From owner-cvs-src-old@FreeBSD.ORG Sat Oct 16 07:12:58 2010 Return-Path: Delivered-To: cvs-src-old@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 231591065675 for ; Sat, 16 Oct 2010 07:12:58 +0000 (UTC) (envelope-from lstewart@FreeBSD.org) Received: from repoman.freebsd.org (repoman.freebsd.org [IPv6:2001:4f8:fff6::29]) by mx1.freebsd.org (Postfix) with ESMTP id E51928FC1E for ; Sat, 16 Oct 2010 07:12:57 +0000 (UTC) Received: from repoman.freebsd.org (localhost [127.0.0.1]) by repoman.freebsd.org (8.14.4/8.14.4) with ESMTP id o9G7CvcY013067 for ; Sat, 16 Oct 2010 07:12:57 GMT (envelope-from lstewart@repoman.freebsd.org) Received: (from svn2cvs@localhost) by repoman.freebsd.org (8.14.4/8.14.4/Submit) id o9G7CvuG013066 for cvs-src-old@freebsd.org; Sat, 16 Oct 2010 07:12:57 GMT (envelope-from lstewart@repoman.freebsd.org) Message-Id: <201010160712.o9G7CvuG013066@repoman.freebsd.org> X-Authentication-Warning: repoman.freebsd.org: svn2cvs set sender to lstewart@repoman.freebsd.org using -f From: Lawrence Stewart Date: Sat, 16 Oct 2010 07:12:39 +0000 (UTC) To: cvs-src-old@freebsd.org X-FreeBSD-CVS-Branch: HEAD Subject: cvs commit: src/sys/netinet tcp_reass.c X-BeenThere: cvs-src-old@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: **OBSOLETE** CVS commit messages for the src tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 16 Oct 2010 07:12:58 -0000 lstewart 2010-10-16 07:12:39 UTC FreeBSD src repository Modified files: sys/netinet tcp_reass.c Log: SVN rev 213913 on 2010-10-16 07:12:39Z by lstewart Retire the system-wide, per-reassembly queue segment limit. The mechanism is far too coarse grained to be useful and the default value significantly degrades TCP performance on moderate to high bandwidth-delay product paths with non-zero loss (e.g. 5+Mbps connections across the public Internet often suffer). Replace the outgoing mechanism with an individual per-queue limit based on the number of MSS segments that fit into the socket's receive buffer. This should strike a good balance between performance and the potential for resource exhaustion when FreeBSD is acting as a TCP receiver. With socket buffer autotuning (which is enabled by default), the reassembly queue tracks the socket buffer and benefits too. As the XXX comment suggests, my testing uncovered some unexpected behaviour which requires further investigation. By using so->so_rcv.sb_hiwat instead of sbspace(&so->so_rcv), we allow more segments to be held across both the socket receive buffer and reassembly queue than we probably should. The tradeoff is better performance in at least one common scenario, versus a devious sender's ability to consume more resources on a FreeBSD receiver. Sponsored by: FreeBSD Foundation Reviewed by: andre, gnn, rpaulo MFC after: 2 weeks Revision Changes Path 1.369 +15 -11 src/sys/netinet/tcp_reass.c