[openib-general] Re: A Couple of User CM Questions

[Rimmer, Todd]

> On Mon, 2005-05-09 at 15:32, Hal Rosenstock wrote:
> > On Mon, 2005-05-09 at 14:39, Libor Michalek wrote: 
> > > On Mon, May 09, 2005 at 01:22:17PM -0400, Hal Rosenstock wrote:
> > > > Hi Libor,
> > > > 
> > > > I have a couple of questions pertaining to the user CM.
> > > > 
> > > > 1. It appears that device (and port) are not supported 
> either incoming
> > > > or outgoing. So if that is correct, I presume only the 
> first device
> > > > (first port) is the only one which can be expected to 
> work. Correct ?
> > > > If so, that should be added to README as a current limitation.
> > 
> > So I am presuming this is a current limitation.
> > 
> > > > 2. Any idea on how device would be supported ? Would 
> this just be with
> > > > an IB device number or a string like "mthca0" ?
> > > 
> > >   I was thinking that the destination GID in the 
> req_event path record
> > > would be sufficient to identify the local device/port, 
> but I had not
> > > taken a look at sidr_req_event. I think this is the 
> correct GID to use
> > > for creating the QP, but I have not taken a look at how 
> easy it is to
> > > turn a GID into a suitable device/portin user verbs. Thoughts?
> > 
> > and source GID (in the REQ primary path) on the active side ? 
> > 
> > That seems like it would work but wouldn't you need to walk the GID
> > tables on each port on each IB device until you find the 
> match ? Sounds
> > a little expensive but would only be done once per connection.
> > 
> > The advantage of GID is this would be common across all IB 
> devices. Not
> > sure what else could be other than a deterministic numbering of IB
> > devices locally which I believe does not currently exist.
> 
> ib_verbs does have a way to get_devices (ibv_get_devices) and specify
> expose a pointer to a struct ibv_device. That would be another
> alternative to using GIDs. That would make it similar to the 
> kernel CM.
> 
> -- Hal

I would recommend GIDs, no searching is necessary and a GID based approach would work for both user and kernel CM.  ibv_device would have challenges for user space listeners.

Todd Rimmer