From: Paolo Abeni <pabeni@redhat.com>
To: Jamal Hadi Salim <jhs@mojatatu.com>, Ilya Maximets <i.maximets@ovn.org>
Cc: netdev@vger.kernel.org, Jiri Pirko <jiri@resnulli.us>,
"David S. Miller" <davem@davemloft.net>,
Eric Dumazet <edumazet@google.com>,
Jakub Kicinski <kuba@kernel.org>, Simon Horman <horms@kernel.org>,
Henrik Steen <henrist@henrist.net>,
Olivier Tilmans <olivier.tilmans@nokia.com>,
Bob Briscoe <research@bobbriscoe.net>,
Olga Albisser <olga@albisser.org>,
GangMin Kim <km.kim1503@gmail.com>,
Eelco Chaudron <echaudro@redhat.com>,
Aaron Conole <aconole@redhat.com>,
Florian Westphal <fw@strlen.de>
Subject: Re: [PATCH net] net_sched: act_ct: drop all packets when not attached to ingress
Date: Wed, 18 Feb 2026 21:43:51 +0100 [thread overview]
Message-ID: <d07fdeb8-2b26-43f0-8fab-28907e88bd06@redhat.com> (raw)
In-Reply-To: <CAM0EoMm3zkJpkHkwHX0YQDSJD_gEMphihZ3CC+b1dMx49bdjWg@mail.gmail.com>
On 2/18/26 7:44 PM, Jamal Hadi Salim wrote:
> On Wed, Feb 18, 2026 at 1:31 PM Jamal Hadi Salim <jhs@mojatatu.com> wrote:
>> On Wed, Feb 18, 2026 at 11:15 AM Ilya Maximets <i.maximets@ovn.org> wrote:
>>> From a user's perspective I'd prefer if RTM_NEWTFILTER just fails when
>>> it contains TCA_ACT_KIND "ct" with TC_H_MIN_EGRESS. This is clear
>>> for the application that makes a request and for the user if they make
>>> the request manually with 'tc filter ...'.
>
>> The challenge is actions could be created as a standalone i.e "tc
>> actions add action ct..." then later bound via tc filter. Don't forget
>> actions can also be shared by multiple filters (which could be a mix
>> of egress/ingress)...
>
> Actually, looking closely at the code - this is doable. Let's see if a
> patch can be cooked.
I discussed a bit the topic with Davide, it looks like the thing you
mentioned above could be handled. The problematic part is AFAICS the
additional indirection level added by (possibly shared) blocks. AFAICS
whatever check we do at ct_init() time, shared block could later
circumvent it - unless act_ct is always forbidden for shared blocks.
TL;DR: I think admission check at init time can be implemented only in a
quite (too much?) restrictive way.
/P
next prev parent reply other threads:[~2026-02-18 20:43 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-17 9:38 [PATCH net] net_sched: act_ct: drop all packets when not attached to ingress Paolo Abeni
2026-02-17 10:42 ` Paolo Abeni
2026-02-17 14:49 ` Ilya Maximets
2026-02-17 15:52 ` Paolo Abeni
2026-02-17 19:37 ` Ilya Maximets
2026-02-18 14:28 ` Jamal Hadi Salim
2026-02-18 16:15 ` Ilya Maximets
2026-02-18 18:31 ` Jamal Hadi Salim
2026-02-18 18:44 ` Jamal Hadi Salim
2026-02-18 20:43 ` Paolo Abeni [this message]
2026-02-19 11:46 ` Ilya Maximets
2026-02-19 14:16 ` Jamal Hadi Salim
2026-02-19 20:13 ` Jamal Hadi Salim
2026-02-20 12:24 ` Victor Nogueira
2026-02-20 13:41 ` Ilya Maximets
2026-02-20 16:12 ` Victor Nogueira
2026-02-17 15:28 ` Jakub Kicinski
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=d07fdeb8-2b26-43f0-8fab-28907e88bd06@redhat.com \
--to=pabeni@redhat.com \
--cc=aconole@redhat.com \
--cc=davem@davemloft.net \
--cc=echaudro@redhat.com \
--cc=edumazet@google.com \
--cc=fw@strlen.de \
--cc=henrist@henrist.net \
--cc=horms@kernel.org \
--cc=i.maximets@ovn.org \
--cc=jhs@mojatatu.com \
--cc=jiri@resnulli.us \
--cc=km.kim1503@gmail.com \
--cc=kuba@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=olga@albisser.org \
--cc=olivier.tilmans@nokia.com \
--cc=research@bobbriscoe.net \
/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