* fp16 result ok

* change -DWITH_NVINFER_PLUGIN toconfig.EnableTensorRtOSS

* auto detect special slice op converter for ernie with trt oss

* ernie oss only support fp16

* fix special_slice_plugin serialize bug

* matmul in tensorrt ok

* ernie unittest ok

* add matmul tensorrt unittest

* remove demo code

* Add rnn_op.
test=develop

* Fix rnn_op grad maker's drop_empty_grad.
test=develop

This PR is follow up of #28213. On that PR we tried to decrease GPU usage, however the CI still randomly failed. So I added retry logic for the initialization of nccl and cusolver. If the initialization failed, we can retry to avoid the random failure.

* add multiple exception type

* define all exception & polish compile pystack

* mapping paddle error to python exception

* polish static mode error format

* fix failed unittests

* fix dytostatic test_error

* fix check_nan_inf failed

* add unittest for coverage

* revert some code try to solve compile error

* refactor enforce & error change

* polish code & add unittest

* add compile limit for paddle enforce

* polish elementwise_op_function.cu.h

* fix failed unittest

* fix windows compile failed

* detail polish

* revert no type constructor

* conv dilated mkldnn support: forward and backward pass

* add mkldnn conv_transpose dilation UT
test=develop

* remove unnecessary PADDLE_ENFORCE

* add int8 and bf16 dilated conv UT

* update according to reviews

* use exhaustive_search for float16

* tune algo only when dtype is float16

* - condidate fix to issue #25537

test=develop

* - UT for transpose NHWC

test=develop

* Add avx instructions check

* Small fix

* Change function name

* Change uint to unsigned int

* test=develop, optimize geo communicator 

* add gloo initializer, test=develop

* Add support for mkldnn ops types selection with FLAGS in dygraph

* use regex to match DNNL verbose

* python3 encoding fix

* include ncclRecv and ncclSend, test=develop

* add gloo initializer, test=develop

* add float64 input to ctc_loss

* modified error message of  warpctc

* update repo and tag of warpctc

* add test for warpctc with float64 input

* modified warpctc.cmake to make sure build always

* resolved sample code bug of warpctc

* add core.ops in warpctc dygraph

* fix a bug of test

* support elementwise add, activation, matmul on Baidu Kunlun
* test=kunlun

* minor
* test=kunlun

* reconstuct the xpu directory
* test=kunlun

* minor
* test=kunlun

* minor
* test=kunlun

* minor
* test=kunlun

* minor
* test=kunlun

* minor
* test=kunlun

fix cpplint error for the autmic max/min

fix cuda atomic for ARCH<350 for the automic_max

* fix tensorrt 6 build error. test=develop

* fix. test=develop

* bug fix

* test=develop

* use iwyu clean include, test=develop, test=win

* compilation error, test=develop

* fix compilation error2, test=develop

* fix compilation error3, test=develop

* fix compilation error4, test=develop

* fix compilation error5, test=develop

* fix compilation error6, test=develop

* fix compilation error7, test=develop

* fix compilation error8, test=develop

* fix compilation error8, test=develop

* fix compilation error10, test=develop

* fix compilation error11, test=develop

Add GPU Kernels of Segment Ops,  support, sum, max, min, mean

* [bug fix]:Memory increases after adapting the cudnn version to 8

* [bug fix]cudnnGetConvolutionForwardAlgorithm not defined

* polish some lost error msg

* add some math file to white list

* polish detail based reviewer commnet

* 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 cudnn dyload error

enhance reduce op which can reduce tensor with arbitrary rank 

* replace sequence length attr to input