GDI Protocol fatal error

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

GDI Protocol fatal error

1,984 Views
jtask
Contributor I

Hi all,

 

I am using a MC9S08QE32 processor along with CodeWarrior 10.6 and a P&E Multilink Universal for debugging.  I can download the program to the chip and run the program through the debugger, but after some time the program stops and I get a pop-up error: "GDI Protocol fatal error: The Debugger is not connected to the PE device" and then I can no longer reconnect to the device and I get the problem occurred pop up: "... Failed to resume target process."  I then have to restart Codewarrior (ending the DE process doesn't always do it) to be able to reconnect to the target board.

 

Some things that I have tried that have not fixed the problem:

  • Upgraded Codewarrior to 10.6
  • Verified stable power supply to board
  • Used different boards (custom boards with same processor) and I still get the problem on all of the boards
  • Disabled COP Watchdog

 

This problem makes debugging very difficult and any help or suggestions at all is greatly appreciated. 

 

Thanks and please let me know if more information is necessary to help.

Labels (1)
0 Kudos
3 Replies

1,009 Views
BlackNight
NXP Employee
NXP Employee

Hi,

I had a problem a while back with overlapping memory (Problem Occurred: Flash programming with overlapping memory), but this was somehow different as it failed while downloading?

The other tips I have:

- I had once a bad USB cable: replacing the cable solved the problem. I had problems as well with bad USB connectors (check the USB solder/connections on the Multilink).

- USB 3.0 port? I have seen a lot of problems with that. If you have USB 3.0, try a different (USB 2.0) port on your machine.

- Do you have a noisy environment? I had once a problem that a motor nearby was injecting noise into the debug cable: shortening the BDM cable and/or shielding it from the motor solved it.

- Then I had once a bad/flacky Multilink: using a different unit identified the problem.

I hope that helps,

Erich

1,009 Views
jtask
Contributor I

Hey Erich,

Thanks for your response and the time that you have put into your website.  I have spent a lot of time over the past few years on your website and it has helped me out on numerous occasions.

This time I seemed to have fixed the problem by moving the workspace over to my local machine, instead of on the company's network.  Maybe the network was going in and out and I kept losing a connection?  Not sure entirely but it is working now.

Thanks again

0 Kudos

1,009 Views
BlackNight
NXP Employee
NXP Employee

Hi James,

glad to hear that :-).

Interesting point about having the workspace on a network (drive?). That would explain it a lot, because Eclipse does a lot of writing/reading from the .metadata. So if there is enough latency on your network, it indeed could create such issues. I would never have thought about using a network drive for the workspace. It is ok to use it for the projects itself (although not ideal).

But good to know this, so I try to remember that case if someone else has the same problem. So many thanks for reporting back what was the (likely) cause and solution!

Thanks,

Erich

0 Kudos