Ada compiler selection for embedded targets

  • Authors:
  • A. Tetewsky;R. Racine

  • Affiliations:
  • The Charles Stark Draper Laboratory, Cambridge, MA;The Charles Stark Draper Laboratory, Cambridge, MA

  • Venue:
  • ACM SIGAda Ada Letters
  • Year:
  • 1987

Quantified Score

Hi-index 0.00

Visualization

Abstract

Although mainframe Ada compilers have reached execution as well as reliability performance characteristics of languages such as FORTRAN, the same is not true of embedded system compilers. Finding a reliable cross-compiler for widely used and powerful micro-processors such as the Motorola 68000 family is extremely difficult, if not impossible. First quarter '87 cross-compiler users face many tough trade-offs and additional project costs due to time lost developing compiler work-arounds. Although the user community appreciates the technological challenges of building an Ada compiler, it is still faced with trying to use incomplete and somewhat unstable systems while trying to deliver efficient, ultra-reliable code. In order to help both vendors and users alike, this article presents a set of evaluation criteria that may be used in selecting an Ada compiler for embedded system applications. Given the progress made with mainframe compilers, the major focus is on robustness, reliability, run-time system, and cross-compiler tools.