James Schloss

@print not functional on AMDGPU

2

Ok, maybe I'm just tired, but... ``` using AMDGPU using ROCKernels using KernelAbstractions @kernel function f_test_kernel!() @print(1, '\n') end function f_test!(AT; numcores = 4, numthreads = 256) if AT ==...

StackOverflow for `get_device` on `ROCArray`

2

``` julia> a = ROCArray(rand(10,10)); julia> KernelAbstractions.get_device(a) ERROR: StackOverflowError: Stacktrace: [1] get_device(A::ROCMatrix{Float64}) (repeats 79984 times) @ KernelAbstractions ~/.julia/packages/KernelAbstractions/DqITC/src/KernelAbstractions.jl:355 ``` Calling `AMDGPU.device(a)` works: ``` julia> AMDGPU.device(a) GPU-XX [AMD Radeon RX 6700...

KernelAbstractions support

16

The KernelAbstractions branch now compiles, so I thought I would put forward a quick draft PR while I figure out all the runtime bugs. Notes: 1. This builds off of...

TODO: - [x] Remove indexing files - [x] Figure out how to `@kernelize` linalg 'n stuff - [x] precompile - [ ] test Just a PR on a PR to...

new GPUArrays interface for KA transition

2

These are the necessary changes for the CUDA tests to pass with https://github.com/JuliaGPU/GPUArrays.jl/pull/525 * [ ] https://github.com/JuliaGPU/KernelAbstractions.jl/issues/490

JuliaGPU/GPUArrays.jl#525

JuliaGPU/GPUArrays.jl#525

JuliaGPU/GPUArrays.jl#525

updating youtube link

1

Valentin's video was moved to the OIST youtube page instead of OIST mini courses. I *think* this was the right link (lecture 4: GPU...), but here's the whole playlist: https://www.youtube.com/playlist?list=PLxZHT0nOVpvxEqYczb4rY1Je2PGjChrxL

Possible SpirV compilation pipeline

2

Ok, this is an info dump, but I think this is the ideal path forward for Fable performance and also fixes #66 . Ideally we... * [ ] Use the...