From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 25113 invoked by alias); 28 Oct 2009 19:02:37 -0000 Received: (qmail 25014 invoked by uid 22791); 28 Oct 2009 19:02:36 -0000 X-SWARE-Spam-Status: No, hits=-1.7 required=5.0 tests=AWL,BAYES_00,SARE_MSGID_LONG40,SPF_HELO_PASS,SPF_PASS X-Spam-Check-By: sourceware.org Received: from smtp-out.google.com (HELO smtp-out.google.com) (216.239.33.17) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Wed, 28 Oct 2009 19:02:29 +0000 Received: from wpaz24.hot.corp.google.com (wpaz24.hot.corp.google.com [172.24.198.88]) by smtp-out.google.com with ESMTP id n9SJ2Q6R025059 for ; Wed, 28 Oct 2009 19:02:26 GMT Received: from pxi6 (pxi6.prod.google.com [10.243.27.6]) by wpaz24.hot.corp.google.com with ESMTP id n9SJ2NWu016729 for ; Wed, 28 Oct 2009 12:02:24 -0700 Received: by pxi6 with SMTP id 6so736621pxi.0 for ; Wed, 28 Oct 2009 12:02:23 -0700 (PDT) MIME-Version: 1.0 Received: by 10.142.250.3 with SMTP id x3mr1552487wfh.23.1256756543424; Wed, 28 Oct 2009 12:02:23 -0700 (PDT) In-Reply-To: <4AE70C5E.4050005@starynkevitch.net> References: <4AE6E471.4020200@starynkevitch.net> <84fc9c000910270518s147e0cc4t7f1cd53049c5d590@mail.gmail.com> <4AE6E5D0.4010401@starynkevitch.net> <4AE7014A.1000903@starynkevitch.net> <4AE70C5E.4050005@starynkevitch.net> Date: Wed, 28 Oct 2009 20:32:00 -0000 Message-ID: <38a0d8450910281202u2d037020wf8a14cbdf969f780@mail.gmail.com> Subject: Re: plugin hooks From: Rafael Espindola To: Basile STARYNKEVITCH Cc: Ian Lance Taylor , Richard Guenther , GCC Mailing List , Grigori Fursin , Albert Cohen , Zbigniew Chamski Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable X-System-Of-Record: true X-IsSubscribed: yes Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org X-SW-Source: 2009-10/txt/msg00594.txt.bz2 > From several face to face discussions with ICI CTuning people (that it > Grigori Fursin, Albert Cohen, Zbignew Chamski; all 3 are in CC of this > reply) I understood that GCC ICI is exactly doing that: it has replaced G= CC > pass manager by its own, which invokes all the existing GCC passes in some > better order (or better number of invocation) and produces better generat= ed > code (the tuning is done by machine learning techniques; they did run the= ir > modified GCC a large number of times and did build a database of measures= .). > > I cannot explain more in details what GCC ICI is actually doing. I hope my > ICI friends will answer here. Perhaps even Zbignew's latest patches are > related to the answer. We talked a bit in the summit about it. I think that the correct solution is to make the gcc pass manager have the flexibility they need so that their plugin can just call into the regular pass manager. > Regards. > > > -- > Basile STARYNKEVITCH =C2=A0 =C2=A0 =C2=A0 =C2=A0 http://starynkevitch.net= /Basile/ > email: basilestarynkevitchnet mobile: +33 6 8501 2359 > 8, rue de la Faiencerie, 92340 Bourg La Reine, France > *** opinions {are only mines, sont seulement les miennes} *** > Cheers, --=20 Rafael =C3=81vila de Esp=C3=ADndola