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, 883 insertions, 0 deletions

diff --git a/drivers/media/platform/amphion/vpu_core.c b/drivers/media/platform/amphion/vpu_core.c
new file mode 100644
index 000000000..f9ec1753f
--- /dev/null
+++ b/drivers/media/platform/amphion/vpu_core.c
@@ -0,0 +1,883 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright 2020-2021 NXP
+ */
+
+#include <linux/init.h>
+#include <linux/interconnect.h>
+#include <linux/ioctl.h>
+#include <linux/list.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/of_address.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+#include <linux/pm_runtime.h>
+#include <linux/pm_domain.h>
+#include <linux/firmware.h>
+#include <linux/vmalloc.h>
+#include "vpu.h"
+#include "vpu_defs.h"
+#include "vpu_core.h"
+#include "vpu_mbox.h"
+#include "vpu_msgs.h"
+#include "vpu_rpc.h"
+#include "vpu_cmds.h"
+
+void csr_writel(struct vpu_core *core, u32 reg, u32 val)
+{
+	writel(val, core->base + reg);
+}
+
+u32 csr_readl(struct vpu_core *core, u32 reg)
+{
+	return readl(core->base + reg);
+}
+
+static int vpu_core_load_firmware(struct vpu_core *core)
+{
+	const struct firmware *pfw = NULL;
+	int ret = 0;
+
+	if (!core->fw.virt) {
+		dev_err(core->dev, "firmware buffer is not ready\n");
+		return -EINVAL;
+	}
+
+	ret = request_firmware(&pfw, core->res->fwname, core->dev);
+	dev_dbg(core->dev, "request_firmware %s : %d\n", core->res->fwname, ret);
+	if (ret) {
+		dev_err(core->dev, "request firmware %s failed, ret = %d\n",
+			core->res->fwname, ret);
+		return ret;
+	}
+
+	if (core->fw.length < pfw->size) {
+		dev_err(core->dev, "firmware buffer size want %zu, but %d\n",
+			pfw->size, core->fw.length);
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	memset(core->fw.virt, 0, core->fw.length);
+	memcpy(core->fw.virt, pfw->data, pfw->size);
+	core->fw.bytesused = pfw->size;
+	ret = vpu_iface_on_firmware_loaded(core);
+exit:
+	release_firmware(pfw);
+	pfw = NULL;
+
+	return ret;
+}
+
+static int vpu_core_boot_done(struct vpu_core *core)
+{
+	u32 fw_version;
+
+	fw_version = vpu_iface_get_version(core);
+	dev_info(core->dev, "%s firmware version : %d.%d.%d\n",
+		 vpu_core_type_desc(core->type),
+		 (fw_version >> 16) & 0xff,
+		 (fw_version >> 8) & 0xff,
+		 fw_version & 0xff);
+	core->supported_instance_count = vpu_iface_get_max_instance_count(core);
+	if (core->res->act_size) {
+		u32 count = core->act.length / core->res->act_size;
+
+		core->supported_instance_count = min(core->supported_instance_count, count);
+	}
+	core->fw_version = fw_version;
+	vpu_core_set_state(core, VPU_CORE_ACTIVE);
+
+	return 0;
+}
+
+static int vpu_core_wait_boot_done(struct vpu_core *core)
+{
+	int ret;
+
+	ret = wait_for_completion_timeout(&core->cmp, VPU_TIMEOUT);
+	if (!ret) {
+		dev_err(core->dev, "boot timeout\n");
+		return -EINVAL;
+	}
+	return vpu_core_boot_done(core);
+}
+
+static int vpu_core_boot(struct vpu_core *core, bool load)
+{
+	int ret;
+
+	reinit_completion(&core->cmp);
+	if (load) {
+		ret = vpu_core_load_firmware(core);
+		if (ret)
+			return ret;
+	}
+
+	vpu_iface_boot_core(core);
+	return vpu_core_wait_boot_done(core);
+}
+
+static int vpu_core_shutdown(struct vpu_core *core)
+{
+	return vpu_iface_shutdown_core(core);
+}
+
+static int vpu_core_restore(struct vpu_core *core)
+{
+	int ret;
+
+	ret = vpu_core_sw_reset(core);
+	if (ret)
+		return ret;
+
+	vpu_core_boot_done(core);
+	return vpu_iface_restore_core(core);
+}
+
+static int __vpu_alloc_dma(struct device *dev, struct vpu_buffer *buf)
+{
+	gfp_t gfp = GFP_KERNEL | GFP_DMA32;
+
+	if (!buf->length)
+		return 0;
+
+	buf->virt = dma_alloc_coherent(dev, buf->length, &buf->phys, gfp);
+	if (!buf->virt)
+		return -ENOMEM;
+
+	buf->dev = dev;
+
+	return 0;
+}
+
+void vpu_free_dma(struct vpu_buffer *buf)
+{
+	if (!buf->virt || !buf->dev)
+		return;
+
+	dma_free_coherent(buf->dev, buf->length, buf->virt, buf->phys);
+	buf->virt = NULL;
+	buf->phys = 0;
+	buf->length = 0;
+	buf->bytesused = 0;
+	buf->dev = NULL;
+}
+
+int vpu_alloc_dma(struct vpu_core *core, struct vpu_buffer *buf)
+{
+	return __vpu_alloc_dma(core->dev, buf);
+}
+
+void vpu_core_set_state(struct vpu_core *core, enum vpu_core_state state)
+{
+	if (state != core->state)
+		vpu_trace(core->dev, "vpu core state change from %d to %d\n", core->state, state);
+	core->state = state;
+	if (core->state == VPU_CORE_DEINIT)
+		core->hang_mask = 0;
+}
+
+static void vpu_core_update_state(struct vpu_core *core)
+{
+	if (!vpu_iface_get_power_state(core)) {
+		if (core->request_count)
+			vpu_core_set_state(core, VPU_CORE_HANG);
+		else
+			vpu_core_set_state(core, VPU_CORE_DEINIT);
+
+	} else if (core->state == VPU_CORE_ACTIVE && core->hang_mask) {
+		vpu_core_set_state(core, VPU_CORE_HANG);
+	}
+}
+
+static struct vpu_core *vpu_core_find_proper_by_type(struct vpu_dev *vpu, u32 type)
+{
+	struct vpu_core *core = NULL;
+	int request_count = INT_MAX;
+	struct vpu_core *c;
+
+	list_for_each_entry(c, &vpu->cores, list) {
+		dev_dbg(c->dev, "instance_mask = 0x%lx, state = %d\n", c->instance_mask, c->state);
+		if (c->type != type)
+			continue;
+		mutex_lock(&c->lock);
+		vpu_core_update_state(c);
+		mutex_unlock(&c->lock);
+		if (c->state == VPU_CORE_DEINIT) {
+			core = c;
+			break;
+		}
+		if (c->state != VPU_CORE_ACTIVE)
+			continue;
+		if (c->request_count < request_count) {
+			request_count = c->request_count;
+			core = c;
+		}
+		if (!request_count)
+			break;
+	}
+
+	return core;
+}
+
+static bool vpu_core_is_exist(struct vpu_dev *vpu, struct vpu_core *core)
+{
+	struct vpu_core *c;
+
+	list_for_each_entry(c, &vpu->cores, list) {
+		if (c == core)
+			return true;
+	}
+
+	return false;
+}
+
+static void vpu_core_get_vpu(struct vpu_core *core)
+{
+	core->vpu->get_vpu(core->vpu);
+	if (core->type == VPU_CORE_TYPE_ENC)
+		core->vpu->get_enc(core->vpu);
+	if (core->type == VPU_CORE_TYPE_DEC)
+		core->vpu->get_dec(core->vpu);
+}
+
+static int vpu_core_register(struct device *dev, struct vpu_core *core)
+{
+	struct vpu_dev *vpu = dev_get_drvdata(dev);
+	int ret = 0;
+
+	dev_dbg(core->dev, "register core %s\n", vpu_core_type_desc(core->type));
+	if (vpu_core_is_exist(vpu, core))
+		return 0;
+
+	core->workqueue = alloc_workqueue("vpu", WQ_UNBOUND | WQ_MEM_RECLAIM, 1);
+	if (!core->workqueue) {
+		dev_err(core->dev, "fail to alloc workqueue\n");
+		return -ENOMEM;
+	}
+	INIT_WORK(&core->msg_work, vpu_msg_run_work);
+	INIT_DELAYED_WORK(&core->msg_delayed_work, vpu_msg_delayed_work);
+	core->msg_buffer_size = roundup_pow_of_two(VPU_MSG_BUFFER_SIZE);
+	core->msg_buffer = vzalloc(core->msg_buffer_size);
+	if (!core->msg_buffer) {
+		dev_err(core->dev, "failed allocate buffer for fifo\n");
+		ret = -ENOMEM;
+		goto error;
+	}
+	ret = kfifo_init(&core->msg_fifo, core->msg_buffer, core->msg_buffer_size);
+	if (ret) {
+		dev_err(core->dev, "failed init kfifo\n");
+		goto error;
+	}
+
+	list_add_tail(&core->list, &vpu->cores);
+	vpu_core_get_vpu(core);
+
+	return 0;
+error:
+	if (core->msg_buffer) {
+		vfree(core->msg_buffer);
+		core->msg_buffer = NULL;
+	}
+	if (core->workqueue) {
+		destroy_workqueue(core->workqueue);
+		core->workqueue = NULL;
+	}
+	return ret;
+}
+
+static void vpu_core_put_vpu(struct vpu_core *core)
+{
+	if (core->type == VPU_CORE_TYPE_ENC)
+		core->vpu->put_enc(core->vpu);
+	if (core->type == VPU_CORE_TYPE_DEC)
+		core->vpu->put_dec(core->vpu);
+	core->vpu->put_vpu(core->vpu);
+}
+
+static int vpu_core_unregister(struct device *dev, struct vpu_core *core)
+{
+	list_del_init(&core->list);
+
+	vpu_core_put_vpu(core);
+	core->vpu = NULL;
+	vfree(core->msg_buffer);
+	core->msg_buffer = NULL;
+
+	if (core->workqueue) {
+		cancel_work_sync(&core->msg_work);
+		cancel_delayed_work_sync(&core->msg_delayed_work);
+		destroy_workqueue(core->workqueue);
+		core->workqueue = NULL;
+	}
+
+	return 0;
+}
+
+static int vpu_core_acquire_instance(struct vpu_core *core)
+{
+	int id;
+
+	id = ffz(core->instance_mask);
+	if (id >= core->supported_instance_count)
+		return -EINVAL;
+
+	set_bit(id, &core->instance_mask);
+
+	return id;
+}
+
+static void vpu_core_release_instance(struct vpu_core *core, int id)
+{
+	if (id < 0 || id >= core->supported_instance_count)
+		return;
+
+	clear_bit(id, &core->instance_mask);
+}
+
+struct vpu_inst *vpu_inst_get(struct vpu_inst *inst)
+{
+	if (!inst)
+		return NULL;
+
+	atomic_inc(&inst->ref_count);
+
+	return inst;
+}
+
+void vpu_inst_put(struct vpu_inst *inst)
+{
+	if (!inst)
+		return;
+	if (atomic_dec_and_test(&inst->ref_count)) {
+		if (inst->release)
+			inst->release(inst);
+	}
+}
+
+struct vpu_core *vpu_request_core(struct vpu_dev *vpu, enum vpu_core_type type)
+{
+	struct vpu_core *core = NULL;
+	int ret;
+
+	mutex_lock(&vpu->lock);
+
+	core = vpu_core_find_proper_by_type(vpu, type);
+	if (!core)
+		goto exit;
+
+	mutex_lock(&core->lock);
+	pm_runtime_resume_and_get(core->dev);
+
+	if (core->state == VPU_CORE_DEINIT) {
+		if (vpu_iface_get_power_state(core))
+			ret = vpu_core_restore(core);
+		else
+			ret = vpu_core_boot(core, true);
+		if (ret) {
+			pm_runtime_put_sync(core->dev);
+			mutex_unlock(&core->lock);
+			core = NULL;
+			goto exit;
+		}
+	}
+
+	core->request_count++;
+
+	mutex_unlock(&core->lock);
+exit:
+	mutex_unlock(&vpu->lock);
+
+	return core;
+}
+
+void vpu_release_core(struct vpu_core *core)
+{
+	if (!core)
+		return;
+
+	mutex_lock(&core->lock);
+	pm_runtime_put_sync(core->dev);
+	if (core->request_count)
+		core->request_count--;
+	mutex_unlock(&core->lock);
+}
+
+int vpu_inst_register(struct vpu_inst *inst)
+{
+	struct vpu_dev *vpu;
+	struct vpu_core *core;
+	int ret = 0;
+
+	vpu = inst->vpu;
+	core = inst->core;
+	if (!core) {
+		core = vpu_request_core(vpu, inst->type);
+		if (!core) {
+			dev_err(vpu->dev, "there is no vpu core for %s\n",
+				vpu_core_type_desc(inst->type));
+			return -EINVAL;
+		}
+		inst->core = core;
+		inst->dev = get_device(core->dev);
+	}
+
+	mutex_lock(&core->lock);
+	if (core->state != VPU_CORE_ACTIVE) {
+		dev_err(core->dev, "vpu core is not active, state = %d\n", core->state);
+		ret = -EINVAL;
+		goto exit;
+	}
+
+	if (inst->id >= 0 && inst->id < core->supported_instance_count)
+		goto exit;
+
+	ret = vpu_core_acquire_instance(core);
+	if (ret < 0)
+		goto exit;
+
+	vpu_trace(inst->dev, "[%d] %p\n", ret, inst);
+	inst->id = ret;
+	list_add_tail(&inst->list, &core->instances);
+	ret = 0;
+	if (core->res->act_size) {
+		inst->act.phys = core->act.phys + core->res->act_size * inst->id;
+		inst->act.virt = core->act.virt + core->res->act_size * inst->id;
+		inst->act.length = core->res->act_size;
+	}
+	vpu_inst_create_dbgfs_file(inst);
+exit:
+	mutex_unlock(&core->lock);
+
+	if (ret)
+		dev_err(core->dev, "register instance fail\n");
+	return ret;
+}
+
+int vpu_inst_unregister(struct vpu_inst *inst)
+{
+	struct vpu_core *core;
+
+	if (!inst->core)
+		return 0;
+
+	core = inst->core;
+	vpu_clear_request(inst);
+	mutex_lock(&core->lock);
+	if (inst->id >= 0 && inst->id < core->supported_instance_count) {
+		vpu_inst_remove_dbgfs_file(inst);
+		list_del_init(&inst->list);
+		vpu_core_release_instance(core, inst->id);
+		inst->id = VPU_INST_NULL_ID;
+	}
+	vpu_core_update_state(core);
+	if (core->state == VPU_CORE_HANG && !core->instance_mask) {
+		int err;
+
+		dev_info(core->dev, "reset hang core\n");
+		mutex_unlock(&core->lock);
+		err = vpu_core_sw_reset(core);
+		mutex_lock(&core->lock);
+		if (!err) {
+			vpu_core_set_state(core, VPU_CORE_ACTIVE);
+			core->hang_mask = 0;
+		}
+	}
+	mutex_unlock(&core->lock);
+
+	return 0;
+}
+
+struct vpu_inst *vpu_core_find_instance(struct vpu_core *core, u32 index)
+{
+	struct vpu_inst *inst = NULL;
+	struct vpu_inst *tmp;
+
+	mutex_lock(&core->lock);
+	if (index >= core->supported_instance_count || !test_bit(index, &core->instance_mask))
+		goto exit;
+	list_for_each_entry(tmp, &core->instances, list) {
+		if (tmp->id == index) {
+			inst = vpu_inst_get(tmp);
+			break;
+		}
+	}
+exit:
+	mutex_unlock(&core->lock);
+
+	return inst;
+}
+
+const struct vpu_core_resources *vpu_get_resource(struct vpu_inst *inst)
+{
+	struct vpu_dev *vpu;
+	struct vpu_core *core = NULL;
+	const struct vpu_core_resources *res = NULL;
+
+	if (!inst || !inst->vpu)
+		return NULL;
+
+	if (inst->core && inst->core->res)
+		return inst->core->res;
+
+	vpu = inst->vpu;
+	mutex_lock(&vpu->lock);
+	list_for_each_entry(core, &vpu->cores, list) {
+		if (core->type == inst->type) {
+			res = core->res;
+			break;
+		}
+	}
+	mutex_unlock(&vpu->lock);
+
+	return res;
+}
+
+static int vpu_core_parse_dt(struct vpu_core *core, struct device_node *np)
+{
+	struct device_node *node;
+	struct resource res;
+	int ret;
+
+	if (of_count_phandle_with_args(np, "memory-region", NULL) < 2) {
+		dev_err(core->dev, "need 2 memory-region for boot and rpc\n");
+		return -ENODEV;
+	}
+
+	node = of_parse_phandle(np, "memory-region", 0);
+	if (!node) {
+		dev_err(core->dev, "boot-region of_parse_phandle error\n");
+		return -ENODEV;
+	}
+	if (of_address_to_resource(node, 0, &res)) {
+		dev_err(core->dev, "boot-region of_address_to_resource error\n");
+		of_node_put(node);
+		return -EINVAL;
+	}
+	core->fw.phys = res.start;
+	core->fw.length = resource_size(&res);
+
+	of_node_put(node);
+
+	node = of_parse_phandle(np, "memory-region", 1);
+	if (!node) {
+		dev_err(core->dev, "rpc-region of_parse_phandle error\n");
+		return -ENODEV;
+	}
+	if (of_address_to_resource(node, 0, &res)) {
+		dev_err(core->dev, "rpc-region of_address_to_resource error\n");
+		of_node_put(node);
+		return -EINVAL;
+	}
+	core->rpc.phys = res.start;
+	core->rpc.length = resource_size(&res);
+
+	if (core->rpc.length < core->res->rpc_size + core->res->fwlog_size) {
+		dev_err(core->dev, "the rpc-region <%pad, 0x%x> is not enough\n",
+			&core->rpc.phys, core->rpc.length);
+		of_node_put(node);
+		return -EINVAL;
+	}
+
+	core->fw.virt = memremap(core->fw.phys, core->fw.length, MEMREMAP_WC);
+	core->rpc.virt = memremap(core->rpc.phys, core->rpc.length, MEMREMAP_WC);
+	memset(core->rpc.virt, 0, core->rpc.length);
+
+	ret = vpu_iface_check_memory_region(core, core->rpc.phys, core->rpc.length);
+	if (ret != VPU_CORE_MEMORY_UNCACHED) {
+		dev_err(core->dev, "rpc region<%pad, 0x%x> isn't uncached\n",
+			&core->rpc.phys, core->rpc.length);
+		of_node_put(node);
+		return -EINVAL;
+	}
+
+	core->log.phys = core->rpc.phys + core->res->rpc_size;
+	core->log.virt = core->rpc.virt + core->res->rpc_size;
+	core->log.length = core->res->fwlog_size;
+	core->act.phys = core->log.phys + core->log.length;
+	core->act.virt = core->log.virt + core->log.length;
+	core->act.length = core->rpc.length - core->res->rpc_size - core->log.length;
+	core->rpc.length = core->res->rpc_size;
+
+	of_node_put(node);
+
+	return 0;
+}
+
+static int vpu_core_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct vpu_core *core;
+	struct vpu_dev *vpu = dev_get_drvdata(dev->parent);
+	struct vpu_shared_addr *iface;
+	u32 iface_data_size;
+	int ret;
+
+	dev_dbg(dev, "probe\n");
+	if (!vpu)
+		return -EINVAL;
+	core = devm_kzalloc(dev, sizeof(*core), GFP_KERNEL);
+	if (!core)
+		return -ENOMEM;
+
+	core->pdev = pdev;
+	core->dev = dev;
+	platform_set_drvdata(pdev, core);
+	core->vpu = vpu;
+	INIT_LIST_HEAD(&core->instances);
+	mutex_init(&core->lock);
+	mutex_init(&core->cmd_lock);
+	init_completion(&core->cmp);
+	init_waitqueue_head(&core->ack_wq);
+	vpu_core_set_state(core, VPU_CORE_DEINIT);
+
+	core->res = of_device_get_match_data(dev);
+	if (!core->res)
+		return -ENODEV;
+
+	core->type = core->res->type;
+	core->id = of_alias_get_id(dev->of_node, "vpu_core");
+	if (core->id < 0) {
+		dev_err(dev, "can't get vpu core id\n");
+		return core->id;
+	}
+	dev_info(core->dev, "[%d] = %s\n", core->id, vpu_core_type_desc(core->type));
+	ret = vpu_core_parse_dt(core, dev->of_node);
+	if (ret)
+		return ret;
+
+	core->base = devm_platform_ioremap_resource(pdev, 0);
+	if (IS_ERR(core->base))
+		return PTR_ERR(core->base);
+
+	if (!vpu_iface_check_codec(core)) {
+		dev_err(core->dev, "is not supported\n");
+		return -EINVAL;
+	}
+
+	ret = vpu_mbox_init(core);
+	if (ret)
+		return ret;
+
+	iface = devm_kzalloc(dev, sizeof(*iface), GFP_KERNEL);
+	if (!iface)
+		return -ENOMEM;
+
+	iface_data_size = vpu_iface_get_data_size(core);
+	if (iface_data_size) {
+		iface->priv = devm_kzalloc(dev, iface_data_size, GFP_KERNEL);
+		if (!iface->priv)
+			return -ENOMEM;
+	}
+
+	ret = vpu_iface_init(core, iface, &core->rpc, core->fw.phys);
+	if (ret) {
+		dev_err(core->dev, "init iface fail, ret = %d\n", ret);
+		return ret;
+	}
+
+	vpu_iface_config_system(core, vpu->res->mreg_base, vpu->base);
+	vpu_iface_set_log_buf(core, &core->log);
+
+	pm_runtime_enable(dev);
+	ret = pm_runtime_resume_and_get(dev);
+	if (ret) {
+		pm_runtime_put_noidle(dev);
+		pm_runtime_set_suspended(dev);
+		goto err_runtime_disable;
+	}
+
+	ret = vpu_core_register(dev->parent, core);
+	if (ret)
+		goto err_core_register;
+	core->parent = dev->parent;
+
+	pm_runtime_put_sync(dev);
+	vpu_core_create_dbgfs_file(core);
+
+	return 0;
+
+err_core_register:
+	pm_runtime_put_sync(dev);
+err_runtime_disable:
+	pm_runtime_disable(dev);
+
+	return ret;
+}
+
+static int vpu_core_remove(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct vpu_core *core = platform_get_drvdata(pdev);
+	int ret;
+
+	vpu_core_remove_dbgfs_file(core);
+	ret = pm_runtime_resume_and_get(dev);
+	WARN_ON(ret < 0);
+
+	vpu_core_shutdown(core);
+	pm_runtime_put_sync(dev);
+	pm_runtime_disable(dev);
+
+	vpu_core_unregister(core->parent, core);
+	memunmap(core->fw.virt);
+	memunmap(core->rpc.virt);
+	mutex_destroy(&core->lock);
+	mutex_destroy(&core->cmd_lock);
+
+	return 0;
+}
+
+static int __maybe_unused vpu_core_runtime_resume(struct device *dev)
+{
+	struct vpu_core *core = dev_get_drvdata(dev);
+
+	return vpu_mbox_request(core);
+}
+
+static int __maybe_unused vpu_core_runtime_suspend(struct device *dev)
+{
+	struct vpu_core *core = dev_get_drvdata(dev);
+
+	vpu_mbox_free(core);
+	return 0;
+}
+
+static void vpu_core_cancel_work(struct vpu_core *core)
+{
+	struct vpu_inst *inst = NULL;
+
+	cancel_work_sync(&core->msg_work);
+	cancel_delayed_work_sync(&core->msg_delayed_work);
+
+	mutex_lock(&core->lock);
+	list_for_each_entry(inst, &core->instances, list)
+		cancel_work_sync(&inst->msg_work);
+	mutex_unlock(&core->lock);
+}
+
+static void vpu_core_resume_work(struct vpu_core *core)
+{
+	struct vpu_inst *inst = NULL;
+	unsigned long delay = msecs_to_jiffies(10);
+
+	queue_work(core->workqueue, &core->msg_work);
+	queue_delayed_work(core->workqueue, &core->msg_delayed_work, delay);
+
+	mutex_lock(&core->lock);
+	list_for_each_entry(inst, &core->instances, list)
+		queue_work(inst->workqueue, &inst->msg_work);
+	mutex_unlock(&core->lock);
+}
+
+static int __maybe_unused vpu_core_resume(struct device *dev)
+{
+	struct vpu_core *core = dev_get_drvdata(dev);
+	int ret = 0;
+
+	mutex_lock(&core->lock);
+	pm_runtime_resume_and_get(dev);
+	vpu_core_get_vpu(core);
+
+	if (core->request_count) {
+		if (!vpu_iface_get_power_state(core))
+			ret = vpu_core_boot(core, false);
+		else
+			ret = vpu_core_sw_reset(core);
+		if (ret) {
+			dev_err(core->dev, "resume fail\n");
+			vpu_core_set_state(core, VPU_CORE_HANG);
+		}
+	}
+	vpu_core_update_state(core);
+	pm_runtime_put_sync(dev);
+	mutex_unlock(&core->lock);
+
+	vpu_core_resume_work(core);
+	return ret;
+}
+
+static int __maybe_unused vpu_core_suspend(struct device *dev)
+{
+	struct vpu_core *core = dev_get_drvdata(dev);
+	int ret = 0;
+
+	mutex_lock(&core->lock);
+	if (core->request_count)
+		ret = vpu_core_snapshot(core);
+	mutex_unlock(&core->lock);
+	if (ret)
+		return ret;
+
+	vpu_core_cancel_work(core);
+
+	mutex_lock(&core->lock);
+	vpu_core_put_vpu(core);
+	mutex_unlock(&core->lock);
+	return ret;
+}
+
+static const struct dev_pm_ops vpu_core_pm_ops = {
+	SET_RUNTIME_PM_OPS(vpu_core_runtime_suspend, vpu_core_runtime_resume, NULL)
+	SET_SYSTEM_SLEEP_PM_OPS(vpu_core_suspend, vpu_core_resume)
+};
+
+static struct vpu_core_resources imx8q_enc = {
+	.type = VPU_CORE_TYPE_ENC,
+	.fwname = "vpu/vpu_fw_imx8_enc.bin",
+	.stride = 16,
+	.max_width = 1920,
+	.max_height = 1920,
+	.min_width = 64,
+	.min_height = 48,
+	.step_width = 2,
+	.step_height = 2,
+	.rpc_size = 0x80000,
+	.fwlog_size = 0x80000,
+	.act_size = 0xc0000,
+};
+
+static struct vpu_core_resources imx8q_dec = {
+	.type = VPU_CORE_TYPE_DEC,
+	.fwname = "vpu/vpu_fw_imx8_dec.bin",
+	.stride = 256,
+	.max_width = 8188,
+	.max_height = 8188,
+	.min_width = 16,
+	.min_height = 16,
+	.step_width = 1,
+	.step_height = 1,
+	.rpc_size = 0x80000,
+	.fwlog_size = 0x80000,
+};
+
+static const struct of_device_id vpu_core_dt_match[] = {
+	{ .compatible = "nxp,imx8q-vpu-encoder", .data = &imx8q_enc },
+	{ .compatible = "nxp,imx8q-vpu-decoder", .data = &imx8q_dec },
+	{}
+};
+MODULE_DEVICE_TABLE(of, vpu_core_dt_match);
+
+static struct platform_driver amphion_vpu_core_driver = {
+	.probe = vpu_core_probe,
+	.remove = vpu_core_remove,
+	.driver = {
+		.name = "amphion-vpu-core",
+		.of_match_table = vpu_core_dt_match,
+		.pm = &vpu_core_pm_ops,
+	},
+};
+
+int __init vpu_core_driver_init(void)
+{
+	return platform_driver_register(&amphion_vpu_core_driver);
+}
+
+void __exit vpu_core_driver_exit(void)
+{
+	platform_driver_unregister(&amphion_vpu_core_driver);
+}