From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner+w=401wt.eu-S1761275AbZANKTJ@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S1761275AbZANKTJ (ORCPT <rfc822;w@1wt.eu>);
	Wed, 14 Jan 2009 05:19:09 -0500
Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1760204AbZANKSu
	(ORCPT <rfc822;linux-kernel-outgoing>);
	Wed, 14 Jan 2009 05:18:50 -0500
Received: from gw1.cosmosbay.com ([86.65.150.130]:54281 "EHLO
	gw1.cosmosbay.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1752413AbZANKSt convert rfc822-to-8bit (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Wed, 14 Jan 2009 05:18:49 -0500
Message-ID: <496DBBD5.7080109@cosmosbay.com>
Date: Wed, 14 Jan 2009 11:17:57 +0100
From: Eric Dumazet <dada1@cosmosbay.com>
User-Agent: Thunderbird 2.0.0.19 (Windows/20081209)
MIME-Version: 1.0
To: Volker.Lendecke@SerNet.DE
CC: Andrew Morton <akpm@linux-foundation.org>, linux-kernel@vger.kernel.org,
       Steven French <sfrench@us.ibm.com>, Jens Axboe <jens.axboe@oracle.com>,
       netdev@vger.kernel.org
Subject: Re: maximum buffer size for splice(2) tcp->pipe?
References: <E1LKrp3-004Tub-IR@intern.SerNet.DE> <20090113123702.ad29cd13.akpm@linux-foundation.org> <496D2078.9080302@cosmosbay.com> <E1LN0MC-00DaJs-M2@intern.SerNet.DE> <496DACBE.3030701@cosmosbay.com> <E1LN2Vu-00DluN-6Z@intern.SerNet.DE>
In-Reply-To: <E1LN2Vu-00DluN-6Z@intern.SerNet.DE>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8BIT
X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-1.6 (gw1.cosmosbay.com [0.0.0.0]); Wed, 14 Jan 2009 11:18:01 +0100 (CET)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

Volker Lendecke a écrit :
> On Wed, Jan 14, 2009 at 10:13:34AM +0100, Eric Dumazet wrote:
>> for (;;) {
>> 	struct pollfd pfd;
>> 	pfd.fd = socket;
>> 	pfd.events = POLLIN;
>> 	if (poll(&pfd, 1, -1) != 1)
>> 		continue;
>> 	res = splice(socket, NULL, pipefds[1], NULL, 65536, SPLICE_F_MOVE|SPLICE_F_NONBLOCK);
>> 	if (res > 0)
>> 		nwritten = splice(pipefds[0], NULL, file_fd, NULL, res, SPLICE_F_MOVE|SPLICE_F_MORE);
>> }
> 
> Doesn't this reduce performance again? I thought the whole
> point of splice() was to increase performance by avoiding
> memory copies. If I have to do a poll syscall for each call
> to splice, doesn't the context switch eat that performance
> advantage again?
> 
> Or was splice designed only for multi-threaded applications
> (which at least Samba is not)?
> 
> Volker

splice() avoids memory copies yes, but on typical 1460 bytes
frames its a small gain.

But if no data is available on socket,
you still have to wait (and have a context switch later).

Waiting in poll() or splice() has same context switch cost.

Only cost is the extra syscall of course, but it is mandatory
if you want to avoid a possible deadlock in current splice()
implementation.