From: Wu Fengguang <fengguang.wu@intel.com>
To: linux-mm <linux-mm@kvack.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
Wu Fengguang <fengguang.wu@intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Rik van Riel <riel@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Li Shaohua <shaohua.li@intel.com>
Subject: [PATCH 04/17] writeback: quit throttling when bdi dirty/writeback pages go down
Date: Sun, 12 Sep 2010 23:49:49 +0800 [thread overview]
Message-ID: <20100912155203.206486412@intel.com> (raw)
In-Reply-To: 20100912154945.758129106@intel.com
[-- Attachment #1: writeback-bdi-throttle-break.patch --]
[-- Type: text/plain, Size: 2269 bytes --]
Tests show that bdi_thresh may take minutes to ramp up on a typical
desktop. The time should be improvable but cannot be eliminated totally.
So when (background_thresh + dirty_thresh)/2 is reached and
balance_dirty_pages() starts to throttle the task, it will suddenly find
the (still low and ramping up) bdi_thresh is exceeded _excessively_. Here
we definitely don't want to stall the task for one minute. So introduce
an alternative way to break out of the loop when the bdi dirty/write
pages has dropped by a reasonable amount.
When dirty_background_ratio is set close to dirty_ratio, bdi_thresh may
also be constantly exceeded due to the task_dirty_limit() gap.
It will take at least 200ms before trying to break out.
(pages_dirtied * 8) is used because in this situation pages_dirtied will
typically be small numbers (eg. 3 pages) due to the fast back off logic.
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
---
mm/page-writeback.c | 15 +++++++++++++++
1 file changed, 15 insertions(+)
--- linux-next.orig/mm/page-writeback.c 2010-09-09 15:51:38.000000000 +0800
+++ linux-next/mm/page-writeback.c 2010-09-12 13:10:02.000000000 +0800
@@ -463,6 +463,7 @@ static void balance_dirty_pages(struct a
{
long nr_reclaimable, bdi_nr_reclaimable;
long nr_writeback, bdi_nr_writeback;
+ long bdi_prev_dirty3 = 0;
unsigned long background_thresh;
unsigned long dirty_thresh;
unsigned long bdi_thresh;
@@ -516,6 +517,20 @@ static void balance_dirty_pages(struct a
bdi_nr_writeback = bdi_stat(bdi, BDI_WRITEBACK);
}
+ /*
+ * bdi_thresh could get exceeded for long time:
+ * - bdi_thresh takes some time to ramp up from the initial 0
+ * - users may set dirty_background_ratio close to dirty_ratio
+ * (at least 1/8 gap is preferred)
+ * So offer a complementary way to break out of the loop when
+ * enough bdi pages have been cleaned during our pause time.
+ */
+ if (nr_reclaimable + nr_writeback <= dirty_thresh &&
+ bdi_prev_dirty3 - (bdi_nr_reclaimable + bdi_nr_writeback) >
+ (long)pages_dirtied * 8)
+ break;
+ bdi_prev_dirty3 = bdi_nr_reclaimable + bdi_nr_writeback;
+
if (bdi_nr_reclaimable + bdi_nr_writeback <=
bdi_thresh - bdi_thresh / DIRTY_SOFT_THROTTLE_RATIO)
goto check_exceeded;
WARNING: multiple messages have this Message-ID (diff)
From: Wu Fengguang <fengguang.wu@intel.com>
To: linux-mm <linux-mm@kvack.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
Wu Fengguang <fengguang.wu@intel.com>,
Andrew Morton <akpm@linux-foundation.org>,
Theodore Ts'o <tytso@mit.edu>, Dave Chinner <david@fromorbit.com>,
Jan Kara <jack@suse.cz>, Peter Zijlstra <a.p.zijlstra@chello.nl>,
Mel Gorman <mel@csn.ul.ie>, Rik van Riel <riel@redhat.com>,
KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>,
Chris Mason <chris.mason@oracle.com>,
Christoph Hellwig <hch@lst.de>, Li Shaohua <shaohua.li@intel.com>
Subject: [PATCH 04/17] writeback: quit throttling when bdi dirty/writeback pages go down
Date: Sun, 12 Sep 2010 23:49:49 +0800 [thread overview]
Message-ID: <20100912155203.206486412@intel.com> (raw)
In-Reply-To: 20100912154945.758129106@intel.com
[-- Attachment #1: writeback-bdi-throttle-break.patch --]
[-- Type: text/plain, Size: 2494 bytes --]
Tests show that bdi_thresh may take minutes to ramp up on a typical
desktop. The time should be improvable but cannot be eliminated totally.
So when (background_thresh + dirty_thresh)/2 is reached and
balance_dirty_pages() starts to throttle the task, it will suddenly find
the (still low and ramping up) bdi_thresh is exceeded _excessively_. Here
we definitely don't want to stall the task for one minute. So introduce
an alternative way to break out of the loop when the bdi dirty/write
pages has dropped by a reasonable amount.
When dirty_background_ratio is set close to dirty_ratio, bdi_thresh may
also be constantly exceeded due to the task_dirty_limit() gap.
It will take at least 200ms before trying to break out.
(pages_dirtied * 8) is used because in this situation pages_dirtied will
typically be small numbers (eg. 3 pages) due to the fast back off logic.
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
---
mm/page-writeback.c | 15 +++++++++++++++
1 file changed, 15 insertions(+)
--- linux-next.orig/mm/page-writeback.c 2010-09-09 15:51:38.000000000 +0800
+++ linux-next/mm/page-writeback.c 2010-09-12 13:10:02.000000000 +0800
@@ -463,6 +463,7 @@ static void balance_dirty_pages(struct a
{
long nr_reclaimable, bdi_nr_reclaimable;
long nr_writeback, bdi_nr_writeback;
+ long bdi_prev_dirty3 = 0;
unsigned long background_thresh;
unsigned long dirty_thresh;
unsigned long bdi_thresh;
@@ -516,6 +517,20 @@ static void balance_dirty_pages(struct a
bdi_nr_writeback = bdi_stat(bdi, BDI_WRITEBACK);
}
+ /*
+ * bdi_thresh could get exceeded for long time:
+ * - bdi_thresh takes some time to ramp up from the initial 0
+ * - users may set dirty_background_ratio close to dirty_ratio
+ * (at least 1/8 gap is preferred)
+ * So offer a complementary way to break out of the loop when
+ * enough bdi pages have been cleaned during our pause time.
+ */
+ if (nr_reclaimable + nr_writeback <= dirty_thresh &&
+ bdi_prev_dirty3 - (bdi_nr_reclaimable + bdi_nr_writeback) >
+ (long)pages_dirtied * 8)
+ break;
+ bdi_prev_dirty3 = bdi_nr_reclaimable + bdi_nr_writeback;
+
if (bdi_nr_reclaimable + bdi_nr_writeback <=
bdi_thresh - bdi_thresh / DIRTY_SOFT_THROTTLE_RATIO)
goto check_exceeded;
--
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>
next prev parent reply other threads:[~2010-09-12 15:55 UTC|newest]
Thread overview: 98+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-09-12 15:49 [PATCH 00/17] [RFC] soft and dynamic dirty throttling limits Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-12 15:49 ` [PATCH 01/17] writeback: remove the internal 5% low bound on dirty_ratio Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-13 9:23 ` Johannes Weiner
2010-09-13 9:23 ` Johannes Weiner
2010-09-13 9:51 ` Mel Gorman
2010-09-13 9:51 ` Mel Gorman
2010-09-13 9:57 ` Wu Fengguang
2010-09-13 9:57 ` Wu Fengguang
2010-09-13 10:10 ` Mel Gorman
2010-09-13 10:10 ` Mel Gorman
2010-09-12 15:49 ` [PATCH 02/17] writeback: IO-less balance_dirty_pages() Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-13 8:45 ` Dave Chinner
2010-09-13 8:45 ` Dave Chinner
2010-09-13 11:38 ` Wu Fengguang
2010-09-13 11:38 ` Wu Fengguang
2010-09-12 15:49 ` [PATCH 03/17] writeback: per-task rate limit to balance_dirty_pages() Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang [this message]
2010-09-12 15:49 ` [PATCH 04/17] writeback: quit throttling when bdi dirty/writeback pages go down Wu Fengguang
2010-09-12 15:49 ` [PATCH 05/17] writeback: quit throttling when signal pending Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-12 20:46 ` Neil Brown
2010-09-12 20:46 ` Neil Brown
2010-09-13 1:55 ` Wu Fengguang
2010-09-13 1:55 ` Wu Fengguang
2010-09-13 3:21 ` Neil Brown
2010-09-13 3:21 ` Neil Brown
2010-09-13 3:48 ` Wu Fengguang
2010-09-13 3:48 ` Wu Fengguang
2010-09-14 8:23 ` KOSAKI Motohiro
2010-09-14 8:23 ` KOSAKI Motohiro
2010-09-14 8:33 ` Wu Fengguang
2010-09-14 8:33 ` Wu Fengguang
2010-09-14 8:44 ` KOSAKI Motohiro
2010-09-14 8:44 ` KOSAKI Motohiro
2010-09-14 9:17 ` Wu Fengguang
2010-09-14 9:17 ` Wu Fengguang
2010-09-14 9:25 ` KOSAKI Motohiro
2010-09-14 9:25 ` KOSAKI Motohiro
2010-09-12 15:49 ` [PATCH 06/17] writeback: move task dirty fraction to balance_dirty_pages() Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-12 15:49 ` [PATCH 07/17] writeback: add trace event for balance_dirty_pages() Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-12 15:49 ` [PATCH 08/17] writeback: account per-bdi accumulated written pages Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-12 15:59 ` Wu Fengguang
2010-09-12 15:59 ` Wu Fengguang
2010-09-14 8:32 ` KOSAKI Motohiro
2010-09-14 8:32 ` KOSAKI Motohiro
2010-09-12 15:49 ` [PATCH 09/17] writeback: bdi write bandwidth estimation Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-12 15:49 ` [PATCH 10/17] writeback: show bdi write bandwidth in debugfs Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-12 15:49 ` [PATCH 11/17] writeback: make nr_to_write a per-file limit Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-12 15:49 ` [PATCH 12/17] writeback: scale IO chunk size up to device bandwidth Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-12 15:49 ` [PATCH 13/17] writeback: reduce per-bdi dirty threshold ramp up time Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-12 16:15 ` Wu Fengguang
2010-09-12 16:15 ` Wu Fengguang
2010-09-12 15:49 ` [PATCH 14/17] vmscan: add scan_control.priority Wu Fengguang
2010-09-12 15:49 ` Wu Fengguang
2010-09-12 15:50 ` [PATCH 15/17] mm: lower soft dirty limits on memory pressure Wu Fengguang
2010-09-12 15:50 ` Wu Fengguang
2010-09-13 9:40 ` Wu Fengguang
2010-09-13 9:40 ` Wu Fengguang
2010-09-12 15:50 ` [PATCH 16/17] mm: create /vm/dirty_pressure in debugfs Wu Fengguang
2010-09-12 15:50 ` Wu Fengguang
2010-09-12 15:50 ` [PATCH 17/17] writeback: consolidate balance_dirty_pages() variable names Wu Fengguang
2010-09-12 15:50 ` Wu Fengguang
2010-10-12 14:17 ` [PATCH 00/17] [RFC] soft and dynamic dirty throttling limits Christoph Hellwig
2010-10-12 14:17 ` Christoph Hellwig
2010-10-13 3:07 ` Dave Chinner
2010-10-13 3:07 ` Dave Chinner
2010-10-13 3:23 ` Wu Fengguang
2010-10-13 3:23 ` Wu Fengguang
2010-10-13 8:26 ` Wu Fengguang
2010-10-13 8:26 ` Wu Fengguang
2010-10-13 9:26 ` Dave Chinner
2010-10-13 9:26 ` Dave Chinner
2010-11-01 6:24 ` Dave Chinner
2010-11-01 6:24 ` Dave Chinner
2010-11-04 3:41 ` Wu Fengguang
2010-11-04 3:41 ` Wu Fengguang
2010-11-04 12:48 ` Dave Chinner
2010-11-04 12:48 ` Dave Chinner
2010-11-04 13:12 ` Christoph Hellwig
2010-11-04 13:12 ` Christoph Hellwig
2010-11-05 14:56 ` Wu Fengguang
2010-11-05 14:56 ` Wu Fengguang
2010-11-06 10:42 ` Dave Chinner
2010-11-06 10:42 ` Dave Chinner
2010-10-14 13:12 ` Wu Fengguang
2010-10-14 13:12 ` Wu Fengguang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20100912155203.206486412@intel.com \
--to=fengguang.wu@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.