topic Re: Harder getting the CPU in debug mode once the external flash is programmed? in ColdFire/68K Microcontrollers and Processors

Harder getting the CPU in debug mode once the external flash is programmed?

s4w2099 — Mon, 27 Dec 2010 07:44:26 GMT

Well the subject pretty much says it all. I made a board with the MCF5249L MPU and everything works great. There is only a little concern about getting into debug mode using the PE Micro BDM Rev D.

If the flash is erased, CodeWarrior and PROGCFZ will connect to it fast and without any problems. However, if the flash has a program in it, I have to manually reset the board by pressing the reset button on board, then the BDM will connect without issues.

Whats really weird is that this happens ONLY when the external flash have been programmed, never once when the flash is erased. My flash device is an : SST39VF3201B in case that helps any.

Does anyone see a reason as to why this might be happening? I am completely lost

Thanks

Re: Harder getting the CPU in debug mode once the external flash is programmed?

TomE — Mon, 27 Dec 2010 08:02:14 GMT

When there's no code in the FLASH the CPU will most likely be getting into a "Fault-on-fault" state and halting. With code in the FLASH it is running. That's one difference.

What do you have JTAG_EN connected to? The manual implies this pin selects JTAG or BDM at any time, but also says "The JTAG_EN does not support dynamic switching between JTAG and BDM modes".

The pod may connect to this pin, check it. Do you have a pull-up or pull-down on this pin? What does the development board for that chip do?

Maybe the pod isn't able to drive CPU Reset properly? How do you have the pod's reset connected to the chip reset?

Tom

Re: Harder getting the CPU in debug mode once the external flash is programmed?

s4w2099 — Tue, 28 Dec 2010 10:41:07 GMT

Hello Tom,

Thank you for your response. I double checked everything and my JTAG_EN (TEST0) is pulled high to VCC so, it is not dynamically switched. I did notice that I accidentally populated a 330 Ohm resistor for !RESET input pin. Despite this the debugger seems to be pulling it low without a problem.

Just to discard any possible problems I installed the latest version of the debugger cable driver. I have not had problems after that either with the flash deleted or full....

I know its weird.... I will keep developing the software and keep an eye on it. Will surely let you know in this thread if this thing starts happening again.

Thanks!