论文信息 - Scheduling arithmetic and load operations in parallel with no spilling

Scheduling arithmetic and load operations in parallel with no spilling

We consider a machine model in which load operations can be performed in parallel with arithmetic operations by two separate functional units. For this model, the evaluation of a set of expression trees is discussed. A dynamic programming algorithm to produce an approximate solution is described and analyzed. For binary trees its worse case cost is at most 9.1% worse than the optimal cost.

Michael Rodeh | David Bernstein | Jeffrey M. Jaffe

[1] Jeffrey D. Ullman,et al. The Generation of Optimal Code for Arithmetic Expressions , 1970, JACM.

[2] Roman R. Redziejowski. On arithmetic expressions and trees , 1969, CACM.

[3] Alfred V. Aho,et al. Code Generation for Expressions with Common Subexpressions , 1977, J. ACM.

[4] Ikuo Nakata,et al. On compiling algorithms for arithmetic expressions , 1967, CACM.

[5] Alfred V. Aho,et al. Code-generation for machines with multiregister operations , 1977, POPL '77.

[6] Ron Y. Pinter,et al. Optimal scheduling of arithmetic operations in parallel with memory access (preliminary version) , 1985, POPL '85.

[7] Kai Hwang,et al. Computer architecture and parallel processing , 1984, McGraw-Hill Series in computer organization and architecture.

[8] John L. Bruno,et al. Code Generation for a One-Register Machine , 1976, J. ACM.

[9] Alfred V. Aho,et al. Optimal code generation for expression trees , 1975, STOC.

[10] George Radin. The 801 Minicomputer , 1983, IBM J. Res. Dev..

[11] JOHN L. HENNESSY,et al. VLSI Processor Architecture , 1984, IEEE Transactions on Computers.

[12] Richard M. Russell,et al. The CRAY-1 computer system , 1978, CACM.

[13] Ron Y. Pinter,et al. Optimal Scheduling of Arithmetic Operations in Parallel with Memory Accesses. , 1985 .