GL screensavers crash the X server in _mesa_update_state_locked

Bug #153986 reported by Dragonmaster Lou
4
Affects Status Importance Assigned to Milestone
mesa (Debian)
Fix Released
Unknown
mesa (Ubuntu)
Invalid
Undecided
Unassigned

Bug Description

Binary package hint: xorg

Summary pretty much describes it. I'm currently using the open source ATI/Radeon xserver that comes with x.org.

My hardware is an IBM ThinkPad T42 with a Mobility Radeon 9600. I'm using an xrandr dual head setup with the built in screen and an external monitor.

Here is an excert of my X.org log:

Backtrace:
0: /usr/bin/X(xf86SigHandler+0x81) [0x80c9581]
1: [0xffffe420]
2: /usr/lib/xorg/modules/extensions//libGLcore.so(_mesa_update_state_locked+0x7e1) [0xb376af91]
3: /usr/lib/xorg/modules/extensions//libGLcore.so(_mesa_update_state+0x2a) [0xb376b0ea]
4: /usr/lib/xorg/modules/extensions//libGLcore.so(_mesa_Clear+0x17e) [0xb36fddbe]
5: /usr/lib/xorg/modules/extensions//libglx.so [0xb7bb9a08]
6: /usr/lib/xorg/modules/extensions//libglx.so(DoRender+0x155) [0xb7bb2625]
7: /usr/lib/xorg/modules/extensions//libglx.so [0xb7bb26bc]
8: /usr/lib/xorg/modules/extensions//libglx.so [0xb7bb6b2c]
9: /usr/bin/X [0x815754e]
10: /usr/bin/X(Dispatch+0x1aa) [0x808f47a]
11: /usr/bin/X(main+0x495) [0x8076f05]
12: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe0) [0xb7cf9050]
13: /usr/bin/X(FontFileCompleteXLFD+0x1e1) [0x8076241]

Fatal server error:
Caught signal 11. Server aborting

Revision history for this message
Dragonmaster Lou (lou-techhouse) wrote :

Here is the entire Xorg log in case it helps.

Revision history for this message
Tormod Volden (tormodvolden) wrote :

If you've been using fglrx before, it might be the reason. Try:
 dpkg -l '*fglrx*'
and
 locate fglrx
to see if there is still some proprietary bits around causing problems.

Changed in xorg:
status: New → Incomplete
Revision history for this message
Dragonmaster Lou (lou-techhouse) wrote :

I had been using fglrx before, but because it's broken with regards to suspend/resume, I switched back to the open source driver.

So apparently I do have some fglrx bits floating around, but I'm not sure how to safely remove them without also removing my madwifi wireless driver. I did a "complete removal" from Synaptic of xorg-driver-fglrx, but the modules are still around. lsmod shows that they aren't loaded, however, and this is the output from dpkg after I thought I removed the last bits in Synaptic:

$ dpkg -l '*fglrx*'
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Installed/Config-f/Unpacked/Failed-cfg/Half-inst/t-aWait/T-pend
|/ Err?=(none)/Hold/Reinst-required/X=both-problems (Status,Err: uppercase=bad)
||/ Name Version Description
+++-==============-==============-============================================
pn fglrx-control <none> (no description available)
un fglrx-driver <none> (no description available)
un xfree86-driver <none> (no description available)
pn xorg-driver-fg <none> (no description available)

I also reinstalled all the mesa DRI packages just in case. X still crashes when using a GL screen saver.

Revision history for this message
Tormod Volden (tormodvolden) wrote :

Could you please try running from a Ubuntu 7.10 live cd, so that we can rule out "proprietary contamination"?

Changed in xorg:
status: Unknown → New
Revision history for this message
Dragonmaster Lou (lou-techhouse) wrote :

Sorry for the delay.

The bug also appears when running off a live CD, so there is no "proprietary contamination" here.

Changed in xorg:
status: Incomplete → Confirmed
Revision history for this message
Tormod Volden (tormodvolden) wrote :

I got a pretty similar backtrace now, using radeonhd on a rv515 card.

Revision history for this message
Bryce Harrington (bryce) wrote :

I notice it's been a long time since this was reported, with no recent confirmations so am guessing people are not still experiencing the issue. The upstream bug also appears DOA so no luck there. I do see that several fixes have gone into the source code file that _mesa_update_state_locked comes from, so perhaps the issue was fixed upstream.

In any case, the Xorg.0.log backtrace doesn't give enough details to analyze the crash any further, so without a more detailed backtrace available I'm going to have to close it at this point. Feel free to reopen with a full backtrace if the issue can still be reproduced against current Karmic.

Changed in mesa (Ubuntu):
status: Confirmed → Invalid
Changed in mesa (Debian):
status: New → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.