ext4: fold ext4_generic_write_end() into ext4_write_end()
authorZheng Liu <wenqing.lz@taobao.com>
Wed, 3 Apr 2013 16:41:17 +0000 (12:41 -0400)
committerTheodore Ts'o <tytso@mit.edu>
Wed, 3 Apr 2013 16:41:17 +0000 (12:41 -0400)
After collapsing the handling of data ordered and data writeback
codepath, ext4_generic_write_end() has only one caller,
ext4_write_end().  So we fold it into ext4_write_end().

Signed-off-by: Zheng Liu <wenqing.lz@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: Lukas Czerner <lczerner@redhat.com>
fs/ext4/inode.c

index 4ee69270a48ad1723c8afe07dd274d92f29f2927..a4ffb470fbf33547b3c52b8440f0e3f5c4c19962 100644 (file)
@@ -1087,14 +1087,32 @@ static int write_end_fn(handle_t *handle, struct buffer_head *bh)
        return ext4_handle_dirty_metadata(handle, NULL, bh);
 }
 
-static int ext4_generic_write_end(struct file *file,
-                                 struct address_space *mapping,
-                                 loff_t pos, unsigned len, unsigned copied,
-                                 struct page *page, void *fsdata)
+/*
+ * We need to pick up the new inode size which generic_commit_write gave us
+ * `file' can be NULL - eg, when called from page_symlink().
+ *
+ * ext4 never places buffers on inode->i_mapping->private_list.  metadata
+ * buffers are managed internally.
+ */
+static int ext4_write_end(struct file *file,
+                         struct address_space *mapping,
+                         loff_t pos, unsigned len, unsigned copied,
+                         struct page *page, void *fsdata)
 {
-       int i_size_changed = 0;
-       struct inode *inode = mapping->host;
        handle_t *handle = ext4_journal_current_handle();
+       struct inode *inode = mapping->host;
+       int ret = 0, ret2;
+       int i_size_changed = 0;
+
+       trace_ext4_write_end(inode, pos, len, copied);
+       if (ext4_test_inode_state(inode, EXT4_STATE_ORDERED_MODE)) {
+               ret = ext4_jbd2_file_inode(handle, inode);
+               if (ret) {
+                       unlock_page(page);
+                       page_cache_release(page);
+                       goto errout;
+               }
+       }
 
        if (ext4_has_inline_data(inode))
                copied = ext4_write_inline_data_end(inode, pos, len,
@@ -1105,7 +1123,7 @@ static int ext4_generic_write_end(struct file *file,
 
        /*
         * No need to use i_size_read() here, the i_size
-        * cannot change under us because we hold i_mutex.
+        * cannot change under us because we hole i_mutex.
         *
         * But it's important to update i_size while still holding page lock:
         * page writeout could otherwise come in and zero beyond i_size.
@@ -1115,10 +1133,10 @@ static int ext4_generic_write_end(struct file *file,
                i_size_changed = 1;
        }
 
-       if (pos + copied >  EXT4_I(inode)->i_disksize) {
+       if (pos + copied > EXT4_I(inode)->i_disksize) {
                /* We need to mark inode dirty even if
                 * new_i_size is less that inode->i_size
-                * bu greater than i_disksize.(hint delalloc)
+                * but greater than i_disksize. (hint delalloc)
                 */
                ext4_update_i_disksize(inode, (pos + copied));
                i_size_changed = 1;
@@ -1135,37 +1153,6 @@ static int ext4_generic_write_end(struct file *file,
        if (i_size_changed)
                ext4_mark_inode_dirty(handle, inode);
 
-       return copied;
-}
-
-/*
- * We need to pick up the new inode size which generic_commit_write gave us
- * `file' can be NULL - eg, when called from page_symlink().
- *
- * ext4 never places buffers on inode->i_mapping->private_list.  metadata
- * buffers are managed internally.
- */
-static int ext4_write_end(struct file *file,
-                         struct address_space *mapping,
-                         loff_t pos, unsigned len, unsigned copied,
-                         struct page *page, void *fsdata)
-{
-       handle_t *handle = ext4_journal_current_handle();
-       struct inode *inode = mapping->host;
-       int ret = 0, ret2;
-
-       trace_ext4_write_end(inode, pos, len, copied);
-       if (ext4_test_inode_state(inode, EXT4_STATE_ORDERED_MODE)) {
-               ret = ext4_jbd2_file_inode(handle, inode);
-               if (ret) {
-                       unlock_page(page);
-                       page_cache_release(page);
-                       goto errout;
-               }
-       }
-
-       copied = ext4_generic_write_end(file, mapping, pos, len, copied,
-                                       page, fsdata);
        if (copied < 0)
                ret = copied;
        if (pos + len > inode->i_size && ext4_can_truncate(inode))