CGO 2025 – Proceedings

Results Reproduced Article: cgo25main-p189-p (type: Full Paper) doi:10.1145/3696443.3708954

Synthesis of Sorting Kernels (doi:10.5281/zenodo.14092980): Recently, AlphaDev has shown significant advances in the synthesis of branchless sorting kernels for arrays of lengths~$3$ to~$5$. In this paper, we propose an enumerative search technique based on A* search and present novel optimality-pre\-serving heuristics and non-optimality-preserving cuts for sorting kernel ...

Tensorize: Fast Synthesis of Tensor Programs from Legacy Code using Symbolic Tracing, Sketching and Solving
Alexander Brauckmann, Luc Jaulmes, José W. de Souza Magalhães, Elizabeth Polgreen, and Michael F. P. O’Boyle
(University of Edinburgh, UK)

Results Reproduced Article: cgo25main-p209-p (type: Full Paper) doi:10.1145/3696443.3708956

Tensorize: Fast Synthesis of Tensor Programs from Legacy Code using Symbolic Tracing, Sketching and Solving (Artifact) (doi:10.5281/zenodo.14095398): This artifact contains the source code of \numpyrize, evaluation benchmarks, and the scripts used to generate the plots shown in Figures 7-10. We provide an automated pipeline to build \numpyrize, run synthesis experiments, and plot results.

Enhancing Deployment-Time Predictive Model Robustness for Code Analysis and Optimization
Huanting Wang, Patrick Lenihan, and Zheng Wang
(University of Leeds, UK)

Results Reproduced Article: cgo25main-p247-p (type: Full Paper) doi:10.1145/3696443.3708959

Enhancing Deployment-time Predictive Model Robustness for Code Analysis and Optimization (doi:10.5281/zenodo.14077780): This code artifact serves as a supporting component for the publication titled "Enhancing Deployment-Time Predictive Model Robustness for Code Analysis and Optimization." Prom is an open-source Python toolkit designed to enhance the robustness and performance of predictive models against changes during deployment. For ...

Optimizations and Transformations (1)

SySTeC: A Symmetric Sparse Tensor Compiler
Radha Patel, Willow Ahrens, and Saman Amarasinghe
(Massachusetts Institute of Technology, USA)

Artifacts Reusable Article: cgo25main-p56-p (type: Full Paper) doi:10.1145/3696443.3708919

SySTeC: A Symmetric Sparse Tensor Compiler: The Artifact (doi:10.5281/zenodo.13821280): This is the artifact to accompany the paper "SySTeC: A Symmetric Sparse Tensor Compiler" to be published in https://conf.researchr.org/home/cgo-2025. A preprint is available on ArXiv at https://arxiv.org/abs/2406.09266

Pattern Matching in AI Compilers and Its Formalization
Joseph W. Cutler, Alex Collins, Bin Fan, Mahesh Ravishankar, and Vinod Grover
(University of Pennsylvania, USA; NVIDIA, USA; NVIDIA, UK; AMD, USA)

Publisher's Version Article: cgo25main-p17-p (type: Full Paper) doi:10.1145/3696443.3708934

Scalar Interpolation: A Better Balance between Vector and Scalar Execution for SuperScalar Architectures
Reza Ghanbari, Henry Kao, João P. L. De Carvalho, Ehsan Amiri, and J. Nelson Amaral
(University of Alberta, Canada; Huawei Technologies, Canada)

Results Reproduced Article: cgo25main-p159-p (type: Full Paper) doi:10.1145/3696443.3708950

Artifact of "Scalar Interpolation: A Better Balance Between Vector and Scalar Execution for SuperScalar Architectures" (doi:10.5281/zenodo.14090974): This is the artifact of the paper "Scalar Interpolation: A Better Balance Between Vector and Scalar Execution for SuperScalar Architectures". The files are: artifact_description.pdf: instructions on how to use this artifact scalar-interpolation.tar.gz: the docker image for running the experiments ...

ML Tools and Optimization

VEGA: Automatically Generating Compiler Backends using a Pre-trained Transformer Model
Ming Zhong, Fang Lv, Lulin Wang, Lei Qiu, Yingying Wang, Ying Liu, Huimin Cui, Xiaobing Feng, and Jingling Xue
(Institute of Computing Technology at Chinese Academy of Sciences, China; University of Chinese Academy of Sciences, China; UNSW, Australia)

Results Reproduced Article: cgo25main-p7-p (type: Full Paper) doi:10.1145/3696443.3708931

Reproduction Package for Article: "VEGA: Automatically Generating Compiler Backends Using a Pre-Trained Transformer Model" (doi:10.5281/zenodo.14064392): Artifact for article: "VEGA: Automatically Generating Compiler Backends Using a Pre-Trained Transformer Model", including a fine-tuned model, scripts and the main results of the paper.

IntelliGen: Instruction-Level Auto-tuning for Tensor Program with Monotonic Memory Optimization
Zixuan Ma, Haojie Wang, Jingze Xing, Shuhong Huang, Liyan Zheng, Chen Zhang, Huanqi Cao, Kezhao Huang, Mingshu Zhai, Shizhi Tang, Penghan Wang, and Jidong Zhai
(Tsinghua University, China; Qingcheng.AI, China)

Publisher's Version Article: cgo25main-p812-p (type: Full Paper) doi:10.1145/3696443.3708967

GraalNN: Context-Sensitive Static Profiling with Graph Neural Networks
Lazar Milikic, Milan Cugurovic, and Vojin Jovanovic
(Oracle Labs, Switzerland; Oracle Labs, Serbia)

Publisher's Version Article: cgo25main-p236-p (type: Full Paper) doi:10.1145/3696443.3708958

LLM-Vectorizer: LLM-Based Verified Loop Vectorizer
Jubi Taneja, Avery Laird, Cong Yan, Madan Musuvathi, and Shuvendu K. Lahiri
(Microsoft Research, USA; University of Toronto, Canada)

Publisher's Version Article: cgo25main-p132-p (type: Full Paper) doi:10.1145/3696443.3708929

Architectures and Code Generation

Calibro: Compilation-Assisted Linking-Time Binary Code Outlining for Code Size Reduction in Android Applications
Zhanhao Liang, Hanming Sun, Wenhan Shang, Mengting Yuan, Jingqin Fu, Jiang Ma, Chun Jason Xue, and Qingan Li
(Wuhan University, China; Wuhan Broadcasting and Television Station, China; Guangdong OPPO Mobile Telecommunications, China; MBZUAI, United Arab Emirates)

Publisher's Version Article: cgo25main-p193-p (type: Full Paper) doi:10.1145/3696443.3708955

A Multi-level Compiler Backend for Accelerated Micro-kernels Targeting RISC-V ISA Extensions
Alexandre Lopoukhine, Federico Ficarelli, Christos Vasiladiotis, Anton Lydike, Josse Van Delm, Alban Dutilleul, Luca Benini, Marian Verhelst, and Tobias Grosser
(University of Cambridge, UK; University of Bologna, Italy; Cineca, Italy; University of Edinburgh, UK; KU Leuven, Belgium; ENS Rennes, France; ETH Zurich, Switzerland)

Results Reproduced Article: cgo25main-p170-p (type: Full Paper) doi:10.1145/3696443.3708952

Artifact of "A Multi-Level Compiler Backend for Accelerated Micro-Kernels Targetting RISC-V ISA Extensions" (doi:10.5281/zenodo.14052014): Supporting artifact for the paper titled "A Multi-Level Compiler Backend for Accelerated Micro-Kernels Targeting RISC-V ISA Extensions" as published in CGO 2025. It can be used to reproduce all results presented in the final version of this paper.

xDSL: Sidekick Compilation for SSA-Based Compilers
Mathieu Fehr, Michel Weber, Christian Ulmann, Alexandre Lopoukhine, Martin Paul Lücke, Théo Degioanni, Christos Vasiladiotis, Michel Steuwer, and Tobias Grosser
(University of Edinburgh, UK; ETH Zurich, Switzerland; University of Cambridge, UK; ENS Rennes, France; Technische Universität Berlin, Germany)

Results Reproduced Article: cgo25main-p91-p (type: Full Paper) doi:10.1145/3696443.3708945

Artifact of "xDSL: Sidekick Compilation for SSA-based Compilers" (doi:10.5281/zenodo.14263271): Supporting artifact for the paper titled “xDSL: Sidekick Compilation for SSA-based compilers” as published in CGO 2025. It contains the source code of xDSL and can be used to reproduce all results presented in the final version of this paper.

ML Compilers

ANT-ACE: An FHE Compiler Framework for Automating Neural Network Inference
Long Li, Jianxin Lai, Peng Yuan, Tianxiang Sui, Yan Liu, Qing Zhu, Xiaojing Zhang, Linjie Xiao, Wenguang Chen, and Jingling Xue
(Ant Group, China; Tsinghua University, China; UNSW, Australia; Ant Group, Australia)

Results Reproduced Article: cgo25main-p95-p (type: Full Paper) doi:10.1145/3696443.3708924

ANT-ACE FHE Compiler (doi:10.5281/zenodo.14625994): ANT-ACE is a Fully Homomorphic Encryption (FHE) Compiler Framework designed for automating Neural Network (NN) Inference. ANT-ACE accepts a pre-trained ONNX model as input and directly generates C/C++ programs to perform NN inference on encrypted data.

CUrator: An Efficient LLM Execution Engine with Optimized Integration of CUDA Libraries
Yoon Noh Lee, Yongseung Yu, and Yongjun Park
(Yonsei University, South Korea)

Results Reproduced Article: cgo25main-p87-p (type: Full Paper) doi:10.1145/3696443.3708944

CUrator: An Efficient LLM Execution Engine with Optimized Integration of CUDA Libraries (doi:10.5281/zenodo.14509993): The artifact provides source codes for cuBLAS, CUTLASS (without Fused Multi-Head Attention (FMHA)), and CUrator with FMHA. It also provides the RTX 4090 CUTLASS tiling configurations used in Figure 7 and Figure 8 in Section4.2. Getting all the oracle tiling settings on each GPU evaluated in this paper takes a long ...

Accelerating LLMs using an Efficient GEMM Library and Target-Aware Optimizations on Real-World PIM Devices
Hyeoncheol Kim, Taehoon Kim, Taehyeong Park, Donghyeon Kim, Yongseung Yu, Hanjun Kim, and Yongjun Park
(Yonsei University, South Korea; Rebellions, South Korea; Hanyang University, South Korea)

Publisher's Version Article: cgo25main-p188-p (type: Full Paper) doi:10.1145/3696443.3708953

MLIR

The MLIR Transform Dialect: Your Compiler Is More Powerful Than You Think
Martin Paul Lücke, Oleksandr Zinenko, William S. Moses, Michel Steuwer, and Albert Cohen
(University of Edinburgh, UK; Google DeepMind, France; University of Illinois at Urbana-Champaign, USA; Google DeepMind, USA; Technische Universität Berlin, Germany)

Info

Results Reproduced Article: cgo25main-p73-p (type: Full Paper) doi:10.1145/3696443.3708922

Artifact for the paper "The MLIR Transform Dialect - Your compiler is more powerful than you think" (doi:10.5281/zenodo.13373470): This represents the artifact to supplement the CGO 2025 paper "The MLIR Transform Dialect - Your compiler is more powerful than you think". It includes the code and scripts to facilitate the reproduction of the experiments presented in the paper. Additionally it contains the tool `mlir-transform-opt` that represents ...

Combining MLIR Dialects with Domain-Specific Architecture for Efficient Regular Expression Matching
Andrea Somaini, Filippo Carloni, Giovanni Agosta, Marco D. Santambrogio, and Davide Conficconi
(Politecnico di Milano, Italy)

Results Reproduced Article: cgo25main-p1-p (type: Full Paper) doi:10.1145/3696443.3708916

Combining MLIR Dialects with Domain-Specific Architecture for Efficient Regular Expression Matching (doi:10.5281/zenodo.13345346): This document provides all the information about the artifact to reproduce the main findings of the paper and potentially modify the proposed work for further experiments. The SW/HW setup comprehends two different environments to reproduce all the main results. The x86 machine is used for experiments on the ...

DialEgg: Dialect-Agnostic MLIR Optimizer using Equality Saturation with Egglog
Abd-El-Aziz Zayed and Christophe Dubach
(McGill University, Canada; Mila, Canada)

Results Reproduced Article: cgo25main-p229-p (type: Full Paper) doi:10.1145/3696443.3708957

dialegg-cgo-artifact (doi:10.6084/m9.figshare.27668247.v3): This artifact contains a source tree and a Dockerfile. Docker will create and build a container that includes DialEgg and its dependencies, LLVM, MLIR, and Egglog. This container is designed to reproduce this paper’s experimental results (fig. 3 and table 2).

Quantum Computing (1)

Synthesis of Quantum Simulators by Compilation
Meisam Tarabkhah, Mahshid Delavar, Mina Doosti, and Amir Shaikhha
(University of Edinburgh, UK; University of Sheffield, UK)

Publisher's Version Article: cgo25main-p133-p (type: Full Paper) doi:10.1145/3696443.3708949

Weaver: A Retargetable Compiler Framework for FPQA Quantum Architectures
Oğuzcan Kırmemiş, Francisco Romão, Emmanouil Giortamis, and Pramod Bhatotia
(TU Munich, Germany)

Results Reproduced Article: cgo25main-p61-p (type: Full Paper) doi:10.1145/3696443.3708965

"Weaver: A Retargetable Compiler Framework for FPQA Quantum Architectures" Artifact (doi:10.5281/zenodo.14368066): Our artifacts include Weaver's framework, focusing on Weaver's proposed FPQA optimization and all the other baseline FPQA compilers for comparison. These include Geyser, Atomique and DPQA. The artifact also includes code to run with Qiskit for a Superconducting compiler baseline.

Program Analysis and Synthesis

Automatic Synthesis of Specialized Hash Functions
Renato B. Hoffmann, Leonardo G. Faé, Dalvan Griebler, Xinliang David Li, and Fernando Magno Quintão Pereira
(PUC-RS, Brazil; Google, USA; Federal University of Minas Gerais, Brazil)

Results Reproduced Article: cgo25main-p51-p (type: Full Paper) doi:10.1145/3696443.3708940

Reproduction Package for Article `Automatic Synthesis of Specialized Hash Functions` (doi:10.5281/zenodo.14396233): This artifact reproduces the research questions (RQ) listed in Section 4 of the paper. All scripts to reproduce the RQs are available in a Docker container. RQ1 and RQ2 can be reproduced with a single script rq1_rq2_benchmark.sh. All other RQs have an individual script rq_benchmark.sh. For a more detailed description ...

Stack Filtering: Elevating Precision and Efficiency in Rust Pointer Analysis
Wei Li, Dongjie He, Wenguang Chen, and Jingling Xue
(UNSW, Australia; Chongqing University, China; Tsinghua University, China)

Results Reproduced Article: cgo25main-p68-p (type: Full Paper) doi:10.1145/3696443.3708921

Artifact for "Stack Filtering: Elevating Precision and Efficiency in Rust Pointer Analysis" (doi:10.5281/zenodo.13634201): This is the artifact for the CGO'25 paper titled "Stack Filtering: Elevating Precision and Efficiency in Rust Pointer Analysis". It includes a docker image and a READEME file.

SkipFlow: Improving the Precision of Points-to Analysis using Primitive Values and Predicate Edges
David Kozak, Codrut Stancu, Tomáš Vojnar, and Christian Wimmer
(Oracle Labs, Czechia; Brno University of Technology, Czechia; Oracle Labs, Switzerland; Masaryk University, Czechia; Oracle Labs, USA)

Results Reproduced Article: cgo25main-p9-p (type: Full Paper) doi:10.1145/3696443.3708932

SkipFlow: Improving the Precision of Points-to Analysis using Primitive Values and Predicate Edges - CGO 25 Artifact (doi:10.5281/zenodo.10900903): This artifact presents our work on improving the precision of points-to analysis using primitive values and predicate edges. The content are two docker images (one for aarch64, one for x64) containing GraalVM nightly build and our benchmarking infrastructure. There is no need for special hardware, everything should ...

Safety and Resilience

FastFlip: Compositional SDC Resiliency Analysis
Keyur Joshi, Rahul Singh, Tommaso Bassetto, Sarita Adve, Darko Marinov, and Sasa Misailovic
(University of Illinois at Urbana-Champaign, USA)

Publisher's Version Article: cgo25main-p45-p (type: Full Paper) doi:10.1145/3696443.3708938

MTE4JNI: A Memory Tagging Method to Protect Java Heap Memory from Illicit Native Code Access
Huinan Chen, Jiang Ma, Chun Jason Xue, and Qingan Li
(Wuhan University, China; Guangdong OPPO Mobile Telecommunications, China; MBZUAI, United Arab Emirates)

Publisher's Version Article: cgo25main-p16-p (type: Full Paper) doi:10.1145/3696443.3708933

Memory Safety Instrumentations in Practice: Usability, Performance, and Security Guarantees
Tina Jung, Fabian Ritter, and Sebastian Hack
(Saarland University, Germany)

Info

Artifacts Available Article: cgo25main-p108-p (type: Full Paper) doi:10.1145/3696443.3708926

Memory Safety Instrumentations in Practice: Usability, Performance, and Security Guarantees (Artifact) (doi:10.5281/zenodo.13345362): The artifact contains a Vagrant VM with pre-installed dependencies for MemInstrument, as well as our MemInstrument development.

Optimizations and Transformations (2)

PreFix: Optimizing the Performance of Heap-Intensive Applications
Chaitanya Mamatha Ananda, Rajiv Gupta, Sriraman Tallam, Han Shen, and Xinliang David Li
(University of California at Riverside, USA; Google, USA)

Publisher's Version Article: cgo25main-p283-p (type: Full Paper) doi:10.1145/3696443.3708960

A Priori Loop Nest Normalization: Automatic Loop Scheduling in Complex Applications
Lukas Trümper, Philipp Schaad, Berke Ates, Alexandru Calotoiu, Marcin Copik, and Torsten Hoefler
(Daisytuner, Germany; ETH Zurich, Switzerland)

Publisher's Version Article: cgo25main-p162-p (type: Full Paper) doi:10.1145/3696443.3708951

An Efficient Polynomial Multiplication Derived Implementation of Convolution in Neural Networks
Haoke Xu, Yulin Zhang, Zitong Cheng, and Xiaoming Li
(University of Delaware, USA; Minzu University of China, China)

Publisher's Version Article: cgo25main-p123-p (type: Full Paper) doi:10.1145/3696443.3708947

Quantum Computing (2)

ASDF: A Compiler for Qwerty, a Basis-Oriented Quantum Programming Language
Austin J. Adams, Sharjeel Khan, Arjun S. Bhamra, Ryan R. Abusaada, Anthony M. Cabrera, Cameron C. Hoechst, Travis S. Humble, Jeffrey S. Young, and Thomas M. Conte
(Georgia Institute of Technology, USA; Oak Ridge National Laboratory, USA)

Info

Results Reproduced Article: cgo25main-p682-p (type: Full Paper) doi:10.1145/3696443.3708966

	Appendices: Appendices with additional technical details
	ASDF: A Compiler for Qwerty, a Basis-Oriented Quantum Programming Language: Artifact (doi:10.5281/zenodo.14505385): This is the artifact the paper "ASDF: A Compiler for Qwerty, a Basis-Oriented Quantum Programming Language" to appear in CGO '25. Please see README for details.

Qubit Movement-Optimized Program Generation on Zoned Neutral Atom Processors
Enhyeok Jang, Youngmin Kim, Hyungseok Kim, Seungwoo Choi, Yipeng Huang, and Won Woo Ro
(Yonsei University, South Korea; Rutgers University, USA)

Publisher's Version Article: cgo25main-p26-p (type: Full Paper) doi:10.1145/3696443.3708937

GPU and Parallelism

Code Generation for Cryptographic Kernels using Multi-word Modular Arithmetic on GPU
Naifeng Zhang and Franz Franchetti
(Carnegie Mellon University, USA)

Results Reproduced Article: cgo25main-p126-p (type: Full Paper) doi:10.1145/3696443.3708948

Artifact for 'Code Generation for Cryptographic Kernels using Multi-word Modular Arithmetic on GPU' (doi:10.5281/zenodo.14564393): Our artifact includes the source code for multi-word modular arithmetic (MoMA), requiring NVIDIA GPUs along with nvcc and nsys nvprof for compilation and performance profiling. While installing the SPIRAL code generation system is highly recommended for full reproducibility and customization, it is not required, as we ...

CuAsmRL: Optimizing GPU SASS Schedules via Deep Reinforcement Learning
Guoliang He and Eiko Yoneki
(University of Cambridge, UK)

Results Reproduced Article: cgo25main-p77-p (type: Full Paper) doi:10.1145/3696443.3708943

artifact for CuAsmRL: Optimizing GPU SASS Schedules via Deep Reinforcement Learning (doi:10.5281/zenodo.14058861): It contains the source code to reproduce the main results

Proteus: Portable Runtime Optimization of GPU Kernel Execution with Just-in-Time Compilation
Giorgis Georgakoudis, Konstantinos Parasyris, and David Beckingsale
(Lawrence Livermore National Laboratory, USA)

Results Reproduced Article: cgo25main-p46-p (type: Full Paper) doi:10.1145/3696443.3708939

Olympus-HPC/proteus CGO25 Artifact (doi:10.5281/zenodo.14087064): This release contains the CGO25 artifact version of Proteus, including build scripts, benchmark programs, experiment workflow scripts, and visualization scripts to re-create figures and tables of the associated manuscript.

Security, Fault Tolerance, and Cryptography

Qiwu: Exploiting Ciphertext-Level SIMD Parallelism in Homomorphic Encryption Programs
Zhongcheng Zhang, Ying Liu, Yuyang Zhang, Zhenchuan Chen, Jiacheng Zhao, Xiaobing Feng, Huimin Cui, and Jingling Xue
(Institute of Computing Technology at Chinese Academy of Sciences, China; University of Chinese Academy of Sciences, China; Zhongguancun Laboratory, China; UNSW, Australia)

Results Reproduced Article: cgo25main-p14-p (type: Full Paper) doi:10.1145/3696443.3708917

Reproduction Package for Article `Qiwu: Exploiting Ciphertext-Level SIMD Parallelism in Homomorphic Encryption Programs` (doi:10.5281/zenodo.13340367): The artifact contains the necessary software components to validate the main results in this paper. We provide a docker image to ease the environment setup. The Docker image includes the source code of Qiwu, scripts for running all benchmarks, drawing figures, and displaying tables. It requires a Linux system equipped ...

Cage: Hardware-Accelerated Safe WebAssembly
Martin Fink, Dimitrios Stavrakakis, Dennis Sprokholt, Soham Chakraborty, Jan-Erik Ekberg, and Pramod Bhatotia
(TU Munich, Germany; TU Delft, Netherlands; Huawei, Finland)

Results Reproduced Article: cgo25main-p64-p (type: Full Paper) doi:10.1145/3696443.3708920

Artifacts for "Cage: Hardware-Accelerated Safe WebAssembly" (doi:10.5281/zenodo.13772996): The artifact contains the source code (LLVM, wasmtime, wasi-libc) and benchmarking scripts required to build and reproduce all results in the paper.

Teapot: Efficiently Uncovering Spectre Gadgets in COTS Binaries
Fangzheng Lin, Zhongfa Wang, and Hiroshi Sasaki
(Institute of Science Tokyo, Japan)

Artifacts Reusable Article: cgo25main-p25-p (type: Full Paper) doi:10.1145/3696443.3708936

Teapot: Efficiently Uncovering Spectre Gadgets in COTS Binaries (doi:10.5281/zenodo.14507732): The artifact contains Teapot, the first Spectre gadget detector for COTS binaries with comparable performance to compiler-based detectors. Teapot is composed of a static binary rewriter and a runtime support library. The binary programs of interest are first statically instrumented by Teapot, then dynamically fuzzed ...

Janitizer: Rethinking Binary Tools for Practical and Comprehensive Security
Mahwish Arif, Sam Ainsworth, and Timothy M. Jones
(University of Cambridge, UK; University of Edinburgh, UK)

Publisher's Version Article: cgo25main-p3-p (type: Full Paper) doi:10.1145/3696443.3708930

Parallaft: Runtime-Based CPU Fault Tolerance via Heterogeneous Parallelism
Boyue Zhang, Sam Ainsworth, Lev Mukhanov, and Timothy M. Jones
(University of Cambridge, UK; University of Edinburgh, UK; Queen Mary University of London, UK)

Results Reproduced Article: cgo25main-p106-p (type: Full Paper) doi:10.1145/3696443.3708946

Artefact evaluation package for "Parallaft: Runtime-based CPU Fault Tolerance via Heterogeneous Parallelism" (doi:10.5281/zenodo.14172159): Parallaft is a run-time CPU fault tolerance technique that exploits heterogeneous parallelism to minimize power and performance overhead on heterogeneous CPUs. This package provides the source code of Parallaft and supporting scripts to reproduce key performance and energy overhead results from our paper "Parallaft: ...

Optimizations and Transformations (3)

Postiz: Extending Post-increment Addressing for Loop Optimization and Code Size Reduction
Enming Fan, Xiaofeng Guan, Fan Hu, Heng Shi, Hao Zhou, and Jianguo Yao
(Shanghai Enflame Technology, China; Shanghai Jiao Tong University, China)

Publisher's Version Article: cgo25main-p21-p (type: Full Paper) doi:10.1145/3696443.3708935

Towards Efficient Compiler Auto-tuning: Leveraging Synergistic Search Spaces
Haolin Pan, Yuanyu Wei, Mingjie Xing, Yanjun Wu, and Chen Zhao
(Institute of Software at Chinese Academy of Sciences, China; Hangzhou Institute for Advanced Study at University of Chinese Academy of Sciences, China; University of Chinese Academy of Sciences, China)

Results Reproduced Article: cgo25main-p323-p (type: Full Paper) doi:10.1145/3696443.3708961

Artifact for Leveraging Synergistic Search Spaces paper at CGO2025 (doi:10.5281/zenodo.14351920): This is a compressed Docker container image file capable of reproducing the main experiments described in the article.

Stardust: Compiling Sparse Tensor Algebra to a Reconfigurable Dataflow Architecture
Olivia Hsu, Alexander Rucker, Tian Zhao, Varun Desai, Kunle Olukotun, and Fredrik Kjolstad
(Stanford University, USA)

Publisher's Version Article: cgo25main-p34-p (type: Full Paper) doi:10.1145/3696443.3708918

Vectron: A Dynamic Programming Auto-vectorization Framework
Sourena Naser Moghaddasi, Haris Smajlović, Ariya Shajii, and Ibrahim Numanagić
(University of Victoria, Canada; Exaloop, USA)

Artifacts Functional Article: cgo25main-p124-p (type: Full Paper) doi:10.1145/3696443.3708963

Reproduction Package for "Vectron: A Dynamic Programming Auto-Vectorization Framework" (doi:10.5281/zenodo.13347575): The Vectron artifact and the exact steps for reproducing its experimental section can be found on GitHub (https://github.com/0xTCG/vectron) in a dockerdirectory. A Zenodo mirror is also provided at doi:10.5281/zenodo.13347575 .

Runtime and System Tools

Honey Potion: An eBPF Backend for Elixir
Kael Soares Augusto, Vinícius Pacheco, Marcos A. Vieira, Rodrigo Geraldo Ribeiro, and Fernando Magno Quintão Pereira
(Federal University of Minas Gerais, Brazil; Cadence, Brazil; Federal University of Ouro Preto, Brazil)

Results Reproduced Article: cgo25main-p81-p (type: Full Paper) doi:10.1145/3696443.3708923

Reproduction package for Honey Potion (doi:10.5281/zenodo.13729837): Reproduction package for Honey Potion

GoFree: Reducing Garbage Collection via Compiler-Inserted Freeing
Haoran Peng, Yu Zhang, Michael D. Ernst, Jinbao Chen, and Boyao Ding
(University of Science and Technology of China, China; University of Washington, USA)

Results Reproduced Article: cgo25main-p97-p (type: Full Paper) doi:10.1145/3696443.3708925

Artifact for CGO'25 paper "GoFree: Reducing Garbage Collection via Compiler-inserted Freeing" (doi:10.6084/m9.figshare.26785357.v1): Artifact for CGO'25 paper "GoFree: Reducing Garbage Collection via Compiler-inserted Freeing". Reusable and reproduces the paper's evaluation section.

Improving Native-Image Startup Performance
Matteo Basso, Aleksandar Prokopec, Andrea Rosà, and Walter Binder
(USI Lugano, Switzerland; Oracle Labs, Switzerland)

Results Reproduced Article: cgo25main-p121-p (type: Full Paper) doi:10.1145/3696443.3708927

Artifact associated to the paper "Improving Native-Image Startup Performance" published in CGO'25 (doi:10.5281/zenodo.13760307): This artifact consists of a ready-to-use Docker image embedding our profiler as well as our modified GraalVM to generate optimized Native-Image binaries that reduce I/O traffic by changing their layout during compilation. A set of tools/scripts can be used to execute the benchmarks, collect, process, and plot page ...

Speeding up the Local C++ Development Cycle with Header Substitution
Nader Al Awar, Zijian Yi, George Biros, and Milos Gligoric
(University of Texas at Austin, USA)