Newsgroups: comp.parallel.pvm
From: bijal@nuc.berkeley.edu (Bijal C. Modi)
Subject: Re: Chosing the best encoding
Organization: University of California
Date: 10 May 1995 08:15:55 GMT
Message-ID: <3opsnr$e2d@agate.berkeley.edu>

In article <Pine.A32.3.91.950510093750.14045A-100000@frodo.cra.enel.it>,
Andrea OMODEO  <andrea@frodo.cra.enel.it> wrote:
>Hello,
>	I have a problem that someone on the net may already have solved.
>I have two PVM tasks involved in point-to-point communications.
>Since the tasks can potentially reside on incompatible hosts, I am forced to
>use pvm_initsend(PvmDataDefault), even if we all know that PvmDataInPlace is
>much faster when hosts are compatible.
  :
>I think this approach is more viable than simply comparing the PVM_ARCH value
>(because there are many different PVM_ARCHs using the same binary
>representation) and it is more suitable as long as new architectures are
>added to the PVM list.
>	Thanks in advance,
>		Andrew.

One can always use pvm_archcode() to see when you can get away with PvmDataRaw,
instead of PvmDataDefault for encoding purpose. See man page for more details.

-Bijal.

--
                *      /\                       *
        *             |()|        *         Bijal Modi                    *
           *          |  |                  bijal@fusion.berkeley.edu
 *              *     |  |     *        *   Fusion Transport Group

