From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH] sendfile() and UDP socket Date: Sun, 21 Sep 2008 21:27:24 -0700 (PDT) Message-ID: <20080921.212724.193719407.davem@davemloft.net> References: <20080922010834.GB3059@2ka.mipt.ru> <20080921.190715.144340989.davem@davemloft.net> <20080922041929.GA20165@2ka.mipt.ru> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: johaahn@gmail.com, netdev@vger.kernel.org To: johnpol@2ka.mipt.ru Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:33350 "EHLO sunset.davemloft.net" rhost-flags-OK-FAIL-OK-OK) by vger.kernel.org with ESMTP id S1750899AbYIVE1g (ORCPT ); Mon, 22 Sep 2008 00:27:36 -0400 In-Reply-To: <20080922041929.GA20165@2ka.mipt.ru> Sender: netdev-owner@vger.kernel.org List-ID: From: Evgeniy Polyakov Date: Mon, 22 Sep 2008 08:19:29 +0400 > Stream socket means that whatever data we put to it, it will become > completely boundary-free, in that regard, that receiving side will not > be able to get original sending packet sizes (without too much efforts). Right. > Datagram just preservs the boundaries and that's what we have with this > patch. Not exactly. Each and every send() operation on a datagram socket must result in exactly one packet. sendfile() was following this rule, when it returned successfully only one single packet was emitted. The new sendfile() semantics are outside of this model.