linux-kernel - Re: 6.12-rc1: general protection fault at pick_task

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] [day] [month] [year] [list]

Message-ID: <20240930-furry-snail-of-promotion-1764e9@leitao>
Date: Mon, 30 Sep 2024 09:56:35 -0700
From: Breno Leitao <leitao@...ian.org>
To: Peter Zijlstra <peterz@...radead.org>
Cc: mingo@...hat.com, juri.lelli@...hat.com, vincent.guittot@...aro.org,
	dietmar.eggemann@....com, rostedt@...dmis.org, bsegall@...gle.com,
	mgorman@...e.de, vschneid@...hat.com, linux-kernel@...r.kernel.org
Subject: Re: 6.12-rc1: general protection fault at pick_task_fair()

Hello Peter,

On Mon, Sep 30, 2024 at 04:48:09PM +0200, Peter Zijlstra wrote:
> On Mon, Sep 30, 2024 at 05:52:49AM -0700, Breno Leitao wrote:
> > Important to say that I am also seeing the following warning before the
> > crash:
> > 
> >  workqueue: drain_vmap_area_work hogged CPU for >20000us 4 times, consider switching to WQ_UNBOUND
> >  ------------[ cut here ]------------
> >            !se->on_rq
> >            WARNING: CPU: 24 PID: 17 at kernel/sched/fair.c:704 dequeue_entity+0xd21/0x17c0
> > 

> This looks to be the same issue as reported here:
> 
>   https://lkml.kernel.org/r/20240930144157.GH5594@noisy.programming.kicks-ass.net

Ok, I though they would be different given that it was crashing at
update_entity_lag() and not at pick_task_fair(). 

> Is there anything you can share about your setup / workload that manages
> to trigger this?

Nothing really. I have a test system where I install a debug kernel
(with some DEBUG+KASAN+LOCKDEP) and keep the host mostly idle, when I
hit it ~2x a day on average.

PS: I have the crashdump if you need me to investigate anything.