This is the mail archive of the
cygwin@cygwin.com
mailing list for the Cygwin project.
Re: gcc bug - temporary objects not destroyed properly
- To: Andrews Harold G Maj USAFA/DFCS <Harold dot Andrews at usafa dot af dot mil>
- Subject: Re: gcc bug - temporary objects not destroyed properly
- From: David Acton <da at alphawave dot net>
- Date: Tue, 9 Oct 2001 20:34:45 +0000
- Cc: "'cygwin at cygwin dot com'" <cygwin at cygwin dot com>
- References: <9BBB0C9AF506D311A68E00902745A53703CC798D@fsxqpz04.usafa.af.mil>
Hi Andy,
> I've seen something similar with gcc 2.95 under Cygwin. The main difference
> was that I had at least one virtual method in my class. The compiler (using
> the -Wall option) mentioned that I should declare a *virtual* destructor.
Yes, I always compile with -Wall -pedantic and it complains vigorously if
you don't accompany virtual functions with a virtual destructor!
> The one I created was was essentially empty, and seemed to fix the problem
> (though I'm not completely certain why that was). This might be something
> that could fix your problem if you want to continue using GCC 2.95.x
Yes, thanks - I'll give it a go. The actual class that caused the original
problem is one of the few in our application without any virtuality so that
might well be significant.
> BTW, I was not able to repeat the error using GCC 3.0.1 (though there were
> some simple-to-fix compiler errors). You might want to consider upgrading.
I did install 3.0.1 and got my code to compile okay with it. But we went back
to 2.95 becaues 3.0.1 seems soooo slow at C++ (and 2.96 so dodgy! :-) It's
another good thing to try though.
Cheers
--
David Acton
Alphawave Ltd
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/