// Copyright (c) Advanced Micro Devices, Inc., or its affiliates. // SPDX-License-Identifier: MIT #include "common.hpp" #include "ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_streamk_v3.hpp" using ADataType = ck::f8_t; using BDataType = ck::f8_t; using AccDataType = float; using CShuffleDataType = ck::half_t; using CDataType = ck::half_t; using ALayout = Row; using BLayout = Col; using CLayout = Row; using AElementOp = PassThrough; using BElementOp = PassThrough; using CElementOp = PassThrough; static constexpr auto GemmDefault = ck::tensor_operation::device::GemmSpecialization::Default; // clang-format off using DeviceGemmV2_Streamk_Instance = ck::tensor_operation::device::DeviceGemm_Xdl_CShuffle_Streamk_V3< ALayout, BLayout, CLayout, ADataType, BDataType, CDataType, AccDataType, CShuffleDataType, PassThrough, PassThrough, PassThrough, GemmDefault, 256, 128, 256, 128, 16, 16, 16, 16, 4, 8, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 16, 16, 1, S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 16, 16, 1, 1, 2, S<1, 32, 1, 8>, 8, ck::BlockGemmPipelineScheduler::Intrawave,ck::BlockGemmPipelineVersion::v3, ck::f8_t>; // clang-format on using ReferenceGemmInstance = ck::tensor_operation::host:: ReferenceGemm; using ReferenceGemmInstanceGPU = ck::tensor_operation::device::ReferenceGemm; #include "run_gemm_example_streamk_v2.inc" int main(int argc, char* argv[]) { return !run_gemm_universal_streamk_example(argc, argv); }