* Optimize the kernel implementation of layernorm with openmp (#20895)

* Add ernie c++ inference test (#21015)

* Add ernie unit test
test=develop

* Add ernie unit test
test=develop

* Add ernie unit test
test=develop

* remove ngraph

* optimize gpu test
test=develop

* optimize codes
test=develop

* fix cmake fails on inference_download_and_uncompress (#21185)

* solve cmake fails on inference_download_and_uncompress
test=develop

* solve cmake fails on inference_download_and_uncompress
test=develop

* Add fc padding to improve mkl GEMM's performance when N and K are multiple of 128. (#20972)

* Add fc padding to solve mkl performance
test=develop

* fix gpu pass and error information
test=develop

* fix fc_fuse_pass_test
test=develop

* fix error information
test=develop

* fix error information
test=develop

* fix name and add fc op padding test
test=develop

* fix attributes
test=develop

* optimize fc padding
test=develop

* fix test
test=develop

* Polish the codes of fc when needs padding (#21378)

test=develop

* Add ernie large c++ inference test (#21365)

* add ernie-large test
test=develop

* add ernie large c++ inference test
test=develop

* Modify padding strategy: remove weight copy in fc padding (#21650)

test=develop

* optimize fc jit (#21878)

test=develop
Co-authored-by: NYihua Xu <yihuaxu@hotmail.com>

* fix the computation for dx (grad for x) for prelu operation. (#20949)

* set the default value of alpha for prelu to 0.25, test=develop

* add the call to __syncthreads(), test=develop

* fix the implementation of cpu prelu, test=develop

* repair the implementation of element mode prelu, test=develop

* modify test_prelu_op.py, test=develop

seems shuffle_sync cannot handle small size

test=develop
Signed-off-by: Nzhaoyuchen <zhaoyuchen01@baidu.com>

* [cherry-pick] All elements in attr(shape) of crop_tensor can be -1 and int32/64 kernel registered (#20756)

* All elements in attr(shape) of crop_tensor can be -1, test=develop, test=document_preview

* fix the bug that attr(offsets) should be initialized, test=develop

* [cherry-pick] maxout supports channel_last input (#20846)

* maxout support channel_last input, test=develop

* modified details of Input(X) and Attr(groups, axis) in doc, test=develop

* [cherry-pick] lrn supports channel_last input, test=develop (#20954)

* delete paddle infershape enforce marco (#20832)

* Polish and arrange code in enforce.h (#20901)

* Enrich the type of error and declare the error type interfaces (#21024)

* Enrich the type of error and declare the error type interfaces, test=develop

* adjust tests to adapt new form, test=develop

* add inference deps with error_codes.pb.h, test=develop

* restore stack iter start pos, test=develop

* polish code based review comments, test=develop

* Add dependency for error_codes.proto (#21084)

* fix activation_functions deps, test=develop, test=document_fix

* add error_codes_proto deps, test=develop, test=document_fix

* try delete enforce.h, test=develop, test=document_fix

* change cuda enforce & add example (#21142)
test=release/1.6

[cherry-pick] fix the bug of conv_transpose: compitable with AnyLayout setting, test=release/1.6 #(20897) (#20918)

Bug Fix: Paddle-TRT cannot handle adaptive pooling in pool2d op converter and "num" attribute in split op converter (#20733) (#20902)

* fix pool2d trt converter, test=develop

* add fix for split op converter, test=develop

* Support fp16 in fused_elemwise_activation_op.
* Fix unit testing in ONLY-CPU mode.

fix conv_transpose's bug: compatible with Anylayout setting: cherry-pick 20589

* Delete PadFuntion, include padding.h instead.

* move function(IsSymmetricPadding) from conv_cudnn_op.cu/conv_transpose_cudnn_op.cu to padding.h.

1.support asymmetric padding;
2.support padding algorithm:"SAME" and "VALID";
3.support channel_last: data_format NHWC and NDHWC;
4.change doc of python API and c++;

test=release/1.6

* fix pool2d pool3d:
1. support asymmetric padding;
2. support padding algorithm:"SAME" and "VALID";
3. support channel_last: data_format NHWC and NDHWC;
4. support inferring shape when input with negative dims in compile time;
5. change doc of python API and c++;
6. fix bug in cuda kernel when Attr(adaptive) is true.

test=develop,test=document_preview

* fix 'tensors' to 'Tensors'. test=develop,test=document_preview

* add test for converage ValueError.test=develop,test=document_preview

* resolve conflict in test_pool2d. test=develop

* make pad and split support fp16
test=develop

* add support of matmul with multiple head even different width and height

Original matmul with multiple head supports only the mat_a.width == mat_b.height,
in that case, mat_b will be horizontally split. In this patch, we extend the
support when mat_a.width != mat_b.height but mat_a.width/head_number == mat_b.height,
in this case, mab_b will be vertically split.

One example is A is [3, 8], B is [2, 16], head_number is 4. In this
case, A will be split as [3, 2], B will be (vertically) split as
[2, 4]. The final result will be 4 matrix of 4 matrix of [3,4], i.e. [3, 16]

test=develop

* add support of matmul with multiple head even different width and height

Original matmul with multiple head supports only the mat_a.width == mat_b.height,
in that case, mat_b will be horizontally split. In this patch, we extend the
support when mat_a.width != mat_b.height but mat_a.width/head_number == mat_b.height,
in this case, mab_b will be vertically split.

One example is A is [3, 8], B is [2, 16], head_number is 4. In this
case, A will be split as [3, 2], B will be (vertically) split as
[2, 4]. The final result will be 4 matrix of 4 matrix of [3,4], i.e. [3, 16]

test=develop

* refactor the code of matmul with multiple head even different width and height

test=develop

* fix softmax ce time limit check failed. test=develop

* refine softmax calc. test=develop

* support 2-level lod of input in sequence_pool test=develop

* fix lod level bug in .cu test=develop

TemporaryAllocator is a singleton used for allocating memory for Cudnn. Since it is a singleton, we can delete it for better performance in memory.

We replace TemporaryAllocator by CUDADeviceContextAllocator and CUDADeviceContextAllocation, which uses stream callback to delete the memory allocated for the stream to avoid singleton.

Also added data_feed_proto to operator to fix CI in CPU compilation

* Refine the codes related to fc op.

* Add GPU implementation for fc functor.

* Apply fc_fuse_pass in GPU inference.
test=develop

* Change the cmake for fc op.

* Change PADDLE_ENFORCE to PADDLE_ENFORCE_EQ.

* Add an attribute to set the activation type in fc_op.

* Enhance the unittest of fc_op.
test=develop

* Remove the declaration of FCOpGrad back to the header file.
test=develop

* Set default value for newly added arguments in test_fc_op.
test=develop

* test=develop,  communicator merge add => merge average

* remove assert.h

* change PADDLE_ASSERT_MSG to PADDLE_ENFORCE

test=develop

* fix tensorrt paddle_enforce

test=develop

test=develop

test=develop

test=develop

* remove unused PADDLE_ASSERT(_IS_NOT_ERROR)

* replace PADDLE_ASSERT with PADDLE_ASSERT_MSG

test=develop

* Implement the operator with sprase matrix multiply

* Update the URL of mklml library.

test=develop

* Disable MKLML implematation when using no-linux.

test=develop

* Ignore the deprecated status for windows

test=develop

* print error code if cuda related API fails

* fix depthwise conv gpu kernel bug, test=develop
* add more depthwise conv test, test=develop

* extend matmul op to support multiple head multiplication

With the support of multiple head, the multiplication of two big matrixes is
split into multiplication of several (head_number) small matrixes. e.g. if
Mat A is [3, 24] and Mat B is [24, 4], when multiple A and B with head_number
as 4, Mat A will be split as 4 matrix of [3, 6] and Mat B will be 4 matrix of
[6, 4]. The result of final matrix will be 4 matrix of [3, 4], i.e. [3, 16].

* add_elementwise_add_inplace_test,test=develop

* rename file, test=develop

* sequnce mask support max length tensor input; test=develop

* add rnn_impl.py; test=develop

* add basic gru lstm unittest; test=develop

* fix api spec; test=develop

* fix sequence_mask op bug;
test=develop
test=document_preview

* change +-*x to elmentwise_op; test=develop

* add mkl flag; test=develop

* fix rnn impl bug; test=develop

* update api spec; test=develop

* fix doc bug; test=develop

* fix lstm bugs; test=develop

test=develop

Optimize the concat and split cuda implementation for cases when the number of inputs/outputs is less than 5. (#17979)

test=develop

* Enable seq_pool op to accept len 0 input

test=develop

* Update sequence_pool's api

test=develop

* Add more unittest cases for seq_pool op

test=develop

* Remove legacy comments

test=develop

* Don't use template in op maker

test=develop

* Enhance fused_elementwise_activation op.
test=develop

* Move the api fused_elementwise_activation to contrib.
test=develop

* Add including files.
test=develop

* Add the support of sigmoid in fused_elementwise_activetion op.

* Update API.spec.
test=develop