[ Fixed up Vincent's address as John's and my reply seems not to be delivered. ]
On Fri, Jul 11, 2025 at 11:48 AM Kuniyuki Iwashima <kuniyu@xxxxxxxxxx> wrote:
>
> On Fri, Jul 11, 2025 at 10:27 AM John Fastabend
> <john.fastabend@xxxxxxxxx> wrote:
> >
> > On 2025-07-09 14:47:57, Vincent Whitchurch via B4 Relay wrote:
> > > From: Vincent Whitchurch <vincent.whitchurch@xxxxxxxxxxxxx>
> > >
> > > The TCP BPF code will need to override splice_read(), so add it to prot.
> > >
> > > Signed-off-by: Vincent Whitchurch <vincent.whitchurch@xxxxxxxxxxxxx>
> > > ---
> > > include/net/inet_common.h | 3 +++
> > > include/net/sock.h | 3 +++
> > > net/ipv4/af_inet.c | 13 ++++++++++++-
> > > net/ipv4/tcp_ipv4.c | 1 +
> > > net/ipv6/af_inet6.c | 2 +-
> > > net/ipv6/tcp_ipv6.c | 1 +
> > > 6 files changed, 21 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/include/net/inet_common.h b/include/net/inet_common.h
> > > index c17a6585d0b0..2a6480d0d575 100644
> > > --- a/include/net/inet_common.h
> > > +++ b/include/net/inet_common.h
> > > @@ -35,6 +35,9 @@ void __inet_accept(struct socket *sock, struct socket *newsock,
> > > struct sock *newsk);
> > > int inet_send_prepare(struct sock *sk);
> > > int inet_sendmsg(struct socket *sock, struct msghdr *msg, size_t size);
> > > +ssize_t inet_splice_read(struct socket *sk, loff_t *ppos,
> > > + struct pipe_inode_info *pipe, size_t len,
> > > + unsigned int flags);
> > > void inet_splice_eof(struct socket *sock);
> > > int inet_recvmsg(struct socket *sock, struct msghdr *msg, size_t size,
> > > int flags);
> > > diff --git a/include/net/sock.h b/include/net/sock.h
> > > index 4c37015b7cf7..4bdebcbcca38 100644
> > > --- a/include/net/sock.h
> > > +++ b/include/net/sock.h
> > > @@ -1280,6 +1280,9 @@ struct proto {
> > > size_t len);
> > > int (*recvmsg)(struct sock *sk, struct msghdr *msg,
> > > size_t len, int flags, int *addr_len);
> > > + ssize_t (*splice_read)(struct socket *sock, loff_t *ppos,
> > > + struct pipe_inode_info *pipe, size_t len,
> > > + unsigned int flags);
> > > void (*splice_eof)(struct socket *sock);
> > > int (*bind)(struct sock *sk,
> > > struct sockaddr *addr, int addr_len);
> > > diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
> > > index 76e38092cd8a..9c521d252f66 100644
> > > --- a/net/ipv4/af_inet.c
> > > +++ b/net/ipv4/af_inet.c
> > > @@ -868,6 +868,17 @@ void inet_splice_eof(struct socket *sock)
> > > }
> > > EXPORT_SYMBOL_GPL(inet_splice_eof);
> > >
> > > +ssize_t inet_splice_read(struct socket *sock, loff_t *ppos,
> > > + struct pipe_inode_info *pipe, size_t len,
> > > + unsigned int flags)
> > > +{
> > > + struct sock *sk = sock->sk;
> > > +
> > > + return INDIRECT_CALL_1(sk->sk_prot->splice_read, tcp_splice_read, sock,
> > > + ppos, pipe, len, flags);
> > > +}
> >
> > Could we do a indirect_call_2 here? something like this?
> >
> > INDIRECT_CALL_2(sk->sk_prot->splice_read, tcp_splice_read ...
> >
> > Otherwise the series looks reasonable to me.
>
> What's the second candidate ?
> I think we should specify the built-in one and cannot use bpf one.
