Re: Status of Ada in GCC?

[Florian Weimer <fw at deneb dot enyo dot de>]

dewar@gnat.com (Robert Dewar) writes:

>> Would this imply that Ada.Strings.Bounded becomes actually usable?
>> (Currently, every instantiation of Generic_Bounded_Length contributes
>> slightly less than 30K of object code on x86.)
>
> fucntion-at-a-time has nothing whatever to do with shared
> generics. Actually Ada.Strings.Bounded is perfectly usable in a
> typical situation where you have very few different instantiations.

I'm not talking about shared generics.  I'm talking about not
generating (and emitting) code for subprograms formally part of
generic instantiations which are never referenced (and known to be
unreferenced at compile time).

Not only would this reduce executable size considerably (I think you
can already use pragma Eliminate for that), compile time would benefit
a lot as well.  During my first Ada experiments, the required compile
time was so *huge* that I avoided the package completely.  (Have you
ever asked your customers if they use more than one instantiation of
this generic per compilation unit?)

Of course, sometimes Moore's Law solves a problem, but even with
today's hardware, using ASB affects compile time in a significant
manner.