Debugging with Hiwave and PEmicro BDM

[Shugie]

I was wondering what the limits are to using the Codewarrior compiler with the PEmicro BDM connected to the chip I'm trying to debug. I'm trying to debug a chip that fails much more often than other chips in the same environment. When the chip fails it give an error over the BDM that the hardware may have reset. Is there a way to get the last 10-20 register values/ changes in memory dumped to a file or just displayed? I would like to do this to compare the values at the different stages right before failure to what happens in a normal chip during those stages. Also being able to see individual pins would be nice, but nowhere in the manuals has this been mentioned as possible. I don't think that stepping through the assembly codes is going to work because the failure seems to be based on environmental changes, but I have to test this more.

[Shugie]

The power supplies look ok. When the micro fails it starts throwing out signals on our outputs that are consistent from failure to failure. I'd like to see what is tripping in software to output these values over and over. Thanks for the advice.

[UcTechnoGeek]

I use the BDM from Witztronics.com, but the P&E BDM should work the same.

You need to enable "trace" (under the BDM menu in the debugger).

Then set some Trigger points.  You have multiple options and combinations with two triggers,  trigger A and B.  You can trigger on address read/writes, memory read/writes (Ports are just memory locations), ect...   You can even set up complex triggers; ie. If trigger A then trigger B...

Find a combination of triggers that will stop the processor when it starts throwing out signals and you can look over the trace buffer to see what it was doing before this happens.

You may have to learn how to use the triggers first, but it will be well worth it in the end.  This is what debuggers are for.

uCTechnoGeek

[BlackNight · Accepted solution]

I had something like you describe a while back. After a long search it turned out that the board power supply had stability problems (power dropped out randomly). I would check first the power supply (e.g. exchange it) and check all the board/chip power related aspects first.