From: Ido Schimmel <ido...@mellanox.com> Date: Thu, 4 Oct 2018 15:48:00 +0000
> First patch works around an hardware issue in Spectrum-2 where a field > indicating the event type is always set to the same value. Since there > are only two event types and they are reported using different queues, > we can use the queue number to derive the event type. > > Second patch prevents a router interface (RIF) leakage when a VLAN > device is deleted from on top a bridge device. Series applied, and patch #2 queued up for -stable. I was thinking about Fixes tags for the situation in patch #2 a bit, and it's probably best to list two Fixes tags. One for the original bug that wasn't fixed properly, and one for the incomplete fix. It's nice that you mentioned the incomplete fix in the commit message but a backporter might miss this chain of commits and dependencies it if it isn't in the Fixes: tags. Thanks.
