|Writing a FAST compiler. email@example.com (1991-08-05)|
|Re: Writing a FAST compiler. firstname.lastname@example.org (1991-08-07)|
|Re: Writing a FAST compiler. email@example.com (1991-08-07)|
|Re: Writing a FAST compiler. firstname.lastname@example.org (1991-08-07)|
|Re: Writing a FAST compiler. email@example.com (1991-08-07)|
|Re: Writing a FAST compiler. firstname.lastname@example.org (Tom Wicklund) (1991-08-08)|
|From:||email@example.com (Clark Brooks)|
|Organization:||California Institute of Technology, Pasadena|
|Date:||7 Aug 91 19:23:40 GMT|
firstname.lastname@example.org (Patrick C Beard) writes:
>Most compiler efficiency issues revolve around creating highly optimized
>code. However, good code generation with fast turn-around time is often
>What are some general approaches I can follow to create
>a compiler that generates slow to moderate code very quickly?
Program in an interpreted language.
FORTH, APL, and sometimes SmallTalk have quick turnaround.
FORTH in particular requires almost no resources : I used to work with
an 8kbyte operating system in FORTH :-)
FORTH code is useful for many applications which are not time-intensive,
such as some user interface stuff. I found about a factor of 50 in run-time
speed from hand-assembly to FORTH code of the same function.
For some applications, there is negligible run-time cost in using APL.
Maybe you'll be lucky.
Return to the
Search the comp.compilers archives again.