merge develop to attn-train-develop-qloop

76f2b6cd · danyao12 · 9b4c780a · 1ee99dca · 76f2b6cd · 76f2b6cd
Commit 76f2b6cd authored Jul 14, 2023 by danyao12
20 changed files
--- a/client_example/15_convnd_bwd_data/conv3d_bwd_data_fp32.cpp
+++ b/client_example/15_convnd_bwd_data/conv3d_bwd_data_fp32.cpp
--- a/client_example/15_gemm_add_multiply/CMakeLists.txt
+++ b/client_example/15_gemm_add_multiply/CMakeLists.txt
+add_executable(client_gemm_add_multiply gemm_add_multiply.cpp)
+target_link_libraries(client_gemm_add_multiply PRIVATE composable_kernel::device_operations)
\ No newline at end of file
--- a/client_example/15_gemm_add_multiply/gemm_add_multiply.cpp
+++ b/client_example/15_gemm_add_multiply/gemm_add_multiply.cpp
--- a/client_example/15_reduce/CMakeLists.txt
+++ b/client_example/15_reduce/CMakeLists.txt
+add_executable(client_reduce_nhwc_c reduce_nhwc_c.cpp)
+target_link_libraries(client_reduce_nhwc_c PRIVATE composable_kernel::device_operations)
--- a/client_example/15_reduce/reduce_nhwc_c.cpp
+++ b/client_example/15_reduce/reduce_nhwc_c.cpp
--- a/client_example/16_convnd_fwd/CMakeLists.txt
+++ b/client_example/16_convnd_fwd/CMakeLists.txt
+add_executable(client_conv3d_fwd_fp16 conv3d_fwd_fp16.cpp)
+add_executable(client_conv3d_fwd_fp32 conv3d_fwd_fp32.cpp)
+target_link_libraries(client_conv3d_fwd_fp16 PRIVATE composable_kernel::device_operations)
+target_link_libraries(client_conv3d_fwd_fp32 PRIVATE composable_kernel::device_operations)
--- a/client_example/16_convnd_fwd/common.hpp
+++ b/client_example/16_convnd_fwd/common.hpp
--- a/client_example/16_convnd_fwd/conv3d_fwd_fp16.cpp
+++ b/client_example/16_convnd_fwd/conv3d_fwd_fp16.cpp
--- a/client_example/16_convnd_fwd/conv3d_fwd_fp32.cpp
+++ b/client_example/16_convnd_fwd/conv3d_fwd_fp32.cpp
--- a/client_example/17_grouped_gemm_fastgelu/CMakeLists.txt
+++ b/client_example/17_grouped_gemm_fastgelu/CMakeLists.txt
+add_executable(client_grouped_gemm_fastgelu grouped_gemm_fastgelu.cpp)
+target_link_libraries(client_grouped_gemm_fastgelu PRIVATE composable_kernel::device_operations)
\ No newline at end of file
--- a/client_example/17_grouped_gemm_fastgelu/grouped_gemm_fastgelu.cpp
+++ b/client_example/17_grouped_gemm_fastgelu/grouped_gemm_fastgelu.cpp
--- a/client_example/18_groupnorm/CMakeLists.txt
+++ b/client_example/18_groupnorm/CMakeLists.txt
+add_executable(client_groupnorm_swish groupnorm_swish.cpp)
+target_link_libraries(client_groupnorm_swish PRIVATE composable_kernel::device_operations)
--- a/client_example/18_groupnorm/groupnorm_swish.cpp
+++ b/client_example/18_groupnorm/groupnorm_swish.cpp
--- a/client_example/19_pool_fwd/CMakeLists.txt
+++ b/client_example/19_pool_fwd/CMakeLists.txt
+add_executable(client_max_pool2d_fwd max_pool2d_fwd.cpp)
+target_link_libraries(client_max_pool2d_fwd PRIVATE composable_kernel::device_operations)
+add_executable(client_avg_pool3d_fwd avg_pool3d_fwd.cpp)
+target_link_libraries(client_avg_pool3d_fwd PRIVATE composable_kernel::device_operations)
\ No newline at end of file
--- a/client_example/19_pool_fwd/avg_pool3d_fwd.cpp
+++ b/client_example/19_pool_fwd/avg_pool3d_fwd.cpp
--- a/client_example/19_pool_fwd/max_pool2d_fwd.cpp
+++ b/client_example/19_pool_fwd/max_pool2d_fwd.cpp
--- a/cmake/EnableCompilerWarnings.cmake
+++ b/cmake/EnableCompilerWarnings.cmake
--- a/cmake/googletest.cmake
+++ b/cmake/googletest.cmake
--- a/doc/markdown/dockerhub.md
+++ b/doc/markdown/dockerhub.md
--- a/docs/API_Reference_Guide.rst
+++ b/docs/API_Reference_Guide.rst