From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 30767 invoked by alias); 25 May 2008 17:19:16 -0000 Received: (qmail 30513 invoked by uid 48); 25 May 2008 17:18:33 -0000 Date: Sun, 25 May 2008 17:19:00 -0000 Message-ID: <20080525171833.30509.qmail@sourceware.org> X-Bugzilla-Reason: CC References: Subject: [Bug middle-end/36326] gimplification of aggregate copies introduces extra aggregate copy In-Reply-To: Reply-To: gcc-bugzilla@gcc.gnu.org To: gcc-bugs@gcc.gnu.org From: "rguenth 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: 2008-05/txt/msg01867.txt.bz2 ------- Comment #2 from rguenth at gcc dot gnu dot org 2008-05-25 17:18 ------- Which was just a workaround. It needs proper fixing so that Index: tree-gimple.c =================================================================== --- tree-gimple.c (revision 135859) +++ tree-gimple.c (working copy) @@ -116,10 +116,7 @@ is_gimple_mem_rhs (tree t) a renamed variable. Also force a temporary if the type doesn't need to be stored in memory, since it's cheap and prevents erroneous tailcalls (PR 17526). */ - if (is_gimple_reg_type (TREE_TYPE (t)) - || (TYPE_MODE (TREE_TYPE (t)) != BLKmode - && (TREE_CODE (t) != CALL_EXPR - || ! aggregate_value_p (t, t)))) + if (is_gimple_reg_type (TREE_TYPE (t))) return is_gimple_val (t); else return is_gimple_formal_tmp_rhs (t); doesn't regress. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36326