diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2018-04-04 00:04:18 +0300 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2018-04-04 00:04:18 +0300 |
commit | 5bb053bef82523a8fd78d650bca81c9f114fa276 (patch) | |
tree | 58c2fe47f60bb69230bb05d57a6c9e3f47f7b1fe /drivers/net/ethernet/mellanox/mlx5 | |
parent | bb2407a7219760926760f0448fddf00d625e5aec (diff) | |
parent | 159f02977b2feb18a4bece5e586c838a6d26d44b (diff) | |
download | linux-5bb053bef82523a8fd78d650bca81c9f114fa276.tar.xz |
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next
Pull networking updates from David Miller:
1) Support offloading wireless authentication to userspace via
NL80211_CMD_EXTERNAL_AUTH, from Srinivas Dasari.
2) A lot of work on network namespace setup/teardown from Kirill Tkhai.
Setup and cleanup of namespaces now all run asynchronously and thus
performance is significantly increased.
3) Add rx/tx timestamping support to mv88e6xxx driver, from Brandon
Streiff.
4) Support zerocopy on RDS sockets, from Sowmini Varadhan.
5) Use denser instruction encoding in x86 eBPF JIT, from Daniel
Borkmann.
6) Support hw offload of vlan filtering in mvpp2 dreiver, from Maxime
Chevallier.
7) Support grafting of child qdiscs in mlxsw driver, from Nogah
Frankel.
8) Add packet forwarding tests to selftests, from Ido Schimmel.
9) Deal with sub-optimal GSO packets better in BBR congestion control,
from Eric Dumazet.
10) Support 5-tuple hashing in ipv6 multipath routing, from David Ahern.
11) Add path MTU tests to selftests, from Stefano Brivio.
12) Various bits of IPSEC offloading support for mlx5, from Aviad
Yehezkel, Yossi Kuperman, and Saeed Mahameed.
13) Support RSS spreading on ntuple filters in SFC driver, from Edward
Cree.
14) Lots of sockmap work from John Fastabend. Applications can use eBPF
to filter sendmsg and sendpage operations.
15) In-kernel receive TLS support, from Dave Watson.
16) Add XDP support to ixgbevf, this is significant because it should
allow optimized XDP usage in various cloud environments. From Tony
Nguyen.
17) Add new Intel E800 series "ice" ethernet driver, from Anirudh
Venkataramanan et al.
18) IP fragmentation match offload support in nfp driver, from Pieter
Jansen van Vuuren.
19) Support XDP redirect in i40e driver, from Björn Töpel.
20) Add BPF_RAW_TRACEPOINT program type for accessing the arguments of
tracepoints in their raw form, from Alexei Starovoitov.
21) Lots of striding RQ improvements to mlx5 driver with many
performance improvements, from Tariq Toukan.
22) Use rhashtable for inet frag reassembly, from Eric Dumazet.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1678 commits)
net: mvneta: improve suspend/resume
net: mvneta: split rxq/txq init and txq deinit into SW and HW parts
ipv6: frags: fix /proc/sys/net/ipv6/ip6frag_low_thresh
net: bgmac: Fix endian access in bgmac_dma_tx_ring_free()
net: bgmac: Correctly annotate register space
route: check sysctl_fib_multipath_use_neigh earlier than hash
fix typo in command value in drivers/net/phy/mdio-bitbang.
sky2: Increase D3 delay to sky2 stops working after suspend
net/mlx5e: Set EQE based as default TX interrupt moderation mode
ibmvnic: Disable irqs before exiting reset from closed state
net: sched: do not emit messages while holding spinlock
vlan: also check phy_driver ts_info for vlan's real device
Bluetooth: Mark expected switch fall-throughs
Bluetooth: Set HCI_QUIRK_SIMULTANEOUS_DISCOVERY for BTUSB_QCA_ROME
Bluetooth: btrsi: remove unused including <linux/version.h>
Bluetooth: hci_bcm: Remove DMI quirk for the MINIX Z83-4
sh_eth: kill useless check in __sh_eth_get_regs()
sh_eth: add sh_eth_cpu_data::no_xdfar flag
ipv6: factorize sk_wmem_alloc updates done by __ip6_append_data()
ipv4: factorize sk_wmem_alloc updates done by __ip_append_data()
...
Diffstat (limited to 'drivers/net/ethernet/mellanox/mlx5')
41 files changed, 3398 insertions, 1291 deletions
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.c index 53e69edaedde..9f1b1939716a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.c @@ -37,24 +37,11 @@ #include "mlx5_core.h" #include "fpga/ipsec.h" -void *mlx5_accel_ipsec_sa_cmd_exec(struct mlx5_core_dev *mdev, - struct mlx5_accel_ipsec_sa *cmd) -{ - if (!MLX5_IPSEC_DEV(mdev)) - return ERR_PTR(-EOPNOTSUPP); - - return mlx5_fpga_ipsec_sa_cmd_exec(mdev, cmd); -} - -int mlx5_accel_ipsec_sa_cmd_wait(void *ctx) -{ - return mlx5_fpga_ipsec_sa_cmd_wait(ctx); -} - u32 mlx5_accel_ipsec_device_caps(struct mlx5_core_dev *mdev) { return mlx5_fpga_ipsec_device_caps(mdev); } +EXPORT_SYMBOL_GPL(mlx5_accel_ipsec_device_caps); unsigned int mlx5_accel_ipsec_counters_count(struct mlx5_core_dev *mdev) { @@ -67,6 +54,21 @@ int mlx5_accel_ipsec_counters_read(struct mlx5_core_dev *mdev, u64 *counters, return mlx5_fpga_ipsec_counters_read(mdev, counters, count); } +void *mlx5_accel_esp_create_hw_context(struct mlx5_core_dev *mdev, + struct mlx5_accel_esp_xfrm *xfrm, + const __be32 saddr[4], + const __be32 daddr[4], + const __be32 spi, bool is_ipv6) +{ + return mlx5_fpga_ipsec_create_sa_ctx(mdev, xfrm, saddr, daddr, + spi, is_ipv6); +} + +void mlx5_accel_esp_free_hw_context(void *context) +{ + mlx5_fpga_ipsec_delete_sa_ctx(context); +} + int mlx5_accel_ipsec_init(struct mlx5_core_dev *mdev) { return mlx5_fpga_ipsec_init(mdev); @@ -76,3 +78,32 @@ void mlx5_accel_ipsec_cleanup(struct mlx5_core_dev *mdev) { mlx5_fpga_ipsec_cleanup(mdev); } + +struct mlx5_accel_esp_xfrm * +mlx5_accel_esp_create_xfrm(struct mlx5_core_dev *mdev, + const struct mlx5_accel_esp_xfrm_attrs *attrs, + u32 flags) +{ + struct mlx5_accel_esp_xfrm *xfrm; + + xfrm = mlx5_fpga_esp_create_xfrm(mdev, attrs, flags); + if (IS_ERR(xfrm)) + return xfrm; + + xfrm->mdev = mdev; + return xfrm; +} +EXPORT_SYMBOL_GPL(mlx5_accel_esp_create_xfrm); + +void mlx5_accel_esp_destroy_xfrm(struct mlx5_accel_esp_xfrm *xfrm) +{ + mlx5_fpga_esp_destroy_xfrm(xfrm); +} +EXPORT_SYMBOL_GPL(mlx5_accel_esp_destroy_xfrm); + +int mlx5_accel_esp_modify_xfrm(struct mlx5_accel_esp_xfrm *xfrm, + const struct mlx5_accel_esp_xfrm_attrs *attrs) +{ + return mlx5_fpga_esp_modify_xfrm(xfrm, attrs); +} +EXPORT_SYMBOL_GPL(mlx5_accel_esp_modify_xfrm); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.h index d6e20fea9554..024dbd22a89b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/accel/ipsec.h @@ -34,89 +34,25 @@ #ifndef __MLX5_ACCEL_IPSEC_H__ #define __MLX5_ACCEL_IPSEC_H__ -#ifdef CONFIG_MLX5_ACCEL - #include <linux/mlx5/driver.h> +#include <linux/mlx5/accel.h> -enum { - MLX5_ACCEL_IPSEC_DEVICE = BIT(1), - MLX5_ACCEL_IPSEC_IPV6 = BIT(2), - MLX5_ACCEL_IPSEC_ESP = BIT(3), - MLX5_ACCEL_IPSEC_LSO = BIT(4), -}; - -#define MLX5_IPSEC_SADB_IP_AH BIT(7) -#define MLX5_IPSEC_SADB_IP_ESP BIT(6) -#define MLX5_IPSEC_SADB_SA_VALID BIT(5) -#define MLX5_IPSEC_SADB_SPI_EN BIT(4) -#define MLX5_IPSEC_SADB_DIR_SX BIT(3) -#define MLX5_IPSEC_SADB_IPV6 BIT(2) - -enum { - MLX5_IPSEC_CMD_ADD_SA = 0, - MLX5_IPSEC_CMD_DEL_SA = 1, -}; - -enum mlx5_accel_ipsec_enc_mode { - MLX5_IPSEC_SADB_MODE_NONE = 0, - MLX5_IPSEC_SADB_MODE_AES_GCM_128_AUTH_128 = 1, - MLX5_IPSEC_SADB_MODE_AES_GCM_256_AUTH_128 = 3, -}; +#ifdef CONFIG_MLX5_ACCEL #define MLX5_IPSEC_DEV(mdev) (mlx5_accel_ipsec_device_caps(mdev) & \ - MLX5_ACCEL_IPSEC_DEVICE) - -struct mlx5_accel_ipsec_sa { - __be32 cmd; - u8 key_enc[32]; - u8 key_auth[32]; - __be32 sip[4]; - __be32 dip[4]; - union { - struct { - __be32 reserved; - u8 salt_iv[8]; - __be32 salt; - } __packed gcm; - struct { - u8 salt[16]; - } __packed cbc; - }; - __be32 spi; - __be32 sw_sa_handle; - __be16 tfclen; - u8 enc_mode; - u8 sip_masklen; - u8 dip_masklen; - u8 flags; - u8 reserved[2]; -} __packed; - -/** - * mlx5_accel_ipsec_sa_cmd_exec - Execute an IPSec SADB command - * @mdev: mlx5 device - * @cmd: command to execute - * May be called from atomic context. Returns context pointer, or error - * Caller must eventually call mlx5_accel_ipsec_sa_cmd_wait from non-atomic - * context, to cleanup the context pointer - */ -void *mlx5_accel_ipsec_sa_cmd_exec(struct mlx5_core_dev *mdev, - struct mlx5_accel_ipsec_sa *cmd); - -/** - * mlx5_accel_ipsec_sa_cmd_wait - Wait for command execution completion - * @context: Context pointer returned from call to mlx5_accel_ipsec_sa_cmd_exec - * Sleeps (killable) until command execution is complete. - * Returns the command result, or -EINTR if killed - */ -int mlx5_accel_ipsec_sa_cmd_wait(void *context); - -u32 mlx5_accel_ipsec_device_caps(struct mlx5_core_dev *mdev); + MLX5_ACCEL_IPSEC_CAP_DEVICE) unsigned int mlx5_accel_ipsec_counters_count(struct mlx5_core_dev *mdev); int mlx5_accel_ipsec_counters_read(struct mlx5_core_dev *mdev, u64 *counters, unsigned int count); +void *mlx5_accel_esp_create_hw_context(struct mlx5_core_dev *mdev, + struct mlx5_accel_esp_xfrm *xfrm, + const __be32 saddr[4], + const __be32 daddr[4], + const __be32 spi, bool is_ipv6); +void mlx5_accel_esp_free_hw_context(void *context); + int mlx5_accel_ipsec_init(struct mlx5_core_dev *mdev); void mlx5_accel_ipsec_cleanup(struct mlx5_core_dev *mdev); @@ -124,6 +60,20 @@ void mlx5_accel_ipsec_cleanup(struct mlx5_core_dev *mdev); #define MLX5_IPSEC_DEV(mdev) false +static inline void * +mlx5_accel_esp_create_hw_context(struct mlx5_core_dev *mdev, + struct mlx5_accel_esp_xfrm *xfrm, + const __be32 saddr[4], + const __be32 daddr[4], + const __be32 spi, bool is_ipv6) +{ + return NULL; +} + +static inline void mlx5_accel_esp_free_hw_context(void *context) +{ +} + static inline int mlx5_accel_ipsec_init(struct mlx5_core_dev *mdev) { return 0; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c index 47239bf7bf43..323ffe8bf7e4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c @@ -71,19 +71,24 @@ static void *mlx5_dma_zalloc_coherent_node(struct mlx5_core_dev *dev, } int mlx5_buf_alloc_node(struct mlx5_core_dev *dev, int size, - struct mlx5_buf *buf, int node) + struct mlx5_frag_buf *buf, int node) { dma_addr_t t; buf->size = size; buf->npages = 1; buf->page_shift = (u8)get_order(size) + PAGE_SHIFT; - buf->direct.buf = mlx5_dma_zalloc_coherent_node(dev, size, - &t, node); - if (!buf->direct.buf) + + buf->frags = kzalloc(sizeof(*buf->frags), GFP_KERNEL); + if (!buf->frags) return -ENOMEM; - buf->direct.map = t; + buf->frags->buf = mlx5_dma_zalloc_coherent_node(dev, size, + &t, node); + if (!buf->frags->buf) + goto err_out; + + buf->frags->map = t; while (t & ((1 << buf->page_shift) - 1)) { --buf->page_shift; @@ -91,18 +96,24 @@ int mlx5_buf_alloc_node(struct mlx5_core_dev *dev, int size, } return 0; +err_out: + kfree(buf->frags); + return -ENOMEM; } -int mlx5_buf_alloc(struct mlx5_core_dev *dev, int size, struct mlx5_buf *buf) +int mlx5_buf_alloc(struct mlx5_core_dev *dev, + int size, struct mlx5_frag_buf *buf) { return mlx5_buf_alloc_node(dev, size, buf, dev->priv.numa_node); } -EXPORT_SYMBOL_GPL(mlx5_buf_alloc); +EXPORT_SYMBOL(mlx5_buf_alloc); -void mlx5_buf_free(struct mlx5_core_dev *dev, struct mlx5_buf *buf) +void mlx5_buf_free(struct mlx5_core_dev *dev, struct mlx5_frag_buf *buf) { - dma_free_coherent(&dev->pdev->dev, buf->size, buf->direct.buf, - buf->direct.map); + dma_free_coherent(&dev->pdev->dev, buf->size, buf->frags->buf, + buf->frags->map); + + kfree(buf->frags); } EXPORT_SYMBOL_GPL(mlx5_buf_free); @@ -147,6 +158,7 @@ err_free_buf: err_out: return -ENOMEM; } +EXPORT_SYMBOL_GPL(mlx5_frag_buf_alloc_node); void mlx5_frag_buf_free(struct mlx5_core_dev *dev, struct mlx5_frag_buf *buf) { @@ -162,6 +174,7 @@ void mlx5_frag_buf_free(struct mlx5_core_dev *dev, struct mlx5_frag_buf *buf) } kfree(buf->frags); } +EXPORT_SYMBOL_GPL(mlx5_frag_buf_free); static struct mlx5_db_pgdir *mlx5_alloc_db_pgdir(struct mlx5_core_dev *dev, int node) @@ -275,13 +288,13 @@ void mlx5_db_free(struct mlx5_core_dev *dev, struct mlx5_db *db) } EXPORT_SYMBOL_GPL(mlx5_db_free); -void mlx5_fill_page_array(struct mlx5_buf *buf, __be64 *pas) +void mlx5_fill_page_array(struct mlx5_frag_buf *buf, __be64 *pas) { u64 addr; int i; for (i = 0; i < buf->npages; i++) { - addr = buf->direct.map + (i << buf->page_shift); + addr = buf->frags->map + (i << buf->page_shift); pas[i] = cpu_to_be64(addr); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index e9a1fbcc4adf..21cd1703a862 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -359,6 +359,7 @@ static int mlx5_internal_err_ret_value(struct mlx5_core_dev *dev, u16 op, case MLX5_CMD_OP_MODIFY_HCA_VPORT_CONTEXT: case MLX5_CMD_OP_QUERY_HCA_VPORT_GID: case MLX5_CMD_OP_QUERY_HCA_VPORT_PKEY: + case MLX5_CMD_OP_QUERY_VNIC_ENV: case MLX5_CMD_OP_QUERY_VPORT_COUNTER: case MLX5_CMD_OP_ALLOC_Q_COUNTER: case MLX5_CMD_OP_QUERY_Q_COUNTER: @@ -501,6 +502,7 @@ const char *mlx5_command_str(int command) MLX5_COMMAND_STR_CASE(MODIFY_HCA_VPORT_CONTEXT); MLX5_COMMAND_STR_CASE(QUERY_HCA_VPORT_GID); MLX5_COMMAND_STR_CASE(QUERY_HCA_VPORT_PKEY); + MLX5_COMMAND_STR_CASE(QUERY_VNIC_ENV); MLX5_COMMAND_STR_CASE(QUERY_VPORT_COUNTER); MLX5_COMMAND_STR_CASE(ALLOC_Q_COUNTER); MLX5_COMMAND_STR_CASE(DEALLOC_Q_COUNTER); @@ -1802,7 +1804,7 @@ int mlx5_cmd_init(struct mlx5_core_dev *dev) cmd->checksum_disabled = 1; cmd->max_reg_cmds = (1 << cmd->log_sz) - 1; - cmd->bitmask = (1 << cmd->max_reg_cmds) - 1; + cmd->bitmask = (1UL << cmd->max_reg_cmds) - 1; cmd->cmdif_rev = ioread32be(&dev->iseg->cmdif_rev_fw_sub) >> 16; if (cmd->cmdif_rev > CMD_IF_REV) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cq.c b/drivers/net/ethernet/mellanox/mlx5/core/cq.c index 1016e05c7ec7..a4179122a279 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cq.c @@ -58,8 +58,7 @@ void mlx5_cq_tasklet_cb(unsigned long data) tasklet_ctx.list) { list_del_init(&mcq->tasklet_ctx.list); mcq->tasklet_ctx.comp(mcq); - if (refcount_dec_and_test(&mcq->refcount)) - complete(&mcq->free); + mlx5_cq_put(mcq); if (time_after(jiffies, end)) break; } @@ -80,69 +79,19 @@ static void mlx5_add_cq_to_tasklet(struct mlx5_core_cq *cq) * still arrive. */ if (list_empty_careful(&cq->tasklet_ctx.list)) { - refcount_inc(&cq->refcount); + mlx5_cq_hold(cq); list_add_tail(&cq->tasklet_ctx.list, &tasklet_ctx->list); } spin_unlock_irqrestore(&tasklet_ctx->lock, flags); } -void mlx5_cq_completion(struct mlx5_core_dev *dev, u32 cqn) -{ - struct mlx5_core_cq *cq; - struct mlx5_cq_table *table = &dev->priv.cq_table; - - spin_lock(&table->lock); - cq = radix_tree_lookup(&table->tree, cqn); - if (likely(cq)) - refcount_inc(&cq->refcount); - spin_unlock(&table->lock); - - if (!cq) { - mlx5_core_warn(dev, "Completion event for bogus CQ 0x%x\n", cqn); - return; - } - - ++cq->arm_sn; - - cq->comp(cq); - - if (refcount_dec_and_test(&cq->refcount)) - complete(&cq->free); -} - -void mlx5_cq_event(struct mlx5_core_dev *dev, u32 cqn, int event_type) -{ - struct mlx5_cq_table *table = &dev->priv.cq_table; - struct mlx5_core_cq *cq; - - spin_lock(&table->lock); - - cq = radix_tree_lookup(&table->tree, cqn); - if (cq) - refcount_inc(&cq->refcount); - - spin_unlock(&table->lock); - - if (!cq) { - mlx5_core_warn(dev, "Async event for bogus CQ 0x%x\n", cqn); - return; - } - - cq->event(cq, event_type); - - if (refcount_dec_and_test(&cq->refcount)) - complete(&cq->free); -} - int mlx5_core_create_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq, u32 *in, int inlen) { - struct mlx5_cq_table *table = &dev->priv.cq_table; + int eqn = MLX5_GET(cqc, MLX5_ADDR_OF(create_cq_in, in, cq_context), c_eqn); + u32 dout[MLX5_ST_SZ_DW(destroy_cq_out)]; u32 out[MLX5_ST_SZ_DW(create_cq_out)]; u32 din[MLX5_ST_SZ_DW(destroy_cq_in)]; - u32 dout[MLX5_ST_SZ_DW(destroy_cq_out)]; - int eqn = MLX5_GET(cqc, MLX5_ADDR_OF(create_cq_in, in, cq_context), - c_eqn); struct mlx5_eq *eq; int err; @@ -159,6 +108,7 @@ int mlx5_core_create_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq, cq->cqn = MLX5_GET(create_cq_out, out, cqn); cq->cons_index = 0; cq->arm_sn = 0; + cq->eq = eq; refcount_set(&cq->refcount, 1); init_completion(&cq->free); if (!cq->comp) @@ -167,12 +117,16 @@ int mlx5_core_create_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq, cq->tasklet_ctx.priv = &eq->tasklet_ctx; INIT_LIST_HEAD(&cq->tasklet_ctx.list); - spin_lock_irq(&table->lock); - err = radix_tree_insert(&table->tree, cq->cqn, cq); - spin_unlock_irq(&table->lock); + /* Add to comp EQ CQ tree to recv comp events */ + err = mlx5_eq_add_cq(eq, cq); if (err) goto err_cmd; + /* Add to async EQ CQ tree to recv async events */ + err = mlx5_eq_add_cq(&dev->priv.eq_table.async_eq, cq); + if (err) + goto err_cq_add; + cq->pid = current->pid; err = mlx5_debug_cq_add(dev, cq); if (err) @@ -183,6 +137,8 @@ int mlx5_core_create_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq, return 0; +err_cq_add: + mlx5_eq_del_cq(eq, cq); err_cmd: memset(din, 0, sizeof(din)); memset(dout, 0, sizeof(dout)); @@ -195,23 +151,17 @@ EXPORT_SYMBOL(mlx5_core_create_cq); int mlx5_core_destroy_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq) { - struct mlx5_cq_table *table = &dev->priv.cq_table; u32 out[MLX5_ST_SZ_DW(destroy_cq_out)] = {0}; u32 in[MLX5_ST_SZ_DW(destroy_cq_in)] = {0}; - struct mlx5_core_cq *tmp; int err; - spin_lock_irq(&table->lock); - tmp = radix_tree_delete(&table->tree, cq->cqn); - spin_unlock_irq(&table->lock); - if (!tmp) { - mlx5_core_warn(dev, "cq 0x%x not found in tree\n", cq->cqn); - return -EINVAL; - } - if (tmp != cq) { - mlx5_core_warn(dev, "corruption on srqn 0x%x\n", cq->cqn); - return -EINVAL; - } + err = mlx5_eq_del_cq(&dev->priv.eq_table.async_eq, cq); + if (err) + return err; + + err = mlx5_eq_del_cq(cq->eq, cq); + if (err) + return err; MLX5_SET(destroy_cq_in, in, opcode, MLX5_CMD_OP_DESTROY_CQ); MLX5_SET(destroy_cq_in, in, cqn, cq->cqn); @@ -222,8 +172,7 @@ int mlx5_core_destroy_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq) synchronize_irq(cq->irqn); mlx5_debug_cq_remove(dev, cq); - if (refcount_dec_and_test(&cq->refcount)) - complete(&cq->free); + mlx5_cq_put(cq); wait_for_completion(&cq->free); return 0; @@ -270,21 +219,3 @@ int mlx5_core_modify_cq_moderation(struct mlx5_core_dev *dev, return mlx5_core_modify_cq(dev, cq, in, sizeof(in)); } EXPORT_SYMBOL(mlx5_core_modify_cq_moderation); - -int mlx5_init_cq_table(struct mlx5_core_dev *dev) -{ - struct mlx5_cq_table *table = &dev->priv.cq_table; - int err; - - memset(table, 0, sizeof(*table)); - spin_lock_init(&table->lock); - INIT_RADIX_TREE(&table->tree, GFP_ATOMIC); - err = mlx5_cq_debugfs_init(dev); - - return err; -} - -void mlx5_cleanup_cq_table(struct mlx5_core_dev *dev) -{ - mlx5_cq_debugfs_cleanup(dev); -} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/dev.c index 17b723218b0c..b994b80d5714 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/dev.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/dev.c @@ -337,6 +337,14 @@ void mlx5_unregister_interface(struct mlx5_interface *intf) } EXPORT_SYMBOL(mlx5_unregister_interface); +void mlx5_reload_interface(struct mlx5_core_dev *mdev, int protocol) +{ + mutex_lock(&mlx5_intf_mutex); + mlx5_remove_dev_by_protocol(mdev, protocol); + mlx5_add_dev_by_protocol(mdev, protocol); + mutex_unlock(&mlx5_intf_mutex); +} + void *mlx5_get_protocol_dev(struct mlx5_core_dev *mdev, int protocol) { struct mlx5_priv *priv = &mdev->priv; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.c index fd509160c8f6..d93ff567b40d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.c @@ -246,6 +246,9 @@ const char *parse_fs_dst(struct trace_seq *p, case MLX5_FLOW_DESTINATION_TYPE_COUNTER: trace_seq_printf(p, "counter_id=%u\n", counter_id); break; + case MLX5_FLOW_DESTINATION_TYPE_PORT: + trace_seq_printf(p, "port\n"); + break; } trace_seq_putc(p, 0); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.h b/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.h index 80eef4163f52..09f178a3fcab 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fs_tracepoint.h @@ -136,6 +136,8 @@ TRACE_EVENT(mlx5_fs_del_fg, {MLX5_FLOW_CONTEXT_ACTION_ENCAP, "ENCAP"},\ {MLX5_FLOW_CONTEXT_ACTION_DECAP, "DECAP"},\ {MLX5_FLOW_CONTEXT_ACTION_MOD_HDR, "MOD_HDR"},\ + {MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH, "VLAN_PUSH"},\ + {MLX5_FLOW_CONTEXT_ACTION_VLAN_POP, "VLAN_POP"},\ {MLX5_FLOW_CONTEXT_ACTION_FWD_NEXT_PRIO, "NEXT_PRIO"} TRACE_EVENT(mlx5_fs_set_fte, @@ -163,9 +165,9 @@ TRACE_EVENT(mlx5_fs_set_fte, fs_get_obj(__entry->fg, fte->node.parent); __entry->group_index = __entry->fg->id; __entry->index = fte->index; - __entry->action = fte->action; + __entry->action = fte->action.action; __entry->mask_enable = __entry->fg->mask.match_criteria_enable; - __entry->flow_tag = fte->flow_tag; + __entry->flow_tag = fte->action.flow_tag; memcpy(__entry->mask_outer, MLX5_ADDR_OF(fte_match_param, &__entry->fg->mask.match_criteria, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h index 4c9360b25532..30cad07be2b5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h @@ -57,24 +57,12 @@ #define MLX5E_ETH_HARD_MTU (ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN) -#define MLX5E_HW2SW_MTU(priv, hwmtu) ((hwmtu) - ((priv)->hard_mtu)) -#define MLX5E_SW2HW_MTU(priv, swmtu) ((swmtu) + ((priv)->hard_mtu)) +#define MLX5E_HW2SW_MTU(params, hwmtu) ((hwmtu) - ((params)->hard_mtu)) +#define MLX5E_SW2HW_MTU(params, swmtu) ((swmtu) + ((params)->hard_mtu)) #define MLX5E_MAX_DSCP 64 #define MLX5E_MAX_NUM_TC 8 -#define MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE 0x6 -#define MLX5E_PARAMS_DEFAULT_LOG_SQ_SIZE 0xa -#define MLX5E_PARAMS_MAXIMUM_LOG_SQ_SIZE 0xd - -#define MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE 0x1 -#define MLX5E_PARAMS_DEFAULT_LOG_RQ_SIZE 0xa -#define MLX5E_PARAMS_MAXIMUM_LOG_RQ_SIZE 0xd - -#define MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE_MPW 0x2 -#define MLX5E_PARAMS_DEFAULT_LOG_RQ_SIZE_MPW 0x3 -#define MLX5E_PARAMS_MAXIMUM_LOG_RQ_SIZE_MPW 0x6 - #define MLX5_RX_HEADROOM NET_SKB_PAD #define MLX5_SKB_FRAG_SZ(len) (SKB_DATA_ALIGN(len) + \ SKB_DATA_ALIGN(sizeof(struct skb_shared_info))) @@ -93,15 +81,31 @@ #define MLX5_MPWRQ_WQE_PAGE_ORDER (MLX5_MPWRQ_LOG_WQE_SZ - PAGE_SHIFT > 0 ? \ MLX5_MPWRQ_LOG_WQE_SZ - PAGE_SHIFT : 0) #define MLX5_MPWRQ_PAGES_PER_WQE BIT(MLX5_MPWRQ_WQE_PAGE_ORDER) -#define MLX5_MPWRQ_STRIDES_PER_PAGE (MLX5_MPWRQ_NUM_STRIDES >> \ - MLX5_MPWRQ_WQE_PAGE_ORDER) #define MLX5_MTT_OCTW(npages) (ALIGN(npages, 8) / 2) -#define MLX5E_REQUIRED_MTTS(wqes) \ - (wqes * ALIGN(MLX5_MPWRQ_PAGES_PER_WQE, 8)) -#define MLX5E_VALID_NUM_MTTS(num_mtts) (MLX5_MTT_OCTW(num_mtts) - 1 <= U16_MAX) +#define MLX5E_REQUIRED_WQE_MTTS (ALIGN(MLX5_MPWRQ_PAGES_PER_WQE, 8)) +#define MLX5E_LOG_ALIGNED_MPWQE_PPW (ilog2(MLX5E_REQUIRED_WQE_MTTS)) +#define MLX5E_REQUIRED_MTTS(wqes) (wqes * MLX5E_REQUIRED_WQE_MTTS) +#define MLX5E_MAX_RQ_NUM_MTTS \ + ((1 << 16) * 2) /* So that MLX5_MTT_OCTW(num_mtts) fits into u16 */ +#define MLX5E_ORDER2_MAX_PACKET_MTU (order_base_2(10 * 1024)) +#define MLX5E_PARAMS_MAXIMUM_LOG_RQ_SIZE_MPW \ + (ilog2(MLX5E_MAX_RQ_NUM_MTTS / MLX5E_REQUIRED_WQE_MTTS)) +#define MLX5E_LOG_MAX_RQ_NUM_PACKETS_MPW \ + (MLX5E_PARAMS_MAXIMUM_LOG_RQ_SIZE_MPW + \ + (MLX5_MPWRQ_LOG_WQE_SZ - MLX5E_ORDER2_MAX_PACKET_MTU)) + +#define MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE 0x6 +#define MLX5E_PARAMS_DEFAULT_LOG_SQ_SIZE 0xa +#define MLX5E_PARAMS_MAXIMUM_LOG_SQ_SIZE 0xd + +#define MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE 0x1 +#define MLX5E_PARAMS_DEFAULT_LOG_RQ_SIZE 0xa +#define MLX5E_PARAMS_MAXIMUM_LOG_RQ_SIZE min_t(u8, 0xd, \ + MLX5E_LOG_MAX_RQ_NUM_PACKETS_MPW) + +#define MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE_MPW 0x2 -#define MLX5_UMR_ALIGN (2048) #define MLX5_MPWRQ_SMALL_PACKET_THRESHOLD (256) #define MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ (64 * 1024) @@ -124,9 +128,15 @@ #define MLX5E_MAX_NUM_SQS (MLX5E_MAX_NUM_CHANNELS * MLX5E_MAX_NUM_TC) #define MLX5E_TX_CQ_POLL_BUDGET 128 #define MLX5E_UPDATE_STATS_INTERVAL 200 /* msecs */ +#define MLX5E_SQ_RECOVER_MIN_INTERVAL 500 /* msecs */ -#define MLX5E_ICOSQ_MAX_WQEBBS \ - (DIV_ROUND_UP(sizeof(struct mlx5e_umr_wqe), MLX5_SEND_WQE_BB)) +#define MLX5E_UMR_WQE_INLINE_SZ \ + (sizeof(struct mlx5e_umr_wqe) + \ + ALIGN(MLX5_MPWRQ_PAGES_PER_WQE * sizeof(struct mlx5_mtt), \ + MLX5_UMR_MTT_ALIGNMENT)) +#define MLX5E_UMR_WQEBBS \ + (DIV_ROUND_UP(MLX5E_UMR_WQE_INLINE_SZ, MLX5_SEND_WQE_BB)) +#define MLX5E_ICOSQ_MAX_WQEBBS MLX5E_UMR_WQEBBS #define MLX5E_XDP_MIN_INLINE (ETH_HLEN + VLAN_HLEN) #define MLX5E_XDP_TX_DS_COUNT \ @@ -156,26 +166,6 @@ static inline u16 mlx5_min_rx_wqes(int wq_type, u32 wq_size) } } -static inline int mlx5_min_log_rq_size(int wq_type) -{ - switch (wq_type) { - case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: - return MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE_MPW; - default: - return MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE; - } -} - -static inline int mlx5_max_log_rq_size(int wq_type) -{ - switch (wq_type) { - case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: - return MLX5E_PARAMS_MAXIMUM_LOG_RQ_SIZE_MPW; - default: - return MLX5E_PARAMS_MAXIMUM_LOG_RQ_SIZE; - } -} - static inline int mlx5e_get_max_num_channels(struct mlx5_core_dev *mdev) { return is_kdump_kernel() ? @@ -198,7 +188,7 @@ struct mlx5e_umr_wqe { struct mlx5_wqe_ctrl_seg ctrl; struct mlx5_wqe_umr_ctrl_seg uctrl; struct mlx5_mkey_seg mkc; - struct mlx5_wqe_data_seg data; + struct mlx5_mtt inline_mtts[0]; }; extern const char mlx5e_self_tests[][ETH_GSTRING_LEN]; @@ -207,12 +197,14 @@ static const char mlx5e_priv_flags[][ETH_GSTRING_LEN] = { "rx_cqe_moder", "tx_cqe_moder", "rx_cqe_compress", + "rx_striding_rq", }; enum mlx5e_priv_flag { MLX5E_PFLAG_RX_CQE_BASED_MODER = (1 << 0), MLX5E_PFLAG_TX_CQE_BASED_MODER = (1 << 1), MLX5E_PFLAG_RX_CQE_COMPRESS = (1 << 2), + MLX5E_PFLAG_RX_STRIDING_RQ = (1 << 3), }; #define MLX5E_SET_PFLAG(params, pflag, enable) \ @@ -232,10 +224,7 @@ enum mlx5e_priv_flag { struct mlx5e_params { u8 log_sq_size; u8 rq_wq_type; - u16 rq_headroom; - u8 mpwqe_log_stride_sz; - u8 mpwqe_log_num_strides; - u8 log_rq_size; + u8 log_rq_mtu_frames; u16 num_channels; u8 num_tc; bool rx_cqe_compress_def; @@ -243,7 +232,6 @@ struct mlx5e_params { struct net_dim_cq_moder tx_cq_moderation; bool lro_en; u32 lro_wqe_sz; - u16 tx_max_inline; u8 tx_min_inline_mode; u8 rss_hfunc; u8 toeplitz_hash_key[40]; @@ -254,6 +242,8 @@ struct mlx5e_params { u32 lro_timeout; u32 pflags; struct bpf_prog *xdp_prog; + unsigned int sw_mtu; + int hard_mtu; }; #ifdef CONFIG_MLX5_CORE_EN_DCB @@ -336,6 +326,7 @@ struct mlx5e_sq_dma { enum { MLX5E_SQ_STATE_ENABLED, + MLX5E_SQ_STATE_RECOVERING, MLX5E_SQ_STATE_IPSEC, }; @@ -369,7 +360,6 @@ struct mlx5e_txqsq { void __iomem *uar_map; struct netdev_queue *txq; u32 sqn; - u16 max_inline; u8 min_inline_mode; u16 edge; struct device *pdev; @@ -383,6 +373,10 @@ struct mlx5e_txqsq { struct mlx5e_channel *channel; int txq_ix; u32 rate_limit; + struct mlx5e_txqsq_recover { + struct work_struct recover_work; + u64 last_recover; + } recover; } ____cacheline_aligned_in_smp; struct mlx5e_xdpsq { @@ -432,7 +426,6 @@ struct mlx5e_icosq { void __iomem *uar_map; u32 sqn; u16 edge; - __be32 mkey_be; unsigned long state; /* control path */ @@ -457,16 +450,13 @@ struct mlx5e_wqe_frag_info { }; struct mlx5e_umr_dma_info { - __be64 *mtt; - dma_addr_t mtt_addr; struct mlx5e_dma_info dma_info[MLX5_MPWRQ_PAGES_PER_WQE]; - struct mlx5e_umr_wqe wqe; }; struct mlx5e_mpw_info { struct mlx5e_umr_dma_info umr; u16 consumed_strides; - u16 skbs_frags[MLX5_MPWRQ_PAGES_PER_WQE]; + DECLARE_BITMAP(xdp_xmit_bitmap, MLX5_MPWRQ_PAGES_PER_WQE); }; /* a single cache unit is capable to serve one napi call (for non-striding rq) @@ -483,9 +473,16 @@ struct mlx5e_page_cache { struct mlx5e_rq; typedef void (*mlx5e_fp_handle_rx_cqe)(struct mlx5e_rq*, struct mlx5_cqe64*); +typedef struct sk_buff * +(*mlx5e_fp_skb_from_cqe_mpwrq)(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, + u16 cqe_bcnt, u32 head_offset, u32 page_idx); typedef bool (*mlx5e_fp_post_rx_wqes)(struct mlx5e_rq *rq); typedef void (*mlx5e_fp_dealloc_wqe)(struct mlx5e_rq*, u16); +enum mlx5e_rq_flag { + MLX5E_RQ_FLAG_XDP_XMIT = BIT(0), +}; + struct mlx5e_rq { /* data path */ struct mlx5_wq_ll wq; @@ -496,12 +493,12 @@ struct mlx5e_rq { u32 frag_sz; /* max possible skb frag_sz */ union { bool page_reuse; - bool xdp_xmit; }; } wqe; struct { + struct mlx5e_umr_wqe umr_wqe; struct mlx5e_mpw_info *info; - void *mtt_no_align; + mlx5e_fp_skb_from_cqe_mpwrq skb_from_cqe_mpwrq; u16 num_strides; u8 log_stride_sz; bool umr_in_progress; @@ -533,7 +530,9 @@ struct mlx5e_rq { /* XDP */ struct bpf_prog *xdp_prog; + unsigned int hw_mtu; struct mlx5e_xdpsq xdpsq; + DECLARE_BITMAP(flags, 8); /* control */ struct mlx5_wq_ctrl wq_ctrl; @@ -766,7 +765,6 @@ struct mlx5e_priv { struct mlx5e_tir inner_indir_tir[MLX5E_NUM_INDIR_TIRS]; struct mlx5e_tir direct_tir[MLX5E_MAX_NUM_CHANNELS]; u32 tx_rates[MLX5E_MAX_NUM_SQS]; - int hard_mtu; struct mlx5e_flow_steering fs; struct mlx5e_vxlan_db vxlan; @@ -781,7 +779,8 @@ struct mlx5e_priv { struct net_device *netdev; struct mlx5e_stats stats; struct hwtstamp_config tstamp; - u16 q_counter; + u16 q_counter; + u16 drop_rq_q_counter; #ifdef CONFIG_MLX5_CORE_EN_DCB struct mlx5e_dcbx dcbx; #endif @@ -831,6 +830,10 @@ bool mlx5e_poll_xdpsq_cq(struct mlx5e_cq *cq); void mlx5e_free_txqsq_descs(struct mlx5e_txqsq *sq); void mlx5e_free_xdpsq_descs(struct mlx5e_xdpsq *sq); +bool mlx5e_check_fragmented_striding_rq_cap(struct mlx5_core_dev *mdev); +bool mlx5e_striding_rq_possible(struct mlx5_core_dev *mdev, + struct mlx5e_params *params); + void mlx5e_page_release(struct mlx5e_rq *rq, struct mlx5e_dma_info *dma_info, bool recycle); void mlx5e_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe); @@ -840,6 +843,12 @@ bool mlx5e_post_rx_mpwqes(struct mlx5e_rq *rq); void mlx5e_dealloc_rx_wqe(struct mlx5e_rq *rq, u16 ix); void mlx5e_dealloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix); void mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi); +struct sk_buff * +mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, + u16 cqe_bcnt, u32 head_offset, u32 page_idx); +struct sk_buff * +mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, + u16 cqe_bcnt, u32 head_offset, u32 page_idx); void mlx5e_update_stats(struct mlx5e_priv *priv); @@ -916,9 +925,9 @@ void mlx5e_set_tx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode); void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode); +void mlx5e_set_rq_type(struct mlx5_core_dev *mdev, struct mlx5e_params *params); void mlx5e_init_rq_type_params(struct mlx5_core_dev *mdev, - struct mlx5e_params *params, - u8 rq_type); + struct mlx5e_params *params); static inline bool mlx5e_tunnel_inner_ft_supported(struct mlx5_core_dev *mdev) { @@ -970,11 +979,6 @@ static inline void mlx5e_cq_arm(struct mlx5e_cq *cq) mlx5_cq_arm(mcq, MLX5_CQ_DB_REQ_NOT, mcq->uar->map, cq->wq.cc); } -static inline u32 mlx5e_get_wqe_mtt_offset(struct mlx5e_rq *rq, u16 wqe_ix) -{ - return wqe_ix * ALIGN(MLX5_MPWRQ_PAGES_PER_WQE, 8); -} - extern const struct ethtool_ops mlx5e_ethtool_ops; #ifdef CONFIG_MLX5_CORE_EN_DCB extern const struct dcbnl_rtnl_ops mlx5e_dcbnl_ops; @@ -1010,7 +1014,6 @@ int mlx5e_rx_flow_steer(struct net_device *dev, const struct sk_buff *skb, u16 rxq_index, u32 flow_id); #endif -u16 mlx5e_get_max_inline_cap(struct mlx5_core_dev *mdev); int mlx5e_create_tir(struct mlx5_core_dev *mdev, struct mlx5e_tir *tir, u32 *in, int inlen); void mlx5e_destroy_tir(struct mlx5_core_dev *mdev, @@ -1061,7 +1064,6 @@ void mlx5e_cleanup_nic_tx(struct mlx5e_priv *priv); int mlx5e_close(struct net_device *netdev); int mlx5e_open(struct net_device *netdev); void mlx5e_update_stats_work(struct work_struct *work); -u32 mlx5e_choose_lro_timeout(struct mlx5_core_dev *mdev, u32 wanted_timeout); int mlx5e_bits_invert(unsigned long a, int size); @@ -1102,7 +1104,7 @@ void mlx5e_detach_netdev(struct mlx5e_priv *priv); void mlx5e_destroy_netdev(struct mlx5e_priv *priv); void mlx5e_build_nic_params(struct mlx5_core_dev *mdev, struct mlx5e_params *params, - u16 max_channels); + u16 max_channels, u16 mtu); u8 mlx5e_params_calculate_tx_min_inline(struct mlx5_core_dev *mdev); void mlx5e_rx_dim_work(struct work_struct *work); #endif /* __MLX5_EN_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c index bac5103efad3..cf58c9637904 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.c @@ -38,17 +38,24 @@ #include <linux/module.h> #include "en.h" -#include "accel/ipsec.h" #include "en_accel/ipsec.h" #include "en_accel/ipsec_rxtx.h" -struct mlx5e_ipsec_sa_entry { - struct hlist_node hlist; /* Item in SADB_RX hashtable */ - unsigned int handle; /* Handle in SADB_RX */ - struct xfrm_state *x; - struct mlx5e_ipsec *ipsec; - void *context; -}; + +static struct mlx5e_ipsec_sa_entry *to_ipsec_sa_entry(struct xfrm_state *x) +{ + struct mlx5e_ipsec_sa_entry *sa; + + if (!x) + return NULL; + + sa = (struct mlx5e_ipsec_sa_entry *)x->xso.offload_handle; + if (!sa) + return NULL; + + WARN_ON(sa->x != x); + return sa; +} struct xfrm_state *mlx5e_ipsec_sadb_rx_lookup(struct mlx5e_ipsec *ipsec, unsigned int handle) @@ -74,18 +81,16 @@ static int mlx5e_ipsec_sadb_rx_add(struct mlx5e_ipsec_sa_entry *sa_entry) unsigned long flags; int ret; - spin_lock_irqsave(&ipsec->sadb_rx_lock, flags); ret = ida_simple_get(&ipsec->halloc, 1, 0, GFP_KERNEL); if (ret < 0) - goto out; + return ret; + spin_lock_irqsave(&ipsec->sadb_rx_lock, flags); sa_entry->handle = ret; hash_add_rcu(ipsec->sadb_rx, &sa_entry->hlist, sa_entry->handle); - ret = 0; - -out: spin_unlock_irqrestore(&ipsec->sadb_rx_lock, flags); - return ret; + + return 0; } static void mlx5e_ipsec_sadb_rx_del(struct mlx5e_ipsec_sa_entry *sa_entry) @@ -101,87 +106,99 @@ static void mlx5e_ipsec_sadb_rx_del(struct mlx5e_ipsec_sa_entry *sa_entry) static void mlx5e_ipsec_sadb_rx_free(struct mlx5e_ipsec_sa_entry *sa_entry) { struct mlx5e_ipsec *ipsec = sa_entry->ipsec; - unsigned long flags; - /* Wait for the hash_del_rcu call in sadb_rx_del to affect data path */ - synchronize_rcu(); - spin_lock_irqsave(&ipsec->sadb_rx_lock, flags); + /* xfrm already doing sync rcu between del and free callbacks */ + ida_simple_remove(&ipsec->halloc, sa_entry->handle); - spin_unlock_irqrestore(&ipsec->sadb_rx_lock, flags); } -static enum mlx5_accel_ipsec_enc_mode mlx5e_ipsec_enc_mode(struct xfrm_state *x) +static bool mlx5e_ipsec_update_esn_state(struct mlx5e_ipsec_sa_entry *sa_entry) { - unsigned int key_len = (x->aead->alg_key_len + 7) / 8 - 4; - - switch (key_len) { - case 16: - return MLX5_IPSEC_SADB_MODE_AES_GCM_128_AUTH_128; - case 32: - return MLX5_IPSEC_SADB_MODE_AES_GCM_256_AUTH_128; - default: - netdev_warn(x->xso.dev, "Bad key len: %d for alg %s\n", - key_len, x->aead->alg_name); - return -1; + struct xfrm_replay_state_esn *replay_esn; + u32 seq_bottom; + u8 overlap; + u32 *esn; + + if (!(sa_entry->x->props.flags & XFRM_STATE_ESN)) { + sa_entry->esn_state.trigger = 0; + return false; + } + + replay_esn = sa_entry->x->replay_esn; + seq_bottom = replay_esn->seq - replay_esn->replay_window + 1; + overlap = sa_entry->esn_state.overlap; + + sa_entry->esn_state.esn = xfrm_replay_seqhi(sa_entry->x, + htonl(seq_bottom)); + esn = &sa_entry->esn_state.esn; + + sa_entry->esn_state.trigger = 1; + if (unlikely(overlap && seq_bottom < MLX5E_IPSEC_ESN_SCOPE_MID)) { + ++(*esn); + sa_entry->esn_state.overlap = 0; + return true; + } else if (unlikely(!overlap && + (seq_bottom >= MLX5E_IPSEC_ESN_SCOPE_MID))) { + sa_entry->esn_state.overlap = 1; + return true; } + + return false; } -static void mlx5e_ipsec_build_hw_sa(u32 op, struct mlx5e_ipsec_sa_entry *sa_entry, - struct mlx5_accel_ipsec_sa *hw_sa) +static void +mlx5e_ipsec_build_accel_xfrm_attrs(struct mlx5e_ipsec_sa_entry *sa_entry, + struct mlx5_accel_esp_xfrm_attrs *attrs) { struct xfrm_state *x = sa_entry->x; + struct aes_gcm_keymat *aes_gcm = &attrs->keymat.aes_gcm; struct aead_geniv_ctx *geniv_ctx; - unsigned int crypto_data_len; struct crypto_aead *aead; - unsigned int key_len; + unsigned int crypto_data_len, key_len; int ivsize; - memset(hw_sa, 0, sizeof(*hw_sa)); - - if (op == MLX5_IPSEC_CMD_ADD_SA) { - crypto_data_len = (x->aead->alg_key_len + 7) / 8; - key_len = crypto_data_len - 4; /* 4 bytes salt at end */ - aead = x->data; - geniv_ctx = crypto_aead_ctx(aead); - ivsize = crypto_aead_ivsize(aead); - - memcpy(&hw_sa->key_enc, x->aead->alg_key, key_len); - /* Duplicate 128 bit key twice according to HW layout */ - if (key_len == 16) - memcpy(&hw_sa->key_enc[16], x->aead->alg_key, key_len); - memcpy(&hw_sa->gcm.salt_iv, geniv_ctx->salt, ivsize); - hw_sa->gcm.salt = *((__be32 *)(x->aead->alg_key + key_len)); - } + memset(attrs, 0, sizeof(*attrs)); - hw_sa->cmd = htonl(op); - hw_sa->flags |= MLX5_IPSEC_SADB_SA_VALID | MLX5_IPSEC_SADB_SPI_EN; - if (x->props.family == AF_INET) { - hw_sa->sip[3] = x->props.saddr.a4; - hw_sa->dip[3] = x->id.daddr.a4; - hw_sa->sip_masklen = 32; - hw_sa->dip_masklen = 32; - } else { - memcpy(hw_sa->sip, x->props.saddr.a6, sizeof(hw_sa->sip)); - memcpy(hw_sa->dip, x->id.daddr.a6, sizeof(hw_sa->dip)); - hw_sa->sip_masklen = 128; - hw_sa->dip_masklen = 128; - hw_sa->flags |= MLX5_IPSEC_SADB_IPV6; - } - hw_sa->spi = x->id.spi; - hw_sa->sw_sa_handle = htonl(sa_entry->handle); - switch (x->id.proto) { - case IPPROTO_ESP: - hw_sa->flags |= MLX5_IPSEC_SADB_IP_ESP; - break; - case IPPROTO_AH: - hw_sa->flags |= MLX5_IPSEC_SADB_IP_AH; - break; - default: - break; + /* key */ + crypto_data_len = (x->aead->alg_key_len + 7) / 8; + key_len = crypto_data_len - 4; /* 4 bytes salt at end */ + + memcpy(aes_gcm->aes_key, x->aead->alg_key, key_len); + aes_gcm->key_len = key_len * 8; + + /* salt and seq_iv */ + aead = x->data; + geniv_ctx = crypto_aead_ctx(aead); + ivsize = crypto_aead_ivsize(aead); + memcpy(&aes_gcm->seq_iv, &geniv_ctx->salt, ivsize); + memcpy(&aes_gcm->salt, x->aead->alg_key + key_len, + sizeof(aes_gcm->salt)); + + /* iv len */ + aes_gcm->icv_len = x->aead->alg_icv_len; + + /* esn */ + if (sa_entry->esn_state.trigger) { + attrs->flags |= MLX5_ACCEL_ESP_FLAGS_ESN_TRIGGERED; + attrs->esn = sa_entry->esn_state.esn; + if (sa_entry->esn_state.overlap) + attrs->flags |= MLX5_ACCEL_ESP_FLAGS_ESN_STATE_OVERLAP; } - hw_sa->enc_mode = mlx5e_ipsec_enc_mode(x); - if (!(x->xso.flags & XFRM_OFFLOAD_INBOUND)) - hw_sa->flags |= MLX5_IPSEC_SADB_DIR_SX; + + /* rx handle */ + attrs->sa_handle = sa_entry->handle; + + /* algo type */ + attrs->keymat_type = MLX5_ACCEL_ESP_KEYMAT_AES_GCM; + + /* action */ + attrs->action = (!(x->xso.flags & XFRM_OFFLOAD_INBOUND)) ? + MLX5_ACCEL_ESP_ACTION_ENCRYPT : + MLX5_ACCEL_ESP_ACTION_DECRYPT; + /* flags */ + attrs->flags |= (x->props.mode == XFRM_MODE_TRANSPORT) ? + MLX5_ACCEL_ESP_FLAGS_TRANSPORT : + MLX5_ACCEL_ESP_FLAGS_TUNNEL; } static inline int mlx5e_xfrm_validate_state(struct xfrm_state *x) @@ -203,7 +220,9 @@ static inline int mlx5e_xfrm_validate_state(struct xfrm_state *x) netdev_info(netdev, "Cannot offload compressed xfrm states\n"); return -EINVAL; } - if (x->props.flags & XFRM_STATE_ESN) { + if (x->props.flags & XFRM_STATE_ESN && + !(mlx5_accel_ipsec_device_caps(priv->mdev) & + MLX5_ACCEL_IPSEC_CAP_ESN)) { netdev_info(netdev, "Cannot offload ESN xfrm states\n"); return -EINVAL; } @@ -251,7 +270,8 @@ static inline int mlx5e_xfrm_validate_state(struct xfrm_state *x) return -EINVAL; } if (x->props.family == AF_INET6 && - !(mlx5_accel_ipsec_device_caps(priv->mdev) & MLX5_ACCEL_IPSEC_IPV6)) { + !(mlx5_accel_ipsec_device_caps(priv->mdev) & + MLX5_ACCEL_IPSEC_CAP_IPV6)) { netdev_info(netdev, "IPv6 xfrm state offload is not supported by this device\n"); return -EINVAL; } @@ -262,9 +282,10 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x) { struct mlx5e_ipsec_sa_entry *sa_entry = NULL; struct net_device *netdev = x->xso.dev; - struct mlx5_accel_ipsec_sa hw_sa; + struct mlx5_accel_esp_xfrm_attrs attrs; struct mlx5e_priv *priv; - void *context; + __be32 saddr[4] = {0}, daddr[4] = {0}, spi; + bool is_ipv6 = false; int err; priv = netdev_priv(netdev); @@ -291,22 +312,49 @@ static int mlx5e_xfrm_add_state(struct xfrm_state *x) netdev_info(netdev, "Failed adding to SADB_RX: %d\n", err); goto err_entry; } + } else { + sa_entry->set_iv_op = (x->props.flags & XFRM_STATE_ESN) ? + mlx5e_ipsec_set_iv_esn : mlx5e_ipsec_set_iv; } - mlx5e_ipsec_build_hw_sa(MLX5_IPSEC_CMD_ADD_SA, sa_entry, &hw_sa); - context = mlx5_accel_ipsec_sa_cmd_exec(sa_entry->ipsec->en_priv->mdev, &hw_sa); - if (IS_ERR(context)) { - err = PTR_ERR(context); + /* check esn */ + mlx5e_ipsec_update_esn_state(sa_entry); + + /* create xfrm */ + mlx5e_ipsec_build_accel_xfrm_attrs(sa_entry, &attrs); + sa_entry->xfrm = + mlx5_accel_esp_create_xfrm(priv->mdev, &attrs, + MLX5_ACCEL_XFRM_FLAG_REQUIRE_METADATA); + if (IS_ERR(sa_entry->xfrm)) { + err = PTR_ERR(sa_entry->xfrm); goto err_sadb_rx; } - err = mlx5_accel_ipsec_sa_cmd_wait(context); - if (err) - goto err_sadb_rx; + /* create hw context */ + if (x->props.family == AF_INET) { + saddr[3] = x->props.saddr.a4; + daddr[3] = x->id.daddr.a4; + } else { + memcpy(saddr, x->props.saddr.a6, sizeof(saddr)); + memcpy(daddr, x->id.daddr.a6, sizeof(daddr)); + is_ipv6 = true; + } + spi = x->id.spi; + sa_entry->hw_context = + mlx5_accel_esp_create_hw_context(priv->mdev, + sa_entry->xfrm, + saddr, daddr, spi, + is_ipv6); + if (IS_ERR(sa_entry->hw_context)) { + err = PTR_ERR(sa_entry->hw_context); + goto err_xfrm; + } x->xso.offload_handle = (unsigned long)sa_entry; goto out; +err_xfrm: + mlx5_accel_esp_destroy_xfrm(sa_entry->xfrm); err_sadb_rx: if (x->xso.flags & XFRM_OFFLOAD_INBOUND) { mlx5e_ipsec_sadb_rx_del(sa_entry); @@ -320,43 +368,26 @@ out: static void mlx5e_xfrm_del_state(struct xfrm_state *x) { - struct mlx5e_ipsec_sa_entry *sa_entry; - struct mlx5_accel_ipsec_sa hw_sa; - void *context; + struct mlx5e_ipsec_sa_entry *sa_entry = to_ipsec_sa_entry(x); - if (!x->xso.offload_handle) + if (!sa_entry) return; - sa_entry = (struct mlx5e_ipsec_sa_entry *)x->xso.offload_handle; - WARN_ON(sa_entry->x != x); - if (x->xso.flags & XFRM_OFFLOAD_INBOUND) mlx5e_ipsec_sadb_rx_del(sa_entry); - - mlx5e_ipsec_build_hw_sa(MLX5_IPSEC_CMD_DEL_SA, sa_entry, &hw_sa); - context = mlx5_accel_ipsec_sa_cmd_exec(sa_entry->ipsec->en_priv->mdev, &hw_sa); - if (IS_ERR(context)) - return; - - sa_entry->context = context; } static void mlx5e_xfrm_free_state(struct xfrm_state *x) { - struct mlx5e_ipsec_sa_entry *sa_entry; - int res; + struct mlx5e_ipsec_sa_entry *sa_entry = to_ipsec_sa_entry(x); - if (!x->xso.offload_handle) + if (!sa_entry) return; - sa_entry = (struct mlx5e_ipsec_sa_entry *)x->xso.offload_handle; - WARN_ON(sa_entry->x != x); - - res = mlx5_accel_ipsec_sa_cmd_wait(sa_entry->context); - sa_entry->context = NULL; - if (res) { - /* Leftover object will leak */ - return; + if (sa_entry->hw_context) { + flush_workqueue(sa_entry->ipsec->wq); + mlx5_accel_esp_free_hw_context(sa_entry->hw_context); + mlx5_accel_esp_destroy_xfrm(sa_entry->xfrm); } if (x->xso.flags & XFRM_OFFLOAD_INBOUND) @@ -383,6 +414,14 @@ int mlx5e_ipsec_init(struct mlx5e_priv *priv) ida_init(&ipsec->halloc); ipsec->en_priv = priv; ipsec->en_priv->ipsec = ipsec; + ipsec->no_trailer = !!(mlx5_accel_ipsec_device_caps(priv->mdev) & + MLX5_ACCEL_IPSEC_CAP_RX_NO_TRAILER); + ipsec->wq = alloc_ordered_workqueue("mlx5e_ipsec: %s", 0, + priv->netdev->name); + if (!ipsec->wq) { + kfree(ipsec); + return -ENOMEM; + } netdev_dbg(priv->netdev, "IPSec attached to netdevice\n"); return 0; } @@ -394,6 +433,9 @@ void mlx5e_ipsec_cleanup(struct mlx5e_priv *priv) if (!ipsec) return; + drain_workqueue(ipsec->wq); + destroy_workqueue(ipsec->wq); + ida_destroy(&ipsec->halloc); kfree(ipsec); priv->ipsec = NULL; @@ -414,11 +456,58 @@ static bool mlx5e_ipsec_offload_ok(struct sk_buff *skb, struct xfrm_state *x) return true; } +struct mlx5e_ipsec_modify_state_work { + struct work_struct work; + struct mlx5_accel_esp_xfrm_attrs attrs; + struct mlx5e_ipsec_sa_entry *sa_entry; +}; + +static void _update_xfrm_state(struct work_struct *work) +{ + int ret; + struct mlx5e_ipsec_modify_state_work *modify_work = + container_of(work, struct mlx5e_ipsec_modify_state_work, work); + struct mlx5e_ipsec_sa_entry *sa_entry = modify_work->sa_entry; + + ret = mlx5_accel_esp_modify_xfrm(sa_entry->xfrm, + &modify_work->attrs); + if (ret) + netdev_warn(sa_entry->ipsec->en_priv->netdev, + "Not an IPSec offload device\n"); + + kfree(modify_work); +} + +static void mlx5e_xfrm_advance_esn_state(struct xfrm_state *x) +{ + struct mlx5e_ipsec_sa_entry *sa_entry = to_ipsec_sa_entry(x); + struct mlx5e_ipsec_modify_state_work *modify_work; + bool need_update; + + if (!sa_entry) + return; + + need_update = mlx5e_ipsec_update_esn_state(sa_entry); + if (!need_update) + return; + + modify_work = kzalloc(sizeof(*modify_work), GFP_ATOMIC); + if (!modify_work) + return; + + mlx5e_ipsec_build_accel_xfrm_attrs(sa_entry, &modify_work->attrs); + modify_work->sa_entry = sa_entry; + + INIT_WORK(&modify_work->work, _update_xfrm_state); + WARN_ON(!queue_work(sa_entry->ipsec->wq, &modify_work->work)); +} + static const struct xfrmdev_ops mlx5e_ipsec_xfrmdev_ops = { .xdo_dev_state_add = mlx5e_xfrm_add_state, .xdo_dev_state_delete = mlx5e_xfrm_del_state, .xdo_dev_state_free = mlx5e_xfrm_free_state, .xdo_dev_offload_ok = mlx5e_ipsec_offload_ok, + .xdo_dev_state_advance_esn = mlx5e_xfrm_advance_esn_state, }; void mlx5e_ipsec_build_netdev(struct mlx5e_priv *priv) @@ -429,7 +518,7 @@ void mlx5e_ipsec_build_netdev(struct mlx5e_priv *priv) if (!priv->ipsec) return; - if (!(mlx5_accel_ipsec_device_caps(mdev) & MLX5_ACCEL_IPSEC_ESP) || + if (!(mlx5_accel_ipsec_device_caps(mdev) & MLX5_ACCEL_IPSEC_CAP_ESP) || !MLX5_CAP_ETH(mdev, swp)) { mlx5_core_dbg(mdev, "mlx5e: ESP and SWP offload not supported\n"); return; @@ -448,7 +537,7 @@ void mlx5e_ipsec_build_netdev(struct mlx5e_priv *priv) netdev->features |= NETIF_F_HW_ESP_TX_CSUM; netdev->hw_enc_features |= NETIF_F_HW_ESP_TX_CSUM; - if (!(mlx5_accel_ipsec_device_caps(mdev) & MLX5_ACCEL_IPSEC_LSO) || + if (!(mlx5_accel_ipsec_device_caps(mdev) & MLX5_ACCEL_IPSEC_CAP_LSO) || !MLX5_CAP_ETH(mdev, swp_lso)) { mlx5_core_dbg(mdev, "mlx5e: ESP LSO not supported\n"); return; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h index 56e00baf16cc..1198fc1eba4c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h @@ -40,7 +40,11 @@ #include <net/xfrm.h> #include <linux/idr.h> +#include "accel/ipsec.h" + #define MLX5E_IPSEC_SADB_RX_BITS 10 +#define MLX5E_IPSEC_ESN_SCOPE_MID 0x80000000L + #define MLX5E_METADATA_ETHER_TYPE (0x8CE4) #define MLX5E_METADATA_ETHER_LEN 8 @@ -77,10 +81,30 @@ struct mlx5e_ipsec_stats { struct mlx5e_ipsec { struct mlx5e_priv *en_priv; DECLARE_HASHTABLE(sadb_rx, MLX5E_IPSEC_SADB_RX_BITS); + bool no_trailer; spinlock_t sadb_rx_lock; /* Protects sadb_rx and halloc */ struct ida halloc; struct mlx5e_ipsec_sw_stats sw_stats; struct mlx5e_ipsec_stats stats; + struct workqueue_struct *wq; +}; + +struct mlx5e_ipsec_esn_state { + u32 esn; + u8 trigger: 1; + u8 overlap: 1; +}; + +struct mlx5e_ipsec_sa_entry { + struct hlist_node hlist; /* Item in SADB_RX hashtable */ + struct mlx5e_ipsec_esn_state esn_state; + unsigned int handle; /* Handle in SADB_RX */ + struct xfrm_state *x; + struct mlx5e_ipsec *ipsec; + struct mlx5_accel_esp_xfrm *xfrm; + void *hw_context; + void (*set_iv_op)(struct sk_buff *skb, struct xfrm_state *x, + struct xfrm_offload *xo); }; void mlx5e_ipsec_build_inverse_table(void); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c index 6a7c8b04447e..c245d8e78509 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.c @@ -42,10 +42,11 @@ enum { MLX5E_IPSEC_RX_SYNDROME_DECRYPTED = 0x11, MLX5E_IPSEC_RX_SYNDROME_AUTH_FAILED = 0x12, + MLX5E_IPSEC_RX_SYNDROME_BAD_PROTO = 0x17, }; struct mlx5e_ipsec_rx_metadata { - unsigned char reserved; + unsigned char nexthdr; __be32 sa_handle; } __packed; @@ -175,7 +176,30 @@ static void mlx5e_ipsec_set_swp(struct sk_buff *skb, } } -static void mlx5e_ipsec_set_iv(struct sk_buff *skb, struct xfrm_offload *xo) +void mlx5e_ipsec_set_iv_esn(struct sk_buff *skb, struct xfrm_state *x, + struct xfrm_offload *xo) +{ + struct xfrm_replay_state_esn *replay_esn = x->replay_esn; + __u32 oseq = replay_esn->oseq; + int iv_offset; + __be64 seqno; + u32 seq_hi; + + if (unlikely(skb_is_gso(skb) && oseq < MLX5E_IPSEC_ESN_SCOPE_MID && + MLX5E_IPSEC_ESN_SCOPE_MID < (oseq - skb_shinfo(skb)->gso_segs))) { + seq_hi = xo->seq.hi - 1; + } else { + seq_hi = xo->seq.hi; + } + + /* Place the SN in the IV field */ + seqno = cpu_to_be64(xo->seq.low + ((u64)seq_hi << 32)); + iv_offset = skb_transport_offset(skb) + sizeof(struct ip_esp_hdr); + skb_store_bits(skb, iv_offset, &seqno, 8); +} + +void mlx5e_ipsec_set_iv(struct sk_buff *skb, struct xfrm_state *x, + struct xfrm_offload *xo) { int iv_offset; __be64 seqno; @@ -227,6 +251,7 @@ struct sk_buff *mlx5e_ipsec_handle_tx_skb(struct net_device *netdev, struct mlx5e_priv *priv = netdev_priv(netdev); struct xfrm_offload *xo = xfrm_offload(skb); struct mlx5e_ipsec_metadata *mdata; + struct mlx5e_ipsec_sa_entry *sa_entry; struct xfrm_state *x; if (!xo) @@ -261,7 +286,8 @@ struct sk_buff *mlx5e_ipsec_handle_tx_skb(struct net_device *netdev, goto drop; } mlx5e_ipsec_set_swp(skb, &wqe->eth, x->props.mode, xo); - mlx5e_ipsec_set_iv(skb, xo); + sa_entry = (struct mlx5e_ipsec_sa_entry *)x->xso.offload_handle; + sa_entry->set_iv_op(skb, x, xo); mlx5e_ipsec_set_metadata(skb, mdata, xo); return skb; @@ -301,10 +327,17 @@ mlx5e_ipsec_build_sp(struct net_device *netdev, struct sk_buff *skb, switch (mdata->syndrome) { case MLX5E_IPSEC_RX_SYNDROME_DECRYPTED: xo->status = CRYPTO_SUCCESS; + if (likely(priv->ipsec->no_trailer)) { + xo->flags |= XFRM_ESP_NO_TRAILER; + xo->proto = mdata->content.rx.nexthdr; + } break; case MLX5E_IPSEC_RX_SYNDROME_AUTH_FAILED: xo->status = CRYPTO_TUNNEL_ESP_AUTH_FAILED; break; + case MLX5E_IPSEC_RX_SYNDROME_BAD_PROTO: + xo->status = CRYPTO_INVALID_PROTOCOL; + break; default: atomic64_inc(&priv->ipsec->sw_stats.ipsec_rx_drop_syndrome); return NULL; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h index e37ae2598dbb..2bfbbef1b054 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec_rxtx.h @@ -37,6 +37,7 @@ #ifdef CONFIG_MLX5_EN_IPSEC #include <linux/skbuff.h> +#include <net/xfrm.h> #include "en.h" struct sk_buff *mlx5e_ipsec_handle_rx_skb(struct net_device *netdev, @@ -46,6 +47,10 @@ void mlx5e_ipsec_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe); void mlx5e_ipsec_inverse_table_init(void); bool mlx5e_ipsec_feature_check(struct sk_buff *skb, struct net_device *netdev, netdev_features_t features); +void mlx5e_ipsec_set_iv_esn(struct sk_buff *skb, struct xfrm_state *x, + struct xfrm_offload *xo); +void mlx5e_ipsec_set_iv(struct sk_buff *skb, struct xfrm_state *x, + struct xfrm_offload *xo); struct sk_buff *mlx5e_ipsec_handle_tx_skb(struct net_device *netdev, struct mlx5e_tx_wqe *wqe, struct sk_buff *skb); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 59ebfdae6695..37fd0245b6c1 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -203,9 +203,6 @@ void mlx5e_ethtool_get_ethtool_stats(struct mlx5e_priv *priv, { int i, idx = 0; - if (!data) - return; - mutex_lock(&priv->state_lock); mlx5e_update_stats(priv); mutex_unlock(&priv->state_lock); @@ -223,60 +220,12 @@ static void mlx5e_get_ethtool_stats(struct net_device *dev, mlx5e_ethtool_get_ethtool_stats(priv, stats, data); } -static u32 mlx5e_rx_wqes_to_packets(struct mlx5e_priv *priv, int rq_wq_type, - int num_wqe) -{ - int packets_per_wqe; - int stride_size; - int num_strides; - int wqe_size; - - if (rq_wq_type != MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) - return num_wqe; - - stride_size = 1 << priv->channels.params.mpwqe_log_stride_sz; - num_strides = 1 << priv->channels.params.mpwqe_log_num_strides; - wqe_size = stride_size * num_strides; - - packets_per_wqe = wqe_size / - ALIGN(ETH_DATA_LEN, stride_size); - return (1 << (order_base_2(num_wqe * packets_per_wqe) - 1)); -} - -static u32 mlx5e_packets_to_rx_wqes(struct mlx5e_priv *priv, int rq_wq_type, - int num_packets) -{ - int packets_per_wqe; - int stride_size; - int num_strides; - int wqe_size; - int num_wqes; - - if (rq_wq_type != MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) - return num_packets; - - stride_size = 1 << priv->channels.params.mpwqe_log_stride_sz; - num_strides = 1 << priv->channels.params.mpwqe_log_num_strides; - wqe_size = stride_size * num_strides; - - num_packets = (1 << order_base_2(num_packets)); - - packets_per_wqe = wqe_size / - ALIGN(ETH_DATA_LEN, stride_size); - num_wqes = DIV_ROUND_UP(num_packets, packets_per_wqe); - return 1 << (order_base_2(num_wqes)); -} - void mlx5e_ethtool_get_ringparam(struct mlx5e_priv *priv, struct ethtool_ringparam *param) { - int rq_wq_type = priv->channels.params.rq_wq_type; - - param->rx_max_pending = mlx5e_rx_wqes_to_packets(priv, rq_wq_type, - 1 << mlx5_max_log_rq_size(rq_wq_type)); + param->rx_max_pending = 1 << MLX5E_PARAMS_MAXIMUM_LOG_RQ_SIZE; param->tx_max_pending = 1 << MLX5E_PARAMS_MAXIMUM_LOG_SQ_SIZE; - param->rx_pending = mlx5e_rx_wqes_to_packets(priv, rq_wq_type, - 1 << priv->channels.params.log_rq_size); + param->rx_pending = 1 << priv->channels.params.log_rq_mtu_frames; param->tx_pending = 1 << priv->channels.params.log_sq_size; } @@ -291,13 +240,9 @@ static void mlx5e_get_ringparam(struct net_device *dev, int mlx5e_ethtool_set_ringparam(struct mlx5e_priv *priv, struct ethtool_ringparam *param) { - int rq_wq_type = priv->channels.params.rq_wq_type; struct mlx5e_channels new_channels = {}; - u32 rx_pending_wqes; - u32 min_rq_size; u8 log_rq_size; u8 log_sq_size; - u32 num_mtts; int err = 0; if (param->rx_jumbo_pending) { @@ -311,23 +256,10 @@ int mlx5e_ethtool_set_ringparam(struct mlx5e_priv *priv, return -EINVAL; } - min_rq_size = mlx5e_rx_wqes_to_packets(priv, rq_wq_type, - 1 << mlx5_min_log_rq_size(rq_wq_type)); - rx_pending_wqes = mlx5e_packets_to_rx_wqes(priv, rq_wq_type, - param->rx_pending); - - if (param->rx_pending < min_rq_size) { + if (param->rx_pending < (1 << MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE)) { netdev_info(priv->netdev, "%s: rx_pending (%d) < min (%d)\n", __func__, param->rx_pending, - min_rq_size); - return -EINVAL; - } - - num_mtts = MLX5E_REQUIRED_MTTS(rx_pending_wqes); - if (priv->channels.params.rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ && - !MLX5E_VALID_NUM_MTTS(num_mtts)) { - netdev_info(priv->netdev, "%s: rx_pending (%d) request can't be satisfied, try to reduce.\n", - __func__, param->rx_pending); + 1 << MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE); return -EINVAL; } @@ -338,17 +270,17 @@ int mlx5e_ethtool_set_ringparam(struct mlx5e_priv *priv, return -EINVAL; } - log_rq_size = order_base_2(rx_pending_wqes); + log_rq_size = order_base_2(param->rx_pending); log_sq_size = order_base_2(param->tx_pending); - if (log_rq_size == priv->channels.params.log_rq_size && + if (log_rq_size == priv->channels.params.log_rq_mtu_frames && log_sq_size == priv->channels.params.log_sq_size) return 0; mutex_lock(&priv->state_lock); new_channels.params = priv->channels.params; - new_channels.params.log_rq_size = log_rq_size; + new_channels.params.log_rq_mtu_frames = log_rq_size; new_channels.params.log_sq_size = log_sq_size; if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { @@ -1083,16 +1015,66 @@ static int mlx5e_get_rxnfc(struct net_device *netdev, return err; } +#define MLX5E_PFC_PREVEN_AUTO_TOUT_MSEC 100 +#define MLX5E_PFC_PREVEN_TOUT_MAX_MSEC 8000 +#define MLX5E_PFC_PREVEN_MINOR_PRECENT 85 +#define MLX5E_PFC_PREVEN_TOUT_MIN_MSEC 80 +#define MLX5E_DEVICE_STALL_MINOR_WATERMARK(critical_tout) \ + max_t(u16, MLX5E_PFC_PREVEN_TOUT_MIN_MSEC, \ + (critical_tout * MLX5E_PFC_PREVEN_MINOR_PRECENT) / 100) + +static int mlx5e_get_pfc_prevention_tout(struct net_device *netdev, + u16 *pfc_prevention_tout) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + + if (!MLX5_CAP_PCAM_FEATURE((priv)->mdev, pfcc_mask) || + !MLX5_CAP_DEBUG((priv)->mdev, stall_detect)) + return -EOPNOTSUPP; + + return mlx5_query_port_stall_watermark(mdev, pfc_prevention_tout, NULL); +} + +static int mlx5e_set_pfc_prevention_tout(struct net_device *netdev, + u16 pfc_preven) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + u16 critical_tout; + u16 minor; + + if (!MLX5_CAP_PCAM_FEATURE((priv)->mdev, pfcc_mask) || + !MLX5_CAP_DEBUG((priv)->mdev, stall_detect)) + return -EOPNOTSUPP; + + critical_tout = (pfc_preven == PFC_STORM_PREVENTION_AUTO) ? + MLX5E_PFC_PREVEN_AUTO_TOUT_MSEC : + pfc_preven; + + if (critical_tout != PFC_STORM_PREVENTION_DISABLE && + (critical_tout > MLX5E_PFC_PREVEN_TOUT_MAX_MSEC || + critical_tout < MLX5E_PFC_PREVEN_TOUT_MIN_MSEC)) { + netdev_info(netdev, "%s: pfc prevention tout not in range (%d-%d)\n", + __func__, MLX5E_PFC_PREVEN_TOUT_MIN_MSEC, + MLX5E_PFC_PREVEN_TOUT_MAX_MSEC); + return -EINVAL; + } + + minor = MLX5E_DEVICE_STALL_MINOR_WATERMARK(critical_tout); + return mlx5_set_port_stall_watermark(mdev, critical_tout, + minor); +} + static int mlx5e_get_tunable(struct net_device *dev, const struct ethtool_tunable *tuna, void *data) { - const struct mlx5e_priv *priv = netdev_priv(dev); - int err = 0; + int err; switch (tuna->id) { - case ETHTOOL_TX_COPYBREAK: - *(u32 *)data = priv->channels.params.tx_max_inline; + case ETHTOOL_PFC_PREVENTION_TOUT: + err = mlx5e_get_pfc_prevention_tout(dev, data); break; default: err = -EINVAL; @@ -1107,34 +1089,13 @@ static int mlx5e_set_tunable(struct net_device *dev, const void *data) { struct mlx5e_priv *priv = netdev_priv(dev); - struct mlx5_core_dev *mdev = priv->mdev; - struct mlx5e_channels new_channels = {}; - int err = 0; - u32 val; + int err; mutex_lock(&priv->state_lock); switch (tuna->id) { - case ETHTOOL_TX_COPYBREAK: - val = *(u32 *)data; - if (val > mlx5e_get_max_inline_cap(mdev)) { - err = -EINVAL; - break; - } - - new_channels.params = priv->channels.params; - new_channels.params.tx_max_inline = val; - - if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { - priv->channels.params = new_channels.params; - break; - } - - err = mlx5e_open_channels(priv, &new_channels); - if (err) - break; - mlx5e_switch_priv_channels(priv, &new_channels, NULL); - + case ETHTOOL_PFC_PREVENTION_TOUT: + err = mlx5e_set_pfc_prevention_tout(dev, *(u16 *)data); break; default: err = -EINVAL; @@ -1524,11 +1485,6 @@ int mlx5e_modify_rx_cqe_compression_locked(struct mlx5e_priv *priv, bool new_val new_channels.params = priv->channels.params; MLX5E_SET_PFLAG(&new_channels.params, MLX5E_PFLAG_RX_CQE_COMPRESS, new_val); - new_channels.params.mpwqe_log_stride_sz = - MLX5E_MPWQE_STRIDE_SZ(priv->mdev, new_val); - new_channels.params.mpwqe_log_num_strides = - MLX5_MPWRQ_LOG_WQE_SZ - new_channels.params.mpwqe_log_stride_sz; - if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { priv->channels.params = new_channels.params; return 0; @@ -1566,6 +1522,38 @@ static int set_pflag_rx_cqe_compress(struct net_device *netdev, return 0; } +static int set_pflag_rx_striding_rq(struct net_device *netdev, bool enable) +{ + struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; + struct mlx5e_channels new_channels = {}; + int err; + + if (enable) { + if (!mlx5e_check_fragmented_striding_rq_cap(mdev)) + return -EOPNOTSUPP; + if (!mlx5e_striding_rq_possible(mdev, &priv->channels.params)) + return -EINVAL; + } + + new_channels.params = priv->channels.params; + + MLX5E_SET_PFLAG(&new_channels.params, MLX5E_PFLAG_RX_STRIDING_RQ, enable); + mlx5e_set_rq_type(mdev, &new_channels.params); + + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { + priv->channels.params = new_channels.params; + return 0; + } + + err = mlx5e_open_channels(priv, &new_channels); + if (err) + return err; + + mlx5e_switch_priv_channels(priv, &new_channels, NULL); + return 0; +} + static int mlx5e_handle_pflag(struct net_device *netdev, u32 wanted_flags, enum mlx5e_priv_flag flag, @@ -1611,6 +1599,12 @@ static int mlx5e_set_priv_flags(struct net_device *netdev, u32 pflags) err = mlx5e_handle_pflag(netdev, pflags, MLX5E_PFLAG_RX_CQE_COMPRESS, set_pflag_rx_cqe_compress); + if (err) + goto out; + + err = mlx5e_handle_pflag(netdev, pflags, + MLX5E_PFLAG_RX_STRIDING_RQ, + set_pflag_rx_striding_rq); out: mutex_unlock(&priv->state_lock); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c index 9b4827d36e3e..0aab3afc6885 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c @@ -71,56 +71,145 @@ struct mlx5e_channel_param { struct mlx5e_cq_param icosq_cq; }; -static bool mlx5e_check_fragmented_striding_rq_cap(struct mlx5_core_dev *mdev) +bool mlx5e_check_fragmented_striding_rq_cap(struct mlx5_core_dev *mdev) { - return MLX5_CAP_GEN(mdev, striding_rq) && + bool striding_rq_umr = MLX5_CAP_GEN(mdev, striding_rq) && MLX5_CAP_GEN(mdev, umr_ptr_rlky) && MLX5_CAP_ETH(mdev, reg_umr_sq); + u16 max_wqe_sz_cap = MLX5_CAP_GEN(mdev, max_wqe_sz_sq); + bool inline_umr = MLX5E_UMR_WQE_INLINE_SZ <= max_wqe_sz_cap; + + if (!striding_rq_umr) + return false; + if (!inline_umr) { + mlx5_core_warn(mdev, "Cannot support Striding RQ: UMR WQE size (%d) exceeds maximum supported (%d).\n", + (int)MLX5E_UMR_WQE_INLINE_SZ, max_wqe_sz_cap); + return false; + } + return true; +} + +static u32 mlx5e_mpwqe_get_linear_frag_sz(struct mlx5e_params *params) +{ + if (!params->xdp_prog) { + u16 hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); + u16 rq_headroom = MLX5_RX_HEADROOM + NET_IP_ALIGN; + + return MLX5_SKB_FRAG_SZ(rq_headroom + hw_mtu); + } + + return PAGE_SIZE; +} + +static u8 mlx5e_mpwqe_log_pkts_per_wqe(struct mlx5e_params *params) +{ + u32 linear_frag_sz = mlx5e_mpwqe_get_linear_frag_sz(params); + + return MLX5_MPWRQ_LOG_WQE_SZ - order_base_2(linear_frag_sz); +} + +static bool mlx5e_rx_mpwqe_is_linear_skb(struct mlx5_core_dev *mdev, + struct mlx5e_params *params) +{ + u32 frag_sz = mlx5e_mpwqe_get_linear_frag_sz(params); + s8 signed_log_num_strides_param; + u8 log_num_strides; + + if (params->lro_en || frag_sz > PAGE_SIZE) + return false; + + if (MLX5_CAP_GEN(mdev, ext_stride_num_range)) + return true; + + log_num_strides = MLX5_MPWRQ_LOG_WQE_SZ - order_base_2(frag_sz); + signed_log_num_strides_param = + (s8)log_num_strides - MLX5_MPWQE_LOG_NUM_STRIDES_BASE; + + return signed_log_num_strides_param >= 0; +} + +static u8 mlx5e_mpwqe_get_log_rq_size(struct mlx5e_params *params) +{ + if (params->log_rq_mtu_frames < + mlx5e_mpwqe_log_pkts_per_wqe(params) + MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE_MPW) + return MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE_MPW; + + return params->log_rq_mtu_frames - mlx5e_mpwqe_log_pkts_per_wqe(params); +} + +static u8 mlx5e_mpwqe_get_log_stride_size(struct mlx5_core_dev *mdev, + struct mlx5e_params *params) +{ + if (mlx5e_rx_mpwqe_is_linear_skb(mdev, params)) + return order_base_2(mlx5e_mpwqe_get_linear_frag_sz(params)); + + return MLX5E_MPWQE_STRIDE_SZ(mdev, + MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS)); +} + +static u8 mlx5e_mpwqe_get_log_num_strides(struct mlx5_core_dev *mdev, + struct mlx5e_params *params) +{ + return MLX5_MPWRQ_LOG_WQE_SZ - + mlx5e_mpwqe_get_log_stride_size(mdev, params); +} + +static u16 mlx5e_get_rq_headroom(struct mlx5_core_dev *mdev, + struct mlx5e_params *params) +{ + u16 linear_rq_headroom = params->xdp_prog ? + XDP_PACKET_HEADROOM : MLX5_RX_HEADROOM; + + linear_rq_headroom += NET_IP_ALIGN; + + if (params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST) + return linear_rq_headroom; + + if (mlx5e_rx_mpwqe_is_linear_skb(mdev, params)) + return linear_rq_headroom; + + return 0; } void mlx5e_init_rq_type_params(struct mlx5_core_dev *mdev, - struct mlx5e_params *params, u8 rq_type) + struct mlx5e_params *params) { - params->rq_wq_type = rq_type; params->lro_wqe_sz = MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ; + params->log_rq_mtu_frames = is_kdump_kernel() ? + MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE : + MLX5E_PARAMS_DEFAULT_LOG_RQ_SIZE; switch (params->rq_wq_type) { case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: - params->log_rq_size = is_kdump_kernel() ? - MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE_MPW : - MLX5E_PARAMS_DEFAULT_LOG_RQ_SIZE_MPW; - params->mpwqe_log_stride_sz = MLX5E_MPWQE_STRIDE_SZ(mdev, - MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS)); - params->mpwqe_log_num_strides = MLX5_MPWRQ_LOG_WQE_SZ - - params->mpwqe_log_stride_sz; break; default: /* MLX5_WQ_TYPE_LINKED_LIST */ - params->log_rq_size = is_kdump_kernel() ? - MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE : - MLX5E_PARAMS_DEFAULT_LOG_RQ_SIZE; - params->rq_headroom = params->xdp_prog ? - XDP_PACKET_HEADROOM : MLX5_RX_HEADROOM; - params->rq_headroom += NET_IP_ALIGN; - /* Extra room needed for build_skb */ - params->lro_wqe_sz -= params->rq_headroom + + params->lro_wqe_sz -= mlx5e_get_rq_headroom(mdev, params) + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)); } mlx5_core_info(mdev, "MLX5E: StrdRq(%d) RqSz(%ld) StrdSz(%ld) RxCqeCmprss(%d)\n", params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ, - BIT(params->log_rq_size), - BIT(params->mpwqe_log_stride_sz), + params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ ? + BIT(mlx5e_mpwqe_get_log_rq_size(params)) : + BIT(params->log_rq_mtu_frames), + BIT(mlx5e_mpwqe_get_log_stride_size(mdev, params)), MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS)); } -static void mlx5e_set_rq_params(struct mlx5_core_dev *mdev, +bool mlx5e_striding_rq_possible(struct mlx5_core_dev *mdev, struct mlx5e_params *params) { - u8 rq_type = mlx5e_check_fragmented_striding_rq_cap(mdev) && - !params->xdp_prog && !MLX5_IPSEC_DEV(mdev) ? - MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ : - MLX5_WQ_TYPE_LINKED_LIST; - mlx5e_init_rq_type_params(mdev, params, rq_type); + return mlx5e_check_fragmented_striding_rq_cap(mdev) && + !MLX5_IPSEC_DEV(mdev) && + !(params->xdp_prog && !mlx5e_rx_mpwqe_is_linear_skb(mdev, params)); +} + +void mlx5e_set_rq_type(struct mlx5_core_dev *mdev, struct mlx5e_params *params) +{ + params->rq_wq_type = mlx5e_striding_rq_possible(mdev, params) && + MLX5E_GET_PFLAG(params, MLX5E_PFLAG_RX_STRIDING_RQ) ? + MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ : + MLX5_WQ_TYPE_LINKED_LIST; } static void mlx5e_update_carrier(struct mlx5e_priv *priv) @@ -153,26 +242,6 @@ static void mlx5e_update_carrier_work(struct work_struct *work) mutex_unlock(&priv->state_lock); } -static void mlx5e_tx_timeout_work(struct work_struct *work) -{ - struct mlx5e_priv *priv = container_of(work, struct mlx5e_priv, - tx_timeout_work); - int err; - - rtnl_lock(); - mutex_lock(&priv->state_lock); - if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) - goto unlock; - mlx5e_close_locked(priv->netdev); - err = mlx5e_open_locked(priv->netdev); - if (err) - netdev_err(priv->netdev, "mlx5e_open_locked failed recovering from a tx_timeout, err(%d).\n", - err); -unlock: - mutex_unlock(&priv->state_lock); - rtnl_unlock(); -} - void mlx5e_update_stats(struct mlx5e_priv *priv) { int i; @@ -235,107 +304,38 @@ static void mlx5e_disable_async_events(struct mlx5e_priv *priv) synchronize_irq(pci_irq_vector(priv->mdev->pdev, MLX5_EQ_VEC_ASYNC)); } -static inline int mlx5e_get_wqe_mtt_sz(void) -{ - /* UMR copies MTTs in units of MLX5_UMR_MTT_ALIGNMENT bytes. - * To avoid copying garbage after the mtt array, we allocate - * a little more. - */ - return ALIGN(MLX5_MPWRQ_PAGES_PER_WQE * sizeof(__be64), - MLX5_UMR_MTT_ALIGNMENT); -} - static inline void mlx5e_build_umr_wqe(struct mlx5e_rq *rq, struct mlx5e_icosq *sq, - struct mlx5e_umr_wqe *wqe, - u16 ix) + struct mlx5e_umr_wqe *wqe) { struct mlx5_wqe_ctrl_seg *cseg = &wqe->ctrl; struct mlx5_wqe_umr_ctrl_seg *ucseg = &wqe->uctrl; - struct mlx5_wqe_data_seg *dseg = &wqe->data; - struct mlx5e_mpw_info *wi = &rq->mpwqe.info[ix]; - u8 ds_cnt = DIV_ROUND_UP(sizeof(*wqe), MLX5_SEND_WQE_DS); - u32 umr_wqe_mtt_offset = mlx5e_get_wqe_mtt_offset(rq, ix); + u8 ds_cnt = DIV_ROUND_UP(MLX5E_UMR_WQE_INLINE_SZ, MLX5_SEND_WQE_DS); cseg->qpn_ds = cpu_to_be32((sq->sqn << MLX5_WQE_CTRL_QPN_SHIFT) | ds_cnt); cseg->fm_ce_se = MLX5_WQE_CTRL_CQ_UPDATE; cseg->imm = rq->mkey_be; - ucseg->flags = MLX5_UMR_TRANSLATION_OFFSET_EN; + ucseg->flags = MLX5_UMR_TRANSLATION_OFFSET_EN | MLX5_UMR_INLINE; ucseg->xlt_octowords = cpu_to_be16(MLX5_MTT_OCTW(MLX5_MPWRQ_PAGES_PER_WQE)); - ucseg->bsf_octowords = - cpu_to_be16(MLX5_MTT_OCTW(umr_wqe_mtt_offset)); ucseg->mkey_mask = cpu_to_be64(MLX5_MKEY_MASK_FREE); - - dseg->lkey = sq->mkey_be; - dseg->addr = cpu_to_be64(wi->umr.mtt_addr); } static int mlx5e_rq_alloc_mpwqe_info(struct mlx5e_rq *rq, struct mlx5e_channel *c) { int wq_sz = mlx5_wq_ll_get_size(&rq->wq); - int mtt_sz = mlx5e_get_wqe_mtt_sz(); - int mtt_alloc = mtt_sz + MLX5_UMR_ALIGN - 1; - int i; rq->mpwqe.info = kzalloc_node(wq_sz * sizeof(*rq->mpwqe.info), GFP_KERNEL, cpu_to_node(c->cpu)); if (!rq->mpwqe.info) - goto err_out; - - /* We allocate more than mtt_sz as we will align the pointer */ - rq->mpwqe.mtt_no_align = kzalloc_node(mtt_alloc * wq_sz, GFP_KERNEL, - cpu_to_node(c->cpu)); - if (unlikely(!rq->mpwqe.mtt_no_align)) - goto err_free_wqe_info; - - for (i = 0; i < wq_sz; i++) { - struct mlx5e_mpw_info *wi = &rq->mpwqe.info[i]; - - wi->umr.mtt = PTR_ALIGN(rq->mpwqe.mtt_no_align + i * mtt_alloc, - MLX5_UMR_ALIGN); - wi->umr.mtt_addr = dma_map_single(c->pdev, wi->umr.mtt, mtt_sz, - PCI_DMA_TODEVICE); - if (unlikely(dma_mapping_error(c->pdev, wi->umr.mtt_addr))) - goto err_unmap_mtts; + return -ENOMEM; - mlx5e_build_umr_wqe(rq, &c->icosq, &wi->umr.wqe, i); - } + mlx5e_build_umr_wqe(rq, &c->icosq, &rq->mpwqe.umr_wqe); return 0; - -err_unmap_mtts: - while (--i >= 0) { - struct mlx5e_mpw_info *wi = &rq->mpwqe.info[i]; - - dma_unmap_single(c->pdev, wi->umr.mtt_addr, mtt_sz, - PCI_DMA_TODEVICE); - } - kfree(rq->mpwqe.mtt_no_align); -err_free_wqe_info: - kfree(rq->mpwqe.info); - -err_out: - return -ENOMEM; -} - -static void mlx5e_rq_free_mpwqe_info(struct mlx5e_rq *rq) -{ - int wq_sz = mlx5_wq_ll_get_size(&rq->wq); - int mtt_sz = mlx5e_get_wqe_mtt_sz(); - int i; - - for (i = 0; i < wq_sz; i++) { - struct mlx5e_mpw_info *wi = &rq->mpwqe.info[i]; - - dma_unmap_single(rq->pdev, wi->umr.mtt_addr, mtt_sz, - PCI_DMA_TODEVICE); - } - kfree(rq->mpwqe.mtt_no_align); - kfree(rq->mpwqe.info); } static int mlx5e_create_umr_mkey(struct mlx5_core_dev *mdev, @@ -347,9 +347,6 @@ static int mlx5e_create_umr_mkey(struct mlx5_core_dev *mdev, u32 *in; int err; - if (!MLX5E_VALID_NUM_MTTS(npages)) - return -EINVAL; - in = kvzalloc(inlen, GFP_KERNEL); if (!in) return -ENOMEM; @@ -382,6 +379,11 @@ static int mlx5e_create_rq_umr_mkey(struct mlx5_core_dev *mdev, struct mlx5e_rq return mlx5e_create_umr_mkey(mdev, num_mtts, PAGE_SHIFT, &rq->umr_mkey); } +static inline u64 mlx5e_get_mpwqe_offset(struct mlx5e_rq *rq, u16 wqe_ix) +{ + return (wqe_ix << MLX5E_LOG_ALIGNED_MPWQE_PPW) << PAGE_SHIFT; +} + static int mlx5e_alloc_rq(struct mlx5e_channel *c, struct mlx5e_params *params, struct mlx5e_rq_param *rqp, @@ -415,6 +417,7 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, rq->channel = c; rq->ix = c->ix; rq->mdev = mdev; + rq->hw_mtu = MLX5E_SW2HW_MTU(params, params->sw_mtu); rq->xdp_prog = params->xdp_prog ? bpf_prog_inc(params->xdp_prog) : NULL; if (IS_ERR(rq->xdp_prog)) { @@ -428,11 +431,10 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, goto err_rq_wq_destroy; rq->buff.map_dir = rq->xdp_prog ? DMA_BIDIRECTIONAL : DMA_FROM_DEVICE; - rq->buff.headroom = params->rq_headroom; + rq->buff.headroom = mlx5e_get_rq_headroom(mdev, params); switch (rq->wq_type) { case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: - rq->post_wqes = mlx5e_post_rx_mpwqes; rq->dealloc_wqe = mlx5e_dealloc_rx_mpwqe; @@ -450,8 +452,12 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, goto err_rq_wq_destroy; } - rq->mpwqe.log_stride_sz = params->mpwqe_log_stride_sz; - rq->mpwqe.num_strides = BIT(params->mpwqe_log_num_strides); + rq->mpwqe.skb_from_cqe_mpwrq = + mlx5e_rx_mpwqe_is_linear_skb(mdev, params) ? + mlx5e_skb_from_cqe_mpwrq_linear : + mlx5e_skb_from_cqe_mpwrq_nonlinear; + rq->mpwqe.log_stride_sz = mlx5e_mpwqe_get_log_stride_size(mdev, params); + rq->mpwqe.num_strides = BIT(mlx5e_mpwqe_get_log_num_strides(mdev, params)); byte_count = rq->mpwqe.num_strides << rq->mpwqe.log_stride_sz; @@ -490,7 +496,7 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, byte_count = params->lro_en ? params->lro_wqe_sz : - MLX5E_SW2HW_MTU(c->priv, c->netdev->mtu); + MLX5E_SW2HW_MTU(params, params->sw_mtu); #ifdef CONFIG_MLX5_EN_IPSEC if (MLX5_IPSEC_DEV(mdev)) byte_count += MLX5E_METADATA_ETHER_LEN; @@ -510,9 +516,9 @@ static int mlx5e_alloc_rq(struct mlx5e_channel *c, struct mlx5e_rx_wqe *wqe = mlx5_wq_ll_get_wqe(&rq->wq, i); if (rq->wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) { - u64 dma_offset = (u64)mlx5e_get_wqe_mtt_offset(rq, i) << PAGE_SHIFT; + u64 dma_offset = mlx5e_get_mpwqe_offset(rq, i); - wqe->data.addr = cpu_to_be64(dma_offset); + wqe->data.addr = cpu_to_be64(dma_offset + rq->buff.headroom); } wqe->data.byte_count = cpu_to_be32(byte_count); @@ -558,7 +564,7 @@ static void mlx5e_free_rq(struct mlx5e_rq *rq) switch (rq->wq_type) { case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: - mlx5e_rq_free_mpwqe_info(rq); + kfree(rq->mpwqe.info); mlx5_core_destroy_mkey(rq->mdev, &rq->umr_mkey); break; default: /* MLX5_WQ_TYPE_LINKED_LIST */ @@ -615,8 +621,7 @@ static int mlx5e_create_rq(struct mlx5e_rq *rq, static int mlx5e_modify_rq_state(struct mlx5e_rq *rq, int curr_state, int next_state) { - struct mlx5e_channel *c = rq->channel; - struct mlx5_core_dev *mdev = c->mdev; + struct mlx5_core_dev *mdev = rq->mdev; void *in; void *rqc; @@ -898,7 +903,6 @@ static int mlx5e_alloc_icosq(struct mlx5e_channel *c, struct mlx5_core_dev *mdev = c->mdev; int err; - sq->mkey_be = c->mkey_be; sq->channel = c; sq->uar_map = mdev->mlx5e_res.bfreg.map; @@ -953,6 +957,7 @@ static int mlx5e_alloc_txqsq_db(struct mlx5e_txqsq *sq, int numa) return 0; } +static void mlx5e_sq_recover(struct work_struct *work); static int mlx5e_alloc_txqsq(struct mlx5e_channel *c, int txq_ix, struct mlx5e_params *params, @@ -970,8 +975,8 @@ static int mlx5e_alloc_txqsq(struct mlx5e_channel *c, sq->channel = c; sq->txq_ix = txq_ix; sq->uar_map = mdev->mlx5e_res.bfreg.map; - sq->max_inline = params->tx_max_inline; sq->min_inline_mode = params->tx_min_inline_mode; + INIT_WORK(&sq->recover.recover_work, mlx5e_sq_recover); if (MLX5_IPSEC_DEV(c->priv->mdev)) set_bit(MLX5E_SQ_STATE_IPSEC, &sq->state); @@ -1038,6 +1043,7 @@ static int mlx5e_create_sq(struct mlx5_core_dev *mdev, MLX5_SET(sqc, sqc, min_wqe_inline_mode, csp->min_inline_mode); MLX5_SET(sqc, sqc, state, MLX5_SQC_STATE_RST); + MLX5_SET(sqc, sqc, flush_in_error_en, 1); MLX5_SET(wq, wq, wq_type, MLX5_WQ_TYPE_CYCLIC); MLX5_SET(wq, wq, uar_page, mdev->mlx5e_res.bfreg.index); @@ -1156,9 +1162,20 @@ err_free_txqsq: return err; } +static void mlx5e_reset_txqsq_cc_pc(struct mlx5e_txqsq *sq) +{ + WARN_ONCE(sq->cc != sq->pc, + "SQ 0x%x: cc (0x%x) != pc (0x%x)\n", + sq->sqn, sq->cc, sq->pc); + sq->cc = 0; + sq->dma_fifo_cc = 0; + sq->pc = 0; +} + static void mlx5e_activate_txqsq(struct mlx5e_txqsq *sq) { sq->txq = netdev_get_tx_queue(sq->channel->netdev, sq->txq_ix); + clear_bit(MLX5E_SQ_STATE_RECOVERING, &sq->state); set_bit(MLX5E_SQ_STATE_ENABLED, &sq->state); netdev_tx_reset_queue(sq->txq); netif_tx_start_queue(sq->txq); @@ -1203,6 +1220,107 @@ static void mlx5e_close_txqsq(struct mlx5e_txqsq *sq) mlx5e_free_txqsq(sq); } +static int mlx5e_wait_for_sq_flush(struct mlx5e_txqsq *sq) +{ + unsigned long exp_time = jiffies + msecs_to_jiffies(2000); + + while (time_before(jiffies, exp_time)) { + if (sq->cc == sq->pc) + return 0; + + msleep(20); + } + + netdev_err(sq->channel->netdev, + "Wait for SQ 0x%x flush timeout (sq cc = 0x%x, sq pc = 0x%x)\n", + sq->sqn, sq->cc, sq->pc); + + return -ETIMEDOUT; +} + +static int mlx5e_sq_to_ready(struct mlx5e_txqsq *sq, int curr_state) +{ + struct mlx5_core_dev *mdev = sq->channel->mdev; + struct net_device *dev = sq->channel->netdev; + struct mlx5e_modify_sq_param msp = {0}; + int err; + + msp.curr_state = curr_state; + msp.next_state = MLX5_SQC_STATE_RST; + + err = mlx5e_modify_sq(mdev, sq->sqn, &msp); + if (err) { + netdev_err(dev, "Failed to move sq 0x%x to reset\n", sq->sqn); + return err; + } + + memset(&msp, 0, sizeof(msp)); + msp.curr_state = MLX5_SQC_STATE_RST; + msp.next_state = MLX5_SQC_STATE_RDY; + + err = mlx5e_modify_sq(mdev, sq->sqn, &msp); + if (err) { + netdev_err(dev, "Failed to move sq 0x%x to ready\n", sq->sqn); + return err; + } + + return 0; +} + +static void mlx5e_sq_recover(struct work_struct *work) +{ + struct mlx5e_txqsq_recover *recover = + container_of(work, struct mlx5e_txqsq_recover, + recover_work); + struct mlx5e_txqsq *sq = container_of(recover, struct mlx5e_txqsq, + recover); + struct mlx5_core_dev *mdev = sq->channel->mdev; + struct net_device *dev = sq->channel->netdev; + u8 state; + int err; + + err = mlx5_core_query_sq_state(mdev, sq->sqn, &state); + if (err) { + netdev_err(dev, "Failed to query SQ 0x%x state. err = %d\n", + sq->sqn, err); + return; + } + + if (state != MLX5_RQC_STATE_ERR) { + netdev_err(dev, "SQ 0x%x not in ERROR state\n", sq->sqn); + return; + } + + netif_tx_disable_queue(sq->txq); + + if (mlx5e_wait_for_sq_flush(sq)) + return; + + /* If the interval between two consecutive recovers per SQ is too + * short, don't recover to avoid infinite loop of ERR_CQE -> recover. + * If we reached this state, there is probably a bug that needs to be + * fixed. let's keep the queue close and let tx timeout cleanup. + */ + if (jiffies_to_msecs(jiffies - recover->last_recover) < + MLX5E_SQ_RECOVER_MIN_INTERVAL) { + netdev_err(dev, "Recover SQ 0x%x canceled, too many error CQEs\n", + sq->sqn); + return; + } + + /* At this point, no new packets will arrive from the stack as TXQ is + * marked with QUEUE_STATE_DRV_XOFF. In addition, NAPI cleared all + * pending WQEs. SQ can safely reset the SQ. + */ + if (mlx5e_sq_to_ready(sq, state)) + return; + + mlx5e_reset_txqsq_cc_pc(sq); + sq->stats.recover++; + recover->last_recover = jiffies; + mlx5e_activate_txqsq(sq); +} + static int mlx5e_open_icosq(struct mlx5e_channel *c, struct mlx5e_params *params, struct mlx5e_sq_param *param, @@ -1743,39 +1861,47 @@ static void mlx5e_build_rq_param(struct mlx5e_priv *priv, struct mlx5e_params *params, struct mlx5e_rq_param *param) { + struct mlx5_core_dev *mdev = priv->mdev; void *rqc = param->rqc; void *wq = MLX5_ADDR_OF(rqc, rqc, wq); switch (params->rq_wq_type) { case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: - MLX5_SET(wq, wq, log_wqe_num_of_strides, params->mpwqe_log_num_strides - 9); - MLX5_SET(wq, wq, log_wqe_stride_size, params->mpwqe_log_stride_sz - 6); + MLX5_SET(wq, wq, log_wqe_num_of_strides, + mlx5e_mpwqe_get_log_num_strides(mdev, params) - + MLX5_MPWQE_LOG_NUM_STRIDES_BASE); + MLX5_SET(wq, wq, log_wqe_stride_size, + mlx5e_mpwqe_get_log_stride_size(mdev, params) - + MLX5_MPWQE_LOG_STRIDE_SZ_BASE); MLX5_SET(wq, wq, wq_type, MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ); + MLX5_SET(wq, wq, log_wq_sz, mlx5e_mpwqe_get_log_rq_size(params)); break; default: /* MLX5_WQ_TYPE_LINKED_LIST */ MLX5_SET(wq, wq, wq_type, MLX5_WQ_TYPE_LINKED_LIST); + MLX5_SET(wq, wq, log_wq_sz, params->log_rq_mtu_frames); } MLX5_SET(wq, wq, end_padding_mode, MLX5_WQ_END_PAD_MODE_ALIGN); MLX5_SET(wq, wq, log_wq_stride, ilog2(sizeof(struct mlx5e_rx_wqe))); - MLX5_SET(wq, wq, log_wq_sz, params->log_rq_size); - MLX5_SET(wq, wq, pd, priv->mdev->mlx5e_res.pdn); + MLX5_SET(wq, wq, pd, mdev->mlx5e_res.pdn); MLX5_SET(rqc, rqc, counter_set_id, priv->q_counter); MLX5_SET(rqc, rqc, vsd, params->vlan_strip_disable); MLX5_SET(rqc, rqc, scatter_fcs, params->scatter_fcs_en); - param->wq.buf_numa_node = dev_to_node(&priv->mdev->pdev->dev); + param->wq.buf_numa_node = dev_to_node(&mdev->pdev->dev); param->wq.linear = 1; } -static void mlx5e_build_drop_rq_param(struct mlx5_core_dev *mdev, +static void mlx5e_build_drop_rq_param(struct mlx5e_priv *priv, struct mlx5e_rq_param *param) { + struct mlx5_core_dev *mdev = priv->mdev; void *rqc = param->rqc; void *wq = MLX5_ADDR_OF(rqc, rqc, wq); MLX5_SET(wq, wq, wq_type, MLX5_WQ_TYPE_LINKED_LIST); MLX5_SET(wq, wq, log_wq_stride, ilog2(sizeof(struct mlx5e_rx_wqe))); + MLX5_SET(rqc, rqc, counter_set_id, priv->drop_rq_q_counter); param->wq.buf_numa_node = dev_to_node(&mdev->pdev->dev); } @@ -1816,15 +1942,17 @@ static void mlx5e_build_rx_cq_param(struct mlx5e_priv *priv, struct mlx5e_params *params, struct mlx5e_cq_param *param) { + struct mlx5_core_dev *mdev = priv->mdev; void *cqc = param->cqc; u8 log_cq_size; switch (params->rq_wq_type) { case MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ: - log_cq_size = params->log_rq_size + params->mpwqe_log_num_strides; + log_cq_size = mlx5e_mpwqe_get_log_rq_size(params) + + mlx5e_mpwqe_get_log_num_strides(mdev, params); break; default: /* MLX5_WQ_TYPE_LINKED_LIST */ - log_cq_size = params->log_rq_size; + log_cq_size = params->log_rq_mtu_frames; } MLX5_SET(cqc, cqc, log_cq_size, log_cq_size); @@ -2375,10 +2503,10 @@ static void mlx5e_build_inner_indir_tir_ctx(struct mlx5e_priv *priv, mlx5e_build_indir_tir_ctx_hash(&priv->channels.params, tt, tirc, true); } -static int mlx5e_set_mtu(struct mlx5e_priv *priv, u16 mtu) +static int mlx5e_set_mtu(struct mlx5_core_dev *mdev, + struct mlx5e_params *params, u16 mtu) { - struct mlx5_core_dev *mdev = priv->mdev; - u16 hw_mtu = MLX5E_SW2HW_MTU(priv, mtu); + u16 hw_mtu = MLX5E_SW2HW_MTU(params, mtu); int err; err = mlx5_set_port_mtu(mdev, hw_mtu, 1); @@ -2390,9 +2518,9 @@ static int mlx5e_set_mtu(struct mlx5e_priv *priv, u16 mtu) return 0; } -static void mlx5e_query_mtu(struct mlx5e_priv *priv, u16 *mtu) +static void mlx5e_query_mtu(struct mlx5_core_dev *mdev, + struct mlx5e_params *params, u16 *mtu) { - struct mlx5_core_dev *mdev = priv->mdev; u16 hw_mtu = 0; int err; @@ -2400,25 +2528,27 @@ static void mlx5e_query_mtu(struct mlx5e_priv *priv, u16 *mtu) if (err || !hw_mtu) /* fallback to port oper mtu */ mlx5_query_port_oper_mtu(mdev, &hw_mtu, 1); - *mtu = MLX5E_HW2SW_MTU(priv, hw_mtu); + *mtu = MLX5E_HW2SW_MTU(params, hw_mtu); } static int mlx5e_set_dev_port_mtu(struct mlx5e_priv *priv) { + struct mlx5e_params *params = &priv->channels.params; struct net_device *netdev = priv->netdev; + struct mlx5_core_dev *mdev = priv->mdev; u16 mtu; int err; - err = mlx5e_set_mtu(priv, netdev->mtu); + err = mlx5e_set_mtu(mdev, params, params->sw_mtu); if (err) return err; - mlx5e_query_mtu(priv, &mtu); - if (mtu != netdev->mtu) + mlx5e_query_mtu(mdev, params, &mtu); + if (mtu != params->sw_mtu) netdev_warn(netdev, "%s: VPort MTU %d is different than netdev mtu %d\n", - __func__, mtu, netdev->mtu); + __func__, mtu, params->sw_mtu); - netdev->mtu = mtu; + params->sw_mtu = mtu; return 0; } @@ -2646,15 +2776,16 @@ static int mlx5e_alloc_drop_cq(struct mlx5_core_dev *mdev, return mlx5e_alloc_cq_common(mdev, param, cq); } -static int mlx5e_open_drop_rq(struct mlx5_core_dev *mdev, +static int mlx5e_open_drop_rq(struct mlx5e_priv *priv, struct mlx5e_rq *drop_rq) { + struct mlx5_core_dev *mdev = priv->mdev; struct mlx5e_cq_param cq_param = {}; struct mlx5e_rq_param rq_param = {}; struct mlx5e_cq *cq = &drop_rq->cq; int err; - mlx5e_build_drop_rq_param(mdev, &rq_param); + mlx5e_build_drop_rq_param(priv, &rq_param); err = mlx5e_alloc_drop_cq(mdev, cq, &cq_param); if (err) @@ -2672,6 +2803,10 @@ static int mlx5e_open_drop_rq(struct mlx5_core_dev *mdev, if (err) goto err_free_rq; + err = mlx5e_modify_rq_state(drop_rq, MLX5_RQC_STATE_RST, MLX5_RQC_STATE_RDY); + if (err) + mlx5_core_warn(priv->mdev, "modify_rq_state failed, rx_if_down_packets won't be counted %d\n", err); + return 0; err_free_rq: @@ -3097,20 +3232,28 @@ typedef int (*mlx5e_feature_handler)(struct net_device *netdev, bool enable); static int set_feature_lro(struct net_device *netdev, bool enable) { struct mlx5e_priv *priv = netdev_priv(netdev); + struct mlx5_core_dev *mdev = priv->mdev; struct mlx5e_channels new_channels = {}; + struct mlx5e_params *old_params; int err = 0; bool reset; mutex_lock(&priv->state_lock); - reset = (priv->channels.params.rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST); - reset = reset && test_bit(MLX5E_STATE_OPENED, &priv->state); + old_params = &priv->channels.params; + reset = test_bit(MLX5E_STATE_OPENED, &priv->state); - new_channels.params = priv->channels.params; + new_channels.params = *old_params; new_channels.params.lro_en = enable; + if (old_params->rq_wq_type != MLX5_WQ_TYPE_LINKED_LIST) { + if (mlx5e_rx_mpwqe_is_linear_skb(mdev, old_params) == + mlx5e_rx_mpwqe_is_linear_skb(mdev, &new_channels.params)) + reset = false; + } + if (!reset) { - priv->channels.params = new_channels.params; + *old_params = new_channels.params; err = mlx5e_modify_tirs_lro(priv); goto out; } @@ -3239,24 +3382,20 @@ static int mlx5e_set_features(struct net_device *netdev, netdev_features_t features) { netdev_features_t oper_features = netdev->features; - int err; + int err = 0; + +#define MLX5E_HANDLE_FEATURE(feature, handler) \ + mlx5e_handle_feature(netdev, &oper_features, features, feature, handler) - err = mlx5e_handle_feature(netdev, &oper_features, features, - NETIF_F_LRO, set_feature_lro); - err |= mlx5e_handle_feature(netdev, &oper_features, features, - NETIF_F_HW_VLAN_CTAG_FILTER, + err |= MLX5E_HANDLE_FEATURE(NETIF_F_LRO, set_feature_lro); + err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_VLAN_CTAG_FILTER, set_feature_cvlan_filter); - err |= mlx5e_handle_feature(netdev, &oper_features, features, - NETIF_F_HW_TC, set_feature_tc_num_filters); - err |= mlx5e_handle_feature(netdev, &oper_features, features, - NETIF_F_RXALL, set_feature_rx_all); - err |= mlx5e_handle_feature(netdev, &oper_features, features, - NETIF_F_RXFCS, set_feature_rx_fcs); - err |= mlx5e_handle_feature(netdev, &oper_features, features, - NETIF_F_HW_VLAN_CTAG_RX, set_feature_rx_vlan); + err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_TC, set_feature_tc_num_filters); + err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXALL, set_feature_rx_all); + err |= MLX5E_HANDLE_FEATURE(NETIF_F_RXFCS, set_feature_rx_fcs); + err |= MLX5E_HANDLE_FEATURE(NETIF_F_HW_VLAN_CTAG_RX, set_feature_rx_vlan); #ifdef CONFIG_RFS_ACCEL - err |= mlx5e_handle_feature(netdev, &oper_features, features, - NETIF_F_NTUPLE, set_feature_arfs); + err |= MLX5E_HANDLE_FEATURE(NETIF_F_NTUPLE, set_feature_arfs); #endif if (err) { @@ -3290,34 +3429,40 @@ static int mlx5e_change_mtu(struct net_device *netdev, int new_mtu) { struct mlx5e_priv *priv = netdev_priv(netdev); struct mlx5e_channels new_channels = {}; - int curr_mtu; + struct mlx5e_params *params; int err = 0; bool reset; mutex_lock(&priv->state_lock); - reset = !priv->channels.params.lro_en && - (priv->channels.params.rq_wq_type != - MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ); + params = &priv->channels.params; + reset = !params->lro_en; reset = reset && test_bit(MLX5E_STATE_OPENED, &priv->state); - curr_mtu = netdev->mtu; - netdev->mtu = new_mtu; + new_channels.params = *params; + new_channels.params.sw_mtu = new_mtu; + + if (params->rq_wq_type != MLX5_WQ_TYPE_LINKED_LIST) { + u8 ppw_old = mlx5e_mpwqe_log_pkts_per_wqe(params); + u8 ppw_new = mlx5e_mpwqe_log_pkts_per_wqe(&new_channels.params); + + reset = reset && (ppw_old != ppw_new); + } if (!reset) { + params->sw_mtu = new_mtu; mlx5e_set_dev_port_mtu(priv); + netdev->mtu = params->sw_mtu; goto out; } - new_channels.params = priv->channels.params; err = mlx5e_open_channels(priv, &new_channels); - if (err) { - netdev->mtu = curr_mtu; + if (err) goto out; - } mlx5e_switch_priv_channels(priv, &new_channels, mlx5e_set_dev_port_mtu); + netdev->mtu = new_channels.params.sw_mtu; out: mutex_unlock(&priv->state_lock); @@ -3607,21 +3752,11 @@ static netdev_features_t mlx5e_features_check(struct sk_buff *skb, static bool mlx5e_tx_timeout_eq_recover(struct net_device *dev, struct mlx5e_txqsq *sq) { - struct mlx5e_priv *priv = netdev_priv(dev); - struct mlx5_core_dev *mdev = priv->mdev; - int irqn_not_used, eqn; - struct mlx5_eq *eq; + struct mlx5_eq *eq = sq->cq.mcq.eq; u32 eqe_count; - if (mlx5_vector2eqn(mdev, sq->cq.mcq.vector, &eqn, &irqn_not_used)) - return false; - - eq = mlx5_eqn2eq(mdev, eqn); - if (IS_ERR(eq)) - return false; - netdev_err(dev, "EQ 0x%x: Cons = 0x%x, irqn = 0x%x\n", - eqn, eq->cons_index, eq->irqn); + eq->eqn, eq->cons_index, eq->irqn); eqe_count = mlx5_eq_poll_irq_disabled(eq); if (!eqe_count) @@ -3632,13 +3767,19 @@ static bool mlx5e_tx_timeout_eq_recover(struct net_device *dev, return true; } -static void mlx5e_tx_timeout(struct net_device *dev) +static void mlx5e_tx_timeout_work(struct work_struct *work) { - struct mlx5e_priv *priv = netdev_priv(dev); + struct mlx5e_priv *priv = container_of(work, struct mlx5e_priv, + tx_timeout_work); + struct net_device *dev = priv->netdev; bool reopen_channels = false; - int i; + int i, err; - netdev_err(dev, "TX timeout detected\n"); + rtnl_lock(); + mutex_lock(&priv->state_lock); + + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) + goto unlock; for (i = 0; i < priv->channels.num * priv->channels.params.num_tc; i++) { struct netdev_queue *dev_queue = netdev_get_tx_queue(dev, i); @@ -3646,7 +3787,9 @@ static void mlx5e_tx_timeout(struct net_device *dev) if (!netif_xmit_stopped(dev_queue)) continue; - netdev_err(dev, "TX timeout on queue: %d, SQ: 0x%x, CQ: 0x%x, SQ Cons: 0x%x SQ Prod: 0x%x, usecs since last trans: %u\n", + + netdev_err(dev, + "TX timeout on queue: %d, SQ: 0x%x, CQ: 0x%x, SQ Cons: 0x%x SQ Prod: 0x%x, usecs since last trans: %u\n", i, sq->sqn, sq->cq.mcq.cqn, sq->cc, sq->pc, jiffies_to_usecs(jiffies - dev_queue->trans_start)); @@ -3659,8 +3802,27 @@ static void mlx5e_tx_timeout(struct net_device *dev) } } - if (reopen_channels && test_bit(MLX5E_STATE_OPENED, &priv->state)) - schedule_work(&priv->tx_timeout_work); + if (!reopen_channels) + goto unlock; + + mlx5e_close_locked(dev); + err = mlx5e_open_locked(dev); + if (err) + netdev_err(priv->netdev, + "mlx5e_open_locked failed recovering from a tx_timeout, err(%d).\n", + err); + +unlock: + mutex_unlock(&priv->state_lock); + rtnl_unlock(); +} + +static void mlx5e_tx_timeout(struct net_device *dev) +{ + struct mlx5e_priv *priv = netdev_priv(dev); + + netdev_err(dev, "TX timeout detected\n"); + queue_work(priv->wq, &priv->tx_timeout_work); } static int mlx5e_xdp_set(struct net_device *netdev, struct bpf_prog *prog) @@ -3710,7 +3872,7 @@ static int mlx5e_xdp_set(struct net_device *netdev, struct bpf_prog *prog) bpf_prog_put(old_prog); if (reset) /* change RQ type according to priv->xdp_prog */ - mlx5e_set_rq_params(priv->mdev, &priv->channels.params); + mlx5e_set_rq_type(priv->mdev, &priv->channels.params); if (was_opened && reset) mlx5e_open_locked(netdev); @@ -3855,15 +4017,6 @@ static int mlx5e_check_required_hca_cap(struct mlx5_core_dev *mdev) return 0; } -u16 mlx5e_get_max_inline_cap(struct mlx5_core_dev *mdev) -{ - int bf_buf_size = (1 << MLX5_CAP_GEN(mdev, log_bf_reg_size)) / 2; - - return bf_buf_size - - sizeof(struct mlx5e_tx_wqe) + - 2 /*sizeof(mlx5e_tx_wqe.inline_hdr_start)*/; -} - void mlx5e_build_default_indir_rqt(u32 *indirection_rqt, int len, int num_channels) { @@ -3903,16 +4056,20 @@ static int mlx5e_get_pci_bw(struct mlx5_core_dev *mdev, u32 *pci_bw) return 0; } -static bool cqe_compress_heuristic(u32 link_speed, u32 pci_bw) +static bool slow_pci_heuristic(struct mlx5_core_dev *mdev) { - return (link_speed && pci_bw && - (pci_bw < 40000) && (pci_bw < link_speed)); -} + u32 link_speed = 0; + u32 pci_bw = 0; -static bool hw_lro_heuristic(u32 link_speed, u32 pci_bw) -{ - return !(link_speed && pci_bw && - (pci_bw <= 16000) && (pci_bw < link_speed)); + mlx5e_get_max_linkspeed(mdev, &link_speed); + mlx5e_get_pci_bw(mdev, &pci_bw); + mlx5_core_dbg_once(mdev, "Max link speed = %d, PCI BW = %d\n", + link_speed, pci_bw); + +#define MLX5E_SLOW_PCI_RATIO (2) + + return link_speed && pci_bw && + link_speed > MLX5E_SLOW_PCI_RATIO * pci_bw; } void mlx5e_set_tx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) @@ -3964,7 +4121,7 @@ void mlx5e_set_rx_cq_mode_params(struct mlx5e_params *params, u8 cq_period_mode) MLX5_CQ_PERIOD_MODE_START_FROM_CQE); } -u32 mlx5e_choose_lro_timeout(struct mlx5_core_dev *mdev, u32 wanted_timeout) +static u32 mlx5e_choose_lro_timeout(struct mlx5_core_dev *mdev, u32 wanted_timeout) { int i; @@ -3978,20 +4135,15 @@ u32 mlx5e_choose_lro_timeout(struct mlx5_core_dev *mdev, u32 wanted_timeout) void mlx5e_build_nic_params(struct mlx5_core_dev *mdev, struct mlx5e_params *params, - u16 max_channels) + u16 max_channels, u16 mtu) { - u8 cq_period_mode = 0; - u32 link_speed = 0; - u32 pci_bw = 0; + u8 rx_cq_period_mode; + params->sw_mtu = mtu; + params->hard_mtu = MLX5E_ETH_HARD_MTU; params->num_channels = max_channels; params->num_tc = 1; - mlx5e_get_max_linkspeed(mdev, &link_speed); - mlx5e_get_pci_bw(mdev, &pci_bw); - mlx5_core_dbg(mdev, "Max link speed = %d, PCI BW = %d\n", - link_speed, pci_bw); - /* SQ */ params->log_sq_size = is_kdump_kernel() ? MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE : @@ -4001,30 +4153,34 @@ void mlx5e_build_nic_params(struct mlx5_core_dev *mdev, params->rx_cqe_compress_def = false; if (MLX5_CAP_GEN(mdev, cqe_compression) && MLX5_CAP_GEN(mdev, vport_group_manager)) - params->rx_cqe_compress_def = cqe_compress_heuristic(link_speed, pci_bw); + params->rx_cqe_compress_def = slow_pci_heuristic(mdev); MLX5E_SET_PFLAG(params, MLX5E_PFLAG_RX_CQE_COMPRESS, params->rx_cqe_compress_def); /* RQ */ - mlx5e_set_rq_params(mdev, params); + if (mlx5e_striding_rq_possible(mdev, params)) + MLX5E_SET_PFLAG(params, MLX5E_PFLAG_RX_STRIDING_RQ, + !slow_pci_heuristic(mdev)); + mlx5e_set_rq_type(mdev, params); + mlx5e_init_rq_type_params(mdev, params); /* HW LRO */ /* TODO: && MLX5_CAP_ETH(mdev, lro_cap) */ if (params->rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) - params->lro_en = hw_lro_heuristic(link_speed, pci_bw); + if (!mlx5e_rx_mpwqe_is_linear_skb(mdev, params)) + params->lro_en = !slow_pci_heuristic(mdev); params->lro_timeout = mlx5e_choose_lro_timeout(mdev, MLX5E_DEFAULT_LRO_TIMEOUT); /* CQ moderation params */ - cq_period_mode = MLX5_CAP_GEN(mdev, cq_period_start_from_cqe) ? + rx_cq_period_mode = MLX5_CAP_GEN(mdev, cq_period_start_from_cqe) ? MLX5_CQ_PERIOD_MODE_START_FROM_CQE : MLX5_CQ_PERIOD_MODE_START_FROM_EQE; params->rx_dim_enabled = MLX5_CAP_GEN(mdev, cq_moderation); - mlx5e_set_rx_cq_mode_params(params, cq_period_mode); - mlx5e_set_tx_cq_mode_params(params, cq_period_mode); + mlx5e_set_rx_cq_mode_params(params, rx_cq_period_mode); + mlx5e_set_tx_cq_mode_params(params, MLX5_CQ_PERIOD_MODE_START_FROM_EQE); /* TX inline */ - params->tx_max_inline = mlx5e_get_max_inline_cap(mdev); params->tx_min_inline_mode = mlx5e_params_calculate_tx_min_inline(mdev); /* RSS */ @@ -4046,9 +4202,9 @@ static void mlx5e_build_nic_netdev_priv(struct mlx5_core_dev *mdev, priv->profile = profile; priv->ppriv = ppriv; priv->msglevel = MLX5E_MSG_LEVEL; - priv->hard_mtu = MLX5E_ETH_HARD_MTU; - mlx5e_build_nic_params(mdev, &priv->channels.params, profile->max_nch(mdev)); + mlx5e_build_nic_params(mdev, &priv->channels.params, + profile->max_nch(mdev), netdev->mtu); mutex_init(&priv->state_lock); @@ -4107,6 +4263,9 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) netdev->vlan_features |= NETIF_F_RXCSUM; netdev->vlan_features |= NETIF_F_RXHASH; + netdev->hw_enc_features |= NETIF_F_HW_VLAN_CTAG_TX; + netdev->hw_enc_features |= NETIF_F_HW_VLAN_CTAG_RX; + if (!!MLX5_CAP_ETH(mdev, lro_cap)) netdev->vlan_features |= NETIF_F_LRO; @@ -4186,7 +4345,7 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev) mlx5e_ipsec_build_netdev(priv); } -static void mlx5e_create_q_counter(struct mlx5e_priv *priv) +static void mlx5e_create_q_counters(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; int err; @@ -4196,14 +4355,21 @@ static void mlx5e_create_q_counter(struct mlx5e_priv *priv) mlx5_core_warn(mdev, "alloc queue counter failed, %d\n", err); priv->q_counter = 0; } + + err = mlx5_core_alloc_q_counter(mdev, &priv->drop_rq_q_counter); + if (err) { + mlx5_core_warn(mdev, "alloc drop RQ counter failed, %d\n", err); + priv->drop_rq_q_counter = 0; + } } -static void mlx5e_destroy_q_counter(struct mlx5e_priv *priv) +static void mlx5e_destroy_q_counters(struct mlx5e_priv *priv) { - if (!priv->q_counter) - return; + if (priv->q_counter) + mlx5_core_dealloc_q_counter(priv->mdev, priv->q_counter); - mlx5_core_dealloc_q_counter(priv->mdev, priv->q_counter); + if (priv->drop_rq_q_counter) + mlx5_core_dealloc_q_counter(priv->mdev, priv->drop_rq_q_counter); } static void mlx5e_nic_init(struct mlx5_core_dev *mdev, @@ -4315,7 +4481,7 @@ static void mlx5e_nic_enable(struct mlx5e_priv *priv) /* MTU range: 68 - hw-specific max */ netdev->min_mtu = ETH_MIN_MTU; mlx5_query_port_max_mtu(priv->mdev, &max_mtu, 1); - netdev->max_mtu = MLX5E_HW2SW_MTU(priv, max_mtu); + netdev->max_mtu = MLX5E_HW2SW_MTU(&priv->channels.params, max_mtu); mlx5e_set_dev_port_mtu(priv); mlx5_lag_add(mdev, netdev); @@ -4436,18 +4602,18 @@ int mlx5e_attach_netdev(struct mlx5e_priv *priv) if (err) goto out; - err = mlx5e_open_drop_rq(mdev, &priv->drop_rq); + mlx5e_create_q_counters(priv); + + err = mlx5e_open_drop_rq(priv, &priv->drop_rq); if (err) { mlx5_core_err(mdev, "open drop rq failed, %d\n", err); - goto err_cleanup_tx; + goto err_destroy_q_counters; } err = profile->init_rx(priv); if (err) goto err_close_drop_rq; - mlx5e_create_q_counter(priv); - if (profile->enable) profile->enable(priv); @@ -4456,7 +4622,8 @@ int mlx5e_attach_netdev(struct mlx5e_priv *priv) err_close_drop_rq: mlx5e_close_drop_rq(&priv->drop_rq); -err_cleanup_tx: +err_destroy_q_counters: + mlx5e_destroy_q_counters(priv); profile->cleanup_tx(priv); out: @@ -4473,9 +4640,9 @@ void mlx5e_detach_netdev(struct mlx5e_priv *priv) profile->disable(priv); flush_workqueue(priv->wq); - mlx5e_destroy_q_counter(priv); profile->cleanup_rx(priv); mlx5e_close_drop_rq(&priv->drop_rq); + mlx5e_destroy_q_counters(priv); profile->cleanup_tx(priv); cancel_delayed_work_sync(&priv->update_stats_work); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c index 500d817d2b0a..d8f68e4d1018 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rep.c @@ -883,14 +883,14 @@ static void mlx5e_build_rep_params(struct mlx5_core_dev *mdev, MLX5_CQ_PERIOD_MODE_START_FROM_CQE : MLX5_CQ_PERIOD_MODE_START_FROM_EQE; + params->hard_mtu = MLX5E_ETH_HARD_MTU; params->log_sq_size = MLX5E_REP_PARAMS_LOG_SQ_SIZE; params->rq_wq_type = MLX5_WQ_TYPE_LINKED_LIST; - params->log_rq_size = MLX5E_REP_PARAMS_LOG_RQ_SIZE; + params->log_rq_mtu_frames = MLX5E_REP_PARAMS_LOG_RQ_SIZE; params->rx_dim_enabled = MLX5_CAP_GEN(mdev, cq_moderation); mlx5e_set_rx_cq_mode_params(params, cq_period_mode); - params->tx_max_inline = mlx5e_get_max_inline_cap(mdev); params->num_tc = 1; params->lro_wqe_sz = MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ; @@ -931,8 +931,6 @@ static void mlx5e_init_rep(struct mlx5_core_dev *mdev, priv->channels.params.num_channels = profile->max_nch(mdev); - priv->hard_mtu = MLX5E_ETH_HARD_MTU; - mlx5e_build_rep_params(mdev, &priv->channels.params); mlx5e_build_rep_netdev(netdev); @@ -1160,6 +1158,15 @@ mlx5e_vport_rep_unload(struct mlx5_eswitch_rep *rep) kfree(ppriv); /* mlx5e_rep_priv */ } +static void *mlx5e_vport_rep_get_proto_dev(struct mlx5_eswitch_rep *rep) +{ + struct mlx5e_rep_priv *rpriv; + + rpriv = mlx5e_rep_to_rep_priv(rep); + + return rpriv->netdev; +} + static void mlx5e_rep_register_vf_vports(struct mlx5e_priv *priv) { struct mlx5_core_dev *mdev = priv->mdev; @@ -1172,6 +1179,7 @@ static void mlx5e_rep_register_vf_vports(struct mlx5e_priv *priv) rep_if.load = mlx5e_vport_rep_load; rep_if.unload = mlx5e_vport_rep_unload; + rep_if.get_proto_dev = mlx5e_vport_rep_get_proto_dev; mlx5_eswitch_register_vport_rep(esw, vport, &rep_if, REP_ETH); } } @@ -1199,6 +1207,7 @@ void mlx5e_register_vport_reps(struct mlx5e_priv *priv) rep_if.load = mlx5e_nic_rep_load; rep_if.unload = mlx5e_nic_rep_unload; + rep_if.get_proto_dev = mlx5e_vport_rep_get_proto_dev; rep_if.priv = rpriv; INIT_LIST_HEAD(&rpriv->vport_sqs_list); mlx5_eswitch_register_vport_rep(esw, 0, &rep_if, REP_ETH); /* UPLINK PF vport*/ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c index e5c3ab46a24a..176645762e49 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c @@ -53,7 +53,7 @@ static inline bool mlx5e_rx_hw_stamp(struct hwtstamp_config *config) static inline void mlx5e_read_cqe_slot(struct mlx5e_cq *cq, u32 cqcc, void *data) { - u32 ci = cqcc & cq->wq.sz_m1; + u32 ci = cqcc & cq->wq.fbc.sz_m1; memcpy(data, mlx5_cqwq_get_wqe(&cq->wq, ci), sizeof(struct mlx5_cqe64)); } @@ -75,9 +75,10 @@ static inline void mlx5e_read_mini_arr_slot(struct mlx5e_cq *cq, u32 cqcc) static inline void mlx5e_cqes_update_owner(struct mlx5e_cq *cq, u32 cqcc, int n) { - u8 op_own = (cqcc >> cq->wq.log_sz) & 1; - u32 wq_sz = 1 << cq->wq.log_sz; - u32 ci = cqcc & cq->wq.sz_m1; + struct mlx5_frag_buf_ctrl *fbc = &cq->wq.fbc; + u8 op_own = (cqcc >> fbc->log_sz) & 1; + u32 wq_sz = 1 << fbc->log_sz; + u32 ci = cqcc & fbc->sz_m1; u32 ci_top = min_t(u32, wq_sz, ci + n); for (; ci < ci_top; ci++, n--) { @@ -102,7 +103,7 @@ static inline void mlx5e_decompress_cqe(struct mlx5e_rq *rq, cq->title.byte_cnt = cq->mini_arr[cq->mini_arr_idx].byte_cnt; cq->title.check_sum = cq->mini_arr[cq->mini_arr_idx].checksum; cq->title.op_own &= 0xf0; - cq->title.op_own |= 0x01 & (cqcc >> cq->wq.log_sz); + cq->title.op_own |= 0x01 & (cqcc >> cq->wq.fbc.log_sz); cq->title.wqe_counter = cpu_to_be16(cq->decmprs_wqe_counter); if (rq->wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ) @@ -295,46 +296,36 @@ void mlx5e_dealloc_rx_wqe(struct mlx5e_rq *rq, u16 ix) mlx5e_free_rx_wqe(rq, wi); } -static inline int mlx5e_mpwqe_strides_per_page(struct mlx5e_rq *rq) -{ - return rq->mpwqe.num_strides >> MLX5_MPWRQ_WQE_PAGE_ORDER; -} - static inline void mlx5e_add_skb_frag_mpwqe(struct mlx5e_rq *rq, struct sk_buff *skb, - struct mlx5e_mpw_info *wi, - u32 page_idx, u32 frag_offset, - u32 len) + struct mlx5e_dma_info *di, + u32 frag_offset, u32 len) { unsigned int truesize = ALIGN(len, BIT(rq->mpwqe.log_stride_sz)); dma_sync_single_for_cpu(rq->pdev, - wi->umr.dma_info[page_idx].addr + frag_offset, + di->addr + frag_offset, len, DMA_FROM_DEVICE); - wi->skbs_frags[page_idx]++; + page_ref_inc(di->page); skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, - wi->umr.dma_info[page_idx].page, frag_offset, - len, truesize); + di->page, frag_offset, len, truesize); } static inline void mlx5e_copy_skb_header_mpwqe(struct device *pdev, struct sk_buff *skb, - struct mlx5e_mpw_info *wi, - u32 page_idx, u32 offset, - u32 headlen) + struct mlx5e_dma_info *dma_info, + u32 offset, u32 headlen) { u16 headlen_pg = min_t(u32, headlen, PAGE_SIZE - offset); - struct mlx5e_dma_info *dma_info = &wi->umr.dma_info[page_idx]; unsigned int len; /* Aligning len to sizeof(long) optimizes memcpy performance */ len = ALIGN(headlen_pg, sizeof(long)); dma_sync_single_for_cpu(pdev, dma_info->addr + offset, len, DMA_FROM_DEVICE); - skb_copy_to_linear_data_offset(skb, 0, - page_address(dma_info->page) + offset, - len); + skb_copy_to_linear_data(skb, page_address(dma_info->page) + offset, len); + if (unlikely(offset + headlen > PAGE_SIZE)) { dma_info++; headlen_pg = len; @@ -347,14 +338,49 @@ mlx5e_copy_skb_header_mpwqe(struct device *pdev, } } -static inline void mlx5e_post_umr_wqe(struct mlx5e_rq *rq, u16 ix) +void mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi) +{ + const bool no_xdp_xmit = + bitmap_empty(wi->xdp_xmit_bitmap, MLX5_MPWRQ_PAGES_PER_WQE); + struct mlx5e_dma_info *dma_info = wi->umr.dma_info; + int i; + + for (i = 0; i < MLX5_MPWRQ_PAGES_PER_WQE; i++) + if (no_xdp_xmit || !test_bit(i, wi->xdp_xmit_bitmap)) + mlx5e_page_release(rq, &dma_info[i], true); +} + +static void mlx5e_post_rx_mpwqe(struct mlx5e_rq *rq) +{ + struct mlx5_wq_ll *wq = &rq->wq; + struct mlx5e_rx_wqe *wqe = mlx5_wq_ll_get_wqe(wq, wq->head); + + rq->mpwqe.umr_in_progress = false; + + mlx5_wq_ll_push(wq, be16_to_cpu(wqe->next.next_wqe_index)); + + /* ensure wqes are visible to device before updating doorbell record */ + dma_wmb(); + + mlx5_wq_ll_update_db_record(wq); +} + +static inline u16 mlx5e_icosq_wrap_cnt(struct mlx5e_icosq *sq) +{ + return sq->pc >> MLX5E_PARAMS_MINIMUM_LOG_SQ_SIZE; +} + +static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix) { struct mlx5e_mpw_info *wi = &rq->mpwqe.info[ix]; + struct mlx5e_dma_info *dma_info = &wi->umr.dma_info[0]; struct mlx5e_icosq *sq = &rq->channel->icosq; struct mlx5_wq_cyc *wq = &sq->wq; - struct mlx5e_umr_wqe *wqe; - u8 num_wqebbs = DIV_ROUND_UP(sizeof(*wqe), MLX5_SEND_WQE_BB); + struct mlx5e_umr_wqe *umr_wqe; + u16 xlt_offset = ix << (MLX5E_LOG_ALIGNED_MPWQE_PPW - 1); + int err; u16 pi; + int i; /* fill sq edge with nops to avoid wqe wrap around */ while ((pi = (sq->pc & wq->sz_m1)) > sq->edge) { @@ -362,90 +388,44 @@ static inline void mlx5e_post_umr_wqe(struct mlx5e_rq *rq, u16 ix) mlx5e_post_nop(wq, sq->sqn, &sq->pc); } - wqe = mlx5_wq_cyc_get_wqe(wq, pi); - memcpy(wqe, &wi->umr.wqe, sizeof(*wqe)); - wqe->ctrl.opmod_idx_opcode = - cpu_to_be32((sq->pc << MLX5_WQE_CTRL_WQE_INDEX_SHIFT) | - MLX5_OPCODE_UMR); - - sq->db.ico_wqe[pi].opcode = MLX5_OPCODE_UMR; - sq->pc += num_wqebbs; - mlx5e_notify_hw(&sq->wq, sq->pc, sq->uar_map, &wqe->ctrl); -} - -static int mlx5e_alloc_rx_umr_mpwqe(struct mlx5e_rq *rq, - u16 ix) -{ - struct mlx5e_mpw_info *wi = &rq->mpwqe.info[ix]; - int pg_strides = mlx5e_mpwqe_strides_per_page(rq); - struct mlx5e_dma_info *dma_info = &wi->umr.dma_info[0]; - int err; - int i; + umr_wqe = mlx5_wq_cyc_get_wqe(wq, pi); + if (unlikely(mlx5e_icosq_wrap_cnt(sq) < 2)) + memcpy(umr_wqe, &rq->mpwqe.umr_wqe, + offsetof(struct mlx5e_umr_wqe, inline_mtts)); for (i = 0; i < MLX5_MPWRQ_PAGES_PER_WQE; i++, dma_info++) { err = mlx5e_page_alloc_mapped(rq, dma_info); if (unlikely(err)) goto err_unmap; - wi->umr.mtt[i] = cpu_to_be64(dma_info->addr | MLX5_EN_WR); - page_ref_add(dma_info->page, pg_strides); + umr_wqe->inline_mtts[i].ptag = cpu_to_be64(dma_info->addr | MLX5_EN_WR); } - memset(wi->skbs_frags, 0, sizeof(*wi->skbs_frags) * MLX5_MPWRQ_PAGES_PER_WQE); + bitmap_zero(wi->xdp_xmit_bitmap, MLX5_MPWRQ_PAGES_PER_WQE); wi->consumed_strides = 0; + rq->mpwqe.umr_in_progress = true; + + umr_wqe->ctrl.opmod_idx_opcode = + cpu_to_be32((sq->pc << MLX5_WQE_CTRL_WQE_INDEX_SHIFT) | + MLX5_OPCODE_UMR); + umr_wqe->uctrl.xlt_offset = cpu_to_be16(xlt_offset); + + sq->db.ico_wqe[pi].opcode = MLX5_OPCODE_UMR; + sq->pc += MLX5E_UMR_WQEBBS; + mlx5e_notify_hw(&sq->wq, sq->pc, sq->uar_map, &umr_wqe->ctrl); + return 0; err_unmap: while (--i >= 0) { dma_info--; - page_ref_sub(dma_info->page, pg_strides); mlx5e_page_release(rq, dma_info, true); } + rq->stats.buff_alloc_err++; return err; } -void mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi) -{ - int pg_strides = mlx5e_mpwqe_strides_per_page(rq); - struct mlx5e_dma_info *dma_info = &wi->umr.dma_info[0]; - int i; - - for (i = 0; i < MLX5_MPWRQ_PAGES_PER_WQE; i++, dma_info++) { - page_ref_sub(dma_info->page, pg_strides - wi->skbs_frags[i]); - mlx5e_page_release(rq, dma_info, true); - } -} - -static void mlx5e_post_rx_mpwqe(struct mlx5e_rq *rq) -{ - struct mlx5_wq_ll *wq = &rq->wq; - struct mlx5e_rx_wqe *wqe = mlx5_wq_ll_get_wqe(wq, wq->head); - - rq->mpwqe.umr_in_progress = false; - - mlx5_wq_ll_push(wq, be16_to_cpu(wqe->next.next_wqe_index)); - - /* ensure wqes are visible to device before updating doorbell record */ - dma_wmb(); - - mlx5_wq_ll_update_db_record(wq); -} - -static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix) -{ - int err; - - err = mlx5e_alloc_rx_umr_mpwqe(rq, ix); - if (unlikely(err)) { - rq->stats.buff_alloc_err++; - return err; - } - rq->mpwqe.umr_in_progress = true; - mlx5e_post_umr_wqe(rq, ix); - return 0; -} - void mlx5e_dealloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix) { struct mlx5e_mpw_info *wi = &rq->mpwqe.info[ix]; @@ -544,7 +524,7 @@ bool mlx5e_post_rx_mpwqes(struct mlx5e_rq *rq) if (!rq->mpwqe.umr_in_progress) mlx5e_alloc_rx_mpwqe(rq, wq->head); - return true; + return false; } static void mlx5e_lro_update_tcp_hdr(struct mlx5_cqe64 *cqe, struct tcphdr *tcp) @@ -766,8 +746,7 @@ static inline bool mlx5e_xmit_xdp_frame(struct mlx5e_rq *rq, prefetchw(wqe); - if (unlikely(dma_len < MLX5E_XDP_MIN_INLINE || - MLX5E_SW2HW_MTU(rq->channel->priv, rq->netdev->mtu) < dma_len)) { + if (unlikely(dma_len < MLX5E_XDP_MIN_INLINE || rq->hw_mtu < dma_len)) { rq->stats.xdp_drop++; return false; } @@ -806,7 +785,7 @@ static inline bool mlx5e_xmit_xdp_frame(struct mlx5e_rq *rq, /* move page to reference to sq responsibility, * and mark so it's not put back in page-cache. */ - rq->wqe.xdp_xmit = true; + __set_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags); /* non-atomic */ sq->db.di[pi] = *di; sq->pc++; @@ -855,6 +834,24 @@ static inline int mlx5e_xdp_handle(struct mlx5e_rq *rq, } static inline +struct sk_buff *mlx5e_build_linear_skb(struct mlx5e_rq *rq, void *va, + u32 frag_size, u16 headroom, + u32 cqe_bcnt) +{ + struct sk_buff *skb = build_skb(va, frag_size); + + if (unlikely(!skb)) { + rq->stats.buff_alloc_err++; + return NULL; + } + + skb_reserve(skb, headroom); + skb_put(skb, cqe_bcnt); + + return skb; +} + +static inline struct sk_buff *skb_from_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe, struct mlx5e_wqe_frag_info *wi, u32 cqe_bcnt) { @@ -869,10 +866,8 @@ struct sk_buff *skb_from_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe, data = va + rx_headroom; frag_size = MLX5_SKB_FRAG_SZ(rx_headroom + cqe_bcnt); - dma_sync_single_range_for_cpu(rq->pdev, - di->addr + wi->offset, - 0, frag_size, - DMA_FROM_DEVICE); + dma_sync_single_range_for_cpu(rq->pdev, di->addr, wi->offset, + frag_size, DMA_FROM_DEVICE); prefetch(data); wi->offset += frag_size; @@ -887,18 +882,13 @@ struct sk_buff *skb_from_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe, if (consumed) return NULL; /* page/packet was consumed by XDP */ - skb = build_skb(va, frag_size); - if (unlikely(!skb)) { - rq->stats.buff_alloc_err++; + skb = mlx5e_build_linear_skb(rq, va, frag_size, rx_headroom, cqe_bcnt); + if (unlikely(!skb)) return NULL; - } /* queue up for recycling/reuse */ page_ref_inc(di->page); - skb_reserve(skb, rx_headroom); - skb_put(skb, cqe_bcnt); - return skb; } @@ -920,9 +910,8 @@ void mlx5e_handle_rx_cqe(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe) skb = skb_from_cqe(rq, cqe, wi, cqe_bcnt); if (!skb) { /* probably for XDP */ - if (rq->wqe.xdp_xmit) { + if (__test_and_clear_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags)) { wi->di.page = NULL; - rq->wqe.xdp_xmit = false; /* do not return page to cache, it will be returned on XDP_TX completion */ goto wq_ll_pop; } @@ -962,9 +951,8 @@ void mlx5e_handle_rx_cqe_rep(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe) skb = skb_from_cqe(rq, cqe, wi, cqe_bcnt); if (!skb) { - if (rq->wqe.xdp_xmit) { + if (__test_and_clear_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags)) { wi->di.page = NULL; - rq->wqe.xdp_xmit = false; /* do not return page to cache, it will be returned on XDP_TX completion */ goto wq_ll_pop; } @@ -987,23 +975,28 @@ wq_ll_pop: } #endif -static inline void mlx5e_mpwqe_fill_rx_skb(struct mlx5e_rq *rq, - struct mlx5_cqe64 *cqe, - struct mlx5e_mpw_info *wi, - u32 cqe_bcnt, - struct sk_buff *skb) +struct sk_buff * +mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, + u16 cqe_bcnt, u32 head_offset, u32 page_idx) { - u16 stride_ix = mpwrq_get_cqe_stride_index(cqe); - u32 wqe_offset = stride_ix << rq->mpwqe.log_stride_sz; - u32 head_offset = wqe_offset & (PAGE_SIZE - 1); - u32 page_idx = wqe_offset >> PAGE_SHIFT; - u32 head_page_idx = page_idx; u16 headlen = min_t(u16, MLX5_MPWRQ_SMALL_PACKET_THRESHOLD, cqe_bcnt); + struct mlx5e_dma_info *di = &wi->umr.dma_info[page_idx]; u32 frag_offset = head_offset + headlen; - u16 byte_cnt = cqe_bcnt - headlen; + u32 byte_cnt = cqe_bcnt - headlen; + struct mlx5e_dma_info *head_di = di; + struct sk_buff *skb; + + skb = napi_alloc_skb(rq->cq.napi, + ALIGN(MLX5_MPWRQ_SMALL_PACKET_THRESHOLD, sizeof(long))); + if (unlikely(!skb)) { + rq->stats.buff_alloc_err++; + return NULL; + } + + prefetchw(skb->data); if (unlikely(frag_offset >= PAGE_SIZE)) { - page_idx++; + di++; frag_offset -= PAGE_SIZE; } @@ -1011,18 +1004,59 @@ static inline void mlx5e_mpwqe_fill_rx_skb(struct mlx5e_rq *rq, u32 pg_consumed_bytes = min_t(u32, PAGE_SIZE - frag_offset, byte_cnt); - mlx5e_add_skb_frag_mpwqe(rq, skb, wi, page_idx, frag_offset, + mlx5e_add_skb_frag_mpwqe(rq, skb, di, frag_offset, pg_consumed_bytes); byte_cnt -= pg_consumed_bytes; frag_offset = 0; - page_idx++; + di++; } /* copy header */ - mlx5e_copy_skb_header_mpwqe(rq->pdev, skb, wi, head_page_idx, + mlx5e_copy_skb_header_mpwqe(rq->pdev, skb, head_di, head_offset, headlen); /* skb linear part was allocated with headlen and aligned to long */ skb->tail += headlen; skb->len += headlen; + + return skb; +} + +struct sk_buff * +mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, + u16 cqe_bcnt, u32 head_offset, u32 page_idx) +{ + struct mlx5e_dma_info *di = &wi->umr.dma_info[page_idx]; + u16 rx_headroom = rq->buff.headroom; + u32 cqe_bcnt32 = cqe_bcnt; + struct sk_buff *skb; + void *va, *data; + u32 frag_size; + bool consumed; + + va = page_address(di->page) + head_offset; + data = va + rx_headroom; + frag_size = MLX5_SKB_FRAG_SZ(rx_headroom + cqe_bcnt32); + + dma_sync_single_range_for_cpu(rq->pdev, di->addr, head_offset, + frag_size, DMA_FROM_DEVICE); + prefetch(data); + + rcu_read_lock(); + consumed = mlx5e_xdp_handle(rq, di, va, &rx_headroom, &cqe_bcnt32); + rcu_read_unlock(); + if (consumed) { + if (__test_and_clear_bit(MLX5E_RQ_FLAG_XDP_XMIT, rq->flags)) + __set_bit(page_idx, wi->xdp_xmit_bitmap); /* non-atomic */ + return NULL; /* page/packet was consumed by XDP */ + } + + skb = mlx5e_build_linear_skb(rq, va, frag_size, rx_headroom, cqe_bcnt32); + if (unlikely(!skb)) + return NULL; + + /* queue up for recycling/reuse */ + page_ref_inc(di->page); + + return skb; } void mlx5e_handle_rx_cqe_mpwrq(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe) @@ -1030,7 +1064,11 @@ void mlx5e_handle_rx_cqe_mpwrq(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe) u16 cstrides = mpwrq_get_cqe_consumed_strides(cqe); u16 wqe_id = be16_to_cpu(cqe->wqe_id); struct mlx5e_mpw_info *wi = &rq->mpwqe.info[wqe_id]; - struct mlx5e_rx_wqe *wqe = mlx5_wq_ll_get_wqe(&rq->wq, wqe_id); + u16 stride_ix = mpwrq_get_cqe_stride_index(cqe); + u32 wqe_offset = stride_ix << rq->mpwqe.log_stride_sz; + u32 head_offset = wqe_offset & (PAGE_SIZE - 1); + u32 page_idx = wqe_offset >> PAGE_SHIFT; + struct mlx5e_rx_wqe *wqe; struct sk_buff *skb; u16 cqe_bcnt; @@ -1046,18 +1084,13 @@ void mlx5e_handle_rx_cqe_mpwrq(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe) goto mpwrq_cqe_out; } - skb = napi_alloc_skb(rq->cq.napi, - ALIGN(MLX5_MPWRQ_SMALL_PACKET_THRESHOLD, - sizeof(long))); - if (unlikely(!skb)) { - rq->stats.buff_alloc_err++; - goto mpwrq_cqe_out; - } - - prefetchw(skb->data); cqe_bcnt = mpwrq_get_cqe_byte_cnt(cqe); - mlx5e_mpwqe_fill_rx_skb(rq, cqe, wi, cqe_bcnt, skb); + skb = rq->mpwqe.skb_from_cqe_mpwrq(rq, wi, cqe_bcnt, head_offset, + page_idx); + if (!skb) + goto mpwrq_cqe_out; + mlx5e_complete_rx_cqe(rq, cqe, cqe_bcnt, skb); napi_gro_receive(rq->cq.napi, skb); @@ -1065,6 +1098,7 @@ mpwrq_cqe_out: if (likely(wi->consumed_strides < rq->mpwqe.num_strides)) return; + wqe = mlx5_wq_ll_get_wqe(&rq->wq, wqe_id); mlx5e_free_rx_mpwqe(rq, wi); mlx5_wq_ll_pop(&rq->wq, cqe->wqe_id, &wqe->next.next_wqe_index); } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c index 5f0f3493d747..b08c94422907 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c @@ -60,6 +60,8 @@ static const struct counter_desc sw_stats_desc[] = { { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_queue_wake) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_queue_dropped) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_xmit_more) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_cqe_err) }, + { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_recover) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_wqe_err) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_mpwqe_filler) }, { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_buff_alloc_err) }, @@ -153,6 +155,8 @@ static void mlx5e_grp_sw_update_stats(struct mlx5e_priv *priv) s->tx_queue_stopped += sq_stats->stopped; s->tx_queue_wake += sq_stats->wake; s->tx_queue_dropped += sq_stats->dropped; + s->tx_cqe_err += sq_stats->cqe_err; + s->tx_recover += sq_stats->recover; s->tx_xmit_more += sq_stats->xmit_more; s->tx_csum_partial_inner += sq_stats->csum_partial_inner; s->tx_csum_none += sq_stats->csum_none; @@ -170,11 +174,24 @@ static const struct counter_desc q_stats_desc[] = { { MLX5E_DECLARE_STAT(struct mlx5e_qcounter_stats, rx_out_of_buffer) }, }; +static const struct counter_desc drop_rq_stats_desc[] = { + { MLX5E_DECLARE_STAT(struct mlx5e_qcounter_stats, rx_if_down_packets) }, +}; + #define NUM_Q_COUNTERS ARRAY_SIZE(q_stats_desc) +#define NUM_DROP_RQ_COUNTERS ARRAY_SIZE(drop_rq_stats_desc) static int mlx5e_grp_q_get_num_stats(struct mlx5e_priv *priv) { - return priv->q_counter ? NUM_Q_COUNTERS : 0; + int num_stats = 0; + + if (priv->q_counter) + num_stats += NUM_Q_COUNTERS; + + if (priv->drop_rq_q_counter) + num_stats += NUM_DROP_RQ_COUNTERS; + + return num_stats; } static int mlx5e_grp_q_fill_strings(struct mlx5e_priv *priv, u8 *data, int idx) @@ -182,7 +199,13 @@ static int mlx5e_grp_q_fill_strings(struct mlx5e_priv *priv, u8 *data, int idx) int i; for (i = 0; i < NUM_Q_COUNTERS && priv->q_counter; i++) - strcpy(data + (idx++) * ETH_GSTRING_LEN, q_stats_desc[i].format); + strcpy(data + (idx++) * ETH_GSTRING_LEN, + q_stats_desc[i].format); + + for (i = 0; i < NUM_DROP_RQ_COUNTERS && priv->drop_rq_q_counter; i++) + strcpy(data + (idx++) * ETH_GSTRING_LEN, + drop_rq_stats_desc[i].format); + return idx; } @@ -191,7 +214,11 @@ static int mlx5e_grp_q_fill_stats(struct mlx5e_priv *priv, u64 *data, int idx) int i; for (i = 0; i < NUM_Q_COUNTERS && priv->q_counter; i++) - data[idx++] = MLX5E_READ_CTR32_CPU(&priv->stats.qcnt, q_stats_desc, i); + data[idx++] = MLX5E_READ_CTR32_CPU(&priv->stats.qcnt, + q_stats_desc, i); + for (i = 0; i < NUM_DROP_RQ_COUNTERS && priv->drop_rq_q_counter; i++) + data[idx++] = MLX5E_READ_CTR32_CPU(&priv->stats.qcnt, + drop_rq_stats_desc, i); return idx; } @@ -199,16 +226,76 @@ static void mlx5e_grp_q_update_stats(struct mlx5e_priv *priv) { struct mlx5e_qcounter_stats *qcnt = &priv->stats.qcnt; u32 out[MLX5_ST_SZ_DW(query_q_counter_out)]; - int err; - if (!priv->q_counter) - return; + if (priv->q_counter && + !mlx5_core_query_q_counter(priv->mdev, priv->q_counter, 0, out, + sizeof(out))) + qcnt->rx_out_of_buffer = MLX5_GET(query_q_counter_out, + out, out_of_buffer); + if (priv->drop_rq_q_counter && + !mlx5_core_query_q_counter(priv->mdev, priv->drop_rq_q_counter, 0, + out, sizeof(out))) + qcnt->rx_if_down_packets = MLX5_GET(query_q_counter_out, out, + out_of_buffer); +} + +#define VNIC_ENV_OFF(c) MLX5_BYTE_OFF(query_vnic_env_out, c) +static const struct counter_desc vnic_env_stats_desc[] = { + { "rx_steer_missed_packets", + VNIC_ENV_OFF(vport_env.nic_receive_steering_discard) }, +}; + +#define NUM_VNIC_ENV_COUNTERS ARRAY_SIZE(vnic_env_stats_desc) + +static int mlx5e_grp_vnic_env_get_num_stats(struct mlx5e_priv *priv) +{ + return MLX5_CAP_GEN(priv->mdev, nic_receive_steering_discard) ? + NUM_VNIC_ENV_COUNTERS : 0; +} + +static int mlx5e_grp_vnic_env_fill_strings(struct mlx5e_priv *priv, u8 *data, + int idx) +{ + int i; + + if (!MLX5_CAP_GEN(priv->mdev, nic_receive_steering_discard)) + return idx; + + for (i = 0; i < NUM_VNIC_ENV_COUNTERS; i++) + strcpy(data + (idx++) * ETH_GSTRING_LEN, + vnic_env_stats_desc[i].format); + return idx; +} + +static int mlx5e_grp_vnic_env_fill_stats(struct mlx5e_priv *priv, u64 *data, + int idx) +{ + int i; + + if (!MLX5_CAP_GEN(priv->mdev, nic_receive_steering_discard)) + return idx; - err = mlx5_core_query_q_counter(priv->mdev, priv->q_counter, 0, out, sizeof(out)); - if (err) + for (i = 0; i < NUM_VNIC_ENV_COUNTERS; i++) + data[idx++] = MLX5E_READ_CTR64_BE(priv->stats.vnic.query_vnic_env_out, + vnic_env_stats_desc, i); + return idx; +} + +static void mlx5e_grp_vnic_env_update_stats(struct mlx5e_priv *priv) +{ + u32 *out = (u32 *)priv->stats.vnic.query_vnic_env_out; + int outlen = MLX5_ST_SZ_BYTES(query_vnic_env_out); + u32 in[MLX5_ST_SZ_DW(query_vnic_env_in)] = {0}; + struct mlx5_core_dev *mdev = priv->mdev; + + if (!MLX5_CAP_GEN(priv->mdev, nic_receive_steering_discard)) return; - qcnt->rx_out_of_buffer = MLX5_GET(query_q_counter_out, out, out_of_buffer); + MLX5_SET(query_vnic_env_in, in, opcode, + MLX5_CMD_OP_QUERY_VNIC_ENV); + MLX5_SET(query_vnic_env_in, in, op_mod, 0); + MLX5_SET(query_vnic_env_in, in, other_vport, 0); + mlx5_cmd_exec(mdev, in, sizeof(in), out, outlen); } #define VPORT_COUNTER_OFF(c) MLX5_BYTE_OFF(query_vport_counter_out, c) @@ -754,7 +841,15 @@ static const struct counter_desc pport_per_prio_pfc_stats_desc[] = { { "rx_%s_pause_transition", PPORT_PER_PRIO_OFF(rx_pause_transition) }, }; +static const struct counter_desc pport_pfc_stall_stats_desc[] = { + { "tx_pause_storm_warning_events ", PPORT_PER_PRIO_OFF(device_stall_minor_watermark_cnt) }, + { "tx_pause_storm_error_events", PPORT_PER_PRIO_OFF(device_stall_critical_watermark_cnt) }, +}; + #define NUM_PPORT_PER_PRIO_PFC_COUNTERS ARRAY_SIZE(pport_per_prio_pfc_stats_desc) +#define NUM_PPORT_PFC_STALL_COUNTERS(priv) (ARRAY_SIZE(pport_pfc_stall_stats_desc) * \ + MLX5_CAP_PCAM_FEATURE((priv)->mdev, pfcc_mask) * \ + MLX5_CAP_DEBUG((priv)->mdev, stall_detect)) static unsigned long mlx5e_query_pfc_combined(struct mlx5e_priv *priv) { @@ -790,7 +885,8 @@ static int mlx5e_grp_per_prio_pfc_get_num_stats(struct mlx5e_priv *priv) { return (mlx5e_query_global_pause_combined(priv) + hweight8(mlx5e_query_pfc_combined(priv))) * - NUM_PPORT_PER_PRIO_PFC_COUNTERS; + NUM_PPORT_PER_PRIO_PFC_COUNTERS + + NUM_PPORT_PFC_STALL_COUNTERS(priv); } static int mlx5e_grp_per_prio_pfc_fill_strings(struct mlx5e_priv *priv, @@ -818,6 +914,10 @@ static int mlx5e_grp_per_prio_pfc_fill_strings(struct mlx5e_priv *priv, } } + for (i = 0; i < NUM_PPORT_PFC_STALL_COUNTERS(priv); i++) + strcpy(data + (idx++) * ETH_GSTRING_LEN, + pport_pfc_stall_stats_desc[i].format); + return idx; } @@ -845,6 +945,10 @@ static int mlx5e_grp_per_prio_pfc_fill_stats(struct mlx5e_priv *priv, } } + for (i = 0; i < NUM_PPORT_PFC_STALL_COUNTERS(priv); i++) + data[idx++] = MLX5E_READ_CTR64_BE(&priv->stats.pport.per_prio_counters[0], + pport_pfc_stall_stats_desc, i); + return idx; } @@ -1003,6 +1107,8 @@ static const struct counter_desc sq_stats_desc[] = { { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, wake) }, { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, dropped) }, { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, xmit_more) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, cqe_err) }, + { MLX5E_DECLARE_TX_STAT(struct mlx5e_sq_stats, recover) }, }; static const struct counter_desc ch_stats_desc[] = { @@ -1095,6 +1201,12 @@ const struct mlx5e_stats_grp mlx5e_stats_grps[] = { .update_stats = mlx5e_grp_q_update_stats, }, { + .get_num_stats = mlx5e_grp_vnic_env_get_num_stats, + .fill_strings = mlx5e_grp_vnic_env_fill_strings, + .fill_stats = mlx5e_grp_vnic_env_fill_stats, + .update_stats = mlx5e_grp_vnic_env_update_stats, + }, + { .get_num_stats = mlx5e_grp_vport_get_num_stats, .fill_strings = mlx5e_grp_vport_fill_strings, .fill_stats = mlx5e_grp_vport_fill_stats, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h index 0b3320a2b072..53111a2df587 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h @@ -78,6 +78,8 @@ struct mlx5e_sw_stats { u64 tx_queue_wake; u64 tx_queue_dropped; u64 tx_xmit_more; + u64 tx_cqe_err; + u64 tx_recover; u64 rx_wqe_err; u64 rx_mpwqe_filler; u64 rx_buff_alloc_err; @@ -97,6 +99,11 @@ struct mlx5e_sw_stats { struct mlx5e_qcounter_stats { u32 rx_out_of_buffer; + u32 rx_if_down_packets; +}; + +struct mlx5e_vnic_env_stats { + __be64 query_vnic_env_out[MLX5_ST_SZ_QW(query_vnic_env_out)]; }; #define VPORT_COUNTER_GET(vstats, c) MLX5_GET64(query_vport_counter_out, \ @@ -192,6 +199,8 @@ struct mlx5e_sq_stats { u64 stopped; u64 wake; u64 dropped; + u64 cqe_err; + u64 recover; }; struct mlx5e_ch_stats { @@ -201,6 +210,7 @@ struct mlx5e_ch_stats { struct mlx5e_stats { struct mlx5e_sw_stats sw; struct mlx5e_qcounter_stats qcnt; + struct mlx5e_vnic_env_stats vnic; struct mlx5e_vport_stats vport; struct mlx5e_pport_stats pport; struct rtnl_link_stats64 vf_vport; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 43234cabf444..4197001f9801 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -675,6 +675,7 @@ mlx5e_tc_add_nic_flow(struct mlx5e_priv *priv, struct mlx5_flow_destination dest[2] = {}; struct mlx5_flow_act flow_act = { .action = attr->action, + .has_flow_tag = true, .flow_tag = attr->flow_tag, .encap_id = 0, }; @@ -2529,12 +2530,17 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, struct tcf_exts *exts, if (tcf_vlan_action(a) == TCA_VLAN_ACT_POP) { attr->action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_POP; } else if (tcf_vlan_action(a) == TCA_VLAN_ACT_PUSH) { - if (tcf_vlan_push_proto(a) != htons(ETH_P_8021Q) || - tcf_vlan_push_prio(a)) - return -EOPNOTSUPP; - attr->action |= MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH; - attr->vlan = tcf_vlan_push_vid(a); + attr->vlan_vid = tcf_vlan_push_vid(a); + if (mlx5_eswitch_vlan_actions_supported(priv->mdev)) { + attr->vlan_prio = tcf_vlan_push_prio(a); + attr->vlan_proto = tcf_vlan_push_proto(a); + if (!attr->vlan_proto) + attr->vlan_proto = htons(ETH_P_8021Q); + } else if (tcf_vlan_push_proto(a) != htons(ETH_P_8021Q) || + tcf_vlan_push_prio(a)) { + return -EOPNOTSUPP; + } } else { /* action is TCA_VLAN_ACT_MODIFY */ return -EOPNOTSUPP; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c index 11b4f1089d1c..20297108528a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c @@ -417,6 +417,18 @@ netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev) return mlx5e_sq_xmit(sq, skb, wqe, pi); } +static void mlx5e_dump_error_cqe(struct mlx5e_txqsq *sq, + struct mlx5_err_cqe *err_cqe) +{ + u32 ci = mlx5_cqwq_get_ci(&sq->cq.wq); + + netdev_err(sq->channel->netdev, + "Error cqe on cqn 0x%x, ci 0x%x, sqn 0x%x, syndrome 0x%x, vendor syndrome 0x%x\n", + sq->cq.mcq.cqn, ci, sq->sqn, err_cqe->syndrome, + err_cqe->vendor_err_synd); + mlx5_dump_err_cqe(sq->cq.mdev, err_cqe); +} + bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) { struct mlx5e_txqsq *sq; @@ -456,6 +468,17 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) wqe_counter = be16_to_cpu(cqe->wqe_counter); + if (unlikely(cqe->op_own >> 4 == MLX5_CQE_REQ_ERR)) { + if (!test_and_set_bit(MLX5E_SQ_STATE_RECOVERING, + &sq->state)) { + mlx5e_dump_error_cqe(sq, + (struct mlx5_err_cqe *)cqe); + queue_work(cq->channel->priv->wq, + &sq->recover.recover_work); + } + sq->stats.cqe_err++; + } + do { struct mlx5e_tx_wqe_info *wi; struct sk_buff *skb; @@ -509,7 +532,9 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget) netdev_tx_completed_queue(sq->txq, npkts, nbytes); if (netif_tx_queue_stopped(sq->txq) && - mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, MLX5E_SQ_STOP_ROOM)) { + mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, + MLX5E_SQ_STOP_ROOM) && + !test_bit(MLX5E_SQ_STATE_RECOVERING, &sq->state)) { netif_tx_wake_queue(sq->txq); sq->stats.wake++; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c index 25106e996a96..c1c94974e16b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c @@ -393,6 +393,51 @@ static void general_event_handler(struct mlx5_core_dev *dev, } } +/* caller must eventually call mlx5_cq_put on the returned cq */ +static struct mlx5_core_cq *mlx5_eq_cq_get(struct mlx5_eq *eq, u32 cqn) +{ + struct mlx5_cq_table *table = &eq->cq_table; + struct mlx5_core_cq *cq = NULL; + + spin_lock(&table->lock); + cq = radix_tree_lookup(&table->tree, cqn); + if (likely(cq)) + mlx5_cq_hold(cq); + spin_unlock(&table->lock); + + return cq; +} + +static void mlx5_eq_cq_completion(struct mlx5_eq *eq, u32 cqn) +{ + struct mlx5_core_cq *cq = mlx5_eq_cq_get(eq, cqn); + + if (unlikely(!cq)) { + mlx5_core_warn(eq->dev, "Completion event for bogus CQ 0x%x\n", cqn); + return; + } + + ++cq->arm_sn; + + cq->comp(cq); + + mlx5_cq_put(cq); +} + +static void mlx5_eq_cq_event(struct mlx5_eq *eq, u32 cqn, int event_type) +{ + struct mlx5_core_cq *cq = mlx5_eq_cq_get(eq, cqn); + + if (unlikely(!cq)) { + mlx5_core_warn(eq->dev, "Async event for bogus CQ 0x%x\n", cqn); + return; + } + + cq->event(cq, event_type); + + mlx5_cq_put(cq); +} + static irqreturn_t mlx5_eq_int(int irq, void *eq_ptr) { struct mlx5_eq *eq = eq_ptr; @@ -415,7 +460,7 @@ static irqreturn_t mlx5_eq_int(int irq, void *eq_ptr) switch (eqe->type) { case MLX5_EVENT_TYPE_COMP: cqn = be32_to_cpu(eqe->data.comp.cqn) & 0xffffff; - mlx5_cq_completion(dev, cqn); + mlx5_eq_cq_completion(eq, cqn); break; case MLX5_EVENT_TYPE_DCT_DRAINED: rsn = be32_to_cpu(eqe->data.dct.dctn) & 0xffffff; @@ -472,7 +517,7 @@ static irqreturn_t mlx5_eq_int(int irq, void *eq_ptr) cqn = be32_to_cpu(eqe->data.cq_err.cqn) & 0xffffff; mlx5_core_warn(dev, "CQ error on CQN 0x%x, syndrome 0x%x\n", cqn, eqe->data.cq_err.syndrome); - mlx5_cq_event(dev, cqn, eqe->type); + mlx5_eq_cq_event(eq, cqn, eqe->type); break; case MLX5_EVENT_TYPE_PAGE_REQUEST: @@ -567,6 +612,7 @@ int mlx5_create_map_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq, u8 vecidx, int nent, u64 mask, const char *name, enum mlx5_eq_type type) { + struct mlx5_cq_table *cq_table = &eq->cq_table; u32 out[MLX5_ST_SZ_DW(create_eq_out)] = {0}; struct mlx5_priv *priv = &dev->priv; irq_handler_t handler; @@ -576,6 +622,11 @@ int mlx5_create_map_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq, u8 vecidx, u32 *in; int err; + /* Init CQ table */ + memset(cq_table, 0, sizeof(*cq_table)); + spin_lock_init(&cq_table->lock); + INIT_RADIX_TREE(&cq_table->tree, GFP_ATOMIC); + eq->type = type; eq->nent = roundup_pow_of_two(nent + MLX5_NUM_SPARE_EQE); eq->cons_index = 0; @@ -669,7 +720,6 @@ err_buf: mlx5_buf_free(dev, &eq->buf); return err; } -EXPORT_SYMBOL_GPL(mlx5_create_map_eq); int mlx5_destroy_unmap_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq) { @@ -696,7 +746,40 @@ int mlx5_destroy_unmap_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq) return err; } -EXPORT_SYMBOL_GPL(mlx5_destroy_unmap_eq); + +int mlx5_eq_add_cq(struct mlx5_eq *eq, struct mlx5_core_cq *cq) +{ + struct mlx5_cq_table *table = &eq->cq_table; + int err; + + spin_lock_irq(&table->lock); + err = radix_tree_insert(&table->tree, cq->cqn, cq); + spin_unlock_irq(&table->lock); + + return err; +} + +int mlx5_eq_del_cq(struct mlx5_eq *eq, struct mlx5_core_cq *cq) +{ + struct mlx5_cq_table *table = &eq->cq_table; + struct mlx5_core_cq *tmp; + + spin_lock_irq(&table->lock); + tmp = radix_tree_delete(&table->tree, cq->cqn); + spin_unlock_irq(&table->lock); + + if (!tmp) { + mlx5_core_warn(eq->dev, "cq 0x%x not found in eq 0x%x tree\n", eq->eqn, cq->cqn); + return -ENOENT; + } + + if (tmp != cq) { + mlx5_core_warn(eq->dev, "corruption on cqn 0x%x in eq 0x%x\n", eq->eqn, cq->cqn); + return -EINVAL; + } + + return 0; +} int mlx5_eq_init(struct mlx5_core_dev *dev) { @@ -840,4 +923,3 @@ int mlx5_core_eq_query(struct mlx5_core_dev *dev, struct mlx5_eq *eq, MLX5_SET(query_eq_in, in, eq_number, eq->eqn); return mlx5_cmd_exec(dev, in, sizeof(in), out, outlen); } -EXPORT_SYMBOL_GPL(mlx5_core_eq_query); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index c2b1d7d351fc..332bc56306bf 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -1619,10 +1619,14 @@ int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode) esw_info(esw->dev, "E-Switch enable SRIOV: nvfs(%d) mode (%d)\n", nvfs, mode); esw->mode = mode; - if (mode == SRIOV_LEGACY) + if (mode == SRIOV_LEGACY) { err = esw_create_legacy_fdb_table(esw, nvfs + 1); - else + } else { + mlx5_reload_interface(esw->dev, MLX5_INTERFACE_PROTOCOL_IB); + err = esw_offloads_init(esw, nvfs + 1); + } + if (err) goto abort; @@ -1644,12 +1648,17 @@ int mlx5_eswitch_enable_sriov(struct mlx5_eswitch *esw, int nvfs, int mode) abort: esw->mode = SRIOV_NONE; + + if (mode == SRIOV_OFFLOADS) + mlx5_reload_interface(esw->dev, MLX5_INTERFACE_PROTOCOL_IB); + return err; } void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw) { struct esw_mc_addr *mc_promisc; + int old_mode; int nvports; int i; @@ -1675,7 +1684,11 @@ void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw) else if (esw->mode == SRIOV_OFFLOADS) esw_offloads_cleanup(esw, nvports); + old_mode = esw->mode; esw->mode = SRIOV_NONE; + + if (old_mode == SRIOV_OFFLOADS) + mlx5_reload_interface(esw->dev, MLX5_INTERFACE_PROTOCOL_IB); } int mlx5_eswitch_init(struct mlx5_core_dev *dev) @@ -2083,17 +2096,19 @@ unlock: return err; } -static void mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev, - int vport_idx, - struct mlx5_vport_drop_stats *stats) +static int mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev, + int vport_idx, + struct mlx5_vport_drop_stats *stats) { struct mlx5_eswitch *esw = dev->priv.eswitch; struct mlx5_vport *vport = &esw->vports[vport_idx]; + u64 rx_discard_vport_down, tx_discard_vport_down; u64 bytes = 0; u16 idx = 0; + int err = 0; if (!vport->enabled || esw->mode != SRIOV_LEGACY) - return; + return 0; if (vport->egress.drop_counter) { idx = vport->egress.drop_counter->id; @@ -2104,6 +2119,23 @@ static void mlx5_eswitch_query_vport_drop_stats(struct mlx5_core_dev *dev, idx = vport->ingress.drop_counter->id; mlx5_fc_query(dev, idx, &stats->tx_dropped, &bytes); } + + if (!MLX5_CAP_GEN(dev, receive_discard_vport_down) && + !MLX5_CAP_GEN(dev, transmit_discard_vport_down)) + return 0; + + err = mlx5_query_vport_down_stats(dev, vport_idx, + &rx_discard_vport_down, + &tx_discard_vport_down); + if (err) + return err; + + if (MLX5_CAP_GEN(dev, receive_discard_vport_down)) + stats->rx_dropped += rx_discard_vport_down; + if (MLX5_CAP_GEN(dev, transmit_discard_vport_down)) + stats->tx_dropped += tx_discard_vport_down; + + return 0; } int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw, @@ -2167,7 +2199,9 @@ int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw, vf_stats->broadcast = MLX5_GET_CTR(out, received_eth_broadcast.packets); - mlx5_eswitch_query_vport_drop_stats(esw->dev, vport, &stats); + err = mlx5_eswitch_query_vport_drop_stats(esw->dev, vport, &stats); + if (err) + goto free_out; vf_stats->rx_dropped = stats.rx_dropped; vf_stats->tx_dropped = stats.tx_dropped; @@ -2175,3 +2209,9 @@ free_out: kvfree(out); return err; } + +u8 mlx5_eswitch_mode(struct mlx5_eswitch *esw) +{ + return esw->mode; +} +EXPORT_SYMBOL_GPL(mlx5_eswitch_mode); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h index 2fa037066b2f..4cd773fa55e3 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h @@ -37,19 +37,9 @@ #include <linux/if_link.h> #include <net/devlink.h> #include <linux/mlx5/device.h> +#include <linux/mlx5/eswitch.h> #include "lib/mpfs.h" -enum { - SRIOV_NONE, - SRIOV_LEGACY, - SRIOV_OFFLOADS -}; - -enum { - REP_ETH, - NUM_REP_TYPES, -}; - #ifdef CONFIG_MLX5_ESWITCH #define MLX5_MAX_UC_PER_VPORT(dev) \ @@ -139,29 +129,13 @@ struct mlx5_eswitch_fdb { struct mlx5_flow_table *fdb; struct mlx5_flow_group *send_to_vport_grp; struct mlx5_flow_group *miss_grp; - struct mlx5_flow_handle *miss_rule; + struct mlx5_flow_handle *miss_rule_uni; + struct mlx5_flow_handle *miss_rule_multi; int vlan_push_pop_refcount; } offloads; }; }; -struct mlx5_eswitch_rep; -struct mlx5_eswitch_rep_if { - int (*load)(struct mlx5_core_dev *dev, - struct mlx5_eswitch_rep *rep); - void (*unload)(struct mlx5_eswitch_rep *rep); - void *priv; - bool valid; -}; - -struct mlx5_eswitch_rep { - struct mlx5_eswitch_rep_if rep_if[NUM_REP_TYPES]; - u16 vport; - u8 hw_id[ETH_ALEN]; - u16 vlan; - u32 vlan_refcount; -}; - struct mlx5_esw_offload { struct mlx5_flow_table *ft_offloads; struct mlx5_flow_group *vport_rx_group; @@ -231,9 +205,6 @@ int mlx5_eswitch_get_vport_config(struct mlx5_eswitch *esw, int mlx5_eswitch_get_vport_stats(struct mlx5_eswitch *esw, int vport, struct ifla_vf_stats *vf_stats); -struct mlx5_flow_handle * -mlx5_eswitch_add_send_to_vport_rule(struct mlx5_eswitch *esw, int vport, - u32 sqn); void mlx5_eswitch_del_send_to_vport_rule(struct mlx5_flow_handle *rule); struct mlx5_flow_spec; @@ -256,15 +227,14 @@ enum { SET_VLAN_INSERT = BIT(1) }; -#define MLX5_FLOW_CONTEXT_ACTION_VLAN_POP 0x4000 -#define MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH 0x8000 - struct mlx5_esw_flow_attr { struct mlx5_eswitch_rep *in_rep; struct mlx5_eswitch_rep *out_rep; int action; - u16 vlan; + __be16 vlan_proto; + u16 vlan_vid; + u8 vlan_prio; bool vlan_handled; u32 encap_id; u32 mod_hdr_id; @@ -278,13 +248,6 @@ int mlx5_devlink_eswitch_inline_mode_get(struct devlink *devlink, u8 *mode); int mlx5_eswitch_inline_mode_get(struct mlx5_eswitch *esw, int nvfs, u8 *mode); int mlx5_devlink_eswitch_encap_mode_set(struct devlink *devlink, u8 encap); int mlx5_devlink_eswitch_encap_mode_get(struct devlink *devlink, u8 *encap); -void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw, - int vport_index, - struct mlx5_eswitch_rep_if *rep_if, - u8 rep_type); -void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw, - int vport_index, - u8 rep_type); void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw, u8 rep_type); int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw, @@ -294,6 +257,12 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw, int __mlx5_eswitch_set_vport_vlan(struct mlx5_eswitch *esw, int vport, u16 vlan, u8 qos, u8 set_flags); +static inline bool mlx5_eswitch_vlan_actions_supported(struct mlx5_core_dev *dev) +{ + return MLX5_CAP_ESW_FLOWTABLE_FDB(dev, pop_vlan) && + MLX5_CAP_ESW_FLOWTABLE_FDB(dev, push_vlan); +} + #define MLX5_DEBUG_ESWITCH_MASK BIT(3) #define esw_info(dev, format, ...) \ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index 99f583a15cc3..35e256eb2f6e 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -58,8 +58,16 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, if (esw->mode != SRIOV_OFFLOADS) return ERR_PTR(-EOPNOTSUPP); - /* per flow vlan pop/push is emulated, don't set that into the firmware */ - flow_act.action = attr->action & ~(MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH | MLX5_FLOW_CONTEXT_ACTION_VLAN_POP); + flow_act.action = attr->action; + /* if per flow vlan pop/push is emulated, don't set that into the firmware */ + if (!mlx5_eswitch_vlan_actions_supported(esw->dev)) + flow_act.action &= ~(MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH | + MLX5_FLOW_CONTEXT_ACTION_VLAN_POP); + else if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH) { + flow_act.vlan.ethtype = ntohs(attr->vlan_proto); + flow_act.vlan.vid = attr->vlan_vid; + flow_act.vlan.prio = attr->vlan_prio; + } if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) { dest[i].type = MLX5_FLOW_DESTINATION_TYPE_VPORT; @@ -88,10 +96,10 @@ mlx5_eswitch_add_offloaded_rule(struct mlx5_eswitch *esw, if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_DECAP) spec->match_criteria_enable |= MLX5_MATCH_INNER_HEADERS; - if (attr->action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) + if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_MOD_HDR) flow_act.modify_id = attr->mod_hdr_id; - if (attr->action & MLX5_FLOW_CONTEXT_ACTION_ENCAP) + if (flow_act.action & MLX5_FLOW_CONTEXT_ACTION_ENCAP) flow_act.encap_id = attr->encap_id; rule = mlx5_add_flow_rules((struct mlx5_flow_table *)esw->fdb_table.fdb, @@ -185,7 +193,7 @@ static int esw_add_vlan_action_check(struct mlx5_esw_flow_attr *attr, /* protects against (1) setting rules with different vlans to push and * (2) setting rules w.o vlans (attr->vlan = 0) && w. vlans to push (!= 0) */ - if (push && in_rep->vlan_refcount && (in_rep->vlan != attr->vlan)) + if (push && in_rep->vlan_refcount && (in_rep->vlan != attr->vlan_vid)) goto out_notsupp; return 0; @@ -202,6 +210,10 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw, bool push, pop, fwd; int err = 0; + /* nop if we're on the vlan push/pop non emulation mode */ + if (mlx5_eswitch_vlan_actions_supported(esw->dev)) + return 0; + push = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH); pop = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_VLAN_POP); fwd = !!(attr->action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST); @@ -239,11 +251,11 @@ int mlx5_eswitch_add_vlan_action(struct mlx5_eswitch *esw, if (vport->vlan_refcount) goto skip_set_push; - err = __mlx5_eswitch_set_vport_vlan(esw, vport->vport, attr->vlan, 0, + err = __mlx5_eswitch_set_vport_vlan(esw, vport->vport, attr->vlan_vid, 0, SET_VLAN_INSERT | SET_VLAN_STRIP); if (err) goto out; - vport->vlan = attr->vlan; + vport->vlan = attr->vlan_vid; skip_set_push: vport->vlan_refcount++; } @@ -261,6 +273,10 @@ int mlx5_eswitch_del_vlan_action(struct mlx5_eswitch *esw, bool push, pop, fwd; int err = 0; + /* nop if we're on the vlan push/pop non emulation mode */ + if (mlx5_eswitch_vlan_actions_supported(esw->dev)) + return 0; + if (!attr->vlan_handled) return 0; @@ -338,6 +354,7 @@ out: kvfree(spec); return flow_rule; } +EXPORT_SYMBOL(mlx5_eswitch_add_send_to_vport_rule); void mlx5_eswitch_del_send_to_vport_rule(struct mlx5_flow_handle *rule) { @@ -350,7 +367,11 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) struct mlx5_flow_destination dest = {}; struct mlx5_flow_handle *flow_rule = NULL; struct mlx5_flow_spec *spec; + void *headers_c; + void *headers_v; int err = 0; + u8 *dmac_c; + u8 *dmac_v; spec = kvzalloc(sizeof(*spec), GFP_KERNEL); if (!spec) { @@ -358,6 +379,13 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) goto out; } + spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS; + headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria, + outer_headers); + dmac_c = MLX5_ADDR_OF(fte_match_param, headers_c, + outer_headers.dmac_47_16); + dmac_c[0] = 0x01; + dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; dest.vport_num = 0; flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; @@ -366,11 +394,28 @@ static int esw_add_fdb_miss_rule(struct mlx5_eswitch *esw) &flow_act, &dest, 1); if (IS_ERR(flow_rule)) { err = PTR_ERR(flow_rule); - esw_warn(esw->dev, "FDB: Failed to add miss flow rule err %d\n", err); + esw_warn(esw->dev, "FDB: Failed to add unicast miss flow rule err %d\n", err); goto out; } - esw->fdb_table.offloads.miss_rule = flow_rule; + esw->fdb_table.offloads.miss_rule_uni = flow_rule; + + headers_v = MLX5_ADDR_OF(fte_match_param, spec->match_value, + outer_headers); + dmac_v = MLX5_ADDR_OF(fte_match_param, headers_v, + outer_headers.dmac_47_16); + dmac_v[0] = 0x01; + flow_rule = mlx5_add_flow_rules(esw->fdb_table.offloads.fdb, spec, + &flow_act, &dest, 1); + if (IS_ERR(flow_rule)) { + err = PTR_ERR(flow_rule); + esw_warn(esw->dev, "FDB: Failed to add multicast miss flow rule err %d\n", err); + mlx5_del_flow_rules(esw->fdb_table.offloads.miss_rule_uni); + goto out; + } + + esw->fdb_table.offloads.miss_rule_multi = flow_rule; + out: kvfree(spec); return err; @@ -426,6 +471,7 @@ static void esw_destroy_offloads_fast_fdb_table(struct mlx5_eswitch *esw) } #define MAX_PF_SQ 256 +#define MAX_SQ_NVPORTS 32 static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw, int nvports) { @@ -438,6 +484,7 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw, int nvports) struct mlx5_flow_group *g; void *match_criteria; u32 *flow_group_in; + u8 *dmac; esw_debug(esw->dev, "Create offloads FDB Tables\n"); flow_group_in = kvzalloc(inlen, GFP_KERNEL); @@ -455,7 +502,7 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw, int nvports) if (err) goto fast_fdb_err; - table_size = nvports + MAX_PF_SQ + 1; + table_size = nvports * MAX_SQ_NVPORTS + MAX_PF_SQ + 2; ft_attr.max_fte = table_size; ft_attr.prio = FDB_SLOW_PATH; @@ -478,7 +525,7 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw, int nvports) MLX5_SET_TO_ONES(fte_match_param, match_criteria, misc_parameters.source_sqn); MLX5_SET_TO_ONES(fte_match_param, match_criteria, misc_parameters.source_port); - ix = nvports + MAX_PF_SQ; + ix = nvports * MAX_SQ_NVPORTS + MAX_PF_SQ; MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, 0); MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, ix - 1); @@ -492,10 +539,16 @@ static int esw_create_offloads_fdb_tables(struct mlx5_eswitch *esw, int nvports) /* create miss group */ memset(flow_group_in, 0, inlen); - MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, 0); + MLX5_SET(create_flow_group_in, flow_group_in, match_criteria_enable, + MLX5_MATCH_OUTER_HEADERS); + match_criteria = MLX5_ADDR_OF(create_flow_group_in, flow_group_in, + match_criteria); + dmac = MLX5_ADDR_OF(fte_match_param, match_criteria, + outer_headers.dmac_47_16); + dmac[0] = 0x01; MLX5_SET(create_flow_group_in, flow_group_in, start_flow_index, ix); - MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, ix + 1); + MLX5_SET(create_flow_group_in, flow_group_in, end_flow_index, ix + 2); g = mlx5_create_flow_group(fdb, flow_group_in); if (IS_ERR(g)) { @@ -531,7 +584,8 @@ static void esw_destroy_offloads_fdb_tables(struct mlx5_eswitch *esw) return; esw_debug(esw->dev, "Destroy offloads FDB Tables\n"); - mlx5_del_flow_rules(esw->fdb_table.offloads.miss_rule); + mlx5_del_flow_rules(esw->fdb_table.offloads.miss_rule_multi); + mlx5_del_flow_rules(esw->fdb_table.offloads.miss_rule_uni); mlx5_destroy_flow_group(esw->fdb_table.offloads.send_to_vport_grp); mlx5_destroy_flow_group(esw->fdb_table.offloads.miss_grp); @@ -789,14 +843,9 @@ int esw_offloads_init(struct mlx5_eswitch *esw, int nvports) { int err; - /* disable PF RoCE so missed packets don't go through RoCE steering */ - mlx5_dev_list_lock(); - mlx5_remove_dev_by_protocol(esw->dev, MLX5_INTERFACE_PROTOCOL_IB); - mlx5_dev_list_unlock(); - err = esw_create_offloads_fdb_tables(esw, nvports); if (err) - goto create_fdb_err; + return err; err = esw_create_offloads_table(esw); if (err) @@ -821,12 +870,6 @@ create_fg_err: create_ft_err: esw_destroy_offloads_fdb_tables(esw); -create_fdb_err: - /* enable back PF RoCE */ - mlx5_dev_list_lock(); - mlx5_add_dev_by_protocol(esw->dev, MLX5_INTERFACE_PROTOCOL_IB); - mlx5_dev_list_unlock(); - return err; } @@ -844,9 +887,7 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw) } /* enable back PF RoCE */ - mlx5_dev_list_lock(); - mlx5_add_dev_by_protocol(esw->dev, MLX5_INTERFACE_PROTOCOL_IB); - mlx5_dev_list_unlock(); + mlx5_reload_interface(esw->dev, MLX5_INTERFACE_PROTOCOL_IB); return err; } @@ -1160,10 +1201,12 @@ void mlx5_eswitch_register_vport_rep(struct mlx5_eswitch *esw, rep_if->load = __rep_if->load; rep_if->unload = __rep_if->unload; + rep_if->get_proto_dev = __rep_if->get_proto_dev; rep_if->priv = __rep_if->priv; rep_if->valid = true; } +EXPORT_SYMBOL(mlx5_eswitch_register_vport_rep); void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw, int vport_index, u8 rep_type) @@ -1178,6 +1221,7 @@ void mlx5_eswitch_unregister_vport_rep(struct mlx5_eswitch *esw, rep->rep_if[rep_type].valid = false; } +EXPORT_SYMBOL(mlx5_eswitch_unregister_vport_rep); void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw, u8 rep_type) { @@ -1188,3 +1232,35 @@ void *mlx5_eswitch_get_uplink_priv(struct mlx5_eswitch *esw, u8 rep_type) rep = &offloads->vport_reps[UPLINK_REP_INDEX]; return rep->rep_if[rep_type].priv; } + +void *mlx5_eswitch_get_proto_dev(struct mlx5_eswitch *esw, + int vport, + u8 rep_type) +{ + struct mlx5_esw_offload *offloads = &esw->offloads; + struct mlx5_eswitch_rep *rep; + + if (vport == FDB_UPLINK_VPORT) + vport = UPLINK_REP_INDEX; + + rep = &offloads->vport_reps[vport]; + + if (rep->rep_if[rep_type].valid && + rep->rep_if[rep_type].get_proto_dev) + return rep->rep_if[rep_type].get_proto_dev(rep); + return NULL; +} +EXPORT_SYMBOL(mlx5_eswitch_get_proto_dev); + +void *mlx5_eswitch_uplink_get_proto_dev(struct mlx5_eswitch *esw, u8 rep_type) +{ + return mlx5_eswitch_get_proto_dev(esw, UPLINK_REP_INDEX, rep_type); +} +EXPORT_SYMBOL(mlx5_eswitch_uplink_get_proto_dev); + +struct mlx5_eswitch_rep *mlx5_eswitch_vport_rep(struct mlx5_eswitch *esw, + int vport) +{ + return &esw->offloads.vport_reps[vport]; +} +EXPORT_SYMBOL(mlx5_eswitch_vport_rep); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c index 35d0e33381ca..0f5da499a223 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c @@ -31,49 +31,91 @@ * */ +#include <linux/rhashtable.h> #include <linux/mlx5/driver.h> +#include <linux/mlx5/fs_helpers.h> +#include <linux/mlx5/fs.h> +#include <linux/rbtree.h> #include "mlx5_core.h" +#include "fs_cmd.h" #include "fpga/ipsec.h" #include "fpga/sdk.h" #include "fpga/core.h" #define SBU_QP_QUEUE_SIZE 8 +#define MLX5_FPGA_IPSEC_CMD_TIMEOUT_MSEC (60 * 1000) -enum mlx5_ipsec_response_syndrome { - MLX5_IPSEC_RESPONSE_SUCCESS = 0, - MLX5_IPSEC_RESPONSE_ILLEGAL_REQUEST = 1, - MLX5_IPSEC_RESPONSE_SADB_ISSUE = 2, - MLX5_IPSEC_RESPONSE_WRITE_RESPONSE_ISSUE = 3, +enum mlx5_fpga_ipsec_cmd_status { + MLX5_FPGA_IPSEC_CMD_PENDING, + MLX5_FPGA_IPSEC_CMD_SEND_FAIL, + MLX5_FPGA_IPSEC_CMD_COMPLETE, }; -enum mlx5_fpga_ipsec_sacmd_status { - MLX5_FPGA_IPSEC_SACMD_PENDING, - MLX5_FPGA_IPSEC_SACMD_SEND_FAIL, - MLX5_FPGA_IPSEC_SACMD_COMPLETE, -}; - -struct mlx5_ipsec_command_context { +struct mlx5_fpga_ipsec_cmd_context { struct mlx5_fpga_dma_buf buf; - struct mlx5_accel_ipsec_sa sa; - enum mlx5_fpga_ipsec_sacmd_status status; + enum mlx5_fpga_ipsec_cmd_status status; + struct mlx5_ifc_fpga_ipsec_cmd_resp resp; int status_code; struct completion complete; struct mlx5_fpga_device *dev; struct list_head list; /* Item in pending_cmds */ + u8 command[0]; +}; + +struct mlx5_fpga_esp_xfrm; + +struct mlx5_fpga_ipsec_sa_ctx { + struct rhash_head hash; + struct mlx5_ifc_fpga_ipsec_sa hw_sa; + struct mlx5_core_dev *dev; + struct mlx5_fpga_esp_xfrm *fpga_xfrm; +}; + +struct mlx5_fpga_esp_xfrm { + unsigned int num_rules; + struct mlx5_fpga_ipsec_sa_ctx *sa_ctx; + struct mutex lock; /* xfrm lock */ + struct mlx5_accel_esp_xfrm accel_xfrm; +}; + +struct mlx5_fpga_ipsec_rule { + struct rb_node node; + struct fs_fte *fte; + struct mlx5_fpga_ipsec_sa_ctx *ctx; }; -struct mlx5_ipsec_sadb_resp { - __be32 syndrome; - __be32 sw_sa_handle; - u8 reserved[24]; -} __packed; +static const struct rhashtable_params rhash_sa = { + .key_len = FIELD_SIZEOF(struct mlx5_fpga_ipsec_sa_ctx, hw_sa), + .key_offset = offsetof(struct mlx5_fpga_ipsec_sa_ctx, hw_sa), + .head_offset = offsetof(struct mlx5_fpga_ipsec_sa_ctx, hash), + .automatic_shrinking = true, + .min_size = 1, +}; struct mlx5_fpga_ipsec { + struct mlx5_fpga_device *fdev; struct list_head pending_cmds; spinlock_t pending_cmds_lock; /* Protects pending_cmds */ u32 caps[MLX5_ST_SZ_DW(ipsec_extended_cap)]; struct mlx5_fpga_conn *conn; + + struct notifier_block fs_notifier_ingress_bypass; + struct notifier_block fs_notifier_egress; + + /* Map hardware SA --> SA context + * (mlx5_fpga_ipsec_sa) (mlx5_fpga_ipsec_sa_ctx) + * We will use this hash to avoid SAs duplication in fpga which + * aren't allowed + */ + struct rhashtable sa_hash; /* hw_sa -> mlx5_fpga_ipsec_sa_ctx */ + struct mutex sa_hash_lock; + + /* Tree holding all rules for this fpga device + * Key for searching a rule (mlx5_fpga_ipsec_rule) is (ft, id) + */ + struct rb_root rules_rb; + struct mutex rules_rb_lock; /* rules lock */ }; static bool mlx5_fpga_is_ipsec_device(struct mlx5_core_dev *mdev) @@ -97,28 +139,29 @@ static void mlx5_fpga_ipsec_send_complete(struct mlx5_fpga_conn *conn, struct mlx5_fpga_dma_buf *buf, u8 status) { - struct mlx5_ipsec_command_context *context; + struct mlx5_fpga_ipsec_cmd_context *context; if (status) { - context = container_of(buf, struct mlx5_ipsec_command_context, + context = container_of(buf, struct mlx5_fpga_ipsec_cmd_context, buf); mlx5_fpga_warn(fdev, "IPSec command send failed with status %u\n", status); - context->status = MLX5_FPGA_IPSEC_SACMD_SEND_FAIL; + context->status = MLX5_FPGA_IPSEC_CMD_SEND_FAIL; complete(&context->complete); } } -static inline int syndrome_to_errno(enum mlx5_ipsec_response_syndrome syndrome) +static inline +int syndrome_to_errno(enum mlx5_ifc_fpga_ipsec_response_syndrome syndrome) { switch (syndrome) { - case MLX5_IPSEC_RESPONSE_SUCCESS: + case MLX5_FPGA_IPSEC_RESPONSE_SUCCESS: return 0; - case MLX5_IPSEC_RESPONSE_SADB_ISSUE: + case MLX5_FPGA_IPSEC_RESPONSE_SADB_ISSUE: return -EEXIST; - case MLX5_IPSEC_RESPONSE_ILLEGAL_REQUEST: + case MLX5_FPGA_IPSEC_RESPONSE_ILLEGAL_REQUEST: return -EINVAL; - case MLX5_IPSEC_RESPONSE_WRITE_RESPONSE_ISSUE: + case MLX5_FPGA_IPSEC_RESPONSE_WRITE_RESPONSE_ISSUE: return -EIO; } return -EIO; @@ -126,9 +169,9 @@ static inline int syndrome_to_errno(enum mlx5_ipsec_response_syndrome syndrome) static void mlx5_fpga_ipsec_recv(void *cb_arg, struct mlx5_fpga_dma_buf *buf) { - struct mlx5_ipsec_sadb_resp *resp = buf->sg[0].data; - struct mlx5_ipsec_command_context *context; - enum mlx5_ipsec_response_syndrome syndrome; + struct mlx5_ifc_fpga_ipsec_cmd_resp *resp = buf->sg[0].data; + struct mlx5_fpga_ipsec_cmd_context *context; + enum mlx5_ifc_fpga_ipsec_response_syndrome syndrome; struct mlx5_fpga_device *fdev = cb_arg; unsigned long flags; @@ -138,12 +181,12 @@ static void mlx5_fpga_ipsec_recv(void *cb_arg, struct mlx5_fpga_dma_buf *buf) return; } - mlx5_fpga_dbg(fdev, "mlx5_ipsec recv_cb syndrome %08x sa_id %x\n", - ntohl(resp->syndrome), ntohl(resp->sw_sa_handle)); + mlx5_fpga_dbg(fdev, "mlx5_ipsec recv_cb syndrome %08x\n", + ntohl(resp->syndrome)); spin_lock_irqsave(&fdev->ipsec->pending_cmds_lock, flags); context = list_first_entry_or_null(&fdev->ipsec->pending_cmds, - struct mlx5_ipsec_command_context, + struct mlx5_fpga_ipsec_cmd_context, list); if (context) list_del(&context->list); @@ -155,51 +198,48 @@ static void mlx5_fpga_ipsec_recv(void *cb_arg, struct mlx5_fpga_dma_buf *buf) } mlx5_fpga_dbg(fdev, "Handling response for %p\n", context); - if (context->sa.sw_sa_handle != resp->sw_sa_handle) { - mlx5_fpga_err(fdev, "mismatch SA handle. cmd 0x%08x vs resp 0x%08x\n", - ntohl(context->sa.sw_sa_handle), - ntohl(resp->sw_sa_handle)); - return; - } - syndrome = ntohl(resp->syndrome); context->status_code = syndrome_to_errno(syndrome); - context->status = MLX5_FPGA_IPSEC_SACMD_COMPLETE; + context->status = MLX5_FPGA_IPSEC_CMD_COMPLETE; + memcpy(&context->resp, resp, sizeof(*resp)); if (context->status_code) - mlx5_fpga_warn(fdev, "IPSec SADB command failed with syndrome %08x\n", + mlx5_fpga_warn(fdev, "IPSec command failed with syndrome %08x\n", syndrome); + complete(&context->complete); } -void *mlx5_fpga_ipsec_sa_cmd_exec(struct mlx5_core_dev *mdev, - struct mlx5_accel_ipsec_sa *cmd) +static void *mlx5_fpga_ipsec_cmd_exec(struct mlx5_core_dev *mdev, + const void *cmd, int cmd_size) { - struct mlx5_ipsec_command_context *context; + struct mlx5_fpga_ipsec_cmd_context *context; struct mlx5_fpga_device *fdev = mdev->fpga; unsigned long flags; - int res = 0; + int res; - BUILD_BUG_ON((sizeof(struct mlx5_accel_ipsec_sa) & 3) != 0); if (!fdev || !fdev->ipsec) return ERR_PTR(-EOPNOTSUPP); - context = kzalloc(sizeof(*context), GFP_ATOMIC); + if (cmd_size & 3) + return ERR_PTR(-EINVAL); + + context = kzalloc(sizeof(*context) + cmd_size, GFP_ATOMIC); if (!context) return ERR_PTR(-ENOMEM); - memcpy(&context->sa, cmd, sizeof(*cmd)); + context->status = MLX5_FPGA_IPSEC_CMD_PENDING; + context->dev = fdev; context->buf.complete = mlx5_fpga_ipsec_send_complete; - context->buf.sg[0].size = sizeof(context->sa); - context->buf.sg[0].data = &context->sa; init_completion(&context->complete); - context->dev = fdev; + memcpy(&context->command, cmd, cmd_size); + context->buf.sg[0].size = cmd_size; + context->buf.sg[0].data = &context->command; + spin_lock_irqsave(&fdev->ipsec->pending_cmds_lock, flags); list_add_tail(&context->list, &fdev->ipsec->pending_cmds); spin_unlock_irqrestore(&fdev->ipsec->pending_cmds_lock, flags); - context->status = MLX5_FPGA_IPSEC_SACMD_PENDING; - res = mlx5_fpga_sbu_conn_sendmsg(fdev->ipsec->conn, &context->buf); if (res) { mlx5_fpga_warn(fdev, "Failure sending IPSec command: %d\n", @@ -214,47 +254,103 @@ void *mlx5_fpga_ipsec_sa_cmd_exec(struct mlx5_core_dev *mdev, return context; } -int mlx5_fpga_ipsec_sa_cmd_wait(void *ctx) +static int mlx5_fpga_ipsec_cmd_wait(void *ctx) { - struct mlx5_ipsec_command_context *context = ctx; + struct mlx5_fpga_ipsec_cmd_context *context = ctx; + unsigned long timeout = + msecs_to_jiffies(MLX5_FPGA_IPSEC_CMD_TIMEOUT_MSEC); int res; - res = wait_for_completion_killable(&context->complete); - if (res) { + res = wait_for_completion_timeout(&context->complete, timeout); + if (!res) { mlx5_fpga_warn(context->dev, "Failure waiting for IPSec command response\n"); - return -EINTR; + return -ETIMEDOUT; } - if (context->status == MLX5_FPGA_IPSEC_SACMD_COMPLETE) + if (context->status == MLX5_FPGA_IPSEC_CMD_COMPLETE) res = context->status_code; else res = -EIO; - kfree(context); return res; } +static inline bool is_v2_sadb_supported(struct mlx5_fpga_ipsec *fipsec) +{ + if (MLX5_GET(ipsec_extended_cap, fipsec->caps, v2_command)) + return true; + return false; +} + +static int mlx5_fpga_ipsec_update_hw_sa(struct mlx5_fpga_device *fdev, + struct mlx5_ifc_fpga_ipsec_sa *hw_sa, + int opcode) +{ + struct mlx5_core_dev *dev = fdev->mdev; + struct mlx5_ifc_fpga_ipsec_sa *sa; + struct mlx5_fpga_ipsec_cmd_context *cmd_context; + size_t sa_cmd_size; + int err; + + hw_sa->ipsec_sa_v1.cmd = htonl(opcode); + if (is_v2_sadb_supported(fdev->ipsec)) + sa_cmd_size = sizeof(*hw_sa); + else + sa_cmd_size = sizeof(hw_sa->ipsec_sa_v1); + + cmd_context = (struct mlx5_fpga_ipsec_cmd_context *) + mlx5_fpga_ipsec_cmd_exec(dev, hw_sa, sa_cmd_size); + if (IS_ERR(cmd_context)) + return PTR_ERR(cmd_context); + + err = mlx5_fpga_ipsec_cmd_wait(cmd_context); + if (err) + goto out; + + sa = (struct mlx5_ifc_fpga_ipsec_sa *)&cmd_context->command; + if (sa->ipsec_sa_v1.sw_sa_handle != cmd_context->resp.sw_sa_handle) { + mlx5_fpga_err(fdev, "mismatch SA handle. cmd 0x%08x vs resp 0x%08x\n", + ntohl(sa->ipsec_sa_v1.sw_sa_handle), + ntohl(cmd_context->resp.sw_sa_handle)); + err = -EIO; + } + +out: + kfree(cmd_context); + return err; +} + u32 mlx5_fpga_ipsec_device_caps(struct mlx5_core_dev *mdev) { struct mlx5_fpga_device *fdev = mdev->fpga; u32 ret = 0; - if (mlx5_fpga_is_ipsec_device(mdev)) - ret |= MLX5_ACCEL_IPSEC_DEVICE; - else + if (mlx5_fpga_is_ipsec_device(mdev)) { + ret |= MLX5_ACCEL_IPSEC_CAP_DEVICE; + ret |= MLX5_ACCEL_IPSEC_CAP_REQUIRED_METADATA; + } else { return ret; + } if (!fdev->ipsec) return ret; if (MLX5_GET(ipsec_extended_cap, fdev->ipsec->caps, esp)) - ret |= MLX5_ACCEL_IPSEC_ESP; + ret |= MLX5_ACCEL_IPSEC_CAP_ESP; if (MLX5_GET(ipsec_extended_cap, fdev->ipsec->caps, ipv6)) - ret |= MLX5_ACCEL_IPSEC_IPV6; + ret |= MLX5_ACCEL_IPSEC_CAP_IPV6; if (MLX5_GET(ipsec_extended_cap, fdev->ipsec->caps, lso)) - ret |= MLX5_ACCEL_IPSEC_LSO; + ret |= MLX5_ACCEL_IPSEC_CAP_LSO; + + if (MLX5_GET(ipsec_extended_cap, fdev->ipsec->caps, rx_no_trailer)) + ret |= MLX5_ACCEL_IPSEC_CAP_RX_NO_TRAILER; + + if (MLX5_GET(ipsec_extended_cap, fdev->ipsec->caps, esn)) { + ret |= MLX5_ACCEL_IPSEC_CAP_ESN; + ret |= MLX5_ACCEL_IPSEC_CAP_TX_IV_IS_ESN; + } return ret; } @@ -318,6 +414,829 @@ out: return ret; } +static int mlx5_fpga_ipsec_set_caps(struct mlx5_core_dev *mdev, u32 flags) +{ + struct mlx5_fpga_ipsec_cmd_context *context; + struct mlx5_ifc_fpga_ipsec_cmd_cap cmd = {0}; + int err; + + cmd.cmd = htonl(MLX5_FPGA_IPSEC_CMD_OP_SET_CAP); + cmd.flags = htonl(flags); + context = mlx5_fpga_ipsec_cmd_exec(mdev, &cmd, sizeof(cmd)); + if (IS_ERR(context)) { + err = PTR_ERR(context); + goto out; + } + + err = mlx5_fpga_ipsec_cmd_wait(context); + if (err) + goto out; + + if ((context->resp.flags & cmd.flags) != cmd.flags) { + mlx5_fpga_err(context->dev, "Failed to set capabilities. cmd 0x%08x vs resp 0x%08x\n", + cmd.flags, + context->resp.flags); + err = -EIO; + } + +out: + return err; +} + +static int mlx5_fpga_ipsec_enable_supported_caps(struct mlx5_core_dev *mdev) +{ + u32 dev_caps = mlx5_fpga_ipsec_device_caps(mdev); + u32 flags = 0; + + if (dev_caps & MLX5_ACCEL_IPSEC_CAP_RX_NO_TRAILER) + flags |= MLX5_FPGA_IPSEC_CAP_NO_TRAILER; + + return mlx5_fpga_ipsec_set_caps(mdev, flags); +} + +static void +mlx5_fpga_ipsec_build_hw_xfrm(struct mlx5_core_dev *mdev, + const struct mlx5_accel_esp_xfrm_attrs *xfrm_attrs, + struct mlx5_ifc_fpga_ipsec_sa *hw_sa) +{ + const struct aes_gcm_keymat *aes_gcm = &xfrm_attrs->keymat.aes_gcm; + + /* key */ + memcpy(&hw_sa->ipsec_sa_v1.key_enc, aes_gcm->aes_key, + aes_gcm->key_len / 8); + /* Duplicate 128 bit key twice according to HW layout */ + if (aes_gcm->key_len == 128) + memcpy(&hw_sa->ipsec_sa_v1.key_enc[16], + aes_gcm->aes_key, aes_gcm->key_len / 8); + + /* salt and seq_iv */ + memcpy(&hw_sa->ipsec_sa_v1.gcm.salt_iv, &aes_gcm->seq_iv, + sizeof(aes_gcm->seq_iv)); + memcpy(&hw_sa->ipsec_sa_v1.gcm.salt, &aes_gcm->salt, + sizeof(aes_gcm->salt)); + + /* esn */ + if (xfrm_attrs->flags & MLX5_ACCEL_ESP_FLAGS_ESN_TRIGGERED) { + hw_sa->ipsec_sa_v1.flags |= MLX5_FPGA_IPSEC_SA_ESN_EN; + hw_sa->ipsec_sa_v1.flags |= + (xfrm_attrs->flags & + MLX5_ACCEL_ESP_FLAGS_ESN_STATE_OVERLAP) ? + MLX5_FPGA_IPSEC_SA_ESN_OVERLAP : 0; + hw_sa->esn = htonl(xfrm_attrs->esn); + } else { + hw_sa->ipsec_sa_v1.flags &= ~MLX5_FPGA_IPSEC_SA_ESN_EN; + hw_sa->ipsec_sa_v1.flags &= + ~(xfrm_attrs->flags & + MLX5_ACCEL_ESP_FLAGS_ESN_STATE_OVERLAP) ? + MLX5_FPGA_IPSEC_SA_ESN_OVERLAP : 0; + hw_sa->esn = 0; + } + + /* rx handle */ + hw_sa->ipsec_sa_v1.sw_sa_handle = htonl(xfrm_attrs->sa_handle); + + /* enc mode */ + switch (aes_gcm->key_len) { + case 128: + hw_sa->ipsec_sa_v1.enc_mode = + MLX5_FPGA_IPSEC_SA_ENC_MODE_AES_GCM_128_AUTH_128; + break; + case 256: + hw_sa->ipsec_sa_v1.enc_mode = + MLX5_FPGA_IPSEC_SA_ENC_MODE_AES_GCM_256_AUTH_128; + break; + } + + /* flags */ + hw_sa->ipsec_sa_v1.flags |= MLX5_FPGA_IPSEC_SA_SA_VALID | + MLX5_FPGA_IPSEC_SA_SPI_EN | + MLX5_FPGA_IPSEC_SA_IP_ESP; + + if (xfrm_attrs->action & MLX5_ACCEL_ESP_ACTION_ENCRYPT) + hw_sa->ipsec_sa_v1.flags |= MLX5_FPGA_IPSEC_SA_DIR_SX; + else + hw_sa->ipsec_sa_v1.flags &= ~MLX5_FPGA_IPSEC_SA_DIR_SX; +} + +static void +mlx5_fpga_ipsec_build_hw_sa(struct mlx5_core_dev *mdev, + struct mlx5_accel_esp_xfrm_attrs *xfrm_attrs, + const __be32 saddr[4], + const __be32 daddr[4], + const __be32 spi, bool is_ipv6, + struct mlx5_ifc_fpga_ipsec_sa *hw_sa) +{ + mlx5_fpga_ipsec_build_hw_xfrm(mdev, xfrm_attrs, hw_sa); + + /* IPs */ + memcpy(hw_sa->ipsec_sa_v1.sip, saddr, sizeof(hw_sa->ipsec_sa_v1.sip)); + memcpy(hw_sa->ipsec_sa_v1.dip, daddr, sizeof(hw_sa->ipsec_sa_v1.dip)); + + /* SPI */ + hw_sa->ipsec_sa_v1.spi = spi; + + /* flags */ + if (is_ipv6) + hw_sa->ipsec_sa_v1.flags |= MLX5_FPGA_IPSEC_SA_IPV6; +} + +static bool is_full_mask(const void *p, size_t len) +{ + WARN_ON(len % 4); + + return !memchr_inv(p, 0xff, len); +} + +static bool validate_fpga_full_mask(struct mlx5_core_dev *dev, + const u32 *match_c, + const u32 *match_v) +{ + const void *misc_params_c = MLX5_ADDR_OF(fte_match_param, + match_c, + misc_parameters); + const void *headers_c = MLX5_ADDR_OF(fte_match_param, + match_c, + outer_headers); + const void *headers_v = MLX5_ADDR_OF(fte_match_param, + match_v, + outer_headers); + + if (mlx5_fs_is_outer_ipv4_flow(dev, headers_c, headers_v)) { + const void *s_ipv4_c = MLX5_ADDR_OF(fte_match_set_lyr_2_4, + headers_c, + src_ipv4_src_ipv6.ipv4_layout.ipv4); + const void *d_ipv4_c = MLX5_ADDR_OF(fte_match_set_lyr_2_4, + headers_c, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4); + + if (!is_full_mask(s_ipv4_c, MLX5_FLD_SZ_BYTES(ipv4_layout, + ipv4)) || + !is_full_mask(d_ipv4_c, MLX5_FLD_SZ_BYTES(ipv4_layout, + ipv4))) + return false; + } else { + const void *s_ipv6_c = MLX5_ADDR_OF(fte_match_set_lyr_2_4, + headers_c, + src_ipv4_src_ipv6.ipv6_layout.ipv6); + const void *d_ipv6_c = MLX5_ADDR_OF(fte_match_set_lyr_2_4, + headers_c, + dst_ipv4_dst_ipv6.ipv6_layout.ipv6); + + if (!is_full_mask(s_ipv6_c, MLX5_FLD_SZ_BYTES(ipv6_layout, + ipv6)) || + !is_full_mask(d_ipv6_c, MLX5_FLD_SZ_BYTES(ipv6_layout, + ipv6))) + return false; + } + + if (!is_full_mask(MLX5_ADDR_OF(fte_match_set_misc, misc_params_c, + outer_esp_spi), + MLX5_FLD_SZ_BYTES(fte_match_set_misc, outer_esp_spi))) + return false; + + return true; +} + +static bool mlx5_is_fpga_ipsec_rule(struct mlx5_core_dev *dev, + u8 match_criteria_enable, + const u32 *match_c, + const u32 *match_v) +{ + u32 ipsec_dev_caps = mlx5_accel_ipsec_device_caps(dev); + bool ipv6_flow; + + ipv6_flow = mlx5_fs_is_outer_ipv6_flow(dev, match_c, match_v); + + if (!(match_criteria_enable & MLX5_MATCH_OUTER_HEADERS) || + mlx5_fs_is_outer_udp_flow(match_c, match_v) || + mlx5_fs_is_outer_tcp_flow(match_c, match_v) || + mlx5_fs_is_vxlan_flow(match_c) || + !(mlx5_fs_is_outer_ipv4_flow(dev, match_c, match_v) || + ipv6_flow)) + return false; + + if (!(ipsec_dev_caps & MLX5_ACCEL_IPSEC_CAP_DEVICE)) + return false; + + if (!(ipsec_dev_caps & MLX5_ACCEL_IPSEC_CAP_ESP) && + mlx5_fs_is_outer_ipsec_flow(match_c)) + return false; + + if (!(ipsec_dev_caps & MLX5_ACCEL_IPSEC_CAP_IPV6) && + ipv6_flow) + return false; + + if (!validate_fpga_full_mask(dev, match_c, match_v)) + return false; + + return true; +} + +static bool mlx5_is_fpga_egress_ipsec_rule(struct mlx5_core_dev *dev, + u8 match_criteria_enable, + const u32 *match_c, + const u32 *match_v, + struct mlx5_flow_act *flow_act) +{ + const void *outer_c = MLX5_ADDR_OF(fte_match_param, match_c, + outer_headers); + bool is_dmac = MLX5_GET(fte_match_set_lyr_2_4, outer_c, dmac_47_16) || + MLX5_GET(fte_match_set_lyr_2_4, outer_c, dmac_15_0); + bool is_smac = MLX5_GET(fte_match_set_lyr_2_4, outer_c, smac_47_16) || + MLX5_GET(fte_match_set_lyr_2_4, outer_c, smac_15_0); + int ret; + + ret = mlx5_is_fpga_ipsec_rule(dev, match_criteria_enable, match_c, + match_v); + if (!ret) + return ret; + + if (is_dmac || is_smac || + (match_criteria_enable & + ~(MLX5_MATCH_OUTER_HEADERS | MLX5_MATCH_MISC_PARAMETERS)) || + (flow_act->action & ~(MLX5_FLOW_CONTEXT_ACTION_ENCRYPT | MLX5_FLOW_CONTEXT_ACTION_ALLOW)) || + flow_act->has_flow_tag) + return false; + + return true; +} + +void *mlx5_fpga_ipsec_create_sa_ctx(struct mlx5_core_dev *mdev, + struct mlx5_accel_esp_xfrm *accel_xfrm, + const __be32 saddr[4], + const __be32 daddr[4], + const __be32 spi, bool is_ipv6) +{ + struct mlx5_fpga_ipsec_sa_ctx *sa_ctx; + struct mlx5_fpga_esp_xfrm *fpga_xfrm = + container_of(accel_xfrm, typeof(*fpga_xfrm), + accel_xfrm); + struct mlx5_fpga_device *fdev = mdev->fpga; + struct mlx5_fpga_ipsec *fipsec = fdev->ipsec; + int opcode, err; + void *context; + + /* alloc SA */ + sa_ctx = kzalloc(sizeof(*sa_ctx), GFP_KERNEL); + if (!sa_ctx) + return ERR_PTR(-ENOMEM); + + sa_ctx->dev = mdev; + + /* build candidate SA */ + mlx5_fpga_ipsec_build_hw_sa(mdev, &accel_xfrm->attrs, + saddr, daddr, spi, is_ipv6, + &sa_ctx->hw_sa); + + mutex_lock(&fpga_xfrm->lock); + + if (fpga_xfrm->sa_ctx) { /* multiple rules for same accel_xfrm */ + /* all rules must be with same IPs and SPI */ + if (memcmp(&sa_ctx->hw_sa, &fpga_xfrm->sa_ctx->hw_sa, + sizeof(sa_ctx->hw_sa))) { + context = ERR_PTR(-EINVAL); + goto exists; + } + + ++fpga_xfrm->num_rules; + context = fpga_xfrm->sa_ctx; + goto exists; + } + + /* This is unbounded fpga_xfrm, try to add to hash */ + mutex_lock(&fipsec->sa_hash_lock); + + err = rhashtable_lookup_insert_fast(&fipsec->sa_hash, &sa_ctx->hash, + rhash_sa); + if (err) { + /* Can't bound different accel_xfrm to already existing sa_ctx. + * This is because we can't support multiple ketmats for + * same IPs and SPI + */ + context = ERR_PTR(-EEXIST); + goto unlock_hash; + } + + /* Bound accel_xfrm to sa_ctx */ + opcode = is_v2_sadb_supported(fdev->ipsec) ? + MLX5_FPGA_IPSEC_CMD_OP_ADD_SA_V2 : + MLX5_FPGA_IPSEC_CMD_OP_ADD_SA; + err = mlx5_fpga_ipsec_update_hw_sa(fdev, &sa_ctx->hw_sa, opcode); + sa_ctx->hw_sa.ipsec_sa_v1.cmd = 0; + if (err) { + context = ERR_PTR(err); + goto delete_hash; + } + + mutex_unlock(&fipsec->sa_hash_lock); + + ++fpga_xfrm->num_rules; + fpga_xfrm->sa_ctx = sa_ctx; + sa_ctx->fpga_xfrm = fpga_xfrm; + + mutex_unlock(&fpga_xfrm->lock); + + return sa_ctx; + +delete_hash: + WARN_ON(rhashtable_remove_fast(&fipsec->sa_hash, &sa_ctx->hash, + rhash_sa)); +unlock_hash: + mutex_unlock(&fipsec->sa_hash_lock); + +exists: + mutex_unlock(&fpga_xfrm->lock); + kfree(sa_ctx); + return context; +} + +static void * +mlx5_fpga_ipsec_fs_create_sa_ctx(struct mlx5_core_dev *mdev, + struct fs_fte *fte, + bool is_egress) +{ + struct mlx5_accel_esp_xfrm *accel_xfrm; + __be32 saddr[4], daddr[4], spi; + struct mlx5_flow_group *fg; + bool is_ipv6 = false; + + fs_get_obj(fg, fte->node.parent); + /* validate */ + if (is_egress && + !mlx5_is_fpga_egress_ipsec_rule(mdev, + fg->mask.match_criteria_enable, + fg->mask.match_criteria, + fte->val, + &fte->action)) + return ERR_PTR(-EINVAL); + else if (!mlx5_is_fpga_ipsec_rule(mdev, + fg->mask.match_criteria_enable, + fg->mask.match_criteria, + fte->val)) + return ERR_PTR(-EINVAL); + + /* get xfrm context */ + accel_xfrm = + (struct mlx5_accel_esp_xfrm *)fte->action.esp_id; + + /* IPs */ + if (mlx5_fs_is_outer_ipv4_flow(mdev, fg->mask.match_criteria, + fte->val)) { + memcpy(&saddr[3], + MLX5_ADDR_OF(fte_match_set_lyr_2_4, + fte->val, + src_ipv4_src_ipv6.ipv4_layout.ipv4), + sizeof(saddr[3])); + memcpy(&daddr[3], + MLX5_ADDR_OF(fte_match_set_lyr_2_4, + fte->val, + dst_ipv4_dst_ipv6.ipv4_layout.ipv4), + sizeof(daddr[3])); + } else { + memcpy(saddr, + MLX5_ADDR_OF(fte_match_param, + fte->val, + outer_headers.src_ipv4_src_ipv6.ipv6_layout.ipv6), + sizeof(saddr)); + memcpy(daddr, + MLX5_ADDR_OF(fte_match_param, + fte->val, + outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6), + sizeof(daddr)); + is_ipv6 = true; + } + + /* SPI */ + spi = MLX5_GET_BE(typeof(spi), + fte_match_param, fte->val, + misc_parameters.outer_esp_spi); + + /* create */ + return mlx5_fpga_ipsec_create_sa_ctx(mdev, accel_xfrm, + saddr, daddr, + spi, is_ipv6); +} + +static void +mlx5_fpga_ipsec_release_sa_ctx(struct mlx5_fpga_ipsec_sa_ctx *sa_ctx) +{ + struct mlx5_fpga_device *fdev = sa_ctx->dev->fpga; + struct mlx5_fpga_ipsec *fipsec = fdev->ipsec; + int opcode = is_v2_sadb_supported(fdev->ipsec) ? + MLX5_FPGA_IPSEC_CMD_OP_DEL_SA_V2 : + MLX5_FPGA_IPSEC_CMD_OP_DEL_SA; + int err; + + err = mlx5_fpga_ipsec_update_hw_sa(fdev, &sa_ctx->hw_sa, opcode); + sa_ctx->hw_sa.ipsec_sa_v1.cmd = 0; + if (err) { + WARN_ON(err); + return; + } + + mutex_lock(&fipsec->sa_hash_lock); + WARN_ON(rhashtable_remove_fast(&fipsec->sa_hash, &sa_ctx->hash, + rhash_sa)); + mutex_unlock(&fipsec->sa_hash_lock); +} + +void mlx5_fpga_ipsec_delete_sa_ctx(void *context) +{ + struct mlx5_fpga_esp_xfrm *fpga_xfrm = + ((struct mlx5_fpga_ipsec_sa_ctx *)context)->fpga_xfrm; + + mutex_lock(&fpga_xfrm->lock); + if (!--fpga_xfrm->num_rules) { + mlx5_fpga_ipsec_release_sa_ctx(fpga_xfrm->sa_ctx); + fpga_xfrm->sa_ctx = NULL; + } + mutex_unlock(&fpga_xfrm->lock); +} + +static inline struct mlx5_fpga_ipsec_rule * +_rule_search(struct rb_root *root, struct fs_fte *fte) +{ + struct rb_node *node = root->rb_node; + + while (node) { + struct mlx5_fpga_ipsec_rule *rule = + container_of(node, struct mlx5_fpga_ipsec_rule, + node); + + if (rule->fte < fte) + node = node->rb_left; + else if (rule->fte > fte) + node = node->rb_right; + else + return rule; + } + return NULL; +} + +static struct mlx5_fpga_ipsec_rule * +rule_search(struct mlx5_fpga_ipsec *ipsec_dev, struct fs_fte *fte) +{ + struct mlx5_fpga_ipsec_rule *rule; + + mutex_lock(&ipsec_dev->rules_rb_lock); + rule = _rule_search(&ipsec_dev->rules_rb, fte); + mutex_unlock(&ipsec_dev->rules_rb_lock); + + return rule; +} + +static inline int _rule_insert(struct rb_root *root, + struct mlx5_fpga_ipsec_rule *rule) +{ + struct rb_node **new = &root->rb_node, *parent = NULL; + + /* Figure out where to put new node */ + while (*new) { + struct mlx5_fpga_ipsec_rule *this = + container_of(*new, struct mlx5_fpga_ipsec_rule, + node); + + parent = *new; + if (rule->fte < this->fte) + new = &((*new)->rb_left); + else if (rule->fte > this->fte) + new = &((*new)->rb_right); + else + return -EEXIST; + } + + /* Add new node and rebalance tree. */ + rb_link_node(&rule->node, parent, new); + rb_insert_color(&rule->node, root); + + return 0; +} + +static int rule_insert(struct mlx5_fpga_ipsec *ipsec_dev, + struct mlx5_fpga_ipsec_rule *rule) +{ + int ret; + + mutex_lock(&ipsec_dev->rules_rb_lock); + ret = _rule_insert(&ipsec_dev->rules_rb, rule); + mutex_unlock(&ipsec_dev->rules_rb_lock); + + return ret; +} + +static inline void _rule_delete(struct mlx5_fpga_ipsec *ipsec_dev, + struct mlx5_fpga_ipsec_rule *rule) +{ + struct rb_root *root = &ipsec_dev->rules_rb; + + mutex_lock(&ipsec_dev->rules_rb_lock); + rb_erase(&rule->node, root); + mutex_unlock(&ipsec_dev->rules_rb_lock); +} + +static void rule_delete(struct mlx5_fpga_ipsec *ipsec_dev, + struct mlx5_fpga_ipsec_rule *rule) +{ + _rule_delete(ipsec_dev, rule); + kfree(rule); +} + +struct mailbox_mod { + uintptr_t saved_esp_id; + u32 saved_action; + u32 saved_outer_esp_spi_value; +}; + +static void restore_spec_mailbox(struct fs_fte *fte, + struct mailbox_mod *mbox_mod) +{ + char *misc_params_v = MLX5_ADDR_OF(fte_match_param, + fte->val, + misc_parameters); + + MLX5_SET(fte_match_set_misc, misc_params_v, outer_esp_spi, + mbox_mod->saved_outer_esp_spi_value); + fte->action.action |= mbox_mod->saved_action; + fte->action.esp_id = (uintptr_t)mbox_mod->saved_esp_id; +} + +static void modify_spec_mailbox(struct mlx5_core_dev *mdev, + struct fs_fte *fte, + struct mailbox_mod *mbox_mod) +{ + char *misc_params_v = MLX5_ADDR_OF(fte_match_param, + fte->val, + misc_parameters); + + mbox_mod->saved_esp_id = fte->action.esp_id; + mbox_mod->saved_action = fte->action.action & + (MLX5_FLOW_CONTEXT_ACTION_ENCRYPT | + MLX5_FLOW_CONTEXT_ACTION_DECRYPT); + mbox_mod->saved_outer_esp_spi_value = + MLX5_GET(fte_match_set_misc, misc_params_v, + outer_esp_spi); + + fte->action.esp_id = 0; + fte->action.action &= ~(MLX5_FLOW_CONTEXT_ACTION_ENCRYPT | + MLX5_FLOW_CONTEXT_ACTION_DECRYPT); + if (!MLX5_CAP_FLOWTABLE(mdev, + flow_table_properties_nic_receive.ft_field_support.outer_esp_spi)) + MLX5_SET(fte_match_set_misc, misc_params_v, outer_esp_spi, 0); +} + +static enum fs_flow_table_type egress_to_fs_ft(bool egress) +{ + return egress ? FS_FT_NIC_TX : FS_FT_NIC_RX; +} + +static int fpga_ipsec_fs_create_flow_group(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + u32 *in, + unsigned int *group_id, + bool is_egress) +{ + int (*create_flow_group)(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, u32 *in, + unsigned int *group_id) = + mlx5_fs_cmd_get_default(egress_to_fs_ft(is_egress))->create_flow_group; + char *misc_params_c = MLX5_ADDR_OF(create_flow_group_in, in, + match_criteria.misc_parameters); + u32 saved_outer_esp_spi_mask; + u8 match_criteria_enable; + int ret; + + if (MLX5_CAP_FLOWTABLE(dev, + flow_table_properties_nic_receive.ft_field_support.outer_esp_spi)) + return create_flow_group(dev, ft, in, group_id); + + match_criteria_enable = + MLX5_GET(create_flow_group_in, in, match_criteria_enable); + saved_outer_esp_spi_mask = + MLX5_GET(fte_match_set_misc, misc_params_c, outer_esp_spi); + if (!match_criteria_enable || !saved_outer_esp_spi_mask) + return create_flow_group(dev, ft, in, group_id); + + MLX5_SET(fte_match_set_misc, misc_params_c, outer_esp_spi, 0); + + if (!(*misc_params_c) && + !memcmp(misc_params_c, misc_params_c + 1, MLX5_ST_SZ_BYTES(fte_match_set_misc) - 1)) + MLX5_SET(create_flow_group_in, in, match_criteria_enable, + match_criteria_enable & ~MLX5_MATCH_MISC_PARAMETERS); + + ret = create_flow_group(dev, ft, in, group_id); + + MLX5_SET(fte_match_set_misc, misc_params_c, outer_esp_spi, saved_outer_esp_spi_mask); + MLX5_SET(create_flow_group_in, in, match_criteria_enable, match_criteria_enable); + + return ret; +} + +static int fpga_ipsec_fs_create_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct mlx5_flow_group *fg, + struct fs_fte *fte, + bool is_egress) +{ + int (*create_fte)(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct mlx5_flow_group *fg, + struct fs_fte *fte) = + mlx5_fs_cmd_get_default(egress_to_fs_ft(is_egress))->create_fte; + struct mlx5_fpga_device *fdev = dev->fpga; + struct mlx5_fpga_ipsec *fipsec = fdev->ipsec; + struct mlx5_fpga_ipsec_rule *rule; + bool is_esp = fte->action.esp_id; + struct mailbox_mod mbox_mod; + int ret; + + if (!is_esp || + !(fte->action.action & + (MLX5_FLOW_CONTEXT_ACTION_ENCRYPT | + MLX5_FLOW_CONTEXT_ACTION_DECRYPT))) + return create_fte(dev, ft, fg, fte); + + rule = kzalloc(sizeof(*rule), GFP_KERNEL); + if (!rule) + return -ENOMEM; + + rule->ctx = mlx5_fpga_ipsec_fs_create_sa_ctx(dev, fte, is_egress); + if (IS_ERR(rule->ctx)) { + int err = PTR_ERR(rule->ctx); + kfree(rule); + return err; + } + + rule->fte = fte; + WARN_ON(rule_insert(fipsec, rule)); + + modify_spec_mailbox(dev, fte, &mbox_mod); + ret = create_fte(dev, ft, fg, fte); + restore_spec_mailbox(fte, &mbox_mod); + if (ret) { + _rule_delete(fipsec, rule); + mlx5_fpga_ipsec_delete_sa_ctx(rule->ctx); + kfree(rule); + } + + return ret; +} + +static int fpga_ipsec_fs_update_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int group_id, + int modify_mask, + struct fs_fte *fte, + bool is_egress) +{ + int (*update_fte)(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int group_id, + int modify_mask, + struct fs_fte *fte) = + mlx5_fs_cmd_get_default(egress_to_fs_ft(is_egress))->update_fte; + bool is_esp = fte->action.esp_id; + struct mailbox_mod mbox_mod; + int ret; + + if (!is_esp || + !(fte->action.action & + (MLX5_FLOW_CONTEXT_ACTION_ENCRYPT | + MLX5_FLOW_CONTEXT_ACTION_DECRYPT))) + return update_fte(dev, ft, group_id, modify_mask, fte); + + modify_spec_mailbox(dev, fte, &mbox_mod); + ret = update_fte(dev, ft, group_id, modify_mask, fte); + restore_spec_mailbox(fte, &mbox_mod); + + return ret; +} + +static int fpga_ipsec_fs_delete_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct fs_fte *fte, + bool is_egress) +{ + int (*delete_fte)(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct fs_fte *fte) = + mlx5_fs_cmd_get_default(egress_to_fs_ft(is_egress))->delete_fte; + struct mlx5_fpga_device *fdev = dev->fpga; + struct mlx5_fpga_ipsec *fipsec = fdev->ipsec; + struct mlx5_fpga_ipsec_rule *rule; + bool is_esp = fte->action.esp_id; + struct mailbox_mod mbox_mod; + int ret; + + if (!is_esp || + !(fte->action.action & + (MLX5_FLOW_CONTEXT_ACTION_ENCRYPT | + MLX5_FLOW_CONTEXT_ACTION_DECRYPT))) + return delete_fte(dev, ft, fte); + + rule = rule_search(fipsec, fte); + if (!rule) + return -ENOENT; + + mlx5_fpga_ipsec_delete_sa_ctx(rule->ctx); + rule_delete(fipsec, rule); + + modify_spec_mailbox(dev, fte, &mbox_mod); + ret = delete_fte(dev, ft, fte); + restore_spec_mailbox(fte, &mbox_mod); + + return ret; +} + +static int +mlx5_fpga_ipsec_fs_create_flow_group_egress(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + u32 *in, + unsigned int *group_id) +{ + return fpga_ipsec_fs_create_flow_group(dev, ft, in, group_id, true); +} + +static int +mlx5_fpga_ipsec_fs_create_fte_egress(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct mlx5_flow_group *fg, + struct fs_fte *fte) +{ + return fpga_ipsec_fs_create_fte(dev, ft, fg, fte, true); +} + +static int +mlx5_fpga_ipsec_fs_update_fte_egress(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int group_id, + int modify_mask, + struct fs_fte *fte) +{ + return fpga_ipsec_fs_update_fte(dev, ft, group_id, modify_mask, fte, + true); +} + +static int +mlx5_fpga_ipsec_fs_delete_fte_egress(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct fs_fte *fte) +{ + return fpga_ipsec_fs_delete_fte(dev, ft, fte, true); +} + +static int +mlx5_fpga_ipsec_fs_create_flow_group_ingress(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + u32 *in, + unsigned int *group_id) +{ + return fpga_ipsec_fs_create_flow_group(dev, ft, in, group_id, false); +} + +static int +mlx5_fpga_ipsec_fs_create_fte_ingress(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct mlx5_flow_group *fg, + struct fs_fte *fte) +{ + return fpga_ipsec_fs_create_fte(dev, ft, fg, fte, false); +} + +static int +mlx5_fpga_ipsec_fs_update_fte_ingress(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int group_id, + int modify_mask, + struct fs_fte *fte) +{ + return fpga_ipsec_fs_update_fte(dev, ft, group_id, modify_mask, fte, + false); +} + +static int +mlx5_fpga_ipsec_fs_delete_fte_ingress(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct fs_fte *fte) +{ + return fpga_ipsec_fs_delete_fte(dev, ft, fte, false); +} + +static struct mlx5_flow_cmds fpga_ipsec_ingress; +static struct mlx5_flow_cmds fpga_ipsec_egress; + +const struct mlx5_flow_cmds *mlx5_fs_cmd_get_default_ipsec_fpga_cmds(enum fs_flow_table_type type) +{ + switch (type) { + case FS_FT_NIC_RX: + return &fpga_ipsec_ingress; + case FS_FT_NIC_TX: + return &fpga_ipsec_egress; + default: + WARN_ON(true); + return NULL; + } +} + int mlx5_fpga_ipsec_init(struct mlx5_core_dev *mdev) { struct mlx5_fpga_conn_attr init_attr = {0}; @@ -332,6 +1251,8 @@ int mlx5_fpga_ipsec_init(struct mlx5_core_dev *mdev) if (!fdev->ipsec) return -ENOMEM; + fdev->ipsec->fdev = fdev; + err = mlx5_fpga_get_sbu_caps(fdev, sizeof(fdev->ipsec->caps), fdev->ipsec->caps); if (err) { @@ -355,14 +1276,47 @@ int mlx5_fpga_ipsec_init(struct mlx5_core_dev *mdev) goto error; } fdev->ipsec->conn = conn; + + err = rhashtable_init(&fdev->ipsec->sa_hash, &rhash_sa); + if (err) + goto err_destroy_conn; + mutex_init(&fdev->ipsec->sa_hash_lock); + + fdev->ipsec->rules_rb = RB_ROOT; + mutex_init(&fdev->ipsec->rules_rb_lock); + + err = mlx5_fpga_ipsec_enable_supported_caps(mdev); + if (err) { + mlx5_fpga_err(fdev, "Failed to enable IPSec extended capabilities: %d\n", + err); + goto err_destroy_hash; + } + return 0; +err_destroy_hash: + rhashtable_destroy(&fdev->ipsec->sa_hash); + +err_destroy_conn: + mlx5_fpga_sbu_conn_destroy(conn); + error: kfree(fdev->ipsec); fdev->ipsec = NULL; return err; } +static void destroy_rules_rb(struct rb_root *root) +{ + struct mlx5_fpga_ipsec_rule *r, *tmp; + + rbtree_postorder_for_each_entry_safe(r, tmp, root, node) { + rb_erase(&r->node, root); + mlx5_fpga_ipsec_delete_sa_ctx(r->ctx); + kfree(r); + } +} + void mlx5_fpga_ipsec_cleanup(struct mlx5_core_dev *mdev) { struct mlx5_fpga_device *fdev = mdev->fpga; @@ -370,7 +1324,209 @@ void mlx5_fpga_ipsec_cleanup(struct mlx5_core_dev *mdev) if (!mlx5_fpga_is_ipsec_device(mdev)) return; + destroy_rules_rb(&fdev->ipsec->rules_rb); + rhashtable_destroy(&fdev->ipsec->sa_hash); + mlx5_fpga_sbu_conn_destroy(fdev->ipsec->conn); kfree(fdev->ipsec); fdev->ipsec = NULL; } + +void mlx5_fpga_ipsec_build_fs_cmds(void) +{ + /* ingress */ + fpga_ipsec_ingress.create_flow_table = + mlx5_fs_cmd_get_default(egress_to_fs_ft(false))->create_flow_table; + fpga_ipsec_ingress.destroy_flow_table = + mlx5_fs_cmd_get_default(egress_to_fs_ft(false))->destroy_flow_table; + fpga_ipsec_ingress.modify_flow_table = + mlx5_fs_cmd_get_default(egress_to_fs_ft(false))->modify_flow_table; + fpga_ipsec_ingress.create_flow_group = + mlx5_fpga_ipsec_fs_create_flow_group_ingress; + fpga_ipsec_ingress.destroy_flow_group = + mlx5_fs_cmd_get_default(egress_to_fs_ft(false))->destroy_flow_group; + fpga_ipsec_ingress.create_fte = + mlx5_fpga_ipsec_fs_create_fte_ingress; + fpga_ipsec_ingress.update_fte = + mlx5_fpga_ipsec_fs_update_fte_ingress; + fpga_ipsec_ingress.delete_fte = + mlx5_fpga_ipsec_fs_delete_fte_ingress; + fpga_ipsec_ingress.update_root_ft = + mlx5_fs_cmd_get_default(egress_to_fs_ft(false))->update_root_ft; + + /* egress */ + fpga_ipsec_egress.create_flow_table = + mlx5_fs_cmd_get_default(egress_to_fs_ft(true))->create_flow_table; + fpga_ipsec_egress.destroy_flow_table = + mlx5_fs_cmd_get_default(egress_to_fs_ft(true))->destroy_flow_table; + fpga_ipsec_egress.modify_flow_table = + mlx5_fs_cmd_get_default(egress_to_fs_ft(true))->modify_flow_table; + fpga_ipsec_egress.create_flow_group = + mlx5_fpga_ipsec_fs_create_flow_group_egress; + fpga_ipsec_egress.destroy_flow_group = + mlx5_fs_cmd_get_default(egress_to_fs_ft(true))->destroy_flow_group; + fpga_ipsec_egress.create_fte = + mlx5_fpga_ipsec_fs_create_fte_egress; + fpga_ipsec_egress.update_fte = + mlx5_fpga_ipsec_fs_update_fte_egress; + fpga_ipsec_egress.delete_fte = + mlx5_fpga_ipsec_fs_delete_fte_egress; + fpga_ipsec_egress.update_root_ft = + mlx5_fs_cmd_get_default(egress_to_fs_ft(true))->update_root_ft; +} + +static int +mlx5_fpga_esp_validate_xfrm_attrs(struct mlx5_core_dev *mdev, + const struct mlx5_accel_esp_xfrm_attrs *attrs) +{ + if (attrs->tfc_pad) { + mlx5_core_err(mdev, "Cannot offload xfrm states with tfc padding\n"); + return -EOPNOTSUPP; + } + + if (attrs->replay_type != MLX5_ACCEL_ESP_REPLAY_NONE) { + mlx5_core_err(mdev, "Cannot offload xfrm states with anti replay\n"); + return -EOPNOTSUPP; + } + + if (attrs->keymat_type != MLX5_ACCEL_ESP_KEYMAT_AES_GCM) { + mlx5_core_err(mdev, "Only aes gcm keymat is supported\n"); + return -EOPNOTSUPP; + } + + if (attrs->keymat.aes_gcm.iv_algo != + MLX5_ACCEL_ESP_AES_GCM_IV_ALGO_SEQ) { + mlx5_core_err(mdev, "Only iv sequence algo is supported\n"); + return -EOPNOTSUPP; + } + + if (attrs->keymat.aes_gcm.icv_len != 128) { + mlx5_core_err(mdev, "Cannot offload xfrm states with AEAD ICV length other than 128bit\n"); + return -EOPNOTSUPP; + } + + if (attrs->keymat.aes_gcm.key_len != 128 && + attrs->keymat.aes_gcm.key_len != 256) { + mlx5_core_err(mdev, "Cannot offload xfrm states with AEAD key length other than 128/256 bit\n"); + return -EOPNOTSUPP; + } + + if ((attrs->flags & MLX5_ACCEL_ESP_FLAGS_ESN_TRIGGERED) && + (!MLX5_GET(ipsec_extended_cap, mdev->fpga->ipsec->caps, + v2_command))) { + mlx5_core_err(mdev, "Cannot offload xfrm states with AEAD key length other than 128/256 bit\n"); + return -EOPNOTSUPP; + } + + return 0; +} + +struct mlx5_accel_esp_xfrm * +mlx5_fpga_esp_create_xfrm(struct mlx5_core_dev *mdev, + const struct mlx5_accel_esp_xfrm_attrs *attrs, + u32 flags) +{ + struct mlx5_fpga_esp_xfrm *fpga_xfrm; + + if (!(flags & MLX5_ACCEL_XFRM_FLAG_REQUIRE_METADATA)) { + mlx5_core_warn(mdev, "Tried to create an esp action without metadata\n"); + return ERR_PTR(-EINVAL); + } + + if (mlx5_fpga_esp_validate_xfrm_attrs(mdev, attrs)) { + mlx5_core_warn(mdev, "Tried to create an esp with unsupported attrs\n"); + return ERR_PTR(-EOPNOTSUPP); + } + + fpga_xfrm = kzalloc(sizeof(*fpga_xfrm), GFP_KERNEL); + if (!fpga_xfrm) + return ERR_PTR(-ENOMEM); + + mutex_init(&fpga_xfrm->lock); + memcpy(&fpga_xfrm->accel_xfrm.attrs, attrs, + sizeof(fpga_xfrm->accel_xfrm.attrs)); + + return &fpga_xfrm->accel_xfrm; +} + +void mlx5_fpga_esp_destroy_xfrm(struct mlx5_accel_esp_xfrm *xfrm) +{ + struct mlx5_fpga_esp_xfrm *fpga_xfrm = + container_of(xfrm, struct mlx5_fpga_esp_xfrm, + accel_xfrm); + /* assuming no sa_ctx are connected to this xfrm_ctx */ + kfree(fpga_xfrm); +} + +int mlx5_fpga_esp_modify_xfrm(struct mlx5_accel_esp_xfrm *xfrm, + const struct mlx5_accel_esp_xfrm_attrs *attrs) +{ + struct mlx5_core_dev *mdev = xfrm->mdev; + struct mlx5_fpga_device *fdev = mdev->fpga; + struct mlx5_fpga_ipsec *fipsec = fdev->ipsec; + struct mlx5_fpga_esp_xfrm *fpga_xfrm; + struct mlx5_ifc_fpga_ipsec_sa org_hw_sa; + + int err = 0; + + if (!memcmp(&xfrm->attrs, attrs, sizeof(xfrm->attrs))) + return 0; + + if (!mlx5_fpga_esp_validate_xfrm_attrs(mdev, attrs)) { + mlx5_core_warn(mdev, "Tried to create an esp with unsupported attrs\n"); + return -EOPNOTSUPP; + } + + if (is_v2_sadb_supported(fipsec)) { + mlx5_core_warn(mdev, "Modify esp is not supported\n"); + return -EOPNOTSUPP; + } + + fpga_xfrm = container_of(xfrm, struct mlx5_fpga_esp_xfrm, accel_xfrm); + + mutex_lock(&fpga_xfrm->lock); + + if (!fpga_xfrm->sa_ctx) + /* Unbounded xfrm, chane only sw attrs */ + goto change_sw_xfrm_attrs; + + /* copy original hw sa */ + memcpy(&org_hw_sa, &fpga_xfrm->sa_ctx->hw_sa, sizeof(org_hw_sa)); + mutex_lock(&fipsec->sa_hash_lock); + /* remove original hw sa from hash */ + WARN_ON(rhashtable_remove_fast(&fipsec->sa_hash, + &fpga_xfrm->sa_ctx->hash, rhash_sa)); + /* update hw_sa with new xfrm attrs*/ + mlx5_fpga_ipsec_build_hw_xfrm(xfrm->mdev, attrs, + &fpga_xfrm->sa_ctx->hw_sa); + /* try to insert new hw_sa to hash */ + err = rhashtable_insert_fast(&fipsec->sa_hash, + &fpga_xfrm->sa_ctx->hash, rhash_sa); + if (err) + goto rollback_sa; + + /* modify device with new hw_sa */ + err = mlx5_fpga_ipsec_update_hw_sa(fdev, &fpga_xfrm->sa_ctx->hw_sa, + MLX5_FPGA_IPSEC_CMD_OP_MOD_SA_V2); + fpga_xfrm->sa_ctx->hw_sa.ipsec_sa_v1.cmd = 0; + if (err) + WARN_ON(rhashtable_remove_fast(&fipsec->sa_hash, + &fpga_xfrm->sa_ctx->hash, + rhash_sa)); +rollback_sa: + if (err) { + /* return original hw_sa to hash */ + memcpy(&fpga_xfrm->sa_ctx->hw_sa, &org_hw_sa, + sizeof(org_hw_sa)); + WARN_ON(rhashtable_insert_fast(&fipsec->sa_hash, + &fpga_xfrm->sa_ctx->hash, + rhash_sa)); + } + mutex_unlock(&fipsec->sa_hash_lock); + +change_sw_xfrm_attrs: + if (!err) + memcpy(&xfrm->attrs, attrs, sizeof(xfrm->attrs)); + mutex_unlock(&fpga_xfrm->lock); + return err; +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.h index 26a3e4b56972..2b5e63b0d4d6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.h @@ -35,33 +35,38 @@ #define __MLX5_FPGA_IPSEC_H__ #include "accel/ipsec.h" +#include "fs_cmd.h" #ifdef CONFIG_MLX5_FPGA -void *mlx5_fpga_ipsec_sa_cmd_exec(struct mlx5_core_dev *mdev, - struct mlx5_accel_ipsec_sa *cmd); -int mlx5_fpga_ipsec_sa_cmd_wait(void *context); - u32 mlx5_fpga_ipsec_device_caps(struct mlx5_core_dev *mdev); unsigned int mlx5_fpga_ipsec_counters_count(struct mlx5_core_dev *mdev); int mlx5_fpga_ipsec_counters_read(struct mlx5_core_dev *mdev, u64 *counters, unsigned int counters_count); +void *mlx5_fpga_ipsec_create_sa_ctx(struct mlx5_core_dev *mdev, + struct mlx5_accel_esp_xfrm *accel_xfrm, + const __be32 saddr[4], + const __be32 daddr[4], + const __be32 spi, bool is_ipv6); +void mlx5_fpga_ipsec_delete_sa_ctx(void *context); + int mlx5_fpga_ipsec_init(struct mlx5_core_dev *mdev); void mlx5_fpga_ipsec_cleanup(struct mlx5_core_dev *mdev); +void mlx5_fpga_ipsec_build_fs_cmds(void); -#else +struct mlx5_accel_esp_xfrm * +mlx5_fpga_esp_create_xfrm(struct mlx5_core_dev *mdev, + const struct mlx5_accel_esp_xfrm_attrs *attrs, + u32 flags); +void mlx5_fpga_esp_destroy_xfrm(struct mlx5_accel_esp_xfrm *xfrm); +int mlx5_fpga_esp_modify_xfrm(struct mlx5_accel_esp_xfrm *xfrm, + const struct mlx5_accel_esp_xfrm_attrs *attrs); -static inline void *mlx5_fpga_ipsec_sa_cmd_exec(struct mlx5_core_dev *mdev, - struct mlx5_accel_ipsec_sa *cmd) -{ - return ERR_PTR(-EOPNOTSUPP); -} +const struct mlx5_flow_cmds * +mlx5_fs_cmd_get_default_ipsec_fpga_cmds(enum fs_flow_table_type type); -static inline int mlx5_fpga_ipsec_sa_cmd_wait(void *context) -{ - return -EOPNOTSUPP; -} +#else static inline u32 mlx5_fpga_ipsec_device_caps(struct mlx5_core_dev *mdev) { @@ -80,6 +85,20 @@ static inline int mlx5_fpga_ipsec_counters_read(struct mlx5_core_dev *mdev, return 0; } +static inline void * +mlx5_fpga_ipsec_create_sa_ctx(struct mlx5_core_dev *mdev, + struct mlx5_accel_esp_xfrm *accel_xfrm, + const __be32 saddr[4], + const __be32 daddr[4], + const __be32 spi, bool is_ipv6) +{ + return NULL; +} + +static inline void mlx5_fpga_ipsec_delete_sa_ctx(void *context) +{ +} + static inline int mlx5_fpga_ipsec_init(struct mlx5_core_dev *mdev) { return 0; @@ -89,6 +108,35 @@ static inline void mlx5_fpga_ipsec_cleanup(struct mlx5_core_dev *mdev) { } +static inline void mlx5_fpga_ipsec_build_fs_cmds(void) +{ +} + +static inline struct mlx5_accel_esp_xfrm * +mlx5_fpga_esp_create_xfrm(struct mlx5_core_dev *mdev, + const struct mlx5_accel_esp_xfrm_attrs *attrs, + u32 flags) +{ + return ERR_PTR(-EOPNOTSUPP); +} + +static inline void mlx5_fpga_esp_destroy_xfrm(struct mlx5_accel_esp_xfrm *xfrm) +{ +} + +static inline int +mlx5_fpga_esp_modify_xfrm(struct mlx5_accel_esp_xfrm *xfrm, + const struct mlx5_accel_esp_xfrm_attrs *attrs) +{ + return -EOPNOTSUPP; +} + +static inline const struct mlx5_flow_cmds * +mlx5_fs_cmd_get_default_ipsec_fpga_cmds(enum fs_flow_table_type type) +{ + return mlx5_fs_cmd_get_default(type); +} + #endif /* CONFIG_MLX5_FPGA */ #endif /* __MLX5_FPGA_SADB_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c index 881e2e55840c..ef5afd7c9325 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.c @@ -39,9 +39,81 @@ #include "mlx5_core.h" #include "eswitch.h" -int mlx5_cmd_update_root_ft(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, u32 underlay_qpn, - bool disconnect) +static int mlx5_cmd_stub_update_root_ft(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + u32 underlay_qpn, + bool disconnect) +{ + return 0; +} + +static int mlx5_cmd_stub_create_flow_table(struct mlx5_core_dev *dev, + u16 vport, + enum fs_flow_table_op_mod op_mod, + enum fs_flow_table_type type, + unsigned int level, + unsigned int log_size, + struct mlx5_flow_table *next_ft, + unsigned int *table_id, u32 flags) +{ + return 0; +} + +static int mlx5_cmd_stub_destroy_flow_table(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft) +{ + return 0; +} + +static int mlx5_cmd_stub_modify_flow_table(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct mlx5_flow_table *next_ft) +{ + return 0; +} + +static int mlx5_cmd_stub_create_flow_group(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + u32 *in, + unsigned int *group_id) +{ + return 0; +} + +static int mlx5_cmd_stub_destroy_flow_group(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int group_id) +{ + return 0; +} + +static int mlx5_cmd_stub_create_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct mlx5_flow_group *group, + struct fs_fte *fte) +{ + return 0; +} + +static int mlx5_cmd_stub_update_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int group_id, + int modify_mask, + struct fs_fte *fte) +{ + return -EOPNOTSUPP; +} + +static int mlx5_cmd_stub_delete_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct fs_fte *fte) +{ + return 0; +} + +static int mlx5_cmd_update_root_ft(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, u32 underlay_qpn, + bool disconnect) { u32 in[MLX5_ST_SZ_DW(set_flow_table_root_in)] = {0}; u32 out[MLX5_ST_SZ_DW(set_flow_table_root_out)] = {0}; @@ -71,12 +143,14 @@ int mlx5_cmd_update_root_ft(struct mlx5_core_dev *dev, return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out)); } -int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev, - u16 vport, - enum fs_flow_table_op_mod op_mod, - enum fs_flow_table_type type, unsigned int level, - unsigned int log_size, struct mlx5_flow_table - *next_ft, unsigned int *table_id, u32 flags) +static int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev, + u16 vport, + enum fs_flow_table_op_mod op_mod, + enum fs_flow_table_type type, + unsigned int level, + unsigned int log_size, + struct mlx5_flow_table *next_ft, + unsigned int *table_id, u32 flags) { int en_encap_decap = !!(flags & MLX5_FLOW_TABLE_TUNNEL_EN); u32 out[MLX5_ST_SZ_DW(create_flow_table_out)] = {0}; @@ -125,8 +199,8 @@ int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev, return err; } -int mlx5_cmd_destroy_flow_table(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft) +static int mlx5_cmd_destroy_flow_table(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft) { u32 in[MLX5_ST_SZ_DW(destroy_flow_table_in)] = {0}; u32 out[MLX5_ST_SZ_DW(destroy_flow_table_out)] = {0}; @@ -143,9 +217,9 @@ int mlx5_cmd_destroy_flow_table(struct mlx5_core_dev *dev, return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out)); } -int mlx5_cmd_modify_flow_table(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, - struct mlx5_flow_table *next_ft) +static int mlx5_cmd_modify_flow_table(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct mlx5_flow_table *next_ft) { u32 in[MLX5_ST_SZ_DW(modify_flow_table_in)] = {0}; u32 out[MLX5_ST_SZ_DW(modify_flow_table_out)] = {0}; @@ -188,10 +262,10 @@ int mlx5_cmd_modify_flow_table(struct mlx5_core_dev *dev, return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out)); } -int mlx5_cmd_create_flow_group(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, - u32 *in, - unsigned int *group_id) +static int mlx5_cmd_create_flow_group(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + u32 *in, + unsigned int *group_id) { u32 out[MLX5_ST_SZ_DW(create_flow_group_out)] = {0}; int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); @@ -213,9 +287,9 @@ int mlx5_cmd_create_flow_group(struct mlx5_core_dev *dev, return err; } -int mlx5_cmd_destroy_flow_group(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, - unsigned int group_id) +static int mlx5_cmd_destroy_flow_group(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int group_id) { u32 out[MLX5_ST_SZ_DW(destroy_flow_group_out)] = {0}; u32 in[MLX5_ST_SZ_DW(destroy_flow_group_in)] = {0}; @@ -243,7 +317,7 @@ static int mlx5_cmd_set_fte(struct mlx5_core_dev *dev, fte->dests_size * MLX5_ST_SZ_BYTES(dest_format_struct); u32 out[MLX5_ST_SZ_DW(set_fte_out)] = {0}; struct mlx5_flow_rule *dst; - void *in_flow_context; + void *in_flow_context, *vlan; void *in_match_value; void *in_dests; u32 *in; @@ -266,16 +340,25 @@ static int mlx5_cmd_set_fte(struct mlx5_core_dev *dev, in_flow_context = MLX5_ADDR_OF(set_fte_in, in, flow_context); MLX5_SET(flow_context, in_flow_context, group_id, group_id); - MLX5_SET(flow_context, in_flow_context, flow_tag, fte->flow_tag); - MLX5_SET(flow_context, in_flow_context, action, fte->action); - MLX5_SET(flow_context, in_flow_context, encap_id, fte->encap_id); - MLX5_SET(flow_context, in_flow_context, modify_header_id, fte->modify_id); + + MLX5_SET(flow_context, in_flow_context, flow_tag, fte->action.flow_tag); + MLX5_SET(flow_context, in_flow_context, action, fte->action.action); + MLX5_SET(flow_context, in_flow_context, encap_id, fte->action.encap_id); + MLX5_SET(flow_context, in_flow_context, modify_header_id, + fte->action.modify_id); + + vlan = MLX5_ADDR_OF(flow_context, in_flow_context, push_vlan); + + MLX5_SET(vlan, vlan, ethtype, fte->action.vlan.ethtype); + MLX5_SET(vlan, vlan, vid, fte->action.vlan.vid); + MLX5_SET(vlan, vlan, prio, fte->action.vlan.prio); + in_match_value = MLX5_ADDR_OF(flow_context, in_flow_context, match_value); memcpy(in_match_value, &fte->val, sizeof(fte->val)); in_dests = MLX5_ADDR_OF(flow_context, in_flow_context, destination); - if (fte->action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) { + if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) { int list_size = 0; list_for_each_entry(dst, &fte->node.children, node.list) { @@ -301,7 +384,7 @@ static int mlx5_cmd_set_fte(struct mlx5_core_dev *dev, list_size); } - if (fte->action & MLX5_FLOW_CONTEXT_ACTION_COUNT) { + if (fte->action.action & MLX5_FLOW_CONTEXT_ACTION_COUNT) { int max_list_size = BIT(MLX5_CAP_FLOWTABLE_TYPE(dev, log_max_flow_counter, ft->type)); @@ -332,19 +415,21 @@ err_out: return err; } -int mlx5_cmd_create_fte(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, - unsigned group_id, - struct fs_fte *fte) +static int mlx5_cmd_create_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct mlx5_flow_group *group, + struct fs_fte *fte) { + unsigned int group_id = group->id; + return mlx5_cmd_set_fte(dev, 0, 0, ft, group_id, fte); } -int mlx5_cmd_update_fte(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, - unsigned group_id, - int modify_mask, - struct fs_fte *fte) +static int mlx5_cmd_update_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int group_id, + int modify_mask, + struct fs_fte *fte) { int opmod; int atomic_mod_cap = MLX5_CAP_FLOWTABLE(dev, @@ -357,9 +442,9 @@ int mlx5_cmd_update_fte(struct mlx5_core_dev *dev, return mlx5_cmd_set_fte(dev, opmod, modify_mask, ft, group_id, fte); } -int mlx5_cmd_delete_fte(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, - unsigned int index) +static int mlx5_cmd_delete_fte(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct fs_fte *fte) { u32 out[MLX5_ST_SZ_DW(delete_fte_out)] = {0}; u32 in[MLX5_ST_SZ_DW(delete_fte_in)] = {0}; @@ -367,7 +452,7 @@ int mlx5_cmd_delete_fte(struct mlx5_core_dev *dev, MLX5_SET(delete_fte_in, in, opcode, MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY); MLX5_SET(delete_fte_in, in, table_type, ft->type); MLX5_SET(delete_fte_in, in, table_id, ft->id); - MLX5_SET(delete_fte_in, in, flow_index, index); + MLX5_SET(delete_fte_in, in, flow_index, fte->index); if (ft->vport) { MLX5_SET(delete_fte_in, in, vport_number, ft->vport); MLX5_SET(delete_fte_in, in, other_vport, 1); @@ -610,3 +695,53 @@ void mlx5_modify_header_dealloc(struct mlx5_core_dev *dev, u32 modify_header_id) mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out)); } + +static const struct mlx5_flow_cmds mlx5_flow_cmds = { + .create_flow_table = mlx5_cmd_create_flow_table, + .destroy_flow_table = mlx5_cmd_destroy_flow_table, + .modify_flow_table = mlx5_cmd_modify_flow_table, + .create_flow_group = mlx5_cmd_create_flow_group, + .destroy_flow_group = mlx5_cmd_destroy_flow_group, + .create_fte = mlx5_cmd_create_fte, + .update_fte = mlx5_cmd_update_fte, + .delete_fte = mlx5_cmd_delete_fte, + .update_root_ft = mlx5_cmd_update_root_ft, +}; + +static const struct mlx5_flow_cmds mlx5_flow_cmd_stubs = { + .create_flow_table = mlx5_cmd_stub_create_flow_table, + .destroy_flow_table = mlx5_cmd_stub_destroy_flow_table, + .modify_flow_table = mlx5_cmd_stub_modify_flow_table, + .create_flow_group = mlx5_cmd_stub_create_flow_group, + .destroy_flow_group = mlx5_cmd_stub_destroy_flow_group, + .create_fte = mlx5_cmd_stub_create_fte, + .update_fte = mlx5_cmd_stub_update_fte, + .delete_fte = mlx5_cmd_stub_delete_fte, + .update_root_ft = mlx5_cmd_stub_update_root_ft, +}; + +static const struct mlx5_flow_cmds *mlx5_fs_cmd_get_fw_cmds(void) +{ + return &mlx5_flow_cmds; +} + +static const struct mlx5_flow_cmds *mlx5_fs_cmd_get_stub_cmds(void) +{ + return &mlx5_flow_cmd_stubs; +} + +const struct mlx5_flow_cmds *mlx5_fs_cmd_get_default(enum fs_flow_table_type type) +{ + switch (type) { + case FS_FT_NIC_RX: + case FS_FT_ESW_EGRESS_ACL: + case FS_FT_ESW_INGRESS_ACL: + case FS_FT_FDB: + case FS_FT_SNIFFER_RX: + case FS_FT_SNIFFER_TX: + return mlx5_fs_cmd_get_fw_cmds(); + case FS_FT_NIC_TX: + default: + return mlx5_fs_cmd_get_stub_cmds(); + } +} diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h index 71e2d0f37ad9..6228ba7bfa1a 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_cmd.h @@ -33,46 +33,52 @@ #ifndef _MLX5_FS_CMD_ #define _MLX5_FS_CMD_ -int mlx5_cmd_create_flow_table(struct mlx5_core_dev *dev, - u16 vport, - enum fs_flow_table_op_mod op_mod, - enum fs_flow_table_type type, unsigned int level, - unsigned int log_size, struct mlx5_flow_table - *next_ft, unsigned int *table_id, u32 flags); +#include "fs_core.h" -int mlx5_cmd_destroy_flow_table(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft); +struct mlx5_flow_cmds { + int (*create_flow_table)(struct mlx5_core_dev *dev, + u16 vport, + enum fs_flow_table_op_mod op_mod, + enum fs_flow_table_type type, + unsigned int level, unsigned int log_size, + struct mlx5_flow_table *next_ft, + unsigned int *table_id, u32 flags); + int (*destroy_flow_table)(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft); -int mlx5_cmd_modify_flow_table(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, - struct mlx5_flow_table *next_ft); + int (*modify_flow_table)(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct mlx5_flow_table *next_ft); -int mlx5_cmd_create_flow_group(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, - u32 *in, unsigned int *group_id); + int (*create_flow_group)(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + u32 *in, + unsigned int *group_id); -int mlx5_cmd_destroy_flow_group(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, - unsigned int group_id); + int (*destroy_flow_group)(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int group_id); -int mlx5_cmd_create_fte(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, - unsigned group_id, - struct fs_fte *fte); + int (*create_fte)(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct mlx5_flow_group *fg, + struct fs_fte *fte); -int mlx5_cmd_update_fte(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, - unsigned group_id, - int modify_mask, - struct fs_fte *fte); + int (*update_fte)(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + unsigned int group_id, + int modify_mask, + struct fs_fte *fte); -int mlx5_cmd_delete_fte(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, - unsigned int index); + int (*delete_fte)(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + struct fs_fte *fte); -int mlx5_cmd_update_root_ft(struct mlx5_core_dev *dev, - struct mlx5_flow_table *ft, u32 underlay_qpn, - bool disconnect); + int (*update_root_ft)(struct mlx5_core_dev *dev, + struct mlx5_flow_table *ft, + u32 underlay_qpn, + bool disconnect); +}; int mlx5_cmd_fc_alloc(struct mlx5_core_dev *dev, u32 *id); int mlx5_cmd_fc_free(struct mlx5_core_dev *dev, u32 id); @@ -90,4 +96,6 @@ void mlx5_cmd_fc_bulk_get(struct mlx5_core_dev *dev, struct mlx5_cmd_fc_bulk *b, u32 id, u64 *packets, u64 *bytes); +const struct mlx5_flow_cmds *mlx5_fs_cmd_get_default(enum fs_flow_table_type type); + #endif diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c index 31fc2cfac3b3..de51e7c39bc8 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.c @@ -37,6 +37,8 @@ #include "fs_core.h" #include "fs_cmd.h" #include "diag/fs_tracepoint.h" +#include "accel/ipsec.h" +#include "fpga/ipsec.h" #define INIT_TREE_NODE_ARRAY_SIZE(...) (sizeof((struct init_tree_node[]){__VA_ARGS__}) /\ sizeof(struct init_tree_node)) @@ -425,15 +427,17 @@ static void del_sw_prio(struct fs_node *node) static void del_hw_flow_table(struct fs_node *node) { + struct mlx5_flow_root_namespace *root; struct mlx5_flow_table *ft; struct mlx5_core_dev *dev; int err; fs_get_obj(ft, node); dev = get_dev(&ft->node); + root = find_root(&ft->node); if (node->active) { - err = mlx5_cmd_destroy_flow_table(dev, ft); + err = root->cmds->destroy_flow_table(dev, ft); if (err) mlx5_core_warn(dev, "flow steering can't destroy ft\n"); } @@ -454,6 +458,7 @@ static void del_sw_flow_table(struct fs_node *node) static void del_sw_hw_rule(struct fs_node *node) { + struct mlx5_flow_root_namespace *root; struct mlx5_flow_rule *rule; struct mlx5_flow_table *ft; struct mlx5_flow_group *fg; @@ -477,19 +482,20 @@ static void del_sw_hw_rule(struct fs_node *node) if (rule->dest_attr.type == MLX5_FLOW_DESTINATION_TYPE_COUNTER && --fte->dests_size) { modify_mask = BIT(MLX5_SET_FTE_MODIFY_ENABLE_MASK_ACTION); - fte->action &= ~MLX5_FLOW_CONTEXT_ACTION_COUNT; + fte->action.action &= ~MLX5_FLOW_CONTEXT_ACTION_COUNT; update_fte = true; goto out; } - if ((fte->action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) && + if ((fte->action.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST) && --fte->dests_size) { modify_mask = BIT(MLX5_SET_FTE_MODIFY_ENABLE_MASK_DESTINATION_LIST), update_fte = true; } out: + root = find_root(&ft->node); if (update_fte && fte->dests_size) { - err = mlx5_cmd_update_fte(dev, ft, fg->id, modify_mask, fte); + err = root->cmds->update_fte(dev, ft, fg->id, modify_mask, fte); if (err) mlx5_core_warn(dev, "%s can't del rule fg id=%d fte_index=%d\n", @@ -500,6 +506,7 @@ out: static void del_hw_fte(struct fs_node *node) { + struct mlx5_flow_root_namespace *root; struct mlx5_flow_table *ft; struct mlx5_flow_group *fg; struct mlx5_core_dev *dev; @@ -512,9 +519,9 @@ static void del_hw_fte(struct fs_node *node) trace_mlx5_fs_del_fte(fte); dev = get_dev(&ft->node); + root = find_root(&ft->node); if (node->active) { - err = mlx5_cmd_delete_fte(dev, ft, - fte->index); + err = root->cmds->delete_fte(dev, ft, fte); if (err) mlx5_core_warn(dev, "flow steering can't delete fte in index %d of flow group id %d\n", @@ -542,6 +549,7 @@ static void del_sw_fte(struct fs_node *node) static void del_hw_flow_group(struct fs_node *node) { + struct mlx5_flow_root_namespace *root; struct mlx5_flow_group *fg; struct mlx5_flow_table *ft; struct mlx5_core_dev *dev; @@ -551,7 +559,8 @@ static void del_hw_flow_group(struct fs_node *node) dev = get_dev(&ft->node); trace_mlx5_fs_del_fg(fg); - if (fg->node.active && mlx5_cmd_destroy_flow_group(dev, ft, fg->id)) + root = find_root(&ft->node); + if (fg->node.active && root->cmds->destroy_flow_group(dev, ft, fg->id)) mlx5_core_warn(dev, "flow steering can't destroy fg %d of ft %d\n", fg->id, ft->id); } @@ -615,10 +624,7 @@ static struct fs_fte *alloc_fte(struct mlx5_flow_table *ft, memcpy(fte->val, match_value, sizeof(fte->val)); fte->node.type = FS_TYPE_FLOW_ENTRY; - fte->flow_tag = flow_act->flow_tag; - fte->action = flow_act->action; - fte->encap_id = flow_act->encap_id; - fte->modify_id = flow_act->modify_id; + fte->action = *flow_act; tree_init_node(&fte->node, del_hw_fte, del_sw_fte); @@ -797,15 +803,14 @@ static int connect_fts_in_prio(struct mlx5_core_dev *dev, struct fs_prio *prio, struct mlx5_flow_table *ft) { + struct mlx5_flow_root_namespace *root = find_root(&prio->node); struct mlx5_flow_table *iter; int i = 0; int err; fs_for_each_ft(iter, prio) { i++; - err = mlx5_cmd_modify_flow_table(dev, - iter, - ft); + err = root->cmds->modify_flow_table(dev, iter, ft); if (err) { mlx5_core_warn(dev, "Failed to modify flow table %d\n", iter->id); @@ -853,12 +858,12 @@ static int update_root_ft_create(struct mlx5_flow_table *ft, struct fs_prio if (list_empty(&root->underlay_qpns)) { /* Don't set any QPN (zero) in case QPN list is empty */ qpn = 0; - err = mlx5_cmd_update_root_ft(root->dev, ft, qpn, false); + err = root->cmds->update_root_ft(root->dev, ft, qpn, false); } else { list_for_each_entry(uqp, &root->underlay_qpns, list) { qpn = uqp->qpn; - err = mlx5_cmd_update_root_ft(root->dev, ft, qpn, - false); + err = root->cmds->update_root_ft(root->dev, ft, + qpn, false); if (err) break; } @@ -877,6 +882,7 @@ static int update_root_ft_create(struct mlx5_flow_table *ft, struct fs_prio static int _mlx5_modify_rule_destination(struct mlx5_flow_rule *rule, struct mlx5_flow_destination *dest) { + struct mlx5_flow_root_namespace *root; struct mlx5_flow_table *ft; struct mlx5_flow_group *fg; struct fs_fte *fte; @@ -884,17 +890,16 @@ static int _mlx5_modify_rule_destination(struct mlx5_flow_rule *rule, int err = 0; fs_get_obj(fte, rule->node.parent); - if (!(fte->action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST)) + if (!(fte->action.action & MLX5_FLOW_CONTEXT_ACTION_FWD_DEST)) return -EINVAL; down_write_ref_node(&fte->node); fs_get_obj(fg, fte->node.parent); fs_get_obj(ft, fg->node.parent); memcpy(&rule->dest_attr, dest, sizeof(*dest)); - err = mlx5_cmd_update_fte(get_dev(&ft->node), - ft, fg->id, - modify_mask, - fte); + root = find_root(&ft->node); + err = root->cmds->update_fte(get_dev(&ft->node), ft, fg->id, + modify_mask, fte); up_write_ref_node(&fte->node); return err; @@ -1035,9 +1040,9 @@ static struct mlx5_flow_table *__mlx5_create_flow_table(struct mlx5_flow_namespa tree_init_node(&ft->node, del_hw_flow_table, del_sw_flow_table); log_table_sz = ft->max_fte ? ilog2(ft->max_fte) : 0; next_ft = find_next_chained_ft(fs_prio); - err = mlx5_cmd_create_flow_table(root->dev, ft->vport, ft->op_mod, ft->type, - ft->level, log_table_sz, next_ft, &ft->id, - ft->flags); + err = root->cmds->create_flow_table(root->dev, ft->vport, ft->op_mod, + ft->type, ft->level, log_table_sz, + next_ft, &ft->id, ft->flags); if (err) goto free_ft; @@ -1053,7 +1058,7 @@ static struct mlx5_flow_table *__mlx5_create_flow_table(struct mlx5_flow_namespa mutex_unlock(&root->chain_lock); return ft; destroy_ft: - mlx5_cmd_destroy_flow_table(root->dev, ft); + root->cmds->destroy_flow_table(root->dev, ft); free_ft: kfree(ft); unlock_root: @@ -1125,6 +1130,7 @@ EXPORT_SYMBOL(mlx5_create_auto_grouped_flow_table); struct mlx5_flow_group *mlx5_create_flow_group(struct mlx5_flow_table *ft, u32 *fg_in) { + struct mlx5_flow_root_namespace *root = find_root(&ft->node); void *match_criteria = MLX5_ADDR_OF(create_flow_group_in, fg_in, match_criteria); u8 match_criteria_enable = MLX5_GET(create_flow_group_in, @@ -1152,7 +1158,7 @@ struct mlx5_flow_group *mlx5_create_flow_group(struct mlx5_flow_table *ft, if (IS_ERR(fg)) return fg; - err = mlx5_cmd_create_flow_group(dev, ft, fg_in, &fg->id); + err = root->cmds->create_flow_group(dev, ft, fg_in, &fg->id); if (err) { tree_put_node(&fg->node); return ERR_PTR(err); @@ -1275,6 +1281,7 @@ add_rule_fte(struct fs_fte *fte, int dest_num, bool update_action) { + struct mlx5_flow_root_namespace *root; struct mlx5_flow_handle *handle; struct mlx5_flow_table *ft; int modify_mask = 0; @@ -1290,12 +1297,13 @@ add_rule_fte(struct fs_fte *fte, modify_mask |= BIT(MLX5_SET_FTE_MODIFY_ENABLE_MASK_ACTION); fs_get_obj(ft, fg->node.parent); + root = find_root(&fg->node); if (!(fte->status & FS_FTE_STATUS_EXISTING)) - err = mlx5_cmd_create_fte(get_dev(&ft->node), - ft, fg->id, fte); + err = root->cmds->create_fte(get_dev(&ft->node), + ft, fg, fte); else - err = mlx5_cmd_update_fte(get_dev(&ft->node), - ft, fg->id, modify_mask, fte); + err = root->cmds->update_fte(get_dev(&ft->node), ft, fg->id, + modify_mask, fte); if (err) goto free_handle; @@ -1360,6 +1368,7 @@ out: static int create_auto_flow_group(struct mlx5_flow_table *ft, struct mlx5_flow_group *fg) { + struct mlx5_flow_root_namespace *root = find_root(&ft->node); struct mlx5_core_dev *dev = get_dev(&ft->node); int inlen = MLX5_ST_SZ_BYTES(create_flow_group_in); void *match_criteria_addr; @@ -1380,7 +1389,7 @@ static int create_auto_flow_group(struct mlx5_flow_table *ft, memcpy(match_criteria_addr, fg->mask.match_criteria, sizeof(fg->mask.match_criteria)); - err = mlx5_cmd_create_flow_group(dev, ft, in, &fg->id); + err = root->cmds->create_flow_group(dev, ft, in, &fg->id); if (!err) { fg->node.active = true; trace_mlx5_fs_add_fg(fg); @@ -1430,7 +1439,9 @@ static bool check_conflicting_actions(u32 action1, u32 action2) if (xored_actions & (MLX5_FLOW_CONTEXT_ACTION_DROP | MLX5_FLOW_CONTEXT_ACTION_ENCAP | MLX5_FLOW_CONTEXT_ACTION_DECAP | - MLX5_FLOW_CONTEXT_ACTION_MOD_HDR)) + MLX5_FLOW_CONTEXT_ACTION_MOD_HDR | + MLX5_FLOW_CONTEXT_ACTION_VLAN_POP | + MLX5_FLOW_CONTEXT_ACTION_VLAN_PUSH)) return true; return false; @@ -1438,16 +1449,17 @@ static bool check_conflicting_actions(u32 action1, u32 action2) static int check_conflicting_ftes(struct fs_fte *fte, const struct mlx5_flow_act *flow_act) { - if (check_conflicting_actions(flow_act->action, fte->action)) { + if (check_conflicting_actions(flow_act->action, fte->action.action)) { mlx5_core_warn(get_dev(&fte->node), "Found two FTEs with conflicting actions\n"); return -EEXIST; } - if (fte->flow_tag != flow_act->flow_tag) { + if (flow_act->has_flow_tag && + fte->action.flow_tag != flow_act->flow_tag) { mlx5_core_warn(get_dev(&fte->node), "FTE flow tag %u already exists with different flow tag %u\n", - fte->flow_tag, + fte->action.flow_tag, flow_act->flow_tag); return -EEXIST; } @@ -1471,12 +1483,12 @@ static struct mlx5_flow_handle *add_rule_fg(struct mlx5_flow_group *fg, if (ret) return ERR_PTR(ret); - old_action = fte->action; - fte->action |= flow_act->action; + old_action = fte->action.action; + fte->action.action |= flow_act->action; handle = add_rule_fte(fte, fg, dest, dest_num, old_action != flow_act->action); if (IS_ERR(handle)) { - fte->action = old_action; + fte->action.action = old_action; return handle; } trace_mlx5_fs_set_fte(fte, false); @@ -1637,7 +1649,6 @@ try_add_to_existing_fg(struct mlx5_flow_table *ft, list_for_each_entry(iter, match_head, list) { nested_down_read_ref_node(&iter->g->node, FS_LOCK_PARENT); - ida_pre_get(&iter->g->fte_allocator, GFP_KERNEL); } search_again_locked: @@ -1919,7 +1930,6 @@ static int update_root_ft_destroy(struct mlx5_flow_table *ft) return 0; new_root_ft = find_next_ft(ft); - if (!new_root_ft) { root->root_ft = NULL; return 0; @@ -1928,13 +1938,14 @@ static int update_root_ft_destroy(struct mlx5_flow_table *ft) if (list_empty(&root->underlay_qpns)) { /* Don't set any QPN (zero) in case QPN list is empty */ qpn = 0; - err = mlx5_cmd_update_root_ft(root->dev, new_root_ft, qpn, - false); + err = root->cmds->update_root_ft(root->dev, new_root_ft, + qpn, false); } else { list_for_each_entry(uqp, &root->underlay_qpns, list) { qpn = uqp->qpn; - err = mlx5_cmd_update_root_ft(root->dev, new_root_ft, - qpn, false); + err = root->cmds->update_root_ft(root->dev, + new_root_ft, qpn, + false); if (err) break; } @@ -2046,6 +2057,11 @@ struct mlx5_flow_namespace *mlx5_get_flow_namespace(struct mlx5_core_dev *dev, return &steering->sniffer_tx_root_ns->ns; else return NULL; + case MLX5_FLOW_NAMESPACE_EGRESS: + if (steering->egress_root_ns) + return &steering->egress_root_ns->ns; + else + return NULL; default: return NULL; } @@ -2236,13 +2252,18 @@ static int init_root_tree(struct mlx5_flow_steering *steering, return 0; } -static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_flow_steering *steering, - enum fs_flow_table_type - table_type) +static struct mlx5_flow_root_namespace +*create_root_ns(struct mlx5_flow_steering *steering, + enum fs_flow_table_type table_type) { + const struct mlx5_flow_cmds *cmds = mlx5_fs_cmd_get_default(table_type); struct mlx5_flow_root_namespace *root_ns; struct mlx5_flow_namespace *ns; + if (mlx5_accel_ipsec_device_caps(steering->dev) & MLX5_ACCEL_IPSEC_CAP_DEVICE && + (table_type == FS_FT_NIC_RX || table_type == FS_FT_NIC_TX)) + cmds = mlx5_fs_cmd_get_default_ipsec_fpga_cmds(table_type); + /* Create the root namespace */ root_ns = kvzalloc(sizeof(*root_ns), GFP_KERNEL); if (!root_ns) @@ -2250,6 +2271,7 @@ static struct mlx5_flow_root_namespace *create_root_ns(struct mlx5_flow_steering root_ns->dev = steering->dev; root_ns->table_type = table_type; + root_ns->cmds = cmds; INIT_LIST_HEAD(&root_ns->underlay_qpns); @@ -2408,6 +2430,7 @@ void mlx5_cleanup_fs(struct mlx5_core_dev *dev) cleanup_root_ns(steering->fdb_root_ns); cleanup_root_ns(steering->sniffer_rx_root_ns); cleanup_root_ns(steering->sniffer_tx_root_ns); + cleanup_root_ns(steering->egress_root_ns); mlx5_cleanup_fc_stats(dev); kmem_cache_destroy(steering->ftes_cache); kmem_cache_destroy(steering->fgs_cache); @@ -2553,6 +2576,20 @@ cleanup_root_ns: return err; } +static int init_egress_root_ns(struct mlx5_flow_steering *steering) +{ + struct fs_prio *prio; + + steering->egress_root_ns = create_root_ns(steering, + FS_FT_NIC_TX); + if (!steering->egress_root_ns) + return -ENOMEM; + + /* create 1 prio*/ + prio = fs_create_prio(&steering->egress_root_ns->ns, 0, 1); + return PTR_ERR_OR_ZERO(prio); +} + int mlx5_init_fs(struct mlx5_core_dev *dev) { struct mlx5_flow_steering *steering; @@ -2618,6 +2655,12 @@ int mlx5_init_fs(struct mlx5_core_dev *dev) goto err; } + if (MLX5_IPSEC_DEV(dev)) { + err = init_egress_root_ns(steering); + if (err) + goto err; + } + return 0; err: mlx5_cleanup_fs(dev); @@ -2641,7 +2684,8 @@ int mlx5_fs_add_rx_underlay_qpn(struct mlx5_core_dev *dev, u32 underlay_qpn) goto update_ft_fail; } - err = mlx5_cmd_update_root_ft(dev, root->root_ft, underlay_qpn, false); + err = root->cmds->update_root_ft(dev, root->root_ft, underlay_qpn, + false); if (err) { mlx5_core_warn(dev, "Failed adding underlay QPN (%u) to root FT err(%d)\n", underlay_qpn, err); @@ -2684,7 +2728,8 @@ int mlx5_fs_remove_rx_underlay_qpn(struct mlx5_core_dev *dev, u32 underlay_qpn) goto out; } - err = mlx5_cmd_update_root_ft(dev, root->root_ft, underlay_qpn, true); + err = root->cmds->update_root_ft(dev, root->root_ft, underlay_qpn, + true); if (err) mlx5_core_warn(dev, "Failed removing underlay QPN (%u) from root FT err(%d)\n", underlay_qpn, err); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h index 05262708f14b..e26d3e9d5f9f 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/fs_core.h @@ -48,6 +48,7 @@ enum fs_node_type { enum fs_flow_table_type { FS_FT_NIC_RX = 0x0, + FS_FT_NIC_TX = 0x1, FS_FT_ESW_EGRESS_ACL = 0x2, FS_FT_ESW_INGRESS_ACL = 0x3, FS_FT_FDB = 0X4, @@ -75,6 +76,7 @@ struct mlx5_flow_steering { struct mlx5_flow_root_namespace **esw_ingress_root_ns; struct mlx5_flow_root_namespace *sniffer_tx_root_ns; struct mlx5_flow_root_namespace *sniffer_rx_root_ns; + struct mlx5_flow_root_namespace *egress_root_ns; }; struct fs_node { @@ -174,11 +176,8 @@ struct fs_fte { struct fs_node node; u32 val[MLX5_ST_SZ_DW_MATCH_PARAM]; u32 dests_size; - u32 flow_tag; u32 index; - u32 action; - u32 encap_id; - u32 modify_id; + struct mlx5_flow_act action; enum fs_fte_status status; struct mlx5_fc *counter; struct rhash_head hash; @@ -224,6 +223,7 @@ struct mlx5_flow_root_namespace { /* Should be held when chaining flow tables */ struct mutex chain_lock; struct list_head underlay_qpns; + const struct mlx5_flow_cmds *cmds; }; int mlx5_init_fc_stats(struct mlx5_core_dev *dev); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c index 9d11e92fb541..70066975f1b5 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c @@ -183,6 +183,9 @@ int mlx5_query_hca_caps(struct mlx5_core_dev *dev) return err; } + if (MLX5_CAP_GEN(dev, debug)) + mlx5_core_get_caps(dev, MLX5_CAP_DEBUG); + if (MLX5_CAP_GEN(dev, pcam_reg)) mlx5_get_pcam_reg(dev); @@ -242,7 +245,7 @@ int mlx5_cmd_force_teardown_hca(struct mlx5_core_dev *dev) force_state = MLX5_GET(teardown_hca_out, out, force_state); if (force_state == MLX5_TEARDOWN_HCA_OUT_FORCE_STATE_FAIL) { - mlx5_core_err(dev, "teardown with force mode failed\n"); + mlx5_core_warn(dev, "teardown with force mode failed, doing normal teardown\n"); return -EIO; } diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c index f953378bd13d..af3bb2f7a504 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib.c @@ -56,14 +56,17 @@ static void mlx5i_build_nic_params(struct mlx5_core_dev *mdev, struct mlx5e_params *params) { /* Override RQ params as IPoIB supports only LINKED LIST RQ for now */ - mlx5e_init_rq_type_params(mdev, params, MLX5_WQ_TYPE_LINKED_LIST); + MLX5E_SET_PFLAG(params, MLX5E_PFLAG_RX_STRIDING_RQ, false); + mlx5e_set_rq_type(mdev, params); + mlx5e_init_rq_type_params(mdev, params); /* RQ size in ipoib by default is 512 */ - params->log_rq_size = is_kdump_kernel() ? + params->log_rq_mtu_frames = is_kdump_kernel() ? MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE : MLX5I_PARAMS_DEFAULT_LOG_RQ_SIZE; params->lro_en = false; + params->hard_mtu = MLX5_IB_GRH_BYTES + MLX5_IPOIB_HARD_LEN; } /* Called directly after IPoIB netdevice was created to initialize SW structs */ @@ -79,10 +82,10 @@ void mlx5i_init(struct mlx5_core_dev *mdev, priv->netdev = netdev; priv->profile = profile; priv->ppriv = ppriv; - priv->hard_mtu = MLX5_IB_GRH_BYTES + MLX5_IPOIB_HARD_LEN; mutex_init(&priv->state_lock); - mlx5e_build_nic_params(mdev, &priv->channels.params, profile->max_nch(mdev)); + mlx5e_build_nic_params(mdev, &priv->channels.params, + profile->max_nch(mdev), netdev->mtu); mlx5i_build_nic_params(mdev, &priv->channels.params); mlx5e_timestamp_init(priv); @@ -366,25 +369,27 @@ static int mlx5i_change_mtu(struct net_device *netdev, int new_mtu) { struct mlx5e_priv *priv = mlx5i_epriv(netdev); struct mlx5e_channels new_channels = {}; - int curr_mtu; + struct mlx5e_params *params; int err = 0; mutex_lock(&priv->state_lock); - curr_mtu = netdev->mtu; - netdev->mtu = new_mtu; + params = &priv->channels.params; - if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) + if (!test_bit(MLX5E_STATE_OPENED, &priv->state)) { + params->sw_mtu = new_mtu; + netdev->mtu = params->sw_mtu; goto out; + } - new_channels.params = priv->channels.params; + new_channels.params = *params; + new_channels.params.sw_mtu = new_mtu; err = mlx5e_open_channels(priv, &new_channels); - if (err) { - netdev->mtu = curr_mtu; + if (err) goto out; - } mlx5e_switch_priv_channels(priv, &new_channels, NULL); + netdev->mtu = new_channels.params.sw_mtu; out: mutex_unlock(&priv->state_lock); @@ -538,7 +543,7 @@ static int mlx5i_detach_mcast(struct net_device *netdev, struct ib_device *hca, err = mlx5_core_detach_mcg(mdev, gid, ipriv->qp.qpn); if (err) - mlx5_core_dbg(mdev, "failed dettaching QPN 0x%x, MGID %pI6\n", + mlx5_core_dbg(mdev, "failed detaching QPN 0x%x, MGID %pI6\n", ipriv->qp.qpn, gid->raw); return err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c index b69e9d847a6b..54a188f41f90 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ipoib_vlan.c @@ -290,7 +290,7 @@ static void mlx5i_pkey_init(struct mlx5_core_dev *mdev, netdev->ethtool_ops = &mlx5i_pkey_ethtool_ops; /* Use dummy rqs */ - priv->channels.params.log_rq_size = MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE; + priv->channels.params.log_rq_mtu_frames = MLX5E_PARAMS_MINIMUM_LOG_RQ_SIZE; } /* Called directly before IPoIB netdevice is destroyed to cleanup SW structs */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index ae391e4b7070..13b6f66310c9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -58,6 +58,7 @@ #include "eswitch.h" #include "lib/mlx5.h" #include "fpga/core.h" +#include "fpga/ipsec.h" #include "accel/ipsec.h" #include "lib/clock.h" @@ -942,9 +943,9 @@ static int mlx5_init_once(struct mlx5_core_dev *dev, struct mlx5_priv *priv) goto out; } - err = mlx5_init_cq_table(dev); + err = mlx5_cq_debugfs_init(dev); if (err) { - dev_err(&pdev->dev, "failed to initialize cq table\n"); + dev_err(&pdev->dev, "failed to initialize cq debugfs\n"); goto err_eq_cleanup; } @@ -1002,7 +1003,7 @@ err_tables_cleanup: mlx5_cleanup_mkey_table(dev); mlx5_cleanup_srq_table(dev); mlx5_cleanup_qp_table(dev); - mlx5_cleanup_cq_table(dev); + mlx5_cq_debugfs_cleanup(dev); err_eq_cleanup: mlx5_eq_cleanup(dev); @@ -1023,7 +1024,7 @@ static void mlx5_cleanup_once(struct mlx5_core_dev *dev) mlx5_cleanup_mkey_table(dev); mlx5_cleanup_srq_table(dev); mlx5_cleanup_qp_table(dev); - mlx5_cleanup_cq_table(dev); + mlx5_cq_debugfs_cleanup(dev); mlx5_eq_cleanup(dev); } @@ -1173,6 +1174,18 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv, goto err_affinity_hints; } + err = mlx5_fpga_device_start(dev); + if (err) { + dev_err(&pdev->dev, "fpga device start failed %d\n", err); + goto err_fpga_start; + } + + err = mlx5_accel_ipsec_init(dev); + if (err) { + dev_err(&pdev->dev, "IPSec device start failed %d\n", err); + goto err_ipsec_start; + } + err = mlx5_init_fs(dev); if (err) { dev_err(&pdev->dev, "Failed to init flow steering\n"); @@ -1191,17 +1204,6 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv, goto err_sriov; } - err = mlx5_fpga_device_start(dev); - if (err) { - dev_err(&pdev->dev, "fpga device start failed %d\n", err); - goto err_fpga_start; - } - err = mlx5_accel_ipsec_init(dev); - if (err) { - dev_err(&pdev->dev, "IPSec device start failed %d\n", err); - goto err_ipsec_start; - } - if (mlx5_device_registered(dev)) { mlx5_attach_device(dev); } else { @@ -1219,17 +1221,18 @@ out: return 0; err_reg_dev: - mlx5_accel_ipsec_cleanup(dev); -err_ipsec_start: - mlx5_fpga_device_stop(dev); - -err_fpga_start: mlx5_sriov_detach(dev); err_sriov: mlx5_cleanup_fs(dev); err_fs: + mlx5_accel_ipsec_cleanup(dev); + +err_ipsec_start: + mlx5_fpga_device_stop(dev); + +err_fpga_start: mlx5_irq_clear_affinity_hints(dev); err_affinity_hints: @@ -1296,11 +1299,10 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv, if (mlx5_device_registered(dev)) mlx5_detach_device(dev); - mlx5_accel_ipsec_cleanup(dev); - mlx5_fpga_device_stop(dev); - mlx5_sriov_detach(dev); mlx5_cleanup_fs(dev); + mlx5_accel_ipsec_cleanup(dev); + mlx5_fpga_device_stop(dev); mlx5_irq_clear_affinity_hints(dev); free_comp_eqs(dev); mlx5_stop_eqs(dev); @@ -1657,6 +1659,7 @@ static int __init init(void) get_random_bytes(&sw_owner_id, sizeof(sw_owner_id)); mlx5_core_verify_params(); + mlx5_fpga_ipsec_build_fs_cmds(); mlx5_register_debugfs(); err = pci_register_driver(&mlx5_core_driver); diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h index 394552f36fcf..7d001fe6e631 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h @@ -38,16 +38,11 @@ #include <linux/sched.h> #include <linux/if_link.h> #include <linux/firmware.h> +#include <linux/mlx5/cq.h> #define DRIVER_NAME "mlx5_core" #define DRIVER_VERSION "5.0-0" -#define MLX5_TOTAL_VPORTS(mdev) (1 + pci_sriov_get_totalvfs(mdev->pdev)) -#define MLX5_VPORT_MANAGER(mdev) \ - (MLX5_CAP_GEN(mdev, vport_group_manager) && \ - (MLX5_CAP_GEN(mdev, port_type) == MLX5_CAP_PORT_TYPE_ETH) && \ - mlx5_core_is_pf(mdev)) - extern uint mlx5_core_debug_mask; #define mlx5_core_dbg(__dev, format, ...) \ @@ -55,6 +50,11 @@ extern uint mlx5_core_debug_mask; __func__, __LINE__, current->pid, \ ##__VA_ARGS__) +#define mlx5_core_dbg_once(__dev, format, ...) \ + dev_dbg_once(&(__dev)->pdev->dev, "%s:%d:(pid %d): " format, \ + __func__, __LINE__, current->pid, \ + ##__VA_ARGS__) + #define mlx5_core_dbg_mask(__dev, mask, format, ...) \ do { \ if ((mask) & mlx5_core_debug_mask) \ @@ -115,9 +115,29 @@ int mlx5_destroy_scheduling_element_cmd(struct mlx5_core_dev *dev, u8 hierarchy, u32 element_id); int mlx5_wait_for_vf_pages(struct mlx5_core_dev *dev); u64 mlx5_read_internal_timer(struct mlx5_core_dev *dev); + +int mlx5_eq_init(struct mlx5_core_dev *dev); +void mlx5_eq_cleanup(struct mlx5_core_dev *dev); +int mlx5_create_map_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq, u8 vecidx, + int nent, u64 mask, const char *name, + enum mlx5_eq_type type); +int mlx5_destroy_unmap_eq(struct mlx5_core_dev *dev, struct mlx5_eq *eq); +int mlx5_eq_add_cq(struct mlx5_eq *eq, struct mlx5_core_cq *cq); +int mlx5_eq_del_cq(struct mlx5_eq *eq, struct mlx5_core_cq *cq); +int mlx5_core_eq_query(struct mlx5_core_dev *dev, struct mlx5_eq *eq, + u32 *out, int outlen); +int mlx5_start_eqs(struct mlx5_core_dev *dev); +void mlx5_stop_eqs(struct mlx5_core_dev *dev); struct mlx5_eq *mlx5_eqn2eq(struct mlx5_core_dev *dev, int eqn); u32 mlx5_eq_poll_irq_disabled(struct mlx5_eq *eq); void mlx5_cq_tasklet_cb(unsigned long data); +void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, u64 vec, bool forced); +int mlx5_debug_eq_add(struct mlx5_core_dev *dev, struct mlx5_eq *eq); +void mlx5_debug_eq_remove(struct mlx5_core_dev *dev, struct mlx5_eq *eq); +int mlx5_eq_debugfs_init(struct mlx5_core_dev *dev); +void mlx5_eq_debugfs_cleanup(struct mlx5_core_dev *dev); +int mlx5_cq_debugfs_init(struct mlx5_core_dev *dev); +void mlx5_cq_debugfs_cleanup(struct mlx5_core_dev *dev); int mlx5_query_pcam_reg(struct mlx5_core_dev *dev, u32 *pcam, u8 feature_group, u8 access_reg_group); @@ -186,4 +206,5 @@ static inline int mlx5_lag_is_lacp_owner(struct mlx5_core_dev *dev) int mlx5_lag_allow(struct mlx5_core_dev *dev); int mlx5_lag_forbid(struct mlx5_core_dev *dev); +void mlx5_reload_interface(struct mlx5_core_dev *mdev, int protocol); #endif /* __MLX5_CORE_H__ */ diff --git a/drivers/net/ethernet/mellanox/mlx5/core/port.c b/drivers/net/ethernet/mellanox/mlx5/core/port.c index c37d00cd472a..fa9d0760dd36 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/port.c @@ -483,6 +483,17 @@ int mlx5_core_query_ib_ppcnt(struct mlx5_core_dev *dev, } EXPORT_SYMBOL_GPL(mlx5_core_query_ib_ppcnt); +static int mlx5_query_pfcc_reg(struct mlx5_core_dev *dev, u32 *out, + u32 out_size) +{ + u32 in[MLX5_ST_SZ_DW(pfcc_reg)] = {0}; + + MLX5_SET(pfcc_reg, in, local_port, 1); + + return mlx5_core_access_reg(dev, in, sizeof(in), out, + out_size, MLX5_REG_PFCC, 0, 0); +} + int mlx5_set_port_pause(struct mlx5_core_dev *dev, u32 rx_pause, u32 tx_pause) { u32 in[MLX5_ST_SZ_DW(pfcc_reg)] = {0}; @@ -500,13 +511,10 @@ EXPORT_SYMBOL_GPL(mlx5_set_port_pause); int mlx5_query_port_pause(struct mlx5_core_dev *dev, u32 *rx_pause, u32 *tx_pause) { - u32 in[MLX5_ST_SZ_DW(pfcc_reg)] = {0}; u32 out[MLX5_ST_SZ_DW(pfcc_reg)]; int err; - MLX5_SET(pfcc_reg, in, local_port, 1); - err = mlx5_core_access_reg(dev, in, sizeof(in), out, - sizeof(out), MLX5_REG_PFCC, 0, 0); + err = mlx5_query_pfcc_reg(dev, out, sizeof(out)); if (err) return err; @@ -520,6 +528,49 @@ int mlx5_query_port_pause(struct mlx5_core_dev *dev, } EXPORT_SYMBOL_GPL(mlx5_query_port_pause); +int mlx5_set_port_stall_watermark(struct mlx5_core_dev *dev, + u16 stall_critical_watermark, + u16 stall_minor_watermark) +{ + u32 in[MLX5_ST_SZ_DW(pfcc_reg)] = {0}; + u32 out[MLX5_ST_SZ_DW(pfcc_reg)]; + + MLX5_SET(pfcc_reg, in, local_port, 1); + MLX5_SET(pfcc_reg, in, pptx_mask_n, 1); + MLX5_SET(pfcc_reg, in, pprx_mask_n, 1); + MLX5_SET(pfcc_reg, in, ppan_mask_n, 1); + MLX5_SET(pfcc_reg, in, critical_stall_mask, 1); + MLX5_SET(pfcc_reg, in, minor_stall_mask, 1); + MLX5_SET(pfcc_reg, in, device_stall_critical_watermark, + stall_critical_watermark); + MLX5_SET(pfcc_reg, in, device_stall_minor_watermark, stall_minor_watermark); + + return mlx5_core_access_reg(dev, in, sizeof(in), out, + sizeof(out), MLX5_REG_PFCC, 0, 1); +} + +int mlx5_query_port_stall_watermark(struct mlx5_core_dev *dev, + u16 *stall_critical_watermark, + u16 *stall_minor_watermark) +{ + u32 out[MLX5_ST_SZ_DW(pfcc_reg)]; + int err; + + err = mlx5_query_pfcc_reg(dev, out, sizeof(out)); + if (err) + return err; + + if (stall_critical_watermark) + *stall_critical_watermark = MLX5_GET(pfcc_reg, out, + device_stall_critical_watermark); + + if (stall_minor_watermark) + *stall_minor_watermark = MLX5_GET(pfcc_reg, out, + device_stall_minor_watermark); + + return 0; +} + int mlx5_set_port_pfc(struct mlx5_core_dev *dev, u8 pfc_en_tx, u8 pfc_en_rx) { u32 in[MLX5_ST_SZ_DW(pfcc_reg)] = {0}; @@ -538,13 +589,10 @@ EXPORT_SYMBOL_GPL(mlx5_set_port_pfc); int mlx5_query_port_pfc(struct mlx5_core_dev *dev, u8 *pfc_en_tx, u8 *pfc_en_rx) { - u32 in[MLX5_ST_SZ_DW(pfcc_reg)] = {0}; u32 out[MLX5_ST_SZ_DW(pfcc_reg)]; int err; - MLX5_SET(pfcc_reg, in, local_port, 1); - err = mlx5_core_access_reg(dev, in, sizeof(in), out, - sizeof(out), MLX5_REG_PFCC, 0, 0); + err = mlx5_query_pfcc_reg(dev, out, sizeof(out)); if (err) return err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/transobj.c b/drivers/net/ethernet/mellanox/mlx5/core/transobj.c index 9e38343a951f..dae1c5c5d27c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/transobj.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/transobj.c @@ -157,6 +157,31 @@ int mlx5_core_query_sq(struct mlx5_core_dev *dev, u32 sqn, u32 *out) } EXPORT_SYMBOL(mlx5_core_query_sq); +int mlx5_core_query_sq_state(struct mlx5_core_dev *dev, u32 sqn, u8 *state) +{ + void *out; + void *sqc; + int inlen; + int err; + + inlen = MLX5_ST_SZ_BYTES(query_sq_out); + out = kvzalloc(inlen, GFP_KERNEL); + if (!out) + return -ENOMEM; + + err = mlx5_core_query_sq(dev, sqn, out); + if (err) + goto out; + + sqc = MLX5_ADDR_OF(query_sq_out, out, sq_context); + *state = MLX5_GET(sqc, sqc, state); + +out: + kvfree(out); + return err; +} +EXPORT_SYMBOL_GPL(mlx5_core_query_sq_state); + int mlx5_core_create_tir(struct mlx5_core_dev *dev, u32 *in, int inlen, u32 *tirn) { @@ -329,27 +354,6 @@ int mlx5_core_destroy_xsrq(struct mlx5_core_dev *dev, u32 xsrqn) return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out)); } -int mlx5_core_query_xsrq(struct mlx5_core_dev *dev, u32 xsrqn, u32 *out) -{ - u32 in[MLX5_ST_SZ_DW(query_xrc_srq_in)] = {0}; - void *srqc; - void *xrc_srqc; - int err; - - MLX5_SET(query_xrc_srq_in, in, opcode, MLX5_CMD_OP_QUERY_XRC_SRQ); - MLX5_SET(query_xrc_srq_in, in, xrc_srqn, xsrqn); - err = mlx5_cmd_exec(dev, in, sizeof(in), out, - MLX5_ST_SZ_BYTES(query_xrc_srq_out)); - if (!err) { - xrc_srqc = MLX5_ADDR_OF(query_xrc_srq_out, out, - xrc_srq_context_entry); - srqc = MLX5_ADDR_OF(query_srq_out, out, srq_context_entry); - memcpy(srqc, xrc_srqc, MLX5_ST_SZ_BYTES(srqc)); - } - - return err; -} - int mlx5_core_arm_xsrq(struct mlx5_core_dev *dev, u32 xsrqn, u16 lwm) { u32 in[MLX5_ST_SZ_DW(arm_xrc_srq_in)] = {0}; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/vport.c b/drivers/net/ethernet/mellanox/mlx5/core/vport.c index dfe36cf6fbea..177e076b8d17 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/vport.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/vport.c @@ -1070,6 +1070,32 @@ free: } EXPORT_SYMBOL_GPL(mlx5_core_query_vport_counter); +int mlx5_query_vport_down_stats(struct mlx5_core_dev *mdev, u16 vport, + u64 *rx_discard_vport_down, + u64 *tx_discard_vport_down) +{ + u32 out[MLX5_ST_SZ_DW(query_vnic_env_out)] = {0}; + u32 in[MLX5_ST_SZ_DW(query_vnic_env_in)] = {0}; + int err; + + MLX5_SET(query_vnic_env_in, in, opcode, + MLX5_CMD_OP_QUERY_VNIC_ENV); + MLX5_SET(query_vnic_env_in, in, op_mod, 0); + MLX5_SET(query_vnic_env_in, in, vport_number, vport); + if (vport) + MLX5_SET(query_vnic_env_in, in, other_vport, 1); + + err = mlx5_cmd_exec(mdev, in, sizeof(in), out, sizeof(out)); + if (err) + return err; + + *rx_discard_vport_down = MLX5_GET64(query_vnic_env_out, out, + vport_env.receive_discard_vport_down); + *tx_discard_vport_down = MLX5_GET64(query_vnic_env_out, out, + vport_env.transmit_discard_vport_down); + return 0; +} + int mlx5_core_modify_hca_vport_context(struct mlx5_core_dev *dev, u8 other_vport, u8 port_num, int vf, diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wq.c b/drivers/net/ethernet/mellanox/mlx5/core/wq.c index 6bcfc25350f5..ea66448ba365 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/wq.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/wq.c @@ -41,7 +41,7 @@ u32 mlx5_wq_cyc_get_size(struct mlx5_wq_cyc *wq) u32 mlx5_cqwq_get_size(struct mlx5_cqwq *wq) { - return wq->sz_m1 + 1; + return wq->fbc.sz_m1 + 1; } u32 mlx5_wq_ll_get_size(struct mlx5_wq_ll *wq) @@ -62,7 +62,7 @@ static u32 mlx5_wq_qp_get_byte_size(struct mlx5_wq_qp *wq) static u32 mlx5_cqwq_get_byte_size(struct mlx5_cqwq *wq) { - return mlx5_cqwq_get_size(wq) << wq->log_stride; + return mlx5_cqwq_get_size(wq) << wq->fbc.log_stride; } static u32 mlx5_wq_ll_get_byte_size(struct mlx5_wq_ll *wq) @@ -92,7 +92,7 @@ int mlx5_wq_cyc_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, goto err_db_free; } - wq->buf = wq_ctrl->buf.direct.buf; + wq->buf = wq_ctrl->buf.frags->buf; wq->db = wq_ctrl->db.db; wq_ctrl->mdev = mdev; @@ -130,7 +130,7 @@ int mlx5_wq_qp_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, goto err_db_free; } - wq->rq.buf = wq_ctrl->buf.direct.buf; + wq->rq.buf = wq_ctrl->buf.frags->buf; wq->sq.buf = wq->rq.buf + mlx5_wq_cyc_get_byte_size(&wq->rq); wq->rq.db = &wq_ctrl->db.db[MLX5_RCV_DBR]; wq->sq.db = &wq_ctrl->db.db[MLX5_SND_DBR]; @@ -151,11 +151,7 @@ int mlx5_cqwq_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, { int err; - wq->log_stride = 6 + MLX5_GET(cqc, cqc, cqe_sz); - wq->log_sz = MLX5_GET(cqc, cqc, log_cq_size); - wq->sz_m1 = (1 << wq->log_sz) - 1; - wq->log_frag_strides = PAGE_SHIFT - wq->log_stride; - wq->frag_sz_m1 = (1 << wq->log_frag_strides) - 1; + mlx5_core_init_cq_frag_buf(&wq->fbc, cqc); err = mlx5_db_alloc_node(mdev, &wq_ctrl->db, param->db_numa_node); if (err) { @@ -172,7 +168,7 @@ int mlx5_cqwq_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, goto err_db_free; } - wq->frag_buf = wq_ctrl->frag_buf; + wq->fbc.frag_buf = wq_ctrl->frag_buf; wq->db = wq_ctrl->db.db; wq_ctrl->mdev = mdev; @@ -209,7 +205,7 @@ int mlx5_wq_ll_create(struct mlx5_core_dev *mdev, struct mlx5_wq_param *param, goto err_db_free; } - wq->buf = wq_ctrl->buf.direct.buf; + wq->buf = wq_ctrl->buf.frags->buf; wq->db = wq_ctrl->db.db; for (i = 0; i < wq->sz_m1; i++) { diff --git a/drivers/net/ethernet/mellanox/mlx5/core/wq.h b/drivers/net/ethernet/mellanox/mlx5/core/wq.h index 718589d0cec2..fca90b94596d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/wq.h +++ b/drivers/net/ethernet/mellanox/mlx5/core/wq.h @@ -45,7 +45,7 @@ struct mlx5_wq_param { struct mlx5_wq_ctrl { struct mlx5_core_dev *mdev; - struct mlx5_buf buf; + struct mlx5_frag_buf buf; struct mlx5_db db; }; @@ -68,14 +68,9 @@ struct mlx5_wq_qp { }; struct mlx5_cqwq { - struct mlx5_frag_buf frag_buf; - __be32 *db; - u32 sz_m1; - u32 frag_sz_m1; - u32 cc; /* consumer counter */ - u8 log_sz; - u8 log_stride; - u8 log_frag_strides; + struct mlx5_frag_buf_ctrl fbc; + __be32 *db; + u32 cc; /* consumer counter */ }; struct mlx5_wq_ll { @@ -131,20 +126,17 @@ static inline int mlx5_wq_cyc_cc_bigger(u16 cc1, u16 cc2) static inline u32 mlx5_cqwq_get_ci(struct mlx5_cqwq *wq) { - return wq->cc & wq->sz_m1; + return wq->cc & wq->fbc.sz_m1; } static inline void *mlx5_cqwq_get_wqe(struct mlx5_cqwq *wq, u32 ix) { - unsigned int frag = (ix >> wq->log_frag_strides); - - return wq->frag_buf.frags[frag].buf + - ((wq->frag_sz_m1 & ix) << wq->log_stride); + return mlx5_frag_buf_get_wqe(&wq->fbc, ix); } static inline u32 mlx5_cqwq_get_wrap_cnt(struct mlx5_cqwq *wq) { - return wq->cc >> wq->log_sz; + return wq->cc >> wq->fbc.log_sz; } static inline void mlx5_cqwq_pop(struct mlx5_cqwq *wq) |