Message ID | 20171110115435.4261-1-privat@egil-hjelmeland.no |
---|---|
Headers | show |
Series | net: dsa: lan9303: IGMP handling | expand |
> skb->offload_fwd_mark calculation is a candidate for consolidation into the > DSA core. The calculation can probably be more polished when done at a point > where DSA has updated skb. Hi Egil Yes, at some point we should do this. But at the moment it is too early. We don't have enough experience with what the different switches need. I would like to see 4 or more devices doing it before we consolidate the code into the core. Andrew
Den 10. nov. 2017 15:07, skrev Andrew Lunn: >> skb->offload_fwd_mark calculation is a candidate for consolidation into the >> DSA core. The calculation can probably be more polished when done at a point >> where DSA has updated skb. > > Hi Egil > > Yes, at some point we should do this. But at the moment it is too > early. We don't have enough experience with what the different > switches need. I would like to see 4 or more devices doing it before > we consolidate the code into the core. > Absolutely. My point was rather I prefer not spend time polishing this snippet right now, because it will probably be refactored later anyway. One steppingstone could be to make reusable helper-functions in net/dsa/dsa_priv.h. But I leave that to device #2 that implement this ;-) > Andrew > Egil
From: Egil Hjelmeland <privat@egil-hjelmeland.no> Date: Fri, 10 Nov 2017 12:54:33 +0100 > Set up the HW switch to trap IGMP packets to CPU port. > And make sure skb->offload_fwd_mark is cleared for incoming IGMP packets. > > skb->offload_fwd_mark calculation is a candidate for consolidation into the > DSA core. The calculation can probably be more polished when done at a point > where DSA has updated skb. Series applied, thank you.