compiled version of cross gpu connection

b74918bc · ThomasNing · 3fcad951 · 1c45ca35 · b74918bc · b74918bc
Commit b74918bc authored Jan 06, 2025 by ThomasNing
20 changed files
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_dl_f32_f32_f32_mk_kn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_dl_f32_f32_f32_mk_kn_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_dl_f32_f32_f32_mk_nk_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_dl_f32_f32_f32_mk_nk_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_bf16_bf16_bf16_km_kn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_bf16_bf16_bf16_km_kn_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_bf16_bf16_bf16_km_nk_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_bf16_bf16_bf16_km_nk_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_bf16_bf16_bf16_mk_kn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_bf16_bf16_bf16_mk_kn_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_bf16_bf16_bf16_mk_nk_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_bf16_bf16_bf16_mk_nk_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_int8_int8_int8_km_kn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_int8_int8_int8_km_kn_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_int8_int8_int8_km_nk_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_int8_int8_int8_km_nk_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_int8_int8_int8_mk_kn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_int8_int8_int8_mk_kn_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_int8_int8_int8_mk_nk_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_wmma_int8_int8_int8_mk_nk_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply/device_gemm_multiply_multiply_xdl_f8_f8_bf16/device_gemm_multiply_multiply_xdl_f8_f8_bf16_mk_nk_mn.hpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply/device_gemm_multiply_multiply_xdl_f8_f8_bf16/device_gemm_multiply_multiply_xdl_f8_f8_bf16_mk_nk_mn.hpp
--- a/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_xdl_universal_f8_f8_bf16/device_gemm_xdl_universal_f8_f8_bf16_mk_kn_mn.hpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_xdl_universal_f8_f8_bf16/device_gemm_xdl_universal_f8_f8_bf16_mk_kn_mn.hpp
--- a/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_xdl_universal_f8_f8_bf16/device_gemm_xdl_universal_f8_f8_bf16_mk_nk_mn.hpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_xdl_universal_f8_f8_bf16/device_gemm_xdl_universal_f8_f8_bf16_mk_nk_mn.hpp
--- a/library/src/tensor_operation_instance/gpu/gemm_universal_batched/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/gemm_universal_batched/CMakeLists.txt
--- a/library/src/tensor_operation_instance/gpu/gemm_universal_batched/device_batched_gemm_xdl_universal_bf16_bf16_bf16/device_batched_gemm_xdl_universal_bf16_bf16_bf16_mk_nk_mn.hpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_universal_batched/device_batched_gemm_xdl_universal_bf16_bf16_bf16/device_batched_gemm_xdl_universal_bf16_bf16_bf16_mk_nk_mn.hpp
--- a/library/src/tensor_operation_instance/gpu/gemm_universal_batched/device_batched_gemm_xdl_universal_bf16_bf16_bf16/device_batched_gemm_xdl_universal_bf16_bf16_bf16_mk_nk_mn_comp_default_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_universal_batched/device_batched_gemm_xdl_universal_bf16_bf16_bf16/device_batched_gemm_xdl_universal_bf16_bf16_bf16_mk_nk_mn_comp_default_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm_universal_batched/device_batched_gemm_xdl_universal_bf16_bf16_bf16/device_batched_gemm_xdl_universal_bf16_bf16_bf16_mk_nk_mn_mem_v1_default_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_universal_batched/device_batched_gemm_xdl_universal_bf16_bf16_bf16/device_batched_gemm_xdl_universal_bf16_bf16_bf16_mk_nk_mn_mem_v1_default_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm_universal_batched/device_batched_gemm_xdl_universal_bf16_bf16_bf16/device_batched_gemm_xdl_universal_bf16_bf16_bf16_mk_nk_mn_mem_v2_default_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_universal_batched/device_batched_gemm_xdl_universal_bf16_bf16_bf16/device_batched_gemm_xdl_universal_bf16_bf16_bf16_mk_nk_mn_mem_v2_default_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm_universal_batched/device_batched_gemm_xdl_universal_f8_f8_bf16/device_batched_gemm_xdl_universal_f8_f8_bf16_mk_nk_mn.hpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_universal_batched/device_batched_gemm_xdl_universal_f8_f8_bf16/device_batched_gemm_xdl_universal_f8_f8_bf16_mk_nk_mn.hpp
--- a/library/src/tensor_operation_instance/gpu/gemm_universal_batched/device_batched_gemm_xdl_universal_f8_f8_bf16/device_batched_gemm_xdl_universal_f8_f8_bf16_mk_nk_mn_comp_default_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_universal_batched/device_batched_gemm_xdl_universal_f8_f8_bf16/device_batched_gemm_xdl_universal_f8_f8_bf16_mk_nk_mn_comp_default_instance.cpp