[En-Nut-Discussion] RFC: "conflicts" statement in configurator scripts

[Harald Kipp]

Hi Ole,

On 5/25/2011 5:03 PM, Ole Reinhardt wrote:

> I just spend two days in hunting an abort exception right on boot time.

Sorry to hear this. Actually it was me who recently did some changes in
this area after I saw in a bug tracker report, that the second flash
controller was not handled in the EFC driver. Not sure anymore, with
which platforms I tested my new setup.


> In short: As it is just defined for the AT49BV chips, this macro is not
> redefined to the settings needed for the internal flash of some AT91
> chips.

I didn't get that. For AT91SAM7X, why can't you simply define this right
in the board configuration (.conf) to override the dev.nut default?


> FLASH_*** constants in at91_efc.c into EFC_*** constants.

Well, I always try to re-use exiting macros, otherwise we may end up if
tons of macro names. Although, in this case I agree, that there's a
potential conflict.


> I would suggest to add a "conflicts" statement (in oposit to the
> requires statement) into the configurator scripts that allows to define
> conflicting configuration options.

In other words, the "require" requires more than just an AND of
non-negated features. Definitely yes.


Regards,

Harald