On 8/2/24 8:29 AM, Alan Maguire wrote:
Add support to test bpf_setsockopt(.., TCP_BPF_SOCK_OPS_CB_FLAGS, ...) in BPF iterator context; use per-socket storage to store the new value and retrieve it in a cgroup/getsockopt program we attach to allow us to query TCP_BPF_SOCK_OPS_CB_FLAGS via getsockopt. Signed-off-by: Alan Maguire <alan.maguire@xxxxxxxxxx> --- .../bpf/prog_tests/bpf_iter_setsockopt.c | 83 +++++++++++++------ .../selftests/bpf/progs/bpf_iter_setsockopt.c | 76 ++++++++++++++---
There are too many code churns to reuse this test to test a new TCP_BPF_SOCK_OPS_CB_FLAGS sockopt. This is not the right test to reuse. It was created mainly to test if the tcp batching logic can survive the bpf-iter's seq_stop.
I don't think it needs a separate bpf_set/getsockopt test specifically for the bpf iter prog. The test in patch 2 should be enough.
pw-bot: cr