Interprocedural dataflow analysis in an executable optimizer

Goodwin, David W.

doi:10.1145/258915.258927

Cited by 21 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Over the years several different solutions have been proposed for computing the mayUse and byPass sets [3,5,9]. Of these solutions, Muth's solution [9] is as precise as the others and it is the most elegant one, as it computes the mayUse and byPass sets with the same equations (albeit on different sets) as the context-sensitive liveness analysis itself.…”

Section: State-of-the-art Analysismentioning

confidence: 99%

“…This analysis is based on link-time data flow analysis of whole programs. Binary rewriting tools such as link-time optimizers [2][3][4][5] or instrumentation tools [6,7], usually use similar data flow analyses as compilers. The most important advantage of applying those analyses at link time is the whole-program overview that is available when a program is being linked.…”

Section: Introductionmentioning

confidence: 99%

“…While it seems obvious that calling convention information is useful to avoid the need for making overly conservative assumptions regarding register spills, we have noticed that existing executable code data flow analyses [3][4][5]9] fail to exploit most if not all of the available information. This contrasts with the fact that previous work has clearly demonstrated the benefits of having more accurate whole-program data flow analyses.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Bidirectional liveness analysis, or how less than half of the Alpha’s registers are used

Sutter

Bus

Bosschere

2006

Journal of Systems Architecture

View full text Add to dashboard Cite

Interprocedural data flow analyses of executable programs suffer from the conservative assumptions that need to be made because no precise control flow graph is available and because registers are spilled onto the stack. This paper discusses the exploitation of calling-conventions in executable code data flow analyses to avoid the propagation of the conservative assumptions throughout a program.Based on this exploitation, the existing backward liveness analyses are improved, and a complementary forward liveness analysis is proposed. For the SPECint2000 programs compiled for the Alpha architecture, the combined forward and improved backward analysis on average finds 62% more free registers than the existing stateof-the-art liveness analysis.With the improved analysis, we are able to show that on average less than half of the registers of the RISC Alpha architecture are used. This contradicts the common wisdom that compilers can exploit large, uniform register files as found on RISC architectures.

show abstract

Section: State-of-the-art Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Bidirectional liveness analysis, or how less than half of the Alpha’s registers are used

Sutter

Bus

Bosschere

2006

Journal of Systems Architecture

View full text Add to dashboard Cite

show abstract

“…While a number of systems have been described for link-time code optimization 5,15,16,27,30,31,33], to the best of our knowledge, any alias analysis carried out by these systems is limited to fairly simple local analyses.…”

Section: Related Workmentioning

confidence: 99%

Alias analysis of executable code

Debray

Muth

Weippert

1998

Proceedings of the 25th ACM SIGPLAN-SIGACT Symposium on Principles of Programming Languages - POPL '98

View full text Add to dashboard Cite

Recent years have seen increasing interest in systems that reason about and manipulate executable code. Such systems can generally bene t from information about aliasing. Unfortunately, most existing alias analyses are formulated in terms of high-level language features, and are unable to cope with features, such as pointer arithmetic, that pervade executable programs. This paper describes a simple algorithm that can be used to obtain aliasing information for executable code. In order to be practical, the algorithm is careful to keep its memory requirements low, sacri cing precision where necessary to achieve this goal. Experimental results indicate that it is nevertheless able to provide a reasonable amount of information about memory references across a variety of benchmark programs.

show abstract

“…The development of efficient and precise dependence analysis algorithms continues as an active area of research in the programming languages and software engineering communities (e.g., [3,8,10,19]). …”

Section: Refining Chainsmentioning

confidence: 99%