From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 14035 invoked by alias); 21 Nov 2007 15:53:12 -0000 Received: (qmail 14027 invoked by uid 22791); 21 Nov 2007 15:53:11 -0000 X-Spam-Check-By: sourceware.org Received: from mtagate8.de.ibm.com (HELO mtagate8.de.ibm.com) (195.212.29.157) by sourceware.org (qpsmtpd/0.31) with ESMTP; Wed, 21 Nov 2007 15:53:04 +0000 Received: from d12nrmr1607.megacenter.de.ibm.com (d12nrmr1607.megacenter.de.ibm.com [9.149.167.49]) by mtagate8.de.ibm.com (8.13.8/8.13.8) with ESMTP id lALFr1Bk387616 for ; Wed, 21 Nov 2007 15:53:01 GMT Received: from d12av04.megacenter.de.ibm.com (d12av04.megacenter.de.ibm.com [9.149.165.229]) by d12nrmr1607.megacenter.de.ibm.com (8.13.8/8.13.8/NCO v8.7) with ESMTP id lALFr1Ld2924704 for ; Wed, 21 Nov 2007 16:53:01 +0100 Received: from d12av04.megacenter.de.ibm.com (loopback [127.0.0.1]) by d12av04.megacenter.de.ibm.com (8.12.11.20060308/8.13.3) with ESMTP id lALFr17A011985 for ; Wed, 21 Nov 2007 16:53:01 +0100 Received: from d12mc102.megacenter.de.ibm.com (d12mc102.megacenter.de.ibm.com [9.149.167.114]) by d12av04.megacenter.de.ibm.com (8.12.11.20060308/8.12.11) with ESMTP id lALFr1UU011978 for ; Wed, 21 Nov 2007 16:53:01 +0100 In-Reply-To: Subject: Re: [PATCH 1/2][Modulo-sched] Fix the direction of the scheduling window To: Ayal Zaks Cc: Andrey Belevantsev , "Alexander Monakov" , gcc-patches@gcc.gnu.org X-Mailer: Lotus Notes Release 7.0 HF277 June 21, 2006 Message-ID: From: Revital1 Eres Date: Wed, 21 Nov 2007 19:02:00 -0000 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII 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: 2007-11/txt/msg01112.txt.bz2 Message-ID: <20071121190200.zksDFCgBKIV5-OsSx2b406toyU6VehzMcuuI1pr5Jx8@z> Ayal Zaks/Haifa/IBM wrote on 21/11/2007 17:37:20: > Revital1 Eres/Haifa/IBM wrote on 20/11/2007 15:47:59: > > > Hello, > > > > When determining the scheduling window of a node all types of edges are > > taken into account. Currently when a node has both predecessors and > > successors it will be set close to it's predecessors. In fact it can > > be that a node has only one successor with true dep edge but it will be > > scheduled close to it's allegedly predecessor; because of the anti-dep > > edge between the two nodes. Trying to avoid this confusion and reduce the > > life range of registers we choose to set the scheduled node close to it's > > predecessors or close to it's successors based on true deps edges only. > > > > This change makes SMS succeed on the attached testcase with -funroll-loops > > on SPU. The testcase was provided by Vladimir. > > > > Bootstrapped and tested together with the follow on patch (2/2) on ppc, > > SPU and x86 with no new regressions. > > > > :ADDPATCH modulo-sched: > > > > OK for mainline? > > Yes, with minor comments below. Thanks, I'll wait for the approval of patch 2/2 before committing the fixed version; as patch 2/2 fixes a bug exposed when testing this patch. Revital