An Empirical Study on the Granularity of Pointer Analysis in C Programs

Chen, Tong; Lin, Jin; Hsu, Wei-Chung; Yew, Pen-Chung

doi:10.1007/11596110_11

Cited by 9 publications

(10 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to build an effective framework for speculative analysis and optimizations, we have developed a comprehensive instrumentation-based profiling environment on ORC to generate a set of alias profile [Chen et al 2002], dependence profile [Chen et al 2004] and edge/path profile. We also provide a set of compiler heuristics to support speculative analysis.…”

Section: Alias Profiling and Compiler Heuristicsmentioning

confidence: 99%

“…Different values of n will give different levels of granularity to the named memory objects. Such granularity can affect the quality of alias profiling [Chen et al 2002]. In our experiments, we set n to 2, since it gives reasonably good results at a low cost.…”

Section: Alias Profiling and Compiler Heuristicsmentioning

confidence: 99%

“…The target set thus obtained represents the best result that a naming scheme can be expected to achieve. However, approximation is introduced into our implementation for the sake of efficiency [Chen et al 2002]. There are also other limitations in our alias profiling.…”

Section: Alias Profiling and Compiler Heuristicsmentioning

confidence: 99%

See 2 more Smart Citations

A compiler framework for speculative optimizations

Lin

Chen

Hsu

et al. 2004

ACM Trans. Archit. Code Optim.

Self Cite

View full text Add to dashboard Cite

, PEN-CHUNG YEW and ROY DZ-CHING JU, TIN-FOOK NGAI, SUN CHAN ________________________________________________________________________ Speculative execution, such as control speculation or data speculation, is an effective way to improve program performance. Using edge/path profile information or simple heuristic rules, existing compiler frameworks can adequately incorporate and exploit control speculation. However, very little has been done so far to allow existing compiler frameworks to incorporate and exploit data speculation effectively in various program transformations beyond instruction scheduling. This paper proposes a speculative SSA form to incorporate information from alias profiling and/or heuristic rules for data speculation, thus allowing existing frameworks to be extended to support both control and data speculation. Such a general framework is very useful for EPIC architectures that provide runtime checking (such as advanced load address table (ALAT)) on data speculation to guarantee the correctness of program execution. We use SSAPRE as one example to illustrate how to incorporate data speculation in partial redundancy elimination (PRE), register promotion, and strength reduction. Our extended framework allows both control and data speculation to be performed on top of SSAPRE and, thus, enables more aggressive speculative optimizations. The proposed framework has been implemented on Intel's Open Research Compiler (ORC). We present experimental data on some SPEC2000 benchmark programs to demonstrate the usefulness of this framework.

show abstract

Section: Alias Profiling and Compiler Heuristicsmentioning

confidence: 99%

Section: Alias Profiling and Compiler Heuristicsmentioning

confidence: 99%

See 1 more Smart Citation

A compiler framework for speculative optimizations

Lin

Chen

Hsu

et al. 2004

ACM Trans. Archit. Code Optim.

Self Cite

View full text Add to dashboard Cite

show abstract

“…If the benefit of aggressive optimizations does not outweigh the cost of long compile times, users will avoid them. Making matters worse, analysis precision drastically affects optimization quality [3,8,28,29,36], and precise analyses tend to be more expensive than their less-precise counterparts [7,15] and scale poorly [16,26]. Despite the performance potential of state-ofthe-art transformations, common compilers optimize small, intraprocedural scopes, instead favoring short compilation times.…”

Section: Introductionmentioning

confidence: 99%

“…Compiler authors should not sacrifice analysis precision for cost since imprecision limits optimization [3,8,28,29,36]. Instead, they should use the most precise analyses and reduce compilation time by exploiting the reduced information of the DAGSCC.…”

Section: Introductionmentioning

confidence: 99%

Fast condensation of the program dependence graph

Johnson

Zaks

et al. 2013

SIGPLAN Not.

View full text Add to dashboard Cite

Aggressive compiler optimizations are formulated around the Program Dependence Graph (PDG). Many techniques, including loop fission and parallelization are concerned primarily with dependence cycles in the PDG. The Directed Acyclic Graph of Strongly Connected Components (DAGSCC) represents these cycles directly. The naïve method to construct the DAGSCC first computes the full PDG. This approach limits adoption of aggressive optimizations because the number of analysis queries grows quadratically with program size, making DAGSCC construction expensive. Consequently, compilers optimize small scopes with weaker but faster analyses.We observe that many PDG edges do not affect the DAGSCC and that ignoring them cannot affect clients of the DAGSCC. Exploiting this insight, we present an algorithm to omit those analysis queries to compute the DAGSCC efficiently. Across 366 hot loops from 20 SPEC2006 benchmarks, this method computes the DAGSCC in half of the time using half as many queries.

show abstract