netdev - Re: [PATCH net 1/1] net: Fix return value of qdisc ingress handling on success

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20220922082349.18fb65d6@kernel.org>
Date:   Thu, 22 Sep 2022 08:23:49 -0700
From:   Jakub Kicinski <kuba@...nel.org>
To:     Daniel Borkmann <daniel@...earbox.net>
Cc:     Paul Blakey <paulb@...dia.com>, Vlad Buslov <vladbu@...dia.com>,
        Oz Shlomo <ozsh@...dia.com>, Roi Dayan <roid@...dia.com>,
        netdev@...r.kernel.org, Saeed Mahameed <saeedm@...dia.com>,
        Eric Dumazet <edumazet@...gle.com>,
        "David S. Miller" <davem@...emloft.net>,
        Paolo Abeni <pabeni@...hat.com>
Subject: Re: [PATCH net 1/1] net: Fix return value of qdisc ingress handling
 on success

On Thu, 22 Sep 2022 16:47:14 +0200 Daniel Borkmann wrote:
> >> Looks reasonable and aligns with sch_handle_egress() fwiw. I think your Fixes tag is wrong
> >> since that commit didn't modify any of the above. This patch should also rather go to net-next
> >> tree to make sure it has enough soak time to catch potential regressions from this change in
> >> behavior.  
> > 
> > I don't think we do "soak time" in networking. Perhaps we can try
> > to use the "CC: stable # after 4 weeks" delay mechanism which Greg
> > promised at LPC?  
> 
> Isn't that implicit? If the commit has Fixes tag and lands in net-next, stable team
> anyway automatically pulls it once everything lands in Linus' tree via merge win and
> then does the backporting for stable.

What I meant is we don't merge fixes into net-next directly.
Perhaps that's my personal view, not shared by other netdev maintainers.

To me the 8 rc release process is fairly arbitrary timing wise.
The fixes continue flowing in after Linus cuts final, plus only 
after a few stable releases the kernel makes it to a wide audience.

Putting a fix in -next gives us anywhere between 0 and 8 weeks of delay.
Explicit delay on the tag seems much more precise and independent of
where we are in the release cycle.

The cases where we put something in -next, later it becomes urgent 
and we can't get it to stable stand out in my memory much more than
problems introduced late in the rc cycle.