1 files changed, 399 insertions, 488 deletions

diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index f03ab5dbda7a..bfae67c593c5 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -144,6 +144,7 @@ struct tree_entry {
  */
 struct btrfs_bio_ctrl {
 	struct bio *bio;
+	int mirror_num;
 	enum btrfs_compression_type compress_type;
 	u32 len_to_stripe_boundary;
 	u32 len_to_oe_boundary;
@@ -178,61 +179,56 @@ static int add_extent_changeset(struct extent_state *state, u32 bits,
 	return ret;
 }
 
-static void submit_one_bio(struct bio *bio, int mirror_num,
-			   enum btrfs_compression_type compress_type)
+static void submit_one_bio(struct btrfs_bio_ctrl *bio_ctrl)
 {
-	struct extent_io_tree *tree = bio->bi_private;
+	struct bio *bio;
+	struct bio_vec *bv;
+	struct inode *inode;
+	int mirror_num;
+
+	if (!bio_ctrl->bio)
+		return;
 
-	bio->bi_private = NULL;
+	bio = bio_ctrl->bio;
+	bv = bio_first_bvec_all(bio);
+	inode = bv->bv_page->mapping->host;
+	mirror_num = bio_ctrl->mirror_num;
 
 	/* Caller should ensure the bio has at least some range added */
 	ASSERT(bio->bi_iter.bi_size);
 
-	if (is_data_inode(tree->private_data))
-		btrfs_submit_data_bio(tree->private_data, bio, mirror_num,
-					    compress_type);
-	else
-		btrfs_submit_metadata_bio(tree->private_data, bio, mirror_num);
-	/*
-	 * Above submission hooks will handle the error by ending the bio,
-	 * which will do the cleanup properly.  So here we should not return
-	 * any error, or the caller of submit_extent_page() will do cleanup
-	 * again, causing problems.
-	 */
-}
+	btrfs_bio(bio)->file_offset = page_offset(bv->bv_page) + bv->bv_offset;
 
-/* Cleanup unsubmitted bios */
-static void end_write_bio(struct extent_page_data *epd, int ret)
-{
-	struct bio *bio = epd->bio_ctrl.bio;
+	if (!is_data_inode(inode))
+		btrfs_submit_metadata_bio(inode, bio, mirror_num);
+	else if (btrfs_op(bio) == BTRFS_MAP_WRITE)
+		btrfs_submit_data_write_bio(inode, bio, mirror_num);
+	else
+		btrfs_submit_data_read_bio(inode, bio, mirror_num,
+					   bio_ctrl->compress_type);
 
-	if (bio) {
-		bio->bi_status = errno_to_blk_status(ret);
-		bio_endio(bio);
-		epd->bio_ctrl.bio = NULL;
-	}
+	/* The bio is owned by the bi_end_io handler now */
+	bio_ctrl->bio = NULL;
 }
 
 /*
- * Submit bio from extent page data via submit_one_bio
- *
- * Return 0 if everything is OK.
- * Return <0 for error.
+ * Submit or fail the current bio in an extent_page_data structure.
  */
-static void flush_write_bio(struct extent_page_data *epd)
+static void submit_write_bio(struct extent_page_data *epd, int ret)
 {
 	struct bio *bio = epd->bio_ctrl.bio;
 
-	if (bio) {
-		submit_one_bio(bio, 0, 0);
-		/*
-		 * Clean up of epd->bio is handled by its endio function.
-		 * And endio is either triggered by successful bio execution
-		 * or the error handler of submit bio hook.
-		 * So at this point, no matter what happened, we don't need
-		 * to clean up epd->bio.
-		 */
+	if (!bio)
+		return;
+
+	if (ret) {
+		ASSERT(ret < 0);
+		bio->bi_status = errno_to_blk_status(ret);
+		bio_endio(bio);
+		/* The bio is owned by the bi_end_io handler now */
 		epd->bio_ctrl.bio = NULL;
+	} else {
+		submit_one_bio(&epd->bio_ctrl);
 	}
 }
 
@@ -376,131 +372,121 @@ void free_extent_state(struct extent_state *state)
 	}
 }
 
-static struct rb_node *tree_insert(struct rb_root *root,
-				   struct rb_node *search_start,
-				   u64 offset,
-				   struct rb_node *node,
-				   struct rb_node ***p_in,
-				   struct rb_node **parent_in)
-{
-	struct rb_node **p;
-	struct rb_node *parent = NULL;
-	struct tree_entry *entry;
-
-	if (p_in && parent_in) {
-		p = *p_in;
-		parent = *parent_in;
-		goto do_insert;
-	}
-
-	p = search_start ? &search_start : &root->rb_node;
-	while (*p) {
-		parent = *p;
-		entry = rb_entry(parent, struct tree_entry, rb_node);
-
-		if (offset < entry->start)
-			p = &(*p)->rb_left;
-		else if (offset > entry->end)
-			p = &(*p)->rb_right;
-		else
-			return parent;
-	}
-
-do_insert:
-	rb_link_node(node, parent, p);
-	rb_insert_color(node, root);
-	return NULL;
-}
-
 /**
  * Search @tree for an entry that contains @offset. Such entry would have
  * entry->start <= offset && entry->end >= offset.
  *
  * @tree:       the tree to search
  * @offset:     offset that should fall within an entry in @tree
- * @next_ret:   pointer to the first entry whose range ends after @offset
- * @prev_ret:   pointer to the first entry whose range begins before @offset
- * @p_ret:      pointer where new node should be anchored (used when inserting an
+ * @node_ret:   pointer where new node should be anchored (used when inserting an
  *	        entry in the tree)
  * @parent_ret: points to entry which would have been the parent of the entry,
  *               containing @offset
  *
- * This function returns a pointer to the entry that contains @offset byte
- * address. If no such entry exists, then NULL is returned and the other
- * pointer arguments to the function are filled, otherwise the found entry is
- * returned and other pointers are left untouched.
+ * Return a pointer to the entry that contains @offset byte address and don't change
+ * @node_ret and @parent_ret.
+ *
+ * If no such entry exists, return pointer to entry that ends before @offset
+ * and fill parameters @node_ret and @parent_ret, ie. does not return NULL.
  */
-static struct rb_node *__etree_search(struct extent_io_tree *tree, u64 offset,
-				      struct rb_node **next_ret,
-				      struct rb_node **prev_ret,
-				      struct rb_node ***p_ret,
-				      struct rb_node **parent_ret)
+static inline struct rb_node *tree_search_for_insert(struct extent_io_tree *tree,
+					             u64 offset,
+						     struct rb_node ***node_ret,
+						     struct rb_node **parent_ret)
 {
 	struct rb_root *root = &tree->state;
-	struct rb_node **n = &root->rb_node;
+	struct rb_node **node = &root->rb_node;
 	struct rb_node *prev = NULL;
-	struct rb_node *orig_prev = NULL;
 	struct tree_entry *entry;
-	struct tree_entry *prev_entry = NULL;
 
-	while (*n) {
-		prev = *n;
+	while (*node) {
+		prev = *node;
 		entry = rb_entry(prev, struct tree_entry, rb_node);
-		prev_entry = entry;
 
 		if (offset < entry->start)
-			n = &(*n)->rb_left;
+			node = &(*node)->rb_left;
 		else if (offset > entry->end)
-			n = &(*n)->rb_right;
+			node = &(*node)->rb_right;
 		else
-			return *n;
+			return *node;
 	}
 
-	if (p_ret)
-		*p_ret = n;
+	if (node_ret)
+		*node_ret = node;
 	if (parent_ret)
 		*parent_ret = prev;
 
-	if (next_ret) {
-		orig_prev = prev;
-		while (prev && offset > prev_entry->end) {
-			prev = rb_next(prev);
-			prev_entry = rb_entry(prev, struct tree_entry, rb_node);
-		}
-		*next_ret = prev;
-		prev = orig_prev;
+	/* Search neighbors until we find the first one past the end */
+	while (prev && offset > entry->end) {
+		prev = rb_next(prev);
+		entry = rb_entry(prev, struct tree_entry, rb_node);
 	}
 
-	if (prev_ret) {
-		prev_entry = rb_entry(prev, struct tree_entry, rb_node);
-		while (prev && offset < prev_entry->start) {
-			prev = rb_prev(prev);
-			prev_entry = rb_entry(prev, struct tree_entry, rb_node);
-		}
-		*prev_ret = prev;
-	}
-	return NULL;
+	return prev;
 }
 
-static inline struct rb_node *
-tree_search_for_insert(struct extent_io_tree *tree,
-		       u64 offset,
-		       struct rb_node ***p_ret,
-		       struct rb_node **parent_ret)
+/*
+ * Inexact rb-tree search, return the next entry if @offset is not found
+ */
+static inline struct rb_node *tree_search(struct extent_io_tree *tree, u64 offset)
 {
-	struct rb_node *next= NULL;
-	struct rb_node *ret;
-
-	ret = __etree_search(tree, offset, &next, NULL, p_ret, parent_ret);
-	if (!ret)
-		return next;
-	return ret;
+	return tree_search_for_insert(tree, offset, NULL, NULL);
 }
 
-static inline struct rb_node *tree_search(struct extent_io_tree *tree,
-					  u64 offset)
+/**
+ * Search offset in the tree or fill neighbor rbtree node pointers.
+ *
+ * @tree:      the tree to search
+ * @offset:    offset that should fall within an entry in @tree
+ * @next_ret:  pointer to the first entry whose range ends after @offset
+ * @prev_ret:  pointer to the first entry whose range begins before @offset
+ *
+ * Return a pointer to the entry that contains @offset byte address. If no
+ * such entry exists, then return NULL and fill @prev_ret and @next_ret.
+ * Otherwise return the found entry and other pointers are left untouched.
+ */
+static struct rb_node *tree_search_prev_next(struct extent_io_tree *tree,
+					     u64 offset,
+					     struct rb_node **prev_ret,
+					     struct rb_node **next_ret)
 {
-	return tree_search_for_insert(tree, offset, NULL, NULL);
+	struct rb_root *root = &tree->state;
+	struct rb_node **node = &root->rb_node;
+	struct rb_node *prev = NULL;
+	struct rb_node *orig_prev = NULL;
+	struct tree_entry *entry;
+
+	ASSERT(prev_ret);
+	ASSERT(next_ret);
+
+	while (*node) {
+		prev = *node;
+		entry = rb_entry(prev, struct tree_entry, rb_node);
+
+		if (offset < entry->start)
+			node = &(*node)->rb_left;
+		else if (offset > entry->end)
+			node = &(*node)->rb_right;
+		else
+			return *node;
+	}
+
+	orig_prev = prev;
+	while (prev && offset > entry->end) {
+		prev = rb_next(prev);
+		entry = rb_entry(prev, struct tree_entry, rb_node);
+	}
+	*next_ret = prev;
+	prev = orig_prev;
+
+	entry = rb_entry(prev, struct tree_entry, rb_node);
+	while (prev && offset < entry->start) {
+		prev = rb_prev(prev);
+		entry = rb_entry(prev, struct tree_entry, rb_node);
+	}
+	*prev_ret = prev;
+
+	return NULL;
 }
 
 /*
@@ -554,7 +540,7 @@ static void merge_state(struct extent_io_tree *tree,
 }
 
 static void set_state_bits(struct extent_io_tree *tree,
-			   struct extent_state *state, u32 *bits,
+			   struct extent_state *state, u32 bits,
 			   struct extent_changeset *changeset);
 
 /*
@@ -568,37 +554,56 @@ static void set_state_bits(struct extent_io_tree *tree,
  * probably isn't what you want to call (see set/clear_extent_bit).
  */
 static int insert_state(struct extent_io_tree *tree,
-			struct extent_state *state, u64 start, u64 end,
-			struct rb_node ***p,
-			struct rb_node **parent,
-			u32 *bits, struct extent_changeset *changeset)
+			struct extent_state *state,
+			u32 bits, struct extent_changeset *changeset)
 {
-	struct rb_node *node;
-
-	if (end < start) {
-		btrfs_err(tree->fs_info,
-			"insert state: end < start %llu %llu", end, start);
-		WARN_ON(1);
-	}
-	state->start = start;
-	state->end = end;
+	struct rb_node **node;
+	struct rb_node *parent;
+	const u64 end = state->end;
 
 	set_state_bits(tree, state, bits, changeset);
 
-	node = tree_insert(&tree->state, NULL, end, &state->rb_node, p, parent);
-	if (node) {
-		struct extent_state *found;
-		found = rb_entry(node, struct extent_state, rb_node);
-		btrfs_err(tree->fs_info,
-		       "found node %llu %llu on insert of %llu %llu",
-		       found->start, found->end, start, end);
-		return -EEXIST;
+	node = &tree->state.rb_node;
+	while (*node) {
+		struct tree_entry *entry;
+
+		parent = *node;
+		entry = rb_entry(parent, struct tree_entry, rb_node);
+
+		if (end < entry->start) {
+			node = &(*node)->rb_left;
+		} else if (end > entry->end) {
+			node = &(*node)->rb_right;
+		} else {
+			btrfs_err(tree->fs_info,
+			       "found node %llu %llu on insert of %llu %llu",
+			       entry->start, entry->end, state->start, end);
+			return -EEXIST;
+		}
 	}
+
+	rb_link_node(&state->rb_node, parent, node);
+	rb_insert_color(&state->rb_node, &tree->state);
+
 	merge_state(tree, state);
 	return 0;
 }
 
 /*
+ * Insert state to @tree to the location given by @node and @parent.
+ */
+static void insert_state_fast(struct extent_io_tree *tree,
+			      struct extent_state *state, struct rb_node **node,
+			      struct rb_node *parent, unsigned bits,
+			      struct extent_changeset *changeset)
+{
+	set_state_bits(tree, state, bits, changeset);
+	rb_link_node(&state->rb_node, parent, node);
+	rb_insert_color(&state->rb_node, &tree->state);
+	merge_state(tree, state);
+}
+
+/*
  * split a given extent state struct in two, inserting the preallocated
  * struct 'prealloc' as the newly created second half.  'split' indicates an
  * offset inside 'orig' where it should be split.
@@ -615,7 +620,8 @@ static int insert_state(struct extent_io_tree *tree,
 static int split_state(struct extent_io_tree *tree, struct extent_state *orig,
 		       struct extent_state *prealloc, u64 split)
 {
-	struct rb_node *node;
+	struct rb_node *parent = NULL;
+	struct rb_node **node;
 
 	if (tree->private_data && is_data_inode(tree->private_data))
 		btrfs_split_delalloc_extent(tree->private_data, orig, split);
@@ -625,12 +631,27 @@ static int split_state(struct extent_io_tree *tree, struct extent_state *orig,
 	prealloc->state = orig->state;
 	orig->start = split;
 
-	node = tree_insert(&tree->state, &orig->rb_node, prealloc->end,
-			   &prealloc->rb_node, NULL, NULL);
-	if (node) {
-		free_extent_state(prealloc);
-		return -EEXIST;
+	parent = &orig->rb_node;
+	node = &parent;
+	while (*node) {
+		struct tree_entry *entry;
+
+		parent = *node;
+		entry = rb_entry(parent, struct tree_entry, rb_node);
+
+		if (prealloc->end < entry->start) {
+			node = &(*node)->rb_left;
+		} else if (prealloc->end > entry->end) {
+			node = &(*node)->rb_right;
+		} else {
+			free_extent_state(prealloc);
+			return -EEXIST;
+		}
 	}
+
+	rb_link_node(&prealloc->rb_node, parent, node);
+	rb_insert_color(&prealloc->rb_node, &tree->state);
+
 	return 0;
 }
 
@@ -652,11 +673,11 @@ static struct extent_state *next_state(struct extent_state *state)
  */
 static struct extent_state *clear_state_bit(struct extent_io_tree *tree,
 					    struct extent_state *state,
-					    u32 *bits, int wake,
+					    u32 bits, int wake,
 					    struct extent_changeset *changeset)
 {
 	struct extent_state *next;
-	u32 bits_to_clear = *bits & ~EXTENT_CTLBITS;
+	u32 bits_to_clear = bits & ~EXTENT_CTLBITS;
 	int ret;
 
 	if ((bits_to_clear & EXTENT_DIRTY) && (state->state & EXTENT_DIRTY)) {
@@ -818,8 +839,7 @@ hit_next:
 		if (err)
 			goto out;
 		if (state->end <= end) {
-			state = clear_state_bit(tree, state, &bits, wake,
-						changeset);
+			state = clear_state_bit(tree, state, bits, wake, changeset);
 			goto next;
 		}
 		goto search_again;
@@ -840,13 +860,13 @@ hit_next:
 		if (wake)
 			wake_up(&state->wq);
 
-		clear_state_bit(tree, prealloc, &bits, wake, changeset);
+		clear_state_bit(tree, prealloc, bits, wake, changeset);
 
 		prealloc = NULL;
 		goto out;
 	}
 
-	state = clear_state_bit(tree, state, &bits, wake, changeset);
+	state = clear_state_bit(tree, state, bits, wake, changeset);
 next:
 	if (last_end == (u64)-1)
 		goto out;
@@ -937,9 +957,9 @@ out:
 
 static void set_state_bits(struct extent_io_tree *tree,
 			   struct extent_state *state,
-			   u32 *bits, struct extent_changeset *changeset)
+			   u32 bits, struct extent_changeset *changeset)
 {
-	u32 bits_to_set = *bits & ~EXTENT_CTLBITS;
+	u32 bits_to_set = bits & ~EXTENT_CTLBITS;
 	int ret;
 
 	if (tree->private_data && is_data_inode(tree->private_data))
@@ -1033,11 +1053,9 @@ again:
 	if (!node) {
 		prealloc = alloc_extent_state_atomic(prealloc);
 		BUG_ON(!prealloc);
-		err = insert_state(tree, prealloc, start, end,
-				   &p, &parent, &bits, changeset);
-		if (err)
-			extent_io_tree_panic(tree, err);
-
+		prealloc->start = start;
+		prealloc->end = end;
+		insert_state_fast(tree, prealloc, p, parent, bits, changeset);
 		cache_state(prealloc, cached_state);
 		prealloc = NULL;
 		goto out;
@@ -1060,7 +1078,7 @@ hit_next:
 			goto out;
 		}
 
-		set_state_bits(tree, state, &bits, changeset);
+		set_state_bits(tree, state, bits, changeset);
 		cache_state(state, cached_state);
 		merge_state(tree, state);
 		if (last_end == (u64)-1)
@@ -1116,7 +1134,7 @@ hit_next:
 		if (err)
 			goto out;
 		if (state->end <= end) {
-			set_state_bits(tree, state, &bits, changeset);
+			set_state_bits(tree, state, bits, changeset);
 			cache_state(state, cached_state);
 			merge_state(tree, state);
 			if (last_end == (u64)-1)
@@ -1150,8 +1168,9 @@ hit_next:
 		 * Avoid to free 'prealloc' if it can be merged with
 		 * the later extent.
 		 */
-		err = insert_state(tree, prealloc, start, this_end,
-				   NULL, NULL, &bits, changeset);
+		prealloc->start = start;
+		prealloc->end = this_end;
+		err = insert_state(tree, prealloc, bits, changeset);
 		if (err)
 			extent_io_tree_panic(tree, err);
 
@@ -1179,7 +1198,7 @@ hit_next:
 		if (err)
 			extent_io_tree_panic(tree, err);
 
-		set_state_bits(tree, prealloc, &bits, changeset);
+		set_state_bits(tree, prealloc, bits, changeset);
 		cache_state(prealloc, cached_state);
 		merge_state(tree, prealloc);
 		prealloc = NULL;
@@ -1274,10 +1293,9 @@ again:
 			err = -ENOMEM;
 			goto out;
 		}
-		err = insert_state(tree, prealloc, start, end,
-				   &p, &parent, &bits, NULL);
-		if (err)
-			extent_io_tree_panic(tree, err);
+		prealloc->start = start;
+		prealloc->end = end;
+		insert_state_fast(tree, prealloc, p, parent, bits, NULL);
 		cache_state(prealloc, cached_state);
 		prealloc = NULL;
 		goto out;
@@ -1294,9 +1312,9 @@ hit_next:
 	 * Just lock what we found and keep going
 	 */
 	if (state->start == start && state->end <= end) {
-		set_state_bits(tree, state, &bits, NULL);
+		set_state_bits(tree, state, bits, NULL);
 		cache_state(state, cached_state);
-		state = clear_state_bit(tree, state, &clear_bits, 0, NULL);
+		state = clear_state_bit(tree, state, clear_bits, 0, NULL);
 		if (last_end == (u64)-1)
 			goto out;
 		start = last_end + 1;
@@ -1335,10 +1353,9 @@ hit_next:
 		if (err)
 			goto out;
 		if (state->end <= end) {
-			set_state_bits(tree, state, &bits, NULL);
+			set_state_bits(tree, state, bits, NULL);
 			cache_state(state, cached_state);
-			state = clear_state_bit(tree, state, &clear_bits, 0,
-						NULL);
+			state = clear_state_bit(tree, state, clear_bits, 0, NULL);
 			if (last_end == (u64)-1)
 				goto out;
 			start = last_end + 1;
@@ -1372,8 +1389,9 @@ hit_next:
 		 * Avoid to free 'prealloc' if it can be merged with
 		 * the later extent.
 		 */
-		err = insert_state(tree, prealloc, start, this_end,
-				   NULL, NULL, &bits, NULL);
+		prealloc->start = start;
+		prealloc->end = this_end;
+		err = insert_state(tree, prealloc, bits, NULL);
 		if (err)
 			extent_io_tree_panic(tree, err);
 		cache_state(prealloc, cached_state);
@@ -1398,9 +1416,9 @@ hit_next:
 		if (err)
 			extent_io_tree_panic(tree, err);
 
-		set_state_bits(tree, prealloc, &bits, NULL);
+		set_state_bits(tree, prealloc, bits, NULL);
 		cache_state(prealloc, cached_state);
-		clear_state_bit(tree, prealloc, &clear_bits, 0, NULL);
+		clear_state_bit(tree, prealloc, clear_bits, 0, NULL);
 		prealloc = NULL;
 		goto out;
 	}
@@ -1674,7 +1692,7 @@ void find_first_clear_extent_bit(struct extent_io_tree *tree, u64 start,
 
 	/* Find first extent with bits cleared */
 	while (1) {
-		node = __etree_search(tree, start, &next, &prev, NULL, NULL);
+		node = tree_search_prev_next(tree, start, &prev, &next);
 		if (!node && !next && !prev) {
 			/*
 			 * Tree is completely empty, send full range and let
@@ -2007,10 +2025,12 @@ noinline_for_stack bool find_lock_delalloc_range(struct inode *inode,
 				    struct page *locked_page, u64 *start,
 				    u64 *end)
 {
+	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
 	const u64 orig_start = *start;
 	const u64 orig_end = *end;
-	u64 max_bytes = BTRFS_MAX_EXTENT_SIZE;
+	/* The sanity tests may not set a valid fs_info. */
+	u64 max_bytes = fs_info ? fs_info->max_extent_size : BTRFS_MAX_EXTENT_SIZE;
 	u64 delalloc_start;
 	u64 delalloc_end;
 	bool found;
@@ -2418,6 +2438,20 @@ int btrfs_repair_eb_io_failure(const struct extent_buffer *eb, int mirror_num)
 	return ret;
 }
 
+static int next_mirror(const struct io_failure_record *failrec, int cur_mirror)
+{
+	if (cur_mirror == failrec->num_copies)
+		return cur_mirror + 1 - failrec->num_copies;
+	return cur_mirror + 1;
+}
+
+static int prev_mirror(const struct io_failure_record *failrec, int cur_mirror)
+{
+	if (cur_mirror == 1)
+		return failrec->num_copies;
+	return cur_mirror - 1;
+}
+
 /*
  * each time an IO finishes, we do a fast check in the IO failure tree
  * to see if we need to process or clean up an io_failure_record
@@ -2430,7 +2464,7 @@ int clean_io_failure(struct btrfs_fs_info *fs_info,
 	u64 private;
 	struct io_failure_record *failrec;
 	struct extent_state *state;
-	int num_copies;
+	int mirror;
 	int ret;
 
 	private = 0;
@@ -2454,20 +2488,19 @@ int clean_io_failure(struct btrfs_fs_info *fs_info,
 					    EXTENT_LOCKED);
 	spin_unlock(&io_tree->lock);
 
-	if (state && state->start <= failrec->start &&
-	    state->end >= failrec->start + failrec->len - 1) {
-		num_copies = btrfs_num_copies(fs_info, failrec->logical,
-					      failrec->len);
-		if (num_copies > 1)  {
-			repair_io_failure(fs_info, ino, start, failrec->len,
-					  failrec->logical, page, pg_offset,
-					  failrec->failed_mirror);
-		}
-	}
+	if (!state || state->start > failrec->start ||
+	    state->end < failrec->start + failrec->len - 1)
+		goto out;
+
+	mirror = failrec->this_mirror;
+	do {
+		mirror = prev_mirror(failrec, mirror);
+		repair_io_failure(fs_info, ino, start, failrec->len,
+				  failrec->logical, page, pg_offset, mirror);
+	} while (mirror != failrec->failed_mirror);
 
 out:
 	free_io_failure(failure_tree, io_tree, failrec);
-
 	return 0;
 }
 
@@ -2506,17 +2539,16 @@ void btrfs_free_io_failure_record(struct btrfs_inode *inode, u64 start, u64 end)
 }
 
 static struct io_failure_record *btrfs_get_io_failure_record(struct inode *inode,
-							     u64 start)
+							     struct btrfs_bio *bbio,
+							     unsigned int bio_offset)
 {
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	u64 start = bbio->file_offset + bio_offset;
 	struct io_failure_record *failrec;
-	struct extent_map *em;
 	struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
 	struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
-	struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree;
 	const u32 sectorsize = fs_info->sectorsize;
 	int ret;
-	u64 logical;
 
 	failrec = get_state_failrec(failure_tree, start);
 	if (!IS_ERR(failrec)) {
@@ -2528,7 +2560,8 @@ static struct io_failure_record *btrfs_get_io_failure_record(struct inode *inode
 		 * (e.g. with a list for failed_mirror) to make
 		 * clean_io_failure() clean all those errors at once.
 		 */
-
+		ASSERT(failrec->this_mirror == bbio->mirror_num);
+		ASSERT(failrec->len == fs_info->sectorsize);
 		return failrec;
 	}
 
@@ -2538,41 +2571,28 @@ static struct io_failure_record *btrfs_get_io_failure_record(struct inode *inode
 
 	failrec->start = start;
 	failrec->len = sectorsize;
-	failrec->this_mirror = 0;
-	failrec->compress_type = BTRFS_COMPRESS_NONE;
+	failrec->failed_mirror = bbio->mirror_num;
+	failrec->this_mirror = bbio->mirror_num;
+	failrec->logical = (bbio->iter.bi_sector << SECTOR_SHIFT) + bio_offset;
 
-	read_lock(&em_tree->lock);
-	em = lookup_extent_mapping(em_tree, start, failrec->len);
-	if (!em) {
-		read_unlock(&em_tree->lock);
-		kfree(failrec);
-		return ERR_PTR(-EIO);
-	}
+	btrfs_debug(fs_info,
+		    "new io failure record logical %llu start %llu",
+		    failrec->logical, start);
 
-	if (em->start > start || em->start + em->len <= start) {
-		free_extent_map(em);
-		em = NULL;
-	}
-	read_unlock(&em_tree->lock);
-	if (!em) {
+	failrec->num_copies = btrfs_num_copies(fs_info, failrec->logical, sectorsize);
+	if (failrec->num_copies == 1) {
+		/*
+		 * We only have a single copy of the data, so don't bother with
+		 * all the retry and error correction code that follows. No
+		 * matter what the error is, it is very likely to persist.
+		 */
+		btrfs_debug(fs_info,
+			"cannot repair logical %llu num_copies %d",
+			failrec->logical, failrec->num_copies);
 		kfree(failrec);
 		return ERR_PTR(-EIO);
 	}
 
-	logical = start - em->start;
-	logical = em->block_start + logical;
-	if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) {
-		logical = em->block_start;
-		failrec->compress_type = em->compress_type;
-	}
-
-	btrfs_debug(fs_info,
-		    "Get IO Failure Record: (new) logical=%llu, start=%llu, len=%llu",
-		    logical, start, failrec->len);
-
-	failrec->logical = logical;
-	free_extent_map(em);
-
 	/* Set the bits in the private failure tree */
 	ret = set_extent_bits(failure_tree, start, start + sectorsize - 1,
 			      EXTENT_LOCKED | EXTENT_DIRTY);
@@ -2589,65 +2609,16 @@ static struct io_failure_record *btrfs_get_io_failure_record(struct inode *inode
 	return failrec;
 }
 
-static bool btrfs_check_repairable(struct inode *inode,
-				   struct io_failure_record *failrec,
-				   int failed_mirror)
-{
-	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
-	int num_copies;
-
-	num_copies = btrfs_num_copies(fs_info, failrec->logical, failrec->len);
-	if (num_copies == 1) {
-		/*
-		 * we only have a single copy of the data, so don't bother with
-		 * all the retry and error correction code that follows. no
-		 * matter what the error is, it is very likely to persist.
-		 */
-		btrfs_debug(fs_info,
-			"Check Repairable: cannot repair, num_copies=%d, next_mirror %d, failed_mirror %d",
-			num_copies, failrec->this_mirror, failed_mirror);
-		return false;
-	}
-
-	/* The failure record should only contain one sector */
-	ASSERT(failrec->len == fs_info->sectorsize);
-
-	/*
-	 * There are two premises:
-	 * a) deliver good data to the caller
-	 * b) correct the bad sectors on disk
-	 *
-	 * Since we're only doing repair for one sector, we only need to get
-	 * a good copy of the failed sector and if we succeed, we have setup
-	 * everything for repair_io_failure to do the rest for us.
-	 */
-	ASSERT(failed_mirror);
-	failrec->failed_mirror = failed_mirror;
-	failrec->this_mirror++;
-	if (failrec->this_mirror == failed_mirror)
-		failrec->this_mirror++;
-
-	if (failrec->this_mirror > num_copies) {
-		btrfs_debug(fs_info,
-			"Check Repairable: (fail) num_copies=%d, next_mirror %d, failed_mirror %d",
-			num_copies, failrec->this_mirror, failed_mirror);
-		return false;
-	}
-
-	return true;
-}
-
-int btrfs_repair_one_sector(struct inode *inode,
-			    struct bio *failed_bio, u32 bio_offset,
-			    struct page *page, unsigned int pgoff,
-			    u64 start, int failed_mirror,
+int btrfs_repair_one_sector(struct inode *inode, struct btrfs_bio *failed_bbio,
+			    u32 bio_offset, struct page *page, unsigned int pgoff,
 			    submit_bio_hook_t *submit_bio_hook)
 {
+	u64 start = failed_bbio->file_offset + bio_offset;
 	struct io_failure_record *failrec;
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 	struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
 	struct extent_io_tree *failure_tree = &BTRFS_I(inode)->io_failure_tree;
-	struct btrfs_bio *failed_bbio = btrfs_bio(failed_bio);
+	struct bio *failed_bio = &failed_bbio->bio;
 	const int icsum = bio_offset >> fs_info->sectorsize_bits;
 	struct bio *repair_bio;
 	struct btrfs_bio *repair_bbio;
@@ -2657,12 +2628,24 @@ int btrfs_repair_one_sector(struct inode *inode,
 
 	BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE);
 
-	failrec = btrfs_get_io_failure_record(inode, start);
+	failrec = btrfs_get_io_failure_record(inode, failed_bbio, bio_offset);
 	if (IS_ERR(failrec))
 		return PTR_ERR(failrec);
 
-
-	if (!btrfs_check_repairable(inode, failrec, failed_mirror)) {
+	/*
+	 * There are two premises:
+	 * a) deliver good data to the caller
+	 * b) correct the bad sectors on disk
+	 *
+	 * Since we're only doing repair for one sector, we only need to get
+	 * a good copy of the failed sector and if we succeed, we have setup
+	 * everything for repair_io_failure to do the rest for us.
+	 */
+	failrec->this_mirror = next_mirror(failrec, failrec->this_mirror);
+	if (failrec->this_mirror == failrec->failed_mirror) {
+		btrfs_debug(fs_info,
+			"failed to repair num_copies %d this_mirror %d failed_mirror %d",
+			failrec->num_copies, failrec->this_mirror, failrec->failed_mirror);
 		free_io_failure(failure_tree, tree, failrec);
 		return -EIO;
 	}
@@ -2695,7 +2678,7 @@ int btrfs_repair_one_sector(struct inode *inode,
 	 * will be handled by the endio on the repair_bio, so we can't return an
 	 * error here.
 	 */
-	submit_bio_hook(inode, repair_bio, failrec->this_mirror, failrec->compress_type);
+	submit_bio_hook(inode, repair_bio, failrec->this_mirror, 0);
 	return BLK_STS_OK;
 }
 
@@ -2727,21 +2710,35 @@ static void end_page_read(struct page *page, bool uptodate, u64 start, u32 len)
 		btrfs_subpage_end_reader(fs_info, page, start, len);
 }
 
-static blk_status_t submit_data_read_repair(struct inode *inode,
-					    struct bio *failed_bio,
-					    u32 bio_offset, struct page *page,
-					    unsigned int pgoff,
-					    u64 start, u64 end,
-					    int failed_mirror,
-					    unsigned int error_bitmap)
+static void end_sector_io(struct page *page, u64 offset, bool uptodate)
+{
+	struct btrfs_inode *inode = BTRFS_I(page->mapping->host);
+	const u32 sectorsize = inode->root->fs_info->sectorsize;
+	struct extent_state *cached = NULL;
+
+	end_page_read(page, uptodate, offset, sectorsize);
+	if (uptodate)
+		set_extent_uptodate(&inode->io_tree, offset,
+				    offset + sectorsize - 1, &cached, GFP_ATOMIC);
+	unlock_extent_cached_atomic(&inode->io_tree, offset,
+				    offset + sectorsize - 1, &cached);
+}
+
+static void submit_data_read_repair(struct inode *inode,
+				    struct btrfs_bio *failed_bbio,
+				    u32 bio_offset, const struct bio_vec *bvec,
+				    unsigned int error_bitmap)
 {
+	const unsigned int pgoff = bvec->bv_offset;
 	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
+	struct page *page = bvec->bv_page;
+	const u64 start = page_offset(bvec->bv_page) + bvec->bv_offset;
+	const u64 end = start + bvec->bv_len - 1;
 	const u32 sectorsize = fs_info->sectorsize;
 	const int nr_bits = (end + 1 - start) >> fs_info->sectorsize_bits;
-	int error = 0;
 	int i;
 
-	BUG_ON(bio_op(failed_bio) == REQ_OP_WRITE);
+	BUG_ON(bio_op(&failed_bbio->bio) == REQ_OP_WRITE);
 
 	/* This repair is only for data */
 	ASSERT(is_data_inode(inode));
@@ -2753,12 +2750,11 @@ static blk_status_t submit_data_read_repair(struct inode *inode,
 	 * We only get called on buffered IO, thus page must be mapped and bio
 	 * must not be cloned.
 	 */
-	ASSERT(page->mapping && !bio_flagged(failed_bio, BIO_CLONED));
+	ASSERT(page->mapping && !bio_flagged(&failed_bbio->bio, BIO_CLONED));
 
 	/* Iterate through all the sectors in the range */
 	for (i = 0; i < nr_bits; i++) {
 		const unsigned int offset = i * sectorsize;
-		struct extent_state *cached = NULL;
 		bool uptodate = false;
 		int ret;
 
@@ -2771,10 +2767,9 @@ static blk_status_t submit_data_read_repair(struct inode *inode,
 			goto next;
 		}
 
-		ret = btrfs_repair_one_sector(inode, failed_bio,
-				bio_offset + offset,
-				page, pgoff + offset, start + offset,
-				failed_mirror, btrfs_submit_data_bio);
+		ret = btrfs_repair_one_sector(inode, failed_bbio,
+				bio_offset + offset, page, pgoff + offset,
+				btrfs_submit_data_read_bio);
 		if (!ret) {
 			/*
 			 * We have submitted the read repair, the page release
@@ -2785,24 +2780,12 @@ static blk_status_t submit_data_read_repair(struct inode *inode,
 			continue;
 		}
 		/*
-		 * Repair failed, just record the error but still continue.
-		 * Or the remaining sectors will not be properly unlocked.
+		 * Continue on failed repair, otherwise the remaining sectors
+		 * will not be properly unlocked.
 		 */
-		if (!error)
-			error = ret;
 next:
-		end_page_read(page, uptodate, start + offset, sectorsize);
-		if (uptodate)
-			set_extent_uptodate(&BTRFS_I(inode)->io_tree,
-					start + offset,
-					start + offset + sectorsize - 1,
-					&cached, GFP_ATOMIC);
-		unlock_extent_cached_atomic(&BTRFS_I(inode)->io_tree,
-				start + offset,
-				start + offset + sectorsize - 1,
-				&cached);
+		end_sector_io(page, start + offset, uptodate);
 	}
-	return errno_to_blk_status(error);
 }
 
 /* lots and lots of room for performance fixes in the end_bio funcs */
@@ -3017,7 +3000,6 @@ static void end_bio_extent_readpage(struct bio *bio)
 	 */
 	u32 bio_offset = 0;
 	int mirror;
-	int ret;
 	struct bvec_iter_all iter_all;
 
 	ASSERT(!bio_flagged(bio, BIO_CLONED));
@@ -3028,6 +3010,7 @@ static void end_bio_extent_readpage(struct bio *bio)
 		struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
 		const u32 sectorsize = fs_info->sectorsize;
 		unsigned int error_bitmap = (unsigned int)-1;
+		bool repair = false;
 		u64 start;
 		u64 end;
 		u32 len;
@@ -3065,57 +3048,23 @@ static void end_bio_extent_readpage(struct bio *bio)
 			if (is_data_inode(inode)) {
 				error_bitmap = btrfs_verify_data_csum(bbio,
 						bio_offset, page, start, end);
-				ret = error_bitmap;
+				if (error_bitmap)
+					uptodate = false;
 			} else {
-				ret = btrfs_validate_metadata_buffer(bbio,
-					page, start, end, mirror);
+				if (btrfs_validate_metadata_buffer(bbio,
+						page, start, end, mirror))
+					uptodate = false;
 			}
-			if (ret)
-				uptodate = false;
-			else
-				clean_io_failure(BTRFS_I(inode)->root->fs_info,
-						 failure_tree, tree, start,
-						 page,
-						 btrfs_ino(BTRFS_I(inode)), 0);
 		}
 
-		if (likely(uptodate))
-			goto readpage_ok;
-
-		if (is_data_inode(inode)) {
-			/*
-			 * If we failed to submit the IO at all we'll have a
-			 * mirror_num == 0, in which case we need to just mark
-			 * the page with an error and unlock it and carry on.
-			 */
-			if (mirror == 0)
-				goto readpage_ok;
-
-			/*
-			 * submit_data_read_repair() will handle all the good
-			 * and bad sectors, we just continue to the next bvec.
-			 */
-			submit_data_read_repair(inode, bio, bio_offset, page,
-						start - page_offset(page),