[hardy] Xorg crash in _mesa_update_texture()

[Olivier]

Binary package hint: xserver-xorg

I experienced a crash of the X server. the relevant part of /var/log/Xorg.0.log.old is

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x7e) [0x80c744e]
1: [0xffffe420]
2: /usr/lib/dri/r300_dri.so(_mesa_update_texture+0x219) [0xaf7e84d9]
3: /usr/lib/dri/r300_dri.so(_mesa_update_state_locked+0x628) [0xaf7d0b78]
4: /usr/lib/dri/r300_dri.so(_mesa_update_state+0x2a) [0xaf7d0c8a]
5: /usr/lib/dri/r300_dri.so(_mesa_GetIntegerv+0x258) [0xaf8b28c8]
6: /usr/lib/xorg/modules/extensions//libglx.so [0xb7bd4bc4]
7: /usr/lib/xorg/modules/extensions//libglx.so [0xb7b9a09c]
8: /usr/lib/xorg/modules/extensions//libglx.so [0xb7b99777]
9: /usr/lib/xorg/modules/extensions//libglx.so [0xb7b9e0a6]
10: /usr/bin/X [0x814feee]
11: /usr/bin/X(Dispatch+0x2cf) [0x808d93f]
12: /usr/bin/X(main+0x48b) [0x80747ab]
13: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb7d2a050]
14: /usr/bin/X(FontFileCompleteXLFD+0x20d) [0x8073b21]

Fatal server error:
Caught signal 11. Server aborting

I am using an up to date hardy with an X600 mobility card with compiz enabled.

Please feel free to ask for any additional info

[Nanley Chery (nanoman)]

Thank you for taking the time to report this bug and helping to make Ubuntu better. Please answer these questions:

* Is this reproducible?
* If so, what specific steps should we take to recreate this bug?

This will help us to find and resolve the problem.

[Olivier (olivier-lacroix)]

I fear this is not really reproductible. At least, I have not encountered this problem since I reported the crash.

Feel free to mark this bug as invalid. I will reopen it if I come across an other crash

[Nanley Chery (nanoman)]

Your wish is my command.

[Olivier (olivier-lacroix)]

Too bad. It happened again

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x7e) [0x80c744e]
1: [0xffffe420]
2: /usr/lib/dri/r300_dri.so [0xaf79d8ec]
3: /usr/lib/dri/r300_dri.so(_mesa_BindTexture+0x33f) [0xaf80ac7f]
4: /usr/lib/xorg/modules/extensions//libglx.so [0xb7bc758f]
5: /usr/lib/xorg/modules/extensions//libglx.so(DoRender+0xdd) [0xb7bbea2d]
6: /usr/lib/xorg/modules/extensions//libglx.so [0xb7bbeb4c]
7: /usr/lib/xorg/modules/extensions//libglx.so [0xb7bc30a6]
8: /usr/bin/X [0x814feee]
9: /usr/bin/X(Dispatch+0x2cf) [0x808d93f]
10: /usr/bin/X(main+0x48b) [0x80747ab]
11: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb7d4f050]
12: /usr/bin/X(FontFileCompleteXLFD+0x20d) [0x8073b21]

Fatal server error:
Caught signal 11. Server aborting

My guess is that r300 have some problems with compositing : The first crash happened while using compiz, and the second one after using kwin4.

How can I help to get more debug info ?

[mneagul (mneagul)]

I can confirm this bug. I'm attaching the log file coresponding to the crash. Currently I'm not able to reproduce the crash. :(

I don't know if this is a r300 issue...

[ogc (hackrez)]

sorry, above is unrelated

[netslayer (netslayer007)]

please see my possible duplicate report, may be helpful in repro: https://bugs.launchpad.net/ubuntu/+bug/221162

[netslayer (netslayer007)]

ogc, we both have (II) AIGLX: Suspending AIGLX clients for VT switch in our xorg log output at the end.. interesting?
attached mine

[Սահակ (petrosyan)]

I am also having frequent X server crashes on a freshly installed Ubuntu 8.04
Backtrace:
0: /usr/bin/X(xf86SigHandler+0x6a) [0x48402a]
1: /lib/libc.so.6 [0x7f825cda2100]
2: /usr/lib/xorg/modules//libxaa.so [0x7f825a311c9c]
3: /usr/lib/xorg/modules//libxaa.so [0x7f825a311e96]
4: /usr/bin/X [0x527e42]
5: /usr/bin/X [0x515c1f]
6: /usr/bin/X(Dispatch+0x2ef) [0x44eaaf]
7: /usr/bin/X(main+0x47d) [0x436b9d]
8: /lib/libc.so.6(__libc_start_main+0xf4) [0x7f825cd8e1c4]
9: /usr/bin/X(FontFileCompleteXLFD+0x279) [0x435ed9]

[Michael (mijobe)]

Same for me. I have a Dell Inspiron 9400 with Intel 945GM.

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x7e) [0x80c780e]
1: [0xb7fa6420]
2: /usr/bin/X [0x80da805]
3: /usr/lib/xorg/modules/extensions//libextmod.so(XvdiPutImage+0x178) [0xb7c0c608]
4: /usr/lib/xorg/modules/extensions//libextmod.so [0xb7c0f476]
5: /usr/bin/X [0x81506de]
6: /usr/bin/X(Dispatch+0x2cf) [0x808d8df]
7: /usr/bin/X(main+0x48b) [0x807471b]
8: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb7d35450]
9: /usr/bin/X(FontFileCompleteXLFD+0x201) [0x8073a91]

Fatal server error:
Caught signal 11. Server aborting

(II) AIGLX: Suspending AIGLX clients for VT switch
(II) intel(0): fbc disabled on plane a
(II) intel(0): xf86UnbindGARTMemory: unbind key 0
(II) intel(0): xf86UnbindGARTMemory: unbind key 1
(II) intel(0): xf86UnbindGARTMemory: unbind key 2
(II) intel(0): xf86UnbindGARTMemory: unbind key 3
(II) intel(0): xf86UnbindGARTMemory: unbind key 4

This is reproducible by viewing a video in full screen mode and than switching into another virtual desktop. If you switch several times between the video and some other desktop you will get the problem. It seems that, if the problem ocurred once, it will occur on each switch untill you reboot.

[Michael (mijobe)]

Some additional info: I am looking TV via DVB-T using kaffeine. It seems that xine is somewhat more stable. This does not mean that the crash does not occur, but that it occurs later. If I use kaffeine within a window and switch it from background to foreground and so on, I run into the same problem.

[johas (johas)]

I am experiencing X crashes on kubuntu 8.04 when I get a kopete message and click on the notification bubble to open the chat window.

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x7e) [0x80c780e]
1: [0xb7ee8420]

Fatal server error:
Caught signal 11. Server aborting

This is not really reproducable but occurs about every 3rd or 4th time I click that "open"-button. I usually have Compiz enabled, but I think it also occured with kwin enabled instead of compiz.

[johas (johas)]

Switching to fullscreen mode (e.g. in gwenview) produces the same crash from time to times.

[km (km-mathcs)]

I get exactly the same backtrace as Michael

2: /usr/bin/X [0x80da805]
3: /usr/lib/xorg/modules/extensions//libextmod.so(XvdiPutImage+0x178) [0xb7bb3608]
4: /usr/lib/xorg/modules/extensions//libextmod.so [0xb7bb6476]
5: /usr/bin/X [0x81506ee]
6: /usr/bin/X(Dispatch+0x2cf) [0x808d8df]
7: /usr/bin/X(main+0x48b) [0x807471b]
8: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb7cdc450]
9: /usr/bin/X(FontFileCompleteXLFD+0x201) [0x8073a91]

except I'm on an Intel 915GM (on a Fujistu 7000 laptop).

I can reproduce it consistently by playing a standard def video file after a high def one. It doesn't matter which of the video players I use (xine, vlc, myth, mplayer).

Are we really in the right bug for this? The initial report is about a Mesa/GLX issue on radeon and this is about a Xvid problem on Intel.

[noamik (spam-noamik)]

I got the same problem on a desktop with GeForce 6800 with XFCE4 and compiz/fusion on Hardy since I upgraded. It happens when starting a video for the first time (VLC, mplayer, ...) or when coming back from fullscreen dvb-t with kaffeine.

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x7e) [0x80c780e]
1: [0xb7fb6420]
2: /usr/bin/X(ProcPolyPoint+0x14a) [0x808aa9a]
3: /usr/bin/X [0x815076e]
4: /usr/bin/X(Dispatch+0x2cf) [0x808d8df]
5: /usr/bin/X(main+0x48b) [0x807471b]
6: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb7d45450]
7: /usr/bin/X(FontFileCompleteXLFD+0x201) [0x8073a91]

Fatal server error:
Caught signal 11. Server aborting

I have the same problem on a fresh install on my IBM x41 with Intel 915 GM ...

[avenger (bole-chen)]

got it on my Lenovo T60 with ati x1300
Linux avenger-ubuntu 2.6.24-19-generic #1 SMP Wed Jun 18 14:43:41 UTC 2008 i686 GNU/Linux

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x7e) [0x80c780e]
1: [0xb7f0e420]
2: /usr/lib/xorg/modules//libxaa.so [0xb75ad6dd]
3: /usr/bin/X [0x8172e24]
4: /usr/bin/X(CompositeGlyphs+0x9a) [0x815a36a]
5: /usr/bin/X [0x8161d29]
6: /usr/bin/X [0x815d085]
7: /usr/bin/X [0x81506ee]
8: /usr/bin/X(Dispatch+0x2cf) [0x808d8df]
9: /usr/bin/X(main+0x48b) [0x807471b]
10: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb7ca3450]
11: /usr/bin/X(FontFileCompleteXLFD+0x201) [0x8073a91]

Fatal server error:
Caught signal 11. Server aborting

(II) AIGLX: Suspending AIGLX clients for VT switch

[avenger (bole-chen)]

Yestoday, i apt-get remove kvpnc, until now, the crash not retun……

[km (km-mathcs)]

>Yestoday, i apt-get remove kvpnc, until now, the crash not retun……

I've never had kvpnc installed. I wonder if we are looking at more than one issue here. Not every segfault has the same cause.

In my case I can reproduce it whenever I play a low res video after a high res video, and the backtrace shows XvdiPutImage at the top (as did Michael's).

[avenger (bole-chen)]

Bad news.

I have crash again,a while ago, when i start use gvim and svn client.

So, i reboot my laptop and go here.

Seems that crash it not regularity.

[Matteo (telegraph-road)]

I have a crash too using only mplayer with xv video output. I have an Intel 945GM on a ThinkPad T60

This is my Backtrace:
0: /usr/bin/X(xf86SigHandler+0x7e) [0x80c780e]
1: [0xb7ef0420]
2: /usr/bin/X [0x80da805]
3: /usr/lib/xorg/modules/extensions//libextmod.so(XvdiPutImage+0x178) [0xb7b5c608]
4: /usr/lib/xorg/modules/extensions//libextmod.so [0xb7b5f476]
5: /usr/bin/X [0x81506ee]
6: /usr/bin/X(Dispatch+0x2cf) [0x808d8df]
7: /usr/bin/X(main+0x48b) [0x807471b]
8: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb7c85450]
9: /usr/bin/X(FontFileCompleteXLFD+0x201) [0x8073a91]

[Moberg (moberg-komogvind)]

I have the same crash and backtrace, it has happened several times during switch between virtual desktops.

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x7e) [0x80c780e]
1: [0xb7f98420]
2: /usr/lib/xorg/modules//libxaa.so [0xb763a5b9]
3: /usr/bin/X [0x8173013]
4: /usr/bin/X(CompositePicture+0x150) [0x815a1f0]
5: /usr/bin/X [0x81601df]
6: /usr/bin/X [0x815d085]
7: /usr/bin/X [0x81506ee]
8: /usr/bin/X(Dispatch+0x2cf) [0x808d8df]
9: /usr/bin/X(main+0x48b) [0x807471b]
10: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb7d2c450]
11: /usr/bin/X(FontFileCompleteXLFD+0x201) [0x8073a91]

Fatal server error:
Caught signal 11. Server aborting

(II) AIGLX: Suspending AIGLX clients for VT switch

[km (km-mathcs)]

Looking at the backtraces I'm still not convinced we are all talking about the same bug. Mine manifests reproducibly playing a low def video after a high def video (but not the reverse).

I'm going to attempt to attach two samples low.mpg and high.mpg. These are just a few seconds long and 2.5MB each. Please play low first then high then low on a fresh ubuntu session. If you are seeing the same segment fault as me the first two will be ok, and the third will cause the segfault. It doesn't matter which player you use, all of mplayer, xine, vlc, myth do the same thing for me.

If you do this please post your hardware config and whether you got the segfault.

I'm unclear about the attachment limits. I don't know if 2.5MB is too large, and it looks like I will have to do a separate post to get the second attachment The HD one is going with this post.

[km (km-mathcs)]

This is to attach low.mpg as per last post

[Bryce Harrington (bryce)]

This bug was generically titled "Xorg crash", and a number of people have commented that "I too have this bug", however in looking at each person's backtrace, clearly each of these crashes is a completely different, unrelated bug. Unless your backtrace matches exactly to Olivier's (which mentions _mesa_update_texture()), you need to assume you're having an unrelated issue, and file a NEW bug report.

Also, when you encounter an X crash, please follow the directions on http://wiki.ubuntu.com/X/Backtracing to obtain a full backtrace of the issue. The standard trace included in your Xorg.0.log is not enough to begin troubleshooting. Olivier, can you please confirm you're still having the crash, and collect a full backtrace?

Everyone else - please file new bugs unless your backtrace matches Olivier's.

[km (km-mathcs)]

There are three of us that posted backtraces that are not related to the mesa bug, and are in fact a problem with the Intel Video driver. This is the bug that I can reproduce with the samples I posted above. As suggested we need to move out of this Mesa bug.

I found that there was in fact an existing bug 226828 with the Intel XvdiPutImage backtrace. I have added onto that one.

Incidentally, as I report at 226828, this problem goes away when I built and replaced intel_drv.,so with a later version (2.3.2).

[Olivier (olivier-lacroix)]

Thanks Bryce for your attention. This crash was kind of random, so I have no backtrace so far. I'll try to get one but this may take a while.

[robeph (robeph)]

I am still having this issue. While my crash log from X is a bit different in the lead up, the symptoms are all otherwise the same. I have noted that this crash occurs during input on many occasions, namely middle clicking. Of the many times it has crashed the majority of the crashes have occurred as I was middle clicking to either open or close a tab in firefox.

When it crashes it drops to a console (akin to hitting ctrl-alt f8 while in X) And often even though you're in a console you can still see a graphical mouse pointer (which will no longer move and remains as you switch from console to console alt-F(n) )

I hope this is the right place to post this as this bug seems to be marked inactive.

I'm using an ati hd3870 (well two of them, in crossfire, but I don't think the drivers support that, but it may have some effect in this, I am not sure.)