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]
Date:   Wed, 25 Oct 2023 09:14:05 -0700
From:   Andres Freund <andres@...razel.de>
To:     Jens Axboe <axboe@...nel.dk>
Cc:     Dave Chinner <david@...morbit.com>, Theodore Ts'o <tytso@....edu>,
        Thorsten Leemhuis <regressions@...mhuis.info>,
        Shreeya Patel <shreeya.patel@...labora.com>,
        linux-ext4@...r.kernel.org,
        Ricardo CaƱuelo <ricardo.canuelo@...labora.com>,
        gustavo.padovan@...labora.com, zsm@...gle.com, garrick@...gle.com,
        Linux regressions mailing list <regressions@...ts.linux.dev>,
        io-uring@...r.kernel.org
Subject: Re: task hung in ext4_fallocate #2

Hi,

On 2023-10-25 09:36:01 -0600, Jens Axboe wrote:
> On 10/25/23 9:31 AM, Andres Freund wrote:
> > Hi,
> > 
> > On 2023-10-24 18:34:05 -0600, Jens Axboe wrote:
> >> Yeah I'm going to do a revert of the io_uring side, which effectively
> >> disables it. Then a revised series can be done, and when done, we could
> >> bring it back.
> > 
> > I'm queueing a test to confirm that the revert actually fixes things.
> > Is there still benefit in testing your other patch in addition
> > upstream?
> 
> Don't think there's much point to testing the quick hack, I believe it
> should work. So testing the most recent revert is useful, though I also
> fully expect that to work.

I'll leave it running for a few hours, just to be sure.


> And then we can test the re-enable once that is sent out, I did prepare a
> series. But timing is obviously unfortunate for that, as it'll miss 6.6 and
> now also 6.7 due to the report timing.

Yea, it's too bad. I'm somewhat lucky to have hit it at all, it was just due
to having procrastinated on talk preparation and having to test things on my
laptop instead of my workstation (where the same workload never triggered the
problem), while travelling.


> FWIW, I wrote a small test case which does seem to trigger it very fast,
> as expected:

Ah, nice.

It's too bad stuff like this can't be discovered by lockdep et al..

Greetings,

Andres Freund

Powered by blists - more mailing lists