>>Walter,
>>
>>Forgive my ignorance, but I am not sure how a C compiler can target a
>>device with no SP (but they claim they have a 1 level subroutine call).
>>How can you handle function params, and recursive functions. You
>>apparently need to emulate the SP using other registers - this sounds
>>inefficient?
>>
>
>
> "Grimme acknowledged that Freescale's strategy is similar to one proposed by
> Atmel Corp. (San Jose, Calif.), which has developed compatible AVR
> microcontrollers across the 8- and 32 bit market."
That was not what I took from the Atmel press releases - has anyone
seen a "Compatible Avr-32" ?
The are certainly not binary compatible, and pin compatible was never
hinted at..... ?
>
>>Is Freescale on drugs?
>>
>
>
> Looks like they are trying to go the original ATtinyAVR approach
> which was abandoned by Atmel 3 years ago...
It will be interesting to watch.
If they _can_ do even a vanilla C, that will make a significant difference.
At the very bottom end, these have a relatively large
code space of 16K - I have not seen RAM figures yet.
Will these devices have on chip debug ?
-jg
Reply by Walter Banks●March 14, 20062006-03-14
The RS08 C compiler handles multiple levels of subroutine calls. Subroutine
nesting beyond 1 level of subroutines are part of the design.
We have implemented several compilers without data stacks that support
full parameter passing and recursive functions.
In non recursive functions parameter access is faster than parameters on a
stack frame. In the C6808 we do recursion detection (including multi link
recursion) and only create stack frames when recursion is needed. The
RS08 C compiler supports recursive functions. Function setup for
recursive functions is marginally more expensive than S08 stack frames
and parameter access is similar.
I have ported and bunch HC08 and S08 C applications to the RS08
by just replacing part specific header files and code changes for part
specific I/O devices.
Walter Banks, Byte Craft Limited
http://www.bytecraft.com
Eric wrote:
> Walter Banks wrote:
> > C compiler support for the RS08 is in the next release of Byte Craft's
> > C6808.
>
> Walter,
>
> Forgive my ignorance, but I am not sure how a C compiler can target a
> device with no SP (but they claim they have a 1 level subroutine call).
> How can you handle function params, and recursive functions. You
> apparently need to emulate the SP using other registers - this sounds
> inefficient?
>
> Is Freescale on drugs?
>
> Eric
Reply by Ulf Samuelsson●March 14, 20062006-03-14
> Walter,
>
> Forgive my ignorance, but I am not sure how a C compiler can target a
> device with no SP (but they claim they have a 1 level subroutine call).
> How can you handle function params, and recursive functions. You
> apparently need to emulate the SP using other registers - this sounds
> inefficient?
>
"Grimme acknowledged that Freescale's strategy is similar to one proposed by
Atmel Corp. (San Jose, Calif.), which has developed compatible AVR
microcontrollers
across the 8- and 32 bit market."
> Is Freescale on drugs?
>
Looks like they are trying to go the original ATtinyAVR approach
which was abandoned by Atmel 3 years ago...
--
Best Regards,
Ulf Samuelsson
This is intended to be my personal opinion which may,
or may bot be shared by my employer Atmel Nordic AB
Reply by Eric●March 14, 20062006-03-14
Walter Banks wrote:
> C compiler support for the RS08 is in the next release of Byte Craft's
> C6808.
Walter,
Forgive my ignorance, but I am not sure how a C compiler can target a
device with no SP (but they claim they have a 1 level subroutine call).
How can you handle function params, and recursive functions. You
apparently need to emulate the SP using other registers - this sounds
inefficient?
Is Freescale on drugs?
Eric
Reply by Walter Banks●March 14, 20062006-03-14
C compiler support for the RS08 is in the next release of Byte Craft's
C6808. Demo's ,C6808 updates and design reference material will be
available on our website as soon as the technical details of the RS08 have
been released.
Walter Banks, Byte Craft Limited
Mike Silva wrote:
> The big mistake here, is that Microchip are staking a chunk of their
> credibility, on a flimsy patent that they _have_ to hope never actually
> reaches open court. What was Steve thinking ?