Timekeeping hosed by factor 3, high lapic[01] interrupt rates
Jens Schweikhardt
schweikh at schweikhardt.net
Mon May 23 10:56:33 PDT 2005
...
# Lets try this:
# 0. If you're overclocking your CPU, don't.
Never did.
# 1. Boot with ACPI enabled and print the two kern.timecount sysctls above.
# I'm curious if its picking up the ACPI timecounter.
Isn't APIC enabled unless I disable it with hint.acpi.0.disabled="1"?
# 2. Shutdown and unplug the machine for about 20 minutes or overnight if
# convenient. Plug it back in, go into BIOS Setup and check the clock. If its
# off or dead then the CMOS battery is dead.
This is my home machine, which I turn off at night. The BIOS clock
looks good. And this would not explain why it's the same system
that works with a different kernel.
# 3. Backout rev 1.218 of src/sys/i386/isa/clock.c so the irq0 interrupt
# handler is reactivated and the RTC fiddled.
Will do so next. I've nailed the change between March 6 and March 30.
1.218 is from 2005/03/24 21:34:16, which would fit.
# > Some time in the past, the system would hang at boot with acpi enabled.
# > So I kept a hint.acpi.0.disabled="1" in /boot/device.hints. But even
# > without that hint, the time dilation effect (hey, it's the Einstein
# > Year!) is the same...
#
# This would imply the source of the problem is not in the timecounter,
# which doesn't make sense.
It's puzzling, but dammit, these are deterministic machines :-) I'm sure
in the end we'll find the cause. Thanks for your patience.
# Are you running ntpd?
Yes.
Regards,
Jens
--
Jens Schweikhardt http://www.schweikhardt.net/
SIGSIG -- signature too long (core dumped)
More information about the freebsd-current
mailing list