From owner-freebsd-alpha@FreeBSD.ORG Fri Jan 14 08:53:46 2005 Return-Path: Delivered-To: freebsd-alpha@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 88F2016A4DC for ; Fri, 14 Jan 2005 08:53:46 +0000 (GMT) Received: from srv1.cosmo-project.de (srv1.cosmo-project.de [213.83.6.106]) by mx1.FreeBSD.org (Postfix) with ESMTP id E1AA143D54 for ; Fri, 14 Jan 2005 08:53:45 +0000 (GMT) (envelope-from ticso@cicely12.cicely.de) Received: from cicely5.cicely.de (cicely5.cicely.de [10.1.1.7]) (authenticated bits=0)j0E8rgHo063483 (version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=OK); Fri, 14 Jan 2005 09:53:43 +0100 (CET) (envelope-from ticso@cicely12.cicely.de) Received: from cicely12.cicely.de (cicely12.cicely.de [IPv6:3ffe:400:8d0:301::12]) by cicely5.cicely.de (8.12.10/8.12.10) with ESMTP id j0E8rOU3000968 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Fri, 14 Jan 2005 09:53:25 +0100 (CET) (envelope-from ticso@cicely12.cicely.de) Received: from cicely12.cicely.de (localhost [127.0.0.1]) by cicely12.cicely.de (8.12.11/8.12.11) with ESMTP id j0E8rOlW043556; Fri, 14 Jan 2005 09:53:24 +0100 (CET) (envelope-from ticso@cicely12.cicely.de) Received: (from ticso@localhost) by cicely12.cicely.de (8.12.11/8.12.11/Submit) id j0E8rOtP043555; Fri, 14 Jan 2005 09:53:24 +0100 (CET) (envelope-from ticso) Date: Fri, 14 Jan 2005 09:53:24 +0100 From: Bernd Walter To: Wilko Bulte Message-ID: <20050114085323.GB43380@cicely12.cicely.de> References: <16866.32790.398095.651691@canoe.dclg.ca> <20050114002651.GD29445@dragon.nuxi.com> <20050114074336.GA97061@freebie.xs4all.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20050114074336.GA97061@freebie.xs4all.nl> X-Operating-System: FreeBSD cicely12.cicely.de 5.2-CURRENT alpha User-Agent: Mutt/1.5.6i X-Spam-Status: No, hits=-4.9 required=3.0 tests=BAYES_00 autolearn=ham version=2.64 X-Spam-Report: * -4.9 BAYES_00 BODY: Bayesian spam probability is 0 to 1% * [score: 0.0000] X-Spam-Checker-Version: SpamAssassin 2.64 (2004-01-11) on cicely12.cicely.de cc: freebsd-alpha@freebsd.org Subject: Re: processor type. X-BeenThere: freebsd-alpha@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: ticso@cicely.de List-Id: Porting FreeBSD to the Alpha List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 14 Jan 2005 08:53:46 -0000 On Fri, Jan 14, 2005 at 08:43:37AM +0100, Wilko Bulte wrote: > On Thu, Jan 13, 2005 at 04:26:51PM -0800, David O'Brien wrote.. > > On Mon, Jan 10, 2005 at 08:16:06AM -0500, David Gilbert wrote: > > > I see in the compiler lines crawling by that gcc is asked to optimize > > > for 'EV5' while being compatible with 'EV4'. My Alpha is an EV4 --- > > > I'm wondering if I would see better performance with a different flag > > > there, but the gcc manual doesn't even acknowledge the existance of > > > the options that are in use, let alone the available options. > > > > It doens't??? > > > > >From /usr/share/mk/bsd.cpu.mk: > > . elif ${MACHINE_ARCH} == "alpha" > > _CPUCFLAGS = -mcpu=ev4 -mtune=ev5 > > > > http://gcc.gnu.org/onlinedocs/gcc-3.4.3/gcc/DEC-Alpha-Options.html#DEC-Alpha-Options > > clearly tells: > > > > -mcpu=cpu_type > > Set the instruction set and instruction scheduling parameters for > > machine type cpu_type. You can specify either the EV style name or > > the corresponding chip number. GCC supports scheduling parameters for > > the EV4, EV5 and EV6 family of processors. > > > > -mtune=cpu_type > > Set only the instruction scheduling parameters for machine type > > cpu_type. The instruction set is not changed. > > > > > > That said, you should remove the -mtune=ev5 or change it to -mtune=ev4. > > My guess is you won't notice a difference, but this this change will > > produce the best code for your machine. > > Any idea how drastic -mtune=ev6 would change things on a DS10 (to name > a random box ;-) ? Theoretically -mtune=ev6 could make a noticeable difference, but using BWX gets you way more than mtune. It could make thing slower on older systems and you still want to do a complete recompile on ev56 and up systems. I don't see any real win for ev6 users. > Am I correct in assuming that -mcpu=ev4 will result in not using the BWX > etc of the newer CPUs? And that -mcpu=ev6 will result in code that does > not run on anything older than EV6 due to missing instructions etc? ev56 is the first BWX implementation and pca56 brings some multimedia related instructions - don't think that ev6 has new instructions. BWX instructions are emulated for userland tools, so ev56 compiled applications should work on any alpha, but ev6 may not. However the kernel and bootcode strictly needs to stay with common instructions and emulation of BWX is naturaly much slower than running code without them. -- B.Walter BWCT http://www.bwct.de bernd@bwct.de info@bwct.de