Re: PATCH to c-commonc.c, preliminary for PR/3865 fix

[Neil Booth <neil at daikokuya dot co dot uk>]

Zack Weinberg wrote:-

> > Sometime in 3.4 I might improve this area a bit.  e.g. it is probably
> > worth having a cpp_define_bool (macroname, on) which defines macroname
> > to be 1 or 0.  We could avoid all the buffer pushing, lexing and popping
> > in this common case (>80%?)
> 
> It's an interesting idea.  Didn't we have problems with that kind of
> optimization before?  That was with macros defined to the empty
> string, though, maybe defined to a single token wouldn't be as much of
> a problem.
> 
> I was mainly thinking about reducing initialization overhead, not
> macro expansion overhead.

I think you misunderstood.  I _was_ talking about init overhead; we
still have to push, lex and pop the string buffer.

Neil.