lkml.org 
[lkml]   [2008]   [Mar]   [7]   [last100]   RSS Feed
Views: [more markup]  [less markup]  [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
FromNeilBrown <>
DateFri, 7 Mar 2008 16:00:53 +1100
Subject[PATCH 002 of 2] md: Reduce CPU wastage on idle md array with a write-intent bitmap. fix
Recent patch titled
  Reduce CPU wastage on idle md array with a write-intent bitmap.

would sometimes leave the array with dirty bitmap bits that stay
dirty.  A subsequent write would sort things out so it isn't a big
problem, but should be fixed nonetheless.

We need to make sure that when the bitmap becomes not "allclean", the
daemon_sleep really does get set to a sensible value.


Signed-off-by: Neil Brown <neilb@suse.de>

### Diffstat output
 ./drivers/md/bitmap.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
diff .prev/drivers/md/bitmap.c ./drivers/md/bitmap.c
--- .prev/drivers/md/bitmap.c	2008-03-07 15:51:23.000000000 +1100
+++ ./drivers/md/bitmap.c	2008-03-07 15:56:57.000000000 +1100
@@ -1045,7 +1045,8 @@ void bitmap_daemon_work(struct bitmap *b
 	if (bitmap == NULL)
 		return;
 	if (time_before(jiffies, bitmap->daemon_lastrun + bitmap->daemon_sleep*HZ))
-		return;
+		goto done;
+
 	bitmap->daemon_lastrun = jiffies;
 	if (bitmap->allclean) {
 		bitmap->mddev->thread->timeout = MAX_SCHEDULE_TIMEOUT;
@@ -1142,6 +1143,7 @@ void bitmap_daemon_work(struct bitmap *b
 		}
 	}
 
+ done:
 	if (bitmap->allclean == 0)
 		bitmap->mddev->thread->timeout = bitmap->daemon_sleep * HZ;
 }

\
 
 \ /
  Last update: 2008-03-07 06:05    [from the cache]
©2003-2008