[Mesa-dev] [PATCH 1/4] r600g: avoid redundant DB register updates

Wed Apr 26 14:36:39 UTC 2017

HEAD recently have triggered a bug in sb compiler, so, just for the safe case, can you tell if reverting this commit

git revert --no-commit eb8aa93c03ee89ffd3041d41b6293e4b282b6ce6

but having my patch still applied changes anything or not?

(I have this commit reverted locally)

On 26.04.2017 17:23, Marc Dietrich wrote:
> Am Mittwoch, 26. April 2017, 16:16:35 CEST schrieb Constantine Kharlamov:
>> Thank you for testing! What is "UH4" btw?
> 
> old GL3.3 benchmark Unigine Heaven 4.0 
> 
> https://benchmark.unigine.com/heaven
> 
> Marc
> 
>>
>> On 26.04.2017 17:03, Marc Dietrich wrote:
>>> Am Dienstag, 25. April 2017, 13:59:44 CEST schrieb Constantine Kharlamov:
>>>> Split out of "avoid redundant CB register updates" for easier review.
>>>>
>>>> Signed-off-by: Constantine Kharlamov <Hi-Angel at yandex.ru>
>>>> Tested-by: Dieter Nützel <Dieter at nuetzel-hh.de>
>>>> ---
>>>
>>> This one generates GPU crashes here (RS880) when running UH4:
>>>
>>> radeon 0000:01:05.0: ring 0 stalled for more than 10240msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d865 on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 10460msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d934 on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 10972msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93d on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 11484msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93d on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 11996msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 12508msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 13020msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 13532msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 14044msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 14556msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 15068msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 15580msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 16092msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 16604msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 17116msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 17628msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 18140msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 18652msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 19164msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 19676msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d93f on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 20188msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d940 on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 20700msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d940 on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 21212msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d940 on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 21724msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d940 on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 22236msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d940 on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 22748msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d940 on ring 0)
>>> radeon 0000:01:05.0: ring 0 stalled for more than 23260msec
>>> radeon 0000:01:05.0: GPU lockup (current fence id 0x000000000000d85e last
>>> fence id 0x000000000000d940 on ring 0)
>>> radeon 0000:01:05.0: Saved 5241 dwords of commands on ring 0.
>>> radeon 0000:01:05.0: GPU softreset: 0x00000008
>>> radeon 0000:01:05.0:   R_008010_GRBM_STATUS      = 0xA0002030
>>> radeon 0000:01:05.0:   R_008014_GRBM_STATUS2     = 0x00000003
>>> radeon 0000:01:05.0:   R_000E50_SRBM_STATUS      = 0x20000040
>>> radeon 0000:01:05.0:   R_008674_CP_STALLED_STAT1 = 0x00000000
>>> radeon 0000:01:05.0:   R_008678_CP_STALLED_STAT2 = 0x00040002
>>> radeon 0000:01:05.0:   R_00867C_CP_BUSY_STAT     = 0x00020186
>>> radeon 0000:01:05.0:   R_008680_CP_STAT          = 0x80078645
>>> radeon 0000:01:05.0:   R_00D034_DMA_STATUS_REG   = 0x44C83D57
>>> radeon 0000:01:05.0: R_008020_GRBM_SOFT_RESET=0x00004001
>>> radeon 0000:01:05.0: SRBM_SOFT_RESET=0x00000100
>>> radeon 0000:01:05.0:   R_008010_GRBM_STATUS      = 0xA0003030
>>> radeon 0000:01:05.0:   R_008014_GRBM_STATUS2     = 0x00000003
>>> radeon 0000:01:05.0:   R_000E50_SRBM_STATUS      = 0x20008040
>>> radeon 0000:01:05.0:   R_008674_CP_STALLED_STAT1 = 0x00000000
>>> radeon 0000:01:05.0:   R_008678_CP_STALLED_STAT2 = 0x00000000
>>> radeon 0000:01:05.0:   R_00867C_CP_BUSY_STAT     = 0x00000000
>>> radeon 0000:01:05.0:   R_008680_CP_STAT          = 0x80100000
>>> radeon 0000:01:05.0:   R_00D034_DMA_STATUS_REG   = 0x44C83D57
>>> radeon 0000:01:05.0: GPU reset succeeded, trying to resume
>>> [drm] PCIE GART of 512M enabled (table at 0x00000000C0146000).
>>> radeon 0000:01:05.0: WB enabled
>>> radeon 0000:01:05.0: fence driver on ring 0 use gpu addr
>>> 0x00000000a0000c00
>>> and cpu addr 0xffff976161e88c00
>>> radeon 0000:01:05.0: fence driver on ring 5 use gpu addr
>>> 0x00000000c0056038
>>> and cpu addr 0xffffa96581816038
>>> [drm] ring test on 0 succeeded in 1 usecs
>>> [drm] ring test on 5 succeeded in 1 usecs
>>> [drm] UVD initialized successfully.
>>> [drm:r600_ib_test [radeon]] *ERROR* radeon: fence wait timed out.
>>> [drm:radeon_ib_ring_tests [radeon]] *ERROR* radeon: failed testing IB on
>>> GFX ring (-110).
>>>
>>>
>>>
>>> _______________________________________________
>>> mesa-dev mailing list
>>> mesa-dev at lists.freedesktop.org
>>> https://lists.freedesktop.org/mailman/listinfo/mesa-dev