ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-11 17:00:18 +00:00
Code Issues Packages Projects Releases Wiki Activity

experimenting new merged tensor copy

Browse Source
This commit is contained in:
Chao Liu
2019-07-30 09:35:54 -05:00
parent 284e7bb317
commit cd8de11218
7 changed files with 137 additions and 25 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
driver/include/device_convolution_implicit_gemm_v1_chwn_cyxk_khwn.hpp
View File

@@ -478,9 +478,9 @@ void device_convolution_implicit_gemm_v1_chwn_cyxk_khwn(InDesc,
GridwiseConvolutionImplicitGemm_v1r1_chwn_cyxk_khwn
#elif 0
GridwiseConvolutionImplicitGemm_v1r2_chwn_cyxk_khwn
#elif 0
GridwiseConvolutionImplicitGemm_v1r3_chwn_cyxk_khwn
#elif 1
GridwiseConvolutionImplicitGemm_v1r3_chwn_cyxk_khwn
#elif 0
GridwiseConvolutionImplicitGemm_v1r3_chwn_cyxk_khwn_lds_double_buffer
#endif
<GridSize,

4
driver/src/driver.cpp
View File

@@ -85,7 +85,7 @@ int main(int argc, char* argv[])
constexpr index_t HPad = 0;
constexpr index_t WPad = 0;
#elif 1
#elif 0
// 3x3, 34x34
constexpr index_t N = 128;
constexpr index_t C = 256;
@@ -520,7 +520,7 @@ int main(int argc, char* argv[])
#if 0
device_convolution_direct_v2_nchw_kcyx_nkhw
(in_nchw_desc, in_nchw, wei_kcyx_desc, wei_kcyx, out_nkhw_desc, out_nkhw_device, nrepeat);
#elif 1
#elif 0
device_convolution_implicit_gemm_v1_chwn_cyxk_khwn(
in_nchw_desc, in_nchw, wei_kcyx_desc, wei_kcyx, out_nkhw_desc, out_nkhw_device, nrepeat);
#elif 0