[ofa-general] Re: Re: [PATCH RFC] sharing userspace IB objects

Gleb Natapov glebn at voltaire.com
Mon Jul 2 06:03:49 PDT 2007


On Mon, Jul 02, 2007 at 04:00:57PM +0300, Michael S. Tsirkin wrote:
> > >>>RD is not supported in hardware today. Implementing RD is extremely 
> > >>>complicated. To solve the scalability issues on MPI like applications
> > >>>we believe that SRC and SSQ are the right solutions. It is much simpler
> > >>>for implementation by both software and hardware. By MPI-like I refer
> > >>>to applications that have some level of trust between two processes of
> > >>>the
> > >>>same application. RD also has some performance issues as it only 
> > >>>supports one message in the air. Those performance issues are solved
> > >>>by design in SRC/SSQ.
> > >>>
> > >>>      
> > >>Didn't know about RD limitation. Is this shortcomings of IB spec or
> > >>general limitation of reliable datagram? RD looks much nice to me then 
> > >>SRC/SSQ.
> > >>    
> > >
> > >I think Dror is referring to number of messages in flight per EEC and
> > >number of messages in flight per QP being limited to 1 per IBA spec.
> > >Number of messages enqueued per EEC/QP is implementation dependent.
> > >
> > >-- Hal
> > >  
> > Correct. The number of messages in flight per EEC is 1 per IB spec.
> > The fact that IB requires SQ WQEs to complete in order, even if their 
> > destination is different EECs, makes it pretty challenging to have an 
> > implementation that can really process more than one message 
> > simultaneously per QP.
> 
> Hmm, I guess this requirement could easily be relaxed - in a way
> similiar to what was done for SRQ - without breaking applications.
Especially as there are no applications that use RD because there is not
HCA that support it.

--
			Gleb.



More information about the general mailing list