mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-04-20 14:59:17 +00:00
Clean up conv example, Instances, profiler and test (#324)
* convnd_fwd fp16 example * update example * update example * update instance * updating refernce conv * update reference conv * update conv fwd profiler * update conv 1d and 3d instance * update include path * clean * update profiler for conv bwd data and weight * update conv bwd weight * clean * update conv example * update profiler for conv bwd weight * update ckprofiler for conv bwd data * fix reference conv bwd data bug; update conv bwd data test * update examples * fix initialization issue * update test for conv fwd * clean * clean * remove test case too sensitive to error threshhold * fix test * clean * fix build * adding conv multiple d * adding conv multiple D * add matrix padder * add gemm padding to convnd * adding group conv * update gemm multi-d * refactor * refactor * refactor * clean * clean * refactor * refactor * reorg * add ds * add bias * clean * add G * adding group * adding group * adding group * update Tensor * clean * update example * update DeviceGemmMultipleD_Xdl_CShuffle * update conv bwd-data and bwd-weight * upate contraction example * update gemm and batch gemm with e permute * fix example build * instance for grouped conv1d * update example * adding group conv instance * update gemm bilinear instance * update gemm+add+add+fastgelu instance * update profiler * update profiler * update test * update test and client example * clean * add grouped conv into profiler * update profiler * clean * add test grouped conv, update all conv test to gtest * update test
This commit is contained in:
Chao Liu
2
example/17_convnd_bwd_data/CMakeLists.txt
Normal file
2
example/17_convnd_bwd_data/CMakeLists.txt
Normal file
@@ -0,0 +1,2 @@
|
||||
add_example_executable(example_convnd_bwd_data_xdl_fp16 convnd_bwd_data_xdl_fp16.cpp)
|
||||
target_link_libraries(example_convnd_bwd_data_xdl_fp16 PRIVATE utility)
|
||||
47
example/17_convnd_bwd_data/README.md
Normal file
47
example/17_convnd_bwd_data/README.md
Normal file
@@ -0,0 +1,47 @@
|
||||
# Instructions for ```example_convnd_bwd_data_xdl```
|
||||
|
||||
## Run ```example_example_convnd_bwd_data_xdl```
|
||||
```bash
|
||||
#arg1: verification (0=no, 1=yes)
|
||||
#arg2: initialization (0=no init, 1=integer value, 2=decimal value)
|
||||
#arg3: run kernel # of times (>1)
|
||||
#arg4: num_dim_spatial(1|2|3)
|
||||
#arg5 to ...: N, K, C, [Z,] [Y,] X, [Di,] [Hi,] Wi, S[z,] [Sy,] Sx, [Dz,] [Dy,] Dx, [LeftPz,] [LeftPy,] LeftPx, [RightPy,] [RightPy,] RightPx
|
||||
./bin/example_convnd_bwd_data_xdl 0 1 5
|
||||
```
|
||||
|
||||
Result
|
||||
```
|
||||
in_n_c_hi_wi: dim 4, lengths {128, 128, 71, 71}, strides {645248, 1, 9088, 128}
|
||||
wei_k_c_y_x: dim 4, lengths {256, 128, 3, 3}, strides {1152, 1, 384, 128}
|
||||
out_n_k_ho_wo: dim 4, lengths {128, 256, 36, 36}, strides {331776, 1, 9216, 256}
|
||||
arg.a_grid_desc_k0_m_k1_container_{128, 175232, 8}
|
||||
arg.b_grid_desc_k0_n_k1_container_{128, 128, 8}
|
||||
arg.c_grid_desc_m_n_container_{ 175232, 128}
|
||||
arg.c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2_container_( 2738, 2, 2, 2, 4, 2 )
|
||||
launch_and_time_kernel: grid_dim {1369, 1, 1}, block_dim {256, 1, 1}
|
||||
Warm up
|
||||
Start running 1 times...
|
||||
arg.a_grid_desc_k0_m_k1_container_{64, 175232, 8}
|
||||
arg.b_grid_desc_k0_n_k1_container_{64, 128, 8}
|
||||
arg.c_grid_desc_m_n_container_{ 175232, 128}
|
||||
arg.c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2_container_( 2738, 2, 2, 2, 4, 2 )
|
||||
launch_and_time_kernel: grid_dim {1369, 1, 1}, block_dim {256, 1, 1}
|
||||
Warm up
|
||||
Start running 1 times...
|
||||
arg.a_grid_desc_k0_m_k1_container_{64, 175232, 8}
|
||||
arg.b_grid_desc_k0_n_k1_container_{64, 128, 8}
|
||||
arg.c_grid_desc_m_n_container_{ 175232, 128}
|
||||
arg.c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2_container_( 2738, 2, 2, 2, 4, 2 )
|
||||
launch_and_time_kernel: grid_dim {1369, 1, 1}, block_dim {256, 1, 1}
|
||||
Warm up
|
||||
Start running 1 times...
|
||||
arg.a_grid_desc_k0_m_k1_container_{32, 175232, 8}
|
||||
arg.b_grid_desc_k0_n_k1_container_{32, 128, 8}
|
||||
arg.c_grid_desc_m_n_container_{ 175232, 128}
|
||||
arg.c_grid_desc_m0_n0_m1_n1_m2_m3_m4_n2_container_( 2738, 2, 2, 2, 4, 2 )
|
||||
launch_and_time_kernel: grid_dim {1369, 1, 1}, block_dim {256, 1, 1}
|
||||
Warm up
|
||||
Start running 1 times...
|
||||
Perf: 1.40031 ms, 69.8734 TFlops, 179.037 GB/s
|
||||
```
|
||||
149
example/17_convnd_bwd_data/convnd_bwd_data_common.hpp
Normal file
149
example/17_convnd_bwd_data/convnd_bwd_data_common.hpp
Normal file
@@ -0,0 +1,149 @@
|
||||
// SPDX-License-Identifier: MIT
|
||||
// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
|
||||
|
||||
#include <iostream>
|
||||
#include <numeric>
|
||||
#include <initializer_list>
|
||||
#include <cstdlib>
|
||||
|
||||
#include "ck/ck.hpp"
|
||||
#include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
|
||||
#include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
|
||||
|
||||
#include "ck/library/utility/check_err.hpp"
|
||||
#include "ck/library/utility/device_memory.hpp"
|
||||
#include "ck/library/utility/host_tensor.hpp"
|
||||
#include "ck/library/utility/host_tensor_generator.hpp"
|
||||
#include "ck/library/utility/convolution_parameter.hpp"
|
||||
#include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
|
||||
#include "ck/library/reference_tensor_operation/cpu/reference_conv_bwd_data.hpp"
|
||||
|
||||
void print_helper_msg()
|
||||
{
|
||||
std::cout << "arg1: verification (0=no, 1=yes)\n"
|
||||
<< "arg2: initialization (0=no init, 1=integer value, 2=decimal value)\n"
|
||||
<< "arg3: time kernel (0=no, 1=yes)\n"
|
||||
<< ck::utils::conv::get_conv_param_parser_helper_msg() << std::endl;
|
||||
}
|
||||
|
||||
template <ck::index_t NDimSpatial,
|
||||
typename InDataType,
|
||||
typename WeiDataType,
|
||||
typename OutDataType,
|
||||
typename InElementOp,
|
||||
typename WeiElementOp,
|
||||
typename OutElementOp,
|
||||
typename DeviceConvNdBwdDataInstance>
|
||||
int run_conv_bwd_data(bool do_verification,
|
||||
int init_method,
|
||||
bool time_kernel,
|
||||
const ck::utils::conv::ConvParam& conv_param,
|
||||
const HostTensorDescriptor& in_g_n_c_wis_desc,
|
||||
const HostTensorDescriptor& wei_g_k_c_xs_desc,
|
||||
const HostTensorDescriptor& out_g_n_k_wos_desc,
|
||||
const InElementOp& in_element_op,
|
||||
const WeiElementOp& wei_element_op,
|
||||
const OutElementOp& out_element_op)
|
||||
{
|
||||
Tensor<InDataType> in_host(in_g_n_c_wis_desc);
|
||||
Tensor<InDataType> in_device(in_g_n_c_wis_desc);
|
||||
Tensor<WeiDataType> wei(wei_g_k_c_xs_desc);
|
||||
Tensor<OutDataType> out(out_g_n_k_wos_desc);
|
||||
|
||||
std::cout << "in: " << in_host.mDesc << std::endl;
|
||||
std::cout << "wei: " << wei.mDesc << std::endl;
|
||||
std::cout << "out: " << out.mDesc << std::endl;
|
||||
|
||||
switch(init_method)
|
||||
{
|
||||
case 0: break;
|
||||
case 1:
|
||||
out.GenerateTensorValue(GeneratorTensor_2<OutDataType>{-5, 5});
|
||||
wei.GenerateTensorValue(GeneratorTensor_2<WeiDataType>{-5, 5});
|
||||
break;
|
||||
default:
|
||||
out.GenerateTensorValue(GeneratorTensor_3<OutDataType>{0.0, 1.0});
|
||||
wei.GenerateTensorValue(GeneratorTensor_3<WeiDataType>{-0.5, 0.5});
|
||||
}
|
||||
|
||||
DeviceMem in_device_buf(sizeof(InDataType) * in_device.mDesc.GetElementSpaceSize());
|
||||
DeviceMem wei_device_buf(sizeof(WeiDataType) * wei.mDesc.GetElementSpaceSize());
|
||||
DeviceMem out_device_buf(sizeof(OutDataType) * out.mDesc.GetElementSpaceSize());
|
||||
|
||||
out_device_buf.ToDevice(out.mData.data());
|
||||
wei_device_buf.ToDevice(wei.mData.data());
|
||||
|
||||
// reset input to zero
|
||||
in_device_buf.SetZero();
|
||||
|
||||
// do GEMM
|
||||
auto conv = DeviceConvNdBwdDataInstance{};
|
||||
auto invoker = conv.MakeInvoker();
|
||||
auto argument = conv.MakeArgument(static_cast<InDataType*>(in_device_buf.GetDeviceBuffer()),
|
||||
static_cast<WeiDataType*>(wei_device_buf.GetDeviceBuffer()),
|
||||
static_cast<OutDataType*>(out_device_buf.GetDeviceBuffer()),
|
||||
conv_param.N_,
|
||||
conv_param.K_,
|
||||
conv_param.C_,
|
||||
conv_param.input_spatial_lengths_,
|
||||
conv_param.filter_spatial_lengths_,
|
||||
conv_param.GetOutputSpatialLengths(),
|
||||
conv_param.conv_filter_strides_,
|
||||
conv_param.conv_filter_dilations_,
|
||||
conv_param.input_left_pads_,
|
||||
conv_param.input_right_pads_,
|
||||
in_element_op,
|
||||
wei_element_op,
|
||||
out_element_op);
|
||||
|
||||
if(!conv.IsSupportedArgument(argument))
|
||||
{
|
||||
throw std::runtime_error(
|
||||
"wrong! device_conv with the specified compilation parameters does "
|
||||
"not support this Conv problem");
|
||||
}
|
||||
|
||||
float ave_time = invoker.Run(argument, StreamConfig{nullptr, time_kernel});
|
||||
|
||||
std::size_t flop = conv_param.GetFlops();
|
||||
std::size_t num_btype = conv_param.GetByte<InDataType, WeiDataType, OutDataType>();
|
||||
|
||||
float tflops = static_cast<float>(flop) / 1.E9 / ave_time;
|
||||
|
||||
float gb_per_sec = num_btype / 1.E6 / ave_time;
|
||||
|
||||
std::cout << "Perf: " << ave_time << " ms, " << tflops << " TFlops, " << gb_per_sec << " GB/s"
|
||||
<< std::endl;
|
||||
|
||||
if(do_verification)
|
||||
{
|
||||
auto ref_conv = ck::tensor_operation::host::ReferenceConvBwdData<NDimSpatial,
|
||||
InDataType,
|
||||
WeiDataType,
|
||||
OutDataType,
|
||||
InElementOp,
|
||||
WeiElementOp,
|
||||
OutElementOp>();
|
||||
|
||||
auto ref_invoker = ref_conv.MakeInvoker();
|
||||
|
||||
auto ref_argument = ref_conv.MakeArgument(in_host,
|
||||
wei,
|
||||
out,
|
||||
conv_param.conv_filter_strides_,
|
||||
conv_param.conv_filter_dilations_,
|
||||
conv_param.input_left_pads_,
|
||||
conv_param.input_right_pads_,
|
||||
in_element_op,
|
||||
wei_element_op,
|
||||
out_element_op);
|
||||
|
||||
ref_invoker.Run(ref_argument);
|
||||
|
||||
in_device_buf.FromDevice(in_device.mData.data());
|
||||
|
||||
return ck::utils::check_err(in_device.mData, in_host.mData) ? 0 : 1;
|
||||
}
|
||||
|
||||
return 0;
|
||||
}
|
||||
207
example/17_convnd_bwd_data/convnd_bwd_data_xdl_fp16.cpp
Normal file
207
example/17_convnd_bwd_data/convnd_bwd_data_xdl_fp16.cpp
Normal file
@@ -0,0 +1,207 @@
|
||||
// SPDX-License-Identifier: MIT
|
||||
// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
|
||||
|
||||
#include "convnd_bwd_data_common.hpp"
|
||||
|
||||
#include "ck/tensor_operation/gpu/device/device_convnd_bwd_data_nwc_kxc_nwk_xdl.hpp"
|
||||
|
||||
using InDataType = ck::half_t;
|
||||
using WeiDataType = ck::half_t;
|
||||
using OutDataType = ck::half_t;
|
||||
using AccDataType = float;
|
||||
|
||||
template <ck::index_t... Is>
|
||||
using S = ck::Sequence<Is...>;
|
||||
|
||||
using InElementOp = ck::tensor_operation::element_wise::PassThrough;
|
||||
using WeiElementOp = ck::tensor_operation::element_wise::PassThrough;
|
||||
using OutElementOp = ck::tensor_operation::element_wise::PassThrough;
|
||||
|
||||
static constexpr auto ConvBwdDefault =
|
||||
ck::tensor_operation::device::ConvolutionBackwardDataSpecialization::Default;
|
||||
|
||||
template <ck::index_t NDimSpatial>
|
||||
using DeviceConvNdBwdDataInstance = ck::tensor_operation::device::DeviceConvNdBwdDataNwcKxcNwk_Xdl<
|
||||
NDimSpatial, // NDimSpatial
|
||||
InDataType, // InDataType
|
||||
WeiDataType, // WeiDataType
|
||||
OutDataType, // OutDataType
|
||||
AccDataType, // AccDataType
|
||||
InElementOp, // InElementwiseOperation
|
||||
WeiElementOp, // WeiElementwiseOperation
|
||||
OutElementOp, // OutElementwiseOperation
|
||||
ConvBwdDefault, // ConvolutionBackwardDataSpecialization
|
||||
256, // BlockSize
|
||||
128, // MPerBlock
|
||||
128, // NPerBlock
|
||||
4, // K0PerBlock
|
||||
8, // K1
|
||||
32, // MPerXdl
|
||||
32, // NPerXdl
|
||||
2, // MXdlPerWave
|
||||
2, // NXdlPerWave
|
||||
S<4, 64, 1>, // ABlockTransferThreadClusterLengths_K0_M_K1
|
||||
S<1, 0, 2>, // ABlockTransferThreadClusterArrangeOrder
|
||||
S<1, 0, 2>, // ABlockTransferSrcAccessOrder
|
||||
2, // ABlockTransferSrcVectorDim
|
||||
8, // ABlockTransferSrcScalarPerVector
|
||||
8, // ABlockTransferDstScalarPerVector_K1
|
||||
true, // ABlockLdsAddExtraM
|
||||
S<4, 64, 1>, // BBlockTransferThreadClusterLengths_K0_N_K1
|
||||
S<2, 0, 1>, // BBlockTransferThreadClusterArrangeOrder
|
||||
S<0, 2, 1>, // BBlockTransferSrcAccessOrder
|
||||
1, // BBlockTransferSrcVectorDim
|
||||
2, // BBlockTransferSrcScalarPerVector
|
||||
8, // BBlockTransferDstScalarPerVector_K1
|
||||
true, // BBlockLdsAddExtraN
|
||||
7,
|
||||
1>; // GemmCThreadTransferDstScalarPerVector
|
||||
|
||||
int main(int argc, char* argv[])
|
||||
{
|
||||
namespace ctc = ck::tensor_layout::convolution;
|
||||
|
||||
print_helper_msg();
|
||||
|
||||
bool do_verification = true;
|
||||
int init_method = 1;
|
||||
bool time_kernel = false;
|
||||
|
||||
ck::utils::conv::ConvParam conv_param{
|
||||
2, 1, 128, 256, 256, {3, 3}, {71, 71}, {2, 2}, {1, 1}, {1, 1}, {1, 1}};
|
||||
|
||||
if(argc == 1)
|
||||
{
|
||||
// use default
|
||||
}
|
||||
else if(argc == 4)
|
||||
{
|
||||
do_verification = std::stoi(argv[1]);
|
||||
init_method = std::stoi(argv[2]);
|
||||
time_kernel = std::stoi(argv[3]);
|
||||
}
|
||||
else
|
||||
{
|
||||
do_verification = std::stoi(argv[1]);
|
||||
init_method = std::stoi(argv[2]);
|
||||
time_kernel = std::stoi(argv[3]);
|
||||
const ck::index_t num_dim_spatial = std::stoi(argv[4]);
|
||||
|
||||
conv_param = ck::utils::conv::parse_conv_param(num_dim_spatial, 5, argv);
|
||||
}
|
||||
|
||||
const auto in_element_op = InElementOp{};
|
||||
const auto wei_element_op = WeiElementOp{};
|
||||
const auto out_element_op = OutElementOp{};
|
||||
|
||||
if(conv_param.num_dim_spatial_ == 1)
|
||||
{
|
||||
using InLayout = ctc::GNWC;
|
||||
using WeiLayout = ctc::GKXC;
|
||||
using OutLayout = ctc::GNWK;
|
||||
|
||||
const auto in_g_n_c_wis_desc =
|
||||
ck::utils::conv::make_input_host_tensor_descriptor_g_n_c_wis_packed<InLayout>(
|
||||
conv_param);
|
||||
|
||||
const auto wei_g_k_c_xs_desc =
|
||||
ck::utils::conv::make_weight_host_tensor_descriptor_g_k_c_xs_packed<WeiLayout>(
|
||||
conv_param);
|
||||
|
||||
const auto out_g_n_k_wos_desc =
|
||||
ck::utils::conv::make_output_host_tensor_descriptor_g_n_k_wos_packed<OutLayout>(
|
||||
conv_param);
|
||||
|
||||
return run_conv_bwd_data<1,
|
||||
InDataType,
|
||||
WeiDataType,
|
||||
OutDataType,
|
||||
InElementOp,
|
||||
WeiElementOp,
|
||||
OutElementOp,
|
||||
DeviceConvNdBwdDataInstance<1>>(do_verification,
|
||||
init_method,
|
||||
time_kernel,
|
||||
conv_param,
|
||||
in_g_n_c_wis_desc,
|
||||
wei_g_k_c_xs_desc,
|
||||
out_g_n_k_wos_desc,
|
||||
in_element_op,
|
||||
wei_element_op,
|
||||
out_element_op);
|
||||
}
|
||||
else if(conv_param.num_dim_spatial_ == 2)
|
||||
{
|
||||
using InLayout = ctc::GNHWC;
|
||||
using WeiLayout = ctc::GKYXC;
|
||||
using OutLayout = ctc::GNHWK;
|
||||
|
||||
const auto in_g_n_c_wis_desc =
|
||||
ck::utils::conv::make_input_host_tensor_descriptor_g_n_c_wis_packed<InLayout>(
|
||||
conv_param);
|
||||
|
||||
const auto wei_g_k_c_xs_desc =
|
||||
ck::utils::conv::make_weight_host_tensor_descriptor_g_k_c_xs_packed<WeiLayout>(
|
||||
conv_param);
|
||||
|
||||
const auto out_g_n_k_wos_desc =
|
||||
ck::utils::conv::make_output_host_tensor_descriptor_g_n_k_wos_packed<OutLayout>(
|
||||
conv_param);
|
||||
|
||||
return run_conv_bwd_data<2,
|
||||
InDataType,
|
||||
WeiDataType,
|
||||
OutDataType,
|
||||
InElementOp,
|
||||
WeiElementOp,
|
||||
OutElementOp,
|
||||
DeviceConvNdBwdDataInstance<2>>(do_verification,
|
||||
init_method,
|
||||
time_kernel,
|
||||
conv_param,
|
||||
in_g_n_c_wis_desc,
|
||||
wei_g_k_c_xs_desc,
|
||||
out_g_n_k_wos_desc,
|
||||
in_element_op,
|
||||
wei_element_op,
|
||||
out_element_op);
|
||||
}
|
||||
else if(conv_param.num_dim_spatial_ == 3)
|
||||
{
|
||||
using InLayout = ctc::GNDHWC;
|
||||
using WeiLayout = ctc::GKZYXC;
|
||||
using OutLayout = ctc::GNDHWK;
|
||||
|
||||
const auto in_g_n_c_wis_desc =
|
||||
ck::utils::conv::make_input_host_tensor_descriptor_g_n_c_wis_packed<InLayout>(
|
||||
conv_param);
|
||||
|
||||
const auto wei_g_k_c_xs_desc =
|
||||
ck::utils::conv::make_weight_host_tensor_descriptor_g_k_c_xs_packed<WeiLayout>(
|
||||
conv_param);
|
||||
|
||||
const auto out_g_n_k_wos_desc =
|
||||
ck::utils::conv::make_output_host_tensor_descriptor_g_n_k_wos_packed<OutLayout>(
|
||||
conv_param);
|
||||
|
||||
return run_conv_bwd_data<3,
|
||||
InDataType,
|
||||
WeiDataType,
|
||||
OutDataType,
|
||||
InElementOp,
|
||||
WeiElementOp,
|
||||
OutElementOp,
|
||||
DeviceConvNdBwdDataInstance<3>>(do_verification,
|
||||
init_method,
|
||||
time_kernel,
|
||||
conv_param,
|
||||
in_g_n_c_wis_desc,
|
||||
wei_g_k_c_xs_desc,
|
||||
out_g_n_k_wos_desc,
|
||||
in_element_op,
|
||||
wei_element_op,
|
||||
out_element_op);
|
||||
}
|
||||
|
||||
return 0;
|
||||
}
|
||||
Reference in New Issue
Block a user