Fw: 96-277r1 - Proposed Change in QErr for SPC-2

Edward A. Gardner gardner at acm.org
Thu Feb 27 13:50:43 PST 1997


* From the SCSI Reflector (scsi at symbios.com), posted by:
* "Edward A. Gardner" <gardner at acm.org>
*
> As I read Ed's note, he is effectively proposing a change in the Basic
queing
> model which replaces QErr = 01b with QErr = 00b as the behavior
**required** of

Just to make things clear, I am not proposing this as a change to Basic
Queuing.  Yes, I think it would be an improvement, and would vote for it if
someone else proposed it, but this is minor enough that I'm more interested
in consensus.  Peter Johannson (who I assume is representing the SBP-2 /
diskboys community) has spoken quite strongly in support of QErr=11 (abort
tasks from issuing initiator), that's good enough for me.

I feel that 99% (at least) of the benefits from Basic Queuing derive from
elimination of Ordered and Head-of-Queue queue tags, ACA and restricted
reordering.  Elimination of task management functions and QErr settings is
almost incidental.  Provided that the result is usable (including by
multi-initiator systems), I almost don't care what QErr setting(s) is(are)
required, although I do have preferences.  These measures of relative
importance are, of course, my own opinions, reasonable people may disagree.

I do wonder about the apparent controversy.  So far every company that
actually buys disk drives has said that they want QErr=00.  While they
don't account for all the low end drives sold, they do account for enough
that market forces will compel support what they want.  The diskboys crowd,
the one group that represents a new market that doesn't care about
backwards compatibility wants QErr=11.  There's not a single customer in
this discussion who's asking for QErr=01, the alleged status quo.  I think
as a matter of market reality any disk vendor that implements basic queuing
will have to implement multiple QErr settings.  As Gerry Houlder notes, the
question is whether any disk vendor will implement Basic Queuing at all
rather than Full Queuing.  Again, these are my opinions, feel free to
disagree, the market will decide.


Edward A. Gardner               gardner at acm.org
Ophidian Designs                719 593-8866 voice
1262 Hofstead Terrace           719 593-8989 fax
Colorado Springs, CO  80907

----------
> From: Charles Monia 237-6757 <monia at AM.SHRMSG.SHR.mts.dec.com>
> To: PJohansson at aol.com
<""PJohansson at aol.com""@SHDNS1.STARCH.MRGATE.SHRMSG.SHR.mts.dec.com>
> Cc: SCSI Reflector <scsi at Symbios.COM>
> Subject: M970228E Re: Fw: 96-277r1 - Proposed Change in QErr for SPC-2
> Date: Thursday, February 27, 1997 09:14
> 
> * From the SCSI Reflector (scsi at symbios.com), posted by:
> * Charles Monia 237-6757 <monia at AM.SHRMSG.SHR.mts.dec.com>
> *
> Hi Peter:
> 
> I'm not clear on what you disagree with.  In the following, you say:
> 
> | [making] QErr == 00b mandatory would REQUIRE that Basic queuing targets
> | permit initiators to set QErr to this value.
> 
> Are you against "QErr = 00b" or against requiring that "Basic" devices
provide
> a choice of behaviors that includes this option?
> 
> 
> Charles
> 
> 
> | * From the SCSI Reflector (scsi at symbios.com), posted by:
> | * PJohansson at aol.com
> | *
> | In a message dated 97-02-26 23:03:30 EST, gardner at acm.org (Edward A.
Gardner)
> | writes:
> | 
> | <<Given recent reflector traffic, today I would vote for QErr=00 being
> | mandatory, with support for other QErr values being optional.  Doing
nothing
> | has to be simpler than aborting any or all commands (after all, it's
the same
> | action as when a command completes normally).  Single host systems can
always
> | issue a TARGET RESET.  And it seems to be the preference of multi-host
> | systems.>>
> | 
> | I've extracted only this portion because I agree with the rest of what
Ed has
> | said.
> | 
> | This one paragraph stands in opposition, though. If I understand such
things
> | correctly, QErr == 00b mandatory would REQUIRE that Basic queuing
targets
> | permit initiators to set QErr to this value. This is not OK. SBP-2 was
> | designed to support and implement the Basic queuing model. SBP-2
devices
> | can't, don't or won't (pick your favorite verb) QErr == 00b. Basic
queuing
>  | was also intended to restrict optionality.
> | 
> | I am still in support of QErr == 01b (only the faulted initiator's task
set
> | is aborted) as a mandatory requirement for Basic queuing.
> | 
> | Incidentally, Ed, a very sincere "THANKS!" for doing all that research
into
> | our archives.
> | 
> | Regards,
> | 
> | Peter Johansson
> | 
> | Congruent Software, Inc.
> | 3998 Whittle Avenue
> | Oakland, CA  94602
> | 
> | (510) 531-5472
> | (510) 531-2942 FAX
> | 
> | pjohansson at aol.com
> | 
> | *
> | * For SCSI Reflector information, send a message with
> | * 'info scsi' (no quotes) in the message body to majordomo at symbios.com
> | 
> 
> ************************************************************ ***** *
> * Charles Monia, Storage Architecture Group                      *
> * Digital Equipment Co.  Internet: charles.monia at shr.mts.dec.com *
> * Tel: (508) 841-6757    X400: c=US; a=MCI; o=Digital; ou=SHR    *
> * Fax: (508) 841-6100                                            *
> ************************************************************ ***** *
> 
> *
> * For SCSI Reflector information, send a message with
> * 'info scsi' (no quotes) in the message body to majordomo at symbios.com
*
* For SCSI Reflector information, send a message with
* 'info scsi' (no quotes) in the message body to majordomo at symbios.com




More information about the T10 mailing list