提交 · 0c68ae0c774a710a83f8a730f3d8d482b2090c80 · 机器未来 / Paddle

26 4月, 2022 5 次提交

[cherry-pick] Adapt BKCL comm for XPUPS (#42266) · 0c68ae0c

由 Fan Zhang 提交于 4月 26, 2022

* XPUPS Adaptation (#40991)

* Adapt XPUPS - 1st version - 3.24

* Adapt XPUPS - update XPU PushSparse -  2nd version - 3.24

* Adapt XPUPS - add XPU PullSparseOp - 3nd version - 3.25

* refactor heter comm kernel

* update. test=develop

* Adapt XPUPS - modify by compilation - 4th version - 3.27

* update calc_shard_offset. test=develop

* update xpu kernel. test=develop

* update args of calc_shard_offset

* update. test=develop

* remove customGradMerger

* update. test=develop

* heter_comm update

* heter_comm update

* update calc_shard_offset. test=develop

* heter_comm update

* update args of calc_shard_offset

* update. test=develop

* remove customGradMerger

* update. test=develop

* fix. test=develop

* update. test=develop

* update. test=develop

* update optimizer kernel

* Adapt XPUPS - use WITH_XPU_KP and modify wrapper kernel function - 5th version - 3.30

* update. test=develop

* update pslib.cmake

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* Adapt XPUPS - modify by kp compilation  - 6th version - 3.30

* update. test=develop

* update. test=develop

* update. test=develop

* update optimizer kernel

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* fix. test=develop

* fix. test=develop

* used by minxu

* update heter_comm_inl

* fix. test=develop

* Adapt XPUPS - modify by kp compilation  - 7th version - 3.30

* fix. test=develop

* add optimizer kernel. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* 3.31 update

* Adapt XPUPS - update kp compilation path  - 8th version - 3.31

* add optimizer kernel. test=develop

* fix kunlun not support size_t. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix kunlun not support size_t. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* update heter_comm_kernel.kps 3.31

* fix. test=develop

* fix. test=develop

* update heter_comm_kernel.kps 3.31

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* update heter_comm.h 3.31

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* update hashtable. test=develop

* update. test=develop

* Adapt XPUPS - update by kp compilation  - 9th version - 4.1

* update hashtable. test=develop

* fix. test=develop

* update hashtable 4.1

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* Adapt XPUPS - update by kp compilation  - 10th version - 4.1

* fix. test=develop

* fix. test=develop

* fix. test=develop

* update. test=develop

* modify by compilation 4.1

* update. test=develop

* update. test=develop

* fix. test=develop

* modify by compilation 4.1

* update. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* modify by compilation 4.1

* fix. test=develop

* fix. test=develop

* fix. test=develop

* modify by compilation 4.1 19:30

* fix. test=develop

* update ps_gpu_wrapper.kps 4.1

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* Adapt XPUPS - update by kp compilation  - 11th version - 4.1

* fix. test=develop

* Adapt XPUPS - update by kp compilation  - 12nd version - 4.2

* fix. test=develop

* fix. test=develop

* modify by compilation 4.2

* 4.2 update

* fix. test=develop

* template init. test=develop

* update 4.6

* fix. test=develop

* template init. test=develop

* 4.6 modify by compilation

* hashtable template init. test=develop

* hashtable template init. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=devlop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=devlop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* Adapt XPUPS - update by kp compilation  - 13nd version - 4.7

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* 4.11 update

* fix. test=develop

* fix. test=develop

* 4.11 update

* update by pre-commit

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* 4.12 update

* fix. test=develop

* Adapt XPUPS - update by kp compilation  - 14th version - 4.13

* 4.13 update

* 4.14 update

* 4.14 update

* 4.14 update

* 4.14 modify by merged latest compilation

* retry CI 4.14

* 4.15 pass static check

* 4.15 modify by gpups CI

* 3.16 update by gpups CI - modify ps_gpu_wrapper.h

* 4.16 update

* 4.16 pass xpu compile

* 4.16 retry CI

* 4.16 update
Co-authored-by: Nzmxdream <zhangminxu01@baidu.com>

* modify ps_gpu_wrapper.cc

* update

* Adapt BKCL comm for XPUPS (#42168)

* Adapt XPUPS - 1st version - 3.24

* Adapt XPUPS - update XPU PushSparse -  2nd version - 3.24

* Adapt XPUPS - add XPU PullSparseOp - 3nd version - 3.25

* refactor heter comm kernel

* update. test=develop

* Adapt XPUPS - modify by compilation - 4th version - 3.27

* update calc_shard_offset. test=develop

* update xpu kernel. test=develop

* update args of calc_shard_offset

* update. test=develop

* remove customGradMerger

* update. test=develop

* heter_comm update

* heter_comm update

* update calc_shard_offset. test=develop

* heter_comm update

* update args of calc_shard_offset

* update. test=develop

* remove customGradMerger

* update. test=develop

* fix. test=develop

* update. test=develop

* update. test=develop

* update optimizer kernel

* Adapt XPUPS - use WITH_XPU_KP and modify wrapper kernel function - 5th version - 3.30

* update. test=develop

* update pslib.cmake

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* Adapt XPUPS - modify by kp compilation  - 6th version - 3.30

* update. test=develop

* update. test=develop

* update. test=develop

* update optimizer kernel

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* update. test=develop

* fix. test=develop

* fix. test=develop

* used by minxu

* update heter_comm_inl

* fix. test=develop

* Adapt XPUPS - modify by kp compilation  - 7th version - 3.30

* fix. test=develop

* add optimizer kernel. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* 3.31 update

* Adapt XPUPS - update kp compilation path  - 8th version - 3.31

* add optimizer kernel. test=develop

* fix kunlun not support size_t. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix kunlun not support size_t. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* update heter_comm_kernel.kps 3.31

* fix. test=develop

* fix. test=develop

* update heter_comm_kernel.kps 3.31

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* update heter_comm.h 3.31

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* update hashtable. test=develop

* update. test=develop

* Adapt XPUPS - update by kp compilation  - 9th version - 4.1

* update hashtable. test=develop

* fix. test=develop

* update hashtable 4.1

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* Adapt XPUPS - update by kp compilation  - 10th version - 4.1

* fix. test=develop

* fix. test=develop

* fix. test=develop

* update. test=develop

* modify by compilation 4.1

* update. test=develop

* update. test=develop

* fix. test=develop

* modify by compilation 4.1

* update. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* modify by compilation 4.1

* fix. test=develop

* fix. test=develop

* fix. test=develop

* modify by compilation 4.1 19:30

* fix. test=develop

* update ps_gpu_wrapper.kps 4.1

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* Adapt XPUPS - update by kp compilation  - 11th version - 4.1

* fix. test=develop

* Adapt XPUPS - update by kp compilation  - 12nd version - 4.2

* fix. test=develop

* fix. test=develop

* modify by compilation 4.2

* 4.2 update

* fix. test=develop

* template init. test=develop

* update 4.6

* fix. test=develop

* template init. test=develop

* 4.6 modify by compilation

* hashtable template init. test=develop

* hashtable template init. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=devlop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=devlop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* Adapt XPUPS - update by kp compilation  - 13nd version - 4.7

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* 4.11 update

* fix. test=develop

* fix. test=develop

* 4.11 update

* update by pre-commit

* fix. test=develop

* fix. test=develop

* fix. test=develop

* fix. test=develop

* 4.12 update

* fix. test=develop

* Adapt XPUPS - update by kp compilation  - 14th version - 4.13

* 4.13 update

* 4.14 update

* 4.14 update

* 4.14 update

* 4.14 modify by merged latest compilation

* retry CI 4.14

* 4.15 pass static check

* 4.15 modify by gpups CI

* 3.16 update by gpups CI - modify ps_gpu_wrapper.h

* 4.16 update

* 4.16 pass xpu compile

* 4.16 retry CI

* 4.16 update

* Adapt XPUPS - adapt BKCL comm for XPUPS - 4.24

* update by compilation

* Adapt XPUPS - register PSGPUTrainer for XPUPS - 4.25

* update device_worker_factory
Co-authored-by: Nzmxdream <zhangminxu01@baidu.com>

* update

* update CMakeLists
Co-authored-by: Nzmxdream <zhangminxu01@baidu.com>

0c68ae0c

H
[Cherry-Pick]Fix compiling ort test cases error on Windows(#42186) (#42247) · 5eba3847
由 heliqi 提交于 4月 26, 2022
```
* fix windows compile test case error
```
5eba3847
W

[Eager] Support numpy.ndarry in CastNumpy2Scalar (#42136) (#42213) · 983fcb56
由 Weilong Wu 提交于 4月 26, 2022

983fcb56
fix python3.10 compile bug on windows (#42140) (#42180) · 42297995
由 zhouweiwei2014 提交于 4月 26, 2022
```
cherry-pick #42140
```
42297995

[Cherry-pick] Optimize dygraph performance part2 (#42224) · ab24b9c0

由 Chen Weihang 提交于 4月 26, 2022

* Add paddle::variant and replace paddle::any (#42139)

* add variant and replace any

* split attribute

* Optimize dygraph GetExpectedKernelType perf (#42154)

* opt dygraph scheduling

* revert part impl

* fix variant compile error (#42203)

* replace any by variant in infermeta (#42181)

ab24b9c0

25 4月, 2022 3 次提交

B
fix FlattenContiguousRangeOpConverter out dim error (#42087) (#42184) · 20e8bf1f
由 baoachun 提交于 4月 25, 2022
```
* fix FlattenContiguousRangeOpConverter out dim error

* update code
```
20e8bf1f

[cherry-pick] Optimize performance of dygraph (#42093, #42103, #42137) (#42171) · 0d537003

由 zyfncg 提交于 4月 25, 2022

* optimiaze performance of PreparePhiData (#42093)

* Dygraph performance optimization (v2) (#42103)

* optimiaze performance of PreparePhiData

* dygraph performance optimization

* optimize performance of dygraph (#42137)

0d537003

[Cherry-Pick][Performance]Remove CudaStreamSychornize in ClipGradByGlobalNorm... · 58d0d15e

由 Aurelius84 提交于 4月 25, 2022

[Cherry-Pick][Performance]Remove CudaStreamSychornize in ClipGradByGlobalNorm and fix shape op (#42170)

* [Performance]Set ShapeKernel with ALL_BACKEND and ALL_LAYOUT (#42138)

* [Performance]Set ShapeKernel with ALL_BACKEND and ALL_LAYOUT

* [Performance]Set ShapeKernel with ALL_BACKEND and ALL_LAYOUT

* [Performance]Remove CudaStreamSychornize in ClipGradByGlobalNorm (#42132)

58d0d15e

24 4月, 2022 2 次提交
- T
  add build pylayer depend pybind (#42135) · dd4ef244
  由 tianshuo78520a 提交于 4月 24, 2022
```
解决编译依赖失败问题
```
  dd4ef244
- W
  [Cherry-pick, Eager] Fix CastPyArg2scalar for max value of int64 (#42098) (#42129) · b543998f
  由 Weilong Wu 提交于 4月 24, 2022
```
* [Eager] Fix CastPyArg2scalar for max value of int64 (#42098)

* [Eager] Fix CastPyArg2Scalar in Long case

* Add more test cases for paddle.clip

* Use PyLong_AsLongLong

* Fix merge conflicts
```
  b543998f
23 4月, 2022 1 次提交

[XPUPS]add hashtable interface (#41987) (#42110) · 6ab441bb

由 zmxdream 提交于 4月 23, 2022

* add hashtable interface. test=develop

* update. test=develop

* update. test=develop

* fix. test=develop

* fix optimizer config for xpups. test=develop

* fix. test=develop

* fix. test=develop

6ab441bb

22 4月, 2022 7 次提交
- A
  
  [Eager]Fix SetDeviceId in eager_final_state_api from python_c_gen.py (#42025) (#42067) · 9f3d9381
  由 Aurelius84 提交于 4月 22, 2022
  
  9f3d9381
- P
  Cherry pick PR41990, add _grad_name and _grad_value for eager tensor (#41990) (#42079) · 3475c2bf
  由 pangyoki 提交于 4月 22, 2022
```
* add _grad_name and _grad_value for eager tensor

* fix paddle_enforce

* fix paddle_enforce 2

* fix grad_name

* _grad_value return lodtensor rather than tensor

* fix
```
  3475c2bf
- H
  fix onnxruntime bug (#42095) (#42104) · 26cc5c54
  由 heliqi 提交于 4月 22, 2022
```
修复ORT在batch变动时，输出shape不对问题
```
  26cc5c54
- B
  
  add mkldnn compute_propagate_scales int8 pass (#41592) (#42080) · 41003161
  由 baoachun 提交于 4月 22, 2022
  
  41003161
- J
  
  Add UT (#42055) · 4f6aba87
  由 Jacek Czaja 提交于 4月 22, 2022
  
  4f6aba87
- B
  [Cherry-pick] sharding for eager tensor (#42054) · 6ad0f061
  由 Baibaifan 提交于 4月 22, 2022
```
* sharding_for_eager_tensor (#41415)

* fix_sharding_copy_right (#41849)
```
  6ad0f061
- A
  [IPU] add mixed-precission support for ipu (#41733) (#41906) · c09b1d68
  由 Allen Guo 提交于 4月 22, 2022
```
add mixed-precission support for ipu

cherry-pick from #41733
```
  c09b1d68
21 4月, 2022 18 次提交
- W
  
  [Eager] Support numpy.narray as input for eager expand (#42043) (#42064) · ef0b5fdc
  由 Weilong Wu 提交于 4月 21, 2022
  
  ef0b5fdc
- Z
  [Cherry-Pick]Move pass optimizations into CINN. (#42047) (#42070) · 2f2f987c
  由 Zhen Wang 提交于 4月 21, 2022
```
* Move pass optimizations into CINN.
```
  2f2f987c
- Z
  
  support bce_loss and bce_loss_grad in XPU, test=kunlun (#41610) · b1ba98ca
  由 zhangyikun02 提交于 4月 13, 2022
  
  b1ba98ca
- [cherry-pick]support multi_layer of bilstm,*test=kunlun (#42076) · 58f6d459
  由 z8hanghuan 提交于 4月 21, 2022
```
* modify xpu.cmake,*test=kunlun (#41832)

* modify xpu.cmake,*test=kunlun

* modify xpu.cmake,*test=kunlun

* modify xpu.cmake,*test=kunlun

* modify xpu.cmake,*test=kunlun

* support bilstm,*test=kunlun

* [cherry-pick]support multi_layer of bilstm,*test=kunlun
```
  58f6d459
- L
  [Cherry-pick] fix the bug for nccl barrier and alltoall (#42042) · 8a12f459
  由 lilong12 提交于 4月 21, 2022
```
* fix_nccl_barrier (#41970)

* be compatible with the old version of alltoall (#42007)
Co-authored-by: NBaibaifan <39549453+Baibaifan@users.noreply.github.com>
```
  8a12f459
- W
  
  fix inf in fused_attention (#41933) (#42032) · 50fd2450
  由 WangXi 提交于 4月 21, 2022
  
  50fd2450
- W
  double accessor and show_scale (#41943) (#42014) · efaef31a
  由 wangguanqun 提交于 4月 21, 2022
```
* double accessor and show_scale

* double accessor and show_scale

* rename

* fix bug in pslib config

* add unittest
```
  efaef31a
- B
  update gpu fp16 op blacklist (#41703) (#42051) · 97104695
  由 baoachun 提交于 4月 21, 2022
```
* update gpu fp16 op blacklist

* update blacklist
```
  97104695
- B
  
  add mkldnn int8 pass [step1] (#41579) (#42045) · 04f20b83
  由 baoachun 提交于 4月 21, 2022
  
  04f20b83
- Z
  [cherry-pick] Adjust the Phi C++ API and yaml (#41576, #41778, #41909) (#41928) · d24a402e
  由 zyfncg 提交于 4月 21, 2022
```
* [PHI] Support some c++ api in paddle namespace (#41778)

* support some c++ api in paddle namespace

* change c++ api namespace in custom op

* [Phi] Support setting size of vector<Tensor> for out in yaml (#41576)

* support setting vector out size in yaml

* support setting size of vector<tensor> for out in yaml

* add data transform config for shape and size (#41909)

* fix api_gen bug
```
  d24a402e
- C
  [Cherry-pick] Optimize dygraph scheduling performance (#42010) · ec1d2a16
  由 Chen Weihang 提交于 4月 21, 2022
```
* [Phi] Support setting size of vector<Tensor> for out in yaml (#41576)

* support setting vector out size in yaml

* support setting size of vector<tensor> for out in yaml

* resolve conflict
Co-authored-by: Nzyfncg <zhangyunfei07@baidu.com>
```
  ec1d2a16
- A
  [Eager]Fix full_like/clip with np.generic type as attribute (#41808) (#41974) · e4cb897e
  由 Aurelius84 提交于 4月 21, 2022
```
* [Eager]Fix full_like/clip with np.generic type as attribute

* support numpy genertic

* remove usless code
```
  e4cb897e
- T
  [cherry-pick] enable auto-tune when using cinn (#41795) (#42006) · f5d356b8
  由 TeFeng Chen 提交于 4月 21, 2022
```
cherry-pick #41795
```
  f5d356b8
- B
  
  update demo_ci ut threshold (#41981) (#42030) · efddf9ea
  由 baoachun 提交于 4月 21, 2022
  
  efddf9ea
- J
  
  fix adaptive pool pass bug (#42022) · 5b9cdd9b
  由 JingZhuangzhuang 提交于 4月 21, 2022
  
  5b9cdd9b
- J
  [Cherry-pick] Enabled test_imperative_star_gan_with_gradient_penalty.py under eager mode (#41994) · af7439ad
  由 Jiabin Yang 提交于 4月 21, 2022
```
* cherry-pick python/paddle/utils/code_gen/backward.yaml

* remove unsupported yaml
Co-authored-by: NZhanlue Yang <jim19930609@gmail.com>
```
  af7439ad
- J
  [Eager] make fast through to linear (#41945) (#41995) · 0c141322
  由 Jiabin Yang 提交于 4月 21, 2022
```
* make fast through to linear

* make fast through to linear

* add to do for later upgrades

* support build once for now
```
  0c141322
- C
  [Cherry-pick] Polish custom op details (#42008) · f637e3d2
  由 Chen Weihang 提交于 4月 21, 2022
```
* polish tensor api details (#41971)

* [CustomOp] Fix custom op pinned input error (#41972)

* fix custom op pinned input error

* fix compile error

* fix inference custom op (#41999)

* resolve conflict
```
  f637e3d2
20 4月, 2022 4 次提交

L

update (#41762) (#41843) · 1e18b57b
由 lilong12 提交于 4月 20, 2022

1e18b57b
H
[Dygraph] Refactor Model Parallel in eager mode (#41761) (#41960) · 5ce7f48d
由 Haohongxiang 提交于 4月 20, 2022
```
* refactor mp in eager mode

* update

* update

* add uts
```
5ce7f48d

cherry pick recent updates in graph-engine to release2.3 (#42027) · ef78c9c2

由 seemingwang 提交于 4月 20, 2022

* gpu_graph engine optimization+ (#41455)

* extract sub-graph

* graph-engine merging

* fix

* fix

* fix heter-ps config

* test performance

* test performance

* test performance

* test

* test

* update bfs

* change cmake

* test

* test gpu speed

* gpu_graph_engine optimization

* add ssd layer to graph_engine

* fix allocation

* fix syntax error

* fix syntax error

* fix pscore class

* fix

* recover test

* recover test

* fix spelling

* recover

* fix

* Cpu gpu graph engine (#41942)

* extract sub-graph

* graph-engine merging

* fix

* fix

* fix heter-ps config

* test performance

* test performance

* test performance

* test

* test

* update bfs

* change cmake

* test

* test gpu speed

* gpu_graph_engine optimization

* add ssd layer to graph_engine

* fix allocation

* fix syntax error

* fix syntax error

* fix pscore class

* fix

* recover test

* recover test

* fix spelling

* recover

* fix

* fix linking problem

* remove comment

ef78c9c2

[cherry-pick] Cherry pick pr of new-exec (#42009) · 80992253

由 Leo Chen 提交于 4月 20, 2022

* [new-exec] shrink downstream map (#41471)

* shrink downstream map

* shrink last live ops of var

* add comment

* fix bug

* add dependency for send/recv to support pp parallel (#41652)

* [new-exec] clear the scope listener after run (#41947)

* clear the listener after run

* only sync variables in program

* refine code

* fit for lod_tensor_blocking_queue

80992253

机器未来 / Paddle 与 Fork 源项目一致

机器未来 / Paddle
与 Fork 源项目一致