Sub-operation Parallelism Optimization in SIMD Processor Core Synthesis

• KAWAZU Hideki

  Department of Computer Science, Waseda University

• UCHIDA Jumpei

  Department of Computer Science, Waseda University

• MIYAOKA Yuichiro

  Department of Computer Science, Waseda University

• TOGAWA Nozomu

  Department of Computer Science, Waseda University

• YANAGISAWA Masao

  Department of Computer Science, Waseda University

• OHTSUKI Tatsuo

  Department of Computer Science, Waseda University

抄録

A b-bit SIMD functional unit has n k-bit sub-functional units in itself, where b=k×n. It can execute n-parallel k-bit operations. However, all the b-bit functional units in a processor core do not necessarily execute n-parallel operations. Depending on an application program, some of them just execute n/2-parallel operations or even n/4-parallel operations. This means that we can modify a b-bit SIMD functional unit so that it has n/2 k-bit sub-functional units or n/4 k-bit sub-functional units. The number of k-bit sub-functional units in a SIMD functional unit is called sub-operation parallelism. We incorporate a sub-operation parallelism optimization algorithm into SIMD functional unit optimization. Our proposed algorithm gradually reduces sub-operation parallelism of a SIMD functional unit while the timing constraint of execution time satisfied. Thereby, we can finally find a processor core with small area under the given timing constraint. We expect that we can obtain processor core configurations of smaller area in the same timing constraint rather than a conventional system. The promising experimental results are also shown.

収録刊行物

• IEICE transactions on fundamentals of electronics, communications and computer sciences

  IEICE transactions on fundamentals of electronics, communications and computer sciences 88 (4), 876-884, 2005-04-01

  一般社団法人電子情報通信学会

キーワード

• processor synthesis
• packed SIMD type operation
• hardware
• software partitioning
• software cosynthesis
• sub-operation parallelism

詳細情報

• CRID: 1571980077395773824
• NII論文ID: 110003213389
• NII書誌ID: AA10826239
• ISSN: 09168508
• 本文言語コード: en
• データソース種別:
  • CiNii Articles