Bug 1011 - [ATI/radeon] OpenGL segfault with ATI closed-source drivers
Summary: [ATI/radeon] OpenGL segfault with ATI closed-source drivers
Status: RESOLVED NOTOURBUG
Alias: None
Product: xorg
Classification: Unclassified
Component: Server/DDX/Xorg (show other bugs)
Version: git
Hardware: x86 (IA32) Linux (All)
: high blocker
Assignee: Xorg Project Team
QA Contact:
URL:
Whiteboard:
Keywords:
: 1282 (view as bug list)
Depends on:
Blocks:
 
Reported: 2004-08-08 00:35 UTC by Stéphan Kochen
Modified: 2005-01-10 19:11 UTC (History)
4 users (show)

See Also:
i915 platform:
i915 features:


Attachments

Description Stéphan Kochen 2004-08-08 00:35:51 UTC
Trying to run any OpenGL related application (even glxinfo) in CVS HEAD
(2004-08-07) with the ATI binary drivers (3.11.1 for XFree86 4.3.0) immediatly
causes a segfault.

Backtrace:
--------------------------------------------------
#0  0xb7ea6a11 in kill () from /lib/libc.so.6
#1  0xb7ea6655 in raise () from /lib/libc.so.6
#2  0xb7ea7cb8 in abort () from /lib/libc.so.6
No symbol table info available.

#3  0x08070e5d in ddxGiveUp () at xf86Init.c:1214
        i = 1
#4  0x08070fa0 in AbortDDX () at xf86Init.c:1269
        i = 1
#5  0x0810ebd7 in AbortServer () at log.c:395
No locals.
#6  0x0810ef6e in FatalError (f=0x8218090 "Caught signal %d.  Server
aborting\n") at log.c:531
        args = 0xbfffecb8 "\v"
        beenhere = 1
#7  0x0808d60b in xf86SigHandler (signo=11) at xf86Events.c:1294
No locals.
#8  <signal handler called>

#9  0xb7d3fe96 in ?? ()
#10 0x08b21290 in ?? ()
#11 0x00000008 in ?? ()
#12 0xbfffefb8 in ?? ()
No symbol table info available.

#13 0x0810d9b9 in Xfree (ptr=0x8b21290) at utils.c:1373
No locals.

#14 0xb7d3ff19 in ?? ()
#15 0x08b21290 in ?? ()
#16 0x00000000 in ?? ()
#17 0x0000000b in ?? ()
#18 0x08b20b28 in ?? ()
#19 0xb7f8daa8 in ?? () from /lib/libc.so.6
#20 0x00000000 in ?? ()
#21 0xbffff458 in ?? ()
No symbol table info available.

#22 0x080d2abc in Dispatch () at dispatch.c:455
        clientReady = (int *) 0xb7f8daa8
        result = 145885992
        client = 0xb
        nready = 0
        icheck = (HWEventQueuePtr *) 0x8b21290
        start_tick = 0
--------------------------------------------------

programs/Xserver/dix/dispatch.c:455:
result = (* client->requestVector[MAJOROP])(client);

'client' looks like the culprit?

(ps. I'm not sure I selected the right component)
Comment 1 Kristian Høgsberg 2004-08-08 06:19:48 UTC
This could be caused by the changes in the driver ABI.  They were just reverted
in CVS, could you try to update and test it again?
Comment 2 Stéphan Kochen 2004-08-08 07:37:49 UTC
Unfortunatly not, it's still segfaulting with latest CVS.

Also, something 'bt full' apparently doesn't print, but just 'bt' does:
Previous frame identical to this frame (corrupt stack?)
Comment 3 Mike A. Harris 2004-08-14 03:20:42 UTC
I've CC'd Hui Yu at ATI for comment.
Comment 4 Chris Bruner 2004-09-16 23:47:51 UTC
Apparently compiling xorg-x11 with -mno-sse2 get's around the problem the 
problem. (Haven't tried it myself, just noticed in forums). 
Comment 5 Chris Bruner 2004-10-12 19:57:00 UTC
 
Compiling xorg-x11 with -mno-sse2 does NOT get around the problem. 
 
 
Comment 6 Chris Bruner 2004-10-12 20:35:38 UTC
Currently using ATI drivers 3.14.1, which work under xorg 6.7.  Under xorg 6.8 
any opengl segfaults. 
 
6.8 seems to have a lot of buggage. 
 
 
Comment 7 Daniel Stone 2005-01-04 07:35:40 UTC
Closing as NOTOURBUG.
Comment 8 Michel Dänzer 2005-01-11 14:11:49 UTC
*** Bug 1282 has been marked as a duplicate of this bug. ***


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.