New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[inductor] Decompose addmm if it's a dot product on cpu #110010
Conversation
Generated code for dot product is often faster (on CPU) than dispatching to aten, since it avoids op dispatch overhead and allows fusion with surrounding ops, which in turn avoids allocations. Differential Revision: [D49595876](https://our.internmc.facebook.com/intern/diff/D49595876/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/110010
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 1 PendingAs of commit 921607c with merge base 92d86cd (): This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Generated code for dot product is often faster (on CPU) than dispatching to aten, since it avoids op dispatch overhead and allows fusion with surrounding ops, which in turn avoids allocations. Differential Revision: [D49595876](https://our.internmc.facebook.com/intern/diff/D49595876/) ghstack-source-id: 201785775 Pull Request resolved: #110010
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM. Thanks!
Generated code for dot product is often faster (on CPU) than dispatching to aten, since it avoids op dispatch overhead and allows fusion with surrounding ops, which in turn avoids allocations. Differential Revision: [D49595876](https://our.internmc.facebook.com/intern/diff/D49595876/) cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng Xia-Weiwen wenzhe-nrv jiayisunx peterbell10 ipiszy yf225 chenyang78 kadeng muchulee8 aakhundov [ghstack-poisoned]
Pull Request resolved: #110010 Generated code for dot product is often faster (on CPU) than dispatching to aten, since it avoids op dispatch overhead and allows fusion with surrounding ops, which in turn avoids allocations. ghstack-source-id: 202094380 @exported-using-ghexport Differential Revision: [D49595876](https://our.internmc.facebook.com/intern/diff/D49595876/)
@pytorchbot merge (Initiating merge automatically since Phabricator Diff has merged) |
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 |
Stack from ghstack (oldest at bottom):
Generated code for dot product is often faster (on CPU) than
dispatching to aten, since it avoids op dispatch overhead and allows fusion
with surrounding ops, which in turn avoids allocations.
Differential Revision: D49595876
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @Xia-Weiwen @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @aakhundov @ColinPeppler