Complete Seagate Comments on SSC-2 r 3

Paul.A.Suhler at seagate.com Paul.A.Suhler at seagate.com
Tue Jul 10 17:04:42 PDT 2001


* From the T10 Reflector (t10 at t10.org), posted by:
* Paul.A.Suhler at seagate.com
*
Folks,

Here is the complete set of our Technical comments on SSC2r3.  It
incorporates comments posted to the reflector on June 26th.  I've deleted
some for which Dave Peterson already has an action item.

Since some companies will not be represented at the 16 July working group,
I'm posting these comments to the reflector, in addition to sending them to
Dave Peterson.  I'd appreciate feedback from anyone who can't attend.  I'll
see that we discuss your comments in the meeting.

Cheers,

Paul Suhler
Seagate Removable Storage Solutions

-=-=-=-

E/T Clause      Page  Comment

T   5.3.1       34    Try to align eight-byte field in ERASE (16) the same
as in other 16-byte commands
                      Fix:  Start LOGICAL BLOCK ADDRESS field in byte 3.
Make byte 11 Reserved.

T   5.3.2       35    Try to align eight-byte field in LOCATE (16) the same
as in other 16-byte commands
                      Fix:  Start LOGICAL BLOCK ADDRESS field in byte 3.
Make byte 11 Reserved.

T   5.4.14      71    Try to align eight-byte field in SPACE (16) the same
as in other 16-byte commands
                      Fix:  Start COUNT field in byte 3.  Make byte 11
Reserved.

T   5.5.3.2     88    Is SWP cleared when medium is unloaded? Is SWP
redundant with ASOCWP?
                      Fix:  TBD in working group

T   5.5.3.2     89    Why does setting SWP, etc. with no medium report
MANUAL INTERVENTION REQUIRED ?
                      Fix:  How about LOGICAL UNIT NOT READY, MEDIUM NOT
PRESENT ?

T   01-146r0          E0:E0 Implicit command enabled
                      Fix:  Also require ((BOT= False) or (BAML=1)) ?  What
ASC/ASCQ?

T   01-146r0          F0:E0 Explicit command enabled will end up in E0 or
E1 or E2
                      Fix:  Split into three transitions, F0:E0, F0:E1, and
F0:E2

-=-=-=-

Additional Comments added July 10th:

T   5.2.9.2     24    SWP field values in the Control and Device Config
mode pages are allowed to be different.  This can lead to confusion.
Changing one should change the other.
                      Fix:  Change fourth and fifth sentences to:  "If both
methods are implemented, a change to the bit in either page shall change
the bit in both pages."  If this change is not acceptable, then let
independence be at the vendor's option.

T   5.3         32++  Do we need the IMMED bit in the explicit address
model commands?

T   5.3.2       36    BT and DEST_TYPE can have conflicting values, e.g,. 0
and 01b, respectively.
                      Fix:  Unified field as in earlier revisions?  TBD in
working group.

T   5.3.3       37    READ(16) and other data-in commands will auto-locate
if the LBA in the CDB is not the current LBA.  Are we SURE that we want it
this way?  This wording has been present since Rob Basham's 00-318r1.
                      Fix:  Don't auto-locate; report sequential
positioning error.

T   5.4.6       53    Third paragraph says min and max fixed block size are
multiples of four.  What about fixed block sizes in between?  This wording
allows odd sizes.

-=-=-=-


*
* 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