Newsgroups: comp.parallel.pvm
From: Steve Simonds <simonds@sgi.com>
Subject: Re: PVM on shared memory computers ?
Organization: Silicon Graphics
Date: 29 May 1996 08:51:03 GMT
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <4oh35n$r8h@fido.asd.sgi.com>

john@elroy.jpl.nasa.gov (John Wright) wrote:
>In article <4ntu7d$o8g@fido.asd.sgi.com>,
>Steve Simonds  <simonds@sgi.com> wrote:
>>
>
>The last time I tried to use SGIMP my software kept pulling a bunch
>of end of buffer, out of data errors.  I modified the pvmgetarch
>command to return SGI5, even for our multiprocessor Onyx and the
>system worked correctly every time.  I believe that the problem was
>in the implementation of the message passing of shared memory and
>using SGI5 forced the messages through the TCP/IP or whatever port.
>I have seen others with this problem but no fix reported to date.
>Does anyone know if this problem has been solved?  I would not
>consider it solved if changes need to be made to the code depending
>on whether it runs on a shared memory system or a distributed system.
>

I have run numerious benchmarks using pvm3.3.8 in shared memory mode - SGIMP64.
I even compiled pvm using the -n32 compiler switch and ran this successfully.

I have not used pvm3.3.10 or 3.3.11 but I have heard that it works fine with
some of the bug fixes.  I think SGI feeds bug fixes back to the public domain
version of PVM.

I would suggest that you get the ARRAY 2.0 software from the locak SGI people.
It includes the SGI version of PVM plus other software.


