linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Nick Piggin <npiggin@gmail.com>
Cc: Jeff Moyer <jmoyer@redhat.com>, Jan Kara <jack@suse.cz>,
	Andrew Morton <akpm@linux-foundation.org>,
	linux-fsdevel <linux-fsdevel@vger.kernel.org>,
	linux-kernel@vger.kernel.org
Subject: Re: [patch] fs: aio fix rcu lookup
Date: Thu, 20 Jan 2011 12:02:37 -0800	[thread overview]
Message-ID: <20110120200237.GC17752@linux.vnet.ibm.com> (raw)
In-Reply-To: <AANLkTi=RdmWV9MLWzDxe6N_kWZQA+a2o8jg-PkB-7QGk@mail.gmail.com>

On Fri, Jan 21, 2011 at 05:31:53AM +1100, Nick Piggin wrote:
> On Thu, Jan 20, 2011 at 3:03 PM, Paul E. McKenney
> <paulmck@linux.vnet.ibm.com> wrote:
> > On Thu, Jan 20, 2011 at 08:20:00AM +1100, Nick Piggin wrote:
> >> On Thu, Jan 20, 2011 at 8:03 AM, Jeff Moyer <jmoyer@redhat.com> wrote:
> >> >> I don't know exactly how all programs use io_destroy -- of the small
> >> >> number that do, probably an even smaller number would care here. But I
> >> >> don't think it simplifies things enough to use synchronize_rcu for it.
> >> >
> >> > Above it sounded like you didn't think AIO should be using RCU at all.
> >>
> >> synchronize_rcu of course, not RCU (typo).
> >
> > I think that Nick is suggesting that call_rcu() be used instead.
> > Perhaps also very sparing use of synchronize_rcu_expedited(), which
> > is faster than synchronize_rcu(), but which which uses more CPU time.
> 
> call_rcu() is the obvious alternative, yes.
> 
> Basically, once we give in to synchronize_rcu() we're basically giving
> up. That's certainly a very good tradeoff for something like filesystem
> unregistration or module unload, it buys big simplifications in real
> fastpaths. But I just don't think it should be taken lightly.

Makes sense to me!

BTW, on your earlier usage classification:

> I think synchronize_rcu should firstly not be used unless it gives a good
> simplification, or speedup in fastpath.
>
> When that is satified, then it is a question of exactly what kind of slow
> path it should be used in. I don't think it should be used in process-
> synchronous code (eg syscalls) except for error cases, resource
> exhaustion, management syscalls (like module unload).

I don't have any feedback either way on your guidance to where
synchronize_rcu() should be used, as I believe that it depends a lot
on the details of usage, and would vary from one part of the kernel
to another, and possibly also over time.

But I am very glad to see that you have been thinking about it and
that you are putting forth some clear guidelines!

							Thanx, Paul

  reply	other threads:[~2011-01-20 20:02 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-01-14  1:35 [patch] fs: aio fix rcu lookup Nick Piggin
2011-01-14 14:52 ` Jeff Moyer
2011-01-14 15:00   ` Nick Piggin
2011-01-17 19:07     ` Jeff Moyer
2011-01-17 23:24       ` Nick Piggin
2011-01-18 17:21         ` Jeff Moyer
2011-01-18 19:01         ` Jan Kara
2011-01-18 22:17           ` Nick Piggin
2011-01-18 23:00             ` Jeff Moyer
2011-01-18 23:05               ` Nick Piggin
2011-01-18 23:52             ` Jan Kara
2011-01-19  0:20               ` Nick Piggin
2011-01-19 13:21                 ` Jan Kara
2011-01-19 16:03                   ` Nick Piggin
2011-01-19 16:50                     ` Jan Kara
2011-01-19 17:37                       ` Nick Piggin
2011-01-20 20:21                         ` Jan Kara
2011-01-19 19:13                   ` Jeff Moyer
2011-01-19 19:46                     ` Jeff Moyer
2011-01-19 20:18                       ` Nick Piggin
2011-01-19 20:32                         ` Jeff Moyer
2011-01-19 20:45                           ` Nick Piggin
2011-01-19 21:03                             ` Jeff Moyer
2011-01-19 21:20                               ` Nick Piggin
2011-01-20  4:03                                 ` Paul E. McKenney
2011-01-20 18:31                                   ` Nick Piggin
2011-01-20 20:02                                     ` Paul E. McKenney [this message]
2011-01-20 20:15                                       ` Eric Dumazet
2011-01-21 21:22                                         ` Paul E. McKenney
2011-01-20 20:16                                     ` Jan Kara
2011-01-20 21:16                                       ` Jeff Moyer
2011-02-01 16:24                                       ` 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=20110120200237.GC17752@linux.vnet.ibm.com \
    --to=paulmck@linux.vnet.ibm.com \
    --cc=akpm@linux-foundation.org \
    --cc=jack@suse.cz \
    --cc=jmoyer@redhat.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=npiggin@gmail.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).