- Bird-users - Mailing Lists

auto-cost reference-bandwidth
by Markus Wigge 13 Jun '17

13 Jun '17

Hello, according to a chat I recently had with Elrond I put my question on the mailing list: Is there an option for the OSPF protocol in bird to calculate the interface cost based on the bandwidth like cisco or quagga or juniper or fortigate or ... do? As far as I remember the definition of the interface cost is as follows: 100.000kbps cost = ----------- <bandwidth> The configurable option "auto-cost reference-bandwidth" then manipulates the "100.000". So when you have 40G or 100G links in your network you set the reference to that on all your routers. Regards, Markus

2 1

Re: Order of BGP communities
by Pavlos Parissis 13 Jun '17

13 Jun '17

On 13/06/2017 01:46 μμ, Ondrej Zajicek wrote: > On Wed, Jun 07, 2017 at 03:49:35PM +0200, Pavlos Parissis wrote: >> Hi, >> >> Today, we noticed a difference in the order that communities were added between 1.6.3 and 1.4.5 version. >> >> Does anyone know why ? > > It is true, one reason is that to have that consistent with extended and > long communities. Another reason is that append is a bit more natural, if > you add several communities and then show it, you end with the same > order. > > Also note that order of communities is not significant and in the future > we perhaps end with sorted community attributes. > Thanks for the clarification. Cheers, Pavlos

1 0

Order of BGP communities
by Pavlos Parissis 13 Jun '17

13 Jun '17

Hi, Today, we noticed a difference in the order that communities were added between 1.6.3 and 1.4.5 version. All exported routes pass through a function, which adds three communities: function match_route_north() { bgp_community.add((64813,lookup_live_id(net.ip))); bgp_community.add(DC_ID); bgp_community.add((64813,lookup_fabric_id(net.ip))); return (net ~ ACAST_PS_ADVERTISE && net ~ ANYCAST_NETWORKS_NORTH && !(net ~ ACAST_PS_DISABLED)); } As you see we add first live_id (which is always 1000) and last the fabric_id(which is always 3). With bird 1.6.3 version we see that live_id community is listed 1st: 1.17.1.143/32 dev lo [direct1 2017-06-07 15:30:29 1496842229] * (240) Type: device unicast univ BGP.origin: IGP BGP.as_path: 61115 BGP.next_hop: 10.1.9.1 BGP.local_pref: 0 BGP.community: (64813,1000) (64815,100) (64813,3) but with 1.4.5 version is listed as last: 1.17.16.143/32 dev lo [direct1 2017-03-10 11:24:36 1489141476] * (240) Type: device unicast univ BGP.origin: IGP BGP.as_path: 6115 BGP.next_hop: 10.1.6.1 BGP.local_pref: 0 BGP.community: (64813,3) (61115,100) (64813,1000) Does anyone know why ? Cheers, Pavlos

2 1

[PATCH 0/3] babel: Add compression and support for subtlvs
by Toke Høiland-Jørgensen 09 Jun '17

09 Jun '17

This patch series adds compression of outgoing IPv6 prefixes to the Babel protocol, and also adds support to the parser for parsing subtlvs as per RFC6126 bis (see https://tools.ietf.org/html/draft-ietf-babel-rfc6126bis-02#section-4.4) These patches apply to 2.0.0-pre1, and on top of the previously posted patch that adds dualstack support to Babel. I'll roll that up into this series as well once the patches have been reviewed. Toke Høiland-Jørgensen (3): lib: Add ip6_common_octets function babel: Implement IPv6 prefix compression on outgoing updates babel: Parse subtlvs and skip TLVs with a mandatory subtlv lib/ip.c | 14 +++++++ lib/ip.h | 1 + proto/babel/babel.h | 5 +++ proto/babel/packets.c | 104 +++++++++++++++++++++++++++++++++++++++++++++++--- 4 files changed, 119 insertions(+), 5 deletions(-) -- 2.13.0

2 3

[PATCH v2] babel: Implement IPv6 prefix compression on outgoing updates
by Toke Høiland-Jørgensen 09 Jun '17

09 Jun '17

Previously, the Babel protocol would never use prefix compression on outgoing updates (but would parse it on incoming ones). This adds compression of IPv6 addresses of outgoing updates. The compression only works to the extent that the FIB is walked in lexicographic order; i.e. a prefix is only compressed if it shares bytes with the previous prefix in the same packet. Signed-off-by: Toke Høiland-Jørgensen <toke(a)toke.dk> --- proto/babel/packets.c | 34 +++++++++++++++++++++++++++++++++- 1 file changed, 33 insertions(+), 1 deletion(-) diff --git a/proto/babel/packets.c b/proto/babel/packets.c index 0f70898e..8e014955 100644 --- a/proto/babel/packets.c +++ b/proto/babel/packets.c @@ -133,6 +133,8 @@ struct babel_write_state { u8 router_id_seen; u8 next_hop_v4_seen; u8 next_hop_v6_seen; + u8 def_ip6_prefix[16]; /* Previous IPv6 prefix in network order (for compression) */ + u8 def_ip6_pxlen; }; @@ -150,6 +152,14 @@ struct babel_write_state { #define NET_SIZE(n) BYTES(net_pxlen(n)) +static inline uint bytes_equal(u8 *b1, u8 *b2, uint maxlen) +{ + uint i; + for(i = 0; *b1 == *b2 && i < maxlen; i++, b1++, b2++) + ; + return i; +} + static inline u16 get_time16(const void *p) @@ -682,9 +692,31 @@ babel_write_update(struct babel_tlv *hdr, union babel_msg *m, } else { + u8 omit, buf[16] = {}; tlv->ae = BABEL_AE_IP6; tlv->plen = net6_pxlen(&msg->net); - put_ip6_px(tlv->addr, &msg->net); + + put_ip6_px(buf, &msg->net); + omit = bytes_equal(buf, state->def_ip6_prefix, state->def_ip6_pxlen/8); + DBG("%d bytes common between %I and previous prefix\n", omit, + net6_prefix(&msg->net)); + + if (0 < omit) + { + omit = MIN(omit, tlv->plen/8); + memcpy(tlv->addr, buf+omit, BYTES(tlv->plen) - omit); + + tlv->omitted = omit; + tlv->length -= omit; + len -= omit; + } + else + { + put_ip6_px(tlv->addr, &msg->net); + tlv->flags |= BABEL_FLAG_DEF_PREFIX; + put_ip6_px(state->def_ip6_prefix, &msg->net); + state->def_ip6_pxlen = tlv->plen; + } } put_time16(&tlv->interval, msg->interval); -- 2.13.0

2 1

[PATCH] babel: Add documentation for dual-stack operation and options
by Toke Høiland-Jørgensen 09 Jun '17

09 Jun '17

This updates the documentation for the Babel protocol to mention the fact that it now supports dual-stack operation, and adds documentation for the new next hop options. Signed-off-by: Toke Høiland-Jørgensen <toke(a)toke.dk> --- doc/bird.sgml | 32 ++++++++++++++++++++++++++------ 1 file changed, 26 insertions(+), 6 deletions(-) diff --git a/doc/bird.sgml b/doc/bird.sgml index 01b59c6a..4a9177f5 100644 --- a/doc/bird.sgml +++ b/doc/bird.sgml @@ -1596,11 +1596,12 @@ networks. Babel is conceptually very simple in its operation and "just works" in its default configuration, though some configuration is possible and in some cases desirable. -<p>While the Babel protocol is dual stack (i.e., can carry both IPv4 and IPv6 -routes over the same IPv6 transport), BIRD presently implements only the IPv6 -subset of the protocol. No Babel extensions are implemented, but the BIRD -implementation can coexist with implementations using the extensions (and will -just ignore extension messages). +<p>The Babel protocol is dual stack; i.e., it can carry both IPv4 and IPv6 +routes over the same IPv6 transport. For sending and receiving Babel packets, +only a link-local IPv6 address is needed. + +<p>Bird does not implement any Babel extensions, but will coexist with +implementations using extensions (and will just ignore extension messages). <p>The Babel protocol implementation in BIRD is currently in alpha stage. @@ -1623,6 +1624,8 @@ protocol babel [<name>] { rx buffer <number>; tx length <number>; check link <switch>; + next hop ipv4 <address>; + next hop ipv6 <address>; }; } </code> @@ -1680,6 +1683,18 @@ protocol babel [<name>] { routes received from them are withdrawn. It is possible that some hardware drivers or platforms do not implement this feature. Default: yes. + + <tag><label id="babel-next-hop-ipv4">next hop ipv4 <m/address/</tag> + Set the next hop address advertised for IPv4 routes advertised on this + interface. If not set, the first IPv4 address found on the interface will + be used, so it should only be necessary to set this option if this + auto-detection fails or finds the wrong address. + + <tag><label id="babel-next-hop-ipv6">next hop ipv6 <m/address/</tag> + Set the next hop address advertised for IPv6 routes advertised on this + interface. If not set, the same link-local address that is used as the + source for Babel packets will be used. In normal operation, it should not + be necessary to set this option. </descrip> <sect1>Attributes @@ -1708,7 +1723,12 @@ protocol babel { # configured on local interfaces, plus re-distribute all routes received # from other babel peers. - export where (source = RTS_DEVICE) || (source = RTS_BABEL); + ipv4 { + export where (source = RTS_DEVICE) || (source = RTS_BABEL); + }; + ipv6 { + export where (source = RTS_DEVICE) || (source = RTS_BABEL); + }; } </code> -- 2.13.0

2 1

Filters and command
by Anibe Onuche 08 Jun '17

08 Jun '17

Hello All, Top of the day greetings. Some questions 1.How do I input a command that shows what the bird route server is a. Advertising to a particular neighbor b. Receiving from a particular neighbor 2.How do I implement a filter that stop the bird server from advertising an ASN from peer A to another peer B. Peer B does not want to receive an ASN from Peer A from the bird server but wants to receive it through it upstream Thanks. Warm Regards A.O

3 2

[PATCH] babel: Add support for dual-stack v4/v6 operation
by Toke Høiland-Jørgensen 08 Jun '17

08 Jun '17

This adds support for dual-stack v4/v6 operation to the Babel protocol. Routing messages will be exchanged over IPv6, but IPv4 routes can be carried in the messages being exchanged. This matches how the reference Babel implementation (babeld) works. The nexthop address for v4 can be configured per interface, and will default to the first available IPv4 address on the given interface. For symmetry, a configuration option to configure the IPv6 nexthop address is also added. Signed-off-by: Toke Høiland-Jørgensen <toke(a)toke.dk> --- This patch is against v2.0.0-pre1. I have done basic testing that Bird can now exchange both v4 and v6 routes with babeld, but have not performed extensive verification of corner cases (i.e, I haven't run it on a real network yet). nest/proto.c | 3 - nest/protocol.h | 3 + proto/babel/babel.c | 151 ++++++++++++++++++++++++++++++++---------- proto/babel/babel.h | 10 ++- proto/babel/config.Y | 6 +- proto/babel/packets.c | 179 +++++++++++++++++++++++++++++++++++++++++++------- 6 files changed, 286 insertions(+), 66 deletions(-) diff --git a/nest/proto.c b/nest/proto.c index 3d764df0..4d46e433 100644 --- a/nest/proto.c +++ b/nest/proto.c @@ -53,9 +53,6 @@ static void channel_reset_limit(struct channel_limit *l); static inline int proto_is_done(struct proto *p) { return (p->proto_state == PS_DOWN) && (p->active_channels == 0); } -static inline int channel_is_active(struct channel *c) -{ return (c->channel_state == CS_START) || (c->channel_state == CS_UP); } - static void proto_log_state_change(struct proto *p) { diff --git a/nest/protocol.h b/nest/protocol.h index f0958678..efeae24b 100644 --- a/nest/protocol.h +++ b/nest/protocol.h @@ -572,6 +572,9 @@ void channel_set_state(struct channel *c, uint state); static inline void channel_init(struct channel *c) { channel_set_state(c, CS_START); } static inline void channel_open(struct channel *c) { channel_set_state(c, CS_UP); } static inline void channel_close(struct channel *c) { channel_set_state(c, CS_FLUSHING); } +static inline int channel_is_active(struct channel *c) +{ return (c->channel_state == CS_START) || (c->channel_state == CS_UP); } + void channel_request_feeding(struct channel *c); void *channel_config_new(const struct channel_class *cc, uint net_type, struct proto_config *proto); diff --git a/proto/babel/babel.c b/proto/babel/babel.c index 1b1d9f62..6cf66fb9 100644 --- a/proto/babel/babel.c +++ b/proto/babel/babel.c @@ -78,13 +78,15 @@ babel_init_entry(void *E) static inline struct babel_entry * babel_find_entry(struct babel_proto *p, const net_addr *n) { - return fib_find(&p->rtable, n); + struct fib *rtable = (n->type == NET_IP4) ? &p->ip4_rtable : &p->ip6_rtable; + return fib_find(rtable, n); } static struct babel_entry * babel_get_entry(struct babel_proto *p, const net_addr *n) { - struct babel_entry *e = fib_get(&p->rtable, n); + struct fib *rtable = (n->type == NET_IP4) ? &p->ip4_rtable : &p->ip6_rtable; + struct babel_entry *e = fib_get(rtable, n); e->proto = p; return e; } @@ -224,15 +226,15 @@ babel_refresh_route(struct babel_route *r) } static void -babel_expire_routes(struct babel_proto *p) +babel_expire_rtable(struct fib *rtable) { struct babel_route *r, *rx; struct fib_iterator fit; - FIB_ITERATE_INIT(&fit, &p->rtable); + FIB_ITERATE_INIT(&fit, rtable); loop: - FIB_ITERATE_START(&p->rtable, &fit, struct babel_entry, e) + FIB_ITERATE_START(rtable, &fit, struct babel_entry, e) { int changed = 0; @@ -253,7 +255,7 @@ loop: /* * We have to restart the iteration because there may be a cascade of * synchronous events babel_select_route() -> nest table change -> - * babel_rt_notify() -> p->rtable change, invalidating hidden variables. + * babel_rt_notify() -> rtable change, invalidating hidden variables. */ FIB_ITERATE_PUT(&fit); @@ -267,13 +269,20 @@ loop: if (EMPTY_LIST(e->sources) && EMPTY_LIST(e->routes)) { FIB_ITERATE_PUT(&fit); - fib_delete(&p->rtable, e); + fib_delete(rtable, e); goto loop; } } FIB_ITERATE_END; } +static void +babel_expire_routes(struct babel_proto *p) +{ + babel_expire_rtable(&p->ip4_rtable); + babel_expire_rtable(&p->ip6_rtable); +} + static struct babel_neighbor * babel_find_neighbor(struct babel_iface *ifa, ip_addr addr) { @@ -468,6 +477,12 @@ static void babel_announce_rte(struct babel_proto *p, struct babel_entry *e) { struct babel_route *r = e->selected_in; + struct channel *c; + + if (e->n.addr->type == NET_IP4) + c = p->ip4_channel; + else + c = p->ip6_channel; if (r) { @@ -490,12 +505,12 @@ babel_announce_rte(struct babel_proto *p, struct babel_entry *e) rte->u.babel.router_id = r->router_id; rte->pflags = 0; - rte_update(&p->p, e->n.addr, rte); + rte_update2(c, e->n.addr, rte, p->p.main_source); } else { /* Retraction */ - rte_update(&p->p, e->n.addr, NULL); + rte_update2(c, e->n.addr, NULL, p->p.main_source); } } @@ -740,11 +755,11 @@ babel_unicast_seqno_request(struct babel_route *r) * transmitted entry is updated. */ static void -babel_send_update(struct babel_iface *ifa, bird_clock_t changed) +babel_send_update_rtable(struct babel_iface *ifa, bird_clock_t changed, struct fib *rtable) { struct babel_proto *p = ifa->proto; - FIB_WALK(&p->rtable, struct babel_entry, e) + FIB_WALK(rtable, struct babel_entry, e) { struct babel_route *r = e->selected_out; @@ -774,6 +789,9 @@ babel_send_update(struct babel_iface *ifa, bird_clock_t changed) msg.update.router_id = r->router_id; net_copy(&msg.update.net, e->n.addr); + msg.update.next_hop = ((e->n.addr->type == NET_IP4) ? + ifa->cf->next_hop_v4 : ifa->cf->next_hop_v6); + babel_enqueue(&msg, ifa); /* Update feasibility distance for redistributed routes */ @@ -794,6 +812,15 @@ babel_send_update(struct babel_iface *ifa, bird_clock_t changed) } static void +babel_send_update(struct babel_iface *ifa, bird_clock_t changed) +{ + struct babel_proto *p = ifa->proto; + + babel_send_update_rtable(ifa, changed, &p->ip4_rtable); + babel_send_update_rtable(ifa, changed, &p->ip6_rtable); +} + +static void babel_trigger_iface_update(struct babel_iface *ifa) { struct babel_proto *p = ifa->proto; @@ -1477,12 +1504,20 @@ babel_add_iface(struct babel_proto *p, struct iface *new, struct babel_iface_con struct ifa *addr; WALK_LIST(addr, new->addrs) + { if (ipa_is_link_local(addr->ip)) ifa->addr = addr->ip; + if (ipa_zero(ifa->cf->next_hop_v4) && ipa_is_ip4(addr->ip)) + ifa->cf->next_hop_v4 = addr->ip; + } + if (ipa_zero(ifa->addr)) log(L_WARN "%s: Cannot find link-local addr on %s", p->p.name, new->name); + if (ipa_zero(ifa->cf->next_hop_v4) && channel_is_active(p->ip4_channel)) + log(L_WARN "%s: Cannot find IPv4 next hop addr on %s", p->p.name, new->name); + init_list(&ifa->neigh_list); ifa->hello_seqno = 1; @@ -1570,6 +1605,26 @@ babel_reconfigure_iface(struct babel_proto *p, struct babel_iface *ifa, struct b (new->tx_priority != old->tx_priority)) return 0; + + if (ipa_zero(new->next_hop_v4)) + { + struct ifa *addr; + WALK_LIST(addr, ifa->iface->addrs) + { + if (ipa_is_ip4(addr->ip)) + { + new->next_hop_v4 = addr->ip; + break; + } + } + } + + /* If no v4 address is found for next_hop we give up on reconfiguring and hope + that we can find one when the interface is brought down and back up + again. */ + if (ipa_zero(new->next_hop_v4) && channel_is_active(p->ip4_channel)) + return 0; + TRACE(D_EVENTS, "Reconfiguring interface %s", ifa->iface->name); ifa->cf = new; @@ -1680,9 +1735,11 @@ babel_dump_iface(struct babel_iface *ifa) { struct babel_neighbor *n; - debug("Babel: Interface %s addr %I rxcost %d type %d hello seqno %d intervals %d %d\n", + debug("Babel: Interface %s addr %I rxcost %d type %d hello seqno %d intervals %d %d", ifa->ifname, ifa->addr, ifa->cf->rxcost, ifa->cf->type, ifa->hello_seqno, ifa->cf->hello_interval, ifa->cf->update_interval); + debug(" next hop v4 %I next hop v6 %I\n", ifa->cf->next_hop_v4, + ipa_zero(ifa->cf->next_hop_v6) ? ifa->addr : ifa->cf->next_hop_v6); WALK_LIST(n, ifa->neigh_list) { debug(" "); babel_dump_neighbor(n); } @@ -1699,7 +1756,12 @@ babel_dump(struct proto *P) WALK_LIST(ifa, p->interfaces) babel_dump_iface(ifa); - FIB_WALK(&p->rtable, struct babel_entry, e) + FIB_WALK(&p->ip4_rtable, struct babel_entry, e) + { + babel_dump_entry(e); + } + FIB_WALK_END; + FIB_WALK(&p->ip6_rtable, struct babel_entry, e) { babel_dump_entry(e); } @@ -1749,8 +1811,9 @@ babel_show_interfaces(struct proto *P, char *iff) } cli_msg(-1023, "%s:", p->p.name); - cli_msg(-1023, "%-10s %-6s %7s %6s %6s", - "Interface", "State", "RX cost", "Nbrs", "Timer"); + cli_msg(-1023, "%-10s %-6s %7s %6s %6s %-15s %s", + "Interface", "State", "RX cost", "Nbrs", "Timer", + "Next hop (v4)", "Next hop (v6)"); WALK_LIST(ifa, p->interfaces) { @@ -1762,8 +1825,11 @@ babel_show_interfaces(struct proto *P, char *iff) nbrs++; int timer = MIN(ifa->next_regular, ifa->next_hello) - now; - cli_msg(-1023, "%-10s %-6s %7u %6u %6u", - ifa->iface->name, (ifa->up ? "Up" : "Down"), ifa->cf->rxcost, nbrs, MAX(timer, 0)); + cli_msg(-1023, "%-10s %-6s %7u %6u %6u %-15I %I", + ifa->iface->name, (ifa->up ? "Up" : "Down"), + ifa->cf->rxcost, nbrs, MAX(timer, 0), + ifa->cf->next_hop_v4, + ipa_zero(ifa->cf->next_hop_v6) ? ifa->addr : ifa->cf->next_hop_v6); } cli_msg(0, ""); @@ -1808,27 +1874,14 @@ babel_show_neighbors(struct proto *P, char *iff) cli_msg(0, ""); } -void -babel_show_entries(struct proto *P) +void babel_show_rtable(struct babel_proto *p, struct fib *rtable) { - struct babel_proto *p = (void *) P; struct babel_source *s = NULL; struct babel_route *r = NULL; char ridbuf[ROUTER_ID_64_LENGTH+1]; - if (p->p.proto_state != PS_UP) - { - cli_msg(-1025, "%s: is not up", p->p.name); - cli_msg(0, ""); - return; - } - - cli_msg(-1025, "%s:", p->p.name); - cli_msg(-1025, "%-29s %-23s %6s %5s %7s %7s", - "Prefix", "Router ID", "Metric", "Seqno", "Expires", "Sources"); - - FIB_WALK(&p->rtable, struct babel_entry, e) + FIB_WALK(rtable, struct babel_entry, e) { r = e->selected_in ? e->selected_in : e->selected_out; @@ -1853,6 +1906,26 @@ babel_show_entries(struct proto *P) } } FIB_WALK_END; +} + +void +babel_show_entries(struct proto *P) +{ + struct babel_proto *p = (void *) P; + + if (p->p.proto_state != PS_UP) + { + cli_msg(-1025, "%s: is not up", p->p.name); + cli_msg(0, ""); + return; + } + + cli_msg(-1025, "%s:", p->p.name); + cli_msg(-1025, "%-29s %-23s %6s %5s %7s %7s", + "Prefix", "Router ID", "Metric", "Seqno", "Expires", "Sources"); + + babel_show_rtable(p, &p->ip4_rtable); + babel_show_rtable(p, &p->ip6_rtable); cli_msg(0, ""); } @@ -2028,8 +2101,10 @@ static struct proto * babel_init(struct proto_config *CF) { struct proto *P = proto_new(CF); + struct babel_proto *p = (void *) P; - P->main_channel = proto_add_channel(P, proto_cf_main_channel(CF)); + proto_configure_channel(P, &p->ip4_channel, proto_cf_find_channel(CF, NET_IP4)); + proto_configure_channel(P, &p->ip6_channel, proto_cf_find_channel(CF, NET_IP6)); P->if_notify = babel_if_notify; P->rt_notify = babel_rt_notify; @@ -2048,8 +2123,11 @@ babel_start(struct proto *P) struct babel_proto *p = (void *) P; struct babel_config *cf = (void *) P->cf; - fib_init(&p->rtable, P->pool, NET_IP6, sizeof(struct babel_entry), + fib_init(&p->ip4_rtable, P->pool, NET_IP4, sizeof(struct babel_entry), OFFSETOF(struct babel_entry, n), 0, babel_init_entry); + fib_init(&p->ip6_rtable, P->pool, NET_IP6, sizeof(struct babel_entry), + OFFSETOF(struct babel_entry, n), 0, babel_init_entry); + init_list(&p->interfaces); p->timer = tm_new_set(P->pool, babel_timer, p, 0, 1); tm_start(p->timer, 2); @@ -2099,7 +2177,8 @@ babel_reconfigure(struct proto *P, struct proto_config *CF) TRACE(D_EVENTS, "Reconfiguring"); - if (!proto_configure_channel(P, &P->main_channel, proto_cf_main_channel(CF))) + if (!(proto_configure_channel(P, &p->ip4_channel, proto_cf_find_channel(CF, NET_IP4)) && + proto_configure_channel(P, &p->ip6_channel, proto_cf_find_channel(CF, NET_IP6)))) return 0; p->p.cf = CF; @@ -2117,7 +2196,7 @@ struct protocol proto_babel = { .template = "babel%d", .attr_class = EAP_BABEL, .preference = DEF_PREF_BABEL, - .channel_mask = NB_IP6, + .channel_mask = NB_IP, .proto_size = sizeof(struct babel_proto), .config_size = sizeof(struct babel_config), .init = babel_init, diff --git a/proto/babel/babel.h b/proto/babel/babel.h index 792c9d60..236c1a3b 100644 --- a/proto/babel/babel.h +++ b/proto/babel/babel.h @@ -116,12 +116,20 @@ struct babel_iface_config { u16 tx_length; /* TX packet length limit (including headers), 0 for MTU */ int tx_tos; int tx_priority; + + ip_addr next_hop_v4; + ip_addr next_hop_v6; }; struct babel_proto { struct proto p; timer *timer; - struct fib rtable; + struct fib ip4_rtable; + struct fib ip6_rtable; + + struct channel *ip4_channel; + struct channel *ip6_channel; + list interfaces; /* Interfaces we really know about (struct babel_iface) */ u64 router_id; u16 update_seqno; /* To be increased on request */ diff --git a/proto/babel/config.Y b/proto/babel/config.Y index cf8983fa..f22ae61c 100644 --- a/proto/babel/config.Y +++ b/proto/babel/config.Y @@ -21,7 +21,8 @@ CF_DEFINES CF_DECLS CF_KEYWORDS(BABEL, METRIC, RXCOST, HELLO, UPDATE, INTERVAL, PORT, WIRED, -WIRELESS, RX, TX, BUFFER, LENGTH, CHECK, LINK, BABEL_METRIC) +WIRELESS, RX, TX, BUFFER, LENGTH, CHECK, LINK, BABEL_METRIC, NEXT, HOP, +IPV4, IPV6) CF_GRAMMAR @@ -30,7 +31,6 @@ CF_ADDTO(proto, babel_proto) babel_proto_start: proto_start BABEL { this_proto = proto_config_new(&proto_babel, $1); - this_proto->net_type = NET_IP6; init_list(&BABEL_CFG->iface_list); }; @@ -98,6 +98,8 @@ babel_iface_item: | TX tos { BABEL_IFACE->tx_tos = $2; } | TX PRIORITY expr { BABEL_IFACE->tx_priority = $3; } | CHECK LINK bool { BABEL_IFACE->check_link = $3; } + | NEXT HOP IPV4 ipa { BABEL_IFACE->next_hop_v4 = $4; if (!ipa_is_ip4($4)) cf_error("Must be an IPv4 address"); } + | NEXT HOP IPV6 ipa { BABEL_IFACE->next_hop_v6 = $4; if (!ipa_is_ip6($4)) cf_error("Must be an IPv6 address"); } ; babel_iface_opts: diff --git a/proto/babel/packets.c b/proto/babel/packets.c index 70cfc196..0f70898e 100644 --- a/proto/babel/packets.c +++ b/proto/babel/packets.c @@ -112,7 +112,8 @@ struct babel_parse_state { struct babel_proto *proto; struct babel_iface *ifa; ip_addr saddr; - ip_addr next_hop; + ip_addr next_hop_v4; + ip_addr next_hop_v6; u64 router_id; /* Router ID used in subsequent updates */ u8 def_ip6_prefix[16]; /* Implicit IPv6 prefix in network order */ u8 def_ip4_prefix[4]; /* Implicit IPv4 prefix in network order */ @@ -130,7 +131,8 @@ enum parse_result { struct babel_write_state { u64 router_id; u8 router_id_seen; -// ip_addr next_hop; + u8 next_hop_v4_seen; + u8 next_hop_v6_seen; }; @@ -163,6 +165,21 @@ put_time16(void *p, u16 v) } static inline void +read_ip4_px(net_addr *n, const void *p, uint plen) +{ + ip4_addr addr = {0}; + memcpy(&addr, p, BYTES(plen)); + net_fill_ip4(n, ip4_ntoh(addr), plen); +} + +static inline void +put_ip4_px(void *p, net_addr *n) +{ + ip4_addr addr = ip4_hton(net4_prefix(n)); + memcpy(p, &addr, NET_SIZE(n)); +} + +static inline void read_ip6_px(net_addr *n, const void *p, uint plen) { ip6_addr addr = IPA_NONE; @@ -432,21 +449,24 @@ babel_read_next_hop(struct babel_tlv *hdr, union babel_msg *m UNUSED, return PARSE_ERROR; case BABEL_AE_IP4: - /* TODO */ + if (TLV_OPT_LENGTH(tlv) < sizeof(ip4_addr)) + return PARSE_ERROR; + + state->next_hop_v4 = ipa_from_ip4(get_ip4(&tlv->addr)); return PARSE_IGNORE; case BABEL_AE_IP6: if (TLV_OPT_LENGTH(tlv) < sizeof(ip6_addr)) return PARSE_ERROR; - state->next_hop = ipa_from_ip6(get_ip6(&tlv->addr)); + state->next_hop_v6 = ipa_from_ip6(get_ip6(&tlv->addr)); return PARSE_IGNORE; case BABEL_AE_IP6_LL: if (TLV_OPT_LENGTH(tlv) < 8) return PARSE_ERROR; - state->next_hop = ipa_from_ip6(get_ip6_ll(&tlv->addr)); + state->next_hop_v6 = ipa_from_ip6(get_ip6_ll(&tlv->addr)); return PARSE_IGNORE; default: @@ -456,6 +476,47 @@ babel_read_next_hop(struct babel_tlv *hdr, union babel_msg *m UNUSED, return PARSE_IGNORE; } +/* This is called directly from babel_write_update() and returns -1 if a next + hop should be written but there is not enough space. */ +static int +babel_write_next_hop(struct babel_tlv *hdr, ip_addr addr, + struct babel_write_state *state, uint max_len) +{ + struct babel_tlv_next_hop *tlv = (void *) hdr; + + + if (ipa_is_ip4(addr) && !state->next_hop_v4_seen) { + uint len = sizeof(struct babel_tlv_next_hop) + sizeof(ip4_addr); + if (len > max_len) + return -1; + + TLV_HDR(tlv, BABEL_TLV_NEXT_HOP, len); + + tlv->ae = BABEL_AE_IP4; + put_ip4(&tlv->addr, ipa_to_ip4(addr)); + state->next_hop_v4_seen = 1; + + return len; + } + else if (ipa_is_ip6(addr) && ip6_nonzero(addr) && !state->next_hop_v6_seen) + { + uint len = sizeof(struct babel_tlv_next_hop) + sizeof(ip6_addr); + if (len > max_len) + return -1; + + TLV_HDR(tlv, BABEL_TLV_NEXT_HOP, len); + + tlv->ae = BABEL_AE_IP6; + put_ip6(&tlv->addr, ipa_to_ip6(addr)); + state->next_hop_v6_seen = 1; + + return len; + } + + return 0; +} + + static int babel_read_update(struct babel_tlv *hdr, union babel_msg *m, struct babel_parse_state *state) @@ -488,8 +549,33 @@ babel_read_update(struct babel_tlv *hdr, union babel_msg *m, break; case BABEL_AE_IP4: - /* TODO */ - return PARSE_IGNORE; + if (tlv->plen > IP4_MAX_PREFIX_LENGTH) + return PARSE_ERROR; + + /* Cannot omit data if there is no saved prefix */ + if (tlv->omitted && !state->def_ip4_prefix_seen) + return PARSE_ERROR; + + /* Need next hop for v4 routes */ + if (ipa_zero(state->next_hop_v4)) + return PARSE_ERROR; + + /* Merge saved prefix and received prefix parts */ + memcpy(buf, state->def_ip4_prefix, tlv->omitted); + memcpy(buf + tlv->omitted, tlv->addr, len); + + ip4_addr prefix4 = get_ip4(buf); + net_fill_ip4(&msg->net, prefix4, tlv->plen); + + if (tlv->flags & BABEL_FLAG_DEF_PREFIX) + { + put_ip4(state->def_ip4_prefix, prefix4); + state->def_ip4_prefix_seen = 1; + } + + msg->next_hop = state->next_hop_v4; + + break; case BABEL_AE_IP6: if (tlv->plen > IP6_MAX_PREFIX_LENGTH) @@ -503,20 +589,23 @@ babel_read_update(struct babel_tlv *hdr, union babel_msg *m, memcpy(buf, state->def_ip6_prefix, tlv->omitted); memcpy(buf + tlv->omitted, tlv->addr, len); - ip6_addr prefix = get_ip6(buf); - net_fill_ip6(&msg->net, prefix, tlv->plen); + ip6_addr prefix6 = get_ip6(buf); + net_fill_ip6(&msg->net, prefix6, tlv->plen); if (tlv->flags & BABEL_FLAG_DEF_PREFIX) { - put_ip6(state->def_ip6_prefix, prefix); + put_ip6(state->def_ip6_prefix, prefix6); state->def_ip6_prefix_seen = 1; } if (tlv->flags & BABEL_FLAG_ROUTER_ID) { - state->router_id = ((u64) _I2(prefix)) << 32 | _I3(prefix); + state->router_id = ((u64) _I2(prefix6)) << 32 | _I3(prefix6); state->router_id_seen = 1; } + + msg->next_hop = state->next_hop_v6; + break; case BABEL_AE_IP6_LL: @@ -535,7 +624,6 @@ babel_read_update(struct babel_tlv *hdr, union babel_msg *m, } msg->router_id = state->router_id; - msg->next_hop = state->next_hop; msg->sender = state->saddr; return PARSE_SUCCESS; @@ -554,7 +642,7 @@ babel_write_update(struct babel_tlv *hdr, union babel_msg *m, * both of them. There is enough space for the Router-ID TLV, because * sizeof(struct babel_tlv_router_id) == sizeof(struct babel_tlv_update). * - * Router ID is not used for retractions, so do not us it in such case. + * Router ID is not used for retractions, so do not use it in such case. */ if ((msg->metric < BABEL_INFINITY) && (!state->router_id_seen || (msg->router_id != state->router_id))) @@ -563,6 +651,16 @@ babel_write_update(struct babel_tlv *hdr, union babel_msg *m, tlv = (struct babel_tlv_update *) NEXT_TLV(tlv); } + int l = babel_write_next_hop((void *)tlv, msg->next_hop, state, max_len - len0); + if (l < 0) + return 0; + + if (l) + { + len0 += l; + tlv = (struct babel_tlv_update *) NEXT_TLV(tlv); + } + uint len = sizeof(struct babel_tlv_update) + NET_SIZE(&msg->net); if (len0 + len > max_len) @@ -576,6 +674,12 @@ babel_write_update(struct babel_tlv *hdr, union babel_msg *m, tlv->ae = BABEL_AE_WILDCARD; tlv->plen = 0; } + else if(msg->net.type == NET_IP4) + { + tlv->ae = BABEL_AE_IP4; + tlv->plen = net4_pxlen(&msg->net); + put_ip4_px(tlv->addr, &msg->net); + } else { tlv->ae = BABEL_AE_IP6; @@ -610,8 +714,14 @@ babel_read_route_request(struct babel_tlv *hdr, union babel_msg *m, return PARSE_SUCCESS; case BABEL_AE_IP4: - /* TODO */ - return PARSE_IGNORE; + if (tlv->plen > IP4_MAX_PREFIX_LENGTH) + return PARSE_ERROR; + + if (TLV_OPT_LENGTH(tlv) < BYTES(tlv->plen)) + return PARSE_ERROR; + + read_ip4_px(&msg->net, tlv->addr, tlv->plen); + return PARSE_SUCCESS; case BABEL_AE_IP6: if (tlv->plen > IP6_MAX_PREFIX_LENGTH) @@ -652,6 +762,12 @@ babel_write_route_request(struct babel_tlv *hdr, union babel_msg *m, tlv->ae = BABEL_AE_WILDCARD; tlv->plen = 0; } + else if (msg->net.type == NET_IP4) + { + tlv->ae = BABEL_AE_IP4; + tlv->plen = net4_pxlen(&msg->net); + put_ip4_px(tlv->addr, &msg->net); + } else { tlv->ae = BABEL_AE_IP6; @@ -684,8 +800,14 @@ babel_read_seqno_request(struct babel_tlv *hdr, union babel_msg *m, return PARSE_ERROR; case BABEL_AE_IP4: - /* TODO */ - return PARSE_IGNORE; + if (tlv->plen > IP4_MAX_PREFIX_LENGTH) + return PARSE_ERROR; + + if (TLV_OPT_LENGTH(tlv) < BYTES(tlv->plen)) + return PARSE_ERROR; + + read_ip4_px(&msg->net, tlv->addr, tlv->plen); + return PARSE_SUCCESS; case BABEL_AE_IP6: if (tlv->plen > IP6_MAX_PREFIX_LENGTH) @@ -720,12 +842,21 @@ babel_write_seqno_request(struct babel_tlv *hdr, union babel_msg *m, return 0; TLV_HDR(tlv, BABEL_TLV_SEQNO_REQUEST, len); - tlv->ae = BABEL_AE_IP6; - tlv->plen = net6_pxlen(&msg->net); + if (msg->net.type == NET_IP4) + { + tlv->ae = BABEL_AE_IP4; + tlv->plen = net4_pxlen(&msg->net); + put_ip4_px(tlv->addr, &msg->net); + } + else + { + tlv->ae = BABEL_AE_IP6; + tlv->plen = net6_pxlen(&msg->net); + put_ip6_px(tlv->addr, &msg->net); + } put_u16(&tlv->seqno, msg->seqno); tlv->hop_count = msg->hop_count; put_u64(&tlv->router_id, msg->router_id); - put_ip6_px(tlv->addr, &msg->net); return len; } @@ -935,10 +1066,10 @@ babel_process_packet(struct babel_pkt_header *pkt, int len, byte *end = (byte *)pkt + plen; struct babel_parse_state state = { - .proto = p, - .ifa = ifa, - .saddr = saddr, - .next_hop = saddr, + .proto = p, + .ifa = ifa, + .saddr = saddr, + .next_hop_v6 = saddr, }; if ((pkt->magic != BABEL_MAGIC) || (pkt->version != BABEL_VERSION)) -- 2.13.0

2 3

Increasing the granularity of internal timers in Bird?
by Toke Høiland-Jørgensen 06 Jun '17

06 Jun '17

Hey I have been looking at changing the internal timers in Bird to operate at millisecond granularity instead of the current one-second steps. I have two motivations for this: - The Babel protocol specifies time in 10ms increments, and I want to be able to schedule things to match this. - I have some ideas for experimenting with low latency route propagation that are not practical the current granularity. Now, looking at the nest timer code, it seems to be quite straight forward to change bird_time_t to represent milliseconds and adjust the functions that use it accordingly (getting system time with millisecond granularity is another matter; but I figure there can be a fallback with rounding for systems that don't support that). However, before I start hacking, I wanted to ask if (a) there is some fundamental reason why this cannot be done, and if not if (b) you would accept patches to this effect? :) Thanks, -Toke

4 3

Re: [PATCH 1/3] lib: Add ip6_common_octets function
by Toke Høiland-Jørgensen 06 Jun '17

06 Jun '17

Ondrej Zajicek <santiago(a)crfreenet.org> writes: > On Tue, Jun 06, 2017 at 01:00:38PM +0200, Toke Høiland-Jørgensen wrote: >> Christian Tacke <Christian.Tacke%2Bbird.network.cz(a)cosmokey.com> writes: >> >> > Hi, >> > >> > just out of curiosity: >> > >> > On Mon, Jun 05, 2017 at 23:49:52 +0200, Toke Høiland-Jørgensen wrote: >> > [...] >> >> +int >> >> +ip6_common_octets(ip6_addr a, ip6_addr b) >> >> +{ >> >> + int i, j, common = 0; >> > [...] >> > >> > Why aren't you using an unsigned int here? >> >> No particular reason; I suppose muscle memory kicked in and gave me an >> 'int'. You are right, of course, that it should probably be uint... :) > > Well, beware of down-iteration in for(): > > for (j=3; j>=0; j--) > > That would not work with uint. Right, but I was mostly referring to the return type. Anyway, that function is not really needed anymore, since the new version of the patch compares buffers instead; so guess you can drop that patch. Will send v2 of the other patch along in a sec... -Toke

1 0