[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <53F1D476.1030801@redhat.com>
Date: Mon, 18 Aug 2014 12:24:54 +0200
From: Paolo Bonzini <pbonzini@...hat.com>
To: Wanpeng Li <wanpeng.li@...ux.intel.com>
CC: Gleb Natapov <gleb@...nel.org>, hpa@...or.com, x86@...nel.org,
kvm@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 3/5] KVM: vmx: don't vmx_segment_cache_clear twice in
enter_pmode
Il 18/08/2014 11:50, Wanpeng Li ha scritto:
> vmx_segment_cache_clear() will be called by vmx_set_segment()
> which lead to vmx_segment_cache_clear() is called twice in
> enter_pmode(). This patch remove the duplicate call site.
>
> Reviewed-by: Yang Zhang <yang.z.zhang@...el.com>
> Signed-off-by: Wanpeng Li <wanpeng.li@...ux.intel.com>
> ---
> arch/x86/kvm/vmx.c | 2 --
> 1 file changed, 2 deletions(-)
>
> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
> index 2963303..c97c44c 100644
> --- a/arch/x86/kvm/vmx.c
> +++ b/arch/x86/kvm/vmx.c
> @@ -3183,8 +3183,6 @@ static void enter_pmode(struct kvm_vcpu *vcpu)
>
> vmx->rmode.vm86_active = 0;
>
> - vmx_segment_cache_clear(vmx);
> -
> vmx_set_segment(vcpu, &vmx->rmode.segs[VCPU_SREG_TR], VCPU_SREG_TR);
>
> flags = vmcs_readl(GUEST_RFLAGS);
>
I think this is not useful and obscures the code a bit.
It would be a bigger improvement to see when fix_pmode_seg really needs
to call vmx_set_segment, and avoid the call. Are you seeing enter_pmode
high in any profiles?
Paolo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists