diff options
author | Eric Dumazet <edumazet@google.com> | 2023-09-13 12:58:35 +0000 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2023-09-16 10:09:43 +0100 |
commit | 41862d12e77f78b4ecb59b028bf44de92991bda2 (patch) | |
tree | e4f3ed2dcbaaf1a8b14bf2da554650b2ca92b0b0 /net/core/sock.c | |
parent | 4fa5ce3e3a10da4ecc438a548fc701dec5f28758 (diff) |
net: use indirect call helpers for sk->sk_prot->release_cb()
When adding sk->sk_prot->release_cb() call from __sk_flush_backlog()
Paolo suggested using indirect call helpers to take care of
CONFIG_RETPOLINE=y case.
It turns out Google had such mitigation for years in release_sock(),
it is time to make this public :)
Suggested-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/core/sock.c')
-rw-r--r-- | net/core/sock.c | 7 |
1 files changed, 5 insertions, 2 deletions
diff --git a/net/core/sock.c b/net/core/sock.c index 213a62ac13f2..a5995750c5c5 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -3003,7 +3003,9 @@ void __sk_flush_backlog(struct sock *sk) __release_sock(sk); if (sk->sk_prot->release_cb) - sk->sk_prot->release_cb(sk); + INDIRECT_CALL_INET_1(sk->sk_prot->release_cb, + tcp_release_cb, sk); + spin_unlock_bh(&sk->sk_lock.slock); } EXPORT_SYMBOL_GPL(__sk_flush_backlog); @@ -3523,7 +3525,8 @@ void release_sock(struct sock *sk) __release_sock(sk); if (sk->sk_prot->release_cb) - sk->sk_prot->release_cb(sk); + INDIRECT_CALL_INET_1(sk->sk_prot->release_cb, + tcp_release_cb, sk); sock_release_ownership(sk); if (waitqueue_active(&sk->sk_lock.wq)) |