[T10] Direction for 17-042 proposal

Elliott, Robert (Persistent Memory) elliott at hpe.com
Thu Apr 13 07:27:59 PDT 2017

VPD bits are typically more static, only changing if some massive configuration change has occurred.  In this case, the changes that activation might cause have not happened yet.

Consider having the REQUEST SENSE command return the microcode activation sense data descriptor while deferred microcode is pending.

Robert Elliott, HPE Persistent Memory

From: t10-bounces at t10.org [mailto:t10-bounces at t10.org] On Behalf Of Gerry Houlder
Sent: Friday, April 7, 2017 3:52 PM
To: T10 Reflector <t10 at t10.org>
Subject: [T10] Direction for 17-042 proposal


This proposal adds a bit to a VPD page to indicate that a deferred download has completed but has not yet been activated. A VPD bit is appropriate because it is a bit that is only changeable by the SCSI target device (not an application client) and is a bit that is desired for discovery process. A slight disadvantage (for making this translatable for SATA) is that the drive must establish a Unit Attention when the INQUIRY data changes -- this is not easily translatable to SATA.

An alternate model for this feature is to use a mode page bit. The bit would be marked as not changeable by the host but the target device would set and clear the bit according to the presence of deferred microcode. There would not be a requirement to establish a Unit Attention for mode parameters changed because this Unit Attention is defined for changes made by an application client. This would eliminate the need to simulate a Unit Attention in the SATA version of this feature (which is not desired in T13).

I am asking for opinions ahead of the April T13 meeting and May T10 meeting as to whether a VPD bit or a mode page bit is the preferred direction for T10. At the moment, it seems that changing to a mode page bit is the better direction. Feel free to give feedback on the T10 reflector.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.t10.org/pipermail/t10/attachments/20170413/9ce212a9/attachment.html>

More information about the T10 mailing list