Optimizing for space and time usage with speculative partial redundancy elimination

Scholz, Bernhard; Horspool, Nigel; Knoop, Jens

doi:10.1145/997163.997195

Cited by 15 publications

(8 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Significant performance gains can be expected from these optimizations, leveraging the benefits of a more application-specific PRE technique that may be derived from our algorithm. Furthermore, our algorithm may find applications in some embedded applications [Scholz et al 2004] where performance improvements can be beneficial. One future work is to evolve MC-PRE into a more application-specific PRE technique in these research areas so that more significant performance gains can be obtained.…”

Section: Resultsmentioning

confidence: 99%

“…Because of the local heuristics used, both algorithms are not computationally optimal. Later, computationally optimal results are achieved by the three different algorithms described in Bodik [1999], Cai and Xue [2003], and Scholz et al [2004], which all rely on finding a minimum cut to obtain the required insertion points. However, none of these three algorithms is also lifetime optimal due to the arbitrary minimum cuts chosen in their algorithms.…”

Section: Related Workmentioning

confidence: 99%

“…The PRE transformation obtained using such a cut will correspond to the Busy Code Motion (BCM) as described in Knoop et al [1992Knoop et al [ , 1994, resulting in the longest lifetimes possible. Note that in all three computationally optimal algorithms for speculative PRE [Bodik 1999;Cai and Xue 2003;Scholz et al 2004], min-cut cuts are arbitrarily chosen. As a result, the notion of isolated computations does not exist.…”

Section: Implementation Detailsmentioning

confidence: 99%

“…There are three computationally optimal algorithms for solving the speculative PRE problem [Bodik 1999;Cai and Xue 2003;Scholz et al 2004]. However, none of them is also lifetime optimal.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A lifetime optimal algorithm for speculative PRE

Xue

Cai

2006

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

A lifetime optimal algorithm, called MC-PRE, is presented for the first time that performs speculative PRE based on edge profiles. In addition to being computationally optimal in the sense that the total number of dynamic computations for an expression in the transformed code is minimized, MC-PRE is also lifetime optimal since the lifetimes of introduced temporaries are also minimized. The key in achieving lifetime optimality lies not only in finding a unique minimum cut on a transformed graph of a given CFG, but also in performing a data-flow analysis directly on the CFG to avoid making unnecessary code insertions and deletions. The lifetime optimal results are rigorously proved. We evaluate our algorithm in GCC against three previously published PRE algorithms, namely, MC-PRE comp (Qiong and Xue's computationally optimal version of MC-PRE), LCM (Knoop, Rüthing, and Steffen's lifetime optimal algorithm for performing nonspeculative classic PRE), and CMP-PRE (Bodik, Gupta, and Soffa's PRE algorithm based on code-motion preventing (CMP) regions, which is speculative but not computationally optimal). We report and analyze our experimental results, obtained from both actual program execution and instrumentation, for all 22 C, C++ and FORTRAN 77 benchmarks from SPECcpu2000 on an Itanium 2 computer system. Our results show that MC-PRE (or MC-PRE comp ) is capable of eliminating more partial redundancies than both LCM and CMP-PRE (especially in functions with complex control flow), and, in addition, MC-PRE inserts temporaries with shorter lifetimes than MC-PRE comp . Each of both benefits has contributed to the performance improvements in benchmark programs at the costs of only small compile-time and code-size increases in some benchmarks.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Implementation Detailsmentioning

confidence: 99%

“…There are three computationally optimal algorithms for solving the speculative PRE problem [Bodik 1999;Cai and Xue 2003;Scholz et al 2004]. However, none of them is also lifetime optimal.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A lifetime optimal algorithm for speculative PRE

Xue

Cai

2006

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

show abstract

“…There are three computationally optimal algorithms for speculative PRE [1,4,25]. Later we also developed a lifetime optimal algorithm [26].…”

Section: Related Work and Conclusionmentioning

confidence: 99%

A Fresh Look at PRE as a Maximum Flow Problem

Xue

Knoop

2006

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. We show that classic PRE is also a maximum flow problem, thereby revealing the missing link between classic and speculative PRE, and more importantly, establishing a common high-level conceptual basis for this important compiler optimisation. To demonstrate this, we formulate a new, simple unidirectional bit-vector algorithm for classic PRE based only on the well-known concepts of availability and anticipatability. Designed to find a unique minimum cut in a flow network derived from a CFG, which is proved simply but rigorously, our algorithm is simple and intuitive, and its optimality is self-evident. This conceptual simplicity also translates into efficiency, as validated by experiments.

show abstract