* support npu profiler

* add python api

* fix bugs

* add wrapper for incomplete type

* update profile proto

* record npu wait

* add xpu placeholder

* support npu allocator

* add npu device context

* fix some compile problem

* fix some compile problem

* add npu info

* compile ok

* fix include dir

* support naive_best_fit_allocator

* run ut ok, bug failed to exit

* call aclrtResetDevice before exit

* fix aclFinilize

* add system allocatot test

* add selected_gpus in gtest

* add tensor_test for npu

* support npu op, initial commit

* add npu stream

* add elementwise_add_op

* compile ok

* fix typo

* fix elementwise_add_op_npu_test

* support op run

* test can run but failed

* change aclopExecuteV2 to aclopCompileAndExecute

[feature] support npu operator

[feature] support npu allocator

* optimize memcpy perf for kunlun

* remove useless unitest for kunlun mean

* minor

remove eigen threadpool for the speed up

* test=develop, optimize geo communicator 

enhance reduce op which can reduce tensor with arbitrary rank 

* Add mechanism for blocking oneDNN cache clearing

* Review changes and Add thread guards

* support Baidu AI Accelerator
  * test=kunlun

* minor
 * test=kunlun

* support xpu op in separate file
 * test=kunlun

* update XPU error message and remove duplicated code

 * test=kunlun

* minor
 * test=kunlun

* minor
 * test=kunlun

* Refine PADDLE_ENFORCE in paddle/fluid/platform
test=develop

* refine PADDLE_ENFORCE
test=develop

* remove REPLACE_ENFORCE_GLOG compile option & add ci rule prohibit LOG(FATAL) using, test=develop

* remove ci test case, test=develop

* replace all LOG(FATAL) & polish message, test=develop

* fix typo, test=develop

* polish error info detail, test=develop

test=develop

* add new macro BOOST_GET_SAFELY & unittests, test=develop

* add different macro type, test=develop

* fix get macro type in executor, test=develop

* four macro part change backup

* using one macro for all case, test=develop

* revert attribute change, test=develop

* change to three func to solve gcc4.8 bug, test=develop

* polish some details, test=develop

* Add cholesky_op forward part. test=develop

* Complete cholesky_op forward part. test=develop

* Add cholesky_op backward part. test=develop

* Complete cholesky_op backward part. test=develop

* Refine cholesky_op error check and docs. test=develop

* Add grad_check unit test for cholesky_op. test=develop

* Fix sample code in cholesky doc. test=develop

* Refine some error messages of cholesky_op. test=develop

* Refine some error messages of cholesky_op. test=develop

* Remove unused input in cholesky_grad. test=develop

* Remove unused input in cholesky_grad. test=develop

* Fix stream for cusolverDnSetStream. test=develop

* Update PADDLE_ENFORCE_CUDA_SUCCESS from cholesky_op to adapt to latest code.
test=develop

* Add CUSOLVER ERROR in enforce.h
test=develop

* Fix the missing return value in cholesky. test=develop

* update eigen, test=develop

* remove patches, test=develop

* add definition of -fabi-version, test=develop

* add patch for TensorBlock.h, test=develop

* test windows, test=develop

* only update eigen for Linux, test=develop

* add code comments, test=develop

* supports thread-binding stream, test=develop

* avoid using thread_local variables in dtor, test=develop

* modify the stream priority enum, test=develop

As nccl comm is not created by CUDADeviceContext, it should be destroyed by the creator as the best practice of RAII.

cmake选项中添加了WITH_NCCL，显示指定是否编译NCCL的部分代码，WITH_NCCL默认打开，但如果WITH_GPU为OFF，则关闭WITH_NCCL

添加了PADDLE_WITH_NCCL定义

单机单卡能够关闭NCCL编译，多卡的话需要默认打开NCCL，如果关闭NCCL，则只能使用单卡
Co-authored-by: N石晓伟 <39303645+Shixiaowei02@users.noreply.github.com>

stack's wait cost a lot of cpu time, use cuda kernel to do memory copy
will reduce cpu time.
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* MKLDNN v1.0 rebase to Paddle 1.6
test=develop

* Add hacky paddle::string::to_string() implementation

* vectorize<int64-t>() -> vectorize() cleanup
test=develop

* PADDLE_ENFORCE and void_cast fixes
test=develop

* Rebase changes
test=develop

* Cosmetics
test=develop

* Delete MKL from mkldnn.cmake
test=develop

* CMake debug commands
test=develop

* Delete MKLDNN_VERBOSE and rebase fixes
test=develop

* Rebase fixes
test=develop

* Temporarily disable int8 resnet101 vgg16 and vgg19 tests
test=develop

* Add libmkldnn.so.1 to python setup
test=develop

* Add libmkldnn.so.1 to inference_lib cmake after rebase
test=develop

* Post rebase fixes + FC int8 changes
test=develop

* Fix LRN NHWC
test=develop

* Fix NHWC conv3d
test=develop

* Windows build fix + next conv3d fix
test=develop

* Fix conv2d on AVX2 machines
test=develop

cudaStreamSynchronize randomly hang when used in multi-thread environment, replace it with cudaStreamQuery API on windows

* Improve topk performance.

give 200000 data to compute topk,
before opt: cost 1s
after opt: cost 0.0028s.

* Refine return value.
* Add cuda util funtions.
* Fix ComputeBlockSize bug & refine comments.
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* refine reallocate of workspace size, test=develop

* add lock to cudnn handle calls, test=develop

TemporaryAllocator is a singleton used for allocating memory for Cudnn. Since it is a singleton, we can delete it for better performance in memory.

We replace TemporaryAllocator by CUDADeviceContextAllocator and CUDADeviceContextAllocation, which uses stream callback to delete the memory allocated for the stream to avoid singleton.

Also added data_feed_proto to operator to fix CI in CPU compilation

Polish fleet API to support cuda collective mode and nccl2 mode