#include "mlx5_defs.h"
#include "mlx5_prm.h"
-static inline volatile struct mlx5_cqe64 *
-get_cqe64(volatile struct mlx5_cqe cqes[],
- unsigned int cqes_n, uint16_t *ci)
- __attribute__((always_inline));
+#ifndef NDEBUG
+/**
+ * Verify or set magic value in CQE.
+ *
+ * @param cqe
+ * Pointer to CQE.
+ *
+ * @return
+ * 0 the first time.
+ */
static inline int
-rx_poll_len(struct rxq *rxq) __attribute__((always_inline));
-
-static volatile struct mlx5_cqe64 *
-get_cqe64(volatile struct mlx5_cqe cqes[],
- unsigned int cqes_n, uint16_t *ci)
+check_cqe64_seen(volatile struct mlx5_cqe64 *cqe)
{
- volatile struct mlx5_cqe64 *cqe;
- uint16_t idx = *ci;
- uint8_t op_own;
-
- cqe = &cqes[idx & (cqes_n - 1)].cqe64;
- op_own = cqe->op_own;
- if (unlikely((op_own & MLX5_CQE_OWNER_MASK) == !(idx & cqes_n))) {
- return NULL;
- } else if (unlikely(op_own & 0x80)) {
- switch (op_own >> 4) {
- case MLX5_CQE_INVALID:
- return NULL; /* No CQE */
- case MLX5_CQE_REQ_ERR:
- return cqe;
- case MLX5_CQE_RESP_ERR:
- ++(*ci);
- return NULL;
- default:
- return NULL;
+ static const uint8_t magic[] = "seen";
+ volatile uint8_t (*buf)[sizeof(cqe->rsvd40)] = &cqe->rsvd40;
+ int ret = 1;
+ unsigned int i;
+
+ 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 */
+
+static inline int
+check_cqe64(volatile struct mlx5_cqe64 *cqe,
+ unsigned int cqes_n, const uint16_t ci)
+ __attribute__((always_inline));
+
+/**
+ * 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_cqe64(volatile struct mlx5_cqe64 *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_cqe64_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_cqe64_seen(cqe))
+ ERROR("unexpected CQE opcode %u (0x%02x)",
+ op_code, op_code);
+ return 1;
}
- if (cqe) {
- *ci = idx + 1;
- return cqe;
- }
- return NULL;
+#endif /* NDEBUG */
+ return 0;
}
+static inline void
+txq_complete(struct txq *txq) __attribute__((always_inline));
+
/**
* Manage TX completions.
*
* When sending a burst, mlx5_tx_burst() posts several WRs.
- * To improve performance, a completion event is only required once every
- * MLX5_PMD_TX_PER_COMP_REQ sends. Doing so discards completion information
- * for other WRs, but this information would not be used anyway.
*
* @param txq
* Pointer to TX queue structure.
*/
-static void
+static inline void
txq_complete(struct txq *txq)
{
const unsigned int elts_n = txq->elts_n;
const unsigned int cqe_n = 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;
- unsigned int wqe_ci = (unsigned int)-1;
- int ret = 0;
+ volatile struct mlx5_cqe64 *cqe = NULL;
+ volatile union mlx5_wqe *wqe;
- while (ret == 0) {
- volatile struct mlx5_cqe64 *cqe;
+ do {
+ volatile struct mlx5_cqe64 *tmp;
- cqe = get_cqe64(*txq->cqes, cqe_n, &cq_ci);
- if (cqe == NULL)
+ tmp = &(*txq->cqes)[cq_ci & cqe_cnt].cqe64;
+ if (check_cqe64(tmp, cqe_n, cq_ci))
break;
- wqe_ci = ntohs(cqe->wqe_counter);
- }
- if (unlikely(wqe_ci == (unsigned int)-1))
+ cqe = tmp;
+#ifndef NDEBUG
+ if (MLX5_CQE_FORMAT(cqe->op_own) == MLX5_COMPRESSED) {
+ if (!check_cqe64_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_cqe64_seen(cqe))
+ ERROR("unexpected error CQE, TX stopped");
+ return;
+ }
+#endif /* NDEBUG */
+ ++cq_ci;
+ } while (1);
+ if (unlikely(cqe == NULL))
return;
+ wqe = &(*txq->wqes)[htons(cqe->wqe_counter) & (txq->wqe_n - 1)];
+ elts_tail = wqe->wqe.ctrl.data[3];
+ assert(elts_tail < txq->wqe_n);
/* Free buffers. */
- elts_tail = (wqe_ci + 1) & (elts_n - 1);
- do {
+ while (elts_free != elts_tail) {
struct rte_mbuf *elt = (*txq->elts)[elts_free];
unsigned int elts_free_next =
(elts_free + 1) & (elts_n - 1);
/* Only one segment needs to be freed. */
rte_pktmbuf_free_seg(elt);
elts_free = elts_free_next;
- } while (elts_free != elts_tail);
+ }
txq->cq_ci = cq_ci;
txq->elts_tail = elts_tail;
/* Update the consumer index. */
* Pointer to TX queue structure.
* @param wqe
* Pointer to the WQE to fill.
- * @param addr
- * Buffer data address.
+ * @param buf
+ * Buffer.
* @param length
* Packet length.
* @param lkey
*/
static inline void
mlx5_wqe_write(struct txq *txq, volatile union mlx5_wqe *wqe,
- uintptr_t addr, uint32_t length, uint32_t lkey)
+ struct rte_mbuf *buf, uint32_t length, uint32_t lkey)
{
+ uintptr_t addr = rte_pktmbuf_mtod(buf, uintptr_t);
+
+ rte_mov16((uint8_t *)&wqe->wqe.eseg.inline_hdr_start,
+ (uint8_t *)addr);
+ addr += 16;
+ length -= 16;
+ /* Need to insert VLAN ? */
+ if (buf->ol_flags & PKT_TX_VLAN_PKT) {
+ uint32_t vlan = htonl(0x81000000 | buf->vlan_tci);
+
+ memcpy((uint8_t *)&wqe->wqe.eseg.inline_hdr_start + 12,
+ &vlan, sizeof(vlan));
+ addr -= sizeof(vlan);
+ length += sizeof(vlan);
+ }
+ /* Write the WQE. */
wqe->wqe.ctrl.data[0] = htonl((txq->wqe_ci << 8) | MLX5_OPCODE_SEND);
wqe->wqe.ctrl.data[1] = htonl((txq->qp_num_8s) | 4);
+ wqe->wqe.ctrl.data[2] = 0;
wqe->wqe.ctrl.data[3] = 0;
wqe->inl.eseg.rsvd0 = 0;
wqe->inl.eseg.rsvd1 = 0;
wqe->inl.eseg.mss = 0;
wqe->inl.eseg.rsvd2 = 0;
- wqe->wqe.eseg.inline_hdr_sz = htons(MLX5_ETH_INLINE_HEADER_SIZE);
- /* Copy the first 16 bytes into inline header. */
- rte_memcpy((uint8_t *)(uintptr_t)wqe->wqe.eseg.inline_hdr_start,
- (uint8_t *)(uintptr_t)addr,
- MLX5_ETH_INLINE_HEADER_SIZE);
- addr += MLX5_ETH_INLINE_HEADER_SIZE;
- length -= MLX5_ETH_INLINE_HEADER_SIZE;
+ wqe->wqe.eseg.inline_hdr_sz = htons(16);
/* Store remaining data in data segment. */
wqe->wqe.dseg.byte_count = htonl(length);
wqe->wqe.dseg.lkey = lkey;
}
/**
- * Write a regular WQE with VLAN.
+ * Write a inline WQE.
*
* @param txq
* Pointer to TX queue structure.
* Packet length.
* @param lkey
* Memory region lkey.
- * @param vlan_tci
- * VLAN field to insert in packet.
*/
static inline void
-mlx5_wqe_write_vlan(struct txq *txq, volatile union mlx5_wqe *wqe,
- uintptr_t addr, uint32_t length, uint32_t lkey,
- uint16_t vlan_tci)
+mlx5_wqe_write_inline(struct txq *txq, volatile union mlx5_wqe *wqe,
+ uintptr_t addr, uint32_t length)
{
- uint32_t vlan = htonl(0x81000000 | vlan_tci);
+ uint32_t size;
+ uint16_t wqe_cnt = txq->wqe_n - 1;
+ uint16_t wqe_ci = txq->wqe_ci + 1;
- wqe->wqe.ctrl.data[0] = htonl((txq->wqe_ci << 8) | MLX5_OPCODE_SEND);
- wqe->wqe.ctrl.data[1] = htonl((txq->qp_num_8s) | 4);
- wqe->wqe.ctrl.data[3] = 0;
+ /* Copy the first 16 bytes into inline header. */
+ rte_memcpy((void *)(uintptr_t)wqe->inl.eseg.inline_hdr_start,
+ (void *)(uintptr_t)addr,
+ MLX5_ETH_INLINE_HEADER_SIZE);
+ addr += MLX5_ETH_INLINE_HEADER_SIZE;
+ length -= MLX5_ETH_INLINE_HEADER_SIZE;
+ size = 3 + ((4 + length + 15) / 16);
+ wqe->inl.byte_cnt = htonl(length | MLX5_INLINE_SEG);
+ rte_memcpy((void *)(uintptr_t)&wqe->inl.data[0],
+ (void *)addr, MLX5_WQE64_INL_DATA);
+ addr += MLX5_WQE64_INL_DATA;
+ length -= MLX5_WQE64_INL_DATA;
+ while (length) {
+ volatile union mlx5_wqe *wqe_next =
+ &(*txq->wqes)[wqe_ci & wqe_cnt];
+ uint32_t copy_bytes = (length > sizeof(*wqe)) ?
+ sizeof(*wqe) :
+ length;
+
+ rte_mov64((uint8_t *)(uintptr_t)&wqe_next->data[0],
+ (uint8_t *)addr);
+ addr += copy_bytes;
+ length -= copy_bytes;
+ ++wqe_ci;
+ }
+ assert(size < 64);
+ wqe->inl.ctrl.data[0] = htonl((txq->wqe_ci << 8) | MLX5_OPCODE_SEND);
+ wqe->inl.ctrl.data[1] = htonl(txq->qp_num_8s | size);
+ wqe->inl.ctrl.data[2] = 0;
+ wqe->inl.ctrl.data[3] = 0;
wqe->inl.eseg.rsvd0 = 0;
wqe->inl.eseg.rsvd1 = 0;
wqe->inl.eseg.mss = 0;
wqe->inl.eseg.rsvd2 = 0;
- wqe->wqe.eseg.inline_hdr_sz = htons(MLX5_ETH_VLAN_INLINE_HEADER_SIZE);
+ wqe->inl.eseg.inline_hdr_sz = htons(MLX5_ETH_INLINE_HEADER_SIZE);
+ /* Increment consumer index. */
+ txq->wqe_ci = wqe_ci;
+}
+
+/**
+ * Write a inline WQE with VLAN.
+ *
+ * @param txq
+ * Pointer to TX queue structure.
+ * @param wqe
+ * Pointer to the WQE to fill.
+ * @param addr
+ * Buffer data address.
+ * @param length
+ * Packet length.
+ * @param lkey
+ * Memory region lkey.
+ * @param vlan_tci
+ * VLAN field to insert in packet.
+ */
+static inline void
+mlx5_wqe_write_inline_vlan(struct txq *txq, volatile union mlx5_wqe *wqe,
+ uintptr_t addr, uint32_t length, uint16_t vlan_tci)
+{
+ uint32_t size;
+ uint32_t wqe_cnt = txq->wqe_n - 1;
+ uint16_t wqe_ci = txq->wqe_ci + 1;
+ uint32_t vlan = htonl(0x81000000 | vlan_tci);
+
/*
* Copy 12 bytes of source & destination MAC address.
* Copy 4 bytes of VLAN.
* Copy 2 bytes of Ether type.
*/
- rte_memcpy((uint8_t *)(uintptr_t)wqe->wqe.eseg.inline_hdr_start,
- (uint8_t *)(uintptr_t)addr, 12);
- rte_memcpy((uint8_t *)((uintptr_t)wqe->wqe.eseg.inline_hdr_start + 12),
+ rte_memcpy((uint8_t *)(uintptr_t)wqe->inl.eseg.inline_hdr_start,
+ (uint8_t *)addr, 12);
+ rte_memcpy((uint8_t *)(uintptr_t)wqe->inl.eseg.inline_hdr_start + 12,
&vlan, sizeof(vlan));
- rte_memcpy((uint8_t *)((uintptr_t)wqe->wqe.eseg.inline_hdr_start + 16),
- (uint8_t *)((uintptr_t)addr + 12), 2);
+ rte_memcpy((uint8_t *)((uintptr_t)wqe->inl.eseg.inline_hdr_start + 16),
+ (uint8_t *)(addr + 12), 2);
addr += MLX5_ETH_VLAN_INLINE_HEADER_SIZE - sizeof(vlan);
length -= MLX5_ETH_VLAN_INLINE_HEADER_SIZE - sizeof(vlan);
- /* Store remaining data in data segment. */
- wqe->wqe.dseg.byte_count = htonl(length);
- wqe->wqe.dseg.lkey = lkey;
- wqe->wqe.dseg.addr = htonll(addr);
+ size = (sizeof(wqe->inl.ctrl.ctrl) +
+ sizeof(wqe->inl.eseg) +
+ sizeof(wqe->inl.byte_cnt) +
+ length + 15) / 16;
+ wqe->inl.byte_cnt = htonl(length | MLX5_INLINE_SEG);
+ rte_memcpy((void *)(uintptr_t)&wqe->inl.data[0],
+ (void *)addr, MLX5_WQE64_INL_DATA);
+ addr += MLX5_WQE64_INL_DATA;
+ length -= MLX5_WQE64_INL_DATA;
+ while (length) {
+ volatile union mlx5_wqe *wqe_next =
+ &(*txq->wqes)[wqe_ci & wqe_cnt];
+ uint32_t copy_bytes = (length > sizeof(*wqe)) ?
+ sizeof(*wqe) :
+ length;
+
+ rte_mov64((uint8_t *)(uintptr_t)&wqe_next->data[0],
+ (uint8_t *)addr);
+ addr += copy_bytes;
+ length -= copy_bytes;
+ ++wqe_ci;
+ }
+ assert(size < 64);
+ wqe->inl.ctrl.data[0] = htonl((txq->wqe_ci << 8) | MLX5_OPCODE_SEND);
+ wqe->inl.ctrl.data[1] = htonl(txq->qp_num_8s | size);
+ wqe->inl.ctrl.data[2] = 0;
+ wqe->inl.ctrl.data[3] = 0;
+ wqe->inl.eseg.rsvd0 = 0;
+ wqe->inl.eseg.rsvd1 = 0;
+ wqe->inl.eseg.mss = 0;
+ wqe->inl.eseg.rsvd2 = 0;
+ wqe->inl.eseg.inline_hdr_sz = htons(MLX5_ETH_VLAN_INLINE_HEADER_SIZE);
/* Increment consumer index. */
- ++txq->wqe_ci;
+ txq->wqe_ci = wqe_ci;
}
/**
rte_prefetch0(cqe);
}
+/**
+ * Prefetch a WQE.
+ *
+ * @param txq
+ * Pointer to TX queue structure.
+ * @param wqe_ci
+ * WQE consumer index.
+ */
+static inline void
+tx_prefetch_wqe(struct txq *txq, uint16_t ci)
+{
+ volatile union mlx5_wqe *wqe;
+
+ wqe = &(*txq->wqes)[ci & (txq->wqe_n - 1)];
+ rte_prefetch0(wqe);
+}
+
/**
* DPDK callback for TX.
*
struct txq *txq = (struct txq *)dpdk_txq;
uint16_t elts_head = txq->elts_head;
const unsigned int elts_n = txq->elts_n;
- unsigned int i;
+ unsigned int i = 0;
+ unsigned int j = 0;
unsigned int max;
- volatile union mlx5_wqe *wqe;
- struct rte_mbuf *buf;
+ unsigned int comp;
+ volatile union mlx5_wqe *wqe = NULL;
if (unlikely(!pkts_n))
return 0;
- buf = pkts[0];
/* Prefetch first packet cacheline. */
tx_prefetch_cqe(txq, txq->cq_ci);
tx_prefetch_cqe(txq, txq->cq_ci + 1);
- rte_prefetch0(buf);
+ rte_prefetch0(*pkts);
/* Start processing. */
txq_complete(txq);
max = (elts_n - (elts_head - txq->elts_tail));
if (max > elts_n)
max -= elts_n;
- assert(max >= 1);
- assert(max <= elts_n);
- /* Always leave one free entry in the ring. */
- --max;
- if (max == 0)
- return 0;
- if (max > pkts_n)
- max = pkts_n;
- for (i = 0; (i != max); ++i) {
- unsigned int elts_head_next = (elts_head + 1) & (elts_n - 1);
- uintptr_t addr;
+ do {
+ struct rte_mbuf *buf = *(pkts++);
+ unsigned int elts_head_next;
uint32_t length;
uint32_t lkey;
+ unsigned int segs_n = buf->nb_segs;
+ volatile struct mlx5_wqe_data_seg *dseg;
+ unsigned int ds = sizeof(*wqe) / 16;
+ /*
+ * 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;
+ --pkts_n;
+ elts_head_next = (elts_head + 1) & (elts_n - 1);
wqe = &(*txq->wqes)[txq->wqe_ci & (txq->wqe_n - 1)];
+ dseg = &wqe->wqe.dseg;
rte_prefetch0(wqe);
- if (i + 1 < max)
- rte_prefetch0(pkts[i + 1]);
- /* Retrieve buffer information. */
- addr = rte_pktmbuf_mtod(buf, uintptr_t);
+ if (pkts_n)
+ rte_prefetch0(*pkts);
length = DATA_LEN(buf);
/* Update element. */
(*txq->elts)[elts_head] = buf;
/* Prefetch next buffer data. */
- if (i + 1 < max)
- rte_prefetch0(rte_pktmbuf_mtod(pkts[i + 1],
+ if (pkts_n)
+ rte_prefetch0(rte_pktmbuf_mtod(*pkts,
volatile void *));
/* Retrieve Memory Region key for this memory pool. */
lkey = txq_mp2mr(txq, txq_mb2mp(buf));
- if (buf->ol_flags & PKT_TX_VLAN_PKT)
- mlx5_wqe_write_vlan(txq, wqe, addr, length, lkey,
- buf->vlan_tci);
- else
- mlx5_wqe_write(txq, wqe, addr, length, lkey);
- /* Request completion if needed. */
- if (unlikely(--txq->elts_comp == 0)) {
- wqe->wqe.ctrl.data[2] = htonl(8);
- txq->elts_comp = txq->elts_comp_cd_init;
- } else {
- wqe->wqe.ctrl.data[2] = 0;
- }
+ mlx5_wqe_write(txq, wqe, buf, length, lkey);
/* Should we enable HW CKSUM offload */
if (buf->ol_flags &
(PKT_TX_IP_CKSUM | PKT_TX_TCP_CKSUM | PKT_TX_UDP_CKSUM)) {
} else {
wqe->wqe.eseg.cs_flags = 0;
}
+ while (--segs_n) {
+ /*
+ * 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(!(sizeof(*wqe) % sizeof(*dseg)));
+ if (!(ds % (sizeof(*wqe) / 16)))
+ dseg = (volatile void *)
+ &(*txq->wqes)[txq->wqe_ci++ &
+ (txq->wqe_n - 1)];
+ else
+ ++dseg;
+ ++ds;
+ buf = buf->next;
+ assert(buf);
+ /* Store segment information. */
+ dseg->byte_count = htonl(DATA_LEN(buf));
+ dseg->lkey = txq_mp2mr(txq, txq_mb2mp(buf));
+ dseg->addr = htonll(rte_pktmbuf_mtod(buf, uintptr_t));
+ (*txq->elts)[elts_head_next] = buf;
+ elts_head_next = (elts_head_next + 1) & (elts_n - 1);
+#ifdef MLX5_PMD_SOFT_COUNTERS
+ length += DATA_LEN(buf);
+#endif
+ ++j;
+ }
+ /* Update DS field in WQE. */
+ wqe->wqe.ctrl.data[1] &= htonl(0xffffffc0);
+ wqe->wqe.ctrl.data[1] |= htonl(ds & 0x3f);
+ elts_head = elts_head_next;
#ifdef MLX5_PMD_SOFT_COUNTERS
/* Increment sent bytes counter. */
txq->stats.obytes += length;
#endif
elts_head = elts_head_next;
- buf = pkts[i + 1];
+ ++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. */
+ comp = txq->elts_comp + i + j;
+ if (comp >= MLX5_TX_COMP_THRESH) {
+ /* Request completion on last WQE. */
+ wqe->wqe.ctrl.data[2] = htonl(8);
+ /* Save elts_head in unused "immediate" field of WQE. */
+ wqe->wqe.ctrl.data[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. */
+ mlx5_tx_dbrec(txq);
+ txq->elts_head = elts_head;
+ return i;
+}
+
+/**
+ * DPDK callback for TX with 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.
+ *
+ * @return
+ * Number of packets successfully transmitted (<= pkts_n).
+ */
+uint16_t
+mlx5_tx_burst_inline(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 unsigned int elts_n = txq->elts_n;
+ unsigned int i = 0;
+ unsigned int j = 0;
+ unsigned int max;
+ unsigned int comp;
+ volatile union mlx5_wqe *wqe = NULL;
+ unsigned int max_inline = txq->max_inline;
+
+ 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;
+ do {
+ struct rte_mbuf *buf = *(pkts++);
+ unsigned int elts_head_next;
+ uintptr_t addr;
+ uint32_t length;
+ uint32_t lkey;
+ unsigned int segs_n = buf->nb_segs;
+ volatile struct mlx5_wqe_data_seg *dseg;
+ unsigned int ds = sizeof(*wqe) / 16;
+
+ /*
+ * 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;
+ --pkts_n;
+ elts_head_next = (elts_head + 1) & (elts_n - 1);
+ wqe = &(*txq->wqes)[txq->wqe_ci & (txq->wqe_n - 1)];
+ dseg = &wqe->wqe.dseg;
+ tx_prefetch_wqe(txq, txq->wqe_ci);
+ tx_prefetch_wqe(txq, txq->wqe_ci + 1);
+ if (pkts_n)
+ rte_prefetch0(*pkts);
+ /* Should we enable HW CKSUM offload */
+ if (buf->ol_flags &
+ (PKT_TX_IP_CKSUM | PKT_TX_TCP_CKSUM | PKT_TX_UDP_CKSUM)) {
+ wqe->inl.eseg.cs_flags =
+ MLX5_ETH_WQE_L3_CSUM |
+ MLX5_ETH_WQE_L4_CSUM;
+ } else {
+ wqe->inl.eseg.cs_flags = 0;
+ }
+ /* Retrieve buffer information. */
+ addr = rte_pktmbuf_mtod(buf, uintptr_t);
+ length = DATA_LEN(buf);
+ /* Update element. */
+ (*txq->elts)[elts_head] = buf;
+ /* Prefetch next buffer data. */
+ if (pkts_n)
+ rte_prefetch0(rte_pktmbuf_mtod(*pkts,
+ volatile void *));
+ if ((length <= max_inline) && (segs_n == 1)) {
+ if (buf->ol_flags & PKT_TX_VLAN_PKT)
+ mlx5_wqe_write_inline_vlan(txq, wqe,
+ addr, length,
+ buf->vlan_tci);
+ else
+ mlx5_wqe_write_inline(txq, wqe, addr, length);
+ goto skip_segs;
+ } else {
+ /* Retrieve Memory Region key for this memory pool. */
+ lkey = txq_mp2mr(txq, txq_mb2mp(buf));
+ mlx5_wqe_write(txq, wqe, buf, length, lkey);
+ }
+ while (--segs_n) {
+ /*
+ * 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(!(sizeof(*wqe) % sizeof(*dseg)));
+ if (!(ds % (sizeof(*wqe) / 16)))
+ dseg = (volatile void *)
+ &(*txq->wqes)[txq->wqe_ci++ &
+ (txq->wqe_n - 1)];
+ else
+ ++dseg;
+ ++ds;
+ buf = buf->next;
+ assert(buf);
+ /* Store segment information. */
+ dseg->byte_count = htonl(DATA_LEN(buf));
+ dseg->lkey = txq_mp2mr(txq, txq_mb2mp(buf));
+ dseg->addr = htonll(rte_pktmbuf_mtod(buf, uintptr_t));
+ (*txq->elts)[elts_head_next] = buf;
+ elts_head_next = (elts_head_next + 1) & (elts_n - 1);
+#ifdef MLX5_PMD_SOFT_COUNTERS
+ length += DATA_LEN(buf);
+#endif
+ ++j;
+ }
+ /* Update DS field in WQE. */
+ wqe->inl.ctrl.data[1] &= htonl(0xffffffc0);
+ wqe->inl.ctrl.data[1] |= htonl(ds & 0x3f);
+skip_segs:
+ 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. */
+ comp = txq->elts_comp + i + j;
+ if (comp >= MLX5_TX_COMP_THRESH) {
+ /* Request completion on last WQE. */
+ wqe->inl.ctrl.data[2] = htonl(8);
+ /* Save elts_head in unused "immediate" field of WQE. */
+ wqe->inl.ctrl.data[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;
return i;
}
+/**
+ * Open a MPW session.
+ *
+ * @param txq
+ * Pointer to TX queue structure.
+ * @param mpw
+ * Pointer to MPW session structure.
+ * @param length
+ * Packet length.
+ */
+static inline void
+mlx5_mpw_new(struct txq *txq, struct mlx5_mpw *mpw, uint32_t length)
+{
+ uint16_t idx = txq->wqe_ci & (txq->wqe_n - 1);
+ volatile struct mlx5_wqe_data_seg (*dseg)[MLX5_MPW_DSEG_MAX] =
+ (volatile struct mlx5_wqe_data_seg (*)[])
+ (uintptr_t)&(*txq->wqes)[(idx + 1) & (txq->wqe_n - 1)];
+
+ mpw->state = MLX5_MPW_STATE_OPENED;
+ mpw->pkts_n = 0;
+ mpw->len = length;
+ mpw->total_len = 0;
+ mpw->wqe = &(*txq->wqes)[idx];
+ mpw->wqe->mpw.eseg.mss = htons(length);
+ mpw->wqe->mpw.eseg.inline_hdr_sz = 0;
+ mpw->wqe->mpw.eseg.rsvd0 = 0;
+ mpw->wqe->mpw.eseg.rsvd1 = 0;
+ mpw->wqe->mpw.eseg.rsvd2 = 0;
+ mpw->wqe->mpw.ctrl.data[0] = htonl((MLX5_OPC_MOD_MPW << 24) |
+ (txq->wqe_ci << 8) |
+ MLX5_OPCODE_LSO_MPW);
+ mpw->wqe->mpw.ctrl.data[2] = 0;
+ mpw->wqe->mpw.ctrl.data[3] = 0;
+ mpw->data.dseg[0] = &mpw->wqe->mpw.dseg[0];
+ mpw->data.dseg[1] = &mpw->wqe->mpw.dseg[1];
+ mpw->data.dseg[2] = &(*dseg)[0];
+ mpw->data.dseg[3] = &(*dseg)[1];
+ mpw->data.dseg[4] = &(*dseg)[2];
+}
+
+/**
+ * Close a MPW session.
+ *
+ * @param txq
+ * Pointer to TX queue structure.
+ * @param mpw
+ * Pointer to MPW session structure.
+ */
+static inline void
+mlx5_mpw_close(struct txq *txq, struct mlx5_mpw *mpw)
+{
+ unsigned int num = mpw->pkts_n;
+
+ /*
+ * Store size in multiple of 16 bytes. Control and Ethernet segments
+ * count as 2.
+ */
+ mpw->wqe->mpw.ctrl.data[1] = htonl(txq->qp_num_8s | (2 + num));
+ mpw->state = MLX5_MPW_STATE_CLOSED;
+ if (num < 3)
+ ++txq->wqe_ci;
+ else
+ txq->wqe_ci += 2;
+ tx_prefetch_wqe(txq, txq->wqe_ci);
+ tx_prefetch_wqe(txq, txq->wqe_ci + 1);
+}
+
+/**
+ * DPDK callback for TX with 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_mpw(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 unsigned int elts_n = txq->elts_n;
+ unsigned int i = 0;
+ unsigned int j = 0;
+ unsigned int max;
+ 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);
+ tx_prefetch_wqe(txq, txq->wqe_ci);
+ tx_prefetch_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;
+ 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 < 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->mpw.eseg.cs_flags != cs_flags)))
+ mlx5_mpw_close(txq, &mpw);
+ if (mpw.state == MLX5_MPW_STATE_CLOSED) {
+ mlx5_mpw_new(txq, &mpw, length);
+ mpw.wqe->mpw.eseg.cs_flags = cs_flags;
+ }
+ /* 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 union mlx5_wqe *wqe = mpw.wqe;
+
+ /* Request completion on last WQE. */
+ wqe->mpw.ctrl.data[2] = htonl(8);
+ /* Save elts_head in unused "immediate" field of WQE. */
+ wqe->mpw.ctrl.data[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);
+ txq->elts_head = elts_head;
+ return i;
+}
+
+/**
+ * Open a MPW inline session.
+ *
+ * @param txq
+ * Pointer to TX queue structure.
+ * @param mpw
+ * Pointer to MPW session structure.
+ * @param length
+ * Packet length.
+ */
+static inline void
+mlx5_mpw_inline_new(struct txq *txq, struct mlx5_mpw *mpw, uint32_t length)
+{
+ uint16_t idx = txq->wqe_ci & (txq->wqe_n - 1);
+
+ mpw->state = MLX5_MPW_INL_STATE_OPENED;
+ mpw->pkts_n = 0;
+ mpw->len = length;
+ mpw->total_len = 0;
+ mpw->wqe = &(*txq->wqes)[idx];
+ mpw->wqe->mpw_inl.ctrl.data[0] = htonl((MLX5_OPC_MOD_MPW << 24) |
+ (txq->wqe_ci << 8) |
+ MLX5_OPCODE_LSO_MPW);
+ mpw->wqe->mpw_inl.ctrl.data[2] = 0;
+ mpw->wqe->mpw_inl.ctrl.data[3] = 0;
+ mpw->wqe->mpw_inl.eseg.mss = htons(length);
+ mpw->wqe->mpw_inl.eseg.inline_hdr_sz = 0;
+ mpw->wqe->mpw_inl.eseg.cs_flags = 0;
+ mpw->wqe->mpw_inl.eseg.rsvd0 = 0;
+ mpw->wqe->mpw_inl.eseg.rsvd1 = 0;
+ mpw->wqe->mpw_inl.eseg.rsvd2 = 0;
+ mpw->data.raw = &mpw->wqe->mpw_inl.data[0];
+}
+
+/**
+ * Close a MPW inline session.
+ *
+ * @param txq
+ * Pointer to TX queue structure.
+ * @param mpw
+ * Pointer to MPW session structure.
+ */
+static inline void
+mlx5_mpw_inline_close(struct txq *txq, struct mlx5_mpw *mpw)
+{
+ unsigned int size;
+
+ size = sizeof(*mpw->wqe) - MLX5_MWQE64_INL_DATA + mpw->total_len;
+ /*
+ * Store size in multiple of 16 bytes. Control and Ethernet segments
+ * count as 2.
+ */
+ mpw->wqe->mpw_inl.ctrl.data[1] =
+ htonl(txq->qp_num_8s | ((size + 15) / 16));
+ mpw->state = MLX5_MPW_STATE_CLOSED;
+ mpw->wqe->mpw_inl.byte_cnt = htonl(mpw->total_len | MLX5_INLINE_SEG);
+ txq->wqe_ci += (size + (sizeof(*mpw->wqe) - 1)) / sizeof(*mpw->wqe);
+}
+
+/**
+ * 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.
+ *
+ * @return
+ * Number of packets successfully transmitted (<= pkts_n).
+ */
+uint16_t
+mlx5_tx_burst_mpw_inline(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 unsigned int elts_n = txq->elts_n;
+ unsigned int i = 0;
+ unsigned int j = 0;
+ unsigned int max;
+ unsigned int comp;
+ unsigned int inline_room = txq->max_inline;
+ 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);
+ tx_prefetch_wqe(txq, txq->wqe_ci);
+ tx_prefetch_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;
+ uintptr_t addr;
+ 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 < 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);
+ /* Start new session if packet differs. */
+ if (mpw.state == MLX5_MPW_STATE_OPENED) {
+ if ((mpw.len != length) ||
+ (segs_n != 1) ||
+ (mpw.wqe->mpw.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->mpw_inl.eseg.cs_flags != cs_flags)) {
+ mlx5_mpw_inline_close(txq, &mpw);
+ inline_room = txq->max_inline;
+ }
+ }
+ if (mpw.state == MLX5_MPW_STATE_CLOSED) {
+ if ((segs_n != 1) ||
+ (length > inline_room)) {
+ mlx5_mpw_new(txq, &mpw, length);
+ mpw.wqe->mpw.eseg.cs_flags = cs_flags;
+ } else {
+ mlx5_mpw_inline_new(txq, &mpw, length);
+ mpw.wqe->mpw_inl.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);
+#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);
+ } else {
+ unsigned int max;
+
+ 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;
+ /* Maximum number of bytes before wrapping. */
+ max = ((uintptr_t)&(*txq->wqes)[txq->wqe_n] -
+ (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)[0];
+ 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);
+ mpw.data.raw += length;
+ }
+ if ((uintptr_t)mpw.data.raw ==
+ (uintptr_t)&(*txq->wqes)[txq->wqe_n])
+ mpw.data.raw =
+ (volatile void *)&(*txq->wqes)[0];
+ ++mpw.pkts_n;
+ ++j;
+ if (mpw.pkts_n == MLX5_MPW_DSEG_MAX) {
+ mlx5_mpw_inline_close(txq, &mpw);
+ inline_room = txq->max_inline;
+ } else {
+ 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;
+#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 union mlx5_wqe *wqe = mpw.wqe;
+
+ /* Request completion on last WQE. */
+ wqe->mpw_inl.ctrl.data[2] = htonl(8);
+ /* Save elts_head in unused "immediate" field of WQE. */
+ wqe->mpw_inl.ctrl.data[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_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);
+ txq->elts_head = elts_head;
+ return i;
+}
+
/**
* Translate RX completion flags to packet type.
*
return pkt_type;
}
+/**
+ * Get size of the next packet for a given CQE. For compressed CQEs, the
+ * consumer index is updated only once all packets of the current one have
+ * been processed.
+ *
+ * @param rxq
+ * Pointer to RX queue.
+ * @param cqe
+ * CQE to process.
+ *
+ * @return
+ * Packet size in bytes (0 if there is none), -1 in case of completion
+ * with error.
+ */
+static inline int
+mlx5_rx_poll_len(struct rxq *rxq, volatile struct mlx5_cqe64 *cqe,
+ uint16_t cqe_cnt)
+{
+ struct rxq_zip *zip = &rxq->zip;
+ uint16_t cqe_n = cqe_cnt + 1;
+ int 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)[zip->ca & cqe_cnt].cqe64);
+
+ len = ntohl((*mc)[zip->ai & 7].byte_cnt);
+ if ((++zip->ai & 7) == 0) {
+ /*
+ * 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)) {
+ uint16_t idx = rxq->cq_ci;
+ uint16_t end = zip->cq_ci;
+
+ while (idx != end) {
+ (*rxq->cqes)[idx & cqe_cnt].cqe64.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_cqe64(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) {
+ volatile struct mlx5_mini_cqe8 (*mc)[8] =
+ (volatile struct mlx5_mini_cqe8 (*)[8])
+ (uintptr_t)(&(*rxq->cqes)[rxq->cq_ci &
+ cqe_cnt].cqe64);
+
+ /* 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 & cqe_cnt;
+ 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);
+ zip->ai = 1;
+ } else {
+ len = ntohl(cqe->byte_cnt);
+ }
+ /* Error while receiving packet. */
+ if (unlikely(MLX5_CQE_OPCODE(op_own) == MLX5_CQE_RESP_ERR))
+ return -1;
+ }
+ return len;
+}
+
/**
* Translate RX completion flags to offload flags.
*
return ol_flags;
}
-/**
- * Get size of the next packet.
- *
- * @param rxq
- * RX queue to fetch packet from.
- *
- * @return
- * Packet size in bytes.
- */
-static inline int __attribute__((always_inline))
-rx_poll_len(struct rxq *rxq)
-{
- volatile struct mlx5_cqe64 *cqe;
-
- cqe = get_cqe64(*rxq->cqes, rxq->elts_n, &rxq->cq_ci);
- if (cqe)
- return ntohl(cqe->byte_cnt);
- return 0;
-}
-
/**
* DPDK callback for RX.
*
mlx5_rx_burst(void *dpdk_rxq, struct rte_mbuf **pkts, uint16_t pkts_n)
{
struct rxq *rxq = dpdk_rxq;
- unsigned int pkts_ret = 0;
- unsigned int i;
- unsigned int rq_ci = rxq->rq_ci;
- const unsigned int elts_n = rxq->elts_n;
- const unsigned int wqe_cnt = elts_n - 1;
+ const unsigned int wqe_cnt = rxq->elts_n - 1;
+ const unsigned int cqe_cnt = rxq->cqe_n - 1;
+ const unsigned int sges_n = rxq->sges_n;
+ struct rte_mbuf *pkt = NULL;
+ struct rte_mbuf *seg = NULL;
+ volatile struct mlx5_cqe64 *cqe =
+ &(*rxq->cqes)[rxq->cq_ci & cqe_cnt].cqe64;
+ unsigned int i = 0;
+ unsigned int rq_ci = rxq->rq_ci << sges_n;
+ int len;
- for (i = 0; (i != pkts_n); ++i) {
+ while (pkts_n) {
unsigned int idx = rq_ci & wqe_cnt;
- struct rte_mbuf *rep;
- struct rte_mbuf *pkt;
- unsigned int len;
volatile struct mlx5_wqe_data_seg *wqe = &(*rxq->wqes)[idx];
- volatile struct mlx5_cqe64 *cqe =
- &(*rxq->cqes)[rxq->cq_ci & wqe_cnt].cqe64;
+ struct rte_mbuf *rep = (*rxq->elts)[idx];
- pkt = (*rxq->elts)[idx];
+ if (pkt)
+ NEXT(seg) = rep;
+ seg = rep;
+ rte_prefetch0(seg);
rte_prefetch0(cqe);
+ rte_prefetch0(wqe);
rep = rte_mbuf_raw_alloc(rxq->mp);
if (unlikely(rep == NULL)) {
++rxq->stats.rx_nombuf;
+ if (!pkt) {
+ /*
+ * no buffers before we even started,
+ * bail out silently.
+ */
+ break;
+ }
+ while (pkt != seg) {
+ assert(pkt != (*rxq->elts)[idx]);
+ seg = NEXT(pkt);
+ rte_mbuf_refcnt_set(pkt, 0);
+ __rte_mbuf_raw_free(pkt);
+ pkt = seg;
+ }
break;
}
- SET_DATA_OFF(rep, RTE_PKTMBUF_HEADROOM);
- NB_SEGS(rep) = 1;
- PORT(rep) = rxq->port_id;
- NEXT(rep) = NULL;
- len = rx_poll_len(rxq);
- if (unlikely(len == 0)) {
- rte_mbuf_refcnt_set(rep, 0);
- __rte_mbuf_raw_free(rep);
- break;
+ if (!pkt) {
+ cqe = &(*rxq->cqes)[rxq->cq_ci & cqe_cnt].cqe64;
+ len = mlx5_rx_poll_len(rxq, cqe, cqe_cnt);
+ if (len == 0) {
+ rte_mbuf_refcnt_set(rep, 0);
+ __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 (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);
+ }
+ 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
* of the buffers are already known, only the buffer address
* changes.
*/
- wqe->addr = htonll((uintptr_t)rep->buf_addr +
- RTE_PKTMBUF_HEADROOM);
- (*rxq->elts)[idx] = rep;
- /* Update pkt information. */
- 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);
- }
- if (rxq->crc_present)
- len -= ETHER_CRC_LEN;
+ wqe->addr = htonll(rte_pktmbuf_mtod(rep, uintptr_t));
+ if (len > DATA_LEN(seg)) {
+ len -= DATA_LEN(seg);
+ ++NB_SEGS(pkt);
+ ++rq_ci;
+ continue;
}
- PKT_LEN(pkt) = len;
- DATA_LEN(pkt) = len;
+ DATA_LEN(seg) = len;
#ifdef MLX5_PMD_SOFT_COUNTERS
/* Increment bytes counter. */
- rxq->stats.ibytes += len;
+ rxq->stats.ibytes += PKT_LEN(pkt);
#endif
/* Return packet. */
*(pkts++) = pkt;
- ++pkts_ret;
+ pkt = NULL;
+ --pkts_n;
+ ++i;
+skip:
+ /* Align consumer index to the next stride. */
+ rq_ci >>= sges_n;
++rq_ci;
+ rq_ci <<= sges_n;
}
- if (unlikely((i == 0) && (rq_ci == rxq->rq_ci)))
+ if (unlikely((i == 0) && ((rq_ci >> sges_n) == rxq->rq_ci)))
return 0;
- /* Repost WRs. */
-#ifdef DEBUG_RECV
- DEBUG("%p: reposting %u WRs", (void *)rxq, i);
-#endif
/* Update the consumer index. */
- rxq->rq_ci = rq_ci;
+ 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);
#ifdef MLX5_PMD_SOFT_COUNTERS
/* Increment packets counter. */
- rxq->stats.ipackets += pkts_ret;
+ rxq->stats.ipackets += i;
#endif
- return pkts_ret;
+ return i;
}
/**
git.droids-corp.org / dpdk.git / blobdiff