Description of problem: X hangs with 100% cpu usage during logout. The keyboard is dead (Ctrl + Alt + F1 or + backspaece doesn't work). The mouse cursor works but otherwise the screen is blank. Loging via ssh and killing X (kill -9) solves the problem. Version-Release number of selected component (if applicable): xorg-x11-drv-sis-0.9.1-7 How reproducible: Login and Logout. Approximately 1 in 10 times X halts.
Created attachment 144614 [details] xorg log
I have booted the system with acpi=off, but it didn't solve the problem
Thanks for the bug report. We have reviewed the information you have provided above, and there is some additional information we require that will be helpful in our diagnosis of this issue. Please attach your X server config file (/etc/X11/xorg.conf) as well to the bug report as individual uncompressed file attachments using the bugzilla file attachment link below. Could you please also try to run without any /etc/X11/xorg.conf whatsoever and let X11 autodetect your display and video card? Attach to this bug /var/log/Xorg.0.log from this attempt as well, please. We will review this issue again once you've had a chance to attach this information. Thanks in advance.
I will attach the requested files in a minute. I have upgraded the sis driver to version 0.9.3 by compiling and installing the package from fedora devel. I\ have also installed the dubug symbols. The trace printed by gdb is: #0 0x0035e402 in __kernel_vsyscall () #1 0x4d85996e in __lll_mutex_lock_wait () from /lib/libc.so.6 #2 0x4d7ea01e in _L_lock_14512 () from /lib/libc.so.6 #3 0x4d7e9544 in free () from /lib/libc.so.6 #4 0x0079770d in _mesa_free () from /usr/lib/xorg/modules/extensions/libGLcore.so #5 0x009242bd in XMesaDestroyVisual () from /usr/lib/xorg/modules/extensions/libGLcore.so #6 0x00937682 in GlxGetMesaProvider () from /usr/lib/xorg/modules/extensions/libGLcore.so #7 0x0051a997 in __glXResetScreens () from /usr/lib/xorg/modules/extensions/libglx.so #8 0x00519bdc in GlxExtensionInit () from /usr/lib/xorg/modules/extensions/libglx.so #9 0x08095676 in CloseDownExtensions () #10 0x0806fa7f in main () I will install the missing symbols and report again when it halts.
Created attachment 144659 [details] xorg.conf
Created attachment 144660 [details] new Xorg.0.log (with xorg.conf, without acpi)
new back trace: #0 0x008f1400 in __kernel_vsyscall () #1 0x4d85996e in __lll_mutex_lock_wait () from /lib/libc.so.6 #2 0x4d7ea01e in _L_lock_14512 () from /lib/libc.so.6 #3 0x4d7e9544 in free () from /lib/libc.so.6 #4 0x00fa270d in _mesa_free (ptr=0x18) at imports.c:116 #5 0x0112f2bd in XMesaDestroyVisual (v=0x18) at xm_api.c:1646 #6 0x01142682 in __glXMesaScreenDestroy (screen=0x18) at xf86glx.c:299 #7 0x00aaa997 in __glXResetScreens () at glxscreens.c:321 #8 0x00aa9bdc in ResetExtension (extEntry=0xa16f778) at glxext.c:49 #9 0x08095676 in CloseDownExtensions () at extension.c:280 #10 0x0806fa7f in main (argc=8, argv=0xbfa52ea4, envp=0xf000e2c3) at main.c:453 Now it looks like a libc bug. The strange thing is that there is only one thread running. What mutex is free waiting for?
I'm also encountering this bug. Concerning the previous stacktrace, the bug seems to be in __glXResetScreens that call __glXResetScreens with aninvalid pointer (0x18) that is later passed to free ...
Should be fixed in 1.1.1-47.6 and later. Please update, and reopen if this is still an issue for you.
were can I found this version? core/development/i386/os/Fedora/ only shows 0.9.3-2
I was looking at the wrong package. The updated one is xorg-x11-server-Xorg. Sorry about that :-(