Skip to contentSkip to conversation

selftests: kvm: Check whether SIDA memop fails for normal guests

PATCH Applied Thomas Huth 5 participants 8 messages Feb 15, 2022 – Feb 21, 2022

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 selftests: kvm: Check whether SIDA memop fails for normal guests

Commit 2c212e1baedc ("KVM: s390: Return error on SIDA memop on normal guest") fixed the behavior of the SIDA memops for normal guests. It would be nice to have a way to test whether the current kernel has the fix applied or not. Thus add a check to the KVM selftests for these two memops.

---
 tools/testing/selftests/kvm/s390x/memop.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
tools/testing/selftests/kvm/s390x/memop.c +15 -0
index 9f49ead380ab..d19c3ffdea3f 100644
@@ -160,6 +160,21 @@ int main(int argc, char *argv[])
 	run->psw_mask &= ~(3UL << (63 - 17));   /* Disable AR mode */
 	vcpu_run(vm, VCPU_ID);                  /* Run to sync new state */
 
+	/* Check that the SIDA calls are rejected for non-protected guests */
+	ksmo.gaddr = 0;
+	ksmo.flags = 0;
+	ksmo.size = 8;
+	ksmo.op = KVM_S390_MEMOP_SIDA_READ;
+	ksmo.buf = (uintptr_t)mem1;
+	ksmo.sida_offset = 0x1c0;
+	rv = _vcpu_ioctl(vm, VCPU_ID, KVM_S390_MEM_OP, &ksmo);
+	TEST_ASSERT(rv == -1 && errno == EINVAL,
+		    "ioctl does not reject SIDA_READ in non-protected mode");
+	ksmo.op = KVM_S390_MEMOP_SIDA_WRITE;
+	rv = _vcpu_ioctl(vm, VCPU_ID, KVM_S390_MEM_OP, &ksmo);
+	TEST_ASSERT(rv == -1 && errno == EINVAL,
+		    "ioctl does not reject SIDA_WRITE in non-protected mode");
+
 	kvm_vm_free(vm);
 
 	return 0;
Signature · 1 line
2.27.0
Reviewed-by on v1 1/1
#1 lore raw
Quoted 7 lines · On Tue, 15 Feb 2022 08:48:24 +0100 Thomas Huth <[email protected]> wrote:

Commit 2c212e1baedc ("KVM: s390: Return error on SIDA memop on normal guest") fixed the behavior of the SIDA memops for normal guests. It would be nice to have a way to test whether the current kernel has the fix applied or not. Thus add a check to the KVM selftests for these two memops.

looks rather straightforward

Quoted 30 lines
---
 tools/testing/selftests/kvm/s390x/memop.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
tools/testing/selftests/kvm/s390x/memop.c +15 -0
index 9f49ead380ab..d19c3ffdea3f 100644
@@ -160,6 +160,21 @@ int main(int argc, char *argv[])
 	run->psw_mask &= ~(3UL << (63 - 17));   /* Disable AR mode */
 	vcpu_run(vm, VCPU_ID);                  /* Run to sync new state */
 
+	/* Check that the SIDA calls are rejected for non-protected guests */
+	ksmo.gaddr = 0;
+	ksmo.flags = 0;
+	ksmo.size = 8;
+	ksmo.op = KVM_S390_MEMOP_SIDA_READ;
+	ksmo.buf = (uintptr_t)mem1;
+	ksmo.sida_offset = 0x1c0;
+	rv = _vcpu_ioctl(vm, VCPU_ID, KVM_S390_MEM_OP, &ksmo);
+	TEST_ASSERT(rv == -1 && errno == EINVAL,
+		    "ioctl does not reject SIDA_READ in non-protected mode");
+	ksmo.op = KVM_S390_MEMOP_SIDA_WRITE;
+	rv = _vcpu_ioctl(vm, VCPU_ID, KVM_S390_MEM_OP, &ksmo);
+	TEST_ASSERT(rv == -1 && errno == EINVAL,
+		    "ioctl does not reject SIDA_WRITE in non-protected mode");
+
 	kvm_vm_free(vm);
 
 	return 0;
#2 lore raw
In reply to Thomas Huth
Quoted 16 lines · Thomas Huth

Commit 2c212e1baedc ("KVM: s390: Return error on SIDA memop on normal guest") fixed the behavior of the SIDA memops for normal guests. It would be nice to have a way to test whether the current kernel has the fix applied or not. Thus add a check to the KVM selftests for these two memops.

---
 tools/testing/selftests/kvm/s390x/memop.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
tools/testing/selftests/kvm/s390x/memop.c +0 -0
index 9f49ead380ab..d19c3ffdea3f 100644

tools/testing/selftests/kvm/s390x/memop.c int main(int argc, char *argv[]) · line 169

Quoted 10 lines
tools/testing/selftests/kvm/s390x/memop.c +7 -0
@@ -160,6 +160,21 @@ int main(int argc, char *argv[])
 	run->psw_mask &= ~(3UL << (63 - 17));   /* Disable AR mode */
 	vcpu_run(vm, VCPU_ID);                  /* Run to sync new state */
 
+	/* Check that the SIDA calls are rejected for non-protected guests */
+	ksmo.gaddr = 0;
+	ksmo.flags = 0;
+	ksmo.size = 8;
+	ksmo.op = KVM_S390_MEMOP_SIDA_READ;
+	ksmo.buf = (uintptr_t)mem1;
+	ksmo.sida_offset = 0x1c0;

What is the rational for that constant?
Any would do, as long as size + offset < PAGE_SIZE, correct?

Quoted 11 lines
tools/testing/selftests/kvm/s390x/memop.c +8 -0
+	rv = _vcpu_ioctl(vm, VCPU_ID, KVM_S390_MEM_OP, &ksmo);
+	TEST_ASSERT(rv == -1 && errno == EINVAL,
+		    "ioctl does not reject SIDA_READ in non-protected mode");
+	ksmo.op = KVM_S390_MEMOP_SIDA_WRITE;
+	rv = _vcpu_ioctl(vm, VCPU_ID, KVM_S390_MEM_OP, &ksmo);
+	TEST_ASSERT(rv == -1 && errno == EINVAL,
+		    "ioctl does not reject SIDA_WRITE in non-protected mode");
+
 	kvm_vm_free(vm);
 
 	return 0;
#3 lore raw
Quoted 20 lines · Janis Schoetterl-Glausch
Quoted 19 lines · Thomas Huth

Commit 2c212e1baedc ("KVM: s390: Return error on SIDA memop on normal guest") fixed the behavior of the SIDA memops for normal guests. It would be nice to have a way to test whether the current kernel has the fix applied or not. Thus add a check to the KVM selftests for these two memops.

---
  tools/testing/selftests/kvm/s390x/memop.c | 15 +++++++++++++++
  1 file changed, 15 insertions(+)
tools/testing/selftests/kvm/s390x/memop.c +0 -0
index 9f49ead380ab..d19c3ffdea3f 100644
@@ -160,6 +160,21 @@ int main(int argc, char *argv[])
  	run->psw_mask &= ~(3UL << (63 - 17));   /* Disable AR mode */
  	vcpu_run(vm, VCPU_ID);                  /* Run to sync new state */
 
Quoted 10 lines

tools/testing/selftests/kvm/s390x/memop.c int main(int argc, char *argv[]) · line 169

Quoted 7 lines
tools/testing/selftests/kvm/s390x/memop.c +7 -0
+	/* Check that the SIDA calls are rejected for non-protected guests */
+	ksmo.gaddr = 0;
+	ksmo.flags = 0;
+	ksmo.size = 8;
+	ksmo.op = KVM_S390_MEMOP_SIDA_READ;
+	ksmo.buf = (uintptr_t)mem1;
+	ksmo.sida_offset = 0x1c0;

What is the rational for that constant?
Any would do, as long as size + offset < PAGE_SIZE, correct?

Right, it's rather a random value.

Thomas

#4 lore raw
Quoted 32 lines · Thomas Huth
Quoted 20 lines · Janis Schoetterl-Glausch
Quoted 19 lines · Thomas Huth

Commit 2c212e1baedc ("KVM: s390: Return error on SIDA memop on normal guest") fixed the behavior of the SIDA memops for normal guests. It would be nice to have a way to test whether the current kernel has the fix applied or not. Thus add a check to the KVM selftests for these two memops.

---
  tools/testing/selftests/kvm/s390x/memop.c | 15 +++++++++++++++

  1 file changed, 15 insertions(+)
tools/testing/selftests/kvm/s390x/memop.c +0 -0
index 9f49ead380ab..d19c3ffdea3f 100644
@@ -160,6 +160,21 @@ int main(int argc, char *argv[])

      run->psw_mask &= ~(3UL << (63 - 17));   /* Disable AR mode */ vcpu_run(vm, VCPU_ID);                  /* Run to sync new state */

Quoted 10 lines
Quoted 7 lines

+    /* Check that the SIDA calls are rejected for non-protected guests */ +    ksmo.gaddr = 0;
+    ksmo.flags = 0;
+    ksmo.size = 8;
+    ksmo.op = KVM_S390_MEMOP_SIDA_READ;
+    ksmo.buf = (uintptr_t)mem1;
+    ksmo.sida_offset = 0x1c0;

What is the rational for that constant?
Any would do, as long as size + offset < PAGE_SIZE, correct?

Quoted 1 line

Right, it's rather a random value.

Ah, ok.

Quoted 3 lines

 Thomas

#5 lore raw
In reply to Thomas Huth

Am 15.02.22 um 08:48 schrieb Thomas Huth:

Quoted 7 lines

Commit 2c212e1baedc ("KVM: s390: Return error on SIDA memop on normal guest") fixed the behavior of the SIDA memops for normal guests. It would be nice to have a way to test whether the current kernel has the fix applied or not. Thus add a check to the KVM selftests for these two memops.

Thanks, applied.

Quoted 30 lines
---
  tools/testing/selftests/kvm/s390x/memop.c | 15 +++++++++++++++
  1 file changed, 15 insertions(+)
tools/testing/selftests/kvm/s390x/memop.c +15 -0
index 9f49ead380ab..d19c3ffdea3f 100644
@@ -160,6 +160,21 @@ int main(int argc, char *argv[])
  	run->psw_mask &= ~(3UL << (63 - 17));   /* Disable AR mode */
  	vcpu_run(vm, VCPU_ID);                  /* Run to sync new state */
 
+	/* Check that the SIDA calls are rejected for non-protected guests */
+	ksmo.gaddr = 0;
+	ksmo.flags = 0;
+	ksmo.size = 8;
+	ksmo.op = KVM_S390_MEMOP_SIDA_READ;
+	ksmo.buf = (uintptr_t)mem1;
+	ksmo.sida_offset = 0x1c0;
+	rv = _vcpu_ioctl(vm, VCPU_ID, KVM_S390_MEM_OP, &ksmo);
+	TEST_ASSERT(rv == -1 && errno == EINVAL,
+		    "ioctl does not reject SIDA_READ in non-protected mode");
+	ksmo.op = KVM_S390_MEMOP_SIDA_WRITE;
+	rv = _vcpu_ioctl(vm, VCPU_ID, KVM_S390_MEM_OP, &ksmo);
+	TEST_ASSERT(rv == -1 && errno == EINVAL,
+		    "ioctl does not reject SIDA_WRITE in non-protected mode");
+
  	kvm_vm_free(vm);
 
  	return 0;
maintainer Reviewed-by on v1 1/1
#6 lore raw
In reply to Thomas Huth
Quoted 19 lines · Thomas Huth

Commit 2c212e1baedc ("KVM: s390: Return error on SIDA memop on normal guest") fixed the behavior of the SIDA memops for normal guests. It would be nice to have a way to test whether the current kernel has the fix applied or not. Thus add a check to the KVM selftests for these two memops.

---
  tools/testing/selftests/kvm/s390x/memop.c | 15 +++++++++++++++
  1 file changed, 15 insertions(+)
tools/testing/selftests/kvm/s390x/memop.c +0 -0
index 9f49ead380ab..d19c3ffdea3f 100644
@@ -160,6 +160,21 @@ int main(int argc, char *argv[])
  	run->psw_mask &= ~(3UL << (63 - 17));   /* Disable AR mode */
  	vcpu_run(vm, VCPU_ID);                  /* Run to sync new state */
 

tools/testing/selftests/kvm/s390x/memop.c int main(int argc, char *argv[]) · line 172

Quoted 10 lines
tools/testing/selftests/kvm/s390x/memop.c +10 -0
+	/* Check that the SIDA calls are rejected for non-protected guests */
+	ksmo.gaddr = 0;
+	ksmo.flags = 0;
+	ksmo.size = 8;
+	ksmo.op = KVM_S390_MEMOP_SIDA_READ;
+	ksmo.buf = (uintptr_t)mem1;
+	ksmo.sida_offset = 0x1c0;
+	rv = _vcpu_ioctl(vm, VCPU_ID, KVM_S390_MEM_OP, &ksmo);
+	TEST_ASSERT(rv == -1 && errno == EINVAL,
+		    "ioctl does not reject SIDA_READ in non-protected mode");

Printing what passed would be a good addition to understand the tests that get run and expected to pass.

tools/testing/selftests/kvm/s390x/memop.c int main(int argc, char *argv[]) · line 177

Quoted 5 lines
tools/testing/selftests/kvm/s390x/memop.c +5 -0
+	ksmo.op = KVM_S390_MEMOP_SIDA_WRITE;
+	rv = _vcpu_ioctl(vm, VCPU_ID, KVM_S390_MEM_OP, &ksmo);
+	TEST_ASSERT(rv == -1 && errno == EINVAL,
+		    "ioctl does not reject SIDA_WRITE in non-protected mode");
+

Same here.

tools/testing/selftests/kvm/s390x/memop.c int main(int argc, char *argv[]) · line 180

Quoted 4 lines
tools/testing/selftests/kvm/s390x/memop.c +0 -0
  	kvm_vm_free(vm);
 
  	return 0;

Something to consider in a follow-on patch and future changes to these tests.

thanks,
-- Shuah

#7 lore raw
Quoted 24 lines · Shuah Khan
Quoted 23 lines · Thomas Huth

Commit 2c212e1baedc ("KVM: s390: Return error on SIDA memop on normal guest") fixed the behavior of the SIDA memops for normal guests. It would be nice to have a way to test whether the current kernel has the fix applied or not. Thus add a check to the KVM selftests for these two memops.

---
  tools/testing/selftests/kvm/s390x/memop.c | 15 +++++++++++++++

  1 file changed, 15 insertions(+)
tools/testing/selftests/kvm/s390x/memop.c +0 -0
index 9f49ead380ab..d19c3ffdea3f 100644
@@ -160,6 +160,21 @@ int main(int argc, char *argv[])

      run->psw_mask &= ~(3UL << (63 - 17));   /* Disable AR mode */ vcpu_run(vm, VCPU_ID);                  /* Run to sync new state */ +    /* Check that the SIDA calls are rejected for non-protected guests */ +    ksmo.gaddr = 0;
+    ksmo.flags = 0;
+    ksmo.size = 8;
+    ksmo.op = KVM_S390_MEMOP_SIDA_READ;

Quoted 8 lines
Quoted 5 lines

+    ksmo.buf = (uintptr_t)mem1;
+    ksmo.sida_offset = 0x1c0;
+    rv = _vcpu_ioctl(vm, VCPU_ID, KVM_S390_MEM_OP, &ksmo); +    TEST_ASSERT(rv == -1 && errno == EINVAL,
+            "ioctl does not reject SIDA_READ in non-protected mode");

Printing what passed would be a good addition to understand the tests that get run and expected to pass.

Yes, I agree ... I'll add that for a follow-up patch to my TODO list.

Quoted 8 lines
Quoted 5 lines

+    ksmo.op = KVM_S390_MEMOP_SIDA_WRITE;
+    rv = _vcpu_ioctl(vm, VCPU_ID, KVM_S390_MEM_OP, &ksmo); +    TEST_ASSERT(rv == -1 && errno == EINVAL,
+            "ioctl does not reject SIDA_WRITE in non-protected mode"); +

Same here.

Quoted 7 lines
Quoted 3 lines

      kvm_vm_free(vm);
      return 0;

Something to consider in a follow-on patch and future changes to these tests.

Thanks!

Thomas