Resending as below conversation didn't show up on the open-fcoe mailing list.
> -----Original Message----- > From: Parikh, Neerav > Sent: Friday, January 06, 2012 11:07 AM > To: James Smart > Cc: devel@open-fcoe.org > Subject: RE: [Open-FCoE] [PATCH 2/4] scsi_transport_fc: Getting FC Port > Speed in sync with FC-GS > > > > > -----Original Message----- > > From: James Smart [mailto:james.sm...@emulex.com] > > Sent: Thursday, January 05, 2012 7:06 PM > > To: Parikh, Neerav > > Cc: devel@open-fcoe.org > > Subject: Re: [Open-FCoE] [PATCH 2/4] scsi_transport_fc: Getting FC > Port > > Speed in sync with FC-GS > > > > > > > > On 1/5/2012 1:10 PM, Parikh, Neerav wrote: > > > The reason I swapped the values was because; as part of adding the > > FDMI > > > support in libfc when I was adding the Port Speed and > SupportedSpeed > > > attributes I wanted to make sure the FC Management service receives > > the > > > correct values as per the standard. > > > Now to do that one option was to modify the #defines for the 10G/4G > > speeds > > > and make them consistent with the standards or add code to convert > to > > > standard values based on the kernel values. I chose to do the prior > > than > > > later and not introduce another set of defines. > > > > Yes - you did the right thing. I was thinking it was just vs hbaapi > > (which I > > have little affection for if you can't tell), but as that morphed > into > > SM-HBA, > > which got carried into FDMI... > > > > > We all love HBAAPI :-) > > > > > > > On the removal of the 10G value from transport, it may be true for > FC > > HBA > > > that 10G is not there anywhere. And you're right that it is > currently > > being > > > used for reporting the current speed/supported_speed by the open- > > FCoE > > > stack when configured on top of 10Gbps NIC/CNA. > > > Do you think it is incorrect to use the 10G speed by FCoE capable > > HBA/CNAs? > > > > I would be a little hesitant to use it. I guessed you were setting it > > for > > fcoe, and it's a simple leap to it, but I'm not sure it's a good one. > > My > > concerns would be : The SM-HBA/FDMI standard certainly wasn't written > > with the > > idea of reporting ethernet speeds. What will happen when 40G and 100G > > enet > > come along and the standard doesn't cover them - what values would > you > > use ?; > > I don't like the confusion it can generate - with an admin on the > other > > side > > of the fabric, using FDMI and seeing 10G - do they know it's really > an > > FCoE > > port (maybe as there isn't any other 10G) or will it confuse them ? > > I'm also > > not sure whether reporting current speed at 10G is valid, as you > > usually have > > only a fraction of the 10G link bandwidth available. I could see an > > admin > > wondering why, if you are 10G, are they seeing only fraction of that > in > > sustained bandwidth. CNA use cases and IOV will carve your bandwidth > > even > > lower. So although it's a simple leap, it can lead to > misinformation. > > > > -- james > > > > > IMHO using the 10G usage is correct as that is the link speed for the > port > being used by FCoE and not the bandwidth. Hence, I guess all the FCoE > HBA/CNA drivers seem to be using that and not just open-FCoE. > I do see some ongoing work on T11 where the 40 GB/s is already part of > SM-HBA-2/FC-GS6 and perhaps it is only time when 100 G will get > included. > > While I agree there is no way today for an Administrator to identify > whether > the connection is over native FC or FCoE (apart from the implied speed > 10G) > but I guess that is a question more on the standards/specs and how they > want > the distinction (or NOT) to be reported. > > Thanks, > Neerav > > > > > > > Thanks, > > > Neerav _______________________________________________ devel mailing list devel@open-fcoe.org https://lists.open-fcoe.org/mailman/listinfo/devel
