[JIT] Add another decomposition api #76493

eellison · 2022-04-27T22:02:16Z

Needed for functorch schema invocation

facebook-github-bot · 2022-04-27T22:02:22Z

🔗 Helpful links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/76493
📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓Need help or want to give feedback on the CI? Visit our office hours
↩️ [fb-only] Re-run with SSH instructions

💊 CI failures summary and remediations

As of commit 4127a62 (more details on the Dr. CI page):

Expand to see more

2/2 failures introduced in this PR

🕵️ 2 new failures recognized by patterns

The following CI failures do not appear to be due to upstream breakages

pull / linux-bionic-rocm5.0-py3.7 / test (default, 2, 2, linux.rocm.gpu) (1/2)

Step: "Teardown ROCm" (full log | diagnosis details | 🔁 rerun)

2022-04-28T04:29:34.9839518Z AssertionError: can only test a child process

2022-04-28T04:29:34.9808509Z   File "/opt/conda/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 1387, in __del__
2022-04-28T04:29:34.9812365Z     if w.is_alive():
2022-04-28T04:29:34.9813344Z   File "/opt/conda/lib/python3.7/multiprocessing/process.py", line 151, in is_alive
2022-04-28T04:29:34.9816411Z     assert self._parent_pid == os.getpid(), 'can only test a child process'
2022-04-28T04:29:34.9817457Z AssertionError: can only test a child process
2022-04-28T04:29:34.9820248Z     self._shutdown_workers()
2022-04-28T04:29:34.9823194Z   File "/opt/conda/lib/python3.7/site-packages/torch/utils/data/dataloader.py", line 1370, in _shutdown_workers
2022-04-28T04:29:34.9835608Z     if w.is_alive():
2022-04-28T04:29:34.9836762Z   File "/opt/conda/lib/python3.7/multiprocessing/process.py", line 151, in is_alive
2022-04-28T04:29:34.9838531Z     assert self._parent_pid == os.getpid(), 'can only test a child process'
2022-04-28T04:29:34.9839518Z AssertionError: can only test a child process
2022-04-28T04:29:35.0170797Z ok (0.057s)
2022-04-28T04:29:35.0204213Z   test_multiple_dataloaders (__main__.TestDataLoaderPersistentWorkers) ... skip: Test is disabled because an issue exists disabling it: https://github.com/pytorch/pytorch/issues/74598 for allplatform(s) . If you're seeing this on your local machine and would like to enable this test, please make sure IN_CI is not set and you are not using the flag --import-disabled-tests. (0.003s)
2022-04-28T04:29:36.9145960Z   test_multiprocessing_contexts (__main__.TestDataLoaderPersistentWorkers) ... [W CudaIPCTypes.cpp:15] Producer process has been terminated before all shared CUDA tensors released. See Note [Sharing CUDA tensors]
2022-04-28T04:29:37.0745274Z [W CudaIPCTypes.cpp:15] Producer process has been terminated before all shared CUDA tensors released. See Note [Sharing CUDA tensors]
2022-04-28T04:29:37.1188406Z [W CudaIPCTypes.cpp:15] Producer process has been terminated before all shared CUDA tensors released. See Note [Sharing CUDA tensors]
2022-04-28T04:29:40.8591456Z [W CudaIPCTypes.cpp:15] Producer process has been terminated before all shared CUDA tensors released. See Note [Sharing CUDA tensors]
2022-04-28T04:29:40.8761599Z [W CudaIPCTypes.cpp:15] Producer process has been terminated before all shared CUDA tensors released. See Note [Sharing CUDA tensors]
2022-04-28T04:29:40.8889783Z [W CudaIPCTypes.cpp:15] Producer process has been terminated before all shared CUDA tensors released. See Note [Sharing CUDA tensors]
2022-04-28T04:29:44.8006872Z ok (9.780s)
2022-04-28T04:29:44.8051774Z   test_multiprocessing_iterdatapipe (__main__.TestDataLoaderPersistentWorkers) ... skip: Test is disabled because an issue exists disabling it: https://github.com/pytorch/pytorch/issues/74498 for allplatform(s) . If you're seeing this on your local machine and would like to enable this test, please make sure IN_CI is not set and you are not using the flag --import-disabled-tests. (0.004s)

pull / linux-bionic-rocm5.0-py3.7 / test (default, 1, 2, linux.rocm.gpu) (2/2)

Step: "Teardown ROCm" (full log | diagnosis details | 🔁 rerun)

2022-04-28T03:46:13.7667280Z test_forward_mod...meError: bias tensor has to be contiguous (0.000s)

2022-04-28T03:46:12.1846044Z   test_forward_mode_AD_nn_functional_bilinear_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.010s)
2022-04-28T03:46:12.1889610Z   test_forward_mode_AD_nn_functional_binary_cross_entropy_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.004s)
2022-04-28T03:46:12.3123614Z   test_forward_mode_AD_nn_functional_binary_cross_entropy_with_logits_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.123s)
2022-04-28T03:46:12.3168107Z   test_forward_mode_AD_nn_functional_celu_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.004s)
2022-04-28T03:46:12.6412931Z   test_forward_mode_AD_nn_functional_conv1d_cuda_complex128 (__main__.TestGradientsCUDA) ... ok (0.324s)
2022-04-28T03:46:12.6783539Z   test_forward_mode_AD_nn_functional_conv1d_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.037s)
2022-04-28T03:46:13.5976523Z   test_forward_mode_AD_nn_functional_conv2d_cuda_complex128 (__main__.TestGradientsCUDA) ... ok (0.919s)
2022-04-28T03:46:13.6973602Z   test_forward_mode_AD_nn_functional_conv2d_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.100s)
2022-04-28T03:46:13.7320325Z   test_forward_mode_AD_nn_functional_conv_transpose1d_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.035s)
2022-04-28T03:46:13.7663127Z   test_forward_mode_AD_nn_functional_conv_transpose2d_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.034s)
2022-04-28T03:46:13.7667280Z   test_forward_mode_AD_nn_functional_conv_transpose3d_cuda_float64 (__main__.TestGradientsCUDA) ... skip: Skipped! RuntimeError: bias tensor has to be contiguous (0.000s)
2022-04-28T03:46:13.8365259Z   test_forward_mode_AD_nn_functional_cosine_embedding_loss_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.070s)
2022-04-28T03:46:13.8949402Z   test_forward_mode_AD_nn_functional_cosine_similarity_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.058s)
2022-04-28T03:46:13.9605486Z   test_forward_mode_AD_nn_functional_cross_entropy_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.066s)
2022-04-28T03:46:13.9667657Z   test_forward_mode_AD_nn_functional_ctc_loss_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.006s)
2022-04-28T03:46:13.9783175Z   test_forward_mode_AD_nn_functional_dropout2d_cuda_float64 (__main__.TestGradientsCUDA) ... expected failure (0.011s)
2022-04-28T03:46:14.0061583Z   test_forward_mode_AD_nn_functional_dropout_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.028s)
2022-04-28T03:46:14.0105641Z   test_forward_mode_AD_nn_functional_elu_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.004s)
2022-04-28T03:46:14.0158312Z   test_forward_mode_AD_nn_functional_embedding_bag_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.005s)
2022-04-28T03:46:14.0196367Z   test_forward_mode_AD_nn_functional_embedding_cuda_float64 (__main__.TestGradientsCUDA) ... ok (0.004s)
2022-04-28T03:46:14.0291828Z   test_forward_mode_AD_nn_functional_feature_alpha_dropout_with_train_cuda_float64 (__main__.TestGradientsCUDA) ... expected failure (0.009s)

This comment was automatically generated by Dr. CI (expand for details).

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

davidberard98

LGTM!

Chillee

LGTM!

eellison · 2022-04-28T00:32:36Z

@pytochbot merge this

eellison · 2022-04-28T02:15:05Z

@pytorchbot merge this

github-actions · 2022-04-28T02:17:04Z

Hey @eellison.
You've committed this PR, but it does not have both a 'release notes: ...' and 'topics: ...' label. Please add one of each to the PR. The 'release notes: ...' label should represent the part of PyTorch that this PR changes (fx, autograd, distributed, etc) and the 'topics: ...' label should represent the kind of PR it is (not user facing, new feature, bug fix, perf improvement, etc). The list of valid labels can be found here for the 'release notes: ...' and here for the 'topics: ...'.
For changes that are 'topic: not user facing' there is no need for a release notes label.

osalpekar · 2022-04-28T03:29:49Z

@pytorchbot revert this please (need to revert this to cleanly revert #76397, which is the actual breaking change)

This reverts commit 32cb037. Reverted #76493 on behalf of https://github.com/osalpekar

Reland of #76397 and #76493 This time I'll get it right 😢 Pull Request resolved: #76539 Approved by: https://github.com/davidberard98, https://github.com/osalpekar

Summary: Reland of #76397 and #76493 This time I'll get it right 😢 Pull Request resolved: #76539 Approved by: https://github.com/davidberard98, https://github.com/osalpekar Test Plan: contbuild & OSS CI, see https://hud.pytorch.org/commit/pytorch/pytorch/e5a55af3056fb06c2f5b570b9ed818ed8fb9366c Reviewed By: osalpekar Differential Revision: D36042323 fbshipit-source-id: 7eefcf6cf142bbbc9509e379e6900de84d5b33e4

Add api

4127a62

facebook-github-bot added the cla signed label Apr 27, 2022

facebook-github-bot added the oncall: jit Add this issue/PR to JIT oncall triage queue label Apr 27, 2022

eellison mentioned this pull request Apr 27, 2022

Add trace decomposition, invoke from C++ pytorch/functorch#740

Merged

davidberard98 approved these changes Apr 27, 2022

View reviewed changes

Chillee approved these changes Apr 27, 2022

View reviewed changes

pytorchmergebot closed this in 32cb037 Apr 28, 2022

pytorchmergebot added the Reverted label Apr 28, 2022

pytorchmergebot added a commit that referenced this pull request Apr 28, 2022

Revert "[JIT] Add another decomposition api"

2af0c42

This reverts commit 32cb037. Reverted #76493 on behalf of https://github.com/osalpekar

eellison mentioned this pull request Apr 28, 2022

Reland reland #76539

Closed

pytorchmergebot pushed a commit that referenced this pull request Apr 28, 2022

Reland reland

e5a55af

Reland of #76397 and #76493 This time I'll get it right 😢 Pull Request resolved: #76539 Approved by: https://github.com/davidberard98, https://github.com/osalpekar

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[JIT] Add another decomposition api #76493

[JIT] Add another decomposition api #76493

Uh oh!

eellison commented Apr 27, 2022 •

edited

Loading

Uh oh!

facebook-github-bot commented Apr 27, 2022 •

edited

Loading

🕵️ 2 new failures recognized by patterns

pull / linux-bionic-rocm5.0-py3.7 / test (default, 2, 2, linux.rocm.gpu) (1/2)

pull / linux-bionic-rocm5.0-py3.7 / test (default, 1, 2, linux.rocm.gpu) (2/2)

Uh oh!

davidberard98 left a comment

Uh oh!

Chillee left a comment

Uh oh!

eellison commented Apr 28, 2022

Uh oh!

eellison commented Apr 28, 2022

Uh oh!

github-actions bot commented Apr 28, 2022

Uh oh!

osalpekar commented Apr 28, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

[JIT] Add another decomposition api #76493

[JIT] Add another decomposition api #76493

Uh oh!

Conversation

eellison commented Apr 27, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

facebook-github-bot commented Apr 27, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful links

💊 CI failures summary and remediations

🕵️ 2 new failures recognized by patterns

pull / linux-bionic-rocm5.0-py3.7 / test (default, 2, 2, linux.rocm.gpu) (1/2)

pull / linux-bionic-rocm5.0-py3.7 / test (default, 1, 2, linux.rocm.gpu) (2/2)

Uh oh!

davidberard98 left a comment

Choose a reason for hiding this comment

Uh oh!

Chillee left a comment

Choose a reason for hiding this comment

Uh oh!

eellison commented Apr 28, 2022

Uh oh!

eellison commented Apr 28, 2022

Uh oh!

github-actions bot commented Apr 28, 2022

Uh oh!

osalpekar commented Apr 28, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

eellison commented Apr 27, 2022 •

edited

Loading

facebook-github-bot commented Apr 27, 2022 •

edited

Loading