Bug 27263

Summary:	Gallium R300g can crash PC when running celestia.
Product:	Mesa	Reporter:	Chris Rankin <rankincj>
Component:	Drivers/DRI/r300	Assignee:	Default DRI bug account <dri-devel>
Status:	RESOLVED FIXED	QA Contact:
Severity:	normal
Priority:	medium
Version:	git
Hardware:	x86 (IA32)
OS:	Linux (All)
Whiteboard:
i915 platform:		i915 features:

Description Chris Rankin 2010-03-23 06:22:05 UTC

I tried running celestia 1.5.1 using the R300g driver from git, vanilla Linux 2.6.33.1 and libdrm 2.4.18, but celestia hung. In one case, the entire PC then locked up shortly afterwards. However, in one case it recovered and recorded these messages in dmesg instead:

[drm:radeon_fence_wait] *ERROR* fence(f04020a0:0x00081CC5) 504ms timeout going to reset GPU
[drm:r300_ga_reset] *ERROR* VAP & CP still busy (RBBM_STATUS=0x84000140)
[drm] GA reset succeed (RBBM_STATUS=0x00000140)
[drm] radeon: cp idle (0x10000000)
[drm] radeon: ring at 0x00000000C8000000
[drm] ring test succeeded in 0 usecs
[drm] GPU reset succeed (RBBM_STATUS=0x00000140)
[drm:radeon_fence_wait] *ERROR* fence(f04020a0:0x00081CC5) 512ms timeout
[drm:radeon_fence_wait] *ERROR* last signaled fence(0x00081CC5)
[drm:radeon_fence_wait] *ERROR* fence(f5967900:0x00081CD2) 508ms timeout
[drm:radeon_fence_wait] *ERROR* last signaled fence(0x00081CD5)
[drm:radeon_fence_wait] *ERROR* fence(f059d0a0:0x00081CE9) 504ms timeout going to reset GPU
[drm:r300_ga_reset] *ERROR* VAP & CP still busy (RBBM_STATUS=0x84000140)
[drm] GA reset succeed (RBBM_STATUS=0x00000140)
[drm] radeon: cp idle (0x10000000)
[drm] radeon: ring at 0x00000000C8000000
[drm] ring test succeeded in 0 usecs
[drm] GPU reset succeed (RBBM_STATUS=0x00000140)
[drm:radeon_fence_wait] *ERROR* fence(f059d0a0:0x00081CE9) 512ms timeout
[drm:radeon_fence_wait] *ERROR* last signaled fence(0x00081CE9)
[drm:radeon_fence_wait] *ERROR* fence(f0bc4900:0x00081CEC) 492ms timeout
[drm:radeon_fence_wait] *ERROR* last signaled fence(0x00081CF9)
[drm:radeon_fence_wait] *ERROR* fence(c4c0b260:0x00081D3F) 504ms timeout going to reset GPU
[drm:r300_ga_reset] *ERROR* VAP & CP still busy (RBBM_STATUS=0x84000140)
[drm] GA reset succeed (RBBM_STATUS=0x00000140)
[drm] radeon: cp idle (0x10000000)
[drm] radeon: ring at 0x00000000C8000000
[drm] ring test succeeded in 0 usecs
[drm] GPU reset succeed (RBBM_STATUS=0x00000140)
[drm:radeon_fence_wait] *ERROR* fence(c4c0b260:0x00081D3F) 512ms timeout
[drm:radeon_fence_wait] *ERROR* last signaled fence(0x00081D3F)
[drm:radeon_fence_wait] *ERROR* fence(c6e62900:0x00081D43) 488ms timeout
[drm:radeon_fence_wait] *ERROR* last signaled fence(0x00081D4F)

Comment 1 Chris Rankin 2010-03-23 06:22:55 UTC

This is with a Radeon 9550 (RV350).

Comment 2 Corbin Simpson 2010-03-26 05:26:32 UTC

38c7a01b6c220ad04c5754602673ad3cf36ad508 should fix the hardlock, although it renders Earth incorrectly. b6df7aed60189d5f28a139c6fe351022ca2907a4 keeps it from crashing with a bad vertex program.

Closing for now; the incorrect rendering will get fixed at a later date.

Comment 3 Chris Rankin 2010-03-26 06:21:23 UTC

(In reply to comment #2)
> 38c7a01b6c220ad04c5754602673ad3cf36ad508 should fix the hardlock, although it
> renders Earth incorrectly. b6df7aed60189d5f28a139c6fe351022ca2907a4 keeps it
> from crashing with a bad vertex program.
> 
> Closing for now; the incorrect rendering will get fixed at a later date.

It is no longer locking my PC up either :-). If the "incorrect rendering" that you're talking about is the "blank" dayside of the Earth then in my case this was due to celestia now using the OpenGL 2.0 render path with Gallium. Switching back to the OpenGL vertex program render path (Ctrl-V) fixed things. Celestia was still a lot slower under Gallium than classic Mesa, though.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.