drbd: use rolling marks for resync speed calculation

The current resync speed as displayed in /proc/drbd fluctuates a lot.
Using an array of rolling marks makes this calculation much more stable.
We used to have this (a long time ago with 0.7), but it got lost somehow.

If "stalled", do not discard the rest of the information, just add a
" (stalled)" tag to the progress line.

This patch also shortens a spinlock critical section somewhat, and
reduces the number of atomic operations in put_ldev.

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

5 files changed, 79 insertions, 45 deletions

diff --git a/drivers/block/drbd/drbd_actlog.c b/drivers/block/drbd/drbd_actlog.c
index 9400845d602e..b895470e53d7 100644
--- a/drivers/block/drbd/drbd_actlog.c
+++ b/drivers/block/drbd/drbd_actlog.c
@@ -965,29 +965,30 @@ void __drbd_set_in_sync(struct drbd_conf *mdev, sector_t sector, int size,
 	 * ok, (capacity & 7) != 0 sometimes, but who cares...
 	 * we count rs_{total,left} in bits, not sectors.
 	 */
-	spin_lock_irqsave(&mdev->al_lock, flags);
 	count = drbd_bm_clear_bits(mdev, sbnr, ebnr);
-	if (count) {
-		/* we need the lock for drbd_try_clear_on_disk_bm */
-		if (jiffies - mdev->rs_mark_time > HZ*10) {
-			/* should be rolling marks,
-			 * but we estimate only anyways. */
-			if (mdev->rs_mark_left != drbd_bm_total_weight(mdev) &&
+	if (count && get_ldev(mdev)) {
+		unsigned long now = jiffies;
+		unsigned long last = mdev->rs_mark_time[mdev->rs_last_mark];
+		int next = (mdev->rs_last_mark + 1) % DRBD_SYNC_MARKS;
+		if (time_after_eq(now, last + DRBD_SYNC_MARK_STEP)) {
+			unsigned long tw = drbd_bm_total_weight(mdev);
+			if (mdev->rs_mark_left[mdev->rs_last_mark] != tw &&
 			    mdev->state.conn != C_PAUSED_SYNC_T &&
 			    mdev->state.conn != C_PAUSED_SYNC_S) {
-				mdev->rs_mark_time = jiffies;
-				mdev->rs_mark_left = drbd_bm_total_weight(mdev);
+				mdev->rs_mark_time[next] = now;
+				mdev->rs_mark_left[next] = tw;
+				mdev->rs_last_mark = next;
 			}
 		}
-		if (get_ldev(mdev)) {
-			drbd_try_clear_on_disk_bm(mdev, sector, count, TRUE);
-			put_ldev(mdev);
-		}
+		spin_lock_irqsave(&mdev->al_lock, flags);
+		drbd_try_clear_on_disk_bm(mdev, sector, count, TRUE);
+		spin_unlock_irqrestore(&mdev->al_lock, flags);
+
 		/* just wake_up unconditional now, various lc_chaged(),
 		 * lc_put() in drbd_try_clear_on_disk_bm(). */
 		wake_up = 1;
+		put_ldev(mdev);
 	}
-	spin_unlock_irqrestore(&mdev->al_lock, flags);
 	if (wake_up)
 		wake_up(&mdev->al_wait);
 }
diff --git a/drivers/block/drbd/drbd_int.h b/drivers/block/drbd/drbd_int.h
index 72d204750408..0fce3f36fc1c 100644
--- a/drivers/block/drbd/drbd_int.h
+++ b/drivers/block/drbd/drbd_int.h
@@ -997,12 +997,16 @@ struct drbd_conf {
 	unsigned long rs_start;
 	/* cumulated time in PausedSyncX state [unit jiffies] */
 	unsigned long rs_paused;
+	/* skipped because csum was equal [unit BM_BLOCK_SIZE] */
+	unsigned long rs_same_csum;
+#define DRBD_SYNC_MARKS 8
+#define DRBD_SYNC_MARK_STEP (3*HZ)
 	/* block not up-to-date at mark [unit BM_BLOCK_SIZE] */
-	unsigned long rs_mark_left;
+	unsigned long rs_mark_left[DRBD_SYNC_MARKS];
 	/* marks's time [unit jiffies] */
-	unsigned long rs_mark_time;
-	/* skipped because csum was equeal [unit BM_BLOCK_SIZE] */
-	unsigned long rs_same_csum;
+	unsigned long rs_mark_time[DRBD_SYNC_MARKS];
+	/* current index into rs_mark_{left,time} */
+	int rs_last_mark;
 
 	/* where does the admin want us to start? (sector) */
 	sector_t ov_start_sector;
@@ -1077,8 +1081,12 @@ struct drbd_conf {
 	u64 ed_uuid; /* UUID of the exposed data */
 	struct mutex state_mutex;
 	char congestion_reason;  /* Why we where congested... */
-	atomic_t rs_sect_in; /* counter to measure the incoming resync data rate */
-	int c_sync_rate; /* current resync rate after delay_probe magic */
+	atomic_t rs_sect_in; /* for incoming resync data rate, SyncTarget */
+	atomic_t rs_sect_ev; /* for submitted resync data rate, both */
+	int rs_last_sect_ev; /* counter to compare with */
+	int rs_last_events;  /* counter of read or write "events" (unit sectors)
+			      * on the lower level device when we last looked. */
+	int c_sync_rate; /* current resync rate after syncer throttle magic */
 	struct fifo_buffer rs_plan_s; /* correction values of resync planer */
 	int rs_in_flight; /* resync sectors in flight (to proxy, in proxy and from proxy) */
 	int rs_planed;    /* resync sectors already planed */
@@ -2072,10 +2080,11 @@ static inline int get_net_conf(struct drbd_conf *mdev)
 
 static inline void put_ldev(struct drbd_conf *mdev)
 {
+	int i = atomic_dec_return(&mdev->local_cnt);
 	__release(local);
-	if (atomic_dec_and_test(&mdev->local_cnt))
+	D_ASSERT(i >= 0);
+	if (i == 0)
 		wake_up(&mdev->misc_wait);
-	D_ASSERT(atomic_read(&mdev->local_cnt) >= 0);
 }
 
 #ifndef __CHECKER__
diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index ed09a840d838..1ff8418ae0fa 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -1064,7 +1064,8 @@ int __drbd_set_state(struct drbd_conf *mdev,
 	if ((os.conn == C_PAUSED_SYNC_T || os.conn == C_PAUSED_SYNC_S) &&
 	    (ns.conn == C_SYNC_TARGET  || ns.conn == C_SYNC_SOURCE)) {
 		dev_info(DEV, "Syncer continues.\n");
-		mdev->rs_paused += (long)jiffies-(long)mdev->rs_mark_time;
+		mdev->rs_paused += (long)jiffies
+				  -(long)mdev->rs_mark_time[mdev->rs_last_mark];
 		if (ns.conn == C_SYNC_TARGET) {
 			if (!test_and_clear_bit(STOP_SYNC_TIMER, &mdev->flags))
 				mod_timer(&mdev->resync_timer, jiffies);
@@ -1078,27 +1079,33 @@ int __drbd_set_state(struct drbd_conf *mdev,
 	if ((os.conn == C_SYNC_TARGET  || os.conn == C_SYNC_SOURCE) &&
 	    (ns.conn == C_PAUSED_SYNC_T || ns.conn == C_PAUSED_SYNC_S)) {
 		dev_info(DEV, "Resync suspended\n");
-		mdev->rs_mark_time = jiffies;
+		mdev->rs_mark_time[mdev->rs_last_mark] = jiffies;
 		if (ns.conn == C_PAUSED_SYNC_T)
 			set_bit(STOP_SYNC_TIMER, &mdev->flags);
 	}
 
 	if (os.conn == C_CONNECTED &&
 	    (ns.conn == C_VERIFY_S || ns.conn == C_VERIFY_T)) {
+		unsigned long now = jiffies;
+		int i;
+
 		mdev->ov_position = 0;
-		mdev->rs_total =
-		mdev->rs_mark_left = drbd_bm_bits(mdev);
+		mdev->rs_total = drbd_bm_bits(mdev);
 		if (mdev->agreed_pro_version >= 90)
 			set_ov_position(mdev, ns.conn);
 		else
 			mdev->ov_start_sector = 0;
 		mdev->ov_left = mdev->rs_total
 			      - BM_SECT_TO_BIT(mdev->ov_position);
-		mdev->rs_start     =
-		mdev->rs_mark_time = jiffies;
+		mdev->rs_start = now;
 		mdev->ov_last_oos_size = 0;
 		mdev->ov_last_oos_start = 0;
 
+		for (i = 0; i < DRBD_SYNC_MARKS; i++) {
+			mdev->rs_mark_left[i] = mdev->rs_total;
+			mdev->rs_mark_time[i] = now;
+		}
+
 		if (ns.conn == C_VERIFY_S) {
 			dev_info(DEV, "Starting Online Verify from sector %llu\n",
 					(unsigned long long)mdev->ov_position);
@@ -2793,6 +2800,7 @@ void drbd_init_set_defaults(struct drbd_conf *mdev)
 
 void drbd_mdev_cleanup(struct drbd_conf *mdev)
 {
+	int i;
 	if (mdev->receiver.t_state != None)
 		dev_err(DEV, "ASSERT FAILED: receiver t_state == %d expected 0.\n",
 				mdev->receiver.t_state);
@@ -2809,9 +2817,12 @@ void drbd_mdev_cleanup(struct drbd_conf *mdev)
 	mdev->p_size       =
 	mdev->rs_start     =
 	mdev->rs_total     =
-	mdev->rs_failed    =
-	mdev->rs_mark_left =
-	mdev->rs_mark_time = 0;
+	mdev->rs_failed    = 0;
+	mdev->rs_last_events = 0;
+	for (i = 0; i < DRBD_SYNC_MARKS; i++) {
+		mdev->rs_mark_left[i] = 0;
+		mdev->rs_mark_time[i] = 0;
+	}
 	D_ASSERT(mdev->net_conf == NULL);
 
 	drbd_set_my_capacity(mdev, 0);
diff --git a/drivers/block/drbd/drbd_proc.c b/drivers/block/drbd/drbd_proc.c
index be3374b68460..c159692c3b56 100644
--- a/drivers/block/drbd/drbd_proc.c
+++ b/drivers/block/drbd/drbd_proc.c
@@ -57,6 +57,7 @@ static void drbd_syncer_progress(struct drbd_conf *mdev, struct seq_file *seq)
 	unsigned long db, dt, dbdt, rt, rs_left;
 	unsigned int res;
 	int i, x, y;
+	int stalled = 0;
 
 	drbd_get_syncer_progress(mdev, &rs_left, &res);
 
@@ -90,18 +91,17 @@ static void drbd_syncer_progress(struct drbd_conf *mdev, struct seq_file *seq)
 	 * db: blocks written from mark until now
 	 * rt: remaining time
 	 */
-	dt = (jiffies - mdev->rs_mark_time) / HZ;
-
-	if (dt > 20) {
-		/* if we made no update to rs_mark_time for too long,
-		 * we are stalled. show that. */
-		seq_printf(seq, "stalled\n");
-		return;
-	}
+	/* Rolling marks. last_mark+1 may just now be modified.  last_mark+2 is
+	 * at least (DRBD_SYNC_MARKS-2)*DRBD_SYNC_MARK_STEP old, and has at
+	 * least DRBD_SYNC_MARK_STEP time before it will be modified. */
+	i = (mdev->rs_last_mark + 2) % DRBD_SYNC_MARKS;
+	dt = (jiffies - mdev->rs_mark_time[i]) / HZ;
+	if (dt > (DRBD_SYNC_MARK_STEP * DRBD_SYNC_MARKS))
+		stalled = 1;
 
 	if (!dt)
 		dt++;
-	db = mdev->rs_mark_left - rs_left;
+	db = mdev->rs_mark_left[i] - rs_left;
 	rt = (dt * (rs_left / (db/100+1)))/100; /* seconds */
 
 	seq_printf(seq, "finish: %lu:%02lu:%02lu",
@@ -128,7 +128,14 @@ static void drbd_syncer_progress(struct drbd_conf *mdev, struct seq_file *seq)
 	else
 		seq_printf(seq, " (%ld)", dbdt);
 
-	seq_printf(seq, " K/sec\n");
+	if (mdev->state.conn == C_SYNC_TARGET) {
+		if (mdev->c_sync_rate > 1000)
+			seq_printf(seq, " want: %d,%03d",
+				   mdev->c_sync_rate / 1000, mdev->c_sync_rate % 1000);
+		else
+			seq_printf(seq, " want: %d", mdev->c_sync_rate);
+	}
+	seq_printf(seq, " K/sec%s\n", stalled ? " (stalled)" : "");
 }
 
 static void resync_dump_detail(struct seq_file *seq, struct lc_element *e)
diff --git a/drivers/block/drbd/drbd_worker.c b/drivers/block/drbd/drbd_worker.c
index 48452fe83603..53b74254b1c2 100644
--- a/drivers/block/drbd/drbd_worker.c
+++ b/drivers/block/drbd/drbd_worker.c
@@ -1481,13 +1481,19 @@ void drbd_start_resync(struct drbd_conf *mdev, enum drbd_conns side)
 		r = SS_UNKNOWN_ERROR;
 
 	if (r == SS_SUCCESS) {
-		mdev->rs_total     =
-		mdev->rs_mark_left = drbd_bm_total_weight(mdev);
+		unsigned long tw = drbd_bm_total_weight(mdev);
+		unsigned long now = jiffies;
+		int i;
+
 		mdev->rs_failed    = 0;
 		mdev->rs_paused    = 0;
-		mdev->rs_start     =
-		mdev->rs_mark_time = jiffies;
 		mdev->rs_same_csum = 0;
+		mdev->rs_total     = tw;
+		mdev->rs_start     = now;
+		for (i = 0; i < DRBD_SYNC_MARKS; i++) {
+			mdev->rs_mark_left[i] = tw;
+			mdev->rs_mark_time[i] = now;
+		}
 		_drbd_pause_after(mdev);
 	}
 	write_unlock_irq(&global_state_lock);