home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!sun-barr!news2me.ebay.sun.com!exodus.Eng.Sun.COM!appserv.Eng.Sun.COM!appserv!khb
- From: khb@chiba.Eng.Sun.COM (Keith Bierman fpgroup)
- Newsgroups: comp.lang.fortran
- Subject: Re: how to quit on an NaN result?
- Date: 17 Aug 92 19:28:06
- Organization: Sun MegaSystems
- Lines: 20
- Message-ID: <KHB.92Aug17192806@chiba.Eng.Sun.COM>
- References: <1992Aug16.200837.14024@samba.oit.unc.edu>
- <KHB.92Aug16143141@chiba.Eng.Sun.COM> <92230.171200HDK@psuvm.psu.edu>
- NNTP-Posting-Host: chiba
- In-reply-to: HDK@psuvm.psu.edu's message of 17 Aug 92 21:12:00 GMT
-
-
-
- >but most exception (interrupt) handlers would set the intermediate result
-
- No. The default is to fully conform to IEEE 754 which mandates
- denormalization. On an IBM S/370 family machine this would, of course,
- not be the case.
-
- The observation about cost is well taken. The cost of doing this *can*
- be non-trivial. On the TI SuperSPARC chip it is "free" (viz. done in
- hw) on the LSI and Cypress SPARC chipsets used in the 4/60|65|75 etc.
- family, it is expensive.
-
-
- --
- ----------------------------------------------------------------
- Keith H. Bierman keith.bierman@Sun.COM| khb@chiba.Eng.Sun.COM
- SunPro 2550 Garcia MTV 12-33 | (415 336 2648) fax 964 0946
- Mountain View, CA 94043 <speaking for myself, not Sun*> Copyright 1992
-
-