Re: [PATCH v6 03/19] x86/VPMU: Minor VPMU cleanup

[Boris Ostrovsky <boris.ostrovsky@oracle.com>]

On 05/19/2014 10:26 AM, Jan Beulich wrote:
>>>> On 13.05.14 at 17:53, <boris.ostrovsky@oracle.com> wrote:
>> Update macros that modify VPMU flags to allow changing multiple bits at once.
>>
>> Make sure that we only touch MSR bitmap on HVM guests (both VMX and SVM).
>> This is needed by subsequent PMU patches.
> This part is at least questionable - why would these bitmaps not
> similarly be used by PVH? And if so, this second change is kind of
> a policy one, while the first change is a purely mechanical one. I.e.


At this patch PVH VPMU won't work at all, bitmaps or not. It is enabled 
in patch 18 and there I replace is_hvm_domain() with 
has_hvm_container_domain(). I didn't want to do this here so that 
PVH-related changes are more explicit in 18.

> I don't think they fit well together into a single patch.

I can break this into two parts.

>
>> --- a/xen/arch/x86/hvm/vmx/vpmu_core2.c
>> +++ b/xen/arch/x86/hvm/vmx/vpmu_core2.c
>> @@ -326,16 +326,14 @@ static int core2_vpmu_save(struct vcpu *v)
>>   {
>>       struct vpmu_struct *vpmu = vcpu_vpmu(v);
>>   
>> -    if ( !vpmu_is_set(vpmu, VPMU_CONTEXT_SAVE) )
>> -        return 0;
>> -
>> -    if ( !vpmu_is_set(vpmu, VPMU_CONTEXT_LOADED) )
>> +    if ( !vpmu_is_set_all(vpmu, VPMU_CONTEXT_SAVE | VPMU_CONTEXT_LOADED) )
> Is this really a good name? How about vpmu_are_all_set() or
> vpmu_all_set()?


vpmu_are_all_set() is better.

-boris

vpmu_are_all_set