From owner-freebsd-current@FreeBSD.ORG Thu Jul 10 14:35:50 2003 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 13F8837B401 for ; Thu, 10 Jul 2003 14:35:50 -0700 (PDT) Received: from mail.speakeasy.net (mail13.speakeasy.net [216.254.0.213]) by mx1.FreeBSD.org (Postfix) with ESMTP id 703B843F93 for ; Thu, 10 Jul 2003 14:35:49 -0700 (PDT) (envelope-from jhb@FreeBSD.org) Received: (qmail 15035 invoked from network); 10 Jul 2003 21:35:48 -0000 Received: from unknown (HELO server.baldwin.cx) ([216.27.160.63]) (envelope-sender )encrypted SMTP for ; 10 Jul 2003 21:35:48 -0000 Received: from laptop.baldwin.cx (gw1.twc.weather.com [216.133.140.1]) by server.baldwin.cx (8.12.9/8.12.9) with ESMTP id h6ALZkGI002859; Thu, 10 Jul 2003 17:35:46 -0400 (EDT) (envelope-from jhb@FreeBSD.org) Message-ID: X-Mailer: XFMail 1.5.4 on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <20030711070631.C27297@gamplex.bde.org> Date: Thu, 10 Jul 2003 17:36:00 -0400 (EDT) From: John Baldwin To: Bruce Evans cc: FreeBSD current users cc: Julian Elischer Subject: Re: small scheduler hack/patch X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 10 Jul 2003 21:35:50 -0000 On 10-Jul-2003 Bruce Evans wrote: > On Thu, 10 Jul 2003, Julian Elischer wrote: > >> I have a small "proof of concept" scheduler hack at: >> http://www.freebsd.org/~julian/it.diff >> ... >> It's only implemented for SMP/i386 as the code to halt the cpu is only >> present (from my quick view) in x86 and it doesn't make sense in UP.. > > ipi_selected() is essentially MI although it is declared in > and takes MD arg types, since it is used in subr_smp.c > and subr_smp.c is MI by definition. Its second arg is u_int64_t on > alphas, int on ia64's, and u_int on other arches. This isn't a problem > since the arg is always a #define'd value like IPI_AST and these values > have MD definitions. Its first arg has type u_int64_t on ia64's and > u_int on other arches. This is bogus for ia64's since subr_smp.c uses > u_int for all bitmaps of CPUs, so systems with more than 32 CPUs cannot > actually work. I'm actually not a big fan of the CPU bitmap thing, but perhaps an MD cpu_bitmap typedef should be defined? -- John Baldwin <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/