-
Notifications
You must be signed in to change notification settings - Fork 712
[Executorch][LLM] Use caching allocator for runner #15656
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
base: gh/kimishpatel/207/base
Are you sure you want to change the base?
Conversation
We observed that on iOS it improves perf by 6% because SDPA op does temp allocations. No significant difference on android though. Differential Revision: [D86120038](https://our.internmc.facebook.com/intern/diff/D86120038/) [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/15656
Note: Links to docs will display an error until the docs builds have been completed. ❌ 116 New Failures, 5 Unrelated FailuresAs of commit 66a37e3 with merge base 15a0fcd ( NEW FAILURES - The following jobs have failed:
BROKEN TRUNK - The following jobs failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
We observed that on iOS it improves perf by 6% because SDPA op does temp allocations. No significant difference on android though. Differential Revision: [D86120038](https://our.internmc.facebook.com/intern/diff/D86120038/) ghstack-source-id: 321483010 Pull Request resolved: #15656
This PR needs a
|
Stack from ghstack (oldest at bottom):
We observed that on iOS it improves perf by 6% because SDPA op does temp allocations.
No significant difference on android though.
Differential Revision: D86120038