[From nobody Sun May 29 05:38:44 2016 Received: from zmta01.collab.prod.int.phx2.redhat.com (LHLO zmta01.collab.prod.int.phx2.redhat.com) (10.5.81.8) by zmail21.collab.prod.int.phx2.redhat.com with LMTP; Mon, 2 Nov 2015 17:27:32 -0500 (EST) Received: from int-mx09.intmail.prod.int.phx2.redhat.com (int-mx09.intmail.prod.int.phx2.redhat.com [10.5.11.22]) by zmta01.collab.prod.int.phx2.redhat.com (Postfix) with ESMTP id CA5EC18550F for <thaller@mail.corp.redhat.com>; Mon, 2 Nov 2015 17:27:32 -0500 (EST) Received: from mx1.redhat.com (ext-mx07.extmail.prod.ext.phx2.redhat.com [10.5.110.31]) by int-mx09.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id tA2MRWX4015797 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO) for <thaller@redhat.com>; Mon, 2 Nov 2015 17:27:32 -0500 Received: from mail-pa0-f46.google.com (mail-pa0-f46.google.com [209.85.220.46]) by mx1.redhat.com (Postfix) with ESMTPS id 9CBE5C0A160F for <thaller@redhat.com>; Mon, 2 Nov 2015 22:27:31 +0000 (UTC) Received: by pacfv9 with SMTP id fv9so167662882pac.3 for <thaller@redhat.com>; Mon, 02 Nov 2015 14:27:30 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=cumulusnetworks.com; s=google; h=from:to:cc:subject:date:message-id; bh=r4/HfWD/97ZRpqr1RxqFf1/DA/7VYm4I3fGiA8RMbXk=; b=LBkTxoFRx5RwmfT+T6LyaPJMEIJ3qZYT3CrVEL3BTmCUnTUuggdzuTKoX3wytbrFXh us4reZ4YiZDe+iGFdDEUtn08bgAurWSTl9xhm48x5/hP0ECVDRWe86mIQ1nIXbiVEbz7 fgrA0nMfMvjKj6q9ghsqKtJQ59iK2zDndiGsQ= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:from:to:cc:subject:date:message-id; bh=r4/HfWD/97ZRpqr1RxqFf1/DA/7VYm4I3fGiA8RMbXk=; b=F/uzN5Cg4V0Yp88aB9dIS9DmPqEworY6neqinbdahD+T8FyvrPCuP7ekVnIEtVlBs/ td1EzHzwTM2CgBgPI5Q1n9Hgg30qxehSVteH/CQtoAzoD2AR/yE4WNcOOwIUL52tdkQ/ m1OaWB9Q3lbMlpKjjkNN8VeRtWfUnfNHRoKqM6bxes61Og16hy5tvZIAYRLXKjyXF4CR ek9gfDD43gFozuKHBYXwrIp9nxIg03aIcZgxnIGTLEWDFHO4NbztbLf06AyhTydg1KxJ 7wA5ouZSMcfC/CT+BV6IQM/SOTYkDyTAVOSqHGzHnIbiMBEkPXRwPdZ7IoYu+TzmxhwE XH6Q== X-Gm-Message-State: ALoCoQlNtDrcSbuV4e2hMXftE4aabUnzwzUxMdv38psWXJIctYG9qHp6uuLkBtXT3ol4fTU5w3Fy X-Received: by 10.68.65.104 with SMTP id w8mr29694479pbs.48.1446501940633; Mon, 02 Nov 2015 14:05:40 -0800 (PST) Received: from hydra-01.cumulusnetworks.com ([216.129.126.126]) by smtp.googlemail.com with ESMTPSA id bz1sm25892966pab.20.2015.11.02.14.05.39 (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Mon, 02 Nov 2015 14:05:40 -0800 (PST) From: Roopa Prabhu <roopa@cumulusnetworks.com> X-Google-Original-From: Roopa Prabhu To: libnl@lists.infradead.org Cc: thaller@redhat.com, roopa@cumulusnetworks.com Subject: [PATCH 6/6] route: cache and object changes to support non-exclusive and append routes Date: Mon, 2 Nov 2015 14:05:29 -0800 Message-Id: <1446501929-32003-7-git-send-email-roopa@cumulusnetworks.com> X-RedHat-Spam-Score: 2.08 ** (ANY_BOUNCE_MESSAGE, BAYES_50, BOUNCE_MESSAGE, DCC_REPUT_13_19, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DSN_NO_MIMEVERSION, RCVD_IN_DNSWL_LOW, RCVD_IN_MSPIKE_H3, RCVD_IN_MSPIKE_WL, URIBL_BLOCKED) 209.85.220.46 mail-pa0-f46.google.com 209.85.220.46 mail-pa0-f46.google.com <> X-Scanned-By: MIMEDefang 2.68 on 10.5.11.22 X-Scanned-By: MIMEDefang 2.78 on 10.5.110.31 Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable From: Roopa Prabhu <roopa@cumulusnetworks.com> Problem (ipv4 only): Todays libnl route cache looks at prefix + tos + priority to lookup a route object. To support route append operation, where routes with same prefix + tos + priority but different nexthop information can co-exist, we need to also look at nexthop info. Else we will wrongly store only one route for all appended routes. This happens Because the libnl cache inclusion process looks up a route by prefix + tos + priority and replaces it with the new object with the same prefix + tos + priority. Only adding nexthop attribute during lookup does not solve the whole problem. Because NLM_F_REPLACE of objects needs special handling. This patch implements route cache callback .co_cache_search_attrs_get and route object callback .oo_hash_attrs_get to return appropriate attributes for searching route objects depending on type of route and the netlink mess= age flags (NLM_F_APPEND or NLM_F_REPLACE). This is used during cache inclusion process. Also adds ROUTE_ATTR_MULTIPATH to the list of route attribute keys= to search. Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com> --- include/netlink/route/route.h | 8 ++++++ lib/route/route.c | 21 ++++++++++++++ lib/route/route_obj.c | 64 ++++++++++++++++++++++++++++++++-----= ---- 3 files changed, 79 insertions(+), 14 deletions(-) diff --git a/include/netlink/route/route.h b/include/netlink/route/route.h index 477250d..5ab17fe 100644 --- a/include/netlink/route/route.h +++ b/include/netlink/route/route.h @@ -46,6 +46,12 @@ struct rtnl_rtcacheinfo uint32_t rtci_tsage; }; =20 +#define rtnl_route_is_likely_ipv6_multipath(route) \ + (route->rt_family =3D=3D AF_INET6 && \ + route->rt_table !=3D RT_TABLE_LOCAL && \ + rtnl_route_get_nnexthops(route) =3D=3D 1 && \ + rtnl_route_nh_get_gateway(rtnl_route_nexthop_n(route, 0))) + extern struct nl_object_ops route_obj_ops; =20 extern struct rtnl_route * rtnl_route_alloc(void); @@ -97,6 +103,8 @@ extern int rtnl_route_get_src_len(struct rtnl_route *); =20 extern void rtnl_route_add_nexthop(struct rtnl_route *, struct rtnl_nexthop *); +extern void rtnl_route_add_nexthop_head(struct rtnl_route *, + struct rtnl_nexthop *); extern void rtnl_route_remove_nexthop(struct rtnl_route *, struct rtnl_nexthop *); extern struct nl_list_head *rtnl_route_get_nexthops(struct rtnl_route *); diff --git a/lib/route/route.c b/lib/route/route.c index 2985187..cb9d243 100644 --- a/lib/route/route.c +++ b/lib/route/route.c @@ -27,6 +27,26 @@ =20 static struct nl_cache_ops rtnl_route_ops; =20 +uint32_t route_cache_search_attr_get(struct nl_cache *cache, + struct nl_object *needle) +{ + struct nl_object_ops *ops =3D needle->ce_ops; + struct rtnl_route *new_route =3D (struct rtnl_route *)needle; + + /* + * If route add req is a replace + * or it is a ipv6 multipath route + * then, we replace the existing object + * in the cache. So, look for the first object + * that matches the hash key attributes + */ + if ((needle->ce_msgflags & NLM_F_REPLACE) || + rtnl_route_is_likely_ipv6_multipath(new_route)) + return ops->oo_hash_attrs_get(needle); + else + return ops->oo_id_attrs_get(needle); +} + static int route_msg_parser(struct nl_cache_ops *ops, struct sockaddr_nl *= who, struct nlmsghdr *nlh, struct nl_parser_param *pp) { @@ -191,6 +211,7 @@ static struct nl_cache_ops rtnl_route_ops =3D { .co_request_update =3D route_request_update, .co_msg_parser =3D route_msg_parser, .co_obj_ops =3D &route_obj_ops, + .co_cache_search_attrs_get =3D route_cache_search_attr_get, }; =20 static void __init route_init(void) diff --git a/lib/route/route_obj.c b/lib/route/route_obj.c index 1d7ce57..7a4ea63 100644 --- a/lib/route/route_obj.c +++ b/lib/route/route_obj.c @@ -294,6 +294,26 @@ static void route_dump_stats(struct nl_object *obj, st= ruct nl_dump_params *p) } } =20 +static uint32_t route_id_attrs_get(struct nl_object *obj) +{ + struct nl_object_ops *ops =3D obj->ce_ops; + + if (!ops) + BUG(); + + return ops->oo_id_attrs | ROUTE_ATTR_MULTIPATH; +} + +static uint32_t route_hash_attrs_get(struct nl_object *obj) +{ + struct nl_object_ops *ops =3D obj->ce_ops; + + if (!ops) + BUG(); + + return ops->oo_id_attrs; +} + static void route_keygen(struct nl_object *obj, uint32_t *hashkey, uint32_t table_sz) { @@ -463,15 +483,9 @@ static int route_update(struct nl_object *old_obj, str= uct nl_object *new_obj) * route with multiple nexthops. The resulting object looks * similar to a ipv4 ECMP route */ - if (new_route->rt_family !=3D AF_INET6 || - new_route->rt_table =3D=3D RT_TABLE_LOCAL) - return -NLE_OPNOTSUPP; =20 - /* - * For routes that are already multipath, - * or dont have a nexthop dont do anything - */ - if (rtnl_route_get_nnexthops(new_route) !=3D 1) + + if (!rtnl_route_is_likely_ipv6_multipath(new_route)) return -NLE_OPNOTSUPP; =20 /* @@ -480,19 +494,31 @@ static int route_update(struct nl_object *old_obj, st= ruct nl_object *new_obj) * filled or nothing at all */ new_nh =3D rtnl_route_nexthop_n(new_route, 0); - if (!new_nh || !rtnl_route_nh_get_gateway(new_nh)) - return -NLE_OPNOTSUPP; =20 switch(action) { case RTM_NEWROUTE : { - struct rtnl_nexthop *cloned_nh; + struct rtnl_nexthop *cloned_nh, *first_nh; =20 - /* - * Add the nexthop to old route - */ cloned_nh =3D rtnl_route_nh_clone(new_nh); if (!cloned_nh) return -NLE_NOMEM; + + if (new_obj->ce_msgflags & NLM_F_REPLACE) { + /* If replace flag, we should replace the first + * nexthop with the new one + */ + first_nh =3D rtnl_route_nexthop_n(old_route, 0); + if (first_nh) + rtnl_route_remove_nexthop(old_route, + first_nh); + rtnl_route_add_nexthop_head(old_route, cloned_nh); + + return 0; + } + + /* + * Add the nexthop to old route + */ rtnl_route_add_nexthop(old_route, cloned_nh); =20 NL_DBG(2, "Route obj %p updated. Added " @@ -838,6 +864,14 @@ void rtnl_route_add_nexthop(struct rtnl_route *route, = struct rtnl_nexthop *nh) route->ce_mask |=3D ROUTE_ATTR_MULTIPATH; } =20 +void rtnl_route_add_nexthop_head(struct rtnl_route *route, + struct rtnl_nexthop *nh) +{ + nl_list_add_head(&nh->rtnh_list, &route->rt_nexthops); + route->rt_nr_nh++; + route->ce_mask |=3D ROUTE_ATTR_MULTIPATH; +} + void rtnl_route_remove_nexthop(struct rtnl_route *route, struct rtnl_nexth= op *nh) { if (route->ce_mask & ROUTE_ATTR_MULTIPATH) { @@ -1310,6 +1344,8 @@ struct nl_object_ops route_obj_ops =3D { .oo_id_attrs =3D (ROUTE_ATTR_FAMILY | ROUTE_ATTR_TOS | ROUTE_ATTR_TABLE | ROUTE_ATTR_DST | ROUTE_ATTR_PRIO), + .oo_id_attrs_get =3D route_id_attrs_get, + .oo_hash_attrs_get =3D route_hash_attrs_get }; /** @endcond */ =20 --=20 1.7.10.4 ]