Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!cs.utexas.edu!sun-barr!newstop!grapevine!male!jethro!exodus!exodus-bb!khb From: khb@chiba.Eng.Sun.COM (Keith Bierman fpgroup) Newsgroups: comp.lang.fortran Subject: Re: matrix multiplication Message-ID: Date: 8 May 91 17:15:39 GMT References: <1991May04.170203.22222@ariel.unm.edu> <23261@lanl.gov> <1991May07.155218.13741@ariel.unm.edu> Sender: news@exodus.Eng.Sun.COM Organization: Sun MegaSystems Lines: 28 In-reply-to: scavo@cie.uoregon.edu's message of 7 May 91 15:52:18 GMT In article <1991May07.155218.13741@ariel.unm.edu> scavo@cie.uoregon.edu (Tom Scavo) writes: As an aside, how did it happen that two competing parameter passing mechanisms got into the Fortran 77 standard? (This The standard was carefully crafted to allow compilers to do the highest performing thing on their platform. If you code according to the standard, you can't tell how parameter passing occurs. On traditional "scalar" uniprocessors, passing addresses is often the best way. On machines like the TMC and MASSPAR, copying is often cheaper. Looking to the future, how does Fortran 90 deal with the parameter passing problem? The same way; the standard doesn't mandate how you do it; just what the effects are. With interface blocks, the coder can communicate with the compiler a bit better (parameters passed just as IN, or OUT may merit special treatment, for example). -- ---------------------------------------------------------------- Keith H. Bierman keith.bierman@Sun.COM| khb@chiba.Eng.Sun.COM SMI 2550 Garcia 12-33 | (415 336 2648) Mountain View, CA 94043