Re: static method hides virtual

Re: static method hides virtual

[Andrew Fitzgibbon]

Also a problem  on
        gcc version egcs-2.91.66 19990314 (egcs-1.1.2 release)

Re: static method hides virtual

[Zack Weinberg]

This bug is fixed in the development version, soon to be egcs 1.2.

zw

static method hides virtual

[Andrew Fitzgibbon]

  Reading specs from /homes/39/pkg/lib/gcc-lib/sparc-sun-solaris2.6/egcs-2.91.60/specs
  gcc version egcs-2.91.60 19981201 (egcs-1.1.1 release)

I'm fairly sure the following is a bug.  The code below prints "bad" unless 
the "static" is deleted.

Paste the code below to reproduce the behaviour.

cat << EOF > bad.cc
#include <iostream.h>

struct UIT { 
  virtual void get()      { cerr << "bad"; }
  AWFSTATIC void get(void*) { cerr << "inconcievably bad"; }
          void p()        { get(); }
};

struct S3d : public UIT {
  void get() { cerr << "ok"; }
};

main()
{
  S3d* s = new S3d;
  s->p(); // Should call S3d::get()   shirley?
}
EOF

egcs -DAWFSTATIC=static bad.cc && ./a.out
# Prints "bad"

egcs -DAWFSTATIC= bad.cc && ./a.out
# Prints "ok"
>From awf@robots.ox.ac.uk Sun Apr 25 10:14:00 1999
From: Andrew Fitzgibbon <awf@robots.ox.ac.uk>
To: egcs-bugs@cygnus.com
Subject: Re: static method hides virtual
Date: Sun, 25 Apr 1999 10:14:00 -0000
Message-id: <199904251714.SAA15923@volumnia.robots.ox.ac.uk>
X-SW-Source: 1999-04/msg00689.html
Content-length: 82

Also a problem  on
        gcc version egcs-2.91.66 19990314 (egcs-1.1.2 release)
>From zack@rabi.columbia.edu Sun Apr 25 10:17:00 1999
From: Zack Weinberg <zack@rabi.columbia.edu>
To: Andrew Fitzgibbon <awf@robots.ox.ac.uk>
Cc: egcs-bugs@egcs.cygnus.com
Subject: Re: static method hides virtual 
Date: Sun, 25 Apr 1999 10:17:00 -0000
Message-id: <199904251717.NAA04126@blastula.phys.columbia.edu>
References: <199904251714.SAA15923@volumnia.robots.ox.ac.uk>
X-SW-Source: 1999-04/msg00690.html
Content-length: 71

This bug is fixed in the development version, soon to be egcs 1.2.

zw
>From law@upchuck.cygnus.com Sun Apr 25 10:45:00 1999
From: Jeffrey A Law <law@upchuck.cygnus.com>
To: mombasa@ptolemy.arc.nasa.gov
Cc: egcs-bugs@egcs.cygnus.com, mombasa@lagoon.arc.nasa.gov
Subject: Re: build from egcs-1.1.2.tar.gz 
Date: Sun, 25 Apr 1999 10:45:00 -0000
Message-id: <5681.925062117@upchuck.cygnus.com>
References: <199904232244.PAA24488@lagoon.arc.nasa.gov>
X-SW-Source: 1999-04/msg00691.html
Content-length: 496

  In message < 199904232244.PAA24488@lagoon.arc.nasa.gov >you write:
  > 
  > Hi,
  > 
  >  I am get the following error on performing the following build
  >  on Solaris 2.5.1, having configured with the following instruction
  > 
  > ../egcs-1.1.2/configure --prefix=/usr/local/Gcs --enable-version-specific-r
  > untime-libs --enable-languages="c++ f77 objc"
--enable-languages is not supported for egcs-1.1.x releases.  Only in the
development snapshots (which will become egcs-1.2.x).

jeff
>From donn@verinet.com Sun Apr 25 11:22:00 1999
From: Donn Terry <donn@verinet.com>
To: egcs-bugs@egcs.cygnus.com
Subject: Regression g++.jason/rfg10.C: "excessive" pass?
Date: Sun, 25 Apr 1999 11:22:00 -0000
Message-id: <37234F64.3915048E@verinet.com>
X-SW-Source: 1999-04/msg00692.html
Content-length: 2393

In g++.jason/rfg10.c:

The alpha 32 compiler appears to pass the intent of this
test, but it's maybe a bit overenthusiastic about it :-).
All the other architectures seem happy, so I'm not quite
sure where to go with this: treat it as a test bug, a compiler
bug (somehow suppress the extra warnings), just ignore it,
or ???.  (I suspect that suppressing the additional warnings
would be tricky to do without overshooting and suppressing
too much.)

Comments, anyone?

Donn


Here's the output.

Executing on host: /home/donn/egcs.bin/gcc/testsuite/../xgcc
-B/home/donn/egcs.b
in/gcc/testsuite/../
/home/donn/egcs.source/gcc/testsuite/g++.old-deja/g++.jason
/rfg10.C  -ansi -pedantic-errors  -I/home/donn/egcs.source/libio
-I/home/donn/eg
cs.bin/alpha-pc-interix/libio -I/home/donn/egcs.source/libstdc++
-I/home/donn/eg
cs.source/libstdc++/stl -lstdc++
-L/home/donn/egcs.bin/alpha-pc-interix//libstd
c++ -L/home/donn/egcs.bin/alpha-pc-interix//libiberty  -lm  -o
/home/donn/egcs.b
in/gcc/testsuite/g++-jason-rfg10-C
spawn /home/donn/egcs.bin/gcc/testsuite/../xgcc
-B/home/donn/egcs.bin/gcc/testsu
ite/../
/home/donn/egcs.source/gcc/testsuite/g++.old-deja/g++.jason/rfg10.C -ans

i -pedantic-errors -I/home/donn/egcs.source/libio
-I/home/donn/egcs.bin/alpha-pc
-interix/libio -I/home/donn/egcs.source/libstdc++
-I/home/donn/egcs.source/libst
dc++/stl -lstdc++ -L/home/donn/egcs.bin/alpha-pc-interix//libstdc++
-L/home/donn
/egcs.bin/alpha-pc-interix//libiberty -lm -o
/home/donn/egcs.bin/gcc/testsuite/g
++-jason-rfg10-C ^M
/home/donn/egcs.source/gcc/testsuite/g++.old-deja/g++.jason/rfg10.C:8:
integer c
onstant out of range^M
/home/donn/egcs.source/gcc/testsuite/g++.old-deja/g++.jason/rfg10.C:8:
warning:
decimal integer constant is so large that it is unsigned^M
/home/donn/egcs.source/gcc/testsuite/g++.old-deja/g++.jason/rfg10.C:9:
overflow
in enumeration values at `blue'^M
compiler exited with status 1
output is:
/home/donn/egcs.source/gcc/testsuite/g++.old-deja/g++.jason/rfg10.C:8:
integer c
onstant out of range^M
/home/donn/egcs.source/gcc/testsuite/g++.old-deja/g++.jason/rfg10.C:8:
warning:
decimal integer constant is so large that it is unsigned^M
/home/donn/egcs.source/gcc/testsuite/g++.old-deja/g++.jason/rfg10.C:9:
overflow
in enumeration values at `blue'^M

PASS: g++.jason/rfg10.C enum overflow (test for errors, line 9)
FAIL: g++.jason/rfg10.C (test for excess errors)