"losing ticks" message/error under moderate load

[Simon Morgan]

Using the latest kernel from Hoary (2.6.10-5) I get the following message/error
in dmesg output as soon as the system is under slightly higher load:

Losing some ticks... checking if CPU frequency changed.

[Simon Morgan (sjmorgan)]

warning: many lost ticks.
Your time source seems to be instable or some driver is hogging interupts
rip default_idle+0x24/0x30

[Chuck Short (zulcss)]

Please post dmesg. Thanks

chuck

[Ben Collins (ben-collins)]

No response, closing.

[Ralf Findeisen (ralf-findeisen)]

Created an attachment (id=4668)
dmesg

dmesg log for bug reopen issue

[Ralf Findeisen (ralf-findeisen)]

Can you please re-open this issue?

I got the same problem. Also the clock seems to be off by a few minutes per
hour. Problem was worse (more time offset, timing related keyboard issues too)
with "nv" driver (open source) for X11, went better with "nvidia" driver (closed
source). Error message from /var/log/messages:

Oct 17 19:24:39 tanja kernel: [ 1392.338993] warning: many lost ticks.
Oct 17 19:24:39 tanja kernel: [ 1392.338994] Your time source seems to be
instable or some driver is hogging interupts
Oct 17 19:24:39 tanja kernel: [ 1392.339011] rip acpi_processor_idle+0x12e/0x380
[processor]

[Matt Zimmerman (mdz)]

(In reply to comment #5)
> Can you please re-open this issue?
>
> I got the same problem. Also the clock seems to be off by a few minutes per
> hour. Problem was worse (more time offset, timing related keyboard issues too)
> with "nv" driver (open source) for X11, went better with "nvidia" driver (closed
> source). Error message from /var/log/messages:
>
> Oct 17 19:24:39 tanja kernel: [ 1392.338993] warning: many lost ticks.
> Oct 17 19:24:39 tanja kernel: [ 1392.338994] Your time source seems to be
> instable or some driver is hogging interupts
> Oct 17 19:24:39 tanja kernel: [ 1392.339011] rip acpi_processor_idle+0x12e/0x380
> [processor]
>

Because the original report was incomplete, and your messages don't match, it's
possible that this is a different bug. Please file a separate report instead.