Re: Efficient bytecode design and interpretation

Martin von Loewis <loewis@informatik.hu-berlin.de>
30 May 2001 00:01:55 -0400

          From comp.compilers

Related articles
Efficient bytecode design and interpretation mg169780@zodiac.mimuw.edu.pl (Michal Gajda) (2001-05-22)
Re: Efficient bytecode design and interpretation anton@mips.complang.tuwien.ac.at (2001-05-29)
Re: Efficient bytecode design and interpretation jonm@fishwife.cis.upenn.edu (2001-05-30)
Re: Efficient bytecode design and interpretation loewis@informatik.hu-berlin.de (Martin von Loewis) (2001-05-30)
Re: Efficient bytecode design and interpretation eugene@datapower.com (Eugene Kuznetsov) (2001-05-30)
Re: Efficient bytecode design and interpretation korek@icm.edu.pl (2001-05-31)
Re: Efficient bytecode design and interpretation usenet.2001-05-30@bagley.org (Doug Bagley) (2001-05-31)
Re: Efficient bytecode design and interpretation anton@mips.complang.tuwien.ac.at (2001-06-03)
Re: Efficient bytecode design and interpretation anton@mips.complang.tuwien.ac.at (2001-06-03)
| List of all articles for this month |

From: Martin von Loewis <loewis@informatik.hu-berlin.de>
Newsgroups: comp.compilers
Date: 30 May 2001 00:01:55 -0400
Organization: Humboldt University Berlin, Department of Computer Science
References: 01-05-068
Keywords: interpreter, performance, practice
Posted-Date: 30 May 2001 00:01:55 EDT

Michal Gajda <mg169780@zodiac.mimuw.edu.pl> writes:


> where can I find information about writting efficient bytecode
> interpreter? Or some at least comparison of a LOT of open source
> bytecode interpreters so that I could peep into their code?


I don't think there is any deep recent analysis of byte code designs.
Instead, you probably have to look at the individual interpreters and
compare them yourself. If you are interested in that approach, people
might want to help contributing documentation for the individual byte
code designs. E.g. the Python byte code documentation can be found
at


http://www.python.org/doc/current/lib/bytecodes.html


However, I'd like to suggest that efficiency of the interpreter is
*not* a matter of properly designing the byte code. Other aspects,
like memory management or thread-safety overhead matter much more for
interpreter efficiency. The closest thing to bytecodes with an impact
for efficiency is probably the quality of the just-in-time compiler,
if the interpreter uses any.


Regards,
Martin


Post a followup to this message

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