Newsgroups: comp.arch
Path: utzoo!henry
From: henry@zoo.toronto.edu (Henry Spencer)
Subject: Re: Object-Oriented Procedure Call (was: Smart I-cache?)
Message-ID: <1990Nov26.205811.27083@zoo.toronto.edu>
Organization: U of Toronto Zoology
References: <2823@crdos1.crd.ge.COM> <1157@cameron.egr.duke.edu> <657720712@lear.cs.duke.edu> <MOSS.90Nov24181741@ibis.cs.umass.edu>
Date: Mon, 26 Nov 90 20:58:11 GMT

In article <MOSS.90Nov24181741@ibis.cs.umass.edu> moss@cs.umass.edu (Eliot Moss) writes:
>Mention of architectural speedups via special branch target cache mechanisms,
>and the fact that they do not apply to "object-oriented" procedure calls
>because they are indirect...

???  The AMD 29k BTC, the major example that I know about, works regardless
of what kind of branch you use.  It even works on interrupts, as I recall.
Caches have the advantage of being able to observe run-time behavior, so
they don't *care* whether the branches are not statically predictable.
-- 
"I'm not sure it's possible            | Henry Spencer at U of Toronto Zoology
to explain how X works."               |  henry@zoo.toronto.edu   utzoo!henry
