Merge commit '30480288' into conv_quant_int8

example/17_convnd_bwd_data/convnd_bwd_data_xdl_fp16.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_bwd_data_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_convnd_bwd_data_nwc_kxc_nwk_xdl.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_convnd_bwd_data_nwc_kxc_nwk_xdl.hpp"
 
 using InDataType  = ck::half_t;
 using WeiDataType = ck::half_t;

example/18_batched_gemm_reduce/batched_gemm_reduce_xdl_fp16.cpp

@@ -9,7 +9,7 @@
 #include "ck/ck.hpp"
 #include "ck/utility/reduction_operator.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/device_batched_gemm_reduce_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_batched_gemm_reduce_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 
 #include "ck/library/utility/check_err.hpp"

example/19_binary_elementwise/broadcast_add_2d_amn_bn.cpp

@@ -6,7 +6,7 @@
 
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/element/binary_element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/device_elementwise.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_elementwise.hpp"
 
 #include "ck/library/utility/check_err.hpp"
 #include "ck/library/utility/device_memory.hpp"

example/19_binary_elementwise/broadcast_add_3d_am_bmnk.cpp

@@ -6,7 +6,7 @@
 
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/element/binary_element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/device_elementwise.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_elementwise.hpp"
 
 #include "ck/library/utility/check_err.hpp"
 #include "ck/library/utility/device_memory.hpp"

example/19_binary_elementwise/elementwise_add_1d.cpp

@@ -5,7 +5,7 @@
 #include <cstdlib>
 
 #include "ck/ck.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/binary_element_wise_operation.hpp"
 #include "ck/library/utility/check_err.hpp"
 #include "ck/library/utility/device_memory.hpp"

example/19_binary_elementwise/elementwise_add_4d.cpp

@@ -6,7 +6,7 @@
 
 #include "ck/ck.hpp"
 #include "ck/tensor_operation/gpu/element/binary_element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/device_elementwise.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_elementwise.hpp"
 
 #include "ck/library/utility/check_err.hpp"
 #include "ck/library/utility/device_memory.hpp"

example/20_convnd_bwd_weight/convnd_bwd_weight_xdl_bf16.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_bwd_weight_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_convnd_bwd_weight_nwc_kxc_nwk_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_convnd_bwd_weight_nwc_kxc_nwk_xdl_cshuffle.hpp"
 
 using InDataType = ck::bhalf_t;
 // bf16 kernel use fp32 atomic add to accumulate Weight tensor into global memory

example/20_convnd_bwd_weight/convnd_bwd_weight_xdl_fp16.cpp

@@ -3,7 +3,7 @@
 
 #include "convnd_bwd_weight_common.hpp"
 
-#include "ck/tensor_operation/gpu/device/device_convnd_bwd_weight_nwc_kxc_nwk_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_convnd_bwd_weight_nwc_kxc_nwk_xdl_cshuffle.hpp"
 
 using InDataType  = ck::half_t;
 using WeiDataType = ck::half_t;

example/21_gemm_layernorm/gemm_bias_relu_add_layernorm_xdl_fp16.cpp

@@ -9,8 +9,8 @@
 #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_multiple_r_xdl_cshuffle.hpp"
-#include "ck/tensor_operation/gpu/device/device_elementwise.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_multiple_r_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_elementwise.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/utility/device_memory.hpp"

example/21_gemm_layernorm/gemm_layernorm_xdl_fp16.cpp

@@ -9,8 +9,8 @@
 #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_multiple_r_xdl_cshuffle.hpp"
-#include "ck/tensor_operation/gpu/device/device_elementwise.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_multiple_r_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_elementwise.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/utility/device_memory.hpp"

example/21_gemm_layernorm/gemm_xdl_layernorm_single_kernel_fp16.cpp

@@ -11,7 +11,7 @@
 #include "ck/library/utility/host_tensor.hpp"
 #include "ck/library/utility/host_tensor_generator.hpp"
 #include "ck/tensor_operation/gpu/device/tensor_layout.hpp"
-#include "ck/tensor_operation/gpu/device/device_gemm_xdl_layernorm_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_gemm_xdl_layernorm_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 #include "ck/utility/reduction_operator.hpp"
 #include "ck/library/reference_tensor_operation/cpu/reference_gemm_layernorm.hpp"

example/22_cgemm/cgemm_xdl_bf16.cpp

@@ -8,7 +8,7 @@
 #include "ck/library/reference_tensor_operation/cpu/reference_cgemm.hpp"
 
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/device_cgemm_4gemm_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_cgemm_4gemm_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 
 using ADataType   = BF16;

example/22_cgemm/cgemm_xdl_fp16.cpp

@@ -6,7 +6,7 @@
 #include "cgemm_xdl_common.hpp"
 #include "ck/library/reference_tensor_operation/cpu/reference_cgemm.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/device_cgemm_4gemm_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_cgemm_4gemm_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 
 using ADataType        = F16;

example/22_cgemm/cgemm_xdl_fp32.cpp

@@ -8,7 +8,7 @@
 #include "ck/library/reference_tensor_operation/cpu/reference_cgemm.hpp"
 
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/device_cgemm_4gemm_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_cgemm_4gemm_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 
 using ADataType   = F32;

example/22_cgemm/cgemm_xdl_int4.cpp

@@ -8,7 +8,7 @@
 #include "ck/library/reference_tensor_operation/cpu/reference_cgemm.hpp"
 
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/device_cgemm_4gemm_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_cgemm_4gemm_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 
 using ADataType        = INT4;

example/22_cgemm/cgemm_xdl_int8.cpp

@@ -8,7 +8,7 @@
 #include "ck/library/reference_tensor_operation/cpu/reference_cgemm.hpp"
 
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
-#include "ck/tensor_operation/gpu/device/device_cgemm_4gemm_xdl_cshuffle.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_cgemm_4gemm_xdl_cshuffle.hpp"
 #include "ck/tensor_operation/gpu/device/gemm_specialization.hpp"
 
 using ADataType   = INT8;

example/24_batched_gemm/batched_gemm_xdl_bfp16.cpp

@@ -6,7 +6,7 @@
 #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_batched_gemm_multi_d_xdl.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_batched_gemm_multi_d_xdl.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/utility/check_err.hpp"

example/24_batched_gemm/batched_gemm_xdl_fp16.cpp

@@ -6,7 +6,7 @@
 #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_batched_gemm_multi_d_xdl.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_batched_gemm_multi_d_xdl.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/utility/check_err.hpp"

example/24_batched_gemm/batched_gemm_xdl_fp32.cpp

@@ -6,7 +6,7 @@
 #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_batched_gemm_multi_d_xdl.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_batched_gemm_multi_d_xdl.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/utility/check_err.hpp"

example/24_batched_gemm/batched_gemm_xdl_int4.cpp

@@ -6,7 +6,7 @@
 #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_batched_gemm_multi_d_xdl.hpp"
+#include "ck/tensor_operation/gpu/device/impl/device_batched_gemm_multi_d_xdl.hpp"
 #include "ck/tensor_operation/gpu/element/element_wise_operation.hpp"
 
 #include "ck/library/utility/check_err.hpp"