[ofa-general] [PATCH] ipoib: refresh path when remote lid changes

Hal Rosenstock hal.rosenstock at gmail.com
Wed Jul 29 03:30:07 PDT 2009


On Wed, Jul 29, 2009 at 5:24 AM, Eli Dorfman (Voltaire) <
dorfman.eli at gmail.com> wrote:

> Hal Rosenstock wrote:
> >
> >
> > On Tue, Jul 28, 2009 at 6:14 AM, Moni Shoua <monis at voltaire.com
> > <mailto:monis at voltaire.com>> wrote:
> >
> >     Jason Gunthorpe wrote:
> >     > On Mon, Jul 27, 2009 at 08:11:42PM +0300, Yossi Etigin wrote:
> >     >>   If the LID of an ipoib neighbour changes without a SM event on
> >     the local node,
> >     >> IPoIB will keep caching the invalid path until the device is
> >     flushed. The patch
> >     >> below will remove the path for every incoming ARP packet where
> >     the sender hardware
> >     >> address does not match the cached lid.
> >     >
> >     > This assumes a LMC=0 fabric.. Otherwise there is no guarentee that
> the
> >     > slid the sender uses is the same one you are using to send.
> >     >
> >     > Jason
> >     > _______________________________________________
> >     > general mailing list
> >     > general at lists.openfabrics.org <mailto:
> general at lists.openfabrics.org>
> >     > http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
> >     >
> >     > To unsubscribe, please visit
> >     http://openib.org/mailman/listinfo/openib-general
> >     >
> >     This issue can be solved by a little bit different check
> >     instead of
> >                    be16_to_cpu(path->pathrec.dlid) == lid)
> >     do
> >                    is_in_same_lid_group(be16_to_cpu(path->pathrec.dlid),
> >     lid, lmc)
> >
> >     what do you think?
> >
> >
> > Is the LMC in the above from the local port ? Unfortunately, LMC is not
> > required to be uniform across the subnet so the remote port's LMC may
> > not be the same as that on the local port.
>
> opensm configures the same LMC to all endports (CA) in the fabric so in
> which case do you suspect that
> it will be different?


I wasn't talking about the simplification that OpenSM uses but rather what
IBA allows. There may be other SMs which do this or OpenSM could be extended
(with additional configuration for this).

-- Hal


>
>
> Eli
> >
> > -- Hal
> >
> >
> >
> >     _______________________________________________
> >     general mailing list
> >     general at lists.openfabrics.org <mailto:general at lists.openfabrics.org>
> >     http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
> >
> >     To unsubscribe, please visit
> >     http://openib.org/mailman/listinfo/openib-general
> >
> >
> >
> > ------------------------------------------------------------------------
>  >
> > _______________________________________________
> > general mailing list
> > general at lists.openfabrics.org
> > http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
> >
> > To unsubscribe, please visit
> http://openib.org/mailman/listinfo/openib-general
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/general/attachments/20090729/0e517e16/attachment.html>


More information about the general mailing list