writeback: separate starting of sync vs opportunistic writeback
bdi_start_writeback() is currently split into two paths, one for
WB_SYNC_NONE and one for WB_SYNC_ALL. Add bdi_sync_writeback()
for WB_SYNC_ALL writeback and let bdi_start_writeback() handle
only WB_SYNC_NONE.
Push down the writeback_control allocation and only accept the
parameters that make sense for each function. This cleans up
the API considerably.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c
index 1c8991b..ee1ce68 100644
--- a/fs/ubifs/budget.c
+++ b/fs/ubifs/budget.c
@@ -54,29 +54,15 @@
* @nr_to_write: how many dirty pages to write-back
*
* This function shrinks UBIFS liability by means of writing back some amount
- * of dirty inodes and their pages. Returns the amount of pages which were
- * written back. The returned value does not include dirty inodes which were
- * synchronized.
+ * of dirty inodes and their pages.
*
* Note, this function synchronizes even VFS inodes which are locked
* (@i_mutex) by the caller of the budgeting function, because write-back does
* not touch @i_mutex.
*/
-static int shrink_liability(struct ubifs_info *c, int nr_to_write)
+static void shrink_liability(struct ubifs_info *c, int nr_to_write)
{
- int nr_written;
-
- nr_written = writeback_inodes_sb(c->vfs_sb);
- if (!nr_written) {
- /*
- * Re-try again but wait on pages/inodes which are being
- * written-back concurrently (e.g., by pdflush).
- */
- nr_written = sync_inodes_sb(c->vfs_sb);
- }
-
- dbg_budg("%d pages were written back", nr_written);
- return nr_written;
+ writeback_inodes_sb(c->vfs_sb);
}
/**