On Tue, Sep 26, 2017 at 11:34:22PM -0700, Wanpeng Li wrote:
> From: Wanpeng Li <wanpeng.li@xxxxxxxxxxx>
>
> PLE_Window: Software can configure this field as an upper bound on the amount of time
> a guest is allowed to execute in a PAUSE LOOP.
>
> KVM doesn't expose the PLE capability to the L1 hypervsior, however, ple_window still
s/hypervsior/hypervisor/
> shows the default value on L1 hypervsior. This patch fixes it by clearing all the
Ditto here.
> PLE related module parameter if there is no PLE capability.
>
> Cc: Paolo Bonzini <pbonzini@xxxxxxxxxx>
> Cc: Radim Krčmář <rkrcmar@xxxxxxxxxx>
> Signed-off-by: Wanpeng Li <wanpeng.li@xxxxxxxxxxx>
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx>
> ---
> arch/x86/kvm/vmx.c | 7 ++++++-
> 1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
> index c83d28b..4d4f9b4 100644
> --- a/arch/x86/kvm/vmx.c
> +++ b/arch/x86/kvm/vmx.c
> @@ -6781,8 +6781,13 @@ static __init int hardware_setup(void)
> if (enable_ept && !cpu_has_vmx_ept_2m_page())
> kvm_disable_largepages();
>
> - if (!cpu_has_vmx_ple())
> + if (!cpu_has_vmx_ple()) {
> ple_gap = 0;
> + ple_window = 0;
> + ple_window_grow = 0;
> + ple_window_max = 0;
> + ple_window_shrink = 0;
> + }
>
> if (!cpu_has_vmx_apicv()) {
> enable_apicv = 0;
> --
> 2.7.4
>
