From: Pablo Neira Ayuso <pablo@netfilter.org>
To: Donald Hunter <donald.hunter@gmail.com>
Cc: Jozsef Kadlecsik <kadlec@netfilter.org>,
Florian Westphal <fw@strlen.de>,
"David S. Miller" <davem@davemloft.net>,
Eric Dumazet <edumazet@google.com>,
Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
Paul Blakey <paulb@nvidia.com>, Vlad Buslov <vladbu@nvidia.com>,
netfilter-devel@vger.kernel.org, coreteam@netfilter.org,
netdev@vger.kernel.org
Subject: Re: [PATCH netfilter] Fix hw flow offload from nftables
Date: Mon, 23 Oct 2023 12:24:17 +0200 [thread overview]
Message-ID: <ZTZJ0WGBHRBvoDca@calendula> (raw)
In-Reply-To: <20231023101347.564898-1-donald.hunter@gmail.com>
On Mon, Oct 23, 2023 at 11:13:47AM +0100, Donald Hunter wrote:
> The NF_FLOW_HW_ESTABLISHED bit was not getting set in any nftables code
> paths. It seems that the state was never correctly maintained but there
> was no negative side-effect until commit 41f2c7c342d3 ("net/sched:
> act_ct: Fix promotion of offloaded unreplied tuple") which uses it as
> part of a flow outdated check. The net result is repeated cycles of
> FLOW_CLS_REPLACE / FLOW_CLS_DESTROY commands and never getting any
> FLOW_CLS_STATS commands for a flow.
>
> This patch sets and clears the NF_FLOW_HW_ESTABLISHED bit for nftables.
>
> Note that I don't have hardware to test this with. I have observed
> the behaviour and verified the fix with modified veth code.
Very strange, this bit did not exists before:
commit 1a441a9b8be8849957a01413a144f84932c324cb
Author: Vlad Buslov <vladbu@nvidia.com>
Date: Wed Feb 1 17:30:57 2023 +0100
netfilter: flowtable: cache info of last offload
I have no idea why this is needed.
> Signed-off-by: Donald Hunter <donald.hunter@gmail.com>
> Fixes: 41f2c7c342d3 ("net/sched: act_ct: Fix promotion of offloaded unreplied tuple")
> ---
> net/netfilter/nf_flow_table_offload.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c
> index 1c26f03fc661..1d017191af80 100644
> --- a/net/netfilter/nf_flow_table_offload.c
> +++ b/net/netfilter/nf_flow_table_offload.c
> @@ -918,6 +918,7 @@ static void flow_offload_work_add(struct flow_offload_work *offload)
> goto out;
>
> set_bit(IPS_HW_OFFLOAD_BIT, &offload->flow->ct->status);
> + set_bit(NF_FLOW_HW_ESTABLISHED, &offload->flow->flags);
>
> out:
> nf_flow_offload_destroy(flow_rule);
> @@ -925,6 +926,7 @@ static void flow_offload_work_add(struct flow_offload_work *offload)
>
> static void flow_offload_work_del(struct flow_offload_work *offload)
> {
> + clear_bit(NF_FLOW_HW_ESTABLISHED, &offload->flow->flags);
> clear_bit(IPS_HW_OFFLOAD_BIT, &offload->flow->ct->status);
> flow_offload_tuple_del(offload, FLOW_OFFLOAD_DIR_ORIGINAL);
> if (test_bit(NF_FLOW_HW_BIDIRECTIONAL, &offload->flow->flags))
> --
> 2.41.0
>
prev parent reply other threads:[~2023-10-23 10:24 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-10-23 10:13 [PATCH netfilter] Fix hw flow offload from nftables Donald Hunter
2023-10-23 10:24 ` Pablo Neira Ayuso [this message]
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=ZTZJ0WGBHRBvoDca@calendula \
--to=pablo@netfilter.org \
--cc=coreteam@netfilter.org \
--cc=davem@davemloft.net \
--cc=donald.hunter@gmail.com \
--cc=edumazet@google.com \
--cc=fw@strlen.de \
--cc=kadlec@netfilter.org \
--cc=kuba@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=netfilter-devel@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=paulb@nvidia.com \
--cc=vladbu@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).