Re: DP256 - DP512 Transition

Hi,

DP512 does not have EEPROM or is it XDP512??

I found this type of issues when my memory map was messed up and were  writing errounious data to te flash secure registers locking up the debugger.

So in order to reverse this, you need a generic clock/osc inserted or to be able to run the free P&E unsecure program in order to unlock the flash lockout.

To prevent this from happening again carefull  analysis needs to be done in order to make sure that the memory mapping was implemented correctly when moved from the DP256 to the DP512.

Hope this helps

Fred



  ----- Original Message -----
  From: Jonathan Masters
  To: 68HC12@...
  Sent: Monday, April 20, 2009 3:46 AM
  Subject: Re: [68HC12] DP256 - DP512 Transition





  jpdi wrote:
  >
  >
  > Hi, Jonathan,
  >
  > Is it DP512 or XDP512 ?
  > XDP512 vs DP256 has
  > - a different scheme for quartz than DP256. Maybe it's why you have speed
  > problem with your BDM, as the CPU run with it security clock, not quartz.
  > It's the first problem I had when moving from DP256 to XDP512.
  > - different page addresses (for me, with DP256, my pages begin at 0x30,
  > XDP512 at 0xE0), so I had to rewrite some code.
  > - some register addresses are different, so I had to rebuild my projects
  > with the "good" include file.
  >
  > I use ICC12 V7 and NoIce debugger.
  > ICC12V7 options are to generate "Linear S2 file", and NoIce load option is
  > "S2 have linear addresses".
  >
  > Hope this help.
  > Best regards.
  > Joel
  >
  > -----Message d'origine-----
  > De : 68HC12@... <mailto:68HC12%40yahoogroups.com>
  > [mailto:68HC12@... <mailto:68HC12%40yahoogroups.com>] De
  > la part de
  > Jonathan Masters
  > Envoyé : lundi 20 avril 2009 10:06
  > À : 68HC12@... <mailto:68HC12%40yahoogroups.com>
  > Objet : [68HC12] DP256 - DP512 Transition
  >
  > Hello All.
  >
  > I have recently started to upgrade an old board that uses DP256B's with
  > DP512. I managed to program the board a couple of days ago with no issues.
  >
  > Today, I ran into a problem with my application code - it couldn't write
  > to paged flash when downloading data from my PC app. This was quickly
  > identified as a problem with the bank select in my flash loading routine.
  >
  > However no matter how hard I try today, I cannot get a P&E Multilink to
  > load my original binary back into the chip. Most of the time I get "ELF
  > Loader Error - Unable to program 00C000-00C200". Sometimes this is
  > different - its all a bit flaky. I have tried two different pods, two
  > different boards with the same results. An original board with a DP256
  > works fine.
  >
  > I have tried using DP512 as the drivative in ICD12 as well as DP256 -
  > neither makes any differences.
  >
  > Finally, I find that communications is frequently lost with the board
  > after the programming failure. Thinking it could be a hardware problem I
  > have changed crystals and reset chips to no avail.
  >
  > I think I have had all these problems before and they just seem to go
  > away. Naturally they come back when under most pressure ;). My previous
  > questions to the group on similar topic related to Quad Core machines.
  >
  > Has anyone got any suggestions where I can start looking?
  >
  > Jonathan Masters
  >
  > ------------------------------------
  >
  > Yahoo! Groups Links
  >
  >
  Joel,

  Thanks, but no it is a 9S12DP512 (Mask 4L00M). I am more or less
  consistently now getting the error and losing connection
  to the device during programming.

  Jonathan.


 

[Non-text portions of this message have been removed]