refactoring block copy

[ROCm/composable_kernel commit: 03eef73c5b]
This commit is contained in:
Chao Liu
2019-03-17 15:36:38 -05:00
parent 6fd0910da8
commit da50d65ba0
5 changed files with 224 additions and 83 deletions

View File

@@ -7,11 +7,11 @@
#include "tensor.hpp"
#include "ConstantTensorDescriptor.hip.hpp"
#include "conv_common.hip.hpp"
#include "device_direct_convolution_1.hpp"
//#include "device_direct_convolution_1.hpp"
#include "device_direct_convolution_2_nchw_kcyx_nkhw.hpp"
#include "device_implicit_gemm_convolution_1_chwn_cyxk_khwn.hpp"
#include "device_implicit_gemm_convolution_1_chwn_cyxk_khwn_padded.hpp"
#include "device_implicit_gemm_convolution_2_chwn_cyxk_khwn.hpp"
//#include "device_implicit_gemm_convolution_1_chwn_cyxk_khwn.hpp"
//#include "device_implicit_gemm_convolution_1_chwn_cyxk_khwn_padded.hpp"
//#include "device_implicit_gemm_convolution_2_chwn_cyxk_khwn.hpp"
struct GeneratorTensor_1
{