Jason Gunthorpe wrote:
> Check that the format of the multicast link address is correct before
> taking it from dev->mc_list to priv->multicast_list. This way we never
> try to send a bogus address to the SA, and prevents badness from
> erronous 'ip maddr addr add', broken bonding drivers, or whatever.
> 
> Signed-off-by: Jason Gunthorpe <[email protected]>
> ---
>  drivers/infiniband/ulp/ipoib/ipoib_multicast.c |   18 ++++++++++++++++++
>  1 files changed, 18 insertions(+), 0 deletions(-)
> 
> Same problem Moni was working on, but lets just address it directly.
> 
> There is work to try and fix the bonding driver but no fixed version
> is in mainline yet. This is a cheap and simple work around that is
> worth having even once the driver is fixed.
> 
The fix is available from at least 2.6.31-rc2. However, I still need to check 
your claim that it doesn't work for you.

> Despite this, I think it is still necessary to do something like Moni
> was trying - to prevent the MCG join queue from head of line blocking
> on a single bad SA response. This can happen even if everything is
> correct.
> 
I'll resend my patch since it has been a long time from the last time I sent it.
_______________________________________________
general mailing list
[email protected]
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general

To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general

Reply via email to