From owner-freebsd-hackers Fri Dec 28 17:12:35 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from mass.dis.org (mass.dis.org [216.240.45.41]) by hub.freebsd.org (Postfix) with ESMTP id 842AB37B41E for ; Fri, 28 Dec 2001 17:12:27 -0800 (PST) Received: from mass.dis.org (localhost [127.0.0.1]) by mass.dis.org (8.11.6/8.11.3) with ESMTP id fBT1K8F16931; Fri, 28 Dec 2001 17:20:08 -0800 (PST) (envelope-from msmith@mass.dis.org) Message-Id: <200112290120.fBT1K8F16931@mass.dis.org> To: Duraid Madina Cc: hackers@freebsd.org, msmith@mass.dis.org Subject: Re: SSE bcopy etc. In-Reply-To: Message from Duraid Madina of "Sat, 29 Dec 2001 11:36:31 +1100." <3C2D100F.80303@fl.net.au> Date: Fri, 28 Dec 2001 17:20:08 -0800 From: Mike Smith Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG > Has anyone considered adding Pentium 2/3/4/Athlon/Athlon XP support > to the low level string/bytecopy routines? If we just supported SSE (1) > that'd get us (okay, me) a pretty nice performance boost on the P2, P3, > P4 and Athlon XP, all in one hit. These days, the "average" new computer > supports SSE, and since the GCC 3.x/binutils folk are on their way to > supporting this stuff in a Standard way (e.g. '-msse', '-march=athlon' > etc.) maybe it's time to code something up? This is less of a win than it seems, since it means that you have to save/restore FPU context if the current process is using the FPU. These saves are expensive, and mean that the SSE-enhanced copy is only faster for large-ish copies. Page-sized copies and zeroes would be improved substantially. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message