* add split_with_num composite rule

* add split_with_num composite rule

* add split composite rule

* update

* update test

* update test

* delete split_with_num_grad

* first commit

* fix code bugs in for_loop

* fix bugs in cuLoadAddStridedInputs.

* optimization for LayerNormBackwardComputeGradInput

* add unitest for validating the optimization

* fix windows ci error

* cuda graph support multi-stream for new executor

* fix windows compile error

* delete create_cuda_graph_stream

* wisemax fp16 support

* add bf16 support 4 elementwise_max

* append broadcast 4 op 4 fp16 / bf16

* fix elewise_max ut bf16 numeric delta

* append fp/bf16 uts

* add fp/bf16 uts

* change bf16 uts delta

* fix some issue

* add prim 4 fp16

* fix conv2d filter

* add fp16 and bf16 test

* update

* [CINN]Enhance CacheKey hash logic by considering input dtypes (#50557)

---------
Co-authored-by: Njiangcheng <thisjiang@qq.com>

* [prim] enable dygraph_to_static to support custom_vjp

* Pr 50885 (#7)

* [CINN]Enhance CacheKey hash logic by considering input dtypes (#50557)

* [CINN]Enhance CacheKey hash logic by considering input dtypes

---------
Co-authored-by: Njiangcheng <thisjiang@qq.com>

* [prim] enable dygraph_to_static to support custom_vjp

* fix code in a dy2static-friendly way.

* [dystatic] add hooker for prim

---------
Co-authored-by: NAurelius84 <zhangliujie@baidu.com>
Co-authored-by: Njiangcheng <thisjiang@qq.com>
Co-authored-by: Ncxxly <chenxx_id@163.com>

* [prim] enable dygraph_to_static to support custom_vjp

* fix cast prim and vjp dtype mapping error bug

* [dy2static-ci] fix dy2static ci errors.

---------
Co-authored-by: NAurelius84 <zhangliujie@baidu.com>
Co-authored-by: Njiangcheng <thisjiang@qq.com>
Co-authored-by: Ncxxly <chenxx_id@163.com>

* [CINN]Enhance CacheKey hash logic by considering input dtypes

* add unittest

* fix typo

* fix typo

* fix map.at

* fix find

* fix test

* fix cinn cache key structure realize

* using ordered map for attributes

* add test by review advice

---------
Co-authored-by: Njiangcheng <thisjiang@qq.com>

* finish task

* add static_check and fix unittest.

* add int32/64

* Update test_cross_op.py

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

* Fix is_paddle_func not take effect for plain paddle API

* fix typo

* fix typo

* fix UT when np >= 1.24

* optimize decription of this change

* add all_gather and fix conflicts

* fix code format

* fix ut

* fix broadcast ut

* refresh

* compat

* register

* testop

* fix

* fix

* fox

* cast

* cast

* fix

* type

* fix

* out

* cast

* fix

* fix

* fix

* broad

* broad

* broad

* fix

* fix

* fix

* fix

* fix

* broad

* broad

* numel

* fix

* fix

* fix

* fix

* cinn

* fix

* fix

* fix

* fix

* first test

* add unsqueeze_op

* add fp16/bf16

* add grad bf16

* test name

* mkldnn->onednn

* fused softplus op + kernel

* remove extra attributes

* add missing handler

* change var name

* squeeze2_op

* add ut

* fix ut

* fix static

* modity ut

* find relevant testcase

* remove with_data_parallel

* trigger CI

* do not apply ParameterServerGraphOptimizer

* remove useless optimizer

* remove with_data_parallel in test_dist_base

* fix test_fleet_base_3

* only reserve changes for GraphExecutionOptimizer

* fix bug

* fix test_minst_dgc_nccl

* fix typo

* fix test_dist_mnist_gradient_merge

* rm TestDistMnistNCCL2DGCMultiCards

* fix optimizer conflicts

* fix dist_mnist

* fix test_dist_hapi

* delete test_fleet_graph_execution_meta_optimizer & test_fleet_graph_executor

* temporally not delete unittest

* fix unittests

* fix ci

* recover prune in python/paddle/hapi/model.py

* Add expand composite rule

* reshape x when dim_in less than dim_out

* add tile op for expand

* remove rensor shape case when comp prim

* enable cinn case

* dim_out can't be 0

* update test case for prim type

* use python to generate cutlass code

* refine CommonConvKernelPart1, CommonConvKernelPart2

* remove useless code in generate_cutlass_code.sh

* add more config in conv2d_residual

* CommonCutlassConvKernelPart1 and CommonCutlassConvKernelPart2

* add group conv support in util.cu

* remove .sh

* refine name

* make name goodgit status!

* add fuse_alpha

* make code easy to understand

* mot fopen generate in py

* use python script to generate conv2d,group=1 cutlass code

* use const &

* use const & && use python script to generate conv2d/group=1 code

* remove with_data_parallel in test_sync_batch_norm_op

* fix debug code

* polish code

* polish code

* polish code

* [XPU] add increment op.

* fix ci

* [XPU] add reduce_min_grad XPU kernel

* add unittest for reduce_min xpu op

* delete old dygraph op test

* delete old dygraph op test