March 2022 - Bird-users - Mailing Lists

birdc show route output question
by Adam Thompson 30 Mar '22

30 Mar '22

Hi, I’m successfully using BIRD with the Go Looking Glass from https://github.com/xddxdd/bird-lg-go. I haven’t needed to use the birdc CLI until now, and I’m a bit confused by it in ways the docs don’t appear to address. (I’m running default distro pkgs on Alma Linux 8.5, one of the post-CentOS RHEL rebuilds.) The output I get looks like: [root@lg2 ~]# birdc show route | head BIRD 2.0.9 ready. Table master4: 223.94.242.0/23 unreachable [uwpg_rtr1_v4 2022-03-26 from 216.73.71.131] * (100) [AS56041?] unreachable [umrtr2_v4 2022-03-26 from 206.211.216.52] (100) [AS56041?] unreachable [umrtr1_v4 2022-03-26 from 206.211.216.51] (100) [AS56041?] 181.123.200.0/22 unreachable [umrtr1_v4 2022-03-26 from 206.211.216.51] * (100) [AS23201i] unreachable [umrtr2_v4 2022-03-26 from 206.211.216.52] (100) [AS23201i] unreachable [uwpg_rtr1_v4 2022-03-26 from 216.73.71.131] (100) [AS23201i] 88.218.186.0/24 unreachable [umrtr1_v4 2022-03-26 from 206.211.216.51] * (100) [AS62240i] unreachable [umrtr2_v4 2022-03-26 from 206.211.216.52] (100) [AS62240i] 1. Is the “unreachable” bit normal on a looking glass / route collector? 2. What’s with the “?” at the end of some ASes? 3. Where’s the AS paths? All I seem to get is the terminal AS. I’m not sure if I’m not using the tool correctly, or if I’m even using the right tool in the first place. Pointers / clues appreciated? Thanks, -Adam Adam Thompson Consultant, Infrastructure Services [MERLIN] 100 - 135 Innovation Drive Winnipeg, MB, R3T 6A8 (204) 977-6824 or 1-800-430-6404 (MB only) athompson(a)merlin.mb.ca<mailto:athompson@merlin.mb.ca> www.merlin.mb.ca<http://www.merlin.mb.ca/>

2 1

RFC: Upgrade filter integer size to 64-bit
by Maria Matejka 28 Mar '22

28 Mar '22

Dear fellow users of BIRD, I'd like to do a quick survey. BIRD now uses a 32-bit unsigned integer inside filters. OTOH, most of the integer route attributes are stored as uintptr_t, which is 32-bit or 64-bit depending on architecture. There are three basic options for further development: 1. Keep it as it is (32-bit filter integer, arch-dependent attribute integer). Pros: * Lesser memory footprint of route attributes on 32-bit archs Cons: * 8-byte attributes (currently only Babel RouterID) stored indirectly at least on 32-bit archs * Explicit checks and transformations between filters and attributes 2. Update both to 64 bits. Pros: * Much more convenient development Cons: * More memory eaten on 32-bit archs (actual ratio depends on how many route attributes you need to store) * Slower computation of arithmetics in filters on 32-bit archs 3. Make it completely arch-dependent, either both 64-bit or both 32-bit. Pros: * Uses both 32-bit and 64-bit archs in their intended way Cons: * Different behavior of filters on different archs * Possibly hard-to-discover architecture-dependent bugs If it was on me, I'd choose to update everything to 64 bits and just hope that the 32-bit use cases aren't affected too much. I suppose that nobody is running BIRD at a low memory 32-bit box having lots of routes in kernel with Linux kernel route metrics set, which is the worst case scenario I can think about. Is there anybody who would try to convince me that I shouldn't make filter integers and route attribute integers both 64 bits wide? Thank you all for your answers. Maria

7 7

[PATCH] Filter: Add support for setting TCP congestion control algorithm
by Trisha Biswas 25 Mar '22

25 Mar '22

Add support for setting the TCP congestion control algorithm per destination by modifying route attributes in BIRD. Previously supported kernel route attributes in BIRD were all integer values. Subsequently, this also adds string handling of kernel metrics. Usage: set krt_cc_algo in the BIRD filter configs. Allowed values can be found with net.ipv4.tcp_allowed_congestion_control. --- diff --git a/doc/bird.sgml b/doc/bird.sgml index 1d5ae056..310aef37 100644 --- a/doc/bird.sgml +++ b/doc/bird.sgml @@ -3377,7 +3377,7 @@ Supported attributes are: <cf/krt_sstresh/, <cf/krt_lock_sstresh/, <cf/krt_cwnd/, <cf/krt_lock_cwnd/, <cf/krt_advmss/, <cf/krt_lock_advmss/, <cf/krt_reordering/, <cf/krt_lock_reordering/, <cf/krt_hoplimit/, <cf/krt_lock_hoplimit/, <cf/krt_rto_min/, <cf/krt_lock_rto_min/, -<cf/krt_initcwnd/, <cf/krt_initrwnd/, <cf/krt_quickack/, +<cf/krt_initcwnd/, <cf/krt_initrwnd/, <cf/krt_quickack/, <cf/krt_cc_algo/, <cf/krt_feature_ecn/, <cf/krt_feature_allfrag/ <sect1>Example diff --git a/filter/config.Y b/filter/config.Y index 8916ea97..8916d08d 100644 --- a/filter/config.Y +++ b/filter/config.Y @@ -178,6 +178,9 @@ f_generate_empty(struct f_dynamic_attr dyn) case EAF_TYPE_LC_SET: empty = f_const_empty_lclist; break; + case EAF_TYPE_CC_ALGO: + empty = f_const_empty_cc_algo; + break; default: cf_error("Can't empty that attribute"); } @@ -816,6 +819,7 @@ term: | '-' EMPTY '-' { $$ = f_new_inst(FI_CONSTANT, f_const_empty_clist); } | '-' '-' EMPTY '-' '-' { $$ = f_new_inst(FI_CONSTANT, f_const_empty_eclist); } | '-' '-' '-' EMPTY '-' '-' '-' { $$ = f_new_inst(FI_CONSTANT, f_const_empty_lclist); } + | '-' '-' '-' '-' EMPTY '-' '-' '-' '-' { $$ = f_new_inst(FI_CONSTANT, f_const_empty_cc_algo); } | PREPEND '(' term ',' term ')' { $$ = f_new_inst(FI_PATH_PREPEND, $3, $5); } | ADD '(' term ',' term ')' { $$ = f_new_inst(FI_CLIST_ADD, $3, $5); } | DELETE '(' term ',' term ')' { $$ = f_new_inst(FI_CLIST_DEL, $3, $5); } diff --git a/filter/data.c b/filter/data.c index 56c1fb17..84e91d59 100644 --- a/filter/data.c +++ b/filter/data.c @@ -91,6 +91,9 @@ const struct f_val f_const_empty_path = { }, f_const_empty_lclist = { .type = T_LCLIST, .val.ad = &null_adata, +}, f_const_empty_cc_algo = { + .type = T_STRING, + .val.ad = &null_adata, }; static struct adata * diff --git a/filter/data.h b/filter/data.h index 4cb6b7a8..46216e92 100644 --- a/filter/data.h +++ b/filter/data.h @@ -262,6 +262,7 @@ trie_match_next_longest_ip6(net_addr_ip6 *n, ip6_addr *found) #define F_CMP_ERROR 999 +#define TCP_CA_NAME_MAX 16 const char *f_type_name(enum f_type t); @@ -297,7 +298,7 @@ undef_value(struct f_val v) (v.val.ad == &null_adata); } -extern const struct f_val f_const_empty_path, f_const_empty_clist, f_const_empty_eclist, f_const_empty_lclist; +extern const struct f_val f_const_empty_path, f_const_empty_clist, f_const_empty_eclist, f_const_empty_lclist, f_const_empty_cc_algo; enum filter_return f_eval(const struct f_line *expr, struct linpool *tmp_pool, struct f_val *pres); diff --git a/filter/f-inst.c b/filter/f-inst.c index 901d2939..735c643a 100644 --- a/filter/f-inst.c +++ b/filter/f-inst.c @@ -709,6 +709,9 @@ case EAF_TYPE_LC_SET: RESULT_(T_LCLIST, ad, e->u.ptr); break; + case EAF_TYPE_CC_ALGO: + RESULT_(T_STRING, s, (const char *) e->u.ptr->data); + break; case EAF_TYPE_UNDEF: RESULT_VOID; break; @@ -758,7 +761,16 @@ case EAF_TYPE_LC_SET: l->attrs[0].u.ptr = v1.val.ad; break; - + case EAF_TYPE_CC_ALGO: + if (v1.type != T_STRING) + runtime( "Setting cc_algo attribute to non-string value" ); + else if (strlen(v1.val.s) >= TCP_CA_NAME_MAX) + runtime( "Setting cc_algo attribute out of bounds (> 15 chars)" ); + /* Store cc_algo string in byte[], making sure to copy the null terminator */ + struct adata *d = lp_alloc_adata(fs->pool, strlen(v1.val.s) + 1); + memcpy(d->data, v1.val.s, d->length); + l->attrs[0].u.ptr = d; + break; case EAF_TYPE_BITFIELD: { /* First, we have to find the old value */ diff --git a/filter/f-util.c b/filter/f-util.c index 410999a6..3cf68270 100644 --- a/filter/f-util.c +++ b/filter/f-util.c @@ -121,6 +121,8 @@ ca_lookup(pool *p, const char *name, int f_type) break; case T_LCLIST: ea_type = EAF_TYPE_LC_SET; + case T_STRING: + ea_type = EAF_TYPE_CC_ALGO; break; default: cf_error("Custom route attribute of unsupported type"); diff --git a/nest/route.h b/nest/route.h index 7930058a..053d2f1e 100644 --- a/nest/route.h +++ b/nest/route.h @@ -587,6 +587,7 @@ const char *ea_custom_name(uint ea); #define EAF_TYPE_INT_SET 0x0a /* Set of u32's (e.g., a community list) */ #define EAF_TYPE_EC_SET 0x0e /* Set of pairs of u32's - ext. community list */ #define EAF_TYPE_LC_SET 0x12 /* Set of triplets of u32's - large community list */ +#define EAF_TYPE_CC_ALGO 0x18 /* String to specify congestion control algorithm */ #define EAF_TYPE_UNDEF 0x1f /* `force undefined' entry */ #define EAF_EMBEDDED 0x01 /* Data stored in eattr.u.data (part of type spec) */ #define EAF_VAR_LENGTH 0x02 /* Attribute length is variable (part of type spec) */ diff --git a/nest/rt-attr.c b/nest/rt-attr.c index c630aa95..b4fdfd85 100644 --- a/nest/rt-attr.c +++ b/nest/rt-attr.c @@ -969,6 +969,9 @@ ea_show(struct cli *c, const eattr *e) case EAF_TYPE_LC_SET: ea_show_lc_set(c, ad, pos, buf, end); return; + case EAF_TYPE_CC_ALGO: + bsnprintf(pos, ad->length, "%s", (char *)ad->data); + return; case EAF_TYPE_UNDEF: default: bsprintf(pos, "<type %02x>", e->type); diff --git a/sysdep/linux/krt-sys.h b/sysdep/linux/krt-sys.h index 8897f889..f265ceb7 100644 --- a/sysdep/linux/krt-sys.h +++ b/sysdep/linux/krt-sys.h @@ -39,7 +39,7 @@ static inline struct ifa * kif_get_primary_ip(struct iface *i UNUSED) { return N #define EA_KRT_SCOPE EA_CODE(PROTOCOL_KERNEL, 0x12) -#define KRT_METRICS_MAX 0x10 /* RTAX_QUICKACK+1 */ +#define KRT_METRICS_MAX 0x11 /* RTAX_CC_ALGO+1 */ #define KRT_METRICS_OFFSET 0x20 /* Offset of EA_KRT_* vs RTAX_* */ #define KRT_FEATURES_MAX 4 @@ -64,7 +64,7 @@ static inline struct ifa * kif_get_primary_ip(struct iface *i UNUSED) { return N #define EA_KRT_RTO_MIN EA_CODE(PROTOCOL_KERNEL, 0x2d) #define EA_KRT_INITRWND EA_CODE(PROTOCOL_KERNEL, 0x2e) #define EA_KRT_QUICKACK EA_CODE(PROTOCOL_KERNEL, 0x2f) - +#define EA_KRT_CC_ALGO EA_CODE(PROTOCOL_KERNEL, 0x30) struct krt_params { u32 table_id; /* Kernel table ID we sync with */ diff --git a/sysdep/linux/netlink.Y b/sysdep/linux/netlink.Y index 487ad1d8..b07fa842 100644 --- a/sysdep/linux/netlink.Y +++ b/sysdep/linux/netlink.Y @@ -14,7 +14,7 @@ CF_KEYWORDS(KERNEL, TABLE, METRIC, NETLINK, RX, BUFFER, KRT_PREFSRC, KRT_REALM, KRT_SCOPE, KRT_MTU, KRT_WINDOW, KRT_RTT, KRT_RTTVAR, KRT_SSTRESH, KRT_CWND, KRT_ADVMSS, KRT_REORDERING, KRT_HOPLIMIT, KRT_INITCWND, KRT_RTO_MIN, KRT_INITRWND, KRT_QUICKACK, - KRT_LOCK_MTU, KRT_LOCK_WINDOW, KRT_LOCK_RTT, KRT_LOCK_RTTVAR, + KRT_CC_ALGO, KRT_LOCK_MTU, KRT_LOCK_WINDOW, KRT_LOCK_RTT, KRT_LOCK_RTTVAR, KRT_LOCK_SSTRESH, KRT_LOCK_CWND, KRT_LOCK_ADVMSS, KRT_LOCK_REORDERING, KRT_LOCK_HOPLIMIT, KRT_LOCK_RTO_MIN, KRT_FEATURE_ECN, KRT_FEATURE_ALLFRAG) @@ -45,6 +45,7 @@ dynamic_attr: KRT_INITCWND { $$ = f_new_dynamic_attr(EAF_TYPE_INT, T_INT, EA_KRT dynamic_attr: KRT_RTO_MIN { $$ = f_new_dynamic_attr(EAF_TYPE_INT, T_INT, EA_KRT_RTO_MIN); } ; dynamic_attr: KRT_INITRWND { $$ = f_new_dynamic_attr(EAF_TYPE_INT, T_INT, EA_KRT_INITRWND); } ; dynamic_attr: KRT_QUICKACK { $$ = f_new_dynamic_attr(EAF_TYPE_INT, T_INT, EA_KRT_QUICKACK); } ; +dynamic_attr: KRT_CC_ALGO { $$ = f_new_dynamic_attr(EAF_TYPE_CC_ALGO, T_STRING, EA_KRT_CC_ALGO); } ; /* Bits of EA_KRT_LOCK, based on RTAX_* constants */ diff --git a/sysdep/linux/netlink.c b/sysdep/linux/netlink.c index 29b744cb..4d4d704b 100644 --- a/sysdep/linux/netlink.c +++ b/sysdep/linux/netlink.c @@ -73,6 +73,10 @@ #define NETLINK_GET_STRICT_CHK 12 #endif +#ifndef TCP_CA_NAME_MAX +#define TCP_CA_NAME_MAX 16 +#endif + #define krt_ipv4(p) ((p)->af == AF_INET) #define krt_ecmp6(p) ((p)->af == AF_INET6) @@ -534,6 +538,9 @@ static inline u16 rta_get_u16(struct rtattr *a) static inline u32 rta_get_u32(struct rtattr *a) { return *(u32 *) RTA_DATA(a); } +static inline char *rta_get_str(struct rtattr *a) +{ return (char *) RTA_DATA(a); } + static inline ip4_addr rta_get_ip4(struct rtattr *a) { return ip4_ntoh(*(ip4_addr *) RTA_DATA(a)); } @@ -624,6 +631,12 @@ nl_add_attr_u32(struct nlmsghdr *h, uint bufsize, int code, u32 data) nl_add_attr(h, bufsize, code, &data, 4); } +static inline void +nl_add_attr_str(struct nlmsghdr *h, unsigned bufsize, int code, char *str) +{ + nl_add_attr(h, bufsize, code, str, strlen(str) + 1); +} + static inline void nl_add_attr_ip4(struct nlmsghdr *h, uint bufsize, int code, ip4_addr ip4) { @@ -880,20 +893,25 @@ err: } static void -nl_add_metrics(struct nlmsghdr *h, uint bufsize, u32 *metrics, int max) +nl_add_metrics(struct nlmsghdr *h, uint bufsize, u32 *metrics, char *cc_algo, int max) { struct rtattr *a = nl_open_attr(h, bufsize, RTA_METRICS); int t; - for (t = 1; t < max; t++) - if (metrics[0] & (1 << t)) - nl_add_attr_u32(h, bufsize, t, metrics[t]); + for (t = 1; t < max; t++) { + if (metrics[0] & (1 << t)) { + if (EA_CODE(PROTOCOL_KERNEL, KRT_METRICS_OFFSET + t) == EA_KRT_CC_ALGO) + nl_add_attr_str(h, bufsize, t, cc_algo); + else + nl_add_attr_u32(h, bufsize, t, metrics[t]); + } + } nl_close_attr(h, a); } static int -nl_parse_metrics(struct rtattr *hdr, u32 *metrics, int max) +nl_parse_metrics(struct rtattr *hdr, u32 *metrics, char *cc_algo, int max) { struct rtattr *a = RTA_DATA(hdr); int len = RTA_PAYLOAD(hdr); @@ -911,7 +929,19 @@ nl_parse_metrics(struct rtattr *hdr, u32 *metrics, int max) return -1; metrics[0] |= 1 << a->rta_type; - metrics[a->rta_type] = rta_get_u32(a); + + if (EA_CODE(PROTOCOL_KERNEL, KRT_METRICS_OFFSET + a->rta_type) == EA_KRT_CC_ALGO) { + char *str = rta_get_str(a); + if (strlen(str) < TCP_CA_NAME_MAX) { + memcpy(cc_algo, str, strlen(str) + 1); + } else { + log(L_ERR "KRT: Received route with cc_algo attribute out of bounds (> 15 chars)"); + return -1; + } + metrics[a->rta_type] = 0; + } else { + metrics[a->rta_type] = rta_get_u32(a); + } } if (len > 0) @@ -1427,6 +1457,7 @@ nl_send_route(struct krt_proto *p, rte *e, int op, int dest, struct nexthop *nh) u32 metrics[KRT_METRICS_MAX]; + char cc_algo[TCP_CA_NAME_MAX]; metrics[0] = 0; struct ea_walk_state ews = { .eattrs = eattrs }; @@ -1434,11 +1465,16 @@ nl_send_route(struct krt_proto *p, rte *e, int op, int dest, struct nexthop *nh) { int id = ea->id - EA_KRT_METRICS; metrics[0] |= 1 << id; - metrics[id] = ea->u.data; + if(ea->id == EA_KRT_CC_ALGO) { + metrics[id] = 0; + memcpy(cc_algo, ea->u.ptr->data, ea->u.ptr->length); + } else { + metrics[id] = ea->u.data; + } } if (metrics[0]) - nl_add_metrics(&r->h, rsize, metrics, KRT_METRICS_MAX); + nl_add_metrics(&r->h, rsize, metrics, cc_algo, KRT_METRICS_MAX); dest: @@ -1907,10 +1943,12 @@ nl_parse_route(struct nl_parse_state *s, struct nlmsghdr *h) if (a[RTA_METRICS]) { u32 metrics[KRT_METRICS_MAX]; + char *cc_algo = lp_alloc(s->pool, TCP_CA_NAME_MAX); ea_list *ea = lp_alloc(s->pool, sizeof(ea_list) + KRT_METRICS_MAX * sizeof(eattr)); + struct adata *d = lp_alloc(s->pool, sizeof(struct adata) + TCP_CA_NAME_MAX); int t, n = 0; - if (nl_parse_metrics(a[RTA_METRICS], metrics, ARRAY_SIZE(metrics)) < 0) + if (nl_parse_metrics(a[RTA_METRICS], metrics, cc_algo, ARRAY_SIZE(metrics)) < 0) { log(L_ERR "KRT: Received route %N with strange RTA_METRICS attribute", net->n.addr); return; @@ -1921,8 +1959,15 @@ nl_parse_route(struct nl_parse_state *s, struct nlmsghdr *h) { ea->attrs[n].id = EA_CODE(PROTOCOL_KERNEL, KRT_METRICS_OFFSET + t); ea->attrs[n].flags = 0; - ea->attrs[n].type = EAF_TYPE_INT; /* FIXME: Some are EAF_TYPE_BITFIELD */ - ea->attrs[n].u.data = metrics[t]; + if (ea->attrs[n].id == EA_KRT_CC_ALGO) { + ea->attrs[n].type = EAF_TYPE_CC_ALGO; + d->length = strlen(cc_algo) + 1; + memcpy(d->data, cc_algo, d->length); + ea->attrs[n].u.ptr = d; + } else { + ea->attrs[n].type = EAF_TYPE_INT; /* FIXME: Some are EAF_TYPE_BITFIELD */ + ea->attrs[n].u.data = metrics[t]; + } n++; } @@ -2225,7 +2270,8 @@ krt_sys_copy_config(struct krt_config *d, struct krt_config *s) static const char *krt_metrics_names[KRT_METRICS_MAX] = { NULL, "lock", "mtu", "window", "rtt", "rttvar", "sstresh", "cwnd", "advmss", - "reordering", "hoplimit", "initcwnd", "features", "rto_min", "initrwnd", "quickack" + "reordering", "hoplimit", "initcwnd", "features", "rto_min", "initrwnd", "quickack", + "cc_algo" }; static const char *krt_features_names[KRT_FEATURES_MAX] = { --- Please also find the patch attached. Thank you, Trisha -- *Trisha Biswas* | Sr. Software Engineer, Network Systems fastly.com | @fastly <https://twitter.com/fastly> | LinkedIn <http://www.linkedin.com/company/fastly>

3 3

Filter based on BGP protocol status ...
by Xavier Trilla 25 Mar '22

25 Mar '22

Hi, I'm quite sure this cannot be done, but I also know there is a lot of BIRD I still don't know, so here it goes: Is there any way to filter a export route based on another BGP session status? For the sake of simplicity let's say we have just two providers A and B, and I only want to export some specific routes to B when A is down. I can do it externally with a quite simple script (For example: Check if the provider is down via CLI and if it's down insert the routes I want to export to a kernel table and export that to provider B) but it would be nice if I could do it directly in Bird. I've been scratching my head around this, but unless there is something like if proto.A == down on the filters I don't really see how to do it (Or maybe some way to raise a global flag based on if I'm receiving routes via provider B, but I don't think that can't be done either.) Thanks for your time! Xavier

7 12

Filter BGP route based on ospf_metric1 ?
by Robert Sander 24 Mar '22

24 Mar '22

Hi, Our routers speak OSPF and have a full mesh of BGP connections for our internal routes. This works quite well. One router sees the other through OSPF: 80.241.60.8/32 via 10.25.19.251 on bond0.19 [ospf1 2022-03-17] * I (150/2474) [80.241.60.8] Type: OSPF unicast univ OSPF.metric1: 2474 OSPF.metric2: 16777215 OSPF.tag: 0x00000000 OSPF.router_id: 80.241.60.8 The BGP session is established to 80.241.60.8 and e.g. this route is learned: 192.168.100.0/22 via 10.25.19.251 on bond0.19 [thor 2022-03-17 from 80.241.60.8] * (100/2474) [i] Type: BGP unicast univ BGP.origin: IGP BGP.as_path: BGP.next_hop: 80.241.60.8 BGP.local_pref: 100 In the first line the ospf_metric1 is shown: (100/2474) But when I try to filter on it (because I want to drop routes with a too high metric), the route attribute ospf_metric1 is "(void)". This seems to be because it's a BGP route and not an OSPF route. Is it somehow possible to achieve this? PS: This is with bird 1.6.8. Kindest Regards -- Robert Sander Heinlein Consulting GmbH Schwedter Str. 8/9b, 10119 Berlin http://www.heinlein-support.de Tel: 030 / 405051-43 Fax: 030 / 405051-19 Zwangsangaben lt. §35a GmbHG: HRB 220009 B / Amtsgericht Berlin-Charlottenburg, Geschäftsführer: Peer Heinlein -- Sitz: Berlin

3 4

[PATCH] Lib: accept 240.0.0.0/4 as a valid range
by Vincent Bernat 16 Mar '22

16 Mar '22

240.0.0.0/4 is marked as reserved and considered invalid by BIRD. At work, we are using this range internally since all RFC 1918 are full and 100.64.0.0/10 is already used too. BIRD complains loudly for each interface using this range. This change makes it possible to use this range. I have used scope "universe". But I would be happy with "site" too. While widely discussed, I don't think 240/4 will become routable on the Internet one day. As a bonus, I added some comments and unrolled a condition for each block. I also have added some hints for the compiler to avoid using jumps in the hotpath (tested on Godbolt, see https://godbolt.org/z/rGjz336K3) --- lib/ip.c | 31 ++++++++++++------------------- sysdep/config.h | 5 +++++ 2 files changed, 17 insertions(+), 19 deletions(-) diff --git a/lib/ip.c b/lib/ip.c index fcc72cafb4de..8f0f32d25d61 100644 --- a/lib/ip.c +++ b/lib/ip.c @@ -80,30 +80,23 @@ ip6_masklen(ip6_addr *a) } int -ip4_classify(ip4_addr ad) +ip4_classify(u32 a) { - u32 a = _I(ad); u32 b = a >> 24U; - if (b && b <= 0xdf) - { - if (b == 0x7f) - return IADDR_HOST | SCOPE_HOST; - else if ((b == 0x0a) || - ((a & 0xffff0000) == 0xc0a80000) || - ((a & 0xfff00000) == 0xac100000)) - return IADDR_HOST | SCOPE_SITE; - else - return IADDR_HOST | SCOPE_UNIVERSE; - } - - if (b >= 0xe0 && b <= 0xef) + if (unlikely(b == 0x00)) + return IADDR_INVALID; /* 0.0.0.0/8 This network */ + if (unlikely(b == 0x7f)) /* 127.0.0.0/8 Loopback */ + return IADDR_HOST | SCOPE_HOST; + if ((b == 0x0a) || /* 10.0.0.0/8 Private-use */ + ((a & 0xffff0000) == 0xc0a80000) || /* 192.168.0.0/16 Private-use */ + ((a & 0xfff00000) == 0xac100000)) /* 172.16.0.0/12 Private-use */ + return IADDR_HOST | SCOPE_SITE; + if (unlikely(b >= 0xe0 && b <= 0xef)) /* 224.0.0.0/4 Multicast */ return IADDR_MULTICAST | SCOPE_UNIVERSE; - - if (a == 0xffffffff) + if (unlikely(a == 0xffffffff)) /* 255.255.255.255 Limited broadcast */ return IADDR_BROADCAST | SCOPE_LINK; - - return IADDR_INVALID; + return IADDR_HOST | SCOPE_UNIVERSE; } int diff --git a/sysdep/config.h b/sysdep/config.h index b0531844af9f..4d73543c3894 100644 --- a/sysdep/config.h +++ b/sysdep/config.h @@ -30,6 +30,11 @@ */ #include "sysdep/paths.h" +/* Likely/unlikely macros */ + +#define likely(x) __builtin_expect((x),1) +#define unlikely(x) __builtin_expect((x),0) + /* Types */ #include <stdint.h> -- 2.35.1

5 6

operator to prepend bgp path multiple times
by Alexander Zubkov 14 Mar '22

14 Mar '22

Hi, I suggest to add a new operator: prepend_times(P, A, N) - that will allow to prepend AS A to path P several times (N). Now we need to use some function like this: function add_prepends(int asn, int num) { if num = 1 then { bgp_path.prepend(asn); } else if num = 2 then { bgp_path.prepend(asn); bgp_path.prepend(asn); } else if num = 3 then { ... } Usually the number of prepends is observable, so this function can fit most of cases, but with available operator it will be a bit more convenient, a little syntactic sugar. But it is not something that is hard to live without, of course. In my patch, I chose to "join" implementations of prepend() and prepend_times(). It is questionable, because it requires some additional supporting stuff if I understood right. So may be it is better to implement it as separate operators.

2 2

constant for empty prefix set
by Alexander Zubkov 14 Mar '22

14 Mar '22

Hi, I propose to add some constant for empty prefix set, for example (as in the patch): define pfx_empty = [ /empty/ ]; It is useful when we define a named prefix list, which we can use somewhere in filters, but this prefix list is generated, for example, and we may need to make it empty. But there is no currently means to create an empty prefix set, as I understand.

2 17

FreeBSD, multiple routing tables, "KRT: Error sending route 192.222.249.232/29 to kernel: File exists"
by Dan Mahoney 13 Mar '22

13 Mar '22

Hey there, We've got a fairly simple setup where we have two fibs. Interface routes are in both fibs the way BSD does it. (Table 1 is the second fib, if it matters). log syslog all; protocol kernel { scan time 20; ipv4 { import all; export all; table master4; }; kernel table 1; } protocol kernel { ipv6 { import all; export all; table master6; }; kernel table 1; } protocol device { scan time 10; } setfib 1 netstat -nr: Routing tables (fib: 1) Internet: Destination Gateway Flags Netif Expire default 192.222.249.233 UGS bge0 127.0.0.1 link#5 UH lo0 127.0.0.2 link#6 UH lo1 192.222.249.232/29 link#1 U bge0 fib 0 would be the same. An old thread seemed to imply this was because the "direct" protocol was missing. Adding the "direct" protocol doesn't stop this error every 20 seconds in syslog (which seems to be related to the scan time). Mar 12 15:24:11 box1 bird[34463]: KRT: Error sending route 192.222.249.232/29 to kernel: File exists Mar 12 15:24:31 box1 bird[34463]: KRT: Error sending route 192.222.249.232/29 to kernel: File exists Mar 12 15:24:51 box1 bird[34463]: KRT: Error sending route 192.222.249.232/29 to kernel: File exists Mar 12 15:25:11 box1 bird[34463]: KRT: Error sending route 192.222.249.232/29 to kernel: File exists Mar 12 15:25:31 box1 bird[34463]: KRT: Error sending route 192.222.249.232/29 to kernel: File exists Is it possible bird isn't differentiating routes learned from each fib and is learning from both, separately, and not properly deduplicating those routes? -Dan

2 2

birdc configure exits with 0 on error
by Vincent Bernat 11 Mar '22

11 Mar '22

Hey! "birdc configure" (or any command in fact) exits with 0 on error. This is a bit annoying as when using "systemctl reload bird", we get no notification there is an error. Looking at the source code, it seems there is no easy way to hack around that. Commands do not report an error code and messages printed are not tagged as errors. Here is a small workaround for systemd: -- Keep it simple to make it faster. - The Elements of Programming Style (Kernighan & Plauger)

2 3