amforth-devel

[Amforth] AmForth JTAG support

[Enoch <ix...@ho...>]

Hello AmForth-ers,

*Background* A brief visit to Atmel's Studio while debugging the wlscope
("wordlist scope") patch convinced me that this magnificant tool is not
really good at Forth level debugging as it has no concept of the Forth
VM.

*Idea* Let us take advantage of Atmel's documented JTAG protocol (e.g.,
AVR067: JTAGICE mkII Communication Protocol) and some reverse engineered
data (from AVaRICE) to explore the possibility of enhancing
amforth-shell.py with several commands that set VM breakpoints and
examine data unintrusively.

*Needed* People with Python and C skills to show interest and share the
burden.

Did anyone attempt this approach before? Any interest?

Thanks, Enoch.

Re: [Amforth] AmForth JTAG support

[Matthias T. <mt...@we...>]

Hi Enoch,

> *Background* A brief visit to Atmel's Studio while debugging the wlscope
> ("wordlist scope") patch convinced me that this magnificant tool is not
> really good at Forth level debugging as it has no concept of the Forth
> VM.

You dig deep into the inner workings of amforth. A rather dark place. A
casual programmer is seldom seen in this area.

> 
> *Idea* Let us take advantage of Atmel's documented JTAG protocol (e.g.,
> AVR067: JTAGICE mkII Communication Protocol) and some reverse engineered
> data (from AVaRICE) to explore the possibility of enhancing
> amforth-shell.py with several commands that set VM breakpoints and
> examine data unintrusively.

The idea sound crazy enough to follow it. Its only beyond what I can do....
I don't know whether the amforth-shell is the right place to start, it may 
need a more modular design.

Keep in mind that some atmegas do not have JTAG (they have e.g. Debug Wire or
nothing at all).

Matthias

Re: [Amforth] AmForth JTAG support

[Enoch <ix...@ho...>]

Hello Matthias,

Matthias Trute <mt...@we...> writes:

> Hi Enoch,
>
>> *Background* A brief visit to Atmel's Studio while debugging the wlscope
>> ("wordlist scope") patch convinced me that this magnificant tool is not
>> really good at Forth level debugging as it has no concept of the Forth
>> VM.
>
> You dig deep into the inner workings of amforth. A rather dark place. A
> casual programmer is seldom seen in this area.
>
>> 
>> *Idea* Let us take advantage of Atmel's documented JTAG protocol (e.g.,
>> AVR067: JTAGICE mkII Communication Protocol) and some reverse engineered
>> data (from AVaRICE) to explore the possibility of enhancing
>> amforth-shell.py with several commands that set VM breakpoints and
>> examine data unintrusively.
>
> The idea sound crazy enough to follow it. Its only beyond what I can do....
> I don't know whether the amforth-shell is the right place to start, it may 
> need a more modular design.
>
> Keep in mind that some atmegas do not have JTAG (they have e.g. Debug Wire or
> nothing at all).

That's one of the reasons why such a project is beyond a single
programmer reach. I just have the JTAG Mk II to play with while Atmel is
pushing Mk III (*), there's the debugWIRE to handle, etc.

Within a few days I hope to find the time to conclusively decide about
this project feasibility or if it requires an AmForth kernel heck to
succeed.

Thanks again for the encouragement!

Regards, Enoch.

(*) IIRC the AVRDUDE / AVaRICE guys are in the process of reverse
engineering this Mk III protocol. 









>
> Matthias
>
>
> ------------------------------------------------------------------------------
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_d2d_mar