[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: RFD : libg++/gcc/egcs upgrades needed for libc6 (READ ME)

On Sun, Feb 22, 1998 at 06:23:46PM +0000, James Troup wrote:
> Marcus Brinkmann <Marcus.Brinkmann@ruhr-uni-bochum.de> writes:
> > Well, it may compile the kernel, but it is broken in respect to C++.
> FUD.  If it's ``broken'', how come we managed to get by with it for
> rex, bo and hamm till now?  Gee, I wonder how dselect is working? I
> heard a rumour that it's written in C++.  And lftp [ ... ad nauseam ].

I think they are working because they are developed with gcc 2.7.2, aren't
they? Thank you for snipping my next sentence, it gave you a pointer where
gcc 2.7.2 really has "flaws", in exception handling, a very important
feature of C++. I didn't take a look at the source code of any debian c++
programm, but you can use C++ in a variety of sorts, just as a better C or
as a complete new language. gcc 2.7.2 does the main things right, but
other's don't. Take a look at the bug reports about gcc, too.

> I realise gcc 2.7.x's C++ is (severely) sub-optimal, but *why*
> sacrifice the safety of the majority of the packages for a very small
> _minority_?  Could you give me numbers on how many packages in hamm
> truly need egcs/2.8 to be built?

Probably none. This may change. libgtkmm for example only builds very
recently with gcc 2.7.2, but truly it is a *very* unimportant package by

I don't have enough experience to be able to justice about the grade that
g++ is sub-optimal or broken, but even a simple program like

main(){ throw 1;}

is segfaulting (after you, surprise, recompiled with the undocumented
-fhandle-exceptions option), although uncaught exceptions should lead to

I don't say that we should sacrifice stability. It would be fine if gcc and
egcc can run side by side (as a matter of fact, I have installed both, gcc
2.7.2 and egcc, and it is working).

> > > IMHO we shouldn't sacrifice the safety of our C compiler for the
> > > dubious goal of giving the egcs team moral support.
> > 
> > The moral support is a side effect. The goal is to put pressure on
> > the gcc group to release stable versions earlier, as I see it.
> My point stands; we shouldn't sacrifice stability of our C compiler to
> put pressure on anyone.  What do we care about here; the integrity of
> our stable distribution (which hamm will become in a matter of months)
> or some misguided political lobbying?

I think you are overstating here a bit. If hamm is coming soon, we don't
have a choice but gcc 2.7, if hamm is taking months, gcc 2.8 should be
pretty stable ? (just guessing here).

As Christian already said: They really should coexist. People who need a
stable C compiler can use gcc 2.7 (or gcc 2.8 if ready), people who want to
cut the edge could use egcs.


"Rhubarb is no Egyptian god."        Debian GNU/Linux        finger brinkmd@ 
Marcus Brinkmann                   http://www.debian.org    master.debian.org
Marcus.Brinkmann@ruhr-uni-bochum.de                        for public  PGP Key
http://homepage.ruhr-uni-bochum.de/Marcus.Brinkmann/       PGP Key ID 36E7CD09

TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-devel-request@lists.debian.org . 
Trouble?  e-mail to templin@bucknell.edu .

Reply to: