[En-Nut-Discussion] Atmega 256 for Ethernut?

[Henrik Maier]

The stack handling problem should not be of a concern anymore as Nut/OS 
supports now a thread model where the thread stack can be kept in 
internal memory.

Should the ATmega256 show a similar problem to the AT90CAN128, one can 
use this feature to overcome it.

More of a concern will be the best way to incorporate an additional CPU 
into the Nut/OS source tree.

With more AVR CPUs on the horizon, the current model of checking against 
ATmega103, ATmega128 and AT90can128 does not make sense any more as the 
list of ifdef conditions will become almost endless.

Possible AVR CPU Nut/Os could support already now are:

Atmega103
Atmega32
Atmega64
Atmega128
Atmega256
at90can32
at90can64
at90can128
atmega256
atmega1281
atmega2561
atmega640
atmega1280
atmega2560
...

I rather suggest to introduce a capability model, where in avr.h for 
each known CPU certain capability macros are set. The goal should be, 
that with introduction of a new AVR cpu only avr.h needs to be changed.

Examples could be:

AVR_CAP_USE_TIMER0 (atmega103, atmega128)
AVR_CAP_USE_TIMER2 (at90can128, atmega256)
AVR_CAP_HAVE_SECOND_UART
AVR_CAP_HAVE_THIRD_UART
AVR_CAP_HAVE_CAN
AVR_CAP_HAVE_SECOND_TWI
AVR_CAP_HAVE_SECOND_SPI
AVR_CAP_I_HAVE_THIS_SPECIAL_FEATURE
...



Henrik


Jean Pierre Gauthier wrote:
>  Is there some news about ATMEGA256 implementation, scheduled date?
> 
> Thanks,
> Jean Pierre
> 
> -----Message d'origine-----
> De : en-nut-discussion-bounces at egnite.de
> [mailto:en-nut-discussion-bounces at egnite.de] De la part de Mark E. Scott Jr.
> Envoyé : jeudi 7 juillet 2005 18:27
> À : Ethernut User Chat (English)
> Objet : RE: [En-Nut-Discussion] Atmega 256 for Ethernut?
> 
> We are working with the mega2560 now.  I had to purchase the CodeVisionAVR
> compiler because GCC doesn't support it.  Hoping not to have to use it for
> long, but my guess was that even if AVR-GCC had support for it tomorrow, I
> wouldn't want to trust my life to it yet.
> 
> I'm not familiar with the stack handling bug of the AT90CAN128 so I can't
> say whether the same problem would exist.
> 
> We JUST got our new boards in with the 2560 populated.
> 
> Mark E. Scott, Jr.
> mscott at awfs.net
> AWS, Inc.
> 512-478-7727 x 122
> 
> -----Original Message-----
> From: en-nut-discussion-bounces at egnite.de
> [mailto:en-nut-discussion-bounces at egnite.de] On Behalf Of Harald Kipp
> Sent: Thursday, July 07, 2005 6:00 AM
> To: Ethernut User Chat (English)
> Subject: Re: [En-Nut-Discussion] Atmega 256 for Ethernut?
> 
> Brett,
> 
> Yes, but no decision about the final date. Probably late.
> 
> Did anyone tested this chip so far? The worst would be, it it contains the
> same stack handling bug like the AT90CAN128.
> 
> Harald
> 
> At 19:40 07.07.2005 +1200, you wrote:
> 
>>Harald
>>
>>Do you have plans to release a atmega256 Ethernut?  Any scheduled 
>>dates?  Please accept my apologies if this has already been answered.
>>
>>Thanks
>>Brett
>