研究生: |
徐叡霆 Hsu, Rei Ting |
---|---|
論文名稱: |
使用高階模擬和合成實現三維圖形處理多核心加速器之實驗平台 An Experimental Study with ESL and HLS of Hardware Accelerators for 3D Graphics Rasterization on a Many-Core System |
指導教授: |
劉靖家
Liou, Jing-Jia |
口試委員: |
金仲達
King, Chung-Ta 黃稚存 Huang, Chih-Tsun |
學位類別: |
碩士 Master |
系所名稱: |
電機資訊學院 - 電機工程學系 Department of Electrical Engineering |
論文出版年: | 2017 |
畢業學年度: | 105 |
語文別: | 英文 |
論文頁數: | 57 |
中文關鍵詞: | 加速器 、多核心平台 、時間精準度較正 |
外文關鍵詞: | accelerator, ESL/RTL Many-core platform, Timing Annotation |
相關次數: | 點閱:3 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
電子系統層級語言平台的設立的目的是為協助當寄存器傳送語言層級平台遇到的瓶頸,如具有複雜的設計的輸入輸出,過長時間的驗證流程,尤其是遭遇到大的多核心平台寄存器傳送語言平台的模擬時間長度讓人無法忍受。然而電子系統層級語言在模擬的時間精準度會限制提升軟體的抽象層級,這篇論文呈現了是有辦法產生一個具有寄存器傳送語言層級的時間準確度的電子系統層級語言多核心模擬平台。
首先我們利用了運行在電子系統層級語言多核心平台上具有平行處理的三維繪圖兼具有光柵化軟體,因電子系統層級語言平台是依賴指令數為時間準度基準的,因此我們透過觀察這隻軟體的指令數分佈,把部分的具有高指令數的軟體透過高階模擬和合成的流程加以實作出RTL層級的硬體加速器,再利用寄存器傳送語言多核心平台驗證高階模擬和合成產生的硬體,我們嘗試把電子系統層級語言平台上所有的元件得到正確的時間延遲(可以在軟體模擬中和寄存器傳送語言層級壓縮到5%以內的誤差)。
最後,我們終於能跑透過電子系統層級語言多核心平台跑出高速的模擬,且此模擬平台兼具RTL層級的模擬時間精準度,透過電子系統層級語言相較寄存器傳送語言所具有的高度彈性,我們能利用這個ESL平台進行數據量更大且更複雜的晶片網路(10X10的網孔結構的晶片網路且具有硬體加速器的模擬)來呈現不同的硬體組態在電子系統層級語言平台上。
ESL platforms have been built to overcome the design challenges posed by RTL
including complex design entries and long verification process, especially for
large many-core architectures. However, accuracy of ESL simulation may limit
the application of raising to a higher abstraction level. In this thesis, we
demonstrate that it's possible to create a fast and accurate ESL platform.
First, we start from a parallel software for 3D graphics rasterization on a
44-mesh architecture (with an RTL definition). By profiling the software on an
ESL platform based on instruction-level accuracy (hence no timing), we
partition the part with a high-instruction count and implement it with a
high-level synthesis flow. After verifying the HLS hardware on RTL, we try to
augment the ESL platform for all components with correct timings (end-to-end
software cycle errors are less than 5\% as compared with RTL). Finally, we are
able to run a fast simulation (up to 10x10 mesh with hardware accelerators) to
show the overall performance for different configuration of hardware on the ESL
platform.
author = {David Yeh and Li-Shiuan Peh and S.Borkar}, title = {Thousand Core Chips - A Technology Perspective}, journal = { IEEE Design and Test of Computers}, isbn={0740-7475}, pages = {272-278}, month = {May}, year = {2007}
author = {Jyu-Yuan Lai and Chih-Tsun Huang and Ting-Shuo Hsu and Jing-Jia Liou and Tung-Hua Yeh and Liang-Chia Cheng and Juin-Ming Lu},
title = {Methodology of exploring ESL/RTL many-core platforms for developing embedded parallel application},
journal = {System-on-Chip Conference (SOCC), 2014 27th IEEE International}, month = {Sept}, year = {2014}
author = {Zhe-Mao Hsu and Jen-Chieh Yeh and I-Yao Chuang},
title = {An accurate system architecture refinement methodology with mixed abstraction-level virtual platform}, journal = { Design, Automation and Test in Europe Conference and Exhibition (DATE), 2010}, isbn = {978-3-9810801-6-2}, month = {March}, year = {2010}
author = {Mingsong Chen and P.Mishra}, title = {Assertion-Based Functional Consistency Checking between TLM and RTL Models}, journal = { VLSI Design and 2013 12th International Conference on Embedded Systems (VLSID)}, isbn = {978-1-4673-4639-9}, month = {Jan},
year = {2013}
author = {Yangjie Cao and Baodong Wuand Yongcai Tao and Lei Shi},
title = {Performance analysis of current parallel programming models for many-core systems},
journal = {Computer Science and Education (ICCSE), 2013 8th}, pages = {132-135}, isbn = {978-1-4673-4463-0},
month = {April}, year = {2013}
author = {R. R. Lee and Y. Lo}, title = {Load Balancing Graphics Rendering Process on a Many-Core Architecture},
journel = {International Research Conference on Information Technology and Computer Sciences (IRCITCS 2013)}, month = {Sept}, year = {2013}
author = {D.Lampret and C.-M. Chen and M. Mlinar and J.Rydberg and M. Ziv-Av and C. Ziomkowski and G. McGary and B. Gardner and R. Mathur and M.Bolado},
title = {OpenRISC 1000 Architecture Manual},
month = {May}, year = {2006}, ee = {opencores.org/or1k/MainPage}
author = {W.S Chen}, title = {Design of Non-Blocking Communication Engine for NoC-based Platform}, address = {Electrical Engineering Department}, school= {National Tsing-Hua University}, year= {2014}
author = {Hsiao-Wei Chien and Jyun-Yuan Lai and Chao-Chieh Wu and Chih-Tsun Huang}, title = {Design of a Scalable Many-Core Processor for Embedded Applications}, journel = {Design Automation Conference (ASP-DAC), 2015 20th Asia and South Pacific}, month = {March}, year = {2015}
author = {D.Lampret and C.-M. Chen and M. Mlinar and J.Rydberg and M. Ziv-Av and C. Ziomkowski and G. McGary and B. Gardner and R. Mathur and M.Bolado}, title = {OpenRISC 1000 Architecture Manual}, month = {May},
year = {2006}, ee = {opencores.org/or1k/MainPage}
title={IEEE standard SystemC Language Reference Manual}, author={Open SystemC Initiative},year = {2003}, month = March,
title={Vivado Design Suit User Guide UG902},
ee ={http://www.xilinx.com/support/documentation/sw_manuals/xilinx2015_4/ug902-vivado-high-level-synthesis.pdf}, year = {2015},
title={WISHBONE System-on-Chip (SoC) Interconnection Architecture for Portable IP Cores},
ee = {https://opencores.org/cdn/downloads/wbspec_b3.pdf},
month = {Sept}, year = {2002}
title = {AXI Reference Guide UG761},
ee = {http://www.xilinx.com/support/documentation/ip_documentation/axi_ref_guide/latest/ug761_axi_reference_guide.pdf},
year = {2012}, ver = {2012.4}
title={{NoC Solution 1.16 NoCcompiler User's Guide}},
author={Arteris S.A.},
year = {2009},
month = Feb,