flashforth-devel

[Flashforth-devel] PIC24-30-33 and C Math Library Functions

[Pete Z. <pza...@pz...>]

Hi Mikael and all,

Where do we stand with interfacing the C Math Library Functions in the 
MPLAB XC16 Standard Compiler into FF for the PIC 16bit processors?

Specifically, has anyone put together a template (register saves, 
fixed-point or floating-point interface, etc)  for the use of some of 
these functions (e.g. log, exp, sin, cos, etc.)?

Is this something we should get together and have a 'work group' address 
for the good of the FF community?
Certainly a section in the wiki could focus on this FF and C interfacing 
and be a great place for folks to post their solutions, helpful hints 
and examples.

Thoughts, comments......?

Pete

Re: [Flashforth-devel] PIC24-30-33 and C Math Library Functions

[Mikael N. <mik...@pp...>]

Hello Pete,
The status is that I have done nothing in this area.

I think the math routines are written in assembly and have a
C-interface description. Maybe also a assembly interface descriptio can 
be found.

In general code in linked libraries is simplest to call from words 
defined when the FF core is compiled and linked to some library.

C4add.c is a simple example of this.

For example the following stuff could be described in the wiki.

1. The interface to the math libraries
2. Linking math routines to FlashForth words
3. Parse fixed point number
4. Print fixed point number
5. Parse floating point number
6. Print fixed point number
7. Linking C-routines to FF.

BR Mike