Bug 92428

Summary: Sword Coast Legends causing GPU lockup. radeon 0000:01:00.0: GPU lockup (current fence id 0x000000000000afd1 last fence id 0x000000000000b194 on ring 3)
Product: Mesa Reporter: James <h0ser>
Component: Drivers/Gallium/radeonsiAssignee: Default DRI bug account <dri-devel>
Status: RESOLVED MOVED QA Contact: Default DRI bug account <dri-devel>
Severity: normal    
Priority: medium CC: h0ser
Version: 10.6   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments: System Logs
Messages printed into the journalctl log about this crash

Description James 2015-10-11 03:05:36 UTC
Created attachment 118821 [details]
System Logs

Bear with me as it's my first time submitting a bug report. When loading a level in Sword Coast Legends, the driver crashes and hard locks the PC. Attached is what I believe are the relevant portions of the logs.
Comment 1 James 2015-10-11 03:06:44 UTC
 glxinfo | grep OpenGL
OpenGL vendor string: X.Org
OpenGL renderer string: Gallium 0.4 on AMD PITCAIRN
OpenGL core profile version string: 3.3 (Core Profile) Mesa 10.6.9 (git-8957b69)
OpenGL core profile shading language version string: 3.30
OpenGL core profile context flags: (none)
OpenGL core profile profile mask: core profile
OpenGL core profile extensions:
OpenGL version string: 3.0 Mesa 10.6.9 (git-8957b69)
OpenGL shading language version string: 1.30
OpenGL context flags: (none)
OpenGL extensions:
OpenGL ES profile version string: OpenGL ES 3.0 Mesa 10.6.9 (git-8957b69)
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.00
OpenGL ES profile extensions:
Comment 2 James 2015-10-11 03:07:21 UTC
01:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Pitcairn PRO [Radeon HD 7850 / R7 265 / R9 270 1024SP] [1002:6819]
	Subsystem: PC Partner Limited / Sapphire Technology Radeon HD 7850 2GB GDDR5 DVI-I/DVI-D/HDMI/DP [174b:e221]
	Kernel driver in use: radeon
	Kernel modules: radeon
Comment 3 Francesco Biscani 2015-10-23 20:22:47 UTC
I am also experiencing a lockup with this game. In my case the lockup happens before I get to the main menu right after starting the game. I have a radeon r9 270x, and I am using the radeonsi driver (version 7.5.0), mesa 11.0.3, llvm 3.7.0, xorg server 1.17.2. The OS is Linux 4.2.2 64 bit.

# glxinfo|grep -i opengl
OpenGL vendor string: X.Org
OpenGL renderer string: Gallium 0.4 on AMD PITCAIRN (DRM 2.43.0, LLVM 3.7.0)
OpenGL core profile version string: 4.1 (Core Profile) Mesa 11.0.3
OpenGL core profile shading language version string: 4.10
OpenGL core profile context flags: (none)
OpenGL core profile profile mask: core profile
OpenGL core profile extensions:
OpenGL version string: 3.0 Mesa 11.0.3
OpenGL shading language version string: 1.30
OpenGL context flags: (none)
OpenGL extensions:
OpenGL ES profile version string: OpenGL ES 3.0 Mesa 11.0.3
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.00


# sudo lspci |grep -i vga
01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Curacao XT [Radeon R7 370 / R9 270X/370 OEM]
Comment 4 l.gambetta 2015-12-18 08:51:45 UTC
I can confirm this bug. The game crashed my PC on the load screen.

My specs:

CPU AMD A10-7850K
GPU ATI Radeon 7850 1Gb RAM
8 GB RAM
Kernel 4.3.3, 64 bit
xf86-video-ati 7.6.1
mesa 11.0.7
Manjaro Linux

$ glxinfo | grep OpenGL
OpenGL vendor string: X.Org
OpenGL renderer string: Gallium 0.4 on AMD PITCAIRN (DRM 2.43.0, LLVM 3.7.0)
OpenGL core profile version string: 4.1 (Core Profile) Mesa 11.0.7
OpenGL core profile shading language version string: 4.10
OpenGL core profile context flags: (none)
OpenGL core profile profile mask: core profile
OpenGL core profile extensions:
OpenGL version string: 3.0 Mesa 11.0.7
OpenGL shading language version string: 1.30
OpenGL context flags: (none)
OpenGL extensions:
OpenGL ES profile version string: OpenGL ES 3.0 Mesa 11.0.7
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.00
OpenGL ES profile extensions:

01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Pitcairn PRO [Radeon HD 7850 / R7 265 / R9 270 1024SP]
Comment 5 l.gambetta 2015-12-18 08:52:45 UTC
Created attachment 120582 [details]
Messages printed into the journalctl log about this crash
Comment 6 Nicolai Hähnle 2015-12-18 20:15:27 UTC
Thanks for the reports. VM faults are tricky to find - can one of you provide an apitrace of a lockup so that we can reproduce this?
Comment 7 James 2015-12-19 06:01:33 UTC
Since the it exceeds the file limit, I've uploaded the apitrace to my dropbox. 

The link is here. https://www.dropbox.com/s/15mfgng35b2rmt2/SwordCoast.x86_64.1.trace?dl=0
Comment 8 Michel Dänzer 2015-12-21 07:34:58 UTC
(In reply to James from comment #7)
> The link is here.
> https://www.dropbox.com/s/15mfgng35b2rmt2/SwordCoast.x86_64.1.trace?dl=0

I don't get any VM faults when replaying that, let alone a lockup, neither on Kaveri nor Verde. Does replaying this apitrace reproduce the problem for you?
Comment 9 James 2015-12-22 04:59:56 UTC
(In reply to Michel Dänzer from comment #8)
> (In reply to James from comment #7)
> > The link is here.
> > https://www.dropbox.com/s/15mfgng35b2rmt2/SwordCoast.x86_64.1.trace?dl=0
> 
> I don't get any VM faults when replaying that, let alone a lockup, neither
> on Kaveri nor Verde. Does replaying this apitrace reproduce the problem for
> you?

Yes, it replays up to the splash screen then hard locks my PC. Although I'm using a 7850  wouldn't be Kaveri or Verde. It would be Pitcarin/GCN. Don't know if that makes a difference.
Comment 10 James 2015-12-22 05:00:25 UTC
(In reply to James from comment #9)
> (In reply to Michel Dänzer from comment #8)
> > (In reply to James from comment #7)
> > > The link is here.
> > > https://www.dropbox.com/s/15mfgng35b2rmt2/SwordCoast.x86_64.1.trace?dl=0
> > 
> > I don't get any VM faults when replaying that, let alone a lockup, neither
> > on Kaveri nor Verde. Does replaying this apitrace reproduce the problem for
> > you?
> 
> Yes, it replays up to the splash screen then hard locks my PC. Although I'm
> using a 7850 so it wouldn't be Kaveri or Verde. It would be Pitcarin/GCN. Don't
> know if that makes a difference.
Comment 11 Marek Olšák 2016-10-17 20:37:04 UTC
Does it still happen with current Mesa git? If so, can you try to run it with this environment variable?

R600_DEBUG=switch_on_eop
Comment 12 GitLab Migration User 2019-09-25 17:53:29 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/mesa/mesa/issues/1224.

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.