[v2,bpf-next,03/17] bpf: net: Avoid sk_getsockopt() taking sk lock when called from bpf

Commit Message

Martin KaFai Lau Sept. 2, 2022, 12:28 a.m. UTC

From: Martin KaFai Lau <[e-mail]>

Similar to the earlier commit that changed sk_setsockopt() to
use sockopt_{lock,release}_sock() such that it can avoid taking
lock when called from bpf.  This patch also changes sk_getsockopt()
to use sockopt_{lock,release}_sock() such that a latter patch can
make bpf_getsockopt(SOL_SOCKET) to reuse sk_getsockopt().

Only sk_get_filter() requires this change and it is used by
the optname SO_GET_FILTER.

The '.getname' implementations in sock->ops->getname() is not
changed also since bpf does not always have the sk->sk_socket
pointer and cannot support SO_PEERNAME.

Signed-off-by: Martin KaFai Lau <[e-mail]>
---
 net/core/filter.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

Patch

diff --git a/net/core/filter.c b/net/core/filter.c
index 0f6f86b9e487..8e7fc71160cd 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -10718,7 +10718,7 @@  int sk_get_filter(struct sock *sk, sockptr_t optval, unsigned int len)
 	struct sk_filter *filter;
 	int ret = 0;
 
-	lock_sock(sk);
+	sockopt_lock_sock(sk);
 	filter = rcu_dereference_protected(sk->sk_filter,
 					   lockdep_sock_is_held(sk));
 	if (!filter)
@@ -10751,7 +10751,7 @@  int sk_get_filter(struct sock *sk, sockptr_t optval, unsigned int len)
 	 */
 	ret = fprog->len;
 out:
-	release_sock(sk);
+	sockopt_release_sock(sk);
 	return ret;
 }