Re: Bug in kernel 2.6.31, Slow wb_kupdate writeout

[Martin Bligh <mbligh@google.com>]

> --- mm.orig/fs/fs-writeback.c
> +++ mm/fs/fs-writeback.c
> @@ -325,7 +325,8 @@ __sync_single_inode(struct inode *inode,
>                                 * soon as the queue becomes uncongested.
>                                 */
>                                inode->i_state |= I_DIRTY_PAGES;
> -                               if (wbc->nr_to_write <= 0) {
> +                               if (wbc->nr_to_write <= 0 ||
> +                                   wbc->encountered_congestion) {
>                                        /*
>                                         * slice used up: queue for next turn
>                                         */
>

That's not sufficient - it only the problem in the wb_kupdate path. If you want
to be more conservative, how about we do this?

--- linux-2.6.30/fs/fs-writeback.c.old  2009-07-29 00:08:29.000000000 -0700
+++ linux-2.6.30/fs/fs-writeback.c      2009-07-29 07:08:48.000000000 -0700
@@ -323,43 +323,14 @@ __sync_single_inode(struct inode *inode,
                         * We didn't write back all the pages.  nfs_writepages(
)
                         * sometimes bales out without doing anything. Redirty
                         * the inode; Move it from s_io onto s_more_io/s_dirty.
+                        * It may well have just encountered congestion
                         */
-                       /*
-                        * akpm: if the caller was the kupdate function we put
-                        * this inode at the head of s_dirty so it gets first
-                        * consideration.  Otherwise, move it to the tail, for
-                        * the reasons described there.  I'm not really sure
-                        * how much sense this makes.  Presumably I had a good
-                        * reasons for doing it this way, and I'd rather not
-                        * muck with it at present.
-                        */
-                       if (wbc->for_kupdate) {
-                               /*
-                                * For the kupdate function we move the inode
-                                * to s_more_io so it will get more writeout as
-                                * soon as the queue becomes uncongested.
-                                */
-                               inode->i_state |= I_DIRTY_PAGES;
-                               if (wbc->nr_to_write <= 0) {
-                                       /*
-                                        * slice used up: queue for next turn
-                                        */
-                                       requeue_io(inode);
-                               } else {
-                                       /*
-                                        * somehow blocked: retry later
-                                        */
-                                       redirty_tail(inode);
-                               }
-                       } else {
-                               /*
-                                * Otherwise fully redirty the inode so that
-                                * other inodes on this superblock will get som
e
-                                * writeout.  Otherwise heavy writing to one
-                                * file would indefinitely suspend writeout of
-                                * all the other files.
-                                */
-                               inode->i_state |= I_DIRTY_PAGES;
+                       inode->i_state |= I_DIRTY_PAGES;
+                       if (wbc->nr_to_write <= 0 ||     /* sliced used up */
+                            wbc->encountered_congestion)
+                               requeue_io(inode);
+                       else {
+                               /* somehow blocked: retry later */
                                redirty_tail(inode);
                        }
                } else if (inode->i_state & I_DIRTY) {

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>