EUVIMEDSkopri EUVIMED
Open access · CC BY 4.0

An LLVM-Based HW/SW Co-Design Framework for Overlapped Chunked Execution of Single-Buffer FPGA Accelerators

Gihyeon Jeon, Daejin Park

Electronics · 2026-11-11 · 15 (22) · 5219

DOI: 10.3390/electronics15225219 ↗

At a glance

Journal article in Electronics, published online on 2026-11-11. Written by Gihyeon Jeon and 1 co-authors. 24 references. The full text is freely available (open access, CC BY 4.0). Funded by: BK21 FOUR project, Basic Science Research Program, National Research Foundation of Korea.

AstrattDettalji biblijografiċiCiteRelated studies

Abstract (original language)

Edge AI accelerators frequently adopt a single-buffer structure to minimize on-chip memory, but such accelerators cannot accept the next block of input while the current block is still being processed, which serializes data transfer, computation, and result retrieval. This paper presents a hardware/software co-design framework that restores system-level overlap for chains of single-buffer accelerators connected through AXI DMA. On the hardware side, a Hardware Performance Manager (HPM) monitors the AXI4-Stream handshakes between the DMA and the accelerator and coordinates the output handshake while a new input packet is starting. On the software side, an LLVM Pass analyzes a sequential chunked offload loop with scalar-evolution (SCEV) expressions, verifies that consecutive accelerator stages form a producer–consumer chain, and replaces the loop with a call to a pipelined runtime. The Pass additionally moves whole-buffer cache maintenance in front of and behind the loop into per-chunk operations that the runtime performs while the accelerators are busy. On a Zybo Z7-20 board with a CORDIC–GEMM chain, the transformed program reduces the total execution time by 30.2–31.3% relative to sequential execution for 201 K–4.8 M input words, and all 1040 measured runs, which also include single-stage loops and a loop with a shorter final chunk, reproduce a bit-exact software reference. An exhaustive chunk-size sweep shows that the default chunk size lies within 0.44% of the best fixed chunk size for every workload size.

Funding

  • BK21 FOUR project
  • Basic Science Research Program
  • National Research Foundation of Korea
  • Ministry of Education
  • Institute of Information and communications Technology Planning and Evaluation (IITP) grant funded by the Korean government
  • IC Design Education Center (IDEC), Korea

Cite

APA

Jeon, G., & Park, D. (2026). An LLVM-Based HW/SW Co-Design Framework for Overlapped Chunked Execution of Single-Buffer FPGA Accelerators. Electronics, 15(22), 5219. https://doi.org/10.3390/electronics15225219

Vancouver

Jeon G, Park D. An LLVM-Based HW/SW Co-Design Framework for Overlapped Chunked Execution of Single-Buffer FPGA Accelerators. Electronics. 2026;15(22):5219. doi:10.3390/electronics15225219

BibTeX
@article{Jeon2026,
  author = {Jeon, Gihyeon and Park, Daejin},
  title = {An LLVM-Based HW/SW Co-Design Framework for Overlapped Chunked Execution of Single-Buffer FPGA Accelerators},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {22},
  pages = {5219},
  doi = {10.3390/electronics15225219}
}
Start for free →