netdev - Re: netconsole: HARDIRQ-safe -> HARDIRQ-unsafe lock order warning

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: <6943b4ab95056019eae5780d23232b3ef237a49f.camel@gmx.de>
Date: Tue, 09 Sep 2025 17:49:24 +0200
From: Mike Galbraith <efault@....de>
To: Calvin Owens <calvin@...nvd.org>, John Ogness
 <john.ogness@...utronix.de>,  Breno Leitao <leitao@...ian.org>
Cc: Simon Horman <horms@...nel.org>, kuba@...nel.org, Pavel Begunkov
 <asml.silence@...il.com>, Johannes Berg <johannes@...solutions.net>, 
 paulmck@...nel.org, LKML <linux-kernel@...r.kernel.org>,
 netdev@...r.kernel.org,  boqun.feng@...il.com
Subject: Re: netconsole: HARDIRQ-safe -> HARDIRQ-unsafe lock order warning

On Mon, 2025-09-08 at 13:27 -0700, Calvin Owens wrote:
> On Friday 09/05 at 14:54 +0206, John Ogness wrote:
> > <snip>
> > 
> > NBCON is meant to deprecate @oops_in_progress. However, it is true that
> > consoles not implementing ->write_atomic() will never print panic
> > output.
> 
> Below is a silly little testcase that makes it more convenient to test
> if crashes are getting out in a few canned cases, in case anyone else
> finds it useful.

I gave it a go.

> Testing this on 6.17-rc5 on a Pi 4b, I don't get any netconsole output
> at all for any crash case over wifi, so that already doesn't work. All
> the cases currently work over ethernet.

With local hackery on top of Breno's patches I captured all testcases
for RT and !RT wired and wireless except for one RT wired instance due
to a GPF_ATOMIC allocation failure.

The wireless stack was unhappy about ending up in an nbcon write_atomic
method with IRQs disabled, thus sent complaints along with the oopsen.

	-Mike