Bug 698237 (xABI1_11) - X is segfaulting at startup in Rawhide (intel, ati, nouveau...)
Summary: X is segfaulting at startup in Rawhide (intel, ati, nouveau...)
Keywords:
Status: CLOSED RAWHIDE
Alias: xABI1_11
Product: Fedora
Classification: Fedora
Component: xorg-x11-server
Version: rawhide
Hardware: Unspecified
OS: Unspecified
unspecified
high
Target Milestone: ---
Assignee: Peter Hutterer
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard: [cat:crash]
: 698227 698334 699177 699277 701623 702182 702956 (view as bug list)
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2011-04-20 13:54 UTC by Daniel Walsh
Modified: 2018-04-11 08:19 UTC (History)
15 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2011-05-16 00:27:28 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)
log file (20.41 KB, text/plain)
2011-04-20 13:54 UTC, Daniel Walsh
no flags Details

Description Daniel Walsh 2011-04-20 13:54:19 UTC
Created attachment 493496 [details]
log file

Description of problem:

X Crashes.

Comment 1 Matěj Cepl 2011-04-21 16:01:12 UTC
Lovely backtrace, indeed!

Backtrace:
[   916.713] 0: X (xorg_backtrace+0x2f) [0x45e17f]
[   916.713] 1: X (0x400000+0x62b96) [0x462b96]
[   916.714] 2: /lib64/libpthread.so.0 (0x3ed0c00000+0xf3c0) [0x3ed0c0f3c0]
[   916.714] 3: /usr/lib64/xorg/modules/drivers/intel_drv.so (0x7f22308b7000+0xeab9) [0x7f22308c5ab9]
[   916.714] 4: X (AddScreen+0x17f) [0x42ee1f]
[   916.714] 5: X (InitOutput+0x282) [0x471ca2]
[   916.714] 6: X (0x400000+0x22a39) [0x422a39]
[   916.714] 7: /lib64/libc.so.6 (__libc_start_main+0xed) [0x3ed082143d]
[   916.714] 8: X (0x400000+0x22ee1) [0x422ee1]
[   916.715] Segmentation fault at address 0xffffffffffecca4c
[   916.715] 
Fatal server error:
[   916.715] Caught signal 11 (Segmentation fault). Server aborting
[   916.715] 

Could we have versions of xorg-x11-server and xorg-x11-drv-intel, please?

Thank you

Comment 2 Daniel Walsh 2011-04-21 16:03:39 UTC
If you back down to xorg-x11-server-Xorg-1.10.1-9.fc15.x86_64 it works.

Comment 3 Daniel Walsh 2011-04-21 16:05:38 UTC
With these versions:

xorg-x11-drv-intel-2.15.0-2.fc16.x86_64
xorg-x11-server-1.10.99.1-3.20110418.fc16

It blows up.

Comment 4 Matěj Cepl 2011-04-22 11:13:12 UTC
(In reply to comment #2)
> If you back down to xorg-x11-server-Xorg-1.10.1-9.fc15.x86_64 it works.

Yes, you're right and from the fedora-devel post by ajax (http://article.gmane.org/gmane.linux.redhat.fedora.devel/147699) it looks like a known bug. We are sorry.

Comment 5 Matěj Cepl 2011-04-22 11:15:52 UTC
*** Bug 698227 has been marked as a duplicate of this bug. ***

Comment 6 Matěj Cepl 2011-04-22 11:22:40 UTC
*** Bug 698334 has been marked as a duplicate of this bug. ***

Comment 7 Matěj Cepl 2011-04-24 21:23:56 UTC
*** Bug 699177 has been marked as a duplicate of this bug. ***

Comment 8 Matěj Cepl 2011-04-27 05:29:09 UTC
*** Bug 699277 has been marked as a duplicate of this bug. ***

Comment 9 Matěj Cepl 2011-05-06 12:16:57 UTC
*** Bug 702182 has been marked as a duplicate of this bug. ***

Comment 10 Matěj Cepl 2011-05-06 12:43:16 UTC
*** Bug 701623 has been marked as a duplicate of this bug. ***

Comment 11 Matthew Miller 2011-05-06 13:08:51 UTC
(In reply to comment #4)
> (In reply to comment #2)
> > If you back down to xorg-x11-server-Xorg-1.10.1-9.fc15.x86_64 it works.
> 
> Yes, you're right and from the fedora-devel post by ajax
> (http://article.gmane.org/gmane.linux.redhat.fedora.devel/147699) it looks like
> a known bug. We are sorry.

That thread just says "Rawhide's X is going to be like that for a while."

Comment 12 Shawn 2011-05-06 17:52:16 UTC
I have run into this with both my NVidia and AMD drivers and getting back down to a previous version of X is a manual process. Shouldn't there be dependencies set within the RPMs to indicate a specific version of X is required and not allow you to drive off the cliff if that version isn't available? Isn't that why we use RPM and Yum?

Comment 13 Peter Hutterer 2011-05-10 04:54:15 UTC
The main drivers are rebuilt, this should hopefully fix the issues.

Intel build: http://koji.fedoraproject.org/koji/taskinfo?taskID=3062113
ATI build: http://koji.fedoraproject.org/koji/taskinfo?taskID=3062122
Nouveau build: http://koji.fedoraproject.org/koji/taskinfo?taskID=3062143

I'll rebuild the other less-used drivers too, but please let us know if this fixes the issue.

Comment 14 Matěj Cepl 2011-05-10 14:17:07 UTC
*** Bug 702956 has been marked as a duplicate of this bug. ***

Comment 15 Shawn 2011-05-11 03:57:14 UTC
I have tried it on my NVidia powered laptop with the nouveau driver and it works nicely. Thank you. I can't test my AMD powered machine until tomorrow but will post the results when I do.

Comment 16 Yanko Kaneti 2011-05-11 05:46:24 UTC
Please redo the nvr to be > F15 versions
xorg-x11-drv-ati-6.14.0-9.20110316gitcdfc007ec.fc16 
xorg-x11-drv-ati-6.14.1-1.20110504gita6d2dba6.fc15

Comment 17 Andy Clark 2011-05-11 06:02:25 UTC
With these packages:
[andy@timelock ~]$ rpm -qa | egrep "(server-common|intel)"
xorg-x11-server-common-1.10.99.1-5.20110510.fc16.i686
xorg-x11-drv-intel-2.15.0-4.fc16.i686
[andy@timelock ~]$
My X Window system works fine. Thank you.

[andy@timelock ~]$ dmesg | grep -i chipset
[    1.478614] agpgart-intel 0000:00:00.0: Intel G41 Chipset
[andy@timelock ~]$

Comment 18 Stijn Hoop 2011-05-13 14:38:55 UTC
Solved for me as well,

[stijn@pounce] <~/Documents/Fedora/Bugs/698237> rpm -qa | egrep "(server-common|intel)"
xorg-x11-drv-intel-2.15.0-4.fc16.x86_64
xorg-x11-server-common-1.10.99.1-5.20110510.fc16.x86_64
[stijn@pounce] <~/Documents/Fedora/Bugs/698237> sudo grep -i chipset /var/log/messages 
May  1 15:01:49 pounce kernel: [    0.976223] agpgart-intel 0000:00:00.0: Intel HD Graphics Chipset

Comment 19 Peter Hutterer 2011-05-16 00:27:28 UTC
All drivers are rebuilt now and should have found their way into the rawhide compose. Reports so far are positive, so I'm closing this bug. Thanks for testing, sorry about this bug, it shouldn't have happened.


Note You need to log in before you can comment on or make changes to this bug.