FC4 - unexplained lockups, system will not start

[Don Levey]

I don't know if this is an FC4 problem, a video driver problem, or what.
Any info or suggestions would, of course, be welcomed.  First, the symptons:

After upgrading from FC2 -> FC4, I thought I'd give the Livna ATI drivers a
try.  Previously, I was geting about 115 FPS from glxgears; after the smooth
install I think I got a 10x boost (though I can't recall right now).
Pushing things to the limit, I wanted to try Tux Racer.  It had been
painfully slow beforehand.  After the new drivers, though, the display
seemed to flip to 640x480.  Restarting the machine caused a hang in the boot
cycle; right after "Enabling Swap Space".  Restarting yet again *usually*
allowed me to continue.  This went on the back burner until I had time to
deal with it.

This morning, a similar problem occurred.  I was browsing using Epiphany
(the default for my system at the moment) when everything locked up.  I've
got the machine in a KVM switch; I couldn't even switch to another machine
(it had locked the KVM).  A hard restart hung the machine in the same
place - twice.  Back on the front burner...

The /var/log/messages file shows nothing incriminating that I can find, from
that time period.  Indeed, the two hard restarts don't show at all.  Only
the third, which allowed a normal startup:

Jul  6 09:25:02 dauphin crond(pam_unix)[26753]: session closed for user root
Jul  6 09:30:02 dauphin crond(pam_unix)[26777]: session opened for user root
by (uid=0)
Jul  6 09:30:02 dauphin crond(pam_unix)[26778]: session opened for user root
by (uid=0)
Jul  6 09:30:02 dauphin crond(pam_unix)[26778]: session closed for user root
Jul  6 09:30:02 dauphin crond(pam_unix)[26777]: session closed for user root
Jul  6 09:35:02 dauphin crond(pam_unix)[26804]: session opened for user root
by (uid=0)
Jul  6 09:35:02 dauphin crond(pam_unix)[26804]: session closed for user root
Jul  6 11:52:01 dauphin syslogd 1.4.1: restart.
Jul  6 11:52:01 dauphin kernel: klogd 1.4.1, log source = /proc/kmsg
started.
Jul  6 11:52:01 dauphin kernel: Linux version 2.6.11-1.1369_FC4
(bhcompile at decompose.build.redhat.com) (gcc version 4.0.0 20050525 (Red Hat
4.0.0-9)) #1 Thu Jun 2 22:55:56 EDT 2005
Jul  6 11:52:01 dauphin kernel: BIOS-provided physical RAM map:
Jul  6 11:52:01 dauphin kernel:  BIOS-e820: 0000000000000000 -
000000000009fc00 (usable)
Jul  6 11:52:01 dauphin kernel:  BIOS-e820: 000000000009fc00 -
00000000000a0000 (reserved)
Jul  6 11:52:01 dauphin kernel:  BIOS-e820: 00000000000f0000 -
0000000000100000 (reserved)
Jul  6 11:52:01 dauphin kernel:  BIOS-e820: 0000000000100000 -
000000003fffc000 (usable)
Jul  6 11:52:01 dauphin kernel:  BIOS-e820: 000000003fffc000 -
000000003ffff000 (ACPI data)
Jul  6 11:52:01 dauphin kernel:  BIOS-e820: 000000003ffff000 -
0000000040000000 (ACPI NVS)
Jul  6 11:52:01 dauphin kernel:  BIOS-e820: 00000000fec00000 -
00000000fec01000 (reserved)
Jul  6 11:52:01 dauphin kernel:  BIOS-e820: 00000000fee00000 -
00000000fee01000 (reserved)
Jul  6 11:52:01 dauphin kernel:  BIOS-e820: 00000000ffff0000 -
0000000100000000 (reserved)


It was at about 9:35 when the lockup occurred, and 11:52 when the third
restart happened.  I'm a little disturbed that the first two restarts were
not recorded at all.

Does anyone have any thoughts on this?  I can post the Xorg.log, the Xorg
config files, or anything else that may have bearing on the situation.

TIA,
 -Don