home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.arch
- Path: sparky!uunet!metaflow!rschnapp
- From: rschnapp@metaflow.com (Russ Schnapp)
- Subject: Re: Not using integers
- Message-ID: <BuKsCu.4uH@metaflow.com>
- Sender: usenet@metaflow.com
- Nntp-Posting-Host: habu
- Organization: Metaflow Technologies Inc.
- References: <1992Sep8.182423.26063@athena.mit.edu> <1265@taniwha.UUCP>
- Date: Mon, 14 Sep 1992 15:58:05 GMT
- Lines: 19
-
- In article <1265@taniwha.UUCP>, paul@taniwha.UUCP (Paul Campbell) writes:
- |> The good old Burroughs 6700 (and family now called the A-series) had/has a
- |> FP format for which a value with a 0 exponent looks like a 'normal' integer
- |> (they use a signed integer rather than a 2's complement one) - the hardware
- |> goes out of its way to keep a 0 exponent for values if it possibly can ...
- |> this means that there are no FP/INT instructions, there's only one ADD, SUB,
- |> MULT etc instructions
-
- As I recall, the A-series is a tagged architecture, so there would be only one
- ADD, SUB, etc., instruction despite the data format. The tag automatically
- indicates the datatype, and therefore, the kind of operation to perform. Or
- didn't the tags go that far? It's been so long since I mucked with A-series
- code generation...
- --
-
- ...Russ Schnapp
- BIX: rschnapp 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!
-