mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-06-29 11:16:59 +00:00
3aee45e115
[CK_TILE] Add CShuffleLds microbenchmark suite (#5383) ## Summary Microbenchmarks isolating LDS store/load operations in CShuffleEpilogue for bank conflict analysis. ## Motivation CShuffleEpilogue performs LDS store (MFMA registers → LDS) and load (LDS → registers for coalesced global writes). This suite isolates each operation to: - Identify which operation causes bank conflicts - Measure pure LDS bandwidth per access pattern - Validate access patterns across MFMA tile sizes and wave layouts ## Components - **Microkernels** (`tile_load_store_microkernels.hpp`): `StoreTile<Setup>`, `LoadTile<Setup>` - **Setup Adapters** (`benchmark_cshuffle_lds.hpp`): Wire CShuffleEpilogue to microkernels - **Template** (`benchmark_template.cpp.in`): Generated benchmarks with timing ## Build ```bash cmake -G Ninja -B build -S . \ -DGPU_TARGETS=gfx950 \ -DBUILD_CK_EXAMPLES=ON \ -DBUILD_CK_TILE_CSHUFFLE_LDS_BENCHMARKS=ON ninja -C build bench_lds_fp8_16x16x128_2x2_fp8 ``` ## New CMake Options | Option | Default | Description | |--------|---------|-------------| | `BUILD_CK_TILE_CSHUFFLE_LDS_BENCHMARKS` | OFF | LDS microbenchmarks | | `BUILD_CK_TILE_FMHA_TESTS` | ON | FMHA tests | | `BUILD_CK_TILE_ENGINE` | ON | Tile engine | | `BUILD_CK_TILE_ENGINE_TESTS` | ON | Tile engine tests | | `BUILD_CK_EXAMPLES` | ON | Examples | | `BUILD_CK_TUTORIALS` | ON | Tutorials | | `BUILD_CK_DEVICE_INSTANCES` | ON | Device instances | | `BUILD_CK_PROFILER` | ON | Profiler | Setting guards to OFF reduces cmake configure from ~150s to ~5s. --------- Made-with: Claude Code, Opus 4.5
101 lines
3.2 KiB
C++
101 lines
3.2 KiB
C++
// Copyright (c) Advanced Micro Devices, Inc., or its affiliates.
|
|
// SPDX-License-Identifier: MIT
|
|
|
|
// clang-format off
|
|
|
|
#include "benchmark_cshuffle_lds.hpp"
|
|
#include "ck_tile/host/kernel_launch.hpp"
|
|
#include <iostream>
|
|
#include <cstdlib>
|
|
#include <cstring>
|
|
|
|
using Epilogue = ck_tile::BenchmarkEpilogue<
|
|
@A_TYPE@, @B_TYPE@, @ACC_TYPE@, @O_TYPE@,
|
|
@M@, @N@, @M_WAVE@, @N_WAVE@, @M_XDL@, @N_XDL@, @K_XDL@>;
|
|
|
|
using StoreSetup = ck_tile::LdsStoreSetup<Epilogue>;
|
|
using LoadSetup = ck_tile::LdsLoadSetup<Epilogue>;
|
|
|
|
void print_help(const char* prog)
|
|
{
|
|
std::cout << "Usage: " << prog << " [options]\n"
|
|
<< "\n"
|
|
<< "LDS microbenchmark for CShuffleEpilogue (@CONFIG_NAME@)\n"
|
|
<< "\n"
|
|
<< "Options:\n"
|
|
<< " -w, --warmup <N> Warmup iterations (default: 3)\n"
|
|
<< " -i, --iters <N> Benchmark iterations (default: 10)\n"
|
|
<< " -h, --help Show this help message\n"
|
|
<< "\n"
|
|
<< "Configuration:\n"
|
|
<< " MFMA tile: @M_XDL@x@N_XDL@x@K_XDL@\n"
|
|
<< " Wave layout: @M_WAVE@x@N_WAVE@\n"
|
|
<< " Block tile: @M@x@N@\n"
|
|
<< std::endl;
|
|
}
|
|
|
|
int main(int argc, char** argv)
|
|
{
|
|
int warmup = 3;
|
|
int iters = 10;
|
|
|
|
for (int i = 1; i < argc; ++i)
|
|
{
|
|
if (std::strcmp(argv[i], "-h") == 0 || std::strcmp(argv[i], "--help") == 0)
|
|
{
|
|
print_help(argv[0]);
|
|
return 0;
|
|
}
|
|
else if ((std::strcmp(argv[i], "-w") == 0 || std::strcmp(argv[i], "--warmup") == 0) && i + 1 < argc)
|
|
{
|
|
int val = std::atoi(argv[++i]);
|
|
if (val <= 0)
|
|
{
|
|
std::cerr << "Error: --warmup requires a positive integer\n";
|
|
return 1;
|
|
}
|
|
warmup = val;
|
|
}
|
|
else if ((std::strcmp(argv[i], "-i") == 0 || std::strcmp(argv[i], "--iters") == 0) && i + 1 < argc)
|
|
{
|
|
int val = std::atoi(argv[++i]);
|
|
if (val <= 0)
|
|
{
|
|
std::cerr << "Error: --iters requires a positive integer\n";
|
|
return 1;
|
|
}
|
|
iters = val;
|
|
}
|
|
else
|
|
{
|
|
std::cerr << "Unknown option: " << argv[i] << "\n";
|
|
print_help(argv[0]);
|
|
return 1;
|
|
}
|
|
}
|
|
|
|
std::cout << "=== @CONFIG_NAME@ ===" << std::endl;
|
|
|
|
ck_tile::stream_config stream{nullptr, true, 0, warmup, iters, true};
|
|
|
|
// Store benchmark
|
|
{
|
|
float ms = ck_tile::launch_kernel(stream,
|
|
ck_tile::make_kernel(ck_tile::StoreTile<StoreSetup>{},
|
|
dim3(1), dim3(StoreSetup::kBlockSize), 0));
|
|
double gb_s = (double(StoreSetup::kBytes) / 1e9) / (ms / 1e3);
|
|
std::cout << "Store: " << ms << " ms, " << gb_s << " GB/s" << std::endl;
|
|
}
|
|
|
|
// Load benchmark
|
|
{
|
|
float ms = ck_tile::launch_kernel(stream,
|
|
ck_tile::make_kernel(ck_tile::LoadTile<LoadSetup>{},
|
|
dim3(1), dim3(LoadSetup::kBlockSize), 0));
|
|
double gb_s = (double(LoadSetup::kBytes) / 1e9) / (ms / 1e3);
|
|
std::cout << "Load: " << ms << " ms, " << gb_s << " GB/s" << std::endl;
|
|
}
|
|
|
|
return 0;
|
|
}
|