[ofw] RE: [IPoIB] [Patch] adapter could be left in INIT state.
alex.estrin at qlogic.com
Wed Feb 11 08:52:49 PST 2009
Also applied in ipoib_cm branch rev 1954.
> -----Original Message-----
> From: Tzachi Dar [mailto:tzachid at mellanox.co.il]
> Sent: Wednesday, February 11, 2009 5:09 AM
> To: Sean Hefty; Alex Estrin; ofw at lists.openfabrics.org
> Cc: Alex Naslednikov
> Subject: RE: [ofw] RE: [IPoIB] [Patch] adapter could be left in INIT state.
> I guess that this can happen even with the IBAL's event reporting, since
> The current ipoib code is processing the notifcations in asyncronious
> I believe that the problem starts from the fact that there is an ipoib
> and an ipoib port which have different states and different locks but
> are actually
> representing the same object.
> In any case, I have applied the patch at 1939 and let's hope that it
> will not create new
> > -----Original Message-----
> > From: ofw-bounces at lists.openfabrics.org
> > [mailto:ofw-bounces at lists.openfabrics.org] On Behalf Of Sean Hefty
> > Sent: Wednesday, February 11, 2009 12:31 AM
> > To: 'Alex Estrin'; ofw at lists.openfabrics.org
> > Cc: Alex Naslednikov
> > Subject: [ofw] RE: [IPoIB] [Patch] adapter could be left in
> > INIT state.
> > >Adapter has a chance to stuck in IB_PNP_PORT_INIT state if
> > PNP events
> > >arrive in following order: first IB_PNP_LID_CHANGE or
> > IB_PNP_SM_CHANGE
> > >- ipoib_port_up() called first time, then later IB_PNP_PORT_ACTIVE -
> > >ipoib_port_up() called again.
> > Is this possible with the IBAL 'event reporting' code? (Nit:
> > can we please stop calling what IBAL does PnP?) If this is
> > possible, should the fix go into IBAL's event reporting?
> > - Sean
> > _______________________________________________
> > ofw mailing list
> > ofw at lists.openfabrics.org
> > http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ofw
More information about the ofw