Abstract
An ILP (Instruction-Level Parallelism) compiler uses aggressive optimizations to reduce a program's running time. These optimizations have been shown to be effective when profile information is available. Unfortunately, users are not always willing or able to profile their programs. A method of overcoming this issue is for an ILP compiler to statically infer the information normally obtained from profiling. This paper investigates one aspect of this inference: the static prediction of conditional-branch direction. The goals of this work are to utilize the source-level information available in a compiler when performing static branch prediction, to identify static-branch-prediction cases in which there is a high confidence that a branch will go in one direction at run time, to gain an intuitive understanding into the reasons why the static-branch-prediction heuristics are effective, and ultimately to improve the accuracy of the static branch prediction. The effectiveness of the static-branch-prediction heuristics developed in this paper is demonstrated on a set of programs from SPEC CINT92, SPEC CINT95, and the IMPACT compiler.
| Original language | English (US) |
|---|---|
| Title of host publication | Proceedings - 1998 International Conference on Parallel Architectures and Compilation Techniques, PACT 1998 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 214-221 |
| Number of pages | 8 |
| ISBN (Electronic) | 0818685913 |
| DOIs | |
| State | Published - Jan 1 1998 |
| Event | 1998 International Conference on Parallel Architectures and Compilation Techniques, PACT 1998 - Paris, France Duration: Oct 12 1998 → Oct 18 1998 |
Publication series
| Name | Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT |
|---|---|
| ISSN (Print) | 1089-795X |
Other
| Other | 1998 International Conference on Parallel Architectures and Compilation Techniques, PACT 1998 |
|---|---|
| Country | France |
| City | Paris |
| Period | 10/12/98 → 10/18/98 |
Fingerprint
ASJC Scopus subject areas
- Software
- Theoretical Computer Science
- Hardware and Architecture
Cite this
Improving static branch prediction in a compiler. / Deitrich, Brian L.; Chen, Ben Chung; Hwu, Wen-Mei W.
Proceedings - 1998 International Conference on Parallel Architectures and Compilation Techniques, PACT 1998. Institute of Electrical and Electronics Engineers Inc., 1998. p. 214-221 (Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Improving static branch prediction in a compiler
AU - Deitrich, Brian L.
AU - Chen, Ben Chung
AU - Hwu, Wen-Mei W
PY - 1998/1/1
Y1 - 1998/1/1
N2 - An ILP (Instruction-Level Parallelism) compiler uses aggressive optimizations to reduce a program's running time. These optimizations have been shown to be effective when profile information is available. Unfortunately, users are not always willing or able to profile their programs. A method of overcoming this issue is for an ILP compiler to statically infer the information normally obtained from profiling. This paper investigates one aspect of this inference: the static prediction of conditional-branch direction. The goals of this work are to utilize the source-level information available in a compiler when performing static branch prediction, to identify static-branch-prediction cases in which there is a high confidence that a branch will go in one direction at run time, to gain an intuitive understanding into the reasons why the static-branch-prediction heuristics are effective, and ultimately to improve the accuracy of the static branch prediction. The effectiveness of the static-branch-prediction heuristics developed in this paper is demonstrated on a set of programs from SPEC CINT92, SPEC CINT95, and the IMPACT compiler.
AB - An ILP (Instruction-Level Parallelism) compiler uses aggressive optimizations to reduce a program's running time. These optimizations have been shown to be effective when profile information is available. Unfortunately, users are not always willing or able to profile their programs. A method of overcoming this issue is for an ILP compiler to statically infer the information normally obtained from profiling. This paper investigates one aspect of this inference: the static prediction of conditional-branch direction. The goals of this work are to utilize the source-level information available in a compiler when performing static branch prediction, to identify static-branch-prediction cases in which there is a high confidence that a branch will go in one direction at run time, to gain an intuitive understanding into the reasons why the static-branch-prediction heuristics are effective, and ultimately to improve the accuracy of the static branch prediction. The effectiveness of the static-branch-prediction heuristics developed in this paper is demonstrated on a set of programs from SPEC CINT92, SPEC CINT95, and the IMPACT compiler.
UR - http://www.scopus.com/inward/record.url?scp=0003320262&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0003320262&partnerID=8YFLogxK
U2 - 10.1109/PACT.1998.727253
DO - 10.1109/PACT.1998.727253
M3 - Conference contribution
AN - SCOPUS:0003320262
T3 - Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT
SP - 214
EP - 221
BT - Proceedings - 1998 International Conference on Parallel Architectures and Compilation Techniques, PACT 1998
PB - Institute of Electrical and Electronics Engineers Inc.
ER -