RapidStream IR: Infrastructure for FPGA High-Level Physical Synthesis

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Lau, Jason, Xiao, Yuanlong, Xie, Yutong, Chi, Yuze, Song, Linghao, Xiang, Shaojie, Lo, Michael, Zhang, Zhiru, Cong, Jason, Guo, Licheng
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912075141349376
author Lau, Jason
Xiao, Yuanlong
Xie, Yutong
Chi, Yuze
Song, Linghao
Xiang, Shaojie
Lo, Michael
Zhang, Zhiru
Cong, Jason
Guo, Licheng
author_facet Lau, Jason
Xiao, Yuanlong
Xie, Yutong
Chi, Yuze
Song, Linghao
Xiang, Shaojie
Lo, Michael
Zhang, Zhiru
Cong, Jason
Guo, Licheng
contents The increasing complexity of large-scale FPGA accelerators poses significant challenges in achieving high performance while maintaining design productivity. High-level synthesis (HLS) has been adopted as a solution, but the mismatch between the high-level description and the physical layout often leads to suboptimal operating frequency. Although existing proposals for high-level physical synthesis, which use coarse-grained design partitioning, floorplanning, and pipelining to improve frequency, have gained traction, they lack a framework enabling (1) pipelining of real-world designs at arbitrary hierarchical levels, (2) integration of HLS blocks, vendor IPs, and handcrafted RTL designs, (3) portability to emerging new target FPGA devices, and (4) extensibility for the easy implementation of new design optimization tools. We present RapidStream IR, a practical high-level physical synthesis (HLPS) infrastructure for representing the composition of complex FPGA designs and exploring physical optimizations. Our approach introduces a flexible intermediate representation (IR) that captures interconnection protocols at arbitrary hierarchical levels, coarse-grained pipelining, and spatial information, enabling the creation of reusable passes for design frequency optimizations. RapidStream IR improves the frequency of a broad set of mixed-source designs by 7% to 62%, including large language models and genomics accelerators, and is portable to user-customizable new FPGA platforms. We further demonstrate its extensibility through case studies, showcasing the ability to facilitate future research.
format Preprint
id arxiv_https___arxiv_org_abs_2410_13079
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle RapidStream IR: Infrastructure for FPGA High-Level Physical Synthesis
Lau, Jason
Xiao, Yuanlong
Xie, Yutong
Chi, Yuze
Song, Linghao
Xiang, Shaojie
Lo, Michael
Zhang, Zhiru
Cong, Jason
Guo, Licheng
Hardware Architecture
Distributed, Parallel, and Cluster Computing
68M99
The increasing complexity of large-scale FPGA accelerators poses significant challenges in achieving high performance while maintaining design productivity. High-level synthesis (HLS) has been adopted as a solution, but the mismatch between the high-level description and the physical layout often leads to suboptimal operating frequency. Although existing proposals for high-level physical synthesis, which use coarse-grained design partitioning, floorplanning, and pipelining to improve frequency, have gained traction, they lack a framework enabling (1) pipelining of real-world designs at arbitrary hierarchical levels, (2) integration of HLS blocks, vendor IPs, and handcrafted RTL designs, (3) portability to emerging new target FPGA devices, and (4) extensibility for the easy implementation of new design optimization tools. We present RapidStream IR, a practical high-level physical synthesis (HLPS) infrastructure for representing the composition of complex FPGA designs and exploring physical optimizations. Our approach introduces a flexible intermediate representation (IR) that captures interconnection protocols at arbitrary hierarchical levels, coarse-grained pipelining, and spatial information, enabling the creation of reusable passes for design frequency optimizations. RapidStream IR improves the frequency of a broad set of mixed-source designs by 7% to 62%, including large language models and genomics accelerators, and is portable to user-customizable new FPGA platforms. We further demonstrate its extensibility through case studies, showcasing the ability to facilitate future research.
title RapidStream IR: Infrastructure for FPGA High-Level Physical Synthesis
topic Hardware Architecture
Distributed, Parallel, and Cluster Computing
68M99
url https://arxiv.org/abs/2410.13079