linux-kernel - Crash in fair scheduler

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-next>] [day] [month] [year] [list]

Message-ID: <1575364273836.74450@mentor.com>
Date:   Tue, 3 Dec 2019 09:11:14 +0000
From:   "Schmid, Carsten" <Carsten_Schmid@...tor.com>
To:     "mingo@...hat.com" <mingo@...hat.com>,
        "peterz@...radead.org" <peterz@...radead.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Crash in fair scheduler

Hi maintainers of the fair scheduler,

we had a crash in the fair scheduler and analysis shows that this could happen again.
Happened on 4.14.86 (LTS series) but failing code path still exists in 5.4-rc2 (and 4.14.147 too).

I have collected relevant data from the core dump and put together with the source.
So i was able to explain why the crash happened, but as i am not an expert on this
scheduler stuff i can't tell if any of the data fed into the code path is ok or not.

In the analysis appended you can find the details; please check if we ran into a corner
case that nobody expected yet or if the data is somehow corrupted (i really can't tell, for me it looks fine).
Can you please check if this is a bug in the scheduler?

To give some "surrounding" info:
It looks like a task has slept for a long time, even through some suspend/resume cycles.
When this task had to wake up, the failure occured.

If you need more data from the core dump, let me know.

Best regards
Carsten
View attachment "fair_scheduler_crash.txt" of type "text/plain" (11368 bytes)