Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

merge Master into zj/develop #21

Merged
merged 346 commits into from
Aug 23, 2022
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
346 commits
Select commit Hold shift + click to select a range
6f24d92
Multi Tensor apply Optimizer (#8373)
MARD1NO Jul 4, 2022
1600dcf
Fix doc and ops template auto gen (#8546)
zhongshsh Jul 4, 2022
66027d0
fix diag 0size tensr shape infer bug (#8557)
BBuf Jul 4, 2022
d2e40b4
Format tensor on cpu (#8548)
liujuncheng Jul 4, 2022
b537eae
Remove useless WITH_CUDAs (#8562)
liujuncheng Jul 4, 2022
81edd93
unique identity (#8509)
strint Jul 4, 2022
55b822e
Add GenericStreamContext (#8560)
liujuncheng Jul 4, 2022
06d73d0
Modify some file and add test (#8556)
small1945 Jul 5, 2022
ef2ce66
Move some op into amp gray list (#8545)
leaves-zwx Jul 5, 2022
b495587
Refine inplace expand runtime_error (#8561)
lixiang007666 Jul 5, 2022
51b9a1e
OneEmbedding use malloc async (#8543)
guo-ran Jul 5, 2022
11e6278
fix cpu aligned_alloc size (#8569)
daquexian Jul 6, 2022
9ac0679
Add flow norm (#8535)
zhongshsh Jul 6, 2022
7ff5981
fix pad_packed_sequence method input requires_grad==True (#8574)
Ldpe2G Jul 6, 2022
91eab12
fix embedding manager tmp buffer (#8585)
guo-ran Jul 6, 2022
28690a2
fix reduce_ops 0size bug (#8551)
BBuf Jul 6, 2022
1531b06
Align Momentum Optimizer (#8549)
MARD1NO Jul 6, 2022
58d6f3c
Fill GetSbp bug and consistent test bug (#8576)
wyg1997 Jul 7, 2022
2a407fb
Dev Fully fused MLP Grad[OneEmbedding] (#8462)
MARD1NO Jul 7, 2022
b136ce4
rename mirrored to local (#8503)
Flowingsun007 Jul 7, 2022
3ba3211
Implement BroadcastElementwiseUnary primitive (#8384)
EsdeathYZH Jul 7, 2022
b076be7
skip cpu autotest for graph global (#8593)
lixiang007666 Jul 7, 2022
cc4a2ae
Add function_library.h Exception (#8241)
Alive1024 Jul 7, 2022
c4b6912
Refactor shrink (#8573)
strint Jul 8, 2022
4fa1520
Speed up SliceKernel (#8589)
wyg1997 Jul 8, 2022
c176952
Update readme and vsn for 0.8.0 (#8600)
jackalcooper Jul 8, 2022
464f85f
modify some file and improve error message (#8592)
small1945 Jul 8, 2022
69fdb27
rename consistent to global (#8505)
Flowingsun007 Jul 8, 2022
6b7e6d1
add module releated container docs (#8580)
BBuf Jul 8, 2022
07eaade
fix rnn util extra memory usage when requires_grad=False (#8603)
Ldpe2G Jul 9, 2022
25ec8d1
use bracket format slice in tensor str (#8489)
liufengwei0103 Jul 9, 2022
a233d7d
Perf TensorInfo constructor (#8606)
wyg1997 Jul 10, 2022
a79415e
print operators' python location when print nn_graph (#8558)
xiacijie Jul 10, 2022
1ed7f9e
Add randint like (#8598)
BBuf Jul 10, 2022
d595892
Add full_like api (#8595)
BBuf Jul 10, 2022
b79be4f
fix cumsum GenBackwardOpConfFn (#8604)
liufengwei0103 Jul 10, 2022
31c922e
revert change (#8613)
Flowingsun007 Jul 11, 2022
44886c1
fix test graph optimization conf CI bug (#8617)
xiacijie Jul 11, 2022
bfaa258
Release pod tensor (#8552)
lixinqi Jul 11, 2022
8dfb3e1
Add param group for optimizer (#8611)
BBuf Jul 11, 2022
6cf1bae
fix broadcast_elementwise_binary cpu (#8625)
guo-ran Jul 11, 2022
9674b41
align exception msg to torch (#8627)
Flowingsun007 Jul 12, 2022
ebbcab6
skip unstable global test in ci, reduce failture rate (#8635)
BBuf Jul 12, 2022
3526beb
fuse embedding interaction (#8586)
guo-ran Jul 12, 2022
601043c
fix flip gen backward opconf (#8605)
liufengwei0103 Jul 12, 2022
5759b7e
Add ONEFLOW_ONE_EMBEDDING_PERSISTENT_TABLE_SNAPSHOT_LOAD_MMAP_LOCKED …
liujuncheng Jul 12, 2022
d8ad86c
Profiling main thread (#8601)
lixinqi Jul 12, 2022
8ffab16
Fully Memory Log V2 with more details (#8565)
chengtbf Jul 12, 2022
8076433
Stream policy (#8590)
lixinqi Jul 13, 2022
3253b23
Add fully support for broadcast matmul (#6937)
MARD1NO Jul 13, 2022
09601e1
Revert "skip cpu autotest for graph global" (#8608)
jackalcooper Jul 13, 2022
dd580f2
OneEmbedding add tmp_buffer allocator (#8588)
guo-ran Jul 13, 2022
a645784
refine error msg for some user ops (#8579)
pingzhuu Jul 14, 2022
4ee6a03
Add tril fill value (#8655)
MARD1NO Jul 15, 2022
8f01ed9
fix_non_pod_data_allocate_bug (#8657)
clackhan Jul 15, 2022
0f3ebdc
Fix norm (#8629)
zhongshsh Jul 15, 2022
d024f82
fix_decorate_mem_leak_bug_in_eager_boxing (#8661)
clackhan Jul 16, 2022
9e38f03
add higher order derivative for leaky_relu and negative op (#8643)
pingzhuu Jul 16, 2022
b8b8eaa
update oneflow intro to show the difference (#8669)
strint Jul 16, 2022
8683fec
Stacked error (#8671)
lixinqi Jul 17, 2022
42fb865
Refactor tensor initializer (#8626)
wyg1997 Jul 17, 2022
2c53513
Fix nn doc (#8650)
zhongshsh Jul 17, 2022
b6bd1ac
Fix reduce max min bool dtype bug (#8651)
BBuf Jul 17, 2022
8877ae6
Remove redundant exception wrapper (#8631)
Flowingsun007 Jul 17, 2022
a3b19ed
Refactor MemoryCase to eliminate determine statements of device_type …
leaves-zwx Jul 17, 2022
5fbf385
fix some data races in c++ api and SteadyVector (#8654)
daquexian Jul 17, 2022
d1a7d24
Fix sin/cos higher order derivative (#8648)
wyg1997 Jul 18, 2022
62b2504
refine_eager_boxing_to_adapt_ep (#8568)
clackhan Jul 18, 2022
28e687f
Fix repeat bug (#8645)
Flowingsun007 Jul 18, 2022
52ab1ee
Instruction policy (#8583)
lixinqi Jul 18, 2022
4d9dac4
handle non-contiguous input (#8665)
Flowingsun007 Jul 18, 2022
dbe120d
rename define CONSISTENT to GLOBAL (#8652)
ouyangyu Jul 18, 2022
2733168
Refine naive interpret (#8672)
lixinqi Jul 18, 2022
e15a8bc
Rebuild Docs V0.8.0 (#8392)
QiaolingChen00 Jul 19, 2022
4856d69
Fix zeros like and ones_like api (#8632)
BBuf Jul 19, 2022
6ccedd3
Fix sbp print bug (#8689)
Yipeng1994 Jul 19, 2022
f57b0a0
eager_local_interpreter_with_infer_cache (#8619)
clackhan Jul 19, 2022
90a6b10
fix gelu nn.Module bug and support tanh mode. (#8693)
BBuf Jul 20, 2022
ec81c76
Fix bug in CrossFeatureInteraction LazyBackward (#8677)
MARD1NO Jul 20, 2022
000072f
fix floating-point scalar tensor in arange (#8673)
shangguanshiyuan Jul 20, 2022
c677eea
Add nn functional fold (#8667)
zhongshsh Jul 20, 2022
3d0de2a
modify some file and improve the error message (#8566)
small1945 Jul 20, 2022
84cbd47
[OneEmbedding] add id_shuffle_copy_out (#8683)
guo-ran Jul 20, 2022
c25ae2b
fix add_param_group step key not match error (#8698)
BBuf Jul 20, 2022
2f1a1ae
add env ONEFLOW_EP_CUDA_DEVICE_FLAGS and ONEFLOW_EP_CUDA_STREAM_FLAGS…
guo-ran Jul 21, 2022
3857e57
fix for docsv0.8 (#8710)
QiaolingChen00 Jul 21, 2022
ba0f0c2
fix repeat op 0-size releated bug (both in FW and AD) (#8707)
BBuf Jul 21, 2022
26150ed
Support Dropout Scale in FusedMLPGrad[OneEmbedding] (#8633)
MARD1NO Jul 21, 2022
ae33d7f
Fix bug of Tensor.type (#8697)
marigoold Jul 21, 2022
e9b7a4b
ONEFLOW_GRAPH_PLACE_TRAINING_STATE_ON_ALL_RANKS (#8706)
guo-ran Jul 21, 2022
d6f1fcb
define_mut_output_shape_and_mut_output_stride_in_infer_ctx (#8709)
clackhan Jul 22, 2022
66df6c0
Add qat conv modules (#8368)
Ldpe2G Jul 22, 2022
737878e
add unsqueeze_multiple_op (#8714)
small1945 Jul 22, 2022
7664464
modify broadcast_like_op.cpp and add test (#8720)
small1945 Jul 22, 2022
4153e74
JIT LR (#8500)
jackalcooper Jul 22, 2022
338a91e
Add logspace (#8599)
zhongshsh Jul 22, 2022
f721e94
Add hann_window (#8615)
zhongshsh Jul 22, 2022
d0356a0
add ONEFLOW_VM_PENDING_HANDLE_WINDOW_SIZE (#8730)
ouyangyu Jul 23, 2022
349b1db
Fix as strided bool type and view bug (#8713)
BBuf Jul 23, 2022
882ba2f
Add functional binary cross entropy (#8708)
zhongshsh Jul 23, 2022
cf27cde
support map_location in flow.load (#8666)
daquexian Jul 23, 2022
b542e15
Add addcdiv (#8581)
zhongshsh Jul 23, 2022
fdd12e1
Inner most dim case for cumsum cumprod op (#8403)
liufengwei0103 Jul 24, 2022
1307edf
Define mut output dtype and mut output is dynamic in infer ctx (#8716)
clackhan Jul 24, 2022
5f615c1
Dev refactor fuse instruction policy (#8624)
ouyangyu Jul 24, 2022
8028e07
fix bug of batchnorm num_batches_tracked global error when loading st…
marigoold Jul 24, 2022
3432969
add launch master port limit (#8563)
ouyangyu Jul 25, 2022
ad5b104
Fix docs import distance (#8691)
laoliu97 Jul 25, 2022
60bc35c
refactor control stream type (#8647)
ouyangyu Jul 25, 2022
60827b0
Define mut output tensor desc (#8717)
clackhan Jul 25, 2022
146288e
Symbolic local tensor meta (#8662)
clackhan Jul 25, 2022
73f84df
Feat general basic communication (#8437)
Yipeng1994 Jul 25, 2022
8795f81
Move nonzero to c plus plus side (#8694)
clackhan Jul 26, 2022
7007647
impl of reflection_pad1d and replication_pad1d (#8724)
Flowingsun007 Jul 26, 2022
5555959
refine infer shape fn (#8733)
Flowingsun007 Jul 26, 2022
a91d570
fuse bce loss ops (#8734)
guo-ran Jul 26, 2022
c7f8e16
Fix tensor argsort bug (#8736)
minasora Jul 26, 2022
4a316ab
refactor op call instruction (#8718)
ouyangyu Jul 26, 2022
70bb9be
oneEmbedding add shuffle p2p kernel (#8705)
guo-ran Jul 26, 2022
d56e712
fix p2p kernel cuda_version (#8756)
guo-ran Jul 27, 2022
abccd44
Dev refactor barrier instruction policy (#8729)
ouyangyu Jul 27, 2022
08b94f4
Fix masked_fill graph grad (#8742)
zhongshsh Jul 27, 2022
b842f8a
refactor EpRecordEventInstructionPolicy and AccessBlobArgCbInstructio…
ouyangyu Jul 27, 2022
d310e5e
fix var cuda kernel illegal memoey (#8751)
BBuf Jul 27, 2022
5b9194c
Fix global tensor clone bug (#8719)
lixiang007666 Jul 28, 2022
bfe0765
Add RMSLayerNorm Module (#8725)
BBuf Jul 28, 2022
f06bd60
refactor lazy job instruction policy (#8735)
ouyangyu Jul 28, 2022
01d72db
refine qat conv module tests (#8748)
Ldpe2G Jul 28, 2022
670adfb
refine oneflow readme introduction (#8779)
xiacijie Jul 28, 2022
eba8f19
remove unused graph resource config API (#8727)
xiacijie Jul 28, 2022
a10c1fb
refactor PadFunctor (#8747)
Flowingsun007 Jul 28, 2022
fef142b
refactor touch tensors instruction type (#8774)
ouyangyu Jul 28, 2022
9f1d6c6
add SparseSoftmaxCrossEntropyMsGrad op (#8758)
marigoold Jul 28, 2022
2170a12
fix gradient shuffle bug and typo (#8759)
guo-ran Jul 29, 2022
d3fba10
United allocators (#8591)
lixinqi Jul 29, 2022
8750602
fix t5 layernorm test bug (#8793)
BBuf Jul 29, 2022
34e8fd5
MLIR sbp dialect attribute for parallel signature (#8492)
jackalcooper Jul 29, 2022
5bad1ca
resolve the bug of using ONEFLOW_PYTHON_BASE_DIR in CMake (#8792)
xiacijie Jul 29, 2022
b38c675
fix amp pass when lbi2ibns size greater than 1 (#8746)
guo-ran Jul 29, 2022
a2e5ba5
Return infinity for different sbps while is_mutable (#8783)
Yipeng1994 Jul 29, 2022
0107361
Refactor ep stream types (#8790)
clackhan Jul 30, 2022
3b547ac
RawReader (#8721)
liujuncheng Jul 30, 2022
5b8206e
Fix kineto and cupti not found (#8786)
mosout Jul 30, 2022
506cb3f
fix dynamic_loss_scale_schedule ods and adjust the round trip pass or…
hjchen2 Jul 30, 2022
881f54b
refactor auto contiguous and check view inplace operation (#8791)
hjchen2 Jul 30, 2022
a61720b
Fix pip install failure in release workflow (#8801)
jackalcooper Jul 31, 2022
4d11231
Dev refactor critical section instruction policy (#8761)
ouyangyu Jul 31, 2022
0afda50
add isfinite (#8023)
liufengwei0103 Jul 31, 2022
829e253
Refactor ccl allreduce (#8760)
clackhan Jul 31, 2022
e25d3c0
_shutdown_workers does nothing if _utils is freed (#8804)
daquexian Aug 1, 2022
1d466b4
refactor_critical_section_and_lazy_job_stream_type (#8805)
clackhan Aug 1, 2022
594a358
mv id_shuffle testcase to expensive dir (#8806)
guo-ran Aug 1, 2022
31df497
Fix bug of init_tmp_buffer_ptr in CallContext (#8811)
clackhan Aug 1, 2022
58576b5
Fix global tensor clone (#8813)
lixiang007666 Aug 1, 2022
6351b25
relax cuda.set_device requirement (#8794)
daquexian Aug 1, 2022
499b4dc
Remove OfBlob, ForeignXXX kernels and other old code (#8785)
daquexian Aug 1, 2022
d531f94
Broadcast tensors (#8745)
lixinqi Aug 1, 2022
59eb330
Remove PhyInstrOperand and InstructionType (#8815)
ouyangyu Aug 1, 2022
104e01d
Tmp compute (#8570)
lixinqi Aug 1, 2022
b9f449f
add double grad for slice op (#8784)
pingzhuu Aug 2, 2022
7e71d4c
scalar math kernel use primitive (#8612)
guo-ran Aug 2, 2022
dad9f88
Rename StreamRole to StreamType (#8816)
ouyangyu Aug 2, 2022
2f77d59
Tensor from numpy support stride (#8808)
Flowingsun007 Aug 2, 2022
3e0cede
Dev AdaDelta Optimizer (#8636)
MARD1NO Aug 2, 2022
c4dc1c4
Sequentialize add n (#8507)
lixinqi Aug 2, 2022
4b12dbb
Sync vm mode guard (#8212)
lixinqi Aug 2, 2022
e297d1f
Fix copy not support broadcast (#8773)
BBuf Aug 2, 2022
b32d46d
fix get default cpu device (#8752)
shangguanshiyuan Aug 2, 2022
9ee553b
separate lazy and eager tensor names (#8826)
hjchen2 Aug 3, 2022
34b6d54
Add Cross Feature Interaction in AMP List[OneEmbedding] (#8807)
MARD1NO Aug 3, 2022
738c4b0
Env var compute on worker thread (#8687)
lixinqi Aug 3, 2022
29c6a3c
Schedule yield (#8796)
lixinqi Aug 3, 2022
39e4fe6
add conv higher order derivative (#8688)
pingzhuu Aug 3, 2022
f0d5359
refine graph lr scheduler test (#8829)
hjchen2 Aug 3, 2022
9f5a1f6
Fix nn init eye bug (#8825)
BBuf Aug 3, 2022
8cde8d1
Fix binary cross entropy with logits op bug (#8819)
BBuf Aug 3, 2022
a8982ef
Fix build failure when accessing https://docs.python.org/3/objects.in…
jackalcooper Aug 4, 2022
23298ea
Primitives check n_dims gt 0 (#8827)
liujuncheng Aug 4, 2022
651a6ea
Default copy eager boxing expr (#8830)
clackhan Aug 4, 2022
f59583e
Support OneEmbedding in cpp api[OneEmbedding] (#8681)
MARD1NO Aug 4, 2022
b9165b9
Stream wait (#8571)
lixinqi Aug 4, 2022
230b510
Refactor ccl all gather and reduce scatter (#8814)
clackhan Aug 4, 2022
991a4e2
Bump nccl up to 2.13.4 (#8738)
guo-ran Aug 4, 2022
67c2036
modify reduce_like_ops.cpp and broadcast_like_op.cpp (#8762)
small1945 Aug 5, 2022
88f147d
Refactor 1n1d sbp (#8755)
pingzhuu Aug 5, 2022
94a4467
Prevent benchmark failure (#8860)
jackalcooper Aug 6, 2022
d12f06b
Feat support more tensor setitem (#8741)
wyg1997 Aug 6, 2022
f45a978
libai support bfloat16 (#8818)
guo-ran Aug 6, 2022
6843c5d
resnet50 support amp data_type bfloat16 (#8812)
guo-ran Aug 6, 2022
694d397
fix wrong paths to keep for op repr locations (#8851)
xiacijie Aug 6, 2022
6e26cbb
Refactor ccl reduce and broadcast (#8823)
clackhan Aug 6, 2022
0db31e9
fix build for cuda_bf16 (#8862)
guo-ran Aug 7, 2022
9ed88b0
remove old serving code (#8781)
daquexian Aug 7, 2022
0ccbae9
add module.requires_grad_ api (#8836)
Flowingsun007 Aug 8, 2022
a0f0985
register l2_normalize double dtype (#8863)
BBuf Aug 8, 2022
d9bad54
Support parameter `p` for flow.bernoulli (#8824)
marigoold Aug 8, 2022
d92efa6
Add clamp_min/max and inplace version functor (#8850)
marigoold Aug 8, 2022
8f91128
Revert "Add clamp_min/max and inplace version functor" (#8879)
marigoold Aug 8, 2022
aef37f3
develop grad acc functional apis (#8567)
hjchen2 Aug 8, 2022
b22e7dc
fix gather not support negative axis (#8875)
BBuf Aug 8, 2022
476a29c
Fix not requires grad tensor inplace (#8872)
wyg1997 Aug 9, 2022
e291399
Add empty like op (#8861)
BBuf Aug 9, 2022
83ab665
Only initialize out_grads for those requires_grad outputs (#8878)
duck7216 Aug 9, 2022
49dd66d
Support slice update kernel with stride (#8810)
wyg1997 Aug 9, 2022
6b20fce
Move copy h2d/d2h to user op (#8809)
howin98 Aug 9, 2022
9a1fc46
Fix deadlock in instruction done (#8897)
lixinqi Aug 10, 2022
952d1a1
refine usage of maybe_wrap_dim (#8892)
Flowingsun007 Aug 10, 2022
978942c
fix nn.init.constant_ (#8899)
Flowingsun007 Aug 11, 2022
1b7c31f
infer output stride in lazy mode for the ops which do not support non…
hjchen2 Aug 11, 2022
dd98909
Fix set_acc_grad and backward wrong value (#8575)
wyg1997 Aug 11, 2022
ef9d83b
fix bug in cosine_similarity when inputs have different dims (#8902)
marigoold Aug 11, 2022
4ccf3ed
register reduce_sum_like backward function and add testcase (#8895)
pingzhuu Aug 11, 2022
27e94d2
support non contiguous inplace (#8867)
hjchen2 Aug 11, 2022
381fb6c
Default no vm sync in clip grad (#8885)
lixinqi Aug 11, 2022
4bdbf95
Add clamp_min/max and inplace version functor (#8882)
marigoold Aug 12, 2022
5086f32
Add double grad for broadcast_matmul_grad_b op (#8844)
pingzhuu Aug 12, 2022
4338d96
Fix non-contiguous tensor initialize functions (#8907)
wyg1997 Aug 12, 2022
7ecb256
Add pairwise_distance api (#8908)
marigoold Aug 12, 2022
f1bbad6
Add error msg for local/global mismatch when loading state_dict (#8894)
marigoold Aug 13, 2022
f619b45
Dev linalg inv (#8183)
simonJJJ Aug 14, 2022
9c8644b
add mse_loss and ls_loss interface (#8884)
xiezipeng-ML Aug 15, 2022
2a3e875
modify setitem to support scalar tensor (#8886)
small1945 Aug 15, 2022
c846b0a
Migrate MLIR JIT Op to User Op TableGen (#8870)
howin98 Aug 15, 2022
f61a161
unified autograd engine (#8510)
hjchen2 Aug 15, 2022
0dcb5fa
fix amp pass (#8925)
hjchen2 Aug 17, 2022
4a33514
Fix stack bug for 129inputs (#8927)
wyg1997 Aug 17, 2022
c0f06fd
support tosa make broadcastable pass (#8923)
howin98 Aug 17, 2022
cd3b126
Fix cumprod grad get sbp (#8929)
wyg1997 Aug 18, 2022
7c34b18
remove the deprecated user op grad registration (#8946)
hjchen2 Aug 18, 2022
338be20
add double grad for activation (#8909)
pingzhuu Aug 19, 2022
d5392bf
nccl logical kernels support nv_bfloat16 (#8962)
liujuncheng Aug 19, 2022
a86efb6
align repr of nn.Embedding (#8922)
Flowingsun007 Aug 19, 2022
dbbc917
replace xor with HashCombine (#8932)
lixinqi Aug 19, 2022
6d899af
fix docstring of oneflow.pow (#8960)
doombeaker Aug 19, 2022
e4016b8
Support empty ndim advancedindexing (#8948)
small1945 Aug 19, 2022
cbea1ad
Fix bug when retain_grad for leaf_tensor (#8956)
wyg1997 Aug 20, 2022
8f5745b
support throttle to fix ninja c1 oom (#8953)
howin98 Aug 20, 2022
0d91cf6
Add relative threshold while testing inversing (#8973)
Yipeng1994 Aug 20, 2022
68f932c
add double grad for div (#8877)
pingzhuu Aug 20, 2022
75fb926
Dev fix bug of linespace (#8941)
Flowingsun007 Aug 20, 2022
4956899
Split axis according to grouped axises (#8919)
Yipeng1994 Aug 20, 2022
b0369f9
Add stack series api (#8901)
marigoold Aug 20, 2022
cd96862
fix requires_grad check and precison in higher derivative (#8963)
pingzhuu Aug 20, 2022
e4f6666
Lx add profile (#8889)
laoliu97 Aug 21, 2022
88035cc
Add dropout1d/2d/3d api (#8880)
BBuf Aug 21, 2022
e3a9b89
Fix scalar logical equal of bool (#8924)
Flowingsun007 Aug 21, 2022
5259a7c
Refactor ccl send and recv (#8855)
clackhan Aug 21, 2022
7040a48
Refine elementwise.cuh (#8968)
liujuncheng Aug 22, 2022
854e6a7
Add lib path for KINETO (#8950)
wyg1997 Aug 22, 2022
931d3b5
Merge branch 'zj/develop'
Ikkyu321 Aug 23, 2022
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
2 changes: 1 addition & 1 deletion .github/actions/whl/action.yml
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@ inputs:
default: "10.2"
python_version:
description: "python_version"
default: "3.6"
default: "3.8"
extra_flags:
description: "flags like --xla"
default: ""
Expand Down
3 changes: 1 addition & 2 deletions .github/workflows/canary.yml
Original file line number Diff line number Diff line change
Expand Up @@ -55,7 +55,7 @@ jobs:
- name: Checkout Oneflow-Inc/oneflow
if: ${{ github.event.inputs.oneflow-ref == '' }}
uses: actions/checkout@v2
- uses: Oneflow-Inc/get-oneflow@single-matrix-for-efficiency
- uses: Oneflow-Inc/get-oneflow@support-iree-ci
name: Build manylinux
id: build-cuda
with:
Expand All @@ -73,7 +73,6 @@ jobs:
retry-failed-build: true
clean-ccache: true
python-versions: |
3.6
3.7
3.8
- name: Upload wheelhouse
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/on_merge.yml
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,6 @@ jobs:
if: github.event.pull_request.merged == true
runs-on: ubuntu-latest
steps:
- uses: Oneflow-Inc/get-oneflow/update-benchmark-history@single-matrix-for-efficiency
- uses: Oneflow-Inc/get-oneflow/update-benchmark-history@support-iree-ci
name: Update benchmark history
timeout-minutes: 10
12 changes: 5 additions & 7 deletions .github/workflows/release.yml
Original file line number Diff line number Diff line change
Expand Up @@ -33,7 +33,7 @@ jobs:
with:
ref: ${{ github.event.pull_request.head.sha }}
repository: ${{github.event.pull_request.head.repo.full_name}}
- uses: Oneflow-Inc/get-oneflow/cache-complete/matrix/build@single-matrix-for-efficiency
- uses: Oneflow-Inc/get-oneflow/cache-complete/matrix/build@support-iree-ci
name: find cache
id: find-cache
timeout-minutes: 5
Expand All @@ -45,7 +45,7 @@ jobs:
release
oneflow-src: ${{ env.ONEFLOW_SRC }}
entries: |
cu115
cu116
cu112
cu102
cpu
Expand All @@ -71,10 +71,10 @@ jobs:
- name: Install dependencies
run: |
python3 -m pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
python3 -m pip install -U pip setuptools wheel --user
python3 -m pip install -U setuptools wheel --user
python3 -m pip install oss2 --user
- uses: actions/checkout@v2
- uses: Oneflow-Inc/get-oneflow@single-matrix-for-efficiency
- uses: Oneflow-Inc/get-oneflow@support-iree-ci
name: Build ${{ matrix.entry }}
if: ${{ matrix.entry !='cpu' }}
with:
Expand All @@ -93,12 +93,11 @@ jobs:
clean-ccache: true
nightly: ${{ github.event_name == 'schedule' }}
python-versions: |
3.6
3.7
3.8
3.9
3.10
- uses: Oneflow-Inc/get-oneflow@single-matrix-for-efficiency
- uses: Oneflow-Inc/get-oneflow@support-iree-ci
name: Build ${{ matrix.entry }}
if: ${{ matrix.entry =='cpu' }}
with:
Expand All @@ -117,7 +116,6 @@ jobs:
clean-ccache: false
nightly: ${{ github.event_name == 'schedule' || github.ref == 'refs/heads/master'}}
python-versions: |
3.6
3.7
3.8
3.9
Expand Down
4 changes: 2 additions & 2 deletions .github/workflows/simple.yml
Original file line number Diff line number Diff line change
Expand Up @@ -245,7 +245,7 @@ jobs:
repository: Oneflow-Inc/conda-env
ref: 30a7f00eb48ee9009d85a848e720823e5054c66b
path: conda-env
- uses: Oneflow-Inc/get-oneflow@single-matrix-for-efficiency
- uses: Oneflow-Inc/get-oneflow@support-iree-ci
name: Build with gcc7
if: ${{ matrix.build-type == 'gcc7'}}
with:
Expand All @@ -254,7 +254,7 @@ jobs:
oneflow-build-env: conda
conda-env-file: conda-env/dev/gcc7/environment-v2.yml
conda-env-name: oneflow-dev-gcc7-v2
- uses: Oneflow-Inc/get-oneflow@single-matrix-for-efficiency
- uses: Oneflow-Inc/get-oneflow@support-iree-ci
name: Build with clang10
if: ${{ matrix.build-type == 'clang10'}}
with:
Expand Down
Loading