Landing: efeeb755192c

← All landings

Project / Subsystem

gcc / gcc/vectorization

Date

2026-04-28

Author

Richard Biener

Commit

efeeb755192cb03e5b24b44cb4fb563c11626a8e

Source

github

Perf win

Breaking

View on GitHub →

All attributes

project: gcc
subsystem: gcc/vectorization
patch_id: —
commit_hash: efeeb755192cb03e5b24b44cb4fb563c11626a8e
source_type: github
headline: Improve masked main loop selection for x86 vectorization.
tldr: The compiler now prefers masked main loops when they reduce epilogue iterations on x86, preventing performance regressions.
author: Richard Biener
outcome: committed
performance_win: false
breaking_change: false
series_id: —
series_parts: []
tags: • gcc
• vectorization
• x86
• optimization
discussion_id_link: —
bugzilla_pr: —
url: https://github.com/gcc-mirror/gcc/commit/efeeb755192cb03e5b24b44cb4fb563c11626a8e
date: 2026-04-28T00:00:00.000Z

This commit overrides vector_costs::better_main_loop_than_p to prevent regressions in gcc.target/i386/vect-partial-vectors-2.c when --param ix86-vect-compare-costs=1 is enabled. The change prioritizes masked main loops (where AVX masking is used) if they minimize the need for vector and scalar epilogue iterations, specifically when a non-masked main loop cannot be vectorized. This is a heuristic that favors smaller icache footprint on x86 for loops with few iterations.