[Intel-wired-lan] [RFC v3 net-next 06/18] net: ipv4: udp: Hook into time based transmission.

[Willem de Bruijn]

On Tue, Mar 6, 2018 at 8:12 PM, Jesus Sanchez-Palencia
<jesus.sanchez-palencia at intel.com> wrote:
> From: Richard Cochran <rcochran at linutronix.de>
>
> For udp packets, copy the desired future transmit time from the CMSG
> cookie into the skb.
>
> Signed-off-by: Richard Cochran <rcochran at linutronix.de>
> Signed-off-by: Jesus Sanchez-Palencia <jesus.sanchez-palencia at intel.com>
> ---
>  net/ipv4/udp.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
>
> diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
> index 3013404d0935..d683bbde526b 100644
> --- a/net/ipv4/udp.c
> +++ b/net/ipv4/udp.c
> @@ -926,6 +926,7 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
>         }
>
>         ipc.sockc.tsflags = sk->sk_tsflags;
> +       ipc.sockc.transmit_time = 0;
>         ipc.addr = inet->inet_saddr;
>         ipc.oif = sk->sk_bound_dev_if;
>
> @@ -1040,8 +1041,10 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
>                                   sizeof(struct udphdr), &ipc, &rt,
>                                   msg->msg_flags);
>                 err = PTR_ERR(skb);
> -               if (!IS_ERR_OR_NULL(skb))
> +               if (!IS_ERR_OR_NULL(skb)) {
> +                       skb->tstamp = ipc.sockc.transmit_time;
>                         err = udp_send_skb(skb, fl4);
> +               }

similar comment to raw: this implements only for a subset of udp requests:
those that can take the fast path.