Merge branch 'develop' into ab_blk_copy_multi_d

library/src/tensor_operation_instance/gpu/gemm_fastgelu/CMakeLists.txt

-if(DTYPES MATCHES "fp16" OR NOT DEFINED DTYPES)
 add_instance_library(device_gemm_fastgelu_instance
    device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_kn_mn_instance.cpp
    device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_km_nk_mn_instance.cpp
    device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_kn_mn_instance.cpp
    device_gemm_fastgelu_xdl_c_shuffle_f16_f16_f16_mk_nk_mn_instance.cpp
 )
-endif()

library/src/tensor_operation_instance/gpu/gemm_multiply_add/CMakeLists.txt

 set(GEMM_MULTIPLY_ADD_INSTANCES)
-
-if(DTYPES MATCHES "fp16" OR NOT DEFINED DTYPES)
-  list(APPEND GEMM_MULTIPLY_ADD_INSTANCES device_gemm_multiply_add_xdl_c_shuffle_f16_f16_f16_f16_f16_mk_kn_mn_mn_mn_instance.cpp)
-  list(APPEND GEMM_MULTIPLY_ADD_INSTANCES device_gemm_multiply_add_xdl_c_shuffle_f16_f16_f16_f16_f16_mk_nk_mn_mn_mn_instance.cpp)
-endif()
-
-if((DTYPES MATCHES "fp16" AND DTYPES MATCHES "fp8") OR NOT DEFINED DTYPES)
-  list(APPEND GEMM_MULTIPLY_ADD_INSTANCES device_gemm_multiply_add_xdl_c_shuffle_f16_f8_f32_f32_f16_mk_kn_mn_mn_mn_instance.cpp)
-  list(APPEND GEMM_MULTIPLY_ADD_INSTANCES device_gemm_multiply_add_xdl_c_shuffle_f16_f8_f32_f32_f16_mk_nk_mn_mn_mn_instance.cpp)
-endif()
-
+list(APPEND GEMM_MULTIPLY_ADD_INSTANCES device_gemm_multiply_add_xdl_c_shuffle_f16_f16_f16_f16_f16_mk_kn_mn_mn_mn_instance.cpp
+                                        device_gemm_multiply_add_xdl_c_shuffle_f16_f16_f16_f16_f16_mk_nk_mn_mn_mn_instance.cpp
+                                        device_gemm_multiply_add_xdl_c_shuffle_f16_f8_f32_f32_f16_mk_kn_mn_mn_mn_instance.cpp
+                                        device_gemm_multiply_add_xdl_c_shuffle_f16_f8_f32_f32_f16_mk_nk_mn_mn_mn_instance.cpp)
 add_instance_library(device_gemm_multiply_add_instance ${GEMM_MULTIPLY_ADD_INSTANCES})

library/src/tensor_operation_instance/gpu/gemm_splitk/CMakeLists.txt

 set(GEMM_SPLITK_INSTANCES)
 
-if(DTYPES MATCHES "fp32" OR NOT DEFINED DTYPES)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f32_f32_f32_mk_kn_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f32_f32_f32_mk_nk_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f32_f32_f32_km_kn_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f32_f32_f32_km_nk_mn_instance.cpp)
-endif()
-
-if(DTYPES MATCHES "fp16" OR NOT DEFINED DTYPES)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f16_f16_f16_mk_kn_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f16_f16_f16_mk_nk_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f16_f16_f16_km_kn_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f16_f16_f16_km_nk_mn_instance.cpp)
-endif()
-
-if((DTYPES MATCHES "fp16" AND DTYPES MATCHES "fp8") OR NOT DEFINED DTYPES)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f8_f16_f16_mk_kn_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f8_f16_f16_mk_nk_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f8_f16_f16_km_kn_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f8_f16_f16_km_nk_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f16_f8_f16_mk_kn_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f16_f8_f16_mk_nk_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f16_f8_f16_km_kn_mn_instance.cpp)
-  list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f16_f8_f16_km_nk_mn_instance.cpp)
-endif()
+list(APPEND GEMM_SPLITK_INSTANCES device_gemm_xdl_splitk_f32_f32_f32_mk_kn_mn_instance.cpp
+                                  device_gemm_xdl_splitk_f32_f32_f32_mk_nk_mn_instance.cpp
+                                  device_gemm_xdl_splitk_f32_f32_f32_km_kn_mn_instance.cpp
+                                  device_gemm_xdl_splitk_f32_f32_f32_km_nk_mn_instance.cpp
+                                  device_gemm_xdl_splitk_f16_f16_f16_mk_kn_mn_instance.cpp
+                                  device_gemm_xdl_splitk_f16_f16_f16_mk_nk_mn_instance.cpp
+                                  device_gemm_xdl_splitk_f16_f16_f16_km_kn_mn_instance.cpp
+                                  device_gemm_xdl_splitk_f16_f16_f16_km_nk_mn_instance.cpp
+                                  device_gemm_xdl_splitk_fp8_f16_f16_mk_kn_mn_instance.cpp
+                                  device_gemm_xdl_splitk_fp8_f16_f16_mk_nk_mn_instance.cpp
+                                  device_gemm_xdl_splitk_fp8_f16_f16_km_kn_mn_instance.cpp
+                                  device_gemm_xdl_splitk_fp8_f16_f16_km_nk_mn_instance.cpp
+                                  device_gemm_xdl_splitk_f16_fp8_f16_mk_kn_mn_instance.cpp
+                                  device_gemm_xdl_splitk_f16_fp8_f16_mk_nk_mn_instance.cpp
+                                  device_gemm_xdl_splitk_f16_fp8_f16_km_kn_mn_instance.cpp
+                                  device_gemm_xdl_splitk_f16_fp8_f16_km_nk_mn_instance.cpp)
 
 add_instance_library(device_gemm_splitk_instance ${GEMM_SPLITK_INSTANCES})

library/src/tensor_operation_instance/gpu/gemm_streamk/CMakeLists.txt

-if(DTYPES MATCHES "fp16" OR NOT DEFINED DTYPES)
 add_instance_library(device_gemm_streamk_instance
    # device_gemm_xdl_streamk_f32_f32_f32_mk_kn_mn_instance.cpp
    # device_gemm_xdl_streamk_f32_f32_f32_mk_nk_mn_instance.cpp
@@ -9,4 +8,3 @@ add_instance_library(device_gemm_streamk_instance
    # device_gemm_xdl_streamk_f16_f16_f16_km_kn_mn_instance.cpp
    # device_gemm_xdl_streamk_f16_f16_f16_km_nk_mn_instance.cpp
 )
-endif()

library/src/tensor_operation_instance/gpu/grouped_conv1d_fwd/CMakeLists.txt

 add_instance_library(device_grouped_conv1d_fwd_instance
-   device_grouped_conv1d_fwd_xdl_gnwc_gkxc_gnwk_bf16_instance.cpp
-   device_grouped_conv1d_fwd_xdl_gnwc_gkxc_gnwk_f16_instance.cpp
-   device_grouped_conv1d_fwd_xdl_gnwc_gkxc_gnwk_f32_instance.cpp
-   device_grouped_conv1d_fwd_xdl_gnwc_gkxc_gnwk_int8_instance.cpp
+   xdl/device_grouped_conv1d_fwd_xdl_gnwc_gkxc_gnwk_bf16_instance.cpp
+   xdl/device_grouped_conv1d_fwd_xdl_gnwc_gkxc_gnwk_f16_instance.cpp
+   xdl/device_grouped_conv1d_fwd_xdl_gnwc_gkxc_gnwk_f32_instance.cpp
+   xdl/device_grouped_conv1d_fwd_xdl_gnwc_gkxc_gnwk_int8_instance.cpp
 )

library/src/tensor_operation_instance/gpu/grouped_conv2d_fwd/CMakeLists.txt

 add_instance_library(device_grouped_conv2d_fwd_instance
    #xdl
    # GNHWC, GKYXC, GNHWK
-   device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp
-   device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
-   device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
+   xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_bf16_instance.cpp
+   xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
+   xdl/device_grouped_conv2d_fwd_xdl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
    # NHWGC, GKYXC, NHWGK
-   device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
-   device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_instance.cpp
-   device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_instance.cpp
+   xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_bf16_instance.cpp
+   xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f16_instance.cpp
+   xdl/device_grouped_conv2d_fwd_xdl_nhwgc_gkyxc_nhwgk_f32_instance.cpp
    #dl
    # GNHWC, GKYXC, GNHWK
-   device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
-   device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
+   dl/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f16_instance.cpp
+   dl/device_grouped_conv2d_fwd_dl_gnhwc_gkyxc_gnhwk_f32_instance.cpp
+   # NHWGC, GKYXC, NHWGK
+   dl/device_grouped_conv2d_fwd_dl_nhwgc_gkyxc_nhwgk_f16_instance.cpp
+   dl/device_grouped_conv2d_fwd_dl_nhwgc_gkyxc_nhwgk_f32_instance.cpp
    # WMMA
-   device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_instance.cpp
-   device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_instance.cpp
+   # GNHWC, GKYXC, GNHWK
+   wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_1x1p0_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_1x1p0_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_1x1s1p0_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_1x1s1p0_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_f16_oddc_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_gnhwc_gkyxc_gnhwk_i8_oddc_instance.cpp
    # NHWGC, GKYXC, NHWGK
-   device_grouped_conv2d_fwd_dl_nhwgc_gkyxc_nhwgk_f16_instance.cpp
-   device_grouped_conv2d_fwd_dl_nhwgc_gkyxc_nhwgk_f32_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_1x1p0_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_1x1p0_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_1x1s1p0_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_1x1s1p0_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_f16_oddc_instance.cpp
+   wmma/device_grouped_conv2d_fwd_wmma_nhwgc_gkyxc_nhwgk_i8_oddc_instance.cpp
 )