Summary: | x11-drivers/xf86-video-intel-2.21.2 random crash black screen and logout | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Paul Littlefield <info> |
Component: | [OLD] Unspecified | Assignee: | Gentoo X packagers <x11> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | me |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: | tarball of Xorg logs |
Description
Paul Littlefield
2013-02-19 16:20:58 UTC
Please attach Xorg.0.log from the crash (will be renamed to Xorg.0.log.old when the server restarts). Also check if it happens both with uxa and sna. OK, will do at the end of my working day :) OK, this is right now before the changeover test. I have sna USE flag enabled and it has been running fine all day, in fact has been fine for weeks. thinkpad ~ # eix -I intel [U] x11-drivers/xf86-video-intel Available versions: ~*2.9.1 2.19.0 (~)2.20.2 2.20.13 (~)2.20.19-r1 (~)2.21.0 (~)2.21.2 {{debug dri glamor (+)sna +udev uxa xvmc}} Installed versions: 2.20.19-r1(10:11:36 19/02/13)(dri sna udev -glamor -uxa -xvmc) Homepage: http://xorg.freedesktop.org/ Description: X.Org driver for Intel cards I will now upgrade to 2.21.2 with 'sna' enabled and see if it borks... My apologies for the delay... work and kids got in the way :) OK, here are the results... x11-drivers/xf86-video-intel-2.20.19-r1 +sna = OK :) x11-drivers/xf86-video-intel-2.21.2 +sna = bork to login screen :( x11-drivers/xf86-video-intel-2.21.3 +sna = bork to login screen :( x11-drivers/xf86-video-intel-2.21.3 -sna +uxa = OK :) I have looked in the .log file and nothing obvious is mentioned :( So, for some reason, USE="sna" is fine in 2.20.19-r1 (and has been for weeks!) but not in the later releases. I cannot really tell what performance benefit or hit I am getting with SNA so I am not really sure if I am missing anything. I just compiled it in because I thought it was better, and has been really smooth for weeks. I am now using 2.21.3 and will for this week to se how it behaves in "real use". :) I hope that this helps. Regards Paully We still need to see Xorg.0.log from the crash. SNA is the future and UXA will go away sooner or later, so using UXA is only for isolating the issue and not a sustainable solution. If you have the time, a full backtrace as described in http://wiki.x.org/wiki/Development/Documentation/ServerDebugging would be good. OK, where do I upload the .log file? As an attachment or some online location? Also, is it worth mentioning I am using the following? Oh yes, and the problem still happened when using stable X server. Portage 2.1.11.50 (default/linux/amd64/13.0/desktop, gcc-4.7.2, glibc-2.15-r3, 3.7.9-gentoo x86_64) Gentoo Base System release 2.1 This is a pre-release version of the X server from The X.Org Foundation. latest version in the X.Org Foundation git repository. X.Org X Server 1.13.2.901 (1.13.3 RC 1) xfce-base/xfdesktop-4.10.0 x11-drivers/xf86-video-intel-2.21.3 Paully The Xorg.0.log is best attached to this bug. Compress it if it is too large for an attachment. You can post the backtrace in a comment. Created attachment 340060 [details]
tarball of Xorg logs
[ 436.163] (EE) Backtrace: [ 436.175] (EE) 0: /usr/bin/X (xorg_backtrace+0x36) [0x58a2a6] [ 436.176] (EE) 1: /usr/bin/X (0x400000+0x18dfb9) [0x58dfb9] [ 436.176] (EE) 2: /lib64/libpthread.so.0 (0x7fe7e31a3000+0x10ae0) [0x7fe7e31b3ae0] [ 436.176] (EE) 3: /usr/lib64/xorg/modules/drivers/intel_drv.so (0x7fe7e0bc7000+0x98d58) [0x7fe7e0 c5fd58] [ 436.176] (EE) 4: /usr/lib64/xorg/modules/drivers/intel_drv.so (0x7fe7e0bc7000+0x9d6dc) [0x7fe7e0c646dc] [ 436.176] (EE) 5: /usr/lib64/xorg/modules/drivers/intel_drv.so (0x7fe7e0bc7000+0x66e16) [0x7fe7e0c2de16] [ 436.176] (EE) 6: /usr/lib64/xorg/modules/drivers/intel_drv.so (0x7fe7e0bc7000+0x6762d) [0x7fe7e0c2e62d] [ 436.176] (EE) 7: /usr/lib64/xorg/modules/drivers/intel_drv.so (0x7fe7e0bc7000+0x729f9) [0x7fe7e0c399f9] [ 436.176] (EE) 8: /usr/bin/X (0x400000+0x10eb8b) [0x50eb8b] [ 436.176] (EE) 9: /usr/bin/X (0x400000+0x3b2d1) [0x43b2d1] [ 436.176] (EE) 10: /usr/bin/X (0x400000+0x29eda) [0x429eda] [ 436.176] (EE) 11: /lib64/libc.so.6 (__libc_start_main+0xf5) [0x7fe7e1e346c5] [ 436.176] (EE) 12: /usr/bin/X (0x400000+0x2a21d) [0x42a21d] [ 436.176] (EE) [ 436.176] (EE) Bus error at address 0x7fe7dc657000 [ 436.176] Fatal server error: [ 436.176] Caught signal 7 (Bus error). Server aborting This crash is firmly in xf86-video-intel code so best report this upstream at https://bugs.freedesktop.org/ with the backtrace as per comment 5 and link the upstream report to this bug. OK, I will try my best. I have briefly skimmed the BackTrace howto. In a nutshell, do I need to recompile the Xorg ebuild with a flag and try it all again? Eeek. Yes, emerge xorg-server, xf86-video-intel with CFLAGS="-g" and FEATURES="splitdebug" and reproduce the crash with gdb attached (best via ssh session). OK, will give it a go later tonight. Oh, I also have display compositing turned on in XFCE windows manager... not for the pointless effects nonsense, but for the REALLY smooth fonts it gives. Worth mentioning as well? It seems I have a very similar or the same issue :( @paul did your open a upstream bug? Can you retest with 2.21.15 ? (In reply to David Heidelberger (okias) from comment #15) > Can you retest with 2.21.15 ? for me the issue seems to be gone long ago, just forgot the mention it here. I'm on ~arch, so using 2.99.x atm Marking as fixed per comment #16. Reporter: if you still see this bug, please reopen. |