From owner-cvs-all@FreeBSD.ORG Wed Oct 1 01:28:43 2003 Return-Path: Delivered-To: cvs-all@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2BD6F16A4B3; Wed, 1 Oct 2003 01:28:43 -0700 (PDT) Received: from cs.rice.edu (cs.rice.edu [128.42.1.30]) by mx1.FreeBSD.org (Postfix) with ESMTP id 5507443FE9; Wed, 1 Oct 2003 01:28:42 -0700 (PDT) (envelope-from alc@cs.rice.edu) Received: from localhost (localhost [127.0.0.1]) by localhost.cs.rice.edu (Postfix) with ESMTP id 6D26E4AB43; Wed, 1 Oct 2003 03:28:41 -0500 (CDT) Received: from cs.rice.edu ([127.0.0.1]) by localhost (cs.rice.edu [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 08207-09; Wed, 1 Oct 2003 03:28:38 -0500 (CDT) Received: by cs.rice.edu (Postfix, from userid 19572) id B307E4AB72; Wed, 1 Oct 2003 03:28:34 -0500 (CDT) Date: Wed, 1 Oct 2003 03:28:34 -0500 From: Alan Cox To: Jeff Roberson Message-ID: <20031001082834.GT20747@cs.rice.edu> References: <200310010525.h915PRwl062742@repoman.freebsd.org> <20031001013745.K99666-100000@mail.chesapeake.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20031001013745.K99666-100000@mail.chesapeake.net> User-Agent: Mutt/1.3.28i X-Virus-Scanned: by amavis-20030314-p2 at cs.rice.edu cc: cvs-src@FreeBSD.org cc: Jeff Roberson cc: src-committers@FreeBSD.org cc: cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/i386/i386 support.s src/sys/i386/include md_var.h X-BeenThere: cvs-all@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: CVS commit messages for the entire tree List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 01 Oct 2003 08:28:43 -0000 On Wed, Oct 01, 2003 at 01:39:03AM -0400, Jeff Roberson wrote: > This moves data 3.8x faster than bzero on my laptop. In a moment I will > commit pmap bits to make use of it. This improved my 'make kernel' times > by almost 2%. It will do even better in workloads that demand a lot of > zero pages. > An sfence at the end of sse2_pagezero() is probably prudent. Alan > On Tue, 30 Sep 2003, Jeff Roberson wrote: > > > jeff 2003/09/30 22:25:27 PDT > > > > FreeBSD src repository > > > > Modified files: > > sys/i386/i386 support.s > > sys/i386/include md_var.h > > Log: > > - Add ss2_pagezero() for zeroing pages using the movnti instruction. This > > instruction is enabled with SSE2 but does not use SSE registers. It is a > > "non-temporal" move which bypasses the cache and does not dirty lines. > > > > Revision Changes Path > > 1.98 +14 -0 src/sys/i386/i386/support.s > > 1.65 +1 -0 src/sys/i386/include/md_var.h > >