[PATCH] KVM: x86: Use active memslots for the per-vCPU MMIO cache

[PATCH] KVM: x86: Use active memslots for the per-vCPU MMIO cache

[Jinu Kim]

KVM tags the per-vCPU MMIO cache with a memslot generation so that a
memslot update invalidates cached MMIO information.  Both the fill and
validation paths use kvm_memslots(), which unconditionally selects address
space 0, even when the vCPU is running in SMM and using address space 1.

Commit 56f17dd3fbc4 ("kvm: x86: fix stale mmio cache bug") added the
memslot generation to the cache key so that a memslot update could not
leave a stale entry valid.  When commit 699023e23965 ("KVM: x86: add SMM
to the MMU role, support SMRAM address space") added the SMM address space,
these helpers were not converted to use the active memslots.

Consequently, an update to the SMM memslots can leave an entry from the old
SMM address space apparently valid after the vCPU returns to the normal
address space.  A guest can then cause an access to valid RAM at the same
GFN to be returned to userspace as KVM_EXIT_MMIO.

Completing that exit through the VMM's RAM address space writes the backing
page without going through KVM's write-tracking path.  If the page backs a
nested EPT table, KVM can continue using shadow translations derived from
the old contents.  The WARN_ON_ONCE() in kvm_mmu_write_protect_fault()
catches this invalid cache and memslot combination.

Use the vCPU's active memslots when caching and validating the entry.
Memslot generations are unique across address spaces, so the generation
also distinguishes the normal and SMM views.  This matches the MMIO SPTE
cache, which already uses kvm_vcpu_memslots().

On an unpatched current-mainline kernel, a regression test that fills the
cache in SMM and updates only the SMM memslots fails the intended RAM write
and triggers the kvm_mmu_write_protect_fault() warning.  With this change,
the same test completes the RAM write without a warning.  The x86/smm_test,
set_memory_region_test, and memslot_modification_stress_test selftests also
pass.

Fixes: 699023e23965 ("KVM: x86: add SMM to the MMU role, support SMRAM address space")
Cc: stable@vger.kernel.org
Signed-off-by: Jinu Kim <kimjw04271234@gmail.com>
---
 arch/x86/kvm/x86.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h
index 9de577ef9c97..a6d86d3dcff6 100644
--- a/arch/x86/kvm/x86.h
+++ b/arch/x86/kvm/x86.h
@@ -313,7 +313,7 @@ static inline bool is_noncanonical_invlpg_address(u64 la, struct kvm_vcpu *vcpu)
 static inline void vcpu_cache_mmio_info(struct kvm_vcpu *vcpu,
 					gva_t gva, gfn_t gfn, unsigned access)
 {
-	u64 gen = kvm_memslots(vcpu->kvm)->generation;
+	u64 gen = kvm_vcpu_memslots(vcpu)->generation;
 
 	if (unlikely(gen & KVM_MEMSLOT_GEN_UPDATE_IN_PROGRESS))
 		return;
@@ -330,7 +330,7 @@ static inline void vcpu_cache_mmio_info(struct kvm_vcpu *vcpu,
 
 static inline bool vcpu_match_mmio_gen(struct kvm_vcpu *vcpu)
 {
-	return vcpu->arch.mmio_gen == kvm_memslots(vcpu->kvm)->generation;
+	return vcpu->arch.mmio_gen == kvm_vcpu_memslots(vcpu)->generation;
 }
 
 /*
-- 
2.43.0

Re: [PATCH] KVM: x86: Use active memslots for the per-vCPU MMIO cache

[Jinu Kim]

Hi Sean, Paolo,

A friendly ping on this patch.
I originally sent this on August 7, which was fairly late in the 7.2
release cycle, so I held off on pinging during the remainder of the
cycle and the 7.3 merge window.
Now that 7.3-rc1 is out, could you please take a look when you get a chance?
Please let me know if you'd prefer that I resend or rebase the patch.

Thanks,
Jinu

2026년 8월 7일 (금) 오후 8:28, Jinu Kim <kimjw04271234@gmail.com>님이 작성:
>
> KVM tags the per-vCPU MMIO cache with a memslot generation so that a
> memslot update invalidates cached MMIO information.  Both the fill and
> validation paths use kvm_memslots(), which unconditionally selects address
> space 0, even when the vCPU is running in SMM and using address space 1.
>
> Commit 56f17dd3fbc4 ("kvm: x86: fix stale mmio cache bug") added the
> memslot generation to the cache key so that a memslot update could not
> leave a stale entry valid.  When commit 699023e23965 ("KVM: x86: add SMM
> to the MMU role, support SMRAM address space") added the SMM address space,
> these helpers were not converted to use the active memslots.
>
> Consequently, an update to the SMM memslots can leave an entry from the old
> SMM address space apparently valid after the vCPU returns to the normal
> address space.  A guest can then cause an access to valid RAM at the same
> GFN to be returned to userspace as KVM_EXIT_MMIO.
>
> Completing that exit through the VMM's RAM address space writes the backing
> page without going through KVM's write-tracking path.  If the page backs a
> nested EPT table, KVM can continue using shadow translations derived from
> the old contents.  The WARN_ON_ONCE() in kvm_mmu_write_protect_fault()
> catches this invalid cache and memslot combination.
>
> Use the vCPU's active memslots when caching and validating the entry.
> Memslot generations are unique across address spaces, so the generation
> also distinguishes the normal and SMM views.  This matches the MMIO SPTE
> cache, which already uses kvm_vcpu_memslots().
>
> On an unpatched current-mainline kernel, a regression test that fills the
> cache in SMM and updates only the SMM memslots fails the intended RAM write
> and triggers the kvm_mmu_write_protect_fault() warning.  With this change,
> the same test completes the RAM write without a warning.  The x86/smm_test,
> set_memory_region_test, and memslot_modification_stress_test selftests also
> pass.
>
> Fixes: 699023e23965 ("KVM: x86: add SMM to the MMU role, support SMRAM address space")
> Cc: stable@vger.kernel.org
> Signed-off-by: Jinu Kim <kimjw04271234@gmail.com>
> ---
>  arch/x86/kvm/x86.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h
> index 9de577ef9c97..a6d86d3dcff6 100644
> --- a/arch/x86/kvm/x86.h
> +++ b/arch/x86/kvm/x86.h
> @@ -313,7 +313,7 @@ static inline bool is_noncanonical_invlpg_address(u64 la, struct kvm_vcpu *vcpu)
>  static inline void vcpu_cache_mmio_info(struct kvm_vcpu *vcpu,
>                                         gva_t gva, gfn_t gfn, unsigned access)
>  {
> -       u64 gen = kvm_memslots(vcpu->kvm)->generation;
> +       u64 gen = kvm_vcpu_memslots(vcpu)->generation;
>
>         if (unlikely(gen & KVM_MEMSLOT_GEN_UPDATE_IN_PROGRESS))
>                 return;
> @@ -330,7 +330,7 @@ static inline void vcpu_cache_mmio_info(struct kvm_vcpu *vcpu,
>
>  static inline bool vcpu_match_mmio_gen(struct kvm_vcpu *vcpu)
>  {
> -       return vcpu->arch.mmio_gen == kvm_memslots(vcpu->kvm)->generation;
> +       return vcpu->arch.mmio_gen == kvm_vcpu_memslots(vcpu)->generation;
>  }
>
>  /*
> --
> 2.43.0
>

Re: [PATCH] KVM: x86: Use active memslots for the per-vCPU MMIO cache

[Sean Christopherson]

On Tue, Sep 01, 2026, Jinu Kim wrote:
> Hi Sean, Paolo,
> 
> A friendly ping on this patch.
> I originally sent this on August 7, which was fairly late in the 7.2
> release cycle, so I held off on pinging during the remainder of the
> cycle and the 7.3 merge window.
> Now that 7.3-rc1 is out, could you please take a look when you get a chance?
> Please let me know if you'd prefer that I resend or rebase the patch.

No RESEND is necessary, this is already on my list of things to grab (or to make
sure Paolo grabs).