From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.1 required=3.0 tests=DKIM_SIGNED,DKIM_VALID, DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH,MAILING_LIST_MULTI, SIGNED_OFF_BY,SPF_PASS,URIBL_BLOCKED autolearn=unavailable autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id D94B4C10F0E for ; Mon, 15 Apr 2019 19:25:51 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 988BA2075B for ; Mon, 15 Apr 2019 19:25:51 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=cumulusnetworks.com header.i=@cumulusnetworks.com header.b="fi9JWwAY" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1730802AbfDOTZu (ORCPT ); Mon, 15 Apr 2019 15:25:50 -0400 Received: from mail-pg1-f193.google.com ([209.85.215.193]:33661 "EHLO mail-pg1-f193.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1729675AbfDOTFJ (ORCPT ); Mon, 15 Apr 2019 15:05:09 -0400 Received: by mail-pg1-f193.google.com with SMTP id k19so9049888pgh.0 for ; Mon, 15 Apr 2019 12:05:08 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cumulusnetworks.com; s=google; h=subject:to:cc:references:from:message-id:date:user-agent :mime-version:in-reply-to:content-language:content-transfer-encoding; bh=u5cVf37ZSBSrmFmZ1QHquBZ0dAcfuAsJNF/mAhYOQi0=; b=fi9JWwAYPBjrScJAmLrkQZtHlLowhuti0lrs8I2kqvgQ6U1kUEufnqFmfndrEYMni0 stuoRS4mi2JBTA5q0ToURWg5gvto2dyhtzJ4YCCrVm3cQOG8mwYtzmdKlpyQF3VTMPvk 5E43X/++Bq697ACGKoPC/+DnDgrY8FgVkEElE= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:subject:to:cc:references:from:message-id:date :user-agent:mime-version:in-reply-to:content-language :content-transfer-encoding; bh=u5cVf37ZSBSrmFmZ1QHquBZ0dAcfuAsJNF/mAhYOQi0=; b=RKhmZvBe5bbrvQItNL0QmadDiGPE+VzsrxShHk+wMCsKcZYZPoRV/xa6S6o1wRODWz DWcjIfXBgk0mqoDqtffCSjlxliBNFz+IoLHXBn1wLm1MnwZhZ3hOWU+IXbmRq3DeZFgn s5DjDpk3iyrMGLqrdK43EzuYEIZaWWqtr9ugkvHaZRM4/dlZbESj0QuTQWMuMWnxyFHW Z6d2CUUUwLfRXfJF/973oOQegEnCbuMLl1o6Hh4s+YCIKbBCqL0lqBvuzWvklPYPCGWJ GAO1Y3lfY+GkA7niBE9CTs1Eow/Jr/oowcmSzVgbhmgy+3NPtl4CNFcafon+vHDk01t/ ZWKQ== X-Gm-Message-State: APjAAAWxm/LHFklWKeFD4Ni+bLW8jCTuJxhu5ZzONb0p55Th6thm1TJb Bx93uXGnaWhjNouNdpOEQSKslw== X-Google-Smtp-Source: APXvYqwN4nJZSJxXA1fDFjhAIfOWCsn2mSlaDELFUJtm8cFX733Mn0I5ZhVT9aqj/IcI9+bz417BqA== X-Received: by 2002:a63:af0a:: with SMTP id w10mr47197612pge.67.1555355108159; Mon, 15 Apr 2019 12:05:08 -0700 (PDT) Received: from ?IPv6:2601:282:800:fd80:91ac:460f:1e02:116c? ([2601:282:800:fd80:91ac:460f:1e02:116c]) by smtp.googlemail.com with ESMTPSA id l184sm119889775pfc.98.2019.04.15.12.05.06 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Mon, 15 Apr 2019 12:05:07 -0700 (PDT) Subject: Re: [PATCH net] net: vrf: remove redundant vrf neigh entry To: linmiaohe , Shrijeet Mukherjee , "David S. Miller" , netdev@vger.kernel.org, linux-kernel@vger.kernel.org Cc: Mingfangsen References: From: David Ahern Message-ID: Date: Mon, 15 Apr 2019 13:05:06 -0600 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.14; rv:52.0) Gecko/20100101 Thunderbird/52.9.1 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 4/10/19 9:39 PM, linmiaohe wrote: > > On 2019/3/22 23:50, David Ahern wrote: >> On 3/22/19 3:10 PM, linmiaohe wrote: >>> From: Miaohe Lin >>> >>> When vrf->rth is created, it wouldn't change in his lifetime.And in >>> func vrf_finish_output, we always create a neigh with ip_hdr(skb)->daddr >>> as key because rth->rt_gateway is equal to 0. But I think we only need >>> one vrf neigh entry because we strip the ethernet header and reset the >>> dst_entry in vrf_process_v4_outbound. >>> So I set rth->rt_gateway to loopback addr(It's ok to set any other >>> valid ip address, just choose one.). And we would only create one vrf >>> neigh entry. This helps saving some memory and improving the hitting >>> rate of neigh lookup. >>> If there is something I misunderstand, it's very nice of you to >>> let me know. Thanks a lot. >>> >>> Signed-off-by: linmiaohe >>> --- >>> drivers/net/vrf.c | 1 + >>> 1 file changed, 1 insertion(+) >>> >>> diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c >>> index 7c1430ed0244..2b0227fb8f53 100644 >>> --- a/drivers/net/vrf.c >>> +++ b/drivers/net/vrf.c >>> @@ -738,6 +738,7 @@ static int vrf_rtable_create(struct net_device *dev) >>> return -ENOMEM; >>> >>> rth->dst.output = vrf_output; >>> + rth->rt_gateway = htonl(INADDR_LOOPBACK); >>> >>> rcu_assign_pointer(vrf->rth, rth); >>> >> >> Did you investigate how neighbor entries are getting created? The vrf >> device has IFF_NOARP set, so neigh entries should not be created. >> >> . >> > Hi,David A.,I investigate how neighbor entries are getting created recently. > But I can't find where neigh entries is not created when vrf device has > IFF_NOARP set. > So I add some printk info,and I ping the different host, here is the output: > > [root@localhost ~]# ip vrf exec vrf1 ping 10.0.0.2 > PING 10.0.0.2 (10.0.0.2) 56(84) bytes of data. > 64 bytes from 10.0.0.2: icmp_seq=1 ttl=64 time=1.78 ms > ^C > --- 10.0.0.2 ping statistics --- > 1 packets transmitted, 1 received, 0% packet loss, time 0ms > rtt min/avg/max/mdev = 1.776/1.776/1.776/0.000 ms > [root@localhost ~]# ip vrf exec vrf1 ping 11.0.0.2 > PING 11.0.0.2 (11.0.0.2) 56(84) bytes of data. > 64 bytes from 11.0.0.2: icmp_seq=1 ttl=64 time=1.59 ms > ^C > --- 11.0.0.2 ping statistics --- > 1 packets transmitted, 1 received, 0% packet loss, time 0ms > rtt min/avg/max/mdev = 1.591/1.591/1.591/0.000 ms > [root@localhost ~]# ip vrf exec vrf1 ping 11.0.0.3 > PING 11.0.0.3 (11.0.0.3) 56(84) bytes of data. > ^C > --- 11.0.0.3 ping statistics --- > 1 packets transmitted, 0 received, 100% packet loss, time 0ms > > Apr 11 11:01:48 localhost kernel: [ 337.311270] VRF: IFF_NOARP is set > Apr 11 11:01:48 localhost kernel: [ 337.311279] VRF: nexthop = 200000a > Apr 11 11:01:48 localhost kernel: [ 337.311284] VRF: neigh = (null) after lookup > Apr 11 11:01:48 localhost kernel: [ 337.311294] VRF: we create a neigh 000000001e8acd79 > Apr 11 11:01:51 localhost kernel: [ 340.026623] VRF: IFF_NOARP is set > Apr 11 11:01:51 localhost kernel: [ 340.026627] VRF: nexthop = 200000b > Apr 11 11:01:51 localhost kernel: [ 340.026631] VRF: neigh = (null) after lookup > Apr 11 11:01:51 localhost kernel: [ 340.026637] VRF: we create a neigh 00000000a0ad96da > Apr 11 11:01:56 localhost kernel: [ 345.157529] VRF: IFF_NOARP is set > Apr 11 11:01:56 localhost kernel: [ 345.157539] VRF: nexthop = 300000b > Apr 11 11:01:56 localhost kernel: [ 345.157544] VRF: neigh = (null) after lookup > Apr 11 11:01:56 localhost kernel: [ 345.157556] VRF: we create a neigh 00000000a5167b56 > > And here is the printk code: > > if (vrf_dev->flags & IFF_NOARP) { > printk(KERN_ERR "VRF: IFF_NOARP is set\n"); > rth = rcu_dereference(vrf->rth); > nexthop = (__force u32)rt_nexthop(rth, ip_hdr(skb)->daddr); > printk(KERN_ERR "VRF: nexthop = %x\n", nexthop); > neigh = __ipv4_neigh_lookup_noref(vrf_dev, nexthop); > printk(KERN_ERR "VRF: neigh = %p after lookup\n", (void *)neigh); > if (unlikely(!neigh)) { > neigh = __neigh_create(&arp_tbl, &nexthop, vrf_dev, false); > printk(KERN_ERR "VRF: we create a neigh %p\n", (void *)neigh); > } > } > > Could you please tell me if I was misunderstanding something again? It's very nice > of you if you can figure me out that. Thanks a lot.I am looking forward to your reply. > In the above statements you are passing vrf_dev to neigh_create, so of course it is creating an entry against that device. That should not be happening. When dst->dev is a vrf device it is for local traffic only (locally originated to a local address) and there are no neighbor entries for that. Otherwise, it is an enslaved device and the neigh_create is always done with it.