Using the FC3t2 SMP kernel on a HP xw4100 workstation with a P4 dual
hyper-threaded CPU the X server will cause the machine to lock up.
To reproduce, boot the smp kernel, at the graphical login prompt log
in, within a few seconds as the Gnome desktop is coming up the system
I have not been able to ascertain if there was a machine check or if
the kernel is deadlocked somehow. Alternate consoles do not work,
alt-sysrq to break into the kernel does not work and the machine does
not respond to pings.
The non-smp kernel works fine. If the X server is not started on the
smp kernel hangs do not occur.
The graphics card is a Radeon 7000 using the standard Xorg open source
ATI radeon driver. Attached is the output of /proc/cpu/info and the X
log from a non-smp session.
This system is located in my cube in the Westord office.
Created attachment 104274 [details]
Xorg log file
This is the X server log file from a non-smp invocation. Its the smp kernel
that is causing the problem but the two log files should be identical as I do
not believe there is anything in the X server which is dependent on the number
of cpu's or smp.
Created attachment 104275 [details]
output of /proc/cpuinfo
The cpuinfo is also from a non-smp kernel so it only shows one cpu. If
its important I can get both log files from a smp run.
(II) Initializing built-in extension COMPOSITE
If you disable COMPOSITE does the problem still occur? I assume
so, but just curious. ;)
Probably a duplicate of bug #138108.
This is most likely a dupe of bug #138108, for which I believe we now
have a fix. xorg-x11-6.8.1-17 or later in rawhide should make the
crashes go away, and we'll be putting out an update shortly for FC3.
Please test either the udpate or the rawhide packages so we can
confirm that this is indeed the same issue. Thanks.
*** This bug has been marked as a duplicate of 138108 ***
Changed to 'CLOSED' state since 'RESOLVED' has been deprecated.