You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

wanghaox 0968c7cd6b Update code and fix conflicts.		7 years ago
..
details	Extract DevPtrCast to device_ptr_cast.h	8 years ago
dynload	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into cmake_speed	7 years ago
.clang-format	Use Google Style in new directories	8 years ago
CMakeLists.txt	remove unused code (#5219)	7 years ago
assert.h	Make paddle/framework buildable and passed	8 years ago
call_once.h	Fix GPU Compile on Linux	7 years ago
cpu_info.cc	Follow comments	8 years ago
cpu_info.h	ENH: Add buddy allocator Free	8 years ago
cpu_info_test.cc	ENH: Polish cpu info interface	8 years ago
cuda_helper.h	Refine accuracy_op CUDA kernel (#4097)	8 years ago
cudnn_helper.h	Cudnn batch norm op (#5067)	7 years ago
cudnn_helper_test.cc	Remove cudnn_helper.cc	8 years ago
device_context.cc	Check errors for the cuda kernel calls. (#5436)	7 years ago
device_context.h	Check errors for the cuda kernel calls. (#5436)	7 years ago
device_context_test.cc	Rename platform::GetDeviceCount into platform::GetCUDADeviceCount	7 years ago
enforce.h	Feature/nccl dso (#5001)	7 years ago
enforce_test.cc	Use PADDLE_WITH_CUDA instead of PADDLE_WITH_GPU	7 years ago
gpu_info.cc	"fix accuracy kernel bug" (#5673)	7 years ago
gpu_info.h	"fix accuracy kernel bug" (#5673)	7 years ago
hostdevice.h	fix framework::LoDTensor => Tensor	7 years ago
macros.h	format	7 years ago
nccl_test.cu	checkin nccl operator	7 years ago
place.cc	Add Skeleton of Double support	7 years ago
place.h	Feature/nccl dso (#5001)	7 years ago
place_test.cc	add paddle/memory/detail/cpu_allocator*	8 years ago
transform.h	CompareOp's kernel device type is decided by input tensor place	7 years ago
transform_test.cu	Refine platform::Transform function and fix prelu_op testing.	8 years ago
variant.h	Use PADDLE_WITH_CUDA instead of PADDLE_WITH_GPU	7 years ago