I'm in the process of moving from CW 7.2 to CW 10.5 after years of using 7.2. I've found there seem to be many ways to break things in CW 10.5 that I've stumbled upon. I've finally got a project set up the way I'd like and it builds great. I can use a task to erase and program my target (MCF55258) using a P&E USB-ML-CF Rev D. The program executes as desired but if I use the Debugger and allow the target to run with no expected breakpoint, the Debugger will fail in about 3 seconds saying it lost its connection with the target. I know in CW 7.2 I could let it run indefinitely and then break back in and suspend. Anyone have an idea what I might have done to get this behavior? (I assume this is an anomaly isn't it?)
Solved! Go to Solution.
Hello All,
Jorge seems to have come close to what I found the problem to be, or more correctly, what I did to eliminate the problem. It was a simple fix, since we merely removed the link between the earth ground lug and the negative terminal of the lab power supply. The only explanation we could come up with, was a ground loop was being created and it introduced noise when the power supply load was increased by activating an MCU output. We zeroed in on this issue by locating the section of code that caused the Debugger to Disconnect and eventually found a single statement. However, when we tried to connect a scope to monitor the BDM signals the problem stopped happening! That's why we ended up examining the power connections.
Regards,:smileyhappy:
Tom Michalski
Hello All,
Jorge seems to have come close to what I found the problem to be, or more correctly, what I did to eliminate the problem. It was a simple fix, since we merely removed the link between the earth ground lug and the negative terminal of the lab power supply. The only explanation we could come up with, was a ground loop was being created and it introduced noise when the power supply load was increased by activating an MCU output. We zeroed in on this issue by locating the section of code that caused the Debugger to Disconnect and eventually found a single statement. However, when we tried to connect a scope to monitor the BDM signals the problem stopped happening! That's why we ended up examining the power connections.
Regards,:smileyhappy:
Tom Michalski
Hi Tom,
I meet this issue now, how to fix it?
Hello Tom:
I have seen connection lost in cases when changing clock configuration (maybe the core too fast for the debugger tool) or when entering a low power mode. Even a watchdog enabled might cause the processor to reset and lose connection with the debugger
Regards!
Jorge Gonzalez
Hello,
Thank you for your post, however please consider moving it to the right community place (e.g. CodeWarrior Development Tools) to get it visible for active members.
For details please see general advice Where to post a Discussion?
Thank you for using Freescale Community.
Thanks for the feedback. I have moved it as you suggested.