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, 465 insertions, 0 deletions

diff --git a/drivers/net/ethernet/chelsio/cxgb3/l2t.c b/drivers/net/ethernet/chelsio/cxgb3/l2t.c
new file mode 100644
index 000000000..9749d1239
--- /dev/null
+++ b/drivers/net/ethernet/chelsio/cxgb3/l2t.c
@@ -0,0 +1,465 @@
+/*
+ * Copyright (c) 2003-2008 Chelsio, Inc. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/if.h>
+#include <linux/if_vlan.h>
+#include <linux/jhash.h>
+#include <linux/slab.h>
+#include <linux/export.h>
+#include <net/neighbour.h>
+#include "common.h"
+#include "t3cdev.h"
+#include "cxgb3_defs.h"
+#include "l2t.h"
+#include "t3_cpl.h"
+#include "firmware_exports.h"
+
+#define VLAN_NONE 0xfff
+
+/*
+ * Module locking notes:  There is a RW lock protecting the L2 table as a
+ * whole plus a spinlock per L2T entry.  Entry lookups and allocations happen
+ * under the protection of the table lock, individual entry changes happen
+ * while holding that entry's spinlock.  The table lock nests outside the
+ * entry locks.  Allocations of new entries take the table lock as writers so
+ * no other lookups can happen while allocating new entries.  Entry updates
+ * take the table lock as readers so multiple entries can be updated in
+ * parallel.  An L2T entry can be dropped by decrementing its reference count
+ * and therefore can happen in parallel with entry allocation but no entry
+ * can change state or increment its ref count during allocation as both of
+ * these perform lookups.
+ */
+
+static inline unsigned int vlan_prio(const struct l2t_entry *e)
+{
+	return e->vlan >> 13;
+}
+
+static inline unsigned int arp_hash(u32 key, int ifindex,
+				    const struct l2t_data *d)
+{
+	return jhash_2words(key, ifindex, 0) & (d->nentries - 1);
+}
+
+static inline void neigh_replace(struct l2t_entry *e, struct neighbour *n)
+{
+	neigh_hold(n);
+	if (e->neigh)
+		neigh_release(e->neigh);
+	e->neigh = n;
+}
+
+/*
+ * Set up an L2T entry and send any packets waiting in the arp queue.  The
+ * supplied skb is used for the CPL_L2T_WRITE_REQ.  Must be called with the
+ * entry locked.
+ */
+static int setup_l2e_send_pending(struct t3cdev *dev, struct sk_buff *skb,
+				  struct l2t_entry *e)
+{
+	struct cpl_l2t_write_req *req;
+	struct sk_buff *tmp;
+
+	if (!skb) {
+		skb = alloc_skb(sizeof(*req), GFP_ATOMIC);
+		if (!skb)
+			return -ENOMEM;
+	}
+
+	req = __skb_put(skb, sizeof(*req));
+	req->wr.wr_hi = htonl(V_WR_OP(FW_WROPCODE_FORWARD));
+	OPCODE_TID(req) = htonl(MK_OPCODE_TID(CPL_L2T_WRITE_REQ, e->idx));
+	req->params = htonl(V_L2T_W_IDX(e->idx) | V_L2T_W_IFF(e->smt_idx) |
+			    V_L2T_W_VLAN(e->vlan & VLAN_VID_MASK) |
+			    V_L2T_W_PRIO(vlan_prio(e)));
+	memcpy(e->dmac, e->neigh->ha, sizeof(e->dmac));
+	memcpy(req->dst_mac, e->dmac, sizeof(req->dst_mac));
+	skb->priority = CPL_PRIORITY_CONTROL;
+	cxgb3_ofld_send(dev, skb);
+
+	skb_queue_walk_safe(&e->arpq, skb, tmp) {
+		__skb_unlink(skb, &e->arpq);
+		cxgb3_ofld_send(dev, skb);
+	}
+	e->state = L2T_STATE_VALID;
+
+	return 0;
+}
+
+/*
+ * Add a packet to the an L2T entry's queue of packets awaiting resolution.
+ * Must be called with the entry's lock held.
+ */
+static inline void arpq_enqueue(struct l2t_entry *e, struct sk_buff *skb)
+{
+	__skb_queue_tail(&e->arpq, skb);
+}
+
+int t3_l2t_send_slow(struct t3cdev *dev, struct sk_buff *skb,
+		     struct l2t_entry *e)
+{
+again:
+	switch (e->state) {
+	case L2T_STATE_STALE:	/* entry is stale, kick off revalidation */
+		neigh_event_send(e->neigh, NULL);
+		spin_lock_bh(&e->lock);
+		if (e->state == L2T_STATE_STALE)
+			e->state = L2T_STATE_VALID;
+		spin_unlock_bh(&e->lock);
+		fallthrough;
+	case L2T_STATE_VALID:	/* fast-path, send the packet on */
+		return cxgb3_ofld_send(dev, skb);
+	case L2T_STATE_RESOLVING:
+		spin_lock_bh(&e->lock);
+		if (e->state != L2T_STATE_RESOLVING) {
+			/* ARP already completed */
+			spin_unlock_bh(&e->lock);
+			goto again;
+		}
+		arpq_enqueue(e, skb);
+		spin_unlock_bh(&e->lock);
+
+		/*
+		 * Only the first packet added to the arpq should kick off
+		 * resolution.  However, because the alloc_skb below can fail,
+		 * we allow each packet added to the arpq to retry resolution
+		 * as a way of recovering from transient memory exhaustion.
+		 * A better way would be to use a work request to retry L2T
+		 * entries when there's no memory.
+		 */
+		if (!neigh_event_send(e->neigh, NULL)) {
+			skb = alloc_skb(sizeof(struct cpl_l2t_write_req),
+					GFP_ATOMIC);
+			if (!skb)
+				break;
+
+			spin_lock_bh(&e->lock);
+			if (!skb_queue_empty(&e->arpq))
+				setup_l2e_send_pending(dev, skb, e);
+			else	/* we lost the race */
+				__kfree_skb(skb);
+			spin_unlock_bh(&e->lock);
+		}
+	}
+	return 0;
+}
+
+EXPORT_SYMBOL(t3_l2t_send_slow);
+
+void t3_l2t_send_event(struct t3cdev *dev, struct l2t_entry *e)
+{
+again:
+	switch (e->state) {
+	case L2T_STATE_STALE:	/* entry is stale, kick off revalidation */
+		neigh_event_send(e->neigh, NULL);
+		spin_lock_bh(&e->lock);
+		if (e->state == L2T_STATE_STALE) {
+			e->state = L2T_STATE_VALID;
+		}
+		spin_unlock_bh(&e->lock);
+		return;
+	case L2T_STATE_VALID:	/* fast-path, send the packet on */
+		return;
+	case L2T_STATE_RESOLVING:
+		spin_lock_bh(&e->lock);
+		if (e->state != L2T_STATE_RESOLVING) {
+			/* ARP already completed */
+			spin_unlock_bh(&e->lock);
+			goto again;
+		}
+		spin_unlock_bh(&e->lock);
+
+		/*
+		 * Only the first packet added to the arpq should kick off
+		 * resolution.  However, because the alloc_skb below can fail,
+		 * we allow each packet added to the arpq to retry resolution
+		 * as a way of recovering from transient memory exhaustion.
+		 * A better way would be to use a work request to retry L2T
+		 * entries when there's no memory.
+		 */
+		neigh_event_send(e->neigh, NULL);
+	}
+}
+
+EXPORT_SYMBOL(t3_l2t_send_event);
+
+/*
+ * Allocate a free L2T entry.  Must be called with l2t_data.lock held.
+ */
+static struct l2t_entry *alloc_l2e(struct l2t_data *d)
+{
+	struct l2t_entry *end, *e, **p;
+
+	if (!atomic_read(&d->nfree))
+		return NULL;
+
+	/* there's definitely a free entry */
+	for (e = d->rover, end = &d->l2tab[d->nentries]; e != end; ++e)
+		if (atomic_read(&e->refcnt) == 0)
+			goto found;
+
+	for (e = &d->l2tab[1]; atomic_read(&e->refcnt); ++e) ;
+found:
+	d->rover = e + 1;
+	atomic_dec(&d->nfree);
+
+	/*
+	 * The entry we found may be an inactive entry that is
+	 * presently in the hash table.  We need to remove it.
+	 */
+	if (e->state != L2T_STATE_UNUSED) {
+		int hash = arp_hash(e->addr, e->ifindex, d);
+
+		for (p = &d->l2tab[hash].first; *p; p = &(*p)->next)
+			if (*p == e) {
+				*p = e->next;
+				break;
+			}
+		e->state = L2T_STATE_UNUSED;
+	}
+	return e;
+}
+
+/*
+ * Called when an L2T entry has no more users.  The entry is left in the hash
+ * table since it is likely to be reused but we also bump nfree to indicate
+ * that the entry can be reallocated for a different neighbor.  We also drop
+ * the existing neighbor reference in case the neighbor is going away and is
+ * waiting on our reference.
+ *
+ * Because entries can be reallocated to other neighbors once their ref count
+ * drops to 0 we need to take the entry's lock to avoid races with a new
+ * incarnation.
+ */
+void t3_l2e_free(struct l2t_data *d, struct l2t_entry *e)
+{
+	spin_lock_bh(&e->lock);
+	if (atomic_read(&e->refcnt) == 0) {	/* hasn't been recycled */
+		if (e->neigh) {
+			neigh_release(e->neigh);
+			e->neigh = NULL;
+		}
+	}
+	spin_unlock_bh(&e->lock);
+	atomic_inc(&d->nfree);
+}
+
+EXPORT_SYMBOL(t3_l2e_free);
+
+/*
+ * Update an L2T entry that was previously used for the same next hop as neigh.
+ * Must be called with softirqs disabled.
+ */
+static inline void reuse_entry(struct l2t_entry *e, struct neighbour *neigh)
+{
+	unsigned int nud_state;
+
+	spin_lock(&e->lock);	/* avoid race with t3_l2t_free */
+
+	if (neigh != e->neigh)
+		neigh_replace(e, neigh);
+	nud_state = neigh->nud_state;
+	if (memcmp(e->dmac, neigh->ha, sizeof(e->dmac)) ||
+	    !(nud_state & NUD_VALID))
+		e->state = L2T_STATE_RESOLVING;
+	else if (nud_state & NUD_CONNECTED)
+		e->state = L2T_STATE_VALID;
+	else
+		e->state = L2T_STATE_STALE;
+	spin_unlock(&e->lock);
+}
+
+struct l2t_entry *t3_l2t_get(struct t3cdev *cdev, struct dst_entry *dst,
+			     struct net_device *dev, const void *daddr)
+{
+	struct l2t_entry *e = NULL;
+	struct neighbour *neigh;
+	struct port_info *p;
+	struct l2t_data *d;
+	int hash;
+	u32 addr;
+	int ifidx;
+	int smt_idx;
+
+	rcu_read_lock();
+	neigh = dst_neigh_lookup(dst, daddr);
+	if (!neigh)
+		goto done_rcu;
+
+	addr = *(u32 *) neigh->primary_key;
+	ifidx = neigh->dev->ifindex;
+
+	if (!dev)
+		dev = neigh->dev;
+	p = netdev_priv(dev);
+	smt_idx = p->port_id;
+
+	d = L2DATA(cdev);
+	if (!d)
+		goto done_rcu;
+
+	hash = arp_hash(addr, ifidx, d);
+
+	write_lock_bh(&d->lock);
+	for (e = d->l2tab[hash].first; e; e = e->next)
+		if (e->addr == addr && e->ifindex == ifidx &&
+		    e->smt_idx == smt_idx) {
+			l2t_hold(d, e);
+			if (atomic_read(&e->refcnt) == 1)
+				reuse_entry(e, neigh);
+			goto done_unlock;
+		}
+
+	/* Need to allocate a new entry */
+	e = alloc_l2e(d);
+	if (e) {
+		spin_lock(&e->lock);	/* avoid race with t3_l2t_free */
+		e->next = d->l2tab[hash].first;
+		d->l2tab[hash].first = e;
+		e->state = L2T_STATE_RESOLVING;
+		e->addr = addr;
+		e->ifindex = ifidx;
+		e->smt_idx = smt_idx;
+		atomic_set(&e->refcnt, 1);
+		neigh_replace(e, neigh);
+		if (is_vlan_dev(neigh->dev))
+			e->vlan = vlan_dev_vlan_id(neigh->dev);
+		else
+			e->vlan = VLAN_NONE;
+		spin_unlock(&e->lock);
+	}
+done_unlock:
+	write_unlock_bh(&d->lock);
+done_rcu:
+	if (neigh)
+		neigh_release(neigh);
+	rcu_read_unlock();
+	return e;
+}
+
+EXPORT_SYMBOL(t3_l2t_get);
+
+/*
+ * Called when address resolution fails for an L2T entry to handle packets
+ * on the arpq head.  If a packet specifies a failure handler it is invoked,
+ * otherwise the packets is sent to the offload device.
+ *
+ * XXX: maybe we should abandon the latter behavior and just require a failure
+ * handler.
+ */
+static void handle_failed_resolution(struct t3cdev *dev, struct sk_buff_head *arpq)
+{
+	struct sk_buff *skb, *tmp;
+
+	skb_queue_walk_safe(arpq, skb, tmp) {
+		struct l2t_skb_cb *cb = L2T_SKB_CB(skb);
+
+		__skb_unlink(skb, arpq);
+		if (cb->arp_failure_handler)
+			cb->arp_failure_handler(dev, skb);
+		else
+			cxgb3_ofld_send(dev, skb);
+	}
+}
+
+/*
+ * Called when the host's ARP layer makes a change to some entry that is
+ * loaded into the HW L2 table.
+ */
+void t3_l2t_update(struct t3cdev *dev, struct neighbour *neigh)
+{
+	struct sk_buff_head arpq;
+	struct l2t_entry *e;
+	struct l2t_data *d = L2DATA(dev);
+	u32 addr = *(u32 *) neigh->primary_key;
+	int ifidx = neigh->dev->ifindex;
+	int hash = arp_hash(addr, ifidx, d);
+
+	read_lock_bh(&d->lock);
+	for (e = d->l2tab[hash].first; e; e = e->next)
+		if (e->addr == addr && e->ifindex == ifidx) {
+			spin_lock(&e->lock);
+			goto found;
+		}
+	read_unlock_bh(&d->lock);
+	return;
+
+found:
+	__skb_queue_head_init(&arpq);
+
+	read_unlock(&d->lock);
+	if (atomic_read(&e->refcnt)) {
+		if (neigh != e->neigh)
+			neigh_replace(e, neigh);
+
+		if (e->state == L2T_STATE_RESOLVING) {
+			if (neigh->nud_state & NUD_FAILED) {
+				skb_queue_splice_init(&e->arpq, &arpq);
+			} else if (neigh->nud_state & (NUD_CONNECTED|NUD_STALE))
+				setup_l2e_send_pending(dev, NULL, e);
+		} else {
+			e->state = neigh->nud_state & NUD_CONNECTED ?
+			    L2T_STATE_VALID : L2T_STATE_STALE;
+			if (!ether_addr_equal(e->dmac, neigh->ha))
+				setup_l2e_send_pending(dev, NULL, e);
+		}
+	}
+	spin_unlock_bh(&e->lock);
+
+	if (!skb_queue_empty(&arpq))
+		handle_failed_resolution(dev, &arpq);
+}
+
+struct l2t_data *t3_init_l2t(unsigned int l2t_capacity)
+{
+	struct l2t_data *d;
+	int i;
+
+	d = kvzalloc(struct_size(d, l2tab, l2t_capacity), GFP_KERNEL);
+	if (!d)
+		return NULL;
+
+	d->nentries = l2t_capacity;
+	d->rover = &d->l2tab[1];	/* entry 0 is not used */
+	atomic_set(&d->nfree, l2t_capacity - 1);
+	rwlock_init(&d->lock);
+
+	for (i = 0; i < l2t_capacity; ++i) {
+		d->l2tab[i].idx = i;
+		d->l2tab[i].state = L2T_STATE_UNUSED;
+		__skb_queue_head_init(&d->l2tab[i].arpq);
+		spin_lock_init(&d->l2tab[i].lock);
+		atomic_set(&d->l2tab[i].refcnt, 0);
+	}
+	return d;
+}