[sdiy] self-modifying code was: Re: My latest project

cheater cheater cheater00 at gmail.com
Thu Feb 18 11:07:37 CET 2010


Neil,

> dsPICs, like most other smaller microcontrollers, and flash based, and maybe
> Harvard as well, so self-modifying code is pretty much a no-no in this
> situation (unless you have a *LONG* time to wait for the flash sector to
> erase before reprogramming it, not mentioning the flash wear issue).
ah, fair enough!

Out of curiosity: how difficult is it to keep program data in RAM?

And similarly.. how does SMC work with FPGA's and its family? Is it
even possible?

Thanks
D.

On Thu, Feb 18, 2010 at 00:45, Neil Johnson <neil.johnson97 at ntlworld.com> wrote:
> Hi,
>
> cheater cheater wrote:
>>
>> Dereferencing pointers is different from modifying the code in that it
>> frees up your stack/registers.
>
> dsPICs, like most other smaller microcontrollers, and flash based, and maybe
> Harvard as well, so self-modifying code is pretty much a no-no in this
> situation (unless you have a *LONG* time to wait for the flash sector to
> erase before reprogramming it, not mentioning the flash wear issue).
>
> Shame it doesn't have conditional execution a la ARM.
>
> Writing code with no if's is very tedious and very sensitive to design
> changes and bugfixes.  Have you looked at a round-robin schedule sitting on
> a timer?  As long as your longest execution path is less than the cycle time
> it'll look after itself.
>
> Neil
> --
> http://www.njohnson.co.uk
>
>
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>




More information about the Synth-diy mailing list