Merge branch 'amd-develop' into amd-master

example/16_gemm_multi_d_multi_reduces/gemm_max_xdl_bf16.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include "gemm_reduce_xdl_common.hpp"
 

example/16_gemm_multi_d_multi_reduces/gemm_max_xdl_fp16.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include "gemm_reduce_xdl_common.hpp"
 

example/16_gemm_multi_d_multi_reduces/gemm_max_xdl_fp32.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include "gemm_reduce_xdl_common.hpp"
 

example/16_gemm_multi_d_multi_reduces/gemm_max_xdl_int4.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include "gemm_reduce_xdl_common.hpp"
 

example/16_gemm_multi_d_multi_reduces/gemm_max_xdl_int8.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include "gemm_reduce_xdl_common.hpp"
 

example/16_gemm_multi_d_multi_reduces/gemm_mean_meansquare_xdl_bf16.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include "gemm_reduce_xdl_common.hpp"
 

example/16_gemm_multi_d_multi_reduces/gemm_mean_meansquare_xdl_fp16.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include "gemm_reduce_xdl_common.hpp"
 

example/16_gemm_multi_d_multi_reduces/gemm_mean_meansquare_xdl_fp32.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include "gemm_reduce_xdl_common.hpp"
 

example/16_gemm_multi_d_multi_reduces/gemm_reduce_xdl_common.hpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include <numeric>
 #include <initializer_list>

example/17_convnd_bwd_data/CMakeLists.txt

-add_example_executable(example_convnd_bwd_data_xdl_fp16 convnd_bwd_data_xdl_fp16.cpp)
-target_link_libraries(example_convnd_bwd_data_xdl_fp16 PRIVATE utility)
-
+list(APPEND gpu_list gfx908 gfx90a gfx940 gfx941 gfx942)
+set(target 0)
+foreach(gpu IN LISTS GPU_TARGETS)
+ if(gpu IN_LIST gpu_list AND target EQUAL 0)
+   add_example_executable(example_convnd_bwd_data_xdl_fp16 convnd_bwd_data_xdl_fp16.cpp)
+   target_link_libraries(example_convnd_bwd_data_xdl_fp16 PRIVATE utility)
+   set(target 1)
+ endif()
+endforeach()
 add_example_executable(example_convnd_bwd_data_dl_fp16 convnd_bwd_data_dl_fp16.cpp)
 target_link_libraries(example_convnd_bwd_data_dl_fp16 PRIVATE utility)

example/17_convnd_bwd_data/convnd_bwd_data_common.hpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include <iostream>
 #include <numeric>
@@ -81,32 +81,33 @@ int run_conv_bwd_data(bool do_verification,
     in_device_buf.SetZero();
 
     // do GEMM
-    auto conv     = DeviceConvNdBwdDataInstance{};
-    auto invoker  = conv.MakeInvoker();
-    auto argument = conv.MakeArgument(static_cast<InDataType*>(in_device_buf.GetDeviceBuffer()),
-                                      static_cast<WeiDataType*>(wei_device_buf.GetDeviceBuffer()),
-                                      static_cast<OutDataType*>(out_device_buf.GetDeviceBuffer()),
-                                      conv_param.N_,
-                                      conv_param.K_,
-                                      conv_param.C_,
-                                      conv_param.input_spatial_lengths_,
-                                      conv_param.filter_spatial_lengths_,
-                                      conv_param.GetOutputSpatialLengths(),
-                                      conv_param.conv_filter_strides_,
-                                      conv_param.conv_filter_dilations_,
-                                      conv_param.input_left_pads_,
-                                      conv_param.input_right_pads_,
-                                      in_element_op,
-                                      wei_element_op,
-                                      out_element_op);
-
-    if(!conv.IsSupportedArgument(argument))
+    auto conv    = DeviceConvNdBwdDataInstance{};
+    auto invoker = conv.MakeInvoker();
+    auto argument =
+        conv.MakeArgumentPointer(static_cast<InDataType*>(in_device_buf.GetDeviceBuffer()),
+                                 static_cast<WeiDataType*>(wei_device_buf.GetDeviceBuffer()),
+                                 static_cast<OutDataType*>(out_device_buf.GetDeviceBuffer()),
+                                 conv_param.N_,
+                                 conv_param.K_,
+                                 conv_param.C_,
+                                 conv_param.input_spatial_lengths_,
+                                 conv_param.filter_spatial_lengths_,
+                                 conv_param.GetOutputSpatialLengths(),
+                                 conv_param.conv_filter_strides_,
+                                 conv_param.conv_filter_dilations_,
+                                 conv_param.input_left_pads_,
+                                 conv_param.input_right_pads_,
+                                 in_element_op,
+                                 wei_element_op,
+                                 out_element_op);
+
+    if(!conv.IsSupportedArgument(argument.get()))
     {
         std::cout << "Not support,please check parameters or device";
         return 0;
     }
 
-    float ave_time = invoker.Run(argument, StreamConfig{nullptr, time_kernel});
+    float ave_time = invoker.Run(argument.get(), StreamConfig{nullptr, time_kernel});
 
     std::size_t flop      = conv_param.GetFlops();
     std::size_t num_btype = conv_param.GetByte<InDataType, WeiDataType, OutDataType>();

example/17_convnd_bwd_data/convnd_bwd_data_dl_fp16.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include "convnd_bwd_data_common.hpp"
 

example/17_convnd_bwd_data/convnd_bwd_data_xdl_fp16.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include "convnd_bwd_data_common.hpp"
 

example/18_batched_gemm_reduce/CMakeLists.txt

-add_example_executable(example_batched_gemm_reduce_xdl_fp16 batched_gemm_reduce_xdl_fp16.cpp)
-
+list(APPEND gpu_list gfx908 gfx90a gfx940 gfx941 gfx942)
+set(target 0)
+foreach(gpu IN LISTS GPU_TARGETS)
+ if(gpu IN_LIST gpu_list AND target EQUAL 0)
+   add_example_executable(example_batched_gemm_reduce_xdl_fp16 batched_gemm_reduce_xdl_fp16.cpp)
+   set(target 1)
+ endif()
+endforeach()

example/18_batched_gemm_reduce/batched_gemm_reduce_xdl_fp16.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include <iostream>
 #include <numeric>

example/19_binary_elementwise/broadcast_add_2d_amn_bn.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include <iostream>
 #include <cstdlib>

example/19_binary_elementwise/broadcast_add_3d_am_bmnk.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include <iostream>
 #include <cstdlib>

example/19_binary_elementwise/elementwise_add_1d.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include <iostream>
 #include <cstdlib>

example/19_binary_elementwise/elementwise_add_4d.cpp

 // SPDX-License-Identifier: MIT
-// Copyright (c) 2018-2022, Advanced Micro Devices, Inc. All rights reserved.
+// Copyright (c) 2018-2023, Advanced Micro Devices, Inc. All rights reserved.
 
 #include <iostream>
 #include <cstdlib>

example/20_grouped_conv_bwd_weight/CMakeLists.txt

-add_custom_target(example_grouped_conv_bwd_weight)
+list(APPEND gpu_list gfx908 gfx90a gfx940 gfx941 gfx942)
+set(target 0)
+foreach(gpu IN LISTS GPU_TARGETS)
+ if(gpu IN_LIST gpu_list AND target EQUAL 0)
+   add_custom_target(example_grouped_conv_bwd_weight)
 
-add_example_executable(example_grouped_conv_bwd_weight_xdl_fp16 grouped_conv_bwd_weight_xdl_fp16.cpp)
-add_example_executable(example_grouped_conv_bwd_weight_xdl_bf16 grouped_conv_bwd_weight_xdl_bf16.cpp)
+   add_example_executable(example_grouped_conv_bwd_weight_xdl_fp16 grouped_conv_bwd_weight_xdl_fp16.cpp)
+   add_example_executable(example_grouped_conv_bwd_weight_xdl_bf16 grouped_conv_bwd_weight_xdl_bf16.cpp)
 
-
-add_dependencies(example_grouped_conv_bwd_weight example_grouped_conv_bwd_weight_xdl_fp16
+   add_dependencies(example_grouped_conv_bwd_weight example_grouped_conv_bwd_weight_xdl_fp16
                                                  example_grouped_conv_bwd_weight_xdl_bf16)
+   set(target 1)
+ endif()
+endforeach()
 
 add_custom_target(example_grouped_conv_bwd_weight_dl)