Re: [Bugme-new] [Bug 15426] New: Running many copies of bonnie++ on different filesystems seems to deadlock in sync

[Andre Noll <maan@systemlinux.org>]

[-- Attachment #1: Type: text/plain, Size: 1577 bytes --]

On 13:03, Jens Axboe wrote:

> Our sync is currently broken in that regard, since it'll wait for too
> long. We have a debated patch going, I have included it below. Any
> chance you could give it a whirl?

That's the same patch you posted two weeks ago and on which Linus
commented he thinks it is broken. He also proposed another, simpler
patch (included below) which I'm running for a few days and which
fixes the sync problem for me.

However, it is not clear at all whether that simpler patch is
correct. Any comments on that?

Thanks
Andre

---
 fs/fs-writeback.c |   15 ++++++++++-----
 1 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 1a7c42c..a0a8424 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -738,11 +738,16 @@ static long wb_writeback(struct bdi_writeback *wb,
 	long wrote = 0;
 	struct inode *inode;
 
-	if (wbc.for_kupdate) {
-		wbc.older_than_this = &oldest_jif;
-		oldest_jif = jiffies -
-				msecs_to_jiffies(dirty_expire_interval * 10);
-	}
+	/*
+	 * We never write back data that was dirtied _after_ we
+	 * started writeback. But kupdate doesn't even want to
+	 * write back recently dirtied stuff, only older data.
+	 */
+	oldest_jif = jiffies-1;
+	wbc.older_than_this = &oldest_jif;
+	if (wbc.for_kupdate)
+		oldest_jif -= msecs_to_jiffies(dirty_expire_interval * 10);
+
 	if (!wbc.range_cyclic) {
 		wbc.range_start = 0;
 		wbc.range_end = LLONG_MAX;
-- 
The only person who always got his work done by Friday was Robinson Crusoe

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]