[dpdk-dev] net/mlx5: implement drop action in hardware classifier
Checks
Commit Message
The current drop action is implemented as a queue tail drop,
requiring to instantiate multiple WQs to maintain high drop rate.
This commit, implements the drop action in hardware classifier.
This enables to reduce the amount of contexts needed for the drop,
without affecting the drop rate.
Signed-off-by: Shachar Beiser <shacharbe@mellanox.com>
---
drivers/net/mlx5/Makefile | 5 +++++
drivers/net/mlx5/mlx5_flow.c | 43 +++++++++++++++++++++++++++++++++++++++++++
2 files changed, 48 insertions(+)
@@ -101,6 +101,11 @@ mlx5_autoconf.h.new: FORCE
mlx5_autoconf.h.new: $(RTE_SDK)/buildtools/auto-config-h.sh
$Q $(RM) -f -- '$@'
$Q sh -- '$<' '$@' \
+ HAVE_VERBS_IBV_EXP_FLOW_SPEC_ACTION_DROP \
+ infiniband/verbs_exp.h \
+ enum IBV_EXP_FLOW_SPEC_ACTION_DROP \
+ $(AUTOCONF_OUTPUT)
+ $Q sh -- '$<' '$@' \
HAVE_VERBS_IBV_EXP_CQ_COMPRESSED_CQE \
infiniband/verbs_exp.h \
enum IBV_EXP_CQ_COMPRESSED_CQE \
@@ -994,6 +994,11 @@ struct mlx5_flow_action {
{
struct rte_flow *rte_flow;
+#ifdef HAVE_VERBS_IBV_EXP_FLOW_SPEC_ACTION_DROP
+ struct ibv_exp_flow_spec_action_drop *drop;
+ unsigned int size = sizeof(struct ibv_exp_flow_spec_action_drop);
+#endif
+
assert(priv->pd);
assert(priv->ctx);
rte_flow = rte_calloc(__func__, 1, sizeof(*rte_flow), 0);
@@ -1007,6 +1012,15 @@ struct mlx5_flow_action {
rte_flow->qp = priv->flow_drop_queue->qp;
if (!priv->started)
return rte_flow;
+#ifdef HAVE_VERBS_IBV_EXP_FLOW_SPEC_ACTION_DROP
+ drop = (void *)((uintptr_t)flow->ibv_attr + flow->offset);
+ *drop = (struct ibv_exp_flow_spec_action_drop){
+ .type = IBV_EXP_FLOW_SPEC_ACTION_DROP,
+ .size = size,
+ };
+ ++flow->ibv_attr->num_of_specs;
+ flow->offset += sizeof(struct ibv_exp_flow_spec_action_drop);
+#endif
rte_flow->ibv_flow = ibv_exp_create_flow(rte_flow->qp,
rte_flow->ibv_attr);
if (!rte_flow->ibv_flow) {
@@ -1370,7 +1384,9 @@ struct rte_flow *
priv_flow_create_drop_queue(struct priv *priv)
{
struct rte_flow_drop *fdq = NULL;
+#ifndef HAVE_VERBS_IBV_EXP_FLOW_SPEC_ACTION_DROP
unsigned int i;
+#endif
assert(priv->pd);
assert(priv->ctx);
@@ -1387,6 +1403,7 @@ struct rte_flow *
WARN("cannot allocate CQ for drop queue");
goto error;
}
+#ifndef HAVE_VERBS_IBV_EXP_FLOW_SPEC_ACTION_DROP
for (i = 0; i != MLX5_DROP_WQ_N; ++i) {
fdq->wqs[i] = ibv_exp_create_wq(priv->ctx,
&(struct ibv_exp_wq_init_attr){
@@ -1401,6 +1418,20 @@ struct rte_flow *
goto error;
}
}
+#else
+ fdq->wqs[0] = ibv_exp_create_wq(priv->ctx,
+ &(struct ibv_exp_wq_init_attr){
+ .wq_type = IBV_EXP_WQT_RQ,
+ .max_recv_wr = 1,
+ .max_recv_sge = 1,
+ .pd = priv->pd,
+ .cq = fdq->cq,
+ });
+ if (!fdq->wqs[0]) {
+ WARN("cannot allocate WQ for drop queue");
+ goto error;
+ }
+#endif
fdq->ind_table = ibv_exp_create_rwq_ind_table(priv->ctx,
&(struct ibv_exp_rwq_ind_table_init_attr){
.pd = priv->pd,
@@ -1441,10 +1472,15 @@ struct rte_flow *
claim_zero(ibv_destroy_qp(fdq->qp));
if (fdq->ind_table)
claim_zero(ibv_exp_destroy_rwq_ind_table(fdq->ind_table));
+#ifndef HAVE_VERBS_IBV_EXP_FLOW_SPEC_ACTION_DROP
for (i = 0; i != MLX5_DROP_WQ_N; ++i) {
if (fdq->wqs[i])
claim_zero(ibv_exp_destroy_wq(fdq->wqs[i]));
}
+#else
+ if (fdq->wqs[0])
+ claim_zero(ibv_exp_destroy_wq(fdq->wqs[0]));
+#endif
if (fdq->cq)
claim_zero(ibv_destroy_cq(fdq->cq));
if (fdq)
@@ -1463,7 +1499,9 @@ struct rte_flow *
priv_flow_delete_drop_queue(struct priv *priv)
{
struct rte_flow_drop *fdq = priv->flow_drop_queue;
+#ifndef HAVE_VERBS_IBV_EXP_FLOW_SPEC_ACTION_DROP
unsigned int i;
+#endif
if (!fdq)
return;
@@ -1471,10 +1509,15 @@ struct rte_flow *
claim_zero(ibv_destroy_qp(fdq->qp));
if (fdq->ind_table)
claim_zero(ibv_exp_destroy_rwq_ind_table(fdq->ind_table));
+#ifndef HAVE_VERBS_IBV_EXP_FLOW_SPEC_ACTION_DROP
for (i = 0; i != MLX5_DROP_WQ_N; ++i) {
if (fdq->wqs[i])
claim_zero(ibv_exp_destroy_wq(fdq->wqs[i]));
}
+#else
+ if (fdq->wqs[0])
+ claim_zero(ibv_exp_destroy_wq(fdq->wqs[0]));
+#endif
if (fdq->cq)
claim_zero(ibv_destroy_cq(fdq->cq));
rte_free(fdq);