--- zzzz-none-000/linux-3.10.107/drivers/net/ethernet/mellanox/mlx4/en_tx.c 2017-06-27 09:49:32.000000000 +0000 +++ scorpion-7490-727/linux-3.10.107/drivers/net/ethernet/mellanox/mlx4/en_tx.c 2021-02-04 17:41:59.000000000 +0000 @@ -37,53 +37,65 @@ #include #include #include +#include #include #include +#include #include #include "mlx4_en.h" -enum { - MAX_INLINE = 104, /* 128 - 16 - 4 - 4 */ - MAX_BF = 256, -}; - -static int inline_thold __read_mostly = MAX_INLINE; - -module_param_named(inline_thold, inline_thold, int, 0444); -MODULE_PARM_DESC(inline_thold, "threshold for using inline data"); - int mlx4_en_create_tx_ring(struct mlx4_en_priv *priv, - struct mlx4_en_tx_ring *ring, int qpn, u32 size, - u16 stride) + struct mlx4_en_tx_ring **pring, u32 size, + u16 stride, int node, int queue_index) { struct mlx4_en_dev *mdev = priv->mdev; + struct mlx4_en_tx_ring *ring; int tmp; int err; + ring = kzalloc_node(sizeof(*ring), GFP_KERNEL, node); + if (!ring) { + ring = kzalloc(sizeof(*ring), GFP_KERNEL); + if (!ring) { + en_err(priv, "Failed allocating TX ring\n"); + return -ENOMEM; + } + } + ring->size = size; ring->size_mask = size - 1; ring->stride = stride; - - inline_thold = min(inline_thold, MAX_INLINE); + ring->full_size = ring->size - HEADROOM - MAX_DESC_TXBBS; tmp = size * sizeof(struct mlx4_en_tx_info); - ring->tx_info = vmalloc(tmp); - if (!ring->tx_info) - return -ENOMEM; + ring->tx_info = kmalloc_node(tmp, GFP_KERNEL | __GFP_NOWARN, node); + if (!ring->tx_info) { + ring->tx_info = vmalloc(tmp); + if (!ring->tx_info) { + err = -ENOMEM; + goto err_ring; + } + } en_dbg(DRV, priv, "Allocated tx_info ring at addr:%p size:%d\n", ring->tx_info, tmp); - ring->bounce_buf = kmalloc(MAX_DESC_SIZE, GFP_KERNEL); + ring->bounce_buf = kmalloc_node(MAX_DESC_SIZE, GFP_KERNEL, node); if (!ring->bounce_buf) { - err = -ENOMEM; - goto err_tx; + ring->bounce_buf = kmalloc(MAX_DESC_SIZE, GFP_KERNEL); + if (!ring->bounce_buf) { + err = -ENOMEM; + goto err_info; + } } ring->buf_size = ALIGN(size * ring->stride, MLX4_EN_PAGE_SIZE); + /* Allocate HW buffers on provided NUMA node */ + set_dev_node(&mdev->dev->persist->pdev->dev, node); err = mlx4_alloc_hwq_res(mdev->dev, &ring->wqres, ring->buf_size, 2 * PAGE_SIZE); + set_dev_node(&mdev->dev->persist->pdev->dev, mdev->dev->numa_node); if (err) { en_err(priv, "Failed allocating hwq resources\n"); goto err_bounce; @@ -97,31 +109,51 @@ ring->buf = ring->wqres.buf.direct.buf; - en_dbg(DRV, priv, "Allocated TX ring (addr:%p) - buf:%p size:%d " - "buf_size:%d dma:%llx\n", ring, ring->buf, ring->size, - ring->buf_size, (unsigned long long) ring->wqres.buf.direct.map); + en_dbg(DRV, priv, "Allocated TX ring (addr:%p) - buf:%p size:%d buf_size:%d dma:%llx\n", + ring, ring->buf, ring->size, ring->buf_size, + (unsigned long long) ring->wqres.buf.direct.map); - ring->qpn = qpn; - err = mlx4_qp_alloc(mdev->dev, ring->qpn, &ring->qp); + err = mlx4_qp_reserve_range(mdev->dev, 1, 1, &ring->qpn, + MLX4_RESERVE_ETH_BF_QP); if (err) { - en_err(priv, "Failed allocating qp %d\n", ring->qpn); + en_err(priv, "failed reserving qp for TX ring\n"); goto err_map; } + + err = mlx4_qp_alloc(mdev->dev, ring->qpn, &ring->qp, GFP_KERNEL); + if (err) { + en_err(priv, "Failed allocating qp %d\n", ring->qpn); + goto err_reserve; + } ring->qp.event = mlx4_en_sqp_event; - err = mlx4_bf_alloc(mdev->dev, &ring->bf); + err = mlx4_bf_alloc(mdev->dev, &ring->bf, node); if (err) { - en_dbg(DRV, priv, "working without blueflame (%d)", err); + en_dbg(DRV, priv, "working without blueflame (%d)\n", err); ring->bf.uar = &mdev->priv_uar; ring->bf.uar->map = mdev->uar_map; ring->bf_enabled = false; - } else - ring->bf_enabled = true; + ring->bf_alloced = false; + priv->pflags &= ~MLX4_EN_PRIV_FLAGS_BLUEFLAME; + } else { + ring->bf_alloced = true; + ring->bf_enabled = !!(priv->pflags & + MLX4_EN_PRIV_FLAGS_BLUEFLAME); + } ring->hwtstamp_tx_type = priv->hwtstamp_config.tx_type; + ring->queue_index = queue_index; + + if (queue_index < priv->num_tx_rings_p_up) + cpumask_set_cpu(cpumask_local_spread(queue_index, + priv->mdev->dev->numa_node), + &ring->affinity_mask); + *pring = ring; return 0; +err_reserve: + mlx4_qp_release_range(mdev->dev, ring->qpn, 1); err_map: mlx4_en_unmap_buffer(&ring->wqres.buf); err_hwq_res: @@ -129,28 +161,35 @@ err_bounce: kfree(ring->bounce_buf); ring->bounce_buf = NULL; -err_tx: - vfree(ring->tx_info); +err_info: + kvfree(ring->tx_info); ring->tx_info = NULL; +err_ring: + kfree(ring); + *pring = NULL; return err; } void mlx4_en_destroy_tx_ring(struct mlx4_en_priv *priv, - struct mlx4_en_tx_ring *ring) + struct mlx4_en_tx_ring **pring) { struct mlx4_en_dev *mdev = priv->mdev; + struct mlx4_en_tx_ring *ring = *pring; en_dbg(DRV, priv, "Destroying tx ring, qpn: %d\n", ring->qpn); - if (ring->bf_enabled) + if (ring->bf_alloced) mlx4_bf_free(mdev->dev, &ring->bf); mlx4_qp_remove(mdev->dev, &ring->qp); mlx4_qp_free(mdev->dev, &ring->qp); + mlx4_qp_release_range(priv->mdev->dev, ring->qpn, 1); mlx4_en_unmap_buffer(&ring->wqres.buf); mlx4_free_hwq_res(mdev->dev, &ring->wqres, ring->buf_size); kfree(ring->bounce_buf); ring->bounce_buf = NULL; - vfree(ring->tx_info); + kvfree(ring->tx_info); ring->tx_info = NULL; + kfree(ring); + *pring = NULL; } int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv, @@ -164,20 +203,23 @@ ring->prod = 0; ring->cons = 0xffffffff; ring->last_nr_txbb = 1; - ring->poll_cnt = 0; memset(ring->tx_info, 0, ring->size * sizeof(struct mlx4_en_tx_info)); memset(ring->buf, 0, ring->buf_size); ring->qp_state = MLX4_QP_STATE_RST; - ring->doorbell_qpn = ring->qp.qpn << 8; + ring->doorbell_qpn = cpu_to_be32(ring->qp.qpn << 8); + ring->mr_key = cpu_to_be32(mdev->mr.key); mlx4_en_fill_qp_context(priv, ring->size, ring->stride, 1, 0, ring->qpn, ring->cqn, user_prio, &ring->context); - if (ring->bf_enabled) + if (ring->bf_alloced) ring->context.usr_page = cpu_to_be32(ring->bf.uar->index); err = mlx4_qp_to_ready(mdev->dev, &ring->wqres.mtt, &ring->context, &ring->qp, &ring->qp_state); + if (!cpumask_empty(&ring->affinity_mask)) + netif_set_xps_queue(priv->dev, &ring->affinity_mask, + ring->queue_index); return err; } @@ -191,6 +233,11 @@ MLX4_QP_STATE_RST, NULL, 0, 0, &ring->qp); } +static inline bool mlx4_en_is_tx_ring_full(struct mlx4_en_tx_ring *ring) +{ + return ring->prod - ring->cons > ring->full_size; +} + static void mlx4_en_stamp_wqe(struct mlx4_en_priv *priv, struct mlx4_en_tx_ring *ring, int index, u8 owner) @@ -229,38 +276,45 @@ struct mlx4_en_tx_ring *ring, int index, u8 owner, u64 timestamp) { - struct mlx4_en_dev *mdev = priv->mdev; struct mlx4_en_tx_info *tx_info = &ring->tx_info[index]; struct mlx4_en_tx_desc *tx_desc = ring->buf + index * TXBB_SIZE; struct mlx4_wqe_data_seg *data = (void *) tx_desc + tx_info->data_offset; - struct sk_buff *skb = tx_info->skb; - struct skb_frag_struct *frag; void *end = ring->buf + ring->buf_size; - int frags = skb_shinfo(skb)->nr_frags; + struct sk_buff *skb = tx_info->skb; + int nr_maps = tx_info->nr_maps; int i; - struct skb_shared_hwtstamps hwts; - if (timestamp) { - mlx4_en_fill_hwtstamps(mdev, &hwts, timestamp); + /* We do not touch skb here, so prefetch skb->users location + * to speedup consume_skb() + */ + prefetchw(&skb->users); + + if (unlikely(timestamp)) { + struct skb_shared_hwtstamps hwts; + + mlx4_en_fill_hwtstamps(priv->mdev, &hwts, timestamp); skb_tstamp_tx(skb, &hwts); } /* Optimize the common case when there are no wraparounds */ if (likely((void *) tx_desc + tx_info->nr_txbb * TXBB_SIZE <= end)) { if (!tx_info->inl) { - if (tx_info->linear) { + if (tx_info->linear) dma_unmap_single(priv->ddev, - (dma_addr_t) be64_to_cpu(data->addr), - be32_to_cpu(data->byte_count), - PCI_DMA_TODEVICE); - ++data; - } - - for (i = 0; i < frags; i++) { - frag = &skb_shinfo(skb)->frags[i]; + tx_info->map0_dma, + tx_info->map0_byte_count, + PCI_DMA_TODEVICE); + else + dma_unmap_page(priv->ddev, + tx_info->map0_dma, + tx_info->map0_byte_count, + PCI_DMA_TODEVICE); + for (i = 1; i < nr_maps; i++) { + data++; dma_unmap_page(priv->ddev, - (dma_addr_t) be64_to_cpu(data[i].addr), - skb_frag_size(frag), PCI_DMA_TODEVICE); + (dma_addr_t)be64_to_cpu(data->addr), + be32_to_cpu(data->byte_count), + PCI_DMA_TODEVICE); } } } else { @@ -269,27 +323,29 @@ data = ring->buf + ((void *)data - end); } - if (tx_info->linear) { + if (tx_info->linear) dma_unmap_single(priv->ddev, - (dma_addr_t) be64_to_cpu(data->addr), - be32_to_cpu(data->byte_count), - PCI_DMA_TODEVICE); - ++data; - } - - for (i = 0; i < frags; i++) { + tx_info->map0_dma, + tx_info->map0_byte_count, + PCI_DMA_TODEVICE); + else + dma_unmap_page(priv->ddev, + tx_info->map0_dma, + tx_info->map0_byte_count, + PCI_DMA_TODEVICE); + for (i = 1; i < nr_maps; i++) { + data++; /* Check for wraparound before unmapping */ if ((void *) data >= end) data = ring->buf; - frag = &skb_shinfo(skb)->frags[i]; dma_unmap_page(priv->ddev, - (dma_addr_t) be64_to_cpu(data->addr), - skb_frag_size(frag), PCI_DMA_TODEVICE); - ++data; + (dma_addr_t)be64_to_cpu(data->addr), + be32_to_cpu(data->byte_count), + PCI_DMA_TODEVICE); } } } - dev_kfree_skb_any(skb); + dev_consume_skb_any(skb); return tx_info->nr_txbb; } @@ -326,11 +382,12 @@ return cnt; } -static void mlx4_en_process_tx_cq(struct net_device *dev, struct mlx4_en_cq *cq) +static bool mlx4_en_process_tx_cq(struct net_device *dev, + struct mlx4_en_cq *cq) { struct mlx4_en_priv *priv = netdev_priv(dev); struct mlx4_cq *mcq = &cq->mcq; - struct mlx4_en_tx_ring *ring = &priv->tx_ring[cq->ring]; + struct mlx4_en_tx_ring *ring = priv->tx_ring[cq->ring]; struct mlx4_cqe *cqe; u16 index; u16 new_index, ring_index, stamp_index; @@ -343,52 +400,71 @@ u32 packets = 0; u32 bytes = 0; int factor = priv->cqe_factor; - u64 timestamp = 0; + int done = 0; + int budget = priv->tx_work_limit; + u32 last_nr_txbb; + u32 ring_cons; if (!priv->port_up) - return; + return true; + + netdev_txq_bql_complete_prefetchw(ring->tx_queue); index = cons_index & size_mask; - cqe = &buf[(index << factor) + factor]; - ring_index = ring->cons & size_mask; + cqe = mlx4_en_get_cqe(buf, index, priv->cqe_size) + factor; + last_nr_txbb = ACCESS_ONCE(ring->last_nr_txbb); + ring_cons = ACCESS_ONCE(ring->cons); + ring_index = ring_cons & size_mask; stamp_index = ring_index; /* Process all completed CQEs */ while (XNOR(cqe->owner_sr_opcode & MLX4_CQE_OWNER_MASK, - cons_index & size)) { + cons_index & size) && (done < budget)) { /* * make sure we read the CQE after we read the * ownership bit */ - rmb(); + dma_rmb(); + + if (unlikely((cqe->owner_sr_opcode & MLX4_CQE_OPCODE_MASK) == + MLX4_CQE_OPCODE_ERROR)) { + struct mlx4_err_cqe *cqe_err = (struct mlx4_err_cqe *)cqe; + + en_err(priv, "CQE error - vendor syndrome: 0x%x syndrome: 0x%x\n", + cqe_err->vendor_err_syndrome, + cqe_err->syndrome); + } /* Skip over last polled CQE */ new_index = be16_to_cpu(cqe->wqe_index) & size_mask; do { - txbbs_skipped += ring->last_nr_txbb; - ring_index = (ring_index + ring->last_nr_txbb) & size_mask; - if (ring->tx_info[ring_index].ts_requested) + u64 timestamp = 0; + + txbbs_skipped += last_nr_txbb; + ring_index = (ring_index + last_nr_txbb) & size_mask; + + if (unlikely(ring->tx_info[ring_index].ts_requested)) timestamp = mlx4_en_get_cqe_ts(cqe); /* free next descriptor */ - ring->last_nr_txbb = mlx4_en_free_tx_desc( + last_nr_txbb = mlx4_en_free_tx_desc( priv, ring, ring_index, - !!((ring->cons + txbbs_skipped) & + !!((ring_cons + txbbs_skipped) & ring->size), timestamp); mlx4_en_stamp_wqe(priv, ring, stamp_index, - !!((ring->cons + txbbs_stamp) & + !!((ring_cons + txbbs_stamp) & ring->size)); stamp_index = ring_index; txbbs_stamp = txbbs_skipped; packets++; bytes += ring->tx_info[ring_index].nr_bytes; - } while (ring_index != new_index); + } while ((++done < budget) && (ring_index != new_index)); ++cons_index; index = cons_index & size_mask; - cqe = &buf[(index << factor) + factor]; + cqe = mlx4_en_get_cqe(buf, index, priv->cqe_size) + factor; } @@ -399,17 +475,21 @@ mcq->cons_index = cons_index; mlx4_cq_set_ci(mcq); wmb(); - ring->cons += txbbs_skipped; + + /* we want to dirty this cache line once */ + ACCESS_ONCE(ring->last_nr_txbb) = last_nr_txbb; + ACCESS_ONCE(ring->cons) = ring_cons + txbbs_skipped; + netdev_tx_completed_queue(ring->tx_queue, packets, bytes); - /* - * Wakeup Tx queue if this stopped, and at least 1 packet - * was completed + /* Wakeup Tx queue if this stopped, and ring is not full. */ - if (netif_tx_queue_stopped(ring->tx_queue) && txbbs_skipped > 0) { + if (netif_tx_queue_stopped(ring->tx_queue) && + !mlx4_en_is_tx_ring_full(ring)) { netif_tx_wake_queue(ring->tx_queue); - priv->port_stats.wake_queue++; + ring->wake_queue++; } + return done < budget; } void mlx4_en_tx_irq(struct mlx4_cq *mcq) @@ -417,10 +497,29 @@ struct mlx4_en_cq *cq = container_of(mcq, struct mlx4_en_cq, mcq); struct mlx4_en_priv *priv = netdev_priv(cq->dev); - mlx4_en_process_tx_cq(cq->dev, cq); - mlx4_en_arm_cq(priv, cq); + if (likely(priv->port_up)) + napi_schedule_irqoff(&cq->napi); + else + mlx4_en_arm_cq(priv, cq); } +/* TX CQ polling - called by NAPI */ +int mlx4_en_poll_tx_cq(struct napi_struct *napi, int budget) +{ + struct mlx4_en_cq *cq = container_of(napi, struct mlx4_en_cq, napi); + struct net_device *dev = cq->dev; + struct mlx4_en_priv *priv = netdev_priv(dev); + int clean_complete; + + clean_complete = mlx4_en_process_tx_cq(dev, cq); + if (!clean_complete) + return budget; + + napi_complete(napi); + mlx4_en_arm_cq(priv, cq); + + return 0; +} static struct mlx4_en_tx_desc *mlx4_en_bounce_to_desc(struct mlx4_en_priv *priv, struct mlx4_en_tx_ring *ring, @@ -450,30 +549,35 @@ return ring->buf + index * TXBB_SIZE; } -static int is_inline(struct sk_buff *skb, void **pfrag) +/* Decide if skb can be inlined in tx descriptor to avoid dma mapping + * + * It seems strange we do not simply use skb_copy_bits(). + * This would allow to inline all skbs iff skb->len <= inline_thold + * + * Note that caller already checked skb was not a gso packet + */ +static bool is_inline(int inline_thold, const struct sk_buff *skb, + const struct skb_shared_info *shinfo, + void **pfrag) { void *ptr; - if (inline_thold && !skb_is_gso(skb) && skb->len <= inline_thold) { - if (skb_shinfo(skb)->nr_frags == 1) { - ptr = skb_frag_address_safe(&skb_shinfo(skb)->frags[0]); - if (unlikely(!ptr)) - return 0; - - if (pfrag) - *pfrag = ptr; - - return 1; - } else if (unlikely(skb_shinfo(skb)->nr_frags)) - return 0; - else - return 1; - } + if (skb->len > inline_thold || !inline_thold) + return false; - return 0; + if (shinfo->nr_frags == 1) { + ptr = skb_frag_address_safe(&shinfo->frags[0]); + if (unlikely(!ptr)) + return false; + *pfrag = ptr; + return true; + } + if (shinfo->nr_frags) + return false; + return true; } -static int inline_size(struct sk_buff *skb) +static int inline_size(const struct sk_buff *skb) { if (skb->len + CTRL_SIZE + sizeof(struct mlx4_wqe_inline_seg) <= MLX4_INLINE_ALIGN) @@ -484,15 +588,23 @@ sizeof(struct mlx4_wqe_inline_seg), 16); } -static int get_real_size(struct sk_buff *skb, struct net_device *dev, - int *lso_header_size) +static int get_real_size(const struct sk_buff *skb, + const struct skb_shared_info *shinfo, + struct net_device *dev, + int *lso_header_size, + bool *inline_ok, + void **pfrag) { struct mlx4_en_priv *priv = netdev_priv(dev); int real_size; - if (skb_is_gso(skb)) { - *lso_header_size = skb_transport_offset(skb) + tcp_hdrlen(skb); - real_size = CTRL_SIZE + skb_shinfo(skb)->nr_frags * DS_SIZE + + if (shinfo->gso_size) { + *inline_ok = false; + if (skb->encapsulation) + *lso_header_size = (skb_inner_transport_header(skb) - skb->data) + inner_tcp_hdrlen(skb); + else + *lso_header_size = skb_transport_offset(skb) + tcp_hdrlen(skb); + real_size = CTRL_SIZE + shinfo->nr_frags * DS_SIZE + ALIGN(*lso_header_size + 4, DS_SIZE); if (unlikely(*lso_header_size != skb_headlen(skb))) { /* We add a segment for the skb linear buffer only if @@ -507,36 +619,50 @@ } } else { *lso_header_size = 0; - if (!is_inline(skb, NULL)) - real_size = CTRL_SIZE + (skb_shinfo(skb)->nr_frags + 1) * DS_SIZE; - else + *inline_ok = is_inline(priv->prof->inline_thold, skb, + shinfo, pfrag); + + if (*inline_ok) real_size = inline_size(skb); + else + real_size = CTRL_SIZE + + (shinfo->nr_frags + 1) * DS_SIZE; } return real_size; } -static void build_inline_wqe(struct mlx4_en_tx_desc *tx_desc, struct sk_buff *skb, - int real_size, u16 *vlan_tag, int tx_ind, void *fragptr) +static void build_inline_wqe(struct mlx4_en_tx_desc *tx_desc, + const struct sk_buff *skb, + const struct skb_shared_info *shinfo, + int real_size, u16 *vlan_tag, + int tx_ind, void *fragptr) { struct mlx4_wqe_inline_seg *inl = &tx_desc->inl; int spc = MLX4_INLINE_ALIGN - CTRL_SIZE - sizeof *inl; + unsigned int hlen = skb_headlen(skb); if (skb->len <= spc) { - inl->byte_count = cpu_to_be32(1 << 31 | skb->len); - skb_copy_from_linear_data(skb, inl + 1, skb_headlen(skb)); - if (skb_shinfo(skb)->nr_frags) - memcpy(((void *)(inl + 1)) + skb_headlen(skb), fragptr, - skb_frag_size(&skb_shinfo(skb)->frags[0])); + if (likely(skb->len >= MIN_PKT_LEN)) { + inl->byte_count = cpu_to_be32(1 << 31 | skb->len); + } else { + inl->byte_count = cpu_to_be32(1 << 31 | MIN_PKT_LEN); + memset(((void *)(inl + 1)) + skb->len, 0, + MIN_PKT_LEN - skb->len); + } + skb_copy_from_linear_data(skb, inl + 1, hlen); + if (shinfo->nr_frags) + memcpy(((void *)(inl + 1)) + hlen, fragptr, + skb_frag_size(&shinfo->frags[0])); } else { inl->byte_count = cpu_to_be32(1 << 31 | spc); - if (skb_headlen(skb) <= spc) { - skb_copy_from_linear_data(skb, inl + 1, skb_headlen(skb)); - if (skb_headlen(skb) < spc) { - memcpy(((void *)(inl + 1)) + skb_headlen(skb), - fragptr, spc - skb_headlen(skb)); - fragptr += spc - skb_headlen(skb); + if (hlen <= spc) { + skb_copy_from_linear_data(skb, inl + 1, hlen); + if (hlen < spc) { + memcpy(((void *)(inl + 1)) + hlen, + fragptr, spc - hlen); + fragptr += spc - hlen; } inl = (void *) (inl + 1) + spc; memcpy(((void *)(inl + 1)), fragptr, skb->len - spc); @@ -544,18 +670,20 @@ skb_copy_from_linear_data(skb, inl + 1, spc); inl = (void *) (inl + 1) + spc; skb_copy_from_linear_data_offset(skb, spc, inl + 1, - skb_headlen(skb) - spc); - if (skb_shinfo(skb)->nr_frags) - memcpy(((void *)(inl + 1)) + skb_headlen(skb) - spc, - fragptr, skb_frag_size(&skb_shinfo(skb)->frags[0])); + hlen - spc); + if (shinfo->nr_frags) + memcpy(((void *)(inl + 1)) + hlen - spc, + fragptr, + skb_frag_size(&shinfo->frags[0])); } - wmb(); + dma_wmb(); inl->byte_count = cpu_to_be32(1 << 31 | (skb->len - spc)); } } -u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb) +u16 mlx4_en_select_queue(struct net_device *dev, struct sk_buff *skb, + void *accel_priv, select_queue_fallback_t fallback) { struct mlx4_en_priv *priv = netdev_priv(dev); u16 rings_p_up = priv->num_tx_rings_p_up; @@ -564,44 +692,55 @@ if (dev->num_tc) return skb_tx_hash(dev, skb); - if (vlan_tx_tag_present(skb)) - up = vlan_tx_tag_get(skb) >> VLAN_PRIO_SHIFT; + if (skb_vlan_tag_present(skb)) + up = skb_vlan_tag_get(skb) >> VLAN_PRIO_SHIFT; - return __skb_tx_hash(dev, skb, rings_p_up) + up * rings_p_up; + return fallback(dev, skb) % rings_p_up + up * rings_p_up; } -static void mlx4_bf_copy(void __iomem *dst, unsigned long *src, unsigned bytecnt) +static void mlx4_bf_copy(void __iomem *dst, const void *src, + unsigned int bytecnt) { __iowrite64_copy(dst, src, bytecnt / 8); } netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) { + struct skb_shared_info *shinfo = skb_shinfo(skb); struct mlx4_en_priv *priv = netdev_priv(dev); - struct mlx4_en_dev *mdev = priv->mdev; + struct device *ddev = priv->ddev; struct mlx4_en_tx_ring *ring; struct mlx4_en_tx_desc *tx_desc; struct mlx4_wqe_data_seg *data; - struct skb_frag_struct *frag; struct mlx4_en_tx_info *tx_info; - struct ethhdr *ethh; int tx_ind = 0; int nr_txbb; int desc_size; int real_size; - dma_addr_t dma; u32 index, bf_index; __be32 op_own; u16 vlan_tag = 0; - int i; + u16 vlan_proto = 0; + int i_frag; int lso_header_size; - void *fragptr; + void *fragptr = NULL; bool bounce = false; + bool send_doorbell; + bool stop_queue; + bool inline_ok; + u32 ring_cons; if (!priv->port_up) goto tx_drop; - real_size = get_real_size(skb, dev, &lso_header_size); + tx_ind = skb_get_queue_mapping(skb); + ring = priv->tx_ring[tx_ind]; + + /* fetch ring->cons far ahead before needing it to avoid stall */ + ring_cons = ACCESS_ONCE(ring->cons); + + real_size = get_real_size(skb, shinfo, dev, &lso_header_size, + &inline_ok, &fragptr); if (unlikely(!real_size)) goto tx_drop; @@ -614,38 +753,16 @@ goto tx_drop; } - tx_ind = skb->queue_mapping; - ring = &priv->tx_ring[tx_ind]; - if (vlan_tx_tag_present(skb)) - vlan_tag = vlan_tx_tag_get(skb); - - /* Check available TXBBs And 2K spare for prefetch */ - if (unlikely(((int)(ring->prod - ring->cons)) > - ring->size - HEADROOM - MAX_DESC_TXBBS)) { - /* every full Tx ring stops queue */ - netif_tx_stop_queue(ring->tx_queue); - priv->port_stats.queue_stopped++; - - /* If queue was emptied after the if, and before the - * stop_queue - need to wake the queue, or else it will remain - * stopped forever. - * Need a memory barrier to make sure ring->cons was not - * updated before queue was stopped. - */ - wmb(); - - if (unlikely(((int)(ring->prod - ring->cons)) <= - ring->size - HEADROOM - MAX_DESC_TXBBS)) { - netif_tx_wake_queue(ring->tx_queue); - priv->port_stats.wake_queue++; - } else { - return NETDEV_TX_BUSY; - } + if (skb_vlan_tag_present(skb)) { + vlan_tag = skb_vlan_tag_get(skb); + vlan_proto = be16_to_cpu(skb->vlan_proto); } + netdev_txq_bql_enqueue_prefetchw(ring->tx_queue); + /* Track current inflight packets for performance analysis */ AVG_PERF_COUNTER(priv->pstats.inflight_avg, - (u32) (ring->prod - ring->cons - 1)); + (u32)(ring->prod - ring_cons - 1)); /* Packet is good - grab an index and transmit it */ index = ring->prod & ring->size_mask; @@ -665,30 +782,91 @@ tx_info->skb = skb; tx_info->nr_txbb = nr_txbb; + data = &tx_desc->data; + if (lso_header_size) + data = ((void *)&tx_desc->lso + ALIGN(lso_header_size + 4, + DS_SIZE)); + + /* valid only for none inline segments */ + tx_info->data_offset = (void *)data - (void *)tx_desc; + + tx_info->inl = inline_ok; + + tx_info->linear = (lso_header_size < skb_headlen(skb) && + !inline_ok) ? 1 : 0; + + tx_info->nr_maps = shinfo->nr_frags + tx_info->linear; + data += tx_info->nr_maps - 1; + + if (!tx_info->inl) { + dma_addr_t dma = 0; + u32 byte_count = 0; + + /* Map fragments if any */ + for (i_frag = shinfo->nr_frags - 1; i_frag >= 0; i_frag--) { + const struct skb_frag_struct *frag; + + frag = &shinfo->frags[i_frag]; + byte_count = skb_frag_size(frag); + dma = skb_frag_dma_map(ddev, frag, + 0, byte_count, + DMA_TO_DEVICE); + if (dma_mapping_error(ddev, dma)) + goto tx_drop_unmap; + + data->addr = cpu_to_be64(dma); + data->lkey = ring->mr_key; + dma_wmb(); + data->byte_count = cpu_to_be32(byte_count); + --data; + } + + /* Map linear part if needed */ + if (tx_info->linear) { + byte_count = skb_headlen(skb) - lso_header_size; + + dma = dma_map_single(ddev, skb->data + + lso_header_size, byte_count, + PCI_DMA_TODEVICE); + if (dma_mapping_error(ddev, dma)) + goto tx_drop_unmap; + + data->addr = cpu_to_be64(dma); + data->lkey = ring->mr_key; + dma_wmb(); + data->byte_count = cpu_to_be32(byte_count); + } + /* tx completion can avoid cache line miss for common cases */ + tx_info->map0_dma = dma; + tx_info->map0_byte_count = byte_count; + } + /* * For timestamping add flag to skb_shinfo and * set flag for further reference */ - if (ring->hwtstamp_tx_type == HWTSTAMP_TX_ON && - skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) { - skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS; + tx_info->ts_requested = 0; + if (unlikely(ring->hwtstamp_tx_type == HWTSTAMP_TX_ON && + shinfo->tx_flags & SKBTX_HW_TSTAMP)) { + shinfo->tx_flags |= SKBTX_IN_PROGRESS; tx_info->ts_requested = 1; } /* Prepare ctrl segement apart opcode+ownership, which depends on * whether LSO is used */ - tx_desc->ctrl.vlan_tag = cpu_to_be16(vlan_tag); - tx_desc->ctrl.ins_vlan = MLX4_WQE_CTRL_INS_VLAN * - !!vlan_tx_tag_present(skb); - tx_desc->ctrl.fence_size = (real_size / 16) & 0x3f; tx_desc->ctrl.srcrb_flags = priv->ctrl_flags; if (likely(skb->ip_summed == CHECKSUM_PARTIAL)) { - tx_desc->ctrl.srcrb_flags |= cpu_to_be32(MLX4_WQE_CTRL_IP_CSUM | - MLX4_WQE_CTRL_TCP_UDP_CSUM); + if (!skb->encapsulation) + tx_desc->ctrl.srcrb_flags |= cpu_to_be32(MLX4_WQE_CTRL_IP_CSUM | + MLX4_WQE_CTRL_TCP_UDP_CSUM); + else + tx_desc->ctrl.srcrb_flags |= cpu_to_be32(MLX4_WQE_CTRL_IP_CSUM); ring->tx_csum++; } if (priv->flags & MLX4_EN_FLAG_ENABLE_HW_LOOPBACK) { + struct ethhdr *ethh; + /* Copy dst mac address to wqe. This allows loopback in eSwitch, * so that VFs and PF can communicate with each other */ @@ -699,6 +877,8 @@ /* Handle LSO (TSO) packets */ if (lso_header_size) { + int i; + /* Mark opcode as LSO */ op_own = cpu_to_be32(MLX4_OPCODE_LSO | (1 << 6)) | ((ring->prod & ring->size) ? @@ -706,17 +886,16 @@ /* Fill in the LSO prefix */ tx_desc->lso.mss_hdr_size = cpu_to_be32( - skb_shinfo(skb)->gso_size << 16 | lso_header_size); + shinfo->gso_size << 16 | lso_header_size); /* Copy headers; * note that we already verified that it is linear */ memcpy(tx_desc->lso.header, skb->data, lso_header_size); - data = ((void *) &tx_desc->lso + - ALIGN(lso_header_size + 4, DS_SIZE)); - priv->port_stats.tso_packets++; - i = ((skb->len - lso_header_size) / skb_shinfo(skb)->gso_size) + - !!((skb->len - lso_header_size) % skb_shinfo(skb)->gso_size); + ring->tso_packets++; + + i = ((skb->len - lso_header_size) / shinfo->gso_size) + + !!((skb->len - lso_header_size) % shinfo->gso_size); tx_info->nr_bytes = skb->len + (i - 1) * lso_header_size; ring->packets += i; } else { @@ -724,86 +903,125 @@ op_own = cpu_to_be32(MLX4_OPCODE_SEND) | ((ring->prod & ring->size) ? cpu_to_be32(MLX4_EN_BIT_DESC_OWN) : 0); - data = &tx_desc->data; tx_info->nr_bytes = max_t(unsigned int, skb->len, ETH_ZLEN); ring->packets++; - } ring->bytes += tx_info->nr_bytes; netdev_tx_sent_queue(ring->tx_queue, tx_info->nr_bytes); AVG_PERF_COUNTER(priv->pstats.tx_pktsz_avg, skb->len); - - /* valid only for none inline segments */ - tx_info->data_offset = (void *) data - (void *) tx_desc; - - tx_info->linear = (lso_header_size < skb_headlen(skb) && !is_inline(skb, NULL)) ? 1 : 0; - data += skb_shinfo(skb)->nr_frags + tx_info->linear - 1; - - if (!is_inline(skb, &fragptr)) { - /* Map fragments */ - for (i = skb_shinfo(skb)->nr_frags - 1; i >= 0; i--) { - frag = &skb_shinfo(skb)->frags[i]; - dma = skb_frag_dma_map(priv->ddev, frag, - 0, skb_frag_size(frag), - DMA_TO_DEVICE); - data->addr = cpu_to_be64(dma); - data->lkey = cpu_to_be32(mdev->mr.key); - wmb(); - data->byte_count = cpu_to_be32(skb_frag_size(frag)); - --data; - } - - /* Map linear part */ - if (tx_info->linear) { - dma = dma_map_single(priv->ddev, skb->data + lso_header_size, - skb_headlen(skb) - lso_header_size, PCI_DMA_TODEVICE); - data->addr = cpu_to_be64(dma); - data->lkey = cpu_to_be32(mdev->mr.key); - wmb(); - data->byte_count = cpu_to_be32(skb_headlen(skb) - lso_header_size); - } - tx_info->inl = 0; - } else { - build_inline_wqe(tx_desc, skb, real_size, &vlan_tag, tx_ind, fragptr); - tx_info->inl = 1; + if (tx_info->inl) + build_inline_wqe(tx_desc, skb, shinfo, real_size, &vlan_tag, + tx_ind, fragptr); + + if (skb->encapsulation) { + struct iphdr *ipv4 = (struct iphdr *)skb_inner_network_header(skb); + if (ipv4->protocol == IPPROTO_TCP || ipv4->protocol == IPPROTO_UDP) + op_own |= cpu_to_be32(MLX4_WQE_CTRL_IIP | MLX4_WQE_CTRL_ILP); + else + op_own |= cpu_to_be32(MLX4_WQE_CTRL_IIP); } ring->prod += nr_txbb; /* If we used a bounce buffer then copy descriptor back into place */ - if (bounce) + if (unlikely(bounce)) tx_desc = mlx4_en_bounce_to_desc(priv, ring, index, desc_size); skb_tx_timestamp(skb); - if (ring->bf_enabled && desc_size <= MAX_BF && !bounce && !vlan_tx_tag_present(skb)) { - *(__be32 *) (&tx_desc->ctrl.vlan_tag) |= cpu_to_be32(ring->doorbell_qpn); + /* Check available TXBBs And 2K spare for prefetch */ + stop_queue = mlx4_en_is_tx_ring_full(ring); + if (unlikely(stop_queue)) { + netif_tx_stop_queue(ring->tx_queue); + ring->queue_stopped++; + } + send_doorbell = !skb->xmit_more || netif_xmit_stopped(ring->tx_queue); + + real_size = (real_size / 16) & 0x3f; + + if (ring->bf_enabled && desc_size <= MAX_BF && !bounce && + !skb_vlan_tag_present(skb) && send_doorbell) { + tx_desc->ctrl.bf_qpn = ring->doorbell_qpn | + cpu_to_be32(real_size); + op_own |= htonl((bf_index & 0xffff) << 8); - /* Ensure new descirptor hits memory - * before setting ownership of this descriptor to HW */ - wmb(); + /* Ensure new descriptor hits memory + * before setting ownership of this descriptor to HW + */ + dma_wmb(); tx_desc->ctrl.owner_opcode = op_own; wmb(); - mlx4_bf_copy(ring->bf.reg + ring->bf.offset, (unsigned long *) &tx_desc->ctrl, - desc_size); + mlx4_bf_copy(ring->bf.reg + ring->bf.offset, &tx_desc->ctrl, + desc_size); wmb(); ring->bf.offset ^= ring->bf.buf_size; } else { - /* Ensure new descirptor hits memory - * before setting ownership of this descriptor to HW */ - wmb(); + tx_desc->ctrl.vlan_tag = cpu_to_be16(vlan_tag); + if (vlan_proto == ETH_P_8021AD) + tx_desc->ctrl.ins_vlan = MLX4_WQE_CTRL_INS_SVLAN; + else if (vlan_proto == ETH_P_8021Q) + tx_desc->ctrl.ins_vlan = MLX4_WQE_CTRL_INS_CVLAN; + else + tx_desc->ctrl.ins_vlan = 0; + + tx_desc->ctrl.fence_size = real_size; + + /* Ensure new descriptor hits memory + * before setting ownership of this descriptor to HW + */ + dma_wmb(); tx_desc->ctrl.owner_opcode = op_own; - wmb(); - iowrite32be(ring->doorbell_qpn, ring->bf.uar->map + MLX4_SEND_DOORBELL); + if (send_doorbell) { + wmb(); + /* Since there is no iowrite*_native() that writes the + * value as is, without byteswapping - using the one + * the doesn't do byteswapping in the relevant arch + * endianness. + */ +#if defined(__LITTLE_ENDIAN) + iowrite32( +#else + iowrite32be( +#endif + ring->doorbell_qpn, + ring->bf.uar->map + MLX4_SEND_DOORBELL); + } else { + ring->xmit_more++; + } } + if (unlikely(stop_queue)) { + /* If queue was emptied after the if (stop_queue) , and before + * the netif_tx_stop_queue() - need to wake the queue, + * or else it will remain stopped forever. + * Need a memory barrier to make sure ring->cons was not + * updated before queue was stopped. + */ + smp_rmb(); + + ring_cons = ACCESS_ONCE(ring->cons); + if (unlikely(!mlx4_en_is_tx_ring_full(ring))) { + netif_tx_wake_queue(ring->tx_queue); + ring->wake_queue++; + } + } return NETDEV_TX_OK; +tx_drop_unmap: + en_err(priv, "DMA mapping error\n"); + + while (++i_frag < shinfo->nr_frags) { + ++data; + dma_unmap_page(ddev, (dma_addr_t) be64_to_cpu(data->addr), + be32_to_cpu(data->byte_count), + PCI_DMA_TODEVICE); + } + tx_drop: dev_kfree_skb_any(skb); priv->stats.tx_dropped++;