From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 89557 invoked by alias); 16 Jul 2015 10:23:24 -0000 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 Received: (qmail 87170 invoked by uid 89); 16 Jul 2015 10:23:23 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-3.2 required=5.0 tests=BAYES_00,FREEMAIL_FROM,RP_MATCHES_RCVD,SPF_PASS autolearn=ham version=3.3.2 X-HELO: fencepost.gnu.org Received: from fencepost.gnu.org (HELO fencepost.gnu.org) (208.118.235.10) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES128-SHA encrypted) ESMTPS; Thu, 16 Jul 2015 10:23:13 +0000 Received: from eggs.gnu.org ([2001:4830:134:3::10]:44183) by fencepost.gnu.org with esmtps (TLS1.0:RSA_AES_256_CBC_SHA1:256) (Exim 4.82) (envelope-from ) id 1ZFgJq-0003kI-L4 for gcc-patches@gnu.org; Thu, 16 Jul 2015 06:23:10 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ZFgJo-0000P2-Ib for gcc-patches@gnu.org; Thu, 16 Jul 2015 06:23:09 -0400 Received: from mail-oi0-x22b.google.com ([2607:f8b0:4003:c06::22b]:33339) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ZFgJo-0000OH-8v for gcc-patches@gnu.org; Thu, 16 Jul 2015 06:23:08 -0400 Received: by oige126 with SMTP id e126so47285141oig.0 for ; Thu, 16 Jul 2015 03:23:07 -0700 (PDT) MIME-Version: 1.0 X-Received: by 10.202.49.212 with SMTP id x203mr7626055oix.81.1437042187607; Thu, 16 Jul 2015 03:23:07 -0700 (PDT) Received: by 10.76.115.167 with HTTP; Thu, 16 Jul 2015 03:23:07 -0700 (PDT) In-Reply-To: <87fv4owgjy.fsf@kepler.schwinge.homeip.net> References: <55A6C1DF.1050108@mentor.com> <87fv4owgjy.fsf@kepler.schwinge.homeip.net> Date: Thu, 16 Jul 2015 10:28:00 -0000 Message-ID: Subject: Re: [RFC, PR66873] Use graphite for parloops From: Richard Biener To: Thomas Schwinge Cc: Tom de Vries , "gcc-patches@gnu.org" Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable X-detected-operating-system: by eggs.gnu.org: Error: Malformed IPv6 address (bad octet value). X-Received-From: 2607:f8b0:4003:c06::22b X-IsSubscribed: yes X-SW-Source: 2015-07/txt/msg01378.txt.bz2 On Thu, Jul 16, 2015 at 12:19 PM, Thomas Schwinge wrote: > Hi Tom! > > On Thu, 16 Jul 2015 10:46:00 +0200, Richard Biener wrote: >> On Wed, Jul 15, 2015 at 10:26 PM, Tom de Vries = wrote: >> > I tried to parallelize this fortran test-case (based on autopar/outer-= 1.c), >> > [...] > >> > So I wondered, why not always use the graphite dependency analysis in >> > parloops. (Of course you could use -floop-parallelize-all, but that al= so >> > changes the heuristic). So I wrote a patch for parloops to use graphite >> > dependency analysis by default (so without -floop-parallelize-all), but >> > while testing found out that all the reduction test-cases started fail= ing >> > because the modifications graphite makes to the code messes up the par= loops >> > reduction analysis. >> > >> > Then I came up with this patch, which: >> > - first runs a parloops pass, restricted to reduction loops only, >> > - then runs graphite dependency analysis >> > - followed by a normal parloops pass run. >> > >> > This way, we get to both: >> > - compile the reduction testcases as before, and >> > - profit from the better graphite dependency analysis otherwise. > >> graphite dependence analysis is too slow to be enabled unconditionally. >> (read: hours in some simple cases - see bugzilla) > > Haha, "cool"! ;-) > > Maybe it is still reasonable to use graphite to analyze the code inside > OpenACC kernels regions -- maybe such code can reasonably be expected to > not have the properties that make its analysis lengthy? So, Tom, could > you please identify and check such PRs, to get an understanding of what > these properties are? Like the one in PR62113 or 53852 or 59121. > > Gr=C3=BC=C3=9Fe, > Thomas