Diagnosing problems running WDTF-based tests

To help you troubleshoot problems running WDTF-based test, you can use a debugger.

Diagnose problems with unresponsive WDTF-based tests (run from Visual Studio)

  1. Configure and connect a kernel debugger to the computer that is running the WDTF-based test. See Configuring a Computer for Driver Deployment, Testing, and Debugging .
  2. Search for Te.exe process and switch context to that process. For information about Te.exe, see Test Authoring and Execution Framework (TAEF).

    !process 0 0 Te.exe 
    
    PROCESS fffffa80093c6340
    
    SessionId: 1 Cid: 1320 Peb: 7f6595b3000 ParentCid: 12a0
    
    DirBase: 21eee000 ObjectTable: fffff8a0035b0a00 HandleCount: 327.
    
    Image: TE.exe
    
    ·         .process /p /r fffffa80093c6340
    
    ·         
    
  3. Run the !process command to identify the threads running under Te.exe.

    !process fffffa80093c6340
    

    Look for threads with WDTF* on the stack.

  4. Repeat step 3 for Te.ProcessHost.exe (if it exists).

Diagnose problems with PnP and power management tests

You can triage unresponsive PnP and power management tests by using the following two commands. These commands can help you identify any PnP and power management threads that could be preventing the tests from making progress.

!pnptriage
!poaction

After confirming that there are active PnP or power management threads that are stuck (see TickCount), follow up with the right component owners. (Figure out the component owners from looking at the stacks of these stuck threads).

 

 

Send comments about this topic to Microsoft

顯示:
© 2014 Microsoft