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
76 lines
2.1 KiB
Plaintext
76 lines
2.1 KiB
Plaintext
//===----------------------------------------------------------------------===//
|
|
//
|
|
// Part of CUDASTF 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) 2022-2024 NVIDIA CORPORATION & AFFILIATES.
|
|
//
|
|
//===----------------------------------------------------------------------===//
|
|
|
|
#include <cuda/experimental/__places/partitions/blocked_partition.cuh>
|
|
#include <cuda/experimental/__places/partitions/cyclic_shape.cuh>
|
|
#include <cuda/experimental/stf.cuh>
|
|
|
|
using namespace cuda::experimental::stf;
|
|
|
|
double X0(int i)
|
|
{
|
|
return sin((double) i);
|
|
}
|
|
|
|
double Y0(int i)
|
|
{
|
|
return cos((double) i);
|
|
}
|
|
|
|
int main()
|
|
{
|
|
stream_ctx ctx;
|
|
|
|
const int N = 128;
|
|
double X[N], Y[N];
|
|
|
|
for (int ind = 0; ind < N; ind++)
|
|
{
|
|
X[ind] = X0(ind);
|
|
Y[ind] = Y0(ind);
|
|
}
|
|
|
|
const double alpha = 3.14;
|
|
|
|
auto handle_X = ctx.logical_data(X, {N});
|
|
auto handle_Y = ctx.logical_data(Y, {N});
|
|
|
|
auto number_devices = 4;
|
|
auto all_devs = exec_place::repeat(exec_place::device(0), number_devices);
|
|
|
|
auto spec = par(16 * 4, par(4));
|
|
ctx.launch(spec, all_devs, handle_X.read(), handle_Y.rw())->*[=] _CCCL_DEVICE(auto th, auto x, auto y) {
|
|
// Blocked partition among elements in the outer most level
|
|
auto outer_sh = blocked_partition::apply(shape(x), pos4(th.rank(0)), dim4(th.size(0)));
|
|
|
|
// Cyclic partition among elements in the remaining levels
|
|
auto inner_sh = cyclic_partition::apply(outer_sh, pos4(th.inner().rank()), dim4(th.inner().size()));
|
|
|
|
for (auto ind : inner_sh)
|
|
{
|
|
y(ind) += alpha * x(ind);
|
|
}
|
|
};
|
|
|
|
ctx.host_launch(handle_X.read(), handle_Y.read())->*[=](auto X, auto Y) {
|
|
for (int ind = 0; ind < N; ind++)
|
|
{
|
|
// Y should be Y0 + alpha X0
|
|
// fprintf(stderr, "Y[%ld] = %lf - expect %lf\n", ind, Y(ind), (Y0(ind) + alpha * X0(ind)));
|
|
EXPECT(fabs(Y(ind) - (Y0(ind) + alpha * X0(ind))) < 0.0001);
|
|
|
|
// X should be X0
|
|
EXPECT(fabs(X(ind) - X0(ind)) < 0.0001);
|
|
}
|
|
};
|
|
|
|
ctx.finalize();
|
|
}
|