linux-kernel - Re: [PATCH 1/2] KVM: x86: Deduplicate MSR interception enabling and disabling

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: <aFsqMnjTCZSBTO3m@google.com>
Date: Tue, 24 Jun 2025 15:44:02 -0700
From: Sean Christopherson <seanjc@...gle.com>
To: Chao Gao <chao.gao@...el.com>
Cc: Dapeng Mi <dapeng1.mi@...ux.intel.com>, linux-kernel@...r.kernel.org, 
	kvm@...r.kernel.org, pbonzini@...hat.com
Subject: Re: [PATCH 1/2] KVM: x86: Deduplicate MSR interception enabling and disabling

On Mon, Jun 16, 2025, Chao Gao wrote:
> >> --- a/arch/x86/kvm/vmx/vmx.h
> >> +++ b/arch/x86/kvm/vmx/vmx.h
> >> @@ -388,21 +388,13 @@ void vmx_ept_load_pdptrs(struct kvm_vcpu *vcpu);
> >>  
> >>  void vmx_disable_intercept_for_msr(struct kvm_vcpu *vcpu, u32 msr, int type);
> >>  void vmx_enable_intercept_for_msr(struct kvm_vcpu *vcpu, u32 msr, int type);
> >> +void vmx_set_intercept_for_msr(struct kvm_vcpu *vcpu, u32 msr, int type, bool enable);
> >>  
> >>  u64 vmx_get_l2_tsc_offset(struct kvm_vcpu *vcpu);
> >>  u64 vmx_get_l2_tsc_multiplier(struct kvm_vcpu *vcpu);
> >>  
> >>  gva_t vmx_get_untagged_addr(struct kvm_vcpu *vcpu, gva_t gva, unsigned int flags);
> >>  
> >> -static inline void vmx_set_intercept_for_msr(struct kvm_vcpu *vcpu, u32 msr,
> >> -					     int type, bool value)
> >> -{
> >> -	if (value)
> >> -		vmx_enable_intercept_for_msr(vcpu, msr, type);
> >> -	else
> >> -		vmx_disable_intercept_for_msr(vcpu, msr, type);
> >> -}
> >> -
> >>  void vmx_update_cpu_dirty_logging(struct kvm_vcpu *vcpu);
> >>  
> >>  /*
> >
> >The change looks good to me. 
> >
> >Reviewed-by: Dapeng Mi <dapeng1.mi@...ux.intel.com>
> 
> Thanks.
> 
> >
> >Just curious, is there a preference on using these 3 interfaces? When
> >should we use the disable/enable interfaces? When should be we use the set
> >interface?  or no preference?
> 
> I think the set API is to reduce boilerplate code. So, use the set API when
> you need to perform conditional logic, such as
> 
> 	if (/*check guest/host caps*/)
> 		//disable intercept
> 	else
> 		//enable intercept
> 
> otherwise, use the disable/enable APIs.

Yep.  The preference is to use the API that is most appropriate for the code.
E.g. if the code unconditionally enables/disables interceptions, then it should
use the appropriate wrapper.