Optimal bitwise register allocation using integer linear programming

DMCA

Optimal bitwise register allocation using integer linear programming (2006)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Rajkishore Barik , Christian Grothoff , Rahul Gupta , Vinayaka P , Raghavendra Udupa

Venue:	In International Workshop on Languages and Compilers for Parallel Computing (LCPC’06), LNCS
Citations:	3 - 0 self

BibTeX

@INPROCEEDINGS{Barik06optimalbitwise,
    author = {Rajkishore Barik and Christian Grothoff and Rahul Gupta and Vinayaka P and Raghavendra Udupa},
    title = {Optimal bitwise register allocation using integer linear programming},
    booktitle = {In International Workshop on Languages and Compilers for Parallel Computing (LCPC’06), LNCS},
    year = {2006},
    publisher = {Springer Verlag}
}

OpenURL

Abstract

Abstract. This paper addresses the problem of optimal global register allocation. The register allocation problem is expressed as an integer linear programming problem and solved optimally. The model is more flexible than previous graph-coloring based methods and thus allows for register allocations with significantly fewer moves and spills. The formulation can also model complex architectural features, such as bit-wise access to registers. With bit-wise access to registers, multiple subword temporaries can be stored in a single register and accessed effi-ciently, resulting in a register allocation problem that cannot be addressed effec-tively with simple graph coloring. The paper describes techniques that can help reduce the problem size of the ILP formulation, making the algorithm feasible in practice. Preliminary empirical results from an implementation prototype are reported. 1

Keyphrases

optimal bitwise register allocation integer linear programming register allocation problem bit-wise access ilp formulation register allocation simple graph coloring complex architectural feature optimal global register allocation single register algorithm feasible implementation prototype multiple subword temporary preliminary empirical result problem size previous graph-coloring

DMCA