-/*-
- * BSD LICENSE
- *
- * Copyright 2015 6WIND S.A.
- * Copyright 2015 Mellanox.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions
- * are met:
- *
- * * Redistributions of source code must retain the above copyright
- * notice, this list of conditions and the following disclaimer.
- * * Redistributions in binary form must reproduce the above copyright
- * notice, this list of conditions and the following disclaimer in
- * the documentation and/or other materials provided with the
- * distribution.
- * * Neither the name of 6WIND S.A. nor the names of its
- * contributors may be used to endorse or promote products derived
- * from this software without specific prior written permission.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
- * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
- * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
- * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
- * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
- * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
- * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
- * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
- * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
- * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
- * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright 2015 6WIND S.A.
+ * Copyright 2015 Mellanox Technologies, Ltd
*/
#include <assert.h>
#pragma GCC diagnostic ignored "-Wpedantic"
#endif
#include <infiniband/verbs.h>
-#include <infiniband/mlx5_hw.h>
-#include <infiniband/arch.h>
+#include <infiniband/mlx5dv.h>
#ifdef PEDANTIC
#pragma GCC diagnostic error "-Wpedantic"
#endif
-/* DPDK headers don't like -pedantic. */
-#ifdef PEDANTIC
-#pragma GCC diagnostic ignored "-Wpedantic"
-#endif
#include <rte_mbuf.h>
#include <rte_mempool.h>
#include <rte_prefetch.h>
#include <rte_common.h>
#include <rte_branch_prediction.h>
#include <rte_ether.h>
-#ifdef PEDANTIC
-#pragma GCC diagnostic error "-Wpedantic"
-#endif
+#include <rte_cycles.h>
#include "mlx5.h"
#include "mlx5_utils.h"
#include "mlx5_defs.h"
#include "mlx5_prm.h"
-static inline int
-check_cqe(volatile struct mlx5_cqe *cqe,
- unsigned int cqes_n, const uint16_t ci)
- __attribute__((always_inline));
+static __rte_always_inline uint32_t
+rxq_cq_to_pkt_type(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cqe);
-static inline void
-txq_complete(struct txq *txq) __attribute__((always_inline));
+static __rte_always_inline int
+mlx5_rx_poll_len(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cqe,
+ uint16_t cqe_cnt, volatile struct mlx5_mini_cqe8 **mcqe);
-static inline uint32_t
-txq_mp2mr(struct txq *txq, struct rte_mempool *mp)
- __attribute__((always_inline));
+static __rte_always_inline uint32_t
+rxq_cq_to_ol_flags(volatile struct mlx5_cqe *cqe);
-static inline void
-mlx5_tx_dbrec(struct txq *txq, volatile struct mlx5_wqe *wqe)
- __attribute__((always_inline));
+static __rte_always_inline void
+rxq_cq_to_mbuf(struct mlx5_rxq_data *rxq, struct rte_mbuf *pkt,
+ volatile struct mlx5_cqe *cqe, uint32_t rss_hash_res);
-static inline uint32_t
-rxq_cq_to_pkt_type(volatile struct mlx5_cqe *cqe)
- __attribute__((always_inline));
+static __rte_always_inline void
+mprq_buf_replace(struct mlx5_rxq_data *rxq, uint16_t rq_idx);
-static inline int
-mlx5_rx_poll_len(struct rxq *rxq, volatile struct mlx5_cqe *cqe,
- uint16_t cqe_cnt, uint32_t *rss_hash)
- __attribute__((always_inline));
+static int
+mlx5_queue_state_modify(struct rte_eth_dev *dev,
+ struct mlx5_mp_arg_queue_state_modify *sm);
-static inline uint32_t
-rxq_cq_to_ol_flags(struct rxq *rxq, volatile struct mlx5_cqe *cqe)
- __attribute__((always_inline));
+uint32_t mlx5_ptype_table[] __rte_cache_aligned = {
+ [0xff] = RTE_PTYPE_ALL_MASK, /* Last entry for errored packet. */
+};
-#ifndef NDEBUG
+uint8_t mlx5_cksum_table[1 << 10] __rte_cache_aligned;
+uint8_t mlx5_swp_types_table[1 << 10] __rte_cache_aligned;
/**
- * Verify or set magic value in CQE.
- *
- * @param cqe
- * Pointer to CQE.
+ * Build a table to translate Rx completion flags to packet type.
*
- * @return
- * 0 the first time.
+ * @note: fix mlx5_dev_supported_ptypes_get() if any change here.
*/
-static inline int
-check_cqe_seen(volatile struct mlx5_cqe *cqe)
+void
+mlx5_set_ptype_table(void)
{
- static const uint8_t magic[] = "seen";
- volatile uint8_t (*buf)[sizeof(cqe->rsvd0)] = &cqe->rsvd0;
- int ret = 1;
unsigned int i;
+ uint32_t (*p)[RTE_DIM(mlx5_ptype_table)] = &mlx5_ptype_table;
- for (i = 0; i < sizeof(magic) && i < sizeof(*buf); ++i)
- if (!ret || (*buf)[i] != magic[i]) {
- ret = 0;
- (*buf)[i] = magic[i];
- }
- return ret;
-}
-
-#endif /* NDEBUG */
-
-/**
- * Check whether CQE is valid.
- *
- * @param cqe
- * Pointer to CQE.
- * @param cqes_n
- * Size of completion queue.
- * @param ci
- * Consumer index.
- *
- * @return
- * 0 on success, 1 on failure.
- */
-static inline int
-check_cqe(volatile struct mlx5_cqe *cqe,
- unsigned int cqes_n, const uint16_t ci)
-{
- uint16_t idx = ci & cqes_n;
- uint8_t op_own = cqe->op_own;
- uint8_t op_owner = MLX5_CQE_OWNER(op_own);
- uint8_t op_code = MLX5_CQE_OPCODE(op_own);
-
- if (unlikely((op_owner != (!!(idx))) || (op_code == MLX5_CQE_INVALID)))
- return 1; /* No CQE. */
-#ifndef NDEBUG
- if ((op_code == MLX5_CQE_RESP_ERR) ||
- (op_code == MLX5_CQE_REQ_ERR)) {
- volatile struct mlx5_err_cqe *err_cqe = (volatile void *)cqe;
- uint8_t syndrome = err_cqe->syndrome;
-
- if ((syndrome == MLX5_CQE_SYNDROME_LOCAL_LENGTH_ERR) ||
- (syndrome == MLX5_CQE_SYNDROME_REMOTE_ABORTED_ERR))
- return 0;
- if (!check_cqe_seen(cqe))
- ERROR("unexpected CQE error %u (0x%02x)"
- " syndrome 0x%02x",
- op_code, op_code, syndrome);
- return 1;
- } else if ((op_code != MLX5_CQE_RESP_SEND) &&
- (op_code != MLX5_CQE_REQ)) {
- if (!check_cqe_seen(cqe))
- ERROR("unexpected CQE opcode %u (0x%02x)",
- op_code, op_code);
- return 1;
- }
-#endif /* NDEBUG */
- return 0;
-}
-
-/**
- * Return the address of the WQE.
- *
- * @param txq
- * Pointer to TX queue structure.
- * @param wqe_ci
- * WQE consumer index.
- *
- * @return
- * WQE address.
- */
-static inline uintptr_t *
-tx_mlx5_wqe(struct txq *txq, uint16_t ci)
-{
- ci &= ((1 << txq->wqe_n) - 1);
- return (uintptr_t *)((uintptr_t)txq->wqes + ci * MLX5_WQE_SIZE);
-}
-
-/**
- * Return the size of tailroom of WQ.
- *
- * @param txq
- * Pointer to TX queue structure.
- * @param addr
- * Pointer to tail of WQ.
- *
- * @return
- * Size of tailroom.
- */
-static inline size_t
-tx_mlx5_wq_tailroom(struct txq *txq, void *addr)
-{
- size_t tailroom;
- tailroom = (uintptr_t)(txq->wqes) +
- (1 << txq->wqe_n) * MLX5_WQE_SIZE -
- (uintptr_t)addr;
- return tailroom;
-}
-
-/**
- * Copy data to tailroom of circular queue.
- *
- * @param dst
- * Pointer to destination.
- * @param src
- * Pointer to source.
- * @param n
- * Number of bytes to copy.
- * @param base
- * Pointer to head of queue.
- * @param tailroom
- * Size of tailroom from dst.
- *
- * @return
- * Pointer after copied data.
- */
-static inline void *
-mlx5_copy_to_wq(void *dst, const void *src, size_t n,
- void *base, size_t tailroom)
-{
- void *ret;
-
- if (n > tailroom) {
- rte_memcpy(dst, src, tailroom);
- rte_memcpy(base, (void *)((uintptr_t)src + tailroom),
- n - tailroom);
- ret = (uint8_t *)base + n - tailroom;
- } else {
- rte_memcpy(dst, src, n);
- ret = (n == tailroom) ? base : (uint8_t *)dst + n;
- }
- return ret;
+ /* Last entry must not be overwritten, reserved for errored packet. */
+ for (i = 0; i < RTE_DIM(mlx5_ptype_table) - 1; ++i)
+ (*p)[i] = RTE_PTYPE_UNKNOWN;
+ /*
+ * The index to the array should have:
+ * bit[1:0] = l3_hdr_type
+ * bit[4:2] = l4_hdr_type
+ * bit[5] = ip_frag
+ * bit[6] = tunneled
+ * bit[7] = outer_l3_type
+ */
+ /* L2 */
+ (*p)[0x00] = RTE_PTYPE_L2_ETHER;
+ /* L3 */
+ (*p)[0x01] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_NONFRAG;
+ (*p)[0x02] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_NONFRAG;
+ /* Fragmented */
+ (*p)[0x21] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_FRAG;
+ (*p)[0x22] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_FRAG;
+ /* TCP */
+ (*p)[0x05] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0x06] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0x0d] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0x0e] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0x11] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0x12] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ /* UDP */
+ (*p)[0x09] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_UDP;
+ (*p)[0x0a] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_UDP;
+ /* Repeat with outer_l3_type being set. Just in case. */
+ (*p)[0x81] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_NONFRAG;
+ (*p)[0x82] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_NONFRAG;
+ (*p)[0xa1] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_FRAG;
+ (*p)[0xa2] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_FRAG;
+ (*p)[0x85] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0x86] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0x8d] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0x8e] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0x91] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0x92] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0x89] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_UDP;
+ (*p)[0x8a] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_UDP;
+ /* Tunneled - L3 */
+ (*p)[0x40] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN;
+ (*p)[0x41] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_NONFRAG;
+ (*p)[0x42] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_NONFRAG;
+ (*p)[0xc0] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN;
+ (*p)[0xc1] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_NONFRAG;
+ (*p)[0xc2] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_NONFRAG;
+ /* Tunneled - Fragmented */
+ (*p)[0x61] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_FRAG;
+ (*p)[0x62] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_FRAG;
+ (*p)[0xe1] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_FRAG;
+ (*p)[0xe2] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_FRAG;
+ /* Tunneled - TCP */
+ (*p)[0x45] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_TCP;
+ (*p)[0x46] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_TCP;
+ (*p)[0x4d] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_TCP;
+ (*p)[0x4e] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_TCP;
+ (*p)[0x51] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_TCP;
+ (*p)[0x52] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_TCP;
+ (*p)[0xc5] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_TCP;
+ (*p)[0xc6] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_TCP;
+ (*p)[0xcd] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_TCP;
+ (*p)[0xce] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_TCP;
+ (*p)[0xd1] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_TCP;
+ (*p)[0xd2] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_TCP;
+ /* Tunneled - UDP */
+ (*p)[0x49] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_UDP;
+ (*p)[0x4a] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_UDP;
+ (*p)[0xc9] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_UDP;
+ (*p)[0xca] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L4_UDP;
}
/**
- * Manage TX completions.
- *
- * When sending a burst, mlx5_tx_burst() posts several WRs.
- *
- * @param txq
- * Pointer to TX queue structure.
+ * Build a table to translate packet to checksum type of Verbs.
*/
-static inline void
-txq_complete(struct txq *txq)
+void
+mlx5_set_cksum_table(void)
{
- const unsigned int elts_n = 1 << txq->elts_n;
- const unsigned int cqe_n = 1 << txq->cqe_n;
- const unsigned int cqe_cnt = cqe_n - 1;
- uint16_t elts_free = txq->elts_tail;
- uint16_t elts_tail;
- uint16_t cq_ci = txq->cq_ci;
- volatile struct mlx5_cqe *cqe = NULL;
- volatile struct mlx5_wqe_ctrl *ctrl;
-
- do {
- volatile struct mlx5_cqe *tmp;
+ unsigned int i;
+ uint8_t v;
- tmp = &(*txq->cqes)[cq_ci & cqe_cnt];
- if (check_cqe(tmp, cqe_n, cq_ci))
- break;
- cqe = tmp;
-#ifndef NDEBUG
- if (MLX5_CQE_FORMAT(cqe->op_own) == MLX5_COMPRESSED) {
- if (!check_cqe_seen(cqe))
- ERROR("unexpected compressed CQE, TX stopped");
- return;
- }
- if ((MLX5_CQE_OPCODE(cqe->op_own) == MLX5_CQE_RESP_ERR) ||
- (MLX5_CQE_OPCODE(cqe->op_own) == MLX5_CQE_REQ_ERR)) {
- if (!check_cqe_seen(cqe))
- ERROR("unexpected error CQE, TX stopped");
- return;
+ /*
+ * The index should have:
+ * bit[0] = PKT_TX_TCP_SEG
+ * bit[2:3] = PKT_TX_UDP_CKSUM, PKT_TX_TCP_CKSUM
+ * bit[4] = PKT_TX_IP_CKSUM
+ * bit[8] = PKT_TX_OUTER_IP_CKSUM
+ * bit[9] = tunnel
+ */
+ for (i = 0; i < RTE_DIM(mlx5_cksum_table); ++i) {
+ v = 0;
+ if (i & (1 << 9)) {
+ /* Tunneled packet. */
+ if (i & (1 << 8)) /* Outer IP. */
+ v |= MLX5_ETH_WQE_L3_CSUM;
+ if (i & (1 << 4)) /* Inner IP. */
+ v |= MLX5_ETH_WQE_L3_INNER_CSUM;
+ if (i & (3 << 2 | 1 << 0)) /* L4 or TSO. */
+ v |= MLX5_ETH_WQE_L4_INNER_CSUM;
+ } else {
+ /* No tunnel. */
+ if (i & (1 << 4)) /* IP. */
+ v |= MLX5_ETH_WQE_L3_CSUM;
+ if (i & (3 << 2 | 1 << 0)) /* L4 or TSO. */
+ v |= MLX5_ETH_WQE_L4_CSUM;
}
-#endif /* NDEBUG */
- ++cq_ci;
- } while (1);
- if (unlikely(cqe == NULL))
- return;
- txq->wqe_pi = ntohs(cqe->wqe_counter);
- ctrl = (volatile struct mlx5_wqe_ctrl *)
- tx_mlx5_wqe(txq, txq->wqe_pi);
- elts_tail = ctrl->ctrl3;
- assert(elts_tail < (1 << txq->wqe_n));
- /* Free buffers. */
- while (elts_free != elts_tail) {
- struct rte_mbuf *elt = (*txq->elts)[elts_free];
- unsigned int elts_free_next =
- (elts_free + 1) & (elts_n - 1);
- struct rte_mbuf *elt_next = (*txq->elts)[elts_free_next];
-
-#ifndef NDEBUG
- /* Poisoning. */
- memset(&(*txq->elts)[elts_free],
- 0x66,
- sizeof((*txq->elts)[elts_free]));
-#endif
- RTE_MBUF_PREFETCH_TO_FREE(elt_next);
- /* Only one segment needs to be freed. */
- rte_pktmbuf_free_seg(elt);
- elts_free = elts_free_next;
+ mlx5_cksum_table[i] = v;
}
- txq->cq_ci = cq_ci;
- txq->elts_tail = elts_tail;
- /* Update the consumer index. */
- rte_wmb();
- *txq->cq_db = htonl(cq_ci);
}
/**
- * Get Memory Pool (MP) from mbuf. If mbuf is indirect, the pool from which
- * the cloned mbuf is allocated is returned instead.
- *
- * @param buf
- * Pointer to mbuf.
- *
- * @return
- * Memory pool where data is located for given mbuf.
- */
-static struct rte_mempool *
-txq_mb2mp(struct rte_mbuf *buf)
-{
- if (unlikely(RTE_MBUF_INDIRECT(buf)))
- return rte_mbuf_from_indirect(buf)->pool;
- return buf->pool;
-}
-
-/**
- * Get Memory Region (MR) <-> Memory Pool (MP) association from txq->mp2mr[].
- * Add MP to txq->mp2mr[] if it's not registered yet. If mp2mr[] is full,
- * remove an entry first.
- *
- * @param txq
- * Pointer to TX queue structure.
- * @param[in] mp
- * Memory Pool for which a Memory Region lkey must be returned.
- *
- * @return
- * mr->lkey on success, (uint32_t)-1 on failure.
+ * Build a table to translate packet type of mbuf to SWP type of Verbs.
*/
-static inline uint32_t
-txq_mp2mr(struct txq *txq, struct rte_mempool *mp)
+void
+mlx5_set_swp_types_table(void)
{
unsigned int i;
- uint32_t lkey = (uint32_t)-1;
+ uint8_t v;
- for (i = 0; (i != RTE_DIM(txq->mp2mr)); ++i) {
- if (unlikely(txq->mp2mr[i].mp == NULL)) {
- /* Unknown MP, add a new MR for it. */
- break;
- }
- if (txq->mp2mr[i].mp == mp) {
- assert(txq->mp2mr[i].lkey != (uint32_t)-1);
- assert(htonl(txq->mp2mr[i].mr->lkey) ==
- txq->mp2mr[i].lkey);
- lkey = txq->mp2mr[i].lkey;
- break;
- }
+ /*
+ * The index should have:
+ * bit[0:1] = PKT_TX_L4_MASK
+ * bit[4] = PKT_TX_IPV6
+ * bit[8] = PKT_TX_OUTER_IPV6
+ * bit[9] = PKT_TX_OUTER_UDP
+ */
+ for (i = 0; i < RTE_DIM(mlx5_swp_types_table); ++i) {
+ v = 0;
+ if (i & (1 << 8))
+ v |= MLX5_ETH_WQE_L3_OUTER_IPV6;
+ if (i & (1 << 9))
+ v |= MLX5_ETH_WQE_L4_OUTER_UDP;
+ if (i & (1 << 4))
+ v |= MLX5_ETH_WQE_L3_INNER_IPV6;
+ if ((i & 3) == (PKT_TX_UDP_CKSUM >> 52))
+ v |= MLX5_ETH_WQE_L4_INNER_UDP;
+ mlx5_swp_types_table[i] = v;
}
- if (unlikely(lkey == (uint32_t)-1))
- lkey = txq_mp2mr_reg(txq, mp, i);
- return lkey;
-}
-
-/**
- * Ring TX queue doorbell.
- *
- * @param txq
- * Pointer to TX queue structure.
- * @param wqe
- * Pointer to the last WQE posted in the NIC.
- */
-static inline void
-mlx5_tx_dbrec(struct txq *txq, volatile struct mlx5_wqe *wqe)
-{
- uint64_t *dst = (uint64_t *)((uintptr_t)txq->bf_reg);
- volatile uint64_t *src = ((volatile uint64_t *)wqe);
-
- rte_wmb();
- *txq->qp_db = htonl(txq->wqe_ci);
- /* Ensure ordering between DB record and BF copy. */
- rte_wmb();
- *dst = *src;
-}
-
-/**
- * DPDK callback to check the status of a tx descriptor.
- *
- * @param tx_queue
- * The tx queue.
- * @param[in] offset
- * The index of the descriptor in the ring.
- *
- * @return
- * The status of the tx descriptor.
- */
-int
-mlx5_tx_descriptor_status(void *tx_queue, uint16_t offset)
-{
- struct txq *txq = tx_queue;
- const unsigned int elts_n = 1 << txq->elts_n;
- const unsigned int elts_cnt = elts_n - 1;
- unsigned int used;
-
- txq_complete(txq);
- used = (txq->elts_head - txq->elts_tail) & elts_cnt;
- if (offset < used)
- return RTE_ETH_TX_DESC_FULL;
- return RTE_ETH_TX_DESC_DONE;
}
/**
- * DPDK callback to check the status of a rx descriptor.
+ * Internal function to compute the number of used descriptors in an RX queue
*
- * @param rx_queue
- * The rx queue.
- * @param[in] offset
- * The index of the descriptor in the ring.
+ * @param rxq
+ * The Rx queue.
*
* @return
- * The status of the tx descriptor.
+ * The number of used rx descriptor.
*/
-int
-mlx5_rx_descriptor_status(void *rx_queue, uint16_t offset)
+static uint32_t
+rx_queue_count(struct mlx5_rxq_data *rxq)
{
- struct rxq *rxq = rx_queue;
struct rxq_zip *zip = &rxq->zip;
volatile struct mlx5_cqe *cqe;
const unsigned int cqe_n = (1 << rxq->cqe_n);
cq_ci = rxq->cq_ci;
}
cqe = &(*rxq->cqes)[cq_ci & cqe_cnt];
- while (check_cqe(cqe, cqe_n, cq_ci) == 0) {
+ while (check_cqe(cqe, cqe_n, cq_ci) != MLX5_CQE_STATUS_HW_OWN) {
int8_t op_own;
unsigned int n;
op_own = cqe->op_own;
if (MLX5_CQE_FORMAT(op_own) == MLX5_COMPRESSED)
- n = ntohl(cqe->byte_cnt);
+ n = rte_be_to_cpu_32(cqe->byte_cnt);
else
n = 1;
cq_ci += n;
cqe = &(*rxq->cqes)[cq_ci & cqe_cnt];
}
used = RTE_MIN(used, (1U << rxq->elts_n) - 1);
- if (offset < used)
- return RTE_ETH_RX_DESC_DONE;
- return RTE_ETH_RX_DESC_AVAIL;
+ return used;
}
/**
- * DPDK callback for TX.
+ * DPDK callback to check the status of a rx descriptor.
*
- * @param dpdk_txq
- * Generic pointer to TX queue structure.
- * @param[in] pkts
- * Packets to transmit.
- * @param pkts_n
- * Number of packets in array.
+ * @param rx_queue
+ * The Rx queue.
+ * @param[in] offset
+ * The index of the descriptor in the ring.
*
* @return
- * Number of packets successfully transmitted (<= pkts_n).
+ * The status of the tx descriptor.
*/
-uint16_t
-mlx5_tx_burst(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
+int
+mlx5_rx_descriptor_status(void *rx_queue, uint16_t offset)
{
- struct txq *txq = (struct txq *)dpdk_txq;
- uint16_t elts_head = txq->elts_head;
- const unsigned int elts_n = 1 << txq->elts_n;
- unsigned int i = 0;
- unsigned int j = 0;
- unsigned int k = 0;
- unsigned int max;
- uint16_t max_wqe;
- unsigned int comp;
- volatile struct mlx5_wqe_v *wqe = NULL;
- unsigned int segs_n = 0;
- struct rte_mbuf *buf = NULL;
- uint8_t *raw;
-
- if (unlikely(!pkts_n))
- return 0;
- /* Prefetch first packet cacheline. */
- rte_prefetch0(*pkts);
- /* Start processing. */
- txq_complete(txq);
- max = (elts_n - (elts_head - txq->elts_tail));
- if (max > elts_n)
- max -= elts_n;
- max_wqe = (1u << txq->wqe_n) - (txq->wqe_ci - txq->wqe_pi);
- if (unlikely(!max_wqe))
- return 0;
- do {
- volatile rte_v128u32_t *dseg = NULL;
- uint32_t length;
- unsigned int ds = 0;
- uintptr_t addr;
- uint64_t naddr;
- uint16_t pkt_inline_sz = MLX5_WQE_DWORD_SIZE + 2;
- uint16_t tso_header_sz = 0;
- uint16_t ehdr;
- uint8_t cs_flags = 0;
- uint64_t tso = 0;
-#ifdef MLX5_PMD_SOFT_COUNTERS
- uint32_t total_length = 0;
-#endif
-
- /* first_seg */
- buf = *(pkts++);
- segs_n = buf->nb_segs;
- /*
- * Make sure there is enough room to store this packet and
- * that one ring entry remains unused.
- */
- assert(segs_n);
- if (max < segs_n + 1)
- break;
- max -= segs_n;
- --segs_n;
- if (!segs_n)
- --pkts_n;
- if (unlikely(--max_wqe == 0))
- break;
- wqe = (volatile struct mlx5_wqe_v *)
- tx_mlx5_wqe(txq, txq->wqe_ci);
- rte_prefetch0(tx_mlx5_wqe(txq, txq->wqe_ci + 1));
- if (pkts_n > 1)
- rte_prefetch0(*pkts);
- addr = rte_pktmbuf_mtod(buf, uintptr_t);
- length = DATA_LEN(buf);
- ehdr = (((uint8_t *)addr)[1] << 8) |
- ((uint8_t *)addr)[0];
-#ifdef MLX5_PMD_SOFT_COUNTERS
- total_length = length;
-#endif
- assert(length >= MLX5_WQE_DWORD_SIZE);
- /* Update element. */
- (*txq->elts)[elts_head] = buf;
- elts_head = (elts_head + 1) & (elts_n - 1);
- /* Prefetch next buffer data. */
- if (pkts_n > 1) {
- volatile void *pkt_addr;
-
- pkt_addr = rte_pktmbuf_mtod(*pkts, volatile void *);
- rte_prefetch0(pkt_addr);
- }
- /* Should we enable HW CKSUM offload */
- if (buf->ol_flags &
- (PKT_TX_IP_CKSUM | PKT_TX_TCP_CKSUM | PKT_TX_UDP_CKSUM)) {
- const uint64_t is_tunneled = buf->ol_flags &
- (PKT_TX_TUNNEL_GRE |
- PKT_TX_TUNNEL_VXLAN);
-
- if (is_tunneled && txq->tunnel_en) {
- cs_flags = MLX5_ETH_WQE_L3_INNER_CSUM |
- MLX5_ETH_WQE_L4_INNER_CSUM;
- if (buf->ol_flags & PKT_TX_OUTER_IP_CKSUM)
- cs_flags |= MLX5_ETH_WQE_L3_CSUM;
- } else {
- cs_flags = MLX5_ETH_WQE_L3_CSUM |
- MLX5_ETH_WQE_L4_CSUM;
- }
- }
- raw = ((uint8_t *)(uintptr_t)wqe) + 2 * MLX5_WQE_DWORD_SIZE;
- /* Replace the Ethernet type by the VLAN if necessary. */
- if (buf->ol_flags & PKT_TX_VLAN_PKT) {
- uint32_t vlan = htonl(0x81000000 | buf->vlan_tci);
- unsigned int len = 2 * ETHER_ADDR_LEN - 2;
-
- addr += 2;
- length -= 2;
- /* Copy Destination and source mac address. */
- memcpy((uint8_t *)raw, ((uint8_t *)addr), len);
- /* Copy VLAN. */
- memcpy((uint8_t *)raw + len, &vlan, sizeof(vlan));
- /* Copy missing two bytes to end the DSeg. */
- memcpy((uint8_t *)raw + len + sizeof(vlan),
- ((uint8_t *)addr) + len, 2);
- addr += len + 2;
- length -= (len + 2);
- } else {
- memcpy((uint8_t *)raw, ((uint8_t *)addr) + 2,
- MLX5_WQE_DWORD_SIZE);
- length -= pkt_inline_sz;
- addr += pkt_inline_sz;
- }
- if (txq->tso_en) {
- tso = buf->ol_flags & PKT_TX_TCP_SEG;
- if (tso) {
- uintptr_t end = (uintptr_t)
- (((uintptr_t)txq->wqes) +
- (1 << txq->wqe_n) *
- MLX5_WQE_SIZE);
- unsigned int copy_b;
- uint8_t vlan_sz = (buf->ol_flags &
- PKT_TX_VLAN_PKT) ? 4 : 0;
- const uint64_t is_tunneled =
- buf->ol_flags &
- (PKT_TX_TUNNEL_GRE |
- PKT_TX_TUNNEL_VXLAN);
-
- tso_header_sz = buf->l2_len + vlan_sz +
- buf->l3_len + buf->l4_len;
-
- if (is_tunneled && txq->tunnel_en) {
- tso_header_sz += buf->outer_l2_len +
- buf->outer_l3_len;
- cs_flags |= MLX5_ETH_WQE_L4_INNER_CSUM;
- } else {
- cs_flags |= MLX5_ETH_WQE_L4_CSUM;
- }
- if (unlikely(tso_header_sz >
- MLX5_MAX_TSO_HEADER))
- break;
- copy_b = tso_header_sz - pkt_inline_sz;
- /* First seg must contain all headers. */
- assert(copy_b <= length);
- raw += MLX5_WQE_DWORD_SIZE;
- if (copy_b &&
- ((end - (uintptr_t)raw) > copy_b)) {
- uint16_t n = (MLX5_WQE_DS(copy_b) -
- 1 + 3) / 4;
-
- if (unlikely(max_wqe < n))
- break;
- max_wqe -= n;
- rte_memcpy((void *)raw,
- (void *)addr, copy_b);
- addr += copy_b;
- length -= copy_b;
- pkt_inline_sz += copy_b;
- /*
- * Another DWORD will be added
- * in the inline part.
- */
- raw += MLX5_WQE_DS(copy_b) *
- MLX5_WQE_DWORD_SIZE -
- MLX5_WQE_DWORD_SIZE;
- } else {
- /* NOP WQE. */
- wqe->ctrl = (rte_v128u32_t){
- htonl(txq->wqe_ci << 8),
- htonl(txq->qp_num_8s | 1),
- 0,
- 0,
- };
- ds = 1;
- total_length = 0;
- pkts--;
- pkts_n++;
- elts_head = (elts_head - 1) &
- (elts_n - 1);
- k++;
- goto next_wqe;
- }
- }
- }
- /* Inline if enough room. */
- if (txq->inline_en || tso) {
- uintptr_t end = (uintptr_t)
- (((uintptr_t)txq->wqes) +
- (1 << txq->wqe_n) * MLX5_WQE_SIZE);
- unsigned int max_inline = txq->max_inline *
- RTE_CACHE_LINE_SIZE -
- (pkt_inline_sz - 2);
- uintptr_t addr_end = (addr + max_inline) &
- ~(RTE_CACHE_LINE_SIZE - 1);
- unsigned int copy_b = (addr_end > addr) ?
- RTE_MIN((addr_end - addr), length) :
- 0;
-
- raw += MLX5_WQE_DWORD_SIZE;
- if (copy_b && ((end - (uintptr_t)raw) > copy_b)) {
- /*
- * One Dseg remains in the current WQE. To
- * keep the computation positive, it is
- * removed after the bytes to Dseg conversion.
- */
- uint16_t n = (MLX5_WQE_DS(copy_b) - 1 + 3) / 4;
-
- if (unlikely(max_wqe < n))
- break;
- max_wqe -= n;
- if (tso) {
- uint32_t inl =
- htonl(copy_b | MLX5_INLINE_SEG);
-
- pkt_inline_sz =
- MLX5_WQE_DS(tso_header_sz) *
- MLX5_WQE_DWORD_SIZE;
- rte_memcpy((void *)raw,
- (void *)&inl, sizeof(inl));
- raw += sizeof(inl);
- pkt_inline_sz += sizeof(inl);
- }
- rte_memcpy((void *)raw, (void *)addr, copy_b);
- addr += copy_b;
- length -= copy_b;
- pkt_inline_sz += copy_b;
- }
- /*
- * 2 DWORDs consumed by the WQE header + ETH segment +
- * the size of the inline part of the packet.
- */
- ds = 2 + MLX5_WQE_DS(pkt_inline_sz - 2);
- if (length > 0) {
- if (ds % (MLX5_WQE_SIZE /
- MLX5_WQE_DWORD_SIZE) == 0) {
- if (unlikely(--max_wqe == 0))
- break;
- dseg = (volatile rte_v128u32_t *)
- tx_mlx5_wqe(txq, txq->wqe_ci +
- ds / 4);
- } else {
- dseg = (volatile rte_v128u32_t *)
- ((uintptr_t)wqe +
- (ds * MLX5_WQE_DWORD_SIZE));
- }
- goto use_dseg;
- } else if (!segs_n) {
- goto next_pkt;
- } else {
- /* dseg will be advance as part of next_seg */
- dseg = (volatile rte_v128u32_t *)
- ((uintptr_t)wqe +
- ((ds - 1) * MLX5_WQE_DWORD_SIZE));
- goto next_seg;
- }
- } else {
- /*
- * No inline has been done in the packet, only the
- * Ethernet Header as been stored.
- */
- dseg = (volatile rte_v128u32_t *)
- ((uintptr_t)wqe + (3 * MLX5_WQE_DWORD_SIZE));
- ds = 3;
-use_dseg:
- /* Add the remaining packet as a simple ds. */
- naddr = htonll(addr);
- *dseg = (rte_v128u32_t){
- htonl(length),
- txq_mp2mr(txq, txq_mb2mp(buf)),
- naddr,
- naddr >> 32,
- };
- ++ds;
- if (!segs_n)
- goto next_pkt;
- }
-next_seg:
- assert(buf);
- assert(ds);
- assert(wqe);
- /*
- * Spill on next WQE when the current one does not have
- * enough room left. Size of WQE must a be a multiple
- * of data segment size.
- */
- assert(!(MLX5_WQE_SIZE % MLX5_WQE_DWORD_SIZE));
- if (!(ds % (MLX5_WQE_SIZE / MLX5_WQE_DWORD_SIZE))) {
- if (unlikely(--max_wqe == 0))
- break;
- dseg = (volatile rte_v128u32_t *)
- tx_mlx5_wqe(txq, txq->wqe_ci + ds / 4);
- rte_prefetch0(tx_mlx5_wqe(txq,
- txq->wqe_ci + ds / 4 + 1));
- } else {
- ++dseg;
- }
- ++ds;
- buf = buf->next;
- assert(buf);
- length = DATA_LEN(buf);
-#ifdef MLX5_PMD_SOFT_COUNTERS
- total_length += length;
-#endif
- /* Store segment information. */
- naddr = htonll(rte_pktmbuf_mtod(buf, uintptr_t));
- *dseg = (rte_v128u32_t){
- htonl(length),
- txq_mp2mr(txq, txq_mb2mp(buf)),
- naddr,
- naddr >> 32,
- };
- (*txq->elts)[elts_head] = buf;
- elts_head = (elts_head + 1) & (elts_n - 1);
- ++j;
- --segs_n;
- if (segs_n)
- goto next_seg;
- else
- --pkts_n;
-next_pkt:
- ++i;
- /* Initialize known and common part of the WQE structure. */
- if (tso) {
- wqe->ctrl = (rte_v128u32_t){
- htonl((txq->wqe_ci << 8) | MLX5_OPCODE_TSO),
- htonl(txq->qp_num_8s | ds),
- 0,
- 0,
- };
- wqe->eseg = (rte_v128u32_t){
- 0,
- cs_flags | (htons(buf->tso_segsz) << 16),
- 0,
- (ehdr << 16) | htons(tso_header_sz),
- };
- } else {
- wqe->ctrl = (rte_v128u32_t){
- htonl((txq->wqe_ci << 8) | MLX5_OPCODE_SEND),
- htonl(txq->qp_num_8s | ds),
- 0,
- 0,
- };
- wqe->eseg = (rte_v128u32_t){
- 0,
- cs_flags,
- 0,
- (ehdr << 16) | htons(pkt_inline_sz),
- };
- }
-next_wqe:
- txq->wqe_ci += (ds + 3) / 4;
-#ifdef MLX5_PMD_SOFT_COUNTERS
- /* Increment sent bytes counter. */
- txq->stats.obytes += total_length;
-#endif
- } while (pkts_n);
- /* Take a shortcut if nothing must be sent. */
- if (unlikely((i + k) == 0))
- return 0;
- /* Check whether completion threshold has been reached. */
- comp = txq->elts_comp + i + j + k;
- if (comp >= MLX5_TX_COMP_THRESH) {
- volatile struct mlx5_wqe_ctrl *w =
- (volatile struct mlx5_wqe_ctrl *)wqe;
-
- /* Request completion on last WQE. */
- w->ctrl2 = htonl(8);
- /* Save elts_head in unused "immediate" field of WQE. */
- w->ctrl3 = elts_head;
- txq->elts_comp = 0;
- } else {
- txq->elts_comp = comp;
+ struct mlx5_rxq_data *rxq = rx_queue;
+ struct mlx5_rxq_ctrl *rxq_ctrl =
+ container_of(rxq, struct mlx5_rxq_ctrl, rxq);
+ struct rte_eth_dev *dev = ETH_DEV(rxq_ctrl->priv);
+
+ if (dev->rx_pkt_burst != mlx5_rx_burst) {
+ rte_errno = ENOTSUP;
+ return -rte_errno;
}
-#ifdef MLX5_PMD_SOFT_COUNTERS
- /* Increment sent packets counter. */
- txq->stats.opackets += i;
-#endif
- /* Ring QP doorbell. */
- mlx5_tx_dbrec(txq, (volatile struct mlx5_wqe *)wqe);
- txq->elts_head = elts_head;
- return i;
+ if (offset >= (1 << rxq->elts_n)) {
+ rte_errno = EINVAL;
+ return -rte_errno;
+ }
+ if (offset < rx_queue_count(rxq))
+ return RTE_ETH_RX_DESC_DONE;
+ return RTE_ETH_RX_DESC_AVAIL;
}
/**
- * Open a MPW session.
+ * DPDK callback to get the number of used descriptors in a RX queue
*
- * @param txq
- * Pointer to TX queue structure.
- * @param mpw
- * Pointer to MPW session structure.
- * @param length
- * Packet length.
+ * @param dev
+ * Pointer to the device structure.
+ *
+ * @param rx_queue_id
+ * The Rx queue.
+ *
+ * @return
+ * The number of used rx descriptor.
+ * -EINVAL if the queue is invalid
*/
-static inline void
-mlx5_mpw_new(struct txq *txq, struct mlx5_mpw *mpw, uint32_t length)
+uint32_t
+mlx5_rx_queue_count(struct rte_eth_dev *dev, uint16_t rx_queue_id)
{
- uint16_t idx = txq->wqe_ci & ((1 << txq->wqe_n) - 1);
- volatile struct mlx5_wqe_data_seg (*dseg)[MLX5_MPW_DSEG_MAX] =
- (volatile struct mlx5_wqe_data_seg (*)[])
- tx_mlx5_wqe(txq, idx + 1);
-
- mpw->state = MLX5_MPW_STATE_OPENED;
- mpw->pkts_n = 0;
- mpw->len = length;
- mpw->total_len = 0;
- mpw->wqe = (volatile struct mlx5_wqe *)tx_mlx5_wqe(txq, idx);
- mpw->wqe->eseg.mss = htons(length);
- mpw->wqe->eseg.inline_hdr_sz = 0;
- mpw->wqe->eseg.rsvd0 = 0;
- mpw->wqe->eseg.rsvd1 = 0;
- mpw->wqe->eseg.rsvd2 = 0;
- mpw->wqe->ctrl[0] = htonl((MLX5_OPC_MOD_MPW << 24) |
- (txq->wqe_ci << 8) | MLX5_OPCODE_TSO);
- mpw->wqe->ctrl[2] = 0;
- mpw->wqe->ctrl[3] = 0;
- mpw->data.dseg[0] = (volatile struct mlx5_wqe_data_seg *)
- (((uintptr_t)mpw->wqe) + (2 * MLX5_WQE_DWORD_SIZE));
- mpw->data.dseg[1] = (volatile struct mlx5_wqe_data_seg *)
- (((uintptr_t)mpw->wqe) + (3 * MLX5_WQE_DWORD_SIZE));
- mpw->data.dseg[2] = &(*dseg)[0];
- mpw->data.dseg[3] = &(*dseg)[1];
- mpw->data.dseg[4] = &(*dseg)[2];
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_rxq_data *rxq;
+
+ if (dev->rx_pkt_burst != mlx5_rx_burst) {
+ rte_errno = ENOTSUP;
+ return -rte_errno;
+ }
+ rxq = (*priv->rxqs)[rx_queue_id];
+ if (!rxq) {
+ rte_errno = EINVAL;
+ return -rte_errno;
+ }
+ return rx_queue_count(rxq);
}
+#define MLX5_SYSTEM_LOG_DIR "/var/log"
/**
- * Close a MPW session.
+ * Dump debug information to log file.
*
- * @param txq
- * Pointer to TX queue structure.
- * @param mpw
- * Pointer to MPW session structure.
+ * @param fname
+ * The file name.
+ * @param hex_title
+ * If not NULL this string is printed as a header to the output
+ * and the output will be in hexadecimal view.
+ * @param buf
+ * This is the buffer address to print out.
+ * @param len
+ * The number of bytes to dump out.
*/
-static inline void
-mlx5_mpw_close(struct txq *txq, struct mlx5_mpw *mpw)
+void
+mlx5_dump_debug_information(const char *fname, const char *hex_title,
+ const void *buf, unsigned int hex_len)
{
- unsigned int num = mpw->pkts_n;
-
- /*
- * Store size in multiple of 16 bytes. Control and Ethernet segments
- * count as 2.
- */
- mpw->wqe->ctrl[1] = htonl(txq->qp_num_8s | (2 + num));
- mpw->state = MLX5_MPW_STATE_CLOSED;
- if (num < 3)
- ++txq->wqe_ci;
+ FILE *fd;
+
+ MKSTR(path, "%s/%s", MLX5_SYSTEM_LOG_DIR, fname);
+ fd = fopen(path, "a+");
+ if (!fd) {
+ DRV_LOG(WARNING, "cannot open %s for debug dump\n",
+ path);
+ MKSTR(path2, "./%s", fname);
+ fd = fopen(path2, "a+");
+ if (!fd) {
+ DRV_LOG(ERR, "cannot open %s for debug dump\n",
+ path2);
+ return;
+ }
+ DRV_LOG(INFO, "New debug dump in file %s\n", path2);
+ } else {
+ DRV_LOG(INFO, "New debug dump in file %s\n", path);
+ }
+ if (hex_title)
+ rte_hexdump(fd, hex_title, buf, hex_len);
else
- txq->wqe_ci += 2;
- rte_prefetch0(tx_mlx5_wqe(txq, txq->wqe_ci));
- rte_prefetch0(tx_mlx5_wqe(txq, txq->wqe_ci + 1));
+ fprintf(fd, "%s", (const char *)buf);
+ fprintf(fd, "\n\n\n");
+ fclose(fd);
}
/**
- * DPDK callback for TX with MPW support.
+ * Move QP from error state to running state and initialize indexes.
*
- * @param dpdk_txq
- * Generic pointer to TX queue structure.
- * @param[in] pkts
- * Packets to transmit.
- * @param pkts_n
- * Number of packets in array.
+ * @param txq_ctrl
+ * Pointer to TX queue control structure.
*
* @return
- * Number of packets successfully transmitted (<= pkts_n).
+ * 0 on success, else -1.
*/
-uint16_t
-mlx5_tx_burst_mpw(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
+static int
+tx_recover_qp(struct mlx5_txq_ctrl *txq_ctrl)
{
- struct txq *txq = (struct txq *)dpdk_txq;
- uint16_t elts_head = txq->elts_head;
- const unsigned int elts_n = 1 << txq->elts_n;
- unsigned int i = 0;
- unsigned int j = 0;
- unsigned int max;
- uint16_t max_wqe;
- unsigned int comp;
- struct mlx5_mpw mpw = {
- .state = MLX5_MPW_STATE_CLOSED,
+ struct mlx5_mp_arg_queue_state_modify sm = {
+ .is_wq = 0,
+ .queue_id = txq_ctrl->txq.idx,
};
- if (unlikely(!pkts_n))
- return 0;
- /* Prefetch first packet cacheline. */
- rte_prefetch0(tx_mlx5_wqe(txq, txq->wqe_ci));
- rte_prefetch0(tx_mlx5_wqe(txq, txq->wqe_ci + 1));
- /* Start processing. */
- txq_complete(txq);
- max = (elts_n - (elts_head - txq->elts_tail));
- if (max > elts_n)
- max -= elts_n;
- max_wqe = (1u << txq->wqe_n) - (txq->wqe_ci - txq->wqe_pi);
- if (unlikely(!max_wqe))
- return 0;
- do {
- struct rte_mbuf *buf = *(pkts++);
- unsigned int elts_head_next;
- uint32_t length;
- unsigned int segs_n = buf->nb_segs;
- uint32_t cs_flags = 0;
+ if (mlx5_queue_state_modify(ETH_DEV(txq_ctrl->priv), &sm))
+ return -1;
+ txq_ctrl->txq.wqe_ci = 0;
+ txq_ctrl->txq.wqe_pi = 0;
+ txq_ctrl->txq.elts_comp = 0;
+ return 0;
+}
- /*
- * Make sure there is enough room to store this packet and
- * that one ring entry remains unused.
- */
- assert(segs_n);
- if (max < segs_n + 1)
- break;
- /* Do not bother with large packets MPW cannot handle. */
- if (segs_n > MLX5_MPW_DSEG_MAX)
- break;
- max -= segs_n;
- --pkts_n;
- /* Should we enable HW CKSUM offload */
- if (buf->ol_flags &
- (PKT_TX_IP_CKSUM | PKT_TX_TCP_CKSUM | PKT_TX_UDP_CKSUM))
- cs_flags = MLX5_ETH_WQE_L3_CSUM | MLX5_ETH_WQE_L4_CSUM;
- /* Retrieve packet information. */
- length = PKT_LEN(buf);
- assert(length);
- /* Start new session if packet differs. */
- if ((mpw.state == MLX5_MPW_STATE_OPENED) &&
- ((mpw.len != length) ||
- (segs_n != 1) ||
- (mpw.wqe->eseg.cs_flags != cs_flags)))
- mlx5_mpw_close(txq, &mpw);
- if (mpw.state == MLX5_MPW_STATE_CLOSED) {
- /*
- * Multi-Packet WQE consumes at most two WQE.
- * mlx5_mpw_new() expects to be able to use such
- * resources.
- */
- if (unlikely(max_wqe < 2))
- break;
- max_wqe -= 2;
- mlx5_mpw_new(txq, &mpw, length);
- mpw.wqe->eseg.cs_flags = cs_flags;
+/* Return 1 if the error CQE is signed otherwise, sign it and return 0. */
+static int
+check_err_cqe_seen(volatile struct mlx5_err_cqe *err_cqe)
+{
+ static const uint8_t magic[] = "seen";
+ int ret = 1;
+ unsigned int i;
+
+ for (i = 0; i < sizeof(magic); ++i)
+ if (!ret || err_cqe->rsvd1[i] != magic[i]) {
+ ret = 0;
+ err_cqe->rsvd1[i] = magic[i];
}
- /* Multi-segment packets must be alone in their MPW. */
- assert((segs_n == 1) || (mpw.pkts_n == 0));
-#if defined(MLX5_PMD_SOFT_COUNTERS) || !defined(NDEBUG)
- length = 0;
-#endif
- do {
- volatile struct mlx5_wqe_data_seg *dseg;
- uintptr_t addr;
-
- elts_head_next = (elts_head + 1) & (elts_n - 1);
- assert(buf);
- (*txq->elts)[elts_head] = buf;
- dseg = mpw.data.dseg[mpw.pkts_n];
- addr = rte_pktmbuf_mtod(buf, uintptr_t);
- *dseg = (struct mlx5_wqe_data_seg){
- .byte_count = htonl(DATA_LEN(buf)),
- .lkey = txq_mp2mr(txq, txq_mb2mp(buf)),
- .addr = htonll(addr),
- };
- elts_head = elts_head_next;
-#if defined(MLX5_PMD_SOFT_COUNTERS) || !defined(NDEBUG)
- length += DATA_LEN(buf);
-#endif
- buf = buf->next;
- ++mpw.pkts_n;
- ++j;
- } while (--segs_n);
- assert(length == mpw.len);
- if (mpw.pkts_n == MLX5_MPW_DSEG_MAX)
- mlx5_mpw_close(txq, &mpw);
- elts_head = elts_head_next;
-#ifdef MLX5_PMD_SOFT_COUNTERS
- /* Increment sent bytes counter. */
- txq->stats.obytes += length;
-#endif
- ++i;
- } while (pkts_n);
- /* Take a shortcut if nothing must be sent. */
- if (unlikely(i == 0))
- return 0;
- /* Check whether completion threshold has been reached. */
- /* "j" includes both packets and segments. */
- comp = txq->elts_comp + j;
- if (comp >= MLX5_TX_COMP_THRESH) {
- volatile struct mlx5_wqe *wqe = mpw.wqe;
-
- /* Request completion on last WQE. */
- wqe->ctrl[2] = htonl(8);
- /* Save elts_head in unused "immediate" field of WQE. */
- wqe->ctrl[3] = elts_head;
- txq->elts_comp = 0;
- } else {
- txq->elts_comp = comp;
- }
-#ifdef MLX5_PMD_SOFT_COUNTERS
- /* Increment sent packets counter. */
- txq->stats.opackets += i;
-#endif
- /* Ring QP doorbell. */
- if (mpw.state == MLX5_MPW_STATE_OPENED)
- mlx5_mpw_close(txq, &mpw);
- mlx5_tx_dbrec(txq, mpw.wqe);
- txq->elts_head = elts_head;
- return i;
+ return ret;
}
/**
- * Open a MPW inline session.
+ * Handle error CQE.
*
* @param txq
* Pointer to TX queue structure.
- * @param mpw
- * Pointer to MPW session structure.
- * @param length
- * Packet length.
+ * @param error_cqe
+ * Pointer to the error CQE.
+ *
+ * @return
+ * The last Tx buffer element to free.
*/
-static inline void
-mlx5_mpw_inline_new(struct txq *txq, struct mlx5_mpw *mpw, uint32_t length)
+uint16_t
+mlx5_tx_error_cqe_handle(struct mlx5_txq_data *txq,
+ volatile struct mlx5_err_cqe *err_cqe)
{
- uint16_t idx = txq->wqe_ci & ((1 << txq->wqe_n) - 1);
- struct mlx5_wqe_inl_small *inl;
-
- mpw->state = MLX5_MPW_INL_STATE_OPENED;
- mpw->pkts_n = 0;
- mpw->len = length;
- mpw->total_len = 0;
- mpw->wqe = (volatile struct mlx5_wqe *)tx_mlx5_wqe(txq, idx);
- mpw->wqe->ctrl[0] = htonl((MLX5_OPC_MOD_MPW << 24) |
- (txq->wqe_ci << 8) |
- MLX5_OPCODE_TSO);
- mpw->wqe->ctrl[2] = 0;
- mpw->wqe->ctrl[3] = 0;
- mpw->wqe->eseg.mss = htons(length);
- mpw->wqe->eseg.inline_hdr_sz = 0;
- mpw->wqe->eseg.cs_flags = 0;
- mpw->wqe->eseg.rsvd0 = 0;
- mpw->wqe->eseg.rsvd1 = 0;
- mpw->wqe->eseg.rsvd2 = 0;
- inl = (struct mlx5_wqe_inl_small *)
- (((uintptr_t)mpw->wqe) + 2 * MLX5_WQE_DWORD_SIZE);
- mpw->data.raw = (uint8_t *)&inl->raw;
+ if (err_cqe->syndrome != MLX5_CQE_SYNDROME_WR_FLUSH_ERR) {
+ const uint16_t wqe_m = ((1 << txq->wqe_n) - 1);
+ struct mlx5_txq_ctrl *txq_ctrl =
+ container_of(txq, struct mlx5_txq_ctrl, txq);
+ uint16_t new_wqe_pi = rte_be_to_cpu_16(err_cqe->wqe_counter);
+ int seen = check_err_cqe_seen(err_cqe);
+
+ if (!seen && txq_ctrl->dump_file_n <
+ txq_ctrl->priv->config.max_dump_files_num) {
+ MKSTR(err_str, "Unexpected CQE error syndrome "
+ "0x%02x CQN = %u SQN = %u wqe_counter = %u "
+ "wq_ci = %u cq_ci = %u", err_cqe->syndrome,
+ txq_ctrl->cqn, txq->qp_num_8s >> 8,
+ rte_be_to_cpu_16(err_cqe->wqe_counter),
+ txq->wqe_ci, txq->cq_ci);
+ MKSTR(name, "dpdk_mlx5_port_%u_txq_%u_index_%u_%u",
+ PORT_ID(txq_ctrl->priv), txq->idx,
+ txq_ctrl->dump_file_n, (uint32_t)rte_rdtsc());
+ mlx5_dump_debug_information(name, NULL, err_str, 0);
+ mlx5_dump_debug_information(name, "MLX5 Error CQ:",
+ (const void *)((uintptr_t)
+ &(*txq->cqes)[0]),
+ sizeof(*err_cqe) *
+ (1 << txq->cqe_n));
+ mlx5_dump_debug_information(name, "MLX5 Error SQ:",
+ (const void *)((uintptr_t)
+ txq->wqes),
+ MLX5_WQE_SIZE *
+ (1 << txq->wqe_n));
+ txq_ctrl->dump_file_n++;
+ }
+ if (!seen)
+ /*
+ * Count errors in WQEs units.
+ * Later it can be improved to count error packets,
+ * for example, by SQ parsing to find how much packets
+ * should be counted for each WQE.
+ */
+ txq->stats.oerrors += ((txq->wqe_ci & wqe_m) -
+ new_wqe_pi) & wqe_m;
+ if (tx_recover_qp(txq_ctrl) == 0) {
+ txq->cq_ci++;
+ /* Release all the remaining buffers. */
+ return txq->elts_head;
+ }
+ /* Recovering failed - try again later on the same WQE. */
+ } else {
+ txq->cq_ci++;
+ }
+ /* Do not release buffers. */
+ return txq->elts_tail;
}
/**
- * Close a MPW inline session.
+ * Translate RX completion flags to packet type.
*
- * @param txq
- * Pointer to TX queue structure.
- * @param mpw
- * Pointer to MPW session structure.
+ * @param[in] rxq
+ * Pointer to RX queue structure.
+ * @param[in] cqe
+ * Pointer to CQE.
+ *
+ * @note: fix mlx5_dev_supported_ptypes_get() if any change here.
+ *
+ * @return
+ * Packet type for struct rte_mbuf.
*/
-static inline void
-mlx5_mpw_inline_close(struct txq *txq, struct mlx5_mpw *mpw)
+static inline uint32_t
+rxq_cq_to_pkt_type(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cqe)
{
- unsigned int size;
- struct mlx5_wqe_inl_small *inl = (struct mlx5_wqe_inl_small *)
- (((uintptr_t)mpw->wqe) + (2 * MLX5_WQE_DWORD_SIZE));
+ uint8_t idx;
+ uint8_t pinfo = cqe->pkt_info;
+ uint16_t ptype = cqe->hdr_type_etc;
- size = MLX5_WQE_SIZE - MLX5_MWQE64_INL_DATA + mpw->total_len;
/*
- * Store size in multiple of 16 bytes. Control and Ethernet segments
- * count as 2.
+ * The index to the array should have:
+ * bit[1:0] = l3_hdr_type
+ * bit[4:2] = l4_hdr_type
+ * bit[5] = ip_frag
+ * bit[6] = tunneled
+ * bit[7] = outer_l3_type
*/
- mpw->wqe->ctrl[1] = htonl(txq->qp_num_8s | MLX5_WQE_DS(size));
- mpw->state = MLX5_MPW_STATE_CLOSED;
- inl->byte_cnt = htonl(mpw->total_len | MLX5_INLINE_SEG);
- txq->wqe_ci += (size + (MLX5_WQE_SIZE - 1)) / MLX5_WQE_SIZE;
+ idx = ((pinfo & 0x3) << 6) | ((ptype & 0xfc00) >> 10);
+ return mlx5_ptype_table[idx] | rxq->tunnel * !!(idx & (1 << 6));
}
/**
- * DPDK callback for TX with MPW inline support.
- *
- * @param dpdk_txq
- * Generic pointer to TX queue structure.
- * @param[in] pkts
- * Packets to transmit.
- * @param pkts_n
- * Number of packets in array.
+ * Initialize Rx WQ and indexes.
*
- * @return
- * Number of packets successfully transmitted (<= pkts_n).
+ * @param[in] rxq
+ * Pointer to RX queue structure.
*/
-uint16_t
-mlx5_tx_burst_mpw_inline(void *dpdk_txq, struct rte_mbuf **pkts,
- uint16_t pkts_n)
+void
+mlx5_rxq_initialize(struct mlx5_rxq_data *rxq)
{
- struct txq *txq = (struct txq *)dpdk_txq;
- uint16_t elts_head = txq->elts_head;
- const unsigned int elts_n = 1 << txq->elts_n;
- unsigned int i = 0;
- unsigned int j = 0;
- unsigned int max;
- uint16_t max_wqe;
- unsigned int comp;
- unsigned int inline_room = txq->max_inline * RTE_CACHE_LINE_SIZE;
- struct mlx5_mpw mpw = {
- .state = MLX5_MPW_STATE_CLOSED,
- };
- /*
- * Compute the maximum number of WQE which can be consumed by inline
- * code.
- * - 2 DSEG for:
- * - 1 control segment,
- * - 1 Ethernet segment,
- * - N Dseg from the inline request.
- */
- const unsigned int wqe_inl_n =
- ((2 * MLX5_WQE_DWORD_SIZE +
- txq->max_inline * RTE_CACHE_LINE_SIZE) +
- RTE_CACHE_LINE_SIZE - 1) / RTE_CACHE_LINE_SIZE;
+ const unsigned int wqe_n = 1 << rxq->elts_n;
+ unsigned int i;
- if (unlikely(!pkts_n))
- return 0;
- /* Prefetch first packet cacheline. */
- rte_prefetch0(tx_mlx5_wqe(txq, txq->wqe_ci));
- rte_prefetch0(tx_mlx5_wqe(txq, txq->wqe_ci + 1));
- /* Start processing. */
- txq_complete(txq);
- max = (elts_n - (elts_head - txq->elts_tail));
- if (max > elts_n)
- max -= elts_n;
- do {
- struct rte_mbuf *buf = *(pkts++);
- unsigned int elts_head_next;
+ for (i = 0; (i != wqe_n); ++i) {
+ volatile struct mlx5_wqe_data_seg *scat;
uintptr_t addr;
- uint32_t length;
- unsigned int segs_n = buf->nb_segs;
- uint32_t cs_flags = 0;
+ uint32_t byte_count;
- /*
- * Make sure there is enough room to store this packet and
- * that one ring entry remains unused.
- */
- assert(segs_n);
- if (max < segs_n + 1)
- break;
- /* Do not bother with large packets MPW cannot handle. */
- if (segs_n > MLX5_MPW_DSEG_MAX)
- break;
- max -= segs_n;
- --pkts_n;
- /*
- * Compute max_wqe in case less WQE were consumed in previous
- * iteration.
- */
- max_wqe = (1u << txq->wqe_n) - (txq->wqe_ci - txq->wqe_pi);
- /* Should we enable HW CKSUM offload */
- if (buf->ol_flags &
- (PKT_TX_IP_CKSUM | PKT_TX_TCP_CKSUM | PKT_TX_UDP_CKSUM))
- cs_flags = MLX5_ETH_WQE_L3_CSUM | MLX5_ETH_WQE_L4_CSUM;
- /* Retrieve packet information. */
- length = PKT_LEN(buf);
- /* Start new session if packet differs. */
- if (mpw.state == MLX5_MPW_STATE_OPENED) {
- if ((mpw.len != length) ||
- (segs_n != 1) ||
- (mpw.wqe->eseg.cs_flags != cs_flags))
- mlx5_mpw_close(txq, &mpw);
- } else if (mpw.state == MLX5_MPW_INL_STATE_OPENED) {
- if ((mpw.len != length) ||
- (segs_n != 1) ||
- (length > inline_room) ||
- (mpw.wqe->eseg.cs_flags != cs_flags)) {
- mlx5_mpw_inline_close(txq, &mpw);
- inline_room =
- txq->max_inline * RTE_CACHE_LINE_SIZE;
- }
- }
- if (mpw.state == MLX5_MPW_STATE_CLOSED) {
- if ((segs_n != 1) ||
- (length > inline_room)) {
- /*
- * Multi-Packet WQE consumes at most two WQE.
- * mlx5_mpw_new() expects to be able to use
- * such resources.
- */
- if (unlikely(max_wqe < 2))
- break;
- max_wqe -= 2;
- mlx5_mpw_new(txq, &mpw, length);
- mpw.wqe->eseg.cs_flags = cs_flags;
- } else {
- if (unlikely(max_wqe < wqe_inl_n))
- break;
- max_wqe -= wqe_inl_n;
- mlx5_mpw_inline_new(txq, &mpw, length);
- mpw.wqe->eseg.cs_flags = cs_flags;
- }
- }
- /* Multi-segment packets must be alone in their MPW. */
- assert((segs_n == 1) || (mpw.pkts_n == 0));
- if (mpw.state == MLX5_MPW_STATE_OPENED) {
- assert(inline_room ==
- txq->max_inline * RTE_CACHE_LINE_SIZE);
-#if defined(MLX5_PMD_SOFT_COUNTERS) || !defined(NDEBUG)
- length = 0;
-#endif
- do {
- volatile struct mlx5_wqe_data_seg *dseg;
-
- elts_head_next =
- (elts_head + 1) & (elts_n - 1);
- assert(buf);
- (*txq->elts)[elts_head] = buf;
- dseg = mpw.data.dseg[mpw.pkts_n];
- addr = rte_pktmbuf_mtod(buf, uintptr_t);
- *dseg = (struct mlx5_wqe_data_seg){
- .byte_count = htonl(DATA_LEN(buf)),
- .lkey = txq_mp2mr(txq, txq_mb2mp(buf)),
- .addr = htonll(addr),
- };
- elts_head = elts_head_next;
-#if defined(MLX5_PMD_SOFT_COUNTERS) || !defined(NDEBUG)
- length += DATA_LEN(buf);
-#endif
- buf = buf->next;
- ++mpw.pkts_n;
- ++j;
- } while (--segs_n);
- assert(length == mpw.len);
- if (mpw.pkts_n == MLX5_MPW_DSEG_MAX)
- mlx5_mpw_close(txq, &mpw);
+ if (mlx5_rxq_mprq_enabled(rxq)) {
+ struct mlx5_mprq_buf *buf = (*rxq->mprq_bufs)[i];
+
+ scat = &((volatile struct mlx5_wqe_mprq *)
+ rxq->wqes)[i].dseg;
+ addr = (uintptr_t)mlx5_mprq_buf_addr(buf);
+ byte_count = (1 << rxq->strd_sz_n) *
+ (1 << rxq->strd_num_n);
} else {
- unsigned int max;
+ struct rte_mbuf *buf = (*rxq->elts)[i];
- assert(mpw.state == MLX5_MPW_INL_STATE_OPENED);
- assert(length <= inline_room);
- assert(length == DATA_LEN(buf));
- elts_head_next = (elts_head + 1) & (elts_n - 1);
+ scat = &((volatile struct mlx5_wqe_data_seg *)
+ rxq->wqes)[i];
addr = rte_pktmbuf_mtod(buf, uintptr_t);
- (*txq->elts)[elts_head] = buf;
- /* Maximum number of bytes before wrapping. */
- max = ((((uintptr_t)(txq->wqes)) +
- (1 << txq->wqe_n) *
- MLX5_WQE_SIZE) -
- (uintptr_t)mpw.data.raw);
- if (length > max) {
- rte_memcpy((void *)(uintptr_t)mpw.data.raw,
- (void *)addr,
- max);
- mpw.data.raw = (volatile void *)txq->wqes;
- rte_memcpy((void *)(uintptr_t)mpw.data.raw,
- (void *)(addr + max),
- length - max);
- mpw.data.raw += length - max;
- } else {
- rte_memcpy((void *)(uintptr_t)mpw.data.raw,
- (void *)addr,
- length);
-
- if (length == max)
- mpw.data.raw =
- (volatile void *)txq->wqes;
- else
- mpw.data.raw += length;
- }
- ++mpw.pkts_n;
- mpw.total_len += length;
- ++j;
- if (mpw.pkts_n == MLX5_MPW_DSEG_MAX) {
- mlx5_mpw_inline_close(txq, &mpw);
- inline_room =
- txq->max_inline * RTE_CACHE_LINE_SIZE;
- } else {
- inline_room -= length;
- }
+ byte_count = DATA_LEN(buf);
}
- elts_head = elts_head_next;
-#ifdef MLX5_PMD_SOFT_COUNTERS
- /* Increment sent bytes counter. */
- txq->stats.obytes += length;
-#endif
- ++i;
- } while (pkts_n);
- /* Take a shortcut if nothing must be sent. */
- if (unlikely(i == 0))
- return 0;
- /* Check whether completion threshold has been reached. */
- /* "j" includes both packets and segments. */
- comp = txq->elts_comp + j;
- if (comp >= MLX5_TX_COMP_THRESH) {
- volatile struct mlx5_wqe *wqe = mpw.wqe;
-
- /* Request completion on last WQE. */
- wqe->ctrl[2] = htonl(8);
- /* Save elts_head in unused "immediate" field of WQE. */
- wqe->ctrl[3] = elts_head;
- txq->elts_comp = 0;
- } else {
- txq->elts_comp = comp;
+ /* scat->addr must be able to store a pointer. */
+ assert(sizeof(scat->addr) >= sizeof(uintptr_t));
+ *scat = (struct mlx5_wqe_data_seg){
+ .addr = rte_cpu_to_be_64(addr),
+ .byte_count = rte_cpu_to_be_32(byte_count),
+ .lkey = mlx5_rx_addr2mr(rxq, addr),
+ };
}
-#ifdef MLX5_PMD_SOFT_COUNTERS
- /* Increment sent packets counter. */
- txq->stats.opackets += i;
-#endif
- /* Ring QP doorbell. */
- if (mpw.state == MLX5_MPW_INL_STATE_OPENED)
- mlx5_mpw_inline_close(txq, &mpw);
- else if (mpw.state == MLX5_MPW_STATE_OPENED)
- mlx5_mpw_close(txq, &mpw);
- mlx5_tx_dbrec(txq, mpw.wqe);
- txq->elts_head = elts_head;
- return i;
+ rxq->consumed_strd = 0;
+ rxq->decompressed = 0;
+ rxq->rq_pi = 0;
+ rxq->zip = (struct rxq_zip){
+ .ai = 0,
+ };
+ /* Update doorbell counter. */
+ rxq->rq_ci = wqe_n >> rxq->sges_n;
+ rte_cio_wmb();
+ *rxq->rq_db = rte_cpu_to_be_32(rxq->rq_ci);
}
/**
- * Open an Enhanced MPW session.
+ * Modify a Verbs queue state.
+ * This must be called from the primary process.
*
- * @param txq
- * Pointer to TX queue structure.
- * @param mpw
- * Pointer to MPW session structure.
- * @param length
- * Packet length.
+ * @param dev
+ * Pointer to Ethernet device.
+ * @param sm
+ * State modify request parameters.
+ *
+ * @return
+ * 0 in case of success else non-zero value and rte_errno is set.
*/
-static inline void
-mlx5_empw_new(struct txq *txq, struct mlx5_mpw *mpw, int padding)
+int
+mlx5_queue_state_modify_primary(struct rte_eth_dev *dev,
+ const struct mlx5_mp_arg_queue_state_modify *sm)
{
- uint16_t idx = txq->wqe_ci & ((1 << txq->wqe_n) - 1);
-
- mpw->state = MLX5_MPW_ENHANCED_STATE_OPENED;
- mpw->pkts_n = 0;
- mpw->total_len = sizeof(struct mlx5_wqe);
- mpw->wqe = (volatile struct mlx5_wqe *)tx_mlx5_wqe(txq, idx);
- mpw->wqe->ctrl[0] = htonl((MLX5_OPC_MOD_ENHANCED_MPSW << 24) |
- (txq->wqe_ci << 8) |
- MLX5_OPCODE_ENHANCED_MPSW);
- mpw->wqe->ctrl[2] = 0;
- mpw->wqe->ctrl[3] = 0;
- memset((void *)(uintptr_t)&mpw->wqe->eseg, 0, MLX5_WQE_DWORD_SIZE);
- if (unlikely(padding)) {
- uintptr_t addr = (uintptr_t)(mpw->wqe + 1);
-
- /* Pad the first 2 DWORDs with zero-length inline header. */
- *(volatile uint32_t *)addr = htonl(MLX5_INLINE_SEG);
- *(volatile uint32_t *)(addr + MLX5_WQE_DWORD_SIZE) =
- htonl(MLX5_INLINE_SEG);
- mpw->total_len += 2 * MLX5_WQE_DWORD_SIZE;
- /* Start from the next WQEBB. */
- mpw->data.raw = (volatile void *)(tx_mlx5_wqe(txq, idx + 1));
+ int ret;
+ struct mlx5_priv *priv = dev->data->dev_private;
+
+ if (sm->is_wq) {
+ struct ibv_wq_attr mod = {
+ .attr_mask = IBV_WQ_ATTR_STATE,
+ .wq_state = sm->state,
+ };
+ struct mlx5_rxq_data *rxq = (*priv->rxqs)[sm->queue_id];
+ struct mlx5_rxq_ctrl *rxq_ctrl =
+ container_of(rxq, struct mlx5_rxq_ctrl, rxq);
+
+ ret = mlx5_glue->modify_wq(rxq_ctrl->ibv->wq, &mod);
+ if (ret) {
+ DRV_LOG(ERR, "Cannot change Rx WQ state to %u - %s\n",
+ sm->state, strerror(errno));
+ rte_errno = errno;
+ return ret;
+ }
} else {
- mpw->data.raw = (volatile void *)(mpw->wqe + 1);
+ struct mlx5_txq_data *txq = (*priv->txqs)[sm->queue_id];
+ struct mlx5_txq_ctrl *txq_ctrl =
+ container_of(txq, struct mlx5_txq_ctrl, txq);
+ struct ibv_qp_attr mod = {
+ .qp_state = IBV_QPS_RESET,
+ .port_num = (uint8_t)priv->ibv_port,
+ };
+ struct ibv_qp *qp = txq_ctrl->ibv->qp;
+
+ ret = mlx5_glue->modify_qp(qp, &mod, IBV_QP_STATE);
+ if (ret) {
+ DRV_LOG(ERR, "Cannot change the Tx QP state to RESET "
+ "%s\n", strerror(errno));
+ rte_errno = errno;
+ return ret;
+ }
+ mod.qp_state = IBV_QPS_INIT;
+ ret = mlx5_glue->modify_qp(qp, &mod,
+ (IBV_QP_STATE | IBV_QP_PORT));
+ if (ret) {
+ DRV_LOG(ERR, "Cannot change Tx QP state to INIT %s\n",
+ strerror(errno));
+ rte_errno = errno;
+ return ret;
+ }
+ mod.qp_state = IBV_QPS_RTR;
+ ret = mlx5_glue->modify_qp(qp, &mod, IBV_QP_STATE);
+ if (ret) {
+ DRV_LOG(ERR, "Cannot change Tx QP state to RTR %s\n",
+ strerror(errno));
+ rte_errno = errno;
+ return ret;
+ }
+ mod.qp_state = IBV_QPS_RTS;
+ ret = mlx5_glue->modify_qp(qp, &mod, IBV_QP_STATE);
+ if (ret) {
+ DRV_LOG(ERR, "Cannot change Tx QP state to RTS %s\n",
+ strerror(errno));
+ rte_errno = errno;
+ return ret;
+ }
}
+ return 0;
}
/**
- * Close an Enhanced MPW session.
+ * Modify a Verbs queue state.
*
- * @param txq
- * Pointer to TX queue structure.
- * @param mpw
- * Pointer to MPW session structure.
+ * @param dev
+ * Pointer to Ethernet device.
+ * @param sm
+ * State modify request parameters.
*
* @return
- * Number of consumed WQEs.
+ * 0 in case of success else non-zero value.
*/
-static inline uint16_t
-mlx5_empw_close(struct txq *txq, struct mlx5_mpw *mpw)
+static int
+mlx5_queue_state_modify(struct rte_eth_dev *dev,
+ struct mlx5_mp_arg_queue_state_modify *sm)
{
- uint16_t ret;
+ int ret = 0;
- /* Store size in multiple of 16 bytes. Control and Ethernet segments
- * count as 2.
- */
- mpw->wqe->ctrl[1] = htonl(txq->qp_num_8s | MLX5_WQE_DS(mpw->total_len));
- mpw->state = MLX5_MPW_STATE_CLOSED;
- ret = (mpw->total_len + (MLX5_WQE_SIZE - 1)) / MLX5_WQE_SIZE;
- txq->wqe_ci += ret;
+ switch (rte_eal_process_type()) {
+ case RTE_PROC_PRIMARY:
+ ret = mlx5_queue_state_modify_primary(dev, sm);
+ break;
+ case RTE_PROC_SECONDARY:
+ ret = mlx5_mp_req_queue_state_modify(dev, sm);
+ break;
+ default:
+ break;
+ }
return ret;
}
/**
- * DPDK callback for TX with Enhanced MPW support.
+ * Handle a Rx error.
+ * The function inserts the RQ state to reset when the first error CQE is
+ * shown, then drains the CQ by the caller function loop. When the CQ is empty,
+ * it moves the RQ state to ready and initializes the RQ.
+ * Next CQE identification and error counting are in the caller responsibility.
*
- * @param dpdk_txq
- * Generic pointer to TX queue structure.
- * @param[in] pkts
- * Packets to transmit.
- * @param pkts_n
- * Number of packets in array.
+ * @param[in] rxq
+ * Pointer to RX queue structure.
+ * @param[in] mbuf_prepare
+ * Whether to prepare mbufs for the RQ.
*
* @return
- * Number of packets successfully transmitted (<= pkts_n).
+ * -1 in case of recovery error, otherwise the CQE status.
*/
-uint16_t
-mlx5_tx_burst_empw(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
+int
+mlx5_rx_err_handle(struct mlx5_rxq_data *rxq, uint8_t mbuf_prepare)
{
- struct txq *txq = (struct txq *)dpdk_txq;
- uint16_t elts_head = txq->elts_head;
- const unsigned int elts_n = 1 << txq->elts_n;
- unsigned int i = 0;
- unsigned int j = 0;
- unsigned int max_elts;
- uint16_t max_wqe;
- unsigned int max_inline = txq->max_inline * RTE_CACHE_LINE_SIZE;
- unsigned int mpw_room = 0;
- unsigned int inl_pad = 0;
- uint32_t inl_hdr;
- struct mlx5_mpw mpw = {
- .state = MLX5_MPW_STATE_CLOSED,
+ const uint16_t cqe_n = 1 << rxq->cqe_n;
+ const uint16_t cqe_mask = cqe_n - 1;
+ const unsigned int wqe_n = 1 << rxq->elts_n;
+ struct mlx5_rxq_ctrl *rxq_ctrl =
+ container_of(rxq, struct mlx5_rxq_ctrl, rxq);
+ union {
+ volatile struct mlx5_cqe *cqe;
+ volatile struct mlx5_err_cqe *err_cqe;
+ } u = {
+ .cqe = &(*rxq->cqes)[rxq->cq_ci & cqe_mask],
};
-
- if (unlikely(!pkts_n))
- return 0;
- /* Start processing. */
- txq_complete(txq);
- max_elts = (elts_n - (elts_head - txq->elts_tail));
- if (max_elts > elts_n)
- max_elts -= elts_n;
- /* A CQE slot must always be available. */
- assert((1u << txq->cqe_n) - (txq->cq_pi - txq->cq_ci));
- max_wqe = (1u << txq->wqe_n) - (txq->wqe_ci - txq->wqe_pi);
- if (unlikely(!max_wqe))
- return 0;
- do {
- struct rte_mbuf *buf = *(pkts++);
- unsigned int elts_head_next;
- uintptr_t addr;
- uint64_t naddr;
- unsigned int n;
- unsigned int do_inline = 0; /* Whether inline is possible. */
- uint32_t length;
- unsigned int segs_n = buf->nb_segs;
- uint32_t cs_flags = 0;
-
- /*
- * Make sure there is enough room to store this packet and
- * that one ring entry remains unused.
- */
- assert(segs_n);
- if (max_elts - j < segs_n + 1)
- break;
- /* Do not bother with large packets MPW cannot handle. */
- if (segs_n > MLX5_MPW_DSEG_MAX)
- break;
- /* Should we enable HW CKSUM offload. */
- if (buf->ol_flags &
- (PKT_TX_IP_CKSUM | PKT_TX_TCP_CKSUM | PKT_TX_UDP_CKSUM))
- cs_flags = MLX5_ETH_WQE_L3_CSUM | MLX5_ETH_WQE_L4_CSUM;
- /* Retrieve packet information. */
- length = PKT_LEN(buf);
- /* Start new session if:
- * - multi-segment packet
- * - no space left even for a dseg
- * - next packet can be inlined with a new WQE
- * - cs_flag differs
- * It can't be MLX5_MPW_STATE_OPENED as always have a single
- * segmented packet.
- */
- if (mpw.state == MLX5_MPW_ENHANCED_STATE_OPENED) {
- if ((segs_n != 1) ||
- (inl_pad + sizeof(struct mlx5_wqe_data_seg) >
- mpw_room) ||
- (length <= txq->inline_max_packet_sz &&
- inl_pad + sizeof(inl_hdr) + length >
- mpw_room) ||
- (mpw.wqe->eseg.cs_flags != cs_flags))
- max_wqe -= mlx5_empw_close(txq, &mpw);
- }
- if (unlikely(mpw.state == MLX5_MPW_STATE_CLOSED)) {
- if (unlikely(segs_n != 1)) {
- /* Fall back to legacy MPW.
- * A MPW session consumes 2 WQEs at most to
- * include MLX5_MPW_DSEG_MAX pointers.
- */
- if (unlikely(max_wqe < 2))
- break;
- mlx5_mpw_new(txq, &mpw, length);
- } else {
- /* In Enhanced MPW, inline as much as the budget
- * is allowed. The remaining space is to be
- * filled with dsegs. If the title WQEBB isn't
- * padded, it will have 2 dsegs there.
- */
- mpw_room = RTE_MIN(MLX5_WQE_SIZE_MAX,
- (max_inline ? max_inline :
- pkts_n * MLX5_WQE_DWORD_SIZE) +
- MLX5_WQE_SIZE);
- if (unlikely(max_wqe * MLX5_WQE_SIZE <
- mpw_room))
- break;
- /* Don't pad the title WQEBB to not waste WQ. */
- mlx5_empw_new(txq, &mpw, 0);
- mpw_room -= mpw.total_len;
- inl_pad = 0;
- do_inline =
- length <= txq->inline_max_packet_sz &&
- sizeof(inl_hdr) + length <= mpw_room &&
- !txq->mpw_hdr_dseg;
- }
- mpw.wqe->eseg.cs_flags = cs_flags;
- } else {
- /* Evaluate whether the next packet can be inlined.
- * Inlininig is possible when:
- * - length is less than configured value
- * - length fits for remaining space
- * - not required to fill the title WQEBB with dsegs
- */
- do_inline =
- length <= txq->inline_max_packet_sz &&
- inl_pad + sizeof(inl_hdr) + length <=
- mpw_room &&
- (!txq->mpw_hdr_dseg ||
- mpw.total_len >= MLX5_WQE_SIZE);
+ struct mlx5_mp_arg_queue_state_modify sm;
+ int ret;
+
+ switch (rxq->err_state) {
+ case MLX5_RXQ_ERR_STATE_NO_ERROR:
+ rxq->err_state = MLX5_RXQ_ERR_STATE_NEED_RESET;
+ /* Fall-through */
+ case MLX5_RXQ_ERR_STATE_NEED_RESET:
+ sm.is_wq = 1;
+ sm.queue_id = rxq->idx;
+ sm.state = IBV_WQS_RESET;
+ if (mlx5_queue_state_modify(ETH_DEV(rxq_ctrl->priv), &sm))
+ return -1;
+ if (rxq_ctrl->dump_file_n <
+ rxq_ctrl->priv->config.max_dump_files_num) {
+ MKSTR(err_str, "Unexpected CQE error syndrome "
+ "0x%02x CQN = %u RQN = %u wqe_counter = %u"
+ " rq_ci = %u cq_ci = %u", u.err_cqe->syndrome,
+ rxq->cqn, rxq_ctrl->wqn,
+ rte_be_to_cpu_16(u.err_cqe->wqe_counter),
+ rxq->rq_ci << rxq->sges_n, rxq->cq_ci);
+ MKSTR(name, "dpdk_mlx5_port_%u_rxq_%u_%u",
+ rxq->port_id, rxq->idx, (uint32_t)rte_rdtsc());
+ mlx5_dump_debug_information(name, NULL, err_str, 0);
+ mlx5_dump_debug_information(name, "MLX5 Error CQ:",
+ (const void *)((uintptr_t)
+ rxq->cqes),
+ sizeof(*u.cqe) * cqe_n);
+ mlx5_dump_debug_information(name, "MLX5 Error RQ:",
+ (const void *)((uintptr_t)
+ rxq->wqes),
+ 16 * wqe_n);
+ rxq_ctrl->dump_file_n++;
}
- /* Multi-segment packets must be alone in their MPW. */
- assert((segs_n == 1) || (mpw.pkts_n == 0));
- if (unlikely(mpw.state == MLX5_MPW_STATE_OPENED)) {
-#if defined(MLX5_PMD_SOFT_COUNTERS) || !defined(NDEBUG)
- length = 0;
-#endif
- do {
- volatile struct mlx5_wqe_data_seg *dseg;
-
- elts_head_next =
- (elts_head + 1) & (elts_n - 1);
- assert(buf);
- (*txq->elts)[elts_head] = buf;
- dseg = mpw.data.dseg[mpw.pkts_n];
- addr = rte_pktmbuf_mtod(buf, uintptr_t);
- *dseg = (struct mlx5_wqe_data_seg){
- .byte_count = htonl(DATA_LEN(buf)),
- .lkey = txq_mp2mr(txq, txq_mb2mp(buf)),
- .addr = htonll(addr),
- };
- elts_head = elts_head_next;
-#if defined(MLX5_PMD_SOFT_COUNTERS) || !defined(NDEBUG)
- length += DATA_LEN(buf);
-#endif
- buf = buf->next;
- ++j;
- ++mpw.pkts_n;
- } while (--segs_n);
- /* A multi-segmented packet takes one MPW session.
- * TODO: Pack more multi-segmented packets if possible.
- */
- mlx5_mpw_close(txq, &mpw);
- if (mpw.pkts_n < 3)
- max_wqe--;
- else
- max_wqe -= 2;
- } else if (do_inline) {
- /* Inline packet into WQE. */
- unsigned int max;
-
- assert(mpw.state == MLX5_MPW_ENHANCED_STATE_OPENED);
- assert(length == DATA_LEN(buf));
- inl_hdr = htonl(length | MLX5_INLINE_SEG);
- addr = rte_pktmbuf_mtod(buf, uintptr_t);
- mpw.data.raw = (volatile void *)
- ((uintptr_t)mpw.data.raw + inl_pad);
- max = tx_mlx5_wq_tailroom(txq,
- (void *)(uintptr_t)mpw.data.raw);
- /* Copy inline header. */
- mpw.data.raw = (volatile void *)
- mlx5_copy_to_wq(
- (void *)(uintptr_t)mpw.data.raw,
- &inl_hdr,
- sizeof(inl_hdr),
- (void *)(uintptr_t)txq->wqes,
- max);
- max = tx_mlx5_wq_tailroom(txq,
- (void *)(uintptr_t)mpw.data.raw);
- /* Copy packet data. */
- mpw.data.raw = (volatile void *)
- mlx5_copy_to_wq(
- (void *)(uintptr_t)mpw.data.raw,
- (void *)addr,
- length,
- (void *)(uintptr_t)txq->wqes,
- max);
- ++mpw.pkts_n;
- mpw.total_len += (inl_pad + sizeof(inl_hdr) + length);
- /* No need to get completion as the entire packet is
- * copied to WQ. Free the buf right away.
+ rxq->err_state = MLX5_RXQ_ERR_STATE_NEED_READY;
+ /* Fall-through */
+ case MLX5_RXQ_ERR_STATE_NEED_READY:
+ ret = check_cqe(u.cqe, cqe_n, rxq->cq_ci);
+ if (ret == MLX5_CQE_STATUS_HW_OWN) {
+ rte_cio_wmb();
+ *rxq->cq_db = rte_cpu_to_be_32(rxq->cq_ci);
+ rte_cio_wmb();
+ /*
+ * The RQ consumer index must be zeroed while moving
+ * from RESET state to RDY state.
*/
- elts_head_next = elts_head;
- rte_pktmbuf_free_seg(buf);
- mpw_room -= (inl_pad + sizeof(inl_hdr) + length);
- /* Add pad in the next packet if any. */
- inl_pad = (((uintptr_t)mpw.data.raw +
- (MLX5_WQE_DWORD_SIZE - 1)) &
- ~(MLX5_WQE_DWORD_SIZE - 1)) -
- (uintptr_t)mpw.data.raw;
- } else {
- /* No inline. Load a dseg of packet pointer. */
- volatile rte_v128u32_t *dseg;
-
- assert(mpw.state == MLX5_MPW_ENHANCED_STATE_OPENED);
- assert((inl_pad + sizeof(*dseg)) <= mpw_room);
- assert(length == DATA_LEN(buf));
- if (!tx_mlx5_wq_tailroom(txq,
- (void *)((uintptr_t)mpw.data.raw
- + inl_pad)))
- dseg = (volatile void *)txq->wqes;
- else
- dseg = (volatile void *)
- ((uintptr_t)mpw.data.raw +
- inl_pad);
- elts_head_next = (elts_head + 1) & (elts_n - 1);
- (*txq->elts)[elts_head] = buf;
- addr = rte_pktmbuf_mtod(buf, uintptr_t);
- for (n = 0; n * RTE_CACHE_LINE_SIZE < length; n++)
- rte_prefetch2((void *)(addr +
- n * RTE_CACHE_LINE_SIZE));
- naddr = htonll(addr);
- *dseg = (rte_v128u32_t) {
- htonl(length),
- txq_mp2mr(txq, txq_mb2mp(buf)),
- naddr,
- naddr >> 32,
- };
- mpw.data.raw = (volatile void *)(dseg + 1);
- mpw.total_len += (inl_pad + sizeof(*dseg));
- ++j;
- ++mpw.pkts_n;
- mpw_room -= (inl_pad + sizeof(*dseg));
- inl_pad = 0;
+ *rxq->rq_db = rte_cpu_to_be_32(0);
+ rte_cio_wmb();
+ sm.is_wq = 1;
+ sm.queue_id = rxq->idx;
+ sm.state = IBV_WQS_RDY;
+ if (mlx5_queue_state_modify(ETH_DEV(rxq_ctrl->priv),
+ &sm))
+ return -1;
+ if (mbuf_prepare) {
+ const uint16_t q_mask = wqe_n - 1;
+ uint16_t elt_idx;
+ struct rte_mbuf **elt;
+ int i;
+ unsigned int n = wqe_n - (rxq->rq_ci -
+ rxq->rq_pi);
+
+ for (i = 0; i < (int)n; ++i) {
+ elt_idx = (rxq->rq_ci + i) & q_mask;
+ elt = &(*rxq->elts)[elt_idx];
+ *elt = rte_mbuf_raw_alloc(rxq->mp);
+ if (!*elt) {
+ for (i--; i >= 0; --i) {
+ elt_idx = (rxq->rq_ci +
+ i) & q_mask;
+ elt = &(*rxq->elts)
+ [elt_idx];
+ rte_pktmbuf_free_seg
+ (*elt);
+ }
+ return -1;
+ }
+ }
+ }
+ mlx5_rxq_initialize(rxq);
+ rxq->err_state = MLX5_RXQ_ERR_STATE_NO_ERROR;
}
- elts_head = elts_head_next;
-#ifdef MLX5_PMD_SOFT_COUNTERS
- /* Increment sent bytes counter. */
- txq->stats.obytes += length;
-#endif
- ++i;
- } while (i < pkts_n);
- /* Take a shortcut if nothing must be sent. */
- if (unlikely(i == 0))
- return 0;
- /* Check whether completion threshold has been reached. */
- if (txq->elts_comp + j >= MLX5_TX_COMP_THRESH ||
- (uint16_t)(txq->wqe_ci - txq->mpw_comp) >=
- (1 << txq->wqe_n) / MLX5_TX_COMP_THRESH_INLINE_DIV) {
- volatile struct mlx5_wqe *wqe = mpw.wqe;
-
- /* Request completion on last WQE. */
- wqe->ctrl[2] = htonl(8);
- /* Save elts_head in unused "immediate" field of WQE. */
- wqe->ctrl[3] = elts_head;
- txq->elts_comp = 0;
- txq->mpw_comp = txq->wqe_ci;
- txq->cq_pi++;
- } else {
- txq->elts_comp += j;
- }
-#ifdef MLX5_PMD_SOFT_COUNTERS
- /* Increment sent packets counter. */
- txq->stats.opackets += i;
-#endif
- if (mpw.state == MLX5_MPW_ENHANCED_STATE_OPENED)
- mlx5_empw_close(txq, &mpw);
- else if (mpw.state == MLX5_MPW_STATE_OPENED)
- mlx5_mpw_close(txq, &mpw);
- /* Ring QP doorbell. */
- mlx5_tx_dbrec(txq, mpw.wqe);
- txq->elts_head = elts_head;
- return i;
-}
-
-/**
- * Translate RX completion flags to packet type.
- *
- * @param[in] cqe
- * Pointer to CQE.
- *
- * @note: fix mlx5_dev_supported_ptypes_get() if any change here.
- *
- * @return
- * Packet type for struct rte_mbuf.
- */
-static inline uint32_t
-rxq_cq_to_pkt_type(volatile struct mlx5_cqe *cqe)
-{
- uint32_t pkt_type;
- uint16_t flags = ntohs(cqe->hdr_type_etc);
-
- if (cqe->pkt_info & MLX5_CQE_RX_TUNNEL_PACKET) {
- pkt_type =
- TRANSPOSE(flags,
- MLX5_CQE_RX_IPV4_PACKET,
- RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN) |
- TRANSPOSE(flags,
- MLX5_CQE_RX_IPV6_PACKET,
- RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN);
- pkt_type |= ((cqe->pkt_info & MLX5_CQE_RX_OUTER_PACKET) ?
- RTE_PTYPE_L3_IPV6_EXT_UNKNOWN :
- RTE_PTYPE_L3_IPV4_EXT_UNKNOWN);
- } else {
- pkt_type =
- TRANSPOSE(flags,
- MLX5_CQE_L3_HDR_TYPE_IPV6,
- RTE_PTYPE_L3_IPV6_EXT_UNKNOWN) |
- TRANSPOSE(flags,
- MLX5_CQE_L3_HDR_TYPE_IPV4,
- RTE_PTYPE_L3_IPV4_EXT_UNKNOWN);
+ return ret;
+ default:
+ return -1;
}
- return pkt_type;
}
/**
* Pointer to RX queue.
* @param cqe
* CQE to process.
- * @param[out] rss_hash
- * Packet RSS Hash result.
+ * @param[out] mcqe
+ * Store pointer to mini-CQE if compressed. Otherwise, the pointer is not
+ * written.
*
* @return
- * Packet size in bytes (0 if there is none), -1 in case of completion
- * with error.
+ * 0 in case of empty CQE, otherwise the packet size in bytes.
*/
static inline int
-mlx5_rx_poll_len(struct rxq *rxq, volatile struct mlx5_cqe *cqe,
- uint16_t cqe_cnt, uint32_t *rss_hash)
+mlx5_rx_poll_len(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cqe,
+ uint16_t cqe_cnt, volatile struct mlx5_mini_cqe8 **mcqe)
{
struct rxq_zip *zip = &rxq->zip;
uint16_t cqe_n = cqe_cnt + 1;
- int len = 0;
+ int len;
uint16_t idx, end;
- /* Process compressed data in the CQE and mini arrays. */
- if (zip->ai) {
- volatile struct mlx5_mini_cqe8 (*mc)[8] =
- (volatile struct mlx5_mini_cqe8 (*)[8])
- (uintptr_t)(&(*rxq->cqes)[zip->ca & cqe_cnt]);
-
- len = ntohl((*mc)[zip->ai & 7].byte_cnt);
- *rss_hash = ntohl((*mc)[zip->ai & 7].rx_hash_result);
- if ((++zip->ai & 7) == 0) {
- /* Invalidate consumed CQEs */
- idx = zip->ca;
- end = zip->na;
- while (idx != end) {
- (*rxq->cqes)[idx & cqe_cnt].op_own =
- MLX5_CQE_INVALIDATE;
- ++idx;
- }
- /*
- * Increment consumer index to skip the number of
- * CQEs consumed. Hardware leaves holes in the CQ
- * ring for software use.
- */
- zip->ca = zip->na;
- zip->na += 8;
- }
- if (unlikely(rxq->zip.ai == rxq->zip.cqe_cnt)) {
- /* Invalidate the rest */
- idx = zip->ca;
- end = zip->cq_ci;
-
- while (idx != end) {
- (*rxq->cqes)[idx & cqe_cnt].op_own =
- MLX5_CQE_INVALIDATE;
- ++idx;
- }
- rxq->cq_ci = zip->cq_ci;
- zip->ai = 0;
- }
- /* No compressed data, get next CQE and verify if it is compressed. */
- } else {
- int ret;
- int8_t op_own;
-
- ret = check_cqe(cqe, cqe_n, rxq->cq_ci);
- if (unlikely(ret == 1))
- return 0;
- ++rxq->cq_ci;
- op_own = cqe->op_own;
- if (MLX5_CQE_FORMAT(op_own) == MLX5_COMPRESSED) {
+ do {
+ len = 0;
+ /* Process compressed data in the CQE and mini arrays. */
+ if (zip->ai) {
volatile struct mlx5_mini_cqe8 (*mc)[8] =
(volatile struct mlx5_mini_cqe8 (*)[8])
- (uintptr_t)(&(*rxq->cqes)[rxq->cq_ci &
- cqe_cnt]);
-
- /* Fix endianness. */
- zip->cqe_cnt = ntohl(cqe->byte_cnt);
- /*
- * Current mini array position is the one returned by
- * check_cqe64().
- *
- * If completion comprises several mini arrays, as a
- * special case the second one is located 7 CQEs after
- * the initial CQE instead of 8 for subsequent ones.
- */
- zip->ca = rxq->cq_ci;
- zip->na = zip->ca + 7;
- /* Compute the next non compressed CQE. */
- --rxq->cq_ci;
- zip->cq_ci = rxq->cq_ci + zip->cqe_cnt;
- /* Get packet size to return. */
- len = ntohl((*mc)[0].byte_cnt);
- *rss_hash = ntohl((*mc)[0].rx_hash_result);
- zip->ai = 1;
- /* Prefetch all the entries to be invalidated */
- idx = zip->ca;
- end = zip->cq_ci;
- while (idx != end) {
- rte_prefetch0(&(*rxq->cqes)[(idx) & cqe_cnt]);
- ++idx;
+ (uintptr_t)(&(*rxq->cqes)[zip->ca &
+ cqe_cnt].pkt_info);
+
+ len = rte_be_to_cpu_32((*mc)[zip->ai & 7].byte_cnt);
+ *mcqe = &(*mc)[zip->ai & 7];
+ if ((++zip->ai & 7) == 0) {
+ /* Invalidate consumed CQEs */
+ idx = zip->ca;
+ end = zip->na;
+ while (idx != end) {
+ (*rxq->cqes)[idx & cqe_cnt].op_own =
+ MLX5_CQE_INVALIDATE;
+ ++idx;
+ }
+ /*
+ * Increment consumer index to skip the number
+ * of CQEs consumed. Hardware leaves holes in
+ * the CQ ring for software use.
+ */
+ zip->ca = zip->na;
+ zip->na += 8;
+ }
+ if (unlikely(rxq->zip.ai == rxq->zip.cqe_cnt)) {
+ /* Invalidate the rest */
+ idx = zip->ca;
+ end = zip->cq_ci;
+
+ while (idx != end) {
+ (*rxq->cqes)[idx & cqe_cnt].op_own =
+ MLX5_CQE_INVALIDATE;
+ ++idx;
+ }
+ rxq->cq_ci = zip->cq_ci;
+ zip->ai = 0;
}
+ /*
+ * No compressed data, get next CQE and verify if it is
+ * compressed.
+ */
+ } else {
+ int ret;
+ int8_t op_own;
+
+ ret = check_cqe(cqe, cqe_n, rxq->cq_ci);
+ if (unlikely(ret != MLX5_CQE_STATUS_SW_OWN)) {
+ if (unlikely(ret == MLX5_CQE_STATUS_ERR ||
+ rxq->err_state)) {
+ ret = mlx5_rx_err_handle(rxq, 0);
+ if (ret == MLX5_CQE_STATUS_HW_OWN ||
+ ret == -1)
+ return 0;
+ } else {
+ return 0;
+ }
+ }
+ ++rxq->cq_ci;
+ op_own = cqe->op_own;
+ if (MLX5_CQE_FORMAT(op_own) == MLX5_COMPRESSED) {
+ volatile struct mlx5_mini_cqe8 (*mc)[8] =
+ (volatile struct mlx5_mini_cqe8 (*)[8])
+ (uintptr_t)(&(*rxq->cqes)
+ [rxq->cq_ci &
+ cqe_cnt].pkt_info);
+
+ /* Fix endianness. */
+ zip->cqe_cnt = rte_be_to_cpu_32(cqe->byte_cnt);
+ /*
+ * Current mini array position is the one
+ * returned by check_cqe64().
+ *
+ * If completion comprises several mini arrays,
+ * as a special case the second one is located
+ * 7 CQEs after the initial CQE instead of 8
+ * for subsequent ones.
+ */
+ zip->ca = rxq->cq_ci;
+ zip->na = zip->ca + 7;
+ /* Compute the next non compressed CQE. */
+ --rxq->cq_ci;
+ zip->cq_ci = rxq->cq_ci + zip->cqe_cnt;
+ /* Get packet size to return. */
+ len = rte_be_to_cpu_32((*mc)[0].byte_cnt);
+ *mcqe = &(*mc)[0];
+ zip->ai = 1;
+ /* Prefetch all to be invalidated */
+ idx = zip->ca;
+ end = zip->cq_ci;
+ while (idx != end) {
+ rte_prefetch0(&(*rxq->cqes)[(idx) &
+ cqe_cnt]);
+ ++idx;
+ }
+ } else {
+ len = rte_be_to_cpu_32(cqe->byte_cnt);
+ }
+ }
+ if (unlikely(rxq->err_state)) {
+ cqe = &(*rxq->cqes)[rxq->cq_ci & cqe_cnt];
+ ++rxq->stats.idropped;
} else {
- len = ntohl(cqe->byte_cnt);
- *rss_hash = ntohl(cqe->rx_hash_res);
+ return len;
}
- /* Error while receiving packet. */
- if (unlikely(MLX5_CQE_OPCODE(op_own) == MLX5_CQE_RESP_ERR))
- return -1;
- }
- return len;
+ } while (1);
}
/**
* Translate RX completion flags to offload flags.
*
- * @param[in] rxq
- * Pointer to RX queue structure.
* @param[in] cqe
* Pointer to CQE.
*
* Offload flags (ol_flags) for struct rte_mbuf.
*/
static inline uint32_t
-rxq_cq_to_ol_flags(struct rxq *rxq, volatile struct mlx5_cqe *cqe)
+rxq_cq_to_ol_flags(volatile struct mlx5_cqe *cqe)
{
uint32_t ol_flags = 0;
- uint16_t flags = ntohs(cqe->hdr_type_etc);
+ uint16_t flags = rte_be_to_cpu_16(cqe->hdr_type_etc);
ol_flags =
TRANSPOSE(flags,
TRANSPOSE(flags,
MLX5_CQE_RX_L4_HDR_VALID,
PKT_RX_L4_CKSUM_GOOD);
- if ((cqe->pkt_info & MLX5_CQE_RX_TUNNEL_PACKET) && (rxq->csum_l2tun))
- ol_flags |=
- TRANSPOSE(flags,
- MLX5_CQE_RX_L3_HDR_VALID,
- PKT_RX_IP_CKSUM_GOOD) |
- TRANSPOSE(flags,
- MLX5_CQE_RX_L4_HDR_VALID,
- PKT_RX_L4_CKSUM_GOOD);
return ol_flags;
}
+/**
+ * Fill in mbuf fields from RX completion flags.
+ * Note that pkt->ol_flags should be initialized outside of this function.
+ *
+ * @param rxq
+ * Pointer to RX queue.
+ * @param pkt
+ * mbuf to fill.
+ * @param cqe
+ * CQE to process.
+ * @param rss_hash_res
+ * Packet RSS Hash result.
+ */
+static inline void
+rxq_cq_to_mbuf(struct mlx5_rxq_data *rxq, struct rte_mbuf *pkt,
+ volatile struct mlx5_cqe *cqe, uint32_t rss_hash_res)
+{
+ /* Update packet information. */
+ pkt->packet_type = rxq_cq_to_pkt_type(rxq, cqe);
+ if (rss_hash_res && rxq->rss_hash) {
+ pkt->hash.rss = rss_hash_res;
+ pkt->ol_flags |= PKT_RX_RSS_HASH;
+ }
+ if (rxq->mark && MLX5_FLOW_MARK_IS_VALID(cqe->sop_drop_qpn)) {
+ pkt->ol_flags |= PKT_RX_FDIR;
+ if (cqe->sop_drop_qpn !=
+ rte_cpu_to_be_32(MLX5_FLOW_MARK_DEFAULT)) {
+ uint32_t mark = cqe->sop_drop_qpn;
+
+ pkt->ol_flags |= PKT_RX_FDIR_ID;
+ pkt->hash.fdir.hi = mlx5_flow_mark_get(mark);
+ }
+ }
+ if (rxq->csum)
+ pkt->ol_flags |= rxq_cq_to_ol_flags(cqe);
+ if (rxq->vlan_strip &&
+ (cqe->hdr_type_etc & rte_cpu_to_be_16(MLX5_CQE_VLAN_STRIPPED))) {
+ pkt->ol_flags |= PKT_RX_VLAN | PKT_RX_VLAN_STRIPPED;
+ pkt->vlan_tci = rte_be_to_cpu_16(cqe->vlan_info);
+ }
+ if (rxq->hw_timestamp) {
+ pkt->timestamp = rte_be_to_cpu_64(cqe->timestamp);
+ pkt->ol_flags |= PKT_RX_TIMESTAMP;
+ }
+}
+
/**
* DPDK callback for RX.
*
uint16_t
mlx5_rx_burst(void *dpdk_rxq, struct rte_mbuf **pkts, uint16_t pkts_n)
{
- struct rxq *rxq = dpdk_rxq;
+ struct mlx5_rxq_data *rxq = dpdk_rxq;
const unsigned int wqe_cnt = (1 << rxq->elts_n) - 1;
const unsigned int cqe_cnt = (1 << rxq->cqe_n) - 1;
const unsigned int sges_n = rxq->sges_n;
&(*rxq->cqes)[rxq->cq_ci & cqe_cnt];
unsigned int i = 0;
unsigned int rq_ci = rxq->rq_ci << sges_n;
- int len; /* keep its value across iterations. */
+ int len = 0; /* keep its value across iterations. */
while (pkts_n) {
unsigned int idx = rq_ci & wqe_cnt;
- volatile struct mlx5_wqe_data_seg *wqe = &(*rxq->wqes)[idx];
+ volatile struct mlx5_wqe_data_seg *wqe =
+ &((volatile struct mlx5_wqe_data_seg *)rxq->wqes)[idx];
struct rte_mbuf *rep = (*rxq->elts)[idx];
- uint32_t rss_hash_res = 0;
+ volatile struct mlx5_mini_cqe8 *mcqe = NULL;
+ uint32_t rss_hash_res;
if (pkt)
NEXT(seg) = rep;
while (pkt != seg) {
assert(pkt != (*rxq->elts)[idx]);
rep = NEXT(pkt);
- rte_mbuf_refcnt_set(pkt, 0);
- __rte_mbuf_raw_free(pkt);
+ NEXT(pkt) = NULL;
+ NB_SEGS(pkt) = 1;
+ rte_mbuf_raw_free(pkt);
pkt = rep;
}
break;
}
if (!pkt) {
cqe = &(*rxq->cqes)[rxq->cq_ci & cqe_cnt];
- len = mlx5_rx_poll_len(rxq, cqe, cqe_cnt,
- &rss_hash_res);
+ len = mlx5_rx_poll_len(rxq, cqe, cqe_cnt, &mcqe);
if (!len) {
- rte_mbuf_refcnt_set(rep, 0);
- __rte_mbuf_raw_free(rep);
+ rte_mbuf_raw_free(rep);
break;
}
- if (unlikely(len == -1)) {
- /* RX error, packet is likely too large. */
- rte_mbuf_refcnt_set(rep, 0);
- __rte_mbuf_raw_free(rep);
- ++rxq->stats.idropped;
- goto skip;
- }
pkt = seg;
assert(len >= (rxq->crc_present << 2));
- /* Update packet information. */
- pkt->packet_type = 0;
pkt->ol_flags = 0;
- if (rss_hash_res && rxq->rss_hash) {
- pkt->hash.rss = rss_hash_res;
- pkt->ol_flags = PKT_RX_RSS_HASH;
- }
- if (rxq->mark && (cqe->sop_drop_qpn !=
- htonl(MLX5_FLOW_MARK_INVALID))) {
- pkt->ol_flags |= PKT_RX_FDIR;
- if (cqe->sop_drop_qpn !=
- htonl(MLX5_FLOW_MARK_DEFAULT)) {
- uint32_t mark = cqe->sop_drop_qpn;
-
- pkt->ol_flags |= PKT_RX_FDIR_ID;
- pkt->hash.fdir.hi =
- mlx5_flow_mark_get(mark);
- }
- }
- if (rxq->csum | rxq->csum_l2tun | rxq->vlan_strip |
- rxq->crc_present) {
- if (rxq->csum) {
- pkt->packet_type =
- rxq_cq_to_pkt_type(cqe);
- pkt->ol_flags |=
- rxq_cq_to_ol_flags(rxq, cqe);
- }
- if (ntohs(cqe->hdr_type_etc) &
- MLX5_CQE_VLAN_STRIPPED) {
- pkt->ol_flags |= PKT_RX_VLAN_PKT |
- PKT_RX_VLAN_STRIPPED;
- pkt->vlan_tci = ntohs(cqe->vlan_info);
- }
- if (rxq->crc_present)
- len -= ETHER_CRC_LEN;
- }
+ /* If compressed, take hash result from mini-CQE. */
+ rss_hash_res = rte_be_to_cpu_32(mcqe == NULL ?
+ cqe->rx_hash_res :
+ mcqe->rx_hash_result);
+ rxq_cq_to_mbuf(rxq, pkt, cqe, rss_hash_res);
+ if (rxq->crc_present)
+ len -= RTE_ETHER_CRC_LEN;
PKT_LEN(pkt) = len;
}
DATA_LEN(rep) = DATA_LEN(seg);
PKT_LEN(rep) = PKT_LEN(seg);
SET_DATA_OFF(rep, DATA_OFF(seg));
- NB_SEGS(rep) = NB_SEGS(seg);
PORT(rep) = PORT(seg);
- NEXT(rep) = NULL;
(*rxq->elts)[idx] = rep;
/*
* Fill NIC descriptor with the new buffer. The lkey and size
* of the buffers are already known, only the buffer address
* changes.
*/
- wqe->addr = htonll(rte_pktmbuf_mtod(rep, uintptr_t));
+ wqe->addr = rte_cpu_to_be_64(rte_pktmbuf_mtod(rep, uintptr_t));
+ /* If there's only one MR, no need to replace LKey in WQE. */
+ if (unlikely(mlx5_mr_btree_len(&rxq->mr_ctrl.cache_bh) > 1))
+ wqe->lkey = mlx5_rx_mb2mr(rxq, rep);
if (len > DATA_LEN(seg)) {
len -= DATA_LEN(seg);
++NB_SEGS(pkt);
pkt = NULL;
--pkts_n;
++i;
-skip:
/* Align consumer index to the next stride. */
rq_ci >>= sges_n;
++rq_ci;
return 0;
/* Update the consumer index. */
rxq->rq_ci = rq_ci >> sges_n;
- rte_wmb();
- *rxq->cq_db = htonl(rxq->cq_ci);
- rte_wmb();
- *rxq->rq_db = htonl(rxq->rq_ci);
+ rte_cio_wmb();
+ *rxq->cq_db = rte_cpu_to_be_32(rxq->cq_ci);
+ rte_cio_wmb();
+ *rxq->rq_db = rte_cpu_to_be_32(rxq->rq_ci);
+#ifdef MLX5_PMD_SOFT_COUNTERS
+ /* Increment packets counter. */
+ rxq->stats.ipackets += i;
+#endif
+ return i;
+}
+
+void
+mlx5_mprq_buf_free_cb(void *addr __rte_unused, void *opaque)
+{
+ struct mlx5_mprq_buf *buf = opaque;
+
+ if (rte_atomic16_read(&buf->refcnt) == 1) {
+ rte_mempool_put(buf->mp, buf);
+ } else if (rte_atomic16_add_return(&buf->refcnt, -1) == 0) {
+ rte_atomic16_set(&buf->refcnt, 1);
+ rte_mempool_put(buf->mp, buf);
+ }
+}
+
+void
+mlx5_mprq_buf_free(struct mlx5_mprq_buf *buf)
+{
+ mlx5_mprq_buf_free_cb(NULL, buf);
+}
+
+static inline void
+mprq_buf_replace(struct mlx5_rxq_data *rxq, uint16_t rq_idx)
+{
+ struct mlx5_mprq_buf *rep = rxq->mprq_repl;
+ volatile struct mlx5_wqe_data_seg *wqe =
+ &((volatile struct mlx5_wqe_mprq *)rxq->wqes)[rq_idx].dseg;
+ void *addr;
+
+ assert(rep != NULL);
+ /* Replace MPRQ buf. */
+ (*rxq->mprq_bufs)[rq_idx] = rep;
+ /* Replace WQE. */
+ addr = mlx5_mprq_buf_addr(rep);
+ wqe->addr = rte_cpu_to_be_64((uintptr_t)addr);
+ /* If there's only one MR, no need to replace LKey in WQE. */
+ if (unlikely(mlx5_mr_btree_len(&rxq->mr_ctrl.cache_bh) > 1))
+ wqe->lkey = mlx5_rx_addr2mr(rxq, (uintptr_t)addr);
+ /* Stash a mbuf for next replacement. */
+ if (likely(!rte_mempool_get(rxq->mprq_mp, (void **)&rep)))
+ rxq->mprq_repl = rep;
+ else
+ rxq->mprq_repl = NULL;
+}
+
+/**
+ * DPDK callback for RX with Multi-Packet RQ support.
+ *
+ * @param dpdk_rxq
+ * Generic pointer to RX queue structure.
+ * @param[out] pkts
+ * Array to store received packets.
+ * @param pkts_n
+ * Maximum number of packets in array.
+ *
+ * @return
+ * Number of packets successfully received (<= pkts_n).
+ */
+uint16_t
+mlx5_rx_burst_mprq(void *dpdk_rxq, struct rte_mbuf **pkts, uint16_t pkts_n)
+{
+ struct mlx5_rxq_data *rxq = dpdk_rxq;
+ const unsigned int strd_n = 1 << rxq->strd_num_n;
+ const unsigned int strd_sz = 1 << rxq->strd_sz_n;
+ const unsigned int strd_shift =
+ MLX5_MPRQ_STRIDE_SHIFT_BYTE * rxq->strd_shift_en;
+ const unsigned int cq_mask = (1 << rxq->cqe_n) - 1;
+ const unsigned int wq_mask = (1 << rxq->elts_n) - 1;
+ volatile struct mlx5_cqe *cqe = &(*rxq->cqes)[rxq->cq_ci & cq_mask];
+ unsigned int i = 0;
+ uint32_t rq_ci = rxq->rq_ci;
+ uint16_t consumed_strd = rxq->consumed_strd;
+ struct mlx5_mprq_buf *buf = (*rxq->mprq_bufs)[rq_ci & wq_mask];
+
+ while (i < pkts_n) {
+ struct rte_mbuf *pkt;
+ void *addr;
+ int ret;
+ unsigned int len;
+ uint16_t strd_cnt;
+ uint16_t strd_idx;
+ uint32_t offset;
+ uint32_t byte_cnt;
+ volatile struct mlx5_mini_cqe8 *mcqe = NULL;
+ uint32_t rss_hash_res = 0;
+
+ if (consumed_strd == strd_n) {
+ /* Replace WQE only if the buffer is still in use. */
+ if (rte_atomic16_read(&buf->refcnt) > 1) {
+ mprq_buf_replace(rxq, rq_ci & wq_mask);
+ /* Release the old buffer. */
+ mlx5_mprq_buf_free(buf);
+ } else if (unlikely(rxq->mprq_repl == NULL)) {
+ struct mlx5_mprq_buf *rep;
+
+ /*
+ * Currently, the MPRQ mempool is out of buffer
+ * and doing memcpy regardless of the size of Rx
+ * packet. Retry allocation to get back to
+ * normal.
+ */
+ if (!rte_mempool_get(rxq->mprq_mp,
+ (void **)&rep))
+ rxq->mprq_repl = rep;
+ }
+ /* Advance to the next WQE. */
+ consumed_strd = 0;
+ ++rq_ci;
+ buf = (*rxq->mprq_bufs)[rq_ci & wq_mask];
+ }
+ cqe = &(*rxq->cqes)[rxq->cq_ci & cq_mask];
+ ret = mlx5_rx_poll_len(rxq, cqe, cq_mask, &mcqe);
+ if (!ret)
+ break;
+ byte_cnt = ret;
+ strd_cnt = (byte_cnt & MLX5_MPRQ_STRIDE_NUM_MASK) >>
+ MLX5_MPRQ_STRIDE_NUM_SHIFT;
+ assert(strd_cnt);
+ consumed_strd += strd_cnt;
+ if (byte_cnt & MLX5_MPRQ_FILLER_MASK)
+ continue;
+ if (mcqe == NULL) {
+ rss_hash_res = rte_be_to_cpu_32(cqe->rx_hash_res);
+ strd_idx = rte_be_to_cpu_16(cqe->wqe_counter);
+ } else {
+ /* mini-CQE for MPRQ doesn't have hash result. */
+ strd_idx = rte_be_to_cpu_16(mcqe->stride_idx);
+ }
+ assert(strd_idx < strd_n);
+ assert(!((rte_be_to_cpu_16(cqe->wqe_id) ^ rq_ci) & wq_mask));
+ /*
+ * Currently configured to receive a packet per a stride. But if
+ * MTU is adjusted through kernel interface, device could
+ * consume multiple strides without raising an error. In this
+ * case, the packet should be dropped because it is bigger than
+ * the max_rx_pkt_len.
+ */
+ if (unlikely(strd_cnt > 1)) {
+ ++rxq->stats.idropped;
+ continue;
+ }
+ pkt = rte_pktmbuf_alloc(rxq->mp);
+ if (unlikely(pkt == NULL)) {
+ ++rxq->stats.rx_nombuf;
+ break;
+ }
+ len = (byte_cnt & MLX5_MPRQ_LEN_MASK) >> MLX5_MPRQ_LEN_SHIFT;
+ assert((int)len >= (rxq->crc_present << 2));
+ if (rxq->crc_present)
+ len -= RTE_ETHER_CRC_LEN;
+ offset = strd_idx * strd_sz + strd_shift;
+ addr = RTE_PTR_ADD(mlx5_mprq_buf_addr(buf), offset);
+ /* Initialize the offload flag. */
+ pkt->ol_flags = 0;
+ /*
+ * Memcpy packets to the target mbuf if:
+ * - The size of packet is smaller than mprq_max_memcpy_len.
+ * - Out of buffer in the Mempool for Multi-Packet RQ.
+ */
+ if (len <= rxq->mprq_max_memcpy_len || rxq->mprq_repl == NULL) {
+ /*
+ * When memcpy'ing packet due to out-of-buffer, the
+ * packet must be smaller than the target mbuf.
+ */
+ if (unlikely(rte_pktmbuf_tailroom(pkt) < len)) {
+ rte_pktmbuf_free_seg(pkt);
+ ++rxq->stats.idropped;
+ continue;
+ }
+ rte_memcpy(rte_pktmbuf_mtod(pkt, void *), addr, len);
+ } else {
+ rte_iova_t buf_iova;
+ struct rte_mbuf_ext_shared_info *shinfo;
+ uint16_t buf_len = strd_cnt * strd_sz;
+
+ /* Increment the refcnt of the whole chunk. */
+ rte_atomic16_add_return(&buf->refcnt, 1);
+ assert((uint16_t)rte_atomic16_read(&buf->refcnt) <=
+ strd_n + 1);
+ addr = RTE_PTR_SUB(addr, RTE_PKTMBUF_HEADROOM);
+ /*
+ * MLX5 device doesn't use iova but it is necessary in a
+ * case where the Rx packet is transmitted via a
+ * different PMD.
+ */
+ buf_iova = rte_mempool_virt2iova(buf) +
+ RTE_PTR_DIFF(addr, buf);
+ shinfo = rte_pktmbuf_ext_shinfo_init_helper(addr,
+ &buf_len, mlx5_mprq_buf_free_cb, buf);
+ /*
+ * EXT_ATTACHED_MBUF will be set to pkt->ol_flags when
+ * attaching the stride to mbuf and more offload flags
+ * will be added below by calling rxq_cq_to_mbuf().
+ * Other fields will be overwritten.
+ */
+ rte_pktmbuf_attach_extbuf(pkt, addr, buf_iova, buf_len,
+ shinfo);
+ rte_pktmbuf_reset_headroom(pkt);
+ assert(pkt->ol_flags == EXT_ATTACHED_MBUF);
+ /*
+ * Prevent potential overflow due to MTU change through
+ * kernel interface.
+ */
+ if (unlikely(rte_pktmbuf_tailroom(pkt) < len)) {
+ rte_pktmbuf_free_seg(pkt);
+ ++rxq->stats.idropped;
+ continue;
+ }
+ }
+ rxq_cq_to_mbuf(rxq, pkt, cqe, rss_hash_res);
+ PKT_LEN(pkt) = len;
+ DATA_LEN(pkt) = len;
+ PORT(pkt) = rxq->port_id;
+#ifdef MLX5_PMD_SOFT_COUNTERS
+ /* Increment bytes counter. */
+ rxq->stats.ibytes += PKT_LEN(pkt);
+#endif
+ /* Return packet. */
+ *(pkts++) = pkt;
+ ++i;
+ }
+ /* Update the consumer indexes. */
+ rxq->consumed_strd = consumed_strd;
+ rte_cio_wmb();
+ *rxq->cq_db = rte_cpu_to_be_32(rxq->cq_ci);
+ if (rq_ci != rxq->rq_ci) {
+ rxq->rq_ci = rq_ci;
+ rte_cio_wmb();
+ *rxq->rq_db = rte_cpu_to_be_32(rxq->rq_ci);
+ }
#ifdef MLX5_PMD_SOFT_COUNTERS
/* Increment packets counter. */
rxq->stats.ipackets += i;
* Number of packets successfully transmitted (<= pkts_n).
*/
uint16_t
-removed_tx_burst(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
+removed_tx_burst(void *dpdk_txq __rte_unused,
+ struct rte_mbuf **pkts __rte_unused,
+ uint16_t pkts_n __rte_unused)
{
- (void)dpdk_txq;
- (void)pkts;
- (void)pkts_n;
+ rte_mb();
return 0;
}
* Number of packets successfully received (<= pkts_n).
*/
uint16_t
-removed_rx_burst(void *dpdk_rxq, struct rte_mbuf **pkts, uint16_t pkts_n)
+removed_rx_burst(void *dpdk_txq __rte_unused,
+ struct rte_mbuf **pkts __rte_unused,
+ uint16_t pkts_n __rte_unused)
+{
+ rte_mb();
+ return 0;
+}
+
+/*
+ * Vectorized Rx/Tx routines are not compiled in when required vector
+ * instructions are not supported on a target architecture. The following null
+ * stubs are needed for linkage when those are not included outside of this file
+ * (e.g. mlx5_rxtx_vec_sse.c for x86).
+ */
+
+__rte_weak uint16_t
+mlx5_rx_burst_vec(void *dpdk_txq __rte_unused,
+ struct rte_mbuf **pkts __rte_unused,
+ uint16_t pkts_n __rte_unused)
{
- (void)dpdk_rxq;
- (void)pkts;
- (void)pkts_n;
return 0;
}
+__rte_weak int
+mlx5_rxq_check_vec_support(struct mlx5_rxq_data *rxq __rte_unused)
+{
+ return -ENOTSUP;
+}
+
+__rte_weak int
+mlx5_check_vec_rx_support(struct rte_eth_dev *dev __rte_unused)
+{
+ return -ENOTSUP;
+}
+
/**
- * DPDK callback for rx queue interrupt enable.
+ * DPDK callback to check the status of a tx descriptor.
*
- * @param dev
- * Pointer to Ethernet device structure.
- * @param rx_queue_id
- * RX queue number
+ * @param tx_queue
+ * The tx queue.
+ * @param[in] offset
+ * The index of the descriptor in the ring.
*
* @return
- * 0 on success, negative on failure.
+ * The status of the tx descriptor.
*/
int
-mlx5_rx_intr_enable(struct rte_eth_dev *dev, uint16_t rx_queue_id)
+mlx5_tx_descriptor_status(void *tx_queue, uint16_t offset)
{
-#ifdef HAVE_UPDATE_CQ_CI
- struct priv *priv = mlx5_get_priv(dev);
- struct rxq *rxq = (*priv->rxqs)[rx_queue_id];
- struct rxq_ctrl *rxq_ctrl = container_of(rxq, struct rxq_ctrl, rxq);
- struct ibv_cq *cq = rxq_ctrl->cq;
- uint16_t ci = rxq->cq_ci;
- int ret = 0;
-
- ibv_mlx5_exp_update_cq_ci(cq, ci);
- ret = ibv_req_notify_cq(cq, 0);
-#else
- int ret = -1;
- (void)dev;
- (void)rx_queue_id;
-#endif
- if (ret)
- WARN("unable to arm interrupt on rx queue %d", rx_queue_id);
- return ret;
+ (void)tx_queue;
+ (void)offset;
+ return RTE_ETH_TX_DESC_FULL;
}
/**
- * DPDK callback for rx queue interrupt disable.
+ * Configure the TX function to use.
*
* @param dev
- * Pointer to Ethernet device structure.
- * @param rx_queue_id
- * RX queue number
+ * Pointer to private data structure.
*
* @return
- * 0 on success, negative on failure.
+ * Pointer to selected Tx burst function.
*/
-int
-mlx5_rx_intr_disable(struct rte_eth_dev *dev, uint16_t rx_queue_id)
+eth_tx_burst_t
+mlx5_select_tx_function(struct rte_eth_dev *dev)
{
-#ifdef HAVE_UPDATE_CQ_CI
- struct priv *priv = mlx5_get_priv(dev);
- struct rxq *rxq = (*priv->rxqs)[rx_queue_id];
- struct rxq_ctrl *rxq_ctrl = container_of(rxq, struct rxq_ctrl, rxq);
- struct ibv_cq *cq = rxq_ctrl->cq;
- struct ibv_cq *ev_cq;
- void *ev_ctx;
- int ret = 0;
-
- ret = ibv_get_cq_event(cq->channel, &ev_cq, &ev_ctx);
- if (ret || ev_cq != cq)
- ret = -1;
- else
- ibv_ack_cq_events(cq, 1);
-#else
- int ret = -1;
(void)dev;
- (void)rx_queue_id;
-#endif
- if (ret)
- WARN("unable to disable interrupt on rx queue %d",
- rx_queue_id);
- return ret;
+ return removed_tx_burst;
}
+
+