[c10d] Remove Option for ProcessGroup and Expose backend Options to reflect the correct code structure #132931
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/132931
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit 13359b2 with merge base 55236d0 ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
…structure" What I noticed is that we still assume PGNCCL is using option from ProcessGroup but we actually should make it use Option from backend class. cc XilunWu H-Huang awgu kwen2501 wanchaol fegin wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
H-Huang
reviewed
Aug 8, 2024
…structure" Although it looks like documents for Options mentioned in #130958 have been all addressed. What I noticed is that we still assume PGNCCL is using option from ProcessGroup but we actually should make it use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" cc XilunWu H-Huang awgu kwen2501 wanchaol fegin wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
…ptions to reflect the correct code structure" We introduced the dispatchable backend for a ProcessGroup and collective in #86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" cc XilunWu H-Huang awgu kwen2501 wanchaol fegin wz337 wconstab d4l3k c-p-i-o ezyang gchanan [ghstack-poisoned]
…ptions to reflect the correct code structure" We introduced the dispatchable backend for a ProcessGroup and collective in #86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" cc XilunWu H-Huang awgu kwen2501 wanchaol fegin wz337 wconstab d4l3k c-p-i-o ezyang gchanan [ghstack-poisoned]
…ptions to reflect the correct code structure" We introduced the dispatchable backend for a ProcessGroup and collective in #86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" cc XilunWu H-Huang awgu kwen2501 wanchaol fegin wz337 wconstab d4l3k c-p-i-o ezyang gchanan [ghstack-poisoned]
wz337
reviewed
Aug 28, 2024
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
|
@pytorchbot revert -c ghfirst -m "This PR is breaking builds internally due to the removal of ProcessGroup::Options" I'd recommend ghimporting this PR to make sure it passes internally before re-merging. See D62008954 for more details |
|
@pytorchbot successfully started a revert job. Check the current status here. |
|
@fduwjj your PR has been successfully reverted. |
pytorchmergebot
added a commit
that referenced
this pull request
Aug 30, 2024
…ons to reflect the correct code structure (#132931)" This reverts commit 65864d0. Reverted #132931 on behalf of https://github.com/ZainRizvi due to This PR is breaking builds internally due to the removal of ProcessGroup::Options ([comment](#132931 (comment)))
fduwjj
added a commit
that referenced
this pull request
Sep 11, 2024
…ons to reflect the correct code structure (#132931) We introduced the dispatchable backend for a ProcessGroup and collective in #86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" We need to make changes to the test to make it aligned with the change. This is try to reland D62008954 by fixing internal errors. Differential Revision: [D62483294](https://our.internmc.facebook.com/intern/diff/D62483294/) [ghstack-poisoned]
fduwjj
added a commit
that referenced
this pull request
Sep 11, 2024
…ons to reflect the correct code structure (#132931) We introduced the dispatchable backend for a ProcessGroup and collective in #86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" We need to mak 8000 e changes to the test to make it aligned with the change. This is try to reland D62008954 by fixing internal errors. Differential Revision: [D62483294](https://our.internmc.facebook.com/intern/diff/D62483294/) ghstack-source-id: 241951792 Pull Request resolved: #135653
|
reland in #135653 |
fduwjj
added a commit
that referenced
this pull request
Sep 11, 2024
…ackend Options to reflect the correct code structure (#132931)" We introduced the dispatchable backend for a ProcessGroup and collective in #86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" We need to make changes to the test to make it aligned with the change. This is try to reland D62008954 by fixing internal errors. Differential Revision: [D62483294](https://our.internmc.facebook.com/intern/diff/D62483294/) cc XilunWu H-Huang awgu kwen2501 wanchaol fegin wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
fduwjj
added a commit
that referenced
this pull request
Sep 11, 2024
…ons to reflect the correct code structure (#132931) Pull Request resolved: #135653 We introduced the dispatchable backend for a ProcessGroup and collective in #86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" We need to make changes to the test to make it aligned with the change. This is try to reland D62008954 by fixing internal errors. Differential Revision: [D62483294](https://our.internmc.facebook.com/intern/diff/D62483294/) ghstack-source-id: 242054823
fduwjj
added a commit
that referenced
this pull request
Sep 11, 2024
…ons to reflect the correct code structu 6D40 re (#132931) Pull Request resolved: #135653 We introduced the dispatchable backend for a ProcessGroup and collective in #86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" We need to make changes to the test to make it aligned with the change. This is try to reland D62008954 by fixing internal errors. ghstack-source-id: 242088446 Differential Revision: [D62483294](https://our.internmc.facebook.com/intern/diff/D62483294/)
fduwjj
added a commit
that referenced
this pull request
Sep 11, 2024
…oup and Expose backend Options to reflect the correct code structure (#132931)" We introduced the dispatchable backend for a ProcessGroup and collective in #86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" We need to make changes to the test to make it aligned with the change. This is try to reland D62008954 by fixing internal errors. Differential Revision: [D62483294](https://our.internmc.facebook.com/intern/diff/D62483294/) cc XilunWu H-Huang awgu kwen2501 wanchaol fegin wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
fduwjj
added a commit
that referenced
this pull request
Sep 11, 2024
…ackend Options to reflect the correct code structure (#132931)" We introduced the dispatchable backend for a ProcessGroup and collective in #86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" We need to make changes to the test to make it aligned with the change. This is try to reland D62008954 by fixing internal errors. Differential Revision: [D62483294](https://our.internmc.facebook.com/intern/diff/D62483294/) cc XilunWu H-Huang awgu kwen2501 wanchaol fegin wz337 wconstab d4l3k c-p-i-o [ghstack-poisoned]
tolleybot
pushed a commit
to tolleybot/pytorch
that referenced
this pull request
Sep 14, 2024
…eflect the correct code structure (pytorch#132931) We introduced the dispatchable backend for a ProcessGroup and collective in pytorch#86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" We need to make changes to the test to make it aligned with the change. Pull Request resolved: pytorch#132931 Approved by: https://github.com/H-Huang
tolleybot
pushed a commit
to tolleybot/pytorch
that referenced
this pull request
Sep 14, 2024
…ons to reflect the correct code structure (pytorch#132931)" This reverts commit 65864d0. Reverted pytorch#132931 on behalf of https://github.com/ZainRizvi due to This PR is breaking builds internally due to the removal of ProcessGroup::Options ([comment](pytorch#132931 (comment)))
pytorchmergebot
pushed a commit
that referenced
this pull request
Sep 16, 2024
…ons to reflect the correct code structure (#132931) (#135653) We introduced the dispatchable backend for a ProcessGroup and collective in #86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" We need to make changes to the test to make it aligned with the change. This is try to reland D62008954 by fixing internal errors. Differential Revision: [D62483294](https://our.internmc.facebook.com/intern/diff/D62483294/) Pull Request resolved: #135653 Approved by: https://github.com/wz337, https://github.com/H-Huang
Chao1Han
pushed a commit
to Chao1Han/pytorch
that referenced
this pull request
Sep 20, 2024
…eflect the correct code structure (pytorch#132931) We introduced the dispatchable backend for a ProcessGroup and collective in pytorch#86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" We need to make changes to the test to make it aligned with the change. Pull Request resolved: pytorch#132931 Approved by: https://github.com/H-Huang
Chao1Han
pushed a commit
to Chao1Han/pytorch
that referenced
this pull request
Sep 20, 2024
…ons to reflect the correct code structure (pytorch#132931)" This reverts commit 65864d0. Reverted pytorch#132931 on behalf of https://github.com/ZainRizvi due to This PR is breaking builds internally due to the removal of ProcessGroup::Options ([comment](pytorch#132931 (comment)))
Chao1Han
pushed a commit
to Chao1Han/pytorch
that referenced
this pull request
Sep 20, 2024
…ons to reflect the correct code structure (pytorch#132931) (pytorch#135653) We introduced the dispatchable backend for a ProcessGroup and collective in pytorch#86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG. Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options" We need to make changes to the test to make it aligned with the change. This is try to reland D62008954 by fixing internal errors. Differential Revision: [D62483294](https://our.internmc.facebook.com/intern/diff/D62483294/) Pull Request resolved: pytorch#135653 Approved by: https://github.com/wz337, https://github.com/H-Huang
njhill
added a commit
to njhill/vllm
that referenced
this pull request
Mar 15, 2025
ProcessGroup.Options was removed in torch 2.6: pytorch/pytorch#132931 Signed-off-by: Nick Hill <nhill@redhat.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Stack from ghstack (oldest at bottom):
We introduced the dispatchable backend for a ProcessGroup and collective in #86225. This PR is a follow-up cleanup to clean up the option of a ProcessGroup and ask users to either set timeout or backend later on or directly create backend after creating a PG.
Also PGNCCL is using option class from ProcessGroup but we actually should use Option from backend class. So this PR is to make the type or name to be aligned with what we are doing in cpp side. I don't change the signature for the public API, so they still use args named "pg_options"
We need to make changes to the test to make it aligned with the change.
cc @XilunWu @H-Huang @awgu @kwen2501 @wanchaol @fegin @wz337 @wconstab @d4l3k @c-p-i-o @ezyang @gchanan