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, 823 insertions, 0 deletions

diff --git a/drivers/net/ethernet/marvell/prestera/prestera_rxtx.c b/drivers/net/ethernet/marvell/prestera/prestera_rxtx.c
new file mode 100644
index 000000000..9277a8fd1
--- /dev/null
+++ b/drivers/net/ethernet/marvell/prestera/prestera_rxtx.c
@@ -0,0 +1,823 @@
+// SPDX-License-Identifier: BSD-3-Clause OR GPL-2.0
+/* Copyright (c) 2019-2020 Marvell International Ltd. All rights reserved */
+
+#include <linux/bitfield.h>
+#include <linux/dmapool.h>
+#include <linux/etherdevice.h>
+#include <linux/if_vlan.h>
+#include <linux/of_address.h>
+#include <linux/of_device.h>
+#include <linux/of.h>
+#include <linux/platform_device.h>
+
+#include "prestera_dsa.h"
+#include "prestera.h"
+#include "prestera_hw.h"
+#include "prestera_rxtx.h"
+#include "prestera_devlink.h"
+
+#define PRESTERA_SDMA_WAIT_MUL		10
+
+struct prestera_sdma_desc {
+	__le32 word1;
+	__le32 word2;
+	__le32 buff;
+	__le32 next;
+} __packed __aligned(16);
+
+#define PRESTERA_SDMA_BUFF_SIZE_MAX	1544
+
+#define PRESTERA_SDMA_RX_DESC_PKT_LEN(desc) \
+	((le32_to_cpu((desc)->word2) >> 16) & GENMASK(13, 0))
+
+#define PRESTERA_SDMA_RX_DESC_OWNER(desc) \
+	((le32_to_cpu((desc)->word1) & BIT(31)) >> 31)
+
+#define PRESTERA_SDMA_RX_DESC_IS_RCVD(desc) \
+	(PRESTERA_SDMA_RX_DESC_OWNER(desc) == PRESTERA_SDMA_RX_DESC_CPU_OWN)
+
+#define PRESTERA_SDMA_RX_DESC_CPU_OWN	0
+#define PRESTERA_SDMA_RX_DESC_DMA_OWN	1
+
+#define PRESTERA_SDMA_RX_QUEUE_NUM	8
+
+#define PRESTERA_SDMA_RX_DESC_PER_Q	1000
+
+#define PRESTERA_SDMA_TX_DESC_PER_Q	1000
+#define PRESTERA_SDMA_TX_MAX_BURST	64
+
+#define PRESTERA_SDMA_TX_DESC_OWNER(desc) \
+	((le32_to_cpu((desc)->word1) & BIT(31)) >> 31)
+
+#define PRESTERA_SDMA_TX_DESC_CPU_OWN	0
+#define PRESTERA_SDMA_TX_DESC_DMA_OWN	1U
+
+#define PRESTERA_SDMA_TX_DESC_IS_SENT(desc) \
+	(PRESTERA_SDMA_TX_DESC_OWNER(desc) == PRESTERA_SDMA_TX_DESC_CPU_OWN)
+
+#define PRESTERA_SDMA_TX_DESC_LAST	BIT(20)
+#define PRESTERA_SDMA_TX_DESC_FIRST	BIT(21)
+#define PRESTERA_SDMA_TX_DESC_CALC_CRC	BIT(12)
+
+#define PRESTERA_SDMA_TX_DESC_SINGLE	\
+	(PRESTERA_SDMA_TX_DESC_FIRST | PRESTERA_SDMA_TX_DESC_LAST)
+
+#define PRESTERA_SDMA_TX_DESC_INIT	\
+	(PRESTERA_SDMA_TX_DESC_SINGLE | PRESTERA_SDMA_TX_DESC_CALC_CRC)
+
+#define PRESTERA_SDMA_RX_INTR_MASK_REG		0x2814
+#define PRESTERA_SDMA_RX_QUEUE_STATUS_REG	0x2680
+#define PRESTERA_SDMA_RX_QUEUE_DESC_REG(n)	(0x260C + (n) * 16)
+
+#define PRESTERA_SDMA_TX_QUEUE_DESC_REG		0x26C0
+#define PRESTERA_SDMA_TX_QUEUE_START_REG	0x2868
+
+struct prestera_sdma_buf {
+	struct prestera_sdma_desc *desc;
+	dma_addr_t desc_dma;
+	struct sk_buff *skb;
+	dma_addr_t buf_dma;
+	bool is_used;
+};
+
+struct prestera_rx_ring {
+	struct prestera_sdma_buf *bufs;
+	int next_rx;
+};
+
+struct prestera_tx_ring {
+	struct prestera_sdma_buf *bufs;
+	int next_tx;
+	int max_burst;
+	int burst;
+};
+
+struct prestera_sdma {
+	struct prestera_rx_ring rx_ring[PRESTERA_SDMA_RX_QUEUE_NUM];
+	struct prestera_tx_ring tx_ring;
+	struct prestera_switch *sw;
+	struct dma_pool *desc_pool;
+	struct work_struct tx_work;
+	struct napi_struct rx_napi;
+	struct net_device napi_dev;
+	u32 map_addr;
+	u64 dma_mask;
+	/* protect SDMA with concurrent access from multiple CPUs */
+	spinlock_t tx_lock;
+};
+
+struct prestera_rxtx {
+	struct prestera_sdma sdma;
+};
+
+static int prestera_sdma_buf_init(struct prestera_sdma *sdma,
+				  struct prestera_sdma_buf *buf)
+{
+	struct prestera_sdma_desc *desc;
+	dma_addr_t dma;
+
+	desc = dma_pool_alloc(sdma->desc_pool, GFP_DMA | GFP_KERNEL, &dma);
+	if (!desc)
+		return -ENOMEM;
+
+	buf->buf_dma = DMA_MAPPING_ERROR;
+	buf->desc_dma = dma;
+	buf->desc = desc;
+	buf->skb = NULL;
+
+	return 0;
+}
+
+static u32 prestera_sdma_map(struct prestera_sdma *sdma, dma_addr_t pa)
+{
+	return sdma->map_addr + pa;
+}
+
+static void prestera_sdma_rx_desc_init(struct prestera_sdma *sdma,
+				       struct prestera_sdma_desc *desc,
+				       dma_addr_t buf)
+{
+	u32 word = le32_to_cpu(desc->word2);
+
+	u32p_replace_bits(&word, PRESTERA_SDMA_BUFF_SIZE_MAX, GENMASK(15, 0));
+	desc->word2 = cpu_to_le32(word);
+
+	desc->buff = cpu_to_le32(prestera_sdma_map(sdma, buf));
+
+	/* make sure buffer is set before reset the descriptor */
+	wmb();
+
+	desc->word1 = cpu_to_le32(0xA0000000);
+}
+
+static void prestera_sdma_rx_desc_set_next(struct prestera_sdma *sdma,
+					   struct prestera_sdma_desc *desc,
+					   dma_addr_t next)
+{
+	desc->next = cpu_to_le32(prestera_sdma_map(sdma, next));
+}
+
+static int prestera_sdma_rx_skb_alloc(struct prestera_sdma *sdma,
+				      struct prestera_sdma_buf *buf)
+{
+	struct device *dev = sdma->sw->dev->dev;
+	struct sk_buff *skb;
+	dma_addr_t dma;
+
+	skb = alloc_skb(PRESTERA_SDMA_BUFF_SIZE_MAX, GFP_DMA | GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+
+	dma = dma_map_single(dev, skb->data, skb->len, DMA_FROM_DEVICE);
+	if (dma_mapping_error(dev, dma))
+		goto err_dma_map;
+
+	if (buf->skb)
+		dma_unmap_single(dev, buf->buf_dma, buf->skb->len,
+				 DMA_FROM_DEVICE);
+
+	buf->buf_dma = dma;
+	buf->skb = skb;
+
+	return 0;
+
+err_dma_map:
+	kfree_skb(skb);
+
+	return -ENOMEM;
+}
+
+static struct sk_buff *prestera_sdma_rx_skb_get(struct prestera_sdma *sdma,
+						struct prestera_sdma_buf *buf)
+{
+	dma_addr_t buf_dma = buf->buf_dma;
+	struct sk_buff *skb = buf->skb;
+	u32 len = skb->len;
+	int err;
+
+	err = prestera_sdma_rx_skb_alloc(sdma, buf);
+	if (err) {
+		buf->buf_dma = buf_dma;
+		buf->skb = skb;
+
+		skb = alloc_skb(skb->len, GFP_ATOMIC);
+		if (skb) {
+			skb_put(skb, len);
+			skb_copy_from_linear_data(buf->skb, skb->data, len);
+		}
+	}
+
+	prestera_sdma_rx_desc_init(sdma, buf->desc, buf->buf_dma);
+
+	return skb;
+}
+
+static int prestera_rxtx_process_skb(struct prestera_sdma *sdma,
+				     struct sk_buff *skb)
+{
+	struct prestera_port *port;
+	struct prestera_dsa dsa;
+	u32 hw_port, dev_id;
+	u8 cpu_code;
+	int err;
+
+	skb_pull(skb, ETH_HLEN);
+
+	/* ethertype field is part of the dsa header */
+	err = prestera_dsa_parse(&dsa, skb->data - ETH_TLEN);
+	if (err)
+		return err;
+
+	dev_id = dsa.hw_dev_num;
+	hw_port = dsa.port_num;
+
+	port = prestera_port_find_by_hwid(sdma->sw, dev_id, hw_port);
+	if (unlikely(!port)) {
+		dev_warn_ratelimited(prestera_dev(sdma->sw), "received pkt for non-existent port(%u, %u)\n",
+				     dev_id, hw_port);
+		return -ENOENT;
+	}
+
+	if (unlikely(!pskb_may_pull(skb, PRESTERA_DSA_HLEN)))
+		return -EINVAL;
+
+	/* remove DSA tag and update checksum */
+	skb_pull_rcsum(skb, PRESTERA_DSA_HLEN);
+
+	memmove(skb->data - ETH_HLEN, skb->data - ETH_HLEN - PRESTERA_DSA_HLEN,
+		ETH_ALEN * 2);
+
+	skb_push(skb, ETH_HLEN);
+
+	skb->protocol = eth_type_trans(skb, port->dev);
+
+	if (dsa.vlan.is_tagged) {
+		u16 tci = dsa.vlan.vid & VLAN_VID_MASK;
+
+		tci |= dsa.vlan.vpt << VLAN_PRIO_SHIFT;
+		if (dsa.vlan.cfi_bit)
+			tci |= VLAN_CFI_MASK;
+
+		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), tci);
+	}
+
+	cpu_code = dsa.cpu_code;
+	prestera_devlink_trap_report(port, skb, cpu_code);
+
+	return 0;
+}
+
+static int prestera_sdma_next_rx_buf_idx(int buf_idx)
+{
+	return (buf_idx + 1) % PRESTERA_SDMA_RX_DESC_PER_Q;
+}
+
+static int prestera_sdma_rx_poll(struct napi_struct *napi, int budget)
+{
+	int qnum = PRESTERA_SDMA_RX_QUEUE_NUM;
+	unsigned int rxq_done_map = 0;
+	struct prestera_sdma *sdma;
+	struct list_head rx_list;
+	unsigned int qmask;
+	int pkts_done = 0;
+	int q;
+
+	qnum = PRESTERA_SDMA_RX_QUEUE_NUM;
+	qmask = GENMASK(qnum - 1, 0);
+
+	INIT_LIST_HEAD(&rx_list);
+
+	sdma = container_of(napi, struct prestera_sdma, rx_napi);
+
+	while (pkts_done < budget && rxq_done_map != qmask) {
+		for (q = 0; q < qnum && pkts_done < budget; q++) {
+			struct prestera_rx_ring *ring = &sdma->rx_ring[q];
+			struct prestera_sdma_desc *desc;
+			struct prestera_sdma_buf *buf;
+			int buf_idx = ring->next_rx;
+			struct sk_buff *skb;
+
+			buf = &ring->bufs[buf_idx];
+			desc = buf->desc;
+
+			if (PRESTERA_SDMA_RX_DESC_IS_RCVD(desc)) {
+				rxq_done_map &= ~BIT(q);
+			} else {
+				rxq_done_map |= BIT(q);
+				continue;
+			}
+
+			pkts_done++;
+
+			__skb_trim(buf->skb, PRESTERA_SDMA_RX_DESC_PKT_LEN(desc));
+
+			skb = prestera_sdma_rx_skb_get(sdma, buf);
+			if (!skb)
+				goto rx_next_buf;
+
+			if (unlikely(prestera_rxtx_process_skb(sdma, skb)))
+				goto rx_next_buf;
+
+			list_add_tail(&skb->list, &rx_list);
+rx_next_buf:
+			ring->next_rx = prestera_sdma_next_rx_buf_idx(buf_idx);
+		}
+	}
+
+	if (pkts_done < budget && napi_complete_done(napi, pkts_done))
+		prestera_write(sdma->sw, PRESTERA_SDMA_RX_INTR_MASK_REG,
+			       GENMASK(9, 2));
+
+	netif_receive_skb_list(&rx_list);
+
+	return pkts_done;
+}
+
+static void prestera_sdma_rx_fini(struct prestera_sdma *sdma)
+{
+	int qnum = PRESTERA_SDMA_RX_QUEUE_NUM;
+	int q, b;
+
+	/* disable all rx queues */
+	prestera_write(sdma->sw, PRESTERA_SDMA_RX_QUEUE_STATUS_REG,
+		       GENMASK(15, 8));
+
+	for (q = 0; q < qnum; q++) {
+		struct prestera_rx_ring *ring = &sdma->rx_ring[q];
+
+		if (!ring->bufs)
+			break;
+
+		for (b = 0; b < PRESTERA_SDMA_RX_DESC_PER_Q; b++) {
+			struct prestera_sdma_buf *buf = &ring->bufs[b];
+
+			if (buf->desc_dma)
+				dma_pool_free(sdma->desc_pool, buf->desc,
+					      buf->desc_dma);
+
+			if (!buf->skb)
+				continue;
+
+			if (buf->buf_dma != DMA_MAPPING_ERROR)
+				dma_unmap_single(sdma->sw->dev->dev,
+						 buf->buf_dma, buf->skb->len,
+						 DMA_FROM_DEVICE);
+			kfree_skb(buf->skb);
+		}
+	}
+}
+
+static int prestera_sdma_rx_init(struct prestera_sdma *sdma)
+{
+	int bnum = PRESTERA_SDMA_RX_DESC_PER_Q;
+	int qnum = PRESTERA_SDMA_RX_QUEUE_NUM;
+	int err;
+	int q;
+
+	/* disable all rx queues */
+	prestera_write(sdma->sw, PRESTERA_SDMA_RX_QUEUE_STATUS_REG,
+		       GENMASK(15, 8));
+
+	for (q = 0; q < qnum; q++) {
+		struct prestera_sdma_buf *head, *tail, *next, *prev;
+		struct prestera_rx_ring *ring = &sdma->rx_ring[q];
+
+		ring->bufs = kmalloc_array(bnum, sizeof(*head), GFP_KERNEL);
+		if (!ring->bufs)
+			return -ENOMEM;
+
+		ring->next_rx = 0;
+
+		tail = &ring->bufs[bnum - 1];
+		head = &ring->bufs[0];
+		next = head;
+		prev = next;
+
+		do {
+			err = prestera_sdma_buf_init(sdma, next);
+			if (err)
+				return err;
+
+			err = prestera_sdma_rx_skb_alloc(sdma, next);
+			if (err)
+				return err;
+
+			prestera_sdma_rx_desc_init(sdma, next->desc,
+						   next->buf_dma);
+
+			prestera_sdma_rx_desc_set_next(sdma, prev->desc,
+						       next->desc_dma);
+
+			prev = next;
+			next++;
+		} while (prev != tail);
+
+		/* join tail with head to make a circular list */
+		prestera_sdma_rx_desc_set_next(sdma, tail->desc, head->desc_dma);
+
+		prestera_write(sdma->sw, PRESTERA_SDMA_RX_QUEUE_DESC_REG(q),
+			       prestera_sdma_map(sdma, head->desc_dma));
+	}
+
+	/* make sure all rx descs are filled before enabling all rx queues */
+	wmb();
+
+	prestera_write(sdma->sw, PRESTERA_SDMA_RX_QUEUE_STATUS_REG,
+		       GENMASK(7, 0));
+
+	return 0;
+}
+
+static void prestera_sdma_tx_desc_init(struct prestera_sdma *sdma,
+				       struct prestera_sdma_desc *desc)
+{
+	desc->word1 = cpu_to_le32(PRESTERA_SDMA_TX_DESC_INIT);
+	desc->word2 = 0;
+}
+
+static void prestera_sdma_tx_desc_set_next(struct prestera_sdma *sdma,
+					   struct prestera_sdma_desc *desc,
+					   dma_addr_t next)
+{
+	desc->next = cpu_to_le32(prestera_sdma_map(sdma, next));
+}
+
+static void prestera_sdma_tx_desc_set_buf(struct prestera_sdma *sdma,
+					  struct prestera_sdma_desc *desc,
+					  dma_addr_t buf, size_t len)
+{
+	u32 word = le32_to_cpu(desc->word2);
+
+	u32p_replace_bits(&word, len + ETH_FCS_LEN, GENMASK(30, 16));
+
+	desc->buff = cpu_to_le32(prestera_sdma_map(sdma, buf));
+	desc->word2 = cpu_to_le32(word);
+}
+
+static void prestera_sdma_tx_desc_xmit(struct prestera_sdma_desc *desc)
+{
+	u32 word = le32_to_cpu(desc->word1);
+
+	word |= PRESTERA_SDMA_TX_DESC_DMA_OWN << 31;
+
+	/* make sure everything is written before enable xmit */
+	wmb();
+
+	desc->word1 = cpu_to_le32(word);
+}
+
+static int prestera_sdma_tx_buf_map(struct prestera_sdma *sdma,
+				    struct prestera_sdma_buf *buf,
+				    struct sk_buff *skb)
+{
+	struct device *dma_dev = sdma->sw->dev->dev;
+	dma_addr_t dma;
+
+	dma = dma_map_single(dma_dev, skb->data, skb->len, DMA_TO_DEVICE);
+	if (dma_mapping_error(dma_dev, dma))
+		return -ENOMEM;
+
+	buf->buf_dma = dma;
+	buf->skb = skb;
+
+	return 0;
+}
+
+static void prestera_sdma_tx_buf_unmap(struct prestera_sdma *sdma,
+				       struct prestera_sdma_buf *buf)
+{
+	struct device *dma_dev = sdma->sw->dev->dev;
+
+	dma_unmap_single(dma_dev, buf->buf_dma, buf->skb->len, DMA_TO_DEVICE);
+}
+
+static void prestera_sdma_tx_recycle_work_fn(struct work_struct *work)
+{
+	int bnum = PRESTERA_SDMA_TX_DESC_PER_Q;
+	struct prestera_tx_ring *tx_ring;
+	struct prestera_sdma *sdma;
+	int b;
+
+	sdma = container_of(work, struct prestera_sdma, tx_work);
+
+	tx_ring = &sdma->tx_ring;
+
+	for (b = 0; b < bnum; b++) {
+		struct prestera_sdma_buf *buf = &tx_ring->bufs[b];
+
+		if (!buf->is_used)
+			continue;
+
+		if (!PRESTERA_SDMA_TX_DESC_IS_SENT(buf->desc))
+			continue;
+
+		prestera_sdma_tx_buf_unmap(sdma, buf);
+		dev_consume_skb_any(buf->skb);
+		buf->skb = NULL;
+
+		/* make sure everything is cleaned up */
+		wmb();
+
+		buf->is_used = false;
+	}
+}
+
+static int prestera_sdma_tx_init(struct prestera_sdma *sdma)
+{
+	struct prestera_sdma_buf *head, *tail, *next, *prev;
+	struct prestera_tx_ring *tx_ring = &sdma->tx_ring;
+	int bnum = PRESTERA_SDMA_TX_DESC_PER_Q;
+	int err;
+
+	INIT_WORK(&sdma->tx_work, prestera_sdma_tx_recycle_work_fn);
+	spin_lock_init(&sdma->tx_lock);
+
+	tx_ring->bufs = kmalloc_array(bnum, sizeof(*head), GFP_KERNEL);
+	if (!tx_ring->bufs)
+		return -ENOMEM;
+
+	tail = &tx_ring->bufs[bnum - 1];
+	head = &tx_ring->bufs[0];
+	next = head;
+	prev = next;
+
+	tx_ring->max_burst = PRESTERA_SDMA_TX_MAX_BURST;
+	tx_ring->burst = tx_ring->max_burst;
+	tx_ring->next_tx = 0;
+
+	do {
+		err = prestera_sdma_buf_init(sdma, next);
+		if (err)
+			return err;
+
+		next->is_used = false;
+
+		prestera_sdma_tx_desc_init(sdma, next->desc);
+
+		prestera_sdma_tx_desc_set_next(sdma, prev->desc,
+					       next->desc_dma);
+
+		prev = next;
+		next++;
+	} while (prev != tail);
+
+	/* join tail with head to make a circular list */
+	prestera_sdma_tx_desc_set_next(sdma, tail->desc, head->desc_dma);
+
+	/* make sure descriptors are written */
+	wmb();
+
+	prestera_write(sdma->sw, PRESTERA_SDMA_TX_QUEUE_DESC_REG,
+		       prestera_sdma_map(sdma, head->desc_dma));
+
+	return 0;
+}
+
+static void prestera_sdma_tx_fini(struct prestera_sdma *sdma)
+{
+	struct prestera_tx_ring *ring = &sdma->tx_ring;
+	int bnum = PRESTERA_SDMA_TX_DESC_PER_Q;
+	int b;
+
+	cancel_work_sync(&sdma->tx_work);
+
+	if (!ring->bufs)
+		return;
+
+	for (b = 0; b < bnum; b++) {
+		struct prestera_sdma_buf *buf = &ring->bufs[b];
+
+		if (buf->desc)
+			dma_pool_free(sdma->desc_pool, buf->desc,
+				      buf->desc_dma);
+
+		if (!buf->skb)
+			continue;
+
+		dma_unmap_single(sdma->sw->dev->dev, buf->buf_dma,
+				 buf->skb->len, DMA_TO_DEVICE);
+
+		dev_consume_skb_any(buf->skb);
+	}
+}
+
+static void prestera_rxtx_handle_event(struct prestera_switch *sw,
+				       struct prestera_event *evt,
+				       void *arg)
+{
+	struct prestera_sdma *sdma = arg;
+
+	if (evt->id != PRESTERA_RXTX_EVENT_RCV_PKT)
+		return;
+
+	prestera_write(sdma->sw, PRESTERA_SDMA_RX_INTR_MASK_REG, 0);
+	napi_schedule(&sdma->rx_napi);
+}
+
+static int prestera_sdma_switch_init(struct prestera_switch *sw)
+{
+	struct prestera_sdma *sdma = &sw->rxtx->sdma;
+	struct device *dev = sw->dev->dev;
+	struct prestera_rxtx_params p;
+	int err;
+
+	p.use_sdma = true;
+
+	err = prestera_hw_rxtx_init(sw, &p);
+	if (err) {
+		dev_err(dev, "failed to init rxtx by hw\n");
+		return err;
+	}
+
+	sdma->dma_mask = dma_get_mask(dev);
+	sdma->map_addr = p.map_addr;
+	sdma->sw = sw;
+
+	sdma->desc_pool = dma_pool_create("desc_pool", dev,
+					  sizeof(struct prestera_sdma_desc),
+					  16, 0);
+	if (!sdma->desc_pool)
+		return -ENOMEM;
+
+	err = prestera_sdma_rx_init(sdma);
+	if (err) {
+		dev_err(dev, "failed to init rx ring\n");
+		goto err_rx_init;
+	}
+
+	err = prestera_sdma_tx_init(sdma);
+	if (err) {
+		dev_err(dev, "failed to init tx ring\n");
+		goto err_tx_init;
+	}
+
+	err = prestera_hw_event_handler_register(sw, PRESTERA_EVENT_TYPE_RXTX,
+						 prestera_rxtx_handle_event,
+						 sdma);
+	if (err)
+		goto err_evt_register;
+
+	init_dummy_netdev(&sdma->napi_dev);
+
+	netif_napi_add(&sdma->napi_dev, &sdma->rx_napi, prestera_sdma_rx_poll);
+	napi_enable(&sdma->rx_napi);
+
+	return 0;
+
+err_evt_register:
+err_tx_init:
+	prestera_sdma_tx_fini(sdma);
+err_rx_init:
+	prestera_sdma_rx_fini(sdma);
+
+	dma_pool_destroy(sdma->desc_pool);
+	return err;
+}
+
+static void prestera_sdma_switch_fini(struct prestera_switch *sw)
+{
+	struct prestera_sdma *sdma = &sw->rxtx->sdma;
+
+	napi_disable(&sdma->rx_napi);
+	netif_napi_del(&sdma->rx_napi);
+	prestera_hw_event_handler_unregister(sw, PRESTERA_EVENT_TYPE_RXTX,
+					     prestera_rxtx_handle_event);
+	prestera_sdma_tx_fini(sdma);
+	prestera_sdma_rx_fini(sdma);
+	dma_pool_destroy(sdma->desc_pool);
+}
+
+static bool prestera_sdma_is_ready(struct prestera_sdma *sdma)
+{
+	return !(prestera_read(sdma->sw, PRESTERA_SDMA_TX_QUEUE_START_REG) & 1);
+}
+
+static int prestera_sdma_tx_wait(struct prestera_sdma *sdma,
+				 struct prestera_tx_ring *tx_ring)
+{
+	int tx_wait_num = PRESTERA_SDMA_WAIT_MUL * tx_ring->max_burst;
+
+	do {
+		if (prestera_sdma_is_ready(sdma))
+			return 0;
+
+		udelay(1);
+	} while (--tx_wait_num);
+
+	return -EBUSY;
+}
+
+static void prestera_sdma_tx_start(struct prestera_sdma *sdma)
+{
+	prestera_write(sdma->sw, PRESTERA_SDMA_TX_QUEUE_START_REG, 1);
+	schedule_work(&sdma->tx_work);
+}
+
+static netdev_tx_t prestera_sdma_xmit(struct prestera_sdma *sdma,
+				      struct sk_buff *skb)
+{
+	struct device *dma_dev = sdma->sw->dev->dev;
+	struct net_device *dev = skb->dev;
+	struct prestera_tx_ring *tx_ring;
+	struct prestera_sdma_buf *buf;
+	int err;
+
+	spin_lock(&sdma->tx_lock);
+
+	tx_ring = &sdma->tx_ring;
+
+	buf = &tx_ring->bufs[tx_ring->next_tx];
+	if (buf->is_used) {
+		schedule_work(&sdma->tx_work);
+		goto drop_skb;
+	}
+
+	if (unlikely(eth_skb_pad(skb)))
+		goto drop_skb_nofree;
+
+	err = prestera_sdma_tx_buf_map(sdma, buf, skb);
+	if (err)
+		goto drop_skb;
+
+	prestera_sdma_tx_desc_set_buf(sdma, buf->desc, buf->buf_dma, skb->len);
+
+	dma_sync_single_for_device(dma_dev, buf->buf_dma, skb->len,
+				   DMA_TO_DEVICE);
+
+	if (tx_ring->burst) {
+		tx_ring->burst--;
+	} else {
+		tx_ring->burst = tx_ring->max_burst;
+
+		err = prestera_sdma_tx_wait(sdma, tx_ring);
+		if (err)
+			goto drop_skb_unmap;
+	}
+
+	tx_ring->next_tx = (tx_ring->next_tx + 1) % PRESTERA_SDMA_TX_DESC_PER_Q;
+	prestera_sdma_tx_desc_xmit(buf->desc);
+	buf->is_used = true;
+
+	prestera_sdma_tx_start(sdma);
+
+	goto tx_done;
+
+drop_skb_unmap:
+	prestera_sdma_tx_buf_unmap(sdma, buf);
+drop_skb:
+	dev_consume_skb_any(skb);
+drop_skb_nofree:
+	dev->stats.tx_dropped++;
+tx_done:
+	spin_unlock(&sdma->tx_lock);
+	return NETDEV_TX_OK;
+}
+
+int prestera_rxtx_switch_init(struct prestera_switch *sw)
+{
+	struct prestera_rxtx *rxtx;
+	int err;
+
+	rxtx = kzalloc(sizeof(*rxtx), GFP_KERNEL);
+	if (!rxtx)
+		return -ENOMEM;
+
+	sw->rxtx = rxtx;
+
+	err = prestera_sdma_switch_init(sw);
+	if (err)
+		kfree(rxtx);
+
+	return err;
+}
+
+void prestera_rxtx_switch_fini(struct prestera_switch *sw)
+{
+	prestera_sdma_switch_fini(sw);
+	kfree(sw->rxtx);
+}
+
+int prestera_rxtx_port_init(struct prestera_port *port)
+{
+	port->dev->needed_headroom = PRESTERA_DSA_HLEN;
+	return 0;
+}
+
+netdev_tx_t prestera_rxtx_xmit(struct prestera_port *port, struct sk_buff *skb)
+{
+	struct prestera_dsa dsa;
+
+	dsa.hw_dev_num = port->dev_id;
+	dsa.port_num = port->hw_id;
+
+	if (skb_cow_head(skb, PRESTERA_DSA_HLEN) < 0)
+		return NET_XMIT_DROP;
+
+	skb_push(skb, PRESTERA_DSA_HLEN);
+	memmove(skb->data, skb->data + PRESTERA_DSA_HLEN, 2 * ETH_ALEN);
+
+	if (prestera_dsa_build(&dsa, skb->data + 2 * ETH_ALEN) != 0)
+		return NET_XMIT_DROP;
+
+	return prestera_sdma_xmit(&port->sw->rxtx->sdma, skb);
+}