Created attachment 21738 [details]
log file showing backtrace
Forwarding this bug from a Ubuntu reporter:
X freeze triggered by Klondike Solitaire on 945G graphics on ASUSTeK Computer.
The X system crashed while playing Solitaire Aisleriot.
Ubuntu 8.10, all packages upgraded as of today 15-11-2008.
Solitaire froze while dragging a card. Then the display disappeared. I saw a dialogue box about a "safe mode" or something I didn't have the time to read, even less to click OK. I didn't get safe mode or any other mode working. Neither Ctrl+Alt+F1 or Alt+F1 worked to log in in another terminal and emit a "shutdown". I had to reset the computer.
This is the third time, I think, since I upgraded to 8.10 Intrepid. It never happened before. I do not know whether the problem lies with /usr/games/sol or with the X system (or gdm). (I do now I was playing Solitaire also on the second crash. I do not remember for the first one.) Anyway it is extremely serious because one could be running something important in the background but it was impossible to regain control of the system without a reset.
Fatal server error:
0: /usr/X11R6/bin/X(xf86SigHandler+0x79) [0x80c3009]
2: /usr/lib/xorg/modules/drivers//intel_drv.so [0xb7a22b50]
3: /usr/X11R6/bin/X [0x80d6b0a]
4: /usr/lib/xorg/modules/extensions//libglx.so [0xb7adabe9]
5: /usr/X11R6/bin/X(AbortDDX+0x79) [0x80a8b09]
6: /usr/X11R6/bin/X(AbortServer+0x28) [0x813c498]
7: /usr/X11R6/bin/X(FatalError+0x63) [0x813caa3]
8: /usr/lib/xorg/modules/drivers//intel_drv.so(I830WaitLpRing+0x201) [0xb7a171d1]
9: /usr/lib/xorg/modules/drivers//intel_drv.so(I830Sync+0x1c3) [0xb7a175e3]
10: /usr/lib/xorg/modules/drivers//intel_drv.so [0xb7a3f7ea]
11: /usr/lib/xorg/modules//libexa.so(exaWaitSync+0x65) [0xb78f1045]
12: /usr/lib/xorg/modules//libexa.so(ExaDoPrepareAccess+0x7e) [0xb78f223e]
13: /usr/lib/xorg/modules//libexa.so(ExaCheckPutImage+0x103) [0xb78f9e03]
14: /usr/lib/xorg/modules//libexa.so [0xb78f3585]
15: /usr/X11R6/bin/X [0x817948d]
16: /usr/X11R6/bin/X(ProcPutImage+0x15e) [0x808951e]
17: /usr/X11R6/bin/X(Dispatch+0x34f) [0x808c89f]
18: /usr/X11R6/bin/X(main+0x47d) [0x8071d1d]
19: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe5) [0xb7bb2685]
20: /usr/X11R6/bin/X [0x8071101]
Saw signal 11. Server aborting.
There is nothing in /var/crash.
Nov 15 17:14:37 library gdm: WARNING: gdm_slave_xioerror_handler : erreur X fatale - Redémarrage de :0
Nov 15 17:14:46 library gdm: WARNING: gdm_slave_xioerror_handler : erreur X fatale - Redémarrage de :1
Nov 15 17:14:52 library gdm: WARNING: gdm_slave_xioerror_handler : erreur X fatale - Redémarrage de :1
Nov 15 17:14:58 library gdm: WARNING: gdm_slave_xioerror_handler : erreur X fatale - Redémarrage de :1
Nov 15 17:14:58 library gdm: CRITICAL: gdm_config_value_get_bool: assertion `value->type == GDM_CONFIG_VALUE_BOOL' failed
Nov 15 17:15:09 library gdm: WARNING: gdm_slave_xioerror_handler : erreur X fatale - Redémarrage de :0
Nov 15 17:15:09 library gdm: CRITICAL: gdm_config_value_get_bool: assertion `value->type == GDM_CONFIG_VALUE_BOOL' failed
Nov 15 17:15:16 library gdm: CRITICAL: gdm_config_value_get_bool: assertion `value->type == GDM_CONFIG_VALUE_BOOL' failed
Nov 15 17:14:36 library kernel: [ 2262.325308] Xorg: segfault at 6574616c ip b7aa3d64 sp bfdc07c0 error 4 in libdri.so[b7aa0000+8000]
Nov 15 17:14:45 library kernel: [ 2271.285551] Xorg: segfault at 6574616c ip b7a6ed64 sp bf88c4f0 error 4 in libdri.so[b7a6b000+8000]
Nov 15 17:14:52 library kernel: [ 2277.757315] Xorg: segfault at 6574616c ip b7a8ad64 sp bf9a6610 error 4 in libdri.so[b7a87000+8000]
Nov 15 17:14:58 library kernel: [ 2284.201318] Xorg: segfault at 6574616c ip b7a92d64 sp bfdaf220 error 4 in libdri.so[b7a8f000+8000]
Nov 15 17:15:01 library kernel: [ 2286.701702] mtrr: base(0xe0000000) is not aligned on a size(0x7b0000) boundary
Nov 15 17:15:09 library kernel: [ 2295.041754] Xorg: segfault at 6574616c ip b7b92d64 sp bfdb0a20 error 4 in libdri.so[b7b8f000+8000]
Two other messages:
Nov 15 17:14:37 library bonobo-activation-server (mic-7003): could not associate with desktop session: Failed to connect to socket /tmp/dbus-FCFw0yqGV2: Connexion refusée
Nov 15 17:14:42 library bonobo-activation-server (mic-7051): could not associate with desktop session: Failed to connect to socket /tmp/dbus-FCFw0yqGV2: Connexion refusée
Please provide complete Xorg log, xorg.conf.
Created attachment 21821 [details]
Created attachment 21822 [details]
I can produce with 8.10, but can't work recent driver. So please try current driver release.
Intel(R) Celeron(R) CPU E1200 @ 1.60GHz, 1600 MHz
Intel(R) Celeron(R) CPU E1200 @ 1.60GHz, 1600 MHz
/dev/input/event1 AT Translated Set 2 keyboard
/dev/input/mice Macintosh mouse button emulation
/dev/input/mice ImExPS/2 Logitech Explorer Mouse
AisleRiot, now 184.108.40.206
I still had a freeze of Aisleriot Solitaire Klondike. This time X didn't crash but froze and I had to reset. See https://bugs.launchpad.net/bugs/298483
I'll be glad to provide more information if needed and if I can.
From my testing, newer driver has no hang problem with sol, so please try recent releases for xf86-video-intel, kernel, mesa, drm, etc.
(In reply to comment #7)
> From my testing, newer driver has no hang problem with sol, so please try
> recent releases for xf86-video-intel, kernel, mesa, drm, etc.
zhenyu, we are required _not_ to use the ambiguous words like "recent release". Would you point Dominique the commit id of each components you used that can work?
sigh, each component has very clear recent releases.
Please try xf86-video-intel 2.6.1, kernel 2.6.28 stable, mesa 7.3, libdrm 2.4.4.
zhenyu, thank you for confirming the exact versions. Since those versions are all in Jaunty, this means the user can avoid having to build components themselves in order to test.
zhenyu, one question. Do you know what change specifically fixed the problem? If not, you mentioned you reproduced the issue on 8.10 - can you post the complete backtrace here for review?
I don't know exactly which commit might fix the problem, the crash backtrace looks just same as yours.
Bryce, would you please ping the bug reporter to verify?
I already did so. The bug is difficult to reproduce (occurs only every few weeks) so would require upgrading to the unstable development version and running on it for some time, which he is unwilling to do without definitive confirmation that the issue was definitely fixed.
we can do nothing if the bug reporter is unwilling to verify our result. I'll mark this as FIXED and please feel free to reopen if he is able to action and reproduce this bug again...