On Fri, 2014-06-06 at 13:51 -0700, Vasu Dev wrote:
> On Thu, 2014-06-05 at 23:30 +0000, Nicholas A. Bellinger wrote:
> > From: Nicholas Bellinger <nab@xxxxxxxxxxxxxxx>
> >
> > This patch changes ft_queue_data_in() to set SAM_STAT_TASK_SET_FULL
> > status upon a lport->tt.seq_send() failure, where it will now stop
> > sending subsequent DataIN, and immediately attempt to send the
> > response with exception status.
> >
> > Sending a response with SAM_STAT_TASK_SET_FULL status is useful in
> > order to signal the initiator that it should try to reduce it's
> > current queue_depth, to lower the number of outstanding I/Os on
> > the wire.
> >
> > Also, add a check to skip sending DataIN if TASK_SET_FULL status
> > has already been set due to a response lport->tt.seq_send()
> > failure, that has asked target-core to requeue a response.
> >
> > Reported-by: Vasu Dev <vasu.dev@xxxxxxxxxxxxxxx>
> > Cc: Vasu Dev <vasu.dev@xxxxxxxxxxxxxxx>
> > Cc: Jun Wu <jwu@xxxxxxxxxxxx>
> > Signed-off-by: Nicholas Bellinger <nab@xxxxxxxxxxxxxxx>
> > ---
> > drivers/target/tcm_fc/tfc_io.c | 14 +++++++++++++-
> > 1 file changed, 13 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/target/tcm_fc/tfc_io.c b/drivers/target/tcm_fc/tfc_io.c
> > index e415af3..140659f 100644
> > --- a/drivers/target/tcm_fc/tfc_io.c
> > +++ b/drivers/target/tcm_fc/tfc_io.c
> > @@ -82,6 +82,10 @@ int ft_queue_data_in(struct se_cmd *se_cmd)
> >
> > if (cmd->aborted)
> > return 0;
> > +
> > + if (se_cmd->scsi_status == SAM_STAT_TASK_SET_FULL)
> > + goto queue_status;
> > +
> > ep = fc_seq_exch(cmd->seq);
> > lport = ep->lp;
> > cmd->seq = lport->tt.seq_start_next(cmd->seq);
> > @@ -178,14 +182,22 @@ int ft_queue_data_in(struct se_cmd *se_cmd)
> > FC_TYPE_FCP, f_ctl, fh_off);
> > error = lport->tt.seq_send(lport, seq, fp);
> > if (error) {
> > - /* XXX For now, initiator will retry */
> > pr_err_ratelimited("%s: Failed to send frame %p, "
> > "xid <0x%x>, remaining %zu, "
> > "lso_max <0x%x>\n",
> > __func__, fp, ep->xid,
> > remaining, lport->lso_max);
> > + /*
> > + * Generate a TASK_SET_FULL status to notify the
> > + * initiator to reduce it's queue_depth, ignoring
> > + * the rest of the data-in and immediately attempt
> > + * to send the response.
> > + */
>
> I see above added check will drop rest of the frames but cannot find
> notifying to initiator with the TASK_SET_FULL status as in comments
> above, may be just the comment needs update otherwise code changes are
> fine.
>
The break aborts the DataIN send loop and invokes ft_queue_status()
below in an attempt to send TASK_SET_FULL status.
If the ft_queue_status() -> lport->tt.seq_send() also fails, then
-ENOMEM will be returned to the target and a delayed re-queue attempt
will be made.
In any event, updating the comment above to be more precise..
> > + se_cmd->scsi_status = SAM_STAT_TASK_SET_FULL;
> > + break;
> > }
> > }
> > +queue_status:
> > return ft_queue_status(se_cmd);
> > }
> >
>
> Reviewed-by: Vasu Dev <vasu.dev@xxxxxxxxx>
>
Thanks Vasu!
--nab
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
