Large language models for compiler optimization
We explore the novel application of Large Language Models to code optimization. We
present a 7B-parameter transformer model trained from scratch to optimize LLVM assembly …
present a 7B-parameter transformer model trained from scratch to optimize LLVM assembly …
TpuGraphs: A Performance Prediction Dataset on Large Tensor Computational Graphs
M Phothilimthana, S Abu-El-Haija… - Advances in …, 2024 - proceedings.neurips.cc
Precise hardware performance models play a crucial role in code optimizations. They can
assist compilers in making heuristic decisions or aid autotuners in identifying the optimal …
assist compilers in making heuristic decisions or aid autotuners in identifying the optimal …
TelaMalloc: Efficient On-Chip Memory Allocation for Production Machine Learning Accelerators
Memory buffer allocation for on-chip memories is a major challenge in modern machine
learning systems that target ML accelerators. In interactive systems such as mobile phones …
learning systems that target ML accelerators. In interactive systems such as mobile phones …
Neural architecture search using property guided synthesis
Neural architecture search (NAS) has become an increasingly important tool within the deep
learning community in recent years, yielding many practical advancements in the design of …
learning community in recent years, yielding many practical advancements in the design of …
Compiler generated feedback for Large Language Models
We introduce a novel paradigm in compiler optimization powered by Large Language
Models with compiler feedback to optimize the code size of LLVM assembly. The model …
Models with compiler feedback to optimize the code size of LLVM assembly. The model …
ALT: Breaking the Wall between Data Layout and Loop Optimizations for Deep Learning Compilation
Deep learning models rely on highly optimized tensor libraries for efficient inference on
heterogeneous hardware. Current deep compilers typically predetermine layouts of tensors …
heterogeneous hardware. Current deep compilers typically predetermine layouts of tensors …
Compiler optimization prediction with new self-improved optimization model
Users may now choose from a vast range of compiler optimizations. These optimizations
interact in a variety of sophisticated ways with one another and with the source code. The …
interact in a variety of sophisticated ways with one another and with the source code. The …
The Droplet Search Algorithm for Kernel Scheduling
Kernel scheduling is the problem of finding the most efficient implementation for a
computational kernel. Identifying this implementation involves experimenting with the …
computational kernel. Identifying this implementation involves experimenting with the …
Optimizing memory mapping using deep reinforcement learning
P Wang, M Sazanovich, B Ilbeyi… - arXiv preprint arXiv …, 2023 - arxiv.org
Resource scheduling and allocation is a critical component of many high impact systems
ranging from congestion control to cloud computing. Finding more optimal solutions to these …
ranging from congestion control to cloud computing. Finding more optimal solutions to these …
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 …