Co-authored-by: Ntianshuo78520a <707759223@qq.com>

* remove -DSUPPORTS_CUDA_FP16 in cuda.cmake

* comile with cuda9

* add some unittest

* notest;test=coverage

* add unittest for trt plugin swish && split

* update ernie unittest

* fix some error message

* remove repeated judgement of CUDA version in mbEltwiseLayerNormOpConverter

* fix comile errror when CUDA_ARCH_NAME < Pascal"

* fix comile error

* update unittest timeout

* compile with cuda9

* update error msg

* fix code style

* add some comments

* add define IF_CUDA_ARCH_SUPPORT_FP16

* rename IF_CUDA_ARCH_SUPPORT_FP16 to CUDA_ARCH_FP16_SUPPORTED

* 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 unittests and op version register for tensorrt_subgraph_pass

* rename to test_trt_subgraph_pass.py

* fix softmax converter diff when padding dim=1

* 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

* polish some lost error msg

* add some math file to white list

* polish detail based reviewer commnet

* 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>

test=develop

* add stack_op to CMakeLists

* add dim=3 support for scale op

* add trt stack op, test=develop

* remove debug message

* add stack plugin serialize

* remove slice, scale op, will add later

* enhence error message

* revise trt ernie test to conver the stack op CI testi, test=develop

* add stack op serialization

* fix test shape after adding stack op

* remove slice op, will add after implementing serialization

* roll back to min_graph=5 to avoid using slice op

* fix scale op output layer

* implement stack op createPlugin

* use workspace and move the defination to .cu

* move stack plugin creator definition to .cu, test=develop

test=develop

* fix trt plugin registry without trt lib

* support trt4

* refine code style

* develop dynamic shape serilization

* add test param for gelu

* fix bugs

* delete redundant comments

* debug

* fix conflict. test=develop

* fix bug. test=develop

* add trt dynamic shape serialized support

* fix ernie serialized bug
test=develop

* fix codestyle
test=develop

* fix bug
test=develop

* fix bug.test=develop

* modify cmakelist test=develop

* fix bug
test=develop

* fix error message.  test=develop

* fix trt register plugin based on pr#25003

* add trt dynload

* fix deserialization bug of not finding plugin registration

* refine code style

* recover engine key in tensorrt_subgraph_pass

* for ci coverage

* add unittest for deserialization
Co-authored-by: Nhaozech <chenhaoze94@gmail.com>

Use vector instruction (LDG.128) to improve qkv transpose. It
provides 1.4X speedup at same GPU base frequency.
test=develop

* fix multhead matmul's instable
test=develop

* fix multihead matmul bug
test=develop

* fix converage problem
test=develop

test=develop

* fix bug of prelu when rank not equal 4, test=develop

* fix prelu inference, test=develop

* fix api, test=develop

* fix shape when mode is chennel, test=develop

* remove debug code, test=develop

* add unittest, test=develop

* parallel move shared data test=develop

* test=develop

* support the head number == 1
test=develop

* fix slice op error.
test=develop

* Optimize the error messages of paddle CUDA API, test=develop

* fix the error messages of paddle CUDA API, test=develop

* Refactoring PADDLE_ENFORCE_CUDA_SUCCESS, and apply to curand/cudnn/cublas/NCCL,test=develop

* remove build_ex_string,test=develop

* merge conflict,test=develop

* refine ernie trt dynamic shape support
1. add slice op converter
2. add emb eltwise layernorm fp16 support
test=develop

* fix dynamic shape test ut
test=develop

* fix comments.
test=develop

* fix comments
test=develop

* add hard_sigmoid trt op converter

* add hard_swish op converter and plugin. test=develop

* add macro to adapt lower trt version. test=develop

* add elementwise pool2d, prelu, shuffle channel
test=develop

* add scale and refine concat eltwise conveter
test=develop

* refine elementwise converter
test=develop

* refine ut test and enforce error.
test=develop

* modify const cast
test=develop

Revert "[Paddle-TRT] Add hard_sigmoid and hard_swish support(support MobilenetV3) (#23536)", test=develop (#23642)

This reverts commit cdc6d4e2.

* add hard_sigmoid trt op converter

* add hard_swish op converter and plugin. test=develop

test=develop

* add dynamic plugin support.
test=develop

* change emb eltwise layernorm to math function
test=develop

* add emb eltwise layernorm
test=develop

* can run dynamic shape ernie
test=develop

* fix ci
test=develop

* add ut for trt ernie dynamic

test=develop

* refine dynamic shape c++ interface.
test=develop

* fix comments
test=develop

* fix comments
test=develop

* add TRT support for instance_norm op

* add gelu plugin

* align trt bert with gpu

* add support for fused fc with relu,

* add unittest for bert trt

Bug Fix: Paddle-TRT cannot handle adaptive pooling in pool2d op converter and "num" attribute in split op converter (#20733)

* fix pool2d trt converter, test=develop

* add fix for split op converter, test=develop

* remove assert.h

* change PADDLE_ASSERT_MSG to PADDLE_ENFORCE

test=develop

* fix tensorrt paddle_enforce

test=develop

* add tensorrt support for windows

* update paddle-trt for:
    1. fix bug: when batch > 2, core in split plugin.
    2. add leaky_relu trt5.0 support (yolov3 from 65ms to 42ms.)
    3. add new attr to dropout.
    4. shuffle channel, swish, relu6 support
    test=develop

* 1. fix ci
test=develop