Skip to contentSkip to conversation

KVM: s390: Check for invalid bits in kvm_valid_regs and kvm_dirty_regs

PATCH In review Thomas Huth 4 participants 11 messages Sep 4, 2019 – Sep 5, 2019

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

v2 cover KVM: s390: Check for invalid bits in kvm_valid_regs and kvm_dirty_regs

Avoid invalid bits in kvm_valid_regs and kvm_dirty_regs on s390x.

v2:
- Split the single patch from v1 into two separate patches (I've kept the Reviewed-bys from v1, but if you don't agree with the patch description of the 2nd patch, please complain)

Thomas Huth (2):
KVM: s390: Disallow invalid bits in kvm_valid_regs and kvm_dirty_regs
KVM: selftests: Test invalid bits in kvm_valid_regs and kvm_dirty_regs

    on s390x
 arch/s390/include/uapi/asm/kvm.h              |  6 ++++
 arch/s390/kvm/kvm-s390.c                      |  4 +++
 .../selftests/kvm/s390x/sync_regs_test.c      | 30 +++++++++++++++++++
 3 files changed, 40 insertions(+)
Signature · 1 line
2.18.1
#1 lore raw

PATCH v2 1/2 KVM: s390: Disallow invalid bits in kvm_valid_regs and kvm_dirty_regs

If unknown bits are set in kvm_valid_regs or kvm_dirty_regs, this clearly indicates that something went wrong in the KVM userspace application. The x86 variant of KVM already contains a check for bad bits, so let's do the same on s390x now, too.

---
 arch/s390/include/uapi/asm/kvm.h | 6 ++++++
 arch/s390/kvm/kvm-s390.c         | 4 ++++
 2 files changed, 10 insertions(+)
arch/s390/include/uapi/asm/kvm.h +6 -0
index 47104e5b47fd..436ec7636927 100644
@@ -231,6 +231,12 @@ struct kvm_guest_debug_arch {
 #define KVM_SYNC_GSCB   (1UL << 9)
 #define KVM_SYNC_BPBC   (1UL << 10)
 #define KVM_SYNC_ETOKEN (1UL << 11)
+
+#define KVM_SYNC_S390_VALID_FIELDS \
+	(KVM_SYNC_PREFIX | KVM_SYNC_GPRS | KVM_SYNC_ACRS | KVM_SYNC_CRS | \
+	 KVM_SYNC_ARCH0 | KVM_SYNC_PFAULT | KVM_SYNC_VRS | KVM_SYNC_RICCB | \
+	 KVM_SYNC_FPRS | KVM_SYNC_GSCB | KVM_SYNC_BPBC | KVM_SYNC_ETOKEN)
+
 /* length and alignment of the sdnx as a power of two */
 #define SDNXC 8
 #define SDNXL (1UL << SDNXC)
arch/s390/kvm/kvm-s390.c +4 -0
index 49d7722229ae..a7d7dedfe527 100644
@@ -3998,6 +3998,10 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	if (kvm_run->immediate_exit)
 		return -EINTR;
 
+	if (kvm_run->kvm_valid_regs & ~KVM_SYNC_S390_VALID_FIELDS ||
+	    kvm_run->kvm_dirty_regs & ~KVM_SYNC_S390_VALID_FIELDS)
+		return -EINVAL;
+
 	vcpu_load(vcpu);
 
 	if (guestdbg_exit_pending(vcpu)) {
Signature · 1 line
2.18.1
Reviewed-by on v2 1/2
#2 lore raw

Re: [PATCH v2 1/2] KVM: s390: Disallow invalid bits in kvm_valid_regs and kvm_dirty_regs

Quoted 19 lines · Thomas Huth

If unknown bits are set in kvm_valid_regs or kvm_dirty_regs, this clearly indicates that something went wrong in the KVM userspace application. The x86 variant of KVM already contains a check for bad bits, so let's do the same on s390x now, too.

---
 arch/s390/include/uapi/asm/kvm.h | 6 ++++++
 arch/s390/kvm/kvm-s390.c         | 4 ++++
 2 files changed, 10 insertions(+)
arch/s390/include/uapi/asm/kvm.h +0 -0
index 47104e5b47fd..436ec7636927 100644

arch/s390/include/uapi/asm/kvm.h struct kvm_guest_debug_arch { · line 239

Quoted 9 lines
arch/s390/include/uapi/asm/kvm.h +6 -0
@@ -231,6 +231,12 @@ struct kvm_guest_debug_arch {
 #define KVM_SYNC_GSCB   (1UL << 9)
 #define KVM_SYNC_BPBC   (1UL << 10)
 #define KVM_SYNC_ETOKEN (1UL << 11)
+
+#define KVM_SYNC_S390_VALID_FIELDS \
+	(KVM_SYNC_PREFIX | KVM_SYNC_GPRS | KVM_SYNC_ACRS | KVM_SYNC_CRS | \
+	 KVM_SYNC_ARCH0 | KVM_SYNC_PFAULT | KVM_SYNC_VRS | KVM_SYNC_RICCB | \
+	 KVM_SYNC_FPRS | KVM_SYNC_GSCB | KVM_SYNC_BPBC | KVM_SYNC_ETOKEN)
+

We didn't care about the S390 for the actual flags, why care now?

Quoted 9 lines
arch/s390/include/uapi/asm/kvm.h +0 -0
 /* length and alignment of the sdnx as a power of two */
 #define SDNXC 8
 #define SDNXL (1UL << SDNXC)
arch/s390/kvm/kvm-s390.c +0 -0
index 49d7722229ae..a7d7dedfe527 100644

arch/s390/kvm/kvm-s390.c int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) · line 4007

Quoted 10 lines
arch/s390/kvm/kvm-s390.c +4 -0
@@ -3998,6 +3998,10 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	if (kvm_run->immediate_exit)
 		return -EINTR;
 
+	if (kvm_run->kvm_valid_regs & ~KVM_SYNC_S390_VALID_FIELDS ||
+	    kvm_run->kvm_dirty_regs & ~KVM_SYNC_S390_VALID_FIELDS)
+		return -EINVAL;
+
 	vcpu_load(vcpu);
 
 	if (guestdbg_exit_pending(vcpu)) {
Signature · 3 lines
Thanks,

David / dhildenb
#3 lore raw

Re: [PATCH v2 1/2] KVM: s390: Disallow invalid bits in kvm_valid_regs and kvm_dirty_regs

Quoted 15 lines · David Hildenbrand
Quoted 14 lines · Thomas Huth

If unknown bits are set in kvm_valid_regs or kvm_dirty_regs, this clearly indicates that something went wrong in the KVM userspace application. The x86 variant of KVM already contains a check for bad bits, so let's do the same on s390x now, too.

---
 arch/s390/include/uapi/asm/kvm.h | 6 ++++++
 arch/s390/kvm/kvm-s390.c         | 4 ++++
 2 files changed, 10 insertions(+)
Quoted 16 lines
Quoted 5 lines
arch/s390/include/uapi/asm/kvm.h +0 -0
index 47104e5b47fd..436ec7636927 100644

arch/s390/include/uapi/asm/kvm.h struct kvm_guest_debug_arch { · line 239

Quoted 9 lines
arch/s390/include/uapi/asm/kvm.h +6 -0
@@ -231,6 +231,12 @@ struct kvm_guest_debug_arch {
 #define KVM_SYNC_GSCB   (1UL << 9)
 #define KVM_SYNC_BPBC   (1UL << 10)
 #define KVM_SYNC_ETOKEN (1UL << 11)
+
+#define KVM_SYNC_S390_VALID_FIELDS \
+	(KVM_SYNC_PREFIX | KVM_SYNC_GPRS | KVM_SYNC_ACRS | KVM_SYNC_CRS | \
+	 KVM_SYNC_ARCH0 | KVM_SYNC_PFAULT | KVM_SYNC_VRS | KVM_SYNC_RICCB | \
+	 KVM_SYNC_FPRS | KVM_SYNC_GSCB | KVM_SYNC_BPBC | KVM_SYNC_ETOKEN)
+

We didn't care about the S390 for the actual flags, why care now?

x86 does the same, and we don't want to be worse than x86, do we? ;-)

Honestly, this was just one of the differences that I noticed while porting the sync_regs_test from x86 to s390x.

Thomas

#4 lore raw

Re: [PATCH v2 1/2] KVM: s390: Disallow invalid bits in kvm_valid_regs and kvm_dirty_regs

Quoted 33 lines · Thomas Huth
Quoted 15 lines · David Hildenbrand
Quoted 14 lines · Thomas Huth

If unknown bits are set in kvm_valid_regs or kvm_dirty_regs, this clearly indicates that something went wrong in the KVM userspace application. The x86 variant of KVM already contains a check for bad bits, so let's do the same on s390x now, too.

---
 arch/s390/include/uapi/asm/kvm.h | 6 ++++++
 arch/s390/kvm/kvm-s390.c         | 4 ++++
 2 files changed, 10 insertions(+)
Quoted 16 lines
Quoted 5 lines
arch/s390/include/uapi/asm/kvm.h +0 -0
index 47104e5b47fd..436ec7636927 100644

arch/s390/include/uapi/asm/kvm.h struct kvm_guest_debug_arch { · line 239

Quoted 9 lines
arch/s390/include/uapi/asm/kvm.h +6 -0
@@ -231,6 +231,12 @@ struct kvm_guest_debug_arch {
 #define KVM_SYNC_GSCB   (1UL << 9)
 #define KVM_SYNC_BPBC   (1UL << 10)
 #define KVM_SYNC_ETOKEN (1UL << 11)
+
+#define KVM_SYNC_S390_VALID_FIELDS \
+	(KVM_SYNC_PREFIX | KVM_SYNC_GPRS | KVM_SYNC_ACRS | KVM_SYNC_CRS | \
+	 KVM_SYNC_ARCH0 | KVM_SYNC_PFAULT | KVM_SYNC_VRS | KVM_SYNC_RICCB | \
+	 KVM_SYNC_FPRS | KVM_SYNC_GSCB | KVM_SYNC_BPBC | KVM_SYNC_ETOKEN)
+

We didn't care about the S390 for the actual flags, why care now?

Quoted 1 line

x86 does the same, and we don't want to be worse than x86, do we? ;-)

Yeah, but they do have X86 in every flag - in contrast to us.

Quoted 6 lines

Honestly, this was just one of the differences that I noticed while porting the sync_regs_test from x86 to s390x.

Thomas

Signature · 3 lines
Thanks,

David / dhildenb
#5 lore raw

Re: [PATCH v2 1/2] KVM: s390: Disallow invalid bits in kvm_valid_regs and kvm_dirty_regs

In reply to David Hildenbrand
Quoted 15 lines · David Hildenbrand
Quoted 14 lines · Thomas Huth

If unknown bits are set in kvm_valid_regs or kvm_dirty_regs, this clearly indicates that something went wrong in the KVM userspace application. The x86 variant of KVM already contains a check for bad bits, so let's do the same on s390x now, too.

---
 arch/s390/include/uapi/asm/kvm.h | 6 ++++++
 arch/s390/kvm/kvm-s390.c         | 4 ++++
 2 files changed, 10 insertions(+)
Quoted 16 lines
Quoted 5 lines
arch/s390/include/uapi/asm/kvm.h +0 -0
index 47104e5b47fd..436ec7636927 100644

arch/s390/include/uapi/asm/kvm.h struct kvm_guest_debug_arch { · line 239

Quoted 9 lines
arch/s390/include/uapi/asm/kvm.h +6 -0
@@ -231,6 +231,12 @@ struct kvm_guest_debug_arch {
 #define KVM_SYNC_GSCB   (1UL << 9)
 #define KVM_SYNC_BPBC   (1UL << 10)
 #define KVM_SYNC_ETOKEN (1UL << 11)
+
+#define KVM_SYNC_S390_VALID_FIELDS \
+	(KVM_SYNC_PREFIX | KVM_SYNC_GPRS | KVM_SYNC_ACRS | KVM_SYNC_CRS | \
+	 KVM_SYNC_ARCH0 | KVM_SYNC_PFAULT | KVM_SYNC_VRS | KVM_SYNC_RICCB | \
+	 KVM_SYNC_FPRS | KVM_SYNC_GSCB | KVM_SYNC_BPBC | KVM_SYNC_ETOKEN)
+

We didn't care about the S390 for the actual flags, why care now?

I think it makes sense to have the interface as defined as possible. If for some reason userspace sets a wrong bit this would be undetected. If we at a later point in time use that bit this would resultin strange problems.

Quoted 24 lines
Quoted 11 lines
arch/s390/include/uapi/asm/kvm.h +0 -0
 /* length and alignment of the sdnx as a power of two */
 #define SDNXC 8
 #define SDNXL (1UL << SDNXC)
arch/s390/kvm/kvm-s390.c +0 -0
index 49d7722229ae..a7d7dedfe527 100644

arch/s390/kvm/kvm-s390.c int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) · line 4007

Quoted 10 lines
arch/s390/kvm/kvm-s390.c +4 -0
@@ -3998,6 +3998,10 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 	if (kvm_run->immediate_exit)
 		return -EINTR;
 
+	if (kvm_run->kvm_valid_regs & ~KVM_SYNC_S390_VALID_FIELDS ||
+	    kvm_run->kvm_dirty_regs & ~KVM_SYNC_S390_VALID_FIELDS)
+		return -EINVAL;
+
 	vcpu_load(vcpu);
 
 	if (guestdbg_exit_pending(vcpu)) {
#6 lore raw

Re: [PATCH v2 1/2] KVM: s390: Disallow invalid bits in kvm_valid_regs and kvm_dirty_regs

Quoted 35 lines · Christian Borntraeger
Quoted 15 lines · David Hildenbrand
Quoted 14 lines · Thomas Huth

If unknown bits are set in kvm_valid_regs or kvm_dirty_regs, this clearly indicates that something went wrong in the KVM userspace application. The x86 variant of KVM already contains a check for bad bits, so let's do the same on s390x now, too.

---
 arch/s390/include/uapi/asm/kvm.h | 6 ++++++
 arch/s390/kvm/kvm-s390.c         | 4 ++++
 2 files changed, 10 insertions(+)
Quoted 16 lines
Quoted 5 lines
arch/s390/include/uapi/asm/kvm.h +0 -0
index 47104e5b47fd..436ec7636927 100644

arch/s390/include/uapi/asm/kvm.h struct kvm_guest_debug_arch { · line 239

Quoted 9 lines
arch/s390/include/uapi/asm/kvm.h +6 -0
@@ -231,6 +231,12 @@ struct kvm_guest_debug_arch {
 #define KVM_SYNC_GSCB   (1UL << 9)
 #define KVM_SYNC_BPBC   (1UL << 10)
 #define KVM_SYNC_ETOKEN (1UL << 11)
+
+#define KVM_SYNC_S390_VALID_FIELDS \
+	(KVM_SYNC_PREFIX | KVM_SYNC_GPRS | KVM_SYNC_ACRS | KVM_SYNC_CRS | \
+	 KVM_SYNC_ARCH0 | KVM_SYNC_PFAULT | KVM_SYNC_VRS | KVM_SYNC_RICCB | \
+	 KVM_SYNC_FPRS | KVM_SYNC_GSCB | KVM_SYNC_BPBC | KVM_SYNC_ETOKEN)
+

We didn't care about the S390 for the actual flags, why care now?

Quoted 3 lines

I think it makes sense to have the interface as defined as possible. If for some reason userspace sets a wrong bit this would be undetected. If we at a later point in time use that bit this would resultin strange problems.

Not arguing about the concept of checking for valid bits. Was just wondering if the "S390" part in the name makes sense at all. But you guys seem to have a consent here.

Signature · 3 lines
Thanks,

David / dhildenb
#7 lore raw

Re: [PATCH v2 1/2] KVM: s390: Disallow invalid bits in kvm_valid_regs and kvm_dirty_regs

Quoted 36 lines · David Hildenbrand
Quoted 35 lines · Christian Borntraeger
Quoted 31 lines · David Hildenbrand
Quoted 19 lines · Thomas Huth

If unknown bits are set in kvm_valid_regs or kvm_dirty_regs, this clearly indicates that something went wrong in the KVM userspace application. The x86 variant of KVM already contains a check for bad bits, so let's do the same on s390x now, too.

---
 arch/s390/include/uapi/asm/kvm.h | 6 ++++++
 arch/s390/kvm/kvm-s390.c         | 4 ++++
 2 files changed, 10 insertions(+)
arch/s390/include/uapi/asm/kvm.h +0 -0
index 47104e5b47fd..436ec7636927 100644

arch/s390/include/uapi/asm/kvm.h struct kvm_guest_debug_arch { · line 239

Quoted 9 lines
arch/s390/include/uapi/asm/kvm.h +6 -0
@@ -231,6 +231,12 @@ struct kvm_guest_debug_arch {
 #define KVM_SYNC_GSCB   (1UL << 9)
 #define KVM_SYNC_BPBC   (1UL << 10)
 #define KVM_SYNC_ETOKEN (1UL << 11)
+
+#define KVM_SYNC_S390_VALID_FIELDS \
+	(KVM_SYNC_PREFIX | KVM_SYNC_GPRS | KVM_SYNC_ACRS | KVM_SYNC_CRS | \
+	 KVM_SYNC_ARCH0 | KVM_SYNC_PFAULT | KVM_SYNC_VRS | KVM_SYNC_RICCB | \
+	 KVM_SYNC_FPRS | KVM_SYNC_GSCB | KVM_SYNC_BPBC | KVM_SYNC_ETOKEN)
+

We didn't care about the S390 for the actual flags, why care now?

Quoted 7 lines
Quoted 3 lines

I think it makes sense to have the interface as defined as possible. If for some reason userspace sets a wrong bit this would be undetected. If we at a later point in time use that bit this would resultin strange problems.

Not arguing about the concept of checking for valid bits. Was just wondering if the "S390" part in the name makes sense at all. But you guys seem to have a consent here.

Oh, I guess we both got your question wrong... Well, I don't care too much whether we've got an "S390" in there or not ... so unless there is a real good reason to remove it, let's simply keep it this way.

Thomas

#8 lore raw

PATCH v2 2/2 KVM: selftests: Test invalid bits in kvm_valid_regs and kvm_dirty_regs on s390x

In reply to Thomas Huth

Now that we disallow invalid bits in kvm_valid_regs and kvm_dirty_regs on s390x, too, we should also check this condition in the selftests. The code has been taken from the x86-version of the sync_regs_test.

---
 .../selftests/kvm/s390x/sync_regs_test.c      | 30 +++++++++++++++++++
 1 file changed, 30 insertions(+)
tools/testing/selftests/kvm/s390x/sync_regs_test.c +30 -0
index bbc93094519b..d5290b4ad636 100644
@@ -85,6 +85,36 @@ int main(int argc, char *argv[])
 
 	run = vcpu_state(vm, VCPU_ID);
 
+	/* Request reading invalid register set from VCPU. */
+	run->kvm_valid_regs = INVALID_SYNC_FIELD;
+	rv = _vcpu_run(vm, VCPU_ID);
+	TEST_ASSERT(rv < 0 && errno == EINVAL,
+		    "Invalid kvm_valid_regs did not cause expected KVM_RUN error: %d\n",
+		    rv);
+	vcpu_state(vm, VCPU_ID)->kvm_valid_regs = 0;
+
+	run->kvm_valid_regs = INVALID_SYNC_FIELD | TEST_SYNC_FIELDS;
+	rv = _vcpu_run(vm, VCPU_ID);
+	TEST_ASSERT(rv < 0 && errno == EINVAL,
+		    "Invalid kvm_valid_regs did not cause expected KVM_RUN error: %d\n",
+		    rv);
+	vcpu_state(vm, VCPU_ID)->kvm_valid_regs = 0;
+
+	/* Request setting invalid register set into VCPU. */
+	run->kvm_dirty_regs = INVALID_SYNC_FIELD;
+	rv = _vcpu_run(vm, VCPU_ID);
+	TEST_ASSERT(rv < 0 && errno == EINVAL,
+		    "Invalid kvm_dirty_regs did not cause expected KVM_RUN error: %d\n",
+		    rv);
+	vcpu_state(vm, VCPU_ID)->kvm_dirty_regs = 0;
+
+	run->kvm_dirty_regs = INVALID_SYNC_FIELD | TEST_SYNC_FIELDS;
+	rv = _vcpu_run(vm, VCPU_ID);
+	TEST_ASSERT(rv < 0 && errno == EINVAL,
+		    "Invalid kvm_dirty_regs did not cause expected KVM_RUN error: %d\n",
+		    rv);
+	vcpu_state(vm, VCPU_ID)->kvm_dirty_regs = 0;
+
 	/* Request and verify all valid register sets. */
 	run->kvm_valid_regs = TEST_SYNC_FIELDS;
 	rv = _vcpu_run(vm, VCPU_ID);
Signature · 1 line
2.18.1
#9 lore raw

Re: [PATCH v2 2/2] KVM: selftests: Test invalid bits in kvm_valid_regs and kvm_dirty_regs on s390x

Quoted 44 lines · Thomas Huth

Now that we disallow invalid bits in kvm_valid_regs and kvm_dirty_regs on s390x, too, we should also check this condition in the selftests. The code has been taken from the x86-version of the sync_regs_test.

---
 .../selftests/kvm/s390x/sync_regs_test.c      | 30 +++++++++++++++++++
 1 file changed, 30 insertions(+)
tools/testing/selftests/kvm/s390x/sync_regs_test.c +23 -0
index bbc93094519b..d5290b4ad636 100644
@@ -85,6 +85,36 @@ int main(int argc, char *argv[])
 
 	run = vcpu_state(vm, VCPU_ID);
 
+	/* Request reading invalid register set from VCPU. */
+	run->kvm_valid_regs = INVALID_SYNC_FIELD;
+	rv = _vcpu_run(vm, VCPU_ID);
+	TEST_ASSERT(rv < 0 && errno == EINVAL,
+		    "Invalid kvm_valid_regs did not cause expected KVM_RUN error: %d\n",
+		    rv);
+	vcpu_state(vm, VCPU_ID)->kvm_valid_regs = 0;
+
+	run->kvm_valid_regs = INVALID_SYNC_FIELD | TEST_SYNC_FIELDS;
+	rv = _vcpu_run(vm, VCPU_ID);
+	TEST_ASSERT(rv < 0 && errno == EINVAL,
+		    "Invalid kvm_valid_regs did not cause expected KVM_RUN error: %d\n",
+		    rv);
+	vcpu_state(vm, VCPU_ID)->kvm_valid_regs = 0;
+
+	/* Request setting invalid register set into VCPU. */
+	run->kvm_dirty_regs = INVALID_SYNC_FIELD;
+	rv = _vcpu_run(vm, VCPU_ID);
+	TEST_ASSERT(rv < 0 && errno == EINVAL,
+		    "Invalid kvm_dirty_regs did not cause expected KVM_RUN error: %d\n",
+		    rv);
+	vcpu_state(vm, VCPU_ID)->kvm_dirty_regs = 0;
+

tools/testing/selftests/kvm/s390x/sync_regs_test.c int main(int argc, char *argv[]) · line 120

Quoted 10 lines
tools/testing/selftests/kvm/s390x/sync_regs_test.c +7 -0
+	run->kvm_dirty_regs = INVALID_SYNC_FIELD | TEST_SYNC_FIELDS;
+	rv = _vcpu_run(vm, VCPU_ID);
+	TEST_ASSERT(rv < 0 && errno == EINVAL,
+		    "Invalid kvm_dirty_regs did not cause expected KVM_RUN error: %d\n",
+		    rv);
+	vcpu_state(vm, VCPU_ID)->kvm_dirty_regs = 0;
+
 	/* Request and verify all valid register sets. */
 	run->kvm_valid_regs = TEST_SYNC_FIELDS;
 	rv = _vcpu_run(vm, VCPU_ID);

Looks sane to me.

Signature · 3 lines
Thanks,

David / dhildenb
maintainer
#10 lore raw
In reply to Thomas Huth
Quoted 10 lines · Thomas Huth

Avoid invalid bits in kvm_valid_regs and kvm_dirty_regs on s390x.

v2:
- Split the single patch from v1 into two separate patches (I've kept the Reviewed-bys from v1, but if you don't agree with the patch description of the 2nd patch, please complain)

Thomas Huth (2):
KVM: s390: Disallow invalid bits in kvm_valid_regs and kvm_dirty_regs
KVM: selftests: Test invalid bits in kvm_valid_regs and kvm_dirty_regs

Quoted 7 lines
    on s390x
 arch/s390/include/uapi/asm/kvm.h              |  6 ++++
 arch/s390/kvm/kvm-s390.c                      |  4 +++
 .../selftests/kvm/s390x/sync_regs_test.c      | 30 +++++++++++++++++++
 3 files changed, 40 insertions(+)

Thanks, both are picked