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, 579 insertions, 0 deletions

diff --git a/drivers/platform/chrome/wilco_ec/event.c b/drivers/platform/chrome/wilco_ec/event.c
new file mode 100644
index 000000000..69ceead8c
--- /dev/null
+++ b/drivers/platform/chrome/wilco_ec/event.c
@@ -0,0 +1,579 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * ACPI event handling for Wilco Embedded Controller
+ *
+ * Copyright 2019 Google LLC
+ *
+ * The Wilco Embedded Controller can create custom events that
+ * are not handled as standard ACPI objects. These events can
+ * contain information about changes in EC controlled features,
+ * such as errors and events in the dock or display. For example,
+ * an event is triggered if the dock is plugged into a display
+ * incorrectly. These events are needed for telemetry and
+ * diagnostics reasons, and for possibly alerting the user.
+
+ * These events are triggered by the EC with an ACPI Notify(0x90),
+ * and then the BIOS reads the event buffer from EC RAM via an
+ * ACPI method. When the OS receives these events via ACPI,
+ * it passes them along to this driver. The events are put into
+ * a queue which can be read by a userspace daemon via a char device
+ * that implements read() and poll(). The event queue acts as a
+ * circular buffer of size 64, so if there are no userspace consumers
+ * the kernel will not run out of memory. The char device will appear at
+ * /dev/wilco_event{n}, where n is some small non-negative integer,
+ * starting from 0. Standard ACPI events such as the battery getting
+ * plugged/unplugged can also come through this path, but they are
+ * dealt with via other paths, and are ignored here.
+
+ * To test, you can tail the binary data with
+ * $ cat /dev/wilco_event0 | hexdump -ve '1/1 "%x\n"'
+ * and then create an event by plugging/unplugging the battery.
+ */
+
+#include <linux/acpi.h>
+#include <linux/cdev.h>
+#include <linux/device.h>
+#include <linux/fs.h>
+#include <linux/idr.h>
+#include <linux/io.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/poll.h>
+#include <linux/spinlock.h>
+#include <linux/uaccess.h>
+#include <linux/wait.h>
+
+/* ACPI Notify event code indicating event data is available. */
+#define EC_ACPI_NOTIFY_EVENT		0x90
+/* ACPI Method to execute to retrieve event data buffer from the EC. */
+#define EC_ACPI_GET_EVENT		"QSET"
+/* Maximum number of words in event data returned by the EC. */
+#define EC_ACPI_MAX_EVENT_WORDS		6
+#define EC_ACPI_MAX_EVENT_SIZE \
+	(sizeof(struct ec_event) + (EC_ACPI_MAX_EVENT_WORDS) * sizeof(u16))
+
+/* Node will appear in /dev/EVENT_DEV_NAME */
+#define EVENT_DEV_NAME		"wilco_event"
+#define EVENT_CLASS_NAME	EVENT_DEV_NAME
+#define DRV_NAME		EVENT_DEV_NAME
+#define EVENT_DEV_NAME_FMT	(EVENT_DEV_NAME "%d")
+static struct class event_class = {
+	.owner	= THIS_MODULE,
+	.name	= EVENT_CLASS_NAME,
+};
+
+/* Keep track of all the device numbers used. */
+#define EVENT_MAX_DEV 128
+static int event_major;
+static DEFINE_IDA(event_ida);
+
+/* Size of circular queue of events. */
+#define MAX_NUM_EVENTS 64
+
+/**
+ * struct ec_event - Extended event returned by the EC.
+ * @size: Number of 16bit words in structure after the size word.
+ * @type: Extended event type, meaningless for us.
+ * @event: Event data words.  Max count is %EC_ACPI_MAX_EVENT_WORDS.
+ */
+struct ec_event {
+	u16 size;
+	u16 type;
+	u16 event[];
+} __packed;
+
+#define ec_event_num_words(ev) (ev->size - 1)
+#define ec_event_size(ev) (sizeof(*ev) + (ec_event_num_words(ev) * sizeof(u16)))
+
+/**
+ * struct ec_event_queue - Circular queue for events.
+ * @capacity: Number of elements the queue can hold.
+ * @head: Next index to write to.
+ * @tail: Next index to read from.
+ * @entries: Array of events.
+ */
+struct ec_event_queue {
+	int capacity;
+	int head;
+	int tail;
+	struct ec_event *entries[];
+};
+
+/* Maximum number of events to store in ec_event_queue */
+static int queue_size = 64;
+module_param(queue_size, int, 0644);
+
+static struct ec_event_queue *event_queue_new(int capacity)
+{
+	struct ec_event_queue *q;
+
+	q = kzalloc(struct_size(q, entries, capacity), GFP_KERNEL);
+	if (!q)
+		return NULL;
+
+	q->capacity = capacity;
+
+	return q;
+}
+
+static inline bool event_queue_empty(struct ec_event_queue *q)
+{
+	/* head==tail when both full and empty, but head==NULL when empty */
+	return q->head == q->tail && !q->entries[q->head];
+}
+
+static inline bool event_queue_full(struct ec_event_queue *q)
+{
+	/* head==tail when both full and empty, but head!=NULL when full */
+	return q->head == q->tail && q->entries[q->head];
+}
+
+static struct ec_event *event_queue_pop(struct ec_event_queue *q)
+{
+	struct ec_event *ev;
+
+	if (event_queue_empty(q))
+		return NULL;
+
+	ev = q->entries[q->tail];
+	q->entries[q->tail] = NULL;
+	q->tail = (q->tail + 1) % q->capacity;
+
+	return ev;
+}
+
+/*
+ * If full, overwrite the oldest event and return it so the caller
+ * can kfree it. If not full, return NULL.
+ */
+static struct ec_event *event_queue_push(struct ec_event_queue *q,
+					 struct ec_event *ev)
+{
+	struct ec_event *popped = NULL;
+
+	if (event_queue_full(q))
+		popped = event_queue_pop(q);
+	q->entries[q->head] = ev;
+	q->head = (q->head + 1) % q->capacity;
+
+	return popped;
+}
+
+static void event_queue_free(struct ec_event_queue *q)
+{
+	struct ec_event *event;
+
+	while ((event = event_queue_pop(q)) != NULL)
+		kfree(event);
+
+	kfree(q);
+}
+
+/**
+ * struct event_device_data - Data for a Wilco EC device that responds to ACPI.
+ * @events: Circular queue of EC events to be provided to userspace.
+ * @queue_lock: Protect the queue from simultaneous read/writes.
+ * @wq: Wait queue to notify processes when events are available or the
+ *	device has been removed.
+ * @cdev: Char dev that userspace reads() and polls() from.
+ * @dev: Device associated with the %cdev.
+ * @exist: Has the device been not been removed? Once a device has been removed,
+ *	   writes, reads, and new opens will fail.
+ * @available: Guarantee only one client can open() file and read from queue.
+ *
+ * There will be one of these structs for each ACPI device registered. This data
+ * is the queue of events received from ACPI that still need to be read from
+ * userspace, the device and char device that userspace is using, a wait queue
+ * used to notify different threads when something has changed, plus a flag
+ * on whether the ACPI device has been removed.
+ */
+struct event_device_data {
+	struct ec_event_queue *events;
+	spinlock_t queue_lock;
+	wait_queue_head_t wq;
+	struct device dev;
+	struct cdev cdev;
+	bool exist;
+	atomic_t available;
+};
+
+/**
+ * enqueue_events() - Place EC events in queue to be read by userspace.
+ * @adev: Device the events came from.
+ * @buf: Buffer of event data.
+ * @length: Length of event data buffer.
+ *
+ * %buf contains a number of ec_event's, packed one after the other.
+ * Each ec_event is of variable length. Start with the first event, copy it
+ * into a persistent ec_event, store that entry in the queue, move on
+ * to the next ec_event in buf, and repeat.
+ *
+ * Return: 0 on success or negative error code on failure.
+ */
+static int enqueue_events(struct acpi_device *adev, const u8 *buf, u32 length)
+{
+	struct event_device_data *dev_data = adev->driver_data;
+	struct ec_event *event, *queue_event, *old_event;
+	size_t num_words, event_size;
+	u32 offset = 0;
+
+	while (offset < length) {
+		event = (struct ec_event *)(buf + offset);
+
+		num_words = ec_event_num_words(event);
+		event_size = ec_event_size(event);
+		if (num_words > EC_ACPI_MAX_EVENT_WORDS) {
+			dev_err(&adev->dev, "Too many event words: %zu > %d\n",
+				num_words, EC_ACPI_MAX_EVENT_WORDS);
+			return -EOVERFLOW;
+		}
+
+		/* Ensure event does not overflow the available buffer */
+		if ((offset + event_size) > length) {
+			dev_err(&adev->dev, "Event exceeds buffer: %zu > %d\n",
+				offset + event_size, length);
+			return -EOVERFLOW;
+		}
+
+		/* Point to the next event in the buffer */
+		offset += event_size;
+
+		/* Copy event into the queue */
+		queue_event = kmemdup(event, event_size, GFP_KERNEL);
+		if (!queue_event)
+			return -ENOMEM;
+		spin_lock(&dev_data->queue_lock);
+		old_event = event_queue_push(dev_data->events, queue_event);
+		spin_unlock(&dev_data->queue_lock);
+		kfree(old_event);
+		wake_up_interruptible(&dev_data->wq);
+	}
+
+	return 0;
+}
+
+/**
+ * event_device_notify() - Callback when EC generates an event over ACPI.
+ * @adev: The device that the event is coming from.
+ * @value: Value passed to Notify() in ACPI.
+ *
+ * This function will read the events from the device and enqueue them.
+ */
+static void event_device_notify(struct acpi_device *adev, u32 value)
+{
+	struct acpi_buffer event_buffer = { ACPI_ALLOCATE_BUFFER, NULL };
+	union acpi_object *obj;
+	acpi_status status;
+
+	if (value != EC_ACPI_NOTIFY_EVENT) {
+		dev_err(&adev->dev, "Invalid event: 0x%08x\n", value);
+		return;
+	}
+
+	/* Execute ACPI method to get event data buffer. */
+	status = acpi_evaluate_object(adev->handle, EC_ACPI_GET_EVENT,
+				      NULL, &event_buffer);
+	if (ACPI_FAILURE(status)) {
+		dev_err(&adev->dev, "Error executing ACPI method %s()\n",
+			EC_ACPI_GET_EVENT);
+		return;
+	}
+
+	obj = (union acpi_object *)event_buffer.pointer;
+	if (!obj) {
+		dev_err(&adev->dev, "Nothing returned from %s()\n",
+			EC_ACPI_GET_EVENT);
+		return;
+	}
+	if (obj->type != ACPI_TYPE_BUFFER) {
+		dev_err(&adev->dev, "Invalid object returned from %s()\n",
+			EC_ACPI_GET_EVENT);
+		kfree(obj);
+		return;
+	}
+	if (obj->buffer.length < sizeof(struct ec_event)) {
+		dev_err(&adev->dev, "Invalid buffer length %d from %s()\n",
+			obj->buffer.length, EC_ACPI_GET_EVENT);
+		kfree(obj);
+		return;
+	}
+
+	enqueue_events(adev, obj->buffer.pointer, obj->buffer.length);
+	kfree(obj);
+}
+
+static int event_open(struct inode *inode, struct file *filp)
+{
+	struct event_device_data *dev_data;
+
+	dev_data = container_of(inode->i_cdev, struct event_device_data, cdev);
+	if (!dev_data->exist)
+		return -ENODEV;
+
+	if (atomic_cmpxchg(&dev_data->available, 1, 0) == 0)
+		return -EBUSY;
+
+	/* Increase refcount on device so dev_data is not freed */
+	get_device(&dev_data->dev);
+	stream_open(inode, filp);
+	filp->private_data = dev_data;
+
+	return 0;
+}
+
+static __poll_t event_poll(struct file *filp, poll_table *wait)
+{
+	struct event_device_data *dev_data = filp->private_data;
+	__poll_t mask = 0;
+
+	poll_wait(filp, &dev_data->wq, wait);
+	if (!dev_data->exist)
+		return EPOLLHUP;
+	if (!event_queue_empty(dev_data->events))
+		mask |= EPOLLIN | EPOLLRDNORM | EPOLLPRI;
+	return mask;
+}
+
+/**
+ * event_read() - Callback for passing event data to userspace via read().
+ * @filp: The file we are reading from.
+ * @buf: Pointer to userspace buffer to fill with one event.
+ * @count: Number of bytes requested. Must be at least EC_ACPI_MAX_EVENT_SIZE.
+ * @pos: File position pointer, irrelevant since we don't support seeking.
+ *
+ * Removes the first event from the queue, places it in the passed buffer.
+ *
+ * If there are no events in the queue, then one of two things happens,
+ * depending on if the file was opened in nonblocking mode: If in nonblocking
+ * mode, then return -EAGAIN to say there's no data. If in blocking mode, then
+ * block until an event is available.
+ *
+ * Return: Number of bytes placed in buffer, negative error code on failure.
+ */
+static ssize_t event_read(struct file *filp, char __user *buf, size_t count,
+			  loff_t *pos)
+{
+	struct event_device_data *dev_data = filp->private_data;
+	struct ec_event *event;
+	ssize_t n_bytes_written = 0;
+	int err;
+
+	/* We only will give them the entire event at once */
+	if (count != 0 && count < EC_ACPI_MAX_EVENT_SIZE)
+		return -EINVAL;
+
+	spin_lock(&dev_data->queue_lock);
+	while (event_queue_empty(dev_data->events)) {
+		spin_unlock(&dev_data->queue_lock);
+		if (filp->f_flags & O_NONBLOCK)
+			return -EAGAIN;
+
+		err = wait_event_interruptible(dev_data->wq,
+					!event_queue_empty(dev_data->events) ||
+					!dev_data->exist);
+		if (err)
+			return err;
+
+		/* Device was removed as we waited? */
+		if (!dev_data->exist)
+			return -ENODEV;
+		spin_lock(&dev_data->queue_lock);
+	}
+	event = event_queue_pop(dev_data->events);
+	spin_unlock(&dev_data->queue_lock);
+	n_bytes_written = ec_event_size(event);
+	if (copy_to_user(buf, event, n_bytes_written))
+		n_bytes_written = -EFAULT;
+	kfree(event);
+
+	return n_bytes_written;
+}
+
+static int event_release(struct inode *inode, struct file *filp)
+{
+	struct event_device_data *dev_data = filp->private_data;
+
+	atomic_set(&dev_data->available, 1);
+	put_device(&dev_data->dev);
+
+	return 0;
+}
+
+static const struct file_operations event_fops = {
+	.open = event_open,
+	.poll  = event_poll,
+	.read = event_read,
+	.release = event_release,
+	.llseek = no_llseek,
+	.owner = THIS_MODULE,
+};
+
+/**
+ * free_device_data() - Callback to free the event_device_data structure.
+ * @d: The device embedded in our device data, which we have been ref counting.
+ *
+ * This is called only after event_device_remove() has been called and all
+ * userspace programs have called event_release() on all the open file
+ * descriptors.
+ */
+static void free_device_data(struct device *d)
+{
+	struct event_device_data *dev_data;
+
+	dev_data = container_of(d, struct event_device_data, dev);
+	event_queue_free(dev_data->events);
+	kfree(dev_data);
+}
+
+static void hangup_device(struct event_device_data *dev_data)
+{
+	dev_data->exist = false;
+	/* Wake up the waiting processes so they can close. */
+	wake_up_interruptible(&dev_data->wq);
+	put_device(&dev_data->dev);
+}
+
+/**
+ * event_device_add() - Callback when creating a new device.
+ * @adev: ACPI device that we will be receiving events from.
+ *
+ * This finds a free minor number for the device, allocates and initializes
+ * some device data, and creates a new device and char dev node.
+ *
+ * The device data is freed in free_device_data(), which is called when
+ * %dev_data->dev is release()ed. This happens after all references to
+ * %dev_data->dev are dropped, which happens once both event_device_remove()
+ * has been called and every open()ed file descriptor has been release()ed.
+ *
+ * Return: 0 on success, negative error code on failure.
+ */
+static int event_device_add(struct acpi_device *adev)
+{
+	struct event_device_data *dev_data;
+	int error, minor;
+
+	minor = ida_alloc_max(&event_ida, EVENT_MAX_DEV-1, GFP_KERNEL);
+	if (minor < 0) {
+		error = minor;
+		dev_err(&adev->dev, "Failed to find minor number: %d\n", error);
+		return error;
+	}
+
+	dev_data = kzalloc(sizeof(*dev_data), GFP_KERNEL);
+	if (!dev_data) {
+		error = -ENOMEM;
+		goto free_minor;
+	}
+
+	/* Initialize the device data. */
+	adev->driver_data = dev_data;
+	dev_data->events = event_queue_new(queue_size);
+	if (!dev_data->events) {
+		kfree(dev_data);
+		error = -ENOMEM;
+		goto free_minor;
+	}
+	spin_lock_init(&dev_data->queue_lock);
+	init_waitqueue_head(&dev_data->wq);
+	dev_data->exist = true;
+	atomic_set(&dev_data->available, 1);
+
+	/* Initialize the device. */
+	dev_data->dev.devt = MKDEV(event_major, minor);
+	dev_data->dev.class = &event_class;
+	dev_data->dev.release = free_device_data;
+	dev_set_name(&dev_data->dev, EVENT_DEV_NAME_FMT, minor);
+	device_initialize(&dev_data->dev);
+
+	/* Initialize the character device, and add it to userspace. */
+	cdev_init(&dev_data->cdev, &event_fops);
+	error = cdev_device_add(&dev_data->cdev, &dev_data->dev);
+	if (error)
+		goto free_dev_data;
+
+	return 0;
+
+free_dev_data:
+	hangup_device(dev_data);
+free_minor:
+	ida_simple_remove(&event_ida, minor);
+	return error;
+}
+
+static void event_device_remove(struct acpi_device *adev)
+{
+	struct event_device_data *dev_data = adev->driver_data;
+
+	cdev_device_del(&dev_data->cdev, &dev_data->dev);
+	ida_simple_remove(&event_ida, MINOR(dev_data->dev.devt));
+	hangup_device(dev_data);
+}
+
+static const struct acpi_device_id event_acpi_ids[] = {
+	{ "GOOG000D", 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(acpi, event_acpi_ids);
+
+static struct acpi_driver event_driver = {
+	.name = DRV_NAME,
+	.class = DRV_NAME,
+	.ids = event_acpi_ids,
+	.ops = {
+		.add = event_device_add,
+		.notify = event_device_notify,
+		.remove = event_device_remove,
+	},
+	.owner = THIS_MODULE,
+};
+
+static int __init event_module_init(void)
+{
+	dev_t dev_num = 0;
+	int ret;
+
+	ret = class_register(&event_class);
+	if (ret) {
+		pr_err(DRV_NAME ": Failed registering class: %d\n", ret);
+		return ret;
+	}
+
+	/* Request device numbers, starting with minor=0. Save the major num. */
+	ret = alloc_chrdev_region(&dev_num, 0, EVENT_MAX_DEV, EVENT_DEV_NAME);
+	if (ret) {
+		pr_err(DRV_NAME ": Failed allocating dev numbers: %d\n", ret);
+		goto destroy_class;
+	}
+	event_major = MAJOR(dev_num);
+
+	ret = acpi_bus_register_driver(&event_driver);
+	if (ret < 0) {
+		pr_err(DRV_NAME ": Failed registering driver: %d\n", ret);
+		goto unregister_region;
+	}
+
+	return 0;
+
+unregister_region:
+	unregister_chrdev_region(MKDEV(event_major, 0), EVENT_MAX_DEV);
+destroy_class:
+	class_unregister(&event_class);
+	ida_destroy(&event_ida);
+	return ret;
+}
+
+static void __exit event_module_exit(void)
+{
+	acpi_bus_unregister_driver(&event_driver);
+	unregister_chrdev_region(MKDEV(event_major, 0), EVENT_MAX_DEV);
+	class_unregister(&event_class);
+	ida_destroy(&event_ida);
+}
+
+module_init(event_module_init);
+module_exit(event_module_exit);
+
+MODULE_AUTHOR("Nick Crews <ncrews@chromium.org>");
+MODULE_DESCRIPTION("Wilco EC ACPI event driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:" DRV_NAME);