From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 1874 invoked by alias); 6 Feb 2010 00:52:15 -0000 Received: (qmail 1614 invoked by alias); 6 Feb 2010 00:51:52 -0000 Date: Sat, 06 Feb 2010 00:52:00 -0000 Message-ID: <20100206005152.1613.qmail@sourceware.org> X-Bugzilla-Reason: CC References: Subject: [Bug middle-end/42973] [4.5 regression] IRA apparently systematically making reload too busy on 2 address instructions with 3 operands In-Reply-To: Reply-To: gcc-bugzilla@gcc.gnu.org To: gcc-bugs@gcc.gnu.org From: "hubicka at ucw dot cz" Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-bugs-owner@gcc.gnu.org X-SW-Source: 2010-02/txt/msg00444.txt.bz2 ------- Comment #3 from hubicka at ucw dot cz 2010-02-06 00:51 ------- Subject: Re: [4.5 regression] IRA apparently systematically making reload too busy on 2 address instructions with 3 operands > The argument for using conflicts is that the vast majority of the time if the > constraints force us to tie a specific source to the destination, then the > other source must not use the same hard reg as the destination. There might be > oddball cases where all the operands could share the same hard reg, but I > suspect those would be extremely rare. This should happen only when all operands are already same pseudo, and at that time there won't be conflict to add, right? Honza -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42973