FCP Target ABTS Requirement
Kurt Chan
kc at core.rose.hp.com
Fri Jun 24 08:05:52 PDT 1994
| While there exist certain restricted environments which can tolerate
| targets which do not terminate ambiguous exchanges, there also exist
| environments in which data integrity problems result when targets do
| not terminate them. Therefore, if FCP does not continue to require all
| targets to terminate ambiguous exchanges, then initiators will have to
| vary their error recovery procedures depending on the exact FCS
| configuration, fabric parameters, and N_port parameters. This causes
| extensive transport-unique programming, which is one thing the
| layering of SCSI-3 is trying to avoid.
|
| Giles Frazier
| IBM Austin
| gfrazier at ausvm6.vnet.ibm.com
The transport-specific programming need not exist in the Host - the
absence of Target-Initiated ABTS is meant to be a Target optimization.
Consider the following rules:
- All hosts must be able to initiate ABTS
- All hosts must be able to accept ABTS
- Not all targets need to initiate ABTS
This allows host SW to remain the same, regardless of topology or
transport. I'm assuming here that transport-specific programming is
acceptable for Targets (i.e., a SCSI-3 SSA Disk isn't expected to work
over FC without modification!).
Again, I'm not sure why the above rules need to be in FCP versus an
interoperability document. Making FCP "architecturally complete" by
requiring all FCP implementations to have the most complicated error
recovery for any topology doesn't help the low cost FC-AL markets. It
is up to individual implementations to add complexity where they need
to address both simple and general solutions. I would prefer to
see the architecture allow some leeway for innovation by not requiring
features which aren't necessary for low cost markets.
Regards,
Kurt Chan Hewlett-Packard System Interconnect Lab
kc at core.rose.hp.com Voice: 916-785-5621 Fax: 916-785-2875
8000 Foothills Blvd MS R5NF Roseville, CA 95747
More information about the T10
mailing list