Skip to content

GitLab

"script/hack_isa.sh" did not exist on "0979fb4af9d78b821334a8021e9b8bf0812d8ae6"
Commit adbda385 authored by Chao Liu's avatar Chao Liu
Browse files

clean up

parent 2c7ccf67
Hide whitespace changes
Inline Side-by-side
Showing with 79 additions and 72 deletions
device_operation/include/device_conv2d_fwd_xdl_bias_activation_add_nhwc_kyxc_nhwk.hpp
View file @ adbda385
......@@ -40,16 +40,16 @@ template <typename InDataType,
ck::index_t ABlockTransferSrcVectorDim,
ck::index_t ABlockTransferSrcScalarPerVector,
ck::index_t ABlockTransferDstScalarPerVector_K1,
bool ABlockLdsAddExtraM,
typename BBlockTransferThreadClusterLengths_K0_N_K1,
typename BBlockTransferThreadClusterArrangeOrder,
typename BBlockTransferSrcAccessOrder,
ck::index_t BBlockTransferSrcVectorDim,
ck::index_t BBlockTransferSrcScalarPerVector,
ck::index_t BBlockTransferDstScalarPerVector_K1,
bool BBlockLdsAddExtraN,
ck::index_t CThreadTransferSrcDstVectorDim,
ck::index_t CThreadTransferDstScalarPerVector,
bool ABlockLdsAddExtraM,
bool BBlockLdsAddExtraN>
ck::index_t CThreadTransferDstScalarPerVector>
struct DeviceConv2dFwdXdl_Bias_Activation_Add_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K
: public DeviceConvFwdBiasActivationAdd<InElementwiseOperation,
WeiElementwiseOperation,
......@@ -256,19 +256,18 @@ struct DeviceConv2dFwdXdl_Bias_Activation_Add_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Out
ABlockTransferSrcScalarPerVector,
ABlockTransferDstScalarPerVector_K1,
false, // AThreadTransferSrcResetCoordinateAfterRun,
ABlockLdsAddExtraM,
BBlockTransferThreadClusterLengths_K0_N_K1,
Sequence<1, 0, 2>, // BBlockTransferThreadClusterArrangeOrder,
Sequence<1, 0, 2>, // BBlockTransferSrcAccessOrder,
2, // BBlockTransferSrcVectorDim,
BBlockTransferSrcScalarPerVector,
BBlockTransferDstScalarPerVector_K1,
false, // BThreadTransferSrcResetCoordinateAfterRun,
false, // BThreadTransferSrcResetCoordinateAfterRun,
BBlockLdsAddExtraN,
Sequence<2, 3, 0, 1, 7, 5, 4, 6>, // CThreadTransferSrcDstAccessOrder,
7, // CThreadTransferSrcDstVectorDim,
CThreadTransferDstScalarPerVector,
false, // CAccessOrderMRepeatNRepeat,
ABlockLdsAddExtraM,
BBlockLdsAddExtraN>;
CThreadTransferDstScalarPerVector>;
using CGridDesc_M0_N0_M1_N1_M2_M3_M4_N2 =
decltype(GridwiseGemm::MakeCGridDescriptor_M0_N0_M1_N1_M2_M3_M4_N2(CGridDesc_M_N{}));
......
device_operation/include/device_conv2d_fwd_xdl_bias_activation_nhwc_kyxc_nhwk.hpp
View file @ adbda385
......@@ -40,16 +40,16 @@ template <typename InDataType,
ck::index_t ABlockTransferSrcVectorDim,
ck::index_t ABlockTransferSrcScalarPerVector,
ck::index_t ABlockTransferDstScalarPerVector_K1,
bool ABlockLdsAddExtraM,
typename BBlockTransferThreadClusterLengths_K0_N_K1,
typename BBlockTransferThreadClusterArrangeOrder,
typename BBlockTransferSrcAccessOrder,
ck::index_t BBlockTransferSrcVectorDim,
ck::index_t BBlockTransferSrcScalarPerVector,
ck::index_t BBlockTransferDstScalarPerVector_K1,
bool BBlockLdsAddExtraN,
ck::index_t CThreadTransferSrcDstVectorDim,
ck::index_t CThreadTransferDstScalarPerVector,
bool ABlockLdsAddExtraM,
bool BBlockLdsAddExtraN>
ck::index_t CThreadTransferDstScalarPerVector>
struct DeviceConv2dFwdXdl_Bias_Activation_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K
: public DeviceConvFwdBiasActivation<InElementwiseOperation,
WeiElementwiseOperation,
......@@ -250,19 +250,18 @@ struct DeviceConv2dFwdXdl_Bias_Activation_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_
ABlockTransferSrcScalarPerVector,
ABlockTransferDstScalarPerVector_K1,
false, // AThreadTransferSrcResetCoordinateAfterRun,
ABlockLdsAddExtraM,
BBlockTransferThreadClusterLengths_K0_N_K1,
Sequence<1, 0, 2>, // BBlockTransferThreadClusterArrangeOrder,
Sequence<1, 0, 2>, // BBlockTransferSrcAccessOrder,
2, // BBlockTransferSrcVectorDim,
BBlockTransferSrcScalarPerVector,
BBlockTransferDstScalarPerVector_K1,
false, // BThreadTransferSrcResetCoordinateAfterRun,
false, // BThreadTransferSrcResetCoordinateAfterRun,
BBlockLdsAddExtraN,
Sequence<2, 3, 0, 1, 7, 5, 4, 6>, // CThreadTransferSrcDstAccessOrder,
7, // CThreadTransferSrcDstVectorDim,
CThreadTransferDstScalarPerVector,
false, // CAccessOrderMRepeatNRepeat,
ABlockLdsAddExtraM,
BBlockLdsAddExtraN>;
CThreadTransferDstScalarPerVector>;
using CGridDesc_M0_N0_M1_N1_M2_M3_M4_N2 =
decltype(GridwiseGemm::MakeCGridDescriptor_M0_N0_M1_N1_M2_M3_M4_N2(CGridDesc_M_N{}));
......
device_operation/include/device_conv2d_fwd_xdl_output_shuffle_bias_activation_add_nhwc_kyxc_nhwk.hpp → device_operation/include/device_conv2d_fwd_xdl_c_shuffle_bias_activation_add_nhwc_kyxc_nhwk.hpp
View file @ adbda385
#ifndef DEVICE_CONV2D_FWD_XDL_OUTPUT_SHUFFLE_BIAS_ACTIVATION_ADD_NHWC_KYXC_NHWK_HPP
#define DEVICE_CONV2D_FWD_XDL_OUTPUT_SHUFFLE_BIAS_ACTIVATION_ADD_NHWC_KYXC_NHWK_HPP
#ifndef DEVICE_CONV2D_FWD_XDL_C_SHUFFLE_BIAS_ACTIVATION_ADD_NHWC_KYXC_NHWK_HPP
#define DEVICE_CONV2D_FWD_XDL_C_SHUFFLE_BIAS_ACTIVATION_ADD_NHWC_KYXC_NHWK_HPP
#include <iostream>
#include <sstream>
......@@ -54,13 +54,13 @@ template <
typename CBlockTransferClusterLengths_MBlock_MRepeat_MWaveMPerXdl_NBlock_NRepeat_NWaveNPerXdl,
index_t CBlockTransferScalarPerVector_NWaveNPerXdl>
struct
DeviceConv2dFwdXdl_Output_Shuffle_Bias_Activation_Add_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K
DeviceConv2dFwdXdl_C_Shuffle_Bias_Activation_Add_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K
: public DeviceConvFwdBiasActivationAdd<InElementwiseOperation,
WeiElementwiseOperation,
OutElementwiseOperation>
{
using DeviceOp =
DeviceConv2dFwdXdl_Output_Shuffle_Bias_Activation_Add_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K;
DeviceConv2dFwdXdl_C_Shuffle_Bias_Activation_Add_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K;
using ADataType = InDataType;
using BDataType = WeiDataType;
......@@ -642,7 +642,7 @@ struct
auto str = std::stringstream();
// clang-format off
str << "DeviceConv2dFwdXdl_Output_Shuffle_Bias_Activation_Add_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K"
str << "DeviceConv2dFwdXdl_C_Shuffle_Bias_Activation_Add_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K"
<< "<"
<< BlockSize << ", "
<< MPerBlock << ", "
......
device_operation/include/device_conv2d_fwd_xdl_output_shuffle_nhwc_kyxc_nhwk.hpp → device_operation/include/device_conv2d_fwd_xdl_c_shuffle_nhwc_kyxc_nhwk.hpp
View file @ adbda385
#ifndef DEVICE_CONV2D_FWD_XDL_OUTPUT_SHUFFLE_NHWC_KYXC_NHWK_HPP
#define DEVICE_CONV2D_FWD_XDL_OUTPUT_SHUFFLE_NHWC_KYXC_NHWK_HPP
#ifndef DEVICE_CONV2D_FWD_XDL_C_SHUFFLE_NHWC_KYXC_NHWK_HPP
#define DEVICE_CONV2D_FWD_XDL_C_SHUFFLE_NHWC_KYXC_NHWK_HPP
#include <iostream>
#include <sstream>
......@@ -52,11 +52,10 @@ template <
index_t CShuffleNRepeatPerShuffle,
typename CBlockTransferClusterLengths_MBlock_MRepeat_MWaveMPerXdl_NBlock_NRepeat_NWaveNPerXdl,
index_t CBlockTransferScalarPerVector_NWaveNPerXdl>
struct DeviceConv2dFwdXdl_Output_Shuffle_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K
struct DeviceConv2dFwdXdl_C_Shuffle_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K
: public DeviceConvFwd<InElementwiseOperation, WeiElementwiseOperation, OutElementwiseOperation>
{
using DeviceOp =
DeviceConv2dFwdXdl_Output_Shuffle_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K;
using DeviceOp = DeviceConv2dFwdXdl_C_Shuffle_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K;
using ADataType = InDataType;
using BDataType = WeiDataType;
......@@ -580,7 +579,7 @@ struct DeviceConv2dFwdXdl_Output_Shuffle_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N
auto str = std::stringstream();
// clang-format off
str << "DeviceConv2dFwdXdl_Output_Shuffle_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K"
str << "DeviceConv2dFwdXdl_C_Shuffle_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K"
<< "<"
<< BlockSize << ", "
<< MPerBlock << ", "
......
device_operation/include/device_conv2d_fwd_xdl_nhwc_kyxc_nhwk.hpp
View file @ adbda385
......@@ -39,16 +39,16 @@ template <typename InDataType,
ck::index_t ABlockTransferSrcVectorDim,
ck::index_t ABlockTransferSrcScalarPerVector,
ck::index_t ABlockTransferDstScalarPerVector_K1,
bool ABlockLdsAddExtraM,
typename BBlockTransferThreadClusterLengths_K0_N_K1,
typename BBlockTransferThreadClusterArrangeOrder,
typename BBlockTransferSrcAccessOrder,
ck::index_t BBlockTransferSrcVectorDim,
ck::index_t BBlockTransferSrcScalarPerVector,
ck::index_t BBlockTransferDstScalarPerVector_K1,
bool BBlockLdsAddExtraN,
ck::index_t CThreadTransferSrcDstVectorDim,
ck::index_t CThreadTransferDstScalarPerVector,
bool ABlockLdsAddExtraM,
bool BBlockLdsAddExtraN>
ck::index_t CThreadTransferDstScalarPerVector>
struct DeviceConv2dFwdXdl_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K
: public DeviceConvFwd<InElementwiseOperation, WeiElementwiseOperation, OutElementwiseOperation>
{
......@@ -235,19 +235,18 @@ struct DeviceConv2dFwdXdl_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K
ABlockTransferSrcScalarPerVector,
ABlockTransferDstScalarPerVector_K1,
false, // AThreadTransferSrcResetCoordinateAfterRun,
ABlockLdsAddExtraM,
BBlockTransferThreadClusterLengths_K0_N_K1,
Sequence<1, 0, 2>, // BBlockTransferThreadClusterArrangeOrder,
Sequence<1, 0, 2>, // BBlockTransferSrcAccessOrder,
2, // BBlockTransferSrcVectorDim,
BBlockTransferSrcScalarPerVector,
BBlockTransferDstScalarPerVector_K1,
false, // BThreadTransferSrcResetCoordinateAfterRun,
false, // BThreadTransferSrcResetCoordinateAfterRun,
BBlockLdsAddExtraN,
Sequence<2, 3, 0, 1, 7, 5, 4, 6>, // CThreadTransferSrcDstAccessOrder,
7, // CThreadTransferSrcDstVectorDim,
CThreadTransferDstScalarPerVector,
false, // CAccessOrderMRepeatNRepeat,
ABlockLdsAddExtraM,
BBlockLdsAddExtraN>;
CThreadTransferDstScalarPerVector>;
// Argument
struct Argument : public BaseArgument
......
device_operation/include/device_conv2d_fwd_xdl_nhwc_kyxc_nhwk_1x1_p0.hpp
View file @ adbda385
......@@ -39,16 +39,16 @@ template <typename InDataType,
ck::index_t ABlockTransferSrcVectorDim,
ck::index_t ABlockTransferSrcScalarPerVector,
ck::index_t ABlockTransferDstScalarPerVector_K1,
bool ABlockLdsAddExtraM,
typename BBlockTransferThreadClusterLengths_K0_N_K1,
typename BBlockTransferThreadClusterArrangeOrder,
typename BBlockTransferSrcAccessOrder,
ck::index_t BBlockTransferSrcVectorDim,
ck::index_t BBlockTransferSrcScalarPerVector,
ck::index_t BBlockTransferDstScalarPerVector_K1,
bool BBlockLdsAddExtraN,
ck::index_t CThreadTransferSrcDstVectorDim,
ck::index_t CThreadTransferDstScalarPerVector,
bool ABlockLdsAddExtraM,
bool BBlockLdsAddExtraN>
ck::index_t CThreadTransferDstScalarPerVector>
struct DeviceConv2dFwdXdl_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K_1x1_P0
: public DeviceConvFwd<InElementwiseOperation, WeiElementwiseOperation, OutElementwiseOperation>
{
......@@ -194,19 +194,18 @@ struct DeviceConv2dFwdXdl_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K_1x1_P0
ABlockTransferSrcScalarPerVector,
ABlockTransferDstScalarPerVector_K1,
false, // AThreadTransferSrcResetCoordinateAfterRun,
ABlockLdsAddExtraM,
BBlockTransferThreadClusterLengths_K0_N_K1,
Sequence<1, 0, 2>, // BBlockTransferThreadClusterArrangeOrder,
Sequence<1, 0, 2>, // BBlockTransferSrcAccessOrder,
2, // BBlockTransferSrcVectorDim,
BBlockTransferSrcScalarPerVector,
BBlockTransferDstScalarPerVector_K1,
false, // BThreadTransferSrcResetCoordinateAfterRun,
false, // BThreadTransferSrcResetCoordinateAfterRun,
BBlockLdsAddExtraN,
Sequence<2, 3, 0, 1, 7, 5, 4, 6>, // CThreadTransferSrcDstAccessOrder,
7, // CThreadTransferSrcDstVectorDim,
CThreadTransferDstScalarPerVector,
false, // CAccessOrderMRepeatNRepeat,
ABlockLdsAddExtraM,
BBlockLdsAddExtraN>;
CThreadTransferDstScalarPerVector>;
// Argument
struct Argument : public BaseArgument
......
device_operation/include/device_conv2d_fwd_xdl_nhwc_kyxc_nhwk_1x1_s1_p0.hpp
View file @ adbda385
......@@ -39,16 +39,16 @@ template <typename InDataType,
ck::index_t ABlockTransferSrcVectorDim,
ck::index_t ABlockTransferSrcScalarPerVector,
ck::index_t ABlockTransferDstScalarPerVector_K1,
bool ABlockLdsAddExtraM,
typename BBlockTransferThreadClusterLengths_K0_N_K1,
typename BBlockTransferThreadClusterArrangeOrder,
typename BBlockTransferSrcAccessOrder,
ck::index_t BBlockTransferSrcVectorDim,
ck::index_t BBlockTransferSrcScalarPerVector,
ck::index_t BBlockTransferDstScalarPerVector_K1,
bool BBlockLdsAddExtraN,
ck::index_t CThreadTransferSrcDstVectorDim,
ck::index_t CThreadTransferDstScalarPerVector,
bool ABlockLdsAddExtraM,
bool BBlockLdsAddExtraN>
ck::index_t CThreadTransferDstScalarPerVector>
struct DeviceConv2dFwdXdl_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K_1x1_S1_P0
: public DeviceConvFwd<InElementwiseOperation, WeiElementwiseOperation, OutElementwiseOperation>
{
......@@ -171,19 +171,18 @@ struct DeviceConv2dFwdXdl_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K_1x1_S1
ABlockTransferSrcScalarPerVector,
ABlockTransferDstScalarPerVector_K1,
false, // AThreadTransferSrcResetCoordinateAfterRun,
ABlockLdsAddExtraM,
BBlockTransferThreadClusterLengths_K0_N_K1,
Sequence<1, 0, 2>, // BBlockTransferThreadClusterArrangeOrder,
Sequence<1, 0, 2>, // BBlockTransferSrcAccessOrder,
2, // BBlockTransferSrcVectorDim,
BBlockTransferSrcScalarPerVector,
BBlockTransferDstScalarPerVector_K1,
false, // BThreadTransferSrcResetCoordinateAfterRun,
false, // BThreadTransferSrcResetCoordinateAfterRun,
BBlockLdsAddExtraN,
Sequence<2, 3, 0, 1, 7, 5, 4, 6>, // CThreadTransferSrcDstAccessOrder,
7, // CThreadTransferSrcDstVectorDim,
CThreadTransferDstScalarPerVector,
false, // CAccessOrderMRepeatNRepeat,
ABlockLdsAddExtraM,
BBlockLdsAddExtraN>;
CThreadTransferDstScalarPerVector>;
// Argument
struct Argument : public BaseArgument
......
device_operation/include/device_gemm_xdl.hpp
View file @ adbda385
......@@ -41,16 +41,16 @@ template <typename ADataType,
ck::index_t ABlockTransferSrcVectorDim,
ck::index_t ABlockTransferSrcScalarPerVector,
ck::index_t ABlockTransferDstScalarPerVector_K1,
bool ABlockLdsAddExtraM,
typename BBlockTransferThreadClusterLengths_K0_N_K1,
typename BBlockTransferThreadClusterArrangeOrder,
typename BBlockTransferSrcAccessOrder,
ck::index_t BBlockTransferSrcVectorDim,
ck::index_t BBlockTransferSrcScalarPerVector,
ck::index_t BBlockTransferDstScalarPerVector_K1,
bool BBlockLdsAddExtraN,
ck::index_t CThreadTransferSrcDstVectorDim,
ck::index_t CThreadTransferDstScalarPerVector,
bool ABlockLdsAddExtraM,
bool BBlockLdsAddExtraN>
ck::index_t CThreadTransferDstScalarPerVector>
struct DeviceGemmXdl
: public DeviceGemm<AElementwiseOperation, BElementwiseOperation, CElementwiseOperation>
{
......@@ -158,19 +158,18 @@ struct DeviceGemmXdl
ABlockTransferSrcScalarPerVector,
ABlockTransferDstScalarPerVector_K1,
false, // AThreadTransferSrcResetCoordinateAfterRun,
ABlockLdsAddExtraM,
BBlockTransferThreadClusterLengths_K0_N_K1,
BBlockTransferThreadClusterArrangeOrder,
BBlockTransferSrcAccessOrder,
BBlockTransferSrcVectorDim,
BBlockTransferSrcScalarPerVector,
BBlockTransferDstScalarPerVector_K1,
false, // BThreadTransferSrcResetCoordinateAfterRun,
false, // BThreadTransferSrcResetCoordinateAfterRun,
BBlockLdsAddExtraN,
Sequence<0, 2, 4, 5, 6, 1, 3, 7>, // CThreadTransferSrcDstAccessOrder,
CThreadTransferSrcDstVectorDim,
CThreadTransferDstScalarPerVector,
false, // CAccessOrderMRepeatNRepeat,
ABlockLdsAddExtraM,
BBlockLdsAddExtraN>;
CThreadTransferDstScalarPerVector>;
// Argument
struct Argument : public BaseArgument
......
example/4_conv2d_fwd_xdl_output_shuffle/conv2d_fwd_xdl_output_shuffle.cpp → example/4_conv2d_fwd_xdl_c_shuffle/conv2d_fwd_xdl_c_shuffle.cpp
View file @ adbda385
......@@ -11,7 +11,7 @@
#include "host_tensor_generator.hpp"
#include "device_tensor.hpp"
#include "tensor_layout.hpp"
#include "device_operation/include/device_conv2d_fwd_xdl_output_shuffle_nhwc_kyxc_nhwk.hpp"
#include "device_operation/include/device_conv2d_fwd_xdl_c_shuffle_nhwc_kyxc_nhwk.hpp"
#include "element_wise_operation.hpp"
using InDataType = ck::half_t;
......@@ -28,10 +28,10 @@ using OutLayout = ck::tensor_layout::convolution::NHWK;
using InElementOp = ck::tensor_operation::element_wise::PassThrough;
using WeiElementOp = ck::tensor_operation::element_wise::PassThrough;
using OutElementOp = ck::tensor_operation::element_wise::PassThrough_v2;
using OutElementOp = ck::tensor_operation::element_wise::PassThrough;
using DeviceConvFwdInstance = ck::tensor_operation::device::
DeviceConv2dFwdXdl_Output_Shuffle_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K
DeviceConv2dFwdXdl_C_Shuffle_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K
// clang-format off
// | InData| WeiData| OutData| AccData| In| Wei| Out| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CShuffle| CShuffle| CBlockTransferClusterLengths| CBlockTransfer|
// | Type| Type| Type| Type| Elementwise| Elementwise| Elementwise| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| MRepeate| NRepeate| _MBlock_MRepeat_MWaveMPerXdl| ScalarPerVector|
......
example/6_conv2d_fwd_xdl_output_shuffle_bias_relu_add/conv2d_fwd_xdl_output_shuffle_bias_relu_add.cpp → example/6_conv2d_fwd_xdl_c_shuffle_bias_relu_add/conv2d_fwd_xdl_c_shuffle_bias_relu_add.cpp
View file @ adbda385
......@@ -11,7 +11,7 @@
#include "host_tensor_generator.hpp"
#include "device_tensor.hpp"
#include "tensor_layout.hpp"
#include "device_conv2d_fwd_xdl_output_shuffle_bias_activation_add_nhwc_kyxc_nhwk.hpp"
#include "device_conv2d_fwd_xdl_c_shuffle_bias_activation_add_nhwc_kyxc_nhwk.hpp"
#include "element_wise_operation.hpp"
using InDataType = ck::half_t;
......@@ -32,7 +32,7 @@ using OutElementOp = ck::tensor_operation::element_wise::AddReluAdd_v2;
// clang-format off
using DeviceConvFwdInstance = ck::tensor_operation::device::
DeviceConv2dFwdXdl_Output_Shuffle_Bias_Activation_Add_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K
DeviceConv2dFwdXdl_C_Shuffle_Bias_Activation_Add_Input_N_Hi_Wi_C_Weight_K_Y_X_C_Output_N_Ho_Wo_K
// | InData| WeiData| OutData| AccData| In| Wei| Out| Block| MPer| NPer| K0Per| K1| MPer| NPer| MXdl| NXdl| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds| BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CShuffle| CShuffle| CBlockTransferClusterLengths| CBlockTransfer|
// | Type| Type| Type| Type| Elementwise| Elementwise| Elementwise| Size| Block| Block| Block| | XDL| XDL| Per| Per| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraM| ThreadCluster| ThreadCluster| SrcAccessOrder| SrcVectorDim| SrcScalar| DstScalar| AddExtraN| MRepeate| NRepeate| _MBlock_MRepeat_MWaveMPerXdl| ScalarPerVector|
// | | | | | Operation| Operation| Operation| | | | | | | | Wave| Wave| Lengths_K0_M_K1| ArrangeOrder| | | PerVector| PerVector_K1| | Lengths_K0_N_K1| ArrangeOrder| | | PerVector| PerVector_K1| | PerShuffle| PerShuffle| _NBlock_NRepeat_NWaveNPerXdl| _NWaveNPerXdl|
......
example/CMakeLists.txt
View file @ adbda385
......@@ -14,23 +14,23 @@ include_directories(BEFORE
set(GEMM_XDL_SOURCE 1_gemm_xdl/gemm_xdl.cpp)
set(GEMM_XDL_BIAS_RELU_ADD_SOURCE 3_gemm_xdl_bias_relu_add/gemm_xdl_bias_relu_add.cpp)
set(CONV2D_FWD_XDL_SOURCE 4_conv2d_fwd_xdl/conv2d_fwd_xdl.cpp)
set(CONV2D_FWD_XDL_OUTPUT_SHUFFLE_SOURCE 4_conv2d_fwd_xdl_output_shuffle/conv2d_fwd_xdl_output_shuffle.cpp)
set(CONV2D_FWD_XDL_C_SHUFFLE_SOURCE 4_conv2d_fwd_xdl_c_shuffle/conv2d_fwd_xdl_c_shuffle.cpp)
set(CONV2D_FWD_XDL_BIAS_RELU_SOURCE 5_conv2d_fwd_xdl_bias_relu/conv2d_fwd_xdl_bias_relu.cpp)
set(CONV2D_FWD_XDL_BIAS_RELU_ADD_SOURCE 6_conv2d_fwd_xdl_bias_relu_add/conv2d_fwd_xdl_bias_relu_add.cpp)
set(CONV2D_FWD_XDL_OUTPUT_SHUFFLE_BIAS_RELU_ADD_SOURCE 6_conv2d_fwd_xdl_output_shuffle_bias_relu_add/conv2d_fwd_xdl_output_shuffle_bias_relu_add.cpp)
set(CONV2D_FWD_XDL_C_SHUFFLE_BIAS_RELU_ADD_SOURCE 6_conv2d_fwd_xdl_c_shuffle_bias_relu_add/conv2d_fwd_xdl_c_shuffle_bias_relu_add.cpp)
add_executable(gemm_xdl ${GEMM_XDL_SOURCE})
add_executable(gemm_xdl_bias_relu_add ${GEMM_XDL_BIAS_RELU_ADD_SOURCE})
add_executable(conv2d_fwd_xdl ${CONV2D_FWD_XDL_SOURCE})
add_executable(conv2d_fwd_xdl_output_shuffle ${CONV2D_FWD_XDL_OUTPUT_SHUFFLE_SOURCE})
add_executable(conv2d_fwd_xdl_c_shuffle ${CONV2D_FWD_XDL_C_SHUFFLE_SOURCE})
add_executable(conv2d_fwd_xdl_bias_relu ${CONV2D_FWD_XDL_BIAS_RELU_SOURCE})
add_executable(conv2d_fwd_xdl_bias_relu_add ${CONV2D_FWD_XDL_BIAS_RELU_ADD_SOURCE})
add_executable(conv2d_fwd_xdl_output_shuffle_bias_relu_add ${CONV2D_FWD_XDL_OUTPUT_SHUFFLE_BIAS_RELU_ADD_SOURCE})
add_executable(conv2d_fwd_xdl_c_shuffle_bias_relu_add ${CONV2D_FWD_XDL_C_SHUFFLE_BIAS_RELU_ADD_SOURCE})
target_link_libraries(gemm_xdl PRIVATE host_tensor)
target_link_libraries(gemm_xdl_bias_relu_add PRIVATE host_tensor)
target_link_libraries(conv2d_fwd_xdl PRIVATE host_tensor)
target_link_libraries(conv2d_fwd_xdl_output_shuffle PRIVATE host_tensor)
target_link_libraries(conv2d_fwd_xdl_c_shuffle PRIVATE host_tensor)
target_link_libraries(conv2d_fwd_xdl_bias_relu PRIVATE host_tensor)
target_link_libraries(conv2d_fwd_xdl_bias_relu_add PRIVATE host_tensor)
target_link_libraries(conv2d_fwd_xdl_output_shuffle_bias_relu_add PRIVATE host_tensor)
target_link_libraries(conv2d_fwd_xdl_c_shuffle_bias_relu_add PRIVATE host_tensor)
profiler/CMakeLists.txt
View file @ adbda385
......@@ -36,6 +36,7 @@ set(DEVICE_CONV2D_FWD_INSTANCE_SOURCE
${PROJECT_SOURCE_DIR}/device_operation/device_conv2d_fwd_xdl_nhwc_kyxc_nhwk_f16_instance.cpp;
${PROJECT_SOURCE_DIR}/device_operation/device_conv2d_fwd_xdl_nhwc_kyxc_nhwk_1x1_p0_f16_instance.cpp;
${PROJECT_SOURCE_DIR}/device_operation/device_conv2d_fwd_xdl_nhwc_kyxc_nhwk_1x1_s1_p0_f16_instance.cpp;
${PROJECT_SOURCE_DIR}/device_operation/device_conv2d_fwd_xdl_c_shuffle_nhwc_kyxc_nhwk_f16_instance.cpp;
)
add_library(device_conv2d_fwd_instance SHARED ${DEVICE_CONV2D_FWD_INSTANCE_SOURCE})
......@@ -58,6 +59,7 @@ install(TARGETS device_conv2d_fwd_bias_relu_instance LIBRARY DESTINATION lib)
# device_conv2d_fwd_bias_relu_add_instance
set(DEVICE_CONV2D_FWD_BIAS_RELU_ADD_INSTANCE_SOURCE
${PROJECT_SOURCE_DIR}/device_operation/device_conv2d_fwd_xdl_bias_relu_add_nhwc_kyxc_nhwk_f16_instance.cpp;
${PROJECT_SOURCE_DIR}/device_operation/device_conv2d_fwd_xdl_c_shuffle_bias_relu_add_nhwc_kyxc_nhwk_f16_instance.cpp;
)
add_library(device_conv2d_fwd_bias_relu_add_instance SHARED ${DEVICE_CONV2D_FWD_BIAS_RELU_ADD_INSTANCE_SOURCE})
......
profiler/include/profile_conv_fwd_bias_relu_add_impl.hpp
View file @ adbda385
......@@ -19,7 +19,10 @@ using DeviceConvFwdBiasReluAddPtr =
ck::tensor_operation::element_wise::PassThrough,
ck::tensor_operation::element_wise::AddReluAdd>;
void add_device_conv2d_fwd_bias_relu_add_xdl_nhwc_kyxc_nhwk_fp16_instances(
void add_device_conv2d_fwd_bias_relu_xdl_nhwc_kyxc_nhwk_fp16_instances(
std::vector<DeviceConvFwdBiasReluAddPtr>&);
void add_device_conv2d_fwd_xdl_c_shuffle_bias_relu_add_nhwc_kyxc_nhwk_f16_instances(
std::vector<DeviceConvFwdBiasReluAddPtr>&);
} // namespace device_conv2d_fwd_bias_activation_add_instance
......@@ -208,7 +211,10 @@ void profile_conv_fwd_bias_relu_add_impl(int do_verification,
ck::is_same_v<ck::remove_cv_t<OutDataType>, ck::half_t>)
{
ck::tensor_operation::device::device_conv2d_fwd_bias_activation_add_instance::
add_device_conv2d_fwd_bias_relu_add_xdl_nhwc_kyxc_nhwk_fp16_instances(op_ptrs);
add_device_conv2d_fwd_bias_relu_xdl_nhwc_kyxc_nhwk_fp16_instances(op_ptrs);
ck::tensor_operation::device::device_conv2d_fwd_bias_activation_add_instance::
add_device_conv2d_fwd_xdl_c_shuffle_bias_relu_add_nhwc_kyxc_nhwk_f16_instances(op_ptrs);
}
if(op_ptrs.size() <= 0)
......
profiler/include/profile_conv_fwd_impl.hpp
View file @ adbda385
......@@ -28,6 +28,9 @@ void add_device_conv2d_fwd_xdl_nhwc_kyxc_nhwk_1x1_p0_fp16_instances(
void add_device_conv2d_fwd_xdl_nhwc_kyxc_nhwk_1x1_s1_p0_fp16_instances(
std::vector<DeviceConvFwdNoOpPtr>&);
void add_device_conv2d_fwd_xdl_c_shuffle_nhwc_kyxc_nhwk_fp16_instances(
std::vector<DeviceConvFwdNoOpPtr>&);
} // namespace device_conv2d_fwd_instance
} // namespace device
} // namespace tensor_operation
......@@ -154,6 +157,9 @@ void profile_conv_fwd_impl(int do_verification,
ck::tensor_operation::device::device_conv2d_fwd_instance::
add_device_conv2d_fwd_xdl_nhwc_kyxc_nhwk_1x1_s1_p0_fp16_instances(conv_ptrs);
ck::tensor_operation::device::device_conv2d_fwd_instance::
add_device_conv2d_fwd_xdl_c_shuffle_nhwc_kyxc_nhwk_fp16_instances(conv_ptrs);
}
if(conv_ptrs.size() <= 0)
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment