Novel VLIW Code Compaction Method for a 3D Geometry Processor
スポンサーリンク
概要
- 論文の詳細を見る
A VLIW(Very Long Instruction Word)architecture with a new code compaction method has been proposed. For a 3D-geometry processor, we consider two types of 2-issue VLIW architectures, the floating-point execution accelerating VLIW(FP-VLIW)and the data-move enhancing VLIW(MV-VLIW)architectures, as expansions of a Single-Streaming Singel Instruction, Multiple Data(SS-SIMD)architecture. To solve the code bloat problem which is common to VLIW architectures, the proposed method makes it possible to compact original codes into the VLIW codes by software tools and decompact the VLIW codes by a simple hardware decompactor composed of an instruction swap circuit on a chip. Speeds and code densities of the two VLIWs with the code compaction are compared to the SS-SIMD with the same instruction set and the same building blocks. The FP-VLIW shows the fastest speed performance in the evaluation results of the viewperf CDRS-03 benchmark programs. It is 36% faster than the SS-SIMD used as reference. The proposed compaction method keeps the 95% code density of the SS-SIMD. One test program shows that the code density of the MV-VLIW is higher than that of the SS-SIMD. This result demonstrates that the merit of compacting nops can be greater than the VLIW penalty. The FP-VLIW architecture with the code compaction achieves 1.36 times the speed performance without significant code-density deterioration.
- 社団法人電子情報通信学会の論文
- 2001-11-01
著者
-
Makino Hiroshi
Renesas Technology Corporation
-
Matsuda Y
Mitsubishi Electric Corp. Itami‐shi Jpn
-
Matsuda Yoshio
System Lsi Laboratory Mitsubishi Electric Corporation
-
SUZUKI Hiroaki
System LSI Development Center, Mitsubishi Electric Corporation
-
KAWAI Hiroyuki
System LSI Development Center, Mitsubishi Electric Corporation
-
MAKINO Hiroshi
System LSI Development Center, Mitsubishi Electric Corporation
-
Makino H
Mitsubishi Electric Corp. Itami‐shi Jpn
-
Kawai Hiroyuki
System Lsi Development Center Mitsubishi Electric Co.
-
Kawai Hiroyuki
System Lsi Development Center Mitsubishi Electric Corporation
-
Matsuda Yoshio
System Lsi Development Center Mitsubishi Electric Corporation
-
Suzuki Hiroaki
System Lsi Development Center Mitsubishi Electric Co.
-
Suzuki Hiroaki
System Lsi Development Center Mitsubishi Electric Corporation
関連論文
- Novel VLIW Code Compaction Method for a 3D Geometry Processor
- A Wide Range 1.0-3.6 V 200 Mbps, Push-Pull Output Buffer Using Parasitic Bipolar Transistors(Low-Power System LSI, IP and Related Technologies)
- Signal Integrity Design and Analysis for a 400 MHz RISC Microcontroller
- A Low Standby Current DSP Core Using Improved ABC-MT-CMOS with Charge Pump Circuit
- Physical Design Methodology for On-Chip 64-Mb DRAM MPEG-2 Encoding with a Multimedia Processor(Special Issue on High-Performance and Low-Power Microprocessors)
- 招待講演 A 65nm embedded SRAM with wafer level burn-in mode, leak-bit redundancy and E-trim fuse for known good die (集積回路)
- A Design of High-Speed 4-2 Compressor for Fast Multiplier (Special Issue on Ultra-High-Speed LSIs)
- A Floating-Point Divider Using Redundant Binary Circuits and an Asynchronous Clock Scheme
- A 286 MHz 64-b Floating Point Multiplier with Enhanced CG Operation
- A 2.6-ns 64-b Fast and Small CMOS Adder (Special Issue on Ultra-High-Speed LSIs)
- Shared Multibuffer ATM Switches with Hierarchical Queueing and Multicast Functions
- A 1.5-V 250-MHz to 3.0-V 622-MHz Operation CMOS Phase-Locked Loop with Precharge Type Phase-Frequency Detector
- A Shared Multibuffer Architecture for High-Speed ATM Switch LSIs (Special Issue on New Architecture LSIs)
- A Highly Parallel DSP Architecture for Image Recognition
- A Low-Power Microcontroller with Body-Tied SOI Technology(Low-Power System LSI, IP and Related Technologies)
- Realistic Scaling Scenario for Sub-100 nm Embedded SRAM Based on 3-Dimensional Interconnect Simulation(the IEEE International Conference on SISPAD '02)
- A Large-Scale, Flip-Flop RAM Imitating a Logic LSI for Fast Development of Process Technology
- A Line-Mode Test with Data Register for ULSI Memory Architecture (Special Issue on LSI Memories)
- A Programmable Geometry Processor with Enhanced Four-Parallel SIMD Type Processing Core for PC-Based 3D Graphics
- A 250 MHz Dual Port Cursor RAM Using Dynamic Data Alignment Architecture
- An Efficient Self-Timed Queue Architecture for ATM Switch LSIs (Special Issue on Multimedia, Analog and Processing LSIs)