From: NeilBrown <neilb@suse.de>
Date: Tue, 25 Sep 2012 07:00:43 +0000 (+1000)
Subject: md: refine reporting of resync/reshape delays.
X-Git-Tag: next-20120926~52^2~6
X-Git-Url: https://git.karo-electronics.de/?a=commitdiff_plain;h=2ea04ba5d0f23f6bca2057dc4e7460e14b203f83;p=karo-tx-linux.git

md: refine reporting of resync/reshape delays.

If 'resync_max' is set to 0 (as is often done when starting a
reshape, so the mdadm can remain in control during a sensitive
period), and if the reshape request is initially delayed because
another array using the same array is resyncing or reshaping etc,
when user-space cannot easily tell when the delay changes from being
due to a conflicting reshape, to being due to resync_max = 0.

So introduce a new state: (curr_resync == 3) to reflect this, make
sure it is visible both via /proc/mdstat and via the "sync_completed"
sysfs attribute, and ensure that the event transition from one delay
state to the other is properly notified.

Signed-off-by: NeilBrown <neilb@suse.de>
---

diff --git a/drivers/md/md.c b/drivers/md/md.c
index f0494d4e7707..a75fd91c9b0f 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -4427,6 +4427,10 @@ sync_completed_show(struct mddev *mddev, char *page)
 	if (!test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
 		return sprintf(page, "none\n");
 
+	if (mddev->curr_resync == 1 ||
+	    mddev->curr_resync == 2)
+		return sprintf(page, "delayed\n");
+
 	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) ||
 	    test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery))
 		max_sectors = mddev->resync_max_sectors;
@@ -6835,7 +6839,11 @@ static void status_resync(struct seq_file *seq, struct mddev * mddev)
 	int scale;
 	unsigned int per_milli;
 
-	resync = mddev->curr_resync - atomic_read(&mddev->recovery_active);
+	if (mddev->curr_resync <= 3)
+		resync = 0;
+	else
+		resync = mddev->curr_resync
+			- atomic_read(&mddev->recovery_active);
 
 	if (test_bit(MD_RECOVERY_SYNC, &mddev->recovery) ||
 	    test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery))
@@ -7061,7 +7069,7 @@ static int md_seq_show(struct seq_file *seq, void *v)
 				if (mddev->curr_resync > 2) {
 					status_resync(seq, mddev);
 					seq_printf(seq, "\n      ");
-				} else if (mddev->curr_resync == 1 || mddev->curr_resync == 2)
+				} else if (mddev->curr_resync >= 1)
 					seq_printf(seq, "\tresync=DELAYED\n      ");
 				else if (mddev->recovery_cp < MaxSector)
 					seq_printf(seq, "\tresync=PENDING\n      ");
@@ -7451,8 +7459,11 @@ void md_do_sync(struct md_thread *thread)
 		       "md: resuming %s of %s from checkpoint.\n",
 		       desc, mdname(mddev));
 		mddev->curr_resync = j;
-	}
+	} else
+		mddev->curr_resync = 3; /* no longer delayed */
 	mddev->curr_resync_completed = j;
+	sysfs_notify(&mddev->kobj, NULL, "sync_completed");
+	md_new_event(mddev);
 
 	blk_start_plug(&plug);
 	while (j < max_sectors) {
@@ -7505,7 +7516,7 @@ void md_do_sync(struct md_thread *thread)
 			break;
 
 		j += sectors;
-		if (j>1) mddev->curr_resync = j;
+		if (j>2) mddev->curr_resync = j;
 		mddev->curr_mark_cnt = io_sectors;
 		if (last_check == 0)
 			/* this is the earliest that rebuild will be
@@ -7627,8 +7638,6 @@ static int remove_and_add_spares(struct mddev *mddev)
 	int spares = 0;
 	int removed = 0;
 
-	mddev->curr_resync_completed = 0;
-
 	rdev_for_each(rdev, mddev)
 		if (rdev->raid_disk >= 0 &&
 		    !test_bit(Blocked, &rdev->flags) &&
@@ -7823,6 +7832,7 @@ void md_check_recovery(struct mddev *mddev)
 		/* Set RUNNING before clearing NEEDED to avoid
 		 * any transients in the value of "sync_action".
 		 */
+		mddev->curr_resync_completed = 0;
 		set_bit(MD_RECOVERY_RUNNING, &mddev->recovery);
 		/* Clear some bits that don't mean anything, but
 		 * might be left set
@@ -7836,7 +7846,7 @@ void md_check_recovery(struct mddev *mddev)
 		/* no recovery is running.
 		 * remove any failed drives, then
 		 * add spares if possible.
-		 * Spare are also removed and re-added, to allow
+		 * Spares are also removed and re-added, to allow
 		 * the personality to fail the re-add.
 		 */