[dtensor][6/N] change to a better/safer op registration #90735
Closed
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
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
wanchaol
requested review from
H-Huang,
awgu,
kwen2501,
mrshenli,
pritamdamania87,
rohan-varma and
zhaojuanmao
as code owners
This was referenced Dec 13, 2022
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/90735
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit d2f2977: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This was referenced Dec 13, 2022
wanchaol
added a commit
that referenced
this pull request
Dec 13, 2022
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch ghstack-source-id: 3c4d812 Pull Request resolved: #90735
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
wanchaol
added a commit
that referenced
this pull request
Dec 20, 2022
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch ghstack-source-id: 7c0eb4c Pull Request resolved: #90735
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
wanchaol
added a commit
that referenced
this pull request
Dec 29, 2022
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch ghstack-source-id: 72918af Pull Request resolved: #90735
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
wanchaol
added a commit
that referenced
this pull request
Jan 3, 2023
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch ghstack-source-id: 2d70971 Pull Request resolved: #90735
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
wanchaol
added a commit
that referenced
this pull request
Jan 4, 2023
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch ghstack-source-id: c25cef7 Pull Request resolved: #90735
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This was referenced Jan 24, 2023
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch [ghstack-poisoned]
Collaborator
Author
|
@wanchaol has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch Differential Revision: [D42876250](https://our.internmc.facebook.com/intern/diff/D42876250) [ghstack-poisoned]
Collaborator
Author
|
@wanchaol has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch Differential Revision: [D42876250](https://our.internmc.facebook.com/intern/diff/D42876250) [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch Differential Revision: [D42876250](https://our.internmc.facebook.com/intern/diff/D42876250) [ghstack-poisoned]
This PR changes the op registration to a better mechanism, now we require the directly overload registration instead of the op key str, this have several benefits: 1. We ensure that the op registration registers the correct op, which means it would be faild if the op registration become wrong (this PR already fixing several op registration errors as we use direct OpOverload registration 2. If the overload name get changed/deleted, we immediately know it at the source code compilation level, which is safer 3. This also keep it consistents with the op registration mechanism with other tensor subclasses within PyTorch Differential Revision: [D42876250](https://our.internmc.facebook.com/intern/diff/D42876250) [ghstack-poisoned]
fduwjj
reviewed
Jan 31, 2023
Contributor
|
@pytorchbot merge (Initiating merge automatically since Phabricator Diff has merged) |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
wanchaol
added a commit
to pytorch/PiPPy
that referenced
this pull request
Feb 1, 2023
This is a joint PR with the stack pytorch/pytorch#90735
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):
This PR changes the op registration to a better mechanism, now
we require the directly overload registration instead of the op
key str, this have several benefits:
means it would be faild if the op registration become wrong (this PR
already fixing several op registration errors as we use direct
OpOverload registration
the source code compilation level, which is safer
other tensor subclasses within PyTorch
Differential Revision: D42876250