Path: utzoo!mnetor!uunet!husc6!mit-eddie!uw-beaver!cornell!rochester!PT.CS.CMU.EDU!K.GP.CS.CMU.EDU!lindsay
From: lindsay@K.GP.CS.CMU.EDU (Donald Lindsay)
Newsgroups: comp.arch
Subject: Re: taken -vs- untaken branches, Fortran FREQUENCY declaration
Message-ID: <615@PT.CS.CMU.EDU>
Date: 9 Jan 88 19:42:12 GMT
References: <496@cresswell.quintus.UUCP> <638@l.cc.purdue.edu> <836@ima.ISC.COM> <645@l.cc.purdue.edu> <8513@ism780c.UUCP>
Sender: netnews@PT.CS.CMU.EDU
Organization: Carnegie-Mellon University, CS/RI
Lines: 10

A FREQUENCY statement (or equivalent) is needed so that the tooling
has a way of feeding its discoveries back to the compiler.

RISC machines aren't the only ones who care about branch prediction.
IBM 370's have dedicated considerable hardware to the matter - conditional
execution of instructions down both possible paths, and the like. 
The new VLIW machines, such as the Multiflow Trace, eliminate this
hardware, so the compiler's analysis becomes crucial.
-- 
	Don		lindsay@k.gp.cs.cmu.edu    CMU Computer Science