From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 22466 invoked by alias); 10 Jul 2008 19:08:54 -0000 Received: (qmail 22445 invoked by uid 22791); 10 Jul 2008 19:08:54 -0000 X-Spam-Check-By: sourceware.org Received: from NaN.false.org (HELO nan.false.org) (208.75.86.248) by sourceware.org (qpsmtpd/0.31) with ESMTP; Thu, 10 Jul 2008 19:08:32 +0000 Received: from nan.false.org (localhost [127.0.0.1]) by nan.false.org (Postfix) with ESMTP id 78FC898415; Thu, 10 Jul 2008 19:08:30 +0000 (GMT) Received: from caradoc.them.org (22.svnf5.xdsl.nauticom.net [209.195.183.55]) by nan.false.org (Postfix) with ESMTP id 6514898337; Thu, 10 Jul 2008 19:08:30 +0000 (GMT) Received: from drow by caradoc.them.org with local (Exim 4.69) (envelope-from ) id 1KH1VF-0001bz-53; Thu, 10 Jul 2008 15:08:29 -0400 Date: Thu, 10 Jul 2008 19:08:00 -0000 From: Daniel Jacobowitz To: Paul Koning Cc: pedro@codesourcery.com, gdb@sourceware.org, gdb-patches@sourceware.org Subject: Re: [remote protocol] support for disabling packet acknowledgement Message-ID: <20080710190828.GA6147@caradoc.them.org> Mail-Followup-To: Paul Koning , pedro@codesourcery.com, gdb@sourceware.org, gdb-patches@sourceware.org References: <200807101823.13728.pedro@codesourcery.com> <18550.22660.364998.466763@gargle.gargle.HOWL> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <18550.22660.364998.466763@gargle.gargle.HOWL> User-Agent: Mutt/1.5.17 (2008-05-11) X-IsSubscribed: yes Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2008-07/txt/msg00156.txt.bz2 On Thu, Jul 10, 2008 at 02:44:20PM -0400, Paul Koning wrote: > For one thing, if you want to work on performance, there are much more > dramatic changes to the protocol that could be done that would help > much more. I can't believe that the cost of acks is significant > compared to all the other bottlenecks. In addition to what Sandra said, the ack is generally in a separate TCP packet from the next response or request. That means there's double the number of TCP packets. Especially over high latency links, I think this will be a big help. I've had plenty of experience of cross-country remote protocol sessions and I like anything that helps. -- Daniel Jacobowitz CodeSourcery