[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CANDhNCrkDxPqyV5BhawdO8H2u+TbMVQjC51jwwY=qn2Q5RhTAg@mail.gmail.com>
Date: Wed, 11 Dec 2024 13:08:50 -0800
From: John Stultz <jstultz@...gle.com>
To: Bert Karwatzki <spasswolf@....de>
Cc: Thomas Gleixner <tglx@...utronix.de>, Metin.Kaya@....com, boqun.feng@...il.com,
bsegall@...gle.com, daniel.lezcano@...aro.org, dave@...olabs.net,
dietmar.eggemann@....com, joelaf@...gle.com, juri.lelli@...hat.com,
kernel-team@...roid.com, kprateek.nayak@....com, linux-kernel@...r.kernel.org,
longman@...hat.com, mgorman@...e.de, mingo@...hat.com, paulmck@...nel.org,
peterz@...radead.org, qyousef@...alina.io, rostedt@...dmis.org,
vincent.guittot@...aro.org, vschneid@...hat.com, will@...nel.org,
xuewen.yan94@...il.com, zezeozue@...gle.com
Subject: Re: commit 894d1b3db41c leads to frequent hangs when booting
On Wed, Dec 11, 2024 at 12:44 PM Bert Karwatzki <spasswolf@....de> wrote:
>
> Am Mittwoch, dem 11.12.2024 um 12:27 -0800 schrieb John Stultz:
> > On Wed, Dec 11, 2024 at 11:19 AM Thomas Gleixner <tglx@...utronix.de> wrote:
> > >
> > > On Wed, Dec 11 2024 at 19:25, Bert Karwatzki wrote:
> > > > I tried to debug this with CONFIG_LOCKDEP=y in v6.13-rc2, but using CONFIG_LOCKDEP=y
> > > > makes the hangs disappear or far less likely, and I get this warning
> > > > (2 examples from two boots):
> > > >
> > > > [ 17.203857] [ T1337] BUG: MAX_LOCKDEP_CHAIN_HLOCKS too low!
> > >
> > > config LOCKDEP_CHAINS_BITS
> > > int "Bitsize for MAX_LOCKDEP_CHAINS"
> > > depends on LOCKDEP && !LOCKDEP_SMALL
> > > range 10 21
> > > default 16
> > > help
> > > Try increasing this value if you hit "BUG: MAX_LOCKDEP_CHAINS too low!" message.
> > >
> > > Can you increase the chain bits config and try again?
> > >
> >
> > Thanks for the report and bisecting this down!
> >
> > Can you double check that the following commit is in your tree? I
> > think it has landed just in the last few days upstream:
> > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=82f9cc094975240885c93effbca7f4603f5de1bf
> >
> > thanks
> > -john
>
> v6.13-rc2 does not contain commit 82f9cc094975, but linux-next-20241210 does
> contain it and shows the same buggy behaviour as v6.13-rc2, I'm currently
:/ hrm.
> testing if linux-next-20241010 can be fixed by the same revert as v6.12-rc2.
> I didn't bisect this in the linux-next tree to avoid an overlapping bug which
> also caused a hang on boot (this has already been fixed):
> https://lore.kernel.org/linux-mm/CAOQ4uxgVNGmLqURdO0wf3vo=K-a2C--ZLKFzXw-22PJdkBjEdA@mail.gmail.com/T/#t
Could you share the config you're using to reproduce this?
I'll try to work up a few smaller reversions (along with some extra
debug details) to see if we can isolate where things are going awry.
thanks
-john
Powered by blists - more mailing lists