Re: iSCSI: synch and steering comments

[Matt Wakeley <matt_wakeley@agilent.com>]

Doug,

You don't have to play your broken record about modifying TCP.

What I stated is correct.  Just because it could require a "change" to TCP is
not relevant.

-Matt

Douglas Otis wrote:
> 
> Matt,
> 
> If you intend to use TCP, there is no method yet to reliably obtain this
> type of header alignment.  Unless you are to modify TCP to become a record
> based protocol, this type of behavior is a modification of TCP and beyond
> the IPS charter.  If you wish this type of behavior, you should consider
> RFC-2960 instead.
> 
> Doug
> 
> > > -I am somewhat confused about the following statement in
> > section 1.2.8.2:
> > >     "The Synch and Steering Layer is required to add to every sent data
> > >     item (IP packet, TCP packet or some other superstructure) enough
> > >     information to enable the receiver to steer it to a memory location
> > >     independent of any other piece. "
> > >
> > >  Clearly from the way I understood the markers in Appendix.C, it doesn't
> > >  comply with this requirement.  A more generic statement would be:
> > >     "The Synch and Steering Layer is required to add adequate
> > >     information to the data stream to enable the receiver to quickly
> > >     steer the stream to its final memory location, even in the face of
> > >     discontiguities in the stream. "
> > >
> > > +++ Markers are but one example and have only the Synch component. The
> > > statement refers to a full steering (RDMA) scheme +++
> >
> > Julian,
> >
> > No "full steering (RDMA) scheme" is required. A Sync mechanism that that
> > implements framing such that the first item in each frame is an
> > iSCSI header
> > is all that is required to enable steering because the iSCSI
> > header contains
> > all the required steering information (task tag, data offset).
> >
> > -Matt
> >