[PATCH v3 0/4] KVM: Miscellaneous SEV/SNP related fixes

[PATCH v3 0/4] KVM: Miscellaneous SEV/SNP related fixes

[Atish Patra]

This series addresses a few issues found during code audit of the
KVM SEV/SNP and CCP driver code. The fixes include a incorrect lock state
and incomplete state handling during intra-host migration for SNP VMs.

To: Sean Christopherson <seanjc@google.com>
To: Paolo Bonzini <pbonzini@redhat.com>
To: Borislav Petkov <bp@alien8.de>
To: Dave Hansen <dave.hansen@linux.intel.com>
To: x86@kernel.org
To: H. Peter Anvin <hpa@zytor.com>
To: Tom Lendacky <thomas.lendacky@amd.com>
To: Peter Gonda <pgonda@google.com>
To: Brijesh Singh <brijesh.singh@amd.com>
To: Youngjae Lee <youngjaelee@meta.com>
To: Ashish Kalra <ashish.kalra@amd.com>
To: Michael Roth <michael.roth@amd.com>
To: John Allen <john.allen@amd.com>
To: Herbert Xu <herbert@gondor.apana.org.au>
Cc: clm@meta.com
Cc: kvm@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-crypto@vger.kernel.org
Cc: stable@vger.kernel.org

Signed-off-by: Atish Patra <atishp@meta.com>
---
Changes in v3:
- Added comments, fixed commit messages and fixes tag as per discussions on v2. 
- sev_init_ex_buffer initialized with zero at allocation to prevent any kernel
  data leak in case of init_ex_file is not present. Reported by Sashiko
- Link to v2: https://lore.kernel.org/r/20260601-sev_snp_fixes-v2-0-611891b28a86@meta.com

Changes in v2:
- Added fixes based on the reports by Sashiko. 
- Added a kselftest for validating SNP VM mirroring/migration rejection. 
- Link to v1: https://lore.kernel.org/r/20260528-sev_snp_fixes-v1-0-d67a08151779@meta.com

---
Atish Patra (4):
      KVM: SEV: Do not allow intra-host migration/mirroring of SNP VMs
      KVM: selftests: Verify SNP VMs are rejected from migration and mirroring
      crypto: ccp: Fix possible deadlock in SEV init failure path
      crypto: ccp: Fix memory leak in SEV INIT_EX path

 arch/x86/kvm/svm/sev.c                             |  6 ++-
 drivers/crypto/ccp/sev-dev.c                       | 19 +++++++--
 .../testing/selftests/kvm/x86/sev_migrate_tests.c  | 47 ++++++++++++++++++++++
 3 files changed, 67 insertions(+), 5 deletions(-)
---
base-commit: e7ae89a0c97ce2b68b0983cd01eda67cf373517d
change-id: 20260525-sev_snp_fixes-0b73789c1a91

Best regards,
-- 
Atish Patra <atishp@meta.com>

[PATCH v3 1/4] KVM: SEV: Do not allow intra-host migration/mirroring of SNP VMs

[Atish Patra]

From: Atish Patra <atishp@meta.com>

The intra-host migration/mirroring feature is not fully implemented for
SEV-SNP VMs. The proper migration requires additional SNP-specific
state such as guest_req_mutex, guest_req_buf, and guest_resp_buf to be
transferred or initialized on the destination.

The SNP VM mirroring requires vmsa features to be copied as well otherwise
ASID would be bound to SNP range while VM is detected as a SEV VM.

Reject SNP source VMs in migration/mirroring until proper SNP state
transfer is implemented.

Fixes: 1dfe571c12cf ("KVM: SEV: Add initial SEV-SNP support")

Reported-by: Chris Mason <clm@meta.com>
Reported-by: Sashiko <sashiko-bot@kernel.org>
Assisted-by: Claude:claude-opus-4-6
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Atish Patra <atishp@meta.com>
---
 arch/x86/kvm/svm/sev.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c
index c2126b3c3072..a34326a77290 100644
--- a/arch/x86/kvm/svm/sev.c
+++ b/arch/x86/kvm/svm/sev.c
@@ -2141,8 +2141,10 @@ int sev_vm_move_enc_context_from(struct kvm *kvm, unsigned int source_fd)
 	if (ret)
 		return ret;
 
+	/* Do not allow SNP VM migration until additional state transfer is implemented  */
 	if (kvm->arch.vm_type != source_kvm->arch.vm_type ||
-	    sev_guest(kvm) || !sev_guest(source_kvm)) {
+	    sev_guest(kvm) || !sev_guest(source_kvm) ||
+	    sev_snp_guest(source_kvm)) {
 		ret = -EINVAL;
 		goto out_unlock;
 	}
@@ -2863,8 +2865,10 @@ int sev_vm_copy_enc_context_from(struct kvm *kvm, unsigned int source_fd)
 	 * disallow out-of-band SEV/SEV-ES init if the target is already an
 	 * SEV guest, or if vCPUs have been created.  KVM relies on vCPUs being
 	 * created after SEV/SEV-ES initialization, e.g. to init intercepts.
+	 * Also do not allow SNP VM mirroring until additional state transfer is implemented.
 	 */
 	if (sev_guest(kvm) || !sev_guest(source_kvm) ||
+	    sev_snp_guest(source_kvm) ||
 	    is_mirroring_enc_context(source_kvm) || kvm->created_vcpus) {
 		ret = -EINVAL;
 		goto e_unlock;

-- 
2.53.0-Meta

[PATCH v3 2/4] KVM: selftests: Verify SNP VMs are rejected from migration and mirroring

[Atish Patra]

From: Atish Patra <atishp@meta.com>

Migration and mirroring of SEV-SNP VMs are not supported yet.

Add two selftests that verify KVM rejects intra-host migration and
mirroring when the source VM is an SNP VM, so the restriction stays enforced
until proper SNP state transfer is implemented.

Signed-off-by: Atish Patra <atishp@meta.com>
---
 .../testing/selftests/kvm/x86/sev_migrate_tests.c  | 47 ++++++++++++++++++++++
 1 file changed, 47 insertions(+)

diff --git a/tools/testing/selftests/kvm/x86/sev_migrate_tests.c b/tools/testing/selftests/kvm/x86/sev_migrate_tests.c
index 6b0928e69051..acef6ab26d3d 100644
--- a/tools/testing/selftests/kvm/x86/sev_migrate_tests.c
+++ b/tools/testing/selftests/kvm/x86/sev_migrate_tests.c
@@ -313,6 +313,49 @@ static void test_sev_mirror_parameters(void)
 	kvm_vm_free(vm_no_vcpu);
 }
 
+static void test_sev_snp_migrate_reject(void)
+{
+	struct kvm_vm *src_vm, *dst_vm;
+	int ret;
+
+	src_vm = vm_create_barebones_type(KVM_X86_SNP_VM);
+	snp_vm_init(src_vm);
+	__vm_vcpu_add(src_vm, 0);
+	vm_sev_launch(src_vm, snp_default_policy(), NULL);
+
+	dst_vm = vm_create_barebones_type(KVM_X86_SNP_VM);
+	__vm_vcpu_add(dst_vm, 0);
+
+	ret = __sev_migrate_from(dst_vm, src_vm);
+	TEST_ASSERT(ret == -1 && errno == EINVAL,
+		    "SNP VM migration should be rejected. ret: %d, errno: %d",
+		    ret, errno);
+
+	kvm_vm_free(src_vm);
+	kvm_vm_free(dst_vm);
+}
+
+static void test_sev_snp_mirror_reject(void)
+{
+	struct kvm_vm *src_vm, *dst_vm;
+	int ret;
+
+	src_vm = vm_create_barebones_type(KVM_X86_SNP_VM);
+	snp_vm_init(src_vm);
+	__vm_vcpu_add(src_vm, 0);
+	vm_sev_launch(src_vm, snp_default_policy(), NULL);
+
+	dst_vm = aux_vm_create(false);
+
+	ret = __sev_mirror_create(dst_vm, src_vm);
+	TEST_ASSERT(ret == -1 && errno == EINVAL,
+		    "SNP VM mirroring should be rejected. ret: %d, errno: %d",
+		    ret, errno);
+
+	kvm_vm_free(src_vm);
+	kvm_vm_free(dst_vm);
+}
+
 static void test_sev_move_copy(void)
 {
 	struct kvm_vm *dst_vm, *dst2_vm, *dst3_vm, *sev_vm, *mirror_vm,
@@ -384,12 +427,16 @@ int main(int argc, char *argv[])
 		test_sev_migrate_parameters();
 		if (kvm_has_cap(KVM_CAP_VM_COPY_ENC_CONTEXT_FROM))
 			test_sev_move_copy();
+		if (kvm_cpu_has(X86_FEATURE_SEV_SNP))
+			test_sev_snp_migrate_reject();
 	}
 	if (kvm_has_cap(KVM_CAP_VM_COPY_ENC_CONTEXT_FROM)) {
 		test_sev_mirror(/* es= */ false);
 		if (have_sev_es)
 			test_sev_mirror(/* es= */ true);
 		test_sev_mirror_parameters();
+		if (kvm_cpu_has(X86_FEATURE_SEV_SNP))
+			test_sev_snp_mirror_reject();
 	}
 	return 0;
 }

-- 
2.53.0-Meta

[PATCH v3 3/4] crypto: ccp: Fix possible deadlock in SEV init failure path

[Atish Patra]

From: Atish Patra <atishp@meta.com>

__sev_platform_init_handle_init_ex_path() calls
rmp_mark_pages_firmware() with locked=false while the parent
function of init_ex_path already acquired the sev_cmd_mutex.
In the case of an RMPUPDATE failure for any page after the first, the cleanup
path would invoke reclaim pages which would result in a deadlock in
sev_do_cmd.

Pass locked=true to honor the lock status of the parent function.

Fixes: 7364a6fbca45 ("crypto: ccp: Handle non-volatile INIT_EX data when SNP is enabled")

Reported-by: Chris Mason <clm@meta.com>
Assisted-by: Claude:claude-opus-4-6
Fixes: 7364a6fbca45 ("crypto: ccp: Handle non-volatile INIT_EX data when SNP is enabled")
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Atish Patra <atishp@meta.com>
---
 drivers/crypto/ccp/sev-dev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/crypto/ccp/sev-dev.c b/drivers/crypto/ccp/sev-dev.c
index d1e9e0ac63b6..3d4793e8e34b 100644
--- a/drivers/crypto/ccp/sev-dev.c
+++ b/drivers/crypto/ccp/sev-dev.c
@@ -1557,7 +1557,7 @@ static int __sev_platform_init_handle_init_ex_path(struct sev_device *sev)
 		unsigned long npages;
 
 		npages = 1UL << get_order(NV_LENGTH);
-		if (rmp_mark_pages_firmware(__pa(sev_init_ex_buffer), npages, false)) {
+		if (rmp_mark_pages_firmware(__pa(sev_init_ex_buffer), npages, true)) {
 			dev_err(sev->dev, "SEV: INIT_EX NV memory page state change failed.\n");
 			return -ENOMEM;
 		}

-- 
2.53.0-Meta

[PATCH v3 4/4] crypto: ccp: Fix memory leak in SEV INIT_EX path

[Atish Patra]

From: Atish Patra <atishp@meta.com>

allocated pages in _init_ext_path are never freed and sev_init_ex_buffer
is left pointing at the leaked memory in case of any failures during the
function..

Fix by adding an error path that frees the pages and clears
sev_init_ex_buffer. Make sure we only free the memory if the failure
happens before the conversion. Otherwise, we may end up trying to free
up converted pages in case of reclaim failure. rmp_mark_pages_firmware
failures should be rare enough to avoid more code complexity to track
down which pages were reclaimed/leaked vs which are not.

Fixes: 7364a6fbca45 ("crypto: ccp: Handle non-volatile INIT_EX data when SNP is enabled")

Reported-by: Sashiko <sashiko-bot@kernel.org>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
Signed-off-by: Atish Patra <atishp@meta.com>
---
 drivers/crypto/ccp/sev-dev.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/drivers/crypto/ccp/sev-dev.c b/drivers/crypto/ccp/sev-dev.c
index 3d4793e8e34b..57b4c1e79589 100644
--- a/drivers/crypto/ccp/sev-dev.c
+++ b/drivers/crypto/ccp/sev-dev.c
@@ -1540,7 +1540,7 @@ static int __sev_platform_init_handle_init_ex_path(struct sev_device *sev)
 	if (sev_init_ex_buffer)
 		return 0;
 
-	page = alloc_pages(GFP_KERNEL, get_order(NV_LENGTH));
+	page = alloc_pages(GFP_KERNEL | __GFP_ZERO, get_order(NV_LENGTH));
 	if (!page) {
 		dev_err(sev->dev, "SEV: INIT_EX NV memory allocation failed\n");
 		return -ENOMEM;
@@ -1550,7 +1550,7 @@ static int __sev_platform_init_handle_init_ex_path(struct sev_device *sev)
 
 	rc = sev_read_init_ex_file();
 	if (rc)
-		return rc;
+		goto err_free;
 
 	/* If SEV-SNP is initialized, transition to firmware page. */
 	if (sev->snp_initialized) {
@@ -1559,11 +1559,22 @@ static int __sev_platform_init_handle_init_ex_path(struct sev_device *sev)
 		npages = 1UL << get_order(NV_LENGTH);
 		if (rmp_mark_pages_firmware(__pa(sev_init_ex_buffer), npages, true)) {
 			dev_err(sev->dev, "SEV: INIT_EX NV memory page state change failed.\n");
-			return -ENOMEM;
+			rc = -ENOMEM;
+			/*
+			 * Pages can be in an inconsistent state, don't release them back to the
+			 * system.
+			 */
+			goto err_reset;
 		}
 	}
 
 	return 0;
+
+err_free:
+	__free_pages(page, get_order(NV_LENGTH));
+err_reset:
+	sev_init_ex_buffer = NULL;
+	return rc;
 }
 
 static int __sev_platform_init_locked(int *error)

-- 
2.53.0-Meta

Re: [PATCH v3 3/4] crypto: ccp: Fix possible deadlock in SEV init failure path

[Herbert Xu]

On Tue, Jun 02, 2026 at 03:36:34PM -0700, Atish Patra wrote:
> From: Atish Patra <atishp@meta.com>
> 
> __sev_platform_init_handle_init_ex_path() calls
> rmp_mark_pages_firmware() with locked=false while the parent
> function of init_ex_path already acquired the sev_cmd_mutex.
> In the case of an RMPUPDATE failure for any page after the first, the cleanup
> path would invoke reclaim pages which would result in a deadlock in
> sev_do_cmd.
> 
> Pass locked=true to honor the lock status of the parent function.
> 
> Fixes: 7364a6fbca45 ("crypto: ccp: Handle non-volatile INIT_EX data when SNP is enabled")
> 
> Reported-by: Chris Mason <clm@meta.com>
> Assisted-by: Claude:claude-opus-4-6
> Fixes: 7364a6fbca45 ("crypto: ccp: Handle non-volatile INIT_EX data when SNP is enabled")
> Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
> Signed-off-by: Atish Patra <atishp@meta.com>
> ---
>  drivers/crypto/ccp/sev-dev.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)

Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Re: [PATCH v3 4/4] crypto: ccp: Fix memory leak in SEV INIT_EX path

[Herbert Xu]

On Tue, Jun 02, 2026 at 03:36:35PM -0700, Atish Patra wrote:
> From: Atish Patra <atishp@meta.com>
> 
> allocated pages in _init_ext_path are never freed and sev_init_ex_buffer
> is left pointing at the leaked memory in case of any failures during the
> function..
> 
> Fix by adding an error path that frees the pages and clears
> sev_init_ex_buffer. Make sure we only free the memory if the failure
> happens before the conversion. Otherwise, we may end up trying to free
> up converted pages in case of reclaim failure. rmp_mark_pages_firmware
> failures should be rare enough to avoid more code complexity to track
> down which pages were reclaimed/leaked vs which are not.
> 
> Fixes: 7364a6fbca45 ("crypto: ccp: Handle non-volatile INIT_EX data when SNP is enabled")
> 
> Reported-by: Sashiko <sashiko-bot@kernel.org>
> Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
> Signed-off-by: Atish Patra <atishp@meta.com>
> ---
>  drivers/crypto/ccp/sev-dev.c | 17 ++++++++++++++---
>  1 file changed, 14 insertions(+), 3 deletions(-)

Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Re: [PATCH v3 3/4] crypto: ccp: Fix possible deadlock in SEV init failure path

[Sean Christopherson]

On Fri, Jul 03, 2026, Herbert Xu wrote:
> On Tue, Jun 02, 2026 at 03:36:34PM -0700, Atish Patra wrote:
> > From: Atish Patra <atishp@meta.com>
> > 
> > __sev_platform_init_handle_init_ex_path() calls
> > rmp_mark_pages_firmware() with locked=false while the parent
> > function of init_ex_path already acquired the sev_cmd_mutex.
> > In the case of an RMPUPDATE failure for any page after the first, the cleanup
> > path would invoke reclaim pages which would result in a deadlock in
> > sev_do_cmd.
> > 
> > Pass locked=true to honor the lock status of the parent function.
> > 
> > Fixes: 7364a6fbca45 ("crypto: ccp: Handle non-volatile INIT_EX data when SNP is enabled")
> > 
> > Reported-by: Chris Mason <clm@meta.com>
> > Assisted-by: Claude:claude-opus-4-6
> > Fixes: 7364a6fbca45 ("crypto: ccp: Handle non-volatile INIT_EX data when SNP is enabled")
> > Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
> > Signed-off-by: Atish Patra <atishp@meta.com>
> > ---
> >  drivers/crypto/ccp/sev-dev.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> Acked-by: Herbert Xu <herbert@gondor.apana.org.au>

Want to take patches 3 and 4 through your tree?  I'll grab 1 and 2; there's no
dependency between the KVM changes and the crypto changes.