linux-kernel - Re: [PATCH v3 4/7] KVM: x86: nSVM: support PAUSE filter threshold and count when cpu

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <5a63e5f2d88aa8a2caf84319c469b265ec430eb2.camel@redhat.com>
Date:   Tue, 22 Mar 2022 18:52:05 +0200
From:   Maxim Levitsky <mlevitsk@...hat.com>
To:     Paolo Bonzini <pbonzini@...hat.com>, kvm@...r.kernel.org
Cc:     Ingo Molnar <mingo@...hat.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Sean Christopherson <seanjc@...gle.com>,
        Borislav Petkov <bp@...en8.de>,
        "H. Peter Anvin" <hpa@...or.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Jim Mattson <jmattson@...gle.com>, x86@...nel.org,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Joerg Roedel <joro@...tes.org>, linux-kernel@...r.kernel.org,
        Wanpeng Li <wanpengli@...cent.com>
Subject: Re: [PATCH v3 4/7] KVM: x86: nSVM: support PAUSE filter threshold
 and count when cpu_pm=on

On Wed, 2022-03-09 at 14:12 +0100, Paolo Bonzini wrote:
> On 3/1/22 15:36, Maxim Levitsky wrote:
> > Allow L1 to use these settings if L0 disables PAUSE interception
> > (AKA cpu_pm=on)
> > 
> > Signed-off-by: Maxim Levitsky <mlevitsk@...hat.com>
> > ---
> >   arch/x86/kvm/svm/nested.c |  6 ++++++
> >   arch/x86/kvm/svm/svm.c    | 17 +++++++++++++++++
> >   arch/x86/kvm/svm/svm.h    |  2 ++
> >   3 files changed, 25 insertions(+)
> > 
> > diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c
> > index 37510cb206190..4cb0bc49986d5 100644
> > --- a/arch/x86/kvm/svm/nested.c
> > +++ b/arch/x86/kvm/svm/nested.c
> > @@ -664,6 +664,12 @@ static void nested_vmcb02_prepare_control(struct vcpu_svm *svm)
> >   	if (!nested_vmcb_needs_vls_intercept(svm))
> >   		svm->vmcb->control.virt_ext |= VIRTUAL_VMLOAD_VMSAVE_ENABLE_MASK;
> >   
> > +	if (svm->pause_filter_enabled)
> > +		svm->vmcb->control.pause_filter_count = svm->nested.ctl.pause_filter_count;
> > +
> > +	if (svm->pause_threshold_enabled)
> > +		svm->vmcb->control.pause_filter_thresh = svm->nested.ctl.pause_filter_thresh;
> 
> I think this should be
> 
> 	if (kvm_pause_in_guest(vcpu->kvm)) {
> 		/* copy from VMCB12 if guest has CPUID, else set to 0 */
> 	} else {
> 		/* copy from VMCB01, unconditionally */
> 	}

> and likewise it should be copied back to VMCB01 unconditionally on 
> vmexit if !kvm_pause_in_guest(vcpu->kvm).


I did something different in the next version of the patches.
Please take a look.


> 
> >   	nested_svm_transition_tlb_flush(vcpu);
> >   
> >   	/* Enter Guest-Mode */
> > diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
> > index 6a571eed32ef4..52198e63c5fc4 100644
> > --- a/arch/x86/kvm/svm/svm.c
> > +++ b/arch/x86/kvm/svm/svm.c
> > @@ -4008,6 +4008,17 @@ static void svm_vcpu_after_set_cpuid(struct kvm_vcpu *vcpu)
> >   
> >   	svm->v_vmload_vmsave_enabled = vls && guest_cpuid_has(vcpu, X86_FEATURE_V_VMSAVE_VMLOAD);
> >   
> > +	if (kvm_pause_in_guest(vcpu->kvm)) {
> > +		svm->pause_filter_enabled = pause_filter_count > 0 &&
> > +					    guest_cpuid_has(vcpu, X86_FEATURE_PAUSEFILTER);
> > +
> > +		svm->pause_threshold_enabled = pause_filter_thresh > 0 &&
> > +					    guest_cpuid_has(vcpu, X86_FEATURE_PFTHRESHOLD);
> 
> Why only if the module parameters are >0?  The module parameter is 
> unused if pause-in-guest is active.

Agree, will do.

> 
> > +	} else {
> > +		svm->pause_filter_enabled = false;
> > +		svm->pause_threshold_enabled = false;
> > +	}
> > +
> >   	svm_recalc_instruction_intercepts(vcpu, svm);
> >   
> >   	/* For sev guests, the memory encryption bit is not reserved in CR3.  */
> > @@ -4763,6 +4774,12 @@ static __init void svm_set_cpu_caps(void)
> >   		if (vls)
> >   			kvm_cpu_cap_set(X86_FEATURE_V_VMSAVE_VMLOAD);
> >   
> > +		if (pause_filter_count)
> > +			kvm_cpu_cap_set(X86_FEATURE_PAUSEFILTER);
> > +
> > +		if (pause_filter_thresh)
> > +			kvm_cpu_cap_set(X86_FEATURE_PFTHRESHOLD);
> 
> Likewise, this should be set using just boot_cpu_has, not the module 
> parameters.

Agree as well + the check above is wrong - it should have been inverted.

> 
> Paolo
> 
> >   		/* Nested VM can receive #VMEXIT instead of triggering #GP */
> >   		kvm_cpu_cap_set(X86_FEATURE_SVME_ADDR_CHK);
> >   	}
> > diff --git a/arch/x86/kvm/svm/svm.h b/arch/x86/kvm/svm/svm.h
> > index a3c93f9c02847..6fa81eb3ffb78 100644
> > --- a/arch/x86/kvm/svm/svm.h
> > +++ b/arch/x86/kvm/svm/svm.h
> > @@ -234,6 +234,8 @@ struct vcpu_svm {
> >   	bool tsc_scaling_enabled          : 1;
> >   	bool lbrv_enabled                 : 1;
> >   	bool v_vmload_vmsave_enabled      : 1;
> > +	bool pause_filter_enabled         : 1;
> > +	bool pause_threshold_enabled      : 1;
> >   
> >   	u32 ldr_reg;
> >   	u32 dfr_reg;


Thanks for the review!

Best regards,
	Maxim Levitsky