56fd68e7dd
CCCL (CUDA C++ Core Libraries) provides: - CUB: device/block/warp-level GPU primitives (reduce, scan, sort, topk) - Thrust: high-level parallel algorithms (transform_reduce, sort, scan) - libcudacxx: CUDA C++ standard library (atomics, barriers, memory) - cudax: experimental features (memory resources, allocators) - Tuning policies: per-SM hardware-specific algorithm parameters Competition optimization vectors mapped to CCCL: - Output TPS (83% weight): warp_reduce, block_reduce, device_topk - Input TPS (14% weight): device_scan, block_load, prefetch - Cache TPS (3% weight): prefix caching strategy patterns - Memory (0.9 util): pooled/cached/buddy allocators Source: https://github.com/NVIDIA/cccl (shallow clone, HEAD only) License: Apache-2.0
47 lines
2.6 KiB
Plaintext
47 lines
2.6 KiB
Plaintext
// SPDX-FileCopyrightText: Copyright (c) 2024, NVIDIA CORPORATION. All rights reserved.
|
|
// SPDX-License-Identifier: BSD-3
|
|
|
|
#include <cub/util_math.cuh>
|
|
|
|
#include <cuda/std/type_traits>
|
|
|
|
#include <c2h/catch2_test_helper.h>
|
|
|
|
C2H_TEST("Tests safe_add_bound_to_max", "[util][math]")
|
|
{
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(0U, cuda::std::numeric_limits<std::uint32_t>::max())
|
|
== cuda::std::numeric_limits<std::uint32_t>::max());
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(cuda::std::numeric_limits<std::uint32_t>::max(), 0U)
|
|
== cuda::std::numeric_limits<std::uint32_t>::max());
|
|
|
|
// We do not overflow
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(std::int32_t{0}, cuda::std::numeric_limits<std::int32_t>::max())
|
|
== cuda::std::numeric_limits<std::int32_t>::max());
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(cuda::std::numeric_limits<std::int32_t>::max(), std::int32_t{0})
|
|
== cuda::std::numeric_limits<std::int32_t>::max());
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(std::int32_t{1}, cuda::std::numeric_limits<std::int32_t>::max())
|
|
== cuda::std::numeric_limits<std::int32_t>::max());
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(cuda::std::numeric_limits<std::int32_t>::max(), std::int32_t{1})
|
|
== cuda::std::numeric_limits<std::int32_t>::max());
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(
|
|
cuda::std::numeric_limits<std::int32_t>::max(), cuda::std::numeric_limits<std::int32_t>::max())
|
|
== cuda::std::numeric_limits<std::int32_t>::max());
|
|
|
|
// We do not overflow
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(std::int64_t{0}, cuda::std::numeric_limits<std::int64_t>::max())
|
|
== cuda::std::numeric_limits<std::int64_t>::max());
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(cuda::std::numeric_limits<std::int64_t>::max(), std::int64_t{0LL})
|
|
== cuda::std::numeric_limits<std::int64_t>::max());
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(std::int64_t{1LL}, cuda::std::numeric_limits<std::int64_t>::max())
|
|
== cuda::std::numeric_limits<std::int64_t>::max());
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(cuda::std::numeric_limits<std::int64_t>::max(), std::int64_t{1LL})
|
|
== cuda::std::numeric_limits<std::int64_t>::max());
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(
|
|
cuda::std::numeric_limits<std::int64_t>::max(), cuda::std::numeric_limits<std::int64_t>::max())
|
|
== cuda::std::numeric_limits<std::int64_t>::max());
|
|
|
|
// We do not underflow for negative rhs (not, lhs must not be negative per documentation)
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(0, -1) == -1);
|
|
REQUIRE(cub::detail::safe_add_bound_to_max(1, -1) == 0);
|
|
}
|