Improving ILP with the Vectorized Computing Mechanism in VLIW DSP Architecture
碩士 === 國立中山大學 === 電機工程學系研究所 === 91 === In order to improving the performance for real-time application, current digital signal processors use VLIW architectures to increase the degree of instruction level parallelism (ILP). Two factors will limit the ILP, one is enough hardware resource for all para...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2003
|
Online Access: | http://ndltd.ncl.edu.tw/handle/58764251280206231113 |
id |
ndltd-TW-091NSYS5442039 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-091NSYS54420392016-06-22T04:20:47Z http://ndltd.ncl.edu.tw/handle/58764251280206231113 Improving ILP with the Vectorized Computing Mechanism in VLIW DSP Architecture VLIWDSP架構之增進指令並行度之向量化運算機制 Te-Shin Yang 楊得鑫 碩士 國立中山大學 電機工程學系研究所 91 In order to improving the performance for real-time application, current digital signal processors use VLIW architectures to increase the degree of instruction level parallelism (ILP). Two factors will limit the ILP, one is enough hardware resource for all parallel instructions. Another is the dependence relations between instructions. This thesis designs a VLIW architecture processing core called DVBTDSP molded by FFT algorithm and uses the software pipelining mechanism to schedule the loop to achieve the highest ILP degree when used to execute FFT butterfly operations. Furthermore, in order to provide the smooth data stream for pipeline operations, we design a mechanism to improve the modulo addressing, which will collect the discrete vectors into one continuous vector. The simulation results show that the DVBTDSP has double performance of the C6200 for the FFT processing, and has good performance for FIR, IIR and DCT algorithm computing. Jih-Ching Chiu 邱日清 2003 學位論文 ; thesis 84 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立中山大學 === 電機工程學系研究所 === 91 === In order to improving the performance for real-time application, current digital signal processors use VLIW architectures to increase the degree of instruction level parallelism (ILP). Two factors will limit the ILP, one is enough hardware resource for all parallel instructions. Another is the dependence relations between instructions. This thesis designs a VLIW architecture processing core called DVBTDSP molded by FFT algorithm and uses the software pipelining mechanism to schedule the loop to achieve the highest ILP degree when used to execute FFT butterfly operations. Furthermore, in order to provide the smooth data stream for pipeline operations, we design a mechanism to improve the modulo addressing, which will collect the discrete vectors into one continuous vector. The simulation results show that the DVBTDSP has double performance of the C6200 for the FFT processing, and has good performance for FIR, IIR and DCT algorithm computing.
|
author2 |
Jih-Ching Chiu |
author_facet |
Jih-Ching Chiu Te-Shin Yang 楊得鑫 |
author |
Te-Shin Yang 楊得鑫 |
spellingShingle |
Te-Shin Yang 楊得鑫 Improving ILP with the Vectorized Computing Mechanism in VLIW DSP Architecture |
author_sort |
Te-Shin Yang |
title |
Improving ILP with the Vectorized Computing Mechanism in VLIW DSP Architecture |
title_short |
Improving ILP with the Vectorized Computing Mechanism in VLIW DSP Architecture |
title_full |
Improving ILP with the Vectorized Computing Mechanism in VLIW DSP Architecture |
title_fullStr |
Improving ILP with the Vectorized Computing Mechanism in VLIW DSP Architecture |
title_full_unstemmed |
Improving ILP with the Vectorized Computing Mechanism in VLIW DSP Architecture |
title_sort |
improving ilp with the vectorized computing mechanism in vliw dsp architecture |
publishDate |
2003 |
url |
http://ndltd.ncl.edu.tw/handle/58764251280206231113 |
work_keys_str_mv |
AT teshinyang improvingilpwiththevectorizedcomputingmechanisminvliwdsparchitecture AT yángdéxīn improvingilpwiththevectorizedcomputingmechanisminvliwdsparchitecture AT teshinyang vliwdspjiàgòuzhīzēngjìnzhǐlìngbìngxíngdùzhīxiàngliànghuàyùnsuànjīzhì AT yángdéxīn vliwdspjiàgòuzhīzēngjìnzhǐlìngbìngxíngdùzhīxiàngliànghuàyùnsuànjīzhì |
_version_ |
1718318530921758720 |