Re: Bloat

[Thor Lancelot Simon <tls%rek.tjls.com@localhost>]

On Sat, Jan 31, 2009 at 01:25:57AM +0200, Antti Kantee wrote:
> On Fri Jan 30 2009 at 13:33:00 -0500, Thor Lancelot Simon wrote:
> > Except that you seem to be making the absolute argument that all uses
> > of inline are bad
> 
> How did you arrive at that conclusion?  I said I'd like to see proof of
> the benefit instead of premature optimization and handwaving.

Clearly, you're not going to see proof that each use of inline is a
beneficial use.  As Homer had Achilles say, "if that's what you want,
you should get used to wanting."

On the other hand, you've been offered at least one reasoned argument,
with a concrete example, establishing that some kinds of short inline
functions in fact offer a general performance benefit.  Instead of
addressing it, you snipped it.

This leads me to the conclusion that any facts or reasoning that actually
run counter to the point of view you came into this discussion with, you're
like to simply ignore.  Okay, I'll stop offering them.  Fine with me --
except I think it's a little silly, because about the vast majority of
uses of "inline" historically in NetBSD, I think we almost certainly agree:
they should go.

Thor