EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LASM: Extensible Competitive Scheduling Framework for VLIW Assembly Optimization

Hongli Zhong, Zhong Liu, Sheng Ma

ACM Transactions on Design Automation of Electronic Systems · 31.01.2027 · 32 (1) · 1-43

DOI: 10.1145/3810954 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ACM Transactions on Design Automation of Electronic Systems, online erschienen am 02.11.2026. Verfasst von Hongli Zhong und 2 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Innovation Research Foundation of NUDT, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Very Long Instruction Word (VLIW) architectures deliver compelling performance-energy-cost advantages for automotive, AI, and high-performance computing. However, a persistent performance-productivity gap limits broader adoption: manual assembly optimization achieves superior performance but requires substantial expertise, while automated compilation underperforms. Existing assembly-level optimization tools commit to single scheduling strategies, unable to adapt to diverse program patterns. This article presents LASM, the first symbolic assembly framework enabling competitive multi-strategy orchestration for complete VLIW programs. Rather than committing to a single algorithm, multiple scheduling strategies simultaneously optimize identical code regions, with the framework automatically selecting the top solution through feasibility-driven ranking. LASM features standardized extension interfaces enabling plugin-style strategy integration and unified scheduling primitives eliminating algorithmic redundancy. We contribute two novel strategies: Longevity-Aware Expanded Modulo Scheduling (LAEMS) with dependency relaxation, proactive reordering, and parallel expansion search, and Front-Back Block Fusion (FBBF) for cross-boundary scheduling. Evaluation on 13 benchmarks across two production VLIW-SIMD processors demonstrates LASM achieves 103.6% and 101.6% of expert assembly performance while reducing manual code by 72.6%. A full YOLOv5 deployment validates production applicability at 95.9% of expert performance. Code is available at https://github.com/urays/lasm.

Förderung

  • Innovation Research Foundation of NUDT
  • National Natural Science Foundation of China

Zitieren

APA

Zhong, H., Liu, Z., & Ma, S. (2027). LASM: Extensible Competitive Scheduling Framework for VLIW Assembly Optimization. ACM Transactions on Design Automation of Electronic Systems, 32(1), 1-43. https://doi.org/10.1145/3810954

Vancouver

Zhong H, Liu Z, Ma S. LASM: Extensible Competitive Scheduling Framework for VLIW Assembly Optimization. ACM Transactions on Design Automation of Electronic Systems. 2027;32(1):1-43. doi:10.1145/3810954

BibTeX
@article{Zhong2027,
  author = {Zhong, Hongli and Liu, Zhong and Ma, Sheng},
  title = {LASM: Extensible Competitive Scheduling Framework for VLIW Assembly Optimization},
  journal = {ACM Transactions on Design Automation of Electronic Systems},
  year = {2027},
  volume = {32},
  number = {1},
  pages = {1--43},
  doi = {10.1145/3810954}
}
Kostenlos starten →