1 files changed, 61 insertions, 50 deletions

diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 201fd8aec59a..cd810e195086 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -29,6 +29,7 @@
 #include <linux/module.h>
 #include <linux/seq_file.h>
 #include <linux/ratelimit.h>
+#include <linux/interval_tree_generic.h>
 
 #include <trace/events/block.h>
 
@@ -50,55 +51,71 @@ static void lower_barrier(struct r1conf *conf, sector_t sector_nr);
 
 #include "raid1-10.c"
 
-static int check_and_add_wb(struct md_rdev *rdev, sector_t lo, sector_t hi)
+#define START(node) ((node)->start)
+#define LAST(node) ((node)->last)
+INTERVAL_TREE_DEFINE(struct serial_info, node, sector_t, _subtree_last,
+		     START, LAST, static inline, raid1_rb);
+
+static int check_and_add_serial(struct md_rdev *rdev, struct r1bio *r1_bio,
+				struct serial_info *si, int idx)
 {
-	struct wb_info *wi, *temp_wi;
 	unsigned long flags;
 	int ret = 0;
-	struct mddev *mddev = rdev->mddev;
-
-	wi = mempool_alloc(mddev->wb_info_pool, GFP_NOIO);
-
-	spin_lock_irqsave(&rdev->wb_list_lock, flags);
-	list_for_each_entry(temp_wi, &rdev->wb_list, list) {
-		/* collision happened */
-		if (hi > temp_wi->lo && lo < temp_wi->hi) {
-			ret = -EBUSY;
-			break;
-		}
+	sector_t lo = r1_bio->sector;
+	sector_t hi = lo + r1_bio->sectors;
+	struct serial_in_rdev *serial = &rdev->serial[idx];
+
+	spin_lock_irqsave(&serial->serial_lock, flags);
+	/* collision happened */
+	if (raid1_rb_iter_first(&serial->serial_rb, lo, hi))
+		ret = -EBUSY;
+	else {
+		si->start = lo;
+		si->last = hi;
+		raid1_rb_insert(si, &serial->serial_rb);
 	}
-
-	if (!ret) {
-		wi->lo = lo;
-		wi->hi = hi;
-		list_add(&wi->list, &rdev->wb_list);
-	} else
-		mempool_free(wi, mddev->wb_info_pool);
-	spin_unlock_irqrestore(&rdev->wb_list_lock, flags);
+	spin_unlock_irqrestore(&serial->serial_lock, flags);
 
 	return ret;
 }
 
-static void remove_wb(struct md_rdev *rdev, sector_t lo, sector_t hi)
+static void wait_for_serialization(struct md_rdev *rdev, struct r1bio *r1_bio)
+{
+	struct mddev *mddev = rdev->mddev;
+	struct serial_info *si;
+	int idx = sector_to_idx(r1_bio->sector);
+	struct serial_in_rdev *serial = &rdev->serial[idx];
+
+	if (WARN_ON(!mddev->serial_info_pool))
+		return;
+	si = mempool_alloc(mddev->serial_info_pool, GFP_NOIO);
+	wait_event(serial->serial_io_wait,
+		   check_and_add_serial(rdev, r1_bio, si, idx) == 0);
+}
+
+static void remove_serial(struct md_rdev *rdev, sector_t lo, sector_t hi)
 {
-	struct wb_info *wi;
+	struct serial_info *si;
 	unsigned long flags;
 	int found = 0;
 	struct mddev *mddev = rdev->mddev;
-
-	spin_lock_irqsave(&rdev->wb_list_lock, flags);
-	list_for_each_entry(wi, &rdev->wb_list, list)
-		if (hi == wi->hi && lo == wi->lo) {
-			list_del(&wi->list);
-			mempool_free(wi, mddev->wb_info_pool);
+	int idx = sector_to_idx(lo);
+	struct serial_in_rdev *serial = &rdev->serial[idx];
+
+	spin_lock_irqsave(&serial->serial_lock, flags);
+	for (si = raid1_rb_iter_first(&serial->serial_rb, lo, hi);
+	     si; si = raid1_rb_iter_next(si, lo, hi)) {
+		if (si->start == lo && si->last == hi) {
+			raid1_rb_remove(si, &serial->serial_rb);
+			mempool_free(si, mddev->serial_info_pool);
 			found = 1;
 			break;
 		}
-
+	}
 	if (!found)
-		WARN(1, "The write behind IO is not recorded\n");
-	spin_unlock_irqrestore(&rdev->wb_list_lock, flags);
-	wake_up(&rdev->wb_io_wait);
+		WARN(1, "The write IO is not recorded for serialization\n");
+	spin_unlock_irqrestore(&serial->serial_lock, flags);
+	wake_up(&serial->serial_io_wait);
 }
 
 /*
@@ -430,6 +447,8 @@ static void raid1_end_write_request(struct bio *bio)
 	int mirror = find_bio_disk(r1_bio, bio);
 	struct md_rdev *rdev = conf->mirrors[mirror].rdev;
 	bool discard_error;
+	sector_t lo = r1_bio->sector;
+	sector_t hi = r1_bio->sector + r1_bio->sectors;
 
 	discard_error = bio->bi_status && bio_op(bio) == REQ_OP_DISCARD;
 
@@ -499,12 +518,8 @@ static void raid1_end_write_request(struct bio *bio)
 	}
 
 	if (behind) {
-		if (test_bit(WBCollisionCheck, &rdev->flags)) {
-			sector_t lo = r1_bio->sector;
-			sector_t hi = r1_bio->sector + r1_bio->sectors;
-
-			remove_wb(rdev, lo, hi);
-		}
+		if (test_bit(CollisionCheck, &rdev->flags))
+			remove_serial(rdev, lo, hi);
 		if (test_bit(WriteMostly, &rdev->flags))
 			atomic_dec(&r1_bio->behind_remaining);
 
@@ -527,7 +542,8 @@ static void raid1_end_write_request(struct bio *bio)
 				call_bio_endio(r1_bio);
 			}
 		}
-	}
+	} else if (rdev->mddev->serialize_policy)
+		remove_serial(rdev, lo, hi);
 	if (r1_bio->bios[mirror] == NULL)
 		rdev_dec_pending(rdev, conf->mddev);
 
@@ -1479,6 +1495,7 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio,
 
 	for (i = 0; i < disks; i++) {
 		struct bio *mbio = NULL;
+		struct md_rdev *rdev = conf->mirrors[i].rdev;
 		if (!r1_bio->bios[i])
 			continue;
 
@@ -1506,18 +1523,12 @@ static void raid1_write_request(struct mddev *mddev, struct bio *bio,
 			mbio = bio_clone_fast(bio, GFP_NOIO, &mddev->bio_set);
 
 		if (r1_bio->behind_master_bio) {
-			struct md_rdev *rdev = conf->mirrors[i].rdev;
-
-			if (test_bit(WBCollisionCheck, &rdev->flags)) {
-				sector_t lo = r1_bio->sector;
-				sector_t hi = r1_bio->sector + r1_bio->sectors;
-
-				wait_event(rdev->wb_io_wait,
-					   check_and_add_wb(rdev, lo, hi) == 0);
-			}
+			if (test_bit(CollisionCheck, &rdev->flags))
+				wait_for_serialization(rdev, r1_bio);
 			if (test_bit(WriteMostly, &rdev->flags))
 				atomic_inc(&r1_bio->behind_remaining);
-		}
+		} else if (mddev->serialize_policy)
+			wait_for_serialization(rdev, r1_bio);
 
 		r1_bio->bios[i] = mbio;