[Bug 85204] [Radeon HD 5650] return from sleep state failed

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=85204

--- Comment #11 from Florent Bondoux <florent.bondoux at gmail.com> ---
I have the same bug with another GPU (Mobility HD 5430, code name is CEDAR).

Log after suspend/resume :

[  139.610475] [drm] PCIE GART of 1024M enabled (table at 0x000000000025E000).
[  139.610628] radeon 0000:01:00.0: WB enabled
[  139.610633] radeon 0000:01:00.0: fence driver on ring 0 use gpu addr
0x0000000020000c00 and cpu addr 0xffff8800b88d7c00
[  139.610635] radeon 0000:01:00.0: fence driver on ring 3 use gpu addr
0x0000000020000c0c and cpu addr 0xffff8800b88d7c0c
[  139.611445] radeon 0000:01:00.0: fence driver on ring 5 use gpu addr
0x000000000005c418 and cpu addr 0xffffc90005c9c418
[  139.628149] [drm] ring test on 0 succeeded in 0 usecs
[  139.628174] [drm] ring test on 3 succeeded in 3 usecs
[  139.805679] [drm] ring test on 5 succeeded in 1 usecs
[  139.805683] [drm] UVD initialized successfully.
[  139.805727] [drm] ib test on ring 0 succeeded in 0 usecs
[  140.457604] [drm] ib test on ring 5 succeeded
[  152.692727] radeon 0000:01:00.0: ring 0 stalled for more than 10463msec
[  152.692737] radeon 0000:01:00.0: GPU lockup (current fence id
0x00000000000015b8 last fence id 0x00000000000016b8 on ring 0)
[  152.692786] radeon 0000:01:00.0: failed to get a new IB (-35)
[  152.692795] [drm:radeon_cs_ib_fill] *ERROR* Failed to get ib !
[  152.750259] radeon 0000:01:00.0: Saved 8183 dwords of commands on ring 0.
[  152.750281] radeon 0000:01:00.0: GPU softreset: 0x00000008
[  152.750284] radeon 0000:01:00.0:   GRBM_STATUS               = 0xA0003828
[  152.750287] radeon 0000:01:00.0:   GRBM_STATUS_SE0           = 0x00000007
[  152.750290] radeon 0000:01:00.0:   GRBM_STATUS_SE1           = 0x00000007
[  152.750293] radeon 0000:01:00.0:   SRBM_STATUS               = 0x200000C0
[  152.750296] radeon 0000:01:00.0:   SRBM_STATUS2              = 0x00000000
[  152.750299] radeon 0000:01:00.0:   R_008674_CP_STALLED_STAT1 = 0x00000000
[  152.750302] radeon 0000:01:00.0:   R_008678_CP_STALLED_STAT2 = 0x00010100
[  152.750304] radeon 0000:01:00.0:   R_00867C_CP_BUSY_STAT     = 0x00020182
[  152.750307] radeon 0000:01:00.0:   R_008680_CP_STAT          = 0x80038243
[  152.750311] radeon 0000:01:00.0:   R_00D034_DMA_STATUS_REG   = 0x44C83D57
[  152.759991] radeon 0000:01:00.0: GRBM_SOFT_RESET=0x00004001
[  152.760049] radeon 0000:01:00.0: SRBM_SOFT_RESET=0x00000100
[  152.761216] radeon 0000:01:00.0:   GRBM_STATUS               = 0x00003828
[  152.761219] radeon 0000:01:00.0:   GRBM_STATUS_SE0           = 0x00000007
[  152.761222] radeon 0000:01:00.0:   GRBM_STATUS_SE1           = 0x00000007
[  152.761225] radeon 0000:01:00.0:   SRBM_STATUS               = 0x200000C0
[  152.761228] radeon 0000:01:00.0:   SRBM_STATUS2              = 0x00000000
[  152.761231] radeon 0000:01:00.0:   R_008674_CP_STALLED_STAT1 = 0x00000000
[  152.761233] radeon 0000:01:00.0:   R_008678_CP_STALLED_STAT2 = 0x00000000
[  152.761236] radeon 0000:01:00.0:   R_00867C_CP_BUSY_STAT     = 0x00000000
[  152.761239] radeon 0000:01:00.0:   R_008680_CP_STAT          = 0x00000000
[  152.761242] radeon 0000:01:00.0:   R_00D034_DMA_STATUS_REG   = 0x44C83D57
[  152.761267] radeon 0000:01:00.0: GPU reset succeeded, trying to resume
[  152.784402] [drm] PCIE GART of 1024M enabled (table at 0x000000000025E000).
[  152.784520] radeon 0000:01:00.0: WB enabled
[  152.784523] radeon 0000:01:00.0: fence driver on ring 0 use gpu addr
0x0000000020000c00 and cpu addr 0xffff8800b88d7c00
[  152.784526] radeon 0000:01:00.0: fence driver on ring 3 use gpu addr
0x0000000020000c0c and cpu addr 0xffff8800b88d7c0c
[  152.785283] radeon 0000:01:00.0: fence driver on ring 5 use gpu addr
0x000000000005c418 and cpu addr 0xffffc90005c9c418
[  152.801936] [drm] ring test on 0 succeeded in 1 usecs
[  152.801945] [drm] ring test on 3 succeeded in 3 usecs
[  152.979461] [drm] ring test on 5 succeeded in 1 usecs
[  152.979468] [drm] UVD initialized successfully.
[  163.207936] radeon 0000:01:00.0: ring 0 stalled for more than 10234msec
[  163.207946] radeon 0000:01:00.0: GPU lockup (current fence id
0x00000000000015b9 last fence id 0x00000000000016b9 on ring 0)
[  163.208075] [drm:r600_ib_test] *ERROR* radeon: fence wait failed (-35).
[  163.208088] [drm:radeon_ib_ring_tests] *ERROR* radeon: failed testing IB on
GFX ring (-35).


Bisecting gave commit [a3eb06dbca08e3fdad7039021ae03b46b215f22a] drm/radeon:
Remove radeon_gart_restore()

commit a3eb06dbca08e3fdad7039021ae03b46b215f22a
Author: Michel Dänzer <michel.daenzer at amd.com>
Date:   Wed Jul 9 20:15:42 2014 +0200

    drm/radeon: Remove radeon_gart_restore()

    Doesn't seem necessary, the GART table memory should be persistent.

    Signed-off-by: Michel Dänzer <michel.daenzer at amd.com>
    Reviewed-by: Christian König <christian.koenig at amd.com>
    Signed-off-by: Alex Deucher <alexander.deucher at amd.com>

That's the same commit as in bug #86267. The attached patches didn't help
either.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/dri-devel/attachments/20141225/f0ef4fc7/attachment-0001.html>