selftests/bpf: Update selftests to always provide "struct_ops" SEC
authorAndrii Nakryiko <andrii@kernel.org>
Tue, 14 Sep 2021 01:47:30 +0000 (18:47 -0700)
committerAlexei Starovoitov <ast@kernel.org>
Tue, 14 Sep 2021 22:49:24 +0000 (15:49 -0700)
Update struct_ops selftests to always specify "struct_ops" section
prefix. Libbpf will require a proper BPF program type set in the next
patch, so this prevents tests breaking.

Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Link: https://lore.kernel.org/bpf/20210914014733.2768-2-andrii@kernel.org
tools/testing/selftests/bpf/progs/bpf_cubic.c

index f62df4d..d9660e7 100644 (file)
@@ -169,11 +169,7 @@ static __always_inline void bictcp_hystart_reset(struct sock *sk)
        ca->sample_cnt = 0;
 }
 
-/* "struct_ops/" prefix is not a requirement
- * It will be recognized as BPF_PROG_TYPE_STRUCT_OPS
- * as long as it is used in one of the func ptr
- * under SEC(".struct_ops").
- */
+/* "struct_ops/" prefix is a requirement */
 SEC("struct_ops/bpf_cubic_init")
 void BPF_PROG(bpf_cubic_init, struct sock *sk)
 {
@@ -188,10 +184,8 @@ void BPF_PROG(bpf_cubic_init, struct sock *sk)
                tcp_sk(sk)->snd_ssthresh = initial_ssthresh;
 }
 
-/* No prefix in SEC will also work.
- * The remaining tcp-cubic functions have an easier way.
- */
-SEC("no-sec-prefix-bictcp_cwnd_event")
+/* "struct_ops" prefix is a requirement */
+SEC("struct_ops/bpf_cubic_cwnd_event")
 void BPF_PROG(bpf_cubic_cwnd_event, struct sock *sk, enum tcp_ca_event event)
 {
        if (event == CA_EVENT_TX_START) {