home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.arch
- Path: sparky!uunet!metaflow!rschnapp
- From: rschnapp@metaflow.com (Russ Schnapp)
- Subject: Re: SuperSPARC doesn't need superscalar compiler?!
- Message-ID: <BrstrI.GJ7@metaflow.com>
- Sender: usenet@metaflow.com
- Nntp-Posting-Host: habu
- Organization: Metaflow Technologies Inc.
- References: <9207201142.AA04083@x1sun6.ccl.itri.org.tw> <MOSS.92Jul21091941@ibis.cs.umass.edu>
- Date: Wed, 22 Jul 1992 16:28:30 GMT
- Lines: 42
-
- In article <MOSS.92Jul21091941@ibis.cs.umass.edu>, moss@cs.umass.edu (Eliot Moss) writes:
- |> >>>>> On 21 Jul 92 01:42:07 GMT, lycmit@X1SUN6.CCL.ITRI.ORG.TW (Yin-Chih Lin)
- |> >>>>> said:
- |>
- |> Yin-Chih> ... one of the Sun Micro employees had adverted that the SuperSPARC
- |> Yin-Chih> (superscalar SPARC which can issue 3 intructions at max performance)
- |> Yin-Chih> does not require the superscalar compiler to obtain the best CPU
- |> Yin-Chih> performance.
- |>
-
- |> Of course, you are right that in general a good schedule for a non-superscalar
- |> may not be the best schedule for a superscalar that does not have enough
- |> resources available on each cycle. For example, a non-superscalar might be
- |> able to issue one register to register multiply every cycle, given one fully
- |> pipelined multiplier. A superscalar that issues N instructions per cycle would
- |> need N fully pipeined units to guarantee not to stall, but most code probbaly
- |> does not need N fully pipelined multipliers.
- |>
-
- ...and there are other reasons why issue might stall on a superscalar
- machine. Even if there were, for instance, three multipliers, a conventional
- superscalar machine could not issue the following sequence in a single frame:
-
- smul %g1,%g2,%g3
- smul %g3,%g4,%g3
- smul %g3,6,%g1
-
- (i.e., calculating %g1 * %g2 * %g4 * 6). There are a couple of kinds of
- data hazards here. While this is a very contrived example, it points out
- one of the additional intricacies of generating code for conventional
- superscalar processors.
-
- There are, of course, solutions to superscalar design that allow
- unrecompiled code to execute at a very substantial performance
- improvement...
-
- ---
-
- ...Russ Schnapp
- Email: uunet!metaflow!rschnapp or rschnapp@metaflow.com
- Metaflow Technologies Voice: 619/452-6608x230; FAX: 619/452-0401
- La Jolla, California Unless otw specified, I`m speaking only for myself!
-