[Inductor] Add fused_attention pattern matcher with additional clone #108141
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/108141
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 8eb66b8 with merge base 60bb02a ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
637d24a to
8eb66b8
Compare
jgong5
requested changes
Aug 29, 2023
Oh, I didn't realize it. Yes, let's get this PR in first then. |
jgong5
approved these changes
Aug 30, 2023
|
@pytorchbot merge |
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 |
Valentine233
added a commit
that referenced
this pull request
Aug 31, 2023
…108141) A previous PR #106274 decomposes `aten.dropout` and would create a `clone()` when `eval()` or `p=0`. This makes many SDPA-related models fail to match fused_attention pattern matchers. This PR adds new fused_attention pattern matchers with an additional clone to re-enable the SDPA op matching. Pull Request resolved: #108141 Approved by: https://github.com/jgong5, https://github.com/eellison
This was referenced Aug 31, 2023
CaoE
pushed a commit
to CaoE/pytorch
that referenced
this pull request
Sep 1, 2023
…ytorch#108141) A previous PR pytorch#106274 decomposes `aten.dropout` and would create a `clone()` when `eval()` or `p=0`. This makes many SDPA-related models fail to match fused_attention pattern matchers. This PR adds new fused_attention pattern matchers with an additional clone to re-enable the SDPA op matching. Pull Request resolved: pytorch#108141 Approved by: https://github.com/jgong5, https://github.com/eellison
atalman
pushed a commit
that referenced
this pull request
Sep 5, 2023
…108141) (#108327) A previous PR #106274 decomposes `aten.dropout` and would create a `clone()` when `eval()` or `p=0`. This makes many SDPA-related models fail to match fused_attention pattern matchers. This PR adds new fused_attention pattern matchers with an additional clone to re-enable the SDPA op matching. Pull Request resolved: #108141 Approved by: https://github.com/jgong5, https://github.com/eellison
eellison
added a commit
that referenced
this pull request
Sep 12, 2023
When dropout is traced in inference, it creates a clone() instead of training pattern of rand() etc. This was partially addressed by manually #108141, however that did not cover all of the patterns that included dropout, and there is no reason we should have to specify them manually. This updates the inference patterns generated to trace with dropout_p = 0.0. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx peterbell10 ipiszy ngimel yf225 chenyang78 kadeng muchulee8 aakhundov [ghstack-poisoned]
eellison
added a commit
that referenced
this pull request
Sep 13, 2023
…0, cleanup" When dropout is traced in inference, it creates a clone() instead of training pattern of rand() etc. This was partially addressed by manually #108141, however that did not cover all of the patterns that included dropout, and there is no reason we should have to specify them manually. This updates the inference patterns generated to trace with dropout_p = 0.0. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx peterbell10 ipiszy ngimel yf225 chenyang78 kadeng muchulee8 aakhundov [ghstack-poisoned]
eellison
added a commit
that referenced
this pull request
Sep 13, 2023
When dropout is traced in inference, it creates a clone() instead of training pattern of rand() etc. This was partially addressed by manually #108141, however that did not cover all of the patterns that included dropout, and there is no reason we should have to specify them manually. This updates the inference patterns generated to trace with dropout_p = 0.0. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx peterbell10 ipiszy ngimel yf225 chenyang78 kadeng muchulee8 aakhundov [ghstack-poisoned]
eellison
added a commit
that referenced
this pull request
Sep 14, 2023
…0, cleanup" When dropout is traced in inference, it creates a clone() instead of training pattern of rand() etc. This was partially addressed by manually #108141, however that did not cover all of the patterns that included dropout, and there is no reason we should have to specify them manually. This updates the inference patterns generated to trace with dropout_p = 0.0. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx peterbell10 ipiszy ngimel yf225 chenyang78 kadeng muchulee8 aakhundov [ghstack-poisoned]
eellison
added a commit
that referenced
this pull request
Sep 14, 2023
When dropout is traced in inference, it creates a clone() instead of training pattern of rand() etc. This was partially addressed by manually #108141, however that did not cover all of the patterns that included dropout, and there is no reason we should have to specify them manually. This updates the inference patterns generated to trace with dropout_p = 0.0. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx peterbell10 ipiszy ngimel yf225 chenyang78 kadeng muchulee8 aakhundov [ghstack-poisoned]
pytorchmergebot
pushed a commit
that referenced
this pull request
Sep 15, 2023
When dropout is traced in inference, it creates a clone() instead of training pattern of rand() etc. This was partially addressed by manually #108141, however that did not cover all of the patterns that included dropout, and there is no reason we should have to specify them manually. This updates the inference patterns generated to trace with dropout_p = 0.0. Pull Request resolved: #109118 Approved by: https://github.com/drisspg, https://github.com/Valentine233
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.
A previous PR #106274 decomposes
aten.dropoutand would create aclone()wheneval()orp=0. This makes many SDPA-related models fail to match fused_attention pattern matchers.This PR adds new fused_attention pattern matchers with an additional clone to re-enable the SDPA op matching.
cc @jgong5 @mingfeima @XiaobingSuper @sanchitintel @ashokei @jingxu10 @voznesenskym @penguinwu @EikanWang @Guobing-Chen @zhuhaozhe @blzheng @Xia-Weiwen @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @ngimel @yf225 @chenyang78 @kadeng @muchulee8 @aakhundov