Re: [avr-gcc-list] -mmcu=avrxmega confusion

[Georg-Johann Lay]

Don Kinzer schrieb:
> All xmega devices have a dedicated boot section *in addition* to the
> stated amount of Flash.  In particular, this means that the xmega64xxx
> devices all have slightly more than 64K of Flash and therefore have
> the ELPM instruction.  For the same reason, the xmega128xxx and higher
> devices all have more than 128K of Flash and therefore need the EIND
> register.  Consequently, the avrxmega4 and avrxmega5 should both have
> ELPM=1 and the avrxmega6 and avrxmega7 devices should have EIND=1.

Thanks, that *in addition* is the missing bit.
Up to now amega is purely theoretical construct for me... no even
a reasonable simulator is available to run the test suite:

* runs from linux command line
* runs /fast/
* has built-in magic like avrtest so that printf ("foo") will print
  "foo" to the host console

> avrxmega2: ELPM=0, EIND=0, RAMPD=0
>     atxmega16a4
>     atxmega16d4
>     atxmega16x1
>     atxmega32a4
>     atxmega32d4
>     atxmega32x1
>
> avrxmega4: ELPM=1, EIND=0, RAMPD=0
>     atxmega64a3
>     atxmega64d3
>
> avrxmega5: ELPM=1, EIND=0, RAMPD=1
>     atxmega64a1
>     atxmega64a1u
>
> avrxmega6: ELPM=1, EIND=1, RAMPD=0
>     atxmega128a3
>     atxmega128b1
>     atxmega128d3
>     atxmega192a3
>     atxmega192d3
>     atxmega256a3
>     atxmega256a3b
>     atxmega256a3bu
>     atxmega256d3
>
> avrxmega7: ELPM=1, EIND=1, RAMPD=1
>     atxmega128a1
>     atxmega128a1u
>
> Don Kinzer

Thanks, now I undestand the layout, and everything is perfectly
reasonable as-is (except some source comments in avr-gcc).

Johann