[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Fresh syntax



Bradley Lucier wrote:
> The first was that the authors of SRFI-77 don't seem to have any
> problem removing quotient and remainder, which match hardware
> operations on nearly every known architecture, and which I'm
> sure many, many more programs use than use syntax-case. If there
> is a hardware architecture that has single instructions for the
> proposed div or mod on small integers, I don't know what it is.

I expect the quotient, remainder, and modulo procedures
of R5RS to be provided by an R5RS-backward-compatibility
library of some sort.

The same compiler techniques that would be used to
compile fixnum+ into a single instruction would be
used to compile quotient into a single instruction
on architectures that make that possible.

Will