Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextgrafted

Pull networking updates from Jakub Kicinski:
 "Core:

   - Add dedicated kmem_cache for typical/small skb->head, avoid having
     to access struct page at kfree time, and improve memory use.

   - Introduce sysctl to set default RPS configuration for new netdevs.

   - Define Netlink protocol specification format which can be used to
     describe messages used by each family and auto-generate parsers.
     Add tools for generating kernel data structures and uAPI headers.

   - Expose all net/core sysctls inside netns.

   - Remove 4s sleep in netpoll if carrier is instantly detected on
     boot.

   - Add configurable limit of MDB entries per port, and port-vlan.

   - Continue populating drop reasons throughout the stack.

   - Retire a handful of legacy Qdiscs and classifiers.

  Protocols:

   - Support IPv4 big TCP (TSO frames larger than 64kB).

   - Add IP_LOCAL_PORT_RANGE socket option, to control local port range
     on socket by socket basis.

   - Track and report in procfs number of MPTCP sockets used.

   - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path
     manager.

   - IPv6: don't check net.ipv6.route.max_size and rely on garbage
     collection to free memory (similarly to IPv4).

   - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986).

   - ICMP: add per-rate limit counters.

   - Add support for user scanning requests in ieee802154.

   - Remove static WEP support.

   - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate
     reporting.

   - WiFi 7 EHT channel puncturing support (client & AP).

  BPF:

   - Add a rbtree data structure following the "next-gen data structure"
     precedent set by recently added linked list, that is, by using
     kfunc + kptr instead of adding a new BPF map type.

   - Expose XDP hints via kfuncs with initial support for RX hash and
     timestamp metadata.

   - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to
     better support decap on GRE tunnel devices not operating in collect
     metadata.

   - Improve x86 JIT's codegen for PROBE_MEM runtime error checks.

   - Remove the need for trace_printk_lock for bpf_trace_printk and
     bpf_trace_vprintk helpers.

   - Extend libbpf's bpf_tracing.h support for tracing arguments of
     kprobes/uprobes and syscall as a special case.

   - Significantly reduce the search time for module symbols by
     livepatch and BPF.

   - Enable cpumasks to be used as kptrs, which is useful for tracing
     programs tracking which tasks end up running on which CPUs in
     different time intervals.

   - Add support for BPF trampoline on s390x and riscv64.

   - Add capability to export the XDP features supported by the NIC.

   - Add __bpf_kfunc tag for marking kernel functions as kfuncs.

   - Add cgroup.memory=nobpf kernel parameter option to disable BPF
     memory accounting for container environments.

  Netfilter:

   - Remove the CLUSTERIP target. It has been marked as obsolete for
     years, and we still have WARN splats wrt races of the out-of-band
     /proc interface installed by this target.

   - Add 'destroy' commands to nf_tables. They are identical to the
     existing 'delete' commands, but do not return an error if the
     referenced object (set, chain, rule...) did not exist.

  Driver API:

   - Improve cpumask_local_spread() locality to help NICs set the right
     IRQ affinity on AMD platforms.

   - Separate C22 and C45 MDIO bus transactions more clearly.

   - Introduce new DCB table to control DSCP rewrite on egress.

   - Support configuration of Physical Layer Collision Avoidance (PLCA)
     Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of
     shared medium Ethernet.

   - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing
     preemption of low priority frames by high priority frames.

   - Add support for controlling MACSec offload using netlink SET.

   - Rework devlink instance refcounts to allow registration and
     de-registration under the instance lock. Split the code into
     multiple files, drop some of the unnecessarily granular locks and
     factor out common parts of netlink operation handling.

   - Add TX frame aggregation parameters (for USB drivers).

   - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning
     messages with notifications for debug.

   - Allow offloading of UDP NEW connections via act_ct.

   - Add support for per action HW stats in TC.

   - Support hardware miss to TC action (continue processing in SW from
     a specific point in the action chain).

   - Warn if old Wireless Extension user space interface is used with
     modern cfg80211/mac80211 drivers. Do not support Wireless
     Extensions for Wi-Fi 7 devices at all. Everyone should switch to
     using nl80211 interface instead.

   - Improve the CAN bit timing configuration. Use extack to return
     error messages directly to user space, update the SJW handling,
     including the definition of a new default value that will benefit
     CAN-FD controllers, by increasing their oscillator tolerance.

  New hardware / drivers:

   - Ethernet:
      - nVidia BlueField-3 support (control traffic driver)
      - Ethernet support for imx93 SoCs
      - Motorcomm yt8531 gigabit Ethernet PHY
      - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA)
      - Microchip LAN8841 PHY (incl. cable diagnostics and PTP)
      - Amlogic gxl MDIO mux

   - WiFi:
      - RealTek RTL8188EU (rtl8xxxu)
      - Qualcomm Wi-Fi 7 devices (ath12k)

   - CAN:
      - Renesas R-Car V4H

  Drivers:

   - Bluetooth:
      - Set Per Platform Antenna Gain (PPAG) for Intel controllers.

   - Ethernet NICs:
      - Intel (1G, igc):
         - support TSN / Qbv / packet scheduling features of i226 model
      - Intel (100G, ice):
         - use GNSS subsystem instead of TTY
         - multi-buffer XDP support
         - extend support for GPIO pins to E823 devices
      - nVidia/Mellanox:
         - update the shared buffer configuration on PFC commands
         - implement PTP adjphase function for HW offset control
         - TC support for Geneve and GRE with VF tunnel offload
         - more efficient crypto key management method
         - multi-port eswitch support
      - Netronome/Corigine:
         - add DCB IEEE support
         - support IPsec offloading for NFP3800
      - Freescale/NXP (enetc):
         - support XDP_REDIRECT for XDP non-linear buffers
         - improve reconfig, avoid link flap and waiting for idle
         - support MAC Merge layer
      - Other NICs:
         - sfc/ef100: add basic devlink support for ef100
         - ionic: rx_push mode operation (writing descriptors via MMIO)
         - bnxt: use the auxiliary bus abstraction for RDMA
         - r8169: disable ASPM and reset bus in case of tx timeout
         - cpsw: support QSGMII mode for J721e CPSW9G
         - cpts: support pulse-per-second output
         - ngbe: add an mdio bus driver
         - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing
         - r8152: handle devices with FW with NCM support
         - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation
         - virtio-net: support multi buffer XDP
         - virtio/vsock: replace virtio_vsock_pkt with sk_buff
         - tsnep: XDP support

   - Ethernet high-speed switches:
      - nVidia/Mellanox (mlxsw):
         - add support for latency TLV (in FW control messages)
      - Microchip (sparx5):
         - separate explicit and implicit traffic forwarding rules, make
           the implicit rules always active
         - add support for egress DSCP rewrite
         - IS0 VCAP support (Ingress Classification)
         - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS
           etc.)
         - ES2 VCAP support (Egress Access Control)
         - support for Per-Stream Filtering and Policing (802.1Q,
           8.6.5.1)

   - Ethernet embedded switches:
      - Marvell (mv88e6xxx):
         - add MAB (port auth) offload support
         - enable PTP receive for mv88e6390
      - NXP (ocelot):
         - support MAC Merge layer
         - support for the the vsc7512 internal copper phys
      - Microchip:
         - lan9303: convert to PHYLINK
         - lan966x: support TC flower filter statistics
         - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x
         - lan937x: support Credit Based Shaper configuration
         - ksz9477: support Energy Efficient Ethernet
      - other:
         - qca8k: convert to regmap read/write API, use bulk operations
         - rswitch: Improve TX timestamp accuracy

   - Intel WiFi (iwlwifi):
      - EHT (Wi-Fi 7) rate reporting
      - STEP equalizer support: transfer some STEP (connection to radio
        on platforms with integrated wifi) related parameters from the
        BIOS to the firmware.

   - Qualcomm 802.11ax WiFi (ath11k):
      - IPQ5018 support
      - Fine Timing Measurement (FTM) responder role support
      - channel 177 support

   - MediaTek WiFi (mt76):
      - per-PHY LED support
      - mt7996: EHT (Wi-Fi 7) support
      - Wireless Ethernet Dispatch (WED) reset support
      - switch to using page pool allocator

   - RealTek WiFi (rtw89):
      - support new version of Bluetooth co-existance

   - Mobile:
      - rmnet: support TX aggregation"

* tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits)
  page_pool: add a comment explaining the fragment counter usage
  net: ethtool: fix __ethtool_dev_mm_supported() implementation
  ethtool: pse-pd: Fix double word in comments
  xsk: add linux/vmalloc.h to xsk.c
  sefltests: netdevsim: wait for devlink instance after netns removal
  selftest: fib_tests: Always cleanup before exit
  net/mlx5e: Align IPsec ASO result memory to be as required by hardware
  net/mlx5e: TC, Set CT miss to the specific ct action instance
  net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG
  net/mlx5: Refactor tc miss handling to a single function
  net/mlx5: Kconfig: Make tc offload depend on tc skb extension
  net/sched: flower: Support hardware miss to tc action
  net/sched: flower: Move filter handle initialization earlier
  net/sched: cls_api: Support hardware miss to tc action
  net/sched: Rename user cookie and act cookie
  sfc: fix builds without CONFIG_RTC_LIB
  sfc: clean up some inconsistent indentings
  net/mlx4_en: Introduce flexible array to silence overflow warning
  net: lan966x: Fix possible deadlock inside PTP
  net/ulp: Remove redundant ->clone() test in inet_clone_ulp().
  ...

1 files changed, 1049 insertions, 0 deletions

diff --git a/net/appletalk/aarp.c b/net/appletalk/aarp.c
new file mode 100644
index 000000000..c7236daa2
--- /dev/null
+++ b/net/appletalk/aarp.c
@@ -0,0 +1,1049 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ *	AARP:		An implementation of the AppleTalk AARP protocol for
+ *			Ethernet 'ELAP'.
+ *
+ *		Alan Cox  <Alan.Cox@linux.org>
+ *
+ *	This doesn't fit cleanly with the IP arp. Potentially we can use
+ *	the generic neighbour discovery code to clean this up.
+ *
+ *	FIXME:
+ *		We ought to handle the retransmits with a single list and a
+ *	separate fast timer for when it is needed.
+ *		Use neighbour discovery code.
+ *		Token Ring Support.
+ *
+ *	References:
+ *		Inside AppleTalk (2nd Ed).
+ *	Fixes:
+ *		Jaume Grau	-	flush caches on AARP_PROBE
+ *		Rob Newberry	-	Added proxy AARP and AARP proc fs,
+ *					moved probing from DDP module.
+ *		Arnaldo C. Melo -	don't mangle rx packets
+ */
+
+#include <linux/if_arp.h>
+#include <linux/slab.h>
+#include <net/sock.h>
+#include <net/datalink.h>
+#include <net/psnap.h>
+#include <linux/atalk.h>
+#include <linux/delay.h>
+#include <linux/init.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/export.h>
+#include <linux/etherdevice.h>
+
+int sysctl_aarp_expiry_time = AARP_EXPIRY_TIME;
+int sysctl_aarp_tick_time = AARP_TICK_TIME;
+int sysctl_aarp_retransmit_limit = AARP_RETRANSMIT_LIMIT;
+int sysctl_aarp_resolve_time = AARP_RESOLVE_TIME;
+
+/* Lists of aarp entries */
+/**
+ *	struct aarp_entry - AARP entry
+ *	@last_sent: Last time we xmitted the aarp request
+ *	@packet_queue: Queue of frames wait for resolution
+ *	@status: Used for proxy AARP
+ *	@expires_at: Entry expiry time
+ *	@target_addr: DDP Address
+ *	@dev:  Device to use
+ *	@hwaddr:  Physical i/f address of target/router
+ *	@xmit_count:  When this hits 10 we give up
+ *	@next: Next entry in chain
+ */
+struct aarp_entry {
+	/* These first two are only used for unresolved entries */
+	unsigned long		last_sent;
+	struct sk_buff_head	packet_queue;
+	int			status;
+	unsigned long		expires_at;
+	struct atalk_addr	target_addr;
+	struct net_device	*dev;
+	char			hwaddr[ETH_ALEN];
+	unsigned short		xmit_count;
+	struct aarp_entry	*next;
+};
+
+/* Hashed list of resolved, unresolved and proxy entries */
+static struct aarp_entry *resolved[AARP_HASH_SIZE];
+static struct aarp_entry *unresolved[AARP_HASH_SIZE];
+static struct aarp_entry *proxies[AARP_HASH_SIZE];
+static int unresolved_count;
+
+/* One lock protects it all. */
+static DEFINE_RWLOCK(aarp_lock);
+
+/* Used to walk the list and purge/kick entries.  */
+static struct timer_list aarp_timer;
+
+/*
+ *	Delete an aarp queue
+ *
+ *	Must run under aarp_lock.
+ */
+static void __aarp_expire(struct aarp_entry *a)
+{
+	skb_queue_purge(&a->packet_queue);
+	kfree(a);
+}
+
+/*
+ *	Send an aarp queue entry request
+ *
+ *	Must run under aarp_lock.
+ */
+static void __aarp_send_query(struct aarp_entry *a)
+{
+	static unsigned char aarp_eth_multicast[ETH_ALEN] =
+					{ 0x09, 0x00, 0x07, 0xFF, 0xFF, 0xFF };
+	struct net_device *dev = a->dev;
+	struct elapaarp *eah;
+	int len = dev->hard_header_len + sizeof(*eah) + aarp_dl->header_length;
+	struct sk_buff *skb = alloc_skb(len, GFP_ATOMIC);
+	struct atalk_addr *sat = atalk_find_dev_addr(dev);
+
+	if (!skb)
+		return;
+
+	if (!sat) {
+		kfree_skb(skb);
+		return;
+	}
+
+	/* Set up the buffer */
+	skb_reserve(skb, dev->hard_header_len + aarp_dl->header_length);
+	skb_reset_network_header(skb);
+	skb_reset_transport_header(skb);
+	skb_put(skb, sizeof(*eah));
+	skb->protocol    = htons(ETH_P_ATALK);
+	skb->dev	 = dev;
+	eah		 = aarp_hdr(skb);
+
+	/* Set up the ARP */
+	eah->hw_type	 = htons(AARP_HW_TYPE_ETHERNET);
+	eah->pa_type	 = htons(ETH_P_ATALK);
+	eah->hw_len	 = ETH_ALEN;
+	eah->pa_len	 = AARP_PA_ALEN;
+	eah->function	 = htons(AARP_REQUEST);
+
+	ether_addr_copy(eah->hw_src, dev->dev_addr);
+
+	eah->pa_src_zero = 0;
+	eah->pa_src_net	 = sat->s_net;
+	eah->pa_src_node = sat->s_node;
+
+	eth_zero_addr(eah->hw_dst);
+
+	eah->pa_dst_zero = 0;
+	eah->pa_dst_net	 = a->target_addr.s_net;
+	eah->pa_dst_node = a->target_addr.s_node;
+
+	/* Send it */
+	aarp_dl->request(aarp_dl, skb, aarp_eth_multicast);
+	/* Update the sending count */
+	a->xmit_count++;
+	a->last_sent = jiffies;
+}
+
+/* This runs under aarp_lock and in softint context, so only atomic memory
+ * allocations can be used. */
+static void aarp_send_reply(struct net_device *dev, struct atalk_addr *us,
+			    struct atalk_addr *them, unsigned char *sha)
+{
+	struct elapaarp *eah;
+	int len = dev->hard_header_len + sizeof(*eah) + aarp_dl->header_length;
+	struct sk_buff *skb = alloc_skb(len, GFP_ATOMIC);
+
+	if (!skb)
+		return;
+
+	/* Set up the buffer */
+	skb_reserve(skb, dev->hard_header_len + aarp_dl->header_length);
+	skb_reset_network_header(skb);
+	skb_reset_transport_header(skb);
+	skb_put(skb, sizeof(*eah));
+	skb->protocol    = htons(ETH_P_ATALK);
+	skb->dev	 = dev;
+	eah		 = aarp_hdr(skb);
+
+	/* Set up the ARP */
+	eah->hw_type	 = htons(AARP_HW_TYPE_ETHERNET);
+	eah->pa_type	 = htons(ETH_P_ATALK);
+	eah->hw_len	 = ETH_ALEN;
+	eah->pa_len	 = AARP_PA_ALEN;
+	eah->function	 = htons(AARP_REPLY);
+
+	ether_addr_copy(eah->hw_src, dev->dev_addr);
+
+	eah->pa_src_zero = 0;
+	eah->pa_src_net	 = us->s_net;
+	eah->pa_src_node = us->s_node;
+
+	if (!sha)
+		eth_zero_addr(eah->hw_dst);
+	else
+		ether_addr_copy(eah->hw_dst, sha);
+
+	eah->pa_dst_zero = 0;
+	eah->pa_dst_net	 = them->s_net;
+	eah->pa_dst_node = them->s_node;
+
+	/* Send it */
+	aarp_dl->request(aarp_dl, skb, sha);
+}
+
+/*
+ *	Send probe frames. Called from aarp_probe_network and
+ *	aarp_proxy_probe_network.
+ */
+
+static void aarp_send_probe(struct net_device *dev, struct atalk_addr *us)
+{
+	struct elapaarp *eah;
+	int len = dev->hard_header_len + sizeof(*eah) + aarp_dl->header_length;
+	struct sk_buff *skb = alloc_skb(len, GFP_ATOMIC);
+	static unsigned char aarp_eth_multicast[ETH_ALEN] =
+					{ 0x09, 0x00, 0x07, 0xFF, 0xFF, 0xFF };
+
+	if (!skb)
+		return;
+
+	/* Set up the buffer */
+	skb_reserve(skb, dev->hard_header_len + aarp_dl->header_length);
+	skb_reset_network_header(skb);
+	skb_reset_transport_header(skb);
+	skb_put(skb, sizeof(*eah));
+	skb->protocol    = htons(ETH_P_ATALK);
+	skb->dev	 = dev;
+	eah		 = aarp_hdr(skb);
+
+	/* Set up the ARP */
+	eah->hw_type	 = htons(AARP_HW_TYPE_ETHERNET);
+	eah->pa_type	 = htons(ETH_P_ATALK);
+	eah->hw_len	 = ETH_ALEN;
+	eah->pa_len	 = AARP_PA_ALEN;
+	eah->function	 = htons(AARP_PROBE);
+
+	ether_addr_copy(eah->hw_src, dev->dev_addr);
+
+	eah->pa_src_zero = 0;
+	eah->pa_src_net	 = us->s_net;
+	eah->pa_src_node = us->s_node;
+
+	eth_zero_addr(eah->hw_dst);
+
+	eah->pa_dst_zero = 0;
+	eah->pa_dst_net	 = us->s_net;
+	eah->pa_dst_node = us->s_node;
+
+	/* Send it */
+	aarp_dl->request(aarp_dl, skb, aarp_eth_multicast);
+}
+
+/*
+ *	Handle an aarp timer expire
+ *
+ *	Must run under the aarp_lock.
+ */
+
+static void __aarp_expire_timer(struct aarp_entry **n)
+{
+	struct aarp_entry *t;
+
+	while (*n)
+		/* Expired ? */
+		if (time_after(jiffies, (*n)->expires_at)) {
+			t = *n;
+			*n = (*n)->next;
+			__aarp_expire(t);
+		} else
+			n = &((*n)->next);
+}
+
+/*
+ *	Kick all pending requests 5 times a second.
+ *
+ *	Must run under the aarp_lock.
+ */
+static void __aarp_kick(struct aarp_entry **n)
+{
+	struct aarp_entry *t;
+
+	while (*n)
+		/* Expired: if this will be the 11th tx, we delete instead. */
+		if ((*n)->xmit_count >= sysctl_aarp_retransmit_limit) {
+			t = *n;
+			*n = (*n)->next;
+			__aarp_expire(t);
+		} else {
+			__aarp_send_query(*n);
+			n = &((*n)->next);
+		}
+}
+
+/*
+ *	A device has gone down. Take all entries referring to the device
+ *	and remove them.
+ *
+ *	Must run under the aarp_lock.
+ */
+static void __aarp_expire_device(struct aarp_entry **n, struct net_device *dev)
+{
+	struct aarp_entry *t;
+
+	while (*n)
+		if ((*n)->dev == dev) {
+			t = *n;
+			*n = (*n)->next;
+			__aarp_expire(t);
+		} else
+			n = &((*n)->next);
+}
+
+/* Handle the timer event */
+static void aarp_expire_timeout(struct timer_list *unused)
+{
+	int ct;
+
+	write_lock_bh(&aarp_lock);
+
+	for (ct = 0; ct < AARP_HASH_SIZE; ct++) {
+		__aarp_expire_timer(&resolved[ct]);
+		__aarp_kick(&unresolved[ct]);
+		__aarp_expire_timer(&unresolved[ct]);
+		__aarp_expire_timer(&proxies[ct]);
+	}
+
+	write_unlock_bh(&aarp_lock);
+	mod_timer(&aarp_timer, jiffies +
+			       (unresolved_count ? sysctl_aarp_tick_time :
+				sysctl_aarp_expiry_time));
+}
+
+/* Network device notifier chain handler. */
+static int aarp_device_event(struct notifier_block *this, unsigned long event,
+			     void *ptr)
+{
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+	int ct;
+
+	if (!net_eq(dev_net(dev), &init_net))
+		return NOTIFY_DONE;
+
+	if (event == NETDEV_DOWN) {
+		write_lock_bh(&aarp_lock);
+
+		for (ct = 0; ct < AARP_HASH_SIZE; ct++) {
+			__aarp_expire_device(&resolved[ct], dev);
+			__aarp_expire_device(&unresolved[ct], dev);
+			__aarp_expire_device(&proxies[ct], dev);
+		}
+
+		write_unlock_bh(&aarp_lock);
+	}
+	return NOTIFY_DONE;
+}
+
+/* Expire all entries in a hash chain */
+static void __aarp_expire_all(struct aarp_entry **n)
+{
+	struct aarp_entry *t;
+
+	while (*n) {
+		t = *n;
+		*n = (*n)->next;
+		__aarp_expire(t);
+	}
+}
+
+/* Cleanup all hash chains -- module unloading */
+static void aarp_purge(void)
+{
+	int ct;
+
+	write_lock_bh(&aarp_lock);
+	for (ct = 0; ct < AARP_HASH_SIZE; ct++) {
+		__aarp_expire_all(&resolved[ct]);
+		__aarp_expire_all(&unresolved[ct]);
+		__aarp_expire_all(&proxies[ct]);
+	}
+	write_unlock_bh(&aarp_lock);
+}
+
+/*
+ *	Create a new aarp entry.  This must use GFP_ATOMIC because it
+ *	runs while holding spinlocks.
+ */
+static struct aarp_entry *aarp_alloc(void)
+{
+	struct aarp_entry *a = kmalloc(sizeof(*a), GFP_ATOMIC);
+
+	if (a)
+		skb_queue_head_init(&a->packet_queue);
+	return a;
+}
+
+/*
+ * Find an entry. We might return an expired but not yet purged entry. We
+ * don't care as it will do no harm.
+ *
+ * This must run under the aarp_lock.
+ */
+static struct aarp_entry *__aarp_find_entry(struct aarp_entry *list,
+					    struct net_device *dev,
+					    struct atalk_addr *sat)
+{
+	while (list) {
+		if (list->target_addr.s_net == sat->s_net &&
+		    list->target_addr.s_node == sat->s_node &&
+		    list->dev == dev)
+			break;
+		list = list->next;
+	}
+
+	return list;
+}
+
+/* Called from the DDP code, and thus must be exported. */
+void aarp_proxy_remove(struct net_device *dev, struct atalk_addr *sa)
+{
+	int hash = sa->s_node % (AARP_HASH_SIZE - 1);
+	struct aarp_entry *a;
+
+	write_lock_bh(&aarp_lock);
+
+	a = __aarp_find_entry(proxies[hash], dev, sa);
+	if (a)
+		a->expires_at = jiffies - 1;
+
+	write_unlock_bh(&aarp_lock);
+}
+
+/* This must run under aarp_lock. */
+static struct atalk_addr *__aarp_proxy_find(struct net_device *dev,
+					    struct atalk_addr *sa)
+{
+	int hash = sa->s_node % (AARP_HASH_SIZE - 1);
+	struct aarp_entry *a = __aarp_find_entry(proxies[hash], dev, sa);
+
+	return a ? sa : NULL;
+}
+
+/*
+ * Probe a Phase 1 device or a device that requires its Net:Node to
+ * be set via an ioctl.
+ */
+static void aarp_send_probe_phase1(struct atalk_iface *iface)
+{
+	struct ifreq atreq;
+	struct sockaddr_at *sa = (struct sockaddr_at *)&atreq.ifr_addr;
+	const struct net_device_ops *ops = iface->dev->netdev_ops;
+
+	sa->sat_addr.s_node = iface->address.s_node;
+	sa->sat_addr.s_net = ntohs(iface->address.s_net);
+
+	/* We pass the Net:Node to the drivers/cards by a Device ioctl. */
+	if (!(ops->ndo_do_ioctl(iface->dev, &atreq, SIOCSIFADDR))) {
+		ops->ndo_do_ioctl(iface->dev, &atreq, SIOCGIFADDR);
+		if (iface->address.s_net != htons(sa->sat_addr.s_net) ||
+		    iface->address.s_node != sa->sat_addr.s_node)
+			iface->status |= ATIF_PROBE_FAIL;
+
+		iface->address.s_net  = htons(sa->sat_addr.s_net);
+		iface->address.s_node = sa->sat_addr.s_node;
+	}
+}
+
+
+void aarp_probe_network(struct atalk_iface *atif)
+{
+	if (atif->dev->type == ARPHRD_LOCALTLK ||
+	    atif->dev->type == ARPHRD_PPP)
+		aarp_send_probe_phase1(atif);
+	else {
+		unsigned int count;
+
+		for (count = 0; count < AARP_RETRANSMIT_LIMIT; count++) {
+			aarp_send_probe(atif->dev, &atif->address);
+
+			/* Defer 1/10th */
+			msleep(100);
+
+			if (atif->status & ATIF_PROBE_FAIL)
+				break;
+		}
+	}
+}
+
+int aarp_proxy_probe_network(struct atalk_iface *atif, struct atalk_addr *sa)
+{
+	int hash, retval = -EPROTONOSUPPORT;
+	struct aarp_entry *entry;
+	unsigned int count;
+
+	/*
+	 * we don't currently support LocalTalk or PPP for proxy AARP;
+	 * if someone wants to try and add it, have fun
+	 */
+	if (atif->dev->type == ARPHRD_LOCALTLK ||
+	    atif->dev->type == ARPHRD_PPP)
+		goto out;
+
+	/*
+	 * create a new AARP entry with the flags set to be published --
+	 * we need this one to hang around even if it's in use
+	 */
+	entry = aarp_alloc();
+	retval = -ENOMEM;
+	if (!entry)
+		goto out;
+
+	entry->expires_at = -1;
+	entry->status = ATIF_PROBE;
+	entry->target_addr.s_node = sa->s_node;
+	entry->target_addr.s_net = sa->s_net;
+	entry->dev = atif->dev;
+
+	write_lock_bh(&aarp_lock);
+
+	hash = sa->s_node % (AARP_HASH_SIZE - 1);
+	entry->next = proxies[hash];
+	proxies[hash] = entry;
+
+	for (count = 0; count < AARP_RETRANSMIT_LIMIT; count++) {
+		aarp_send_probe(atif->dev, sa);
+
+		/* Defer 1/10th */
+		write_unlock_bh(&aarp_lock);
+		msleep(100);
+		write_lock_bh(&aarp_lock);
+
+		if (entry->status & ATIF_PROBE_FAIL)
+			break;
+	}
+
+	if (entry->status & ATIF_PROBE_FAIL) {
+		entry->expires_at = jiffies - 1; /* free the entry */
+		retval = -EADDRINUSE; /* return network full */
+	} else { /* clear the probing flag */
+		entry->status &= ~ATIF_PROBE;
+		retval = 1;
+	}
+
+	write_unlock_bh(&aarp_lock);
+out:
+	return retval;
+}
+
+/* Send a DDP frame */
+int aarp_send_ddp(struct net_device *dev, struct sk_buff *skb,
+		  struct atalk_addr *sa, void *hwaddr)
+{
+	static char ddp_eth_multicast[ETH_ALEN] =
+		{ 0x09, 0x00, 0x07, 0xFF, 0xFF, 0xFF };
+	int hash;
+	struct aarp_entry *a;
+
+	skb_reset_network_header(skb);
+
+	/* Check for LocalTalk first */
+	if (dev->type == ARPHRD_LOCALTLK) {
+		struct atalk_addr *at = atalk_find_dev_addr(dev);
+		struct ddpehdr *ddp = (struct ddpehdr *)skb->data;
+		int ft = 2;
+
+		/*
+		 * Compressible ?
+		 *
+		 * IFF: src_net == dest_net == device_net
+		 * (zero matches anything)
+		 */
+
+		if ((!ddp->deh_snet || at->s_net == ddp->deh_snet) &&
+		    (!ddp->deh_dnet || at->s_net == ddp->deh_dnet)) {
+			skb_pull(skb, sizeof(*ddp) - 4);
+
+			/*
+			 *	The upper two remaining bytes are the port
+			 *	numbers	we just happen to need. Now put the
+			 *	length in the lower two.
+			 */
+			*((__be16 *)skb->data) = htons(skb->len);
+			ft = 1;
+		}
+		/*
+		 * Nice and easy. No AARP type protocols occur here so we can
+		 * just shovel it out with a 3 byte LLAP header
+		 */
+
+		skb_push(skb, 3);
+		skb->data[0] = sa->s_node;
+		skb->data[1] = at->s_node;
+		skb->data[2] = ft;
+		skb->dev     = dev;
+		goto sendit;
+	}
+
+	/* On a PPP link we neither compress nor aarp.  */
+	if (dev->type == ARPHRD_PPP) {
+		skb->protocol = htons(ETH_P_PPPTALK);
+		skb->dev = dev;
+		goto sendit;
+	}
+
+	/* Non ELAP we cannot do. */
+	if (dev->type != ARPHRD_ETHER)
+		goto free_it;
+
+	skb->dev = dev;
+	skb->protocol = htons(ETH_P_ATALK);
+	hash = sa->s_node % (AARP_HASH_SIZE - 1);
+
+	/* Do we have a resolved entry? */
+	if (sa->s_node == ATADDR_BCAST) {
+		/* Send it */
+		ddp_dl->request(ddp_dl, skb, ddp_eth_multicast);
+		goto sent;
+	}
+
+	write_lock_bh(&aarp_lock);
+	a = __aarp_find_entry(resolved[hash], dev, sa);
+
+	if (a) { /* Return 1 and fill in the address */
+		a->expires_at = jiffies + (sysctl_aarp_expiry_time * 10);
+		ddp_dl->request(ddp_dl, skb, a->hwaddr);
+		write_unlock_bh(&aarp_lock);
+		goto sent;
+	}
+
+	/* Do we have an unresolved entry: This is the less common path */
+	a = __aarp_find_entry(unresolved[hash], dev, sa);
+	if (a) { /* Queue onto the unresolved queue */
+		skb_queue_tail(&a->packet_queue, skb);
+		goto out_unlock;
+	}
+
+	/* Allocate a new entry */
+	a = aarp_alloc();
+	if (!a) {
+		/* Whoops slipped... good job it's an unreliable protocol 8) */
+		write_unlock_bh(&aarp_lock);
+		goto free_it;
+	}
+
+	/* Set up the queue */
+	skb_queue_tail(&a->packet_queue, skb);
+	a->expires_at	 = jiffies + sysctl_aarp_resolve_time;
+	a->dev		 = dev;
+	a->next		 = unresolved[hash];
+	a->target_addr	 = *sa;
+	a->xmit_count	 = 0;
+	unresolved[hash] = a;
+	unresolved_count++;
+
+	/* Send an initial request for the address */
+	__aarp_send_query(a);
+
+	/*
+	 * Switch to fast timer if needed (That is if this is the first
+	 * unresolved entry to get added)
+	 */
+
+	if (unresolved_count == 1)
+		mod_timer(&aarp_timer, jiffies + sysctl_aarp_tick_time);
+
+	/* Now finally, it is safe to drop the lock. */
+out_unlock:
+	write_unlock_bh(&aarp_lock);
+
+	/* Tell the ddp layer we have taken over for this frame. */
+	goto sent;
+
+sendit:
+	if (skb->sk)
+		skb->priority = skb->sk->sk_priority;
+	if (dev_queue_xmit(skb))
+		goto drop;
+sent:
+	return NET_XMIT_SUCCESS;
+free_it:
+	kfree_skb(skb);
+drop:
+	return NET_XMIT_DROP;
+}
+EXPORT_SYMBOL(aarp_send_ddp);
+
+/*
+ *	An entry in the aarp unresolved queue has become resolved. Send
+ *	all the frames queued under it.
+ *
+ *	Must run under aarp_lock.
+ */
+static void __aarp_resolved(struct aarp_entry **list, struct aarp_entry *a,
+			    int hash)
+{
+	struct sk_buff *skb;
+
+	while (*list)
+		if (*list == a) {
+			unresolved_count--;
+			*list = a->next;
+
+			/* Move into the resolved list */
+			a->next = resolved[hash];
+			resolved[hash] = a;
+
+			/* Kick frames off */
+			while ((skb = skb_dequeue(&a->packet_queue)) != NULL) {
+				a->expires_at = jiffies +
+						sysctl_aarp_expiry_time * 10;
+				ddp_dl->request(ddp_dl, skb, a->hwaddr);
+			}
+		} else
+			list = &((*list)->next);
+}
+
+/*
+ *	This is called by the SNAP driver whenever we see an AARP SNAP
+ *	frame. We currently only support Ethernet.
+ */
+static int aarp_rcv(struct sk_buff *skb, struct net_device *dev,
+		    struct packet_type *pt, struct net_device *orig_dev)
+{
+	struct elapaarp *ea = aarp_hdr(skb);
+	int hash, ret = 0;
+	__u16 function;
+	struct aarp_entry *a;
+	struct atalk_addr sa, *ma, da;
+	struct atalk_iface *ifa;
+
+	if (!net_eq(dev_net(dev), &init_net))
+		goto out0;
+
+	/* We only do Ethernet SNAP AARP. */
+	if (dev->type != ARPHRD_ETHER)
+		goto out0;
+
+	/* Frame size ok? */
+	if (!skb_pull(skb, sizeof(*ea)))
+		goto out0;
+
+	function = ntohs(ea->function);
+
+	/* Sanity check fields. */
+	if (function < AARP_REQUEST || function > AARP_PROBE ||
+	    ea->hw_len != ETH_ALEN || ea->pa_len != AARP_PA_ALEN ||
+	    ea->pa_src_zero || ea->pa_dst_zero)
+		goto out0;
+
+	/* Looks good. */
+	hash = ea->pa_src_node % (AARP_HASH_SIZE - 1);
+
+	/* Build an address. */
+	sa.s_node = ea->pa_src_node;
+	sa.s_net = ea->pa_src_net;
+
+	/* Process the packet. Check for replies of me. */
+	ifa = atalk_find_dev(dev);
+	if (!ifa)
+		goto out1;
+
+	if (ifa->status & ATIF_PROBE &&
+	    ifa->address.s_node == ea->pa_dst_node &&
+	    ifa->address.s_net == ea->pa_dst_net) {
+		ifa->status |= ATIF_PROBE_FAIL; /* Fail the probe (in use) */
+		goto out1;
+	}
+
+	/* Check for replies of proxy AARP entries */
+	da.s_node = ea->pa_dst_node;
+	da.s_net  = ea->pa_dst_net;
+
+	write_lock_bh(&aarp_lock);
+	a = __aarp_find_entry(proxies[hash], dev, &da);
+
+	if (a && a->status & ATIF_PROBE) {
+		a->status |= ATIF_PROBE_FAIL;
+		/*
+		 * we do not respond to probe or request packets of
+		 * this address while we are probing this address
+		 */
+		goto unlock;
+	}
+
+	switch (function) {
+	case AARP_REPLY:
+		if (!unresolved_count)	/* Speed up */
+			break;
+
+		/* Find the entry.  */
+		a = __aarp_find_entry(unresolved[hash], dev, &sa);
+		if (!a || dev != a->dev)
+			break;
+
+		/* We can fill one in - this is good. */
+		ether_addr_copy(a->hwaddr, ea->hw_src);
+		__aarp_resolved(&unresolved[hash], a, hash);
+		if (!unresolved_count)
+			mod_timer(&aarp_timer,
+				  jiffies + sysctl_aarp_expiry_time);
+		break;
+
+	case AARP_REQUEST:
+	case AARP_PROBE:
+
+		/*
+		 * If it is my address set ma to my address and reply.
+		 * We can treat probe and request the same.  Probe
+		 * simply means we shouldn't cache the querying host,
+		 * as in a probe they are proposing an address not
+		 * using one.
+		 *
+		 * Support for proxy-AARP added. We check if the
+		 * address is one of our proxies before we toss the
+		 * packet out.
+		 */
+
+		sa.s_node = ea->pa_dst_node;
+		sa.s_net  = ea->pa_dst_net;
+
+		/* See if we have a matching proxy. */
+		ma = __aarp_proxy_find(dev, &sa);
+		if (!ma)
+			ma = &ifa->address;
+		else { /* We need to make a copy of the entry. */
+			da.s_node = sa.s_node;
+			da.s_net = sa.s_net;
+			ma = &da;
+		}
+
+		if (function == AARP_PROBE) {
+			/*
+			 * A probe implies someone trying to get an
+			 * address. So as a precaution flush any
+			 * entries we have for this address.
+			 */
+			a = __aarp_find_entry(resolved[sa.s_node %
+						       (AARP_HASH_SIZE - 1)],
+					      skb->dev, &sa);
+
+			/*
+			 * Make it expire next tick - that avoids us
+			 * getting into a probe/flush/learn/probe/
+			 * flush/learn cycle during probing of a slow
+			 * to respond host addr.
+			 */
+			if (a) {
+				a->expires_at = jiffies - 1;
+				mod_timer(&aarp_timer, jiffies +
+					  sysctl_aarp_tick_time);
+			}
+		}
+
+		if (sa.s_node != ma->s_node)
+			break;
+
+		if (sa.s_net && ma->s_net && sa.s_net != ma->s_net)
+			break;
+
+		sa.s_node = ea->pa_src_node;
+		sa.s_net = ea->pa_src_net;
+
+		/* aarp_my_address has found the address to use for us.
+		 */
+		aarp_send_reply(dev, ma, &sa, ea->hw_src);
+		break;
+	}
+
+unlock:
+	write_unlock_bh(&aarp_lock);
+out1:
+	ret = 1;
+out0:
+	kfree_skb(skb);
+	return ret;
+}
+
+static struct notifier_block aarp_notifier = {
+	.notifier_call = aarp_device_event,
+};
+
+static unsigned char aarp_snap_id[] = { 0x00, 0x00, 0x00, 0x80, 0xF3 };
+
+int __init aarp_proto_init(void)
+{
+	int rc;
+
+	aarp_dl = register_snap_client(aarp_snap_id, aarp_rcv);
+	if (!aarp_dl) {
+		printk(KERN_CRIT "Unable to register AARP with SNAP.\n");
+		return -ENOMEM;
+	}
+	timer_setup(&aarp_timer, aarp_expire_timeout, 0);
+	aarp_timer.expires  = jiffies + sysctl_aarp_expiry_time;
+	add_timer(&aarp_timer);
+	rc = register_netdevice_notifier(&aarp_notifier);
+	if (rc) {
+		del_timer_sync(&aarp_timer);
+		unregister_snap_client(aarp_dl);
+	}
+	return rc;
+}
+
+/* Remove the AARP entries associated with a device. */
+void aarp_device_down(struct net_device *dev)
+{
+	int ct;
+
+	write_lock_bh(&aarp_lock);
+
+	for (ct = 0; ct < AARP_HASH_SIZE; ct++) {
+		__aarp_expire_device(&resolved[ct], dev);
+		__aarp_expire_device(&unresolved[ct], dev);
+		__aarp_expire_device(&proxies[ct], dev);
+	}
+
+	write_unlock_bh(&aarp_lock);
+}
+
+#ifdef CONFIG_PROC_FS
+/*
+ * Get the aarp entry that is in the chain described
+ * by the iterator.
+ * If pos is set then skip till that index.
+ * pos = 1 is the first entry
+ */
+static struct aarp_entry *iter_next(struct aarp_iter_state *iter, loff_t *pos)
+{
+	int ct = iter->bucket;
+	struct aarp_entry **table = iter->table;
+	loff_t off = 0;
+	struct aarp_entry *entry;
+
+ rescan:
+	while (ct < AARP_HASH_SIZE) {
+		for (entry = table[ct]; entry; entry = entry->next) {
+			if (!pos || ++off == *pos) {
+				iter->table = table;
+				iter->bucket = ct;
+				return entry;
+			}
+		}
+		++ct;
+	}
+
+	if (table == resolved) {
+		ct = 0;
+		table = unresolved;
+		goto rescan;
+	}
+	if (table == unresolved) {
+		ct = 0;
+		table = proxies;
+		goto rescan;
+	}
+	return NULL;
+}
+
+static void *aarp_seq_start(struct seq_file *seq, loff_t *pos)
+	__acquires(aarp_lock)
+{
+	struct aarp_iter_state *iter = seq->private;
+
+	read_lock_bh(&aarp_lock);
+	iter->table     = resolved;
+	iter->bucket    = 0;
+
+	return *pos ? iter_next(iter, pos) : SEQ_START_TOKEN;
+}
+
+static void *aarp_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+	struct aarp_entry *entry = v;
+	struct aarp_iter_state *iter = seq->private;
+
+	++*pos;
+
+	/* first line after header */
+	if (v == SEQ_START_TOKEN)
+		entry = iter_next(iter, NULL);
+
+	/* next entry in current bucket */
+	else if (entry->next)
+		entry = entry->next;
+
+	/* next bucket or table */
+	else {
+		++iter->bucket;
+		entry = iter_next(iter, NULL);
+	}
+	return entry;
+}
+
+static void aarp_seq_stop(struct seq_file *seq, void *v)
+	__releases(aarp_lock)
+{
+	read_unlock_bh(&aarp_lock);
+}
+
+static const char *dt2str(unsigned long ticks)
+{
+	static char buf[32];
+
+	sprintf(buf, "%ld.%02ld", ticks / HZ, ((ticks % HZ) * 100) / HZ);
+
+	return buf;
+}
+
+static int aarp_seq_show(struct seq_file *seq, void *v)
+{
+	struct aarp_iter_state *iter = seq->private;
+	struct aarp_entry *entry = v;
+	unsigned long now = jiffies;
+
+	if (v == SEQ_START_TOKEN)
+		seq_puts(seq,
+			 "Address  Interface   Hardware Address"
+			 "   Expires LastSend  Retry Status\n");
+	else {
+		seq_printf(seq, "%04X:%02X  %-12s",
+			   ntohs(entry->target_addr.s_net),
+			   (unsigned int) entry->target_addr.s_node,
+			   entry->dev ? entry->dev->name : "????");
+		seq_printf(seq, "%pM", entry->hwaddr);
+		seq_printf(seq, " %8s",
+			   dt2str((long)entry->expires_at - (long)now));
+		if (iter->table == unresolved)
+			seq_printf(seq, " %8s %6hu",
+				   dt2str(now - entry->last_sent),
+				   entry->xmit_count);
+		else
+			seq_puts(seq, "                ");
+		seq_printf(seq, " %s\n",
+			   (iter->table == resolved) ? "resolved"
+			   : (iter->table == unresolved) ? "unresolved"
+			   : (iter->table == proxies) ? "proxies"
+			   : "unknown");
+	}
+	return 0;
+}
+
+const struct seq_operations aarp_seq_ops = {
+	.start  = aarp_seq_start,
+	.next   = aarp_seq_next,
+	.stop   = aarp_seq_stop,
+	.show   = aarp_seq_show,
+};
+#endif
+
+/* General module cleanup. Called from cleanup_module() in ddp.c. */
+void aarp_cleanup_module(void)
+{
+	del_timer_sync(&aarp_timer);
+	unregister_netdevice_notifier(&aarp_notifier);
+	unregister_snap_client(aarp_dl);
+	aarp_purge();
+}