[Bug 206475] amdgpu under load drop signal to monitor until hard reset

bugzilla-daemon at bugzilla.kernel.org bugzilla-daemon at bugzilla.kernel.org
Fri May 22 12:55:35 UTC 2020


andrewammerlaan at riseup.net (andrewammerlaan at riseup.net) changed:

           What    |Removed                     |Added
                 CC|                            |andrewammerlaan at riseup.net

--- Comment #10 from andrewammerlaan at riseup.net (andrewammerlaan at riseup.net) ---
Created attachment 289235
  --> https://bugzilla.kernel.org/attachment.cgi?id=289235&action=edit

I think I ran into this issue as well. It has happened twice. Both times it
happened 10 to 20 minutes *after* playing minecraft. Both times I was in a full
screen video meeting. Everything works, except the screen goes black, I could
finish the meeting, but without seeing anything. 

Only the monitors connected to my RX 590 go black, the one connected to the
iGPU just freezes, and after a while the cursor becomes usable again on that
monitor, though all applications remain frozen, and switching to tty does not
work. REISUB'ing the machine makes it boot on the iGPU. It needs to be
completely switched on and off to boot from the amdgpu.

It looks like it does a graphics reset (why though?):
15554.332021] amdgpu 0000:01:00.0: GPU reset begin!

And from that point onwards everyting goes wrong:
[15554.332296] amdgpu: [powerplay] 
[15554.332296]  last message was failed ret is 65535
[15554.332297] amdgpu: [powerplay] 
[15554.332297]  failed to send message 261 ret is 65535 
[15554.332297] amdgpu: [powerplay] 
[15554.332297]  last message was failed ret is 65535

This is kernel 5.6.14

Full log is attached.

You are receiving this mail because:
You are watching the assignee of the bug.

More information about the dri-devel mailing list