Re: [PATCH net-next] tcp: batch calls to sk_flush_backlog()

2019-08-09 Thread David Miller
From: Eric Dumazet Date: Fri, 9 Aug 2019 05:04:47 -0700 > Starting from commit d41a69f1d390 ("tcp: make tcp_sendmsg() aware of socket > backlog") > loopback flows got hurt, because for each skb sent, the socket receives an > immediate ACK and sk_flush_backlog() causes extra work. > > Intent wa

Re: [PATCH net-next] tcp: batch calls to sk_flush_backlog()

2019-08-09 Thread Soheil Hassas Yeganeh
On Fri, Aug 9, 2019 at 8:04 AM Eric Dumazet wrote: > > Starting from commit d41a69f1d390 ("tcp: make tcp_sendmsg() aware of socket > backlog") > loopback flows got hurt, because for each skb sent, the socket receives an > immediate ACK and sk_flush_backlog() causes extra work. > > Intent was to n

[PATCH net-next] tcp: batch calls to sk_flush_backlog()

2019-08-09 Thread Eric Dumazet
Starting from commit d41a69f1d390 ("tcp: make tcp_sendmsg() aware of socket backlog") loopback flows got hurt, because for each skb sent, the socket receives an immediate ACK and sk_flush_backlog() causes extra work. Intent was to not let the backlog grow too much, but we went a bit too far. We