TY - GEN
T1 - Register allocation in structured programs
AU - Kannan, Sampath
AU - Proebsting, Todd
N1 - Funding Information:
in part by NSF Grant
PY - 1995/1/22
Y1 - 1995/1/22
N2 - In this paper we look at the register allocation problem. In the literature this problem is frequently reduced to the general graph coloring problem and the solutions to the problem are obtained from graph coloring heuristics. Hence, no algorithm with a good performance guarantee is known. Here we show that when attention is restricted to structured programs which we define to be programs whose control-flow graphs are series-parallel, there is an efficient algorithm that produces a solution which is within a factor of 2 of the optimal solution. We note that even with the above restriction the resulting coloring problem is NP-complete. We also consider how to delete a minimum number of edges from arbitrary control-flow graphs to make them series-parallel and apply our algorithm. We show that this problem is Max SNP hard. However, we define the notion of an approximate articulation point and give efficient algorithms to find approximate articulation points. We present a heuristic for the edge deletion problem based on this notion which seems to work well when the given graph is close to series-parallel.
AB - In this paper we look at the register allocation problem. In the literature this problem is frequently reduced to the general graph coloring problem and the solutions to the problem are obtained from graph coloring heuristics. Hence, no algorithm with a good performance guarantee is known. Here we show that when attention is restricted to structured programs which we define to be programs whose control-flow graphs are series-parallel, there is an efficient algorithm that produces a solution which is within a factor of 2 of the optimal solution. We note that even with the above restriction the resulting coloring problem is NP-complete. We also consider how to delete a minimum number of edges from arbitrary control-flow graphs to make them series-parallel and apply our algorithm. We show that this problem is Max SNP hard. However, we define the notion of an approximate articulation point and give efficient algorithms to find approximate articulation points. We present a heuristic for the edge deletion problem based on this notion which seems to work well when the given graph is close to series-parallel.
UR - http://www.scopus.com/inward/record.url?scp=84912068039&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84912068039&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84912068039
T3 - Proceedings of the Annual ACM-SIAM Symposium on Discrete Algorithms
SP - 360
EP - 368
BT - Proceedings of the 6th Annual ACM-SIAM Symposium on Discrete Algorithms, SODA 1995
PB - Association for Computing Machinery
T2 - 6th Annual ACM-SIAM Symposium on Discrete Algorithms, SODA 1995
Y2 - 22 January 1995 through 24 January 1995
ER -