Related articles |
---|
Why do intermediate codes have >, >=? alan@ez2.ezlink.com (1996-05-21) |
Re: Why do intermediate codes have >, >=? mark@omnifest.uwm.edu (1996-05-25) |
Re: Why do intermediate codes have >, >=? henry@zoo.toronto.edu (Henry Spencer) (1996-06-13) |
Re: Why do intermediate codes have >, >=? cwf@research.bell-labs.com (Chris Fraser) (1996-06-21) |
From: | Chris Fraser <cwf@research.bell-labs.com> |
Newsgroups: | comp.compilers |
Date: | 21 Jun 1996 17:07:41 -0400 |
Organization: | Compilers Central |
Keywords: | design |
Dave might remember something that I've forgotten, but I think lcc has those
extra codes mainly because I instinctively avoid asymmetry, but that's a
poor rationale. I'd be hard-pressed to refute Alan's point.
Henry's note about side effects can't explain lcc's redundant intermediate
codes, because lcc's front end moves side effects into separate trees or
dags before generating the IR.
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.