From owner-freebsd-arch@FreeBSD.ORG Fri Oct 5 08:04:28 2012 Return-Path: Delivered-To: freebsd-arch@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 508401065670; Fri, 5 Oct 2012 08:04:28 +0000 (UTC) (envelope-from des@des.no) Received: from smtp.des.no (smtp.des.no [194.63.250.102]) by mx1.freebsd.org (Postfix) with ESMTP id 0D5F08FC14; Fri, 5 Oct 2012 08:04:27 +0000 (UTC) Received: from ds4.des.no (smtp.des.no [194.63.250.102]) by smtp.des.no (Postfix) with ESMTP id 50BE265D6; Fri, 5 Oct 2012 10:04:27 +0200 (CEST) Received: by ds4.des.no (Postfix, from userid 1001) id 26358812D; Fri, 5 Oct 2012 10:04:27 +0200 (CEST) From: =?utf-8?Q?Dag-Erling_Sm=C3=B8rgrav?= To: Garrett Cooper References: <506C385C.3020400@FreeBSD.org> <506DEB4C.5020508@andric.com> <86haq9hq2c.fsf@ds4.des.no> <20121005033244.GL35915@deviant.kiev.zoral.com.ua> <86y5jll7kc.fsf@ds4.des.no> Date: Fri, 05 Oct 2012 10:04:26 +0200 In-Reply-To: (Garrett Cooper's message of "Fri, 5 Oct 2012 00:40:51 -0700") Message-ID: <86txu9l4z9.fsf@ds4.des.no> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.4 (berkeley-unix) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Cc: Konstantin Belousov , Dimitry Andric , Andriy Gapon , freebsd-arch@freebsd.org Subject: Re: x86 boot code build X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 05 Oct 2012 08:04:28 -0000 Garrett Cooper writes: > I would target the appropriate architecture (amd64) where it matters > (amd64), and target the lowest sane common denominator on i386. In > reality, what does a couple MB mean on amd64 vs i386? 1) Nobody mentioned amd64 - this is about i386. 2) It's not a question of *size* but of *performance*. By targeting the least capable platform that our users are likely to encounter (pentium-mmx) instead of one which is virtually eradicated (486), we can use features that are available on the former but not the latter. Someone said they'd like to target SSE2, but that would leave many common embedded systems out in the cold. If we do that, we should provide two sets of binaries; one set for sse2-capable machines (which covers all i386 desktop and server machines made in the last ten years) and one set for pentium-mmx (which covers the soekris and other popular SFF / embedded systems). DES --=20 Dag-Erling Sm=C3=B8rgrav - des@des.no