Landing: 9d42eeefb750
Project / Subsystem
gcc / aarch64
Date
2026-08-24
Author
Dhruv Chawla
Commit
9d42eeefb75065ab7c17926ae4f65bdfea6e8e77
Source
github
Perf win
No
Breaking
No
All attributes
- project
- gcc
- subsystem
- aarch64
- patch_id
- —
- commit_hash
- 9d42eeefb75065ab7c17926ae4f65bdfea6e8e77
- source_type
- github
- headline
- aarch64: Port NEON reduction intrinsics to pragma framework
- tldr
- NEON reduction intrinsics (vaddv, vmaxv, vminv, etc.) are now handled by GCC's pragma-based framework.
- author
- Dhruv Chawla
- outcome
- committed
- performance_win
- false
- breaking_change
- false
- series_id
- —
- series_parts
- []
- tags
-
- • aarch64
- • neon
- • intrinsics
- • compiler framework
- discussion_id_link
- —
- bugzilla_pr
- —
- date
- 2026-08-24T00:00:00.000Z
Several NEON reduction intrinsics for aarch64, including vaddv, vmaxv, vminv, vmaxnmv, and vminnmv, have been ported to GCC’s pragma-based framework. Some floating-point variants requiring special NaN handling (e.g., vmaxv_f32) were not ported in this patch due to complexities in mapping them to existing instruction patterns. The changes involve adding new function bases and groups and removing many previously defined functions from arm_neon.h.