Path: utzoo!mnetor!uunet!lll-winken!lll-tis!ptsfa!pacbell!att-ih!ihnp4!inuxc!iuvax!pur-ee!uiucdcs!uxc.cso.uiuc.edu!ccvaxa!aglew
From: aglew@ccvaxa.UUCP
Newsgroups: comp.arch
Subject: Re: Cache lookahead
Message-ID: <28200127@ccvaxa>
Date: 20 Mar 88 23:01:00 GMT
References: <2769@mmintl.UUCP>
Lines: 19
Nf-ID: #R:mmintl.UUCP:2769:ccvaxa:28200127:000:686
Nf-From: ccvaxa.UUCP!aglew    Mar 20 17:01:00 1988


I have a paper from Purdue EE entitled "Compiler Driven Cache Policy"
which is basically about providing compiler control of cache.
Their policy is called SCP - Software Cache Policy.

Authors are Chi, C., and Dietz, H.G.
June 1987.

Chi talks about a machines that have advisory bits, like "read this and
keep in cache", but his algorithm is general to machines that use explicit
instructions. It is not general to all codes, although this is promised for
a later report.

---

Someone from the CEDAR group at the U of I gave a talk showing that, in
his research, software controlled caching was not a win.

The trouble is, this is not a disproof - all we need is one existence
proof.