Related articles |
---|
Register Files storm@binkley.cs.mcgill.ca (1992-10-05) |
Re: Register Files preston@helena.cs.rice.edu (1992-10-05) |
Re: Register Files ssimmons@convex.com (1992-10-06) |
Re: Register Files cliffc@cs.rice.edu (1992-10-06) |
Re: Register Files hobbs@steven.enet.dec.com (1992-10-06) |
Re: Register Files kendall@centerline.com (1992-10-07) |
Re: Register Files bart@cs.uoregon.edu (1992-10-07) |
[3 later articles] |
Newsgroups: | comp.compilers |
From: | storm@binkley.cs.mcgill.ca (Marc WANDSCHNEIDER) |
Organization: | Compilers Central |
Date: | Mon, 5 Oct 1992 02:30:51 GMT |
Keywords: | registers |
I recently read an article on the Design and Implementation of the new DEC
Alpha Chip (to be renamed something like VPX so that people think it's at
least RELATED to the VAX ;-)) and there was a comment that caughy my eye.
Apparantly, one of the big decisions was whether or not it would be a good
idea to split the register file up into two -- one for integer and one for
floating point registers.
They only decided to do this AFTER they got the okay from the Compiler
writers.
Now, the quesiton is: What does this matter to the compiler writer, and
how can s/he take advantage of it. I don't see how it would affect
instruction ordering for pipelines, etc....
Any ideas???
Toodlepip!
marc 'em.
--
Return to the
comp.compilers page.
Search the
comp.compilers archives again.