-
Notifications
You must be signed in to change notification settings - Fork 608
Manual LICM in op_convolution #3274
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
Conversation
Summary: Otherwise, we require LTO for decent performance for this thing in kernels. Reviewed By: kimishpatel, manuelcandales Differential Revision: D56503573
Summary: Hoisting loads of these constants manually speeds up the kernel a lot. Depending on the types for CTYPE and CTYPE_BIAS, the compiler might not be able to prove this is safe on its own. Reviewed By: kimishpatel, manuelcandales Differential Revision: D56503574
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/3274
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 2b1426e with merge base b7b40ac ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This pull request was exported from Phabricator. Differential Revision: D56503574 |
Summary: Hoisting loads of these constants manually speeds up the kernel a lot. Depending on the types for CTYPE and CTYPE_BIAS, the compiler might not be able to prove this is safe on its own. Reviewed By: kimishpatel, manuelcandales Differential Revision: D56503574
Summary: Hoisting loads of these constants manually speeds up the kernel a lot. Depending on the types for CTYPE and CTYPE_BIAS, the compiler might not be able to prove this is safe on its own. Reviewed By: kimishpatel, manuelcandales Differential Revision: D56503574
Summary: Hoisting loads of these constants manually speeds up the kernel a lot. Depending on the types for CTYPE and CTYPE_BIAS, the compiler might not be able to prove this is safe on its own. Reviewed By: kimishpatel, manuelcandales Differential Revision: D56503574
to exported stacked diffs you want to use ghstack else it will combine changes from diff underneat |
This pull request has been merged in 987a1f9. |
Summary: Hoisting loads of these constants manually speeds up the kernel a lot. Depending on the types for CTYPE and CTYPE_BIAS, the compiler might not be able to prove this is safe on its own.
Reviewed By: kimishpatel, manuelcandales
Differential Revision: D56503574