s390/vfio-ap: fix stale pqap_hook pointer on error in vfio_ap_mdev_set_kvm()

In vfio_ap_mdev_set_kvm(), kvm->arch.crypto.pqap_hook is set to
&matrix_mdev->pqap_hook before the update locks are acquired and the
mdev list is checked for a conflicting assignment. If another mdev is
already attached to the same KVM instance, the function returns -EPERM
without restoring the hook pointer, leaving kvm->arch.crypto.pqap_hook
pointing at the failing matrix_mdev instead of the mdev that legitimately
owns the KVM.

Since matrix_mdev->kvm is never set on this error path,
vfio_ap_mdev_unset_kvm() will not clean up the hook when matrix_mdev
is later closed. If matrix_mdev is subsequently freed, any PQAP
instruction executed by the guest will dereference the stale pointer
through pqap_hook_rwsem, resulting in a use-after-free.

Since kvm->arch.crypto.pqap_hook is only set in the vfio_ap_mdev_set_kvm()
function and is cleared in the vfio_ap_mdev_unset_kvm() function, a check
for 'kvm->arch.crypto.pqap_hook != NULL' is all that is needed to determine
whether it belongs to another mdev. This will alleviate the need to iterate
the matrix_dev->mdev_list list to see if the kvm object is assigned to
another mdev.This was introduced in v3 to alleviate the need to take the
mdevs_lock while iterating the list; however, this did not prevent a
potential race condition.

The pqap_hook_rwsem(write) is now performed inside
get_update_locks_for_kvm(), which is updated to acquire
pqap_hook_rwsem(write) between kvm->lock and mdevs_lock. This ordering
is consistent with the PQAP intercept path, which acquires pqap_hook_rwsem
in read mode while srcu is held under vcpu->mutex, establishing the
dependency: kvm->lock -> vcpu->mutex -> srcu -> pqap_hook_rwsem(read).

The pqap_hook_rwsem is now released inside the
release_update_locks_for_kvm(), which is updated to release
pqap_hook_rwsem(write) between mdevs_lock and kvm->lock.

Additionally, kvm_put_kvm() in vfio_ap_mdev_unset_kvm() is moved
after release_update_locks_for_kvm(). Previously it was called while
kvm->lock was held; if it were ever the last reference, kvm_destroy_vm()
would run under kvm->lock, which would deadlock.

Fixes: 86956e7076 ("s390/vfio-ap: replace open coded locks for VFIO_GROUP_NOTIFY_SET_KVM notification")
Cc: stable@vger.kernel.org
Co-developed-by: Matthew Rosato <mjrosato@linux.ibm.com>
Signed-off-by: Matthew Rosato <mjrosato@linux.ibm.com>
Signed-off-by: Anthony Krowiak <akrowiak@linux.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@linux.ibm.com>
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
Message-ID: <20260806173435.105044-1-akrowiak@linux.ibm.com>

drivers/s390/crypto/vfio_ap_ops.c

@@ -48,15 +48,19 @@ static void vfio_ap_mdev_reset_queue(struct vfio_ap_queue *q);
  * 1. matrix_dev->guests_lock: required to use the KVM pointer to update a KVM
  *			       guest's APCB.
  * 2. kvm->lock:	       required to update a guest's APCB
- * 3. matrix_dev->mdevs_lock:  required to access data stored in a matrix_mdev
+ * 3. kvm->arch.crypto.pqap_hook_rwsem: required to update pqap_hook and
+ *					serialize against PQAP intercepts
+ * 4. matrix_dev->mdevs_lock:  required to access data stored in a matrix_mdev
  *
- * Note: If @kvm is NULL, the KVM lock will not be taken.
+ * Note: If @kvm is NULL, the KVM lock and pqap_hook_rwsem will not be taken.
  */
 static inline void get_update_locks_for_kvm(struct kvm *kvm)
 {
 	mutex_lock(&matrix_dev->guests_lock);
-	if (kvm)
+	if (kvm) {
 		mutex_lock(&kvm->lock);
+		down_write(&kvm->arch.crypto.pqap_hook_rwsem);
+	}
 	mutex_lock(&matrix_dev->mdevs_lock);
 }
 
@@ -68,16 +72,19 @@ static inline void get_update_locks_for_kvm(struct kvm *kvm)
  *
  * The proper unlocking order is:
  * 1. matrix_dev->mdevs_lock
- * 2. kvm->lock
- * 3. matrix_dev->guests_lock
+ * 2. kvm->arch.crypto.pqap_hook_rwsem
+ * 3. kvm->lock
+ * 4. matrix_dev->guests_lock
  *
- * Note: If @kvm is NULL, the KVM lock will not be released.
+ * Note: If @kvm is NULL, the KVM lock and pqap_hook_rwsem will not be released.
  */
 static inline void release_update_locks_for_kvm(struct kvm *kvm)
 {
 	mutex_unlock(&matrix_dev->mdevs_lock);
-	if (kvm)
+	if (kvm) {
+		up_write(&kvm->arch.crypto.pqap_hook_rwsem);
 		mutex_unlock(&kvm->lock);
+	}
 	mutex_unlock(&matrix_dev->guests_lock);
 }
 
@@ -1821,26 +1828,17 @@ static const struct attribute_group *vfio_ap_mdev_attr_groups[] = {
 static int vfio_ap_mdev_set_kvm(struct ap_matrix_mdev *matrix_mdev,
 				struct kvm *kvm)
 {
-	struct ap_matrix_mdev *m;
-
 	if (kvm->arch.crypto.crycbd) {
-		down_write(&kvm->arch.crypto.pqap_hook_rwsem);
-		kvm->arch.crypto.pqap_hook = &matrix_mdev->pqap_hook;
-		up_write(&kvm->arch.crypto.pqap_hook_rwsem);
-
 		get_update_locks_for_kvm(kvm);
-
-		list_for_each_entry(m, &matrix_dev->mdev_list, node) {
-			if (m != matrix_mdev && m->kvm == kvm) {
-				release_update_locks_for_kvm(kvm);
-				return -EPERM;
-			}
+		if (kvm->arch.crypto.pqap_hook) {
+			release_update_locks_for_kvm(kvm);
+			return -EPERM;
 		}
+		kvm->arch.crypto.pqap_hook = &matrix_mdev->pqap_hook;
 
 		kvm_get_kvm(kvm);
 		matrix_mdev->kvm = kvm;
 		vfio_ap_mdev_update_guest_apcb(matrix_mdev);
-
 		release_update_locks_for_kvm(kvm);
 	}
 
@@ -1883,18 +1881,15 @@ static void vfio_ap_mdev_unset_kvm(struct ap_matrix_mdev *matrix_mdev)
 	struct kvm *kvm = matrix_mdev->kvm;
 
 	if (kvm && kvm->arch.crypto.crycbd) {
-		down_write(&kvm->arch.crypto.pqap_hook_rwsem);
-		kvm->arch.crypto.pqap_hook = NULL;
-		up_write(&kvm->arch.crypto.pqap_hook_rwsem);
-
 		get_update_locks_for_kvm(kvm);
+		kvm->arch.crypto.pqap_hook = NULL;
 
 		kvm_arch_crypto_clear_masks(kvm);
 		vfio_ap_mdev_reset_queues(matrix_mdev);
-		kvm_put_kvm(kvm);
 		matrix_mdev->kvm = NULL;
 
 		release_update_locks_for_kvm(kvm);
+		kvm_put_kvm(kvm);
 	}
 }