Re: [gforth] Gforth's fsl-util.*

[Bernd Paysan]

Am Sonntag, 25. November 2012, 10:35:22 schrieb Charles Montgomery:
> The admittedly slow high-level versions of } and }} are provided for
> the sake of portability that comes from being Standard.  If you should
> come up with improvements for a specific architecture, or other things
> as suggested by Bernd, they could be made available to other FSL users
> by adding them in an auxiliary file on the taygeta site.  (Such files
> don't need to go through the reviewing process used for the Library
> Algorithms.)

No, } and }} are broken by design, as they do not address the actual problem: 
Iterate through vectors and arrays.  If you write a scientific library, you 
should have good higher order primitives, and iteration through a vector or 
array is such a higher order primitive.

Don't think in Fortran.  Think in Forth.  Factor (this implies common factors 
for vector/matrix operations like v* and faxpy), and extend the language.

AFAIK, we install the fsl-util.fth, because FSL loads that particular 
filename.  As we in Gforth don't use that file extension (we use .fs), we 
accidently duplicated the files...

-- 
Bernd Paysan
"If you want it done right, you have to do it yourself"
http://bernd-paysan.de/

signature.asc
Description: This is a digitally signed message part.