Comment 50 for bug 986524

Revision history for this message
wdesmet (kromagg) wrote : Re: radeon 0000:02:00.0: GPU lockup CP stall for more than 10000msec

Not sure if this is the same bug, but mine starts slightly differently, i.e.:
kernel: [79731.820027] radeon 0000:01:00.0: GPU lockup CP stall for more than 10000msec
kernel: [79731.820033] radeon 0000:01:00.0: GPU lockup (waiting for 0x000000000006efb7 last fence id 0x000000000006efb6)
kernel: [79731.820090] radeon 0000:01:00.0: couldn't schedule ib
kernel: [79731.820093] [drm:radeon_cs_ib_chunk] *ERROR* Failed to schedule IB !
kernel: [79731.820119] radeon 0000:01:00.0: couldn't schedule ib
kernel: [79731.820123] [drm:radeon_cs_ib_chunk] *ERROR* Failed to schedule IB !
kernel: [79731.821373] radeon 0000:01:00.0: couldn't schedule ib

And then just keeps going. After a while, it will start to output lines about failed allocations:
kernel: [79799.398689] [drm:radeon_gem_object_create] *ERROR* Failed to allocate GEM object (14327808, 6, 4096, -12)
kernel: [79799.398766] [drm:radeon_gem_object_create] *ERROR* Failed to allocate GEM object (14327808, 6, 4096, -12)
kernel: [79799.398849] [drm:radeon_gem_object_create] *ERROR* Failed to allocate GEM object (14327808, 6, 4096, -12)
kernel: [79799.398929] [drm:radeon_gem_object_create] *ERROR* Failed to allocate GEM object (14327808, 6, 4096, -12)
kernel: [79799.398971] [drm:radeon_gem_object_create] *ERROR* Failed to allocate GEM object (14327808, 6, 4096, -12)

The system quickly becomes unusable when this happens. The only solution is to reboot and wait till it happens again. It seems to always happen while scrolling or selecting text in google chrome (which I imagine can be taxing on some sites).