Dan and Mark,
Thanks for talking about this a little more and making
it a much clearer for me. I do have VECTOR CATCH
enabled. I am going to keep my fingers crossed. :-)
--- Dan Malek <dan at embeddededge.com> wrote:
>
> On Jan 26, 2005, at 12:33 PM, Mark Chambers wrote:
>
> > P.S. I've read the BDI manual - I don't see it.
> My manual
> > says the BDI can only set one hardware breakpoint,
> and
> > stuff like VECTOR CATCH assumes valid memory.
>
> The BDI2000 lets you set as many hardware
> breakpoints
> as the processor supports. The VECTOR CATCH is
> equivalent
> to setting of the bits in the DER. I don't know
> where you
> determined it requires valid memory, as it catches
> the vector
> before it starts to execute the instructions in the
> trap handler.
> I don't know the COP details, there may be versions
> that don't
> have sufficient hardware support, and in that case
> it would
> have to be implemented by stuffing breakpoints into
> all of
> the vector tables, but that is just speculation on
> my part and
> have never seen that done. I've used VECTOR CATCH
> on
> 603 cores and had it trap properly before memory was
> initialized.
> Using VECTOR CATCH seems to have no effect on
> setting
> other breakpoints.
>
> Yes, you can manually set bits in the DER and get a
> different
> control over the debugging environment, but you have
> to
> be conscious of what the BDI2000 is also trying to
> do with
> that and other debug registers, too :-)
>
>
> -- Dan
>
>
__________________________________
Do you Yahoo!?
Yahoo! Mail - Easier than ever with enhanced search. Learn more.
http://info.mail.yahoo.com/mail_250
