Speculative Register Promotion Using Advanced Load Address Table (ALAT)

Loading...
Thumbnail Image

View/Download File

Persistent link to this item

Statistics
View Statistics

Journal Title

Journal ISSN

Volume Title

Title

Speculative Register Promotion Using Advanced Load Address Table (ALAT)

Published Date

2002-09-19

Publisher

Type

Report

Abstract

Register promotion is intended to allocate more references into registers when aliases do not appear in a region of code. When possible aliases are assumed by the compilers, speculative register promotion with hardware support is able to further improve register promotion. This paper addresses how to use the advanced loads address table (ALAT) in speculative register promotion. A compiler algorithm for speculative data flow analysis is proposed. Then the traditional partial redundancy elimination can be applied. The transformations based on the ALAT arediscussed. Experiments with SPEC CPU2000 are conducted to reveal the potential of speculative register promotion, the precision of our alias speculation, and the efficiency of ALAT. The results show that our algorithm is able to speculatively locate the candidates with high precision. The execution time on Itanium machines of one benchmark, equake, can be reduced 17% by the speculation register promotion. The tradeoffs in the hardware support for speculative register promotion are analyzed.

Keywords

Description

Related to

Replaces

License

Series/Report Number

Funding information

Isbn identifier

Doi identifier

Previously Published Citation

Suggested citation

Chen, Tong; Lin, Jin; Hsu, Wei-Chung; Yew, Pen-Chung. (2002). Speculative Register Promotion Using Advanced Load Address Table (ALAT). Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/215535.

Content distributed via the University Digital Conservancy may be subject to additional license and use restrictions applied by the depositor. By using these files, users agree to the Terms of Use. Materials in the UDC may contain content that is disturbing and/or harmful. For more information, please see our statement on harmful content in digital repositories.