Debug issue

[LordMark]

Hi everyone.

I've been working for a long time on a board based on MCF52259, designed by me. I'm using the 26-pin BDM connector and USB programmer by P&Emicro. The board has always worked fine... till now.

I want to debug, so I set the breakpoint where I need, click on debug button, click it again to skip the debug of hardware initialization and I expect to reach the breakpoint I set. But what I get is the error message "Source text or disassembly not available" (as you can see in the attached screen). By stepping over every single line in the hardware initialization, I noticed that the error occurs at the calling of function "pll_init()". I don't know if it it important.

Anyway I obtained a new board, the same board and obviously the error doesn't occur. So, there must be an hardware problem that causes the old board to fail debugging. Please note that the old board lets me do all the other things, like reading, erasing, writing flash, running application, executing all tests successfully and so on.

Finally I checked electrically the lines between the connector and the MCF52259 and they are good.

What do you think?

Regards,
Marco 

[JeffB]

If possilble monitor pin 24 of the debugger header with a scope.  This is the system clock for the debugger.

The pll_init() function is probably setting up the internall pll to run at 80MHz then switching the oscillator from the internal 8MHz clock to the external crystal.

Step through the pll_init() code and see if this clock signal is stopping.  This will cause the debugger to croak.

[accutime]

I have the same problem with my 52559 board (still working on it)

I have been reading through forums and found the following :-

http://cache.freescale.com/files/soft_dev_tools/doc/ref_manual/USBMLCFESUM.pdf

There is a jumper inside the unit which selects the default as to whether communications with the target ColdFire processor should be synchronous to the PSTCLK signal (clocked) or not. The default jumper position (pins 2+3 close to the USB connector) selects synchronous communications. This is required by some devices (such as the 5272) which do not have on-chip synchronization circuitry. Communications can be set to be asynchronous to the PSTCLK (pins 1+2), in which case the PSTCLK signal will be terminated to ground with a 50 Ohm load to decrease EMI. The jumper should only be changed while the USB-ML-CF unit is not powered. Asynchronous mode may be useful (most ColdFire devices support it) in cases where the PSTCLK frequency is very high or the amount of noise emission (EMI) is a consideration.

http://www.pemicro.com/forums/lofiversion/index.php?t2798.html

I encountered the same problem. I continued to use the 48MHz crystal but set the PLL for 64MHz. The Multilink would work at that speed. But what I found was that if I set the PLL to anything other than 80MHz, I actually got only 48MHz. Is there an upgrade for the Multilink so it will work at 80MHz?

On the first comment I have not managed to open my USB-CW-ML - Has anyone else ?

But I do feel you problem is with the PLL.

It would be good to have some kind of statement from Freescale

Kind Regards

Paul H