Re: [avrdude-dev] avrdude on Windows

[E. Weddington]

On 13 Feb 2003 at 11:35, Brian Dean wrote:

> On Thu, Feb 13, 2003 at 09:24:37AM -0700, E. Weddington wrote:
> 
> > I suspected as much that this would be the best way to go. I'm
> > currently in the middle of writing a ppiwin.c file that provides the
> > 8 functions previously talked about. Brian, I assume that we would
> > want to have one ppi.h that provides the prototypes?
> 
> Yes, here's my current copy of ppi.h:
> 
> #ifndef __ppi_h__
> #define __ppi_h__
> 
> #include "config.h"
> 
> /*
>  * PPI registers
>  */
> enum {
>   PPIDATA,
>   PPICTRL,
>   PPISTATUS
> };
> 

Suggestion:

enum {
  PPIDATA = 0,
  PPICTRL = 1,
  PPISTATUS = 2
};


> 
> int ppi_getpinmask(int pin);
> 
> #if !defined(ppi_claim)
> #  define ppi_claim(pgm)
> #endif
> 
> #if !defined(ppi_release)
> #  define ppi_release(pgm)
> #endif
> 
> int ppi_setpin    (int fd, int pin, int value);
> 
> int ppi_getpin    (int fd, int pin);
> 
> int ppi_set       (int fd, int reg, int bit);
> 
> int ppi_clr       (int fd, int reg, int bit);
> 
> int ppi_getall    (int fd, int reg);
> 
> int ppi_setall    (int fd, int reg, int val);
> 
> int ppi_pulsepin  (int fd, int pin);
> 
> #endif
> 

Realize that 

ppi_pulsepin
ppi_getpin
ppi_setpin
ppi_getpinmask

is going to be the same code no matter what the platform. Why 
duplicate this?

I'd still rather see it contain:
ppi_open
ppi_close
ppi_set
ppi_clr
ppi_get
ppi_toggle
ppi_setall
ppi_clrall

As these are the functions that are different between platforms.
 
> These represent the functions and enums needed by my new 'par.c' file.
> I've got the split done and working.  Shall I go ahead an commit this?
> Is it going to cause anyone too much disruption (mostly thinking of
> Ted's autoconf work)?  If there are any objections, let me know soon.
> 

Give Ted a couple more hours to weigh in. He's an hour behind me.

Eric