Posted SCSI / ATA Translation revision-4 (SAT-r4)

Sheffield, Robert L robert.l.sheffield at
Wed May 18 08:38:57 PDT 2005

* From the T10 Reflector (t10 at, posted by:
* "Sheffield, Robert L" <robert.l.sheffield at>

I have posted revision-4 of the SCSI / ATA Translation draft standard
(SAT-r4) to the T10 website:

It incorporates the following changes:

Incorporated SAT WG proposals approved in the May plenary as follows:
a) 05-109r0 REQUEST SENSE Correction Proposal [Overby] (with notes from
draft review 3 May 2005); and
b) 05-137r1 Standard INQUIRY - PRODUCT REVISION LEVEL [Bellamy].

Incorporated changes identified during review of the draft from SAT WG
meetings and teleconferences in Apirl and
May 2005 as follows:
a) deleted template text from clause 10;
b) added autosense to definitions;
c) made "should" and "shall" keyword definitions consistent with SAS;
d) clarified definition of the "unspecified" keyword;
e) simplified Table-1 to better show the use of the "implemented",
"emulated", and "unspecified" keywords;
f) deleted text and editor's note in 8.3.1 MODE SELECT (6) command
overview that discussed sharing
mode page header and block descriptor;
g) made significant changes to WRITE BUFFER download microcode modes -
supporting modes 5h, 6h, and
7h, and specifying that the SATL issues MICROCODE DOWNLOAD only on the
first WRITE BUFFER, and
continues the data transfer with subsequent WRITE BUFFER commands;
h) Removed the "Is NCQ enabled", "Is Queuing enabled", and "and SCSI OP
Codes" columns in the write
command summary table 58, modified the last two rows and table footnotes
to clarify the use of NCQ and
the non-packet device Overlapped feature set;
i) Clarified the transfer length for the WRITE (12) command, adding that
multiple ATA commands may be
sent to satisfy the specified transfer length;
j) Deleted the Format Device and Rigid Disk Geometry pages from the list
of commonly used SCSI mode
pages in Table 62 (because SBC-2 made them obsolete);
k) Moved the INQUIRY VPD pages subclause from the primary commands
clause (8) to the device parameters
clause (10), and modified the heading levels to match;
l) Added a glossary entry for ATA queued commands and used it in the
commands overview clause (7) in the
description of the requirement for a SATL to queue internally any
commands that are not ATA queued
m) Miscellaneous editorial changes; and
n) Added table-50 listing the mode page translations defined in the SAT

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

More information about the T10 mailing list