[SR] Avoid allocating rstd/mean in layer_norm #73606
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
The single-output overload of `layer_norm` internally allocates two tensors. As an optimization, we previously added `static_runtime::layer_norm`. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely. Differential Revision: [D34562131](https://our.internmc.facebook.com/intern/diff/D34562131/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D34562131/)! [ghstack-poisoned]
Contributor
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit 92ec7d6 (more details on the Dr. CI page): 💚 💚 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. |
CI Flow Status⚛️ CI FlowRuleset - Version:
|
facebook-github-bot
added
cla signed
oncall: jit
The single-output overload of `layer_norm` internally allocates two tensors. As an optimization, we previously added `static_runtime::layer_norm`. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely. Differential Revision: [D34562131](https://our.internmc.facebook.com/intern/diff/D34562131/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D34562131/)! [ghstack-poisoned]
mikeiovine
pushed a commit
that referenced
this pull request
Mar 1, 2022
Pull Request resolved: #73606 The single-output overload of `layer_norm` internally allocates two tensors. As an optimization, we previously added `static_runtime::layer_norm`. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely. ghstack-source-id: 150273000 Differential Revision: [D34562131](https://our.internmc.facebook.com/intern/diff/D34562131/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D34562131/)!
The single-output overload of `layer_norm` internally allocates two tensors. As an optimization, we previously added `static_runtime::layer_norm`. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely. Differential Revision: [D34562131](https://our.internmc.facebook.com/intern/diff/D34562131/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D34562131/)! [ghstack-poisoned]
mikeiovine
pushed a commit
that referenced
this pull request
Mar 8, 2022
Pull Request resolved: #73606 The single-output overload of `layer_norm` internally allocates two tensors. As an optimization, we previously added `static_runtime::layer_norm`. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely. ghstack-source-id: 150814731 Differential Revision: [D34562131](https://our.internmc.facebook.com/intern/diff/D34562131/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D34562131/)!
The single-output overload of `layer_norm` internally allocates two tensors. As an optimization, we previously added `static_runtime::layer_norm`. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely. Differential Revision: [D34562131](https://our.internmc.facebook.com/intern/diff/D34562131/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D34562131/)! [ghstack-poisoned]
mikeiovine
pushed a commit
that referenced
this pull request
Mar 10, 2022
Pull Request resolved: #73606 The single-output overload of `layer_norm` internally allocates two tensors. As an optimization, we previously added `static_runtime::layer_norm`. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely. ghstack-source-id: 151045157 Differential Revision: [D34562131](https://our.internmc.facebook.com/intern/diff/D34562131/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D34562131/)!
The single-output overload of `layer_norm` internally allocates two tensors. As an optimization, we previously added `static_runtime::layer_norm`. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely. Differential Revision: [D34562131](https://our.internmc.facebook.com/intern/diff/D34562131/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D34562131/)! [ghstack-poisoned]
mikeiovine
pushed a commit
that referenced
this pull request
Mar 14, 2022
Pull Request resolved: #73606 The single-output overload of `layer_norm` internally allocates two tensors. As an optimization, we previously added `static_runtime::layer_norm`. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely. ghstack-source-id: 151281719 Differential Revision: [D34562131](https://our.internmc.facebook.com/intern/diff/D34562131/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D34562131/)!
The single-output overload of `layer_norm` internally allocates two tensors. As an optimization, we previously added `static_runtime::layer_norm`. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely. Differential Revision: [D34562131](https://our.internmc.facebook.com/intern/diff/D34562131/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D34562131/)! [ghstack-poisoned]
mikeiovine
pushed a commit
that referenced
this pull request
Mar 15, 2022
Pull Request resolved: #73606 The single-output overload of `layer_norm` internally allocates two tensors. As an optimization, we previously added `static_runtime::layer_norm`. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely. ghstack-source-id: 151394116 Differential Revision: [D34562131](https://our.internmc.facebook.com/intern/diff/D34562131/) **NOTE FOR REVIEWERS**: This PR has internal Facebook specific changes or comments, please review them on [Phabricator](https://our.internmc.facebook.com/intern/diff/D34562131/)!
facebook-github-bot
pushed a commit
that referenced
this pull request
Mar 15, 2022
Summary: Pull Request resolved: #73606 The single-output overload of `layer_norm` internally allocates two tensors. As an optimization, we previously added `static_runtime::layer_norm`. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely. ghstack-source-id: 151394116 Test Plan: Existing unit tests Reviewed By: hlu1 Differential Revision: D34562131 fbshipit-source-id: c6a6560e60db43b0b100aedc54ea4265acb347de
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):
The single-output overload of
layer_norminternally allocates two tensors. As an optimization, we previously addedstatic_runtime::layer_norm. This variant of layer norm had two extra outputs to make the memory planner aware of these extra tensors. But these outputs were unused; it's actually better for us to avoid the allocation and associated computations entirely.Differential Revision: D34562131
NOTE FOR REVIEWERS: This PR has internal Facebook specific changes or comments, please review them on Phabricator!