Bug 92087

Summary: [Radeon/Tonga] lockup in valley demo
Product: DRI Reporter: Edward O'Callaghan <edward.ocallaghan>
Component: DRM/AMDgpuAssignee: Default DRI bug account <dri-devel>
Status: RESOLVED DUPLICATE QA Contact:
Severity: major    
Priority: medium    
Version: unspecified   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:

Description Edward O'Callaghan 2015-09-23 09:04:01 UTC
Hi,

Upon running the valley demo I experience I lockup after a while. I can recover by switching to another virtual term and killing X.

dmesg shows this repeatedly with only a difference of the value after the fault id? of 146 after the lockup:

--------------------

VM fault (0x00, vmid 0) at page 0, read from '' (0x00000000) (0)
amdgpu 0000:01:00.0: GPU fault detected: 146 0x02704004
amdgpu 0000:01:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR   0x00000000
amdgpu 0000:01:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x00000000

--------------------

This is with mesa 11.0.1 (8ae8feca) + kern 4.3.0-rc2 + libdrm 2.4.65

No idea how to get a card state dump with amdgpu?

Edward.
Comment 1 Andy Furniss 2015-09-24 15:41:55 UTC
There are a couple of related bugs already open.

The VM fault seems to have appeared with more recent kernels, I can get those now even without locking.

If you leave valley for a couple of minutes after lock you will likely get a hung task trace like those in the bugs below (Though not a GPU reset as that is not enabled on more recent kernels)

It would probably be best to mark this bug a DUP of one of these.

https://bugs.freedesktop.org/show_bug.cgi?id=91278

https://bugs.freedesktop.org/show_bug.cgi?id=91790
Comment 2 Alex Deucher 2015-09-24 15:43:44 UTC

*** This bug has been marked as a duplicate of bug 91278 ***

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.