linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@infradead.org>
To: Jan Kara <jack@suse.cz>
Cc: Wu Fengguang <fengguang.wu@intel.com>,
	"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
	Christoph Hellwig <hch@infradead.org>,
	Dave Chinner <david@fromorbit.com>
Subject: Re: [PATCH 1/2] writeback: Improve busyloop prevention
Date: Thu, 20 Oct 2011 05:46:49 -0400	[thread overview]
Message-ID: <20111020094649.GA11291@infradead.org> (raw)
In-Reply-To: <20111018005128.GI4528@quack.suse.cz>

On Tue, Oct 18, 2011 at 02:51:28AM +0200, Jan Kara wrote:
> > It's still puzzling why dirty pages remain at 286 and does not get
> > cleaned by either flusher threads for local XFS and NFSROOT for so
> > long time..
>   I was looking at this as well. So the reason why pages were not cleaned
> by the flusher thread is that there were 2 dirty inodes and the inode with
> dirty pages had i_dirtied_whan newer than the time when we started this
> background writeback. Thus the running background writeback work always
> included only the other inode which has no dirty pages but I_DIRTY_SYNC set.
> Apparently XFS is stubborn and refuses to write the inode although we try
> rather hard. That is probably because dd writing to this inode is stuck in
> balance_dirty_pages() and holds ilock - which is a bit unfortunate behavior
> but what can we do...

Stop writing data from balance_dirty_pages()?

Anyway, XFS tries very hard to not block in a non-block ->write_inode
or ->writepages, which generally is a good thing to avoid getting stuck
in the flusher thread.  For cases like this where an inode is long
beyond it's due time it might make sense to simply do a synchronous
write_inode from the flusher thread to force the inode out.


  parent reply	other threads:[~2011-10-20  9:46 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-10-12 20:57 [PATCH 0/2 v4] writeback: Improve busyloop prevention and inode requeueing Jan Kara
2011-10-12 20:57 ` [PATCH 1/2] writeback: Improve busyloop prevention Jan Kara
2011-10-13 14:26   ` Wu Fengguang
2011-10-13 20:13     ` Jan Kara
2011-10-14  7:18       ` Christoph Hellwig
2011-10-14 19:31         ` Chris Mason
     [not found]     ` <20111013143939.GA9691@localhost>
2011-10-13 20:18       ` Jan Kara
2011-10-14 16:00         ` Wu Fengguang
2011-10-14 16:28           ` Wu Fengguang
2011-10-18  0:51             ` Jan Kara
2011-10-18 14:35               ` Wu Fengguang
2011-10-19 11:56                 ` Jan Kara
2011-10-19 13:25                   ` Wu Fengguang
2011-10-19 13:30                   ` Wu Fengguang
2011-10-19 13:35                   ` Wu Fengguang
2011-10-20 12:09                   ` Wu Fengguang
2011-10-20 12:33                     ` Wu Fengguang
2011-10-20 13:39                       ` Wu Fengguang
2011-10-20 22:26                         ` Jan Kara
2011-10-22  4:20                           ` Wu Fengguang
2011-10-24 15:45                             ` Jan Kara
     [not found]                           ` <20111027063133.GA10146@localhost>
2011-10-27 20:31                             ` Jan Kara
     [not found]                               ` <20111101134231.GA31718@localhost>
2011-11-01 21:53                                 ` Jan Kara
2011-11-02 17:25                                   ` Wu Fengguang
     [not found]                               ` <20111102185603.GA4034@localhost>
2011-11-03  1:51                                 ` Jan Kara
2011-11-03 14:52                                   ` Wu Fengguang
     [not found]                                   ` <20111104152054.GA11577@localhost>
2011-11-08 23:52                                     ` Jan Kara
2011-11-09 13:51                                       ` Wu Fengguang
2011-11-10 14:50                                       ` Jan Kara
2011-12-05  8:02                                         ` Wu Fengguang
2011-12-07 10:13                                           ` Jan Kara
2011-12-07 11:45                                             ` Wu Fengguang
     [not found]                           ` <20111027064745.GA14017@localhost>
2011-10-27 20:50                             ` Jan Kara
2011-10-20  9:46               ` Christoph Hellwig [this message]
2011-10-20 15:32                 ` Jan Kara
2011-10-15 12:41           ` Wu Fengguang
2011-10-12 20:57 ` [PATCH 2/2] writeback: Replace some redirty_tail() calls with requeue_io() Jan Kara
2011-10-13 14:30   ` Wu Fengguang
2011-10-13 14:15 ` [PATCH 0/2 v4] writeback: Improve busyloop prevention and inode requeueing Wu Fengguang
  -- strict thread matches above, loose matches on Subject: below --
2011-09-08  0:44 [PATCH 1/2] writeback: Improve busyloop prevention Jan Kara
2011-09-08  0:57 ` Wu Fengguang
2011-09-08 13:49   ` Jan Kara

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=20111020094649.GA11291@infradead.org \
    --to=hch@infradead.org \
    --cc=david@fromorbit.com \
    --cc=fengguang.wu@intel.com \
    --cc=jack@suse.cz \
    --cc=linux-fsdevel@vger.kernel.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 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).