Path: utzoo!attcan!uunet!maverick.ksu.ksu.edu!ux1.cso.uiuc.edu!ux1.cso.uiuc.edu!aglew
From: aglew@crhc.uiuc.edu (Andy Glew)
Newsgroups: comp.arch
Subject: Re: Why FP at all? (was: Re: Killer Micro II)
Message-ID: <AGLEW.90Sep6233218@dwarfs.crhc.uiuc.edu>
Date: 7 Sep 90 03:32:18 GMT
References: <527@llnl.LLNL.GOV> <603@array.UUCP> <2482@l.cc.purdue.edu>
	<STEPHEN.90Sep5000536@estragon.uchicago.edu> <1632@lhr.Morgan.COM>
	<STEPHEN.90Sep6215928@estragon.uchicago.edu>
Sender: news@ux1.cso.uiuc.edu (News)
Organization: Center for Reliable and High-Performance Computing University of
	Illinois at Urbana Champaign
Lines: 16
In-Reply-To: stephen@estragon.uchicago.edu's message of 6 Sep 90 21:59:28


>Is it so absurd to suggest, in sum, that exposing separate mantissa
>and exponent to the optimiser might result in *speedup* due to
>constant propagation and expression-rearrangement, while at the same
>time increasing expressivity by allowing an INDEPENDENT choice of
>mantissa and exponent sizes?

Someone at MIT(?) (MS thesis?) had a paper on "Micro-optimization of
floating point" in a conference within the last few years.

Note that some of the same ideas can be applied to hardware: for
example, you can combine the normalization post-shift with the
alignment pre-shift for forwarded FP operands, saving a cycle.

--
Andy Glew, a-glew@uiuc.edu [get ph nameserver from uxc.cso.uiuc.edu:net/qi]