Merge branch 'vxlan-convert-fdb-table-to-rhashtable'

Ido Schimmel says:

====================
vxlan: Convert FDB table to rhashtable

The VXLAN driver currently stores FDB entries in a hash table with a
fixed number of buckets (256), resulting in reduced performance as the
number of entries grows. This patchset solves the issue by converting
the driver to use rhashtable which maintains a 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%

The first patches are preparations for the conversion in the last patch.
Specifically, the series is structured as follows:

Patch #1 adds RCU read-side critical sections in the Tx path when
accessing FDB entries. Targeting at net-next as I am not aware of any
issues due to this omission despite the code being structured that way
for a long time. Without it, traces will be generated when converting
FDB lookup to rhashtable_lookup().

Patch #2-#5 simplify the creation of the default FDB entry (all-zeroes).
Current code assumes that insertion into the hash table cannot fail,
which will no longer be true with rhashtable.

Patches #6-#10 add FDB entries to a linked list for entry traversal
instead of traversing over them using the fixed size hash table which is
removed in the last patch.

Patches #11-#12 add wrappers for FDB lookup that make it clear when each
should be used along with lockdep annotations. Needed as a preparation
for rhashtable_lookup() that must be called from an RCU read-side
critical section.

Patch #13 treats dst cache initialization errors as non-fatal. See more
info in the commit message. The current code happens to work because
insertion into the fixed size hash table is slow enough for the per-CPU
allocator to be able to create new chunks of per-CPU memory.

Patch #14 adds an FDB key structure that includes the MAC address and
source VNI. To be used as rhashtable key.

Patch #15 does the conversion to rhashtable.
====================

Link: https://patch.msgid.link/20250415121143.345227-1-idosch@nvidia.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

drivers/net/vxlan/vxlan_private.h

@@ -24,18 +24,23 @@ struct vxlan_net {
 	struct notifier_block nexthop_notifier_block;
 };
 
+struct vxlan_fdb_key {
+	u8 eth_addr[ETH_ALEN];
+	__be32 vni;
+};
+
 /* Forwarding table entry */
 struct vxlan_fdb {
-	struct hlist_node hlist;	/* linked list of entries */
+	struct rhash_head rhnode;
 	struct rcu_head	  rcu;
 	unsigned long	  updated;	/* jiffies */
 	unsigned long	  used;
 	struct list_head  remotes;
-	u8		  eth_addr[ETH_ALEN];
+	struct vxlan_fdb_key key;
 	u16		  state;	/* see ndm_state */
-	__be32		  vni;
 	u16		  flags;	/* see ndm_flags and below */
 	struct list_head  nh_list;
+	struct hlist_node fdb_node;
 	struct nexthop __rcu *nh;
 	struct vxlan_dev  __rcu *vdev;
 };

drivers/net/vxlan/vxlan_vnifilter.c

@@ -482,11 +482,9 @@ static int vxlan_update_default_fdb_entry(struct vxlan_dev *vxlan, __be32 vni,
 					  struct netlink_ext_ack *extack)
 {
 	struct vxlan_rdst *dst = &vxlan->default_dst;
-	u32 hash_index;
 	int err = 0;
 
-	hash_index = fdb_head_index(vxlan, all_zeros_mac, vni);
-	spin_lock_bh(&vxlan->hash_lock[hash_index]);
+	spin_lock_bh(&vxlan->hash_lock);
 	if (remote_ip && !vxlan_addr_any(remote_ip)) {
 		err = vxlan_fdb_update(vxlan, all_zeros_mac,
 				       remote_ip,
@@ -498,7 +496,7 @@ static int vxlan_update_default_fdb_entry(struct vxlan_dev *vxlan, __be32 vni,
 				       dst->remote_ifindex,
 				       NTF_SELF, 0, true, extack);
 		if (err) {
-			spin_unlock_bh(&vxlan->hash_lock[hash_index]);
+			spin_unlock_bh(&vxlan->hash_lock);
 			return err;
 		}
 	}
@@ -511,7 +509,7 @@ static int vxlan_update_default_fdb_entry(struct vxlan_dev *vxlan, __be32 vni,
 				   dst->remote_ifindex,
 				   true);
 	}
-	spin_unlock_bh(&vxlan->hash_lock[hash_index]);
+	spin_unlock_bh(&vxlan->hash_lock);
 
 	return err;
 }

include/net/vxlan.h

@@ -296,7 +296,7 @@ struct vxlan_dev {
 	struct vxlan_rdst default_dst;	/* default destination */
 
 	struct timer_list age_timer;
-	spinlock_t	  hash_lock[FDB_HASH_SIZE];
+	spinlock_t	  hash_lock;
 	unsigned int	  addrcnt;
 	struct gro_cells  gro_cells;
 
@@ -304,9 +304,10 @@ struct vxlan_dev {
 
 	struct vxlan_vni_group  __rcu *vnigrp;
 
-	struct hlist_head fdb_head[FDB_HASH_SIZE];
+	struct rhashtable fdb_hash_tbl;
 
 	struct rhashtable mdb_tbl;
+	struct hlist_head fdb_list;
 	struct hlist_head mdb_list;
 	unsigned int mdb_seq;
 };