#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 "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 inline void
-txq_complete(struct txq *txq) __attribute__((always_inline));
-
-static inline uint32_t
-txq_mp2mr(struct txq *txq, struct rte_mempool *mp)
- __attribute__((always_inline));
-
-static inline void
-mlx5_tx_dbrec(struct txq *txq) __attribute__((always_inline));
-
-static inline uint32_t
-rxq_cq_to_pkt_type(volatile struct mlx5_cqe *cqe)
- __attribute__((always_inline));
+static __rte_always_inline uint32_t
+rxq_cq_to_pkt_type(volatile struct mlx5_cqe *cqe);
-static inline int
+static __rte_always_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));
+ uint16_t cqe_cnt, uint32_t *rss_hash);
-static inline uint32_t
-rxq_cq_to_ol_flags(struct rxq *rxq, volatile struct mlx5_cqe *cqe)
- __attribute__((always_inline));
+static __rte_always_inline uint32_t
+rxq_cq_to_ol_flags(struct rxq *rxq, volatile struct mlx5_cqe *cqe);
-#ifndef NDEBUG
+uint32_t mlx5_ptype_table[] __rte_cache_aligned = {
+ [0xff] = RTE_PTYPE_ALL_MASK, /* Last entry for errored packet. */
+};
/**
- * 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;
+ /* 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
+ */
+ /* 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;
+ /* 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)[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)[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)[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_L4_TCP;
+ (*p)[0x46] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0xc5] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_TCP;
+ (*p)[0xc6] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_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_L4_UDP;
+ (*p)[0x4a] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_UDP;
+ (*p)[0xc9] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_L4_UDP;
+ (*p)[0xca] = RTE_PTYPE_L2_ETHER | RTE_PTYPE_L3_IPV6_EXT_UNKNOWN |
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN |
+ RTE_PTYPE_L4_UDP;
}
/**
- * Return the address of the WQE.
+ * Return the size of tailroom of WQ.
*
* @param txq
* Pointer to TX queue structure.
- * @param wqe_ci
- * WQE consumer index.
+ * @param addr
+ * Pointer to tail of WQ.
*
* @return
- * WQE address.
+ * Size of tailroom.
*/
-static inline uintptr_t *
-tx_mlx5_wqe(struct txq *txq, uint16_t ci)
+static inline size_t
+tx_mlx5_wq_tailroom(struct txq *txq, void *addr)
{
- ci &= ((1 << txq->wqe_n) - 1);
- return (uintptr_t *)((uintptr_t)txq->wqes + ci * MLX5_WQE_SIZE);
+ size_t tailroom;
+ tailroom = (uintptr_t)(txq->wqes) +
+ (1 << txq->wqe_n) * MLX5_WQE_SIZE -
+ (uintptr_t)addr;
+ return tailroom;
}
/**
- * Manage TX completions.
+ * Copy data to tailroom of circular queue.
*
- * When sending a burst, mlx5_tx_burst() posts several WRs.
+ * @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.
*
- * @param txq
- * Pointer to TX queue structure.
+ * @return
+ * Pointer after copied data.
*/
-static inline void
-txq_complete(struct txq *txq)
+static inline void *
+mlx5_copy_to_wq(void *dst, const void *src, size_t n,
+ void *base, size_t tailroom)
{
- 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;
+ void *ret;
- 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;
- }
-#endif /* NDEBUG */
- ++cq_ci;
- } while (1);
- if (unlikely(cqe == NULL))
- return;
- ctrl = (volatile struct mlx5_wqe_ctrl *)
- tx_mlx5_wqe(txq, ntohs(cqe->wqe_counter));
- 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;
+ 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;
}
- txq->cq_ci = cq_ci;
- txq->elts_tail = elts_tail;
- /* Update the consumer index. */
- rte_wmb();
- *txq->cq_db = htonl(cq_ci);
+ return ret;
}
/**
- * Get Memory Pool (MP) from mbuf. If mbuf is indirect, the pool from which
- * the cloned mbuf is allocated is returned instead.
+ * DPDK callback to check the status of a tx descriptor.
*
- * @param buf
- * Pointer to mbuf.
+ * @param tx_queue
+ * The tx queue.
+ * @param[in] offset
+ * The index of the descriptor in the ring.
*
* @return
- * Memory pool where data is located for given mbuf.
+ * The status of the tx descriptor.
*/
-static struct rte_mempool *
-txq_mb2mp(struct rte_mbuf *buf)
+int
+mlx5_tx_descriptor_status(void *tx_queue, uint16_t offset)
{
- if (unlikely(RTE_MBUF_INDIRECT(buf)))
- return rte_mbuf_from_indirect(buf)->pool;
- return buf->pool;
+ struct txq *txq = tx_queue;
+ uint16_t used;
+
+ mlx5_tx_complete(txq);
+ used = txq->elts_head - txq->elts_tail;
+ if (offset < used)
+ return RTE_ETH_TX_DESC_FULL;
+ return RTE_ETH_TX_DESC_DONE;
}
/**
- * 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.
+ * DPDK callback to check the status of a rx descriptor.
*
- * @param txq
- * Pointer to TX queue structure.
- * @param[in] mp
- * Memory Pool for which a Memory Region lkey must be returned.
+ * @param rx_queue
+ * The rx queue.
+ * @param[in] offset
+ * The index of the descriptor in the ring.
*
* @return
- * mr->lkey on success, (uint32_t)-1 on failure.
+ * The status of the tx descriptor.
*/
-static inline uint32_t
-txq_mp2mr(struct txq *txq, struct rte_mempool *mp)
+int
+mlx5_rx_descriptor_status(void *rx_queue, uint16_t offset)
{
- unsigned int i;
- uint32_t lkey = (uint32_t)-1;
+ 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);
+ const unsigned int cqe_cnt = cqe_n - 1;
+ unsigned int cq_ci;
+ unsigned int used;
- 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;
- }
+ /* if we are processing a compressed cqe */
+ if (zip->ai) {
+ used = zip->cqe_cnt - zip->ca;
+ cq_ci = zip->cq_ci;
+ } else {
+ used = 0;
+ cq_ci = rxq->cq_ci;
}
- 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.
- */
-static inline void
-mlx5_tx_dbrec(struct txq *txq)
-{
- uint8_t *dst = (uint8_t *)((uintptr_t)txq->bf_reg + txq->bf_offset);
- uint32_t data[4] = {
- htonl((txq->wqe_ci << 8) | MLX5_OPCODE_SEND),
- htonl(txq->qp_num_8s),
- 0,
- 0,
- };
- rte_wmb();
- *txq->qp_db = htonl(txq->wqe_ci);
- /* Ensure ordering between DB record and BF copy. */
- rte_wmb();
- memcpy(dst, (uint8_t *)data, 16);
- txq->bf_offset ^= (1 << txq->bf_buf_size);
-}
-
-/**
- * Prefetch a CQE.
- *
- * @param txq
- * Pointer to TX queue structure.
- * @param cqe_ci
- * CQE consumer index.
- */
-static inline void
-tx_prefetch_cqe(struct txq *txq, uint16_t ci)
-{
- volatile struct mlx5_cqe *cqe;
+ cqe = &(*rxq->cqes)[cq_ci & cqe_cnt];
+ while (check_cqe(cqe, cqe_n, cq_ci) == 0) {
+ int8_t op_own;
+ unsigned int n;
- cqe = &(*txq->cqes)[ci & ((1 << txq->cqe_n) - 1)];
- rte_prefetch0(cqe);
+ op_own = cqe->op_own;
+ if (MLX5_CQE_FORMAT(op_own) == MLX5_COMPRESSED)
+ n = ntohl(cqe->byte_cnt);
+ else
+ n = 1;
+ cq_ci += n;
+ used += 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;
}
/**
{
struct txq *txq = (struct txq *)dpdk_txq;
uint16_t elts_head = txq->elts_head;
- const unsigned int elts_n = 1 << txq->elts_n;
+ const uint16_t elts_n = 1 << txq->elts_n;
+ const uint16_t elts_m = elts_n - 1;
unsigned int i = 0;
unsigned int j = 0;
- unsigned int max;
+ unsigned int k = 0;
+ uint16_t max_elts;
+ unsigned int max_inline = txq->max_inline;
+ const unsigned int inline_en = !!max_inline && txq->inline_en;
+ uint16_t max_wqe;
unsigned int comp;
volatile struct mlx5_wqe_v *wqe = NULL;
+ volatile struct mlx5_wqe_ctrl *last_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. */
- tx_prefetch_cqe(txq, txq->cq_ci);
- tx_prefetch_cqe(txq, txq->cq_ci + 1);
rte_prefetch0(*pkts);
/* Start processing. */
- txq_complete(txq);
- max = (elts_n - (elts_head - txq->elts_tail));
- if (max > elts_n)
- max -= elts_n;
+ mlx5_tx_complete(txq);
+ max_elts = (elts_n - (elts_head - txq->elts_tail));
+ 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;
+ unsigned int sg = 0; /* counter of additional segs attached. */
uintptr_t addr;
uint64_t naddr;
- uint16_t pkt_inline_sz = MLX5_WQE_DWORD_SIZE;
+ 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;
+ uint16_t tso_segsz = 0;
#ifdef MLX5_PMD_SOFT_COUNTERS
uint32_t total_length = 0;
#endif
/* first_seg */
- buf = *(pkts++);
+ 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)
+ if (max_elts < segs_n)
break;
- max -= segs_n;
+ max_elts -= 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);
+ if (pkts_n - i > 1)
+ rte_prefetch0(*(pkts + 1));
addr = rte_pktmbuf_mtod(buf, uintptr_t);
length = DATA_LEN(buf);
ehdr = (((uint8_t *)addr)[1] << 8) |
#ifdef MLX5_PMD_SOFT_COUNTERS
total_length = length;
#endif
- assert(length >= MLX5_WQE_DWORD_SIZE);
+ if (length < (MLX5_WQE_DWORD_SIZE + 2))
+ break;
/* Update element. */
- (*txq->elts)[elts_head] = buf;
- elts_head = (elts_head + 1) & (elts_n - 1);
+ (*txq->elts)[elts_head & elts_m] = buf;
/* Prefetch next buffer data. */
- if (pkts_n > 1) {
- volatile void *pkt_addr;
-
- pkt_addr = rte_pktmbuf_mtod(*pkts, volatile void *);
- rte_prefetch0(pkt_addr);
- }
+ if (pkts_n - i > 1)
+ rte_prefetch0(
+ rte_pktmbuf_mtod(*(pkts + 1), volatile void *));
/* 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;
+ 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;
- /*
- * Start by copying the Ethernet header minus the first two
- * bytes which will be appended at the end of the Ethernet
- * segment.
- */
- memcpy((uint8_t *)raw, ((uint8_t *)addr) + 2, 16);
- length -= MLX5_WQE_DWORD_SIZE;
- addr += 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);
-
- memcpy((uint8_t *)(raw + MLX5_WQE_DWORD_SIZE - 2 -
- sizeof(vlan)),
- &vlan, sizeof(vlan));
- addr -= sizeof(vlan);
- length += sizeof(vlan);
+ 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;
+ }
+ raw += MLX5_WQE_DWORD_SIZE;
+ 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;
+ tso_segsz = buf->tso_segsz;
+
+ 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);
+ 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;
+ /* Include padding for TSO header. */
+ copy_b = MLX5_WQE_DS(copy_b) *
+ MLX5_WQE_DWORD_SIZE;
+ pkt_inline_sz += copy_b;
+ raw += copy_b;
+ } 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;
+ k++;
+ goto next_wqe;
+ }
+ }
}
/* Inline if enough room. */
- if (txq->max_inline != 0) {
+ if (inline_en || tso) {
+ uint32_t inl;
uintptr_t end = (uintptr_t)
(((uintptr_t)txq->wqes) +
(1 << txq->wqe_n) * MLX5_WQE_SIZE);
- uint16_t max_inline =
- txq->max_inline * RTE_CACHE_LINE_SIZE;
- uint16_t room;
-
- /*
- * raw starts two bytes before the boundary to
- * continue the above copy of packet data.
- */
- raw += MLX5_WQE_DWORD_SIZE - 2;
- room = end - (uintptr_t)raw;
- if (room > max_inline) {
- uintptr_t addr_end = (addr + max_inline) &
- ~(RTE_CACHE_LINE_SIZE - 1);
- uint16_t copy_b = ((addr_end - addr) > length) ?
- length :
- (addr_end - addr);
-
+ unsigned int inline_room = max_inline *
+ RTE_CACHE_LINE_SIZE -
+ (pkt_inline_sz - 2) -
+ !!tso * sizeof(inl);
+ uintptr_t addr_end = (addr + inline_room) &
+ ~(RTE_CACHE_LINE_SIZE - 1);
+ unsigned int copy_b = (addr_end > addr) ?
+ RTE_MIN((addr_end - addr), length) :
+ 0;
+
+ 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) {
+ inl = htonl(copy_b | MLX5_INLINE_SEG);
+ 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;
- /* Sanity check. */
- assert(addr <= addr_end);
}
/*
- * 2 DWORDs consumed by the WQE header + 1 DSEG +
+ * 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) {
- dseg = (volatile rte_v128u32_t *)
- ((uintptr_t)wqe +
- (ds * MLX5_WQE_DWORD_SIZE));
- if ((uintptr_t)dseg >= end)
+ 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 *)
- txq->wqes;
+ ((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 {
naddr = htonll(addr);
*dseg = (rte_v128u32_t){
htonl(length),
- txq_mp2mr(txq, txq_mb2mp(buf)),
+ mlx5_tx_mb2mr(txq, buf),
naddr,
naddr >> 32,
};
*/
assert(!(MLX5_WQE_SIZE % MLX5_WQE_DWORD_SIZE));
if (!(ds % (MLX5_WQE_SIZE / MLX5_WQE_DWORD_SIZE))) {
- unsigned int n = (txq->wqe_ci + ((ds + 3) / 4)) &
- ((1 << txq->wqe_n) - 1);
-
+ if (unlikely(--max_wqe == 0))
+ break;
dseg = (volatile rte_v128u32_t *)
- tx_mlx5_wqe(txq, n);
- rte_prefetch0(tx_mlx5_wqe(txq, n + 1));
+ tx_mlx5_wqe(txq, txq->wqe_ci + ds / 4);
+ rte_prefetch0(tx_mlx5_wqe(txq,
+ txq->wqe_ci + ds / 4 + 1));
} else {
++dseg;
}
naddr = htonll(rte_pktmbuf_mtod(buf, uintptr_t));
*dseg = (rte_v128u32_t){
htonl(length),
- txq_mp2mr(txq, txq_mb2mp(buf)),
+ mlx5_tx_mb2mr(txq, buf),
naddr,
naddr >> 32,
};
- (*txq->elts)[elts_head] = buf;
- elts_head = (elts_head + 1) & (elts_n - 1);
- ++j;
- --segs_n;
- if (segs_n)
+ (*txq->elts)[++elts_head & elts_m] = buf;
+ ++sg;
+ /* Advance counter only if all segs are successfully posted. */
+ if (sg < segs_n)
goto next_seg;
else
- --pkts_n;
+ j += sg;
next_pkt:
+ ++elts_head;
+ ++pkts;
++i;
/* Initialize known and common part of the WQE structure. */
- 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),
- };
+ 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(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;
+ /* Save the last successful WQE for completion request */
+ last_wqe = (volatile struct mlx5_wqe_ctrl *)wqe;
#ifdef MLX5_PMD_SOFT_COUNTERS
/* Increment sent bytes counter. */
txq->stats.obytes += total_length;
#endif
- } while (pkts_n);
+ } while (i < pkts_n);
/* Take a shortcut if nothing must be sent. */
- if (unlikely(i == 0))
+ if (unlikely((i + k) == 0))
return 0;
+ txq->elts_head += (i + j);
/* Check whether completion threshold has been reached. */
- comp = txq->elts_comp + i + j;
+ 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);
+ last_wqe->ctrl2 = htonl(8);
/* Save elts_head in unused "immediate" field of WQE. */
- w->ctrl3 = elts_head;
+ last_wqe->ctrl3 = txq->elts_head;
txq->elts_comp = 0;
} else {
txq->elts_comp = comp;
txq->stats.opackets += i;
#endif
/* Ring QP doorbell. */
- mlx5_tx_dbrec(txq);
- txq->elts_head = elts_head;
+ mlx5_tx_dbrec(txq, (volatile struct mlx5_wqe *)last_wqe);
return i;
}
{
struct txq *txq = (struct txq *)dpdk_txq;
uint16_t elts_head = txq->elts_head;
- const unsigned int elts_n = 1 << txq->elts_n;
+ const uint16_t elts_n = 1 << txq->elts_n;
+ const uint16_t elts_m = elts_n - 1;
unsigned int i = 0;
unsigned int j = 0;
- unsigned int max;
+ uint16_t max_elts;
+ uint16_t max_wqe;
unsigned int comp;
struct mlx5_mpw mpw = {
.state = MLX5_MPW_STATE_CLOSED,
if (unlikely(!pkts_n))
return 0;
/* Prefetch first packet cacheline. */
- tx_prefetch_cqe(txq, txq->cq_ci);
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;
+ mlx5_tx_complete(txq);
+ max_elts = (elts_n - (elts_head - txq->elts_tail));
+ 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;
* that one ring entry remains unused.
*/
assert(segs_n);
- if (max < segs_n + 1)
+ if (max_elts < segs_n)
break;
/* Do not bother with large packets MPW cannot handle. */
if (segs_n > MLX5_MPW_DSEG_MAX)
break;
- max -= segs_n;
+ max_elts -= segs_n;
--pkts_n;
/* Should we enable HW CKSUM offload */
if (buf->ol_flags &
(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;
}
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;
+ (*txq->elts)[elts_head++ & elts_m] = 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)),
+ .lkey = mlx5_tx_mb2mr(txq, buf),
.addr = htonll(addr),
};
- elts_head = elts_head_next;
#if defined(MLX5_PMD_SOFT_COUNTERS) || !defined(NDEBUG)
length += DATA_LEN(buf);
#endif
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;
/* Ring QP doorbell. */
if (mpw.state == MLX5_MPW_STATE_OPENED)
mlx5_mpw_close(txq, &mpw);
- mlx5_tx_dbrec(txq);
+ mlx5_tx_dbrec(txq, mpw.wqe);
txq->elts_head = elts_head;
return i;
}
{
struct txq *txq = (struct txq *)dpdk_txq;
uint16_t elts_head = txq->elts_head;
- const unsigned int elts_n = 1 << txq->elts_n;
+ const uint16_t elts_n = 1 << txq->elts_n;
+ const uint16_t elts_m = elts_n - 1;
unsigned int i = 0;
unsigned int j = 0;
- unsigned int max;
+ uint16_t max_elts;
+ 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;
if (unlikely(!pkts_n))
return 0;
/* Prefetch first packet cacheline. */
- tx_prefetch_cqe(txq, txq->cq_ci);
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;
+ mlx5_tx_complete(txq);
+ max_elts = (elts_n - (elts_head - txq->elts_tail));
do {
struct rte_mbuf *buf = *(pkts++);
- unsigned int elts_head_next;
uintptr_t addr;
uint32_t length;
unsigned int segs_n = buf->nb_segs;
* that one ring entry remains unused.
*/
assert(segs_n);
- if (max < segs_n + 1)
+ if (max_elts < segs_n)
break;
/* Do not bother with large packets MPW cannot handle. */
if (segs_n > MLX5_MPW_DSEG_MAX)
break;
- max -= segs_n;
+ max_elts -= 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))
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;
}
do {
volatile struct mlx5_wqe_data_seg *dseg;
- elts_head_next =
- (elts_head + 1) & (elts_n - 1);
assert(buf);
- (*txq->elts)[elts_head] = buf;
+ (*txq->elts)[elts_head++ & elts_m] = 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)),
+ .lkey = mlx5_tx_mb2mr(txq, buf),
.addr = htonll(addr),
};
- elts_head = elts_head_next;
#if defined(MLX5_PMD_SOFT_COUNTERS) || !defined(NDEBUG)
length += DATA_LEN(buf);
#endif
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);
addr = rte_pktmbuf_mtod(buf, uintptr_t);
- (*txq->elts)[elts_head] = buf;
+ (*txq->elts)[elts_head++ & elts_m] = buf;
/* Maximum number of bytes before wrapping. */
max = ((((uintptr_t)(txq->wqes)) +
(1 << txq->wqe_n) *
rte_memcpy((void *)(uintptr_t)mpw.data.raw,
(void *)addr,
length);
- mpw.data.raw += length;
+
+ if (length == max)
+ mpw.data.raw =
+ (volatile void *)txq->wqes;
+ else
+ mpw.data.raw += length;
}
- if ((uintptr_t)mpw.data.raw ==
- (uintptr_t)tx_mlx5_wqe(txq, 1 << txq->wqe_n))
- mpw.data.raw = (volatile void *)txq->wqes;
++mpw.pkts_n;
+ mpw.total_len += length;
++j;
if (mpw.pkts_n == MLX5_MPW_DSEG_MAX) {
mlx5_mpw_inline_close(txq, &mpw);
inline_room -= length;
}
}
- mpw.total_len += length;
- elts_head = elts_head_next;
#ifdef MLX5_PMD_SOFT_COUNTERS
/* Increment sent bytes counter. */
txq->stats.obytes += length;
mlx5_mpw_inline_close(txq, &mpw);
else if (mpw.state == MLX5_MPW_STATE_OPENED)
mlx5_mpw_close(txq, &mpw);
- mlx5_tx_dbrec(txq);
+ mlx5_tx_dbrec(txq, mpw.wqe);
+ txq->elts_head = elts_head;
+ return i;
+}
+
+/**
+ * Open an Enhanced MPW session.
+ *
+ * @param txq
+ * Pointer to TX queue structure.
+ * @param mpw
+ * Pointer to MPW session structure.
+ * @param length
+ * Packet length.
+ */
+static inline void
+mlx5_empw_new(struct txq *txq, struct mlx5_mpw *mpw, int padding)
+{
+ 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));
+ } else {
+ mpw->data.raw = (volatile void *)(mpw->wqe + 1);
+ }
+}
+
+/**
+ * Close an Enhanced MPW session.
+ *
+ * @param txq
+ * Pointer to TX queue structure.
+ * @param mpw
+ * Pointer to MPW session structure.
+ *
+ * @return
+ * Number of consumed WQEs.
+ */
+static inline uint16_t
+mlx5_empw_close(struct txq *txq, struct mlx5_mpw *mpw)
+{
+ uint16_t ret;
+
+ /* 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;
+ return ret;
+}
+
+/**
+ * DPDK callback for TX with Enhanced MPW support.
+ *
+ * @param dpdk_txq
+ * Generic pointer to TX queue structure.
+ * @param[in] pkts
+ * Packets to transmit.
+ * @param pkts_n
+ * Number of packets in array.
+ *
+ * @return
+ * Number of packets successfully transmitted (<= pkts_n).
+ */
+uint16_t
+mlx5_tx_burst_empw(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
+{
+ struct txq *txq = (struct txq *)dpdk_txq;
+ uint16_t elts_head = txq->elts_head;
+ const uint16_t elts_n = 1 << txq->elts_n;
+ const uint16_t elts_m = elts_n - 1;
+ unsigned int i = 0;
+ unsigned int j = 0;
+ uint16_t 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,
+ };
+
+ if (unlikely(!pkts_n))
+ return 0;
+ /* Start processing. */
+ mlx5_tx_complete(txq);
+ max_elts = (elts_n - (elts_head - txq->elts_tail));
+ /* 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++);
+ 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)
+ 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);
+ }
+ /* 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;
+
+ assert(buf);
+ (*txq->elts)[elts_head++ & elts_m] = 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 = mlx5_tx_mb2mr(txq, buf),
+ .addr = htonll(addr),
+ };
+#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.
+ */
+ 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);
+ (*txq->elts)[elts_head++ & elts_m] = 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),
+ mlx5_tx_mb2mr(txq, 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;
+ }
+#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;
}
static inline uint32_t
rxq_cq_to_pkt_type(volatile struct mlx5_cqe *cqe)
{
- uint32_t pkt_type;
- uint8_t flags = cqe->l4_hdr_type_etc;
+ uint8_t idx;
+ uint8_t pinfo = cqe->pkt_info;
+ uint16_t ptype = cqe->hdr_type_etc;
- if (cqe->pkt_info & MLX5_CQE_RX_TUNNEL_PACKET)
- pkt_type =
- TRANSPOSE(flags,
- MLX5_CQE_RX_OUTER_IPV4_PACKET,
- RTE_PTYPE_L3_IPV4) |
- TRANSPOSE(flags,
- MLX5_CQE_RX_OUTER_IPV6_PACKET,
- RTE_PTYPE_L3_IPV6) |
- TRANSPOSE(flags,
- MLX5_CQE_RX_IPV4_PACKET,
- RTE_PTYPE_INNER_L3_IPV4) |
- TRANSPOSE(flags,
- MLX5_CQE_RX_IPV6_PACKET,
- RTE_PTYPE_INNER_L3_IPV6);
- else
- pkt_type =
- TRANSPOSE(flags,
- MLX5_CQE_L3_HDR_TYPE_IPV6,
- RTE_PTYPE_L3_IPV6) |
- TRANSPOSE(flags,
- MLX5_CQE_L3_HDR_TYPE_IPV4,
- RTE_PTYPE_L3_IPV4);
- return pkt_type;
+ /*
+ * 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
+ */
+ idx = ((pinfo & 0x3) << 6) | ((ptype & 0xfc00) >> 10);
+ return mlx5_ptype_table[idx];
}
/**
struct rxq_zip *zip = &rxq->zip;
uint16_t cqe_n = cqe_cnt + 1;
int len = 0;
+ 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]);
+ (uintptr_t)(&(*rxq->cqes)[zip->ca & cqe_cnt].pkt_info);
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
zip->na += 8;
}
if (unlikely(rxq->zip.ai == rxq->zip.cqe_cnt)) {
- uint16_t idx = rxq->cq_ci + 1;
- uint16_t end = zip->cq_ci;
+ /* Invalidate the rest */
+ idx = zip->ca;
+ end = zip->cq_ci;
while (idx != end) {
(*rxq->cqes)[idx & cqe_cnt].op_own =
volatile struct mlx5_mini_cqe8 (*mc)[8] =
(volatile struct mlx5_mini_cqe8 (*)[8])
(uintptr_t)(&(*rxq->cqes)[rxq->cq_ci &
- cqe_cnt]);
+ cqe_cnt].pkt_info);
/* Fix endianness. */
zip->cqe_cnt = ntohl(cqe->byte_cnt);
* special case the second one is located 7 CQEs after
* the initial CQE instead of 8 for subsequent ones.
*/
- zip->ca = rxq->cq_ci & cqe_cnt;
+ zip->ca = rxq->cq_ci;
zip->na = zip->ca + 7;
/* Compute the next non compressed CQE. */
--rxq->cq_ci;
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;
+ }
} else {
len = ntohl(cqe->byte_cnt);
*rss_hash = ntohl(cqe->rx_hash_res);
rxq_cq_to_ol_flags(struct rxq *rxq, volatile struct mlx5_cqe *cqe)
{
uint32_t ol_flags = 0;
- uint8_t l3_hdr = (cqe->l4_hdr_type_etc) & MLX5_CQE_L3_HDR_TYPE_MASK;
- uint8_t l4_hdr = (cqe->l4_hdr_type_etc) & MLX5_CQE_L4_HDR_TYPE_MASK;
-
- if ((l3_hdr == MLX5_CQE_L3_HDR_TYPE_IPV4) ||
- (l3_hdr == MLX5_CQE_L3_HDR_TYPE_IPV6))
- ol_flags |= TRANSPOSE(cqe->hds_ip_ext,
- MLX5_CQE_L3_OK,
- PKT_RX_IP_CKSUM_GOOD);
- if ((l4_hdr == MLX5_CQE_L4_HDR_TYPE_TCP) ||
- (l4_hdr == MLX5_CQE_L4_HDR_TYPE_TCP_EMP_ACK) ||
- (l4_hdr == MLX5_CQE_L4_HDR_TYPE_TCP_ACK) ||
- (l4_hdr == MLX5_CQE_L4_HDR_TYPE_UDP))
- ol_flags |= TRANSPOSE(cqe->hds_ip_ext,
- MLX5_CQE_L4_OK,
- PKT_RX_L4_CKSUM_GOOD);
+ uint16_t flags = ntohs(cqe->hdr_type_etc);
+
+ 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);
if ((cqe->pkt_info & MLX5_CQE_RX_TUNNEL_PACKET) && (rxq->csum_l2tun))
ol_flags |=
- TRANSPOSE(cqe->l4_hdr_type_etc,
- MLX5_CQE_RX_OUTER_IP_CSUM_OK,
+ TRANSPOSE(flags,
+ MLX5_CQE_RX_L3_HDR_VALID,
PKT_RX_IP_CKSUM_GOOD) |
- TRANSPOSE(cqe->l4_hdr_type_etc,
- MLX5_CQE_RX_OUTER_TCP_UDP_CSUM_OK,
+ TRANSPOSE(flags,
+ MLX5_CQE_RX_L4_HDR_VALID,
PKT_RX_L4_CKSUM_GOOD);
return ol_flags;
}
&(*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;
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;
len = mlx5_rx_poll_len(rxq, cqe, cqe_cnt,
&rss_hash_res);
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);
+ 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->packet_type = rxq_cq_to_pkt_type(cqe);
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)) ||
- (cqe->sop_drop_qpn !=
- htonl(MLX5_FLOW_MARK_DEFAULT)))) {
- pkt->hash.fdir.hi =
- mlx5_flow_mark_get(cqe->sop_drop_qpn);
- pkt->ol_flags &= ~PKT_RX_RSS_HASH;
- pkt->ol_flags |= PKT_RX_FDIR | PKT_RX_FDIR_ID;
- }
- 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 (cqe->l4_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);
+ MLX5_FLOW_MARK_IS_VALID(cqe->sop_drop_qpn)) {
+ 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->crc_present)
- len -= ETHER_CRC_LEN;
}
+ if (rxq->csum | rxq->csum_l2tun)
+ pkt->ol_flags |= rxq_cq_to_ol_flags(rxq, cqe);
+ if (rxq->vlan_strip &&
+ (cqe->hdr_type_etc &
+ htons(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;
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
(void)pkts_n;
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).
+ */
+
+uint16_t __attribute__((weak))
+mlx5_tx_burst_raw_vec(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
+{
+ (void)dpdk_txq;
+ (void)pkts;
+ (void)pkts_n;
+ return 0;
+}
+
+uint16_t __attribute__((weak))
+mlx5_tx_burst_vec(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
+{
+ (void)dpdk_txq;
+ (void)pkts;
+ (void)pkts_n;
+ return 0;
+}
+
+uint16_t __attribute__((weak))
+mlx5_rx_burst_vec(void *dpdk_rxq, struct rte_mbuf **pkts, uint16_t pkts_n)
+{
+ (void)dpdk_rxq;
+ (void)pkts;
+ (void)pkts_n;
+ return 0;
+}
+
+int __attribute__((weak))
+priv_check_raw_vec_tx_support(struct priv *priv)
+{
+ (void)priv;
+ return -ENOTSUP;
+}
+
+int __attribute__((weak))
+priv_check_vec_tx_support(struct priv *priv)
+{
+ (void)priv;
+ return -ENOTSUP;
+}
+
+int __attribute__((weak))
+rxq_check_vec_support(struct rxq *rxq)
+{
+ (void)rxq;
+ return -ENOTSUP;
+}
+
+int __attribute__((weak))
+priv_check_vec_rx_support(struct priv *priv)
+{
+ (void)priv;
+ return -ENOTSUP;
+}