* [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct
@ 2019-03-27 3:29 David Ahern
2019-03-27 3:29 ` [PATCH net-next 01/13] ipv4: Define fib_get_nhs when CONFIG_IP_ROUTE_MULTIPATH is disabled David Ahern
` (12 more replies)
0 siblings, 13 replies; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
First set of three with the end goal of enabling IPv6 gateways with IPv4
routes.
This set refactors ipv4 and ipv6 code to create init and release
helpers for each protocol and moving common elements to a fib_nh_common
struct.
David Ahern (13):
ipv4: Define fib_get_nhs when CONFIG_IP_ROUTE_MULTIPATH is disabled
ipv4: Move IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN to helper
ipv4: Create init helper for fib_nh
ipv4: Create cleanup helper for fib_nh
ipv6: Create init helper for fib6_nh
ipv6: Create cleanup helper for fib6_nh
ipv6: Move gateway checks to a fib6_nh setting
ipv6: Refactor fib6_ignore_linkdown
ipv6: Change rt6_add_nexthop and rt6_nexthop_info to take fib6_nh
ipv4: Rename fib_nh entries
ipv6: Rename fib6_nh entries
net: Add fib_nh_common and update fib_nh and fib6_nh
net: Use common nexthop init and release helpers
drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c | 12 +-
.../net/ethernet/mellanox/mlxsw/spectrum_router.c | 56 +--
drivers/net/ethernet/rocker/rocker_ofdpa.c | 10 +-
include/linux/inetdevice.h | 14 +
include/net/addrconf.h | 8 +
include/net/ip6_fib.h | 18 +-
include/net/ip6_route.h | 12 +-
include/net/ip_fib.h | 58 ++-
include/trace/events/fib.h | 7 +-
include/trace/events/fib6.h | 6 +-
net/core/filter.c | 16 +-
net/ipv4/fib_frontend.c | 10 +-
net/ipv4/fib_semantics.c | 470 +++++++++++----------
net/ipv4/fib_trie.c | 14 +-
net/ipv4/route.c | 18 +-
net/ipv6/addrconf.c | 27 +-
net/ipv6/ip6_fib.c | 18 +-
net/ipv6/ndisc.c | 8 +-
net/ipv6/route.c | 440 +++++++++----------
19 files changed, 654 insertions(+), 568 deletions(-)
--
2.11.0
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH net-next 01/13] ipv4: Define fib_get_nhs when CONFIG_IP_ROUTE_MULTIPATH is disabled
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 7:40 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 02/13] ipv4: Move IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN to helper David Ahern
` (11 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
Define fib_get_nhs to return EINVAL when CONFIG_IP_ROUTE_MULTIPATH is
not enabled and remove the ifdef check for CONFIG_IP_ROUTE_MULTIPATH
in fib_create_info.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
net/ipv4/fib_semantics.c | 15 +++++++++------
1 file changed, 9 insertions(+), 6 deletions(-)
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 8e185b5a2bf6..b5dbbdfd1e49 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -601,6 +601,15 @@ static void fib_rebalance(struct fib_info *fi)
}
#else /* CONFIG_IP_ROUTE_MULTIPATH */
+static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh,
+ int remaining, struct fib_config *cfg,
+ struct netlink_ext_ack *extack)
+{
+ NL_SET_ERR_MSG(extack, "Multipath support not enabled in kernel");
+
+ return -EINVAL;
+}
+
#define fib_rebalance(fi) do { } while (0)
#endif /* CONFIG_IP_ROUTE_MULTIPATH */
@@ -1102,7 +1111,6 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
} endfor_nexthops(fi)
if (cfg->fc_mp) {
-#ifdef CONFIG_IP_ROUTE_MULTIPATH
err = fib_get_nhs(fi, cfg->fc_mp, cfg->fc_mp_len, cfg, extack);
if (err != 0)
goto failure;
@@ -1123,11 +1131,6 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
goto err_inval;
}
#endif
-#else
- NL_SET_ERR_MSG(extack,
- "Multipath support not enabled in kernel");
- goto err_inval;
-#endif
} else {
struct fib_nh *nh = fi->fib_nh;
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 02/13] ipv4: Move IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN to helper
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
2019-03-27 3:29 ` [PATCH net-next 01/13] ipv4: Define fib_get_nhs when CONFIG_IP_ROUTE_MULTIPATH is disabled David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 7:44 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 03/13] ipv4: Create init helper for fib_nh David Ahern
` (10 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
in_dev lookup followed by IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN check
is called in several places, some with the rcu lock and others with the
rtnl held.
Move the check to a helper similar to what IPv6 has. Since the helper
can be invoked from either context use rcu_dereference_rtnl to
dereference ip_ptr.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
include/linux/inetdevice.h | 14 ++++++++++++++
net/ipv4/fib_semantics.c | 31 +++++++------------------------
net/ipv4/fib_trie.c | 4 +---
3 files changed, 22 insertions(+), 27 deletions(-)
diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
index a64f21a97369..35e0390fd7b8 100644
--- a/include/linux/inetdevice.h
+++ b/include/linux/inetdevice.h
@@ -237,6 +237,20 @@ static inline struct in_device *__in_dev_get_rtnl(const struct net_device *dev)
return rtnl_dereference(dev->ip_ptr);
}
+/* called with rc_read_lock or rtnl held */
+static inline bool ip_ignore_linkdown(const struct net_device *dev)
+{
+ struct in_device *in_dev;
+ bool rc = false;
+
+ in_dev = rcu_dereference_rtnl(dev->ip_ptr);
+ if (in_dev &&
+ IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev))
+ rc = true;
+
+ return rc;
+}
+
static inline struct neigh_parms *__in_dev_arp_parms_get_rcu(const struct net_device *dev)
{
struct in_device *in_dev = __in_dev_get_rcu(dev);
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index b5dbbdfd1e49..78631eb255f7 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -558,7 +558,6 @@ static void fib_rebalance(struct fib_info *fi)
{
int total;
int w;
- struct in_device *in_dev;
if (fi->fib_nhs < 2)
return;
@@ -568,10 +567,7 @@ static void fib_rebalance(struct fib_info *fi)
if (nh->nh_flags & RTNH_F_DEAD)
continue;
- in_dev = __in_dev_get_rtnl(nh->nh_dev);
-
- if (in_dev &&
- IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) &&
+ if (ip_ignore_linkdown(nh->nh_dev) &&
nh->nh_flags & RTNH_F_LINKDOWN)
continue;
@@ -582,12 +578,9 @@ static void fib_rebalance(struct fib_info *fi)
change_nexthops(fi) {
int upper_bound;
- in_dev = __in_dev_get_rtnl(nexthop_nh->nh_dev);
-
if (nexthop_nh->nh_flags & RTNH_F_DEAD) {
upper_bound = -1;
- } else if (in_dev &&
- IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) &&
+ } else if (ip_ignore_linkdown(nexthop_nh->nh_dev) &&
nexthop_nh->nh_flags & RTNH_F_LINKDOWN) {
upper_bound = -1;
} else {
@@ -1325,12 +1318,8 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event,
nla_put_u32(skb, RTA_OIF, fi->fib_nh->nh_oif))
goto nla_put_failure;
if (fi->fib_nh->nh_flags & RTNH_F_LINKDOWN) {
- struct in_device *in_dev;
-
rcu_read_lock();
- in_dev = __in_dev_get_rcu(fi->fib_nh->nh_dev);
- if (in_dev &&
- IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev))
+ if (ip_ignore_linkdown(fi->fib_nh->nh_dev))
rtm->rtm_flags |= RTNH_F_DEAD;
rcu_read_unlock();
}
@@ -1361,12 +1350,8 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event,
rtnh->rtnh_flags = nh->nh_flags & 0xFF;
if (nh->nh_flags & RTNH_F_LINKDOWN) {
- struct in_device *in_dev;
-
rcu_read_lock();
- in_dev = __in_dev_get_rcu(nh->nh_dev);
- if (in_dev &&
- IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev))
+ if (ip_ignore_linkdown(nh->nh_dev))
rtnh->rtnh_flags |= RTNH_F_DEAD;
rcu_read_unlock();
}
@@ -1433,7 +1418,7 @@ int fib_sync_down_addr(struct net_device *dev, __be32 local)
static int call_fib_nh_notifiers(struct fib_nh *fib_nh,
enum fib_event_type event_type)
{
- struct in_device *in_dev = __in_dev_get_rtnl(fib_nh->nh_dev);
+ bool ignore_link_down = ip_ignore_linkdown(fib_nh->nh_dev);
struct fib_nh_notifier_info info = {
.fib_nh = fib_nh,
};
@@ -1442,14 +1427,12 @@ static int call_fib_nh_notifiers(struct fib_nh *fib_nh,
case FIB_EVENT_NH_ADD:
if (fib_nh->nh_flags & RTNH_F_DEAD)
break;
- if (IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) &&
- fib_nh->nh_flags & RTNH_F_LINKDOWN)
+ if (ignore_link_down && fib_nh->nh_flags & RTNH_F_LINKDOWN)
break;
return call_fib4_notifiers(dev_net(fib_nh->nh_dev), event_type,
&info.info);
case FIB_EVENT_NH_DEL:
- if ((in_dev && IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) &&
- fib_nh->nh_flags & RTNH_F_LINKDOWN) ||
+ if ((ignore_link_down && fib_nh->nh_flags & RTNH_F_LINKDOWN) ||
(fib_nh->nh_flags & RTNH_F_DEAD))
return call_fib4_notifiers(dev_net(fib_nh->nh_dev),
event_type, &info.info);
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 1704f432de1f..656d3d19f112 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1471,12 +1471,10 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi4 *flp,
continue;
for (nhsel = 0; nhsel < fi->fib_nhs; nhsel++) {
const struct fib_nh *nh = &fi->fib_nh[nhsel];
- struct in_device *in_dev = __in_dev_get_rcu(nh->nh_dev);
if (nh->nh_flags & RTNH_F_DEAD)
continue;
- if (in_dev &&
- IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) &&
+ if (ip_ignore_linkdown(nh->nh_dev) &&
nh->nh_flags & RTNH_F_LINKDOWN &&
!(fib_flags & FIB_LOOKUP_IGNORE_LINKSTATE))
continue;
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 03/13] ipv4: Create init helper for fib_nh
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
2019-03-27 3:29 ` [PATCH net-next 01/13] ipv4: Define fib_get_nhs when CONFIG_IP_ROUTE_MULTIPATH is disabled David Ahern
2019-03-27 3:29 ` [PATCH net-next 02/13] ipv4: Move IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN to helper David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 8:12 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 04/13] ipv4: Create cleanup " David Ahern
` (9 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
Consolidate the fib_nh initialization which is duplicated between
fib_create_info for single path and fib_get_nhs for multipath.
Export the helper to allow for use with nexthop objects in the
future.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
include/net/ip_fib.h | 4 ++
net/ipv4/fib_semantics.c | 177 +++++++++++++++++++++++------------------------
2 files changed, 92 insertions(+), 89 deletions(-)
diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 9c8214d2116d..1af1f552644a 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -416,6 +416,10 @@ void fib_select_multipath(struct fib_result *res, int hash);
void fib_select_path(struct net *net, struct fib_result *res,
struct flowi4 *fl4, const struct sk_buff *skb);
+int fib_nh_init(struct net *net, struct fib_nh *fib_nh,
+ struct fib_config *cfg, int nh_weight,
+ struct netlink_ext_ack *extack);
+
/* Exported by fib_trie.c */
void fib_trie_init(void);
struct fib_table *fib_trie_table(u32 id, struct fib_table *alias);
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 78631eb255f7..773e451549df 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -457,6 +457,51 @@ static int fib_detect_death(struct fib_info *fi, int order,
return 1;
}
+int fib_nh_init(struct net *net, struct fib_nh *nh,
+ struct fib_config *cfg, int nh_weight,
+ struct netlink_ext_ack *extack)
+{
+ int err = -ENOMEM;
+
+ nh->nh_pcpu_rth_output = alloc_percpu(struct rtable __rcu *);
+ if (!nh->nh_pcpu_rth_output)
+ goto failure;
+
+ if (cfg->fc_encap) {
+ struct lwtunnel_state *lwtstate;
+
+ err = -EINVAL;
+ if (cfg->fc_encap_type == LWTUNNEL_ENCAP_NONE) {
+ NL_SET_ERR_MSG(extack, "LWT encap type not specified");
+ goto failure;
+ }
+ err = lwtunnel_build_state(cfg->fc_encap_type,
+ cfg->fc_encap, AF_INET, cfg,
+ &lwtstate, extack);
+ if (err)
+ goto failure;
+
+ nh->nh_lwtstate = lwtstate_get(lwtstate);
+ }
+
+ nh->nh_oif = cfg->fc_oif;
+ nh->nh_gw = cfg->fc_gw;
+ nh->nh_flags = cfg->fc_flags;
+
+#ifdef CONFIG_IP_ROUTE_CLASSID
+ nh->nh_tclassid = cfg->fc_flow;
+ if (nh->nh_tclassid)
+ net->ipv4.fib_num_tclassid_users++;
+#endif
+#ifdef CONFIG_IP_ROUTE_MULTIPATH
+ nh->nh_weight = nh_weight;
+#endif
+ err = 0;
+
+failure:
+ return err;
+}
+
#ifdef CONFIG_IP_ROUTE_MULTIPATH
static int fib_count_nexthops(struct rtnexthop *rtnh, int remaining,
@@ -483,11 +528,15 @@ static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh,
int remaining, struct fib_config *cfg,
struct netlink_ext_ack *extack)
{
+ struct net *net = fi->fib_net;
+ struct fib_config fib_cfg;
int ret;
change_nexthops(fi) {
int attrlen;
+ memset(&fib_cfg, 0, sizeof(fib_cfg));
+
if (!rtnh_ok(rtnh, remaining)) {
NL_SET_ERR_MSG(extack,
"Invalid nexthop configuration - extra data after nexthop");
@@ -500,56 +549,54 @@ static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh,
return -EINVAL;
}
- nexthop_nh->nh_flags =
- (cfg->fc_flags & ~0xFF) | rtnh->rtnh_flags;
- nexthop_nh->nh_oif = rtnh->rtnh_ifindex;
- nexthop_nh->nh_weight = rtnh->rtnh_hops + 1;
+ fib_cfg.fc_flags = (cfg->fc_flags & ~0xFF) | rtnh->rtnh_flags;
+ fib_cfg.fc_oif = rtnh->rtnh_ifindex;
attrlen = rtnh_attrlen(rtnh);
if (attrlen > 0) {
struct nlattr *nla, *attrs = rtnh_attrs(rtnh);
nla = nla_find(attrs, attrlen, RTA_GATEWAY);
- nexthop_nh->nh_gw = nla ? nla_get_in_addr(nla) : 0;
-#ifdef CONFIG_IP_ROUTE_CLASSID
+ if (nla)
+ fib_cfg.fc_gw = nla_get_in_addr(nla);
+
nla = nla_find(attrs, attrlen, RTA_FLOW);
- nexthop_nh->nh_tclassid = nla ? nla_get_u32(nla) : 0;
- if (nexthop_nh->nh_tclassid)
- fi->fib_net->ipv4.fib_num_tclassid_users++;
-#endif
- nla = nla_find(attrs, attrlen, RTA_ENCAP);
- if (nla) {
- struct lwtunnel_state *lwtstate;
- struct nlattr *nla_entype;
-
- nla_entype = nla_find(attrs, attrlen,
- RTA_ENCAP_TYPE);
- if (!nla_entype) {
- NL_SET_BAD_ATTR(extack, nla);
- NL_SET_ERR_MSG(extack,
- "Encap type is missing");
- goto err_inval;
- }
+ if (nla)
+ fib_cfg.fc_flow = nla_get_u32(nla);
- ret = lwtunnel_build_state(nla_get_u16(
- nla_entype),
- nla, AF_INET, cfg,
- &lwtstate, extack);
- if (ret)
- goto errout;
- nexthop_nh->nh_lwtstate =
- lwtstate_get(lwtstate);
- }
+ fib_cfg.fc_encap = nla_find(attrs, attrlen, RTA_ENCAP);
+ nla = nla_find(attrs, attrlen, RTA_ENCAP_TYPE);
+ if (nla)
+ fib_cfg.fc_encap_type = nla_get_u16(nla);
}
+ ret = fib_nh_init(net, nexthop_nh, &fib_cfg,
+ rtnh->rtnh_hops + 1, extack);
+ if (ret)
+ goto errout;
+
rtnh = rtnh_next(rtnh, &remaining);
} endfor_nexthops(fi);
- return 0;
-
-err_inval:
ret = -EINVAL;
-
+ if (cfg->fc_oif && fi->fib_nh->nh_oif != cfg->fc_oif) {
+ NL_SET_ERR_MSG(extack,
+ "Nexthop device index does not match RTA_OIF");
+ goto errout;
+ }
+ if (cfg->fc_gw && fi->fib_nh->nh_gw != cfg->fc_gw) {
+ NL_SET_ERR_MSG(extack,
+ "Nexthop gateway does not match RTA_GATEWAY");
+ goto errout;
+ }
+#ifdef CONFIG_IP_ROUTE_CLASSID
+ if (cfg->fc_flow && fi->fib_nh->nh_tclassid != cfg->fc_flow) {
+ NL_SET_ERR_MSG(extack,
+ "Nexthop class id does not match RTA_FLOW");
+ goto errout;
+ }
+#endif
+ ret = 0;
errout:
return ret;
}
@@ -1098,63 +1145,15 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
fi->fib_nhs = nhs;
change_nexthops(fi) {
nexthop_nh->nh_parent = fi;
- nexthop_nh->nh_pcpu_rth_output = alloc_percpu(struct rtable __rcu *);
- if (!nexthop_nh->nh_pcpu_rth_output)
- goto failure;
} endfor_nexthops(fi)
- if (cfg->fc_mp) {
+ if (cfg->fc_mp)
err = fib_get_nhs(fi, cfg->fc_mp, cfg->fc_mp_len, cfg, extack);
- if (err != 0)
- goto failure;
- if (cfg->fc_oif && fi->fib_nh->nh_oif != cfg->fc_oif) {
- NL_SET_ERR_MSG(extack,
- "Nexthop device index does not match RTA_OIF");
- goto err_inval;
- }
- if (cfg->fc_gw && fi->fib_nh->nh_gw != cfg->fc_gw) {
- NL_SET_ERR_MSG(extack,
- "Nexthop gateway does not match RTA_GATEWAY");
- goto err_inval;
- }
-#ifdef CONFIG_IP_ROUTE_CLASSID
- if (cfg->fc_flow && fi->fib_nh->nh_tclassid != cfg->fc_flow) {
- NL_SET_ERR_MSG(extack,
- "Nexthop class id does not match RTA_FLOW");
- goto err_inval;
- }
-#endif
- } else {
- struct fib_nh *nh = fi->fib_nh;
-
- if (cfg->fc_encap) {
- struct lwtunnel_state *lwtstate;
-
- if (cfg->fc_encap_type == LWTUNNEL_ENCAP_NONE) {
- NL_SET_ERR_MSG(extack,
- "LWT encap type not specified");
- goto err_inval;
- }
- err = lwtunnel_build_state(cfg->fc_encap_type,
- cfg->fc_encap, AF_INET, cfg,
- &lwtstate, extack);
- if (err)
- goto failure;
+ else
+ err = fib_nh_init(net, fi->fib_nh, cfg, 1, extack);
- nh->nh_lwtstate = lwtstate_get(lwtstate);
- }
- nh->nh_oif = cfg->fc_oif;
- nh->nh_gw = cfg->fc_gw;
- nh->nh_flags = cfg->fc_flags;
-#ifdef CONFIG_IP_ROUTE_CLASSID
- nh->nh_tclassid = cfg->fc_flow;
- if (nh->nh_tclassid)
- fi->fib_net->ipv4.fib_num_tclassid_users++;
-#endif
-#ifdef CONFIG_IP_ROUTE_MULTIPATH
- nh->nh_weight = 1;
-#endif
- }
+ if (err != 0)
+ goto failure;
if (fib_props[cfg->fc_type].error) {
if (cfg->fc_gw || cfg->fc_oif || cfg->fc_mp) {
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 04/13] ipv4: Create cleanup helper for fib_nh
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
` (2 preceding siblings ...)
2019-03-27 3:29 ` [PATCH net-next 03/13] ipv4: Create init helper for fib_nh David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 8:17 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 05/13] ipv6: Create init helper for fib6_nh David Ahern
` (8 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
Move the fib_nh cleanup code from free_fib_info_rcu into a new helper,
fib_nh_release. Move classid accounting into fib_nh_release which is
called per fib_nh to make accounting symmetrical with fib_nh_init.
Export the helper to allow for use with nexthop objects in the
future.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
include/net/ip_fib.h | 1 +
net/ipv4/fib_semantics.c | 29 +++++++++++++++++------------
2 files changed, 18 insertions(+), 12 deletions(-)
diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 1af1f552644a..5a4df0ba175e 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -419,6 +419,7 @@ void fib_select_path(struct net *net, struct fib_result *res,
int fib_nh_init(struct net *net, struct fib_nh *fib_nh,
struct fib_config *cfg, int nh_weight,
struct netlink_ext_ack *extack);
+void fib_nh_release(struct net *net, struct fib_nh *fib_nh);
/* Exported by fib_trie.c */
void fib_trie_init(void);
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 773e451549df..72cf51e1eaa9 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -204,18 +204,28 @@ static void rt_fibinfo_free_cpus(struct rtable __rcu * __percpu *rtp)
free_percpu(rtp);
}
+void fib_nh_release(struct net *net, struct fib_nh *fib_nh)
+{
+#ifdef CONFIG_IP_ROUTE_CLASSID
+ if (fib_nh->nh_tclassid)
+ net->ipv4.fib_num_tclassid_users--;
+#endif
+ if (fib_nh->nh_dev)
+ dev_put(fib_nh->nh_dev);
+
+ lwtstate_put(fib_nh->nh_lwtstate);
+ free_nh_exceptions(fib_nh);
+ rt_fibinfo_free_cpus(fib_nh->nh_pcpu_rth_output);
+ rt_fibinfo_free(&fib_nh->nh_rth_input);
+}
+
/* Release a nexthop info record */
static void free_fib_info_rcu(struct rcu_head *head)
{
struct fib_info *fi = container_of(head, struct fib_info, rcu);
change_nexthops(fi) {
- if (nexthop_nh->nh_dev)
- dev_put(nexthop_nh->nh_dev);
- lwtstate_put(nexthop_nh->nh_lwtstate);
- free_nh_exceptions(nexthop_nh);
- rt_fibinfo_free_cpus(nexthop_nh->nh_pcpu_rth_output);
- rt_fibinfo_free(&nexthop_nh->nh_rth_input);
+ fib_nh_release(fi->fib_net, nexthop_nh);
} endfor_nexthops(fi);
ip_fib_metrics_put(fi->fib_metrics);
@@ -230,12 +240,7 @@ void free_fib_info(struct fib_info *fi)
return;
}
fib_info_cnt--;
-#ifdef CONFIG_IP_ROUTE_CLASSID
- change_nexthops(fi) {
- if (nexthop_nh->nh_tclassid)
- fi->fib_net->ipv4.fib_num_tclassid_users--;
- } endfor_nexthops(fi);
-#endif
+
call_rcu(&fi->rcu, free_fib_info_rcu);
}
EXPORT_SYMBOL_GPL(free_fib_info);
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 05/13] ipv6: Create init helper for fib6_nh
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
` (3 preceding siblings ...)
2019-03-27 3:29 ` [PATCH net-next 04/13] ipv4: Create cleanup " David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 8:34 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 06/13] ipv6: Create cleanup " David Ahern
` (7 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
Similar to IPv4, consolidate the fib6_nh initialization into a helper.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
include/net/ip6_fib.h | 4 +
net/ipv6/route.c | 231 +++++++++++++++++++++++++++-----------------------
2 files changed, 127 insertions(+), 108 deletions(-)
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 2acb78a762ee..ce1f81345c8e 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -444,6 +444,10 @@ static inline struct net_device *fib6_info_nh_dev(const struct fib6_info *f6i)
return f6i->fib6_nh.nh_dev;
}
+int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
+ struct fib6_config *cfg, gfp_t gfp_flags,
+ struct netlink_ext_ack *extack);
+
static inline
struct lwtunnel_state *fib6_info_nh_lwt(const struct fib6_info *f6i)
{
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index b804be3cbf05..269ad7b9cdbb 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2896,16 +2896,129 @@ static int ip6_validate_gw(struct net *net, struct fib6_config *cfg,
return err;
}
+int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
+ struct fib6_config *cfg, gfp_t gfp_flags,
+ struct netlink_ext_ack *extack)
+{
+ struct net_device *dev = NULL;
+ struct inet6_dev *idev = NULL;
+ int addr_type;
+ int err;
+
+ err = -ENODEV;
+ if (cfg->fc_ifindex) {
+ dev = dev_get_by_index(net, cfg->fc_ifindex);
+ if (!dev)
+ goto out;
+ idev = in6_dev_get(dev);
+ if (!idev)
+ goto out;
+ }
+
+ if (cfg->fc_flags & RTNH_F_ONLINK) {
+ if (!dev) {
+ NL_SET_ERR_MSG(extack,
+ "Nexthop device required for onlink");
+ goto out;
+ }
+
+ if (!(dev->flags & IFF_UP)) {
+ NL_SET_ERR_MSG(extack, "Nexthop device is not up");
+ err = -ENETDOWN;
+ goto out;
+ }
+
+ fib6_nh->nh_flags |= RTNH_F_ONLINK;
+ }
+
+ if (cfg->fc_encap) {
+ struct lwtunnel_state *lwtstate;
+
+ err = lwtunnel_build_state(cfg->fc_encap_type,
+ cfg->fc_encap, AF_INET6, cfg,
+ &lwtstate, extack);
+ if (err)
+ goto out;
+
+ fib6_nh->nh_lwtstate = lwtstate_get(lwtstate);
+ }
+
+ fib6_nh->nh_weight = 1;
+
+ /* We cannot add true routes via loopback here,
+ * they would result in kernel looping; promote them to reject routes
+ */
+ addr_type = ipv6_addr_type(&cfg->fc_dst);
+ if ((cfg->fc_flags & RTF_REJECT) ||
+ (dev && (dev->flags & IFF_LOOPBACK) &&
+ !(addr_type & IPV6_ADDR_LOOPBACK) &&
+ !(cfg->fc_flags & RTF_LOCAL))) {
+ /* hold loopback dev/idev if we haven't done so. */
+ if (dev != net->loopback_dev) {
+ if (dev) {
+ dev_put(dev);
+ in6_dev_put(idev);
+ }
+ dev = net->loopback_dev;
+ dev_hold(dev);
+ idev = in6_dev_get(dev);
+ if (!idev) {
+ err = -ENODEV;
+ goto out;
+ }
+ }
+ cfg->fc_flags = RTF_REJECT | RTF_NONEXTHOP;
+ goto set_dev;
+ }
+
+ if (cfg->fc_flags & RTF_GATEWAY) {
+ err = ip6_validate_gw(net, cfg, &dev, &idev, extack);
+ if (err)
+ goto out;
+
+ fib6_nh->nh_gw = cfg->fc_gateway;
+ }
+
+ err = -ENODEV;
+ if (!dev)
+ goto out;
+
+ if (idev->cnf.disable_ipv6) {
+ NL_SET_ERR_MSG(extack, "IPv6 is disabled on nexthop device");
+ err = -EACCES;
+ goto out;
+ }
+
+ if (!(dev->flags & IFF_UP) && !cfg->fc_ignore_dev_down) {
+ NL_SET_ERR_MSG(extack, "Nexthop device is not up");
+ err = -ENETDOWN;
+ goto out;
+ }
+
+ if (!(cfg->fc_flags & (RTF_LOCAL | RTF_ANYCAST)) &&
+ !netif_carrier_ok(dev))
+ fib6_nh->nh_flags |= RTNH_F_LINKDOWN;
+
+set_dev:
+ fib6_nh->nh_dev = dev;
+ err = 0;
+out:
+ if (idev)
+ in6_dev_put(idev);
+
+ if (err && dev)
+ dev_put(dev);
+
+ return err;
+}
+
static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
gfp_t gfp_flags,
struct netlink_ext_ack *extack)
{
struct net *net = cfg->fc_nlinfo.nl_net;
struct fib6_info *rt = NULL;
- struct net_device *dev = NULL;
- struct inet6_dev *idev = NULL;
struct fib6_table *table;
- int addr_type;
int err = -EINVAL;
/* RTF_PCPU is an internal flag; can not be set by userspace */
@@ -2940,30 +3053,6 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
goto out;
}
#endif
- if (cfg->fc_ifindex) {
- err = -ENODEV;
- dev = dev_get_by_index(net, cfg->fc_ifindex);
- if (!dev)
- goto out;
- idev = in6_dev_get(dev);
- if (!idev)
- goto out;
- }
-
- if (cfg->fc_flags & RTNH_F_ONLINK) {
- if (!dev) {
- NL_SET_ERR_MSG(extack,
- "Nexthop device required for onlink");
- err = -ENODEV;
- goto out;
- }
-
- if (!(dev->flags & IFF_UP)) {
- NL_SET_ERR_MSG(extack, "Nexthop device is not up");
- err = -ENETDOWN;
- goto out;
- }
- }
err = -ENOBUFS;
if (cfg->fc_nlinfo.nlh &&
@@ -3007,18 +3096,9 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
cfg->fc_protocol = RTPROT_BOOT;
rt->fib6_protocol = cfg->fc_protocol;
- addr_type = ipv6_addr_type(&cfg->fc_dst);
-
- if (cfg->fc_encap) {
- struct lwtunnel_state *lwtstate;
-
- err = lwtunnel_build_state(cfg->fc_encap_type,
- cfg->fc_encap, AF_INET6, cfg,
- &lwtstate, extack);
- if (err)
- goto out;
- rt->fib6_nh.nh_lwtstate = lwtstate_get(lwtstate);
- }
+ rt->fib6_table = table;
+ rt->fib6_metric = cfg->fc_metric;
+ rt->fib6_type = cfg->fc_type;
ipv6_addr_prefix(&rt->fib6_dst.addr, &cfg->fc_dst, cfg->fc_dst_len);
rt->fib6_dst.plen = cfg->fc_dst_len;
@@ -3029,62 +3109,13 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
ipv6_addr_prefix(&rt->fib6_src.addr, &cfg->fc_src, cfg->fc_src_len);
rt->fib6_src.plen = cfg->fc_src_len;
#endif
-
- rt->fib6_metric = cfg->fc_metric;
- rt->fib6_nh.nh_weight = 1;
-
- rt->fib6_type = cfg->fc_type;
-
- /* We cannot add true routes via loopback here,
- they would result in kernel looping; promote them to reject routes
- */
- if ((cfg->fc_flags & RTF_REJECT) ||
- (dev && (dev->flags & IFF_LOOPBACK) &&
- !(addr_type & IPV6_ADDR_LOOPBACK) &&
- !(cfg->fc_flags & RTF_LOCAL))) {
- /* hold loopback dev/idev if we haven't done so. */
- if (dev != net->loopback_dev) {
- if (dev) {
- dev_put(dev);
- in6_dev_put(idev);
- }
- dev = net->loopback_dev;
- dev_hold(dev);
- idev = in6_dev_get(dev);
- if (!idev) {
- err = -ENODEV;
- goto out;
- }
- }
- rt->fib6_flags = RTF_REJECT|RTF_NONEXTHOP;
- goto install_route;
- }
-
- if (cfg->fc_flags & RTF_GATEWAY) {
- err = ip6_validate_gw(net, cfg, &dev, &idev, extack);
- if (err)
- goto out;
-
- rt->fib6_nh.nh_gw = cfg->fc_gateway;
- }
-
- err = -ENODEV;
- if (!dev)
- goto out;
-
- if (idev->cnf.disable_ipv6) {
- NL_SET_ERR_MSG(extack, "IPv6 is disabled on nexthop device");
- err = -EACCES;
- goto out;
- }
-
- if (!(dev->flags & IFF_UP) && !cfg->fc_ignore_dev_down) {
- NL_SET_ERR_MSG(extack, "Nexthop device is not up");
- err = -ENETDOWN;
+ err = fib6_nh_init(net, &rt->fib6_nh, cfg, gfp_flags, extack);
+ if (err)
goto out;
- }
if (!ipv6_addr_any(&cfg->fc_prefsrc)) {
+ struct net_device *dev = fib6_info_nh_dev(rt);
+
if (!ipv6_chk_addr(net, &cfg->fc_prefsrc, dev, 0)) {
NL_SET_ERR_MSG(extack, "Invalid source address");
err = -EINVAL;
@@ -3097,24 +3128,8 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
rt->fib6_flags = cfg->fc_flags;
-install_route:
- if (!(rt->fib6_flags & (RTF_LOCAL | RTF_ANYCAST)) &&
- !netif_carrier_ok(dev))
- rt->fib6_nh.nh_flags |= RTNH_F_LINKDOWN;
- rt->fib6_nh.nh_flags |= (cfg->fc_flags & RTNH_F_ONLINK);
- rt->fib6_nh.nh_dev = dev;
- rt->fib6_table = table;
-
- if (idev)
- in6_dev_put(idev);
-
return rt;
out:
- if (dev)
- dev_put(dev);
- if (idev)
- in6_dev_put(idev);
-
fib6_info_release(rt);
return ERR_PTR(err);
}
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 06/13] ipv6: Create cleanup helper for fib6_nh
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
` (4 preceding siblings ...)
2019-03-27 3:29 ` [PATCH net-next 05/13] ipv6: Create init helper for fib6_nh David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 8:38 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 07/13] ipv6: Move gateway checks to a fib6_nh setting David Ahern
` (6 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
Move the fib6_nh cleanup code to a new helper, fib6_nh_release.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
include/net/ip6_fib.h | 1 +
net/ipv6/ip6_fib.c | 5 +----
net/ipv6/route.c | 8 ++++++++
3 files changed, 10 insertions(+), 4 deletions(-)
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index ce1f81345c8e..2d2a468b3d6d 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -447,6 +447,7 @@ static inline struct net_device *fib6_info_nh_dev(const struct fib6_info *f6i)
int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
struct fib6_config *cfg, gfp_t gfp_flags,
struct netlink_ext_ack *extack);
+void fib6_nh_release(struct fib6_nh *fib6_nh);
static inline
struct lwtunnel_state *fib6_info_nh_lwt(const struct fib6_info *f6i)
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 6613d8dbb0e5..db886085369b 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -199,10 +199,7 @@ void fib6_info_destroy_rcu(struct rcu_head *head)
free_percpu(f6i->rt6i_pcpu);
}
- lwtstate_put(f6i->fib6_nh.nh_lwtstate);
-
- if (f6i->fib6_nh.nh_dev)
- dev_put(f6i->fib6_nh.nh_dev);
+ fib6_nh_release(&f6i->fib6_nh);
ip_fib_metrics_put(f6i->fib6_metrics);
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 269ad7b9cdbb..4923a5f88e9b 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -3012,6 +3012,14 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
return err;
}
+void fib6_nh_release(struct fib6_nh *fib6_nh)
+{
+ if (fib6_nh->nh_dev)
+ dev_put(fib6_nh->nh_dev);
+
+ lwtstate_put(fib6_nh->nh_lwtstate);
+}
+
static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
gfp_t gfp_flags,
struct netlink_ext_ack *extack)
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 07/13] ipv6: Move gateway checks to a fib6_nh setting
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
` (5 preceding siblings ...)
2019-03-27 3:29 ` [PATCH net-next 06/13] ipv6: Create cleanup " David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 9:08 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 08/13] ipv6: Refactor fib6_ignore_linkdown David Ahern
` (5 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
The gateway setting is not per fib6_info entry but per-fib6_nh. Add a new
fib_nh_has_gw flag to fib6_nh and convert references to RTF_GATEWAY to
the new flag. For IPv6 address the flag is cheaper than checking that
nh_gw is non-0 like IPv4 does.
While this increases fib6_nh by 8-bytes, the effective allocation size of
a fib6_info is unchanged. The 8 bytes is recovered later with the
fib_nh_common change.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
include/net/ip6_fib.h | 1 +
include/net/ip6_route.h | 4 ++--
net/core/filter.c | 2 +-
net/ipv6/addrconf.c | 25 +++++++++++++------------
net/ipv6/ip6_fib.c | 9 ++++++---
net/ipv6/route.c | 26 ++++++++++++++++----------
6 files changed, 39 insertions(+), 28 deletions(-)
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 2d2a468b3d6d..3b04b318cf13 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -126,6 +126,7 @@ struct rt6_exception {
struct fib6_nh {
struct in6_addr nh_gw;
+ bool fib_nh_has_gw;
struct net_device *nh_dev;
struct lwtunnel_state *nh_lwtstate;
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 7ab119936e69..95cd8a2f6284 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -68,8 +68,8 @@ static inline bool rt6_need_strict(const struct in6_addr *daddr)
static inline bool rt6_qualify_for_ecmp(const struct fib6_info *f6i)
{
- return (f6i->fib6_flags & (RTF_GATEWAY|RTF_ADDRCONF|RTF_DYNAMIC)) ==
- RTF_GATEWAY;
+ return !(f6i->fib6_flags & (RTF_ADDRCONF|RTF_DYNAMIC)) &&
+ f6i->fib6_nh.fib_nh_has_gw;
}
void ip6_route_input(struct sk_buff *skb);
diff --git a/net/core/filter.c b/net/core/filter.c
index 647c63a7b25b..248207171475 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -4677,7 +4677,7 @@ static int bpf_ipv6_fib_lookup(struct net *net, struct bpf_fib_lookup *params,
if (f6i->fib6_nh.nh_lwtstate)
return BPF_FIB_LKUP_RET_UNSUPP_LWT;
- if (f6i->fib6_flags & RTF_GATEWAY)
+ if (f6i->fib6_nh.fib_nh_has_gw)
*dst = f6i->fib6_nh.nh_gw;
dev = f6i->fib6_nh.nh_dev;
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 4ae17a966ae3..c5ac08fc6cc9 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -173,7 +173,8 @@ static int addrconf_ifdown(struct net_device *dev, int how);
static struct fib6_info *addrconf_get_prefix_route(const struct in6_addr *pfx,
int plen,
const struct net_device *dev,
- u32 flags, u32 noflags);
+ u32 flags, u32 noflags,
+ bool no_gw);
static void addrconf_dad_start(struct inet6_ifaddr *ifp);
static void addrconf_dad_work(struct work_struct *w);
@@ -1230,10 +1231,8 @@ cleanup_prefix_route(struct inet6_ifaddr *ifp, unsigned long expires, bool del_r
{
struct fib6_info *f6i;
- f6i = addrconf_get_prefix_route(&ifp->addr,
- ifp->prefix_len,
- ifp->idev->dev,
- 0, RTF_GATEWAY | RTF_DEFAULT);
+ f6i = addrconf_get_prefix_route(&ifp->addr, ifp->prefix_len,
+ ifp->idev->dev, 0, RTF_DEFAULT, true);
if (f6i) {
if (del_rt)
ip6_del_rt(dev_net(ifp->idev->dev), f6i);
@@ -2402,7 +2401,8 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, u32 metric,
static struct fib6_info *addrconf_get_prefix_route(const struct in6_addr *pfx,
int plen,
const struct net_device *dev,
- u32 flags, u32 noflags)
+ u32 flags, u32 noflags,
+ bool no_gw)
{
struct fib6_node *fn;
struct fib6_info *rt = NULL;
@@ -2421,6 +2421,8 @@ static struct fib6_info *addrconf_get_prefix_route(const struct in6_addr *pfx,
for_each_fib6_node_rt_rcu(fn) {
if (rt->fib6_nh.nh_dev->ifindex != dev->ifindex)
continue;
+ if (no_gw && rt->fib6_nh.fib_nh_has_gw)
+ continue;
if ((rt->fib6_flags & flags) != flags)
continue;
if ((rt->fib6_flags & noflags) != 0)
@@ -2717,7 +2719,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao)
pinfo->prefix_len,
dev,
RTF_ADDRCONF | RTF_PREFIX_RT,
- RTF_GATEWAY | RTF_DEFAULT);
+ RTF_DEFAULT, true);
if (rt) {
/* Autoconf prefix route */
@@ -4588,10 +4590,8 @@ static int modify_prefix_route(struct inet6_ifaddr *ifp,
struct fib6_info *f6i;
u32 prio;
- f6i = addrconf_get_prefix_route(&ifp->addr,
- ifp->prefix_len,
- ifp->idev->dev,
- 0, RTF_GATEWAY | RTF_DEFAULT);
+ f6i = addrconf_get_prefix_route(&ifp->addr, ifp->prefix_len,
+ ifp->idev->dev, 0, RTF_DEFAULT, true);
if (!f6i)
return -ENOENT;
@@ -5972,7 +5972,8 @@ static void __ipv6_ifa_notify(int event, struct inet6_ifaddr *ifp)
struct fib6_info *rt;
rt = addrconf_get_prefix_route(&ifp->peer_addr, 128,
- ifp->idev->dev, 0, 0);
+ ifp->idev->dev, 0, 0,
+ false);
if (rt)
ip6_del_rt(net, rt);
}
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index db886085369b..91ce84ecdb57 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -2294,6 +2294,7 @@ static int ipv6_route_seq_show(struct seq_file *seq, void *v)
{
struct fib6_info *rt = v;
struct ipv6_route_iter *iter = seq->private;
+ unsigned int flags = rt->fib6_flags;
const struct net_device *dev;
seq_printf(seq, "%pi6 %02x ", &rt->fib6_dst.addr, rt->fib6_dst.plen);
@@ -2303,15 +2304,17 @@ static int ipv6_route_seq_show(struct seq_file *seq, void *v)
#else
seq_puts(seq, "00000000000000000000000000000000 00 ");
#endif
- if (rt->fib6_flags & RTF_GATEWAY)
+ if (rt->fib6_nh.fib_nh_has_gw) {
+ flags |= RTF_GATEWAY;
seq_printf(seq, "%pi6", &rt->fib6_nh.nh_gw);
- else
+ } else {
seq_puts(seq, "00000000000000000000000000000000");
+ }
dev = rt->fib6_nh.nh_dev;
seq_printf(seq, " %08x %08x %08x %08x %8s\n",
rt->fib6_metric, atomic_read(&rt->fib6_ref), 0,
- rt->fib6_flags, dev ? dev->name : "");
+ flags, dev ? dev->name : "");
iter->w.leaf = NULL;
return 0;
}
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 4923a5f88e9b..b99c05b580e6 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -533,7 +533,7 @@ static void rt6_probe(struct fib6_info *rt)
* Router Reachability Probe MUST be rate-limited
* to no more than one per minute.
*/
- if (!rt || !(rt->fib6_flags & RTF_GATEWAY))
+ if (!rt || !rt->fib6_nh.fib_nh_has_gw)
return;
nh_gw = &rt->fib6_nh.nh_gw;
@@ -595,7 +595,7 @@ static inline enum rt6_nud_state rt6_check_neigh(struct fib6_info *rt)
struct neighbour *neigh;
if (rt->fib6_flags & RTF_NONEXTHOP ||
- !(rt->fib6_flags & RTF_GATEWAY))
+ !rt->fib6_nh.fib_nh_has_gw)
return RT6_NUD_SUCCEED;
rcu_read_lock_bh();
@@ -784,7 +784,7 @@ static struct fib6_info *rt6_select(struct net *net, struct fib6_node *fn,
static bool rt6_is_gw_or_nonexthop(const struct fib6_info *rt)
{
- return (rt->fib6_flags & (RTF_NONEXTHOP | RTF_GATEWAY));
+ return (rt->fib6_flags & RTF_NONEXTHOP) || rt->fib6_nh.fib_nh_has_gw;
}
#ifdef CONFIG_IPV6_ROUTE_INFO
@@ -989,8 +989,11 @@ static void ip6_rt_copy_init(struct rt6_info *rt, struct fib6_info *ort)
rt->rt6i_dst = ort->fib6_dst;
rt->rt6i_idev = dev ? in6_dev_get(dev) : NULL;
- rt->rt6i_gateway = ort->fib6_nh.nh_gw;
rt->rt6i_flags = ort->fib6_flags;
+ if (ort->fib6_nh.fib_nh_has_gw) {
+ rt->rt6i_gateway = ort->fib6_nh.nh_gw;
+ rt->rt6i_flags |= RTF_GATEWAY;
+ }
rt6_set_from(rt, ort);
#ifdef CONFIG_IPV6_SUBTREES
rt->rt6i_src = ort->fib6_src;
@@ -1870,7 +1873,7 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table,
rcu_read_unlock();
return rt;
} else if (unlikely((fl6->flowi6_flags & FLOWI_FLAG_KNOWN_NH) &&
- !(f6i->fib6_flags & RTF_GATEWAY))) {
+ !f6i->fib6_nh.fib_nh_has_gw)) {
/* Create a RTF_CACHE clone which will not be
* owned by the fib6 tree. It is for the special case where
* the daddr in the skb during the neighbor look-up is different
@@ -2440,7 +2443,7 @@ static struct rt6_info *__ip6_route_redirect(struct net *net,
continue;
if (rt->fib6_flags & RTF_REJECT)
break;
- if (!(rt->fib6_flags & RTF_GATEWAY))
+ if (!rt->fib6_nh.fib_nh_has_gw)
continue;
if (fl6->flowi6_oif != rt->fib6_nh.nh_dev->ifindex)
continue;
@@ -2977,6 +2980,7 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
goto out;
fib6_nh->nh_gw = cfg->fc_gateway;
+ fib6_nh->fib_nh_has_gw = 1;
}
err = -ENODEV;
@@ -3134,7 +3138,7 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
} else
rt->fib6_prefsrc.plen = 0;
- rt->fib6_flags = cfg->fc_flags;
+ rt->fib6_flags = cfg->fc_flags & ~RTF_GATEWAY;
return rt;
out:
@@ -3470,7 +3474,8 @@ static struct fib6_info *rt6_get_route_info(struct net *net,
for_each_fib6_node_rt_rcu(fn) {
if (rt->fib6_nh.nh_dev->ifindex != ifindex)
continue;
- if ((rt->fib6_flags & (RTF_ROUTEINFO|RTF_GATEWAY)) != (RTF_ROUTEINFO|RTF_GATEWAY))
+ if (!(rt->fib6_flags & RTF_ROUTEINFO) ||
+ !rt->fib6_nh.fib_nh_has_gw)
continue;
if (!ipv6_addr_equal(&rt->fib6_nh.nh_gw, gwaddr))
continue;
@@ -3791,7 +3796,7 @@ void rt6_remove_prefsrc(struct inet6_ifaddr *ifp)
fib6_clean_all(net, fib6_remove_prefsrc, &adni);
}
-#define RTF_RA_ROUTER (RTF_ADDRCONF | RTF_DEFAULT | RTF_GATEWAY)
+#define RTF_RA_ROUTER (RTF_ADDRCONF | RTF_DEFAULT)
/* Remove routers and update dst entries when gateway turn into host. */
static int fib6_clean_tohost(struct fib6_info *rt, void *arg)
@@ -3799,6 +3804,7 @@ static int fib6_clean_tohost(struct fib6_info *rt, void *arg)
struct in6_addr *gateway = (struct in6_addr *)arg;
if (((rt->fib6_flags & RTF_RA_ROUTER) == RTF_RA_ROUTER) &&
+ rt->fib6_nh.fib_nh_has_gw &&
ipv6_addr_equal(gateway, &rt->fib6_nh.nh_gw)) {
return -1;
}
@@ -4587,7 +4593,7 @@ static int rt6_nexthop_info(struct sk_buff *skb, struct fib6_info *rt,
rcu_read_unlock();
}
- if (rt->fib6_flags & RTF_GATEWAY) {
+ if (rt->fib6_nh.fib_nh_has_gw) {
if (nla_put_in6_addr(skb, RTA_GATEWAY, &rt->fib6_nh.nh_gw) < 0)
goto nla_put_failure;
}
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 08/13] ipv6: Refactor fib6_ignore_linkdown
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
` (6 preceding siblings ...)
2019-03-27 3:29 ` [PATCH net-next 07/13] ipv6: Move gateway checks to a fib6_nh setting David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 9:19 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 09/13] ipv6: Change rt6_add_nexthop and rt6_nexthop_info to take fib6_nh David Ahern
` (4 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
fib6_ignore_linkdown takes a fib6_info but only looks at the net_device
and its IPv6 config. Change it to take a net_device over a fib6_info as
its input argument.
In addition, move it to a header file to make the check inline and usable
later with IPv4 code without going through the ipv6 stub, and rename to
ip6_ignore_linkdown since it is only checking the setting based on the
ipv6 struct on a device.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
include/net/addrconf.h | 8 ++++++++
net/ipv6/route.c | 21 +++------------------
2 files changed, 11 insertions(+), 18 deletions(-)
diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 269ec27385e9..ec8e6784a6f7 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -425,6 +425,14 @@ static inline void in6_dev_hold(struct inet6_dev *idev)
refcount_inc(&idev->refcnt);
}
+/* called with rcu_read_lock held */
+static inline bool ip6_ignore_linkdown(const struct net_device *dev)
+{
+ const struct inet6_dev *idev = __in6_dev_get(dev);
+
+ return !!idev->cnf.ignore_routes_with_linkdown;
+}
+
void inet6_ifa_finish_destroy(struct inet6_ifaddr *ifp);
static inline void in6_ifa_put(struct inet6_ifaddr *ifp)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index b99c05b580e6..14b1c22c5a8d 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -639,21 +639,6 @@ static int rt6_score_route(struct fib6_info *rt, int oif, int strict)
return m;
}
-/* called with rc_read_lock held */
-static inline bool fib6_ignore_linkdown(const struct fib6_info *f6i)
-{
- const struct net_device *dev = fib6_info_nh_dev(f6i);
- bool rc = false;
-
- if (dev) {
- const struct inet6_dev *idev = __in6_dev_get(dev);
-
- rc = !!idev->cnf.ignore_routes_with_linkdown;
- }
-
- return rc;
-}
-
static struct fib6_info *find_match(struct fib6_info *rt, int oif, int strict,
int *mpri, struct fib6_info *match,
bool *do_rr)
@@ -664,7 +649,7 @@ static struct fib6_info *find_match(struct fib6_info *rt, int oif, int strict,
if (rt->fib6_nh.nh_flags & RTNH_F_DEAD)
goto out;
- if (fib6_ignore_linkdown(rt) &&
+ if (ip6_ignore_linkdown(rt->fib6_nh.nh_dev) &&
rt->fib6_nh.nh_flags & RTNH_F_LINKDOWN &&
!(strict & RT6_LOOKUP_F_IGNORE_LINKSTATE))
goto out;
@@ -3855,7 +3840,7 @@ static bool rt6_is_dead(const struct fib6_info *rt)
{
if (rt->fib6_nh.nh_flags & RTNH_F_DEAD ||
(rt->fib6_nh.nh_flags & RTNH_F_LINKDOWN &&
- fib6_ignore_linkdown(rt)))
+ ip6_ignore_linkdown(rt->fib6_nh.nh_dev)))
return true;
return false;
@@ -4588,7 +4573,7 @@ static int rt6_nexthop_info(struct sk_buff *skb, struct fib6_info *rt,
*flags |= RTNH_F_LINKDOWN;
rcu_read_lock();
- if (fib6_ignore_linkdown(rt))
+ if (ip6_ignore_linkdown(rt->fib6_nh.nh_dev))
*flags |= RTNH_F_DEAD;
rcu_read_unlock();
}
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 09/13] ipv6: Change rt6_add_nexthop and rt6_nexthop_info to take fib6_nh
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
` (7 preceding siblings ...)
2019-03-27 3:29 ` [PATCH net-next 08/13] ipv6: Refactor fib6_ignore_linkdown David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 9:24 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 10/13] ipv4: Rename fib_nh entries David Ahern
` (3 subsequent siblings)
12 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
rt6_add_nexthop and rt6_nexthop_info only need the fib6_info for the
gateway flag and the nexthop weight, and the presence of a gateway is now
per-nexthop. Update the signatures to take a fib6_nh and nexthop weight
and better align with the ipv4 versions.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
net/ipv6/route.c | 39 ++++++++++++++++++++-------------------
1 file changed, 20 insertions(+), 19 deletions(-)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 14b1c22c5a8d..79fd43bd9e4e 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -4563,37 +4563,37 @@ static size_t rt6_nlmsg_size(struct fib6_info *rt)
+ nexthop_len;
}
-static int rt6_nexthop_info(struct sk_buff *skb, struct fib6_info *rt,
+static int rt6_nexthop_info(struct sk_buff *skb, const struct fib6_nh *fib6_nh,
unsigned int *flags, bool skip_oif)
{
- if (rt->fib6_nh.nh_flags & RTNH_F_DEAD)
+ if (fib6_nh->nh_flags & RTNH_F_DEAD)
*flags |= RTNH_F_DEAD;
- if (rt->fib6_nh.nh_flags & RTNH_F_LINKDOWN) {
+ if (fib6_nh->nh_flags & RTNH_F_LINKDOWN) {
*flags |= RTNH_F_LINKDOWN;
rcu_read_lock();
- if (ip6_ignore_linkdown(rt->fib6_nh.nh_dev))
+ if (ip6_ignore_linkdown(fib6_nh->nh_dev))
*flags |= RTNH_F_DEAD;
rcu_read_unlock();
}
- if (rt->fib6_nh.fib_nh_has_gw) {
- if (nla_put_in6_addr(skb, RTA_GATEWAY, &rt->fib6_nh.nh_gw) < 0)
+ if (fib6_nh->fib_nh_has_gw) {
+ if (nla_put_in6_addr(skb, RTA_GATEWAY, &fib6_nh->nh_gw) < 0)
goto nla_put_failure;
}
- *flags |= (rt->fib6_nh.nh_flags & RTNH_F_ONLINK);
- if (rt->fib6_nh.nh_flags & RTNH_F_OFFLOAD)
+ *flags |= (fib6_nh->nh_flags & RTNH_F_ONLINK);
+ if (fib6_nh->nh_flags & RTNH_F_OFFLOAD)
*flags |= RTNH_F_OFFLOAD;
/* not needed for multipath encoding b/c it has a rtnexthop struct */
- if (!skip_oif && rt->fib6_nh.nh_dev &&
- nla_put_u32(skb, RTA_OIF, rt->fib6_nh.nh_dev->ifindex))
+ if (!skip_oif && fib6_nh->nh_dev &&
+ nla_put_u32(skb, RTA_OIF, fib6_nh->nh_dev->ifindex))
goto nla_put_failure;
- if (rt->fib6_nh.nh_lwtstate &&
- lwtunnel_fill_encap(skb, rt->fib6_nh.nh_lwtstate) < 0)
+ if (fib6_nh->nh_lwtstate &&
+ lwtunnel_fill_encap(skb, fib6_nh->nh_lwtstate) < 0)
goto nla_put_failure;
return 0;
@@ -4603,9 +4603,9 @@ static int rt6_nexthop_info(struct sk_buff *skb, struct fib6_info *rt,
}
/* add multipath next hop */
-static int rt6_add_nexthop(struct sk_buff *skb, struct fib6_info *rt)
+static int rt6_add_nexthop(struct sk_buff *skb, const struct fib6_nh *fib6_nh)
{
- const struct net_device *dev = rt->fib6_nh.nh_dev;
+ const struct net_device *dev = fib6_nh->nh_dev;
struct rtnexthop *rtnh;
unsigned int flags = 0;
@@ -4613,10 +4613,10 @@ static int rt6_add_nexthop(struct sk_buff *skb, struct fib6_info *rt)
if (!rtnh)
goto nla_put_failure;
- rtnh->rtnh_hops = rt->fib6_nh.nh_weight - 1;
+ rtnh->rtnh_hops = fib6_nh->nh_weight - 1;
rtnh->rtnh_ifindex = dev ? dev->ifindex : 0;
- if (rt6_nexthop_info(skb, rt, &flags, true) < 0)
+ if (rt6_nexthop_info(skb, fib6_nh, &flags, true) < 0)
goto nla_put_failure;
rtnh->rtnh_flags = flags;
@@ -4746,18 +4746,19 @@ static int rt6_fill_node(struct net *net, struct sk_buff *skb,
if (!mp)
goto nla_put_failure;
- if (rt6_add_nexthop(skb, rt) < 0)
+ if (rt6_add_nexthop(skb, &rt->fib6_nh) < 0)
goto nla_put_failure;
list_for_each_entry_safe(sibling, next_sibling,
&rt->fib6_siblings, fib6_siblings) {
- if (rt6_add_nexthop(skb, sibling) < 0)
+ if (rt6_add_nexthop(skb, &sibling->fib6_nh) < 0)
goto nla_put_failure;
}
nla_nest_end(skb, mp);
} else {
- if (rt6_nexthop_info(skb, rt, &rtm->rtm_flags, false) < 0)
+ if (rt6_nexthop_info(skb, &rt->fib6_nh, &rtm->rtm_flags,
+ false) < 0)
goto nla_put_failure;
}
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 10/13] ipv4: Rename fib_nh entries
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
` (8 preceding siblings ...)
2019-03-27 3:29 ` [PATCH net-next 09/13] ipv6: Change rt6_add_nexthop and rt6_nexthop_info to take fib6_nh David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 3:29 ` [PATCH net-next 11/13] ipv6: Rename fib6_nh entries David Ahern
` (2 subsequent siblings)
12 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
Rename fib_nh entries that will be moved to a fib_nh_common struct.
Specifically, the device, oif, gateway, flags, scope, lwtstate,
nh_weight and nh_upper_bound are common with all nexthop definitions.
In the process shorten fib_nh_lwtstate to fib_nh_lws to avoid really
long lines.
Rename only; no functional change intended.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c | 12 +-
.../net/ethernet/mellanox/mlxsw/spectrum_router.c | 22 +-
drivers/net/ethernet/rocker/rocker_ofdpa.c | 10 +-
include/net/ip_fib.h | 24 +--
include/trace/events/fib.h | 7 +-
net/core/filter.c | 8 +-
net/ipv4/fib_frontend.c | 10 +-
net/ipv4/fib_semantics.c | 229 +++++++++++----------
net/ipv4/fib_trie.c | 12 +-
net/ipv4/route.c | 18 +-
10 files changed, 178 insertions(+), 174 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c
index 5633f8572800..8212bfd05733 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c
@@ -122,7 +122,7 @@ static void mlx5_lag_fib_route_event(struct mlx5_lag *ldev,
/* Handle add/replace event */
if (fi->fib_nhs == 1) {
if (__mlx5_lag_is_active(ldev)) {
- struct net_device *nh_dev = fi->fib_nh[0].nh_dev;
+ struct net_device *nh_dev = fi->fib_nh[0].fib_nh_dev;
int i = mlx5_lag_dev_get_netdev_idx(ldev, nh_dev);
mlx5_lag_set_port_affinity(ldev, ++i);
@@ -134,10 +134,10 @@ static void mlx5_lag_fib_route_event(struct mlx5_lag *ldev,
return;
/* Verify next hops are ports of the same hca */
- if (!(fi->fib_nh[0].nh_dev == ldev->pf[0].netdev &&
- fi->fib_nh[1].nh_dev == ldev->pf[1].netdev) &&
- !(fi->fib_nh[0].nh_dev == ldev->pf[1].netdev &&
- fi->fib_nh[1].nh_dev == ldev->pf[0].netdev)) {
+ if (!(fi->fib_nh[0].fib_nh_dev == ldev->pf[0].netdev &&
+ fi->fib_nh[1].fib_nh_dev == ldev->pf[1].netdev) &&
+ !(fi->fib_nh[0].fib_nh_dev == ldev->pf[1].netdev &&
+ fi->fib_nh[1].fib_nh_dev == ldev->pf[0].netdev)) {
mlx5_core_warn(ldev->pf[0].dev, "Multipath offload require two ports of the same HCA\n");
return;
}
@@ -167,7 +167,7 @@ static void mlx5_lag_fib_nexthop_event(struct mlx5_lag *ldev,
/* nh added/removed */
if (event == FIB_EVENT_NH_DEL) {
- int i = mlx5_lag_dev_get_netdev_idx(ldev, fib_nh->nh_dev);
+ int i = mlx5_lag_dev_get_netdev_idx(ldev, fib_nh->fib_nh_dev);
if (i >= 0) {
i = (i + 1) % 2 + 1; /* peer port */
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 52fed8c7bf1e..06e1a98fcd6a 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -3610,7 +3610,7 @@ static bool mlxsw_sp_nexthop4_ipip_type(const struct mlxsw_sp *mlxsw_sp,
const struct fib_nh *fib_nh,
enum mlxsw_sp_ipip_type *p_ipipt)
{
- struct net_device *dev = fib_nh->nh_dev;
+ struct net_device *dev = fib_nh->fib_nh_dev;
return dev &&
fib_nh->nh_parent->fib_type == RTN_UNICAST &&
@@ -3637,7 +3637,7 @@ static int mlxsw_sp_nexthop4_type_init(struct mlxsw_sp *mlxsw_sp,
struct fib_nh *fib_nh)
{
const struct mlxsw_sp_ipip_ops *ipip_ops;
- struct net_device *dev = fib_nh->nh_dev;
+ struct net_device *dev = fib_nh->fib_nh_dev;
struct mlxsw_sp_ipip_entry *ipip_entry;
struct mlxsw_sp_rif *rif;
int err;
@@ -3681,18 +3681,18 @@ static int mlxsw_sp_nexthop4_init(struct mlxsw_sp *mlxsw_sp,
struct mlxsw_sp_nexthop *nh,
struct fib_nh *fib_nh)
{
- struct net_device *dev = fib_nh->nh_dev;
+ struct net_device *dev = fib_nh->fib_nh_dev;
struct in_device *in_dev;
int err;
nh->nh_grp = nh_grp;
nh->key.fib_nh = fib_nh;
#ifdef CONFIG_IP_ROUTE_MULTIPATH
- nh->nh_weight = fib_nh->nh_weight;
+ nh->nh_weight = fib_nh->fib_nh_weight;
#else
nh->nh_weight = 1;
#endif
- memcpy(&nh->gw_addr, &fib_nh->nh_gw, sizeof(fib_nh->nh_gw));
+ memcpy(&nh->gw_addr, &fib_nh->fib_nh_gw4, sizeof(fib_nh->fib_nh_gw4));
err = mlxsw_sp_nexthop_insert(mlxsw_sp, nh);
if (err)
return err;
@@ -3705,7 +3705,7 @@ static int mlxsw_sp_nexthop4_init(struct mlxsw_sp *mlxsw_sp,
in_dev = __in_dev_get_rtnl(dev);
if (in_dev && IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev) &&
- fib_nh->nh_flags & RTNH_F_LINKDOWN)
+ fib_nh->fib_nh_flags & RTNH_F_LINKDOWN)
return 0;
err = mlxsw_sp_nexthop4_type_init(mlxsw_sp, nh, fib_nh);
@@ -3804,7 +3804,7 @@ static void mlxsw_sp_nexthop_rif_gone_sync(struct mlxsw_sp *mlxsw_sp,
static bool mlxsw_sp_fi_is_gateway(const struct mlxsw_sp *mlxsw_sp,
const struct fib_info *fi)
{
- return fi->fib_nh->nh_scope == RT_SCOPE_LINK ||
+ return fi->fib_nh->fib_nh_scope == RT_SCOPE_LINK ||
mlxsw_sp_nexthop4_ipip_type(mlxsw_sp, fi->fib_nh, NULL);
}
@@ -3966,7 +3966,7 @@ mlxsw_sp_fib4_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry)
fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_BLACKHOLE ||
fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_IPIP_DECAP ||
fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_NVE_DECAP) {
- nh_grp->nexthops->key.fib_nh->nh_flags |= RTNH_F_OFFLOAD;
+ nh_grp->nexthops->key.fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
return;
}
@@ -3974,9 +3974,9 @@ mlxsw_sp_fib4_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry)
struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];
if (nh->offloaded)
- nh->key.fib_nh->nh_flags |= RTNH_F_OFFLOAD;
+ nh->key.fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
else
- nh->key.fib_nh->nh_flags &= ~RTNH_F_OFFLOAD;
+ nh->key.fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
}
}
@@ -3992,7 +3992,7 @@ mlxsw_sp_fib4_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry)
for (i = 0; i < nh_grp->count; i++) {
struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];
- nh->key.fib_nh->nh_flags &= ~RTNH_F_OFFLOAD;
+ nh->key.fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
}
}
diff --git a/drivers/net/ethernet/rocker/rocker_ofdpa.c b/drivers/net/ethernet/rocker/rocker_ofdpa.c
index fa296a7c255d..30a49802fb51 100644
--- a/drivers/net/ethernet/rocker/rocker_ofdpa.c
+++ b/drivers/net/ethernet/rocker/rocker_ofdpa.c
@@ -2288,11 +2288,11 @@ static int ofdpa_port_fib_ipv4(struct ofdpa_port *ofdpa_port, __be32 dst,
nh = fi->fib_nh;
nh_on_port = (fi->fib_dev == ofdpa_port->dev);
- has_gw = !!nh->nh_gw;
+ has_gw = !!nh->fib_nh_gw4;
if (has_gw && nh_on_port) {
err = ofdpa_port_ipv4_nh(ofdpa_port, flags,
- nh->nh_gw, &index);
+ nh->fib_nh_gw4, &index);
if (err)
return err;
@@ -2749,7 +2749,7 @@ static int ofdpa_fib4_add(struct rocker *rocker,
fen_info->tb_id, 0);
if (err)
return err;
- fen_info->fi->fib_nh->nh_flags |= RTNH_F_OFFLOAD;
+ fen_info->fi->fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
return 0;
}
@@ -2764,7 +2764,7 @@ static int ofdpa_fib4_del(struct rocker *rocker,
ofdpa_port = ofdpa_port_dev_lower_find(fen_info->fi->fib_dev, rocker);
if (!ofdpa_port)
return 0;
- fen_info->fi->fib_nh->nh_flags &= ~RTNH_F_OFFLOAD;
+ fen_info->fi->fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
return ofdpa_port_fib_ipv4(ofdpa_port, htonl(fen_info->dst),
fen_info->dst_len, fen_info->fi,
fen_info->tb_id, OFDPA_OP_FLAG_REMOVE);
@@ -2791,7 +2791,7 @@ static void ofdpa_fib4_abort(struct rocker *rocker)
rocker);
if (!ofdpa_port)
continue;
- flow_entry->fi->fib_nh->nh_flags &= ~RTNH_F_OFFLOAD;
+ flow_entry->fi->fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
ofdpa_flow_tbl_del(ofdpa_port, OFDPA_OP_FLAG_REMOVE,
flow_entry);
}
diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 5a4df0ba175e..029acd333d29 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -77,26 +77,26 @@ struct fnhe_hash_bucket {
#define FNHE_RECLAIM_DEPTH 5
struct fib_nh {
- struct net_device *nh_dev;
+ struct net_device *fib_nh_dev;
struct hlist_node nh_hash;
struct fib_info *nh_parent;
- unsigned int nh_flags;
- unsigned char nh_scope;
+ unsigned int fib_nh_flags;
+ unsigned char fib_nh_scope;
#ifdef CONFIG_IP_ROUTE_MULTIPATH
- int nh_weight;
- atomic_t nh_upper_bound;
+ int fib_nh_weight;
+ atomic_t fib_nh_upper_bound;
#endif
#ifdef CONFIG_IP_ROUTE_CLASSID
__u32 nh_tclassid;
#endif
- int nh_oif;
- __be32 nh_gw;
+ int fib_nh_oif;
+ __be32 fib_nh_gw4;
__be32 nh_saddr;
int nh_saddr_genid;
struct rtable __rcu * __percpu *nh_pcpu_rth_output;
struct rtable __rcu *nh_rth_input;
struct fnhe_hash_bucket __rcu *nh_exceptions;
- struct lwtunnel_state *nh_lwtstate;
+ struct lwtunnel_state *fib_nh_lws;
};
/*
@@ -125,7 +125,7 @@ struct fib_info {
int fib_nhs;
struct rcu_head rcu;
struct fib_nh fib_nh[0];
-#define fib_dev fib_nh[0].nh_dev
+#define fib_dev fib_nh[0].fib_nh_dev
};
@@ -180,9 +180,9 @@ __be32 fib_info_update_nh_saddr(struct net *net, struct fib_nh *nh);
atomic_read(&(net)->ipv4.dev_addr_genid)) ? \
FIB_RES_NH(res).nh_saddr : \
fib_info_update_nh_saddr((net), &FIB_RES_NH(res)))
-#define FIB_RES_GW(res) (FIB_RES_NH(res).nh_gw)
-#define FIB_RES_DEV(res) (FIB_RES_NH(res).nh_dev)
-#define FIB_RES_OIF(res) (FIB_RES_NH(res).nh_oif)
+#define FIB_RES_GW(res) (FIB_RES_NH(res).fib_nh_gw4)
+#define FIB_RES_DEV(res) (FIB_RES_NH(res).fib_nh_dev)
+#define FIB_RES_OIF(res) (FIB_RES_NH(res).fib_nh_oif)
#define FIB_RES_PREFSRC(net, res) ((res).fi->fib_prefsrc ? : \
FIB_RES_SADDR(net, res))
diff --git a/include/trace/events/fib.h b/include/trace/events/fib.h
index 6271bab63bfb..61ea7a24c8e5 100644
--- a/include/trace/events/fib.h
+++ b/include/trace/events/fib.h
@@ -63,13 +63,16 @@ TRACE_EVENT(fib_table_lookup,
}
if (nh) {
+ struct net_device *dev;
+
p32 = (__be32 *) __entry->saddr;
*p32 = nh->nh_saddr;
p32 = (__be32 *) __entry->gw;
- *p32 = nh->nh_gw;
+ *p32 = nh->fib_nh_gw4;
- __assign_str(name, nh->nh_dev ? nh->nh_dev->name : "-");
+ dev = nh->fib_nh_dev;
+ __assign_str(name, dev ? dev->name : "-");
} else {
p32 = (__be32 *) __entry->saddr;
*p32 = 0;
diff --git a/net/core/filter.c b/net/core/filter.c
index 248207171475..7adcba70092e 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -4560,12 +4560,12 @@ static int bpf_ipv4_fib_lookup(struct net *net, struct bpf_fib_lookup *params,
nh = &res.fi->fib_nh[res.nh_sel];
/* do not handle lwt encaps right now */
- if (nh->nh_lwtstate)
+ if (nh->fib_nh_lws)
return BPF_FIB_LKUP_RET_UNSUPP_LWT;
- dev = nh->nh_dev;
- if (nh->nh_gw)
- params->ipv4_dst = nh->nh_gw;
+ dev = nh->fib_nh_dev;
+ if (nh->fib_nh_gw4)
+ params->ipv4_dst = nh->fib_nh_gw4;
params->rt_metric = res.fi->fib_priority;
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index ed14ec245584..ffbe24397dbe 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -324,16 +324,16 @@ bool fib_info_nh_uses_dev(struct fib_info *fi, const struct net_device *dev)
for (ret = 0; ret < fi->fib_nhs; ret++) {
struct fib_nh *nh = &fi->fib_nh[ret];
- if (nh->nh_dev == dev) {
+ if (nh->fib_nh_dev == dev) {
dev_match = true;
break;
- } else if (l3mdev_master_ifindex_rcu(nh->nh_dev) == dev->ifindex) {
+ } else if (l3mdev_master_ifindex_rcu(nh->fib_nh_dev) == dev->ifindex) {
dev_match = true;
break;
}
}
#else
- if (fi->fib_nh[0].nh_dev == dev)
+ if (fi->fib_nh[0].fib_nh_dev == dev)
dev_match = true;
#endif
@@ -390,7 +390,7 @@ static int __fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst,
dev_match = fib_info_nh_uses_dev(res.fi, dev);
if (dev_match) {
- ret = FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST;
+ ret = FIB_RES_NH(res).fib_nh_scope >= RT_SCOPE_HOST;
return ret;
}
if (no_addr)
@@ -402,7 +402,7 @@ static int __fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst,
ret = 0;
if (fib_lookup(net, &fl4, &res, FIB_LOOKUP_IGNORE_LINKSTATE) == 0) {
if (res.type == RTN_UNICAST)
- ret = FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST;
+ ret = FIB_RES_NH(res).fib_nh_scope >= RT_SCOPE_HOST;
}
return ret;
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 72cf51e1eaa9..46e941f505d5 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -210,10 +210,10 @@ void fib_nh_release(struct net *net, struct fib_nh *fib_nh)
if (fib_nh->nh_tclassid)
net->ipv4.fib_num_tclassid_users--;
#endif
- if (fib_nh->nh_dev)
- dev_put(fib_nh->nh_dev);
+ if (fib_nh->fib_nh_dev)
+ dev_put(fib_nh->fib_nh_dev);
- lwtstate_put(fib_nh->nh_lwtstate);
+ lwtstate_put(fib_nh->fib_nh_lws);
free_nh_exceptions(fib_nh);
rt_fibinfo_free_cpus(fib_nh->nh_pcpu_rth_output);
rt_fibinfo_free(&fib_nh->nh_rth_input);
@@ -253,7 +253,7 @@ void fib_release_info(struct fib_info *fi)
if (fi->fib_prefsrc)
hlist_del(&fi->fib_lhash);
change_nexthops(fi) {
- if (!nexthop_nh->nh_dev)
+ if (!nexthop_nh->fib_nh_dev)
continue;
hlist_del(&nexthop_nh->nh_hash);
} endfor_nexthops(fi)
@@ -268,17 +268,17 @@ static inline int nh_comp(const struct fib_info *fi, const struct fib_info *ofi)
const struct fib_nh *onh = ofi->fib_nh;
for_nexthops(fi) {
- if (nh->nh_oif != onh->nh_oif ||
- nh->nh_gw != onh->nh_gw ||
- nh->nh_scope != onh->nh_scope ||
+ if (nh->fib_nh_oif != onh->fib_nh_oif ||
+ nh->fib_nh_gw4 != onh->fib_nh_gw4 ||
+ nh->fib_nh_scope != onh->fib_nh_scope ||
#ifdef CONFIG_IP_ROUTE_MULTIPATH
- nh->nh_weight != onh->nh_weight ||
+ nh->fib_nh_weight != onh->fib_nh_weight ||
#endif
#ifdef CONFIG_IP_ROUTE_CLASSID
nh->nh_tclassid != onh->nh_tclassid ||
#endif
- lwtunnel_cmp_encap(nh->nh_lwtstate, onh->nh_lwtstate) ||
- ((nh->nh_flags ^ onh->nh_flags) & ~RTNH_COMPARE_MASK))
+ lwtunnel_cmp_encap(nh->fib_nh_lws, onh->fib_nh_lws) ||
+ ((nh->fib_nh_flags ^ onh->fib_nh_flags) & ~RTNH_COMPARE_MASK))
return -1;
onh++;
} endfor_nexthops(fi);
@@ -303,7 +303,7 @@ static inline unsigned int fib_info_hashfn(const struct fib_info *fi)
val ^= (__force u32)fi->fib_prefsrc;
val ^= fi->fib_priority;
for_nexthops(fi) {
- val ^= fib_devindex_hashfn(nh->nh_oif);
+ val ^= fib_devindex_hashfn(nh->fib_nh_oif);
} endfor_nexthops(fi)
return (val ^ (val >> 7) ^ (val >> 12)) & mask;
@@ -352,9 +352,9 @@ int ip_fib_check_default(__be32 gw, struct net_device *dev)
hash = fib_devindex_hashfn(dev->ifindex);
head = &fib_info_devhash[hash];
hlist_for_each_entry(nh, head, nh_hash) {
- if (nh->nh_dev == dev &&
- nh->nh_gw == gw &&
- !(nh->nh_flags & RTNH_F_DEAD)) {
+ if (nh->fib_nh_dev == dev &&
+ nh->fib_nh_gw4 == gw &&
+ !(nh->fib_nh_flags & RTNH_F_DEAD)) {
spin_unlock(&fib_info_lock);
return 0;
}
@@ -389,10 +389,10 @@ static inline size_t fib_nlmsg_size(struct fib_info *fi)
/* grab encap info */
for_nexthops(fi) {
- if (nh->nh_lwtstate) {
+ if (nh->fib_nh_lws) {
/* RTA_ENCAP_TYPE */
nh_encapsize += lwtunnel_get_encap_size(
- nh->nh_lwtstate);
+ nh->fib_nh_lws);
/* RTA_ENCAP */
nh_encapsize += nla_total_size(2);
}
@@ -443,7 +443,7 @@ static int fib_detect_death(struct fib_info *fi, int order,
struct neighbour *n;
int state = NUD_NONE;
- n = neigh_lookup(&arp_tbl, &fi->fib_nh[0].nh_gw, fi->fib_dev);
+ n = neigh_lookup(&arp_tbl, &fi->fib_nh[0].fib_nh_gw4, fi->fib_dev);
if (n) {
state = n->nud_state;
neigh_release(n);
@@ -486,12 +486,12 @@ int fib_nh_init(struct net *net, struct fib_nh *nh,
if (err)
goto failure;
- nh->nh_lwtstate = lwtstate_get(lwtstate);
+ nh->fib_nh_lws = lwtstate_get(lwtstate);
}
- nh->nh_oif = cfg->fc_oif;
- nh->nh_gw = cfg->fc_gw;
- nh->nh_flags = cfg->fc_flags;
+ nh->fib_nh_oif = cfg->fc_oif;
+ nh->fib_nh_gw4 = cfg->fc_gw;
+ nh->fib_nh_flags = cfg->fc_flags;
#ifdef CONFIG_IP_ROUTE_CLASSID
nh->nh_tclassid = cfg->fc_flow;
@@ -499,7 +499,7 @@ int fib_nh_init(struct net *net, struct fib_nh *nh,
net->ipv4.fib_num_tclassid_users++;
#endif
#ifdef CONFIG_IP_ROUTE_MULTIPATH
- nh->nh_weight = nh_weight;
+ nh->fib_nh_weight = nh_weight;
#endif
err = 0;
@@ -584,12 +584,12 @@ static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh,
} endfor_nexthops(fi);
ret = -EINVAL;
- if (cfg->fc_oif && fi->fib_nh->nh_oif != cfg->fc_oif) {
+ if (cfg->fc_oif && fi->fib_nh->fib_nh_oif != cfg->fc_oif) {
NL_SET_ERR_MSG(extack,
"Nexthop device index does not match RTA_OIF");
goto errout;
}
- if (cfg->fc_gw && fi->fib_nh->nh_gw != cfg->fc_gw) {
+ if (cfg->fc_gw && fi->fib_nh->fib_nh_gw4 != cfg->fc_gw) {
NL_SET_ERR_MSG(extack,
"Nexthop gateway does not match RTA_GATEWAY");
goto errout;
@@ -616,32 +616,32 @@ static void fib_rebalance(struct fib_info *fi)
total = 0;
for_nexthops(fi) {
- if (nh->nh_flags & RTNH_F_DEAD)
+ if (nh->fib_nh_flags & RTNH_F_DEAD)
continue;
- if (ip_ignore_linkdown(nh->nh_dev) &&
- nh->nh_flags & RTNH_F_LINKDOWN)
+ if (ip_ignore_linkdown(nh->fib_nh_dev) &&
+ nh->fib_nh_flags & RTNH_F_LINKDOWN)
continue;
- total += nh->nh_weight;
+ total += nh->fib_nh_weight;
} endfor_nexthops(fi);
w = 0;
change_nexthops(fi) {
int upper_bound;
- if (nexthop_nh->nh_flags & RTNH_F_DEAD) {
+ if (nexthop_nh->fib_nh_flags & RTNH_F_DEAD) {
upper_bound = -1;
- } else if (ip_ignore_linkdown(nexthop_nh->nh_dev) &&
- nexthop_nh->nh_flags & RTNH_F_LINKDOWN) {
+ } else if (ip_ignore_linkdown(nexthop_nh->fib_nh_dev) &&
+ nexthop_nh->fib_nh_flags & RTNH_F_LINKDOWN) {
upper_bound = -1;
} else {
- w += nexthop_nh->nh_weight;
+ w += nexthop_nh->fib_nh_weight;
upper_bound = DIV_ROUND_CLOSEST_ULL((u64)w << 31,
total) - 1;
}
- atomic_set(&nexthop_nh->nh_upper_bound, upper_bound);
+ atomic_set(&nexthop_nh->fib_nh_upper_bound, upper_bound);
} endfor_nexthops(fi);
}
#else /* CONFIG_IP_ROUTE_MULTIPATH */
@@ -674,7 +674,7 @@ static int fib_encap_match(u16 encap_type,
ret = lwtunnel_build_state(encap_type, encap, AF_INET,
cfg, &lwtstate, extack);
if (!ret) {
- result = lwtunnel_cmp_encap(lwtstate, nh->nh_lwtstate);
+ result = lwtunnel_cmp_encap(lwtstate, nh->fib_nh_lws);
lwtstate_free(lwtstate);
}
@@ -703,8 +703,8 @@ int fib_nh_match(struct fib_config *cfg, struct fib_info *fi,
cfg->fc_flow != fi->fib_nh->nh_tclassid)
return 1;
#endif
- if ((!cfg->fc_oif || cfg->fc_oif == fi->fib_nh->nh_oif) &&
- (!cfg->fc_gw || cfg->fc_gw == fi->fib_nh->nh_gw))
+ if ((!cfg->fc_oif || cfg->fc_oif == fi->fib_nh->fib_nh_oif) &&
+ (!cfg->fc_gw || cfg->fc_gw == fi->fib_nh->fib_nh_gw4))
return 0;
return 1;
}
@@ -722,7 +722,7 @@ int fib_nh_match(struct fib_config *cfg, struct fib_info *fi,
if (!rtnh_ok(rtnh, remaining))
return -EINVAL;
- if (rtnh->rtnh_ifindex && rtnh->rtnh_ifindex != nh->nh_oif)
+ if (rtnh->rtnh_ifindex && rtnh->rtnh_ifindex != nh->fib_nh_oif)
return 1;
attrlen = rtnh_attrlen(rtnh);
@@ -730,7 +730,7 @@ int fib_nh_match(struct fib_config *cfg, struct fib_info *fi,
struct nlattr *nla, *attrs = rtnh_attrs(rtnh);
nla = nla_find(attrs, attrlen, RTA_GATEWAY);
- if (nla && nla_get_in_addr(nla) != nh->nh_gw)
+ if (nla && nla_get_in_addr(nla) != nh->fib_nh_gw4)
return 1;
#ifdef CONFIG_IP_ROUTE_CLASSID
nla = nla_find(attrs, attrlen, RTA_FLOW);
@@ -837,10 +837,10 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_nh *nh,
struct net_device *dev;
net = cfg->fc_nlinfo.nl_net;
- if (nh->nh_gw) {
+ if (nh->fib_nh_gw4) {
struct fib_result res;
- if (nh->nh_flags & RTNH_F_ONLINK) {
+ if (nh->fib_nh_flags & RTNH_F_ONLINK) {
unsigned int addr_type;
if (cfg->fc_scope >= RT_SCOPE_LINK) {
@@ -848,7 +848,7 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_nh *nh,
"Nexthop has invalid scope");
return -EINVAL;
}
- dev = __dev_get_by_index(net, nh->nh_oif);
+ dev = __dev_get_by_index(net, nh->fib_nh_oif);
if (!dev) {
NL_SET_ERR_MSG(extack, "Nexthop device required for onlink");
return -ENODEV;
@@ -858,26 +858,27 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_nh *nh,
"Nexthop device is not up");
return -ENETDOWN;
}
- addr_type = inet_addr_type_dev_table(net, dev, nh->nh_gw);
+ addr_type = inet_addr_type_dev_table(net, dev,
+ nh->fib_nh_gw4);
if (addr_type != RTN_UNICAST) {
NL_SET_ERR_MSG(extack,
"Nexthop has invalid gateway");
return -EINVAL;
}
if (!netif_carrier_ok(dev))
- nh->nh_flags |= RTNH_F_LINKDOWN;
- nh->nh_dev = dev;
+ nh->fib_nh_flags |= RTNH_F_LINKDOWN;
+ nh->fib_nh_dev = dev;
dev_hold(dev);
- nh->nh_scope = RT_SCOPE_LINK;
+ nh->fib_nh_scope = RT_SCOPE_LINK;
return 0;
}
rcu_read_lock();
{
struct fib_table *tbl = NULL;
struct flowi4 fl4 = {
- .daddr = nh->nh_gw,
+ .daddr = nh->fib_nh_gw4,
.flowi4_scope = cfg->fc_scope + 1,
- .flowi4_oif = nh->nh_oif,
+ .flowi4_oif = nh->fib_nh_oif,
.flowi4_iif = LOOPBACK_IFINDEX,
};
@@ -914,9 +915,9 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_nh *nh,
NL_SET_ERR_MSG(extack, "Nexthop has invalid gateway");
goto out;
}
- nh->nh_scope = res.scope;
- nh->nh_oif = FIB_RES_OIF(res);
- nh->nh_dev = dev = FIB_RES_DEV(res);
+ nh->fib_nh_scope = res.scope;
+ nh->fib_nh_oif = FIB_RES_OIF(res);
+ nh->fib_nh_dev = dev = FIB_RES_DEV(res);
if (!dev) {
NL_SET_ERR_MSG(extack,
"No egress device for nexthop gateway");
@@ -924,19 +925,19 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_nh *nh,
}
dev_hold(dev);
if (!netif_carrier_ok(dev))
- nh->nh_flags |= RTNH_F_LINKDOWN;
+ nh->fib_nh_flags |= RTNH_F_LINKDOWN;
err = (dev->flags & IFF_UP) ? 0 : -ENETDOWN;
} else {
struct in_device *in_dev;
- if (nh->nh_flags & (RTNH_F_PERVASIVE | RTNH_F_ONLINK)) {
+ if (nh->fib_nh_flags & (RTNH_F_PERVASIVE | RTNH_F_ONLINK)) {
NL_SET_ERR_MSG(extack,
"Invalid flags for nexthop - PERVASIVE and ONLINK can not be set");
return -EINVAL;
}
rcu_read_lock();
err = -ENODEV;
- in_dev = inetdev_by_index(net, nh->nh_oif);
+ in_dev = inetdev_by_index(net, nh->fib_nh_oif);
if (!in_dev)
goto out;
err = -ENETDOWN;
@@ -944,11 +945,11 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_nh *nh,
NL_SET_ERR_MSG(extack, "Device for nexthop is not up");
goto out;
}
- nh->nh_dev = in_dev->dev;
- dev_hold(nh->nh_dev);
- nh->nh_scope = RT_SCOPE_HOST;
- if (!netif_carrier_ok(nh->nh_dev))
- nh->nh_flags |= RTNH_F_LINKDOWN;
+ nh->fib_nh_dev = in_dev->dev;
+ dev_hold(nh->fib_nh_dev);
+ nh->fib_nh_scope = RT_SCOPE_HOST;
+ if (!netif_carrier_ok(nh->fib_nh_dev))
+ nh->fib_nh_flags |= RTNH_F_LINKDOWN;
err = 0;
}
out:
@@ -1040,8 +1041,8 @@ static void fib_info_hash_move(struct hlist_head *new_info_hash,
__be32 fib_info_update_nh_saddr(struct net *net, struct fib_nh *nh)
{
- nh->nh_saddr = inet_select_addr(nh->nh_dev,
- nh->nh_gw,
+ nh->nh_saddr = inet_select_addr(nh->fib_nh_dev,
+ nh->fib_nh_gw4,
nh->nh_parent->fib_scope);
nh->nh_saddr_genid = atomic_read(&net->ipv4.dev_addr_genid);
@@ -1195,15 +1196,15 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
"Route with host scope can not have multiple nexthops");
goto err_inval;
}
- if (nh->nh_gw) {
+ if (nh->fib_nh_gw4) {
NL_SET_ERR_MSG(extack,
"Route with host scope can not have a gateway");
goto err_inval;
}
- nh->nh_scope = RT_SCOPE_NOWHERE;
- nh->nh_dev = dev_get_by_index(net, fi->fib_nh->nh_oif);
+ nh->fib_nh_scope = RT_SCOPE_NOWHERE;
+ nh->fib_nh_dev = dev_get_by_index(net, fi->fib_nh->fib_nh_oif);
err = -ENODEV;
- if (!nh->nh_dev)
+ if (!nh->fib_nh_dev)
goto failure;
} else {
int linkdown = 0;
@@ -1212,7 +1213,7 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
err = fib_check_nh(cfg, nexthop_nh, extack);
if (err != 0)
goto failure;
- if (nexthop_nh->nh_flags & RTNH_F_LINKDOWN)
+ if (nexthop_nh->fib_nh_flags & RTNH_F_LINKDOWN)
linkdown++;
} endfor_nexthops(fi)
if (linkdown == fi->fib_nhs)
@@ -1254,9 +1255,9 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
struct hlist_head *head;
unsigned int hash;
- if (!nexthop_nh->nh_dev)
+ if (!nexthop_nh->fib_nh_dev)
continue;
- hash = fib_devindex_hashfn(nexthop_nh->nh_dev->ifindex);
+ hash = fib_devindex_hashfn(nexthop_nh->fib_nh_dev->ifindex);
head = &fib_info_devhash[hash];
hlist_add_head(&nexthop_nh->nh_hash, head);
} endfor_nexthops(fi)
@@ -1315,27 +1316,27 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event,
nla_put_in_addr(skb, RTA_PREFSRC, fi->fib_prefsrc))
goto nla_put_failure;
if (fi->fib_nhs == 1) {
- if (fi->fib_nh->nh_gw &&
- nla_put_in_addr(skb, RTA_GATEWAY, fi->fib_nh->nh_gw))
+ if (fi->fib_nh->fib_nh_gw4 &&
+ nla_put_in_addr(skb, RTA_GATEWAY, fi->fib_nh->fib_nh_gw4))
goto nla_put_failure;
- if (fi->fib_nh->nh_oif &&
- nla_put_u32(skb, RTA_OIF, fi->fib_nh->nh_oif))
+ if (fi->fib_nh->fib_nh_oif &&
+ nla_put_u32(skb, RTA_OIF, fi->fib_nh->fib_nh_oif))
goto nla_put_failure;
- if (fi->fib_nh->nh_flags & RTNH_F_LINKDOWN) {
+ if (fi->fib_nh->fib_nh_flags & RTNH_F_LINKDOWN) {
rcu_read_lock();
- if (ip_ignore_linkdown(fi->fib_nh->nh_dev))
+ if (ip_ignore_linkdown(fi->fib_nh->fib_nh_dev))
rtm->rtm_flags |= RTNH_F_DEAD;
rcu_read_unlock();
}
- if (fi->fib_nh->nh_flags & RTNH_F_OFFLOAD)
+ if (fi->fib_nh->fib_nh_flags & RTNH_F_OFFLOAD)
rtm->rtm_flags |= RTNH_F_OFFLOAD;
#ifdef CONFIG_IP_ROUTE_CLASSID
if (fi->fib_nh[0].nh_tclassid &&
nla_put_u32(skb, RTA_FLOW, fi->fib_nh[0].nh_tclassid))
goto nla_put_failure;
#endif
- if (fi->fib_nh->nh_lwtstate &&
- lwtunnel_fill_encap(skb, fi->fib_nh->nh_lwtstate) < 0)
+ if (fi->fib_nh->fib_nh_lws &&
+ lwtunnel_fill_encap(skb, fi->fib_nh->fib_nh_lws) < 0)
goto nla_put_failure;
}
#ifdef CONFIG_IP_ROUTE_MULTIPATH
@@ -1352,26 +1353,26 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event,
if (!rtnh)
goto nla_put_failure;
- rtnh->rtnh_flags = nh->nh_flags & 0xFF;
- if (nh->nh_flags & RTNH_F_LINKDOWN) {
+ rtnh->rtnh_flags = nh->fib_nh_flags & 0xFF;
+ if (nh->fib_nh_flags & RTNH_F_LINKDOWN) {
rcu_read_lock();
- if (ip_ignore_linkdown(nh->nh_dev))
+ if (ip_ignore_linkdown(nh->fib_nh_dev))
rtnh->rtnh_flags |= RTNH_F_DEAD;
rcu_read_unlock();
}
- rtnh->rtnh_hops = nh->nh_weight - 1;
- rtnh->rtnh_ifindex = nh->nh_oif;
+ rtnh->rtnh_hops = nh->fib_nh_weight - 1;
+ rtnh->rtnh_ifindex = nh->fib_nh_oif;
- if (nh->nh_gw &&
- nla_put_in_addr(skb, RTA_GATEWAY, nh->nh_gw))
+ if (nh->fib_nh_gw4 &&
+ nla_put_in_addr(skb, RTA_GATEWAY, nh->fib_nh_gw4))
goto nla_put_failure;
#ifdef CONFIG_IP_ROUTE_CLASSID
if (nh->nh_tclassid &&
nla_put_u32(skb, RTA_FLOW, nh->nh_tclassid))
goto nla_put_failure;
#endif
- if (nh->nh_lwtstate &&
- lwtunnel_fill_encap(skb, nh->nh_lwtstate) < 0)
+ if (nh->fib_nh_lws &&
+ lwtunnel_fill_encap(skb, nh->fib_nh_lws) < 0)
goto nla_put_failure;
/* length of rtnetlink header + attributes */
@@ -1419,26 +1420,26 @@ int fib_sync_down_addr(struct net_device *dev, __be32 local)
return ret;
}
-static int call_fib_nh_notifiers(struct fib_nh *fib_nh,
+static int call_fib_nh_notifiers(struct fib_nh *nh,
enum fib_event_type event_type)
{
- bool ignore_link_down = ip_ignore_linkdown(fib_nh->nh_dev);
+ bool ignore_link_down = ip_ignore_linkdown(nh->fib_nh_dev);
struct fib_nh_notifier_info info = {
- .fib_nh = fib_nh,
+ .fib_nh = nh,
};
switch (event_type) {
case FIB_EVENT_NH_ADD:
- if (fib_nh->nh_flags & RTNH_F_DEAD)
+ if (nh->fib_nh_flags & RTNH_F_DEAD)
break;
- if (ignore_link_down && fib_nh->nh_flags & RTNH_F_LINKDOWN)
+ if (ignore_link_down && nh->fib_nh_flags & RTNH_F_LINKDOWN)
break;
- return call_fib4_notifiers(dev_net(fib_nh->nh_dev), event_type,
+ return call_fib4_notifiers(dev_net(nh->fib_nh_dev), event_type,
&info.info);
case FIB_EVENT_NH_DEL:
- if ((ignore_link_down && fib_nh->nh_flags & RTNH_F_LINKDOWN) ||
- (fib_nh->nh_flags & RTNH_F_DEAD))
- return call_fib4_notifiers(dev_net(fib_nh->nh_dev),
+ if ((ignore_link_down && nh->fib_nh_flags & RTNH_F_LINKDOWN) ||
+ (nh->fib_nh_flags & RTNH_F_DEAD))
+ return call_fib4_notifiers(dev_net(nh->fib_nh_dev),
event_type, &info.info);
default:
break;
@@ -1492,7 +1493,7 @@ void fib_sync_mtu(struct net_device *dev, u32 orig_mtu)
struct fib_nh *nh;
hlist_for_each_entry(nh, head, nh_hash) {
- if (nh->nh_dev == dev)
+ if (nh->fib_nh_dev == dev)
nh_update_mtu(nh, dev->mtu, orig_mtu);
}
}
@@ -1520,22 +1521,22 @@ int fib_sync_down_dev(struct net_device *dev, unsigned long event, bool force)
int dead;
BUG_ON(!fi->fib_nhs);
- if (nh->nh_dev != dev || fi == prev_fi)
+ if (nh->fib_nh_dev != dev || fi == prev_fi)
continue;
prev_fi = fi;
dead = 0;
change_nexthops(fi) {
- if (nexthop_nh->nh_flags & RTNH_F_DEAD)
+ if (nexthop_nh->fib_nh_flags & RTNH_F_DEAD)
dead++;
- else if (nexthop_nh->nh_dev == dev &&
- nexthop_nh->nh_scope != scope) {
+ else if (nexthop_nh->fib_nh_dev == dev &&
+ nexthop_nh->fib_nh_scope != scope) {
switch (event) {
case NETDEV_DOWN:
case NETDEV_UNREGISTER:
- nexthop_nh->nh_flags |= RTNH_F_DEAD;
+ nexthop_nh->fib_nh_flags |= RTNH_F_DEAD;
/* fall through */
case NETDEV_CHANGE:
- nexthop_nh->nh_flags |= RTNH_F_LINKDOWN;
+ nexthop_nh->fib_nh_flags |= RTNH_F_LINKDOWN;
break;
}
call_fib_nh_notifiers(nexthop_nh,
@@ -1544,7 +1545,7 @@ int fib_sync_down_dev(struct net_device *dev, unsigned long event, bool force)
}
#ifdef CONFIG_IP_ROUTE_MULTIPATH
if (event == NETDEV_UNREGISTER &&
- nexthop_nh->nh_dev == dev) {
+ nexthop_nh->fib_nh_dev == dev) {
dead = fi->fib_nhs;
break;
}
@@ -1604,8 +1605,8 @@ static void fib_select_default(const struct flowi4 *flp, struct fib_result *res)
if (next_fi->fib_scope != res->scope ||
fa->fa_type != RTN_UNICAST)
continue;
- if (!next_fi->fib_nh[0].nh_gw ||
- next_fi->fib_nh[0].nh_scope != RT_SCOPE_LINK)
+ if (!next_fi->fib_nh[0].fib_nh_gw4 ||
+ next_fi->fib_nh[0].fib_nh_scope != RT_SCOPE_LINK)
continue;
fib_alias_accessed(fa);
@@ -1676,24 +1677,24 @@ int fib_sync_up(struct net_device *dev, unsigned int nh_flags)
int alive;
BUG_ON(!fi->fib_nhs);
- if (nh->nh_dev != dev || fi == prev_fi)
+ if (nh->fib_nh_dev != dev || fi == prev_fi)
continue;
prev_fi = fi;
alive = 0;
change_nexthops(fi) {
- if (!(nexthop_nh->nh_flags & nh_flags)) {
+ if (!(nexthop_nh->fib_nh_flags & nh_flags)) {
alive++;
continue;
}
- if (!nexthop_nh->nh_dev ||
- !(nexthop_nh->nh_dev->flags & IFF_UP))
+ if (!nexthop_nh->fib_nh_dev ||
+ !(nexthop_nh->fib_nh_dev->flags & IFF_UP))
continue;
- if (nexthop_nh->nh_dev != dev ||
+ if (nexthop_nh->fib_nh_dev != dev ||
!__in_dev_get_rtnl(dev))
continue;
alive++;
- nexthop_nh->nh_flags &= ~nh_flags;
+ nexthop_nh->fib_nh_flags &= ~nh_flags;
call_fib_nh_notifiers(nexthop_nh, FIB_EVENT_NH_ADD);
} endfor_nexthops(fi)
@@ -1713,13 +1714,13 @@ static bool fib_good_nh(const struct fib_nh *nh)
{
int state = NUD_REACHABLE;
- if (nh->nh_scope == RT_SCOPE_LINK) {
+ if (nh->fib_nh_scope == RT_SCOPE_LINK) {
struct neighbour *n;
rcu_read_lock_bh();
- n = __ipv4_neigh_lookup_noref(nh->nh_dev,
- (__force u32)nh->nh_gw);
+ n = __ipv4_neigh_lookup_noref(nh->fib_nh_dev,
+ (__force u32)nh->fib_nh_gw4);
if (n)
state = n->nud_state;
@@ -1745,7 +1746,7 @@ void fib_select_multipath(struct fib_result *res, int hash)
}
}
- if (hash > atomic_read(&nh->nh_upper_bound))
+ if (hash > atomic_read(&nh->fib_nh_upper_bound))
continue;
res->nh_sel = nhsel;
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 656d3d19f112..1e3b492690f9 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1472,15 +1472,15 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi4 *flp,
for (nhsel = 0; nhsel < fi->fib_nhs; nhsel++) {
const struct fib_nh *nh = &fi->fib_nh[nhsel];
- if (nh->nh_flags & RTNH_F_DEAD)
+ if (nh->fib_nh_flags & RTNH_F_DEAD)
continue;
- if (ip_ignore_linkdown(nh->nh_dev) &&
- nh->nh_flags & RTNH_F_LINKDOWN &&
+ if (ip_ignore_linkdown(nh->fib_nh_dev) &&
+ nh->fib_nh_flags & RTNH_F_LINKDOWN &&
!(fib_flags & FIB_LOOKUP_IGNORE_LINKSTATE))
continue;
if (!(flp->flowi4_flags & FLOWI_FLAG_SKIP_NH_OIF)) {
if (flp->flowi4_oif &&
- flp->flowi4_oif != nh->nh_oif)
+ flp->flowi4_oif != nh->fib_nh_oif)
continue;
}
@@ -2651,7 +2651,7 @@ static unsigned int fib_flag_trans(int type, __be32 mask, const struct fib_info
if (type == RTN_UNREACHABLE || type == RTN_PROHIBIT)
flags = RTF_REJECT;
- if (fi && fi->fib_nh->nh_gw)
+ if (fi && fi->fib_nh->fib_nh_gw4)
flags |= RTF_GATEWAY;
if (mask == htonl(0xFFFFFFFF))
flags |= RTF_HOST;
@@ -2702,7 +2702,7 @@ static int fib_route_seq_show(struct seq_file *seq, void *v)
"%d\t%08X\t%d\t%u\t%u",
fi->fib_dev ? fi->fib_dev->name : "*",
prefix,
- fi->fib_nh->nh_gw, flags, 0, 0,
+ fi->fib_nh->fib_nh_gw4, flags, 0, 0,
fi->fib_priority,
mask,
(fi->fib_advmss ?
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 14c7fdacaa72..365f0d035dfb 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -642,7 +642,7 @@ static void update_or_create_fnhe(struct fib_nh *nh, __be32 daddr, __be32 gw,
unsigned int i;
int depth;
- genid = fnhe_genid(dev_net(nh->nh_dev));
+ genid = fnhe_genid(dev_net(nh->fib_nh_dev));
hval = fnhe_hashfun(daddr);
spin_lock_bh(&fnhe_lock);
@@ -1354,7 +1354,7 @@ u32 ip_mtu_from_fib_result(struct fib_result *res, __be32 daddr)
{
struct fib_info *fi = res->fi;
struct fib_nh *nh = &fi->fib_nh[res->nh_sel];
- struct net_device *dev = nh->nh_dev;
+ struct net_device *dev = nh->fib_nh_dev;
u32 mtu = 0;
if (dev_net(dev)->ipv4.sysctl_ip_fwd_use_pmtu ||
@@ -1372,7 +1372,7 @@ u32 ip_mtu_from_fib_result(struct fib_result *res, __be32 daddr)
if (likely(!mtu))
mtu = min(READ_ONCE(dev->mtu), IP_MAX_MTU);
- return mtu - lwtunnel_headroom(nh->nh_lwtstate, mtu);
+ return mtu - lwtunnel_headroom(nh->fib_nh_lws, mtu);
}
static bool rt_bind_exception(struct rtable *rt, struct fib_nh_exception *fnhe,
@@ -1529,8 +1529,8 @@ static void rt_set_nexthop(struct rtable *rt, __be32 daddr,
if (fi) {
struct fib_nh *nh = &FIB_RES_NH(*res);
- if (nh->nh_gw && nh->nh_scope == RT_SCOPE_LINK) {
- rt->rt_gateway = nh->nh_gw;
+ if (nh->fib_nh_gw4 && nh->fib_nh_scope == RT_SCOPE_LINK) {
+ rt->rt_gateway = nh->fib_nh_gw4;
rt->rt_uses_gateway = 1;
}
ip_dst_init_metrics(&rt->dst, fi->fib_metrics);
@@ -1538,7 +1538,7 @@ static void rt_set_nexthop(struct rtable *rt, __be32 daddr,
#ifdef CONFIG_IP_ROUTE_CLASSID
rt->dst.tclassid = nh->nh_tclassid;
#endif
- rt->dst.lwtstate = lwtstate_get(nh->nh_lwtstate);
+ rt->dst.lwtstate = lwtstate_get(nh->fib_nh_lws);
if (unlikely(fnhe))
cached = rt_bind_exception(rt, fnhe, daddr, do_cache);
else if (do_cache)
@@ -2073,7 +2073,7 @@ out: return err;
if (do_cache) {
struct fib_nh *nh = &FIB_RES_NH(*res);
- rth->dst.lwtstate = lwtstate_get(nh->nh_lwtstate);
+ rth->dst.lwtstate = lwtstate_get(nh->fib_nh_lws);
if (lwtunnel_input_redirect(rth->dst.lwtstate)) {
WARN_ON(rth->dst.input == lwtunnel_input);
rth->dst.lwtstate->orig_input = rth->dst.input;
@@ -2262,8 +2262,8 @@ static struct rtable *__mkroute_output(const struct fib_result *res,
} else {
if (unlikely(fl4->flowi4_flags &
FLOWI_FLAG_KNOWN_NH &&
- !(nh->nh_gw &&
- nh->nh_scope == RT_SCOPE_LINK))) {
+ !(nh->fib_nh_gw4 &&
+ nh->fib_nh_scope == RT_SCOPE_LINK))) {
do_cache = false;
goto add;
}
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 11/13] ipv6: Rename fib6_nh entries
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
` (9 preceding siblings ...)
2019-03-27 3:29 ` [PATCH net-next 10/13] ipv4: Rename fib_nh entries David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 3:29 ` [PATCH net-next 12/13] net: Add fib_nh_common and update fib_nh and fib6_nh David Ahern
2019-03-27 3:29 ` [PATCH net-next 13/13] net: Use common nexthop init and release helpers David Ahern
12 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
Rename fib6_nh entries that will be moved to a fib_nh_common struct.
Specifically, the device, gateway, flags, and lwtstate are common
with all nexthop definitions. In some places new temporary variables
are declared or local variables renamed to maintain line lengths.
Rename only; no functional change intended.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
.../net/ethernet/mellanox/mlxsw/spectrum_router.c | 34 ++--
include/net/ip6_fib.h | 16 +-
include/net/ip6_route.h | 8 +-
include/trace/events/fib6.h | 6 +-
net/core/filter.c | 6 +-
net/ipv6/addrconf.c | 2 +-
net/ipv6/ip6_fib.c | 4 +-
net/ipv6/ndisc.c | 8 +-
net/ipv6/route.c | 175 +++++++++++----------
9 files changed, 135 insertions(+), 124 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 06e1a98fcd6a..298aa8d450f2 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -2873,12 +2873,13 @@ mlxsw_sp_nexthop6_group_cmp(const struct mlxsw_sp_nexthop_group *nh_grp,
return false;
list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
+ struct fib6_nh *fib6_nh = &mlxsw_sp_rt6->rt->fib6_nh;
struct in6_addr *gw;
int ifindex, weight;
- ifindex = mlxsw_sp_rt6->rt->fib6_nh.nh_dev->ifindex;
- weight = mlxsw_sp_rt6->rt->fib6_nh.nh_weight;
- gw = &mlxsw_sp_rt6->rt->fib6_nh.nh_gw;
+ ifindex = fib6_nh->fib_nh_dev->ifindex;
+ weight = fib6_nh->fib_nh_weight;
+ gw = &fib6_nh->fib_nh_gw6;
if (!mlxsw_sp_nexthop6_group_has_nexthop(nh_grp, gw, ifindex,
weight))
return false;
@@ -2944,7 +2945,7 @@ mlxsw_sp_nexthop6_group_hash(struct mlxsw_sp_fib6_entry *fib6_entry, u32 seed)
struct net_device *dev;
list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
- dev = mlxsw_sp_rt6->rt->fib6_nh.nh_dev;
+ dev = mlxsw_sp_rt6->rt->fib6_nh.fib_nh_dev;
val ^= dev->ifindex;
}
@@ -3946,9 +3947,9 @@ mlxsw_sp_rt6_nexthop(struct mlxsw_sp_nexthop_group *nh_grp,
struct mlxsw_sp_nexthop *nh = &nh_grp->nexthops[i];
struct fib6_info *rt = mlxsw_sp_rt6->rt;
- if (nh->rif && nh->rif->dev == rt->fib6_nh.nh_dev &&
+ if (nh->rif && nh->rif->dev == rt->fib6_nh.fib_nh_dev &&
ipv6_addr_equal((const struct in6_addr *) &nh->gw_addr,
- &rt->fib6_nh.nh_gw))
+ &rt->fib6_nh.fib_nh_gw6))
return nh;
continue;
}
@@ -4008,19 +4009,20 @@ mlxsw_sp_fib6_entry_offload_set(struct mlxsw_sp_fib_entry *fib_entry)
if (fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_LOCAL ||
fib_entry->type == MLXSW_SP_FIB_ENTRY_TYPE_BLACKHOLE) {
list_first_entry(&fib6_entry->rt6_list, struct mlxsw_sp_rt6,
- list)->rt->fib6_nh.nh_flags |= RTNH_F_OFFLOAD;
+ list)->rt->fib6_nh.fib_nh_flags |= RTNH_F_OFFLOAD;
return;
}
list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
struct mlxsw_sp_nexthop_group *nh_grp = fib_entry->nh_group;
+ struct fib6_nh *fib6_nh = &mlxsw_sp_rt6->rt->fib6_nh;
struct mlxsw_sp_nexthop *nh;
nh = mlxsw_sp_rt6_nexthop(nh_grp, mlxsw_sp_rt6);
if (nh && nh->offloaded)
- mlxsw_sp_rt6->rt->fib6_nh.nh_flags |= RTNH_F_OFFLOAD;
+ fib6_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
else
- mlxsw_sp_rt6->rt->fib6_nh.nh_flags &= ~RTNH_F_OFFLOAD;
+ fib6_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
}
}
@@ -4035,7 +4037,7 @@ mlxsw_sp_fib6_entry_offload_unset(struct mlxsw_sp_fib_entry *fib_entry)
list_for_each_entry(mlxsw_sp_rt6, &fib6_entry->rt6_list, list) {
struct fib6_info *rt = mlxsw_sp_rt6->rt;
- rt->fib6_nh.nh_flags &= ~RTNH_F_OFFLOAD;
+ rt->fib6_nh.fib_nh_flags &= ~RTNH_F_OFFLOAD;
}
}
@@ -4972,8 +4974,8 @@ static bool mlxsw_sp_nexthop6_ipip_type(const struct mlxsw_sp *mlxsw_sp,
const struct fib6_info *rt,
enum mlxsw_sp_ipip_type *ret)
{
- return rt->fib6_nh.nh_dev &&
- mlxsw_sp_netdev_ipip_type(mlxsw_sp, rt->fib6_nh.nh_dev, ret);
+ return rt->fib6_nh.fib_nh_dev &&
+ mlxsw_sp_netdev_ipip_type(mlxsw_sp, rt->fib6_nh.fib_nh_dev, ret);
}
static int mlxsw_sp_nexthop6_type_init(struct mlxsw_sp *mlxsw_sp,
@@ -4983,7 +4985,7 @@ static int mlxsw_sp_nexthop6_type_init(struct mlxsw_sp *mlxsw_sp,
{
const struct mlxsw_sp_ipip_ops *ipip_ops;
struct mlxsw_sp_ipip_entry *ipip_entry;
- struct net_device *dev = rt->fib6_nh.nh_dev;
+ struct net_device *dev = rt->fib6_nh.fib_nh_dev;
struct mlxsw_sp_rif *rif;
int err;
@@ -5026,11 +5028,11 @@ static int mlxsw_sp_nexthop6_init(struct mlxsw_sp *mlxsw_sp,
struct mlxsw_sp_nexthop *nh,
const struct fib6_info *rt)
{
- struct net_device *dev = rt->fib6_nh.nh_dev;
+ struct net_device *dev = rt->fib6_nh.fib_nh_dev;
nh->nh_grp = nh_grp;
- nh->nh_weight = rt->fib6_nh.nh_weight;
- memcpy(&nh->gw_addr, &rt->fib6_nh.nh_gw, sizeof(nh->gw_addr));
+ nh->nh_weight = rt->fib6_nh.fib_nh_weight;
+ memcpy(&nh->gw_addr, &rt->fib6_nh.fib_nh_gw6, sizeof(nh->gw_addr));
mlxsw_sp_nexthop_counter_alloc(mlxsw_sp, nh);
list_add_tail(&nh->router_list_node, &mlxsw_sp->router->nexthop_list);
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index 3b04b318cf13..aff8570725c8 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -125,14 +125,14 @@ struct rt6_exception {
#define FIB6_MAX_DEPTH 5
struct fib6_nh {
- struct in6_addr nh_gw;
+ struct in6_addr fib_nh_gw6;
bool fib_nh_has_gw;
- struct net_device *nh_dev;
- struct lwtunnel_state *nh_lwtstate;
+ struct net_device *fib_nh_dev;
+ struct lwtunnel_state *fib_nh_lws;
- unsigned int nh_flags;
- atomic_t nh_upper_bound;
- int nh_weight;
+ unsigned int fib_nh_flags;
+ atomic_t fib_nh_upper_bound;
+ int fib_nh_weight;
};
struct fib6_info {
@@ -442,7 +442,7 @@ void rt6_get_prefsrc(const struct rt6_info *rt, struct in6_addr *addr)
static inline struct net_device *fib6_info_nh_dev(const struct fib6_info *f6i)
{
- return f6i->fib6_nh.nh_dev;
+ return f6i->fib6_nh.fib_nh_dev;
}
int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
@@ -453,7 +453,7 @@ void fib6_nh_release(struct fib6_nh *fib6_nh);
static inline
struct lwtunnel_state *fib6_info_nh_lwt(const struct fib6_info *f6i)
{
- return f6i->fib6_nh.nh_lwtstate;
+ return f6i->fib6_nh.fib_nh_lws;
}
void inet6_rt_notify(int event, struct fib6_info *rt, struct nl_info *info,
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 95cd8a2f6284..342180a7285c 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -274,9 +274,11 @@ static inline struct in6_addr *rt6_nexthop(struct rt6_info *rt,
static inline bool rt6_duplicate_nexthop(struct fib6_info *a, struct fib6_info *b)
{
- return a->fib6_nh.nh_dev == b->fib6_nh.nh_dev &&
- ipv6_addr_equal(&a->fib6_nh.nh_gw, &b->fib6_nh.nh_gw) &&
- !lwtunnel_cmp_encap(a->fib6_nh.nh_lwtstate, b->fib6_nh.nh_lwtstate);
+ struct fib6_nh *nha = &a->fib6_nh, *nhb = &b->fib6_nh;
+
+ return nha->fib_nh_dev == nhb->fib_nh_dev &&
+ ipv6_addr_equal(&nha->fib_nh_gw6, &nhb->fib_nh_gw6) &&
+ !lwtunnel_cmp_encap(nha->fib_nh_lws, nhb->fib_nh_lws);
}
static inline unsigned int ip6_dst_mtu_forward(const struct dst_entry *dst)
diff --git a/include/trace/events/fib6.h b/include/trace/events/fib6.h
index b088b54d699c..6d05ebdd669c 100644
--- a/include/trace/events/fib6.h
+++ b/include/trace/events/fib6.h
@@ -62,8 +62,8 @@ TRACE_EVENT(fib6_table_lookup,
__entry->dport = 0;
}
- if (f6i->fib6_nh.nh_dev) {
- __assign_str(name, f6i->fib6_nh.nh_dev);
+ if (f6i->fib6_nh.fib_nh_dev) {
+ __assign_str(name, f6i->fib6_nh.fib_nh_dev);
} else {
__assign_str(name, "-");
}
@@ -75,7 +75,7 @@ TRACE_EVENT(fib6_table_lookup,
} else if (f6i) {
in6 = (struct in6_addr *)__entry->gw;
- *in6 = f6i->fib6_nh.nh_gw;
+ *in6 = f6i->fib6_nh.fib_nh_gw6;
}
),
diff --git a/net/core/filter.c b/net/core/filter.c
index 7adcba70092e..49b407296d5a 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -4674,13 +4674,13 @@ static int bpf_ipv6_fib_lookup(struct net *net, struct bpf_fib_lookup *params,
return BPF_FIB_LKUP_RET_FRAG_NEEDED;
}
- if (f6i->fib6_nh.nh_lwtstate)
+ if (f6i->fib6_nh.fib_nh_lws)
return BPF_FIB_LKUP_RET_UNSUPP_LWT;
if (f6i->fib6_nh.fib_nh_has_gw)
- *dst = f6i->fib6_nh.nh_gw;
+ *dst = f6i->fib6_nh.fib_nh_gw6;
- dev = f6i->fib6_nh.nh_dev;
+ dev = f6i->fib6_nh.fib_nh_dev;
params->rt_metric = f6i->fib6_metric;
/* xdp and cls_bpf programs are run in RCU-bh so rcu_read_lock_bh is
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index c5ac08fc6cc9..2e8d1d2d8d3d 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -2419,7 +2419,7 @@ static struct fib6_info *addrconf_get_prefix_route(const struct in6_addr *pfx,
goto out;
for_each_fib6_node_rt_rcu(fn) {
- if (rt->fib6_nh.nh_dev->ifindex != dev->ifindex)
+ if (rt->fib6_nh.fib_nh_dev->ifindex != dev->ifindex)
continue;
if (no_gw && rt->fib6_nh.fib_nh_has_gw)
continue;
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 91ce84ecdb57..8c00609a1513 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -2306,12 +2306,12 @@ static int ipv6_route_seq_show(struct seq_file *seq, void *v)
#endif
if (rt->fib6_nh.fib_nh_has_gw) {
flags |= RTF_GATEWAY;
- seq_printf(seq, "%pi6", &rt->fib6_nh.nh_gw);
+ seq_printf(seq, "%pi6", &rt->fib6_nh.fib_nh_gw6);
} else {
seq_puts(seq, "00000000000000000000000000000000");
}
- dev = rt->fib6_nh.nh_dev;
+ dev = rt->fib6_nh.fib_nh_dev;
seq_printf(seq, " %08x %08x %08x %08x %8s\n",
rt->fib6_metric, atomic_read(&rt->fib6_ref), 0,
flags, dev ? dev->name : "");
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 659ecf4e4b3c..66c8b294e02b 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1276,8 +1276,8 @@ static void ndisc_router_discovery(struct sk_buff *skb)
rt = rt6_get_dflt_router(net, &ipv6_hdr(skb)->saddr, skb->dev);
if (rt) {
- neigh = ip6_neigh_lookup(&rt->fib6_nh.nh_gw,
- rt->fib6_nh.nh_dev, NULL,
+ neigh = ip6_neigh_lookup(&rt->fib6_nh.fib_nh_gw6,
+ rt->fib6_nh.fib_nh_dev, NULL,
&ipv6_hdr(skb)->saddr);
if (!neigh) {
ND_PRINTK(0, err,
@@ -1306,8 +1306,8 @@ static void ndisc_router_discovery(struct sk_buff *skb)
return;
}
- neigh = ip6_neigh_lookup(&rt->fib6_nh.nh_gw,
- rt->fib6_nh.nh_dev, NULL,
+ neigh = ip6_neigh_lookup(&rt->fib6_nh.fib_nh_gw6,
+ rt->fib6_nh.fib_nh_dev, NULL,
&ipv6_hdr(skb)->saddr);
if (!neigh) {
ND_PRINTK(0, err,
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 79fd43bd9e4e..b9218e3f0522 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -441,14 +441,14 @@ struct fib6_info *fib6_multipath_select(const struct net *net,
if (!fl6->mp_hash)
fl6->mp_hash = rt6_multipath_hash(net, fl6, skb, NULL);
- if (fl6->mp_hash <= atomic_read(&match->fib6_nh.nh_upper_bound))
+ if (fl6->mp_hash <= atomic_read(&match->fib6_nh.fib_nh_upper_bound))
return match;
list_for_each_entry_safe(sibling, next_sibling, &match->fib6_siblings,
fib6_siblings) {
int nh_upper_bound;
- nh_upper_bound = atomic_read(&sibling->fib6_nh.nh_upper_bound);
+ nh_upper_bound = atomic_read(&sibling->fib6_nh.fib_nh_upper_bound);
if (fl6->mp_hash > nh_upper_bound)
continue;
if (rt6_score_route(sibling, oif, strict) < 0)
@@ -473,13 +473,13 @@ static inline struct fib6_info *rt6_device_match(struct net *net,
struct fib6_info *sprt;
if (!oif && ipv6_addr_any(saddr) &&
- !(rt->fib6_nh.nh_flags & RTNH_F_DEAD))
+ !(rt->fib6_nh.fib_nh_flags & RTNH_F_DEAD))
return rt;
for (sprt = rt; sprt; sprt = rcu_dereference(sprt->fib6_next)) {
- const struct net_device *dev = sprt->fib6_nh.nh_dev;
+ const struct net_device *dev = sprt->fib6_nh.fib_nh_dev;
- if (sprt->fib6_nh.nh_flags & RTNH_F_DEAD)
+ if (sprt->fib6_nh.fib_nh_flags & RTNH_F_DEAD)
continue;
if (oif) {
@@ -495,7 +495,7 @@ static inline struct fib6_info *rt6_device_match(struct net *net,
if (oif && flags & RT6_LOOKUP_F_IFACE)
return net->ipv6.fib6_null_entry;
- return rt->fib6_nh.nh_flags & RTNH_F_DEAD ? net->ipv6.fib6_null_entry : rt;
+ return rt->fib6_nh.fib_nh_flags & RTNH_F_DEAD ? net->ipv6.fib6_null_entry : rt;
}
#ifdef CONFIG_IPV6_ROUTER_PREF
@@ -536,8 +536,8 @@ static void rt6_probe(struct fib6_info *rt)
if (!rt || !rt->fib6_nh.fib_nh_has_gw)
return;
- nh_gw = &rt->fib6_nh.nh_gw;
- dev = rt->fib6_nh.nh_dev;
+ nh_gw = &rt->fib6_nh.fib_nh_gw6;
+ dev = rt->fib6_nh.fib_nh_dev;
rcu_read_lock_bh();
idev = __in6_dev_get(dev);
neigh = __ipv6_neigh_lookup_noref(dev, nh_gw);
@@ -582,7 +582,7 @@ static inline void rt6_probe(struct fib6_info *rt)
*/
static inline int rt6_check_dev(struct fib6_info *rt, int oif)
{
- const struct net_device *dev = rt->fib6_nh.nh_dev;
+ const struct net_device *dev = rt->fib6_nh.fib_nh_dev;
if (!oif || dev->ifindex == oif)
return 2;
@@ -599,8 +599,8 @@ static inline enum rt6_nud_state rt6_check_neigh(struct fib6_info *rt)
return RT6_NUD_SUCCEED;
rcu_read_lock_bh();
- neigh = __ipv6_neigh_lookup_noref(rt->fib6_nh.nh_dev,
- &rt->fib6_nh.nh_gw);
+ neigh = __ipv6_neigh_lookup_noref(rt->fib6_nh.fib_nh_dev,
+ &rt->fib6_nh.fib_nh_gw6);
if (neigh) {
read_lock(&neigh->lock);
if (neigh->nud_state & NUD_VALID)
@@ -646,11 +646,11 @@ static struct fib6_info *find_match(struct fib6_info *rt, int oif, int strict,
int m;
bool match_do_rr = false;
- if (rt->fib6_nh.nh_flags & RTNH_F_DEAD)
+ if (rt->fib6_nh.fib_nh_flags & RTNH_F_DEAD)
goto out;
- if (ip6_ignore_linkdown(rt->fib6_nh.nh_dev) &&
- rt->fib6_nh.nh_flags & RTNH_F_LINKDOWN &&
+ if (ip6_ignore_linkdown(rt->fib6_nh.fib_nh_dev) &&
+ rt->fib6_nh.fib_nh_flags & RTNH_F_LINKDOWN &&
!(strict & RT6_LOOKUP_F_IGNORE_LINKSTATE))
goto out;
@@ -855,7 +855,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
/* called with rcu_lock held */
static struct net_device *ip6_rt_get_dev_rcu(struct fib6_info *rt)
{
- struct net_device *dev = rt->fib6_nh.nh_dev;
+ struct net_device *dev = rt->fib6_nh.fib_nh_dev;
if (rt->fib6_flags & (RTF_LOCAL | RTF_ANYCAST)) {
/* for copies of local routes, dst->dev needs to be the
@@ -949,8 +949,8 @@ static void ip6_rt_init_dst(struct rt6_info *rt, struct fib6_info *ort)
rt->dst.input = ip6_forward;
}
- if (ort->fib6_nh.nh_lwtstate) {
- rt->dst.lwtstate = lwtstate_get(ort->fib6_nh.nh_lwtstate);
+ if (ort->fib6_nh.fib_nh_lws) {
+ rt->dst.lwtstate = lwtstate_get(ort->fib6_nh.fib_nh_lws);
lwtunnel_set_redirect(&rt->dst);
}
@@ -976,7 +976,7 @@ static void ip6_rt_copy_init(struct rt6_info *rt, struct fib6_info *ort)
rt->rt6i_idev = dev ? in6_dev_get(dev) : NULL;
rt->rt6i_flags = ort->fib6_flags;
if (ort->fib6_nh.fib_nh_has_gw) {
- rt->rt6i_gateway = ort->fib6_nh.nh_gw;
+ rt->rt6i_gateway = ort->fib6_nh.fib_nh_gw6;
rt->rt6i_flags |= RTF_GATEWAY;
}
rt6_set_from(rt, ort);
@@ -1023,7 +1023,7 @@ static bool ip6_hold_safe(struct net *net, struct rt6_info **prt)
static struct rt6_info *ip6_create_rt_rcu(struct fib6_info *rt)
{
unsigned short flags = fib6_info_dst_flags(rt);
- struct net_device *dev = rt->fib6_nh.nh_dev;
+ struct net_device *dev = rt->fib6_nh.fib_nh_dev;
struct rt6_info *nrt;
if (!fib6_info_hold_safe(rt))
@@ -1405,7 +1405,7 @@ static unsigned int fib6_mtu(const struct fib6_info *rt)
mtu = min_t(unsigned int, mtu, IP6_MAX_MTU);
- return mtu - lwtunnel_headroom(rt->fib6_nh.nh_lwtstate, mtu);
+ return mtu - lwtunnel_headroom(rt->fib6_nh.fib_nh_lws, mtu);
}
static int rt6_insert_exception(struct rt6_info *nrt,
@@ -2422,7 +2422,7 @@ static struct rt6_info *__ip6_route_redirect(struct net *net,
fn = fib6_node_lookup(&table->tb6_root, &fl6->daddr, &fl6->saddr);
restart:
for_each_fib6_node_rt_rcu(fn) {
- if (rt->fib6_nh.nh_flags & RTNH_F_DEAD)
+ if (rt->fib6_nh.fib_nh_flags & RTNH_F_DEAD)
continue;
if (fib6_check_expired(rt))
continue;
@@ -2430,14 +2430,14 @@ static struct rt6_info *__ip6_route_redirect(struct net *net,
break;
if (!rt->fib6_nh.fib_nh_has_gw)
continue;
- if (fl6->flowi6_oif != rt->fib6_nh.nh_dev->ifindex)
+ if (fl6->flowi6_oif != rt->fib6_nh.fib_nh_dev->ifindex)
continue;
/* rt_cache's gateway might be different from its 'parent'
* in the case of an ip redirect.
* So we keep searching in the exception table if the gateway
* is different.
*/
- if (!ipv6_addr_equal(&rdfl->gateway, &rt->fib6_nh.nh_gw)) {
+ if (!ipv6_addr_equal(&rdfl->gateway, &rt->fib6_nh.fib_nh_gw6)) {
rt_cache = rt6_find_cached_rt(rt,
&fl6->daddr,
&fl6->saddr);
@@ -2916,7 +2916,7 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
goto out;
}
- fib6_nh->nh_flags |= RTNH_F_ONLINK;
+ fib6_nh->fib_nh_flags |= RTNH_F_ONLINK;
}
if (cfg->fc_encap) {
@@ -2928,10 +2928,10 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
if (err)
goto out;
- fib6_nh->nh_lwtstate = lwtstate_get(lwtstate);
+ fib6_nh->fib_nh_lws = lwtstate_get(lwtstate);
}
- fib6_nh->nh_weight = 1;
+ fib6_nh->fib_nh_weight = 1;
/* We cannot add true routes via loopback here,
* they would result in kernel looping; promote them to reject routes
@@ -2964,7 +2964,7 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
if (err)
goto out;
- fib6_nh->nh_gw = cfg->fc_gateway;
+ fib6_nh->fib_nh_gw6 = cfg->fc_gateway;
fib6_nh->fib_nh_has_gw = 1;
}
@@ -2986,10 +2986,10 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
if (!(cfg->fc_flags & (RTF_LOCAL | RTF_ANYCAST)) &&
!netif_carrier_ok(dev))
- fib6_nh->nh_flags |= RTNH_F_LINKDOWN;
+ fib6_nh->fib_nh_flags |= RTNH_F_LINKDOWN;
set_dev:
- fib6_nh->nh_dev = dev;
+ fib6_nh->fib_nh_dev = dev;
err = 0;
out:
if (idev)
@@ -3003,10 +3003,10 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
void fib6_nh_release(struct fib6_nh *fib6_nh)
{
- if (fib6_nh->nh_dev)
- dev_put(fib6_nh->nh_dev);
+ if (fib6_nh->fib_nh_dev)
+ dev_put(fib6_nh->fib_nh_dev);
- lwtstate_put(fib6_nh->nh_lwtstate);
+ lwtstate_put(fib6_nh->fib_nh_lws);
}
static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
@@ -3267,6 +3267,8 @@ static int ip6_route_del(struct fib6_config *cfg,
if (fn) {
for_each_fib6_node_rt_rcu(fn) {
+ struct fib6_nh *nh;
+
if (cfg->fc_flags & RTF_CACHE) {
int rc;
@@ -3281,12 +3283,14 @@ static int ip6_route_del(struct fib6_config *cfg,
}
continue;
}
+
+ nh = &rt->fib6_nh;
if (cfg->fc_ifindex &&
- (!rt->fib6_nh.nh_dev ||
- rt->fib6_nh.nh_dev->ifindex != cfg->fc_ifindex))
+ (!nh->fib_nh_dev ||
+ nh->fib_nh_dev->ifindex != cfg->fc_ifindex))
continue;
if (cfg->fc_flags & RTF_GATEWAY &&
- !ipv6_addr_equal(&cfg->fc_gateway, &rt->fib6_nh.nh_gw))
+ !ipv6_addr_equal(&cfg->fc_gateway, &nh->fib_nh_gw6))
continue;
if (cfg->fc_metric && cfg->fc_metric != rt->fib6_metric)
continue;
@@ -3457,12 +3461,12 @@ static struct fib6_info *rt6_get_route_info(struct net *net,
goto out;
for_each_fib6_node_rt_rcu(fn) {
- if (rt->fib6_nh.nh_dev->ifindex != ifindex)
+ if (rt->fib6_nh.fib_nh_dev->ifindex != ifindex)
continue;
if (!(rt->fib6_flags & RTF_ROUTEINFO) ||
!rt->fib6_nh.fib_nh_has_gw)
continue;
- if (!ipv6_addr_equal(&rt->fib6_nh.nh_gw, gwaddr))
+ if (!ipv6_addr_equal(&rt->fib6_nh.fib_nh_gw6, gwaddr))
continue;
if (!fib6_info_hold_safe(rt))
continue;
@@ -3520,9 +3524,11 @@ struct fib6_info *rt6_get_dflt_router(struct net *net,
rcu_read_lock();
for_each_fib6_node_rt_rcu(&table->tb6_root) {
- if (dev == rt->fib6_nh.nh_dev &&
+ struct fib6_nh *nh = &rt->fib6_nh;
+
+ if (dev == nh->fib_nh_dev &&
((rt->fib6_flags & (RTF_ADDRCONF | RTF_DEFAULT)) == (RTF_ADDRCONF | RTF_DEFAULT)) &&
- ipv6_addr_equal(&rt->fib6_nh.nh_gw, addr))
+ ipv6_addr_equal(&nh->fib_nh_gw6, addr))
break;
}
if (rt && !fib6_info_hold_safe(rt))
@@ -3759,7 +3765,7 @@ static int fib6_remove_prefsrc(struct fib6_info *rt, void *arg)
struct net *net = ((struct arg_dev_net_ip *)arg)->net;
struct in6_addr *addr = ((struct arg_dev_net_ip *)arg)->addr;
- if (((void *)rt->fib6_nh.nh_dev == dev || !dev) &&
+ if (((void *)rt->fib6_nh.fib_nh_dev == dev || !dev) &&
rt != net->ipv6.fib6_null_entry &&
ipv6_addr_equal(addr, &rt->fib6_prefsrc.addr)) {
spin_lock_bh(&rt6_exception_lock);
@@ -3790,7 +3796,7 @@ static int fib6_clean_tohost(struct fib6_info *rt, void *arg)
if (((rt->fib6_flags & RTF_RA_ROUTER) == RTF_RA_ROUTER) &&
rt->fib6_nh.fib_nh_has_gw &&
- ipv6_addr_equal(gateway, &rt->fib6_nh.nh_gw)) {
+ ipv6_addr_equal(gateway, &rt->fib6_nh.fib_nh_gw6)) {
return -1;
}
@@ -3838,9 +3844,9 @@ static struct fib6_info *rt6_multipath_first_sibling(const struct fib6_info *rt)
static bool rt6_is_dead(const struct fib6_info *rt)
{
- if (rt->fib6_nh.nh_flags & RTNH_F_DEAD ||
- (rt->fib6_nh.nh_flags & RTNH_F_LINKDOWN &&
- ip6_ignore_linkdown(rt->fib6_nh.nh_dev)))
+ if (rt->fib6_nh.fib_nh_flags & RTNH_F_DEAD ||
+ (rt->fib6_nh.fib_nh_flags & RTNH_F_LINKDOWN &&
+ ip6_ignore_linkdown(rt->fib6_nh.fib_nh_dev)))
return true;
return false;
@@ -3852,11 +3858,11 @@ static int rt6_multipath_total_weight(const struct fib6_info *rt)
int total = 0;
if (!rt6_is_dead(rt))
- total += rt->fib6_nh.nh_weight;
+ total += rt->fib6_nh.fib_nh_weight;
list_for_each_entry(iter, &rt->fib6_siblings, fib6_siblings) {
if (!rt6_is_dead(iter))
- total += iter->fib6_nh.nh_weight;
+ total += iter->fib6_nh.fib_nh_weight;
}
return total;
@@ -3867,11 +3873,11 @@ static void rt6_upper_bound_set(struct fib6_info *rt, int *weight, int total)
int upper_bound = -1;
if (!rt6_is_dead(rt)) {
- *weight += rt->fib6_nh.nh_weight;
+ *weight += rt->fib6_nh.fib_nh_weight;
upper_bound = DIV_ROUND_CLOSEST_ULL((u64) (*weight) << 31,
total) - 1;
}
- atomic_set(&rt->fib6_nh.nh_upper_bound, upper_bound);
+ atomic_set(&rt->fib6_nh.fib_nh_upper_bound, upper_bound);
}
static void rt6_multipath_upper_bound_set(struct fib6_info *rt, int total)
@@ -3914,8 +3920,9 @@ static int fib6_ifup(struct fib6_info *rt, void *p_arg)
const struct arg_netdev_event *arg = p_arg;
struct net *net = dev_net(arg->dev);
- if (rt != net->ipv6.fib6_null_entry && rt->fib6_nh.nh_dev == arg->dev) {
- rt->fib6_nh.nh_flags &= ~arg->nh_flags;
+ if (rt != net->ipv6.fib6_null_entry &&
+ rt->fib6_nh.fib_nh_dev == arg->dev) {
+ rt->fib6_nh.fib_nh_flags &= ~arg->nh_flags;
fib6_update_sernum_upto_root(net, rt);
rt6_multipath_rebalance(rt);
}
@@ -3943,10 +3950,10 @@ static bool rt6_multipath_uses_dev(const struct fib6_info *rt,
{
struct fib6_info *iter;
- if (rt->fib6_nh.nh_dev == dev)
+ if (rt->fib6_nh.fib_nh_dev == dev)
return true;
list_for_each_entry(iter, &rt->fib6_siblings, fib6_siblings)
- if (iter->fib6_nh.nh_dev == dev)
+ if (iter->fib6_nh.fib_nh_dev == dev)
return true;
return false;
@@ -3967,12 +3974,12 @@ static unsigned int rt6_multipath_dead_count(const struct fib6_info *rt,
struct fib6_info *iter;
unsigned int dead = 0;
- if (rt->fib6_nh.nh_dev == down_dev ||
- rt->fib6_nh.nh_flags & RTNH_F_DEAD)
+ if (rt->fib6_nh.fib_nh_dev == down_dev ||
+ rt->fib6_nh.fib_nh_flags & RTNH_F_DEAD)
dead++;
list_for_each_entry(iter, &rt->fib6_siblings, fib6_siblings)
- if (iter->fib6_nh.nh_dev == down_dev ||
- iter->fib6_nh.nh_flags & RTNH_F_DEAD)
+ if (iter->fib6_nh.fib_nh_dev == down_dev ||
+ iter->fib6_nh.fib_nh_flags & RTNH_F_DEAD)
dead++;
return dead;
@@ -3984,11 +3991,11 @@ static void rt6_multipath_nh_flags_set(struct fib6_info *rt,
{
struct fib6_info *iter;
- if (rt->fib6_nh.nh_dev == dev)
- rt->fib6_nh.nh_flags |= nh_flags;
+ if (rt->fib6_nh.fib_nh_dev == dev)
+ rt->fib6_nh.fib_nh_flags |= nh_flags;
list_for_each_entry(iter, &rt->fib6_siblings, fib6_siblings)
- if (iter->fib6_nh.nh_dev == dev)
- iter->fib6_nh.nh_flags |= nh_flags;
+ if (iter->fib6_nh.fib_nh_dev == dev)
+ iter->fib6_nh.fib_nh_flags |= nh_flags;
}
/* called with write lock held for table with rt */
@@ -4003,12 +4010,12 @@ static int fib6_ifdown(struct fib6_info *rt, void *p_arg)
switch (arg->event) {
case NETDEV_UNREGISTER:
- return rt->fib6_nh.nh_dev == dev ? -1 : 0;
+ return rt->fib6_nh.fib_nh_dev == dev ? -1 : 0;
case NETDEV_DOWN:
if (rt->should_flush)
return -1;
if (!rt->fib6_nsiblings)
- return rt->fib6_nh.nh_dev == dev ? -1 : 0;
+ return rt->fib6_nh.fib_nh_dev == dev ? -1 : 0;
if (rt6_multipath_uses_dev(rt, dev)) {
unsigned int count;
@@ -4024,10 +4031,10 @@ static int fib6_ifdown(struct fib6_info *rt, void *p_arg)
}
return -2;
case NETDEV_CHANGE:
- if (rt->fib6_nh.nh_dev != dev ||
+ if (rt->fib6_nh.fib_nh_dev != dev ||
rt->fib6_flags & (RTF_LOCAL | RTF_ANYCAST))
break;
- rt->fib6_nh.nh_flags |= RTNH_F_LINKDOWN;
+ rt->fib6_nh.fib_nh_flags |= RTNH_F_LINKDOWN;
rt6_multipath_rebalance(rt);
break;
}
@@ -4083,7 +4090,7 @@ static int rt6_mtu_change_route(struct fib6_info *rt, void *p_arg)
Since RFC 1981 doesn't include administrative MTU increase
update PMTU increase is a MUST. (i.e. jumbo frame)
*/
- if (rt->fib6_nh.nh_dev == arg->dev &&
+ if (rt->fib6_nh.fib_nh_dev == arg->dev &&
!fib6_metric_locked(rt, RTAX_MTU)) {
u32 mtu = rt->fib6_pmtu;
@@ -4374,7 +4381,7 @@ static int ip6_route_multipath_add(struct fib6_config *cfg,
goto cleanup;
}
- rt->fib6_nh.nh_weight = rtnh->rtnh_hops + 1;
+ rt->fib6_nh.fib_nh_weight = rtnh->rtnh_hops + 1;
err = ip6_route_info_append(info->nl_net, &rt6_nh_list,
rt, &r_cfg);
@@ -4541,7 +4548,7 @@ static size_t rt6_nlmsg_size(struct fib6_info *rt)
nexthop_len = nla_total_size(0) /* RTA_MULTIPATH */
+ NLA_ALIGN(sizeof(struct rtnexthop))
+ nla_total_size(16) /* RTA_GATEWAY */
- + lwtunnel_get_encap_size(rt->fib6_nh.nh_lwtstate);
+ + lwtunnel_get_encap_size(rt->fib6_nh.fib_nh_lws);
nexthop_len *= rt->fib6_nsiblings;
}
@@ -4559,41 +4566,41 @@ static size_t rt6_nlmsg_size(struct fib6_info *rt)
+ nla_total_size(sizeof(struct rta_cacheinfo))
+ nla_total_size(TCP_CA_NAME_MAX) /* RTAX_CC_ALGO */
+ nla_total_size(1) /* RTA_PREF */
- + lwtunnel_get_encap_size(rt->fib6_nh.nh_lwtstate)
+ + lwtunnel_get_encap_size(rt->fib6_nh.fib_nh_lws)
+ nexthop_len;
}
static int rt6_nexthop_info(struct sk_buff *skb, const struct fib6_nh *fib6_nh,
unsigned int *flags, bool skip_oif)
{
- if (fib6_nh->nh_flags & RTNH_F_DEAD)
+ if (fib6_nh->fib_nh_flags & RTNH_F_DEAD)
*flags |= RTNH_F_DEAD;
- if (fib6_nh->nh_flags & RTNH_F_LINKDOWN) {
+ if (fib6_nh->fib_nh_flags & RTNH_F_LINKDOWN) {
*flags |= RTNH_F_LINKDOWN;
rcu_read_lock();
- if (ip6_ignore_linkdown(fib6_nh->nh_dev))
+ if (ip6_ignore_linkdown(fib6_nh->fib_nh_dev))
*flags |= RTNH_F_DEAD;
rcu_read_unlock();
}
if (fib6_nh->fib_nh_has_gw) {
- if (nla_put_in6_addr(skb, RTA_GATEWAY, &fib6_nh->nh_gw) < 0)
+ if (nla_put_in6_addr(skb, RTA_GATEWAY, &fib6_nh->fib_nh_gw6) < 0)
goto nla_put_failure;
}
- *flags |= (fib6_nh->nh_flags & RTNH_F_ONLINK);
- if (fib6_nh->nh_flags & RTNH_F_OFFLOAD)
+ *flags |= (fib6_nh->fib_nh_flags & RTNH_F_ONLINK);
+ if (fib6_nh->fib_nh_flags & RTNH_F_OFFLOAD)
*flags |= RTNH_F_OFFLOAD;
/* not needed for multipath encoding b/c it has a rtnexthop struct */
- if (!skip_oif && fib6_nh->nh_dev &&
- nla_put_u32(skb, RTA_OIF, fib6_nh->nh_dev->ifindex))
+ if (!skip_oif && fib6_nh->fib_nh_dev &&
+ nla_put_u32(skb, RTA_OIF, fib6_nh->fib_nh_dev->ifindex))
goto nla_put_failure;
- if (fib6_nh->nh_lwtstate &&
- lwtunnel_fill_encap(skb, fib6_nh->nh_lwtstate) < 0)
+ if (fib6_nh->fib_nh_lws &&
+ lwtunnel_fill_encap(skb, fib6_nh->fib_nh_lws) < 0)
goto nla_put_failure;
return 0;
@@ -4605,7 +4612,7 @@ static int rt6_nexthop_info(struct sk_buff *skb, const struct fib6_nh *fib6_nh,
/* add multipath next hop */
static int rt6_add_nexthop(struct sk_buff *skb, const struct fib6_nh *fib6_nh)
{
- const struct net_device *dev = fib6_nh->nh_dev;
+ const struct net_device *dev = fib6_nh->fib_nh_dev;
struct rtnexthop *rtnh;
unsigned int flags = 0;
@@ -4613,7 +4620,7 @@ static int rt6_add_nexthop(struct sk_buff *skb, const struct fib6_nh *fib6_nh)
if (!rtnh)
goto nla_put_failure;
- rtnh->rtnh_hops = fib6_nh->nh_weight - 1;
+ rtnh->rtnh_hops = fib6_nh->fib_nh_weight - 1;
rtnh->rtnh_ifindex = dev ? dev->ifindex : 0;
if (rt6_nexthop_info(skb, fib6_nh, &flags, true) < 0)
@@ -4785,7 +4792,7 @@ static int rt6_fill_node(struct net *net, struct sk_buff *skb,
static bool fib6_info_uses_dev(const struct fib6_info *f6i,
const struct net_device *dev)
{
- if (f6i->fib6_nh.nh_dev == dev)
+ if (f6i->fib6_nh.fib_nh_dev == dev)
return true;
if (f6i->fib6_nsiblings) {
@@ -4793,7 +4800,7 @@ static bool fib6_info_uses_dev(const struct fib6_info *f6i,
list_for_each_entry_safe(sibling, next_sibling,
&f6i->fib6_siblings, fib6_siblings) {
- if (sibling->fib6_nh.nh_dev == dev)
+ if (sibling->fib6_nh.fib_nh_dev == dev)
return true;
}
}
@@ -5078,7 +5085,7 @@ static int ip6_route_dev_notify(struct notifier_block *this,
return NOTIFY_OK;
if (event == NETDEV_REGISTER) {
- net->ipv6.fib6_null_entry->fib6_nh.nh_dev = dev;
+ net->ipv6.fib6_null_entry->fib6_nh.fib_nh_dev = dev;
net->ipv6.ip6_null_entry->dst.dev = dev;
net->ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(dev);
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
@@ -5413,7 +5420,7 @@ void __init ip6_route_init_special_entries(void)
/* Registering of the loopback is done before this portion of code,
* the loopback reference in rt6_info will not be taken, do it
* manually for init_net */
- init_net.ipv6.fib6_null_entry->fib6_nh.nh_dev = init_net.loopback_dev;
+ init_net.ipv6.fib6_null_entry->fib6_nh.fib_nh_dev = init_net.loopback_dev;
init_net.ipv6.ip6_null_entry->dst.dev = init_net.loopback_dev;
init_net.ipv6.ip6_null_entry->rt6i_idev = in6_dev_get(init_net.loopback_dev);
#ifdef CONFIG_IPV6_MULTIPLE_TABLES
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 12/13] net: Add fib_nh_common and update fib_nh and fib6_nh
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
` (10 preceding siblings ...)
2019-03-27 3:29 ` [PATCH net-next 11/13] ipv6: Rename fib6_nh entries David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 9:54 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 13/13] net: Use common nexthop init and release helpers David Ahern
12 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
Add fib_nh_common struct with common nexthop attributes. Convert
fib_nh and fib6_nh to use it. Use macros to move existing
fib_nh_* references to the new nh_common.nhc_*.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
include/net/ip6_fib.h | 10 ++--------
include/net/ip_fib.h | 41 +++++++++++++++++++++++++++++++----------
net/ipv4/fib_semantics.c | 7 ++++++-
net/ipv6/route.c | 3 +++
4 files changed, 42 insertions(+), 19 deletions(-)
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index aff8570725c8..58dbb4e82908 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -19,6 +19,7 @@
#include <linux/notifier.h>
#include <net/dst.h>
#include <net/flow.h>
+#include <net/ip_fib.h>
#include <net/netlink.h>
#include <net/inetpeer.h>
#include <net/fib_notifier.h>
@@ -125,14 +126,7 @@ struct rt6_exception {
#define FIB6_MAX_DEPTH 5
struct fib6_nh {
- struct in6_addr fib_nh_gw6;
- bool fib_nh_has_gw;
- struct net_device *fib_nh_dev;
- struct lwtunnel_state *fib_nh_lws;
-
- unsigned int fib_nh_flags;
- atomic_t fib_nh_upper_bound;
- int fib_nh_weight;
+ struct fib_nh_common nh_common;
};
struct fib6_info {
diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 029acd333d29..70548b1a6322 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -76,27 +76,48 @@ struct fnhe_hash_bucket {
#define FNHE_HASH_SIZE (1 << FNHE_HASH_SHIFT)
#define FNHE_RECLAIM_DEPTH 5
+struct fib_nh_common {
+ struct net_device *nhc_dev;
+ int nhc_oif;
+ unsigned int nhc_flags;
+ struct lwtunnel_state *nhc_lwtstate;
+ unsigned char nhc_scope;
+ u8 nhc_family;
+ u8 nhc_has_gw:1,
+ unused:7;
+ union {
+ __be32 ipv4;
+ struct in6_addr ipv6;
+ } nhc_gw;
+
+ int nhc_weight;
+ atomic_t nhc_upper_bound;
+};
+
struct fib_nh {
- struct net_device *fib_nh_dev;
+ struct fib_nh_common nh_common;
struct hlist_node nh_hash;
struct fib_info *nh_parent;
- unsigned int fib_nh_flags;
- unsigned char fib_nh_scope;
-#ifdef CONFIG_IP_ROUTE_MULTIPATH
- int fib_nh_weight;
- atomic_t fib_nh_upper_bound;
-#endif
#ifdef CONFIG_IP_ROUTE_CLASSID
__u32 nh_tclassid;
#endif
- int fib_nh_oif;
- __be32 fib_nh_gw4;
__be32 nh_saddr;
int nh_saddr_genid;
struct rtable __rcu * __percpu *nh_pcpu_rth_output;
struct rtable __rcu *nh_rth_input;
struct fnhe_hash_bucket __rcu *nh_exceptions;
- struct lwtunnel_state *fib_nh_lws;
+#define fib_nh_family nh_common.nhc_family
+#define fib_nh_dev nh_common.nhc_dev
+#define fib_nh_oif nh_common.nhc_oif
+#define fib_nh_flags nh_common.nhc_flags
+#define fib_nh_lws nh_common.nhc_lwtstate
+#define fib_nh_scope nh_common.nhc_scope
+#define fib_nh_family nh_common.nhc_family
+#define fib_nh_has_gw nh_common.nhc_has_gw
+#define fib_nh_gw4 nh_common.nhc_gw.ipv4
+#define fib_nh_gw6 nh_common.nhc_gw.ipv6
+#define fib_nh_weight nh_common.nhc_weight
+#define fib_nh_upper_bound nh_common.nhc_upper_bound
};
/*
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 46e941f505d5..b44ee0404260 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -468,6 +468,8 @@ int fib_nh_init(struct net *net, struct fib_nh *nh,
{
int err = -ENOMEM;
+ nh->fib_nh_family = AF_INET;
+
nh->nh_pcpu_rth_output = alloc_percpu(struct rtable __rcu *);
if (!nh->nh_pcpu_rth_output)
goto failure;
@@ -490,7 +492,10 @@ int fib_nh_init(struct net *net, struct fib_nh *nh,
}
nh->fib_nh_oif = cfg->fc_oif;
- nh->fib_nh_gw4 = cfg->fc_gw;
+ if (cfg->fc_gw) {
+ nh->fib_nh_gw4 = cfg->fc_gw;
+ nh->fib_nh_has_gw = 1;
+ }
nh->fib_nh_flags = cfg->fc_flags;
#ifdef CONFIG_IP_ROUTE_CLASSID
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index b9218e3f0522..490834c5758b 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2893,6 +2893,8 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
int addr_type;
int err;
+ fib6_nh->fib_nh_family = AF_INET6;
+
err = -ENODEV;
if (cfg->fc_ifindex) {
dev = dev_get_by_index(net, cfg->fc_ifindex);
@@ -2990,6 +2992,7 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
set_dev:
fib6_nh->fib_nh_dev = dev;
+ fib6_nh->fib_nh_oif = dev->ifindex;
err = 0;
out:
if (idev)
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 13/13] net: Use common nexthop init and release helpers
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
` (11 preceding siblings ...)
2019-03-27 3:29 ` [PATCH net-next 12/13] net: Add fib_nh_common and update fib_nh and fib6_nh David Ahern
@ 2019-03-27 3:29 ` David Ahern
2019-03-27 9:46 ` Ido Schimmel
12 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-03-27 3:29 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, jiri, saeedm, David Ahern
From: David Ahern <dsahern@gmail.com>
With fib_nh_common in place, move common initialization and release
code into helpers used by both ipv4 and ipv6. For the moment, the init
is just the lwt encap and the release is both the net dev reference and
the the lwt state reference. More will be added later.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
include/net/ip_fib.h | 4 ++++
net/ipv4/fib_semantics.c | 60 +++++++++++++++++++++++++++++++-----------------
net/ipv6/route.c | 21 ++++-------------
3 files changed, 48 insertions(+), 37 deletions(-)
diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 70548b1a6322..12a6d759cf57 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -441,6 +441,10 @@ int fib_nh_init(struct net *net, struct fib_nh *fib_nh,
struct fib_config *cfg, int nh_weight,
struct netlink_ext_ack *extack);
void fib_nh_release(struct net *net, struct fib_nh *fib_nh);
+int fib_nh_common_init(struct fib_nh_common *nhc, struct nlattr *fc_encap,
+ u16 fc_encap_type, void *cfg, gfp_t gfp_flags,
+ struct netlink_ext_ack *extack);
+void fib_nh_common_release(struct fib_nh_common *nhc);
/* Exported by fib_trie.c */
void fib_trie_init(void);
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index b44ee0404260..d5738bc8904c 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -204,16 +204,22 @@ static void rt_fibinfo_free_cpus(struct rtable __rcu * __percpu *rtp)
free_percpu(rtp);
}
+void fib_nh_common_release(struct fib_nh_common *nhc)
+{
+ if (nhc->nhc_dev)
+ dev_put(nhc->nhc_dev);
+
+ lwtstate_put(nhc->nhc_lwtstate);
+}
+EXPORT_SYMBOL_GPL(fib_nh_common_release);
+
void fib_nh_release(struct net *net, struct fib_nh *fib_nh)
{
#ifdef CONFIG_IP_ROUTE_CLASSID
if (fib_nh->nh_tclassid)
net->ipv4.fib_num_tclassid_users--;
#endif
- if (fib_nh->fib_nh_dev)
- dev_put(fib_nh->fib_nh_dev);
-
- lwtstate_put(fib_nh->fib_nh_lws);
+ fib_nh_common_release(&fib_nh->nh_common);
free_nh_exceptions(fib_nh);
rt_fibinfo_free_cpus(fib_nh->nh_pcpu_rth_output);
rt_fibinfo_free(&fib_nh->nh_rth_input);
@@ -462,6 +468,30 @@ static int fib_detect_death(struct fib_info *fi, int order,
return 1;
}
+int fib_nh_common_init(struct fib_nh_common *nhc, struct nlattr *encap,
+ u16 encap_type, void *cfg, gfp_t gfp_flags,
+ struct netlink_ext_ack *extack)
+{
+ if (encap) {
+ struct lwtunnel_state *lwtstate;
+ int err;
+
+ if (encap_type == LWTUNNEL_ENCAP_NONE) {
+ NL_SET_ERR_MSG(extack, "LWT encap type not specified");
+ return -EINVAL;
+ }
+ err = lwtunnel_build_state(encap_type, encap, nhc->nhc_family,
+ cfg, &lwtstate, extack);
+ if (err)
+ return err;
+
+ nhc->nhc_lwtstate = lwtstate_get(lwtstate);
+ }
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(fib_nh_common_init);
+
int fib_nh_init(struct net *net, struct fib_nh *nh,
struct fib_config *cfg, int nh_weight,
struct netlink_ext_ack *extack)
@@ -474,26 +504,14 @@ int fib_nh_init(struct net *net, struct fib_nh *nh,
if (!nh->nh_pcpu_rth_output)
goto failure;
- if (cfg->fc_encap) {
- struct lwtunnel_state *lwtstate;
-
- err = -EINVAL;
- if (cfg->fc_encap_type == LWTUNNEL_ENCAP_NONE) {
- NL_SET_ERR_MSG(extack, "LWT encap type not specified");
- goto failure;
- }
- err = lwtunnel_build_state(cfg->fc_encap_type,
- cfg->fc_encap, AF_INET, cfg,
- &lwtstate, extack);
- if (err)
- goto failure;
-
- nh->fib_nh_lws = lwtstate_get(lwtstate);
- }
+ err = fib_nh_common_init(&nh->nh_common, cfg->fc_encap,
+ cfg->fc_encap_type, cfg, GFP_KERNEL, extack);
+ if (err)
+ goto failure;
nh->fib_nh_oif = cfg->fc_oif;
if (cfg->fc_gw) {
- nh->fib_nh_gw4 = cfg->fc_gw;
+ nh->fib_nh_gw4 = cfg->fc_gw;
nh->fib_nh_has_gw = 1;
}
nh->fib_nh_flags = cfg->fc_flags;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 490834c5758b..250c439242ff 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2921,18 +2921,6 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
fib6_nh->fib_nh_flags |= RTNH_F_ONLINK;
}
- if (cfg->fc_encap) {
- struct lwtunnel_state *lwtstate;
-
- err = lwtunnel_build_state(cfg->fc_encap_type,
- cfg->fc_encap, AF_INET6, cfg,
- &lwtstate, extack);
- if (err)
- goto out;
-
- fib6_nh->fib_nh_lws = lwtstate_get(lwtstate);
- }
-
fib6_nh->fib_nh_weight = 1;
/* We cannot add true routes via loopback here,
@@ -2990,6 +2978,10 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
!netif_carrier_ok(dev))
fib6_nh->fib_nh_flags |= RTNH_F_LINKDOWN;
+ err = fib_nh_common_init(&fib6_nh->nh_common, cfg->fc_encap,
+ cfg->fc_encap_type, cfg, gfp_flags, extack);
+ if (err)
+ goto out;
set_dev:
fib6_nh->fib_nh_dev = dev;
fib6_nh->fib_nh_oif = dev->ifindex;
@@ -3006,10 +2998,7 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
void fib6_nh_release(struct fib6_nh *fib6_nh)
{
- if (fib6_nh->fib_nh_dev)
- dev_put(fib6_nh->fib_nh_dev);
-
- lwtstate_put(fib6_nh->fib_nh_lws);
+ fib_nh_common_release(&fib6_nh->nh_common);
}
static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 01/13] ipv4: Define fib_get_nhs when CONFIG_IP_ROUTE_MULTIPATH is disabled
2019-03-27 3:29 ` [PATCH net-next 01/13] ipv4: Define fib_get_nhs when CONFIG_IP_ROUTE_MULTIPATH is disabled David Ahern
@ 2019-03-27 7:40 ` Ido Schimmel
0 siblings, 0 replies; 29+ messages in thread
From: Ido Schimmel @ 2019-03-27 7:40 UTC (permalink / raw)
To: David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed, David Ahern
On Tue, Mar 26, 2019 at 08:29:30PM -0700, David Ahern wrote:
> From: David Ahern <dsahern@gmail.com>
>
> Define fib_get_nhs to return EINVAL when CONFIG_IP_ROUTE_MULTIPATH is
> not enabled and remove the ifdef check for CONFIG_IP_ROUTE_MULTIPATH
> in fib_create_info.
>
> Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 02/13] ipv4: Move IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN to helper
2019-03-27 3:29 ` [PATCH net-next 02/13] ipv4: Move IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN to helper David Ahern
@ 2019-03-27 7:44 ` Ido Schimmel
0 siblings, 0 replies; 29+ messages in thread
From: Ido Schimmel @ 2019-03-27 7:44 UTC (permalink / raw)
To: David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed, David Ahern
On Tue, Mar 26, 2019 at 08:29:31PM -0700, David Ahern wrote:
> From: David Ahern <dsahern@gmail.com>
>
> in_dev lookup followed by IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN check
> is called in several places, some with the rcu lock and others with the
> rtnl held.
>
> Move the check to a helper similar to what IPv6 has. Since the helper
> can be invoked from either context use rcu_dereference_rtnl to
> dereference ip_ptr.
>
> Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
See one nit below.
> ---
> include/linux/inetdevice.h | 14 ++++++++++++++
> net/ipv4/fib_semantics.c | 31 +++++++------------------------
> net/ipv4/fib_trie.c | 4 +---
> 3 files changed, 22 insertions(+), 27 deletions(-)
>
> diff --git a/include/linux/inetdevice.h b/include/linux/inetdevice.h
> index a64f21a97369..35e0390fd7b8 100644
> --- a/include/linux/inetdevice.h
> +++ b/include/linux/inetdevice.h
> @@ -237,6 +237,20 @@ static inline struct in_device *__in_dev_get_rtnl(const struct net_device *dev)
> return rtnl_dereference(dev->ip_ptr);
> }
>
> +/* called with rc_read_lock or rtnl held */
s/rc_read_lock/rcu_read_lock/
> +static inline bool ip_ignore_linkdown(const struct net_device *dev)
> +{
> + struct in_device *in_dev;
> + bool rc = false;
> +
> + in_dev = rcu_dereference_rtnl(dev->ip_ptr);
> + if (in_dev &&
> + IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN(in_dev))
> + rc = true;
> +
> + return rc;
> +}
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 03/13] ipv4: Create init helper for fib_nh
2019-03-27 3:29 ` [PATCH net-next 03/13] ipv4: Create init helper for fib_nh David Ahern
@ 2019-03-27 8:12 ` Ido Schimmel
2019-03-27 14:17 ` David Ahern
0 siblings, 1 reply; 29+ messages in thread
From: Ido Schimmel @ 2019-03-27 8:12 UTC (permalink / raw)
To: David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed, David Ahern
On Tue, Mar 26, 2019 at 08:29:32PM -0700, David Ahern wrote:
> +int fib_nh_init(struct net *net, struct fib_nh *nh,
> + struct fib_config *cfg, int nh_weight,
> + struct netlink_ext_ack *extack)
> +{
> + int err = -ENOMEM;
> +
> + nh->nh_pcpu_rth_output = alloc_percpu(struct rtable __rcu *);
> + if (!nh->nh_pcpu_rth_output)
> + goto failure;
> +
> + if (cfg->fc_encap) {
> + struct lwtunnel_state *lwtstate;
> +
> + err = -EINVAL;
> + if (cfg->fc_encap_type == LWTUNNEL_ENCAP_NONE) {
> + NL_SET_ERR_MSG(extack, "LWT encap type not specified");
> + goto failure;
This is very confusing and probably error-prone. You call
alloc_percpu(), but don't free it in the error path and instead rely on
the call to free_fib_info() in the error path of fib_create_info().
Better to free it here and NULL-ify the pointer. Then the call to
rt_fibinfo_free_cpus() in free_fib_info_rcu() is basically a NOP.
> + }
> + err = lwtunnel_build_state(cfg->fc_encap_type,
> + cfg->fc_encap, AF_INET, cfg,
> + &lwtstate, extack);
> + if (err)
> + goto failure;
> +
> + nh->nh_lwtstate = lwtstate_get(lwtstate);
> + }
> +
> + nh->nh_oif = cfg->fc_oif;
> + nh->nh_gw = cfg->fc_gw;
> + nh->nh_flags = cfg->fc_flags;
> +
> +#ifdef CONFIG_IP_ROUTE_CLASSID
> + nh->nh_tclassid = cfg->fc_flow;
> + if (nh->nh_tclassid)
> + net->ipv4.fib_num_tclassid_users++;
> +#endif
> +#ifdef CONFIG_IP_ROUTE_MULTIPATH
> + nh->nh_weight = nh_weight;
> +#endif
> + err = 0;
> +
> +failure:
> + return err;
> +}
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 04/13] ipv4: Create cleanup helper for fib_nh
2019-03-27 3:29 ` [PATCH net-next 04/13] ipv4: Create cleanup " David Ahern
@ 2019-03-27 8:17 ` Ido Schimmel
0 siblings, 0 replies; 29+ messages in thread
From: Ido Schimmel @ 2019-03-27 8:17 UTC (permalink / raw)
To: David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed, David Ahern
On Tue, Mar 26, 2019 at 08:29:33PM -0700, David Ahern wrote:
> From: David Ahern <dsahern@gmail.com>
>
> Move the fib_nh cleanup code from free_fib_info_rcu into a new helper,
> fib_nh_release. Move classid accounting into fib_nh_release which is
> called per fib_nh to make accounting symmetrical with fib_nh_init.
> Export the helper to allow for use with nexthop objects in the
> future.
>
> Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 05/13] ipv6: Create init helper for fib6_nh
2019-03-27 3:29 ` [PATCH net-next 05/13] ipv6: Create init helper for fib6_nh David Ahern
@ 2019-03-27 8:34 ` Ido Schimmel
0 siblings, 0 replies; 29+ messages in thread
From: Ido Schimmel @ 2019-03-27 8:34 UTC (permalink / raw)
To: David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed, David Ahern
On Tue, Mar 26, 2019 at 08:29:34PM -0700, David Ahern wrote:
> +int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
> + struct fib6_config *cfg, gfp_t gfp_flags,
> + struct netlink_ext_ack *extack)
> +{
> + struct net_device *dev = NULL;
> + struct inet6_dev *idev = NULL;
> + int addr_type;
> + int err;
> +
> + err = -ENODEV;
> + if (cfg->fc_ifindex) {
> + dev = dev_get_by_index(net, cfg->fc_ifindex);
> + if (!dev)
> + goto out;
> + idev = in6_dev_get(dev);
> + if (!idev)
> + goto out;
> + }
> +
> + if (cfg->fc_flags & RTNH_F_ONLINK) {
> + if (!dev) {
> + NL_SET_ERR_MSG(extack,
> + "Nexthop device required for onlink");
> + goto out;
> + }
> +
> + if (!(dev->flags & IFF_UP)) {
> + NL_SET_ERR_MSG(extack, "Nexthop device is not up");
> + err = -ENETDOWN;
> + goto out;
> + }
> +
> + fib6_nh->nh_flags |= RTNH_F_ONLINK;
> + }
> +
> + if (cfg->fc_encap) {
> + struct lwtunnel_state *lwtstate;
> +
> + err = lwtunnel_build_state(cfg->fc_encap_type,
> + cfg->fc_encap, AF_INET6, cfg,
> + &lwtstate, extack);
> + if (err)
> + goto out;
> +
> + fib6_nh->nh_lwtstate = lwtstate_get(lwtstate);
> + }
> +
> + fib6_nh->nh_weight = 1;
> +
> + /* We cannot add true routes via loopback here,
> + * they would result in kernel looping; promote them to reject routes
> + */
> + addr_type = ipv6_addr_type(&cfg->fc_dst);
> + if ((cfg->fc_flags & RTF_REJECT) ||
> + (dev && (dev->flags & IFF_LOOPBACK) &&
> + !(addr_type & IPV6_ADDR_LOOPBACK) &&
> + !(cfg->fc_flags & RTF_LOCAL))) {
> + /* hold loopback dev/idev if we haven't done so. */
> + if (dev != net->loopback_dev) {
> + if (dev) {
> + dev_put(dev);
> + in6_dev_put(idev);
> + }
> + dev = net->loopback_dev;
> + dev_hold(dev);
> + idev = in6_dev_get(dev);
> + if (!idev) {
> + err = -ENODEV;
> + goto out;
Same comment as previous patch. Earlier in the function you take a
reference on the LWT object, but don't drop it in error path and instead
rely on the error path of ip6_route_info_create() to take care of that.
> + }
> + }
> + cfg->fc_flags = RTF_REJECT | RTF_NONEXTHOP;
> + goto set_dev;
> + }
> +
> + if (cfg->fc_flags & RTF_GATEWAY) {
> + err = ip6_validate_gw(net, cfg, &dev, &idev, extack);
> + if (err)
> + goto out;
> +
> + fib6_nh->nh_gw = cfg->fc_gateway;
> + }
> +
> + err = -ENODEV;
> + if (!dev)
> + goto out;
> +
> + if (idev->cnf.disable_ipv6) {
> + NL_SET_ERR_MSG(extack, "IPv6 is disabled on nexthop device");
> + err = -EACCES;
> + goto out;
> + }
> +
> + if (!(dev->flags & IFF_UP) && !cfg->fc_ignore_dev_down) {
> + NL_SET_ERR_MSG(extack, "Nexthop device is not up");
> + err = -ENETDOWN;
> + goto out;
> + }
> +
> + if (!(cfg->fc_flags & (RTF_LOCAL | RTF_ANYCAST)) &&
> + !netif_carrier_ok(dev))
> + fib6_nh->nh_flags |= RTNH_F_LINKDOWN;
> +
> +set_dev:
> + fib6_nh->nh_dev = dev;
> + err = 0;
> +out:
> + if (idev)
> + in6_dev_put(idev);
> +
> + if (err && dev)
> + dev_put(dev);
> +
> + return err;
> +}
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 06/13] ipv6: Create cleanup helper for fib6_nh
2019-03-27 3:29 ` [PATCH net-next 06/13] ipv6: Create cleanup " David Ahern
@ 2019-03-27 8:38 ` Ido Schimmel
2019-03-27 14:17 ` David Ahern
0 siblings, 1 reply; 29+ messages in thread
From: Ido Schimmel @ 2019-03-27 8:38 UTC (permalink / raw)
To: David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed, David Ahern
On Tue, Mar 26, 2019 at 08:29:35PM -0700, David Ahern wrote:
> diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
> index 6613d8dbb0e5..db886085369b 100644
> --- a/net/ipv6/ip6_fib.c
> +++ b/net/ipv6/ip6_fib.c
> @@ -199,10 +199,7 @@ void fib6_info_destroy_rcu(struct rcu_head *head)
> free_percpu(f6i->rt6i_pcpu);
> }
>
> - lwtstate_put(f6i->fib6_nh.nh_lwtstate);
> -
> - if (f6i->fib6_nh.nh_dev)
> - dev_put(f6i->fib6_nh.nh_dev);
> + fib6_nh_release(&f6i->fib6_nh);
>
> ip_fib_metrics_put(f6i->fib6_metrics);
>
> diff --git a/net/ipv6/route.c b/net/ipv6/route.c
> index 269ad7b9cdbb..4923a5f88e9b 100644
> --- a/net/ipv6/route.c
> +++ b/net/ipv6/route.c
> @@ -3012,6 +3012,14 @@ int fib6_nh_init(struct net *net, struct fib6_nh *fib6_nh,
> return err;
> }
>
> +void fib6_nh_release(struct fib6_nh *fib6_nh)
> +{
> + if (fib6_nh->nh_dev)
> + dev_put(fib6_nh->nh_dev);
> +
> + lwtstate_put(fib6_nh->nh_lwtstate);
Why did you flip the order?
lwtstate_put()
dev_put()
is symmetric with fib6_nh_init()
> +}
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 07/13] ipv6: Move gateway checks to a fib6_nh setting
2019-03-27 3:29 ` [PATCH net-next 07/13] ipv6: Move gateway checks to a fib6_nh setting David Ahern
@ 2019-03-27 9:08 ` Ido Schimmel
2019-03-27 14:21 ` David Ahern
0 siblings, 1 reply; 29+ messages in thread
From: Ido Schimmel @ 2019-03-27 9:08 UTC (permalink / raw)
To: David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed, David Ahern
On Tue, Mar 26, 2019 at 08:29:36PM -0700, David Ahern wrote:
> @@ -989,8 +989,11 @@ static void ip6_rt_copy_init(struct rt6_info *rt, struct fib6_info *ort)
>
> rt->rt6i_dst = ort->fib6_dst;
> rt->rt6i_idev = dev ? in6_dev_get(dev) : NULL;
> - rt->rt6i_gateway = ort->fib6_nh.nh_gw;
> rt->rt6i_flags = ort->fib6_flags;
> + if (ort->fib6_nh.fib_nh_has_gw) {
> + rt->rt6i_gateway = ort->fib6_nh.nh_gw;
> + rt->rt6i_flags |= RTF_GATEWAY;
Not sure I understand why this line is needed. Can you explain?
> + }
> rt6_set_from(rt, ort);
> #ifdef CONFIG_IPV6_SUBTREES
> rt->rt6i_src = ort->fib6_src;
> @@ -1870,7 +1873,7 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table,
...
> @@ -3134,7 +3138,7 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
> } else
> rt->fib6_prefsrc.plen = 0;
>
> - rt->fib6_flags = cfg->fc_flags;
> + rt->fib6_flags = cfg->fc_flags & ~RTF_GATEWAY;
This is going to break mlxsw. See:
mlxsw_sp_fib6_rt_can_mp()
mlxsw_sp_rt6_is_gateway()
>
> return rt;
> out:
> @@ -3470,7 +3474,8 @@ static struct fib6_info *rt6_get_route_info(struct net *net,
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 08/13] ipv6: Refactor fib6_ignore_linkdown
2019-03-27 3:29 ` [PATCH net-next 08/13] ipv6: Refactor fib6_ignore_linkdown David Ahern
@ 2019-03-27 9:19 ` Ido Schimmel
0 siblings, 0 replies; 29+ messages in thread
From: Ido Schimmel @ 2019-03-27 9:19 UTC (permalink / raw)
To: David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed, David Ahern
On Tue, Mar 26, 2019 at 08:29:37PM -0700, David Ahern wrote:
> From: David Ahern <dsahern@gmail.com>
>
> fib6_ignore_linkdown takes a fib6_info but only looks at the net_device
> and its IPv6 config. Change it to take a net_device over a fib6_info as
> its input argument.
>
> In addition, move it to a header file to make the check inline and usable
> later with IPv4 code without going through the ipv6 stub, and rename to
> ip6_ignore_linkdown since it is only checking the setting based on the
> ipv6 struct on a device.
>
> Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 09/13] ipv6: Change rt6_add_nexthop and rt6_nexthop_info to take fib6_nh
2019-03-27 3:29 ` [PATCH net-next 09/13] ipv6: Change rt6_add_nexthop and rt6_nexthop_info to take fib6_nh David Ahern
@ 2019-03-27 9:24 ` Ido Schimmel
0 siblings, 0 replies; 29+ messages in thread
From: Ido Schimmel @ 2019-03-27 9:24 UTC (permalink / raw)
To: David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed, David Ahern
On Tue, Mar 26, 2019 at 08:29:38PM -0700, David Ahern wrote:
> From: David Ahern <dsahern@gmail.com>
>
> rt6_add_nexthop and rt6_nexthop_info only need the fib6_info for the
> gateway flag and the nexthop weight, and the presence of a gateway is now
> per-nexthop. Update the signatures to take a fib6_nh and nexthop weight
> and better align with the ipv4 versions.
>
> Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 13/13] net: Use common nexthop init and release helpers
2019-03-27 3:29 ` [PATCH net-next 13/13] net: Use common nexthop init and release helpers David Ahern
@ 2019-03-27 9:46 ` Ido Schimmel
2019-03-27 14:22 ` David Ahern
0 siblings, 1 reply; 29+ messages in thread
From: Ido Schimmel @ 2019-03-27 9:46 UTC (permalink / raw)
To: David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed, David Ahern
On Tue, Mar 26, 2019 at 08:29:42PM -0700, David Ahern wrote:
> From: David Ahern <dsahern@gmail.com>
>
> With fib_nh_common in place, move common initialization and release
> code into helpers used by both ipv4 and ipv6. For the moment, the init
> is just the lwt encap and the release is both the net dev reference and
s/net dev/netdev/
> the the lwt state reference. More will be added later.
>
> Signed-off-by: David Ahern <dsahern@gmail.com>
> ---
> include/net/ip_fib.h | 4 ++++
> net/ipv4/fib_semantics.c | 60 +++++++++++++++++++++++++++++++-----------------
I don't really mind, but shouldn't the common parts be placed under
net/core/ ? Like how the FIB rules have a common code in
net/core/fib_rules.c ?
> net/ipv6/route.c | 21 ++++-------------
> 3 files changed, 48 insertions(+), 37 deletions(-)
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 12/13] net: Add fib_nh_common and update fib_nh and fib6_nh
2019-03-27 3:29 ` [PATCH net-next 12/13] net: Add fib_nh_common and update fib_nh and fib6_nh David Ahern
@ 2019-03-27 9:54 ` Ido Schimmel
0 siblings, 0 replies; 29+ messages in thread
From: Ido Schimmel @ 2019-03-27 9:54 UTC (permalink / raw)
To: David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed, David Ahern
On Tue, Mar 26, 2019 at 08:29:41PM -0700, David Ahern wrote:
> From: David Ahern <dsahern@gmail.com>
>
> Add fib_nh_common struct with common nexthop attributes. Convert
> fib_nh and fib6_nh to use it. Use macros to move existing
> fib_nh_* references to the new nh_common.nhc_*.
>
> Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 03/13] ipv4: Create init helper for fib_nh
2019-03-27 8:12 ` Ido Schimmel
@ 2019-03-27 14:17 ` David Ahern
0 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-03-27 14:17 UTC (permalink / raw)
To: Ido Schimmel, David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed
On 3/27/19 2:12 AM, Ido Schimmel wrote:
> On Tue, Mar 26, 2019 at 08:29:32PM -0700, David Ahern wrote:
>> +int fib_nh_init(struct net *net, struct fib_nh *nh,
>> + struct fib_config *cfg, int nh_weight,
>> + struct netlink_ext_ack *extack)
>> +{
>> + int err = -ENOMEM;
>> +
>> + nh->nh_pcpu_rth_output = alloc_percpu(struct rtable __rcu *);
>> + if (!nh->nh_pcpu_rth_output)
>> + goto failure;
>> +
>> + if (cfg->fc_encap) {
>> + struct lwtunnel_state *lwtstate;
>> +
>> + err = -EINVAL;
>> + if (cfg->fc_encap_type == LWTUNNEL_ENCAP_NONE) {
>> + NL_SET_ERR_MSG(extack, "LWT encap type not specified");
>> + goto failure;
>
> This is very confusing and probably error-prone. You call
> alloc_percpu(), but don't free it in the error path and instead rely on
> the call to free_fib_info() in the error path of fib_create_info().
> Better to free it here and NULL-ify the pointer. Then the call to
> rt_fibinfo_free_cpus() in free_fib_info_rcu() is basically a NOP.
>
Sure. I will change it.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 06/13] ipv6: Create cleanup helper for fib6_nh
2019-03-27 8:38 ` Ido Schimmel
@ 2019-03-27 14:17 ` David Ahern
0 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-03-27 14:17 UTC (permalink / raw)
To: Ido Schimmel, David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed
On 3/27/19 2:38 AM, Ido Schimmel wrote:
> Why did you flip the order?
>
> lwtstate_put()
> dev_put()
>
> is symmetric with fib6_nh_init()
no real reason. Will reverse them
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 07/13] ipv6: Move gateway checks to a fib6_nh setting
2019-03-27 9:08 ` Ido Schimmel
@ 2019-03-27 14:21 ` David Ahern
0 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-03-27 14:21 UTC (permalink / raw)
To: Ido Schimmel, David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed
On 3/27/19 3:08 AM, Ido Schimmel wrote:
> On Tue, Mar 26, 2019 at 08:29:36PM -0700, David Ahern wrote:
>> @@ -989,8 +989,11 @@ static void ip6_rt_copy_init(struct rt6_info *rt, struct fib6_info *ort)
>>
>> rt->rt6i_dst = ort->fib6_dst;
>> rt->rt6i_idev = dev ? in6_dev_get(dev) : NULL;
>> - rt->rt6i_gateway = ort->fib6_nh.nh_gw;
>> rt->rt6i_flags = ort->fib6_flags;
>> + if (ort->fib6_nh.fib_nh_has_gw) {
>> + rt->rt6i_gateway = ort->fib6_nh.nh_gw;
>> + rt->rt6i_flags |= RTF_GATEWAY;
>
> Not sure I understand why this line is needed. Can you explain?
The RTF_GATEWAY flag in an rt6_info still has many users.
>
>> + }
>> rt6_set_from(rt, ort);
>> #ifdef CONFIG_IPV6_SUBTREES
>> rt->rt6i_src = ort->fib6_src;
>> @@ -1870,7 +1873,7 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table,
>
> ...
>
>> @@ -3134,7 +3138,7 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
>> } else
>> rt->fib6_prefsrc.plen = 0;
>>
>> - rt->fib6_flags = cfg->fc_flags;
>> + rt->fib6_flags = cfg->fc_flags & ~RTF_GATEWAY;
>
> This is going to break mlxsw. See:
>
> mlxsw_sp_fib6_rt_can_mp()
> mlxsw_sp_rt6_is_gateway()
hmmm... not sure how I missed those. Will rescan the code for any other
missed checks on a fib6_info.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 13/13] net: Use common nexthop init and release helpers
2019-03-27 9:46 ` Ido Schimmel
@ 2019-03-27 14:22 ` David Ahern
0 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-03-27 14:22 UTC (permalink / raw)
To: Ido Schimmel, David Ahern; +Cc: davem, netdev, Jiri Pirko, Saeed Mahameed
On 3/27/19 3:46 AM, Ido Schimmel wrote:
> On Tue, Mar 26, 2019 at 08:29:42PM -0700, David Ahern wrote:
>> From: David Ahern <dsahern@gmail.com>
>>
>> With fib_nh_common in place, move common initialization and release
>> code into helpers used by both ipv4 and ipv6. For the moment, the init
>> is just the lwt encap and the release is both the net dev reference and
>
> s/net dev/netdev/
>
>> the the lwt state reference. More will be added later.
>>
>> Signed-off-by: David Ahern <dsahern@gmail.com>
>> ---
>> include/net/ip_fib.h | 4 ++++
>> net/ipv4/fib_semantics.c | 60 +++++++++++++++++++++++++++++++-----------------
>
> I don't really mind, but shouldn't the common parts be placed under
> net/core/ ? Like how the FIB rules have a common code in
> net/core/fib_rules.c ?
>
net/core is for common code across networking protocols.
Code common between just IPv4 and IPv6 is currently in net/ipv4
^ permalink raw reply [flat|nested] 29+ messages in thread
end of thread, other threads:[~2019-03-27 14:22 UTC | newest]
Thread overview: 29+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-03-27 3:29 [PATCH net-next 00/13] net: Move fib_nh and fib6_nh to a common struct David Ahern
2019-03-27 3:29 ` [PATCH net-next 01/13] ipv4: Define fib_get_nhs when CONFIG_IP_ROUTE_MULTIPATH is disabled David Ahern
2019-03-27 7:40 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 02/13] ipv4: Move IN_DEV_IGNORE_ROUTES_WITH_LINKDOWN to helper David Ahern
2019-03-27 7:44 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 03/13] ipv4: Create init helper for fib_nh David Ahern
2019-03-27 8:12 ` Ido Schimmel
2019-03-27 14:17 ` David Ahern
2019-03-27 3:29 ` [PATCH net-next 04/13] ipv4: Create cleanup " David Ahern
2019-03-27 8:17 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 05/13] ipv6: Create init helper for fib6_nh David Ahern
2019-03-27 8:34 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 06/13] ipv6: Create cleanup " David Ahern
2019-03-27 8:38 ` Ido Schimmel
2019-03-27 14:17 ` David Ahern
2019-03-27 3:29 ` [PATCH net-next 07/13] ipv6: Move gateway checks to a fib6_nh setting David Ahern
2019-03-27 9:08 ` Ido Schimmel
2019-03-27 14:21 ` David Ahern
2019-03-27 3:29 ` [PATCH net-next 08/13] ipv6: Refactor fib6_ignore_linkdown David Ahern
2019-03-27 9:19 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 09/13] ipv6: Change rt6_add_nexthop and rt6_nexthop_info to take fib6_nh David Ahern
2019-03-27 9:24 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 10/13] ipv4: Rename fib_nh entries David Ahern
2019-03-27 3:29 ` [PATCH net-next 11/13] ipv6: Rename fib6_nh entries David Ahern
2019-03-27 3:29 ` [PATCH net-next 12/13] net: Add fib_nh_common and update fib_nh and fib6_nh David Ahern
2019-03-27 9:54 ` Ido Schimmel
2019-03-27 3:29 ` [PATCH net-next 13/13] net: Use common nexthop init and release helpers David Ahern
2019-03-27 9:46 ` Ido Schimmel
2019-03-27 14:22 ` David Ahern
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.