From: John McCutchan <john@johnmccutchan.com>
To: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Andrew Morton <akpm@osdl.org>,
holt@sgi.com, linux-kernel@vger.kernel.org, rml@novell.com,
arnd@arndb.de, hch@lst.de, Dipankar Sarma <dipankar@in.ibm.com>
Subject: Re: udevd is killing file write performance.
Date: Sun, 26 Feb 2006 11:55:18 -0500 [thread overview]
Message-ID: <1140972918.15634.1.camel@localhost.localdomain> (raw)
In-Reply-To: <43FEB0BF.6080403@yahoo.com.au>
On Fri, 2006-24-02 at 18:07 +1100, Nick Piggin wrote:
> Andrew Morton wrote:
> > John McCutchan <john@johnmccutchan.com> wrote:
> >
> >> > > @@ -538,7 +537,7 @@
> >> > > struct dentry *parent;
> >> > > struct inode *inode;
> >> > >
> >> > > - if (!atomic_read (&inotify_watches))
> >> > > + if (!atomic_read (&dentry->d_sb->s_inotify_watches))
> >> > > return;
> >> > >
> >> >
> >> > What happens here if we're watching a mountpoint - the parent is on a
> >> > different fs?
> >>
> >> There are four cases to consider here.
> >>
> >> Case 1: parent fs watched and child fs watched
> >> correct results
> >> Case 2: parent fs watched and child fs not watched
> >> We may not deliver an event that should be delivered.
> >> Case 3: parent fs not watched and child fs watched
> >> We take d_lock when we don't need to
> >> Case 4: parent fs not watched and child fs not watched
> >> correct results
> >>
> >> Case 2 screws us. We have to take the lock to even look at the parent's
> >> dentry->d_sb->s_inotify_watches. I don't know of a way around this one.
> >
> >
> > Yeah. There are a lot of "screw"s in this thread.
> >
> > I wonder if RCU can save us - if we do an rcu_read_lock() we at least know
> > that the dentries won't get deallocated. Then we can take a look at
> > d_parent (which might not be the parent any more). Once in a million years
> > we might send a false event or miss sending an event, depending on where
> > our dentry suddenly got moved to. Not very nice, but at least it won't
> > oops.
> >
> > (hopefully cc's Dipankar)
>
> I saw this problem when testing my lockless pagecache a while back.
>
> Attached is a first implementation of what was my idea then of how
> to solve it... note it is pretty rough and I never got around to doing
> much testing of it.
>
> Basically: moves work out of inotify event time and to inotify attach
> /detach time while staying out of the core VFS.
This looks really good. There might be some corner cases but it looks
like it will solve this problem nicely.
--
John McCutchan <john@johnmccutchan.com>
next prev parent reply other threads:[~2006-02-26 16:55 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-02-22 13:42 udevd is killing file write performance Robin Holt
2006-02-22 13:55 ` Andrew Morton
2006-02-22 16:48 ` John McCutchan
2006-02-22 17:50 ` Robin Holt
2006-02-22 20:05 ` Andrew Morton
2006-02-22 21:50 ` Jeff V. Merkey
2006-02-23 12:56 ` Robin Holt
2006-02-23 13:42 ` David Chinner
2006-02-22 22:52 ` John McCutchan
2006-02-22 23:12 ` Andrew Morton
2006-02-22 23:41 ` John McCutchan
2006-02-24 0:14 ` Andrew Morton
2006-02-24 0:14 ` Andrew Morton
2006-02-24 5:47 ` John McCutchan
2006-02-24 6:00 ` Andrew Morton
2006-02-24 7:07 ` Nick Piggin
2006-02-24 7:16 ` Andrew Morton
2006-02-24 7:19 ` Nick Piggin
2006-02-26 16:58 ` John McCutchan
2006-02-24 18:56 ` Robin Holt
2006-02-25 2:44 ` Nick Piggin
2006-02-25 15:53 ` [patch] inotify: lock avoidance with parent watch status in dentry Nick Piggin
2006-02-28 0:48 ` Andrew Morton
2006-02-26 16:55 ` John McCutchan [this message]
2006-02-27 10:11 ` udevd is killing file write performance Nick Piggin
2006-02-27 20:17 ` John McCutchan
2006-02-23 20:38 ` Benjamin LaHaise
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=1140972918.15634.1.camel@localhost.localdomain \
--to=john@johnmccutchan.com \
--cc=akpm@osdl.org \
--cc=arnd@arndb.de \
--cc=dipankar@in.ibm.com \
--cc=hch@lst.de \
--cc=holt@sgi.com \
--cc=linux-kernel@vger.kernel.org \
--cc=nickpiggin@yahoo.com.au \
--cc=rml@novell.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;
as well as URLs for NNTP newsgroup(s).