Merge remote-tracking branch 'origin/develop' into bwroblew/warp_wise_dpp8

03cd2692 · Bartlomiej Wroblewski · bf445c31 · f5ec04f0 · 03cd2692 · 03cd2692
Commit 03cd2692 authored Aug 31, 2023 by Bartlomiej Wroblewski
20 changed files
--- a/library/src/tensor_operation_instance/gpu/grouped_gemm_bias/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/grouped_gemm_bias/CMakeLists.txt
+add_instance_library(device_grouped_gemm_bias_instance
+   device_grouped_gemm_xdl_fixed_nk_bias_f16_f16_f16_mk_kn_mn_instance.cpp
+   device_grouped_gemm_xdl_fixed_nk_bias_f16_f16_f16_mk_nk_mn_instance.cpp
+   device_grouped_gemm_xdl_fixed_nk_bias_f16_f16_f32_mk_kn_mn_instance.cpp
+   device_grouped_gemm_xdl_fixed_nk_bias_f16_f16_f32_mk_nk_mn_instance.cpp
+)
--- a/library/src/tensor_operation_instance/gpu/grouped_gemm_bias/device_grouped_gemm_xdl_fixed_nk_bias_f16_f16_f16_mk_kn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_gemm_bias/device_grouped_gemm_xdl_fixed_nk_bias_f16_f16_f16_mk_kn_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_gemm_bias/device_grouped_gemm_xdl_fixed_nk_bias_f16_f16_f16_mk_nk_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_gemm_bias/device_grouped_gemm_xdl_fixed_nk_bias_f16_f16_f16_mk_nk_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_gemm_bias/device_grouped_gemm_xdl_fixed_nk_bias_f16_f16_f32_mk_kn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_gemm_bias/device_grouped_gemm_xdl_fixed_nk_bias_f16_f16_f32_mk_kn_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/grouped_gemm_bias/device_grouped_gemm_xdl_fixed_nk_bias_f16_f16_f32_mk_nk_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/grouped_gemm_bias/device_grouped_gemm_xdl_fixed_nk_bias_f16_f16_f32_mk_nk_mn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/max_pool_bwd/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/max_pool_bwd/CMakeLists.txt
--- a/library/src/tensor_operation_instance/gpu/max_pool_bwd/device_max_pool_bwd_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/max_pool_bwd/device_max_pool_bwd_bf16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/max_pool_bwd/device_max_pool_bwd_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/max_pool_bwd/device_max_pool_bwd_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/max_pool_bwd/device_max_pool_bwd_f32_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/max_pool_bwd/device_max_pool_bwd_f32_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/max_pool_bwd/max_pool_bwd_instance_common.hpp
+++ b/library/src/tensor_operation_instance/gpu/max_pool_bwd/max_pool_bwd_instance_common.hpp
--- a/library/src/tensor_operation_instance/gpu/pool3d_fwd/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/pool3d_fwd/CMakeLists.txt
--- a/library/src/tensor_operation_instance/gpu/pool3d_fwd/device_avg_pool3d_fwd_ndhwc_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/pool3d_fwd/device_avg_pool3d_fwd_ndhwc_bf16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/pool3d_fwd/device_max_pool3d_fwd_ndhwc_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/pool3d_fwd/device_max_pool3d_fwd_ndhwc_bf16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/pool3d_fwd/pool_fwd_instance_common.hpp
+++ b/library/src/tensor_operation_instance/gpu/pool3d_fwd/pool_fwd_instance_common.hpp
--- a/library/src/utility/device_memory.cpp
+++ b/library/src/utility/device_memory.cpp
--- a/profiler/include/profiler/profile_avg_pool3d_bwd_impl.hpp
+++ b/profiler/include/profiler/profile_avg_pool3d_bwd_impl.hpp
--- a/profiler/include/profiler/profile_gemm_multiply_add_impl.hpp
+++ b/profiler/include/profiler/profile_gemm_multiply_add_impl.hpp
--- a/profiler/include/profiler/profile_max_pool3d_bwd_impl.hpp
+++ b/profiler/include/profiler/profile_max_pool3d_bwd_impl.hpp
--- a/profiler/src/CMakeLists.txt
+++ b/profiler/src/CMakeLists.txt
--- a/profiler/src/profile_avg_pool3d_bwd.cpp
+++ b/profiler/src/profile_avg_pool3d_bwd.cpp