//TEST:SIMPLE(filecheck=CUDA): -target cuda -line-directive-mode none
//TEST:SIMPLE(filecheck=TORCH): -target torch -line-directive-mode none
// Verify that we can output a cuda device function with [CudaKernel].
struct MySubType
{
TorchTensor<float> array[2];
}
struct MyType
{
float2 v;
MySubType sub[2];
}
struct MyInput
{
TorchTensor<float> inValues;
float normalVal;
}
// CUDA: __global__ void myKernel(TensorView inValues_[[#]], TensorView outValues_[[#]])
[CudaKernel]
void myKernel(TensorView<float> inValues, TensorView<float> outValues)
{
if (cudaThreadIdx().x > 0)
return;
outValues.store(cudaThreadIdx().x, sin(inValues.load(cudaThreadIdx().x)));
}
// TORCH: {{^SLANG_PRELUDE_EXPORT$}}
// TORCH-NEXT: void myKernel(TensorView {{[[:alnum:]_]+}}, TensorView {{[[:alnum:]_]+}});
//
// TORCH: {{^SLANG_PRELUDE_EXPORT$}}
// TORCH-NEXT: std::tuple<std::tuple<float, float>, std::tuple<std::tuple<std::tuple<torch::Tensor, torch::Tensor>>, std::tuple<std::tuple<torch::Tensor, torch::Tensor>>>> runCompute(std::tuple<torch::Tensor, float> input_[[#]])
[TorchEntryPoint]
export __extern_cpp MyType runCompute(MyInput input)
{
MyType rs;
var outValues = TorchTensor<float>.alloc(1);
let inValues = input.inValues;
__dispatch_kernel(myKernel, uint3(1, 1, 1), uint3(32, 1, 1))(inValues, outValues);
rs.v = float2(1.0, 2.0);
rs.sub[0].array[0] = outValues;
rs.sub[0].array[1] = inValues;
rs.sub[1].array[0] = inValues;
rs.sub[1].array[1] = outValues;
return rs;
}