dedf08166a
Added 863 files from NVIDIA/cccl sparse checkout: - c2h/ (27 files): Catch2 test helpers — generators, validators, runner - nvbench_helper/ (10 files): Benchmark harness utilities - cmake/ (29 files): CMake presets and build helpers - cudax/ (794 files): Experimental CUDA extensions - AGENTS.md: NVIDIA's official AI agent instructions for CCCL - CMakePresets.json: Standardized build configurations - cccl-version.json: Version tracking Also added CCCL_ASSET_MAP.md mapping all 4295 CCCL files to competition value and PRD items. cccl_upstream now covers 100% of competition-critical assets: - 27 tuning headers (SM80/90/100 benchmark data) - 32 dispatch headers (algorithm implementations) - 60 Thrust examples (correctness verification) - 217 CUB Catch2 tests (regression matrix) - 153 CUB benchmarks (parameter space search) - 18 CUB examples (API verification) - 27 test helpers + benchmark harness - 794 cudax experimental extensions
102 lines
3.3 KiB
Plaintext
102 lines
3.3 KiB
Plaintext
//===----------------------------------------------------------------------===//
|
|
//
|
|
// Part of CUDA Experimental in CUDA C++ Core Libraries,
|
|
// 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) 2026 NVIDIA CORPORATION & AFFILIATES.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include <cuda/devices>
|
|
#include <cuda/hierarchy>
|
|
#include <cuda/launch>
|
|
#include <cuda/std/cstddef>
|
|
#include <cuda/std/type_traits>
|
|
#include <cuda/std/utility>
|
|
#include <cuda/stream>
|
|
#include <cuda/warp>
|
|
|
|
#include <cuda/experimental/group.cuh>
|
|
|
|
#include "group_testing.cuh"
|
|
|
|
namespace
|
|
{
|
|
template <class Config>
|
|
__device__ void test_identity_mapping(Config config)
|
|
{
|
|
using Mapping = cudax::identity_mapping;
|
|
|
|
// Test default constructor.
|
|
{
|
|
static_assert(cuda::std::is_trivially_default_constructible_v<Mapping>);
|
|
static_assert(cuda::std::is_empty_v<Mapping>);
|
|
|
|
[[maybe_unused]] cudax::identity_mapping mapping;
|
|
}
|
|
|
|
// Test map(...).
|
|
{
|
|
const cudax::this_warp parent_group{config};
|
|
const ThreadsInWarpMappingResult prev_mapping_result;
|
|
|
|
static_assert(cudax::__group_mapping_result<decltype(cuda::std::declval<const Mapping>().map(
|
|
cuda::gpu_thread, parent_group, prev_mapping_result))>);
|
|
static_assert(
|
|
noexcept(cuda::std::declval<const Mapping>().map(cuda::gpu_thread, parent_group, prev_mapping_result)));
|
|
|
|
const Mapping mapping;
|
|
auto result = mapping.map(cuda::gpu_thread, parent_group, prev_mapping_result);
|
|
using Result = decltype(result);
|
|
|
|
static_assert(cuda::std::is_same_v<Result, ThreadsInWarpMappingResult>);
|
|
|
|
static_assert(Result::static_group_count() == ThreadsInWarpMappingResult::static_group_count());
|
|
CHECK(result.group_count() == prev_mapping_result.group_count());
|
|
CHECK(result.group_rank() == prev_mapping_result.group_rank());
|
|
|
|
static_assert(Result::static_unit_count() == ThreadsInWarpMappingResult::static_unit_count());
|
|
CHECK(result.unit_count() == prev_mapping_result.unit_count());
|
|
CHECK(result.unit_rank() == prev_mapping_result.unit_rank());
|
|
|
|
CHECK(result.lane_mask() == prev_mapping_result.lane_mask());
|
|
|
|
CHECK(result.is_valid() == prev_mapping_result.is_valid());
|
|
static_assert(Result::is_always_exhaustive() == ThreadsInWarpMappingResult::is_always_exhaustive());
|
|
static_assert(Result::is_always_contiguous() == ThreadsInWarpMappingResult::is_always_contiguous());
|
|
}
|
|
}
|
|
|
|
struct TestKernel
|
|
{
|
|
template <class Config>
|
|
__device__ void operator()(const Config& config)
|
|
{
|
|
test_identity_mapping(config);
|
|
test_identity_mapping(config);
|
|
test_identity_mapping(config);
|
|
test_identity_mapping(config);
|
|
test_identity_mapping(config);
|
|
}
|
|
};
|
|
} // namespace
|
|
|
|
C2H_TEST("Identity mapping", "[group]")
|
|
{
|
|
const auto device = cuda::devices[0];
|
|
|
|
const cuda::stream stream{device};
|
|
|
|
{
|
|
const auto config = cuda::make_config(cuda::grid_dims<1>(), cuda::block_dims<8, 4>());
|
|
cuda::launch(stream, config, TestKernel{});
|
|
}
|
|
{
|
|
const auto config = cuda::make_config(cuda::grid_dims<1>(), cuda::block_dims(dim3{8, 4}));
|
|
cuda::launch(stream, config, TestKernel{});
|
|
}
|
|
|
|
stream.sync();
|
|
}
|