linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Wu Fengguang <fengguang.wu@intel.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Jan Kara <jack@suse.cz>, Wu Fengguang <fengguang.wu@intel.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: <linux-fsdevel@vger.kernel.org>
Cc: LKML <linux-kernel@vger.kernel.org>
Subject: [PATCH 10/18] writeback: avoid extra sync work at enqueue time
Date: Fri, 20 May 2011 05:45:40 +0800	[thread overview]
Message-ID: <20110519214745.534355616@intel.com> (raw)
In-Reply-To: 20110519214530.939830917@intel.com

[-- Attachment #1: writeback-kill-wb_start.patch --]
[-- Type: text/plain, Size: 2179 bytes --]

This removes writeback_control.wb_start and does more straightforward
sync livelock prevention by setting .older_than_this to prevent extra
inodes from being enqueued in the first place.

Acked-by: Jan Kara <jack@suse.cz>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
---
 fs/fs-writeback.c         |   16 +++-------------
 include/linux/writeback.h |    3 ---
 2 files changed, 3 insertions(+), 16 deletions(-)

--- linux-next.orig/fs/fs-writeback.c	2011-05-20 05:12:41.000000000 +0800
+++ linux-next/fs/fs-writeback.c	2011-05-20 05:12:42.000000000 +0800
@@ -544,15 +544,6 @@ static int writeback_sb_inodes(struct su
 			continue;
 		}
 
-		/*
-		 * Was this inode dirtied after sync_sb_inodes was called?
-		 * This keeps sync from extra jobs and livelock.
-		 */
-		if (inode_dirtied_after(inode, wbc->wb_start)) {
-			spin_unlock(&inode->i_lock);
-			return 1;
-		}
-
 		__iget(inode);
 
 		pages_skipped = wbc->pages_skipped;
@@ -585,9 +576,6 @@ static void __writeback_inodes_wb(struct
 {
 	int ret = 0;
 
-	if (!wbc->wb_start)
-		wbc->wb_start = jiffies; /* livelock avoidance */
-
 	while (!list_empty(&wb->b_io)) {
 		struct inode *inode = wb_inode(wb->b_io.prev);
 		struct super_block *sb = inode->i_sb;
@@ -686,7 +674,9 @@ static long wb_writeback(struct bdi_writ
 	if (wbc.sync_mode == WB_SYNC_ALL || wbc.tagged_writepages)
 		write_chunk = LONG_MAX;
 
-	wbc.wb_start = jiffies; /* livelock avoidance */
+	oldest_jif = jiffies;
+	wbc.older_than_this = &oldest_jif;
+
 	spin_lock(&wb->list_lock);
 	for (;;) {
 		/*
--- linux-next.orig/include/linux/writeback.h	2011-05-20 05:11:32.000000000 +0800
+++ linux-next/include/linux/writeback.h	2011-05-20 05:12:42.000000000 +0800
@@ -26,9 +26,6 @@ struct writeback_control {
 	enum writeback_sync_modes sync_mode;
 	unsigned long *older_than_this;	/* If !NULL, only write back inodes
 					   older than this */
-	unsigned long wb_start;         /* Time writeback_inodes_wb was
-					   called. This is needed to avoid
-					   extra jobs and livelock */
 	long nr_to_write;		/* Write this many pages, and decrement
 					   this for each page written */
 	long pages_skipped;		/* Pages which were not written */



  parent reply	other threads:[~2011-05-19 21:54 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-05-19 21:45 [PATCH 00/18] writeback fixes and cleanups for 2.6.40 (v3) Wu Fengguang
2011-05-19 21:45 ` [PATCH 01/18] writeback: introduce .tagged_writepages for the WB_SYNC_NONE sync stage Wu Fengguang
2011-05-19 21:45 ` [PATCH 02/18] writeback: update dirtied_when for synced inode to prevent livelock Wu Fengguang
2011-05-19 21:45 ` [PATCH 03/18] writeback: introduce writeback_control.inodes_cleaned Wu Fengguang
2011-05-19 21:45 ` [PATCH 04/18] writeback: try more writeback as long as something was written Wu Fengguang
2011-05-19 21:45 ` [PATCH 05/18] writeback: the kupdate expire timestamp should be a moving target Wu Fengguang
2011-05-19 21:45 ` [PATCH 06/18] writeback: sync expired inodes first in background writeback Wu Fengguang
2011-05-19 21:45 ` [PATCH 07/18] writeback: refill b_io iff empty Wu Fengguang
2011-05-19 21:45 ` [PATCH 08/18] writeback: split inode_wb_list_lock into bdi_writeback.list_lock Wu Fengguang
2011-05-19 21:45 ` [PATCH 09/18] writeback: elevate queue_io() into wb_writeback() Wu Fengguang
2011-05-19 21:45 ` Wu Fengguang [this message]
2011-05-19 21:45 ` [PATCH 11/18] writeback: add bdi_dirty_limit() kernel-doc Wu Fengguang
2011-05-19 21:45 ` [PATCH 12/18] writeback: skip balance_dirty_pages() for in-memory fs Wu Fengguang
2011-05-19 21:45 ` [PATCH 13/18] writeback: remove writeback_control.more_io Wu Fengguang
2011-05-19 21:45 ` [PATCH 14/18] writeback: make writeback_control.nr_to_write straight Wu Fengguang
2011-05-19 22:06   ` Wu Fengguang
2011-05-19 23:29     ` Dave Chinner
2011-05-20  4:07       ` Wu Fengguang
2011-05-20  6:52         ` Dave Chinner
2011-05-20  7:15           ` Wu Fengguang
2011-05-20  7:26             ` Wu Fengguang
2011-05-19 21:45 ` [PATCH 15/18] writeback: remove .nonblocking and .encountered_congestion Wu Fengguang
2011-05-19 21:45 ` [PATCH 16/18] writeback: trace event writeback_single_inode Wu Fengguang
2011-05-19 21:45 ` [PATCH 17/18] writeback: trace event writeback_queue_io Wu Fengguang
2011-05-19 21:45 ` [PATCH 18/18] writeback: rearrange the wb_writeback() loop Wu Fengguang
2011-05-23  9:07 ` [PATCH 00/18] writeback fixes and cleanups for 2.6.40 (v3) Christoph Hellwig
2011-05-23  9:28   ` Wu Fengguang
2011-05-24  3:28     ` Wu Fengguang
2011-06-01 22:31       ` Andrew Morton
2011-06-02  2:29         ` Wu Fengguang
2011-06-07 12:13           ` writeback merge status, was " Christoph Hellwig
2011-06-07 20:15             ` Andrew Morton
2011-06-07 21:11               ` Wu Fengguang
  -- strict thread matches above, loose matches on Subject: below --
2011-05-24  5:14 [PATCH 00/18] writeback fixes and cleanups for 2.6.40 (v4) Wu Fengguang
2011-05-24  5:14 ` [PATCH 10/18] writeback: avoid extra sync work at enqueue time 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=20110519214745.534355616@intel.com \
    --to=fengguang.wu@intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=jack@suse.cz \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).