Landing: d373724d1dde

← All landings

Project / Subsystem

gcc / amdgcn

Date

2026-09-15

Author

Andrew Stubbs

Commit

d373724d1ddeef9b322b6b6078fc1de8dbe27791

Source

github

Perf win

Yes

Breaking

View on GitHub →

All attributes

project: gcc
subsystem: amdgcn
patch_id: —
commit_hash: d373724d1ddeef9b322b6b6078fc1de8dbe27791
source_type: github
headline: amdgcn: Spill scalars to both SGPRs and AVGPRs for performance
tldr: AMDGCN backend now spills scalars to both SGPRs and AVGPRs, improving performance across various benchmarks.
author: Andrew Stubbs
outcome: committed
performance_win: true
breaking_change: false
series_id: —
series_parts: []
tags: • gcc
• amdgcn
• gpu
• optimization
• registers
discussion_id_link: —
bugzilla_pr: —
url: https://github.com/gcc-mirror/gcc/commit/d373724d1ddeef9b322b6b6078fc1de8dbe27791
date: 2026-09-15T00:00:00.000Z

The AMD GCN backend in GCC has been updated to allow scalars residing in Vector General Purpose Registers (VGPRs) to be spilled into either Scalar General Purpose Registers (SGPRs) or Additional VGPRs (AVGPRs). This change addresses performance regressions observed when scalars were exclusively spilled to SGPRs, by providing flexibility in register allocation. The new approach improves performance across a range of benchmarks by better managing register pressure.