George Zagaris

Consider removing the axom aliases to fixed width types

2

Axom defines the following aliases: https://github.com/LLNL/axom/blob/fb2bb2a1f4e37e8a965e8799bca7896b64084e3a/src/axom/core/Types.hpp#L29-L46 This was mainly done in the past in order to support fixed width types with and without C++11. Since, we now require C++11, these...

Currently, to log to a file, the application has to write code that looks like this: ``` std::ofstream file; file.open( "log.dat" ); slic::addStreamToAllMsgLevels( new slic::GenericOutputStream(&file) ); ... file.close(); ``` This...

BVH query tolerance should be relative to either the AABB bounds or the total "scene" bounds

1

Currently, the BVH has a default tolerance or the caller can set a tolerance in an ad hoc way. @mclarsen suggests that a better approach may be to set the...

BVH API Semantics when using asynchronous execution policies

5

The BVH may be instantiated using asynchronous CUDA execution policy as follows: ``` spin::BVH bvh< NDIMS, axom::CUDA_EXEC >( aabbs, N ); ``` Then constructing the BVH on the GPU can...

Overhead of first reduction with CUDA backend

6

Hi everyone, We've observed that the execution time of the first reduction is notably high (in some cases, slower than sequential). However, subsequent reductions do not exhibit this behavior, suggesting...

GenerateTopology + User CellConnections not saving properly

1

dax::cont::UniformGrid inputGrid; inputGrid.SetExtent(minE,maxE); inputGrid.SetOrigin( dax::make_Vector3(this->Origin[0], this->Origin[1], this->Origin[2])); inputGrid.SetSpacing( dax::make_Vector3(this->Spacing[0], this->Spacing[1], this->Spacing[2])); typedef dax::cont::GenerateTopology GenerateT; typedef GenerateT::ClassifyResultType ClassifyResultType; dax::cont::Scheduler scheduler; //for every input cell generate 5 output cells, using a constant...

`blt_hip_smoke` does not compile

2

Hello, I am trying to compile with HIP and it seems that CMake attempts to compile the `blt_hip_smoke` test using the host-compiler. My host-config looks like this: ``` #--------------------------------------------------------------------------------------------------- #...

Runtime error when copying to constant device memory with HIP

9

**Describe the bug** Copying data from the host to constant device memory using HIP throws a runtime error. The code works fine with CUDA , but, Umpire throws the following...

# Summary - This PR attempts to fix issues encountered when using constant device memory with HIP. - A reproducer/unit regression test is also added. See #912 for more details

Question: Hypre GPU support

7

Hello, I am exploring using Hypre in my application and I am particularly interested in its GPU capabilities: - Which of the solvers and preconditioners in Hypre are available on...