implement __attribute__((cleanup(function)))

[Daniel Jacobowitz]

On Tue, May 06, 2003 at 02:57:06PM +0200, Gerald Pfeifer wrote:
> On Tue, 6 May 2003, Michael Matz wrote:
> > I hereby request as formerly as I know that the steering committee
> > discusses and resolves the issue about try/finally in GCC.
> 
> I have forwarded your message with some explanations (for those who
> might not have followed the issue closely).
> 
> > _My_ personal opinion is that we will accept it if we can define it.
> > There were useful definitions of it in the past (I remember to have
> > written one some months ago, when it was a hot topic on gcc-patches@).
> > As we can't seem to resolve it by normal discussions and that leads to
> > (seemingly) frustration of developers I think the committee should do,
> > what it was supposed to do (IMHO).
> 
> What would help, I guess, are references/statements by those in favor
> of try/finally and those against it.
> 
> If you (Plural) send self-contained explanations of your point of view
> to the gcc list or me, I'll happily forward them as well.
> 
> On Tue, 6 May 2003, Daniel Jacobowitz wrote:
> > At the risk of being obvious, I don't believe that discussion is
> > appropriate for 3.3.  I also don't see any actual risk in adding
> > __attribute__((cleanup)) on the branch.  So can we have this discussion
> > in a few weeks, please?
> 
> Alexandre argued that the extension may not without problems.
> 
> I have no strong opinion either way yet, but in any case, should we risk
> shipping something in GCC 3.3 when we are not confident that we won't
> completely revert directions and remove the extension in GCC 3.4 in favor
> of a different one?

I'm confident that, even if we revert position on try/finally, that
will not force us to remove __attribute__((cleanup)).  It's really a
very simple extension...

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer