[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: small integers
- To: Kim Barrett <kab@CAMBRIDGE.APPLE.COM>
- Subject: Re: small integers
- From: Scott_Fahlman@SEF-PMAX.SLISP.CS.CMU.EDU
- Date: Sat, 10 Oct 92 22:24:16 -0400
- Cc: info-dylan@CAMBRIDGE.APPLE.COM
- In-reply-to: Your message of Sat, 10 Oct 92 19:40:45. <9210102337.AA00459@cambridge.apple.com>
Define a module that imports the Dylan module variable "+" under the name
"generalized+" and the Dylan module variable "fix+" as "+" ...
Hmmm...maybe. There would be a bunch of substitutions to make in the
module header -- not just +.
The problem is to make this easy enough so that the C programmers will feel
that they can get the efficient policy they want without a lot of
complicated gyrations. This kind of module hackery is certainly more
hassle than just changing the overflow policy with a declaration, but maybe
it can be turned into a no-brainer -- just insert this magic header and the
module becomes fixnum-only. It would be essential to pre-define all the
generic functions for fixnum-only arithmetic, and to make sure the compiler
understands what to do with them.
A disadvantage is that the whole module would have to have a single default
policy for overflow-handling. You couldn't switch back and forth for
certain critical functions. But maybe that's OK.
I'm not crazy about this way of doing it, but it could work.
Scott E. Fahlman
School of Computer Science
Carnegie Mellon University
5000 Forbes Avenue
Pittsburgh, PA 15213