Hi Santosh,
On Thu, 2013-11-07 at 11:55 +0530, santosh kulkarni wrote:
> Hi Nicholas,
>
> On Wednesday 06 November 2013 11:57 PM, Nicholas A. Bellinger wrote:
<SNIP>
> >> I am facing few issues and needed to clarify few of my doubts.
> >> 1. Our suite currently checks the behavior of a target when a Data-Out
> >> PDU with DataSegmentLength set to 0 is sent out. And what i see in the
> >> target is as follows
> >>
> >> ~# dmesg
> >> [599509.602032] DataOUT payload is ZERO, protocol error.
> >> [599675.065060] DataOUT payload is ZERO, protocol error.
> >>
> >> But the RFC states the following
> >>
> >>
> >>
> >> 10.7.7. DataSegmentLength
> >> This is the data payload length of a SCSI Data-In or SCSI Data-Out PDU.
> >> "The sending of 0 length data segments should be avoided, but initiators
> >> and targets MUST be able to properly receive 0 length data
> >>
> >> segments."
> >> Is this the default behavior of LIO. It would be great if someone could
> >> clarify my doubts.
> >>
> > Here is a quick patch to drop the sending of the REJECT, and ignore zero
> > length ISCSI_OP_SCSI_DATA_OUTs.
> >
> > Care to test..?
> >
> > --nab
> >
> > diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c
> > index e183352..ab64cbb 100644
> > --- a/drivers/target/iscsi/iscsi_target.c
> > +++ b/drivers/target/iscsi/iscsi_target.c
> > @@ -1249,9 +1249,8 @@ iscsit_check_dataout_hdr(struct iscsi_conn *conn, unsigned char *buf,
> > int rc;
> >
> > if (!payload_length) {
> > - pr_err("DataOUT payload is ZERO, protocol error.\n");
> > - return iscsit_add_reject(conn, ISCSI_REASON_PROTOCOL_ERROR,
> > - buf);
> > + pr_warn("DataOUT payload is ZERO, ignoring.\n");
> > + return 0;
> > }
> >
> > /* iSCSI write */
> > @@ -1481,7 +1480,7 @@ EXPORT_SYMBOL(iscsit_check_dataout_payload);
> >
> > static int iscsit_handle_data_out(struct iscsi_conn *conn, unsigned char *buf)
> > {
> > - struct iscsi_cmd *cmd;
> > + struct iscsi_cmd *cmd = NULL;
> > struct iscsi_data *hdr = (struct iscsi_data *)buf;
> > int rc;
> > bool data_crc_failed = false;
> >
> >
> >
> Thanks for the quick fix.Will try to apply it and test it out.
>
Have you been able to confirm with this patch..?
It's been included into for-next, as it's a pretty obvious change to me,
but I'd still like to see a test confirmation.
Thanks,
--nab
--
To unsubscribe from this list: send the line "unsubscribe target-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
