Efficient exploitation of instruction-level parallelism for superscalar processors by the conjugate register file scheme

Meng Chou Chang, Feipei Lai

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

This paper introduces a novel superscalar micro-architecture, called IAS-S, and its related software techniques. We treat two basic problems in superscalar machines. First, we seek a feasible hardware platform which allows the compiler to perform more aggressive instruction scheduling. Second, we develop a good way of communication between the instruction scheduler and register allocator to avoid inadequate register allocation resulting in poor instruction schedules. For the first part. IAS-S employs the Conjugate Register File (CRF) scheme to support multilevel instruction boosting so that a greater amount of instruction-level parallelism in a program can be identified at compile time. For the second part, the instruction scheduling in the IAS-S compiler consists of two passes, prepass and postpass, and a scheduling-conflict graph is built for the register allocator during the prepass scheduling. In this manner, the register allocator can take the potential benefit for later postpass instruction scheduling into account and thus prevent inadequate register allocation.

Original languageEnglish
Pages (from-to)278-293
Number of pages16
JournalIEEE Transactions on Computers
Volume45
Issue number3
DOIs
Publication statusPublished - 1996 Dec 1

Fingerprint

Instruction Scheduling
Superscalar Processor
Instruction Level Parallelism
Exploitation
Register Allocation
Superscalar
Scheduling
Compiler
Boosting
Scheduler
Schedule
Hardware
Software
Graph in graph theory
Communication

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computational Theory and Mathematics

Cite this

@article{e26e3cad604d490489e28a1fff58b52f,
title = "Efficient exploitation of instruction-level parallelism for superscalar processors by the conjugate register file scheme",
abstract = "This paper introduces a novel superscalar micro-architecture, called IAS-S, and its related software techniques. We treat two basic problems in superscalar machines. First, we seek a feasible hardware platform which allows the compiler to perform more aggressive instruction scheduling. Second, we develop a good way of communication between the instruction scheduler and register allocator to avoid inadequate register allocation resulting in poor instruction schedules. For the first part. IAS-S employs the Conjugate Register File (CRF) scheme to support multilevel instruction boosting so that a greater amount of instruction-level parallelism in a program can be identified at compile time. For the second part, the instruction scheduling in the IAS-S compiler consists of two passes, prepass and postpass, and a scheduling-conflict graph is built for the register allocator during the prepass scheduling. In this manner, the register allocator can take the potential benefit for later postpass instruction scheduling into account and thus prevent inadequate register allocation.",
author = "Chang, {Meng Chou} and Feipei Lai",
year = "1996",
month = "12",
day = "1",
doi = "10.1109/12.485567",
language = "English",
volume = "45",
pages = "278--293",
journal = "IEEE Transactions on Computers",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "3",

}

Efficient exploitation of instruction-level parallelism for superscalar processors by the conjugate register file scheme. / Chang, Meng Chou; Lai, Feipei.

In: IEEE Transactions on Computers, Vol. 45, No. 3, 01.12.1996, p. 278-293.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Efficient exploitation of instruction-level parallelism for superscalar processors by the conjugate register file scheme

AU - Chang, Meng Chou

AU - Lai, Feipei

PY - 1996/12/1

Y1 - 1996/12/1

N2 - This paper introduces a novel superscalar micro-architecture, called IAS-S, and its related software techniques. We treat two basic problems in superscalar machines. First, we seek a feasible hardware platform which allows the compiler to perform more aggressive instruction scheduling. Second, we develop a good way of communication between the instruction scheduler and register allocator to avoid inadequate register allocation resulting in poor instruction schedules. For the first part. IAS-S employs the Conjugate Register File (CRF) scheme to support multilevel instruction boosting so that a greater amount of instruction-level parallelism in a program can be identified at compile time. For the second part, the instruction scheduling in the IAS-S compiler consists of two passes, prepass and postpass, and a scheduling-conflict graph is built for the register allocator during the prepass scheduling. In this manner, the register allocator can take the potential benefit for later postpass instruction scheduling into account and thus prevent inadequate register allocation.

AB - This paper introduces a novel superscalar micro-architecture, called IAS-S, and its related software techniques. We treat two basic problems in superscalar machines. First, we seek a feasible hardware platform which allows the compiler to perform more aggressive instruction scheduling. Second, we develop a good way of communication between the instruction scheduler and register allocator to avoid inadequate register allocation resulting in poor instruction schedules. For the first part. IAS-S employs the Conjugate Register File (CRF) scheme to support multilevel instruction boosting so that a greater amount of instruction-level parallelism in a program can be identified at compile time. For the second part, the instruction scheduling in the IAS-S compiler consists of two passes, prepass and postpass, and a scheduling-conflict graph is built for the register allocator during the prepass scheduling. In this manner, the register allocator can take the potential benefit for later postpass instruction scheduling into account and thus prevent inadequate register allocation.

UR - http://www.scopus.com/inward/record.url?scp=0001247141&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0001247141&partnerID=8YFLogxK

U2 - 10.1109/12.485567

DO - 10.1109/12.485567

M3 - Article

AN - SCOPUS:0001247141

VL - 45

SP - 278

EP - 293

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

SN - 0018-9340

IS - 3

ER -