Newsgroups: comp.parallel.pvm
From: tony@aurora.cs.msstate.edu (Tony Skjellum)
Subject: Re: Is PVM Thread Safe?
Organization: Mississippi State University
Date: 4 Jul 1995 15:04:17 GMT
Message-ID: <3tbl9h$fb@NNTP.MsState.Edu>

The PVM API has unsafe aspects to it.  

In pseudo-calls:
	Initialize a buffer
	Pack
	Pack
	...
	Pack
	Send or Mcast
One needs to know which message a given pack applies to; in the
thread world, this is not well-defined, unless we restrict ourselves
to the PVM 2.x model of one message at a time, which is unlikely
to be your choice.

How have you modified the API to accomodate thread safety? Are you
using a thread ID in calls?
-Tony Skjellum

Honbo Zhou (hzhou@cs.utk.edu) wrote:
: >I've also modified PVM to "become" thread safe but it is very inefficient
: >as there are a lot of "critical" regions specially in the buffer alloc
: >area and in the message buffer handling so I'm interested to see how
: >you've modified PVM. 

: In my MT PVM implementation on SMP, I use a general linked list to 
: link all the resources of every PVM task (they are thread now!) together
: so as to make them easier to manage and search and the modified code is 
: as efficient as the original PVM. 

: >	So if you do have the code in a shar package please do mail
: >it direct if possible or let me know of the ftp site etc.

: I am going to establish a home page for the stuff I did and the source
: will be available soon.

: Honbo 

--
Anthony Skjellum, Asst. Professor, MSU/CS/ERC, Ph: (601)325-8435; FAX: (601)325-8997.
Mosaic: http://www.erc.msstate.edu/~tony; e-mail: tony@cs.msstate.edu
Maxim:  "There is no lifeguard at the gene pool." - C. H. Baldwin
Quote:  "What a rain of ashes falls on him that sees the new and cannot leave the old."-Shak.


