* make bfloat16 implicit convert to float/double

* fix bfloat16_test ut compile

* use fused mlp in multi transformer
* Restruct code
* use cublaslt to fuse ffn
* fix conflict

* Migrate elementwise_div

* Migrate elementwise grad kernels

* fix:fix the bug of trt_8.0.3.4

* fix: fix the bug of trt_8.0

* fix: notes

* move vol2col from fluid to phi

* update copyright year

* Skip tests that use fused_ops on H100

* Add error message to FusedOps on H100

Removed ResidualData and Bias from ExtraAttrProperties because it's not an attribute.
Removed bug with checking for ResidualData attribute in matmul_elementwise_add_fuse_pass
Removed residualData from list of matmul outputs in cpu_bfloat16_pass.cc because it's input
Co-authored-by: NTomasz Socha <tomasz.socha@intel.com>

* Delete caching from requantize_mkldnn_op and changed to Acquire API
* Fixed codestyle and implementation

* move "paddle/phi/backends/gpu/gpu_device_function.h" to phi

* update copyright years

* rm "fluid/platform/device/gpu/gpu_device_function.h" in phi

* rm dependence to "gpu_device_function.h" in fluid

* rm gpu_device_function.h etc in fluid

* fix rocm-complie bugs

* fix cuda_helper_test.cu bugs

* fix doc of NPUPlace

* fix doc of NPUPlace, test=document_fix

* add fc-residual quantization

* revert removal of check for use_mkldnn

* fix bug

* add disable_logs

* review fix

call twice AreScalesPresntForNodes instead of if-else

* rewrite residual input to output

* revert fc mkldnn taking residual data

* format fix

* fix LoDTensor->DenseTensor

* LoDTensor->DenseTensor

* output->input

* revert changes to unsupported script

revert changes to unsupported script

* remove fc residualdata from output blocklist in cpu_bfloat16_pass.cc

* cleanup unused code

* unify is_int8 is_bfloat16

* Simplify matmul_v2 FWD kernel

* remove RunKernel methods

* remove import namespace

* remove headers

* clean fluid/phi cross imports

* remove fluid axpy_handler

* delete fluid methods

* activations

* OneDNNMemDesc

* MKLDNNFormatForSize

* MatchShapeToLayout

* MKLDNNMemoryFormat

* MKLDNNFormat

* ReorderMKLDNNHandler

* to_void_cast

* review suggestions

* interpolate

* remove fluid depedency

* init

* ExecuteMatMulV2

* rm fluid kernel

* matmul_grad

* remove mutable_data

* mul_grad

* use mma for QK dot computing in fused_multi_transformer.
* Update fused_multi_transformer_op.cu.h

* move threadpool

fix cmake

* fix make

* move cross_entropy from fluid to phi

* replace mutable_data with Alloc

* use .template

* refactor: replace Collective & PointToPoint with NCCLEnv

* refactor: rename to RunFnInNCCLEnv

* refactor: pass std::function by value

* [CustomPlace] fix amp

* [CustomPlace] fix amp

* fix ut because of too long time matmul fp16

* fused qkvBiasAdd and transpose with split qkv

* fix typo

* fix format

* fix name

* add annotation

* fix comment

* cleanup unused code

* unify is_int8 is_bfloat16

* Simplify matmul_v2 FWD kernel

* remove RunKernel methods

* remove import namespace

* remove headers

* clean fluid/phi cross imports

* remove fluid axpy_handler

* delete fluid methods

* activations

* OneDNNMemDesc

* MKLDNNFormatForSize

* MatchShapeToLayout

* MKLDNNMemoryFormat

* MKLDNNFormat

* ReorderMKLDNNHandler

* to_void_cast

* review suggestions

* interpolate

* remove fluid depedency

* init

* ExecuteMatMulV2

* rm fluid kernel

* matmul_grad

* remove mutable_data

* Migrate conv_transpose to phi

* Move handler to kernel

* kernel m

* Fix formatting

* handler

* remove fluid

* revert tcp_store

* tcp_store

* remove unused

* Fix declaration

* add dnn input

* Fix typo
Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com>

* fix bug of zero_allocator in host

* fix test compile bug

* add unittest

* update test

* Add quick gelu and fused bias add kernel

* fix annotation

* remove useless code

* add fast gelu option and set it in multi transformer op

* add flag to restrict if use fast gelu approximate

* fix flags conflict

* fix use tanh function instead

* add cudart version limit

* use phi fast tanh func

* fix comment

* correct sync behavior for XPU distributed training

XPU support event mechanism similar to cuda event, so it is advisable to
use an event to sync compute/comm streams for performance. However this
mechanism is never fully tested, and inconsistent loss/ending_epochs are
reported. Therefore, this PR replaces event sync with stream waiting as
a temporary solution.

* remove compile warning

* fix device id issue for xpu eager

xpu device id is not correctly set in eager mode, thus vars are on dev0 unless
XPUDeviceGurad is called, leading to this error message for all node rank != 0:
"NotImplementedError: (Unimplemented) Place Place(xpu:0) is not supported."

* fix typo

* fix pybind error

* Refactor conv_kernel and conv_grad_kernel to provide interface for CUDNNv8 implementation

* Fix macro

* Add implementation for conv_kernel and conv_grad_kernel

* Modification after rebase onto latest develop

* Modify plan cache to comply with the API of phi::autotune

* Refactor to reduce duplicate code

* Review fix:
- move functions in  conv_kernel_impl_v8.h and conv_grad_kernel_impl_v8.h to conv_kernel.cu and conv_grad_kernelk.cu
- add const specifier for input tensor
- add logging when plans fail to execute
- move CudnnConvBwdFilterV8 and CudnnConvBwdDataV8 to conv_cudnn_frontend.h

* - move plan building outside of cache

* Fix ROCM build

* remove "gpu_primitives.h" in fluid namespace

* fix PR-CI-GpuPS fail

* fix PR-CI-GpuPS fail