Created attachment 99321 [details] /sys/class/drm/card0/error output I am on Kernel Linux 3.15.0-999-generic #201404260253 SMP Sat Apr 26 06:54:32 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux This bug happened the first time today during regular use having chromium, terminals and vim open (using the dwm window manager). The XServer seemed to hang and was not responsive for a few seconds. Soon after it started to react properly again and the dmesg output suggested to file a new bug report (it may be a duplicate of bug 75991 however). Please find the dmesg output, the /sys/class/drm/card0/error output and the Xorg.0.log attached.
Created attachment 99323 [details] dmesg output
Created attachment 99324 [details] Xorg log
Something (userspace) scribbled over offset 0 which had a pending batchbuffer at the time. It looks like output from the GPU, so most likely mesa.
Silvan, can you please retest with latest mesa git? There have been a bunch of bugfixes recently which could match.
I have been running the same kernel for a few weeks now and this is the first time I hit this bug. Sadly, I do not know how to reproduce it. With the bug not being reproducible at the moment would you still recommend to install the newest mesa (the currently installed version is 10.1)?
Can't hurt, but Chris said on irc that it's likely a different bug.
I compiled and installed the newest git version ( ff5456d1acf6f627a6837be3f3f37c6a268c9e8e ) of Mesa using the instructions on the site (basically just 'sudo make install'; the tests did not run through because of failed check in the gallium llvmpipe drivers that apparently is known). Is there a way to check which version of mesa is actually in use? I tried glxinfo but it errors out name of display: :0 Error: couldn't find RGB GLX visual or fbconfig Error: couldn't find RGB GLX visual or fbconfig which looks like mesa is not working at all...
*** Bug 79545 has been marked as a duplicate of this bug. ***
Created attachment 100311 [details] GPU crash dump Just happened to me again (see my duplicate bug), this time with another line in dmesg: [18280.992041] Watchdog[12283]: segfault at 0 ip 00007fe9bef9db38 sp 00007fe9ab14a720 error 6 in chromium[7fe9bb280000+5a14000] Crash dump attached
Created attachment 100610 [details] /sys/class/drm/card0/error stuck on render ring stuck on blitter ring 3.13.0-29-generic kernel, ubuntu 14.04, similar to initial report conditions firefox browser open, terminal open, scrolling through a browser viewed pdf page/screen froze, mouse pointer was responsive but pointer graphic stuck on text entry/capital 'I' normal service resumed ~10s later. Dmesg told me to do it.... [ 8499.125636] [drm] stuck on render ring [ 8499.125640] [drm] stuck on blitter ring x220 i5 problem encountered not encountered on x200s with same os.
(In reply to comment #10) > Dmesg told me to do it.... I am sure it told you to file a new bug so that you do not erroneously add your GPU hang to the wrong bug... Yours was bug 54226.
Closed after more than one year of inactivity. Feel free to reopen if needed. Thanks
Won't fix is the wrong answer; if we could we would. As initially suggested this was likely a bug in mesa that we needed confirmation.
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.