OCEANS: Optimizing compilers for embedded applications
This paper describes the recently funded ESPRIT project OCEANS. Its aim is to investigate and develop advanced compiler infrastructure for embedded VLIW processors, such as the Philips TriMedia. Such processors promise high performance at low unit cost. This paper outlines the project's aims, presents the compiler infrastructure and its application to a typical case study.
KeywordsCache Line Register Allocation Assembly Code Software Pipeline Very Long Instruction Word
Unable to display preview. Download preview PDF.
- 1.G. Araujo et al. Challenges in Code Generation for Embedded Processors. In Code Generation for Embedded Processors. Kluwer Academic Publishers, pp. 49–64, 1995.Google Scholar
- 2.A. J. C. Bik, H. A. G. Wijshoff. MT1: A Prototype Restructuring Compiler. Technical Report 93-32, Department of Computer Science, Leiden University, Oct. 1993.Google Scholar
- 4.S. Carr. Combining Optimizations for Cache and Instruction-Level Parallelism. Proceedings of PACT'96.Google Scholar
- 5.B. Case. Philips Hope to Displace DSPs with VLIW. Microprocessor Report, 8(16), 5 Dec. 1994, pp. 12–15. See also http://www.trimedia-philips.com/Google Scholar
- 6.C. Eisenbeis, S. Lelait, B. Marmol. The meeting graph: a new model for loop cyclic register allocation. Proceedings of PACT'95.Google Scholar
- 7.E. Rohou, F. Bodin, A. Seznec, G. Le Fol, F. Charot, F. Raimbault. SALTO: System for Assembly-Language Transformation and Optimization. Technical Report 1032, IRISA, June 1996. See also http://www.irisa.fr/capsGoogle Scholar
- 8.J. Wang, C. Eisenbeis, M. Jourdan, B. Su. Decomposed Software Pipelining: a New Perspective and a New Approach. International Journal on Parallel Processing, 22(3), 1994, pp. 357–379. *** DIRECT SUPPORT *** A0008C42 00047Google Scholar