USB TAP BDM has encountered a problem.  Failed to resume target process.  Bad arguments passed to function.

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

USB TAP BDM has encountered a problem.  Failed to resume target process.  Bad arguments passed to function.

4,381 Views
josephclifton
Contributor I

We had a student group get the following error when trying to program the MC56F8037 eval board:  "Lab7_MC56F8037_Internal_PFlash_SDM_Freescale USB TAP BDM has encountered a problem.  Failed to resume target process.  Bad arguments passed to function."   After that error, the eval board became unusable.  We tried programming after that on different PCs, barebones program, etc. and with different USB Taps.  It would not accept any program and always gives the same error.  We had this happen on 3 boards.  I have been giving this same lab for 4 or 5 years and haven't seen anything like this.  I read the post where someone had a similar issue ( https://community.freescale.com/thread/302492), but the "fix" was to restart the PC.  We did that and also tried different PCs and different programs and different USB Taps. So it appears as if downloading their program onto a board rendered the board unusable.  I should note that we had the Motor Control board attached.  The Motor Control board still works fine.  The fact that this happened for 3 different MC56F8037 eval boards when trying to program seems strange. Was wondering if the ever came up before.  We are running Windows 7.   I was hoping that somehow they put the microcontroller in some mode that we could do something to recover from, but I know that is a long shot.  Has anyone had something similar to this happen (other than the link shown)?

Labels (1)
0 Kudos
39 Replies

2,230 Views
johnlwinters
NXP Employee
NXP Employee

Hi Joseph,

Please try upgrading the tool you are using to program the DSC since you are using Windows 7.

Please follow the suggestions in this article:

Using the CodeWarrior TAP with the DSC

Also, please read off all the numbers on your USB TAP to make sure it is correct (the first one that did not work.).

Also, BDM is not supported by these devices.

0 Kudos

2,230 Views
josephclifton
Contributor I

Hi John,

We have several boards and several USB TAPs. All the rest work just fine. It just went bad for one group. The other 6 groups and the many before them was fine. I am not sure we can upgrade, since we have the University license. Unless we can get that under the University license also. What is it that the upgrade will allow us to do that the other doesn’t?

Also, please read off all the numbers on your USB TAP to make sure it is correct (the first one that did not work.).

The USB Tap works just fine. We can take another project and another board of the exact same type and it programs just fine. So we know the USP Tap is fine. It was their particular project that when programmed onto a board that caused the problems. I will send the numbers on Monday.

Also, BDM is not supported by these devices.

I am not sure what you mean by this.

Thanks

Joe

0 Kudos

2,230 Views
johnlwinters
NXP Employee
NXP Employee

Hi,

It is likely that the problem project reprogrammed the JTAG signals to be GPIO instead. I suggest you get help from the technical information center on how to get control of the devices. You can ignore the comment about BDM for now.

Best Regards,

John L. Winters

Senior Applications Engineer

AMR/EU MCU & MPU AE Group

Freescale Inc.

2100 Elliot Road

This email and any associated attachments have been classified as:

Freescale Confidential and Proprietary

0 Kudos

2,230 Views
josephclifton
Contributor I

I dug around in the technical Info Center and could not find what I needed. I think that maybe they locked the flash. The data sheet: http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=56F803X&tab=Documentation_Tab&pspll=1... Sheets&fpsp=1&linkline=Data Sheets

section 7.2 discusses this. It indicates how to get out of it, but it appears to refer to an older version of CodeWarrior. I have Eclipse-based CodeWarrior for MCU Version: 10.2.

I searched the program and the help and the web and your site for how to do something similar to what it says in 7.2.3 and can’t find anything. I also did a lot of searching to see if I could issue JTAG commands directly via Code Warrior but didn’t find anything. I also tried to see if there was a way to erase the flash from the version of Code Warrior I have and I can’t find that either.

So please let me know how I can do this (or if I can do this) using the Code Warrior version I have.

Thanks.

0 Kudos

2,230 Views
johnlwinters
NXP Employee
NXP Employee

I just noticed that your CodeWarrior, 10.2, is one of the early buggy versions.

Please upgrage to CodeWarrior 10.6 where this issue can be supported.

0 Kudos

2,230 Views
josephclifton
Contributor I

Upgrading to 10.6 won't "fix the eval boards". Over the weekend, I did install 10.6 and made a small barebones to try to program the "broken" boards and also see if 10.6 had any additional "tools" that might help "unlock" the boards. The boards in question won't program.

Just to be clear, we have other eval boards, exactly the same type, that program fine with 10.2. There is something in the project I sent that has a problem. Converting to 10.2 will most likely mask that.

Also, we were given special permission to put 10.2 on the labs with the "freebie" license. Can we get the same permission for 1.6?

0 Kudos

2,230 Views
johnlwinters
NXP Employee
NXP Employee

Yes the same free use is allowed for small projects using 10.6.

The zip file did not open for me.

Best Regards,

John L. Winters

Senior Applications Engineer

AMR/EU MCU & MPU AE Group

Freescale Inc.

2100 Elliot Road

This email and any associated attachments have been classified as:

Freescale Confidential and Proprietary

0 Kudos

2,229 Views
josephclifton
Contributor I

Good! We will switch to the new one next year.

What is the size of the ZIP? Maybe it got partially stripped by the emailer? The file should be 555,732 bytes.

When I look at it, everything is relative to the Lab7_old. The long names are what CodeWarrior puts into the project. The worst violators are the Eclipse .metadata .plugins

0 Kudos

2,228 Views
johnlwinters
NXP Employee
NXP Employee


Capture.PNG

0 Kudos

2,229 Views
josephclifton
Contributor I

I have attached it as 2.zip

That is as short as I can get it.

If you extract to c:\2

Then that is as short as you can get it.

0 Kudos

2,229 Views
TICS_Fiona
NXP Employee
NXP Employee

I opened your project in CodeWarrior v10.6, and the connection settings in debugger is not configured, please see the screenshot in below:

connection.png

Would you please create a new project for 56F8037 from <File> menu -> New->Bareboard project, can CodeWarrior v10.6 program with this project into your board?

0 Kudos

2,229 Views
josephclifton
Contributor I

Hi Xin,

Would you please create a new project for 56F8037 from <File> menu -> New->Bareboard project, can CodeWarrior v10.6 program with this project into your board?

Tried it – multiple times. Please note that I had tried several times to program the boards with a barebones project and a small LED blinking project, and nothing worked. We have 10 boards. I can program the others just fine with a barebones project. It is just those programmed with the project I sent you that don’t work. Once they were programmed with it, they would no longer take any program. I tried with an older (non-eclipse-based) IDE. I tried with your newer 10.6 version. I tried with 10.2, with a new barebones project and with a small little light-blinking project from last year. All of these worked on the “good boards” but not on the ones that were first programmed with what I sent you.

Did you try programming a 56F8037 eval board with the project I sent you? The students in that group went through 2 boards. I had assumed they had some bad serial connections and it was something they were doing with the hardware. I brought them into my office, made sure all the connections were okay Then I tried programming another board. Same thing happened as happened to them. The board would no longer take a program. Please note that I have 7 other groups and everything worked fine for them and fine when they came to demo their projects. The 3 boards “became bad” when programmed with the project I sent you. The other boards programmed by the other groups are just fine.

Also please note that when programmed the “fatal” times, the board was connected to the Freescale Motor daughter board. I think there is something in that project that is doing something to the boards, but I cannot tell what it is. So if you tried to program an eval board connected to a daughter Motor controller board and got the same result, then we could know for sure.

Just to be clear, we are using the 56F8037 Evaluation Module (MC56F8037EVM) connected to a APMOTOR56F8000. I have done this lab for 4 or 5 years. We switched to the Eclipsed-based 10.2 IDE last year. I have had many groups do it. This is the first time any group has produced a project that has done this to a board.

I have attached the lab I use in case you want to see it.

If you wanted to discuss this over the phone, we could attempt to arrange a time.

Last note: I had messed around with the target connections on the project I had sent, trying to make something work.

Thanks,

Joe

0 Kudos

2,229 Views
TICS_Fiona
NXP Employee
NXP Employee

Since you have several boards and several USB TAPs.  All the rest work just fine. It does not look like to be a CodeWarrior or  USBTAP problem.

Could you please supply your 56F8037’s reversion, or date codes which is printed on the mask.

The Lock/Unlock operation will damage 56F8037 of reversion A. This is a known problem and fixed in higher reversion with date codes date 0701.

Fiona Kuang

Technical Information & Commercial Support

-----------------------------------------------------------------------------------------------------------------------

Note: If this post answers your question, please click the Correct Answer button. Thank you!

-----------------------------------------------------------------------------------------------------------------------

0 Kudos

2,229 Views
josephclifton
Contributor I

Hi,

Of the 4 “bad” boards, one board has markings:

MC56F8037V

1M67E (or it might be 1W67E), my magnifier isn’t good enough to tell

XAA0827A

One I can’t read.

The other two are nearly the same as the first one except one has C as the last letter (XAA0827C) and one has a last letter of D.

We just purchased 5 new replacement boards. I checked one and it has:

MC56F8037V

4M67E (or W)

XAC1233N

Thanks,

Joe

0 Kudos

2,229 Views
TICS_Fiona
NXP Employee
NXP Employee

The code '0827' means this chip was manufactured in the 27th week in Y2008.  So the problem is not caused by the lock/unlock  operation.

Fiona Kuang

Technical Information & Commercial Support

-----------------------------------------------------------------------------------------------------------------------

Note: If this post answers your question, please click the Correct Answer button. Thank you!

-----------------------------------------------------------------------------------------------------------------------

0 Kudos

2,229 Views
TICS_Fiona
NXP Employee
NXP Employee

Is the application configuring JTAG or RESET pin as GPIO?

Fiona Kuang

Technical Information & Commercial Support

-----------------------------------------------------------------------------------------------------------------------

Note: If this post answers your question, please click the Correct Answer button. Thank you!

-----------------------------------------------------------------------------------------------------------------------

Best Regards

Fiona

0 Kudos

2,229 Views
johnlwinters
NXP Employee
NXP Employee

No

Best Regards,

John L. Winters

Senior Applications Engineer

AMR/EU MCU & MPU AE Group

Freescale Inc.

2100 Elliot Road

This email and any associated attachments have been classified as:

Freescale Confidential and Proprietary

0 Kudos

2,229 Views
johnlwinters
NXP Employee
NXP Employee

I was able to unzip this one at close to root level.

I imported and attempted to build the project.

Our Technical Information Center (TIC) will have to help you further as this SW is beyond me.

I don't have a license to use C++, nor do I support it personally.  The TIC will be able to sugges the next steps.

0 Kudos

2,229 Views
josephclifton
Contributor I

OK, thanks John for what you have done so far. You have helped eliminate several possibilities.

0 Kudos

2,229 Views
josephclifton
Contributor I

You could gain back more chars by extracting to say a directory c:\1

If you think that is still not enough, I could compress at c:\1

0 Kudos