From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 22798 invoked by alias); 18 Dec 2003 21:44:48 -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 22775 invoked by uid 48); 18 Dec 2003 21:44:47 -0000 Date: Thu, 18 Dec 2003 22:19:00 -0000 Message-ID: <20031218214447.22774.qmail@sources.redhat.com> From: "reichelt at gcc dot gnu dot org" To: gcc-bugs@gcc.gnu.org In-Reply-To: <20030314113600.10079.stip@mathematik.uni-ulm.de> References: <20030314113600.10079.stip@mathematik.uni-ulm.de> Reply-To: gcc-bugzilla@gcc.gnu.org Subject: [Bug c++/10079] ICE (segfault) while substitute return type containing unresolved qualified id X-Bugzilla-Reason: CC X-SW-Source: 2003-12/txt/msg01966.txt.bz2 List-Id: ------- Additional Comments From reichelt at gcc dot gnu dot org 2003-12-18 21:44 ------- Sorry, the testcase didn't go in yet because I got confused about the validity of the testcase: The PR was marked as "ice-on-invalid-code", but this should be "ice-on-valid-code", since the second version of "func" is more specialized than the first one. In fact mainline does compile the testcase without problems. Giovanni's guess that the segfault happens while trying to format the error message about the ambiguity of the call added to this confusion. If you really make ambiguous functions - i.e. use template void func(T,U) {} as second function, you get an error message about the ambiguity of the call (and then a segfault with the 3.3 branch). So Giovanni's guess was wrong. I'll mark the PR as "ice-on-valid-code" and add a small testcase to the testsuite Real Soon Now(tm). -- What |Removed |Added ---------------------------------------------------------------------------- Keywords|ice-on-invalid-code |ice-on-valid-code http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10079