[PATCH] Too many disk writes in 2.4.8-pre8

[Peter Osterlund]

There is a problem in sync_old_buffers() in kernel 2.4.8-pre8.
write_some_buffers() is always called at least once, which means that
under low/moderate I/O load, b_flushtime doesn't have any effect.
Everything is flushed to disk as soon as possible.

The result is unnecessary I/O and slowdown in a lot of cases. For example,
while configuring gcc, the system is paging out 400-500Kb/s constantly,
while in previous kernels the pageout rate was more like 10-20Kb/s.

The patch below fixes this problem:

--- linux/fs/buffer.c.orig	Fri Aug 10 21:42:52 2001
+++ linux/fs/buffer.c	Fri Aug 10 21:34:47 2001
@@ -2601,20 +2601,20 @@
 	sync_supers(0);
 	unlock_kernel();

-	spin_lock(&lru_list_lock);
 	for (;;) {
-		if (write_some_buffers(NODEV)) {
-			struct buffer_head *bh;
+		struct buffer_head *bh;

-			spin_lock(&lru_list_lock);
-			bh = lru_list[BUF_DIRTY];
-			if (bh && !time_before(jiffies, bh->b_flushtime))
-				continue;
+		spin_lock(&lru_list_lock);
+		bh = lru_list[BUF_DIRTY];
+		if (!bh || time_before(jiffies, bh->b_flushtime)) {
 			spin_unlock(&lru_list_lock);
+			break;
 		}
-		run_task_queue(&tq_disk);
-		return 0;
+		if (!write_some_buffers(NODEV))
+			break;
 	}
+	run_task_queue(&tq_disk);
+	return 0;
 }

 int block_sync_page(struct page *page)

-- 
Peter Österlund             [e-mail]
Sköndalsvägen 35            http://home1.swipnet.se/~w-15919
S-128 66 Sköndal            [phone]
Sweden