From owner-freebsd-current@FreeBSD.ORG Tue Jun 1 14:31:16 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9029D16A4CE; Tue, 1 Jun 2004 14:31:16 -0700 (PDT) Received: from lakermmtao10.cox.net (lakermmtao10.cox.net [68.230.240.29]) by mx1.FreeBSD.org (Postfix) with ESMTP id 08A9343D2D; Tue, 1 Jun 2004 14:31:16 -0700 (PDT) (envelope-from conrads@cox.net) Received: from ip68-11-70-23.no.no.cox.net ([68.11.70.23]) by lakermmtao10.cox.netESMTP <20040601213114.QVHV15473.lakermmtao10.cox.net@ip68-11-70-23.no.no.cox.net>; Tue, 1 Jun 2004 17:31:14 -0400 Received: from ip68-11-70-23.no.no.cox.net (localhost.no.no.cox.net [127.0.0.1])i51LVEHT030212; Tue, 1 Jun 2004 16:31:14 -0500 (CDT) (envelope-from conrads@ip68-11-70-23.no.no.cox.net) Received: (from conrads@localhost)i51LV9xh030207; Tue, 1 Jun 2004 16:31:09 -0500 (CDT) (envelope-from conrads) Message-ID: X-Mailer: XFMail 1.5.5 on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <20040601063642.GD69572@cell.sick.ru> Date: Tue, 01 Jun 2004 16:31:09 -0500 (CDT) From: Conrad Sabatier To: Gleb Smirnoff cc: freebsd-current@freebsd.org cc: Bosko Milekic Subject: Re: [HEADS-UP] mbuma is in the tree X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: conrads@cox.net List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 01 Jun 2004 21:31:16 -0000 On 01-Jun-2004 Gleb Smirnoff wrote: > Bosko, > > On Mon, May 31, 2004 at 02:51:01PM -0700, Bosko Milekic wrote: > B> In order to avoid having to type everything again, I'll refer > B> to the commit log. PLEASE READ IT IN FULL: > B> > B> Bring in mbuma to replace mballoc. > B> > B> mbuma is an Mbuf & Cluster allocator built on top of a number of > B> extensions to the UMA framework, all included herein. > > Have you done any performance tests? How this new allocator affects > network performance? > > How stable is it? :) Yesterday I was planning to upgrade CURRENT on > my production router. Should I do it? I went ahead and upgraded as soon as I saw the announcement, mainly just for the new capability of using an unlimited mbufs setting. It seems to be working just fine. I've experienced no problems with it whatsoever so far. No negative impact on network performance. Nice work there, Bosko! (now, if we could just get the ACPI-related hangs at shutdown fixed) :-) -- Conrad Sabatier - "In Unix veritas"