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, 642 insertions, 0 deletions

diff --git a/arch/mips/include/asm/octeon/cvmx-pko.h b/arch/mips/include/asm/octeon/cvmx-pko.h
new file mode 100644
index 000000000..5fec8476e
--- /dev/null
+++ b/arch/mips/include/asm/octeon/cvmx-pko.h
@@ -0,0 +1,642 @@
+/***********************license start***************
+ * Author: Cavium Networks
+ *
+ * Contact: support@caviumnetworks.com
+ * This file is part of the OCTEON SDK
+ *
+ * Copyright (c) 2003-2008 Cavium Networks
+ *
+ * This file is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License, Version 2, as
+ * published by the Free Software Foundation.
+ *
+ * This file is distributed in the hope that it will be useful, but
+ * AS-IS and WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, TITLE, or
+ * NONINFRINGEMENT.  See the GNU General Public License for more
+ * details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this file; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ * or visit http://www.gnu.org/licenses/.
+ *
+ * This file may also be available under a different license from Cavium.
+ * Contact Cavium Networks for more information
+ ***********************license end**************************************/
+
+/**
+ *
+ * Interface to the hardware Packet Output unit.
+ *
+ * Starting with SDK 1.7.0, the PKO output functions now support
+ * two types of locking. CVMX_PKO_LOCK_ATOMIC_TAG continues to
+ * function similarly to previous SDKs by using POW atomic tags
+ * to preserve ordering and exclusivity. As a new option, you
+ * can now pass CVMX_PKO_LOCK_CMD_QUEUE which uses a ll/sc
+ * memory based locking instead. This locking has the advantage
+ * of not affecting the tag state but doesn't preserve packet
+ * ordering. CVMX_PKO_LOCK_CMD_QUEUE is appropriate in most
+ * generic code while CVMX_PKO_LOCK_CMD_QUEUE should be used
+ * with hand tuned fast path code.
+ *
+ * Some of other SDK differences visible to the command queuing:
+ * - PKO indexes are no longer stored in the FAU. A large
+ *   percentage of the FAU register block used to be tied up
+ *   maintaining PKO queue pointers. These are now stored in a
+ *   global named block.
+ * - The PKO <b>use_locking</b> parameter can now have a global
+ *   effect. Since all application use the same named block,
+ *   queue locking correctly applies across all operating
+ *   systems when using CVMX_PKO_LOCK_CMD_QUEUE.
+ * - PKO 3 word commands are now supported. Use
+ *   cvmx_pko_send_packet_finish3().
+ *
+ */
+
+#ifndef __CVMX_PKO_H__
+#define __CVMX_PKO_H__
+
+#include <asm/octeon/cvmx-fpa.h>
+#include <asm/octeon/cvmx-pow.h>
+#include <asm/octeon/cvmx-cmd-queue.h>
+#include <asm/octeon/cvmx-pko-defs.h>
+
+/* Adjust the command buffer size by 1 word so that in the case of using only
+ * two word PKO commands no command words stradle buffers.  The useful values
+ * for this are 0 and 1. */
+#define CVMX_PKO_COMMAND_BUFFER_SIZE_ADJUST (1)
+
+#define CVMX_PKO_MAX_OUTPUT_QUEUES_STATIC 256
+#define CVMX_PKO_MAX_OUTPUT_QUEUES	((OCTEON_IS_MODEL(OCTEON_CN31XX) || \
+	OCTEON_IS_MODEL(OCTEON_CN3010) || OCTEON_IS_MODEL(OCTEON_CN3005) || \
+	OCTEON_IS_MODEL(OCTEON_CN50XX)) ? 32 : \
+		(OCTEON_IS_MODEL(OCTEON_CN58XX) || \
+		OCTEON_IS_MODEL(OCTEON_CN56XX)) ? 256 : 128)
+#define CVMX_PKO_NUM_OUTPUT_PORTS	40
+/* use this for queues that are not used */
+#define CVMX_PKO_MEM_QUEUE_PTRS_ILLEGAL_PID 63
+#define CVMX_PKO_QUEUE_STATIC_PRIORITY	9
+#define CVMX_PKO_ILLEGAL_QUEUE	0xFFFF
+#define CVMX_PKO_MAX_QUEUE_DEPTH 0
+
+typedef enum {
+	CVMX_PKO_SUCCESS,
+	CVMX_PKO_INVALID_PORT,
+	CVMX_PKO_INVALID_QUEUE,
+	CVMX_PKO_INVALID_PRIORITY,
+	CVMX_PKO_NO_MEMORY,
+	CVMX_PKO_PORT_ALREADY_SETUP,
+	CVMX_PKO_CMD_QUEUE_INIT_ERROR
+} cvmx_pko_status_t;
+
+/**
+ * This enumeration represents the differnet locking modes supported by PKO.
+ */
+typedef enum {
+	/*
+	 * PKO doesn't do any locking. It is the responsibility of the
+	 * application to make sure that no other core is accessing
+	 * the same queue at the same time
+	 */
+	CVMX_PKO_LOCK_NONE = 0,
+	/*
+	 * PKO performs an atomic tagswitch to insure exclusive access
+	 * to the output queue. This will maintain packet ordering on
+	 * output.
+	 */
+	CVMX_PKO_LOCK_ATOMIC_TAG = 1,
+	/*
+	 * PKO uses the common command queue locks to insure exclusive
+	 * access to the output queue. This is a memory based
+	 * ll/sc. This is the most portable locking mechanism.
+	 */
+	CVMX_PKO_LOCK_CMD_QUEUE = 2,
+} cvmx_pko_lock_t;
+
+typedef struct {
+	uint32_t packets;
+	uint64_t octets;
+	uint64_t doorbell;
+} cvmx_pko_port_status_t;
+
+/**
+ * This structure defines the address to use on a packet enqueue
+ */
+typedef union {
+	uint64_t u64;
+	struct {
+#ifdef __BIG_ENDIAN_BITFIELD
+		/* Must CVMX_IO_SEG */
+		uint64_t mem_space:2;
+		/* Must be zero */
+		uint64_t reserved:13;
+		/* Must be one */
+		uint64_t is_io:1;
+		/* The ID of the device on the non-coherent bus */
+		uint64_t did:8;
+		/* Must be zero */
+		uint64_t reserved2:4;
+		/* Must be zero */
+		uint64_t reserved3:18;
+		/*
+		 * The hardware likes to have the output port in
+		 * addition to the output queue,
+		 */
+		uint64_t port:6;
+		/*
+		 * The output queue to send the packet to (0-127 are
+		 * legal)
+		 */
+		uint64_t queue:9;
+		/* Must be zero */
+		uint64_t reserved4:3;
+#else
+	        uint64_t reserved4:3;
+	        uint64_t queue:9;
+	        uint64_t port:9;
+	        uint64_t reserved3:15;
+	        uint64_t reserved2:4;
+	        uint64_t did:8;
+	        uint64_t is_io:1;
+	        uint64_t reserved:13;
+	        uint64_t mem_space:2;
+#endif
+	} s;
+} cvmx_pko_doorbell_address_t;
+
+/**
+ * Structure of the first packet output command word.
+ */
+union cvmx_pko_command_word0 {
+	uint64_t u64;
+	struct {
+#ifdef __BIG_ENDIAN_BITFIELD
+		/*
+		 * The size of the reg1 operation - could be 8, 16,
+		 * 32, or 64 bits.
+		 */
+		uint64_t size1:2;
+		/*
+		 * The size of the reg0 operation - could be 8, 16,
+		 * 32, or 64 bits.
+		 */
+		uint64_t size0:2;
+		/*
+		 * If set, subtract 1, if clear, subtract packet
+		 * size.
+		 */
+		uint64_t subone1:1;
+		/*
+		 * The register, subtract will be done if reg1 is
+		 * non-zero.
+		 */
+		uint64_t reg1:11;
+		/* If set, subtract 1, if clear, subtract packet size */
+		uint64_t subone0:1;
+		/* The register, subtract will be done if reg0 is non-zero */
+		uint64_t reg0:11;
+		/*
+		 * When set, interpret segment pointer and segment
+		 * bytes in little endian order.
+		 */
+		uint64_t le:1;
+		/*
+		 * When set, packet data not allocated in L2 cache by
+		 * PKO.
+		 */
+		uint64_t n2:1;
+		/*
+		 * If set and rsp is set, word3 contains a pointer to
+		 * a work queue entry.
+		 */
+		uint64_t wqp:1;
+		/* If set, the hardware will send a response when done */
+		uint64_t rsp:1;
+		/*
+		 * If set, the supplied pkt_ptr is really a pointer to
+		 * a list of pkt_ptr's.
+		 */
+		uint64_t gather:1;
+		/*
+		 * If ipoffp1 is non zero, (ipoffp1-1) is the number
+		 * of bytes to IP header, and the hardware will
+		 * calculate and insert the UDP/TCP checksum.
+		 */
+		uint64_t ipoffp1:7;
+		/*
+		 * If set, ignore the I bit (force to zero) from all
+		 * pointer structures.
+		 */
+		uint64_t ignore_i:1;
+		/*
+		 * If clear, the hardware will attempt to free the
+		 * buffers containing the packet.
+		 */
+		uint64_t dontfree:1;
+		/*
+		 * The total number of segs in the packet, if gather
+		 * set, also gather list length.
+		 */
+		uint64_t segs:6;
+		/* Including L2, but no trailing CRC */
+		uint64_t total_bytes:16;
+#else
+	        uint64_t total_bytes:16;
+	        uint64_t segs:6;
+	        uint64_t dontfree:1;
+	        uint64_t ignore_i:1;
+	        uint64_t ipoffp1:7;
+	        uint64_t gather:1;
+	        uint64_t rsp:1;
+	        uint64_t wqp:1;
+	        uint64_t n2:1;
+	        uint64_t le:1;
+	        uint64_t reg0:11;
+	        uint64_t subone0:1;
+	        uint64_t reg1:11;
+	        uint64_t subone1:1;
+	        uint64_t size0:2;
+	        uint64_t size1:2;
+#endif
+	} s;
+};
+
+/* CSR typedefs have been moved to cvmx-csr-*.h */
+
+/**
+ * Definition of internal state for Packet output processing
+ */
+typedef struct {
+	/* ptr to start of buffer, offset kept in FAU reg */
+	uint64_t *start_ptr;
+} cvmx_pko_state_elem_t;
+
+/**
+ * Call before any other calls to initialize the packet
+ * output system.
+ */
+extern void cvmx_pko_initialize_global(void);
+
+/**
+ * Enables the packet output hardware. It must already be
+ * configured.
+ */
+extern void cvmx_pko_enable(void);
+
+/**
+ * Disables the packet output. Does not affect any configuration.
+ */
+extern void cvmx_pko_disable(void);
+
+/**
+ * Shutdown and free resources required by packet output.
+ */
+
+extern void cvmx_pko_shutdown(void);
+
+/**
+ * Configure a output port and the associated queues for use.
+ *
+ * @port:	Port to configure.
+ * @base_queue: First queue number to associate with this port.
+ * @num_queues: Number of queues t oassociate with this port
+ * @priority:	Array of priority levels for each queue. Values are
+ *		     allowed to be 1-8. A value of 8 get 8 times the traffic
+ *		     of a value of 1. There must be num_queues elements in the
+ *		     array.
+ */
+extern cvmx_pko_status_t cvmx_pko_config_port(uint64_t port,
+					      uint64_t base_queue,
+					      uint64_t num_queues,
+					      const uint64_t priority[]);
+
+/**
+ * Ring the packet output doorbell. This tells the packet
+ * output hardware that "len" command words have been added
+ * to its pending list.	 This command includes the required
+ * CVMX_SYNCWS before the doorbell ring.
+ *
+ * @port:   Port the packet is for
+ * @queue:  Queue the packet is for
+ * @len:    Length of the command in 64 bit words
+ */
+static inline void cvmx_pko_doorbell(uint64_t port, uint64_t queue,
+				     uint64_t len)
+{
+	cvmx_pko_doorbell_address_t ptr;
+
+	ptr.u64 = 0;
+	ptr.s.mem_space = CVMX_IO_SEG;
+	ptr.s.did = CVMX_OCT_DID_PKT_SEND;
+	ptr.s.is_io = 1;
+	ptr.s.port = port;
+	ptr.s.queue = queue;
+	/*
+	 * Need to make sure output queue data is in DRAM before
+	 * doorbell write.
+	 */
+	CVMX_SYNCWS;
+	cvmx_write_io(ptr.u64, len);
+}
+
+/**
+ * Prepare to send a packet.  This may initiate a tag switch to
+ * get exclusive access to the output queue structure, and
+ * performs other prep work for the packet send operation.
+ *
+ * cvmx_pko_send_packet_finish() MUST be called after this function is called,
+ * and must be called with the same port/queue/use_locking arguments.
+ *
+ * The use_locking parameter allows the caller to use three
+ * possible locking modes.
+ * - CVMX_PKO_LOCK_NONE
+ *	- PKO doesn't do any locking. It is the responsibility
+ *	    of the application to make sure that no other core
+ *	    is accessing the same queue at the same time.
+ * - CVMX_PKO_LOCK_ATOMIC_TAG
+ *	- PKO performs an atomic tagswitch to insure exclusive
+ *	    access to the output queue. This will maintain
+ *	    packet ordering on output.
+ * - CVMX_PKO_LOCK_CMD_QUEUE
+ *	- PKO uses the common command queue locks to insure
+ *	    exclusive access to the output queue. This is a
+ *	    memory based ll/sc. This is the most portable
+ *	    locking mechanism.
+ *
+ * NOTE: If atomic locking is used, the POW entry CANNOT be
+ * descheduled, as it does not contain a valid WQE pointer.
+ *
+ * @port:   Port to send it on
+ * @queue:  Queue to use
+ * @use_locking: CVMX_PKO_LOCK_NONE, CVMX_PKO_LOCK_ATOMIC_TAG, or
+ *		 CVMX_PKO_LOCK_CMD_QUEUE
+ */
+
+static inline void cvmx_pko_send_packet_prepare(uint64_t port, uint64_t queue,
+						cvmx_pko_lock_t use_locking)
+{
+	if (use_locking == CVMX_PKO_LOCK_ATOMIC_TAG) {
+		/*
+		 * Must do a full switch here to handle all cases.  We
+		 * use a fake WQE pointer, as the POW does not access
+		 * this memory.	 The WQE pointer and group are only
+		 * used if this work is descheduled, which is not
+		 * supported by the
+		 * cvmx_pko_send_packet_prepare/cvmx_pko_send_packet_finish
+		 * combination.	 Note that this is a special case in
+		 * which these fake values can be used - this is not a
+		 * general technique.
+		 */
+		uint32_t tag =
+		    CVMX_TAG_SW_BITS_INTERNAL << CVMX_TAG_SW_SHIFT |
+		    CVMX_TAG_SUBGROUP_PKO << CVMX_TAG_SUBGROUP_SHIFT |
+		    (CVMX_TAG_SUBGROUP_MASK & queue);
+		cvmx_pow_tag_sw_full((struct cvmx_wqe *) cvmx_phys_to_ptr(0x80), tag,
+				     CVMX_POW_TAG_TYPE_ATOMIC, 0);
+	}
+}
+
+/**
+ * Complete packet output. cvmx_pko_send_packet_prepare() must be
+ * called exactly once before this, and the same parameters must be
+ * passed to both cvmx_pko_send_packet_prepare() and
+ * cvmx_pko_send_packet_finish().
+ *
+ * @port:   Port to send it on
+ * @queue:  Queue to use
+ * @pko_command:
+ *		 PKO HW command word
+ * @packet: Packet to send
+ * @use_locking: CVMX_PKO_LOCK_NONE, CVMX_PKO_LOCK_ATOMIC_TAG, or
+ *		 CVMX_PKO_LOCK_CMD_QUEUE
+ *
+ * Returns: CVMX_PKO_SUCCESS on success, or error code on
+ * failure of output
+ */
+static inline cvmx_pko_status_t cvmx_pko_send_packet_finish(
+	uint64_t port,
+	uint64_t queue,
+	union cvmx_pko_command_word0 pko_command,
+	union cvmx_buf_ptr packet,
+	cvmx_pko_lock_t use_locking)
+{
+	cvmx_cmd_queue_result_t result;
+	if (use_locking == CVMX_PKO_LOCK_ATOMIC_TAG)
+		cvmx_pow_tag_sw_wait();
+	result = cvmx_cmd_queue_write2(CVMX_CMD_QUEUE_PKO(queue),
+				       (use_locking == CVMX_PKO_LOCK_CMD_QUEUE),
+				       pko_command.u64, packet.u64);
+	if (likely(result == CVMX_CMD_QUEUE_SUCCESS)) {
+		cvmx_pko_doorbell(port, queue, 2);
+		return CVMX_PKO_SUCCESS;
+	} else if ((result == CVMX_CMD_QUEUE_NO_MEMORY)
+		   || (result == CVMX_CMD_QUEUE_FULL)) {
+		return CVMX_PKO_NO_MEMORY;
+	} else {
+		return CVMX_PKO_INVALID_QUEUE;
+	}
+}
+
+/**
+ * Complete packet output. cvmx_pko_send_packet_prepare() must be
+ * called exactly once before this, and the same parameters must be
+ * passed to both cvmx_pko_send_packet_prepare() and
+ * cvmx_pko_send_packet_finish().
+ *
+ * @port:   Port to send it on
+ * @queue:  Queue to use
+ * @pko_command:
+ *		 PKO HW command word
+ * @packet: Packet to send
+ * @addr: Plysical address of a work queue entry or physical address
+ *	  to zero on complete.
+ * @use_locking: CVMX_PKO_LOCK_NONE, CVMX_PKO_LOCK_ATOMIC_TAG, or
+ *		 CVMX_PKO_LOCK_CMD_QUEUE
+ *
+ * Returns: CVMX_PKO_SUCCESS on success, or error code on
+ * failure of output
+ */
+static inline cvmx_pko_status_t cvmx_pko_send_packet_finish3(
+	uint64_t port,
+	uint64_t queue,
+	union cvmx_pko_command_word0 pko_command,
+	union cvmx_buf_ptr packet,
+	uint64_t addr,
+	cvmx_pko_lock_t use_locking)
+{
+	cvmx_cmd_queue_result_t result;
+	if (use_locking == CVMX_PKO_LOCK_ATOMIC_TAG)
+		cvmx_pow_tag_sw_wait();
+	result = cvmx_cmd_queue_write3(CVMX_CMD_QUEUE_PKO(queue),
+				       (use_locking == CVMX_PKO_LOCK_CMD_QUEUE),
+				       pko_command.u64, packet.u64, addr);
+	if (likely(result == CVMX_CMD_QUEUE_SUCCESS)) {
+		cvmx_pko_doorbell(port, queue, 3);
+		return CVMX_PKO_SUCCESS;
+	} else if ((result == CVMX_CMD_QUEUE_NO_MEMORY)
+		   || (result == CVMX_CMD_QUEUE_FULL)) {
+		return CVMX_PKO_NO_MEMORY;
+	} else {
+		return CVMX_PKO_INVALID_QUEUE;
+	}
+}
+
+/**
+ * Return the pko output queue associated with a port and a specific core.
+ * In normal mode (PKO lockless operation is disabled), the value returned
+ * is the base queue.
+ *
+ * @port:   Port number
+ * @core:   Core to get queue for
+ *
+ * Returns Core-specific output queue
+ */
+static inline int cvmx_pko_get_base_queue_per_core(int port, int core)
+{
+#ifndef CVMX_HELPER_PKO_MAX_PORTS_INTERFACE0
+#define CVMX_HELPER_PKO_MAX_PORTS_INTERFACE0 16
+#endif
+#ifndef CVMX_HELPER_PKO_MAX_PORTS_INTERFACE1
+#define CVMX_HELPER_PKO_MAX_PORTS_INTERFACE1 16
+#endif
+
+	if (port < CVMX_PKO_MAX_PORTS_INTERFACE0)
+		return port * CVMX_PKO_QUEUES_PER_PORT_INTERFACE0 + core;
+	else if (port >= 16 && port < 16 + CVMX_PKO_MAX_PORTS_INTERFACE1)
+		return CVMX_PKO_MAX_PORTS_INTERFACE0 *
+		    CVMX_PKO_QUEUES_PER_PORT_INTERFACE0 + (port -
+							   16) *
+		    CVMX_PKO_QUEUES_PER_PORT_INTERFACE1 + core;
+	else if ((port >= 32) && (port < 36))
+		return CVMX_PKO_MAX_PORTS_INTERFACE0 *
+		    CVMX_PKO_QUEUES_PER_PORT_INTERFACE0 +
+		    CVMX_PKO_MAX_PORTS_INTERFACE1 *
+		    CVMX_PKO_QUEUES_PER_PORT_INTERFACE1 + (port -
+							   32) *
+		    CVMX_PKO_QUEUES_PER_PORT_PCI;
+	else if ((port >= 36) && (port < 40))
+		return CVMX_PKO_MAX_PORTS_INTERFACE0 *
+		    CVMX_PKO_QUEUES_PER_PORT_INTERFACE0 +
+		    CVMX_PKO_MAX_PORTS_INTERFACE1 *
+		    CVMX_PKO_QUEUES_PER_PORT_INTERFACE1 +
+		    4 * CVMX_PKO_QUEUES_PER_PORT_PCI + (port -
+							36) *
+		    CVMX_PKO_QUEUES_PER_PORT_LOOP;
+	else
+		/* Given the limit on the number of ports we can map to
+		 * CVMX_MAX_OUTPUT_QUEUES_STATIC queues (currently 256,
+		 * divided among all cores), the remaining unmapped ports
+		 * are assigned an illegal queue number */
+		return CVMX_PKO_ILLEGAL_QUEUE;
+}
+
+/**
+ * For a given port number, return the base pko output queue
+ * for the port.
+ *
+ * @port:   Port number
+ * Returns Base output queue
+ */
+static inline int cvmx_pko_get_base_queue(int port)
+{
+	if (OCTEON_IS_MODEL(OCTEON_CN68XX))
+		return port;
+
+	return cvmx_pko_get_base_queue_per_core(port, 0);
+}
+
+/**
+ * For a given port number, return the number of pko output queues.
+ *
+ * @port:   Port number
+ * Returns Number of output queues
+ */
+static inline int cvmx_pko_get_num_queues(int port)
+{
+	if (port < 16)
+		return CVMX_PKO_QUEUES_PER_PORT_INTERFACE0;
+	else if (port < 32)
+		return CVMX_PKO_QUEUES_PER_PORT_INTERFACE1;
+	else if (port < 36)
+		return CVMX_PKO_QUEUES_PER_PORT_PCI;
+	else if (port < 40)
+		return CVMX_PKO_QUEUES_PER_PORT_LOOP;
+	else
+		return 0;
+}
+
+/**
+ * Get the status counters for a port.
+ *
+ * @port_num: Port number to get statistics for.
+ * @clear:    Set to 1 to clear the counters after they are read
+ * @status:   Where to put the results.
+ */
+static inline void cvmx_pko_get_port_status(uint64_t port_num, uint64_t clear,
+					    cvmx_pko_port_status_t *status)
+{
+	union cvmx_pko_reg_read_idx pko_reg_read_idx;
+	union cvmx_pko_mem_count0 pko_mem_count0;
+	union cvmx_pko_mem_count1 pko_mem_count1;
+
+	pko_reg_read_idx.u64 = 0;
+	pko_reg_read_idx.s.index = port_num;
+	cvmx_write_csr(CVMX_PKO_REG_READ_IDX, pko_reg_read_idx.u64);
+
+	pko_mem_count0.u64 = cvmx_read_csr(CVMX_PKO_MEM_COUNT0);
+	status->packets = pko_mem_count0.s.count;
+	if (clear) {
+		pko_mem_count0.s.count = port_num;
+		cvmx_write_csr(CVMX_PKO_MEM_COUNT0, pko_mem_count0.u64);
+	}
+
+	pko_mem_count1.u64 = cvmx_read_csr(CVMX_PKO_MEM_COUNT1);
+	status->octets = pko_mem_count1.s.count;
+	if (clear) {
+		pko_mem_count1.s.count = port_num;
+		cvmx_write_csr(CVMX_PKO_MEM_COUNT1, pko_mem_count1.u64);
+	}
+
+	if (OCTEON_IS_MODEL(OCTEON_CN3XXX)) {
+		union cvmx_pko_mem_debug9 debug9;
+		pko_reg_read_idx.s.index = cvmx_pko_get_base_queue(port_num);
+		cvmx_write_csr(CVMX_PKO_REG_READ_IDX, pko_reg_read_idx.u64);
+		debug9.u64 = cvmx_read_csr(CVMX_PKO_MEM_DEBUG9);
+		status->doorbell = debug9.cn38xx.doorbell;
+	} else {
+		union cvmx_pko_mem_debug8 debug8;
+		pko_reg_read_idx.s.index = cvmx_pko_get_base_queue(port_num);
+		cvmx_write_csr(CVMX_PKO_REG_READ_IDX, pko_reg_read_idx.u64);
+		debug8.u64 = cvmx_read_csr(CVMX_PKO_MEM_DEBUG8);
+		status->doorbell = debug8.cn50xx.doorbell;
+	}
+}
+
+/**
+ * Rate limit a PKO port to a max packets/sec. This function is only
+ * supported on CN57XX, CN56XX, CN55XX, and CN54XX.
+ *
+ * @port:      Port to rate limit
+ * @packets_s: Maximum packet/sec
+ * @burst:     Maximum number of packets to burst in a row before rate
+ *		    limiting cuts in.
+ *
+ * Returns Zero on success, negative on failure
+ */
+extern int cvmx_pko_rate_limit_packets(int port, int packets_s, int burst);
+
+/**
+ * Rate limit a PKO port to a max bits/sec. This function is only
+ * supported on CN57XX, CN56XX, CN55XX, and CN54XX.
+ *
+ * @port:   Port to rate limit
+ * @bits_s: PKO rate limit in bits/sec
+ * @burst:  Maximum number of bits to burst before rate
+ *		 limiting cuts in.
+ *
+ * Returns Zero on success, negative on failure
+ */
+extern int cvmx_pko_rate_limit_bits(int port, uint64_t bits_s, int burst);
+
+#endif /* __CVMX_PKO_H__ */