From mboxrd@z Thu Jan 1 00:00:00 1970 From: Samudrala, Sridhar Date: Fri, 30 Dec 2016 09:04:38 -0800 Subject: [Intel-wired-lan] [net-next PATCH 5/6] i40e: Add TX and RX support in switchdev mode. In-Reply-To: References: <1483078863-22026-1-git-send-email-sridhar.samudrala@intel.com> <1483078863-22026-6-git-send-email-sridhar.samudrala@intel.com> Message-ID: <586693A6.5090105@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: intel-wired-lan@osuosl.org List-ID: On 12/30/2016 7:31 AM, Or Gerlitz wrote: > On Fri, Dec 30, 2016 at 8:21 AM, Sridhar Samudrala > wrote: >> In switchdev mode, broadcast filter is not enabled on VFs, The broadcasts and > nit ", The broadcast" --> ", the broadcasts" or ". The broadcasts" > >> unknown frames from VFs are received by the PF and passed to corresponding VF >> port representator netdev. >> A host based switching entity like a linux bridge or OVS redirects these frames >> to the right VFs via VFPR netdevs. Any frames sent via VFPR netdevs are sent as >> directed transmits to the corresponding VFs. To enable directed transmit, skb >> metadata dst is used to pass the VF id and the frame is requeued to call the PFs >> transmit routine. > So the representator netdevs do or don't have TX/RX rings of their > own? if the latter, was there any special locking you had to do for > that? Atleast in the inital implementation, we are not creating any TX/RX rings for VFPR netdevs. skb->dev is set to the PF device and skb is requeued via dev_queue_xmit(). > > Are you exposing switchdev ops for the representators? didn't see that > or maybe it's in the 4th patch which didn't make it to the list? Not at this time. In the future patches when we offload fdb/vlan functionality, we could use switchdev ops.