LLVM Instruction Latency Measurement for Software-Hardware Interface for Multi-many-core
DOI:
https://doi.org/10.24297/ijct.v22i.9231Keywords:
estimation, multi-many-core, embedded system, SHIMAbstract
The increasing scale and complexity of embedded systems and the use of multi-many-core processors have resulted in a corresponding increase in the demand for software development with a high degree of parallelism. The degree of parallelism in software and the accuracy of performance estimation in the early design stages of model-based development can be improved by estimating performance of blocks in models and utilizing the estimate for parallelization. Research is therefore being performed on a software performance estimation technique that uses the IEEE2804-2019 hardware feature description called software-hardware interface for multi-many-core (SHIM). In SHIM, each LLVM-IR instruction is associated with the execution cycle of the target processor. Because several types of assembly instruction sequences for the target processor are generated from a given LLVM-IR instruction, it is not easy to estimate the number of execution cycles. In this study, we propose a regression analysis method to estimate the execution cycles for each LLVM-IR instruction. It is observed that our method estimated the execution cycles within the target error of ±20% in experiments using a Raspberry Pi3 Model B+.
Downloads
References
EEMBC, "CoreMark", (Accessed 2022-05-01). https://www.eembc.org/coremark/.
EMC, "The Multicore Association Specifications", (Accessed 2022-05-01). https://www.embeddedmulticore.org/themulticore- association-specifications/.
eSOL Co. Ltd., "eMBP (Model Based Parallelizer)", (Accessed 2022-05-01).
https://www.esol.co.jp/embedded/mbp.html.
Gondo, M., Arakawa, F., & Edahiro, M. (2014, April). Establishing a standard interface between multi-manycore
and software tools-SHIM. In 2014 IEEE COOL Chips XVII (pp. 1-3). IEEE.
Hwang, Y., Abdi, S., & Gajski, D. (2008, March). Cycle-approximate retargetable performance estimation at the
transaction level. In Proceedings of the conference on Design, automation and test in Europe (pp. 3-8).
IEEE, "IEEE Standard for Software-Hardware Interface for Multi-Many-Core", IEEE 2804-2019, (Accessed
-05-01). https://standards.ieee.org/ieee/2804/7477/.
LLVM project, "The LLVM Compiler Infrastructure",(Accessed 2022-05-01). https://llvm.org/.
LLVM project, "LLVM Language Reference Manual", (Accessed 2022-05-01). https://llvm.org/docs/LangRef.html.
LLVM project, "llvm-cov", (Accessed 2022-05-01). https://llvm.org/docs/CommandGuide/llvm-cov.html.
Patel, R., & Rajawat, A. (2013). Recent trends in embedded system software performance estimation. Design
Automation for Embedded Systems, 17(1), 193-213.
Ray, A., Srikanthan, T., & Wu, J. (2010). Rapid techniques for performance estimation of processors. Journal of
Research and Practice in Information Technology, 42(2), 147-165.
Renesas electronics, "RH850/E1M-S2", (Accessed 2022-05-01). https://www.renesas.com/jp/en/products/microcontrollersmicroprocessors/ rh850-automotive-mcus.
SHIM Working Group, "SHIM Latency Measurement and Insertion", (Accessed 2022-05-01).
https://github.com/openshim/shim2/tree/master/shim-measure.
Wang, S., Zhong, G., & Mitra, T. (2017). CGPredict: Embedded GPU performance estimation from single-threaded
applications. ACM Transactions on Embedded Computing Systems (TECS), 16(5s), 1-22.
Wijesundera, D., Prakash, A., Srikanthan, T., & Ihalage, A. (2018). Framework for rapid performance estimation
of embedded soft core processors. ACM Transactions on Reconfigurable Technology and Systems (TRETS), 11(2), 1-21.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2022 Hiro Mikami, Kei Torigoe, Makoto Inokawa, Masato Edahiro
This work is licensed under a Creative Commons Attribution 4.0 International License.
