1 files changed, 147 insertions, 30 deletions

diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index c616a16a9f36..b543a46fc809 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -50,10 +50,9 @@ static __le32 ext4_extent_block_csum(struct inode *inode,
 				     struct ext4_extent_header *eh)
 {
 	struct ext4_inode_info *ei = EXT4_I(inode);
-	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 	__u32 csum;
 
-	csum = ext4_chksum(sbi, ei->i_csum_seed, (__u8 *)eh,
+	csum = ext4_chksum(ei->i_csum_seed, (__u8 *)eh,
 			   EXT4_EXTENT_TAIL_OFFSET(eh));
 	return cpu_to_le32(csum);
 }
@@ -611,6 +610,8 @@ int ext4_ext_precache(struct inode *inode)
 	if (!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
 		return 0;	/* not an extent-mapped inode */
 
+	ext4_check_map_extents_env(inode);
+
 	down_read(&ei->i_data_sem);
 	depth = ext_depth(inode);
 
@@ -1530,7 +1531,7 @@ static int ext4_ext_search_left(struct inode *inode,
 static int ext4_ext_search_right(struct inode *inode,
 				 struct ext4_ext_path *path,
 				 ext4_lblk_t *logical, ext4_fsblk_t *phys,
-				 struct ext4_extent *ret_ex)
+				 struct ext4_extent *ret_ex, int flags)
 {
 	struct buffer_head *bh = NULL;
 	struct ext4_extent_header *eh;
@@ -1604,7 +1605,8 @@ got_index:
 	ix++;
 	while (++depth < path->p_depth) {
 		/* subtract from p_depth to get proper eh_depth */
-		bh = read_extent_tree_block(inode, ix, path->p_depth - depth, 0);
+		bh = read_extent_tree_block(inode, ix, path->p_depth - depth,
+					    flags);
 		if (IS_ERR(bh))
 			return PTR_ERR(bh);
 		eh = ext_block_hdr(bh);
@@ -1612,7 +1614,7 @@ got_index:
 		put_bh(bh);
 	}
 
-	bh = read_extent_tree_block(inode, ix, path->p_depth - depth, 0);
+	bh = read_extent_tree_block(inode, ix, path->p_depth - depth, flags);
 	if (IS_ERR(bh))
 		return PTR_ERR(bh);
 	eh = ext_block_hdr(bh);
@@ -2396,18 +2398,20 @@ int ext4_ext_calc_credits_for_single_extent(struct inode *inode, int nrblocks,
 int ext4_ext_index_trans_blocks(struct inode *inode, int extents)
 {
 	int index;
-	int depth;
 
 	/* If we are converting the inline data, only one is needed here. */
 	if (ext4_has_inline_data(inode))
 		return 1;
 
-	depth = ext_depth(inode);
-
+	/*
+	 * Extent tree can change between the time we estimate credits and
+	 * the time we actually modify the tree. Assume the worst case.
+	 */
 	if (extents <= 1)
-		index = depth * 2;
+		index = (EXT4_MAX_EXTENT_DEPTH * 2) + extents;
 	else
-		index = depth * 3;
+		index = (EXT4_MAX_EXTENT_DEPTH * 3) +
+			DIV_ROUND_UP(extents, ext4_ext_space_block(inode, 0));
 
 	return index;
 }
@@ -2821,6 +2825,7 @@ int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start,
 	struct partial_cluster partial;
 	handle_t *handle;
 	int i = 0, err = 0;
+	int flags = EXT4_EX_NOCACHE | EXT4_EX_NOFAIL;
 
 	partial.pclu = 0;
 	partial.lblk = 0;
@@ -2851,8 +2856,7 @@ again:
 		ext4_fsblk_t pblk;
 
 		/* find extent for or closest extent to this block */
-		path = ext4_find_extent(inode, end, NULL,
-					EXT4_EX_NOCACHE | EXT4_EX_NOFAIL);
+		path = ext4_find_extent(inode, end, NULL, flags);
 		if (IS_ERR(path)) {
 			ext4_journal_stop(handle);
 			return PTR_ERR(path);
@@ -2918,7 +2922,7 @@ again:
 			 */
 			lblk = ex_end + 1;
 			err = ext4_ext_search_right(inode, path, &lblk, &pblk,
-						    NULL);
+						    NULL, flags);
 			if (err < 0)
 				goto out;
 			if (pblk) {
@@ -2994,8 +2998,7 @@ again:
 				  i + 1, ext4_idx_pblock(path[i].p_idx));
 			memset(path + i + 1, 0, sizeof(*path));
 			bh = read_extent_tree_block(inode, path[i].p_idx,
-						    depth - i - 1,
-						    EXT4_EX_NOCACHE);
+						    depth - i - 1, flags);
 			if (IS_ERR(bh)) {
 				/* should we reset i_size? */
 				err = PTR_ERR(bh);
@@ -4202,7 +4205,7 @@ 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_find_extent(inode, map->m_lblk, NULL, 0);
+	path = ext4_find_extent(inode, map->m_lblk, NULL, flags);
 	if (IS_ERR(path)) {
 		err = PTR_ERR(path);
 		goto out;
@@ -4314,7 +4317,8 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 	if (err)
 		goto out;
 	ar.lright = map->m_lblk;
-	err = ext4_ext_search_right(inode, path, &ar.lright, &ar.pright, &ex2);
+	err = ext4_ext_search_right(inode, path, &ar.lright, &ar.pright,
+				    &ex2, flags);
 	if (err < 0)
 		goto out;
 
@@ -4433,6 +4437,20 @@ got_allocated_blocks:
 	allocated = map->m_len;
 	ext4_ext_show_leaf(inode, path);
 out:
+	/*
+	 * We never use EXT4_GET_BLOCKS_QUERY_LAST_IN_LEAF with CREATE flag.
+	 * So we know that the depth used here is correct, since there was no
+	 * block allocation done if EXT4_GET_BLOCKS_QUERY_LAST_IN_LEAF is set.
+	 * If tomorrow we start using this QUERY flag with CREATE, then we will
+	 * need to re-calculate the depth as it might have changed due to block
+	 * allocation.
+	 */
+	if (flags & EXT4_GET_BLOCKS_QUERY_LAST_IN_LEAF) {
+		WARN_ON_ONCE(flags & EXT4_GET_BLOCKS_CREATE);
+		if (!err && ex && (ex == EXT_LAST_EXTENT(path[depth].p_hdr)))
+			map->m_flags |= EXT4_MAP_QUERY_LAST_IN_LEAF;
+	}
+
 	ext4_free_ext_path(path);
 
 	trace_ext4_ext_map_blocks_exit(inode, flags, map,
@@ -4781,6 +4799,93 @@ out_inode_lock:
 }
 
 /*
+ * This function converts a range of blocks to written extents. The caller of
+ * this function will pass the start offset and the size. all unwritten extents
+ * within this range will be converted to written extents.
+ *
+ * This function is called from the direct IO end io call back function for
+ * atomic writes, to convert the unwritten extents after IO is completed.
+ *
+ * Note that the requirement for atomic writes is that all conversion should
+ * happen atomically in a single fs journal transaction. We mainly only allocate
+ * unwritten extents either on a hole on a pre-exiting unwritten extent range in
+ * ext4_map_blocks_atomic_write(). The only case where we can have multiple
+ * unwritten extents in a range [offset, offset+len) is when there is a split
+ * unwritten extent between two leaf nodes which was cached in extent status
+ * cache during ext4_iomap_alloc() time. That will allow
+ * ext4_map_blocks_atomic_write() to return the unwritten extent range w/o going
+ * into the slow path. That means we might need a loop for conversion of this
+ * unwritten extent split across leaf block within a single journal transaction.
+ * Split extents across leaf nodes is a rare case, but let's still handle that
+ * to meet the requirements of multi-fsblock atomic writes.
+ *
+ * Returns 0 on success.
+ */
+int ext4_convert_unwritten_extents_atomic(handle_t *handle, struct inode *inode,
+					  loff_t offset, ssize_t len)
+{
+	unsigned int max_blocks;
+	int ret = 0, ret2 = 0, ret3 = 0;
+	struct ext4_map_blocks map;
+	unsigned int blkbits = inode->i_blkbits;
+	unsigned int credits = 0;
+	int flags = EXT4_GET_BLOCKS_IO_CONVERT_EXT | EXT4_EX_NOCACHE;
+
+	map.m_lblk = offset >> blkbits;
+	max_blocks = EXT4_MAX_BLOCKS(len, offset, blkbits);
+
+	if (!handle) {
+		/*
+		 * TODO: An optimization can be added later by having an extent
+		 * status flag e.g. EXTENT_STATUS_SPLIT_LEAF. If we query that
+		 * it can tell if the extent in the cache is a split extent.
+		 * But for now let's assume pextents as 2 always.
+		 */
+		credits = ext4_meta_trans_blocks(inode, max_blocks, 2);
+	}
+
+	if (credits) {
+		handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, credits);
+		if (IS_ERR(handle)) {
+			ret = PTR_ERR(handle);
+			return ret;
+		}
+	}
+
+	while (ret >= 0 && ret < max_blocks) {
+		map.m_lblk += ret;
+		map.m_len = (max_blocks -= ret);
+		ret = ext4_map_blocks(handle, inode, &map, flags);
+		if (ret != max_blocks)
+			ext4_msg(inode->i_sb, KERN_INFO,
+				     "inode #%lu: block %u: len %u: "
+				     "split block mapping found for atomic write, "
+				     "ret = %d",
+				     inode->i_ino, map.m_lblk,
+				     map.m_len, ret);
+		if (ret <= 0)
+			break;
+	}
+
+	ret2 = ext4_mark_inode_dirty(handle, inode);
+
+	if (credits) {
+		ret3 = ext4_journal_stop(handle);
+		if (unlikely(ret3))
+			ret2 = ret3;
+	}
+
+	if (ret <= 0 || ret2)
+		ext4_warning(inode->i_sb,
+			     "inode #%lu: block %u: len %u: "
+			     "returned %d or %d",
+			     inode->i_ino, map.m_lblk,
+			     map.m_len, ret, ret2);
+
+	return ret > 0 ? ret2 : ret;
+}
+
+/*
  * This function convert a range of blocks to written extents
  * The caller of this function will pass the start offset and the size.
  * all unwritten extents within this range will be converted to
@@ -4819,8 +4924,14 @@ int ext4_convert_unwritten_extents(handle_t *handle, struct inode *inode,
 				break;
 			}
 		}
+		/*
+		 * Do not cache any unrelated extents, as it does not hold the
+		 * i_rwsem or invalidate_lock, which could corrupt the extent
+		 * status tree.
+		 */
 		ret = ext4_map_blocks(handle, inode, &map,
-				      EXT4_GET_BLOCKS_IO_CONVERT_EXT);
+				      EXT4_GET_BLOCKS_IO_CONVERT_EXT |
+				      EXT4_EX_NOCACHE);
 		if (ret <= 0)
 			ext4_warning(inode->i_sb,
 				     "inode #%lu: block %u: len %u: "
@@ -4931,12 +5042,7 @@ static const struct iomap_ops ext4_iomap_xattr_ops = {
 
 static int ext4_fiemap_check_ranges(struct inode *inode, u64 start, u64 *len)
 {
-	u64 maxbytes;
-
-	if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
-		maxbytes = inode->i_sb->s_maxbytes;
-	else
-		maxbytes = EXT4_SB(inode->i_sb)->s_bitmap_maxbytes;
+	u64 maxbytes = ext4_get_maxbytes(inode);
 
 	if (*len == 0)
 		return -EINVAL;
@@ -4956,10 +5062,11 @@ int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 {
 	int error = 0;
 
+	inode_lock_shared(inode);
 	if (fieinfo->fi_flags & FIEMAP_FLAG_CACHE) {
 		error = ext4_ext_precache(inode);
 		if (error)
-			return error;
+			goto unlock;
 		fieinfo->fi_flags &= ~FIEMAP_FLAG_CACHE;
 	}
 
@@ -4970,15 +5077,19 @@ int ext4_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	 */
 	error = ext4_fiemap_check_ranges(inode, start, &len);
 	if (error)
-		return error;
+		goto unlock;
 
 	if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) {
 		fieinfo->fi_flags &= ~FIEMAP_FLAG_XATTR;
-		return iomap_fiemap(inode, fieinfo, start, len,
-				    &ext4_iomap_xattr_ops);
+		error = iomap_fiemap(inode, fieinfo, start, len,
+				     &ext4_iomap_xattr_ops);
+	} else {
+		error = iomap_fiemap(inode, fieinfo, start, len,
+				     &ext4_iomap_report_ops);
 	}
-
-	return iomap_fiemap(inode, fieinfo, start, len, &ext4_iomap_report_ops);
+unlock:
+	inode_unlock_shared(inode);
+	return error;
 }
 
 int ext4_get_es_cache(struct inode *inode, struct fiemap_extent_info *fieinfo,
@@ -4999,7 +5110,9 @@ int ext4_get_es_cache(struct inode *inode, struct fiemap_extent_info *fieinfo,
 	}
 
 	if (fieinfo->fi_flags & FIEMAP_FLAG_CACHE) {
+		inode_lock_shared(inode);
 		error = ext4_ext_precache(inode);
+		inode_unlock_shared(inode);
 		if (error)
 			return error;
 		fieinfo->fi_flags &= ~FIEMAP_FLAG_CACHE;
@@ -5328,6 +5441,8 @@ static int ext4_collapse_range(struct file *file, loff_t offset, loff_t len)
 	start_lblk = offset >> inode->i_blkbits;
 	end_lblk = (offset + len) >> inode->i_blkbits;
 
+	ext4_check_map_extents_env(inode);
+
 	down_write(&EXT4_I(inode)->i_data_sem);
 	ext4_discard_preallocations(inode);
 	ext4_es_remove_extent(inode, start_lblk, EXT_MAX_BLOCKS - start_lblk);
@@ -5429,6 +5544,8 @@ static int ext4_insert_range(struct file *file, loff_t offset, loff_t len)
 	start_lblk = offset >> inode->i_blkbits;
 	len_lblk = len >> inode->i_blkbits;
 
+	ext4_check_map_extents_env(inode);
+
 	down_write(&EXT4_I(inode)->i_data_sem);
 	ext4_discard_preallocations(inode);