Andrejs Cainikovs <AndrejsC@GlobalAutomationSystems.com> wrote:

Our boards flash driver is the same as in board/atmel/at91rm9200dk/flash.c
The only thing that was modified is the flash identification routine, as 
there was no support for AT49BV642D.

This chip is supported by the CFI driver (the ATNGW100 board relies on
that.)

The flash organization for this chip is the same as in AT49BV6416 which 
exists in the sources, so we haven't changed a lot.

This chip is not yet supported by the CFI driver. It needs a fixup due
to reversed erase region information, but since the CFI driver only
reads the low 8 bits of the ID, there's currently no way to tell it
apart from AT49BV642D, which doesn't need this fixup.

If someone can figure out how to read a 16-bit ID when appropriate, I
suspect we can get rid of quite a few board-specific flash drivers
(including the one for ATSTK1000).

Unfortunately, I don't have the time to look into this myself at the
moment.

Haavard