GitLab

Fixes #370

18a0cd35 vtkm::worklet::Invoker now supports scatter types
afc3f530 Remove unneeded ScatterType as it was the default
a1ea509f All scatter types now inherit from a common base
77378993 DispatcherBase: Simplify remove_cvref and remove_pointer_and_decay.
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Kenneth Moreland <kmorel@sandia.gov>
Merge-request: !1673

Fixes #297

This is required for vtkm::worklet::Invoker to launch worklets
with a non default scatter type.

designed easier to use remove_cvref and remove_pointer_and_decay
functions for the DispatcherBase.

b74ab327 Improve VTK DataSet Reader
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Robert Maynard <robert.maynard@kitware.com>
Merge-request: !1670

1. Support reading Tensors (Vec<T, 9>)
2. Support reading files having `METADATA`.

80e5f274 Update update-common.sh to the SHA1: b094fcba
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !1669

This update allows for the update-common script to be run from
any directory

fa3722bf FindFirstSetBit now correctly compiles with the Intel compiler
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !1668

Fixes paraview/paraview#18993

d1ce4a0b Fix the default launch sizes for Tesla hardware.
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !1667

674fe1fb StealArray now returns the array and free function as a Pair.
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !1664

The 8x8x8 is a better launch strategy for most VTK-m kernels.
The current problem is that a couple of VTK-m kernels use a
high number of registers and this number of threads combines to
require too many registers.

What we should do in the longer run is have more controls over
kernel launches on a per kernel basis. This will require VTK-m
to extract the number of registers being used by each kernel

41b8236a For GCC 4.8.4 'half' shadows a global variable with that name
770912f9 Correct compiler issues found with GCC 4.8.5 + CUDA 9.2 on summit
b248b2c9 Correct unused-parameter warnings from defaulted methods.
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Allison Vacanti <allison.vacanti@kitware.com>
Merge-request: !1666

GCC 4.8.5 warns about defaulted methods when parameters have names.

45c92a05 added missing MinAndMax BinaryOperator tests, updated formatting
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Kenneth Moreland <kmorel@sandia.gov>
Merge-request: !1662

This helps reduces bugs when the callers ask to steal arrays
without getting the free function, or ask for the free function
after stealing the array.

065d1178 Testing Device Adapter now uses ArrayHandle for all device transfers
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Kenneth Moreland <kmorel@sandia.gov>
Merge-request: !1651

3dda0ab2 Correct unused-parameter warnings from defaulted methods.
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Kenneth Moreland <kmorel@sandia.gov>
Merge-request: !1663

The consistent API for control to execution memory transfers is
the ArrayHandle class. Previously the tests would verify memory
transfer by calling the ArrayManagerExecution class directly. This
is problematic as the class isn't used by ArrayHandle<T, StorageBasic>.

GCC 6 warns about defaulted methods when parameters have names.

27426b7b Fix warnings found with clang-8 inside the OpenMP device adapter.
cec9af64 Update the lsan suppression to capture tbb leaks.
8ef2c4be TransferInfo doesn't leak when holding VirtualObjectTransferShareWithControl
Acked-by: Kitware Robot <kwrobot@kitware.com>
Acked-by: Kenneth Moreland <kmorel@sandia.gov>
Merge-request: !1661

d468784f Correct more host/device warnings from methods that should be host only
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !1660

Additionally document that we only want to suppress the known
leak in loguru::set_thread_name.

When TransferInfo is given memory from VirtualObjectTransferShareWithControl
it doesn't have a bound function ptr for the destruction. In those cases
we need to make sure the HostCopyOfDevice is properly deleted, otherwise
we will cause a memory leak.

bdabfbe1 Make sure ArrayPortalUniformPointCoordinates constructor is explicit
b3d951b5 vtkm::Range Include function now requires half as many min/max calls
ddaa0df2 ArrayHandleVirtualCoordinates now calls the proper parent constructor
61e80037 Make sure all execution side CellLocator objects have explicit destructors
307898ff Cleanup the CellLocatorBoundingIntervalHierarchy.cxx style.
0f31c69f Remove unnecessary constructor from ParameterContainer
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !1657

967581cf Merge branch 'upstream-taotuple' into update_tao_cuda_pragmas
90784608 taotuple 2019-04-24 (57001975)
Acked-by: Kitware Robot <kwrobot@kitware.com>
Merge-request: !1659

* upstream-taotuple:
  taotuple 2019-04-24 (57001975)

Code extracted from:

    https://gitlab.kitware.com/third-party/taotuple.git

at commit 5700197521301b5dd102f81d0073f83e15e5a7e5 (for/vtk-m).

We have had problems in the past with host/device default constructors

Previously it was calling the ArrayHandle<T,StorageTagVirtual>
constructor and not the ArrayHandleVirtual constructor which
generated a warning with some compilers