FS#18512 - [nvidia-utils] libglx.so Segmentation fault
Attached to Project:
Arch Linux
Opened by Daniele Marinello (marinz) - Monday, 01 March 2010, 14:18 GMT
Last edited by Jan de Groot (JGC) - Sunday, 07 March 2010, 13:48 GMT
Opened by Daniele Marinello (marinz) - Monday, 01 March 2010, 14:18 GMT
Last edited by Jan de Groot (JGC) - Sunday, 07 March 2010, 13:48 GMT
|
Details
Description:
After last pacman -Syu, I've reboot and X doesn't start .... .... (II) Initializing extension GLX Backtrace: 0: /usr/bin/X (xorg_backtrace+0x28) [0x45a8d8] 1: /usr/bin/X (0x400000+0x61a19) [0x461a19] 2: /lib/libpthread.so.0 (0x7fb3ef762000+0xee80) [0x7fb3ef770e80] 3: /lib/libc.so.6 (_IO_vfprintf+0x409e) [0x7fb3eea34cae] 4: /lib/libc.so.6 (vsnprintf+0xa2) [0x7fb3eea5a572] 5: /usr/bin/X (LogVWrite+0x7f) [0x456b8f] 6: /usr/bin/X (LogVMessageVerb+0x64) [0x4571e4] 7: /usr/bin/X (xf86Msg+0x97) [0x4f3107] 8: /usr/lib/xorg/modules/extensions/libglx.so (0x7fb3ed1fd000+0x288b6f) [0x7fb3ed485b6f] Segmentation fault at address 0x1 Fatal server error: Caught signal 11 (Segmentation fault). Server aborting Here's the discussion: http://bbs.archlinux.org/viewtopic.php?id=92254 |
This task depends upon
Closed by Jan de Groot (JGC)
Sunday, 07 March 2010, 13:48 GMT
Reason for closing: Fixed
Additional comments about closing: Nvidia has been downgraded due to other bugs that could wreck hardware. Current version works fine.
Sunday, 07 March 2010, 13:48 GMT
Reason for closing: Fixed
Additional comments about closing: Nvidia has been downgraded due to other bugs that could wreck hardware. Current version works fine.
I have 2 nvidia 8500gt in sli, latest version of xorg from the repos and I'm using gnome.
if it helps:
$ lspci | grep VGA
02:00.0 VGA compatible controller: nVidia Corporation G84 [GeForce 8600 GT] (rev a1)
Reverting to 190.53 helps, then the X-Server starts without segfaulting.
Additinal info:
$ pacman -Q nvidia nvidia-utils
nvidia 190.53-4
nvidia-utils 190.53-1
$ pacman -Q kernel26
kernel26 2.6.32.9-1
$ ls -l /boot | grep -E '(kernel26|vmlinuz26)'
-rw-r--r-- 1 root root 8999631 Feb 24 19:37 kernel26-fallback.img
-rw-r--r-- 1 root root 3316199 Feb 24 19:36 kernel26.img
-rw-r--r-- 1 root root 3230329 Feb 17 01:35 kernel26.img.mdadm
-rw-r--r-- 1 root root 1972672 Feb 23 19:45 vmlinuz26
$ uname -a
Linux hadante 2.6.32-ARCH #1 SMP PREEMPT Tue Feb 23 19:43:46 CET 2010 x86_64 AMD Athlon(tm) 64 X2 Dual Core Processor 3800+ AuthenticAMD GNU/Linux
And creating a new config also doesn't help:
$ X -configure
$ X -config xorg.conf.new
also segfaults, so I guess it's not a config issue.
I did an strace -f X, I can provide it if it helps, but it doesn't really help. It only confirms that the segfault occurs on loading the glx extension:
2065 read(6, "# Registry of protocol names use"..., 4096) = 4096
2065 read(6, "reateContext\nR004 GLX:DestroyCon"..., 4096) = 4096
2065 read(6, "ANDR:GetCrtcGammaSize\nR023 RANDR"..., 4096) = 4096
2065 read(6, "X11:ChangeWindowAttributes\nR003 "..., 4096) = 4096
2065 read(6, "etDefinedIndices\nR010 X3D-PEX:Ge"..., 4096) = 4096
2065 read(6, "e\nR024 XFIXES:GetCursorName\nR025"..., 4096) = 4096
2065 read(6, "teImmediate\nR017 XIE:CreatePhoto"..., 4096) = 4096
2065 read(6, "ARD:SetNamedIndicator\nR017 XKEYB"..., 4096) = 2574
2065 read(6, "", 4096) = 0
2065 write(0, "(II) Initializing extension GLX\n", 32) = 32
2065 --- SIGSEGV (Segmentation fault) @ 0 (0) ---
2065 write(2, "\nBacktrace:\n", 12) = 12
2065 write(0, "\nBacktrace:\n", 12) = 12
then the backtrace follows, exactly the same as in the original bug report.
Let me know if you need more information.
nvidia 6800 nu, nvidia 8600gt, 9800gt or gts 250.
I am running KDE and have found this downgrade in [extra] to be a bit excessive.