Use XNN fused RoPE for HF-style RoPE patterns#18627
Use XNN fused RoPE for HF-style RoPE patterns#18627GregoryComer wants to merge 1 commit intopytorch:mainfrom
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/18627
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 Awaiting Approval, 5 New Failures, 2 Unrelated FailuresAs of commit 1594c71 with merge base aa7c8ce ( 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. |
This PR needs a
|
Summary
Pattern match HF style RoPE and delegate to XNNPACK. Relax slice constraints and add unsqueeze, as well, since XNNPACK supports them and it's easiest to not have to conditionally partition them for RoPE only when fusable.