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, 850 insertions, 0 deletions

diff --git a/drivers/net/ethernet/netronome/nfp/abm/qdisc.c b/drivers/net/ethernet/netronome/nfp/abm/qdisc.c
new file mode 100644
index 000000000..2a5cc6422
--- /dev/null
+++ b/drivers/net/ethernet/netronome/nfp/abm/qdisc.c
@@ -0,0 +1,850 @@
+// SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+/* Copyright (C) 2018 Netronome Systems, Inc. */
+
+#include <linux/rtnetlink.h>
+#include <net/pkt_cls.h>
+#include <net/pkt_sched.h>
+#include <net/red.h>
+
+#include "../nfpcore/nfp_cpp.h"
+#include "../nfp_app.h"
+#include "../nfp_main.h"
+#include "../nfp_net.h"
+#include "../nfp_port.h"
+#include "main.h"
+
+static bool nfp_abm_qdisc_is_red(struct nfp_qdisc *qdisc)
+{
+	return qdisc->type == NFP_QDISC_RED || qdisc->type == NFP_QDISC_GRED;
+}
+
+static bool nfp_abm_qdisc_child_valid(struct nfp_qdisc *qdisc, unsigned int id)
+{
+	return qdisc->children[id] &&
+	       qdisc->children[id] != NFP_QDISC_UNTRACKED;
+}
+
+static void *nfp_abm_qdisc_tree_deref_slot(void __rcu **slot)
+{
+	return rtnl_dereference(*slot);
+}
+
+static void
+nfp_abm_stats_propagate(struct nfp_alink_stats *parent,
+			struct nfp_alink_stats *child)
+{
+	parent->tx_pkts		+= child->tx_pkts;
+	parent->tx_bytes	+= child->tx_bytes;
+	parent->backlog_pkts	+= child->backlog_pkts;
+	parent->backlog_bytes	+= child->backlog_bytes;
+	parent->overlimits	+= child->overlimits;
+	parent->drops		+= child->drops;
+}
+
+static void
+nfp_abm_stats_update_red(struct nfp_abm_link *alink, struct nfp_qdisc *qdisc,
+			 unsigned int queue)
+{
+	struct nfp_cpp *cpp = alink->abm->app->cpp;
+	unsigned int i;
+	int err;
+
+	if (!qdisc->offloaded)
+		return;
+
+	for (i = 0; i < qdisc->red.num_bands; i++) {
+		err = nfp_abm_ctrl_read_q_stats(alink, i, queue,
+						&qdisc->red.band[i].stats);
+		if (err)
+			nfp_err(cpp, "RED stats (%d, %d) read failed with error %d\n",
+				i, queue, err);
+
+		err = nfp_abm_ctrl_read_q_xstats(alink, i, queue,
+						 &qdisc->red.band[i].xstats);
+		if (err)
+			nfp_err(cpp, "RED xstats (%d, %d) read failed with error %d\n",
+				i, queue, err);
+	}
+}
+
+static void
+nfp_abm_stats_update_mq(struct nfp_abm_link *alink, struct nfp_qdisc *qdisc)
+{
+	unsigned int i;
+
+	if (qdisc->type != NFP_QDISC_MQ)
+		return;
+
+	for (i = 0; i < alink->total_queues; i++)
+		if (nfp_abm_qdisc_child_valid(qdisc, i))
+			nfp_abm_stats_update_red(alink, qdisc->children[i], i);
+}
+
+static void __nfp_abm_stats_update(struct nfp_abm_link *alink, u64 time_now)
+{
+	alink->last_stats_update = time_now;
+	if (alink->root_qdisc)
+		nfp_abm_stats_update_mq(alink, alink->root_qdisc);
+}
+
+static void nfp_abm_stats_update(struct nfp_abm_link *alink)
+{
+	u64 now;
+
+	/* Limit the frequency of updates - stats of non-leaf qdiscs are a sum
+	 * of all their leafs, so we would read the same stat multiple times
+	 * for every dump.
+	 */
+	now = ktime_get();
+	if (now - alink->last_stats_update < NFP_ABM_STATS_REFRESH_IVAL)
+		return;
+
+	__nfp_abm_stats_update(alink, now);
+}
+
+static void
+nfp_abm_qdisc_unlink_children(struct nfp_qdisc *qdisc,
+			      unsigned int start, unsigned int end)
+{
+	unsigned int i;
+
+	for (i = start; i < end; i++)
+		if (nfp_abm_qdisc_child_valid(qdisc, i)) {
+			qdisc->children[i]->use_cnt--;
+			qdisc->children[i] = NULL;
+		}
+}
+
+static void
+nfp_abm_qdisc_offload_stop(struct nfp_abm_link *alink, struct nfp_qdisc *qdisc)
+{
+	unsigned int i;
+
+	/* Don't complain when qdisc is getting unlinked */
+	if (qdisc->use_cnt)
+		nfp_warn(alink->abm->app->cpp, "Offload of '%08x' stopped\n",
+			 qdisc->handle);
+
+	if (!nfp_abm_qdisc_is_red(qdisc))
+		return;
+
+	for (i = 0; i < qdisc->red.num_bands; i++) {
+		qdisc->red.band[i].stats.backlog_pkts = 0;
+		qdisc->red.band[i].stats.backlog_bytes = 0;
+	}
+}
+
+static int
+__nfp_abm_stats_init(struct nfp_abm_link *alink, unsigned int band,
+		     unsigned int queue, struct nfp_alink_stats *prev_stats,
+		     struct nfp_alink_xstats *prev_xstats)
+{
+	u64 backlog_pkts, backlog_bytes;
+	int err;
+
+	/* Don't touch the backlog, backlog can only be reset after it has
+	 * been reported back to the tc qdisc stats.
+	 */
+	backlog_pkts = prev_stats->backlog_pkts;
+	backlog_bytes = prev_stats->backlog_bytes;
+
+	err = nfp_abm_ctrl_read_q_stats(alink, band, queue, prev_stats);
+	if (err) {
+		nfp_err(alink->abm->app->cpp,
+			"RED stats init (%d, %d) failed with error %d\n",
+			band, queue, err);
+		return err;
+	}
+
+	err = nfp_abm_ctrl_read_q_xstats(alink, band, queue, prev_xstats);
+	if (err) {
+		nfp_err(alink->abm->app->cpp,
+			"RED xstats init (%d, %d) failed with error %d\n",
+			band, queue, err);
+		return err;
+	}
+
+	prev_stats->backlog_pkts = backlog_pkts;
+	prev_stats->backlog_bytes = backlog_bytes;
+	return 0;
+}
+
+static int
+nfp_abm_stats_init(struct nfp_abm_link *alink, struct nfp_qdisc *qdisc,
+		   unsigned int queue)
+{
+	unsigned int i;
+	int err;
+
+	for (i = 0; i < qdisc->red.num_bands; i++) {
+		err = __nfp_abm_stats_init(alink, i, queue,
+					   &qdisc->red.band[i].prev_stats,
+					   &qdisc->red.band[i].prev_xstats);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static void
+nfp_abm_offload_compile_red(struct nfp_abm_link *alink, struct nfp_qdisc *qdisc,
+			    unsigned int queue)
+{
+	bool good_red, good_gred;
+	unsigned int i;
+
+	good_red = qdisc->type == NFP_QDISC_RED &&
+		   qdisc->params_ok &&
+		   qdisc->use_cnt == 1 &&
+		   !alink->has_prio &&
+		   !qdisc->children[0];
+	good_gred = qdisc->type == NFP_QDISC_GRED &&
+		    qdisc->params_ok &&
+		    qdisc->use_cnt == 1;
+	qdisc->offload_mark = good_red || good_gred;
+
+	/* If we are starting offload init prev_stats */
+	if (qdisc->offload_mark && !qdisc->offloaded)
+		if (nfp_abm_stats_init(alink, qdisc, queue))
+			qdisc->offload_mark = false;
+
+	if (!qdisc->offload_mark)
+		return;
+
+	for (i = 0; i < alink->abm->num_bands; i++) {
+		enum nfp_abm_q_action act;
+
+		nfp_abm_ctrl_set_q_lvl(alink, i, queue,
+				       qdisc->red.band[i].threshold);
+		act = qdisc->red.band[i].ecn ?
+			NFP_ABM_ACT_MARK_DROP : NFP_ABM_ACT_DROP;
+		nfp_abm_ctrl_set_q_act(alink, i, queue, act);
+	}
+}
+
+static void
+nfp_abm_offload_compile_mq(struct nfp_abm_link *alink, struct nfp_qdisc *qdisc)
+{
+	unsigned int i;
+
+	qdisc->offload_mark = qdisc->type == NFP_QDISC_MQ;
+	if (!qdisc->offload_mark)
+		return;
+
+	for (i = 0; i < alink->total_queues; i++) {
+		struct nfp_qdisc *child = qdisc->children[i];
+
+		if (!nfp_abm_qdisc_child_valid(qdisc, i))
+			continue;
+
+		nfp_abm_offload_compile_red(alink, child, i);
+	}
+}
+
+void nfp_abm_qdisc_offload_update(struct nfp_abm_link *alink)
+{
+	struct nfp_abm *abm = alink->abm;
+	struct radix_tree_iter iter;
+	struct nfp_qdisc *qdisc;
+	void __rcu **slot;
+	size_t i;
+
+	/* Mark all thresholds as unconfigured */
+	for (i = 0; i < abm->num_bands; i++)
+		__bitmap_set(abm->threshold_undef,
+			     i * NFP_NET_MAX_RX_RINGS + alink->queue_base,
+			     alink->total_queues);
+
+	/* Clear offload marks */
+	radix_tree_for_each_slot(slot, &alink->qdiscs, &iter, 0) {
+		qdisc = nfp_abm_qdisc_tree_deref_slot(slot);
+		qdisc->offload_mark = false;
+	}
+
+	if (alink->root_qdisc)
+		nfp_abm_offload_compile_mq(alink, alink->root_qdisc);
+
+	/* Refresh offload status */
+	radix_tree_for_each_slot(slot, &alink->qdiscs, &iter, 0) {
+		qdisc = nfp_abm_qdisc_tree_deref_slot(slot);
+		if (!qdisc->offload_mark && qdisc->offloaded)
+			nfp_abm_qdisc_offload_stop(alink, qdisc);
+		qdisc->offloaded = qdisc->offload_mark;
+	}
+
+	/* Reset the unconfigured thresholds */
+	for (i = 0; i < abm->num_thresholds; i++)
+		if (test_bit(i, abm->threshold_undef))
+			__nfp_abm_ctrl_set_q_lvl(abm, i, NFP_ABM_LVL_INFINITY);
+
+	__nfp_abm_stats_update(alink, ktime_get());
+}
+
+static void
+nfp_abm_qdisc_clear_mq(struct net_device *netdev, struct nfp_abm_link *alink,
+		       struct nfp_qdisc *qdisc)
+{
+	struct radix_tree_iter iter;
+	unsigned int mq_refs = 0;
+	void __rcu **slot;
+
+	if (!qdisc->use_cnt)
+		return;
+	/* MQ doesn't notify well on destruction, we need special handling of
+	 * MQ's children.
+	 */
+	if (qdisc->type == NFP_QDISC_MQ &&
+	    qdisc == alink->root_qdisc &&
+	    netdev->reg_state == NETREG_UNREGISTERING)
+		return;
+
+	/* Count refs held by MQ instances and clear pointers */
+	radix_tree_for_each_slot(slot, &alink->qdiscs, &iter, 0) {
+		struct nfp_qdisc *mq = nfp_abm_qdisc_tree_deref_slot(slot);
+		unsigned int i;
+
+		if (mq->type != NFP_QDISC_MQ || mq->netdev != netdev)
+			continue;
+		for (i = 0; i < mq->num_children; i++)
+			if (mq->children[i] == qdisc) {
+				mq->children[i] = NULL;
+				mq_refs++;
+			}
+	}
+
+	WARN(qdisc->use_cnt != mq_refs, "non-zero qdisc use count: %d (- %d)\n",
+	     qdisc->use_cnt, mq_refs);
+}
+
+static void
+nfp_abm_qdisc_free(struct net_device *netdev, struct nfp_abm_link *alink,
+		   struct nfp_qdisc *qdisc)
+{
+	struct nfp_port *port = nfp_port_from_netdev(netdev);
+
+	if (!qdisc)
+		return;
+	nfp_abm_qdisc_clear_mq(netdev, alink, qdisc);
+	WARN_ON(radix_tree_delete(&alink->qdiscs,
+				  TC_H_MAJ(qdisc->handle)) != qdisc);
+
+	kfree(qdisc->children);
+	kfree(qdisc);
+
+	port->tc_offload_cnt--;
+}
+
+static struct nfp_qdisc *
+nfp_abm_qdisc_alloc(struct net_device *netdev, struct nfp_abm_link *alink,
+		    enum nfp_qdisc_type type, u32 parent_handle, u32 handle,
+		    unsigned int children)
+{
+	struct nfp_port *port = nfp_port_from_netdev(netdev);
+	struct nfp_qdisc *qdisc;
+	int err;
+
+	qdisc = kzalloc(sizeof(*qdisc), GFP_KERNEL);
+	if (!qdisc)
+		return NULL;
+
+	if (children) {
+		qdisc->children = kcalloc(children, sizeof(void *), GFP_KERNEL);
+		if (!qdisc->children)
+			goto err_free_qdisc;
+	}
+
+	qdisc->netdev = netdev;
+	qdisc->type = type;
+	qdisc->parent_handle = parent_handle;
+	qdisc->handle = handle;
+	qdisc->num_children = children;
+
+	err = radix_tree_insert(&alink->qdiscs, TC_H_MAJ(qdisc->handle), qdisc);
+	if (err) {
+		nfp_err(alink->abm->app->cpp,
+			"Qdisc insertion into radix tree failed: %d\n", err);
+		goto err_free_child_tbl;
+	}
+
+	port->tc_offload_cnt++;
+	return qdisc;
+
+err_free_child_tbl:
+	kfree(qdisc->children);
+err_free_qdisc:
+	kfree(qdisc);
+	return NULL;
+}
+
+static struct nfp_qdisc *
+nfp_abm_qdisc_find(struct nfp_abm_link *alink, u32 handle)
+{
+	return radix_tree_lookup(&alink->qdiscs, TC_H_MAJ(handle));
+}
+
+static int
+nfp_abm_qdisc_replace(struct net_device *netdev, struct nfp_abm_link *alink,
+		      enum nfp_qdisc_type type, u32 parent_handle, u32 handle,
+		      unsigned int children, struct nfp_qdisc **qdisc)
+{
+	*qdisc = nfp_abm_qdisc_find(alink, handle);
+	if (*qdisc) {
+		if (WARN_ON((*qdisc)->type != type))
+			return -EINVAL;
+		return 1;
+	}
+
+	*qdisc = nfp_abm_qdisc_alloc(netdev, alink, type, parent_handle, handle,
+				     children);
+	return *qdisc ? 0 : -ENOMEM;
+}
+
+static void
+nfp_abm_qdisc_destroy(struct net_device *netdev, struct nfp_abm_link *alink,
+		      u32 handle)
+{
+	struct nfp_qdisc *qdisc;
+
+	qdisc = nfp_abm_qdisc_find(alink, handle);
+	if (!qdisc)
+		return;
+
+	/* We don't get TC_SETUP_ROOT_QDISC w/ MQ when netdev is unregistered */
+	if (alink->root_qdisc == qdisc)
+		qdisc->use_cnt--;
+
+	nfp_abm_qdisc_unlink_children(qdisc, 0, qdisc->num_children);
+	nfp_abm_qdisc_free(netdev, alink, qdisc);
+
+	if (alink->root_qdisc == qdisc) {
+		alink->root_qdisc = NULL;
+		/* Only root change matters, other changes are acted upon on
+		 * the graft notification.
+		 */
+		nfp_abm_qdisc_offload_update(alink);
+	}
+}
+
+static int
+nfp_abm_qdisc_graft(struct nfp_abm_link *alink, u32 handle, u32 child_handle,
+		    unsigned int id)
+{
+	struct nfp_qdisc *parent, *child;
+
+	parent = nfp_abm_qdisc_find(alink, handle);
+	if (!parent)
+		return 0;
+
+	if (WARN(id >= parent->num_children,
+		 "graft child out of bound %d >= %d\n",
+		 id, parent->num_children))
+		return -EINVAL;
+
+	nfp_abm_qdisc_unlink_children(parent, id, id + 1);
+
+	child = nfp_abm_qdisc_find(alink, child_handle);
+	if (child)
+		child->use_cnt++;
+	else
+		child = NFP_QDISC_UNTRACKED;
+	parent->children[id] = child;
+
+	nfp_abm_qdisc_offload_update(alink);
+
+	return 0;
+}
+
+static void
+nfp_abm_stats_calculate(struct nfp_alink_stats *new,
+			struct nfp_alink_stats *old,
+			struct gnet_stats_basic_sync *bstats,
+			struct gnet_stats_queue *qstats)
+{
+	_bstats_update(bstats, new->tx_bytes - old->tx_bytes,
+		       new->tx_pkts - old->tx_pkts);
+	qstats->qlen += new->backlog_pkts - old->backlog_pkts;
+	qstats->backlog += new->backlog_bytes - old->backlog_bytes;
+	qstats->overlimits += new->overlimits - old->overlimits;
+	qstats->drops += new->drops - old->drops;
+}
+
+static void
+nfp_abm_stats_red_calculate(struct nfp_alink_xstats *new,
+			    struct nfp_alink_xstats *old,
+			    struct red_stats *stats)
+{
+	stats->forced_mark += new->ecn_marked - old->ecn_marked;
+	stats->pdrop += new->pdrop - old->pdrop;
+}
+
+static int
+nfp_abm_gred_stats(struct nfp_abm_link *alink, u32 handle,
+		   struct tc_gred_qopt_offload_stats *stats)
+{
+	struct nfp_qdisc *qdisc;
+	unsigned int i;
+
+	nfp_abm_stats_update(alink);
+
+	qdisc = nfp_abm_qdisc_find(alink, handle);
+	if (!qdisc)
+		return -EOPNOTSUPP;
+	/* If the qdisc offload has stopped we may need to adjust the backlog
+	 * counters back so carry on even if qdisc is not currently offloaded.
+	 */
+
+	for (i = 0; i < qdisc->red.num_bands; i++) {
+		if (!stats->xstats[i])
+			continue;
+
+		nfp_abm_stats_calculate(&qdisc->red.band[i].stats,
+					&qdisc->red.band[i].prev_stats,
+					&stats->bstats[i], &stats->qstats[i]);
+		qdisc->red.band[i].prev_stats = qdisc->red.band[i].stats;
+
+		nfp_abm_stats_red_calculate(&qdisc->red.band[i].xstats,
+					    &qdisc->red.band[i].prev_xstats,
+					    stats->xstats[i]);
+		qdisc->red.band[i].prev_xstats = qdisc->red.band[i].xstats;
+	}
+
+	return qdisc->offloaded ? 0 : -EOPNOTSUPP;
+}
+
+static bool
+nfp_abm_gred_check_params(struct nfp_abm_link *alink,
+			  struct tc_gred_qopt_offload *opt)
+{
+	struct nfp_cpp *cpp = alink->abm->app->cpp;
+	struct nfp_abm *abm = alink->abm;
+	unsigned int i;
+
+	if (opt->set.grio_on || opt->set.wred_on) {
+		nfp_warn(cpp, "GRED offload failed - GRIO and WRED not supported (p:%08x h:%08x)\n",
+			 opt->parent, opt->handle);
+		return false;
+	}
+	if (opt->set.dp_def != alink->def_band) {
+		nfp_warn(cpp, "GRED offload failed - default band must be %d (p:%08x h:%08x)\n",
+			 alink->def_band, opt->parent, opt->handle);
+		return false;
+	}
+	if (opt->set.dp_cnt != abm->num_bands) {
+		nfp_warn(cpp, "GRED offload failed - band count must be %d (p:%08x h:%08x)\n",
+			 abm->num_bands, opt->parent, opt->handle);
+		return false;
+	}
+
+	for (i = 0; i < abm->num_bands; i++) {
+		struct tc_gred_vq_qopt_offload_params *band = &opt->set.tab[i];
+
+		if (!band->present)
+			return false;
+		if (!band->is_ecn && !nfp_abm_has_drop(abm)) {
+			nfp_warn(cpp, "GRED offload failed - drop is not supported (ECN option required) (p:%08x h:%08x vq:%d)\n",
+				 opt->parent, opt->handle, i);
+			return false;
+		}
+		if (band->is_ecn && !nfp_abm_has_mark(abm)) {
+			nfp_warn(cpp, "GRED offload failed - ECN marking not supported (p:%08x h:%08x vq:%d)\n",
+				 opt->parent, opt->handle, i);
+			return false;
+		}
+		if (band->is_harddrop) {
+			nfp_warn(cpp, "GRED offload failed - harddrop is not supported (p:%08x h:%08x vq:%d)\n",
+				 opt->parent, opt->handle, i);
+			return false;
+		}
+		if (band->min != band->max) {
+			nfp_warn(cpp, "GRED offload failed - threshold mismatch (p:%08x h:%08x vq:%d)\n",
+				 opt->parent, opt->handle, i);
+			return false;
+		}
+		if (band->min > S32_MAX) {
+			nfp_warn(cpp, "GRED offload failed - threshold too large %d > %d (p:%08x h:%08x vq:%d)\n",
+				 band->min, S32_MAX, opt->parent, opt->handle,
+				 i);
+			return false;
+		}
+	}
+
+	return true;
+}
+
+static int
+nfp_abm_gred_replace(struct net_device *netdev, struct nfp_abm_link *alink,
+		     struct tc_gred_qopt_offload *opt)
+{
+	struct nfp_qdisc *qdisc;
+	unsigned int i;
+	int ret;
+
+	ret = nfp_abm_qdisc_replace(netdev, alink, NFP_QDISC_GRED, opt->parent,
+				    opt->handle, 0, &qdisc);
+	if (ret < 0)
+		return ret;
+
+	qdisc->params_ok = nfp_abm_gred_check_params(alink, opt);
+	if (qdisc->params_ok) {
+		qdisc->red.num_bands = opt->set.dp_cnt;
+		for (i = 0; i < qdisc->red.num_bands; i++) {
+			qdisc->red.band[i].ecn = opt->set.tab[i].is_ecn;
+			qdisc->red.band[i].threshold = opt->set.tab[i].min;
+		}
+	}
+
+	if (qdisc->use_cnt)
+		nfp_abm_qdisc_offload_update(alink);
+
+	return 0;
+}
+
+int nfp_abm_setup_tc_gred(struct net_device *netdev, struct nfp_abm_link *alink,
+			  struct tc_gred_qopt_offload *opt)
+{
+	switch (opt->command) {
+	case TC_GRED_REPLACE:
+		return nfp_abm_gred_replace(netdev, alink, opt);
+	case TC_GRED_DESTROY:
+		nfp_abm_qdisc_destroy(netdev, alink, opt->handle);
+		return 0;
+	case TC_GRED_STATS:
+		return nfp_abm_gred_stats(alink, opt->handle, &opt->stats);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int
+nfp_abm_red_xstats(struct nfp_abm_link *alink, struct tc_red_qopt_offload *opt)
+{
+	struct nfp_qdisc *qdisc;
+
+	nfp_abm_stats_update(alink);
+
+	qdisc = nfp_abm_qdisc_find(alink, opt->handle);
+	if (!qdisc || !qdisc->offloaded)
+		return -EOPNOTSUPP;
+
+	nfp_abm_stats_red_calculate(&qdisc->red.band[0].xstats,
+				    &qdisc->red.band[0].prev_xstats,
+				    opt->xstats);
+	qdisc->red.band[0].prev_xstats = qdisc->red.band[0].xstats;
+	return 0;
+}
+
+static int
+nfp_abm_red_stats(struct nfp_abm_link *alink, u32 handle,
+		  struct tc_qopt_offload_stats *stats)
+{
+	struct nfp_qdisc *qdisc;
+
+	nfp_abm_stats_update(alink);
+
+	qdisc = nfp_abm_qdisc_find(alink, handle);
+	if (!qdisc)
+		return -EOPNOTSUPP;
+	/* If the qdisc offload has stopped we may need to adjust the backlog
+	 * counters back so carry on even if qdisc is not currently offloaded.
+	 */
+
+	nfp_abm_stats_calculate(&qdisc->red.band[0].stats,
+				&qdisc->red.band[0].prev_stats,
+				stats->bstats, stats->qstats);
+	qdisc->red.band[0].prev_stats = qdisc->red.band[0].stats;
+
+	return qdisc->offloaded ? 0 : -EOPNOTSUPP;
+}
+
+static bool
+nfp_abm_red_check_params(struct nfp_abm_link *alink,
+			 struct tc_red_qopt_offload *opt)
+{
+	struct nfp_cpp *cpp = alink->abm->app->cpp;
+	struct nfp_abm *abm = alink->abm;
+
+	if (!opt->set.is_ecn && !nfp_abm_has_drop(abm)) {
+		nfp_warn(cpp, "RED offload failed - drop is not supported (ECN option required) (p:%08x h:%08x)\n",
+			 opt->parent, opt->handle);
+		return false;
+	}
+	if (opt->set.is_ecn && !nfp_abm_has_mark(abm)) {
+		nfp_warn(cpp, "RED offload failed - ECN marking not supported (p:%08x h:%08x)\n",
+			 opt->parent, opt->handle);
+		return false;
+	}
+	if (opt->set.is_harddrop) {
+		nfp_warn(cpp, "RED offload failed - harddrop is not supported (p:%08x h:%08x)\n",
+			 opt->parent, opt->handle);
+		return false;
+	}
+	if (opt->set.min != opt->set.max) {
+		nfp_warn(cpp, "RED offload failed - unsupported min/max parameters (p:%08x h:%08x)\n",
+			 opt->parent, opt->handle);
+		return false;
+	}
+	if (opt->set.min > NFP_ABM_LVL_INFINITY) {
+		nfp_warn(cpp, "RED offload failed - threshold too large %d > %d (p:%08x h:%08x)\n",
+			 opt->set.min, NFP_ABM_LVL_INFINITY, opt->parent,
+			 opt->handle);
+		return false;
+	}
+
+	return true;
+}
+
+static int
+nfp_abm_red_replace(struct net_device *netdev, struct nfp_abm_link *alink,
+		    struct tc_red_qopt_offload *opt)
+{
+	struct nfp_qdisc *qdisc;
+	int ret;
+
+	ret = nfp_abm_qdisc_replace(netdev, alink, NFP_QDISC_RED, opt->parent,
+				    opt->handle, 1, &qdisc);
+	if (ret < 0)
+		return ret;
+
+	/* If limit != 0 child gets reset */
+	if (opt->set.limit) {
+		if (nfp_abm_qdisc_child_valid(qdisc, 0))
+			qdisc->children[0]->use_cnt--;
+		qdisc->children[0] = NULL;
+	} else {
+		/* Qdisc was just allocated without a limit will use noop_qdisc,
+		 * i.e. a block hole.
+		 */
+		if (!ret)
+			qdisc->children[0] = NFP_QDISC_UNTRACKED;
+	}
+
+	qdisc->params_ok = nfp_abm_red_check_params(alink, opt);
+	if (qdisc->params_ok) {
+		qdisc->red.num_bands = 1;
+		qdisc->red.band[0].ecn = opt->set.is_ecn;
+		qdisc->red.band[0].threshold = opt->set.min;
+	}
+
+	if (qdisc->use_cnt == 1)
+		nfp_abm_qdisc_offload_update(alink);
+
+	return 0;
+}
+
+int nfp_abm_setup_tc_red(struct net_device *netdev, struct nfp_abm_link *alink,
+			 struct tc_red_qopt_offload *opt)
+{
+	switch (opt->command) {
+	case TC_RED_REPLACE:
+		return nfp_abm_red_replace(netdev, alink, opt);
+	case TC_RED_DESTROY:
+		nfp_abm_qdisc_destroy(netdev, alink, opt->handle);
+		return 0;
+	case TC_RED_STATS:
+		return nfp_abm_red_stats(alink, opt->handle, &opt->stats);
+	case TC_RED_XSTATS:
+		return nfp_abm_red_xstats(alink, opt);
+	case TC_RED_GRAFT:
+		return nfp_abm_qdisc_graft(alink, opt->handle,
+					   opt->child_handle, 0);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+static int
+nfp_abm_mq_create(struct net_device *netdev, struct nfp_abm_link *alink,
+		  struct tc_mq_qopt_offload *opt)
+{
+	struct nfp_qdisc *qdisc;
+	int ret;
+
+	ret = nfp_abm_qdisc_replace(netdev, alink, NFP_QDISC_MQ,
+				    TC_H_ROOT, opt->handle, alink->total_queues,
+				    &qdisc);
+	if (ret < 0)
+		return ret;
+
+	qdisc->params_ok = true;
+	qdisc->offloaded = true;
+	nfp_abm_qdisc_offload_update(alink);
+	return 0;
+}
+
+static int
+nfp_abm_mq_stats(struct nfp_abm_link *alink, u32 handle,
+		 struct tc_qopt_offload_stats *stats)
+{
+	struct nfp_qdisc *qdisc, *red;
+	unsigned int i, j;
+
+	qdisc = nfp_abm_qdisc_find(alink, handle);
+	if (!qdisc)
+		return -EOPNOTSUPP;
+
+	nfp_abm_stats_update(alink);
+
+	/* MQ stats are summed over the children in the core, so we need
+	 * to add up the unreported child values.
+	 */
+	memset(&qdisc->mq.stats, 0, sizeof(qdisc->mq.stats));
+	memset(&qdisc->mq.prev_stats, 0, sizeof(qdisc->mq.prev_stats));
+
+	for (i = 0; i < qdisc->num_children; i++) {
+		if (!nfp_abm_qdisc_child_valid(qdisc, i))
+			continue;
+
+		if (!nfp_abm_qdisc_is_red(qdisc->children[i]))
+			continue;
+		red = qdisc->children[i];
+
+		for (j = 0; j < red->red.num_bands; j++) {
+			nfp_abm_stats_propagate(&qdisc->mq.stats,
+						&red->red.band[j].stats);
+			nfp_abm_stats_propagate(&qdisc->mq.prev_stats,
+						&red->red.band[j].prev_stats);
+		}
+	}
+
+	nfp_abm_stats_calculate(&qdisc->mq.stats, &qdisc->mq.prev_stats,
+				stats->bstats, stats->qstats);
+
+	return qdisc->offloaded ? 0 : -EOPNOTSUPP;
+}
+
+int nfp_abm_setup_tc_mq(struct net_device *netdev, struct nfp_abm_link *alink,
+			struct tc_mq_qopt_offload *opt)
+{
+	switch (opt->command) {
+	case TC_MQ_CREATE:
+		return nfp_abm_mq_create(netdev, alink, opt);
+	case TC_MQ_DESTROY:
+		nfp_abm_qdisc_destroy(netdev, alink, opt->handle);
+		return 0;
+	case TC_MQ_STATS:
+		return nfp_abm_mq_stats(alink, opt->handle, &opt->stats);
+	case TC_MQ_GRAFT:
+		return nfp_abm_qdisc_graft(alink, opt->handle,
+					   opt->graft_params.child_handle,
+					   opt->graft_params.queue);
+	default:
+		return -EOPNOTSUPP;
+	}
+}
+
+int nfp_abm_setup_root(struct net_device *netdev, struct nfp_abm_link *alink,
+		       struct tc_root_qopt_offload *opt)
+{
+	if (opt->ingress)
+		return -EOPNOTSUPP;
+	if (alink->root_qdisc)
+		alink->root_qdisc->use_cnt--;
+	alink->root_qdisc = nfp_abm_qdisc_find(alink, opt->handle);
+	if (alink->root_qdisc)
+		alink->root_qdisc->use_cnt++;
+
+	nfp_abm_qdisc_offload_update(alink);
+
+	return 0;
+}