TS 4.0.1/CVI 8.5:Great River Tech Driver Fails with Teradyne M920

Question

I have an application that I developed that uses 2 Great River Technology (GRT) PCIe cards (for hotlink video data) and a Teradyne M920 (pattern generator/logic analyzer) running in a VXI chassis.  The development platform is WindowsXP, LabWindowsCVI 8.5, and Test Stand 4.0.1.  The problem is that when I run the sequence, one or both of the GRT cards fail to initialize and I get a hardware error message returned from the GRT driver (which is an XP installation).  Once this happens, the only recovery is to power down the PC (a reboot is not enough).  Note that the initialization failure occurs about 90% of the time.  When it doesn't fail, the sequence fails further into the testing as soon as I try using the GRT cards.  The Teradyne always works, but during my debugging efforts, I found that if I didn't load the Teradyne driver, the GRT cards work ok.  So the problem appears to be related to the combination of CVI/TS when used with GRT/Teradyne drivers.  Remove TS and all is well, except of course that we prefer to use TS - this is mature software that has been migrated from NT to XP.  The NT version was very different - older versions of CVI/TS and older versions of many of the device drivers we use.  Also didn't use the GRT cards.
I modified the CVI software to run as a stand-alone executable.  It runs fine - 100% of the time (unless of course it had already failed using TS and not repowering the PC).  I also ran the test software supplied by GRT and Teradyne.  Again - runs 100%. 
Any ideas?
Thanks, Pete

Answer 1

Hello Pete Bratko,
Thank you for posting on the NI Discussion Forums.  I am assuming that you are doing all of your driver function calls (initializing calls, etc.) with code written in LabWindows/CVI.  Could it be that you are passing some incorrect parameters into the step that calls this code?  All TestStand does when calling the code module is run the external code, and pass it input parameters and read output parameters if they are applicable. 
What does the error message from the driver say? (A screen shot would be great).  Is there any hint as to what is causing it?  Does the same error appear at both initialization and other calls?  Some more information here would be helpful.
 If you are indeed making calls to the DLL from code you have written CVI, then you can step into the code from TestStand by following these steps:
(1)    Go to Configure » Adapters…
(2)    Highlight LabWindows/CVI and select Configure…
(3)    Under Step Execution, select the radio button that says Execute Steps in an External Instance of CVI.
(4)    Select OK and then Done to let the settings take effect.
(5)    Place a breakpoint at the step that the CVI is called and run the sequence.
(6)    When the sequence editor gets to the breakpoint, select Step Into (F8) to step into the CVI code.
Once in the CVI code, you can step through it and find out exactly what line the error is coming from.
The fact that the drivers do not play well together is also strange, and raises a red flag, but it sounds to me like there is something wrong with the code that is being called (is it a .c or .dll file?)  Try out the above steps and let us know what you find.
Chris_G
Sr Test Engineer
Medtronic, Inc.