From owner-freebsd-net Fri Mar 26 6:56:48 1999 Delivered-To: freebsd-net@freebsd.org Received: from ausmail1.austin.ibm.com (ausmail1.austin.ibm.com [192.35.232.12]) by hub.freebsd.org (Postfix) with ESMTP id 70A9515508 for ; Fri, 26 Mar 1999 06:56:46 -0800 (PST) (envelope-from marquard@austin.ibm.com) Received: from netmail.austin.ibm.com (netmail.austin.ibm.com [9.53.250.98]) by ausmail1.austin.ibm.com (8.9.1/8.8.5) with ESMTP id IAA30138 for ; Fri, 26 Mar 1999 08:51:20 -0600 Received: from mojave.austin.ibm.com (mojave.austin.ibm.com [9.53.150.76]) by netmail.austin.ibm.com (8.8.5/8.8.5) with ESMTP id IAA56990 for ; Fri, 26 Mar 1999 08:56:27 -0600 Received: (from marquard@localhost) by mojave.austin.ibm.com (AIX4.3/UCB 8.8.8/8.7-client1.01) id IAA33876; Fri, 26 Mar 1999 08:56:25 -0600 To: freebsd-net@freebsd.org Subject: Re: mbuf clusters and socket send buffers (was Re: 3.1-STABLE dies on 40+ connects) References: <199903260054.QAA22060@biggusdiskus.flyingfox.com> From: Dave Marquardt Date: 26 Mar 1999 08:56:25 -0600 In-Reply-To: Jim Shankland's message of "Thu, 25 Mar 1999 16:54:16 -0800 (PST)" Message-ID: Lines: 15 X-Mailer: Gnus v5.6.2/Emacs 19.34 Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Jim Shankland writes: > A thought related to this discussion: does it make sense to allow the > send buffers to be larger than the peer's advertised window size? > In other words, why "preposition" those bytes in the kernel before > the peer has indicated a willingness to accept them? Yes, it absolutely makes sense. Let's say you send everything the receiver allows you to send. If your socket send buffer has unsent data in it, then when the receiver ACKs, TCP can start sending right away. If you don't have more data in the socket send buffer, you have to wake up the application and copy more data in from it. So, in other words, having a larger send buffer allows streaming. -Dave To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message