* fix sqrt grad

* fix sqrt grad

* Add comments for #50886

* [Tensor Operants & Prim-Relevant] Tensor supports logical operants

* add prim dynamic unit test

* add prim static unit test

* tmp gather vjp

* support gather

* remove useless code

* fix compiling error

* fix ut

* add eager test

* add eager test

* add seed

* small change

* fix cpu error

* fix transpose op compat

* remove tensor index case

* fix prim_cinn

* small commit

* add cumsum prim backward

* small commit

* skip aixs=None test case

* fix op generante eror

* fix static test error

* remove unused code

* fix static test error

* small commit

* skip cpu float16 test case

* skip eager cpu cumsum float16 test case

* add eager and static UT

* fix ut

* add composite backward rule

* fix error

* fix type error and format error

* add try cpu+float16 test

* fix test bugs

* remove test for cpu+float16 and make y[0] be the grad arg

* add cinn test

* fix UT

* fix the wrong dim of v in test cases

* change y[0] to y[1] for grad in UT

* reshape flatten out

* Disable cinn single test

* use scatter_nd_add

* modify the reshape part of topk_grad

* delete useless build file

* to make the syntax right

* modify bug

* try use of put_along_axis

* remove cinn test

* reformat todo

* add silu composite rule

* fix code style.

* add cinn test

* fix composite grad maker code gen

* add prim in cumsum op test

* remove old test

* fix typro

* pass the static test

* fix typro

* modify optest and delete old test files

* remove normal test_top_k_op test

* fix typro

* pass axis=None test case

* buffer comment

* for debug

* add silu fp16 unit test.

* add static guard

* remove forward prim test

* remove same name axis

* modify the test_top_v2_op.py to pass all local tests

* delete the useless testcase

* fix mistake

* add more testcases to test dtype16 and dtype32

---------
Co-authored-by: NJiabinYang <360788950@qq.com>
Co-authored-by: NGGBond8488 <857631483@qq.com>
Co-authored-by: Nzxcd <228587199@qq.com>
Co-authored-by: NCharles-hit <wanghao107@baidu.com>

* Add gru qat int8 test

* Change place of model downloading

* Update paddle/fluid/inference/tests/api/CMakeLists.txt
Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com>

* Correct flags names and add description

---------
Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com>

* Added file

* Tests separated and rewritten, fixed fc_lstm_fuse_pass

* Resolve conflicts

* [Extension Operants] Extension supports tensor operants

* Polish fluid init_tensor_operants

* modify name

* merge develop

* original code

* build modify

* success 2*2

* fused dim=1 failed

* success

* modify static

* success for static except dim=1

* delete log

* tmp modify

* success

* success

* add fp1664

* delete fp16 cpu test

* stop windows test

* review modify

* modify tanh test

* modify tanh

* fix_conflixt

* modift static prim

* fix_conflict

* Update test_static_prim.cc

* update

* bug fix

* add cumsum prim backward

* skip aixs=None test case

* fix op generante eror

* fix static test error

* remove unused code

* fix static test error

* skip cpu float16 test case

* skip eager cpu cumsum float16 test case

* add cinn test

* reshape flatten out

* Disable cinn single test

* remove cinn test

* reformat todo

* add prim in cumsum op test

* remove old test

* fix typro

* fix typro

* fix typro

* pass axis=None test case

* remove forward prim test

* remove same name axis

* add unittest for nn.DropOut2D

* add fp16

* add fp16 in docs of temporal_shift_op.cc

* Update test_dropout_op.py

* forbid tensorrt_engine op's output is a persistable var

* fix concat bug

* recommit for ci

* support transpose and reshape

* support reshpe, transpose, cast vjp

* merge develop

* recover unused file

* remove prim base

* support problem

* remove additional status settting

* remove additional status settting

* fix ut

* fix ut

* fix ut

* fix no grad branch

* add more test

* disable fp16 in cpu

* fix test

* add inferface of get registered phi kernels

* change KernelType to KernelKey

* add test

* refactor code

* remove utils

* remove utils

* remove utils

* remove utils

* Update get_data_from_tensor.h

* Update rnn_functor.h

* Update rnn_grad_kernel.cu.cc

* Update rnn_kernel.cu.cc

* Update rnn_kernel.cc

* Update rnn_grad_kernel.cu.cc

* Update rnn_functor.h

* Update rnn_kernel.cu.cc

* Update rnn_kernel.cc

* remove utils

* Update rnn_functor.h

* remove utils

* remove utils

* remove utils

* remove utils

* remove utils

* Update rnn_functor.h

* Update unsqueeze_op.h

* Update utils.h

* roll back

* Update tensor_utils.h

* Update tensor_utils.h

* Update tensor_utils.h

* Update tensor_utils.h

* Update tensor_utils.h

* use TensorToVector

* use TensorToVector

* use TensorToVector

* use TensorToVector

* use TensorToVector

* Update rnn_kernel.cc

* Update rnn_grad_kernel.cc

* Update rnn_functor.h

* Update rnn_grad_kernel.cu.cc

* Update rnn_kernel.cu.cc

* Update rnn_functor.h

* Update rnn_grad_kernel.cu.cc

* Update rnn_kernel.cu.cc

* Update rnn_functor.h

* Update rnn_grad_kernel.cu.cc

* Update rnn_kernel.cu.cc

* add TensorToVector

* roll back

* Update tensor_utils.h

* Update rnn_functor.h

* Update rnn_grad_kernel.cu.cc

* Update tensor_utils.h

* Update rnn_kernel.cu.cc

* Update rnn_grad_kernel.cc

* Update rnn_kernel.cc

* Update rnn_grad_kernel.cu.cc

* Update rnn_kernel.cu.cc

* Update rnn_grad_kernel.cc

* Update rnn_kernel.cc

* TensorCopySync to phi::Copy

* fix codestyle

* rnn_kernel.cc: add ;

* replace all GetDataFromTensor with phi::GetVectorFromTensor

* delete include of util.h

* add sm version check

* use GetGPUComputeCapability

* [Tensor Operants & Prim] Tensor pow API uses elementwise_pow

* unittest change to fill_constant+elementwise_pow

* [Error Msg] Polish error message when GPU kernel not found

* Only test in GPU environment

* [kunlun] support reduce_scatter

* uncomment unittest

* update xccl to 1.0.10

* register bfloat16 datatype for squared l2 norm

* register bfloat16 datatype for softmax with upper triangular mask

* register bfloat16 for tril triu cuda kernel

* [mv fleet] mv fleet to distributed

* [mv fleet] for ci

* [mv fleet] for ci

* [mv fleet] solve ci of version

* Enable matmul + bias fusion in fused_gat_attention.

* Add a variable to control whether using fused matmul + bias.

* allow fall back to fp16 when int8

* refine code

* refine code

* refine code

* ConvertToFusedOp

* change static to inline
Co-authored-by: NTomasz Socha <tomasz.socha@intel.com>

---------
Co-authored-by: NTomasz Socha <tomasz.socha@intel.com>