From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-patches-return-401702-listarch-gcc-patches=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 97980 invoked by alias); 30 Jun 2015 16:04:08 -0000
Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-patches.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-patches/>
List-Post: <mailto:gcc-patches@gcc.gnu.org>
List-Help: <mailto:gcc-patches-help@gcc.gnu.org>
Sender: gcc-patches-owner@gcc.gnu.org
Received: (qmail 97970 invoked by uid 89); 30 Jun 2015 16:04:08 -0000
Authentication-Results: sourceware.org; auth=none
X-Virus-Found: No
X-Spam-SWARE-Status: No, score=-1.8 required=5.0 tests=AWL,BAYES_00,SPF_PASS autolearn=ham version=3.3.2
X-HELO: eu-smtp-delivery-143.mimecast.com
Received: from eu-smtp-delivery-143.mimecast.com (HELO eu-smtp-delivery-143.mimecast.com) (146.101.78.143) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with ESMTP; Tue, 30 Jun 2015 16:04:06 +0000
Received: from cam-owa2.Emea.Arm.com (fw-tnat.cambridge.arm.com [217.140.96.140]) by eu-smtp-1.mimecast.com with ESMTP id uk-mta-35-jvpqjXoxR-2zVVIjZt9UZg-1
Received: from [10.2.207.65] ([10.1.2.79]) by cam-owa2.Emea.Arm.com with Microsoft SMTPSVC(6.0.3790.3959);	 Tue, 30 Jun 2015 17:04:01 +0100
Message-ID: <5592BDF1.7000207@arm.com>
Date: Tue, 30 Jun 2015 16:10:00 -0000
From: Alan Lawrence <alan.lawrence@arm.com>
User-Agent: Thunderbird 2.0.0.24 (X11/20101213)
MIME-Version: 1.0
To: Jeff Law <law@redhat.com>
CC: "gcc-patches@gcc.gnu.org" <gcc-patches@gcc.gnu.org>,  Richard Biener <richard.guenther@gmail.com>
Subject: Re: [PATCH/RFC] Make loop-header-copying more aggressive, rerun before tree-if-conversion
References: <555F4E54.6080600@arm.com> <5565E839.2090402@redhat.com> <55845395.3020306@arm.com> <5589BFF8.5050200@redhat.com>
In-Reply-To: <5589BFF8.5050200@redhat.com>
X-MC-Unique: jvpqjXoxR-2zVVIjZt9UZg-1
Content-Type: text/plain; charset=WINDOWS-1252; format=flowed
Content-Transfer-Encoding: quoted-printable
X-IsSubscribed: yes
X-SW-Source: 2015-06/txt/msg02237.txt.bz2

Jeff Law wrote:
> Thanks.  Does running the phi-only propagator after the loop header=20
> copying help?  At first glance it would seem that it ought to propagate=20
> the values of those degenerate PHIs then eliminate those PHIs.
>=20
> It was written to cleanup after jump threading which has a tendency to=20
> create very similar code to what you've shown below and to do so very=20
> quickly.

Thanks for the tip - this fixes up some examples, but not at all. Other exa=
mples=20
require also a call to rewrite_into_loop_closed_ssa and recomputing=20
dominators...maybe I can get everything to work with all of those, but my=20
feeling is to keep it as a pass: if the first pass_ch justifies being a pas=
s in=20
its own right, then surely a *more aggressive* version of that, does too...