mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-06-29 19:28:33 +00:00
717f2efef7
[CK] add composable kernel support on gfx1250 (#6978) ## Motivation Add composable kernel support on gfx1250. ## Technical Details <!-- Explain the changes along with any relevant GitHub links. --> ## Test Plan <!-- Explain any relevant testing done to verify this PR. --> ## Test Result <!-- Briefly summarize test outcomes. --> ## Submission Checklist - [ ] Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests. --------- Co-authored-by: Qun Lin <qlin@amd.com> Co-authored-by: jialuo12_amdeng <jia.luo@amd.com> Co-authored-by: Andriy Roshchenko <andriy.roshchenko@amd.com> Co-authored-by: hsivasun_amdeng <haresh.sivasuntharampillai@amd.com>
50 lines
1.6 KiB
C++
50 lines
1.6 KiB
C++
// SPDX-License-Identifier: MIT
|
|
// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
#define CK_ENABLE_DYNAMIC_WARP_SIZE 1
|
|
#include "common.hpp"
|
|
|
|
#include "ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp"
|
|
|
|
using ADataType = float;
|
|
using BDataType = float;
|
|
using AccDataType = float;
|
|
using CShuffleDataType = float;
|
|
using CDataType = float;
|
|
|
|
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 DeviceGemmV2Instance =
|
|
ck::tensor_operation::device::DeviceGemm_Xdl_CShuffleV3<
|
|
ALayout, BLayout, CLayout,
|
|
ADataType, BDataType, CDataType, AccDataType, CShuffleDataType,
|
|
PassThrough, PassThrough, PassThrough, GemmDefault,
|
|
128,
|
|
64, 64,
|
|
64, 4, 4,
|
|
16, 16,
|
|
2, 4,
|
|
S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>,
|
|
2, 2, 2, 0,
|
|
S<4, 32, 1>, S<1, 0, 2>, S<1, 0, 2>,
|
|
2, 2, 2, 0,
|
|
1, 2, S<1, 32, 1, 4>, 2,
|
|
ck::BlockGemmPipelineScheduler::Intrawave,ck::BlockGemmPipelineVersion::v3>;
|
|
// clang-format on
|
|
|
|
using ReferenceGemmInstance = ck::tensor_operation::host::
|
|
ReferenceGemm<ADataType, BDataType, CDataType, AccDataType, AElementOp, BElementOp, CElementOp>;
|
|
|
|
#include "run_gemm_example_v2.inc"
|
|
|
|
int main(int argc, char* argv[]) { return !run_gemm_splitk_example(argc, argv); }
|