linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Wu Fengguang <fengguang.wu@intel.com>
To: Jan Kara <jack@suse.cz>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Dave Chinner <david@fromorbit.com>,
	Christoph Hellwig <hch@infradead.org>, Mel Gorman <mel@csn.ul.ie>,
	Chris Mason <chris.mason@oracle.com>,
	Jens Axboe <jens.axboe@oracle.com>,
	LKML <linux-kernel@vger.kernel.org>,
	"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>
Subject: Re: [PATCH 5/6] writeback: try more writeback as long as something was written
Date: Mon, 26 Jul 2010 20:39:37 +0800	[thread overview]
Message-ID: <20100726123937.GB11947@localhost> (raw)
In-Reply-To: <20100723173953.GB20540@quack.suse.cz>

On Sat, Jul 24, 2010 at 01:39:54AM +0800, Jan Kara wrote:
> On Thu 22-07-10 13:09:33, Wu Fengguang wrote:
> > writeback_inodes_wb()/__writeback_inodes_sb() are not agressive in that
> > they only populate b_io when necessary at entrance time. When the queued
> > set of inodes are all synced, they just return, possibly with
> > wbc.nr_to_write > 0.
> > 
> > For kupdate and background writeback, there may be more eligible inodes
> > sitting in b_dirty when the current set of b_io inodes are completed. So
> > it is necessary to try another round of writeback as long as we made some
> > progress in this round. When there are no more eligible inodes, no more
> > inodes will be enqueued in queue_io(), hence nothing could/will be
> > synced and we may safely bail.
> > 
> > This will livelock sync when there are heavy dirtiers. However in that case
> > sync will already be livelocked w/o this patch, as the current livelock
> > avoidance code is virtually a no-op (for one thing, wb_time should be
> > set statically at sync start time and be used in move_expired_inodes()).
> > The sync livelock problem will be addressed in other patches.
>   Hmm, any reason why you don't solve this problem by just removing the
> condition before queue_io()? It would also make the logic simpler - always

Yeah I'll remove queue_io() in the coming sync livelock patchset.
This patchset does the below. Though awkward, it avoids unnecessary
behavior changes for non-background cases.

-       if (!wbc->for_kupdate || list_empty(&wb->b_io))
+       if (!(wbc->for_kupdate || wbc->for_background) || list_empty(&wb->b_io))
                queue_io(wb, wbc);


> queue all inodes that are eligible for writeback...

Thanks,
Fengguang

--
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>

  reply	other threads:[~2010-07-26 12:39 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-07-22  5:09 [PATCH 0/6] [RFC] writeback: try to write older pages first Wu Fengguang
2010-07-22  5:09 ` [PATCH 1/6] writeback: pass writeback_control down to move_expired_inodes() Wu Fengguang
2010-07-23 18:16   ` Jan Kara
2010-07-26 10:44   ` Mel Gorman
2010-08-01 15:23   ` Minchan Kim
2010-07-22  5:09 ` [PATCH 2/6] writeback: the kupdate expire timestamp should be a moving target Wu Fengguang
2010-07-23 18:17   ` Jan Kara
2010-07-26 10:52   ` Mel Gorman
2010-07-26 11:32     ` Wu Fengguang
2010-08-01 15:29   ` Minchan Kim
2010-07-22  5:09 ` [PATCH 3/6] writeback: kill writeback_control.more_io Wu Fengguang
2010-07-23 18:24   ` Jan Kara
2010-07-26 10:53   ` Mel Gorman
2010-08-01 15:34   ` Minchan Kim
2010-08-05 14:50     ` Wu Fengguang
2010-08-05 14:55       ` Wu Fengguang
2010-08-05 14:56       ` Minchan Kim
2010-08-05 15:26         ` Wu Fengguang
2010-07-22  5:09 ` [PATCH 4/6] writeback: sync expired inodes first in background writeback Wu Fengguang
2010-07-23 18:15   ` Jan Kara
2010-07-26 11:51     ` Wu Fengguang
2010-07-26 12:12       ` Jan Kara
2010-07-26 12:29         ` Wu Fengguang
2010-07-26 10:57   ` Mel Gorman
2010-07-26 12:00     ` Wu Fengguang
2010-07-26 12:20       ` Jan Kara
2010-07-26 12:31         ` Wu Fengguang
2010-07-26 12:39           ` Jan Kara
2010-07-26 12:47             ` Wu Fengguang
2010-07-26 12:56     ` Wu Fengguang
2010-07-26 12:59       ` Mel Gorman
2010-07-26 13:11         ` Wu Fengguang
2010-07-27  9:45           ` Mel Gorman
2010-08-01 15:15           ` Minchan Kim
2010-07-22  5:09 ` [PATCH 5/6] writeback: try more writeback as long as something was written Wu Fengguang
2010-07-23 17:39   ` Jan Kara
2010-07-26 12:39     ` Wu Fengguang [this message]
2010-07-26 11:01   ` Mel Gorman
2010-07-26 11:39     ` Wu Fengguang
2010-07-22  5:09 ` [PATCH 6/6] writeback: introduce writeback_control.inodes_written Wu Fengguang
2010-07-26 11:04   ` Mel Gorman
2010-07-23 10:24 ` [PATCH 0/6] [RFC] writeback: try to write older pages first Mel Gorman
2010-07-26  7:18   ` Wu Fengguang
2010-07-26 10:42     ` Mel Gorman
2010-07-26 10:28 ` Itaru Kitayama
2010-07-26 11:47   ` Wu Fengguang
  -- strict thread matches above, loose matches on Subject: below --
2011-04-19  3:00 [PATCH 0/6] writeback: moving expire targets for background/kupdate works Wu Fengguang
2011-04-19  3:00 ` [PATCH 5/6] writeback: try more writeback as long as something was written Wu Fengguang
2011-04-19 10:20   ` Jan Kara
2011-04-19 11:16     ` Wu Fengguang
2011-04-19 21:10       ` Jan Kara
2011-04-20  7:50         ` Wu Fengguang
2011-04-20 15:22           ` Jan Kara
2011-04-21  3:33             ` Wu Fengguang
2011-04-21  4:39               ` Christoph Hellwig
2011-04-21  6:05                 ` Wu Fengguang
2011-04-21 16:41                   ` Jan Kara
2011-04-22  2:32                     ` Wu Fengguang
2011-04-22 21:23                       ` Jan Kara
2011-04-21  7:09               ` Dave Chinner
2011-04-21  7:14                 ` Christoph Hellwig
2011-04-21  7:52                   ` Dave Chinner
2011-04-21  8:00                     ` Christoph Hellwig

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=20100726123937.GB11947@localhost \
    --to=fengguang.wu@intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=chris.mason@oracle.com \
    --cc=david@fromorbit.com \
    --cc=hch@infradead.org \
    --cc=jack@suse.cz \
    --cc=jens.axboe@oracle.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mel@csn.ul.ie \
    /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).