Re: [boost] Signals Library Review

From: Fernando Cacciola (fcacciola_at_[hidden])
Date: 2002-02-26 11:02:32

----- Original Message -----
From: "David Abrahams" <david.abrahams_at_[hidden]>
To: <boost_at_[hidden]>
Sent: Tuesday, February 26, 2002 12:58 PM
Subject: Re: [boost] Signals Library Review

>
> ----- Original Message -----
> From: "Steven Kirk" <steven.kirk_at_[hidden]>
>
>
> > > Ouch! I just checked the size of boost::signal<void> on the compilers
I
> > have
> > > easy access to, and here are the results:
> > >
> > > Borland C++ 5.5.1/Windows 2000: 80
> > > MSVC 6.0sp5/Windows 2000: 28
> > > GCC 2.95.3/Linux 2.4.x: 24
> > > GCC 3.0.3/Linux 2.4.x: 16
> > > Comeau 4.2.45.2/Linux 2.4.x: 20
> > >
> > Oh now those sizes sound far more reasonable! In that case consider my
> > objections dropped. I'm still not sure how you're getting 80 bytes on
BCC
> > though. On a default Jam build, which I notice doesn't switch on the
empty
> > base class optimisation or empty member optimisation, I get 136.
Enabling
> > these optimisations, I get 120. Still quite some way behind GCC3.0.3
>
> Please post a patch for the toolset which enables these by default.
There's
> no reason for them to be disabled unless they cause bugs.
>
> Thanks!
> -Dave
>
Unfortunately, they do cause spurious bugs, mostly ICEs. I have them turned
off by default.
I think we better leave them off by now...

Fernando Cacciola
Sierra s.r.l.
fcacciola_at_[hidden]
www.gosierra.com