In a message dated 99-11-28 19:24:52 EST, shimura at pure.cpdc.canon.co.jp
writes:
<<... [the table] specifies the value of "sbp_status" as eight (resources
unavailable) for negative residual case.
I thought that the "sbp_status" of "resources unavailable" means insufficient
SBP-2 resources in the "target" has prevented the target from processing the
ORB.>>
I don't think it necessarily has this connotation of target resources, but
that doesn't necessarily mean I disagree with your conclusion!
<<In T2I case, target stores a negative residual completion status because
the buffer provided by the "initiator" was too small to transfer the service
data.
In I2T case, a negative residual would be returned because of the lack of
target's resource, but lacking resource would not be an SBP-2 resource but a
receiving buffer for the service data outside SBP-2.
Thus, from the above observation, appropriate "sbp_status" seems to be zero
(no additional info. to report).>>
A zero value for sbp_status when residual is negative seems OK to me. I'm
making the change in PPDT_r08 with the usual caution that nothing is final in
that document until reviewed by the working group.
<<By the way, there seems to be a few editorial things in the same table. The
"sbp_status" values "4x" and "2x" in case of "resp" equal TRANSPORT_FAILURE
will be "8x(10rr xxxx): page table" and "4x(01rr xxxx)": data buffer)
respectively. Also, the "sbp_status" value "FF(unspecified)" will need to be
added in conformity with the SBP-2 definition.>>
Thank you, Shimura-san. Fixed in the forthcoming draft.
Regards,
Peter Johansson
Congruent Software, Inc.
98 Colorado Avenue
Berkeley, CA 94707
(510) 527-3926
(510) 527-3856 FAX
pjohansson at aol.com