Newsgroups: comp.parallel.pvm
From: wrankin@ee.duke.edu (William T. Rankin)
Subject: Question, PvmDataInPlace|PvmDataRaw
Keywords: multiple encoding packing
Organization: Duke University EE Dept.; Durham, NC
Date: 16 Jan 1995 16:19:23 GMT
Message-ID: <3fe6ab$ac5@news.duke.edu>

I have a question to the PVM group

PVM supports different packing/encoding strategies to optimize 
communications.  Two of the ones I use are PvmDataInPlace for 
efficently moving around some larger static data structures, and 
PvmDataRaw when I know I'm on a homogeneous processor set.

However, while logically it would seem that the two encoding
schemes address independant issues, they cannot be used together.
From a cursory glance through the PVM 3.3.6 source, the code that
performs the encoding and packing will not allow me to do something
like:

   pvm_initsend(PvmDataInPlace|PvmDataRaw);

for the case where I am moving static data around on a homogeneous
cluster.

Are there any plans to implement this enhancement?  If not, then can I
request it?  And finally, if I go ahead and make this enhancement to 
the code, would any of the PVM folks be interested in it?

Jst a simple question, eh? :-)

Thanks for the help,

-bill



-- 
----                                /       __/    /    /
bill rankin                        /              /    /
wrankin@ee.duke.edu               ___  /    /    /    /
philosopher/coffee-drinker       /    /    /    /    /
molecules-R-us                  /    /    /    /    /
                             _______/  __/  __/  __/

"All I wanna do is have a little fun before I die."  - Sheryl Crow

