From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-prs-return-21329-listarch-gcc-prs=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 1651 invoked by alias); 23 Aug 2002 16:36:01 -0000
Mailing-List: contact gcc-prs-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Archive: <http://gcc.gnu.org/ml/gcc-prs/>
List-Post: <mailto:gcc-prs@gcc.gnu.org>
List-Help: <mailto:gcc-prs-help@gcc.gnu.org>
Sender: gcc-prs-owner@gcc.gnu.org
Received: (qmail 1595 invoked by uid 71); 23 Aug 2002 16:36:00 -0000
Resent-Date: 23 Aug 2002 16:36:00 -0000
Resent-Message-ID: <20020823163600.1594.qmail@sources.redhat.com>
Resent-From: gcc-gnats@gcc.gnu.org (GNATS Filer)
Resent-Cc: gcc-prs@gcc.gnu.org, gcc-bugs@gcc.gnu.org
Resent-Reply-To: gcc-gnats@gcc.gnu.org, leick.robinson@motorola.com
Received: (qmail 29174 invoked by uid 61); 23 Aug 2002 16:29:32 -0000
Message-Id: <20020823162932.29173.qmail@sources.redhat.com>
Date: Fri, 23 Aug 2002 10:26:00 -0000
From: leick.robinson@motorola.com
Reply-To: leick.robinson@motorola.com
To: gcc-gnats@gcc.gnu.org
X-Send-Pr-Version: gnatsweb-2.9.3 (1.1.1.1.2.31)
Subject: c++/7704: template match difference on function pointer template args
X-SW-Source: 2002-08/txt/msg00500.txt.bz2
List-Id: <gcc-prs.sourceware.org>


>Number:         7704
>Category:       c++
>Synopsis:       template match difference on function pointer template args
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          wrong-code
>Submitter-Id:   net
>Arrival-Date:   Fri Aug 23 09:36:00 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator:     leick.robinson@motorola.com
>Release:        gcc 2.95.2 or gcc 3.0.4
>Organization:
>Environment:
SUNW,Ultra-60
SunOS 5.8
>Description:
It's not clear whether the bug is in gcc 2.95.2 or 3.0.4.

There's a difference in the way these releases match 
template arguments to the corresponding (partial) template 
specialization.

In the code below, gcc 2.95.2 matches the function pointer
argument ( A<void (*) ()> ) to the last specialization
( class A<retval (*) (...)> ).

On the other hand, gcc 3.0.4 matches it to the specialization
for pointers ( A<U*> ).

Which is the correct behavior?
>How-To-Repeat:
#include <iostream>
using namespace std;

template <class T>
class A {
public:
   static void foo()
   {
      cout << "Default A" << endl;
   }
};

// Template specialization for a pointer
template <class U>
class A<U*> {
public:
   static void foo()
   {
      cout << "A for pointers" << endl;
   }
};

// Template specialization for function pointers (?)
template <class retval>
class A<retval (*) (...)> {
public:
   static void foo()
   {
      cout << "A for function pointers" << endl;
   };
};

int main()
{
   A<int>::foo();

   A<int*>::foo();

   A<void (*) ()>::foo();
}

// g++ 2.95.2 produces:
//
// Default A
// A for pointers
// A for function pointers
//
// g++ 3.0.4 produces:
//
// Default A
// A for pointers
// A for pointers
//
>Fix:

>Release-Note:
>Audit-Trail:
>Unformatted: