[Devel] Re: rdmsr_on_cpu, wrmsr_on_cpu et al

[Dave Jones]

On Wed, Jan 17, 2007 at 04:26:13PM +0300, Alexey Dobriyan wrote:
 > Hi, Dave!
 > 
 > There was OpenVZ specific bug rendering some cpufreq drivers unusable
 > on SMP. In short, when cpufreq code thinks it confined itself to
 > needed cpu by means of set_cpus_allowed() to execute rdmsr, some
 > "virtual cpu" feature can migrate process to anywhere. This triggers
 > bugons and does wrong things in general.
 > 
 > This got fixed by introducing rdmsr_on_cpu and wrmsr_on_cpu executing
 > rdmsr and wrmsr on given physical cpu by means of smp_call_function_single().
 > 
 > So, understanding p4-clockmod is not the only driver that was affected,
 > is something like this acceptable for start?

This is probably useful outside of cpufreq too, so should probably
go somewhere in include/asm/processor.h for eg.
Looks ok on a quick glance, but I'd like to see if anyone on lkml
has any comments, especially the usual x86/x86-64 experts.

		Dave

-- 
http://www.codemonkey.org.uk