drbd: introduce stop-sector to online verify

We now can schedule only a specific range of sectors for online verify,
or interrupt a running verify without interrupting the connection.

Had to bump the protocol version differently, we are now 101.
Added verify_can_do_stop_sector() { protocol >= 97 && protocol != 100; }

Also, the return value convention for worker callbacks has changed,
we returned "true/false" for "keep the connection up" in 8.3,
we return 0 for success and <= for failure in 8.4.
Affected: receive_state()

Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>

1 files changed, 27 insertions, 6 deletions

diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c
index 9d7e1fb0f431..1c9c6fd332c3 100644
--- a/drivers/block/drbd/drbd_worker.c
+++ b/drivers/block/drbd/drbd_worker.c
@@ -692,6 +692,7 @@ static int w_make_ov_request(struct drbd_work *w, int cancel)
 	int number, i, size;
 	sector_t sector;
 	const sector_t capacity = drbd_get_capacity(mdev->this_bdev);
+	bool stop_sector_reached = false;
 
 	if (unlikely(cancel))
 		return 1;
@@ -700,9 +701,17 @@ static int w_make_ov_request(struct drbd_work *w, int cancel)
 
 	sector = mdev->ov_position;
 	for (i = 0; i < number; i++) {
-		if (sector >= capacity) {
+		if (sector >= capacity)
 			return 1;
-		}
+
+		/* We check for "finished" only in the reply path:
+		 * w_e_end_ov_reply().
+		 * We need to send at least one request out. */
+		stop_sector_reached = i > 0
+			&& verify_can_do_stop_sector(mdev)
+			&& sector >= mdev->ov_stop_sector;
+		if (stop_sector_reached)
+			break;
 
 		size = BM_BLOCK_SIZE;
 
@@ -726,7 +735,8 @@ static int w_make_ov_request(struct drbd_work *w, int cancel)
 
  requeue:
 	mdev->rs_in_flight += (i << (BM_BLOCK_SHIFT - 9));
-	mod_timer(&mdev->resync_timer, jiffies + SLEEP_TIME);
+	if (i == 0 || !stop_sector_reached)
+		mod_timer(&mdev->resync_timer, jiffies + SLEEP_TIME);
 	return 1;
 }
 
@@ -792,7 +802,12 @@ int drbd_resync_finished(struct drbd_conf *mdev)
 	dt = (jiffies - mdev->rs_start - mdev->rs_paused) / HZ;
 	if (dt <= 0)
 		dt = 1;
+	
 	db = mdev->rs_total;
+	/* adjust for verify start and stop sectors, respective reached position */
+	if (mdev->state.conn == C_VERIFY_S || mdev->state.conn == C_VERIFY_T)
+		db -= mdev->ov_left;
+
 	dbdt = Bit2KB(db/dt);
 	mdev->rs_paused /= HZ;
 
@@ -815,7 +830,7 @@ int drbd_resync_finished(struct drbd_conf *mdev)
 	ns.conn = C_CONNECTED;
 
 	dev_info(DEV, "%s done (total %lu sec; paused %lu sec; %lu K/sec)\n",
-	     verify_done ? "Online verify " : "Resync",
+	     verify_done ? "Online verify" : "Resync",
 	     dt + mdev->rs_paused, mdev->rs_paused, dbdt);
 
 	n_oos = drbd_bm_total_weight(mdev);
@@ -896,7 +911,9 @@ out:
 	mdev->rs_total  = 0;
 	mdev->rs_failed = 0;
 	mdev->rs_paused = 0;
-	if (verify_done)
+
+	/* reset start sector, if we reached end of device */
+	if (verify_done && mdev->ov_left == 0)
 		mdev->ov_start_sector = 0;
 
 	drbd_md_sync(mdev);
@@ -1144,6 +1161,7 @@ int w_e_end_ov_reply(struct drbd_work *w, int cancel)
 	unsigned int size = peer_req->i.size;
 	int digest_size;
 	int err, eq = 0;
+	bool stop_sector_reached = false;
 
 	if (unlikely(cancel)) {
 		drbd_free_peer_req(mdev, peer_req);
@@ -1194,7 +1212,10 @@ int w_e_end_ov_reply(struct drbd_work *w, int cancel)
 	if ((mdev->ov_left & 0x200) == 0x200)
 		drbd_advance_rs_marks(mdev, mdev->ov_left);
 
-	if (mdev->ov_left == 0) {
+	stop_sector_reached = verify_can_do_stop_sector(mdev) &&
+		(sector + (size>>9)) >= mdev->ov_stop_sector;
+
+	if (mdev->ov_left == 0 || stop_sector_reached) {
 		ov_out_of_sync_print(mdev);
 		drbd_resync_finished(mdev);
 	}