Merge remote-tracking branch 'origin/develop' into cpu_avx2

07a673c6 · carlushuang · c0f698d5 · ac0d8066 · 07a673c6 · 07a673c6
Commit 07a673c6 authored Apr 14, 2022 by carlushuang
20 changed files
--- a/include/ck/tensor_operation/gpu/device/device_gemm_xdl_cshuffle.hpp
+++ b/include/ck/tensor_operation/gpu/device/device_gemm_xdl_cshuffle.hpp
--- a/include/ck/tensor_operation/gpu/device/device_gemm_xdl_splitk.hpp
+++ b/include/ck/tensor_operation/gpu/device/device_gemm_xdl_splitk.hpp
--- a/include/ck/tensor_operation/gpu/device/device_gemm_xdl_splitk_c_shuffle.hpp
+++ b/include/ck/tensor_operation/gpu/device/device_gemm_xdl_splitk_c_shuffle.hpp
--- a/include/ck/tensor_operation/gpu/device/device_grouped_gemm_xdl.hpp
+++ b/include/ck/tensor_operation/gpu/device/device_grouped_gemm_xdl.hpp
--- a/include/ck/tensor_operation/gpu/device/device_pool2d_fwd.hpp
+++ b/include/ck/tensor_operation/gpu/device/device_pool2d_fwd.hpp
--- a/include/ck/tensor_operation/gpu/device/device_pool2d_fwd_nhwc_nhwc.hpp
+++ b/include/ck/tensor_operation/gpu/device/device_pool2d_fwd_nhwc_nhwc.hpp
--- a/include/ck/tensor_operation/gpu/device/gemm_specialization.hpp
+++ b/include/ck/tensor_operation/gpu/device/gemm_specialization.hpp
--- a/include/ck/tensor_operation/gpu/device/reduction_operator_mapping.hpp
+++ b/include/ck/tensor_operation/gpu/device/reduction_operator_mapping.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_2d_reduction_blockwise.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_2d_reduction_blockwise.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_2d_reduction_multiblock_atomic_add.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_2d_reduction_multiblock_atomic_add.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_2d_reduction_multiblock_partial_reduce.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_2d_reduction_multiblock_partial_reduce.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_2d_reduction_threadwise.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_2d_reduction_threadwise.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_contraction_dlops_v1r2.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_contraction_dlops_v1r2.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_dlops_v1r2.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_dlops_v1r2.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_dlops_v1r3.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_dlops_v1r3.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_dlops_v2.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_dlops_v2.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_dlops_v3.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_dlops_v3.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_reduce_xdl_cshuffle_v1.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_reduce_xdl_cshuffle_v1.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_v1.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdl_cshuffle_v1.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r3.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r3.hpp