* Add migrations

* Fix build errors

* Remove elementwise_mul from migration

* Migrate mul_mkldnn_op to matmul_kernel

* Review fixes - changed mutable_data, changed ctx to dev_ctx, fixed namespaces

* switched some funcs to phi

* Deleted not needed phi:: and changed place checking according to standards

* [Eager] Fix paddle.grad interface

* [Eager] Support minimum SubGraph for GeneralGrad

* Add needed_nodes to prune grad graph more thoroughly

* [Eager] Add grad_node_trans_mapping_ to record which grad_node has been transformed to AccumulationNode

* [Eager] Fix paddle.grad interface

* Polish code

* remove potential_stop_node

* Add endding_nodes to enhance genSugraph logic

* clear endding_nodes_

* polish code

* rename endding_nodes to endding_nades_

* Refactor grad interface

* Add register_hook case to fix coverage-ci

* Fix code format

* Refactor general_grad

* Add more code comments

* call clear directly to release GradSlotMeta

* fix a mistake

* fix matmul/ multiply kernel logic and optional input in yaml, fill zeros logic and so on.

* fix batch_norm_double_grad yaml optional config

* fix tanh_triple_grad yaml and kernels

* fix MultiplyTripleGradKernel optional logic

* fix merge mistake

* fix compile error

* remove legacy attr for bn

* polish code

* fix some kernel

* merge develop

* fix error

* remote log

* fix kernel with full like

* hide value log behind

* hide value log behind

* fix matmul_triple grad

* fix xpu compile error

* fix xpu compile error

* fix xpu ut

* fix xpu ut

* fix_xpu_compile_error
Co-authored-by: NWeilong Wu <veyron_wu@163.com>

* profile reduce kernel for fp16 and reduceHigherdim

* use reinterpret_cast

* fix for CI on ROCm

* add Macro for ROCm

* ROCm CI config

* ROCm CI config

* unit test repair

* pull

* add common_funcs.h

* reduceType

* Update reduce_function.h

* not higher

* rename

* Fix fuse_gemm_epilogue

* update tests

* Update CMakeLists.txt

* Update CMakeLists.txt

* Update CMakeLists.txt

* fix random seed

* use assert_allclose

* Update test_dist_fuse_gemm_epilogue_pass.py

* Update cpp_pass.py

* Update test_dist_fuse_gemm_epilogue_pass.py

* fix codestyle

* update seed and atol

* fix capi kernel registration macro error

* update

[Eager, Performance Optimization] modify AllocateFrom to reduce deconstruction of shared_ptr (#48548)

* [Eager] Fix paddle.grad interface

* [Eager] Support minimum SubGraph for GeneralGrad

* Add needed_nodes to prune grad graph more thoroughly

* [Eager] Add grad_node_trans_mapping_ to record which grad_node has been transformed to AccumulationNode

* [Eager] Fix paddle.grad interface

* Polish code

* remove potential_stop_node

* Add endding_nodes to enhance genSugraph logic

* clear endding_nodes_

* polish code

* rename endding_nodes to endding_nades_

* Refactor grad interface

* Add register_hook case to fix coverage-ci

* Fix code format

* Refactor general_grad

* Add more code comments

* call clear directly to release GradSlotMeta

* fix a mistake

* fix matmul/ multiply kernel logic and optional input in yaml, fill zeros logic and so on.

* fix batch_norm_double_grad yaml optional config

* fix tanh_triple_grad yaml and kernels

* fix MultiplyTripleGradKernel optional logic

* fix merge mistake

* fix compile error

* remove legacy attr for bn

* polish code

* fix some kernel

* merge develop

* fix error

* remote log

* fix kernel with full like

* hide value log behind

* hide value log behind

* fix matmul_triple grad
Co-authored-by: NWeilong Wu <veyron_wu@163.com>

* add stat tool

* add roll and roll_grad kernels and strided_slice and strided_slice_grad kernels, test=kunlun

* add silu, unfold and their grads,test=kunlun

* fix boardcasting superlink

* Update bitwise_op.cc

* fix typo errors(from 48186)

* Update python/paddle/distribution/uniform.py
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

* Update math.py

* Update math.py

* refix

* Update logic.py

* BaseTransform api doc; test=docs_preview

* Update python/paddle/vision/transforms/transforms.py

* for text block; test=docs_preview

* Update transforms.py
Co-authored-by: NLigoml <39876205+Ligoml@users.noreply.github.com>

* rename kernel for top_k, slogdeterminant, generate_proposals_v2

* fix bug

* general optimization for no_varlen multihead

* update memory_optimize pass

* do not link python lib in tensor wrapper

* instance norm support dynamic shape
* update unittest

* remove conv_act_set from graph_pattern_detector.cc

* Support FP16 in generic TensorRT plugin.
* Support FP16 for Pad3D.

* fuse-mt passes compatible with structured pruning

* fix typo error

* pass CI-coverage

* fix error log for yaml check

* remove grad_op of increment

* migrate transpose_op.cu.h and gpu_utils.h

* format code style

* fix some problems

* format code

* reset tranpose_op.cc

* test commit

* recover transpose_op.h

* delete transpose_op.h

* adjust header files order in transpose_op.cc

* fix bug of pylayer

* fix bug

* [Perf]Fix interploate OutSize data transform problem

* fix code style

* fix grad

* fix phi kernel

* add activation support
* fix cublasLt bug
* remove useless code and fix test random range

* add fuse act add grad pass

* polish code

* refine code

* add test

* refine code

* rename some kernel name

* fix compile problem

* fix bug of eigen_dependency

* fix xpu compile

* delete unnecessary shape and slice op
Co-authored-by: NYour Name <you@example.com>

some legacy code still use xpu_wait() for stream sync -- it only syncs
default stream. this PR replaces them with dev_ctx.Wait() to ensure
that correct stream is always used

* fix mma_tensorcore (__CUDA_ARCH__)

* disable tensorcore by default.

disable tensorcore by default, because the judgment of __CUDA_ARCH__ will cause undefined behavior in some environments, can manually enable it on a machine that supports tensorcore.