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, 655 insertions, 0 deletions

diff --git a/net/sunrpc/xprtmultipath.c b/net/sunrpc/xprtmultipath.c
new file mode 100644
index 000000000..701250b30
--- /dev/null
+++ b/net/sunrpc/xprtmultipath.c
@@ -0,0 +1,655 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Multipath support for RPC
+ *
+ * Copyright (c) 2015, 2016, Primary Data, Inc. All rights reserved.
+ *
+ * Trond Myklebust <trond.myklebust@primarydata.com>
+ *
+ */
+#include <linux/atomic.h>
+#include <linux/types.h>
+#include <linux/kref.h>
+#include <linux/list.h>
+#include <linux/rcupdate.h>
+#include <linux/rculist.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+#include <linux/sunrpc/xprt.h>
+#include <linux/sunrpc/addr.h>
+#include <linux/sunrpc/xprtmultipath.h>
+
+#include "sysfs.h"
+
+typedef struct rpc_xprt *(*xprt_switch_find_xprt_t)(struct rpc_xprt_switch *xps,
+		const struct rpc_xprt *cur);
+
+static const struct rpc_xprt_iter_ops rpc_xprt_iter_singular;
+static const struct rpc_xprt_iter_ops rpc_xprt_iter_roundrobin;
+static const struct rpc_xprt_iter_ops rpc_xprt_iter_listall;
+static const struct rpc_xprt_iter_ops rpc_xprt_iter_listoffline;
+
+static void xprt_switch_add_xprt_locked(struct rpc_xprt_switch *xps,
+		struct rpc_xprt *xprt)
+{
+	if (unlikely(xprt_get(xprt) == NULL))
+		return;
+	list_add_tail_rcu(&xprt->xprt_switch, &xps->xps_xprt_list);
+	smp_wmb();
+	if (xps->xps_nxprts == 0)
+		xps->xps_net = xprt->xprt_net;
+	xps->xps_nxprts++;
+	xps->xps_nactive++;
+}
+
+/**
+ * rpc_xprt_switch_add_xprt - Add a new rpc_xprt to an rpc_xprt_switch
+ * @xps: pointer to struct rpc_xprt_switch
+ * @xprt: pointer to struct rpc_xprt
+ *
+ * Adds xprt to the end of the list of struct rpc_xprt in xps.
+ */
+void rpc_xprt_switch_add_xprt(struct rpc_xprt_switch *xps,
+		struct rpc_xprt *xprt)
+{
+	if (xprt == NULL)
+		return;
+	spin_lock(&xps->xps_lock);
+	if (xps->xps_net == xprt->xprt_net || xps->xps_net == NULL)
+		xprt_switch_add_xprt_locked(xps, xprt);
+	spin_unlock(&xps->xps_lock);
+	rpc_sysfs_xprt_setup(xps, xprt, GFP_KERNEL);
+}
+
+static void xprt_switch_remove_xprt_locked(struct rpc_xprt_switch *xps,
+		struct rpc_xprt *xprt, bool offline)
+{
+	if (unlikely(xprt == NULL))
+		return;
+	if (!test_bit(XPRT_OFFLINE, &xprt->state) && offline)
+		xps->xps_nactive--;
+	xps->xps_nxprts--;
+	if (xps->xps_nxprts == 0)
+		xps->xps_net = NULL;
+	smp_wmb();
+	list_del_rcu(&xprt->xprt_switch);
+}
+
+/**
+ * rpc_xprt_switch_remove_xprt - Removes an rpc_xprt from a rpc_xprt_switch
+ * @xps: pointer to struct rpc_xprt_switch
+ * @xprt: pointer to struct rpc_xprt
+ * @offline: indicates if the xprt that's being removed is in an offline state
+ *
+ * Removes xprt from the list of struct rpc_xprt in xps.
+ */
+void rpc_xprt_switch_remove_xprt(struct rpc_xprt_switch *xps,
+		struct rpc_xprt *xprt, bool offline)
+{
+	spin_lock(&xps->xps_lock);
+	xprt_switch_remove_xprt_locked(xps, xprt, offline);
+	spin_unlock(&xps->xps_lock);
+	xprt_put(xprt);
+}
+
+static DEFINE_IDA(rpc_xprtswitch_ids);
+
+void xprt_multipath_cleanup_ids(void)
+{
+	ida_destroy(&rpc_xprtswitch_ids);
+}
+
+static int xprt_switch_alloc_id(struct rpc_xprt_switch *xps, gfp_t gfp_flags)
+{
+	int id;
+
+	id = ida_alloc(&rpc_xprtswitch_ids, gfp_flags);
+	if (id < 0)
+		return id;
+
+	xps->xps_id = id;
+	return 0;
+}
+
+static void xprt_switch_free_id(struct rpc_xprt_switch *xps)
+{
+	ida_free(&rpc_xprtswitch_ids, xps->xps_id);
+}
+
+/**
+ * xprt_switch_alloc - Allocate a new struct rpc_xprt_switch
+ * @xprt: pointer to struct rpc_xprt
+ * @gfp_flags: allocation flags
+ *
+ * On success, returns an initialised struct rpc_xprt_switch, containing
+ * the entry xprt. Returns NULL on failure.
+ */
+struct rpc_xprt_switch *xprt_switch_alloc(struct rpc_xprt *xprt,
+		gfp_t gfp_flags)
+{
+	struct rpc_xprt_switch *xps;
+
+	xps = kmalloc(sizeof(*xps), gfp_flags);
+	if (xps != NULL) {
+		spin_lock_init(&xps->xps_lock);
+		kref_init(&xps->xps_kref);
+		xprt_switch_alloc_id(xps, gfp_flags);
+		xps->xps_nxprts = xps->xps_nactive = 0;
+		atomic_long_set(&xps->xps_queuelen, 0);
+		xps->xps_net = NULL;
+		INIT_LIST_HEAD(&xps->xps_xprt_list);
+		xps->xps_iter_ops = &rpc_xprt_iter_singular;
+		rpc_sysfs_xprt_switch_setup(xps, xprt, gfp_flags);
+		xprt_switch_add_xprt_locked(xps, xprt);
+		xps->xps_nunique_destaddr_xprts = 1;
+		rpc_sysfs_xprt_setup(xps, xprt, gfp_flags);
+	}
+
+	return xps;
+}
+
+static void xprt_switch_free_entries(struct rpc_xprt_switch *xps)
+{
+	spin_lock(&xps->xps_lock);
+	while (!list_empty(&xps->xps_xprt_list)) {
+		struct rpc_xprt *xprt;
+
+		xprt = list_first_entry(&xps->xps_xprt_list,
+				struct rpc_xprt, xprt_switch);
+		xprt_switch_remove_xprt_locked(xps, xprt, true);
+		spin_unlock(&xps->xps_lock);
+		xprt_put(xprt);
+		spin_lock(&xps->xps_lock);
+	}
+	spin_unlock(&xps->xps_lock);
+}
+
+static void xprt_switch_free(struct kref *kref)
+{
+	struct rpc_xprt_switch *xps = container_of(kref,
+			struct rpc_xprt_switch, xps_kref);
+
+	xprt_switch_free_entries(xps);
+	rpc_sysfs_xprt_switch_destroy(xps);
+	xprt_switch_free_id(xps);
+	kfree_rcu(xps, xps_rcu);
+}
+
+/**
+ * xprt_switch_get - Return a reference to a rpc_xprt_switch
+ * @xps: pointer to struct rpc_xprt_switch
+ *
+ * Returns a reference to xps unless the refcount is already zero.
+ */
+struct rpc_xprt_switch *xprt_switch_get(struct rpc_xprt_switch *xps)
+{
+	if (xps != NULL && kref_get_unless_zero(&xps->xps_kref))
+		return xps;
+	return NULL;
+}
+
+/**
+ * xprt_switch_put - Release a reference to a rpc_xprt_switch
+ * @xps: pointer to struct rpc_xprt_switch
+ *
+ * Release the reference to xps, and free it once the refcount is zero.
+ */
+void xprt_switch_put(struct rpc_xprt_switch *xps)
+{
+	if (xps != NULL)
+		kref_put(&xps->xps_kref, xprt_switch_free);
+}
+
+/**
+ * rpc_xprt_switch_set_roundrobin - Set a round-robin policy on rpc_xprt_switch
+ * @xps: pointer to struct rpc_xprt_switch
+ *
+ * Sets a round-robin default policy for iterators acting on xps.
+ */
+void rpc_xprt_switch_set_roundrobin(struct rpc_xprt_switch *xps)
+{
+	if (READ_ONCE(xps->xps_iter_ops) != &rpc_xprt_iter_roundrobin)
+		WRITE_ONCE(xps->xps_iter_ops, &rpc_xprt_iter_roundrobin);
+}
+
+static
+const struct rpc_xprt_iter_ops *xprt_iter_ops(const struct rpc_xprt_iter *xpi)
+{
+	if (xpi->xpi_ops != NULL)
+		return xpi->xpi_ops;
+	return rcu_dereference(xpi->xpi_xpswitch)->xps_iter_ops;
+}
+
+static
+void xprt_iter_no_rewind(struct rpc_xprt_iter *xpi)
+{
+}
+
+static
+void xprt_iter_default_rewind(struct rpc_xprt_iter *xpi)
+{
+	WRITE_ONCE(xpi->xpi_cursor, NULL);
+}
+
+static
+bool xprt_is_active(const struct rpc_xprt *xprt)
+{
+	return (kref_read(&xprt->kref) != 0 &&
+		!test_bit(XPRT_OFFLINE, &xprt->state));
+}
+
+static
+struct rpc_xprt *xprt_switch_find_first_entry(struct list_head *head)
+{
+	struct rpc_xprt *pos;
+
+	list_for_each_entry_rcu(pos, head, xprt_switch) {
+		if (xprt_is_active(pos))
+			return pos;
+	}
+	return NULL;
+}
+
+static
+struct rpc_xprt *xprt_switch_find_first_entry_offline(struct list_head *head)
+{
+	struct rpc_xprt *pos;
+
+	list_for_each_entry_rcu(pos, head, xprt_switch) {
+		if (!xprt_is_active(pos))
+			return pos;
+	}
+	return NULL;
+}
+
+static
+struct rpc_xprt *xprt_iter_first_entry(struct rpc_xprt_iter *xpi)
+{
+	struct rpc_xprt_switch *xps = rcu_dereference(xpi->xpi_xpswitch);
+
+	if (xps == NULL)
+		return NULL;
+	return xprt_switch_find_first_entry(&xps->xps_xprt_list);
+}
+
+static
+struct rpc_xprt *_xprt_switch_find_current_entry(struct list_head *head,
+						 const struct rpc_xprt *cur,
+						 bool find_active)
+{
+	struct rpc_xprt *pos;
+	bool found = false;
+
+	list_for_each_entry_rcu(pos, head, xprt_switch) {
+		if (cur == pos)
+			found = true;
+		if (found && ((find_active && xprt_is_active(pos)) ||
+			      (!find_active && xprt_is_active(pos))))
+			return pos;
+	}
+	return NULL;
+}
+
+static
+struct rpc_xprt *xprt_switch_find_current_entry(struct list_head *head,
+						const struct rpc_xprt *cur)
+{
+	return _xprt_switch_find_current_entry(head, cur, true);
+}
+
+static
+struct rpc_xprt * _xprt_iter_current_entry(struct rpc_xprt_iter *xpi,
+		struct rpc_xprt *first_entry(struct list_head *head),
+		struct rpc_xprt *current_entry(struct list_head *head,
+					       const struct rpc_xprt *cur))
+{
+	struct rpc_xprt_switch *xps = rcu_dereference(xpi->xpi_xpswitch);
+	struct list_head *head;
+
+	if (xps == NULL)
+		return NULL;
+	head = &xps->xps_xprt_list;
+	if (xpi->xpi_cursor == NULL || xps->xps_nxprts < 2)
+		return first_entry(head);
+	return current_entry(head, xpi->xpi_cursor);
+}
+
+static
+struct rpc_xprt *xprt_iter_current_entry(struct rpc_xprt_iter *xpi)
+{
+	return _xprt_iter_current_entry(xpi, xprt_switch_find_first_entry,
+			xprt_switch_find_current_entry);
+}
+
+static
+struct rpc_xprt *xprt_switch_find_current_entry_offline(struct list_head *head,
+		const struct rpc_xprt *cur)
+{
+	return _xprt_switch_find_current_entry(head, cur, false);
+}
+
+static
+struct rpc_xprt *xprt_iter_current_entry_offline(struct rpc_xprt_iter *xpi)
+{
+	return _xprt_iter_current_entry(xpi,
+			xprt_switch_find_first_entry_offline,
+			xprt_switch_find_current_entry_offline);
+}
+
+bool rpc_xprt_switch_has_addr(struct rpc_xprt_switch *xps,
+			      const struct sockaddr *sap)
+{
+	struct list_head *head;
+	struct rpc_xprt *pos;
+
+	if (xps == NULL || sap == NULL)
+		return false;
+
+	head = &xps->xps_xprt_list;
+	list_for_each_entry_rcu(pos, head, xprt_switch) {
+		if (rpc_cmp_addr_port(sap, (struct sockaddr *)&pos->addr)) {
+			pr_info("RPC:   addr %s already in xprt switch\n",
+				pos->address_strings[RPC_DISPLAY_ADDR]);
+			return true;
+		}
+	}
+	return false;
+}
+
+static
+struct rpc_xprt *xprt_switch_find_next_entry(struct list_head *head,
+		const struct rpc_xprt *cur, bool check_active)
+{
+	struct rpc_xprt *pos, *prev = NULL;
+	bool found = false;
+
+	list_for_each_entry_rcu(pos, head, xprt_switch) {
+		if (cur == prev)
+			found = true;
+		/* for request to return active transports return only
+		 * active, for request to return offline transports
+		 * return only offline
+		 */
+		if (found && ((check_active && xprt_is_active(pos)) ||
+			      (!check_active && !xprt_is_active(pos))))
+			return pos;
+		prev = pos;
+	}
+	return NULL;
+}
+
+static
+struct rpc_xprt *xprt_switch_set_next_cursor(struct rpc_xprt_switch *xps,
+		struct rpc_xprt **cursor,
+		xprt_switch_find_xprt_t find_next)
+{
+	struct rpc_xprt *pos, *old;
+
+	old = smp_load_acquire(cursor);
+	pos = find_next(xps, old);
+	smp_store_release(cursor, pos);
+	return pos;
+}
+
+static
+struct rpc_xprt *xprt_iter_next_entry_multiple(struct rpc_xprt_iter *xpi,
+		xprt_switch_find_xprt_t find_next)
+{
+	struct rpc_xprt_switch *xps = rcu_dereference(xpi->xpi_xpswitch);
+
+	if (xps == NULL)
+		return NULL;
+	return xprt_switch_set_next_cursor(xps, &xpi->xpi_cursor, find_next);
+}
+
+static
+struct rpc_xprt *__xprt_switch_find_next_entry_roundrobin(struct list_head *head,
+		const struct rpc_xprt *cur)
+{
+	struct rpc_xprt *ret;
+
+	ret = xprt_switch_find_next_entry(head, cur, true);
+	if (ret != NULL)
+		return ret;
+	return xprt_switch_find_first_entry(head);
+}
+
+static
+struct rpc_xprt *xprt_switch_find_next_entry_roundrobin(struct rpc_xprt_switch *xps,
+		const struct rpc_xprt *cur)
+{
+	struct list_head *head = &xps->xps_xprt_list;
+	struct rpc_xprt *xprt;
+	unsigned int nactive;
+
+	for (;;) {
+		unsigned long xprt_queuelen, xps_queuelen;
+
+		xprt = __xprt_switch_find_next_entry_roundrobin(head, cur);
+		if (!xprt)
+			break;
+		xprt_queuelen = atomic_long_read(&xprt->queuelen);
+		xps_queuelen = atomic_long_read(&xps->xps_queuelen);
+		nactive = READ_ONCE(xps->xps_nactive);
+		/* Exit loop if xprt_queuelen <= average queue length */
+		if (xprt_queuelen * nactive <= xps_queuelen)
+			break;
+		cur = xprt;
+	}
+	return xprt;
+}
+
+static
+struct rpc_xprt *xprt_iter_next_entry_roundrobin(struct rpc_xprt_iter *xpi)
+{
+	return xprt_iter_next_entry_multiple(xpi,
+			xprt_switch_find_next_entry_roundrobin);
+}
+
+static
+struct rpc_xprt *xprt_switch_find_next_entry_all(struct rpc_xprt_switch *xps,
+		const struct rpc_xprt *cur)
+{
+	return xprt_switch_find_next_entry(&xps->xps_xprt_list, cur, true);
+}
+
+static
+struct rpc_xprt *xprt_switch_find_next_entry_offline(struct rpc_xprt_switch *xps,
+		const struct rpc_xprt *cur)
+{
+	return xprt_switch_find_next_entry(&xps->xps_xprt_list, cur, false);
+}
+
+static
+struct rpc_xprt *xprt_iter_next_entry_all(struct rpc_xprt_iter *xpi)
+{
+	return xprt_iter_next_entry_multiple(xpi,
+			xprt_switch_find_next_entry_all);
+}
+
+static
+struct rpc_xprt *xprt_iter_next_entry_offline(struct rpc_xprt_iter *xpi)
+{
+	return xprt_iter_next_entry_multiple(xpi,
+			xprt_switch_find_next_entry_offline);
+}
+
+/*
+ * xprt_iter_rewind - Resets the xprt iterator
+ * @xpi: pointer to rpc_xprt_iter
+ *
+ * Resets xpi to ensure that it points to the first entry in the list
+ * of transports.
+ */
+void xprt_iter_rewind(struct rpc_xprt_iter *xpi)
+{
+	rcu_read_lock();
+	xprt_iter_ops(xpi)->xpi_rewind(xpi);
+	rcu_read_unlock();
+}
+
+static void __xprt_iter_init(struct rpc_xprt_iter *xpi,
+		struct rpc_xprt_switch *xps,
+		const struct rpc_xprt_iter_ops *ops)
+{
+	rcu_assign_pointer(xpi->xpi_xpswitch, xprt_switch_get(xps));
+	xpi->xpi_cursor = NULL;
+	xpi->xpi_ops = ops;
+}
+
+/**
+ * xprt_iter_init - Initialise an xprt iterator
+ * @xpi: pointer to rpc_xprt_iter
+ * @xps: pointer to rpc_xprt_switch
+ *
+ * Initialises the iterator to use the default iterator ops
+ * as set in xps. This function is mainly intended for internal
+ * use in the rpc_client.
+ */
+void xprt_iter_init(struct rpc_xprt_iter *xpi,
+		struct rpc_xprt_switch *xps)
+{
+	__xprt_iter_init(xpi, xps, NULL);
+}
+
+/**
+ * xprt_iter_init_listall - Initialise an xprt iterator
+ * @xpi: pointer to rpc_xprt_iter
+ * @xps: pointer to rpc_xprt_switch
+ *
+ * Initialises the iterator to iterate once through the entire list
+ * of entries in xps.
+ */
+void xprt_iter_init_listall(struct rpc_xprt_iter *xpi,
+		struct rpc_xprt_switch *xps)
+{
+	__xprt_iter_init(xpi, xps, &rpc_xprt_iter_listall);
+}
+
+void xprt_iter_init_listoffline(struct rpc_xprt_iter *xpi,
+		struct rpc_xprt_switch *xps)
+{
+	__xprt_iter_init(xpi, xps, &rpc_xprt_iter_listoffline);
+}
+
+/**
+ * xprt_iter_xchg_switch - Atomically swap out the rpc_xprt_switch
+ * @xpi: pointer to rpc_xprt_iter
+ * @newswitch: pointer to a new rpc_xprt_switch or NULL
+ *
+ * Swaps out the existing xpi->xpi_xpswitch with a new value.
+ */
+struct rpc_xprt_switch *xprt_iter_xchg_switch(struct rpc_xprt_iter *xpi,
+		struct rpc_xprt_switch *newswitch)
+{
+	struct rpc_xprt_switch __rcu *oldswitch;
+
+	/* Atomically swap out the old xpswitch */
+	oldswitch = xchg(&xpi->xpi_xpswitch, RCU_INITIALIZER(newswitch));
+	if (newswitch != NULL)
+		xprt_iter_rewind(xpi);
+	return rcu_dereference_protected(oldswitch, true);
+}
+
+/**
+ * xprt_iter_destroy - Destroys the xprt iterator
+ * @xpi: pointer to rpc_xprt_iter
+ */
+void xprt_iter_destroy(struct rpc_xprt_iter *xpi)
+{
+	xprt_switch_put(xprt_iter_xchg_switch(xpi, NULL));
+}
+
+/**
+ * xprt_iter_xprt - Returns the rpc_xprt pointed to by the cursor
+ * @xpi: pointer to rpc_xprt_iter
+ *
+ * Returns a pointer to the struct rpc_xprt that is currently
+ * pointed to by the cursor.
+ * Caller must be holding rcu_read_lock().
+ */
+struct rpc_xprt *xprt_iter_xprt(struct rpc_xprt_iter *xpi)
+{
+	WARN_ON_ONCE(!rcu_read_lock_held());
+	return xprt_iter_ops(xpi)->xpi_xprt(xpi);
+}
+
+static
+struct rpc_xprt *xprt_iter_get_helper(struct rpc_xprt_iter *xpi,
+		struct rpc_xprt *(*fn)(struct rpc_xprt_iter *))
+{
+	struct rpc_xprt *ret;
+
+	do {
+		ret = fn(xpi);
+		if (ret == NULL)
+			break;
+		ret = xprt_get(ret);
+	} while (ret == NULL);
+	return ret;
+}
+
+/**
+ * xprt_iter_get_xprt - Returns the rpc_xprt pointed to by the cursor
+ * @xpi: pointer to rpc_xprt_iter
+ *
+ * Returns a reference to the struct rpc_xprt that is currently
+ * pointed to by the cursor.
+ */
+struct rpc_xprt *xprt_iter_get_xprt(struct rpc_xprt_iter *xpi)
+{
+	struct rpc_xprt *xprt;
+
+	rcu_read_lock();
+	xprt = xprt_iter_get_helper(xpi, xprt_iter_ops(xpi)->xpi_xprt);
+	rcu_read_unlock();
+	return xprt;
+}
+
+/**
+ * xprt_iter_get_next - Returns the next rpc_xprt following the cursor
+ * @xpi: pointer to rpc_xprt_iter
+ *
+ * Returns a reference to the struct rpc_xprt that immediately follows the
+ * entry pointed to by the cursor.
+ */
+struct rpc_xprt *xprt_iter_get_next(struct rpc_xprt_iter *xpi)
+{
+	struct rpc_xprt *xprt;
+
+	rcu_read_lock();
+	xprt = xprt_iter_get_helper(xpi, xprt_iter_ops(xpi)->xpi_next);
+	rcu_read_unlock();
+	return xprt;
+}
+
+/* Policy for always returning the first entry in the rpc_xprt_switch */
+static
+const struct rpc_xprt_iter_ops rpc_xprt_iter_singular = {
+	.xpi_rewind = xprt_iter_no_rewind,
+	.xpi_xprt = xprt_iter_first_entry,
+	.xpi_next = xprt_iter_first_entry,
+};
+
+/* Policy for round-robin iteration of entries in the rpc_xprt_switch */
+static
+const struct rpc_xprt_iter_ops rpc_xprt_iter_roundrobin = {
+	.xpi_rewind = xprt_iter_default_rewind,
+	.xpi_xprt = xprt_iter_current_entry,
+	.xpi_next = xprt_iter_next_entry_roundrobin,
+};
+
+/* Policy for once-through iteration of entries in the rpc_xprt_switch */
+static
+const struct rpc_xprt_iter_ops rpc_xprt_iter_listall = {
+	.xpi_rewind = xprt_iter_default_rewind,
+	.xpi_xprt = xprt_iter_current_entry,
+	.xpi_next = xprt_iter_next_entry_all,
+};
+
+static
+const struct rpc_xprt_iter_ops rpc_xprt_iter_listoffline = {
+	.xpi_rewind = xprt_iter_default_rewind,
+	.xpi_xprt = xprt_iter_current_entry_offline,
+	.xpi_next = xprt_iter_next_entry_offline,
+};