Related articles |
---|
Optimizing Across && And || bart@cs.uoregon.edu (1995-02-15) |
Re: Optimizing Across && And || preston@tera.com (1995-02-18) |
Re: Optimizing Across && And || bill@amber.ssd.csd.harris.com (1995-02-19) |
Re: Optimizing Across && And || bart@cs.uoregon.edu (1995-02-21) |
Re: Optimizing Across && And || whalley@fork.cs.fsu.edu (David Whalley) (1995-02-22) |
Re: Optimizing Across && And || bart@cs.uoregon.edu (1995-02-23) |
Re: Optimizing Across && And || bill@amber.ssd.csd.harris.com (1995-02-24) |
Re: Optimizing Across && And || glew@ichips.intel.com (1995-02-27) |
Re: Optimizing Across && And || bill@amber.ssd.csd.harris.com (1995-02-28) |
Re: Optimizing Across && And || bill@amber.ssd.csd.harris.com (1995-02-28) |
Re: Optimizing Across && And || ryg@summit.novell.com (1995-03-03) |
[8 later articles] |
Newsgroups: | comp.compilers |
From: | David Whalley <whalley@fork.cs.fsu.edu> |
Keywords: | C, optimize |
Organization: | Compilers Central |
References: | 95-02-123 |
Date: | Wed, 22 Feb 1995 12:56:42 GMT |
There have been a number of articles recently about optimizing
the common subexpressions that are embedded in boolean expressions.
It was noted that this is difficult since the code has already
been generated as conditional branches. It has even been stated that:
>I wouldn't expect many compilers to do this, though, because of the
>low likelihood of an opportunity for it.
Well, we have implemented a general optimization for this. We have
also found that this type of optimization can be exploited quite
frequently. The optimization is described in the paper "Avoiding
Conditional Branches in Loops by Code Replication," which will appear
in PLD&I '95.
You can obtain a copy of the original submission by anonymously ftping
the paper from ftp.cs.fsu.edu. It is in pub/whalley/papers/pldi95.ps.Z.
Dave
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.