public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@infradead.org>
To: Dave Chinner <david@fromorbit.com>
Cc: xfs@oss.sgi.com
Subject: Re: [PATCH 4/6] xfs: introduce background inode reclaim work
Date: Thu, 10 Mar 2011 12:40:54 -0500	[thread overview]
Message-ID: <20110310174054.GD19609@infradead.org> (raw)
In-Reply-To: <1299715529-11026-5-git-send-email-david@fromorbit.com>

Why do we still keep the inode reclaim in the syncer work?  If we
already have this one doing it I don't think we need it there as well.

>  /*
> + * Queue a new inode reclaim pass if there isn't one already in progress.
> + * Wait for completion of the flush if necessary.

No, it doesn't wait ever.

> + */
> +static void
> +xfs_syncd_queue_reclaim(
> +	struct xfs_mount        *mp)
> +{
> +	queue_delayed_work(xfs_syncd_wq, &mp->m_reclaim_work,
> +			xfs_syncd_centisecs / 5 * msecs_to_jiffies(10));

What explanation is there for the magic 5?  E.g. why do we neeed to run
it exactly 5 times as often as the normal sync work? Should it have it's
own tunable?  And isn't ever 6 seconds by default a little often on
systems trying to saver power, especiall if there aren't any inodes to
reclaim?  Should we trigger starting this work off having reclaimable
inodes tagged in the radix tree?

> +	/* first unpin all the dirty and stale inodes. */
> +	xfs_log_force(mp, XFS_LOG_SYNC);

So we force out the log every 6 seconds.  That's a lot more often than
most other filesystem and might have adverse performance impact.

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

  reply	other threads:[~2011-03-10 17:38 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-03-10  0:05 [PATCH 0/6] xfs: Reduce OOM kill problems under heavy load V2 Dave Chinner
2011-03-10  0:05 ` [PATCH 1/6] xfs: introduce inode cluster buffer trylocks for xfs_iflush Dave Chinner
2011-03-10 17:31   ` Christoph Hellwig
2011-03-10  0:05 ` [PATCH 2/6] xfs: introduce a xfssyncd workqueue Dave Chinner
2011-03-10 17:34   ` Christoph Hellwig
2011-03-18  3:39     ` Dave Chinner
2011-03-10  0:05 ` [PATCH 3/6] xfs: convert ENOSPC inode flushing to use new syncd workqueue Dave Chinner
2011-03-10 17:35   ` Christoph Hellwig
2011-03-18  3:39     ` Dave Chinner
2011-03-10  0:05 ` [PATCH 4/6] xfs: introduce background inode reclaim work Dave Chinner
2011-03-10 17:40   ` Christoph Hellwig [this message]
2011-03-18  4:00     ` Dave Chinner
2011-03-10  0:05 ` [PATCH 5/6] xfs: convert the xfsaild threads to a workqueue Dave Chinner
2011-03-10 17:48   ` Christoph Hellwig
2011-03-18  4:06     ` Dave Chinner
2011-03-19 13:45       ` Christoph Hellwig
2011-04-03  0:38         ` Dave Chinner
2011-04-03 10:59           ` Christoph Hellwig
2011-04-04  2:11             ` Dave Chinner
2011-03-10  0:05 ` [PATCH 6/6] xfs: push the AIL from memory reclaim and periodic sync Dave Chinner
2011-03-10 21:31   ` Christoph Hellwig
2011-03-18  4:07     ` Dave Chinner

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=20110310174054.GD19609@infradead.org \
    --to=hch@infradead.org \
    --cc=david@fromorbit.com \
    --cc=xfs@oss.sgi.com \
    /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