* [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
@ 2019-05-30 3:07 David Ahern
2019-05-30 3:07 ` [PATCH net-next 1/7] ipv4: Use accessors for fib_info nexthop data David Ahern
` (7 more replies)
0 siblings, 8 replies; 29+ messages in thread
From: David Ahern @ 2019-05-30 3:07 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, saeedm, kafai, weiwan, David Ahern
From: David Ahern <dsahern@gmail.com>
This sets adds 'struct nexthop' to fib_info and fib6_info. IPv4
already handles multiple fib_nh entries in a single fib_info, so
the conversion to use a nexthop struct is fairly mechanical. IPv6
using a nexthop struct with a fib6_info impacts a lot of core logic
which is built around the assumption of a single, builtin fib6_nh
per fib6_info. To make this easier to review, this set adds
nexthop to fib6_info and adds checks in most places fib6_info is
used. The next set finishes the IPv6 conversion, walking through
the places that need to consider all fib6_nh within a nexthop struct.
Offload drivers - mlx5, mlxsw and rocker - are changed to fail FIB
entries using nexthop objects. That limitation can be removed once
the drivers are updated to properly support separate nexthops.
This set starts by adding accessors for fib_nh and fib_nhs in a
fib_info. This makes it easier to extract the number of nexthops
in the fib entry and a specific fib_nh once the entry references
a struct nexthop. Patch 2 converts more of IPv4 code to use
fib_nh_common allowing a struct nexthop to use a fib6_nh with an
IPv4 entry.
Patches 3 and 4 add 'struct nexthop' to fib{6}_info and update
references to both take a different path when it is set. New
exported functions are added to the nexthop code to validate a
nexthop struct when configured for use with a fib entry. IPv4
is allowed to use a nexthop with either v4 or v6 entries. IPv6
is limited to v6 entries only. In both cases list_heads track
the fib entries using a nexthop struct for fast correlation on
events (e.g., device events or nexthop events like delete or
replace).
The last 3 patches add hooks to drivers listening for FIB
notificationas. All 3 of them reject the routes as unsupported,
returning an error message to the user via extack. For mlxsw
at least this is a stop gap measure until the driver is updated for
proper support.
David Ahern (7):
ipv4: Use accessors for fib_info nexthop data
ipv4: Prepare for fib6_nh from a nexthop object
ipv4: Plumb support for nexthop object in a fib_info
ipv6: Plumb support for nexthop object in a fib6_info
mlxsw: Fail attempts to use routes with nexthop objects
mlx5: Fail attempts to use routes with nexthop objects
rocker: Fail attempts to use routes with nexthop objects
drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c | 33 ++-
.../net/ethernet/mellanox/mlxsw/spectrum_router.c | 33 ++-
drivers/net/ethernet/rocker/rocker_main.c | 4 +
drivers/net/ethernet/rocker/rocker_ofdpa.c | 25 +-
include/net/ip6_fib.h | 11 +-
include/net/ip6_route.h | 13 +-
include/net/ip_fib.h | 25 +-
include/net/nexthop.h | 113 +++++++++
net/core/filter.c | 3 +-
net/ipv4/fib_frontend.c | 15 +-
net/ipv4/fib_lookup.h | 1 +
net/ipv4/fib_rules.c | 8 +-
net/ipv4/fib_semantics.c | 257 ++++++++++++++-------
net/ipv4/fib_trie.c | 38 ++-
net/ipv4/nexthop.c | 111 ++++++++-
net/ipv4/route.c | 5 +-
net/ipv6/addrconf.c | 5 +
net/ipv6/ip6_fib.c | 22 +-
net/ipv6/ndisc.c | 3 +-
net/ipv6/route.c | 156 +++++++++++--
20 files changed, 706 insertions(+), 175 deletions(-)
--
2.11.0
^ permalink raw reply [flat|nested] 29+ messages in thread
* [PATCH net-next 1/7] ipv4: Use accessors for fib_info nexthop data
2019-05-30 3:07 [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info David Ahern
@ 2019-05-30 3:07 ` David Ahern
2019-05-30 3:07 ` [PATCH net-next 2/7] ipv4: Prepare for fib6_nh from a nexthop object David Ahern
` (6 subsequent siblings)
7 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-05-30 3:07 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, saeedm, kafai, weiwan, David Ahern
From: David Ahern <dsahern@gmail.com>
Use helpers to access fib_nh and fib_nhs fields of a fib_info. Drop the
fib_dev macro which is an alias for the first nexthop. Replacements:
fib_dev --> fib_info_nh(fi, 0)->fib_nh_dev
fi->fib_nh --> fib_info_nh(fi, 0)
fi->fib_nh[i] --> fib_info_nh(fi, i)
fi->fib_nhs --> fib_info_num_path(fi)
Move the existing fib_info_nhc to nexthop.h and define the new ones
there. A later patch adds a check if a fib_info uses a nexthop object,
and defining the helpers in nexthop.h avoid circular header
dependencies.
After this all remaining open coded references to fi->fib_nhs and
fi->fib_nh are in:
- fib_create_info and helpers used to lookup an existing fib_info
entry, and
- the netdev event functions fib_sync_down_dev and fib_sync_up.
The latter two will not be reused for nexthops, and the fib_create_info
will be updated to handle a nexthop in a fib_info.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c | 29 ++++++----
.../net/ethernet/mellanox/mlxsw/spectrum_router.c | 19 ++++---
drivers/net/ethernet/rocker/rocker_ofdpa.c | 25 +++++---
include/net/ip_fib.h | 6 --
include/net/nexthop.h | 15 +++++
net/core/filter.c | 3 +-
net/ipv4/fib_frontend.c | 11 ++--
net/ipv4/fib_lookup.h | 1 +
net/ipv4/fib_rules.c | 8 ++-
net/ipv4/fib_semantics.c | 66 ++++++++++++----------
net/ipv4/fib_trie.c | 26 +++++----
net/ipv4/route.c | 3 +-
12 files changed, 132 insertions(+), 80 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c
index 8212bfd05733..2cbfaa8da7fc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c
@@ -2,6 +2,7 @@
/* Copyright (c) 2019 Mellanox Technologies. */
#include <linux/netdevice.h>
+#include <net/nexthop.h>
#include "lag.h"
#include "lag_mp.h"
#include "mlx5_core.h"
@@ -110,6 +111,8 @@ static void mlx5_lag_fib_route_event(struct mlx5_lag *ldev,
struct fib_info *fi)
{
struct lag_mp *mp = &ldev->lag_mp;
+ struct fib_nh *fib_nh0, *fib_nh1;
+ unsigned int nhs;
/* Handle delete event */
if (event == FIB_EVENT_ENTRY_DEL) {
@@ -120,9 +123,11 @@ static void mlx5_lag_fib_route_event(struct mlx5_lag *ldev,
}
/* Handle add/replace event */
- if (fi->fib_nhs == 1) {
+ nhs = fib_info_num_path(fi);
+ if (nhs == 1) {
if (__mlx5_lag_is_active(ldev)) {
- struct net_device *nh_dev = fi->fib_nh[0].fib_nh_dev;
+ struct fib_nh *nh = fib_info_nh(fi, 0);
+ struct net_device *nh_dev = nh->fib_nh_dev;
int i = mlx5_lag_dev_get_netdev_idx(ldev, nh_dev);
mlx5_lag_set_port_affinity(ldev, ++i);
@@ -130,14 +135,16 @@ static void mlx5_lag_fib_route_event(struct mlx5_lag *ldev,
return;
}
- if (fi->fib_nhs != 2)
+ if (nhs != 2)
return;
/* Verify next hops are ports of the same hca */
- 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)) {
+ fib_nh0 = fib_info_nh(fi, 0);
+ fib_nh1 = fib_info_nh(fi, 1);
+ if (!(fib_nh0->fib_nh_dev == ldev->pf[0].netdev &&
+ fib_nh1->fib_nh_dev == ldev->pf[1].netdev) &&
+ !(fib_nh0->fib_nh_dev == ldev->pf[1].netdev &&
+ fib_nh1->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;
}
@@ -174,7 +181,7 @@ static void mlx5_lag_fib_nexthop_event(struct mlx5_lag *ldev,
mlx5_lag_set_port_affinity(ldev, i);
}
} else if (event == FIB_EVENT_NH_ADD &&
- fi->fib_nhs == 2) {
+ fib_info_num_path(fi) == 2) {
mlx5_lag_set_port_affinity(ldev, 0);
}
}
@@ -238,6 +245,7 @@ static int mlx5_lag_fib_event(struct notifier_block *nb,
struct mlx5_fib_event_work *fib_work;
struct fib_entry_notifier_info *fen_info;
struct fib_nh_notifier_info *fnh_info;
+ struct net_device *fib_dev;
struct fib_info *fi;
if (info->family != AF_INET)
@@ -254,8 +262,9 @@ static int mlx5_lag_fib_event(struct notifier_block *nb,
fen_info = container_of(info, struct fib_entry_notifier_info,
info);
fi = fen_info->fi;
- if (fi->fib_dev != ldev->pf[0].netdev &&
- fi->fib_dev != ldev->pf[1].netdev) {
+ fib_dev = fib_info_nh(fen_info->fi, 0)->fib_nh_dev;
+ if (fib_dev != ldev->pf[0].netdev &&
+ fib_dev != ldev->pf[1].netdev) {
return NOTIFY_DONE;
}
fib_work = mlx5_lag_init_fib_work(ldev, event);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 0ec52be7cc33..4f781358aef1 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -21,6 +21,7 @@
#include <net/arp.h>
#include <net/ip_fib.h>
#include <net/ip6_fib.h>
+#include <net/nexthop.h>
#include <net/fib_rules.h>
#include <net/ip_tunnels.h>
#include <net/l3mdev.h>
@@ -3816,23 +3817,25 @@ 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)
+ struct fib_info *fi)
{
- return fi->fib_nh->fib_nh_scope == RT_SCOPE_LINK ||
- mlxsw_sp_nexthop4_ipip_type(mlxsw_sp, fi->fib_nh, NULL);
+ const struct fib_nh *nh = fib_info_nh(fi, 0);
+
+ return nh->fib_nh_scope == RT_SCOPE_LINK ||
+ mlxsw_sp_nexthop4_ipip_type(mlxsw_sp, nh, NULL);
}
static struct mlxsw_sp_nexthop_group *
mlxsw_sp_nexthop4_group_create(struct mlxsw_sp *mlxsw_sp, struct fib_info *fi)
{
+ unsigned int nhs = fib_info_num_path(fi);
struct mlxsw_sp_nexthop_group *nh_grp;
struct mlxsw_sp_nexthop *nh;
struct fib_nh *fib_nh;
int i;
int err;
- nh_grp = kzalloc(struct_size(nh_grp, nexthops, fi->fib_nhs),
- GFP_KERNEL);
+ nh_grp = kzalloc(struct_size(nh_grp, nexthops, nhs), GFP_KERNEL);
if (!nh_grp)
return ERR_PTR(-ENOMEM);
nh_grp->priv = fi;
@@ -3840,11 +3843,11 @@ mlxsw_sp_nexthop4_group_create(struct mlxsw_sp *mlxsw_sp, struct fib_info *fi)
nh_grp->neigh_tbl = &arp_tbl;
nh_grp->gateway = mlxsw_sp_fi_is_gateway(mlxsw_sp, fi);
- nh_grp->count = fi->fib_nhs;
+ nh_grp->count = nhs;
fib_info_hold(fi);
for (i = 0; i < nh_grp->count; i++) {
nh = &nh_grp->nexthops[i];
- fib_nh = &fi->fib_nh[i];
+ fib_nh = fib_info_nh(fi, i);
err = mlxsw_sp_nexthop4_init(mlxsw_sp, nh_grp, nh, fib_nh);
if (err)
goto err_nexthop4_init;
@@ -4282,9 +4285,9 @@ mlxsw_sp_fib4_entry_type_set(struct mlxsw_sp *mlxsw_sp,
const struct fib_entry_notifier_info *fen_info,
struct mlxsw_sp_fib_entry *fib_entry)
{
+ struct net_device *dev = fib_info_nh(fen_info->fi, 0)->fib_nh_dev;
union mlxsw_sp_l3addr dip = { .addr4 = htonl(fen_info->dst) };
u32 tb_id = mlxsw_sp_fix_tb_id(fen_info->tb_id);
- struct net_device *dev = fen_info->fi->fib_dev;
struct mlxsw_sp_ipip_entry *ipip_entry;
struct fib_info *fi = fen_info->fi;
diff --git a/drivers/net/ethernet/rocker/rocker_ofdpa.c b/drivers/net/ethernet/rocker/rocker_ofdpa.c
index 30a49802fb51..47ed9d41047f 100644
--- a/drivers/net/ethernet/rocker/rocker_ofdpa.c
+++ b/drivers/net/ethernet/rocker/rocker_ofdpa.c
@@ -22,6 +22,7 @@
#include <net/neighbour.h>
#include <net/switchdev.h>
#include <net/ip_fib.h>
+#include <net/nexthop.h>
#include <net/arp.h>
#include "rocker.h"
@@ -2286,8 +2287,8 @@ static int ofdpa_port_fib_ipv4(struct ofdpa_port *ofdpa_port, __be32 dst,
/* XXX support ECMP */
- nh = fi->fib_nh;
- nh_on_port = (fi->fib_dev == ofdpa_port->dev);
+ nh = fib_info_nh(fi, 0);
+ nh_on_port = (nh->fib_nh_dev == ofdpa_port->dev);
has_gw = !!nh->fib_nh_gw4;
if (has_gw && nh_on_port) {
@@ -2737,11 +2738,13 @@ static int ofdpa_fib4_add(struct rocker *rocker,
{
struct ofdpa *ofdpa = rocker->wpriv;
struct ofdpa_port *ofdpa_port;
+ struct fib_nh *nh;
int err;
if (ofdpa->fib_aborted)
return 0;
- ofdpa_port = ofdpa_port_dev_lower_find(fen_info->fi->fib_dev, rocker);
+ nh = fib_info_nh(fen_info->fi, 0);
+ ofdpa_port = ofdpa_port_dev_lower_find(nh->fib_nh_dev, rocker);
if (!ofdpa_port)
return 0;
err = ofdpa_port_fib_ipv4(ofdpa_port, htonl(fen_info->dst),
@@ -2749,7 +2752,7 @@ static int ofdpa_fib4_add(struct rocker *rocker,
fen_info->tb_id, 0);
if (err)
return err;
- fen_info->fi->fib_nh->fib_nh_flags |= RTNH_F_OFFLOAD;
+ nh->fib_nh_flags |= RTNH_F_OFFLOAD;
return 0;
}
@@ -2758,13 +2761,15 @@ static int ofdpa_fib4_del(struct rocker *rocker,
{
struct ofdpa *ofdpa = rocker->wpriv;
struct ofdpa_port *ofdpa_port;
+ struct fib_nh *nh;
if (ofdpa->fib_aborted)
return 0;
- ofdpa_port = ofdpa_port_dev_lower_find(fen_info->fi->fib_dev, rocker);
+ nh = fib_info_nh(fen_info->fi, 0);
+ ofdpa_port = ofdpa_port_dev_lower_find(nh->fib_nh_dev, rocker);
if (!ofdpa_port)
return 0;
- fen_info->fi->fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
+ 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);
@@ -2784,14 +2789,16 @@ static void ofdpa_fib4_abort(struct rocker *rocker)
spin_lock_irqsave(&ofdpa->flow_tbl_lock, flags);
hash_for_each_safe(ofdpa->flow_tbl, bkt, tmp, flow_entry, entry) {
+ struct fib_nh *nh;
+
if (flow_entry->key.tbl_id !=
ROCKER_OF_DPA_TABLE_ID_UNICAST_ROUTING)
continue;
- ofdpa_port = ofdpa_port_dev_lower_find(flow_entry->fi->fib_dev,
- rocker);
+ nh = fib_info_nh(flow_entry->fi, 0);
+ ofdpa_port = ofdpa_port_dev_lower_find(nh->fib_nh_dev, rocker);
if (!ofdpa_port)
continue;
- flow_entry->fi->fib_nh->fib_nh_flags &= ~RTNH_F_OFFLOAD;
+ 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 70ba0302c8c9..42b1a806f6f5 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -153,7 +153,6 @@ struct fib_info {
bool nh_updated;
struct rcu_head rcu;
struct fib_nh fib_nh[0];
-#define fib_dev fib_nh[0].fib_nh_dev
};
@@ -190,11 +189,6 @@ struct fib_result_nl {
int err;
};
-static inline struct fib_nh_common *fib_info_nhc(struct fib_info *fi, int nhsel)
-{
- return &fi->fib_nh[nhsel].nh_common;
-}
-
#ifdef CONFIG_IP_MULTIPLE_TABLES
#define FIB_TABLE_HASHSZ 256
#else
diff --git a/include/net/nexthop.h b/include/net/nexthop.h
index 6e1b8f53624c..e501d77b82c8 100644
--- a/include/net/nexthop.h
+++ b/include/net/nexthop.h
@@ -192,4 +192,19 @@ static inline bool nexthop_is_blackhole(const struct nexthop *nh)
nhi = rcu_dereference_rtnl(nh->nh_info);
return nhi->reject_nh;
}
+
+static inline unsigned int fib_info_num_path(const struct fib_info *fi)
+{
+ return fi->fib_nhs;
+}
+
+static inline struct fib_nh_common *fib_info_nhc(struct fib_info *fi, int nhsel)
+{
+ return &fi->fib_nh[nhsel].nh_common;
+}
+
+static inline struct fib_nh *fib_info_nh(struct fib_info *fi, int nhsel)
+{
+ return &fi->fib_nh[nhsel];
+}
#endif
diff --git a/net/core/filter.c b/net/core/filter.c
index 55bfc941d17a..2ae72bbfa6d2 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -66,6 +66,7 @@
#include <net/inet_hashtables.h>
#include <net/inet6_hashtables.h>
#include <net/ip_fib.h>
+#include <net/nexthop.h>
#include <net/flow.h>
#include <net/arp.h>
#include <net/ipv6.h>
@@ -4674,7 +4675,7 @@ static int bpf_ipv4_fib_lookup(struct net *net, struct bpf_fib_lookup *params,
if (res.type != RTN_UNICAST)
return BPF_FIB_LKUP_RET_NOT_FWDED;
- if (res.fi->fib_nhs > 1)
+ if (fib_info_num_path(res.fi) > 1)
fib_select_path(net, &res, &fl4, NULL);
if (check_mtu) {
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 76055c66326a..ab369959ce0b 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -43,6 +43,7 @@
#include <net/sock.h>
#include <net/arp.h>
#include <net/ip_fib.h>
+#include <net/nexthop.h>
#include <net/rtnetlink.h>
#include <net/xfrm.h>
#include <net/l3mdev.h>
@@ -234,7 +235,9 @@ static inline unsigned int __inet_dev_addr_type(struct net *net,
if (table) {
ret = RTN_UNICAST;
if (!fib_table_lookup(table, &fl4, &res, FIB_LOOKUP_NOREF)) {
- if (!dev || dev == res.fi->fib_dev)
+ struct fib_nh *nh = fib_info_nh(res.fi, 0);
+
+ if (!dev || dev == nh->fib_nh_dev)
ret = res.type;
}
}
@@ -321,8 +324,8 @@ bool fib_info_nh_uses_dev(struct fib_info *fi, const struct net_device *dev)
#ifdef CONFIG_IP_ROUTE_MULTIPATH
int ret;
- for (ret = 0; ret < fi->fib_nhs; ret++) {
- struct fib_nh *nh = &fi->fib_nh[ret];
+ for (ret = 0; ret < fib_info_num_path(fi); ret++) {
+ const struct fib_nh *nh = fib_info_nh(fi, ret);
if (nh->fib_nh_dev == dev) {
dev_match = true;
@@ -333,7 +336,7 @@ bool fib_info_nh_uses_dev(struct fib_info *fi, const struct net_device *dev)
}
}
#else
- if (fi->fib_nh[0].fib_nh_dev == dev)
+ if (fib_info_nh(fi, 0)->fib_nh_dev == dev)
dev_match = true;
#endif
diff --git a/net/ipv4/fib_lookup.h b/net/ipv4/fib_lookup.h
index 7945f0534db7..a68b5e21ec51 100644
--- a/net/ipv4/fib_lookup.h
+++ b/net/ipv4/fib_lookup.h
@@ -5,6 +5,7 @@
#include <linux/types.h>
#include <linux/list.h>
#include <net/ip_fib.h>
+#include <net/nexthop.h>
struct fib_alias {
struct hlist_node fa_list;
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index cfec3af54c8d..ab06fd73b343 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -31,6 +31,7 @@
#include <net/route.h>
#include <net/tcp.h>
#include <net/ip_fib.h>
+#include <net/nexthop.h>
#include <net/fib_rules.h>
struct fib4_rule {
@@ -145,8 +146,11 @@ static bool fib4_rule_suppress(struct fib_rule *rule, struct fib_lookup_arg *arg
struct fib_result *result = (struct fib_result *) arg->result;
struct net_device *dev = NULL;
- if (result->fi)
- dev = result->fi->fib_dev;
+ if (result->fi) {
+ struct fib_nh *nh = fib_info_nh(result->fi, 0);
+
+ dev = nh->fib_nh_dev;
+ }
/* do not accept result if the route does
* not meet the required prefix length
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 78648072783e..a37ff07718a8 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -42,6 +42,7 @@
#include <net/sock.h>
#include <net/ip_fib.h>
#include <net/ip6_fib.h>
+#include <net/nexthop.h>
#include <net/netlink.h>
#include <net/rtnh.h>
#include <net/lwtunnel.h>
@@ -65,13 +66,13 @@ static struct hlist_head fib_info_devhash[DEVINDEX_HASHSIZE];
#define for_nexthops(fi) { \
int nhsel; const struct fib_nh *nh; \
for (nhsel = 0, nh = (fi)->fib_nh; \
- nhsel < (fi)->fib_nhs; \
+ nhsel < fib_info_num_path((fi)); \
nh++, nhsel++)
#define change_nexthops(fi) { \
int nhsel; struct fib_nh *nexthop_nh; \
for (nhsel = 0, nexthop_nh = (struct fib_nh *)((fi)->fib_nh); \
- nhsel < (fi)->fib_nhs; \
+ nhsel < fib_info_num_path((fi)); \
nexthop_nh++, nhsel++)
#else /* CONFIG_IP_ROUTE_MULTIPATH */
@@ -271,11 +272,13 @@ void fib_release_info(struct fib_info *fi)
spin_unlock_bh(&fib_info_lock);
}
-static inline int nh_comp(const struct fib_info *fi, const struct fib_info *ofi)
+static inline int nh_comp(struct fib_info *fi, struct fib_info *ofi)
{
- const struct fib_nh *onh = ofi->fib_nh;
+ const struct fib_nh *onh;
for_nexthops(fi) {
+ onh = fib_info_nh(ofi, nhsel);
+
if (nh->fib_nh_oif != onh->fib_nh_oif ||
nh->fib_nh_gw_family != onh->fib_nh_gw_family ||
nh->fib_nh_scope != onh->fib_nh_scope ||
@@ -296,8 +299,6 @@ static inline int nh_comp(const struct fib_info *fi, const struct fib_info *ofi)
if (nh->fib_nh_gw_family == AF_INET6 &&
ipv6_addr_cmp(&nh->fib_nh_gw6, &onh->fib_nh_gw6))
return -1;
-
- onh++;
} endfor_nexthops(fi);
return 0;
}
@@ -326,7 +327,7 @@ static inline unsigned int fib_info_hashfn(const struct fib_info *fi)
return (val ^ (val >> 7) ^ (val >> 12)) & mask;
}
-static struct fib_info *fib_find_info(const struct fib_info *nfi)
+static struct fib_info *fib_find_info(struct fib_info *nfi)
{
struct hlist_head *head;
struct fib_info *fi;
@@ -390,13 +391,14 @@ static inline size_t fib_nlmsg_size(struct fib_info *fi)
+ nla_total_size(4) /* RTA_PRIORITY */
+ nla_total_size(4) /* RTA_PREFSRC */
+ nla_total_size(TCP_CA_NAME_MAX); /* RTAX_CC_ALGO */
+ unsigned int nhs = fib_info_num_path(fi);
/* space for nested metrics */
payload += nla_total_size((RTAX_MAX * nla_total_size(4)));
- if (fi->fib_nhs) {
+ if (nhs) {
size_t nh_encapsize = 0;
- /* Also handles the special case fib_nhs == 1 */
+ /* Also handles the special case nhs == 1 */
/* each nexthop is packed in an attribute */
size_t nhsize = nla_total_size(sizeof(struct rtnexthop));
@@ -416,8 +418,7 @@ static inline size_t fib_nlmsg_size(struct fib_info *fi)
} endfor_nexthops(fi);
/* all nexthops are packed in a nested attribute */
- payload += nla_total_size((fi->fib_nhs * nhsize) +
- nh_encapsize);
+ payload += nla_total_size((nhs * nhsize) + nh_encapsize);
}
@@ -584,6 +585,7 @@ static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh,
{
struct net *net = fi->fib_net;
struct fib_config fib_cfg;
+ struct fib_nh *nh;
int ret;
change_nexthops(fi) {
@@ -646,24 +648,25 @@ static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh,
} endfor_nexthops(fi);
ret = -EINVAL;
- if (cfg->fc_oif && fi->fib_nh->fib_nh_oif != cfg->fc_oif) {
+ nh = fib_info_nh(fi, 0);
+ if (cfg->fc_oif && 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_family) {
- if (cfg->fc_gw_family != fi->fib_nh->fib_nh_gw_family ||
+ if (cfg->fc_gw_family != nh->fib_nh_gw_family ||
(cfg->fc_gw_family == AF_INET &&
- fi->fib_nh->fib_nh_gw4 != cfg->fc_gw4) ||
+ nh->fib_nh_gw4 != cfg->fc_gw4) ||
(cfg->fc_gw_family == AF_INET6 &&
- ipv6_addr_cmp(&fi->fib_nh->fib_nh_gw6, &cfg->fc_gw6))) {
+ ipv6_addr_cmp(&nh->fib_nh_gw6, &cfg->fc_gw6))) {
NL_SET_ERR_MSG(extack,
"Nexthop gateway does not match RTA_GATEWAY or RTA_VIA");
goto errout;
}
}
#ifdef CONFIG_IP_ROUTE_CLASSID
- if (cfg->fc_flow && fi->fib_nh->nh_tclassid != cfg->fc_flow) {
+ if (cfg->fc_flow && nh->nh_tclassid != cfg->fc_flow) {
NL_SET_ERR_MSG(extack,
"Nexthop class id does not match RTA_FLOW");
goto errout;
@@ -679,7 +682,7 @@ static void fib_rebalance(struct fib_info *fi)
int total;
int w;
- if (fi->fib_nhs < 2)
+ if (fib_info_num_path(fi) < 2)
return;
total = 0;
@@ -761,27 +764,29 @@ int fib_nh_match(struct fib_config *cfg, struct fib_info *fi,
return 1;
if (cfg->fc_oif || cfg->fc_gw_family) {
+ struct fib_nh *nh = fib_info_nh(fi, 0);
+
if (cfg->fc_encap) {
if (fib_encap_match(cfg->fc_encap_type, cfg->fc_encap,
- fi->fib_nh, cfg, extack))
+ nh, cfg, extack))
return 1;
}
#ifdef CONFIG_IP_ROUTE_CLASSID
if (cfg->fc_flow &&
- cfg->fc_flow != fi->fib_nh->nh_tclassid)
+ cfg->fc_flow != nh->nh_tclassid)
return 1;
#endif
- if ((cfg->fc_oif && cfg->fc_oif != fi->fib_nh->fib_nh_oif) ||
+ if ((cfg->fc_oif && cfg->fc_oif != nh->fib_nh_oif) ||
(cfg->fc_gw_family &&
- cfg->fc_gw_family != fi->fib_nh->fib_nh_gw_family))
+ cfg->fc_gw_family != nh->fib_nh_gw_family))
return 1;
if (cfg->fc_gw_family == AF_INET &&
- cfg->fc_gw4 != fi->fib_nh->fib_nh_gw4)
+ cfg->fc_gw4 != nh->fib_nh_gw4)
return 1;
if (cfg->fc_gw_family == AF_INET6 &&
- ipv6_addr_cmp(&cfg->fc_gw6, &fi->fib_nh->fib_nh_gw6))
+ ipv6_addr_cmp(&cfg->fc_gw6, &nh->fib_nh_gw6))
return 1;
return 0;
@@ -1366,7 +1371,7 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
goto err_inval;
}
nh->fib_nh_scope = RT_SCOPE_NOWHERE;
- nh->fib_nh_dev = dev_get_by_index(net, fi->fib_nh->fib_nh_oif);
+ nh->fib_nh_dev = dev_get_by_index(net, nh->fib_nh_oif);
err = -ENODEV;
if (!nh->fib_nh_dev)
goto failure;
@@ -1583,6 +1588,7 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event,
u32 tb_id, u8 type, __be32 dst, int dst_len, u8 tos,
struct fib_info *fi, unsigned int flags)
{
+ unsigned int nhs = fib_info_num_path(fi);
struct nlmsghdr *nlh;
struct rtmsg *rtm;
@@ -1618,8 +1624,8 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event,
if (fi->fib_prefsrc &&
nla_put_in_addr(skb, RTA_PREFSRC, fi->fib_prefsrc))
goto nla_put_failure;
- if (fi->fib_nhs == 1) {
- struct fib_nh *nh = &fi->fib_nh[0];
+ if (nhs == 1) {
+ const struct fib_nh *nh = fib_info_nh(fi, 0);
unsigned char flags = 0;
if (fib_nexthop_info(skb, &nh->nh_common, &flags, false) < 0)
@@ -1838,6 +1844,7 @@ static void fib_select_default(const struct flowi4 *flp, struct fib_result *res)
hlist_for_each_entry_rcu(fa, fa_head, fa_list) {
struct fib_info *next_fi = fa->fa_info;
+ struct fib_nh *nh;
if (fa->fa_slen != slen)
continue;
@@ -1859,8 +1866,9 @@ 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].fib_nh_gw4 ||
- next_fi->fib_nh[0].fib_nh_scope != RT_SCOPE_LINK)
+
+ nh = fib_info_nh(next_fi, 0);
+ if (!nh->fib_nh_gw4 || nh->fib_nh_scope != RT_SCOPE_LINK)
continue;
fib_alias_accessed(fa);
@@ -2024,7 +2032,7 @@ void fib_select_path(struct net *net, struct fib_result *res,
goto check_saddr;
#ifdef CONFIG_IP_ROUTE_MULTIPATH
- if (res->fi->fib_nhs > 1) {
+ if (fib_info_num_path(res->fi) > 1) {
int h = fib_multipath_hash(net, fl4, skb, NULL);
fib_select_multipath(res, h);
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index b53ecef89d59..5c8a4d21b8e0 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1469,7 +1469,7 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi4 *flp,
}
if (fi->fib_flags & RTNH_F_DEAD)
continue;
- for (nhsel = 0; nhsel < fi->fib_nhs; nhsel++) {
+ for (nhsel = 0; nhsel < fib_info_num_path(fi); nhsel++) {
struct fib_nh_common *nhc = fib_info_nhc(fi, nhsel);
if (nhc->nhc_flags & RTNH_F_DEAD)
@@ -2717,14 +2717,18 @@ static void fib_route_seq_stop(struct seq_file *seq, void *v)
rcu_read_unlock();
}
-static unsigned int fib_flag_trans(int type, __be32 mask, const struct fib_info *fi)
+static unsigned int fib_flag_trans(int type, __be32 mask, struct fib_info *fi)
{
unsigned int flags = 0;
if (type == RTN_UNREACHABLE || type == RTN_PROHIBIT)
flags = RTF_REJECT;
- if (fi && fi->fib_nh->fib_nh_gw4)
- flags |= RTF_GATEWAY;
+ if (fi) {
+ const struct fib_nh *nh = fib_info_nh(fi, 0);
+
+ if (nh->fib_nh_gw4)
+ flags |= RTF_GATEWAY;
+ }
if (mask == htonl(0xFFFFFFFF))
flags |= RTF_HOST;
flags |= RTF_UP;
@@ -2755,7 +2759,7 @@ static int fib_route_seq_show(struct seq_file *seq, void *v)
prefix = htonl(l->key);
hlist_for_each_entry_rcu(fa, &l->leaf, fa_list) {
- const struct fib_info *fi = fa->fa_info;
+ struct fib_info *fi = fa->fa_info;
__be32 mask = inet_make_mask(KEYLENGTH - fa->fa_slen);
unsigned int flags = fib_flag_trans(fa->fa_type, mask, fi);
@@ -2768,26 +2772,28 @@ static int fib_route_seq_show(struct seq_file *seq, void *v)
seq_setwidth(seq, 127);
- if (fi)
+ if (fi) {
+ struct fib_nh *nh = fib_info_nh(fi, 0);
+
seq_printf(seq,
"%s\t%08X\t%08X\t%04X\t%d\t%u\t"
"%d\t%08X\t%d\t%u\t%u",
- fi->fib_dev ? fi->fib_dev->name : "*",
+ nh->fib_nh_dev ? nh->fib_nh_dev->name : "*",
prefix,
- fi->fib_nh->fib_nh_gw4, flags, 0, 0,
+ nh->fib_nh_gw4, flags, 0, 0,
fi->fib_priority,
mask,
(fi->fib_advmss ?
fi->fib_advmss + 40 : 0),
fi->fib_window,
fi->fib_rtt >> 3);
- else
+ } else {
seq_printf(seq,
"*\t%08X\t%08X\t%04X\t%d\t%u\t"
"%d\t%08X\t%d\t%u\t%u",
prefix, 0, flags, 0, 0, 0,
mask, 0, 0, 0);
-
+ }
seq_pad(seq, '\n');
}
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 11ddc276776e..05a6a8ecb574 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -99,6 +99,7 @@
#include <net/inetpeer.h>
#include <net/sock.h>
#include <net/ip_fib.h>
+#include <net/nexthop.h>
#include <net/arp.h>
#include <net/tcp.h>
#include <net/icmp.h>
@@ -1950,7 +1951,7 @@ static int ip_mkroute_input(struct sk_buff *skb,
struct flow_keys *hkeys)
{
#ifdef CONFIG_IP_ROUTE_MULTIPATH
- if (res->fi && res->fi->fib_nhs > 1) {
+ if (res->fi && fib_info_num_path(res->fi) > 1) {
int h = fib_multipath_hash(res->fi->fib_net, NULL, skb, hkeys);
fib_select_multipath(res, h);
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 2/7] ipv4: Prepare for fib6_nh from a nexthop object
2019-05-30 3:07 [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info David Ahern
2019-05-30 3:07 ` [PATCH net-next 1/7] ipv4: Use accessors for fib_info nexthop data David Ahern
@ 2019-05-30 3:07 ` David Ahern
2019-05-30 3:07 ` [PATCH net-next 3/7] ipv4: Plumb support for nexthop object in a fib_info David Ahern
` (5 subsequent siblings)
7 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-05-30 3:07 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, saeedm, kafai, weiwan, David Ahern
From: David Ahern <dsahern@gmail.com>
Convert more IPv4 code to use fib_nh_common over fib_nh to enable routes
to use a fib6_nh based nexthop. In the end, only code not using a
nexthop object in a fib_info should directly access fib_nh in a fib_info
without checking the famiy and going through fib_nh_common. Those
functions will be marked when it is not directly evident.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
include/net/ip_fib.h | 15 +++++++++----
net/ipv4/fib_frontend.c | 12 +++++------
net/ipv4/fib_rules.c | 4 ++--
net/ipv4/fib_semantics.c | 55 +++++++++++++++++++++++++++++++++---------------
net/ipv4/fib_trie.c | 15 +++++++------
net/ipv4/nexthop.c | 3 ++-
net/ipv4/route.c | 2 +-
7 files changed, 69 insertions(+), 37 deletions(-)
diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 42b1a806f6f5..7da8ea784029 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -195,8 +195,8 @@ struct fib_result_nl {
#define FIB_TABLE_HASHSZ 2
#endif
-__be32 fib_info_update_nh_saddr(struct net *net, struct fib_nh *nh,
- unsigned char scope);
+__be32 fib_info_update_nhc_saddr(struct net *net, struct fib_nh_common *nhc,
+ unsigned char scope);
__be32 fib_result_prefsrc(struct net *net, struct fib_result *res);
#define FIB_RES_NHC(res) ((res).nhc)
@@ -455,11 +455,18 @@ static inline void fib_combine_itag(u32 *itag, const struct fib_result *res)
{
#ifdef CONFIG_IP_ROUTE_CLASSID
struct fib_nh_common *nhc = res->nhc;
- struct fib_nh *nh = container_of(nhc, struct fib_nh, nh_common);
#ifdef CONFIG_IP_MULTIPLE_TABLES
u32 rtag;
#endif
- *itag = nh->nh_tclassid << 16;
+ if (nhc->nhc_family == AF_INET) {
+ struct fib_nh *nh;
+
+ nh = container_of(nhc, struct fib_nh, nh_common);
+ *itag = nh->nh_tclassid << 16;
+ } else {
+ *itag = 0;
+ }
+
#ifdef CONFIG_IP_MULTIPLE_TABLES
rtag = res->tclassid;
if (*itag == 0)
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index ab369959ce0b..8e49baa00d20 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -235,9 +235,9 @@ static inline unsigned int __inet_dev_addr_type(struct net *net,
if (table) {
ret = RTN_UNICAST;
if (!fib_table_lookup(table, &fl4, &res, FIB_LOOKUP_NOREF)) {
- struct fib_nh *nh = fib_info_nh(res.fi, 0);
+ struct fib_nh_common *nhc = fib_info_nhc(res.fi, 0);
- if (!dev || dev == nh->fib_nh_dev)
+ if (!dev || dev == nhc->nhc_dev)
ret = res.type;
}
}
@@ -325,18 +325,18 @@ bool fib_info_nh_uses_dev(struct fib_info *fi, const struct net_device *dev)
int ret;
for (ret = 0; ret < fib_info_num_path(fi); ret++) {
- const struct fib_nh *nh = fib_info_nh(fi, ret);
+ const struct fib_nh_common *nhc = fib_info_nhc(fi, ret);
- if (nh->fib_nh_dev == dev) {
+ if (nhc->nhc_dev == dev) {
dev_match = true;
break;
- } else if (l3mdev_master_ifindex_rcu(nh->fib_nh_dev) == dev->ifindex) {
+ } else if (l3mdev_master_ifindex_rcu(nhc->nhc_dev) == dev->ifindex) {
dev_match = true;
break;
}
}
#else
- if (fib_info_nh(fi, 0)->fib_nh_dev == dev)
+ if (fib_info_nhc(fi, 0)->nhc_dev == dev)
dev_match = true;
#endif
diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c
index ab06fd73b343..88807c138df4 100644
--- a/net/ipv4/fib_rules.c
+++ b/net/ipv4/fib_rules.c
@@ -147,9 +147,9 @@ static bool fib4_rule_suppress(struct fib_rule *rule, struct fib_lookup_arg *arg
struct net_device *dev = NULL;
if (result->fi) {
- struct fib_nh *nh = fib_info_nh(result->fi, 0);
+ struct fib_nh_common *nhc = fib_info_nhc(result->fi, 0);
- dev = nh->fib_nh_dev;
+ dev = nhc->nhc_dev;
}
/* do not accept result if the route does
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index a37ff07718a8..bc768113fb4f 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -61,6 +61,9 @@ static unsigned int fib_info_cnt;
#define DEVINDEX_HASHSIZE (1U << DEVINDEX_HASHBITS)
static struct hlist_head fib_info_devhash[DEVINDEX_HASHSIZE];
+/* for_nexthops and change_nexthops only used when nexthop object
+ * is not used with a fib_info
+ */
#ifdef CONFIG_IP_ROUTE_MULTIPATH
#define for_nexthops(fi) { \
@@ -402,20 +405,23 @@ static inline size_t fib_nlmsg_size(struct fib_info *fi)
/* each nexthop is packed in an attribute */
size_t nhsize = nla_total_size(sizeof(struct rtnexthop));
+ unsigned int i;
/* may contain flow and gateway attribute */
nhsize += 2 * nla_total_size(4);
/* grab encap info */
- for_nexthops(fi) {
- if (nh->fib_nh_lws) {
+ for (i = 0; i < fib_info_num_path(fi); i++) {
+ struct fib_nh_common *nhc = fib_info_nhc(fi, i);
+
+ if (nhc->nhc_lwtstate) {
/* RTA_ENCAP_TYPE */
nh_encapsize += lwtunnel_get_encap_size(
- nh->fib_nh_lws);
+ nhc->nhc_lwtstate);
/* RTA_ENCAP */
nh_encapsize += nla_total_size(2);
}
- } endfor_nexthops(fi);
+ }
/* all nexthops are packed in a nested attribute */
payload += nla_total_size((nhs * nhsize) + nh_encapsize);
@@ -1194,9 +1200,15 @@ static void fib_info_hash_move(struct hlist_head *new_info_hash,
fib_info_hash_free(old_laddrhash, bytes);
}
-__be32 fib_info_update_nh_saddr(struct net *net, struct fib_nh *nh,
- unsigned char scope)
+__be32 fib_info_update_nhc_saddr(struct net *net, struct fib_nh_common *nhc,
+ unsigned char scope)
{
+ struct fib_nh *nh;
+
+ if (nhc->nhc_family != AF_INET)
+ return inet_select_addr(nhc->nhc_dev, 0, scope);
+
+ nh = container_of(nhc, struct fib_nh, nh_common);
nh->nh_saddr = inet_select_addr(nh->fib_nh_dev, nh->fib_nh_gw4, scope);
nh->nh_saddr_genid = atomic_read(&net->ipv4.dev_addr_genid);
@@ -1206,16 +1218,19 @@ __be32 fib_info_update_nh_saddr(struct net *net, struct fib_nh *nh,
__be32 fib_result_prefsrc(struct net *net, struct fib_result *res)
{
struct fib_nh_common *nhc = res->nhc;
- struct fib_nh *nh;
if (res->fi->fib_prefsrc)
return res->fi->fib_prefsrc;
- nh = container_of(nhc, struct fib_nh, nh_common);
- if (nh->nh_saddr_genid == atomic_read(&net->ipv4.dev_addr_genid))
- return nh->nh_saddr;
+ if (nhc->nhc_family == AF_INET) {
+ struct fib_nh *nh;
+
+ nh = container_of(nhc, struct fib_nh, nh_common);
+ if (nh->nh_saddr_genid == atomic_read(&net->ipv4.dev_addr_genid))
+ return nh->nh_saddr;
+ }
- return fib_info_update_nh_saddr(net, nh, res->fi->fib_scope);
+ return fib_info_update_nhc_saddr(net, nhc, res->fi->fib_scope);
}
static bool fib_valid_prefsrc(struct fib_config *cfg, __be32 fib_prefsrc)
@@ -1397,7 +1412,8 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
}
change_nexthops(fi) {
- fib_info_update_nh_saddr(net, nexthop_nh, fi->fib_scope);
+ fib_info_update_nhc_saddr(net, &nexthop_nh->nh_common,
+ fi->fib_scope);
if (nexthop_nh->fib_nh_gw_family == AF_INET6)
fi->fib_nh_is_v6 = true;
} endfor_nexthops(fi)
@@ -1625,17 +1641,22 @@ 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 (nhs == 1) {
- const struct fib_nh *nh = fib_info_nh(fi, 0);
+ const struct fib_nh_common *nhc = fib_info_nhc(fi, 0);
unsigned char flags = 0;
- if (fib_nexthop_info(skb, &nh->nh_common, &flags, false) < 0)
+ if (fib_nexthop_info(skb, nhc, &flags, false) < 0)
goto nla_put_failure;
rtm->rtm_flags = flags;
#ifdef CONFIG_IP_ROUTE_CLASSID
- if (nh->nh_tclassid &&
- nla_put_u32(skb, RTA_FLOW, nh->nh_tclassid))
- goto nla_put_failure;
+ if (nhc->nhc_family == AF_INET) {
+ struct fib_nh *nh;
+
+ nh = container_of(nhc, struct fib_nh, nh_common);
+ if (nh->nh_tclassid &&
+ nla_put_u32(skb, RTA_FLOW, nh->nh_tclassid))
+ goto nla_put_failure;
+ }
#endif
} else {
if (fib_add_multipath(skb, fi) < 0)
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 5c8a4d21b8e0..d704d1606b8f 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -2724,9 +2724,9 @@ static unsigned int fib_flag_trans(int type, __be32 mask, struct fib_info *fi)
if (type == RTN_UNREACHABLE || type == RTN_PROHIBIT)
flags = RTF_REJECT;
if (fi) {
- const struct fib_nh *nh = fib_info_nh(fi, 0);
+ const struct fib_nh_common *nhc = fib_info_nhc(fi, 0);
- if (nh->fib_nh_gw4)
+ if (nhc->nhc_gw.ipv4)
flags |= RTF_GATEWAY;
}
if (mask == htonl(0xFFFFFFFF))
@@ -2773,14 +2773,17 @@ static int fib_route_seq_show(struct seq_file *seq, void *v)
seq_setwidth(seq, 127);
if (fi) {
- struct fib_nh *nh = fib_info_nh(fi, 0);
+ struct fib_nh_common *nhc = fib_info_nhc(fi, 0);
+ __be32 gw = 0;
+
+ if (nhc->nhc_gw_family == AF_INET)
+ gw = nhc->nhc_gw.ipv4;
seq_printf(seq,
"%s\t%08X\t%08X\t%04X\t%d\t%u\t"
"%d\t%08X\t%d\t%u\t%u",
- nh->fib_nh_dev ? nh->fib_nh_dev->name : "*",
- prefix,
- nh->fib_nh_gw4, flags, 0, 0,
+ nhc->nhc_dev ? nhc->nhc_dev->name : "*",
+ prefix, gw, flags, 0, 0,
fi->fib_priority,
mask,
(fi->fib_advmss ?
diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c
index 1af8a329dacb..d426829b7d32 100644
--- a/net/ipv4/nexthop.c
+++ b/net/ipv4/nexthop.c
@@ -815,7 +815,8 @@ static int nh_create_ipv4(struct net *net, struct nexthop *nh,
err = fib_check_nh(net, fib_nh, tb_id, 0, extack);
if (!err) {
nh->nh_flags = fib_nh->fib_nh_flags;
- fib_info_update_nh_saddr(net, fib_nh, fib_nh->fib_nh_scope);
+ fib_info_update_nhc_saddr(net, &fib_nh->nh_common,
+ fib_nh->fib_nh_scope);
} else {
fib_nh_release(net, fib_nh);
}
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 05a6a8ecb574..4a1168451f3a 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1585,7 +1585,7 @@ static void rt_set_nexthop(struct rtable *rt, __be32 daddr,
ip_dst_init_metrics(&rt->dst, fi->fib_metrics);
#ifdef CONFIG_IP_ROUTE_CLASSID
- {
+ if (nhc->nhc_family == AF_INET) {
struct fib_nh *nh;
nh = container_of(nhc, struct fib_nh, nh_common);
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 3/7] ipv4: Plumb support for nexthop object in a fib_info
2019-05-30 3:07 [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info David Ahern
2019-05-30 3:07 ` [PATCH net-next 1/7] ipv4: Use accessors for fib_info nexthop data David Ahern
2019-05-30 3:07 ` [PATCH net-next 2/7] ipv4: Prepare for fib6_nh from a nexthop object David Ahern
@ 2019-05-30 3:07 ` David Ahern
2019-05-30 3:07 ` [PATCH net-next 4/7] ipv6: Plumb support for nexthop object in a fib6_info David Ahern
` (4 subsequent siblings)
7 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-05-30 3:07 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, saeedm, kafai, weiwan, David Ahern
From: David Ahern <dsahern@gmail.com>
Add 'struct nexthop' and nh_list list_head to fib_info. nh_list is the
fib_info side of the nexthop <-> fib_info relationship.
Add fi_list list_head to 'struct nexthop' to track fib_info entries
using a nexthop instance. Add __remove_nexthop_fib and add it to
__remove_nexthop to walk the new list_head and mark those fib entries
as dead when the nexthop is deleted.
Add a few nexthop helpers for use when a nexthop is added to fib_info.
Update the fib_info functions to check for fi->nh and take a different
path as needed.
The bulk of the changes are in fib_semantics.c and most of that is
moving the existing change_nexthops into an else branch.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
include/net/ip_fib.h | 4 ++
include/net/nexthop.h | 48 ++++++++++++++++
net/ipv4/fib_semantics.c | 142 +++++++++++++++++++++++++++++++++++------------
net/ipv4/fib_trie.c | 7 +++
net/ipv4/nexthop.c | 64 +++++++++++++++++++++
5 files changed, 229 insertions(+), 36 deletions(-)
diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 7da8ea784029..071d280de389 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -129,9 +129,12 @@ struct fib_nh {
* This structure contains data shared by many of routes.
*/
+struct nexthop;
+
struct fib_info {
struct hlist_node fib_hash;
struct hlist_node fib_lhash;
+ struct list_head nh_list;
struct net *fib_net;
int fib_treeref;
refcount_t fib_clntref;
@@ -151,6 +154,7 @@ struct fib_info {
int fib_nhs;
bool fib_nh_is_v6;
bool nh_updated;
+ struct nexthop *nh;
struct rcu_head rcu;
struct fib_nh fib_nh[0];
};
diff --git a/include/net/nexthop.h b/include/net/nexthop.h
index e501d77b82c8..2912a2d7a515 100644
--- a/include/net/nexthop.h
+++ b/include/net/nexthop.h
@@ -77,6 +77,7 @@ struct nh_group {
struct nexthop {
struct rb_node rb_node; /* entry on netns rbtree */
+ struct list_head fi_list; /* v4 entries using nh */
struct list_head grp_list; /* nh group entries using this nh */
struct net *net;
@@ -110,6 +111,12 @@ static inline void nexthop_put(struct nexthop *nh)
call_rcu(&nh->rcu, nexthop_free_rcu);
}
+static inline bool nexthop_cmp(const struct nexthop *nh1,
+ const struct nexthop *nh2)
+{
+ return nh1 == nh2;
+}
+
static inline bool nexthop_is_multipath(const struct nexthop *nh)
{
if (nh->is_group) {
@@ -193,18 +200,59 @@ static inline bool nexthop_is_blackhole(const struct nexthop *nh)
return nhi->reject_nh;
}
+static inline void nexthop_path_fib_result(struct fib_result *res, int hash)
+{
+ struct nh_info *nhi;
+ struct nexthop *nh;
+
+ nh = nexthop_select_path(res->fi->nh, hash);
+ nhi = rcu_dereference(nh->nh_info);
+ res->nhc = &nhi->fib_nhc;
+}
+
+/* called with rcu read lock or rtnl held */
+static inline
+struct fib_nh_common *nexthop_fib_nhc(struct nexthop *nh, int nhsel)
+{
+ struct nh_info *nhi;
+
+ BUILD_BUG_ON(offsetof(struct fib_nh, nh_common) != 0);
+ BUILD_BUG_ON(offsetof(struct fib6_nh, nh_common) != 0);
+
+ if (nexthop_is_multipath(nh)) {
+ nh = nexthop_mpath_select(nh, nhsel);
+ if (!nh)
+ return NULL;
+ }
+
+ nhi = rcu_dereference_rtnl(nh->nh_info);
+ return &nhi->fib_nhc;
+}
+
static inline unsigned int fib_info_num_path(const struct fib_info *fi)
{
+ if (unlikely(fi->nh))
+ return nexthop_num_path(fi->nh);
+
return fi->fib_nhs;
}
+int fib_check_nexthop(struct nexthop *nh, u8 scope,
+ struct netlink_ext_ack *extack);
+
static inline struct fib_nh_common *fib_info_nhc(struct fib_info *fi, int nhsel)
{
+ if (unlikely(fi->nh))
+ return nexthop_fib_nhc(fi->nh, nhsel);
+
return &fi->fib_nh[nhsel].nh_common;
}
+/* only used when fib_nh is built into fib_info */
static inline struct fib_nh *fib_info_nh(struct fib_info *fi, int nhsel)
{
+ WARN_ON(fi->nh);
+
return &fi->fib_nh[nhsel];
}
#endif
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index bc768113fb4f..4a3564dc0ee3 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -236,9 +236,13 @@ static void free_fib_info_rcu(struct rcu_head *head)
{
struct fib_info *fi = container_of(head, struct fib_info, rcu);
- change_nexthops(fi) {
- fib_nh_release(fi->fib_net, nexthop_nh);
- } endfor_nexthops(fi);
+ if (fi->nh) {
+ nexthop_put(fi->nh);
+ } else {
+ change_nexthops(fi) {
+ fib_nh_release(fi->fib_net, nexthop_nh);
+ } endfor_nexthops(fi);
+ }
ip_fib_metrics_put(fi->fib_metrics);
@@ -264,11 +268,15 @@ void fib_release_info(struct fib_info *fi)
hlist_del(&fi->fib_hash);
if (fi->fib_prefsrc)
hlist_del(&fi->fib_lhash);
- change_nexthops(fi) {
- if (!nexthop_nh->fib_nh_dev)
- continue;
- hlist_del(&nexthop_nh->nh_hash);
- } endfor_nexthops(fi)
+ if (fi->nh) {
+ list_del(&fi->nh_list);
+ } else {
+ change_nexthops(fi) {
+ if (!nexthop_nh->fib_nh_dev)
+ continue;
+ hlist_del(&nexthop_nh->nh_hash);
+ } endfor_nexthops(fi)
+ }
fi->fib_dead = 1;
fib_info_put(fi);
}
@@ -279,6 +287,12 @@ static inline int nh_comp(struct fib_info *fi, struct fib_info *ofi)
{
const struct fib_nh *onh;
+ if (fi->nh || ofi->nh)
+ return nexthop_cmp(fi->nh, ofi->nh) ? 0 : -1;
+
+ if (ofi->fib_nhs == 0)
+ return 0;
+
for_nexthops(fi) {
onh = fib_info_nh(ofi, nhsel);
@@ -323,9 +337,14 @@ static inline unsigned int fib_info_hashfn(const struct fib_info *fi)
val ^= (fi->fib_protocol << 8) | fi->fib_scope;
val ^= (__force u32)fi->fib_prefsrc;
val ^= fi->fib_priority;
- for_nexthops(fi) {
- val ^= fib_devindex_hashfn(nh->fib_nh_oif);
- } endfor_nexthops(fi)
+
+ if (fi->nh) {
+ val ^= fib_devindex_hashfn(fi->nh->id);
+ } else {
+ for_nexthops(fi) {
+ val ^= fib_devindex_hashfn(nh->fib_nh_oif);
+ } endfor_nexthops(fi)
+ }
return (val ^ (val >> 7) ^ (val >> 12)) & mask;
}
@@ -352,7 +371,7 @@ static struct fib_info *fib_find_info(struct fib_info *nfi)
memcmp(nfi->fib_metrics, fi->fib_metrics,
sizeof(u32) * RTAX_MAX) == 0 &&
!((nfi->fib_flags ^ fi->fib_flags) & ~RTNH_COMPARE_MASK) &&
- (nfi->fib_nhs == 0 || nh_comp(fi, nfi) == 0))
+ nh_comp(fi, nfi) == 0)
return fi;
}
@@ -399,6 +418,9 @@ static inline size_t fib_nlmsg_size(struct fib_info *fi)
/* space for nested metrics */
payload += nla_total_size((RTAX_MAX * nla_total_size(4)));
+ if (fi->nh)
+ payload += nla_total_size(4); /* RTA_NH_ID */
+
if (nhs) {
size_t nh_encapsize = 0;
/* Also handles the special case nhs == 1 */
@@ -585,6 +607,7 @@ static int fib_count_nexthops(struct rtnexthop *rtnh, int remaining,
return nhs;
}
+/* only called when fib_nh is integrated into fib_info */
static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh,
int remaining, struct fib_config *cfg,
struct netlink_ext_ack *extack)
@@ -683,6 +706,7 @@ static int fib_get_nhs(struct fib_info *fi, struct rtnexthop *rtnh,
return ret;
}
+/* only called when fib_nh is integrated into fib_info */
static void fib_rebalance(struct fib_info *fi)
{
int total;
@@ -1262,6 +1286,7 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
{
int err;
struct fib_info *fi = NULL;
+ struct nexthop *nh = NULL;
struct fib_info *ofi;
int nhs = 1;
struct net *net = cfg->fc_nlinfo.nl_net;
@@ -1333,14 +1358,25 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
fi->fib_tb_id = cfg->fc_table;
fi->fib_nhs = nhs;
- change_nexthops(fi) {
- nexthop_nh->nh_parent = fi;
- } endfor_nexthops(fi)
+ if (nh) {
+ if (!nexthop_get(nh)) {
+ NL_SET_ERR_MSG(extack, "Nexthop has been deleted");
+ err = -EINVAL;
+ } else {
+ err = 0;
+ fi->nh = nh;
+ }
+ } else {
+ change_nexthops(fi) {
+ nexthop_nh->nh_parent = fi;
+ } endfor_nexthops(fi)
- if (cfg->fc_mp)
- err = fib_get_nhs(fi, cfg->fc_mp, cfg->fc_mp_len, cfg, extack);
- else
- err = fib_nh_init(net, fi->fib_nh, cfg, 1, extack);
+ if (cfg->fc_mp)
+ err = fib_get_nhs(fi, cfg->fc_mp, cfg->fc_mp_len, cfg,
+ extack);
+ else
+ err = fib_nh_init(net, fi->fib_nh, cfg, 1, extack);
+ }
if (err != 0)
goto failure;
@@ -1371,7 +1407,11 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
goto err_inval;
}
- if (cfg->fc_scope == RT_SCOPE_HOST) {
+ if (fi->nh) {
+ err = fib_check_nexthop(fi->nh, cfg->fc_scope, extack);
+ if (err)
+ goto failure;
+ } else if (cfg->fc_scope == RT_SCOPE_HOST) {
struct fib_nh *nh = fi->fib_nh;
/* Local address is added. */
@@ -1411,14 +1451,16 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
goto err_inval;
}
- change_nexthops(fi) {
- fib_info_update_nhc_saddr(net, &nexthop_nh->nh_common,
- fi->fib_scope);
- if (nexthop_nh->fib_nh_gw_family == AF_INET6)
- fi->fib_nh_is_v6 = true;
- } endfor_nexthops(fi)
+ if (!fi->nh) {
+ change_nexthops(fi) {
+ fib_info_update_nhc_saddr(net, &nexthop_nh->nh_common,
+ fi->fib_scope);
+ if (nexthop_nh->fib_nh_gw_family == AF_INET6)
+ fi->fib_nh_is_v6 = true;
+ } endfor_nexthops(fi)
- fib_rebalance(fi);
+ fib_rebalance(fi);
+ }
link_it:
ofi = fib_find_info(fi);
@@ -1440,16 +1482,20 @@ struct fib_info *fib_create_info(struct fib_config *cfg,
head = &fib_info_laddrhash[fib_laddr_hashfn(fi->fib_prefsrc)];
hlist_add_head(&fi->fib_lhash, head);
}
- change_nexthops(fi) {
- struct hlist_head *head;
- unsigned int hash;
+ if (fi->nh) {
+ list_add(&fi->nh_list, &nh->fi_list);
+ } else {
+ change_nexthops(fi) {
+ struct hlist_head *head;
+ unsigned int hash;
- if (!nexthop_nh->fib_nh_dev)
- continue;
- 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)
+ if (!nexthop_nh->fib_nh_dev)
+ continue;
+ 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)
+ }
spin_unlock_bh(&fib_info_lock);
return fi;
@@ -1576,6 +1622,12 @@ static int fib_add_multipath(struct sk_buff *skb, struct fib_info *fi)
if (!mp)
goto nla_put_failure;
+ if (unlikely(fi->nh)) {
+ if (nexthop_mpath_fill_node(skb, fi->nh) < 0)
+ goto nla_put_failure;
+ goto mp_end;
+ }
+
for_nexthops(fi) {
if (fib_add_nexthop(skb, &nh->nh_common, nh->fib_nh_weight) < 0)
goto nla_put_failure;
@@ -1586,6 +1638,7 @@ static int fib_add_multipath(struct sk_buff *skb, struct fib_info *fi)
#endif
} endfor_nexthops(fi);
+mp_end:
nla_nest_end(skb, mp);
return 0;
@@ -1640,6 +1693,14 @@ int fib_dump_info(struct sk_buff *skb, u32 portid, u32 seq, int event,
if (fi->fib_prefsrc &&
nla_put_in_addr(skb, RTA_PREFSRC, fi->fib_prefsrc))
goto nla_put_failure;
+
+ if (fi->nh) {
+ if (nla_put_u32(skb, RTA_NH_ID, fi->nh->id))
+ goto nla_put_failure;
+ if (nexthop_is_blackhole(fi->nh))
+ rtm->rtm_type = RTN_BLACKHOLE;
+ }
+
if (nhs == 1) {
const struct fib_nh_common *nhc = fib_info_nhc(fi, 0);
unsigned char flags = 0;
@@ -1784,6 +1845,8 @@ void fib_sync_mtu(struct net_device *dev, u32 orig_mtu)
* NETDEV_DOWN 0 LINKDOWN|DEAD Link down, not for scope host
* NETDEV_DOWN 1 LINKDOWN|DEAD Last address removed
* NETDEV_UNREGISTER 1 LINKDOWN|DEAD Device removed
+ *
+ * only used when fib_nh is built into fib_info
*/
int fib_sync_down_dev(struct net_device *dev, unsigned long event, bool force)
{
@@ -1931,6 +1994,8 @@ static void fib_select_default(const struct flowi4 *flp, struct fib_result *res)
/*
* Dead device goes up. We wake up dead nexthops.
* It takes sense only on multipath routes.
+ *
+ * only used when fib_nh is built into fib_info
*/
int fib_sync_up(struct net_device *dev, unsigned char nh_flags)
{
@@ -2025,6 +2090,11 @@ void fib_select_multipath(struct fib_result *res, int hash)
struct net *net = fi->fib_net;
bool first = false;
+ if (unlikely(res->fi->nh)) {
+ nexthop_path_fib_result(res, hash);
+ return;
+ }
+
change_nexthops(fi) {
if (net->ipv4.sysctl_fib_multipath_use_neigh) {
if (!fib_good_nh(nexthop_nh))
diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index d704d1606b8f..716f2d66cb3f 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -1461,6 +1461,7 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi4 *flp,
fib_alias_accessed(fa);
err = fib_props[fa->fa_type].error;
if (unlikely(err < 0)) {
+out_reject:
#ifdef CONFIG_IP_FIB_TRIE_STATS
this_cpu_inc(stats->semantic_match_passed);
#endif
@@ -1469,6 +1470,12 @@ int fib_table_lookup(struct fib_table *tb, const struct flowi4 *flp,
}
if (fi->fib_flags & RTNH_F_DEAD)
continue;
+
+ if (unlikely(fi->nh && nexthop_is_blackhole(fi->nh))) {
+ err = fib_props[RTN_BLACKHOLE].error;
+ goto out_reject;
+ }
+
for (nhsel = 0; nhsel < fib_info_num_path(fi); nhsel++) {
struct fib_nh_common *nhc = fib_info_nhc(fi, nhsel);
diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c
index d426829b7d32..504dd147f296 100644
--- a/net/ipv4/nexthop.c
+++ b/net/ipv4/nexthop.c
@@ -105,6 +105,7 @@ static struct nexthop *nexthop_alloc(void)
nh = kzalloc(sizeof(struct nexthop), GFP_KERNEL);
if (nh) {
+ INIT_LIST_HEAD(&nh->fi_list);
INIT_LIST_HEAD(&nh->grp_list);
}
return nh;
@@ -515,6 +516,54 @@ struct nexthop *nexthop_select_path(struct nexthop *nh, int hash)
}
EXPORT_SYMBOL_GPL(nexthop_select_path);
+static int nexthop_check_scope(struct nexthop *nh, u8 scope,
+ struct netlink_ext_ack *extack)
+{
+ struct nh_info *nhi;
+
+ nhi = rtnl_dereference(nh->nh_info);
+ if (scope == RT_SCOPE_HOST && nhi->fib_nhc.nhc_gw_family) {
+ NL_SET_ERR_MSG(extack,
+ "Route with host scope can not have a gateway");
+ return -EINVAL;
+ }
+
+ if (nhi->fib_nhc.nhc_flags & RTNH_F_ONLINK && scope >= RT_SCOPE_LINK) {
+ NL_SET_ERR_MSG(extack, "Scope mismatch with nexthop");
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+/* Invoked by fib add code to verify nexthop by id is ok with
+ * config for prefix; parts of fib_check_nh not done when nexthop
+ * object is used.
+ */
+int fib_check_nexthop(struct nexthop *nh, u8 scope,
+ struct netlink_ext_ack *extack)
+{
+ int err = 0;
+
+ if (nh->is_group) {
+ struct nh_group *nhg;
+
+ if (scope == RT_SCOPE_HOST) {
+ NL_SET_ERR_MSG(extack, "Route with host scope can not have multiple nexthops");
+ err = -EINVAL;
+ goto out;
+ }
+
+ nhg = rtnl_dereference(nh->nh_grp);
+ /* all nexthops in a group have the same scope */
+ err = nexthop_check_scope(nhg->nh_entries[0].nh, scope, extack);
+ } else {
+ err = nexthop_check_scope(nh, scope, extack);
+ }
+out:
+ return err;
+}
+
static void nh_group_rebalance(struct nh_group *nhg)
{
int total = 0;
@@ -607,9 +656,24 @@ static void remove_nexthop_group(struct nexthop *nh, struct nl_info *nlinfo)
}
}
+static void __remove_nexthop_fib(struct net *net, struct nexthop *nh)
+{
+ bool do_flush = false;
+ struct fib_info *fi;
+
+ list_for_each_entry(fi, &nh->fi_list, nh_list) {
+ fi->fib_flags |= RTNH_F_DEAD;
+ do_flush = true;
+ }
+ if (do_flush)
+ fib_flush(net);
+}
+
static void __remove_nexthop(struct net *net, struct nexthop *nh,
struct nl_info *nlinfo)
{
+ __remove_nexthop_fib(net, nh);
+
if (nh->is_group) {
remove_nexthop_group(nh, nlinfo);
} else {
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 4/7] ipv6: Plumb support for nexthop object in a fib6_info
2019-05-30 3:07 [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info David Ahern
` (2 preceding siblings ...)
2019-05-30 3:07 ` [PATCH net-next 3/7] ipv4: Plumb support for nexthop object in a fib_info David Ahern
@ 2019-05-30 3:07 ` David Ahern
2019-05-30 3:07 ` [PATCH net-next 5/7] mlxsw: Fail attempts to use routes with nexthop objects David Ahern
` (3 subsequent siblings)
7 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-05-30 3:07 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, saeedm, kafai, weiwan, David Ahern
From: David Ahern <dsahern@gmail.com>
Add struct nexthop and nh_list list_head to fib6_info. nh_list is the
fib6_info side of the nexthop <-> fib_info relationship. Since a fib6_info
referencing a nexthop object can not have 'sibling' entries (the old way
of doing multipath routes), the nh_list is a union with fib6_siblings.
Add f6i_list list_head to 'struct nexthop' to track fib6_info entries
using a nexthop instance. Update __remove_nexthop_fib to walk f6_list
and delete fib entries using the nexthop.
Add a few nexthop helpers for use when a nexthop is added to fib6_info.
Update the fib6_info references to check for nh and take a different path
as needed.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
include/net/ip6_fib.h | 11 ++--
include/net/ip6_route.h | 13 +++-
include/net/nexthop.h | 50 ++++++++++++++++
net/ipv4/nexthop.c | 44 ++++++++++++++
net/ipv6/addrconf.c | 5 ++
net/ipv6/ip6_fib.c | 22 +++++--
net/ipv6/ndisc.c | 3 +-
net/ipv6/route.c | 156 +++++++++++++++++++++++++++++++++++++++++-------
8 files changed, 268 insertions(+), 36 deletions(-)
diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
index ebe5d65f97e0..1a8acd51b277 100644
--- a/include/net/ip6_fib.h
+++ b/include/net/ip6_fib.h
@@ -146,7 +146,10 @@ struct fib6_info {
* destination, but not the same gateway. nsiblings is just a cache
* to speed up lookup.
*/
- struct list_head fib6_siblings;
+ union {
+ struct list_head fib6_siblings;
+ struct list_head nh_list;
+ };
unsigned int fib6_nsiblings;
refcount_t fib6_ref;
@@ -170,6 +173,7 @@ struct fib6_info {
unused:3;
struct rcu_head rcu;
+ struct nexthop *nh;
struct fib6_nh fib6_nh[0];
};
@@ -441,11 +445,6 @@ void rt6_get_prefsrc(const struct rt6_info *rt, struct in6_addr *addr)
rcu_read_unlock();
}
-static inline struct net_device *fib6_info_nh_dev(const struct fib6_info *f6i)
-{
- return f6i->fib6_nh->fib_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);
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index a6ce6ea856b9..7375a165fd98 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -27,6 +27,7 @@ struct route_info {
#include <linux/ip.h>
#include <linux/ipv6.h>
#include <linux/route.h>
+#include <net/nexthop.h>
#define RT6_LOOKUP_F_IFACE 0x00000001
#define RT6_LOOKUP_F_REACHABLE 0x00000002
@@ -66,10 +67,13 @@ static inline bool rt6_need_strict(const struct in6_addr *daddr)
(IPV6_ADDR_MULTICAST | IPV6_ADDR_LINKLOCAL | IPV6_ADDR_LOOPBACK);
}
+/* fib entries using a nexthop object can not be coalesced into
+ * a multipath route
+ */
static inline bool rt6_qualify_for_ecmp(const struct fib6_info *f6i)
{
/* the RTF_ADDRCONF flag filters out RA's */
- return !(f6i->fib6_flags & RTF_ADDRCONF) &&
+ return !(f6i->fib6_flags & RTF_ADDRCONF) && !f6i->nh &&
f6i->fib6_nh->fib_nh_gw_family;
}
@@ -275,8 +279,13 @@ 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)
{
- struct fib6_nh *nha = a->fib6_nh, *nhb = b->fib6_nh;
+ struct fib6_nh *nha, *nhb;
+
+ if (a->nh || b->nh)
+ return nexthop_cmp(a->nh, b->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);
diff --git a/include/net/nexthop.h b/include/net/nexthop.h
index 2912a2d7a515..aff7b2410057 100644
--- a/include/net/nexthop.h
+++ b/include/net/nexthop.h
@@ -10,6 +10,7 @@
#define __LINUX_NEXTHOP_H
#include <linux/netdevice.h>
+#include <linux/route.h>
#include <linux/types.h>
#include <net/ip_fib.h>
#include <net/ip6_fib.h>
@@ -78,6 +79,7 @@ struct nh_group {
struct nexthop {
struct rb_node rb_node; /* entry on netns rbtree */
struct list_head fi_list; /* v4 entries using nh */
+ struct list_head f6i_list; /* v6 entries using nh */
struct list_head grp_list; /* nh group entries using this nh */
struct net *net;
@@ -255,4 +257,52 @@ static inline struct fib_nh *fib_info_nh(struct fib_info *fi, int nhsel)
return &fi->fib_nh[nhsel];
}
+
+/*
+ * IPv6 variants
+ */
+int fib6_check_nexthop(struct nexthop *nh, struct fib6_config *cfg,
+ struct netlink_ext_ack *extack);
+
+static inline struct fib6_nh *nexthop_fib6_nh(struct nexthop *nh)
+{
+ struct nh_info *nhi;
+
+ if (nexthop_is_multipath(nh)) {
+ nh = nexthop_mpath_select(nh, 0);
+ if (!nh)
+ return NULL;
+ }
+
+ nhi = rcu_dereference_rtnl(nh->nh_info);
+ if (nhi->family == AF_INET6)
+ return &nhi->fib6_nh;
+
+ return NULL;
+}
+
+static inline struct net_device *fib6_info_nh_dev(struct fib6_info *f6i)
+{
+ struct fib6_nh *fib6_nh;
+
+ fib6_nh = f6i->nh ? nexthop_fib6_nh(f6i->nh) : f6i->fib6_nh;
+ return fib6_nh->fib_nh_dev;
+}
+
+static inline void nexthop_path_fib6_result(struct fib6_result *res, int hash)
+{
+ struct nexthop *nh = res->f6i->nh;
+ struct nh_info *nhi;
+
+ nh = nexthop_select_path(nh, hash);
+
+ nhi = rcu_dereference_rtnl(nh->nh_info);
+ if (nhi->reject_nh) {
+ res->fib6_type = RTN_BLACKHOLE;
+ res->fib6_flags |= RTF_REJECT;
+ res->nh = nexthop_fib6_nh(nh);
+ } else {
+ res->nh = &nhi->fib6_nh;
+ }
+}
#endif
diff --git a/net/ipv4/nexthop.c b/net/ipv4/nexthop.c
index 504dd147f296..b1a7749099e7 100644
--- a/net/ipv4/nexthop.c
+++ b/net/ipv4/nexthop.c
@@ -106,6 +106,7 @@ static struct nexthop *nexthop_alloc(void)
nh = kzalloc(sizeof(struct nexthop), GFP_KERNEL);
if (nh) {
INIT_LIST_HEAD(&nh->fi_list);
+ INIT_LIST_HEAD(&nh->f6i_list);
INIT_LIST_HEAD(&nh->grp_list);
}
return nh;
@@ -516,6 +517,41 @@ struct nexthop *nexthop_select_path(struct nexthop *nh, int hash)
}
EXPORT_SYMBOL_GPL(nexthop_select_path);
+int fib6_check_nexthop(struct nexthop *nh, struct fib6_config *cfg,
+ struct netlink_ext_ack *extack)
+{
+ struct nh_info *nhi;
+
+ /* fib6_src is unique to a fib6_info and limits the ability to cache
+ * routes in fib6_nh within a nexthop that is potentially shared
+ * across multiple fib entries. If the config wants to use source
+ * routing it can not use nexthop objects. mlxsw also does not allow
+ * fib6_src on routes.
+ */
+ if (!ipv6_addr_any(&cfg->fc_src)) {
+ NL_SET_ERR_MSG(extack, "IPv6 routes using source address can not use nexthop objects");
+ return -EINVAL;
+ }
+
+ if (nh->is_group) {
+ struct nh_group *nhg;
+
+ nhg = rtnl_dereference(nh->nh_grp);
+ if (nhg->has_v4)
+ goto no_v4_nh;
+ } else {
+ nhi = rtnl_dereference(nh->nh_info);
+ if (nhi->family == AF_INET)
+ goto no_v4_nh;
+ }
+
+ return 0;
+no_v4_nh:
+ NL_SET_ERR_MSG(extack, "IPv6 routes can not use an IPv4 nexthop");
+ return -EINVAL;
+}
+EXPORT_SYMBOL_GPL(fib6_check_nexthop);
+
static int nexthop_check_scope(struct nexthop *nh, u8 scope,
struct netlink_ext_ack *extack)
{
@@ -658,6 +694,7 @@ static void remove_nexthop_group(struct nexthop *nh, struct nl_info *nlinfo)
static void __remove_nexthop_fib(struct net *net, struct nexthop *nh)
{
+ struct fib6_info *f6i, *tmp;
bool do_flush = false;
struct fib_info *fi;
@@ -667,6 +704,13 @@ static void __remove_nexthop_fib(struct net *net, struct nexthop *nh)
}
if (do_flush)
fib_flush(net);
+
+ /* ip6_del_rt removes the entry from this list hence the _safe */
+ list_for_each_entry_safe(f6i, tmp, &nh->f6i_list, nh_list) {
+ /* __ip6_del_rt does a release, so do a hold here */
+ fib6_info_hold(f6i);
+ ipv6_stub->ip6_del_rt(net, f6i);
+ }
}
static void __remove_nexthop(struct net *net, struct nexthop *nh,
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 683613e7355b..bc2a33f1a6ce 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -2421,6 +2421,10 @@ static struct fib6_info *addrconf_get_prefix_route(const struct in6_addr *pfx,
goto out;
for_each_fib6_node_rt_rcu(fn) {
+ /* prefix routes only use builtin fib6_nh */
+ if (rt->nh)
+ continue;
+
if (rt->fib6_nh->fib_nh_dev->ifindex != dev->ifindex)
continue;
if (no_gw && rt->fib6_nh->fib_nh_gw_family)
@@ -6341,6 +6345,7 @@ void addrconf_disable_policy_idev(struct inet6_dev *idev, int val)
list_for_each_entry(ifa, &idev->addr_list, if_list) {
spin_lock(&ifa->lock);
if (ifa->rt) {
+ /* host routes only use builtin fib6_nh */
struct fib6_nh *nh = ifa->rt->fib6_nh;
int cpu;
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index cdfb8500ccae..02feda73a98e 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -159,6 +159,7 @@ struct fib6_info *fib6_info_alloc(gfp_t gfp_flags, bool with_fib6_nh)
if (!f6i)
return NULL;
+ /* fib6_siblings is a union with nh_list, so this initializes both */
INIT_LIST_HEAD(&f6i->fib6_siblings);
refcount_set(&f6i->fib6_ref, 1);
@@ -171,7 +172,11 @@ void fib6_info_destroy_rcu(struct rcu_head *head)
WARN_ON(f6i->fib6_node);
- fib6_nh_release(f6i->fib6_nh);
+ if (f6i->nh)
+ nexthop_put(f6i->nh);
+ else
+ fib6_nh_release(f6i->fib6_nh);
+
ip_fib_metrics_put(f6i->fib6_metrics);
kfree(f6i);
}
@@ -927,6 +932,9 @@ static void fib6_purge_rt(struct fib6_info *rt, struct fib6_node *fn,
fib6_drop_pcpu_from(rt, table);
+ if (rt->nh && !list_empty(&rt->nh_list))
+ list_del_init(&rt->nh_list);
+
if (refcount_read(&rt->fib6_ref) != 1) {
/* This route is used as dummy address holder in some split
* nodes. It is not leaked, but it still holds other resources,
@@ -1334,6 +1342,8 @@ int fib6_add(struct fib6_node *root, struct fib6_info *rt,
err = fib6_add_rt2node(fn, rt, info, extack);
if (!err) {
+ if (rt->nh)
+ list_add(&rt->nh_list, &rt->nh->f6i_list);
__fib6_update_sernum_upto_root(rt, sernum);
fib6_start_gc(info->nl_net, rt);
}
@@ -2295,9 +2305,13 @@ static int ipv6_route_seq_show(struct seq_file *seq, void *v)
{
struct fib6_info *rt = v;
struct ipv6_route_iter *iter = seq->private;
+ struct fib6_nh *fib6_nh = rt->fib6_nh;
unsigned int flags = rt->fib6_flags;
const struct net_device *dev;
+ if (rt->nh)
+ fib6_nh = nexthop_fib6_nh(rt->nh);
+
seq_printf(seq, "%pi6 %02x ", &rt->fib6_dst.addr, rt->fib6_dst.plen);
#ifdef CONFIG_IPV6_SUBTREES
@@ -2305,14 +2319,14 @@ static int ipv6_route_seq_show(struct seq_file *seq, void *v)
#else
seq_puts(seq, "00000000000000000000000000000000 00 ");
#endif
- if (rt->fib6_nh->fib_nh_gw_family) {
+ if (fib6_nh->fib_nh_gw_family) {
flags |= RTF_GATEWAY;
- seq_printf(seq, "%pi6", &rt->fib6_nh->fib_nh_gw6);
+ seq_printf(seq, "%pi6", &fib6_nh->fib_nh_gw6);
} else {
seq_puts(seq, "00000000000000000000000000000000");
}
- dev = rt->fib6_nh->fib_nh_dev;
+ dev = fib6_nh->fib_nh_dev;
seq_printf(seq, " %08x %08x %08x %08x %8s\n",
rt->fib6_metric, refcount_read(&rt->fib6_ref), 0,
flags, dev ? dev->name : "");
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index f874dde1ee85..6e3c51109c83 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1289,9 +1289,8 @@ static void ndisc_router_discovery(struct sk_buff *skb)
!in6_dev->cnf.accept_ra_rtr_pref)
pref = ICMPV6_ROUTER_PREF_MEDIUM;
#endif
-
+ /* routes added from RAs do not use nexthop objects */
rt = rt6_get_dflt_router(net, &ipv6_hdr(skb)->saddr, skb->dev);
-
if (rt) {
neigh = ip6_neigh_lookup(&rt->fib6_nh->fib_nh_gw6,
rt->fib6_nh->fib_nh_dev, NULL,
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index fdf598926883..d1fe4d91f622 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -432,15 +432,21 @@ void fib6_select_path(const struct net *net, struct fib6_result *res,
struct fib6_info *sibling, *next_sibling;
struct fib6_info *match = res->f6i;
- if (!match->fib6_nsiblings || have_oif_match)
+ if ((!match->fib6_nsiblings && !match->nh) || have_oif_match)
goto out;
/* We might have already computed the hash for ICMPv6 errors. In such
* case it will always be non-zero. Otherwise now is the time to do it.
*/
- if (!fl6->mp_hash)
+ if (!fl6->mp_hash &&
+ (!match->nh || nexthop_is_multipath(match->nh)))
fl6->mp_hash = rt6_multipath_hash(net, fl6, skb, NULL);
+ if (unlikely(match->nh)) {
+ nexthop_path_fib6_result(res, fl6->mp_hash);
+ return;
+ }
+
if (fl6->mp_hash <= atomic_read(&match->fib6_nh->fib_nh_upper_bound))
goto out;
@@ -496,7 +502,13 @@ static void rt6_device_match(struct net *net, struct fib6_result *res,
struct fib6_nh *nh;
if (!oif && ipv6_addr_any(saddr)) {
- nh = f6i->fib6_nh;
+ if (unlikely(f6i->nh)) {
+ nh = nexthop_fib6_nh(f6i->nh);
+ if (nexthop_is_blackhole(f6i->nh))
+ goto out_blackhole;
+ } else {
+ nh = f6i->fib6_nh;
+ }
if (!(nh->fib_nh_flags & RTNH_F_DEAD))
goto out;
}
@@ -515,7 +527,14 @@ static void rt6_device_match(struct net *net, struct fib6_result *res,
goto out;
}
- nh = f6i->fib6_nh;
+ if (unlikely(f6i->nh)) {
+ nh = nexthop_fib6_nh(f6i->nh);
+ if (nexthop_is_blackhole(f6i->nh))
+ goto out_blackhole;
+ } else {
+ nh = f6i->fib6_nh;
+ }
+
if (nh->fib_nh_flags & RTNH_F_DEAD) {
res->f6i = net->ipv6.fib6_null_entry;
nh = res->f6i->fib6_nh;
@@ -524,6 +543,12 @@ static void rt6_device_match(struct net *net, struct fib6_result *res,
res->nh = nh;
res->fib6_type = res->f6i->fib6_type;
res->fib6_flags = res->f6i->fib6_flags;
+ return;
+
+out_blackhole:
+ res->fib6_flags |= RTF_REJECT;
+ res->fib6_type = RTN_BLACKHOLE;
+ res->nh = nh;
}
#ifdef CONFIG_IPV6_ROUTER_PREF
@@ -1117,6 +1142,8 @@ static struct rt6_info *ip6_pol_route_lookup(struct net *net,
rt = net->ipv6.ip6_null_entry;
dst_hold(&rt->dst);
goto out;
+ } else if (res.fib6_flags & RTF_REJECT) {
+ goto do_create;
}
fib6_select_path(net, &res, fl6, fl6->flowi6_oif,
@@ -1128,6 +1155,7 @@ static struct rt6_info *ip6_pol_route_lookup(struct net *net,
if (ip6_hold_safe(net, &rt))
dst_use_noref(&rt->dst, jiffies);
} else {
+do_create:
rt = ip6_create_rt_rcu(&res);
}
@@ -1982,6 +2010,14 @@ struct rt6_info *ip6_pol_route(struct net *net, struct fib6_table *table,
rcu_read_unlock();
dst_hold(&rt->dst);
return rt;
+ } else if (res.fib6_flags & RTF_REJECT) {
+ rt = ip6_create_rt_rcu(&res);
+ rcu_read_unlock();
+ if (!rt) {
+ rt = net->ipv6.ip6_null_entry;
+ dst_hold(&rt->dst);
+ }
+ return rt;
}
fib6_select_path(net, &res, fl6, oif, false, skb, strict);
@@ -3211,7 +3247,9 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
{
struct net *net = cfg->fc_nlinfo.nl_net;
struct fib6_info *rt = NULL;
+ struct nexthop *nh = NULL;
struct fib6_table *table;
+ struct fib6_nh *fib6_nh;
int err = -EINVAL;
int addr_type;
@@ -3264,7 +3302,7 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,
goto out;
err = -ENOMEM;
- rt = fib6_info_alloc(gfp_flags, true);
+ rt = fib6_info_alloc(gfp_flags, !nh);
if (!rt)
goto out;
@@ -3304,19 +3342,35 @@ 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
- err = fib6_nh_init(net, rt->fib6_nh, cfg, gfp_flags, extack);
- if (err)
- goto out;
+ if (nh) {
+ if (!nexthop_get(nh)) {
+ NL_SET_ERR_MSG(extack, "Nexthop has been deleted");
+ goto out;
+ }
+ if (rt->fib6_src.plen) {
+ NL_SET_ERR_MSG(extack, "Nexthops can not be used wtih source routing");
+ goto out;
+ }
+ rt->nh = nh;
+ fib6_nh = nexthop_fib6_nh(rt->nh);
+ } else {
+ err = fib6_nh_init(net, rt->fib6_nh, cfg, gfp_flags, extack);
+ if (err)
+ goto out;
- /* 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 (fib6_is_reject(cfg->fc_flags, rt->fib6_nh->fib_nh_dev, addr_type))
- rt->fib6_flags = RTF_REJECT | RTF_NONEXTHOP;
+ fib6_nh = rt->fib6_nh;
+
+ /* 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 (fib6_is_reject(cfg->fc_flags, rt->fib6_nh->fib_nh_dev,
+ addr_type))
+ rt->fib6_flags = RTF_REJECT | RTF_NONEXTHOP;
+ }
if (!ipv6_addr_any(&cfg->fc_prefsrc)) {
- struct net_device *dev = fib6_info_nh_dev(rt);
+ struct net_device *dev = fib6_nh->fib_nh_dev;
if (!ipv6_chk_addr(net, &cfg->fc_prefsrc, dev, 0)) {
NL_SET_ERR_MSG(extack, "Invalid source address");
@@ -3672,6 +3726,9 @@ static struct fib6_info *rt6_get_route_info(struct net *net,
goto out;
for_each_fib6_node_rt_rcu(fn) {
+ /* these routes do not use nexthops */
+ if (rt->nh)
+ continue;
if (rt->fib6_nh->fib_nh_dev->ifindex != ifindex)
continue;
if (!(rt->fib6_flags & RTF_ROUTEINFO) ||
@@ -3735,8 +3792,13 @@ struct fib6_info *rt6_get_dflt_router(struct net *net,
rcu_read_lock();
for_each_fib6_node_rt_rcu(&table->tb6_root) {
- struct fib6_nh *nh = rt->fib6_nh;
+ struct fib6_nh *nh;
+
+ /* RA routes do not use nexthops */
+ if (rt->nh)
+ continue;
+ nh = rt->fib6_nh;
if (dev == nh->fib_nh_dev &&
((rt->fib6_flags & (RTF_ADDRCONF | RTF_DEFAULT)) == (RTF_ADDRCONF | RTF_DEFAULT)) &&
ipv6_addr_equal(&nh->fib_nh_gw6, addr))
@@ -3987,7 +4049,8 @@ 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->fib_nh_dev == dev || !dev) &&
+ if (!rt->nh &&
+ ((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);
@@ -4015,8 +4078,13 @@ void rt6_remove_prefsrc(struct inet6_ifaddr *ifp)
static int fib6_clean_tohost(struct fib6_info *rt, void *arg)
{
struct in6_addr *gateway = (struct in6_addr *)arg;
- struct fib6_nh *nh = rt->fib6_nh;
+ struct fib6_nh *nh;
+ /* RA routes do not use nexthops */
+ if (rt->nh)
+ return 0;
+
+ nh = rt->fib6_nh;
if (((rt->fib6_flags & RTF_RA_ROUTER) == RTF_RA_ROUTER) &&
nh->fib_nh_gw_family && ipv6_addr_equal(gateway, &nh->fib_nh_gw6))
return -1;
@@ -4063,6 +4131,7 @@ static struct fib6_info *rt6_multipath_first_sibling(const struct fib6_info *rt)
return NULL;
}
+/* only called for fib entries with builtin fib6_nh */
static bool rt6_is_dead(const struct fib6_info *rt)
{
if (rt->fib6_nh->fib_nh_flags & RTNH_F_DEAD ||
@@ -4141,7 +4210,7 @@ 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 &&
+ if (rt != net->ipv6.fib6_null_entry && !rt->nh &&
rt->fib6_nh->fib_nh_dev == arg->dev) {
rt->fib6_nh->fib_nh_flags &= ~arg->nh_flags;
fib6_update_sernum_upto_root(net, rt);
@@ -4166,6 +4235,7 @@ void rt6_sync_up(struct net_device *dev, unsigned char nh_flags)
fib6_clean_all(dev_net(dev), fib6_ifup, &arg);
}
+/* only called for fib entries with inline fib6_nh */
static bool rt6_multipath_uses_dev(const struct fib6_info *rt,
const struct net_device *dev)
{
@@ -4226,7 +4296,7 @@ static int fib6_ifdown(struct fib6_info *rt, void *p_arg)
const struct net_device *dev = arg->dev;
struct net *net = dev_net(dev);
- if (rt == net->ipv6.fib6_null_entry)
+ if (rt == net->ipv6.fib6_null_entry || rt->nh)
return 0;
switch (arg->event) {
@@ -4780,6 +4850,9 @@ static size_t rt6_nlmsg_size(struct fib6_info *rt)
{
int nexthop_len = 0;
+ if (rt->nh)
+ nexthop_len += nla_total_size(4); /* RTA_NH_ID */
+
if (rt->fib6_nsiblings) {
nexthop_len = nla_total_size(0) /* RTA_MULTIPATH */
+ NLA_ALIGN(sizeof(struct rtnexthop))
@@ -4806,6 +4879,35 @@ static size_t rt6_nlmsg_size(struct fib6_info *rt)
+ nexthop_len;
}
+static int rt6_fill_node_nexthop(struct sk_buff *skb, struct nexthop *nh,
+ unsigned char *flags)
+{
+ if (nexthop_is_multipath(nh)) {
+ struct nlattr *mp;
+
+ mp = nla_nest_start(skb, RTA_MULTIPATH);
+ if (!mp)
+ goto nla_put_failure;
+
+ if (nexthop_mpath_fill_node(skb, nh))
+ goto nla_put_failure;
+
+ nla_nest_end(skb, mp);
+ } else {
+ struct fib6_nh *fib6_nh;
+
+ fib6_nh = nexthop_fib6_nh(nh);
+ if (fib_nexthop_info(skb, &fib6_nh->nh_common,
+ flags, false) < 0)
+ goto nla_put_failure;
+ }
+
+ return 0;
+
+nla_put_failure:
+ return -EMSGSIZE;
+}
+
static int rt6_fill_node(struct net *net, struct sk_buff *skb,
struct fib6_info *rt, struct dst_entry *dst,
struct in6_addr *dest, struct in6_addr *src,
@@ -4815,6 +4917,7 @@ static int rt6_fill_node(struct net *net, struct sk_buff *skb,
struct rt6_info *rt6 = (struct rt6_info *)dst;
struct rt6key *rt6_dst, *rt6_src;
u32 *pmetrics, table, rt6_flags;
+ unsigned char nh_flags = 0;
struct nlmsghdr *nlh;
struct rtmsg *rtm;
long expires = 0;
@@ -4934,9 +5037,18 @@ static int rt6_fill_node(struct net *net, struct sk_buff *skb,
}
nla_nest_end(skb, mp);
- } else {
- unsigned char nh_flags = 0;
+ } else if (rt->nh) {
+ if (nla_put_u32(skb, RTA_NH_ID, rt->nh->id))
+ goto nla_put_failure;
+
+ if (nexthop_is_blackhole(rt->nh))
+ rtm->rtm_type = RTN_BLACKHOLE;
+ if (rt6_fill_node_nexthop(skb, rt->nh, &nh_flags) < 0)
+ goto nla_put_failure;
+
+ rtm->rtm_flags |= nh_flags;
+ } else {
if (fib_nexthop_info(skb, &rt->fib6_nh->nh_common,
&nh_flags, false) < 0)
goto nla_put_failure;
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 5/7] mlxsw: Fail attempts to use routes with nexthop objects
2019-05-30 3:07 [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info David Ahern
` (3 preceding siblings ...)
2019-05-30 3:07 ` [PATCH net-next 4/7] ipv6: Plumb support for nexthop object in a fib6_info David Ahern
@ 2019-05-30 3:07 ` David Ahern
2019-05-30 8:10 ` Ido Schimmel
2019-05-30 3:07 ` [PATCH net-next 6/7] mlx5: " David Ahern
` (2 subsequent siblings)
7 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-05-30 3:07 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, saeedm, kafai, weiwan, David Ahern
From: David Ahern <dsahern@gmail.com>
Fail attempts to use nexthop objects with routes until support can be
properly added.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
index 4f781358aef1..23f17ea52061 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c
@@ -6122,6 +6122,20 @@ static int mlxsw_sp_router_fib_event(struct notifier_block *nb,
NL_SET_ERR_MSG_MOD(info->extack, "IPv6 gateway with IPv4 route is not supported");
return notifier_from_errno(-EINVAL);
}
+ if (fen_info->fi->nh) {
+ NL_SET_ERR_MSG_MOD(info->extack, "IPv4 route with nexthop objects is not supported");
+ return notifier_from_errno(-EINVAL);
+ }
+ } else if (info->family == AF_INET6) {
+ struct fib6_entry_notifier_info *fen6_info;
+
+ fen6_info = container_of(info,
+ struct fib6_entry_notifier_info,
+ info);
+ if (fen6_info->rt->nh) {
+ NL_SET_ERR_MSG_MOD(info->extack, "IPv6 route with nexthop objects is not supported");
+ return notifier_from_errno(-EINVAL);
+ }
}
break;
}
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 6/7] mlx5: Fail attempts to use routes with nexthop objects
2019-05-30 3:07 [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info David Ahern
` (4 preceding siblings ...)
2019-05-30 3:07 ` [PATCH net-next 5/7] mlxsw: Fail attempts to use routes with nexthop objects David Ahern
@ 2019-05-30 3:07 ` David Ahern
2019-05-30 3:08 ` [PATCH net-next 7/7] rocker: " David Ahern
2019-05-30 15:06 ` [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info Alexei Starovoitov
7 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-05-30 3:07 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, saeedm, kafai, weiwan, David Ahern
From: David Ahern <dsahern@gmail.com>
Fail attempts to use nexthop objects with routes until support can be
properly added.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c
index 2cbfaa8da7fc..e69766393990 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c
@@ -262,6 +262,10 @@ static int mlx5_lag_fib_event(struct notifier_block *nb,
fen_info = container_of(info, struct fib_entry_notifier_info,
info);
fi = fen_info->fi;
+ if (fi->nh) {
+ NL_SET_ERR_MSG_MOD(info->extack, "IPv4 route with nexthop objects is not supported");
+ return notifier_from_errno(-EINVAL);
+ }
fib_dev = fib_info_nh(fen_info->fi, 0)->fib_nh_dev;
if (fib_dev != ldev->pf[0].netdev &&
fib_dev != ldev->pf[1].netdev) {
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* [PATCH net-next 7/7] rocker: Fail attempts to use routes with nexthop objects
2019-05-30 3:07 [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info David Ahern
` (5 preceding siblings ...)
2019-05-30 3:07 ` [PATCH net-next 6/7] mlx5: " David Ahern
@ 2019-05-30 3:08 ` David Ahern
2019-05-30 15:06 ` [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info Alexei Starovoitov
7 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-05-30 3:08 UTC (permalink / raw)
To: davem, netdev; +Cc: idosch, saeedm, kafai, weiwan, David Ahern
From: David Ahern <dsahern@gmail.com>
Fail attempts to use nexthop objects with routes until support can be
properly added.
Signed-off-by: David Ahern <dsahern@gmail.com>
---
drivers/net/ethernet/rocker/rocker_main.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ethernet/rocker/rocker_main.c b/drivers/net/ethernet/rocker/rocker_main.c
index 7ae6c124bfe9..45b3325c3a38 100644
--- a/drivers/net/ethernet/rocker/rocker_main.c
+++ b/drivers/net/ethernet/rocker/rocker_main.c
@@ -2214,6 +2214,10 @@ static int rocker_router_fib_event(struct notifier_block *nb,
NL_SET_ERR_MSG_MOD(info->extack, "IPv6 gateway with IPv4 route is not supported");
return notifier_from_errno(-EINVAL);
}
+ if (fen_info->fi->nh) {
+ NL_SET_ERR_MSG_MOD(info->extack, "IPv4 route with nexthop objects is not supported");
+ return notifier_from_errno(-EINVAL);
+ }
}
memcpy(&fib_work->fen_info, ptr, sizeof(fib_work->fen_info));
--
2.11.0
^ permalink raw reply related [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 5/7] mlxsw: Fail attempts to use routes with nexthop objects
2019-05-30 3:07 ` [PATCH net-next 5/7] mlxsw: Fail attempts to use routes with nexthop objects David Ahern
@ 2019-05-30 8:10 ` Ido Schimmel
0 siblings, 0 replies; 29+ messages in thread
From: Ido Schimmel @ 2019-05-30 8:10 UTC (permalink / raw)
To: David Ahern; +Cc: davem, netdev, idosch, saeedm, kafai, weiwan, David Ahern
On Wed, May 29, 2019 at 08:07:58PM -0700, David Ahern wrote:
> From: David Ahern <dsahern@gmail.com>
>
> Fail attempts to use nexthop objects with routes until support can be
> properly added.
>
> 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 0/7] net: add struct nexthop to fib{6}_info
2019-05-30 3:07 [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info David Ahern
` (6 preceding siblings ...)
2019-05-30 3:08 ` [PATCH net-next 7/7] rocker: " David Ahern
@ 2019-05-30 15:06 ` Alexei Starovoitov
2019-05-30 15:15 ` David Ahern
7 siblings, 1 reply; 29+ messages in thread
From: Alexei Starovoitov @ 2019-05-30 15:06 UTC (permalink / raw)
To: David Ahern
Cc: David S. Miller, Network Development, Ido Schimmel,
Saeed Mahameed, Martin KaFai Lau, Wei Wang, David Ahern
On Wed, May 29, 2019 at 8:10 PM David Ahern <dsahern@kernel.org> wrote:
>
> From: David Ahern <dsahern@gmail.com>
>
> This sets adds 'struct nexthop' to fib_info and fib6_info. IPv4
> already handles multiple fib_nh entries in a single fib_info, so
> the conversion to use a nexthop struct is fairly mechanical. IPv6
> using a nexthop struct with a fib6_info impacts a lot of core logic
> which is built around the assumption of a single, builtin fib6_nh
> per fib6_info. To make this easier to review, this set adds
> nexthop to fib6_info and adds checks in most places fib6_info is
> used. The next set finishes the IPv6 conversion, walking through
> the places that need to consider all fib6_nh within a nexthop struct.
>
> Offload drivers - mlx5, mlxsw and rocker - are changed to fail FIB
> entries using nexthop objects. That limitation can be removed once
> the drivers are updated to properly support separate nexthops.
>
> This set starts by adding accessors for fib_nh and fib_nhs in a
> fib_info. This makes it easier to extract the number of nexthops
> in the fib entry and a specific fib_nh once the entry references
> a struct nexthop. Patch 2 converts more of IPv4 code to use
> fib_nh_common allowing a struct nexthop to use a fib6_nh with an
> IPv4 entry.
>
> Patches 3 and 4 add 'struct nexthop' to fib{6}_info and update
> references to both take a different path when it is set. New
> exported functions are added to the nexthop code to validate a
> nexthop struct when configured for use with a fib entry. IPv4
> is allowed to use a nexthop with either v4 or v6 entries. IPv6
> is limited to v6 entries only. In both cases list_heads track
> the fib entries using a nexthop struct for fast correlation on
> events (e.g., device events or nexthop events like delete or
> replace).
>
> The last 3 patches add hooks to drivers listening for FIB
> notificationas. All 3 of them reject the routes as unsupported,
> returning an error message to the user via extack. For mlxsw
> at least this is a stop gap measure until the driver is updated for
> proper support.
>
> David Ahern (7):
> ipv4: Use accessors for fib_info nexthop data
> ipv4: Prepare for fib6_nh from a nexthop object
> ipv4: Plumb support for nexthop object in a fib_info
> ipv6: Plumb support for nexthop object in a fib6_info
> mlxsw: Fail attempts to use routes with nexthop objects
> mlx5: Fail attempts to use routes with nexthop objects
> rocker: Fail attempts to use routes with nexthop objects
>
> drivers/net/ethernet/mellanox/mlx5/core/lag_mp.c | 33 ++-
> .../net/ethernet/mellanox/mlxsw/spectrum_router.c | 33 ++-
> drivers/net/ethernet/rocker/rocker_main.c | 4 +
> drivers/net/ethernet/rocker/rocker_ofdpa.c | 25 +-
> include/net/ip6_fib.h | 11 +-
> include/net/ip6_route.h | 13 +-
> include/net/ip_fib.h | 25 +-
> include/net/nexthop.h | 113 +++++++++
> net/core/filter.c | 3 +-
> net/ipv4/fib_frontend.c | 15 +-
> net/ipv4/fib_lookup.h | 1 +
> net/ipv4/fib_rules.c | 8 +-
> net/ipv4/fib_semantics.c | 257 ++++++++++++++-------
> net/ipv4/fib_trie.c | 38 ++-
> net/ipv4/nexthop.c | 111 ++++++++-
> net/ipv4/route.c | 5 +-
> net/ipv6/addrconf.c | 5 +
> net/ipv6/ip6_fib.c | 22 +-
> net/ipv6/ndisc.c | 3 +-
> net/ipv6/route.c | 156 +++++++++++--
> 20 files changed, 706 insertions(+), 175 deletions(-)
Huge number of core changes and zero tests.
Nacked-by: Alexei Starovoitov <ast@kernel.org>
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-30 15:06 ` [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info Alexei Starovoitov
@ 2019-05-30 15:15 ` David Ahern
2019-05-30 15:18 ` Alexei Starovoitov
0 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-05-30 15:15 UTC (permalink / raw)
To: Alexei Starovoitov, David Ahern
Cc: David S. Miller, Network Development, Ido Schimmel,
Saeed Mahameed, Martin KaFai Lau, Wei Wang
On 5/30/19 9:06 AM, Alexei Starovoitov wrote:
> Huge number of core changes and zero tests.
As mentioned in a past response, there are a number of tests under
selftests that exercise the code paths affected by this change.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-30 15:15 ` David Ahern
@ 2019-05-30 15:18 ` Alexei Starovoitov
2019-05-30 18:01 ` David Miller
0 siblings, 1 reply; 29+ messages in thread
From: Alexei Starovoitov @ 2019-05-30 15:18 UTC (permalink / raw)
To: David Ahern
Cc: David Ahern, David S. Miller, Network Development, Ido Schimmel,
Saeed Mahameed, Martin KaFai Lau, Wei Wang
On Thu, May 30, 2019 at 8:16 AM David Ahern <dsahern@gmail.com> wrote:
>
> On 5/30/19 9:06 AM, Alexei Starovoitov wrote:
> > Huge number of core changes and zero tests.
>
> As mentioned in a past response, there are a number of tests under
> selftests that exercise the code paths affected by this change.
I see zero new tests added.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-30 15:18 ` Alexei Starovoitov
@ 2019-05-30 18:01 ` David Miller
2019-05-30 18:27 ` Alexei Starovoitov
0 siblings, 1 reply; 29+ messages in thread
From: David Miller @ 2019-05-30 18:01 UTC (permalink / raw)
To: alexei.starovoitov
Cc: dsahern, dsahern, netdev, idosch, saeedm, kafai, weiwan
From: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Date: Thu, 30 May 2019 08:18:10 -0700
> On Thu, May 30, 2019 at 8:16 AM David Ahern <dsahern@gmail.com> wrote:
>>
>> On 5/30/19 9:06 AM, Alexei Starovoitov wrote:
>> > Huge number of core changes and zero tests.
>>
>> As mentioned in a past response, there are a number of tests under
>> selftests that exercise the code paths affected by this change.
>
> I see zero new tests added.
If the existing tests give sufficient coverage, your objections are not
reasonable Alexei.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-30 18:01 ` David Miller
@ 2019-05-30 18:27 ` Alexei Starovoitov
2019-05-30 18:52 ` David Ahern
2019-05-30 18:56 ` David Miller
0 siblings, 2 replies; 29+ messages in thread
From: Alexei Starovoitov @ 2019-05-30 18:27 UTC (permalink / raw)
To: David Miller
Cc: David Ahern, David Ahern, Network Development, Ido Schimmel,
Saeed Mahameed, Martin KaFai Lau, Wei Wang
On Thu, May 30, 2019 at 11:01 AM David Miller <davem@davemloft.net> wrote:
>
> From: Alexei Starovoitov <alexei.starovoitov@gmail.com>
> Date: Thu, 30 May 2019 08:18:10 -0700
>
> > On Thu, May 30, 2019 at 8:16 AM David Ahern <dsahern@gmail.com> wrote:
> >>
> >> On 5/30/19 9:06 AM, Alexei Starovoitov wrote:
> >> > Huge number of core changes and zero tests.
> >>
> >> As mentioned in a past response, there are a number of tests under
> >> selftests that exercise the code paths affected by this change.
> >
> > I see zero new tests added.
>
> If the existing tests give sufficient coverage, your objections are not
> reasonable Alexei.
I completely disagree. Existing tests are not sufficient.
It is a new feature for the kernel with corresponding iproute2 new features,
yet there are zero tests.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-30 18:27 ` Alexei Starovoitov
@ 2019-05-30 18:52 ` David Ahern
2019-05-30 18:57 ` David Miller
2019-05-30 20:01 ` Alexei Starovoitov
2019-05-30 18:56 ` David Miller
1 sibling, 2 replies; 29+ messages in thread
From: David Ahern @ 2019-05-30 18:52 UTC (permalink / raw)
To: Alexei Starovoitov, David Miller
Cc: David Ahern, Network Development, Ido Schimmel, Saeed Mahameed,
Martin KaFai Lau, Wei Wang
On 5/30/19 12:27 PM, Alexei Starovoitov wrote:
> On Thu, May 30, 2019 at 11:01 AM David Miller <davem@davemloft.net> wrote:
>>
>> From: Alexei Starovoitov <alexei.starovoitov@gmail.com>
>> Date: Thu, 30 May 2019 08:18:10 -0700
>>
>>> On Thu, May 30, 2019 at 8:16 AM David Ahern <dsahern@gmail.com> wrote:
>>>>
>>>> On 5/30/19 9:06 AM, Alexei Starovoitov wrote:
>>>>> Huge number of core changes and zero tests.
>>>>
>>>> As mentioned in a past response, there are a number of tests under
>>>> selftests that exercise the code paths affected by this change.
>>>
>>> I see zero new tests added.
>>
>> If the existing tests give sufficient coverage, your objections are not
>> reasonable Alexei.
>
> I completely disagree. Existing tests are not sufficient.
> It is a new feature for the kernel with corresponding iproute2 new features,
> yet there are zero tests.
>
Your objection was based on changes to core code ("Huge number of core
changes and zero tests"), not new code paths.
The nexthop code paths are not live yet. More changes are needed before
that can happen. I have been sending the patches in small, reviewable
increments to be kind to reviewers than a single 27 patch set with
everything (the remaining set which is over the limit BTW).
Once iproute2 has the nexthop command (patches sent to the list) AND the
RTA_NHID patches are in, I have this as the final set:
8c0b06b9813e selftests: Add test cases for nexthop objects
ea5c19e4dc7c selftests: Add nexthop objects to router_multipath.sh
3be7b15d1e56 selftests: pmtu: Move running of test into a new function
a896b2206ea5 selftests: pmtu: Move route installs to a new function
cfa48193d0b8 selftests: pmtu: Add support for routing via nexthop objects
3d09a79208b9 selftests: icmp_redirect: Add support for routing via
nexthop objects
That includes:
1. a test script doing functional validation of the nexthop code
(net/ipv4/nexthop.c), along with stack integration (e.g., v6 nexthop
with v4 routes) and traffic (ping).
2. updates to existing exception tests (pmtu and redirect) with the
nexthop obects used for routing
3. updates to the existing router_multipath script with the nexthop
obects used for routing and validating balanced selection in paths.
As always, my patches can be viewed by anyone:
https://github.com/dsahern/linux
Latest branch is:
https://github.com/dsahern/linux/tree/5.2-next-nexthops-v14
In addition to the functional and runtime tests listed above:
1. I have all kinds of MPLS and network functional tests that I run. I
have already committed to sending those for inclusion, it is a matter of
time.
2. The FRR team has had a kernel with these patches for 5+ months adding
support for nexthops to FRR. They have their own test suites used for
their development.
3. The patches are included in our 4.19 kernel and subjected to our
suite of smoke and validation tests.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-30 18:27 ` Alexei Starovoitov
2019-05-30 18:52 ` David Ahern
@ 2019-05-30 18:56 ` David Miller
1 sibling, 0 replies; 29+ messages in thread
From: David Miller @ 2019-05-30 18:56 UTC (permalink / raw)
To: alexei.starovoitov
Cc: dsahern, dsahern, netdev, idosch, saeedm, kafai, weiwan
From: Alexei Starovoitov <alexei.starovoitov@gmail.com>
Date: Thu, 30 May 2019 11:27:23 -0700
> On Thu, May 30, 2019 at 11:01 AM David Miller <davem@davemloft.net> wrote:
>>
>> From: Alexei Starovoitov <alexei.starovoitov@gmail.com>
>> Date: Thu, 30 May 2019 08:18:10 -0700
>>
>> > On Thu, May 30, 2019 at 8:16 AM David Ahern <dsahern@gmail.com> wrote:
>> >>
>> >> On 5/30/19 9:06 AM, Alexei Starovoitov wrote:
>> >> > Huge number of core changes and zero tests.
>> >>
>> >> As mentioned in a past response, there are a number of tests under
>> >> selftests that exercise the code paths affected by this change.
>> >
>> > I see zero new tests added.
>>
>> If the existing tests give sufficient coverage, your objections are not
>> reasonable Alexei.
>
> I completely disagree. Existing tests are not sufficient.
> It is a new feature for the kernel with corresponding iproute2 new features,
> yet there are zero tests.
Ok, that's tree, and I agree.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-30 18:52 ` David Ahern
@ 2019-05-30 18:57 ` David Miller
2019-05-30 18:58 ` David Ahern
2019-05-30 20:01 ` Alexei Starovoitov
1 sibling, 1 reply; 29+ messages in thread
From: David Miller @ 2019-05-30 18:57 UTC (permalink / raw)
To: dsahern
Cc: alexei.starovoitov, dsahern, netdev, idosch, saeedm, kafai, weiwan
From: David Ahern <dsahern@gmail.com>
Date: Thu, 30 May 2019 12:52:43 -0600
> The nexthop code paths are not live yet. More changes are needed before
> that can happen. I have been sending the patches in small, reviewable
> increments to be kind to reviewers than a single 27 patch set with
> everything (the remaining set which is over the limit BTW).
We definitely need tests using the new netlink attributes even if the
code paths aren't live yet.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-30 18:57 ` David Miller
@ 2019-05-30 18:58 ` David Ahern
0 siblings, 0 replies; 29+ messages in thread
From: David Ahern @ 2019-05-30 18:58 UTC (permalink / raw)
To: David Miller
Cc: alexei.starovoitov, dsahern, netdev, idosch, saeedm, kafai, weiwan
On 5/30/19 12:57 PM, David Miller wrote:
> From: David Ahern <dsahern@gmail.com>
> Date: Thu, 30 May 2019 12:52:43 -0600
>
>> The nexthop code paths are not live yet. More changes are needed before
>> that can happen. I have been sending the patches in small, reviewable
>> increments to be kind to reviewers than a single 27 patch set with
>> everything (the remaining set which is over the limit BTW).
>
> We definitely need tests using the new netlink attributes even if the
> code paths aren't live yet.
>
https://github.com/dsahern/linux/commit/8c0b06b9813e74c561c8626de862fb649f615cb1
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-30 18:52 ` David Ahern
2019-05-30 18:57 ` David Miller
@ 2019-05-30 20:01 ` Alexei Starovoitov
2019-05-31 21:29 ` David Miller
1 sibling, 1 reply; 29+ messages in thread
From: Alexei Starovoitov @ 2019-05-30 20:01 UTC (permalink / raw)
To: David Ahern
Cc: David Miller, David Ahern, Network Development, Ido Schimmel,
Saeed Mahameed, Martin KaFai Lau, Wei Wang
On Thu, May 30, 2019 at 11:52 AM David Ahern <dsahern@gmail.com> wrote:
>
> On 5/30/19 12:27 PM, Alexei Starovoitov wrote:
> > On Thu, May 30, 2019 at 11:01 AM David Miller <davem@davemloft.net> wrote:
> >>
> >> From: Alexei Starovoitov <alexei.starovoitov@gmail.com>
> >> Date: Thu, 30 May 2019 08:18:10 -0700
> >>
> >>> On Thu, May 30, 2019 at 8:16 AM David Ahern <dsahern@gmail.com> wrote:
> >>>>
> >>>> On 5/30/19 9:06 AM, Alexei Starovoitov wrote:
> >>>>> Huge number of core changes and zero tests.
> >>>>
> >>>> As mentioned in a past response, there are a number of tests under
> >>>> selftests that exercise the code paths affected by this change.
> >>>
> >>> I see zero new tests added.
> >>
> >> If the existing tests give sufficient coverage, your objections are not
> >> reasonable Alexei.
> >
> > I completely disagree. Existing tests are not sufficient.
> > It is a new feature for the kernel with corresponding iproute2 new features,
> > yet there are zero tests.
> >
>
> Your objection was based on changes to core code ("Huge number of core
> changes and zero tests"), not new code paths.
>
> The nexthop code paths are not live yet. More changes are needed before
> that can happen. I have been sending the patches in small, reviewable
> increments to be kind to reviewers than a single 27 patch set with
> everything (the remaining set which is over the limit BTW).
>
> Once iproute2 has the nexthop command (patches sent to the list) AND the
> RTA_NHID patches are in, I have this as the final set:
>
> 8c0b06b9813e selftests: Add test cases for nexthop objects
> ea5c19e4dc7c selftests: Add nexthop objects to router_multipath.sh
> 3be7b15d1e56 selftests: pmtu: Move running of test into a new function
> a896b2206ea5 selftests: pmtu: Move route installs to a new function
> cfa48193d0b8 selftests: pmtu: Add support for routing via nexthop objects
> 3d09a79208b9 selftests: icmp_redirect: Add support for routing via
> nexthop objects
cool. that's exactly what I was asking about.
Could you please post them to the mailing list as RFC or some
not-yet-to-be-merged tag?
And in the main patch set, just say in cover letter:
"here look at this set posted separately, because of such and such"
> As always, my patches can be viewed by anyone:
> https://github.com/dsahern/linux
sure, but it's counter to the standard code review procedure we have.
Pointing a link to your tree doesn't make them reviewable on the
public mailing list.
I very much prefer to do code reviews by looking at the tests first.
It gives me a better idea on api, use cases, etc.
> That includes:
> 1. a test script doing functional validation of the nexthop code
> (net/ipv4/nexthop.c), along with stack integration (e.g., v6 nexthop
> with v4 routes) and traffic (ping).
>
> 2. updates to existing exception tests (pmtu and redirect) with the
> nexthop obects used for routing
>
> 3. updates to the existing router_multipath script with the nexthop
> obects used for routing and validating balanced selection in paths.
great.
>
> Latest branch is:
> https://github.com/dsahern/linux/tree/5.2-next-nexthops-v14
>
> In addition to the functional and runtime tests listed above:
> 1. I have all kinds of MPLS and network functional tests that I run. I
> have already committed to sending those for inclusion, it is a matter of
> time.
cool. first time I hear.
> 2. The FRR team has had a kernel with these patches for 5+ months adding
> support for nexthops to FRR. They have their own test suites used for
> their development.
what is FRR team? First time I hear about them as well.
> 3. The patches are included in our 4.19 kernel and subjected to our
> suite of smoke and validation tests.
it's also great to hear, but your kernel tree is not a substitute
for netdev community with many companies contributing and running
those kernels, syzbots, buildbots, etc.
I have no doubt that you test your patches, but these patches also touch
the very core of networking that we all are care deeply about.
Hence we want to be able to run all those tests on our kernels too.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-30 20:01 ` Alexei Starovoitov
@ 2019-05-31 21:29 ` David Miller
2019-05-31 21:38 ` David Ahern
0 siblings, 1 reply; 29+ messages in thread
From: David Miller @ 2019-05-31 21:29 UTC (permalink / raw)
To: alexei.starovoitov
Cc: dsahern, dsahern, netdev, idosch, saeedm, kafai, weiwan
David, can you add some supplementary information to your cover letter
et al. which seems to be part of what Alexei is asking for and seems
quite reasonable?
Thanks.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-31 21:29 ` David Miller
@ 2019-05-31 21:38 ` David Ahern
2019-06-01 1:04 ` Eric Dumazet
` (2 more replies)
0 siblings, 3 replies; 29+ messages in thread
From: David Ahern @ 2019-05-31 21:38 UTC (permalink / raw)
To: David Miller, alexei.starovoitov
Cc: dsahern, netdev, idosch, saeedm, kafai, weiwan
On 5/31/19 3:29 PM, David Miller wrote:
> David, can you add some supplementary information to your cover letter
> et al. which seems to be part of what Alexei is asking for and seems
> quite reasonable?
>
It is not clear to me what more is wanted in the cover letter. His
complaints were on lack of tests. I sent those separately; they really
aren't tied to this specific set (most of them apply to the previous set
which provided the uapi and core implementation). This set is mostly
mechanical in adding wrappers to 2 fields and shoving existing code into
the else branch of 'if (fi->nh) { new code } else { existing code }'
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-31 21:38 ` David Ahern
@ 2019-06-01 1:04 ` Eric Dumazet
2019-06-01 2:29 ` David Ahern
2019-06-01 2:51 ` Alexei Starovoitov
2019-06-03 16:50 ` Willem de Bruijn
2 siblings, 1 reply; 29+ messages in thread
From: Eric Dumazet @ 2019-06-01 1:04 UTC (permalink / raw)
To: David Ahern, David Miller, alexei.starovoitov
Cc: dsahern, netdev, idosch, saeedm, kafai, weiwan
On 5/31/19 2:38 PM, David Ahern wrote:
> On 5/31/19 3:29 PM, David Miller wrote:
>> David, can you add some supplementary information to your cover letter
>> et al. which seems to be part of what Alexei is asking for and seems
>> quite reasonable?
>>
>
> It is not clear to me what more is wanted in the cover letter. His
> complaints were on lack of tests. I sent those separately; they really
> aren't tied to this specific set (most of them apply to the previous set
> which provided the uapi and core implementation). This set is mostly
> mechanical in adding wrappers to 2 fields and shoving existing code into
> the else branch of 'if (fi->nh) { new code } else { existing code }'
>
Hi David
I have a bunch (about 15 ) of syzbot reports, probably caused to your latest patch series.
Do we want to stabilize first, or do you expect this new patch series to fix
these issues ?
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-06-01 1:04 ` Eric Dumazet
@ 2019-06-01 2:29 ` David Ahern
2019-06-01 2:34 ` Eric Dumazet
0 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-06-01 2:29 UTC (permalink / raw)
To: Eric Dumazet, David Miller, alexei.starovoitov
Cc: dsahern, netdev, idosch, saeedm, kafai, weiwan
On 5/31/19 7:04 PM, Eric Dumazet wrote:
>
> I have a bunch (about 15 ) of syzbot reports, probably caused to your latest patch series.
>
> Do we want to stabilize first, or do you expect this new patch series to fix
> these issues ?
>
Please forward. I will take a look.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-06-01 2:29 ` David Ahern
@ 2019-06-01 2:34 ` Eric Dumazet
2019-06-01 2:38 ` Eric Dumazet
0 siblings, 1 reply; 29+ messages in thread
From: Eric Dumazet @ 2019-06-01 2:34 UTC (permalink / raw)
To: David Ahern, Eric Dumazet, David Miller, alexei.starovoitov
Cc: dsahern, netdev, idosch, saeedm, kafai, weiwan
On 5/31/19 7:29 PM, David Ahern wrote:
> On 5/31/19 7:04 PM, Eric Dumazet wrote:
>>
>> I have a bunch (about 15 ) of syzbot reports, probably caused to your latest patch series.
>>
>> Do we want to stabilize first, or do you expect this new patch series to fix
>> these issues ?
>>
>
> Please forward. I will take a look.
>
I will release one of them, I suspect they are duplicates.
This is why I was looking at rt6_get_pcpu_route()
I thought we were lacking a READ_ONCE() but this does not seem needed.
https://patchwork.ozlabs.org/patch/1108632/
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-06-01 2:34 ` Eric Dumazet
@ 2019-06-01 2:38 ` Eric Dumazet
0 siblings, 0 replies; 29+ messages in thread
From: Eric Dumazet @ 2019-06-01 2:38 UTC (permalink / raw)
To: David Ahern, David Miller, alexei.starovoitov
Cc: dsahern, netdev, idosch, saeedm, kafai, weiwan
On 5/31/19 7:34 PM, Eric Dumazet wrote:
>
>
> On 5/31/19 7:29 PM, David Ahern wrote:
>> On 5/31/19 7:04 PM, Eric Dumazet wrote:
>>>
>>> I have a bunch (about 15 ) of syzbot reports, probably caused to your latest patch series.
>>>
>>> Do we want to stabilize first, or do you expect this new patch series to fix
>>> these issues ?
>>>
>>
>> Please forward. I will take a look.
>>
>
> I will release one of them, I suspect they are duplicates.
>
> This is why I was looking at rt6_get_pcpu_route()
>
> I thought we were lacking a READ_ONCE() but this does not seem needed.
>
> https://patchwork.ozlabs.org/patch/1108632/
>
I will release a second report, please not they are no repro yet.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-31 21:38 ` David Ahern
2019-06-01 1:04 ` Eric Dumazet
@ 2019-06-01 2:51 ` Alexei Starovoitov
2019-06-01 2:59 ` David Ahern
2019-06-03 16:50 ` Willem de Bruijn
2 siblings, 1 reply; 29+ messages in thread
From: Alexei Starovoitov @ 2019-06-01 2:51 UTC (permalink / raw)
To: David Ahern
Cc: David Miller, David Ahern, Network Development, Ido Schimmel,
Saeed Mahameed, Martin KaFai Lau, Wei Wang
On Fri, May 31, 2019 at 2:38 PM David Ahern <dsahern@gmail.com> wrote:
>
> On 5/31/19 3:29 PM, David Miller wrote:
> > David, can you add some supplementary information to your cover letter
> > et al. which seems to be part of what Alexei is asking for and seems
> > quite reasonable?
> >
>
> It is not clear to me what more is wanted in the cover letter. His
> complaints were on lack of tests. I sent those separately; they really
> aren't tied to this specific set (most of them apply to the previous set
> which provided the uapi and core implementation). This set is mostly
> mechanical in adding wrappers to 2 fields and shoving existing code into
> the else branch of 'if (fi->nh) { new code } else { existing code }'
Are you talking about this one?
https://patchwork.ozlabs.org/patch/1107892/
From single sentence of commit log it's not clear at all
whether they're related to this thread.
Will they fail if run w/o this set?
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-06-01 2:51 ` Alexei Starovoitov
@ 2019-06-01 2:59 ` David Ahern
2019-06-01 3:10 ` Alexei Starovoitov
0 siblings, 1 reply; 29+ messages in thread
From: David Ahern @ 2019-06-01 2:59 UTC (permalink / raw)
To: Alexei Starovoitov
Cc: David Miller, David Ahern, Network Development, Ido Schimmel,
Saeed Mahameed, Martin KaFai Lau, Wei Wang
On 5/31/19 8:51 PM, Alexei Starovoitov wrote:
> From single sentence of commit log it's not clear at all
> whether they're related to this thread.
> Will they fail if run w/o this set?
>
New code in this set (if (fi->nh) {}) can not be tested yet. It can not
be tested until the final patch that wires up nexthops with fib entries.
Old code in this set (else {}) is tested by existing selftests.
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-06-01 2:59 ` David Ahern
@ 2019-06-01 3:10 ` Alexei Starovoitov
0 siblings, 0 replies; 29+ messages in thread
From: Alexei Starovoitov @ 2019-06-01 3:10 UTC (permalink / raw)
To: David Ahern
Cc: David Miller, David Ahern, Network Development, Ido Schimmel,
Saeed Mahameed, Martin KaFai Lau, Wei Wang
On Fri, May 31, 2019 at 7:59 PM David Ahern <dsahern@gmail.com> wrote:
>
> On 5/31/19 8:51 PM, Alexei Starovoitov wrote:
> > From single sentence of commit log it's not clear at all
> > whether they're related to this thread.
> > Will they fail if run w/o this set?
> >
>
> New code in this set (if (fi->nh) {}) can not be tested yet. It can not
> be tested until the final patch that wires up nexthops with fib entries.
Could you post the whole thing please?
This set to review/apply and whatever number of
supporting sets _including_ final tests as RFC ?
^ permalink raw reply [flat|nested] 29+ messages in thread
* Re: [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info
2019-05-31 21:38 ` David Ahern
2019-06-01 1:04 ` Eric Dumazet
2019-06-01 2:51 ` Alexei Starovoitov
@ 2019-06-03 16:50 ` Willem de Bruijn
2 siblings, 0 replies; 29+ messages in thread
From: Willem de Bruijn @ 2019-06-03 16:50 UTC (permalink / raw)
To: David Ahern
Cc: David Miller, Alexei Starovoitov, David Ahern,
Network Development, Ido Schimmel, Saeed Mahameed,
Martin KaFai Lau, weiwan
On Fri, May 31, 2019 at 5:38 PM David Ahern <dsahern@gmail.com> wrote:
>
> On 5/31/19 3:29 PM, David Miller wrote:
> > David, can you add some supplementary information to your cover letter
> > et al. which seems to be part of what Alexei is asking for and seems
> > quite reasonable?
> >
>
> It is not clear to me what more is wanted in the cover letter.
A bit late and people in this thread probably already do know the context.
But for those like me that did not: a summary or pointer to the
overall goal and design would be informative. A search dug up the RFC
from last year. That definitely helped me understand the intent a bit
better.
net: Improve route scalability via support for nexthop objects
https://lwn.net/Articles/763950/
^ permalink raw reply [flat|nested] 29+ messages in thread
end of thread, other threads:[~2019-06-03 16:50 UTC | newest]
Thread overview: 29+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-05-30 3:07 [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info David Ahern
2019-05-30 3:07 ` [PATCH net-next 1/7] ipv4: Use accessors for fib_info nexthop data David Ahern
2019-05-30 3:07 ` [PATCH net-next 2/7] ipv4: Prepare for fib6_nh from a nexthop object David Ahern
2019-05-30 3:07 ` [PATCH net-next 3/7] ipv4: Plumb support for nexthop object in a fib_info David Ahern
2019-05-30 3:07 ` [PATCH net-next 4/7] ipv6: Plumb support for nexthop object in a fib6_info David Ahern
2019-05-30 3:07 ` [PATCH net-next 5/7] mlxsw: Fail attempts to use routes with nexthop objects David Ahern
2019-05-30 8:10 ` Ido Schimmel
2019-05-30 3:07 ` [PATCH net-next 6/7] mlx5: " David Ahern
2019-05-30 3:08 ` [PATCH net-next 7/7] rocker: " David Ahern
2019-05-30 15:06 ` [PATCH net-next 0/7] net: add struct nexthop to fib{6}_info Alexei Starovoitov
2019-05-30 15:15 ` David Ahern
2019-05-30 15:18 ` Alexei Starovoitov
2019-05-30 18:01 ` David Miller
2019-05-30 18:27 ` Alexei Starovoitov
2019-05-30 18:52 ` David Ahern
2019-05-30 18:57 ` David Miller
2019-05-30 18:58 ` David Ahern
2019-05-30 20:01 ` Alexei Starovoitov
2019-05-31 21:29 ` David Miller
2019-05-31 21:38 ` David Ahern
2019-06-01 1:04 ` Eric Dumazet
2019-06-01 2:29 ` David Ahern
2019-06-01 2:34 ` Eric Dumazet
2019-06-01 2:38 ` Eric Dumazet
2019-06-01 2:51 ` Alexei Starovoitov
2019-06-01 2:59 ` David Ahern
2019-06-01 3:10 ` Alexei Starovoitov
2019-06-03 16:50 ` Willem de Bruijn
2019-05-30 18:56 ` David Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).