ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-18 20:09:25 +00:00
Code Issues Packages Projects Releases Wiki Activity

refactored implicit gemm v1r3

Browse Source 
[ROCm/composable_kernel commit: 284e7bb317]
This commit is contained in:
Chao Liu
2019-07-29 15:25:38 -05:00
parent 360f15cf24
commit dc2cdacd15
4 changed files with 6 additions and 61 deletions
Download Patch File Download Diff File Expand all files Collapse all files

42
composable_kernel/include/kernel_algorithm/gridwise_convolution_implicit_gemm_v1r3_chwn_cyxk_khwn.hpp
View File

@@ -98,8 +98,7 @@ struct GridwiseConvolutionImplicitGemm_v1r3_chwn_cyxk_khwn
const index_t wi_block_data_begin = wo_block_data_begin;
// global tensor view
constexpr auto wei_c_k_global_desc =
make_ConstantTensorDescriptor(Sequence<C, K>{}, Sequence<Y * X * K, 1>{});
constexpr auto wei_c_k_global_desc = wei_c_y_x_k_global_desc.Extract(I0, I3);
// LDS tensor view
// be careful of alignment
@@ -212,44 +211,6 @@ struct GridwiseConvolutionImplicitGemm_v1r3_chwn_cyxk_khwn
// set threadwise output tensor to 0
threadwise_matrix_set_zero(c_k_wn_thread_mtx_desc, p_out_thread);
#if 1
const Float* p_in_global_block_offset =
p_in_global +
in_c_h_w_n_global_desc.GetOffsetFromMultiIndex(
0, hi_block_data_begin, wi_block_data_begin, n_block_data_begin);
const Float* p_wei_global_block_offset =
p_wei_global +
wei_c_y_x_k_global_desc.GetOffsetFromMultiIndex(0, 0, 0, k_block_data_begin);
for(index_t c_block_data_begin = 0; c_block_data_begin < C; c_block_data_begin += CPerBlock,
p_in_global_block_offset += CPerBlock * in_c_h_w_n_global_desc.GetStride(I0),
p_wei_global_block_offset += CPerBlock * wei_c_y_x_k_global_desc.GetStride(I0))
{
for(index_t y = 0; y < Y; ++y)
{
#pragma unroll
for(index_t x = 0; x < X; ++x)
{
blockwise_in_copy.Run(
p_in_global_block_offset +
in_c_h_w_n_global_desc.GetOffsetFromMultiIndex(0, y, x, 0),
p_in_block);
blockwise_wei_copy.Run(
p_wei_global_block_offset +
wei_c_y_x_k_global_desc.GetOffsetFromMultiIndex(0, y, x, 0),
p_wei_block);
__syncthreads();
blockwise_batch_gemm.Run(p_wei_block, p_in_block, p_out_thread);
__syncthreads();
}
}
}
#else
for(index_t y = 0; y < Y; ++y)
{
for(index_t x = 0; x < X; ++x)
@@ -282,7 +243,6 @@ struct GridwiseConvolutionImplicitGemm_v1r3_chwn_cyxk_khwn
}
}
}
#endif
// output: register to global mem,
const auto c_thread_mtx_begin =

14
composable_kernel/include/kernel_algorithm/gridwise_convolution_implicit_gemm_v1r3_chwn_cyxk_khwn_lds_double_buffer.hpp
View File

@@ -128,17 +128,8 @@ struct GridwiseConvolutionImplicitGemm_v1r3_chwn_cyxk_khwn_lds_double_buffer
constexpr auto out_k_h_w_n_thread_desc = make_ConstantTensorDescriptor_packed(
Sequence<KPerThread, HoPerThread, WoPerThread, NPerThread>{});
// blockwise copy
// input: format is [C, Hi, Wi, N]
#if 0
const auto blockwise_in_copy =
Blockwise4dTensorCopy1<BlockSize,
Float,
decltype(in_c_h_w_n_global_desc),
decltype(in_c_h_w_n_block_desc),
decltype(in_c_h_w_n_block_desc.GetLengths()),
InBlockCopyDataPerRead_N>{};
#else
// blockwise copy
// input: format is [C, Hi, Wi, N]
const auto blockwise_in_copy =
Blockwise4dTensorCopy3<BlockSize,
Float,
@@ -147,7 +138,6 @@ struct GridwiseConvolutionImplicitGemm_v1r3_chwn_cyxk_khwn_lds_double_buffer
decltype(in_c_h_w_n_block_desc.GetLengths()),
InBlockCopyClusterLengths_CHWN,
InBlockCopyDataPerRead_N>{};
#endif
// blockwise wei copy
// format is [CPerBlock, X * KPerBlock]