From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 9165 invoked by alias); 25 Jul 2005 18:18:19 -0000 Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-bugs-owner@gcc.gnu.org Received: (qmail 8355 invoked by uid 48); 25 Jul 2005 18:18:05 -0000 Date: Mon, 25 Jul 2005 18:52:00 -0000 Message-ID: <20050725181805.8354.qmail@sourceware.org> From: "rth at gcc dot gnu dot org" To: gcc-bugs@gcc.gnu.org In-Reply-To: <20050520220116.21687.pinskia@gcc.gnu.org> References: <20050520220116.21687.pinskia@gcc.gnu.org> Reply-To: gcc-bugzilla@gcc.gnu.org Subject: [Bug c++/21687] [4.0/4.1 Regression] ICE in GC with local class inside a template function X-Bugzilla-Reason: CC X-SW-Source: 2005-07/txt/msg03255.txt.bz2 List-Id: ------- Additional Comments From rth at gcc dot gnu dot org 2005-07-25 18:18 ------- Whatever you might think, this is not my bug. This is the C++ front end calling ggc_collect in the middle of parsing a function. I can't tell whether or not there is a ggc_push_context missing when we begin parsing the local class, or if we simply shouldn't have the collect in expand_or_defer_fn. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21687