From owner-freebsd-alpha Sat Jul 17 1:28:44 1999 Delivered-To: freebsd-alpha@freebsd.org Received: from herring.nlsystems.com (nlsys.demon.co.uk [158.152.125.33]) by hub.freebsd.org (Postfix) with ESMTP id 8BBFF14E17 for ; Sat, 17 Jul 1999 01:28:32 -0700 (PDT) (envelope-from dfr@nlsystems.com) Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id JAA14108; Sat, 17 Jul 1999 09:27:51 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Sat, 17 Jul 1999 09:27:51 +0100 (BST) From: Doug Rabson To: Hidetoshi Shimokawa Cc: jdp@polstra.com, alpha@freebsd.org Subject: Re: alpha/12623: strtod(3) FPE on alphaev56 In-Reply-To: <14223.8578.753480.80014H@ett.sat.t.u-tokyo.ac.jp> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 16 Jul 1999, Hidetoshi Shimokawa wrote: > At Fri, 16 Jul 1999 09:45:35 +0100 (BST), > Doug Rabson wrote: > > I agree. As long as there is a -mno-ieee option for those programs which > > need to squeeze the last drop of performance then this is the right thing > > to do. > > I agree too. > > However, I'm very suprised when I found -mieee has BIG performance > penatly on my system(21164/600MHz). > > A well-optimized program multiplying two 1000 * 1000 matrixes. > > % cc -mcpu=ev56 -O2 matrix2a.c > % time ./a.out > ./a.out 5.95s user 0.10s system 99% cpu 6.052 total > % cc -mieee -mcpu=ev56 -O2 matrix2a.c > % time ./a.out > ./a.out 21.86s user 0.08s system 99% cpu 21.976 total This is quite a severe penalty. However, such code is hopefully fairly rare and the developer can still build the program using -mno-ieee for maximum performance so perhaps it isn't a problem? I really must build my 3D code on the alpha soon and see what kind of penalty I get for the maths in that. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 442 9037 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message