by Bill Pottenger, Rudolf Eigenmann
In ACM Int. Conf. on Supercomputing (ICS’95
http://polaris.cs.uiuc.edu/reports/1396.ps.gz
Add To MetaCart
Abstract:
The elimination of induction variables and the parallelization of reductions in FORTRAN programs has been shown to be integral to performance improvement on parallel computers [9, 10]. As part of the Polaris project, compiler passes that recognize these idioms have been implemented and evaluated. Developing these techniques to the point necessary for achieving significant speedups on real applications has prompted solutions to problems that have not been addressed in previous reports. These include analysis capabilities to disprove zero-trip loops, symbolic handling facilities to compute closed forms of recurrences, and interfaces to other compilation passes, such as the datadependence test. In comparison, the analysis phase of induction variables, which has received most attention so far, has turned out to be relatively straightforward. 1
Citations
|
352
|
The omega test: a fast and practical integer programming algorithm for dependence analysis
– Pugh
- 1991
|
|
344
|
Dependence Analysis for Supercomputing
– Banerjee
- 1988
|
|
296
|
Advanced compiler optimizations for supercomputers
– Padua, Wolfe
- 1986
|
|
213
|
The Perfect Club Benchmarks: Effective Performance Evaluation of Supercomputers
– Berry, Chen, et al.
- 1989
|
|
123
|
Automatic Array Privatization
– Tu, Padua
- 1993
|
|
102
|
Experience in the automatic parallelization of four perfect benchmark programs
– Hoeflinger, Li, et al.
- 1992
|
|
91
|
Beyond induction variables: Detecting and classifying sequences using a demand-driven SSA form
– Gerlek, Stoltz, et al.
- 1995
|
|
83
|
Beyond induction variables
– Wolfe
- 1992
|
|
77
|
The Range Test: A Dependence Test for Symbolic, Non-linear Expressions
– Blume, Eigenmann
- 1994
|
|
59
|
Symbolic range propagation
– Blume, Eigenmann
- 1995
|
|
37
|
Automatic Detection of Parallelism: A Grand Challenge for High-Performance Computing
– Blume, Eigenmann, et al.
- 1994
|
|
35
|
On the automatic parallelization of the perfect benchmarks
– Hoeflinger, Padua
- 1998
|
|
35
|
Symbolic program analysis and optimization for parallelizing compilers
– Haghighat, Polychronopoulos
- 1992
|
|
32
|
Polaris: The Next Generation in Parallelizing Compilers
– Blume, Eigenmann, et al.
- 1994
|
|
24
|
The PERFECT club benchmarks: E ective performance evaluation of supercomputers
– Sameh, Clementi, et al.
- 1989
|
|
12
|
Automatic Recognition of Induction & Recurrence Relations by Abstract Interpretation
– Ammarguellat, Harrison
- 1990
|
|
11
|
High-Level Pattern-Matching Extensions to C++ for Fortran Program Manipulation in Polaris
– Weatherford
- 1994
|
|
10
|
Demand-Driven Symbolic Analysis
– Tu, Padua
- 1994
|
|
5
|
Fortran 77 Programmer's Guide and associated man pages. Silicon Graphics
– Documentation
- 1994
|
|
2
|
Personal communication with author
– Harrison
- 1994
|
|
2
|
Automatic Parallelization and Manual Improvements of the Perfect Club Program OCEAN for Cedar
– Hoeflinger
- 1992
|
|
2
|
Personal communication with authors
– Wolfe, Gerlek
- 1994
|
|
2
|
Automatic Parallelization and Manual Improvements of the Perfect Club Program OCEAN for Cedar
– inger
- 1992
|
|
2
|
andDavid Padua. Demand-Driven Symbolic Analysis
– Tu
- 1994
|
