From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH net-next 03/13] net: sched: extend Qdisc with rcu Date: Thu, 6 Sep 2018 02:21:53 -0700 Message-ID: <5809e3d4-78cc-1ccf-53e2-fa52e97d8607@gmail.com> References: <1536220742-25650-1-git-send-email-vladbu@mellanox.com> <1536220742-25650-4-git-send-email-vladbu@mellanox.com> <4a230727-1072-3bc9-1590-2e59c2162460@virtuozzo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Cc: jhs@mojatatu.com, xiyou.wangcong@gmail.com, jiri@resnulli.us, davem@davemloft.net, stephen@networkplumber.org, paulmck@linux.vnet.ibm.com, nicolas.dichtel@6wind.com, leon@kernel.org, gregkh@linuxfoundation.org, mark.rutland@arm.com, fw@strlen.de, dsahern@gmail.com, lucien.xin@gmail.com, jakub.kicinski@netronome.com, christian.brauner@ubuntu.com, jbenc@redhat.com To: Kirill Tkhai , Eric Dumazet , Vlad Buslov , netdev@vger.kernel.org Return-path: Received: from mail-wr1-f65.google.com ([209.85.221.65]:44633 "EHLO mail-wr1-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727962AbeIFN4a (ORCPT ); Thu, 6 Sep 2018 09:56:30 -0400 Received: by mail-wr1-f65.google.com with SMTP id v16-v6so10538878wro.11 for ; Thu, 06 Sep 2018 02:21:57 -0700 (PDT) In-Reply-To: <4a230727-1072-3bc9-1590-2e59c2162460@virtuozzo.com> Content-Language: en-US Sender: netdev-owner@vger.kernel.org List-ID: On 09/06/2018 01:39 AM, Kirill Tkhai wrote: > On 06.09.2018 11:30, Eric Dumazet wrote: >> >> >> On 09/06/2018 12:58 AM, Vlad Buslov wrote: >> >> ... >> >>> diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h >>> index 18e22a5a6550..239c73f29471 100644 >>> --- a/include/net/sch_generic.h >>> +++ b/include/net/sch_generic.h >>> @@ -90,6 +90,7 @@ struct Qdisc { >>> struct gnet_stats_queue __percpu *cpu_qstats; >>> int padded; >>> refcount_t refcnt; >>> + struct rcu_head rcu; >>> >>> /* >>> * For performance sake on SMP, we put highly modified fields at the end >> >> Probably better to move this at the end of struct Qdisc, >> not risking unexpected performance regressions in fast path. > > Do you mean regressions on UP? On SMP it looks like this field > fits in the unused gap created by: > > struct sk_buff_head gso_skb ____cacheline_aligned_in_smp; > > Kirill Oh right, we might have holes there, and in the last cache line, so it does not matter for now.