DSP Code Optimization Methods Utilizing Addressing Operations at the Codes without Memory Accesses
スポンサーリンク
概要
- 論文の詳細を見る
Many digital signal processors (DSPs) employ indirect addressing using address registers (ARs) to indicate their memory addresses, which often leads to overhead. This paper presents methods to efficiently allocate addresses for variables in a given program so that overhead in AR update operations is reduced. Memory addressing model is generalized in such a way that AR can be updated at the codes without memory accesses. An efficient memory address allocation is obtained by a method based on the graph linearization algorithm, which takes account of the number of possible AR update operations for every memory access. In order to utilize multiple ARs, methods to assign variables into ARs are also investigated. The proposed methods are applied to the compiler for _μPD77230 (NEC) and generated codes for several examples prove effectiveness of these methods.
- 社団法人電子情報通信学会の論文
- 1997-12-25
著者
-
Sugino Nobuhiko
Department Of Information Processing Interdisciplinary Graduate School Of Science And Engineering To
-
Nishihara Akinori
Center For Research And Development Of Educational Technology Tokyo Institute Of Technology
-
MIYAZAKI Hironobu
Department of Physical Electronics, Faculty of Engineering, Tokyo Institute of Technology
-
Miyazaki Hironobu
Sheet & Coil Div. Yawata Works Nippon Steel
関連論文
- Design Method for 2-Channel Signal Word Decomposed Filters with Minimum Output Error and Their Effective VLSI Implementation(Digital Signal Processing, Papers Selected from the 19th Symposium on Signal Processing)
- DSP Code Optimization Utilizing Memory Addressing Operation (Special Section on Digital Signal Processing)
- LMS-Based Algorithms with Multi-Band Decomposition of the Estimation Error Applied to System Identification (Special Section on Digital Signal Processing)
- Modular Array Structures for Design and Multiplierless Realization of Two-Dimensional Linear Phase FIR Digital Filters
- Design of Checkerboard-Distortion-Free Multidimensional Multirate Filters(Special Section on Digital Signal Processing)
- Eigensignals of Downsamplers in Time and Transform Domains(Digital Signal Processing)
- Discrete Wirtinger-Type Inequalities for Gauging the Power of Sinusoids Buried in Noise
- Explicit Formula for Predictive FIR Filters and Differentiators Using Hahn Orthogonal Polynomials(Papers Selected from the 21st Symposium on Signal Processing)
- Memory Allocation Method for Indirect Addressing DSPs with ±2 Update Operations
- Stable Single-Bit Noise-Shaping Quantizer Based on Σ△ Modulation and Successive Data Coding into Pre-Optimized Binary Vectors(Special Section on Digital Signal Processing)
- Multiple-Image-Based Restoration for Motion Blur with Non-uniform Point Spread Function
- Restoration of Images Degraded by Linear Motion Blurred Objects in Still Background
- Identification of Piecewise Linear Uniform Motion Blur
- A Change Detection Method for Image Sequences Based on Physical Models(Image, Papers Selected from the 19th Symposium on Signal Processing)
- DSP Code Optimization Methods Utilizing Addressing Operations at the Codes without Memory Accesses
- Performance Enhancement on Digital Signal Processors with Complex Arithmetic Capability (Special Section on VLSI for Digital Signal Processing)
- Analytic Constructions of Periodic and Non-periodic Complementary Sequences(Digital Signal Processing)
- Development of Nonchromate-Treated Tinplate Having Excellent Resistance to Yellow Staining and Smudging
- Design of High-Order Noise-Shaping FIR Filters for Overload-Free Stable Single- and Multi-Bit Data Converters(Digital Signal Processing)