Re: Questions about GPLPV stability tests

[Andreas Kinzler <ml-xen-devel@hfp.de>]

>> The patch actually only disables HPET broadcast which has some downsides
>> because it effectively disables C3 states. Without C3 states Nehalem and
>> later CPUs cannot enter turbo-mode. So there is some loss of performance.
> Would disabling any low CPU power states and turbo clocks in the BIOS,
> help as well? Just curious. I have seen other 'weird' performance

I would leave the BIOS to its default and I'd guess that playing with 
BIOS settings only makes it worse.

> issues between machines using the same hardware. Some CPU intensive
> algorithm could be twice as slow running on Dom0 compared to the same
> kernel without Xen. On other identical systems I didn't see that
> issue. I didn't have time to investigate, but I felt there may have
> been BIOS setting differences.

Xen does not use the performance governor by default which has a notable 
performance impact in my tests. My Xen boot config is:

kernel /xen.gz dom0_mem=1024m dom0_max_vcpus=2 cpufreq=xen:performance
module /linux-2.6.32.36-pvops0-ak2 root=/dev/sda5

Regards Andreas