This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: [c++ patch] (was Re: member templates and default parameter)


On Mon, Mar 26, 2001 at 08:06:34AM -0800, Mark Mitchell wrote:
> 
> I think the key point is that everything *might* become part of the C
> frnot-end some point.  Especially things in the C++ front-end.
> Especially since the new C++ parser may just happen to understand C as
> well...

That doesn't sound like "just" a C front-end.  That sounds more like a
very intelligent generic front- or middle-end.

If GCC moves to having a generic front-end with small language-specific
bits laying on top of it for each compiler (which sounds like a wonderful
thing), then I would hope that K&R would be dropped for the big generic bits.
Keep it for a tiny bootstrapping compiler or some such, but let the bulk
of the code at least move to C89.


> Now, GDB did switch to ANSI/ISO C at some point; we *could* decide
> that we don't need to support K&R host compilers any more.  But,
> that's a whole 'nother can of worms.

I find it hard to believe that there are any useful platforms on which
a K&R C compiler is the only native compiler /and/ that no precompiled
gcc binary exists for such a machine.  Retrocomputing for fun, sure,
but useful platform?

Anyhow, this is off-topic for gcc-patches; I'm just constantly amazed that
K&R keeps popping up as a requirement.


-- 
pedwards at disaster dot jaj dot com  |  pme at sources dot redhat dot com
devphil at several other less interesting addresses in various dot domains
The gods do not protect fools.  Fools are protected by more capable fools.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]