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

4769425e · Chao Liu · b548c0be · ba58a93f · 4769425e · 4769425e
Commit 4769425e authored May 23, 2022 by Chao Liu
20 changed files
--- a/include/ck/tensor_operation/gpu/grid/gridwise_binary_elementwise_1d.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_binary_elementwise_1d.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_bwd_weight.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_bwd_weight.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
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r4.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r4.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r4r2.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v2r4r2.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v3r1.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v3r1.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v3r2.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v3r2.hpp
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v3r3.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_v3r3.hpp
--- a/include/ck/utility/amd_buffer_addressing.hpp
+++ b/include/ck/utility/amd_buffer_addressing.hpp
--- a/include/ck/utility/common_header.hpp
+++ b/include/ck/utility/common_header.hpp
@@ -32,7 +32,7 @@
 #include "debug.hpp"

 #include "amd_buffer_addressing.hpp"
-#include "generic_memory_space_atomic_add.hpp"
+#include "generic_memory_space_atomic.hpp"
 #include "get_id.hpp"
 #include "synchronization.hpp"
 #include "amd_address_space.hpp"

--- a/include/ck/utility/dynamic_buffer.hpp
+++ b/include/ck/utility/dynamic_buffer.hpp
--- a/include/ck/utility/generic_memory_space_atomic_add.hpp
+++ b/include/ck/utility/generic_memory_space_atomic_add.hpp
--- a/include/ck/utility/get_id.hpp
+++ b/include/ck/utility/get_id.hpp
--- a/include/ck/utility/statically_indexed_array_multi_index.hpp
+++ b/include/ck/utility/statically_indexed_array_multi_index.hpp
--- a/include/ck/utility/type.hpp
+++ b/include/ck/utility/type.hpp
--- a/library/include/ck/library/host_tensor/device.hpp
+++ b/library/include/ck/library/host_tensor/device.hpp
--- a/library/include/ck/library/reference_tensor_operation/cpu/reference_conv_backward_weight.hpp
+++ b/library/include/ck/library/reference_tensor_operation/cpu/reference_conv_backward_weight.hpp
--- a/library/src/tensor_operation_instance/gpu/batched_gemm_reduce/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/batched_gemm_reduce/CMakeLists.txt
-set(DEVICE_BATCHED_GEMM_REDUCE_INSTANCE_SOURCE 
+set(DEVICE_BATCHED_GEMM_REDUCE_INSTANCE_SOURCE
    device_batched_gemm_reduce_xdl_cshuffle_f16_f16_f16_f32_f32_gmk_gkn_gmn_instance.cpp
    device_batched_gemm_reduce_xdl_cshuffle_f16_f16_f16_f32_f32_gmk_gnk_gmn_instance.cpp
    device_batched_gemm_reduce_xdl_cshuffle_f16_f16_f16_f32_f32_gkm_gkn_gmn_instance.cpp