Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!apple!snorkelwacker.mit.edu!ai-lab!just-right!misha From: misha@just-right.ai.mit.edu (Mike Bolotski) Newsgroups: comp.lang.misc Subject: Re: Optimizing Message-ID: <11896@life.ai.mit.edu> Date: 15 Nov 90 20:25:19 GMT References: <5020:Nov1518:23:0790@kramden.acf.nyu.edu> Sender: news@ai.mit.edu Organization: MIT Artificial Intelligence Laboratory Lines: 23 In article <5020:Nov1518:23:0790@kramden.acf.nyu.edu> brnstnd@kramden.acf.nyu.edu (Dan Bernstein) writes: >What I'd love to see is a way to turn on optimizer debugging output for >a section of code. The optimizer would print out everything that it >looks for while optimizing, what it finds, and what it's not sure about. >Then the programmer can make assertions to fill in the missing gaps. This is something that everyone in the group will hopefully agree on. The Lucid Common Lisp compiler can do this, to an extent. It spits out, on request, optimizations it made and optimizations that it could have made, had it additional information. Granted, most of these are type declarations, are therefore inapplicable to C, but it's a definite step in the right direction. Unfortunately, I have yet to see a C compiler with a similar feature. And by the way, Dan, are we ever going to see a public admission of error on the "optimal addition sequence" issue? Or do you claim to be more an authority on the issue than Aho, Sethi, or Ullman? Mike Bolotski Artificial Intelligence Laboratory, MIT misha@ai.mit.edu Cambridge, MA