Re: [avrdude-dev] New pin definition data structure

[René Liebscher]

Hi,

it should be zero initialized in the programmer structure as this is 
done in pgm_new (pgm.c), later the parser sets only the bits coming from 
the config file.

René



On 05.05.2013 16:33, Hannes Weisbach wrote:
> Am 05.05.2013 um 16:00 schrieb René Liebscher:
>
> > Hi,
> >
> > I added some more functions for working with the new pin definitions. 
> > Especially there is now a generic function for checking for valid pin 
> > definitions to be used by any bitbanging programmer.
> >
> > It is currently only used in ftdi_syncbb (search for pins_check in ft245r.c), 
> > but might also used in avrftdi and others.
> >
> > For avrftdi I have attached an example how to it (after you have changed the 
> > pin numbering there to be 0-based). Also one could remove then the pin 
> > validation code from the add_pins function.
> Hi,
>
> is there a reason why pindef_t.mask and pindef_t.inverse are not zero 
> initialised? For avrftdi it would make things really easy, because all pins 
> could be OR'ed (minus MISO) to build the data direction mask.
>
> Best regards,
> Hannes