proxmox-mirrors/mirror_frr
1
0
Fork 0
mirror of https://git.proxmox.com/git/mirror_frr synced 2025-11-04 03:29:06 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
61d9ffe168
mirror_frr/zebra/zebra_rib.c
Stephen Worley 986a6617cc zebra: Optimize the fib/notified nexthop matching 
Optimize the fib and notified nexthop group comparison algorithm
to assume ordering. There were some pretty serious performance hits with
this on high ecmp routes.

Signed-off-by: Stephen Worley <sworley@cumulusnetworks.com>
2019-10-25 11:13:42 -04:00

3711 lines
95 KiB
C
Raw Blame History

/* Routing Information Base.
* Copyright (C) 1997, 98, 99, 2001 Kunihiro Ishiguro
*
* This file is part of GNU Zebra.
*
* GNU Zebra is free software; you can redistribute it and/or modify it
* under the terms of the GNU General Public License as published by the
* Free Software Foundation; either version 2, or (at your option) any
* later version.
*
* GNU Zebra is distributed in the hope that it will be useful, but
* WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* General Public License for more details.
*
* You should have received a copy of the GNU General Public License along
* with this program; see the file COPYING; if not, write to the Free Software
* Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include <zebra.h>
#include "command.h"
#include "if.h"
#include "linklist.h"
#include "log.h"
#include "memory.h"
#include "mpls.h"
#include "nexthop.h"
#include "prefix.h"
#include "prefix.h"
#include "routemap.h"
#include "sockunion.h"
#include "srcdest_table.h"
#include "table.h"
#include "thread.h"
#include "vrf.h"
#include "workqueue.h"
#include "nexthop_group_private.h"
#include "frr_pthread.h"
#include "zebra/zebra_router.h"
#include "zebra/connected.h"
#include "zebra/debug.h"
#include "zebra/interface.h"
#include "zebra/redistribute.h"
#include "zebra/rib.h"
#include "zebra/rt.h"
#include "zebra/zapi_msg.h"
#include "zebra/zebra_errors.h"
#include "zebra/zebra_memory.h"
#include "zebra/zebra_ns.h"
#include "zebra/zebra_rnh.h"
#include "zebra/zebra_routemap.h"
#include "zebra/zebra_vrf.h"
#include "zebra/zebra_vxlan.h"
#include "zebra/zapi_msg.h"
#include "zebra/zebra_dplane.h"
DEFINE_MTYPE_STATIC(ZEBRA, RIB_UPDATE_CTX, "Rib update context object");
/*
* Event, list, and mutex for delivery of dataplane results
*/
static pthread_mutex_t dplane_mutex;
static struct thread *t_dplane;
static struct dplane_ctx_q rib_dplane_q;
DEFINE_HOOK(rib_update, (struct route_node * rn, const char *reason),
(rn, reason))
/* Should we allow non Quagga processes to delete our routes */
extern int allow_delete;
/* Each route type's string and default distance value. */
static const struct {
int key;
uint8_t distance;
uint8_t meta_q_map;
} route_info[ZEBRA_ROUTE_MAX] = {
[ZEBRA_ROUTE_NHG] = {ZEBRA_ROUTE_NHG, 255 /* Uneeded for nhg's */, 0},
[ZEBRA_ROUTE_SYSTEM] = {ZEBRA_ROUTE_SYSTEM, 0, 5},
[ZEBRA_ROUTE_KERNEL] = {ZEBRA_ROUTE_KERNEL, 0, 1},
[ZEBRA_ROUTE_CONNECT] = {ZEBRA_ROUTE_CONNECT, 0, 1},
[ZEBRA_ROUTE_STATIC] = {ZEBRA_ROUTE_STATIC, 1, 2},
[ZEBRA_ROUTE_RIP] = {ZEBRA_ROUTE_RIP, 120, 3},
[ZEBRA_ROUTE_RIPNG] = {ZEBRA_ROUTE_RIPNG, 120, 3},
[ZEBRA_ROUTE_OSPF] = {ZEBRA_ROUTE_OSPF, 110, 3},
[ZEBRA_ROUTE_OSPF6] = {ZEBRA_ROUTE_OSPF6, 110, 3},
[ZEBRA_ROUTE_ISIS] = {ZEBRA_ROUTE_ISIS, 115, 3},
[ZEBRA_ROUTE_BGP] = {ZEBRA_ROUTE_BGP, 20 /* IBGP is 200. */, 4},
[ZEBRA_ROUTE_PIM] = {ZEBRA_ROUTE_PIM, 255, 5},
[ZEBRA_ROUTE_EIGRP] = {ZEBRA_ROUTE_EIGRP, 90, 3},
[ZEBRA_ROUTE_NHRP] = {ZEBRA_ROUTE_NHRP, 10, 3},
[ZEBRA_ROUTE_HSLS] = {ZEBRA_ROUTE_HSLS, 255, 5},
[ZEBRA_ROUTE_OLSR] = {ZEBRA_ROUTE_OLSR, 255, 5},
[ZEBRA_ROUTE_TABLE] = {ZEBRA_ROUTE_TABLE, 150, 2},
[ZEBRA_ROUTE_LDP] = {ZEBRA_ROUTE_LDP, 150, 5},
[ZEBRA_ROUTE_VNC] = {ZEBRA_ROUTE_VNC, 20, 4},
[ZEBRA_ROUTE_VNC_DIRECT] = {ZEBRA_ROUTE_VNC_DIRECT, 20, 4},
[ZEBRA_ROUTE_VNC_DIRECT_RH] = {ZEBRA_ROUTE_VNC_DIRECT_RH, 20, 4},
[ZEBRA_ROUTE_BGP_DIRECT] = {ZEBRA_ROUTE_BGP_DIRECT, 20, 4},
[ZEBRA_ROUTE_BGP_DIRECT_EXT] = {ZEBRA_ROUTE_BGP_DIRECT_EXT, 20, 4},
[ZEBRA_ROUTE_BABEL] = {ZEBRA_ROUTE_BABEL, 100, 3},
[ZEBRA_ROUTE_SHARP] = {ZEBRA_ROUTE_SHARP, 150, 5},
[ZEBRA_ROUTE_PBR] = {ZEBRA_ROUTE_PBR, 200, 5},
[ZEBRA_ROUTE_BFD] = {ZEBRA_ROUTE_BFD, 255, 5},
[ZEBRA_ROUTE_OPENFABRIC] = {ZEBRA_ROUTE_OPENFABRIC, 115, 3},
[ZEBRA_ROUTE_VRRP] = {ZEBRA_ROUTE_VRRP, 255, 5}
/* Any new route type added to zebra, should be mirrored here */
/* no entry/default: 150 */
};
static void __attribute__((format(printf, 5, 6)))
_rnode_zlog(const char *_func, vrf_id_t vrf_id, struct route_node *rn,
int priority, const char *msgfmt, ...)
{
char buf[SRCDEST2STR_BUFFER + sizeof(" (MRIB)")];
char msgbuf[512];
va_list ap;
va_start(ap, msgfmt);
vsnprintf(msgbuf, sizeof(msgbuf), msgfmt, ap);
va_end(ap);
if (rn) {
rib_table_info_t *info = srcdest_rnode_table_info(rn);
srcdest_rnode2str(rn, buf, sizeof(buf));
if (info->safi == SAFI_MULTICAST)
strlcat(buf, " (MRIB)", sizeof(buf));
} else {
snprintf(buf, sizeof(buf), "{(route_node *) NULL}");
}
zlog(priority, "%s: %d:%s: %s", _func, vrf_id, buf, msgbuf);
}
#define rnode_debug(node, vrf_id, ...) \
_rnode_zlog(__func__, vrf_id, node, LOG_DEBUG, __VA_ARGS__)
#define rnode_info(node, ...) \
_rnode_zlog(__func__, vrf_id, node, LOG_INFO, __VA_ARGS__)
uint8_t route_distance(int type)
{
uint8_t distance;
if ((unsigned)type >= array_size(route_info))
distance = 150;
else
distance = route_info[type].distance;
return distance;
}
int is_zebra_valid_kernel_table(uint32_t table_id)
{
#ifdef linux
if ((table_id == RT_TABLE_UNSPEC) || (table_id == RT_TABLE_LOCAL)
|| (table_id == RT_TABLE_COMPAT))
return 0;
#endif
return 1;
}
int is_zebra_main_routing_table(uint32_t table_id)
{
if (table_id == RT_TABLE_MAIN)
return 1;
return 0;
}
int zebra_check_addr(const struct prefix *p)
{
if (p->family == AF_INET) {
uint32_t addr;
addr = p->u.prefix4.s_addr;
addr = ntohl(addr);
if (IPV4_NET127(addr) || IN_CLASSD(addr)
|| IPV4_LINKLOCAL(addr))
return 0;
}
if (p->family == AF_INET6) {
if (IN6_IS_ADDR_LOOPBACK(&p->u.prefix6))
return 0;
if (IN6_IS_ADDR_LINKLOCAL(&p->u.prefix6))
return 0;
}
return 1;
}
/* Add nexthop to the end of a rib node's nexthop list */
void route_entry_nexthop_add(struct route_entry *re, struct nexthop *nexthop)
{
_nexthop_group_add_sorted(re->ng, nexthop);
}
/**
* copy_nexthop - copy a nexthop to the rib structure.
*/
void route_entry_copy_nexthops(struct route_entry *re, struct nexthop *nh)
{
assert(!re->ng->nexthop);
copy_nexthops(&re->ng->nexthop, nh, NULL);
}
/* Delete specified nexthop from the list. */
void route_entry_nexthop_delete(struct route_entry *re, struct nexthop *nexthop)
{
if (nexthop->next)
nexthop->next->prev = nexthop->prev;
if (nexthop->prev)
nexthop->prev->next = nexthop->next;
else
re->ng->nexthop = nexthop->next;
}
struct nexthop *route_entry_nexthop_ifindex_add(struct route_entry *re,
ifindex_t ifindex,
vrf_id_t nh_vrf_id)
{
struct nexthop *nexthop;
nexthop = nexthop_new();
nexthop->type = NEXTHOP_TYPE_IFINDEX;
nexthop->ifindex = ifindex;
nexthop->vrf_id = nh_vrf_id;
route_entry_nexthop_add(re, nexthop);
return nexthop;
}
struct nexthop *route_entry_nexthop_ipv4_add(struct route_entry *re,
struct in_addr *ipv4,
struct in_addr *src,
vrf_id_t nh_vrf_id)
{
struct nexthop *nexthop;
nexthop = nexthop_new();
nexthop->type = NEXTHOP_TYPE_IPV4;
nexthop->vrf_id = nh_vrf_id;
nexthop->gate.ipv4 = *ipv4;
if (src)
nexthop->src.ipv4 = *src;
route_entry_nexthop_add(re, nexthop);
return nexthop;
}
struct nexthop *route_entry_nexthop_ipv4_ifindex_add(struct route_entry *re,
struct in_addr *ipv4,
struct in_addr *src,
ifindex_t ifindex,
vrf_id_t nh_vrf_id)
{
struct nexthop *nexthop;
struct interface *ifp;
nexthop = nexthop_new();
nexthop->vrf_id = nh_vrf_id;
nexthop->type = NEXTHOP_TYPE_IPV4_IFINDEX;
nexthop->gate.ipv4 = *ipv4;
if (src)
nexthop->src.ipv4 = *src;
nexthop->ifindex = ifindex;
ifp = if_lookup_by_index(nexthop->ifindex, nh_vrf_id);
/*Pending: need to think if null ifp here is ok during bootup?
There was a crash because ifp here was coming to be NULL */
if (ifp)
if (connected_is_unnumbered(ifp))
SET_FLAG(nexthop->flags, NEXTHOP_FLAG_ONLINK);
route_entry_nexthop_add(re, nexthop);
return nexthop;
}
struct nexthop *route_entry_nexthop_ipv6_add(struct route_entry *re,
struct in6_addr *ipv6,
vrf_id_t nh_vrf_id)
{
struct nexthop *nexthop;
nexthop = nexthop_new();
nexthop->vrf_id = nh_vrf_id;
nexthop->type = NEXTHOP_TYPE_IPV6;
nexthop->gate.ipv6 = *ipv6;
route_entry_nexthop_add(re, nexthop);
return nexthop;
}
struct nexthop *route_entry_nexthop_ipv6_ifindex_add(struct route_entry *re,
struct in6_addr *ipv6,
ifindex_t ifindex,
vrf_id_t nh_vrf_id)
{
struct nexthop *nexthop;
nexthop = nexthop_new();
nexthop->vrf_id = nh_vrf_id;
nexthop->type = NEXTHOP_TYPE_IPV6_IFINDEX;
nexthop->gate.ipv6 = *ipv6;
nexthop->ifindex = ifindex;
route_entry_nexthop_add(re, nexthop);
return nexthop;
}
struct nexthop *route_entry_nexthop_blackhole_add(struct route_entry *re,
enum blackhole_type bh_type)
{
struct nexthop *nexthop;
nexthop = nexthop_new();
nexthop->vrf_id = VRF_DEFAULT;
nexthop->type = NEXTHOP_TYPE_BLACKHOLE;
nexthop->bh_type = bh_type;
route_entry_nexthop_add(re, nexthop);
return nexthop;
}
struct route_entry *rib_match(afi_t afi, safi_t safi, vrf_id_t vrf_id,
union g_addr *addr, struct route_node **rn_out)
{
struct prefix p;
struct route_table *table;
struct route_node *rn;
struct route_entry *match = NULL;
/* Lookup table. */
table = zebra_vrf_table(afi, safi, vrf_id);
if (!table)
return 0;
memset(&p, 0, sizeof(struct prefix));
p.family = afi;
if (afi == AFI_IP) {
p.u.prefix4 = addr->ipv4;
p.prefixlen = IPV4_MAX_PREFIXLEN;
} else {
p.u.prefix6 = addr->ipv6;
p.prefixlen = IPV6_MAX_PREFIXLEN;
}
rn = route_node_match(table, (struct prefix *)&p);
while (rn) {
rib_dest_t *dest;
route_unlock_node(rn);
dest = rib_dest_from_rnode(rn);
if (dest && dest->selected_fib
&& !CHECK_FLAG(dest->selected_fib->status,
ROUTE_ENTRY_REMOVED))
match = dest->selected_fib;
/* If there is no selected route or matched route is EGP, go up
tree. */
if (!match) {
do {
rn = rn->parent;
} while (rn && rn->info == NULL);
if (rn)
route_lock_node(rn);
} else {
if (match->type != ZEBRA_ROUTE_CONNECT) {
if (!CHECK_FLAG(match->status,
ROUTE_ENTRY_INSTALLED))
return NULL;
}
if (rn_out)
*rn_out = rn;
return match;
}
}
return NULL;
}
struct route_entry *rib_match_ipv4_multicast(vrf_id_t vrf_id,
struct in_addr addr,
struct route_node **rn_out)
{
struct route_entry *re = NULL, *mre = NULL, *ure = NULL;
struct route_node *m_rn = NULL, *u_rn = NULL;
union g_addr gaddr = {.ipv4 = addr};
switch (zrouter.ipv4_multicast_mode) {
case MCAST_MRIB_ONLY:
return rib_match(AFI_IP, SAFI_MULTICAST, vrf_id, &gaddr,
rn_out);
case MCAST_URIB_ONLY:
return rib_match(AFI_IP, SAFI_UNICAST, vrf_id, &gaddr, rn_out);
case MCAST_NO_CONFIG:
case MCAST_MIX_MRIB_FIRST:
re = mre = rib_match(AFI_IP, SAFI_MULTICAST, vrf_id, &gaddr,
&m_rn);
if (!mre)
re = ure = rib_match(AFI_IP, SAFI_UNICAST, vrf_id,
&gaddr, &u_rn);
break;
case MCAST_MIX_DISTANCE:
mre = rib_match(AFI_IP, SAFI_MULTICAST, vrf_id, &gaddr, &m_rn);
ure = rib_match(AFI_IP, SAFI_UNICAST, vrf_id, &gaddr, &u_rn);
if (mre && ure)
re = ure->distance < mre->distance ? ure : mre;
else if (mre)
re = mre;
else if (ure)
re = ure;
break;
case MCAST_MIX_PFXLEN:
mre = rib_match(AFI_IP, SAFI_MULTICAST, vrf_id, &gaddr, &m_rn);
ure = rib_match(AFI_IP, SAFI_UNICAST, vrf_id, &gaddr, &u_rn);
if (mre && ure)
re = u_rn->p.prefixlen > m_rn->p.prefixlen ? ure : mre;
else if (mre)
re = mre;
else if (ure)
re = ure;
break;
}
if (rn_out)
*rn_out = (re == mre) ? m_rn : u_rn;
if (IS_ZEBRA_DEBUG_RIB) {
char buf[BUFSIZ];
inet_ntop(AF_INET, &addr, buf, BUFSIZ);
zlog_debug("%s: %s: vrf: %u found %s, using %s",
__func__, buf, vrf_id,
mre ? (ure ? "MRIB+URIB" : "MRIB")
: ure ? "URIB" : "nothing",
re == ure ? "URIB" : re == mre ? "MRIB" : "none");
}
return re;
}
struct route_entry *rib_lookup_ipv4(struct prefix_ipv4 *p, vrf_id_t vrf_id)
{
struct route_table *table;
struct route_node *rn;
struct route_entry *match = NULL;
rib_dest_t *dest;
/* Lookup table. */
table = zebra_vrf_table(AFI_IP, SAFI_UNICAST, vrf_id);
if (!table)
return 0;
rn = route_node_lookup(table, (struct prefix *)p);
/* No route for this prefix. */
if (!rn)
return NULL;
/* Unlock node. */
route_unlock_node(rn);
dest = rib_dest_from_rnode(rn);
if (dest && dest->selected_fib
&& !CHECK_FLAG(dest->selected_fib->status, ROUTE_ENTRY_REMOVED))
match = dest->selected_fib;
if (!match)
return NULL;
if (match->type == ZEBRA_ROUTE_CONNECT)
return match;
if (CHECK_FLAG(match->status, ROUTE_ENTRY_INSTALLED))
return match;
return NULL;
}
/*
* Is this RIB labeled-unicast? It must be of type BGP and all paths
* (nexthops) must have a label.
*/
int zebra_rib_labeled_unicast(struct route_entry *re)
{
struct nexthop *nexthop = NULL;
if (re->type != ZEBRA_ROUTE_BGP)
return 0;
for (ALL_NEXTHOPS_PTR(re->ng, nexthop))
if (!nexthop->nh_label || !nexthop->nh_label->num_labels)
return 0;
return 1;
}
/* Update flag indicates whether this is a "replace" or not. Currently, this
* is only used for IPv4.
*/
void rib_install_kernel(struct route_node *rn, struct route_entry *re,
struct route_entry *old)
{
struct nexthop *nexthop;
rib_table_info_t *info = srcdest_rnode_table_info(rn);
struct zebra_vrf *zvrf = vrf_info_lookup(re->vrf_id);
const struct prefix *p, *src_p;
enum zebra_dplane_result ret;
rib_dest_t *dest = rib_dest_from_rnode(rn);
srcdest_rnode_prefixes(rn, &p, &src_p);
if (info->safi != SAFI_UNICAST) {
for (ALL_NEXTHOPS_PTR(re->ng, nexthop))
SET_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB);
return;
}
/*
* Install the resolved nexthop object first.
*/
zebra_nhg_install_kernel(zebra_nhg_lookup_id(re->nhe_id));
/*
* If this is a replace to a new RE let the originator of the RE
* know that they've lost
*/
if (old && (old != re) && (old->type != re->type))
zsend_route_notify_owner(old, p, ZAPI_ROUTE_BETTER_ADMIN_WON);
/* Update fib selection */
dest->selected_fib = re;
/*
* Make sure we update the FPM any time we send new information to
* the kernel.
*/
hook_call(rib_update, rn, "installing in kernel");
/* Send add or update */
if (old)
ret = dplane_route_update(rn, re, old);
else
ret = dplane_route_add(rn, re);
switch (ret) {
case ZEBRA_DPLANE_REQUEST_QUEUED:
SET_FLAG(re->status, ROUTE_ENTRY_QUEUED);
if (old) {
SET_FLAG(old->status, ROUTE_ENTRY_QUEUED);
/* Free old FIB nexthop group */
if (old->fib_ng.nexthop) {
nexthops_free(old->fib_ng.nexthop);
old->fib_ng.nexthop = NULL;
}
if (!RIB_SYSTEM_ROUTE(old)) {
/* Clear old route's FIB flags */
for (ALL_NEXTHOPS_PTR(old->ng, nexthop)) {
UNSET_FLAG(nexthop->flags,
NEXTHOP_FLAG_FIB);
}
}
}
if (zvrf)
zvrf->installs_queued++;
break;
case ZEBRA_DPLANE_REQUEST_FAILURE:
{
char str[SRCDEST2STR_BUFFER];
srcdest_rnode2str(rn, str, sizeof(str));
flog_err(EC_ZEBRA_DP_INSTALL_FAIL,
"%u:%s: Failed to enqueue dataplane install",
re->vrf_id, str);
break;
}
case ZEBRA_DPLANE_REQUEST_SUCCESS:
if (zvrf)
zvrf->installs++;
break;
}
return;
}
/* Uninstall the route from kernel. */
void rib_uninstall_kernel(struct route_node *rn, struct route_entry *re)
{
struct nexthop *nexthop;
rib_table_info_t *info = srcdest_rnode_table_info(rn);
struct zebra_vrf *zvrf = vrf_info_lookup(re->vrf_id);
if (info->safi != SAFI_UNICAST) {
UNSET_FLAG(re->status, ROUTE_ENTRY_INSTALLED);
for (ALL_NEXTHOPS_PTR(re->ng, nexthop))
UNSET_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB);
return;
}
/*
* Make sure we update the FPM any time we send new information to
* the dataplane.
*/
hook_call(rib_update, rn, "uninstalling from kernel");
switch (dplane_route_delete(rn, re)) {
case ZEBRA_DPLANE_REQUEST_QUEUED:
if (zvrf)
zvrf->removals_queued++;
break;
case ZEBRA_DPLANE_REQUEST_FAILURE:
{
char str[SRCDEST2STR_BUFFER];
srcdest_rnode2str(rn, str, sizeof(str));
flog_err(EC_ZEBRA_DP_INSTALL_FAIL,
"%u:%s: Failed to enqueue dataplane uninstall",
re->vrf_id, str);
break;
}
case ZEBRA_DPLANE_REQUEST_SUCCESS:
if (zvrf)
zvrf->removals++;
break;
}
return;
}
/* Uninstall the route from kernel. */
static void rib_uninstall(struct route_node *rn, struct route_entry *re)
{
rib_table_info_t *info = srcdest_rnode_table_info(rn);
rib_dest_t *dest = rib_dest_from_rnode(rn);
struct nexthop *nexthop;
if (dest && dest->selected_fib == re) {
if (info->safi == SAFI_UNICAST)
hook_call(rib_update, rn, "rib_uninstall");
/* If labeled-unicast route, uninstall transit LSP. */
if (zebra_rib_labeled_unicast(re))
zebra_mpls_lsp_uninstall(info->zvrf, rn, re);
rib_uninstall_kernel(rn, re);
dest->selected_fib = NULL;
/* Free FIB nexthop group, if present */
if (re->fib_ng.nexthop) {
nexthops_free(re->fib_ng.nexthop);
re->fib_ng.nexthop = NULL;
}
for (ALL_NEXTHOPS_PTR(re->ng, nexthop))
UNSET_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB);
}
if (CHECK_FLAG(re->flags, ZEBRA_FLAG_SELECTED)) {
const struct prefix *p, *src_p;
srcdest_rnode_prefixes(rn, &p, &src_p);
redistribute_delete(p, src_p, re, NULL);
UNSET_FLAG(re->flags, ZEBRA_FLAG_SELECTED);
}
}
/*
* rib_can_delete_dest
*
* Returns true if the given dest can be deleted from the table.
*/
static int rib_can_delete_dest(rib_dest_t *dest)
{
if (re_list_first(&dest->routes)) {
return 0;
}
/*
* Unresolved rnh's are stored on the default route's list
*
* dest->rnode can also be the source prefix node in an
* ipv6 sourcedest table. Fortunately the prefix of a
* source prefix node can never be the default prefix.
*/
if (is_default_prefix(&dest->rnode->p))
return 0;
/*
* Don't delete the dest if we have to update the FPM about this
* prefix.
*/
if (CHECK_FLAG(dest->flags, RIB_DEST_UPDATE_FPM)
|| CHECK_FLAG(dest->flags, RIB_DEST_SENT_TO_FPM))
return 0;
return 1;
}
void zebra_rib_evaluate_rn_nexthops(struct route_node *rn, uint32_t seq)
{
rib_dest_t *dest = rib_dest_from_rnode(rn);
struct rnh *rnh;
/*
* We are storing the rnh's associated withb
* the tracked nexthop as a list of the rn's.
* Unresolved rnh's are placed at the top
* of the tree list.( 0.0.0.0/0 for v4 and 0::0/0 for v6 )
* As such for each rn we need to walk up the tree
* and see if any rnh's need to see if they
* would match a more specific route
*/
while (rn) {
if (IS_ZEBRA_DEBUG_NHT_DETAILED) {
char buf[PREFIX_STRLEN];
zlog_debug("%s: %s Being examined for Nexthop Tracking Count: %zd",
__PRETTY_FUNCTION__,
srcdest_rnode2str(rn, buf, sizeof(buf)),
dest ? rnh_list_count(&dest->nht) : 0);
}
if (!dest) {
rn = rn->parent;
if (rn)
dest = rib_dest_from_rnode(rn);
continue;
}
/*
* If we have any rnh's stored in the nht list
* then we know that this route node was used for
* nht resolution and as such we need to call the
* nexthop tracking evaluation code
*/
frr_each_safe(rnh_list, &dest->nht, rnh) {
struct zebra_vrf *zvrf =
zebra_vrf_lookup_by_id(rnh->vrf_id);
struct prefix *p = &rnh->node->p;
if (IS_ZEBRA_DEBUG_NHT_DETAILED) {
char buf1[PREFIX_STRLEN];
char buf2[PREFIX_STRLEN];
zlog_debug("%u:%s has Nexthop(%s) Type: %s depending on it, evaluating %u:%u",
zvrf->vrf->vrf_id,
srcdest_rnode2str(rn, buf1,
sizeof(buf1)),
prefix2str(p, buf2, sizeof(buf2)),
rnh_type2str(rnh->type),
seq, rnh->seqno);
}
/*
* If we have evaluated this node on this pass
* already, due to following the tree up
* then we know that we can move onto the next
* rnh to process.
*
* Additionally we call zebra_evaluate_rnh
* when we gc the dest. In this case we know
* that there must be no other re's where
* we were originally as such we know that
* that sequence number is ok to respect.
*/
if (rnh->seqno == seq) {
if (IS_ZEBRA_DEBUG_NHT_DETAILED)
zlog_debug(
"\tNode processed and moved already");
continue;
}
rnh->seqno = seq;
zebra_evaluate_rnh(zvrf, family2afi(p->family), 0,
rnh->type, p);
}
rn = rn->parent;
if (rn)
dest = rib_dest_from_rnode(rn);
}
}
/*
* rib_gc_dest
*
* Garbage collect the rib dest corresponding to the given route node
* if appropriate.
*
* Returns true if the dest was deleted, false otherwise.
*/
int rib_gc_dest(struct route_node *rn)
{
rib_dest_t *dest;
dest = rib_dest_from_rnode(rn);
if (!dest)
return 0;
if (!rib_can_delete_dest(dest))
return 0;
if (IS_ZEBRA_DEBUG_RIB) {
struct zebra_vrf *zvrf;
zvrf = rib_dest_vrf(dest);
rnode_debug(rn, zvrf_id(zvrf), "removing dest from table");
}
zebra_rib_evaluate_rn_nexthops(rn, zebra_router_get_next_sequence());
dest->rnode = NULL;
rnh_list_fini(&dest->nht);
XFREE(MTYPE_RIB_DEST, dest);
rn->info = NULL;
/*
* Release the one reference that we keep on the route node.
*/
route_unlock_node(rn);
return 1;
}
static void rib_process_add_fib(struct zebra_vrf *zvrf, struct route_node *rn,
struct route_entry *new)
{
hook_call(rib_update, rn, "new route selected");
/* Update real nexthop. This may actually determine if nexthop is active
* or not. */
if (!nexthop_group_active_nexthop_num(new->ng)) {
UNSET_FLAG(new->status, ROUTE_ENTRY_CHANGED);
return;
}
if (IS_ZEBRA_DEBUG_RIB) {
char buf[SRCDEST2STR_BUFFER];
srcdest_rnode2str(rn, buf, sizeof(buf));
zlog_debug("%u:%s: Adding route rn %p, re %p (%s)",
zvrf_id(zvrf), buf, rn, new,
zebra_route_string(new->type));
}
/* If labeled-unicast route, install transit LSP. */
if (zebra_rib_labeled_unicast(new))
zebra_mpls_lsp_install(zvrf, rn, new);
rib_install_kernel(rn, new, NULL);
UNSET_FLAG(new->status, ROUTE_ENTRY_CHANGED);
}
static void rib_process_del_fib(struct zebra_vrf *zvrf, struct route_node *rn,
struct route_entry *old)
{
hook_call(rib_update, rn, "removing existing route");
/* Uninstall from kernel. */
if (IS_ZEBRA_DEBUG_RIB) {
char buf[SRCDEST2STR_BUFFER];
srcdest_rnode2str(rn, buf, sizeof(buf));
zlog_debug("%u:%s: Deleting route rn %p, re %p (%s)",
zvrf_id(zvrf), buf, rn, old,
zebra_route_string(old->type));
}
/* If labeled-unicast route, uninstall transit LSP. */
if (zebra_rib_labeled_unicast(old))
zebra_mpls_lsp_uninstall(zvrf, rn, old);
rib_uninstall_kernel(rn, old);
/* Update nexthop for route, reset changed flag. */
/* Note: this code also handles the Linux case when an interface goes
* down, causing the kernel to delete routes without sending DELROUTE
* notifications
*/
if (RIB_KERNEL_ROUTE(old))
SET_FLAG(old->status, ROUTE_ENTRY_REMOVED);
else
UNSET_FLAG(old->status, ROUTE_ENTRY_CHANGED);
}
static void rib_process_update_fib(struct zebra_vrf *zvrf,
struct route_node *rn,
struct route_entry *old,
struct route_entry *new)
{
int nh_active = 0;
/*
* We have to install or update if a new route has been selected or
* something has changed.
*/
if (new != old || CHECK_FLAG(new->status, ROUTE_ENTRY_CHANGED)) {
hook_call(rib_update, rn, "updating existing route");
/* Update the nexthop; we could determine here that nexthop is
* inactive. */
if (nexthop_group_active_nexthop_num(new->ng))
nh_active = 1;
/* If nexthop is active, install the selected route, if
* appropriate. If
* the install succeeds, cleanup flags for prior route, if
* different from
* newly selected.
*/
if (nh_active) {
if (IS_ZEBRA_DEBUG_RIB) {
char buf[SRCDEST2STR_BUFFER];
srcdest_rnode2str(rn, buf, sizeof(buf));
if (new != old)
zlog_debug(
"%u:%s: Updating route rn %p, re %p (%s) old %p (%s)",
zvrf_id(zvrf), buf, rn, new,
zebra_route_string(new->type),
old,
zebra_route_string(old->type));
else
zlog_debug(
"%u:%s: Updating route rn %p, re %p (%s)",
zvrf_id(zvrf), buf, rn, new,
zebra_route_string(new->type));
}
/* If labeled-unicast route, uninstall transit LSP. */
if (zebra_rib_labeled_unicast(old))
zebra_mpls_lsp_uninstall(zvrf, rn, old);
/*
* Non-system route should be installed.
* If labeled-unicast route, install transit
* LSP.
*/
if (zebra_rib_labeled_unicast(new))
zebra_mpls_lsp_install(zvrf, rn, new);
rib_install_kernel(rn, new, old);
}
/*
* If nexthop for selected route is not active or install
* failed, we
* may need to uninstall and delete for redistribution.
*/
if (!nh_active) {
if (IS_ZEBRA_DEBUG_RIB) {
char buf[SRCDEST2STR_BUFFER];
srcdest_rnode2str(rn, buf, sizeof(buf));
if (new != old)
zlog_debug(
"%u:%s: Deleting route rn %p, re %p (%s) old %p (%s) - nexthop inactive",
zvrf_id(zvrf), buf, rn, new,
zebra_route_string(new->type),
old,
zebra_route_string(old->type));
else
zlog_debug(
"%u:%s: Deleting route rn %p, re %p (%s) - nexthop inactive",
zvrf_id(zvrf), buf, rn, new,
zebra_route_string(new->type));
}
/* If labeled-unicast route, uninstall transit LSP. */
if (zebra_rib_labeled_unicast(old))
zebra_mpls_lsp_uninstall(zvrf, rn, old);
rib_uninstall_kernel(rn, old);
}
} else {
/*
* Same route selected; check if in the FIB and if not,
* re-install. This is housekeeping code to deal with
* race conditions in kernel with linux netlink reporting
* interface up before IPv4 or IPv6 protocol is ready
* to add routes.
*/
if (!CHECK_FLAG(new->status, ROUTE_ENTRY_INSTALLED) ||
RIB_SYSTEM_ROUTE(new))
rib_install_kernel(rn, new, NULL);
}
/* Update prior route. */
if (new != old)
UNSET_FLAG(old->status, ROUTE_ENTRY_CHANGED);
/* Clear changed flag. */
UNSET_FLAG(new->status, ROUTE_ENTRY_CHANGED);
}
/* Check if 'alternate' RIB entry is better than 'current'. */
static struct route_entry *rib_choose_best(struct route_entry *current,
struct route_entry *alternate)
{
if (current == NULL)
return alternate;
/* filter route selection in following order:
* - connected beats other types
* - if both connected, loopback or vrf wins
* - lower distance beats higher
* - lower metric beats higher for equal distance
* - last, hence oldest, route wins tie break.
*/
/* Connected routes. Check to see if either are a vrf
* or loopback interface. If not, pick the last connected
* route of the set of lowest metric connected routes.
*/
if (alternate->type == ZEBRA_ROUTE_CONNECT) {
if (current->type != ZEBRA_ROUTE_CONNECT)
return alternate;
/* both are connected. are either loop or vrf? */
struct nexthop *nexthop = NULL;
for (ALL_NEXTHOPS_PTR(alternate->ng, nexthop)) {
struct interface *ifp = if_lookup_by_index(
nexthop->ifindex, alternate->vrf_id);
if (ifp && if_is_loopback_or_vrf(ifp))
return alternate;
}
for (ALL_NEXTHOPS_PTR(current->ng, nexthop)) {
struct interface *ifp = if_lookup_by_index(
nexthop->ifindex, current->vrf_id);
if (ifp && if_is_loopback_or_vrf(ifp))
return current;
}
/* Neither are loop or vrf so pick best metric */
if (alternate->metric <= current->metric)
return alternate;
return current;
}
if (current->type == ZEBRA_ROUTE_CONNECT)
return current;
/* higher distance loses */
if (alternate->distance < current->distance)
return alternate;
if (current->distance < alternate->distance)
return current;
/* metric tie-breaks equal distance */
if (alternate->metric <= current->metric)
return alternate;
return current;
}
/* Core function for processing nexthop group contexts's off metaq */
static void rib_nhg_process(struct nhg_ctx *ctx)
{
nhg_ctx_process(ctx);
}
/* Core function for processing routing information base. */
static void rib_process(struct route_node *rn)
{
struct route_entry *re;
struct route_entry *next;
struct route_entry *old_selected = NULL;
struct route_entry *new_selected = NULL;
struct route_entry *old_fib = NULL;
struct route_entry *new_fib = NULL;
struct route_entry *best = NULL;
char buf[SRCDEST2STR_BUFFER];
rib_dest_t *dest;
struct zebra_vrf *zvrf = NULL;
const struct prefix *p, *src_p;
srcdest_rnode_prefixes(rn, &p, &src_p);
vrf_id_t vrf_id = VRF_UNKNOWN;
assert(rn);
dest = rib_dest_from_rnode(rn);
if (dest) {
zvrf = rib_dest_vrf(dest);
vrf_id = zvrf_id(zvrf);
}
if (IS_ZEBRA_DEBUG_RIB)
srcdest_rnode2str(rn, buf, sizeof(buf));
if (IS_ZEBRA_DEBUG_RIB_DETAILED)
zlog_debug("%u:%s: Processing rn %p", vrf_id, buf, rn);
/*
* we can have rn's that have a NULL info pointer
* (dest). As such let's not let the deref happen
* additionally we know RNODE_FOREACH_RE_SAFE
* will not iterate so we are ok.
*/
if (dest)
old_fib = dest->selected_fib;
RNODE_FOREACH_RE_SAFE (rn, re, next) {
if (IS_ZEBRA_DEBUG_RIB_DETAILED)
zlog_debug(
"%u:%s: Examine re %p (%s) status %x flags %x dist %d metric %d",
vrf_id, buf, re, zebra_route_string(re->type),
re->status, re->flags, re->distance,
re->metric);
/* Currently selected re. */
if (CHECK_FLAG(re->flags, ZEBRA_FLAG_SELECTED)) {
assert(old_selected == NULL);
old_selected = re;
}
/* Skip deleted entries from selection */
if (CHECK_FLAG(re->status, ROUTE_ENTRY_REMOVED))
continue;
/* Skip unreachable nexthop. */
/* This first call to nexthop_active_update is merely to
* determine if there's any change to nexthops associated
* with this RIB entry. Now, rib_process() can be invoked due
* to an external event such as link down or due to
* next-hop-tracking evaluation. In the latter case,
* a decision has already been made that the NHs have changed.
* So, no need to invoke a potentially expensive call again.
* Further, since the change might be in a recursive NH which
* is not caught in the nexthop_active_update() code. Thus, we
* might miss changes to recursive NHs.
*/
if (CHECK_FLAG(re->status, ROUTE_ENTRY_CHANGED)
&& !nexthop_active_update(rn, re)) {
if (re->type == ZEBRA_ROUTE_TABLE) {
/* XXX: HERE BE DRAGONS!!!!!
* In all honesty, I have not yet figured out
* what this part does or why the
* ROUTE_ENTRY_CHANGED test above is correct
* or why we need to delete a route here, and
* also not whether this concerns both selected
* and fib route, or only selected
* or only fib
*
* This entry was denied by the 'ip protocol
* table' route-map, we need to delete it */
if (re != old_selected) {
if (IS_ZEBRA_DEBUG_RIB)
zlog_debug(
"%s: %u:%s: imported via import-table but denied "
"by the ip protocol table route-map",
__func__, vrf_id, buf);
rib_unlink(rn, re);
} else
SET_FLAG(re->status,
ROUTE_ENTRY_REMOVED);
}
continue;
}
/* Infinite distance. */
if (re->distance == DISTANCE_INFINITY) {
UNSET_FLAG(re->status, ROUTE_ENTRY_CHANGED);
continue;
}
if (CHECK_FLAG(re->flags, ZEBRA_FLAG_FIB_OVERRIDE)) {
best = rib_choose_best(new_fib, re);
if (new_fib && best != new_fib)
UNSET_FLAG(new_fib->status,
ROUTE_ENTRY_CHANGED);
new_fib = best;
} else {
best = rib_choose_best(new_selected, re);
if (new_selected && best != new_selected)
UNSET_FLAG(new_selected->status,
ROUTE_ENTRY_CHANGED);
new_selected = best;
}
if (best != re)
UNSET_FLAG(re->status, ROUTE_ENTRY_CHANGED);
} /* RNODE_FOREACH_RE */
/* If no FIB override route, use the selected route also for FIB */
if (new_fib == NULL)
new_fib = new_selected;
/* After the cycle is finished, the following pointers will be set:
* old_selected --- RE entry currently having SELECTED
* new_selected --- RE entry that is newly SELECTED
* old_fib --- RE entry currently in kernel FIB
* new_fib --- RE entry that is newly to be in kernel FIB
*
* new_selected will get SELECTED flag, and is going to be redistributed
* the zclients. new_fib (which can be new_selected) will be installed
* in kernel.
*/
if (IS_ZEBRA_DEBUG_RIB_DETAILED) {
zlog_debug(
"%u:%s: After processing: old_selected %p new_selected %p old_fib %p new_fib %p",
vrf_id, buf, (void *)old_selected, (void *)new_selected,
(void *)old_fib, (void *)new_fib);
}
/* Buffer ROUTE_ENTRY_CHANGED here, because it will get cleared if
* fib == selected */
bool selected_changed = new_selected && CHECK_FLAG(new_selected->status,
ROUTE_ENTRY_CHANGED);
/* Update fib according to selection results */
if (new_fib && old_fib)
rib_process_update_fib(zvrf, rn, old_fib, new_fib);
else if (new_fib)
rib_process_add_fib(zvrf, rn, new_fib);
else if (old_fib)
rib_process_del_fib(zvrf, rn, old_fib);
/* Update SELECTED entry */
if (old_selected != new_selected || selected_changed) {
if (new_selected && new_selected != new_fib)
UNSET_FLAG(new_selected->status, ROUTE_ENTRY_CHANGED);
if (new_selected)
SET_FLAG(new_selected->flags, ZEBRA_FLAG_SELECTED);
if (old_selected) {
/*
* If we're removing the old entry, we should tell
* redist subscribers about that *if* they aren't
* going to see a redist for the new entry.
*/
if (!new_selected || CHECK_FLAG(old_selected->status,
ROUTE_ENTRY_REMOVED))
redistribute_delete(p, src_p,
old_selected,
new_selected);
if (old_selected != new_selected)
UNSET_FLAG(old_selected->flags,
ZEBRA_FLAG_SELECTED);
}
}
/* Remove all RE entries queued for removal */
RNODE_FOREACH_RE_SAFE (rn, re, next) {
if (CHECK_FLAG(re->status, ROUTE_ENTRY_REMOVED)) {
if (IS_ZEBRA_DEBUG_RIB) {
rnode_debug(rn, vrf_id, "rn %p, removing re %p",
(void *)rn, (void *)re);
}
rib_unlink(rn, re);
}
}
/*
* Check if the dest can be deleted now.
*/
rib_gc_dest(rn);
}
static void zebra_rib_evaluate_mpls(struct route_node *rn)
{
rib_dest_t *dest = rib_dest_from_rnode(rn);
struct zebra_vrf *zvrf = vrf_info_lookup(VRF_DEFAULT);
if (!dest)
return;
if (CHECK_FLAG(dest->flags, RIB_DEST_UPDATE_LSPS)) {
if (IS_ZEBRA_DEBUG_MPLS)
zlog_debug(
"%u: Scheduling all LSPs upon RIB completion",
zvrf_id(zvrf));
zebra_mpls_lsp_schedule(zvrf);
mpls_unmark_lsps_for_processing(rn);
}
}
/*
* Utility to match route with dplane context data
*/
static bool rib_route_match_ctx(const struct route_entry *re,
const struct zebra_dplane_ctx *ctx,
bool is_update)
{
bool result = false;
if (is_update) {
/*
* In 'update' case, we test info about the 'previous' or
* 'old' route
*/
if ((re->type == dplane_ctx_get_old_type(ctx)) &&
(re->instance == dplane_ctx_get_old_instance(ctx))) {
result = true;
/* TODO -- we're using this extra test, but it's not
* exactly clear why.
*/
if (re->type == ZEBRA_ROUTE_STATIC &&
(re->distance != dplane_ctx_get_old_distance(ctx) ||
re->tag != dplane_ctx_get_old_tag(ctx))) {
result = false;
}
}
} else {
/*
* Ordinary, single-route case using primary context info
*/
if ((dplane_ctx_get_op(ctx) != DPLANE_OP_ROUTE_DELETE) &&
CHECK_FLAG(re->status, ROUTE_ENTRY_REMOVED)) {
/* Skip route that's been deleted */
goto done;
}
if ((re->type == dplane_ctx_get_type(ctx)) &&
(re->instance == dplane_ctx_get_instance(ctx))) {
result = true;
/* TODO -- we're using this extra test, but it's not
* exactly clear why.
*/
if (re->type == ZEBRA_ROUTE_STATIC &&
(re->distance != dplane_ctx_get_distance(ctx) ||
re->tag != dplane_ctx_get_tag(ctx))) {
result = false;
}
}
}
done:
return (result);
}
static void zebra_rib_fixup_system(struct route_node *rn)
{
struct route_entry *re;
RNODE_FOREACH_RE(rn, re) {
struct nexthop *nhop;
if (!RIB_SYSTEM_ROUTE(re))
continue;
if (CHECK_FLAG(re->status, ROUTE_ENTRY_REMOVED))
continue;
SET_FLAG(re->status, ROUTE_ENTRY_INSTALLED);
UNSET_FLAG(re->status, ROUTE_ENTRY_QUEUED);
for (ALL_NEXTHOPS_PTR(re->ng, nhop)) {
if (CHECK_FLAG(nhop->flags, NEXTHOP_FLAG_RECURSIVE))
continue;
SET_FLAG(nhop->flags, NEXTHOP_FLAG_FIB);
}
}
}
/*
* Update a route from a dplane context. This consolidates common code
* that can be used in processing of results from FIB updates, and in
* async notification processing.
* The return is 'true' if the installed nexthops changed; 'false' otherwise.
*/
static bool rib_update_re_from_ctx(struct route_entry *re,
struct route_node *rn,
struct zebra_dplane_ctx *ctx)
{
char dest_str[PREFIX_STRLEN] = "";
char nh_str[NEXTHOP_STRLEN];
struct nexthop *nexthop, *ctx_nexthop;
bool matched;
const struct nexthop_group *ctxnhg;
bool is_selected = false; /* Is 're' currently the selected re? */
bool changed_p = false; /* Change to nexthops? */
rib_dest_t *dest;
/* Note well: only capturing the prefix string if debug is enabled here;
* unconditional log messages will have to generate the string.
*/
if (IS_ZEBRA_DEBUG_RIB)
prefix2str(&(rn->p), dest_str, sizeof(dest_str));
dest = rib_dest_from_rnode(rn);
if (dest)
is_selected = (re == dest->selected_fib);
if (IS_ZEBRA_DEBUG_RIB_DETAILED)
zlog_debug("update_from_ctx: %u:%s: %sSELECTED",
re->vrf_id, dest_str, (is_selected ? "" : "NOT "));
/* Update zebra's nexthop FIB flag for each nexthop that was installed.
* If the installed set differs from the set requested by the rib/owner,
* we use the fib-specific nexthop-group to record the actual FIB
* status.
*/
/* Check both fib group and notif group for equivalence.
*
* Let's assume the nexthops are ordered here to save time.
*/
if (nexthop_group_equal(&re->fib_ng, dplane_ctx_get_ng(ctx)) == false) {
if (IS_ZEBRA_DEBUG_RIB_DETAILED) {
zlog_debug(
"%u:%s update_from_ctx: notif nh and fib nh mismatch",
re->vrf_id, dest_str);
}
matched = false;
} else
matched = true;
/* If the new FIB set matches the existing FIB set, we're done. */
if (matched) {
if (IS_ZEBRA_DEBUG_RIB)
zlog_debug("%u:%s update_from_ctx(): existing fib nhg, no change",
re->vrf_id, dest_str);
goto done;
} else if (re->fib_ng.nexthop) {
/*
* Free stale fib list and move on to check the rib nhg.
*/
if (IS_ZEBRA_DEBUG_RIB)
zlog_debug("%u:%s update_from_ctx(): replacing fib nhg",
re->vrf_id, dest_str);
nexthops_free(re->fib_ng.nexthop);
re->fib_ng.nexthop = NULL;
/* Note that the installed nexthops have changed */
changed_p = true;
} else {
if (IS_ZEBRA_DEBUG_RIB)
zlog_debug("%u:%s update_from_ctx(): no fib nhg",
re->vrf_id, dest_str);
}
/*
* Compare with the rib nexthop group. The comparison here is different:
* the RIB group may be a superset of the list installed in the FIB. We
* walk the RIB group, looking for the 'installable' candidate
* nexthops, and then check those against the set
* that is actually installed.
*
* Assume nexthops are ordered here as well.
*/
matched = true;
ctx_nexthop = dplane_ctx_get_ng(ctx)->nexthop;
/* Get the first `installed` one to check against.
* If the dataplane doesn't set these to be what was actually installed,
* it will just be whatever was in re->ng?
*/
if (CHECK_FLAG(ctx_nexthop->flags, NEXTHOP_FLAG_RECURSIVE)
|| !CHECK_FLAG(ctx_nexthop->flags, NEXTHOP_FLAG_ACTIVE))
ctx_nexthop = nexthop_next_active_resolved(ctx_nexthop);
for (ALL_NEXTHOPS_PTR(re->ng, nexthop)) {
if (CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_RECURSIVE))
continue;
if (!CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_ACTIVE))
continue;
/* Check for a FIB nexthop corresponding to the RIB nexthop */
if (nexthop_same(ctx_nexthop, nexthop) == false) {
/* If the FIB doesn't know about the nexthop,
* it's not installed
*/
if (IS_ZEBRA_DEBUG_RIB_DETAILED) {
nexthop2str(nexthop, nh_str, sizeof(nh_str));
zlog_debug(
"update_from_ctx: no notif match for rib nh %s",
nh_str);
}
matched = false;
if (CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB))
changed_p = true;
UNSET_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB);
/* Keep checking nexthops */
continue;
}
if (CHECK_FLAG(ctx_nexthop->flags, NEXTHOP_FLAG_FIB)) {
if (!CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB))
changed_p = true;
SET_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB);
} else {
if (CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB))
changed_p = true;
UNSET_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB);
}
ctx_nexthop = nexthop_next_active_resolved(ctx_nexthop);
}
/* If all nexthops were processed, we're done */
if (matched) {
if (IS_ZEBRA_DEBUG_RIB)
zlog_debug("%u:%s update_from_ctx(): rib nhg matched, changed '%s'",
re->vrf_id, dest_str,
(changed_p ? "true" : "false"));
goto done;
}
/* FIB nexthop set differs from the RIB set:
* create a fib-specific nexthop-group
*/
if (IS_ZEBRA_DEBUG_RIB)
zlog_debug("%u:%s update_from_ctx(): changed %s, adding new fib nhg",
re->vrf_id, dest_str,
(changed_p ? "true" : "false"));
ctxnhg = dplane_ctx_get_ng(ctx);
if (ctxnhg->nexthop)
copy_nexthops(&(re->fib_ng.nexthop), ctxnhg->nexthop, NULL);
else {
/* Bit of a special case when the fib has _no_ installed
* nexthops.
*/
nexthop = nexthop_new();
nexthop->type = NEXTHOP_TYPE_IPV4;
_nexthop_add(&(re->fib_ng.nexthop), nexthop);
}
done:
return changed_p;
}
/*
* Helper to locate a zebra route-node from a dplane context. This is used
* when processing dplane results, e.g. Note well: the route-node is returned
* with a ref held - route_unlock_node() must be called eventually.
*/
static struct route_node *
rib_find_rn_from_ctx(const struct zebra_dplane_ctx *ctx)
{
struct route_table *table = NULL;
struct route_node *rn = NULL;
const struct prefix *dest_pfx, *src_pfx;
/* Locate rn and re(s) from ctx */
table = zebra_vrf_table_with_table_id(dplane_ctx_get_afi(ctx),
dplane_ctx_get_safi(ctx),
dplane_ctx_get_vrf(ctx),
dplane_ctx_get_table(ctx));
if (table == NULL) {
if (IS_ZEBRA_DEBUG_DPLANE) {
zlog_debug("Failed to find route for ctx: no table for afi %d, safi %d, vrf %u",
dplane_ctx_get_afi(ctx),
dplane_ctx_get_safi(ctx),
dplane_ctx_get_vrf(ctx));
}
goto done;
}
dest_pfx = dplane_ctx_get_dest(ctx);
src_pfx = dplane_ctx_get_src(ctx);
rn = srcdest_rnode_get(table, dest_pfx,
src_pfx ? (struct prefix_ipv6 *)src_pfx : NULL);
done:
return rn;
}
/*
* Route-update results processing after async dataplane update.
*/
static void rib_process_result(struct zebra_dplane_ctx *ctx)
{
struct zebra_vrf *zvrf = NULL;
struct route_node *rn = NULL;
struct route_entry *re = NULL, *old_re = NULL, *rib;
bool is_update = false;
char dest_str[PREFIX_STRLEN] = "";
enum dplane_op_e op;
enum zebra_dplane_result status;
const struct prefix *dest_pfx, *src_pfx;
uint32_t seq;
bool fib_changed = false;
zvrf = vrf_info_lookup(dplane_ctx_get_vrf(ctx));
dest_pfx = dplane_ctx_get_dest(ctx);
/* Note well: only capturing the prefix string if debug is enabled here;
* unconditional log messages will have to generate the string.
*/
if (IS_ZEBRA_DEBUG_DPLANE)
prefix2str(dest_pfx, dest_str, sizeof(dest_str));
/* Locate rn and re(s) from ctx */
rn = rib_find_rn_from_ctx(ctx);
if (rn == NULL) {
if (IS_ZEBRA_DEBUG_DPLANE) {
zlog_debug("Failed to process dplane results: no route for %u:%s",
dplane_ctx_get_vrf(ctx), dest_str);
}
goto done;
}
srcdest_rnode_prefixes(rn, &dest_pfx, &src_pfx);
op = dplane_ctx_get_op(ctx);
status = dplane_ctx_get_status(ctx);
if (IS_ZEBRA_DEBUG_DPLANE_DETAIL)
zlog_debug("%u:%s Processing dplane ctx %p, op %s result %s",
dplane_ctx_get_vrf(ctx), dest_str, ctx,
dplane_op2str(op), dplane_res2str(status));
/*
* Update is a bit of a special case, where we may have both old and new
* routes to post-process.
*/
is_update = dplane_ctx_is_update(ctx);
/*
* Take a pass through the routes, look for matches with the context
* info.
*/
RNODE_FOREACH_RE(rn, rib) {
if (re == NULL) {
if (rib_route_match_ctx(rib, ctx, false))
re = rib;
}
/* Check for old route match */
if (is_update && (old_re == NULL)) {
if (rib_route_match_ctx(rib, ctx, true /*is_update*/))
old_re = rib;
}
/* Have we found the routes we need to work on? */
if (re && ((!is_update || old_re)))
break;
}
seq = dplane_ctx_get_seq(ctx);
/*
* Check sequence number(s) to detect stale results before continuing
*/
if (re) {
if (re->dplane_sequence != seq) {
if (IS_ZEBRA_DEBUG_DPLANE_DETAIL)
zlog_debug("%u:%s Stale dplane result for re %p",
dplane_ctx_get_vrf(ctx),
dest_str, re);
} else
UNSET_FLAG(re->status, ROUTE_ENTRY_QUEUED);
}
if (old_re) {
if (old_re->dplane_sequence != dplane_ctx_get_old_seq(ctx)) {
if (IS_ZEBRA_DEBUG_DPLANE_DETAIL)
zlog_debug("%u:%s Stale dplane result for old_re %p",
dplane_ctx_get_vrf(ctx),
dest_str, old_re);
} else
UNSET_FLAG(old_re->status, ROUTE_ENTRY_QUEUED);
}
switch (op) {
case DPLANE_OP_ROUTE_INSTALL:
case DPLANE_OP_ROUTE_UPDATE:
if (status == ZEBRA_DPLANE_REQUEST_SUCCESS) {
if (re) {
UNSET_FLAG(re->status, ROUTE_ENTRY_FAILED);
SET_FLAG(re->status, ROUTE_ENTRY_INSTALLED);
}
/*
* On an update operation from the same route type
* context retrieval currently has no way to know
* which was the old and which was the new.
* So don't unset our flags that we just set.
* We know redistribution is ok because the
* old_re in this case is used for nothing
* more than knowing whom to contact if necessary.
*/
if (old_re && old_re != re) {
UNSET_FLAG(old_re->status, ROUTE_ENTRY_FAILED);
UNSET_FLAG(old_re->status,
ROUTE_ENTRY_INSTALLED);
}
/* Update zebra route based on the results in
* the context struct.
*/
if (re) {
fib_changed =
rib_update_re_from_ctx(re, rn, ctx);
if (!fib_changed) {
if (IS_ZEBRA_DEBUG_DPLANE_DETAIL)
zlog_debug("%u:%s no fib change for re",
dplane_ctx_get_vrf(
ctx),
dest_str);
}
/* Redistribute */
redistribute_update(dest_pfx, src_pfx,
re, old_re);
}
/*
* System routes are weird in that they
* allow multiple to be installed that match
* to the same prefix, so after we get the
* result we need to clean them up so that
* we can actually use them.
*/
if ((re && RIB_SYSTEM_ROUTE(re)) ||
(old_re && RIB_SYSTEM_ROUTE(old_re)))
zebra_rib_fixup_system(rn);
if (zvrf)
zvrf->installs++;
/* Notify route owner */
zsend_route_notify_owner_ctx(ctx, ZAPI_ROUTE_INSTALLED);
} else {
if (re) {
SET_FLAG(re->status, ROUTE_ENTRY_FAILED);
UNSET_FLAG(re->status, ROUTE_ENTRY_INSTALLED);
} if (old_re)
SET_FLAG(old_re->status, ROUTE_ENTRY_FAILED);
if (re)
zsend_route_notify_owner(re, dest_pfx,
ZAPI_ROUTE_FAIL_INSTALL);
zlog_warn("%u:%s: Route install failed",
dplane_ctx_get_vrf(ctx),
prefix2str(dest_pfx,
dest_str, sizeof(dest_str)));
}
break;
case DPLANE_OP_ROUTE_DELETE:
if (re)
SET_FLAG(re->status, ROUTE_ENTRY_FAILED);
/*
* In the delete case, the zebra core datastructs were
* updated (or removed) at the time the delete was issued,
* so we're just notifying the route owner.
*/
if (status == ZEBRA_DPLANE_REQUEST_SUCCESS) {
if (re) {
UNSET_FLAG(re->status, ROUTE_ENTRY_INSTALLED);
UNSET_FLAG(re->status, ROUTE_ENTRY_FAILED);
}
zsend_route_notify_owner_ctx(ctx, ZAPI_ROUTE_REMOVED);
if (zvrf)
zvrf->removals++;
} else {
if (re)
SET_FLAG(re->status, ROUTE_ENTRY_FAILED);
zsend_route_notify_owner_ctx(ctx,
ZAPI_ROUTE_REMOVE_FAIL);
zlog_warn("%u:%s: Route Deletion failure",
dplane_ctx_get_vrf(ctx),
prefix2str(dest_pfx,
dest_str, sizeof(dest_str)));
}
/*
* System routes are weird in that they
* allow multiple to be installed that match
* to the same prefix, so after we get the
* result we need to clean them up so that
* we can actually use them.
*/
if ((re && RIB_SYSTEM_ROUTE(re)) ||
(old_re && RIB_SYSTEM_ROUTE(old_re)))
zebra_rib_fixup_system(rn);
break;
default:
break;
}
zebra_rib_evaluate_rn_nexthops(rn, seq);
zebra_rib_evaluate_mpls(rn);
done:
if (rn)
route_unlock_node(rn);
/* Return context to dataplane module */
dplane_ctx_fini(&ctx);
}
/*
* Handle notification from async dataplane: the dataplane has detected
* some change to a route, and notifies zebra so that the control plane
* can reflect that change.
*/
static void rib_process_dplane_notify(struct zebra_dplane_ctx *ctx)
{
struct route_node *rn = NULL;
struct route_entry *re = NULL;
struct nexthop *nexthop;
char dest_str[PREFIX_STRLEN] = "";
const struct prefix *dest_pfx, *src_pfx;
rib_dest_t *dest;
bool fib_changed = false;
bool debug_p = IS_ZEBRA_DEBUG_DPLANE | IS_ZEBRA_DEBUG_RIB;
int start_count, end_count;
dest_pfx = dplane_ctx_get_dest(ctx);
/* Note well: only capturing the prefix string if debug is enabled here;
* unconditional log messages will have to generate the string.
*/
if (debug_p)
prefix2str(dest_pfx, dest_str, sizeof(dest_str));
/* Locate rn and re(s) from ctx */
rn = rib_find_rn_from_ctx(ctx);
if (rn == NULL) {
if (debug_p) {
zlog_debug("Failed to process dplane notification: no routes for %u:%s",
dplane_ctx_get_vrf(ctx), dest_str);
}
goto done;
}
dest = rib_dest_from_rnode(rn);
srcdest_rnode_prefixes(rn, &dest_pfx, &src_pfx);
if (debug_p)
zlog_debug("%u:%s Processing dplane notif ctx %p",
dplane_ctx_get_vrf(ctx), dest_str, ctx);
/*
* Take a pass through the routes, look for matches with the context
* info.
*/
RNODE_FOREACH_RE(rn, re) {
if (rib_route_match_ctx(re, ctx, false /*!update*/))
break;
}
/* No match? Nothing we can do */
if (re == NULL) {
if (debug_p)
zlog_debug("%u:%s Unable to process dplane notification: no entry for type %s",
dplane_ctx_get_vrf(ctx), dest_str,
zebra_route_string(
dplane_ctx_get_type(ctx)));
goto done;
}
/* Ensure we clear the QUEUED flag */
UNSET_FLAG(re->status, ROUTE_ENTRY_QUEUED);
/* Is this a notification that ... matters? We only really care about
* the route that is currently selected for installation.
*/
if (re != dest->selected_fib) {
/* TODO -- don't skip processing entirely? We might like to
* at least report on the event.
*/
if (debug_p)
zlog_debug("%u:%s dplane notif, but type %s not selected_fib",
dplane_ctx_get_vrf(ctx), dest_str,
zebra_route_string(
dplane_ctx_get_type(ctx)));
goto done;
}
/* We'll want to determine whether the installation status of the
* route has changed: we'll check the status before processing,
* and then again if there's been a change.
*/
start_count = 0;
for (ALL_NEXTHOPS_PTR(rib_active_nhg(re), nexthop)) {
if (CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB))
start_count++;
}
/* Update zebra's nexthop FIB flags based on the context struct's
* nexthops.
*/
fib_changed = rib_update_re_from_ctx(re, rn, ctx);
if (!fib_changed) {
if (debug_p)
zlog_debug("%u:%s No change from dplane notification",
dplane_ctx_get_vrf(ctx), dest_str);
goto done;
}
/*
* Perform follow-up work if the actual status of the prefix
* changed.
*/
end_count = 0;
for (ALL_NEXTHOPS_PTR(rib_active_nhg(re), nexthop)) {
if (CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB))
end_count++;
}
/* Various fib transitions: changed nexthops; from installed to
* not-installed; or not-installed to installed.
*/
if (start_count > 0 && end_count > 0) {
if (debug_p)
zlog_debug("%u:%s applied nexthop changes from dplane notification",
dplane_ctx_get_vrf(ctx), dest_str);
/* Changed nexthops - update kernel/others */
dplane_route_notif_update(rn, re,
DPLANE_OP_ROUTE_UPDATE, ctx);
} else if (start_count == 0 && end_count > 0) {
if (debug_p)
zlog_debug("%u:%s installed transition from dplane notification",
dplane_ctx_get_vrf(ctx), dest_str);
/* We expect this to be the selected route, so we want
* to tell others about this transition.
*/
SET_FLAG(re->status, ROUTE_ENTRY_INSTALLED);
/* Changed nexthops - update kernel/others */
dplane_route_notif_update(rn, re, DPLANE_OP_ROUTE_INSTALL, ctx);
/* Redistribute, lsp, and nht update */
redistribute_update(dest_pfx, src_pfx, re, NULL);
zebra_rib_evaluate_rn_nexthops(
rn, zebra_router_get_next_sequence());
zebra_rib_evaluate_mpls(rn);
} else if (start_count > 0 && end_count == 0) {
if (debug_p)
zlog_debug("%u:%s un-installed transition from dplane notification",
dplane_ctx_get_vrf(ctx), dest_str);
/* Transition from _something_ installed to _nothing_
* installed.
*/
/* We expect this to be the selected route, so we want
* to tell others about this transistion.
*/
UNSET_FLAG(re->status, ROUTE_ENTRY_INSTALLED);
/* Changed nexthops - update kernel/others */
dplane_route_notif_update(rn, re, DPLANE_OP_ROUTE_DELETE, ctx);
/* Redistribute, lsp, and nht update */
redistribute_delete(dest_pfx, src_pfx, re, NULL);
zebra_rib_evaluate_rn_nexthops(
rn, zebra_router_get_next_sequence());
zebra_rib_evaluate_mpls(rn);
}
done:
if (rn)
route_unlock_node(rn);
/* Return context to dataplane module */
dplane_ctx_fini(&ctx);
}
static void process_subq_nhg(struct listnode *lnode)
{
struct nhg_ctx *ctx = NULL;
uint8_t qindex = route_info[ZEBRA_ROUTE_NHG].meta_q_map;
ctx = listgetdata(lnode);
if (!ctx)
return;
if (IS_ZEBRA_DEBUG_RIB_DETAILED)
zlog_debug("NHG Context id=%u dequeued from sub-queue %u",
ctx->id, qindex);
rib_nhg_process(ctx);
}
static void process_subq_route(struct listnode *lnode, uint8_t qindex)
{
struct route_node *rnode = NULL;
rib_dest_t *dest = NULL;
struct zebra_vrf *zvrf = NULL;
rnode = listgetdata(lnode);
dest = rib_dest_from_rnode(rnode);
if (dest)
zvrf = rib_dest_vrf(dest);
rib_process(rnode);
if (IS_ZEBRA_DEBUG_RIB_DETAILED) {
char buf[SRCDEST2STR_BUFFER];
srcdest_rnode2str(rnode, buf, sizeof(buf));
zlog_debug("%u:%s: rn %p dequeued from sub-queue %u",
zvrf ? zvrf_id(zvrf) : 0, buf, rnode, qindex);
}
if (rnode->info)
UNSET_FLAG(rib_dest_from_rnode(rnode)->flags,
RIB_ROUTE_QUEUED(qindex));
#if 0
else
{
zlog_debug ("%s: called for route_node (%p, %d) with no ribs",
__func__, rnode, rnode->lock);
zlog_backtrace(LOG_DEBUG);
}
#endif
route_unlock_node(rnode);
}
/* Take a list of route_node structs and return 1, if there was a record
* picked from it and processed by rib_process(). Don't process more,
* than one RN record; operate only in the specified sub-queue.
*/
static unsigned int process_subq(struct list *subq, uint8_t qindex)
{
struct listnode *lnode = listhead(subq);
if (!lnode)
return 0;
if (qindex == route_info[ZEBRA_ROUTE_NHG].meta_q_map)
process_subq_nhg(lnode);
else
process_subq_route(lnode, qindex);
list_delete_node(subq, lnode);
return 1;
}
/* Dispatch the meta queue by picking, processing and unlocking the next RN from
* a non-empty sub-queue with lowest priority. wq is equal to zebra->ribq and
* data
* is pointed to the meta queue structure.
*/
static wq_item_status meta_queue_process(struct work_queue *dummy, void *data)
{
struct meta_queue *mq = data;
unsigned i;
uint32_t queue_len, queue_limit;
/* Ensure there's room for more dataplane updates */
queue_limit = dplane_get_in_queue_limit();
queue_len = dplane_get_in_queue_len();
if (queue_len > queue_limit) {
if (IS_ZEBRA_DEBUG_RIB_DETAILED)
zlog_debug("rib queue: dplane queue len %u, limit %u, retrying",
queue_len, queue_limit);
/* Ensure that the meta-queue is actually enqueued */
if (work_queue_empty(zrouter.ribq))
work_queue_add(zrouter.ribq, zrouter.mq);
return WQ_QUEUE_BLOCKED;
}
for (i = 0; i < MQ_SIZE; i++)
if (process_subq(mq->subq[i], i)) {
mq->size--;
break;
}
return mq->size ? WQ_REQUEUE : WQ_SUCCESS;
}
/*
* Look into the RN and queue it into the highest priority queue
* at this point in time for processing.
*
* We will enqueue a route node only once per invocation.
*
* There are two possibilities here that should be kept in mind.
* If the original invocation has not been pulled off for processing
* yet, A subsuquent invocation can have a route entry with a better
* meta queue index value and we can have a situation where
* we might have the same node enqueued 2 times. Not necessarily
* an optimal situation but it should be ok.
*
* The other possibility is that the original invocation has not
* been pulled off for processing yet, A subsusquent invocation
* doesn't have a route_entry with a better meta-queue and the
* original metaqueue index value will win and we'll end up with
* the route node enqueued once.
*/
static int rib_meta_queue_add(struct meta_queue *mq, void *data)
{
struct route_node *rn = NULL;
struct route_entry *re = NULL, *curr_re = NULL;
uint8_t qindex = MQ_SIZE, curr_qindex = MQ_SIZE;
rn = (struct route_node *)data;
RNODE_FOREACH_RE (rn, curr_re) {
curr_qindex = route_info[curr_re->type].meta_q_map;
if (curr_qindex <= qindex) {
re = curr_re;
qindex = curr_qindex;
}
}
if (!re)
return -1;
/* Invariant: at this point we always have rn->info set. */
if (CHECK_FLAG(rib_dest_from_rnode(rn)->flags,
RIB_ROUTE_QUEUED(qindex))) {
if (IS_ZEBRA_DEBUG_RIB_DETAILED)
rnode_debug(rn, re->vrf_id,
"rn %p is already queued in sub-queue %u",
(void *)rn, qindex);
return -1;
}
SET_FLAG(rib_dest_from_rnode(rn)->flags, RIB_ROUTE_QUEUED(qindex));
listnode_add(mq->subq[qindex], rn);
route_lock_node(rn);
mq->size++;
if (IS_ZEBRA_DEBUG_RIB_DETAILED)
rnode_debug(rn, re->vrf_id, "queued rn %p into sub-queue %u",
(void *)rn, qindex);
return 0;
}
static int rib_meta_queue_nhg_add(struct meta_queue *mq, void *data)
{
struct nhg_ctx *ctx = NULL;
uint8_t qindex = route_info[ZEBRA_ROUTE_NHG].meta_q_map;
ctx = (struct nhg_ctx *)data;
if (!ctx)
return -1;
listnode_add(mq->subq[qindex], ctx);
mq->size++;
if (IS_ZEBRA_DEBUG_RIB_DETAILED)
zlog_debug("NHG Context id=%u queued into sub-queue %u",
ctx->id, qindex);
return 0;
}
static int mq_add_handler(void *data,
int (*mq_add_func)(struct meta_queue *mq, void *data))
{
if (zrouter.ribq == NULL) {
flog_err(EC_ZEBRA_WQ_NONEXISTENT,
"%s: work_queue does not exist!", __func__);
return -1;
}
/*
* The RIB queue should normally be either empty or holding the only
* work_queue_item element. In the latter case this element would
* hold a pointer to the meta queue structure, which must be used to
* actually queue the route nodes to process. So create the MQ
* holder, if necessary, then push the work into it in any case.
* This semantics was introduced after 0.99.9 release.
*/
if (work_queue_empty(zrouter.ribq))
work_queue_add(zrouter.ribq, zrouter.mq);
return mq_add_func(zrouter.mq, data);
}
/* Add route_node to work queue and schedule processing */
int rib_queue_add(struct route_node *rn)
{
assert(rn);
/* Pointless to queue a route_node with no RIB entries to add or remove
*/
if (!rnode_to_ribs(rn)) {
zlog_debug("%s: called for route_node (%p, %d) with no ribs",
__func__, (void *)rn, rn->lock);
zlog_backtrace(LOG_DEBUG);
return -1;
}
return mq_add_handler(rn, &rib_meta_queue_add);
}
int rib_queue_nhg_add(struct nhg_ctx *ctx)
{
assert(ctx);
return mq_add_handler(ctx, &rib_meta_queue_nhg_add);
}
/* Create new meta queue.
A destructor function doesn't seem to be necessary here.
*/
static struct meta_queue *meta_queue_new(void)
{
struct meta_queue *new;
unsigned i;
new = XCALLOC(MTYPE_WORK_QUEUE, sizeof(struct meta_queue));
for (i = 0; i < MQ_SIZE; i++) {
new->subq[i] = list_new();
assert(new->subq[i]);
}
return new;
}
void meta_queue_free(struct meta_queue *mq)
{
unsigned i;
for (i = 0; i < MQ_SIZE; i++)
list_delete(&mq->subq[i]);
XFREE(MTYPE_WORK_QUEUE, mq);
}
/* initialise zebra rib work queue */
static void rib_queue_init(void)
{
if (!(zrouter.ribq = work_queue_new(zrouter.master,
"route_node processing"))) {
flog_err(EC_ZEBRA_WQ_NONEXISTENT,
"%s: could not initialise work queue!", __func__);
return;
}
/* fill in the work queue spec */
zrouter.ribq->spec.workfunc = &meta_queue_process;
zrouter.ribq->spec.errorfunc = NULL;
zrouter.ribq->spec.completion_func = NULL;
/* XXX: TODO: These should be runtime configurable via vty */
zrouter.ribq->spec.max_retries = 3;
zrouter.ribq->spec.hold = ZEBRA_RIB_PROCESS_HOLD_TIME;
zrouter.ribq->spec.retry = ZEBRA_RIB_PROCESS_RETRY_TIME;
if (!(zrouter.mq = meta_queue_new())) {
flog_err(EC_ZEBRA_WQ_NONEXISTENT,
"%s: could not initialise meta queue!", __func__);
return;
}
return;
}
rib_dest_t *zebra_rib_create_dest(struct route_node *rn)
{
rib_dest_t *dest;
dest = XCALLOC(MTYPE_RIB_DEST, sizeof(rib_dest_t));
rnh_list_init(&dest->nht);
route_lock_node(rn); /* rn route table reference */
rn->info = dest;
dest->rnode = rn;
return dest;
}
/* RIB updates are processed via a queue of pointers to route_nodes.
*
* The queue length is bounded by the maximal size of the routing table,
* as a route_node will not be requeued, if already queued.
*
* REs are submitted via rib_addnode or rib_delnode which set minimal
* state, or static_install_route (when an existing RE is updated)
* and then submit route_node to queue for best-path selection later.
* Order of add/delete state changes are preserved for any given RE.
*
* Deleted REs are reaped during best-path selection.
*
* rib_addnode
* |-> rib_link or unset ROUTE_ENTRY_REMOVE |->Update kernel with
* |-------->| | best RE, if required
* | |
* static_install->|->rib_addqueue...... -> rib_process
* | |
* |-------->| |-> rib_unlink
* |-> set ROUTE_ENTRY_REMOVE |
* rib_delnode (RE freed)
*
* The 'info' pointer of a route_node points to a rib_dest_t
* ('dest'). Queueing state for a route_node is kept on the dest. The
* dest is created on-demand by rib_link() and is kept around at least
* as long as there are ribs hanging off it (@see rib_gc_dest()).
*
* Refcounting (aka "locking" throughout the GNU Zebra and Quagga code):
*
* - route_nodes: refcounted by:
* - dest attached to route_node:
* - managed by: rib_link/rib_gc_dest
* - route_node processing queue
* - managed by: rib_addqueue, rib_process.
*
*/
/* Add RE to head of the route node. */
static void rib_link(struct route_node *rn, struct route_entry *re, int process)
{
rib_dest_t *dest;
afi_t afi;
const char *rmap_name;
assert(re && rn);
dest = rib_dest_from_rnode(rn);
if (!dest) {
if (IS_ZEBRA_DEBUG_RIB_DETAILED)
rnode_debug(rn, re->vrf_id, "rn %p adding dest", rn);
dest = zebra_rib_create_dest(rn);
}
re_list_add_head(&dest->routes, re);
afi = (rn->p.family == AF_INET)
? AFI_IP
: (rn->p.family == AF_INET6) ? AFI_IP6 : AFI_MAX;
if (is_zebra_import_table_enabled(afi, re->vrf_id, re->table)) {
struct zebra_vrf *zvrf = zebra_vrf_lookup_by_id(re->vrf_id);
rmap_name = zebra_get_import_table_route_map(afi, re->table);
zebra_add_import_table_entry(zvrf, rn, re, rmap_name);
} else if (process)
rib_queue_add(rn);
}
static void rib_addnode(struct route_node *rn,
struct route_entry *re, int process)
{
/* RE node has been un-removed before route-node is processed.
* route_node must hence already be on the queue for processing..
*/
if (CHECK_FLAG(re->status, ROUTE_ENTRY_REMOVED)) {
if (IS_ZEBRA_DEBUG_RIB)
rnode_debug(rn, re->vrf_id, "rn %p, un-removed re %p",
(void *)rn, (void *)re);
UNSET_FLAG(re->status, ROUTE_ENTRY_REMOVED);
return;
}
rib_link(rn, re, process);
}
/*
* rib_unlink
*
* Detach a rib structure from a route_node.
*
* Note that a call to rib_unlink() should be followed by a call to
* rib_gc_dest() at some point. This allows a rib_dest_t that is no
* longer required to be deleted.
*/
void rib_unlink(struct route_node *rn, struct route_entry *re)
{
rib_dest_t *dest;
struct nhg_hash_entry *nhe = NULL;
assert(rn && re);
if (IS_ZEBRA_DEBUG_RIB)
rnode_debug(rn, re->vrf_id, "rn %p, re %p", (void *)rn,
(void *)re);
dest = rib_dest_from_rnode(rn);
re_list_del(&dest->routes, re);
if (dest->selected_fib == re)
dest->selected_fib = NULL;
if (re->nhe_id) {
nhe = zebra_nhg_lookup_id(re->nhe_id);
if (nhe)
zebra_nhg_decrement_ref(nhe);
} else if (re->ng)
nexthop_group_free_delete(&re->ng);
nexthops_free(re->fib_ng.nexthop);
XFREE(MTYPE_RE, re);
}
void rib_delnode(struct route_node *rn, struct route_entry *re)
{
afi_t afi;
if (IS_ZEBRA_DEBUG_RIB)
rnode_debug(rn, re->vrf_id, "rn %p, re %p, removing",
(void *)rn, (void *)re);
SET_FLAG(re->status, ROUTE_ENTRY_REMOVED);
afi = (rn->p.family == AF_INET)
? AFI_IP
: (rn->p.family == AF_INET6) ? AFI_IP6 : AFI_MAX;
if (is_zebra_import_table_enabled(afi, re->vrf_id, re->table)) {
struct zebra_vrf *zvrf = zebra_vrf_lookup_by_id(re->vrf_id);
zebra_del_import_table_entry(zvrf, rn, re);
/* Just clean up if non main table */
if (IS_ZEBRA_DEBUG_RIB) {
char buf[SRCDEST2STR_BUFFER];
srcdest_rnode2str(rn, buf, sizeof(buf));
zlog_debug("%u:%s: Freeing route rn %p, re %p (%s)",
re->vrf_id, buf, rn, re,
zebra_route_string(re->type));
}
rib_unlink(rn, re);
} else {
rib_queue_add(rn);
}
}
/* This function dumps the contents of a given RE entry into
* standard debug log. Calling function name and IP prefix in
* question are passed as 1st and 2nd arguments.
*/
void _route_entry_dump(const char *func, union prefixconstptr pp,
union prefixconstptr src_pp,
const struct route_entry *re)
{
const struct prefix *src_p = src_pp.p;
bool is_srcdst = src_p && src_p->prefixlen;
char straddr[PREFIX_STRLEN];
char srcaddr[PREFIX_STRLEN];
char nhname[PREFIX_STRLEN];
struct nexthop *nexthop;
zlog_debug("%s: dumping RE entry %p for %s%s%s vrf %u", func,
(const void *)re, prefix2str(pp, straddr, sizeof(straddr)),
is_srcdst ? " from " : "",
is_srcdst ? prefix2str(src_pp, srcaddr, sizeof(srcaddr))
: "",
re->vrf_id);
zlog_debug("%s: uptime == %lu, type == %u, instance == %d, table == %d",
straddr, (unsigned long)re->uptime, re->type, re->instance,
re->table);
zlog_debug(
"%s: metric == %u, mtu == %u, distance == %u, flags == %u, status == %u",
straddr, re->metric, re->mtu, re->distance, re->flags, re->status);
zlog_debug("%s: nexthop_num == %u, nexthop_active_num == %u", straddr,
nexthop_group_nexthop_num(re->ng),
nexthop_group_active_nexthop_num(re->ng));
for (ALL_NEXTHOPS_PTR(re->ng, nexthop)) {
struct interface *ifp;
struct vrf *vrf = vrf_lookup_by_id(nexthop->vrf_id);
switch (nexthop->type) {
case NEXTHOP_TYPE_BLACKHOLE:
sprintf(nhname, "Blackhole");
break;
case NEXTHOP_TYPE_IFINDEX:
ifp = if_lookup_by_index(nexthop->ifindex,
nexthop->vrf_id);
sprintf(nhname, "%s", ifp ? ifp->name : "Unknown");
break;
case NEXTHOP_TYPE_IPV4:
/* fallthrough */
case NEXTHOP_TYPE_IPV4_IFINDEX:
inet_ntop(AF_INET, &nexthop->gate, nhname,
INET6_ADDRSTRLEN);
break;
case NEXTHOP_TYPE_IPV6:
case NEXTHOP_TYPE_IPV6_IFINDEX:
inet_ntop(AF_INET6, &nexthop->gate, nhname,
INET6_ADDRSTRLEN);
break;
}
zlog_debug("%s: %s %s[%u] vrf %s(%u) with flags %s%s%s%s%s%s",
straddr, (nexthop->rparent ? " NH" : "NH"), nhname,
nexthop->ifindex, vrf ? vrf->name : "Unknown",
nexthop->vrf_id,
(CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_ACTIVE)
? "ACTIVE "
: ""),
(CHECK_FLAG(re->status, ROUTE_ENTRY_INSTALLED)
? "FIB "
: ""),
(CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_RECURSIVE)
? "RECURSIVE "
: ""),
(CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_ONLINK)
? "ONLINK "
: ""),
(CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_MATCHED)
? "MATCHED "
: ""),
(CHECK_FLAG(nexthop->flags, NEXTHOP_FLAG_DUPLICATE)
? "DUPLICATE "
: ""));
}
zlog_debug("%s: dump complete", straddr);
}
/* This is an exported helper to rtm_read() to dump the strange
* RE entry found by rib_lookup_ipv4_route()
*/
void rib_lookup_and_dump(struct prefix_ipv4 *p, vrf_id_t vrf_id)
{
struct route_table *table;
struct route_node *rn;
struct route_entry *re;
char prefix_buf[INET_ADDRSTRLEN];
/* Lookup table. */
table = zebra_vrf_table(AFI_IP, SAFI_UNICAST, vrf_id);
if (!table) {
flog_err(EC_ZEBRA_TABLE_LOOKUP_FAILED,
"%s:%u zebra_vrf_table() returned NULL", __func__,
vrf_id);
return;
}
/* Scan the RIB table for exactly matching RE entry. */
rn = route_node_lookup(table, (struct prefix *)p);
/* No route for this prefix. */
if (!rn) {
zlog_debug("%s:%u lookup failed for %s", __func__, vrf_id,
prefix2str((struct prefix *)p, prefix_buf,
sizeof(prefix_buf)));
return;
}
/* Unlock node. */
route_unlock_node(rn);
/* let's go */
RNODE_FOREACH_RE (rn, re) {
zlog_debug("%s:%u rn %p, re %p: %s, %s",
__func__, vrf_id,
(void *)rn, (void *)re,
(CHECK_FLAG(re->status, ROUTE_ENTRY_REMOVED)
? "removed"
: "NOT removed"),
(CHECK_FLAG(re->flags, ZEBRA_FLAG_SELECTED)
? "selected"
: "NOT selected"));
route_entry_dump(p, NULL, re);
}
}
/* Check if requested address assignment will fail due to another
* route being installed by zebra in FIB already. Take necessary
* actions, if needed: remove such a route from FIB and deSELECT
* corresponding RE entry. Then put affected RN into RIBQ head.
*/
void rib_lookup_and_pushup(struct prefix_ipv4 *p, vrf_id_t vrf_id)
{
struct route_table *table;
struct route_node *rn;
rib_dest_t *dest;
if (NULL == (table = zebra_vrf_table(AFI_IP, SAFI_UNICAST, vrf_id))) {
flog_err(EC_ZEBRA_TABLE_LOOKUP_FAILED,
"%s:%u zebra_vrf_table() returned NULL", __func__,
vrf_id);
return;
}
/* No matches would be the simplest case. */
if (NULL == (rn = route_node_lookup(table, (struct prefix *)p)))
return;
/* Unlock node. */
route_unlock_node(rn);
dest = rib_dest_from_rnode(rn);
/* Check all RE entries. In case any changes have to be done, requeue
* the RN into RIBQ head. If the routing message about the new connected
* route (generated by the IP address we are going to assign very soon)
* comes before the RIBQ is processed, the new RE entry will join
* RIBQ record already on head. This is necessary for proper
* revalidation
* of the rest of the RE.
*/
if (dest->selected_fib) {
if (IS_ZEBRA_DEBUG_RIB) {
char buf[PREFIX_STRLEN];
zlog_debug("%u:%s: freeing way for connected prefix",
dest->selected_fib->vrf_id,
prefix2str(&rn->p, buf, sizeof(buf)));
route_entry_dump(&rn->p, NULL, dest->selected_fib);
}
rib_uninstall(rn, dest->selected_fib);
rib_queue_add(rn);
}
}
int rib_add_multipath(afi_t afi, safi_t safi, struct prefix *p,
struct prefix_ipv6 *src_p, struct route_entry *re)
{
struct nhg_hash_entry *nhe = NULL;
struct route_table *table;
struct route_node *rn;
struct route_entry *same = NULL;
int ret = 0;
if (!re)
return 0;
assert(!src_p || !src_p->prefixlen || afi == AFI_IP6);
/* Lookup table. */
table = zebra_vrf_table_with_table_id(afi, safi, re->vrf_id, re->table);
if (!table) {
if (re->ng)
nexthop_group_free_delete(&re->ng);
XFREE(MTYPE_RE, re);
return 0;
}
if (re->nhe_id) {
nhe = zebra_nhg_lookup_id(re->nhe_id);
if (!nhe) {
flog_err(
EC_ZEBRA_TABLE_LOOKUP_FAILED,
"Zebra failed to find the nexthop hash entry for id=%u in a route entry",
re->nhe_id);
XFREE(MTYPE_RE, re);
return -1;
}
} else {
nhe = zebra_nhg_rib_find(0, re->ng, afi);
/*
* The nexthops got copied over into an nhe,
* so free them now.
*/
nexthop_group_free_delete(&re->ng);
if (!nhe) {
char buf[PREFIX_STRLEN] = "";
char buf2[PREFIX_STRLEN] = "";
flog_err(
EC_ZEBRA_TABLE_LOOKUP_FAILED,
"Zebra failed to find or create a nexthop hash entry for %s%s%s",
prefix2str(p, buf, sizeof(buf)),
src_p ? " from " : "",
src_p ? prefix2str(src_p, buf2, sizeof(buf2))
: "");
XFREE(MTYPE_RE, re);
return -1;
}
}
/*
* Attach the re to the nhe's nexthop group.
*
* TODO: This will need to change when we start getting IDs from upper
* level protocols, as the refcnt might be wrong, since it checks
* if old_id != new_id.
*/
zebra_nhg_re_update_ref(re, nhe);
/* Make it sure prefixlen is applied to the prefix. */
apply_mask(p);
if (src_p)
apply_mask_ipv6(src_p);
/* Set default distance by route type. */
if (re->distance == 0)
re->distance = route_distance(re->type);
/* Lookup route node.*/
rn = srcdest_rnode_get(table, p, src_p);
/*
* If same type of route are installed, treat it as a implicit
* withdraw.
* If the user has specified the No route replace semantics
* for the install don't do a route replace.
*/
RNODE_FOREACH_RE (rn, same) {
if (CHECK_FLAG(same->status, ROUTE_ENTRY_REMOVED))
continue;
if (same->type != re->type)
continue;
if (same->instance != re->instance)
continue;
if (same->type == ZEBRA_ROUTE_KERNEL
&& same->metric != re->metric)
continue;
if (CHECK_FLAG(re->flags, ZEBRA_FLAG_RR_USE_DISTANCE) &&
same->distance != re->distance)
continue;
/*
* We should allow duplicate connected routes
* because of IPv6 link-local routes and unnumbered
* interfaces on Linux.
*/
if (same->type != ZEBRA_ROUTE_CONNECT)
break;
}
/* If this route is kernel/connected route, notify the dataplane. */
if (RIB_SYSTEM_ROUTE(re)) {
/* Notify dataplane */
dplane_sys_route_add(rn, re);
}
/* Link new re to node.*/
if (IS_ZEBRA_DEBUG_RIB) {
rnode_debug(rn, re->vrf_id,
"Inserting route rn %p, re %p (%s) existing %p",
rn, re, zebra_route_string(re->type), same);
if (IS_ZEBRA_DEBUG_RIB_DETAILED)
route_entry_dump(p, src_p, re);
}
SET_FLAG(re->status, ROUTE_ENTRY_CHANGED);
rib_addnode(rn, re, 1);
ret = 1;
/* Free implicit route.*/
if (same) {
rib_delnode(rn, same);
ret = -1;
}
route_unlock_node(rn);
return ret;
}
void rib_delete(afi_t afi, safi_t safi, vrf_id_t vrf_id, int type,
unsigned short instance, int flags, struct prefix *p,
struct prefix_ipv6 *src_p, const struct nexthop *nh,
uint32_t nhe_id, uint32_t table_id, uint32_t metric,
uint8_t distance, bool fromkernel)
{
struct route_table *table;
struct route_node *rn;
struct route_entry *re;
struct route_entry *fib = NULL;
struct route_entry *same = NULL;
struct nexthop *rtnh;
char buf2[INET6_ADDRSTRLEN];
rib_dest_t *dest;
assert(!src_p || !src_p->prefixlen || afi == AFI_IP6);
/* Lookup table. */
table = zebra_vrf_table_with_table_id(afi, safi, vrf_id, table_id);
if (!table)
return;
/* Apply mask. */
apply_mask(p);
if (src_p)
apply_mask_ipv6(src_p);
/* Lookup route node. */
rn = srcdest_rnode_lookup(table, p, src_p);
if (!rn) {
char dst_buf[PREFIX_STRLEN], src_buf[PREFIX_STRLEN];
prefix2str(p, dst_buf, sizeof(dst_buf));
if (src_p && src_p->prefixlen)
prefix2str(src_p, src_buf, sizeof(src_buf));
else
src_buf[0] = '\0';
if (IS_ZEBRA_DEBUG_RIB) {
struct vrf *vrf = vrf_lookup_by_id(vrf_id);
zlog_debug("%s[%d]:%s%s%s doesn't exist in rib",
vrf->name, table_id, dst_buf,
(src_buf[0] != '\0') ? " from " : "",
src_buf);
}
return;
}
dest = rib_dest_from_rnode(rn);
fib = dest->selected_fib;
/* Lookup same type route. */
RNODE_FOREACH_RE (rn, re) {
if (CHECK_FLAG(re->status, ROUTE_ENTRY_REMOVED))
continue;
if (re->type != type)
continue;
if (re->instance != instance)
continue;
if (CHECK_FLAG(re->flags, ZEBRA_FLAG_RR_USE_DISTANCE) &&
distance != re->distance)
continue;
if (re->type == ZEBRA_ROUTE_KERNEL && re->metric != metric)
continue;
if (re->type == ZEBRA_ROUTE_CONNECT && (rtnh = re->ng->nexthop)
&& rtnh->type == NEXTHOP_TYPE_IFINDEX && nh) {
if (rtnh->ifindex != nh->ifindex)
continue;
same = re;
break;
}
/* Make sure that the route found has the same gateway. */
else if (nhe_id && re->nhe_id == nhe_id) {
same = re;
break;
} else {
if (nh == NULL) {
same = re;
break;
}
for (ALL_NEXTHOPS_PTR(re->ng, rtnh))
/*
* No guarantee all kernel send nh with labels
* on delete.
*/
if (nexthop_same_no_labels(rtnh, nh)) {
same = re;
break;
}
if (same)
break;
}
}
/* If same type of route can't be found and this message is from
kernel. */
if (!same) {
/*
* In the past(HA!) we could get here because
* we were receiving a route delete from the
* kernel and we're not marking the proto
* as coming from it's appropriate originator.
* Now that we are properly noticing the fact
* that the kernel has deleted our route we
* are not going to get called in this path
* I am going to leave this here because
* this might still work this way on non-linux
* platforms as well as some weird state I have
* not properly thought of yet.
* If we can show that this code path is
* dead then we can remove it.
*/
if (fib && CHECK_FLAG(flags, ZEBRA_FLAG_SELFROUTE)) {
if (IS_ZEBRA_DEBUG_RIB) {
rnode_debug(rn, vrf_id,
"rn %p, re %p (%s) was deleted from kernel, adding",
rn, fib,
zebra_route_string(fib->type));
}
if (allow_delete) {
UNSET_FLAG(fib->status, ROUTE_ENTRY_INSTALLED);
/* Unset flags. */
for (rtnh = fib->ng->nexthop; rtnh;
rtnh = rtnh->next)
UNSET_FLAG(rtnh->flags,
NEXTHOP_FLAG_FIB);
/*
* This is a non FRR route
* as such we should mark
* it as deleted
*/
dest->selected_fib = NULL;
} else {
/* This means someone else, other than Zebra,
* has deleted
* a Zebra router from the kernel. We will add
* it back */
rib_install_kernel(rn, fib, NULL);
}
} else {
if (IS_ZEBRA_DEBUG_RIB) {
if (nh)
rnode_debug(
rn, vrf_id,
"via %s ifindex %d type %d "
"doesn't exist in rib",
inet_ntop(afi2family(afi),
&nh->gate, buf2,
sizeof(buf2)),
nh->ifindex, type);
else
rnode_debug(
rn, vrf_id,
"type %d doesn't exist in rib",
type);
}
route_unlock_node(rn);
return;
}
}
if (same) {
if (fromkernel && CHECK_FLAG(flags, ZEBRA_FLAG_SELFROUTE)
&& !allow_delete) {
rib_install_kernel(rn, same, NULL);
route_unlock_node(rn);
return;
}
/* Special handling for IPv4 or IPv6 routes sourced from
* EVPN - the nexthop (and associated MAC) need to be
* uninstalled if no more refs.
*/
if (CHECK_FLAG(flags, ZEBRA_FLAG_EVPN_ROUTE)) {
struct nexthop *tmp_nh;
for (ALL_NEXTHOPS_PTR(re->ng, tmp_nh)) {
struct ipaddr vtep_ip;
memset(&vtep_ip, 0, sizeof(struct ipaddr));
if (afi == AFI_IP) {
vtep_ip.ipa_type = IPADDR_V4;
memcpy(&(vtep_ip.ipaddr_v4),
&(tmp_nh->gate.ipv4),
sizeof(struct in_addr));
} else {
vtep_ip.ipa_type = IPADDR_V6;
memcpy(&(vtep_ip.ipaddr_v6),
&(tmp_nh->gate.ipv6),
sizeof(struct in6_addr));
}
zebra_vxlan_evpn_vrf_route_del(re->vrf_id,
&vtep_ip, p);
}
}
/* Notify dplane if system route changes */
if (RIB_SYSTEM_ROUTE(re))
dplane_sys_route_del(rn, same);
rib_delnode(rn, same);
}
route_unlock_node(rn);
return;
}
int rib_add(afi_t afi, safi_t safi, vrf_id_t vrf_id, int type,
unsigned short instance, int flags, struct prefix *p,
struct prefix_ipv6 *src_p, const struct nexthop *nh,
uint32_t nhe_id, uint32_t table_id, uint32_t metric, uint32_t mtu,
uint8_t distance, route_tag_t tag)
{
struct route_entry *re = NULL;
struct nexthop *nexthop = NULL;
/* Allocate new route_entry structure. */
re = XCALLOC(MTYPE_RE, sizeof(struct route_entry));
re->type = type;
re->instance = instance;
re->distance = distance;
re->flags = flags;
re->metric = metric;
re->mtu = mtu;
re->table = table_id;
re->vrf_id = vrf_id;
re->uptime = monotime(NULL);
re->tag = tag;
re->nhe_id = nhe_id;
if (!nhe_id) {
re->ng = nexthop_group_new();
/* Add nexthop. */
nexthop = nexthop_new();
*nexthop = *nh;
route_entry_nexthop_add(re, nexthop);
}
return rib_add_multipath(afi, safi, p, src_p, re);
}
static const char *rib_update_event2str(rib_update_event_t event)
{
const char *ret = "UNKNOWN";
switch (event) {
case RIB_UPDATE_KERNEL:
ret = "RIB_UPDATE_KERNEL";
break;
case RIB_UPDATE_RMAP_CHANGE:
ret = "RIB_UPDATE_RMAP_CHANGE";
break;
case RIB_UPDATE_OTHER:
ret = "RIB_UPDATE_OTHER";
break;
case RIB_UPDATE_MAX:
break;
}
return ret;
}
/* Schedule route nodes to be processed if they match the type */
static void rib_update_route_node(struct route_node *rn, int type)
{
struct route_entry *re, *next;
bool re_changed = false;
RNODE_FOREACH_RE_SAFE (rn, re, next) {
if (type == ZEBRA_ROUTE_ALL || type == re->type) {
SET_FLAG(re->status, ROUTE_ENTRY_CHANGED);
re_changed = true;
}
}
if (re_changed)
rib_queue_add(rn);
}
/* Schedule routes of a particular table (address-family) based on event. */
void rib_update_table(struct route_table *table, rib_update_event_t event)
{
struct route_node *rn;
if (IS_ZEBRA_DEBUG_EVENT) {
struct zebra_vrf *zvrf;
struct vrf *vrf;
zvrf = table->info ? ((rib_table_info_t *)table->info)->zvrf
: NULL;
vrf = zvrf ? zvrf->vrf : NULL;
zlog_debug("%s: %s VRF %s Table %u event %s", __func__,
table->info ? afi2str(
((rib_table_info_t *)table->info)->afi)
: "Unknown",
vrf ? vrf->name : "Unknown",
zvrf ? zvrf->table_id : 0,
rib_update_event2str(event));
}
/* Walk all routes and queue for processing, if appropriate for
* the trigger event.
*/
for (rn = route_top(table); rn; rn = srcdest_route_next(rn)) {
/*
* If we are looking at a route node and the node
* has already been queued we don't
* need to queue it up again
*/
if (rn->info
&& CHECK_FLAG(rib_dest_from_rnode(rn)->flags,
RIB_ROUTE_ANY_QUEUED))
continue;
switch (event) {
case RIB_UPDATE_KERNEL:
rib_update_route_node(rn, ZEBRA_ROUTE_KERNEL);
break;
case RIB_UPDATE_RMAP_CHANGE:
case RIB_UPDATE_OTHER:
rib_update_route_node(rn, ZEBRA_ROUTE_ALL);
break;
default:
break;
}
}
}
static void rib_update_handle_vrf(vrf_id_t vrf_id, rib_update_event_t event)
{
struct route_table *table;
if (IS_ZEBRA_DEBUG_EVENT)
zlog_debug("%s: Handling VRF %s event %s", __func__,
vrf_id_to_name(vrf_id), rib_update_event2str(event));
/* Process routes of interested address-families. */
table = zebra_vrf_table(AFI_IP, SAFI_UNICAST, vrf_id);
if (table)
rib_update_table(table, event);
table = zebra_vrf_table(AFI_IP6, SAFI_UNICAST, vrf_id);
if (table)
rib_update_table(table, event);
}
static void rib_update_handle_vrf_all(rib_update_event_t event)
{
struct zebra_router_table *zrt;
if (IS_ZEBRA_DEBUG_EVENT)
zlog_debug("%s: Handling VRF (ALL) event %s", __func__,
rib_update_event2str(event));
/* Just iterate over all the route tables, rather than vrf lookups */
RB_FOREACH (zrt, zebra_router_table_head, &zrouter.tables)
rib_update_table(zrt->table, event);
}
struct rib_update_ctx {
rib_update_event_t event;
bool vrf_all;
vrf_id_t vrf_id;
};
static struct rib_update_ctx *rib_update_ctx_init(vrf_id_t vrf_id,
rib_update_event_t event)
{
struct rib_update_ctx *ctx;
ctx = XCALLOC(MTYPE_RIB_UPDATE_CTX, sizeof(struct rib_update_ctx));
ctx->event = event;
ctx->vrf_id = vrf_id;
return ctx;
}
static void rib_update_ctx_fini(struct rib_update_ctx **ctx)
{
XFREE(MTYPE_RIB_UPDATE_CTX, *ctx);
*ctx = NULL;
}
static int rib_update_handler(struct thread *thread)
{
struct rib_update_ctx *ctx;
ctx = THREAD_ARG(thread);
if (ctx->vrf_all)
rib_update_handle_vrf_all(ctx->event);
else
rib_update_handle_vrf(ctx->vrf_id, ctx->event);
rib_update_ctx_fini(&ctx);
return 0;
}
/*
* Thread list to ensure we don't schedule a ton of events
* if interfaces are flapping for instance.
*/
static struct thread *t_rib_update_threads[RIB_UPDATE_MAX];
/* Schedule a RIB update event for specific vrf */
void rib_update_vrf(vrf_id_t vrf_id, rib_update_event_t event)
{
struct rib_update_ctx *ctx;
ctx = rib_update_ctx_init(vrf_id, event);
/* Don't worry about making sure multiple rib updates for specific vrf
* are scheduled at once for now. If it becomes a problem, we can use a
* lookup of some sort to keep track of running threads via t_vrf_id
* like how we are doing it in t_rib_update_threads[].
*/
thread_add_event(zrouter.master, rib_update_handler, ctx, 0, NULL);
if (IS_ZEBRA_DEBUG_EVENT)
zlog_debug("%s: Scheduled VRF %s, event %s", __func__,
vrf_id_to_name(ctx->vrf_id),
rib_update_event2str(event));
}
/* Schedule a RIB update event for all vrfs */
void rib_update(rib_update_event_t event)
{
struct rib_update_ctx *ctx;
ctx = rib_update_ctx_init(0, event);
ctx->vrf_all = true;
if (!thread_add_event(zrouter.master, rib_update_handler, ctx, 0,
&t_rib_update_threads[event]))
rib_update_ctx_fini(&ctx); /* Already scheduled */
else if (IS_ZEBRA_DEBUG_EVENT)
zlog_debug("%s: Schedued VRF (ALL), event %s", __func__,
rib_update_event2str(event));
}
/* Delete self installed routes after zebra is relaunched. */
void rib_sweep_table(struct route_table *table)
{
struct route_node *rn;
struct route_entry *re;
struct route_entry *next;
struct nexthop *nexthop;
if (!table)
return;
for (rn = route_top(table); rn; rn = srcdest_route_next(rn)) {
RNODE_FOREACH_RE_SAFE (rn, re, next) {
if (IS_ZEBRA_DEBUG_RIB)
route_entry_dump(&rn->p, NULL, re);
if (CHECK_FLAG(re->status, ROUTE_ENTRY_REMOVED))
continue;
if (!CHECK_FLAG(re->flags, ZEBRA_FLAG_SELFROUTE))
continue;
/*
* If routes are older than startup_time then
* we know we read them in from the kernel.
* As such we can safely remove them.
*/
if (zrouter.startup_time < re->uptime)
continue;
/*
* So we are starting up and have received
* routes from the kernel that we have installed
* from a previous run of zebra but not cleaned
* up ( say a kill -9 )
* But since we haven't actually installed
* them yet( we received them from the kernel )
* we don't think they are active.
* So let's pretend they are active to actually
* remove them.
* In all honesty I'm not sure if we should
* mark them as active when we receive them
* This is startup only so probably ok.
*
* If we ever decide to move rib_sweep_table
* to a different spot (ie startup )
* this decision needs to be revisited
*/
SET_FLAG(re->status, ROUTE_ENTRY_INSTALLED);
for (ALL_NEXTHOPS_PTR(re->ng, nexthop))
SET_FLAG(nexthop->flags, NEXTHOP_FLAG_FIB);
rib_uninstall_kernel(rn, re);
rib_delnode(rn, re);
}
}
}
/* Sweep all RIB tables. */
int rib_sweep_route(struct thread *t)
{
struct vrf *vrf;
struct zebra_vrf *zvrf;
RB_FOREACH (vrf, vrf_id_head, &vrfs_by_id) {
if ((zvrf = vrf->info) == NULL)
continue;
rib_sweep_table(zvrf->table[AFI_IP][SAFI_UNICAST]);
rib_sweep_table(zvrf->table[AFI_IP6][SAFI_UNICAST]);
}
zebra_router_sweep_route();
zebra_router_sweep_nhgs();
return 0;
}
/* Remove specific by protocol routes from 'table'. */
unsigned long rib_score_proto_table(uint8_t proto, unsigned short instance,
struct route_table *table)
{
struct route_node *rn;
struct route_entry *re;
struct route_entry *next;
unsigned long n = 0;
if (table)
for (rn = route_top(table); rn; rn = srcdest_route_next(rn))
RNODE_FOREACH_RE_SAFE (rn, re, next) {
if (CHECK_FLAG(re->status, ROUTE_ENTRY_REMOVED))
continue;
if (re->type == proto
&& re->instance == instance) {
rib_delnode(rn, re);
n++;
}
}
return n;
}
/* Remove specific by protocol routes. */
unsigned long rib_score_proto(uint8_t proto, unsigned short instance)
{
struct vrf *vrf;
struct zebra_vrf *zvrf;
struct other_route_table *ort;
unsigned long cnt = 0;
RB_FOREACH (vrf, vrf_id_head, &vrfs_by_id) {
zvrf = vrf->info;
if (!zvrf)
continue;
cnt += rib_score_proto_table(proto, instance,
zvrf->table[AFI_IP][SAFI_UNICAST])
+ rib_score_proto_table(
proto, instance,
zvrf->table[AFI_IP6][SAFI_UNICAST]);
frr_each(otable, &zvrf->other_tables, ort) cnt +=
rib_score_proto_table(proto, instance, ort->table);
}
return cnt;
}
/* Close RIB and clean up kernel routes. */
void rib_close_table(struct route_table *table)
{
struct route_node *rn;
rib_table_info_t *info;
rib_dest_t *dest;
if (!table)
return;
info = route_table_get_info(table);
for (rn = route_top(table); rn; rn = srcdest_route_next(rn)) {
dest = rib_dest_from_rnode(rn);
if (dest && dest->selected_fib) {
if (info->safi == SAFI_UNICAST)
hook_call(rib_update, rn, NULL);
rib_uninstall_kernel(rn, dest->selected_fib);
dest->selected_fib = NULL;
}
}
}
/*
* Handler for async dataplane results after a pseudowire installation
*/
static int handle_pw_result(struct zebra_dplane_ctx *ctx)
{
struct zebra_pw *pw;
struct zebra_vrf *vrf;
/* The pseudowire code assumes success - we act on an error
* result for installation attempts here.
*/
if (dplane_ctx_get_op(ctx) != DPLANE_OP_PW_INSTALL)
goto done;
if (dplane_ctx_get_status(ctx) != ZEBRA_DPLANE_REQUEST_SUCCESS) {
vrf = zebra_vrf_lookup_by_id(dplane_ctx_get_vrf(ctx));
pw = zebra_pw_find(vrf, dplane_ctx_get_ifname(ctx));
if (pw)
zebra_pw_install_failure(pw);
}
done:
return 0;
}
/*
* Handle results from the dataplane system. Dequeue update context
* structs, dispatch to appropriate internal handlers.
*/
static int rib_process_dplane_results(struct thread *thread)
{
struct zebra_dplane_ctx *ctx;
struct dplane_ctx_q ctxlist;
bool shut_p = false;
/* Dequeue a list of completed updates with one lock/unlock cycle */
do {
TAILQ_INIT(&ctxlist);
/* Take lock controlling queue of results */
frr_with_mutex(&dplane_mutex) {
/* Dequeue list of context structs */
dplane_ctx_list_append(&ctxlist, &rib_dplane_q);
}
/* Dequeue context block */
ctx = dplane_ctx_dequeue(&ctxlist);
/* If we've emptied the results queue, we're done */
if (ctx == NULL)
break;
/* If zebra is shutting down, avoid processing results,
* just drain the results queue.
*/
shut_p = atomic_load_explicit(&zrouter.in_shutdown,
memory_order_relaxed);
if (shut_p) {
while (ctx) {
dplane_ctx_fini(&ctx);
ctx = dplane_ctx_dequeue(&ctxlist);
}
continue;
}
while (ctx) {
switch (dplane_ctx_get_op(ctx)) {
case DPLANE_OP_ROUTE_INSTALL:
case DPLANE_OP_ROUTE_UPDATE:
case DPLANE_OP_ROUTE_DELETE:
{
/* Bit of special case for route updates
* that were generated by async notifications:
* we don't want to continue processing these
* in the rib.
*/
if (dplane_ctx_get_notif_provider(ctx) == 0)
rib_process_result(ctx);
else
dplane_ctx_fini(&ctx);
}
break;
case DPLANE_OP_ROUTE_NOTIFY:
rib_process_dplane_notify(ctx);
break;
case DPLANE_OP_NH_INSTALL:
case DPLANE_OP_NH_UPDATE:
case DPLANE_OP_NH_DELETE:
zebra_nhg_dplane_result(ctx);
break;
case DPLANE_OP_LSP_INSTALL:
case DPLANE_OP_LSP_UPDATE:
case DPLANE_OP_LSP_DELETE:
{
/* Bit of special case for LSP updates
* that were generated by async notifications:
* we don't want to continue processing these.
*/
if (dplane_ctx_get_notif_provider(ctx) == 0)
zebra_mpls_lsp_dplane_result(ctx);
else
dplane_ctx_fini(&ctx);
}
break;
case DPLANE_OP_LSP_NOTIFY:
zebra_mpls_process_dplane_notify(ctx);
break;
case DPLANE_OP_PW_INSTALL:
case DPLANE_OP_PW_UNINSTALL:
handle_pw_result(ctx);
break;
case DPLANE_OP_SYS_ROUTE_ADD:
case DPLANE_OP_SYS_ROUTE_DELETE:
/* No further processing in zebra for these. */
dplane_ctx_fini(&ctx);
break;
case DPLANE_OP_MAC_INSTALL:
case DPLANE_OP_MAC_DELETE:
zebra_vxlan_handle_result(ctx);
break;
/* Some op codes not handled here */
case DPLANE_OP_ADDR_INSTALL:
case DPLANE_OP_ADDR_UNINSTALL:
case DPLANE_OP_NEIGH_INSTALL:
case DPLANE_OP_NEIGH_UPDATE:
case DPLANE_OP_NEIGH_DELETE:
case DPLANE_OP_VTEP_ADD:
case DPLANE_OP_VTEP_DELETE:
case DPLANE_OP_NONE:
/* Don't expect this: just return the struct? */
dplane_ctx_fini(&ctx);
break;
} /* Dispatch by op code */
ctx = dplane_ctx_dequeue(&ctxlist);
}
} while (1);
return 0;
}
/*
* Results are returned from the dataplane subsystem, in the context of
* the dataplane pthread. We enqueue the results here for processing by
* the main thread later.
*/
static int rib_dplane_results(struct dplane_ctx_q *ctxlist)
{
/* Take lock controlling queue of results */
frr_with_mutex(&dplane_mutex) {
/* Enqueue context blocks */
dplane_ctx_list_append(&rib_dplane_q, ctxlist);
}
/* Ensure event is signalled to zebra main pthread */
thread_add_event(zrouter.master, rib_process_dplane_results, NULL, 0,
&t_dplane);
return 0;
}
/*
* Ensure there are no empty slots in the route_info array.
* Every route type in zebra should be present there.
*/
static void check_route_info(void)
{
int len = array_size(route_info);
/*
* ZEBRA_ROUTE_SYSTEM is special cased since
* its key is 0 anyway.
*
* ZEBRA_ROUTE_ALL is also ignored.
*/
for (int i = 0; i < len; i++) {
if (i == ZEBRA_ROUTE_SYSTEM || i == ZEBRA_ROUTE_ALL)
continue;
assert(route_info[i].key);
assert(route_info[i].meta_q_map < MQ_SIZE);
}
}
/* Routing information base initialize. */
void rib_init(void)
{
check_route_info();
rib_queue_init();
/* Init dataplane, and register for results */
pthread_mutex_init(&dplane_mutex, NULL);
TAILQ_INIT(&rib_dplane_q);
zebra_dplane_init(rib_dplane_results);
}
/*
* vrf_id_get_next
*
* Get the first vrf id that is greater than the given vrf id if any.
*
* Returns true if a vrf id was found, false otherwise.
*/
static inline int vrf_id_get_next(vrf_id_t vrf_id, vrf_id_t *next_id_p)
{
struct vrf *vrf;
vrf = vrf_lookup_by_id(vrf_id);
if (vrf) {
vrf = RB_NEXT(vrf_id_head, vrf);
if (vrf) {
*next_id_p = vrf->vrf_id;
return 1;
}
}
return 0;
}
/*
* rib_tables_iter_next
*
* Returns the next table in the iteration.
*/
struct route_table *rib_tables_iter_next(rib_tables_iter_t *iter)
{
struct route_table *table;
/*
* Array that helps us go over all AFI/SAFI combinations via one
* index.
*/
static struct {
afi_t afi;
safi_t safi;
} afi_safis[] = {
{AFI_IP, SAFI_UNICAST}, {AFI_IP, SAFI_MULTICAST},
{AFI_IP, SAFI_LABELED_UNICAST}, {AFI_IP6, SAFI_UNICAST},
{AFI_IP6, SAFI_MULTICAST}, {AFI_IP6, SAFI_LABELED_UNICAST},
};
table = NULL;
switch (iter->state) {
case RIB_TABLES_ITER_S_INIT:
iter->vrf_id = VRF_DEFAULT;
iter->afi_safi_ix = -1;
/* Fall through */
case RIB_TABLES_ITER_S_ITERATING:
iter->afi_safi_ix++;
while (1) {
while (iter->afi_safi_ix
< (int)array_size(afi_safis)) {
table = zebra_vrf_table(
afi_safis[iter->afi_safi_ix].afi,
afi_safis[iter->afi_safi_ix].safi,
iter->vrf_id);
if (table)
break;
iter->afi_safi_ix++;
}
/*
* Found another table in this vrf.
*/
if (table)
break;
/*
* Done with all tables in the current vrf, go to the
* next
* one.
*/
if (!vrf_id_get_next(iter->vrf_id, &iter->vrf_id))
break;
iter->afi_safi_ix = 0;
}
break;
case RIB_TABLES_ITER_S_DONE:
return NULL;
}
if (table)
iter->state = RIB_TABLES_ITER_S_ITERATING;
else
iter->state = RIB_TABLES_ITER_S_DONE;
return table;
}
Reference in New Issue View Git Blame Copy Permalink