From owner-freebsd-standards@FreeBSD.ORG  Fri Jun  4 19:54:12 2004
Return-Path: <owner-freebsd-standards@FreeBSD.ORG>
Delivered-To: freebsd-standards@freebsd.org
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP
	id 9073816A4CE; Fri,  4 Jun 2004 19:54:12 -0700 (PDT)
Received: from VARK.homeunix.com
	(ar59.lsanca2-4.27.98.47.lsanca2.dsl-verizon.net [4.27.98.47])
	by mx1.FreeBSD.org (Postfix) with ESMTP
	id 3C07843D31; Fri,  4 Jun 2004 19:54:12 -0700 (PDT)
	(envelope-from das@FreeBSD.ORG)
Received: from VARK.homeunix.com (localhost [127.0.0.1])
	by VARK.homeunix.com (8.12.11/8.12.10) with ESMTP id i552rx92004039;
	Fri, 4 Jun 2004 19:53:59 -0700 (PDT)
	(envelope-from das@FreeBSD.ORG)
Received: (from das@localhost)
	by VARK.homeunix.com (8.12.11/8.12.10/Submit) id i552rx2x004038;
	Fri, 4 Jun 2004 19:53:59 -0700 (PDT)
	(envelope-from das@FreeBSD.ORG)
Date: Fri, 4 Jun 2004 19:53:59 -0700
From: David Schultz <das@FreeBSD.ORG>
To: "Steven G. Kargl" <kargl@troutmask.apl.washington.edu>
Message-ID: <20040605025359.GA3084@VARK.homeunix.com>
Mail-Followup-To: "Steven G. Kargl" <kargl@troutmask.apl.washington.edu>,
	FreeBSD-gnats-submit@FreeBSD.ORG, freebsd-standards@FreeBSD.ORG
References: <200311291810.hATIAIWu084953@freefall.freebsd.org>
	<200312101711.hBAHBoEL007529@troutmask.apl.washington.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <200312101711.hBAHBoEL007529@troutmask.apl.washington.edu>
cc: FreeBSD-gnats-submit@FreeBSD.ORG
cc: freebsd-standards@FreeBSD.ORG
Subject: Re: standards/59797: Implement C99's round[f]() math fucntions
X-BeenThere: freebsd-standards@freebsd.org
X-Mailman-Version: 2.1.1
Precedence: list
List-Id: Standards compliance <freebsd-standards.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-standards>,
	<mailto:freebsd-standards-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/freebsd-standards>
List-Post: <mailto:freebsd-standards@freebsd.org>
List-Help: <mailto:freebsd-standards-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/freebsd-standards>,
	<mailto:freebsd-standards-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Sat, 05 Jun 2004 02:54:12 -0000

Sorry, I've put this off way too long.  The good news is that I'm
now going to do something about it.  The bad news is that I found
a significant bug in the proposed implementation.  Namely, round()
and roundf() often get the wrong answer for halfway cases.  In
IEEE-754 round-to-nearest mode, numbers that are halfway between
two representable numbers are supposed to be rounded to even.  For
instance, 0.5 becomes 0, and 1.5 becomes 2.  I don't see an easy
way to make this work in the present implementation without
fiddling with the underlying bits.  Perhaps we need to implement
it more similarly to fdlibm's rint().

BTW, benchmarking shows that using the sample implementation that
appears in the C99 standard results in a slowdown of two orders of
magnitude over your round() implementation and four orders of
magnitude over the x87 frndint instruction.  Just setting the
rounding mode and calling rint() also results in a significant
slowdown.  Thus, we definitely want something that's in the spirit
of what you wrote, but perhaps one that operates on the bits directly.