From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 15168 invoked by alias); 21 Feb 2008 22:43:48 -0000 Received: (qmail 15134 invoked by uid 22791); 21 Feb 2008 22:43:45 -0000 X-Spam-Check-By: sourceware.org Received: from mtagate1.de.ibm.com (HELO mtagate1.de.ibm.com) (195.212.29.150) by sourceware.org (qpsmtpd/0.31) with ESMTP; Thu, 21 Feb 2008 22:43:10 +0000 Received: from d12nrmr1607.megacenter.de.ibm.com (d12nrmr1607.megacenter.de.ibm.com [9.149.167.49]) by mtagate1.de.ibm.com (8.13.8/8.13.8) with ESMTP id m1LMh7pZ097738 for ; Thu, 21 Feb 2008 22:43:07 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 m1LMh68t2068496 for ; Thu, 21 Feb 2008 23:43:06 +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 m1LMh6Mt011532 for ; Thu, 21 Feb 2008 23:43:06 +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 m1LMh6Hx011529 for ; Thu, 21 Feb 2008 23:43:06 +0100 In-Reply-To: Subject: Re: [PING][patch][modulo-sched] Fix order params calculation To: Revital1 Eres Cc: gcc-patches@gcc.gnu.org X-Mailer: Lotus Notes Release 7.0 HF277 June 21, 2006 Message-ID: From: Ayal Zaks Date: Thu, 21 Feb 2008 23:43:00 -0000 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII 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: 2008-02/txt/msg00945.txt.bz2 Revital1 Eres/Haifa/IBM wrote on 19/02/2008 09:49:31: > Hello, > > http://gcc.gnu.org/ml/gcc-patches/2008-01/msg01065.html > > Thanks, > Revital > Hello, > > The following patch contains a fix to the calculation of the order params > in case of latency zero. > > It is currently bootstrapped and tested on ppc and SPU. > > OK for mainline once testing completes? > OK. Please add the explanation/testcase why this fix is needed -- we want the ASAP/ALAP/HEIGHT parameters to maintain the direction of dependencies, including those having zero latency. Ayal. > :ADDPATCH modulo-sched: > > Thanks, > Revital > > 2008-01-23 Ayal Zaks > > * modulo-sched.c (calculate_order_params): Fix calculate of > order params. > > > Index: modulo-sched.c > =================================================================== > --- modulo-sched.c (revision 131717) > +++ modulo-sched.c (working copy) > @@ -2174,7 +2174,7 @@ > for (e = u_node->in; e; e = e->next_in) > if (e->distance == 0); > ASAP (u_node) = MAX (ASAP (u_node),) > - ASAP (e->src) + e->latency); > + ASAP (e->src) + e->latency + 1); > max_asap = MAX (max_asap, ASAP (u_node)); > } > > @@ -2188,9 +2188,9 @@ > if (e->distance == 0) > { > ALAP (u_node) = MIN (ALAP (u_node),; > - ALAP (e->dest) - e->latency);; > + ALAP (e->dest) - e->latency - 1); > HEIGHT (u_node) = MAX (HEIGHT (u_node), > - HEIGHT (e->dest) + e->latency); > + HEIGHT (e->dest) + e->latency + 1); > } > } > if (dump_file) > > > Thanks, > Revital