Re: [PATCH net-next 00/17] tcp/dccp: lockless listener

2015-10-03 Thread Eric Dumazet
On Sat, 2015-10-03 at 04:48 -0700, David Miller wrote: > This looks amazing, series applied, thanks Eric! Thanks David. While splitting my patches, it seems I somehow missed one change, adding SLAB_DESTROY_BY_RCU flag for request kmem_cache. I will send the patch in few minutes. -- To

Re: [PATCH net-next 00/17] tcp/dccp: lockless listener

2015-10-03 Thread David Miller
From: Eric Dumazet Date: Fri, 2 Oct 2015 11:43:22 -0700 > TCP listener refactoring : this is becoming interesting ! > > This patch series takes the steps to use normal TCP/DCCP ehash > table to store SYN_RECV requests, instead of the private per-listener > hash table we

[PATCH net-next 00/17] tcp/dccp: lockless listener

2015-10-02 Thread Eric Dumazet
TCP listener refactoring : this is becoming interesting ! This patch series takes the steps to use normal TCP/DCCP ehash table to store SYN_RECV requests, instead of the private per-listener hash table we had until now. SYNACK skb are now attached to their syn_recv request socket, so that we no