Bug 18250

Created attachment 19886 [details]
/var/log/dmesg with drm.ko debug=1

Is this a regression, or is your first attempt at using nouveau on this machine? 

Also, is it reproducible? If so, using ssh to attach gdb remotely before the crash would help get a better back-trace (gdb /usr/bin/Xorg `pidof Xorg` , then "cont", and after the crash "bt"). If it's a regression, can you use git bisect on the ddx to point to a suspect commit?

I don't know if it's a regression: it's the 1st time I noticed this error, after using nouveau for about 20 hours.
I've used it before (with ~3rd Oct. and ~19th Oct.-git), but it crashed in a few hours with "DMA queue hang", both of times, so I can't tell when did this one start.

At the moment I don't know if it reproducible, either. As I said it's the first time, and btw I've not used any graphic-intensive program (except for kde 4.1 as a desktop manager, no compiz, no 3D, no games).
But as soon as I can, after next crash, I will send a more complete gdb  backtrace.
Thank you.

Created attachment 19919 [details]
gdb output with bt and bt full

This is what gdb displays.

BTW I've no errors in Xorg.0.log

Created attachment 19920 [details]
concerning lines in dmesg (last)

Before and after the "Segmentation fault" I've got these lines in /var/log/dmesg

Created attachment 19933 [details]
Xorg.0.log

I've had another crash, but of a different kind, with last git after about 12-13 hours.

Fatal server error:
DMA queue hang: dmaPut=e0, current=ec, status=0
preceeded by:
II) NOUVEAU(0): Fifo dump (lockup 0x00ec,0x00e0):



Before and after that, repeated messages:
"""
tossed event which came in late
mieqEnequeue: out-of-order valuator event; dropping.
"""

Created attachment 19934 [details]
dmesg (only drm-related parts)

In dmesg file I see same errors as before:
PGRAPH_ERROR - nSource: METHOD_CNT, nStatus: INVALID_STATE
[..]

preceeded by:
[drm:nouveau_fifo_free] *ERROR* Failed to idle channel 1 before destroy.Prepare for strangeness..

Created attachment 19935 [details]
backtrace

gdb backtrace

I don't know if I can be helpful (maybe no: if so, sorry!), but enabling
Option "ShadowFB"
in xorg.conf file I don't see any X crash / DMA queue hang (so when acceleration is disabled this driver seems to be stable).

It appears that this bug report has laid dormant for quite a while. Sorry we haven't gotten to it. Since we fix bugs all the time, chances are pretty good that your issue has been fixed with the latest software. Please give it a shot. (Linux kernel 3.10.7, xf86-video-nouveau 1.0.9, mesa 9.1.6, or their git versions.) If upgrading to the latest isn't an option for you, your distro's bugzilla is probably the right destination for your bug report.

In an effort to clean up our bug list, we're pre-emptively closing all bugs that haven't seen updates since 2011. If the original issue remains, please make sure to provide fresh info, see http://nouveau.freedesktop.org/wiki/Bugs/ for what we need to see, and re-open this one.

Thanks,

The Nouveau Team