DPDK patches and discussions
 help / color / mirror / Atom feed
From: Dariusz Sosnowski <dsosnowski@nvidia.com>
To: Matan Azrad <matan@nvidia.com>,
	Viacheslav Ovsiienko <viacheslavo@nvidia.com>,
	Ori Kam <orika@nvidia.com>, Suanming Mou <suanmingm@nvidia.com>
Cc: <dev@dpdk.org>, Raslan Darawsheh <rasland@nvidia.com>, <stable@dpdk.org>
Subject: [PATCH] net/mlx5: fix offset field size in conntrack action
Date: Mon, 20 Nov 2023 10:24:57 +0200	[thread overview]
Message-ID: <20231120082458.9020-1-dsosnowski@nvidia.com> (raw)

In mlx5 PMD, struct mlx5_aso_ct_action represents a single ASO CT object
inside a ASO CT bulk, which is used to implement conntrack flow action.
offset field, which is 16 bits wide, in this struct represents
an offset of the given object inside a bulk.

When SW Steering flow engine is used, ASO CT bulks are allocated on
demand, each having 64 objects. In this case offset field can fit any
valid object offset.

When HW Steering flow engine is used, a single ASO CT bulk is allocated
during flow engine configuration. Since HW supports bulks bigger
than 2^16 this offset field is too small to fit all valid values.
This causes some ASO CT objects to be incorrectly reused on multiple
connections, when more than 2^16 conntrack flow actions are used.

This patch fixes this behavior. offset field size is increased
to 32 bits.

Fixes: 463170a7c934 ("net/mlx5: support connection tracking with HWS")
Cc: suanmingm@nvidia.com
Cc: stable@dpdk.org

Signed-off-by: Dariusz Sosnowski <dsosnowski@nvidia.com>
Acked-by: Ori Kam <orika@nvidia.com>
---
 drivers/net/mlx5/mlx5.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/mlx5/mlx5.h b/drivers/net/mlx5/mlx5.h
index 76bf7d0f4f..263ebead7f 100644
--- a/drivers/net/mlx5/mlx5.h
+++ b/drivers/net/mlx5/mlx5.h
@@ -1286,7 +1286,7 @@ struct mlx5_aso_ct_action {
 	/* General action object for reply dir. */
 	void *dr_action_rply;
 	uint32_t refcnt; /* Action used count in device flows. */
-	uint16_t offset; /* Offset of ASO CT in DevX objects bulk. */
+	uint32_t offset; /* Offset of ASO CT in DevX objects bulk. */
 	uint16_t peer; /* The only peer port index could also use this CT. */
 	enum mlx5_aso_ct_state state; /* ASO CT state. */
 	bool is_original; /* The direction of the DR action to be used. */
-- 
2.25.1


             reply	other threads:[~2023-11-20  8:25 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-11-20  8:24 Dariusz Sosnowski [this message]
2023-11-21 15:11 ` Raslan Darawsheh

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20231120082458.9020-1-dsosnowski@nvidia.com \
    --to=dsosnowski@nvidia.com \
    --cc=dev@dpdk.org \
    --cc=matan@nvidia.com \
    --cc=orika@nvidia.com \
    --cc=rasland@nvidia.com \
    --cc=stable@dpdk.org \
    --cc=suanmingm@nvidia.com \
    --cc=viacheslavo@nvidia.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).