From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Ahern Subject: [PATCH net-next v2 2/2] net: l3mdev: prefer VRF master for source address selection Date: Mon, 22 Feb 2016 10:23:51 -0800 Message-ID: <1456165431-867-2-git-send-email-dsa@cumulusnetworks.com> References: <1456165431-867-1-git-send-email-dsa@cumulusnetworks.com> Cc: david@opensourcerouting.org, David Lamparter , David Ahern To: netdev@vger.kernel.org Return-path: Received: from mail-pa0-f51.google.com ([209.85.220.51]:36017 "EHLO mail-pa0-f51.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755340AbcBVSYA (ORCPT ); Mon, 22 Feb 2016 13:24:00 -0500 Received: by mail-pa0-f51.google.com with SMTP id yy13so94954465pab.3 for ; Mon, 22 Feb 2016 10:24:00 -0800 (PST) In-Reply-To: <1456165431-867-1-git-send-email-dsa@cumulusnetworks.com> Sender: netdev-owner@vger.kernel.org List-ID: From: David Lamparter When selecting an address in context of a VRF, the vrf master should be preferred for address selection. If it isn't, the user has a hard time getting the system to select to their preference - the code will pick the address off the first in-VRF interface it can find, which on a router could well be a non-routable address. Signed-off-by: David Lamparter Signed-off-by: David Ahern [dsa: Fixed comment style ] --- v2 - Fixed comment per Dave's response net/ipv4/devinet.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 614904c29cbd..01119fbf3b30 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -1217,6 +1217,24 @@ __be32 inet_select_addr(const struct net_device *dev, __be32 dst, int scope) no_in_dev: master_idx = l3mdev_master_ifindex_rcu(dev); + /* For VRFs, the VRF device takes the place of the loopback device, + * with addresses on it being preferred. Note in such cases the + * loopback device will be among the devices that fail the master_idx + * equality check in the loop below. + */ + if (master_idx && + (dev = dev_get_by_index_rcu(net, master_idx)) && + (in_dev = __in_dev_get_rcu(dev))) { + + for_primary_ifa(in_dev) { + if (ifa->ifa_scope != RT_SCOPE_LINK && + ifa->ifa_scope <= scope) { + addr = ifa->ifa_local; + goto out_unlock; + } + } endfor_ifa(in_dev); + } + /* Not loopback addresses on loopback should be preferred in this case. It is important that lo is the first interface in dev_base list. -- 2.1.4