* support bfloat16 for amp_decorate

* add check_finite for bf16

* fix bug

* add ut

* add ut

* refine code

* add softmax infer kernel

paddle::platform::CudaAtomicAdd
https://github.com/PaddlePaddle/Paddle/issues/45881

* fix softmax int64

* follow comments

* migrate load kernel

* remove load op

* fix test failed

* add fusion dir and fuse_softmax_mask kernel

* remove fusion kernel dir

* migrate infershape

* fix code errror

* modify slice infershape

* code style

* modify slice_unittest

* simplify size op

* trans to cuda manuly

* fix copy error

* gaussian random

* mkldnn to onednn renaming

* fix merge conflicts

* remove fluid code

* onednn renaming

* Move classes from mkldnn_reuse.h to onednn_reuse.h

* Migrate pool+grad, clip+grad and cast oneDNN kernels to PHI

* Refactor grad kernels into separate files

* Fix CI failures

* Fix Codestyle

* Implement reviewer suggestions

* Add new lines after includes for readability
Co-authored-by: NSilv3S <slawomir.siwek@intel.com>

* add save kernel

* add save_sr_kernel

* remove original save_op

* add save gpu kernel

* remove combine kernel

* add port.h include

* add save selected rows test

* remove useless kernel.h

* [XPU] update xdnn to 0906. test=kunlun

* [XPU] update xdnn to 0907. test=kunlun

* gaussian random

* mkldnn to onednn renaming

* fix merge conflicts

* Migrate reduce_op oneDNN kernels to phi

* Remove unnecessary header

* remove fluid code

* onednn renaming

* Change std::vector<int64_t> to IntArray

* Fix code style

* Move classes from mkldnn_reuse.h to onednn_reuse.h

* Move more functions from mkldnn_helper.h to onednn_helpper.h

* Change MKLDNN to OneDNN in VLOG message

* Implement reviewer suggestions
Co-authored-by: NSilv3S <slawomir.siwek@intel.com>

Adapt tensor output_size for conv2d_transpose and depthwise_conv2d_transpose

* first commit

* merged with develop

* merged with develop

* fix merge sequential one dims bugs

* scale kernel

* endline

* add inplace

* fix merge conflicts

* Merge conflicts

* fix amp kernel

* update to remove PADDLE_WITH_XPU macro

* add tensor array

* fix ci bugs

* fix ci bugs

* fix ci bugs

* fix ci bugs

* update by comment

* update code

[Eager, Performance optimization] reduce_all interface move reduce_all flag from python to C++ (#45744)

* [Eager, Performance optimization] move reduce_all flag from python to c++

* polish reduce_all

* fix ci error

* fix errors

* [Eager, Performance optimization] reduce_max / min polish

* polish reduce_max / min

* update min/max kernel reduce_all logic

* fix a mistake

* fix ci errors

* fix errors

* fix grad error of grounorm op when cuda version==11.7

* gaussian random

* mkldnn to onednn renaming

* fix merge conflicts

* remove fluid code

* onednn renaming

* Move classes from mkldnn_reuse.h to onednn_reuse.h

* Move more functions from mkldnn_helper.h to onednn_helpper.h

* Change MKLDNN to OneDNN in VLOG message
Co-authored-by: NSilv3S <slawomir.siwek@intel.com>

* fix compile error in gcc540

* [OpAttr]Fix complation error of XPU from Pool2dGradKernel

* test=kunlun

* migrate memcpy to phi

* fix typo error

* fix typo error

* fix  bug and testcase

* fix typo, uniform_random_kernel.cc header

* fix Alloc pinned bug

* change GPUContext::GetPinnedPlace

* add GetPinnedPlace function

* add GetPinnedPlace function

* restore default throw error

* fix Unimplemented error

* skip StandaloneExecutor testcase

* delete memcpy_sig