Skip to content

GitLab

Commit 5ce19234 authored by Chao Liu's avatar Chao Liu
Browse files

added GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn

parent 19f17df4
Hide whitespace changes
Inline Side-by-side
Showing with 134 additions and 405 deletions
driver/device_convolution_implicit_gemm_v1_chwn_cyxk_khwn.hpp
View file @ 5ce19234
...@@ -111,7 +111,7 @@ void device_convolution_implicit_gemm_v1_chwn_cyxk_khwn(InDesc, ...@@ -111,7 +111,7 @@ void device_convolution_implicit_gemm_v1_chwn_cyxk_khwn(InDesc,
constexpr index_t OutThreadCopyDataPerWrite = 2; constexpr index_t OutThreadCopyDataPerWrite = 2;
constexpr index_t BlockSize = 128; constexpr index_t BlockSize = 128;
#elif 0 #elif 1
// for 3x3, 34x34, v1r2, Pascal, in-block-copy1 // for 3x3, 34x34, v1r2, Pascal, in-block-copy1
constexpr index_t NPerBlock = 4; constexpr index_t NPerBlock = 4;
constexpr index_t KPerBlock = 64; constexpr index_t KPerBlock = 64;
......
driver/device_convolution_implicit_gemm_v1_nchw_cyxk_khwn.hpp
View file @ 5ce19234
...@@ -62,208 +62,10 @@ void device_convolution_implicit_gemm_v1_nchw_cyxk_khwn(InDesc, ...@@ -62,208 +62,10 @@ void device_convolution_implicit_gemm_v1_nchw_cyxk_khwn(InDesc,
wei_cyxk_device_buf.ToDevice(wei_cyxk.mData.data()); wei_cyxk_device_buf.ToDevice(wei_cyxk.mData.data());
out_khwn_device_buf.ToDevice(out_khwn.mData.data()); out_khwn_device_buf.ToDevice(out_khwn.mData.data());
#if 0 #if 1
// for 3x3, 34x34, v1r1, Pascal // for 3x3, 28x28, v1r2, Pascal
constexpr index_t NPerBlock = 16;
constexpr index_t KPerBlock = 64;
constexpr index_t CPerBlock = 4;
constexpr index_t HoPerBlock = 2;
constexpr index_t WoPerBlock = 4;
constexpr index_t NPerThread = 4;
constexpr index_t KPerThread = 8;
constexpr index_t HoPerThread = 1;
constexpr index_t WoPerThread = 2;
constexpr index_t InBlockCopy_ThreadPerDimC = 4;
constexpr index_t InBlockCopy_ThreadPerDimH = 4;
constexpr index_t InBlockCopy_ThreadPerDimW = 2;
constexpr index_t InBlockCopy_ThreadPerDimN = 4;
constexpr index_t InBlockCopyDataPerRead = 4;
constexpr index_t WeiBlockCopyDataPerRead = 4;
constexpr index_t GemmMPerThreadSubC = 4;
constexpr index_t GemmNPerThreadSubC = 4;
constexpr index_t GemmMLevel0Cluster = 4;
constexpr index_t GemmNLevel0Cluster = 2;
constexpr index_t GemmMLevel1Cluster = 2;
constexpr index_t GemmNLevel1Cluster = 4;
constexpr index_t GemmKPerThreadLoop = 1;
constexpr index_t GemmDataPerReadA = 4;
constexpr index_t GemmDataPerReadB = 4;
constexpr index_t OutThreadCopyDataPerWrite = 2;
constexpr index_t BlockSize = 128;
#elif 0
// for 3x3, 34x34, v1r2, Pascal, in-block-copy1
constexpr index_t NPerBlock = 4;
constexpr index_t KPerBlock = 64;
constexpr index_t CPerBlock = 8;
constexpr index_t HoPerBlock = 4;
constexpr index_t WoPerBlock = 8;
constexpr index_t NPerThread = 4;
constexpr index_t KPerThread = 8;
constexpr index_t HoPerThread = 1;
constexpr index_t WoPerThread = 2;
constexpr index_t InBlockCopy_ThreadPerDimC = 4;
constexpr index_t InBlockCopy_ThreadPerDimH = 4;
constexpr index_t InBlockCopy_ThreadPerDimW = 2;
constexpr index_t InBlockCopy_ThreadPerDimN = 1;
constexpr index_t InBlockCopyDataPerRead = 4;
constexpr index_t WeiBlockCopyDataPerRead = 4;
constexpr index_t GemmMPerThreadSubC = 4;
constexpr index_t GemmNPerThreadSubC = 4;
constexpr index_t GemmMLevel0Cluster = 4;
constexpr index_t GemmNLevel0Cluster = 2;
constexpr index_t GemmMLevel1Cluster = 2;
constexpr index_t GemmNLevel1Cluster = 2;
constexpr index_t GemmKPerThreadLoop = 1;
constexpr index_t GemmDataPerReadA = 4;
constexpr index_t GemmDataPerReadB = 4;
constexpr index_t OutThreadCopyDataPerWrite = 2;
constexpr index_t BlockSize = 128;
#elif 0
// for 3x3, 34x34, v1r1, Vega 20
constexpr index_t NPerBlock = 16;
constexpr index_t KPerBlock = 128;
constexpr index_t CPerBlock = 4;
constexpr index_t HoPerBlock = 2;
constexpr index_t WoPerBlock = 4;
constexpr index_t NPerThread = 4;
constexpr index_t KPerThread = 8;
constexpr index_t HoPerThread = 1;
constexpr index_t WoPerThread = 2;
constexpr index_t GemmMPerThreadSubC = 4;
constexpr index_t GemmNPerThreadSubC = 4;
constexpr index_t GemmMLevel0Cluster = 4;
constexpr index_t GemmNLevel0Cluster = 4;
constexpr index_t GemmMLevel1Cluster = 4;
constexpr index_t GemmNLevel1Cluster = 2;
constexpr index_t GemmKPerThreadLoop = 1;
constexpr index_t GemmDataPerReadA = 4;
constexpr index_t GemmDataPerReadB = 4;
constexpr index_t InBlockCopy_ThreadPerDimC = 4;
constexpr index_t InBlockCopy_ThreadPerDimH = 4;
constexpr index_t InBlockCopy_ThreadPerDimW = 2;
constexpr index_t InBlockCopy_ThreadPerDimN = 8;
constexpr index_t InBlockCopyDataPerRead = 2;
constexpr index_t WeiBlockCopyDataPerRead = 2;
constexpr index_t OutThreadCopyDataPerWrite = 4;
constexpr index_t BlockSize = 256;
#elif 0
// for 3x3, 56x56, v1, Pascal
constexpr index_t NPerBlock = 32;
constexpr index_t KPerBlock = 64;
constexpr index_t CPerBlock = 4;
constexpr index_t HoPerBlock = 2;
constexpr index_t WoPerBlock = 2;
constexpr index_t NPerThread = 4;
constexpr index_t KPerThread = 8;
constexpr index_t HoPerThread = 1;
constexpr index_t WoPerThread = 2;
constexpr index_t InBlockCopy_ThreadPerDimC = 1;
constexpr index_t InBlockCopy_ThreadPerDimH = 4;
constexpr index_t InBlockCopy_ThreadPerDimW = 4;
constexpr index_t InBlockCopy_ThreadPerDimN = 8;
constexpr index_t InBlockCopyDataPerRead = 4;
constexpr index_t WeiBlockCopyDataPerRead = 4;
constexpr index_t GemmMPerThreadSubC = 4;
constexpr index_t GemmNPerThreadSubC = 4;
constexpr index_t GemmMLevel0Cluster = 4;
constexpr index_t GemmNLevel0Cluster = 2;
constexpr index_t GemmMLevel1Cluster = 2;
constexpr index_t GemmNLevel1Cluster = 4;
constexpr index_t GemmKPerThreadLoop = 1;
constexpr index_t OutThreadCopyDataPerWrite = 2;
constexpr index_t BlockSize = 128;
#elif 0
// for 3x3, 56x56, v1r2, Pascal
constexpr index_t NPerBlock = 16;
constexpr index_t KPerBlock = 128;
constexpr index_t CPerBlock = 8;
constexpr index_t HoPerBlock = 2;
constexpr index_t WoPerBlock = 2;
constexpr index_t NPerThread = 4;
constexpr index_t KPerThread = 8;
constexpr index_t HoPerThread = 1;
constexpr index_t WoPerThread = 2;
constexpr index_t GemmMPerThreadSubC = 4;
constexpr index_t GemmNPerThreadSubC = 4;
constexpr index_t GemmMLevel0Cluster = 4;
constexpr index_t GemmNLevel0Cluster = 2;
constexpr index_t GemmMLevel1Cluster = 4;
constexpr index_t GemmNLevel1Cluster = 2;
constexpr index_t GemmKPerThreadLoop = 1;
constexpr index_t GemmDataPerReadA = 1;
constexpr index_t GemmDataPerReadB = 1;
constexpr index_t InBlockCopy_ThreadPerDimC = 1;
constexpr index_t InBlockCopy_ThreadPerDimH = 2;
constexpr index_t InBlockCopy_ThreadPerDimW = 4;
constexpr index_t InBlockCopy_ThreadPerDimN = 4;
constexpr index_t InBlockCopyDataPerRead = 4;
constexpr index_t WeiBlockCopyDataPerRead = 4;
constexpr index_t OutThreadCopyDataPerWrite = 4;
constexpr index_t BlockSize = 128; constexpr index_t BlockSize = 128;
#elif 0
// for 3x3, 28x28, v1r1, Pacal
constexpr index_t NPerBlock = 32;
constexpr index_t KPerBlock = 64;
constexpr index_t CPerBlock = 4;
constexpr index_t HoPerBlock = 2;
constexpr index_t WoPerBlock = 2;
constexpr index_t NPerThread = 4;
constexpr index_t KPerThread = 8;
constexpr index_t HoPerThread = 1;
constexpr index_t WoPerThread = 2;
constexpr index_t InBlockCopy_ThreadPerDimC = 1;
constexpr index_t InBlockCopy_ThreadPerDimH = 4;
constexpr index_t InBlockCopy_ThreadPerDimW = 4;
constexpr index_t InBlockCopy_ThreadPerDimN = 8;
constexpr index_t InBlockCopyDataPerRead = 4;
constexpr index_t WeiBlockCopyDataPerRead = 4;
constexpr index_t GemmMPerThreadSubC = 4;
constexpr index_t GemmNPerThreadSubC = 4;
constexpr index_t GemmMLevel0Cluster = 4;
constexpr index_t GemmNLevel0Cluster = 2;
constexpr index_t GemmMLevel1Cluster = 2;
constexpr index_t GemmNLevel1Cluster = 4;
constexpr index_t GemmKPerThreadLoop = 1;
constexpr index_t GemmDataPerReadA = 4;
constexpr index_t GemmDataPerReadB = 4;
constexpr index_t OutThreadCopyDataPerWrite = 2;
constexpr index_t BlockSize = 128;
#elif 1
// for 3x3, 28x28, v1r2, Pascal
constexpr index_t NPerBlock = 16; constexpr index_t NPerBlock = 16;
constexpr index_t KPerBlock = 128; constexpr index_t KPerBlock = 128;
constexpr index_t CPerBlock = 8; constexpr index_t CPerBlock = 8;
...@@ -275,14 +77,6 @@ void device_convolution_implicit_gemm_v1_nchw_cyxk_khwn(InDesc, ...@@ -275,14 +77,6 @@ void device_convolution_implicit_gemm_v1_nchw_cyxk_khwn(InDesc,
constexpr index_t HoPerThread = 1; constexpr index_t HoPerThread = 1;
constexpr index_t WoPerThread = 2; constexpr index_t WoPerThread = 2;
constexpr index_t InBlockCopy_ThreadPerDimN = 4;
constexpr index_t InBlockCopy_ThreadPerDimC = 8;
constexpr index_t InBlockCopy_ThreadPerDimH = 2;
constexpr index_t InBlockCopy_ThreadPerDimW = 2;
constexpr index_t InBlockCopyDataPerRead = 2;
constexpr index_t WeiBlockCopyDataPerRead = 4;
constexpr index_t GemmMPerThreadSubC = 4; constexpr index_t GemmMPerThreadSubC = 4;
constexpr index_t GemmNPerThreadSubC = 4; constexpr index_t GemmNPerThreadSubC = 4;
constexpr index_t GemmMLevel0Cluster = 4; constexpr index_t GemmMLevel0Cluster = 4;
...@@ -293,73 +87,16 @@ void device_convolution_implicit_gemm_v1_nchw_cyxk_khwn(InDesc, ...@@ -293,73 +87,16 @@ void device_convolution_implicit_gemm_v1_nchw_cyxk_khwn(InDesc,
constexpr index_t GemmDataPerReadA = 4; constexpr index_t GemmDataPerReadA = 4;
constexpr index_t GemmDataPerReadB = 4; constexpr index_t GemmDataPerReadB = 4;
constexpr index_t OutThreadCopyDataPerWrite = 2; using InBlockReorderSrcSubLengths_NCHW = Sequence<4, 1, 1, 2>;
using InBlockReorderSrcClusterLengths_NCHW = Sequence<4, 8, 2, 2>;
constexpr index_t BlockSize = 128; using InBlockReorderMapThreadCluster2SrcCluster = Sequence<1, 2, 3, 0>;
#elif 0 constexpr index_t InBlockReorderDataPerRead_W = 2;
// for 1x1, 28x28 constexpr index_t InBlockReorderDataPerWrite_N = 4;
constexpr index_t NPerBlock = 16;
constexpr index_t KPerBlock = 128;
constexpr index_t CPerBlock = 8;
constexpr index_t HoPerBlock = 2;
constexpr index_t WoPerBlock = 2;
constexpr index_t NPerThread = 4;
constexpr index_t KPerThread = 16;
constexpr index_t CPerThread = 1;
constexpr index_t HoPerThread = 1;
constexpr index_t WoPerThread = 1;
constexpr index_t InBlockCopy_ThreadPerDimC = 8;
constexpr index_t InBlockCopy_ThreadPerDimH = 2;
constexpr index_t InBlockCopy_ThreadPerDimW = 2;
constexpr index_t InBlockCopy_ThreadPerDimN = 4;
constexpr index_t InBlockCopyDataPerRead = 4;
constexpr index_t WeiBlockCopyDataPerRead = 4;
constexpr index_t GemmMPerThreadSubC = 4;
constexpr index_t GemmNPerThreadSubC = 4;
constexpr index_t GemmMLevel0Cluster = 4;
constexpr index_t GemmNLevel0Cluster = 2;
constexpr index_t GemmMLevel1Cluster = 2;
constexpr index_t GemmNLevel1Cluster = 4;
constexpr index_t GemmKPerThreadLoop = 1;
constexpr index_t OutThreadCopyDataPerWrite = 2;
constexpr index_t BlockSize = 128; using WeiBlockCopyClusterLengths = Sequence<4, 1, 32>;
#elif 1 constexpr index_t WeiBlockCopyDataPerRead_C = 4;
// for 1x1, 14x14, Pascal
constexpr index_t NPerBlock = 16;
constexpr index_t KPerBlock = 128;
constexpr index_t CPerBlock = 8;
constexpr index_t HoPerBlock = 2;
constexpr index_t WoPerBlock = 2;
constexpr index_t NPerThread = 8;
constexpr index_t KPerThread = 8;
constexpr index_t HoPerThread = 1;
constexpr index_t WoPerThread = 1;
constexpr index_t GemmMPerThreadSubC = 4;
constexpr index_t GemmNPerThreadSubC = 4;
constexpr index_t GemmMLevel0Cluster = 4;
constexpr index_t GemmNLevel0Cluster = 2;
constexpr index_t GemmMLevel1Cluster = 4;
constexpr index_t GemmNLevel1Cluster = 2;
constexpr index_t GemmKPerThreadLoop = 1;
constexpr index_t InBlockCopy_ThreadPerDimC = 8;
constexpr index_t InBlockCopy_ThreadPerDimH = 2;
constexpr index_t InBlockCopy_ThreadPerDimW = 2;
constexpr index_t InBlockCopy_ThreadPerDimN = 4;
constexpr index_t InBlockCopyDataPerRead = 4;
constexpr index_t WeiBlockCopyDataPerRead = 4; constexpr index_t OutThreadCopyDataPerWrite_N = 2;
constexpr index_t OutThreadCopyDataPerWrite = 2;
constexpr index_t BlockSize = 128;
#endif #endif
constexpr index_t GridSize = constexpr index_t GridSize =
...@@ -398,13 +135,14 @@ void device_convolution_implicit_gemm_v1_nchw_cyxk_khwn(InDesc, ...@@ -398,13 +135,14 @@ void device_convolution_implicit_gemm_v1_nchw_cyxk_khwn(InDesc,
GemmKPerThreadLoop, GemmKPerThreadLoop,
GemmDataPerReadA, GemmDataPerReadA,
GemmDataPerReadB, GemmDataPerReadB,
Sequence<InBlockCopy_ThreadPerDimN, InBlockReorderSrcSubLengths_NCHW,
InBlockCopy_ThreadPerDimC, InBlockReorderSrcClusterLengths_NCHW,
InBlockCopy_ThreadPerDimH, InBlockReorderMapThreadCluster2SrcCluster,
InBlockCopy_ThreadPerDimW>, InBlockReorderDataPerRead_W,
InBlockCopyDataPerRead, InBlockReorderDataPerWrite_N,
WeiBlockCopyDataPerRead, WeiBlockCopyClusterLengths,
OutThreadCopyDataPerWrite>{}; WeiBlockCopyDataPerRead_C,
OutThreadCopyDataPerWrite_N>{};
float time = launch_kernel(run_gridwise_convolution<decltype(gridwise_conv), T>, float time = launch_kernel(run_gridwise_convolution<decltype(gridwise_conv), T>,
dim3(GridSize), dim3(GridSize),
......
driver/driver.hip.cpp
View file @ 5ce19234
...@@ -673,7 +673,7 @@ int main(int argc, char* argv[]) ...@@ -673,7 +673,7 @@ int main(int argc, char* argv[])
device_direct_convolution_2_nchw_kcyx_nkhw device_direct_convolution_2_nchw_kcyx_nkhw
#elif 0 #elif 0
device_direct_convolution_2_vectorized_nchw_kcyx_nkhw device_direct_convolution_2_vectorized_nchw_kcyx_nkhw
#elif 1 #elif 0
device_convolution_implicit_gemm_v1_chwn_cyxk_khwn device_convolution_implicit_gemm_v1_chwn_cyxk_khwn
#elif 1 #elif 1
device_convolution_implicit_gemm_v1_nchw_cyxk_khwn device_convolution_implicit_gemm_v1_nchw_cyxk_khwn
......
src/include/blockwise_3d_tensor_op.hip.hpp
View file @ 5ce19234
...@@ -231,7 +231,7 @@ struct Blockwise3dTensorCopy3 ...@@ -231,7 +231,7 @@ struct Blockwise3dTensorCopy3
} }
} }
__device__ constexpr index_t GetRegisterClipboardSize() const __device__ static constexpr index_t GetRegisterClipboardSize()
{ {
static_assert(is_same<Float, float>::value, "wrong! only support float!\n"); static_assert(is_same<Float, float>::value, "wrong! only support float!\n");
......
src/include/blockwise_4d_tensor_op.hip.hpp
View file @ 5ce19234
...@@ -761,7 +761,6 @@ struct Blockwise4dTensorCopyReorder1 ...@@ -761,7 +761,6 @@ struct Blockwise4dTensorCopyReorder1
} }
}; };
#if 1
template <index_t BlockSize, template <index_t BlockSize,
class Float, class Float,
class SrcDesc, class SrcDesc,
...@@ -1070,36 +1069,17 @@ struct Blockwise4dTensorCopyReorder3 ...@@ -1070,36 +1069,17 @@ struct Blockwise4dTensorCopyReorder3
} }
#endif #endif
#if 1 threadwise_4d_tensor_copy_reorder_given_dst2src_v2(
threadwise_4d_tensor_copy_reorder_by_get_dst_from_src(
thread_tensor_desc, thread_tensor_desc,
p_clipboard + clipboard_offset, p_clipboard + clipboard_offset,
DstDesc{}, DstDesc{},
p_dst + dst_offset + mDstMyThreadOffset, p_dst + dst_offset + mDstMyThreadOffset,
thread_sub_tensor_lengths, thread_sub_tensor_lengths,
MapDst2Src{}); MapDst2Src{});
#endif
} }
} }
} }
} }
#if 0
if(get_block_1d_id() == 0)
{
printf("tid %5u, "
"data: %f %f %f %f %f %f %f %f\n",
get_thread_local_1d_id(),
p_clipboard[0],
p_clipboard[1],
p_clipboard[2],
p_clipboard[3],
p_clipboard[4],
p_clipboard[5],
p_clipboard[6],
p_clipboard[7]);
}
#endif
} }
__device__ void Run(const Float* __restrict__ p_src, Float* __restrict__ p_dst) const __device__ void Run(const Float* __restrict__ p_src, Float* __restrict__ p_dst) const
...@@ -1110,4 +1090,3 @@ struct Blockwise4dTensorCopyReorder3 ...@@ -1110,4 +1090,3 @@ struct Blockwise4dTensorCopyReorder3
RunStoreRegisterClipboard(p_clipboard, p_dst); RunStoreRegisterClipboard(p_clipboard, p_dst);
} }
}; };
#endif
src/include/gridwise_convolution_implicit_gemm_v1r2_nchw_cyxk_khwn.hip.hpp
View file @ 5ce19234
...@@ -33,10 +33,14 @@ template <index_t GridSize, ...@@ -33,10 +33,14 @@ template <index_t GridSize,
index_t GemmKPerThreadLoop, index_t GemmKPerThreadLoop,
index_t GemmDataPerReadA, index_t GemmDataPerReadA,
index_t GemmDataPerReadB, index_t GemmDataPerReadB,
class InBlockCopyThreadPerDims, class InBlockReorderSrcSubLengths_NCHW,
index_t InBlockCopyDataPerRead, class InBlockReorderSrcClusterLengths_NCHW,
index_t WeiBlockCopyDataPerRead, class InBlockReorderMapThreadCluster2SrcCluster,
index_t OutThreadCopyDataPerWrite> index_t InBlockReorderDataPerRead_W,
index_t InBlockReorderDataPerWrite_N,
class WeiBlockCopyClusterLengths_KXC,
index_t WeiBlockCopyDataPerRead_C,
index_t OutThreadCopyDataPerWrite_N>
struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn
{ {
__device__ void Run(const Float* const __restrict__ p_in_global, __device__ void Run(const Float* const __restrict__ p_in_global,
...@@ -101,8 +105,10 @@ struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn ...@@ -101,8 +105,10 @@ struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn
// LDS tensor view // LDS tensor view
// be careful of alignment // be careful of alignment
constexpr index_t max_align = mod_conv::max( constexpr index_t max_align = mod_conv::max(InBlockReorderDataPerWrite_N,
InBlockCopyDataPerRead, WeiBlockCopyDataPerRead, GemmDataPerReadA, GemmDataPerReadB); WeiBlockCopyDataPerRead_C,
GemmDataPerReadA,
GemmDataPerReadB);
constexpr auto in_c_h_w_n_block_desc = make_ConstantTensorDescriptor_aligned( constexpr auto in_c_h_w_n_block_desc = make_ConstantTensorDescriptor_aligned(
Sequence<CPerBlock, HoPerBlock, WiPerBlock, NPerBlock>{}, Number<max_align>{}); Sequence<CPerBlock, HoPerBlock, WiPerBlock, NPerBlock>{}, Number<max_align>{});
...@@ -117,68 +123,38 @@ struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn ...@@ -117,68 +123,38 @@ struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn
// blockwise copy // blockwise copy
// input: format is [N, C, Hi, Wi] to [C, Hi, Wi, N] // input: format is [N, C, Hi, Wi] to [C, Hi, Wi, N]
auto map_chwn2nchw = Sequence<1, 2, 3, 0>{}; auto map_chwn2nchw = Sequence<1, 2, 3, 0>{};
#if 0
const auto blockwise_in_copy_reorder =
Blockwise4dTensorCopyReorder1<BlockSize,
Float,
decltype(in_n_c_h_w_global_desc),
decltype(in_c_h_w_n_block_desc),
Sequence<NPerBlock, CPerBlock, HoPerBlock, WiPerBlock>,
decltype(map_chwn2nchw)>{};
#else
auto map_thread_cluster_2_src_cluster = Sequence<1, 2, 0, 3>{};
const auto blockwise_in_copy_reorder = const auto blockwise_in_copy_reorder =
Blockwise4dTensorCopyReorder3<BlockSize, Blockwise4dTensorCopyReorder3<BlockSize,
Float, Float,
decltype(in_n_c_h_w_global_desc), decltype(in_n_c_h_w_global_desc),
decltype(in_c_h_w_n_block_desc), decltype(in_c_h_w_n_block_desc),
Sequence<NPerBlock, CPerBlock, HoPerBlock, WiPerBlock>, Sequence<NPerBlock, CPerBlock, HoPerBlock, WiPerBlock>,
Sequence<4, 1, 1, 2>, InBlockReorderSrcSubLengths_NCHW,
Sequence<4, 8, 2, 2>, InBlockReorderSrcClusterLengths_NCHW,
decltype(map_chwn2nchw), decltype(map_chwn2nchw),
decltype(map_thread_cluster_2_src_cluster), InBlockReorderMapThreadCluster2SrcCluster,
2, InBlockReorderDataPerRead_W,
4>{}; InBlockReorderDataPerWrite_N>{};
#if 0
if(get_thread_local_1d_id() == 0 && get_block_1d_id() == 0)
{
printf("size %u\n", blockwise_in_copy_reorder.GetRegisterClipboardSize());
}
#endif
#endif
// blockwise wei copy // blockwise wei copy
// format is [CPerBlock, X * KPerBlock] // format is [CPerBlock, X * KPerBlock]
const auto blockwise_wei_copy = const auto blockwise_wei_copy =
#if 0
Blockwise3dTensorCopy1<BlockSize,
Float,
decltype(wei_c_x_k_global_desc),
decltype(wei_c_x_k_block_desc),
decltype(wei_c_x_k_block_desc.GetLengths()),
WeiBlockCopyDataPerRead>{};
#else
Blockwise3dTensorCopy3<BlockSize, Blockwise3dTensorCopy3<BlockSize,
Float, Float,
decltype(wei_c_x_k_global_desc), decltype(wei_c_x_k_global_desc),
decltype(wei_c_x_k_block_desc), decltype(wei_c_x_k_block_desc),
decltype(wei_c_x_k_block_desc.GetLengths()), decltype(wei_c_x_k_block_desc.GetLengths()),
Sequence<4, 1, 32>, Sequence<4, 1, 32>,
WeiBlockCopyDataPerRead>{}; WeiBlockCopyDataPerRead_C>{};
#endif
// a series of blockwise batched GEMM // a series of blockwise batched GEMM
// C_matrix += transpose(A_matrix) * B_matrix // C_matrix += transpose(A_matrix) * B_matrix
// A_matrix and B_matrix saved in LDS, C_matrix saved in register // A_matrix and B_matrix saved in LDS, C_matrix saved in register
// A_matrix[C,K] is a sub-matrix of wei_block[C,K] // A_matrix[C,K] is a sub-matrix of wei_block[C,K]
// B_matrix[C,Wo*N] is a sub-matrix of in_block[C,Hi,Wi,N] // B_matrix[C,Wo*N] is a sub-matrix of in_block[C,Hi,Wi,N]
// C_matrix[K,Wo*N] is a sub-matrix of out_block[K,Ho,Wo,N] // C_matrix[K,Wo*N] is a sub-matrix of out_block[K,Ho,Wo,N]
constexpr auto a_c_k_block_mtx_desc = constexpr auto a_c_k_block_mtx_desc = make_ConstantMatrixDescriptor(
make_ConstantMatrixDescriptor(Number<CPerBlock>{}, Number<CPerBlock>{}, Number<KPerBlock>{}, Number<wei_c_x_k_block_desc.GetStride(I0)>{});
Number<KPerBlock>{},
Number<wei_c_x_k_block_desc.GetStride(I0)>{});
constexpr auto b_c_wn_block_mtx_desc = constexpr auto b_c_wn_block_mtx_desc =
make_ConstantMatrixDescriptor(Number<CPerBlock>{}, make_ConstantMatrixDescriptor(Number<CPerBlock>{},
...@@ -252,6 +228,7 @@ struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn ...@@ -252,6 +228,7 @@ struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn
{ {
for(index_t y = 0; y < Y; ++y) for(index_t y = 0; y < Y; ++y)
{ {
#if 1
blockwise_in_copy_reorder.Run(p_in_global_block_offset + blockwise_in_copy_reorder.Run(p_in_global_block_offset +
in_n_c_h_w_global_desc.Get1dIndex(0, 0, y, 0), in_n_c_h_w_global_desc.Get1dIndex(0, 0, y, 0),
p_in_block); p_in_block);
...@@ -259,6 +236,23 @@ struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn ...@@ -259,6 +236,23 @@ struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn
blockwise_wei_copy.Run(p_wei_global_block_offset + blockwise_wei_copy.Run(p_wei_global_block_offset +
wei_c_y_x_k_global_desc.Get1dIndex(0, y, 0, 0), wei_c_y_x_k_global_desc.Get1dIndex(0, y, 0, 0),
p_wei_block); p_wei_block);
#else
Float p_in_clipboard[blockwise_in_copy_reorder.GetRegisterClipboardSize()];
Float p_wei_clipboard[blockwise_wei_copy.GetRegisterClipboardSize()];
blockwise_in_copy_reorder.RunLoadRegisterClipboard(
p_in_global_block_offset + in_n_c_h_w_global_desc.Get1dIndex(0, 0, y, 0),
p_in_clipboard);
blockwise_wei_copy.RunLoadRegisterClipboard(
p_wei_global_block_offset + wei_c_y_x_k_global_desc.Get1dIndex(0, y, 0, 0),
p_wei_clipboard);
blockwise_wei_copy.RunStoreRegisterClipboard(p_wei_clipboard, p_wei_block);
blockwise_in_copy_reorder.RunStoreRegisterClipboard(p_in_clipboard, p_in_block);
#endif
__syncthreads(); __syncthreads();
...@@ -274,42 +268,7 @@ struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn ...@@ -274,42 +268,7 @@ struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn
} }
} }
// output: register to global mem, // output: register to global mem,
#if 0
const auto c_thread_mtx_begin =
blockwise_batch_gemm.GetBeginOfThreadMatrixC(get_thread_local_1d_id());
for(index_t k = 0; k < out_khwn_thread_desc.GetLength(I0); ++k)
{
for(index_t ho = 0; ho < out_khwn_thread_desc.GetLength(I1); ++ho)
{
for(index_t wo = 0; wo < out_khwn_thread_desc.GetLength(I2); ++wo)
{
for(index_t n = 0; n < out_khwn_thread_desc.GetLength(I3); ++n)
{
const index_t b = out_khwn_thread_desc.Get1dIndex(0, 0, wo, n);
const auto c_thread_mtx_distance =
blockwise_batch_gemm.GetDistanceFromBeginOfThreadMatrixC(ho, k, b);
const index_t ho_thread =
c_thread_mtx_begin.batch + c_thread_mtx_distance.batch;
const index_t k_thread = c_thread_mtx_begin.row + c_thread_mtx_distance.row;
const index_t b_thread = c_thread_mtx_begin.col + c_thread_mtx_distance.col;
const index_t wo_thread = b_thread / NPerBlock;
const index_t n_thread = b_thread % NPerBlock;
p_out_global[out_khwn_global_desc.Get1dIndex(k_block_data_begin + k_thread,
ho_block_data_begin + ho_thread,
wo_block_data_begin + wo_thread,
n_block_data_begin + n_thread)] =
p_out_thread[out_khwn_thread_desc.Get1dIndex(k, ho, wo, n)];
}
}
}
}
#elif 1
const auto c_thread_mtx_begin = const auto c_thread_mtx_begin =
blockwise_batch_gemm.GetBeginOfThreadMatrixC(get_thread_local_1d_id()); blockwise_batch_gemm.GetBeginOfThreadMatrixC(get_thread_local_1d_id());
...@@ -356,7 +315,6 @@ struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn ...@@ -356,7 +315,6 @@ struct GridwiseConvolutionImplicitGemm_v1r2_nchw_cyxk_khwn
wo_block_data_begin + wo_thread_data_begin, wo_block_data_begin + wo_thread_data_begin,
n_block_data_begin + n_thread_data_begin), n_block_data_begin + n_thread_data_begin),
out_10d_thread_desc.GetLengths(), out_10d_thread_desc.GetLengths(),
Number<OutThreadCopyDataPerWrite>{}); Number<OutThreadCopyDataPerWrite_N>{});
#endif
} }
}; };
src/include/threadwise_4d_tensor_op.hip.hpp
View file @ 5ce19234
...@@ -44,7 +44,7 @@ template <class SrcData, ...@@ -44,7 +44,7 @@ template <class SrcData,
class SrcOpLengths, class SrcOpLengths,
class MapDst2Src, class MapDst2Src,
class F> class F>
__device__ void threadwise_4d_tensor_pointwise_operation_binary_reorder_by_get_dst_from_src( __device__ void threadwise_4d_tensor_pointwise_operation_binary_reorder_given_dst2src(
SrcDesc, SrcDesc,
const SrcData* __restrict__ p_src, const SrcData* __restrict__ p_src,
DstDesc, DstDesc,
...@@ -82,9 +82,9 @@ __device__ void threadwise_4d_tensor_pointwise_operation_binary_reorder_by_get_d ...@@ -82,9 +82,9 @@ __device__ void threadwise_4d_tensor_pointwise_operation_binary_reorder_by_get_d
const index_t bindex = const index_t bindex =
dst_desc.Get1dIndex(did[IR0], did[IR1], did[IR2], did[IR3]); dst_desc.Get1dIndex(did[IR0], did[IR1], did[IR2], did[IR3]);
#if 1
f(p_src[aindex], p_dst[bindex]); f(p_src[aindex], p_dst[bindex]);
#else
#if 0
if(get_block_1d_id() == 0) if(get_block_1d_id() == 0)
{ {
printf("tid %5u, " printf("tid %5u, "
...@@ -126,17 +126,16 @@ template <class SrcData, ...@@ -126,17 +126,16 @@ template <class SrcData,
class DstDesc, class DstDesc,
class SrcOpLengths, class SrcOpLengths,
class MapDst2Src> class MapDst2Src>
__device__ void __device__ void threadwise_4d_tensor_copy_reorder_given_dst2src(SrcDesc,
threadwise_4d_tensor_copy_reorder_by_get_dst_from_src(SrcDesc, const SrcData* __restrict__ p_src,
const SrcData* __restrict__ p_src, DstDesc,
DstDesc, DstData* __restrict__ p_dst,
DstData* __restrict__ p_dst, SrcOpLengths,
SrcOpLengths, MapDst2Src)
MapDst2Src)
{ {
auto f_copy = [](const SrcData& src, DstData& dst) { dst = static_cast<DstData>(src); }; auto f_copy = [](const SrcData& src, DstData& dst) { dst = static_cast<DstData>(src); };
threadwise_4d_tensor_pointwise_operation_binary_reorder_by_get_dst_from_src( threadwise_4d_tensor_pointwise_operation_binary_reorder_given_dst2src(
SrcDesc{}, p_src, DstDesc{}, p_dst, SrcOpLengths{}, MapDst2Src{}, f_copy); SrcDesc{}, p_src, DstDesc{}, p_dst, SrcOpLengths{}, MapDst2Src{}, f_copy);
} }
...@@ -146,7 +145,7 @@ __device__ void threadwise_4d_tensor_copy( ...@@ -146,7 +145,7 @@ __device__ void threadwise_4d_tensor_copy(
{ {
auto dst_from_src_reorder = Sequence<0, 1, 2, 3>{}; auto dst_from_src_reorder = Sequence<0, 1, 2, 3>{};
threadwise_4d_tensor_copy_reorder_by_get_dst_from_src( threadwise_4d_tensor_copy_reorder_given_dst2src(
SrcDesc{}, p_src, DstDesc{}, p_dst, SrcOpLengths{}, dst_from_src_reorder); SrcDesc{}, p_src, DstDesc{}, p_dst, SrcOpLengths{}, dst_from_src_reorder);
} }
...@@ -212,6 +211,61 @@ __device__ void threadwise_4d_tensor_copy_v2(SrcDesc, ...@@ -212,6 +211,61 @@ __device__ void threadwise_4d_tensor_copy_v2(SrcDesc,
} }
} }
template <class SrcData,
class DstData,
class SrcDesc,
class DstDesc,
class SrcOpLengths,
class MapDst2Src>
__device__ void
threadwise_4d_tensor_copy_reorder_given_dst2src_v2(SrcDesc,
const SrcData* __restrict__ p_src,
DstDesc,
DstData* __restrict__ p_dst,
SrcOpLengths,
MapDst2Src)
{
constexpr auto I0 = Number<0>{};
constexpr auto I1 = Number<1>{};
constexpr auto I2 = Number<2>{};
constexpr auto I3 = Number<3>{};
constexpr index_t IR0 = MapDst2Src{}.Get(I0);
constexpr index_t IR1 = MapDst2Src{}.Get(I1);
constexpr index_t IR2 = MapDst2Src{}.Get(I2);
constexpr index_t IR3 = MapDst2Src{}.Get(I3);
constexpr auto src_desc = SrcDesc{};
constexpr auto dst_desc = DstDesc{};
// ref_desc has dst_desc's ordering
constexpr auto ref_desc =
make_ConstantTensorDescriptor(SrcOpLengths{}.ReorderGivenNew2Old(MapDst2Src{}));
for(index_t did0 = 0; did0 < ref_desc.GetLength(I0); ++did0)
{
for(index_t did1 = 0; did1 < ref_desc.GetLength(I1); ++did1)
{
for(index_t did2 = 0; did2 < ref_desc.GetLength(I2); ++did2)
{
for(index_t did3 = 0; did3 < ref_desc.GetLength(I3); ++did3)
{
const auto dst_multi_id = Array<index_t, 4>{did0, did1, did2, did3};
const auto src_multi_id =
reorder_array_given_old2new(dst_multi_id, MapDst2Src{});
const index_t dst_index = dst_desc.Get1dIndex(dst_multi_id);
const index_t src_index = src_desc.Get1dIndex(src_multi_id);
p_dst[dst_index] = p_src[src_index];
}
}
}
}
}
template <class Float, class Desc, class IDim, class NShift> template <class Float, class Desc, class IDim, class NShift>
__device__ void threadwise_4d_tensor_shift_down(Desc, Float* __restrict__ p, IDim, NShift) __device__ void threadwise_4d_tensor_shift_down(Desc, Float* __restrict__ p, IDim, NShift)
{ {
......
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