Predicate Analysis Based on Path Information

Predicated execution leads new challenges to traditional optimizers. Limited by predicate representations, current predicate analysis methods could only perform locally or have low efficiency. A new representation based on path information is proposed in this paper, based on which a global predicate relation query system is constructed. Experiment results indicate that precise and efficient global predicate analysis could be achieved with this system and the performance of optimized codes can also be improved.

[1]  Ken Kennedy,et al.  Conversion of control dependence to data dependence , 1983, POPL '83.

[2]  W. Hwu,et al.  Accurate and efficient predicate analysis with binary decision diagrams , 2000, Proceedings 33rd Annual IEEE/ACM International Symposium on Microarchitecture. MICRO-33 2000.

[3]  Scott A. Mahlke,et al.  Characterizing the impact of predicated execution on branch prediction , 1994, Proceedings of MICRO-27. The 27th Annual IEEE/ACM International Symposium on Microarchitecture.

[4]  Dirk Grunwald,et al.  A system level perspective on branch architecture performance , 1995, Proceedings of the 28th Annual International Symposium on Microarchitecture.

[5]  Scott Mahlke,et al.  Effective compiler support for predicated execution using the hyperblock , 1992, MICRO 1992.

[6]  Gary S. Tyson,et al.  The effects of predicated execution on branch prediction , 1994, Proceedings of MICRO-27. The 27th Annual IEEE/ACM International Symposium on Microarchitecture.

[7]  Richard Johnson,et al.  Analysis techniques for predicated code , 1996, Proceedings of the 29th Annual IEEE/ACM International Symposium on Microarchitecture. MICRO 29.

[8]  Alexandre E. Eichenberger,et al.  Register allocation for predicated code , 1995, Proceedings of the 28th Annual International Symposium on Microarchitecture.