SRP WG agenda request

Holt, Keith kholt at lsil.com
Mon Feb 19 09:42:39 PST 2001


* From the T10 Reflector (t10 at t10.org), posted by:
* "Holt, Keith" <kholt at lsil.com>
*
This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C09A9B.5A41E450
Content-Type: text/plain; charset="iso-8859-1"

Time permitting, there are a couple of items that I would like to
discuss at the Feb. 20-21 WG meeting. 

1. SRP_CMD overall data length for scatter/gather 
Our original discussions on the need for an overall data length focused
on the fact that this would be a duplicate of the transfer length
specified in the CDB.  For IB-FC SAN bridging, the IOC would be required
to sum up the total transfer lengths specified in each individual
scatter/gather element to obtain the total transfer length needed for
the FCP_DL.  The IOC must obtain the indirect scatter/gather list from
host memory via RDMA prior to this calculation.  This will make resource
management problematic for the IOC.

Proposal:  Add an overall data transfer length when an indirect list is
used. 

2. IU flow control 
I'd like to better understand the proposed usage of the SRP_RSP Request
Limit Delta field.  The fact that credits can be taken away from the
initiator seems to imply task set credit management capability, not just
simple flow control.  If credit reduction is indeed needed, then I would
like to require a positive response from the initiator.  Without this,
buffer management between the target protocol and transport units is
made more difficult.

Keith Holt 
LSI Logic 




------_=_NextPart_001_01C09A9B.5A41E450
Content-Type: text/html; charset="iso-8859-1"
Content-Transfer-Encoding: Quoted-Printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">

 SRP WG agenda request Time permitting, there are a couple of = items that I would like to discuss at the Feb. 20-21 WG meeting. 1. SRP_CMD overall data length for = scatter/gather 
Our original discussions on the need = for an overall data length focused on the fact that this would be a = duplicate of the transfer length specified in the CDB.  For IB-FC = SAN bridging, the IOC would be required to sum up the total transfer = lengths specified in each individual scatter/gather element to obtain = the total transfer length needed for the FCP_DL.  The IOC must = obtain the indirect scatter/gather list from host memory via RDMA prior = to this calculation.  This will make resource management = problematic for the IOC. Proposal:  Add an overall data = transfer length when an indirect list is used. 2. IU flow control 
I'd like to better understand the = proposed usage of the SRP_RSP Request Limit Delta field.  The fact = that credits can be taken away from the initiator seems to imply task = set credit management capability, not just simple flow control.  = If credit reduction is indeed needed, then I would like to require a = positive response from the initiator.  Without this, buffer = management between the target protocol and transport units is made more = difficult. Keith Holt 
LSI Logic 


------_=_NextPart_001_01C09A9B.5A41E450--




More information about the T10 mailing list