[CI] Move CUDA-11.6 to Python-3.10 configuration by malfet · Pull Request #81233 · pytorch/pytorch

malfet · 2022-07-11T13:44:16Z

Second attempt of landing the change after #66530

Skip nan hashes comparison validation in jit/test_hash.py, as it behaves differently in 3.10 vs other pythons
Skip tensor_fx assert tests
Skip initializing uint8 tensors from negative values in TestScript.test_torch_tensor_as_tensor

Final step in closing #66424

facebook-github-bot · 2022-07-11T13:44:24Z

🔗 Helpful links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/81233
📄 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

✅ No Failures (0 Pending)

As of commit 023fbf5 (more details on the Dr. CI page):

Expand to see more

💚 💚 Looks good so far! There are no failures yet. 💚 💚

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.

malfet · 2022-07-11T17:00:39Z

Will have to disable most of test_fx for py-3.10 (as they are incompatible right now)

Copy-n-paste the list from https://github.com/PyCQA/isort/blob/main/isort/stdlibs/py310.py Tested locally and in #81233 Pull Request resolved: #81261 Approved by: https://github.com/suo

Summary: Copy-n-paste the list from https://github.com/PyCQA/isort/blob/main/isort/stdlibs/py310.py Tested locally and in #81233 Pull Request resolved: #81261 Approved by: https://github.com/suo Test Plan: contbuild & OSS CI, see https://hud.pytorch.org/commit/pytorch/pytorch/9ed76c8c89f6875f36619cd27fdd4af01d765a5b Reviewed By: DanilBaibak Differential Revision: D37781957 Pulled By: malfet fbshipit-source-id: e39d94335950022fbdbe7b053674136694b89fad

Thus avoiding `TypeError: 'float' object cannot be interpreted as an integer` when trying to create integer tensor from floating point values Use `c10::checked_convert` to detect overflows during tensor construction from scalars. Modify sparse_csr test that violated this rule Fixes #69319 Tested in #81233 Pull Request resolved: #81372 Approved by: https://github.com/ezyang, https://github.com/ngimel

Second attempt of landing the change after #66530

As hashes of nans are not equal in 3.10

…st_torch_tensor_as_tensor

malfet · 2022-07-16T19:41:55Z

@pytorchbot merge

malfet · 2022-07-16T20:06:39Z

@pytorchbot merge -f

pytorchmergebot · 2022-07-16T20:40:59Z

@pytorchbot successfully started a merge job. Check the current status here

janeyx99 · 2022-07-18T17:14:29Z

@pytorchbot revert -m "this should have been reverted along with 81372 for breaking internal builds" -c "ghfirst"

pytorchmergebot · 2022-07-18T17:15:47Z

@pytorchbot successfully started a revert job. Check the current status here

pytorchmergebot · 2022-07-18T17:15:53Z

@malfet your PR has been successfully reverted.

This reverts commit 7ccf693. Reverted #81233 on behalf of https://github.com/janeyx99 due to this should have been reverted along with 81372 for breaking internal builds

Summary: Second attempt of landing the change after #66530 Skip nan hashes comparison validation in `jit/test_hash.py`, as it behaves differently in 3.10 vs other pythons Skip tensor_fx assert tests Skip initializing uint8 tensors from negative values in `TestScript.test_torch_tensor_as_tensor` Final step in closing #66424 Pull Request resolved: #81233 Approved by: https://github.com/seemethere Test Plan: contbuild & OSS CI, see https://hud.pytorch.org/commit/pytorch/pytorch/7ccf693cf6ca28a215d7a79b45cf6f47dfd22029 Reviewed By: DanilBaibak Differential Revision: D37919654 Pulled By: malfet fbshipit-source-id: 20d2cc1258e7411a669d96ad7b423b92f4f8593b

Summary: Copy-n-paste the list from https://github.com/PyCQA/isort/blob/main/isort/stdlibs/py310.py Tested locally and in pytorch#81233 Pull Request resolved: pytorch#81261 Approved by: https://github.com/suo Test Plan: contbuild & OSS CI, see https://hud.pytorch.org/commit/pytorch/pytorch/9ed76c8c89f6875f36619cd27fdd4af01d765a5b Reviewed By: DanilBaibak Differential Revision: D37781957 Pulled By: malfet fbshipit-source-id: e39d94335950022fbdbe7b053674136694b89fad

Summary: Copy-n-paste the list from https://github.com/PyCQA/isort/blob/main/isort/stdlibs/py310.py Tested locally and in #81233 Pull Request resolved: #81261 Approved by: https://github.com/suo Test Plan: contbuild & OSS CI, see https://hud.pytorch.org/commit/pytorch/pytorch/9ed76c8c89f6875f36619cd27fdd4af01d765a5b Reviewed By: DanilBaibak Differential Revision: D37781957 Pulled By: malfet fbshipit-source-id: e39d94335950022fbdbe7b053674136694b89fad Co-authored-by: Nikita Shulga (Meta Employee) <[email protected]>

Thus avoiding `TypeError: 'float' object cannot be interpreted as an integer` when trying to create integer tensor from floating point values Use `c10::checked_convert` to detect overflows during tensor construction from scalars. Modify sparse_csr test that violated this rule Fixes #69319 Tested in #81233 Pull Request resolved: #81372 Approved by: https://github.com/ezyang, https://github.com/ngimel (cherry picked from commit 69d7334)

Second attempt of landing the change after #66530 Skip nan hashes comparison validation in `jit/test_hash.py`, as it behaves differently in 3.10 vs other pythons Skip tensor_fx assert tests Skip initializing uint8 tensors from negative values in `TestScript.test_torch_tensor_as_tensor` Final step in closing #66424 Pull Request resolved: #81233 Approved by: https://github.com/seemethere (cherry picked from commit 7ccf693)

This is a re-land of #81372 and #81233 with the exception that it does not force the range-checks on older Python runtime versions and as such should not affect the internal workloads, which were the reason for revert, see #81372 (comment) - [Py3.10] Allow floats to be imported as Long (#81372) - [CI] Move CUDA-11.6 to Python-3.10 configuration (#81233) - Don't do anything about range checks for pre-py3.10 Pull Request resolved: #82329 Approved by: https://github.com/kit1980

Summary: This is a re-land of #81372 and #81233 with the exception that it does not force the range-checks on older Python runtime versions and as such should not affect the internal workloads, which were the reason for revert, see #81372 (comment) - [Py3.10] Allow floats to be imported as Long (#81372) - [CI] Move CUDA-11.6 to Python-3.10 configuration (#81233) - Don't do anything about range checks for pre-py3.10 Pull Request resolved: #82329 Approved by: https://github.com/kit1980 Test Plan: contbuild & OSS CI, see https://hud.pytorch.org/commit/pytorch/pytorch/d80fe49de09399e77a384bd9573c7bb69887dd20 Reviewed By: osalpekar Differential Revision: D38234293 Pulled By: malfet fbshipit-source-id: 3f7ffe729ab2a3ef692bc95f5398dfdf395b2397

malfet requested review from a team and jeffdaily as code owners July 11, 2022 13:44

facebook-github-bot added the cla signed label Jul 11, 2022

seemethere approved these changes Jul 11, 2022

View reviewed changes

malfet force-pushed the malfet/add-py310-config branch 2 times, most recently from c7e0b81 to f55061a Compare July 11, 2022 15:15

malfet mentioned this pull request Jul 11, 2022

Add 3.10 stdlib to torch.package #81261

Closed

malfet force-pushed the malfet/add-py310-config branch 2 times, most recently from a4dc7b7 to 420a2fa Compare July 12, 2022 13:32

malfet requested review from mruberry and ngimel as code owners July 12, 2022 18:19

malfet force-pushed the malfet/add-py310-config branch from ece3d74 to 8733d1c Compare July 12, 2022 22:59

malfet mentioned this pull request Jul 12, 2022

[Py3.10] Allow floats to be imported as Long #81372

Closed

malfet force-pushed the malfet/add-py310-config branch from 49f39fe to b0d3c9f Compare July 15, 2022 17:19

malfet added 3 commits July 15, 2022 16:25

[CI] Move CUDA-11.6 to Python-3.10 configuration

b1c188c

Second attempt of landing the change after #66530

Fix test_fx for 3.10

08033e1

Skip nan comparison from test_hash_float

e7220c2

As hashes of nans are not equal in 3.10

malfet force-pushed the malfet/add-py310-config branch from b0d3c9f to e7220c2 Compare July 15, 2022 23:25

malfet added 2 commits July 15, 2022 18:28

Skip initializing uint8 tensors from negative values in TestScript.te…

2d9ce7f

…st_torch_tensor_as_tensor

Fix lint

023fbf5

malfet added the topic: not user facing topic category label Jul 16, 2022

pytorchmergebot added the Merged label Jul 16, 2022

pytorchmergebot closed this in 7ccf693 Jul 16, 2022

malfet deleted the malfet/add-py310-config branch July 17, 2022 18:12

pytorchmergebot added the Reverted label Jul 18, 2022

atalman mentioned this pull request Jul 21, 2022

Add 3.10 stdlib to torch.package (#81261) (#81261) #81879

Merged

janeyx99 mentioned this pull request Jul 22, 2022

[Meta] CI Revert Tracker #66178

Closed

malfet mentioned this pull request Jul 27, 2022

[Reland] Add py-3.10 config #82329

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[CI] Move CUDA-11.6 to Python-3.10 configuration#81233

[CI] Move CUDA-11.6 to Python-3.10 configuration#81233
malfet wants to merge 5 commits intomasterfrom
malfet/add-py310-config

malfet commented Jul 11, 2022 •

edited

Loading

Uh oh!

facebook-github-bot commented Jul 11, 2022 •

edited

Loading

Uh oh!

malfet commented Jul 11, 2022

Uh oh!

malfet commented Jul 16, 2022

Uh oh!

malfet commented Jul 16, 2022

Uh oh!

pytorchmergebot commented Jul 16, 2022

Uh oh!

janeyx99 commented Jul 18, 2022

Uh oh!

pytorchmergebot commented Jul 18, 2022

Uh oh!

pytorchmergebot commented Jul 18, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Conversation

malfet commented Jul 11, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

facebook-github-bot commented Jul 11, 2022 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful links

✅ No Failures (0 Pending)

Uh oh!

malfet commented Jul 11, 2022

Uh oh!

malfet commented Jul 16, 2022

Uh oh!

malfet commented Jul 16, 2022

Uh oh!

pytorchmergebot commented Jul 16, 2022

Uh oh!

janeyx99 commented Jul 18, 2022

Uh oh!

pytorchmergebot commented Jul 18, 2022

Uh oh!

pytorchmergebot commented Jul 18, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

malfet commented Jul 11, 2022 •

edited

Loading

facebook-github-bot commented Jul 11, 2022 •

edited

Loading