From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 858 invoked by alias); 29 Dec 2007 22:31:24 -0000 Received: (qmail 824 invoked by uid 48); 29 Dec 2007 22:31:09 -0000 Date: Sat, 29 Dec 2007 22:31:00 -0000 Message-ID: <20071229223109.823.qmail@sourceware.org> X-Bugzilla-Reason: CC References: Subject: [Bug tree-optimization/31081] [4.3 Regression] Inliner messes up SSA for abnormals In-Reply-To: Reply-To: gcc-bugzilla@gcc.gnu.org To: gcc-bugs@gcc.gnu.org From: "hubicka at gcc dot gnu dot org" 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: 2007-12/txt/msg02390.txt.bz2 ------- Comment #15 from hubicka at gcc dot gnu dot org 2007-12-29 22:31 ------- Well, I would personally not worry that much about pesimization. The current behaviour of increasing register pressure is worse. We definitly can insert the initialization to common dominator of all uses of the initialized SSA name, but the dominator tree nor the uses are readily available and I am not convinced it is worth the effort. I did some behchmarking on C++ benchmarks with patch initializing to 0 and found no regressionons. I will send patch shortly (the version I tested had little bug causing occasitonal ICEs) -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31081