Bug 84614 - radeon gpu crash /kernel crash when using dynamic light in borderlands 2
Summary: radeon gpu crash /kernel crash when using dynamic light in borderlands 2
Status: RESOLVED FIXED
Alias: None
Product: Mesa
Classification: Unclassified
Component: Drivers/Gallium/radeonsi (show other bugs)
Version: git
Hardware: x86-64 (AMD64) Linux (All)
: medium major
Assignee: Default DRI bug account
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-10-03 00:37 UTC by Paulo Dias
Modified: 2015-08-28 15:26 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
xorg after crash (49.53 KB, text/plain)
2014-10-08 00:46 UTC, Paulo Dias
Details

Description Paulo Dias 2014-10-03 00:37:22 UTC
if i activate dynamic light in borderlands 2, it works and it quite beautiful, that is , till i start running with my char. then the kernel/gpu crashes with a lot of:

Oct  2 21:14:40 kerberos kernel: [  717.282422] radeon 0000:03:00.0: GPU fault detected: 147 0x0a820801
Oct  2 21:14:40 kerberos kernel: [  717.282429] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0BB00854
Oct  2 21:14:40 kerberos kernel: [  717.282432] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008001
Oct  2 21:14:40 kerberos kernel: [  717.282435] VM fault (0x01, vmid 1) at page 196085844, read from TC (8)
Oct  2 21:14:40 kerberos kernel: [  717.358511] radeon 0000:03:00.0: GPU fault detected: 147 0x0a420801
Oct  2 21:14:40 kerberos kernel: [  717.358518] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0BB00852
Oct  2 21:14:40 kerberos kernel: [  717.358520] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008001
Oct  2 21:14:40 kerberos kernel: [  717.358523] VM fault (0x01, vmid 1) at page 196085842, read from TC (8)
Oct  2 21:14:40 kerberos kernel: [  717.808432] radeon 0000:03:00.0: GPU fault detected: 147 0x0a420801
Oct  2 21:14:40 kerberos kernel: [  717.808439] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0BB00852
Oct  2 21:14:40 kerberos kernel: [  717.808441] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008001
Oct  2 21:14:40 kerberos kernel: [  717.808444] VM fault (0x01, vmid 1) at page 196085842, read from TC (8)
Oct  2 21:14:40 kerberos kernel: [  717.897558] radeon 0000:03:00.0: GPU fault detected: 147 0x0a420801
Oct  2 21:14:40 kerberos kernel: [  717.897564] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0BB00852
Oct  2 21:14:40 kerberos kernel: [  717.897566] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008001
Oct  2 21:14:40 kerberos kernel: [  717.897568] VM fault (0x01, vmid 1) at page 196085842, read from TC (8)
Oct  2 21:14:40 kerberos kernel: [  717.897610] radeon 0000:03:00.0: GPU fault detected: 147 0x0a420801
Oct  2 21:14:40 kerberos kernel: [  717.897612] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0BB00852
Oct  2 21:14:40 kerberos kernel: [  717.897614] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008001
Oct  2 21:14:40 kerberos kernel: [  717.897616] VM fault (0x01, vmid 1) at page 196085842, read from TC (8)
Oct  2 21:14:42 kerberos kernel: [  719.466784] radeon 0000:03:00.0: GPU fault detected: 147 0x0a420801
Oct  2 21:14:42 kerberos kernel: [  719.466790] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0BB00852
Oct  2 21:14:42 kerberos kernel: [  719.466793] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008001
Oct  2 21:14:42 kerberos kernel: [  719.466796] VM fault (0x01, vmid 1) at page 196085842, read from TC (8)
Oct  2 21:14:42 kerberos kernel: [  719.466822] radeon 0000:03:00.0: GPU fault detected: 147 0x0a420801
Oct  2 21:14:42 kerberos kernel: [  719.466824] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0BB00852
Oct  2 21:14:42 kerberos kernel: [  719.466827] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008001
Oct  2 21:14:42 kerberos kernel: [  719.466829] VM fault (0x01, vmid 1) at page 196085842, read from TC (8)
Oct  2 21:14:42 kerberos kernel: [  719.635386] radeon 0000:03:00.0: GPU fault detected: 147 0x0a620801
Oct  2 21:14:42 kerberos kernel: [  719.635393] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0BB00853
Oct  2 21:14:42 kerberos kernel: [  719.635395] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008001
Oct  2 21:14:42 kerberos kernel: [  719.635398] VM fault (0x01, vmid 1) at page 196085843, read from TC (8)
Oct  2 21:14:42 kerberos kernel: [  719.635416] radeon 0000:03:00.0: GPU fault detected: 147 0x0a820801
Oct  2 21:14:42 kerberos kernel: [  719.635418] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0BB00854
Oct  2 21:14:42 kerberos kernel: [  719.635420] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008001
Oct  2 21:14:42 kerberos kernel: [  719.635421] VM fault (0x01, vmid 1) at page 196085844, read from TC (8)
Oct  2 21:14:42 kerberos kernel: [  719.975165] radeon 0000:03:00.0: GPU fault detected: 147 0x0a420801
Oct  2 21:14:42 kerberos kernel: [  719.975172] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0BB00852
Oct  2 21:14:42 kerberos kernel: [  719.975175] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008001
Oct  2 21:14:42 kerberos kernel: [  719.975178] VM fault (0x01, vmid 1) at page 196085842, read from TC (8)
Oct  2 21:14:42 kerberos kernel: [  719.975227] radeon 0000:03:00.0: GPU fault detected: 147 0x0a420801
Oct  2 21:14:42 kerberos kernel: [  719.975230] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0BB00852
Oct  2 21:14:42 kerberos kernel: [  719.975232] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008001

llvm 5.2, kernel 3.17 rc7, latest microcode, mesa 10.4 git from oibaf ppa, same for radeon driver.
Comment 1 Michel Dänzer 2014-10-03 01:01:21 UTC
Probably a radeonsi issue. Can you try creating an apitrace which reproduces the problem?
Comment 2 Paulo Dias 2014-10-03 03:48:08 UTC
how can i do that? the kernel freezes and not even sysrq keys work anymore, i need to power off the pc.
Comment 3 Filipp Andjelo 2014-10-04 09:22:01 UTC
Almost the same problem here, full system crash, but I can't even reach the main menu. System is completely dead on the first rendering frame of the scene you see in the main menu.

Running HD7850 on linux 3.16.3, mesa 10.3 and llvm 3.5  (all stock)
Comment 4 Michel Dänzer 2014-10-06 06:54:33 UTC
(In reply to Filipp Andjelo from comment #3)
> Almost the same problem here, full system crash, but I can't even reach the
> main menu. System is completely dead on the first rendering frame of the
> scene you see in the main menu.

Almost, but not quite the same problem then. Please file your own report.

Paulo, please attach the full /var/log/Xorg.0.log and the output of dmesg showing the radeon driver initialization.


(In reply to Paulo Dias from comment #2)
> how can i do that? the kernel freezes and not even sysrq keys work anymore,
> i need to power off the pc.

Since you were able to retrieve the GPU fault messages from the kernel, you might be able to retrieve a working apitrace as well.
Comment 5 Paulo Dias 2014-10-06 16:46:56 UTC
Well with the latest update from oibaf ppa, im having the same problem has Flip Andjelo, as soon as borderlands enters the main menu, the gpu crashes

this is what i get from dmesg:


Oct  4 18:44:06 kerberos kernel: [  117.140004] [drm] probing gen 2 caps for device 8086:9c18 = 5323c42/0
Oct  4 18:44:06 kerberos kernel: [  117.140010] [drm] PCIE gen 2 link speeds already enabled
Oct  4 18:44:06 kerberos kernel: [  117.142751] [drm] PCIE GART of 1024M enabled (table at 0x0000000000276000).
Oct  4 18:44:06 kerberos kernel: [  117.142935] radeon 0000:03:00.0: WB enabled
Oct  4 18:44:06 kerberos kernel: [  117.142938] radeon 0000:03:00.0: fence driver on ring 0 use gpu addr 0x0000000080000c00 and cpu addr 0xffff880035b4bc00
Oct  4 18:44:06 kerberos kernel: [  117.142940] radeon 0000:03:00.0: fence driver on ring 1 use gpu addr 0x0000000080000c04 and cpu addr 0xffff880035b4bc04
Oct  4 18:44:06 kerberos kernel: [  117.142942] radeon 0000:03:00.0: fence driver on ring 2 use gpu addr 0x0000000080000c08 and cpu addr 0xffff880035b4bc08
Oct  4 18:44:06 kerberos kernel: [  117.142943] radeon 0000:03:00.0: fence driver on ring 3 use gpu addr 0x0000000080000c0c and cpu addr 0xffff880035b4bc0c
Oct  4 18:44:06 kerberos kernel: [  117.142944] radeon 0000:03:00.0: fence driver on ring 4 use gpu addr 0x0000000080000c10 and cpu addr 0xffff880035b4bc10
Oct  4 18:44:06 kerberos kernel: [  117.144485] radeon 0000:03:00.0: fence driver on ring 5 use gpu addr 0x0000000000075a18 and cpu addr 0xffffc900160b5a18
Oct  4 18:44:07 kerberos kernel: [  117.745896] [drm] ring test on 0 succeeded in 1 usecs
Oct  4 18:44:07 kerberos kernel: [  117.745904] [drm] ring test on 1 succeeded in 1 usecs
Oct  4 18:44:07 kerberos kernel: [  117.745909] [drm] ring test on 2 succeeded in 1 usecs
Oct  4 18:44:07 kerberos kernel: [  117.745924] [drm] ring test on 3 succeeded in 3 usecs
Oct  4 18:44:07 kerberos kernel: [  117.745932] [drm] ring test on 4 succeeded in 1 usecs
Oct  4 18:44:07 kerberos kernel: [  117.931477] [drm] ring test on 5 succeeded in 2 usecs
Oct  4 18:44:07 kerberos kernel: [  117.931486] [drm] UVD initialized successfully.
Oct  4 18:44:07 kerberos kernel: [  117.931522] [drm] ib test on ring 0 succeeded in 0 usecs
Oct  4 18:44:07 kerberos kernel: [  117.931555] [drm] ib test on ring 1 succeeded in 0 usecs
Oct  4 18:44:07 kerberos kernel: [  117.931586] [drm] ib test on ring 2 succeeded in 0 usecs
Oct  4 18:44:07 kerberos kernel: [  117.931618] [drm] ib test on ring 3 succeeded in 0 usecs
Oct  4 18:44:07 kerberos kernel: [  117.931649] [drm] ib test on ring 4 succeeded in 0 usecs
Oct  4 18:44:07 kerberos kernel: [  118.082210] [drm] ib test on ring 5 succeeded
Oct  4 18:44:19 kerberos kernel: [  130.035395] [drm] probing gen 2 caps for device 8086:9c18 = 5323c42/0
Oct  4 18:44:19 kerberos kernel: [  130.035401] [drm] PCIE gen 2 link speeds already enabled
Oct  4 18:44:19 kerberos kernel: [  130.038144] [drm] PCIE GART of 1024M enabled (table at 0x0000000000276000).
Oct  4 18:44:19 kerberos kernel: [  130.038256] radeon 0000:03:00.0: WB enabled
Oct  4 18:44:19 kerberos kernel: [  130.038259] radeon 0000:03:00.0: fence driver on ring 0 use gpu addr 0x0000000080000c00 and cpu addr 0xffff880035b4bc00
Oct  4 18:44:19 kerberos kernel: [  130.038261] radeon 0000:03:00.0: fence driver on ring 1 use gpu addr 0x0000000080000c04 and cpu addr 0xffff880035b4bc04
Oct  4 18:44:19 kerberos kernel: [  130.038263] radeon 0000:03:00.0: fence driver on ring 2 use gpu addr 0x0000000080000c08 and cpu addr 0xffff880035b4bc08
Oct  4 18:44:19 kerberos kernel: [  130.038264] radeon 0000:03:00.0: fence driver on ring 3 use gpu addr 0x0000000080000c0c and cpu addr 0xffff880035b4bc0c
Oct  4 18:44:19 kerberos kernel: [  130.038266] radeon 0000:03:00.0: fence driver on ring 4 use gpu addr 0x0000000080000c10 and cpu addr 0xffff880035b4bc10
Oct  4 18:44:19 kerberos kernel: [  130.039837] radeon 0000:03:00.0: fence driver on ring 5 use gpu addr 0x0000000000075a18 and cpu addr 0xffffc900160b5a18
Oct  4 18:44:20 kerberos kernel: [  130.640603] [drm] ring test on 0 succeeded in 1 usecs
Oct  4 18:44:20 kerberos kernel: [  130.640610] [drm] ring test on 1 succeeded in 1 usecs
Oct  4 18:44:20 kerberos kernel: [  130.640615] [drm] ring test on 2 succeeded in 1 usecs
Oct  4 18:44:20 kerberos kernel: [  130.640629] [drm] ring test on 3 succeeded in 3 usecs
Oct  4 18:44:20 kerberos kernel: [  130.640637] [drm] ring test on 4 succeeded in 1 usecs
Oct  4 18:44:20 kerberos kernel: [  130.826174] [drm] ring test on 5 succeeded in 2 usecs
Oct  4 18:44:20 kerberos kernel: [  130.826183] [drm] UVD initialized successfully.
Oct  4 18:44:20 kerberos kernel: [  130.826218] [drm] ib test on ring 0 succeeded in 0 usecs
Oct  4 18:44:20 kerberos kernel: [  130.826250] [drm] ib test on ring 1 succeeded in 0 usecs
Oct  4 18:44:20 kerberos kernel: [  130.826281] [drm] ib test on ring 2 succeeded in 0 usecs
Oct  4 18:44:20 kerberos kernel: [  130.826312] [drm] ib test on ring 3 succeeded in 0 usecs
Oct  4 18:44:20 kerberos kernel: [  130.826344] [drm] ib test on ring 4 succeeded in 0 usecs
Oct  4 18:44:20 kerberos kernel: [  130.976944] [drm] ib test on ring 5 succeeded
Oct  4 18:44:29 kerberos kernel: [  140.043075] [drm] probing gen 2 caps for device 8086:9c18 = 5323c42/0
Oct  4 18:44:29 kerberos kernel: [  140.043080] [drm] PCIE gen 2 link speeds already enabled
Oct  4 18:44:29 kerberos kernel: [  140.045826] [drm] PCIE GART of 1024M enabled (table at 0x0000000000276000).
Oct  4 18:44:29 kerberos kernel: [  140.045952] radeon 0000:03:00.0: WB enabled
Oct  4 18:44:29 kerberos kernel: [  140.045955] radeon 0000:03:00.0: fence driver on ring 0 use gpu addr 0x0000000080000c00 and cpu addr 0xffff880035b4bc00
Oct  4 18:44:29 kerberos kernel: [  140.045958] radeon 0000:03:00.0: fence driver on ring 1 use gpu addr 0x0000000080000c04 and cpu addr 0xffff880035b4bc04
Oct  4 18:44:29 kerberos kernel: [  140.045960] radeon 0000:03:00.0: fence driver on ring 2 use gpu addr 0x0000000080000c08 and cpu addr 0xffff880035b4bc08
Oct  4 18:44:29 kerberos kernel: [  140.045962] radeon 0000:03:00.0: fence driver on ring 3 use gpu addr 0x0000000080000c0c and cpu addr 0xffff880035b4bc0c
Oct  4 18:44:29 kerberos kernel: [  140.045964] radeon 0000:03:00.0: fence driver on ring 4 use gpu addr 0x0000000080000c10 and cpu addr 0xffff880035b4bc10
Oct  4 18:44:29 kerberos kernel: [  140.047513] radeon 0000:03:00.0: fence driver on ring 5 use gpu addr 0x0000000000075a18 and cpu addr 0xffffc900160b5a18
Oct  4 18:44:30 kerberos kernel: [  140.649833] [drm] ring test on 0 succeeded in 1 usecs
Oct  4 18:44:30 kerberos kernel: [  140.649841] [drm] ring test on 1 succeeded in 1 usecs
Oct  4 18:44:30 kerberos kernel: [  140.649846] [drm] ring test on 2 succeeded in 1 usecs
Oct  4 18:44:30 kerberos kernel: [  140.649861] [drm] ring test on 3 succeeded in 3 usecs
Oct  4 18:44:30 kerberos kernel: [  140.649869] [drm] ring test on 4 succeeded in 1 usecs
Oct  4 18:44:30 kerberos kernel: [  140.835485] [drm] ring test on 5 succeeded in 2 usecs
Oct  4 18:44:30 kerberos kernel: [  140.835494] [drm] UVD initialized successfully.
Oct  4 18:44:30 kerberos kernel: [  140.835532] [drm] ib test on ring 0 succeeded in 0 usecs
Oct  4 18:44:30 kerberos kernel: [  140.835574] [drm] ib test on ring 1 succeeded in 0 usecs
Oct  4 18:44:30 kerberos kernel: [  140.835605] [drm] ib test on ring 2 succeeded in 0 usecs
Oct  4 18:44:30 kerberos kernel: [  140.835637] [drm] ib test on ring 3 succeeded in 0 usecs
Oct  4 18:44:30 kerberos kernel: [  140.835669] [drm] ib test on ring 4 succeeded in 0 usecs
Oct  4 18:44:30 kerberos kernel: [  140.986291] [drm] ib test on ring 5 succeeded
Oct  4 18:45:18 kerberos kernel: [  188.774142] radeon 0000:03:00.0: GPU fault detected: 146 0x0a020804
Oct  4 18:45:18 kerberos kernel: [  188.774147] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x0000F1D0
Oct  4 18:45:18 kerberos kernel: [  188.774149] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008004

xorg doesnt show anything beyond the ordinary, same for syslog
Comment 6 Michel Dänzer 2014-10-07 01:45:23 UTC
(In reply to Paulo Dias from comment #5)
> Well with the latest update from oibaf ppa, im having the same problem has
> Flip Andjelo, as soon as borderlands enters the main menu, the gpu crashes

Sounds like there was another regression in Mesa then, can one of you bisect that?
Comment 7 Paulo Dias 2014-10-07 15:13:12 UTC
im gonna try this latest patch before trying a bisect, http://cgit.freedesktop.org/mesa/mesa/commit/?id=16b53005a7df4249fecb6641af0934c32181fdea

if i were to guess i would say the offending patch might be http://cgit.freedesktop.org/mesa/mesa/commit/?id=ed03747e6a1a0655bfd0b5e4a52455fbd37f97c5 or something after this one. anyway more new later after i tested the above commit
Comment 8 Paulo Dias 2014-10-08 00:46:29 UTC
Created attachment 107529 [details]
xorg after crash
Comment 9 Paulo Dias 2014-10-08 00:47:22 UTC
still crashing:

Oct  7 21:35:01 kerberos kernel: [  750.357204] radeon 0000:03:00.0: WB enabled
Oct  7 21:35:01 kerberos kernel: [  750.357206] radeon 0000:03:00.0: fence driver on ring 0 use gpu addr 0x0000000080000c00 and cpu addr 0xffff8800354e2c00
Oct  7 21:35:01 kerberos kernel: [  750.357208] radeon 0000:03:00.0: fence driver on ring 1 use gpu addr 0x0000000080000c04 and cpu addr 0xffff8800354e2c04
Oct  7 21:35:01 kerberos kernel: [  750.357210] radeon 0000:03:00.0: fence driver on ring 2 use gpu addr 0x0000000080000c08 and cpu addr 0xffff8800354e2c08
Oct  7 21:35:01 kerberos kernel: [  750.357212] radeon 0000:03:00.0: fence driver on ring 3 use gpu addr 0x0000000080000c0c and cpu addr 0xffff8800354e2c0c
Oct  7 21:35:01 kerberos kernel: [  750.357213] radeon 0000:03:00.0: fence driver on ring 4 use gpu addr 0x0000000080000c10 and cpu addr 0xffff8800354e2c10
Oct  7 21:35:01 kerberos kernel: [  750.358756] radeon 0000:03:00.0: fence driver on ring 5 use gpu addr 0x0000000000075a18 and cpu addr 0xffffc90005935a18
Oct  7 21:35:02 kerberos kernel: [  750.957112] [drm] ring test on 0 succeeded in 1 usecs
Oct  7 21:35:02 kerberos kernel: [  750.957119] [drm] ring test on 1 succeeded in 1 usecs
Oct  7 21:35:02 kerberos kernel: [  750.957125] [drm] ring test on 2 succeeded in 1 usecs
Oct  7 21:35:02 kerberos kernel: [  750.957140] [drm] ring test on 3 succeeded in 3 usecs
Oct  7 21:35:02 kerberos kernel: [  750.957148] [drm] ring test on 4 succeeded in 1 usecs
Oct  7 21:35:02 kerberos kernel: [  751.142691] [drm] ring test on 5 succeeded in 2 usecs
Oct  7 21:35:02 kerberos kernel: [  751.142700] [drm] UVD initialized successfully.
Oct  7 21:35:02 kerberos kernel: [  751.142738] [drm] ib test on ring 0 succeeded in 0 usecs
Oct  7 21:35:02 kerberos kernel: [  751.142771] [drm] ib test on ring 1 succeeded in 0 usecs
Oct  7 21:35:02 kerberos kernel: [  751.142803] [drm] ib test on ring 2 succeeded in 0 usecs
Oct  7 21:35:02 kerberos kernel: [  751.142836] [drm] ib test on ring 3 succeeded in 0 usecs
Oct  7 21:35:02 kerberos kernel: [  751.142868] [drm] ib test on ring 4 succeeded in 0 usecs
Oct  7 21:35:02 kerberos kernel: [  751.293429] [drm] ib test on ring 5 succeeded
Oct  7 21:35:15 kerberos kernel: [  763.948153] [drm] probing gen 2 caps for device 8086:9c18 = 5323c42/0
Oct  7 21:35:15 kerberos kernel: [  763.948158] [drm] PCIE gen 2 link speeds already enabled
Oct  7 21:35:15 kerberos kernel: [  763.950935] [drm] PCIE GART of 1024M enabled (table at 0x0000000000276000).
Oct  7 21:35:15 kerberos kernel: [  763.951050] radeon 0000:03:00.0: WB enabled
Oct  7 21:35:15 kerberos kernel: [  763.951055] radeon 0000:03:00.0: fence driver on ring 0 use gpu addr 0x0000000080000c00 and cpu addr 0xffff8800354e2c00
Oct  7 21:35:15 kerberos kernel: [  763.951059] radeon 0000:03:00.0: fence driver on ring 1 use gpu addr 0x0000000080000c04 and cpu addr 0xffff8800354e2c04
Oct  7 21:35:15 kerberos kernel: [  763.951062] radeon 0000:03:00.0: fence driver on ring 2 use gpu addr 0x0000000080000c08 and cpu addr 0xffff8800354e2c08
Oct  7 21:35:15 kerberos kernel: [  763.951065] radeon 0000:03:00.0: fence driver on ring 3 use gpu addr 0x0000000080000c0c and cpu addr 0xffff8800354e2c0c
Oct  7 21:35:15 kerberos kernel: [  763.951068] radeon 0000:03:00.0: fence driver on ring 4 use gpu addr 0x0000000080000c10 and cpu addr 0xffff8800354e2c10
Oct  7 21:35:15 kerberos kernel: [  763.952623] radeon 0000:03:00.0: fence driver on ring 5 use gpu addr 0x0000000000075a18 and cpu addr 0xffffc90005935a18
Oct  7 21:35:15 kerberos kernel: [  764.556516] [drm] ring test on 0 succeeded in 1 usecs
Oct  7 21:35:15 kerberos kernel: [  764.556524] [drm] ring test on 1 succeeded in 1 usecs
Oct  7 21:35:15 kerberos kernel: [  764.556529] [drm] ring test on 2 succeeded in 1 usecs
Oct  7 21:35:15 kerberos kernel: [  764.556544] [drm] ring test on 3 succeeded in 3 usecs
Oct  7 21:35:15 kerberos kernel: [  764.556552] [drm] ring test on 4 succeeded in 1 usecs
Oct  7 21:35:16 kerberos kernel: [  764.742082] [drm] ring test on 5 succeeded in 2 usecs
Oct  7 21:35:16 kerberos kernel: [  764.742090] [drm] UVD initialized successfully.
Oct  7 21:35:16 kerberos kernel: [  764.742128] [drm] ib test on ring 0 succeeded in 0 usecs
Oct  7 21:35:16 kerberos kernel: [  764.742160] [drm] ib test on ring 1 succeeded in 0 usecs
Oct  7 21:35:16 kerberos kernel: [  764.742192] [drm] ib test on ring 2 succeeded in 0 usecs
Oct  7 21:35:16 kerberos kernel: [  764.742225] [drm] ib test on ring 3 succeeded in 0 usecs
Oct  7 21:35:16 kerberos kernel: [  764.742257] [drm] ib test on ring 4 succeeded in 0 usecs
Oct  7 21:35:16 kerberos kernel: [  764.892884] [drm] ib test on ring 5 succeeded
Oct  7 21:35:27 kerberos kernel: [  775.808225] [drm] probing gen 2 caps for device 8086:9c18 = 5323c42/0
Oct  7 21:35:27 kerberos kernel: [  775.808232] [drm] PCIE gen 2 link speeds already enabled
Oct  7 21:35:27 kerberos kernel: [  775.811032] [drm] PCIE GART of 1024M enabled (table at 0x0000000000276000).
Oct  7 21:35:27 kerberos kernel: [  775.811147] radeon 0000:03:00.0: WB enabled
Oct  7 21:35:27 kerberos kernel: [  775.811151] radeon 0000:03:00.0: fence driver on ring 0 use gpu addr 0x0000000080000c00 and cpu addr 0xffff8800354e2c00
Oct  7 21:35:27 kerberos kernel: [  775.811153] radeon 0000:03:00.0: fence driver on ring 1 use gpu addr 0x0000000080000c04 and cpu addr 0xffff8800354e2c04
Oct  7 21:35:27 kerberos kernel: [  775.811156] radeon 0000:03:00.0: fence driver on ring 2 use gpu addr 0x0000000080000c08 and cpu addr 0xffff8800354e2c08
Oct  7 21:35:27 kerberos kernel: [  775.811158] radeon 0000:03:00.0: fence driver on ring 3 use gpu addr 0x0000000080000c0c and cpu addr 0xffff8800354e2c0c
Oct  7 21:35:27 kerberos kernel: [  775.811160] radeon 0000:03:00.0: fence driver on ring 4 use gpu addr 0x0000000080000c10 and cpu addr 0xffff8800354e2c10
Oct  7 21:35:27 kerberos kernel: [  775.812713] radeon 0000:03:00.0: fence driver on ring 5 use gpu addr 0x0000000000075a18 and cpu addr 0xffffc90005935a18
Oct  7 21:35:27 kerberos kernel: [  776.416088] [drm] ring test on 0 succeeded in 1 usecs
Oct  7 21:35:27 kerberos kernel: [  776.416097] [drm] ring test on 1 succeeded in 1 usecs
Oct  7 21:35:27 kerberos kernel: [  776.416103] [drm] ring test on 2 succeeded in 1 usecs
Oct  7 21:35:27 kerberos kernel: [  776.416119] [drm] ring test on 3 succeeded in 3 usecs
Oct  7 21:35:27 kerberos kernel: [  776.416128] [drm] ring test on 4 succeeded in 1 usecs
Oct  7 21:35:27 kerberos kernel: [  776.601682] [drm] ring test on 5 succeeded in 2 usecs
Oct  7 21:35:27 kerberos kernel: [  776.601691] [drm] UVD initialized successfully.
Oct  7 21:35:27 kerberos kernel: [  776.601727] [drm] ib test on ring 0 succeeded in 0 usecs
Oct  7 21:35:27 kerberos kernel: [  776.601760] [drm] ib test on ring 1 succeeded in 0 usecs
Oct  7 21:35:27 kerberos kernel: [  776.601791] [drm] ib test on ring 2 succeeded in 0 usecs
Oct  7 21:35:27 kerberos kernel: [  776.601823] [drm] ib test on ring 3 succeeded in 0 usecs
Oct  7 21:35:27 kerberos kernel: [  776.601854] [drm] ib test on ring 4 succeeded in 0 usecs
Oct  7 21:35:28 kerberos kernel: [  776.752454] [drm] ib test on ring 5 succeeded
Oct  7 21:36:13 kerberos kernel: [  821.976487] radeon 0000:03:00.0: GPU fault detected: 147 0x0aa20801
Oct  7 21:36:13 kerberos kernel: [  821.976493] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x01600855
Oct  7 21:36:13 kerberos kernel: [  821.976494] radeon 0000:03:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x02008001
Oct  7 21:36:13 kerberos kernel: [  821.976497] VM fault (0x01, vmid 1) at page 23070805, read from TC (8)
Comment 10 Paulo Dias 2014-10-08 01:04:37 UTC
i think i should mention that these crashes only happen if i enable

DynamicLights=True in .local/share/aspyr-media/borderlands\ 2/willowgame/config/willowengine.ini

turning it to false, the game runs, but i get stalls for more then 4 seconds randomly, otherwise is very fast,even with everything maxed out, and of course, without dynamic lights, the game looks like >.<

the intel driver works fine with dynlights on, without a glitch, only slowerofc
Comment 11 Eduardo 2014-10-10 05:05:06 UTC
I'm not sure if it is the same error, but I'm having a very similar issue with the game Brutal Legends, even in the lowest quality.
I created an apitrace that reproduces the error every time.
Comment 12 Eduardo 2014-10-10 05:19:28 UTC
The APITrace
https://drive.google.com/folderview?id=0B8TMGw8Zhm5PclBsOUZVa3JKQjA&usp=drive_web
(it's quite big, even after gzipping, and still uploading)

Specs:
Arch Linux 3.16.4-1-ARCH x86_64
Mesa 10.3.0
AMD Radeon HD 7730M
Comment 13 Michel Dänzer 2014-10-10 08:26:24 UTC
(In reply to Eduardo from comment #11)
> I'm not sure if it is the same error, but I'm having a very similar issue
> with the game Brutal Legends, even in the lowest quality.
> I created an apitrace that reproduces the error every time.

Looks like that might be bug 83500. If so, the environment variable R600_DEBUG=nodma might work around it.
Comment 14 José Suárez 2014-10-10 10:17:53 UTC
I am running BL2 with radeonsi git as of yesterday from oibaf on my Radeon HD 7870 but recompiled with llvm 3.6. with no problems (other than the stutters which was fixed yesterday in mesa git). Could you guys maybe try recompiling mesa with llvm 3.6?
Comment 15 Paulo Dias 2014-10-10 16:30:32 UTC
(In reply to José Suárez from comment #14)
> I am running BL2 with radeonsi git as of yesterday from oibaf on my Radeon
> HD 7870 but recompiled with llvm 3.6. with no problems (other than the
> stutters which was fixed yesterday in mesa git). Could you guys maybe try
> recompiling mesa with llvm 3.6?

Before i test with llvm 3.6, did you enabled DynamicLights=True in .local/share/aspyr-media/borderlands\ 2/willowgame/config/willowengine.ini ?

because that setting is the one crashing the game for me, without it, it works just fine (with the stutters, didnt test the latest patch yet)
Comment 16 Eduardo 2014-10-10 19:00:02 UTC
(In reply to Michel Dänzer from comment #13)
> (In reply to Eduardo from comment #11)
> > I'm not sure if it is the same error, but I'm having a very similar issue
> > with the game Brutal Legends, even in the lowest quality.
> > I created an apitrace that reproduces the error every time.
> 
> Looks like that might be bug 83500. If so, the environment variable
> R600_DEBUG=nodma might work around it.

I'm afraid not. I've tested with R600_DEBUG=nodma and the same crash occurs, both while replaying the apitrace and while playing the game.
I've uploaded new apitraces with R600_DEBUG=nodma in https://drive.google.com/folderview?id=0B8TMGw8Zhm5PclBsOUZVa3JKQjA&usp=drive_web
I should have mentioned earlier, but I'm using PRIME (DRI_PRIME=1).
I will try with llvm 3.6 and the lastest git latter.
Comment 17 José Suárez 2014-10-11 20:36:08 UTC
(In reply to Paulo Dias from comment #15)
> (In reply to José Suárez from comment #14)
> > I am running BL2 with radeonsi git as of yesterday from oibaf on my Radeon
> > HD 7870 but recompiled with llvm 3.6. with no problems (other than the
> > stutters which was fixed yesterday in mesa git). Could you guys maybe try
> > recompiling mesa with llvm 3.6?
> 
> Before i test with llvm 3.6, did you enabled DynamicLights=True in
> .local/share/aspyr-media/borderlands\ 2/willowgame/config/willowengine.ini ?
> 
> because that setting is the one crashing the game for me, without it, it
> works just fine (with the stutters, didnt test the latest patch yet)

Yes, I have tried llvm-3.6 with DynamicLights on. For approximately 15 minutes without any crash. I was experiencing heavy stutter (heavier than without DynamicLight), though. Thus I turned it off again. Since the stutter problem has been fixed (or at least mitigated), I'll give it a try again.
Comment 18 Michel Dänzer 2014-10-14 03:07:28 UTC
(In reply to Eduardo from comment #16)
> I'm afraid not. I've tested with R600_DEBUG=nodma and the same crash occurs,
> both while replaying the apitrace and while playing the game.

Unless your problem is fixed with an LLVM 3.6 snapshot as well, please file your own bug report about Brütal Legend then.
Comment 19 Marek Olšák 2015-08-24 22:17:04 UTC
Is this issue still present with current Mesa git?
Comment 20 Eduardo 2015-08-26 14:08:25 UTC
For me, the issue is solved. But I'm not the original reporter.
Comment 21 Marek Olšák 2015-08-28 15:26:07 UTC
(In reply to Eduardo from comment #20)
> For me, the issue is solved. But I'm not the original reporter.

OK, thanks. The original reporter can re-open this if needed. Closing.


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.