diff mbox

[dpdk-dev,1/2] net/mlx4: fix UDP flow rule limitation enforcement

Message ID 20180504151421.24765-1-adrien.mazarguil@6wind.com (mailing list archive)
State Accepted, archived
Delegated to: Shahaf Shuler
Headers

Checks

Context Check Description
ci/checkpatch success coding style OK
ci/Intel-compilation success Compilation OK

Commit Message

Adrien Mazarguil May 4, 2018, 3:17 p.m. UTC 
  For some unknown reason, priorities do not have any effect on flow rules
that happen to match UDP destination ports. Those are always matched first
regardless.

This patch is a workaround that enforces this limitation at the PMD level;
such flow rules can only be created at the highest priority level for
correctness.

Fixes: a5171594fc3b ("net/mlx4: expose support for flow rule priorities")
Cc: stable@dpdk.org

Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
---
 drivers/net/mlx4/mlx4_flow.c | 9 +++++++++
 drivers/net/mlx4/mlx4_flow.h | 1 +
 2 files changed, 10 insertions(+)

Comments

Shahaf Shuler May 8, 2018, 8:04 a.m. UTC | #1 
Friday, May 4, 2018 6:18 PM, Adrien Mazarguil:
> Subject: [PATCH 1/2] net/mlx4: fix UDP flow rule limitation enforcement
> 
> For some unknown reason, priorities do not have any effect on flow rules
> that happen to match UDP destination ports. Those are always matched first
> regardless.
> 
> This patch is a workaround that enforces this limitation at the PMD level; such
> flow rules can only be created at the highest priority level for correctness.
> 
> Fixes: a5171594fc3b ("net/mlx4: expose support for flow rule priorities")
> Cc: stable@dpdk.org
> 
> Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
> ---

Series applied to next-net-mlx, thanks.
diff mbox

Patch

diff --git a/drivers/net/mlx4/mlx4_flow.c b/drivers/net/mlx4/mlx4_flow.c
index bebad074e..3f754b480 100644
--- a/drivers/net/mlx4/mlx4_flow.c
+++ b/drivers/net/mlx4/mlx4_flow.c
@@ -357,6 +357,9 @@  mlx4_flow_merge_ipv4(struct rte_flow *flow,
  * Additional mlx4-specific constraints on supported fields:
  *
  * - No support for partial masks.
+ * - Due to HW/FW limitation, flow rule priority is not taken into account
+ *   when matching UDP destination ports, doing is therefore only supported
+ *   at the highest priority level (0).
  *
  * @param[in, out] flow
  *   Flow rule handle to update.
@@ -388,6 +391,11 @@  mlx4_flow_merge_udp(struct rte_flow *flow,
 		msg = "mlx4 does not support matching partial UDP fields";
 		goto error;
 	}
+	if (mask && mask->hdr.dst_port && flow->priority) {
+		msg = "combining UDP destination port matching with a nonzero"
+			" priority level is not supported";
+		goto error;
+	}
 	if (!flow->ibv_attr)
 		return 0;
 	++flow->ibv_attr->num_of_specs;
@@ -658,6 +666,7 @@  mlx4_flow_prepare(struct priv *priv,
 fill:
 	overlap = 0;
 	proc = mlx4_flow_proc_item_list;
+	flow->priority = attr->priority;
 	/* Go over pattern. */
 	for (item = pattern; item->type; ++item) {
 		const struct mlx4_flow_proc_item *next = NULL;
diff --git a/drivers/net/mlx4/mlx4_flow.h b/drivers/net/mlx4/mlx4_flow.h
index 7b83d74b0..2c8dff362 100644
--- a/drivers/net/mlx4/mlx4_flow.h
+++ b/drivers/net/mlx4/mlx4_flow.h
@@ -42,6 +42,7 @@  struct rte_flow {
 	uint32_t promisc:1; /**< This rule matches everything. */
 	uint32_t allmulti:1; /**< This rule matches all multicast traffic. */
 	uint32_t drop:1; /**< This rule drops packets. */
+	uint32_t priority; /**< Flow rule priority. */
 	struct mlx4_rss *rss; /**< Rx target. */
 };