_get_queue_mapping(skb); + if (q_no >= oct->num_iqs) { + netdev_err(netdev, "Invalid Tx skb->queue_mapping=%d\n", q_no); + q_no = q_no % oct->num_iqs; + } + + iq = oct->iq[q_no]; + + shinfo = skb_shinfo(skb); + nr_frags = shinfo->nr_frags; + + wi = iq->host_write_index; + hw_desc = &iq->desc_ring[wi]; + hw_desc->ih64 = 0; + + tx_buffer = iq->buff_info + wi; + tx_buffer->skb = skb; + + ih = &hw_desc->ih; + ih->tlen = skb->len; + ih->pkind = oct->fw_info.pkind; + ih->fsz = oct->fw_info.fsz; + ih->tlen = skb->len + ih->fsz; + + if (!nr_frags) { + tx_buffer->gather = 0; + tx_buffer->dma = dma_map_single(iq->dev, skb->data, + skb->len, DMA_TO_DEVICE); + if (dma_mapping_error(iq->dev, tx_buffer->dma)) + goto dma_map_err; + hw_desc->dptr = tx_buffer->dma; + } else { + /* Scatter/Gather */ + dma_addr_t dma; + u16 len; + + sglist = tx_buffer->sglist; + + ih->gsz = nr_frags + 1; + ih->gather = 1; + tx_buffer->gather = 1; + + len = skb_headlen(skb); + dma = dma_map_single(iq->dev, skb->data, len, DMA_TO_DEVICE); + if (dma_mapping_error(iq->dev, dma)) + goto dma_map_err; + + memset(sglist, 0, OCTEP_VF_SGLIST_SIZE_PER_PKT); + sglist[0].len[3] = len; + sglist[0].dma_ptr[0] = dma; + + si = 1; /* entry 0 is main skb, mapped above */ + frag = &shinfo->frags[0]; + while (nr_frags--) { + len = skb_frag_size(frag); + dma = skb_frag_dma_map(iq->dev, frag, 0, + len, DMA_TO_DEVICE); + if (dma_mapping_error(iq->dev, dma)) + goto dma_map_sg_err; + + sglist[si >> 2].len[3 - (si & 3)] = len; + sglist[si >> 2].dma_ptr[si & 3] = dma; + + frag++; + si++; + } + hw_desc->dptr = tx_buffer->sglist_dma; + } + if (oct->fw_info.tx_ol_flags) { + if ((feat & (NETIF_F_TSO)) && (skb_is_gso(skb))) { + hw_desc->txm.ol_flags = OCTEP_VF_TX_OFFLOAD_CKSUM; + hw_desc->txm.ol_flags |= OCTEP_VF_TX_OFFLOAD_TSO; + hw_desc->txm.gso_size = skb_shinfo(skb)->gso_size; + hw_desc->txm.gso_segs = skb_shinfo(skb)->gso_segs; + } else if (feat & (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM)) { + hw_desc->txm.ol_flags = OCTEP_VF_TX_OFFLOAD_CKSUM; + } + /* due to ESR txm will be swapped by hw */ + hw_desc->txm64[0] = (__force u64)cpu_to_be64(hw_desc->txm64[0]); + } + + xmit_more = netdev_xmit_more(); + + netdev_tx_sent_queue(iq->netdev_q, skb->len); + + skb_tx_timestamp(skb); + iq->fill_cnt++; + wi++; + iq->host_write_index = wi & iq->ring_size_mask; + + /* octep_iq_full_check stops the queue and returns + * true if so, in case the queue has become full + * by inserting current packet. If so, we can + * go ahead and ring doorbell. + */ + if (!octep_vf_iq_full_check(iq) && xmit_more && + iq->fill_cnt < iq->fill_threshold) + return NETDEV_TX_OK; + + goto ring_dbell; + +dma_map_sg_err: + if (si > 0) { + dma_unmap_single(iq->dev, sglist[0].dma_ptr[0], + sglist[0].len[0], DMA_TO_DEVICE); + sglist[0].len[0] = 0; + } + while (si > 1) { + dma_unmap_page(iq->dev, sglist[si >> 2].dma_ptr[si & 3], + sglist[si >> 2].len[si & 3], DMA_TO_DEVICE); + sglist[si >> 2].len[si & 3] = 0; + si--; + } + tx_buffer->gather = 0; +dma_map_err: + dev_kfree_skb_any(skb); +ring_dbell: + /* Flush the hw descriptors before writing to doorbell */ + smp_wmb(); + writel(iq->fill_cnt, iq->doorbell_reg); + iq->stats.instr_posted += iq->fill_cnt; + iq->fill_cnt = 0; + return NETDEV_TX_OK; +} + +int octep_vf_get_if_stats(struct octep_vf_device *oct) +{ + struct octep_vf_iface_rxtx_stats vf_stats; + int ret, size; + + memset(&vf_stats, 0, sizeof(struct octep_vf_iface_rxtx_stats)); + ret = octep_vf_mbox_bulk_read(oct, OCTEP_PFVF_MBOX_CMD_GET_STATS, + (u8 *)&vf_stats, &size); + + if (ret) + return ret; + + memcpy(&oct->iface_rx_stats, &vf_stats.iface_rx_stats, + sizeof(struct octep_vf_iface_rx_stats)); + memcpy(&oct->iface_tx_stats, &vf_stats.iface_tx_stats, + sizeof(struct octep_vf_iface_tx_stats)); + + return 0; +} + +int octep_vf_get_link_info(struct octep_vf_device *oct) +{ + int ret, size; + + ret = octep_vf_mbox_bulk_read(oct, OCTEP_PFVF_MBOX_CMD_GET_LINK_INFO, + (u8 *)&oct->link_info, &size); + if (ret) { + dev_err(&oct->pdev->dev, "Get VF link info failed via VF Mbox\n"); + return ret; + } + return 0; +} + +/** + * octep_vf_get_stats64() - Get Octeon network device statistics. + * + * @netdev: kernel network device. + * @stats: pointer to stats structure to be filled in. + */ +static void octep_vf_get_stats64(struct net_device *netdev, + struct rtnl_link_stats64 *stats) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + u64 tx_packets, tx_bytes, rx_packets, rx_bytes; + int q; + + tx_packets = 0; + tx_bytes = 0; + rx_packets = 0; + rx_bytes = 0; + for (q = 0; q < oct->num_oqs; q++) { + struct octep_vf_iq *iq = oct->iq[q]; + struct octep_vf_oq *oq = oct->oq[q]; + + tx_packets += iq->stats.instr_completed; + tx_bytes += iq->stats.bytes_sent; + rx_packets += oq->stats.packets; + rx_bytes += oq->stats.bytes; + } + stats->tx_packets = tx_packets; + stats->tx_bytes = tx_bytes; + stats->rx_packets = rx_packets; + stats->rx_bytes = rx_bytes; + if (!octep_vf_get_if_stats(oct)) { + stats->multicast = oct->iface_rx_stats.mcast_pkts; + stats->rx_errors = oct->iface_rx_stats.err_pkts; + stats->rx_dropped = oct->iface_rx_stats.dropped_pkts_fifo_full + + oct->iface_rx_stats.err_pkts; + stats->rx_missed_errors = oct->iface_rx_stats.dropped_pkts_fifo_full; + stats->tx_dropped = oct->iface_tx_stats.dropped; + } +} + +/** + * octep_vf_tx_timeout_task - work queue task to Handle Tx queue timeout. + * + * @work: pointer to Tx queue timeout work_struct + * + * Stop and start the device so that it frees up all queue resources + * and restarts the queues, that potentially clears a Tx queue timeout + * condition. + **/ +static void octep_vf_tx_timeout_task(struct work_struct *work) +{ + struct octep_vf_device *oct = container_of(work, struct octep_vf_device, + tx_timeout_task); + struct net_device *netdev = oct->netdev; + + rtnl_lock(); + if (netif_running(netdev)) { + octep_vf_stop(netdev); + octep_vf_open(netdev); + } + rtnl_unlock(); + netdev_put(netdev, NULL); +} + +/** + * octep_vf_tx_timeout() - Handle Tx Queue timeout. + * + * @netdev: pointer to kernel network device. + * @txqueue: Timed out Tx queue number. + * + * Schedule a work to handle Tx queue timeout. + */ +static void octep_vf_tx_timeout(struct net_device *netdev, unsigned int txqueue) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + + netdev_hold(netdev, NULL, GFP_ATOMIC); + schedule_work(&oct->tx_timeout_task); +} + +static int octep_vf_set_mac(struct net_device *netdev, void *p) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + struct sockaddr *addr = (struct sockaddr *)p; + int err; + + if (!is_valid_ether_addr(addr->sa_data)) + return -EADDRNOTAVAIL; + + err = octep_vf_mbox_set_mac_addr(oct, addr->sa_data); + if (err) + return err; + + memcpy(oct->mac_addr, addr->sa_data, ETH_ALEN); + eth_hw_addr_set(netdev, addr->sa_data); + + return 0; +} + +static int octep_vf_change_mtu(struct net_device *netdev, int new_mtu) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + struct octep_vf_iface_link_info *link_info; + int err; + + link_info = &oct->link_info; + if (link_info->mtu == new_mtu) + return 0; + + err = octep_vf_mbox_set_mtu(oct, new_mtu); + if (!err) { + oct->link_info.mtu = new_mtu; + netdev->mtu = new_mtu; + } + return err; +} + +static int octep_vf_set_features(struct net_device *netdev, + netdev_features_t features) +{ + struct octep_vf_device *oct = netdev_priv(netdev); + u16 rx_offloads = 0, tx_offloads = 0; + int err; + + /* We only support features received from firmware */ + if ((features & netdev->hw_features) != features) + return -EINVAL; + + if (features & NETIF_F_TSO) + tx_offloads |= OCTEP_VF_TX_OFFLOAD_TSO; + + if (features & NETIF_F_TSO6) + tx_offloads |= OCTEP_VF_TX_OFFLOAD_TSO; + + if (features & NETIF_F_IP_CSUM) + tx_offloads |= OCTEP_VF_TX_OFFLOAD_CKSUM; + + if (features & NETIF_F_IPV6_CSUM) + tx_offloads |= OCTEP_VF_TX_OFFLOAD_CKSUM; + + if (features & NETIF_F_RXCSUM) + rx_offloads |= OCTEP_VF_RX_OFFLOAD_CKSUM; + + err = octep_vf_mbox_set_offloads(oct, tx_offloads, rx_offloads); + if (!err) + netdev->features = features; + + return err; +} + +static const struct net_device_ops octep_vf_netdev_ops = { + .ndo_open = octep_vf_open, + .ndo_stop = octep_vf_stop, + .ndo_start_xmit = octep_vf_start_xmit, + .ndo_get_stats64 = octep_vf_get_stats64, + .ndo_tx_timeout = octep_vf_tx_timeout, + .ndo_set_mac_address = octep_vf_set_mac, + .ndo_change_mtu = octep_vf_change_mtu, + .ndo_set_features = octep_vf_set_features, +}; + +static const char *octep_vf_devid_to_str(struct octep_vf_device *oct) +{ + switch (oct->chip_id) { + case OCTEP_PCI_DEVICE_ID_CN93_VF: + return "CN93XX"; + case OCTEP_PCI_DEVICE_ID_CNF95N_VF: + return "CNF95N"; + case OCTEP_PCI_DEVICE_ID_CN10KA_VF: + return "CN10KA"; + case OCTEP_PCI_DEVICE_ID_CNF10KA_VF: + return "CNF10KA"; + case OCTEP_PCI_DEVICE_ID_CNF10KB_VF: + return "CNF10KB"; + case OCTEP_PCI_DEVICE_ID_CN10KB_VF: + return "CN10KB"; + default: + return "Unsupported"; + } +} + +/** + * octep_vf_device_setup() - Setup Octeon Device. + * + * @oct: Octeon device private data structure. + * + * Setup Octeon device hardware operations, configuration, etc ... + */ +int octep_vf_device_setup(struct octep_vf_device *oct) +{ + struct pci_dev *pdev = oct->pdev; + + /* allocate memory for oct->conf */ + oct->conf = kzalloc(sizeof(*oct->conf), GFP_KERNEL); + if (!oct->conf) + return -ENOMEM; + + /* Map BAR region 0 */ + oct->mmio.hw_addr = ioremap(pci_resource_start(oct->pdev, 0), + pci_resource_len(oct->pdev, 0)); + if (!oct->mmio.hw_addr) { + dev_err(&pdev->dev, + "Failed to remap BAR0; start=0x%llx len=0x%llx\n", + pci_resource_start(oct->pdev, 0), + pci_resource_len(oct->pdev, 0)); + goto ioremap_err; + } + oct->mmio.mapped = 1; + + oct->chip_id = pdev->device; + oct->rev_id = pdev->revision; + dev_info(&pdev->dev, "chip_id = 0x%x\n", pdev->device); + + switch (oct->chip_id) { + case OCTEP_PCI_DEVICE_ID_CN93_VF: + case OCTEP_PCI_DEVICE_ID_CNF95N_VF: + case OCTEP_PCI_DEVICE_ID_CN98_VF: + dev_info(&pdev->dev, "Setting up OCTEON %s VF PASS%d.%d\n", + octep_vf_devid_to_str(oct), OCTEP_VF_MAJOR_REV(oct), + OCTEP_VF_MINOR_REV(oct)); + octep_vf_device_setup_cn93(oct); + break; + case OCTEP_PCI_DEVICE_ID_CNF10KA_VF: + case OCTEP_PCI_DEVICE_ID_CN10KA_VF: + case OCTEP_PCI_DEVICE_ID_CNF10KB_VF: + case OCTEP_PCI_DEVICE_ID_CN10KB_VF: + dev_info(&pdev->dev, "Setting up OCTEON %s VF PASS%d.%d\n", + octep_vf_devid_to_str(oct), OCTEP_VF_MAJOR_REV(oct), + OCTEP_VF_MINOR_REV(oct)); + octep_vf_device_setup_cnxk(oct); + break; + default: + dev_err(&pdev->dev, "Unsupported device\n"); + goto unsupported_dev; + } + + return 0; + +unsupported_dev: + iounmap(oct->mmio.hw_addr); +ioremap_err: + kfree(oct->conf); + return -EOPNOTSUPP; +} + +/** + * octep_vf_device_cleanup() - Cleanup Octeon Device. + * + * @oct: Octeon device private data structure. + * + * Cleanup Octeon device allocated resources. + */ +static void octep_vf_device_cleanup(struct octep_vf_device *oct) +{ + dev_info(&oct->pdev->dev, "Cleaning up Octeon Device ...\n"); + + if (oct->mmio.mapped) + iounmap(oct->mmio.hw_addr); + + kfree(oct->conf); + oct->conf = NULL; +} + +static int octep_vf_get_mac_addr(struct octep_vf_device *oct, u8 *addr) +{ + return octep_vf_mbox_get_mac_addr(oct, addr); +} + +/** + * octep_vf_probe() - Octeon PCI device probe handler. + * + * @pdev: PCI device structure. + * @ent: entry in Octeon PCI device ID table. + * + * Initializes and enables the Octeon PCI device for network operations. + * Initializes Octeon private data structure and registers a network device. + */ +static int octep_vf_probe(struct pci_dev *pdev, const struct pci_device_id *ent) +{ + struct octep_vf_device *octep_vf_dev; + struct net_device *netdev; + int err; + + err = pci_enable_device(pdev); + if (err) { + dev_err(&pdev->dev, "Failed to enable PCI device\n"); + return err; + } + + err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64)); + if (err) { + dev_err(&pdev->dev, "Failed to set DMA mask !!\n"); + goto disable_pci_device; + } + + err = pci_request_mem_regions(pdev, OCTEP_VF_DRV_NAME); + if (err) { + dev_err(&pdev->dev, "Failed to map PCI memory regions\n"); + goto disable_pci_device; + } + + pci_set_master(pdev); + + netdev = alloc_etherdev_mq(sizeof(struct octep_vf_device), + OCTEP_VF_MAX_QUEUES); + if (!netdev) { + dev_err(&pdev->dev, "Failed to allocate netdev\n"); + err = -ENOMEM; + goto mem_regions_release; + } + SET_NETDEV_DEV(netdev, &pdev->dev); + + octep_vf_dev = netdev_priv(netdev); + octep_vf_dev->netdev = netdev; + octep_vf_dev->pdev = pdev; + octep_vf_dev->dev = &pdev->dev; + pci_set_drvdata(pdev, octep_vf_dev); + + err = octep_vf_device_setup(octep_vf_dev); + if (err) { + dev_err(&pdev->dev, "Device setup failed\n"); + goto netdevice_free; + } + INIT_WORK(&octep_vf_dev->tx_timeout_task, octep_vf_tx_timeout_task); + + netdev->netdev_ops = &octep_vf_netdev_ops; + octep_vf_set_ethtool_ops(netdev); + netif_carrier_off(netdev); + + if (octep_vf_setup_mbox(octep_vf_dev)) { + dev_err(&pdev->dev, "VF Mailbox setup failed\n"); + err = -ENOMEM; + goto device_cleanup; + } + + if (octep_vf_mbox_version_check(octep_vf_dev)) { + dev_err(&pdev->dev, "PF VF Mailbox version mismatch\n"); + err = -EINVAL; + goto delete_mbox; + } + + if (octep_vf_mbox_get_fw_info(octep_vf_dev)) { + dev_err(&pdev->dev, "unable to get fw info\n"); + err = -EINVAL; + goto delete_mbox; + } + + netdev->hw_features = NETIF_F_SG; + if (OCTEP_VF_TX_IP_CSUM(octep_vf_dev->fw_info.tx_ol_flags)) + netdev->hw_features |= (NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM); + + if (OCTEP_VF_RX_IP_CSUM(octep_vf_dev->fw_info.rx_ol_flags)) + netdev->hw_features |= NETIF_F_RXCSUM; + + netdev->min_mtu = OCTEP_VF_MIN_MTU; + netdev->max_mtu = OCTEP_VF_MAX_MTU; + netdev->mtu = OCTEP_VF_DEFAULT_MTU; + + if (OCTEP_VF_TX_TSO(octep_vf_dev->fw_info.tx_ol_flags)) { + netdev->hw_features |= NETIF_F_TSO; + netif_set_tso_max_size(netdev, netdev->max_mtu); + } + + netdev->features |= netdev->hw_features; + octep_vf_get_mac_addr(octep_vf_dev, octep_vf_dev->mac_addr); + eth_hw_addr_set(netdev, octep_vf_dev->mac_addr); + err = register_netdev(netdev); + if (err) { + dev_err(&pdev->dev, "Failed to register netdev\n"); + goto delete_mbox; + } + dev_info(&pdev->dev, "Device probe successful\n"); + return 0; + +delete_mbox: + octep_vf_delete_mbox(octep_vf_dev); +device_cleanup: + octep_vf_device_cleanup(octep_vf_dev); +netdevice_free: + free_netdev(netdev); +mem_regions_release: + pci_release_mem_regions(pdev); +disable_pci_device: + pci_disable_device(pdev); + dev_err(&pdev->dev, "Device probe failed\n"); + return err; +} + +/** + * octep_vf_remove() - Remove Octeon PCI device from driver control. + * + * @pdev: PCI device structure of the Octeon device. + * + * Cleanup all resources allocated for the Octeon device. + * Unregister from network device and disable the PCI device. + */ +static void octep_vf_remove(struct pci_dev *pdev) +{ + struct octep_vf_device *oct = pci_get_drvdata(pdev); + struct net_device *netdev; + + if (!oct) + return; + + octep_vf_mbox_dev_remove(oct); + cancel_work_sync(&oct->tx_timeout_task); + netdev = oct->netdev; + if (netdev->reg_state == NETREG_REGISTERED) + unregister_netdev(netdev); + octep_vf_delete_mbox(oct); + octep_vf_device_cleanup(oct); + pci_release_mem_regions(pdev); + free_netdev(netdev); + pci_disable_device(pdev); +} + +static struct pci_driver octep_vf_driver = { + .name = OCTEP_VF_DRV_NAME, + .id_table = octep_vf_pci_id_tbl, + .probe = octep_vf_probe, + .remove = octep_vf_remove, +}; + +/** + * octep_vf_init_module() - Module initialization. + * + * create common resource for the driver and register PCI driver. + */ +static int __init octep_vf_init_module(void) +{ + int ret; + + pr_info("%s: Loading %s ...\n", OCTEP_VF_DRV_NAME, OCTEP_VF_DRV_STRING); + + ret = pci_register_driver(&octep_vf_driver); + if (ret < 0) { + pr_err("%s: Failed to register PCI driver; err=%d\n", + OCTEP_VF_DRV_NAME, ret); + return ret; + } + + return ret; +} + +/** + * octep_vf_exit_module() - Module exit routine. + * + * unregister the driver with PCI subsystem and cleanup common resources. + */ +static void __exit octep_vf_exit_module(void) +{ + pr_info("%s: Unloading ...\n", OCTEP_VF_DRV_NAME); + + pci_unregister_driver(&octep_vf_driver); + + pr_info("%s: Unloading complete\n", OCTEP_VF_DRV_NAME); +} + +module_init(octep_vf_init_module); +module_exit(octep_vf_exit_module); diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.h b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.h new file mode 100644 index 000000000000..5769f62545cd --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_main.h @@ -0,0 +1,334 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ + +#ifndef _OCTEP_VF_MAIN_H_ +#define _OCTEP_VF_MAIN_H_ + +#include "octep_vf_tx.h" +#include "octep_vf_rx.h" +#include "octep_vf_mbox.h" + +#define OCTEP_VF_DRV_NAME "octeon_ep_vf" +#define OCTEP_VF_DRV_STRING "Marvell Octeon EndPoint NIC VF Driver" + +#define OCTEP_PCI_DEVICE_ID_CN93_VF 0xB203 //93xx VF +#define OCTEP_PCI_DEVICE_ID_CNF95N_VF 0xB403 //95N VF +#define OCTEP_PCI_DEVICE_ID_CN98_VF 0xB103 +#define OCTEP_PCI_DEVICE_ID_CN10KA_VF 0xB903 +#define OCTEP_PCI_DEVICE_ID_CNF10KA_VF 0xBA03 +#define OCTEP_PCI_DEVICE_ID_CNF10KB_VF 0xBC03 +#define OCTEP_PCI_DEVICE_ID_CN10KB_VF 0xBD03 + +#define OCTEP_VF_MAX_QUEUES 63 +#define OCTEP_VF_MAX_IQ OCTEP_VF_MAX_QUEUES +#define OCTEP_VF_MAX_OQ OCTEP_VF_MAX_QUEUES + +#define OCTEP_VF_MAX_MSIX_VECTORS OCTEP_VF_MAX_OQ + +#define OCTEP_VF_IQ_INTR_RESEND_BIT 59 +#define OCTEP_VF_OQ_INTR_RESEND_BIT 59 + +#define IQ_INSTR_PENDING(iq) ({ typeof(iq) iq__ = (iq); \ + ((iq__)->host_write_index - (iq__)->flush_index) & \ + (iq__)->ring_size_mask; \ + }) +#define IQ_INSTR_SPACE(iq) ({ typeof(iq) iq_ = (iq); \ + (iq_)->max_count - IQ_INSTR_PENDING(iq_); \ + }) + +/* PCI address space mapping information. + * Each of the 3 address spaces given by BAR0, BAR2 and BAR4 of + * Octeon gets mapped to different physical address spaces in + * the kernel. + */ +struct octep_vf_mmio { + /* The physical address to which the PCI address space is mapped. */ + u8 __iomem *hw_addr; + + /* Flag indicating the mapping was successful. */ + int mapped; +}; + +struct octep_vf_hw_ops { + void (*setup_iq_regs)(struct octep_vf_device *oct, int q); + void (*setup_oq_regs)(struct octep_vf_device *oct, int q); + void (*setup_mbox_regs)(struct octep_vf_device *oct, int mbox); + + irqreturn_t (*non_ioq_intr_handler)(void *ioq_vector); + irqreturn_t (*ioq_intr_handler)(void *ioq_vector); + void (*reinit_regs)(struct octep_vf_device *oct); + u32 (*update_iq_read_idx)(struct octep_vf_iq *iq); + + void (*enable_interrupts)(struct octep_vf_device *oct); + void (*disable_interrupts)(struct octep_vf_device *oct); + + void (*enable_io_queues)(struct octep_vf_device *oct); + void (*disable_io_queues)(struct octep_vf_device *oct); + void (*enable_iq)(struct octep_vf_device *oct, int q); + void (*disable_iq)(struct octep_vf_device *oct, int q); + void (*enable_oq)(struct octep_vf_device *oct, int q); + void (*disable_oq)(struct octep_vf_device *oct, int q); + void (*reset_io_queues)(struct octep_vf_device *oct); + void (*dump_registers)(struct octep_vf_device *oct); +}; + +/* Octeon mailbox data */ +struct octep_vf_mbox_data { + /* Holds the offset of received data via mailbox. */ + u32 data_index; + + /* Holds the received data via mailbox. */ + u8 recv_data[OCTEP_PFVF_MBOX_MAX_DATA_BUF_SIZE]; +}; + +/* wrappers around work structs */ +struct octep_vf_mbox_wk { + struct work_struct work; + void *ctxptr; +}; + +/* Octeon device mailbox */ +struct octep_vf_mbox { + /* A mutex to protect access to this q_mbox. */ + struct mutex lock; + + u32 state; + + /* SLI_MAC_PF_MBOX_INT for PF, SLI_PKT_MBOX_INT for VF. */ + u8 __iomem *mbox_int_reg; + + /* SLI_PKT_PF_VF_MBOX_SIG(0) for PF, + * SLI_PKT_PF_VF_MBOX_SIG(1) for VF. + */ + u8 __iomem *mbox_write_reg; + + /* SLI_PKT_PF_VF_MBOX_SIG(1) for PF, + * SLI_PKT_PF_VF_MBOX_SIG(0) for VF. + */ + u8 __iomem *mbox_read_reg; + + /* Octeon mailbox data */ + struct octep_vf_mbox_data mbox_data; + + /* Octeon mailbox work handler to process Mbox messages */ + struct octep_vf_mbox_wk wk; +}; + +/* Tx/Rx queue vector per interrupt. */ +struct octep_vf_ioq_vector { + char name[OCTEP_VF_MSIX_NAME_SIZE]; + struct napi_struct napi; + struct octep_vf_device *octep_vf_dev; + struct octep_vf_iq *iq; + struct octep_vf_oq *oq; + cpumask_t affinity_mask; +}; + +/* Octeon hardware/firmware offload capability flags. */ +#define OCTEP_VF_CAP_TX_CHECKSUM BIT(0) +#define OCTEP_VF_CAP_RX_CHECKSUM BIT(1) +#define OCTEP_VF_CAP_TSO BIT(2) + +/* Link modes */ +enum octep_vf_link_mode_bit_indices { + OCTEP_VF_LINK_MODE_10GBASE_T = 0, + OCTEP_VF_LINK_MODE_10GBASE_R, + OCTEP_VF_LINK_MODE_10GBASE_CR, + OCTEP_VF_LINK_MODE_10GBASE_KR, + OCTEP_VF_LINK_MODE_10GBASE_LR, + OCTEP_VF_LINK_MODE_10GBASE_SR, + OCTEP_VF_LINK_MODE_25GBASE_CR, + OCTEP_VF_LINK_MODE_25GBASE_KR, + OCTEP_VF_LINK_MODE_25GBASE_SR, + OCTEP_VF_LINK_MODE_40GBASE_CR4, + OCTEP_VF_LINK_MODE_40GBASE_KR4, + OCTEP_VF_LINK_MODE_40GBASE_LR4, + OCTEP_VF_LINK_MODE_40GBASE_SR4, + OCTEP_VF_LINK_MODE_50GBASE_CR2, + OCTEP_VF_LINK_MODE_50GBASE_KR2, + OCTEP_VF_LINK_MODE_50GBASE_SR2, + OCTEP_VF_LINK_MODE_50GBASE_CR, + OCTEP_VF_LINK_MODE_50GBASE_KR, + OCTEP_VF_LINK_MODE_50GBASE_LR, + OCTEP_VF_LINK_MODE_50GBASE_SR, + OCTEP_VF_LINK_MODE_100GBASE_CR4, + OCTEP_VF_LINK_MODE_100GBASE_KR4, + OCTEP_VF_LINK_MODE_100GBASE_LR4, + OCTEP_VF_LINK_MODE_100GBASE_SR4, + OCTEP_VF_LINK_MODE_NBITS +}; + +/* Hardware interface link state information. */ +struct octep_vf_iface_link_info { + /* Bitmap of Supported link speeds/modes. */ + u64 supported_modes; + + /* Bitmap of Advertised link speeds/modes. */ + u64 advertised_modes; + + /* Negotiated link speed in Mbps. */ + u32 speed; + + /* MTU */ + u16 mtu; + + /* Autonegotiation state. */ +#define OCTEP_VF_LINK_MODE_AUTONEG_SUPPORTED BIT(0) +#define OCTEP_VF_LINK_MODE_AUTONEG_ADVERTISED BIT(1) + u8 autoneg; + + /* Pause frames setting. */ +#define OCTEP_VF_LINK_MODE_PAUSE_SUPPORTED BIT(0) +#define OCTEP_VF_LINK_MODE_PAUSE_ADVERTISED BIT(1) + u8 pause; + + /* Admin state of the link (ifconfig <iface> up/down */ + u8 admin_up; + + /* Operational state of the link: physical link is up down */ + u8 oper_up; +}; + +/* Hardware interface stats information. */ +struct octep_vf_iface_rxtx_stats { + /* Hardware Interface Rx statistics */ + struct octep_vf_iface_rx_stats iface_rx_stats; + + /* Hardware Interface Tx statistics */ + struct octep_vf_iface_tx_stats iface_tx_stats; +}; + +struct octep_vf_fw_info { + /* pkind value to be used in every Tx hardware descriptor */ + u8 pkind; + /* front size data */ + u8 fsz; + /* supported rx offloads OCTEP_VF_RX_OFFLOAD_* */ + u16 rx_ol_flags; + /* supported tx offloads OCTEP_VF_TX_OFFLOAD_* */ + u16 tx_ol_flags; +}; + +/* The Octeon device specific private data structure. + * Each Octeon device has this structure to represent all its components. + */ +struct octep_vf_device { + struct octep_vf_config *conf; + + /* Octeon Chip type. */ + u16 chip_id; + u16 rev_id; + + /* Device capabilities enabled */ + u64 caps_enabled; + /* Device capabilities supported */ + u64 caps_supported; + + /* Pointer to basic Linux device */ + struct device *dev; + /* Linux PCI device pointer */ + struct pci_dev *pdev; + /* Netdev corresponding to the Octeon device */ + struct net_device *netdev; + + /* memory mapped io range */ + struct octep_vf_mmio mmio; + + /* MAC address */ + u8 mac_addr[ETH_ALEN]; + + /* Tx queues (IQ: Instruction Queue) */ + u16 num_iqs; + /* Pointers to Octeon Tx queues */ + struct octep_vf_iq *iq[OCTEP_VF_MAX_IQ]; + + /* Rx queues (OQ: Output Queue) */ + u16 num_oqs; + /* Pointers to Octeon Rx queues */ + struct octep_vf_oq *oq[OCTEP_VF_MAX_OQ]; + + /* Hardware port number of the PCIe interface */ + u16 pcie_port; + + /* Hardware operations */ + struct octep_vf_hw_ops hw_ops; + + /* IRQ info */ + u16 num_irqs; + u16 num_non_ioq_irqs; + char *non_ioq_irq_names; + struct msix_entry *msix_entries; + /* IOq information of it's corresponding MSI-X interrupt. */ + struct octep_vf_ioq_vector *ioq_vector[OCTEP_VF_MAX_QUEUES]; + + /* Hardware Interface Tx statistics */ + struct octep_vf_iface_tx_stats iface_tx_stats; + /* Hardware Interface Rx statistics */ + struct octep_vf_iface_rx_stats iface_rx_stats; + + /* Hardware Interface Link info like supported modes, aneg support */ + struct octep_vf_iface_link_info link_info; + + /* Mailbox to talk to VFs */ + struct octep_vf_mbox *mbox; + + /* Work entry to handle Tx timeout */ + struct work_struct tx_timeout_task; + + /* offset for iface stats */ + u32 ctrl_mbox_ifstats_offset; + + /* Negotiated Mbox version */ + u32 mbox_neg_ver; + + /* firmware info */ + struct octep_vf_fw_info fw_info; +}; + +static inline u16 OCTEP_VF_MAJOR_REV(struct octep_vf_device *oct) +{ + u16 rev = (oct->rev_id & 0xC) >> 2; + + return (rev == 0) ? 1 : rev; +} + +static inline u16 OCTEP_VF_MINOR_REV(struct octep_vf_device *oct) +{ + return (oct->rev_id & 0x3); +} + +/* Octeon CSR read/write access APIs */ +#define octep_vf_write_csr(octep_vf_dev, reg_off, value) \ + writel(value, (octep_vf_dev)->mmio.hw_addr + (reg_off)) + +#define octep_vf_write_csr64(octep_vf_dev, reg_off, val64) \ + writeq(val64, (octep_vf_dev)->mmio.hw_addr + (reg_off)) + +#define octep_vf_read_csr(octep_vf_dev, reg_off) \ + readl((octep_vf_dev)->mmio.hw_addr + (reg_off)) + +#define octep_vf_read_csr64(octep_vf_dev, reg_off) \ + readq((octep_vf_dev)->mmio.hw_addr + (reg_off)) + +extern struct workqueue_struct *octep_vf_wq; + +int octep_vf_device_setup(struct octep_vf_device *oct); +int octep_vf_setup_iqs(struct octep_vf_device *oct); +void octep_vf_free_iqs(struct octep_vf_device *oct); +void octep_vf_clean_iqs(struct octep_vf_device *oct); +int octep_vf_setup_oqs(struct octep_vf_device *oct); +void octep_vf_free_oqs(struct octep_vf_device *oct); +void octep_vf_oq_dbell_init(struct octep_vf_device *oct); +void octep_vf_device_setup_cn93(struct octep_vf_device *oct); +void octep_vf_device_setup_cnxk(struct octep_vf_device *oct); +int octep_vf_iq_process_completions(struct octep_vf_iq *iq, u16 budget); +int octep_vf_oq_process_rx(struct octep_vf_oq *oq, int budget); +void octep_vf_set_ethtool_ops(struct net_device *netdev); +int octep_vf_get_link_info(struct octep_vf_device *oct); +int octep_vf_get_if_stats(struct octep_vf_device *oct); +void octep_vf_mbox_work(struct work_struct *work); +#endif /* _OCTEP_VF_MAIN_H_ */ diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_mbox.c b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_mbox.c new file mode 100644 index 000000000000..2eab21e43048 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_mbox.c @@ -0,0 +1,430 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ +#include <linux/types.h> +#include <linux/pci.h> +#include <linux/netdevice.h> +#include "octep_vf_config.h" +#include "octep_vf_main.h" + +/* When a new command is implemented, the below table should be updated + * with new command and it's version info. + */ +static u32 pfvf_cmd_versions[OCTEP_PFVF_MBOX_CMD_MAX] = { + [0 ... OCTEP_PFVF_MBOX_CMD_DEV_REMOVE] = OCTEP_PFVF_MBOX_VERSION_V1, + [OCTEP_PFVF_MBOX_CMD_GET_FW_INFO ... OCTEP_PFVF_MBOX_NOTIF_LINK_STATUS] = + OCTEP_PFVF_MBOX_VERSION_V2 +}; + +int octep_vf_setup_mbox(struct octep_vf_device *oct) +{ + int ring = 0; + + oct->mbox = vzalloc(sizeof(*oct->mbox)); + if (!oct->mbox) + return -1; + + mutex_init(&oct->mbox->lock); + + oct->hw_ops.setup_mbox_regs(oct, ring); + INIT_WORK(&oct->mbox->wk.work, octep_vf_mbox_work); + oct->mbox->wk.ctxptr = oct; + oct->mbox_neg_ver = OCTEP_PFVF_MBOX_VERSION_CURRENT; + dev_info(&oct->pdev->dev, "setup vf mbox successfully\n"); + return 0; +} + +void octep_vf_delete_mbox(struct octep_vf_device *oct) +{ + if (oct->mbox) { + if (work_pending(&oct->mbox->wk.work)) + cancel_work_sync(&oct->mbox->wk.work); + + mutex_destroy(&oct->mbox->lock); + vfree(oct->mbox); + oct->mbox = NULL; + dev_info(&oct->pdev->dev, "Deleted vf mbox successfully\n"); + } +} + +int octep_vf_mbox_version_check(struct octep_vf_device *oct) +{ + union octep_pfvf_mbox_word cmd; + union octep_pfvf_mbox_word rsp; + int ret; + + cmd.u64 = 0; + cmd.s_version.opcode = OCTEP_PFVF_MBOX_CMD_VERSION; + cmd.s_version.version = OCTEP_PFVF_MBOX_VERSION_CURRENT; + ret = octep_vf_mbox_send_cmd(oct, cmd, &rsp); + if (ret == OCTEP_PFVF_MBOX_CMD_STATUS_NACK) { + dev_err(&oct->pdev->dev, + "VF Mbox version is incompatible with PF\n"); + return -EINVAL; + } + oct->mbox_neg_ver = (u32)rsp.s_version.version; + dev_dbg(&oct->pdev->dev, + "VF Mbox version:%u Negotiated VF version with PF:%u\n", + (u32)cmd.s_version.version, + (u32)rsp.s_version.version); + return 0; +} + +void octep_vf_mbox_work(struct work_struct *work) +{ + struct octep_vf_mbox_wk *wk = container_of(work, struct octep_vf_mbox_wk, work); + struct octep_vf_iface_link_info *link_info; + struct octep_vf_device *oct = NULL; + struct octep_vf_mbox *mbox = NULL; + union octep_pfvf_mbox_word *notif; + u64 pf_vf_data; + + oct = (struct octep_vf_device *)wk->ctxptr; + link_info = &oct->link_info; + mbox = oct->mbox; + pf_vf_data = readq(mbox->mbox_read_reg); + + notif = (union octep_pfvf_mbox_word *)&pf_vf_data; + + switch (notif->s.opcode) { + case OCTEP_PFVF_MBOX_NOTIF_LINK_STATUS: + if (notif->s_link_status.status) { + link_info->oper_up = OCTEP_PFVF_LINK_STATUS_UP; + netif_carrier_on(oct->netdev); + dev_info(&oct->pdev->dev, "netif_carrier_on\n"); + } else { + link_info->oper_up = OCTEP_PFVF_LINK_STATUS_DOWN; + netif_carrier_off(oct->netdev); + dev_info(&oct->pdev->dev, "netif_carrier_off\n"); + } + break; + default: + dev_err(&oct->pdev->dev, + "Received unsupported notif %d\n", notif->s.opcode); + break; + } +} + +static int __octep_vf_mbox_send_cmd(struct octep_vf_device *oct, + union octep_pfvf_mbox_word cmd, + union octep_pfvf_mbox_word *rsp) +{ + struct octep_vf_mbox *mbox = oct->mbox; + u64 reg_val = 0ull; + int count; + + if (!mbox) + return OCTEP_PFVF_MBOX_CMD_STATUS_NOT_SETUP; + + cmd.s.type = OCTEP_PFVF_MBOX_TYPE_CMD; + writeq(cmd.u64, mbox->mbox_write_reg); + + /* No response for notification messages */ + if (!rsp) + return 0; + + for (count = 0; count < OCTEP_PFVF_MBOX_TIMEOUT_WAIT_COUNT; count++) { + usleep_range(1000, 1500); + reg_val = readq(mbox->mbox_write_reg); + if (reg_val != cmd.u64) { + rsp->u64 = reg_val; + break; + } + } + if (count == OCTEP_PFVF_MBOX_TIMEOUT_WAIT_COUNT) { + dev_err(&oct->pdev->dev, "mbox send command timed out\n"); + return OCTEP_PFVF_MBOX_CMD_STATUS_TIMEDOUT; + } + if (rsp->s.type != OCTEP_PFVF_MBOX_TYPE_RSP_ACK) { + dev_err(&oct->pdev->dev, "mbox_send: Received NACK\n"); + return OCTEP_PFVF_MBOX_CMD_STATUS_NACK; + } + rsp->u64 = reg_val; + return 0; +} + +int octep_vf_mbox_send_cmd(struct octep_vf_device *oct, union octep_pfvf_mbox_word cmd, + union octep_pfvf_mbox_word *rsp) +{ + struct octep_vf_mbox *mbox = oct->mbox; + int ret; + + if (!mbox) + return OCTEP_PFVF_MBOX_CMD_STATUS_NOT_SETUP; + mutex_lock(&mbox->lock); + if (pfvf_cmd_versions[cmd.s.opcode] > oct->mbox_neg_ver) { + dev_dbg(&oct->pdev->dev, "CMD:%d not supported in Version:%d\n", + cmd.s.opcode, oct->mbox_neg_ver); + mutex_unlock(&mbox->lock); + return -EOPNOTSUPP; + } + ret = __octep_vf_mbox_send_cmd(oct, cmd, rsp); + mutex_unlock(&mbox->lock); + return ret; +} + +int octep_vf_mbox_bulk_read(struct octep_vf_device *oct, enum octep_pfvf_mbox_opcode opcode, + u8 *data, int *size) +{ + struct octep_vf_mbox *mbox = oct->mbox; + union octep_pfvf_mbox_word cmd; + union octep_pfvf_mbox_word rsp; + int data_len = 0, tmp_len = 0; + int read_cnt, i = 0, ret; + + if (!mbox) + return OCTEP_PFVF_MBOX_CMD_STATUS_NOT_SETUP; + + mutex_lock(&mbox->lock); + cmd.u64 = 0; + cmd.s_data.opcode = opcode; + cmd.s_data.frag = 0; + /* Send cmd to read data from PF */ + ret = __octep_vf_mbox_send_cmd(oct, cmd, &rsp); + if (ret) { + dev_err(&oct->pdev->dev, "send mbox cmd fail for data request\n"); + mutex_unlock(&mbox->lock); + return ret; + } + /* PF sends the data length of requested CMD + * in ACK + */ + data_len = *((int32_t *)rsp.s_data.data); + tmp_len = data_len; + cmd.u64 = 0; + rsp.u64 = 0; + cmd.s_data.opcode = opcode; + cmd.s_data.frag = 1; + while (data_len) { + ret = __octep_vf_mbox_send_cmd(oct, cmd, &rsp); + if (ret) { + dev_err(&oct->pdev->dev, "send mbox cmd fail for data request\n"); + mutex_unlock(&mbox->lock); + mbox->mbox_data.data_index = 0; + memset(mbox->mbox_data.recv_data, 0, OCTEP_PFVF_MBOX_MAX_DATA_BUF_SIZE); + return ret; + } + if (data_len > OCTEP_PFVF_MBOX_MAX_DATA_SIZE) { + data_len -= OCTEP_PFVF_MBOX_MAX_DATA_SIZE; + read_cnt = OCTEP_PFVF_MBOX_MAX_DATA_SIZE; + } else { + read_cnt = data_len; + data_len = 0; + } + for (i = 0; i < read_cnt; i++) { + mbox->mbox_data.recv_data[mbox->mbox_data.data_index] = + rsp.s_data.data[i]; + mbox->mbox_data.data_index++; + } + cmd.u64 = 0; + rsp.u64 = 0; + cmd.s_data.opcode = opcode; + cmd.s_data.frag = 1; + } + memcpy(data, mbox->mbox_data.recv_data, tmp_len); + *size = tmp_len; + mbox->mbox_data.data_index = 0; + memset(mbox->mbox_data.recv_data, 0, OCTEP_PFVF_MBOX_MAX_DATA_BUF_SIZE); + mutex_unlock(&mbox->lock); + return 0; +} + +int octep_vf_mbox_set_mtu(struct octep_vf_device *oct, int mtu) +{ + int frame_size = mtu + ETH_HLEN + ETH_FCS_LEN; + union octep_pfvf_mbox_word cmd; + union octep_pfvf_mbox_word rsp; + int ret = 0; + + if (mtu < ETH_MIN_MTU || frame_size > ETH_MAX_MTU) { + dev_err(&oct->pdev->dev, + "Failed to set MTU to %d MIN MTU:%d MAX MTU:%d\n", + mtu, ETH_MIN_MTU, ETH_MAX_MTU); + return -EINVAL; + } + + cmd.u64 = 0; + cmd.s_set_mtu.opcode = OCTEP_PFVF_MBOX_CMD_SET_MTU; + cmd.s_set_mtu.mtu = mtu; + + ret = octep_vf_mbox_send_cmd(oct, cmd, &rsp); + if (ret) { + dev_err(&oct->pdev->dev, "Mbox send failed; err=%d\n", ret); + return ret; + } + if (rsp.s_set_mtu.type != OCTEP_PFVF_MBOX_TYPE_RSP_ACK) { + dev_err(&oct->pdev->dev, "Received Mbox NACK from PF for MTU:%d\n", mtu); + return -EINVAL; + } + + return 0; +} + +int octep_vf_mbox_set_mac_addr(struct octep_vf_device *oct, char *mac_addr) +{ + union octep_pfvf_mbox_word cmd; + union octep_pfvf_mbox_word rsp; + int i, ret; + + cmd.u64 = 0; + cmd.s_set_mac.opcode = OCTEP_PFVF_MBOX_CMD_SET_MAC_ADDR; + for (i = 0; i < ETH_ALEN; i++) + cmd.s_set_mac.mac_addr[i] = mac_addr[i]; + ret = octep_vf_mbox_send_cmd(oct, cmd, &rsp); + if (ret) { + dev_err(&oct->pdev->dev, "Mbox send failed; err = %d\n", ret); + return ret; + } + if (rsp.s_set_mac.type != OCTEP_PFVF_MBOX_TYPE_RSP_ACK) { + dev_err(&oct->pdev->dev, "received NACK\n"); + return -EINVAL; + } + return 0; +} + +int octep_vf_mbox_get_mac_addr(struct octep_vf_device *oct, char *mac_addr) +{ + union octep_pfvf_mbox_word cmd; + union octep_pfvf_mbox_word rsp; + int i, ret; + + cmd.u64 = 0; + cmd.s_set_mac.opcode = OCTEP_PFVF_MBOX_CMD_GET_MAC_ADDR; + ret = octep_vf_mbox_send_cmd(oct, cmd, &rsp); + if (ret) { + dev_err(&oct->pdev->dev, "get_mac: mbox send failed; err = %d\n", ret); + return ret; + } + if (rsp.s_set_mac.type != OCTEP_PFVF_MBOX_TYPE_RSP_ACK) { + dev_err(&oct->pdev->dev, "get_mac: received NACK\n"); + return -EINVAL; + } + for (i = 0; i < ETH_ALEN; i++) + mac_addr[i] = rsp.s_set_mac.mac_addr[i]; + return 0; +} + +int octep_vf_mbox_set_rx_state(struct octep_vf_device *oct, bool state) +{ + union octep_pfvf_mbox_word cmd; + union octep_pfvf_mbox_word rsp; + int ret; + + cmd.u64 = 0; + cmd.s_link_state.opcode = OCTEP_PFVF_MBOX_CMD_SET_RX_STATE; + cmd.s_link_state.state = state; + ret = octep_vf_mbox_send_cmd(oct, cmd, &rsp); + if (ret) { + dev_err(&oct->pdev->dev, "Set Rx state via VF Mbox send failed\n"); + return ret; + } + if (rsp.s_link_state.type != OCTEP_PFVF_MBOX_TYPE_RSP_ACK) { + dev_err(&oct->pdev->dev, "Set Rx state received NACK\n"); + return -EINVAL; + } + return 0; +} + +int octep_vf_mbox_set_link_status(struct octep_vf_device *oct, bool status) +{ + union octep_pfvf_mbox_word cmd; + union octep_pfvf_mbox_word rsp; + int ret; + + cmd.u64 = 0; + cmd.s_link_status.opcode = OCTEP_PFVF_MBOX_CMD_SET_LINK_STATUS; + cmd.s_link_status.status = status; + ret = octep_vf_mbox_send_cmd(oct, cmd, &rsp); + if (ret) { + dev_err(&oct->pdev->dev, "Set link status via VF Mbox send failed\n"); + return ret; + } + if (rsp.s_link_status.type != OCTEP_PFVF_MBOX_TYPE_RSP_ACK) { + dev_err(&oct->pdev->dev, "Set link status received NACK\n"); + return -EINVAL; + } + return 0; +} + +int octep_vf_mbox_get_link_status(struct octep_vf_device *oct, u8 *oper_up) +{ + union octep_pfvf_mbox_word cmd; + union octep_pfvf_mbox_word rsp; + int ret; + + cmd.u64 = 0; + cmd.s_link_status.opcode = OCTEP_PFVF_MBOX_CMD_GET_LINK_STATUS; + ret = octep_vf_mbox_send_cmd(oct, cmd, &rsp); + if (ret) { + dev_err(&oct->pdev->dev, "Get link status via VF Mbox send failed\n"); + return ret; + } + if (rsp.s_link_status.type != OCTEP_PFVF_MBOX_TYPE_RSP_ACK) { + dev_err(&oct->pdev->dev, "Get link status received NACK\n"); + return -EINVAL; + } + *oper_up = rsp.s_link_status.status; + return 0; +} + +int octep_vf_mbox_dev_remove(struct octep_vf_device *oct) +{ + union octep_pfvf_mbox_word cmd; + int ret; + + cmd.u64 = 0; + cmd.s.opcode = OCTEP_PFVF_MBOX_CMD_DEV_REMOVE; + ret = octep_vf_mbox_send_cmd(oct, cmd, NULL); + return ret; +} + +int octep_vf_mbox_get_fw_info(struct octep_vf_device *oct) +{ + union octep_pfvf_mbox_word cmd; + union octep_pfvf_mbox_word rsp; + int ret; + + cmd.u64 = 0; + cmd.s_fw_info.opcode = OCTEP_PFVF_MBOX_CMD_GET_FW_INFO; + ret = octep_vf_mbox_send_cmd(oct, cmd, &rsp); + if (ret) { + dev_err(&oct->pdev->dev, "Get link status via VF Mbox send failed\n"); + return ret; + } + if (rsp.s_fw_info.type != OCTEP_PFVF_MBOX_TYPE_RSP_ACK) { + dev_err(&oct->pdev->dev, "Get link status received NACK\n"); + return -EINVAL; + } + oct->fw_info.pkind = rsp.s_fw_info.pkind; + oct->fw_info.fsz = rsp.s_fw_info.fsz; + oct->fw_info.rx_ol_flags = rsp.s_fw_info.rx_ol_flags; + oct->fw_info.tx_ol_flags = rsp.s_fw_info.tx_ol_flags; + + return 0; +} + +int octep_vf_mbox_set_offloads(struct octep_vf_device *oct, u16 tx_offloads, + u16 rx_offloads) +{ + union octep_pfvf_mbox_word cmd; + union octep_pfvf_mbox_word rsp; + int ret; + + cmd.u64 = 0; + cmd.s_offloads.opcode = OCTEP_PFVF_MBOX_CMD_SET_OFFLOADS; + cmd.s_offloads.rx_ol_flags = rx_offloads; + cmd.s_offloads.tx_ol_flags = tx_offloads; + ret = octep_vf_mbox_send_cmd(oct, cmd, &rsp); + if (ret) { + dev_err(&oct->pdev->dev, "Set offloads via VF Mbox send failed\n"); + return ret; + } + if (rsp.s_link_state.type != OCTEP_PFVF_MBOX_TYPE_RSP_ACK) { + dev_err(&oct->pdev->dev, "Set offloads received NACK\n"); + return -EINVAL; + } + return 0; +} diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_mbox.h b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_mbox.h new file mode 100644 index 000000000000..9b5efad37eab --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_mbox.h @@ -0,0 +1,166 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Marvell Octeon EP (EndPoint) Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ +#ifndef _OCTEP_VF_MBOX_H_ +#define _OCTEP_VF_MBOX_H_ + +/* When a new command is implemented, VF Mbox version should be bumped. + */ +enum octep_pfvf_mbox_version { + OCTEP_PFVF_MBOX_VERSION_V0, + OCTEP_PFVF_MBOX_VERSION_V1, + OCTEP_PFVF_MBOX_VERSION_V2 +}; + +#define OCTEP_PFVF_MBOX_VERSION_CURRENT OCTEP_PFVF_MBOX_VERSION_V2 + +enum octep_pfvf_mbox_opcode { + OCTEP_PFVF_MBOX_CMD_VERSION, + OCTEP_PFVF_MBOX_CMD_SET_MTU, + OCTEP_PFVF_MBOX_CMD_SET_MAC_ADDR, + OCTEP_PFVF_MBOX_CMD_GET_MAC_ADDR, + OCTEP_PFVF_MBOX_CMD_GET_LINK_INFO, + OCTEP_PFVF_MBOX_CMD_GET_STATS, + OCTEP_PFVF_MBOX_CMD_SET_RX_STATE, + OCTEP_PFVF_MBOX_CMD_SET_LINK_STATUS, + OCTEP_PFVF_MBOX_CMD_GET_LINK_STATUS, + OCTEP_PFVF_MBOX_CMD_GET_MTU, + OCTEP_PFVF_MBOX_CMD_DEV_REMOVE, + OCTEP_PFVF_MBOX_CMD_GET_FW_INFO, + OCTEP_PFVF_MBOX_CMD_SET_OFFLOADS, + OCTEP_PFVF_MBOX_NOTIF_LINK_STATUS, + OCTEP_PFVF_MBOX_CMD_MAX, +}; + +enum octep_pfvf_mbox_word_type { + OCTEP_PFVF_MBOX_TYPE_CMD, + OCTEP_PFVF_MBOX_TYPE_RSP_ACK, + OCTEP_PFVF_MBOX_TYPE_RSP_NACK, +}; + +enum octep_pfvf_mbox_cmd_status { + OCTEP_PFVF_MBOX_CMD_STATUS_NOT_SETUP = 1, + OCTEP_PFVF_MBOX_CMD_STATUS_TIMEDOUT = 2, + OCTEP_PFVF_MBOX_CMD_STATUS_NACK = 3, + OCTEP_PFVF_MBOX_CMD_STATUS_BUSY = 4, + OCTEP_PFVF_MBOX_CMD_STATUS_ERR = 5 +}; + +enum octep_pfvf_link_status { + OCTEP_PFVF_LINK_STATUS_DOWN, + OCTEP_PFVF_LINK_STATUS_UP, +}; + +enum octep_pfvf_link_speed { + OCTEP_PFVF_LINK_SPEED_NONE, + OCTEP_PFVF_LINK_SPEED_1000, + OCTEP_PFVF_LINK_SPEED_10000, + OCTEP_PFVF_LINK_SPEED_25000, + OCTEP_PFVF_LINK_SPEED_40000, + OCTEP_PFVF_LINK_SPEED_50000, + OCTEP_PFVF_LINK_SPEED_100000, + OCTEP_PFVF_LINK_SPEED_LAST, +}; + +enum octep_pfvf_link_duplex { + OCTEP_PFVF_LINK_HALF_DUPLEX, + OCTEP_PFVF_LINK_FULL_DUPLEX, +}; + +enum octep_pfvf_link_autoneg { + OCTEP_PFVF_LINK_AUTONEG, + OCTEP_PFVF_LINK_FIXED, +}; + +#define OCTEP_PFVF_MBOX_TIMEOUT_WAIT_COUNT 8000 +#define OCTEP_PFVF_MBOX_TIMEOUT_WAIT_UDELAY 1000 +#define OCTEP_PFVF_MBOX_MAX_RETRIES 2 +#define OCTEP_PFVF_MBOX_VERSION 0 +#define OCTEP_PFVF_MBOX_MAX_DATA_SIZE 6 +#define OCTEP_PFVF_MBOX_MAX_DATA_BUF_SIZE 320 +#define OCTEP_PFVF_MBOX_MORE_FRAG_FLAG 1 + +union octep_pfvf_mbox_word { + u64 u64; + struct { + u64 opcode:8; + u64 type:2; + u64 rsvd:6; + u64 data:48; + } s; + struct { + u64 opcode:8; + u64 type:2; + u64 frag:1; + u64 rsvd:5; + u8 data[6]; + } s_data; + struct { + u64 opcode:8; + u64 type:2; + u64 rsvd:6; + u64 version:48; + } s_version; + struct { + u64 opcode:8; + u64 type:2; + u64 rsvd:6; + u8 mac_addr[6]; + } s_set_mac; + struct { + u64 opcode:8; + u64 type:2; + u64 rsvd:6; + u64 mtu:48; + } s_set_mtu; + struct { + u64 opcode:8; + u64 type:2; + u64 state:1; + u64 rsvd:53; + } s_link_state; + struct { + u64 opcode:8; + u64 type:2; + u64 status:1; + u64 rsvd:53; + } s_link_status; + struct { + u64 opcode:8; + u64 type:2; + u64 pkind:8; + u64 fsz:8; + u64 rx_ol_flags:16; + u64 tx_ol_flags:16; + u64 rsvd:6; + } s_fw_info; + struct { + u64 opcode:8; + u64 type:2; + u64 rsvd:22; + u64 rx_ol_flags:16; + u64 tx_ol_flags:16; + } s_offloads; +} __packed; + +int octep_vf_setup_mbox(struct octep_vf_device *oct); +void octep_vf_delete_mbox(struct octep_vf_device *oct); +int octep_vf_mbox_send_cmd(struct octep_vf_device *oct, union octep_pfvf_mbox_word cmd, + union octep_pfvf_mbox_word *rsp); +int octep_vf_mbox_bulk_read(struct octep_vf_device *oct, enum octep_pfvf_mbox_opcode opcode, + u8 *data, int *size); +int octep_vf_mbox_set_mtu(struct octep_vf_device *oct, int mtu); +int octep_vf_mbox_set_mac_addr(struct octep_vf_device *oct, char *mac_addr); +int octep_vf_mbox_get_mac_addr(struct octep_vf_device *oct, char *mac_addr); +int octep_vf_mbox_version_check(struct octep_vf_device *oct); +int octep_vf_mbox_set_rx_state(struct octep_vf_device *oct, bool state); +int octep_vf_mbox_set_link_status(struct octep_vf_device *oct, bool status); +int octep_vf_mbox_get_link_status(struct octep_vf_device *oct, u8 *oper_up); +int octep_vf_mbox_dev_remove(struct octep_vf_device *oct); +int octep_vf_mbox_get_fw_info(struct octep_vf_device *oct); +int octep_vf_mbox_set_offloads(struct octep_vf_device *oct, u16 tx_offloads, u16 rx_offloads); + +#endif diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_regs_cn9k.h b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_regs_cn9k.h new file mode 100644 index 000000000000..25e2a876ebba --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_regs_cn9k.h @@ -0,0 +1,154 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ +#ifndef _OCTEP_VF_REGS_CN9K_H_ +#define _OCTEP_VF_REGS_CN9K_H_ + +/*############################ RST #########################*/ +#define CN93_VF_CONFIG_XPANSION_BAR 0x38 +#define CN93_VF_CONFIG_PCIE_CAP 0x70 +#define CN93_VF_CONFIG_PCIE_DEVCAP 0x74 +#define CN93_VF_CONFIG_PCIE_DEVCTL 0x78 +#define CN93_VF_CONFIG_PCIE_LINKCAP 0x7C +#define CN93_VF_CONFIG_PCIE_LINKCTL 0x80 +#define CN93_VF_CONFIG_PCIE_SLOTCAP 0x84 +#define CN93_VF_CONFIG_PCIE_SLOTCTL 0x88 + +#define CN93_VF_RING_OFFSET BIT_ULL(17) + +/*###################### RING IN REGISTERS #########################*/ +#define CN93_VF_SDP_R_IN_CONTROL_START 0x10000 +#define CN93_VF_SDP_R_IN_ENABLE_START 0x10010 +#define CN93_VF_SDP_R_IN_INSTR_BADDR_START 0x10020 +#define CN93_VF_SDP_R_IN_INSTR_RSIZE_START 0x10030 +#define CN93_VF_SDP_R_IN_INSTR_DBELL_START 0x10040 +#define CN93_VF_SDP_R_IN_CNTS_START 0x10050 +#define CN93_VF_SDP_R_IN_INT_LEVELS_START 0x10060 +#define CN93_VF_SDP_R_IN_PKT_CNT_START 0x10080 +#define CN93_VF_SDP_R_IN_BYTE_CNT_START 0x10090 + +#define CN93_VF_SDP_R_IN_CONTROL(ring) \ + (CN93_VF_SDP_R_IN_CONTROL_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_IN_ENABLE(ring) \ + (CN93_VF_SDP_R_IN_ENABLE_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_IN_INSTR_BADDR(ring) \ + (CN93_VF_SDP_R_IN_INSTR_BADDR_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_IN_INSTR_RSIZE(ring) \ + (CN93_VF_SDP_R_IN_INSTR_RSIZE_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_IN_INSTR_DBELL(ring) \ + (CN93_VF_SDP_R_IN_INSTR_DBELL_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_IN_CNTS(ring) \ + (CN93_VF_SDP_R_IN_CNTS_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_IN_INT_LEVELS(ring) \ + (CN93_VF_SDP_R_IN_INT_LEVELS_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_IN_PKT_CNT(ring) \ + (CN93_VF_SDP_R_IN_PKT_CNT_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_IN_BYTE_CNT(ring) \ + (CN93_VF_SDP_R_IN_BYTE_CNT_START + ((ring) * CN93_VF_RING_OFFSET)) + +/*------------------ R_IN Masks ----------------*/ + +/** Rings per Virtual Function **/ +#define CN93_VF_R_IN_CTL_RPVF_MASK (0xF) +#define CN93_VF_R_IN_CTL_RPVF_POS (48) + +/* Number of instructions to be read in one MAC read request. + * setting to Max value(4) + **/ +#define CN93_VF_R_IN_CTL_IDLE BIT_ULL(28) +#define CN93_VF_R_IN_CTL_RDSIZE (0x3ULL << 25) +#define CN93_VF_R_IN_CTL_IS_64B BIT_ULL(24) +#define CN93_VF_R_IN_CTL_D_NSR BIT_ULL(8) +#define CN93_VF_R_IN_CTL_D_ESR BIT_ULL(6) +#define CN93_VF_R_IN_CTL_D_ROR BIT_ULL(5) +#define CN93_VF_R_IN_CTL_NSR BIT_ULL(3) +#define CN93_VF_R_IN_CTL_ESR BIT_ULL(1) +#define CN93_VF_R_IN_CTL_ROR BIT_ULL(0) + +#define CN93_VF_R_IN_CTL_MASK (CN93_VF_R_IN_CTL_RDSIZE | CN93_VF_R_IN_CTL_IS_64B) + +/*###################### RING OUT REGISTERS #########################*/ +#define CN93_VF_SDP_R_OUT_CNTS_START 0x10100 +#define CN93_VF_SDP_R_OUT_INT_LEVELS_START 0x10110 +#define CN93_VF_SDP_R_OUT_SLIST_BADDR_START 0x10120 +#define CN93_VF_SDP_R_OUT_SLIST_RSIZE_START 0x10130 +#define CN93_VF_SDP_R_OUT_SLIST_DBELL_START 0x10140 +#define CN93_VF_SDP_R_OUT_CONTROL_START 0x10150 +#define CN93_VF_SDP_R_OUT_ENABLE_START 0x10160 +#define CN93_VF_SDP_R_OUT_PKT_CNT_START 0x10180 +#define CN93_VF_SDP_R_OUT_BYTE_CNT_START 0x10190 + +#define CN93_VF_SDP_R_OUT_CONTROL(ring) \ + (CN93_VF_SDP_R_OUT_CONTROL_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_OUT_ENABLE(ring) \ + (CN93_VF_SDP_R_OUT_ENABLE_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_OUT_SLIST_BADDR(ring) \ + (CN93_VF_SDP_R_OUT_SLIST_BADDR_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_OUT_SLIST_RSIZE(ring) \ + (CN93_VF_SDP_R_OUT_SLIST_RSIZE_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_OUT_SLIST_DBELL(ring) \ + (CN93_VF_SDP_R_OUT_SLIST_DBELL_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_OUT_CNTS(ring) \ + (CN93_VF_SDP_R_OUT_CNTS_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_OUT_INT_LEVELS(ring) \ + (CN93_VF_SDP_R_OUT_INT_LEVELS_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_OUT_PKT_CNT(ring) \ + (CN93_VF_SDP_R_OUT_PKT_CNT_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_OUT_BYTE_CNT(ring) \ + (CN93_VF_SDP_R_OUT_BYTE_CNT_START + ((ring) * CN93_VF_RING_OFFSET)) + +/*------------------ R_OUT Masks ----------------*/ +#define CN93_VF_R_OUT_INT_LEVELS_BMODE BIT_ULL(63) +#define CN93_VF_R_OUT_INT_LEVELS_TIMET (32) + +#define CN93_VF_R_OUT_CTL_IDLE BIT_ULL(40) +#define CN93_VF_R_OUT_CTL_ES_I BIT_ULL(34) +#define CN93_VF_R_OUT_CTL_NSR_I BIT_ULL(33) +#define CN93_VF_R_OUT_CTL_ROR_I BIT_ULL(32) +#define CN93_VF_R_OUT_CTL_ES_D BIT_ULL(30) +#define CN93_VF_R_OUT_CTL_NSR_D BIT_ULL(29) +#define CN93_VF_R_OUT_CTL_ROR_D BIT_ULL(28) +#define CN93_VF_R_OUT_CTL_ES_P BIT_ULL(26) +#define CN93_VF_R_OUT_CTL_NSR_P BIT_ULL(25) +#define CN93_VF_R_OUT_CTL_ROR_P BIT_ULL(24) +#define CN93_VF_R_OUT_CTL_IMODE BIT_ULL(23) + +/* ##################### Mail Box Registers ########################## */ +/* SDP PF to VF Mailbox Data Register */ +#define CN93_VF_SDP_R_MBOX_PF_VF_DATA_START 0x10210 +/* SDP Packet PF to VF Mailbox Interrupt Register */ +#define CN93_VF_SDP_R_MBOX_PF_VF_INT_START 0x10220 +/* SDP VF to PF Mailbox Data Register */ +#define CN93_VF_SDP_R_MBOX_VF_PF_DATA_START 0x10230 + +#define CN93_VF_SDP_R_MBOX_PF_VF_INT_ENAB BIT_ULL(1) +#define CN93_VF_SDP_R_MBOX_PF_VF_INT_STATUS BIT_ULL(0) + +#define CN93_VF_SDP_R_MBOX_PF_VF_DATA(ring) \ + (CN93_VF_SDP_R_MBOX_PF_VF_DATA_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_MBOX_PF_VF_INT(ring) \ + (CN93_VF_SDP_R_MBOX_PF_VF_INT_START + ((ring) * CN93_VF_RING_OFFSET)) + +#define CN93_VF_SDP_R_MBOX_VF_PF_DATA(ring) \ + (CN93_VF_SDP_R_MBOX_VF_PF_DATA_START + ((ring) * CN93_VF_RING_OFFSET)) +#endif /* _OCTEP_VF_REGS_CN9K_H_ */ diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_regs_cnxk.h b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_regs_cnxk.h new file mode 100644 index 000000000000..2e156745ef64 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_regs_cnxk.h @@ -0,0 +1,162 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ +#ifndef _OCTEP_VF_REGS_CNXK_H_ +#define _OCTEP_VF_REGS_CNXK_H_ + +/*############################ RST #########################*/ +#define CNXK_VF_CONFIG_XPANSION_BAR 0x38 +#define CNXK_VF_CONFIG_PCIE_CAP 0x70 +#define CNXK_VF_CONFIG_PCIE_DEVCAP 0x74 +#define CNXK_VF_CONFIG_PCIE_DEVCTL 0x78 +#define CNXK_VF_CONFIG_PCIE_LINKCAP 0x7C +#define CNXK_VF_CONFIG_PCIE_LINKCTL 0x80 +#define CNXK_VF_CONFIG_PCIE_SLOTCAP 0x84 +#define CNXK_VF_CONFIG_PCIE_SLOTCTL 0x88 + +#define CNXK_VF_RING_OFFSET (0x1ULL << 17) + +/*###################### RING IN REGISTERS #########################*/ +#define CNXK_VF_SDP_R_IN_CONTROL_START 0x10000 +#define CNXK_VF_SDP_R_IN_ENABLE_START 0x10010 +#define CNXK_VF_SDP_R_IN_INSTR_BADDR_START 0x10020 +#define CNXK_VF_SDP_R_IN_INSTR_RSIZE_START 0x10030 +#define CNXK_VF_SDP_R_IN_INSTR_DBELL_START 0x10040 +#define CNXK_VF_SDP_R_IN_CNTS_START 0x10050 +#define CNXK_VF_SDP_R_IN_INT_LEVELS_START 0x10060 +#define CNXK_VF_SDP_R_IN_PKT_CNT_START 0x10080 +#define CNXK_VF_SDP_R_IN_BYTE_CNT_START 0x10090 +#define CNXK_VF_SDP_R_ERR_TYPE_START 0x10400 + +#define CNXK_VF_SDP_R_ERR_TYPE(ring) \ + (CNXK_VF_SDP_R_ERR_TYPE_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_IN_CONTROL(ring) \ + (CNXK_VF_SDP_R_IN_CONTROL_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_IN_ENABLE(ring) \ + (CNXK_VF_SDP_R_IN_ENABLE_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_IN_INSTR_BADDR(ring) \ + (CNXK_VF_SDP_R_IN_INSTR_BADDR_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_IN_INSTR_RSIZE(ring) \ + (CNXK_VF_SDP_R_IN_INSTR_RSIZE_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_IN_INSTR_DBELL(ring) \ + (CNXK_VF_SDP_R_IN_INSTR_DBELL_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_IN_CNTS(ring) \ + (CNXK_VF_SDP_R_IN_CNTS_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_IN_INT_LEVELS(ring) \ + (CNXK_VF_SDP_R_IN_INT_LEVELS_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_IN_PKT_CNT(ring) \ + (CNXK_VF_SDP_R_IN_PKT_CNT_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_IN_BYTE_CNT(ring) \ + (CNXK_VF_SDP_R_IN_BYTE_CNT_START + ((ring) * CNXK_VF_RING_OFFSET)) + +/*------------------ R_IN Masks ----------------*/ + +/** Rings per Virtual Function **/ +#define CNXK_VF_R_IN_CTL_RPVF_MASK (0xF) +#define CNXK_VF_R_IN_CTL_RPVF_POS (48) + +/* Number of instructions to be read in one MAC read request. + * setting to Max value(4) + **/ +#define CNXK_VF_R_IN_CTL_IDLE (0x1ULL << 28) +#define CNXK_VF_R_IN_CTL_RDSIZE (0x3ULL << 25) +#define CNXK_VF_R_IN_CTL_IS_64B (0x1ULL << 24) +#define CNXK_VF_R_IN_CTL_D_NSR (0x1ULL << 8) +#define CNXK_VF_R_IN_CTL_D_ESR (0x1ULL << 6) +#define CNXK_VF_R_IN_CTL_D_ROR (0x1ULL << 5) +#define CNXK_VF_R_IN_CTL_NSR (0x1ULL << 3) +#define CNXK_VF_R_IN_CTL_ESR (0x1ULL << 1) +#define CNXK_VF_R_IN_CTL_ROR (0x1ULL << 0) + +#define CNXK_VF_R_IN_CTL_MASK (CNXK_VF_R_IN_CTL_RDSIZE | CNXK_VF_R_IN_CTL_IS_64B) + +/*###################### RING OUT REGISTERS #########################*/ +#define CNXK_VF_SDP_R_OUT_CNTS_START 0x10100 +#define CNXK_VF_SDP_R_OUT_INT_LEVELS_START 0x10110 +#define CNXK_VF_SDP_R_OUT_SLIST_BADDR_START 0x10120 +#define CNXK_VF_SDP_R_OUT_SLIST_RSIZE_START 0x10130 +#define CNXK_VF_SDP_R_OUT_SLIST_DBELL_START 0x10140 +#define CNXK_VF_SDP_R_OUT_CONTROL_START 0x10150 +#define CNXK_VF_SDP_R_OUT_WMARK_START 0x10160 +#define CNXK_VF_SDP_R_OUT_ENABLE_START 0x10170 +#define CNXK_VF_SDP_R_OUT_PKT_CNT_START 0x10180 +#define CNXK_VF_SDP_R_OUT_BYTE_CNT_START 0x10190 + +#define CNXK_VF_SDP_R_OUT_CONTROL(ring) \ + (CNXK_VF_SDP_R_OUT_CONTROL_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_OUT_ENABLE(ring) \ + (CNXK_VF_SDP_R_OUT_ENABLE_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_OUT_SLIST_BADDR(ring) \ + (CNXK_VF_SDP_R_OUT_SLIST_BADDR_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_OUT_SLIST_RSIZE(ring) \ + (CNXK_VF_SDP_R_OUT_SLIST_RSIZE_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_OUT_SLIST_DBELL(ring) \ + (CNXK_VF_SDP_R_OUT_SLIST_DBELL_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_OUT_WMARK(ring) \ + (CNXK_VF_SDP_R_OUT_WMARK_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_OUT_CNTS(ring) \ + (CNXK_VF_SDP_R_OUT_CNTS_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_OUT_INT_LEVELS(ring) \ + (CNXK_VF_SDP_R_OUT_INT_LEVELS_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_OUT_PKT_CNT(ring) \ + (CNXK_VF_SDP_R_OUT_PKT_CNT_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_OUT_BYTE_CNT(ring) \ + (CNXK_VF_SDP_R_OUT_BYTE_CNT_START + ((ring) * CNXK_VF_RING_OFFSET)) + +/*------------------ R_OUT Masks ----------------*/ +#define CNXK_VF_R_OUT_INT_LEVELS_BMODE BIT_ULL(63) +#define CNXK_VF_R_OUT_INT_LEVELS_TIMET (32) + +#define CNXK_VF_R_OUT_CTL_IDLE BIT_ULL(40) +#define CNXK_VF_R_OUT_CTL_ES_I BIT_ULL(34) +#define CNXK_VF_R_OUT_CTL_NSR_I BIT_ULL(33) +#define CNXK_VF_R_OUT_CTL_ROR_I BIT_ULL(32) +#define CNXK_VF_R_OUT_CTL_ES_D BIT_ULL(30) +#define CNXK_VF_R_OUT_CTL_NSR_D BIT_ULL(29) +#define CNXK_VF_R_OUT_CTL_ROR_D BIT_ULL(28) +#define CNXK_VF_R_OUT_CTL_ES_P BIT_ULL(26) +#define CNXK_VF_R_OUT_CTL_NSR_P BIT_ULL(25) +#define CNXK_VF_R_OUT_CTL_ROR_P BIT_ULL(24) +#define CNXK_VF_R_OUT_CTL_IMODE BIT_ULL(23) + +/* ##################### Mail Box Registers ########################## */ +/* SDP PF to VF Mailbox Data Register */ +#define CNXK_VF_SDP_R_MBOX_PF_VF_DATA_START 0x10210 +/* SDP Packet PF to VF Mailbox Interrupt Register */ +#define CNXK_VF_SDP_R_MBOX_PF_VF_INT_START 0x10220 +/* SDP VF to PF Mailbox Data Register */ +#define CNXK_VF_SDP_R_MBOX_VF_PF_DATA_START 0x10230 + +#define CNXK_VF_SDP_R_MBOX_PF_VF_INT_ENAB BIT_ULL(1) +#define CNXK_VF_SDP_R_MBOX_PF_VF_INT_STATUS BIT_ULL(0) + +#define CNXK_VF_SDP_R_MBOX_PF_VF_DATA(ring) \ + (CNXK_VF_SDP_R_MBOX_PF_VF_DATA_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_MBOX_PF_VF_INT(ring) \ + (CNXK_VF_SDP_R_MBOX_PF_VF_INT_START + ((ring) * CNXK_VF_RING_OFFSET)) + +#define CNXK_VF_SDP_R_MBOX_VF_PF_DATA(ring) \ + (CNXK_VF_SDP_R_MBOX_VF_PF_DATA_START + ((ring) * CNXK_VF_RING_OFFSET)) +#endif /* _OCTEP_VF_REGS_CNXK_H_ */ diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_rx.c b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_rx.c new file mode 100644 index 000000000000..82821bc28634 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_rx.c @@ -0,0 +1,510 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ + +#include <linux/pci.h> +#include <linux/etherdevice.h> +#include <linux/vmalloc.h> + +#include "octep_vf_config.h" +#include "octep_vf_main.h" + +static void octep_vf_oq_reset_indices(struct octep_vf_oq *oq) +{ + oq->host_read_idx = 0; + oq->host_refill_idx = 0; + oq->refill_count = 0; + oq->last_pkt_count = 0; + oq->pkts_pending = 0; +} + +/** + * octep_vf_oq_fill_ring_buffers() - fill initial receive buffers for Rx ring. + * + * @oq: Octeon Rx queue data structure. + * + * Return: 0, if successfully filled receive buffers for all descriptors. + * -ENOMEM, if failed to allocate a buffer or failed to map for DMA. + */ +static int octep_vf_oq_fill_ring_buffers(struct octep_vf_oq *oq) +{ + struct octep_vf_oq_desc_hw *desc_ring = oq->desc_ring; + struct page *page; + u32 i; + + for (i = 0; i < oq->max_count; i++) { + page = dev_alloc_page(); + if (unlikely(!page)) { + dev_err(oq->dev, "Rx buffer alloc failed\n"); + goto rx_buf_alloc_err; + } + desc_ring[i].buffer_ptr = dma_map_page(oq->dev, page, 0, + PAGE_SIZE, + DMA_FROM_DEVICE); + if (dma_mapping_error(oq->dev, desc_ring[i].buffer_ptr)) { + dev_err(oq->dev, + "OQ-%d buffer alloc: DMA mapping error!\n", + oq->q_no); + goto dma_map_err; + } + oq->buff_info[i].page = page; + } + + return 0; + +dma_map_err: + put_page(page); +rx_buf_alloc_err: + while (i) { + i--; + dma_unmap_page(oq->dev, desc_ring[i].buffer_ptr, PAGE_SIZE, DMA_FROM_DEVICE); + put_page(oq->buff_info[i].page); + oq->buff_info[i].page = NULL; + } + + return -ENOMEM; +} + +/** + * octep_vf_oq_refill() - refill buffers for used Rx ring descriptors. + * + * @oct: Octeon device private data structure. + * @oq: Octeon Rx queue data structure. + * + * Return: number of descriptors successfully refilled with receive buffers. + */ +static int octep_vf_oq_refill(struct octep_vf_device *oct, struct octep_vf_oq *oq) +{ + struct octep_vf_oq_desc_hw *desc_ring = oq->desc_ring; + struct page *page; + u32 refill_idx, i; + + refill_idx = oq->host_refill_idx; + for (i = 0; i < oq->refill_count; i++) { + page = dev_alloc_page(); + if (unlikely(!page)) { + dev_err(oq->dev, "refill: rx buffer alloc failed\n"); + oq->stats.alloc_failures++; + break; + } + + desc_ring[refill_idx].buffer_ptr = dma_map_page(oq->dev, page, 0, + PAGE_SIZE, DMA_FROM_DEVICE); + if (dma_mapping_error(oq->dev, desc_ring[refill_idx].buffer_ptr)) { + dev_err(oq->dev, + "OQ-%d buffer refill: DMA mapping error!\n", + oq->q_no); + put_page(page); + oq->stats.alloc_failures++; + break; + } + oq->buff_info[refill_idx].page = page; + refill_idx++; + if (refill_idx == oq->max_count) + refill_idx = 0; + } + oq->host_refill_idx = refill_idx; + oq->refill_count -= i; + + return i; +} + +/** + * octep_vf_setup_oq() - Setup a Rx queue. + * + * @oct: Octeon device private data structure. + * @q_no: Rx queue number to be setup. + * + * Allocate resources for a Rx queue. + */ +static int octep_vf_setup_oq(struct octep_vf_device *oct, int q_no) +{ + struct octep_vf_oq *oq; + u32 desc_ring_size; + + oq = vzalloc(sizeof(*oq)); + if (!oq) + goto create_oq_fail; + oct->oq[q_no] = oq; + + oq->octep_vf_dev = oct; + oq->netdev = oct->netdev; + oq->dev = &oct->pdev->dev; + oq->q_no = q_no; + oq->max_count = CFG_GET_OQ_NUM_DESC(oct->conf); + oq->ring_size_mask = oq->max_count - 1; + oq->buffer_size = CFG_GET_OQ_BUF_SIZE(oct->conf); + oq->max_single_buffer_size = oq->buffer_size - OCTEP_VF_OQ_RESP_HW_SIZE; + + /* When the hardware/firmware supports additional capabilities, + * additional header is filled-in by Octeon after length field in + * Rx packets. this header contains additional packet information. + */ + if (oct->fw_info.rx_ol_flags) + oq->max_single_buffer_size -= OCTEP_VF_OQ_RESP_HW_EXT_SIZE; + + oq->refill_threshold = CFG_GET_OQ_REFILL_THRESHOLD(oct->conf); + + desc_ring_size = oq->max_count * OCTEP_VF_OQ_DESC_SIZE; + oq->desc_ring = dma_alloc_coherent(oq->dev, desc_ring_size, + &oq->desc_ring_dma, GFP_KERNEL); + + if (unlikely(!oq->desc_ring)) { + dev_err(oq->dev, + "Failed to allocate DMA memory for OQ-%d !!\n", q_no); + goto desc_dma_alloc_err; + } + + oq->buff_info = vzalloc(oq->max_count * OCTEP_VF_OQ_RECVBUF_SIZE); + + if (unlikely(!oq->buff_info)) { + dev_err(&oct->pdev->dev, + "Failed to allocate buffer info for OQ-%d\n", q_no); + goto buf_list_err; + } + + if (octep_vf_oq_fill_ring_buffers(oq)) + goto oq_fill_buff_err; + + octep_vf_oq_reset_indices(oq); + oct->hw_ops.setup_oq_regs(oct, q_no); + oct->num_oqs++; + + return 0; + +oq_fill_buff_err: + vfree(oq->buff_info); + oq->buff_info = NULL; +buf_list_err: + dma_free_coherent(oq->dev, desc_ring_size, + oq->desc_ring, oq->desc_ring_dma); + oq->desc_ring = NULL; +desc_dma_alloc_err: + vfree(oq); + oct->oq[q_no] = NULL; +create_oq_fail: + return -ENOMEM; +} + +/** + * octep_vf_oq_free_ring_buffers() - Free ring buffers. + * + * @oq: Octeon Rx queue data structure. + * + * Free receive buffers in unused Rx queue descriptors. + */ +static void octep_vf_oq_free_ring_buffers(struct octep_vf_oq *oq) +{ + struct octep_vf_oq_desc_hw *desc_ring = oq->desc_ring; + int i; + + if (!oq->desc_ring || !oq->buff_info) + return; + + for (i = 0; i < oq->max_count; i++) { + if (oq->buff_info[i].page) { + dma_unmap_page(oq->dev, desc_ring[i].buffer_ptr, + PAGE_SIZE, DMA_FROM_DEVICE); + put_page(oq->buff_info[i].page); + oq->buff_info[i].page = NULL; + desc_ring[i].buffer_ptr = 0; + } + } + octep_vf_oq_reset_indices(oq); +} + +/** + * octep_vf_free_oq() - Free Rx queue resources. + * + * @oq: Octeon Rx queue data structure. + * + * Free all resources of a Rx queue. + */ +static int octep_vf_free_oq(struct octep_vf_oq *oq) +{ + struct octep_vf_device *oct = oq->octep_vf_dev; + int q_no = oq->q_no; + + octep_vf_oq_free_ring_buffers(oq); + + vfree(oq->buff_info); + + if (oq->desc_ring) + dma_free_coherent(oq->dev, + oq->max_count * OCTEP_VF_OQ_DESC_SIZE, + oq->desc_ring, oq->desc_ring_dma); + + vfree(oq); + oct->oq[q_no] = NULL; + oct->num_oqs--; + return 0; +} + +/** + * octep_vf_setup_oqs() - setup resources for all Rx queues. + * + * @oct: Octeon device private data structure. + */ +int octep_vf_setup_oqs(struct octep_vf_device *oct) +{ + int i, retval = 0; + + oct->num_oqs = 0; + for (i = 0; i < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); i++) { + retval = octep_vf_setup_oq(oct, i); + if (retval) { + dev_err(&oct->pdev->dev, + "Failed to setup OQ(RxQ)-%d.\n", i); + goto oq_setup_err; + } + dev_dbg(&oct->pdev->dev, "Successfully setup OQ(RxQ)-%d.\n", i); + } + + return 0; + +oq_setup_err: + while (i) { + i--; + octep_vf_free_oq(oct->oq[i]); + } + return retval; +} + +/** + * octep_vf_oq_dbell_init() - Initialize Rx queue doorbell. + * + * @oct: Octeon device private data structure. + * + * Write number of descriptors to Rx queue doorbell register. + */ +void octep_vf_oq_dbell_init(struct octep_vf_device *oct) +{ + int i; + + for (i = 0; i < oct->num_oqs; i++) + writel(oct->oq[i]->max_count, oct->oq[i]->pkts_credit_reg); +} + +/** + * octep_vf_free_oqs() - Free resources of all Rx queues. + * + * @oct: Octeon device private data structure. + */ +void octep_vf_free_oqs(struct octep_vf_device *oct) +{ + int i; + + for (i = 0; i < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); i++) { + if (!oct->oq[i]) + continue; + octep_vf_free_oq(oct->oq[i]); + dev_dbg(&oct->pdev->dev, + "Successfully freed OQ(RxQ)-%d.\n", i); + } +} + +/** + * octep_vf_oq_check_hw_for_pkts() - Check for new Rx packets. + * + * @oct: Octeon device private data structure. + * @oq: Octeon Rx queue data structure. + * + * Return: packets received after previous check. + */ +static int octep_vf_oq_check_hw_for_pkts(struct octep_vf_device *oct, + struct octep_vf_oq *oq) +{ + u32 pkt_count, new_pkts; + + pkt_count = readl(oq->pkts_sent_reg); + new_pkts = pkt_count - oq->last_pkt_count; + + /* Clear the hardware packets counter register if the rx queue is + * being processed continuously with-in a single interrupt and + * reached half its max value. + * this counter is not cleared every time read, to save write cycles. + */ + if (unlikely(pkt_count > 0xF0000000U)) { + writel(pkt_count, oq->pkts_sent_reg); + pkt_count = readl(oq->pkts_sent_reg); + new_pkts += pkt_count; + } + oq->last_pkt_count = pkt_count; + oq->pkts_pending += new_pkts; + return new_pkts; +} + +/** + * __octep_vf_oq_process_rx() - Process hardware Rx queue and push to stack. + * + * @oct: Octeon device private data structure. + * @oq: Octeon Rx queue data structure. + * @pkts_to_process: number of packets to be processed. + * + * Process the new packets in Rx queue. + * Packets larger than single Rx buffer arrive in consecutive descriptors. + * But, count returned by the API only accounts full packets, not fragments. + * + * Return: number of packets processed and pushed to stack. + */ +static int __octep_vf_oq_process_rx(struct octep_vf_device *oct, + struct octep_vf_oq *oq, u16 pkts_to_process) +{ + struct octep_vf_oq_resp_hw_ext *resp_hw_ext = NULL; + netdev_features_t feat = oq->netdev->features; + struct octep_vf_rx_buffer *buff_info; + struct octep_vf_oq_resp_hw *resp_hw; + u32 pkt, rx_bytes, desc_used; + u16 data_offset, rx_ol_flags; + struct sk_buff *skb; + u32 read_idx; + + read_idx = oq->host_read_idx; + rx_bytes = 0; + desc_used = 0; + for (pkt = 0; pkt < pkts_to_process; pkt++) { + buff_info = (struct octep_vf_rx_buffer *)&oq->buff_info[read_idx]; + dma_unmap_page(oq->dev, oq->desc_ring[read_idx].buffer_ptr, + PAGE_SIZE, DMA_FROM_DEVICE); + resp_hw = page_address(buff_info->page); + buff_info->page = NULL; + + /* Swap the length field that is in Big-Endian to CPU */ + buff_info->len = be64_to_cpu(resp_hw->length); + if (oct->fw_info.rx_ol_flags) { + /* Extended response header is immediately after + * response header (resp_hw) + */ + resp_hw_ext = (struct octep_vf_oq_resp_hw_ext *) + (resp_hw + 1); + buff_info->len -= OCTEP_VF_OQ_RESP_HW_EXT_SIZE; + /* Packet Data is immediately after + * extended response header. + */ + data_offset = OCTEP_VF_OQ_RESP_HW_SIZE + + OCTEP_VF_OQ_RESP_HW_EXT_SIZE; + rx_ol_flags = resp_hw_ext->rx_ol_flags; + } else { + /* Data is immediately after + * Hardware Rx response header. + */ + data_offset = OCTEP_VF_OQ_RESP_HW_SIZE; + rx_ol_flags = 0; + } + rx_bytes += buff_info->len; + + if (buff_info->len <= oq->max_single_buffer_size) { + skb = napi_build_skb((void *)resp_hw, PAGE_SIZE); + skb_reserve(skb, data_offset); + skb_put(skb, buff_info->len); + read_idx++; + desc_used++; + if (read_idx == oq->max_count) + read_idx = 0; + } else { + struct skb_shared_info *shinfo; + u16 data_len; + + skb = napi_build_skb((void *)resp_hw, PAGE_SIZE); + skb_reserve(skb, data_offset); + /* Head fragment includes response header(s); + * subsequent fragments contains only data. + */ + skb_put(skb, oq->max_single_buffer_size); + read_idx++; + desc_used++; + if (read_idx == oq->max_count) + read_idx = 0; + + shinfo = skb_shinfo(skb); + data_len = buff_info->len - oq->max_single_buffer_size; + while (data_len) { + dma_unmap_page(oq->dev, oq->desc_ring[read_idx].buffer_ptr, + PAGE_SIZE, DMA_FROM_DEVICE); + buff_info = (struct octep_vf_rx_buffer *) + &oq->buff_info[read_idx]; + if (data_len < oq->buffer_size) { + buff_info->len = data_len; + data_len = 0; + } else { + buff_info->len = oq->buffer_size; + data_len -= oq->buffer_size; + } + + skb_add_rx_frag(skb, shinfo->nr_frags, + buff_info->page, 0, + buff_info->len, + buff_info->len); + buff_info->page = NULL; + read_idx++; + desc_used++; + if (read_idx == oq->max_count) + read_idx = 0; + } + } + + skb->dev = oq->netdev; + skb->protocol = eth_type_trans(skb, skb->dev); + if (feat & NETIF_F_RXCSUM && + OCTEP_VF_RX_CSUM_VERIFIED(rx_ol_flags)) + skb->ip_summed = CHECKSUM_UNNECESSARY; + else + skb->ip_summed = CHECKSUM_NONE; + napi_gro_receive(oq->napi, skb); + } + + oq->host_read_idx = read_idx; + oq->refill_count += desc_used; + oq->stats.packets += pkt; + oq->stats.bytes += rx_bytes; + + return pkt; +} + +/** + * octep_vf_oq_process_rx() - Process Rx queue. + * + * @oq: Octeon Rx queue data structure. + * @budget: max number of packets can be processed in one invocation. + * + * Check for newly received packets and process them. + * Keeps checking for new packets until budget is used or no new packets seen. + * + * Return: number of packets processed. + */ +int octep_vf_oq_process_rx(struct octep_vf_oq *oq, int budget) +{ + u32 pkts_available, pkts_processed, total_pkts_processed; + struct octep_vf_device *oct = oq->octep_vf_dev; + + pkts_available = 0; + pkts_processed = 0; + total_pkts_processed = 0; + while (total_pkts_processed < budget) { + /* update pending count only when current one exhausted */ + if (oq->pkts_pending == 0) + octep_vf_oq_check_hw_for_pkts(oct, oq); + pkts_available = min(budget - total_pkts_processed, + oq->pkts_pending); + if (!pkts_available) + break; + + pkts_processed = __octep_vf_oq_process_rx(oct, oq, + pkts_available); + oq->pkts_pending -= pkts_processed; + total_pkts_processed += pkts_processed; + } + + if (oq->refill_count >= oq->refill_threshold) { + u32 desc_refilled = octep_vf_oq_refill(oct, oq); + + /* flush pending writes before updating credits */ + smp_wmb(); + writel(desc_refilled, oq->pkts_credit_reg); + } + + return total_pkts_processed; +} diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_rx.h b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_rx.h new file mode 100644 index 000000000000..fe46838b5200 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_rx.h @@ -0,0 +1,224 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ + +#ifndef _OCTEP_VF_RX_H_ +#define _OCTEP_VF_RX_H_ + +/* struct octep_vf_oq_desc_hw - Octeon Hardware OQ descriptor format. + * + * The descriptor ring is made of descriptors which have 2 64-bit values: + * + * @buffer_ptr: DMA address of the skb->data + * @info_ptr: DMA address of host memory, used to update pkt count by hw. + * This is currently unused to save pci writes. + */ +struct octep_vf_oq_desc_hw { + dma_addr_t buffer_ptr; + u64 info_ptr; +}; + +static_assert(sizeof(struct octep_vf_oq_desc_hw) == 16); + +#define OCTEP_VF_OQ_DESC_SIZE (sizeof(struct octep_vf_oq_desc_hw)) + +/* Rx offload flags */ +#define OCTEP_VF_RX_OFFLOAD_VLAN_STRIP BIT(0) +#define OCTEP_VF_RX_OFFLOAD_IPV4_CKSUM BIT(1) +#define OCTEP_VF_RX_OFFLOAD_UDP_CKSUM BIT(2) +#define OCTEP_VF_RX_OFFLOAD_TCP_CKSUM BIT(3) + +#define OCTEP_VF_RX_OFFLOAD_CKSUM (OCTEP_VF_RX_OFFLOAD_IPV4_CKSUM | \ + OCTEP_VF_RX_OFFLOAD_UDP_CKSUM | \ + OCTEP_VF_RX_OFFLOAD_TCP_CKSUM) + +#define OCTEP_VF_RX_IP_CSUM(flags) ((flags) & \ + (OCTEP_VF_RX_OFFLOAD_IPV4_CKSUM | \ + OCTEP_VF_RX_OFFLOAD_TCP_CKSUM | \ + OCTEP_VF_RX_OFFLOAD_UDP_CKSUM)) + +/* bit 0 is vlan strip */ +#define OCTEP_VF_RX_CSUM_IP_VERIFIED BIT(1) +#define OCTEP_VF_RX_CSUM_L4_VERIFIED BIT(2) + +#define OCTEP_VF_RX_CSUM_VERIFIED(flags) ((flags) & \ + (OCTEP_VF_RX_CSUM_L4_VERIFIED | \ + OCTEP_VF_RX_CSUM_IP_VERIFIED)) + +/* Extended Response Header in packet data received from Hardware. + * Includes metadata like checksum status. + * this is valid only if hardware/firmware published support for this. + * This is at offset 0 of packet data (skb->data). + */ +struct octep_vf_oq_resp_hw_ext { + /* Reserved. */ + u64 rsvd:48; + + /* rx offload flags */ + u16 rx_ol_flags; +}; + +static_assert(sizeof(struct octep_vf_oq_resp_hw_ext) == 8); + +#define OCTEP_VF_OQ_RESP_HW_EXT_SIZE (sizeof(struct octep_vf_oq_resp_hw_ext)) + +/* Length of Rx packet DMA'ed by Octeon to Host. + * this is in bigendian; so need to be converted to cpu endian. + * Octeon writes this at the beginning of Rx buffer (skb->data). + */ +struct octep_vf_oq_resp_hw { + /* The Length of the packet. */ + __be64 length; +}; + +static_assert(sizeof(struct octep_vf_oq_resp_hw) == 8); + +#define OCTEP_VF_OQ_RESP_HW_SIZE (sizeof(struct octep_vf_oq_resp_hw)) + +/* Pointer to data buffer. + * Driver keeps a pointer to the data buffer that it made available to + * the Octeon device. Since the descriptor ring keeps physical (bus) + * addresses, this field is required for the driver to keep track of + * the virtual address pointers. The fields are operated by + * OS-dependent routines. + */ +struct octep_vf_rx_buffer { + struct page *page; + + /* length from rx hardware descriptor after converting to cpu endian */ + u64 len; +}; + +#define OCTEP_VF_OQ_RECVBUF_SIZE (sizeof(struct octep_vf_rx_buffer)) + +/* Output Queue statistics. Each output queue has four stats fields. */ +struct octep_vf_oq_stats { + /* Number of packets received from the Device. */ + u64 packets; + + /* Number of bytes received from the Device. */ + u64 bytes; + + /* Number of times failed to allocate buffers. */ + u64 alloc_failures; +}; + +#define OCTEP_VF_OQ_STATS_SIZE (sizeof(struct octep_vf_oq_stats)) + +/* Hardware interface Rx statistics */ +struct octep_vf_iface_rx_stats { + /* Received packets */ + u64 pkts; + + /* Octets of received packets */ + u64 octets; + + /* Received PAUSE and Control packets */ + u64 pause_pkts; + + /* Received PAUSE and Control octets */ + u64 pause_octets; + + /* Filtered DMAC0 packets */ + u64 dmac0_pkts; + + /* Filtered DMAC0 octets */ + u64 dmac0_octets; + + /* Packets dropped due to RX FIFO full */ + u64 dropped_pkts_fifo_full; + + /* Octets dropped due to RX FIFO full */ + u64 dropped_octets_fifo_full; + + /* Error packets */ + u64 err_pkts; + + /* Filtered DMAC1 packets */ + u64 dmac1_pkts; + + /* Filtered DMAC1 octets */ + u64 dmac1_octets; + + /* NCSI-bound packets dropped */ + u64 ncsi_dropped_pkts; + + /* NCSI-bound octets dropped */ + u64 ncsi_dropped_octets; + + /* Multicast packets received. */ + u64 mcast_pkts; + + /* Broadcast packets received. */ + u64 bcast_pkts; + +}; + +/* The Descriptor Ring Output Queue structure. + * This structure has all the information required to implement a + * Octeon OQ. + */ +struct octep_vf_oq { + u32 q_no; + + struct octep_vf_device *octep_vf_dev; + struct net_device *netdev; + struct device *dev; + + struct napi_struct *napi; + + /* The receive buffer list. This list has the virtual addresses + * of the buffers. + */ + struct octep_vf_rx_buffer *buff_info; + + /* Pointer to the mapped packet credit register. + * Host writes number of info/buffer ptrs available to this register + */ + u8 __iomem *pkts_credit_reg; + + /* Pointer to the mapped packet sent register. + * Octeon writes the number of packets DMA'ed to host memory + * in this register. + */ + u8 __iomem *pkts_sent_reg; + + /* Statistics for this OQ. */ + struct octep_vf_oq_stats stats; + + /* Packets pending to be processed */ + u32 pkts_pending; + u32 last_pkt_count; + + /* Index in the ring where the driver should read the next packet */ + u32 host_read_idx; + + /* Number of descriptors in this ring. */ + u32 max_count; + u32 ring_size_mask; + + /* The number of descriptors pending refill. */ + u32 refill_count; + + /* Index in the ring where the driver will refill the + * descriptor's buffer + */ + u32 host_refill_idx; + u32 refill_threshold; + + /* The size of each buffer pointed by the buffer pointer. */ + u32 buffer_size; + u32 max_single_buffer_size; + + /* The 8B aligned descriptor ring starts at this address. */ + struct octep_vf_oq_desc_hw *desc_ring; + + /* DMA mapped address of the OQ descriptor ring. */ + dma_addr_t desc_ring_dma; +}; + +#define OCTEP_VF_OQ_SIZE (sizeof(struct octep_vf_oq)) +#endif /* _OCTEP_VF_RX_H_ */ diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_tx.c b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_tx.c new file mode 100644 index 000000000000..47a5c054fdb6 --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_tx.c @@ -0,0 +1,330 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ + +#include <linux/pci.h> +#include <linux/etherdevice.h> +#include <linux/vmalloc.h> +#include <net/netdev_queues.h> + +#include "octep_vf_config.h" +#include "octep_vf_main.h" + +/* Reset various index of Tx queue data structure. */ +static void octep_vf_iq_reset_indices(struct octep_vf_iq *iq) +{ + iq->fill_cnt = 0; + iq->host_write_index = 0; + iq->octep_vf_read_index = 0; + iq->flush_index = 0; + iq->pkts_processed = 0; + iq->pkt_in_done = 0; +} + +/** + * octep_vf_iq_process_completions() - Process Tx queue completions. + * + * @iq: Octeon Tx queue data structure. + * @budget: max number of completions to be processed in one invocation. + */ +int octep_vf_iq_process_completions(struct octep_vf_iq *iq, u16 budget) +{ + u32 compl_pkts, compl_bytes, compl_sg; + struct octep_vf_device *oct = iq->octep_vf_dev; + struct octep_vf_tx_buffer *tx_buffer; + struct skb_shared_info *shinfo; + u32 fi = iq->flush_index; + struct sk_buff *skb; + u8 frags, i; + + compl_pkts = 0; + compl_sg = 0; + compl_bytes = 0; + iq->octep_vf_read_index = oct->hw_ops.update_iq_read_idx(iq); + + while (likely(budget && (fi != iq->octep_vf_read_index))) { + tx_buffer = iq->buff_info + fi; + skb = tx_buffer->skb; + + fi++; + if (unlikely(fi == iq->max_count)) + fi = 0; + compl_bytes += skb->len; + compl_pkts++; + budget--; + + if (!tx_buffer->gather) { + dma_unmap_single(iq->dev, tx_buffer->dma, + tx_buffer->skb->len, DMA_TO_DEVICE); + dev_kfree_skb_any(skb); + continue; + } + + /* Scatter/Gather */ + shinfo = skb_shinfo(skb); + frags = shinfo->nr_frags; + compl_sg++; + + dma_unmap_single(iq->dev, tx_buffer->sglist[0].dma_ptr[0], + tx_buffer->sglist[0].len[3], DMA_TO_DEVICE); + + i = 1; /* entry 0 is main skb, unmapped above */ + while (frags--) { + dma_unmap_page(iq->dev, tx_buffer->sglist[i >> 2].dma_ptr[i & 3], + tx_buffer->sglist[i >> 2].len[3 - (i & 3)], DMA_TO_DEVICE); + i++; + } + + dev_kfree_skb_any(skb); + } + + iq->pkts_processed += compl_pkts; + iq->stats.instr_completed += compl_pkts; + iq->stats.bytes_sent += compl_bytes; + iq->stats.sgentry_sent += compl_sg; + iq->flush_index = fi; + + netif_subqueue_completed_wake(iq->netdev, iq->q_no, compl_pkts, + compl_bytes, IQ_INSTR_SPACE(iq), + OCTEP_VF_WAKE_QUEUE_THRESHOLD); + + return !budget; +} + +/** + * octep_vf_iq_free_pending() - Free Tx buffers for pending completions. + * + * @iq: Octeon Tx queue data structure. + */ +static void octep_vf_iq_free_pending(struct octep_vf_iq *iq) +{ + struct octep_vf_tx_buffer *tx_buffer; + struct skb_shared_info *shinfo; + u32 fi = iq->flush_index; + struct sk_buff *skb; + u8 frags, i; + + while (fi != iq->host_write_index) { + tx_buffer = iq->buff_info + fi; + skb = tx_buffer->skb; + + fi++; + if (unlikely(fi == iq->max_count)) + fi = 0; + + if (!tx_buffer->gather) { + dma_unmap_single(iq->dev, tx_buffer->dma, + tx_buffer->skb->len, DMA_TO_DEVICE); + dev_kfree_skb_any(skb); + continue; + } + + /* Scatter/Gather */ + shinfo = skb_shinfo(skb); + frags = shinfo->nr_frags; + + dma_unmap_single(iq->dev, + tx_buffer->sglist[0].dma_ptr[0], + tx_buffer->sglist[0].len[0], + DMA_TO_DEVICE); + + i = 1; /* entry 0 is main skb, unmapped above */ + while (frags--) { + dma_unmap_page(iq->dev, tx_buffer->sglist[i >> 2].dma_ptr[i & 3], + tx_buffer->sglist[i >> 2].len[i & 3], DMA_TO_DEVICE); + i++; + } + + dev_kfree_skb_any(skb); + } + + iq->flush_index = fi; + netdev_tx_reset_queue(netdev_get_tx_queue(iq->netdev, iq->q_no)); +} + +/** + * octep_vf_clean_iqs() - Clean Tx queues to shutdown the device. + * + * @oct: Octeon device private data structure. + * + * Free the buffers in Tx queue descriptors pending completion and + * reset queue indices + */ +void octep_vf_clean_iqs(struct octep_vf_device *oct) +{ + int i; + + for (i = 0; i < oct->num_iqs; i++) { + octep_vf_iq_free_pending(oct->iq[i]); + octep_vf_iq_reset_indices(oct->iq[i]); + } +} + +/** + * octep_vf_setup_iq() - Setup a Tx queue. + * + * @oct: Octeon device private data structure. + * @q_no: Tx queue number to be setup. + * + * Allocate resources for a Tx queue. + */ +static int octep_vf_setup_iq(struct octep_vf_device *oct, int q_no) +{ + u32 desc_ring_size, buff_info_size, sglist_size; + struct octep_vf_iq *iq; + int i; + + iq = vzalloc(sizeof(*iq)); + if (!iq) + goto iq_alloc_err; + oct->iq[q_no] = iq; + + iq->octep_vf_dev = oct; + iq->netdev = oct->netdev; + iq->dev = &oct->pdev->dev; + iq->q_no = q_no; + iq->max_count = CFG_GET_IQ_NUM_DESC(oct->conf); + iq->ring_size_mask = iq->max_count - 1; + iq->fill_threshold = CFG_GET_IQ_DB_MIN(oct->conf); + iq->netdev_q = netdev_get_tx_queue(iq->netdev, q_no); + + /* Allocate memory for hardware queue descriptors */ + desc_ring_size = OCTEP_VF_IQ_DESC_SIZE * CFG_GET_IQ_NUM_DESC(oct->conf); + iq->desc_ring = dma_alloc_coherent(iq->dev, desc_ring_size, + &iq->desc_ring_dma, GFP_KERNEL); + if (unlikely(!iq->desc_ring)) { + dev_err(iq->dev, + "Failed to allocate DMA memory for IQ-%d\n", q_no); + goto desc_dma_alloc_err; + } + + /* Allocate memory for hardware SGLIST descriptors */ + sglist_size = OCTEP_VF_SGLIST_SIZE_PER_PKT * + CFG_GET_IQ_NUM_DESC(oct->conf); + iq->sglist = dma_alloc_coherent(iq->dev, sglist_size, + &iq->sglist_dma, GFP_KERNEL); + if (unlikely(!iq->sglist)) { + dev_err(iq->dev, + "Failed to allocate DMA memory for IQ-%d SGLIST\n", + q_no); + goto sglist_alloc_err; + } + + /* allocate memory to manage Tx packets pending completion */ + buff_info_size = OCTEP_VF_IQ_TXBUFF_INFO_SIZE * iq->max_count; + iq->buff_info = vzalloc(buff_info_size); + if (!iq->buff_info) { + dev_err(iq->dev, + "Failed to allocate buff info for IQ-%d\n", q_no); + goto buff_info_err; + } + + /* Setup sglist addresses in tx_buffer entries */ + for (i = 0; i < CFG_GET_IQ_NUM_DESC(oct->conf); i++) { + struct octep_vf_tx_buffer *tx_buffer; + + tx_buffer = &iq->buff_info[i]; + tx_buffer->sglist = + &iq->sglist[i * OCTEP_VF_SGLIST_ENTRIES_PER_PKT]; + tx_buffer->sglist_dma = + iq->sglist_dma + (i * OCTEP_VF_SGLIST_SIZE_PER_PKT); + } + + octep_vf_iq_reset_indices(iq); + oct->hw_ops.setup_iq_regs(oct, q_no); + + oct->num_iqs++; + return 0; + +buff_info_err: + dma_free_coherent(iq->dev, sglist_size, iq->sglist, iq->sglist_dma); +sglist_alloc_err: + dma_free_coherent(iq->dev, desc_ring_size, + iq->desc_ring, iq->desc_ring_dma); +desc_dma_alloc_err: + vfree(iq); + oct->iq[q_no] = NULL; +iq_alloc_err: + return -1; +} + +/** + * octep_vf_free_iq() - Free Tx queue resources. + * + * @iq: Octeon Tx queue data structure. + * + * Free all the resources allocated for a Tx queue. + */ +static void octep_vf_free_iq(struct octep_vf_iq *iq) +{ + struct octep_vf_device *oct = iq->octep_vf_dev; + u64 desc_ring_size, sglist_size; + int q_no = iq->q_no; + + desc_ring_size = OCTEP_VF_IQ_DESC_SIZE * CFG_GET_IQ_NUM_DESC(oct->conf); + + vfree(iq->buff_info); + + if (iq->desc_ring) + dma_free_coherent(iq->dev, desc_ring_size, + iq->desc_ring, iq->desc_ring_dma); + + sglist_size = OCTEP_VF_SGLIST_SIZE_PER_PKT * + CFG_GET_IQ_NUM_DESC(oct->conf); + if (iq->sglist) + dma_free_coherent(iq->dev, sglist_size, + iq->sglist, iq->sglist_dma); + + vfree(iq); + oct->iq[q_no] = NULL; + oct->num_iqs--; +} + +/** + * octep_vf_setup_iqs() - setup resources for all Tx queues. + * + * @oct: Octeon device private data structure. + */ +int octep_vf_setup_iqs(struct octep_vf_device *oct) +{ + int i; + + oct->num_iqs = 0; + for (i = 0; i < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); i++) { + if (octep_vf_setup_iq(oct, i)) { + dev_err(&oct->pdev->dev, + "Failed to setup IQ(TxQ)-%d.\n", i); + goto iq_setup_err; + } + dev_dbg(&oct->pdev->dev, "Successfully setup IQ(TxQ)-%d.\n", i); + } + + return 0; + +iq_setup_err: + while (i) { + i--; + octep_vf_free_iq(oct->iq[i]); + } + return -1; +} + +/** + * octep_vf_free_iqs() - Free resources of all Tx queues. + * + * @oct: Octeon device private data structure. + */ +void octep_vf_free_iqs(struct octep_vf_device *oct) +{ + int i; + + for (i = 0; i < CFG_GET_PORTS_ACTIVE_IO_RINGS(oct->conf); i++) { + octep_vf_free_iq(oct->iq[i]); + dev_dbg(&oct->pdev->dev, + "Successfully destroyed IQ(TxQ)-%d.\n", i); + } + oct->num_iqs = 0; +} diff --git a/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_tx.h b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_tx.h new file mode 100644 index 000000000000..f338b975103c --- /dev/null +++ b/drivers/net/ethernet/marvell/octeon_ep_vf/octep_vf_tx.h @@ -0,0 +1,276 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* Marvell Octeon EP (EndPoint) VF Ethernet Driver + * + * Copyright (C) 2020 Marvell. + * + */ + +#ifndef _OCTEP_VF_TX_H_ +#define _OCTEP_VF_TX_H_ + +#define IQ_SEND_OK 0 +#define IQ_SEND_STOP 1 +#define IQ_SEND_FAILED -1 + +#define TX_BUFTYPE_NONE 0 +#define TX_BUFTYPE_NET 1 +#define TX_BUFTYPE_NET_SG 2 +#define NUM_TX_BUFTYPES 3 + +/* Hardware format for Scatter/Gather list + * + * 63 48|47 32|31 16|15 0 + * ----------------------------------------- + * | Len 0 | Len 1 | Len 2 | Len 3 | + * ----------------------------------------- + * | Ptr 0 | + * ----------------------------------------- + * | Ptr 1 | + * ----------------------------------------- + * | Ptr 2 | + * ----------------------------------------- + * | Ptr 3 | + * ----------------------------------------- + */ +struct octep_vf_tx_sglist_desc { + u16 len[4]; + dma_addr_t dma_ptr[4]; +}; + +static_assert(sizeof(struct octep_vf_tx_sglist_desc) == 40); + +/* Each Scatter/Gather entry sent to hardwar hold four pointers. + * So, number of entries required is (MAX_SKB_FRAGS + 1)/4, where '+1' + * is for main skb which also goes as a gather buffer to Octeon hardware. + * To allocate sufficient SGLIST entries for a packet with max fragments, + * align by adding 3 before calcuating max SGLIST entries per packet. + */ +#define OCTEP_VF_SGLIST_ENTRIES_PER_PKT ((MAX_SKB_FRAGS + 1 + 3) / 4) +#define OCTEP_VF_SGLIST_SIZE_PER_PKT \ + (OCTEP_VF_SGLIST_ENTRIES_PER_PKT * sizeof(struct octep_vf_tx_sglist_desc)) + +struct octep_vf_tx_buffer { + struct sk_buff *skb; + dma_addr_t dma; + struct octep_vf_tx_sglist_desc *sglist; + dma_addr_t sglist_dma; + u8 gather; +}; + +#define OCTEP_VF_IQ_TXBUFF_INFO_SIZE (sizeof(struct octep_vf_tx_buffer)) + +/* VF Hardware interface Tx statistics */ +struct octep_vf_iface_tx_stats { + /* Total frames sent on the interface */ + u64 pkts; + + /* Total octets sent on the interface */ + u64 octs; + + /* Packets sent to a broadcast DMAC */ + u64 bcst; + + /* Packets sent to the multicast DMAC */ + u64 mcst; + + /* Packets dropped */ + u64 dropped; + + /* Reserved */ + u64 reserved[13]; +}; + +/* VF Input Queue statistics */ +struct octep_vf_iq_stats { + /* Instructions posted to this queue. */ + u64 instr_posted; + + /* Instructions copied by hardware for processing. */ + u64 instr_completed; + + /* Instructions that could not be processed. */ + u64 instr_dropped; + + /* Bytes sent through this queue. */ + u64 bytes_sent; + + /* Gather entries sent through this queue. */ + u64 sgentry_sent; + + /* Number of transmit failures due to TX_BUSY */ + u64 tx_busy; + + /* Number of times the queue is restarted */ + u64 restart_cnt; +}; + +/* The instruction (input) queue. + * The input queue is used to post raw (instruction) mode data or packet + * data to Octeon device from the host. Each input queue (up to 4) for + * a Octeon device has one such structure to represent it. + */ +struct octep_vf_iq { + u32 q_no; + + struct octep_vf_device *octep_vf_dev; + struct net_device *netdev; + struct device *dev; + struct netdev_queue *netdev_q; + + /* Index in input ring where driver should write the next packet */ + u16 host_write_index; + + /* Index in input ring where Octeon is expected to read next packet */ + u16 octep_vf_read_index; + + /* This index aids in finding the window in the queue where Octeon + * has read the commands. + */ + u16 flush_index; + + /* Statistics for this input queue. */ + struct octep_vf_iq_stats stats; + + /* Pointer to the Virtual Base addr of the input ring. */ + struct octep_vf_tx_desc_hw *desc_ring; + + /* DMA mapped base address of the input descriptor ring. */ + dma_addr_t desc_ring_dma; + + /* Info of Tx buffers pending completion. */ + struct octep_vf_tx_buffer *buff_info; + + /* Base pointer to Scatter/Gather lists for all ring descriptors. */ + struct octep_vf_tx_sglist_desc *sglist; + + /* DMA mapped addr of Scatter Gather Lists */ + dma_addr_t sglist_dma; + + /* Octeon doorbell register for the ring. */ + u8 __iomem *doorbell_reg; + + /* Octeon instruction count register for this ring. */ + u8 __iomem *inst_cnt_reg; + + /* interrupt level register for this ring */ + u8 __iomem *intr_lvl_reg; + + /* Maximum no. of instructions in this queue. */ + u32 max_count; + u32 ring_size_mask; + + u32 pkt_in_done; + u32 pkts_processed; + + u32 status; + + /* Number of instructions pending to be posted to Octeon. */ + u32 fill_cnt; + + /* The max. number of instructions that can be held pending by the + * driver before ringing doorbell. + */ + u32 fill_threshold; +}; + +/* Hardware Tx Instruction Header */ +struct octep_vf_instr_hdr { + /* Data Len */ + u64 tlen:16; + + /* Reserved */ + u64 rsvd:20; + + /* PKIND for SDP */ + u64 pkind:6; + + /* Front Data size */ + u64 fsz:6; + + /* No. of entries in gather list */ + u64 gsz:14; + + /* Gather indicator 1=gather*/ + u64 gather:1; + + /* Reserved3 */ + u64 reserved3:1; +}; + +static_assert(sizeof(struct octep_vf_instr_hdr) == 8); + +/* Tx offload flags */ +#define OCTEP_VF_TX_OFFLOAD_VLAN_INSERT BIT(0) +#define OCTEP_VF_TX_OFFLOAD_IPV4_CKSUM BIT(1) +#define OCTEP_VF_TX_OFFLOAD_UDP_CKSUM BIT(2) +#define OCTEP_VF_TX_OFFLOAD_TCP_CKSUM BIT(3) +#define OCTEP_VF_TX_OFFLOAD_SCTP_CKSUM BIT(4) +#define OCTEP_VF_TX_OFFLOAD_TCP_TSO BIT(5) +#define OCTEP_VF_TX_OFFLOAD_UDP_TSO BIT(6) + +#define OCTEP_VF_TX_OFFLOAD_CKSUM (OCTEP_VF_TX_OFFLOAD_IPV4_CKSUM | \ + OCTEP_VF_TX_OFFLOAD_UDP_CKSUM | \ + OCTEP_VF_TX_OFFLOAD_TCP_CKSUM) + +#define OCTEP_VF_TX_OFFLOAD_TSO (OCTEP_VF_TX_OFFLOAD_TCP_TSO | \ + OCTEP_VF_TX_OFFLOAD_UDP_TSO) + +#define OCTEP_VF_TX_IP_CSUM(flags) ((flags) & \ + (OCTEP_VF_TX_OFFLOAD_IPV4_CKSUM | \ + OCTEP_VF_TX_OFFLOAD_TCP_CKSUM | \ + OCTEP_VF_TX_OFFLOAD_UDP_CKSUM)) + +#define OCTEP_VF_TX_TSO(flags) ((flags) & \ + (OCTEP_VF_TX_OFFLOAD_TCP_TSO | \ + OCTEP_VF_TX_OFFLOAD_UDP_TSO)) + +struct tx_mdata { + /* offload flags */ + u16 ol_flags; + + /* gso size */ + u16 gso_size; + + /* gso flags */ + u16 gso_segs; + + /* reserved */ + u16 rsvd1; + + /* reserved */ + u64 rsvd2; +}; + +static_assert(sizeof(struct tx_mdata) == 16); + +/* 64-byte Tx instruction format. + * Format of instruction for a 64-byte mode input queue. + * + * only first 16-bytes (dptr and ih) are mandatory; rest are optional + * and filled by the driver based on firmware/hardware capabilities. + * These optional headers together called Front Data and its size is + * described by ih->fsz. + */ +struct octep_vf_tx_desc_hw { + /* Pointer where the input data is available. */ + u64 dptr; + + /* Instruction Header. */ + union { + struct octep_vf_instr_hdr ih; + u64 ih64; + }; + + union { + u64 txm64[2]; + struct tx_mdata txm; + }; + + /* Additional headers available in a 64-byte instruction. */ + u64 exhdr[4]; +}; + +static_assert(sizeof(struct octep_vf_tx_desc_hw) == 64); + +#define OCTEP_VF_IQ_DESC_SIZE (sizeof(struct octep_vf_tx_desc_hw)) +#endif /* _OCTEP_VF_TX_H_ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c index 078f56a3cbb2..fd4ef6431142 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c @@ -935,6 +935,7 @@ void mlx5e_ptp_activate_channel(struct mlx5e_ptp *c) if (test_bit(MLX5E_PTP_STATE_RX, c->state)) { mlx5e_ptp_rx_set_fs(c->priv); mlx5e_activate_rq(&c->rq); + netif_queue_set_napi(c->netdev, c->rq.ix, NETDEV_QUEUE_TYPE_RX, &c->napi); } mlx5e_trigger_napi_sched(&c->napi); } @@ -943,8 +944,10 @@ void mlx5e_ptp_deactivate_channel(struct mlx5e_ptp *c) { int tc; - if (test_bit(MLX5E_PTP_STATE_RX, c->state)) + if (test_bit(MLX5E_PTP_STATE_RX, c->state)) { + netif_queue_set_napi(c->netdev, c->rq.ix, NETDEV_QUEUE_TYPE_RX, NULL); mlx5e_deactivate_rq(&c->rq); + } if (test_bit(MLX5E_PTP_STATE_TX, c->state)) { for (tc = 0; tc < c->num_tc; tc++) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index c8e8f512803e..be809556b2e1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -1806,6 +1806,7 @@ void mlx5e_activate_txqsq(struct mlx5e_txqsq *sq) set_bit(MLX5E_SQ_STATE_ENABLED, &sq->state); netdev_tx_reset_queue(sq->txq); netif_tx_start_queue(sq->txq); + netif_queue_set_napi(sq->netdev, sq->txq_ix, NETDEV_QUEUE_TYPE_TX, sq->cq.napi); } void mlx5e_tx_disable_queue(struct netdev_queue *txq) @@ -1819,6 +1820,7 @@ void mlx5e_deactivate_txqsq(struct mlx5e_txqsq *sq) { struct mlx5_wq_cyc *wq = &sq->wq; + netif_queue_set_napi(sq->netdev, sq->txq_ix, NETDEV_QUEUE_TYPE_TX, NULL); clear_bit(MLX5E_SQ_STATE_ENABLED, &sq->state); synchronize_net(); /* Sync with NAPI to prevent netif_tx_wake_queue. */ @@ -2560,6 +2562,7 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix, c->lag_port = mlx5e_enumerate_lag_port(priv->mdev, ix); netif_napi_add(netdev, &c->napi, mlx5e_napi_poll); + netif_napi_set_irq(&c->napi, irq); err = mlx5e_open_queues(c, params, cparam); if (unlikely(err)) @@ -2602,12 +2605,16 @@ static void mlx5e_activate_channel(struct mlx5e_channel *c) mlx5e_activate_xsk(c); else mlx5e_activate_rq(&c->rq); + + netif_queue_set_napi(c->netdev, c->ix, NETDEV_QUEUE_TYPE_RX, &c->napi); } static void mlx5e_deactivate_channel(struct mlx5e_channel *c) { int tc; + netif_queue_set_napi(c->netdev, c->ix, NETDEV_QUEUE_TYPE_RX, NULL); + if (test_bit(MLX5E_CHANNEL_STATE_XSK, c->state)) mlx5e_deactivate_xsk(c); else diff --git a/drivers/net/ethernet/realtek/r8169.h b/drivers/net/ethernet/realtek/r8169.h index c921456edeb9..4c043052198d 100644 --- a/drivers/net/ethernet/realtek/r8169.h +++ b/drivers/net/ethernet/realtek/r8169.h @@ -85,3 +85,6 @@ void r8169_get_led_name(struct rtl8169_private *tp, int idx, int rtl8168_get_led_mode(struct rtl8169_private *tp); int rtl8168_led_mod_ctrl(struct rtl8169_private *tp, u16 mask, u16 val); void rtl8168_init_leds(struct net_device *ndev); +int rtl8125_get_led_mode(struct rtl8169_private *tp, int index); +int rtl8125_set_led_mode(struct rtl8169_private *tp, int index, u16 mode); +void rtl8125_init_leds(struct net_device *ndev); diff --git a/drivers/net/ethernet/realtek/r8169_leds.c b/drivers/net/ethernet/realtek/r8169_leds.c index 78078eca3c43..7c5dc9d0df85 100644 --- a/drivers/net/ethernet/realtek/r8169_leds.c +++ b/drivers/net/ethernet/realtek/r8169_leds.c @@ -18,7 +18,14 @@ #define RTL8168_LED_CTRL_LINK_100 BIT(1) #define RTL8168_LED_CTRL_LINK_10 BIT(0) +#define RTL8125_LED_CTRL_ACT BIT(9) +#define RTL8125_LED_CTRL_LINK_2500 BIT(5) +#define RTL8125_LED_CTRL_LINK_1000 BIT(3) +#define RTL8125_LED_CTRL_LINK_100 BIT(1) +#define RTL8125_LED_CTRL_LINK_10 BIT(0) + #define RTL8168_NUM_LEDS 3 +#define RTL8125_NUM_LEDS 4 struct r8169_led_classdev { struct led_classdev led; @@ -156,3 +163,102 @@ void rtl8168_init_leds(struct net_device *ndev) for (i = 0; i < RTL8168_NUM_LEDS; i++) rtl8168_setup_ldev(leds + i, ndev, i); } + +static int rtl8125_led_hw_control_is_supported(struct led_classdev *led_cdev, + unsigned long flags) +{ + struct r8169_led_classdev *ldev = lcdev_to_r8169_ldev(led_cdev); + struct rtl8169_private *tp = netdev_priv(ldev->ndev); + + if (!r8169_trigger_mode_is_valid(flags)) { + /* Switch LED off to indicate that mode isn't supported */ + rtl8125_set_led_mode(tp, ldev->index, 0); + return -EOPNOTSUPP; + } + + return 0; +} + +static int rtl8125_led_hw_control_set(struct led_classdev *led_cdev, + unsigned long flags) +{ + struct r8169_led_classdev *ldev = lcdev_to_r8169_ldev(led_cdev); + struct rtl8169_private *tp = netdev_priv(ldev->ndev); + u16 mode = 0; + + if (flags & BIT(TRIGGER_NETDEV_LINK_10)) + mode |= RTL8125_LED_CTRL_LINK_10; + if (flags & BIT(TRIGGER_NETDEV_LINK_100)) + mode |= RTL8125_LED_CTRL_LINK_100; + if (flags & BIT(TRIGGER_NETDEV_LINK_1000)) + mode |= RTL8125_LED_CTRL_LINK_1000; + if (flags & BIT(TRIGGER_NETDEV_LINK_2500)) + mode |= RTL8125_LED_CTRL_LINK_2500; + if (flags & (BIT(TRIGGER_NETDEV_TX) | BIT(TRIGGER_NETDEV_RX))) + mode |= RTL8125_LED_CTRL_ACT; + + return rtl8125_set_led_mode(tp, ldev->index, mode); +} + +static int rtl8125_led_hw_control_get(struct led_classdev *led_cdev, + unsigned long *flags) +{ + struct r8169_led_classdev *ldev = lcdev_to_r8169_ldev(led_cdev); + struct rtl8169_private *tp = netdev_priv(ldev->ndev); + int mode; + + mode = rtl8125_get_led_mode(tp, ldev->index); + if (mode < 0) + return mode; + + if (mode & RTL8125_LED_CTRL_LINK_10) + *flags |= BIT(TRIGGER_NETDEV_LINK_10); + if (mode & RTL8125_LED_CTRL_LINK_100) + *flags |= BIT(TRIGGER_NETDEV_LINK_100); + if (mode & RTL8125_LED_CTRL_LINK_1000) + *flags |= BIT(TRIGGER_NETDEV_LINK_1000); + if (mode & RTL8125_LED_CTRL_LINK_2500) + *flags |= BIT(TRIGGER_NETDEV_LINK_2500); + if (mode & RTL8125_LED_CTRL_ACT) + *flags |= BIT(TRIGGER_NETDEV_TX) | BIT(TRIGGER_NETDEV_RX); + + return 0; +} + +static void rtl8125_setup_led_ldev(struct r8169_led_classdev *ldev, + struct net_device *ndev, int index) +{ + struct rtl8169_private *tp = netdev_priv(ndev); + struct led_classdev *led_cdev = &ldev->led; + char led_name[LED_MAX_NAME_SIZE]; + + ldev->ndev = ndev; + ldev->index = index; + + r8169_get_led_name(tp, index, led_name, LED_MAX_NAME_SIZE); + led_cdev->name = led_name; + led_cdev->hw_control_trigger = "netdev"; + led_cdev->flags |= LED_RETAIN_AT_SHUTDOWN; + led_cdev->hw_control_is_supported = rtl8125_led_hw_control_is_supported; + led_cdev->hw_control_set = rtl8125_led_hw_control_set; + led_cdev->hw_control_get = rtl8125_led_hw_control_get; + led_cdev->hw_control_get_device = r8169_led_hw_control_get_device; + + /* ignore errors */ + devm_led_classdev_register(&ndev->dev, led_cdev); +} + +void rtl8125_init_leds(struct net_device *ndev) +{ + /* bind resource mgmt to netdev */ + struct device *dev = &ndev->dev; + struct r8169_led_classdev *leds; + int i; + + leds = devm_kcalloc(dev, RTL8125_NUM_LEDS, sizeof(*leds), GFP_KERNEL); + if (!leds) + return; + + for (i = 0; i < RTL8125_NUM_LEDS; i++) + rtl8125_setup_led_ldev(leds + i, ndev, i); +} diff --git a/drivers/net/ethernet/realtek/r8169_main.c b/drivers/net/ethernet/realtek/r8169_main.c index b43db3c49d82..7d3f6d59be8c 100644 --- a/drivers/net/ethernet/realtek/r8169_main.c +++ b/drivers/net/ethernet/realtek/r8169_main.c @@ -330,17 +330,23 @@ enum rtl8168_registers { }; enum rtl8125_registers { + LEDSEL0 = 0x18, INT_CFG0_8125 = 0x34, #define INT_CFG0_ENABLE_8125 BIT(0) #define INT_CFG0_CLKREQEN BIT(3) IntrMask_8125 = 0x38, IntrStatus_8125 = 0x3c, INT_CFG1_8125 = 0x7a, + LEDSEL2 = 0x84, + LEDSEL1 = 0x86, TxPoll_8125 = 0x90, + LEDSEL3 = 0x96, MAC0_BKP = 0x19e0, EEE_TXIDLE_TIMER_8125 = 0x6048, }; +#define LEDSEL_MASK_8125 0x23f + #define RX_VLAN_INNER_8125 BIT(22) #define RX_VLAN_OUTER_8125 BIT(23) #define RX_VLAN_8125 (RX_VLAN_INNER_8125 | RX_VLAN_OUTER_8125) @@ -613,6 +619,7 @@ struct rtl8169_private { struct page *Rx_databuff[NUM_RX_DESC]; /* Rx data buffers */ struct ring_info tx_skb[NUM_TX_DESC]; /* Tx data buffers */ u16 cp_cmd; + u16 tx_lpi_timer; u32 irq_mask; int irq; struct clk *clk; @@ -830,6 +837,51 @@ int rtl8168_get_led_mode(struct rtl8169_private *tp) return ret; } +static int rtl8125_get_led_reg(int index) +{ + static const int led_regs[] = { LEDSEL0, LEDSEL1, LEDSEL2, LEDSEL3 }; + + return led_regs[index]; +} + +int rtl8125_set_led_mode(struct rtl8169_private *tp, int index, u16 mode) +{ + int reg = rtl8125_get_led_reg(index); + struct device *dev = tp_to_dev(tp); + int ret; + u16 val; + + ret = pm_runtime_resume_and_get(dev); + if (ret < 0) + return ret; + + mutex_lock(&tp->led_lock); + val = RTL_R16(tp, reg) & ~LEDSEL_MASK_8125; + RTL_W16(tp, reg, val | mode); + mutex_unlock(&tp->led_lock); + + pm_runtime_put_sync(dev); + + return 0; +} + +int rtl8125_get_led_mode(struct rtl8169_private *tp, int index) +{ + int reg = rtl8125_get_led_reg(index); + struct device *dev = tp_to_dev(tp); + int ret; + + ret = pm_runtime_resume_and_get(dev); + if (ret < 0) + return ret; + + ret = RTL_R16(tp, reg); + + pm_runtime_put_sync(dev); + + return ret; +} + void r8169_get_led_name(struct rtl8169_private *tp, int idx, char *buf, int buf_len) { @@ -1980,14 +2032,55 @@ static int rtl_set_coalesce(struct net_device *dev, return 0; } +static void rtl_set_eee_txidle_timer(struct rtl8169_private *tp) +{ + unsigned int timer_val = READ_ONCE(tp->dev->mtu) + ETH_HLEN + 0x20; + + switch (tp->mac_version) { + case RTL_GIGA_MAC_VER_46: + case RTL_GIGA_MAC_VER_48: + tp->tx_lpi_timer = timer_val; + r8168_mac_ocp_write(tp, 0xe048, timer_val); + break; + case RTL_GIGA_MAC_VER_61: + case RTL_GIGA_MAC_VER_63: + case RTL_GIGA_MAC_VER_65: + tp->tx_lpi_timer = timer_val; + RTL_W16(tp, EEE_TXIDLE_TIMER_8125, timer_val); + break; + default: + break; + } +} + +static unsigned int r8169_get_tx_lpi_timer_us(struct rtl8169_private *tp) +{ + unsigned int speed = tp->phydev->speed; + unsigned int timer = tp->tx_lpi_timer; + + if (!timer || speed == SPEED_UNKNOWN) + return 0; + + /* tx_lpi_timer value is in bytes */ + return DIV_ROUND_CLOSEST(timer * BITS_PER_BYTE, speed); +} + static int rtl8169_get_eee(struct net_device *dev, struct ethtool_keee *data) { struct rtl8169_private *tp = netdev_priv(dev); + int ret; if (!rtl_supports_eee(tp)) return -EOPNOTSUPP; - return phy_ethtool_get_eee(tp->phydev, data); + ret = phy_ethtool_get_eee(tp->phydev, data); + if (ret) + return ret; + + data->tx_lpi_timer = r8169_get_tx_lpi_timer_us(tp); + data->tx_lpi_enabled = data->tx_lpi_timer ? data->eee_enabled : false; + + return 0; } static int rtl8169_set_eee(struct net_device *dev, struct ethtool_keee *data) @@ -2238,14 +2331,8 @@ static void rtl8125a_config_eee_mac(struct rtl8169_private *tp) r8168_mac_ocp_modify(tp, 0xeb62, 0, BIT(2) | BIT(1)); } -static void rtl8125_set_eee_txidle_timer(struct rtl8169_private *tp) -{ - RTL_W16(tp, EEE_TXIDLE_TIMER_8125, tp->dev->mtu + ETH_HLEN + 0x20); -} - static void rtl8125b_config_eee_mac(struct rtl8169_private *tp) { - rtl8125_set_eee_txidle_timer(tp); r8168_mac_ocp_modify(tp, 0xe040, 0, BIT(1) | BIT(0)); } @@ -3778,6 +3865,8 @@ static void rtl_hw_start(struct rtl8169_private *tp) rtl_hw_aspm_clkreq_enable(tp, false); RTL_W16(tp, CPlusCmd, tp->cp_cmd); + rtl_set_eee_txidle_timer(tp); + if (tp->mac_version <= RTL_GIGA_MAC_VER_06) rtl_hw_start_8169(tp); else if (rtl_is_8125(tp)) @@ -3811,14 +3900,7 @@ static int rtl8169_change_mtu(struct net_device *dev, int new_mtu) dev->mtu = new_mtu; netdev_update_features(dev); rtl_jumbo_config(tp); - - switch (tp->mac_version) { - case RTL_GIGA_MAC_VER_61 ... RTL_GIGA_MAC_VER_65: - rtl8125_set_eee_txidle_timer(tp); - break; - default: - break; - } + rtl_set_eee_txidle_timer(tp); return 0; } @@ -5233,11 +5315,6 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) raw_spin_lock_init(&tp->mac_ocp_lock); mutex_init(&tp->led_lock); - dev->tstats = devm_netdev_alloc_pcpu_stats(&pdev->dev, - struct pcpu_sw_netstats); - if (!dev->tstats) - return -ENOMEM; - /* Get the *optional* external "ether_clk" used on some boards */ tp->clk = devm_clk_get_optional_enabled(&pdev->dev, "ether_clk"); if (IS_ERR(tp->clk)) @@ -5352,6 +5429,8 @@ 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; + dev->pcpu_stat_type = NETDEV_PCPU_STAT_TSTATS; + netdev_sw_irq_coalesce_default_on(dev); /* configure chip for default features */ @@ -5388,10 +5467,12 @@ static int rtl_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) if (rc) return rc; - if (IS_ENABLED(CONFIG_R8169_LEDS) && - tp->mac_version > RTL_GIGA_MAC_VER_06 && - tp->mac_version < RTL_GIGA_MAC_VER_61) - rtl8168_init_leds(dev); + if (IS_ENABLED(CONFIG_R8169_LEDS)) { + if (rtl_is_8125(tp)) + rtl8125_init_leds(dev); + else if (tp->mac_version > RTL_GIGA_MAC_VER_06) + rtl8168_init_leds(dev); + } netdev_info(dev, "%s, %pM, XID %03x, IRQ %d\n", rtl_chip_infos[chipset].name, dev->dev_addr, xid, tp->irq); diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h index 07fe852001a0..a6fefe675ef1 100644 --- a/drivers/net/ethernet/stmicro/stmmac/common.h +++ b/drivers/net/ethernet/stmicro/stmmac/common.h @@ -371,6 +371,7 @@ enum request_irq_err { REQ_IRQ_ERR_ALL, REQ_IRQ_ERR_TX, REQ_IRQ_ERR_RX, + REQ_IRQ_ERR_SFTY, REQ_IRQ_ERR_SFTY_UE, REQ_IRQ_ERR_SFTY_CE, REQ_IRQ_ERR_LPI, diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c index 31631e3f89d0..2691a250a5a7 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-qcom-ethqos.c @@ -728,7 +728,7 @@ static int qcom_ethqos_probe(struct platform_device *pdev) struct stmmac_resources stmmac_res; struct device *dev = &pdev->dev; struct qcom_ethqos *ethqos; - int ret; + int ret, i; ret = stmmac_get_platform_resources(pdev, &stmmac_res); if (ret) @@ -822,6 +822,10 @@ static int qcom_ethqos_probe(struct platform_device *pdev) plat_dat->serdes_powerdown = qcom_ethqos_serdes_powerdown; } + /* Enable TSO on queue0 and enable TBS on rest of the queues */ + for (i = 1; i < plat_dat->tx_queues_to_use; i++) + plat_dat->tx_queues_cfg[i].tbs_en = 1; + return devm_stmmac_pltfr_probe(pdev, plat_dat, &stmmac_res); } diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h index f155e4841c62..dddcaa9220cc 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h @@ -31,6 +31,7 @@ struct stmmac_resources { int wol_irq; int lpi_irq; int irq; + int sfty_irq; int sfty_ce_irq; int sfty_ue_irq; int rx_irq[MTL_MAX_RX_QUEUES]; @@ -298,6 +299,7 @@ struct stmmac_priv { void __iomem *ptpaddr; void __iomem *estaddr; unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)]; + int sfty_irq; int sfty_ce_irq; int sfty_ue_irq; int rx_irq[MTL_MAX_RX_QUEUES]; @@ -306,6 +308,7 @@ struct stmmac_priv { char int_name_mac[IFNAMSIZ + 9]; char int_name_wol[IFNAMSIZ + 9]; char int_name_lpi[IFNAMSIZ + 9]; + char int_name_sfty[IFNAMSIZ + 10]; char int_name_sfty_ce[IFNAMSIZ + 10]; char int_name_sfty_ue[IFNAMSIZ + 10]; char int_name_rx_irq[MTL_MAX_TX_QUEUES][IFNAMSIZ + 14]; diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c index 5236956cc7e4..ae2ffa9595d6 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c @@ -3597,6 +3597,10 @@ static void stmmac_free_irq(struct net_device *dev, if (priv->wol_irq > 0 && priv->wol_irq != dev->irq) free_irq(priv->wol_irq, dev); fallthrough; + case REQ_IRQ_ERR_SFTY: + if (priv->sfty_irq > 0 && priv->sfty_irq != dev->irq) + free_irq(priv->sfty_irq, dev); + fallthrough; case REQ_IRQ_ERR_WOL: free_irq(dev->irq, dev); fallthrough; @@ -3667,6 +3671,23 @@ static int stmmac_request_irq_multi_msi(struct net_device *dev) } } + /* Request the common Safety Feature Correctible/Uncorrectible + * Error line in case of another line is used + */ + if (priv->sfty_irq > 0 && priv->sfty_irq != dev->irq) { + int_name = priv->int_name_sfty; + sprintf(int_name, "%s:%s", dev->name, "safety"); + ret = request_irq(priv->sfty_irq, stmmac_safety_interrupt, + 0, int_name, dev); + if (unlikely(ret < 0)) { + netdev_err(priv->dev, + "%s: alloc sfty MSI %d (error: %d)\n", + __func__, priv->sfty_irq, ret); + irq_err = REQ_IRQ_ERR_SFTY; + goto irq_error; + } + } + /* Request the Safety Feature Correctible Error line in * case of another line is used */ @@ -3804,6 +3825,21 @@ static int stmmac_request_irq_single(struct net_device *dev) } } + /* Request the common Safety Feature Correctible/Uncorrectible + * Error line in case of another line is used + */ + if (priv->sfty_irq > 0 && priv->sfty_irq != dev->irq) { + ret = request_irq(priv->sfty_irq, stmmac_safety_interrupt, + IRQF_SHARED, dev->name, dev); + if (unlikely(ret < 0)) { + netdev_err(priv->dev, + "%s: ERROR: allocating the sfty IRQ %d (%d)\n", + __func__, priv->sfty_irq, ret); + irq_err = REQ_IRQ_ERR_SFTY; + goto irq_error; + } + } + return 0; irq_error: @@ -6025,10 +6061,8 @@ static void stmmac_common_interrupt(struct stmmac_priv *priv) priv->tx_path_in_lpi_mode = false; } - for (queue = 0; queue < queues_count; queue++) { - status = stmmac_host_mtl_irq_status(priv, priv->hw, - queue); - } + for (queue = 0; queue < queues_count; queue++) + stmmac_host_mtl_irq_status(priv, priv->hw, queue); /* PCS link status */ if (priv->hw->pcs && @@ -6063,8 +6097,8 @@ static irqreturn_t stmmac_interrupt(int irq, void *dev_id) if (test_bit(STMMAC_DOWN, &priv->state)) return IRQ_HANDLED; - /* Check if a fatal error happened */ - if (stmmac_safety_feat_interrupt(priv)) + /* Check ASP error if it isn't delivered via an individual IRQ */ + if (priv->sfty_irq <= 0 && stmmac_safety_feat_interrupt(priv)) return IRQ_HANDLED; /* To handle Common interrupts */ @@ -7513,6 +7547,7 @@ int stmmac_dvr_probe(struct device *device, priv->dev->irq = res->irq; priv->wol_irq = res->wol_irq; priv->lpi_irq = res->lpi_irq; + priv->sfty_irq = res->sfty_irq; priv->sfty_ce_irq = res->sfty_ce_irq; priv->sfty_ue_irq = res->sfty_ue_irq; for (i = 0; i < MTL_MAX_RX_QUEUES; i++) diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 70eadc83ca68..54797edc9b38 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c @@ -743,6 +743,14 @@ int stmmac_get_platform_resources(struct platform_device *pdev, dev_info(&pdev->dev, "IRQ eth_lpi not found\n"); } + stmmac_res->sfty_irq = + platform_get_irq_byname_optional(pdev, "sfty"); + if (stmmac_res->sfty_irq < 0) { + if (stmmac_res->sfty_irq == -EPROBE_DEFER) + return -EPROBE_DEFER; + dev_info(&pdev->dev, "IRQ sfty not found\n"); + } + stmmac_res->addr = devm_platform_ioremap_resource(pdev, 0); return PTR_ERR_OR_ZERO(stmmac_res->addr); diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index 23e97c2e4f6f..256602a72356 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -335,6 +335,7 @@ static int geneve_init(struct net_device *dev) gro_cells_destroy(&geneve->gro_cells); return err; } + netdev_lockdep_set_classes(dev); return 0; } diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c index f6d53e63ef4e..f6eab66c2660 100644 --- a/drivers/net/loopback.c +++ b/drivers/net/loopback.c @@ -144,6 +144,7 @@ static int loopback_dev_init(struct net_device *dev) dev->lstats = netdev_alloc_pcpu_stats(struct pcpu_lstats); if (!dev->lstats) return -ENOMEM; + netdev_lockdep_set_classes(dev); return 0; } diff --git a/drivers/net/phy/aquantia/aquantia_main.c b/drivers/net/phy/aquantia/aquantia_main.c index 97a2fafa15ca..e1f092cbfdce 100644 --- a/drivers/net/phy/aquantia/aquantia_main.c +++ b/drivers/net/phy/aquantia/aquantia_main.c @@ -727,6 +727,15 @@ static int aqr113c_config_init(struct phy_device *phydev) if (ret < 0) return ret; + ret = phy_clear_bits_mmd(phydev, MDIO_MMD_PMAPMD, MDIO_PMA_TXDIS, + MDIO_PMD_TXDIS_GLOBAL); + if (ret) + return ret; + + ret = aqr107_wait_processor_intensive_op(phydev); + if (ret) + return ret; + return aqr107_fill_interface_modes(phydev); } diff --git a/drivers/net/tun.c b/drivers/net/tun.c index b472f2c972d8..bc80fc1d576e 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1926,7 +1926,7 @@ static ssize_t tun_get_user(struct tun_struct *tun, struct tun_file *tfile, rcu_read_lock(); xdp_prog = rcu_dereference(tun->xdp_prog); if (xdp_prog) { - ret = do_xdp_generic(xdp_prog, skb); + ret = do_xdp_generic(xdp_prog, &skb); if (ret != XDP_PASS) { rcu_read_unlock(); local_bh_enable(); @@ -2516,7 +2516,7 @@ build: skb_record_rx_queue(skb, tfile->queue_index); if (skb_xdp) { - ret = do_xdp_generic(xdp_prog, skb); + ret = do_xdp_generic(xdp_prog, &skb); if (ret != XDP_PASS) { ret = 0; goto out; diff --git a/drivers/net/veth.c b/drivers/net/veth.c index 578e36ea1589..500b9dfccd08 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -729,80 +729,10 @@ static int veth_convert_skb_to_xdp_buff(struct veth_rq *rq, if (skb_shared(skb) || skb_head_is_locked(skb) || skb_shinfo(skb)->nr_frags || skb_headroom(skb) < XDP_PACKET_HEADROOM) { - u32 size, len, max_head_size, off, truesize, page_offset; - struct sk_buff *nskb; - struct page *page; - int i, head_off; - void *va; - - /* We need a private copy of the skb and data buffers since - * the ebpf program can modify it. We segment the original skb - * into order-0 pages without linearize it. - * - * Make sure we have enough space for linear and paged area - */ - max_head_size = SKB_WITH_OVERHEAD(PAGE_SIZE - - VETH_XDP_HEADROOM); - if (skb->len > PAGE_SIZE * MAX_SKB_FRAGS + max_head_size) - goto drop; - - size = min_t(u32, skb->len, max_head_size); - truesize = SKB_HEAD_ALIGN(size) + VETH_XDP_HEADROOM; - - /* Allocate skb head */ - va = page_pool_dev_alloc_va(rq->page_pool, &truesize); - if (!va) - goto drop; - - nskb = napi_build_skb(va, truesize); - if (!nskb) { - page_pool_free_va(rq->page_pool, va, true); + if (skb_pp_cow_data(rq->page_pool, pskb, XDP_PACKET_HEADROOM)) goto drop; - } - - skb_reserve(nskb, VETH_XDP_HEADROOM); - skb_copy_header(nskb, skb); - skb_mark_for_recycle(nskb); - - if (skb_copy_bits(skb, 0, nskb->data, size)) { - consume_skb(nskb); - goto drop; - } - skb_put(nskb, size); - head_off = skb_headroom(nskb) - skb_headroom(skb); - skb_headers_offset_update(nskb, head_off); - - /* Allocate paged area of new skb */ - off = size; - len = skb->len - off; - - for (i = 0; i < MAX_SKB_FRAGS && off < skb->len; i++) { - size = min_t(u32, len, PAGE_SIZE); - truesize = size; - - page = page_pool_dev_alloc(rq->page_pool, &page_offset, - &truesize); - if (!page) { - consume_skb(nskb); - goto drop; - } - - skb_add_rx_frag(nskb, i, page, page_offset, size, - truesize); - if (skb_copy_bits(skb, off, - page_address(page) + page_offset, - size)) { - consume_skb(nskb); - goto drop; - } - - len -= size; - off += size; - } - - consume_skb(skb); - skb = nskb; + skb = *pskb; } /* SKB "head" area always have tailroom for skb_shared_info */ @@ -1499,6 +1429,7 @@ static void veth_free_queues(struct net_device *dev) static int veth_dev_init(struct net_device *dev) { + netdev_lockdep_set_classes(dev); return veth_alloc_queues(dev); } diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c index 11707647afb9..705a6fd6ab6c 100644 --- a/drivers/net/vxlan/vxlan_core.c +++ b/drivers/net/vxlan/vxlan_core.c @@ -2855,6 +2855,7 @@ static int vxlan_init(struct net_device *dev) if (err) goto err_gro_cells_destroy; + netdev_lockdep_set_classes(dev); return 0; err_gro_cells_destroy: diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 07cefa32eafa..a3f9c95da51e 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -3958,7 +3958,7 @@ static inline void dev_consume_skb_any(struct sk_buff *skb) u32 bpf_prog_run_generic_xdp(struct sk_buff *skb, struct xdp_buff *xdp, struct bpf_prog *xdp_prog); void generic_xdp_tx(struct sk_buff *skb, struct bpf_prog *xdp_prog); -int do_xdp_generic(struct bpf_prog *xdp_prog, struct sk_buff *skb); +int do_xdp_generic(struct bpf_prog *xdp_prog, struct sk_buff **pskb); int netif_rx(struct sk_buff *skb); int __netif_rx(struct sk_buff *skb); diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 2dde34c29203..696e7680656f 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -3446,6 +3446,10 @@ static inline void skb_frag_ref(struct sk_buff *skb, int f) __skb_frag_ref(&skb_shinfo(skb)->frags[f]); } +int skb_pp_cow_data(struct page_pool *pool, struct sk_buff **pskb, + unsigned int headroom); +int skb_cow_data_for_xdp(struct page_pool *pool, struct sk_buff **pskb, + struct bpf_prog *prog); bool napi_pp_put_page(struct page *page, bool napi_safe); static inline void diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 360b12e61850..323c94f1845b 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -173,6 +173,9 @@ struct fib6_info { refcount_t fib6_ref; unsigned long expires; + + struct hlist_node gc_link; + struct dst_metrics *fib6_metrics; #define fib6_pmtu fib6_metrics->metrics[RTAX_MTU-1] @@ -241,12 +244,18 @@ static inline bool fib6_requires_src(const struct fib6_info *rt) return rt->fib6_src.plen > 0; } +/* The callers should hold f6i->fib6_table->tb6_lock if a route has ever + * been added to a table before. + */ static inline void fib6_clean_expires(struct fib6_info *f6i) { f6i->fib6_flags &= ~RTF_EXPIRES; f6i->expires = 0; } +/* The callers should hold f6i->fib6_table->tb6_lock if a route has ever + * been added to a table before. + */ static inline void fib6_set_expires(struct fib6_info *f6i, unsigned long expires) { @@ -327,8 +336,10 @@ static inline bool fib6_info_hold_safe(struct fib6_info *f6i) static inline void fib6_info_release(struct fib6_info *f6i) { - if (f6i && refcount_dec_and_test(&f6i->fib6_ref)) + if (f6i && refcount_dec_and_test(&f6i->fib6_ref)) { + DEBUG_NET_WARN_ON_ONCE(!hlist_unhashed(&f6i->gc_link)); call_rcu(&f6i->rcu, fib6_info_destroy_rcu); + } } enum fib6_walk_state { @@ -382,6 +393,7 @@ struct fib6_table { struct inet_peer_base tb6_peers; unsigned int flags; unsigned int fib_seq; + struct hlist_head tb6_gc_hlist; /* GC candidates */ #define RT6_TABLE_HAS_DFLT_ROUTER BIT(0) }; @@ -498,6 +510,38 @@ void fib6_gc_cleanup(void); int fib6_init(void); +/* Add the route to the gc list if it is not already there + * + * The callers should hold f6i->fib6_table->tb6_lock. + */ +static inline void fib6_add_gc_list(struct fib6_info *f6i) +{ + /* If fib6_node is null, the f6i is not in (or removed from) the + * table. + * + * There is a gap between finding the f6i from the table and + * calling this function without the protection of the tb6_lock. + * This check makes sure the f6i is not added to the gc list when + * it is not on the table. + */ + if (!rcu_dereference_protected(f6i->fib6_node, + lockdep_is_held(&f6i->fib6_table->tb6_lock))) + return; + + if (hlist_unhashed(&f6i->gc_link)) + hlist_add_head(&f6i->gc_link, &f6i->fib6_table->tb6_gc_hlist); +} + +/* Remove the route from the gc list if it is on the list. + * + * The callers should hold f6i->fib6_table->tb6_lock. + */ +static inline void fib6_remove_gc_list(struct fib6_info *f6i) +{ + if (!hlist_unhashed(&f6i->gc_link)) + hlist_del_init(&f6i->gc_link); +} + struct ipv6_route_iter { struct seq_net_private p; struct fib6_walker w; diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 28b065790261..52a51c69aa9d 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -170,7 +170,8 @@ struct fib6_info *rt6_get_dflt_router(struct net *net, struct fib6_info *rt6_add_dflt_router(struct net *net, const struct in6_addr *gwaddr, struct net_device *dev, unsigned int pref, - u32 defrtr_usr_metric); + u32 defrtr_usr_metric, + int lifetime); void rt6_purge_dflt_routers(struct net *net); diff --git a/include/net/page_pool/types.h b/include/net/page_pool/types.h index 76481c465375..3828396ae60c 100644 --- a/include/net/page_pool/types.h +++ b/include/net/page_pool/types.h @@ -128,6 +128,7 @@ struct page_pool_stats { struct page_pool { struct page_pool_params_fast p; + int cpuid; bool has_init_callback; long frag_users; @@ -203,6 +204,8 @@ struct page *page_pool_alloc_pages(struct page_pool *pool, gfp_t gfp); struct page *page_pool_alloc_frag(struct page_pool *pool, unsigned int *offset, unsigned int size, gfp_t gfp); struct page_pool *page_pool_create(const struct page_pool_params *params); +struct page_pool *page_pool_create_percpu(const struct page_pool_params *params, + int cpuid); struct xdp_mem_info; diff --git a/include/net/route.h b/include/net/route.h index 980ab474eabd..d4a0147942f1 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -37,9 +37,6 @@ #define RTO_ONLINK 0x01 -#define RT_CONN_FLAGS(sk) (RT_TOS(READ_ONCE(inet_sk(sk)->tos)) | sock_flag(sk, SOCK_LOCALROUTE)) -#define RT_CONN_FLAGS_TOS(sk,tos) (RT_TOS(tos) | sock_flag(sk, SOCK_LOCALROUTE)) - static inline __u8 ip_sock_rt_scope(const struct sock *sk) { if (sock_flag(sk, SOCK_LOCALROUTE)) @@ -163,8 +160,8 @@ static inline struct rtable *ip_route_output_ports(struct net *net, struct flowi __u8 proto, __u8 tos, int oif) { flowi4_init_output(fl4, oif, sk ? READ_ONCE(sk->sk_mark) : 0, tos, - RT_SCOPE_UNIVERSE, proto, - sk ? inet_sk_flowi_flags(sk) : 0, + sk ? ip_sock_rt_scope(sk) : RT_SCOPE_UNIVERSE, + proto, sk ? inet_sk_flowi_flags(sk) : 0, daddr, saddr, dport, sport, sock_net_uid(net, sk)); if (sk) security_sk_classify_flow(sk, flowi4_to_flowi_common(fl4)); diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 407b2335f091..790b54a7cbe3 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -504,28 +504,6 @@ static void vlan_dev_set_rx_mode(struct net_device *vlan_dev) dev_uc_sync(vlan_dev_priv(vlan_dev)->real_dev, vlan_dev); } -/* - * vlan network devices have devices nesting below it, and are a special - * "super class" of normal network devices; split their locks off into a - * separate class since they always nest. - */ -static struct lock_class_key vlan_netdev_xmit_lock_key; -static struct lock_class_key vlan_netdev_addr_lock_key; - -static void vlan_dev_set_lockdep_one(struct net_device *dev, - struct netdev_queue *txq, - void *unused) -{ - lockdep_set_class(&txq->_xmit_lock, &vlan_netdev_xmit_lock_key); -} - -static void vlan_dev_set_lockdep_class(struct net_device *dev) -{ - lockdep_set_class(&dev->addr_list_lock, - &vlan_netdev_addr_lock_key); - netdev_for_each_tx_queue(dev, vlan_dev_set_lockdep_one, NULL); -} - static __be16 vlan_parse_protocol(const struct sk_buff *skb) { struct vlan_ethhdr *veth = (struct vlan_ethhdr *)(skb->data); @@ -627,7 +605,7 @@ static int vlan_dev_init(struct net_device *dev) SET_NETDEV_DEVTYPE(dev, &vlan_type); - vlan_dev_set_lockdep_class(dev); + netdev_lockdep_set_classes(dev); vlan->vlan_pcpu_stats = netdev_alloc_pcpu_stats(struct vlan_pcpu_stats); if (!vlan->vlan_pcpu_stats) diff --git a/net/8021q/vlanproc.c b/net/8021q/vlanproc.c index 7825c129742a..87b959da00cd 100644 --- a/net/8021q/vlanproc.c +++ b/net/8021q/vlanproc.c @@ -163,48 +163,34 @@ void vlan_proc_rem_dev(struct net_device *vlandev) * The following few functions build the content of /proc/net/vlan/config */ -/* start read of /proc/net/vlan/config */ -static void *vlan_seq_start(struct seq_file *seq, loff_t *pos) - __acquires(rcu) +static void *vlan_seq_from_index(struct seq_file *seq, loff_t *pos) { + unsigned long ifindex = *pos; struct net_device *dev; - struct net *net = seq_file_net(seq); - loff_t i = 1; - - rcu_read_lock(); - if (*pos == 0) - return SEQ_START_TOKEN; - for_each_netdev_rcu(net, dev) { + for_each_netdev_dump(seq_file_net(seq), dev, ifindex) { if (!is_vlan_dev(dev)) continue; - - if (i++ == *pos) - return dev; + *pos = dev->ifindex; + return dev; } + return NULL; +} + +static void *vlan_seq_start(struct seq_file *seq, loff_t *pos) + __acquires(rcu) +{ + rcu_read_lock(); + if (*pos == 0) + return SEQ_START_TOKEN; - return NULL; + return vlan_seq_from_index(seq, pos); } static void *vlan_seq_next(struct seq_file *seq, void *v, loff_t *pos) { - struct net_device *dev; - struct net *net = seq_file_net(seq); - ++*pos; - - dev = v; - if (v == SEQ_START_TOKEN) - dev = net_device_entry(&net->dev_base_head); - - for_each_netdev_continue_rcu(net, dev) { - if (!is_vlan_dev(dev)) - continue; - - return dev; - } - - return NULL; + return vlan_seq_from_index(seq, pos); } static void vlan_seq_stop(struct seq_file *seq, void *v) diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 65cee0ad3c1b..717e9750614c 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -108,13 +108,6 @@ out: return NETDEV_TX_OK; } -static struct lock_class_key bridge_netdev_addr_lock_key; - -static void br_set_lockdep_class(struct net_device *dev) -{ - lockdep_set_class(&dev->addr_list_lock, &bridge_netdev_addr_lock_key); -} - static int br_dev_init(struct net_device *dev) { struct net_bridge *br = netdev_priv(dev); @@ -153,7 +146,7 @@ static int br_dev_init(struct net_device *dev) br_fdb_hash_fini(br); } - br_set_lockdep_class(dev); + netdev_lockdep_set_classes(dev); return err; } diff --git a/net/bridge/br_vlan.c b/net/bridge/br_vlan.c index 15f44d026e75..9c2fffb827ab 100644 --- a/net/bridge/br_vlan.c +++ b/net/bridge/br_vlan.c @@ -841,7 +841,7 @@ void br_vlan_flush(struct net_bridge *br) vg = br_vlan_group(br); __vlan_flush(br, NULL, vg); RCU_INIT_POINTER(br->vlgrp, NULL); - synchronize_rcu(); + synchronize_net(); __vlan_group_free(vg); } @@ -1372,7 +1372,7 @@ void nbp_vlan_flush(struct net_bridge_port *port) vg = nbp_vlan_group(port); __vlan_flush(port->br, port, vg); RCU_INIT_POINTER(port->vlgrp, NULL); - synchronize_rcu(); + synchronize_net(); __vlan_group_free(vg); } diff --git a/net/core/dev.c b/net/core/dev.c index 31f2c97d1990..2d02ca8a3da5 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -153,6 +153,8 @@ #include <linux/prandom.h> #include <linux/once_lite.h> #include <net/netdev_rx_queue.h> +#include <net/page_pool/types.h> +#include <net/page_pool/helpers.h> #include "dev.h" #include "net-sysfs.h" @@ -450,6 +452,12 @@ static RAW_NOTIFIER_HEAD(netdev_chain); DEFINE_PER_CPU_ALIGNED(struct softnet_data, softnet_data); EXPORT_PER_CPU_SYMBOL(softnet_data); +/* Page_pool has a lockless array/stack to alloc/recycle pages. + * PP consumers must pay attention to run APIs in the appropriate context + * (e.g. NAPI context). + */ +static DEFINE_PER_CPU_ALIGNED(struct page_pool *, system_page_pool); + #ifdef CONFIG_LOCKDEP /* * register_netdevice() inits txq->_xmit_lock and sets lockdep class @@ -1239,7 +1247,7 @@ rollback: netdev_name_node_del(dev->name_node); write_unlock(&dev_base_lock); - synchronize_rcu(); + synchronize_net(); write_lock(&dev_base_lock); netdev_name_node_add(net, dev->name_node); @@ -4866,6 +4874,12 @@ u32 bpf_prog_run_generic_xdp(struct sk_buff *skb, struct xdp_buff *xdp, xdp_init_buff(xdp, frame_sz, &rxqueue->xdp_rxq); xdp_prepare_buff(xdp, hard_start, skb_headroom(skb) - mac_len, skb_headlen(skb) + mac_len, true); + if (skb_is_nonlinear(skb)) { + skb_shinfo(skb)->xdp_frags_size = skb->data_len; + xdp_buff_set_frags_flag(xdp); + } else { + xdp_buff_clear_frags_flag(xdp); + } orig_data_end = xdp->data_end; orig_data = xdp->data; @@ -4895,6 +4909,14 @@ u32 bpf_prog_run_generic_xdp(struct sk_buff *skb, struct xdp_buff *xdp, skb->len += off; /* positive on grow, negative on shrink */ } + /* XDP frag metadata (e.g. nr_frags) are updated in eBPF helpers + * (e.g. bpf_xdp_adjust_tail), we need to update data_len here. + */ + if (xdp_buff_has_frags(xdp)) + skb->data_len = skb_shinfo(skb)->xdp_frags_size; + else + skb->data_len = 0; + /* check if XDP changed eth hdr such SKB needs update */ eth = (struct ethhdr *)xdp->data; if ((orig_eth_type != eth->h_proto) || @@ -4928,11 +4950,35 @@ u32 bpf_prog_run_generic_xdp(struct sk_buff *skb, struct xdp_buff *xdp, return act; } -static u32 netif_receive_generic_xdp(struct sk_buff *skb, +static int +netif_skb_check_for_xdp(struct sk_buff **pskb, struct bpf_prog *prog) +{ + struct sk_buff *skb = *pskb; + int err, hroom, troom; + + if (!skb_cow_data_for_xdp(this_cpu_read(system_page_pool), pskb, prog)) + return 0; + + /* In case we have to go down the path and also linearize, + * then lets do the pskb_expand_head() work just once here. + */ + hroom = XDP_PACKET_HEADROOM - skb_headroom(skb); + troom = skb->tail + skb->data_len - skb->end; + err = pskb_expand_head(skb, + hroom > 0 ? ALIGN(hroom, NET_SKB_PAD) : 0, + troom > 0 ? troom + 128 : 0, GFP_ATOMIC); + if (err) + return err; + + return skb_linearize(skb); +} + +static u32 netif_receive_generic_xdp(struct sk_buff **pskb, struct xdp_buff *xdp, struct bpf_prog *xdp_prog) { - u32 act = XDP_DROP; + struct sk_buff *skb = *pskb; + u32 mac_len, act = XDP_DROP; /* Reinjected packets coming from act_mirred or similar should * not get XDP generic processing. @@ -4940,41 +4986,36 @@ static u32 netif_receive_generic_xdp(struct sk_buff *skb, if (skb_is_redirected(skb)) return XDP_PASS; - /* XDP packets must be linear and must have sufficient headroom - * of XDP_PACKET_HEADROOM bytes. This is the guarantee that also - * native XDP provides, thus we need to do it here as well. + /* XDP packets must have sufficient headroom of XDP_PACKET_HEADROOM + * bytes. This is the guarantee that also native XDP provides, + * thus we need to do it here as well. */ + mac_len = skb->data - skb_mac_header(skb); + __skb_push(skb, mac_len); + if (skb_cloned(skb) || skb_is_nonlinear(skb) || skb_headroom(skb) < XDP_PACKET_HEADROOM) { - int hroom = XDP_PACKET_HEADROOM - skb_headroom(skb); - int troom = skb->tail + skb->data_len - skb->end; - - /* In case we have to go down the path and also linearize, - * then lets do the pskb_expand_head() work just once here. - */ - if (pskb_expand_head(skb, - hroom > 0 ? ALIGN(hroom, NET_SKB_PAD) : 0, - troom > 0 ? troom + 128 : 0, GFP_ATOMIC)) - goto do_drop; - if (skb_linearize(skb)) + if (netif_skb_check_for_xdp(pskb, xdp_prog)) goto do_drop; } - act = bpf_prog_run_generic_xdp(skb, xdp, xdp_prog); + __skb_pull(*pskb, mac_len); + + act = bpf_prog_run_generic_xdp(*pskb, xdp, xdp_prog); switch (act) { case XDP_REDIRECT: case XDP_TX: case XDP_PASS: break; default: - bpf_warn_invalid_xdp_action(skb->dev, xdp_prog, act); + bpf_warn_invalid_xdp_action((*pskb)->dev, xdp_prog, act); fallthrough; case XDP_ABORTED: - trace_xdp_exception(skb->dev, xdp_prog, act); + trace_xdp_exception((*pskb)->dev, xdp_prog, act); fallthrough; case XDP_DROP: do_drop: - kfree_skb(skb); + kfree_skb(*pskb); break; } @@ -5012,24 +5053,24 @@ void generic_xdp_tx(struct sk_buff *skb, struct bpf_prog *xdp_prog) static DEFINE_STATIC_KEY_FALSE(generic_xdp_needed_key); -int do_xdp_generic(struct bpf_prog *xdp_prog, struct sk_buff *skb) +int do_xdp_generic(struct bpf_prog *xdp_prog, struct sk_buff **pskb) { if (xdp_prog) { struct xdp_buff xdp; u32 act; int err; - act = netif_receive_generic_xdp(skb, &xdp, xdp_prog); + act = netif_receive_generic_xdp(pskb, &xdp, xdp_prog); if (act != XDP_PASS) { switch (act) { case XDP_REDIRECT: - err = xdp_do_generic_redirect(skb->dev, skb, + err = xdp_do_generic_redirect((*pskb)->dev, *pskb, &xdp, xdp_prog); if (err) goto out_redir; break; case XDP_TX: - generic_xdp_tx(skb, xdp_prog); + generic_xdp_tx(*pskb, xdp_prog); break; } return XDP_DROP; @@ -5037,7 +5078,7 @@ int do_xdp_generic(struct bpf_prog *xdp_prog, struct sk_buff *skb) } return XDP_PASS; out_redir: - kfree_skb_reason(skb, SKB_DROP_REASON_XDP); + kfree_skb_reason(*pskb, SKB_DROP_REASON_XDP); return XDP_DROP; } EXPORT_SYMBOL_GPL(do_xdp_generic); @@ -5360,7 +5401,8 @@ another_round: int ret2; migrate_disable(); - ret2 = do_xdp_generic(rcu_dereference(skb->dev->xdp_prog), skb); + ret2 = do_xdp_generic(rcu_dereference(skb->dev->xdp_prog), + &skb); migrate_enable(); if (ret2 != XDP_PASS) { @@ -11724,6 +11766,27 @@ static void __init net_dev_struct_check(void) * */ +/* We allocate 256 pages for each CPU if PAGE_SHIFT is 12 */ +#define SYSTEM_PERCPU_PAGE_POOL_SIZE ((1 << 20) / PAGE_SIZE) + +static int net_page_pool_create(int cpuid) +{ +#if IS_ENABLED(CONFIG_PAGE_POOL) + struct page_pool_params page_pool_params = { + .pool_size = SYSTEM_PERCPU_PAGE_POOL_SIZE, + .nid = NUMA_NO_NODE, + }; + struct page_pool *pp_ptr; + + pp_ptr = page_pool_create_percpu(&page_pool_params, cpuid); + if (IS_ERR(pp_ptr)) + return -ENOMEM; + + per_cpu(system_page_pool, cpuid) = pp_ptr; +#endif + return 0; +} + /* * This is called single threaded during boot, so no need * to take the rtnl semaphore. @@ -11776,6 +11839,9 @@ static int __init net_dev_init(void) init_gro_hash(&sd->backlog); sd->backlog.poll = process_backlog; sd->backlog.weight = weight_p; + + if (net_page_pool_create(i)) + goto out; } dev_boot_phase = 0; @@ -11803,6 +11869,19 @@ static int __init net_dev_init(void) WARN_ON(rc < 0); rc = 0; out: + if (rc < 0) { + for_each_possible_cpu(i) { + struct page_pool *pp_ptr; + + pp_ptr = per_cpu(system_page_pool, i); + if (!pp_ptr) + continue; + + page_pool_destroy(pp_ptr); + per_cpu(system_page_pool, i) = NULL; + } + } + return rc; } diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c index 233ec0cdd011..f0540c557515 100644 --- a/net/core/net_namespace.c +++ b/net/core/net_namespace.c @@ -622,7 +622,7 @@ static void cleanup_net(struct work_struct *work) * the rcu_barrier() below isn't sufficient alone. * Also the pre_exit() and exit() methods need this barrier. */ - synchronize_rcu(); + synchronize_rcu_expedited(); rtnl_lock(); list_for_each_entry_reverse(ops, &pernet_list, list) { diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 4933762e5a6b..89c835fcf094 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -171,13 +171,16 @@ static void page_pool_producer_unlock(struct page_pool *pool, } static int page_pool_init(struct page_pool *pool, - const struct page_pool_params *params) + const struct page_pool_params *params, + int cpuid) { unsigned int ring_qsize = 1024; /* Default */ memcpy(&pool->p, ¶ms->fast, sizeof(pool->p)); memcpy(&pool->slow, ¶ms->slow, sizeof(pool->slow)); + pool->cpuid = cpuid; + /* Validate only known flags were used */ if (pool->p.flags & ~(PP_FLAG_ALL)) return -EINVAL; @@ -253,10 +256,12 @@ static void page_pool_uninit(struct page_pool *pool) } /** - * page_pool_create() - create a page pool. + * page_pool_create_percpu() - create a page pool for a given cpu. * @params: parameters, see struct page_pool_params + * @cpuid: cpu identifier */ -struct page_pool *page_pool_create(const struct page_pool_params *params) +struct page_pool * +page_pool_create_percpu(const struct page_pool_params *params, int cpuid) { struct page_pool *pool; int err; @@ -265,7 +270,7 @@ struct page_pool *page_pool_create(const struct page_pool_params *params) if (!pool) return ERR_PTR(-ENOMEM); - err = page_pool_init(pool, params); + err = page_pool_init(pool, params, cpuid); if (err < 0) goto err_free; @@ -282,6 +287,16 @@ err_free: kfree(pool); return ERR_PTR(err); } +EXPORT_SYMBOL(page_pool_create_percpu); + +/** + * page_pool_create() - create a page pool + * @params: parameters, see struct page_pool_params + */ +struct page_pool *page_pool_create(const struct page_pool_params *params) +{ + return page_pool_create_percpu(params, -1); +} EXPORT_SYMBOL(page_pool_create); static void page_pool_return_page(struct page_pool *pool, struct page *page); diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 31f433950c8d..6f1c5537e842 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -2188,25 +2188,22 @@ static int rtnl_valid_dump_ifinfo_req(const struct nlmsghdr *nlh, static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) { + const struct rtnl_link_ops *kind_ops = NULL; struct netlink_ext_ack *extack = cb->extack; const struct nlmsghdr *nlh = cb->nlh; struct net *net = sock_net(skb->sk); - struct net *tgt_net = net; - int h, s_h; - int idx = 0, s_idx; - struct net_device *dev; - struct hlist_head *head; + unsigned int flags = NLM_F_MULTI; struct nlattr *tb[IFLA_MAX+1]; + struct { + unsigned long ifindex; + } *ctx = (void *)cb->ctx; + struct net *tgt_net = net; u32 ext_filter_mask = 0; - const struct rtnl_link_ops *kind_ops = NULL; - unsigned int flags = NLM_F_MULTI; + struct net_device *dev; int master_idx = 0; int netnsid = -1; int err, i; - s_h = cb->args[0]; - s_idx = cb->args[1]; - err = rtnl_valid_dump_ifinfo_req(nlh, cb->strict_check, tb, extack); if (err < 0) { if (cb->strict_check) @@ -2250,36 +2247,21 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) flags |= NLM_F_DUMP_FILTERED; walk_entries: - for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) { - idx = 0; - head = &tgt_net->dev_index_head[h]; - hlist_for_each_entry(dev, head, index_hlist) { - if (link_dump_filtered(dev, master_idx, kind_ops)) - goto cont; - if (idx < s_idx) - goto cont; - err = rtnl_fill_ifinfo(skb, dev, net, - RTM_NEWLINK, - NETLINK_CB(cb->skb).portid, - nlh->nlmsg_seq, 0, flags, - ext_filter_mask, 0, NULL, 0, - netnsid, GFP_KERNEL); - - if (err < 0) { - if (likely(skb->len)) - goto out; - - goto out_err; - } -cont: - idx++; + err = 0; + for_each_netdev_dump(tgt_net, dev, ctx->ifindex) { + if (link_dump_filtered(dev, master_idx, kind_ops)) + continue; + err = rtnl_fill_ifinfo(skb, dev, net, RTM_NEWLINK, + NETLINK_CB(cb->skb).portid, + nlh->nlmsg_seq, 0, flags, + ext_filter_mask, 0, NULL, 0, + netnsid, GFP_KERNEL); + if (err < 0) { + if (likely(skb->len)) + err = skb->len; + break; } } -out: - err = skb->len; -out_err: - cb->args[1] = idx; - cb->args[0] = h; cb->seq = tgt_net->dev_base_seq; nl_dump_check_consistent(cb, nlmsg_hdr(skb)); if (netnsid >= 0) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index edbbef563d4d..0d9a489e6ae1 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -895,6 +895,98 @@ static bool is_pp_page(struct page *page) return (page->pp_magic & ~0x3UL) == PP_SIGNATURE; } +int skb_pp_cow_data(struct page_pool *pool, struct sk_buff **pskb, + unsigned int headroom) +{ +#if IS_ENABLED(CONFIG_PAGE_POOL) + u32 size, truesize, len, max_head_size, off; + struct sk_buff *skb = *pskb, *nskb; + int err, i, head_off; + void *data; + + /* XDP does not support fraglist so we need to linearize + * the skb. + */ + if (skb_has_frag_list(skb)) + return -EOPNOTSUPP; + + max_head_size = SKB_WITH_OVERHEAD(PAGE_SIZE - headroom); + if (skb->len > max_head_size + MAX_SKB_FRAGS * PAGE_SIZE) + return -ENOMEM; + + size = min_t(u32, skb->len, max_head_size); + truesize = SKB_HEAD_ALIGN(size) + headroom; + data = page_pool_dev_alloc_va(pool, &truesize); + if (!data) + return -ENOMEM; + + nskb = napi_build_skb(data, truesize); + if (!nskb) { + page_pool_free_va(pool, data, true); + return -ENOMEM; + } + + skb_reserve(nskb, headroom); + skb_copy_header(nskb, skb); + skb_mark_for_recycle(nskb); + + err = skb_copy_bits(skb, 0, nskb->data, size); + if (err) { + consume_skb(nskb); + return err; + } + skb_put(nskb, size); + + head_off = skb_headroom(nskb) - skb_headroom(skb); + skb_headers_offset_update(nskb, head_off); + + off = size; + len = skb->len - off; + for (i = 0; i < MAX_SKB_FRAGS && off < skb->len; i++) { + struct page *page; + u32 page_off; + + size = min_t(u32, len, PAGE_SIZE); + truesize = size; + + page = page_pool_dev_alloc(pool, &page_off, &truesize); + if (!data) { + consume_skb(nskb); + return -ENOMEM; + } + + skb_add_rx_frag(nskb, i, page, page_off, size, truesize); + err = skb_copy_bits(skb, off, page_address(page) + page_off, + size); + if (err) { + consume_skb(nskb); + return err; + } + + len -= size; + off += size; + } + + consume_skb(skb); + *pskb = nskb; + + return 0; +#else + return -EOPNOTSUPP; +#endif +} +EXPORT_SYMBOL(skb_pp_cow_data); + +int skb_cow_data_for_xdp(struct page_pool *pool, struct sk_buff **pskb, + struct bpf_prog *prog) +{ + if (!prog->aux->xdp_has_frags) + return -EINVAL; + + return skb_pp_cow_data(pool, pskb, XDP_PACKET_HEADROOM); +} +EXPORT_SYMBOL(skb_cow_data_for_xdp); + #if IS_ENABLED(CONFIG_PAGE_POOL) bool napi_pp_put_page(struct page *page, bool napi_safe) { @@ -923,9 +1015,10 @@ bool napi_pp_put_page(struct page *page, bool napi_safe) */ if (napi_safe || in_softirq()) { const struct napi_struct *napi = READ_ONCE(pp->p.napi); + unsigned int cpuid = smp_processor_id(); - allow_direct = napi && - READ_ONCE(napi->list_owner) == smp_processor_id(); + allow_direct = napi && READ_ONCE(napi->list_owner) == cpuid; + allow_direct |= (pp->cpuid == cpuid); } /* Driver set this to memory recycling info. Reset it on recycle. diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index a5a820ee2026..ad278009e469 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1326,7 +1326,7 @@ int inet_sk_rebuild_header(struct sock *sk) fl4 = &inet->cork.fl.u.ip4; rt = ip_route_output_ports(sock_net(sk), fl4, sk, daddr, inet->inet_saddr, inet->inet_dport, inet->inet_sport, - sk->sk_protocol, RT_CONN_FLAGS(sk), + sk->sk_protocol, ip_sock_rt_tos(sk), sk->sk_bound_dev_if); if (!IS_ERR(rt)) { err = 0; diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c index 2cc50cbfc2a3..cc6d0bd7b0a9 100644 --- a/net/ipv4/datagram.c +++ b/net/ipv4/datagram.c @@ -119,7 +119,7 @@ void ip4_datagram_release_cb(struct sock *sk) rt = ip_route_output_ports(sock_net(sk), &fl4, sk, daddr, inet->inet_saddr, inet->inet_dport, inet->inet_sport, sk->sk_protocol, - RT_CONN_FLAGS(sk), sk->sk_bound_dev_if); + ip_sock_rt_tos(sk), sk->sk_bound_dev_if); dst = !IS_ERR(rt) ? &rt->dst : NULL; sk_dst_set(sk, dst); diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 3ff35f811765..0fc7ab5832d1 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -501,7 +501,7 @@ static void tnode_free(struct key_vector *tn) if (tnode_free_size >= READ_ONCE(sysctl_fib_sync_mem)) { tnode_free_size = 0; - synchronize_rcu(); + synchronize_net(); } } diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 459af1f89739..747ed7344cbe 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1467,7 +1467,7 @@ static struct dst_entry *inet_csk_rebuild_route(struct sock *sk, struct flowi *f rt = ip_route_output_ports(sock_net(sk), fl4, sk, daddr, inet->inet_saddr, inet->inet_dport, inet->inet_sport, sk->sk_protocol, - RT_CONN_FLAGS(sk), sk->sk_bound_dev_if); + ip_sock_rt_tos(sk), sk->sk_bound_dev_if); if (IS_ERR(rt)) rt = NULL; if (rt) diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 41537d18eecf..5b5a0adb927f 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -493,7 +493,7 @@ int __ip_queue_xmit(struct sock *sk, struct sk_buff *skb, struct flowi *fl, inet->inet_dport, inet->inet_sport, sk->sk_protocol, - RT_CONN_FLAGS_TOS(sk, tos), + RT_TOS(tos), sk->sk_bound_dev_if); if (IS_ERR(rt)) goto no_route; diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c index 00da0b80320f..58de780ab0e2 100644 --- a/net/ipv4/ip_tunnel.c +++ b/net/ipv4/ip_tunnel.c @@ -1269,6 +1269,7 @@ int ip_tunnel_init(struct net_device *dev) if (tunnel->collect_md) netif_keep_dst(dev); + netdev_lockdep_set_classes(dev); return 0; } EXPORT_SYMBOL_GPL(ip_tunnel_init); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index d63f5d063f07..ca1b719323c0 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1255,6 +1255,7 @@ static void cleanup_prefix_route(struct inet6_ifaddr *ifp, unsigned long expires, bool del_rt, bool del_peer) { + struct fib6_table *table; struct fib6_info *f6i; f6i = addrconf_get_prefix_route(del_peer ? &ifp->peer_addr : &ifp->addr, @@ -1264,8 +1265,15 @@ cleanup_prefix_route(struct inet6_ifaddr *ifp, unsigned long expires, if (del_rt) ip6_del_rt(dev_net(ifp->idev->dev), f6i, false); else { - if (!(f6i->fib6_flags & RTF_EXPIRES)) + if (!(f6i->fib6_flags & RTF_EXPIRES)) { + table = f6i->fib6_table; + spin_lock_bh(&table->tb6_lock); + fib6_set_expires(f6i, expires); + fib6_add_gc_list(f6i); + + spin_unlock_bh(&table->tb6_lock); + } fib6_info_release(f6i); } } @@ -2706,6 +2714,7 @@ EXPORT_SYMBOL_GPL(addrconf_prefix_rcv_add_addr); void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) { struct prefix_info *pinfo; + struct fib6_table *table; __u32 valid_lft; __u32 prefered_lft; int addr_type, err; @@ -2782,11 +2791,20 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) if (valid_lft == 0) { ip6_del_rt(net, rt, false); rt = NULL; - } else if (addrconf_finite_timeout(rt_expires)) { - /* not infinity */ - fib6_set_expires(rt, jiffies + rt_expires); } else { - fib6_clean_expires(rt); + table = rt->fib6_table; + spin_lock_bh(&table->tb6_lock); + + if (addrconf_finite_timeout(rt_expires)) { + /* not infinity */ + fib6_set_expires(rt, jiffies + rt_expires); + fib6_add_gc_list(rt); + } else { + fib6_clean_expires(rt); + fib6_remove_gc_list(rt); + } + + spin_unlock_bh(&table->tb6_lock); } } else if (valid_lft) { clock_t expires = 0; @@ -4741,6 +4759,7 @@ static int modify_prefix_route(struct inet6_ifaddr *ifp, unsigned long expires, u32 flags, bool modify_peer) { + struct fib6_table *table; struct fib6_info *f6i; u32 prio; @@ -4761,10 +4780,18 @@ static int modify_prefix_route(struct inet6_ifaddr *ifp, ifp->rt_priority, ifp->idev->dev, expires, flags, GFP_KERNEL); } else { - if (!expires) + table = f6i->fib6_table; + spin_lock_bh(&table->tb6_lock); + + if (!(flags & RTF_EXPIRES)) { fib6_clean_expires(f6i); - else + fib6_remove_gc_list(f6i); + } else { fib6_set_expires(f6i, expires); + fib6_add_gc_list(f6i); + } + + spin_unlock_bh(&table->tb6_lock); fib6_info_release(f6i); } diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 38a0348b1d17..805bbf26b3ef 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -160,6 +160,8 @@ struct fib6_info *fib6_info_alloc(gfp_t gfp_flags, bool with_fib6_nh) INIT_LIST_HEAD(&f6i->fib6_siblings); refcount_set(&f6i->fib6_ref, 1); + INIT_HLIST_NODE(&f6i->gc_link); + return f6i; } @@ -246,6 +248,7 @@ static struct fib6_table *fib6_alloc_table(struct net *net, u32 id) net->ipv6.fib6_null_entry); table->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; inet_peer_base_init(&table->tb6_peers); + INIT_HLIST_HEAD(&table->tb6_gc_hlist); } return table; @@ -1055,6 +1058,9 @@ static void fib6_purge_rt(struct fib6_info *rt, struct fib6_node *fn, lockdep_is_held(&table->tb6_lock)); } } + + fib6_clean_expires(rt); + fib6_remove_gc_list(rt); } /* @@ -1115,10 +1121,13 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct fib6_info *rt, rt->fib6_nsiblings = 0; if (!(iter->fib6_flags & RTF_EXPIRES)) return -EEXIST; - if (!(rt->fib6_flags & RTF_EXPIRES)) + if (!(rt->fib6_flags & RTF_EXPIRES)) { fib6_clean_expires(iter); - else + fib6_remove_gc_list(iter); + } else { fib6_set_expires(iter, rt->expires); + fib6_add_gc_list(iter); + } if (rt->fib6_pmtu) fib6_metric_set(iter, RTAX_MTU, @@ -1477,6 +1486,10 @@ int fib6_add(struct fib6_node *root, struct fib6_info *rt, if (rt->nh) list_add(&rt->nh_list, &rt->nh->f6i_list); __fib6_update_sernum_upto_root(rt, fib6_new_sernum(info->nl_net)); + + if (rt->fib6_flags & RTF_EXPIRES) + fib6_add_gc_list(rt); + fib6_start_gc(info->nl_net, rt); } @@ -2280,9 +2293,8 @@ static void fib6_flush_trees(struct net *net) * Garbage collection */ -static int fib6_age(struct fib6_info *rt, void *arg) +static int fib6_age(struct fib6_info *rt, struct fib6_gc_args *gc_args) { - struct fib6_gc_args *gc_args = arg; unsigned long now = jiffies; /* @@ -2307,6 +2319,42 @@ static int fib6_age(struct fib6_info *rt, void *arg) return 0; } +static void fib6_gc_table(struct net *net, + struct fib6_table *tb6, + struct fib6_gc_args *gc_args) +{ + struct fib6_info *rt; + struct hlist_node *n; + struct nl_info info = { + .nl_net = net, + .skip_notify = false, + }; + + hlist_for_each_entry_safe(rt, n, &tb6->tb6_gc_hlist, gc_link) + if (fib6_age(rt, gc_args) == -1) + fib6_del(rt, &info); +} + +static void fib6_gc_all(struct net *net, struct fib6_gc_args *gc_args) +{ + struct fib6_table *table; + struct hlist_head *head; + unsigned int h; + + rcu_read_lock(); + for (h = 0; h < FIB6_TABLE_HASHSZ; h++) { + head = &net->ipv6.fib_table_hash[h]; + hlist_for_each_entry_rcu(table, head, tb6_hlist) { + spin_lock_bh(&table->tb6_lock); + + fib6_gc_table(net, table, gc_args); + + spin_unlock_bh(&table->tb6_lock); + } + } + rcu_read_unlock(); +} + void fib6_run_gc(unsigned long expires, struct net *net, bool force) { struct fib6_gc_args gc_args; @@ -2322,7 +2370,7 @@ void fib6_run_gc(unsigned long expires, struct net *net, bool force) net->ipv6.sysctl.ip6_rt_gc_interval; gc_args.more = 0; - fib6_clean_all(net, fib6_age, &gc_args); + fib6_gc_all(net, &gc_args); now = jiffies; net->ipv6.ip6_rt_last_gc = now; @@ -2382,6 +2430,7 @@ static int __net_init fib6_net_init(struct net *net) net->ipv6.fib6_main_tbl->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; inet_peer_base_init(&net->ipv6.fib6_main_tbl->tb6_peers); + INIT_HLIST_HEAD(&net->ipv6.fib6_main_tbl->tb6_gc_hlist); #ifdef CONFIG_IPV6_MULTIPLE_TABLES net->ipv6.fib6_local_tbl = kzalloc(sizeof(*net->ipv6.fib6_local_tbl), @@ -2394,6 +2443,7 @@ static int __net_init fib6_net_init(struct net *net) net->ipv6.fib6_local_tbl->tb6_root.fn_flags = RTN_ROOT | RTN_TL_ROOT | RTN_RTINFO; inet_peer_base_init(&net->ipv6.fib6_local_tbl->tb6_peers); + INIT_HLIST_HEAD(&net->ipv6.fib6_local_tbl->tb6_gc_hlist); #endif fib6_tables_init(net); diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c index 428f03e9da45..5e97e0aa8e07 100644 --- a/net/ipv6/ip6_gre.c +++ b/net/ipv6/ip6_gre.c @@ -1511,6 +1511,7 @@ static int ip6gre_tunnel_init_common(struct net_device *dev) ip6gre_tnl_init_features(dev); netdev_hold(dev, &tunnel->dev_tracker, GFP_KERNEL); + netdev_lockdep_set_classes(dev); return 0; cleanup_dst_cache_init: @@ -1901,6 +1902,7 @@ static int ip6erspan_tap_init(struct net_device *dev) ip6erspan_tnl_link_config(tunnel, 1); netdev_hold(dev, &tunnel->dev_tracker, GFP_KERNEL); + netdev_lockdep_set_classes(dev); return 0; cleanup_dst_cache_init: diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index bfb0a6c601c1..44406c28445d 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -1898,6 +1898,7 @@ ip6_tnl_dev_init_gen(struct net_device *dev) dev->max_mtu = IP6_MAX_MTU - dev->hard_header_len; netdev_hold(dev, &t->dev_tracker, GFP_KERNEL); + netdev_lockdep_set_classes(dev); return 0; destroy_dst: diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index cfe1b1ad4d85..7f4f976aa24a 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -935,6 +935,7 @@ static inline int vti6_dev_init_gen(struct net_device *dev) if (!dev->tstats) return -ENOMEM; netdev_hold(dev, &t->dev_tracker, GFP_KERNEL); + netdev_lockdep_set_classes(dev); return 0; } diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index bc6e0a0bad3c..76ee1615ff2a 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -2719,7 +2719,6 @@ void ipv6_mc_down(struct inet6_dev *idev) /* Should stop work after group drop. or we will * start work again in mld_ifc_event() */ - synchronize_net(); mld_query_stop_work(idev); mld_report_stop_work(idev); diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index a19999b30bc0..73cb31afe935 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -1237,6 +1237,7 @@ static enum skb_drop_reason ndisc_router_discovery(struct sk_buff *skb) struct ndisc_options ndopts; struct fib6_info *rt = NULL; struct inet6_dev *in6_dev; + struct fib6_table *table; u32 defrtr_usr_metric; unsigned int pref = 0; __u32 old_if_flags; @@ -1382,7 +1383,8 @@ static enum skb_drop_reason ndisc_router_discovery(struct sk_buff *skb) neigh_release(neigh); rt = rt6_add_dflt_router(net, &ipv6_hdr(skb)->saddr, - skb->dev, pref, defrtr_usr_metric); + skb->dev, pref, defrtr_usr_metric, + lifetime); if (!rt) { ND_PRINTK(0, err, "RA: %s failed to add default route\n", @@ -1409,8 +1411,15 @@ static enum skb_drop_reason ndisc_router_discovery(struct sk_buff *skb) inet6_rt_notify(RTM_NEWROUTE, rt, &nlinfo, NLM_F_REPLACE); } - if (rt) + if (rt) { + table = rt->fib6_table; + spin_lock_bh(&table->tb6_lock); + fib6_set_expires(rt, jiffies + (HZ * lifetime)); + fib6_add_gc_list(rt); + + spin_unlock_bh(&table->tb6_lock); + } if (in6_dev->cnf.accept_ra_min_hop_limit < 256 && ra_msg->icmph.icmp6_hop_limit) { if (in6_dev->cnf.accept_ra_min_hop_limit <= ra_msg->icmph.icmp6_hop_limit) { diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 63b4c6056582..707d65bc9c0e 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -931,6 +931,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, struct net *net = dev_net(dev); struct route_info *rinfo = (struct route_info *) opt; struct in6_addr prefix_buf, *prefix; + struct fib6_table *table; unsigned int pref; unsigned long lifetime; struct fib6_info *rt; @@ -989,10 +990,18 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, (rt->fib6_flags & ~RTF_PREF_MASK) | RTF_PREF(pref); if (rt) { - if (!addrconf_finite_timeout(lifetime)) + table = rt->fib6_table; + spin_lock_bh(&table->tb6_lock); + + if (!addrconf_finite_timeout(lifetime)) { fib6_clean_expires(rt); - else + fib6_remove_gc_list(rt); + } else { fib6_set_expires(rt, jiffies + HZ * lifetime); + fib6_add_gc_list(rt); + } + + spin_unlock_bh(&table->tb6_lock); fib6_info_release(rt); } @@ -3765,8 +3774,6 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg, if (cfg->fc_flags & RTF_EXPIRES) fib6_set_expires(rt, jiffies + clock_t_to_jiffies(cfg->fc_expires)); - else - fib6_clean_expires(rt); if (cfg->fc_protocol == RTPROT_UNSPEC) cfg->fc_protocol = RTPROT_BOOT; @@ -4355,7 +4362,8 @@ struct fib6_info *rt6_add_dflt_router(struct net *net, const struct in6_addr *gwaddr, struct net_device *dev, unsigned int pref, - u32 defrtr_usr_metric) + u32 defrtr_usr_metric, + int lifetime) { struct fib6_config cfg = { .fc_table = l3mdev_fib_table(dev) ? : RT6_TABLE_DFLT, @@ -4368,6 +4376,7 @@ struct fib6_info *rt6_add_dflt_router(struct net *net, .fc_nlinfo.portid = 0, .fc_nlinfo.nlh = NULL, .fc_nlinfo.nl_net = net, + .fc_expires = jiffies_to_clock_t(lifetime * HZ), }; cfg.fc_gateway = *gwaddr; diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index 61b2b71fa8be..b2da1f1b5fec 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -1460,6 +1460,7 @@ static int ipip6_tunnel_init(struct net_device *dev) return err; } netdev_hold(dev, &tunnel->dev_tracker, GFP_KERNEL); + netdev_lockdep_set_classes(dev); return 0; } diff --git a/net/l2tp/l2tp_ip.c b/net/l2tp/l2tp_ip.c index 9a2a9ed3ba47..970af3983d11 100644 --- a/net/l2tp/l2tp_ip.c +++ b/net/l2tp/l2tp_ip.c @@ -478,7 +478,7 @@ static int l2tp_ip_sendmsg(struct sock *sk, struct msghdr *msg, size_t len) rt = ip_route_output_ports(sock_net(sk), fl4, sk, daddr, inet->inet_saddr, inet->inet_dport, inet->inet_sport, - sk->sk_protocol, RT_CONN_FLAGS(sk), + sk->sk_protocol, ip_sock_rt_tos(sk), sk->sk_bound_dev_if); if (IS_ERR(rt)) goto no_route; diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 2e5f3864d353..90e6bd2c3000 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -2530,7 +2530,7 @@ void nf_conntrack_cleanup_net_list(struct list_head *net_exit_list) * netfilter framework. Roll on, two-stage module * delete... */ - synchronize_net(); + synchronize_rcu_expedited(); i_see_dead_people: busy = 0; list_for_each_entry(net, net_exit_list, exit_list) { diff --git a/net/sched/Kconfig b/net/sched/Kconfig index 470c70deffe2..8180d0c12fce 100644 --- a/net/sched/Kconfig +++ b/net/sched/Kconfig @@ -737,16 +737,6 @@ config NET_ACT_SAMPLE To compile this code as a module, choose M here: the module will be called act_sample. -config NET_ACT_IPT - tristate "IPtables targets" - depends on NET_CLS_ACT && NETFILTER && NETFILTER_XTABLES - help - Say Y here to be able to invoke iptables targets after successful - classification. - - To compile this code as a module, choose M here: the - module will be called act_ipt. - config NET_ACT_NAT tristate "Stateless NAT" depends on NET_CLS_ACT diff --git a/net/sched/sch_codel.c b/net/sched/sch_codel.c index 61904d3a593b..ecb3f164bb25 100644 --- a/net/sched/sch_codel.c +++ b/net/sched/sch_codel.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause /* * Codel - The Controlled-Delay Active Queue Management algorithm * @@ -7,37 +8,6 @@ * Implemented on linux by : * Copyright (C) 2012 Michael D. Taht <dave.taht@bufferbloat.net> * Copyright (C) 2012,2015 Eric Dumazet <edumazet@google.com> - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions, and the following disclaimer, - * without modification. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * 3. The names of the authors may not be used to endorse or promote products - * derived from this software without specific prior written permission. - * - * Alternatively, provided that this notice is retained in full, this - * software may be distributed under the terms of the GNU General - * Public License ("GPL") version 2, in which case the provisions of the - * GPL apply INSTEAD OF those given above. - * - * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS - * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT - * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR - * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT - * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, - * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT - * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, - * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY - * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT - * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE - * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH - * DAMAGE. - * */ #include <linux/module.h> diff --git a/tools/testing/selftests/net/fib_tests.sh b/tools/testing/selftests/net/fib_tests.sh index b3ecccbbfcd2..3ec1050e47a2 100755 --- a/tools/testing/selftests/net/fib_tests.sh +++ b/tools/testing/selftests/net/fib_tests.sh @@ -743,6 +743,43 @@ fib_notify_test() cleanup &> /dev/null } +# Create a new dummy_10 to remove all associated routes. +reset_dummy_10() +{ + $IP link del dev dummy_10 + + $IP link add dummy_10 type dummy + $IP link set dev dummy_10 up + $IP -6 address add 2001:10::1/64 dev dummy_10 +} + +check_rt_num() +{ + local expected=$1 + local num=$2 + + if [ $num -ne $expected ]; then + echo "FAIL: Expected $expected routes, got $num" + ret=1 + else + ret=0 + fi +} + +check_rt_num_clean() +{ + local expected=$1 + local num=$2 + + if [ $num -ne $expected ]; then + log_test 1 0 "expected $expected routes, got $num" + set +e + cleanup &> /dev/null + return 1 + fi + return 0 +} + fib6_gc_test() { setup @@ -751,7 +788,7 @@ fib6_gc_test() echo "Fib6 garbage collection test" set -e - EXPIRE=3 + EXPIRE=5 # Check expiration of routes every $EXPIRE seconds (GC) $NS_EXEC sysctl -wq net.ipv6.route.gc_interval=$EXPIRE @@ -763,44 +800,114 @@ fib6_gc_test() $NS_EXEC sysctl -wq net.ipv6.route.flush=1 # Temporary routes - for i in $(seq 1 1000); do + for i in $(seq 1 5); do # Expire route after $EXPIRE seconds $IP -6 route add 2001:20::$i \ via 2001:10::2 dev dummy_10 expires $EXPIRE done - sleep $(($EXPIRE * 2)) - N_EXP_SLEEP=$($IP -6 route list |grep expires|wc -l) - if [ $N_EXP_SLEEP -ne 0 ]; then - echo "FAIL: expected 0 routes with expires, got $N_EXP_SLEEP" - ret=1 - else - ret=0 - fi + sleep $(($EXPIRE * 2 + 1)) + $NS_EXEC sysctl -wq net.ipv6.route.flush=1 + check_rt_num 0 $($IP -6 route list |grep expires|wc -l) + log_test $ret 0 "ipv6 route garbage collection" + + reset_dummy_10 # Permanent routes - for i in $(seq 1 5000); do + for i in $(seq 1 5); do $IP -6 route add 2001:30::$i \ via 2001:10::2 dev dummy_10 done # Temporary routes - for i in $(seq 1 1000); do + for i in $(seq 1 5); do # Expire route after $EXPIRE seconds $IP -6 route add 2001:20::$i \ via 2001:10::2 dev dummy_10 expires $EXPIRE done - sleep $(($EXPIRE * 2)) - N_EXP_SLEEP=$($IP -6 route list |grep expires|wc -l) - if [ $N_EXP_SLEEP -ne 0 ]; then - echo "FAIL: expected 0 routes with expires," \ - "got $N_EXP_SLEEP (5000 permanent routes)" - ret=1 - else - ret=0 + sleep $(($EXPIRE * 2 + 1)) + check_rt_num 0 $($IP -6 route list |grep expires|wc -l) + log_test $ret 0 "ipv6 route garbage collection (with permanent routes)" + + reset_dummy_10 + + # Permanent routes + for i in $(seq 1 5); do + $IP -6 route add 2001:20::$i \ + via 2001:10::2 dev dummy_10 + done + # Replace with temporary routes + for i in $(seq 1 5); do + # Expire route after $EXPIRE seconds + $IP -6 route replace 2001:20::$i \ + via 2001:10::2 dev dummy_10 expires $EXPIRE + done + check_rt_num_clean 5 $($IP -6 route list |grep expires|wc -l) || return + # Wait for GC + sleep $(($EXPIRE * 2 + 1)) + $NS_EXEC sysctl -wq net.ipv6.route.flush=1 + check_rt_num 0 $($IP -6 route list |grep expires|wc -l) + log_test $ret 0 "ipv6 route garbage collection (replace with expires)" + + reset_dummy_10 + + # Temporary routes + for i in $(seq 1 5); do + # Expire route after $EXPIRE seconds + $IP -6 route add 2001:20::$i \ + via 2001:10::2 dev dummy_10 expires $EXPIRE + done + # Replace with permanent routes + for i in $(seq 1 5); do + $IP -6 route replace 2001:20::$i \ + via 2001:10::2 dev dummy_10 + done + check_rt_num_clean 0 $($IP -6 route list |grep expires|wc -l) || return + + # Wait for GC + sleep $(($EXPIRE * 2 + 1)) + + check_rt_num 5 $($IP -6 route list |grep -v expires|grep 2001:20::|wc -l) + log_test $ret 0 "ipv6 route garbage collection (replace with permanent)" + + # ra6 is required for the next test. (ipv6toolkit) + if [ ! -x "$(command -v ra6)" ]; then + echo "SKIP: ra6 not found." + set +e + cleanup &> /dev/null + return fi - set +e + # Delete dummy_10 and remove all routes + $IP link del dev dummy_10 - log_test $ret 0 "ipv6 route garbage collection" + # Create a pair of veth devices to send a RA message from one + # device to another. + $IP link add veth1 type veth peer name veth2 + $IP link set dev veth1 up + $IP link set dev veth2 up + $IP -6 address add 2001:10::1/64 dev veth1 nodad + $IP -6 address add 2001:10::2/64 dev veth2 nodad + + # Make veth1 ready to receive RA messages. + $NS_EXEC sysctl -wq net.ipv6.conf.veth1.accept_ra=2 + + # Send a RA message with a route from veth2 to veth1. + $NS_EXEC ra6 -i veth2 -d 2001:10::1 -t $EXPIRE + + # Wait for the RA message. + sleep 1 + + # systemd may mess up the test. You syould make sure that + # systemd-networkd.service and systemd-networkd.socket are stopped. + check_rt_num_clean 1 $($IP -6 route list|grep expires|wc -l) || return + + # Wait for GC + sleep $(($EXPIRE * 2 + 1)) + + $NS_EXEC sysctl -wq net.ipv6.route.flush=1 + check_rt_num 0 $($IP -6 route list |grep expires|wc -l) + log_test $ret 0 "ipv6 route garbage collection (RA message)" + + set +e cleanup &> /dev/null } diff --git a/tools/testing/selftests/net/txtimestamp.c b/tools/testing/selftests/net/txtimestamp.c index 10f2fde3686b..ec60a16c9307 100644 --- a/tools/testing/selftests/net/txtimestamp.c +++ b/tools/testing/selftests/net/txtimestamp.c @@ -163,7 +163,8 @@ static void validate_timestamp(struct timespec *cur, int min_delay) if (cur64 < start64 + min_delay || cur64 > start64 + max_delay) { fprintf(stderr, "ERROR: %" PRId64 " us expected between %d and %d\n", cur64 - start64, min_delay, max_delay); - test_failed = true; + if (!getenv("KSFT_MACHINE_SLOW")) + test_failed = true; } } |