Hi Bart,
On Tue, 2019-12-10 at 13:00 -0500, Bart Van Assche wrote:
> On 12/10/19 5:52 AM, Martin Wilck wrote:
> > Hello Bart,
> >
> > On Mon, 2019-12-09 at 10:02 -0800, Bart Van Assche wrote:
> > > Restore point-to-point for qla25xx and older. Although this patch
> > > initializes
> > > a field (s_id) that has been marked as "reserved" in the firmware
> > > manual, it
> > > works fine on my setup.
> > >
> > > Cc: Quinn Tran <qutran@xxxxxxxxxxx>
> > > Cc: Martin Wilck <mwilck@xxxxxxxx>
> > > Cc: Daniel Wagner <dwagner@xxxxxxx>
> > > Cc: Roman Bolshakov <r.bolshakov@xxxxxxxxx>
> > > Fixes: 0aabb6b699f7 ("scsi: qla2xxx: Fix Nport ID display value")
> > > Fixes: edd05de19759 ("scsi: qla2xxx: Changes to support N2N
> > > logins")
> > > Signed-off-by: Bart Van Assche <bvanassche@xxxxxxx>
> > > ---
> > > drivers/scsi/qla2xxx/qla_iocb.c | 14 ++++++++++----
> > > 1 file changed, 10 insertions(+), 4 deletions(-)
> >
> > Having followed the discussion between you and Roman, I guess this
> > is
> > ok. However I'd like to understand better in what ways the N2N
> > topology
> > was broken for you. After all, this patch affects only the LOGO
> > payload. Was it a logout / relogin issue? Were wrong ports being
> > logged
> > out?
>
> Hi Martin,
>
> Without this patch N2N login fails for 25xx adapters. With this patch
> N2N login succeeds for 25xx adapters. You may have noticed that
> Martin
> Petersen asked Himanshu for advice in another e-mail thread about how
> to
> address this regression.
Of course I did. It still kind of baffles me that a change in the LOGO
iocb causes LOGIN to fail, but whatever.
Martin
