aboutsummaryrefslogtreecommitdiff
path: root/drivers/net/ethernet/mellanox/mlx4/en_rx.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/net/ethernet/mellanox/mlx4/en_rx.c')
-rw-r--r--drivers/net/ethernet/mellanox/mlx4/en_rx.c103
1 files changed, 83 insertions, 20 deletions
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
index 07a1d0fbae4..5535862f27c 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c
@@ -40,6 +40,7 @@
#include <linux/if_ether.h>
#include <linux/if_vlan.h>
#include <linux/vmalloc.h>
+#include <linux/irq.h>
#include "mlx4_en.h"
@@ -270,13 +271,11 @@ static int mlx4_en_fill_rx_buffers(struct mlx4_en_priv *priv)
ring->actual_size,
GFP_KERNEL)) {
if (ring->actual_size < MLX4_EN_MIN_RX_SIZE) {
- en_err(priv, "Failed to allocate "
- "enough rx buffers\n");
+ en_err(priv, "Failed to allocate enough rx buffers\n");
return -ENOMEM;
} else {
new_size = rounddown_pow_of_two(ring->actual_size);
- en_warn(priv, "Only %d buffers allocated "
- "reducing ring size to %d",
+ en_warn(priv, "Only %d buffers allocated reducing ring size to %d\n",
ring->actual_size, new_size);
goto reduce_rings;
}
@@ -318,6 +317,31 @@ static void mlx4_en_free_rx_buf(struct mlx4_en_priv *priv,
}
}
+void mlx4_en_set_num_rx_rings(struct mlx4_en_dev *mdev)
+{
+ int i;
+ int num_of_eqs;
+ int num_rx_rings;
+ struct mlx4_dev *dev = mdev->dev;
+
+ mlx4_foreach_port(i, dev, MLX4_PORT_TYPE_ETH) {
+ if (!dev->caps.comp_pool)
+ num_of_eqs = max_t(int, MIN_RX_RINGS,
+ min_t(int,
+ dev->caps.num_comp_vectors,
+ DEF_RX_RINGS));
+ else
+ num_of_eqs = min_t(int, MAX_MSIX_P_PORT,
+ dev->caps.comp_pool/
+ dev->caps.num_ports) - 1;
+
+ num_rx_rings = min_t(int, num_of_eqs,
+ netif_get_num_default_rss_queues());
+ mdev->profile.prof[i].rx_ring_num =
+ rounddown_pow_of_two(num_rx_rings);
+ }
+}
+
int mlx4_en_create_rx_ring(struct mlx4_en_priv *priv,
struct mlx4_en_rx_ring **pring,
u32 size, u16 stride, int node)
@@ -631,10 +655,14 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
int ip_summed;
int factor = priv->cqe_factor;
u64 timestamp;
+ bool l2_tunnel;
if (!priv->port_up)
return 0;
+ if (budget <= 0)
+ return polled;
+
/* We assume a 1:1 mapping between CQEs and Rx descriptors, so Rx
* descriptor offset can be deduced from the CQE index instead of
* reading 'cqe->index' */
@@ -656,10 +684,9 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
/* Drop packet on bad receive or bad checksum */
if (unlikely((cqe->owner_sr_opcode & MLX4_CQE_OPCODE_MASK) ==
MLX4_CQE_OPCODE_ERROR)) {
- en_err(priv, "CQE completed in error - vendor "
- "syndrom:%d syndrom:%d\n",
- ((struct mlx4_err_cqe *) cqe)->vendor_err_syndrome,
- ((struct mlx4_err_cqe *) cqe)->syndrome);
+ en_err(priv, "CQE completed in error - vendor syndrom:%d syndrom:%d\n",
+ ((struct mlx4_err_cqe *)cqe)->vendor_err_syndrome,
+ ((struct mlx4_err_cqe *)cqe)->syndrome);
goto next;
}
if (unlikely(cqe->badfcs_enc & MLX4_CQE_BAD_FCS)) {
@@ -709,6 +736,8 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
length -= ring->fcs_del;
ring->bytes += length;
ring->packets++;
+ l2_tunnel = (dev->hw_enc_features & NETIF_F_RXCSUM) &&
+ (cqe->vlan_my_qpn & cpu_to_be32(MLX4_CQE_L2_TUNNEL));
if (likely(dev->features & NETIF_F_RXCSUM)) {
if ((cqe->status & cpu_to_be16(MLX4_CQE_STATUS_IPOK)) &&
@@ -721,7 +750,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
* - not an IP fragment
* - no LLS polling in progress
*/
- if (!mlx4_en_cq_ll_polling(cq) &&
+ if (!mlx4_en_cq_busy_polling(cq) &&
(dev->features & NETIF_F_GRO)) {
struct sk_buff *gro_skb = napi_get_frags(&cq->napi);
if (!gro_skb)
@@ -738,6 +767,8 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
gro_skb->data_len = length;
gro_skb->ip_summed = CHECKSUM_UNNECESSARY;
+ if (l2_tunnel)
+ gro_skb->encapsulation = 1;
if ((cqe->vlan_my_qpn &
cpu_to_be32(MLX4_CQE_VLAN_PRESENT_MASK)) &&
(dev->features & NETIF_F_HW_VLAN_CTAG_RX)) {
@@ -747,9 +778,12 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
}
if (dev->features & NETIF_F_RXHASH)
- gro_skb->rxhash = be32_to_cpu(cqe->immed_rss_invalid);
+ skb_set_hash(gro_skb,
+ be32_to_cpu(cqe->immed_rss_invalid),
+ PKT_HASH_TYPE_L3);
skb_record_rx_queue(gro_skb, cq->ring);
+ skb_mark_napi_id(gro_skb, &cq->napi);
if (ring->hwtstamp_rx_filter == HWTSTAMP_FILTER_ALL) {
timestamp = mlx4_en_get_cqe_ts(cqe);
@@ -788,8 +822,13 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
skb->protocol = eth_type_trans(skb, dev);
skb_record_rx_queue(skb, cq->ring);
+ if (l2_tunnel)
+ skb->encapsulation = 1;
+
if (dev->features & NETIF_F_RXHASH)
- skb->rxhash = be32_to_cpu(cqe->immed_rss_invalid);
+ skb_set_hash(skb,
+ be32_to_cpu(cqe->immed_rss_invalid),
+ PKT_HASH_TYPE_L3);
if ((be32_to_cpu(cqe->vlan_my_qpn) &
MLX4_CQE_VLAN_PRESENT_MASK) &&
@@ -804,8 +843,10 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud
skb_mark_napi_id(skb, &cq->napi);
- /* Push it up the stack */
- netif_receive_skb(skb);
+ if (!mlx4_en_cq_busy_polling(cq))
+ napi_gro_receive(&cq->napi, skb);
+ else
+ netif_receive_skb(skb);
next:
for (nr = 0; nr < priv->num_frags; nr++)
@@ -856,9 +897,25 @@ int mlx4_en_poll_rx_cq(struct napi_struct *napi, int budget)
mlx4_en_cq_unlock_napi(cq);
/* If we used up all the quota - we're probably not done yet... */
- if (done == budget)
+ if (done == budget) {
+ int cpu_curr;
+ const struct cpumask *aff;
+
INC_PERF_COUNTER(priv->pstats.napi_quota);
- else {
+
+ cpu_curr = smp_processor_id();
+ aff = irq_desc_get_irq_data(cq->irq_desc)->affinity;
+
+ if (unlikely(!cpumask_test_cpu(cpu_curr, aff))) {
+ /* Current cpu is not according to smp_irq_affinity -
+ * probably affinity changed. need to stop this NAPI
+ * poll, and restart it on the right CPU
+ */
+ napi_complete(napi);
+ mlx4_en_arm_cq(priv, cq);
+ return 0;
+ }
+ } else {
/* Done for now */
napi_complete(napi);
mlx4_en_arm_cq(priv, cq);
@@ -902,8 +959,8 @@ void mlx4_en_calc_rx_buf(struct net_device *dev)
priv->rx_skb_size = eff_mtu;
priv->log_rx_info = ROUNDUP_LOG2(i * sizeof(struct mlx4_en_rx_alloc));
- en_dbg(DRV, priv, "Rx buffer scatter-list (effective-mtu:%d "
- "num_frags:%d):\n", eff_mtu, priv->num_frags);
+ en_dbg(DRV, priv, "Rx buffer scatter-list (effective-mtu:%d num_frags:%d):\n",
+ eff_mtu, priv->num_frags);
for (i = 0; i < priv->num_frags; i++) {
en_err(priv,
" frag:%d - size:%d prefix:%d align:%d stride:%d\n",
@@ -930,7 +987,7 @@ static int mlx4_en_config_rss_qp(struct mlx4_en_priv *priv, int qpn,
if (!context)
return -ENOMEM;
- err = mlx4_qp_alloc(mdev->dev, qpn, qp);
+ err = mlx4_qp_alloc(mdev->dev, qpn, qp, GFP_KERNEL);
if (err) {
en_err(priv, "Failed to allocate qp #%x\n", qpn);
goto out;
@@ -970,7 +1027,7 @@ int mlx4_en_create_drop_qp(struct mlx4_en_priv *priv)
en_err(priv, "Failed reserving drop qpn\n");
return err;
}
- err = mlx4_qp_alloc(priv->mdev->dev, qpn, &priv->drop_qp);
+ err = mlx4_qp_alloc(priv->mdev->dev, qpn, &priv->drop_qp, GFP_KERNEL);
if (err) {
en_err(priv, "Failed allocating drop qp\n");
mlx4_qp_release_range(priv->mdev->dev, qpn, 1);
@@ -1029,7 +1086,7 @@ int mlx4_en_config_rss_steer(struct mlx4_en_priv *priv)
}
/* Configure RSS indirection qp */
- err = mlx4_qp_alloc(mdev->dev, priv->base_qpn, &rss_map->indir_qp);
+ err = mlx4_qp_alloc(mdev->dev, priv->base_qpn, &rss_map->indir_qp, GFP_KERNEL);
if (err) {
en_err(priv, "Failed to allocate RSS indirection QP\n");
goto rss_err;
@@ -1053,6 +1110,12 @@ int mlx4_en_config_rss_steer(struct mlx4_en_priv *priv)
rss_mask |= MLX4_RSS_UDP_IPV4 | MLX4_RSS_UDP_IPV6;
rss_context->base_qpn_udp = rss_context->default_qpn;
}
+
+ if (mdev->dev->caps.tunnel_offload_mode == MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) {
+ en_info(priv, "Setting RSS context tunnel type to RSS on inner headers\n");
+ rss_mask |= MLX4_RSS_BY_INNER_HEADERS;
+ }
+
rss_context->flags = rss_mask;
rss_context->hash_fn = MLX4_RSS_HASH_TOP;
for (i = 0; i < 10; i++)