Re: debug/543

Re: debug/543

[neil]

The following reply was made to PR debug/543; it has been noted by GNATS.

From: neil@gcc.gnu.org
To: gcc-gnats@gcc.gnu.org, nobody@gcc.gnu.org, robertlipe@usa.net
Cc:  
Subject: Re: debug/543
Date: 24 Jan 2001 21:47:16 -0000

 Synopsis: dwarfout.c ICE during bootstrap simplified
 
 State-Changed-From-To: open->feedback
 State-Changed-By: neil
 State-Changed-When: Wed Jan 24 13:47:16 2001
 State-Changed-Why:
     Hi Robert,
     
     This compiles fine on 586 Linux with CVS GCC.  Is this
     target specific, or is it fixed for you too?
 
 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view&pr=543&database=gcc
>From pcarlini@unitus.it Sun Apr 01 00:00:00 2001
From: pcarlini@unitus.it
To: gcc-gnats@gcc.gnu.org
Cc: bkoz@redhat.com
Subject: libstdc++/1704: Null output from Josuttis example
Date: Sun, 01 Apr 2001 00:00:00 -0000
Message-id: <20010119204609.478.qmail@sourceware.cygnus.com>
X-SW-Source: 2001-q1/msg00499.html
Content-length: 1700

>Number:         1704
>Category:       libstdc++
>Synopsis:       Null output from Josuttis example
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Fri Jan 19 12:56:00 PST 2001
>Closed-Date:
>Last-Modified:
>Originator:     Paolo Carlini
>Release:        gcc version 2.97 20010119 (experimental) or any snapshot in the last weeks
>Organization:
>Environment:
Intel PII, GLIBC2.2.1
>Description:
The following code (by N. Josuttis, p.669- of his book) does not produce the expected output (i.e. 31 HEXADECIMAL: 1F):


#include <iostream>
#include <streambuf>
#include <locale>
#include <cstdio>

class outbuf : public std::streambuf
{
  protected:
    /* central output function
     * - print characters in uppercase mode
     */
    virtual int_type overflow (int_type c) {
        if (c != EOF) {
            // convert lowercase to uppercase
            c = std::toupper(c,getloc());

            // and write the character to the standard output
            if (std::putchar(c) == EOF) {
                return EOF;
            }
        }
        return c;
    }
};

int main()
{
    outbuf ob;                // create special output buffer
    std::ostream out(&ob);    // initialize output stream with that output buffer

    out << "31 hexadecimal: " << std::hex << 31 << std::endl;
}
>How-To-Repeat:
The next example in the book too (outbuf1x.cpp & outbuf1x.hpp) does not produce the expected output

P.S.: Thanks to Benjamin Kosnik for having fixed the bug I previously reported!
		
>Fix:
Dunno, unfortunately...
>Release-Note:
>Audit-Trail:
>Unformatted:
>From mps@sonartech.com.au Sun Apr 01 00:00:00 2001
From: mps@sonartech.com.au
To: gcc-gnats@gcc.gnu.org
Subject: c++/2136: g++ seems to disregard '::' in '::delete'
Date: Sun, 01 Apr 2001 00:00:00 -0000
Message-id: <20010301092659.17769.qmail@sourceware.cygnus.com>
X-SW-Source: 2001-q1/msg01836.html
Content-length: 932

>Number:         2136
>Category:       c++
>Synopsis:       g++ seems to disregard '::' in '::delete'
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          rejects-legal
>Submitter-Id:   net
>Arrival-Date:   Thu Mar 01 01:36:00 PST 2001
>Closed-Date:
>Last-Modified:
>Originator:     Michael Shepanski
>Release:        2.95.2 19991024
>Organization:
>Environment:
AIX power-pc
>Description:
#include  <sys/types.h>

class C {
public:
    int x;

    // A lot of stuff that main() should NOT use:
    static char pool [100];	
    void *operator new (size_t, float) { return pool; }
    void operator delete (void *, float) {}
};

int main ()
{
    C	*p = ::new C;
    ::delete p;     // confused by the presence of C::delete
}
>How-To-Repeat:
Attempt to compile the code in the "description" section
>Fix:

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

Re: debug/543

[Robert Lipe]

The following reply was made to PR debug/543; it has been noted by GNATS.

From: Robert Lipe <robertlipe@usa.net>
To: neil@gcc.gnu.org
Cc: gcc-gnats@gcc.gnu.org, nobody@gcc.gnu.org
Subject: Re: debug/543
Date: Fri, 26 Jan 2001 10:36:34 -0600

 neil@gcc.gnu.org wrote:
 > Synopsis: dwarfout.c ICE during bootstrap simplified
 > 
 > State-Changed-From-To: open->feedback
 > State-Changed-By: neil
 > State-Changed-When: Wed Jan 24 13:47:16 2001
 > State-Changed-Why:
 >     Hi Robert,
 >     
 >     This compiles fine on 586 Linux with CVS GCC.  Is this
 >     target specific, or is it fixed for you too?
 > 
 > http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view&pr=543&database=gcc
 
 Well, OpenServer no longer uses Dwarf-I by default, so I can't pinpoint
 a precise date this was fixed.    The testcase does not fail for me now
 on any target I checked, and I did verify that the original did fail on
 Linux.
 
 So I'm guessing this problem is now gone and should be closed.
 
 Thanx,
 RJL
>From nathan@gcc.gnu.org Sun Apr 01 00:00:00 2001
From: nathan@gcc.gnu.org
To: nobody@gcc.gnu.org
Cc: gcc-prs@gcc.gnu.org
Subject: Re: c++/1803
Date: Sun, 01 Apr 2001 00:00:00 -0000
Message-id: <20010129164601.9849.qmail@sourceware.cygnus.com>
X-SW-Source: 2001-q1/msg00804.html
Content-length: 623

The following reply was made to PR c++/1803; it has been noted by GNATS.

From: nathan@gcc.gnu.org
To: gcc-gnats@gcc.gnu.org, joseph.barreca@transport.alstom.com,
  nobody@gcc.gnu.org
Cc:  
Subject: Re: c++/1803
Date: 29 Jan 2001 16:45:33 -0000

 Synopsis: usage of variable declaration inside "for" loop corrupts other variable on stack
 
 State-Changed-From-To: open->feedback
 State-Changed-By: nathan
 State-Changed-When: Mon Jan 29 08:45:33 2001
 State-Changed-Why:
     There was no attached jdb1.C, please provide preprocessed
     source code
 
 http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view&pr=1803&database=gcc
>From porten@kde.org Sun Apr 01 00:00:00 2001
From: porten@kde.org
To: gcc-gnats@gcc.gnu.org
Subject: preprocessor/2444: Internal error, <new> vs a directory named 'new'
Date: Sun, 01 Apr 2001 00:00:00 -0000
Message-id: <20010331202924.8861.qmail@sourceware.cygnus.com>
X-SW-Source: 2001-q1/msg02695.html
Content-length: 772

>Number:         2444
>Category:       preprocessor
>Synopsis:       Internal error, <new> vs a directory named 'new'
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sat Mar 31 12:36:01 PST 2001
>Closed-Date:
>Last-Modified:
>Originator:     Harri Porten
>Release:        gcc version 3.1 20010326 (experimental)
>Organization:
>Environment:
i686-pc-linux-gnu
>Description:
The preprocessor appearantly gets confused when any
STL headers are included and a directory named 'new'
exists in the include path.
>How-To-Repeat:
mkdir new
echo "#include <vector>" > test.cpp
g++ -I. -c test.cpp
>Fix:

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