On 4/8/25 4:28 PM, Eric Woudstra wrote: > In case of a bridge in the forward-fastpath or bridge-fastpath the fdb is > used to create the tuple. In case of roaming at layer 2 level, for example > 802.11r, the destination device is changed in the fdb. The destination > device of a direct transmitting tuple is no longer valid and traffic is > send to the wrong destination. Also the hardware offloaded fastpath is not > valid anymore. > > This flowentry needs to be torn down asap. Also make sure that the flow > entry is not being used, when marked for teardown. > > Changes in v2: > - Unchanged, only tags RFC net-next to PATCH nf-next. > > Eric Woudstra (3): > netfilter: flow: Add bridge_vid member > netfilter: nf_flow_table_core: teardown direct xmit when destination > changed > netfilter: nf_flow_table_ip: don't follow fastpath when marked > teardown > > include/net/netfilter/nf_flow_table.h | 2 + > net/netfilter/nf_flow_table_core.c | 66 +++++++++++++++++++++++++++ > net/netfilter/nf_flow_table_ip.c | 6 +++ > net/netfilter/nft_flow_offload.c | 3 ++ > 4 files changed, 77 insertions(+) > Hi Pablo, I understand if you are busy, but this patch-set could be reviewed totally separate from my other submissions. It addresses the issue of L2 roaming for any fastpath. I'll wait for any other comments, before sending the fix for 'static'.
