Re: [syzbot] [xfs?] INFO: task hung in __fdget_pos (4)

From: Mateusz Guzik
Date: Mon Sep 04 2023 - 02:09:13 EST


On 9/4/23, Theodore Ts'o <tytso@xxxxxxx> wrote:
> On Mon, Sep 04, 2023 at 11:45:03AM +1000, Dave Chinner wrote:
>> Entirely possible - this is syzbot we are talking about here.
>> Especially if reiser or ntfs has been tested back before the logs we
>> have start, as both are known to corrupt memory and/or leak locks
>> when trying to parse corrupt filesystem images that syzbot feeds
>> them. That's why we largely ignore syzbot reports that involve
>> those filesystems...
>>
>> Unfortunately, the logs from what was being done around when the
>> tasks actually hung are long gone (seems like only the last 20-30s
>> of log activity is reported) so when the hung task timer goes off
>> at 143s, there is nothing left to tell us what might have caused it.
>>
>> IOWs, it's entirely possible that it is a memory corruption that
>> has resulted in a leaked lock somewhere...
>
> ... and this is why I ignore any syzbot report that doesn't have a C
> reproducer. Life is too short to waste time with what is very likely
> syzbot noise.... And I'd much rather opt out of the gamification of
> syzbot dashboards designed to use dark patterns to guilt developers to
> work on "issues" that very likely have no real impact on real life
> actual user impact, if it might cause developers and maintainers to
> burn out and quit.
>
> Basically, if syzbot won't prioritize things for us, it's encumbent on
> us to prioritize things for our own mental health. And so syzbot
> issues without a real reproducer are very low on my priority list; I
> have things I can work on that are much more likely to make real world
> impact. Even ones that have a real reproducer, there are certain
> classes of bugs (e.g., "locking bugs" that require a badly corrupted
> file system, or things that are just denial of service attacks if
> you're too stupid to insert a USB thumb drive found in a parking lock
> --- made worse by GNOME who has decided to default mount any random
> USB thumb drive inserted into a system, even a server system that has
> GNOME installed, thanks to some idiotic decision made by some random
> Red Hat product manager), that I just ignore because I don't have
> infinite amounts of time to coddle stupid Red Hat distro tricks.
>

Hello everyone.

When I first stumbled upon this report I was almost completely
oblivious to syzbot vs fsdevel -- I only knew you guys are not fond of
ntfs reports, which made sense but was not indicating there are much
bigger issues.

Given this and other responses I poked around the history and that
made it apparent there are long-standing non-technical problems going
here, none of which I intend to deal with.

That is to say I'm bailing from this thread.

Cheers and sorry for poking the nest,
--
Mateusz Guzik <mjguzik gmail.com>