Overlapped loop support in the Cydra 5
- 1 April 1989
- proceedings article
- Published by Association for Computing Machinery (ACM)
- Vol. 17 (2) , 26-38
- https://doi.org/10.1145/70082.68185
Abstract
The CydraTM 5 architecture adds unique support for overlapping successive iterations of a loop to a very long instruction word (VLIW) base. This architecture allows highly parallel loop execution for a much larger class of loops than can be vectorized, without requiring the unrolling of loops usually used by compilers for VLIW machines. This paper discusses the Cydra 5 loop scheduling model, the special architectural features which support it, and the loop compilation techniques used to take full advantage of the architecture.Keywords
This publication has 9 references indexed in Scilit:
- The Cydra 5 departmental supercomputer: design philosophies, decisions, and trade-offsComputer, 1989
- Estimating interlock and improving balance for pipelined architecturesJournal of Parallel and Distributed Computing, 1988
- Automatic translation of FORTRAN programs to vector formACM Transactions on Programming Languages and Systems, 1987
- A VLIW architecture for a trace scheduling compilerPublished by Association for Computing Machinery (ACM) ,1987
- A Fortran compiler for the FPS-164 scientific computerPublished by Association for Computing Machinery (ACM) ,1984
- Conversion of control dependence to data dependencePublished by Association for Computing Machinery (ACM) ,1983
- Very Long Instruction Word architectures and the ELI-512Published by Association for Computing Machinery (ACM) ,1983
- A composite hoisting-strength reduction transformation for global program optimization part iiInternational Journal of Computer Mathematics, 1982
- Dependence graphs and compiler optimizationsPublished by Association for Computing Machinery (ACM) ,1981