home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!munnari.oz.au!ariel.ucs.unimelb.EDU.AU!ucsvc.ucs.unimelb.edu.au!mits.com.au!ratty.mits.com.au!lewis
- Newsgroups: comp.sys.amiga.hardware
- Subject: Re: Data/Instruction Cache & BURST modes on 68030? Why/when?
- Message-ID: <1992Dec24.110249.1@ratty.mits.com.au>
- From: lewis@ratty.mits.com.au
- Date: 24 Dec 92 11:02:49 +1000
- References: <hellerS.724958427@batman> <72192@cup.portal.com> <smcgerty.725067816@unix1.tcd.ie>
- Organization: MITS
- Nntp-Posting-Host: ratty
- Nntp-Posting-User: lewis
- Lines: 45
-
- In article <smcgerty.725067816@unix1.tcd.ie>, smcgerty@unix1.tcd.ie (Stephen John McGerty) writes:
- > In <72192@cup.portal.com> Tony-Preston@cup.portal.com (ANTHONY FRANCIS PRESTON) writes:
- >>Data cache is an area of memory(varies in size with different processors,
- >>generally, more is better) that the cpu saves data that might be used over
- >>again.
-
- [ Stuff cut ]
-
- > With this in mind, why is it that Motorola have seperate instruction
- > and data caches? On the i486 there is a single 8k cache used for
- > both instructions and data. On the M68040, there is one 4k cache
- > for data, and another 4k cache for instructions. Which is better?
- >
- > I hear that the Intel 586 is going to have seperate instruction
- > and data caches, as in the M68040, so this would seem to imply that
- > this is the prefered arrangement...?
- >
- > The only thing I can think of is that you might not want your instruction
- > cache getting overwritten with less frequently used data, in a vector
- > operation or something. But this isn't really a serious factor, or
- > so I'm told...
- >
- > Anyone any inspired thoughts?
- >
- > : / T | / Stephen John McGerty (C.Sci) "Theory must never Amiga // :
- > : / | |/ smcgerty@unix1.tcd.ie precede creation" \\// :
-
- The manufacturers decide on whether to use a unified or split cache
- arrangement by getting instruction and data trace data for real programs,
- and running this data through simulations of the difference cache
- arrangements, sizes, associativity, etc. This helps them to find which
- setup suits the processor design best.
-
- I recall from reading an IEEE Micro article on the i486 design that they
- found the single 8k cache performed better for their simulations, and in
- another article, I seem to remember a similar statement from the Motorola
- 68040 designers saying that their simulations showed the split 4k caches to
- perform better.
-
- --
- David Lewis,
-
- Internet: lewis@mits.com.au
- Phone: +61 3 613 9415 Fax: +61 3 613 9550
- "No call alligator long mouth till you pass him." - Jamaican Proverb.
-