"Knopf" <popsch@gmx.net> writes:

>> >   Does anyone know software profiling tools for microchip PICs? Can I
>> > hack this together with an ICD2 or do I really have to get an ICE?
>>
>> I'd recommend an I/O line and an oscilloscope.
>
> Ok. That's a feasible approach, however, the advantage of standard
> profiling software is, that you have all the data at once and can look
> into the 10 most expensive functions and optimize them.
>
> Hm. I might add a C macro for entering and leaving a function and it
> will measure a timestamp and export the data via the usart. Then I'll
> dump it into excel and create my charts. Is the cpp directive
> "__FUNCTION__" part of ansi C?

I don't think so, because the preprocessor does not understand
functions. C99 (and gcc) has __func__.

<http://gcc.gnu.org/onlinedocs/gcc/Function-Names.html#Function-Names>

-- 

John Devereux

> >   Does anyone know software profiling tools for microchip PICs? Can I
> > hack this together with an ICD2 or do I really have to get an ICE?
>
> I'd recommend an I/O line and an oscilloscope.

Ok. That's a feasible approach, however, the advantage of standard
profiling software is, that you have all the data at once and can look
into the 10 most expensive functions and optimize them.

Hm. I might add a C macro for entering and leaving a function and it
will measure a timestamp and export the data via the usart. Then I'll
dump it into excel and create my charts. Is the cpp directive
"__FUNCTION__" part of ansi C?

  Popsch

Knopf <popsch@gmx.net> wrote:
> Hi.

>   Does anyone know software profiling tools for microchip PICs? Can I
> hack this together with an ICD2 or do I really have to get an ICE?

I'd recommend an I/O line and an oscilloscope.

ttyl,

--buddy

Hi.

  Does anyone know software profiling tools for microchip PICs? Can I
hack this together with an ICD2 or do I really have to get an ICE?

Thanks,
  popsch