[ofa-general] Re: [PATCH 0/2] Opensm support for external routing engines

[Ira Weiny]

On Thu, 18 Jun 2009 13:38:36 -0700
Al Chu <chu11 at llnl.gov> wrote:

> Hey Sasha,
> 
> On Thu, 2009-06-18 at 17:36 +0300, Sasha Khapyorsky wrote:
> > On 15:54 Thu 18 Jun     , Vincent Ficet wrote:
> > > The idea is to include non-open source routing algorithms into opensm on
> > > demand, which is permitted by the BSD license.
> > 
> > It is permitted, but I don't think that we as open source community
> > need to support such efforts.
> 
> I agree with this on principle, however, we already support a plugin
> system that allows users to develop proprietary perfmgr plugins.  So I
> feel that we already are down this path.

Also it should be noted that the "event plugin" (formerly perfmgr plugin) was
added to reduce the dependencies OpenSM would require when more advanced
monitoring was desired.  My plugin (and others) use Open Source packages such
as MySQL which could not have been added directly to OpenSM.  The intent was
never to facilitate proprietary extensions.  However, we never thought to
prevent them either.

Ira

> 
> Is the primary issue that this module "imports" new behavior into
> opensm?  I don't know the perfmgr plugin code that well, but my
> understanding is that you can "import new behavior" into opensm via the
> perfmgr plugin code too?  So what is really different??
> 
> Al
> 
> > I'm not a lawyer and cannot say for sure, but maybe in the case of dual
> > GPL/BSD license choice you will need anyway to relicense and to
> > redistribute OpenSM for doing this.
> > 
> > > It also enables to try a new algorithm on the fly without having to
> > > reinstall opensm, which can prove useful on a running cluster.
> > 
> > Even with your patches you will need to restart OpenSM, so it should not
> > be a big deal to build and run OpenSM even on a running cluster
> > especially if you are going to experiment with new routing engines.
> > Alternatively you can load routing tables from file.
> > 
> > > The event-based perfmgr plugin does not make it easy to integrate a new
> > > routing algorithm because the list of available routing algorithms is
> > > statically declared.
> > 
> > Eventually it is only link list with names and callbacks. It should be
> > easy to add new entry there.
> > 
> > > AFAIK, the latest Voltaire UFM also includes or will include proprietary
> > > routing algorithms (mesh + fat tree), so it could also be useful for
> > > Voltaire.
> > 
> > I don't know about this, but even if it is so. It is not our charter as
> > open source project to care about such practices and they (UFM) will need
> > to support their stuff by themselves.
> > 
> > Sasha
> > _______________________________________________
> > general mailing list
> > general at lists.openfabrics.org
> > http://  lists.openfabrics.org/cgi-bin/mailman/listinfo/general
> > 
> > To unsubscribe, please visit http://  openib.org/mailman/listinfo/openib-general
> > 
> -- 
> Albert Chu
> chu11 at llnl.gov
> Computer Scientist
> High Performance Systems Division
> Lawrence Livermore National Laboratory
> 
> _______________________________________________
> general mailing list
> general at lists.openfabrics.org
> http:// lists.openfabrics.org/cgi-bin/mailman/listinfo/general
> 
> To unsubscribe, please visit http:// openib.org/mailman/listinfo/openib-general
> 


-- 
Ira Weiny
Math Programmer/Computer Scientist
Lawrence Livermore National Lab
weiny2 at llnl.gov