Merge branch 'develop' into amd-develop

705d5a08 · Jun Liu · d4ad52d6 · 8f84a012 · 705d5a08 · 705d5a08
Commit 705d5a08 authored Sep 08, 2023 by Jun Liu
20 changed files
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv2d_fwd_wmma_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv2d_fwd_wmma_instance.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_dl_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_dl_instance.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_gemm_bias.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_gemm_bias.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/image_to_column.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/image_to_column.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/image_to_column/device_image_to_column_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/image_to_column/device_image_to_column_instance.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/max_pool_bwd.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/max_pool_bwd.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/pool3d_fwd.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/pool3d_fwd.hpp
--- a/library/include/ck/library/utility/check_err.hpp
+++ b/library/include/ck/library/utility/check_err.hpp
--- a/library/include/ck/library/utility/device_memory.hpp
+++ b/library/include/ck/library/utility/device_memory.hpp
@@ -26,7 +26,9 @@ struct DeviceMem
    void* GetDeviceBuffer() const;
    std::size_t GetBufferSize() const;
    void ToDevice(const void* p) const;
+    void ToDevice(const void* p, const std::size_t cpySize) const;
    void FromDevice(void* p) const;
+    void FromDevice(void* p, const std::size_t cpySize) const;
    void SetZero() const;
    template <typename T>
    void SetValue(T x) const;

--- a/library/include/ck/library/utility/host_tensor_generator.hpp
+++ b/library/include/ck/library/utility/host_tensor_generator.hpp
--- a/library/src/tensor_operation_instance/gpu/avg_pool3d_bwd/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/avg_pool3d_bwd/CMakeLists.txt
--- a/library/src/tensor_operation_instance/gpu/avg_pool3d_bwd/avg_pool3d_bwd_ndhwc_instance_common.hpp
+++ b/library/src/tensor_operation_instance/gpu/avg_pool3d_bwd/avg_pool3d_bwd_ndhwc_instance_common.hpp
--- a/library/src/tensor_operation_instance/gpu/avg_pool3d_bwd/device_avg_pool3d_bwd_ndhwc_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/avg_pool3d_bwd/device_avg_pool3d_bwd_ndhwc_bf16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/avg_pool3d_bwd/device_avg_pool3d_bwd_ndhwc_f16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/avg_pool3d_bwd/device_avg_pool3d_bwd_ndhwc_f16_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/avg_pool3d_bwd/device_avg_pool3d_bwd_ndhwc_f32_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/avg_pool3d_bwd/device_avg_pool3d_bwd_ndhwc_f32_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/batched_gemm/device_batched_gemm_xdl_f16_f16_f16_gmk_gkn_gmn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/batched_gemm/device_batched_gemm_xdl_f16_f16_f16_gmk_gkn_gmn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/batched_gemm/device_batched_gemm_xdl_f16_f16_f16_gmk_gnk_gmn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/batched_gemm/device_batched_gemm_xdl_f16_f16_f16_gmk_gnk_gmn_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/gemm/CMakeLists.txt
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_dl_dpp8_f16_f16_f16_km_kn_mn_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_dl_dpp8_f16_f16_f16_km_kn_mn_instance.cpp