* rm paddle_install_dir

* update test=document_fix

* update test=document_fix

* update

* add test

* update

* update

* update

* update

* update

* update

* update

* update

* update

* add todo commont

* support register single .cu file

* add register GPU kernel function

* update

* fix bug

* Revert "affine_channel_op"

* fix patch error

* fix patch error

* [CppExtension Cuda] Add cuda unit test for CppExtension

* update extra_compile_args for CUDAExtension

* add debug info

* Add patch to fix CUDA12 compile error

* patch for all env

* add windows judgement

* Try to fix setup function not found error

* fix mix_relu_and_extension include file

* fix setup compile error

* remove useless debug comments

* add sleep, debug CI-build

* add space to disable cmake cache

* remove debug info

* add space to pass CI-build

* add flash randomness control

* fix VLOG undefied

* remove c++14 assert and remove include tensor.h in phi

* remove code with PADDLE_WITH_ASCEND

* try pass codestyle

* rename PD_REGISTER_GENERAL_KERNEL

* Update feed_op.cc

* fix

* Update strings_empty_kernel.cc

* fix ninja error

* fix_ninja_error_qa

* slight modify

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

* add arch 90 for each paddle_known_gpu_archs12

* for comments

* Delete the keyword WITH_ASCEND_INT64 in configure.cmake and CMakeList

* try pass Static-Check

* mv WITH_ASCEND_CL

* mv WITH_ASCEND

* rollback

* remove WITH_ASCEND

* remove WITH_ASCEND

* delete [-Wno-error=terminate], test=develop

* remove GPUps[-Wterminate],test=develop

* remove some -Wno-, test=develop

* fixed glog

* add

* add bfloat16 test for logical op

* rm useless code

* add uint16

* deal with comments

* fixed code style

* fixed code style

* fixed for ci

* deal with comments

* fixed for ci

* delete [-Wno-error=terminate], test=develop

* remove GPUps[-Wterminate],test=develop

* add kernel register macro for all backend

* fix msvc bug

* fix

---------
Co-authored-by: Nzhangyuqin1998 <2368719379@qq.com>

* fix error,test=document_fix

* test

* fix gcc12_error

* fix gcc12_error

* fix gcc12_error

* fix_gcc12_py3_error

* fix_range-loop-construct_error

* fix_gcc12_error

* autofix
Co-authored-by: NLiyulingyue <83450930+Liyulingyue@users.noreply.github.com>

* revert changes in python/paddle/distributed/fleet/utils/hybrid_parallel_util.py

* empty commit, trigger ci

* fix test_slice

---------
Co-authored-by: NSigureMo <sigure.qaq@gmail.com>

* update assign_pos

* update attention_lstm

* update barrier

* update batch_fc

* update beam_search

* update beam_search_decode

* update bilateral_slice

* fix bug

* Handle Structure kernel for InterpreterCore::RunOperator

* fix bug

* fix rocm compile

* fix rocm compile

* Revert "fix rocm compile"

* test

* revert test and update cmake

---------
Co-authored-by: Nchenruibiao <chenruibiao@baidu.com>

* Fix flashattn build error on jetson

* Fix nvcc not found on jetson

* [kunlun] support min/max in dygraph mode

* update xccl to 1.0.13

* fix kp compile

* test

* Revert "test"

This reverts commit 3a1cbfaa0f23e6e06d3dcd8d0b0c28aa63a98e70.

* update copyright

* update cmake

* update cmake

* update cmake

* update cmake

* Fix jetson conv2d_fusion not found error

* Fix jetson conv2d_fusion not found error

* Add comment test=document_fix

* fix generate_kernels.py in CUDA 12.0

* fix attrs bug