From: John Gardiner Myers <jgmyers@netscape.com>
To: Mark Mielke <mark@mark.mielke.cc>
Cc: Davide Libenzi <davidel@xmailserver.org>,
Benjamin LaHaise <bcrl@redhat.com>, Dan Kegel <dank@kegel.com>,
Shailabh Nagar <nagar@watson.ibm.com>,
linux-kernel <linux-kernel@vger.kernel.org>,
linux-aio <linux-aio@kvack.org>, Andrew Morton <akpm@digeo.com>,
David Miller <davem@redhat.com>,
Linus Torvalds <torvalds@transmeta.com>,
Stephen Tweedie <sct@redhat.com>
Subject: epoll (was Re: [PATCH] async poll for 2.5)
Date: Wed, 16 Oct 2002 16:48:32 -0700 [thread overview]
Message-ID: <3DADFAD0.1070108@netscape.com> (raw)
In-Reply-To: 20021016200620.GA9067@mark.mielke.cc
[-- Attachment #1: Type: text/plain, Size: 841 bytes --]
Mark Mielke wrote:
>Not to enter into any of the other discussions on this issue, I wouldn't
>usually do what you suggest above. [...] if I did I
>a recv() or read() of 2K, and I only received 1K, there is no reason why
>another system call should be invoked on the resource that likely will not
>have any data ready.
>
>
You're into the minutiae here. Sure, you can optimize the read() in
some cases, but Mr. Libenzi's example of a correct code scheme is no
better than mine when it comes to this.
In some situations, I've found that optimizing in the other direction is
useful. When a peer is feeding you data slowly you can increase
throughput by having the thread block on read for a couple of
milliseconds before going back to the pool. A large part of that was
that the particular event delivery subsystem was expensive.
[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/x-pkcs7-signature, Size: 3537 bytes --]
next prev parent reply other threads:[~2002-10-16 23:42 UTC|newest]
Thread overview: 138+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-10-14 22:36 [PATCH] async poll for 2.5 Shailabh Nagar
2002-10-14 22:54 ` John Myers
2002-10-15 15:05 ` Benjamin LaHaise
2002-10-15 17:06 ` Dan Kegel
2002-10-15 17:03 ` Benjamin LaHaise
2002-10-15 17:18 ` Dan Kegel
2002-10-16 2:11 ` Lincoln Dale
2002-10-15 18:09 ` Shailabh Nagar
2002-10-15 18:53 ` Dan Kegel
2002-10-15 18:57 ` Benjamin LaHaise
2002-10-15 20:25 ` John Gardiner Myers
2002-10-15 21:09 ` Dan Kegel
2002-10-15 21:50 ` John Myers
2002-10-15 22:33 ` Davide Libenzi
2002-10-15 22:56 ` John Gardiner Myers
2002-10-15 23:23 ` Davide Libenzi
2002-10-16 19:16 ` John Myers
2002-10-15 21:11 ` Davide Libenzi
2002-10-15 22:01 ` John Gardiner Myers
2002-10-15 22:27 ` Davide Libenzi
2002-10-15 22:36 ` John Gardiner Myers
2002-10-15 22:41 ` Benjamin LaHaise
2002-10-15 23:26 ` John Gardiner Myers
2002-10-15 23:05 ` Davide Libenzi
2002-10-15 23:33 ` John Gardiner Myers
2002-10-16 0:05 ` Davide Libenzi
2002-10-16 0:15 ` John Myers
2002-10-16 14:25 ` Davide Libenzi
2002-10-16 18:15 ` John Gardiner Myers
2002-10-16 19:20 ` Davide Libenzi
2002-10-16 23:31 ` epoll (was Re: [PATCH] async poll for 2.5) John Gardiner Myers
2002-10-16 23:51 ` Davide Libenzi
2002-10-17 18:06 ` John Gardiner Myers
2002-10-17 18:33 ` Davide Libenzi
2002-10-18 19:02 ` John Gardiner Myers
2002-10-18 19:52 ` Davide Libenzi
2002-10-19 0:55 ` John Myers
2002-10-19 5:40 ` Davide Libenzi
2002-10-19 6:59 ` Mark Mielke
2002-10-19 17:26 ` Davide Libenzi
2002-10-19 17:48 ` Dan Kegel
2002-10-19 18:52 ` Charles 'Buck' Krasic
2002-10-19 20:18 ` Charles 'Buck' Krasic
2002-10-19 21:08 ` Dan Kegel
2002-10-22 19:35 ` John Gardiner Myers
2002-10-22 20:06 ` Davide Libenzi
2002-10-22 21:54 ` Erich Nahum
2002-10-22 22:17 ` Dan Kegel
2002-10-22 22:25 ` Davide Libenzi
2002-10-18 21:01 ` Charles 'Buck' Krasic
2002-10-18 21:33 ` Davide Libenzi
2002-10-19 1:05 ` John Myers
2002-10-19 1:27 ` Tervel Atanassov
2002-10-19 18:52 ` John G. Myers
2002-10-19 4:07 ` Charles 'Buck' Krasic
2002-10-16 20:06 ` [PATCH] async poll for 2.5 Mark Mielke
2002-10-16 23:48 ` John Gardiner Myers [this message]
2002-10-17 0:23 ` epoll (was Re: [PATCH] async poll for 2.5) Davide Libenzi
2002-10-17 17:45 ` John Myers
2002-10-16 2:45 ` [PATCH] async poll for 2.5 Charles 'Buck' Krasic
2002-10-16 14:28 ` Davide Libenzi
2002-10-17 18:47 ` Charles 'Buck' Krasic
2002-10-17 19:20 ` Davide Libenzi
2002-10-18 3:30 ` Dan Kegel
2002-10-16 18:29 ` John Gardiner Myers
2002-10-16 20:39 ` Charles 'Buck' Krasic
2002-10-17 17:59 ` epoll (was Re: [PATCH] async poll for 2.5) John Gardiner Myers
2002-10-21 16:58 ` [PATCH] async poll for 2.5 Alan Cox
2002-10-21 16:50 ` Benjamin LaHaise
2002-10-16 19:59 ` Dan Kegel
2002-10-16 20:03 ` Dan Kegel
2002-10-17 17:43 ` epoll (was Re: [PATCH] async poll for 2.5) John Myers
2002-10-18 17:00 ` Mark Mielke
2002-10-18 17:28 ` Dan Kegel
2002-10-18 17:41 ` Davide Libenzi
2002-10-18 18:55 ` Mark Mielke
2002-10-18 19:16 ` Davide Libenzi
2002-10-19 6:56 ` Mark Mielke
2002-10-19 16:10 ` Charles 'Buck' Krasic
2002-10-22 17:22 ` Mark Mielke
2002-10-22 17:46 ` Dan Kegel
2002-10-22 17:47 ` Davide Libenzi
2002-10-22 18:13 ` Alan Cox
2002-10-22 18:18 ` Davide Libenzi
2002-10-22 18:37 ` Benjamin LaHaise
2002-10-22 19:49 ` Davide Libenzi
2002-10-22 19:22 ` John Gardiner Myers
2002-10-22 19:28 ` Benjamin LaHaise
2002-10-22 19:50 ` John Gardiner Myers
2002-10-22 20:00 ` Benjamin LaHaise
2002-10-23 11:10 ` Latest aio code " Suparna Bhattacharya
2002-10-22 20:23 ` async poll John Myers
2002-10-22 18:42 ` epoll (was Re: [PATCH] async poll for 2.5) Charles 'Buck' Krasic
2002-10-22 19:35 ` Davide Libenzi
2002-10-23 16:49 ` Dan Kegel
2002-10-23 17:39 ` Benjamin LaHaise
2002-10-23 18:47 ` Davide Libenzi
2002-10-23 21:18 ` Benjamin LaHaise
2002-10-23 21:35 ` Davide Libenzi
2002-10-23 21:39 ` John Gardiner Myers
2002-10-23 21:54 ` Davide Libenzi
2002-10-23 17:49 ` Charles 'Buck' Krasic
2002-10-23 18:14 ` Davide Libenzi
2002-10-23 18:32 ` Charles 'Buck' Krasic
2002-10-23 20:36 ` async poll John Myers
2002-10-23 20:57 ` Dan Kegel
2002-10-23 21:13 ` Charles 'Buck' Krasic
2002-10-23 21:23 ` John Gardiner Myers
2002-10-23 21:51 ` Davide Libenzi
2002-10-23 21:51 ` bert hubert
2002-10-23 22:10 ` Davide Libenzi
2002-10-23 21:54 ` John Gardiner Myers
2002-10-23 22:22 ` Davide Libenzi
2002-10-23 22:29 ` John Gardiner Myers
2002-10-23 22:50 ` Davide Libenzi
2002-10-24 7:32 ` Eduardo Pérez
2002-10-24 15:05 ` Charles 'Buck' Krasic
2002-10-23 22:24 ` Dan Kegel
2002-10-23 22:30 ` Davide Libenzi
2002-10-23 22:53 ` Davide Libenzi
2002-10-19 17:19 ` epoll (was Re: [PATCH] async poll for 2.5) Davide Libenzi
2002-10-18 18:55 ` Chris Friesen
2002-10-18 19:00 ` Mark Mielke
2002-10-15 17:38 ` [PATCH] async poll for 2.5 Shailabh Nagar
2002-10-15 17:50 ` Benjamin LaHaise
2002-10-15 18:16 ` Davide Libenzi
2002-10-15 18:18 ` Shailabh Nagar
2002-10-15 19:00 ` Davide Libenzi
2002-10-15 19:02 ` Benjamin LaHaise
2002-10-15 18:59 ` Shailabh Nagar
2002-10-15 19:16 ` Davide Libenzi
2002-10-15 19:12 ` Benjamin LaHaise
2002-10-15 19:31 ` Davide Libenzi
2002-10-15 19:38 ` Dan Kegel
2002-10-15 19:55 ` Davide Libenzi
2002-10-15 20:36 ` John Gardiner Myers
2002-10-15 20:39 ` Benjamin LaHaise
2002-10-15 19:02 ` Davide Libenzi
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=3DADFAD0.1070108@netscape.com \
--to=jgmyers@netscape.com \
--cc=akpm@digeo.com \
--cc=bcrl@redhat.com \
--cc=dank@kegel.com \
--cc=davem@redhat.com \
--cc=davidel@xmailserver.org \
--cc=linux-aio@kvack.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mark@mark.mielke.cc \
--cc=nagar@watson.ibm.com \
--cc=sct@redhat.com \
--cc=torvalds@transmeta.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.