Skip to content

[PyTorch] IValue::toTensor can now return const Tensor& #48868

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.

Sign up for GitHub

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

Jump to bottom
Closed
wants to merge 13 commits into from
Closed

[PyTorch] IValue::toTensor can now return const Tensor& #48868

wants to merge 13 commits into from

Conversation

swolchok
Copy link
Contributor

@swolchok swolchok commented Dec 4, 2020 •
edited
Loading

Stack from ghstack:

Building on the previous diff, we can make toTensor() return a
const Tensor&, which should make it easier to avoid reference
counting.

Differential Revision: D25325379

NOTE FOR REVIEWERS: This PR has internal Facebook specific changes or comments, please review them on Phabricator!

@swolchok
[PyTorch] IValue::toTensor can now return const Tensor&
3be0274 
Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
@facebook-github-bot facebook-github-bot added cla signed oncall: jit Add this issue/PR to JIT oncall triage queue labels Dec 4, 2020
@swolchok swolchok mentioned this pull request Dec 4, 2020
Closed
@dr-ci
Copy link

dr-ci bot commented Dec 4, 2020 •
edited by facebook-github-bot
Loading

💊 CI failures summary and remediations

As of commit 1325e5d (more details on the Dr. CI page):

  • 2/2 failures possibly* introduced in this PR
    • 1/2 non-CircleCI failure(s)

🕵️ 1 new failure recognized by patterns

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

See CircleCI build pytorch_linux_bionic_py3_8_gcc9_coverage_test1 (1/1)

Step: "Run tests" (full log | diagnosis details | 🔁 rerun)

Jan 05 01:22:24 [E request_callback_no_python.cpp:636] Received error while processing request type 258: RuntimeError: Can not pickle torch.futures.Future 
Jan 05 01:22:24 At:
Jan 05 01:22:24   /opt/conda/lib/python3.8/site-packages/torch/distributed/rpc/internal.py(120): serialize
Jan 05 01:22:24   /opt/conda/lib/python3.8/site-packages/torch/distributed/rpc/internal.py(172): serialize
Jan 05 01:22:24 
Jan 05 01:22:24 [E request_callback_no_python.cpp:636] Received error while processing request type 258: RuntimeError: Can not pickle torch.futures.Future
Jan 05 01:22:24 
Jan 05 01:22:24 At:
Jan 05 01:22:24   /opt/conda/lib/python3.8/site-packages/torch/distributed/rpc/internal.py(120): serialize
Jan 05 01:22:24   /opt/conda/lib/python3.8/site-packages/torch/distributed/rpc/internal.py(172): serialize
Jan 05 01:22:24 
Jan 05 01:22:24 [E request_callback_no_python.cpp:636] Received error while processing request type 258: RuntimeError: Can not pickle torch.futures.Future
Jan 05 01:22:24 
Jan 05 01:22:24 At:
Jan 05 01:22:24   /opt/conda/lib/python3.8/site-packages/torch/distributed/rpc/internal.py(120): serialize
Jan 05 01:22:24   /opt/conda/lib/python3.8/site-packages/torch/distributed/rpc/internal.py(172): serialize
Jan 05 01:22:24 
Jan 05 01:22:24 [W tensorpipe_agent.cpp:547] RPC agent for worker0 encountered error when reading incoming request from worker3: EOF: end of file (this is expected to happen during shutdown)
Jan 05 01:22:24 [W tensorpipe_agent.cpp:547] RPC agent for worker0 encountered error when reading incoming request from worker2: EOF: end of file (this is expected to happen during shutdown)
Jan 05 01:22:25 ok (2.551s)
Jan 05 01:22:27   test_return_future_remote (__main__.TensorPipeRpcTestWithSpawn) ... [W tensorpipe_agent.cpp:547] RPC agent for worker1 encountered error when reading incoming request from worker0: EOF: end of file (this is expected to happen during shutdown)
Jan 05 01:22:27 ok (2.554s)

1 job timed out:

  • pytorch_linux_bionic_py3_8_gcc9_coverage_test1
This comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.

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

This comment has been revised 92 times.

@swolchok
remove unnecessary std::move on "[PyTorch] IValue::toTensor can now r…
da70154 
…eturn const Tensor&"

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 4, 2020
@swolchok
[PyTorch] IValue::toTensor can now return const Tensor&
465a113 
Pull Request resolved: #48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 117914173

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!
@swolchok
Update on "[PyTorch] IValue::toTensor can now return const Tensor&"
03771d7 
Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 12, 2020
@swolchok
[PyTorch] IValue::toTensor can now return const Tensor&
e16a0d6 
Pull Request resolved: #48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 118468561

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!
@swolchok
Update on "[PyTorch] IValue::toTensor can now return const Tensor&"
2c2637f 
Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 15, 2020
@swolchok
[PyTorch] IValue::toTensor can now return const Tensor&
50348fd 
Pull Request resolved: #48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 118658273

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!
@swolchok
Remove Elvis operator. Fix ASAN violation. on "[PyTorch] IValue::toTe…
4199e87 
…nsor can now return const Tensor&"

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
@swolchok
Remove Elvis operator. Fix ASAN violation. on "[PyTorch] IValue::toTe…
cde257a 
…nsor can now return const Tensor&"

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 16, 2020
@swolchok
[PyTorch] IValue::toTensor can now return const Tensor&
88b863f 
Pull Request resolved: #48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 118718144

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!
@swolchok
Update on "[PyTorch] IValue::toTensor can now return const Tensor&"
a8f5b14 
Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 16, 2020
@swolchok
[PyTorch] IValue::toTensor can now return const Tensor&
72ac9c8 
Pull Request resolved: #48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 118754514

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!
@swolchok
rebase, add a few more tests on "[PyTorch] IValue::toTensor can now r…
946b780 
…eturn const Tensor&"

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 18, 2020
@swolchok
[PyTorch] IValue::toTensor can now return const Tensor&
7394923 
Pull Request resolved: #48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 118936155

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!
@swolchok
fix isAliasOf and isPtrType on "[PyTorch] IValue::toTensor can now re…
04d59a0 
…turn const Tensor&"

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 19, 2020
@swolchok
[PyTorch] IValue::toTensor can now return const Tensor&
519b99f 
Pull Request resolved: #48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 118965482

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!
@swolchok
Update on "[PyTorch] IValue::toTensor can now return const Tensor&"
3568ca5 
Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
@swolchok swolchok mentioned this pull request Dec 22, 2020
Closed
swolchok added a commit that referenced this pull request Dec 22, 2020
@swolchok
[PyTorch] IValue::toTensor can now return const Tensor&
8e47669 
Pull Request resolved: #48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 119023481

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!
@swolchok
add another isPtrType() test & fix it on "[PyTorch] IValue::toTensor …
a79c5be 
…can now return const Tensor&"

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 22, 2020
@swolchok
[PyTorch] IValue::toTensor can now return const Tensor&
9a86459 
Pull Request resolved: #48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 119032050

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!
@swolchok
Update on "[PyTorch] IValue::toTensor can now return const Tensor&"
216ad53 
Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Dec 22, 2020
@swolchok
[PyTorch] IValue::toTensor can now return const Tensor&
f4192dc 
Pull Request resolved: #48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 119032050

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!
@swolchok
Update on "[PyTorch] IValue::toTensor can now return const Tensor&"
1325e5d 
Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!

[ghstack-poisoned]
swolchok added a commit that referenced this pull request Jan 4, 2021
@swolchok
[PyTorch] IValue::toTensor can now return const Tensor&
d74cc93 
Pull Request resolved: #48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 119327372

Differential Revision: [D25325379](https://our.internmc.facebook.com/intern/diff/D25325379/)

**NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D25325379/)!
@swolchok swolchok mentioned this pull request Jan 5, 2021
Closed
@facebook-github-bot
Copy link
Contributor

This pull request has been merged in a440491.

facebook-github-bot pushed a commit that referenced this pull request Jan 6, 2021
@swolchok @facebook-github-bot
[PyTorch] IValue::toTensor can now return const Tensor& (#48868)
480a756 
Summary:
Pull Request resolved: #48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 119327372

Test Plan: internal benchmarks.

Reviewed By: bwasti

Differential Revision: D25325379

fbshipit-source-id: ca699632901691bcee432f595f75b0a4416d55dd
@facebook-github-bot facebook-github-bot deleted the gh/swolchok/33/head branch January 10, 2021 15:17
hwangdeyu pushed a commit to hwangdeyu/pytorch that referenced this pull request Jan 14, 2021
@swolchok
[PyTorch] IValue::toTensor can now return const Tensor& (pytorch#48868)
de28db8 
Summary:
Pull Request resolved: pytorch#48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 119327372

Test Plan: internal benchmarks.

Reviewed By: bwasti

Differential Revision: D25325379

fbshipit-source-id: ca699632901691bcee432f595f75b0a4416d55dd
jjsjann123 pushed a commit to jjsjann123/nvfuser that referenced this pull request Oct 29, 2022
@swolchok @facebook-github-bot
[PyTorch] IValue::toTensor can now return const Tensor& (#48868)
99e07a9 
Summary:
Pull Request resolved: pytorch/pytorch#48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 119327372

Test Plan: internal benchmarks.

Reviewed By: bwasti

Differential Revision: D25325379

fbshipit-source-id: ca699632901691bcee432f595f75b0a4416d55dd
jjsjann123 pushed a commit to jjsjann123/nvfuser that referenced this pull request Nov 10, 2022
@swolchok @facebook-github-bot
[PyTorch] IValue::toTensor can now return const Tensor& (#48868)
3baa577 
Summary:
Pull Request resolved: pytorch/pytorch#48868

Building on the previous diff, we can make `toTensor()` return a
`const Tensor&`, which should make it easier to avoid reference
counting.
ghstack-source-id: 119327372

Test Plan: internal benchmarks.

Reviewed By: bwasti

Differential Revision: D25325379

fbshipit-source-id: ca699632901691bcee432f595f75b0a4416d55dd
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@bwasti bwasti Awaiting requested review from bwasti

@smessmer smessmer Awaiting requested review from smessmer

@ZolotukhinM ZolotukhinM Awaiting requested review from ZolotukhinM

@hlu1 hlu1 Awaiting requested review from hlu1

Assignees
No one assigned
Labels
cla signed Merged oncall: jit Add this issue/PR to JIT oncall triage queue
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@swolchok @facebook-github-bot