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, 3 Aug 2001 13:23:02 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 3 Aug 2001 13:22:53 -0400 Received: from minus.inr.ac.ru ([193.233.7.97]:57357 "HELO ms2.inr.ac.ru") by vger.kernel.org with SMTP id ; Fri, 3 Aug 2001 13:22:34 -0400 Message-Id: <200108022231.CAA00375@mops.inr.ac.ru> Subject: Re: [PATCH] Inbound Connection Control mechanism: Prioritized Accept To: samudrala@us.ibm.com (Sridhar Samudrala) Date: Fri, 3 Aug 2001 02:31:51 +0400 (MSD) Cc: kuznet@ms2.inr.ac.ru, thiemo@sics.se, dmfreim@us.ibm.com, hadi@cybeus.ca, linux-kernel@vger.kernel.org, linux-net@vger.kernel.org, diffserv-general@lists.sourceforge.net, rusty@rustcorp.com.au In-Reply-To: from "Sridhar Samudrala" at Aug 2, 1 10:17:11 am From: Alexey Kuznetsov X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hello! > This looks like an elegant way of prioritizing without penalizing low priority > connections in the absence of high priority ones. > There may be an issue with sockets in accept queue which have received data. > Is it OK to move a socket which has already received some data back to SYN-RECV > state and expect the data to be resent? No, if you ACKed it. 500msec is maximum... Not so big number. But inside 500msec there are no problems with HTTP, where this ACK is better to maximally delay to piggyback to reply in any case. Alexey