[Nouveau] [Bug 80865] [NVE7] Hard hang (GPC0/TPC0/MP trap: MULTIPLE_WARP_ERRORS MEM_OUT_OF_BOUNDS)

[bugzilla-daemon at freedesktop.org]

https://bugs.freedesktop.org/show_bug.cgi?id=80865

Ilia Mirkin <imirkin at alum.mit.edu> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|Hard hang (GPC0/TPC0/MP     |[NVE7] Hard hang
                   |trap: MULTIPLE_WARP_ERRORS  |(GPC0/TPC0/MP trap:
                   |MEM_OUT_OF_BOUNDS)          |MULTIPLE_WARP_ERRORS
                   |                            |MEM_OUT_OF_BOUNDS)

--- Comment #5 from Ilia Mirkin <imirkin at alum.mit.edu> ---
Those errors are most likely to happen when e.g. you're accessing a UBO out of
its bounds. Or if the driver messed up binding said UBO... or some code
generation error. However it should be ~harmless, except for the fact that the
render in question will be messed up.

BTW, what mesa version are you using?

The DISP errors at the end are the reason (or at least a demonstration) that
the GPU hung. I suspect the rest of the system was totally fine.

This also looks worrying:

[160330.118473] nouveau E[   PFIFO][0000:01:00.0] write fault at 0x00004d6000
[PTE] from GR/GPC0/PROP_0 on channel 0x007f971000 [Xorg[981]]
[160330.118474] nouveau E[   PFIFO][0000:01:00.0] PGRAPH engine fault on
channel 2, recovering...

I'm guessing the errors stressed the ctxsw logic and it couldn't handle some
condition.

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/nouveau/attachments/20140703/1a00a78a/attachment.html>