From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-prs-return-23571-listarch-gcc-prs=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 30849 invoked by alias); 28 Oct 2002 21:27:46 -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 30830 invoked by uid 61); 28 Oct 2002 21:27:46 -0000
Date: Mon, 28 Oct 2002 13:27:00 -0000
Message-ID: <20021028212746.30829.qmail@sources.redhat.com>
To: gcc-bugs@gcc.gnu.org, gcc-prs@gcc.gnu.org, martin.dorey@bluearc.com,
  nobody@gcc.gnu.org
From: bangerth@dealii.org
Reply-To: bangerth@dealii.org, gcc-bugs@gcc.gnu.org, gcc-prs@gcc.gnu.org,
  martin.dorey@bluearc.com, nobody@gcc.gnu.org, gcc-gnats@gcc.gnu.org
Subject: Re: c++/8332: strlen/template interaction causes ICE
X-SW-Source: 2002-10/txt/msg01177.txt.bz2
List-Id: <gcc-prs.sourceware.org>

Synopsis: strlen/template interaction causes ICE

State-Changed-From-To: open->analyzed
State-Changed-By: bangerth
State-Changed-When: Mon Oct 28 13:27:44 2002
State-Changed-Why:
    I _can_ reproduce this. The bug goes away, if strlen is 
    renamed, or if its return type is changed to "int", so I
    assume this has something to do with built-ins. However,
    since the code also SegFaults if we include <string.h>, I
    have no reason to believe that this code is wrong. I'd think
    this is a serious regression from previous releases!
    
    For more information: the SegFault occurs here:
    #0  write_builtin_type (type=0x0) at ../../gcc-3.2/gcc/cp/mangle.c:1569
    #1  0x080bbfc5 in write_type (type=0x401888c0)
        at ../../gcc-3.2/gcc/cp/mangle.c:1399
    #2  0x080bc8df in write_template_args (args=0x4019ec80)
        at ../../gcc-3.2/gcc/cp/mangle.c:1789
    #3  0x080bb13f in write_name (decl=0x401aac40, ignore_local_scope=0)
        at ../../gcc-3.2/gcc/cp/mangle.c:719
    #4  0x080bafd3 in write_encoding (decl=0x401aac40)
        at ../../gcc-3.2/gcc/cp/mangle.c:654
    #5  0x080bd586 in mangle_decl_string (decl=0x401aac40)
        at ../../gcc-3.2/gcc/cp/mangle.c:630
    
    The code in question is
          /* TYPE may still be wchar_t, since that isn't in
    	 integer_type_nodes.  */
          if (type == wchar_type_node)
    	write_char ('w');
          else if (TYPE_FOR_JAVA (type))     <<<<< here
    	write_java_integer_type_codes (type);
          else
    and the reason is that type==0. On one frame higher, this
    code is called from
    	  write_builtin_type (TYPE_MAIN_VARIANT (type));
    Here, type is still!=0, but given the nesting of macros,
    this is the point where I leave the rest to someone else.
    
    Regards
      Wolfgang

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=8332