CRUSH: A Credit-Based Approach for Functional Unit Sharing in Dynamically Scheduled HLS
Jiahui Xu, Lana Josipovic
2025年份
1被引次数
摘要
Dynamically scheduled high-level synthesis (HLS) automatically translates software code (e.g., C/C++) to dataflow circuits-networks of compute units that communicate via handshake signals. These signals schedule the circuit during runtime, allowing them to handle irregular control flow or unpredictable memory accesses efficiently, thus giving them performance merit over statically scheduled circuits produced by standard HLS tools.
问问这篇 Paper
问问你的智能体。
Lune 读过与它相关的顶会 Paper,每个回答都会注明依据哪几篇。
相关 Paper
- PipeLink: A Pipelined Resource Sharing System for Dataflow High-Level SynthesisRui Li, Lincoln Berkley, Rajit ManoharDAC 2025
- Optimizing Recovery Logic in Speculative High-Level SynthesisDylan Leothaud, Jean-Michel Gorius, Simon Rokicki, Steven DerrienDAC 2025
- A Mechanized Semantics for Dataflow CircuitsTony Law, Delphine Demange, Sandrine BlazyOOPSLA 2025 · 被引用 2 次
- An Iterative Method for Mapping-Aware Frequency Regulation in Dataflow CircuitsCarmine Rizzi, Andrea Guerrieri, Lana JosipovicDAC 2023 · 被引用 5 次
- NeuroSchedule: A Novel Effective GNN-based Scheduling Method for High-level SynthesisJun Zeng, Mingyang Kou, Hailong YaoNeurIPS 2022 · 被引用 5 次
