VTK issueshttps://gitlab.kitware.com/vtk/vtk/-/issues2018-12-05T16:19:01-05:00https://gitlab.kitware.com/vtk/vtk/-/issues/17454Some VTK-m filters aren't working with CUDA2018-12-05T16:19:01-05:00Allison Vacantialliepiper16@gmail.comSome VTK-m filters aren't working with CUDAThe following filters are disabled on CUDA pending further investigation:
- `vtkmClip`
- `vtkmContour`
They are crashing with illegal memory access on the CUDA device when running the VTK tests on `adora`:
https://open.cdash.org/viewTe...The following filters are disabled on CUDA pending further investigation:
- `vtkmClip`
- `vtkmContour`
They are crashing with illegal memory access on the CUDA device when running the VTK tests on `adora`:
https://open.cdash.org/viewTest.php?onlyfailed&buildid=5649252https://gitlab.kitware.com/vtk/vtk/-/issues/17269VTKm Clip Compute Scalars has no effect2018-10-23T11:11:55-04:00Mathieu Westphal (Kitware)VTKm Clip Compute Scalars has no effectVTKm Clip Compute Scalars properties does not seem to have any effect
How to reproduce :
* Open ParaView
* Load VTKm Filters plugin
* Create a wavelet, Apply
* Add a VTKm Clip filter
* Set Value to 100
* Compute Scalars is checked...VTKm Clip Compute Scalars properties does not seem to have any effect
How to reproduce :
* Open ParaView
* Load VTKm Filters plugin
* Create a wavelet, Apply
* Add a VTKm Clip filter
* Set Value to 100
* Compute Scalars is checked by default
* Apply
* As expected, RTData array is available in output
* Uncheck compute scalars, Apply
* RTData is still available but it should not be
Initially reported here : paraview/paraview#18072https://gitlab.kitware.com/vtk/vtk/-/issues/17268VTKm Treshold, Clip, Contour does not support certain type of data array2018-10-23T11:11:46-04:00Mathieu Westphal (Kitware)VTKm Treshold, Clip, Contour does not support certain type of data arrayWhen using VTKm Treshold, Clip or Contour, certain type of data array are unsupported :
How to reproduce :
* Open ParaView
* Create a wavelet, Apply
* Add a random Attribute, set Attribute type to **type**, Apply
* Add a VTKm Tresh...When using VTKm Treshold, Clip or Contour, certain type of data array are unsupported :
How to reproduce :
* Open ParaView
* Create a wavelet, Apply
* Add a random Attribute, set Attribute type to **type**, Apply
* Add a VTKm Treshold
With **type** in Unsigned Long, Long, Char, it segfaults.
With **type** in Unsigned Int, Unsigned Short, Short, it show an error like
```
Warning: In /home/glow/work/paraview/paraviewThird/paraview/VTK/Accelerators/Vtkm/vtkmThreshold.cxx, line 105
vtkmThreshold (0x55c47cd4c470): Unable to use VTKm to convert field( RandomPointVectors ) to the Threshold output: Could not find appropriate cast for array in CastAndCall1.
Array: valueType=N4vtkm3VecIjLi3EEE storageType=N6tovtkm23vtkAOSArrayContainerTagE numValues=9261 [(239,165,240) (164,70,254) (183,142,249) ... (213,51,13) (209,137,214) (166,169,18)]
TypeList: N6tovtkm14FieldTypeInVTKE
StorageList: N6tovtkm14ArrayListInVTKE
Warning: In /home/glow/work/paraview/paraviewThird/paraview/VTK/Accelerators/Vtkm/vtkmThreshold.cxx, line 105
vtkmThreshold (0x55c47cd4c470): Unable to use VTKm to convert field( RandomCellVectors ) to the Threshold output: Could not find appropriate cast for array in CastAndCall1.
Array: valueType=N4vtkm3VecIjLi3EEE storageType=N6tovtkm23vtkAOSArrayContainerTagE numValues=8000 [(217,34,19) (187,183,117) (103,181,252) ... (42,248,137) (136,240,116) (179,186,241)]
TypeList: N6tovtkm14FieldTypeInVTKE
StorageList: N6tovtkm14ArrayListInVTKE
```
Initially reported here paraview/paraview#18075https://gitlab.kitware.com/vtk/vtk/-/issues/17260vtkmPolyDataNormals does not respect AutoOrientNormalsOn()2018-10-23T11:12:03-04:00Maik FroechtenichtvtkmPolyDataNormals does not respect AutoOrientNormalsOn()Using [vtkmPolyDataNormals](Accelerators/Vtkm/vtkmPolyDataNormals.cxx) with AutoOrientNormalsOn() (https://www.vtk.org/doc/nightly/html/classvtkmPolyDataNormals.html) does not affect the output of the filter, which leads to wrong results...Using [vtkmPolyDataNormals](Accelerators/Vtkm/vtkmPolyDataNormals.cxx) with AutoOrientNormalsOn() (https://www.vtk.org/doc/nightly/html/classvtkmPolyDataNormals.html) does not affect the output of the filter, which leads to wrong results.
As far as I understand is [SurfaceNormals](vtkm/filter/SurfaceNormals.h), which is called, not able to handle the requested AutoOrientNormalsOn setting.
As a workaround, the flag should be added in [vtkmPolyDataNormals.cxx](Accelerators/Vtkm/vtkmPolyDataNormals.cxx#L86) to the existing list of exceptions that makes sure to fallback to the serial vtkPolyDataNormals.https://gitlab.kitware.com/vtk/vtk/-/issues/17120Simplify the VTK-m and VTK SMP build options2018-10-23T12:35:27-04:00Robert MaynardSimplify the VTK-m and VTK SMP build optionsVTK-m and VTK SMP both need to know what backends they should built with. Currently VTK-m defers to VTKSMP but that should be inverted.
Instead we should find what of all backends can be built with VTK-m since it supports multiple at th...VTK-m and VTK SMP both need to know what backends they should built with. Currently VTK-m defers to VTKSMP but that should be inverted.
Instead we should find what of all backends can be built with VTK-m since it supports multiple at the same time.