From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 10 Aug 2001 20:00:23 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 10 Aug 2001 20:00:13 -0400 Received: from zero.aec.at ([195.3.98.22]:24075 "HELO zero.aec.at") by vger.kernel.org with SMTP id ; Fri, 10 Aug 2001 20:00:04 -0400 To: lk@tantalophile.demon.co.uk (Jamie Lokier) cc: linux-kernel@vger.kernel.org Subject: Re: Half Duplex and Zero Copy IP In-Reply-To: <20010810095313.A6219@home.com> <20010811005037.A5591@thefinal.cern.ch> From: Andi Kleen Date: 11 Aug 2001 02:00:14 +0200 In-Reply-To: lk@tantalophile.demon.co.uk's message of "Fri, 10 Aug 2001 23:52:38 +0000 (UTC)" Message-ID: User-Agent: Gnus/5.0700000000000003 (Pterodactyl Gnus v0.83) Emacs/20.2 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org In article <20010811005037.A5591@thefinal.cern.ch>, lk@tantalophile.demon.co.uk (Jamie Lokier) writes: > This means that you cannot use mmap() on a packet socket to zero-copy > read whole packets. In fact there is no way to zero-copy read whole > packets into user space, without modifying the kernel. [...] Yes, sorry, I should have written single copy. Zero copy read is currently only possible for kernel clients with hacks. -Andi