Bug 81576

Summary:	GPU lock-ups while playing XCOM Enemy Unknown
Product:	Mesa	Reporter:	Frederik Himpe <fhimpe>
Component:	Drivers/Gallium/radeonsi	Assignee:	Default DRI bug account <dri-devel>
Status:	RESOLVED DUPLICATE	QA Contact:
Severity:	normal
Priority:	medium	CC:	ansla80
Version:	unspecified
Hardware:	Other
OS:	All
See Also:	https://bugs.freedesktop.org/show_bug.cgi?id=85334
Whiteboard:
i915 platform:		i915 features:
Attachments:	complete kernel log lspci -vvv

Description Frederik Himpe 2014-07-20 21:59:46 UTC

While playing XCOM Enemy Unknown on my Debian Wheezy system with Radeon HD 7850 card and the free drivers, the game often hangs for several seconds, and a GPU lockup message is printed in dmesg. This happens several times, untile it happens that the system does not become responsive again, and I have to hard reset the machine. The problem happens with both Linux 3.14.12 and 3.16-rc5.

Here is an extract of the kernel log:

Jul 19 23:55:17 piranha kernel: [ 7753.573031] radeon 0000:01:00.0: GPU lockup CP stall for more than 10360msec
Jul 19 23:55:17 piranha kernel: [ 7753.573043] radeon 0000:01:00.0: GPU lockup (waiting for 0x000000000007c746 last fence id 0x000000000007c745 on ring 3)
Jul 19 23:55:17 piranha kernel: [ 7753.573051] radeon 0000:01:00.0: failed to get a new IB (-35)
Jul 19 23:55:17 piranha kernel: [ 7753.717120] radeon 0000:01:00.0: GPU lockup CP stall for more than 10000msec
Jul 19 23:55:17 piranha kernel: [ 7753.717130] radeon 0000:01:00.0: GPU lockup (waiting for 0x00000000001e19a6 last fence id 0x00000000001e19a1 on ring 0)
Jul 19 23:55:17 piranha kernel: [ 7753.717210] radeon 0000:01:00.0: failed to get a new IB (-35)
Jul 19 23:55:17 piranha kernel: [ 7753.718031] radeon 0000:01:00.0: sa_manager is not empty, clearing anyway
Jul 19 23:55:18 piranha kernel: [ 7754.184904] radeon 0000:01:00.0: Saved 1024 dwords of commands on ring 0.
Jul 19 23:55:18 piranha kernel: [ 7754.185033] radeon 0000:01:00.0: GPU softreset: 0x0000006D
Jul 19 23:55:18 piranha kernel: [ 7754.185035] radeon 0000:01:00.0:   GRBM_STATUS               = 0xE7D20028
Jul 19 23:55:18 piranha kernel: [ 7754.185036] radeon 0000:01:00.0:   GRBM_STATUS_SE0           = 0xEDC00000
Jul 19 23:55:18 piranha kernel: [ 7754.185038] radeon 0000:01:00.0:   GRBM_STATUS_SE1           = 0xEDC00000
Jul 19 23:55:18 piranha kernel: [ 7754.185040] radeon 0000:01:00.0:   SRBM_STATUS               = 0x20040AC0
Jul 19 23:55:18 piranha kernel: [ 7754.185157] radeon 0000:01:00.0:   SRBM_STATUS2              = 0x00000000
Jul 19 23:55:18 piranha kernel: [ 7754.185159] radeon 0000:01:00.0:   R_008674_CP_STALLED_STAT1 = 0x00000000
Jul 19 23:55:18 piranha kernel: [ 7754.185160] radeon 0000:01:00.0:   R_008678_CP_STALLED_STAT2 = 0x00010800
Jul 19 23:55:18 piranha kernel: [ 7754.185162] radeon 0000:01:00.0:   R_00867C_CP_BUSY_STAT     = 0x00008006
Jul 19 23:55:18 piranha kernel: [ 7754.185164] radeon 0000:01:00.0:   R_008680_CP_STAT          = 0x80038647
Jul 19 23:55:18 piranha kernel: [ 7754.185165] radeon 0000:01:00.0:   R_00D034_DMA_STATUS_REG   = 0x44483146
Jul 19 23:55:18 piranha kernel: [ 7754.185167] radeon 0000:01:00.0:   R_00D834_DMA_STATUS_REG   = 0x44C83D57
Jul 19 23:55:18 piranha kernel: [ 7754.185169] radeon 0000:01:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x00000000
Jul 19 23:55:18 piranha kernel: [ 7754.185171] radeon 0000:01:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x00000000
Jul 19 23:55:18 piranha kernel: [ 7754.649830] radeon 0000:01:00.0: GRBM_SOFT_RESET=0x0000DDFF
Jul 19 23:55:18 piranha kernel: [ 7754.649884] radeon 0000:01:00.0: SRBM_SOFT_RESET=0x00100140
Jul 19 23:55:18 piranha kernel: [ 7754.651031] radeon 0000:01:00.0:   GRBM_STATUS               = 0x00003028
Jul 19 23:55:18 piranha kernel: [ 7754.651033] radeon 0000:01:00.0:   GRBM_STATUS_SE0           = 0x00000006
Jul 19 23:55:18 piranha kernel: [ 7754.651035] radeon 0000:01:00.0:   GRBM_STATUS_SE1           = 0x00000006
Jul 19 23:55:18 piranha kernel: [ 7754.651036] radeon 0000:01:00.0:   SRBM_STATUS               = 0x200400C0
Jul 19 23:55:18 piranha kernel: [ 7754.651147] radeon 0000:01:00.0:   SRBM_STATUS2              = 0x00000000
Jul 19 23:55:18 piranha kernel: [ 7754.651149] radeon 0000:01:00.0:   R_008674_CP_STALLED_STAT1 = 0x00000000
Jul 19 23:55:18 piranha kernel: [ 7754.651151] radeon 0000:01:00.0:   R_008678_CP_STALLED_STAT2 = 0x00000000
Jul 19 23:55:18 piranha kernel: [ 7754.651152] radeon 0000:01:00.0:   R_00867C_CP_BUSY_STAT     = 0x00000000
Jul 19 23:55:18 piranha kernel: [ 7754.651154] radeon 0000:01:00.0:   R_008680_CP_STAT          = 0x00000000
Jul 19 23:55:18 piranha kernel: [ 7754.651156] radeon 0000:01:00.0:   R_00D034_DMA_STATUS_REG   = 0x44C83D57
Jul 19 23:55:18 piranha kernel: [ 7754.651158] radeon 0000:01:00.0:   R_00D834_DMA_STATUS_REG   = 0x44C83D57
Jul 19 23:55:18 piranha kernel: [ 7754.651403] radeon 0000:01:00.0: GPU reset succeeded, trying to resume
Jul 19 23:55:18 piranha kernel: [ 7754.677135] [drm] probing gen 2 caps for device 8086:151 = 261ad03/e
Jul 19 23:55:18 piranha kernel: [ 7754.677143] [drm] PCIE gen 3 link speeds already enabled
Jul 19 23:55:18 piranha kernel: [ 7754.679075] [drm] PCIE GART of 1024M enabled (table at 0x0000000000276000).
Jul 19 23:55:18 piranha kernel: [ 7754.679178] radeon 0000:01:00.0: WB enabled
Jul 19 23:55:18 piranha kernel: [ 7754.679181] radeon 0000:01:00.0: fence driver on ring 0 use gpu addr 0x0000000080000c00 and cpu addr 0xffff88041d0dfc00
Jul 19 23:55:18 piranha kernel: [ 7754.679183] radeon 0000:01:00.0: fence driver on ring 1 use gpu addr 0x0000000080000c04 and cpu addr 0xffff88041d0dfc04
Jul 19 23:55:18 piranha kernel: [ 7754.679185] radeon 0000:01:00.0: fence driver on ring 2 use gpu addr 0x0000000080000c08 and cpu addr 0xffff88041d0dfc08
Jul 19 23:55:18 piranha kernel: [ 7754.679187] radeon 0000:01:00.0: fence driver on ring 3 use gpu addr 0x0000000080000c0c and cpu addr 0xffff88041d0dfc0c
Jul 19 23:55:18 piranha kernel: [ 7754.679188] radeon 0000:01:00.0: fence driver on ring 4 use gpu addr 0x0000000080000c10 and cpu addr 0xffff88041d0dfc10
Jul 19 23:55:18 piranha kernel: [ 7754.679801] radeon 0000:01:00.0: fence driver on ring 5 use gpu addr 0x0000000000075a18 and cpu addr 0xffffc90005db5a18
Jul 19 23:55:18 piranha kernel: [ 7754.835517] [drm] ring test on 0 succeeded in 4 usecs
Jul 19 23:55:18 piranha kernel: [ 7754.835522] [drm] ring test on 1 succeeded in 1 usecs
Jul 19 23:55:18 piranha kernel: [ 7754.835526] [drm] ring test on 2 succeeded in 1 usecs
Jul 19 23:55:18 piranha kernel: [ 7754.835588] [drm] ring test on 3 succeeded in 2 usecs
Jul 19 23:55:18 piranha kernel: [ 7754.835598] [drm] ring test on 4 succeeded in 1 usecs
Jul 19 23:55:19 piranha kernel: [ 7755.011345] [drm] ring test on 5 succeeded in 2 usecs
Jul 19 23:55:19 piranha kernel: [ 7755.011352] [drm] UVD initialized successfully.
Jul 19 23:55:19 piranha kernel: [ 7755.013070] [drm:r600_ib_test] *ERROR* radeon: fence wait failed (-35).
Jul 19 23:55:19 piranha kernel: [ 7755.013072] [drm:radeon_ib_ring_tests] *ERROR* radeon: failed testing IB on GFX ring (-35).
Jul 19 23:55:19 piranha kernel: [ 7755.013074] radeon 0000:01:00.0: ib ring test failed (-35).
Jul 19 23:55:19 piranha kernel: [ 7755.476969] radeon 0000:01:00.0: GPU softreset: 0x00000048
Jul 19 23:55:19 piranha kernel: [ 7755.476976] radeon 0000:01:00.0:   GRBM_STATUS               = 0xA0000028
Jul 19 23:55:19 piranha kernel: [ 7755.476978] radeon 0000:01:00.0:   GRBM_STATUS_SE0           = 0x00000000
Jul 19 23:55:19 piranha kernel: [ 7755.476980] radeon 0000:01:00.0:   GRBM_STATUS_SE1           = 0x00000000
Jul 19 23:55:19 piranha kernel: [ 7755.476981] radeon 0000:01:00.0:   SRBM_STATUS               = 0x200400C0
Jul 19 23:55:19 piranha kernel: [ 7755.477092] radeon 0000:01:00.0:   SRBM_STATUS2              = 0x00000000
Jul 19 23:55:19 piranha kernel: [ 7755.477094] radeon 0000:01:00.0:   R_008674_CP_STALLED_STAT1 = 0x00000000
Jul 19 23:55:19 piranha kernel: [ 7755.477096] radeon 0000:01:00.0:   R_008678_CP_STALLED_STAT2 = 0x00000000
Jul 19 23:55:19 piranha kernel: [ 7755.477097] radeon 0000:01:00.0:   R_00867C_CP_BUSY_STAT     = 0x00000006
Jul 19 23:55:19 piranha kernel: [ 7755.477099] radeon 0000:01:00.0:   R_008680_CP_STAT          = 0x80028647
Jul 19 23:55:19 piranha kernel: [ 7755.477101] radeon 0000:01:00.0:   R_00D034_DMA_STATUS_REG   = 0x44C83D57
Jul 19 23:55:19 piranha kernel: [ 7755.477103] radeon 0000:01:00.0:   R_00D834_DMA_STATUS_REG   = 0x44C83D57
Jul 19 23:55:19 piranha kernel: [ 7755.477105] radeon 0000:01:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x00000000
Jul 19 23:55:19 piranha kernel: [ 7755.477107] radeon 0000:01:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x00000000
Jul 19 23:55:20 piranha kernel: [ 7755.932045] radeon 0000:01:00.0: GRBM_SOFT_RESET=0x0000DDFF
Jul 19 23:55:20 piranha kernel: [ 7755.932100] radeon 0000:01:00.0: SRBM_SOFT_RESET=0x00000100
Jul 19 23:55:20 piranha kernel: [ 7755.933247] radeon 0000:01:00.0:   GRBM_STATUS               = 0x00003028
Jul 19 23:55:20 piranha kernel: [ 7755.933248] radeon 0000:01:00.0:   GRBM_STATUS_SE0           = 0x00000006
Jul 19 23:55:20 piranha kernel: [ 7755.933250] radeon 0000:01:00.0:   GRBM_STATUS_SE1           = 0x00000006
Jul 19 23:55:20 piranha kernel: [ 7755.933251] radeon 0000:01:00.0:   SRBM_STATUS               = 0x200400C0
Jul 19 23:55:20 piranha kernel: [ 7755.933362] radeon 0000:01:00.0:   SRBM_STATUS2              = 0x00000000
Jul 19 23:55:20 piranha kernel: [ 7755.933364] radeon 0000:01:00.0:   R_008674_CP_STALLED_STAT1 = 0x00000000
Jul 19 23:55:20 piranha kernel: [ 7755.933366] radeon 0000:01:00.0:   R_008678_CP_STALLED_STAT2 = 0x00000000
Jul 19 23:55:20 piranha kernel: [ 7755.933367] radeon 0000:01:00.0:   R_00867C_CP_BUSY_STAT     = 0x00000000
Jul 19 23:55:20 piranha kernel: [ 7755.933369] radeon 0000:01:00.0:   R_008680_CP_STAT          = 0x00000000
Jul 19 23:55:20 piranha kernel: [ 7755.933371] radeon 0000:01:00.0:   R_00D034_DMA_STATUS_REG   = 0x44C83D57
Jul 19 23:55:20 piranha kernel: [ 7755.933373] radeon 0000:01:00.0:   R_00D834_DMA_STATUS_REG   = 0x44C83D57
Jul 19 23:55:20 piranha kernel: [ 7755.933617] radeon 0000:01:00.0: GPU reset succeeded, trying to resume
Jul 19 23:55:20 piranha kernel: [ 7755.944532] [drm] probing gen 2 caps for device 8086:151 = 261ad03/e
Jul 19 23:55:20 piranha kernel: [ 7755.944538] [drm] PCIE gen 3 link speeds already enabled
Jul 19 23:55:20 piranha kernel: [ 7755.946534] [drm] PCIE GART of 1024M enabled (table at 0x0000000000276000).
Jul 19 23:55:20 piranha kernel: [ 7755.946630] radeon 0000:01:00.0: WB enabled
Jul 19 23:55:20 piranha kernel: [ 7755.946633] radeon 0000:01:00.0: fence driver on ring 0 use gpu addr 0x0000000080000c00 and cpu addr 0xffff88041d0dfc00
Jul 19 23:55:20 piranha kernel: [ 7755.946635] radeon 0000:01:00.0: fence driver on ring 1 use gpu addr 0x0000000080000c04 and cpu addr 0xffff88041d0dfc04
Jul 19 23:55:20 piranha kernel: [ 7755.946637] radeon 0000:01:00.0: fence driver on ring 2 use gpu addr 0x0000000080000c08 and cpu addr 0xffff88041d0dfc08
Jul 19 23:55:20 piranha kernel: [ 7755.946638] radeon 0000:01:00.0: fence driver on ring 3 use gpu addr 0x0000000080000c0c and cpu addr 0xffff88041d0dfc0c
Jul 19 23:55:20 piranha kernel: [ 7755.946640] radeon 0000:01:00.0: fence driver on ring 4 use gpu addr 0x0000000080000c10 and cpu addr 0xffff88041d0dfc10
Jul 19 23:55:20 piranha kernel: [ 7755.947252] radeon 0000:01:00.0: fence driver on ring 5 use gpu addr 0x0000000000075a18 and cpu addr 0xffffc90005db5a18
Jul 19 23:55:20 piranha kernel: [ 7756.102387] [drm] ring test on 0 succeeded in 4 usecs
Jul 19 23:55:20 piranha kernel: [ 7756.102395] [drm] ring test on 1 succeeded in 1 usecs
Jul 19 23:55:20 piranha kernel: [ 7756.102398] [drm] ring test on 2 succeeded in 1 usecs
Jul 19 23:55:20 piranha kernel: [ 7756.102461] [drm] ring test on 3 succeeded in 2 usecs
Jul 19 23:55:20 piranha kernel: [ 7756.102497] [drm] ring test on 4 succeeded in 2 usecs
Jul 19 23:55:20 piranha kernel: [ 7756.278275] [drm] ring test on 5 succeeded in 2 usecs
Jul 19 23:55:20 piranha kernel: [ 7756.278281] [drm] UVD initialized successfully.
Jul 19 23:55:20 piranha kernel: [ 7756.280029] [drm] ib test on ring 0 succeeded in 0 usecs
Jul 19 23:55:20 piranha kernel: [ 7756.280076] [drm] ib test on ring 1 succeeded in 0 usecs
Jul 19 23:55:20 piranha kernel: [ 7756.280119] [drm] ib test on ring 2 succeeded in 0 usecs
Jul 19 23:55:20 piranha kernel: [ 7756.280155] [drm] ib test on ring 3 succeeded in 0 usecs
Jul 19 23:55:20 piranha kernel: [ 7756.280189] [drm] ib test on ring 4 succeeded in 0 usecs
Jul 19 23:55:20 piranha kernel: [ 7756.430256] [drm:uvd_v1_0_ib_test] *ERROR* radeon: failed to get create msg (-22).
Jul 19 23:55:20 piranha kernel: [ 7756.430264] [drm:radeon_ib_ring_tests] *ERROR* radeon: failed testing IB on ring 5 (-22).
Jul 19 23:55:20 piranha kernel: [ 7756.430284] [drm:radeon_pm_resume_dpm] *ERROR* radeon: dpm resume failed

I'm using xserver 1.16.0 and mesa 10.2.3.

Comment 1 Frederik Himpe 2014-07-20 22:00:17 UTC

Created attachment 103156 [details]
complete kernel log

Comment 2 Frederik Himpe 2014-07-20 22:02:18 UTC

Created attachment 103158 [details]
lspci -vvv

Comment 3 Andrei Slavoiu 2014-08-10 13:55:46 UTC

I get the same on a Kaveri (A10-7850K)

Comment 4 Alex Deucher 2014-10-26 17:49:45 UTC


*** This bug has been marked as a duplicate of bug 80419 ***

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.