Re: [PATCH] KVM: x86: Break kvm_for_each_vcpu loop after finding the VP_INDEX

[Paolo Bonzini <pbonzini@redhat.com>]

Il 27/02/2014 07:08, Takuya Yoshikawa ha scritto:
> No need to scan the entire VCPU array.
>
> Signed-off-by: Takuya Yoshikawa <yoshikawa_takuya_b1@lab.ntt.co.jp>
> ---
>  BTW, this looks like hyperv support forces us to stick to the current
>  implementation which stores VCPUs in an array, or at least something
>  we can index them; not a good thing.
>
>  arch/x86/kvm/x86.c |    7 +++++--
>  1 file changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 4cca458..773eba7 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -2328,9 +2328,12 @@ static int get_msr_hyperv(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata)
>  	case HV_X64_MSR_VP_INDEX: {
>  		int r;
>  		struct kvm_vcpu *v;
> -		kvm_for_each_vcpu(r, v, vcpu->kvm)
> -			if (v == vcpu)
> +		kvm_for_each_vcpu(r, v, vcpu->kvm) {
> +			if (v == vcpu) {
>  				data = r;
> +				break;
> +			}
> +		}
>  		break;
>  	}
>  	case HV_X64_MSR_EOI:
>

Applying to kvm/queue, thanks.

Paolo