From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 3376 invoked by alias); 27 Apr 2004 10:25:30 -0000 Mailing-List: contact java-prs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: java-prs-owner@gcc.gnu.org Received: (qmail 3357 invoked by uid 48); 27 Apr 2004 10:25:28 -0000 Date: Tue, 27 Apr 2004 10:25:00 -0000 Message-ID: <20040427102528.3356.qmail@sources.redhat.com> From: "rmathew at gcc dot gnu dot org" To: java-prs@gcc.gnu.org In-Reply-To: <20040425154237.15133.konqueror@gmx.de> References: <20040425154237.15133.konqueror@gmx.de> Reply-To: gcc-bugzilla@gcc.gnu.org Subject: [Bug java/15133] gcjh generates wrong method signatures X-Bugzilla-Reason: CC X-SW-Source: 2004-q2/txt/msg00054.txt.bz2 List-Id: ------- Additional Comments From rmathew at gcc dot gnu dot org 2004-04-27 10:25 ------- The problem is that disambiguation using arguments should only happen for *native* overloaded methods - in this case, even a non-native method triggers the disambiguation logic. More specifically, in gjavah.c we should also keep track of whether a method seen before was native and use this information in overloaded_jni_method_exists_p(). I have a patch for this which I will submit shortly, assuming it passes all tests - it certainly resolves this particular testcase and also behaves correctly when more overloaded native methods are present. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15133