ceph: properly release page upon error

When ceph_update_writeable_page fails (including -EAGAIN), it
unlocks (w/ unlock_page) the page but does not 'release'
(w/ page_cache_release) properly.

Upon error, properly set *pagep to NULL, indicating an error.
Signed-off-by: NTaesoo Kim <tsgatesv@gmail.com>
Signed-off-by: NYan, Zheng <zyan@redhat.com>

fs/ceph/addr.c

@@ -1146,6 +1146,10 @@ static int ceph_write_begin(struct file *file, struct address_space *mapping,
 		     inode, page, (int)pos, (int)len);
 
 		r = ceph_update_writeable_page(file, pos, len, page);
+		if (r < 0)
+			page_cache_release(page);
+		else
+			*pagep = page;
 	} while (r == -EAGAIN);
 
 	return r;