On Wed, Jan 26, 2022 at 12:22:13AM +0000, Pavel Begunkov wrote:
#define BPF_CGROUP_RUN_PROG_INET_INGRESS(sk, skb) \From reading sk_filter_trim_cap() where this will be called, sk cannot be NULL.
({ \
int __ret = 0; \
- if (cgroup_bpf_enabled(CGROUP_INET_INGRESS)) \
+ if (cgroup_bpf_enabled(CGROUP_INET_INGRESS) && sk && \
If yes, the new sk test is not needed.
Others lgtm.
+ cgroup_bpf_sock_enabled(sk, CGROUP_INET_INGRESS)) \
__ret = __cgroup_bpf_run_filter_skb(sk, skb, \
CGROUP_INET_INGRESS); \
\