From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from pigeon.gentoo.org ([208.92.234.80] helo=lists.gentoo.org) by finch.gentoo.org with esmtp (Exim 4.60) (envelope-from ) id 1Qao66-0006bs-Or for garchives@archives.gentoo.org; Sun, 26 Jun 2011 12:05:54 +0000 Received: from pigeon.gentoo.org (localhost [127.0.0.1]) by pigeon.gentoo.org (Postfix) with SMTP id 5FA8E1C1BE for ; Sun, 26 Jun 2011 12:05:54 +0000 (UTC) Received: from eagle.jhcloos.com (eagle.jhcloos.com [207.210.242.212]) by pigeon.gentoo.org (Postfix) with ESMTP id 136911C15B for ; Sun, 26 Jun 2011 11:36:14 +0000 (UTC) Received: by eagle.jhcloos.com (Postfix, from userid 10) id 659F0400B4; Sun, 26 Jun 2011 11:35:49 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=jhcloos.com; s=eagle; t=1309088173; bh=tQUGgrw7/dsatXKcY4FfNAKT5JxT0S/byJyx1NUyExY=; h=From:To:Subject:In-Reply-To:References:Date:Message-ID: MIME-Version:Content-Type; b=TOOfbDr9JbnufP37NYZ2VSACEL38huRba5BiSQEA1pQp3wgW31jx+aAs+ELg89JUk aOL5IME2NFMjlHpWCiXl1huXh26AKdV5vDtKapxztMScbgJeMHaE7SxbAN4vX6XdFL zmMzsLJN3s2N6lKtGPNrmx2QVZ5B+Brnz8zaJriE= Received: by carbon.jhcloos.org (Postfix, from userid 500) id 2B400260042; Sun, 26 Jun 2011 11:33:52 +0000 (UTC) From: James Cloos To: gentoo-amd64@lists.gentoo.org Subject: Re: [gentoo-amd64] Possible Math Problem - Request For Verification [Solved] In-Reply-To: <20110626003948.bd4b0b83.frank.peters@comcast.net> (Frank Peters's message of "Sun, 26 Jun 2011 00:39:48 -0400") References: <20110625134111.5c6b08dd.frank.peters@comcast.net> <20110625234418.a66f8d7b.frank.peters@comcast.net> <20110626003948.bd4b0b83.frank.peters@comcast.net> User-Agent: Gnus/5.110018 (No Gnus v0.18) Emacs/24.0.50 (gnu/linux) Face: iVBORw0KGgoAAAANSUhEUgAAABAAAAAQAgMAAABinRfyAAAACVBMVEX///8ZGXBQKKnCrDQ3 AAAAJElEQVQImWNgQAAXzwQg4SKASgAlXIEEiwsSIYBEcLaAtMEAADJnB+kKcKioAAAAAElFTkSu QmCC Copyright: Copyright 2011 James Cloos OpenPGP: ED7DAEA6; url=http://jhcloos.com/public_key/0xED7DAEA6.asc OpenPGP-Fingerprint: E9E9 F828 61A4 6EA9 0F2B 63E7 997A 9F17 ED7D AEA6 Date: Sun, 26 Jun 2011 07:33:52 -0400 Message-ID: Precedence: bulk List-Post: List-Help: List-Unsubscribe: List-Subscribe: List-Id: Gentoo Linux mail X-BeenThere: gentoo-amd64@lists.gentoo.org Reply-to: gentoo-amd64@lists.gentoo.org MIME-Version: 1.0 Content-Type: text/plain X-Hashcash: 1:30:110626:gentoo-amd64@lists.gentoo.org::tUz1PhIbdw4On53F:0000000000000000000000000000000d26cX X-Archives-Salt: X-Archives-Hash: d63ce034950942edeedb0fc7cf00e127 >>>>> "FP" == Frank Peters writes: FP> All I need now is for someone to come forward and say that he is using FP> gcc-4.5.2 and has no problem. Then I'll really start to bang my head. Using gcc-4.6.0 (compiled with +graphite), using: -O3 -ggdb3 -march=amdfam10 -ftree-vectorize a number of the cabs(3) and cpow(3) tests returned (0,0) (for each of float and double, and each of c and fortran) instead of the correct results. The sin and cos tests had max errors in the +- 5 ulp range for doubles and the +- 1.5 range for floats. Your obsercation that graphite may be to blame is likely correct. You may be interested in crlibm (correctly rounded lib m): http://lipforge.ens-lyon.fr/projects/crlibm http://lipforge.ens-lyon.fr/www/crlibm/ -JimC -- James Cloos OpenPGP: 1024D/ED7DAEA6