000015331 001__ 15331 000015331 005__ 20250115134147.0 000015331 020__ $$a979-8-3315-3007-5 000015331 0247_ $$2DOI$$a10.1109/FPL64840.2024.00014 000015331 037__ $$aCONFERENCE 000015331 039_9 $$a2025-01-15 13:41:47$$b0$$c2025-01-14 16:19:40$$d1001044$$c2025-01-14 12:00:34$$d0$$y2025-01-14 12:00:17$$z1000099 000015331 041__ $$aeng 000015331 245__ $$aDynaRapid :$$bfast-tracking from C to routed circuits 000015331 260__ $$aPiscataway, NJ, USA$$bInstitute of Electrical and Electronics Engineers 000015331 269__ $$a2024-09 000015331 300__ $$a9 p. 000015331 506__ $$avisible 000015331 520__ $$9eng$$aAdvancements in design automation technologies, such as high-level synthesis (HLS), have raised the input abstraction level and made the design entry process for FPGAs more friendly to software programmers. In contrast, the backend compilation process for implementing designs on FPGAs is considerably more lengthy compared to software compilation: while software code compilation may take just a few seconds, FPGA compilation times can often span from several minutes to hours due to the complexity of the underlying toolchain and ever-growing device capacities. In this paper, we present DynaRapid, a fast compilation tool that generates—in a matter of seconds—fully legal placed-and-routed designs for commercial FPGAs. Elastic circuits created by the HLS tool Dynamatic are made exclusively of a limited number of reusable components; we exploit this fact to create a library of placed and routed building blocks, and then stitch together instances of them as needed through RapidWright. Our approach accelerates the C-to-FPGA implementation process by a geomean 20× with only 10% of degradation in operating frequency compared to a conventional commercial off-the-shelf implementation flow. 000015331 540__ $$acorrect 000015331 592__ $$aHEI-VS 000015331 592__ $$bInstitut Systèmes industriels 000015331 592__ $$cIngénierie et Architecture 000015331 6531_ $$9eng$$adegradation 000015331 6531_ $$9eng$$acodes 000015331 6531_ $$9eng$$adesign automation 000015331 6531_ $$9eng$$alaw 000015331 6531_ $$9eng$$acircuits 000015331 6531_ $$9eng$$asoftware 000015331 6531_ $$9eng$$alibraries 000015331 6531_ $$9eng$$alogic 000015331 6531_ $$9eng$$afield programmable gate array 000015331 6531_ $$9eng$$astandards 000015331 655_7 $$apublished full paper 000015331 700__ $$aGuerrieri, Andrea$$uSchool of Engineering, HES-SO Valais-Wallis, HEI, HES-SO University of Applied Sciences and Arts Western Switzerland ; EPFL, Lausanne, Switzerland 000015331 700__ $$aGuha, Srijeet$$uEPFL, Lausanne, Switzerland 000015331 700__ $$aLavin, Chris$$uAMD Research and Advanced Development 000015331 700__ $$aHung, Eddie$$uAMD Research and Advanced Development 000015331 700__ $$aJosipovic, Lana$$uETH Zürich, Zürich, Switzerland 000015331 700__ $$aIenne, Paolo$$uEPFL, Lausanne, Switzerland 000015331 711__ $$a2024 34th International Conference on Field-Programmable Logic and Applications (FPL)$$cTorino, Italy$$d2024-09-02$$m2024-09-06 000015331 773__ $$tProceedings of the 2024 34th International Conference on Field-Programmable Logic and Applications (FPL), 2-6 September 2024, Torino, Italy$$j2024$$q24-32 000015331 8564_ $$uhttps://arodes.hes-so.ch/record/15331/files/Guerrieri_2024_DynaRapid_fast-tracking_C_routed_circuits.pdf$$yPublished version$$94be456e2-c2e0-446d-940c-8e8ac34a2432$$s2738977 000015331 906__ $$aNONE 000015331 909CO $$ooai:hesso.tind.io:15331$$pGLOBAL_SET 000015331 950__ $$aaucun 000015331 980__ $$aconference 000015331 981__ $$aconference