The `ReleaseResourcesExecution` method makes changes by booting data off
of the execution environment. But logically the array does not change.
It remains the same size with the same contents. Thus, it makes sense
for this to be a const method.
Also modified some deprecated methods a bit to remove some unnecessary
templates.
19944653b Add scope log for functions that help map fields in filters
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Robert Maynard <robert.maynard@kitware.com>
Merge-request: !2286
The `UnitTestParticleAdvectionFilter` was only testing advection in
structured cell sets (with either uniform or rectilinear points). This
did not test all paths (i.e. grid evaluators and locators) of particle
advection. Changed the test to also try grids with explicit cells.
Note that the explicit cells are created by just converting the
rectilinear grids to explicit grids. It would be better to get at least
one small dataset that starts as unstructured.
f5c5b6188 Log the computation of ranges in fields and arrays
bef55f8ee Log the conversion of num components to offsets
420b9d397 Log the building of arrays for scatters and masks
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !2283
c26e193fe Use kokkos provided routines for sort and copy
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Robert Maynard <robert.maynard@kitware.com>
Merge-request: !2277
Previously vtk-m allowed users to issue atomic loads on constant
values which is problematic for the following reasons:
- can be a source of undefined behavior
- not supported by kokkos
This issue was detected when using kokkos HIP atomic implementation
When you invoke a worklet, a scoped performance log is automatically
generated. The previous version gave the name of the worklet, which is
technically everything you need to know. However, it is also convenient
to know what type of worklet it is (e.g. a field map worklet or a
topology map worklet). This can be determined by looking at the
definition of the worklet in the source code, but that is not practical
if, for example, you want to perform automated analysis.
This change prints out the type of the dispatcher instead of the
worklet. The full type of the dispatcher includes the worklet as a
template parameter, so you get both the dispatcher/worklet type and the
worklet itself. So previously you would get log lines like
```
Invoking Worklet: 'vtkm::worklet::vtkm::worklet::ExternalFaces::NumFacesPerCell'
```
They now look like this:
```
Invoking Worklet: 'vtkm::worklet::DispatcherMapTopology<vtkm::worklet::ExternalFaces::NumFacesPerCell>'
```
268a07bb0 remove test on large graph and sample data
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Kenneth Moreland <kmorel@sandia.gov>
Merge-request: !2278
- It also remove termination on bench error
- It disables passing fields in Countour to skip an error.
Signed-off-by: Vicente Adolfo Bolea Sanchez <vicente.bolea@kitware.com>
932f8b82f Merge branch 'concurrent_union_find' of gitlab.kitware.com:ollielo/vtk-m into concurrent_union_find
eaff27195 minor changes on comments on findRoot()
9fef540ba minor changes on comments on findRoot()
916661e33 Merge branch 'concurrent_union_find' of gitlab.kitware.com:ollielo/vtk-m into concurrent_union_find
a036b0f98 Minor change based on code review.
9e68022ac Minor change based on code review.
1185f70bc Data Race Resolution
aed7e02da Merge branch 'concurrent_union_find' of gitlab.kitware.com:ollielo/vtk-m into concurrent_union_find
...
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Kenneth Moreland <kmorel@sandia.gov>
Merge-request: !2248
9bd6f3e6d Disable VTKM_ASSERT when using HIP
68b5edfcb VTK-m defines VTKM_EXEC / VTKM_EXEC_CONT when compiling with HIP
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Kenneth Moreland <kmorel@sandia.gov>
Merge-request: !2270
d0c9fb274 Correct missing `VTKM_EXEC_CONT` in UnitTestArrayHandleExtrude
1365cfa9e Update ConnectivityExtrude to follow clangs rules on markup
09df45bb1 Update vtkm/Geometry.hxx to follow clangs rules on function markup
461616a77 Refactor some VTK-m device adapter to be alphabetical
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Kenneth Moreland <kmorel@sandia.gov>
Merge-request: !2268
2278fdfd9 Deprecation macro has to come before VTKM_EXEC or VTKM_EXEC_CONT
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Kenneth Moreland <kmorel@sandia.gov>
Merge-request: !2266
Although `vtkm::internal::Variant` respected the trivially copyable
attribute of the types it contains, it was never totally trivial (i.e.
`std::is_trivial<Variant<...>>` was never true). The reason was that
`Variant` was initializing its `Index` parameter to signify that it was
not initialized. However, the fact that `Index` was initialized meant
that it was not trivially constructed.
Now, `Variant` type checks its types to see if they are all trivially
constructible. If so, it makes itself trivially constructible.
This means that `Index` may or may not be valid if `Variant` is
constructed without an argument. This in turn means that the result of
`Variant::IsValid` becomes undefined. That should be OK in practice.
`Index` will "point" to an uninitialized object, but that object is
trivially constructed anyway. However, that could cause problems if
developers used `IsValid` to determine if something is selected.