Red Hat Bugzilla – Bug 55605
kernel 2.4.9-7 constantly outputs messages to syslog about clock timer
Last modified: 2007-04-18 12:37:59 EDT
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (compatible; MSIE 5.5; Linux)
Description of problem:
When running kernel 2.4.9-7 on my Asus A7A266 motherboard (ALi chipset),
the following messages gets printed by the kernel every couple minutes, and
is very annoying. There's nothing wrong with my hardware, and this should
not be printed.
Nov 2 15:05:24 discovery kernel: probable hardware bug: clock timer
configuration lost - probably a VIA686a motherboard.
Nov 2 15:05:24 discovery kernel: probable hardware bug: restoring chip
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1.Turn on system.
2.Look at /var/log/messages or dmesg.
Souldn't you have 2.4.9-13 (instead of -7)? Yes, -13 is the official update
release at ftp.redhat.com. -7 was never released, it was in rawhide, it has
some extra debugging checks. Close the bug.
-7 was also released as official update before -13 was.
Anyway I'll kill this message for the next kernel we'll release.
I didn't know that anyone else was experiencing this. I have the same MOBO,
and sought assistance from ASUS with no help forthcoming from them
(lousy service ASUS has).
My experience didn't show until after I upgraded to the -13 kernel, though.
It didn't show with the 2.4.7-10 or 2.4.9-7. (Actually, I don't know how I got the
2.4.9-7 kernel, but I've been running the Redhat Network Update each time
I get a kernel annoucement--It must have been pushed down with the updater.)
I haven't checked the logs since I installed the -13 kernel, but then, if
it's been removed, I won't see it anyway.
Is it really a kernel issue? Or is the MOBO the problem?
This message is removed in 2.4.9-21 at least.
It's basically a check for a bug in some chipsets that set the timer back to the
(windows) value of 18.2Hz. (Linux uses 100Hz).
The kernel tries to detect this but sometimes it detects it wrong; no big deal
since setting it to 100 if it already was 100 is harmless.