From patchwork Sun Aug 22 14:06:42 2010 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Tobias Burnus X-Patchwork-Id: 62388 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Received: from sourceware.org (server1.sourceware.org [209.132.180.131]) by ozlabs.org (Postfix) with SMTP id E7C14B70D1 for ; Mon, 23 Aug 2010 00:06:56 +1000 (EST) Received: (qmail 10269 invoked by alias); 22 Aug 2010 14:06:52 -0000 Received: (qmail 10250 invoked by uid 22791); 22 Aug 2010 14:06:51 -0000 X-SWARE-Spam-Status: No, hits=-1.9 required=5.0 tests=AWL, BAYES_00, RCVD_IN_DNSWL_NONE X-Spam-Check-By: sourceware.org Received: from mx02.qsc.de (HELO mx02.qsc.de) (213.148.130.14) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Sun, 22 Aug 2010 14:06:45 +0000 Received: from [192.168.178.22] (port-92-204-8-164.dynamic.qsc.de [92.204.8.164]) by mx02.qsc.de (Postfix) with ESMTP id 7F3A51E43A; Sun, 22 Aug 2010 16:06:42 +0200 (CEST) Message-ID: <4C712EF2.4020404@net-b.de> Date: Sun, 22 Aug 2010 16:06:42 +0200 From: Tobias Burnus User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; de; rv:1.9.2.7) Gecko/20100714 SUSE/3.1.1 Thunderbird/3.1.1 MIME-Version: 1.0 To: gcc patches , gfortran Subject: Re: [Patch, Fortran, committed] PR 45367/36158 - Increase numeric tolerance in bessel_{6, 7}.f90 References: <4C70F41C.8050801@net-b.de> In-Reply-To: <4C70F41C.8050801@net-b.de> Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Unsubscribe: List-Archive: List-Post: List-Help: Sender: gcc-patches-owner@gcc.gnu.org Delivered-To: mailing list gcc-patches@gcc.gnu.org Tobias Burnus wrote: > I increased now the tolerance based on Dominique's values and cross > checked using -m32 on x86-64-linux. I hope it will now work on most > systems Dominique found additional failures for bessel_6.f90. Let's see how many systems still fail with the numbers below. I have still the hope that there are reasonable numbers which work on all systems with working jn/yn functions ... Committed as Rev. 163460 Tobias Index: gcc/testsuite/ChangeLog =================================================================== --- gcc/testsuite/ChangeLog (Revision 163458) +++ gcc/testsuite/ChangeLog (Arbeitskopie) @@ -1,6 +1,12 @@ 2010-08-22 Tobias Burnus + Dominique d'Humieres PR fortran/45367 + * gfortran.dg/bessel_6.f90: Further reduce required accuracy. + +2010-08-22 Tobias Burnus + + PR fortran/45367 * gfortran.dg/bessel_6.f90: Fix numeric tolerence. 2010-08-22 Tobias Burnus Index: gcc/testsuite/gfortran.dg/bessel_6.f90 =================================================================== --- gcc/testsuite/gfortran.dg/bessel_6.f90 (Revision 163458) +++ gcc/testsuite/gfortran.dg/bessel_6.f90 (Arbeitskopie) @@ -8,12 +8,12 @@ implicit none real,parameter :: values(*) = [0.0, 0.5, 1.0, 0.9, 1.8,2.0,3.0,4.0,4.25,8.0,34.53, 475.78] real,parameter :: myeps(size(values)) = epsilon(0.0) & - * [2, 7, 5, 6, 9, 12, 12, 7, 7, 8, 75, 6 ] + * [2, 7, 5, 6, 9, 12, 12, 7, 7, 8, 75, 15 ] ! The following is sufficient for me - the values above are a bit ! more tolerant ! * [0, 5, 3, 4, 6, 7, 7, 5, 5, 6, 66, 4 ] integer,parameter :: mymax(size(values)) = & - [100, 17, 23, 21, 27, 28, 32, 35, 36, 41, 49, 50 ] + [100, 17, 23, 21, 27, 28, 32, 35, 36, 41, 47, 37 ] integer, parameter :: Nmax = 100 real :: rec(0:Nmax), lib(0:Nmax) integer :: i