From owner-cvs-all Wed Jul 17 22:26:11 2002 Delivered-To: cvs-all@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 44FC437B400; Wed, 17 Jul 2002 22:26:07 -0700 (PDT) Received: from apollo.backplane.com (apollo.backplane.com [216.240.41.2]) by mx1.FreeBSD.org (Postfix) with ESMTP id E7F1443E3B; Wed, 17 Jul 2002 22:26:06 -0700 (PDT) (envelope-from dillon@apollo.backplane.com) Received: from apollo.backplane.com (localhost [127.0.0.1]) by apollo.backplane.com (8.12.5/8.12.4) with ESMTP id g6I5Q6CV015938; Wed, 17 Jul 2002 22:26:06 -0700 (PDT) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.12.5/8.12.4/Submit) id g6I5Q6ND015937; Wed, 17 Jul 2002 22:26:06 -0700 (PDT) (envelope-from dillon) Date: Wed, 17 Jul 2002 22:26:06 -0700 (PDT) From: Matthew Dillon Message-Id: <200207180526.g6I5Q6ND015937@apollo.backplane.com> To: Luigi Rizzo Cc: Mike Silbersack , cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/sys/netinet tcp_timer.h References: <20020717220320.F83856-100000@patrocles.silby.com> <200207180445.g6I4jkD4006497@apollo.backplane.com> <20020717220630.A17698@iguana.icir.org> Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG :as someone mentioned, one big source of variability in RTT measurements :are delayed acks, which depend on the other side. Delayed acks should have no effect whatsoever. The transmitter should 'P'ush the last data block and this should cause the receiver to ack it immediately. In a continuous stream of data an ack is required to occur at least once every two packets. :I think that it is nice to be able to control the min. timeout, :but you should put it into a sysctl variable, and use a very :conservative value (200...1000)ms as default. I am working up a patch set to be able to set the minimum rtt and a slop value that is added after the rtt calculation (so it applies to everything, not just the minimum). I still think 200ms is too much. How about 50ms? :You can handle it in a way similar to sched_quantum, so you read/write :it in terms of fractions of second but the internal value is stored :in ticks, and you can apply some consistency checks on the values :passed by userland. What is the point of recomputing hz/50 all the times? : :just a side comment -- "polling issues" are resolved within 1 tick, but :1 tick on the sender side, so local HZ values do not count. : : cheers : luigi Right. There are already a number of TCP variables, like mss setting. I'll just use the same sysctl conversion format. I think it's in milliseconds. -Matt Matthew Dillon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message