From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: zero copy for relay server Date: Tue, 29 Mar 2011 16:13:37 +0200 Message-ID: <1301408017.3161.154.camel@edumazet-laptop> References: <1301331138.3182.43.camel@edumazet-laptop> <1301337257.2506.8.camel@edumazet-laptop> <1301372590.2506.57.camel@edumazet-laptop> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Viral Mehta , "netdev@vger.kernel.org" To: Changli Gao Return-path: Received: from mail-bw0-f46.google.com ([209.85.214.46]:38702 "EHLO mail-bw0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753463Ab1C2ONm (ORCPT ); Tue, 29 Mar 2011 10:13:42 -0400 Received: by bwz15 with SMTP id 15so206688bwz.19 for ; Tue, 29 Mar 2011 07:13:41 -0700 (PDT) In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: Le mardi 29 mars 2011 =C3=A0 19:28 +0800, Changli Gao a =C3=A9crit : > It is a waste of fd using pipe buffer with two fds. In fact, I had > ever posted a patch, which extends pipe(2) to return a O_RDWR fd when > NULL is passed in. >=20 Oh well, one extra fd is less than 256 bytes. Adding a syscall means you force users to have latest kernel. All this is hypothetical, nobody gave performance numbers demonstrating splice()/splice() is slow enough we have to optimize it and add kernel bloat...