[Bug 110848] Everything using GPU gets stuck after running+killing parallel Media loads (after running 3D benchmarks)

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=110848

--- Comment #32 from Chris Wilson <chris at chris-wilson.co.uk> ---
(In reply to David Weinehall from comment #31)
> Hmmm, so you're saying that given enough time the machine could/should
> recover?
> 
> I left the machine on 4h after the test concluded (hung), but no luck. The
> desktop clock of the machine stopped updating at the end of the task, the
> machine stopped replying to ping, and when I checked 4h later the machine
> still didn't respond to keyboard input.

There are a few corner cases where oom just eats itself, but nowadays they
usually result in a panic + reboot. (Do you have reboot on panic set?) Give or
take a few HW faults, a lockup usually means a panic. I would personally use
remote logging (via netconsole) in case it is a panic. If not, and with lockdep
and kernel NMI watchdog running, it just quietly hangs, we found ourselves a
more "interesting" issue.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the QA Contact for the bug.
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx-bugs/attachments/20190802/ecb03a6d/attachment.html>