| Message ID | 20180906135459.15529-1-vincent.whitchurch@axis.com |
|---|---|
| State | Accepted, archived |
| Delegated to: | David Miller |
| Headers | show |
| Series | tcp: really ignore MSG_ZEROCOPY if no SO_ZEROCOPY | expand |
On Thu, Sep 6, 2018 at 9:58 AM Vincent Whitchurch <vincent.whitchurch@axis.com> wrote: > > According to the documentation in msg_zerocopy.rst, the SO_ZEROCOPY > flag was introduced because send(2) ignores unknown message flags and > any legacy application which was accidentally passing the equivalent of > MSG_ZEROCOPY earlier should not see any new behaviour. > > Before commit f214f915e7db ("tcp: enable MSG_ZEROCOPY"), a send(2) call > which passed the equivalent of MSG_ZEROCOPY without setting SO_ZEROCOPY > would succeed. However, after that commit, it fails with -ENOBUFS. So > it appears that the SO_ZEROCOPY flag fails to fulfill its intended > purpose. Fix it. > > Fixes: f214f915e7db ("tcp: enable MSG_ZEROCOPY") > > Signed-off-by: Vincent Whitchurch <vincent.whitchurch@axis.com> Acked-by: Willem de Bruijn <willemb@google.com> Good catch, thanks for fixing this. Please remember to mark patches with PATCH net
From: Vincent Whitchurch <vincent.whitchurch@axis.com> Date: Thu, 6 Sep 2018 15:54:59 +0200 > According to the documentation in msg_zerocopy.rst, the SO_ZEROCOPY > flag was introduced because send(2) ignores unknown message flags and > any legacy application which was accidentally passing the equivalent of > MSG_ZEROCOPY earlier should not see any new behaviour. > > Before commit f214f915e7db ("tcp: enable MSG_ZEROCOPY"), a send(2) call > which passed the equivalent of MSG_ZEROCOPY without setting SO_ZEROCOPY > would succeed. However, after that commit, it fails with -ENOBUFS. So > it appears that the SO_ZEROCOPY flag fails to fulfill its intended > purpose. Fix it. > > Fixes: f214f915e7db ("tcp: enable MSG_ZEROCOPY") > Signed-off-by: Vincent Whitchurch <vincent.whitchurch@axis.com> Applied and queued up for -stable, thanks.
diff --git a/net/core/skbuff.c b/net/core/skbuff.c index c996c09d095f..b2c807f67aba 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -939,9 +939,6 @@ struct ubuf_info *sock_zerocopy_alloc(struct sock *sk, size_t size) WARN_ON_ONCE(!in_task()); - if (!sock_flag(sk, SOCK_ZEROCOPY)) - return NULL; - skb = sock_omalloc(sk, 0, GFP_KERNEL); if (!skb) return NULL; diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index b8af2fec5ad5..10c6246396cc 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -1185,7 +1185,7 @@ int tcp_sendmsg_locked(struct sock *sk, struct msghdr *msg, size_t size) flags = msg->msg_flags; - if (flags & MSG_ZEROCOPY && size) { + if (flags & MSG_ZEROCOPY && size && sock_flag(sk, SOCK_ZEROCOPY)) { if (sk->sk_state != TCP_ESTABLISHED) { err = -EINVAL; goto out_err;
According to the documentation in msg_zerocopy.rst, the SO_ZEROCOPY flag was introduced because send(2) ignores unknown message flags and any legacy application which was accidentally passing the equivalent of MSG_ZEROCOPY earlier should not see any new behaviour. Before commit f214f915e7db ("tcp: enable MSG_ZEROCOPY"), a send(2) call which passed the equivalent of MSG_ZEROCOPY without setting SO_ZEROCOPY would succeed. However, after that commit, it fails with -ENOBUFS. So it appears that the SO_ZEROCOPY flag fails to fulfill its intended purpose. Fix it. Fixes: f214f915e7db ("tcp: enable MSG_ZEROCOPY") Signed-off-by: Vincent Whitchurch <vincent.whitchurch@axis.com> --- net/core/skbuff.c | 3 --- net/ipv4/tcp.c | 2 +- 2 files changed, 1 insertion(+), 4 deletions(-)