Command Deadline Timeout Change

Paul Suhler Paul.Suhler at hgst.com
Tue Aug 19 20:15:20 PDT 2014


Formatted message: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r1408192_f.htm">HTML-formatted message</a>

Thanks, Gerry.	That is indeed a severe limit.

;-)

Paul

From: Gerry Houlder [mailto:gerry.houlder at seagate.com]

Sent: Tuesday, August 19, 2014 11:11 AM

To: Paul Suhler

Cc: T10 E-mail Reflector (t10 at t10.org)

Subject: Re: Command Deadline Timeout Change

These commands are only valid for devices that implement type 2 Protection
Information. Those commands are illegal for type 0 and type 1 PI
configurations. This would severely limit the systems that could use the
feature. other than that, it is an interesting idea.

On Tue, Aug 19, 2014 at 12:18 PM, Paul Suhler
<Paul.Suhler at hgst.com> wrote:

During an internal discussion yesterday, it was suggested that I simplify the
proposal by having deadlines only apply to the READ(32) and WRITE(32)
commands, because those have some reserved bytes, one of which could be used
to replicate the ICC field in the READ / WRITE FPDMA QUEUED commands.  This
would have advantages:

1)	Simplified by no longer piggybacking the timeout value on the
priority feature.

2)	Allows a different timeout per command (rather than per priority)

At least one other company had expressed an interest in this functionality;
is it acceptable to have this only for the two commands?

Thanks,

Paul

Paul A. Suhler, PhD

Research Staff Member

HGST Research

paul.suhler at hgst.com

o: 949-476-1180 x275448

m: 949-241-6443

3001 Daimler St.

Santa Ana, CA 92705-5812

www.hgst.com<<a href="http://www.hgst.com/&gt">http://www.hgst.com/&gt;




More information about the T10 mailing list