Merge commit '30480288' into conv_quant_int8

client_example/03_gemm_layernorm/gemm_add_add_layernorm.cpp

@@ -8,7 +8,7 @@
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
 #include "ck/tensor_operation/gpu/device/device_gemm_reduce.hpp"
-#include "ck/tensor_operation/gpu/device/device_elementwise.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_elementwise.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/tensor_operation_instance/gpu/device_elementwise_instance.hpp"

example/01_gemm/gemm_dl_fp16.cpp

@@ -3,7 +3,7 @@
 
 #include "common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_gemm_dl.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_dl.hpp"
 
 using ADataType   = ck::half_t;
 using BDataType   = ck::half_t;

example/01_gemm/gemm_dl_fp32.cpp

@@ -3,7 +3,7 @@
 
 #include "common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_gemm_dl.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_dl.hpp"
 
 using ADataType   = float;
 using BDataType   = float;

example/01_gemm/gemm_dl_int4.cpp

@@ -7,7 +7,7 @@
 
 #include "common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_gemm_dl.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_dl.hpp"
 
 using ADataType       = ck::int4_t;
 using BDataType       = ck::int4_t;

example/01_gemm/gemm_dl_int8.cpp

@@ -3,7 +3,7 @@
 
 #include "common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_gemm_dl.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_dl.hpp"
 
 using ADataType   = int8_t;
 using BDataType   = int8_t;

example/01_gemm/gemm_xdl_bf16.cpp

@@ -3,7 +3,7 @@
 
 #include "common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_gemm_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle.hpp"
 
 using ADataType        = ck::bhalf_t;
 using BDataType        = ck::bhalf_t;

example/01_gemm/gemm_xdl_fp16.cpp

@@ -3,8 +3,8 @@
 
 #include "common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_gemm_xdl.hpp"
-#include "ck/tensor_operation/gpu/device/device_gemm_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_xdl.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle.hpp"
 
 using ADataType        = ck::half_t;
 using BDataType        = ck::half_t;

example/01_gemm/gemm_xdl_fp64.cpp

@@ -3,7 +3,7 @@
 
 #include "common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_gemm_xdl.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_xdl.hpp"
 
 using ADataType   = double;
 using BDataType   = double;

example/01_gemm/gemm_xdl_int4.cpp

@@ -7,7 +7,7 @@
 
 #include "common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_gemm_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle.hpp"
 
 using ADataType        = ck::int4_t;
 using BDataType        = ck::int4_t;

example/01_gemm/gemm_xdl_int8.cpp

@@ -3,7 +3,7 @@
 
 #include "common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_gemm_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle.hpp"
 
 using ADataType        = int8_t;
 using BDataType        = int8_t;

example/01_gemm/gemm_xdl_skip_b_lds_fp16.cpp

@@ -3,8 +3,8 @@
 
 #include "common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_gemm_xdl.hpp"
-#include "ck/tensor_operation/gpu/device/device_gemm_xdl_skip_b_lds.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_xdl.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_xdl_skip_b_lds.hpp"
 
 using F16 = ck::half_t;
 using F32 = float;

example/02_gemm_bilinear/gemm_bilinear_xdl_fp16.cpp

@@ -8,7 +8,7 @@
 
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
-#include "ck/tensor_operation/gpu/device/device_gemm_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/utility/device_memory.hpp"

example/03_gemm_bias_relu/gemm_bias_relu_xdl_fp16.cpp

@@ -9,7 +9,7 @@
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
-#include "ck/tensor_operation/gpu/device/device_gemm_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/utility/device_memory.hpp"

example/04_gemm_add_add_fastgelu/common.hpp

@@ -12,7 +12,7 @@
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
-#include "ck/tensor_operation/gpu/device/device_gemm_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 #include "ck/utility/data_type.hpp"
 

example/09_convnd_fwd/convnd_fwd_xdl_bf16.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_fwd_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
 
 #include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
 

example/09_convnd_fwd/convnd_fwd_xdl_fp16.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_fwd_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
 
 #include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
 

example/09_convnd_fwd/convnd_fwd_xdl_fp32.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_fwd_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
 
 #include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
 

example/09_convnd_fwd/convnd_fwd_xdl_fp64.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_fwd_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
 
 #include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
 

example/09_convnd_fwd/convnd_fwd_xdl_int8.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_fwd_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_xdl_cshuffle.hpp"
 
 #include "ck/library/utility/convolution_host_tensor_descriptor_helper.hpp"
 

example/10_convnd_fwd_multiple_d_multiple_reduce/common.hpp

@@ -12,7 +12,7 @@
 #include <vector>
 
 #include "ck/ck.hpp"
-#include "ck/tensor_operation/gpu/device/device_grouped_conv_fwd_multiple_d_multiple_r_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_grouped_conv_fwd_multiple_d_multiple_r_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"