Re: Biweekly KVM Test report, kernel 0c7771... userspace 1223a0...

[Avi Kivity <avi@redhat.com>]

Amit Shah wrote:
> On (Mon) Mar 30 2009 [10:07:58], Avi Kivity wrote:
>
>   
>>> 1. perfctr wrmsr warning when booting 64bit RHEl5.3
>>> https://sourceforge.net/tracker/?func=detail&aid=2721640&group_id=180599&atid=893831
>>>       
>> This is the architectural performance counting msr which was enabled in  
>> 4f76231 (KVM: x86: Ignore reads to EVNTSEL MSRs).  Amit, can you check  
>> if appropriate cpuid leaf 10 reporting will fix this?
>>     
>
> We already report 0s for the cpuid leaf 10; we need to report 0x3f in
> EBX for leaf 10 to denote events corresponding to the bits aren't
> available.
>
> I checked and it didn't help (we can't rely on guests to abide by cpuid
> flags)
>   

I see this in the code:

>         /*
>          * Check whether the Architectural PerfMon supports
>          * Unhalted Core Cycles Event or not.
>          * NOTE: Corresponding bit = 0 in ebx indicates event present.
>          */
>         cpuid(10, &(eax.full), &ebx, &unused, &unused);
>         if ((eax.split.mask_length < 
> (ARCH_PERFMON_UNHALTED_CORE_CYCLES_INDEX+1)) ||
>             (ebx & ARCH_PERFMON_UNHALTED_CORE_CYCLES_PRESENT))
>                 return 0;
>

So I think it can be done.

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