Accredited Standards Committee*
InterNational
Committee for Information Technology Standards (INCITS)
Doc.
No.: T13/e05156r0
Date: October
26, 2005 3:19 PM
Reply
to: Mark Overby
To: T13 Membership
From: Mark
Overby
Subject: T13 Plenary Meeting (Boulder, CO) --
August 23-26, 2005
Agenda
1. Opening Remarks
2.
Approval of Agenda
3.
Attendance and Membership
3.1 Membership
3.2 Fees
3.3 Attendance
4. Approval of the Previous
Minutes
5. Document Distribution
6. Review of Previous Action
Items
7. Old Business
7.1 ATA/ATAPI-7
7.2 EDD-3
7.3 HBA Standard
7.4 ATA/ATAPI-8 Projects
7.5 Method to Disable Data
Transfer with Error Project
7.6 HBA-2 Project
7.7 A/V Technical Report
Project
7.8 T13 Reflector and
Website Setup
7.9 INCITS Membership and
Website Policy
7.10 International
Participation by T13
8. New Business
8.1 Items from the
Reflector
9. International Representative
Reports
10. Liaison Reports
10.1 T10
10.2 T11
10.3 Trusted Computing
Group
10.4 SFF Committee
10.5 IDEMA
10.6 MMCA
10.7 CFA
10.8 SATA-IO
10.9 CE-ATA
11. Call for Patents
12. Review of Action Items
13. Meeting Schedule
13.1 Meeting Length
13.2 Authorization of
Ad-Hoc Working Groups
13.3 Long Term Meeting
Schedule
14. Adjournment
Results of
Meeting
1. Opening Remarks
Dan
Colegrove, the T13 Chair, called the meeting to order at 0900 Tuesday, August
23rd, 2005. Dan Colegrove thanked Mark Evans of Maxtor who hosted this meeting.
As
is customary, the people in attendance introduced themselves, and a copy of the
attendance list was circulated. Copies of the agenda were made available to
those in attendance.
The
Chair briefly reviewed the meeting rules: T13 is a Technical Committee of
INCITS and operates under INCITS rules. Anyone with interest in the work of the
committee is free to attend and take part in discussions. However, motions and
votes are limited to member organizations, one vote per member organization.
The
Chair also noted that T13 operates under the INCITS antitrust guidelines. There
should never be discussion of the following topics at any INCITS or an INCITS
subgroup meeting:
“Any company's prices or pricing
policies;
“Specific R&D, sales and
marketing plans;
“Any company's confidential product,
product development or production strategies;
“Whether certain suppliers or
customers will be served;
“Prices paid to input sources; or
“Complaints about individual firms
or other actions that might tend to hinder a competitor in any market.
Policies
and Procedures and antitrust guidelines for INCITS are available at the web
site at www.incits.org. T13 operates under these policies and procedures
augmented by the T13 procedures at the T13 web site at www.t13.org.
2. Approval of Agenda
Numerous corrections or
modifications were made to the agenda and the agenda was republished with the
corrections and additions. Mark Overby moved, Curtis Steven seconded, that the
agenda be approved as revised. The motion passed by unanimous consent.
3. Attendance and Membership
The
Chair reviewed the rules for membership and attendance:
3.1 Membership
An individual from an
organization new to T13 must attend at least one T13 plenary meeting before
requesting voting membership. After that meeting, the representative may
request membership from the T13 chair stating expertise and material interest
in the scope of the committee and pay the required fees to INCITS. The time
when membership becomes effective has just changed. Membership now becomes
effective at the start of the next meeting the organization attends assuming
the organization attends one of the next two meetings.
As was noted in the June
Attendance Jeopardy report, Net Cell and Pacific Digital lost membership for
lack of attendance at the end of the June Plenary. No members lost membership for
lack of attendance at this meeting.
Allegent and CPR tools, having
paid their fees and attending a prior meeting, were eligible to vote at the
beginning of this meeting.
3.2 Fees
Current INCITS
memberships fees for Principal or additional Alternates are 800.00 USD. The
First Alternate is free with a Principal membership.
Mailings
of committee documents are available to any interested party after each plenary
meeting. The cost for mailings for one year is 300.00 USD for mailing in the
USA and 400.00 USD for mailing to a location outside of the USA.
Persons
with questions or requests regarding membership fees and/or subscriptions for
mailings should contact Parthenia Purnell, ITI Standards Operations, at
ppurnell@itic.org.
3.3 Attendance
In addition to payment of fees,
there are also minimum attendance and participation requirements to maintain
voting membership status on T13. A representative from a member organization
must attend two of every three plenary meetings and vote on at least every
other letter ballot. If a member organization fails to have a representative in
attendance at two of the last three T13 plenary meetings or vote on every other
letter ballot, they are warned that failing to attend the next T13 meeting or
to vote on the next letter ballot will result in their voting membership being
terminated.
Attendance
at a working group meeting does not affect voting membership status in T13.
The
T13 voting membership during the meeting was 20 organizations.
The
number of eligible voting members required for establishing a quorum was 7. 15
Member organizations were present at the start of the meeting, thus a quorum
was present.
The
following individuals attended:
Candy
Alvarez CPR Tools
Tony
Alvarez CPR Tools
Kenneth
Hirata Emulex
Eric
Lanning Finisar
Dan
Colegrove Hitachi GST
Martin
Hassner Hitachi GST
Steve
Livacarri IBM
Wayne
Bellamy HP*
Robert
Strong Intel
Joe
Breher Lingua Data*
Tim
Thompson LSI Logic
Mark
Evans Maxtor
Frank
Shu Microsoft
Mark
Overby NVIDIA
Karen
Zelenko Phoenix Technologies
Marc
Noblitt Seagate
Jim
Hatfield Seagate
Aaron
Wilson ST Microelectronics
Yutaka
Arakawa Toshiba
Mladen
Luksic Western Digital
Curtis
Steven Western Digital
Rob
Condie Western Digital
*Non
Members
4. Approval of the minutes from the previous T13
plenary meeting (e05126r0)
Jim Hatfield noted that Chandra
from Phoenix Technologies was present and not Karen Zelenko. Dan Colegrove
noted that the motion to approve Mark Overby as the IR for T13 was out of order
due to INCITS procedure and that the new procedure would be covered later in
the agenda for this meeting. Jim Hatfield, Curtis Stevens seconding, that the
minutes be approved as amended. The motion passed with unanimous consent.
5. Document Distribution
The agenda was distributed by
hard copy.
6. Review of Previous Action Items
1. Dan Colegrove to add the obsolescence of clause 14.4.8 of ATA/ATAPI-7 to
the current errata document. [Complete]
2. Dan Colegrove to forward HBA errata to INCITS. [Complete]
3. Dan Colegrove to post the letter from the chairs of T10, T11, and T13 to
the website about INCITS membership policy. [Complete]
4. Jim Hatfield to write an informative annex on the use of the security
freeze lock and why it is important for ATA-8 ACS. [Carryover]
5. Mark Overby to begin process with CE-ATA to establish a liaison with T13.
[Complete]
6. Marc Noblitt to begin process with MMCA about establishing a liaison with
T13. [Marc not available to do so. Curtis agreed to try]
7. Dan Colegrove will query CFA if it uses subcommand codes for SET
FEATURES. [Carryover]
8. Dan Colegrove to check with INCITS about the anti-trust implications for
the SATA-IO proposal. [Work ongoing - Continue to Carryover]
9. Mark Evans will research if a separate letter is needed from Maxtor for a
Quantum patent related to UDMA that was released for ATA-5 and subsequent
project. [Carryover]
10. John Masiewicz to create an item for the ATA-7 errata to incorporate the
BIST retransmission clarification. [Carryover]
11. Dan Colegrove to check with INCITS legal about legal agreement
possibilities with SATA-IO. [Work ongoing. Covered during SATA-IO discussion]
7. Old Business
7.1 ATA/ATAPI-7: 1532D
7.1.1
INCITS Status
INCITS has
published the document as an official standard.
7.1.2
ATA/ATAPI-7
Errata: e05108r3 [Colegrove]
Dan Colegrove
reviewed the current list of ATA/ATAPI-7 errata. No changes were made to the
errata list and the list is being left open for errata that were mentioned
during this meeting.
7.2 EDD-3: 1572D
7.2.1
INCITS Status
[Stevens]
Curtis Steven
stated that SCT has been published as INCITS TR-38-2005.
7.3 HBA Standard: D1510
7.3.1
Errata to HBA
Standard (e05114r0) [Overby]
Dan Colegrove
stated that INCITS was concerned that this was a technical change not an error.
Dan clarified to INCITS that the bit was always supposed to be read only and
was marked read-write in error. Dan will monitor the current status with
INCITS.
7.4 ATA/ATAPI-8 Projects
7.4.1
ATA-8 Schedule
Review
Dan Colegrove
reviewed the current schedule for the various ATA-8 projects.
|
Table 1: ATA-8
Project Schedule |
||||
|
Project |
Proposal Cut Off |
Stabilization |
Letter Ballot |
Forward to INCITS |
|
Architecture |
April 2005 |
August 2005 |
October 2005 |
April 2006 |
|
Commands |
June 2005 |
October 2005 |
December 2005 |
April 2006 |
|
Serial |
June 2005 |
October 2005 |
December 2005 |
April 2006 |
|
Parallel |
February 2005 |
June 2005 |
August 2005 |
April 2006 |
No changes were made.
7.4.2
AT Attachment
8 - Architecture Model (ATA8-AAM)
7.4.2.1 Architecture Model and Diagram [Evans]
Mark Evans conducted line
by line review of revision 0b of the AAM document. Mark agreed to publish the
changes in revision 1 of the project.
7.4.3
AT Attachment
8 - Command Set (ATA8-ACS)
7.4.3.1 Download
Microcode Enhancements (e04132r0) [Livaccari]
Steve Livaccari reviewed
his proposal. Curtis Stevens and Dan Colegrove both noted that for the
segmented download they want more limitations on how the segmented download
works. They both noted that they wanted to see the download come in sequential
chunks. Curtis additionally noted that he does not want to see offsets. Mark
Overby pointed out that with those rules on the chunked download it is
impossible to do a translation of SCSI WRITE BUFFER mode 07h and this makes it
a problem. Mark Evans noted that it would be good to have specifics on what to
do on reset and intervening commands. Steve Livacarri proposed a straw poll on
if the restrictions on sequential access should be added or if the flexibility
in offsets and chunk size should be allowed. 5 people voted for the
restrictions and 5 people voted for the flexibility option.
On Wednesday Steve
presented revision 1 of his proposal. A lengthy discussion ensued about if
statements needed to be in this proposal regarding command completion,
verification of the firmware, and resets. Steve updated the document to
indicate that the firmware shall be verified and applied. Rob Strong pointed
out that the restriction on the host to not send a command during this sequence
was still in place. Mark Overby asked that the behavior be the same as queuing
where receiving a non-queued command causes the queue to be aborted. Steve
agreed to make that change. Dan and Curtis both requested that buffer offset
checking not be required. A compromise was reached in that the device “may”
abort the command if the offsets overlap or are not contiguous.
Steve Livacarri moved, Greg
Elkins seconded, that e05132r1 be incorporated into ATA8-ACS. Curtis Stevens
called the question. The motion to call the question passed with 13 ayes, 1
nay, and 0 abstentions. The main motion passed with 7 ayes, 4 nays, and 4
abstentions. Jim Hatfield and Rob Strong stated he voted no because much new content
was added and wanted more time to review. Mark Overby voted no because of the
restriction in sequence sending that prevents mode 07h (SCSI) from being emulated.
7.4.3.2 Reserving
OpCode and ID Data for NV Cache (e05151r0) [Shu]
Frank Shu presented
e05151r0. This proposal requests a reservation of specific op-codes and for
IDENTIFY data for their proposal for NV cache.
Frank noted that because he believed that the NV cache proposal would
not be completed in time for ATA-8 he would like to see op codes and identify
device bits set aside for the NV cache. Curtis Stevens stated that he had
concerns that the opcodes would be reserved for an activity going on with a few
proprietary implementations and companies. Frank countered that he was
concerned that the opposition to this seemed to be putting a roadblock to
Microsoft doing the right thing. Jim Hatfield and Dan Colegrove stated that they
appreciated the fact that Microsoft is trying to do the right thing. Mark
Overby suggested that we add them and if no one is going to adopt the
technology or if the technology was rejected then it could be removed before
final or as part of letter ballot comments. Jim noted that he was concerned
over the reservation of 4 features codes, since these are in short supply.
Curtis stated that he work
with Frank to get the necessary resources in ATA-8 so that Frank could modify
the proposal.
Frank asked that we defer
the proposal to later during the plenary week so that he can modify the
proposal to request what is needed based on the feedback for voting.
Discussion of the proposal
resumed on Wednesday as Frank presented a revised version of his proposal
(revision 1). Curtis Stevens noted that the proposal should just reserve the
opcodes and the details of each opcode should be in the separate NV cache
proposal (e05106). Curtis provided Frank with significant feedback on how to
modify the proposal to only reserve the resources needed instead of bringing in
pieces of the technical implementation. Frank made such changes while
presenting the proposal.
Frank Shu moved, Curtis
Stevens seconded, that e05151r1 be approved for inclusion into ATA8-ACS. The
motion passed with 14 ayes, 1 nay, and 0 abstentions. Rob Strong stated the
reason for his no vote was that he felt that NV cache proposal was not at a
level of maturity that was ready for inclusion into ATA-8.
7.4.3.3 NV
Ram Buffer Control (e05106r2) [Obr]
Frank Shu requested that
this be deferred to the next meeting.
7.4.3.4 Add SCT
to ATA8 Proposal (e05109r2) [Stevens]
Curtis Stevens presented
revision 2 of his proposal to integrate SCT into ATA-8. Curtis changed the
proposal to show how SCT would be broken up to incorporate into ATA-8. Curtis
also reviewed the BIST proposal that he has incorporated into this proposal as
part of incorporating all of SCT into ATA-8. Curtis requested that the
membership review the document and he intended to bring it to a proposal vote
at the next plenary meeting.
7.4.3.5 SCT
BIST Proposal [Stevens] (e04148r0) [Stevens]
Curtis Stevens stated that
he was making this proposal as part of his “Add SCT to ATA-8 proposal”
(e05109r0) and this item could be removed from future agendas.
7.4.3.6 Long
Physical Sector Alignment (e05119r0) [Barras]
Larry Barras was not
present, so this agenda item was deferred to the next meeting.
7.4.3.7 Long Physical Sector Informative Annex [Colegrove]
Curtis Stevens made an
announcement about bigsector.org. This is a website that has been setup by
companies involved in the move to larger physical sectors to provide
information about the technology as well as drives and hosts that are moving
this way. Curtis also reviewed his document e05122r3, also on the bigsector.org
website, that provides an overview of the impacts of the large physical sector
implementations. Dan Colegrove asked what should be put into an annex to
describe large physical sectors. Curtis stated that the information could be
made into an annex, however, he could not drive such an effort; but would be
willing to work with someone to turn this information into an annex. Curtis
suggested that he could provide the overview and implementation information and
start such a proposal. Dan assigned a document number (e05158) to Curtis to
provide a skeleton annex. He stated a target of December for inclusion into
ATA-8 with one or two conference calls during the intervening time.
Curtis began working on the
skeleton with input from attendees during the discussion of this agenda item.
Attendees provided feedback to Curtis for the initial draft of the document.
Dan Colegrove stated that
he believed that ATA8 should specify a specific offset. However, Marc Noblitt,
Mark Evans, and Curtis Stevens all stated that they believed this was not
possible, especially at this time because it is unclear what that alignment is
needed for the market.
Martin Hassner reviewed the
history of the large sector size and the reasoning behind moving from a
512-byte physical sector to a larger physical sector.
7.4.3.8 SMART
Error Log Sequences (e05141r1) [Chen]
Joseph Chen was not
present, so this agenda item was deferred to the next meeting.
7.4.3.9 Self
Throttling (e05136r0) [Livaccari]
Steve Livaccari reviewed
his proposal to have a method in the drive to throttle I/O at the drive level
to improve reliability at the cost of performance. The general concept is that
if the drive detected a condition that the frequency of I/O is reducing
reliability, it would slow the completion of I/O in order to maintain the
reliability. A lengthy discussion ensued about the need for this and for
alternate mechanisms in order to determine temperature and host intervention.
Frank Shu pointed out that alternative methods require host intervention.
Curtis Stevens replied that hosts are the best ones to know the environment and
be able to control the I/O rate, temperature, fans, and other such devices.
Steve agreed to turn a
revision of the proposal to incorporate feedback from this meeting and bring
the proposal to the next meeting.
7.4.3.10 Write
Uncorrectable (e02126r5) [Livaccari]
Steve Livaccari presented
the current revision of the proposal that included changes to make it more
reflect the ATA-8 style. Steve also noted that he included commands to include
large LBA drives. A discussion was continued on the need for READ RAW and the
effect that permuted ECC would have on this ability to read the ECC that goes
with the logical block. Steve agreed to remove READ RAW from proposal and to
split that into a second proposal. Mark Evans noted that the command
description should be cleaned up to just specify the behavior that is desired
without specifying an implementation. Jim Hatfield noted some missing language
to clarify the behavior of the features register. Steve agreed to make those
changes.
Steve Livacarri moved, Greg
Elkins seconded, that e02126r5, as modified, be approved for inclusion into
ATA-8 and for Curtis Stevens assign the TBD opcodes. Mark Evans moved, Mark
Overby seconded, that the motion be deferred to Wednesday so that the changes
could be seen in the document. The motion passed with unanimous consent.
On Wednesday, discussion of
the motion resumed. Steve displayed the changes that he had made in the
document based on the feedback from the Tuesday discussion. The motion passed
with 12 ayes, 0 nays, and 2 abstentions.
7.4.3.10.1 Read Raw (e05157r0) [Livaccari]
Steve Livacarri presented
his proposal, created from removing READ RAW from e02126r5. The proposal allows
for a host to read the contents of a sector without applying any ECC error.
Mark Evans, Marc Noblitt, and Curtis Stevens all stated that they believed that
this was not possible with their drives. Marc also stated that he had concerns
about exposing the raw error rate for the drives. Steve requested that drive
vendors go back and check if this could be supported or what level of raw data
could be returned. Curtis suggested that the data bytes could be returned after
correction even if they were known to still be bad. Steve will post e05157r1
for review for the next meeting.
7.4.3.11 Trusted
Computing Commands (e05139r2) [Hatfield]
Jim Hatfield presented the
latest revision of the trusted computing proposal for the commands that the
trusted computing group was requesting. The task file contents are defined by
this group and the data stream is defined by the trusted computing working
group. Jim noted that this proposal has a sibling for T10 to incorporate
similar commands into SPC. Jim reviewed the changes from the previous document
and why changes were made. He also stated that this will not be the final
revision. Jim stated that he would take these comments back for discussion as
well with the T10 proposal at the next T10 plenary. Curtis Stevens stated that
he would like to see this ready for integration by October 2005. Jim thought
that would be possible.
7.4.3.12 Proactive
Reporting and Correcting (e05142r1) [Shu]
Frank Shu presented the
latest revision of Microsoft position about the need for further and more
detailed error reporting. Microsoft’s position is that the drive should report
actions that the host can take to correct conditions that are affecting drive
reliability. There was significant discussion from the attendees about what
exactly could be reported and what the host could do. In addition, some
attendees queried about how, considering the latencies involved in the bus
could timely information be reported. As well, devices are not permitted to
initiate uncommanded operations in ATA environments. Frank agreed to review the
feedback and continue working on his proposal.
7.4.3.13 ATA
Feature Summary (e05144r0) [Chen]
Joseph Chen was not present
so this item was deferred to the next meeting.
7.4.3.14 DF Bit
Recovery (e05127r1)
[Stevens]
Curtis Stevens presented
the latest revision of the DF bit proposal. Curtis noted that the changes were
to clean up language to add clarity. Curtis Stevens moved, Mark Overby
seconded, that e05127r2 be incorporated into ATA-8. Steve Livacarri requested
some clarifying language that indicates that the device, once reset, can accept
some commands without asserting the DF bit. Curtis added such language and
amended the motion to reflect r2 be integrated instead of r1. The motion passed
with 14 ayes, 0 nays, and 0 abstentions.
7.4.3.15 Read Write Log Ext DMA (e05129r0) [Stevens]
Curtis Stevens presented a
proposal to add the capability to be able to read and write logs using DMA
protocols instead of PIO. Curtis Stevens moved, Tim Thompson seconded, that
e05129r0 be incorporated into ATA-8. Jim Hatfield requested a minor change to
call the command DMA EXT instead of EXT DMA. Curtis agreed to make that change
and amended to motion to incorporate e05129r1 instead of e05129r0. The motion
passed with 14 ayes, 1 nay, and 0 abstentions. Jim Hatfield noted he voted no
because he did not believe this was needed.
7.4.3.16 Report
Transport Standard (e05132r0 )
[Stevens]
Curtis Stevens reviewed his
proposal for having a mechanism to report the transport standard in the ATA-8.
Dan Colegrove noted that because of the uncertainty around the ATA-8 serial
transport document, it is unclear if this mechanism is needed at this point
because the necessary commands may appear in ATA-8 instead of referencing a
separate document. Curtis responded that this mechanism is analogous to
reporting compliance to a specific ATA revision that we have in preceding ATA
standards. Jim Hatfield provided a list of serial ATA transport specifications
from SATA-IO that would be useful to assign values to. Curtis updated the table
to include those specifications and assigned values to them. Curtis Stevens
moved, Greg Elkins seconded, that e05132r1 be incorporated into ATA-8. The
motion passed with 13 ayes, 0 nays, and 1 abstention.
7.4.3.17 Media
Serial Number Endienness (e05140r0)
[Hatfield]
Jim Hatfield reviewed
e05140r0 and asked for feedback. Dan Colegrove stated that he had not checked
with CFA yet on their preference for media number endien type. Dan agreed to
take an action item to check with CFA to make sure there was no issue. Jim
Hatfield moved, Mark Overby seconded, that e05140r0 be incorporated into
ATA8-ACS. The motion passed with 13 ayes, 0 nays, and 0 abstentions.
7.4.3.18 SATA Device BIST Control (e05149r0) [Chen] (* New
Business)
As Joseph Chen was not
present, this item was deferred to the next meeting.
7.4.3.19 ACS Editor Questions
7.4.3.19.1 Write Read Verify Incorporation (e04129r6)
[Stevens]
Curtis Stevens reported
that this has been integrated with no further questions.
7.4.3.19.2 HPA State Diagram Descriptions (e03124r5) [Stevens]
Curtis Stevens reported
that there exists a problem where if the 28-bit SET MAX ADDRESS is used on a
drive that uses a 48-bit command set it is not clear how you clear that HPA
since setting SET MAX ADDRESS to all F’s would still be a legal size for an HPA. Curtis proposed adding
language that would reflect that if SET MAX ADDRESS is used with an address
that equals SET NATIVE MAX ADDRESS, then it would clear the HPA. Karen Zelenko
asked if the drive should reject 28-bit SET MAX ADDRESS if it supports the
48-bit versions. Discussion ensued and the general consensus was that it was
premature to make such a restriction. Curtis added text to include text on how
to clear 28-bit HPA into section 7.50.2.2 of ATA8-ACS.
Karen Zelenko requested
that the group review the HPA state diagrams (e03124r5) with the new transition
text. Curtis requested that we do this offline and be ready at the next meeting
to consider for incorporation. Dan Colegrove stated that comments must be ready
by the next meeting.
7.4.3.19.3 Logical Sector Alignment (e05120r4) [Stevens]
Curtis Stevens reported
that this has been integrated and there are no questions.
7.4.3.19.4 SATA ICRC Reporting (e05133r0) [Stevens]
Curtis Stevens reported
that this has been integrated and there are no questions.
7.4.3.19.5 DCO Interaction with SMART [Stevens]
Will be covered during
draft review.
7.4.3.19.6 Extend Supported and Enabled Bits (e05130r0)
[Stevens]
Curtis Stevens reported
that this has been integrated and there are no questions.
7.4.3.19.7 Identify Device Requirements [Stevens]
Curtis Stevens reported
that this has been integrated and there are no questions.
7.4.3.20 Draft Review [Stevens]
Curtis Stevens conducted a
change bar review of the command set document and received input from the
attendees and agreed to publish a new revision with the input and changes.
7.4.4
AT
Attachment-8 - Parallel Transport (ATA8-APT)
7.4.4.1 Review of Draft
Mark Overby conducted a
change bar review of the parallel document and received input from the
attendees. Mark agreed to post revision 1 of the document with the changes that
were accepted during this meeting. Mark stated that he would post revision 1a
with change bars for further changes, and 1b with the change bars off.
7.4.5
AT
Attachment-8 - Serial Transport (ATA8-AST)
7.4.5.1 Review of Draft
Due to the ongoing
discussions with SATA-IO and the uncertainty about the need for the ATA8-AST
project, this item was deferred to the next meeting.
7.5 Method to Disable Data Transfer after Error
Detection TR (e05104r2) [Livaccari]
7.5.1
Status
Steve Livacarri stated that
revision 2 of the technical report should be final. Steve reviewed the draft
with the group and requested the chair to issue a letter ballot to approve the
draft for further processing. Dan agreed to issue the letter ballot.
7.6 HBA-2 Project Proposal [Overby]
Mark
Overby moved, Mark Evans seconding, that e05115r1 be approved as a new project
proposal and be forwarded to INCITS as such. The motion passed with 14 ayes, 0
nays, and 0 abstentions.
7.7 A/V Technical Report project [Colegrove]
Dan Colegrove asked that this
be deferred to the next meeting.
7.8 T13 reflector, FTP, and web site setup
There were no
issues to report.
7.9 INCITS Membership and Meeting Attendance Policy
Dan Colegrove posted
the letter that the T committee chairs sent to INCITS about the membership
policy and INCITS has not taken action yet.
7.10 International Standards Participation by T13
Dan Colegrove agreed to
research the correct procedure for requesting an IR and the committee could
take action at the next meeting.
8. New Business
8.1 Items from the Reflector
8.1.1
Stream
Questions [Stevens]
Curtis Stevens reviewed
questions he had sent to the reflector before the meeting. Curtis queried about
if there were 8 streams or 16 streams as there was confusion on configuring
streams if there was 1 bit for each direction or 1 bit for each stream period.
Dan Colegrove stated that he believed that there were just 8 streams period.
Dan and Curtis agreed that the language about the read / write bit in CONFIGURE
STREAM needed clarification. Dan stated his preference for clarification is to
state clearly that there is a total of 8 streams and that the read/write bit is
advisory and a hint to the device about the expected direction of the stream. Greg
Elkins stated that he would like to see this as a proposal and Mark Overby
concurred. Dan Colegrove agreed to bring the proposal forward.
Curtis stated that some drives
error when the CCTO occurs and other drives do not, if the flush bit is set.
Mark Evans stated that the flush explanation in the standard says that the
flush operation shall not complete until all data has been flushed (regardless
of the CCTL). Dan stated that he believes that the CCTL should take priority
over the flush operation. Mark believed that there was a conflict between the
shall statements. A further discussion ensued as attempts were made to fix the
language that over setting error outputs, when to report the error, and the
effect of other bits (e.g. WC) on this flush operation. Mark Evans agreed to
bring a proposal for ATA-8 to correct the behavior (e05154).
Curtis stated that the spec is
stating behavior for read continuous and read stream PIO that is not being
implemented. Curtis states that his belief that the SE bit is not used in PIO.
Curtis stated that he would like to obsolete the SE bit for PIO and would bring
a proposal to do so (e05159).
Word 87, bit 4 is supposed to
be set to 1 if a configure stream has been executed. Curtis stated that he
believes that this is supposed to be an indicator that the drive is in
streaming mode and no statement exists about how that bit gets cleared. Dan
believes that this bit exists to indicate to other applications that some
application is doing streaming and was unsure if it was necessary to go back
and clear the bit if there are no streams outstanding. Dan agreed to add some
clarifying language.
Curtis asked if the delayed LBA
log required for streaming. Dan stated that he believed that the delayed LBA
log was intended to be part of the streaming feature set. Dan agreed to
research the issue with his engineers and possibly propose some changes.
Curtis also noted that if an
ICRC error occurs, continuous reads or writes will not work because the drives
will stop transmitting data.
8.1.2
Read Look
Ahead [Chen]
The group revisited the issue
discussed at the last meeting about if the read ahead disable affects the
entire read cache. Curtis Stevens stated the he believed that the wording
should be changed. Curtis Stevens moved, Mark Evans, that the description of
the read look ahead disable include words to indicate that the entire read
cache may be disabled. The motion passed with unanimous consent.
8.1.3
Count Register
[Shastry]
Curtis Stevens acknowledged
that there was an error in the READ * EXT commands where 0 in count register
indicates 64k sectors to be transferred, not 256. Curtis had already addressed
this problem in the draft he was working on.
8.1.4
PIO “DRQ Data
Block” Definition [Krebs]
Mark Overby agreed
to review this discussion and respond to the reflector.
8.1.5
LBA parameter
in FIS [Shastry]
The committee discussed this
and felt that ATA-7 was clear that the D->H FIS contains the standard return
values from a command (PIO exceptions noted).
8.1.6
Questions on
text for SECURITY ERASE UNIT [Overby]
The committee resolved the
question that before SECURITY ERASE UNIT is sent, a password must be set. Mark Overby agreed to bring a proposal
(e05162) to clarify this language.
8.1.7
READ LOG EXT
[Stevens / Overby]
Curtis
Stevens and Mark Overby both identified a problem where ATAPI devices are
allowed to implement READ LOG, but not WRITE LOG in ATA-7. There was discussion
on why ATAPI devices were prohibited, originally, from implementing the GPL
feature set. There was acknowledgement that ATA-7 was unclear. Mark and Curtis
both reported that SATA ATAPI devices wanted this ability. The consensus was
that this should be fixed in ATA-8. Mark Overby agreed to bring a proposal to
allow for this behavior (e05161).
8.1.8
FLUSH CACHE
EXT and FLUSH CACHE behavior [Overby]
Mark Overby
discussed the issue that it was unclear what the behavior of FLUSH CACHE and
FLUSH CACHE EXT if the write cache was disabled. There was consensus that the
command shall complete without error if write cache is disabled (write cache
being disabled is not a valid reason to abort the command). Mark agreed to
bring a proposal to ATA-8 to clarify this (e05160).
8.1.9
Security
supported / enabled question [Stevens]
Curtis
Stevens stated that he believed that the text was unclear about when the
identify bits should get set for security being enabled versus security being
disabled. Mladen suggested that the name of the bit in IDENTIFY DEVICE be
changed to say that a password as set. Curtis Stevens suggested that the text
for the description could be modified to make that clear. Mark Overby agreed to
add this to his proposal to clarify when SECURITY ERASE UNIT can be sent.
9. International Standardization Reports
Mark Overby reported that Gary
Robinson stated that the work item for ATA-7 had been approved by ISO committee
SC25 and that work was beginning on ATA-7 volume 1.
10. Liaison Reports
10.1 T10 [Colegrove]
10.1.1
T10 Liaison
Report [Lohmeyer]
Dan Colegrove
presented the liaison report from T10 and SAS 1.1 is currently in comment
resolution. And SAS 2 is just beginning.
10.1.2
SCSI to ATA
Translation (SAT) Working Group Report [Sheffield]
The current revision of the
draft is revision 5 that will remain that way until the July meeting and is
targeting a September letter ballot.
10.1.2.1 Working group meeting before T13 plenary
co-located.
SAT was requesting time for
both the October and December meetings.
10.2 T11 [Colegrove]
Dan Colegrove
stated there were no activities that he was aware of impacting this committee.
10.3 Trusted Computing Group [Hatfield]
Jim Hatfield stated
that he believed the information was covered during the discussion of the
proposal.
10.4 SFF [Colegrove]
Dan Colegrove
stated that SFF is working on 30mm x 40mm form factor and 54mm x 71mm form
factors specifications (SFF-813x and SFF-814x).
10.5 IDEMA [Colegrove]
Dan Colegrove
reported that IDEMA CE group is continuing to meet and focusing on reliability
specifications.
10.6 MMCA [Colegrove]
No liaison report
was available. Curtis Stevens stated he might be able to provide a report at
the next meeting.
10.7 CFA [Hatfield]
Dan Colegrove
stated no meeting has been held since the last T13 meeting and there was
nothing to report. Jim Hatfield reported that the A/V team has not met recently
as well, but there has been a change in the lead of the A/V group.
10.8 Serial ATA IO [Dees]
10.8.1
Liaison Report
Brian Dees
was not present, so no liaison report was provided.
10.8.2
Copyright
Clearance for ATA8 Project
Dan Colegrove and Marc Noblitt
reviewed the current draft of a license agreement between SATA-IO and INCITS to
license command set material. Dan created a document to collect feedback from
the attendees about the license agreement. Dan stated that he would format that
feedback, post the document, and send the feedback to SATA-IO.
Feedback from the attendees was
to have a definitive time for review to be completed by SATA-IO so that the
letter ballot, public review, and publishing process could proceed without
worry of encumbrance. Additional questions arose around the derivative works
clause and further clarification was requested. A definition of the normative and
informative material was requested.
There was also a discussion
about having a rule instead of a license agreement. Marc stated that the board
of SATA-IO felt that to be insufficient.
Dan published the draft license
as document number e05152r0 and will publish the committees questions and
concerns in document e05153r0 and send to SATA-IO.
10.9 CE-ATA [Overby]
Mark Overby
reported that a proposal would be discussed at the CE-ATA digital working group
to incorporate the signature for CE-ATA into ATA-8. Mark expected that proposal
to be ready for the October meeting.
11. Call for Patents
The Chair
explained the patent policy. A document is available from ANSI, “Procedures for
the Development and Coordination of American National Standards”, at no charge.
This document is also on the INCITS web site at
www.incits.org/help/ansi_sdo.html. Section 1.2.11 of that document contains the
ANSI patent policy. Amy Marasco manages patent issues for ANSI and can be
contacted at amarasco@ansi.org or 212-642-4954.
The Chair then
called for patents and/or patent applications that may be necessary to comply
with the requirements in T13 standards, draft standards, technical reports, or
draft technical reports.
There was no
response to the call for patents.
12. Review of Action Items
1. Jim Hatfield to write an informative annex on the use of the security
freeze lock and why it is important for ATA-8 ACS. (carryover)
2. Curtis Stevens to inquire with MMCA about establishing a liaison with
T13.
3. Dan Colegrove will query CFA on if SET FEATURES subcommand codes are
used.
4. Dan Colegrove to check with INCITS about the anti-trust implications for
the SATA-IO proposal.
5. Mark Evans will research if a separate letter is needed from Maxtor for a
Quantum patent related to UDMA that was released for ATA-5 and subsequent
project.
6. John Masiewicz to create an item for the ATA-7 errata to incorporate the
BIST retransmission clarification.
7. Dan Colegrove to check with CFA if they have a preference for the
removable media serial number endien type.
8. Dan Colegrove to issue a letter ballot for the Method to Disable Data
Transfer After Error technical report for forwarding to INCITS for further
processing.
9. Dan Colegrove to ask INCITS legal about the need for a copyright release
to incorporate material from ATA-7 into HBA-2 that was part of SATA material.
10. Dan Colegrove to forward e05115r0 to INCITS as a new project.
13.1 Meeting Length
The meeting length
continues to be 3.5 days due to the extra material. Dan Colegrove noted that
for October, Tuesday will run from 0900 to 1800, Wednesday from 0800 to 1700,
Thursday from 0900 to 1700, and Friday from 0900 to 1200.
13.2 Authorization of Ad-Hoc Working Groups
The committee voted to approve
up to two teleconference calls for the purposes of discussing the NV cache
proposal by Nathan Obr or Frank Shu. Dan
Colegrove moved, Mark Overby seconded, that two teleconferences be authorized
to discuss the NV Cache proposal. The motion passed with unanimous consent.
13.3 Long-Term Meeting Schedule
|
Table 2: 2005
Dates |
|||
|
Date |
Location |
Host |
Contact |
|
October 18-21, 2005 |
Las Vegas, NV |
Fujitsu |
Sumit Puri |
|
December 13-16, 2005 |
Las Vegas, NV |
Hitachi GST |
Dan Colegrove |
|
Table 3: 2006
Dates |
|||
|
Date |
Location |
Host |
Contact |
|
February 21-24, 2006 |
Orange County, CA |
Western Digital |
Curtis Stevens |
|
April 18-21, 2006 |
Cupertino, CA |
Apple |
Larry Barras |
|
June 20-23, 2006 |
Seattle, WA |
NVIDIA |
Mark Overby |
|
August 22-25, 2006 |
Boulder, CO |
Maxtor |
Mark Evans |
|
October 24 - 27, 2006 |
Las Vegas, NV |
Fujitsu |
Sumit Puri |
|
December 12-15, 2006 |
Las Vegas, NV |
Hitachi Global |
Dan Colegrove |
|
Table 4: 2007
Dates |
|||
|
Date |
Location |
Host |
Contact |
|
February 20-23, 2007 |
Orange County, CA |
Western Digital |
Curtis Stevens |
|
April 24-27, 2007 |
|
|
|
|
June 19-22, 2007 |
|
|
|
|
August 21-24, 2007 |
|
|
|
|
October 23-26, 2007 |
|
|
|
|
December 11-14, 2007 |
|
|
|
14. Adjournment
The meeting
was adjourned at 1200, Friday August 26, 2005.