On Wed, Jan 4, 2017 at 7:55 AM, Eric Dumazet <eric.duma...@gmail.com> wrote:
>
> On Tue, 2017-01-03 at 10:22 -0500, Soheil Hassas Yeganeh wrote:
> > On Mon, Jan 2, 2017 at 3:23 PM, Soheil Hassas Yeganeh <soh...@google.com>
> > wrote:
> > > On Mon, Jan 2, 2017 at 3:20 PM, Soheil Hassas Yeganeh
> > > <soheil.k...@gmail.com> wrote:
> > >> From: Soheil Hassas Yeganeh <soh...@google.com>
> > >>
> > >> For TCP sockets, tx timestamps are only captured when the user data
> > >> is successfully and fully written to the socket. In many cases,
> > >> however, TCP writes can be partial for which no timestamp is
> > >> collected.
> > >>
> > >> Collect timestamps when the user data is partially copied into
> > >> the socket.
> > >>
> > >> Signed-off-by: Soheil Hassas Yeganeh <soh...@google.com>
> > >> Cc: Willem de Bruijn <will...@google.com>
> > >> Cc: Yuchung Cheng <ych...@google.com>
> > >> Cc: Eric Dumazet <eduma...@google.com>
> > >> Cc: Neal Cardwell <ncardw...@google.com>
> > >> Cc: Martin KaFai Lau <ka...@fb.com>
> > >> ---
> > >> net/ipv4/tcp.c | 8 ++++++--
> > >> 1 file changed, 6 insertions(+), 2 deletions(-)
> > >>
> > >> diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
> > >> index 2e3807d..c207b16 100644
> > >> --- a/net/ipv4/tcp.c
> > >> +++ b/net/ipv4/tcp.c
> > >> @@ -992,8 +992,10 @@ static ssize_t do_tcp_sendpages(struct sock *sk,
> > >> struct page *page, int offset,
> > >> return copied;
> > >>
> > >> do_error:
> > >> - if (copied)
> > >> + if (copied) {
> > >> + tcp_tx_timestamp(sk, sk->sk_tsflags,
> > >> tcp_write_queue_tail(sk));
> > >> goto out;
> > >> + }
> > >> out_err:
> > >> /* make sure we wake any epoll edge trigger waiter */
> > >> if (unlikely(skb_queue_len(&sk->sk_write_queue) == 0 &&
> > >> @@ -1329,8 +1331,10 @@ int tcp_sendmsg(struct sock *sk, struct msghdr
> > >> *msg, size_t size)
> > >> }
> > >>
> > >> do_error:
> > >> - if (copied + copied_syn)
> > >> + if (copied + copied_syn) {
> > >> + tcp_tx_timestamp(sk, sk->sk_tsflags,
> > >> tcp_write_queue_tail(sk));
> >
> > Thanks to Willem for noting that this should be sockc.tsflags and not
> > sk->sk_tsflags. I'll send V2 to fix.
>
> Also, why not factorizing a bit and have a single point calling
> tcp_tx_timestamp() ?
>
> This would ease code review quite a bit.
Thanks Eric! will do in V2.
> diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
> index
> 4a044964da6670829e5c47fef52d2cd76360b59f..11357f3bd1f82fa29129dd3ecf4d270feb4a6b1d
> 100644
> --- a/net/ipv4/tcp.c
> +++ b/net/ipv4/tcp.c
> @@ -958,10 +958,8 @@ static ssize_t do_tcp_sendpages(struct sock *sk, struct
> page *page, int offset,
> copied += copy;
> offset += copy;
> size -= copy;
> - if (!size) {
> - tcp_tx_timestamp(sk, sk->sk_tsflags, skb);
> + if (!size)
> goto out;
> - }
>
> if (skb->len < size_goal || (flags & MSG_OOB))
> continue;
> @@ -987,8 +985,11 @@ static ssize_t do_tcp_sendpages(struct sock *sk, struct
> page *page, int offset,
> }
>
> out:
> - if (copied && !(flags & MSG_SENDPAGE_NOTLAST))
> - tcp_push(sk, flags, mss_now, tp->nonagle, size_goal);
> + if (copied) {
> + tcp_tx_timestamp(sk, sk->sk_tsflags,
> tcp_write_queue_tail(sk));
> + if (!(flags & MSG_SENDPAGE_NOTLAST))
> + tcp_push(sk, flags, mss_now, tp->nonagle, size_goal);
> + }
> return copied;
>
> do_error:
> @@ -1281,7 +1282,6 @@ int tcp_sendmsg(struct sock *sk, struct msghdr *msg,
> size_t size)
>
> copied += copy;
> if (!msg_data_left(msg)) {
> - tcp_tx_timestamp(sk, sockc.tsflags, skb);
> if (unlikely(flags & MSG_EOR))
> TCP_SKB_CB(skb)->eor = 1;
> goto out;
> @@ -1312,8 +1312,10 @@ int tcp_sendmsg(struct sock *sk, struct msghdr *msg,
> size_t size)
> }
>
> out:
> - if (copied)
> + if (copied) {
> + tcp_tx_timestamp(sk, sockc.tsflags, tcp_write_queue_tail(sk));
> tcp_push(sk, flags, mss_now, tp->nonagle, size_goal);
> + }
> out_nopush:
> release_sock(sk);
> return copied + copied_syn;
>
>
>
