Re: Compiler requirements...help needed

"Travers Naran" <tnaran@direct.ca>
4 Nov 2000 01:40:12 -0500

          From comp.compilers

Related articles
Compiler requirements...help needed neal.a.benedict@intel.com (Benedict, Neal A) (2000-11-01)
Re: Compiler requirements...help needed smoleski@surakware.com (Sebastian Moleski) (2000-11-04)
Re: Compiler requirements...help needed tnaran@direct.ca (Travers Naran) (2000-11-04)
Re: Compiler requirements...help needed rbenedik@fsmat.htu.tuwien.ac.at (Ronald Benedik) (2000-11-05)
Re: Compiler requirements...help needed clc5q@cs.virginia.edu (Clark L. Coleman) (2000-11-07)
| List of all articles for this month |
From: "Travers Naran" <tnaran@direct.ca>
Newsgroups: comp.compilers
Date: 4 Nov 2000 01:40:12 -0500
Organization: Look Communications - http://www.look.ca
References: 00-11-018
Keywords: design, practice
Posted-Date: 04 Nov 2000 01:40:12 EST

"Benedict, Neal A" <neal.a.benedict@intel.com> wrote in message


> Hi,
>
> My name is Neal Benedict. I am with the compiler group at Intel. We
> are attempting to build a better compiler to take advantage of IA
> architecture. I am attempting to gather requirements from a broad
> community of users and could use your help. If anyone could spare a
> little time in which to discuss compiler features I world greatly
> appreciate it.


I don't know how many other people would find it useful, but here are
some things I've always wanted in a "C" compiler (which I assume he is
talking about).


* ISO/ANSI Standards conformance (no brainer)


* Support for using "C" as a back-end target language. Specifically,
the kind of pragmas that support generating non-C debugging code that
could convince a debugger to open my original source language instead
of the "C" language itself. I'm pretty sure this is standard, but
I've run across "C" compilers that refused to do this.


* I wish I knew more about the IA architecture, but flexible and
*swappable* optimization methods would be nice. It would be useful if
you could write your optimization modules and plug it into the
compiler.


* Documentation on how the optimizers work, what they're assumptions
are, etc. It would make it easier when tuning code to know what will
make the optimizer work really well or make it double your loop
length.


Back when I used to wrestle with the assembly level output of the gcc
compiler we were using, these features would have or did come in
handy.


> [I'd like a compiler that can compile programs of any size in
> arbitrary combinations of C, Pascal, Perl, Common Lisp, and Intercal in
> under a second, produce object programs of under 100 bytes with runtimes
> too short to measure, and fix any bugs in the source code while it's
> compiling them. Perhaps you could give us more hints about what areas
> you were planning to work on. -John]


Maybe he wants to implement PL/1? :-)
--
--------
Travers Naran: Computer Programmer & P/T Meddler In Time & Space
New Westminster, British Columbia, Canada, North America, Earth, yada, yada


Post a followup to this message

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