CC 2026 – Proceedings

Info Article: cc26main-p53-p (type: Full Paper) doi:10.1145/3771775.3786276

It’s about Time: Temporal Abstractions for Asynchronous GPU Tensor Computations
Bastian Hagedorn and Vinod Grover
(NVIDIA, Germany; NVIDIA, USA)

Publisher's Version Article: cc26main-p56-p (type: Full Paper) doi:10.1145/3771775.3786277

Supplementary Material for: It’s about Time: Temporal Abstractions for Asynchronous GPU Tensor Computations: The supplementary material provides additional technical details including formal type inference rules for the Async Graphene type system, illustrative examples comparing different abstraction approaches, detailed lowering traces showing how high-level primitives translate to CUDA code, and a complete reference ...

Optimizing Sparse Tensor Compilation for Sparse Output
Shideh Hashemian, Michael F. P. O’Boyle, and Amir Shaikhha
(University of Edinburgh, UK)

Publisher's Version Article: cc26main-p16-p (type: Full Paper) doi:10.1145/3771775.3786267

RIFS: Run-Time Invariant Function Specialization
Saba Jamilan, Snehasish Kumar, and Heiner Litz
(University of California at Santa Cruz, USA; Google, USA)

Artifacts Available Article: cc26main-p46-p (type: Full Paper) doi:10.1145/3771775.3786274

RIFS (doi:10.5281/zenodo.18353335): The LLVM Passes, bash scripts, codes for RIFS

Optimizations for Safety and More

DiTOX: Fault Detection and Localization in the ONNX Optimizer
Nikolaos Louloudakis and Ajitha Rajan
(University of Edinburgh, UK)

Info

Results Reproduced Article: cc26main-p10-p (type: Full Paper) doi:10.1145/3771775.3786265

	Appendix: This appendix accompanies the paper "DiTOX: Fault Detection and Localization in the ONNX Optimizer." It provides additional details about the DiTOX system, including general information, installation and usage instructions, and guidance for reproducing several of the experiments presented in the paper.
	DiTOX: Differential Testing of the ONNX Optimizer (doi:10.5281/zenodo.17990604): DiTOX is a utility that enables differential testing of the ONNX Optimizer, by fetching automatically real-life ONNX models from the ONNX Model Hub and performing full and per-pass differential testing on the ONNX Optimizer.

SSMR: Statically Detecting Speculation Safe Memory Regions to Mitigate Transient Execution Attacks
Ange-Thierry Ishimwe, Sam McDiarmid-Sterling, Zack McKevitt, and Tamara Silbergleit Lehman
(University of Colorado Boulder, USA)

Artifacts Available Article: cc26main-p38-p (type: Full Paper) doi:10.1145/3771775.3786272

[Artifact] SSMR: Statically Detecting Speculation Safe Memory Regions to Mitigate Transient Execution Attacks (doi:10.5281/zenodo.17959090): This artifact provides source code for the SSMR (Speculation-Safe Memory Region) implementation. It demonstrates how SSMR is integrated across the compiler and how it is supported in the microarchitecture. The artifact contains: LLVM modifications that add compiler and analysis support for SSMR. gem5 modifications ...

CHEHAB: Automatic Compiler Code Optimization for Fully Homomorphic Encryption
Abdessamed Seddiki, Arab Mohammed, Zakaria Hebbal, Aimad Chabounia, Eduardo Chielle, Karima Benatchba, Challal Yacine, Djamel Eddine Menacer, Michail Maniatakos, and Riyadh Baghdadi
(New York University Abu Dhabi, United Arab Emirates; École Nationale Supérieure d’Informatique, Algeria; University of Doha for Science and Technology, Qatar)

Publisher's Version Article: cc26main-p28-p (type: Full Paper) doi:10.1145/3771775.3786269

Appendix: A pdf file containing extra details about the paper

Parallel and Customizable Equality Saturation
Jonathan Van der Cruysse, Abd-El-Aziz Zayed, Mai Jacob Peng, and Christophe Dubach
(McGill University, Canada)

Results Reproduced Article: cc26main-p15-p (type: Full Paper) doi:10.1145/3771775.3786266

	Supplementary material for "Parallel and Customizable Equality Saturation": Artifact instructions and implementation details for generalized metadata as discussed in the paper.
	Artifact for "Parallel and Customizable Equality Saturation" (doi:10.5281/zenodo.17955956): This repository contains the evaluation artifact for the CC ’26 paper "Parallel and Customizable Equality Saturation"

Code Generation and Tuning

Accelerating Sparse Algebra with Program Synthesis
José Wesley de Souza Magalhães, Shideh Hashemian, Alexander Brauckmann, Jackson Woodruff, Elizabeth Polgreen, and Michael F. P. O’Boyle
(University of Edinburgh, UK)

Publisher's Version Article: cc26main-p63-p (type: Full Paper) doi:10.1145/3771775.3786281

Schedgehammer: Auto-tuning Compiler Optimizations beyond Numerical Parameters
Johannes Lenfers, Sven Spehr, Justus Dieckmann, Johannes Jansen, Martin Paul Lücke, and Sergei Gorlatch
(University of Münster, Germany; Daisytuner, Germany; AMD, Germany)

Results Reproduced Article: cc26main-p64-p (type: Full Paper) doi:10.1145/3771775.3786282

Artifact "Schedgehammer: Auto-tuning Compiler Optimizations beyond Numerical Parameters" (doi:10.5281/zenodo.17970249): This artifact provides a complete implementation, bench- mark suites, and reproduction scripts for Schedgehammer, a framework for general-purpose auto-scheduling that ab- stracts user-schedulable languages as a standalone, graph- structured problem. It is designed to demonstrate flexibility in auto-scheduling tasks ...

TinyGen: Portable and Compact Code Generation for Tiny Machine Learning
Gaeun Ko and Seonyeong Heo
(Kyung Hee University, Republic of Korea)

Info

Results Reproduced Article: cc26main-p57-p (type: Full Paper) doi:10.1145/3771775.3786278

CPerfSmith: A Randomized C Program Generator for Performance-Oriented Compiler Testing
Yashwanth Boda, Abhijit Chunduri, Ruchi Kumari, and Awanish Pandey
(IIT Roorkee, India)

Publisher's Version Article: cc26main-p35-p (type: Full Paper) doi:10.1145/3771775.3786271

Tools

Inside VOLT: Designing an Open-Source GPU Compiler (Tool)
Shinnung Jeong, Chihyo Ahn, Huanzhi Pu, Jisheng Zhao, Hyesoon Kim, and Blaise Tine
(Georgia Institute of Technology, USA; University of California at Los Angeles, USA)

Info

Results Reproduced Article: cc26main-p49-p (type: Full Paper) doi:10.1145/3771775.3786275

VOLT: Vortex-Optimized Lightweight Toolchain (doi:10.5281/zenodo.18071633): VOLT is an end-to-end, open-source compiler framework for the Vortex open-source GPU and its variants. The artifact DOI covers the toolchain, documentation, and evaluation scripts. The artifact evaluation uses the Vortex cycle-level simulator (SimX). Documentation is provided under Volt/docs and covers (i) component ...

Nsight Python: A Python-First Profiling Toolkit for Seamless GPU Kernel Analysis (Tool)
Bastian Hagedorn, Alexander Collins, Tony Mongkolsmai, and Vinod Grover
(NVIDIA, Germany; NVIDIA, UK; NVIDIA, USA)

Results Reproduced Article: cc26main-p42-p (type: Full Paper) doi:10.1145/3771775.3786273

Nsight Python CC'26 Artifact (doi:10.5281/zenodo.17942373): This artifact provides a complete environment for reproducing the experimental results presented in our paper. The artifact includes five experiments corresponding to the key results in the paper. Each experiment is self-contained with Python scripts that generate performance plots and CSV data files using the Nsight ...

Analysis

HORIZON: Estimating Alias Analysis Precision Bounds and Their Impact on Performance
Khushboo Chitre, Piyus Kedia, and Rahul Purandare
(IIIT Delhi, India; University of Nebraska-Lincoln, USA)

Results Reproduced Article: cc26main-p33-p (type: Full Paper) doi:10.1145/3771775.3786270

Reproduction package for article "HORIZON: Estimating Alias Analysis Precision Bounds and Their Impact on Performance" (doi:10.5281/zenodo.17999887): Horizon artifact.

Type Deduction Analysis: Reconstructing Transparent Pointer Types in LLVM-IR
Niccolò Nicolosi, Gabriele Magnani, Emilio Corigliano, Davide Baroffio, Federico Reghenzani, and Giovanni Agosta
(Politecnico di Milano, Italy)

Artifacts Available Article: cc26main-p26-p (type: Full Paper) doi:10.1145/3771775.3786268

Type Deduction Analysis Supplementary Material (doi:10.5281/zenodo.17570563): This repository contains the supplementary material of Type Deduction Analysis, including the source code and the raw results of the experimental evaluation.

Compact Representation and Interleaved Solving for Scalable Constraint-Based Points-to Analysis
Ramya Kasaraneni and V. Krishna Nandivada
(IIT Madras, India)

Results Reproduced Article: cc26main-p61-p (type: Full Paper) doi:10.1145/3771775.3786280

Compact Representation and Interleaved Solving for Scalable Constraint-Based Points-to Analysis - artifact (doi:10.6084/m9.figshare.30925898.v3): Compact Representation and Interleaved Solving for Scalable Constraint-Based Points-to Analysis - artifact

Practical MHP Analysis for Java
A. Samuel Moses and V. Krishna Nandivada
(IIT Madras, India)