fb335f6a5f
CUTLASS 2.0 Substantially refactored for - Better performance, particularly for native Turing Tensor Cores - Robust and durable templates spanning the design space - Encapsulated functionality embodying modern C++11 programming techniques - Optimized containers and data types for efficient, generic, portable device code Updates to: - Quick start guide - Documentation - Utilities - CUTLASS Profiler Native Turing Tensor Cores - Efficient GEMM kernels targeting Turing Tensor Cores - Mixed-precision floating point, 8-bit integer, 4-bit integer, and binarized operands Coverage of existing CUTLASS functionality: - GEMM kernels targeting CUDA and Tensor Cores in NVIDIA GPUs - Volta Tensor Cores through native mma.sync and through WMMA API - Optimizations such as parallel reductions, threadblock rasterization, and intra-threadblock reductions - Batched GEMM operations - Complex-valued GEMMs Note: this commit and all that follow require a host compiler supporting C++11 or greater.
190 lines
5.9 KiB
Plaintext
190 lines
5.9 KiB
Plaintext
/***************************************************************************************************
|
|
* Copyright (c) 2017-2019, NVIDIA CORPORATION. All rights reserved.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without modification, are permitted
|
|
* provided that the following conditions are met:
|
|
* * Redistributions of source code must retain the above copyright notice, this list of
|
|
* conditions and the following disclaimer.
|
|
* * Redistributions in binary form must reproduce the above copyright notice, this list of
|
|
* conditions and the following disclaimer in the documentation and/or other materials
|
|
* provided with the distribution.
|
|
* * Neither the name of the NVIDIA CORPORATION nor the names of its contributors may be used
|
|
* to endorse or promote products derived from this software without specific prior written
|
|
* permission.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR
|
|
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND
|
|
* FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
|
|
* BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
|
|
* OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
|
|
* STRICT LIABILITY, OR TOR (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
|
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
*
|
|
**************************************************************************************************/
|
|
/*! \file
|
|
\brief Tests for device-wide GEMM interface
|
|
*/
|
|
|
|
#include <iostream>
|
|
|
|
#include "cutlass/cutlass.h"
|
|
#include "cutlass/gemm/device/gemm.h"
|
|
|
|
#include "../../common/cutlass_unit_test.h"
|
|
|
|
#include "cutlass/util/host_tensor.h"
|
|
#include "cutlass/util/tensor_view_io.h"
|
|
#include "cutlass/util/reference/host/tensor_fill.h"
|
|
#include "cutlass/util/reference/host/tensor_copy.h"
|
|
#include "cutlass/util/reference/host/tensor_compare.h"
|
|
#include "cutlass/util/reference/host/gemm.h"
|
|
|
|
#include "testbed.h"
|
|
|
|
////////////////////////////////////////////////////////////////////////////////
|
|
|
|
/////////////////////////////////////
|
|
// NT
|
|
/////////////////////////////////////
|
|
|
|
TEST(SM61_Device_Gemm_s8n_s8t_simt_op_dp4a_perf, 128x256x32_64x64x8) {
|
|
|
|
using ElementOutput = int8_t;
|
|
using ElementAccumulator = int32_t;
|
|
using ElementCompute = float;
|
|
|
|
using Gemm = cutlass::gemm::device::Gemm<
|
|
int8_t,
|
|
cutlass::layout::ColumnMajor,
|
|
int8_t,
|
|
cutlass::layout::RowMajor,
|
|
ElementOutput,
|
|
cutlass::layout::RowMajor,
|
|
int32_t,
|
|
cutlass::arch::OpClassSimt,
|
|
cutlass::arch::Sm61,
|
|
cutlass::gemm::GemmShape<128, 256, 32>,
|
|
cutlass::gemm::GemmShape<64, 64, 32>,
|
|
cutlass::gemm::GemmShape<1, 1, 4>,
|
|
cutlass::epilogue::thread::LinearCombinationClamp<
|
|
ElementOutput,
|
|
1,
|
|
ElementAccumulator,
|
|
ElementCompute
|
|
>,
|
|
cutlass::gemm::threadblock::GemmIdentityThreadblockSwizzle,
|
|
2
|
|
>;
|
|
|
|
EXPECT_TRUE(test::gemm::device::TestGemmPerf<Gemm>());
|
|
}
|
|
|
|
/////////////////////////////////////
|
|
// TT
|
|
/////////////////////////////////////
|
|
|
|
TEST(SM61_Device_Gemm_s8t_s8t_simt_op_dp4a_perf, 128x256x32_64x64x8) {
|
|
|
|
using ElementOutput = int32_t;
|
|
using ElementAccumulator = int32_t;
|
|
using ElementCompute = float;
|
|
|
|
using Gemm = cutlass::gemm::device::Gemm<
|
|
int8_t,
|
|
cutlass::layout::RowMajor,
|
|
int8_t,
|
|
cutlass::layout::RowMajor,
|
|
ElementOutput,
|
|
cutlass::layout::RowMajor,
|
|
int32_t,
|
|
cutlass::arch::OpClassSimt,
|
|
cutlass::arch::Sm61,
|
|
cutlass::gemm::GemmShape<128, 256, 32>,
|
|
cutlass::gemm::GemmShape<64, 64, 32>,
|
|
cutlass::gemm::GemmShape<1, 1, 4>,
|
|
cutlass::epilogue::thread::LinearCombinationClamp<
|
|
ElementOutput,
|
|
1,
|
|
ElementAccumulator,
|
|
ElementCompute
|
|
>,
|
|
cutlass::gemm::threadblock::GemmIdentityThreadblockSwizzle,
|
|
2
|
|
>;
|
|
|
|
EXPECT_TRUE(test::gemm::device::TestGemmPerf<Gemm>());
|
|
}
|
|
|
|
/////////////////////////////////////
|
|
// NN
|
|
/////////////////////////////////////
|
|
|
|
TEST(SM61_Device_Gemm_s8n_s8n_simt_op_dp4a_perf, 128x256x32_64x64x8) {
|
|
|
|
using ElementOutput = float;
|
|
using ElementAccumulator = int32_t;
|
|
using ElementCompute = float;
|
|
|
|
using Gemm = cutlass::gemm::device::Gemm<
|
|
int8_t,
|
|
cutlass::layout::ColumnMajor,
|
|
int8_t,
|
|
cutlass::layout::ColumnMajor,
|
|
ElementOutput,
|
|
cutlass::layout::RowMajor,
|
|
int32_t,
|
|
cutlass::arch::OpClassSimt,
|
|
cutlass::arch::Sm61,
|
|
cutlass::gemm::GemmShape<128, 256, 32>,
|
|
cutlass::gemm::GemmShape<64, 64, 32>,
|
|
cutlass::gemm::GemmShape<1, 1, 4>,
|
|
cutlass::epilogue::thread::LinearCombinationClamp<
|
|
ElementOutput,
|
|
1,
|
|
ElementAccumulator,
|
|
ElementCompute
|
|
>,
|
|
cutlass::gemm::threadblock::GemmIdentityThreadblockSwizzle,
|
|
2
|
|
>;
|
|
|
|
EXPECT_TRUE(test::gemm::device::TestGemmPerf<Gemm>());
|
|
}
|
|
|
|
/////////////////////////////////////
|
|
// TN
|
|
/////////////////////////////////////
|
|
|
|
TEST(SM61_Device_Gemm_s8t_s8n_simt_op_dp4a_perf, 128x256x32_64x64x8) {
|
|
|
|
using ElementOutput = int8_t;
|
|
using ElementAccumulator = int32_t;
|
|
using ElementCompute = float;
|
|
|
|
using Gemm = cutlass::gemm::device::Gemm<
|
|
int8_t,
|
|
cutlass::layout::RowMajor,
|
|
int8_t,
|
|
cutlass::layout::ColumnMajor,
|
|
ElementOutput,
|
|
cutlass::layout::RowMajor,
|
|
int32_t,
|
|
cutlass::arch::OpClassSimt,
|
|
cutlass::arch::Sm61,
|
|
cutlass::gemm::GemmShape<128, 256, 32>,
|
|
cutlass::gemm::GemmShape<64, 64, 32>,
|
|
cutlass::gemm::GemmShape<1, 1, 4>,
|
|
cutlass::epilogue::thread::LinearCombinationClamp<
|
|
ElementOutput,
|
|
1,
|
|
ElementAccumulator,
|
|
ElementCompute
|
|
>,
|
|
cutlass::gemm::threadblock::GemmIdentityThreadblockSwizzle,
|
|
2
|
|
>;
|
|
|
|
EXPECT_TRUE(test::gemm::device::TestGemmPerf<Gemm>());
|
|
}
|