BDM trouble on 9S12XD256 with CW12_V4_6

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

BDM trouble on 9S12XD256 with CW12_V4_6

Jump to solution
3,930 Views
Bee
Contributor III
I'm using P&E's MultiLink cable on a 9S12XD part and got the latest CW 4.6 downloaded and installed. It started working as expected for about 3 days. Then, for some unknown set of events, when I got into debug and press the Start/Continue icon, it runs for about 1 sec and stops at an instruction. Pressing the icon does the same: runs a few instructions and stops. By 'stops' I mean it seems to hit a breakpoint or some kind of internal trigger that I haven't set. I can't find the cause. Tried reinstalling it, and the same thing happens. Have not tried to completely uninstall and download again because that takes a long time.

Does anyone have a hint what is causing this behavior? Should I be asking on the hardware forum?

Thanks.
Labels (1)
Tags (1)
0 Kudos
1 Solution
810 Views
Bee
Contributor III
It was running fine until sometime last week. There is a 1K pullup on the Reset# line. I don't use Watchdog. No stack messing. Stops after a few instructions at a time so it can't be stack overflow.

Unhandled interrupt might be a place, but shouldn't I see it go to the vector or PC=FFFF?

Pod should be OK (guessing) because I can load code and erase, etc.

Thanks.

View solution in original post

0 Kudos
4 Replies
810 Views
Lundin
Senior Contributor IV
That usually happens when the BDM loses the contact with the target or the target's program counter. The most common cause is when the target gets a reset.

The most likely causes of that behavior:

- Watchdog reset.
- No pull-up on the reset line of the PCB.
- A bug in the code. Array out-of-bounds, pointer bug etc.
- Asm code messing around with the stack.
- Asm code jumping to places it shouldn't jump to.
- Stack overflow.


Less likely:

- Clock/oscillator problems.
- Unhandled interrupts.


Unlikely:

- Broken BDM-pod.

Message Edited by Lundin on 2007-10-17 12:54 PM
0 Kudos
811 Views
Bee
Contributor III
It was running fine until sometime last week. There is a 1K pullup on the Reset# line. I don't use Watchdog. No stack messing. Stops after a few instructions at a time so it can't be stack overflow.

Unhandled interrupt might be a place, but shouldn't I see it go to the vector or PC=FFFF?

Pod should be OK (guessing) because I can load code and erase, etc.

Thanks.
0 Kudos
810 Views
Alban
Senior Contributor II
Hi Bee,

What does the Command Window (bottom right by default) say ?
It's not a problem of CodeWarrior but of your code and/or connection to the target.

Cheers,
Alban.
0 Kudos
810 Views
Bee
Contributor III
The Command window says: DBG Trigger occurred.

I don't know how to set/unset trigger, so I'm not sure how this got there. What do I do to turn it off?
0 Kudos