Reservation question

Elliott, Robert (Hou) Robert.Elliott at COMPAQ.com
Mon Aug 21 18:10:17 PDT 2000


* From the T10 Reflector (t10 at t10.org), posted by:
* "Elliott, Robert (Hou)" <Robert.Elliott at compaq.com>
*
> -----Original Message-----
> From: gop at us.ibm.com [mailto:gop at us.ibm.com]
> Sent: Monday, August 21, 2000 9:03 AM
> To: t10 at t10.org
> Subject: Reservation question
> 
> For your Monday morning SCSI question I have the following:
> 
> Is a RESERVATION (that's a normal reservation not a persistent one)
> preserved through a START/STOP command being issued with a stop?
> 

The reservation should definitely not be broken.  The RESERVE(10) rule 
does not list STOP as a reason to break a reservation:
	"This command shall request that the entire logical unit be
	reserved for the exclusive use of the initiator until the
	reservation is superseded by another valid RESERVE command
	from the same initiator or until released by a RELEASE command 
	from the same initiator that made the reservation, by a 
	TARGET RESET task management function performed by any 
	initiator, by a hard reset condition, or by a power on cycle."

I don't think it's necessary for the device to properly check 
reservations for every command while stopped, however.  If the 
device stores its reservation table on the media and needs to 
perform a media access to check reservations, then it should 
reject most commands with a NOT READY status.  Even a non-media 
access command subject to reservation conflict may result in
this status.  A START STOP UNIT with START=1 is always allowed, 
and will re-enable reservation checking.  

I expect most devices cache their reservation table in memory and
can check reservation conflicts even while stopped.  I don't think
SBC should mandate that.  The device may only be able to cache part 
of its table.

---
PC: Robert.Elliott at compaq.com
UNIX: relliott at hobbit.eng.hou.compaq.com

*
* For T10 Reflector information, send a message with
* 'info t10' (no quotes) in the message body to majordomo at t10.org




More information about the T10 mailing list