https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c1 Stefan Dirsch changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |NEEDINFO InfoProvider| |bphilips@novell.com --- Comment #1 from Stefan Dirsch 2010-11-10 06:10:46 UTC --- Could you update kernel and X.Org zypper ar -f \ http://download.opensuse.org/repositories/Kernel:/HEAD/openSUSE_Factory/ \ Kernel:HEAD zypper ar -f \ http://download.opensuse.org/repositories/X11:/XOrg/openSUSE_Factory/ \ X11:XOrg zypper mr -p 90 Kernel:HEAD zypper mr -p 90 X11:XOrg zypper ref zypper in libtalloc2 zypper dup -r Kernel:HEAD zypper dup -r X11:XOrg zypper in xorg-x11-server-debuginfo xorg-x11-server-debugsource \ xorg-x11-driver-video-debuginfo xorg-x11-driver-video-debugsource \ xorg-x11-driver-input-debuginfo xorg-x11-driver-input-debugsource \ libpixman-1-0-debuginfo libpixman-1-0-debugsource \ libpciaccess0-debuginfo libpciaccess0-debugsource \ Mesa-debuginfo Mesa-debugsource and attach /var/log/Xorg.0.log? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c3 --- Comment #3 from Brandon Philips 2010-11-10 20:51:56 UTC --- Created an attachment (id=399491) --> (http://bugzilla.novell.com/attachment.cgi?id=399491) gdb backtrace -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c5 Brandon Philips changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |ASSIGNED InfoProvider|bphilips@novell.com | --- Comment #5 from Brandon Philips 2010-11-10 21:52:15 UTC --- (In reply to comment #4)

Any chance to debug this any further? The SIGABRT appears to happen in malloc. Not sure why.

I don't know where to start. Any pointers on how to run valgrind over Xorg successfully? Replacing /usr/bin/Xorg with a script that calls valgrind over /usr/bin/Xorg.orig doesn't seem to get me up to a working session. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c7 --- Comment #7 from Stefan Dirsch 2010-11-24 22:48:12 UTC --- 2 weeks later. Any news on that one? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c9 Brandon Philips changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |REOPENED Resolution|NORESPONSE | --- Comment #9 from Brandon Philips 2010-12-04 00:33:45 UTC --- (In reply to comment #6)

What about attaching gdb to Xorg process and setting a breakpoint in RegionCreate() ?

Sorry I tried this earlier but forgot to respond. RegionCreate is called constantly particularly with rect=0x0, size=0. Any suggestions on how to narrow it down a bit? It seems like the corruption is happening long before RegionCreate since this function call happens so often in Xorg. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c11 --- Comment #11 from Brandon Philips 2010-12-07 04:32:32 UTC --- Created an attachment (id=403685) --> (http://bugzilla.novell.com/attachment.cgi?id=403685) Screenshot of corruption Adding MALLOC_CHECK_ options must adjust the timing just enough to avoid the bug. I can't trigger it when this is set. However, when X starts there are some very noticeable rendering errors all over the screen. See attached. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c Brandon Philips changed: What |Removed |Added ---------------------------------------------------------------------------- Attachment #403686|application/octet-stream |text/plain mime type| | -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c13 Stefan Dirsch changed: What |Removed |Added ---------------------------------------------------------------------------- Summary|Xorg: SIGABRT at very start |nouveau [NV98] SIGABRT of |of session w/ emacs and |Xorg due to mailloc |dual head |failures at very start of | |session w/ emacs and dual | |head --- Comment #13 from Stefan Dirsch 2010-12-07 07:33:30 UTC --- Well, I'm afraid that's deep inside Kernel DRM. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c14 Brandon Philips changed: What |Removed |Added ---------------------------------------------------------------------------- Component|Kernel |X.Org Product|openSUSE 11.4 |openSUSE 11.3 Target Milestone|--- |Factory --- Comment #14 from Brandon Philips 2010-12-07 21:12:39 UTC --- (In reply to comment #13)

Well, I'm afraid that's deep inside Kernel DRM.

Right, but I don't think it is the root cause. It may just be a symptom of Xorg crashing in some crazy state. Is there anything else I can try to debug Xorg? There is still some sort of bug that is being papered over by using MALLOC_CHECK_. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c15 Brandon Philips changed: What |Removed |Added ---------------------------------------------------------------------------- Priority|P3 - Medium |P2 - High --- Comment #15 from Brandon Philips 2010-12-07 21:17:10 UTC --- Increasing to high priority since shipping a NVIDIA driver that crashes when launching applications is fairly critical. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c17 --- Comment #17 from Jeff Mahoney 2010-12-07 23:23:57 UTC --- Strangely, the problem goes away for me when I add the following to /etc/init.d/xdm. export LIBC_FATAL_STDERR_=1 Stefan, what do you think the chances are that writing to the underlying vt while the X server is up is causing this issue? When glibc is aborting, at least using these code paths, it writes to the controlling terminal directly - bypassing stderr. gimp was reliably crashing for me. As was totem, usually in the first few seconds of playing a video. With LIBC_FATAL_STDERR_=1, gimp starts (with visual corruption) and totem works fine. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c19 --- Comment #19 from Jeff Mahoney 2010-12-08 01:04:28 UTC --- Seems mine is different as mine is in free(), not malloc. But, if it'll help: (gdb) list *+0x8528 0x8528 is in exaDestroyPixmap_mixed (exa_mixed.c:259). 254 pExaScr->info->DestroyPixmap(pScreen, pExaPixmap->driverPriv); 255 pExaPixmap->driverPriv = NULL; 256 257 if (pExaPixmap->pDamage) { 258 free(pExaPixmap->sys_ptr); 259 pExaPixmap->sys_ptr = NULL; 260 pExaPixmap->pDamage = NULL; 261 } 262 } 263 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=652523 https://bugzilla.novell.com/show_bug.cgi?id=652523#c21 Brandon Philips changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |CLOSED See Also| |https://bugs.freedesktop.or | |g/show_bug.cgi?id=32453 Resolution| |UPSTREAM --- Comment #21 from Brandon Philips 2010-12-17 00:04:16 UTC --- Filed upstream. The backtrace changed slightly: https://bugs.freedesktop.org/show_bug.cgi?id=32453 Closing as upstream, again :( -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.