Skip to main content

llvm-optimization

Implement or analyze LLVM optimization passes, pipelines, and generated-code quality. Use for IR transformation and pass behavior, including NeverC compiler performance, not general application tuning.

Informations de source

Dépôt
NeverSight/NeverC
Dernière activité de la source
13 septembre 2026 à 11:17
Langue détectée de SKILL.md
anglais
Étoiles
92
Forks
11

Options d'installation

Le prompt qui vérifie d'abord la source est sélectionné par défaut. Vous pouvez passer à une commande directe ou télécharger une copie locale.

Vérifiez les fichiers source

Lisez SKILL.md et les fichiers associés affichés par SkillsMP avant de décider de l'installer.

Explorateur de fichiers
6 fichiers

Affichage de SKILL.md

SKILL.md
Instructions source · Aperçu en lecture seule
name
llvm-optimization
description
Implement or analyze LLVM optimization passes, pipelines, and generated-code quality. Use for IR transformation and pass behavior, including NeverC compiler performance, not general application tuning.
# LLVM optimization Preserve semantics first, then measure the effect at the boundary the change intends to improve. ## Topic routing - [Pass development](references/pass-development.md) for pipelines, built-in passes, custom passes, instruction patterns, and dominance. - [Loops, vectorization, and LTO](references/loops-vectorization-and-lto.md) for specialized optimization modes. - [Debugging and correctness](references/debugging-and-correctness.md) for missed transforms, invalid IR, and semantic verification. - [NeverC performance](references/neverc-performance.md) for repository-specific compile-time and memory optimization evidence. - [Resources](references/resources.md) for upstream references. Use current pass-manager APIs and repository pipelines as authority. Do not infer success from IR shape alone when runtime behavior, code size, or compile-time performance is the intended outcome.
Voir sur GitHub