From mboxrd@z Thu Jan  1 00:00:00 1970
From: "Martin v. Loewis" <martin@mira.isdn.cs.tu-berlin.de>
To: jls@sco.com
Cc: egcs@egcs.cygnus.com, robertl@sco.com, martin@mira.isdn.cs.tu-berlin.de
Subject: Re: incorrect C++ EH tests in egcs tree
Date: Mon, 22 Mar 1999 12:44:00 -0000
Message-id: <199903222039.VAA07030@mira.isdn.cs.tu-berlin.de>
In-reply-to: < 199903221606.LAA23180@kauai.newjersey.sco.com > (jls@sco.com)
References: <199903221606.LAA23180@kauai.newjersey.sco.com> <199903221606.LAA23180@kauai.newjersey.sco.com>
X-SW-Source: 1999-03/msg00726.html

> This test uses global replacements of operator new and operator delete
> to detect whether the storage acquired in main() is released.  But by
> using *global* replacements, it picks up possible new's done by the system
> startup library that are not (yet) delete'd by the time the return 
> statement in main() is evaluated.  Thus this test may get a spurious failure
> on some platforms.

I agree that the test case is bogus. Your modification test a slightly
different feature, though, so I propose a different fix.

Just initialize newed after main enters.

int main ()
{
  newed = created = 0;
  try {
    foo (new B (A ()));
  } catch (...) { }

  return !(!newed && !created);
}

Would this work as well?

regards,
Martin

From mboxrd@z Thu Jan  1 00:00:00 1970
From: "Martin v. Loewis" <martin@mira.isdn.cs.tu-berlin.de>
To: jls@sco.com
Cc: egcs@egcs.cygnus.com, robertl@sco.com, martin@mira.isdn.cs.tu-berlin.de
Subject: Re: incorrect C++ EH tests in egcs tree
Date: Wed, 31 Mar 1999 23:46:00 -0000
Message-ID: <199903222039.VAA07030@mira.isdn.cs.tu-berlin.de>
References: <199903221606.LAA23180@kauai.newjersey.sco.com>
X-SW-Source: 1999-03n/msg00731.html
Message-ID: <19990331234600.61HQU1L9pAjvqU-66kxiW2OHAWz0SpW1jFtP9GuAukY@z>

> This test uses global replacements of operator new and operator delete
> to detect whether the storage acquired in main() is released.  But by
> using *global* replacements, it picks up possible new's done by the system
> startup library that are not (yet) delete'd by the time the return 
> statement in main() is evaluated.  Thus this test may get a spurious failure
> on some platforms.

I agree that the test case is bogus. Your modification test a slightly
different feature, though, so I propose a different fix.

Just initialize newed after main enters.

int main ()
{
  newed = created = 0;
  try {
    foo (new B (A ()));
  } catch (...) { }

  return !(!newed && !created);
}

Would this work as well?

regards,
Martin