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);
+ DRV_LOG(WARNING, "cannot open %s for debug dump", path);
MKSTR(path2, "./%s", fname);
fd = fopen(path2, "a+");
if (!fd) {
- DRV_LOG(ERR, "cannot open %s for debug dump\n",
- path2);
+ DRV_LOG(ERR, "cannot open %s for debug dump", path2);
return;
}
- DRV_LOG(INFO, "New debug dump in file %s\n", path2);
+ DRV_LOG(INFO, "New debug dump in file %s", path2);
} else {
- DRV_LOG(INFO, "New debug dump in file %s\n", path);
+ DRV_LOG(INFO, "New debug dump in file %s", path);
}
if (hex_title)
rte_hexdump(fd, hex_title, buf, hex_len);
* Pointer to the error CQE.
*
* @return
- * The last Tx buffer element to free.
+ * Negative value if queue recovery failed,
+ * the last Tx buffer element to free otherwise.
*/
-uint16_t
+int
mlx5_tx_error_cqe_handle(struct mlx5_txq_data *restrict txq,
volatile struct mlx5_err_cqe *err_cqe)
{
return txq->elts_head;
}
/* Recovering failed - try again later on the same WQE. */
+ return -1;
} else {
txq->cq_ci++;
}
&rq_attr);
}
if (ret) {
- DRV_LOG(ERR, "Cannot change Rx WQ state to %u - %s\n",
+ DRV_LOG(ERR, "Cannot change Rx WQ state to %u - %s",
sm->state, strerror(errno));
rte_errno = errno;
return ret;
.qp_state = IBV_QPS_RESET,
.port_num = (uint8_t)priv->ibv_port,
};
- struct ibv_qp *qp = txq_ctrl->ibv->qp;
+ struct ibv_qp *qp = txq_ctrl->obj->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));
+ "%s", strerror(errno));
rte_errno = errno;
return ret;
}
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",
+ DRV_LOG(ERR, "Cannot change Tx QP state to INIT %s",
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",
+ DRV_LOG(ERR, "Cannot change Tx QP state to RTR %s",
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",
+ DRV_LOG(ERR, "Cannot change Tx QP state to RTS %s",
strerror(errno));
rte_errno = errno;
return ret;
*
* @param[in] rxq
* Pointer to RX queue structure.
- * @param[in] mbuf_prepare
- * Whether to prepare mbufs for the RQ.
+ * @param[in] vec
+ * 1 when called from vectorized Rx burst, need to prepare mbufs for the RQ.
+ * 0 when called from non-vectorized Rx burst.
*
* @return
* -1 in case of recovery error, otherwise the CQE status.
*/
int
-mlx5_rx_err_handle(struct mlx5_rxq_data *rxq, uint8_t mbuf_prepare)
+mlx5_rx_err_handle(struct mlx5_rxq_data *rxq, uint8_t vec)
{
const uint16_t cqe_n = 1 << rxq->cqe_n;
const uint16_t cqe_mask = cqe_n - 1;
if (mlx5_queue_state_modify(ETH_DEV(rxq_ctrl->priv),
&sm))
return -1;
- if (mbuf_prepare) {
+ if (vec) {
const uint16_t q_mask = wqe_n - 1;
uint16_t elt_idx;
struct rte_mbuf **elt;
return -1;
}
}
+ for (i = 0; i < (int)wqe_n; ++i) {
+ elt = &(*rxq->elts)[i];
+ DATA_LEN(*elt) =
+ (uint16_t)((*elt)->buf_len -
+ rte_pktmbuf_headroom(*elt));
+ }
+ /* Padding with a fake mbuf for vec Rx. */
+ for (i = 0; i < MLX5_VPMD_DESCS_PER_LOOP; ++i)
+ (*rxq->elts)[wqe_n + i] =
+ &rxq->fake_mbuf;
}
mlx5_rxq_initialize(rxq);
rxq->err_state = MLX5_RXQ_ERR_STATE_NO_ERROR;
(pkts_n - part) * sizeof(struct rte_mbuf *));
}
+/**
+ * Update completion queue consuming index via doorbell
+ * and flush the completed data buffers.
+ *
+ * @param txq
+ * Pointer to TX queue structure.
+ * @param valid CQE pointer
+ * if not NULL update txq->wqe_pi and flush the buffers
+ * @param itail
+ * if not negative - flush the buffers till this index.
+ * @param olx
+ * Configured Tx offloads mask. It is fully defined at
+ * compile time and may be used for optimization.
+ */
+static __rte_always_inline void
+mlx5_tx_comp_flush(struct mlx5_txq_data *restrict txq,
+ volatile struct mlx5_cqe *last_cqe,
+ int itail,
+ unsigned int olx __rte_unused)
+{
+ uint16_t tail;
+
+ if (likely(last_cqe != NULL)) {
+ txq->wqe_pi = rte_be_to_cpu_16(last_cqe->wqe_counter);
+ tail = ((volatile struct mlx5_wqe_cseg *)
+ (txq->wqes + (txq->wqe_pi & txq->wqe_m)))->misc;
+ } else if (itail >= 0) {
+ tail = (uint16_t)itail;
+ } else {
+ return;
+ }
+ rte_compiler_barrier();
+ *txq->cq_db = rte_cpu_to_be_32(txq->cq_ci);
+ if (likely(tail != txq->elts_tail)) {
+ mlx5_tx_free_elts(txq, tail, olx);
+ assert(tail == txq->elts_tail);
+ }
+}
+
/**
* Manage TX completions. This routine checks the CQ for
* arrived CQEs, deduces the last accomplished WQE in SQ,
unsigned int olx __rte_unused)
{
unsigned int count = MLX5_TX_COMP_MAX_CQE;
- bool update = false;
- uint16_t tail = txq->elts_tail;
+ volatile struct mlx5_cqe *last_cqe = NULL;
int ret;
+ static_assert(MLX5_CQE_STATUS_HW_OWN < 0, "Must be negative value");
+ static_assert(MLX5_CQE_STATUS_SW_OWN < 0, "Must be negative value");
do {
volatile struct mlx5_cqe *cqe;
assert(ret == MLX5_CQE_STATUS_HW_OWN);
break;
}
- /* Some error occurred, try to restart. */
+ /*
+ * Some error occurred, try to restart.
+ * We have no barrier after WQE related Doorbell
+ * written, make sure all writes are completed
+ * here, before we might perform SQ reset.
+ */
rte_wmb();
- tail = mlx5_tx_error_cqe_handle
+ ret = mlx5_tx_error_cqe_handle
(txq, (volatile struct mlx5_err_cqe *)cqe);
- if (likely(tail != txq->elts_tail)) {
- mlx5_tx_free_elts(txq, tail, olx);
- assert(tail == txq->elts_tail);
- }
- /* Allow flushing all CQEs from the queue. */
- count = txq->cqe_s;
- } else {
- volatile struct mlx5_wqe_cseg *cseg;
-
- /* Normal transmit completion. */
- ++txq->cq_ci;
- rte_cio_rmb();
- txq->wqe_pi = rte_be_to_cpu_16(cqe->wqe_counter);
- cseg = (volatile struct mlx5_wqe_cseg *)
- (txq->wqes + (txq->wqe_pi & txq->wqe_m));
- tail = cseg->misc;
+ /*
+ * Flush buffers, update consuming index
+ * if recovery succeeded. Otherwise
+ * just try to recover later.
+ */
+ last_cqe = NULL;
+ break;
}
+ /* Normal transmit completion. */
+ ++txq->cq_ci;
+ last_cqe = cqe;
#ifndef NDEBUG
if (txq->cq_pi)
--txq->cq_pi;
#endif
- update = true;
/*
* We have to restrict the amount of processed CQEs
* in one tx_burst routine call. The CQ may be large
* latency.
*/
} while (--count);
- if (likely(tail != txq->elts_tail)) {
- /* Free data buffers from elts. */
- mlx5_tx_free_elts(txq, tail, olx);
- assert(tail == txq->elts_tail);
- }
- if (likely(update)) {
- /* Update the consumer index. */
- rte_compiler_barrier();
- *txq->cq_db =
- rte_cpu_to_be_32(txq->cq_ci);
- }
+ mlx5_tx_comp_flush(txq, last_cqe, ret, olx);
}
/**
* Pointer to TX queue structure.
* @param loc
* Pointer to burst routine local context.
+ * @param multi,
+ * Routine is called from multi-segment sending loop,
+ * do not correct the elts_head according to the pkts_copy.
* @param olx
* Configured Tx offloads mask. It is fully defined at
* compile time and may be used for optimization.
static __rte_always_inline void
mlx5_tx_request_completion(struct mlx5_txq_data *restrict txq,
struct mlx5_txq_local *restrict loc,
+ bool multi,
unsigned int olx)
{
uint16_t head = txq->elts_head;
unsigned int part;
- part = MLX5_TXOFF_CONFIG(INLINE) ? 0 : loc->pkts_sent -
- (MLX5_TXOFF_CONFIG(MULTI) ? loc->pkts_copy : 0);
+ part = (MLX5_TXOFF_CONFIG(INLINE) || multi) ?
+ 0 : loc->pkts_sent - loc->pkts_copy;
head += part;
if ((uint16_t)(head - txq->elts_comp) >= MLX5_TX_COMP_THRESH ||
(MLX5_TXOFF_CONFIG(INLINE) &&
es->swp_offs = txq_mbuf_to_swp(loc, &es->swp_flags, olx);
/* Fill metadata field if needed. */
es->metadata = MLX5_TXOFF_CONFIG(METADATA) ?
- loc->mbuf->ol_flags & PKT_TX_METADATA ?
- loc->mbuf->tx_metadata : 0 : 0;
+ loc->mbuf->ol_flags & PKT_TX_DYNF_METADATA ?
+ *RTE_FLOW_DYNF_METADATA(loc->mbuf) : 0 : 0;
/* Engage VLAN tag insertion feature if requested. */
if (MLX5_TXOFF_CONFIG(VLAN) &&
loc->mbuf->ol_flags & PKT_TX_VLAN_PKT) {
es->swp_offs = txq_mbuf_to_swp(loc, &es->swp_flags, olx);
/* Fill metadata field if needed. */
es->metadata = MLX5_TXOFF_CONFIG(METADATA) ?
- loc->mbuf->ol_flags & PKT_TX_METADATA ?
- loc->mbuf->tx_metadata : 0 : 0;
+ loc->mbuf->ol_flags & PKT_TX_DYNF_METADATA ?
+ *RTE_FLOW_DYNF_METADATA(loc->mbuf) : 0 : 0;
static_assert(MLX5_ESEG_MIN_INLINE_SIZE ==
(sizeof(uint16_t) +
sizeof(rte_v128u32_t)),
es->swp_offs = txq_mbuf_to_swp(loc, &es->swp_flags, olx);
/* Fill metadata field if needed. */
es->metadata = MLX5_TXOFF_CONFIG(METADATA) ?
- loc->mbuf->ol_flags & PKT_TX_METADATA ?
- loc->mbuf->tx_metadata : 0 : 0;
+ loc->mbuf->ol_flags & PKT_TX_DYNF_METADATA ?
+ *RTE_FLOW_DYNF_METADATA(loc->mbuf) : 0 : 0;
static_assert(MLX5_ESEG_MIN_INLINE_SIZE ==
(sizeof(uint16_t) +
sizeof(rte_v128u32_t)),
es->swp_offs = txq_mbuf_to_swp(loc, &es->swp_flags, olx);
/* Fill metadata field if needed. */
es->metadata = MLX5_TXOFF_CONFIG(METADATA) ?
- loc->mbuf->ol_flags & PKT_TX_METADATA ?
- loc->mbuf->tx_metadata : 0 : 0;
+ loc->mbuf->ol_flags & PKT_TX_DYNF_METADATA ?
+ *RTE_FLOW_DYNF_METADATA(loc->mbuf) : 0 : 0;
static_assert(MLX5_ESEG_MIN_INLINE_SIZE ==
(sizeof(uint16_t) +
sizeof(rte_v128u32_t)),
sizeof(struct rte_vlan_hdr) +
2 * RTE_ETHER_ADDR_LEN),
"invalid Ethernet Segment data size");
- assert(inlen > MLX5_ESEG_MIN_INLINE_SIZE);
+ assert(inlen >= MLX5_ESEG_MIN_INLINE_SIZE);
es->inline_hdr_sz = rte_cpu_to_be_16(inlen);
pdst = (uint8_t *)&es->inline_data;
if (MLX5_TXOFF_CONFIG(VLAN) && vlan) {
/* Unrolled implementation of generic rte_memcpy. */
dst = (uintptr_t)&dseg->inline_data[0];
src = (uintptr_t)buf;
+ if (len & 0x08) {
#ifdef RTE_ARCH_STRICT_ALIGN
- memcpy(dst, src, len);
+ assert(dst == RTE_PTR_ALIGN(dst, sizeof(uint32_t)));
+ *(uint32_t *)dst = *(unaligned_uint32_t *)src;
+ dst += sizeof(uint32_t);
+ src += sizeof(uint32_t);
+ *(uint32_t *)dst = *(unaligned_uint32_t *)src;
+ dst += sizeof(uint32_t);
+ src += sizeof(uint32_t);
#else
- if (len & 0x08) {
- *(uint64_t *)dst = *(uint64_t *)src;
+ *(uint64_t *)dst = *(unaligned_uint64_t *)src;
dst += sizeof(uint64_t);
src += sizeof(uint64_t);
+#endif
}
if (len & 0x04) {
- *(uint32_t *)dst = *(uint32_t *)src;
+ *(uint32_t *)dst = *(unaligned_uint32_t *)src;
dst += sizeof(uint32_t);
src += sizeof(uint32_t);
}
if (len & 0x02) {
- *(uint16_t *)dst = *(uint16_t *)src;
+ *(uint16_t *)dst = *(unaligned_uint16_t *)src;
dst += sizeof(uint16_t);
src += sizeof(uint16_t);
}
if (len & 0x01)
*(uint8_t *)dst = *(uint8_t *)src;
-#endif
}
/**
memcpy(pdst, buf, MLX5_DSEG_MIN_INLINE_SIZE);
buf += MLX5_DSEG_MIN_INLINE_SIZE;
pdst += MLX5_DSEG_MIN_INLINE_SIZE;
+ len -= MLX5_DSEG_MIN_INLINE_SIZE;
/* Insert VLAN ethertype + VLAN tag. Pointer is aligned. */
assert(pdst == RTE_PTR_ALIGN(pdst, MLX5_WSEG_SIZE));
+ if (unlikely(pdst >= (uint8_t *)txq->wqes_end))
+ pdst = (uint8_t *)txq->wqes;
*(uint32_t *)pdst = rte_cpu_to_be_32((RTE_ETHER_TYPE_VLAN << 16) |
loc->mbuf->vlan_tci);
pdst += sizeof(struct rte_vlan_hdr);
- if (unlikely(pdst >= (uint8_t *)txq->wqes_end))
- pdst = (uint8_t *)txq->wqes;
/*
* The WQEBB space availability is checked by caller.
* Here we should be aware of WQE ring buffer wraparound only.
txq->wqe_ci += (ds + 3) / 4;
loc->wqe_free -= (ds + 3) / 4;
/* Request CQE generation if limits are reached. */
- mlx5_tx_request_completion(txq, loc, olx);
+ mlx5_tx_request_completion(txq, loc, true, olx);
return MLX5_TXCMP_CODE_MULTI;
}
txq->wqe_ci += (ds + 3) / 4;
loc->wqe_free -= (ds + 3) / 4;
/* Request CQE generation if limits are reached. */
- mlx5_tx_request_completion(txq, loc, olx);
+ mlx5_tx_request_completion(txq, loc, true, olx);
return MLX5_TXCMP_CODE_MULTI;
}
txq->wqe_ci += (ds + 3) / 4;
loc->wqe_free -= (ds + 3) / 4;
/* Request CQE generation if limits are reached. */
- mlx5_tx_request_completion(txq, loc, olx);
+ mlx5_tx_request_completion(txq, loc, true, olx);
return MLX5_TXCMP_CODE_MULTI;
}
continue;
/* Here ends the series of multi-segment packets. */
if (MLX5_TXOFF_CONFIG(TSO) &&
- unlikely(!(loc->mbuf->ol_flags & PKT_TX_TCP_SEG)))
+ unlikely(loc->mbuf->ol_flags & PKT_TX_TCP_SEG))
return MLX5_TXCMP_CODE_TSO;
return MLX5_TXCMP_CODE_SINGLE;
}
++loc->pkts_sent;
--pkts_n;
/* Request CQE generation if limits are reached. */
- mlx5_tx_request_completion(txq, loc, olx);
+ mlx5_tx_request_completion(txq, loc, false, olx);
if (unlikely(!pkts_n || !loc->elts_free || !loc->wqe_free))
return MLX5_TXCMP_CODE_EXIT;
loc->mbuf = *pkts++;
if (MLX5_TXOFF_CONFIG(MULTI) &&
unlikely(NB_SEGS(loc->mbuf) > 1))
return MLX5_TXCMP_CODE_MULTI;
- if (unlikely(!(loc->mbuf->ol_flags & PKT_TX_TCP_SEG)))
+ if (likely(!(loc->mbuf->ol_flags & PKT_TX_TCP_SEG)))
return MLX5_TXCMP_CODE_SINGLE;
/* Continue with the next TSO packet. */
}
return false;
/* Fill metadata field if needed. */
if (MLX5_TXOFF_CONFIG(METADATA) &&
- es->metadata != (loc->mbuf->ol_flags & PKT_TX_METADATA ?
- loc->mbuf->tx_metadata : 0))
+ es->metadata != (loc->mbuf->ol_flags & PKT_TX_DYNF_METADATA ?
+ *RTE_FLOW_DYNF_METADATA(loc->mbuf) : 0))
return false;
/* There must be no VLAN packets in eMPW loop. */
if (MLX5_TXOFF_CONFIG(VLAN))
txq->wqe_ci += (ds + 3) / 4;
loc->wqe_free -= (ds + 3) / 4;
/* Request CQE generation if limits are reached. */
- mlx5_tx_request_completion(txq, loc, olx);
+ mlx5_tx_request_completion(txq, loc, false, olx);
}
/*
txq->wqe_ci += (len + 3) / 4;
loc->wqe_free -= (len + 3) / 4;
/* Request CQE generation if limits are reached. */
- mlx5_tx_request_completion(txq, loc, olx);
+ mlx5_tx_request_completion(txq, loc, false, olx);
}
/**
loc->wqe_free -= (2 + part + 3) / 4;
pkts_n -= part;
/* Request CQE generation if limits are reached. */
- mlx5_tx_request_completion(txq, loc, olx);
+ mlx5_tx_request_completion(txq, loc, false, olx);
if (unlikely(!pkts_n || !loc->elts_free || !loc->wqe_free))
return MLX5_TXCMP_CODE_EXIT;
loc->mbuf = *pkts++;
++loc->pkts_sent;
--pkts_n;
/* Request CQE generation if limits are reached. */
- mlx5_tx_request_completion(txq, loc, olx);
+ mlx5_tx_request_completion(txq, loc, false, olx);
if (unlikely(!pkts_n || !loc->elts_free || !loc->wqe_free))
return MLX5_TXCMP_CODE_EXIT;
loc->mbuf = *pkts++;
assert(txq->elts_s >= (uint16_t)(txq->elts_head - txq->elts_tail));
assert(txq->wqe_s >= (uint16_t)(txq->wqe_ci - txq->wqe_pi));
+ if (unlikely(!pkts_n))
+ return 0;
+ loc.pkts_sent = 0;
+ loc.pkts_copy = 0;
+ loc.wqe_last = NULL;
+
+send_loop:
+ loc.pkts_loop = loc.pkts_sent;
/*
* Check if there are some CQEs, if any:
* - process an encountered errors
* - free related mbufs
* - doorbell the NIC about processed CQEs
*/
- if (unlikely(!pkts_n))
- return 0;
- rte_prefetch0(*pkts);
+ rte_prefetch0(*(pkts + loc.pkts_sent));
mlx5_tx_handle_completion(txq, olx);
/*
* Calculate the number of available resources - elts and WQEs.
loc.wqe_free = txq->wqe_s -
(uint16_t)(txq->wqe_ci - txq->wqe_pi);
if (unlikely(!loc.elts_free || !loc.wqe_free))
- return 0;
- loc.pkts_sent = 0;
- loc.pkts_copy = 0;
- loc.wqe_last = NULL;
+ goto burst_exit;
for (;;) {
/*
* Fetch the packet from array. Usually this is
*/
assert(MLX5_TXOFF_CONFIG(INLINE) || loc.pkts_sent >= loc.pkts_copy);
/* Take a shortcut if nothing is sent. */
- if (unlikely(loc.pkts_sent == 0))
- return 0;
+ if (unlikely(loc.pkts_sent == loc.pkts_loop))
+ goto burst_exit;
/*
* Ring QP doorbell immediately after WQE building completion
* to improve latencies. The pure software related data treatment
*/
mlx5_tx_dbrec_cond_wmb(txq, loc.wqe_last, 0);
/* Not all of the mbufs may be stored into elts yet. */
- part = MLX5_TXOFF_CONFIG(INLINE) ? 0 : loc.pkts_sent -
- (MLX5_TXOFF_CONFIG(MULTI) ? loc.pkts_copy : 0);
+ part = MLX5_TXOFF_CONFIG(INLINE) ? 0 : loc.pkts_sent - loc.pkts_copy;
if (!MLX5_TXOFF_CONFIG(INLINE) && part) {
/*
* There are some single-segment mbufs not stored in elts.
* inlined mbufs.
*/
mlx5_tx_copy_elts(txq, pkts + loc.pkts_copy, part, olx);
+ loc.pkts_copy = loc.pkts_sent;
}
+ assert(txq->elts_s >= (uint16_t)(txq->elts_head - txq->elts_tail));
+ assert(txq->wqe_s >= (uint16_t)(txq->wqe_ci - txq->wqe_pi));
+ if (pkts_n > loc.pkts_sent) {
+ /*
+ * If burst size is large there might be no enough CQE
+ * fetched from completion queue and no enough resources
+ * freed to send all the packets.
+ */
+ goto send_loop;
+ }
+burst_exit:
#ifdef MLX5_PMD_SOFT_COUNTERS
/* Increment sent packets counter. */
txq->stats.opackets += loc.pkts_sent;
#endif
- assert(txq->elts_s >= (uint16_t)(txq->elts_head - txq->elts_tail));
- assert(txq->wqe_s >= (uint16_t)(txq->wqe_ci - txq->wqe_pi));
return loc.pkts_sent;
}
*/
olx |= MLX5_TXOFF_CONFIG_EMPW;
}
- if (tx_offloads & DEV_TX_OFFLOAD_MATCH_METADATA) {
+ if (rte_flow_dynf_metadata_avail()) {
/* We should support Flow metadata. */
olx |= MLX5_TXOFF_CONFIG_METADATA;
}
DRV_LOG(DEBUG, "\tEMPW (Enhanced MPW)");
return txoff_func[m].func;
}
-
-
git.droids-corp.org / dpdk.git / blobdiff