Re: [PATCH 3/7] s390/vfio_ap: use TAPQ to verify reset in progress completes

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 




On 12/15/22 5:51 AM, Harald Freudenberger wrote:
On 2022-12-13 16:44, Tony Krowiak wrote:
To eliminate the repeated calls to the PQAP(ZAPQ) function to verify that a reset in progress completed successfully and ensure that error response codes get appropriately logged, let's call the apq_reset_check() function
when the ZAPQ response code indicates that a reset that is already in
progress.

Signed-off-by: Tony Krowiak <akrowiak@xxxxxxxxxxxxx>
---
 drivers/s390/crypto/vfio_ap_ops.c | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

diff --git a/drivers/s390/crypto/vfio_ap_ops.c
b/drivers/s390/crypto/vfio_ap_ops.c
index a5530a46cf31..5bf2d93ae8af 100644
--- a/drivers/s390/crypto/vfio_ap_ops.c
+++ b/drivers/s390/crypto/vfio_ap_ops.c
@@ -33,7 +33,7 @@
 static int vfio_ap_mdev_reset_queues(struct ap_queue_table *qtable);
 static struct vfio_ap_queue *vfio_ap_find_queue(int apqn);
 static const struct vfio_device_ops vfio_ap_matrix_dev_ops;
-static int vfio_ap_mdev_reset_queue(struct vfio_ap_queue *q, unsigned
int retry);
+static int vfio_ap_mdev_reset_queue(struct vfio_ap_queue *q);

 /**
  * get_update_locks_for_kvm: Acquire the locks required to dynamically update a @@ -1632,8 +1632,7 @@ static int apq_reset_check(struct vfio_ap_queue *q)
     return ret;
 }

-static int vfio_ap_mdev_reset_queue(struct vfio_ap_queue *q,
-                    unsigned int retry)
+static int vfio_ap_mdev_reset_queue(struct vfio_ap_queue *q)
 {
     struct ap_queue_status status;
     int ret;
@@ -1648,12 +1647,15 @@ static int vfio_ap_mdev_reset_queue(struct
vfio_ap_queue *q,
         ret = 0;
         break;
     case AP_RESPONSE_RESET_IN_PROGRESS:
-        if (retry--) {
-            msleep(20);
-            goto retry_zapq;
-        }
-        ret = -EBUSY;
-        break;
+        /*
+         * There is a reset issued by another process in progress. Let's wait +         * for that to complete. Since we have no idea whether it was a RAPQ or +         * ZAPQ, then if it completes successfully, let's issue the ZAPQ.
+         */
+        ret = apq_reset_check(q);
+        if (ret)
+            break;
+        goto retry_zapq;
     case AP_RESPONSE_Q_NOT_AVAIL:
     case AP_RESPONSE_DECONFIGURED:
     case AP_RESPONSE_CHECKSTOPPED:
@@ -1688,7 +1690,7 @@ static int vfio_ap_mdev_reset_queues(struct
ap_queue_table *qtable)
     struct vfio_ap_queue *q;

     hash_for_each(qtable->queues, loop_cursor, q, mdev_qnode) {
-        ret = vfio_ap_mdev_reset_queue(q, 1);
+        ret = vfio_ap_mdev_reset_queue(q);
         /*
          * Regardless whether a queue turns out to be busy, or
          * is not operational, we need to continue resetting
@@ -1931,7 +1933,7 @@ void vfio_ap_mdev_remove_queue(struct ap_device *apdev)
         }
     }

-    vfio_ap_mdev_reset_queue(q, 1);
+    vfio_ap_mdev_reset_queue(q);
     dev_set_drvdata(&apdev->device, NULL);
     kfree(q);
     release_update_locks_for_mdev(matrix_mdev);

Reviewed-by: Harald Freudenberger <freude@xxxxxxxxxxxxx>


Thanks for the review.





[Index of Archives]     [KVM ARM]     [KVM ia64]     [KVM ppc]     [Virtualization Tools]     [Spice Development]     [Libvirt]     [Libvirt Users]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite Questions]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux