Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Mon, 16 May 2022 13:58:02 +0000
From:   Sean Christopherson <seanjc@google.com>
To:     Yanfei Xu <yanfei.xu@intel.com>
Cc:     pbonzini@redhat.com, vkuznets@redhat.com, wanpengli@tencent.com,
        jmattson@google.com, joro@8bytes.org, tglx@linutronix.de,
        mingo@redhat.com, bp@alien8.de, dave.hansen@linux.intel.com,
        x86@kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] KVM: x86: Fix the intel_pt PMI handling wrongly
 considered from guest
Message-ID: <YoJYah+Ct90aj1I5@google.com>
References: <20220515171633.902901-1-yanfei.xu@intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20220515171633.902901-1-yanfei.xu@intel.com>
Precedence: bulk

On Mon, May 16, 2022, Yanfei Xu wrote:
> When kernel handles the vm-exit caused by external interrupts and PMI,
> it always set a type of kvm_intr_type to handling_intr_from_guest to
> tell if it's dealing an IRQ or NMI.
> However, the further type judgment is missing in kvm_arch_pmi_in_guest().
> It could make the PMI of intel_pt wrongly considered it comes from a
> guest once the PMI breaks the handling of vm-exit of external interrupts.
> 
> Fixes: db215756ae59 ("KVM: x86: More precisely identify NMI from guest when handling PMI")
> Signed-off-by: Yanfei Xu <yanfei.xu@intel.com>
> ---
>  arch/x86/include/asm/kvm_host.h | 8 +++++++-
>  arch/x86/kvm/x86.h              | 6 ------
>  2 files changed, 7 insertions(+), 7 deletions(-)
> 
> diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
> index 4ff36610af6a..308cf19f123d 100644
> --- a/arch/x86/include/asm/kvm_host.h
> +++ b/arch/x86/include/asm/kvm_host.h
> @@ -1582,8 +1582,14 @@ static inline int kvm_arch_flush_remote_tlb(struct kvm *kvm)
>  		return -ENOTSUPP;
>  }
>  
> +enum kvm_intr_type {
> +	/* Values are arbitrary, but must be non-zero. */
> +	KVM_HANDLING_IRQ = 1,
> +	KVM_HANDLING_NMI,
> +};
> +
>  #define kvm_arch_pmi_in_guest(vcpu) \
> -	((vcpu) && (vcpu)->arch.handling_intr_from_guest)
> +	((vcpu) && (vcpu)->arch.handling_intr_from_guest == KVM_HANDLING_NMI)

My understanding is that this isn't correct as a general change, as perf events
can use regular IRQs in some cases.  See commit dd60d217062f4 ("KVM: x86: Fix perf
timer mode IP reporting").

I assume there's got to be a way to know which mode perf is using, e.g. we should
be able to make this look something like:

	((vcpu) && (vcpu)->arch.handling_intr_from_guest == kvm_pmi_vector)


>  void kvm_mmu_x86_module_init(void);
>  int kvm_mmu_vendor_module_init(void);
> diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h
> index 588792f00334..3bdf1bc76863 100644
> --- a/arch/x86/kvm/x86.h
> +++ b/arch/x86/kvm/x86.h
> @@ -344,12 +344,6 @@ static inline bool kvm_cstate_in_guest(struct kvm *kvm)
>  	return kvm->arch.cstate_in_guest;
>  }
>  
> -enum kvm_intr_type {
> -	/* Values are arbitrary, but must be non-zero. */
> -	KVM_HANDLING_IRQ = 1,
> -	KVM_HANDLING_NMI,
> -};
> -
>  static inline void kvm_before_interrupt(struct kvm_vcpu *vcpu,
>  					enum kvm_intr_type intr)
>  {
> -- 
> 2.32.0
>