Supersonic: Learning to generate source code optimizations in C/C++
Z Chen, S Fang, M Monperrus - IEEE Transactions on Software …, 2024 - ieeexplore.ieee.org
Software optimization refines programs for resource efficiency while preserving functionality.
Traditionally, it is a process done by developers and compilers. This paper introduces a third …
Traditionally, it is a process done by developers and compilers. This paper introduces a third …
Mlgoperf: An ml guided inliner to optimize performance
AH Ashouri, M Elhoushi, Y Hua, X Wang… - arXiv preprint arXiv …, 2022 - arxiv.org
For the past 25 years, we have witnessed an extensive application of Machine Learning to
the Compiler space; the selection and the phase-ordering problem. However, limited works …
the Compiler space; the selection and the phase-ordering problem. However, limited works …
GraalSP: Polyglot, efficient, and robust machine learning-based static profiler
M Čugurović, MV Janičić, V Jovanović… - Journal of Systems and …, 2024 - Elsevier
Compilers use profiles to apply profile-guided optimizations and produce efficient programs.
Dynamic profilers collect high-quality profiles but require identifying suitable profile …
Dynamic profilers collect high-quality profiles but require identifying suitable profile …
Facilitating hardware-aware neural architecture search with learning-based predictive models
Neural architecture search (NAS), which automatically explores the efficient model design,
has achieved ground-breaking advances in recent years. To achieve the optimal model …
has achieved ground-breaking advances in recent years. To achieve the optimal model …
Predicting dynamic properties of heap allocations using neural networks trained on static code: An intellectual abstract
Memory allocators and runtime systems can leverage dynamic properties of heap
allocations–such as object lifetimes, hotness or access correlations–to improve performance …
allocations–such as object lifetimes, hotness or access correlations–to improve performance …
ACPO: AI-Enabled Compiler-Driven Program Optimization
AH Ashouri, MA Manzoor, DM Vu, R Zhang… - arXiv preprint arXiv …, 2023 - arxiv.org
The key to performance optimization of a program is to decide correctly when a certain
transformation should be applied by a compiler. Traditionally, such profitability decisions are …
transformation should be applied by a compiler. Traditionally, such profitability decisions are …
Stale Profile Matching
A Ayupov, M Panchenko, S Pupyrev - Proceedings of the 33rd ACM …, 2024 - dl.acm.org
Profile-guided optimizations rely on profile data for directing compilers to generate optimized
code. To achieve the maximum performance boost, profile data needs to be collected on the …
code. To achieve the maximum performance boost, profile data needs to be collected on the …
Identifying and Exploiting Sparse Branch Correlations for Optimizing Branch Prediction
Branch prediction is arguably one of the most important speculative mechanisms within a
high-performance processor architecture. A common approach to improve branch prediction …
high-performance processor architecture. A common approach to improve branch prediction …