linux-kernel - Repeated 'watchdog soft lockup' messages

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>] [day] [month] [year] [list]

Message-ID: <4e624f005f3744eb82adfa1e5ef0d683@AcuMS.aculab.com>
Date:   Thu, 12 Nov 2020 10:50:24 +0000
From:   David Laight <David.Laight@...LAB.COM>
To:     "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "Peter Zijlstra" <peterz@...radead.org>
Subject: Repeated 'watchdog soft lockup' messages

Due to a coding fubar (in my own driver) I managed to sleep
a process that held a spinlock.
Unluckily all 8 cpus ended up trying to acquire the spinlock
before the process woke.
As might be expected the system froze.

I'd got a serial console connected so could see the kernel messages.
It was almost continuously generating output saying that each
cpu was 'stuck for 22s!' (sometimes 23s).

What I don't understand is what was resetting the watchdog
so that I got multiple messages for each cpu at intervals
of (typically) 28 seconds.

The watchdog code seems to only output the trace once.
But that clearly wasn't working.
Any idea why?

kernel is 5.9.0

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)