merge upstream

4525c5d7 · coderfeli · a8d88d8d · 44828b7c · 4525c5d7 · 4525c5d7
Commit 4525c5d7 authored Dec 02, 2024 by coderfeli
20 changed files
--- a/library/src/tensor_operation_instance/gpu/grouped_gemm/device_grouped_gemm_xdl_splitk_f16_f16_f16_mk_nk_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_gemm/device_grouped_gemm_xdl_splitk_f16_f16_f16_mk_nk_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_gemm/device_grouped_gemm_xdl_splitk_f16_f16_f16_mk_nk_mn_irregular_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_gemm/device_grouped_gemm_xdl_splitk_f16_f16_f16_mk_nk_mn_irregular_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_gemm_tile_loop/device_grouped_gemm_xdl_tile_loop_multiply_bf16_i8_bf16_mk_kn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_gemm_tile_loop/device_grouped_gemm_xdl_tile_loop_multiply_bf16_i8_bf16_mk_kn_mn_instance.cpp
--- a/modified_files.txt
+++ b/modified_files.txt
--- a/profiler/include/profiler/profile_grouped_gemm_impl.hpp
+++ b/profiler/include/profiler/profile_grouped_gemm_impl.hpp
--- a/profiler/include/profiler/profile_grouped_gemm_multiply_tile_loop_impl.hpp
+++ b/profiler/include/profiler/profile_grouped_gemm_multiply_tile_loop_impl.hpp
--- a/profiler/include/profiler/profile_grouped_gemm_tile_loop_impl.hpp
+++ b/profiler/include/profiler/profile_grouped_gemm_tile_loop_impl.hpp
--- a/profiler/include/profiler/profile_grouped_gemm_two_stage_impl.hpp
+++ b/profiler/include/profiler/profile_grouped_gemm_two_stage_impl.hpp
--- a/profiler/src/CMakeLists.txt
+++ b/profiler/src/CMakeLists.txt
--- a/profiler/src/profile_gemm_universal_streamk.cpp
+++ b/profiler/src/profile_gemm_universal_streamk.cpp
--- a/profiler/src/profile_grouped_gemm.cpp
+++ b/profiler/src/profile_grouped_gemm.cpp
--- a/profiler/src/profile_grouped_gemm_fixed_nk.cpp
+++ b/profiler/src/profile_grouped_gemm_fixed_nk.cpp
--- a/profiler/src/profile_grouped_gemm_two_stage.cpp
+++ b/profiler/src/profile_grouped_gemm_two_stage.cpp
--- a/python/ck4inductor/batched_universal_gemm/gen_instances.py
+++ b/python/ck4inductor/batched_universal_gemm/gen_instances.py
--- a/python/ck4inductor/batched_universal_gemm/op.py
+++ b/python/ck4inductor/batched_universal_gemm/op.py
--- a/python/ck4inductor/grouped_conv_fwd/gen_instances.py
+++ b/python/ck4inductor/grouped_conv_fwd/gen_instances.py
--- a/test/ck_tile/CMakeLists.txt
+++ b/test/ck_tile/CMakeLists.txt
--- a/test/ck_tile/batched_gemm/CMakeLists.txt
+++ b/test/ck_tile/batched_gemm/CMakeLists.txt
--- a/test/ck_tile/batched_gemm/test_batched_gemm.cpp
+++ b/test/ck_tile/batched_gemm/test_batched_gemm.cpp
--- a/test/ck_tile/batched_gemm/test_batched_gemm_ut_cases.inc
+++ b/test/ck_tile/batched_gemm/test_batched_gemm_ut_cases.inc