[PT2E][X86] Add Inductor fusion passes of float8 qconv for X86Inductor backend#3261
Merged
Xia-Weiwen merged 4 commits intopytorch:mainfrom Nov 26, 2025
Merged
[PT2E][X86] Add Inductor fusion passes of float8 qconv for X86Inductor backend#3261Xia-Weiwen merged 4 commits intopytorch:mainfrom
Xia-Weiwen merged 4 commits intopytorch:mainfrom
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/3261
Note: Links to docs will display an error until the docs builds have been completed. ❗ 2 Active SEVsThere are 2 currently active SEVs. If your PR is affected, please view them below: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
00600ee to
d1163ba
Compare
d1163ba to
fa88f2f
Compare
Xia-Weiwen
approved these changes
Nov 19, 2025
4e834a6 to
c92f1ee
Compare
jerryzh168
approved these changes
Nov 26, 2025
Contributor
jerryzh168
left a comment
There was a problem hiding this comment.
stamping since this doesn't touch the main torchao API or user API
can you show any performance metrics before and after the fusion as well
Collaborator
Thanks for reviewing. This fusion passes are actually for lowering, i.e., to fuse the pattern |
andrewor14
added a commit
that referenced
this pull request
Nov 26, 2025
andrewor14
added a commit
that referenced
this pull request
Nov 26, 2025
jcaip
added a commit
that referenced
this pull request
Dec 2, 2025
namgyu-youn
pushed a commit
to namgyu-youn/ao
that referenced
this pull request
Dec 19, 2025
…r backend (pytorch#3261) * [Inductor][float8] Register qconv weight prepack pass for float8 * [Inductor][float8] Register qconv-unary fusion pass for float8 * [Inductor][float8] Register qconv-binary fusion pass for float8 * add comments
namgyu-youn
pushed a commit
to namgyu-youn/ao
that referenced
this pull request
Dec 19, 2025
namgyu-youn
pushed a commit
to namgyu-youn/ao
that referenced
this pull request
Dec 19, 2025
Revert "Fix style after pytorch#3261 (pytorch#3397)" This reverts commit 316ef03.
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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.
Summary:
This PR aims to support QConv weight prepacking, QConv unary fusion, and QConv binary fusion for fp8 data type in Inductor, to lower the ref quant model to X86Inductor backend.