add CUDAKernels

Author: mcabbott

Project.toml

@@ -10,6 +10,7 @@ Requires = "ae029012-a4dd-5104-9daa-d747884805df"
 
 [compat]
 CUDA = "1, 2"
+CUDAKernels = "0.1"
 DiffRules = "1"
 FillArrays = "0.10"
 ForwardDiff = "0.10"
@@ -26,6 +27,7 @@ julia = "1.3"
 
 [extras]
 CUDA = "052768ef-5323-5732-b1bb-66c8b64840ba"
+CUDAKernels = "72cfdca4-0801-4ab0-bf6a-d52aa10adc57"
 FillArrays = "1a297f60-69ca-5386-bcde-b61e274b549b"
 ForwardDiff = "f6369f11-7733-5829-9624-2563aa707210"
 KernelAbstractions = "63c18a36-062a-441e-b654-da1e3ab1ce7c"
@@ -42,4 +44,4 @@ VectorizationBase = "3d5dd08c-fd9d-11e8-17fa-ed2836048c2f"
 Zygote = "e88e6eb3-aa80-5325-afca-941959d7151f"
 
 [targets]
-test = ["Test", "CUDA", "FillArrays", "ForwardDiff", "KernelAbstractions", "LinearAlgebra", "LoopVectorization", "NamedDims", "OffsetArrays", "Printf", "Random", "TensorOperations", "Tracker", "VectorizationBase", "Zygote"]
+test = ["Test", "CUDA", "CUDAKernels", "FillArrays", "ForwardDiff", "KernelAbstractions", "LinearAlgebra", "LoopVectorization", "NamedDims", "OffsetArrays", "Printf", "Random", "TensorOperations", "Tracker", "VectorizationBase", "Zygote"]

test/cuda.jl

@@ -1,6 +1,6 @@
 
 using Tullio, Test
-using CUDA, KernelAbstractions
+using CUDA, CUDAKernels, KernelAbstractions
 CUDA.allowscalar(false)
 using Tracker
 @tullio grad=Base