linux/drivers/platform/raspberrypi/vchiq-interface/vchiq_arm.c

// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
/*
 * Copyright (c) 2014 Raspberry Pi (Trading) Ltd. All rights reserved.
 * Copyright (c) 2010-2012 Broadcom. All rights reserved.
 */

#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/sched/signal.h>
#include <linux/types.h>
#include <linux/errno.h>
#include <linux/cdev.h>
#include <linux/fs.h>
#include <linux/device.h>
#include <linux/device/bus.h>
#include <linux/mm.h>
#include <linux/pagemap.h>
#include <linux/bug.h>
#include <linux/completion.h>
#include <linux/list.h>
#include <linux/of.h>
#include <linux/platform_device.h>
#include <linux/compat.h>
#include <linux/dma-mapping.h>
#include <linux/rcupdate.h>
#include <linux/delay.h>
#include <linux/slab.h>
#include <linux/interrupt.h>
#include <linux/io.h>
#include <linux/uaccess.h>
#include <soc/bcm2835/raspberrypi-firmware.h>

#include <linux/raspberrypi/vchiq_core.h>
#include <linux/raspberrypi/vchiq_arm.h>
#include <linux/raspberrypi/vchiq_bus.h>
#include <linux/raspberrypi/vchiq_debugfs.h>

#include "vchiq_ioctl.h"

#define DEVICE_NAME "vchiq"

#define TOTAL_SLOTS (VCHIQ_SLOT_ZERO_SLOTS + 2 * 32)

#define MAX_FRAGMENTS (VCHIQ_NUM_CURRENT_BULKS * 2)

#define VCHIQ_PLATFORM_FRAGMENTS_OFFSET_IDX 0
#define VCHIQ_PLATFORM_FRAGMENTS_COUNT_IDX  1

#define BELL0	0x00

#define ARM_DS_ACTIVE	BIT(2)

/* Override the default prefix, which would be vchiq_arm (from the filename) */
#undef MODULE_PARAM_PREFIX
#define MODULE_PARAM_PREFIX DEVICE_NAME "."

#define KEEPALIVE_VER 1
#define KEEPALIVE_VER_MIN KEEPALIVE_VER

/*
 * The devices implemented in the VCHIQ firmware are not discoverable,
 * so we need to maintain a list of them in order to register them with
 * the interface.
 */
static struct vchiq_device *bcm2835_audio;

static const struct vchiq_platform_info bcm2835_info = {
	.cache_line_size = 32,
};

static const struct vchiq_platform_info bcm2836_info = {
	.cache_line_size = 64,
};

struct vchiq_arm_state {
	/*
	 * Keepalive-related data
	 *
	 * The keepalive mechanism was retro-fitted to VCHIQ to allow active
	 * services to prevent the system from suspending.
	 * This feature is not used on Raspberry Pi devices.
	 */
	struct task_struct *ka_thread;
	struct completion ka_evt;
	atomic_t ka_use_count;
	atomic_t ka_use_ack_count;
	atomic_t ka_release_count;

	rwlock_t susp_res_lock;

	struct vchiq_state *state;

	/*
	 * Global use count for videocore.
	 * This is equal to the sum of the use counts for all services.  When
	 * this hits zero the videocore suspend procedure will be initiated.
	 */
	int videocore_use_count;

	/*
	 * Use count to track requests from videocore peer.
	 * This use count is not associated with a service, so needs to be
	 * tracked separately with the state.
	 */
	int peer_use_count;

	/*
	 * Flag to indicate that the first vchiq connect has made it through.
	 * This means that both sides should be fully ready, and we should
	 * be able to suspend after this point.
	 */
	int first_connect;
};

static int
vchiq_blocking_bulk_transfer(struct vchiq_instance *instance, unsigned int handle,
			     struct vchiq_bulk *bulk_params);

static irqreturn_t
vchiq_doorbell_irq(int irq, void *dev_id)
{
	struct vchiq_state *state = dev_id;
	struct vchiq_drv_mgmt *mgmt;
	irqreturn_t ret = IRQ_NONE;
	unsigned int status;

	mgmt = dev_get_drvdata(state->dev);

	/* Read (and clear) the doorbell */
	status = readl(mgmt->regs + BELL0);

	if (status & ARM_DS_ACTIVE) {  /* Was the doorbell rung? */
		remote_event_pollall(state);
		ret = IRQ_HANDLED;
	}

	return ret;
}

/*
 * This function is called by the vchiq stack once it has been connected to
 * the videocore and clients can start to use the stack.
 */
static void vchiq_call_connected_callbacks(struct vchiq_drv_mgmt *drv_mgmt)
{
	int i;

	if (mutex_lock_killable(&drv_mgmt->connected_mutex))
		return;

	for (i = 0; i < drv_mgmt->num_deferred_callbacks; i++)
		drv_mgmt->deferred_callback[i]();

	drv_mgmt->num_deferred_callbacks = 0;
	drv_mgmt->connected = true;
	mutex_unlock(&drv_mgmt->connected_mutex);
}

/*
 * This function is used to defer initialization until the vchiq stack is
 * initialized. If the stack is already initialized, then the callback will
 * be made immediately, otherwise it will be deferred until
 * vchiq_call_connected_callbacks is called.
 */
void vchiq_add_connected_callback(struct vchiq_device *device, void (*callback)(void))
{
	struct vchiq_drv_mgmt *drv_mgmt = device->drv_mgmt;

	if (mutex_lock_killable(&drv_mgmt->connected_mutex))
		return;

	if (drv_mgmt->connected) {
		/* We're already connected. Call the callback immediately. */
		callback();
	} else {
		if (drv_mgmt->num_deferred_callbacks >= VCHIQ_DRV_MAX_CALLBACKS) {
			dev_err(&device->dev,
				"core: deferred callbacks(%d) exceeded the maximum limit(%d)\n",
				drv_mgmt->num_deferred_callbacks, VCHIQ_DRV_MAX_CALLBACKS);
		} else {
			drv_mgmt->deferred_callback[drv_mgmt->num_deferred_callbacks] =
				callback;
			drv_mgmt->num_deferred_callbacks++;
		}
	}
	mutex_unlock(&drv_mgmt->connected_mutex);
}
EXPORT_SYMBOL(vchiq_add_connected_callback);

static int vchiq_platform_init(struct platform_device *pdev, struct vchiq_state *state)
{
	struct device *dev = &pdev->dev;
	struct vchiq_drv_mgmt *drv_mgmt = platform_get_drvdata(pdev);
	struct rpi_firmware *fw = drv_mgmt->fw;
	struct vchiq_slot_zero *vchiq_slot_zero;
	void *slot_mem;
	dma_addr_t slot_phys;
	u32 channelbase;
	int slot_mem_size, frag_mem_size;
	int err, irq, i;

	/*
	 * VCHI messages between the CPU and firmware use
	 * 32-bit bus addresses.
	 */
	err = dma_set_mask_and_coherent(dev, DMA_BIT_MASK(32));

	if (err < 0)
		return err;

	drv_mgmt->fragments_size = 2 * drv_mgmt->info->cache_line_size;

	/* Allocate space for the channels in coherent memory */
	slot_mem_size = PAGE_ALIGN(TOTAL_SLOTS * VCHIQ_SLOT_SIZE);
	frag_mem_size = PAGE_ALIGN(drv_mgmt->fragments_size * MAX_FRAGMENTS);

	slot_mem = dmam_alloc_coherent(dev, slot_mem_size + frag_mem_size,
				       &slot_phys, GFP_KERNEL);
	if (!slot_mem) {
		dev_err(dev, "could not allocate DMA memory\n");
		return -ENOMEM;
	}

	WARN_ON(((unsigned long)slot_mem & (PAGE_SIZE - 1)) != 0);

	vchiq_slot_zero = vchiq_init_slots(dev, slot_mem, slot_mem_size);
	if (!vchiq_slot_zero)
		return -ENOMEM;

	vchiq_slot_zero->platform_data[VCHIQ_PLATFORM_FRAGMENTS_OFFSET_IDX] =
		(int)slot_phys + slot_mem_size;
	vchiq_slot_zero->platform_data[VCHIQ_PLATFORM_FRAGMENTS_COUNT_IDX] =
		MAX_FRAGMENTS;

	drv_mgmt->fragments_base = (char *)slot_mem + slot_mem_size;

	drv_mgmt->free_fragments = drv_mgmt->fragments_base;
	for (i = 0; i < (MAX_FRAGMENTS - 1); i++) {
		*(char **)&drv_mgmt->fragments_base[i * drv_mgmt->fragments_size] =
			&drv_mgmt->fragments_base[(i + 1) * drv_mgmt->fragments_size];
	}
	*(char **)&drv_mgmt->fragments_base[i * drv_mgmt->fragments_size] = NULL;
	sema_init(&drv_mgmt->free_fragments_sema, MAX_FRAGMENTS);
	sema_init(&drv_mgmt->free_fragments_mutex, 1);

	err = vchiq_init_state(state, vchiq_slot_zero, dev);
	if (err)
		return err;

	drv_mgmt->regs = devm_platform_ioremap_resource(pdev, 0);
	if (IS_ERR(drv_mgmt->regs))
		return PTR_ERR(drv_mgmt->regs);

	irq = platform_get_irq(pdev, 0);
	if (irq <= 0)
		return irq;

	err = devm_request_irq(dev, irq, vchiq_doorbell_irq, IRQF_IRQPOLL,
			       "VCHIQ doorbell", state);
	if (err) {
		dev_err(dev, "failed to register irq=%d\n", irq);
		return err;
	}

	/* Send the base address of the slots to VideoCore */
	channelbase = slot_phys;
	err = rpi_firmware_property(fw, RPI_FIRMWARE_VCHIQ_INIT,
				    &channelbase, sizeof(channelbase));
	if (err) {
		dev_err(dev, "failed to send firmware property: %d\n", err);
		return err;
	}

	if (channelbase) {
		dev_err(dev, "failed to set channelbase (response: %x)\n",
			channelbase);
		return -ENXIO;
	}

	dev_dbg(&pdev->dev, "arm: vchiq_init - done (slots %p, phys %pad)\n",
		vchiq_slot_zero, &slot_phys);

	mutex_init(&drv_mgmt->connected_mutex);
	vchiq_call_connected_callbacks(drv_mgmt);

	return 0;
}

int
vchiq_platform_init_state(struct vchiq_state *state)
{
	struct vchiq_arm_state *platform_state;

	platform_state = devm_kzalloc(state->dev, sizeof(*platform_state), GFP_KERNEL);
	if (!platform_state)
		return -ENOMEM;

	rwlock_init(&platform_state->susp_res_lock);

	init_completion(&platform_state->ka_evt);
	atomic_set(&platform_state->ka_use_count, 0);
	atomic_set(&platform_state->ka_use_ack_count, 0);
	atomic_set(&platform_state->ka_release_count, 0);

	platform_state->state = state;

	state->platform_state = (struct opaque_platform_state *)platform_state;

	return 0;
}

static struct vchiq_arm_state *vchiq_platform_get_arm_state(struct vchiq_state *state)
{
	return (struct vchiq_arm_state *)state->platform_state;
}

static void
vchiq_platform_uninit(struct vchiq_drv_mgmt *mgmt)
{
	struct vchiq_arm_state *arm_state;

	kthread_stop(mgmt->state.sync_thread);
	kthread_stop(mgmt->state.recycle_thread);
	kthread_stop(mgmt->state.slot_handler_thread);

	arm_state = vchiq_platform_get_arm_state(&mgmt->state);
	if (!IS_ERR_OR_NULL(arm_state->ka_thread))
		kthread_stop(arm_state->ka_thread);
}

void vchiq_dump_platform_state(struct seq_file *f)
{
	seq_puts(f, "  Platform: 2835 (VC master)\n");
}

#define VCHIQ_INIT_RETRIES 10
int vchiq_initialise(struct vchiq_state *state, struct vchiq_instance **instance_out)
{
	struct vchiq_instance *instance = NULL;
	int i, ret;

	/*
	 * VideoCore may not be ready due to boot up timing.
	 * It may never be ready if kernel and firmware are mismatched,so don't
	 * block forever.
	 */
	for (i = 0; i < VCHIQ_INIT_RETRIES; i++) {
		if (vchiq_remote_initialised(state))
			break;
		usleep_range(500, 600);
	}
	if (i == VCHIQ_INIT_RETRIES) {
		dev_err(state->dev, "core: %s: Videocore not initialized\n", __func__);
		ret = -ENOTCONN;
		goto failed;
	} else if (i > 0) {
		dev_warn(state->dev, "core: %s: videocore initialized after %d retries\n",
			 __func__, i);
	}

	instance = kzalloc(sizeof(*instance), GFP_KERNEL);
	if (!instance) {
		ret = -ENOMEM;
		goto failed;
	}

	instance->connected = 0;
	instance->state = state;
	mutex_init(&instance->bulk_waiter_list_mutex);
	INIT_LIST_HEAD(&instance->bulk_waiter_list);

	*instance_out = instance;

	ret = 0;

failed:
	dev_dbg(state->dev, "core: (%p): returning %d\n", instance, ret);

	return ret;
}
EXPORT_SYMBOL(vchiq_initialise);

void free_bulk_waiter(struct vchiq_instance *instance)
{
	struct bulk_waiter_node *waiter, *next;

	list_for_each_entry_safe(waiter, next,
				 &instance->bulk_waiter_list, list) {
		list_del(&waiter->list);
		dev_dbg(instance->state->dev,
			"arm: bulk_waiter - cleaned up %p for pid %d\n",
			waiter, waiter->pid);
		kfree(waiter);
	}
}

int vchiq_shutdown(struct vchiq_instance *instance)
{
	struct vchiq_state *state = instance->state;
	int ret = 0;

	mutex_lock(&state->mutex);

	/* Remove all services */
	vchiq_shutdown_internal(state, instance);

	mutex_unlock(&state->mutex);

	dev_dbg(state->dev, "core: (%p): returning %d\n", instance, ret);

	free_bulk_waiter(instance);
	kfree(instance);

	return ret;
}
EXPORT_SYMBOL(vchiq_shutdown);

static int vchiq_is_connected(struct vchiq_instance *instance)
{
	return instance->connected;
}

int vchiq_connect(struct vchiq_instance *instance)
{
	struct vchiq_state *state = instance->state;
	int ret;

	if (mutex_lock_killable(&state->mutex)) {
		dev_dbg(state->dev,
			"core: call to mutex_lock failed\n");
		ret = -EAGAIN;
		goto failed;
	}
	ret = vchiq_connect_internal(state, instance);

	if (!ret)
		instance->connected = 1;

	mutex_unlock(&state->mutex);

failed:
	dev_dbg(state->dev, "core: (%p): returning %d\n", instance, ret);

	return ret;
}
EXPORT_SYMBOL(vchiq_connect);

static int
vchiq_add_service(struct vchiq_instance *instance,
		  const struct vchiq_service_params_kernel *params,
		  unsigned int *phandle)
{
	struct vchiq_state *state = instance->state;
	struct vchiq_service *service = NULL;
	int srvstate, ret;

	*phandle = VCHIQ_SERVICE_HANDLE_INVALID;

	srvstate = vchiq_is_connected(instance)
		? VCHIQ_SRVSTATE_LISTENING
		: VCHIQ_SRVSTATE_HIDDEN;

	service = vchiq_add_service_internal(state, params, srvstate, instance, NULL);

	if (service) {
		*phandle = service->handle;
		ret = 0;
	} else {
		ret = -EINVAL;
	}

	dev_dbg(state->dev, "core: (%p): returning %d\n", instance, ret);

	return ret;
}

int
vchiq_open_service(struct vchiq_instance *instance,
		   const struct vchiq_service_params_kernel *params,
		   unsigned int *phandle)
{
	struct vchiq_state   *state = instance->state;
	struct vchiq_service *service = NULL;
	int ret = -EINVAL;

	*phandle = VCHIQ_SERVICE_HANDLE_INVALID;

	if (!vchiq_is_connected(instance))
		goto failed;

	service = vchiq_add_service_internal(state, params, VCHIQ_SRVSTATE_OPENING, instance, NULL);

	if (service) {
		*phandle = service->handle;
		ret = vchiq_open_service_internal(service, current->pid);
		if (ret) {
			vchiq_remove_service(instance, service->handle);
			*phandle = VCHIQ_SERVICE_HANDLE_INVALID;
		}
	}

failed:
	dev_dbg(state->dev, "core: (%p): returning %d\n", instance, ret);

	return ret;
}
EXPORT_SYMBOL(vchiq_open_service);

int
vchiq_bulk_transmit(struct vchiq_instance *instance, unsigned int handle, const void *data,
		    unsigned int size, void *userdata, enum vchiq_bulk_mode mode)
{
	struct vchiq_bulk bulk_params = {};
	int ret;

	switch (mode) {
	case VCHIQ_BULK_MODE_NOCALLBACK:
	case VCHIQ_BULK_MODE_CALLBACK:

		bulk_params.offset = (void *)data;
		bulk_params.mode = mode;
		bulk_params.size = size;
		bulk_params.cb_data = userdata;
		bulk_params.dir = VCHIQ_BULK_TRANSMIT;

		ret = vchiq_bulk_xfer_callback(instance, handle, &bulk_params);
		break;
	case VCHIQ_BULK_MODE_BLOCKING:
		bulk_params.offset = (void *)data;
		bulk_params.mode = mode;
		bulk_params.size = size;
		bulk_params.dir = VCHIQ_BULK_TRANSMIT;

		ret = vchiq_blocking_bulk_transfer(instance, handle, &bulk_params);
		break;
	default:
		return -EINVAL;
	}

	return ret;
}
EXPORT_SYMBOL(vchiq_bulk_transmit);

int vchiq_bulk_receive(struct vchiq_instance *instance, unsigned int handle,
		       void *data, unsigned int size, void *userdata,
		       enum vchiq_bulk_mode mode)
{
	struct vchiq_bulk bulk_params = {};
	int ret;

	switch (mode) {
	case VCHIQ_BULK_MODE_NOCALLBACK:
	case VCHIQ_BULK_MODE_CALLBACK:

		bulk_params.offset = (void *)data;
		bulk_params.mode = mode;
		bulk_params.size = size;
		bulk_params.cb_data = userdata;
		bulk_params.dir = VCHIQ_BULK_RECEIVE;

		ret = vchiq_bulk_xfer_callback(instance, handle, &bulk_params);
		break;
	case VCHIQ_BULK_MODE_BLOCKING:
		bulk_params.offset = (void *)data;
		bulk_params.mode = mode;
		bulk_params.size = size;
		bulk_params.dir = VCHIQ_BULK_RECEIVE;

		ret = vchiq_blocking_bulk_transfer(instance, handle, &bulk_params);
		break;
	default:
		return -EINVAL;
	}

	return ret;
}
EXPORT_SYMBOL(vchiq_bulk_receive);

static int
vchiq_blocking_bulk_transfer(struct vchiq_instance *instance, unsigned int handle,
			     struct vchiq_bulk *bulk_params)
{
	struct vchiq_service *service;
	struct bulk_waiter_node *waiter = NULL, *iter;
	int ret;

	service = find_service_by_handle(instance, handle);
	if (!service)
		return -EINVAL;

	vchiq_service_put(service);

	mutex_lock(&instance->bulk_waiter_list_mutex);
	list_for_each_entry(iter, &instance->bulk_waiter_list, list) {
		if (iter->pid == current->pid) {
			list_del(&iter->list);
			waiter = iter;
			break;
		}
	}
	mutex_unlock(&instance->bulk_waiter_list_mutex);

	if (waiter) {
		struct vchiq_bulk *bulk = waiter->bulk_waiter.bulk;

		if (bulk) {
			/* This thread has an outstanding bulk transfer. */
			/* FIXME: why compare a dma address to a pointer? */
			if ((bulk->dma_addr != (dma_addr_t)(uintptr_t)bulk_params->dma_addr) ||
			    (bulk->size != bulk_params->size)) {
				/*
				 * This is not a retry of the previous one.
				 * Cancel the signal when the transfer completes.
				 */
				spin_lock(&service->state->bulk_waiter_spinlock);
				bulk->waiter = NULL;
				spin_unlock(&service->state->bulk_waiter_spinlock);
			}
		}
	} else {
		waiter = kzalloc(sizeof(*waiter), GFP_KERNEL);
		if (!waiter)
			return -ENOMEM;
	}

	bulk_params->waiter = &waiter->bulk_waiter;

	ret = vchiq_bulk_xfer_blocking(instance, handle, bulk_params);
	if ((ret != -EAGAIN) || fatal_signal_pending(current) || !waiter->bulk_waiter.bulk) {
		struct vchiq_bulk *bulk = waiter->bulk_waiter.bulk;

		if (bulk) {
			/* Cancel the signal when the transfer completes. */
			spin_lock(&service->state->bulk_waiter_spinlock);
			bulk->waiter = NULL;
			spin_unlock(&service->state->bulk_waiter_spinlock);
		}
		kfree(waiter);
	} else {
		waiter->pid = current->pid;
		mutex_lock(&instance->bulk_waiter_list_mutex);
		list_add(&waiter->list, &instance->bulk_waiter_list);
		mutex_unlock(&instance->bulk_waiter_list_mutex);
		dev_dbg(instance->state->dev, "arm: saved bulk_waiter %p for pid %d\n",
			waiter, current->pid);
	}

	return ret;
}

static int
add_completion(struct vchiq_instance *instance, enum vchiq_reason reason,
	       struct vchiq_header *header, struct user_service *user_service,
	       void *cb_data, void __user *cb_userdata)
{
	struct vchiq_completion_data_kernel *completion;
	struct vchiq_drv_mgmt *mgmt = dev_get_drvdata(instance->state->dev);
	int insert;

	DEBUG_INITIALISE(mgmt->state.local);

	insert = instance->completion_insert;
	while ((insert - instance->completion_remove) >= MAX_COMPLETIONS) {
		/* Out of space - wait for the client */
		DEBUG_TRACE(SERVICE_CALLBACK_LINE);
		dev_dbg(instance->state->dev, "core: completion queue full\n");
		DEBUG_COUNT(COMPLETION_QUEUE_FULL_COUNT);
		if (wait_for_completion_interruptible(&instance->remove_event)) {
			dev_dbg(instance->state->dev, "arm: service_callback interrupted\n");
			return -EAGAIN;
		} else if (instance->closing) {
			dev_dbg(instance->state->dev, "arm: service_callback closing\n");
			return 0;
		}
		DEBUG_TRACE(SERVICE_CALLBACK_LINE);
	}

	completion = &instance->completions[insert & (MAX_COMPLETIONS - 1)];

	completion->header = header;
	completion->reason = reason;
	/* N.B. service_userdata is updated while processing AWAIT_COMPLETION */
	completion->service_userdata = user_service->service;
	completion->cb_data = cb_data;
	completion->cb_userdata = cb_userdata;

	if (reason == VCHIQ_SERVICE_CLOSED) {
		/*
		 * Take an extra reference, to be held until
		 * this CLOSED notification is delivered.
		 */
		vchiq_service_get(user_service->service);
		if (instance->use_close_delivered)
			user_service->close_pending = 1;
	}

	/*
	 * A write barrier is needed here to ensure that the entire completion
	 * record is written out before the insert point.
	 */
	wmb();

	if (reason == VCHIQ_MESSAGE_AVAILABLE)
		user_service->message_available_pos = insert;

	insert++;
	instance->completion_insert = insert;

	complete(&instance->insert_event);

	return 0;
}

static int
service_single_message(struct vchiq_instance *instance,
		       enum vchiq_reason reason, struct vchiq_service *service,
		       void *cb_data, void __user *cb_userdata)
{
	struct user_service *user_service;

	user_service = (struct user_service *)service->base.userdata;

	dev_dbg(service->state->dev, "arm: msg queue full\n");
	/*
	 * If there is no MESSAGE_AVAILABLE in the completion
	 * queue, add one
	 */
	if ((user_service->message_available_pos -
	     instance->completion_remove) < 0) {
		int ret;

		dev_dbg(instance->state->dev,
			"arm: Inserting extra MESSAGE_AVAILABLE\n");
		ret = add_completion(instance, reason, NULL, user_service,
				     cb_data, cb_userdata);
		if (ret)
			return ret;
	}

	if (wait_for_completion_interruptible(&user_service->remove_event)) {
		dev_dbg(instance->state->dev, "arm: interrupted\n");
		return -EAGAIN;
	} else if (instance->closing) {
		dev_dbg(instance->state->dev, "arm: closing\n");
		return -EINVAL;
	}

	return 0;
}

int
service_callback(struct vchiq_instance *instance, enum vchiq_reason reason,
		 struct vchiq_header *header, unsigned int handle,
		 void *cb_data, void __user *cb_userdata)
{
	/*
	 * How do we ensure the callback goes to the right client?
	 * The service_user data points to a user_service record
	 * containing the original callback and the user state structure, which
	 * contains a circular buffer for completion records.
	 */
	struct vchiq_drv_mgmt *mgmt = dev_get_drvdata(instance->state->dev);
	struct user_service *user_service;
	struct vchiq_service *service;
	bool skip_completion = false;

	DEBUG_INITIALISE(mgmt->state.local);

	DEBUG_TRACE(SERVICE_CALLBACK_LINE);

	rcu_read_lock();
	service = handle_to_service(instance, handle);
	if (WARN_ON(!service)) {
		rcu_read_unlock();
		return 0;
	}

	user_service = (struct user_service *)service->base.userdata;

	if (instance->closing) {
		rcu_read_unlock();
		return 0;
	}

	/*
	 * As hopping around different synchronization mechanism,
	 * taking an extra reference results in simpler implementation.
	 */
	vchiq_service_get(service);
	rcu_read_unlock();

	dev_dbg(service->state->dev,
		"arm: service %p(%d,%p), reason %d, header %p, instance %p, cb_data %p, cb_userdata %p\n",
		user_service, service->localport, user_service->userdata,
		reason, header, instance, cb_data, cb_userdata);

	if (header && user_service->is_vchi) {
		spin_lock(&service->state->msg_queue_spinlock);
		while (user_service->msg_insert ==
			(user_service->msg_remove + MSG_QUEUE_SIZE)) {
			int ret;

			spin_unlock(&service->state->msg_queue_spinlock);
			DEBUG_TRACE(SERVICE_CALLBACK_LINE);
			DEBUG_COUNT(MSG_QUEUE_FULL_COUNT);

			ret = service_single_message(instance, reason, service,
						     cb_data, cb_userdata);
			if (ret) {
				DEBUG_TRACE(SERVICE_CALLBACK_LINE);
				vchiq_service_put(service);
				return ret;
			}
			DEBUG_TRACE(SERVICE_CALLBACK_LINE);
			spin_lock(&service->state->msg_queue_spinlock);
		}

		user_service->msg_queue[user_service->msg_insert &
			(MSG_QUEUE_SIZE - 1)] = header;
		user_service->msg_insert++;

		/*
		 * If there is a thread waiting in DEQUEUE_MESSAGE, or if
		 * there is a MESSAGE_AVAILABLE in the completion queue then
		 * bypass the completion queue.
		 */
		if (((user_service->message_available_pos -
			instance->completion_remove) >= 0) ||
			user_service->dequeue_pending) {
			user_service->dequeue_pending = 0;
			skip_completion = true;
		}

		spin_unlock(&service->state->msg_queue_spinlock);
		complete(&user_service->insert_event);

		header = NULL;
	}
	DEBUG_TRACE(SERVICE_CALLBACK_LINE);
	vchiq_service_put(service);

	if (skip_completion)
		return 0;

	return add_completion(instance, reason, header, user_service,
			      cb_data, cb_userdata);
}

void vchiq_dump_platform_instances(struct vchiq_state *state, struct seq_file *f)
{
	int i;

	if (!vchiq_remote_initialised(state))
		return;

	/*
	 * There is no list of instances, so instead scan all services,
	 * marking those that have been dumped.
	 */

	rcu_read_lock();
	for (i = 0; i < state->unused_service; i++) {
		struct vchiq_service *service;
		struct vchiq_instance *instance;

		service = rcu_dereference(state->services[i]);
		if (!service || service->base.callback != service_callback)
			continue;

		instance = service->instance;
		if (instance)
			instance->mark = 0;
	}
	rcu_read_unlock();

	for (i = 0; i < state->unused_service; i++) {
		struct vchiq_service *service;
		struct vchiq_instance *instance;

		rcu_read_lock();
		service = rcu_dereference(state->services[i]);
		if (!service || service->base.callback != service_callback) {
			rcu_read_unlock();
			continue;
		}

		instance = service->instance;
		if (!instance || instance->mark) {
			rcu_read_unlock();
			continue;
		}
		rcu_read_unlock();

		seq_printf(f, "Instance %pK: pid %d,%s completions %d/%d\n",
			   instance, instance->pid,
			   instance->connected ? " connected, " :
			   "",
			   instance->completion_insert -
			   instance->completion_remove,
			   MAX_COMPLETIONS);
		instance->mark = 1;
	}
}

void vchiq_dump_platform_service_state(struct seq_file *f,
				       struct vchiq_service *service)
{
	struct user_service *user_service =
			(struct user_service *)service->base.userdata;

	seq_printf(f, "  instance %pK", service->instance);

	if ((service->base.callback == service_callback) && user_service->is_vchi) {
		seq_printf(f, ", %d/%d messages",
			   user_service->msg_insert - user_service->msg_remove,
			   MSG_QUEUE_SIZE);

		if (user_service->dequeue_pending)
			seq_puts(f, " (dequeue pending)");
	}

	seq_puts(f, "\n");
}

/*
 * Autosuspend related functionality
 */

static int
vchiq_keepalive_vchiq_callback(struct vchiq_instance *instance,
			       enum vchiq_reason reason,
			       struct vchiq_header *header,
			       unsigned int service_user,
			       void *cb_data, void __user *cb_userdata)
{
	dev_err(instance->state->dev, "suspend: %s: callback reason %d\n",
		__func__, reason);
	return 0;
}

static int
vchiq_keepalive_thread_func(void *v)
{
	struct vchiq_state *state = (struct vchiq_state *)v;
	struct vchiq_arm_state *arm_state = vchiq_platform_get_arm_state(state);
	struct vchiq_instance *instance;
	unsigned int ka_handle;
	int ret;

	struct vchiq_service_params_kernel params = {
		.fourcc      = VCHIQ_MAKE_FOURCC('K', 'E', 'E', 'P'),
		.callback    = vchiq_keepalive_vchiq_callback,
		.version     = KEEPALIVE_VER,
		.version_min = KEEPALIVE_VER_MIN
	};

	ret = vchiq_initialise(state, &instance);
	if (ret) {
		dev_err(state->dev, "suspend: %s: vchiq_initialise failed %d\n", __func__, ret);
		goto exit;
	}

	ret = vchiq_connect(instance);
	if (ret) {
		dev_err(state->dev, "suspend: %s: vchiq_connect failed %d\n", __func__, ret);
		goto shutdown;
	}

	ret = vchiq_add_service(instance, &params, &ka_handle);
	if (ret) {
		dev_err(state->dev, "suspend: %s: vchiq_open_service failed %d\n",
			__func__, ret);
		goto shutdown;
	}

	while (!kthread_should_stop()) {
		long rc = 0, uc = 0;

		if (wait_for_completion_interruptible(&arm_state->ka_evt)) {
			dev_dbg(state->dev, "suspend: %s: interrupted\n", __func__);
			flush_signals(current);
			continue;
		}

		/*
		 * read and clear counters.  Do release_count then use_count to
		 * prevent getting more releases than uses
		 */
		rc = atomic_xchg(&arm_state->ka_release_count, 0);
		uc = atomic_xchg(&arm_state->ka_use_count, 0);

		/*
		 * Call use/release service the requisite number of times.
		 * Process use before release so use counts don't go negative
		 */
		while (uc--) {
			atomic_inc(&arm_state->ka_use_ack_count);
			ret = vchiq_use_service(instance, ka_handle);
			if (ret) {
				dev_err(state->dev, "suspend: %s: vchiq_use_service error %d\n",
					__func__, ret);
			}
		}
		while (rc--) {
			ret = vchiq_release_service(instance, ka_handle);
			if (ret) {
				dev_err(state->dev, "suspend: %s: vchiq_release_service error %d\n",
					__func__, ret);
			}
		}
	}

shutdown:
	vchiq_shutdown(instance);
exit:
	return 0;
}

int
vchiq_use_internal(struct vchiq_state *state, struct vchiq_service *service,
		   enum USE_TYPE_E use_type)
{
	struct vchiq_arm_state *arm_state = vchiq_platform_get_arm_state(state);
	int ret = 0;
	char entity[64];
	int *entity_uc;
	int local_uc;

	if (!arm_state) {
		ret = -EINVAL;
		goto out;
	}

	if (use_type == USE_TYPE_VCHIQ) {
		snprintf(entity, sizeof(entity), "VCHIQ:   ");
		entity_uc = &arm_state->peer_use_count;
	} else if (service) {
		snprintf(entity, sizeof(entity), "%p4cc:%03d",
			 &service->base.fourcc,
			 service->client_id);
		entity_uc = &service->service_use_count;
	} else {
		dev_err(state->dev, "suspend: %s: null service ptr\n", __func__);
		ret = -EINVAL;
		goto out;
	}

	write_lock_bh(&arm_state->susp_res_lock);
	local_uc = ++arm_state->videocore_use_count;
	++(*entity_uc);

	dev_dbg(state->dev, "suspend: %s count %d, state count %d\n",
		entity, *entity_uc, local_uc);

	write_unlock_bh(&arm_state->susp_res_lock);

	if (!ret) {
		int ret = 0;
		long ack_cnt = atomic_xchg(&arm_state->ka_use_ack_count, 0);

		while (ack_cnt && !ret) {
			/* Send the use notify to videocore */
			ret = vchiq_send_remote_use_active(state);
			if (!ret)
				ack_cnt--;
			else
				atomic_add(ack_cnt, &arm_state->ka_use_ack_count);
		}
	}

out:
	dev_dbg(state->dev, "suspend: exit %d\n", ret);
	return ret;
}

int
vchiq_release_internal(struct vchiq_state *state, struct vchiq_service *service)
{
	struct vchiq_arm_state *arm_state = vchiq_platform_get_arm_state(state);
	int ret = 0;
	char entity[64];
	int *entity_uc;

	if (!arm_state) {
		ret = -EINVAL;
		goto out;
	}

	if (service) {
		snprintf(entity, sizeof(entity), "%p4cc:%03d",
			 &service->base.fourcc,
			 service->client_id);
		entity_uc = &service->service_use_count;
	} else {
		snprintf(entity, sizeof(entity), "PEER:   ");
		entity_uc = &arm_state->peer_use_count;
	}

	write_lock_bh(&arm_state->susp_res_lock);
	if (!arm_state->videocore_use_count || !(*entity_uc)) {
		WARN_ON(!arm_state->videocore_use_count);
		WARN_ON(!(*entity_uc));
		ret = -EINVAL;
		goto unlock;
	}
	--arm_state->videocore_use_count;
	--(*entity_uc);

	dev_dbg(state->dev, "suspend: %s count %d, state count %d\n",
		entity, *entity_uc, arm_state->videocore_use_count);

unlock:
	write_unlock_bh(&arm_state->susp_res_lock);

out:
	dev_dbg(state->dev, "suspend: exit %d\n", ret);
	return ret;
}

void
vchiq_on_remote_use(struct vchiq_state *state)
{
	struct vchiq_arm_state *arm_state = vchiq_platform_get_arm_state(state);

	atomic_inc(&arm_state->ka_use_count);
	complete(&arm_state->ka_evt);
}

void
vchiq_on_remote_release(struct vchiq_state *state)
{
	struct vchiq_arm_state *arm_state = vchiq_platform_get_arm_state(state);

	atomic_inc(&arm_state->ka_release_count);
	complete(&arm_state->ka_evt);
}

int
vchiq_use_service_internal(struct vchiq_service *service)
{
	return vchiq_use_internal(service->state, service, USE_TYPE_SERVICE);
}

int
vchiq_release_service_internal(struct vchiq_service *service)
{
	return vchiq_release_internal(service->state, service);
}

struct vchiq_debugfs_node *
vchiq_instance_get_debugfs_node(struct vchiq_instance *instance)
{
	return &instance->debugfs_node;
}

int
vchiq_instance_get_use_count(struct vchiq_instance *instance)
{
	struct vchiq_service *service;
	int use_count = 0, i;

	i = 0;
	rcu_read_lock();
	while ((service = __next_service_by_instance(instance->state,
						     instance, &i)))
		use_count += service->service_use_count;
	rcu_read_unlock();
	return use_count;
}

int
vchiq_instance_get_pid(struct vchiq_instance *instance)
{
	return instance->pid;
}

int
vchiq_instance_get_trace(struct vchiq_instance *instance)
{
	return instance->trace;
}

void
vchiq_instance_set_trace(struct vchiq_instance *instance, int trace)
{
	struct vchiq_service *service;
	int i;

	i = 0;
	rcu_read_lock();
	while ((service = __next_service_by_instance(instance->state,
						     instance, &i)))
		service->trace = trace;
	rcu_read_unlock();
	instance->trace = (trace != 0);
}

int
vchiq_use_service(struct vchiq_instance *instance, unsigned int handle)
{
	int ret = -EINVAL;
	struct vchiq_service *service = find_service_by_handle(instance, handle);

	if (service) {
		ret = vchiq_use_internal(service->state, service, USE_TYPE_SERVICE);
		vchiq_service_put(service);
	}
	return ret;
}
EXPORT_SYMBOL(vchiq_use_service);

int
vchiq_release_service(struct vchiq_instance *instance, unsigned int handle)
{
	int ret = -EINVAL;
	struct vchiq_service *service = find_service_by_handle(instance, handle);

	if (service) {
		ret = vchiq_release_internal(service->state, service);
		vchiq_service_put(service);
	}
	return ret;
}
EXPORT_SYMBOL(vchiq_release_service);

struct service_data_struct {
	int fourcc;
	int clientid;
	int use_count;
};

void
vchiq_dump_service_use_state(struct vchiq_state *state)
{
	struct vchiq_arm_state *arm_state = vchiq_platform_get_arm_state(state);
	struct service_data_struct *service_data;
	int i, found = 0;
	/*
	 * If there's more than 64 services, only dump ones with
	 * non-zero counts
	 */
	int only_nonzero = 0;
	static const char *nz = "<-- preventing suspend";

	int peer_count;
	int vc_use_count;
	int active_services;

	if (!arm_state)
		return;

	service_data = kmalloc_array(MAX_SERVICES, sizeof(*service_data),
				     GFP_KERNEL);
	if (!service_data)
		return;

	read_lock_bh(&arm_state->susp_res_lock);
	peer_count = arm_state->peer_use_count;
	vc_use_count = arm_state->videocore_use_count;
	active_services = state->unused_service;
	if (active_services > MAX_SERVICES)
		only_nonzero = 1;

	rcu_read_lock();
	for (i = 0; i < active_services; i++) {
		struct vchiq_service *service_ptr =
			rcu_dereference(state->services[i]);

		if (!service_ptr)
			continue;

		if (only_nonzero && !service_ptr->service_use_count)
			continue;

		if (service_ptr->srvstate == VCHIQ_SRVSTATE_FREE)
			continue;

		service_data[found].fourcc = service_ptr->base.fourcc;
		service_data[found].clientid = service_ptr->client_id;
		service_data[found].use_count = service_ptr->service_use_count;
		found++;
		if (found >= MAX_SERVICES)
			break;
	}
	rcu_read_unlock();

	read_unlock_bh(&arm_state->susp_res_lock);

	if (only_nonzero)
		dev_warn(state->dev,
			 "suspend: Too many active services (%d). Only dumping up to first %d services with non-zero use-count\n",
			 active_services, found);

	for (i = 0; i < found; i++) {
		dev_warn(state->dev,
			 "suspend: %p4cc:%d service count %d %s\n",
			 &service_data[i].fourcc,
			 service_data[i].clientid, service_data[i].use_count,
			 service_data[i].use_count ? nz : "");
	}
	dev_warn(state->dev, "suspend: VCHIQ use count %d\n", peer_count);
	dev_warn(state->dev, "suspend: Overall vchiq instance use count %d\n", vc_use_count);

	kfree(service_data);
}

int
vchiq_check_service(struct vchiq_service *service)
{
	struct vchiq_arm_state *arm_state;
	int ret = -EINVAL;

	if (!service || !service->state)
		goto out;

	arm_state = vchiq_platform_get_arm_state(service->state);

	read_lock_bh(&arm_state->susp_res_lock);
	if (service->service_use_count)
		ret = 0;
	read_unlock_bh(&arm_state->susp_res_lock);

	if (ret) {
		dev_err(service->state->dev,
			"suspend: %s:  %p4cc:%d service count %d, state count %d\n",
			__func__, &service->base.fourcc, service->client_id,
			service->service_use_count, arm_state->videocore_use_count);
		vchiq_dump_service_use_state(service->state);
	}
out:
	return ret;
}

void vchiq_platform_conn_state_changed(struct vchiq_state *state,
				       enum vchiq_connstate oldstate,
				       enum vchiq_connstate newstate)
{
	struct vchiq_arm_state *arm_state = vchiq_platform_get_arm_state(state);
	char threadname[16];

	dev_dbg(state->dev, "suspend: %d: %s->%s\n",
		state->id, get_conn_state_name(oldstate), get_conn_state_name(newstate));
	if (state->conn_state != VCHIQ_CONNSTATE_CONNECTED)
		return;

	write_lock_bh(&arm_state->susp_res_lock);
	if (arm_state->first_connect) {
		write_unlock_bh(&arm_state->susp_res_lock);
		return;
	}

	arm_state->first_connect = 1;
	write_unlock_bh(&arm_state->susp_res_lock);
	snprintf(threadname, sizeof(threadname), "vchiq-keep/%d",
		 state->id);
	arm_state->ka_thread = kthread_create(&vchiq_keepalive_thread_func,
					      (void *)state,
					      threadname);
	if (IS_ERR(arm_state->ka_thread)) {
		dev_err(state->dev, "suspend: Couldn't create thread %s\n",
			threadname);
	} else {
		wake_up_process(arm_state->ka_thread);
	}
}

static const struct of_device_id vchiq_of_match[] = {
	{ .compatible = "brcm,bcm2835-vchiq", .data = &bcm2835_info },
	{ .compatible = "brcm,bcm2836-vchiq", .data = &bcm2836_info },
	{},
};
MODULE_DEVICE_TABLE(of, vchiq_of_match);

static int vchiq_probe(struct platform_device *pdev)
{
	const struct vchiq_platform_info *info;
	struct vchiq_drv_mgmt *mgmt;
	int ret;

	info = of_device_get_match_data(&pdev->dev);
	if (!info)
		return -EINVAL;

	struct device_node *fw_node __free(device_node) =
		of_find_compatible_node(NULL, NULL, "raspberrypi,bcm2835-firmware");
	if (!fw_node) {
		dev_err(&pdev->dev, "Missing firmware node\n");
		return -ENOENT;
	}

	mgmt = devm_kzalloc(&pdev->dev, sizeof(*mgmt), GFP_KERNEL);
	if (!mgmt)
		return -ENOMEM;

	mgmt->fw = devm_rpi_firmware_get(&pdev->dev, fw_node);
	if (!mgmt->fw)
		return -EPROBE_DEFER;

	mgmt->info = info;
	platform_set_drvdata(pdev, mgmt);

	ret = vchiq_platform_init(pdev, &mgmt->state);
	if (ret) {
		dev_err(&pdev->dev, "arm: Could not initialize vchiq platform\n");
		return ret;
	}

	dev_dbg(&pdev->dev, "arm: platform initialised - version %d (min %d)\n",
		VCHIQ_VERSION, VCHIQ_VERSION_MIN);

	/*
	 * Simply exit on error since the function handles cleanup in
	 * cases of failure.
	 */
	ret = vchiq_register_chrdev(&pdev->dev);
	if (ret) {
		dev_err(&pdev->dev, "arm: Failed to initialize vchiq cdev\n");
		vchiq_platform_uninit(mgmt);
		return ret;
	}

	vchiq_debugfs_init(&mgmt->state);

	bcm2835_audio = vchiq_device_register(&pdev->dev, "bcm2835-audio");

	return 0;
}

static void vchiq_remove(struct platform_device *pdev)
{
	struct vchiq_drv_mgmt *mgmt = dev_get_drvdata(&pdev->dev);

	vchiq_device_unregister(bcm2835_audio);
	vchiq_debugfs_deinit();
	vchiq_deregister_chrdev();
	vchiq_platform_uninit(mgmt);
}

static struct platform_driver vchiq_driver = {
	.driver = {
		.name = "bcm2835_vchiq",
		.of_match_table = vchiq_of_match,
	},
	.probe = vchiq_probe,
	.remove = vchiq_remove,
};

static int __init vchiq_driver_init(void)
{
	int ret;

	ret = bus_register(&vchiq_bus_type);
	if (ret) {
		pr_err("Failed to register %s\n", vchiq_bus_type.name);
		return ret;
	}

	ret = platform_driver_register(&vchiq_driver);
	if (ret) {
		pr_err("Failed to register vchiq driver\n");
		bus_unregister(&vchiq_bus_type);
	}

	return ret;
}
module_init(vchiq_driver_init);

static void __exit vchiq_driver_exit(void)
{
	bus_unregister(&vchiq_bus_type);
	platform_driver_unregister(&vchiq_driver);
}
module_exit(vchiq_driver_exit);

MODULE_LICENSE("Dual BSD/GPL");
MODULE_DESCRIPTION("Videocore VCHIQ driver");
MODULE_AUTHOR("Broadcom Corporation");