* add double grad compute for batch norm,test=develop

* fix unittest, test=develop

* remove unuse tensor,test=develop

* add format,test=develop

* update, test=develop

* fix bug sequececonv_eltadd_relu_fuse_pass, output error when sequence_conv's padding_start > 0

* fix seqconv_eltadd_relu_fuse_pass unitest error

* support use add instead of sum to do gradient accumulation

* add inplace addto pass

* add grad_add op and inplace addto pass

* remove debug code

* code refine

* fix bug when sereral sum ops inserts at same op_idx

* fix Flags type

* add addto attribute for conv3d

* fix ut

* code clean

* fix type

Fix bug: shapes of Topk outputs are wrong when the parameter k is Tensor 

* Finished ChannelWiseQuantDequantAbsMaxOp and Passed unittests.

* Finished channel-wise quantize strategy in imperative quantization.

* Added Cuda code of ChannelWiseQuantDequantMaxAbsOP
Add Cuda code of ChannelWiseQuantDequantMaxAbsOp

* Add quant_axis for channel_wise quant.

* fixed a bug in unnitests, which will not trigger axis = 1 case and cannot meet the coverage rate requirement.

* Added some assert infomation and fixed some coding style mistakes.

* refine error msg in var_handle.h, test=develop

* refine all_reduce_op_handle

* fix some error msg

* refine variable_visitor

* refine threaded_ssa_graph_executor

* refine inplace related files

* refine executor related files

* refine fetch_op_handle.cc

* fix bug

* follow comments

* fix adam

* rmsprop support double

* fix worker endpoints

* fix gloo wrapper for hdfs

* GPU fleetrun support gloo

* parameterserver fleetrun support gloo

* fix get server endpoint

* Support python3.8

test=notest

* pass compatibility

* add compatibility registry

* add unittests for different padding

* add assert

* drop errmsg

* fix cudnn dyload error

https://github.com/PaddlePaddle/Paddle/pull/27332

* fix op print error info problem. test=develop

* fix build error

* fix format

* fix error msg info

* fix format

and optimize the c++ implmentation of empty op as PR#26659 reviews,
and add bool for shape op.

paddle/fluid/operators/distributed_ops OP报错信息优化

* Fix install pr requirements.txt

* test=document_fix

enhance reduce op which can reduce tensor with arbitrary rank 

* fix floordiv

* add adaptivelsgd

* Todo fix the code to avoid the conflict.

Error description optimize for the math dir

fix judge cache file of inference api more accurate

test=develop

* fix error message

* polish framework error message part 7

* fix typo

* polish by reviewes comment

update the error message check for the some ops

add the error message check for the some operator

* optimize slice TRT plugin

This patch removes unnecessary barrier for data transfer of needed offset,
so data transfer can be overlap with GPU kernel execution.

This patch also fixes incorrect name of slice plugin. That is, replaces
"layernorm" with "slice"

test=develop

* add serialize/deserialize to slice plugin

* add static shape slice trt plugin

* fix slice trt op convertor dynamic shape bug

* fix format by clang-format

* fix pylint format error

* fix problems commented by peiyang
Co-authored-by: NRyan Jeng <rjeng@nvidia.com>