- rte_errno = ret;
- ERROR("%p: QP state to IBV_QPS_RTS failed: %s",
- (void *)dev, strerror(rte_errno));
- goto error;
- }
- /* Clean up txq in case we're reinitializing it. */
- DEBUG("%p: cleaning-up old txq just in case", (void *)txq);
- txq_cleanup(txq);
- *txq = tmpl;
- DEBUG("%p: txq updated with %p", (void *)txq, (void *)&tmpl);
- /* Pre-register known mempools. */
- rte_mempool_walk(txq_mp2mr_iter, txq);
- return 0;
-error:
- ret = rte_errno;
- txq_cleanup(&tmpl);
- rte_errno = ret;
- assert(rte_errno > 0);
- return -rte_errno;
-}
-
-/**
- * DPDK callback to configure a TX queue.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param idx
- * TX queue index.
- * @param desc
- * Number of descriptors to configure in queue.
- * @param socket
- * NUMA socket on which memory must be allocated.
- * @param[in] conf
- * Thresholds parameters.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-mlx4_tx_queue_setup(struct rte_eth_dev *dev, uint16_t idx, uint16_t desc,
- unsigned int socket, const struct rte_eth_txconf *conf)
-{
- struct priv *priv = dev->data->dev_private;
- struct txq *txq = (*priv->txqs)[idx];
- int ret;
-
- DEBUG("%p: configuring queue %u for %u descriptors",
- (void *)dev, idx, desc);
- if (idx >= priv->txqs_n) {
- rte_errno = EOVERFLOW;
- ERROR("%p: queue index out of range (%u >= %u)",
- (void *)dev, idx, priv->txqs_n);
- return -rte_errno;
- }
- if (txq != NULL) {
- DEBUG("%p: reusing already allocated queue index %u (%p)",
- (void *)dev, idx, (void *)txq);
- if (priv->started) {
- rte_errno = EEXIST;
- return -rte_errno;
- }
- (*priv->txqs)[idx] = NULL;
- txq_cleanup(txq);
- } else {
- txq = rte_calloc_socket("TXQ", 1, sizeof(*txq), 0, socket);
- if (txq == NULL) {
- rte_errno = ENOMEM;
- ERROR("%p: unable to allocate queue index %u",
- (void *)dev, idx);
- return -rte_errno;
- }
- }
- ret = txq_setup(dev, txq, desc, socket, conf);
- if (ret)
- rte_free(txq);
- else {
- txq->stats.idx = idx;
- DEBUG("%p: adding TX queue %p to list",
- (void *)dev, (void *)txq);
- (*priv->txqs)[idx] = txq;
- /* Update send callback. */
- dev->tx_pkt_burst = mlx4_tx_burst;
- }
- return ret;
-}
-
-/**
- * DPDK callback to release a TX queue.
- *
- * @param dpdk_txq
- * Generic TX queue pointer.
- */
-static void
-mlx4_tx_queue_release(void *dpdk_txq)
-{
- struct txq *txq = (struct txq *)dpdk_txq;
- struct priv *priv;
- unsigned int i;
-
- if (txq == NULL)
- return;
- priv = txq->priv;
- for (i = 0; (i != priv->txqs_n); ++i)
- if ((*priv->txqs)[i] == txq) {
- DEBUG("%p: removing TX queue %p from list",
- (void *)priv->dev, (void *)txq);
- (*priv->txqs)[i] = NULL;
- break;
- }
- txq_cleanup(txq);
- rte_free(txq);
-}
-
-/* RX queues handling. */
-
-/**
- * Allocate RX queue elements.
- *
- * @param rxq
- * Pointer to RX queue structure.
- * @param elts_n
- * Number of elements to allocate.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-rxq_alloc_elts(struct rxq *rxq, unsigned int elts_n)
-{
- unsigned int i;
- struct rxq_elt (*elts)[elts_n] =
- rte_calloc_socket("RXQ elements", 1, sizeof(*elts), 0,
- rxq->socket);
-
- if (elts == NULL) {
- rte_errno = ENOMEM;
- ERROR("%p: can't allocate packets array", (void *)rxq);
- goto error;
- }
- /* For each WR (packet). */
- for (i = 0; (i != elts_n); ++i) {
- struct rxq_elt *elt = &(*elts)[i];
- struct ibv_recv_wr *wr = &elt->wr;
- struct ibv_sge *sge = &(*elts)[i].sge;
- struct rte_mbuf *buf = rte_pktmbuf_alloc(rxq->mp);
-
- if (buf == NULL) {
- rte_errno = ENOMEM;
- ERROR("%p: empty mbuf pool", (void *)rxq);
- goto error;
- }
- elt->buf = buf;
- wr->next = &(*elts)[(i + 1)].wr;
- wr->sg_list = sge;
- wr->num_sge = 1;
- /* Headroom is reserved by rte_pktmbuf_alloc(). */
- assert(buf->data_off == RTE_PKTMBUF_HEADROOM);
- /* Buffer is supposed to be empty. */
- assert(rte_pktmbuf_data_len(buf) == 0);
- assert(rte_pktmbuf_pkt_len(buf) == 0);
- /* sge->addr must be able to store a pointer. */
- assert(sizeof(sge->addr) >= sizeof(uintptr_t));
- /* SGE keeps its headroom. */
- sge->addr = (uintptr_t)
- ((uint8_t *)buf->buf_addr + RTE_PKTMBUF_HEADROOM);
- sge->length = (buf->buf_len - RTE_PKTMBUF_HEADROOM);
- sge->lkey = rxq->mr->lkey;
- /* Redundant check for tailroom. */
- assert(sge->length == rte_pktmbuf_tailroom(buf));
- }
- /* The last WR pointer must be NULL. */
- (*elts)[(i - 1)].wr.next = NULL;
- DEBUG("%p: allocated and configured %u single-segment WRs",
- (void *)rxq, elts_n);
- rxq->elts_n = elts_n;
- rxq->elts_head = 0;
- rxq->elts = elts;
- return 0;
-error:
- if (elts != NULL) {
- for (i = 0; (i != RTE_DIM(*elts)); ++i)
- rte_pktmbuf_free_seg((*elts)[i].buf);
- rte_free(elts);
- }
- DEBUG("%p: failed, freed everything", (void *)rxq);
- assert(rte_errno > 0);
- return -rte_errno;
-}
-
-/**
- * Free RX queue elements.
- *
- * @param rxq
- * Pointer to RX queue structure.
- */
-static void
-rxq_free_elts(struct rxq *rxq)
-{
- unsigned int i;
- unsigned int elts_n = rxq->elts_n;
- struct rxq_elt (*elts)[elts_n] = rxq->elts;
-
- DEBUG("%p: freeing WRs", (void *)rxq);
- rxq->elts_n = 0;
- rxq->elts = NULL;
- if (elts == NULL)
- return;
- for (i = 0; (i != RTE_DIM(*elts)); ++i)
- rte_pktmbuf_free_seg((*elts)[i].buf);
- rte_free(elts);
-}
-
-/**
- * Unregister a MAC address.
- *
- * @param priv
- * Pointer to private structure.
- */
-static void
-priv_mac_addr_del(struct priv *priv)
-{
-#ifndef NDEBUG
- uint8_t (*mac)[ETHER_ADDR_LEN] = &priv->mac.addr_bytes;
-#endif
-
- if (!priv->mac_flow)
- return;
- DEBUG("%p: removing MAC address %02x:%02x:%02x:%02x:%02x:%02x",
- (void *)priv,
- (*mac)[0], (*mac)[1], (*mac)[2], (*mac)[3], (*mac)[4], (*mac)[5]);
- claim_zero(ibv_destroy_flow(priv->mac_flow));
- priv->mac_flow = NULL;
-}
-
-/**
- * Register a MAC address.
- *
- * The MAC address is registered in queue 0.
- *
- * @param priv
- * Pointer to private structure.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-priv_mac_addr_add(struct priv *priv)
-{
- uint8_t (*mac)[ETHER_ADDR_LEN] = &priv->mac.addr_bytes;
- struct rxq *rxq;
- struct ibv_flow *flow;
-
- /* If device isn't started, this is all we need to do. */
- if (!priv->started)
- return 0;
- if (priv->isolated)
- return 0;
- if (*priv->rxqs && (*priv->rxqs)[0])
- rxq = (*priv->rxqs)[0];
- else
- return 0;
-
- /* Allocate flow specification on the stack. */
- struct __attribute__((packed)) {
- struct ibv_flow_attr attr;
- struct ibv_flow_spec_eth spec;
- } data;
- struct ibv_flow_attr *attr = &data.attr;
- struct ibv_flow_spec_eth *spec = &data.spec;
-
- if (priv->mac_flow)
- priv_mac_addr_del(priv);
- /*
- * No padding must be inserted by the compiler between attr and spec.
- * This layout is expected by libibverbs.
- */
- assert(((uint8_t *)attr + sizeof(*attr)) == (uint8_t *)spec);
- *attr = (struct ibv_flow_attr){
- .type = IBV_FLOW_ATTR_NORMAL,
- .priority = 3,
- .num_of_specs = 1,
- .port = priv->port,
- .flags = 0
- };
- *spec = (struct ibv_flow_spec_eth){
- .type = IBV_FLOW_SPEC_ETH,
- .size = sizeof(*spec),
- .val = {
- .dst_mac = {
- (*mac)[0], (*mac)[1], (*mac)[2],
- (*mac)[3], (*mac)[4], (*mac)[5]
- },
- },
- .mask = {
- .dst_mac = "\xff\xff\xff\xff\xff\xff",
- }
- };
- DEBUG("%p: adding MAC address %02x:%02x:%02x:%02x:%02x:%02x",
- (void *)priv,
- (*mac)[0], (*mac)[1], (*mac)[2], (*mac)[3], (*mac)[4], (*mac)[5]);
- /* Create related flow. */
- flow = ibv_create_flow(rxq->qp, attr);
- if (flow == NULL) {
- rte_errno = errno ? errno : EINVAL;
- ERROR("%p: flow configuration failed, errno=%d: %s",
- (void *)rxq, rte_errno, strerror(errno));
- return -rte_errno;
- }
- assert(priv->mac_flow == NULL);
- priv->mac_flow = flow;
- return 0;
-}
-
-/**
- * Clean up a RX queue.
- *
- * Destroy objects, free allocated memory and reset the structure for reuse.
- *
- * @param rxq
- * Pointer to RX queue structure.
- */
-static void
-rxq_cleanup(struct rxq *rxq)
-{
- DEBUG("cleaning up %p", (void *)rxq);
- rxq_free_elts(rxq);
- if (rxq->qp != NULL)
- claim_zero(ibv_destroy_qp(rxq->qp));
- if (rxq->cq != NULL)
- claim_zero(ibv_destroy_cq(rxq->cq));
- if (rxq->channel != NULL)
- claim_zero(ibv_destroy_comp_channel(rxq->channel));
- if (rxq->mr != NULL)
- claim_zero(ibv_dereg_mr(rxq->mr));
- memset(rxq, 0, sizeof(*rxq));
-}
-
-/**
- * DPDK callback for RX.
- *
- * The following function doesn't manage scattered packets.
- *
- * @param dpdk_rxq
- * Generic pointer to RX queue structure.
- * @param[out] pkts
- * Array to store received packets.
- * @param pkts_n
- * Maximum number of packets in array.
- *
- * @return
- * Number of packets successfully received (<= pkts_n).
- */
-static uint16_t
-mlx4_rx_burst(void *dpdk_rxq, struct rte_mbuf **pkts, uint16_t pkts_n)
-{
- struct rxq *rxq = (struct rxq *)dpdk_rxq;
- struct rxq_elt (*elts)[rxq->elts_n] = rxq->elts;
- const unsigned int elts_n = rxq->elts_n;
- unsigned int elts_head = rxq->elts_head;
- struct ibv_wc wcs[pkts_n];
- struct ibv_recv_wr *wr_head = NULL;
- struct ibv_recv_wr **wr_next = &wr_head;
- struct ibv_recv_wr *wr_bad = NULL;
- unsigned int i;
- unsigned int pkts_ret = 0;
- int ret;
-
- ret = ibv_poll_cq(rxq->cq, pkts_n, wcs);
- if (unlikely(ret == 0))
- return 0;
- if (unlikely(ret < 0)) {
- DEBUG("rxq=%p, ibv_poll_cq() failed (wc_n=%d)",
- (void *)rxq, ret);
- return 0;
- }
- assert(ret <= (int)pkts_n);
- /* For each work completion. */
- for (i = 0; i != (unsigned int)ret; ++i) {
- struct ibv_wc *wc = &wcs[i];
- struct rxq_elt *elt = &(*elts)[elts_head];
- struct ibv_recv_wr *wr = &elt->wr;
- uint32_t len = wc->byte_len;
- struct rte_mbuf *seg = elt->buf;
- struct rte_mbuf *rep;
-
- /* Sanity checks. */
- assert(wr->sg_list == &elt->sge);
- assert(wr->num_sge == 1);
- assert(elts_head < rxq->elts_n);
- assert(rxq->elts_head < rxq->elts_n);
- /*
- * Fetch initial bytes of packet descriptor into a
- * cacheline while allocating rep.
- */
- rte_mbuf_prefetch_part1(seg);
- rte_mbuf_prefetch_part2(seg);
- /* Link completed WRs together for repost. */
- *wr_next = wr;
- wr_next = &wr->next;
- if (unlikely(wc->status != IBV_WC_SUCCESS)) {
- /* Whatever, just repost the offending WR. */
- DEBUG("rxq=%p: bad work completion status (%d): %s",
- (void *)rxq, wc->status,
- ibv_wc_status_str(wc->status));
- /* Increment dropped packets counter. */
- ++rxq->stats.idropped;
- goto repost;
- }
- rep = rte_mbuf_raw_alloc(rxq->mp);
- if (unlikely(rep == NULL)) {
- /*
- * Unable to allocate a replacement mbuf,
- * repost WR.
- */
- DEBUG("rxq=%p: can't allocate a new mbuf",
- (void *)rxq);
- /* Increase out of memory counters. */
- ++rxq->stats.rx_nombuf;
- ++rxq->priv->dev->data->rx_mbuf_alloc_failed;
- goto repost;
- }
- /* Reconfigure sge to use rep instead of seg. */
- elt->sge.addr = (uintptr_t)rep->buf_addr + RTE_PKTMBUF_HEADROOM;
- assert(elt->sge.lkey == rxq->mr->lkey);
- elt->buf = rep;
- /* Update seg information. */
- seg->data_off = RTE_PKTMBUF_HEADROOM;
- seg->nb_segs = 1;
- seg->port = rxq->port_id;
- seg->next = NULL;
- seg->pkt_len = len;
- seg->data_len = len;
- seg->packet_type = 0;
- seg->ol_flags = 0;
- /* Return packet. */
- *(pkts++) = seg;
- ++pkts_ret;
- /* Increase bytes counter. */
- rxq->stats.ibytes += len;
-repost:
- if (++elts_head >= elts_n)
- elts_head = 0;
- continue;
- }
- if (unlikely(i == 0))
- return 0;
- /* Repost WRs. */
- *wr_next = NULL;
- assert(wr_head);
- ret = ibv_post_recv(rxq->qp, wr_head, &wr_bad);
- if (unlikely(ret)) {
- /* Inability to repost WRs is fatal. */
- DEBUG("%p: recv_burst(): failed (ret=%d)",
- (void *)rxq->priv,
- ret);
- abort();
- }
- rxq->elts_head = elts_head;
- /* Increase packets counter. */
- rxq->stats.ipackets += pkts_ret;
- return pkts_ret;
-}
-
-/**
- * Allocate a Queue Pair.
- * Optionally setup inline receive if supported.
- *
- * @param priv
- * Pointer to private structure.
- * @param cq
- * Completion queue to associate with QP.
- * @param desc
- * Number of descriptors in QP (hint only).
- *
- * @return
- * QP pointer or NULL in case of error and rte_errno is set.
- */
-static struct ibv_qp *
-rxq_setup_qp(struct priv *priv, struct ibv_cq *cq, uint16_t desc)
-{
- struct ibv_qp *qp;
- struct ibv_qp_init_attr attr = {
- /* CQ to be associated with the send queue. */
- .send_cq = cq,
- /* CQ to be associated with the receive queue. */
- .recv_cq = cq,
- .cap = {
- /* Max number of outstanding WRs. */
- .max_recv_wr = ((priv->device_attr.max_qp_wr < desc) ?
- priv->device_attr.max_qp_wr :
- desc),
- /* Max number of scatter/gather elements in a WR. */
- .max_recv_sge = 1,
- },
- .qp_type = IBV_QPT_RAW_PACKET,
- };
-
- qp = ibv_create_qp(priv->pd, &attr);
- if (!qp)
- rte_errno = errno ? errno : EINVAL;
- return qp;
-}
-
-/**
- * Configure a RX queue.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param rxq
- * Pointer to RX queue structure.
- * @param desc
- * Number of descriptors to configure in queue.
- * @param socket
- * NUMA socket on which memory must be allocated.
- * @param[in] conf
- * Thresholds parameters.
- * @param mp
- * Memory pool for buffer allocations.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-rxq_setup(struct rte_eth_dev *dev, struct rxq *rxq, uint16_t desc,
- unsigned int socket, const struct rte_eth_rxconf *conf,
- struct rte_mempool *mp)
-{
- struct priv *priv = dev->data->dev_private;
- struct rxq tmpl = {
- .priv = priv,
- .mp = mp,
- .socket = socket
- };
- struct ibv_qp_attr mod;
- struct ibv_recv_wr *bad_wr;
- unsigned int mb_len;
- int ret;
-
- (void)conf; /* Thresholds configuration (ignored). */
- mb_len = rte_pktmbuf_data_room_size(mp);
- if (desc == 0) {
- rte_errno = EINVAL;
- ERROR("%p: invalid number of Rx descriptors", (void *)dev);
- goto error;
- }
- /* Enable scattered packets support for this queue if necessary. */
- assert(mb_len >= RTE_PKTMBUF_HEADROOM);
- if (dev->data->dev_conf.rxmode.max_rx_pkt_len <=
- (mb_len - RTE_PKTMBUF_HEADROOM)) {
- ;
- } else if (dev->data->dev_conf.rxmode.enable_scatter) {
- WARN("%p: scattered mode has been requested but is"
- " not supported, this may lead to packet loss",
- (void *)dev);
- } else {
- WARN("%p: the requested maximum Rx packet size (%u) is"
- " larger than a single mbuf (%u) and scattered"
- " mode has not been requested",
- (void *)dev,
- dev->data->dev_conf.rxmode.max_rx_pkt_len,
- mb_len - RTE_PKTMBUF_HEADROOM);
- }
- /* Use the entire RX mempool as the memory region. */
- tmpl.mr = mlx4_mp2mr(priv->pd, mp);
- if (tmpl.mr == NULL) {
- rte_errno = EINVAL;
- ERROR("%p: MR creation failure: %s",
- (void *)dev, strerror(rte_errno));
- goto error;
- }
- if (dev->data->dev_conf.intr_conf.rxq) {
- tmpl.channel = ibv_create_comp_channel(priv->ctx);
- if (tmpl.channel == NULL) {
- rte_errno = ENOMEM;
- ERROR("%p: Rx interrupt completion channel creation"
- " failure: %s",
- (void *)dev, strerror(rte_errno));
- goto error;
- }
- if (mlx4_fd_set_non_blocking(tmpl.channel->fd) < 0) {
- ERROR("%p: unable to make Rx interrupt completion"
- " channel non-blocking: %s",
- (void *)dev, strerror(rte_errno));
- goto error;
- }
- }
- tmpl.cq = ibv_create_cq(priv->ctx, desc, NULL, tmpl.channel, 0);
- if (tmpl.cq == NULL) {
- rte_errno = ENOMEM;
- ERROR("%p: CQ creation failure: %s",
- (void *)dev, strerror(rte_errno));
- goto error;
- }
- DEBUG("priv->device_attr.max_qp_wr is %d",
- priv->device_attr.max_qp_wr);
- DEBUG("priv->device_attr.max_sge is %d",
- priv->device_attr.max_sge);
- tmpl.qp = rxq_setup_qp(priv, tmpl.cq, desc);
- if (tmpl.qp == NULL) {
- ERROR("%p: QP creation failure: %s",
- (void *)dev, strerror(rte_errno));
- goto error;
- }
- mod = (struct ibv_qp_attr){
- /* Move the QP to this state. */
- .qp_state = IBV_QPS_INIT,
- /* Primary port number. */
- .port_num = priv->port
- };
- ret = ibv_modify_qp(tmpl.qp, &mod, IBV_QP_STATE | IBV_QP_PORT);
- if (ret) {
- rte_errno = ret;
- ERROR("%p: QP state to IBV_QPS_INIT failed: %s",
- (void *)dev, strerror(rte_errno));
- goto error;
- }
- ret = rxq_alloc_elts(&tmpl, desc);
- if (ret) {
- ERROR("%p: RXQ allocation failed: %s",
- (void *)dev, strerror(rte_errno));
- goto error;
- }
- ret = ibv_post_recv(tmpl.qp, &(*tmpl.elts)[0].wr, &bad_wr);
- if (ret) {
- rte_errno = ret;
- ERROR("%p: ibv_post_recv() failed for WR %p: %s",
- (void *)dev,
- (void *)bad_wr,
- strerror(rte_errno));
- goto error;
- }
- mod = (struct ibv_qp_attr){
- .qp_state = IBV_QPS_RTR
- };
- ret = ibv_modify_qp(tmpl.qp, &mod, IBV_QP_STATE);
- if (ret) {
- rte_errno = ret;
- ERROR("%p: QP state to IBV_QPS_RTR failed: %s",
- (void *)dev, strerror(rte_errno));
- goto error;
- }
- /* Save port ID. */
- tmpl.port_id = dev->data->port_id;
- DEBUG("%p: RTE port ID: %u", (void *)rxq, tmpl.port_id);
- /* Clean up rxq in case we're reinitializing it. */
- DEBUG("%p: cleaning-up old rxq just in case", (void *)rxq);
- rxq_cleanup(rxq);
- *rxq = tmpl;
- DEBUG("%p: rxq updated with %p", (void *)rxq, (void *)&tmpl);
- return 0;
-error:
- ret = rte_errno;
- rxq_cleanup(&tmpl);
- rte_errno = ret;
- assert(rte_errno > 0);
- return -rte_errno;
-}
-
-/**
- * DPDK callback to configure a RX queue.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param idx
- * RX queue index.
- * @param desc
- * Number of descriptors to configure in queue.
- * @param socket
- * NUMA socket on which memory must be allocated.
- * @param[in] conf
- * Thresholds parameters.
- * @param mp
- * Memory pool for buffer allocations.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-mlx4_rx_queue_setup(struct rte_eth_dev *dev, uint16_t idx, uint16_t desc,
- unsigned int socket, const struct rte_eth_rxconf *conf,
- struct rte_mempool *mp)
-{
- struct priv *priv = dev->data->dev_private;
- struct rxq *rxq = (*priv->rxqs)[idx];
- int ret;
-
- DEBUG("%p: configuring queue %u for %u descriptors",
- (void *)dev, idx, desc);
- if (idx >= priv->rxqs_n) {
- rte_errno = EOVERFLOW;
- ERROR("%p: queue index out of range (%u >= %u)",
- (void *)dev, idx, priv->rxqs_n);
- return -rte_errno;
- }
- if (rxq != NULL) {
- DEBUG("%p: reusing already allocated queue index %u (%p)",
- (void *)dev, idx, (void *)rxq);
- if (priv->started) {
- rte_errno = EEXIST;
- return -rte_errno;
- }
- (*priv->rxqs)[idx] = NULL;
- if (idx == 0)
- priv_mac_addr_del(priv);
- rxq_cleanup(rxq);
- } else {
- rxq = rte_calloc_socket("RXQ", 1, sizeof(*rxq), 0, socket);
- if (rxq == NULL) {
- rte_errno = ENOMEM;
- ERROR("%p: unable to allocate queue index %u",
- (void *)dev, idx);
- return -rte_errno;
- }
- }
- ret = rxq_setup(dev, rxq, desc, socket, conf, mp);
- if (ret)
- rte_free(rxq);
- else {
- rxq->stats.idx = idx;
- DEBUG("%p: adding RX queue %p to list",
- (void *)dev, (void *)rxq);
- (*priv->rxqs)[idx] = rxq;
- /* Update receive callback. */
- dev->rx_pkt_burst = mlx4_rx_burst;
- }
- return ret;
-}
-
-/**
- * DPDK callback to release a RX queue.
- *
- * @param dpdk_rxq
- * Generic RX queue pointer.
- */
-static void
-mlx4_rx_queue_release(void *dpdk_rxq)
-{
- struct rxq *rxq = (struct rxq *)dpdk_rxq;
- struct priv *priv;
- unsigned int i;
-
- if (rxq == NULL)
- return;
- priv = rxq->priv;
- for (i = 0; (i != priv->rxqs_n); ++i)
- if ((*priv->rxqs)[i] == rxq) {
- DEBUG("%p: removing RX queue %p from list",
- (void *)priv->dev, (void *)rxq);
- (*priv->rxqs)[i] = NULL;
- if (i == 0)
- priv_mac_addr_del(priv);
- break;
- }
- rxq_cleanup(rxq);
- rte_free(rxq);
-}
-
-static int priv_interrupt_handler_install(struct priv *priv);
-static int priv_removal_interrupt_handler_install(struct priv *priv);
-static int priv_link_interrupt_handler_install(struct priv *priv);
-
-/**
- * DPDK callback to start the device.
- *
- * Simulate device start by attaching all configured flows.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-mlx4_dev_start(struct rte_eth_dev *dev)
-{
- struct priv *priv = dev->data->dev_private;
- int ret;
-
- if (priv->started)
- return 0;
- DEBUG("%p: attaching configured flows to all RX queues", (void *)dev);
- priv->started = 1;
- ret = priv_mac_addr_add(priv);
- if (ret)
- goto err;
- ret = priv_link_interrupt_handler_install(priv);
- if (ret) {
- ERROR("%p: LSC handler install failed",
- (void *)dev);
- goto err;
- }
- ret = priv_removal_interrupt_handler_install(priv);
- if (ret) {
- ERROR("%p: RMV handler install failed",
- (void *)dev);
- goto err;
- }
- ret = priv_rx_intr_vec_enable(priv);
- if (ret) {
- ERROR("%p: Rx interrupt vector creation failed",
- (void *)dev);
- goto err;
- }
- ret = mlx4_priv_flow_start(priv);
- if (ret) {
- ERROR("%p: flow start failed: %s",
- (void *)dev, strerror(ret));
- goto err;
- }
- return 0;
-err:
- /* Rollback. */
- priv_mac_addr_del(priv);
- priv->started = 0;
- return ret;
-}
-
-/**
- * DPDK callback to stop the device.
- *
- * Simulate device stop by detaching all configured flows.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- */
-static void
-mlx4_dev_stop(struct rte_eth_dev *dev)
-{
- struct priv *priv = dev->data->dev_private;
-
- if (!priv->started)
- return;
- DEBUG("%p: detaching flows from all RX queues", (void *)dev);
- priv->started = 0;
- mlx4_priv_flow_stop(priv);
- priv_mac_addr_del(priv);
-}
-
-/**
- * Dummy DPDK callback for TX.
- *
- * This function is used to temporarily replace the real callback during
- * unsafe control operations on the queue, or in case of error.
- *
- * @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).
- */
-static uint16_t
-removed_tx_burst(void *dpdk_txq, struct rte_mbuf **pkts, uint16_t pkts_n)
-{
- (void)dpdk_txq;
- (void)pkts;
- (void)pkts_n;
- return 0;
-}
-
-/**
- * Dummy DPDK callback for RX.
- *
- * This function is used to temporarily replace the real callback during
- * unsafe control operations on the queue, or in case of error.
- *
- * @param dpdk_rxq
- * Generic pointer to RX queue structure.
- * @param[out] pkts
- * Array to store received packets.
- * @param pkts_n
- * Maximum number of packets in array.
- *
- * @return
- * Number of packets successfully received (<= pkts_n).
- */
-static uint16_t
-removed_rx_burst(void *dpdk_rxq, struct rte_mbuf **pkts, uint16_t pkts_n)
-{
- (void)dpdk_rxq;
- (void)pkts;
- (void)pkts_n;
- return 0;
-}
-
-static int priv_interrupt_handler_uninstall(struct priv *priv);
-static int priv_removal_interrupt_handler_uninstall(struct priv *priv);
-static int priv_link_interrupt_handler_uninstall(struct priv *priv);
-
-/**
- * DPDK callback to close the device.
- *
- * Destroy all queues and objects, free memory.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- */
-static void
-mlx4_dev_close(struct rte_eth_dev *dev)
-{
- struct priv *priv = dev->data->dev_private;
- void *tmp;
- unsigned int i;
-
- if (priv == NULL)
- return;
- DEBUG("%p: closing device \"%s\"",
- (void *)dev,
- ((priv->ctx != NULL) ? priv->ctx->device->name : ""));
- priv_mac_addr_del(priv);
- /*
- * Prevent crashes when queues are still in use. This is unfortunately
- * still required for DPDK 1.3 because some programs (such as testpmd)
- * never release them before closing the device.
- */
- dev->rx_pkt_burst = removed_rx_burst;
- dev->tx_pkt_burst = removed_tx_burst;
- if (priv->rxqs != NULL) {
- /* XXX race condition if mlx4_rx_burst() is still running. */
- usleep(1000);
- for (i = 0; (i != priv->rxqs_n); ++i) {
- tmp = (*priv->rxqs)[i];
- if (tmp == NULL)
- continue;
- (*priv->rxqs)[i] = NULL;
- rxq_cleanup(tmp);
- rte_free(tmp);
- }
- priv->rxqs_n = 0;
- priv->rxqs = NULL;
- }
- if (priv->txqs != NULL) {
- /* XXX race condition if mlx4_tx_burst() is still running. */
- usleep(1000);
- for (i = 0; (i != priv->txqs_n); ++i) {
- tmp = (*priv->txqs)[i];
- if (tmp == NULL)
- continue;
- (*priv->txqs)[i] = NULL;
- txq_cleanup(tmp);
- rte_free(tmp);
- }
- priv->txqs_n = 0;
- priv->txqs = NULL;
- }
- if (priv->pd != NULL) {
- assert(priv->ctx != NULL);
- claim_zero(ibv_dealloc_pd(priv->pd));
- claim_zero(ibv_close_device(priv->ctx));
- } else
- assert(priv->ctx == NULL);
- priv_removal_interrupt_handler_uninstall(priv);
- priv_link_interrupt_handler_uninstall(priv);
- priv_rx_intr_vec_disable(priv);
- memset(priv, 0, sizeof(*priv));
-}
-
-/**
- * Change the link state (UP / DOWN).
- *
- * @param priv
- * Pointer to Ethernet device private data.
- * @param up
- * Nonzero for link up, otherwise link down.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-priv_set_link(struct priv *priv, int up)
-{
- struct rte_eth_dev *dev = priv->dev;
- int err;
-
- if (up) {
- err = priv_set_flags(priv, ~IFF_UP, IFF_UP);
- if (err)
- return err;
- dev->rx_pkt_burst = mlx4_rx_burst;
- } else {
- err = priv_set_flags(priv, ~IFF_UP, ~IFF_UP);
- if (err)
- return err;
- dev->rx_pkt_burst = removed_rx_burst;
- dev->tx_pkt_burst = removed_tx_burst;
- }
- return 0;
-}
-
-/**
- * DPDK callback to bring the link DOWN.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-mlx4_set_link_down(struct rte_eth_dev *dev)
-{
- struct priv *priv = dev->data->dev_private;
-
- return priv_set_link(priv, 0);
-}
-
-/**
- * DPDK callback to bring the link UP.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-mlx4_set_link_up(struct rte_eth_dev *dev)
-{
- struct priv *priv = dev->data->dev_private;
-
- return priv_set_link(priv, 1);
-}
-
-/**
- * DPDK callback to get information about the device.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param[out] info
- * Info structure output buffer.
- */
-static void
-mlx4_dev_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
-{
- struct priv *priv = dev->data->dev_private;
- unsigned int max;
- char ifname[IF_NAMESIZE];
-
- info->pci_dev = RTE_ETH_DEV_TO_PCI(dev);
- if (priv == NULL)
- return;
- /* FIXME: we should ask the device for these values. */
- info->min_rx_bufsize = 32;
- info->max_rx_pktlen = 65536;
- /*
- * Since we need one CQ per QP, the limit is the minimum number
- * between the two values.
- */
- max = ((priv->device_attr.max_cq > priv->device_attr.max_qp) ?
- priv->device_attr.max_qp : priv->device_attr.max_cq);
- /* If max >= 65535 then max = 0, max_rx_queues is uint16_t. */
- if (max >= 65535)
- max = 65535;
- info->max_rx_queues = max;
- info->max_tx_queues = max;
- /* Last array entry is reserved for broadcast. */
- info->max_mac_addrs = 1;
- info->rx_offload_capa = 0;
- info->tx_offload_capa = 0;
- if (priv_get_ifname(priv, &ifname) == 0)
- info->if_index = if_nametoindex(ifname);
- info->speed_capa =
- ETH_LINK_SPEED_1G |
- ETH_LINK_SPEED_10G |
- ETH_LINK_SPEED_20G |
- ETH_LINK_SPEED_40G |
- ETH_LINK_SPEED_56G;
-}
-
-/**
- * DPDK callback to get device statistics.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param[out] stats
- * Stats structure output buffer.
- */
-static void
-mlx4_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *stats)
-{
- struct priv *priv = dev->data->dev_private;
- struct rte_eth_stats tmp = {0};
- unsigned int i;
- unsigned int idx;
-
- if (priv == NULL)
- return;
- /* Add software counters. */
- for (i = 0; (i != priv->rxqs_n); ++i) {
- struct rxq *rxq = (*priv->rxqs)[i];
-
- if (rxq == NULL)
- continue;
- idx = rxq->stats.idx;
- if (idx < RTE_ETHDEV_QUEUE_STAT_CNTRS) {
- tmp.q_ipackets[idx] += rxq->stats.ipackets;
- tmp.q_ibytes[idx] += rxq->stats.ibytes;
- tmp.q_errors[idx] += (rxq->stats.idropped +
- rxq->stats.rx_nombuf);
- }
- tmp.ipackets += rxq->stats.ipackets;
- tmp.ibytes += rxq->stats.ibytes;
- tmp.ierrors += rxq->stats.idropped;
- tmp.rx_nombuf += rxq->stats.rx_nombuf;
- }
- for (i = 0; (i != priv->txqs_n); ++i) {
- struct txq *txq = (*priv->txqs)[i];
-
- if (txq == NULL)
- continue;
- idx = txq->stats.idx;
- if (idx < RTE_ETHDEV_QUEUE_STAT_CNTRS) {
- tmp.q_opackets[idx] += txq->stats.opackets;
- tmp.q_obytes[idx] += txq->stats.obytes;
- tmp.q_errors[idx] += txq->stats.odropped;
- }
- tmp.opackets += txq->stats.opackets;
- tmp.obytes += txq->stats.obytes;
- tmp.oerrors += txq->stats.odropped;
- }
- *stats = tmp;
-}
-
-/**
- * DPDK callback to clear device statistics.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- */
-static void
-mlx4_stats_reset(struct rte_eth_dev *dev)
-{
- struct priv *priv = dev->data->dev_private;
- unsigned int i;
- unsigned int idx;
-
- if (priv == NULL)
- return;
- for (i = 0; (i != priv->rxqs_n); ++i) {
- if ((*priv->rxqs)[i] == NULL)
- continue;
- idx = (*priv->rxqs)[i]->stats.idx;
- (*priv->rxqs)[i]->stats =
- (struct mlx4_rxq_stats){ .idx = idx };
- }
- for (i = 0; (i != priv->txqs_n); ++i) {
- if ((*priv->txqs)[i] == NULL)
- continue;
- idx = (*priv->txqs)[i]->stats.idx;
- (*priv->txqs)[i]->stats =
- (struct mlx4_txq_stats){ .idx = idx };
- }
-}
-
-/**
- * DPDK callback to retrieve physical link information.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param wait_to_complete
- * Wait for request completion (ignored).
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-mlx4_link_update(struct rte_eth_dev *dev, int wait_to_complete)
-{
- const struct priv *priv = dev->data->dev_private;
- struct ethtool_cmd edata = {
- .cmd = ETHTOOL_GSET
- };
- struct ifreq ifr;
- struct rte_eth_link dev_link;
- int link_speed = 0;
-
- if (priv == NULL) {
- rte_errno = EINVAL;
- return -rte_errno;
- }
- (void)wait_to_complete;
- if (priv_ifreq(priv, SIOCGIFFLAGS, &ifr)) {
- WARN("ioctl(SIOCGIFFLAGS) failed: %s", strerror(rte_errno));
- return -rte_errno;
- }
- memset(&dev_link, 0, sizeof(dev_link));
- dev_link.link_status = ((ifr.ifr_flags & IFF_UP) &&
- (ifr.ifr_flags & IFF_RUNNING));
- ifr.ifr_data = (void *)&edata;
- if (priv_ifreq(priv, SIOCETHTOOL, &ifr)) {
- WARN("ioctl(SIOCETHTOOL, ETHTOOL_GSET) failed: %s",
- strerror(rte_errno));
- return -rte_errno;
- }
- link_speed = ethtool_cmd_speed(&edata);
- if (link_speed == -1)
- dev_link.link_speed = 0;
- else
- dev_link.link_speed = link_speed;
- dev_link.link_duplex = ((edata.duplex == DUPLEX_HALF) ?
- ETH_LINK_HALF_DUPLEX : ETH_LINK_FULL_DUPLEX);
- dev_link.link_autoneg = !(dev->data->dev_conf.link_speeds &
- ETH_LINK_SPEED_FIXED);
- dev->data->dev_link = dev_link;
- return 0;
-}
-
-/**
- * DPDK callback to get flow control status.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param[out] fc_conf
- * Flow control output buffer.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-mlx4_dev_get_flow_ctrl(struct rte_eth_dev *dev, struct rte_eth_fc_conf *fc_conf)
-{
- struct priv *priv = dev->data->dev_private;
- struct ifreq ifr;
- struct ethtool_pauseparam ethpause = {
- .cmd = ETHTOOL_GPAUSEPARAM
- };
- int ret;
-
- ifr.ifr_data = (void *)ðpause;
- if (priv_ifreq(priv, SIOCETHTOOL, &ifr)) {
- ret = rte_errno;
- WARN("ioctl(SIOCETHTOOL, ETHTOOL_GPAUSEPARAM)"
- " failed: %s",
- strerror(rte_errno));
- goto out;
- }
- fc_conf->autoneg = ethpause.autoneg;
- if (ethpause.rx_pause && ethpause.tx_pause)
- fc_conf->mode = RTE_FC_FULL;
- else if (ethpause.rx_pause)
- fc_conf->mode = RTE_FC_RX_PAUSE;
- else if (ethpause.tx_pause)
- fc_conf->mode = RTE_FC_TX_PAUSE;
- else
- fc_conf->mode = RTE_FC_NONE;
- ret = 0;
-out:
- assert(ret >= 0);
- return -ret;
-}
-
-/**
- * DPDK callback to modify flow control parameters.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param[in] fc_conf
- * Flow control parameters.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-mlx4_dev_set_flow_ctrl(struct rte_eth_dev *dev, struct rte_eth_fc_conf *fc_conf)
-{
- struct priv *priv = dev->data->dev_private;
- struct ifreq ifr;
- struct ethtool_pauseparam ethpause = {
- .cmd = ETHTOOL_SPAUSEPARAM
- };
- int ret;
-
- ifr.ifr_data = (void *)ðpause;
- ethpause.autoneg = fc_conf->autoneg;
- if (((fc_conf->mode & RTE_FC_FULL) == RTE_FC_FULL) ||
- (fc_conf->mode & RTE_FC_RX_PAUSE))
- ethpause.rx_pause = 1;
- else
- ethpause.rx_pause = 0;
- if (((fc_conf->mode & RTE_FC_FULL) == RTE_FC_FULL) ||
- (fc_conf->mode & RTE_FC_TX_PAUSE))
- ethpause.tx_pause = 1;
- else
- ethpause.tx_pause = 0;
- if (priv_ifreq(priv, SIOCETHTOOL, &ifr)) {
- ret = rte_errno;
- WARN("ioctl(SIOCETHTOOL, ETHTOOL_SPAUSEPARAM)"
- " failed: %s",
- strerror(rte_errno));
- goto out;
- }
- ret = 0;
-out:
- assert(ret >= 0);
- return -ret;
-}
-
-const struct rte_flow_ops mlx4_flow_ops = {
- .validate = mlx4_flow_validate,
- .create = mlx4_flow_create,
- .destroy = mlx4_flow_destroy,
- .flush = mlx4_flow_flush,
- .query = NULL,
- .isolate = mlx4_flow_isolate,
-};
-
-/**
- * Manage filter operations.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param filter_type
- * Filter type.
- * @param filter_op
- * Operation to perform.
- * @param arg
- * Pointer to operation-specific structure.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-mlx4_dev_filter_ctrl(struct rte_eth_dev *dev,
- enum rte_filter_type filter_type,
- enum rte_filter_op filter_op,
- void *arg)
-{
- switch (filter_type) {
- case RTE_ETH_FILTER_GENERIC:
- if (filter_op != RTE_ETH_FILTER_GET)
- break;
- *(const void **)arg = &mlx4_flow_ops;
- return 0;
- default:
- ERROR("%p: filter type (%d) not supported",
- (void *)dev, filter_type);
- break;
- }
- rte_errno = ENOTSUP;
- return -rte_errno;
-}
-
-static const struct eth_dev_ops mlx4_dev_ops = {
- .dev_configure = mlx4_dev_configure,
- .dev_start = mlx4_dev_start,
- .dev_stop = mlx4_dev_stop,
- .dev_set_link_down = mlx4_set_link_down,
- .dev_set_link_up = mlx4_set_link_up,
- .dev_close = mlx4_dev_close,
- .link_update = mlx4_link_update,
- .stats_get = mlx4_stats_get,
- .stats_reset = mlx4_stats_reset,
- .dev_infos_get = mlx4_dev_infos_get,
- .rx_queue_setup = mlx4_rx_queue_setup,
- .tx_queue_setup = mlx4_tx_queue_setup,
- .rx_queue_release = mlx4_rx_queue_release,
- .tx_queue_release = mlx4_tx_queue_release,
- .flow_ctrl_get = mlx4_dev_get_flow_ctrl,
- .flow_ctrl_set = mlx4_dev_set_flow_ctrl,
- .mtu_set = mlx4_dev_set_mtu,
- .filter_ctrl = mlx4_dev_filter_ctrl,
- .rx_queue_intr_enable = mlx4_rx_intr_enable,
- .rx_queue_intr_disable = mlx4_rx_intr_disable,
-};
-
-/**
- * Get PCI information from struct ibv_device.
- *
- * @param device
- * Pointer to Ethernet device structure.
- * @param[out] pci_addr
- * PCI bus address output buffer.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-mlx4_ibv_device_to_pci_addr(const struct ibv_device *device,
- struct rte_pci_addr *pci_addr)
-{
- FILE *file;
- char line[32];
- MKSTR(path, "%s/device/uevent", device->ibdev_path);
-
- file = fopen(path, "rb");
- if (file == NULL) {
- rte_errno = errno;
- return -rte_errno;
- }
- while (fgets(line, sizeof(line), file) == line) {
- size_t len = strlen(line);
- int ret;
-
- /* Truncate long lines. */
- if (len == (sizeof(line) - 1))
- while (line[(len - 1)] != '\n') {
- ret = fgetc(file);
- if (ret == EOF)
- break;
- line[(len - 1)] = ret;
- }
- /* Extract information. */
- if (sscanf(line,
- "PCI_SLOT_NAME="
- "%" SCNx32 ":%" SCNx8 ":%" SCNx8 ".%" SCNx8 "\n",
- &pci_addr->domain,
- &pci_addr->bus,
- &pci_addr->devid,
- &pci_addr->function) == 4) {
- ret = 0;
- break;
- }
- }
- fclose(file);
- return 0;
-}
-
-/**
- * Get MAC address by querying netdevice.
- *
- * @param[in] priv
- * struct priv for the requested device.
- * @param[out] mac
- * MAC address output buffer.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-priv_get_mac(struct priv *priv, uint8_t (*mac)[ETHER_ADDR_LEN])
-{
- struct ifreq request;
- int ret = priv_ifreq(priv, SIOCGIFHWADDR, &request);
-
- if (ret)
- return ret;
- memcpy(mac, request.ifr_hwaddr.sa_data, ETHER_ADDR_LEN);
- return 0;
-}
-
-static void mlx4_link_status_alarm(struct priv *priv);
-
-/**
- * Collect interrupt events.
- *
- * @param priv
- * Pointer to private structure.
- * @param events
- * Pointer to event flags holder.
- *
- * @return
- * Number of events.
- */
-static int
-priv_collect_interrupt_events(struct priv *priv, uint32_t *events)
-{
- struct ibv_async_event event;
- int port_change = 0;
- struct rte_eth_link *link = &priv->dev->data->dev_link;
- int ret = 0;
-
- *events = 0;
- /* Read all message and acknowledge them. */
- for (;;) {
- if (ibv_get_async_event(priv->ctx, &event))
- break;
- if ((event.event_type == IBV_EVENT_PORT_ACTIVE ||
- event.event_type == IBV_EVENT_PORT_ERR) &&
- (priv->intr_conf.lsc == 1)) {
- port_change = 1;
- ret++;
- } else if (event.event_type == IBV_EVENT_DEVICE_FATAL &&
- priv->intr_conf.rmv == 1) {
- *events |= (1 << RTE_ETH_EVENT_INTR_RMV);
- ret++;
- } else
- DEBUG("event type %d on port %d not handled",
- event.event_type, event.element.port_num);
- ibv_ack_async_event(&event);
- }
- if (!port_change)
- return ret;
- mlx4_link_update(priv->dev, 0);
- if (((link->link_speed == 0) && link->link_status) ||
- ((link->link_speed != 0) && !link->link_status)) {
- if (!priv->intr_alarm) {
- /* Inconsistent status, check again later. */
- priv->intr_alarm = 1;
- rte_eal_alarm_set(MLX4_INTR_ALARM_TIMEOUT,
- (void (*)(void *))
- mlx4_link_status_alarm,
- priv);
- }
- } else {
- *events |= (1 << RTE_ETH_EVENT_INTR_LSC);
- }
- return ret;
-}
-
-/**
- * Process scheduled link status check.
- *
- * @param priv
- * Pointer to private structure.
- */
-static void
-mlx4_link_status_alarm(struct priv *priv)
-{
- uint32_t events;
- int ret;
-
- assert(priv->intr_alarm == 1);
- priv->intr_alarm = 0;
- ret = priv_collect_interrupt_events(priv, &events);
- if (ret > 0 && events & (1 << RTE_ETH_EVENT_INTR_LSC))
- _rte_eth_dev_callback_process(priv->dev,
- RTE_ETH_EVENT_INTR_LSC,
- NULL, NULL);
-}
-
-/**
- * Handle interrupts from the NIC.
- *
- * @param priv
- * Pointer to private structure.
- */
-static void
-mlx4_interrupt_handler(struct priv *priv)
-{
- int ret;
- uint32_t ev;
- int i;
-
- ret = priv_collect_interrupt_events(priv, &ev);
- if (ret > 0) {
- for (i = RTE_ETH_EVENT_UNKNOWN;
- i < RTE_ETH_EVENT_MAX;
- i++) {
- if (ev & (1 << i)) {
- ev &= ~(1 << i);
- _rte_eth_dev_callback_process(priv->dev, i,
- NULL, NULL);
- ret--;
- }
- }
- if (ret)
- WARN("%d event%s not processed", ret,
- (ret > 1 ? "s were" : " was"));
- }
-}
-
-/**
- * Uninstall interrupt handler.
- *
- * @param priv
- * Pointer to private structure.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-priv_interrupt_handler_uninstall(struct priv *priv)
-{
- int ret;
-
- if (priv->intr_conf.lsc ||
- priv->intr_conf.rmv)
- return 0;
- ret = rte_intr_callback_unregister(&priv->intr_handle,
- (void (*)(void *))
- mlx4_interrupt_handler,
- priv);
- if (ret < 0) {
- rte_errno = ret;
- ERROR("rte_intr_callback_unregister failed with %d %s",
- ret, strerror(rte_errno));
- }
- priv->intr_handle.fd = -1;
- return ret;
-}
-
-/**
- * Install interrupt handler.
- *
- * @param priv
- * Pointer to private structure.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-priv_interrupt_handler_install(struct priv *priv)
-{
- int rc;
-
- /*
- * Check whether the interrupt handler has already been installed
- * for either type of interrupt.
- */
- if (priv->intr_conf.lsc &&
- priv->intr_conf.rmv &&
- priv->intr_handle.fd)
- return 0;
- priv->intr_handle.fd = priv->ctx->async_fd;
- rc = rte_intr_callback_register(&priv->intr_handle,
- (void (*)(void *))
- mlx4_interrupt_handler,
- priv);
- if (!rc)
- return 0;
- rte_errno = -rc;
- ERROR("rte_intr_callback_register failed (rte_errno: %s)",
- strerror(rte_errno));
- priv->intr_handle.fd = -1;
- return -rte_errno;
-}
-
-/**
- * Uninstall interrupt handler.
- *
- * @param priv
- * Pointer to private structure.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-priv_removal_interrupt_handler_uninstall(struct priv *priv)
-{
- if (priv->dev->data->dev_conf.intr_conf.rmv) {
- priv->intr_conf.rmv = 0;
- return priv_interrupt_handler_uninstall(priv);
- }
- return 0;
-}
-
-/**
- * Uninstall interrupt handler.
- *
- * @param priv
- * Pointer to private structure.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-priv_link_interrupt_handler_uninstall(struct priv *priv)
-{
- int ret = 0;
-
- if (priv->dev->data->dev_conf.intr_conf.lsc) {
- priv->intr_conf.lsc = 0;
- ret = priv_interrupt_handler_uninstall(priv);
- if (ret)
- return ret;
- }
- if (priv->intr_alarm)
- if (rte_eal_alarm_cancel((void (*)(void *))
- mlx4_link_status_alarm,
- priv)) {
- ERROR("rte_eal_alarm_cancel failed "
- " (rte_errno: %s)", strerror(rte_errno));
- return -rte_errno;
- }
- priv->intr_alarm = 0;
- return 0;
-}
-
-/**
- * Install link interrupt handler.
- *
- * @param priv
- * Pointer to private structure.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-priv_link_interrupt_handler_install(struct priv *priv)
-{
- int ret;
-
- if (priv->dev->data->dev_conf.intr_conf.lsc) {
- ret = priv_interrupt_handler_install(priv);
- if (ret)
- return ret;
- priv->intr_conf.lsc = 1;
- }
- return 0;
-}
-
-/**
- * Install removal interrupt handler.
- *
- * @param priv
- * Pointer to private structure.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-priv_removal_interrupt_handler_install(struct priv *priv)
-{
- int ret;
-
- if (priv->dev->data->dev_conf.intr_conf.rmv) {
- ret = priv_interrupt_handler_install(priv);
- if (ret)
- return ret;
- priv->intr_conf.rmv = 1;
- }
- return 0;
-}
-
-/**
- * Allocate queue vector and fill epoll fd list for Rx interrupts.
- *
- * @param priv
- * Pointer to private structure.
- *
- * @return
- * 0 on success, negative errno value otherwise and rte_errno is set.
- */
-static int
-priv_rx_intr_vec_enable(struct priv *priv)
-{
- unsigned int i;
- unsigned int rxqs_n = priv->rxqs_n;
- unsigned int n = RTE_MIN(rxqs_n, (uint32_t)RTE_MAX_RXTX_INTR_VEC_ID);
- unsigned int count = 0;
- struct rte_intr_handle *intr_handle = &priv->intr_handle;
-
- if (!priv->dev->data->dev_conf.intr_conf.rxq)
- return 0;
- priv_rx_intr_vec_disable(priv);
- intr_handle->intr_vec = malloc(sizeof(intr_handle->intr_vec[rxqs_n]));
- if (intr_handle->intr_vec == NULL) {
- rte_errno = ENOMEM;
- ERROR("failed to allocate memory for interrupt vector,"
- " Rx interrupts will not be supported");
- return -rte_errno;