True-Time Simulator and Real Time Debugger locks up.

[jcdammeyer]

Hi,

I'm using the HSC12X Starter kit and have loaded the CAN Demo.  It works!!!  Unplug CAN0 from CAN1 and the displays don't update.  Plug in the CAN cable and the display updates again with the pot is turned.

So far so good.  Trouble is, the Real Time Debugger has locked up.  I can't even kill it from Windows Task manager.  After I shut down all the other applications I can reboot the PC from the Task Manager Window.  Meanwhile the board does have the updated program because it can run alone.

So first things first.  AFAIK, an app like the debugger just shouldn't hang so there's something odd on my system.  I'm running the Code Warrier IDE 5.7.0. Build 2211

BTW,  when I try and kill it with task manager Windows tells me I've tried to close an unresponsive program HiWave.exe.  I can't kill that process either.

Suggestions?

Thanks

John

szAppName : hiwave.exe szAppVer : 6.1.44.6345 szModName : hungapp

szModVer : 0.0.0.0 offset : 00000000

[stanish]

Hi John,

Which version of Windows do you use?
You might try to install recent CodeWarrior v4.7 (Special Edition)  and update v4.7.1 available here:

CodeWarrior v4.7:
http://www.freescale.com/lgfiles/devsuites/HC12/CWHCS12X_v4.7_Special.exe?WT_TYPE=IDE%20-%20Debug,%2...

Update v4.7.1:
https://www.freescale.com/webapp/Download?colCode=HCS12-V4-7-PATCH&nodeId=0127260061788213D24AF9&loc...

Stanish

[jcdammeyer]

I've downloaded the newer verson and the patch.  Should I uninstall the current version or is the installation smart enough to know?

I've also discovered that when it locks up, disconnecting the USB and then reconnecting will re-establish communications.  The inability to kill the app when it goes brain dead appears to be linked to the USB since it will not allow termination until the USB cable is unplugged.  That suggests the hangup is on the PC side.

I'm using  WIN-XP SP2

Thanks

John

[stanish]

It's not necessary to uninstall current version. Multiple versions of CodeWarrior should coexist on the one PC. You should only install the newer version into different directory.

Stanish

[J2MEJediMaster]

The Hiwave.exe app is the Debugger program. Does the program seize up at random, or does it occur when there is a lot of activity (such as the CAN) running on the board? Are you using your PC's USB port to power the board? If so, maybe the board's activity draws too much power, at which point the USB driver typically shuts the port down. Then the debugger gets confused.

---Tom

[jcdammeyer]

Below is the output from the command window (Communication Show Protocol checked).  The application has just stopped at a break point.  I click on the Data:1 window and an hourglass appears.  At this point the debugger is now locked up.  Up till this point I have been able to single step and start and stop the debugging.  See below this dump for more comments.

GDI:   Reading From Memory Address: 0x031, count: 256
GDI:   DiMemoryRead()==> DI_OK
GDI:   00 FF 00 00 00 00 7C 00 00 F1 00 40 00 80 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
GDI:   00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
GDI:   00 00 00 00 00 FF FF 00 00 00 00 00 00 00 00 0F 0F 00 20 05 00 00 00 80 00 00 00 00 00 FF FF 00
GDI:   00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
GDI:   00 00 80 00 00 00 00 00 04 00 00 C0 00 00 00 00 04 00 00 C0 00 00 00 00 01 00 0C C0 00 00 00 00
GDI:   04 00 00 C0 00 00 00 04 00 00 20 00 00 00 00 00 00 00 80 00 00 00 00 00 00 00 00 00 00 00 00 04
GDI:   00 00 20 00 00 00 00 04 00 00 20 00 00 00 00 00 FE 00 10 FF D0 00 FF 00 8F 00 C6 00 00 00 00 00
GDI:   00 00 00 FF C0 00 00 00 00 00 00 00 FF FF FF 00 FF 00 00 00 00 01 10 07 00 00 00 00 00 00 00 00
----------------------

So next I unplug power to the target board and my system tells me that the USB is no longer connected.  Control returns back to the debugger and the command window now shows this:

GDI:   DiDirectAddMenuItem()==> DI_OK
No Link To Target

in> 

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

At this point, the only solution is to close the debugger and restart.  Plugging in the USB to the target board no longer establishes communication with the app.  It tries ans spits out a large number of messages but ultimately ends up with No Link To Target.

I'm at wits end here.  I have project deadlines but can't work with the target hardware.

John

[J2MEJediMaster]

I'm stumped as to what's going on here. I am puzzled as to why you're plugging/unplugging a CAN cable, but I'm not familiar with the demo programs.

I suggest you file on Service Request on this problem.  Click here to do that.

---Tom

[jcdammeyer]

Hi,

As a last resort I changed USB cables and tried a slower USB port and now the system is stable.  The other cable and port still work fine with other USB devices but obviously there is some strange interaction.

Thanks

John

[J2MEJediMaster]

Sorry I did not really get you a solution there. And thanks for posting your discovery of where the real problem lies. Did you switch from a USB 2.0 port to a USB 1.1 port?

---Tom

[jcdammeyer]

The problem hasn't really been completely solved.  On the target system in the field (I used the softec board) there is a laptop connected to the debuggging port so we can use Windows Remote Terminal to get via internet into the laptop and upgrade the 9S12 software. 

The other day I tried to do this and the system locked up completely.  I was able to reboot the laptop remotely (I had to do this twice before it was stable) and then when I tried to once again use CodeWarrier to get into the 9S12 it just wouldn't talk to the board.  

A few weeks later a tech was out at the site on a non-related service call and I had him cycle power on the 9S12.  After that I was able to load a newer version of my application into the 9S12.  That suggests that the problem is in the usb debugger on the 9S12 since rebooting the PC didn't do anything.  Obviously in the lab, unplugging the USB cable has the same result.

I"m now faced with needing to buildtwo more systems but I really don't want to use the Softec board since it's too large and is overkill.  My needs are relatively simple.

9S12 with 5 CAN ports.

8 diagnostic LEDs

FTDI USB245 parallel port interface

RS232 Serial port.

USB Debugging and code upload port.

Nice extra's would be an SD card interface and Ethernet interface but not required.

The key feature is because of the remote location the debugging port needs to be permanently connected to the PC.

Any suggestions on an external BDM module?

Thanks

John

[jcdammeyer]

The Board is powered from the external supply.

The lockup occurs (or is noticed) when single stepping either of the two demo programs so I'm not sure how the CAN activity could be the only reason.  However, the other demo program is running 500uS interrupt so perhaps that causes extra USB activity.

I'm connected directly to a PC USB port (not a USB hub expander).  I'll try the other version of CodeWarrier and report back.

Thanks

John