[hyperactor] net: zero copy framer #907
Open
+447
−188
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 change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. Differential Revision: [D80365228](https://our.internmc.facebook.com/intern/diff/D80365228/) **NOTE FOR REVIEWERS**: This PR has internal Meta-specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D80365228/)! [ghstack-poisoned]
This was referenced Aug 14, 2025
mariusae
added a commit
that referenced
this pull request
Aug 18, 2025
This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. Differential Revision: [D80365228](https://our.internmc.facebook.com/intern/diff/D80365228/) **NOTE FOR REVIEWERS**: This PR has internal Meta-specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D80365228/)! ghstack-source-id: 303699221 Pull Request resolved: #907
meta-cla
bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D80365228 |
This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. Differential Revision: [D80365228](https://our.internmc.facebook.com/intern/diff/D80365228/) **NOTE FOR REVIEWERS**: This PR has internal Meta-specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D80365228/)! [ghstack-poisoned]
|
This pull request was exported from Phabricator. Differential Revision: D80365228 |
This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. Differential Revision: [D80365228](https://our.internmc.facebook.com/intern/diff/D80365228/) **NOTE FOR REVIEWERS**: This PR has internal Meta-specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D80365228/)! [ghstack-poisoned]
mariusae
added a commit
that referenced
this pull request
Aug 18, 2025
Pull Request resolved: #907 This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303752472 @exported-using-ghexport Differential Revision: [D80365228](https://our.internmc.facebook.com/intern/diff/D80365228/) **NOTE FOR REVIEWERS**: This PR has internal Meta-specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D80365228/)!
|
This pull request was exported from Phabricator. Differential Revision: D80365228 |
This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. Differential Revision: [D80365228](https://our.internmc.facebook.com/intern/diff/D80365228/) **NOTE FOR REVIEWERS**: This PR has internal Meta-specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D80365228/)! [ghstack-poisoned]
mariusae
added a commit
that referenced
this pull request
Aug 18, 2025
Pull Request resolved: #907 This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303773787 @exported-using-ghexport Differential Revision: [D80365228](https://our.internmc.facebook.com/intern/diff/D80365228/) **NOTE FOR REVIEWERS**: This PR has internal Meta-specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D80365228/)!
|
This pull request was exported from Phabricator. Differential Revision: D80365228 |
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303773787 exported-using-ghexport Reviewed By: shayne-fletcher, eliothedeman Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303773787 exported-using-ghexport Reviewed By: shayne-fletcher, eliothedeman Differential Revision: D80365228
This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. Differential Revision: [D80365228](https://our.internmc.facebook.com/intern/diff/D80365228/) **NOTE FOR REVIEWERS**: This PR has internal Meta-specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D80365228/)! [ghstack-poisoned]
mariusae
added a commit
that referenced
this pull request
Aug 18, 2025
Pull Request resolved: #907 This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303814000 @exported-using-ghexport Differential Revision: [D80365228](https://our.internmc.facebook.com/intern/diff/D80365228/) **NOTE FOR REVIEWERS**: This PR has internal Meta-specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D80365228/)!
|
This pull request was exported from Phabricator. Differential Revision: D80365228 |
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: Pull Request resolved: meta-pytorch#907 This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303699221 exported-using-ghexport Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: Pull Request resolved: meta-pytorch#907 This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303699221 exported-using-ghexport Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303814000 exported-using-ghexport Reviewed By: vidhyav Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303814000 exported-using-ghexport Reviewed By: vidhyav Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303814000 exported-using-ghexport Reviewed By: vidhyav Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303814000 exported-using-ghexport Reviewed By: vidhyav Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303814000 exported-using-ghexport Reviewed By: vidhyav Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: Pull Request resolved: meta-pytorch#907 This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303699221 exported-using-ghexport Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303814000 exported-using-ghexport Reviewed By: vidhyav Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303814000 exported-using-ghexport Reviewed By: vidhyav Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303814000 exported-using-ghexport Reviewed By: vidhyav Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303814000 exported-using-ghexport Reviewed By: vidhyav Differential Revision: D80365228
shayne-fletcher
pushed a commit
to shayne-fletcher/monarch-1
that referenced
this pull request
Aug 18, 2025
Summary: This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages. We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe). In this way, we have an easily hackable framer that does not introduce additional queues. ghstack-source-id: 303814000 exported-using-ghexport Reviewed By: vidhyav Differential Revision: D80365228
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):
Message#852This change introduces a zero-copy framer, which will also be easily extendable to doing vectorized framing of multipart messages.
We eschew tokio's framer (which relies on queuing for cancellation safety), in favor of a simple(r) implementation: the reader maintains a simple state machine, while the writer requires the caller to maintain an explicit write state (since this has to be driven across selects in order to be made cancellation safe).
In this way, we have an easily hackable framer that does not introduce additional queues.
Differential Revision: D80365228
NOTE FOR REVIEWERS: This PR has internal Meta-specific changes or comments, please review them on Phabricator!