Bug 816873

Summary: radeon [m76/rv630] gpu&system hang while playing quake3
Product: [openSUSE] openSUSE 12.3 Reporter: Iakov Karpov <srid>
Component: X.OrgAssignee: E-mail List <xorg-maintainer-bugs>
Status: RESOLVED WONTFIX QA Contact: E-mail List <xorg-maintainer-bugs>
Severity: Normal    
Priority: P3 - Medium    
Version: Final   
Target Milestone: ---   
Hardware: x86-64   
OS: openSUSE 12.3   
Whiteboard:
Found By: --- Services Priority:
Business Priority: Blocker: ---
Marketing QA Status: --- IT Deployment: ---
Attachments: part of /var/log/messages, since first mention of gpu problems till reboot

Description Iakov Karpov 2013-04-24 06:49:14 UTC
Created attachment 536618 [details]
part of /var/log/messages, since first mention of gpu problems till reboot

User-Agent:       Mozilla/5.0 (X11; Linux x86_64; rv:20.0) Gecko/20100101 Firefox/20.0

I'm running openSUSE 12.3 on Intel Santa Rosa -based laptop with ATI M76 video (also known as rv630 or Mobility Radeon HD2600), with opensource driver:

$glxinfo | grep renderer
OpenGL renderer string: Gallium 0.4 on AMD RV630

Last time I've played Quake 3 I've got a gpu hang. The picture has stopped, the screen started to flicker. I tried to switch to console but a few seconds later screen went dark and system was unresponsive to any key (including power button) and was not reachable by network.
The ioquake3 package is from "games" repository. All available updates are installed.

Reproducible: Always

Steps to Reproduce:
1.
2.
3.
Comment 1 Stefan Dirsch 2013-04-24 08:37:34 UTC
2013-04-24T10:16:57.728702+04:00 n-day0 kernel: [13944.762054] radeon 0000:01:00.0: GPU lockup CP stall for more than 10000msec
2013-04-24T10:16:57.728732+04:00 n-day0 kernel: [13944.762114] radeon 0000:01:00.0: GPU lockup (waiting for 0x00000000000e9c5f last fence id 0x00000000000e9bf4)
2013-04-24T10:16:57.728737+04:00 n-day0 kernel: [13944.762172] [drm:r600_ib_test] *ERROR* radeon: fence wait failed (-35).
2013-04-24T10:16:57.728740+04:00 n-day0 kernel: [13944.762214] [drm:radeon_ib_ring_tests] *ERROR* radeon: failed testing IB on GFX ring (-35).
2013-04-24T10:16:57.728744+04:00 n-day0 kernel: [13944.762263] radeon 0000:01:00.0: ib ring test failed (-35).
2013-04-24T10:16:57.728747+04:00 n-day0 kernel: [13944.763349] radeon 0000:01:00.0: GPU softreset 
2013-04-24T10:16:57.728750+04:00 n-day0 kernel: [13944.763354] radeon 0000:01:00.0:   R_008010_GRBM_STATUS=0xA0003030
2013-04-24T10:16:57.728753+04:00 n-day0 kernel: [13944.763358] radeon 0000:01:00.0:   R_008014_GRBM_STATUS2=0x00000003
2013-04-24T10:16:57.728756+04:00 n-day0 kernel: [13944.763362] radeon 0000:01:00.0:   R_000E50_SRBM_STATUS=0x200210C0
2013-04-24T10:16:57.728759+04:00 n-day0 kernel: [13944.763365] radeon 0000:01:00.0:   R_008674_CP_STALLED_STAT1 = 0x00000000
2013-04-24T10:16:57.728762+04:00 n-day0 kernel: [13944.763369] radeon 0000:01:00.0:   R_008678_CP_STALLED_STAT2 = 0x00000000
2013-04-24T10:16:57.728765+04:00 n-day0 kernel: [13944.763373] radeon 0000:01:00.0:   R_00867C_CP_BUSY_STAT     = 0x00020986
2013-04-24T10:16:57.728769+04:00 n-day0 kernel: [13944.763377] radeon 0000:01:00.0:   R_008680_CP_STAT          = 0x800286C5
2013-04-24T10:16:57.948456+04:00 n-day0 kernel: [13944.983221] radeon 0000:01:00.0: Wait for MC idle timedout !
2013-04-24T10:16:57.948494+04:00 n-day0 kernel: [13944.983267] radeon 0000:01:00.0:   R_008020_GRBM_SOFT_RE


Thanks for the debug output. Apparently the GPU reset works several times before the machine freezes.
Comment 2 Stefan Dirsch 2015-01-07 14:37:52 UTC
Product is no longer supported. In case the issue is still reproducable on a maintainerd product (at that momement: openSUSE 13.1 or later), feel free to reopen.