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, 643 insertions, 0 deletions

diff --git a/drivers/usb/mtu3/mtu3_qmu.c b/drivers/usb/mtu3/mtu3_qmu.c
new file mode 100644
index 000000000..2ea3157dd
--- /dev/null
+++ b/drivers/usb/mtu3/mtu3_qmu.c
@@ -0,0 +1,643 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * mtu3_qmu.c - Queue Management Unit driver for device controller
+ *
+ * Copyright (C) 2016 MediaTek Inc.
+ *
+ * Author: Chunfeng Yun <chunfeng.yun@mediatek.com>
+ */
+
+/*
+ * Queue Management Unit (QMU) is designed to unload SW effort
+ * to serve DMA interrupts.
+ * By preparing General Purpose Descriptor (GPD) and Buffer Descriptor (BD),
+ * SW links data buffers and triggers QMU to send / receive data to
+ * host / from device at a time.
+ * And now only GPD is supported.
+ *
+ * For more detailed information, please refer to QMU Programming Guide
+ */
+
+#include <linux/dmapool.h>
+#include <linux/iopoll.h>
+
+#include "mtu3.h"
+#include "mtu3_trace.h"
+
+#define QMU_CHECKSUM_LEN	16
+
+#define GPD_FLAGS_HWO	BIT(0)
+#define GPD_FLAGS_BDP	BIT(1)
+#define GPD_FLAGS_BPS	BIT(2)
+#define GPD_FLAGS_ZLP	BIT(6)
+#define GPD_FLAGS_IOC	BIT(7)
+#define GET_GPD_HWO(gpd)	(le32_to_cpu((gpd)->dw0_info) & GPD_FLAGS_HWO)
+
+#define GPD_RX_BUF_LEN_OG(x)	(((x) & 0xffff) << 16)
+#define GPD_RX_BUF_LEN_EL(x)	(((x) & 0xfffff) << 12)
+#define GPD_RX_BUF_LEN(mtu, x)	\
+({				\
+	typeof(x) x_ = (x);	\
+	((mtu)->gen2cp) ? GPD_RX_BUF_LEN_EL(x_) : GPD_RX_BUF_LEN_OG(x_); \
+})
+
+#define GPD_DATA_LEN_OG(x)	((x) & 0xffff)
+#define GPD_DATA_LEN_EL(x)	((x) & 0xfffff)
+#define GPD_DATA_LEN(mtu, x)	\
+({				\
+	typeof(x) x_ = (x);	\
+	((mtu)->gen2cp) ? GPD_DATA_LEN_EL(x_) : GPD_DATA_LEN_OG(x_); \
+})
+
+#define GPD_EXT_FLAG_ZLP	BIT(29)
+#define GPD_EXT_NGP_OG(x)	(((x) & 0xf) << 20)
+#define GPD_EXT_BUF_OG(x)	(((x) & 0xf) << 16)
+#define GPD_EXT_NGP_EL(x)	(((x) & 0xf) << 28)
+#define GPD_EXT_BUF_EL(x)	(((x) & 0xf) << 24)
+#define GPD_EXT_NGP(mtu, x)	\
+({				\
+	typeof(x) x_ = (x);	\
+	((mtu)->gen2cp) ? GPD_EXT_NGP_EL(x_) : GPD_EXT_NGP_OG(x_); \
+})
+
+#define GPD_EXT_BUF(mtu, x)	\
+({				\
+	typeof(x) x_ = (x);	\
+	((mtu)->gen2cp) ? GPD_EXT_BUF_EL(x_) : GPD_EXT_BUF_OG(x_); \
+})
+
+#define HILO_GEN64(hi, lo) (((u64)(hi) << 32) + (lo))
+#define HILO_DMA(hi, lo)	\
+	((dma_addr_t)HILO_GEN64((le32_to_cpu(hi)), (le32_to_cpu(lo))))
+
+static dma_addr_t read_txq_cur_addr(void __iomem *mbase, u8 epnum)
+{
+	u32 txcpr;
+	u32 txhiar;
+
+	txcpr = mtu3_readl(mbase, USB_QMU_TQCPR(epnum));
+	txhiar = mtu3_readl(mbase, USB_QMU_TQHIAR(epnum));
+
+	return HILO_DMA(QMU_CUR_GPD_ADDR_HI(txhiar), txcpr);
+}
+
+static dma_addr_t read_rxq_cur_addr(void __iomem *mbase, u8 epnum)
+{
+	u32 rxcpr;
+	u32 rxhiar;
+
+	rxcpr = mtu3_readl(mbase, USB_QMU_RQCPR(epnum));
+	rxhiar = mtu3_readl(mbase, USB_QMU_RQHIAR(epnum));
+
+	return HILO_DMA(QMU_CUR_GPD_ADDR_HI(rxhiar), rxcpr);
+}
+
+static void write_txq_start_addr(void __iomem *mbase, u8 epnum, dma_addr_t dma)
+{
+	u32 tqhiar;
+
+	mtu3_writel(mbase, USB_QMU_TQSAR(epnum),
+		    cpu_to_le32(lower_32_bits(dma)));
+	tqhiar = mtu3_readl(mbase, USB_QMU_TQHIAR(epnum));
+	tqhiar &= ~QMU_START_ADDR_HI_MSK;
+	tqhiar |= QMU_START_ADDR_HI(upper_32_bits(dma));
+	mtu3_writel(mbase, USB_QMU_TQHIAR(epnum), tqhiar);
+}
+
+static void write_rxq_start_addr(void __iomem *mbase, u8 epnum, dma_addr_t dma)
+{
+	u32 rqhiar;
+
+	mtu3_writel(mbase, USB_QMU_RQSAR(epnum),
+		    cpu_to_le32(lower_32_bits(dma)));
+	rqhiar = mtu3_readl(mbase, USB_QMU_RQHIAR(epnum));
+	rqhiar &= ~QMU_START_ADDR_HI_MSK;
+	rqhiar |= QMU_START_ADDR_HI(upper_32_bits(dma));
+	mtu3_writel(mbase, USB_QMU_RQHIAR(epnum), rqhiar);
+}
+
+static struct qmu_gpd *gpd_dma_to_virt(struct mtu3_gpd_ring *ring,
+		dma_addr_t dma_addr)
+{
+	dma_addr_t dma_base = ring->dma;
+	struct qmu_gpd *gpd_head = ring->start;
+	u32 offset = (dma_addr - dma_base) / sizeof(*gpd_head);
+
+	if (offset >= MAX_GPD_NUM)
+		return NULL;
+
+	return gpd_head + offset;
+}
+
+static dma_addr_t gpd_virt_to_dma(struct mtu3_gpd_ring *ring,
+		struct qmu_gpd *gpd)
+{
+	dma_addr_t dma_base = ring->dma;
+	struct qmu_gpd *gpd_head = ring->start;
+	u32 offset;
+
+	offset = gpd - gpd_head;
+	if (offset >= MAX_GPD_NUM)
+		return 0;
+
+	return dma_base + (offset * sizeof(*gpd));
+}
+
+static void gpd_ring_init(struct mtu3_gpd_ring *ring, struct qmu_gpd *gpd)
+{
+	ring->start = gpd;
+	ring->enqueue = gpd;
+	ring->dequeue = gpd;
+	ring->end = gpd + MAX_GPD_NUM - 1;
+}
+
+static void reset_gpd_list(struct mtu3_ep *mep)
+{
+	struct mtu3_gpd_ring *ring = &mep->gpd_ring;
+	struct qmu_gpd *gpd = ring->start;
+
+	if (gpd) {
+		gpd->dw0_info &= cpu_to_le32(~GPD_FLAGS_HWO);
+		gpd_ring_init(ring, gpd);
+	}
+}
+
+int mtu3_gpd_ring_alloc(struct mtu3_ep *mep)
+{
+	struct qmu_gpd *gpd;
+	struct mtu3_gpd_ring *ring = &mep->gpd_ring;
+
+	/* software own all gpds as default */
+	gpd = dma_pool_zalloc(mep->mtu->qmu_gpd_pool, GFP_ATOMIC, &ring->dma);
+	if (gpd == NULL)
+		return -ENOMEM;
+
+	gpd_ring_init(ring, gpd);
+
+	return 0;
+}
+
+void mtu3_gpd_ring_free(struct mtu3_ep *mep)
+{
+	struct mtu3_gpd_ring *ring = &mep->gpd_ring;
+
+	dma_pool_free(mep->mtu->qmu_gpd_pool,
+			ring->start, ring->dma);
+	memset(ring, 0, sizeof(*ring));
+}
+
+void mtu3_qmu_resume(struct mtu3_ep *mep)
+{
+	struct mtu3 *mtu = mep->mtu;
+	void __iomem *mbase = mtu->mac_base;
+	int epnum = mep->epnum;
+	u32 offset;
+
+	offset = mep->is_in ? USB_QMU_TQCSR(epnum) : USB_QMU_RQCSR(epnum);
+
+	mtu3_writel(mbase, offset, QMU_Q_RESUME);
+	if (!(mtu3_readl(mbase, offset) & QMU_Q_ACTIVE))
+		mtu3_writel(mbase, offset, QMU_Q_RESUME);
+}
+
+static struct qmu_gpd *advance_enq_gpd(struct mtu3_gpd_ring *ring)
+{
+	if (ring->enqueue < ring->end)
+		ring->enqueue++;
+	else
+		ring->enqueue = ring->start;
+
+	return ring->enqueue;
+}
+
+static struct qmu_gpd *advance_deq_gpd(struct mtu3_gpd_ring *ring)
+{
+	if (ring->dequeue < ring->end)
+		ring->dequeue++;
+	else
+		ring->dequeue = ring->start;
+
+	return ring->dequeue;
+}
+
+/* check if a ring is emtpy */
+static int gpd_ring_empty(struct mtu3_gpd_ring *ring)
+{
+	struct qmu_gpd *enq = ring->enqueue;
+	struct qmu_gpd *next;
+
+	if (ring->enqueue < ring->end)
+		next = enq + 1;
+	else
+		next = ring->start;
+
+	/* one gpd is reserved to simplify gpd preparation */
+	return next == ring->dequeue;
+}
+
+int mtu3_prepare_transfer(struct mtu3_ep *mep)
+{
+	return gpd_ring_empty(&mep->gpd_ring);
+}
+
+static int mtu3_prepare_tx_gpd(struct mtu3_ep *mep, struct mtu3_request *mreq)
+{
+	struct qmu_gpd *enq;
+	struct mtu3_gpd_ring *ring = &mep->gpd_ring;
+	struct qmu_gpd *gpd = ring->enqueue;
+	struct usb_request *req = &mreq->request;
+	struct mtu3 *mtu = mep->mtu;
+	dma_addr_t enq_dma;
+	u32 ext_addr;
+
+	gpd->dw0_info = 0;	/* SW own it */
+	gpd->buffer = cpu_to_le32(lower_32_bits(req->dma));
+	ext_addr = GPD_EXT_BUF(mtu, upper_32_bits(req->dma));
+	gpd->dw3_info = cpu_to_le32(GPD_DATA_LEN(mtu, req->length));
+
+	/* get the next GPD */
+	enq = advance_enq_gpd(ring);
+	enq_dma = gpd_virt_to_dma(ring, enq);
+	dev_dbg(mep->mtu->dev, "TX-EP%d queue gpd=%p, enq=%p, qdma=%pad\n",
+		mep->epnum, gpd, enq, &enq_dma);
+
+	enq->dw0_info &= cpu_to_le32(~GPD_FLAGS_HWO);
+	gpd->next_gpd = cpu_to_le32(lower_32_bits(enq_dma));
+	ext_addr |= GPD_EXT_NGP(mtu, upper_32_bits(enq_dma));
+	gpd->dw0_info = cpu_to_le32(ext_addr);
+
+	if (req->zero) {
+		if (mtu->gen2cp)
+			gpd->dw0_info |= cpu_to_le32(GPD_FLAGS_ZLP);
+		else
+			gpd->dw3_info |= cpu_to_le32(GPD_EXT_FLAG_ZLP);
+	}
+
+	/* prevent reorder, make sure GPD's HWO is set last */
+	mb();
+	gpd->dw0_info |= cpu_to_le32(GPD_FLAGS_IOC | GPD_FLAGS_HWO);
+
+	mreq->gpd = gpd;
+	trace_mtu3_prepare_gpd(mep, gpd);
+
+	return 0;
+}
+
+static int mtu3_prepare_rx_gpd(struct mtu3_ep *mep, struct mtu3_request *mreq)
+{
+	struct qmu_gpd *enq;
+	struct mtu3_gpd_ring *ring = &mep->gpd_ring;
+	struct qmu_gpd *gpd = ring->enqueue;
+	struct usb_request *req = &mreq->request;
+	struct mtu3 *mtu = mep->mtu;
+	dma_addr_t enq_dma;
+	u32 ext_addr;
+
+	gpd->dw0_info = 0;	/* SW own it */
+	gpd->buffer = cpu_to_le32(lower_32_bits(req->dma));
+	ext_addr = GPD_EXT_BUF(mtu, upper_32_bits(req->dma));
+	gpd->dw0_info = cpu_to_le32(GPD_RX_BUF_LEN(mtu, req->length));
+
+	/* get the next GPD */
+	enq = advance_enq_gpd(ring);
+	enq_dma = gpd_virt_to_dma(ring, enq);
+	dev_dbg(mep->mtu->dev, "RX-EP%d queue gpd=%p, enq=%p, qdma=%pad\n",
+		mep->epnum, gpd, enq, &enq_dma);
+
+	enq->dw0_info &= cpu_to_le32(~GPD_FLAGS_HWO);
+	gpd->next_gpd = cpu_to_le32(lower_32_bits(enq_dma));
+	ext_addr |= GPD_EXT_NGP(mtu, upper_32_bits(enq_dma));
+	gpd->dw3_info = cpu_to_le32(ext_addr);
+	/* prevent reorder, make sure GPD's HWO is set last */
+	mb();
+	gpd->dw0_info |= cpu_to_le32(GPD_FLAGS_IOC | GPD_FLAGS_HWO);
+
+	mreq->gpd = gpd;
+	trace_mtu3_prepare_gpd(mep, gpd);
+
+	return 0;
+}
+
+void mtu3_insert_gpd(struct mtu3_ep *mep, struct mtu3_request *mreq)
+{
+
+	if (mep->is_in)
+		mtu3_prepare_tx_gpd(mep, mreq);
+	else
+		mtu3_prepare_rx_gpd(mep, mreq);
+}
+
+int mtu3_qmu_start(struct mtu3_ep *mep)
+{
+	struct mtu3 *mtu = mep->mtu;
+	void __iomem *mbase = mtu->mac_base;
+	struct mtu3_gpd_ring *ring = &mep->gpd_ring;
+	u8 epnum = mep->epnum;
+
+	if (mep->is_in) {
+		/* set QMU start address */
+		write_txq_start_addr(mbase, epnum, ring->dma);
+		mtu3_setbits(mbase, MU3D_EP_TXCR0(epnum), TX_DMAREQEN);
+		/* send zero length packet according to ZLP flag in GPD */
+		mtu3_setbits(mbase, U3D_QCR1, QMU_TX_ZLP(epnum));
+		mtu3_writel(mbase, U3D_TQERRIESR0,
+				QMU_TX_LEN_ERR(epnum) | QMU_TX_CS_ERR(epnum));
+
+		if (mtu3_readl(mbase, USB_QMU_TQCSR(epnum)) & QMU_Q_ACTIVE) {
+			dev_warn(mtu->dev, "Tx %d Active Now!\n", epnum);
+			return 0;
+		}
+		mtu3_writel(mbase, USB_QMU_TQCSR(epnum), QMU_Q_START);
+
+	} else {
+		write_rxq_start_addr(mbase, epnum, ring->dma);
+		mtu3_setbits(mbase, MU3D_EP_RXCR0(epnum), RX_DMAREQEN);
+		/* don't expect ZLP */
+		mtu3_clrbits(mbase, U3D_QCR3, QMU_RX_ZLP(epnum));
+		/* move to next GPD when receive ZLP */
+		mtu3_setbits(mbase, U3D_QCR3, QMU_RX_COZ(epnum));
+		mtu3_writel(mbase, U3D_RQERRIESR0,
+				QMU_RX_LEN_ERR(epnum) | QMU_RX_CS_ERR(epnum));
+		mtu3_writel(mbase, U3D_RQERRIESR1, QMU_RX_ZLP_ERR(epnum));
+
+		if (mtu3_readl(mbase, USB_QMU_RQCSR(epnum)) & QMU_Q_ACTIVE) {
+			dev_warn(mtu->dev, "Rx %d Active Now!\n", epnum);
+			return 0;
+		}
+		mtu3_writel(mbase, USB_QMU_RQCSR(epnum), QMU_Q_START);
+	}
+
+	return 0;
+}
+
+/* may called in atomic context */
+void mtu3_qmu_stop(struct mtu3_ep *mep)
+{
+	struct mtu3 *mtu = mep->mtu;
+	void __iomem *mbase = mtu->mac_base;
+	int epnum = mep->epnum;
+	u32 value = 0;
+	u32 qcsr;
+	int ret;
+
+	qcsr = mep->is_in ? USB_QMU_TQCSR(epnum) : USB_QMU_RQCSR(epnum);
+
+	if (!(mtu3_readl(mbase, qcsr) & QMU_Q_ACTIVE)) {
+		dev_dbg(mtu->dev, "%s's qmu is inactive now!\n", mep->name);
+		return;
+	}
+	mtu3_writel(mbase, qcsr, QMU_Q_STOP);
+
+	ret = readl_poll_timeout_atomic(mbase + qcsr, value,
+			!(value & QMU_Q_ACTIVE), 1, 1000);
+	if (ret) {
+		dev_err(mtu->dev, "stop %s's qmu failed\n", mep->name);
+		return;
+	}
+
+	dev_dbg(mtu->dev, "%s's qmu stop now!\n", mep->name);
+}
+
+void mtu3_qmu_flush(struct mtu3_ep *mep)
+{
+
+	dev_dbg(mep->mtu->dev, "%s flush QMU %s\n", __func__,
+		((mep->is_in) ? "TX" : "RX"));
+
+	/*Stop QMU */
+	mtu3_qmu_stop(mep);
+	reset_gpd_list(mep);
+}
+
+/*
+ * QMU can't transfer zero length packet directly (a hardware limit
+ * on old SoCs), so when needs to send ZLP, we intentionally trigger
+ * a length error interrupt, and in the ISR sends a ZLP by BMU.
+ */
+static void qmu_tx_zlp_error_handler(struct mtu3 *mtu, u8 epnum)
+{
+	struct mtu3_ep *mep = mtu->in_eps + epnum;
+	struct mtu3_gpd_ring *ring = &mep->gpd_ring;
+	void __iomem *mbase = mtu->mac_base;
+	struct qmu_gpd *gpd_current = NULL;
+	struct mtu3_request *mreq;
+	dma_addr_t cur_gpd_dma;
+	u32 txcsr = 0;
+	int ret;
+
+	mreq = next_request(mep);
+	if (mreq && mreq->request.length != 0)
+		return;
+
+	cur_gpd_dma = read_txq_cur_addr(mbase, epnum);
+	gpd_current = gpd_dma_to_virt(ring, cur_gpd_dma);
+
+	if (GPD_DATA_LEN(mtu, le32_to_cpu(gpd_current->dw3_info)) != 0) {
+		dev_err(mtu->dev, "TX EP%d buffer length error(!=0)\n", epnum);
+		return;
+	}
+
+	dev_dbg(mtu->dev, "%s send ZLP for req=%p\n", __func__, mreq);
+	trace_mtu3_zlp_exp_gpd(mep, gpd_current);
+
+	mtu3_clrbits(mbase, MU3D_EP_TXCR0(mep->epnum), TX_DMAREQEN);
+
+	ret = readl_poll_timeout_atomic(mbase + MU3D_EP_TXCR0(mep->epnum),
+			txcsr, !(txcsr & TX_FIFOFULL), 1, 1000);
+	if (ret) {
+		dev_err(mtu->dev, "%s wait for fifo empty fail\n", __func__);
+		return;
+	}
+	mtu3_setbits(mbase, MU3D_EP_TXCR0(mep->epnum), TX_TXPKTRDY);
+	/* prevent reorder, make sure GPD's HWO is set last */
+	mb();
+	/* by pass the current GDP */
+	gpd_current->dw0_info |= cpu_to_le32(GPD_FLAGS_BPS | GPD_FLAGS_HWO);
+
+	/*enable DMAREQEN, switch back to QMU mode */
+	mtu3_setbits(mbase, MU3D_EP_TXCR0(mep->epnum), TX_DMAREQEN);
+	mtu3_qmu_resume(mep);
+}
+
+/*
+ * NOTE: request list maybe is already empty as following case:
+ * queue_tx --> qmu_interrupt(clear interrupt pending, schedule tasklet)-->
+ * queue_tx --> process_tasklet(meanwhile, the second one is transferred,
+ * tasklet process both of them)-->qmu_interrupt for second one.
+ * To avoid upper case, put qmu_done_tx in ISR directly to process it.
+ */
+static void qmu_done_tx(struct mtu3 *mtu, u8 epnum)
+{
+	struct mtu3_ep *mep = mtu->in_eps + epnum;
+	struct mtu3_gpd_ring *ring = &mep->gpd_ring;
+	void __iomem *mbase = mtu->mac_base;
+	struct qmu_gpd *gpd = ring->dequeue;
+	struct qmu_gpd *gpd_current = NULL;
+	struct usb_request *request = NULL;
+	struct mtu3_request *mreq;
+	dma_addr_t cur_gpd_dma;
+
+	/*transfer phy address got from QMU register to virtual address */
+	cur_gpd_dma = read_txq_cur_addr(mbase, epnum);
+	gpd_current = gpd_dma_to_virt(ring, cur_gpd_dma);
+
+	dev_dbg(mtu->dev, "%s EP%d, last=%p, current=%p, enq=%p\n",
+		__func__, epnum, gpd, gpd_current, ring->enqueue);
+
+	while (gpd != gpd_current && !GET_GPD_HWO(gpd)) {
+
+		mreq = next_request(mep);
+
+		if (mreq == NULL || mreq->gpd != gpd) {
+			dev_err(mtu->dev, "no correct TX req is found\n");
+			break;
+		}
+
+		request = &mreq->request;
+		request->actual = GPD_DATA_LEN(mtu, le32_to_cpu(gpd->dw3_info));
+		trace_mtu3_complete_gpd(mep, gpd);
+		mtu3_req_complete(mep, request, 0);
+
+		gpd = advance_deq_gpd(ring);
+	}
+
+	dev_dbg(mtu->dev, "%s EP%d, deq=%p, enq=%p, complete\n",
+		__func__, epnum, ring->dequeue, ring->enqueue);
+
+}
+
+static void qmu_done_rx(struct mtu3 *mtu, u8 epnum)
+{
+	struct mtu3_ep *mep = mtu->out_eps + epnum;
+	struct mtu3_gpd_ring *ring = &mep->gpd_ring;
+	void __iomem *mbase = mtu->mac_base;
+	struct qmu_gpd *gpd = ring->dequeue;
+	struct qmu_gpd *gpd_current = NULL;
+	struct usb_request *req = NULL;
+	struct mtu3_request *mreq;
+	dma_addr_t cur_gpd_dma;
+
+	cur_gpd_dma = read_rxq_cur_addr(mbase, epnum);
+	gpd_current = gpd_dma_to_virt(ring, cur_gpd_dma);
+
+	dev_dbg(mtu->dev, "%s EP%d, last=%p, current=%p, enq=%p\n",
+		__func__, epnum, gpd, gpd_current, ring->enqueue);
+
+	while (gpd != gpd_current && !GET_GPD_HWO(gpd)) {
+
+		mreq = next_request(mep);
+
+		if (mreq == NULL || mreq->gpd != gpd) {
+			dev_err(mtu->dev, "no correct RX req is found\n");
+			break;
+		}
+		req = &mreq->request;
+
+		req->actual = GPD_DATA_LEN(mtu, le32_to_cpu(gpd->dw3_info));
+		trace_mtu3_complete_gpd(mep, gpd);
+		mtu3_req_complete(mep, req, 0);
+
+		gpd = advance_deq_gpd(ring);
+	}
+
+	dev_dbg(mtu->dev, "%s EP%d, deq=%p, enq=%p, complete\n",
+		__func__, epnum, ring->dequeue, ring->enqueue);
+}
+
+static void qmu_done_isr(struct mtu3 *mtu, u32 done_status)
+{
+	int i;
+
+	for (i = 1; i < mtu->num_eps; i++) {
+		if (done_status & QMU_RX_DONE_INT(i))
+			qmu_done_rx(mtu, i);
+		if (done_status & QMU_TX_DONE_INT(i))
+			qmu_done_tx(mtu, i);
+	}
+}
+
+static void qmu_exception_isr(struct mtu3 *mtu, u32 qmu_status)
+{
+	void __iomem *mbase = mtu->mac_base;
+	u32 errval;
+	int i;
+
+	if ((qmu_status & RXQ_CSERR_INT) || (qmu_status & RXQ_LENERR_INT)) {
+		errval = mtu3_readl(mbase, U3D_RQERRIR0);
+		for (i = 1; i < mtu->num_eps; i++) {
+			if (errval & QMU_RX_CS_ERR(i))
+				dev_err(mtu->dev, "Rx %d CS error!\n", i);
+
+			if (errval & QMU_RX_LEN_ERR(i))
+				dev_err(mtu->dev, "RX %d Length error\n", i);
+		}
+		mtu3_writel(mbase, U3D_RQERRIR0, errval);
+	}
+
+	if (qmu_status & RXQ_ZLPERR_INT) {
+		errval = mtu3_readl(mbase, U3D_RQERRIR1);
+		for (i = 1; i < mtu->num_eps; i++) {
+			if (errval & QMU_RX_ZLP_ERR(i))
+				dev_dbg(mtu->dev, "RX EP%d Recv ZLP\n", i);
+		}
+		mtu3_writel(mbase, U3D_RQERRIR1, errval);
+	}
+
+	if ((qmu_status & TXQ_CSERR_INT) || (qmu_status & TXQ_LENERR_INT)) {
+		errval = mtu3_readl(mbase, U3D_TQERRIR0);
+		for (i = 1; i < mtu->num_eps; i++) {
+			if (errval & QMU_TX_CS_ERR(i))
+				dev_err(mtu->dev, "Tx %d checksum error!\n", i);
+
+			if (errval & QMU_TX_LEN_ERR(i))
+				qmu_tx_zlp_error_handler(mtu, i);
+		}
+		mtu3_writel(mbase, U3D_TQERRIR0, errval);
+	}
+}
+
+irqreturn_t mtu3_qmu_isr(struct mtu3 *mtu)
+{
+	void __iomem *mbase = mtu->mac_base;
+	u32 qmu_status;
+	u32 qmu_done_status;
+
+	/* U3D_QISAR1 is read update */
+	qmu_status = mtu3_readl(mbase, U3D_QISAR1);
+	qmu_status &= mtu3_readl(mbase, U3D_QIER1);
+
+	qmu_done_status = mtu3_readl(mbase, U3D_QISAR0);
+	qmu_done_status &= mtu3_readl(mbase, U3D_QIER0);
+	mtu3_writel(mbase, U3D_QISAR0, qmu_done_status); /* W1C */
+	dev_dbg(mtu->dev, "=== QMUdone[tx=%x, rx=%x] QMUexp[%x] ===\n",
+		(qmu_done_status & 0xFFFF), qmu_done_status >> 16,
+		qmu_status);
+	trace_mtu3_qmu_isr(qmu_done_status, qmu_status);
+
+	if (qmu_done_status)
+		qmu_done_isr(mtu, qmu_done_status);
+
+	if (qmu_status)
+		qmu_exception_isr(mtu, qmu_status);
+
+	return IRQ_HANDLED;
+}
+
+int mtu3_qmu_init(struct mtu3 *mtu)
+{
+
+	compiletime_assert(QMU_GPD_SIZE == 16, "QMU_GPD size SHOULD be 16B");
+
+	mtu->qmu_gpd_pool = dma_pool_create("QMU_GPD", mtu->dev,
+			QMU_GPD_RING_SIZE, QMU_GPD_SIZE, 0);
+
+	if (!mtu->qmu_gpd_pool)
+		return -ENOMEM;
+
+	return 0;
+}
+
+void mtu3_qmu_exit(struct mtu3 *mtu)
+{
+	dma_pool_destroy(mtu->qmu_gpd_pool);
+}