From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 29135 invoked by alias); 20 Aug 2014 02:32:44 -0000 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 Received: (qmail 29080 invoked by uid 48); 20 Aug 2014 02:32:29 -0000 From: "amker at gcc dot gnu.org" To: gcc-bugs@gcc.gnu.org Subject: [Bug rtl-optimization/62151] [5 Regression] wrong code at -O2 and -O3 on x86_64-linux-gnu Date: Wed, 20 Aug 2014 02:32:00 -0000 X-Bugzilla-Reason: CC X-Bugzilla-Type: changed X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: gcc X-Bugzilla-Component: rtl-optimization X-Bugzilla-Version: 5.0 X-Bugzilla-Keywords: wrong-code X-Bugzilla-Severity: normal X-Bugzilla-Who: amker at gcc dot gnu.org X-Bugzilla-Status: NEW X-Bugzilla-Priority: P1 X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org X-Bugzilla-Target-Milestone: 5.0 X-Bugzilla-Flags: X-Bugzilla-Changed-Fields: Message-ID: In-Reply-To: References: Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/ Auto-Submitted: auto-generated MIME-Version: 1.0 X-SW-Source: 2014-08/txt/msg01326.txt.bz2 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62151 --- Comment #9 from amker at gcc dot gnu.org --- (In reply to Eric Botcazou from comment #8) > > I will try to test a patch, meanwhile, I am wondering if any combine expert > > has something to share. > > distribute_notes is certainly an endless source of bugs. One patch is sent at https://gcc.gnu.org/ml/gcc-patches/2014-08/msg01718.html It's a scratch patch for discussion's purpose. Though I think it's right to find distribution place for REG_DEAD note starting from i2 if it kills the register, I am not sure if it's fixing the issue in a case by case manner. So it's highly appreciated if any expert in this area can share some opinions, or review the patch? > > > BTW, combine pass seems to be another ad-hoc implementation like reload, I saw > > several "Don't know" in comments during this investigation! > > You would need to define what an "ad-hoc implementation" is exactly. See > the top of the file for the origin of the pass. Err, I may used the wrong word. These are just frustrating words after seeing several "Don't know" comments in the code. Please ignore them. Thanks.