[PATCH v3 1/1] KVM: x86: Introduce has_protected_pmu state for TDX VMs

[PATCH v3 1/1] KVM: x86: Introduce has_protected_pmu state for TDX VMs

[Vishal Annapurve]

PMU state for TDX VMs is virtualized by TDX Module [1]. Host has following
toggles to control the PMU functionality exposed to TDX VMs:
1) Configure TD_PARAMS to allow guests to use performance monitoring.
2) Restrict the TD to a subset of the PEBS counters if supported.
3) Limit the TD to setup a certain perfmon events using basic/enhanced
   event filtering.

KVM will need to be enlightened to support these toggles. Introduce
has_protected_pmu state to track the pmu state for such scenarios
and explicitly set the has_protected_pmu flag for TDX VMs.

If pmu state is protected:
1) Disable KVM's PMU virtualization framework as additional
enlightenment is needed within KVM to control/manage the
visibility of PMU state to such VMs.
2) Disallow userspace VMM from toggling PMU virtualization state
using KVM_CAP_PMU_CAPABILITY.

[1] Section 15.2: https://cdrdv2.intel.com/v1/dl/getContent/733575

Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Vishal Annapurve <vannapurve@google.com>

---
v2 -> v3:
- Squashed two patches into a single patch as per feedback from Sean.
- Dropped the cover letter.

v2: https://lore.kernel.org/kvm/20260507003613.1784851-1-vannapurve@google.com/#t

 arch/x86/include/asm/kvm_host.h | 1 +
 arch/x86/kvm/vmx/tdx.c          | 6 ++++++
 arch/x86/kvm/x86.c              | 8 ++++++--
 3 files changed, 13 insertions(+), 2 deletions(-)

diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index c470e40a00aa..8371dcaaed1a 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -1422,6 +1422,7 @@ struct kvm_arch {
 	bool has_private_mem;
 	bool has_protected_state;
 	bool has_protected_eoi;
+	bool has_protected_pmu;
 	bool pre_fault_allowed;
 	struct hlist_head *mmu_page_hash;
 	struct list_head active_mmu_pages;
diff --git a/arch/x86/kvm/vmx/tdx.c b/arch/x86/kvm/vmx/tdx.c
index 1e47c194af53..eb4b4518e6f0 100644
--- a/arch/x86/kvm/vmx/tdx.c
+++ b/arch/x86/kvm/vmx/tdx.c
@@ -638,6 +638,12 @@ int tdx_vm_init(struct kvm *kvm)
 	kvm->arch.has_private_mem = true;
 	kvm->arch.disabled_quirks |= KVM_X86_QUIRK_IGNORE_GUEST_PAT;
 
+	/*
+	 * PMU support is provided by the TDX-Module (if enabled for the VM).
+	 * From KVM's perspective, the VM doesn't have a virtual PMU.
+	 */
+	kvm->arch.has_protected_pmu = true;
+
 	/*
 	 * Because guest TD is protected, VMM can't parse the instruction in TD.
 	 * Instead, guest uses MMIO hypercall.  For unmodified device driver,
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 0a1b63c63d1a..99a383455d46 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -6910,7 +6910,8 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
 			break;
 
 		mutex_lock(&kvm->lock);
-		if (!kvm->created_vcpus && !kvm->arch.created_mediated_pmu) {
+		if (!kvm->created_vcpus && !kvm->arch.created_mediated_pmu &&
+		    !kvm->arch.has_protected_pmu) {
 			kvm->arch.enable_pmu = !(cap->args[0] & KVM_PMU_CAP_DISABLE);
 			r = 0;
 		}
@@ -13375,7 +13376,10 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	kvm->arch.default_tsc_khz = max_tsc_khz ? : tsc_khz;
 	kvm->arch.apic_bus_cycle_ns = APIC_BUS_CYCLE_NS_DEFAULT;
 	kvm->arch.guest_can_read_msr_platform_info = true;
-	kvm->arch.enable_pmu = enable_pmu;
+	if (kvm->arch.has_protected_pmu)
+		kvm->arch.enable_pmu = false;
+	else
+		kvm->arch.enable_pmu = enable_pmu;
 
 #if IS_ENABLED(CONFIG_HYPERV)
 	spin_lock_init(&kvm->arch.hv_root_tdp_lock);
-- 
2.54.0.563.g4f69b47b94-goog

Re: [PATCH v3 1/1] KVM: x86: Introduce has_protected_pmu state for TDX VMs

[Sean Christopherson]

On Thu, May 07, 2026, Vishal Annapurve wrote:
> PMU state for TDX VMs is virtualized by TDX Module [1]. Host has following
> toggles to control the PMU functionality exposed to TDX VMs:
> 1) Configure TD_PARAMS to allow guests to use performance monitoring.
> 2) Restrict the TD to a subset of the PEBS counters if supported.
> 3) Limit the TD to setup a certain perfmon events using basic/enhanced
>    event filtering.
> 
> KVM will need to be enlightened to support these toggles. Introduce
> has_protected_pmu state to track the pmu state for such scenarios
> and explicitly set the has_protected_pmu flag for TDX VMs.
> 
> If pmu state is protected:
> 1) Disable KVM's PMU virtualization framework as additional
> enlightenment is needed within KVM to control/manage the
> visibility of PMU state to such VMs.
> 2) Disallow userspace VMM from toggling PMU virtualization state
> using KVM_CAP_PMU_CAPABILITY.
> 
> [1] Section 15.2: https://cdrdv2.intel.com/v1/dl/getContent/733575
> 
> Suggested-by: Sean Christopherson <seanjc@google.com>
> Signed-off-by: Vishal Annapurve <vannapurve@google.com>
> 
> ---
> v2 -> v3:
> - Squashed two patches into a single patch as per feedback from Sean.
> - Dropped the cover letter.

...

> @@ -13375,7 +13376,10 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
>  	kvm->arch.default_tsc_khz = max_tsc_khz ? : tsc_khz;
>  	kvm->arch.apic_bus_cycle_ns = APIC_BUS_CYCLE_NS_DEFAULT;
>  	kvm->arch.guest_can_read_msr_platform_info = true;
> -	kvm->arch.enable_pmu = enable_pmu;
> +	if (kvm->arch.has_protected_pmu)
> +		kvm->arch.enable_pmu = false;
> +	else
> +		kvm->arch.enable_pmu = enable_pmu;

Once more, with feeling :-)

	kvm->arch.enable_pmu = enable_pmu && !kvm->arch.has_protected_pmu;

No need for a v4, I'll fixup when applying (assuming this is indeed the directin
we take).

>  
>  #if IS_ENABLED(CONFIG_HYPERV)
>  	spin_lock_init(&kvm->arch.hv_root_tdp_lock);
> -- 
> 2.54.0.563.g4f69b47b94-goog
> 

Re: [PATCH v3 1/1] KVM: x86: Introduce has_protected_pmu state for TDX VMs

[Sean Christopherson]

On Thu, May 07, 2026, Vishal Annapurve wrote:
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 0a1b63c63d1a..99a383455d46 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -6910,7 +6910,8 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
>  			break;
>  
>  		mutex_lock(&kvm->lock);
> -		if (!kvm->created_vcpus && !kvm->arch.created_mediated_pmu) {
> +		if (!kvm->created_vcpus && !kvm->arch.created_mediated_pmu &&
> +		    !kvm->arch.has_protected_pmu) {

An offlist Sashiko review pointed out that this will prevent doing
KVM_PMU_CAP_DISABLE on a TDX VM, which could prove problematic if userspace is
already explicitly disabling vPMU support for TDX.  I was thinking that wouldn't
be a problem since KVM already disallows KVM_PMU_CAP_DISABLE if enable_pmu=false,
but in that case KVM doesn't advertise KVM_CAP_PMU_VALID_MASK:

	r = enable_pmu ? KVM_CAP_PMU_VALID_MASK : 0;

And unless I'm misreading QEMU code, this will indeed be a problem.

Argh.  I was going to suggest something sightly fancier, but we rather stupidly
didn't provide a flag for ENABLE, and so we can't feed has_protected_pmu into the
valid mask.  Lame.

The other super minor detail is that kvm->arch.has_protected_pmu isn't guarded by
kvm->lock, and I really hope we never get to that point (e.g. for SNP).

So instead of disallowing KVM_CAP_PMU_CAPABILITY entirely, this?

diff --git arch/x86/kvm/x86.c arch/x86/kvm/x86.c
index dc69b8cebe0b..b833a2596430 100644
--- arch/x86/kvm/x86.c
+++ arch/x86/kvm/x86.c
@@ -6911,6 +6911,13 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
                if (!enable_pmu || (cap->args[0] & ~KVM_CAP_PMU_VALID_MASK))
                        break;
 
+               if (kvm->arch.has_protected_pmu) {
+                       WARN_ON_ONCE(kvm->arch.enable_pmu);
+                       if (cap->args[0] == KVM_PMU_CAP_DISABLE)
+                               r = 0;
+                       break;
+               }
+
                mutex_lock(&kvm->lock);
                if (!kvm->created_vcpus && !kvm->arch.created_mediated_pmu) {
                        kvm->arch.enable_pmu = !(cap->args[0] & KVM_PMU_CAP_DISABLE);


>  			kvm->arch.enable_pmu = !(cap->args[0] & KVM_PMU_CAP_DISABLE);
>  			r = 0;
>  		}
> @@ -13375,7 +13376,10 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
>  	kvm->arch.default_tsc_khz = max_tsc_khz ? : tsc_khz;
>  	kvm->arch.apic_bus_cycle_ns = APIC_BUS_CYCLE_NS_DEFAULT;
>  	kvm->arch.guest_can_read_msr_platform_info = true;
> -	kvm->arch.enable_pmu = enable_pmu;
> +	if (kvm->arch.has_protected_pmu)
> +		kvm->arch.enable_pmu = false;
> +	else
> +		kvm->arch.enable_pmu = enable_pmu;
>  
>  #if IS_ENABLED(CONFIG_HYPERV)
>  	spin_lock_init(&kvm->arch.hv_root_tdp_lock);
> -- 
> 2.54.0.563.g4f69b47b94-goog
> 

Re: [PATCH v3 1/1] KVM: x86: Introduce has_protected_pmu state for TDX VMs

[Vishal Annapurve]

On Thu, May 7, 2026 at 1:11 PM Sean Christopherson <seanjc@google.com> wrote:
>
> On Thu, May 07, 2026, Vishal Annapurve wrote:
> > diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> > index 0a1b63c63d1a..99a383455d46 100644
> > --- a/arch/x86/kvm/x86.c
> > +++ b/arch/x86/kvm/x86.c
> > @@ -6910,7 +6910,8 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
> >                       break;
> >
> >               mutex_lock(&kvm->lock);
> > -             if (!kvm->created_vcpus && !kvm->arch.created_mediated_pmu) {
> > +             if (!kvm->created_vcpus && !kvm->arch.created_mediated_pmu &&
> > +                 !kvm->arch.has_protected_pmu) {
>
> An offlist Sashiko review pointed out that this will prevent doing
> KVM_PMU_CAP_DISABLE on a TDX VM, which could prove problematic if userspace is
> already explicitly disabling vPMU support for TDX.  I was thinking that wouldn't
> be a problem since KVM already disallows KVM_PMU_CAP_DISABLE if enable_pmu=false,
> but in that case KVM doesn't advertise KVM_CAP_PMU_VALID_MASK:
>
>         r = enable_pmu ? KVM_CAP_PMU_VALID_MASK : 0;
>
> And unless I'm misreading QEMU code, this will indeed be a problem.
>
> Argh.  I was going to suggest something sightly fancier, but we rather stupidly
> didn't provide a flag for ENABLE, and so we can't feed has_protected_pmu into the
> valid mask.  Lame.
>
> The other super minor detail is that kvm->arch.has_protected_pmu isn't guarded by
> kvm->lock, and I really hope we never get to that point (e.g. for SNP).
>
> So instead of disallowing KVM_CAP_PMU_CAPABILITY entirely, this?
>
> diff --git arch/x86/kvm/x86.c arch/x86/kvm/x86.c
> index dc69b8cebe0b..b833a2596430 100644
> --- arch/x86/kvm/x86.c
> +++ arch/x86/kvm/x86.c
> @@ -6911,6 +6911,13 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
>                 if (!enable_pmu || (cap->args[0] & ~KVM_CAP_PMU_VALID_MASK))
>                         break;
>
> +               if (kvm->arch.has_protected_pmu) {
> +                       WARN_ON_ONCE(kvm->arch.enable_pmu);
> +                       if (cap->args[0] == KVM_PMU_CAP_DISABLE)
> +                               r = 0;
> +                       break;
> +               }
> +
>                 mutex_lock(&kvm->lock);
>                 if (!kvm->created_vcpus && !kvm->arch.created_mediated_pmu) {
>                         kvm->arch.enable_pmu = !(cap->args[0] & KVM_PMU_CAP_DISABLE);
>

Incorporated this in v4 [1].

[1] https://lore.kernel.org/kvm/20260507212957.2629561-1-vannapurve@google.com/.

>
> >                       kvm->arch.enable_pmu = !(cap->args[0] & KVM_PMU_CAP_DISABLE);
> >                       r = 0;
> >               }
> > @@ -13375,7 +13376,10 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
> >       kvm->arch.default_tsc_khz = max_tsc_khz ? : tsc_khz;
> >       kvm->arch.apic_bus_cycle_ns = APIC_BUS_CYCLE_NS_DEFAULT;
> >       kvm->arch.guest_can_read_msr_platform_info = true;
> > -     kvm->arch.enable_pmu = enable_pmu;
> > +     if (kvm->arch.has_protected_pmu)
> > +             kvm->arch.enable_pmu = false;
> > +     else
> > +             kvm->arch.enable_pmu = enable_pmu;
> >
> >  #if IS_ENABLED(CONFIG_HYPERV)
> >       spin_lock_init(&kvm->arch.hv_root_tdp_lock);
> > --
> > 2.54.0.563.g4f69b47b94-goog
> >