[Static Runtime] Use composite op for TE fusion #74126
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
When we perform fusion without the composite op, `TensorExprDynamicGroup`, it ends up not reusing the output tensor buffers. So, until we figure out a way to do that with `TensorExprGroup` op, it seems strictly better to use composite op, even though it involves going to the JIT. Differential Revision: [D34831280](https://our.internmc.facebook.com/intern/diff/D34831280/) [ghstack-poisoned]
🔗 Helpful links
💊 CI failures summary and remediationsAs of commit f907fdb (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
navahgar
added a commit
that referenced
this pull request
Mar 11, 2022
When we perform fusion without the composite op, `TensorExprDynamicGroup`, it ends up not reusing the output tensor buffers. So, until we figure out a way to do that with `TensorExprGroup` op, it seems strictly better to use composite op, even though it involves going to the JIT. Differential Revision: [D34831280](https://our.internmc.facebook.com/intern/diff/D34831280/) ghstack-source-id: 151191941 Pull Request resolved: #74126
facebook-github-bot
pushed a commit
that referenced
this pull request
Mar 15, 2022
Summary: Pull Request resolved: #74126 When we perform fusion without the composite op, `TensorExprDynamicGroup`, it ends up not reusing the output tensor buffers. So, until we figure out a way to do that with `TensorExprGroup` op, it seems strictly better to use composite op, even though it involves going to the JIT. ghstack-source-id: 151191941 Test Plan: Tested locally with `ptvsc2_predictor_bench` on the Video model. Performance analysis with `caffe2/caffe2/fb/predictor/bench:limb` on the Video model locally showed an improvement of ~1% with this change. Reviewed By: mikeiovine Differential Revision: D34831280 fbshipit-source-id: e523878364b519ccd51b78d52d9f6c9d3e8def17
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):
When we perform fusion without the composite op,
TensorExprDynamicGroup, it ends up not reusing the output tensor buffers. So, until we figure out a way to do that withTensorExprGroupop, it seems strictly better to use composite op, even though it involves going to the JIT.Differential Revision: D34831280