Compiler optimizations with DSP-Specific semantic descriptions

Authors:
Yung-Chia Lin;Yuan-Shin Hwang;Jenq Kuen Lee
Affiliations:
Department of Computer Science, National Tsing Hua University, Hsinchu, Taiwan;Department of Computer Science, National Taiwan Ocean University, Keelung, Taiwan;Department of Computer Science, National Tsing Hua University, Hsinchu, Taiwan
Venue:
LCPC'02 Proceedings of the 15th international conference on Languages and Compilers for Parallel Computing
Year:
2002

Citing 12
Cited 4

Fortran 90D/HPF compiler for distributed memory MIMD computers: design, implementation, and performance results

Proceedings of the 1993 ACM/IEEE conference on Supercomputing
ISDL: an instruction set description language for retargetability

DAC '97 Proceedings of the 34th annual Design Automation Conference
Efficient support of parallel sparse computation for array intrinsic functions of Fortran 90

ICS '98 Proceedings of the 12th international conference on Supercomputing
A function-composition approach to synthesize Fortran 90 array operations

Journal of Parallel and Distributed Computing
EXPRESSION: a language for architecture exploration through compiler/simulator retargetability

DATE '99 Proceedings of the conference on Design, automation and test in Europe
Predicting performance potential of modern DSPs

Proceedings of the 37th Annual Design Automation Conference
Compiler optimization on instruction scheduling for low power

ISSS '00 Proceedings of the 13th international symposium on System synthesis
Integrating Automatic Data Alignment and Array Operation Synthesis to Optimize Data Parallel Programs

LCPC '97 Proceedings of the 10th International Workshop on Languages and Compilers for Parallel Computing
Probabilistic Inference Schemes for Sparsity Structures of Fortran 90 Array Intrinsics

ICPP '02 Proceedings of the 2001 International Conference on Parallel Processing
A New Approach to DSP Intrinsic Functions

HICSS '00 Proceedings of the 33rd Hawaii International Conference on System Sciences-Volume 8 - Volume 8
Exploiting Multiple Levels of Parallelism in OpenMP: A Case Study

ICPP '99 Proceedings of the 1999 International Conference on Parallel Processing
Probabilistic points-to analysis

LCPC'01 Proceedings of the 14th international conference on Languages and compilers for parallel computing

Compiler support for speculative multithreading architecture with probabilistic points-to analysis

Proceedings of the ninth ACM SIGPLAN symposium on Principles and practice of parallel programming
Effective Code Generation for Distributed and Ping-Pong Register Files: A Case Study on PAC VLIW DSP Cores

Journal of Signal Processing Systems
Extendable pattern-oriented optimization directives

CGO '11 Proceedings of the 9th Annual IEEE/ACM International Symposium on Code Generation and Optimization
Extendable pattern-oriented optimization directives

ACM Transactions on Architecture and Code Optimization (TACO)

Quantified Score

Hi-index	0.00

Visualization

Abstract

Due to the specialized architecture and stream-based instruction set, traditional DSP compilers usually yield poor-quality object codes. Lack of an insight into the DSP architecture and the specific semantics of DSP applications, a compiler would have trouble selecting appropriate special instructions to exploit advanced hardware features. In order to extract optimal performance from DSPs, we propose a set of user-specified directives called Digital Signal Processing Interface (DSPI), which can facilitate code generation by relaying DSP specific semantics to compilers. We have implemented a prototype compiler based on the SPAM and SUIF compiler toolkits and integrated the DSPI into the prototype compiler. The compiler is currently targeted to TI's TMS320C6X DSP and will be extended to a retargetable compiler toolkit for embedded systems and System-on-a-Chip (SoC) platforms. Preliminary experimental results show that by incorporating DSPI directives significant performance improvements can be achieved in several DSP applications.