Locally-generated MC packets have so far not been subject to MC routing.
Instead an MC-enabled installation would maintain the MC routing tables,
and separately from that the list of interfaces to send packets to as part
of the VXLAN FDB and MDB.
In a previous patch, a ip_mr_output() and ip6_mr_output() routines were
added for IPv4 and IPv6. All locally generated MC traffic is now passed
through these functions. For reasons of backward compatibility, an SKB
(IPCB / IP6CB) flag guards the actual MC routing.
This patch adds logic to set the flag, and the UAPI to enable the behavior.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Link: https://patch.msgid.link/d899655bb7e9b2521ee8c793e67056b9fd02ba12.1750113335.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
FDB entries are currently stored in a hash table with a fixed number of
buckets (256), resulting in performance degradation as the number of
entries grows. Solve this by converting the driver to use rhashtable
which maintains more or less constant performance regardless of the
number of entries.
Measured transmitted packets per second using a single pktgen thread
with varying number of entries when the transmitted packet always hits
the default entry (worst case):
Number of entries | Improvement
------------------|------------
1k | +1.12%
4k | +9.22%
16k | +55%
64k | +585%
256k | +2460%
In addition, the change reduces the size of the VXLAN device structure
from 2584 bytes to 672 bytes.
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20250415121143.345227-16-idosch@nvidia.com
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Currently, FDB entries are stored in a hash table with a fixed number of
buckets. The table is used for both lookups and entry traversal.
Subsequent patches will convert the table to rhashtable which is not
suitable for entry traversal.
In preparation for this conversion, add FDB entries to a linked list.
Subsequent patches will convert the driver to use this list when
traversing entries during dump, flush, etc.
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20250415121143.345227-8-idosch@nvidia.com
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Currently, the VXLAN driver stores FDB entries in a hash table with a
fixed number of buckets (256). Subsequent patches are going to convert
this table to rhashtable with a linked list for entry traversal, as
rhashtable is more scalable.
In preparation for this conversion, move from a per-bucket spin lock to
a single spin lock that protects the entire FDB table.
The per-bucket spin locks were introduced by commit fe1e0713bb
("vxlan: Use FDB_HASH_SIZE hash_locks to reduce contention") citing
"huge contention when inserting/deleting vxlan_fdbs into the fdb_head".
It is not clear from the commit message which code path was holding the
spin lock for long periods of time, but the obvious suspect is the FDB
cleanup routine (vxlan_cleanup()) that periodically traverses the entire
table in order to delete aged-out entries.
This will be solved by subsequent patches that will convert the FDB
cleanup routine to traverse the linked list of FDB entries using RCU,
only acquiring the spin lock when deleting an aged-out entry.
The change reduces the size of the VXLAN device structure from 3600
bytes to 2576 bytes.
Reviewed-by: Petr Machata <petrm@nvidia.com>
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/20250415121143.345227-7-idosch@nvidia.com
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
In order to make it possible to configure which bits in VXLAN header should
be considered reserved, introduce a new field vxlan_config::reserved_bits.
Have it cover the whole header, except for the VNI-present bit and the bits
for VNI itself, and have individual enabled features clear more bits off
reserved_bits.
(This is expressed as first constructing a used_bits set, and then
inverting it to get the reserved_bits. The set of used_bits will be useful
on its own for validation of user-set reserved_bits in a following patch.)
The patch also moves a comment relevant to the validation from the unparsed
validation site up to the new site. Logically this patch should add the new
comment, and a later patch that removes the unparsed bits would remove the
old comment. But keeping both legs in the same patch is better from the
history spelunking point of view.
Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Link: https://patch.msgid.link/984dbf98d5940d3900268dbffaf70961f731d4a4.1733412063.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
VXLAN-GPE does not add an extra inner Ethernet header. Take that into
account when calculating header length.
This causes problems in skb_tunnel_check_pmtu, where incorrect PMTU is
cached.
In the collect_md mode (which is the only mode that VXLAN-GPE
supports), there's no magic auto-setting of the tunnel interface MTU.
It can't be, since the destination and thus the underlying interface
may be different for each packet.
So, the administrator is responsible for setting the correct tunnel
interface MTU. Apparently, the administrators are capable enough to
calculate that the maximum MTU for VXLAN-GPE is (their_lower_MTU - 36).
They set the tunnel interface MTU to 1464. If you run a TCP stream over
such interface, it's then segmented according to the MTU 1464, i.e.
producing 1514 bytes frames. Which is okay, this still fits the lower
MTU.
However, skb_tunnel_check_pmtu (called from vxlan_xmit_one) uses 50 as
the header size and thus incorrectly calculates the frame size to be
1528. This leads to ICMP too big message being generated (locally),
PMTU of 1450 to be cached and the TCP stream to be resegmented.
The fix is to use the correct actual header size, especially for
skb_tunnel_check_pmtu calculation.
Fixes: e1e5314de0 ("vxlan: implement GPE")
Signed-off-by: Jiri Benc <jbenc@redhat.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If a packet needs to be encapsulated towards a local destination IP, the
packet will undergo a "local bypass" and be injected into the Rx path as
if it was received by the target VXLAN device without undergoing
encapsulation. If such a device does not exist, the packet will be
dropped.
There are scenarios where we do not want to perform such a bypass, but
instead want the packet to be encapsulated and locally received by a
user space program for post-processing.
To that end, add a new VXLAN device attribute that controls whether a
"local bypass" is performed or not. Default to performing a bypass to
maintain existing behavior.
Signed-off-by: Vladimir Nikishkin <vladimir@nikishkin.pw>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add an internal flag to indicate whether MDB entries are configured or
not. Set the flag after installing the first MDB entry and clear it
before deleting the last one.
The flag will be consulted by the data path which will only perform an
MDB lookup if the flag is set, thereby keeping the MDB overhead to a
minimum when the MDB is not used.
Another option would have been to use a static key, but it is global and
not per-device, unlike the current approach.
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Implement MDB control path support, enabling the creation, deletion,
replacement and dumping of MDB entries in a similar fashion to the
bridge driver. Unlike the bridge driver, each entry stores a list of
remote VTEPs to which matched packets need to be replicated to and not a
list of bridge ports.
The motivating use case is the installation of MDB entries by a user
space control plane in response to received EVPN routes. As such, only
allow permanent MDB entries to be installed and do not implement
snooping functionality, avoiding a lot of unnecessary complexity.
Since entries can only be modified by user space under RTNL, use RTNL as
the write lock. Use RCU to ensure that MDB entries and remotes are not
freed while being accessed from the data path during transmission.
In terms of uAPI, reuse the existing MDB netlink interface, but add a
few new attributes to request and response messages:
* IP address of the destination VXLAN tunnel endpoint where the
multicast receivers reside.
* UDP destination port number to use to connect to the remote VXLAN
tunnel endpoint.
* VXLAN VNI Network Identifier to use to connect to the remote VXLAN
tunnel endpoint. Required when Ingress Replication (IR) is used and
the remote VTEP is not a member of originating broadcast domain
(VLAN/VNI) [1].
* Source VNI Network Identifier the MDB entry belongs to. Used only when
the VXLAN device is in external mode.
* Interface index of the outgoing interface to reach the remote VXLAN
tunnel endpoint. This is required when the underlay destination IP is
multicast (P2MP), as the multicast routing tables are not consulted.
All the new attributes are added under the 'MDBA_SET_ENTRY_ATTRS' nest
which is strictly validated by the bridge driver, thereby automatically
rejecting the new attributes.
[1] https://datatracker.ietf.org/doc/html/draft-ietf-bess-evpn-irb-mcast#section-3.2.2
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add per-vni statistics for vni filter mode. Counting Rx/Tx
bytes/packets/drops/errors at the appropriate places.
This patch changes vxlan_vs_find_vni to also return the
vxlan_vni_node in cases where the vni belongs to a vni
filtering vxlan device
Signed-off-by: Nikolay Aleksandrov <nikolay@nvidia.com>
Signed-off-by: Roopa Prabhu <roopa@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds vnifiltering support to collect metadata device.
Motivation:
You can only use a single vxlan collect metadata device for a given
vxlan udp port in the system today. The vxlan collect metadata device
terminates all received vxlan packets. As shown in the below diagram,
there are use-cases where you need to support multiple such vxlan devices in
independent bridge domains. Each vxlan device must terminate the vni's
it is configured for.
Example usecase: In a service provider network a service provider
typically supports multiple bridge domains with overlapping vlans.
One bridge domain per customer. Vlans in each bridge domain are
mapped to globally unique vxlan ranges assigned to each customer.
vnifiltering support in collect metadata devices terminates only configured
vnis. This is similar to vlan filtering in bridge driver. The vni filtering
capability is provided by a new flag on collect metadata device.
In the below pic:
- customer1 is mapped to br1 bridge domain
- customer2 is mapped to br2 bridge domain
- customer1 vlan 10-11 is mapped to vni 1001-1002
- customer2 vlan 10-11 is mapped to vni 2001-2002
- br1 and br2 are vlan filtering bridges
- vxlan1 and vxlan2 are collect metadata devices with
vnifiltering enabled
┌──────────────────────────────────────────────────────────────────┐
│ switch │
│ │
│ ┌───────────┐ ┌───────────┐ │
│ │ │ │ │ │
│ │ br1 │ │ br2 │ │
│ └┬─────────┬┘ └──┬───────┬┘ │
│ vlans│ │ vlans │ │ │
│ 10,11│ │ 10,11│ │ │
│ │ vlanvnimap: │ vlanvnimap: │
│ │ 10-1001,11-1002 │ 10-2001,11-2002 │
│ │ │ │ │ │
│ ┌──────┴┐ ┌──┴─────────┐ ┌───┴────┐ │ │
│ │ swp1 │ │vxlan1 │ │ swp2 │ ┌┴─────────────┐ │
│ │ │ │ vnifilter:│ │ │ │vxlan2 │ │
│ └───┬───┘ │ 1001,1002│ └───┬────┘ │ vnifilter: │ │
│ │ └────────────┘ │ │ 2001,2002 │ │
│ │ │ └──────────────┘ │
│ │ │ │
└───────┼──────────────────────────────────┼───────────────────────┘
│ │
│ │
┌─────┴───────┐ │
│ customer1 │ ┌─────┴──────┐
│ host/VM │ │customer2 │
└─────────────┘ │ host/VM │
└────────────┘
With this implementation, vxlan dst metadata device can
be associated with range of vnis.
struct vxlan_vni_node is introduced to represent
a configured vni. We start with vni and its
associated remote_ip in this structure. This
structure can be extended to bring in other
per vni attributes if there are usecases for it.
A vni inherits an attribute from the base vxlan device
if there is no per vni attributes defined.
struct vxlan_dev gets a new rhashtable for
vnis called vxlan_vni_group. vxlan_vnifilter.c
implements the necessary netlink api, notifications
and helper functions to process and manage lifecycle
of vxlan_vni_node.
This patch also adds new helper functions in vxlan_multicast.c
to handle per vni remote_ip multicast groups which are part
of vxlan_vni_group.
Fix build problems:
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Roopa Prabhu <roopa@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As we can see from vxlan_build/parse_gbp_hdr(), when processing metadata
on vxlan rx/tx path, only dont_learn/policy_applied/policy_id fields can
be set to or parse from the packet for vxlan gbp option.
So we'd better do the mask when set it in act_tunnel_key and cls_flower.
Otherwise, when users don't know these bits, they may configure with a
value which can never be matched.
Reported-by: Shuang Li <shuali@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Todays vxlan mac fdb entries can point to multiple remote
ips (rdsts) with the sole purpose of replicating
broadcast-multicast and unknown unicast packets to those remote ips.
E-VPN multihoming [1,2,3] requires bridged vxlan traffic to be
load balanced to remote switches (vteps) belonging to the
same multi-homed ethernet segment (E-VPN multihoming is analogous
to multi-homed LAG implementations, but with the inter-switch
peerlink replaced with a vxlan tunnel). In other words it needs
support for mac ecmp. Furthermore, for faster convergence, E-VPN
multihoming needs the ability to update fdb ecmp nexthops independent
of the fdb entries.
New route nexthop API is perfect for this usecase.
This patch extends the vxlan fdb code to take a nexthop id
pointing to an ecmp nexthop group.
Changes include:
- New NDA_NH_ID attribute for fdbs
- Use the newly added fdb nexthop groups
- makes vxlan rdsts and nexthop handling code mutually
exclusive
- since this is a new use-case and the requirement is for ecmp
nexthop groups, the fdb add and update path checks that the
nexthop is really an ecmp nexthop group. This check can be relaxed
in the future, if we want to introduce replication fdb nexthop groups
and allow its use in lieu of current rdst lists.
- fdb update requests with nexthop id's only allowed for existing
fdb's that have nexthop id's
- learning will not override an existing fdb entry with nexthop
group
- I have wrapped the switchdev offload code around the presence of
rdst
[1] E-VPN RFC https://tools.ietf.org/html/rfc7432
[2] E-VPN with vxlan https://tools.ietf.org/html/rfc8365
[3] http://vger.kernel.org/lpc_net2018_talks/scaling_bridge_fdb_database_slidesV3.pdf
Includes a null check fix in vxlan_xmit from Nikolay
v2 - Fixed build issue:
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
IN_MULTICAST's primary intent is as a uapi macro.
Elsewhere in the kernel we use ipv4_is_multicast consistently.
This patch unifies linux's multicast checks to use that function
rather than this macro.
Signed-off-by: Dave Taht <dave.taht@gmail.com>
Reviewed-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: David S. Miller <davem@davemloft.net>
The monolithic hash_lock could cause huge contention when
inserting/deletiing vxlan_fdbs into the fdb_head.
Use FDB_HASH_SIZE hash_locks to protect insertions/deletions
of vxlan_fdbs into the fdb_head hash table.
Suggested-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Litao jiao <jiaolitao@raisecom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds extack coverage in vxlan link
create and changelink paths. Introduces a new helper
vxlan_nl2flags to consolidate flag attribute validation.
thanks to Johannes Berg for some tips to construct the
generic vxlan flag extack strings.
Signed-off-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are four sources of VXLAN switchdev notifier calls:
- the changelink() link operation, which already supports extack,
- ndo_fdb_add() which got extack support in a previous patch,
- FDB updates due to packet forwarding,
- and vxlan_fdb_replay().
Extend vxlan_fdb_switchdev_call_notifiers() to include extack in the
switchdev message that it sends, and propagate the argument upwards to
the callers. For the first two cases, pass in the extack gotten through
the operation. For case #3, pass in NULL.
To cover the last case, extend vxlan_fdb_replay() to take extack
argument, which might come from whatever operation necessitated the FDB
replay.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a driver unoffloads all FDB entries en bloc, it's inefficient to
send the switchdev notification one by one. Add a helper that walks the
FDB table, unsetting the offload flag on RDST with a given VNI.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a VXLAN device becomes relevant to a driver (such as when it is
attached to an offloaded bridge), the driver will generally need to walk
the existing FDB entries and offload them.
Add a function vxlan_fdb_replay() to call a given notifier block for
each FDB entry with a given VNI.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Petr Machata
Ido Schimmel
Alce Lafranque
Benjamin Poirier
Jiri Benc
Vladimir Nikishkin
Gavin Li
Nikolay Aleksandrov
Roopa Prabhu
Hao Chen
Xin Long
Roopa Prabhu
Taehee Yoo
Dave Taht
Litao jiao
Moshe Shemesh
Petr Machata