Re: kvm guest: hrtimer: interrupt too slow

[Avi Kivity <avi@redhat.com>]

On 09/29/2009 03:58 PM, Michael Tokarev wrote:
> Avi Kivity wrote:
>> On 09/29/2009 03:12 PM, Michael Tokarev wrote:
> []
>>> The thing is that after some uptime, kvm
>>> guest prints something like this:
>>>
>>> hrtimer: interrupt too slow, forcing clock min delta to 461487495 ns
> []
>> What happens if you use hpet or pmtimer as guest clocksource?
>
> For all the guests I have handy, only 2 clocksources
> are available: kvm-clock and acpi_pm.  The host itself
> has hpet turned off because it itself had issues with
> hpet, and after many tries we finally turned it off
> (there were several long and painy threads on lkml
> about this).
>
> Dunno why pmtimer isn't available.

acpi_pm is pmtimer.

>
> I tried switching to acpi_pm on a running guest (which
> is still runnining in very slow mode), but that did not
> make any difference - i.e., it did not become "fast"
> again.
>
>> Please post host /proc/cpuinfo.
>
> Here's the cpuinfo from host (for last core):
>
> flags        : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge 
> mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext 
> fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc rep_good 
> nonstop_tsc extd_apicid pni monitor cx16 popcnt lahf_lm cmp_legacy svm 
> extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs

Has both constant_tsc and nonstop_tsc, wonder why it's making so much 
trouble.


-- 
error compiling committee.c: too many arguments to function