Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!snorkelwacker.mit.edu!bloom-beacon!eru!hagbard!sunic!mcsun!ukc!pyrltd!pyra.co.uk!tonys
From: tonys@pyra.co.uk (Tony Shaughnessy)
Newsgroups: comp.unix.internals
Subject: Re: Fundamental defect of the concept of shared libraries
Message-ID: <674816585.AA7847@flaccid>
Date: 21 May 91 09:03:06 GMT
References: <MWM.91May17132439@raven.pa.dec.com> <213@titccy.cc.titech.ac.jp> <1991May20.090857@wsl.dec.com> <216@titccy.cc.titech.ac.jp>
Sender: news@pyra.co.uk (News system)
Reply-To: tonys@pyra.co.uk (Tony Shaughnessy)
Organization: Pyramid Technology Ltd, Farnborough, ENGLAND
Lines: 25

In article <216@titccy.cc.titech.ac.jp> mohta@necom830.cc.titech.ac.jp (Masataka Ohta) writes:
>In article <1991May20.090857@wsl.dec.com> mcnally@wsl.dec.com writes:
>
>>|> You don't know about hardware enough. Because address translation is time
>>|> consuming, fast cache is always indexed by virtual address. Thesedays,
>>|> virtually indexed cache is quite common.
>
>>|> So, if you want shared libraries, you can put it only on slower machines.
>
>>How about MIPS R3000/R4000?  Maybe that's not fast enough.
>
>The primary cache of R4000 is virtually indexed and physically tagged.
>That is, it can't map different virtual addresses to a physical address.
>
>							Masataka Ohta

I quote from the book "MIPS Risc Architecture" by Gerry Kane, Prentice Hall,
1989, page 4-1.

	"The mapping of these extended, process-unique virtual addresses to
	physical addresses need not be one-to-one; virtual addresses of two
	or more different processes may map to the same physical address."

Tony Shaughnessy
tonys@pyra.co.uk