On Wed, Jun 06, 2018 at 11:52:02AM +0200, Borislav Petkov wrote:
> On Wed, Jun 06, 2018 at 10:26:15AM +0200, Christian König wrote:
> > Well what did you do to trigger the lockup? Looks like an application send
> > something to the hardware to crash the GFX block.
>
> So what I observed was (in that order): machine was building a kernel so
> was busy, X didn't respond for a couple of seconds (which it should not
> do) and I switched to tty8 to see what dmesg says (I am routing dmesg to
> tty8).
>
> At that moment it showed the lockup splat and then X restarted showing
> me the login prompt again. I'm guessing the X restart was caused by the
> GPU reset.
>
> So if I had to guess, maybe pressing Ctrl+Alt+F8 caused the modeset
> change and thus reset? Hmm, this is just me guessing with my
> non-knowledge about graphics. :)
Happened again, this time it became really pretty, borderline
psychedelic, see attached picture. :-)
Same issue - kernel build running.
[395849.320285] radeon 0000:1d:00.0: ring 0 stalled for more than 10052msec
[395849.320296] radeon 0000:1d:00.0: GPU lockup (current fence id 0x000000000176f423 last fence id 0x000000000176f50f on ring 0)
[395849.320324] radeon 0000:1d:00.0: failed to get a new IB (-35)
[395849.320372] [drm:radeon_cs_ioctl [radeon]] *ERROR* Failed to get ib !
[395849.327413] radeon 0000:1d:00.0: Saved 7545 dwords of commands on ring 0.
[395849.327428] radeon 0000:1d:00.0: GPU softreset: 0x00000008
[395849.327431] radeon 0000:1d:00.0: R_008010_GRBM_STATUS = 0xA0003030
[395849.327433] radeon 0000:1d:00.0: R_008014_GRBM_STATUS2 = 0x00000003
[395849.327435] radeon 0000:1d:00.0: R_000E50_SRBM_STATUS = 0x200010C0
[395849.327437] radeon 0000:1d:00.0: R_008674_CP_STALLED_STAT1 = 0x00000000
[395849.327439] radeon 0000:1d:00.0: R_008678_CP_STALLED_STAT2 = 0x00000000
[395849.327441] radeon 0000:1d:00.0: R_00867C_CP_BUSY_STAT = 0x00020186
[395849.327443] radeon 0000:1d:00.0: R_008680_CP_STAT = 0x80028645
[395849.327445] radeon 0000:1d:00.0: R_00D034_DMA_STATUS_REG = 0x44C83D57
[395849.395006] radeon 0000:1d:00.0: R_008020_GRBM_SOFT_RESET=0x00004001
[395849.395064] radeon 0000:1d:00.0: SRBM_SOFT_RESET=0x00000100
[395849.397160] radeon 0000:1d:00.0: R_008010_GRBM_STATUS = 0xA0003030
[395849.397165] radeon 0000:1d:00.0: R_008014_GRBM_STATUS2 = 0x00000003
[395849.397168] radeon 0000:1d:00.0: R_000E50_SRBM_STATUS = 0x200080C0
[395849.397170] radeon 0000:1d:00.0: R_008674_CP_STALLED_STAT1 = 0x00000000
[395849.397172] radeon 0000:1d:00.0: R_008678_CP_STALLED_STAT2 = 0x00000000
[395849.397175] radeon 0000:1d:00.0: R_00867C_CP_BUSY_STAT = 0x00000000
[395849.397177] radeon 0000:1d:00.0: R_008680_CP_STAT = 0x80100000
[395849.397179] radeon 0000:1d:00.0: R_00D034_DMA_STATUS_REG = 0x44C83D57
[395849.397187] radeon 0000:1d:00.0: GPU reset succeeded, trying to resume
[395849.412961] [drm] PCIE gen 2 link speeds already enabled
[395849.415304] [drm] PCIE GART of 512M enabled (table at 0x0000000000142000).
[395849.415326] radeon 0000:1d:00.0: WB enabled
[395849.415330] radeon 0000:1d:00.0: fence driver on ring 0 use gpu addr 0x0000000020000c00 and cpu addr 0x0000000057ff3827
[395849.415517] radeon 0000:1d:00.0: fence driver on ring 5 use gpu addr 0x00000000000521d0 and cpu addr 0x0000000047dfb920
[395849.446405] [drm] ring test on 0 succeeded in 0 usecs
[395849.621116] [drm] ring test on 5 succeeded in 1 usecs
[395849.621126] [drm] UVD initialized successfully.
[395849.672434] [drm] ib test on ring 0 succeeded in 0 usecs
[395850.344286] [drm] ib test on ring 5 succeeded
--
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gpu-stall.jpg
Type: image/jpeg
Size: 3080309 bytes
Desc: not available
URL: <https://lists.freedesktop.org/archives/amd-gfx/attachments/20180623/f2b33974/attachment-0001.jpg>
