On Wed, Sep 23, 2009 at 12:46:02PM -0400, John Cowan wrote:
> I'm not sure I follow you.

Sorry, my fault: terminology thinko.  I thought that you were
arguing against *literal* floats.  My C/assembler background
hasn't quite let go yet, and I still think of most such uses as
"immediate".

> By "immediate floats" I mean ones that are
> stuffed into the same space as a pointer, but are distinguishable from
> it, exploiting the fact that pointers on modern hardware almost always
> have 000 or 0000 in the bottom few bits.  This is commonly done in Scheme
> for fixnums, characters, #t, #f, and ().
> 
> Some implementations have tried to put chopped versions of 32-bit
> IEEE single-floats into 32-bit pointer space in the same way; it is
> this practice I was warning against, because of the cumulative error
> resulting from truncating instead of proper rounding.

Sure, that would be bad.

In support of that, I keep trying to figure out how a compiled
scheme could manage to feel like scheme while not using type
tags: at the moment I'm into a combination of boxing (with bibop
meta-data) and ... it seems very difficult.  I'm not much of a
compiler guru, I'm afraid.

Sorry for the noise.

Cheers,

-- 
Andrew

_______________________________________________
r6rs-discuss mailing list
[email protected]
http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss

Reply via email to