Compilers vs. Interpreters (WAS: Unsafe Optimizations)

pardo@june.cs.washington.edu (David Keppel)
Sat, 23 Jun 90 21:38:59 GMT

          From comp.compilers

Related articles
Compilers vs. Interpreters (WAS: Unsafe Optimizations) pardo@june.cs.washington.edu (1990-06-23)
| List of all articles for this month |
Newsgroups: comp.compilers
From: pardo@june.cs.washington.edu (David Keppel)
Keywords: C, interpreter, debug
Organization: University of Washington, Computer Science, Seattle
References: <1990Jun22.172118.2020@esegue.segue.boston.ma.us>
Date: Sat, 23 Jun 90 21:38:59 GMT

David S. Harrison <davidh@dent.Berkeley.EDU> writes:
>[Some advantages of interpreters:
> * Develop testing while compiler under development.
> * Sanity check compare to compiled code (had better run on both).
> * Useful for debugging.]
>[Why don't most compiled languages also have an interpreter?]


There's more code to maintain, sometimes a large ammount of code.


Something that MIGHT be useful would be to write an instruction set
simulator for a simple (imaginary) architecture, and start by
retargeting your compiler to that. Then, only the ISA interpreter
has to be ported to get the system `up and running'.


Debuggers for interpreted code and debuggers for compiled code tend
to be different for various reasons. The approach above doesn't
address that.


;-D on ( Take my symbol table ... please ) Pardo
--
pardo@cs.washington.edu
        {rutgers,cornell,ucsd,ubc-cs,tektronix}!uw-beaver!june!pardo
--


Post a followup to this message

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