* XPU support eager mode

* add unittest for XPU eager mode

* minor bugfix

* minor bugfix, test=kunlun

* correct copyright info

* 1. remove unsed vars/funcs
2. ProcessGroupBKCL inherit from ProcessGroupStream

* bugfix for fp16 in eager mode multi-card, test=kunlun

* rebase & fix a few issues

* use new processgroup interface, test=kunlun

* fix compile issue, test=kunlun

* skip_merge_layernorm

* add UT

* modify comments

* fix insert concat op

* fix fp16 assert

* [Fix][F821] fix TestUnpoolOpException

* fix TestUnpoolOpException

* fix TestUnpool3DOpException

* remove unused variables

* fix the regexp does not match the C++ traceback

* add missing error message for gpu unpool_kernel

* Revert "add missing error message for gpu unpool_kernel"

This reverts commit 17ef7a127e1c3ee00f9102c37ad8cea35953f20c.

* assertion indices_value_error errors are only reported on the CPU

* for test

* run test_exception in dygraph mode

* Get params and grads in cpp to avoid gpu idel time

* Using python param instead of cpp return param to fix test_asp_optimize_dynamic.py

* Get grads from cpp and construct params_grads on python

* Check meta and remove comments

* Analysis API interface for disabling fc passes

* Unit tests corrected

* Python API added

* test runs only when PADDLE_WITH_MKLDNN

* Fc op changed to relu in matmul_op_test

* Disable fc passes in tests where acc drops

* code formating

* Unit test for analysisConf added

* Unit test gpu added

* fc passes disabled when iterations=0 in gru test

* style

* passes disabled when fp32 in gru test

* fc passes disabled in lstm test

* Import from inference, not fluid in doc

* fix flake8 CodeStyle E266

* fix comments

* fix paddle.get_default_dtype 

Chinese and English return values are inconsistent

* fix paddle.matmul 文档评估 #4407

把函数的输出改成正确的

* fix paddle.std文档评估 #4370

增加了一个unbiased=False的代码示例，没有增加numpy,怕引起误会。

* fix paddle.load文档测评 #4455 

只把代码拆分了5段

* try

* try

* try

* Update io.py

* Update io.py

* Update creation.py

* Update creation.py

* [Docs]add name description

* [Docs]fix broadcasting issue
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

* [CodeStyle][py2] remove unnecessary u-prefix in string literal

* `"{}".format(x)` -> `x`

* remove duplicated dtype literals

* revert changes in data_feeder.py

* remove u-prefix in data_feeder

* revert remove duplicated dtype literals in data_feeder

* remove unnecessary convert to str

* for test

* add some comments

* refine comment

* restore a removed str conversion

* re-trigger all ci, empty commit

* fix scale kernel configuration error where input numel is 0

* fix code stype

* add unit test case for scale op when numel of input x is zero

* fix ci codestyle check

* add cpu and gpu unit test case for scale op when numel of input x is zero

* add uninitialized judgment for input of scale

* upgrade scale and slice op convert for Paddle-TensorRT

[Auto Parallel] Sharding Optimization：Partition Algorithm & Stage2 Parameter Bucket communication  (#47180)

* partition param by order

* add logging

* reorder opt

* config

* stage2 bucket

* update unitest

* [Zero-Dim] support input 0D Tensor for sundary api

* fix comment

* softplus+activation

* fc + elementwise_add test refactored

* rename MKLDNN to OneDNN

* fc+activation tests refactored

* remove softplus ut

* whitespace

* whitespace

* codestyle

* codestyle

* add more cases to fc+act

* remove softplus+hard_sigmoid pass

* remove softplus + hard_sigmoid UT

* add approximate for gelu

* swish beta range

* new codestyle

* reduce number of tests

* [Zero-Dim] support input 0D Tensor for distribution api

* fix comment

* fix tensor_array slice bugs in _getitem_impl_

* fix when var is a paddle.Tensor

* code format

* allow fold fill_constant && allow nms3 into trt in int8 model
* use unordered_map
* fix CI failing

* [CodeStyle][py2][U004] unecessary explicit `object` inheritance in class definition

* fix an increment

* Split kernel registered, tests for uint/int added

* Split quantized

* Split output scales calculated only once

* NearestInterp test fix reversed

* DequantizeOutputs corrected

* disable distributed xpu tests

* test=kunlun

* test=document_fix;test=kunlun

* test=document_fix;test=kunlun

* test=document_fix;test=kunlun

* test=document_fix;test=kunlun

* support pow_double_grad op

* add unit test for pow double grad

* fix pow double grad

* optimize pow double grad kernel

* fix pow double grad kernel

* fix cast bug

*test=kunlun

* fp16 pass support assign op

* choose assign op exec mode

* add unittest

* add cmakelist

add roll and roll_grad kernels and strided_slice and strided_slice_grad kernels, test=kunlun (#47368)

* add stat tool

* add roll and roll_grad kernels and strided_slice and strided_slice_grad kernels, test=kunlun