Leandro Lacerda f1eb869bae
[Offload][UnitTests] Build device code as C++ (#151714)
This commit refactors the `add_offload_test_device_code` CMake function
to compile device code using the C++ compiler (`CMAKE_CXX_COMPILER`)
instead of the C compiler.

This change enables the use of C++ features, such as templates, within
device-side test kernels. This will allow for more advanced and reusable
kernel wrappers, reducing boilerplate code in the conformance test
suite.

As part of this change:
- All `.c` files for device code in `unittests/` have been renamed to
`.cpp`.
- Kernel definitions are now wrapped in `extern "C"` to ensure C linkage
and prevent name mangling.

This change affects the `OffloadAPI` and `Conformance` test suites.

cc @callumfare @RossBrunton @jhuber6
2025-08-04 07:00:51 -05:00

41 lines
1.2 KiB
C++

//===----------------------------------------------------------------------===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
///
/// \file
/// This file contains the implementation of the device kernels that wrap the
/// math functions from the llvm-libm provider.
///
//===----------------------------------------------------------------------===//
#include <gpuintrin.h>
#include <math.h>
#include <stddef.h>
#include <stdint.h>
typedef _Float16 float16;
extern "C" {
__gpu_kernel void hypotf16Kernel(const float16 *X, float16 *Y, float16 *Out,
size_t NumElements) {
uint32_t Index =
__gpu_num_threads_x() * __gpu_block_id_x() + __gpu_thread_id_x();
if (Index < NumElements)
Out[Index] = hypotf16(X[Index], Y[Index]);
}
__gpu_kernel void logfKernel(const float *X, float *Out, size_t NumElements) {
uint32_t Index =
__gpu_num_threads_x() * __gpu_block_id_x() + __gpu_thread_id_x();
if (Index < NumElements)
Out[Index] = logf(X[Index]);
}
} // extern "C"