7 files changed, 108 insertions, 20 deletions

diff --git a/drivers/md/persistent-data/dm-array.c b/drivers/md/persistent-data/dm-array.c
index 1d75b1dc1e2e..e64b61ad0ef3 100644
--- a/drivers/md/persistent-data/dm-array.c
+++ b/drivers/md/persistent-data/dm-array.c
@@ -645,8 +645,10 @@ static int array_resize(struct dm_array_info *info, dm_block_t root,
 	int r;
 	struct resize resize;
 
-	if (old_size == new_size)
+	if (old_size == new_size) {
+		*new_root = root;
 		return 0;
+	}
 
 	resize.info = info;
 	resize.root = root;
diff --git a/drivers/md/persistent-data/dm-btree-internal.h b/drivers/md/persistent-data/dm-btree-internal.h
index 37d367bb9aa8..bf2b80d5c470 100644
--- a/drivers/md/persistent-data/dm-btree-internal.h
+++ b/drivers/md/persistent-data/dm-btree-internal.h
@@ -42,6 +42,12 @@ struct btree_node {
 } __packed;
 
 
+/*
+ * Locks a block using the btree node validator.
+ */
+int bn_read_lock(struct dm_btree_info *info, dm_block_t b,
+		 struct dm_block **result);
+
 void inc_children(struct dm_transaction_manager *tm, struct btree_node *n,
 		  struct dm_btree_value_type *vt);
 
diff --git a/drivers/md/persistent-data/dm-btree-spine.c b/drivers/md/persistent-data/dm-btree-spine.c
index cf9fd676ae44..1b5e13ec7f96 100644
--- a/drivers/md/persistent-data/dm-btree-spine.c
+++ b/drivers/md/persistent-data/dm-btree-spine.c
@@ -92,7 +92,7 @@ struct dm_block_validator btree_node_validator = {
 
 /*----------------------------------------------------------------*/
 
-static int bn_read_lock(struct dm_btree_info *info, dm_block_t b,
+int bn_read_lock(struct dm_btree_info *info, dm_block_t b,
 		 struct dm_block **result)
 {
 	return dm_tm_read_lock(info->tm, b, &btree_node_validator, result);
diff --git a/drivers/md/persistent-data/dm-btree.c b/drivers/md/persistent-data/dm-btree.c
index 416060c25709..200ac12a1d40 100644
--- a/drivers/md/persistent-data/dm-btree.c
+++ b/drivers/md/persistent-data/dm-btree.c
@@ -847,22 +847,26 @@ EXPORT_SYMBOL_GPL(dm_btree_find_lowest_key);
  * FIXME: We shouldn't use a recursive algorithm when we have limited stack
  * space.  Also this only works for single level trees.
  */
-static int walk_node(struct ro_spine *s, dm_block_t block,
+static int walk_node(struct dm_btree_info *info, dm_block_t block,
 		     int (*fn)(void *context, uint64_t *keys, void *leaf),
 		     void *context)
 {
 	int r;
 	unsigned i, nr;
+	struct dm_block *node;
 	struct btree_node *n;
 	uint64_t keys;
 
-	r = ro_step(s, block);
-	n = ro_node(s);
+	r = bn_read_lock(info, block, &node);
+	if (r)
+		return r;
+
+	n = dm_block_data(node);
 
 	nr = le32_to_cpu(n->header.nr_entries);
 	for (i = 0; i < nr; i++) {
 		if (le32_to_cpu(n->header.flags) & INTERNAL_NODE) {
-			r = walk_node(s, value64(n, i), fn, context);
+			r = walk_node(info, value64(n, i), fn, context);
 			if (r)
 				goto out;
 		} else {
@@ -874,7 +878,7 @@ static int walk_node(struct ro_spine *s, dm_block_t block,
 	}
 
 out:
-	ro_pop(s);
+	dm_tm_unlock(info->tm, node);
 	return r;
 }
 
@@ -882,15 +886,7 @@ int dm_btree_walk(struct dm_btree_info *info, dm_block_t root,
 		  int (*fn)(void *context, uint64_t *keys, void *leaf),
 		  void *context)
 {
-	int r;
-	struct ro_spine spine;
-
 	BUG_ON(info->levels > 1);
-
-	init_ro_spine(&spine, info);
-	r = walk_node(&spine, root, fn, context);
-	exit_ro_spine(&spine);
-
-	return r;
+	return walk_node(info, root, fn, context);
 }
 EXPORT_SYMBOL_GPL(dm_btree_walk);
diff --git a/drivers/md/persistent-data/dm-space-map-metadata.c b/drivers/md/persistent-data/dm-space-map-metadata.c
index 786b689bdfc7..e8a904298887 100644
--- a/drivers/md/persistent-data/dm-space-map-metadata.c
+++ b/drivers/md/persistent-data/dm-space-map-metadata.c
@@ -564,7 +564,9 @@ static int sm_bootstrap_get_nr_blocks(struct dm_space_map *sm, dm_block_t *count
 {
 	struct sm_metadata *smm = container_of(sm, struct sm_metadata, sm);
 
-	return smm->ll.nr_blocks;
+	*count = smm->ll.nr_blocks;
+
+	return 0;
 }
 
 static int sm_bootstrap_get_nr_free(struct dm_space_map *sm, dm_block_t *count)
@@ -581,7 +583,9 @@ static int sm_bootstrap_get_count(struct dm_space_map *sm, dm_block_t b,
 {
 	struct sm_metadata *smm = container_of(sm, struct sm_metadata, sm);
 
-	return b < smm->begin ? 1 : 0;
+	*result = (b < smm->begin) ? 1 : 0;
+
+	return 0;
 }
 
 static int sm_bootstrap_count_is_more_than_one(struct dm_space_map *sm,
diff --git a/drivers/md/persistent-data/dm-transaction-manager.c b/drivers/md/persistent-data/dm-transaction-manager.c
index 3bc30a0ae3d6..9cb797d800cf 100644
--- a/drivers/md/persistent-data/dm-transaction-manager.c
+++ b/drivers/md/persistent-data/dm-transaction-manager.c
@@ -10,6 +10,8 @@
 #include "dm-persistent-data-internal.h"
 
 #include <linux/export.h>
+#include <linux/mutex.h>
+#include <linux/hash.h>
 #include <linux/slab.h>
 #include <linux/device-mapper.h>
 
@@ -17,6 +19,61 @@
 
 /*----------------------------------------------------------------*/
 
+#define PREFETCH_SIZE 128
+#define PREFETCH_BITS 7
+#define PREFETCH_SENTINEL ((dm_block_t) -1ULL)
+
+struct prefetch_set {
+	struct mutex lock;
+	dm_block_t blocks[PREFETCH_SIZE];
+};
+
+static unsigned prefetch_hash(dm_block_t b)
+{
+	return hash_64(b, PREFETCH_BITS);
+}
+
+static void prefetch_wipe(struct prefetch_set *p)
+{
+	unsigned i;
+	for (i = 0; i < PREFETCH_SIZE; i++)
+		p->blocks[i] = PREFETCH_SENTINEL;
+}
+
+static void prefetch_init(struct prefetch_set *p)
+{
+	mutex_init(&p->lock);
+	prefetch_wipe(p);
+}
+
+static void prefetch_add(struct prefetch_set *p, dm_block_t b)
+{
+	unsigned h = prefetch_hash(b);
+
+	mutex_lock(&p->lock);
+	if (p->blocks[h] == PREFETCH_SENTINEL)
+		p->blocks[h] = b;
+
+	mutex_unlock(&p->lock);
+}
+
+static void prefetch_issue(struct prefetch_set *p, struct dm_block_manager *bm)
+{
+	unsigned i;
+
+	mutex_lock(&p->lock);
+
+	for (i = 0; i < PREFETCH_SIZE; i++)
+		if (p->blocks[i] != PREFETCH_SENTINEL) {
+			dm_bm_prefetch(bm, p->blocks[i]);
+			p->blocks[i] = PREFETCH_SENTINEL;
+		}
+
+	mutex_unlock(&p->lock);
+}
+
+/*----------------------------------------------------------------*/
+
 struct shadow_info {
 	struct hlist_node hlist;
 	dm_block_t where;
@@ -37,6 +94,8 @@ struct dm_transaction_manager {
 
 	spinlock_t lock;
 	struct hlist_head buckets[DM_HASH_SIZE];
+
+	struct prefetch_set prefetches;
 };
 
 /*----------------------------------------------------------------*/
@@ -117,6 +176,8 @@ static struct dm_transaction_manager *dm_tm_create(struct dm_block_manager *bm,
 	for (i = 0; i < DM_HASH_SIZE; i++)
 		INIT_HLIST_HEAD(tm->buckets + i);
 
+	prefetch_init(&tm->prefetches);
+
 	return tm;
 }
 
@@ -268,8 +329,14 @@ int dm_tm_read_lock(struct dm_transaction_manager *tm, dm_block_t b,
 		    struct dm_block_validator *v,
 		    struct dm_block **blk)
 {
-	if (tm->is_clone)
-		return dm_bm_read_try_lock(tm->real->bm, b, v, blk);
+	if (tm->is_clone) {
+		int r = dm_bm_read_try_lock(tm->real->bm, b, v, blk);
+
+		if (r == -EWOULDBLOCK)
+			prefetch_add(&tm->real->prefetches, b);
+
+		return r;
+	}
 
 	return dm_bm_read_lock(tm->bm, b, v, blk);
 }
@@ -317,6 +384,12 @@ struct dm_block_manager *dm_tm_get_bm(struct dm_transaction_manager *tm)
 	return tm->bm;
 }
 
+void dm_tm_issue_prefetches(struct dm_transaction_manager *tm)
+{
+	prefetch_issue(&tm->prefetches, tm->bm);
+}
+EXPORT_SYMBOL_GPL(dm_tm_issue_prefetches);
+
 /*----------------------------------------------------------------*/
 
 static int dm_tm_create_internal(struct dm_block_manager *bm,
diff --git a/drivers/md/persistent-data/dm-transaction-manager.h b/drivers/md/persistent-data/dm-transaction-manager.h
index 2772ed2a781a..2e0d4d66fb1b 100644
--- a/drivers/md/persistent-data/dm-transaction-manager.h
+++ b/drivers/md/persistent-data/dm-transaction-manager.h
@@ -109,6 +109,13 @@ int dm_tm_ref(struct dm_transaction_manager *tm, dm_block_t b,
 struct dm_block_manager *dm_tm_get_bm(struct dm_transaction_manager *tm);
 
 /*
+ * If you're using a non-blocking clone the tm will build up a list of
+ * requested blocks that weren't in core.  This call will request those
+ * blocks to be prefetched.
+ */
+void dm_tm_issue_prefetches(struct dm_transaction_manager *tm);
+
+/*
  * A little utility that ties the knot by producing a transaction manager
  * that has a space map managed by the transaction manager...
  *