Special features of a VLIW architecture

This paper describes some of the special features of a very long instruction word architecture that is based on a percolation scheduling (PS) compiler. The architecture exploits fine-grain (instruction-level) parallelism in order to speed up program execution. Recent developments in low-level code transformations have been successful in exposing this parallelism. In order to take advantage of the capabilities of the PS compiler, the authors have devised a novel multi-way branch scheme that improves the overall performance of the architecture and allows speculative execution of the operations in the branch and branch delay slots. They have also adopted a modified instruction execution pipeline, which reduces the frequency of data hazards in the processor.<<ETX>>

[1]  B. Ramakrishna Rau,et al.  The Cydra 5 departmental supercomputer: design philosophies, decisions, and trade-offs , 1989, Computer.

[2]  Alexander Aiken,et al.  Perfect Pipelining: A New Loop Parallelization Technique , 1988, ESOP.

[3]  Kemal Ebcioglu,et al.  A global resource-constrained parallelization technique , 1989 .

[4]  Gerrit A. Slavenburg,et al.  CREATE-LIFE: a modular design approach for high performance ASICs , 1990, Digest of Papers Compcon Spring '90. Thirty-Fifth IEEE Computer Society International Conference on Intellectual Leverage.

[5]  Robert P. Colwell,et al.  A VLIW architecture for a trace scheduling compiler , 1987, ASPLOS 1987.

[6]  Michael J. Flynn,et al.  Detection and Parallel Execution of Independent Instructions , 1970, IEEE Transactions on Computers.