CC 2023 – Proceedings

Frontmatter

Vector and Parallelism

Java Vector API: Benchmarking and Performance Analysis
Matteo Basso, Andrea Rosà, Luca Omini, and Walter Binder
(USI Lugano, Lugano, Switzerland)


@InProceedings{CC23p1, author = {Matteo Basso and Andrea Rosà and Luca Omini and Walter Binder}, title = {Java Vector API: Benchmarking and Performance Analysis}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {1-0}, doi = {10.1145/3578360.3580265}, year = {2023}, } Publisher's Version Published Artifact Artifacts Available Artifacts Reusable Results Reproduced Article: cc23main-p58-p doi:10.1145/3578360.3580265

Compiling Discrete Probabilistic Programs for Vectorized Exact Inference
Jingwen Pan and Amir Shaikhha
(University of Edinburgh, Edinburgh, UK)


@InProceedings{CC23p15, author = {Jingwen Pan and Amir Shaikhha}, title = {Compiling Discrete Probabilistic Programs for Vectorized Exact Inference}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {15-14}, doi = {10.1145/3578360.3580258}, year = {2023}, } Publisher's Version Article: cc23main-p25-p doi:10.1145/3578360.3580258

A Multi-threaded Fast Hardware Compiler for HDLs
Sheng-Hong Wang, Hunter James Coffman, Kenneth Mayer, Sakshi Garg, and Jose Renau
(University of California, Santa Cruz, USA)


@InProceedings{CC23p29, author = {Sheng-Hong Wang and Hunter James Coffman and Kenneth Mayer and Sakshi Garg and Jose Renau}, title = {A Multi-threaded Fast Hardware Compiler for HDLs}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {29-28}, doi = {10.1145/3578360.3580254}, year = {2023}, } Publisher's Version Article: cc23main-p14-p doi:10.1145/3578360.3580254

Scheduling and Tuning

Efficiently Learning Locality Optimizations by Decomposing Transformation Domains
Tharindu R. Patabandi and Mary Hall
(University of Utah, USA)


@InProceedings{CC23p43, author = {Tharindu R. Patabandi and Mary Hall}, title = {Efficiently Learning Locality Optimizations by Decomposing Transformation Domains}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {43-42}, doi = {10.1145/3578360.3580272}, year = {2023}, } Publisher's Version Published Artifact Artifacts Available Artifacts Reusable Results Reproduced Article: cc23main-p73-p doi:10.1145/3578360.3580272

A Deep Learning Model for Loop Interchange
Lina Mezdour, Khadidja Kadem, Massinissa Merouani, Amina Selma Haichour, Saman Amarasinghe, and Riyadh Baghdadi
(NYU Abu Dhabi, Abu Dhabi, United Arab Emirates; ESI, Algiers, Algeria; Massachusetts Institute of Technology, USA)


@InProceedings{CC23p57, author = {Lina Mezdour and Khadidja Kadem and Massinissa Merouani and Amina Selma Haichour and Saman Amarasinghe and Riyadh Baghdadi}, title = {A Deep Learning Model for Loop Interchange}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {57-56}, doi = {10.1145/3578360.3580257}, year = {2023}, } Publisher's Version Published Artifact Artifacts Available Artifacts Reusable Article: cc23main-p23-p doi:10.1145/3578360.3580257

(De/Re)-Compositions Expressed Systematically via MDH-Based Schedules
Ari Rasch, Richard Schulze, Denys Shabalin, Anne Elster, Sergei Gorlatch, and Mary Hall
(University of Muenster, Muenster, Germany; Google, Switzerland; NTNU, Trondheim, Norway; University of Utah, USA)


@InProceedings{CC23p71, author = {Ari Rasch and Richard Schulze and Denys Shabalin and Anne Elster and Sergei Gorlatch and Mary Hall}, title = {(De/Re)-Compositions Expressed Systematically via MDH-Based Schedules}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {71-70}, doi = {10.1145/3578360.3580269}, year = {2023}, } Publisher's Version Published Artifact Artifacts Available Artifacts Reusable Results Reproduced Article: cc23main-p62-p doi:10.1145/3578360.3580269

Code Generation and Synthesis

A Sound and Complete Algorithm for Code Generation in Distance-Based ISA
Shu Sugita, Toru Koizumi, Ryota Shioya, Hidetsugu Irie, and Shuichi Sakai
(University of Tokyo, Tokyo, Japan)


@InProceedings{CC23p85, author = {Shu Sugita and Toru Koizumi and Ryota Shioya and Hidetsugu Irie and Shuichi Sakai}, title = {A Sound and Complete Algorithm for Code Generation in Distance-Based ISA}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {85-84}, doi = {10.1145/3578360.3580263}, year = {2023}, } Publisher's Version Article: cc23main-p51-p doi:10.1145/3578360.3580263

Matching Linear Algebra and Tensor Code to Specialized Hardware Accelerators
Pablo Antonio Martínez, Jackson Woodruff, Jordi Armengol-Estapé, Gregorio Bernabé, José Manuel García, and Michael F. P. O’Boyle
(University of Murcia, Murcia, Spain; University of Edinburgh, Edinburgh, UK)


@InProceedings{CC23p99, author = {Pablo Antonio Martínez and Jackson Woodruff and Jordi Armengol-Estapé and Gregorio Bernabé and José Manuel García and Michael F. P. O’Boyle}, title = {Matching Linear Algebra and Tensor Code to Specialized Hardware Accelerators}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {99-98}, doi = {10.1145/3578360.3580262}, year = {2023}, } Publisher's Version Published Artifact Artifacts Available Artifacts Reusable Results Reproduced Article: cc23main-p40-p doi:10.1145/3578360.3580262

Torchy: A Tracing JIT Compiler for PyTorch
Nuno P. Lopes
(INESC-ID, Lisbon, Portugal; Instituto Superior Técnico - University of Lisbon, Lisbon, Portugal)


@InProceedings{CC23p113, author = {Nuno P. Lopes}, title = {Torchy: A Tracing JIT Compiler for PyTorch}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {113-112}, doi = {10.1145/3578360.3580266}, year = {2023}, } Publisher's Version Article: cc23main-p59-p doi:10.1145/3578360.3580266

Backend

A Symbolic Emulator for Shuffle Synthesis on the NVIDIA PTX Code
Kazuaki Matsumura, Simon Garcia De Gonzalo, and Antonio J. Peña
(Barcelona Supercomputing Center, Barcelona, Spain; Sandia National Laboratories, USA)


@InProceedings{CC23p127, author = {Kazuaki Matsumura and Simon Garcia De Gonzalo and Antonio J. Peña}, title = {A Symbolic Emulator for Shuffle Synthesis on the NVIDIA PTX Code}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {127-126}, doi = {10.1145/3578360.3580253}, year = {2023}, } Publisher's Version Published Artifact Artifacts Available Artifacts Reusable Results Reproduced Article: cc23main-p5-p doi:10.1145/3578360.3580253

Register Allocation for Compressed ISAs in LLVM
Andreas Fried, Maximilian Stemmer-Grabow, and Julian Wachter
(KIT, Karlsruhe, Germany)


@InProceedings{CC23p141, author = {Andreas Fried and Maximilian Stemmer-Grabow and Julian Wachter}, title = {Register Allocation for Compressed ISAs in LLVM}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {141-140}, doi = {10.1145/3578360.3580261}, year = {2023}, } Publisher's Version Published Artifact Artifacts Available Artifacts Reusable Results Reproduced Article: cc23main-p38-p doi:10.1145/3578360.3580261

RL4ReAl: Reinforcement Learning for Register Allocation
S. VenkataKeerthy, Siddharth Jain, Anilava Kundu, Rohit Aggarwal, Albert Cohen, and Ramakrishna Upadrasta
(IIT Hyderabad, Hyderabad, India; Google, France)


@InProceedings{CC23p155, author = {S. VenkataKeerthy and Siddharth Jain and Anilava Kundu and Rohit Aggarwal and Albert Cohen and Ramakrishna Upadrasta}, title = {RL4ReAl: Reinforcement Learning for Register Allocation}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {155-154}, doi = {10.1145/3578360.3580273}, year = {2023}, } Publisher's Version Published Artifact Artifacts Available Artifacts Reusable Results Reproduced Article: cc23main-p74-p doi:10.1145/3578360.3580273

Code Size and Bugs

Automatically Localizing Dynamic Code Generation Bugs in JIT Compiler Back-End
HeuiChan Lim and Saumya Debray
(University of Arizona, USA)


@InProceedings{CC23p169, author = {HeuiChan Lim and Saumya Debray}, title = {Automatically Localizing Dynamic Code Generation Bugs in JIT Compiler Back-End}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {169-168}, doi = {10.1145/3578360.3580260}, year = {2023}, } Publisher's Version Article: cc23main-p31-p doi:10.1145/3578360.3580260

HyBF: A Hybrid Branch Fusion Strategy for Code Size Reduction
Rodrigo C. O. Rocha, Charitha Saumya, Kirshanthan Sundararajah, Pavlos Petoumenos, Milind Kulkarni, and Michael F. P. O’Boyle
(University of Edinburgh, Edinburgh, UK; Purdue University, USA; University of Manchester, Manchester, UK)


@InProceedings{CC23p183, author = {Rodrigo C. O. Rocha and Charitha Saumya and Kirshanthan Sundararajah and Pavlos Petoumenos and Milind Kulkarni and Michael F. P. O’Boyle}, title = {HyBF: A Hybrid Branch Fusion Strategy for Code Size Reduction}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {183-182}, doi = {10.1145/3578360.3580267}, year = {2023}, } Publisher's Version Published Artifact Archive submitted (520 kB) Artifacts Available Artifacts Reusable Results Reproduced Article: cc23main-p60-p doi:10.1145/3578360.3580267

Linker Code Size Optimization for Native Mobile Applications
Gai Liu, Umar Farooq, Chengyan Zhao, Xia Liu, and Nian Sun
(ByteDance, USA; ByteDance, China)


@InProceedings{CC23p197, author = {Gai Liu and Umar Farooq and Chengyan Zhao and Xia Liu and Nian Sun}, title = {Linker Code Size Optimization for Native Mobile Applications}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {197-196}, doi = {10.1145/3578360.3580256}, year = {2023}, } Publisher's Version Article: cc23main-p22-p doi:10.1145/3578360.3580256

Domain Specific Languages

Building a Compiled Query Engine in Python
Hesam Shahrokhi and Amir Shaikhha
(University of Edinburgh, Edinburgh, UK)


@InProceedings{CC23p211, author = {Hesam Shahrokhi and Amir Shaikhha}, title = {Building a Compiled Query Engine in Python}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {211-210}, doi = {10.1145/3578360.3580264}, year = {2023}, } Publisher's Version Article: cc23main-p53-p doi:10.1145/3578360.3580264

Codon: A Compiler for High-Performance Pythonic Applications and DSLs
Ariya Shajii, Gabriel Ramirez, Haris Smajlović, Jessica Ray, Bonnie Berger, Saman Amarasinghe, and Ibrahim Numanagić
(Exaloop, USA; Massachusetts Institute of Technology, USA; University of Victoria, Canada)


@InProceedings{CC23p225, author = {Ariya Shajii and Gabriel Ramirez and Haris Smajlović and Jessica Ray and Bonnie Berger and Saman Amarasinghe and Ibrahim Numanagić}, title = {Codon: A Compiler for High-Performance Pythonic Applications and DSLs}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {225-224}, doi = {10.1145/3578360.3580275}, year = {2023}, } Publisher's Version Archive submitted (960 kB) Article: cc23main-p69-p doi:10.1145/3578360.3580275

MOD2IR: High-Performance Code Generation for a Biophysically Detailed Neuronal Simulation DSL
George Mitenkov, Ioannis Magkanaris, Omar Awile, Pramod Kumbhar, Felix Schürmann, and Alastair F. Donaldson
(Imperial College London, London, UK; EPFL, Lausanne, Switzerland)


@InProceedings{CC23p239, author = {George Mitenkov and Ioannis Magkanaris and Omar Awile and Pramod Kumbhar and Felix Schürmann and Alastair F. Donaldson}, title = {MOD2IR: High-Performance Code Generation for a Biophysically Detailed Neuronal Simulation DSL}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {239-238}, doi = {10.1145/3578360.3580268}, year = {2023}, } Publisher's Version Published Artifact Artifacts Available Artifacts Reusable Results Reproduced Article: cc23main-p61-p doi:10.1145/3578360.3580268

Optimizations

A Hotspot-Driven Semi-automated Competitive Analysis Framework for Identifying Compiler Key Optimizations
Wenlong Mu, Yilei Zhang, Bo Huang, Jianmei Guo, and Shiqiang Cui
(East China Normal University, China; Hangzhou Hongjun Microelectronics Technology, China)


@InProceedings{CC23p253, author = {Wenlong Mu and Yilei Zhang and Bo Huang and Jianmei Guo and Shiqiang Cui}, title = {A Hotspot-Driven Semi-automated Competitive Analysis Framework for Identifying Compiler Key Optimizations}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {253-252}, doi = {10.1145/3578360.3580255}, year = {2023}, } Publisher's Version Published Artifact Artifacts Available Artifacts Reusable Results Reproduced Article: cc23main-p20-p doi:10.1145/3578360.3580255

LAGrad: Statically Optimized Differentiable Programming in MLIR
Mai Jacob Peng and Christophe Dubach
(McGill University, Canada; Mila, Canada)


@InProceedings{CC23p267, author = {Mai Jacob Peng and Christophe Dubach}, title = {LAGrad: Statically Optimized Differentiable Programming in MLIR}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {267-266}, doi = {10.1145/3578360.3580259}, year = {2023}, } Publisher's Version Published Artifact Artifacts Available Artifacts Reusable Results Reproduced Article: cc23main-p26-p doi:10.1145/3578360.3580259

Lazy Evaluation for the Lazy: Automatically Transforming Call-by-Value into Call-by-Need
Breno Campos Ferreira Guimarães and Fernando Magno Quintão Pereira
(Federal University of Minas Gerais, Brazil)


@InProceedings{CC23p281, author = {Breno Campos Ferreira Guimarães and Fernando Magno Quintão Pereira}, title = {Lazy Evaluation for the Lazy: Automatically Transforming Call-by-Value into Call-by-Need}, booktitle = {Proc.\ CC}, publisher = {ACM}, pages = {281-280}, doi = {10.1145/3578360.3580270}, year = {2023}, } Publisher's Version Article: cc23main-p65-p doi:10.1145/3578360.3580270

proc time: 0.05