Cong Wang wrote:
> On Wed, Mar 24, 2021 at 1:59 PM John Fastabend <john.fastabend@xxxxxxxxx> wrote:
> > diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c
> > index 47b7c5334c34..ecb5634b4c4a 100644
> > --- a/net/tls/tls_main.c
> > +++ b/net/tls/tls_main.c
> > @@ -754,6 +754,12 @@ static void tls_update(struct sock *sk, struct proto *p,
> >
> > ctx = tls_get_ctx(sk);
> > if (likely(ctx)) {
> > + /* TLS does not have an unhash proto in SW cases, but we need
> > + * to ensure we stop using the sock_map unhash routine because
> > + * the associated psock is being removed. So use the original
> > + * unhash handler.
> > + */
> > + WRITE_ONCE(sk->sk_prot->unhash, p->unhash);
> > ctx->sk_write_space = write_space;
> > ctx->sk_proto = p;
>
> It looks awkward to update sk->sk_proto inside tls_update(),
> at least when ctx!=NULL.
hmm. It doesn't strike me as paticularly awkward but OK.
>
> What is wrong with updating it in sk_psock_restore_proto()
> when inet_csk_has_ulp() is true? It looks better to me.
It could be wrong if inet_csk_has_ulp has an unhash callback
already assigned. But, because we know inet_csk_has_ulp()
really means is_tls_attached() it would be fine.
>
> diff --git a/include/linux/skmsg.h b/include/linux/skmsg.h
> index 6c09d94be2e9..da5dc3ef0ee3 100644
> --- a/include/linux/skmsg.h
> +++ b/include/linux/skmsg.h
> @@ -360,8 +360,8 @@ static inline void sk_psock_update_proto(struct sock *sk,
> static inline void sk_psock_restore_proto(struct sock *sk,
> struct sk_psock *psock)
> {
> - sk->sk_prot->unhash = psock->saved_unhash;
> if (inet_csk_has_ulp(sk)) {
> + sk->sk_prot->unhash = psock->sk_proto->unhash;
> tcp_update_ulp(sk, psock->sk_proto, psock->saved_write_space);
> } else {
> sk->sk_write_space = psock->saved_write_space;
>
>
> sk_psock_restore_proto() is the only caller of tcp_update_ulp()
> so should be equivalent.
Agree it is equivalent. I don't mind moving the assignment around
if folks think its nicer.
>
> Thanks.
