[bpf-next,06/14] bpf: net: Avoid do_ipv6_setsockopt() taking sk lock when called from bpf

Commit Message

Martin KaFai Lau July 27, 2022, 6:09 a.m. UTC

Similar to the earlier patch that avoids sock_setsockopt() from
taking sk lock when called from bpf.  This patch changes
do_ipv6_setsockopt() to use the {lock,release}_sock_sockopt().

Signed-off-by: Martin KaFai Lau <[e-mail]>
---
 net/ipv6/ipv6_sockglue.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

Patch

diff --git a/net/ipv6/ipv6_sockglue.c b/net/ipv6/ipv6_sockglue.c
index 222f6bf220ba..4559f02ab4a8 100644
--- a/net/ipv6/ipv6_sockglue.c
+++ b/net/ipv6/ipv6_sockglue.c
@@ -417,7 +417,7 @@  static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 
 	if (needs_rtnl)
 		rtnl_lock();
-	lock_sock(sk);
+	lock_sock_sockopt(sk, optval);
 
 	switch (optname) {
 
@@ -994,14 +994,14 @@  static int do_ipv6_setsockopt(struct sock *sk, int level, int optname,
 		break;
 	}
 
-	release_sock(sk);
+	release_sock_sockopt(sk, optval);
 	if (needs_rtnl)
 		rtnl_unlock();
 
 	return retv;
 
 e_inval:
-	release_sock(sk);
+	release_sock_sockopt(sk, optval);
 	if (needs_rtnl)
 		rtnl_unlock();
 	return -EINVAL;