* add get_cublas_handle() api

* update format

* add unittests

* alter function name

[BUG FIX] Fix softmax cross entropy overflow problem.

* Add tf32 support for A100 tensor core acceleration for cuBLAS (#28732)

* Fixed an error

* Fixed an error

* [Dy2stat] Enable jit.save to Save Without Running (#29579)

Enable jit.save to Save Without Running.

* Modify CublasHandleHolder to Fix Random Unittest Failure. test=develop (#29617)

Modify CublasHandleHolder from using PADDLE_ENFORCE_CUDA_SUCCESS to PADDLE_RETRY_CUDA_SUCCESS to fix random unittest failure. We checked that the unittest log showed CUDA allocation error at this file, which may due to GPU not enough. We fixed similar failure in the past, so we applied PADDLE_RETRY_CUDA_SUCCESS here.

* fix softmax_with_cross_entropy cuda kernel overflow bug, test=develop

* replace old macro & for condition, test=develop

* polish details, test=develop

* Optimize the error messages of paddle CUDA API, test=develop

* fix the error messages of paddle CUDA API, test=develop

* Refactoring PADDLE_ENFORCE_CUDA_SUCCESS, and apply to curand/cudnn/cublas/NCCL,test=develop

* remove build_ex_string,test=develop

* merge conflict,test=develop

* fix warnings oof gcc 8 compilation, test=develop

* fix boost::bad_get, test=develop

* refine PADDLE_ENFORCE, test=develop

test=develop

test=develop

test=develop

test=develop

* "re-commit "

* "picked up"

* "fix ci"

* "fix pdb hang up issue in cuda 9"

* add cuda_device_functions.h

* move reduceSum to elementwise_op_function.h

* Support int64 for sum op

* Refine code

* refind accuracy_op

* follow comments

* follow comments

1. finish lookup table CPU and GPU kernel
2. Add some cuda helper
3. Add some math funtor