Bug 94331

Summary: Random panic with radeon
Product: Drivers Reporter: sulamiification
Component: Video(DRI - non Intel)Assignee: drivers_video-dri
Status: NEW ---    
Severity: high CC: szg00000
Priority: P1    
Hardware: x86-64   
OS: Linux   
Kernel Version: 4.0.0-rc1-00151-ga38ecbb Subsystem:
Regression: No Bisected commit-id:
Attachments: Kernel log

Description sulamiification 2015-03-05 11:45:55 UTC 
Created attachment 169221 [details]
Kernel log

Yesterday my 4.0.0-rc1 randomly panicked. Because of the panic messsage I am assuming it is at least related to radeon.

Hardware:
- Intel Xeon E3-1245v3
- AMD Radeon HD7970
- ASRock B85M Pro4

Software:
- Mainline Kernel built on Monday (4.0.0-rc1-00151-ga38ecbb)
- radeon driver

The panic happened after about one day of uptime, I was running cwm + compton, an OpenBSD VM in qemu/kvm without graphics in the background and a HTML5 YouTube video in Firefox.

I added the log up until then as an attachment, but I cannot find anything suspicious. The "wm" segfault is my fault, I am working on a Wayland compositor. The bridge mess as well.

I made a photo of the panic for later examination, which can be found here:

https://i.imgur.com/I7xBwWD.jpg

So far I have not been able to reproduce this issue.
Comment 1 Michel Dänzer 2015-03-06 01:28:34 UTC 
(In reply to sulamiification from comment #0)
> Because of the panic messsage I am assuming it is at least related to radeon.

Why is that? If you mean because of 'drm_kms_helper', that message appears regardless of what causes the panic.
Comment 2 sulamiification 2015-03-07 16:28:00 UTC 
(In reply to Michel Dänzer from comment #1)
> (In reply to sulamiification from comment #0)
> > Because of the panic messsage I am assuming it is at least related to
> radeon.
> 
> Why is that? If you mean because of 'drm_kms_helper', that message appears
> regardless of what causes the panic.

To be honest, it is just my best bet, "Compositor" also plays a role in that. If there is any way to find out what actually happened to cause this page fault I would be happy to help out in any way I can.

I am also puzzled by the "<<EOE>> <UNK>" block, what does it say?