Fix neon64 ee and oe store logic #10
Draft
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
Co-authored-by: czaplejewicz <[email protected]>
|
Cursor Agent can help with this pull request. Just |
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Outdated suggestions cannot be applied.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.
Fixes
neon64_eeandneon64_oeARM64 NEON FFT kernels to prevent writing uninitialized data.The original ARM32 kernels process 2 complex numbers (16 bytes) per iteration. The ARM64 port incorrectly mixed 2-lane (64-bit) arithmetic with 4-lane (128-bit) stores, causing the upper 64 bits of the 128-bit registers to be uninitialized and written to memory. This resulted in large L2 errors for N=32 FFTs. The fix ensures all operations (loads, arithmetic, and stores) are consistently 2-lane to match the ARM32 behavior and prevent writing garbage data.