Bug 138 - X segfaults on ATI RC410 [Radeon Xpress 200] on startup (in driCreateNewScreen ()) when KMS is disabled
Summary: X segfaults on ATI RC410 [Radeon Xpress 200] on startup (in driCreateNewScree...
Status: RESOLVED FIXED
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: Cauldron
Hardware: x86_64 Linux
Priority: Normal major
Target Milestone: ---
Assignee: Mageia Bug Squad
QA Contact:
URL:
Whiteboard:
Keywords: NEEDINFO
: 382 (view as bug list)
Depends on:
Blocks:
 
Reported: 2011-02-20 19:08 CET by Reinout van Schouwen
Modified: 2011-03-27 14:53 CEST (History)
4 users (show)

See Also:
Source RPM: x11-driver-video-ati
CVE:
Status comment:


Attachments
Xorg log (47.22 KB, text/plain)
2011-02-20 19:09 CET, Reinout van Schouwen
Details
lspcidrake -v (2.56 KB, text/plain)
2011-02-20 19:11 CET, Reinout van Schouwen
Details
Xorg.0.log + Full Xorg gdb log with backtrace (38.78 KB, text/plain)
2011-02-21 00:55 CET, Reinout van Schouwen
Details
Backtrace of crash with 2.6.38.1 (3.88 KB, text/plain)
2011-03-16 15:48 CET, John Balcaen
Details

Description Reinout van Schouwen 2011-02-20 19:08:58 CET
Description of problem:
Upon booting Mageia, instead of the login screen, I get a console.

Version-Release number of selected component (if applicable):
6.14.0-2.mga1

How reproducible:
Always

Steps to Reproduce:
1. Boot Mageia
2.
3.

Graphics card is identified as:
01:05.0 VGA compatible controller: ATI Technologies Inc RC410 [Radeon Xpress 200]



Reproducible: 

Steps to Reproduce:
Comment 1 Reinout van Schouwen 2011-02-20 19:09:54 CET
Created attachment 17 [details]
Xorg log
Comment 2 D Morgan 2011-02-20 19:11:28 CET
From the bugreport : 

Backtrace:
[    24.185] 0: /usr/bin/Xorg (xorg_backtrace+0x28) [0x4a0b38]
[    24.185] 1: /usr/bin/Xorg (0x400000+0x61129) [0x461129]
[    24.185] 2: /lib64/libpthread.so.0 (0x7f25d7c47000+0xf460) [0x7f25d7c56460]
[    24.185] Segmentation fault at address (nil)
[    24.185] 
Fatal server error:
[    24.185] Caught signal 11 (Segmentation fault). Server aborting
[    24.185] 
[    24.185] 
Please consult the The X.Org Foundation support 
	 at http://bugs.mageia.org
 for help.

CC: (none) => dmorganec

Comment 3 Reinout van Schouwen 2011-02-20 19:11:44 CET
Created attachment 18 [details]
lspcidrake -v
Comment 4 Reinout van Schouwen 2011-02-21 00:55:23 CET
Created attachment 19 [details]
Xorg.0.log + Full Xorg gdb log with backtrace
Comment 5 Reinout van Schouwen 2011-02-21 01:05:22 CET
possible clue: http://lists.freedesktop.org/archives/mesa-dev/2011-January/004834.html
Sandro CAZZANIGA 2011-02-28 08:09:07 CET

CC: (none) => cazzaniga.sandro
Severity: critical => major

Thierry Vignaud 2011-03-16 14:56:40 CET

Attachment 19 description: Full Xorg gdb log with backtrace => Xorg.0.log

Thierry Vignaud 2011-03-16 14:57:24 CET

Attachment 19 description: Xorg.0.log => Xorg.0.log + Full Xorg gdb log with backtrace

Comment 6 Thierry Vignaud 2011-03-16 15:02:42 CET
Your trace looks like it has been edited (no commands, C line not printed, ...).

Can you do a "print driDriverAPI.InitScreen" from GDB?

Also, if you think this patch fixes it, can you try it?

BTW, Bug #383 is a duplicate.

Keywords: (none) => NEEDINFO
CC: (none) => thierry.vignaud
Summary: X crashes on startup => X segfaults on ATI RC410 [Radeon Xpress 200] on startup (in driCreateNewScreen ())

Thierry Vignaud 2011-03-16 15:04:43 CET

Summary: X segfaults on ATI RC410 [Radeon Xpress 200] on startup (in driCreateNewScreen ()) => X segfaults on ATI RC410 [Radeon Xpress 200] on startup (in driCreateNewScreen ()) when KMS is disabled

Comment 7 Thierry Vignaud 2011-03-16 15:04:49 CET
*** Bug 382 has been marked as a duplicate of this bug. ***

CC: (none) => balcaen.john

Comment 8 Thierry Vignaud 2011-03-16 15:05:51 CET
Reinout, you're not using KMS (which is prefered now, still the driver shouldn't segfault when using UMS).
Does it works once you enable KMS?
Comment 9 John Balcaen 2011-03-16 15:48:15 CET
Created attachment 125 [details]
Backtrace of crash with 2.6.38.1

I'll give a try to the patch linked & i'll report here.
Comment 10 John Balcaen 2011-03-16 18:36:20 CET
I'm not able to compile mesa with patch (or without) in a local iurt chroot

it's failing with

g++ -O2 -g -pipe  -Wp,-D_FORTIFY_SOURCE=2 -fstack-protector --param=ssp-buffer-size=4 -fomit-frame-pointer -march=i586 -mtune=generic -fasynchronous-unwind-tables -Wall -Wmissing-prototypes -std=c99 -ffast-math -fvisibility=hidden -fno-strict-aliasing  -fPIC  -DUSE_X86_ASM -DUSE_MMX_ASM -DUSE_3DNOW_ASM -DUSE_SSE_ASM -D_GNU_SOURCE -DPTHREADS -DHAVE_POSIX_MEMALIGN -DUSE_EXTERNAL_DXTN_LIB=1 -DIN_DRI_DRIVER -DGLX_DIRECT_RENDERING -DGLX_INDIRECT_RENDERING -DHAVE_ALIAS -I/usr/include/libdrm -I/usr/include/nouveau   -DFEATURE_GL=1 -o nouveau_vieux_dri.so.test ../../../../../src/mesa/drivers/dri/common/dri_test.o nouveau_vieux_dri.so.tmp   -ldrm   -lexpat -lm -lpthread -ldl -ltalloc   -ldrm_nouveau  
nouveau_vieux_dri.so.tmp: undefined reference to `BIND_RING'
nouveau_vieux_dri.so.tmp: undefined reference to `BEGIN_RING_NI'
nouveau_vieux_dri.so.tmp: undefined reference to `BEGIN_RING'
collect2: ld returned 1 exit status
gmake[6]: *** [nouveau_vieux_dri.so] Error 1
gmake[6]: Leaving directory `/home/mikala/rpm/BUILD/Mesa-7.10/src/mesa/drivers/dri/nouveau'
gmake[5]: *** [lib] Error 2
gmake[5]: Leaving directory `/home/mikala/rpm/BUILD/Mesa-7.10/src/mesa/drivers/dri/nouveau'
gmake[4]: *** [subdirs] Error 1
gmake[4]: Leaving directory `/home/mikala/rpm/BUILD/Mesa-7.10/src/mesa/drivers/dri'
gmake[3]: *** [default] Error 1
gmake[3]: Leaving directory `/home/mikala/rpm/BUILD/Mesa-7.10/src/mesa/drivers'
gmake[2]: *** [driver_subdirs] Error 2
gmake[2]: Leaving directory `/home/mikala/rpm/BUILD/Mesa-7.10/src/mesa'
make[1]: *** [subdirs] Error 1
make[1]: Leaving directory `/home/mikala/rpm/BUILD/Mesa-7.10/src'
make: *** [default] Error 1
error: Bad exit status from /home/mikala/rpm/tmp/rpm-tmp.bv13O1 (%build)
Comment 11 Reinout van Schouwen 2011-03-26 16:07:27 CET
(In reply to comment #8)
> Reinout, you're not using KMS (which is prefered now, still the driver
> shouldn't segfault when using UMS).
> Does it works once you enable KMS?

I don't know why the driver didn't select KMS (I didn't use UMS on purpose) but I'm happy to report that with updated Mageia packages, the problem does not occur any more.
Comment 12 Thierry Vignaud 2011-03-27 14:53:56 CEST
Closing then

Status: NEW => RESOLVED
Resolution: (none) => FIXED


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