In a message dated 00-04-13 16:33:23 EDT, alan_berkema at hp.com writes:
<< A few of us were looking at p1394.3 and believe we noticed a couple of
minor
errors?>>
I assume that you're looking at Draft 1.1, February 10, 2000?
<<Page 63: Text descriptor leaf offset (3). Believe that this offset should
be 2 instead of 3.>>
A value of 3 is correct in figure F-1. A value of zero would refer to the
entry itself; a value of 3 skips the intervening Node_Capabilities and
instance directory entries and points you to the start of the descriptor leaf.
<<Page 65: Text descriptor offset (2). Believe that this offset should be 1
instead of 2>>
For the same reasons, 2 is the correct value in figure F-3.
<<Page Various: csr_offset Believe that this should be mar_offset for the
SBP-2 Management Agent Register.>>
Then you would have to explain what the name 'mar_offset" means! I don't
think there's anything wrong with the generic "csr_offset". The intent was
to emphasize that the entry was pointing to "a" CSR, not "some particular"
CSR. ANSI NCITS 325-1998 (SBP-2) itself does this; see figure 44 in 7.4.7
on page 45.
At 07:05 PM 5/8/00 +0900, Akihiro Shimura wrote:
<<Page 66: "keyword leaf offset(2)" in Figure F-4 should be "(3)"
Page 67: better to denote "keyword leaf offset" with "(4)"
Page 69: "keyword leaf offset(5)" in Figure F-7 should be "(6)">>
I agree and have made the changes in a not-yet-published PPDT_r12.
I PROPOSE to delay publication until the ballot group has been formed. At
the time Greg LeClair initiates the ballot, I will post it in order that it
may be balloted. Of course, if we uncover anything significant between now
and then we may have to post and vote on an interim version.
Regards,
Peter Johansson
Congruent Software, Inc.
98 Colorado Avenue
Berkeley, CA 94707
(510) 527-3926
(510) 527-3856 FAX
PJohansson at ACM.org