bpf: export bpf_sock for BPF_PROG_TYPE_CGROUP_SOCK_ADDR prog type
authorStanislav Fomichev <sdf@google.com>
Wed, 12 Jun 2019 17:30:37 +0000 (10:30 -0700)
committerDaniel Borkmann <daniel@iogearbox.net>
Fri, 14 Jun 2019 23:21:56 +0000 (01:21 +0200)
commitfb85c4a730af221339c1dde1a434b73da0dfc3ed
treefce01722704a787ffdcfdeba1b5dc3d305359c84
parentd30bd78ce8ba1f093aa4877ed06601d0debd872b
bpf: export bpf_sock for BPF_PROG_TYPE_CGROUP_SOCK_ADDR prog type

And let it use bpf_sk_storage_{get,delete} helpers to access socket
storage. Kernel context (struct bpf_sock_addr_kern) already has sk
member, so I just expose it to the BPF hooks. Using PTR_TO_SOCKET
instead of PTR_TO_SOCK_COMMON should be safe because the hook is
called on bind/connect.

Cc: Martin Lau <kafai@fb.com>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
include/uapi/linux/bpf.h
net/core/filter.c