Determining application client support for CRN.

Dave Peterson dap at cisco.com
Tue Sep 25 11:26:50 PDT 2001


* From the T10 Reflector (t10 at t10.org), posted by:
* "Dave Peterson" <dap at cisco.com>
*
CRN is used on a logical unit basis not on a target/port basis.
CRN is also not applicable to parallel SCSI.
As such the EPDC bit was placed in the protocol specific logical unit
control page.

Dave

> -----Original Message-----
> From: santoshr at cup.hp.com [mailto:santoshr at cup.hp.com]
> Sent: Tuesday, September 25, 2001 12:52 PM
> To: Dave Peterson
> Cc: T10 Reflector
> Subject: Re: Determining application client support for CRN.
>
>
> Dave Peterson wrote:
>
> > CRN at the application client is optional, thus CRN support is not to be
> > "assumed" by the device server.
> > Application client determines support (or non-support) of CRN
> at the device
> > server via the EPDC bit in the logical unit control page. The
> application
> > enables precise delivery (i.e. CRN) by successfully setting the
> EPDC bit to
> > one.
> > The application client is then allowed to use the CRN field as
> specified in
> > FCP-2 clause 4.3
>
> Dave,
>
> If the "EPDC" bit determines the CRN support in the device
> server, why is this
> bit placed in a protocol specific mode page (LUN Control Mode page. a.k.a.
> Protocol Specific LUN Page) and not in a SCSI ULP mode page such as the
> "Control Mode Page" ?
>
> My initial thoughts were that the EPDC was only placed in a FCP-2
> specific mode
> page to verify the target FCP layer's (target SCSI LLP) ability
> to handle the
> CRN in its FCP_CMD IU and had nothing to do with the application
> client. (which
> belongs to the target SCSI ULP).
>
> Regards,
> Santosh
>
>
> > <* Is there any need for new SCSI transports to have a bit similar to
> > <     the EPDC bit of FCP-2, if they are guaranteed to support the CRN
> > <     field in their SCSI Command encapsulation IUs from their initial
> > <     revs ? I ask this in the context of iscsi. I also noticed that SRP
> > <     has chosen not to define the Protocol Specific LUN Page [which
> > <     would have contained something equivalent to the EPDC bit].
> >
> >
>

*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org




More information about the T10 mailing list