caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: Gerd Stolpmann <gerd@gerd-stolpmann.de>
To: Goswin von Brederlow <goswin-v-b@web.de>
Cc: Florian Weimer <fw@deneb.enyo.de>, caml-list@inria.fr
Subject: Re: [Caml-list] Re: How to read different ints from a Bigarray?
Date: Fri, 30 Oct 2009 01:48:55 +0100	[thread overview]
Message-ID: <1256863735.24202.127.camel@flake.lan.gerd-stolpmann.de> (raw)
In-Reply-To: <87r5sl3gxb.fsf@frosties.localdomain>


Am Freitag, den 30.10.2009, 00:43 +0100 schrieb Goswin von Brederlow:
> Gerd Stolpmann <gerd@gerd-stolpmann.de> writes:
> 
> > Am Donnerstag, den 29.10.2009, 21:40 +0100 schrieb Florian Weimer:
> >> * Goswin von Brederlow:
> >> 
> >> > - The data is passed to libaio and needs to be kept alive and unmoved
> >> >   as long as libaio knows it.
> >> 
> >> It also has to be aligned to a 512-byte boundary, so you can use
> >> O_DIRECT.  Linux does not support truely asynchronous I/O without
> >> O_DIRECT AFAIK, which rarely makes it worth the trouble.
> >
> > Right. There is also the question whether aio for regular files (i.e.
> > files backed by page cache) is continued to be supported at all - it is
> > well known that Linus Torvalds doesn't like it. It can happen that at
> > some day aio will be restricted to block devices only.
> >
> > So I wouldn't use it for production code, but it is of course still an
> > interesting interface.
> >
> > Gerd
> 
> Damn. That seems so stupid. Then writing asynchronous will only be
> possible with creating a pot full of worker thread, each one writing
> one chunk. So you get all those chunks in random order submitted to
> the kernel, the kernel has to reorder them, fit them back together,
> write them and then wake up the right thread for each piece
> completed. So much extra work while libaio has all the data already in
> perfect structures for the kernel.

Well, this is exactly the implementation of the POSIX aio functions in
glibc. They are mapped to a bunch of threads.

> And how will you do barriers when writing with threads? Wait for all
> threads to complete every time you hit a barrier and thereby stalling
> the pipeline?

You can't implement barriers. When you have page-cache backed I/O (i.e.
non-direct I/O, no matter of aio or sync I/O) there is no control when
data is written. Ok, there is fsync but this is very coarse-grained
control.

Gerd
-- 
------------------------------------------------------------
Gerd Stolpmann, Bad Nauheimer Str.3, 64289 Darmstadt,Germany 
gerd@gerd-stolpmann.de          http://www.gerd-stolpmann.de
Phone: +49-6151-153855                  Fax: +49-6151-997714
------------------------------------------------------------


  reply	other threads:[~2009-10-30  0:43 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-10-28 13:54 Goswin von Brederlow
2009-10-28 14:16 ` Sylvain Le Gall
2009-10-28 15:00   ` [Caml-list] " Goswin von Brederlow
2009-10-28 15:17     ` Sylvain Le Gall
2009-10-28 17:57       ` [Caml-list] " Goswin von Brederlow
2009-10-28 18:19         ` Sylvain Le Gall
2009-10-28 21:05           ` [Caml-list] " Goswin von Brederlow
2009-10-28 21:26             ` Sylvain Le Gall
2009-10-28 22:48         ` [Caml-list] " blue storm
2009-10-29  9:50           ` Goswin von Brederlow
2009-10-29 10:34             ` Goswin von Brederlow
2009-10-29 12:20             ` Richard Jones
2009-10-29 17:07               ` Goswin von Brederlow
2009-10-30 20:30                 ` Richard Jones
2009-11-01 15:11                   ` Goswin von Brederlow
2009-11-01 19:57                     ` Richard Jones
2009-11-02 16:11                       ` Goswin von Brederlow
2009-11-02 16:33                         ` Mauricio Fernandez
2009-11-02 20:27                           ` Richard Jones
2009-11-03 13:18                             ` Goswin von Brederlow
2009-11-02 20:48                           ` Goswin von Brederlow
2009-10-29 20:40     ` Florian Weimer
2009-10-29 21:04       ` Gerd Stolpmann
2009-10-29 23:43         ` Goswin von Brederlow
2009-10-30  0:48           ` Gerd Stolpmann [this message]
2009-10-29 23:38       ` Goswin von Brederlow
2009-10-28 15:37 ` [Caml-list] " Olivier Andrieu
2009-10-28 16:05   ` Sylvain Le Gall
2009-10-28 15:43 ` [Caml-list] " Gerd Stolpmann
2009-10-28 16:06   ` Sylvain Le Gall
2009-10-28 18:09   ` [Caml-list] " Goswin von Brederlow
2009-10-28 17:09 ` Xavier Leroy
2009-10-28 19:05   ` Goswin von Brederlow
2009-10-29 17:05   ` Goswin von Brederlow
2009-10-29 18:42     ` Christophe TROESTLER
2009-10-29 19:03       ` Goswin von Brederlow
2009-10-29 18:48     ` Sylvain Le Gall
2009-10-29 23:25       ` [Caml-list] " Goswin von Brederlow
2009-11-03 17:16 Charles Forsyth

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=1256863735.24202.127.camel@flake.lan.gerd-stolpmann.de \
    --to=gerd@gerd-stolpmann.de \
    --cc=caml-list@inria.fr \
    --cc=fw@deneb.enyo.de \
    --cc=goswin-v-b@web.de \
    /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).