Landing: ad8686f649cf

← All landings

Project / Subsystem

gcc / aarch64

Date

2026-07-17

Author

Pengfei Li

Commit

ad8686f649cfeca15c4334866b4403cebc5097c9

Source

github

Perf win

Yes

Breaking

View on GitHub →

All attributes

project: gcc
subsystem: aarch64
patch_id: —
commit_hash: ad8686f649cfeca15c4334866b4403cebc5097c9
source_type: github
headline: Update Neoverse V2 cost model for SVE.
tldr: Improves AArch64 vectorizer performance on Neoverse V2 by correcting SVE load/store costs.
author: Pengfei Li
outcome: committed
performance_win: true
breaking_change: false
series_id: —
series_parts: []
tags: • aarch64
• performance
• sve
• optimization
discussion_id_link: —
bugzilla_pr: —
url: https://github.com/gcc-mirror/gcc/commit/ad8686f649cfeca15c4334866b4403cebc5097c9
date: 2026-07-17T00:00:00.000Z

The AArch64 Neoverse V2 cost model has been updated to more accurately reflect the performance characteristics of SVE structure loads and stores. Previously, these costs were underestimated, leading the vectorizer to make suboptimal choices between AdvSIMD and SVE. This refinement aligns the cost model with the Neoverse V2 Software Optimization Guide and improves SPEC CPU 2026 performance by approximately 10.8% on Neoverse V2 systems.