s390/vfio-ap: fix kernel doc and signature of group notifier functions

The vfio_ap device driver registers a group notifier function to handle
the VFIO_GROUP_NOTIFY_SET_KVM event signalling the KVM pointer has been
set or cleared. There are two helper functions invoked by the handler
function: One called when the KVM pointer has been set, and the other
when the pointer is cleared.

The kernel doc for both of these functions contains a comment introduced
by commit 0cc00c8d40 (s390/vfio-ap: fix circular lockdep when
setting/clearing crypto masks) that is no longer valid. This patch removes
this comment from the kernel doc of each helper function.

Commit 86956e7076 (s390/vfio-ap: replace open coded locks for
VFIO_GROUP_NOTIFY_SET_KVM notification) added a parameter to the signature
of the helper function that handles the event indicating the KVM pointer
has been cleared. The parameter added was the KVM pointer itself.
One of the function's primary purposes is to clear the KVM pointer from the
ap_matrix_mdev instance in which it is stored. Since the callers of this
function derive the KVM pointer passed to the function from the
ap_matrix_mdev object itself, it is completely unnecessary to include this
parameter in the function's signature since it can simply be retrieved from
the ap_matrix_mdev object which is also passed in. This patch removes the
KVM pointer from the function's signature.

Signed-off-by: Tony Krowiak <akrowiak@linux.ibm.com>
Signed-off-by: Vasily Gorbik <gor@linux.ibm.com>
This commit is contained in:
Tony Krowiak 2022-03-18 16:17:35 -04:00 committed by Vasily Gorbik
parent ed0192bc64
commit 7107822004

View File

@ -1189,13 +1189,6 @@ static const struct attribute_group *vfio_ap_mdev_attr_groups[] = {
* @matrix_mdev: a mediated matrix device
* @kvm: reference to KVM instance
*
* Note: The matrix_dev->lock must be taken prior to calling
* this function; however, the lock will be temporarily released while the
* guest's AP configuration is set to avoid a potential lockdep splat.
* The kvm->lock is taken to set the guest's AP configuration which, under
* certain circumstances, will result in a circular lock dependency if this is
* done under the @matrix_mdev->lock.
*
* Return: 0 if no other mediated matrix device has a reference to @kvm;
* otherwise, returns an -EPERM.
*/
@ -1269,18 +1262,11 @@ static int vfio_ap_mdev_iommu_notifier(struct notifier_block *nb,
* by @matrix_mdev.
*
* @matrix_mdev: a matrix mediated device
* @kvm: the pointer to the kvm structure being unset.
*
* Note: The matrix_dev->lock must be taken prior to calling
* this function; however, the lock will be temporarily released while the
* guest's AP configuration is cleared to avoid a potential lockdep splat.
* The kvm->lock is taken to clear the guest's AP configuration which, under
* certain circumstances, will result in a circular lock dependency if this is
* done under the @matrix_mdev->lock.
*/
static void vfio_ap_mdev_unset_kvm(struct ap_matrix_mdev *matrix_mdev,
struct kvm *kvm)
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;
@ -1311,7 +1297,7 @@ static int vfio_ap_mdev_group_notifier(struct notifier_block *nb,
matrix_mdev = container_of(nb, struct ap_matrix_mdev, group_notifier);
if (!data)
vfio_ap_mdev_unset_kvm(matrix_mdev, matrix_mdev->kvm);
vfio_ap_mdev_unset_kvm(matrix_mdev);
else if (vfio_ap_mdev_set_kvm(matrix_mdev, data))
notify_rc = NOTIFY_DONE;
@ -1448,7 +1434,7 @@ static void vfio_ap_mdev_close_device(struct vfio_device *vdev)
&matrix_mdev->iommu_notifier);
vfio_unregister_notifier(vdev->dev, VFIO_GROUP_NOTIFY,
&matrix_mdev->group_notifier);
vfio_ap_mdev_unset_kvm(matrix_mdev, matrix_mdev->kvm);
vfio_ap_mdev_unset_kvm(matrix_mdev);
}
static int vfio_ap_mdev_get_device_info(unsigned long arg)