From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751219AbeCHUlM (ORCPT ); Thu, 8 Mar 2018 15:41:12 -0500 Received: from mx3-rdu2.redhat.com ([66.187.233.73]:38442 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751157AbeCHUlK (ORCPT ); Thu, 8 Mar 2018 15:41:10 -0500 Date: Thu, 8 Mar 2018 21:40:29 +0100 From: Radim =?utf-8?B?S3LEjW3DocWZ?= To: Wanpeng Li Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org, Paolo Bonzini Subject: Re: [PATCH 2/3] KVM: X86: Provides userspace with a capability to not intercept HLT Message-ID: <20180308204029.GK12290@flask> References: <1519897782-8124-1-git-send-email-wanpengli@tencent.com> <1519897782-8124-2-git-send-email-wanpengli@tencent.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <1519897782-8124-2-git-send-email-wanpengli@tencent.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 2018-03-01 17:49+0800, Wanpeng Li: > From: Wanpeng Li > > If host CPUs are dedicated to a VM, we can avoid VM exits on HLT. > This patch adds the per-VM non-HLT-exiting capability. > > Cc: Paolo Bonzini > Cc: Radim Krčmář > Signed-off-by: Wanpeng Li > --- > diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h > index dff3a5d..fcf8339 100644 > --- a/arch/x86/kvm/svm.c > +++ b/arch/x86/kvm/svm.c > @@ -1394,6 +1394,9 @@ static void init_vmcb(struct vcpu_svm *svm) > set_intercept(svm, INTERCEPT_MWAIT); > } > > + if (!kvm_hlt_in_guest(svm->vcpu.kvm)) > + set_intercept(svm, INTERCEPT_HLT); We unconditionally set INTERCEPT_HLT just above, so that line has to be removed. > + > control->iopm_base_pa = __sme_set(iopm_base); > control->msrpm_base_pa = __sme_set(__pa(svm->msrpm)); > control->int_ctl = V_INTR_MASKING_MASK; > diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c > @@ -2525,6 +2525,19 @@ static int nested_vmx_check_exception(struct kvm_vcpu *vcpu, unsigned long *exit > return 0; > } > > +static void vmx_clear_hlt(struct kvm_vcpu *vcpu) > +{ > + /* > + * Ensure that we clear the HLT state in the VMCS. We don't need to > + * explicitly skip the instruction because if the HLT state is set, > + * then the instruction is already executing and RIP has already been > + * advanced. > + */ > + if (kvm_hlt_in_guest(vcpu->kvm) && > + vmcs_read32(GUEST_ACTIVITY_STATE) == GUEST_ACTIVITY_HLT) > + vmcs_write32(GUEST_ACTIVITY_STATE, GUEST_ACTIVITY_ACTIVE); > +} The clearing seems to be still missing around SMM -- I think you need to call vmx_clear_hlt() from pre_enter_smm(). Thanks.