Re: register allocation: basic blocks, liveness and next use

Max Hailperin <max@gustavus.edu>
Sun, 23 Mar 2008 12:04:59 -0500

          From comp.compilers
Related articles
register allocation: basic blocks, liveness and next use kevin.phillips83@yahoo.com (kphillips) (2008-03-22)
Re: register allocation: basic blocks, liveness and next use gene.ressler@gmail.com (Gene) (2008-03-22)
Re: register allocation: basic blocks, liveness and next use max@gustavus.edu (Max Hailperin) (2008-03-23)
Re: register allocation: basic blocks, liveness and next use max@gustavus.edu (Max Hailperin) (2008-03-23)
Re: register allocation: basic blocks, liveness and next use max@gustavus.edu (Max Hailperin) (2008-03-23)
Re: register allocation: basic blocks, liveness and next use kevin.phillips83@yahoo.com (kphillips) (2008-03-23)
Re: register allocation: basic blocks, liveness and next use gene.ressler@gmail.com (Gene) (2008-03-23)
Re: register allocation: basic blocks, liveness and next use cfc@shell01.TheWorld.com (Chris F Clark) (2008-03-23)
Re: register allocation: basic blocks, liveness and next use kevin.phillips83@yahoo.com (kphillips) (2008-03-27)
Re: register allocation: basic blocks, liveness and next use jeffrey.kenton@comcast.net (Jeff Kenton) (2008-04-03)
| List of all articles for this month |
From: Max Hailperin <max@gustavus.edu>
Newsgroups: comp.compilers
Date: Sun, 23 Mar 2008 12:04:59 -0500
Organization: Compilers Central
References: 08-03-084 08-03-088
Keywords: registers
Posted-Date: 23 Mar 2008 13:24:21 EDT

This is a correction to my own earlier post. I accidentally missed a
"not" in the following:

> ... then the simplest
> solution would be to treat procedure calls as block boundaries. The
> blocks won't be basic blocks any more -- but they probably will be
> what you want.

I meant to suggest *not* treating procedure calls as block
boundaries.

Post a followup to this message

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