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, 629 insertions, 0 deletions

diff --git a/drivers/soc/fsl/qbman/qman_test_stash.c b/drivers/soc/fsl/qbman/qman_test_stash.c
new file mode 100644
index 000000000..b7e8e5ec8
--- /dev/null
+++ b/drivers/soc/fsl/qbman/qman_test_stash.c
@@ -0,0 +1,629 @@
+/* Copyright 2009 - 2016 Freescale Semiconductor, Inc.
+ *
+ * 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.
+ *     * Neither the name of Freescale Semiconductor nor the
+ *	 names of its contributors may be used to endorse or promote products
+ *	 derived from this software without specific prior written permission.
+ *
+ * ALTERNATIVELY, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") as published by the Free Software
+ * Foundation, either version 2 of that License or (at your option) any
+ * later version.
+ *
+ * THIS SOFTWARE IS PROVIDED BY Freescale Semiconductor ``AS IS'' AND ANY
+ * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
+ * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+ * DISCLAIMED. IN NO EVENT SHALL Freescale Semiconductor BE LIABLE FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
+ * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+ * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+ * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include "qman_test.h"
+
+#include <linux/dma-mapping.h>
+#include <linux/delay.h>
+
+/*
+ * Algorithm:
+ *
+ * Each cpu will have HP_PER_CPU "handlers" set up, each of which incorporates
+ * an rx/tx pair of FQ objects (both of which are stashed on dequeue). The
+ * organisation of FQIDs is such that the HP_PER_CPU*NUM_CPUS handlers will
+ * shuttle a "hot potato" frame around them such that every forwarding action
+ * moves it from one cpu to another. (The use of more than one handler per cpu
+ * is to allow enough handlers/FQs to truly test the significance of caching -
+ * ie. when cache-expiries are occurring.)
+ *
+ * The "hot potato" frame content will be HP_NUM_WORDS*4 bytes in size, and the
+ * first and last words of the frame data will undergo a transformation step on
+ * each forwarding action. To achieve this, each handler will be assigned a
+ * 32-bit "mixer", that is produced using a 32-bit LFSR. When a frame is
+ * received by a handler, the mixer of the expected sender is XOR'd into all
+ * words of the entire frame, which is then validated against the original
+ * values. Then, before forwarding, the entire frame is XOR'd with the mixer of
+ * the current handler. Apart from validating that the frame is taking the
+ * expected path, this also provides some quasi-realistic overheads to each
+ * forwarding action - dereferencing *all* the frame data, computation, and
+ * conditional branching. There is a "special" handler designated to act as the
+ * instigator of the test by creating an enqueuing the "hot potato" frame, and
+ * to determine when the test has completed by counting HP_LOOPS iterations.
+ *
+ * Init phases:
+ *
+ * 1. prepare each cpu's 'hp_cpu' struct using on_each_cpu(,,1) and link them
+ *    into 'hp_cpu_list'. Specifically, set processor_id, allocate HP_PER_CPU
+ *    handlers and link-list them (but do no other handler setup).
+ *
+ * 2. scan over 'hp_cpu_list' HP_PER_CPU times, the first time sets each
+ *    hp_cpu's 'iterator' to point to its first handler. With each loop,
+ *    allocate rx/tx FQIDs and mixer values to the hp_cpu's iterator handler
+ *    and advance the iterator for the next loop. This includes a final fixup,
+ *    which connects the last handler to the first (and which is why phase 2
+ *    and 3 are separate).
+ *
+ * 3. scan over 'hp_cpu_list' HP_PER_CPU times, the first time sets each
+ *    hp_cpu's 'iterator' to point to its first handler. With each loop,
+ *    initialise FQ objects and advance the iterator for the next loop.
+ *    Moreover, do this initialisation on the cpu it applies to so that Rx FQ
+ *    initialisation targets the correct cpu.
+ */
+
+/*
+ * helper to run something on all cpus (can't use on_each_cpu(), as that invokes
+ * the fn from irq context, which is too restrictive).
+ */
+struct bstrap {
+	int (*fn)(void);
+	atomic_t started;
+};
+static int bstrap_fn(void *bs)
+{
+	struct bstrap *bstrap = bs;
+	int err;
+
+	atomic_inc(&bstrap->started);
+	err = bstrap->fn();
+	if (err)
+		return err;
+	while (!kthread_should_stop())
+		msleep(20);
+	return 0;
+}
+static int on_all_cpus(int (*fn)(void))
+{
+	int cpu;
+
+	for_each_cpu(cpu, cpu_online_mask) {
+		struct bstrap bstrap = {
+			.fn = fn,
+			.started = ATOMIC_INIT(0)
+		};
+		struct task_struct *k = kthread_create(bstrap_fn, &bstrap,
+			"hotpotato%d", cpu);
+		int ret;
+
+		if (IS_ERR(k))
+			return -ENOMEM;
+		kthread_bind(k, cpu);
+		wake_up_process(k);
+		/*
+		 * If we call kthread_stop() before the "wake up" has had an
+		 * effect, then the thread may exit with -EINTR without ever
+		 * running the function. So poll until it's started before
+		 * requesting it to stop.
+		 */
+		while (!atomic_read(&bstrap.started))
+			msleep(20);
+		ret = kthread_stop(k);
+		if (ret)
+			return ret;
+	}
+	return 0;
+}
+
+struct hp_handler {
+
+	/* The following data is stashed when 'rx' is dequeued; */
+	/* -------------- */
+	/* The Rx FQ, dequeues of which will stash the entire hp_handler */
+	struct qman_fq rx;
+	/* The Tx FQ we should forward to */
+	struct qman_fq tx;
+	/* The value we XOR post-dequeue, prior to validating */
+	u32 rx_mixer;
+	/* The value we XOR pre-enqueue, after validating */
+	u32 tx_mixer;
+	/* what the hotpotato address should be on dequeue */
+	dma_addr_t addr;
+	u32 *frame_ptr;
+
+	/* The following data isn't (necessarily) stashed on dequeue; */
+	/* -------------- */
+	u32 fqid_rx, fqid_tx;
+	/* list node for linking us into 'hp_cpu' */
+	struct list_head node;
+	/* Just to check ... */
+	unsigned int processor_id;
+} ____cacheline_aligned;
+
+struct hp_cpu {
+	/* identify the cpu we run on; */
+	unsigned int processor_id;
+	/* root node for the per-cpu list of handlers */
+	struct list_head handlers;
+	/* list node for linking us into 'hp_cpu_list' */
+	struct list_head node;
+	/*
+	 * when repeatedly scanning 'hp_list', each time linking the n'th
+	 * handlers together, this is used as per-cpu iterator state
+	 */
+	struct hp_handler *iterator;
+};
+
+/* Each cpu has one of these */
+static DEFINE_PER_CPU(struct hp_cpu, hp_cpus);
+
+/* links together the hp_cpu structs, in first-come first-serve order. */
+static LIST_HEAD(hp_cpu_list);
+static DEFINE_SPINLOCK(hp_lock);
+
+static unsigned int hp_cpu_list_length;
+
+/* the "special" handler, that starts and terminates the test. */
+static struct hp_handler *special_handler;
+static int loop_counter;
+
+/* handlers are allocated out of this, so they're properly aligned. */
+static struct kmem_cache *hp_handler_slab;
+
+/* this is the frame data */
+static void *__frame_ptr;
+static u32 *frame_ptr;
+static dma_addr_t frame_dma;
+
+/* needed for dma_map*() */
+static const struct qm_portal_config *pcfg;
+
+/* the main function waits on this */
+static DECLARE_WAIT_QUEUE_HEAD(queue);
+
+#define HP_PER_CPU	2
+#define HP_LOOPS	8
+/* 80 bytes, like a small ethernet frame, and bleeds into a second cacheline */
+#define HP_NUM_WORDS	80
+/* First word of the LFSR-based frame data */
+#define HP_FIRST_WORD	0xabbaf00d
+
+static inline u32 do_lfsr(u32 prev)
+{
+	return (prev >> 1) ^ (-(prev & 1u) & 0xd0000001u);
+}
+
+static int allocate_frame_data(void)
+{
+	u32 lfsr = HP_FIRST_WORD;
+	int loop;
+
+	if (!qman_dma_portal) {
+		pr_crit("portal not available\n");
+		return -EIO;
+	}
+
+	pcfg = qman_get_qm_portal_config(qman_dma_portal);
+
+	__frame_ptr = kmalloc(4 * HP_NUM_WORDS, GFP_KERNEL);
+	if (!__frame_ptr)
+		return -ENOMEM;
+
+	frame_ptr = PTR_ALIGN(__frame_ptr, 64);
+	for (loop = 0; loop < HP_NUM_WORDS; loop++) {
+		frame_ptr[loop] = lfsr;
+		lfsr = do_lfsr(lfsr);
+	}
+
+	frame_dma = dma_map_single(pcfg->dev, frame_ptr, 4 * HP_NUM_WORDS,
+				   DMA_BIDIRECTIONAL);
+	if (dma_mapping_error(pcfg->dev, frame_dma)) {
+		pr_crit("dma mapping failure\n");
+		kfree(__frame_ptr);
+		return -EIO;
+	}
+
+	return 0;
+}
+
+static void deallocate_frame_data(void)
+{
+	dma_unmap_single(pcfg->dev, frame_dma, 4 * HP_NUM_WORDS,
+			 DMA_BIDIRECTIONAL);
+	kfree(__frame_ptr);
+}
+
+static inline int process_frame_data(struct hp_handler *handler,
+				     const struct qm_fd *fd)
+{
+	u32 *p = handler->frame_ptr;
+	u32 lfsr = HP_FIRST_WORD;
+	int loop;
+
+	if (qm_fd_addr_get64(fd) != handler->addr) {
+		pr_crit("bad frame address, [%llX != %llX]\n",
+			qm_fd_addr_get64(fd), handler->addr);
+		return -EIO;
+	}
+	for (loop = 0; loop < HP_NUM_WORDS; loop++, p++) {
+		*p ^= handler->rx_mixer;
+		if (*p != lfsr) {
+			pr_crit("corrupt frame data");
+			return -EIO;
+		}
+		*p ^= handler->tx_mixer;
+		lfsr = do_lfsr(lfsr);
+	}
+	return 0;
+}
+
+static enum qman_cb_dqrr_result normal_dqrr(struct qman_portal *portal,
+					    struct qman_fq *fq,
+					    const struct qm_dqrr_entry *dqrr,
+					    bool sched_napi)
+{
+	struct hp_handler *handler = (struct hp_handler *)fq;
+
+	if (process_frame_data(handler, &dqrr->fd)) {
+		WARN_ON(1);
+		goto skip;
+	}
+	if (qman_enqueue(&handler->tx, &dqrr->fd)) {
+		pr_crit("qman_enqueue() failed");
+		WARN_ON(1);
+	}
+skip:
+	return qman_cb_dqrr_consume;
+}
+
+static enum qman_cb_dqrr_result special_dqrr(struct qman_portal *portal,
+					     struct qman_fq *fq,
+					     const struct qm_dqrr_entry *dqrr,
+					     bool sched_napi)
+{
+	struct hp_handler *handler = (struct hp_handler *)fq;
+
+	process_frame_data(handler, &dqrr->fd);
+	if (++loop_counter < HP_LOOPS) {
+		if (qman_enqueue(&handler->tx, &dqrr->fd)) {
+			pr_crit("qman_enqueue() failed");
+			WARN_ON(1);
+			goto skip;
+		}
+	} else {
+		pr_info("Received final (%dth) frame\n", loop_counter);
+		wake_up(&queue);
+	}
+skip:
+	return qman_cb_dqrr_consume;
+}
+
+static int create_per_cpu_handlers(void)
+{
+	struct hp_handler *handler;
+	int loop;
+	struct hp_cpu *hp_cpu = this_cpu_ptr(&hp_cpus);
+
+	hp_cpu->processor_id = smp_processor_id();
+	spin_lock(&hp_lock);
+	list_add_tail(&hp_cpu->node, &hp_cpu_list);
+	hp_cpu_list_length++;
+	spin_unlock(&hp_lock);
+	INIT_LIST_HEAD(&hp_cpu->handlers);
+	for (loop = 0; loop < HP_PER_CPU; loop++) {
+		handler = kmem_cache_alloc(hp_handler_slab, GFP_KERNEL);
+		if (!handler) {
+			pr_crit("kmem_cache_alloc() failed");
+			WARN_ON(1);
+			return -EIO;
+		}
+		handler->processor_id = hp_cpu->processor_id;
+		handler->addr = frame_dma;
+		handler->frame_ptr = frame_ptr;
+		list_add_tail(&handler->node, &hp_cpu->handlers);
+	}
+	return 0;
+}
+
+static int destroy_per_cpu_handlers(void)
+{
+	struct list_head *loop, *tmp;
+	struct hp_cpu *hp_cpu = this_cpu_ptr(&hp_cpus);
+
+	spin_lock(&hp_lock);
+	list_del(&hp_cpu->node);
+	spin_unlock(&hp_lock);
+	list_for_each_safe(loop, tmp, &hp_cpu->handlers) {
+		u32 flags = 0;
+		struct hp_handler *handler = list_entry(loop, struct hp_handler,
+							node);
+		if (qman_retire_fq(&handler->rx, &flags) ||
+		    (flags & QMAN_FQ_STATE_BLOCKOOS)) {
+			pr_crit("qman_retire_fq(rx) failed, flags: %x", flags);
+			WARN_ON(1);
+			return -EIO;
+		}
+		if (qman_oos_fq(&handler->rx)) {
+			pr_crit("qman_oos_fq(rx) failed");
+			WARN_ON(1);
+			return -EIO;
+		}
+		qman_destroy_fq(&handler->rx);
+		qman_destroy_fq(&handler->tx);
+		qman_release_fqid(handler->fqid_rx);
+		list_del(&handler->node);
+		kmem_cache_free(hp_handler_slab, handler);
+	}
+	return 0;
+}
+
+static inline u8 num_cachelines(u32 offset)
+{
+	u8 res = (offset + (L1_CACHE_BYTES - 1))
+			 / (L1_CACHE_BYTES);
+	if (res > 3)
+		return 3;
+	return res;
+}
+#define STASH_DATA_CL \
+	num_cachelines(HP_NUM_WORDS * 4)
+#define STASH_CTX_CL \
+	num_cachelines(offsetof(struct hp_handler, fqid_rx))
+
+static int init_handler(void *h)
+{
+	struct qm_mcc_initfq opts;
+	struct hp_handler *handler = h;
+	int err;
+
+	if (handler->processor_id != smp_processor_id()) {
+		err = -EIO;
+		goto failed;
+	}
+	/* Set up rx */
+	memset(&handler->rx, 0, sizeof(handler->rx));
+	if (handler == special_handler)
+		handler->rx.cb.dqrr = special_dqrr;
+	else
+		handler->rx.cb.dqrr = normal_dqrr;
+	err = qman_create_fq(handler->fqid_rx, 0, &handler->rx);
+	if (err) {
+		pr_crit("qman_create_fq(rx) failed");
+		goto failed;
+	}
+	memset(&opts, 0, sizeof(opts));
+	opts.we_mask = cpu_to_be16(QM_INITFQ_WE_FQCTRL |
+				   QM_INITFQ_WE_CONTEXTA);
+	opts.fqd.fq_ctrl = cpu_to_be16(QM_FQCTRL_CTXASTASHING);
+	qm_fqd_set_stashing(&opts.fqd, 0, STASH_DATA_CL, STASH_CTX_CL);
+	err = qman_init_fq(&handler->rx, QMAN_INITFQ_FLAG_SCHED |
+			   QMAN_INITFQ_FLAG_LOCAL, &opts);
+	if (err) {
+		pr_crit("qman_init_fq(rx) failed");
+		goto failed;
+	}
+	/* Set up tx */
+	memset(&handler->tx, 0, sizeof(handler->tx));
+	err = qman_create_fq(handler->fqid_tx, QMAN_FQ_FLAG_NO_MODIFY,
+			     &handler->tx);
+	if (err) {
+		pr_crit("qman_create_fq(tx) failed");
+		goto failed;
+	}
+
+	return 0;
+failed:
+	return err;
+}
+
+static void init_handler_cb(void *h)
+{
+	if (init_handler(h))
+		WARN_ON(1);
+}
+
+static int init_phase2(void)
+{
+	int loop;
+	u32 fqid = 0;
+	u32 lfsr = 0xdeadbeef;
+	struct hp_cpu *hp_cpu;
+	struct hp_handler *handler;
+
+	for (loop = 0; loop < HP_PER_CPU; loop++) {
+		list_for_each_entry(hp_cpu, &hp_cpu_list, node) {
+			int err;
+
+			if (!loop)
+				hp_cpu->iterator = list_first_entry(
+						&hp_cpu->handlers,
+						struct hp_handler, node);
+			else
+				hp_cpu->iterator = list_entry(
+						hp_cpu->iterator->node.next,
+						struct hp_handler, node);
+			/* Rx FQID is the previous handler's Tx FQID */
+			hp_cpu->iterator->fqid_rx = fqid;
+			/* Allocate new FQID for Tx */
+			err = qman_alloc_fqid(&fqid);
+			if (err) {
+				pr_crit("qman_alloc_fqid() failed");
+				return err;
+			}
+			hp_cpu->iterator->fqid_tx = fqid;
+			/* Rx mixer is the previous handler's Tx mixer */
+			hp_cpu->iterator->rx_mixer = lfsr;
+			/* Get new mixer for Tx */
+			lfsr = do_lfsr(lfsr);
+			hp_cpu->iterator->tx_mixer = lfsr;
+		}
+	}
+	/* Fix up the first handler (fqid_rx==0, rx_mixer=0xdeadbeef) */
+	hp_cpu = list_first_entry(&hp_cpu_list, struct hp_cpu, node);
+	handler = list_first_entry(&hp_cpu->handlers, struct hp_handler, node);
+	if (handler->fqid_rx != 0 || handler->rx_mixer != 0xdeadbeef)
+		return 1;
+	handler->fqid_rx = fqid;
+	handler->rx_mixer = lfsr;
+	/* and tag it as our "special" handler */
+	special_handler = handler;
+	return 0;
+}
+
+static int init_phase3(void)
+{
+	int loop, err;
+	struct hp_cpu *hp_cpu;
+
+	for (loop = 0; loop < HP_PER_CPU; loop++) {
+		list_for_each_entry(hp_cpu, &hp_cpu_list, node) {
+			if (!loop)
+				hp_cpu->iterator = list_first_entry(
+						&hp_cpu->handlers,
+						struct hp_handler, node);
+			else
+				hp_cpu->iterator = list_entry(
+						hp_cpu->iterator->node.next,
+						struct hp_handler, node);
+			preempt_disable();
+			if (hp_cpu->processor_id == smp_processor_id()) {
+				err = init_handler(hp_cpu->iterator);
+				if (err)
+					return err;
+			} else {
+				smp_call_function_single(hp_cpu->processor_id,
+					init_handler_cb, hp_cpu->iterator, 1);
+			}
+			preempt_enable();
+		}
+	}
+	return 0;
+}
+
+static int send_first_frame(void *ignore)
+{
+	u32 *p = special_handler->frame_ptr;
+	u32 lfsr = HP_FIRST_WORD;
+	int loop, err;
+	struct qm_fd fd;
+
+	if (special_handler->processor_id != smp_processor_id()) {
+		err = -EIO;
+		goto failed;
+	}
+	memset(&fd, 0, sizeof(fd));
+	qm_fd_addr_set64(&fd, special_handler->addr);
+	qm_fd_set_contig_big(&fd, HP_NUM_WORDS * 4);
+	for (loop = 0; loop < HP_NUM_WORDS; loop++, p++) {
+		if (*p != lfsr) {
+			err = -EIO;
+			pr_crit("corrupt frame data");
+			goto failed;
+		}
+		*p ^= special_handler->tx_mixer;
+		lfsr = do_lfsr(lfsr);
+	}
+	pr_info("Sending first frame\n");
+	err = qman_enqueue(&special_handler->tx, &fd);
+	if (err) {
+		pr_crit("qman_enqueue() failed");
+		goto failed;
+	}
+
+	return 0;
+failed:
+	return err;
+}
+
+static void send_first_frame_cb(void *ignore)
+{
+	if (send_first_frame(NULL))
+		WARN_ON(1);
+}
+
+int qman_test_stash(void)
+{
+	int err;
+
+	if (cpumask_weight(cpu_online_mask) < 2) {
+		pr_info("%s(): skip - only 1 CPU\n", __func__);
+		return 0;
+	}
+
+	pr_info("%s(): Starting\n", __func__);
+
+	hp_cpu_list_length = 0;
+	loop_counter = 0;
+	hp_handler_slab = kmem_cache_create("hp_handler_slab",
+			sizeof(struct hp_handler), L1_CACHE_BYTES,
+			SLAB_HWCACHE_ALIGN, NULL);
+	if (!hp_handler_slab) {
+		err = -EIO;
+		pr_crit("kmem_cache_create() failed");
+		goto failed;
+	}
+
+	err = allocate_frame_data();
+	if (err)
+		goto failed;
+
+	/* Init phase 1 */
+	pr_info("Creating %d handlers per cpu...\n", HP_PER_CPU);
+	if (on_all_cpus(create_per_cpu_handlers)) {
+		err = -EIO;
+		pr_crit("on_each_cpu() failed");
+		goto failed;
+	}
+	pr_info("Number of cpus: %d, total of %d handlers\n",
+		hp_cpu_list_length, hp_cpu_list_length * HP_PER_CPU);
+
+	err = init_phase2();
+	if (err)
+		goto failed;
+
+	err = init_phase3();
+	if (err)
+		goto failed;
+
+	preempt_disable();
+	if (special_handler->processor_id == smp_processor_id()) {
+		err = send_first_frame(NULL);
+		if (err)
+			goto failed;
+	} else {
+		smp_call_function_single(special_handler->processor_id,
+					 send_first_frame_cb, NULL, 1);
+	}
+	preempt_enable();
+
+	wait_event(queue, loop_counter == HP_LOOPS);
+	deallocate_frame_data();
+	if (on_all_cpus(destroy_per_cpu_handlers)) {
+		err = -EIO;
+		pr_crit("on_each_cpu() failed");
+		goto failed;
+	}
+	kmem_cache_destroy(hp_handler_slab);
+	pr_info("%s(): Finished\n", __func__);
+
+	return 0;
+failed:
+	WARN_ON(1);
+	return err;
+}