Hi Thinh,
On Wed, Jul 06, 2022 at 04:35:01PM -0700, Thinh Nguyen wrote:
> According to SAM-4 r14 section 7.2, for ABORT TASK function, a response
> of FUNCTION COMPLETE shall indicate that the command was aborted or was
> not in the task set. Currently we respond with TASK DOES NOT EXIST when
> there's no command in the task set. Fix the response to FUNCTION
> COMPLETE instead.
SAM does not describe a response status encoding. But other specs
do describe. For example, iSCSI RFC7143 11.6.1. Response
0 - Function complete
1 - Task does not exist
The mapping of the response code into a SCSI service response code
value, if needed, is outside the scope of this document. However, in
symbolic terms, Response values 0 and 1 map to the SCSI service
response of FUNCTION COMPLETE.
So, the current code is according to specs.
Moreover, TMR_TASK_DOES_NOT_EXIST is used in several fabric drivers to
handle some corner cases.
>
> Fixes: 3d28934aaae5 ("target: Add TMR_ABORT_TASK task management support")
> Signed-off-by: Thinh Nguyen <Thinh.Nguyen@xxxxxxxxxxxx>
> ---
> drivers/target/target_core_tmr.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/target/target_core_tmr.c b/drivers/target/target_core_tmr.c
> index 2af80d0998bf..724ddabda488 100644
> --- a/drivers/target/target_core_tmr.c
> +++ b/drivers/target/target_core_tmr.c
> @@ -170,9 +170,9 @@ void core_tmr_abort_task(
> if (dev->transport->tmr_notify)
> dev->transport->tmr_notify(dev, TMR_ABORT_TASK, &aborted_list);
>
> - printk("ABORT_TASK: Sending TMR_TASK_DOES_NOT_EXIST for ref_tag: %lld\n",
> + printk("ABORT_TASK: Sending TMR_FUNCTION_COMPLETE for ref_tag: %lld\n",
> tmr->ref_task_tag);
> - tmr->response = TMR_TASK_DOES_NOT_EXIST;
> + tmr->response = TMR_FUNCTION_COMPLETE;
> atomic_long_inc(&dev->aborts_no_task);
> }
>
