Skip to main navigation Skip to search Skip to main content

Dr.avx: A Dynamic Compilation System for Seamlessly Executing Hardware-Unsupported Vectorization Instructions

  • Yue Tang*
  • , Mianzhi Wu
  • , Yufeng Li
  • , Haoyu Liao
  • , Jianmei Guo
  • , Bo Huang
  • *Corresponding author for this work
  • East China Normal University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Modern processors are breaking a fundamental rule: backward compatibility within their own ISA families. We term this Generational ISA Fragmentation (GIF), where newer processors cannot execute instructions supported by prior generations within the same ISA family. This phenomenon is exemplified by Intel's removal of AVX-512 from Alder Lake processors after years of deployment, ARM's inconsistent support for SVE across cores, and RISC-V's incompatible vector specifications. GIF causes illegal instruction crashes when running applications optimized for earlier processors on newer hardware, threatening the foundation of software portability that has underpinned decades of computing evolution.We introduce Dr.avx, a dynamic compilation system that enables seamless execution of AVX-512 instructions on hardware that lacks native support. Dr.avx addresses the most instructive GIF instance, x86 AVX-512 fragmentation, by targeting the integer and floating-point operations that dominate real workloads. Our rewrite engine performs a fine-grained classification of AVX-512 opcode-operand patterns and employs three complementary strategies: Instr Mirroring, AVX Lowering, and Scalar Fallback. Experiments show that Dr.avx incurs a geometric mean overhead of 1.44× on SPEC CINT2017 relative to native AVX-512 execution, 17.3% better than Intel's closed-source SDE. On production databases, Dr.avx sustains 75%-88% (MySQL) and 86%-99% (MongoDB) of native throughput, yielding 2.0-2.7× higher throughput than SDE. For LLM inference (llama.cpp), Dr.avx keeps 95%-99% of native tokens/s and delivers 2.5-4.8× speedup over SDE. Unlike Intel's proprietary SDE, which provides no visibility into its implementation details, Dr.avx achieves functional correctness while providing an open, extensible, near-native performance implementation. Our work offers both a remedy for AVX-512 fragmentation and a blueprint for addressing similar compatibility challenges emerging across all major ISAs.

Original languageEnglish
Title of host publicationCGO 2026 - Proceedings of the 2026 IEEE/ACM International Symposium on Code Generation and Optimization
EditorsStephen M. Blackburn, Albert Cohen, Timothy M. Jones
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages403-415
Number of pages13
ISBN (Electronic)9798331592882
DOIs
StatePublished - 2026
Event24th IEEE/ACM International Symposium on Code Generation and Optimization, CGO 2026 - Sydney, Australia
Duration: 31 Jan 20264 Feb 2026

Publication series

NameCGO 2026 - Proceedings of the 2026 IEEE/ACM International Symposium on Code Generation and Optimization

Conference

Conference24th IEEE/ACM International Symposium on Code Generation and Optimization, CGO 2026
Country/TerritoryAustralia
CitySydney
Period31/01/264/02/26

Keywords

  • AVX-512
  • Dynamic Binary Translation
  • ISA compatibility
  • SIMD

Fingerprint

Dive into the research topics of 'Dr.avx: A Dynamic Compilation System for Seamlessly Executing Hardware-Unsupported Vectorization Instructions'. Together they form a unique fingerprint.

Cite this