This small series corrects some inconveniences for a BPF TCP CC that implements and uses tcp_congestion_ops.cong_control(). Until now, such a CC did not have all necessary write access to struct sock and unnecessarily needed to implement cong_avoid(). --- v2: - Drop redundant check for required functions and just rely on tcp_register_congestion_control() (Martin KaFai Lau) Jörn-Thorben Hinz (2): bpf: Allow a TCP CC to write sk_pacing_rate and sk_pacing_status bpf: Require only one of cong_avoid() and cong_control() from a TCP CC net/ipv4/bpf_tcp_ca.c | 38 ++++++-------------------------------- 1 file changed, 6 insertions(+), 32 deletions(-) -- 2.30.2