[IGSTK-Developers] Re: A new UMR appeared on the dashboard today

Luis Ibanez luis.ibanez at kitware.com
Thu Feb 2 12:54:22 EST 2006


Hi David,

Yes,
This is related to the issue that we posted yesterday to the mailing
list.

We should discuss it at the tcon today, because it is a non-trivial
one. It revolves around the fact that we don't have a service for
getting time from the real time clock. The TimeStamps need such
service, and the current way of getting it, is to use a static
RealTimeClock that is shared by all the TimeStamps.  The construction
and destruction of this clock is not correctly set up at this point.

The option that seems to be the best compromise is to have an 
initialization and a remove method for setting up this static clock.

Application developers would have to remember to call the
initialization method at the beginning of their applications,
and to call the remove method at the end of the application too.
This could easily be done if we have an igstkApplication class,
since the constructor and destructor of this class could take
care of calling the appropriate methods.


Any other suggestions are welcome.



    Thanks


       Luis



------------------
David Gobbi wrote:
> Yup, all of the dynamic analysis errors were fixed, but this one just 
> popped up:
> 
> 
> igstkSpatialObjectTest
> 
> UMR ==17143== Use of uninitialised value of size 8
> ==17143==    at 0x879EBFF: 
> igstk::TimeStamp::SetStartTimeNowAndExpireAfter(double) 
> (igstkTimeStamp.cxx:67)
> ==17143==    by 0x87C8C52: igstk::Transform::SetToIdentity(double) 
> (igstkTransform.cxx:167)
> ==17143==    by 0x83F0655: 
> igstk::SpatialObjectTest::MyTracker::InternalUpdateStatus() 
> (igstkSpatialObjectTest.cxx:117)
> ==17143==    by 0x87ACADB: 
> igstk::Tracker::AttemptToUpdateStatusProcessing() (igstkTracker.cxx:665)
> ==17143==    by 0x87B7B17: igstk::StateMachine::ProcessInput(unsigned 
> long const&) (igstkStateMachine.txx:368)
> ==17143==    by 0x87B4BC6: igstk::StateMachine::ProcessInputs() 
> (igstkStateMachine.txx:298)
> ==17143==    by 0x87A7E11: igstk::Tracker::UpdateStatus() 
> (igstkTracker.cxx:276)
> ==17143==    by 0x87BD63D: 
> itk::SimpleMemberCommand::Execute(itk::Object*, itk::EventObject const&) 
> (itkCommand.h:248)
> ==17143==    by 0x89F1B61: 
> itk::SubjectImplementation::InvokeEvent(itk::EventObject const&, 
> itk::Object*) (itkSmartPointer.h:71)
> ==17143==    by 0x89F2B84: itk::Object::InvokeEvent(itk::EventObject 
> const&) (itkObject.cxx:450)
> ==17143==    by 0x87565CF: igstk::PulseGenerator::EmitPulseProcessing() 
> (igstkPulseGenerator.cxx:265)
> ==17143==    by 0x875DD67: igstk::StateMachine::ProcessInput(unsigned 
> long const&) (igstkStateMachine.txx:368)
> ==17143==    by 0x875B9C0: igstk::StateMachine::ProcessInputs() 
> (igstkStateMachine.txx:298)
> ==17143==    by 0x87562DB: igstk::PulseGenerator::CallbackTimer() 
> (igstkPulseGenerator.cxx:243)
> ==17143==    by 0x8756055: 
> igstk::PulseGenerator::CallbackTimerGlobal(void*) 
> (igstkPulseGenerator.cxx:230)
> ==17143==    by 0x8756EE5: igstk::PulseGenerator::InvokeTimeoutActions() 
> (igstkPulseGenerator.cxx:408)
> ==17143==    by 0x8756D4F: igstk::PulseGenerator::CheckTimeouts() 
> (igstkPulseGenerator.cxx:358)
> ==17143==    by 0x83EEAC0: igstkSpatialObjectTest(int, char**) 
> (igstkSpatialObjectTest.cxx:192)
> ==17143==    by 0x8319117: main (igstkTestMain.h:133)
> 
> 
> 
> 




More information about the IGSTK-Developers mailing list