Interpreters and computationally intensive programs

Ronald Bodkin <rjbodkin@theory.lcs.mit.edu> writes:


      [Interpretation] is sometimes impractical... on any project where
      there is a serious amount of computation, you can't afford to test
      and run interpreted code because of the delay involved in
      testing/using the program.


For this reason, a practical interpreter allows the mixing of interpreted
and compiled code. An ideal interpreter (not ours, unfortunately) allows
you to swap between interpreted and compiled code in the middle of
execution, so that you can execute at full speed for a while, then swap
some of your program into interpreted code, then continue executing. This
assumes that interpreted execution gives you some benefit, such as better
debugging, runtime checking, or exact arithmetic (as in the following
citation).


For an examination of the issues, see B. Chase and R. Hood, "Selective
Interpretation as a Technique for Debugging Computationally Intensive
Programs", some ACM publication (sorry), 1987, pp. 113-124. This is one
of the papers from the Rn project at Rice. The paper seems to have been
written before much was working; can anyone at Rice give us an update?
Has the Fortran interpreter with its exact arithmetic proven useful?


----
Sam Kendall
CenterLine Software, Inc. (formerly Saber Software)
--

Return to the comp.compilers page.
Search the comp.compilers archives again.