Re: [PATCH net-next 2/2] tcp: do not aggressively quick ack after ECN events

2018-05-22 Thread Neal Cardwell
On Mon, May 21, 2018 at 6:09 PM Eric Dumazet wrote: > ECN signals currently forces TCP to enter quickack mode for > up to 16 (TCP_MAX_QUICKACKS) following incoming packets. > We believe this is not needed, and only sending one immediate ack > for the current packet should be enough. > This shou

Re: [PATCH net-next 2/2] tcp: do not aggressively quick ack after ECN events

2018-05-22 Thread Yuchung Cheng
On Mon, May 21, 2018 at 3:08 PM, Eric Dumazet wrote: > ECN signals currently forces TCP to enter quickack mode for > up to 16 (TCP_MAX_QUICKACKS) following incoming packets. > > We believe this is not needed, and only sending one immediate ack > for the current packet should be enough. > > This sh

Re: [PATCH net-next 2/2] tcp: do not aggressively quick ack after ECN events

2018-05-22 Thread Soheil Hassas Yeganeh
On Mon, May 21, 2018 at 6:08 PM, Eric Dumazet wrote: > ECN signals currently forces TCP to enter quickack mode for > up to 16 (TCP_MAX_QUICKACKS) following incoming packets. > > We believe this is not needed, and only sending one immediate ack > for the current packet should be enough. > > This sh

[PATCH net-next 2/2] tcp: do not aggressively quick ack after ECN events

2018-05-21 Thread Eric Dumazet
ECN signals currently forces TCP to enter quickack mode for up to 16 (TCP_MAX_QUICKACKS) following incoming packets. We believe this is not needed, and only sending one immediate ack for the current packet should be enough. This should reduce the extra load noticed in DCTCP environments, after co