On 11/21/2018 02:40 PM, Yuchung Cheng wrote: > On Wed, Nov 21, 2018 at 9:52 AM, Eric Dumazet <eduma...@google.com> wrote: >> Under high stress, and if GRO or coalescing does not help, >> we better make room in backlog queue to be able to keep latest >> packet coming. >> >> This generally helps fast recovery, given that we often receive >> packets in order. > > I like the benefit of fast recovery but I am a bit leery about head > drop causing HoLB on large read, while tail drops can be repaired by > RACK and TLP already. Hmm - This is very different pattern here. We have a train of packets coming, the last packet is not a TLP probe... Consider this train coming from an old stack without burst control nor pacing. This patch guarantees last packet will be processed, and either : 1) We are a receiver, we will send a SACK. Sender will typically start recovery 2) We are a sender, we will process the most recent ACK sent by the receiver.
- [PATCH net-next 0/3] tcp: take a bit more care of backlog st... Eric Dumazet
- [PATCH net-next 1/3] tcp: remove hdrlen argument from t... Eric Dumazet
- Re: [PATCH net-next 1/3] tcp: remove hdrlen argumen... Yuchung Cheng
- [PATCH net-next 3/3] tcp: implement head drops in backl... Eric Dumazet
- Re: [PATCH net-next 3/3] tcp: implement head drops ... Yuchung Cheng
- Re: [PATCH net-next 3/3] tcp: implement head dr... Eric Dumazet
- Re: [PATCH net-next 3/3] tcp: implement hea... Yuchung Cheng
- Re: [PATCH net-next 3/3] tcp: implemen... Eric Dumazet
- Re: [PATCH net-next 3/3] tcp: impl... Eric Dumazet
- Re: [PATCH net-next 3/3] tcp: ... Yuchung Cheng
- Re: [PATCH net-next 3/3] tcp: ... Eric Dumazet
- [PATCH net-next 2/3] tcp: implement coalescing on backl... Eric Dumazet
- Re: [PATCH net-next 2/3] tcp: implement coalescing ... Yuchung Cheng
- Re: [PATCH net-next 2/3] tcp: implement coalesc... Eric Dumazet
- Re: [PATCH net-next 2/3] tcp: implement coa... Yuchung Cheng
- Re: [PATCH net-next 2/3] tcp: implement coalescing ... Neal Cardwell