Re: [boost] Optimization of compilation time and consumed memory

From: Paul Mensonides (pmenso57_at_[hidden])
Date: 2002-04-17 20:32:31

----- Original Message -----
From: "David Abrahams" <david.abrahams_at_[hidden]>
To: <boost_at_[hidden]>
Sent: Wednesday, April 17, 2002 5:01 PM
Subject: Re: [boost] Optimization of compilation time and consumed memory

>
> ----- Original Message -----
> From: "Paul Mensonides" <pmenso57_at_[hidden]>
>
> > You also slow down compile times with extra instantiations of the
> templates that
> > support the abstraction between the algorithms and the lists.  I think
> that
> > speed of compilation is more about pure number of instantiations than
> the depth
> > of instantiation
>
> What is your basis for that conclusion? Are you speculating?
> I happen to know that in one popular compiler implementation, name
> mangling is the limiting factor, such that instantiation depth has a
> direct impact on speed.

Okay.  I was speculating.  However, it isn't worth bending over backward for the
sake of the name mangling inefficiency of compiler X.  We should be trying to
push compiler vendors not coddle them.

By the way, which compiler are you referring to?  I primarily use Comeau C++,
which is definitely not the most efficient compiler out there--though it is the
most compliant. :)

> > .  That compiler-defined limit is really only there to protect
> > the compiler from recursing indefinitely:
>
> But in some compilers it is hard-coded (e.g. recent CodeWarrior
> releases)

But it is likely that this will not remain as meta-programming becomes more
widespread.

Paul Mensonides