derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
259 lines
14 KiB
C++
259 lines
14 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
|
|
// SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
// <cmath>
|
|
|
|
#include <cuda/std/cassert>
|
|
#include <cuda/std/cmath>
|
|
#include <cuda/std/limits>
|
|
#include <cuda/std/type_traits>
|
|
|
|
#include "comparison.h"
|
|
#include "test_macros.h"
|
|
|
|
TEST_DIAG_SUPPRESS_MSVC(4244) // conversion from 'double' to 'float', possible loss of data
|
|
TEST_DIAG_SUPPRESS_MSVC(4305) // 'argument': truncation from 'T' to 'float'
|
|
TEST_DIAG_SUPPRESS_MSVC(4146) // unary minus operator applied to unsigned type, result still unsigned
|
|
|
|
template <typename T>
|
|
__host__ __device__ void test_hypot(T val)
|
|
{
|
|
using ret = cuda::std::conditional_t<cuda::std::is_integral_v<T>, double, T>;
|
|
static_assert(cuda::std::is_same_v<decltype(cuda::std::hypot(T{}, T{})), ret>, "");
|
|
|
|
// hypot(x, y), hypot(y, x), and hypot(x, -y) are equivalent.
|
|
assert(eq(cuda::std::hypot(T(1.0), T(0.25)), cuda::std::hypot(T(0.25), T(1.0))));
|
|
if constexpr (cuda::std::is_signed_v<T>)
|
|
{
|
|
assert(eq(cuda::std::hypot(T(1.0), T(0.25)), cuda::std::hypot(T(1.0), T(-0.25))));
|
|
}
|
|
|
|
// If one of the arguments is ±0, std::hypot(x, y) is equivalent to std::fabs called with the non-zero argument.
|
|
assert(eq(cuda::std::hypot(val, T(0.25)), T(0.25)));
|
|
assert(eq(cuda::std::hypot(-val, T(0.25)), T(0.25)));
|
|
|
|
assert(is_about(cuda::std::hypot(T(3.0), T(4.0)), T(5.0)));
|
|
if constexpr (cuda::std::is_signed_v<T>)
|
|
{
|
|
assert(eq(cuda::std::hypot(val, T(-0.25)), T(0.25)));
|
|
assert(eq(cuda::std::hypot(-val, T(-0.25)), T(0.25)));
|
|
}
|
|
|
|
if constexpr (!cuda::std::is_integral_v<T>)
|
|
{
|
|
// If one of the arguments is ±∞, std::hypot(x, y) returns +∞ even if the other argument is NaN.
|
|
assert(eq(cuda::std::hypot(cuda::std::numeric_limits<T>::infinity(), cuda::std::numeric_limits<T>::quiet_NaN()),
|
|
cuda::std::numeric_limits<T>::infinity()));
|
|
assert(eq(cuda::std::hypot(-cuda::std::numeric_limits<T>::infinity(), cuda::std::numeric_limits<T>::quiet_NaN()),
|
|
cuda::std::numeric_limits<T>::infinity()));
|
|
assert(eq(cuda::std::hypot(cuda::std::numeric_limits<T>::quiet_NaN(), cuda::std::numeric_limits<T>::infinity()),
|
|
cuda::std::numeric_limits<T>::infinity()));
|
|
assert(eq(cuda::std::hypot(cuda::std::numeric_limits<T>::quiet_NaN(), -cuda::std::numeric_limits<T>::infinity()),
|
|
cuda::std::numeric_limits<T>::infinity()));
|
|
|
|
// Otherwise, if any of the arguments is NaN, NaN is returned.
|
|
assert(cuda::std::isnan(cuda::std::hypot(T(3.0), cuda::std::numeric_limits<T>::quiet_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypot(T(3.0), -cuda::std::numeric_limits<T>::quiet_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypot(T(3.0), cuda::std::numeric_limits<T>::signaling_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypot(T(3.0), -cuda::std::numeric_limits<T>::signaling_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypot(cuda::std::numeric_limits<T>::quiet_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypot(-cuda::std::numeric_limits<T>::quiet_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypot(cuda::std::numeric_limits<T>::signaling_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypot(-cuda::std::numeric_limits<T>::signaling_NaN(), T(3.0))));
|
|
|
|
// Some random value
|
|
const T expected = T(2.236067977499789805051477742381394);
|
|
assert(is_about(cuda::std::hypot(T(1.0), T(2.0)), expected));
|
|
}
|
|
|
|
if constexpr (cuda::std::is_same_v<T, float>)
|
|
{
|
|
// If one of the arguments is ±∞, std::hypot(x, y) returns +∞ even if the other argument is NaN.
|
|
assert(eq(cuda::std::hypotf(cuda::std::numeric_limits<T>::infinity(), cuda::std::numeric_limits<T>::quiet_NaN()),
|
|
cuda::std::numeric_limits<T>::infinity()));
|
|
assert(eq(cuda::std::hypotf(-cuda::std::numeric_limits<T>::infinity(), cuda::std::numeric_limits<T>::quiet_NaN()),
|
|
cuda::std::numeric_limits<T>::infinity()));
|
|
assert(eq(cuda::std::hypotf(cuda::std::numeric_limits<T>::quiet_NaN(), cuda::std::numeric_limits<T>::infinity()),
|
|
cuda::std::numeric_limits<T>::infinity()));
|
|
assert(eq(cuda::std::hypotf(cuda::std::numeric_limits<T>::quiet_NaN(), -cuda::std::numeric_limits<T>::infinity()),
|
|
cuda::std::numeric_limits<T>::infinity()));
|
|
|
|
// Otherwise, if any of the arguments is NaN, NaN is returned.
|
|
assert(cuda::std::isnan(cuda::std::hypotf(T(3.0), cuda::std::numeric_limits<T>::quiet_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(T(3.0), -cuda::std::numeric_limits<T>::quiet_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(T(3.0), cuda::std::numeric_limits<T>::signaling_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(T(3.0), -cuda::std::numeric_limits<T>::signaling_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(cuda::std::numeric_limits<T>::quiet_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(-cuda::std::numeric_limits<T>::quiet_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(cuda::std::numeric_limits<T>::signaling_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(-cuda::std::numeric_limits<T>::signaling_NaN(), T(3.0))));
|
|
|
|
// Some random value
|
|
const T expected = T(2.236067977499789805051477742381394);
|
|
assert(is_about(cuda::std::hypotf(T(1.0), T(2.0)), expected));
|
|
assert(is_about(cuda::std::hypotf(T(3.0), T(4.0)), T(5.0)));
|
|
}
|
|
#if _CCCL_HAS_LONG_DOUBLE()
|
|
else if constexpr (cuda::std::is_same_v<T, long double>)
|
|
{
|
|
// If one of the arguments is ±∞, std::hypot(x, y) returns +∞ even if the other argument is NaN.
|
|
assert(eq(cuda::std::hypotl(cuda::std::numeric_limits<T>::infinity(), cuda::std::numeric_limits<T>::quiet_NaN()),
|
|
cuda::std::numeric_limits<T>::infinity()));
|
|
assert(eq(cuda::std::hypotl(-cuda::std::numeric_limits<T>::infinity(), cuda::std::numeric_limits<T>::quiet_NaN()),
|
|
cuda::std::numeric_limits<T>::infinity()));
|
|
assert(eq(cuda::std::hypotl(cuda::std::numeric_limits<T>::quiet_NaN(), cuda::std::numeric_limits<T>::infinity()),
|
|
cuda::std::numeric_limits<T>::infinity()));
|
|
assert(eq(cuda::std::hypotl(cuda::std::numeric_limits<T>::quiet_NaN(), -cuda::std::numeric_limits<T>::infinity()),
|
|
cuda::std::numeric_limits<T>::infinity()));
|
|
|
|
// Otherwise, if any of the arguments is NaN, NaN is returned.
|
|
assert(cuda::std::isnan(cuda::std::hypotl(T(3.0), cuda::std::numeric_limits<T>::quiet_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(T(3.0), -cuda::std::numeric_limits<T>::quiet_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(T(3.0), cuda::std::numeric_limits<T>::signaling_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(T(3.0), -cuda::std::numeric_limits<T>::signaling_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(cuda::std::numeric_limits<T>::quiet_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(-cuda::std::numeric_limits<T>::quiet_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(cuda::std::numeric_limits<T>::signaling_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(-cuda::std::numeric_limits<T>::signaling_NaN(), T(3.0))));
|
|
|
|
// Some random value
|
|
const T expected = T(2.236067977499789805051477742381394);
|
|
assert(is_about(cuda::std::hypotl(T(1.0), T(2.0)), expected));
|
|
assert(is_about(cuda::std::hypotl(T(3.0), T(4.0)), T(5.0)));
|
|
}
|
|
#endif // _CCCL_HAS_LONG_DOUBLE()
|
|
}
|
|
|
|
template <typename T>
|
|
__host__ __device__ void test_hypot3(T val)
|
|
{
|
|
using ret = cuda::std::conditional_t<cuda::std::is_integral_v<T>, double, T>;
|
|
static_assert(cuda::std::is_same_v<decltype(cuda::std::hypot(T{}, T{}, T{})), ret>, "");
|
|
|
|
// hypot(x, y), hypot(y, x), and hypot(x, -y) are equivalent.
|
|
assert(eq(cuda::std::hypot(T(1.0), T(0.25), T(2.0)), cuda::std::hypot(T(0.25), T(1.0), T(2.0))));
|
|
if constexpr (cuda::std::is_signed_v<T>)
|
|
{
|
|
assert(eq(cuda::std::hypot(T(1.0), T(0.25), T(2.0)), cuda::std::hypot(T(1.0), T(-0.25), T(2.0))));
|
|
}
|
|
|
|
// If one of the arguments is ±0, std::hypot(x, y) is equivalent to std::fabs called with the non-zero argument.
|
|
assert(eq(cuda::std::hypot(val, T(0.25), T(2.0)), cuda::std::hypot(T(0.25), T(2.0))));
|
|
assert(eq(cuda::std::hypot(-val, T(0.25), T(2.0)), cuda::std::hypot(T(0.25), T(2.0))));
|
|
if constexpr (cuda::std::is_signed_v<T>)
|
|
{
|
|
assert(eq(cuda::std::hypot(val, T(-0.25), T(2.0)), cuda::std::hypot(T(0.25), T(2.0))));
|
|
assert(eq(cuda::std::hypot(-val, T(-0.25), T(2.0)), cuda::std::hypot(T(0.25), T(2.0))));
|
|
}
|
|
|
|
if constexpr (!cuda::std::is_integral_v<T>)
|
|
{
|
|
// If any of the arguments is NaN, NaN is returned.
|
|
assert(cuda::std::isnan(cuda::std::hypot(cuda::std::numeric_limits<T>::quiet_NaN(), T(0.25), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypot(-cuda::std::numeric_limits<T>::quiet_NaN(), T(0.25), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypot(cuda::std::numeric_limits<T>::signaling_NaN(), T(0.25), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypot(-cuda::std::numeric_limits<T>::signaling_NaN(), T(0.25), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypot(T(0.25), cuda::std::numeric_limits<T>::quiet_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypot(T(0.25), -cuda::std::numeric_limits<T>::quiet_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypot(T(0.25), cuda::std::numeric_limits<T>::signaling_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypot(T(0.25), -cuda::std::numeric_limits<T>::signaling_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypot(T(0.25), T(3.0), cuda::std::numeric_limits<T>::quiet_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypot(T(0.25), T(3.0), -cuda::std::numeric_limits<T>::quiet_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypot(T(0.25), T(3.0), cuda::std::numeric_limits<T>::signaling_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypot(T(0.25), T(3.0), -cuda::std::numeric_limits<T>::signaling_NaN())));
|
|
|
|
// Some random value
|
|
assert(is_about(cuda::std::hypot(T(2.0), T(3.0), T(6.0)), T(7.0)));
|
|
}
|
|
|
|
if constexpr (cuda::std::is_same_v<T, float>)
|
|
{
|
|
// If any of the arguments is NaN, NaN is returned.
|
|
assert(cuda::std::isnan(cuda::std::hypotf(cuda::std::numeric_limits<T>::quiet_NaN(), T(0.25), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(-cuda::std::numeric_limits<T>::quiet_NaN(), T(0.25), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(cuda::std::numeric_limits<T>::signaling_NaN(), T(0.25), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(-cuda::std::numeric_limits<T>::signaling_NaN(), T(0.25), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(T(0.25), cuda::std::numeric_limits<T>::quiet_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(T(0.25), -cuda::std::numeric_limits<T>::quiet_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(T(0.25), cuda::std::numeric_limits<T>::signaling_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(T(0.25), -cuda::std::numeric_limits<T>::signaling_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(T(0.25), T(3.0), cuda::std::numeric_limits<T>::quiet_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(T(0.25), T(3.0), -cuda::std::numeric_limits<T>::quiet_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(T(0.25), T(3.0), cuda::std::numeric_limits<T>::signaling_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotf(T(0.25), T(3.0), -cuda::std::numeric_limits<T>::signaling_NaN())));
|
|
|
|
// Some random value
|
|
assert(is_about(cuda::std::hypotf(T(2.0), T(3.0), T(6.0)), T(7.0)));
|
|
}
|
|
#if _CCCL_HAS_LONG_DOUBLE()
|
|
else if constexpr (cuda::std::is_same_v<T, long double>)
|
|
{
|
|
// If any of the arguments is NaN, NaN is returned.
|
|
assert(cuda::std::isnan(cuda::std::hypotl(cuda::std::numeric_limits<T>::quiet_NaN(), T(0.25), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(-cuda::std::numeric_limits<T>::quiet_NaN(), T(0.25), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(cuda::std::numeric_limits<T>::signaling_NaN(), T(0.25), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(-cuda::std::numeric_limits<T>::signaling_NaN(), T(0.25), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(T(0.25), cuda::std::numeric_limits<T>::quiet_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(T(0.25), -cuda::std::numeric_limits<T>::quiet_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(T(0.25), cuda::std::numeric_limits<T>::signaling_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(T(0.25), -cuda::std::numeric_limits<T>::signaling_NaN(), T(3.0))));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(T(0.25), T(3.0), cuda::std::numeric_limits<T>::quiet_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(T(0.25), T(3.0), -cuda::std::numeric_limits<T>::quiet_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(T(0.25), T(3.0), cuda::std::numeric_limits<T>::signaling_NaN())));
|
|
assert(cuda::std::isnan(cuda::std::hypotl(T(0.25), T(3.0), -cuda::std::numeric_limits<T>::signaling_NaN())));
|
|
|
|
// Some random value
|
|
assert(is_about(cuda::std::hypotl(T(2.0), T(3.0), T(6.0)), T(7.0)));
|
|
}
|
|
#endif // _CCCL_HAS_LONG_DOUBLE()
|
|
}
|
|
|
|
template <typename T>
|
|
__host__ __device__ void test(const T val)
|
|
{
|
|
test_hypot<T>(val);
|
|
test_hypot3<T>(val);
|
|
}
|
|
|
|
__host__ __device__ void test(const float val)
|
|
{
|
|
test<float>(val);
|
|
test<double>(val);
|
|
#if _CCCL_HAS_LONG_DOUBLE()
|
|
test<long double>();
|
|
#endif // _CCCL_HAS_LONG_DOUBLE()
|
|
|
|
#if _LIBCUDACXX_HAS_NVFP16()
|
|
test<__half>(val);
|
|
#endif // _LIBCUDACXX_HAS_NVFP16()
|
|
#if _LIBCUDACXX_HAS_NVBF16()
|
|
test<__nv_bfloat16>(val);
|
|
#endif // _LIBCUDACXX_HAS_NVBF16()
|
|
|
|
test<unsigned short>(static_cast<unsigned short>(val));
|
|
test<int>(static_cast<int>(val));
|
|
test<unsigned int>(static_cast<unsigned int>(val));
|
|
test<long>(static_cast<long>(val));
|
|
test<unsigned long>(static_cast<unsigned long>(val));
|
|
test<long long>(static_cast<long long>(val));
|
|
test<unsigned long long>(static_cast<unsigned long long>(val));
|
|
}
|
|
|
|
__global__ void test_global_kernel(float* val)
|
|
{
|
|
test(*val);
|
|
}
|
|
|
|
int main(int, char**)
|
|
{
|
|
volatile float val = 0.0f;
|
|
test(val);
|
|
return 0;
|
|
}
|