* fix flash_attention

* Update mp_layers.py

* add fp16 unittest

* support bf16 and add unittest

* fix according to review

* with printf

* add DropOutNdForwardKernel

* PR comment

* remove init

* delete fluid in context pool

* fix custom op bugs

* fix profiler bugs

* fix ci bugs

* fix window compile bugs

* fix windows bugs

* fix window bugs

* resgister for ftt_r2c, ftt_c2_r

* fix clang-format

* Add output defs for argsort kernel

* Update argsort_kernel.cc

* Update argsort_kernel.cu

* Update argsort_kernel.cc

* finish task

* fix some question.

* fix error

* change unittest:zeroDim.

* add multinomial output defs

* fix register on gpu

* [AMP OP&Test] Support float & bfloat16 when using cub

* fix compile error

* fix

* fix rocm compile error

* rename flash_attn_raw to flash_attn_unpadded

* fix static api

* fix static return

* add fp32 grad plus fp16 param in adamw

* add python UT

* fix test case

* in test_adamw_op py file, force the moment2 value LE 0

* add a compare option

* remove bf16 fused adam kernel case

* remove contexts in tensor_utils

* update from_blob

* update from_blob

* update from_blob

* fix bug

* fix bug

* add rnn and searchsorted output defs

* add gpu kernel

* fix atan2

* fix

* fix

* fix

* fix error

* fix error

* fix

* skip copy beta pow when use_global_beta_pow=1

* revert infermeta

* register dtype and place

* refine code

* Add output defs for generate_proposals,instance_norm kernel

* fix

* untracked files

* prelu_perf

* remove unused files

* upd

* fix bug

* Revert "Revert "【Hackathon No.67】remove operator.h in blas.h (#50989)" (#51467)"

This reverts commit b9d91531.

* remove cout

* add header

* fix missing header

* fix refer fluid error

* fix missing header

* 更新 repeat_interleave_grad_kernel_impl.h

Change to phi style datatype.

* 更新 repeat_interleave_grad_kernel_impl.h

Fix missing header

* datatype fluid -> phi

* paddle::experimental -> phi

* fix reference error

* fix reference error

* fix reference error

* fix errors

* fix missing FLAGS

* fix missing headers

* fix missing headers

* fix missing headers

* fix missing headers

* fix missing header

* fix missing header

* fix errors

* Delete randperm raw op

* fix

* Delete hardswish_raw op

* fix ut

* add fp16 and bf16 test

* update

* remove device_context include

* fix bug

* fix bug

* finish task

* add static_check and fix unittest.

* add int32/64

* Update test_cross_op.py

---------
Co-authored-by: NZhang Ting <Douyaer2020@qq.com>

* add all_gather and fix conflicts

* fix code format

* fix ut

* fix broadcast ut

* Replace paddle::experimental::DataType as phi::DataType

* restore custom_device.cc

* Add output defs for mode kernel

* fix bug

* add fp16/bf16

* add grad bf16

* test name

* Add output defs for conv3d_coo distribute_fpn_proposals kernel

* fix

* add register of KthvalueKernel

add register of KthvalueKernel

* Update kthvalue_kernel.cc

* Update kthvalue_kernel.cu

* Update interpreter_util.cc

* Update auc_kernel.cc

* Update auc_kernel.cu

* Update auc_kernel.cc

* Update auc_kernel.cu

* Update interpreter_util.cc

* Update bincount_kernel.cc

* Update bincount_kernel.cu