From: Pablo Neira Ayuso <pablo@netfilter.org>
To: Shashank Mohan Jain <jain.sm@gmail.com>
Cc: Florian Westphal <fw@strlen.de>, Phil Sutter <phil@nwl.cc>,
netfilter-devel@vger.kernel.org, coreteam@netfilter.org,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH nf] netfilter: ctnetlink: apply port filters to SCTP and GRE entries
Date: Wed, 18 Nov 2026 12:47:05 +0100 [thread overview]
Message-ID: <av2QOZbkZZnXBKme@chamomile> (raw)
In-Reply-To: <20261118042950.83129-1-jain.sm@gmail.com>
Hi,
This is nf-next material.
This is a feature that is missing. BTW, GRE has no ports.
On Wed, Nov 18, 2026 at 09:59:50AM +0530, Shashank Mohan Jain wrote:
> ctnetlink_filter_match_tuple() only compares CTA_PROTO_SRC_PORT and
> CTA_PROTO_DST_PORT for TCP and UDP. SCTP and GRE also parse these
> attributes into the filter tuple (both use nf_ct_port_nlattr_to_tuple()),
> so a filter such as "protonum SCTP, src port X" is accepted, but the
> port part is then ignored and every SCTP entry matches.
>
> CTA_FILTER is also honoured by IPCTNL_MSG_CT_DELETE
> (ctnetlink_flush_conntrack()), so a filtered flush that asks for one
> SCTP port deletes all SCTP conntrack entries in the netns, and a
> filtered dump returns all of them. The same applies to GRE keys.
>
> The port and the GRE key share the same offset in
> nf_conntrack_man_proto and in the dst union, so the existing
> comparison works for both.
>
> Found by reading the code with LLM assistance. Tested on x86_64 under
> QEMU: with three SCTP and three GRE entries, a delete filtered on one
> source port (or GRE key) removed all three entries of that protocol
> before this change and only the matching one with it. Filtered dumps
> show the same difference.
>
> Fixes: cb8aa9a3affb ("netfilter: ctnetlink: add kernel side filtering for dump")
> Cc: stable@vger.kernel.org
> Assisted-by: LLM
> Signed-off-by: Shashank Mohan Jain <jain.sm@gmail.com>
> ---
> Prepared with Claude Code (Anthropic), model Claude Opus 5.5
> (claude-opus-5-5).
>
> Based on nf.git main 734bef548a96. Tested with a small raw ctnetlink
> program in a busybox initramfs under QEMU (x86_64, TCG). It adds three
> SCTP, three GRE and three TCP entries, then sends dumps and
> IPCTNL_MSG_CT_DELETE requests with CTA_FILTER set to protonum plus
> source port. TCP behaves the same before and after. Also built for
> i386; W=1 shows no warnings for nf_conntrack_netlink.c before or after.
>
> net/netfilter/nf_conntrack_netlink.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
> index 1f8d94a05..5e8e0b93b 100644
> --- a/net/netfilter/nf_conntrack_netlink.c
> +++ b/net/netfilter/nf_conntrack_netlink.c
> @@ -1127,6 +1127,8 @@ static int ctnetlink_filter_match_tuple(struct nf_conntrack_tuple *filter_tuple,
> switch (ct_tuple->dst.protonum) {
> case IPPROTO_TCP:
> case IPPROTO_UDP:
> + case IPPROTO_SCTP:
> + case IPPROTO_GRE:
> if ((flags & CTA_FILTER_FLAG(CTA_PROTO_SRC_PORT)) &&
> filter_tuple->src.u.tcp.port != ct_tuple->src.u.tcp.port)
> return 0;
> --
> 2.43.0
>
prev parent reply other threads:[~2026-11-18 11:47 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-18 4:29 Shashank Mohan Jain
2026-11-18 11:47 ` 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=av2QOZbkZZnXBKme@chamomile \
--to=pablo@netfilter.org \
--cc=coreteam@netfilter.org \
--cc=fw@strlen.de \
--cc=jain.sm@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=netfilter-devel@vger.kernel.org \
--cc=phil@nwl.cc \
/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
all inboxes | Powered by JetHome®