cleaned up comments, rearranged/renamed files

example/44_elementwise_permute/CMakeLists.txt

 add_example_executable(example_elementwise_permute_4D_fp16 elementwise_permute_4D_fp16.cpp)
 add_example_executable(example_elementwise_permute_4D_fp16_2d elementwise_permute_4D_fp16_2d.cpp)
+add_example_executable(example_elementwise_permute_4D_fp32_row elementwise_permute_4D_fp32_row.cpp)
+add_example_executable(example_elementwise_permute_4D_fp16_row elementwise_permute_4D_fp16_row.cpp)
+add_example_executable(example_elementwise_permute_4D_fp32_col elementwise_permute_4D_fp32_col.cpp)
+add_example_executable(example_elementwise_permute_4D_fp16_col elementwise_permute_4D_fp16_col.cpp)

example/65_hip_tensor_permute/elementwise_permute_4D_fp16_col.cpp → example/44_elementwise_permute/elementwise_permute_4D_fp16_col.cpp

@@ -3,7 +3,7 @@
 
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/element/binary_element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/impl/device_elementwise_impl_ht.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_elementwise_scale_impl.hpp"
 
 #include "ck/library/utility/algorithm.hpp"
 #include "ck/library/utility/check_err.hpp"

example/65_hip_tensor_permute/elementwise_permute_4D_fp16_ht.cpp → example/44_elementwise_permute/elementwise_permute_4D_fp16_row.cpp

@@ -3,7 +3,7 @@
 
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/element/binary_element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/impl/device_elementwise_impl_ht.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_elementwise_scale_impl.hpp"
 
 #include "ck/library/utility/algorithm.hpp"
 #include "ck/library/utility/check_err.hpp"

example/65_hip_tensor_permute/elementwise_permute_4D_fp32_col.cpp → example/44_elementwise_permute/elementwise_permute_4D_fp32_col.cpp

@@ -3,7 +3,7 @@
 
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/element/binary_element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/impl/device_elementwise_impl_ht.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_elementwise_scale_impl.hpp"
 
 #include "ck/library/utility/algorithm.hpp"
 #include "ck/library/utility/check_err.hpp"
@@ -27,7 +27,7 @@ using DeviceElementwisePermuteInstance =
                                                         UnaryOp,              // UnaryOp
                                                         Scale,                // Scalar
                                                         4,                    // NumDim
-                                                        8,                    // MPerThread
+                                                        1,                    // MPerThread
                                                         ck::Sequence<1>,  // InScalarPerVectorSeq
                                                         ck::Sequence<1>>; // OutScalarPerVectorSeq
 
@@ -60,10 +60,11 @@ int main()
     bool do_verification = true;
     bool time_kernel     = true;
 
-    std::vector<std::size_t> nchw = {4, 2, 1, 8};
-    std::vector<std::size_t> nhwc = {4, 1, 8, 2};
+    std::vector<std::size_t> nchw = {5, 4, 2, 3};
+    std::vector<std::size_t> nhwc = {5, 2, 3, 4};
     Tensor<ADataType> a(nchw);
     Tensor<BDataType> b(nhwc);
+
     float scale = 1.f;
     auto i      = 0;
     for(std::size_t w = 0; w < a.mDesc.GetLengths()[3]; ++w)

example/65_hip_tensor_permute/elementwise_permute_4D_fp32_ht.cpp → example/44_elementwise_permute/elementwise_permute_4D_fp32_row.cpp

@@ -3,7 +3,7 @@
 
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/element/binary_element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/impl/device_elementwise_impl_ht.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_elementwise_scale_impl.hpp"
 
 #include "ck/library/utility/algorithm.hpp"
 #include "ck/library/utility/check_err.hpp"

example/65_hip_tensor_permute/CMakeLists.txt

-add_example_executable(example_elementwise_permute_4D_fp32_ht elementwise_permute_4D_fp32_ht.cpp)
-add_example_executable(example_elementwise_permute_4D_fp16_ht elementwise_permute_4D_fp16_ht.cpp)
-add_example_executable(example_elementwise_permute_4D_fp32_col elementwise_permute_4D_fp32_col.cpp)
-add_example_executable(example_elementwise_permute_4D_fp16_col elementwise_permute_4D_fp16_col.cpp)

include/ck/tensor_operation/gpu/device/impl/device_elementwise_impl_ht.hpp → include/ck/tensor_operation/gpu/device/impl/device_elementwise_scale_impl.hpp

@@ -8,8 +8,8 @@
 
 #include "ck/utility/math.hpp"
 #include "ck/utility/sequence.hpp"
-#include "ck/tensor_operation/gpu/device/device_elementwise_ht.hpp"
-#include "ck/tensor_operation/gpu/grid/gridwise_elementwise_1d_ht.hpp"
+#include "ck/tensor_operation/gpu/device/device_elementwise_scale.hpp"
+#include "ck/tensor_operation/gpu/grid/gridwise_elementwise_1d_scale.hpp"
 #include "ck/tensor_description/tensor_descriptor_helper.hpp"
 
 #include "ck/host_utility/kernel_launch.hpp"

include/ck/tensor_operation/gpu/grid/gridwise_elementwise_1d_ht.hpp → include/ck/tensor_operation/gpu/grid/gridwise_elementwise_1d_scale.hpp

@@ -7,11 +7,6 @@
 #include "ck/utility/data_type.hpp"
 #include "ck/tensor_operation/gpu/thread/threadwise_tensor_slice_transfer.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
-#define UNUSED(expr)  \
-    do                \
-    {                 \
-        (void)(expr); \
-    } while(0)
 
 namespace ck {
 
@@ -163,7 +158,6 @@ struct GridwiseElementwise_1D
             },
             Number<NumOutput>{});
 
-        // const auto& scalar = ScalarMult;
         index_t num_iter = M / (loop_step);
         do
         {
@@ -211,7 +205,6 @@ struct GridwiseElementwise_1D
                     Number<NumInput>{});
 
                 unpack2(elementwise_op, out_data_refs, in_data_refs);
-                UNUSED(scale_op);
             });
 
             static_for<0, NumOutput, 1>{}([&](auto I) {