Results 8 issues of Lucas C Wilcox

cc: @andreasnoack , @ViralBShah , @eschnett , @amitmurthy A pull request from @steven-varga with an MPI-3 function brought up this issue. What should be the minimum supported MPI version for...

Hi, I am having an issue with `roots` and `BigFloat`. Please see below for minimum example reproducing the error. I am happy to debug further but I am not sure...

Enable the forced inlining option from https://github.com/JuliaGPU/GPUCompiler.jl/pull/335. When I try to use this new parameter I get the following error ```julia julia> asm = sprint(io->CUDA.code_ptx(io, g, Tuple{Int64}; always_inline=true)) ERROR: LLVM...

Now that an abstract launch function, #20, is being considered to address #3 we should think about if other functionality should be supported. Many distributed codes use - device array...

enhancement
help wanted

Here are some changes I made to the CFD2D cylinder example for the guys at HyPerComp. As a bonus I have also included some VTK output functions.

**Is your feature request related to a problem? Please describe.** To get kernel performance matching `clang` we have had to add [fast-math flags](https://llvm.org/docs/LangRef.html#fast-math-flags) such as `contract` (which `clang` and `nvcc`...

enhancement
cuda kernels
upstream

This updates StructsOfArrays to work with Julia 1.4.1 and turns it into a package. In addition basic CUDA support has been added (thanks @vchuravy!) so that `StructOfArrays` can be passed...

I tried to add precompilation to my package (via PrecompileTools in ) and am now generating `illegal hardware instruction` errors. I have not been able to minimize it but @vchuravy...