Created attachment 110408 [details]
gpu crash dump
easiest way to trigger the problem is by watching youtube videos with chromium browser in gnome3 - but also occurs when loading openstreetmap widget in the browser.
This is with a Fujitsu Esprimo Mobile U9200 notebook, Intel 965GM integrated graphics, using Mageia 4
* Screen goes black, switching to different virtual consoles or attempt to switch brightness doesn't bring it back on.
* suspend to RAM and waking it up brings back the display
although gnome-session fails to resume, shows the "cannot recover, log-out" screen. You can kill the gnome-session-failed process from a VT and launch metacity or some other window manager to clean up/save your documents though.
Created attachment 110409 [details]
journalctl log from crash with traces and suspend/resume cycle
the suggestion from bug#80568#c62 (setting always_flush_cache=true) seems to workaround my problem as well, so very likely duplicate.
I suspect this may be another duplicate of the bug 80568, fixed (worked-around) by this commit:
Author: Kenneth Graunke <email@example.com>
Date: Sat Jan 17 23:21:15 2015 -0800
i965: Work around mysterious Gen4 GPU hangs with minimal state changes.
Gen4 hardware appears to GPU hang frequently when using Chromium, and
also when running 'glmark2 -b ideas'. Most of the error states contain
3DPRIMITIVE commands in quick succession, with very few state packets
between them - usually VERTEX_BUFFERS/ELEMENTS and CONSTANT_BUFFER.
I trimmed an apitrace of the glmark2 hang down to two draw calls with a
glUniformMatrix4fv call between the two. Either draw by itself works
fine, but together, they hang the GPU. Removing the glUniform call
makes the hangs disappear. In the hardware state, this translates to
removing the CONSTANT_BUFFER packet between the two 3DPRIMITIVE packets.
Flushing before emitting CONSTANT_BUFFER packets also appears to make
the hangs disappear. I observed a slowdown in glxgears by doing it all
the time, so I've chosen to only do it when BRW_NEW_BATCH and
BRW_NEW_PSP are unset (i.e. we haven't done a CS_URB_STATE change or
already flushed the whole pipeline).
I'd much rather understand the problem, but at this point, I don't see
how we'd ever be able to track it down further. We have no real tools,
and the hardware people moved on years ago. I've analyzed 20+ error
states and read every scrap of documentation I could find.
Signed-off-by: Kenneth Graunke <firstname.lastname@example.org>
Acked-by: Matt Turner <email@example.com>
Cc: "10.4 10.3" <firstname.lastname@example.org>
It's in git, and backports are in Mesa 10.4.x for x > 3. Please try upgrading to >10.4.3. If it's resolved by such an upgrade, please mark as a duplicate of bug 80568.
No reply. Marking as duplicate.
*** This bug has been marked as a duplicate of bug 80568 ***