On 11/6/23 10:32 PM, Jakub Kicinski wrote:
On Fri, 3 Nov 2023 23:27:46 +0100 Daniel Borkmann wrote:
ndo_get_peer_dev is used in tcx BPF fast path, therefore make use of
indirect call wrapper and therefore optimize the bpf_redirect_peer()
internal handling a bit. Add a small skb_get_peer_dev() wrapper which
utilizes the INDIRECT_CALL_1() macro instead of open coding.
Why don't we kill the ndo and put the pointer in struct net_device?
I can take a stab at this some time after LPC and probably makes sense
also after Coco's cacheline optimizations landed.
Thanks,
Daniel
