* Delete the /paddle/fluid/platform/device/npu directory

* clear Cmakelists

* Try removing npu in the header file

* decouple memory copy

* fix ci bugs

* fix ci compile bugs

* fix rocm compile

* fix ci bugs

* decouple memory

* deal with conflict

* fix xpu compile bugs

* fix xpu bugs

* deal with xpu bugs

* fix cmake bugs

* fix windows bugs

* fix ci bugs

* fix ci bugs

* delete redundance code

* add code for pybind

* fix py3 bugs

* fix ci bugs

* platform::CUDAPinnedDeviceContext -> phi::GPUPinnedContext

* replace platform::TraceEventCollector

* move DeviceContextPool to phi

* add EmplaceExternalContextFunc

* update namespace

* update cmake

* fix bugs and create context_pool_impl.h

* replace platform::is_xxx_place

* fix bugs

* update generator

* fix bugs

* fix bugs

* fix bugs

* fix bugs

* fix bugs

* fix bugs

* fix bugs

* fix enforce usage

* Revert "fix enforce usage"

This reverts commit 5f521f08a69713cee506e64a00ec6d9fba709e27.

* fix bugs

* rm XPUDeviceContext and CustomDeviceContext

* fix bugs

* fix fix context init bug

* fix bugs after merge

* fix bugs

* fix name

* fix mutable_data

* update and fix bugs

* fix bugs

* update

* fix bugs

* fix name

* fix bugs

* merge

* fix bugs

* create context_pool in phi/backends

* create context_pool in phi/backends

* fix bugs

* fix xpu bugs

* fix rocm bugs

* fix bugs

* fix bugs

* fix bugs

* fix xpu bugs

* update

* update

* fix bugs

* fix bugs

* Support stream priority for standalone executor

* Fix compile error

* Fix compile error

* Fix compile error

* Fix compile error

* Fix compile error

* revert to use default XPU stream for computing

XPUContext now has a null stream by default. If you want to use a separate stream
 (e.g. in async collective communication), you should create a dedicated XPUContext
and invoke its XPUContext::CreateStream()

* minor

* add extra attr property set

* add type_info for all context

* add onednn context to all context

* fix context compile error

* simplify conv kernel args

* pass runtime attr into dev_ctx

* fix marco error

* clear conv_grad_kernel extra args

* merge conv_grad_grad into conv_grad

* clear conv2d_grad_grad extra attrs

* clear yaml and eager extra attr

* fix conv1d error

* change to thread local

* fix npu compile failed

* try to fix windows compile failed

* add conv2d onednn phi kernel

* fix ci bugs (#36)

* fix compile bugs (#38)

* fix extra input transform bug (#39)

* support dynamic created attr (#40)

* reset extra info gen code

* rm conv_grad_grad kernel

* reimpl pass attr adapting

* add int attr support

* remove vector inputnames creating

* fix map at error

* Update paddle/phi/kernels/onednn/conv_grad_kernel.cc
Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com>

* remove useless extra attrs

* replace mkldnn_engine by onednn_engine
Co-authored-by: NYuanRisheng <yuanrisheng@baidu.com>
Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com>

* [CustomDevice] add custom_device_resource_pool & device_event_custom_device

* update

* update

* update

* update

* [CustomDevice] add dy2static support

* update

* remove cudaDeviceContext

* remove more template

* fix rocm compile

* remove alias name CUDADeviceContext

* fix compile

* fix tests

* revert changes

* remove cudaDeviceContext

* remove more template

* fix rocm compile

* init

* move CUDAStream to phi

* fix compilation

* merge develop

* add stream_owned_ member

* split cuda_stream.h

* fix cpu compile

* fix constructor

* fix bug

* fix windows compile

* fix inference test_levit

* fix windows tests

* adapt mkldnn kernel in PHI

* fix ci compile bugs

* fix compile bugs

* fix compile bugs

* fix compile bugs

* fix compile bugs

* delete comment

* fix compile bugs in windows-inference

* delete code for converage

* modify code by review

* modify code by review

* add todo

* fix compile bugs

* fix compile bugs

* fix compile bugs

* fix unittest bugsx

* fix init()

* delete test_device_context

* replace CPUDeviceContext with CPUContext

* fix test_scalar

* remove dot_op.cc

* fix compile

* unify cpu context

* fix init()

* delete test_device_context

* fix test_scalar

* Refactor DeviceContextPool

* Adjust header file order

* - prototype of reimplemented fixes

* - compilation fixes

* - compilation fix

* - cosmetic info

* - hopefully fix

* - compilation fix

* - supported for nested blocking of cache clearing

* - fix

* - Unit test to changes

* - Compilation fix to windows (hopefully)

* - Moved resetting layout to ResetBlob

* - fixes after review

* add xpu buffer_reader, *test=kunlun

* xpu buffer_reader, use XPUDeviceGuard, *test=kunlun

* modify xpu.cmake, *test=kunlun

* modify xpu.cmake, *test=kunlun

* modify xpu.cmake, *test=kunlun

* add xpu buffer_reader, *test=kunlun

* add xpu buffer reader, *test=kunlun

* add xpu buffer reader, *test=kunlun

* add phi device context pool

* change year

* fix compile error

* fix operator = error

* refine init impl

* polish details

* refine init impl

* Added cuBlasLtHandle_t to device context.

* Added fused_gemm_epilogue op.

1. Added fused_gemm_epilogue op to leverage cuBlastLt Epilogue.
2. Support fusion Act(X*Y + bias), X'dims >=2 and Y'dims shoule be 2.
2. Act currently only be supported ReLU. (Will add GeLU in the future).

* Added UT to fused_gemm_epilogue op.

* Added LinearAct Pattern

1. Added LinearAct into graph_pattern_detector.* to define (2.)'s
pattern.
2. LinearAct is used to detect act(element_add(matmul_v2(x, w), bias)).
3. act currently only support ReLU (Will support GeLU in the future).

* Added FuseGemmEpiloguePass

1, Added FuseGemmEpiloguePass to handle nn.Linear + Act{ReLU}
fusion (GeLU will be supported in the future).
2. Only support matmul_v2 from nn.Linear.

* Added pybind to BuildStrageter.fuse_gemm_epilogue_.

* Added UT for fuse_gemm_epilogue_pass.

* GeLU support and EpilogueSingleton

1. Added GeLU support to fused_gemm_epilogue op.
2. Added EpilogueSingleton to cache auxiliary pointer.
3. Added related UTs.

* Rename cublaslt_epilogue_opto gemm_epilogue_op.*.

* Added both train and infer pattern to LinearAct.

1. Added support of fwd graph with grap_ops linking to LinearAct.
2. Added related changes to fuse_gemm_epilogue_pass for above
modification.

* Changed CUDA requirement from 11.4 to 11.6 for fuse_gemm_epilogue_pass.

* Added identity activation support to gemm_epilogue_op.

* Added Linear Fusion (matmul_v2 + ele_add)

1. Added matmul_v2 + ele_add pattern to LinearActPattern.
2. Added matmul_v2 + ele_add support to fuse_gemm_epilogue_pass.

* Rename gemm_epilogue_op.* to fused_gemm_epilogue_op.*

* Add fused_gemm_epilogue_grad op.

1. Added fused_gemm_epilogue_grad to support backward epilogue fusion.

* Add UTs to fused_gemm_epilogue_grad_op.

* Change attribute name in fused_gemm_epilogue_grad_op for clearing.

* Allow DX and DBias be dispensable to fused_gemm_epilogue_grad op.

* Added ElementwiseAdd+Matmul+Act graph pattern detection.

* Fuse backward of Linear( Act(x))

1. Added backward fusion pass to Linear( Act(x)).
2. Added backward fusion pass to Linear(x).

* Added UTs to backward fusion of Linear(Act(x)).

* Complete document of arguments to fused_gemm_epilogue_op.

* Made arguments of some functions pass by reference.

* Modify code with review comments.

1. Made arguments of some function pass by reference.
2. Removed redundant code.
3. Followed Google code style to change code.

* Made 'const' code style be consistent

* Fixed random seed of python UTs.

* Set Compiling constrains to cuBlasLt

1. Require CUDA 11.6+
2. Remove fuse_gemm_epilogue related tests when CUDA < 11.6.

* Code Reivew from Paddle

1. Changed arguments name is_first_gemm to without_x_gradient for
clearing.
2. Applied PADDLE_THROW in fused_gemm_epilogue_op.

* Remove EpilogueSingleton

1. Applied ReserveSpace to replace Epilogue for passing auxiliary
pointers between FWD and BWD.

* Fix a logical error and enhance UTs.

1. Added act op count checking in UTs.
2. Fix issue to fuse backward or ReLU(Linear(X)).
3. TODO: solve GELU fusion issues.

* Fix Linear and GeLU fusion issues.

1. Modified graph_detech_pattern to fit with both linear wiht gelu or
relu.
2. Modified data range in Uts to allow negative values.

* Removed fused_gemm_epilogue_op.h.

* Rename namespace pten to phi.

* Rename name of arguments in fused_gemm_epilogue_op

1. bias -> Bias.
2. out -> Out.
3. reserve_space -> ReserveSpace.

* Change EpiloguePassActivationCache as local variable.

1. Removed singleton in EpiloguePassActivationCache.
2. Made EpiloguePassActivationCache as an argument to each pass
functions.

* [KP] Add elementwise add xpu, test=develop

* modify the File Permissions

* modify the copyright time

* modify code style

* modify code style

* rename pten dir to phi

* rename namespace to phi

* rename infrt pten dir to phi

* resolve conflict

* rename pten to phi in cmake

* revert all infrt change

* change needed files

* fix infrt failed

* fix inference failed