[avrdude-dev] New pin definition data structure, updated ftdi_syncbb to support buff

[René Liebscher]

Hi,

as I already mentioned I checked in now the new pin definitions data 
structures.

All existing bitbang programmer should work as before as the convert the 
new data into the existing data fields. (See their initpgm functions.)

For ftdi_syncbb I use now the new format internally, so it uses now the 
pin numbers 0-based. Also it should now support the buff pins and 
inverted pins for all its supported pins. (This needs to be tested as my 
hardware has no need for a buff pin and any inverter.) Also there are 
currently more exit() in the code than I really want, but this will be 
changed next.

I guess avrftdi could be adapted very easy by changing the add_pins and 
set_pins functions (add_pin and set_pin might be removed). Another place 
to change is the pin setup in avrftdi_open. (Maybe looking at ft245r.c 
and its usage of the SET_BITS_0 macro and its check_pins function might 
help. )

BTW: I don't like the fact that in avrftdi the default pins are set if 
MOSI,MISO,SCK, or RESET does not fit the expectations. If some one has 
really connected the pins as he describes in the config file, than we 
should rather exit because of wrong connections, instead to change it by 
ourself and possible destroy the connected hardware. (Maybe he could 
still use synchronous bitbanging using ftdi_syncbb in this configuration.)

Also the add_pin(s) and set_pin(s) functions seem to use different 
return code for errors (1 vs. -1). Maybe we should introduce some 
generic defines in avrdude.h so it is really clear what the value for OK 
is, what values for error (any negative number? so you can code the 
reason in it?)


Kind regards

René


On 29.04.2013 20:51, René Liebscher wrote:
> Hi,
>
> I have here now a patch which introduces the new pin definition format 
> into the programmer structure.
>
> At the moment the config parser writes the new definitions and all 
> bitbanging programmers call in their initpgm
> function a function which creates the corresponding values in the old 
> pinno elements of the programmers structure.
>
> So all existing code should run as usual and we can start to convert 
> one programmer after another. If it uses the new format, we can remove 
> then the call to the converter function, and if the last programmer 
> removed it, we can also remove the old data field.
>
> May be you want have a look in the patch if there are any obvious 
> problems. Otherwise I will check in it the next days.
>
> I think all changes should be done before releasing version 6 as we 
> have changed already the config file format. When changing now the pin 
> numbers for some programmers from 1-based to 0-based this will result 
> in nobody having an old version 5 config file lying around which would 
> be parsed without the need to manually correcting it. (Also the new 
> programmers which would be 0-based (avrftdi,ftdi_syncbb,linuxgpio) 
> would all be new in version 6.)
>
>
> Kind regards
>
> René Liebscher
>
>
> On 05.01.2013 17:35, Hannes Weisbach wrote:
> > > You have the same behaviour for ftdi based prgrammers, where you 
> > > have to
> > > write 4 when you actually mean ADBUS3.
> > >
> > > I would prefer a much more general approach to this problem.
> > >
> > > You have to remember that there are not only single pins but also list
> > > of pins are possible for vcc and buf pin definitions. And these define
> > > their pins as 1<< pin1 | 1 << pin2 | ... . So setting the default to -1
> > > would set there all bits if there is no definition in the conf-file.
> > >
> > > I would like to have the pin definition as follows:
> > >
> > > typedef struct programmer_t {
> > > ...
> > >   unsigned int pinno[N_PINS];
> > > ...
> > > }
> > >
> > > changed to
> > >
> > > typedef struct programmer_t {
> > > ...
> > >   pin_def_t pinno[N_PINS];
> > > ...
> > > }
> > >
> > > with pin_def_t defined as:
> > >
> > > #define N_MAX_PINS 32 /* or 256 if GPIO is compiled in */
> > >
> > > struct {
> > >     uint32_t mask   [N_MAX_PINS/32];
> > >     uint32_t inverse[N_MAX_PINS/32];
> > > } pin_def_t;
> > >
> > > Then you have a single pin defined as:
> > > mask = { 0b0001000, .... }
> > > inverse = { 0b0000000, .... }
> > >
> > > an inverted pin as:
> > >
> > > mask = { 0b0001000, .... }
> > > inverse = { 0b0001000, .... }
> > >
> > > pin lists as:
> > >
> > > mask = { 0b0010110, .... }
> > > inverse = { 0b0000000, .... }
> > >
> > > and it would even be possible to define a mixed inverted pin list as
> > > wished in bug #37727 Add support for LM3S811 dev board as a programmer
> > >
> > > mask = { 0b0010110, .... }
> > > inverse = { 0b0010010, .... }
> > >
> > > However you had to change the code for the following programmer types
> > > avrftdi, ftdi_syncbb, par and serbb. (These are the only ones which
> > > currently use pin definitions in the config file.)
> > >
> > > I would also propose to change then pin numbers of ftdi based
> > > programmers to 0 based numbers in the config file. But leave them for
> > > parallel and serial port 1-based as they correspond to real pins at
> > > their connectors.
> > I think it's a good idea and would provide the necessary changes to 
> > avrftdi, if this proposal is generally accepted and you provide the 
> > definitions.
> >
> > Best regards,
> > Hannes