On Mon, Jun 28, 2004 at 01:09:50PM +0200, Jan-Erik Soderholm wrote:
> Kyrre Aalerud wrote :
>
> Hi !
> I know this whole thread is about bootloading, but...
>
> > Reason is safety.  My loader will be installed in a chip that
> > will later be molded in epoxy.  If the bootloader (or jump to it)
> > should fail, the entire circuit would be a brick, including some
> > expensive sensors and high current FET's.
>
> Now, did you say why you could not either :
>
> 1. Make sure you don't have to update the firmware after the
>    module is moulded. This would require quite a bit of testing and
>    verificaton, but you'll have to do that with your bootloader anyway.
>
> 2. Since you will have a number of cables/connectors on your
>    module anyway (not ?),

Um, not. The whole board is potted in epoxy with a radio transceiver.
Kyrre, just curious as to if power is internal or external to the potted unit?

>  why not also design in an ICSP
>    cable/connector ? That way you could always flash the
>    module with a fresh HEX file.

See above.

>
>
> And besides, IMHO, this isn't a generic bootloader, it's a
> a more or less application specific/integrated "feature" to be
> able to update *this* specific firmware.

Bingo.

> I'm not saying that this
> is a bad thing, OTOH this might very well be better, in particular
> if you are not able to re-load the bootloader if something
> goes wrong. But with the limitiations (pointed out by Rob H.)
> it's close to useless for most other (generic) uses.

Bingo again. But my belief is that you have to take baby steps. A working
bootloader is a better platform for improvement than starting from nothing.

That's why I've taken great interest in both Kyrre's and Tato's bootloaders
of late. They can be starting points to get to the general purpose
functionality that is needed for other projects.

BAJ

--
http://www.piclist.com#nomail Going offline? Don't AutoReply us!
email listserv@mitvma.mit.edu with SET PICList DIGEST in the body