Lpc2000

John, you seem to have some of your facts wrong.

--- In lpc2000@yahoogroups.com, "John Heenan" <l10@...> wrote:
 > Fact 1. JTAG hardware operates independently. Nothing JTAG does,
 > including reset signals, will make the slightest bit of difference
 > unless the signals JTAG passes are not blocked and so are allowed to
 > act. Simple invertor and OR logic gates are perfectly adequate to
 > block signals.

Could you explain what your point is in the above paragraph please.  I read 
it a few times but cannot make out what you intend to point out.

 > Fact 2. The boot loader decides if CRP (Code Read Protection) is
 > enabled and so decides if it will allow signals from JTAG to act.

Wrong.  JTAG signals are enabled or disabled on reset by pulling specific 
GPIO pins low on boot [see notes on page 120 of 2294 user manual]. The boot 
loader *disables* it, and then at a later time enables it if it thinks that 
CRP is not enabled.

 > Fact 3. Even if CRP is an afterthought, so what if it works.

The problem is afterthoughts like this are more likely to fail.  In this 
case, where the hardware designers have not included security features, it 
is usually not possible to add this as an afterthought using 
software.  Indeed, if it works for LPC, then it would also work for the 
other ARM cores ... but ... one wonders none of the others are claiming 
they have CRP.

 > Fact 4. A lot of contradictory statements from Philips have been
 > quoted including an apology and clarification (P3) that parallel
 > programming uses either JTAG and/or ISP.

I noticed this too, and this is perhaps the main reason why I have rated 
Philips as not credible.

 > Fact 5. Use of an incomplete statement (C1); admitted to be in
 > a 'different context' that contradicts directly Philips apparently
 > later clarification in Fact 3. Reuse of C1 as C5.

When you examine evidence, there is the notion of "contemporary" evidence 
for which I give more weight, than further evidence in the process of 
defensive examination that contradicts it.  This is quite common practice 
in industry.

It is quite possible that had I posed the first question in relation to CRP 
issues, I would have got quite a different answer and the availability of 
parallel programming option would not have been raised at all.

 > John Heenan

Send instant messages to your online friends http://au.messenger.yahoo.com