Re: autoconf for type sizes

["Kaveh R. Ghazi" <ghazi at caip dot rutgers dot edu>]

 > From: Richard Henderson <rth@redhat.com>
 > 
 > On Mon, Mar 12, 2001 at 03:38:17PM -0500, Kaveh R. Ghazi wrote:
 > > 2.  (Optional) while you're at it, why not use WORD_BIT/LONG_BIT (with
 > >     defaults of 32 for both) and skip the sizeof test for int/long.
 > 
 > Why, when you are guaranteed that CHAR_BIT*sizeof(type) works?
 > r~

Because IMHO, the compile-time sizeof check is a gross hack and I wish
to reduce our reliance on it.  (BTW, I originally wrote it.)

I'm especially concerned because it may not work on some non-gcc
compilers.  We just don't know and while the unreleased upcoming
autoconf uses similar hacks, it hasn't been widely tested on non-gcc
compilers AFAIK.

Zack's floating stuff falls into a similar category, except it relies
too heavily on dark corners and edge cass of float rounding.

My instincts say we're going to get bit by either one or both.  But I
guess only time will tell...

		--Kaveh

PS: I don't mean to be a wet blanket Zack, I think you're doing great
stuff and I generally support the goal and appreciate the work you've
done cleaning out all the cruft in those files.

--
Kaveh R. Ghazi			Engagement Manager / Project Services
ghazi@caip.rutgers.edu		Qwest Internet Solutions