Received-SPF: pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 2620:137:e000::3:5 as permitted sender) client-ip=2620:137:e000::3:5;
Feedback-ID: id4a34324:Fastmail
Date:   Tue, 10 Oct 2023 09:25:32 -0700
From:   Andres Freund <andres@anarazel.de>
To:     Thorsten Leemhuis <regressions@leemhuis.info>
Cc:     Shreeya Patel <shreeya.patel@collabora.com>,
        Theodore Ts'o <tytso@mit.edu>, linux-ext4@vger.kernel.org,
        Ricardo =?utf-8?Q?Ca=C3=B1uelo?= <ricardo.canuelo@collabora.com>,
        "gustavo.padovan@collabora.com" <gustavo.padovan@collabora.com>,
        groeck@google.com, zsm@google.com, garrick@google.com,
        Linux regressions mailing list <regressions@lists.linux.dev>
Subject: Re: [syzbot] INFO: task hung in ext4_fallocate
Message-ID: <20231010162532.kgtphbupib5gwwft@awork3.anarazel.de>
References: <d89989ef-e53b-050e-2916-a4f06433798b@collabora.com>
 <ZQjKOjrjDYsoXBXj@mit.edu>
 <20231003141138.owt6qwqyf4slgqgp@alap3.anarazel.de>
 <20231003232505.GA444157@mit.edu>
 <20231004004247.zkswbseowwwc6vvk@alap3.anarazel.de>
 <604bc623-a090-005b-cbfc-39805a8a7b20@collabora.com>
 <c9632187-bcbc-483f-8b18-1fa403b46ebb@leemhuis.info>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <c9632187-bcbc-483f-8b18-1fa403b46ebb@leemhuis.info>
Precedence: bulk

Hi,

On 2023-10-10 11:21:47 +0200, Thorsten Leemhuis wrote:
> Andres Freund: as it seems you have a different issue, have you or
> somebody else done anything to get us closer to a solution (like a
> bisection)? Doesn't look like it, but it's quite possible that I missed
> something.

I've indeed not performed a bisection - unfortunately it doesn't quite seem
viable at this point, I've only hit the issue after using the kernel
interactively for several hours. Being confident that some kernel version
doesn't hit the issue would require using a kernel version for quite a long
time...


> I just want to ensure that this is not lost if you care about the problems
> you encountered; makes we wonder if reporting it separately (e.g. in a new
> thread) might be wise.

Yes, probably a good idea. I'll try to do so soon.

Greetings,

Andres Freund