Question about CLEAR TASK SET and ABORT TASK SET

Gerry Houlder gerry.houlder at seagate.com
Wed Jan 16 10:19:41 PST 2013


Formatted message: <a href="http://www.t10.org/cgi-bin/ac.pl?t=r&f=r1301163_f.htm">HTML-formatted message</a>

No we don't need any changes here. As described in current SAM-5 wording,
ABORT TASK SET is already equivalent to an abort I_T nexus function, so
there is no point in adding a new function.
On Wed, Jan 16, 2013 at 11:22 AM, Mark Evans <Mark.Evans at wdc.com> wrote:
>  Hi Gerry,****
>
> ** **
>
> I was wrong in what I wrote about ABORT TASK SET.  That TMF aborts all
> commands received on an I_T nexus.  If TST is set to 001b, this means all
> tasks in that particular task set.****
>
> ** **
>
> Unfortunately, the new definition for CLEAR TASK SET in SAM-2 makes it so
> that, when TST is set to 001b, the only way to abort all tasks in all task
> sets is by sending a LOGICAL UNIT RESET.  This TMF potentially causes lots
> of other stuff to happen besides just aborting all commands in all task
> sets.****
>
> ** **
>
> Maybe we should define a new TMF and modify what definitions we have such
> that there is a true TMF hierarchy:****
>
> ** **
>
> **1)	   **ABORT TASK aborts a specific task from a specific I_T nexus;***
> *
>
> **2)	   **ABORT I_T NEXUS aborts all tasks from a specific I_T nexus,
> regardless of task set;****
>
> **3)	   **ABORT TASK SET aborts all tasks in a task set, regardless of
> I_T nexus;****
>
> **4)	   **CLEAR TASK SET aborts all tasks from all I_T nexuses,
> regardless of task set; and****
>
> **5)	   **LOGICAL UNIT RESET aborts all tasks from all I_T nexuses,
> regardless of task set, and may cause lots of other stuff to happen.****
>
> ** **
>
> Please feel free to call or send an email to me with any comments or
> questions that you have about this stuff. ****
>
> ** **
>
> Regards, ****
>
> ** **
>
> Mark Evans
> Western Digital Corporation
> 5863 Rue Ferrari
> San Jose, CA 95138
> Email: mark.evans at wdc.com
> Home office: 541.563.7880
> Cell: 408.391.7805****
>
>  ****
>
>  ****
>
> ** **
>



More information about the T10 mailing list