* convert multihead to oss

* fix:bug

* fix:delete const cast

* fix:don't support bias_qk

* add vit pass

* fix:convert bug and add preln_residual_bias

* support length=-1

* add UT for convert

* add no_bias_qk support for gpu_multihead_op

* delete infer_shape depends on bias_qk

* oss just can be used in T4 and A*

* fix:change api for ROCM CI

* fix_shape

* code style

* fix assert

* fix to_tensor badreturn

* add select_p

* fix bugs

* add custom test for select_p; modify select_p primrules

* modify according to xiaoxu's comment

* add eq_p, select_p, pow_p, use autograd to test grad of high order

* add requirement of autograd, modify expected type of eq

* modify according to xiaoxu's comment

* import primops to use primops.pow

* [XPU] add some collective ops. test=kunlun

* use XPUOpTestWrapper. test=kunlun

* skip kl1 for collective ops. fix typo: deivce -> device. test=kunlun

* convert_fp16 support multi block

* update

* update

* remove v2_transpose_reshape

* matmul_transpose_reshape

* reshape_transpose_matmul

* Add int8 support for matmulV2

* restore ut

* adjust old ut

* restore parallel UT ruels

* remove mkldnn code from base ops

* move enforces to pass

* remove duplicated functions

* delete duplicated enforces

* feedback from review

* add comments to variables

* enable eltwise support

* dynamic attribute

* remove fusepass tests from op test

* remove fuse pass cases from op test

* revert introduction of dynamic attributes

* style
Co-authored-by: Nwozna <joanna.wozna@intel.com>

* transfer memcpy_h2d from fluid to phi

* use UnchangedInferMeta instead

* restore test_standalone_executor

* add newline to fix codestyle check

* rename pt -> phi

* simplify logic and add check

* make the comment more clear

* remove useless comment

* refine code

* trt engine input data type should be consistent with trt input bindings type

* fix some bugs

* fix some bugs

* fix some bugs

* enhance grid_sampler to support 3d input

* add init file

* add op definition and infermeta

* add kernel definition funcs

* add broadcast infer shape

* add gpu forward kernel

* delete SUB and DIV

* add x_grad

* add template

* add e_grad for min and max

* fix small bug

* temp commit

* temp commit

* add e_grad for sum and mean

* fix some compile bug

* fix compile bugs

* fix compile problem

* add sum forward unittest

* fix broadcast error, add kernel sig, register e_grad, change unit test

* fix grad

* add temp grad fix

* temp commit

* add min max unittest

* add max, min unittest, fix mul bug

* add cpu forward sum and mean

* add forward min max, fix mean unittest

* add cpu backward min max

* fix code-style

* add backward sum mean

* fix rocm ci

* set uniitest timeout

* fix bug of x broadcast to e, gpu grad

* fix bug of x broadcast to e, cpu grad

* rename BOOST_GET_CONST macro

* fix rocm ci

* mv graph_send_e_recv to graph_send_ue_recv

* move out_size to IntArray

* add eager op test

* fix max pool type bug, add unittest for api

* revise api doc

* add fp16 for atomic min and max, add unittest

* add unittest

* add fp16 support for graph_send_recv

* fix unittest fp16 bug

* change OutSizeTensor to Out_size

* move E to Y

* add copyright, fix comment

* review code

* fix thread block size

* fix thread block size

* change api attribute name: pool_type to reduce_op, compute_type to message_op

* change api attribute name, move pool_type to reduce_op, move compute_type to message_op

* make affine_grid_op support 5d_input on cpu and gpu

* add class center sample infershape

* add yaml

* modify unittest

* modify unittest

* remove comment

* fix bug of adaptive pool2d_grad, *test=kunlun

* fix bug of adaptive pool2d_grad, *test=kunlun

* fix bug of adaptive pool2d_grad, *test=kunlun

* [OpAttr]Support VarDesc* and vector<VarDesc*> in Attribute

* add unittest for inference predictor

* change out_size to INTArray

* fix out_size eager bug

* add unittest for out_size tensor

* add deprecated for paddle.incubate.graph_send_recv, add paddle.geometric.send_u_recv and unittests

* fix lowest bug

* fix according review comment

* add default value in yaml

* change api file name

* change name

* move api(erfinv) from legacy_api.yaml to api.yaml

* change inplace_map key

* migrate class center sample kernel

* fix Resize ddim error

* set buffer ptr

* add header

* add header

* remove comment

* remove header

* add margin infer

* migrate yaml

* modify unittests script

* fix a bug in transpose2 about mkldnn

* fix bug

* add log_p for auto_grad

* add log_p_op.cc in prim_op_test srcs

* fix bug of wrong op name; add test in test_primops

* add test case of log in testprimapi

* fix bug of test_without_guard

* no need to fix test_without_guard

* transfer kernel, make complete

* add fill_sig file

* fix code style

* fix fill_sig, add yaml, modify python API

* fix inplace, add inplace testcase

* deprecated_op_names append fill

* resolve comments, add test_backward

* for test

* Revert "for test"

This reverts commit baf58738ca485a06073d771e20e3644d8811bf31.

* fix utf8 bug on windows

* clean tensor.h

* fix gather_nd

fix 5 operator makers with typos which pass string literal to argument 'generated', remove generated as parameter of AddAttr (#44935)

* move mkldnn activation kernel

* fix compile bugs

* fix compile bugs

* deal with conflict

* fix compile bugs

* fix windows compile bugs

* mkldnn unittest fix

* change mutable to alloc

* fix unittest bugs

* modify code according comment

* migrate kernel

* fix sig order

* remove header files

* remove header

* remove header

* modify logits grad

* add fused_multi_transformer post_layer_norm

* add test post_layer_norm

* update trt workspace size param

* update

* update

* update

* use int64_t

* use int64_t

* upate

* update

* move fft kernels to phi, done with cufft, pocketfft, mkl_cdft, hipfft
* make stft_op use fft from phi/kernels/funcs, clean code

* Add unit tests

* matmul_v2 + activation

* matmuls + elementwise_add

* matmul_v2 postops

* transform matmul to v2

* opcompat

* fix fusing matmul with multipe outs

* add shape constraints

* remove unused vars

* change pass order

* - Unit tests to be debugged

- fix

- refactor

- diagnostic

- more diagnostic

- fix

- Fix number two

- fix

- fix

- fix

- alpha added

- more fixes

- compilation fix

- removed diagnostic code

- cosmetic fixes

* lint

* add alpha constraint

* merge matmul refactor

* trigger CI

* - fix

* - another fix

* code style

* add support for matmul+elementwise_add+activation

* code style

* fix bfloat16 bugs

* change append_binary to append_sum
Co-authored-by: NJacek Czaja <jacek.czaja@intel.com>