Seokhyun Lee
Abstract
Centralized controllers commonly used in high-level synthesis often require long wires and cause high load capacitance, and that is why critical paths typically occur on paths from controllers to data registers instead of paths from data registers to data registers. However, conventional high-level synthesis has focused on delays within a datapath, making it difficult to solve the timing closure problem during physical synthesis. This article presents hardware architecture with a distributed controller, which makes the timing closure problem much easier. A novel critical-path-aware high-level synthesis flow is also presented for synthesizing such hardware through datapath partitioning, register binding, and controller optimization. We explore the design space related to the number of partitions, which is an important design parameter for target architecture. According to our experiments, the proposed approach reduces the critical path delay excluding FUs by 29.3% and that including FUs by 10.0%, with 2.2% area overhead on average compared to centralized controller architecture.
- S.-Y. Abe, M. Yanagisawa, and N. Togawa. 2012. An energy-efficient high-level synthesis algorithm for huddle-based distributed-register architecture. In Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS). 576--579.Google Scholar
- A. Avakian and I. Ouaiss. 2005. Optimizing register binding in FPGAs using simulated annealing. In Proceedings of the international Conference on Reconfigurable Computing and FPGAs (ReConFig). 8--16. Google ScholarDigital Library
- M. Celik, L. Pileggi, and A. Odabasioglu. 2002. IC Interconnect Analysis. Kluwer Academic Publishers, 25-38.Google Scholar
- J. Cong, Y. Fan, and J. Xu. 2009. Simultaneous resource binding and interconnection optimization based on a distributed register-file microarchitecture. ACM Trans. Des. Automat. Electron. Syst. 14, 3. Google ScholarDigital Library
- J. Cong, Y. Fan, G. Han, X. Yang, and Z. Zhang. 2003. Architectural synthesis integrated with global placement for multi-cycle communication. In Proceedings of the International Conference on Computer Aided Design (ICCAD). 536--543. Google ScholarDigital Library
- P. Coussy and A. Morawiec. 2008. High-Level Synthesis: From Algorithm to Digital Circuit. Springer Publishing Company, Inc. Google ScholarDigital Library
- A. E. Dunlop and B. W. Kernighan. 1985. A procedure for placement of standard-cell VLSI circuits. IEEE Trans. Comput. Aid. Des. Integr. Circuits Syst. 4, 1, 92--98. Google ScholarDigital Library
- C. Fiduccia and R. Mattheyses. 1982. A linear time heuristic for improving network partitions. In Proceedings of the Design Automation Conference (DAC). 175--181. Google ScholarDigital Library
- M. R. Garey and D. S. Johnson. 1979. Computers and Intractability: A Guide to the Theory of NP-Completeness. W. H. Freeman. Google ScholarDigital Library
- GAUT. 2006. GAUT -- High-level synthesis tool. http://hls-labsticc.univ-ubs.fr.Google Scholar
- Z. Gu, J. Wang, R. P. Dick, and H. Zhou. 2007. Unified incremental physical-level and high-level synthesis. IEEE Trans. Comput. Aid. Des. Integr. Circuits Syst. 26, 9, 1576--1588. Google ScholarDigital Library
- C. Huang, S. Ravi, A. Raghunathan, and N. K. Jha. 2005. Generation of distributed logic-memory architectures through high-level synthesis. IEEE Trans. Comput. Aid. Des. Integr. Circuits Syst. 24, 11, 1694--1711. Google ScholarDigital Library
- S. C. Y. Huang and W. H. Wolf. 1994. Performance-driven synthesis in controller-datapath systems. IEEE Trans. Very Large Scale Integr. (VLSI) Syst. 2, 1, 68--80. Google ScholarDigital Library
- J. Jeon, Y. Ahn, and K. Choi. 2001. CDFG toolkit user's guide. Tech. rep. SNU-EE-TR-2002-8. Department of Electrical Engineering, SeoulNational University. http://dal.snu.ac.kr/index.php/Software/CDFG.Google Scholar
- D. Kim, J. Jung, S. Lee, J. Jeon, and K. Choi. 2001. Behavior-to-placed RTL synthesis with performance-driven placement. In Proceedings of the International Conference on Computer Aided Design (ICCAD). 320--325. Google ScholarDigital Library
- T. Kim and X. Liu. 2010. A global interconnect reduction technique during high level synthesis. In Proceedings of the Asia South Pacific Design Automation Conference (ASPDAC). 695--700. Google ScholarDigital Library
- P. Kollig and B. M. Al-Hashimi. 1997. Simultaneous scheduling, allocation and binding in high level synthesis. Electron. Lett. 33, 18, 1516--1518.Google ScholarCross Ref
- V. Krishnan and S. Katkoori. 2008. Clock period minimization with iterative binding based on stochastic wirelength estimation during high-level synthesis. In Proceedings of VLSI Design. 641--646. Google ScholarDigital Library
- S. Lee and K. Choi. 2011. High Level synthesis with distributed controller for fast timing closure. In Proceedings of the International Conference on Computer Aided Design (ICCAD). 193--199. Google ScholarDigital Library
- J. Li, M. Chen, J. Li, and W. Li. 2006. Minimum clique partition problem with constrained weight for interval graphs. In Proceedings of the 12th Annual International Conference on Computing and Combinatorics (COCOON). 459--468. Google ScholarDigital Library
- S. Mitra, L. J. Avra, and E. J. McCluskey. 1999. An output encoding problem and a solution technique. IEEE Trans. Comput. Aid. Des. Integr. Circuits Syst. 18, 6, 761--768. Google ScholarDigital Library
- A. Ohchi, N. Togawa, M. Yanagisawa, and T. Ohtsuki. 2010. Performance-driven high-level synthesis with floorplan for GDR architectures and its evaluation. In Proceedings of the IEEE International Symposium on Circuits and Systems (ISCAS). 921--924.Google Scholar
- C. Papachristou and Y. Alzazefi. 1999. A method of distributed controller design for RTL circuits. In Proceedings of the Design, Automation, and Test in Europe (DATE). 774--775. Google ScholarDigital Library
- S. Park, K. Kim, H. Chang, J. Jeon, and K. Choi. 1999. Backward-annotation of post-layout delay information into high-level synthesis process for performance optimization. In Proceedings of the 6th International Conference on VLSI and CAD (ICVC). 25--28.Google Scholar
- A. Seawright and W. Meyer. 1998. Partitioning and optimizing controllers synthesized from hierarchical high-level descriptions. In Proceedings of the Design Automation Conference (DAC). 770--775. Google ScholarDigital Library
- SUIF. 1994. The SUIF 1.x Compiler System. http://suif.stanford.edu/suif/suif1/index.html.Google Scholar
- Synopsys. 2010a. Synopsys Design Compiler. http://www.synopsys.com.Google Scholar
- Synopsys. 2010b. Synopsys IC Compiler. http://www.synopsys.com.Google Scholar
- TSMC. 2009. TSMC Standard Cell Libraries. http://www.synopsys.com/dw/tsmc.php.Google Scholar
- G. K Wallace. 1992. The JPEG still picture compression standard. IEEE Trans. Consum. Electron. 38, 1, 18--34. Google ScholarDigital Library
- V. Zivojnovic, J. Martinez, C. Schlger, and H. Meyr. 1994. DSPstone: A DSP-oriented benchmarking methodology. In Proceedings of the International Conference on Signal Processing Applications and Technology (ICSPAT). 715--720.Google Scholar
Index Terms
- Critical-path-aware high-level synthesis with distributed controller for fast timing closure
Recommendations
High-level synthesis with distributed controller for fast timing closure
ICCAD '11: Proceedings of the International Conference on Computer-Aided DesignCentralized controllers commonly used in high-level synthesis often cause long wires and high load capacitance and that is why critical paths typically occur on paths from controllers to data registers. However, conventional high level synthesis has ...
Layout-driven RTL binding techniques for high-level synthesis
ISSS '96: Proceedings of the 9th international symposium on System synthesisThe importance of effective and efficient accounting of layout effects is well-established in high-level synthesis (HLS), since it allows more realistic exploration of the design space and the generation of solutions with predictable metrics. This ...
High-level synthesis with distributed controller for fast timing closure
ICCAD '11: Proceedings of the 2011 IEEE/ACM International Conference on Computer-Aided DesignCentralized controllers commonly used in high-level synthesis often cause long wires and high load capacitance and that is why critical paths typically occur on paths from controllers to data registers. However, conventional high level synthesis has ...
Comments