Understanding and Finding JIT Compiler Performance Bugs

Zijian Yi, Chen Ding, August Shi, Milos Gligoric. 2026. Proceedings of the ACM on Programming Languages.

Read the paper · doi:10.1145/3798245 · arXiv:2603.06551

What this paper does with SQLancer

The citations here are to papers by SQLancer's authors rather than to SQLancer itself, listing database engines among the specialised systems that domain-specific tools target. The first study of performance bugs in just-in-time compilers, which prior automated work had not targeted, having focused on functional bugs. The authors manually analyse 191 bug reports across four Java and JavaScript JIT compilers to characterise triggers, manifestation patterns and root causes, then propose layered differential performance testing implemented as Jittery. It found 12 previously unknown performance bugs in Oracle HotSpot and Graal, 11 confirmed. Written by claude-opus-5 from the 3 places this paper refers to SQLancer. The quotations below are the paper's own words, stored verbatim when the text was extracted.

How it was classified

uses infrastructure — no

SQLancer is cited, not used; nothing in the mentions describes reusing its code.

extends technique — no

No technique is extended; the citation is background.

compares with — no

No empirical comparison against SQLancer is reported in the mentions.

describes as state of the art — no

The citation does not characterise SQLancer as the state of the art.

SQLancer publications it cites (2)

Bibliography entries that resolved to a SQLancer publication, or to a paper by one of the project's authors. A sentence citing one of these numbers is a reference to SQLancer even when it never writes the name.

#	Entry	Matched as
5	Jinsheng Ba and Manuel Rigger. 2024. Cert: Finding performance issues in database systems through the lens of cardinality estimation. InProceedings of the IEEE/ACM 46th International Conference on Software Engineering...	sqlancer publication · CERT
80	Theodoros Theodoridis, Manuel Rigger, and Zhendong Su. 2022. Finding missed optimizations through the lens of dead code elimination. InInternational Conference on Architectural Support for Programming Languages and Op...	project authored

Every place it refers to SQLancer (3)

3 sentences, each stored verbatim from the extracted text with where it was found and how. “Citation marker” means the sentence names no tool at all and was reached through a reference number that resolved to a SQLancer publication.

Id	Sentence	Found by	Where
M1	There has also been prior work on identifyinglong compilationbugs in C [ 46] and Markdown [ 52] compilers and on reducing compilation time in C++ [ 1,27], as well as studies onhigh-order performance bugs that primarily target missed optimizations in AOT compilers [ 6,30,55,80,81].	citation marker project authored background	1 Introduction page 2
M2	Other domain-specific tools target specialized systems, such as for database engines [ 5,43] and for machine learning libraries [ 82].	citation marker background	6.5 Discovered Bugs page 24
M3	[ 80,81] introduced techniques using dead-code markers and refined information to assess how effectively compilers eliminate redundant computations or exploit precise data-flow facts.	citation marker project authored background	6.5 Discovered Bugs page 24

This page is rendered from _data/papers/paper_doi_10_1145_3798245.json, extracted from arxiv. 29 pages, 99 references parsed.