On Mon, 2017-09-18 at 22:06 -0700, Eric Dumazet wrote:
> From: Eric Dumazet <eduma...@google.com>
...
> --- a/include/net/tcp.h
> +++ b/include/net/tcp.h
> @@ -797,11 +797,6 @@ struct tcp_skb_cb {
> u16 tcp_gso_segs;
> u16 tcp_gso_size;
> };
> -
> - /* Used to stash the receive timestamp while this skb is in the
> - * out of order queue, as skb->tstamp is overwritten by the
> - * rbnode.
> - */
> ktime_t swtstamp;
> };
I forgot that in upstream kernels, swtstamp can also be removed.
It is used in our kernels to store the time at which the skb was moved
into socket receive queue (after a possible stay in out of order queue)
Will send a V2, removing this field from net-next, before we add it
again later if we upstream our instrumentation code.
