From 7d7ea89e756ea18a3b08cd396e2a4c0c12d473a8 Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Fri, 16 Aug 2013 21:20:41 -0400
Subject: ext4: refactor code to read the extent tree block

Refactor out the code needed to read the extent tree block into a
single read_extent_tree_block() function.  In addition to simplifying
the code, it also makes sure that we call the ext4_ext_load_extent
tracepoint whenever we need to read an extent tree block from disk.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Zheng Liu <wenqing.lz@taobao.com>
---
 fs/ext4/extents.c | 97 ++++++++++++++++++++++++-------------------------------
 1 file changed, 43 insertions(+), 54 deletions(-)

(limited to 'fs/ext4/extents.c')

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 72ba4705d4f..a40be59ddce 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -464,25 +464,39 @@ int ext4_ext_check_inode(struct inode *inode)
 	return ext4_ext_check(inode, ext_inode_hdr(inode), ext_depth(inode));
 }
 
-static int __ext4_ext_check_block(const char *function, unsigned int line,
-				  struct inode *inode,
-				  struct ext4_extent_header *eh,
-				  int depth,
-				  struct buffer_head *bh)
+static struct buffer_head *
+__read_extent_tree_block(const char *function, unsigned int line,
+			 struct inode *inode, ext4_fsblk_t pblk, int depth)
 {
-	int ret;
+	struct buffer_head		*bh;
+	int				err;
 
+	bh = sb_getblk(inode->i_sb, pblk);
+	if (unlikely(!bh))
+		return ERR_PTR(-ENOMEM);
+
+	if (!bh_uptodate_or_lock(bh)) {
+		trace_ext4_ext_load_extent(inode, pblk, _RET_IP_);
+		err = bh_submit_read(bh);
+		if (err < 0)
+			goto errout;
+	}
 	if (buffer_verified(bh))
-		return 0;
-	ret = ext4_ext_check(inode, eh, depth);
-	if (ret)
-		return ret;
+		return bh;
+	err = __ext4_ext_check(function, line, inode,
+			       ext_block_hdr(bh), depth);
+	if (err)
+		goto errout;
 	set_buffer_verified(bh);
-	return ret;
+	return bh;
+errout:
+	put_bh(bh);
+	return ERR_PTR(err);
+
 }
 
-#define ext4_ext_check_block(inode, eh, depth, bh)	\
-	__ext4_ext_check_block(__func__, __LINE__, inode, eh, depth, bh)
+#define read_extent_tree_block(inode, pblk, depth)		\
+	__read_extent_tree_block(__func__, __LINE__, (inode), (pblk), (depth))
 
 #ifdef EXT_DEBUG
 static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path)
@@ -748,20 +762,12 @@ ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block,
 		path[ppos].p_depth = i;
 		path[ppos].p_ext = NULL;
 
-		bh = sb_getblk(inode->i_sb, path[ppos].p_block);
-		if (unlikely(!bh)) {
-			ret = -ENOMEM;
+		bh = read_extent_tree_block(inode, path[ppos].p_block, --i);
+		if (IS_ERR(bh)) {
+			ret = PTR_ERR(bh);
 			goto err;
 		}
-		if (!bh_uptodate_or_lock(bh)) {
-			trace_ext4_ext_load_extent(inode, block,
-						path[ppos].p_block);
-			ret = bh_submit_read(bh);
-			if (ret < 0) {
-				put_bh(bh);
-				goto err;
-			}
-		}
+
 		eh = ext_block_hdr(bh);
 		ppos++;
 		if (unlikely(ppos > depth)) {
@@ -773,11 +779,6 @@ ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block,
 		}
 		path[ppos].p_bh = bh;
 		path[ppos].p_hdr = eh;
-		i--;
-
-		ret = ext4_ext_check_block(inode, eh, i, bh);
-		if (ret < 0)
-			goto err;
 	}
 
 	path[ppos].p_depth = i;
@@ -1412,29 +1413,21 @@ got_index:
 	ix++;
 	block = ext4_idx_pblock(ix);
 	while (++depth < path->p_depth) {
-		bh = sb_bread(inode->i_sb, block);
-		if (bh == NULL)
-			return -EIO;
-		eh = ext_block_hdr(bh);
 		/* subtract from p_depth to get proper eh_depth */
-		if (ext4_ext_check_block(inode, eh,
-					 path->p_depth - depth, bh)) {
-			put_bh(bh);
-			return -EIO;
-		}
+		bh = read_extent_tree_block(inode, block,
+					    path->p_depth - depth);
+		if (IS_ERR(bh))
+			return PTR_ERR(bh);
+		eh = ext_block_hdr(bh);
 		ix = EXT_FIRST_INDEX(eh);
 		block = ext4_idx_pblock(ix);
 		put_bh(bh);
 	}
 
-	bh = sb_bread(inode->i_sb, block);
-	if (bh == NULL)
-		return -EIO;
+	bh = read_extent_tree_block(inode, block, path->p_depth - depth);
+	if (IS_ERR(bh))
+		return PTR_ERR(bh);
 	eh = ext_block_hdr(bh);
-	if (ext4_ext_check_block(inode, eh, path->p_depth - depth, bh)) {
-		put_bh(bh);
-		return -EIO;
-	}
 	ex = EXT_FIRST_EXTENT(eh);
 found_extent:
 	*logical = le32_to_cpu(ex->ee_block);
@@ -2829,10 +2822,11 @@ again:
 			ext_debug("move to level %d (block %llu)\n",
 				  i + 1, ext4_idx_pblock(path[i].p_idx));
 			memset(path + i + 1, 0, sizeof(*path));
-			bh = sb_bread(sb, ext4_idx_pblock(path[i].p_idx));
-			if (!bh) {
+			bh = read_extent_tree_block(inode,
+				ext4_idx_pblock(path[i].p_idx), depth - i - 1);
+			if (IS_ERR(bh)) {
 				/* should we reset i_size? */
-				err = -EIO;
+				err = PTR_ERR(bh);
 				break;
 			}
 			/* Yield here to deal with large extent trees.
@@ -2842,11 +2836,6 @@ again:
 				err = -EIO;
 				break;
 			}
-			if (ext4_ext_check_block(inode, ext_block_hdr(bh),
-							depth - i - 1, bh)) {
-				err = -EIO;
-				break;
-			}
 			path[i + 1].p_bh = bh;
 
 			/* save actual number of indexes since this
-- 
cgit v1.2.3-70-g09d2


From c349179b4808f7c8e1ff1b4dd967c047eefd24bc Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Fri, 16 Aug 2013 21:21:41 -0400
Subject: ext4: print the block number of invalid extent tree blocks

When we find an invalid extent tree block, report the block number of
the bad block for debugging purposes.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Zheng Liu <wenqing.lz@taobao.com>
---
 fs/ext4/extents.c | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

(limited to 'fs/ext4/extents.c')

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index a40be59ddce..6e7b7d928f4 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -407,7 +407,7 @@ static int ext4_valid_extent_entries(struct inode *inode,
 
 static int __ext4_ext_check(const char *function, unsigned int line,
 			    struct inode *inode, struct ext4_extent_header *eh,
-			    int depth)
+			    int depth, ext4_fsblk_t pblk)
 {
 	const char *error_msg;
 	int max = 0;
@@ -447,21 +447,21 @@ static int __ext4_ext_check(const char *function, unsigned int line,
 
 corrupted:
 	ext4_error_inode(inode, function, line, 0,
-			"bad header/extent: %s - magic %x, "
-			"entries %u, max %u(%u), depth %u(%u)",
-			error_msg, le16_to_cpu(eh->eh_magic),
-			le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max),
-			max, le16_to_cpu(eh->eh_depth), depth);
-
+			 "pblk %llu bad header/extent: %s - magic %x, "
+			 "entries %u, max %u(%u), depth %u(%u)",
+			 (unsigned long long) pblk, error_msg,
+			 le16_to_cpu(eh->eh_magic),
+			 le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max),
+			 max, le16_to_cpu(eh->eh_depth), depth);
 	return -EIO;
 }
 
-#define ext4_ext_check(inode, eh, depth)	\
-	__ext4_ext_check(__func__, __LINE__, inode, eh, depth)
+#define ext4_ext_check(inode, eh, depth, pblk)			\
+	__ext4_ext_check(__func__, __LINE__, (inode), (eh), (depth), (pblk))
 
 int ext4_ext_check_inode(struct inode *inode)
 {
-	return ext4_ext_check(inode, ext_inode_hdr(inode), ext_depth(inode));
+	return ext4_ext_check(inode, ext_inode_hdr(inode), ext_depth(inode), 0);
 }
 
 static struct buffer_head *
@@ -484,7 +484,7 @@ __read_extent_tree_block(const char *function, unsigned int line,
 	if (buffer_verified(bh))
 		return bh;
 	err = __ext4_ext_check(function, line, inode,
-			       ext_block_hdr(bh), depth);
+			       ext_block_hdr(bh), depth, pblk);
 	if (err)
 		goto errout;
 	set_buffer_verified(bh);
@@ -2775,7 +2775,7 @@ again:
 		path[0].p_hdr = ext_inode_hdr(inode);
 		i = 0;
 
-		if (ext4_ext_check(inode, path[0].p_hdr, depth)) {
+		if (ext4_ext_check(inode, path[0].p_hdr, depth, 0)) {
 			err = -EIO;
 			goto out;
 		}
-- 
cgit v1.2.3-70-g09d2


From 107a7bd31ac003e42c0f966aa8e5b26947de6024 Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Fri, 16 Aug 2013 21:23:41 -0400
Subject: ext4: cache all of an extent tree's leaf block upon reading

When we read in an extent tree leaf block from disk, arrange to have
all of its entries cached.  In nearly all cases the in-memory
representation will be more compact than the on-disk representation in
the buffer cache, and it allows us to get the information without
having to traverse the extent tree for successive extents.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Zheng Liu <wenqing.lz@taobao.com>
---
 fs/ext4/ext4.h              | 14 +++++++-
 fs/ext4/extents.c           | 87 +++++++++++++++++++++++++++++++--------------
 fs/ext4/extents_status.c    | 37 ++++++++++++++++++-
 fs/ext4/extents_status.h    |  3 ++
 fs/ext4/migrate.c           |  2 +-
 fs/ext4/move_extent.c       |  2 +-
 include/trace/events/ext4.h | 14 +++++++-
 7 files changed, 127 insertions(+), 32 deletions(-)

(limited to 'fs/ext4/extents.c')

diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 0ab26fbf338..c74b1948feb 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -560,6 +560,17 @@ enum {
 	/* Do not put hole in extent cache */
 #define EXT4_GET_BLOCKS_NO_PUT_HOLE		0x0200
 
+/*
+ * The bit position of this flag must not overlap with any of the
+ * EXT4_GET_BLOCKS_*.  It is used by ext4_ext_find_extent(),
+ * read_extent_tree_block(), ext4_split_extent_at(),
+ * ext4_ext_insert_extent(), and ext4_ext_create_new_leaf() to
+ * indicate that the we shouldn't be caching the extents when reading
+ * from the extent tree while a truncate or punch hole operation
+ * is in progress.
+ */
+#define EXT4_EX_NOCACHE				0x0400
+
 /*
  * Flags used by ext4_free_blocks
  */
@@ -2684,7 +2695,8 @@ extern int ext4_ext_insert_extent(handle_t *, struct inode *,
 				  struct ext4_ext_path *,
 				  struct ext4_extent *, int);
 extern struct ext4_ext_path *ext4_ext_find_extent(struct inode *, ext4_lblk_t,
-						  struct ext4_ext_path *);
+						  struct ext4_ext_path *,
+						  int flags);
 extern void ext4_ext_drop_refs(struct ext4_ext_path *);
 extern int ext4_ext_check_inode(struct inode *inode);
 extern int ext4_find_delalloc_range(struct inode *inode,
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 6e7b7d928f4..08c1ac97647 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -466,7 +466,8 @@ int ext4_ext_check_inode(struct inode *inode)
 
 static struct buffer_head *
 __read_extent_tree_block(const char *function, unsigned int line,
-			 struct inode *inode, ext4_fsblk_t pblk, int depth)
+			 struct inode *inode, ext4_fsblk_t pblk, int depth,
+			 int flags)
 {
 	struct buffer_head		*bh;
 	int				err;
@@ -488,6 +489,32 @@ __read_extent_tree_block(const char *function, unsigned int line,
 	if (err)
 		goto errout;
 	set_buffer_verified(bh);
+	/*
+	 * If this is a leaf block, cache all of its entries
+	 */
+	if (!(flags & EXT4_EX_NOCACHE) && depth == 0) {
+		struct ext4_extent_header *eh = ext_block_hdr(bh);
+		struct ext4_extent *ex = EXT_FIRST_EXTENT(eh);
+		ext4_lblk_t prev = 0;
+		int i;
+
+		for (i = le16_to_cpu(eh->eh_entries); i > 0; i--, ex++) {
+			unsigned int status = EXTENT_STATUS_WRITTEN;
+			ext4_lblk_t lblk = le32_to_cpu(ex->ee_block);
+			int len = ext4_ext_get_actual_len(ex);
+
+			if (prev && (prev != lblk))
+				ext4_es_cache_extent(inode, prev,
+						     lblk - prev, ~0,
+						     EXTENT_STATUS_HOLE);
+
+			if (ext4_ext_is_uninitialized(ex))
+				status = EXTENT_STATUS_UNWRITTEN;
+			ext4_es_cache_extent(inode, lblk, len,
+					     ext4_ext_pblock(ex), status);
+			prev = lblk + len;
+		}
+	}
 	return bh;
 errout:
 	put_bh(bh);
@@ -495,8 +522,9 @@ errout:
 
 }
 
-#define read_extent_tree_block(inode, pblk, depth)		\
-	__read_extent_tree_block(__func__, __LINE__, (inode), (pblk), (depth))
+#define read_extent_tree_block(inode, pblk, depth, flags)		\
+	__read_extent_tree_block(__func__, __LINE__, (inode), (pblk),   \
+				 (depth), (flags))
 
 #ifdef EXT_DEBUG
 static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path)
@@ -730,7 +758,7 @@ int ext4_ext_tree_init(handle_t *handle, struct inode *inode)
 
 struct ext4_ext_path *
 ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block,
-					struct ext4_ext_path *path)
+		     struct ext4_ext_path *path, int flags)
 {
 	struct ext4_extent_header *eh;
 	struct buffer_head *bh;
@@ -762,7 +790,8 @@ ext4_ext_find_extent(struct inode *inode, ext4_lblk_t block,
 		path[ppos].p_depth = i;
 		path[ppos].p_ext = NULL;
 
-		bh = read_extent_tree_block(inode, path[ppos].p_block, --i);
+		bh = read_extent_tree_block(inode, path[ppos].p_block, --i,
+					    flags);
 		if (IS_ERR(bh)) {
 			ret = PTR_ERR(bh);
 			goto err;
@@ -1199,7 +1228,8 @@ out:
  * if no free index is found, then it requests in-depth growing.
  */
 static int ext4_ext_create_new_leaf(handle_t *handle, struct inode *inode,
-				    unsigned int flags,
+				    unsigned int mb_flags,
+				    unsigned int gb_flags,
 				    struct ext4_ext_path *path,
 				    struct ext4_extent *newext)
 {
@@ -1221,7 +1251,7 @@ repeat:
 	if (EXT_HAS_FREE_INDEX(curp)) {
 		/* if we found index with free entry, then use that
 		 * entry: create all needed subtree and add new leaf */
-		err = ext4_ext_split(handle, inode, flags, path, newext, i);
+		err = ext4_ext_split(handle, inode, mb_flags, path, newext, i);
 		if (err)
 			goto out;
 
@@ -1229,12 +1259,12 @@ repeat:
 		ext4_ext_drop_refs(path);
 		path = ext4_ext_find_extent(inode,
 				    (ext4_lblk_t)le32_to_cpu(newext->ee_block),
-				    path);
+				    path, gb_flags);
 		if (IS_ERR(path))
 			err = PTR_ERR(path);
 	} else {
 		/* tree is full, time to grow in depth */
-		err = ext4_ext_grow_indepth(handle, inode, flags, newext);
+		err = ext4_ext_grow_indepth(handle, inode, mb_flags, newext);
 		if (err)
 			goto out;
 
@@ -1242,7 +1272,7 @@ repeat:
 		ext4_ext_drop_refs(path);
 		path = ext4_ext_find_extent(inode,
 				   (ext4_lblk_t)le32_to_cpu(newext->ee_block),
-				    path);
+				    path, gb_flags);
 		if (IS_ERR(path)) {
 			err = PTR_ERR(path);
 			goto out;
@@ -1415,7 +1445,7 @@ got_index:
 	while (++depth < path->p_depth) {
 		/* subtract from p_depth to get proper eh_depth */
 		bh = read_extent_tree_block(inode, block,
-					    path->p_depth - depth);
+					    path->p_depth - depth, 0);
 		if (IS_ERR(bh))
 			return PTR_ERR(bh);
 		eh = ext_block_hdr(bh);
@@ -1424,7 +1454,7 @@ got_index:
 		put_bh(bh);
 	}
 
-	bh = read_extent_tree_block(inode, block, path->p_depth - depth);
+	bh = read_extent_tree_block(inode, block, path->p_depth - depth, 0);
 	if (IS_ERR(bh))
 		return PTR_ERR(bh);
 	eh = ext_block_hdr(bh);
@@ -1786,7 +1816,7 @@ out:
  */
 int ext4_ext_insert_extent(handle_t *handle, struct inode *inode,
 				struct ext4_ext_path *path,
-				struct ext4_extent *newext, int flag)
+				struct ext4_extent *newext, int gb_flags)
 {
 	struct ext4_extent_header *eh;
 	struct ext4_extent *ex, *fex;
@@ -1795,7 +1825,7 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode,
 	int depth, len, err;
 	ext4_lblk_t next;
 	unsigned uninitialized = 0;
-	int flags = 0;
+	int mb_flags = 0;
 
 	if (unlikely(ext4_ext_get_actual_len(newext) == 0)) {
 		EXT4_ERROR_INODE(inode, "ext4_ext_get_actual_len(newext) == 0");
@@ -1810,7 +1840,7 @@ int ext4_ext_insert_extent(handle_t *handle, struct inode *inode,
 	}
 
 	/* try to insert block into found extent and return */
-	if (ex && !(flag & EXT4_GET_BLOCKS_PRE_IO)) {
+	if (ex && !(gb_flags & EXT4_GET_BLOCKS_PRE_IO)) {
 
 		/*
 		 * Try to see whether we should rather test the extent on
@@ -1913,7 +1943,7 @@ prepend:
 	if (next != EXT_MAX_BLOCKS) {
 		ext_debug("next leaf block - %u\n", next);
 		BUG_ON(npath != NULL);
-		npath = ext4_ext_find_extent(inode, next, NULL);
+		npath = ext4_ext_find_extent(inode, next, NULL, 0);
 		if (IS_ERR(npath))
 			return PTR_ERR(npath);
 		BUG_ON(npath->p_depth != path->p_depth);
@@ -1932,9 +1962,10 @@ prepend:
 	 * There is no free space in the found leaf.
 	 * We're gonna add a new leaf in the tree.
 	 */
-	if (flag & EXT4_GET_BLOCKS_METADATA_NOFAIL)
-		flags = EXT4_MB_USE_RESERVED;
-	err = ext4_ext_create_new_leaf(handle, inode, flags, path, newext);
+	if (gb_flags & EXT4_GET_BLOCKS_METADATA_NOFAIL)
+		mb_flags = EXT4_MB_USE_RESERVED;
+	err = ext4_ext_create_new_leaf(handle, inode, mb_flags, gb_flags,
+				       path, newext);
 	if (err)
 		goto cleanup;
 	depth = ext_depth(inode);
@@ -2000,7 +2031,7 @@ has_space:
 
 merge:
 	/* try to merge extents */
-	if (!(flag & EXT4_GET_BLOCKS_PRE_IO))
+	if (!(gb_flags & EXT4_GET_BLOCKS_PRE_IO))
 		ext4_ext_try_to_merge(handle, inode, path, nearex);
 
 
@@ -2043,7 +2074,7 @@ static int ext4_fill_fiemap_extents(struct inode *inode,
 			path = NULL;
 		}
 
-		path = ext4_ext_find_extent(inode, block, path);
+		path = ext4_ext_find_extent(inode, block, path, 0);
 		if (IS_ERR(path)) {
 			up_read(&EXT4_I(inode)->i_data_sem);
 			err = PTR_ERR(path);
@@ -2705,7 +2736,7 @@ again:
 		ext4_lblk_t ee_block;
 
 		/* find extent for this block */
-		path = ext4_ext_find_extent(inode, end, NULL);
+		path = ext4_ext_find_extent(inode, end, NULL, EXT4_EX_NOCACHE);
 		if (IS_ERR(path)) {
 			ext4_journal_stop(handle);
 			return PTR_ERR(path);
@@ -2747,6 +2778,7 @@ again:
 			 */
 			err = ext4_split_extent_at(handle, inode, path,
 					end + 1, split_flag,
+					EXT4_EX_NOCACHE |
 					EXT4_GET_BLOCKS_PRE_IO |
 					EXT4_GET_BLOCKS_METADATA_NOFAIL);
 
@@ -2823,7 +2855,8 @@ again:
 				  i + 1, ext4_idx_pblock(path[i].p_idx));
 			memset(path + i + 1, 0, sizeof(*path));
 			bh = read_extent_tree_block(inode,
-				ext4_idx_pblock(path[i].p_idx), depth - i - 1);
+				ext4_idx_pblock(path[i].p_idx), depth - i - 1,
+				EXT4_EX_NOCACHE);
 			if (IS_ERR(bh)) {
 				/* should we reset i_size? */
 				err = PTR_ERR(bh);
@@ -3170,7 +3203,7 @@ static int ext4_split_extent(handle_t *handle,
 	 * result in split of original leaf or extent zeroout.
 	 */
 	ext4_ext_drop_refs(path);
-	path = ext4_ext_find_extent(inode, map->m_lblk, path);
+	path = ext4_ext_find_extent(inode, map->m_lblk, path, 0);
 	if (IS_ERR(path))
 		return PTR_ERR(path);
 	depth = ext_depth(inode);
@@ -3554,7 +3587,7 @@ static int ext4_convert_unwritten_extents_endio(handle_t *handle,
 		if (err < 0)
 			goto out;
 		ext4_ext_drop_refs(path);
-		path = ext4_ext_find_extent(inode, map->m_lblk, path);
+		path = ext4_ext_find_extent(inode, map->m_lblk, path, 0);
 		if (IS_ERR(path)) {
 			err = PTR_ERR(path);
 			goto out;
@@ -4041,7 +4074,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 	trace_ext4_ext_map_blocks_enter(inode, map->m_lblk, map->m_len, flags);
 
 	/* find extent for this block */
-	path = ext4_ext_find_extent(inode, map->m_lblk, NULL);
+	path = ext4_ext_find_extent(inode, map->m_lblk, NULL, 0);
 	if (IS_ERR(path)) {
 		err = PTR_ERR(path);
 		path = NULL;
@@ -4760,6 +4793,6 @@ int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 		error = ext4_fill_fiemap_extents(inode, start_blk,
 						 len_blks, fieinfo);
 	}
-
+	ext4_es_lru_add(inode);
 	return error;
 }
diff --git a/fs/ext4/extents_status.c b/fs/ext4/extents_status.c
index ded2615b63e..1dc5df016e2 100644
--- a/fs/ext4/extents_status.c
+++ b/fs/ext4/extents_status.c
@@ -419,7 +419,7 @@ static void ext4_es_insert_extent_ext_check(struct inode *inode,
 	unsigned short ee_len;
 	int depth, ee_status, es_status;
 
-	path = ext4_ext_find_extent(inode, es->es_lblk, NULL);
+	path = ext4_ext_find_extent(inode, es->es_lblk, NULL, EXT4_EX_NOCACHE);
 	if (IS_ERR(path))
 		return;
 
@@ -683,6 +683,41 @@ error:
 	return err;
 }
 
+/*
+ * ext4_es_cache_extent() inserts information into the extent status
+ * tree if and only if there isn't information about the range in
+ * question already.
+ */
+void ext4_es_cache_extent(struct inode *inode, ext4_lblk_t lblk,
+			  ext4_lblk_t len, ext4_fsblk_t pblk,
+			  unsigned int status)
+{
+	struct extent_status *es;
+	struct extent_status newes;
+	ext4_lblk_t end = lblk + len - 1;
+
+	newes.es_lblk = lblk;
+	newes.es_len = len;
+	ext4_es_store_pblock(&newes, pblk);
+	ext4_es_store_status(&newes, status);
+	trace_ext4_es_cache_extent(inode, &newes);
+
+	if (!len)
+		return;
+
+	BUG_ON(end < lblk);
+
+	write_lock(&EXT4_I(inode)->i_es_lock);
+
+	es = __es_tree_search(&EXT4_I(inode)->i_es_tree.root, lblk);
+	if (es && ((es->es_lblk <= lblk) || (es->es_lblk <= end)))
+		goto out;
+
+	__es_insert_extent(inode, &newes);
+out:
+	write_unlock(&EXT4_I(inode)->i_es_lock);
+}
+
 /*
  * ext4_es_lookup_extent() looks up an extent in extent status tree.
  *
diff --git a/fs/ext4/extents_status.h b/fs/ext4/extents_status.h
index d72af848f98..3e83aef3653 100644
--- a/fs/ext4/extents_status.h
+++ b/fs/ext4/extents_status.h
@@ -71,6 +71,9 @@ extern void ext4_es_init_tree(struct ext4_es_tree *tree);
 extern int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk,
 				 ext4_lblk_t len, ext4_fsblk_t pblk,
 				 unsigned int status);
+extern void ext4_es_cache_extent(struct inode *inode, ext4_lblk_t lblk,
+				 ext4_lblk_t len, ext4_fsblk_t pblk,
+				 unsigned int status);
 extern int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
 				 ext4_lblk_t len);
 extern void ext4_es_find_delayed_extent_range(struct inode *inode,
diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c
index 49e8bdff916..f99bdb8548b 100644
--- a/fs/ext4/migrate.c
+++ b/fs/ext4/migrate.c
@@ -39,7 +39,7 @@ static int finish_range(handle_t *handle, struct inode *inode,
 	newext.ee_block = cpu_to_le32(lb->first_block);
 	newext.ee_len   = cpu_to_le16(lb->last_block - lb->first_block + 1);
 	ext4_ext_store_pblock(&newext, lb->first_pblock);
-	path = ext4_ext_find_extent(inode, lb->first_block, NULL);
+	path = ext4_ext_find_extent(inode, lb->first_block, NULL, 0);
 
 	if (IS_ERR(path)) {
 		retval = PTR_ERR(path);
diff --git a/fs/ext4/move_extent.c b/fs/ext4/move_extent.c
index e86dddbd829..7fa4d855dbd 100644
--- a/fs/ext4/move_extent.c
+++ b/fs/ext4/move_extent.c
@@ -37,7 +37,7 @@ get_ext_path(struct inode *inode, ext4_lblk_t lblock,
 	int ret = 0;
 	struct ext4_ext_path *path;
 
-	path = ext4_ext_find_extent(inode, lblock, *orig_path);
+	path = ext4_ext_find_extent(inode, lblock, *orig_path, EXT4_EX_NOCACHE);
 	if (IS_ERR(path))
 		ret = PTR_ERR(path);
 	else if (path[ext_depth(inode)].p_ext == NULL)
diff --git a/include/trace/events/ext4.h b/include/trace/events/ext4.h
index 47a355b251e..d892b55d91a 100644
--- a/include/trace/events/ext4.h
+++ b/include/trace/events/ext4.h
@@ -2192,7 +2192,7 @@ TRACE_EVENT(ext4_ext_remove_space_done,
 		  (unsigned short) __entry->eh_entries)
 );
 
-TRACE_EVENT(ext4_es_insert_extent,
+DECLARE_EVENT_CLASS(ext4__es_extent,
 	TP_PROTO(struct inode *inode, struct extent_status *es),
 
 	TP_ARGS(inode, es),
@@ -2222,6 +2222,18 @@ TRACE_EVENT(ext4_es_insert_extent,
 		  __entry->pblk, show_extent_status(__entry->status))
 );
 
+DEFINE_EVENT(ext4__es_extent, ext4_es_insert_extent,
+	TP_PROTO(struct inode *inode, struct extent_status *es),
+
+	TP_ARGS(inode, es)
+);
+
+DEFINE_EVENT(ext4__es_extent, ext4_es_cache_extent,
+	TP_PROTO(struct inode *inode, struct extent_status *es),
+
+	TP_ARGS(inode, es)
+);
+
 TRACE_EVENT(ext4_es_remove_extent,
 	TP_PROTO(struct inode *inode, ext4_lblk_t lblk, ext4_lblk_t len),
 
-- 
cgit v1.2.3-70-g09d2


From 7869a4a6c5caa7b2e5c41ccaf46eb3371f88eea7 Mon Sep 17 00:00:00 2001
From: Theodore Ts'o <tytso@mit.edu>
Date: Fri, 16 Aug 2013 22:05:14 -0400
Subject: ext4: add support for extent pre-caching

Add a new fiemap flag which forces the all of the extents in an inode
to be cached in the extent_status tree.  This is critically important
when using AIO to a preallocated file, since if we need to read in
blocks from the extent tree, the io_submit(2) system call becomes
synchronous, and the AIO is no longer "A", which is bad.

In addition, for most files which have an external leaf tree block,
the cost of caching the information in the extent status tree will be
less than caching the entire 4k block in the buffer cache.  So it is
generally a win to keep the extent information cached.

Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/ext4/ext4.h              | 17 ++++++-----
 fs/ext4/extents.c           | 73 ++++++++++++++++++++++++++++++++++++++++++++-
 fs/ext4/extents_status.c    | 72 +++++++++++++++++++++++++++++++-------------
 fs/ext4/ioctl.c             |  3 ++
 include/uapi/linux/fiemap.h |  1 +
 5 files changed, 137 insertions(+), 29 deletions(-)

(limited to 'fs/ext4/extents.c')

diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index c74b1948feb..635135e6148 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -561,15 +561,16 @@ enum {
 #define EXT4_GET_BLOCKS_NO_PUT_HOLE		0x0200
 
 /*
- * The bit position of this flag must not overlap with any of the
- * EXT4_GET_BLOCKS_*.  It is used by ext4_ext_find_extent(),
+ * The bit position of these flags must not overlap with any of the
+ * EXT4_GET_BLOCKS_*.  They are used by ext4_ext_find_extent(),
  * read_extent_tree_block(), ext4_split_extent_at(),
- * ext4_ext_insert_extent(), and ext4_ext_create_new_leaf() to
- * indicate that the we shouldn't be caching the extents when reading
- * from the extent tree while a truncate or punch hole operation
- * is in progress.
+ * ext4_ext_insert_extent(), and ext4_ext_create_new_leaf().
+ * EXT4_EX_NOCACHE is used to indicate that the we shouldn't be
+ * caching the extents when reading from the extent tree while a
+ * truncate or punch hole operation is in progress.
  */
 #define EXT4_EX_NOCACHE				0x0400
+#define EXT4_EX_FORCE_CACHE			0x0800
 
 /*
  * Flags used by ext4_free_blocks
@@ -601,6 +602,7 @@ enum {
 #define EXT4_IOC_MOVE_EXT		_IOWR('f', 15, struct move_extent)
 #define EXT4_IOC_RESIZE_FS		_IOW('f', 16, __u64)
 #define EXT4_IOC_SWAP_BOOT		_IO('f', 17)
+#define EXT4_IOC_PRECACHE_EXTENTS	_IO('f', 18)
 
 #if defined(__KERNEL__) && defined(CONFIG_COMPAT)
 /*
@@ -1386,6 +1388,7 @@ enum {
 					   nolocking */
 	EXT4_STATE_MAY_INLINE_DATA,	/* may have in-inode data */
 	EXT4_STATE_ORDERED_MODE,	/* data=ordered mode */
+	EXT4_STATE_EXT_PRECACHED,	/* extents have been precached */
 };
 
 #define EXT4_INODE_BIT_FNS(name, field, offset)				\
@@ -2705,7 +2708,7 @@ extern int ext4_find_delalloc_range(struct inode *inode,
 extern int ext4_find_delalloc_cluster(struct inode *inode, ext4_lblk_t lblk);
 extern int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			__u64 start, __u64 len);
-
+extern int ext4_ext_precache(struct inode *inode);
 
 /* move_extent.c */
 extern void ext4_double_down_write_data_sem(struct inode *first,
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 08c1ac97647..01838875fca 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -482,7 +482,7 @@ __read_extent_tree_block(const char *function, unsigned int line,
 		if (err < 0)
 			goto errout;
 	}
-	if (buffer_verified(bh))
+	if (buffer_verified(bh) && !(flags & EXT4_EX_FORCE_CACHE))
 		return bh;
 	err = __ext4_ext_check(function, line, inode,
 			       ext_block_hdr(bh), depth, pblk);
@@ -526,6 +526,71 @@ errout:
 	__read_extent_tree_block(__func__, __LINE__, (inode), (pblk),   \
 				 (depth), (flags))
 
+/*
+ * This function is called to cache a file's extent information in the
+ * extent status tree
+ */
+int ext4_ext_precache(struct inode *inode)
+{
+	struct ext4_inode_info *ei = EXT4_I(inode);
+	struct ext4_ext_path *path = NULL;
+	struct buffer_head *bh;
+	int i = 0, depth, ret = 0;
+
+	if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
+		return 0;	/* not an extent-mapped inode */
+
+	down_read(&ei->i_data_sem);
+	depth = ext_depth(inode);
+
+	path = kzalloc(sizeof(struct ext4_ext_path) * (depth + 1),
+		       GFP_NOFS);
+	if (path == NULL) {
+		up_read(&ei->i_data_sem);
+		return -ENOMEM;
+	}
+
+	/* Don't cache anything if there are no external extent blocks */
+	if (depth == 0)
+		goto out;
+	path[0].p_hdr = ext_inode_hdr(inode);
+	ret = ext4_ext_check(inode, path[0].p_hdr, depth, 0);
+	if (ret)
+		goto out;
+	path[0].p_idx = EXT_FIRST_INDEX(path[0].p_hdr);
+	while (i >= 0) {
+		/*
+		 * If this is a leaf block or we've reached the end of
+		 * the index block, go up
+		 */
+		if ((i == depth) ||
+		    path[i].p_idx > EXT_LAST_INDEX(path[i].p_hdr)) {
+			brelse(path[i].p_bh);
+			path[i].p_bh = NULL;
+			i--;
+			continue;
+		}
+		bh = read_extent_tree_block(inode,
+					    ext4_idx_pblock(path[i].p_idx++),
+					    depth - i - 1,
+					    EXT4_EX_FORCE_CACHE);
+		if (IS_ERR(bh)) {
+			ret = PTR_ERR(bh);
+			break;
+		}
+		i++;
+		path[i].p_bh = bh;
+		path[i].p_hdr = ext_block_hdr(bh);
+		path[i].p_idx = EXT_FIRST_INDEX(path[i].p_hdr);
+	}
+	ext4_set_inode_state(inode, EXT4_STATE_EXT_PRECACHED);
+out:
+	up_read(&ei->i_data_sem);
+	ext4_ext_drop_refs(path);
+	kfree(path);
+	return ret;
+}
+
 #ifdef EXT_DEBUG
 static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path)
 {
@@ -4766,6 +4831,12 @@ int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			return error;
 	}
 
+	if (fieinfo->fi_flags & FIEMAP_FLAG_CACHE) {
+		error = ext4_ext_precache(inode);
+		if (error)
+			return error;
+	}
+
 	/* fallback to generic here if not in extents fmt */
 	if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)))
 		return generic_block_fiemap(inode, fieinfo, start, len,
diff --git a/fs/ext4/extents_status.c b/fs/ext4/extents_status.c
index 1dc5df016e2..0e88a367b53 100644
--- a/fs/ext4/extents_status.c
+++ b/fs/ext4/extents_status.c
@@ -710,11 +710,8 @@ void ext4_es_cache_extent(struct inode *inode, ext4_lblk_t lblk,
 	write_lock(&EXT4_I(inode)->i_es_lock);
 
 	es = __es_tree_search(&EXT4_I(inode)->i_es_tree.root, lblk);
-	if (es && ((es->es_lblk <= lblk) || (es->es_lblk <= end)))
-		goto out;
-
-	__es_insert_extent(inode, &newes);
-out:
+	if (!es || es->es_lblk > end)
+		__es_insert_extent(inode, &newes);
 	write_unlock(&EXT4_I(inode)->i_es_lock);
 }
 
@@ -930,6 +927,12 @@ static int ext4_inode_touch_time_cmp(void *priv, struct list_head *a,
 	eia = list_entry(a, struct ext4_inode_info, i_es_lru);
 	eib = list_entry(b, struct ext4_inode_info, i_es_lru);
 
+	if (ext4_test_inode_state(&eia->vfs_inode, EXT4_STATE_EXT_PRECACHED) &&
+	    !ext4_test_inode_state(&eib->vfs_inode, EXT4_STATE_EXT_PRECACHED))
+		return 1;
+	if (!ext4_test_inode_state(&eia->vfs_inode, EXT4_STATE_EXT_PRECACHED) &&
+	    ext4_test_inode_state(&eib->vfs_inode, EXT4_STATE_EXT_PRECACHED))
+		return -1;
 	if (eia->i_touch_when == eib->i_touch_when)
 		return 0;
 	if (time_after(eia->i_touch_when, eib->i_touch_when))
@@ -943,21 +946,13 @@ static int __ext4_es_shrink(struct ext4_sb_info *sbi, int nr_to_scan,
 {
 	struct ext4_inode_info *ei;
 	struct list_head *cur, *tmp;
-	LIST_HEAD(skiped);
+	LIST_HEAD(skipped);
 	int ret, nr_shrunk = 0;
+	int retried = 0, skip_precached = 1, nr_skipped = 0;
 
 	spin_lock(&sbi->s_es_lru_lock);
 
-	/*
-	 * If the inode that is at the head of LRU list is newer than
-	 * last_sorted time, that means that we need to sort this list.
-	 */
-	ei = list_first_entry(&sbi->s_es_lru, struct ext4_inode_info, i_es_lru);
-	if (sbi->s_es_last_sorted < ei->i_touch_when) {
-		list_sort(NULL, &sbi->s_es_lru, ext4_inode_touch_time_cmp);
-		sbi->s_es_last_sorted = jiffies;
-	}
-
+retry:
 	list_for_each_safe(cur, tmp, &sbi->s_es_lru) {
 		/*
 		 * If we have already reclaimed all extents from extent
@@ -968,9 +963,16 @@ static int __ext4_es_shrink(struct ext4_sb_info *sbi, int nr_to_scan,
 
 		ei = list_entry(cur, struct ext4_inode_info, i_es_lru);
 
-		/* Skip the inode that is newer than the last_sorted time */
-		if (sbi->s_es_last_sorted < ei->i_touch_when) {
-			list_move_tail(cur, &skiped);
+		/*
+		 * Skip the inode that is newer than the last_sorted
+		 * time.  Normally we try hard to avoid shrinking
+		 * precached inodes, but we will as a last resort.
+		 */
+		if ((sbi->s_es_last_sorted < ei->i_touch_when) ||
+		    (skip_precached && ext4_test_inode_state(&ei->vfs_inode,
+						EXT4_STATE_EXT_PRECACHED))) {
+			nr_skipped++;
+			list_move_tail(cur, &skipped);
 			continue;
 		}
 
@@ -990,11 +992,33 @@ static int __ext4_es_shrink(struct ext4_sb_info *sbi, int nr_to_scan,
 	}
 
 	/* Move the newer inodes into the tail of the LRU list. */
-	list_splice_tail(&skiped, &sbi->s_es_lru);
+	list_splice_tail(&skipped, &sbi->s_es_lru);
+	INIT_LIST_HEAD(&skipped);
+
+	/*
+	 * If we skipped any inodes, and we weren't able to make any
+	 * forward progress, sort the list and try again.
+	 */
+	if ((nr_shrunk == 0) && nr_skipped && !retried) {
+		retried++;
+		list_sort(NULL, &sbi->s_es_lru, ext4_inode_touch_time_cmp);
+		sbi->s_es_last_sorted = jiffies;
+		ei = list_first_entry(&sbi->s_es_lru, struct ext4_inode_info,
+				      i_es_lru);
+		/*
+		 * If there are no non-precached inodes left on the
+		 * list, start releasing precached extents.
+		 */
+		if (ext4_test_inode_state(&ei->vfs_inode,
+					  EXT4_STATE_EXT_PRECACHED))
+			skip_precached = 0;
+		goto retry;
+	}
+
 	spin_unlock(&sbi->s_es_lru_lock);
 
 	if (locked_ei && nr_shrunk == 0)
-		nr_shrunk = __es_try_to_reclaim_extents(ei, nr_to_scan);
+		nr_shrunk = __es_try_to_reclaim_extents(locked_ei, nr_to_scan);
 
 	return nr_shrunk;
 }
@@ -1069,10 +1093,16 @@ static int __es_try_to_reclaim_extents(struct ext4_inode_info *ei,
 	struct rb_node *node;
 	struct extent_status *es;
 	int nr_shrunk = 0;
+	static DEFINE_RATELIMIT_STATE(_rs, DEFAULT_RATELIMIT_INTERVAL,
+				      DEFAULT_RATELIMIT_BURST);
 
 	if (ei->i_es_lru_nr == 0)
 		return 0;
 
+	if (ext4_test_inode_state(inode, EXT4_STATE_EXT_PRECACHED) &&
+	    __ratelimit(&_rs))
+		ext4_warning(inode->i_sb, "forced shrink of precached extents");
+
 	node = rb_first(&tree->root);
 	while (node != NULL) {
 		es = rb_entry(node, struct extent_status, rb_node);
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index c0427e2f664..5498f75a164 100644
--- a/fs/ext4/ioctl.c
+++ b/fs/ext4/ioctl.c
@@ -624,6 +624,8 @@ resizefs_out:
 
 		return 0;
 	}
+	case EXT4_IOC_PRECACHE_EXTENTS:
+		return ext4_ext_precache(inode);
 
 	default:
 		return -ENOTTY;
@@ -688,6 +690,7 @@ long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case EXT4_IOC_MOVE_EXT:
 	case FITRIM:
 	case EXT4_IOC_RESIZE_FS:
+	case EXT4_IOC_PRECACHE_EXTENTS:
 		break;
 	default:
 		return -ENOIOCTLCMD;
diff --git a/include/uapi/linux/fiemap.h b/include/uapi/linux/fiemap.h
index d830747f5c0..0c51d617dae 100644
--- a/include/uapi/linux/fiemap.h
+++ b/include/uapi/linux/fiemap.h
@@ -40,6 +40,7 @@ struct fiemap {
 
 #define FIEMAP_FLAG_SYNC	0x00000001 /* sync file data before map */
 #define FIEMAP_FLAG_XATTR	0x00000002 /* map extended attribute tree */
+#define FIEMAP_FLAG_CACHE	0x00000004 /* request caching of the extents */
 
 #define FIEMAP_FLAGS_COMPAT	(FIEMAP_FLAG_SYNC | FIEMAP_FLAG_XATTR)
 
-- 
cgit v1.2.3-70-g09d2


From 7d7345322d60edb0fa49a64a89b31360f01d09cb Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Sat, 17 Aug 2013 09:36:54 -0400
Subject: ext4: fix warning in ext4_da_update_reserve_space()

reaim workfile.dbase test easily triggers warning in
ext4_da_update_reserve_space():

EXT4-fs warning (device ram0): ext4_da_update_reserve_space:365:
ino 12, allocated 1 with only 0 reserved metadata blocks (releasing 1
blocks with reserved 9 data blocks)

The problem is that (one of) tests creates file and then randomly writes
to it with O_SYNC. That results in writing back pages of the file in
random order so we create extents for written blocks say 0, 2, 4, 6, 8
- this last allocation also allocates new block for extents. Then we
writeout block 1 so we have extents 0-2, 4, 6, 8 and we release
indirect extent block because extents fit in the inode again. Then we
writeout block 10 and we need to allocate indirect extent block again
which triggers the warning because we don't have the reservation
anymore.

Fix the problem by giving back freed metadata blocks resulting from
extent merging into inode's reservation pool.

Signed-off-by: Jan Kara <jack@suse.cz>
---
 fs/ext4/ext4.h    |  1 +
 fs/ext4/extents.c |  3 ++-
 fs/ext4/mballoc.c | 21 +++++++++++++++++----
 3 files changed, 20 insertions(+), 5 deletions(-)

(limited to 'fs/ext4/extents.c')

diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 635135e6148..58dede76f75 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -581,6 +581,7 @@ enum {
 #define EXT4_FREE_BLOCKS_NO_QUOT_UPDATE	0x0008
 #define EXT4_FREE_BLOCKS_NOFREE_FIRST_CLUSTER	0x0010
 #define EXT4_FREE_BLOCKS_NOFREE_LAST_CLUSTER	0x0020
+#define EXT4_FREE_BLOCKS_RESERVE		0x0040
 
 /*
  * ioctl commands
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 01838875fca..62b21ccea88 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -1793,7 +1793,8 @@ static void ext4_ext_try_to_merge_up(handle_t *handle,
 
 	brelse(path[1].p_bh);
 	ext4_free_blocks(handle, inode, NULL, blk, 1,
-			 EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET);
+			 EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET |
+			 EXT4_FREE_BLOCKS_RESERVE);
 }
 
 /*
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 4bbbf13bd74..aa7d058e9e4 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -4585,6 +4585,7 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode,
 	struct buffer_head *gd_bh;
 	ext4_group_t block_group;
 	struct ext4_sb_info *sbi;
+	struct ext4_inode_info *ei = EXT4_I(inode);
 	struct ext4_buddy e4b;
 	unsigned int count_clusters;
 	int err = 0;
@@ -4784,7 +4785,6 @@ do_more:
 	ext4_block_bitmap_csum_set(sb, block_group, gdp, bitmap_bh);
 	ext4_group_desc_csum_set(sb, block_group, gdp);
 	ext4_unlock_group(sb, block_group);
-	percpu_counter_add(&sbi->s_freeclusters_counter, count_clusters);
 
 	if (sbi->s_log_groups_per_flex) {
 		ext4_group_t flex_group = ext4_flex_group(sbi, block_group);
@@ -4792,10 +4792,23 @@ do_more:
 			     &sbi->s_flex_groups[flex_group].free_clusters);
 	}
 
-	ext4_mb_unload_buddy(&e4b);
-
-	if (!(flags & EXT4_FREE_BLOCKS_NO_QUOT_UPDATE))
+	if (flags & EXT4_FREE_BLOCKS_RESERVE && ei->i_reserved_data_blocks) {
+		percpu_counter_add(&sbi->s_dirtyclusters_counter,
+				   count_clusters);
+		spin_lock(&ei->i_block_reservation_lock);
+		if (flags & EXT4_FREE_BLOCKS_METADATA)
+			ei->i_reserved_meta_blocks += count_clusters;
+		else
+			ei->i_reserved_data_blocks += count_clusters;
+		spin_unlock(&ei->i_block_reservation_lock);
+		if (!(flags & EXT4_FREE_BLOCKS_NO_QUOT_UPDATE))
+			dquot_reclaim_block(inode,
+					EXT4_C2B(sbi, count_clusters));
+	} else if (!(flags & EXT4_FREE_BLOCKS_NO_QUOT_UPDATE))
 		dquot_free_block(inode, EXT4_C2B(sbi, count_clusters));
+	percpu_counter_add(&sbi->s_freeclusters_counter, count_clusters);
+
+	ext4_mb_unload_buddy(&e4b);
 
 	/* We dirtied the bitmap block */
 	BUFFER_TRACE(bitmap_bh, "dirtied bitmap block");
-- 
cgit v1.2.3-70-g09d2


From 27b1b22882d32aa711ab4801700dad997440d940 Mon Sep 17 00:00:00 2001
From: Andi Shyti <andi@etezian.org>
Date: Wed, 28 Aug 2013 14:00:00 -0400
Subject: ext4: fix use of potentially uninitialized variables in debugging
 code

If ext_debugging is enabled and path[depth].p_ext is NULL, len
and lblock are printed non initialized

Signed-off-by: Andi Shyti <andi@etezian.org>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/ext4/extents.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

(limited to 'fs/ext4/extents.c')

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 62b21ccea88..e7580aeb6fb 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -2285,8 +2285,8 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path,
 				ext4_lblk_t block)
 {
 	int depth = ext_depth(inode);
-	unsigned long len;
-	ext4_lblk_t lblock;
+	unsigned long len = 0;
+	ext4_lblk_t lblock = 0;
 	struct ext4_extent *ex;
 
 	ex = path[depth].p_ext;
@@ -2323,7 +2323,6 @@ ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path,
 			ext4_es_insert_extent(inode, lblock, len, ~0,
 					      EXTENT_STATUS_HOLE);
 	} else {
-		lblock = len = 0;
 		BUG();
 	}
 
-- 
cgit v1.2.3-70-g09d2


From 70261f568f3c08552f034742e3d5cb78c3877766 Mon Sep 17 00:00:00 2001
From: Anatol Pomozov <anatol.pomozov@gmail.com>
Date: Wed, 28 Aug 2013 14:40:12 -0400
Subject: ext4: Fix misspellings using 'codespell' tool

Signed-off-by: Anatol Pomozov <anatol.pomozov@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/ext3/dir.c       | 2 +-
 fs/ext4/dir.c       | 2 +-
 fs/ext4/ext4_jbd2.h | 2 +-
 fs/ext4/extents.c   | 2 +-
 fs/ext4/inode.c     | 8 ++++----
 fs/ext4/migrate.c   | 2 +-
 6 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'fs/ext4/extents.c')

diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c
index f522425aaa2..bafdd48eefd 100644
--- a/fs/ext3/dir.c
+++ b/fs/ext3/dir.c
@@ -41,7 +41,7 @@ static unsigned char get_dtype(struct super_block *sb, int filetype)
 
 /**
  * Check if the given dir-inode refers to an htree-indexed directory
- * (or a directory which chould potentially get coverted to use htree
+ * (or a directory which could potentially get converted to use htree
  * indexing).
  *
  * Return 1 if it is a dx dir, 0 if not
diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
index 3c7d288ae94..680bb338891 100644
--- a/fs/ext4/dir.c
+++ b/fs/ext4/dir.c
@@ -33,7 +33,7 @@ static int ext4_dx_readdir(struct file *, struct dir_context *);
 
 /**
  * Check if the given dir-inode refers to an htree-indexed directory
- * (or a directory which chould potentially get coverted to use htree
+ * (or a directory which could potentially get converted to use htree
  * indexing).
  *
  * Return 1 if it is a dx dir, 0 if not
diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h
index 2877258d949..81cfefa9dc0 100644
--- a/fs/ext4/ext4_jbd2.h
+++ b/fs/ext4/ext4_jbd2.h
@@ -197,7 +197,7 @@ static inline void ext4_journal_callback_add(handle_t *handle,
  * ext4_journal_callback_del: delete a registered callback
  * @handle: active journal transaction handle on which callback was registered
  * @jce: registered journal callback entry to unregister
- * Return true if object was sucessfully removed
+ * Return true if object was successfully removed
  */
 static inline bool ext4_journal_callback_try_del(handle_t *handle,
 					     struct ext4_journal_cb_entry *jce)
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index e7580aeb6fb..916e88498c8 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -3220,7 +3220,7 @@ fix_extent_len:
  * ext4_split_extents() splits an extent and mark extent which is covered
  * by @map as split_flags indicates
  *
- * It may result in splitting the extent into multiple extents (upto three)
+ * It may result in splitting the extent into multiple extents (up to three)
  * There are three possibilities:
  *   a> There is no split required
  *   b> Splits in two extents: Split is happening at either end of the extent
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 47c8e465437..9115f280751 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -1916,7 +1916,7 @@ static int mpage_submit_page(struct mpage_da_data *mpd, struct page *page)
 /*
  * mballoc gives us at most this number of blocks...
  * XXX: That seems to be only a limitation of ext4_mb_normalize_request().
- * The rest of mballoc seems to handle chunks upto full group size.
+ * The rest of mballoc seems to handle chunks up to full group size.
  */
 #define MAX_WRITEPAGES_EXTENT_LEN 2048
 
@@ -2057,7 +2057,7 @@ static int mpage_map_and_submit_buffers(struct mpage_da_data *mpd)
 
 			if (page->index > end)
 				break;
-			/* Upto 'end' pages must be contiguous */
+			/* Up to 'end' pages must be contiguous */
 			BUG_ON(page->index != start);
 			bh = head = page_buffers(page);
 			do {
@@ -2256,7 +2256,7 @@ static int mpage_map_and_submit_extent(handle_t *handle,
 /*
  * Calculate the total number of credits to reserve for one writepages
  * iteration. This is called from ext4_writepages(). We map an extent of
- * upto MAX_WRITEPAGES_EXTENT_LEN blocks and then we go on and finish mapping
+ * up to MAX_WRITEPAGES_EXTENT_LEN blocks and then we go on and finish mapping
  * the last partial page. So in total we can map MAX_WRITEPAGES_EXTENT_LEN +
  * bpp - 1 blocks in bpp different extents.
  */
@@ -2443,7 +2443,7 @@ static int ext4_writepages(struct address_space *mapping,
 
 	if (ext4_should_dioread_nolock(inode)) {
 		/*
-		 * We may need to convert upto one extent per block in
+		 * We may need to convert up to one extent per block in
 		 * the page and we may dirty the inode.
 		 */
 		rsv_blocks = 1 + (PAGE_CACHE_SIZE >> inode->i_blkbits);
diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c
index f99bdb8548b..2ae73a80c19 100644
--- a/fs/ext4/migrate.c
+++ b/fs/ext4/migrate.c
@@ -494,7 +494,7 @@ int ext4_ext_migrate(struct inode *inode)
 	 * superblock modification.
 	 *
 	 * For the tmp_inode we already have committed the
-	 * trascation that created the inode. Later as and
+	 * transaction that created the inode. Later as and
 	 * when we add extents we extent the journal
 	 */
 	/*
-- 
cgit v1.2.3-70-g09d2


From d7b2a00c2e2eedf460ce2a15237f28de40412d86 Mon Sep 17 00:00:00 2001
From: Zheng Liu <wenqing.lz@taobao.com>
Date: Wed, 28 Aug 2013 14:47:06 -0400
Subject: ext4: isolate ext4_extents.h file

After applied the commit (4a092d73), we have reduced the number of
source files that need to #include ext4_extents.h.  But we can do
better.

This commit defines ext4_zeroout_es() in extents.c and move
EXT_MAX_BLOCKS into ext4.h in order not to include ext4_extents.h in
indirect.c and ioctl.c.  Meanwhile we just need to include this file in
extent_status.c when ES_AGGRESSIVE_TEST is defined.  Otherwise, this
commit removes a duplicated declaration in trace/events/ext4.h.

After applied this patch, we just need to include ext4_extents.h file
in {super,migrate,move_extents,extents}.c, and it is easy for us to
define a new extent disk layout.

Signed-off-by: Zheng Liu <wenqing.lz@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
 fs/ext4/ext4.h              |  6 ++++++
 fs/ext4/ext4_extents.h      |  6 ------
 fs/ext4/extents.c           | 21 +++++++++++++++++++--
 fs/ext4/extents_status.c    | 20 ++------------------
 fs/ext4/extents_status.h    |  1 -
 fs/ext4/indirect.c          |  1 -
 fs/ext4/ioctl.c             |  1 -
 include/trace/events/ext4.h |  1 -
 8 files changed, 27 insertions(+), 30 deletions(-)

(limited to 'fs/ext4/extents.c')

diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 3dbc56eb484..28896655ad4 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -2686,6 +2686,12 @@ extern int ext4_check_blockref(const char *, unsigned int,
 struct ext4_ext_path;
 struct ext4_extent;
 
+/*
+ * Maximum number of logical blocks in a file; ext4_extent's ee_block is
+ * __le32.
+ */
+#define EXT_MAX_BLOCKS	0xffffffff
+
 extern int ext4_ext_tree_init(handle_t *handle, struct inode *);
 extern int ext4_ext_writepage_trans_blocks(struct inode *, int);
 extern int ext4_ext_index_trans_blocks(struct inode *inode, int extents);
diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h
index 51bc821ade9..5074fe23f19 100644
--- a/fs/ext4/ext4_extents.h
+++ b/fs/ext4/ext4_extents.h
@@ -133,12 +133,6 @@ struct ext4_ext_path {
  * structure for external API
  */
 
-/*
- * Maximum number of logical blocks in a file; ext4_extent's ee_block is
- * __le32.
- */
-#define EXT_MAX_BLOCKS	0xffffffff
-
 /*
  * EXT_INIT_MAX_LEN is the maximum number of blocks we can have in an
  * initialized extent. This is 2^15 and not (2^16 - 1), since we use the
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 916e88498c8..54d52afcdb1 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -3048,6 +3048,23 @@ void ext4_ext_release(struct super_block *sb)
 #endif
 }
 
+static int ext4_zeroout_es(struct inode *inode, struct ext4_extent *ex)
+{
+	ext4_lblk_t  ee_block;
+	ext4_fsblk_t ee_pblock;
+	unsigned int ee_len;
+
+	ee_block  = le32_to_cpu(ex->ee_block);
+	ee_len    = ext4_ext_get_actual_len(ex);
+	ee_pblock = ext4_ext_pblock(ex);
+
+	if (ee_len == 0)
+		return 0;
+
+	return ext4_es_insert_extent(inode, ee_block, ee_len, ee_pblock,
+				     EXTENT_STATUS_WRITTEN);
+}
+
 /* FIXME!! we need to try to merge to left or right after zero-out  */
 static int ext4_ext_zeroout(struct inode *inode, struct ext4_extent *ex)
 {
@@ -3200,7 +3217,7 @@ static int ext4_split_extent_at(handle_t *handle,
 			goto fix_extent_len;
 
 		/* update extent status tree */
-		err = ext4_es_zeroout(inode, &zero_ex);
+		err = ext4_zeroout_es(inode, &zero_ex);
 
 		goto out;
 	} else if (err)
@@ -3551,7 +3568,7 @@ static int ext4_ext_convert_to_initialized(handle_t *handle,
 out:
 	/* If we have gotten a failure, don't zero out status tree */
 	if (!err)
-		err = ext4_es_zeroout(inode, &zero_ex);
+		err = ext4_zeroout_es(inode, &zero_ex);
 	return err ? err : allocated;
 }
 
diff --git a/fs/ext4/extents_status.c b/fs/ext4/extents_status.c
index 0e88a367b53..2d1bdbe78c0 100644
--- a/fs/ext4/extents_status.c
+++ b/fs/ext4/extents_status.c
@@ -13,7 +13,6 @@
 #include <linux/list_sort.h>
 #include "ext4.h"
 #include "extents_status.h"
-#include "ext4_extents.h"
 
 #include <trace/events/ext4.h>
 
@@ -409,6 +408,8 @@ ext4_es_try_to_merge_right(struct inode *inode, struct extent_status *es)
 }
 
 #ifdef ES_AGGRESSIVE_TEST
+#include "ext4_extents.h"	/* Needed when ES_AGGRESSIVE_TEST is defined */
+
 static void ext4_es_insert_extent_ext_check(struct inode *inode,
 					    struct extent_status *es)
 {
@@ -903,23 +904,6 @@ int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
 	return err;
 }
 
-int ext4_es_zeroout(struct inode *inode, struct ext4_extent *ex)
-{
-	ext4_lblk_t  ee_block;
-	ext4_fsblk_t ee_pblock;
-	unsigned int ee_len;
-
-	ee_block  = le32_to_cpu(ex->ee_block);
-	ee_len    = ext4_ext_get_actual_len(ex);
-	ee_pblock = ext4_ext_pblock(ex);
-
-	if (ee_len == 0)
-		return 0;
-
-	return ext4_es_insert_extent(inode, ee_block, ee_len, ee_pblock,
-				     EXTENT_STATUS_WRITTEN);
-}
-
 static int ext4_inode_touch_time_cmp(void *priv, struct list_head *a,
 				     struct list_head *b)
 {
diff --git a/fs/ext4/extents_status.h b/fs/ext4/extents_status.h
index 3e83aef3653..167f4ab8ecc 100644
--- a/fs/ext4/extents_status.h
+++ b/fs/ext4/extents_status.h
@@ -81,7 +81,6 @@ extern void ext4_es_find_delayed_extent_range(struct inode *inode,
 					struct extent_status *es);
 extern int ext4_es_lookup_extent(struct inode *inode, ext4_lblk_t lblk,
 				 struct extent_status *es);
-extern int ext4_es_zeroout(struct inode *inode, struct ext4_extent *ex);
 
 static inline int ext4_es_is_written(struct extent_status *es)
 {
diff --git a/fs/ext4/indirect.c b/fs/ext4/indirect.c
index 87b30cd357e..594009f5f52 100644
--- a/fs/ext4/indirect.c
+++ b/fs/ext4/indirect.c
@@ -23,7 +23,6 @@
 #include <linux/aio.h>
 #include "ext4_jbd2.h"
 #include "truncate.h"
-#include "ext4_extents.h"	/* Needed for EXT_MAX_BLOCKS */
 
 #include <trace/events/ext4.h>
 
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index 5498f75a164..a569d335f80 100644
--- a/fs/ext4/ioctl.c
+++ b/fs/ext4/ioctl.c
@@ -17,7 +17,6 @@
 #include <asm/uaccess.h>
 #include "ext4_jbd2.h"
 #include "ext4.h"
-#include "ext4_extents.h"
 
 #define MAX_32_NUM ((((unsigned long long) 1) << 32) - 1)
 
diff --git a/include/trace/events/ext4.h b/include/trace/events/ext4.h
index d892b55d91a..197d3125df2 100644
--- a/include/trace/events/ext4.h
+++ b/include/trace/events/ext4.h
@@ -14,7 +14,6 @@ struct ext4_prealloc_space;
 struct ext4_inode_info;
 struct mpage_da_data;
 struct ext4_map_blocks;
-struct ext4_extent;
 struct extent_status;
 
 #define EXT4_I(inode) (container_of(inode, struct ext4_inode_info, vfs_inode))
-- 
cgit v1.2.3-70-g09d2