1 files changed, 203 insertions, 93 deletions

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 72ba4705d4fa..54d52afcdb19 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;
@@ -447,42 +447,149 @@ 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 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 flags)
 {
-	int ret;
+	struct buffer_head		*bh;
+	int				err;
 
-	if (buffer_verified(bh))
-		return 0;
-	ret = ext4_ext_check(inode, eh, depth);
-	if (ret)
-		return ret;
+	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) && !(flags & EXT4_EX_FORCE_CACHE))
+		return bh;
+	err = __ext4_ext_check(function, line, inode,
+			       ext_block_hdr(bh), depth, pblk);
+	if (err)
+		goto errout;
 	set_buffer_verified(bh);
-	return ret;
+	/*
+	 * 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);
+	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, flags)		\
+	__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)
@@ -716,7 +823,7 @@ 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;
@@ -748,20 +855,13 @@ 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,
+					    flags);
+		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 +873,6 @@ 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;
@@ -1198,7 +1293,8 @@ 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)
 {
@@ -1220,7 +1316,7 @@ 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;
 
@@ -1228,12 +1324,12 @@ 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;
 
@@ -1241,7 +1337,7 @@ 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;
@@ -1412,29 +1508,21 @@ 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, 0);
+		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, 0);
+	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);
@@ -1705,7 +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);
 }
 
 /*
@@ -1793,7 +1882,7 @@ 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;
@@ -1802,7 +1891,7 @@ 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");
@@ -1817,7 +1906,7 @@ 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
@@ -1920,7 +2009,7 @@ 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);
@@ -1939,9 +2028,10 @@ 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);
@@ -2007,7 +2097,7 @@ 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);
 
 
@@ -2050,7 +2140,7 @@ 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);
@@ -2195,8 +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;
@@ -2233,7 +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();
 	}
 
@@ -2712,7 +2801,7 @@ 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);
@@ -2754,6 +2843,7 @@ 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);
 
@@ -2782,7 +2872,7 @@ 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;
 		}
@@ -2829,10 +2919,12 @@ 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,
+				EXT4_EX_NOCACHE);
+			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 +2934,6 @@ 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
@@ -2961,6 +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)
 {
@@ -3113,7 +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)
@@ -3133,7 +3237,7 @@ 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
@@ -3181,7 +3285,7 @@ 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);
@@ -3464,7 +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;
 }
 
@@ -3565,7 +3669,7 @@ 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;
@@ -4052,7 +4156,7 @@ 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;
@@ -4744,6 +4848,12 @@ 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,
@@ -4771,6 +4881,6 @@ 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;
 }