On Tue, 16 Jan 2007 21:05:20 +0900
sho@tnes.nec.co.jp wrote:

Move the blocks on the temporary inode to the original inode
by a page.
1. Read the file data from the old blocks to the page
2. Move the block on the temporary inode to the original inode
3. Write the file data on the page into the new blocks

...

+
+/**
+ * ext4_ext_replace_branches - replace original extents with new extents.
+ * @org_inode    Original inode
+ * @dest_inode   temporary inode
+ * @from_page    Page offset
+ * @count_page   Page count to be replaced
+ * @delete_start block offset for deletion
+ *
+ * This function returns 0 if succeed, otherwise returns error value.
+ * Replace extents for blocks from "from" to "from+count-1".
+ */
+static int
+ext4_ext_replace_branches(struct inode *org_inode, struct inode *dest_inode,
+			pgoff_t from_page,  pgoff_t dest_from_page,
+			pgoff_t count_page, unsigned long *delete_start) 
+{
+	struct ext4_ext_path *org_path = NULL;
+	struct ext4_ext_path *dest_path = NULL;
+	struct ext4_extent   *oext, *dext;
+	struct ext4_extent   tmp_ext;
+	int	err = 0;
+	int	depth;
+	unsigned long from, count, dest_off, diff, replaced_count = 0;

These should be sector_t, shouldn't they?

+	handle_t *handle = NULL;
+	unsigned jnum;
+
+	from = from_page << (PAGE_CACHE_SHIFT - dest_inode->i_blkbits);

In which case one needs to be very careful to avoid overflows in
expressions such as this one.

+	wait_on_page_locked(page);
+	lock_page(page);

The wait_on_page_locked() is unneeded here.