- 17 Apr, 2019 4 commits
-
-
Nickolas Davis authored
-
Nickolas Davis authored
-
Robert Maynard authored
047b6465 VTK-m now provides better scheduling parameters controls Acked-by:
Kitware Robot <kwrobot@kitware.com> Acked-by:
Kenneth Moreland <kmorel@sandia.gov> Merge-request: !1643
-
Robert Maynard authored
VTK-m now offers a more GPU aware set of defaults for kernel scheduling. When VTK-m first launches a kernel we do system introspection and determine what GPU's are on the machine and than match this information to a preset table of values. The implementation is designed in a way that allows for VTK-m to offer both specific presets for a given GPU ( V100 ) or for an entire generation of cards ( Pascal ). Currently VTK-m offers preset tables for the following GPU's: - Tesla V100 - Tesla P100 If the hardware doesn't match a specific GPU card we than try to find the nearest know hardware generation and use those defaults. Currently we offer defaults for - Older than Pascal Hardware - Pascal Hardware - Volta+ Hardware Some users have workloads that don't align with the defaults provided by VTK-m. When that is the cause, it is possible to override the defaults by binding a custom function to `vtkm::cont::cuda::InitScheduleParameters`. As shown below: ```cpp ScheduleParameters CustomScheduleValues(char const* name, int major, int minor, int multiProcessorCount, int maxThreadsPerMultiProcessor, int maxThreadsPerBlock) { ScheduleParameters params { 64 * multiProcessorCount, //1d blocks 64, //1d threads per block 64 * multiProcessorCount, //2d blocks { 8, 8, 1 }, //2d threads per block 64 * multiProcessorCount, //3d blocks { 4, 4, 4 } }; //3d threads per block return params; } vtkm::cont::cuda::InitScheduleParameters(&CustomScheduleValues); ```
-
- 15 Apr, 2019 2 commits
-
-
Robert Maynard authored
ff30684c Removes the default device macros from VTK-m Acked-by:
Kitware Robot <kwrobot@kitware.com> Merge-request: !1641
-
Robert Maynard authored
Fixes #116
-
- 12 Apr, 2019 4 commits
-
-
Robert Maynard authored
e54cbe85 contourtree_augmented doesn't call abort in VTK-m worklets Acked-by:
Kitware Robot <kwrobot@kitware.com> Merge-request: !1642
-
Robert Maynard authored
-
Robert Maynard authored
71018e3b Always export vtkm_cuda no matter VTKm_INSTALL_ONLY_LIBRARIES Acked-by:
Kitware Robot <kwrobot@kitware.com> Merge-request: !1640
-
Robert Maynard authored
-
- 11 Apr, 2019 8 commits
-
-
Robert Maynard authored
661fb64d AtomicInterfaceControl functions are marked with VTKM_SUPPRESS_EXEC_WARNINGS 0c70f9b9 Add BitFieldIn/Out/InOut worklet signature tags. a66510e8 Add ArrayHandleBitField, a boolean-valued AH backed by a BitField. 56cc5c3d Add support for BitFields. d01b9738 Allow VTKM_SUPPRESS_EXEC_WARNINGS to be used inside macros. 2f2ca937 Add bit operations FindFirstSetBit and CountSetBits to Math.h. Acked-by:
Kitware Robot <kwrobot@kitware.com> Merge-request: !1629
-
Robert Maynard authored
46843ff4 CellSetExplicit has no dependency on default vtk-m device Acked-by:
Kitware Robot <kwrobot@kitware.com> Acked-by:
Kenneth Moreland <kmorel@sandia.gov> Merge-request: !1636
-
Robert Maynard authored
-
Allison Vacanti authored
These provide an appropriate BitPortal for use in worklets.
-
Allison Vacanti authored
-
Allison Vacanti authored
BitFields are: - Stored in memory using a contiguous buffer of bits. - Accessible via portals, a la ArrayHandle. - Portals operate on individual bits or words. - Operations may be atomic for safe use from concurrent kernels. The new BitFieldToUnorderedSet device algorithm produces an ArrayHandle containing the indices of all set bits, in no particular order. The new AtomicInterface classes provide an abstraction into bitwise atomic operations across control and execution environments and are used to implement the BitPortals.
-
Allison Vacanti authored
-
Allison Vacanti authored
-
- 10 Apr, 2019 10 commits
-
-
Robert Maynard authored
89ec4aae Reduction on CUDA handles different input and output types better Acked-by:
Kitware Robot <kwrobot@kitware.com> Merge-request: !1638
-
Sujin Philip authored
b391acf4 Remove SimplePolymorphicContainer Acked-by:
Kitware Robot <kwrobot@kitware.com> Acked-by:
Kenneth Moreland <kmorel@sandia.gov> Merge-request: !1634
-
Robert Maynard authored
1d980ed1 Thrust Patches tried to apply CUDA 10.1 patches to CUDA 10.0 Acked-by:
Kitware Robot <kwrobot@kitware.com> Merge-request: !1639
-
Robert Maynard authored
-
Robert Maynard authored
When reducing an input type that differs from the output type you need to write a custom binary operator that also implements how to do the unary transformation.
-
Robert Maynard authored
The default vtk-m device is an outdated concept and this is the first step of removing it from VTK-m.
-
Sujin Philip authored
The last place remaining where it is being used is DynamicCellSet, which can instead use `vtkm::cont::CellSet` as it itself is polymorphic.
-
Robert Maynard authored
f05940aa ThrustPatches now only patches fixes for relevant cuda versions 20d6201a Suppress thrust::mr::stateless_resource_allocator host/device warnings Acked-by:
Kitware Robot <kwrobot@kitware.com> Merge-request: !1637
-
Robert Maynard authored
Rather than always patch Thrust functions, we now only patch based on the Thrust version.
-
Robert Maynard authored
-
- 09 Apr, 2019 9 commits
-
-
Robert Maynard authored
0130088b Suppress more self-assign-overloaded warnings found by clang fa545585 UnitTestArrayPortalValueReference doesn't warn when compiled with appleclang 5cc0d03f Merge branch 'upstream-diy' into clang_warnings dbd3781d diy 2019-04-09 (f7a68da4) 5c2f2ebc suppress warnings found by Wself-assign-overloaded 77426f04 Correct casting long to long long warning from clang. Acked-by:
Kitware Robot <kwrobot@kitware.com> Merge-request: !1635
-
Robert Maynard authored
-
Robert Maynard authored
-
Robert Maynard authored
* upstream-diy: diy 2019-04-09 (f7a68da4)
-
Kitware Robot authored
Code extracted from: https://gitlab.kitware.com/third-party/diy2.git at commit f7a68da4fbe93ae7a8699b66f125a98ace842998 (for/vtk-m).
-
Robert Maynard authored
-
Robert Maynard authored
-
Sujin Philip authored
dae779b9 ArrayHandleVirtual bugfixes Acked-by:
Kitware Robot <kwrobot@kitware.com> Acked-by:
Robert Maynard <robert.maynard@kitware.com> Merge-request: !1632
-
Sujin Philip authored
2bbf501c Fix a bug in MakeTestDataset Acked-by:
Kitware Robot <kwrobot@kitware.com> Acked-by:
Kenneth Moreland <kmorel@sandia.gov> Merge-request: !1633
-
- 08 Apr, 2019 2 commits
-
-
Sujin Philip authored
In Make3DExplicitDataSetZoo, only 25 cells are added, but the variable `nCells` is set to 27.
-
Sujin Philip authored
After the change to `ArrayHandleVirtual` where it became a subclass of `vtkm::cont::ArrayHandle`, a few extra changes are required. 1. Functions with `ArrayHandleVirtual` as parameters will not be callable with the superclass `ArrayHandle`. This is fixed by changing the argument types to the superclass. 2. Add Serialization classes specializations for the superclass, as "is-a" relation is not considered for class template parameters.
-
- 05 Apr, 2019 1 commit
-
-
Robert Maynard authored
364bdce5 Correctly override the proper PrepareForExecution method Acked-by:
Kitware Robot <kwrobot@kitware.com> Merge-request: !1631
-