| 1 |
//===----------------------------------------------------------------------===// |
| 2 |
// |
| 3 |
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. |
| 4 |
// See https://llvm.org/LICENSE.txt for license information. |
| 5 |
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception |
| 6 |
// |
| 7 |
//===----------------------------------------------------------------------===// |
| 8 |
/// |
| 9 |
/// \file |
| 10 |
/// This file contains the definition of the runKernelBody, a template helper |
| 11 |
/// that executes the per-thread logic of a math function's kernel wrapper. |
| 12 |
/// |
| 13 |
//===----------------------------------------------------------------------===// |
| 14 |
|
| 15 |
#ifndef CONFORMANCE_DEVICE_CODE_KERNELRUNNER_HPP |
| 16 |
#define CONFORMANCE_DEVICE_CODE_KERNELRUNNER_HPP |
| 17 |
|
| 18 |
#include <gpuintrin.h> |
| 19 |
#include <stddef.h> |
| 20 |
#include <stdint.h> |
| 21 |
|
| 22 |
namespace kernels { |
| 23 |
|
| 24 |
template <auto Func, typename OutType, typename... InTypes> |
| 25 |
void runKernelBody(size_t NumElements, OutType *Out, const InTypes *...Ins) { |
| 26 |
uint32_t Index = |
| 27 |
__gpu_num_threads_x() * __gpu_block_id_x() + __gpu_thread_id_x(); |
| 28 |
|
| 29 |
if (Index < NumElements) { |
| 30 |
Out[Index] = Func(Ins[Index]...); |
| 31 |
} |
| 32 |
} |
| 33 |
} // namespace kernels |
| 34 |
|
| 35 |
#endif // CONFORMANCE_DEVICE_CODE_KERNELRUNNER_HPP |