From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 30525 invoked by alias); 4 Nov 2011 09:23:40 -0000 Received: (qmail 30507 invoked by uid 22791); 4 Nov 2011 09:23:39 -0000 X-SWARE-Spam-Status: No, hits=-2.3 required=5.0 tests=AWL,BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FROM,RCVD_IN_DNSWL_LOW X-Spam-Check-By: sourceware.org Received: from mail-qw0-f47.google.com (HELO mail-qw0-f47.google.com) (209.85.216.47) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Fri, 04 Nov 2011 09:23:24 +0000 Received: by qam2 with SMTP id 2so2047587qam.20 for ; Fri, 04 Nov 2011 02:23:23 -0700 (PDT) MIME-Version: 1.0 Received: by 10.182.227.41 with SMTP id rx9mr2716379obc.12.1320398603691; Fri, 04 Nov 2011 02:23:23 -0700 (PDT) Received: by 10.182.17.232 with HTTP; Fri, 4 Nov 2011 02:23:23 -0700 (PDT) In-Reply-To: <201111040202.49532.mikael.morin@sfr.fr> References: <20111027232818.18581.901@gimli.local> <201111040202.49532.mikael.morin@sfr.fr> Date: Fri, 04 Nov 2011 09:39:00 -0000 Message-ID: Subject: Re: [Patch, fortran] [00/66] PR fortran/43829 Inline sum and product (AKA scalarization of reductions) From: Richard Guenther To: Mikael Morin Cc: fortran@gcc.gnu.org, Paul Richard Thomas , GCC patches Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-IsSubscribed: yes Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-patches-owner@gcc.gnu.org X-SW-Source: 2011-11/txt/msg00507.txt.bz2 On Fri, Nov 4, 2011 at 2:02 AM, Mikael Morin wrote: > On Tuesday 01 November 2011 22:07:48 Paul Richard Thomas wrote: >> The only, slight worry that I have is that it is going to make Richi's >> middle end scalarization nearly impossible to use for gfortran. >> However, the enhanced capability that this patch brings makes it a >> worthy addition to gfortran. >> > I think that Richi's middle-end arrays and gfortran's scalarizer are more= or > less incompatible, regardless of this patch. And if they are made to coex= ist > side by side at some point, this patch won't make it either better/easier= or > worse/harder IMHO. Indeed. In the ideal form the current scalarizer could defer scalarizing innermost loop (nests) to the middle-end, my hackish patches from earlier this year of course were just hacks (to eventually get some interest from you folks). But of course not even the middle-end parts are completely ready (partly also due to lack of good test coverage possibility - a chicken and egg problem :/). So I think improving the Fortran scalarizer is time that is very well spent. Thanks, Richard. > >> OK for trunk. >> > Thanks for the review! > Committed as follows. > > Mikael > > patch revision > =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D > 01 =A0 =A0180842 > 02 =A0 =A0180843 > 03 =A0 =A0180844 > 04 =A0 =A0180846 > 05 =A0 =A0180847 > 06 =A0 =A0180848 > 08 =A0 =A0180849 > 07 =A0 =A0180850 > 09 =A0 =A0180851 > 10 =A0 =A0180852 > 11 =A0 =A0180853 > 12 =A0 =A0180855 > 13 =A0 =A0180856 > 14 =A0 =A0180857 > 15 =A0 =A0180858 > 16 =A0 =A0180859 > 17 =A0 =A0180860 > 18 =A0 =A0180861 > 19 =A0 =A0180862 > 20 =A0 =A0180864 > 21 =A0 =A0180865 > 22 =A0 =A0180866 > 23 =A0 =A0180867 > 24 =A0 =A0180868 > 25 =A0 =A0180869 > 26 =A0 =A0180870 > 27 =A0 =A0180872 > 28 =A0 =A0180873 > 29 =A0 =A0180875 > 30 =A0 =A0180877 > 31 =A0 =A0180880 > 32 =A0 =A0180881 > 33 =A0 =A0180882 > 34 =A0 =A0180883 > 35 =A0 =A0180884 > 36 =A0 =A0180885 > 37 =A0 =A0180886 > 38 =A0 =A0180887 > 39 =A0 =A0180888 > 40 =A0 =A0180889 > 41 =A0 =A0180890 > 42 =A0 =A0180891 > 43 =A0 =A0480892 > 44 =A0 =A0180893 > 45 =A0 =A0180894 > 46 =A0 =A0180895 > 47 =A0 =A0180897 > 48 =A0 =A0180898 > 49 =A0 =A0180899 > 50 =A0 =A0180900 > 51 =A0 =A0180901 > 52 =A0 =A0180902 > 53 =A0 =A0180903 > 54 =A0 =A0180904 > 55 =A0 =A0180905 > 56 =A0 =A0180906 > 57 =A0 =A0180907 > 58 =A0 =A0180908 > 59 =A0 =A0180909 > 60 =A0 =A0180910 > 61 =A0 =A0180911 > 62 =A0 =A0180913 > 63 =A0 =A0180917 > 64 =A0 =A0180918 > 65 =A0 =A0180919 > 66 =A0 =A0180920 > tests 180922 >