Hi! After introduction of block callbacks classifiers can no longer track offload state. cls_bpf used to do that in an attempt to move common code from drivers to the core. Remove that functionality and fix drivers.
The user-visible bug this is fixing is that trying to offload a second filter would trigger a spurious DESTROY and in turn disable the already installed one. Jakub Kicinski (2): cls_bpf: fix offload assumptions after callback conversion nfp: bpf: keep track of the offloaded program drivers/net/ethernet/netronome/nfp/bpf/main.c | 55 ++++++++++++---- drivers/net/ethernet/netronome/nfp/bpf/main.h | 8 +++ include/net/pkt_cls.h | 5 +- net/sched/cls_bpf.c | 93 +++++++++++---------------- 4 files changed, 92 insertions(+), 69 deletions(-) -- 2.15.1
