[T10] Another depop quirk

Joe Breher Joe.Breher at wdc.com
Thu Mar 30 21:10:46 PDT 2017


I have submitted proposals for an errata to repurposing depopulation to both T10 and T13:

Title: Repurposing Depopulation errata - STARTING ELEMENT
Description: If no element matching criteria, STARTING ELEMENT specifies a lower bound

T13 PDF Document File Number: f17116r0: http://www.t13.org/documents/UploadedDocuments//docs2017/f17116r0-Repurposing_Depopulation_errata_-_STARTING_ELEMENT.pdf

T10/17-067r0: http://www.t10.org/cgi-bin/ac.pl?t=d&f=17-067r0.pdf


Joe Breher
Storage Architecture Technologist
Standards Setting Organization
San Jose Research Center
Western Digital
+1 (478) 2-Breher
+1 (478) 227-3437


[cid:3888D38D-876F-48C4-94D6-BF5B836DAD27 at gateway.innflux.com]






On Mar 24, 2017, at 9:08 AM, Mark Carlson <Mark.Carlson at taec.toshiba.com<mailto:Mark.Carlson at taec.toshiba.com>> wrote:

Yes, I really don't like the idea of probing around for a valid starting element.

-- mark

On Mar 24, 2017, at 8:33 AM, Joe Breher <Joe.Breher at wdc.com<mailto:Joe.Breher at wdc.com>> wrote:

Gentlemen -

Early implementations have uncovered another wrinkle that requires fixing in Repurposing Depopulation.

>From sbc4r13:

A storage element is a type of physical element. Physical elements are associated with a unique element
identifier that is assigned by the device server. The element identifier shall be non-zero. The association of
element identifiers to physical elements shall persist through all events (e.g., across all resets) except
Microcode Change (see SAM-6). The association of element identifiers to physical elements may persist
through Microcode Change events.

As specified above, the ELEMENT IDENTIFIER is non-zero for all physical elements in the device. The initial proposal included a restriction that the element space be contiguous, but this was removed in committee deliberations.

As the ELEMENT IDENTIFIERS are at the discretion of the device, there is no a priori mechanism for the host to learn a ‘safe’ STARTING ELEMENT to specify for the GET PHYSICAL ELEMENT STATUS command. This makes the following requirement difficult:

The STARTING ELEMENT field specifies the element identifier of the first physical element addressed by this
command. If the specified starting element is invalid, then the device server shall terminate the command with
CHECK CONDITION status with the sense key set to ILLEGAL REQUEST and the additional sense code set
to INVALID FIELD IN CDB.

I think the proper way to fix the standard is to add a requirement that a starting element of zero shall be interpreted by the device as starting at the first valid element identifier (since the requirement to have all identifiers contiguous had been removed in committee deliberations as per above).

If all are in agreement, I will draft proposals for both T10 and T13 to allow a STARTING ELEMENT of zero to be interpreted as ‘start with the first valid element’.

Agreed?

Joe Breher
Storage Architecture Technologist
Standards Setting Organization
San Jose Research Center
Western Digital
+1 (478) 2-Breher
+1 (478) 227-3437


<WDC_Logo_email.png>






Western Digital Corporation (and its subsidiaries) E-mail Confidentiality Notice & Disclaimer:
This e-mail and any files transmitted with it may contain confidential or legally privileged information of WDC and/or its affiliates, and are intended solely for the use of the individual or entity to which they are addressed. If you are not the intended recipient, any disclosure, copying, distribution or any action taken or omitted to be taken in reliance on it, is prohibited. If you have received this e-mail in error, please notify the sender immediately and delete the e-mail in its entirety from your system.
<WDC_Logo_email.png>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.t10.org/pipermail/t10/attachments/20170331/63db9161/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: WDC_Logo_email.png
Type: image/png
Size: 6257 bytes
Desc: WDC_Logo_email.png
URL: <http://www.t10.org/pipermail/t10/attachments/20170331/63db9161/attachment.png>


More information about the T10 mailing list