Implications of structured programming for machine architecture
Communications of the ACM
Systematically derived instruction sets for high-level language support
ACM-SE 20 Proceedings of the 20th annual Southeast regional conference
Advances in Computer Architecture
Advances in Computer Architecture
The theory of parsing, translation, and compiling
The theory of parsing, translation, and compiling
The Theory of Parsing, Translation, and Compiling
The Theory of Parsing, Translation, and Compiling
Levels of representation of programs and the architecture of universal host machines
MICRO 11 Proceedings of the 11th annual workshop on Microprogramming
A study of language directed computer design
A study of language directed computer design
Interpreting machines: Architecture and programming of the B1700/B1800 series (Operating and programming systems series)
The contour model of block structured processes
ACM SIGPLAN Notices
Two-level hybrid interpreter/native code execution for combined space-time program efficiency
SIGPLAN '87 Papers of the Symposium on Interpreters and interpretive techniques
Viewing instruction set design as an optimization problem
MICRO 24 Proceedings of the 24th annual international symposium on Microarchitecture
Synthesis of instruction sets for pipelined microprocessors
DAC '94 Proceedings of the 31st annual Design Automation Conference
Optimal code generation for expressions on super scalar machines
ACM '86 Proceedings of 1986 ACM Fall joint computer conference
Processor Acceleration Through Automated Instruction Set Customization
Proceedings of the 36th annual IEEE/ACM International Symposium on Microarchitecture
International Journal of Parallel Programming - Special issue: Workshop on application specific processors (WASP)
Automated Custom Instruction Generation for Domain-Specific Processor Acceleration
IEEE Transactions on Computers
| Hi-index | 0.00 |
Conventional instruction sets or directly interpretable languages (DILs) have not been designed with high-level languages (HLLs) in mind. The modern design problem is to derive a space-time efficient DIL for a HLL processing system. In this paper, we present our approach to the problem of designing well-matched, space-time efficient DILs. A systematic, syntax- and semantics-directed DIL design methodology is presented. It calls for an incremental transformation of the source HLL, until a suitable target DIL is obtained. At the heart of the methodology is a canonic set of language transformations. An experimental study, involving several systematically derived DILs is carried out in order to characterize the relative merits and disadvantages of various sequences of transformations. Various space, time and interpretability trade-offs implied by the transformations are studied.