* add fused_attention_pass. test=develop

* support fp16. test=develop

* fix format. test=develop

* Fix quantize model deploy bugs when using MKLDNN (#45920)

* fix immutable op quantize bugs

* fix

* fix build bug

* fix test

* notest,test=inference

* fix ppyoloe acc drop bugs

* fix test

* fix test

* add test

* fix

* fix

* fix test

* fix refined name bug

* fix test

* bias fix

* fix matmul weight dequant bug

* re-ci

* fix tester

* fix test

* fix tester

* update weight dequantize func

* update code

* update test for converage

* update test

* update cmake

* update cmakelist

* update code

* rerun ci

* remove useless code

* re-ci

* update code

* update code

* fix header

* update code for log

* stride_to_24

* fix CI failing

* Enhance the layernorm shift partation fuse op when shift size > 0 (roll shifting)
* fix cherry-pick test

Fix some operators when the tensor.numel() > INT32_MAX

support pure bfloat16 for more ops

* Add unsigned int8 propagation

* Add or modify unit tests

* Correct concat scale checking

* Apply review suggestions

* Corrections
Co-authored-by: Njoanna.wozna.intel <joanna.wozna@intel.com>

* Fix bug of reduce_sum op. When input.numel() > INT32_MAX, its result
is wrong.

* Support allow_partial switch, which can be configure in
pipeline_configs. If sent tensor are not the same from
different hosts, they shouldn't been sent partially and
then concated as a whole tensor.

* Change name allow_partial to enable_partial_send_recv.

* Add global variable _enable_partial_send_recv

[CherryPick][Dy2St]Fix recurrent op eager deletion pass error in dy2st

Construct exec and ctx only once in cond op to speed up

Add symbolic shape deduction function for unfold, scatter_nd_add, p_norm, grid_sampler, pad3d, etc (#46291) (#47003)

新增sparse.is_same_shape、sparse.reshape、sparse.transpose 三个API

* [Dygraph] Fix performance of pp+mp by using send/recv_calc_stream instead of send/recv (#46116)

* [Dygraph] Fix Perf of FusedFeedForward and FusedAttention with AllReduce (#46780)

* update

* draft with debug print
* remove debug print
* bug fix for ci

* Support both use_calc_stream and sync_op in send recv APIs (#46023)

* Support both use_calc_stream and sync_op in allgather API (#46295)

* Support both use_calc_stream and sync_op in collective communication API (#46761)

* Move group and all reduce from collective to communication (#45848)

* Completes bfloat16 dtype for collective api in eager mode (#45844)

* Fix collective APIs cannot be recognized when building docs (#46962)
Co-authored-by: NLiYuRio <63526175+LiYuRio@users.noreply.github.com>

cherry-pick : #46322, #46245
Sparse API 支持静态图

* paddle-inference support custom-ops
Co-authored-by: NZhixin Yao <zhixiny@graphcore.ai>

* fix tolower
Co-authored-by: NZhixin Yao <zhixiny@graphcore.ai>
Co-authored-by: NZhixin Yao <zhixiny@graphcore.ai>

* [BUG]Fix expand_as_v2 bug while X and Y with different dtype

* fix commit

* fix reshape2 opteller;
add elementwise min/max register for tensorrt

Fix set_value failure when source tensor is fp16 Dtype and destiny value is a number
(dev PR link:#46801)

* Revert pool+grad oneDNN kernel conversion (#45989)

* [PHI] transpose2_grad op migration (#46139)

* op migrated, Copy(OneDNNContext, ...) added

* mutable_data & op registration in fluid removed

* refactoring

* OneDNNGetDataType to uppercase

* missing cpu check added, handler moved to .h file

* name changed to transpose_grad

* Copy changed back to TensorCopy

* Resizing corrected, Copy(OneDNNContext) removed
Co-authored-by: NPiotr Paturej <48731682+piotrekobi@users.noreply.github.com>
Co-authored-by: NPaulina Gacek <paulina.gacek@intel.com>

Cherry-pick 46541
保证Reset50 TSM deeplabv3模型零修改下实现Layout自动调优

* [PHI] Migrate gelu kernels (#45596)

* gaussian random

* mkldnn to onednn renaming

* fix merge conflicts

* remove fluid code

* onednn renaming

* gelu fwd

* sort activations

* gelu gradient

* remove unused macros

* merge conflicts

* fix merge conflicts

* remove extra contraint from gelu op

* [PHI] relu6_grad kernel (#46501)

* Relu6

* remove fluid handler

* add individual kernel signature

* coding style

* replace bounded_relu with clip

* whitespace

* code style

Co-authored-by: NPiotr Paturej <48731682+piotrekobi@users.noreply.github.com>

* fix gather op convert to only support int32 index as input.
* add ut

[cherry-pick] [PHI] Migrate concat+grad, expand+grad, fill_constant … oneDNN kernels (#45863) (#46727)

* [PHI] Migrate concat+grad, expand+grad, fill_constant, nearest_interp and bilinear_interp oneDNN kernels (#45863)

* Migrate concat+grad, expand+grad, fill_constant, nearest_interp_v2 and bilinear_interp_v2 oneDNN kernels to PHI

* Remove old namespace variable

* Fix invalid out dims error

* Add mutable_data method to concat output

* Add check for -1 dim before computing out_dims

* Capitalize oneDNNGetDataType function name

* Change fill_constant kernel to correct PHI kernel

* Attempt to fix dims error

* Fix fill_constant (full) kernel

* update dependencies
Co-authored-by: NPiotr Paturej <48731682+piotrekobi@users.noreply.github.com>

* [PHI] Migrate sgd and stack oneDNN kernels (#46374)

* Convert slice+grad oneDNN fluid kernels to PHI

* Change mutable_data to Alloc

* Refactor licences

* update dependencies
Co-authored-by: NPiotr Paturej <48731682+piotrekobi@users.noreply.github.com>

* Convert split, pad and pad3d kernels

* Convert slice+grad oneDNN fluid kernels to PHI

* change out->mutable_data to dev_ctx.Alloc
Co-authored-by: NPiotr Paturej <48731682+piotrekobi@users.noreply.github.com>

* init

* remove softmaxop

* merge dev

* correct dir

* style

* First approach

* Shape kernel corrected

* Compilation error fixed

* Resize corrected

* Registered types added

* Mistake corrected & types added

* sum kernel deleted
Co-authored-by: NPaulina Gacek <paulina.gacek.pl@gmail.com>

* set flag of clip_extra in save_inference_model to true (#46151)

* open the clip_extra flag in paddle.static.save_inference_model, test=allcase (#46456)

* Open the clip_extra flag in TracedLayer.save_inference_model (#46473)

* open the clip_extra flag in paddle.static.save_inference_model, test=allcase

* set the defalut value of clip_extra in TracedLayer from False to True, test=allcase

* update english doc of paddle.static.save_inference_model, test=document_fix (#46484)

* Fix clip_extra logic in remove_training_info (#46534)

* fix clip_extra code in remove_training_info

* revert rnn opmaker clear