Related articles |
---|
Reasons why you don't prove your programs are correct steve@hubcap.clemson.edu (1990-01-05) |
Re: Reasons why you don't prove your programs are correct keithd@anvil.oz.au (1990-01-12) |
Re: Reasons why you don't prove your programs are correct mayer@iuvax.cs.indiana.edu (Mayer Goldberg) (1990-01-17) |
Re: Reasons why you don't prove your programs are correct pardo@cs.washington.edu (1990-01-22) |
Re: Reasons why you don't prove your programs are correct pardo@cs.washington.edu (1990-01-24) |
Re: Reasons why you don't prove your programs are correct gateley@m2.csc.ti.com (1990-01-26) |
From: | gateley@m2.csc.ti.com (John Gateley) |
Newsgroups: | comp.compilers |
Date: | 26 Jan 90 18:45:25 GMT |
References: | <7578@hubcap.clemson.edu> <1990Jan16.232703.2368@esegue.segue.boston.ma.us> <1990Jan22.234119.7002@esegue.segue.boston.ma.us> <1990Jan24.220734.223@esegue.segue.boston.ma.us> |
Organization: | TI Computer Science Center, Dallas |
In article <1990Jan24.220734.223@esegue.segue.boston.ma.us> pardo@cs.washington.edu (David Keppel) writes:
>A related issue: do you prove your compiler correct?
The scheme 311 compiler had a proof of correctness. It is published
as "The Scheme 311 Compiler, An Exercise in Denotational Semantics" in
a conference proceedings (my copy doesn't have the name). The stuff
at the bottom is: (c) 1984 ACM 0-89791-142-3/84/008/0356 $00.75
and it is probably one of: the lisp and functional programing conference,
POPL, or the compiler conference.
John
gateley@m2.csc.ti.com
Return to the
comp.compilers page.
Search the
comp.compilers archives again.