* Add GaussianNLLLoss API.

* Change `rotl` `atol`.Check `var` in dynamic graph

* remove assertTrue

* update unittest

* update unittest for ci-covarage.add broadcast with same dim.

* Supply static err print.

* Repair note and example.

* Split unitest.

* empty commit.

* for standard commit.

* for standard commit.

* Add int dynamic graph test.

* Repair parameters name.

* Repair unitest parameters name.

* Repair unitest parameters name

* Repair unitest parameters name

* Repair unitest parameters name

* add square in code-block

* fit few notes.

* fit few notes.

* fit few notes.

* fit few notes.

* add few interpretations.

* add few interpretations.

* add few interpretations.

* fix import.

* fix space.

* empty commit for ci.

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

* Remove climits.

* Fix bug of c_softmax_with_cross_entropy_op. Support ignore_index is
negative number.

* add auto parallel tuner options in launch

* add ut for launch in auto_parallel tuner

fix code format

* fix ci-converage

* add get_kernels.py

* update

* update

* update

* update

* update

* update test=document_fix

---------
Co-authored-by: NYuanRisheng <yuanrisheng@baidu.com>

* rename PD_REGISTER_GENERAL_KERNEL

* Update feed_op.cc

* fix

* Update strings_empty_kernel.cc

* adjust test_elementwise_sub's tolerance, max_relative_error of grad and
atol/rtol of output to 1e-3

* fix the dtype in setUp

* fix the elementwise_sub optest

* modify elementwise_sub optest

* fix and add bf16/fp16 to elementwise_sub

* fix elementwise_sub bugs

* fix bugs

* fix elementwise_sub op

* fix the data type

* fix elementwise_sub

* fix elementwise

* fix elementwise_sub

* fix bugs

* fix elementwise sub

* fix elementwise_sub

* remove scalar and vector

* add autogen code support for expand_as op

* bug fixed

* Support static graph code-gen for yolo_box

* Support static graph code-gen for yolo_box

* Support static graph code-gen for yolo_box

* Update op_compat.yaml

* fix

* fix

* [enforce.h Decouple logging.h] Delete glog/logging.h from enforce.h

* Add logging.h for profiler.cc

* Add logging.h for gloo_utils.h

* Add logging.h for addmm_kernel_impl.h

* Add logging.h for addmm_grad_kernel_impl.h

* Add logging.h for p_send_kernel.cu

* Add logging.h for determinant_grad_kernel_impl.h

* Add logging.h for p_recv_kernel.cu

* Add logging.h for elementwise_grad_base.h

* Add logging.h for transfer_layout_kernel.cc

* Add logging.h for eigvals_kernel.cc and index_select_impl.h

* Add logging.h for all files in kernel directory

* Add logging.h for xpu_info.cc

* Add logging.h for xpu

* Fix _param_grad_names when like 'param@GRAD@GRAD'

* [CINN] optest add cinn check test

* replace set self.check_cinn to pass check_cinn by function parameter

* fix ci bug

* add cinn atol/rtol

* fix bug in cross_entropy in static mode

* fix ci-coverage

* rem cncl in new test

* rem cncl in build sh

* rem cncl in old test

* fix ninja error

* fix_ninja_error_qa

---------
Co-authored-by: NZhang Jun <ewalker@live.cn>

* add autogen code for clip_by_norm op

* bug fixed

* slight modify

* support cuda12+ arch, Hopper arch and discard 30 arch

* add arch 90 for each paddle_known_gpu_archs12

* for comments

* [Auto Parallel] Speedup the completion process

* [Auto Parallel] Skip the property of dist_context when deepcopying

* [Auto Parallel] Remove the unnecessary print

* [Auto Parallel] Move some changes from 2.4 branch to develop

* Update engine.py

* [Auto Parallel] Fix a bug

* add Value OpResult OpOperand class

* add Value OpResult OpOperand class

* fix bug

* fix bug

* add utils

* refine code

* add ptr offset and reset method

* add value impl

* fix bug

* refine comment of ValueImpl

* refine code of OpResult

* refine code of Value

* add some comment

* fix cpu compile bug

* refine code

* add op

* add method for op & test value

* refine unittest

* refine code by comment

* refine code

* refine code

* refine code

* refine code

* Optimize performance of unique kernel

* fix ci