Re: [PATCH v2 net-next] tcp: fix slow start after idle vs TSO/GSO

2015-08-25 Thread David Miller
From: Eric Dumazet Date: Fri, 21 Aug 2015 12:30:00 -0700 > From: Eric Dumazet > > slow start after idle might reduce cwnd, but we perform this > after first packet was cooked and sent. > > With TSO/GSO, it means that we might send a full TSO packet > even if cwnd should have been reduced to IW

Re: [PATCH v2 net-next] tcp: fix slow start after idle vs TSO/GSO

2015-08-21 Thread Neal Cardwell
On Fri, Aug 21, 2015 at 3:30 PM, Eric Dumazet wrote: > From: Eric Dumazet > > slow start after idle might reduce cwnd, but we perform this > after first packet was cooked and sent. > > With TSO/GSO, it means that we might send a full TSO packet > even if cwnd should have been reduced to IW10. > >

[PATCH v2 net-next] tcp: fix slow start after idle vs TSO/GSO

2015-08-21 Thread Eric Dumazet
From: Eric Dumazet slow start after idle might reduce cwnd, but we perform this after first packet was cooked and sent. With TSO/GSO, it means that we might send a full TSO packet even if cwnd should have been reduced to IW10. Moving the SSAI check in skb_entail() makes sense, because we slight