netdev - Re: [net,v3] tcp: correct handling of extreme memory squeeze

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: <CANn89i+YRqgg7YncrvRhisqBP8PZcrykNnUUF+tguaMEJG340Q@mail.gmail.com>
Date: Tue, 28 Jan 2025 18:11:49 +0100
From: Eric Dumazet <edumazet@...gle.com>
To: Jon Maloy <jmaloy@...hat.com>
Cc: netdev@...r.kernel.org, davem@...emloft.net, kuba@...nel.org, 
	passt-dev@...st.top, sbrivio@...hat.com, lvivier@...hat.com, 
	dgibson@...hat.com, memnglong8.dong@...il.com, kerneljasonxing@...il.com, 
	ncardwell@...gle.com, eric.dumazet@...il.com
Subject: Re: [net,v3] tcp: correct handling of extreme memory squeeze

On Tue, Jan 28, 2025 at 5:51 PM Jon Maloy <jmaloy@...hat.com> wrote:

> I clearly stated in a previous comment that this was the case, and that
> it has been fixed now. My reason for posting this is because I still
> think this is a bug, just as I think the way we use rcv_ssthresh in
> _tcp_select)window() is a bug that eventually should be fixed.

I was referring to a wrong statement in the changelog, claiming a
'deadlock situation' ...

It is pretty clear there is no deadlock here, unless the remote TCP
stack is _absolutely_ _broken_.

If you still want to capture this in an official changelog, it would
be nice to clarify this,
to avoid yet another CVE to be filled based on scary sentences
misleading many teams
in the world.

Keep changelogs accurate and factual, so that we can find useful
signals in them.

All your __tcp_cleanup_rbuf() repetitions are simply noise. It does not matter
if it is called once or ten times.