Landing: 78ed779c28d1
Project / Subsystem
gcc / aarch64
Date
2026-08-25
Author
Dhruv Chawla
Commit
78ed779c28d1254e08b6c0a758bedfc8fcb1747c
Source
github
Perf win
Yes
Breaking
No
All attributes
- project
- gcc
- subsystem
- aarch64
- patch_id
- —
- commit_hash
- 78ed779c28d1254e08b6c0a758bedfc8fcb1747c
- source_type
- github
- headline
- Port NEON square-root intrinsics to pragma-based framework
- tldr
- Integrates NEON square-root intrinsics into GCC's pragma framework for unified handling and optimization.
- author
- Dhruv Chawla
- outcome
- committed
- performance_win
- true
- breaking_change
- false
- series_id
- —
- series_parts
- []
- tags
-
- • aarch64
- • neon
- • intrinsics
- • optimization
- discussion_id_link
- —
- bugzilla_pr
- —
- date
- 2026-08-25T00:00:00.000Z
This commit ports the NEON square-root intrinsics (vsqrt, vsqrtq) for floating-point types to the pragma-based framework on aarch64. By moving these intrinsics into this unified system, GCC can manage their lowering and optimization more consistently. The direct function definitions in arm_neon.h are removed, and new function bases and definition groups are added in aarch64-neon-builtins-base.cc and .def. A new test case is included to verify the correct functioning of these ported square-root operations.