[Devel] [PATCH] Revert "KVM: nested VMX: disable perf cpuid reporting"

Denis V. Lunev den at virtuozzo.com
Mon Jul 24 17:33:53 MSK 2017


On 07/24/2017 05:31 PM, Denis Plotnikov wrote:
> From: Jim Mattson <jmattson at google.com>
>
> This reverts commit bc6134942dbbf31c25e9bd7c876be5da81c9e1ce.
>
> A CPUID instruction executed in VMX non-root mode always causes a
> VM-exit, regardless of the leaf being queried.
>
> Fixes: bc6134942dbb ("KVM: nested VMX: disable perf cpuid reporting")
> Signed-off-by: Jim Mattson <jmattson at google.com>
> [The issue solved by bc6134942dbb has been resolved with ff651cb613b4
>  ("KVM: nVMX: Add nested msr load/restore algorithm").]
> Signed-off-by: Radim Krčmář <rkrcmar at redhat.com>
>
> (cherry picked from commit 0b4c208d443ba2af82b4c70f99ca8df31e9a0020)
>
> fix #PSBM-68274
>
> Signed-off-by: Denis Plotnikov <dplotnikov at virtuozzo.com>
> ---
>  arch/x86/kvm/cpuid.c | 6 ------
>  arch/x86/kvm/vmx.c   | 2 --
>  2 files changed, 8 deletions(-)
>
> diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
> index e245bf66..f6be926 100644
> --- a/arch/x86/kvm/cpuid.c
> +++ b/arch/x86/kvm/cpuid.c
> @@ -831,12 +831,6 @@ void kvm_cpuid(struct kvm_vcpu *vcpu, u32 *eax, u32 *ebx, u32 *ecx, u32 *edx)
>  	if (!best)
>  		best = check_cpuid_limit(vcpu, function, index);
>  
> -	/*
> -	 * Perfmon not yet supported for L2 guest.
> -	 */
> -	if (is_guest_mode(vcpu) && function == 0xa)
> -		best = NULL;
> -
>  	if (best) {
>  		*eax = best->eax;
>  		*ebx = best->ebx;
> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
> index 16ca8fa..6fd1049 100644
> --- a/arch/x86/kvm/vmx.c
> +++ b/arch/x86/kvm/vmx.c
> @@ -7663,8 +7663,6 @@ static bool nested_vmx_exit_handled(struct kvm_vcpu *vcpu)
>  	case EXIT_REASON_TASK_SWITCH:
>  		return true;
>  	case EXIT_REASON_CPUID:
> -		if (kvm_register_read(vcpu, VCPU_REGS_RAX) == 0xa)
> -			return false;
>  		return true;
>  	case EXIT_REASON_HLT:
>  		return nested_cpu_has(vmcs12, CPU_BASED_HLT_EXITING);
up6 pls


More information about the Devel mailing list