* Add dlnne engine runtime

* Fix log

* Remove <const_cast> and remove unrelated modify with dlnne, +clang-format

* Fix CMakeList format error

* Add copyright message

* Fix dlnne CMakeList.txt

* Add some paddlepaddle_pass to support more networks

* Fix some format bug

* Add delete dropout_op pass

* Fix some format bug

* Fix format bug

* Revert "Revert "[PsCore] optimize performance of large kv (#32535)" (#32599)"

This reverts commit 809ac036.

* brpc dep

* fix two error message

* fix two error message

* fix error

* fix error

* fix error

* fix error

* fix some error message

* fix some error

* fix error

* fix some error

* fix some error

* fix some error

* fix one error

* fix some error

* fix seven error message

* fix error

* fix error

* fix error

* fix error

* fix some error message

* fix error

* fix some error

* fix some error

Reduce the time cost for the elementwise_add test case (#32628)

* - Added clearing oneDNN per executor

* - Executor is nt always having FLAGS_use_mkldnn set to true

* optimize update_loss_scaling_op by fused for loop to one kernel, test=develop

* remove useless while loop and optimize variable name, test=develop

* optimize variable name from out_addrs_tensor to out_addrs_mem, test=develop

* optimize variable name for readable by change prefix identifier from t_ to local_

* add alltoall api, test=develop

* support cuda11.2 and using gcc5.4 in cuda10.1

* fix manylinux py36 bug

* support cuda11.2

* fix python36 pip version problem in ubuntu

* save cuda11.0

* update 2.0 public api in nn

* replace Chinese character cause error in ci;synchronization with pr:#32588 to avoid 'ascii' codec in python2

* numbers used in paddle.nn.functional.norm but not imported

Co-authored-by: NXiaoguangHu <46782768+XiaoguangHu01@users.noreply.github.com>

* edit paddle.save/load API

* Update io.py

edit doc

* delete cpython-37.pyc

* Update io.py

edit doc

* Update io.py

recommit

* Update io.py

recommit

* Update io.py

recommit

* Update io.py

recommit

* clear 'BasicEngine' when an exception occurs in the backward.

* deal with conflict.

* deal with conflict.

* [OPs] Bug fix, fix the segment mean for illegal syncthreads usage.

* jit.save/load support function.

* delete unnittest test_jit_load_model_incomplete.

* edit code according to CI

* Modify the documentation.

* add note to doc.

* fixed docs.

* Fixed docs. test=document_fix

code bak.

fixed docs. test=document_fix

* Revert to previous version of python/paddle/fluid/backward.py

* fixed bugs.

* test=document_fix. Fixed examples.

* fix cross_entropy calculation error

* add unittest and fix static

* UnicodeDecodeError: 'ascii' codec can't decode byte 0xe2 in position 1788: ordinal not in range(128)

test=document_fix

str(doc) in python2

test=document_fix

* update md5 function in count_api_without_core_ops.py

str in py2 is different.

test=document_fix

* Support list and tuple for parameters of layer_norm, multiprocess_reader, DatasetFolder and ImageFolder.

* add unittest for layer_norm.

* add require gpu for example.

This reverts commit 4b7242b0.

Co-authored-by: NYang Zhang <yangzhang@live.com>

* fix amp bug

* fix name of wordsize

* add sendrecv, test=develop

* clear CUDA compile environment on windows

* fix Windows CI

* fix Windows CI

* fix Windows CI

* new optimize for where_index_op with prefix sum version.

* write a scan prefix sum kernel with stream for where index op.

* optimize where_index by using cub::DeviceScan::InclusiveSum instead of imperfect self-kernel.

* remove CheckTrue struct and rename stide_array for readable.

* optimize variable name for readable.

* optimize function name and annotation.

* optimize pull sparse

* optimize pull sparse

* change macro

* format

* fix bn docs

* fix unittest

A follow up PR of #32420, we changed the doc of python/paddle/fluid/layers/tensor.py in that PR and we are changing python/paddle/tensor/creation.py in this PR.