From owner-freebsd-current Sat Nov 15 10:00:37 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id KAA21536 for current-outgoing; Sat, 15 Nov 1997 10:00:37 -0800 (PST) (envelope-from owner-freebsd-current) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id KAA21497 for ; Sat, 15 Nov 1997 10:00:21 -0800 (PST) (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.7/8.8.5) id NAA04220; Sat, 15 Nov 1997 13:00:16 -0500 (EST) From: "John S. Dyson" Message-Id: <199711151800.NAA04220@dyson.iquest.net> Subject: Re: bcopy via npx not in 686 version? In-Reply-To: from Evan Champion at "Nov 15, 97 11:42:17 am" To: evanc@synapse.net (Evan Champion) Date: Sat, 15 Nov 1997 13:00:16 -0500 (EST) Cc: freebsd-current@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Evan Champion said: > I was just looking through LINT and came upon the npx0 options to optimize > bcopy/bzero/copyin/copyout. These seem to be the only 586-specific > options in FreeBSD (Everything else is ifdef'd for both 586 and 686). > > I'm just curious if this was the desired behaviour, or if it was > accidently left out? > There isn't an advantage using the FPU for bcopys on P6 processors. In fact, movs[*] instructions are optimized on a P6. -- John dyson@freebsd.org jdyson@nc.com