diff options
Diffstat (limited to 'drivers/gpu/host1x')
95 files changed, 8035 insertions, 8043 deletions
diff --git a/drivers/gpu/host1x/Kconfig b/drivers/gpu/host1x/Kconfig index ccfd42b23606..e6c78ae2003a 100644 --- a/drivers/gpu/host1x/Kconfig +++ b/drivers/gpu/host1x/Kconfig @@ -1,6 +1,14 @@ +# SPDX-License-Identifier: GPL-2.0-only + +config TEGRA_HOST1X_CONTEXT_BUS + bool + config TEGRA_HOST1X tristate "NVIDIA Tegra host1x driver" - depends on ARCH_TEGRA || ARCH_MULTIPLATFORM + depends on ARCH_TEGRA || COMPILE_TEST + select DMA_SHARED_BUFFER + select TEGRA_HOST1X_CONTEXT_BUS + select IOMMU_IOVA help Driver for the NVIDIA Tegra host1x hardware. @@ -19,6 +27,4 @@ config TEGRA_HOST1X_FIREWALL If unsure, choose Y. -source "drivers/gpu/host1x/drm/Kconfig" - endif diff --git a/drivers/gpu/host1x/Makefile b/drivers/gpu/host1x/Makefile index 3b037b6e0298..ee5286ffe08d 100644 --- a/drivers/gpu/host1x/Makefile +++ b/drivers/gpu/host1x/Makefile @@ -1,6 +1,6 @@ -ccflags-y = -Idrivers/gpu/host1x - +# SPDX-License-Identifier: GPL-2.0 host1x-y = \ + bus.o \ syncpt.o \ dev.o \ intr.o \ @@ -8,13 +8,18 @@ host1x-y = \ channel.o \ job.o \ debug.o \ - hw/host1x01.o + mipi.o \ + fence.o \ + hw/host1x01.o \ + hw/host1x02.o \ + hw/host1x04.o \ + hw/host1x05.o \ + hw/host1x06.o \ + hw/host1x07.o \ + hw/host1x08.o -ccflags-y += -Iinclude/drm -ccflags-$(CONFIG_DRM_TEGRA_DEBUG) += -DDEBUG +host1x-$(CONFIG_IOMMU_API) += \ + context.o -host1x-$(CONFIG_DRM_TEGRA) += drm/drm.o drm/fb.o drm/dc.o -host1x-$(CONFIG_DRM_TEGRA) += drm/output.o drm/rgb.o drm/hdmi.o -host1x-$(CONFIG_DRM_TEGRA) += drm/gem.o -host1x-$(CONFIG_DRM_TEGRA) += drm/gr2d.o obj-$(CONFIG_TEGRA_HOST1X) += host1x.o +obj-$(CONFIG_TEGRA_HOST1X_CONTEXT_BUS) += context_bus.o diff --git a/drivers/gpu/host1x/bus.c b/drivers/gpu/host1x/bus.c new file mode 100644 index 000000000000..723a80895cd4 --- /dev/null +++ b/drivers/gpu/host1x/bus.c @@ -0,0 +1,964 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2012 Avionic Design GmbH + * Copyright (C) 2012-2013, NVIDIA Corporation + */ + +#include <linux/debugfs.h> +#include <linux/dma-mapping.h> +#include <linux/host1x.h> +#include <linux/of.h> +#include <linux/seq_file.h> +#include <linux/slab.h> +#include <linux/of_device.h> + +#include "bus.h" +#include "dev.h" + +static DEFINE_MUTEX(clients_lock); +static LIST_HEAD(clients); + +static DEFINE_MUTEX(drivers_lock); +static LIST_HEAD(drivers); + +static DEFINE_MUTEX(devices_lock); +static LIST_HEAD(devices); + +struct host1x_subdev { + struct host1x_client *client; + struct device_node *np; + struct list_head list; +}; + +/** + * host1x_subdev_add() - add a new subdevice with an associated device node + * @device: host1x device to add the subdevice to + * @driver: host1x driver containing the subdevices + * @np: device node + */ +static int host1x_subdev_add(struct host1x_device *device, + struct host1x_driver *driver, + struct device_node *np) +{ + struct host1x_subdev *subdev; + int err; + + subdev = kzalloc(sizeof(*subdev), GFP_KERNEL); + if (!subdev) + return -ENOMEM; + + INIT_LIST_HEAD(&subdev->list); + subdev->np = of_node_get(np); + + mutex_lock(&device->subdevs_lock); + list_add_tail(&subdev->list, &device->subdevs); + mutex_unlock(&device->subdevs_lock); + + /* recursively add children */ + for_each_child_of_node_scoped(np, child) { + if (of_match_node(driver->subdevs, child) && + of_device_is_available(child)) { + err = host1x_subdev_add(device, driver, child); + if (err < 0) { + /* XXX cleanup? */ + return err; + } + } + } + + return 0; +} + +/** + * host1x_subdev_del() - remove subdevice + * @subdev: subdevice to remove + */ +static void host1x_subdev_del(struct host1x_subdev *subdev) +{ + list_del(&subdev->list); + of_node_put(subdev->np); + kfree(subdev); +} + +/** + * host1x_device_parse_dt() - scan device tree and add matching subdevices + * @device: host1x logical device + * @driver: host1x driver + */ +static int host1x_device_parse_dt(struct host1x_device *device, + struct host1x_driver *driver) +{ + int err; + + for_each_child_of_node_scoped(device->dev.parent->of_node, np) { + if (of_match_node(driver->subdevs, np) && + of_device_is_available(np)) { + err = host1x_subdev_add(device, driver, np); + if (err < 0) + return err; + } + } + + return 0; +} + +static void host1x_subdev_register(struct host1x_device *device, + struct host1x_subdev *subdev, + struct host1x_client *client) +{ + int err; + + /* + * Move the subdevice to the list of active (registered) subdevices + * and associate it with a client. At the same time, associate the + * client with its parent device. + */ + mutex_lock(&device->subdevs_lock); + mutex_lock(&device->clients_lock); + list_move_tail(&client->list, &device->clients); + list_move_tail(&subdev->list, &device->active); + client->host = &device->dev; + subdev->client = client; + mutex_unlock(&device->clients_lock); + mutex_unlock(&device->subdevs_lock); + + if (list_empty(&device->subdevs)) { + err = device_add(&device->dev); + if (err < 0) + dev_err(&device->dev, "failed to add: %d\n", err); + else + device->registered = true; + } +} + +static void __host1x_subdev_unregister(struct host1x_device *device, + struct host1x_subdev *subdev) +{ + struct host1x_client *client = subdev->client; + + /* + * If all subdevices have been activated, we're about to remove the + * first active subdevice, so unload the driver first. + */ + if (list_empty(&device->subdevs)) { + if (device->registered) { + device->registered = false; + device_del(&device->dev); + } + } + + /* + * Move the subdevice back to the list of idle subdevices and remove + * it from list of clients. + */ + mutex_lock(&device->clients_lock); + subdev->client = NULL; + client->host = NULL; + list_move_tail(&subdev->list, &device->subdevs); + /* + * XXX: Perhaps don't do this here, but rather explicitly remove it + * when the device is about to be deleted. + * + * This is somewhat complicated by the fact that this function is + * used to remove the subdevice when a client is unregistered but + * also when the composite device is about to be removed. + */ + list_del_init(&client->list); + mutex_unlock(&device->clients_lock); +} + +static void host1x_subdev_unregister(struct host1x_device *device, + struct host1x_subdev *subdev) +{ + mutex_lock(&device->subdevs_lock); + __host1x_subdev_unregister(device, subdev); + mutex_unlock(&device->subdevs_lock); +} + +/** + * host1x_device_init() - initialize a host1x logical device + * @device: host1x logical device + * + * The driver for the host1x logical device can call this during execution of + * its &host1x_driver.probe implementation to initialize each of its clients. + * The client drivers access the subsystem specific driver data using the + * &host1x_client.parent field and driver data associated with it (usually by + * calling dev_get_drvdata()). + */ +int host1x_device_init(struct host1x_device *device) +{ + struct host1x_client *client; + int err; + + mutex_lock(&device->clients_lock); + + list_for_each_entry(client, &device->clients, list) { + if (client->ops && client->ops->early_init) { + err = client->ops->early_init(client); + if (err < 0) { + dev_err(&device->dev, "failed to early initialize %s: %d\n", + dev_name(client->dev), err); + goto teardown_late; + } + } + } + + list_for_each_entry(client, &device->clients, list) { + if (client->ops && client->ops->init) { + err = client->ops->init(client); + if (err < 0) { + dev_err(&device->dev, + "failed to initialize %s: %d\n", + dev_name(client->dev), err); + goto teardown; + } + } + } + + mutex_unlock(&device->clients_lock); + + return 0; + +teardown: + list_for_each_entry_continue_reverse(client, &device->clients, list) + if (client->ops->exit) + client->ops->exit(client); + + /* reset client to end of list for late teardown */ + client = list_entry(&device->clients, struct host1x_client, list); + +teardown_late: + list_for_each_entry_continue_reverse(client, &device->clients, list) + if (client->ops->late_exit) + client->ops->late_exit(client); + + mutex_unlock(&device->clients_lock); + return err; +} +EXPORT_SYMBOL(host1x_device_init); + +/** + * host1x_device_exit() - uninitialize host1x logical device + * @device: host1x logical device + * + * When the driver for a host1x logical device is unloaded, it can call this + * function to tear down each of its clients. Typically this is done after a + * subsystem-specific data structure is removed and the functionality can no + * longer be used. + */ +int host1x_device_exit(struct host1x_device *device) +{ + struct host1x_client *client; + int err; + + mutex_lock(&device->clients_lock); + + list_for_each_entry_reverse(client, &device->clients, list) { + if (client->ops && client->ops->exit) { + err = client->ops->exit(client); + if (err < 0) { + dev_err(&device->dev, + "failed to cleanup %s: %d\n", + dev_name(client->dev), err); + mutex_unlock(&device->clients_lock); + return err; + } + } + } + + list_for_each_entry_reverse(client, &device->clients, list) { + if (client->ops && client->ops->late_exit) { + err = client->ops->late_exit(client); + if (err < 0) { + dev_err(&device->dev, "failed to late cleanup %s: %d\n", + dev_name(client->dev), err); + mutex_unlock(&device->clients_lock); + return err; + } + } + } + + mutex_unlock(&device->clients_lock); + + return 0; +} +EXPORT_SYMBOL(host1x_device_exit); + +static int host1x_add_client(struct host1x *host1x, + struct host1x_client *client) +{ + struct host1x_device *device; + struct host1x_subdev *subdev; + + mutex_lock(&host1x->devices_lock); + + list_for_each_entry(device, &host1x->devices, list) { + list_for_each_entry(subdev, &device->subdevs, list) { + if (subdev->np == client->dev->of_node) { + host1x_subdev_register(device, subdev, client); + mutex_unlock(&host1x->devices_lock); + return 0; + } + } + } + + mutex_unlock(&host1x->devices_lock); + return -ENODEV; +} + +static int host1x_del_client(struct host1x *host1x, + struct host1x_client *client) +{ + struct host1x_device *device, *dt; + struct host1x_subdev *subdev; + + mutex_lock(&host1x->devices_lock); + + list_for_each_entry_safe(device, dt, &host1x->devices, list) { + list_for_each_entry(subdev, &device->active, list) { + if (subdev->client == client) { + host1x_subdev_unregister(device, subdev); + mutex_unlock(&host1x->devices_lock); + return 0; + } + } + } + + mutex_unlock(&host1x->devices_lock); + return -ENODEV; +} + +static int host1x_device_match(struct device *dev, const struct device_driver *drv) +{ + return strcmp(dev_name(dev), drv->name) == 0; +} + +/* + * Note that this is really only needed for backwards compatibility + * with libdrm, which parses this information from sysfs and will + * fail if it can't find the OF_FULLNAME, specifically. + */ +static int host1x_device_uevent(const struct device *dev, + struct kobj_uevent_env *env) +{ + of_device_uevent(dev->parent, env); + + return 0; +} + +static const struct dev_pm_ops host1x_device_pm_ops = { + .suspend = pm_generic_suspend, + .resume = pm_generic_resume, + .freeze = pm_generic_freeze, + .thaw = pm_generic_thaw, + .poweroff = pm_generic_poweroff, + .restore = pm_generic_restore, +}; + +const struct bus_type host1x_bus_type = { + .name = "host1x", + .match = host1x_device_match, + .uevent = host1x_device_uevent, + .pm = &host1x_device_pm_ops, +}; + +static void __host1x_device_del(struct host1x_device *device) +{ + struct host1x_subdev *subdev, *sd; + struct host1x_client *client, *cl; + + mutex_lock(&device->subdevs_lock); + + /* unregister subdevices */ + list_for_each_entry_safe(subdev, sd, &device->active, list) { + /* + * host1x_subdev_unregister() will remove the client from + * any lists, so we'll need to manually add it back to the + * list of idle clients. + * + * XXX: Alternatively, perhaps don't remove the client from + * any lists in host1x_subdev_unregister() and instead do + * that explicitly from host1x_unregister_client()? + */ + client = subdev->client; + + __host1x_subdev_unregister(device, subdev); + + /* add the client to the list of idle clients */ + mutex_lock(&clients_lock); + list_add_tail(&client->list, &clients); + mutex_unlock(&clients_lock); + } + + /* remove subdevices */ + list_for_each_entry_safe(subdev, sd, &device->subdevs, list) + host1x_subdev_del(subdev); + + mutex_unlock(&device->subdevs_lock); + + /* move clients to idle list */ + mutex_lock(&clients_lock); + mutex_lock(&device->clients_lock); + + list_for_each_entry_safe(client, cl, &device->clients, list) + list_move_tail(&client->list, &clients); + + mutex_unlock(&device->clients_lock); + mutex_unlock(&clients_lock); + + /* finally remove the device */ + list_del_init(&device->list); +} + +static void host1x_device_release(struct device *dev) +{ + struct host1x_device *device = to_host1x_device(dev); + + __host1x_device_del(device); + kfree(device); +} + +static int host1x_device_add(struct host1x *host1x, + struct host1x_driver *driver) +{ + struct host1x_client *client, *tmp; + struct host1x_subdev *subdev; + struct host1x_device *device; + int err; + + device = kzalloc(sizeof(*device), GFP_KERNEL); + if (!device) + return -ENOMEM; + + device_initialize(&device->dev); + + mutex_init(&device->subdevs_lock); + INIT_LIST_HEAD(&device->subdevs); + INIT_LIST_HEAD(&device->active); + mutex_init(&device->clients_lock); + INIT_LIST_HEAD(&device->clients); + INIT_LIST_HEAD(&device->list); + device->driver = driver; + + device->dev.coherent_dma_mask = host1x->dev->coherent_dma_mask; + device->dev.dma_mask = &device->dev.coherent_dma_mask; + dev_set_name(&device->dev, "%s", driver->driver.name); + device->dev.release = host1x_device_release; + device->dev.bus = &host1x_bus_type; + device->dev.parent = host1x->dev; + + device->dev.dma_parms = &device->dma_parms; + dma_set_max_seg_size(&device->dev, UINT_MAX); + + err = host1x_device_parse_dt(device, driver); + if (err < 0) { + kfree(device); + return err; + } + + list_add_tail(&device->list, &host1x->devices); + + mutex_lock(&clients_lock); + + list_for_each_entry_safe(client, tmp, &clients, list) { + list_for_each_entry(subdev, &device->subdevs, list) { + if (subdev->np == client->dev->of_node) { + host1x_subdev_register(device, subdev, client); + break; + } + } + } + + mutex_unlock(&clients_lock); + + /* + * Add device even if there are no subdevs to ensure syncpoint functionality + * is available regardless of whether any engine subdevices are present + */ + if (list_empty(&device->subdevs)) { + err = device_add(&device->dev); + if (err < 0) + dev_err(&device->dev, "failed to add device: %d\n", err); + else + device->registered = true; + } + + return 0; +} + +/* + * Removes a device by first unregistering any subdevices and then removing + * itself from the list of devices. + * + * This function must be called with the host1x->devices_lock held. + */ +static void host1x_device_del(struct host1x *host1x, + struct host1x_device *device) +{ + if (device->registered) { + device->registered = false; + device_del(&device->dev); + } + + put_device(&device->dev); +} + +static void host1x_attach_driver(struct host1x *host1x, + struct host1x_driver *driver) +{ + struct host1x_device *device; + int err; + + mutex_lock(&host1x->devices_lock); + + list_for_each_entry(device, &host1x->devices, list) { + if (device->driver == driver) { + mutex_unlock(&host1x->devices_lock); + return; + } + } + + err = host1x_device_add(host1x, driver); + if (err < 0) + dev_err(host1x->dev, "failed to allocate device: %d\n", err); + + mutex_unlock(&host1x->devices_lock); +} + +static void host1x_detach_driver(struct host1x *host1x, + struct host1x_driver *driver) +{ + struct host1x_device *device, *tmp; + + mutex_lock(&host1x->devices_lock); + + list_for_each_entry_safe(device, tmp, &host1x->devices, list) + if (device->driver == driver) + host1x_device_del(host1x, device); + + mutex_unlock(&host1x->devices_lock); +} + +static int host1x_devices_show(struct seq_file *s, void *data) +{ + struct host1x *host1x = s->private; + struct host1x_device *device; + + mutex_lock(&host1x->devices_lock); + + list_for_each_entry(device, &host1x->devices, list) { + struct host1x_subdev *subdev; + + seq_printf(s, "%s\n", dev_name(&device->dev)); + + mutex_lock(&device->subdevs_lock); + + list_for_each_entry(subdev, &device->active, list) + seq_printf(s, " %pOFf: %s\n", subdev->np, + dev_name(subdev->client->dev)); + + list_for_each_entry(subdev, &device->subdevs, list) + seq_printf(s, " %pOFf:\n", subdev->np); + + mutex_unlock(&device->subdevs_lock); + } + + mutex_unlock(&host1x->devices_lock); + + return 0; +} +DEFINE_SHOW_ATTRIBUTE(host1x_devices); + +/** + * host1x_register() - register a host1x controller + * @host1x: host1x controller + * + * The host1x controller driver uses this to register a host1x controller with + * the infrastructure. Note that all Tegra SoC generations have only ever come + * with a single host1x instance, so this function is somewhat academic. + */ +int host1x_register(struct host1x *host1x) +{ + struct host1x_driver *driver; + + mutex_lock(&devices_lock); + list_add_tail(&host1x->list, &devices); + mutex_unlock(&devices_lock); + + mutex_lock(&drivers_lock); + + list_for_each_entry(driver, &drivers, list) + host1x_attach_driver(host1x, driver); + + mutex_unlock(&drivers_lock); + + debugfs_create_file("devices", S_IRUGO, host1x->debugfs, host1x, + &host1x_devices_fops); + + return 0; +} + +/** + * host1x_unregister() - unregister a host1x controller + * @host1x: host1x controller + * + * The host1x controller driver uses this to remove a host1x controller from + * the infrastructure. + */ +int host1x_unregister(struct host1x *host1x) +{ + struct host1x_driver *driver; + + mutex_lock(&drivers_lock); + + list_for_each_entry(driver, &drivers, list) + host1x_detach_driver(host1x, driver); + + mutex_unlock(&drivers_lock); + + mutex_lock(&devices_lock); + list_del_init(&host1x->list); + mutex_unlock(&devices_lock); + + return 0; +} + +static int host1x_device_probe(struct device *dev) +{ + struct host1x_driver *driver = to_host1x_driver(dev->driver); + struct host1x_device *device = to_host1x_device(dev); + + if (driver->probe) + return driver->probe(device); + + return 0; +} + +static int host1x_device_remove(struct device *dev) +{ + struct host1x_driver *driver = to_host1x_driver(dev->driver); + struct host1x_device *device = to_host1x_device(dev); + + if (driver->remove) + return driver->remove(device); + + return 0; +} + +static void host1x_device_shutdown(struct device *dev) +{ + struct host1x_driver *driver = to_host1x_driver(dev->driver); + struct host1x_device *device = to_host1x_device(dev); + + if (driver->shutdown) + driver->shutdown(device); +} + +/** + * host1x_driver_register_full() - register a host1x driver + * @driver: host1x driver + * @owner: owner module + * + * Drivers for host1x logical devices call this function to register a driver + * with the infrastructure. Note that since these drive logical devices, the + * registration of the driver actually triggers tho logical device creation. + * A logical device will be created for each host1x instance. + */ +int host1x_driver_register_full(struct host1x_driver *driver, + struct module *owner) +{ + struct host1x *host1x; + + INIT_LIST_HEAD(&driver->list); + + mutex_lock(&drivers_lock); + list_add_tail(&driver->list, &drivers); + mutex_unlock(&drivers_lock); + + mutex_lock(&devices_lock); + + list_for_each_entry(host1x, &devices, list) + host1x_attach_driver(host1x, driver); + + mutex_unlock(&devices_lock); + + driver->driver.bus = &host1x_bus_type; + driver->driver.owner = owner; + driver->driver.probe = host1x_device_probe; + driver->driver.remove = host1x_device_remove; + driver->driver.shutdown = host1x_device_shutdown; + + return driver_register(&driver->driver); +} +EXPORT_SYMBOL(host1x_driver_register_full); + +/** + * host1x_driver_unregister() - unregister a host1x driver + * @driver: host1x driver + * + * Unbinds the driver from each of the host1x logical devices that it is + * bound to, effectively removing the subsystem devices that they represent. + */ +void host1x_driver_unregister(struct host1x_driver *driver) +{ + struct host1x *host1x; + + driver_unregister(&driver->driver); + + mutex_lock(&devices_lock); + + list_for_each_entry(host1x, &devices, list) + host1x_detach_driver(host1x, driver); + + mutex_unlock(&devices_lock); + + mutex_lock(&drivers_lock); + list_del_init(&driver->list); + mutex_unlock(&drivers_lock); +} +EXPORT_SYMBOL(host1x_driver_unregister); + +/** + * __host1x_client_init() - initialize a host1x client + * @client: host1x client + * @key: lock class key for the client-specific mutex + */ +void __host1x_client_init(struct host1x_client *client, struct lock_class_key *key) +{ + host1x_bo_cache_init(&client->cache); + INIT_LIST_HEAD(&client->list); + __mutex_init(&client->lock, "host1x client lock", key); + client->usecount = 0; +} +EXPORT_SYMBOL(__host1x_client_init); + +/** + * host1x_client_exit() - uninitialize a host1x client + * @client: host1x client + */ +void host1x_client_exit(struct host1x_client *client) +{ + mutex_destroy(&client->lock); +} +EXPORT_SYMBOL(host1x_client_exit); + +/** + * __host1x_client_register() - register a host1x client + * @client: host1x client + * + * Registers a host1x client with each host1x controller instance. Note that + * each client will only match their parent host1x controller and will only be + * associated with that instance. Once all clients have been registered with + * their parent host1x controller, the infrastructure will set up the logical + * device and call host1x_device_init(), which will in turn call each client's + * &host1x_client_ops.init implementation. + */ +int __host1x_client_register(struct host1x_client *client) +{ + struct host1x *host1x; + int err; + + mutex_lock(&devices_lock); + + list_for_each_entry(host1x, &devices, list) { + err = host1x_add_client(host1x, client); + if (!err) { + mutex_unlock(&devices_lock); + return 0; + } + } + + mutex_unlock(&devices_lock); + + mutex_lock(&clients_lock); + list_add_tail(&client->list, &clients); + mutex_unlock(&clients_lock); + + return 0; +} +EXPORT_SYMBOL(__host1x_client_register); + +/** + * host1x_client_unregister() - unregister a host1x client + * @client: host1x client + * + * Removes a host1x client from its host1x controller instance. If a logical + * device has already been initialized, it will be torn down. + */ +void host1x_client_unregister(struct host1x_client *client) +{ + struct host1x_client *c; + struct host1x *host1x; + int err; + + mutex_lock(&devices_lock); + + list_for_each_entry(host1x, &devices, list) { + err = host1x_del_client(host1x, client); + if (!err) { + mutex_unlock(&devices_lock); + return; + } + } + + mutex_unlock(&devices_lock); + mutex_lock(&clients_lock); + + list_for_each_entry(c, &clients, list) { + if (c == client) { + list_del_init(&c->list); + break; + } + } + + mutex_unlock(&clients_lock); + + host1x_bo_cache_destroy(&client->cache); +} +EXPORT_SYMBOL(host1x_client_unregister); + +int host1x_client_suspend(struct host1x_client *client) +{ + int err = 0; + + mutex_lock(&client->lock); + + if (client->usecount == 1) { + if (client->ops && client->ops->suspend) { + err = client->ops->suspend(client); + if (err < 0) + goto unlock; + } + } + + client->usecount--; + dev_dbg(client->dev, "use count: %u\n", client->usecount); + + if (client->parent) { + err = host1x_client_suspend(client->parent); + if (err < 0) + goto resume; + } + + goto unlock; + +resume: + if (client->usecount == 0) + if (client->ops && client->ops->resume) + client->ops->resume(client); + + client->usecount++; +unlock: + mutex_unlock(&client->lock); + return err; +} +EXPORT_SYMBOL(host1x_client_suspend); + +int host1x_client_resume(struct host1x_client *client) +{ + int err = 0; + + mutex_lock(&client->lock); + + if (client->parent) { + err = host1x_client_resume(client->parent); + if (err < 0) + goto unlock; + } + + if (client->usecount == 0) { + if (client->ops && client->ops->resume) { + err = client->ops->resume(client); + if (err < 0) + goto suspend; + } + } + + client->usecount++; + dev_dbg(client->dev, "use count: %u\n", client->usecount); + + goto unlock; + +suspend: + if (client->parent) + host1x_client_suspend(client->parent); +unlock: + mutex_unlock(&client->lock); + return err; +} +EXPORT_SYMBOL(host1x_client_resume); + +struct host1x_bo_mapping *host1x_bo_pin(struct device *dev, struct host1x_bo *bo, + enum dma_data_direction dir, + struct host1x_bo_cache *cache) +{ + struct host1x_bo_mapping *mapping; + + if (cache) { + mutex_lock(&cache->lock); + + list_for_each_entry(mapping, &cache->mappings, entry) { + if (mapping->bo == bo && mapping->direction == dir) { + kref_get(&mapping->ref); + goto unlock; + } + } + } + + mapping = bo->ops->pin(dev, bo, dir); + if (IS_ERR(mapping)) + goto unlock; + + spin_lock(&mapping->bo->lock); + list_add_tail(&mapping->list, &bo->mappings); + spin_unlock(&mapping->bo->lock); + + if (cache) { + INIT_LIST_HEAD(&mapping->entry); + mapping->cache = cache; + + list_add_tail(&mapping->entry, &cache->mappings); + + /* bump reference count to track the copy in the cache */ + kref_get(&mapping->ref); + } + +unlock: + if (cache) + mutex_unlock(&cache->lock); + + return mapping; +} +EXPORT_SYMBOL(host1x_bo_pin); + +static void __host1x_bo_unpin(struct kref *ref) +{ + struct host1x_bo_mapping *mapping = to_host1x_bo_mapping(ref); + + /* + * When the last reference of the mapping goes away, make sure to remove the mapping from + * the cache. + */ + if (mapping->cache) + list_del(&mapping->entry); + + spin_lock(&mapping->bo->lock); + list_del(&mapping->list); + spin_unlock(&mapping->bo->lock); + + mapping->bo->ops->unpin(mapping); +} + +void host1x_bo_unpin(struct host1x_bo_mapping *mapping) +{ + struct host1x_bo_cache *cache = mapping->cache; + + if (cache) + mutex_lock(&cache->lock); + + kref_put(&mapping->ref, __host1x_bo_unpin); + + if (cache) + mutex_unlock(&cache->lock); +} +EXPORT_SYMBOL(host1x_bo_unpin); diff --git a/drivers/gpu/host1x/bus.h b/drivers/gpu/host1x/bus.h new file mode 100644 index 000000000000..a80ceadfeb34 --- /dev/null +++ b/drivers/gpu/host1x/bus.h @@ -0,0 +1,18 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2012 Avionic Design GmbH + * Copyright (C) 2012-2013, NVIDIA Corporation + */ + +#ifndef HOST1X_BUS_H +#define HOST1X_BUS_H + +struct bus_type; +struct host1x; + +extern const struct bus_type host1x_bus_type; + +int host1x_register(struct host1x *host1x); +int host1x_unregister(struct host1x *host1x); + +#endif diff --git a/drivers/gpu/host1x/cdma.c b/drivers/gpu/host1x/cdma.c index de72172d3b5f..ba2e572567c0 100644 --- a/drivers/gpu/host1x/cdma.c +++ b/drivers/gpu/host1x/cdma.c @@ -1,25 +1,15 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Tegra host1x Command DMA * * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ #include <asm/cacheflush.h> #include <linux/device.h> #include <linux/dma-mapping.h> +#include <linux/host1x.h> #include <linux/interrupt.h> #include <linux/kernel.h> #include <linux/kfifo.h> @@ -30,7 +20,6 @@ #include "channel.h" #include "dev.h" #include "debug.h" -#include "host1x_bo.h" #include "job.h" /* @@ -41,7 +30,17 @@ * means that the push buffer is full, not empty. */ -#define HOST1X_PUSHBUFFER_SLOTS 512 +/* + * Typically the commands written into the push buffer are a pair of words. We + * use slots to represent each of these pairs and to simplify things. Note the + * strange number of slots allocated here. 512 slots will fit exactly within a + * single memory page. We also need one additional word at the end of the push + * buffer for the RESTART opcode that will instruct the CDMA to jump back to + * the beginning of the push buffer. With 512 slots, this means that we'll use + * 2 memory pages and waste 4092 bytes of the second page that will never be + * used. + */ +#define HOST1X_PUSHBUFFER_SLOTS 511 /* * Clean up push buffer resources @@ -51,9 +50,15 @@ static void host1x_pushbuffer_destroy(struct push_buffer *pb) struct host1x_cdma *cdma = pb_to_cdma(pb); struct host1x *host1x = cdma_to_host1x(cdma); - if (pb->phys != 0) - dma_free_writecombine(host1x->dev, pb->size_bytes + 4, - pb->mapped, pb->phys); + if (!pb->mapped) + return; + + if (host1x->domain) { + iommu_unmap(host1x->domain, pb->dma, pb->alloc_size); + free_iova(&host1x->iova, iova_pfn(&host1x->iova, pb->dma)); + } + + dma_free_wc(host1x->dev, pb->alloc_size, pb->mapped, pb->phys); pb->mapped = NULL; pb->phys = 0; @@ -66,28 +71,64 @@ static int host1x_pushbuffer_init(struct push_buffer *pb) { struct host1x_cdma *cdma = pb_to_cdma(pb); struct host1x *host1x = cdma_to_host1x(cdma); + struct iova *alloc; + u32 size; + int err; pb->mapped = NULL; pb->phys = 0; - pb->size_bytes = HOST1X_PUSHBUFFER_SLOTS * 8; + pb->size = HOST1X_PUSHBUFFER_SLOTS * 8; + + size = pb->size + 4; /* initialize buffer pointers */ - pb->fence = pb->size_bytes - 8; + pb->fence = pb->size - 8; pb->pos = 0; - /* allocate and map pushbuffer memory */ - pb->mapped = dma_alloc_writecombine(host1x->dev, pb->size_bytes + 4, - &pb->phys, GFP_KERNEL); - if (!pb->mapped) - goto fail; + if (host1x->domain) { + unsigned long shift; + + size = iova_align(&host1x->iova, size); + + pb->mapped = dma_alloc_wc(host1x->dev, size, &pb->phys, + GFP_KERNEL); + if (!pb->mapped) + return -ENOMEM; + + shift = iova_shift(&host1x->iova); + alloc = alloc_iova(&host1x->iova, size >> shift, + host1x->iova_end >> shift, true); + if (!alloc) { + err = -ENOMEM; + goto iommu_free_mem; + } + + pb->dma = iova_dma_addr(&host1x->iova, alloc); + err = iommu_map(host1x->domain, pb->dma, pb->phys, size, + IOMMU_READ, GFP_KERNEL); + if (err) + goto iommu_free_iova; + } else { + pb->mapped = dma_alloc_wc(host1x->dev, size, &pb->phys, + GFP_KERNEL); + if (!pb->mapped) + return -ENOMEM; + + pb->dma = pb->phys; + } + + pb->alloc_size = size; host1x_hw_pushbuffer_init(host1x, pb); return 0; -fail: - host1x_pushbuffer_destroy(pb); - return -ENOMEM; +iommu_free_iova: + __free_iova(&host1x->iova, alloc); +iommu_free_mem: + dma_free_wc(host1x->dev, size, pb->mapped, pb->phys); + + return err; } /* @@ -96,12 +137,15 @@ fail: */ static void host1x_pushbuffer_push(struct push_buffer *pb, u32 op1, u32 op2) { - u32 pos = pb->pos; - u32 *p = (u32 *)((u32)pb->mapped + pos); - WARN_ON(pos == pb->fence); + u32 *p = (u32 *)((void *)pb->mapped + pb->pos); + + WARN_ON(pb->pos == pb->fence); *(p++) = op1; *(p++) = op2; - pb->pos = (pos + 8) & (pb->size_bytes - 1); + pb->pos += 8; + + if (pb->pos >= pb->size) + pb->pos -= pb->size; } /* @@ -111,7 +155,10 @@ static void host1x_pushbuffer_push(struct push_buffer *pb, u32 op1, u32 op2) static void host1x_pushbuffer_pop(struct push_buffer *pb, unsigned int slots) { /* Advance the next write position */ - pb->fence = (pb->fence + slots * 8) & (pb->size_bytes - 1); + pb->fence += slots * 8; + + if (pb->fence >= pb->size) + pb->fence -= pb->size; } /* @@ -119,7 +166,12 @@ static void host1x_pushbuffer_pop(struct push_buffer *pb, unsigned int slots) */ static u32 host1x_pushbuffer_space(struct push_buffer *pb) { - return ((pb->fence - pb->pos) & (pb->size_bytes - 1)) / 8; + unsigned int fence = pb->fence; + + if (pb->fence < pb->pos) + fence += pb->size; + + return (fence - pb->pos) / 8; } /* @@ -134,14 +186,19 @@ unsigned int host1x_cdma_wait_locked(struct host1x_cdma *cdma, enum cdma_event event) { for (;;) { + struct push_buffer *pb = &cdma->push_buffer; unsigned int space; - if (event == CDMA_EVENT_SYNC_QUEUE_EMPTY) + switch (event) { + case CDMA_EVENT_SYNC_QUEUE_EMPTY: space = list_empty(&cdma->sync_queue) ? 1 : 0; - else if (event == CDMA_EVENT_PUSH_BUFFER_SPACE) { - struct push_buffer *pb = &cdma->push_buffer; + break; + + case CDMA_EVENT_PUSH_BUFFER_SPACE: space = host1x_pushbuffer_space(pb); - } else { + break; + + default: WARN_ON(1); return -EINVAL; } @@ -159,31 +216,68 @@ unsigned int host1x_cdma_wait_locked(struct host1x_cdma *cdma, mutex_lock(&cdma->lock); continue; } + cdma->event = event; mutex_unlock(&cdma->lock); - down(&cdma->sem); + wait_for_completion(&cdma->complete); mutex_lock(&cdma->lock); } + return 0; } /* + * Sleep (if necessary) until the push buffer has enough free space. + * + * Must be called with the cdma lock held. + */ +static int host1x_cdma_wait_pushbuffer_space(struct host1x *host1x, + struct host1x_cdma *cdma, + unsigned int needed) +{ + while (true) { + struct push_buffer *pb = &cdma->push_buffer; + unsigned int space; + + space = host1x_pushbuffer_space(pb); + if (space >= needed) + break; + + trace_host1x_wait_cdma(dev_name(cdma_to_channel(cdma)->dev), + CDMA_EVENT_PUSH_BUFFER_SPACE); + + /* If somebody has managed to already start waiting, yield */ + if (cdma->event != CDMA_EVENT_NONE) { + mutex_unlock(&cdma->lock); + schedule(); + mutex_lock(&cdma->lock); + continue; + } + + cdma->event = CDMA_EVENT_PUSH_BUFFER_SPACE; + + mutex_unlock(&cdma->lock); + wait_for_completion(&cdma->complete); + mutex_lock(&cdma->lock); + } + + return 0; +} +/* * Start timer that tracks the time spent by the job. * Must be called with the cdma lock held. */ static void cdma_start_timer_locked(struct host1x_cdma *cdma, struct host1x_job *job) { - struct host1x *host = cdma_to_host1x(cdma); - if (cdma->timeout.client) { /* timer already started */ return; } cdma->timeout.client = job->client; - cdma->timeout.syncpt = host1x_syncpt_get(host, job->syncpt_id); + cdma->timeout.syncpt = job->syncpt; cdma->timeout.syncpt_val = job->syncpt_end; cdma->timeout.start_ktime = ktime_get(); @@ -198,7 +292,7 @@ static void cdma_start_timer_locked(struct host1x_cdma *cdma, static void stop_cdma_timer_locked(struct host1x_cdma *cdma) { cancel_delayed_work(&cdma->timeout.wq); - cdma->timeout.client = 0; + cdma->timeout.client = NULL; } /* @@ -214,26 +308,22 @@ static void stop_cdma_timer_locked(struct host1x_cdma *cdma) static void update_cdma_locked(struct host1x_cdma *cdma) { bool signal = false; - struct host1x *host1x = cdma_to_host1x(cdma); struct host1x_job *job, *n; - /* If CDMA is stopped, queue is cleared and we can return */ - if (!cdma->running) - return; - /* * Walk the sync queue, reading the sync point registers as necessary, * to consume as many sync queue entries as possible without blocking */ list_for_each_entry_safe(job, n, &cdma->sync_queue, list) { - struct host1x_syncpt *sp = - host1x_syncpt_get(host1x, job->syncpt_id); + struct host1x_syncpt *sp = job->syncpt; /* Check whether this syncpt has completed, and bail if not */ - if (!host1x_syncpt_is_expired(sp, job->syncpt_end)) { + if (!host1x_syncpt_is_expired(sp, job->syncpt_end) && + !job->cancelled) { /* Start timer on next pending syncpt */ if (job->timeout) cdma_start_timer_locked(cdma, job); + break; } @@ -247,7 +337,9 @@ static void update_cdma_locked(struct host1x_cdma *cdma) /* Pop push buffer slots */ if (job->num_slots) { struct push_buffer *pb = &cdma->push_buffer; + host1x_pushbuffer_pop(pb, job->num_slots); + if (cdma->event == CDMA_EVENT_PUSH_BUFFER_SPACE) signal = true; } @@ -262,18 +354,16 @@ static void update_cdma_locked(struct host1x_cdma *cdma) if (signal) { cdma->event = CDMA_EVENT_NONE; - up(&cdma->sem); + complete(&cdma->complete); } } void host1x_cdma_update_sync_queue(struct host1x_cdma *cdma, struct device *dev) { - u32 restart_addr; - u32 syncpt_incrs; - struct host1x_job *job = NULL; - u32 syncpt_val; struct host1x *host1x = cdma_to_host1x(cdma); + u32 restart_addr, syncpt_incrs, syncpt_val; + struct host1x_job *job, *next_job = NULL; syncpt_val = host1x_syncpt_load(cdma->timeout.syncpt); @@ -291,40 +381,40 @@ void host1x_cdma_update_sync_queue(struct host1x_cdma *cdma, __func__); list_for_each_entry(job, &cdma->sync_queue, list) { - if (syncpt_val < job->syncpt_end) - break; + if (syncpt_val < job->syncpt_end) { + + if (!list_is_last(&job->list, &cdma->sync_queue)) + next_job = list_next_entry(job, list); + + goto syncpt_incr; + } host1x_job_dump(dev, job); } + /* all jobs have been completed */ + job = NULL; + +syncpt_incr: + /* - * Walk the sync_queue, first incrementing with the CPU syncpts that - * are partially executed (the first buffer) or fully skipped while - * still in the current context (slots are also NOP-ed). + * Increment with CPU the remaining syncpts of a partially executed job. * - * At the point contexts are interleaved, syncpt increments must be - * done inline with the pushbuffer from a GATHER buffer to maintain - * the order (slots are modified to be a GATHER of syncpt incrs). - * - * Note: save in restart_addr the location where the timed out buffer - * started in the PB, so we can start the refetch from there (with the - * modified NOP-ed PB slots). This lets things appear to have completed - * properly for this buffer and resources are freed. + * CDMA will continue execution starting with the next job or will get + * into idle state. */ - - dev_dbg(dev, "%s: perform CPU incr on pending same ctx buffers\n", - __func__); - - if (!list_empty(&cdma->sync_queue)) - restart_addr = job->first_get; + if (next_job) + restart_addr = next_job->first_get; else restart_addr = cdma->last_pos; - /* do CPU increments as long as this context continues */ - list_for_each_entry_from(job, &cdma->sync_queue, list) { - /* different context, gets us out of this loop */ - if (job->client != cdma->timeout.client) - break; + if (!job) + goto resume; + + /* do CPU increments for the remaining syncpts */ + if (job->syncpt_recovery) { + dev_dbg(dev, "%s: perform CPU incr on pending buffers\n", + __func__); /* won't need a timeout when replayed */ job->timeout = 0; @@ -339,23 +429,74 @@ void host1x_cdma_update_sync_queue(struct host1x_cdma *cdma, syncpt_incrs, job->syncpt_end, job->num_slots); - syncpt_val += syncpt_incrs; - } + dev_dbg(dev, "%s: finished sync_queue modification\n", + __func__); + } else { + struct host1x_job *failed_job = job; + + host1x_job_dump(dev, job); + + host1x_syncpt_set_locked(job->syncpt); + failed_job->cancelled = true; + + list_for_each_entry_continue(job, &cdma->sync_queue, list) { + unsigned int i; + + if (job->syncpt != failed_job->syncpt) + continue; + + for (i = 0; i < job->num_slots; i++) { + unsigned int slot = (job->first_get/8 + i) % + HOST1X_PUSHBUFFER_SLOTS; + u32 *mapped = cdma->push_buffer.mapped; + + /* + * Overwrite opcodes with 0 word writes + * to offset 0xbad. This does nothing but + * has a easily detected signature in debug + * traces. + * + * On systems with MLOCK enforcement enabled, + * the above 0 word writes would fall foul of + * the enforcement. As such, in the first slot + * put a RESTART_W opcode to the beginning + * of the next job. We don't use this for older + * chips since those only support the RESTART + * opcode with inconvenient alignment requirements. + */ + if (i == 0 && host1x->info->has_wide_gather) { + unsigned int next_job = (job->first_get/8 + job->num_slots) + % HOST1X_PUSHBUFFER_SLOTS; + mapped[2*slot+0] = (0xd << 28) | (next_job * 2); + mapped[2*slot+1] = 0x0; + } else { + mapped[2*slot+0] = 0x1bad0000; + mapped[2*slot+1] = 0x1bad0000; + } + } - /* The following sumbits from the same client may be dependent on the - * failed submit and therefore they may fail. Force a small timeout - * to make the queue cleanup faster */ + job->cancelled = true; + } - list_for_each_entry_from(job, &cdma->sync_queue, list) - if (job->client == cdma->timeout.client) - job->timeout = min_t(unsigned int, job->timeout, 500); + wmb(); - dev_dbg(dev, "%s: finished sync_queue modification\n", __func__); + update_cdma_locked(cdma); + } +resume: /* roll back DMAGET and start up channel again */ host1x_hw_cdma_resume(host1x, cdma, restart_addr); } +static void cdma_update_work(struct work_struct *work) +{ + struct host1x_cdma *cdma = container_of(work, struct host1x_cdma, update_work); + + mutex_lock(&cdma->lock); + update_cdma_locked(cdma); + mutex_unlock(&cdma->lock); +} + /* * Create a cdma */ @@ -364,7 +505,8 @@ int host1x_cdma_init(struct host1x_cdma *cdma) int err; mutex_init(&cdma->lock); - sema_init(&cdma->sem, 0); + init_completion(&cdma->complete); + INIT_WORK(&cdma->update_work, cdma_update_work); INIT_LIST_HEAD(&cdma->sync_queue); @@ -375,6 +517,7 @@ int host1x_cdma_init(struct host1x_cdma *cdma) err = host1x_pushbuffer_init(&cdma->push_buffer); if (err) return err; + return 0; } @@ -406,18 +549,29 @@ int host1x_cdma_begin(struct host1x_cdma *cdma, struct host1x_job *job) mutex_lock(&cdma->lock); + /* + * Check if syncpoint was locked due to previous job timeout. + * This needs to be done within the cdma lock to avoid a race + * with the timeout handler. + */ + if (job->syncpt->locked) { + mutex_unlock(&cdma->lock); + return -EPERM; + } + if (job->timeout) { /* init state on first submit with timeout value */ if (!cdma->timeout.initialized) { int err; - err = host1x_hw_cdma_timeout_init(host1x, cdma, - job->syncpt_id); + + err = host1x_hw_cdma_timeout_init(host1x, cdma); if (err) { mutex_unlock(&cdma->lock); return err; } } } + if (!cdma->running) host1x_hw_cdma_start(host1x, cdma); @@ -435,7 +589,6 @@ int host1x_cdma_begin(struct host1x_cdma *cdma, struct host1x_job *job) */ void host1x_cdma_push(struct host1x_cdma *cdma, u32 op1, u32 op2) { - struct host1x *host1x = cdma_to_host1x(cdma); struct push_buffer *pb = &cdma->push_buffer; u32 slots_free = cdma->slots_free; @@ -443,17 +596,62 @@ void host1x_cdma_push(struct host1x_cdma *cdma, u32 op1, u32 op2) trace_host1x_cdma_push(dev_name(cdma_to_channel(cdma)->dev), op1, op2); - if (slots_free == 0) { - host1x_hw_cdma_flush(host1x, cdma); + if (slots_free == 0) slots_free = host1x_cdma_wait_locked(cdma, CDMA_EVENT_PUSH_BUFFER_SPACE); - } + cdma->slots_free = slots_free - 1; cdma->slots_used++; host1x_pushbuffer_push(pb, op1, op2); } /* + * Push four words into two consecutive push buffer slots. Note that extra + * care needs to be taken not to split the two slots across the end of the + * push buffer. Otherwise the RESTART opcode at the end of the push buffer + * that ensures processing will restart at the beginning will break up the + * four words. + * + * Blocks as necessary if the push buffer is full. + */ +void host1x_cdma_push_wide(struct host1x_cdma *cdma, u32 op1, u32 op2, + u32 op3, u32 op4) +{ + struct host1x_channel *channel = cdma_to_channel(cdma); + struct host1x *host1x = cdma_to_host1x(cdma); + struct push_buffer *pb = &cdma->push_buffer; + unsigned int space, needed = 2, extra = 0; + + if (host1x_debug_trace_cmdbuf) + trace_host1x_cdma_push_wide(dev_name(channel->dev), op1, op2, + op3, op4); + + /* compute number of extra slots needed for padding */ + if (pb->pos + 16 > pb->size) { + extra = (pb->size - pb->pos) / 8; + needed += extra; + } + + host1x_cdma_wait_pushbuffer_space(host1x, cdma, needed); + space = host1x_pushbuffer_space(pb); + + cdma->slots_free = space - needed; + cdma->slots_used += needed; + + if (extra > 0) { + /* + * If there isn't enough space at the tail of the pushbuffer, + * insert a RESTART(0) here to go back to the beginning. + * The code above adjusted the indexes appropriately. + */ + host1x_pushbuffer_push(pb, (0x5 << 28), 0xdead0000); + } + + host1x_pushbuffer_push(pb, op1, op2); + host1x_pushbuffer_push(pb, op3, op4); +} + +/* * End a cdma submit * Kick off DMA, add job to the sync queue, and a number of slots to be freed * from the pushbuffer. The handles for a submit must all be pinned at the same @@ -485,7 +683,5 @@ void host1x_cdma_end(struct host1x_cdma *cdma, */ void host1x_cdma_update(struct host1x_cdma *cdma) { - mutex_lock(&cdma->lock); - update_cdma_locked(cdma); - mutex_unlock(&cdma->lock); + schedule_work(&cdma->update_work); } diff --git a/drivers/gpu/host1x/cdma.h b/drivers/gpu/host1x/cdma.h index 313c4b784348..7fd8168af4f9 100644 --- a/drivers/gpu/host1x/cdma.h +++ b/drivers/gpu/host1x/cdma.h @@ -1,27 +1,17 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ /* * Tegra host1x Command DMA * * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ #ifndef __HOST1X_CDMA_H #define __HOST1X_CDMA_H #include <linux/sched.h> -#include <linux/semaphore.h> +#include <linux/completion.h> #include <linux/list.h> +#include <linux/workqueue.h> struct host1x_syncpt; struct host1x_userctx_timeout; @@ -42,11 +32,13 @@ struct host1x_job; */ struct push_buffer { - u32 *mapped; /* mapped pushbuffer memory */ + void *mapped; /* mapped pushbuffer memory */ + dma_addr_t dma; /* device address of pushbuffer */ dma_addr_t phys; /* physical address of pushbuffer */ u32 fence; /* index we've written */ u32 pos; /* index to write to */ - u32 size_bytes; + u32 size; + u32 alloc_size; }; struct buffer_timeout { @@ -56,7 +48,7 @@ struct buffer_timeout { u32 syncpt_val; /* syncpt value when completed */ ktime_t start_ktime; /* starting time */ /* context timeout information */ - int client; + struct host1x_client *client; }; enum cdma_event { @@ -67,8 +59,8 @@ enum cdma_event { struct host1x_cdma { struct mutex lock; /* controls access to shared state */ - struct semaphore sem; /* signalled when event occurs */ - enum cdma_event event; /* event that sem is waiting for */ + struct completion complete; /* signalled when event occurs */ + enum cdma_event event; /* event that complete is waiting for */ unsigned int slots_used; /* pb slots used in current submit */ unsigned int slots_free; /* pb slots free in current submit */ unsigned int first_get; /* DMAGET value, where submit begins */ @@ -78,6 +70,7 @@ struct host1x_cdma { struct buffer_timeout timeout; /* channel's timeout state/wq */ bool running; bool torndown; + struct work_struct update_work; }; #define cdma_to_channel(cdma) container_of(cdma, struct host1x_channel, cdma) @@ -86,9 +79,10 @@ struct host1x_cdma { int host1x_cdma_init(struct host1x_cdma *cdma); int host1x_cdma_deinit(struct host1x_cdma *cdma); -void host1x_cdma_stop(struct host1x_cdma *cdma); int host1x_cdma_begin(struct host1x_cdma *cdma, struct host1x_job *job); void host1x_cdma_push(struct host1x_cdma *cdma, u32 op1, u32 op2); +void host1x_cdma_push_wide(struct host1x_cdma *cdma, u32 op1, u32 op2, + u32 op3, u32 op4); void host1x_cdma_end(struct host1x_cdma *cdma, struct host1x_job *job); void host1x_cdma_update(struct host1x_cdma *cdma); void host1x_cdma_peek(struct host1x_cdma *cdma, u32 dmaget, int slot, diff --git a/drivers/gpu/host1x/channel.c b/drivers/gpu/host1x/channel.c index 83ea51b9f0fc..08077afe4cde 100644 --- a/drivers/gpu/host1x/channel.c +++ b/drivers/gpu/host1x/channel.c @@ -1,19 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Tegra host1x Channel * * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ #include <linux/slab.h> @@ -24,103 +13,174 @@ #include "job.h" /* Constructor for the host1x device list */ -int host1x_channel_list_init(struct host1x *host) +int host1x_channel_list_init(struct host1x_channel_list *chlist, + unsigned int num_channels) { - INIT_LIST_HEAD(&host->chlist.list); - mutex_init(&host->chlist_mutex); - - if (host->info->nb_channels > BITS_PER_LONG) { - WARN(1, "host1x hardware has more channels than supported by the driver\n"); - return -ENOSYS; + chlist->channels = kcalloc(num_channels, sizeof(struct host1x_channel), + GFP_KERNEL); + if (!chlist->channels) + return -ENOMEM; + + chlist->allocated_channels = bitmap_zalloc(num_channels, GFP_KERNEL); + if (!chlist->allocated_channels) { + kfree(chlist->channels); + return -ENOMEM; } + mutex_init(&chlist->lock); + return 0; } +void host1x_channel_list_free(struct host1x_channel_list *chlist) +{ + bitmap_free(chlist->allocated_channels); + kfree(chlist->channels); +} + int host1x_job_submit(struct host1x_job *job) { struct host1x *host = dev_get_drvdata(job->channel->dev->parent); return host1x_hw_channel_submit(host, job); } +EXPORT_SYMBOL(host1x_job_submit); struct host1x_channel *host1x_channel_get(struct host1x_channel *channel) { - int err = 0; + kref_get(&channel->refcount); + + return channel; +} +EXPORT_SYMBOL(host1x_channel_get); + +/** + * host1x_channel_get_index() - Attempt to get channel reference by index + * @host: Host1x device object + * @index: Index of channel + * + * If channel number @index is currently allocated, increase its refcount + * and return a pointer to it. Otherwise, return NULL. + */ +struct host1x_channel *host1x_channel_get_index(struct host1x *host, + unsigned int index) +{ + struct host1x_channel *ch = &host->channel_list.channels[index]; + + if (!kref_get_unless_zero(&ch->refcount)) + return NULL; + + return ch; +} + +void host1x_channel_stop(struct host1x_channel *channel) +{ + struct host1x *host = dev_get_drvdata(channel->dev->parent); + + host1x_hw_cdma_stop(host, &channel->cdma); +} +EXPORT_SYMBOL(host1x_channel_stop); + +/** + * host1x_channel_stop_all() - disable CDMA on allocated channels + * @host: host1x instance + * + * Stop CDMA on allocated channels + */ +void host1x_channel_stop_all(struct host1x *host) +{ + struct host1x_channel_list *chlist = &host->channel_list; + int bit; - mutex_lock(&channel->reflock); + mutex_lock(&chlist->lock); - if (channel->refcount == 0) - err = host1x_cdma_init(&channel->cdma); + for_each_set_bit(bit, chlist->allocated_channels, host->info->nb_channels) + host1x_channel_stop(&chlist->channels[bit]); - if (!err) - channel->refcount++; + mutex_unlock(&chlist->lock); +} - mutex_unlock(&channel->reflock); +static void release_channel(struct kref *kref) +{ + struct host1x_channel *channel = + container_of(kref, struct host1x_channel, refcount); + struct host1x *host = dev_get_drvdata(channel->dev->parent); + struct host1x_channel_list *chlist = &host->channel_list; + + host1x_hw_cdma_stop(host, &channel->cdma); + host1x_cdma_deinit(&channel->cdma); - return err ? NULL : channel; + clear_bit(channel->id, chlist->allocated_channels); } void host1x_channel_put(struct host1x_channel *channel) { - mutex_lock(&channel->reflock); + kref_put(&channel->refcount, release_channel); +} +EXPORT_SYMBOL(host1x_channel_put); - if (channel->refcount == 1) { - struct host1x *host = dev_get_drvdata(channel->dev->parent); +static struct host1x_channel *acquire_unused_channel(struct host1x *host) +{ + struct host1x_channel_list *chlist = &host->channel_list; + unsigned int max_channels = host->info->nb_channels; + unsigned int index; + + mutex_lock(&chlist->lock); - host1x_hw_cdma_stop(host, &channel->cdma); - host1x_cdma_deinit(&channel->cdma); + index = find_first_zero_bit(chlist->allocated_channels, max_channels); + if (index >= max_channels) { + mutex_unlock(&chlist->lock); + dev_err(host->dev, "failed to find free channel\n"); + return NULL; } - channel->refcount--; + chlist->channels[index].id = index; + + set_bit(index, chlist->allocated_channels); - mutex_unlock(&channel->reflock); + mutex_unlock(&chlist->lock); + + return &chlist->channels[index]; } -struct host1x_channel *host1x_channel_request(struct device *dev) +/** + * host1x_channel_request() - Allocate a channel + * @client: Host1x client this channel will be used to send commands to + * + * Allocates a new host1x channel for @client. May return NULL if CDMA + * initialization fails. + */ +struct host1x_channel *host1x_channel_request(struct host1x_client *client) { - struct host1x *host = dev_get_drvdata(dev->parent); - int max_channels = host->info->nb_channels; - struct host1x_channel *channel = NULL; - int index, err; + struct host1x *host = dev_get_drvdata(client->dev->parent); + struct host1x_channel_list *chlist = &host->channel_list; + struct host1x_channel *channel; + int err; - mutex_lock(&host->chlist_mutex); + channel = acquire_unused_channel(host); + if (!channel) + return NULL; - index = find_first_zero_bit(&host->allocated_channels, max_channels); - if (index >= max_channels) - goto fail; + kref_init(&channel->refcount); + mutex_init(&channel->submitlock); + channel->client = client; + channel->dev = client->dev; - channel = kzalloc(sizeof(*channel), GFP_KERNEL); - if (!channel) + err = host1x_hw_channel_init(host, channel, channel->id); + if (err < 0) goto fail; - err = host1x_hw_channel_init(host, channel, index); + err = host1x_cdma_init(&channel->cdma); if (err < 0) goto fail; - /* Link device to host1x_channel */ - channel->dev = dev; - - /* Add to channel list */ - list_add_tail(&channel->list, &host->chlist.list); - - host->allocated_channels |= BIT(index); - - mutex_unlock(&host->chlist_mutex); return channel; fail: - dev_err(dev, "failed to init channel\n"); - kfree(channel); - mutex_unlock(&host->chlist_mutex); - return NULL; -} + clear_bit(channel->id, chlist->allocated_channels); -void host1x_channel_free(struct host1x_channel *channel) -{ - struct host1x *host = dev_get_drvdata(channel->dev->parent); + dev_err(client->dev, "failed to initialize channel\n"); - host->allocated_channels &= ~BIT(channel->id); - list_del(&channel->list); - kfree(channel); + return NULL; } +EXPORT_SYMBOL(host1x_channel_request); diff --git a/drivers/gpu/host1x/channel.h b/drivers/gpu/host1x/channel.h index 48723b8eea42..d7aede204d83 100644 --- a/drivers/gpu/host1x/channel.h +++ b/drivers/gpu/host1x/channel.h @@ -1,52 +1,45 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ /* * Tegra host1x Channel * * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ #ifndef __HOST1X_CHANNEL_H #define __HOST1X_CHANNEL_H #include <linux/io.h> +#include <linux/kref.h> +#include <linux/mutex.h> #include "cdma.h" struct host1x; +struct host1x_channel; -struct host1x_channel { - struct list_head list; +struct host1x_channel_list { + struct host1x_channel *channels; - unsigned int refcount; + struct mutex lock; + unsigned long *allocated_channels; +}; + +struct host1x_channel { + struct kref refcount; unsigned int id; - struct mutex reflock; struct mutex submitlock; void __iomem *regs; + struct host1x_client *client; struct device *dev; struct host1x_cdma cdma; }; /* channel list operations */ -int host1x_channel_list_init(struct host1x *host); - -struct host1x_channel *host1x_channel_request(struct device *dev); -void host1x_channel_free(struct host1x_channel *channel); -struct host1x_channel *host1x_channel_get(struct host1x_channel *channel); -void host1x_channel_put(struct host1x_channel *channel); -int host1x_job_submit(struct host1x_job *job); - -#define host1x_for_each_channel(host, channel) \ - list_for_each_entry(channel, &host->chlist.list, list) +int host1x_channel_list_init(struct host1x_channel_list *chlist, + unsigned int num_channels); +void host1x_channel_list_free(struct host1x_channel_list *chlist); +struct host1x_channel *host1x_channel_get_index(struct host1x *host, + unsigned int index); +void host1x_channel_stop_all(struct host1x *host); #endif diff --git a/drivers/gpu/host1x/context.c b/drivers/gpu/host1x/context.c new file mode 100644 index 000000000000..a6f6779662a3 --- /dev/null +++ b/drivers/gpu/host1x/context.c @@ -0,0 +1,177 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2021, NVIDIA Corporation. + */ + +#include <linux/device.h> +#include <linux/kref.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/pid.h> +#include <linux/slab.h> + +#include "context.h" +#include "dev.h" + +static void host1x_memory_context_release(struct device *dev) +{ + /* context device is freed in host1x_memory_context_list_free() */ +} + +int host1x_memory_context_list_init(struct host1x *host1x) +{ + struct host1x_memory_context_list *cdl = &host1x->context_list; + struct device_node *node = host1x->dev->of_node; + struct host1x_memory_context *ctx; + unsigned int i; + int err; + + cdl->devs = NULL; + cdl->len = 0; + mutex_init(&cdl->lock); + + err = of_property_count_u32_elems(node, "iommu-map"); + if (err < 0) + return 0; + + cdl->len = err / 4; + cdl->devs = kcalloc(cdl->len, sizeof(*cdl->devs), GFP_KERNEL); + if (!cdl->devs) + return -ENOMEM; + + for (i = 0; i < cdl->len; i++) { + ctx = &cdl->devs[i]; + + ctx->host = host1x; + + device_initialize(&ctx->dev); + + /* + * Due to an issue with T194 NVENC, only 38 bits can be used. + * Anyway, 256GiB of IOVA ought to be enough for anyone. + */ + ctx->dma_mask = DMA_BIT_MASK(38); + ctx->dev.dma_mask = &ctx->dma_mask; + ctx->dev.coherent_dma_mask = ctx->dma_mask; + dev_set_name(&ctx->dev, "host1x-ctx.%d", i); + ctx->dev.bus = &host1x_context_device_bus_type; + ctx->dev.parent = host1x->dev; + ctx->dev.release = host1x_memory_context_release; + + ctx->dev.dma_parms = &ctx->dma_parms; + dma_set_max_seg_size(&ctx->dev, UINT_MAX); + + err = device_add(&ctx->dev); + if (err) { + dev_err(host1x->dev, "could not add context device %d: %d\n", i, err); + put_device(&ctx->dev); + goto unreg_devices; + } + + err = of_dma_configure_id(&ctx->dev, node, true, &i); + if (err) { + dev_err(host1x->dev, "IOMMU configuration failed for context device %d: %d\n", + i, err); + device_unregister(&ctx->dev); + goto unreg_devices; + } + + if (!tegra_dev_iommu_get_stream_id(&ctx->dev, &ctx->stream_id) || + !device_iommu_mapped(&ctx->dev)) { + dev_err(host1x->dev, "Context device %d has no IOMMU!\n", i); + device_unregister(&ctx->dev); + + /* + * This means that if IOMMU is disabled but context devices + * are defined in the device tree, Host1x will fail to probe. + * That's probably OK in this time and age. + */ + err = -EINVAL; + + goto unreg_devices; + } + } + + return 0; + +unreg_devices: + while (i--) + device_unregister(&cdl->devs[i].dev); + + kfree(cdl->devs); + cdl->devs = NULL; + cdl->len = 0; + + return err; +} + +void host1x_memory_context_list_free(struct host1x_memory_context_list *cdl) +{ + unsigned int i; + + for (i = 0; i < cdl->len; i++) + device_unregister(&cdl->devs[i].dev); + + kfree(cdl->devs); + cdl->len = 0; +} + +struct host1x_memory_context *host1x_memory_context_alloc(struct host1x *host1x, + struct device *dev, + struct pid *pid) +{ + struct host1x_memory_context_list *cdl = &host1x->context_list; + struct host1x_memory_context *free = NULL; + int i; + + if (!cdl->len) + return ERR_PTR(-EOPNOTSUPP); + + mutex_lock(&cdl->lock); + + for (i = 0; i < cdl->len; i++) { + struct host1x_memory_context *cd = &cdl->devs[i]; + + if (cd->dev.iommu->iommu_dev != dev->iommu->iommu_dev) + continue; + + if (cd->owner == pid) { + refcount_inc(&cd->ref); + mutex_unlock(&cdl->lock); + return cd; + } else if (!cd->owner && !free) { + free = cd; + } + } + + if (!free) { + mutex_unlock(&cdl->lock); + return ERR_PTR(-EBUSY); + } + + refcount_set(&free->ref, 1); + free->owner = get_pid(pid); + + mutex_unlock(&cdl->lock); + + return free; +} +EXPORT_SYMBOL_GPL(host1x_memory_context_alloc); + +void host1x_memory_context_get(struct host1x_memory_context *cd) +{ + refcount_inc(&cd->ref); +} +EXPORT_SYMBOL_GPL(host1x_memory_context_get); + +void host1x_memory_context_put(struct host1x_memory_context *cd) +{ + struct host1x_memory_context_list *cdl = &cd->host->context_list; + + if (refcount_dec_and_mutex_lock(&cd->ref, &cdl->lock)) { + put_pid(cd->owner); + cd->owner = NULL; + mutex_unlock(&cdl->lock); + } +} +EXPORT_SYMBOL_GPL(host1x_memory_context_put); diff --git a/drivers/gpu/host1x/context.h b/drivers/gpu/host1x/context.h new file mode 100644 index 000000000000..3e03bc1d3bac --- /dev/null +++ b/drivers/gpu/host1x/context.h @@ -0,0 +1,38 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Host1x context devices + * + * Copyright (c) 2020, NVIDIA Corporation. + */ + +#ifndef __HOST1X_CONTEXT_H +#define __HOST1X_CONTEXT_H + +#include <linux/mutex.h> +#include <linux/refcount.h> + +struct host1x; + +extern struct bus_type host1x_context_device_bus_type; + +struct host1x_memory_context_list { + struct mutex lock; + struct host1x_memory_context *devs; + unsigned int len; +}; + +#ifdef CONFIG_IOMMU_API +int host1x_memory_context_list_init(struct host1x *host1x); +void host1x_memory_context_list_free(struct host1x_memory_context_list *cdl); +#else +static inline int host1x_memory_context_list_init(struct host1x *host1x) +{ + return 0; +} + +static inline void host1x_memory_context_list_free(struct host1x_memory_context_list *cdl) +{ +} +#endif + +#endif diff --git a/drivers/gpu/host1x/context_bus.c b/drivers/gpu/host1x/context_bus.c new file mode 100644 index 000000000000..7cd0e1a5edd1 --- /dev/null +++ b/drivers/gpu/host1x/context_bus.c @@ -0,0 +1,26 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2021, NVIDIA Corporation. + */ + +#include <linux/device.h> +#include <linux/of.h> + +const struct bus_type host1x_context_device_bus_type = { + .name = "host1x-context", +}; +EXPORT_SYMBOL_GPL(host1x_context_device_bus_type); + +static int __init host1x_context_device_bus_init(void) +{ + int err; + + err = bus_register(&host1x_context_device_bus_type); + if (err < 0) { + pr_err("bus type registration failed: %d\n", err); + return err; + } + + return 0; +} +postcore_initcall(host1x_context_device_bus_init); diff --git a/drivers/gpu/host1x/debug.c b/drivers/gpu/host1x/debug.c index 3ec7d77de24d..6433c00d5d7e 100644 --- a/drivers/gpu/host1x/debug.c +++ b/drivers/gpu/host1x/debug.c @@ -1,21 +1,13 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Copyright (C) 2010 Google, Inc. * Author: Erik Gilling <konkers@android.com> * * Copyright (C) 2011-2013 NVIDIA Corporation - * - * This software is licensed under the terms of the GNU General Public - * License version 2, as published by the Free Software Foundation, and - * may be copied, distributed, and modified under those terms. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * */ #include <linux/debugfs.h> +#include <linux/pm_runtime.h> #include <linux/seq_file.h> #include <linux/uaccess.h> @@ -25,6 +17,8 @@ #include "debug.h" #include "channel.h" +static DEFINE_MUTEX(debug_lock); + unsigned int host1x_debug_trace_cmdbuf; static pid_t host1x_debug_force_timeout_pid; @@ -39,85 +33,126 @@ void host1x_debug_output(struct output *o, const char *fmt, ...) va_start(args, fmt); len = vsnprintf(o->buf, sizeof(o->buf), fmt, args); va_end(args); - o->fn(o->ctx, o->buf, len); + + o->fn(o->ctx, o->buf, len, false); +} + +void host1x_debug_cont(struct output *o, const char *fmt, ...) +{ + va_list args; + int len; + + va_start(args, fmt); + len = vsnprintf(o->buf, sizeof(o->buf), fmt, args); + va_end(args); + + o->fn(o->ctx, o->buf, len, true); } -static int show_channels(struct host1x_channel *ch, void *data, bool show_fifo) +static int show_channel(struct host1x_channel *ch, void *data, bool show_fifo) { struct host1x *m = dev_get_drvdata(ch->dev->parent); struct output *o = data; + int err; - mutex_lock(&ch->reflock); - if (ch->refcount) { - mutex_lock(&ch->cdma.lock); - if (show_fifo) - host1x_hw_show_channel_fifo(m, ch, o); - host1x_hw_show_channel_cdma(m, ch, o); - mutex_unlock(&ch->cdma.lock); - } - mutex_unlock(&ch->reflock); + err = pm_runtime_resume_and_get(m->dev); + if (err < 0) + return err; + + mutex_lock(&ch->cdma.lock); + mutex_lock(&debug_lock); + + if (show_fifo) + host1x_hw_show_channel_fifo(m, ch, o); + + host1x_hw_show_channel_cdma(m, ch, o); + + mutex_unlock(&debug_lock); + mutex_unlock(&ch->cdma.lock); + + pm_runtime_put(m->dev); return 0; } -static void show_syncpts(struct host1x *m, struct output *o) +static void show_syncpts(struct host1x *m, struct output *o, bool show_all) { - int i; + unsigned long irqflags; + struct list_head *pos; + unsigned int i; + int err; + host1x_debug_output(o, "---- syncpts ----\n"); + + err = pm_runtime_resume_and_get(m->dev); + if (err < 0) + return; + for (i = 0; i < host1x_syncpt_nb_pts(m); i++) { u32 max = host1x_syncpt_read_max(m->syncpt + i); u32 min = host1x_syncpt_load(m->syncpt + i); - if (!min && !max) + unsigned int waiters = 0; + + spin_lock_irqsave(&m->syncpt[i].fences.lock, irqflags); + list_for_each(pos, &m->syncpt[i].fences.list) + waiters++; + spin_unlock_irqrestore(&m->syncpt[i].fences.lock, irqflags); + + if (!kref_read(&m->syncpt[i].ref)) continue; - host1x_debug_output(o, "id %d (%s) min %d max %d\n", - i, m->syncpt[i].name, min, max); + + if (!show_all && !min && !max && !waiters) + continue; + + host1x_debug_output(o, + "id %u (%s) min %d max %d (%d waiters)\n", + i, m->syncpt[i].name, min, max, waiters); } for (i = 0; i < host1x_syncpt_nb_bases(m); i++) { u32 base_val; + base_val = host1x_syncpt_load_wait_base(m->syncpt + i); if (base_val) - host1x_debug_output(o, "waitbase id %d val %d\n", i, + host1x_debug_output(o, "waitbase id %u val %d\n", i, base_val); } + pm_runtime_put(m->dev); + host1x_debug_output(o, "\n"); } -static void show_all(struct host1x *m, struct output *o) +static void show_all(struct host1x *m, struct output *o, bool show_fifo) { - struct host1x_channel *ch; + unsigned int i; host1x_hw_show_mlocks(m, o); - show_syncpts(m, o); + show_syncpts(m, o, true); host1x_debug_output(o, "---- channels ----\n"); - host1x_for_each_channel(m, ch) - show_channels(ch, o, true); -} - -#ifdef CONFIG_DEBUG_FS -static void show_all_no_fifo(struct host1x *host1x, struct output *o) -{ - struct host1x_channel *ch; - - host1x_hw_show_mlocks(host1x, o); - show_syncpts(host1x, o); - host1x_debug_output(o, "---- channels ----\n"); + for (i = 0; i < m->info->nb_channels; ++i) { + struct host1x_channel *ch = host1x_channel_get_index(m, i); - host1x_for_each_channel(host1x, ch) - show_channels(ch, o, false); + if (ch) { + show_channel(ch, o, show_fifo); + host1x_channel_put(ch); + } + } } -static int host1x_debug_show_all(struct seq_file *s, void *unused) +static int host1x_debug_all_show(struct seq_file *s, void *unused) { struct output o = { .fn = write_to_seqfile, .ctx = s }; - show_all(s->private, &o); + + show_all(s->private, &o, true); + return 0; } +DEFINE_SHOW_ATTRIBUTE(host1x_debug_all); static int host1x_debug_show(struct seq_file *s, void *unused) { @@ -125,41 +160,17 @@ static int host1x_debug_show(struct seq_file *s, void *unused) .fn = write_to_seqfile, .ctx = s }; - show_all_no_fifo(s->private, &o); - return 0; -} - -static int host1x_debug_open_all(struct inode *inode, struct file *file) -{ - return single_open(file, host1x_debug_show_all, inode->i_private); -} -static const struct file_operations host1x_debug_all_fops = { - .open = host1x_debug_open_all, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; + show_all(s->private, &o, false); -static int host1x_debug_open(struct inode *inode, struct file *file) -{ - return single_open(file, host1x_debug_show, inode->i_private); + return 0; } +DEFINE_SHOW_ATTRIBUTE(host1x_debug); -static const struct file_operations host1x_debug_fops = { - .open = host1x_debug_open, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; - -void host1x_debug_init(struct host1x *host1x) +static void host1x_debugfs_init(struct host1x *host1x) { struct dentry *de = debugfs_create_dir("tegra-host1x", NULL); - if (!de) - return; - /* Store the created entry */ host1x->debugfs = de; @@ -180,31 +191,28 @@ void host1x_debug_init(struct host1x *host1x) &host1x_debug_force_timeout_channel); } -void host1x_debug_deinit(struct host1x *host1x) +static void host1x_debugfs_exit(struct host1x *host1x) { debugfs_remove_recursive(host1x->debugfs); } -#else + void host1x_debug_init(struct host1x *host1x) { + if (IS_ENABLED(CONFIG_DEBUG_FS)) + host1x_debugfs_init(host1x); } + void host1x_debug_deinit(struct host1x *host1x) { + if (IS_ENABLED(CONFIG_DEBUG_FS)) + host1x_debugfs_exit(host1x); } -#endif void host1x_debug_dump(struct host1x *host1x) { struct output o = { .fn = write_to_printk }; - show_all(host1x, &o); -} -void host1x_debug_dump_syncpts(struct host1x *host1x) -{ - struct output o = { - .fn = write_to_printk - }; - show_syncpts(host1x, &o); + show_all(host1x, &o, true); } diff --git a/drivers/gpu/host1x/debug.h b/drivers/gpu/host1x/debug.h index 4595b2e0799f..c43c61d876a9 100644 --- a/drivers/gpu/host1x/debug.h +++ b/drivers/gpu/host1x/debug.h @@ -1,19 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ /* * Tegra host1x Debug * * Copyright (c) 2011-2013 NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ #ifndef __HOST1X_DEBUG_H #define __HOST1X_DEBUG_H @@ -24,28 +13,33 @@ struct host1x; struct output { - void (*fn)(void *ctx, const char *str, size_t len); + void (*fn)(void *ctx, const char *str, size_t len, bool cont); void *ctx; char buf[256]; }; -static inline void write_to_seqfile(void *ctx, const char *str, size_t len) +static inline void write_to_seqfile(void *ctx, const char *str, size_t len, + bool cont) { seq_write((struct seq_file *)ctx, str, len); } -static inline void write_to_printk(void *ctx, const char *str, size_t len) +static inline void write_to_printk(void *ctx, const char *str, size_t len, + bool cont) { - pr_info("%s", str); + if (cont) + pr_cont("%s", str); + else + pr_info("%s", str); } void __printf(2, 3) host1x_debug_output(struct output *o, const char *fmt, ...); +void __printf(2, 3) host1x_debug_cont(struct output *o, const char *fmt, ...); extern unsigned int host1x_debug_trace_cmdbuf; void host1x_debug_init(struct host1x *host1x); void host1x_debug_deinit(struct host1x *host1x); void host1x_debug_dump(struct host1x *host1x); -void host1x_debug_dump_syncpts(struct host1x *host1x); #endif diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c index 28e28a23d444..3f475f0e6545 100644 --- a/drivers/gpu/host1x/dev.c +++ b/drivers/gpu/host1x/dev.c @@ -1,49 +1,60 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Tegra host1x driver * * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ -#include <linux/module.h> -#include <linux/list.h> -#include <linux/slab.h> -#include <linux/of.h> -#include <linux/of_device.h> #include <linux/clk.h> +#include <linux/delay.h> +#include <linux/dma-mapping.h> #include <linux/io.h> +#include <linux/list.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/of_platform.h> +#include <linux/platform_device.h> +#include <linux/pm_runtime.h> +#include <linux/slab.h> + +#include <soc/tegra/common.h> #define CREATE_TRACE_POINTS #include <trace/events/host1x.h> +#undef CREATE_TRACE_POINTS -#include "dev.h" -#include "intr.h" +#if IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU) +#include <asm/dma-iommu.h> +#endif + +#include "bus.h" #include "channel.h" +#include "context.h" #include "debug.h" +#include "dev.h" +#include "intr.h" + #include "hw/host1x01.h" -#include "host1x_client.h" +#include "hw/host1x02.h" +#include "hw/host1x04.h" +#include "hw/host1x05.h" +#include "hw/host1x06.h" +#include "hw/host1x07.h" +#include "hw/host1x08.h" + +void host1x_common_writel(struct host1x *host1x, u32 v, u32 r) +{ + writel(v, host1x->common_regs + r); +} -void host1x_set_drm_data(struct device *dev, void *data) +void host1x_hypervisor_writel(struct host1x *host1x, u32 v, u32 r) { - struct host1x *host1x = dev_get_drvdata(dev); - host1x->drm_data = data; + writel(v, host1x->hv_regs + r); } -void *host1x_get_drm_data(struct device *dev) +u32 host1x_hypervisor_readl(struct host1x *host1x, u32 r) { - struct host1x *host1x = dev_get_drvdata(dev); - return host1x->drm_data; + return readl(host1x->hv_regs + r); } void host1x_sync_writel(struct host1x *host1x, u32 v, u32 r) @@ -60,6 +71,15 @@ u32 host1x_sync_readl(struct host1x *host1x, u32 r) return readl(sync_regs + r); } +#ifdef CONFIG_64BIT +u64 host1x_sync_readq(struct host1x *host1x, u32 r) +{ + void __iomem *sync_regs = host1x->regs + host1x->info->sync_offset; + + return readq(sync_regs + r); +} +#endif + void host1x_ch_writel(struct host1x_channel *ch, u32 v, u32 r) { writel(v, ch->regs + r); @@ -71,58 +91,501 @@ u32 host1x_ch_readl(struct host1x_channel *ch, u32 r) } static const struct host1x_info host1x01_info = { - .nb_channels = 8, - .nb_pts = 32, - .nb_mlocks = 16, - .nb_bases = 8, - .init = host1x01_init, - .sync_offset = 0x3000, + .nb_channels = 8, + .nb_pts = 32, + .nb_mlocks = 16, + .nb_bases = 8, + .init = host1x01_init, + .sync_offset = 0x3000, + .dma_mask = DMA_BIT_MASK(32), + .has_wide_gather = false, + .has_hypervisor = false, + .num_sid_entries = 0, + .sid_table = NULL, + .reserve_vblank_syncpts = true, +}; + +static const struct host1x_info host1x02_info = { + .nb_channels = 9, + .nb_pts = 32, + .nb_mlocks = 16, + .nb_bases = 12, + .init = host1x02_init, + .sync_offset = 0x3000, + .dma_mask = DMA_BIT_MASK(32), + .has_wide_gather = false, + .has_hypervisor = false, + .num_sid_entries = 0, + .sid_table = NULL, + .reserve_vblank_syncpts = true, +}; + +static const struct host1x_info host1x04_info = { + .nb_channels = 12, + .nb_pts = 192, + .nb_mlocks = 16, + .nb_bases = 64, + .init = host1x04_init, + .sync_offset = 0x2100, + .dma_mask = DMA_BIT_MASK(34), + .has_wide_gather = false, + .has_hypervisor = false, + .num_sid_entries = 0, + .sid_table = NULL, + .reserve_vblank_syncpts = false, +}; + +static const struct host1x_info host1x05_info = { + .nb_channels = 14, + .nb_pts = 192, + .nb_mlocks = 16, + .nb_bases = 64, + .init = host1x05_init, + .sync_offset = 0x2100, + .dma_mask = DMA_BIT_MASK(34), + .has_wide_gather = false, + .has_hypervisor = false, + .num_sid_entries = 0, + .sid_table = NULL, + .reserve_vblank_syncpts = false, +}; + +static const struct host1x_sid_entry tegra186_sid_table[] = { + { /* SE1 */ .base = 0x1ac8, .offset = 0x90, .limit = 0x90 }, + { /* SE2 */ .base = 0x1ad0, .offset = 0x90, .limit = 0x90 }, + { /* SE3 */ .base = 0x1ad8, .offset = 0x90, .limit = 0x90 }, + { /* SE4 */ .base = 0x1ae0, .offset = 0x90, .limit = 0x90 }, + { /* ISP */ .base = 0x1ae8, .offset = 0x50, .limit = 0x50 }, + { /* VIC */ .base = 0x1af0, .offset = 0x30, .limit = 0x34 }, + { /* NVENC */ .base = 0x1af8, .offset = 0x30, .limit = 0x34 }, + { /* NVDEC */ .base = 0x1b00, .offset = 0x30, .limit = 0x34 }, + { /* NVJPG */ .base = 0x1b08, .offset = 0x30, .limit = 0x34 }, + { /* TSEC */ .base = 0x1b10, .offset = 0x30, .limit = 0x34 }, + { /* TSECB */ .base = 0x1b18, .offset = 0x30, .limit = 0x34 }, + { /* VI 0 */ .base = 0x1b80, .offset = 0x10000, .limit = 0x10000 }, + { /* VI 1 */ .base = 0x1b88, .offset = 0x20000, .limit = 0x20000 }, + { /* VI 2 */ .base = 0x1b90, .offset = 0x30000, .limit = 0x30000 }, + { /* VI 3 */ .base = 0x1b98, .offset = 0x40000, .limit = 0x40000 }, + { /* VI 4 */ .base = 0x1ba0, .offset = 0x50000, .limit = 0x50000 }, + { /* VI 5 */ .base = 0x1ba8, .offset = 0x60000, .limit = 0x60000 }, + { /* VI 6 */ .base = 0x1bb0, .offset = 0x70000, .limit = 0x70000 }, + { /* VI 7 */ .base = 0x1bb8, .offset = 0x80000, .limit = 0x80000 }, + { /* VI 8 */ .base = 0x1bc0, .offset = 0x90000, .limit = 0x90000 }, + { /* VI 9 */ .base = 0x1bc8, .offset = 0xa0000, .limit = 0xa0000 }, + { /* VI 10 */ .base = 0x1bd0, .offset = 0xb0000, .limit = 0xb0000 }, + { /* VI 11 */ .base = 0x1bd8, .offset = 0xc0000, .limit = 0xc0000 }, +}; + +static const struct host1x_info host1x06_info = { + .nb_channels = 63, + .nb_pts = 576, + .nb_mlocks = 24, + .nb_bases = 16, + .init = host1x06_init, + .sync_offset = 0x0, + .dma_mask = DMA_BIT_MASK(40), + .has_wide_gather = true, + .has_hypervisor = true, + .num_sid_entries = ARRAY_SIZE(tegra186_sid_table), + .sid_table = tegra186_sid_table, + .reserve_vblank_syncpts = false, + .skip_reset_assert = true, }; -static struct of_device_id host1x_of_match[] = { +static const struct host1x_sid_entry tegra194_sid_table[] = { + { /* SE1 */ .base = 0x1ac8, .offset = 0x90, .limit = 0x90 }, + { /* SE2 */ .base = 0x1ad0, .offset = 0x90, .limit = 0x90 }, + { /* SE3 */ .base = 0x1ad8, .offset = 0x90, .limit = 0x90 }, + { /* SE4 */ .base = 0x1ae0, .offset = 0x90, .limit = 0x90 }, + { /* ISP */ .base = 0x1ae8, .offset = 0x800, .limit = 0x800 }, + { /* VIC */ .base = 0x1af0, .offset = 0x30, .limit = 0x34 }, + { /* NVENC */ .base = 0x1af8, .offset = 0x30, .limit = 0x34 }, + { /* NVDEC */ .base = 0x1b00, .offset = 0x30, .limit = 0x34 }, + { /* NVJPG */ .base = 0x1b08, .offset = 0x30, .limit = 0x34 }, + { /* TSEC */ .base = 0x1b10, .offset = 0x30, .limit = 0x34 }, + { /* TSECB */ .base = 0x1b18, .offset = 0x30, .limit = 0x34 }, + { /* VI */ .base = 0x1b80, .offset = 0x800, .limit = 0x800 }, + { /* VI_THI */ .base = 0x1b88, .offset = 0x30, .limit = 0x34 }, + { /* ISP_THI */ .base = 0x1b90, .offset = 0x30, .limit = 0x34 }, + { /* PVA0_CLUSTER */ .base = 0x1b98, .offset = 0x0, .limit = 0x0 }, + { /* PVA0_CLUSTER */ .base = 0x1ba0, .offset = 0x0, .limit = 0x0 }, + { /* NVDLA0 */ .base = 0x1ba8, .offset = 0x30, .limit = 0x34 }, + { /* NVDLA1 */ .base = 0x1bb0, .offset = 0x30, .limit = 0x34 }, + { /* NVENC1 */ .base = 0x1bb8, .offset = 0x30, .limit = 0x34 }, + { /* NVDEC1 */ .base = 0x1bc0, .offset = 0x30, .limit = 0x34 }, +}; + +static const struct host1x_info host1x07_info = { + .nb_channels = 63, + .nb_pts = 704, + .nb_mlocks = 32, + .nb_bases = 0, + .init = host1x07_init, + .sync_offset = 0x0, + .dma_mask = DMA_BIT_MASK(40), + .has_wide_gather = true, + .has_hypervisor = true, + .num_sid_entries = ARRAY_SIZE(tegra194_sid_table), + .sid_table = tegra194_sid_table, + .reserve_vblank_syncpts = false, +}; + +/* + * Tegra234 has two stream ID protection tables, one for setting stream IDs + * through the channel path via SETSTREAMID, and one for setting them via + * MMIO. We program each engine's data stream ID in the channel path table + * and firmware stream ID in the MMIO path table. + */ +static const struct host1x_sid_entry tegra234_sid_table[] = { + { /* SE1 MMIO */ .base = 0x1650, .offset = 0x90, .limit = 0x90 }, + { /* SE1 ch */ .base = 0x1730, .offset = 0x90, .limit = 0x90 }, + { /* SE2 MMIO */ .base = 0x1658, .offset = 0x90, .limit = 0x90 }, + { /* SE2 ch */ .base = 0x1738, .offset = 0x90, .limit = 0x90 }, + { /* SE4 MMIO */ .base = 0x1660, .offset = 0x90, .limit = 0x90 }, + { /* SE4 ch */ .base = 0x1740, .offset = 0x90, .limit = 0x90 }, + { /* ISP MMIO */ .base = 0x1680, .offset = 0x800, .limit = 0x800 }, + { /* VIC MMIO */ .base = 0x1688, .offset = 0x34, .limit = 0x34 }, + { /* VIC ch */ .base = 0x17b8, .offset = 0x30, .limit = 0x30 }, + { /* NVENC MMIO */ .base = 0x1690, .offset = 0x34, .limit = 0x34 }, + { /* NVENC ch */ .base = 0x17c0, .offset = 0x30, .limit = 0x30 }, + { /* NVDEC MMIO */ .base = 0x1698, .offset = 0x34, .limit = 0x34 }, + { /* NVDEC ch */ .base = 0x17c8, .offset = 0x30, .limit = 0x30 }, + { /* NVJPG MMIO */ .base = 0x16a0, .offset = 0x34, .limit = 0x34 }, + { /* NVJPG ch */ .base = 0x17d0, .offset = 0x30, .limit = 0x30 }, + { /* TSEC MMIO */ .base = 0x16a8, .offset = 0x30, .limit = 0x34 }, + { /* NVJPG1 MMIO */ .base = 0x16b0, .offset = 0x34, .limit = 0x34 }, + { /* NVJPG1 ch */ .base = 0x17a8, .offset = 0x30, .limit = 0x30 }, + { /* VI MMIO */ .base = 0x16b8, .offset = 0x800, .limit = 0x800 }, + { /* VI_THI MMIO */ .base = 0x16c0, .offset = 0x30, .limit = 0x34 }, + { /* ISP_THI MMIO */ .base = 0x16c8, .offset = 0x30, .limit = 0x34 }, + { /* NVDLA MMIO */ .base = 0x16d8, .offset = 0x30, .limit = 0x34 }, + { /* NVDLA ch */ .base = 0x17e0, .offset = 0x30, .limit = 0x34 }, + { /* NVDLA1 MMIO */ .base = 0x16e0, .offset = 0x30, .limit = 0x34 }, + { /* NVDLA1 ch */ .base = 0x17e8, .offset = 0x30, .limit = 0x34 }, + { /* OFA MMIO */ .base = 0x16e8, .offset = 0x34, .limit = 0x34 }, + { /* OFA ch */ .base = 0x1768, .offset = 0x30, .limit = 0x30 }, + { /* VI2 MMIO */ .base = 0x16f0, .offset = 0x800, .limit = 0x800 }, + { /* VI2_THI MMIO */ .base = 0x16f8, .offset = 0x30, .limit = 0x34 }, +}; + +static const struct host1x_info host1x08_info = { + .nb_channels = 63, + .nb_pts = 1024, + .nb_mlocks = 24, + .nb_bases = 0, + .init = host1x08_init, + .sync_offset = 0x0, + .dma_mask = DMA_BIT_MASK(40), + .has_wide_gather = true, + .has_hypervisor = true, + .has_common = true, + .num_sid_entries = ARRAY_SIZE(tegra234_sid_table), + .sid_table = tegra234_sid_table, + .streamid_vm_table = { 0x1004, 128 }, + .classid_vm_table = { 0x1404, 25 }, + .mmio_vm_table = { 0x1504, 25 }, + .reserve_vblank_syncpts = false, +}; + +static const struct of_device_id host1x_of_match[] = { + { .compatible = "nvidia,tegra234-host1x", .data = &host1x08_info, }, + { .compatible = "nvidia,tegra194-host1x", .data = &host1x07_info, }, + { .compatible = "nvidia,tegra186-host1x", .data = &host1x06_info, }, + { .compatible = "nvidia,tegra210-host1x", .data = &host1x05_info, }, + { .compatible = "nvidia,tegra124-host1x", .data = &host1x04_info, }, + { .compatible = "nvidia,tegra114-host1x", .data = &host1x02_info, }, { .compatible = "nvidia,tegra30-host1x", .data = &host1x01_info, }, { .compatible = "nvidia,tegra20-host1x", .data = &host1x01_info, }, { }, }; MODULE_DEVICE_TABLE(of, host1x_of_match); -static int host1x_probe(struct platform_device *pdev) +static void host1x_setup_virtualization_tables(struct host1x *host) { - const struct of_device_id *id; - struct host1x *host; - struct resource *regs; - int syncpt_irq; + const struct host1x_info *info = host->info; + unsigned int i; + + if (!info->has_hypervisor) + return; + + for (i = 0; i < info->num_sid_entries; i++) { + const struct host1x_sid_entry *entry = &info->sid_table[i]; + + host1x_hypervisor_writel(host, entry->offset, entry->base); + host1x_hypervisor_writel(host, entry->limit, entry->base + 4); + } + + for (i = 0; i < info->streamid_vm_table.count; i++) { + /* Allow access to all stream IDs to all VMs. */ + host1x_hypervisor_writel(host, 0xff, info->streamid_vm_table.base + 4 * i); + } + + for (i = 0; i < info->classid_vm_table.count; i++) { + /* Allow access to all classes to all VMs. */ + host1x_hypervisor_writel(host, 0xff, info->classid_vm_table.base + 4 * i); + } + + for (i = 0; i < info->mmio_vm_table.count; i++) { + /* Use VM1 (that's us) as originator VMID for engine MMIO accesses. */ + host1x_hypervisor_writel(host, 0x1, info->mmio_vm_table.base + 4 * i); + } +} + +static bool host1x_wants_iommu(struct host1x *host1x) +{ + /* Our IOMMU usage policy doesn't currently play well with GART */ + if (of_machine_is_compatible("nvidia,tegra20")) + return false; + + /* + * If we support addressing a maximum of 32 bits of physical memory + * and if the host1x firewall is enabled, there's no need to enable + * IOMMU support. This can happen for example on Tegra20, Tegra30 + * and Tegra114. + * + * Tegra124 and later can address up to 34 bits of physical memory and + * many platforms come equipped with more than 2 GiB of system memory, + * which requires crossing the 4 GiB boundary. But there's a catch: on + * SoCs before Tegra186 (i.e. Tegra124 and Tegra210), the host1x can + * only address up to 32 bits of memory in GATHER opcodes, which means + * that command buffers need to either be in the first 2 GiB of system + * memory (which could quickly lead to memory exhaustion), or command + * buffers need to be treated differently from other buffers (which is + * not possible with the current ABI). + * + * A third option is to use the IOMMU in these cases to make sure all + * buffers will be mapped into a 32-bit IOVA space that host1x can + * address. This allows all of the system memory to be used and works + * within the limitations of the host1x on these SoCs. + * + * In summary, default to enable IOMMU on Tegra124 and later. For any + * of the earlier SoCs, only use the IOMMU for additional safety when + * the host1x firewall is disabled. + */ + if (host1x->info->dma_mask <= DMA_BIT_MASK(32)) { + if (IS_ENABLED(CONFIG_TEGRA_HOST1X_FIREWALL)) + return false; + } + + return true; +} + +/* + * Returns ERR_PTR on failure, NULL if the translation is IDENTITY, otherwise a + * valid paging domain. + */ +static struct iommu_domain *host1x_iommu_attach(struct host1x *host) +{ + struct iommu_domain *domain = iommu_get_domain_for_dev(host->dev); int err; - id = of_match_device(host1x_of_match, &pdev->dev); - if (!id) - return -EINVAL; +#if IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU) + if (host->dev->archdata.mapping) { + struct dma_iommu_mapping *mapping = + to_dma_iommu_mapping(host->dev); + arm_iommu_detach_device(host->dev); + arm_iommu_release_mapping(mapping); + + domain = iommu_get_domain_for_dev(host->dev); + } +#endif - regs = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (!regs) { - dev_err(&pdev->dev, "failed to get registers\n"); - return -ENXIO; + /* + * We may not always want to enable IOMMU support (for example if the + * host1x firewall is already enabled and we don't support addressing + * more than 32 bits of physical memory), so check for that first. + * + * Similarly, if host1x is already attached to an IOMMU (via the DMA + * API), don't try to attach again. + */ + if (domain && domain->type == IOMMU_DOMAIN_IDENTITY) + domain = NULL; + if (!host1x_wants_iommu(host) || domain) + return domain; + + host->group = iommu_group_get(host->dev); + if (host->group) { + struct iommu_domain_geometry *geometry; + dma_addr_t start, end; + unsigned long order; + + err = iova_cache_get(); + if (err < 0) + goto put_group; + + host->domain = iommu_paging_domain_alloc(host->dev); + if (IS_ERR(host->domain)) { + err = PTR_ERR(host->domain); + host->domain = NULL; + goto put_cache; + } + + err = iommu_attach_group(host->domain, host->group); + if (err) { + if (err == -ENODEV) + err = 0; + + goto free_domain; + } + + geometry = &host->domain->geometry; + start = geometry->aperture_start & host->info->dma_mask; + end = geometry->aperture_end & host->info->dma_mask; + + order = __ffs(host->domain->pgsize_bitmap); + init_iova_domain(&host->iova, 1UL << order, start >> order); + host->iova_end = end; + + domain = host->domain; } - syncpt_irq = platform_get_irq(pdev, 0); - if (syncpt_irq < 0) { - dev_err(&pdev->dev, "failed to get IRQ\n"); - return -ENXIO; + return domain; + +free_domain: + iommu_domain_free(host->domain); + host->domain = NULL; +put_cache: + iova_cache_put(); +put_group: + iommu_group_put(host->group); + host->group = NULL; + + return ERR_PTR(err); +} + +static int host1x_iommu_init(struct host1x *host) +{ + u64 mask = host->info->dma_mask; + struct iommu_domain *domain; + int err; + + domain = host1x_iommu_attach(host); + if (IS_ERR(domain)) { + err = PTR_ERR(domain); + dev_err(host->dev, "failed to attach to IOMMU: %d\n", err); + return err; + } + + /* + * If we're not behind an IOMMU make sure we don't get push buffers + * that are allocated outside of the range addressable by the GATHER + * opcode. + * + * Newer generations of Tegra (Tegra186 and later) support a wide + * variant of the GATHER opcode that allows addressing more bits. + */ + if (!domain && !host->info->has_wide_gather) + mask = DMA_BIT_MASK(32); + + err = dma_coerce_mask_and_coherent(host->dev, mask); + if (err < 0) { + dev_err(host->dev, "failed to set DMA mask: %d\n", err); + return err; + } + + return 0; +} + +static void host1x_iommu_exit(struct host1x *host) +{ + if (host->domain) { + put_iova_domain(&host->iova); + iommu_detach_group(host->domain, host->group); + + iommu_domain_free(host->domain); + host->domain = NULL; + + iova_cache_put(); + + iommu_group_put(host->group); + host->group = NULL; + } +} + +static int host1x_get_resets(struct host1x *host) +{ + int err; + + host->resets[0].id = "mc"; + host->resets[1].id = "host1x"; + host->nresets = ARRAY_SIZE(host->resets); + + err = devm_reset_control_bulk_get_optional_exclusive_released( + host->dev, host->nresets, host->resets); + if (err) { + dev_err(host->dev, "failed to get reset: %d\n", err); + return err; } + return 0; +} + +static int host1x_probe(struct platform_device *pdev) +{ + struct host1x *host; + int err, i; + host = devm_kzalloc(&pdev->dev, sizeof(*host), GFP_KERNEL); if (!host) return -ENOMEM; + host->info = of_device_get_match_data(&pdev->dev); + + if (host->info->has_hypervisor) { + host->regs = devm_platform_ioremap_resource_byname(pdev, "vm"); + if (IS_ERR(host->regs)) + return PTR_ERR(host->regs); + + host->hv_regs = devm_platform_ioremap_resource_byname(pdev, "hypervisor"); + if (IS_ERR(host->hv_regs)) + return PTR_ERR(host->hv_regs); + + if (host->info->has_common) { + host->common_regs = devm_platform_ioremap_resource_byname(pdev, "common"); + if (IS_ERR(host->common_regs)) + return PTR_ERR(host->common_regs); + } + } else { + host->regs = devm_platform_ioremap_resource(pdev, 0); + if (IS_ERR(host->regs)) + return PTR_ERR(host->regs); + } + + for (i = 0; i < ARRAY_SIZE(host->syncpt_irqs); i++) { + char irq_name[] = "syncptX"; + + sprintf(irq_name, "syncpt%d", i); + + err = platform_get_irq_byname_optional(pdev, irq_name); + if (err == -ENXIO) + break; + if (err < 0) + return err; + + host->syncpt_irqs[i] = err; + } + + host->num_syncpt_irqs = i; + + /* Device tree without irq names */ + if (i == 0) { + host->syncpt_irqs[0] = platform_get_irq(pdev, 0); + if (host->syncpt_irqs[0] < 0) + return host->syncpt_irqs[0]; + + host->num_syncpt_irqs = 1; + } + + mutex_init(&host->devices_lock); + INIT_LIST_HEAD(&host->devices); + INIT_LIST_HEAD(&host->list); host->dev = &pdev->dev; - host->info = id->data; /* set common host1x device data */ platform_set_drvdata(pdev, host); - host->regs = devm_ioremap_resource(&pdev->dev, regs); - if (IS_ERR(host->regs)) - return PTR_ERR(host->regs); + host->dev->dma_parms = &host->dma_parms; + dma_set_max_seg_size(host->dev, UINT_MAX); if (host->info->init) { err = host->info->init(host); @@ -131,115 +594,237 @@ static int host1x_probe(struct platform_device *pdev) } host->clk = devm_clk_get(&pdev->dev, NULL); - if (IS_ERR(host->clk)) { - dev_err(&pdev->dev, "failed to get clock\n"); - err = PTR_ERR(host->clk); + if (IS_ERR(host->clk)) + return dev_err_probe(&pdev->dev, PTR_ERR(host->clk), "failed to get clock\n"); + + err = host1x_get_resets(host); + if (err) return err; + + host1x_bo_cache_init(&host->cache); + + err = host1x_iommu_init(host); + if (err < 0) { + dev_err(&pdev->dev, "failed to setup IOMMU: %d\n", err); + goto destroy_cache; } - err = host1x_channel_list_init(host); + err = host1x_channel_list_init(&host->channel_list, + host->info->nb_channels); if (err) { dev_err(&pdev->dev, "failed to initialize channel list\n"); - return err; + goto iommu_exit; } - err = clk_prepare_enable(host->clk); - if (err < 0) { - dev_err(&pdev->dev, "failed to enable clock\n"); - return err; + err = host1x_memory_context_list_init(host); + if (err) { + dev_err(&pdev->dev, "failed to initialize context list\n"); + goto free_channels; } err = host1x_syncpt_init(host); if (err) { dev_err(&pdev->dev, "failed to initialize syncpts\n"); - return err; + goto free_contexts; } - err = host1x_intr_init(host, syncpt_irq); + mutex_init(&host->intr_mutex); + + pm_runtime_enable(&pdev->dev); + + err = devm_tegra_core_dev_init_opp_table_common(&pdev->dev); + if (err) + goto pm_disable; + + /* the driver's code isn't ready yet for the dynamic RPM */ + err = pm_runtime_resume_and_get(&pdev->dev); + if (err) + goto pm_disable; + + err = host1x_intr_init(host); if (err) { dev_err(&pdev->dev, "failed to initialize interrupts\n"); - goto fail_deinit_syncpt; + goto pm_put; } host1x_debug_init(host); - host1x_drm_alloc(pdev); + err = host1x_register(host); + if (err < 0) + goto deinit_debugfs; + + err = devm_of_platform_populate(&pdev->dev); + if (err < 0) + goto unregister; return 0; -fail_deinit_syncpt: +unregister: + host1x_unregister(host); +deinit_debugfs: + host1x_debug_deinit(host); + host1x_intr_deinit(host); +pm_put: + pm_runtime_put_sync_suspend(&pdev->dev); +pm_disable: + pm_runtime_disable(&pdev->dev); host1x_syncpt_deinit(host); +free_contexts: + host1x_memory_context_list_free(&host->context_list); +free_channels: + host1x_channel_list_free(&host->channel_list); +iommu_exit: + host1x_iommu_exit(host); +destroy_cache: + host1x_bo_cache_destroy(&host->cache); + return err; } -static int __exit host1x_remove(struct platform_device *pdev) +static void host1x_remove(struct platform_device *pdev) { struct host1x *host = platform_get_drvdata(pdev); + host1x_unregister(host); + host1x_debug_deinit(host); + + pm_runtime_force_suspend(&pdev->dev); + host1x_intr_deinit(host); host1x_syncpt_deinit(host); + host1x_memory_context_list_free(&host->context_list); + host1x_channel_list_free(&host->channel_list); + host1x_iommu_exit(host); + host1x_bo_cache_destroy(&host->cache); +} + +static int __maybe_unused host1x_runtime_suspend(struct device *dev) +{ + struct host1x *host = dev_get_drvdata(dev); + int err; + + host1x_channel_stop_all(host); + host1x_intr_stop(host); + host1x_syncpt_save(host); + + if (!host->info->skip_reset_assert) { + err = reset_control_bulk_assert(host->nresets, host->resets); + if (err) { + dev_err(dev, "failed to assert reset: %d\n", err); + goto resume_host1x; + } + + usleep_range(1000, 2000); + } + clk_disable_unprepare(host->clk); + reset_control_bulk_release(host->nresets, host->resets); return 0; + +resume_host1x: + host1x_setup_virtualization_tables(host); + host1x_syncpt_restore(host); + host1x_intr_start(host); + + return err; } +static int __maybe_unused host1x_runtime_resume(struct device *dev) +{ + struct host1x *host = dev_get_drvdata(dev); + int err; + + err = reset_control_bulk_acquire(host->nresets, host->resets); + if (err) { + dev_err(dev, "failed to acquire reset: %d\n", err); + return err; + } + + err = clk_prepare_enable(host->clk); + if (err) { + dev_err(dev, "failed to enable clock: %d\n", err); + goto release_reset; + } + + err = reset_control_bulk_deassert(host->nresets, host->resets); + if (err < 0) { + dev_err(dev, "failed to deassert reset: %d\n", err); + goto disable_clk; + } + + host1x_setup_virtualization_tables(host); + host1x_syncpt_restore(host); + host1x_intr_start(host); + + return 0; + +disable_clk: + clk_disable_unprepare(host->clk); +release_reset: + reset_control_bulk_release(host->nresets, host->resets); + + return err; +} + +static const struct dev_pm_ops host1x_pm_ops = { + SET_RUNTIME_PM_OPS(host1x_runtime_suspend, host1x_runtime_resume, + NULL) + SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, pm_runtime_force_resume) +}; + static struct platform_driver tegra_host1x_driver = { - .probe = host1x_probe, - .remove = __exit_p(host1x_remove), .driver = { - .owner = THIS_MODULE, .name = "tegra-host1x", .of_match_table = host1x_of_match, + .pm = &host1x_pm_ops, }, + .probe = host1x_probe, + .remove = host1x_remove, +}; + +static struct platform_driver * const drivers[] = { + &tegra_host1x_driver, + &tegra_mipi_driver, }; static int __init tegra_host1x_init(void) { int err; - err = platform_driver_register(&tegra_host1x_driver); + err = bus_register(&host1x_bus_type); if (err < 0) return err; -#ifdef CONFIG_DRM_TEGRA - err = platform_driver_register(&tegra_dc_driver); - if (err < 0) - goto unregister_host1x; - - err = platform_driver_register(&tegra_hdmi_driver); + err = platform_register_drivers(drivers, ARRAY_SIZE(drivers)); if (err < 0) - goto unregister_dc; + bus_unregister(&host1x_bus_type); - err = platform_driver_register(&tegra_gr2d_driver); - if (err < 0) - goto unregister_hdmi; -#endif - - return 0; - -#ifdef CONFIG_DRM_TEGRA -unregister_hdmi: - platform_driver_unregister(&tegra_hdmi_driver); -unregister_dc: - platform_driver_unregister(&tegra_dc_driver); -unregister_host1x: - platform_driver_unregister(&tegra_host1x_driver); return err; -#endif } module_init(tegra_host1x_init); static void __exit tegra_host1x_exit(void) { -#ifdef CONFIG_DRM_TEGRA - platform_driver_unregister(&tegra_gr2d_driver); - platform_driver_unregister(&tegra_hdmi_driver); - platform_driver_unregister(&tegra_dc_driver); -#endif - platform_driver_unregister(&tegra_host1x_driver); + platform_unregister_drivers(drivers, ARRAY_SIZE(drivers)); + bus_unregister(&host1x_bus_type); } module_exit(tegra_host1x_exit); +/** + * host1x_get_dma_mask() - query the supported DMA mask for host1x + * @host1x: host1x instance + * + * Note that this returns the supported DMA mask for host1x, which can be + * different from the applicable DMA mask under certain circumstances. + */ +u64 host1x_get_dma_mask(struct host1x *host1x) +{ + return host1x->info->dma_mask; +} +EXPORT_SYMBOL(host1x_get_dma_mask); + +MODULE_SOFTDEP("post: tegra-drm"); MODULE_AUTHOR("Thierry Reding <thierry.reding@avionic-design.de>"); MODULE_AUTHOR("Terje Bergstrom <tbergstrom@nvidia.com>"); MODULE_DESCRIPTION("Host1x driver for Tegra products"); diff --git a/drivers/gpu/host1x/dev.h b/drivers/gpu/host1x/dev.h index 790ddf114e58..ef44618ed88a 100644 --- a/drivers/gpu/host1x/dev.h +++ b/drivers/gpu/host1x/dev.h @@ -1,32 +1,27 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ /* - * Copyright (c) 2012-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (c) 2012-2015, NVIDIA Corporation. */ #ifndef HOST1X_DEV_H #define HOST1X_DEV_H -#include <linux/platform_device.h> #include <linux/device.h> +#include <linux/iommu.h> +#include <linux/iova.h> +#include <linux/irqreturn.h> +#include <linux/platform_device.h> +#include <linux/reset.h> +#include "cdma.h" #include "channel.h" -#include "syncpt.h" +#include "context.h" #include "intr.h" -#include "cdma.h" #include "job.h" +#include "syncpt.h" struct host1x_syncpt; +struct host1x_syncpt_base; struct host1x_channel; struct host1x_cdma; struct host1x_job; @@ -44,7 +39,7 @@ struct host1x_cdma_ops { void (*start)(struct host1x_cdma *cdma); void (*stop)(struct host1x_cdma *cdma); void (*flush)(struct host1x_cdma *cdma); - int (*timeout_init)(struct host1x_cdma *cdma, u32 syncpt_id); + int (*timeout_init)(struct host1x_cdma *cdma); void (*timeout_destroy)(struct host1x_cdma *cdma); void (*freeze)(struct host1x_cdma *cdma); void (*resume)(struct host1x_cdma *cdma, u32 getptr); @@ -74,40 +69,84 @@ struct host1x_syncpt_ops { void (*load_wait_base)(struct host1x_syncpt *syncpt); u32 (*load)(struct host1x_syncpt *syncpt); int (*cpu_incr)(struct host1x_syncpt *syncpt); - int (*patch_wait)(struct host1x_syncpt *syncpt, void *patch_addr); + void (*assign_to_channel)(struct host1x_syncpt *syncpt, + struct host1x_channel *channel); + void (*enable_protection)(struct host1x *host); }; struct host1x_intr_ops { - int (*init_host_sync)(struct host1x *host, u32 cpm, - void (*syncpt_thresh_work)(struct work_struct *work)); + int (*init_host_sync)(struct host1x *host, u32 cpm); void (*set_syncpt_threshold)( - struct host1x *host, u32 id, u32 thresh); - void (*enable_syncpt_intr)(struct host1x *host, u32 id); - void (*disable_syncpt_intr)(struct host1x *host, u32 id); + struct host1x *host, unsigned int id, u32 thresh); + void (*enable_syncpt_intr)(struct host1x *host, unsigned int id); + void (*disable_syncpt_intr)(struct host1x *host, unsigned int id); void (*disable_all_syncpt_intrs)(struct host1x *host); int (*free_syncpt_irq)(struct host1x *host); + irqreturn_t (*isr)(int irq, void *dev_id); +}; + +struct host1x_sid_entry { + unsigned int base; + unsigned int offset; + unsigned int limit; +}; + +struct host1x_table_desc { + unsigned int base; + unsigned int count; }; struct host1x_info { - int nb_channels; /* host1x: num channels supported */ - int nb_pts; /* host1x: num syncpoints supported */ - int nb_bases; /* host1x: num syncpoints supported */ - int nb_mlocks; /* host1x: number of mlocks */ - int (*init)(struct host1x *); /* initialize per SoC ops */ - int sync_offset; + unsigned int nb_channels; /* host1x: number of channels supported */ + unsigned int nb_pts; /* host1x: number of syncpoints supported */ + unsigned int nb_bases; /* host1x: number of syncpoint bases supported */ + unsigned int nb_mlocks; /* host1x: number of mlocks supported */ + int (*init)(struct host1x *host1x); /* initialize per SoC ops */ + unsigned int sync_offset; /* offset of syncpoint registers */ + u64 dma_mask; /* mask of addressable memory */ + bool has_wide_gather; /* supports GATHER_W opcode */ + bool has_hypervisor; /* has hypervisor registers */ + bool has_common; /* has common registers separate from hypervisor */ + unsigned int num_sid_entries; + const struct host1x_sid_entry *sid_table; + struct host1x_table_desc streamid_vm_table; + struct host1x_table_desc classid_vm_table; + struct host1x_table_desc mmio_vm_table; + /* + * On T20-T148, the boot chain may setup DC to increment syncpoints + * 26/27 on VBLANK. As such we cannot use these syncpoints until + * the display driver disables VBLANK increments. + */ + bool reserve_vblank_syncpts; + /* + * On Tegra186, secure world applications may require access to + * host1x during suspend/resume. To allow this, we need to leave + * host1x not in reset. + */ + bool skip_reset_assert; }; struct host1x { const struct host1x_info *info; void __iomem *regs; + void __iomem *hv_regs; /* hypervisor region */ + void __iomem *common_regs; + int syncpt_irqs[8]; + int num_syncpt_irqs; struct host1x_syncpt *syncpt; + struct host1x_syncpt_base *bases; struct device *dev; struct clk *clk; + struct reset_control_bulk_data resets[2]; + unsigned int nresets; + + struct iommu_group *group; + struct iommu_domain *domain; + struct iova_domain iova; + dma_addr_t iova_end; struct mutex intr_mutex; - struct workqueue_struct *intr_wq; - int intr_syncpt_irq; const struct host1x_syncpt_ops *syncpt_op; const struct host1x_intr_ops *intr_op; @@ -118,19 +157,32 @@ struct host1x { struct host1x_syncpt *nop_sp; - struct mutex chlist_mutex; - struct host1x_channel chlist; - unsigned long allocated_channels; - unsigned int num_allocated_channels; + struct mutex syncpt_mutex; + + struct host1x_channel_list channel_list; + struct host1x_memory_context_list context_list; struct dentry *debugfs; - void *drm_data; + struct mutex devices_lock; + struct list_head devices; + + struct list_head list; + + struct device_dma_parameters dma_parms; + + struct host1x_bo_cache cache; }; -void host1x_sync_writel(struct host1x *host1x, u32 r, u32 v); +void host1x_common_writel(struct host1x *host1x, u32 v, u32 r); +void host1x_hypervisor_writel(struct host1x *host1x, u32 v, u32 r); +u32 host1x_hypervisor_readl(struct host1x *host1x, u32 r); +void host1x_sync_writel(struct host1x *host1x, u32 v, u32 r); u32 host1x_sync_readl(struct host1x *host1x, u32 r); -void host1x_ch_writel(struct host1x_channel *ch, u32 r, u32 v); +#ifdef CONFIG_64BIT +u64 host1x_sync_readq(struct host1x *host1x, u32 r); +#endif +void host1x_ch_writel(struct host1x_channel *ch, u32 v, u32 r); u32 host1x_ch_readl(struct host1x_channel *ch, u32 r); static inline void host1x_hw_syncpt_restore(struct host1x *host, @@ -163,33 +215,38 @@ static inline int host1x_hw_syncpt_cpu_incr(struct host1x *host, return host->syncpt_op->cpu_incr(sp); } -static inline int host1x_hw_syncpt_patch_wait(struct host1x *host, - struct host1x_syncpt *sp, - void *patch_addr) +static inline void host1x_hw_syncpt_assign_to_channel( + struct host1x *host, struct host1x_syncpt *sp, + struct host1x_channel *ch) +{ + return host->syncpt_op->assign_to_channel(sp, ch); +} + +static inline void host1x_hw_syncpt_enable_protection(struct host1x *host) { - return host->syncpt_op->patch_wait(sp, patch_addr); + return host->syncpt_op->enable_protection(host); } -static inline int host1x_hw_intr_init_host_sync(struct host1x *host, u32 cpm, - void (*syncpt_thresh_work)(struct work_struct *)) +static inline int host1x_hw_intr_init_host_sync(struct host1x *host, u32 cpm) { - return host->intr_op->init_host_sync(host, cpm, syncpt_thresh_work); + return host->intr_op->init_host_sync(host, cpm); } static inline void host1x_hw_intr_set_syncpt_threshold(struct host1x *host, - u32 id, u32 thresh) + unsigned int id, + u32 thresh) { host->intr_op->set_syncpt_threshold(host, id, thresh); } static inline void host1x_hw_intr_enable_syncpt_intr(struct host1x *host, - u32 id) + unsigned int id) { host->intr_op->enable_syncpt_intr(host, id); } static inline void host1x_hw_intr_disable_syncpt_intr(struct host1x *host, - u32 id) + unsigned int id) { host->intr_op->disable_syncpt_intr(host, id); } @@ -206,9 +263,9 @@ static inline int host1x_hw_intr_free_syncpt_irq(struct host1x *host) static inline int host1x_hw_channel_init(struct host1x *host, struct host1x_channel *channel, - int chid) + unsigned int id) { - return host->channel_op->init(channel, host, chid); + return host->channel_op->init(channel, host, id); } static inline int host1x_hw_channel_submit(struct host1x *host, @@ -236,10 +293,9 @@ static inline void host1x_hw_cdma_flush(struct host1x *host, } static inline int host1x_hw_cdma_timeout_init(struct host1x *host, - struct host1x_cdma *cdma, - u32 syncpt_id) + struct host1x_cdma *cdma) { - return host->cdma_op->timeout_init(cdma, syncpt_id); + return host->cdma_op->timeout_init(cdma); } static inline void host1x_hw_cdma_timeout_destroy(struct host1x *host, @@ -301,8 +357,6 @@ static inline void host1x_hw_show_mlocks(struct host1x *host, struct output *o) host->debug_op->show_mlocks(host, o); } -extern struct platform_driver tegra_hdmi_driver; -extern struct platform_driver tegra_dc_driver; -extern struct platform_driver tegra_gr2d_driver; +extern struct platform_driver tegra_mipi_driver; #endif diff --git a/drivers/gpu/host1x/drm/Kconfig b/drivers/gpu/host1x/drm/Kconfig deleted file mode 100644 index 69853a4de40a..000000000000 --- a/drivers/gpu/host1x/drm/Kconfig +++ /dev/null @@ -1,29 +0,0 @@ -config DRM_TEGRA - bool "NVIDIA Tegra DRM" - depends on DRM - select DRM_KMS_HELPER - select FB_SYS_FILLRECT - select FB_SYS_COPYAREA - select FB_SYS_IMAGEBLIT - help - Choose this option if you have an NVIDIA Tegra SoC. - - To compile this driver as a module, choose M here: the module - will be called tegra-drm. - -if DRM_TEGRA - -config DRM_TEGRA_STAGING - bool "Enable HOST1X interface" - depends on STAGING - help - Say yes if HOST1X should be available for userspace DRM users. - - If unsure, choose N. - -config DRM_TEGRA_DEBUG - bool "NVIDIA Tegra DRM debug support" - help - Say yes here to enable debugging support. - -endif diff --git a/drivers/gpu/host1x/drm/dc.c b/drivers/gpu/host1x/drm/dc.c deleted file mode 100644 index 5360e5a57ecc..000000000000 --- a/drivers/gpu/host1x/drm/dc.c +++ /dev/null @@ -1,1200 +0,0 @@ -/* - * Copyright (C) 2012 Avionic Design GmbH - * Copyright (C) 2012 NVIDIA CORPORATION. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#include <linux/clk.h> -#include <linux/debugfs.h> -#include <linux/module.h> -#include <linux/of.h> -#include <linux/platform_device.h> -#include <linux/clk/tegra.h> - -#include "host1x_client.h" -#include "dc.h" -#include "drm.h" -#include "gem.h" - -struct tegra_plane { - struct drm_plane base; - unsigned int index; -}; - -static inline struct tegra_plane *to_tegra_plane(struct drm_plane *plane) -{ - return container_of(plane, struct tegra_plane, base); -} - -static int tegra_plane_update(struct drm_plane *plane, struct drm_crtc *crtc, - struct drm_framebuffer *fb, int crtc_x, - int crtc_y, unsigned int crtc_w, - unsigned int crtc_h, uint32_t src_x, - uint32_t src_y, uint32_t src_w, uint32_t src_h) -{ - struct tegra_plane *p = to_tegra_plane(plane); - struct tegra_dc *dc = to_tegra_dc(crtc); - struct tegra_dc_window window; - unsigned int i; - - memset(&window, 0, sizeof(window)); - window.src.x = src_x >> 16; - window.src.y = src_y >> 16; - window.src.w = src_w >> 16; - window.src.h = src_h >> 16; - window.dst.x = crtc_x; - window.dst.y = crtc_y; - window.dst.w = crtc_w; - window.dst.h = crtc_h; - window.format = tegra_dc_format(fb->pixel_format); - window.bits_per_pixel = fb->bits_per_pixel; - - for (i = 0; i < drm_format_num_planes(fb->pixel_format); i++) { - struct tegra_bo *bo = tegra_fb_get_plane(fb, i); - - window.base[i] = bo->paddr + fb->offsets[i]; - - /* - * Tegra doesn't support different strides for U and V planes - * so we display a warning if the user tries to display a - * framebuffer with such a configuration. - */ - if (i >= 2) { - if (fb->pitches[i] != window.stride[1]) - DRM_ERROR("unsupported UV-plane configuration\n"); - } else { - window.stride[i] = fb->pitches[i]; - } - } - - return tegra_dc_setup_window(dc, p->index, &window); -} - -static int tegra_plane_disable(struct drm_plane *plane) -{ - struct tegra_dc *dc = to_tegra_dc(plane->crtc); - struct tegra_plane *p = to_tegra_plane(plane); - unsigned long value; - - if (!plane->crtc) - return 0; - - value = WINDOW_A_SELECT << p->index; - tegra_dc_writel(dc, value, DC_CMD_DISPLAY_WINDOW_HEADER); - - value = tegra_dc_readl(dc, DC_WIN_WIN_OPTIONS); - value &= ~WIN_ENABLE; - tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS); - - tegra_dc_writel(dc, WIN_A_UPDATE << p->index, DC_CMD_STATE_CONTROL); - tegra_dc_writel(dc, WIN_A_ACT_REQ << p->index, DC_CMD_STATE_CONTROL); - - return 0; -} - -static void tegra_plane_destroy(struct drm_plane *plane) -{ - tegra_plane_disable(plane); - drm_plane_cleanup(plane); -} - -static const struct drm_plane_funcs tegra_plane_funcs = { - .update_plane = tegra_plane_update, - .disable_plane = tegra_plane_disable, - .destroy = tegra_plane_destroy, -}; - -static const uint32_t plane_formats[] = { - DRM_FORMAT_XBGR8888, - DRM_FORMAT_XRGB8888, - DRM_FORMAT_RGB565, - DRM_FORMAT_UYVY, - DRM_FORMAT_YUV420, - DRM_FORMAT_YUV422, -}; - -static int tegra_dc_add_planes(struct drm_device *drm, struct tegra_dc *dc) -{ - unsigned int i; - int err = 0; - - for (i = 0; i < 2; i++) { - struct tegra_plane *plane; - - plane = devm_kzalloc(drm->dev, sizeof(*plane), GFP_KERNEL); - if (!plane) - return -ENOMEM; - - plane->index = 1 + i; - - err = drm_plane_init(drm, &plane->base, 1 << dc->pipe, - &tegra_plane_funcs, plane_formats, - ARRAY_SIZE(plane_formats), false); - if (err < 0) - return err; - } - - return 0; -} - -static int tegra_dc_set_base(struct tegra_dc *dc, int x, int y, - struct drm_framebuffer *fb) -{ - unsigned int format = tegra_dc_format(fb->pixel_format); - struct tegra_bo *bo = tegra_fb_get_plane(fb, 0); - unsigned long value; - - tegra_dc_writel(dc, WINDOW_A_SELECT, DC_CMD_DISPLAY_WINDOW_HEADER); - - value = fb->offsets[0] + y * fb->pitches[0] + - x * fb->bits_per_pixel / 8; - - tegra_dc_writel(dc, bo->paddr + value, DC_WINBUF_START_ADDR); - tegra_dc_writel(dc, fb->pitches[0], DC_WIN_LINE_STRIDE); - tegra_dc_writel(dc, format, DC_WIN_COLOR_DEPTH); - - value = GENERAL_UPDATE | WIN_A_UPDATE; - tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL); - - value = GENERAL_ACT_REQ | WIN_A_ACT_REQ; - tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL); - - return 0; -} - -void tegra_dc_enable_vblank(struct tegra_dc *dc) -{ - unsigned long value, flags; - - spin_lock_irqsave(&dc->lock, flags); - - value = tegra_dc_readl(dc, DC_CMD_INT_MASK); - value |= VBLANK_INT; - tegra_dc_writel(dc, value, DC_CMD_INT_MASK); - - spin_unlock_irqrestore(&dc->lock, flags); -} - -void tegra_dc_disable_vblank(struct tegra_dc *dc) -{ - unsigned long value, flags; - - spin_lock_irqsave(&dc->lock, flags); - - value = tegra_dc_readl(dc, DC_CMD_INT_MASK); - value &= ~VBLANK_INT; - tegra_dc_writel(dc, value, DC_CMD_INT_MASK); - - spin_unlock_irqrestore(&dc->lock, flags); -} - -static void tegra_dc_finish_page_flip(struct tegra_dc *dc) -{ - struct drm_device *drm = dc->base.dev; - struct drm_crtc *crtc = &dc->base; - unsigned long flags, base; - struct tegra_bo *bo; - - if (!dc->event) - return; - - bo = tegra_fb_get_plane(crtc->fb, 0); - - /* check if new start address has been latched */ - tegra_dc_writel(dc, READ_MUX, DC_CMD_STATE_ACCESS); - base = tegra_dc_readl(dc, DC_WINBUF_START_ADDR); - tegra_dc_writel(dc, 0, DC_CMD_STATE_ACCESS); - - if (base == bo->paddr + crtc->fb->offsets[0]) { - spin_lock_irqsave(&drm->event_lock, flags); - drm_send_vblank_event(drm, dc->pipe, dc->event); - drm_vblank_put(drm, dc->pipe); - dc->event = NULL; - spin_unlock_irqrestore(&drm->event_lock, flags); - } -} - -void tegra_dc_cancel_page_flip(struct drm_crtc *crtc, struct drm_file *file) -{ - struct tegra_dc *dc = to_tegra_dc(crtc); - struct drm_device *drm = crtc->dev; - unsigned long flags; - - spin_lock_irqsave(&drm->event_lock, flags); - - if (dc->event && dc->event->base.file_priv == file) { - dc->event->base.destroy(&dc->event->base); - drm_vblank_put(drm, dc->pipe); - dc->event = NULL; - } - - spin_unlock_irqrestore(&drm->event_lock, flags); -} - -static int tegra_dc_page_flip(struct drm_crtc *crtc, struct drm_framebuffer *fb, - struct drm_pending_vblank_event *event) -{ - struct tegra_dc *dc = to_tegra_dc(crtc); - struct drm_device *drm = crtc->dev; - - if (dc->event) - return -EBUSY; - - if (event) { - event->pipe = dc->pipe; - dc->event = event; - drm_vblank_get(drm, dc->pipe); - } - - tegra_dc_set_base(dc, 0, 0, fb); - crtc->fb = fb; - - return 0; -} - -static const struct drm_crtc_funcs tegra_crtc_funcs = { - .page_flip = tegra_dc_page_flip, - .set_config = drm_crtc_helper_set_config, - .destroy = drm_crtc_cleanup, -}; - -static void tegra_crtc_disable(struct drm_crtc *crtc) -{ - struct drm_device *drm = crtc->dev; - struct drm_plane *plane; - - list_for_each_entry(plane, &drm->mode_config.plane_list, head) { - if (plane->crtc == crtc) { - tegra_plane_disable(plane); - plane->crtc = NULL; - - if (plane->fb) { - drm_framebuffer_unreference(plane->fb); - plane->fb = NULL; - } - } - } -} - -static bool tegra_crtc_mode_fixup(struct drm_crtc *crtc, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted) -{ - return true; -} - -static inline u32 compute_dda_inc(unsigned int in, unsigned int out, bool v, - unsigned int bpp) -{ - fixed20_12 outf = dfixed_init(out); - fixed20_12 inf = dfixed_init(in); - u32 dda_inc; - int max; - - if (v) - max = 15; - else { - switch (bpp) { - case 2: - max = 8; - break; - - default: - WARN_ON_ONCE(1); - /* fallthrough */ - case 4: - max = 4; - break; - } - } - - outf.full = max_t(u32, outf.full - dfixed_const(1), dfixed_const(1)); - inf.full -= dfixed_const(1); - - dda_inc = dfixed_div(inf, outf); - dda_inc = min_t(u32, dda_inc, dfixed_const(max)); - - return dda_inc; -} - -static inline u32 compute_initial_dda(unsigned int in) -{ - fixed20_12 inf = dfixed_init(in); - return dfixed_frac(inf); -} - -static int tegra_dc_set_timings(struct tegra_dc *dc, - struct drm_display_mode *mode) -{ - /* TODO: For HDMI compliance, h & v ref_to_sync should be set to 1 */ - unsigned int h_ref_to_sync = 0; - unsigned int v_ref_to_sync = 0; - unsigned long value; - - tegra_dc_writel(dc, 0x0, DC_DISP_DISP_TIMING_OPTIONS); - - value = (v_ref_to_sync << 16) | h_ref_to_sync; - tegra_dc_writel(dc, value, DC_DISP_REF_TO_SYNC); - - value = ((mode->vsync_end - mode->vsync_start) << 16) | - ((mode->hsync_end - mode->hsync_start) << 0); - tegra_dc_writel(dc, value, DC_DISP_SYNC_WIDTH); - - value = ((mode->vtotal - mode->vsync_end) << 16) | - ((mode->htotal - mode->hsync_end) << 0); - tegra_dc_writel(dc, value, DC_DISP_BACK_PORCH); - - value = ((mode->vsync_start - mode->vdisplay) << 16) | - ((mode->hsync_start - mode->hdisplay) << 0); - tegra_dc_writel(dc, value, DC_DISP_FRONT_PORCH); - - value = (mode->vdisplay << 16) | mode->hdisplay; - tegra_dc_writel(dc, value, DC_DISP_ACTIVE); - - return 0; -} - -static int tegra_crtc_setup_clk(struct drm_crtc *crtc, - struct drm_display_mode *mode, - unsigned long *div) -{ - unsigned long pclk = mode->clock * 1000, rate; - struct tegra_dc *dc = to_tegra_dc(crtc); - struct tegra_output *output = NULL; - struct drm_encoder *encoder; - long err; - - list_for_each_entry(encoder, &crtc->dev->mode_config.encoder_list, head) - if (encoder->crtc == crtc) { - output = encoder_to_output(encoder); - break; - } - - if (!output) - return -ENODEV; - - /* - * This assumes that the display controller will divide its parent - * clock by 2 to generate the pixel clock. - */ - err = tegra_output_setup_clock(output, dc->clk, pclk * 2); - if (err < 0) { - dev_err(dc->dev, "failed to setup clock: %ld\n", err); - return err; - } - - rate = clk_get_rate(dc->clk); - *div = (rate * 2 / pclk) - 2; - - DRM_DEBUG_KMS("rate: %lu, div: %lu\n", rate, *div); - - return 0; -} - -static bool tegra_dc_format_is_yuv(unsigned int format, bool *planar) -{ - switch (format) { - case WIN_COLOR_DEPTH_YCbCr422: - case WIN_COLOR_DEPTH_YUV422: - if (planar) - *planar = false; - - return true; - - case WIN_COLOR_DEPTH_YCbCr420P: - case WIN_COLOR_DEPTH_YUV420P: - case WIN_COLOR_DEPTH_YCbCr422P: - case WIN_COLOR_DEPTH_YUV422P: - case WIN_COLOR_DEPTH_YCbCr422R: - case WIN_COLOR_DEPTH_YUV422R: - case WIN_COLOR_DEPTH_YCbCr422RA: - case WIN_COLOR_DEPTH_YUV422RA: - if (planar) - *planar = true; - - return true; - } - - return false; -} - -int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index, - const struct tegra_dc_window *window) -{ - unsigned h_offset, v_offset, h_size, v_size, h_dda, v_dda, bpp; - unsigned long value; - bool yuv, planar; - - /* - * For YUV planar modes, the number of bytes per pixel takes into - * account only the luma component and therefore is 1. - */ - yuv = tegra_dc_format_is_yuv(window->format, &planar); - if (!yuv) - bpp = window->bits_per_pixel / 8; - else - bpp = planar ? 1 : 2; - - value = WINDOW_A_SELECT << index; - tegra_dc_writel(dc, value, DC_CMD_DISPLAY_WINDOW_HEADER); - - tegra_dc_writel(dc, window->format, DC_WIN_COLOR_DEPTH); - tegra_dc_writel(dc, 0, DC_WIN_BYTE_SWAP); - - value = V_POSITION(window->dst.y) | H_POSITION(window->dst.x); - tegra_dc_writel(dc, value, DC_WIN_POSITION); - - value = V_SIZE(window->dst.h) | H_SIZE(window->dst.w); - tegra_dc_writel(dc, value, DC_WIN_SIZE); - - h_offset = window->src.x * bpp; - v_offset = window->src.y; - h_size = window->src.w * bpp; - v_size = window->src.h; - - value = V_PRESCALED_SIZE(v_size) | H_PRESCALED_SIZE(h_size); - tegra_dc_writel(dc, value, DC_WIN_PRESCALED_SIZE); - - /* - * For DDA computations the number of bytes per pixel for YUV planar - * modes needs to take into account all Y, U and V components. - */ - if (yuv && planar) - bpp = 2; - - h_dda = compute_dda_inc(window->src.w, window->dst.w, false, bpp); - v_dda = compute_dda_inc(window->src.h, window->dst.h, true, bpp); - - value = V_DDA_INC(v_dda) | H_DDA_INC(h_dda); - tegra_dc_writel(dc, value, DC_WIN_DDA_INC); - - h_dda = compute_initial_dda(window->src.x); - v_dda = compute_initial_dda(window->src.y); - - tegra_dc_writel(dc, h_dda, DC_WIN_H_INITIAL_DDA); - tegra_dc_writel(dc, v_dda, DC_WIN_V_INITIAL_DDA); - - tegra_dc_writel(dc, 0, DC_WIN_UV_BUF_STRIDE); - tegra_dc_writel(dc, 0, DC_WIN_BUF_STRIDE); - - tegra_dc_writel(dc, window->base[0], DC_WINBUF_START_ADDR); - - if (yuv && planar) { - tegra_dc_writel(dc, window->base[1], DC_WINBUF_START_ADDR_U); - tegra_dc_writel(dc, window->base[2], DC_WINBUF_START_ADDR_V); - value = window->stride[1] << 16 | window->stride[0]; - tegra_dc_writel(dc, value, DC_WIN_LINE_STRIDE); - } else { - tegra_dc_writel(dc, window->stride[0], DC_WIN_LINE_STRIDE); - } - - tegra_dc_writel(dc, h_offset, DC_WINBUF_ADDR_H_OFFSET); - tegra_dc_writel(dc, v_offset, DC_WINBUF_ADDR_V_OFFSET); - - value = WIN_ENABLE; - - if (yuv) { - /* setup default colorspace conversion coefficients */ - tegra_dc_writel(dc, 0x00f0, DC_WIN_CSC_YOF); - tegra_dc_writel(dc, 0x012a, DC_WIN_CSC_KYRGB); - tegra_dc_writel(dc, 0x0000, DC_WIN_CSC_KUR); - tegra_dc_writel(dc, 0x0198, DC_WIN_CSC_KVR); - tegra_dc_writel(dc, 0x039b, DC_WIN_CSC_KUG); - tegra_dc_writel(dc, 0x032f, DC_WIN_CSC_KVG); - tegra_dc_writel(dc, 0x0204, DC_WIN_CSC_KUB); - tegra_dc_writel(dc, 0x0000, DC_WIN_CSC_KVB); - - value |= CSC_ENABLE; - } else if (window->bits_per_pixel < 24) { - value |= COLOR_EXPAND; - } - - tegra_dc_writel(dc, value, DC_WIN_WIN_OPTIONS); - - /* - * Disable blending and assume Window A is the bottom-most window, - * Window C is the top-most window and Window B is in the middle. - */ - tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_NOKEY); - tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_1WIN); - - switch (index) { - case 0: - tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_X); - tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_Y); - tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_3WIN_XY); - break; - - case 1: - tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_X); - tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_2WIN_Y); - tegra_dc_writel(dc, 0x000000, DC_WIN_BLEND_3WIN_XY); - break; - - case 2: - tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_X); - tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_2WIN_Y); - tegra_dc_writel(dc, 0xffff00, DC_WIN_BLEND_3WIN_XY); - break; - } - - tegra_dc_writel(dc, WIN_A_UPDATE << index, DC_CMD_STATE_CONTROL); - tegra_dc_writel(dc, WIN_A_ACT_REQ << index, DC_CMD_STATE_CONTROL); - - return 0; -} - -unsigned int tegra_dc_format(uint32_t format) -{ - switch (format) { - case DRM_FORMAT_XBGR8888: - return WIN_COLOR_DEPTH_R8G8B8A8; - - case DRM_FORMAT_XRGB8888: - return WIN_COLOR_DEPTH_B8G8R8A8; - - case DRM_FORMAT_RGB565: - return WIN_COLOR_DEPTH_B5G6R5; - - case DRM_FORMAT_UYVY: - return WIN_COLOR_DEPTH_YCbCr422; - - case DRM_FORMAT_YUV420: - return WIN_COLOR_DEPTH_YCbCr420P; - - case DRM_FORMAT_YUV422: - return WIN_COLOR_DEPTH_YCbCr422P; - - default: - break; - } - - WARN(1, "unsupported pixel format %u, using default\n", format); - return WIN_COLOR_DEPTH_B8G8R8A8; -} - -static int tegra_crtc_mode_set(struct drm_crtc *crtc, - struct drm_display_mode *mode, - struct drm_display_mode *adjusted, - int x, int y, struct drm_framebuffer *old_fb) -{ - struct tegra_bo *bo = tegra_fb_get_plane(crtc->fb, 0); - struct tegra_dc *dc = to_tegra_dc(crtc); - struct tegra_dc_window window; - unsigned long div, value; - int err; - - drm_vblank_pre_modeset(crtc->dev, dc->pipe); - - err = tegra_crtc_setup_clk(crtc, mode, &div); - if (err) { - dev_err(dc->dev, "failed to setup clock for CRTC: %d\n", err); - return err; - } - - /* program display mode */ - tegra_dc_set_timings(dc, mode); - - value = DE_SELECT_ACTIVE | DE_CONTROL_NORMAL; - tegra_dc_writel(dc, value, DC_DISP_DATA_ENABLE_OPTIONS); - - value = tegra_dc_readl(dc, DC_COM_PIN_OUTPUT_POLARITY(1)); - value &= ~LVS_OUTPUT_POLARITY_LOW; - value &= ~LHS_OUTPUT_POLARITY_LOW; - tegra_dc_writel(dc, value, DC_COM_PIN_OUTPUT_POLARITY(1)); - - value = DISP_DATA_FORMAT_DF1P1C | DISP_ALIGNMENT_MSB | - DISP_ORDER_RED_BLUE; - tegra_dc_writel(dc, value, DC_DISP_DISP_INTERFACE_CONTROL); - - tegra_dc_writel(dc, 0x00010001, DC_DISP_SHIFT_CLOCK_OPTIONS); - - value = SHIFT_CLK_DIVIDER(div) | PIXEL_CLK_DIVIDER_PCD1; - tegra_dc_writel(dc, value, DC_DISP_DISP_CLOCK_CONTROL); - - /* setup window parameters */ - memset(&window, 0, sizeof(window)); - window.src.x = 0; - window.src.y = 0; - window.src.w = mode->hdisplay; - window.src.h = mode->vdisplay; - window.dst.x = 0; - window.dst.y = 0; - window.dst.w = mode->hdisplay; - window.dst.h = mode->vdisplay; - window.format = tegra_dc_format(crtc->fb->pixel_format); - window.bits_per_pixel = crtc->fb->bits_per_pixel; - window.stride[0] = crtc->fb->pitches[0]; - window.base[0] = bo->paddr; - - err = tegra_dc_setup_window(dc, 0, &window); - if (err < 0) - dev_err(dc->dev, "failed to enable root plane\n"); - - return 0; -} - -static int tegra_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y, - struct drm_framebuffer *old_fb) -{ - struct tegra_dc *dc = to_tegra_dc(crtc); - - return tegra_dc_set_base(dc, x, y, crtc->fb); -} - -static void tegra_crtc_prepare(struct drm_crtc *crtc) -{ - struct tegra_dc *dc = to_tegra_dc(crtc); - unsigned int syncpt; - unsigned long value; - - /* hardware initialization */ - tegra_periph_reset_deassert(dc->clk); - usleep_range(10000, 20000); - - if (dc->pipe) - syncpt = SYNCPT_VBLANK1; - else - syncpt = SYNCPT_VBLANK0; - - /* initialize display controller */ - tegra_dc_writel(dc, 0x00000100, DC_CMD_GENERAL_INCR_SYNCPT_CNTRL); - tegra_dc_writel(dc, 0x100 | syncpt, DC_CMD_CONT_SYNCPT_VSYNC); - - value = WIN_A_UF_INT | WIN_B_UF_INT | WIN_C_UF_INT | WIN_A_OF_INT; - tegra_dc_writel(dc, value, DC_CMD_INT_TYPE); - - value = WIN_A_UF_INT | WIN_B_UF_INT | WIN_C_UF_INT | - WIN_A_OF_INT | WIN_B_OF_INT | WIN_C_OF_INT; - tegra_dc_writel(dc, value, DC_CMD_INT_POLARITY); - - value = PW0_ENABLE | PW1_ENABLE | PW2_ENABLE | PW3_ENABLE | - PW4_ENABLE | PM0_ENABLE | PM1_ENABLE; - tegra_dc_writel(dc, value, DC_CMD_DISPLAY_POWER_CONTROL); - - value = tegra_dc_readl(dc, DC_CMD_DISPLAY_COMMAND); - value |= DISP_CTRL_MODE_C_DISPLAY; - tegra_dc_writel(dc, value, DC_CMD_DISPLAY_COMMAND); - - /* initialize timer */ - value = CURSOR_THRESHOLD(0) | WINDOW_A_THRESHOLD(0x20) | - WINDOW_B_THRESHOLD(0x20) | WINDOW_C_THRESHOLD(0x20); - tegra_dc_writel(dc, value, DC_DISP_DISP_MEM_HIGH_PRIORITY); - - value = CURSOR_THRESHOLD(0) | WINDOW_A_THRESHOLD(1) | - WINDOW_B_THRESHOLD(1) | WINDOW_C_THRESHOLD(1); - tegra_dc_writel(dc, value, DC_DISP_DISP_MEM_HIGH_PRIORITY_TIMER); - - value = VBLANK_INT | WIN_A_UF_INT | WIN_B_UF_INT | WIN_C_UF_INT; - tegra_dc_writel(dc, value, DC_CMD_INT_ENABLE); - - value = WIN_A_UF_INT | WIN_B_UF_INT | WIN_C_UF_INT; - tegra_dc_writel(dc, value, DC_CMD_INT_MASK); -} - -static void tegra_crtc_commit(struct drm_crtc *crtc) -{ - struct tegra_dc *dc = to_tegra_dc(crtc); - unsigned long value; - - value = GENERAL_UPDATE | WIN_A_UPDATE; - tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL); - - value = GENERAL_ACT_REQ | WIN_A_ACT_REQ; - tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL); - - drm_vblank_post_modeset(crtc->dev, dc->pipe); -} - -static void tegra_crtc_load_lut(struct drm_crtc *crtc) -{ -} - -static const struct drm_crtc_helper_funcs tegra_crtc_helper_funcs = { - .disable = tegra_crtc_disable, - .mode_fixup = tegra_crtc_mode_fixup, - .mode_set = tegra_crtc_mode_set, - .mode_set_base = tegra_crtc_mode_set_base, - .prepare = tegra_crtc_prepare, - .commit = tegra_crtc_commit, - .load_lut = tegra_crtc_load_lut, -}; - -static irqreturn_t tegra_dc_irq(int irq, void *data) -{ - struct tegra_dc *dc = data; - unsigned long status; - - status = tegra_dc_readl(dc, DC_CMD_INT_STATUS); - tegra_dc_writel(dc, status, DC_CMD_INT_STATUS); - - if (status & FRAME_END_INT) { - /* - dev_dbg(dc->dev, "%s(): frame end\n", __func__); - */ - } - - if (status & VBLANK_INT) { - /* - dev_dbg(dc->dev, "%s(): vertical blank\n", __func__); - */ - drm_handle_vblank(dc->base.dev, dc->pipe); - tegra_dc_finish_page_flip(dc); - } - - if (status & (WIN_A_UF_INT | WIN_B_UF_INT | WIN_C_UF_INT)) { - /* - dev_dbg(dc->dev, "%s(): underflow\n", __func__); - */ - } - - return IRQ_HANDLED; -} - -static int tegra_dc_show_regs(struct seq_file *s, void *data) -{ - struct drm_info_node *node = s->private; - struct tegra_dc *dc = node->info_ent->data; - -#define DUMP_REG(name) \ - seq_printf(s, "%-40s %#05x %08lx\n", #name, name, \ - tegra_dc_readl(dc, name)) - - DUMP_REG(DC_CMD_GENERAL_INCR_SYNCPT); - DUMP_REG(DC_CMD_GENERAL_INCR_SYNCPT_CNTRL); - DUMP_REG(DC_CMD_GENERAL_INCR_SYNCPT_ERROR); - DUMP_REG(DC_CMD_WIN_A_INCR_SYNCPT); - DUMP_REG(DC_CMD_WIN_A_INCR_SYNCPT_CNTRL); - DUMP_REG(DC_CMD_WIN_A_INCR_SYNCPT_ERROR); - DUMP_REG(DC_CMD_WIN_B_INCR_SYNCPT); - DUMP_REG(DC_CMD_WIN_B_INCR_SYNCPT_CNTRL); - DUMP_REG(DC_CMD_WIN_B_INCR_SYNCPT_ERROR); - DUMP_REG(DC_CMD_WIN_C_INCR_SYNCPT); - DUMP_REG(DC_CMD_WIN_C_INCR_SYNCPT_CNTRL); - DUMP_REG(DC_CMD_WIN_C_INCR_SYNCPT_ERROR); - DUMP_REG(DC_CMD_CONT_SYNCPT_VSYNC); - DUMP_REG(DC_CMD_DISPLAY_COMMAND_OPTION0); - DUMP_REG(DC_CMD_DISPLAY_COMMAND); - DUMP_REG(DC_CMD_SIGNAL_RAISE); - DUMP_REG(DC_CMD_DISPLAY_POWER_CONTROL); - DUMP_REG(DC_CMD_INT_STATUS); - DUMP_REG(DC_CMD_INT_MASK); - DUMP_REG(DC_CMD_INT_ENABLE); - DUMP_REG(DC_CMD_INT_TYPE); - DUMP_REG(DC_CMD_INT_POLARITY); - DUMP_REG(DC_CMD_SIGNAL_RAISE1); - DUMP_REG(DC_CMD_SIGNAL_RAISE2); - DUMP_REG(DC_CMD_SIGNAL_RAISE3); - DUMP_REG(DC_CMD_STATE_ACCESS); - DUMP_REG(DC_CMD_STATE_CONTROL); - DUMP_REG(DC_CMD_DISPLAY_WINDOW_HEADER); - DUMP_REG(DC_CMD_REG_ACT_CONTROL); - DUMP_REG(DC_COM_CRC_CONTROL); - DUMP_REG(DC_COM_CRC_CHECKSUM); - DUMP_REG(DC_COM_PIN_OUTPUT_ENABLE(0)); - DUMP_REG(DC_COM_PIN_OUTPUT_ENABLE(1)); - DUMP_REG(DC_COM_PIN_OUTPUT_ENABLE(2)); - DUMP_REG(DC_COM_PIN_OUTPUT_ENABLE(3)); - DUMP_REG(DC_COM_PIN_OUTPUT_POLARITY(0)); - DUMP_REG(DC_COM_PIN_OUTPUT_POLARITY(1)); - DUMP_REG(DC_COM_PIN_OUTPUT_POLARITY(2)); - DUMP_REG(DC_COM_PIN_OUTPUT_POLARITY(3)); - DUMP_REG(DC_COM_PIN_OUTPUT_DATA(0)); - DUMP_REG(DC_COM_PIN_OUTPUT_DATA(1)); - DUMP_REG(DC_COM_PIN_OUTPUT_DATA(2)); - DUMP_REG(DC_COM_PIN_OUTPUT_DATA(3)); - DUMP_REG(DC_COM_PIN_INPUT_ENABLE(0)); - DUMP_REG(DC_COM_PIN_INPUT_ENABLE(1)); - DUMP_REG(DC_COM_PIN_INPUT_ENABLE(2)); - DUMP_REG(DC_COM_PIN_INPUT_ENABLE(3)); - DUMP_REG(DC_COM_PIN_INPUT_DATA(0)); - DUMP_REG(DC_COM_PIN_INPUT_DATA(1)); - DUMP_REG(DC_COM_PIN_OUTPUT_SELECT(0)); - DUMP_REG(DC_COM_PIN_OUTPUT_SELECT(1)); - DUMP_REG(DC_COM_PIN_OUTPUT_SELECT(2)); - DUMP_REG(DC_COM_PIN_OUTPUT_SELECT(3)); - DUMP_REG(DC_COM_PIN_OUTPUT_SELECT(4)); - DUMP_REG(DC_COM_PIN_OUTPUT_SELECT(5)); - DUMP_REG(DC_COM_PIN_OUTPUT_SELECT(6)); - DUMP_REG(DC_COM_PIN_MISC_CONTROL); - DUMP_REG(DC_COM_PIN_PM0_CONTROL); - DUMP_REG(DC_COM_PIN_PM0_DUTY_CYCLE); - DUMP_REG(DC_COM_PIN_PM1_CONTROL); - DUMP_REG(DC_COM_PIN_PM1_DUTY_CYCLE); - DUMP_REG(DC_COM_SPI_CONTROL); - DUMP_REG(DC_COM_SPI_START_BYTE); - DUMP_REG(DC_COM_HSPI_WRITE_DATA_AB); - DUMP_REG(DC_COM_HSPI_WRITE_DATA_CD); - DUMP_REG(DC_COM_HSPI_CS_DC); - DUMP_REG(DC_COM_SCRATCH_REGISTER_A); - DUMP_REG(DC_COM_SCRATCH_REGISTER_B); - DUMP_REG(DC_COM_GPIO_CTRL); - DUMP_REG(DC_COM_GPIO_DEBOUNCE_COUNTER); - DUMP_REG(DC_COM_CRC_CHECKSUM_LATCHED); - DUMP_REG(DC_DISP_DISP_SIGNAL_OPTIONS0); - DUMP_REG(DC_DISP_DISP_SIGNAL_OPTIONS1); - DUMP_REG(DC_DISP_DISP_WIN_OPTIONS); - DUMP_REG(DC_DISP_DISP_MEM_HIGH_PRIORITY); - DUMP_REG(DC_DISP_DISP_MEM_HIGH_PRIORITY_TIMER); - DUMP_REG(DC_DISP_DISP_TIMING_OPTIONS); - DUMP_REG(DC_DISP_REF_TO_SYNC); - DUMP_REG(DC_DISP_SYNC_WIDTH); - DUMP_REG(DC_DISP_BACK_PORCH); - DUMP_REG(DC_DISP_ACTIVE); - DUMP_REG(DC_DISP_FRONT_PORCH); - DUMP_REG(DC_DISP_H_PULSE0_CONTROL); - DUMP_REG(DC_DISP_H_PULSE0_POSITION_A); - DUMP_REG(DC_DISP_H_PULSE0_POSITION_B); - DUMP_REG(DC_DISP_H_PULSE0_POSITION_C); - DUMP_REG(DC_DISP_H_PULSE0_POSITION_D); - DUMP_REG(DC_DISP_H_PULSE1_CONTROL); - DUMP_REG(DC_DISP_H_PULSE1_POSITION_A); - DUMP_REG(DC_DISP_H_PULSE1_POSITION_B); - DUMP_REG(DC_DISP_H_PULSE1_POSITION_C); - DUMP_REG(DC_DISP_H_PULSE1_POSITION_D); - DUMP_REG(DC_DISP_H_PULSE2_CONTROL); - DUMP_REG(DC_DISP_H_PULSE2_POSITION_A); - DUMP_REG(DC_DISP_H_PULSE2_POSITION_B); - DUMP_REG(DC_DISP_H_PULSE2_POSITION_C); - DUMP_REG(DC_DISP_H_PULSE2_POSITION_D); - DUMP_REG(DC_DISP_V_PULSE0_CONTROL); - DUMP_REG(DC_DISP_V_PULSE0_POSITION_A); - DUMP_REG(DC_DISP_V_PULSE0_POSITION_B); - DUMP_REG(DC_DISP_V_PULSE0_POSITION_C); - DUMP_REG(DC_DISP_V_PULSE1_CONTROL); - DUMP_REG(DC_DISP_V_PULSE1_POSITION_A); - DUMP_REG(DC_DISP_V_PULSE1_POSITION_B); - DUMP_REG(DC_DISP_V_PULSE1_POSITION_C); - DUMP_REG(DC_DISP_V_PULSE2_CONTROL); - DUMP_REG(DC_DISP_V_PULSE2_POSITION_A); - DUMP_REG(DC_DISP_V_PULSE3_CONTROL); - DUMP_REG(DC_DISP_V_PULSE3_POSITION_A); - DUMP_REG(DC_DISP_M0_CONTROL); - DUMP_REG(DC_DISP_M1_CONTROL); - DUMP_REG(DC_DISP_DI_CONTROL); - DUMP_REG(DC_DISP_PP_CONTROL); - DUMP_REG(DC_DISP_PP_SELECT_A); - DUMP_REG(DC_DISP_PP_SELECT_B); - DUMP_REG(DC_DISP_PP_SELECT_C); - DUMP_REG(DC_DISP_PP_SELECT_D); - DUMP_REG(DC_DISP_DISP_CLOCK_CONTROL); - DUMP_REG(DC_DISP_DISP_INTERFACE_CONTROL); - DUMP_REG(DC_DISP_DISP_COLOR_CONTROL); - DUMP_REG(DC_DISP_SHIFT_CLOCK_OPTIONS); - DUMP_REG(DC_DISP_DATA_ENABLE_OPTIONS); - DUMP_REG(DC_DISP_SERIAL_INTERFACE_OPTIONS); - DUMP_REG(DC_DISP_LCD_SPI_OPTIONS); - DUMP_REG(DC_DISP_BORDER_COLOR); - DUMP_REG(DC_DISP_COLOR_KEY0_LOWER); - DUMP_REG(DC_DISP_COLOR_KEY0_UPPER); - DUMP_REG(DC_DISP_COLOR_KEY1_LOWER); - DUMP_REG(DC_DISP_COLOR_KEY1_UPPER); - DUMP_REG(DC_DISP_CURSOR_FOREGROUND); - DUMP_REG(DC_DISP_CURSOR_BACKGROUND); - DUMP_REG(DC_DISP_CURSOR_START_ADDR); - DUMP_REG(DC_DISP_CURSOR_START_ADDR_NS); - DUMP_REG(DC_DISP_CURSOR_POSITION); - DUMP_REG(DC_DISP_CURSOR_POSITION_NS); - DUMP_REG(DC_DISP_INIT_SEQ_CONTROL); - DUMP_REG(DC_DISP_SPI_INIT_SEQ_DATA_A); - DUMP_REG(DC_DISP_SPI_INIT_SEQ_DATA_B); - DUMP_REG(DC_DISP_SPI_INIT_SEQ_DATA_C); - DUMP_REG(DC_DISP_SPI_INIT_SEQ_DATA_D); - DUMP_REG(DC_DISP_DC_MCCIF_FIFOCTRL); - DUMP_REG(DC_DISP_MCCIF_DISPLAY0A_HYST); - DUMP_REG(DC_DISP_MCCIF_DISPLAY0B_HYST); - DUMP_REG(DC_DISP_MCCIF_DISPLAY1A_HYST); - DUMP_REG(DC_DISP_MCCIF_DISPLAY1B_HYST); - DUMP_REG(DC_DISP_DAC_CRT_CTRL); - DUMP_REG(DC_DISP_DISP_MISC_CONTROL); - DUMP_REG(DC_DISP_SD_CONTROL); - DUMP_REG(DC_DISP_SD_CSC_COEFF); - DUMP_REG(DC_DISP_SD_LUT(0)); - DUMP_REG(DC_DISP_SD_LUT(1)); - DUMP_REG(DC_DISP_SD_LUT(2)); - DUMP_REG(DC_DISP_SD_LUT(3)); - DUMP_REG(DC_DISP_SD_LUT(4)); - DUMP_REG(DC_DISP_SD_LUT(5)); - DUMP_REG(DC_DISP_SD_LUT(6)); - DUMP_REG(DC_DISP_SD_LUT(7)); - DUMP_REG(DC_DISP_SD_LUT(8)); - DUMP_REG(DC_DISP_SD_FLICKER_CONTROL); - DUMP_REG(DC_DISP_DC_PIXEL_COUNT); - DUMP_REG(DC_DISP_SD_HISTOGRAM(0)); - DUMP_REG(DC_DISP_SD_HISTOGRAM(1)); - DUMP_REG(DC_DISP_SD_HISTOGRAM(2)); - DUMP_REG(DC_DISP_SD_HISTOGRAM(3)); - DUMP_REG(DC_DISP_SD_HISTOGRAM(4)); - DUMP_REG(DC_DISP_SD_HISTOGRAM(5)); - DUMP_REG(DC_DISP_SD_HISTOGRAM(6)); - DUMP_REG(DC_DISP_SD_HISTOGRAM(7)); - DUMP_REG(DC_DISP_SD_BL_TF(0)); - DUMP_REG(DC_DISP_SD_BL_TF(1)); - DUMP_REG(DC_DISP_SD_BL_TF(2)); - DUMP_REG(DC_DISP_SD_BL_TF(3)); - DUMP_REG(DC_DISP_SD_BL_CONTROL); - DUMP_REG(DC_DISP_SD_HW_K_VALUES); - DUMP_REG(DC_DISP_SD_MAN_K_VALUES); - DUMP_REG(DC_WIN_WIN_OPTIONS); - DUMP_REG(DC_WIN_BYTE_SWAP); - DUMP_REG(DC_WIN_BUFFER_CONTROL); - DUMP_REG(DC_WIN_COLOR_DEPTH); - DUMP_REG(DC_WIN_POSITION); - DUMP_REG(DC_WIN_SIZE); - DUMP_REG(DC_WIN_PRESCALED_SIZE); - DUMP_REG(DC_WIN_H_INITIAL_DDA); - DUMP_REG(DC_WIN_V_INITIAL_DDA); - DUMP_REG(DC_WIN_DDA_INC); - DUMP_REG(DC_WIN_LINE_STRIDE); - DUMP_REG(DC_WIN_BUF_STRIDE); - DUMP_REG(DC_WIN_UV_BUF_STRIDE); - DUMP_REG(DC_WIN_BUFFER_ADDR_MODE); - DUMP_REG(DC_WIN_DV_CONTROL); - DUMP_REG(DC_WIN_BLEND_NOKEY); - DUMP_REG(DC_WIN_BLEND_1WIN); - DUMP_REG(DC_WIN_BLEND_2WIN_X); - DUMP_REG(DC_WIN_BLEND_2WIN_Y); - DUMP_REG(DC_WIN_BLEND_3WIN_XY); - DUMP_REG(DC_WIN_HP_FETCH_CONTROL); - DUMP_REG(DC_WINBUF_START_ADDR); - DUMP_REG(DC_WINBUF_START_ADDR_NS); - DUMP_REG(DC_WINBUF_START_ADDR_U); - DUMP_REG(DC_WINBUF_START_ADDR_U_NS); - DUMP_REG(DC_WINBUF_START_ADDR_V); - DUMP_REG(DC_WINBUF_START_ADDR_V_NS); - DUMP_REG(DC_WINBUF_ADDR_H_OFFSET); - DUMP_REG(DC_WINBUF_ADDR_H_OFFSET_NS); - DUMP_REG(DC_WINBUF_ADDR_V_OFFSET); - DUMP_REG(DC_WINBUF_ADDR_V_OFFSET_NS); - DUMP_REG(DC_WINBUF_UFLOW_STATUS); - DUMP_REG(DC_WINBUF_AD_UFLOW_STATUS); - DUMP_REG(DC_WINBUF_BD_UFLOW_STATUS); - DUMP_REG(DC_WINBUF_CD_UFLOW_STATUS); - -#undef DUMP_REG - - return 0; -} - -static struct drm_info_list debugfs_files[] = { - { "regs", tegra_dc_show_regs, 0, NULL }, -}; - -static int tegra_dc_debugfs_init(struct tegra_dc *dc, struct drm_minor *minor) -{ - unsigned int i; - char *name; - int err; - - name = kasprintf(GFP_KERNEL, "dc.%d", dc->pipe); - dc->debugfs = debugfs_create_dir(name, minor->debugfs_root); - kfree(name); - - if (!dc->debugfs) - return -ENOMEM; - - dc->debugfs_files = kmemdup(debugfs_files, sizeof(debugfs_files), - GFP_KERNEL); - if (!dc->debugfs_files) { - err = -ENOMEM; - goto remove; - } - - for (i = 0; i < ARRAY_SIZE(debugfs_files); i++) - dc->debugfs_files[i].data = dc; - - err = drm_debugfs_create_files(dc->debugfs_files, - ARRAY_SIZE(debugfs_files), - dc->debugfs, minor); - if (err < 0) - goto free; - - dc->minor = minor; - - return 0; - -free: - kfree(dc->debugfs_files); - dc->debugfs_files = NULL; -remove: - debugfs_remove(dc->debugfs); - dc->debugfs = NULL; - - return err; -} - -static int tegra_dc_debugfs_exit(struct tegra_dc *dc) -{ - drm_debugfs_remove_files(dc->debugfs_files, ARRAY_SIZE(debugfs_files), - dc->minor); - dc->minor = NULL; - - kfree(dc->debugfs_files); - dc->debugfs_files = NULL; - - debugfs_remove(dc->debugfs); - dc->debugfs = NULL; - - return 0; -} - -static int tegra_dc_drm_init(struct host1x_client *client, - struct drm_device *drm) -{ - struct tegra_dc *dc = host1x_client_to_dc(client); - int err; - - dc->pipe = drm->mode_config.num_crtc; - - drm_crtc_init(drm, &dc->base, &tegra_crtc_funcs); - drm_mode_crtc_set_gamma_size(&dc->base, 256); - drm_crtc_helper_add(&dc->base, &tegra_crtc_helper_funcs); - - err = tegra_dc_rgb_init(drm, dc); - if (err < 0 && err != -ENODEV) { - dev_err(dc->dev, "failed to initialize RGB output: %d\n", err); - return err; - } - - err = tegra_dc_add_planes(drm, dc); - if (err < 0) - return err; - - if (IS_ENABLED(CONFIG_DEBUG_FS)) { - err = tegra_dc_debugfs_init(dc, drm->primary); - if (err < 0) - dev_err(dc->dev, "debugfs setup failed: %d\n", err); - } - - err = devm_request_irq(dc->dev, dc->irq, tegra_dc_irq, 0, - dev_name(dc->dev), dc); - if (err < 0) { - dev_err(dc->dev, "failed to request IRQ#%u: %d\n", dc->irq, - err); - return err; - } - - return 0; -} - -static int tegra_dc_drm_exit(struct host1x_client *client) -{ - struct tegra_dc *dc = host1x_client_to_dc(client); - int err; - - devm_free_irq(dc->dev, dc->irq, dc); - - if (IS_ENABLED(CONFIG_DEBUG_FS)) { - err = tegra_dc_debugfs_exit(dc); - if (err < 0) - dev_err(dc->dev, "debugfs cleanup failed: %d\n", err); - } - - err = tegra_dc_rgb_exit(dc); - if (err) { - dev_err(dc->dev, "failed to shutdown RGB output: %d\n", err); - return err; - } - - return 0; -} - -static const struct host1x_client_ops dc_client_ops = { - .drm_init = tegra_dc_drm_init, - .drm_exit = tegra_dc_drm_exit, -}; - -static int tegra_dc_probe(struct platform_device *pdev) -{ - struct host1x_drm *host1x = host1x_get_drm_data(pdev->dev.parent); - struct resource *regs; - struct tegra_dc *dc; - int err; - - dc = devm_kzalloc(&pdev->dev, sizeof(*dc), GFP_KERNEL); - if (!dc) - return -ENOMEM; - - spin_lock_init(&dc->lock); - INIT_LIST_HEAD(&dc->list); - dc->dev = &pdev->dev; - - dc->clk = devm_clk_get(&pdev->dev, NULL); - if (IS_ERR(dc->clk)) { - dev_err(&pdev->dev, "failed to get clock\n"); - return PTR_ERR(dc->clk); - } - - err = clk_prepare_enable(dc->clk); - if (err < 0) - return err; - - regs = platform_get_resource(pdev, IORESOURCE_MEM, 0); - dc->regs = devm_ioremap_resource(&pdev->dev, regs); - if (IS_ERR(dc->regs)) - return PTR_ERR(dc->regs); - - dc->irq = platform_get_irq(pdev, 0); - if (dc->irq < 0) { - dev_err(&pdev->dev, "failed to get IRQ\n"); - return -ENXIO; - } - - INIT_LIST_HEAD(&dc->client.list); - dc->client.ops = &dc_client_ops; - dc->client.dev = &pdev->dev; - - err = tegra_dc_rgb_probe(dc); - if (err < 0 && err != -ENODEV) { - dev_err(&pdev->dev, "failed to probe RGB output: %d\n", err); - return err; - } - - err = host1x_register_client(host1x, &dc->client); - if (err < 0) { - dev_err(&pdev->dev, "failed to register host1x client: %d\n", - err); - return err; - } - - platform_set_drvdata(pdev, dc); - - return 0; -} - -static int tegra_dc_remove(struct platform_device *pdev) -{ - struct host1x_drm *host1x = host1x_get_drm_data(pdev->dev.parent); - struct tegra_dc *dc = platform_get_drvdata(pdev); - int err; - - err = host1x_unregister_client(host1x, &dc->client); - if (err < 0) { - dev_err(&pdev->dev, "failed to unregister host1x client: %d\n", - err); - return err; - } - - clk_disable_unprepare(dc->clk); - - return 0; -} - -static struct of_device_id tegra_dc_of_match[] = { - { .compatible = "nvidia,tegra30-dc", }, - { .compatible = "nvidia,tegra20-dc", }, - { }, -}; - -struct platform_driver tegra_dc_driver = { - .driver = { - .name = "tegra-dc", - .owner = THIS_MODULE, - .of_match_table = tegra_dc_of_match, - }, - .probe = tegra_dc_probe, - .remove = tegra_dc_remove, -}; diff --git a/drivers/gpu/host1x/drm/dc.h b/drivers/gpu/host1x/drm/dc.h deleted file mode 100644 index 79eaec9aac77..000000000000 --- a/drivers/gpu/host1x/drm/dc.h +++ /dev/null @@ -1,400 +0,0 @@ -/* - * Copyright (C) 2012 Avionic Design GmbH - * Copyright (C) 2012 NVIDIA CORPORATION. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#ifndef TEGRA_DC_H -#define TEGRA_DC_H 1 - -#define DC_CMD_GENERAL_INCR_SYNCPT 0x000 -#define DC_CMD_GENERAL_INCR_SYNCPT_CNTRL 0x001 -#define DC_CMD_GENERAL_INCR_SYNCPT_ERROR 0x002 -#define DC_CMD_WIN_A_INCR_SYNCPT 0x008 -#define DC_CMD_WIN_A_INCR_SYNCPT_CNTRL 0x009 -#define DC_CMD_WIN_A_INCR_SYNCPT_ERROR 0x00a -#define DC_CMD_WIN_B_INCR_SYNCPT 0x010 -#define DC_CMD_WIN_B_INCR_SYNCPT_CNTRL 0x011 -#define DC_CMD_WIN_B_INCR_SYNCPT_ERROR 0x012 -#define DC_CMD_WIN_C_INCR_SYNCPT 0x018 -#define DC_CMD_WIN_C_INCR_SYNCPT_CNTRL 0x019 -#define DC_CMD_WIN_C_INCR_SYNCPT_ERROR 0x01a -#define DC_CMD_CONT_SYNCPT_VSYNC 0x028 -#define DC_CMD_DISPLAY_COMMAND_OPTION0 0x031 -#define DC_CMD_DISPLAY_COMMAND 0x032 -#define DISP_CTRL_MODE_STOP (0 << 5) -#define DISP_CTRL_MODE_C_DISPLAY (1 << 5) -#define DISP_CTRL_MODE_NC_DISPLAY (2 << 5) -#define DC_CMD_SIGNAL_RAISE 0x033 -#define DC_CMD_DISPLAY_POWER_CONTROL 0x036 -#define PW0_ENABLE (1 << 0) -#define PW1_ENABLE (1 << 2) -#define PW2_ENABLE (1 << 4) -#define PW3_ENABLE (1 << 6) -#define PW4_ENABLE (1 << 8) -#define PM0_ENABLE (1 << 16) -#define PM1_ENABLE (1 << 18) - -#define DC_CMD_INT_STATUS 0x037 -#define DC_CMD_INT_MASK 0x038 -#define DC_CMD_INT_ENABLE 0x039 -#define DC_CMD_INT_TYPE 0x03a -#define DC_CMD_INT_POLARITY 0x03b -#define CTXSW_INT (1 << 0) -#define FRAME_END_INT (1 << 1) -#define VBLANK_INT (1 << 2) -#define WIN_A_UF_INT (1 << 8) -#define WIN_B_UF_INT (1 << 9) -#define WIN_C_UF_INT (1 << 10) -#define WIN_A_OF_INT (1 << 14) -#define WIN_B_OF_INT (1 << 15) -#define WIN_C_OF_INT (1 << 16) - -#define DC_CMD_SIGNAL_RAISE1 0x03c -#define DC_CMD_SIGNAL_RAISE2 0x03d -#define DC_CMD_SIGNAL_RAISE3 0x03e - -#define DC_CMD_STATE_ACCESS 0x040 -#define READ_MUX (1 << 0) -#define WRITE_MUX (1 << 2) - -#define DC_CMD_STATE_CONTROL 0x041 -#define GENERAL_ACT_REQ (1 << 0) -#define WIN_A_ACT_REQ (1 << 1) -#define WIN_B_ACT_REQ (1 << 2) -#define WIN_C_ACT_REQ (1 << 3) -#define GENERAL_UPDATE (1 << 8) -#define WIN_A_UPDATE (1 << 9) -#define WIN_B_UPDATE (1 << 10) -#define WIN_C_UPDATE (1 << 11) -#define NC_HOST_TRIG (1 << 24) - -#define DC_CMD_DISPLAY_WINDOW_HEADER 0x042 -#define WINDOW_A_SELECT (1 << 4) -#define WINDOW_B_SELECT (1 << 5) -#define WINDOW_C_SELECT (1 << 6) - -#define DC_CMD_REG_ACT_CONTROL 0x043 - -#define DC_COM_CRC_CONTROL 0x300 -#define DC_COM_CRC_CHECKSUM 0x301 -#define DC_COM_PIN_OUTPUT_ENABLE(x) (0x302 + (x)) -#define DC_COM_PIN_OUTPUT_POLARITY(x) (0x306 + (x)) -#define LVS_OUTPUT_POLARITY_LOW (1 << 28) -#define LHS_OUTPUT_POLARITY_LOW (1 << 30) -#define DC_COM_PIN_OUTPUT_DATA(x) (0x30a + (x)) -#define DC_COM_PIN_INPUT_ENABLE(x) (0x30e + (x)) -#define DC_COM_PIN_INPUT_DATA(x) (0x312 + (x)) -#define DC_COM_PIN_OUTPUT_SELECT(x) (0x314 + (x)) - -#define DC_COM_PIN_MISC_CONTROL 0x31b -#define DC_COM_PIN_PM0_CONTROL 0x31c -#define DC_COM_PIN_PM0_DUTY_CYCLE 0x31d -#define DC_COM_PIN_PM1_CONTROL 0x31e -#define DC_COM_PIN_PM1_DUTY_CYCLE 0x31f - -#define DC_COM_SPI_CONTROL 0x320 -#define DC_COM_SPI_START_BYTE 0x321 -#define DC_COM_HSPI_WRITE_DATA_AB 0x322 -#define DC_COM_HSPI_WRITE_DATA_CD 0x323 -#define DC_COM_HSPI_CS_DC 0x324 -#define DC_COM_SCRATCH_REGISTER_A 0x325 -#define DC_COM_SCRATCH_REGISTER_B 0x326 -#define DC_COM_GPIO_CTRL 0x327 -#define DC_COM_GPIO_DEBOUNCE_COUNTER 0x328 -#define DC_COM_CRC_CHECKSUM_LATCHED 0x329 - -#define DC_DISP_DISP_SIGNAL_OPTIONS0 0x400 -#define H_PULSE_0_ENABLE (1 << 8) -#define H_PULSE_1_ENABLE (1 << 10) -#define H_PULSE_2_ENABLE (1 << 12) - -#define DC_DISP_DISP_SIGNAL_OPTIONS1 0x401 - -#define DC_DISP_DISP_WIN_OPTIONS 0x402 -#define HDMI_ENABLE (1 << 30) - -#define DC_DISP_DISP_MEM_HIGH_PRIORITY 0x403 -#define CURSOR_THRESHOLD(x) (((x) & 0x03) << 24) -#define WINDOW_A_THRESHOLD(x) (((x) & 0x7f) << 16) -#define WINDOW_B_THRESHOLD(x) (((x) & 0x7f) << 8) -#define WINDOW_C_THRESHOLD(x) (((x) & 0xff) << 0) - -#define DC_DISP_DISP_MEM_HIGH_PRIORITY_TIMER 0x404 -#define CURSOR_DELAY(x) (((x) & 0x3f) << 24) -#define WINDOW_A_DELAY(x) (((x) & 0x3f) << 16) -#define WINDOW_B_DELAY(x) (((x) & 0x3f) << 8) -#define WINDOW_C_DELAY(x) (((x) & 0x3f) << 0) - -#define DC_DISP_DISP_TIMING_OPTIONS 0x405 -#define VSYNC_H_POSITION(x) ((x) & 0xfff) - -#define DC_DISP_REF_TO_SYNC 0x406 -#define DC_DISP_SYNC_WIDTH 0x407 -#define DC_DISP_BACK_PORCH 0x408 -#define DC_DISP_ACTIVE 0x409 -#define DC_DISP_FRONT_PORCH 0x40a -#define DC_DISP_H_PULSE0_CONTROL 0x40b -#define DC_DISP_H_PULSE0_POSITION_A 0x40c -#define DC_DISP_H_PULSE0_POSITION_B 0x40d -#define DC_DISP_H_PULSE0_POSITION_C 0x40e -#define DC_DISP_H_PULSE0_POSITION_D 0x40f -#define DC_DISP_H_PULSE1_CONTROL 0x410 -#define DC_DISP_H_PULSE1_POSITION_A 0x411 -#define DC_DISP_H_PULSE1_POSITION_B 0x412 -#define DC_DISP_H_PULSE1_POSITION_C 0x413 -#define DC_DISP_H_PULSE1_POSITION_D 0x414 -#define DC_DISP_H_PULSE2_CONTROL 0x415 -#define DC_DISP_H_PULSE2_POSITION_A 0x416 -#define DC_DISP_H_PULSE2_POSITION_B 0x417 -#define DC_DISP_H_PULSE2_POSITION_C 0x418 -#define DC_DISP_H_PULSE2_POSITION_D 0x419 -#define DC_DISP_V_PULSE0_CONTROL 0x41a -#define DC_DISP_V_PULSE0_POSITION_A 0x41b -#define DC_DISP_V_PULSE0_POSITION_B 0x41c -#define DC_DISP_V_PULSE0_POSITION_C 0x41d -#define DC_DISP_V_PULSE1_CONTROL 0x41e -#define DC_DISP_V_PULSE1_POSITION_A 0x41f -#define DC_DISP_V_PULSE1_POSITION_B 0x420 -#define DC_DISP_V_PULSE1_POSITION_C 0x421 -#define DC_DISP_V_PULSE2_CONTROL 0x422 -#define DC_DISP_V_PULSE2_POSITION_A 0x423 -#define DC_DISP_V_PULSE3_CONTROL 0x424 -#define DC_DISP_V_PULSE3_POSITION_A 0x425 -#define DC_DISP_M0_CONTROL 0x426 -#define DC_DISP_M1_CONTROL 0x427 -#define DC_DISP_DI_CONTROL 0x428 -#define DC_DISP_PP_CONTROL 0x429 -#define DC_DISP_PP_SELECT_A 0x42a -#define DC_DISP_PP_SELECT_B 0x42b -#define DC_DISP_PP_SELECT_C 0x42c -#define DC_DISP_PP_SELECT_D 0x42d - -#define PULSE_MODE_NORMAL (0 << 3) -#define PULSE_MODE_ONE_CLOCK (1 << 3) -#define PULSE_POLARITY_HIGH (0 << 4) -#define PULSE_POLARITY_LOW (1 << 4) -#define PULSE_QUAL_ALWAYS (0 << 6) -#define PULSE_QUAL_VACTIVE (2 << 6) -#define PULSE_QUAL_VACTIVE1 (3 << 6) -#define PULSE_LAST_START_A (0 << 8) -#define PULSE_LAST_END_A (1 << 8) -#define PULSE_LAST_START_B (2 << 8) -#define PULSE_LAST_END_B (3 << 8) -#define PULSE_LAST_START_C (4 << 8) -#define PULSE_LAST_END_C (5 << 8) -#define PULSE_LAST_START_D (6 << 8) -#define PULSE_LAST_END_D (7 << 8) - -#define PULSE_START(x) (((x) & 0xfff) << 0) -#define PULSE_END(x) (((x) & 0xfff) << 16) - -#define DC_DISP_DISP_CLOCK_CONTROL 0x42e -#define PIXEL_CLK_DIVIDER_PCD1 (0 << 8) -#define PIXEL_CLK_DIVIDER_PCD1H (1 << 8) -#define PIXEL_CLK_DIVIDER_PCD2 (2 << 8) -#define PIXEL_CLK_DIVIDER_PCD3 (3 << 8) -#define PIXEL_CLK_DIVIDER_PCD4 (4 << 8) -#define PIXEL_CLK_DIVIDER_PCD6 (5 << 8) -#define PIXEL_CLK_DIVIDER_PCD8 (6 << 8) -#define PIXEL_CLK_DIVIDER_PCD9 (7 << 8) -#define PIXEL_CLK_DIVIDER_PCD12 (8 << 8) -#define PIXEL_CLK_DIVIDER_PCD16 (9 << 8) -#define PIXEL_CLK_DIVIDER_PCD18 (10 << 8) -#define PIXEL_CLK_DIVIDER_PCD24 (11 << 8) -#define PIXEL_CLK_DIVIDER_PCD13 (12 << 8) -#define SHIFT_CLK_DIVIDER(x) ((x) & 0xff) - -#define DC_DISP_DISP_INTERFACE_CONTROL 0x42f -#define DISP_DATA_FORMAT_DF1P1C (0 << 0) -#define DISP_DATA_FORMAT_DF1P2C24B (1 << 0) -#define DISP_DATA_FORMAT_DF1P2C18B (2 << 0) -#define DISP_DATA_FORMAT_DF1P2C16B (3 << 0) -#define DISP_DATA_FORMAT_DF2S (4 << 0) -#define DISP_DATA_FORMAT_DF3S (5 << 0) -#define DISP_DATA_FORMAT_DFSPI (6 << 0) -#define DISP_DATA_FORMAT_DF1P3C24B (7 << 0) -#define DISP_DATA_FORMAT_DF1P3C18B (8 << 0) -#define DISP_ALIGNMENT_MSB (0 << 8) -#define DISP_ALIGNMENT_LSB (1 << 8) -#define DISP_ORDER_RED_BLUE (0 << 9) -#define DISP_ORDER_BLUE_RED (1 << 9) - -#define DC_DISP_DISP_COLOR_CONTROL 0x430 -#define BASE_COLOR_SIZE666 (0 << 0) -#define BASE_COLOR_SIZE111 (1 << 0) -#define BASE_COLOR_SIZE222 (2 << 0) -#define BASE_COLOR_SIZE333 (3 << 0) -#define BASE_COLOR_SIZE444 (4 << 0) -#define BASE_COLOR_SIZE555 (5 << 0) -#define BASE_COLOR_SIZE565 (6 << 0) -#define BASE_COLOR_SIZE332 (7 << 0) -#define BASE_COLOR_SIZE888 (8 << 0) -#define DITHER_CONTROL_DISABLE (0 << 8) -#define DITHER_CONTROL_ORDERED (2 << 8) -#define DITHER_CONTROL_ERRDIFF (3 << 8) - -#define DC_DISP_SHIFT_CLOCK_OPTIONS 0x431 - -#define DC_DISP_DATA_ENABLE_OPTIONS 0x432 -#define DE_SELECT_ACTIVE_BLANK (0 << 0) -#define DE_SELECT_ACTIVE (1 << 0) -#define DE_SELECT_ACTIVE_IS (2 << 0) -#define DE_CONTROL_ONECLK (0 << 2) -#define DE_CONTROL_NORMAL (1 << 2) -#define DE_CONTROL_EARLY_EXT (2 << 2) -#define DE_CONTROL_EARLY (3 << 2) -#define DE_CONTROL_ACTIVE_BLANK (4 << 2) - -#define DC_DISP_SERIAL_INTERFACE_OPTIONS 0x433 -#define DC_DISP_LCD_SPI_OPTIONS 0x434 -#define DC_DISP_BORDER_COLOR 0x435 -#define DC_DISP_COLOR_KEY0_LOWER 0x436 -#define DC_DISP_COLOR_KEY0_UPPER 0x437 -#define DC_DISP_COLOR_KEY1_LOWER 0x438 -#define DC_DISP_COLOR_KEY1_UPPER 0x439 - -#define DC_DISP_CURSOR_FOREGROUND 0x43c -#define DC_DISP_CURSOR_BACKGROUND 0x43d - -#define DC_DISP_CURSOR_START_ADDR 0x43e -#define DC_DISP_CURSOR_START_ADDR_NS 0x43f - -#define DC_DISP_CURSOR_POSITION 0x440 -#define DC_DISP_CURSOR_POSITION_NS 0x441 - -#define DC_DISP_INIT_SEQ_CONTROL 0x442 -#define DC_DISP_SPI_INIT_SEQ_DATA_A 0x443 -#define DC_DISP_SPI_INIT_SEQ_DATA_B 0x444 -#define DC_DISP_SPI_INIT_SEQ_DATA_C 0x445 -#define DC_DISP_SPI_INIT_SEQ_DATA_D 0x446 - -#define DC_DISP_DC_MCCIF_FIFOCTRL 0x480 -#define DC_DISP_MCCIF_DISPLAY0A_HYST 0x481 -#define DC_DISP_MCCIF_DISPLAY0B_HYST 0x482 -#define DC_DISP_MCCIF_DISPLAY1A_HYST 0x483 -#define DC_DISP_MCCIF_DISPLAY1B_HYST 0x484 - -#define DC_DISP_DAC_CRT_CTRL 0x4c0 -#define DC_DISP_DISP_MISC_CONTROL 0x4c1 -#define DC_DISP_SD_CONTROL 0x4c2 -#define DC_DISP_SD_CSC_COEFF 0x4c3 -#define DC_DISP_SD_LUT(x) (0x4c4 + (x)) -#define DC_DISP_SD_FLICKER_CONTROL 0x4cd -#define DC_DISP_DC_PIXEL_COUNT 0x4ce -#define DC_DISP_SD_HISTOGRAM(x) (0x4cf + (x)) -#define DC_DISP_SD_BL_PARAMETERS 0x4d7 -#define DC_DISP_SD_BL_TF(x) (0x4d8 + (x)) -#define DC_DISP_SD_BL_CONTROL 0x4dc -#define DC_DISP_SD_HW_K_VALUES 0x4dd -#define DC_DISP_SD_MAN_K_VALUES 0x4de - -#define DC_WIN_CSC_YOF 0x611 -#define DC_WIN_CSC_KYRGB 0x612 -#define DC_WIN_CSC_KUR 0x613 -#define DC_WIN_CSC_KVR 0x614 -#define DC_WIN_CSC_KUG 0x615 -#define DC_WIN_CSC_KVG 0x616 -#define DC_WIN_CSC_KUB 0x617 -#define DC_WIN_CSC_KVB 0x618 - -#define DC_WIN_WIN_OPTIONS 0x700 -#define COLOR_EXPAND (1 << 6) -#define CSC_ENABLE (1 << 18) -#define WIN_ENABLE (1 << 30) - -#define DC_WIN_BYTE_SWAP 0x701 -#define BYTE_SWAP_NOSWAP (0 << 0) -#define BYTE_SWAP_SWAP2 (1 << 0) -#define BYTE_SWAP_SWAP4 (2 << 0) -#define BYTE_SWAP_SWAP4HW (3 << 0) - -#define DC_WIN_BUFFER_CONTROL 0x702 -#define BUFFER_CONTROL_HOST (0 << 0) -#define BUFFER_CONTROL_VI (1 << 0) -#define BUFFER_CONTROL_EPP (2 << 0) -#define BUFFER_CONTROL_MPEGE (3 << 0) -#define BUFFER_CONTROL_SB2D (4 << 0) - -#define DC_WIN_COLOR_DEPTH 0x703 -#define WIN_COLOR_DEPTH_P1 0 -#define WIN_COLOR_DEPTH_P2 1 -#define WIN_COLOR_DEPTH_P4 2 -#define WIN_COLOR_DEPTH_P8 3 -#define WIN_COLOR_DEPTH_B4G4R4A4 4 -#define WIN_COLOR_DEPTH_B5G5R5A 5 -#define WIN_COLOR_DEPTH_B5G6R5 6 -#define WIN_COLOR_DEPTH_AB5G5R5 7 -#define WIN_COLOR_DEPTH_B8G8R8A8 12 -#define WIN_COLOR_DEPTH_R8G8B8A8 13 -#define WIN_COLOR_DEPTH_B6x2G6x2R6x2A8 14 -#define WIN_COLOR_DEPTH_R6x2G6x2B6x2A8 15 -#define WIN_COLOR_DEPTH_YCbCr422 16 -#define WIN_COLOR_DEPTH_YUV422 17 -#define WIN_COLOR_DEPTH_YCbCr420P 18 -#define WIN_COLOR_DEPTH_YUV420P 19 -#define WIN_COLOR_DEPTH_YCbCr422P 20 -#define WIN_COLOR_DEPTH_YUV422P 21 -#define WIN_COLOR_DEPTH_YCbCr422R 22 -#define WIN_COLOR_DEPTH_YUV422R 23 -#define WIN_COLOR_DEPTH_YCbCr422RA 24 -#define WIN_COLOR_DEPTH_YUV422RA 25 - -#define DC_WIN_POSITION 0x704 -#define H_POSITION(x) (((x) & 0x1fff) << 0) -#define V_POSITION(x) (((x) & 0x1fff) << 16) - -#define DC_WIN_SIZE 0x705 -#define H_SIZE(x) (((x) & 0x1fff) << 0) -#define V_SIZE(x) (((x) & 0x1fff) << 16) - -#define DC_WIN_PRESCALED_SIZE 0x706 -#define H_PRESCALED_SIZE(x) (((x) & 0x7fff) << 0) -#define V_PRESCALED_SIZE(x) (((x) & 0x1fff) << 16) - -#define DC_WIN_H_INITIAL_DDA 0x707 -#define DC_WIN_V_INITIAL_DDA 0x708 -#define DC_WIN_DDA_INC 0x709 -#define H_DDA_INC(x) (((x) & 0xffff) << 0) -#define V_DDA_INC(x) (((x) & 0xffff) << 16) - -#define DC_WIN_LINE_STRIDE 0x70a -#define DC_WIN_BUF_STRIDE 0x70b -#define DC_WIN_UV_BUF_STRIDE 0x70c -#define DC_WIN_BUFFER_ADDR_MODE 0x70d -#define DC_WIN_DV_CONTROL 0x70e - -#define DC_WIN_BLEND_NOKEY 0x70f -#define DC_WIN_BLEND_1WIN 0x710 -#define DC_WIN_BLEND_2WIN_X 0x711 -#define DC_WIN_BLEND_2WIN_Y 0x712 -#define DC_WIN_BLEND_3WIN_XY 0x713 - -#define DC_WIN_HP_FETCH_CONTROL 0x714 - -#define DC_WINBUF_START_ADDR 0x800 -#define DC_WINBUF_START_ADDR_NS 0x801 -#define DC_WINBUF_START_ADDR_U 0x802 -#define DC_WINBUF_START_ADDR_U_NS 0x803 -#define DC_WINBUF_START_ADDR_V 0x804 -#define DC_WINBUF_START_ADDR_V_NS 0x805 - -#define DC_WINBUF_ADDR_H_OFFSET 0x806 -#define DC_WINBUF_ADDR_H_OFFSET_NS 0x807 -#define DC_WINBUF_ADDR_V_OFFSET 0x808 -#define DC_WINBUF_ADDR_V_OFFSET_NS 0x809 - -#define DC_WINBUF_UFLOW_STATUS 0x80a - -#define DC_WINBUF_AD_UFLOW_STATUS 0xbca -#define DC_WINBUF_BD_UFLOW_STATUS 0xdca -#define DC_WINBUF_CD_UFLOW_STATUS 0xfca - -/* synchronization points */ -#define SYNCPT_VBLANK0 26 -#define SYNCPT_VBLANK1 27 - -#endif /* TEGRA_DC_H */ diff --git a/drivers/gpu/host1x/drm/drm.c b/drivers/gpu/host1x/drm/drm.c deleted file mode 100644 index e184b00faacd..000000000000 --- a/drivers/gpu/host1x/drm/drm.c +++ /dev/null @@ -1,648 +0,0 @@ -/* - * Copyright (C) 2012 Avionic Design GmbH - * Copyright (C) 2012-2013 NVIDIA CORPORATION. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#include <linux/module.h> -#include <linux/of_address.h> -#include <linux/of_platform.h> - -#include <linux/dma-mapping.h> -#include <asm/dma-iommu.h> - -#include <drm/drm.h> -#include <drm/drmP.h> - -#include "host1x_client.h" -#include "dev.h" -#include "drm.h" -#include "gem.h" -#include "syncpt.h" - -#define DRIVER_NAME "tegra" -#define DRIVER_DESC "NVIDIA Tegra graphics" -#define DRIVER_DATE "20120330" -#define DRIVER_MAJOR 0 -#define DRIVER_MINOR 0 -#define DRIVER_PATCHLEVEL 0 - -struct host1x_drm_client { - struct host1x_client *client; - struct device_node *np; - struct list_head list; -}; - -static int host1x_add_drm_client(struct host1x_drm *host1x, - struct device_node *np) -{ - struct host1x_drm_client *client; - - client = kzalloc(sizeof(*client), GFP_KERNEL); - if (!client) - return -ENOMEM; - - INIT_LIST_HEAD(&client->list); - client->np = of_node_get(np); - - list_add_tail(&client->list, &host1x->drm_clients); - - return 0; -} - -static int host1x_activate_drm_client(struct host1x_drm *host1x, - struct host1x_drm_client *drm, - struct host1x_client *client) -{ - mutex_lock(&host1x->drm_clients_lock); - list_del_init(&drm->list); - list_add_tail(&drm->list, &host1x->drm_active); - drm->client = client; - mutex_unlock(&host1x->drm_clients_lock); - - return 0; -} - -static int host1x_remove_drm_client(struct host1x_drm *host1x, - struct host1x_drm_client *client) -{ - mutex_lock(&host1x->drm_clients_lock); - list_del_init(&client->list); - mutex_unlock(&host1x->drm_clients_lock); - - of_node_put(client->np); - kfree(client); - - return 0; -} - -static int host1x_parse_dt(struct host1x_drm *host1x) -{ - static const char * const compat[] = { - "nvidia,tegra20-dc", - "nvidia,tegra20-hdmi", - "nvidia,tegra20-gr2d", - "nvidia,tegra30-dc", - "nvidia,tegra30-hdmi", - "nvidia,tegra30-gr2d", - }; - unsigned int i; - int err; - - for (i = 0; i < ARRAY_SIZE(compat); i++) { - struct device_node *np; - - for_each_child_of_node(host1x->dev->of_node, np) { - if (of_device_is_compatible(np, compat[i]) && - of_device_is_available(np)) { - err = host1x_add_drm_client(host1x, np); - if (err < 0) - return err; - } - } - } - - return 0; -} - -int host1x_drm_alloc(struct platform_device *pdev) -{ - struct host1x_drm *host1x; - int err; - - host1x = devm_kzalloc(&pdev->dev, sizeof(*host1x), GFP_KERNEL); - if (!host1x) - return -ENOMEM; - - mutex_init(&host1x->drm_clients_lock); - INIT_LIST_HEAD(&host1x->drm_clients); - INIT_LIST_HEAD(&host1x->drm_active); - mutex_init(&host1x->clients_lock); - INIT_LIST_HEAD(&host1x->clients); - host1x->dev = &pdev->dev; - - err = host1x_parse_dt(host1x); - if (err < 0) { - dev_err(&pdev->dev, "failed to parse DT: %d\n", err); - return err; - } - - host1x_set_drm_data(&pdev->dev, host1x); - - return 0; -} - -int host1x_drm_init(struct host1x_drm *host1x, struct drm_device *drm) -{ - struct host1x_client *client; - - mutex_lock(&host1x->clients_lock); - - list_for_each_entry(client, &host1x->clients, list) { - if (client->ops && client->ops->drm_init) { - int err = client->ops->drm_init(client, drm); - if (err < 0) { - dev_err(host1x->dev, - "DRM setup failed for %s: %d\n", - dev_name(client->dev), err); - mutex_unlock(&host1x->clients_lock); - return err; - } - } - } - - mutex_unlock(&host1x->clients_lock); - - return 0; -} - -int host1x_drm_exit(struct host1x_drm *host1x) -{ - struct platform_device *pdev = to_platform_device(host1x->dev); - struct host1x_client *client; - - if (!host1x->drm) - return 0; - - mutex_lock(&host1x->clients_lock); - - list_for_each_entry_reverse(client, &host1x->clients, list) { - if (client->ops && client->ops->drm_exit) { - int err = client->ops->drm_exit(client); - if (err < 0) { - dev_err(host1x->dev, - "DRM cleanup failed for %s: %d\n", - dev_name(client->dev), err); - mutex_unlock(&host1x->clients_lock); - return err; - } - } - } - - mutex_unlock(&host1x->clients_lock); - - drm_platform_exit(&tegra_drm_driver, pdev); - host1x->drm = NULL; - - return 0; -} - -int host1x_register_client(struct host1x_drm *host1x, - struct host1x_client *client) -{ - struct host1x_drm_client *drm, *tmp; - int err; - - mutex_lock(&host1x->clients_lock); - list_add_tail(&client->list, &host1x->clients); - mutex_unlock(&host1x->clients_lock); - - list_for_each_entry_safe(drm, tmp, &host1x->drm_clients, list) - if (drm->np == client->dev->of_node) - host1x_activate_drm_client(host1x, drm, client); - - if (list_empty(&host1x->drm_clients)) { - struct platform_device *pdev = to_platform_device(host1x->dev); - - err = drm_platform_init(&tegra_drm_driver, pdev); - if (err < 0) { - dev_err(host1x->dev, "drm_platform_init(): %d\n", err); - return err; - } - } - - return 0; -} - -int host1x_unregister_client(struct host1x_drm *host1x, - struct host1x_client *client) -{ - struct host1x_drm_client *drm, *tmp; - int err; - - list_for_each_entry_safe(drm, tmp, &host1x->drm_active, list) { - if (drm->client == client) { - err = host1x_drm_exit(host1x); - if (err < 0) { - dev_err(host1x->dev, "host1x_drm_exit(): %d\n", - err); - return err; - } - - host1x_remove_drm_client(host1x, drm); - break; - } - } - - mutex_lock(&host1x->clients_lock); - list_del_init(&client->list); - mutex_unlock(&host1x->clients_lock); - - return 0; -} - -static int tegra_drm_load(struct drm_device *drm, unsigned long flags) -{ - struct host1x_drm *host1x; - int err; - - host1x = host1x_get_drm_data(drm->dev); - drm->dev_private = host1x; - host1x->drm = drm; - - drm_mode_config_init(drm); - - err = host1x_drm_init(host1x, drm); - if (err < 0) - return err; - - /* - * We don't use the drm_irq_install() helpers provided by the DRM - * core, so we need to set this manually in order to allow the - * DRM_IOCTL_WAIT_VBLANK to operate correctly. - */ - drm->irq_enabled = 1; - - err = drm_vblank_init(drm, drm->mode_config.num_crtc); - if (err < 0) - return err; - - err = tegra_drm_fb_init(drm); - if (err < 0) - return err; - - drm_kms_helper_poll_init(drm); - - return 0; -} - -static int tegra_drm_unload(struct drm_device *drm) -{ - drm_kms_helper_poll_fini(drm); - tegra_drm_fb_exit(drm); - - drm_mode_config_cleanup(drm); - - return 0; -} - -static int tegra_drm_open(struct drm_device *drm, struct drm_file *filp) -{ - struct host1x_drm_file *fpriv; - - fpriv = kzalloc(sizeof(*fpriv), GFP_KERNEL); - if (!fpriv) - return -ENOMEM; - - INIT_LIST_HEAD(&fpriv->contexts); - filp->driver_priv = fpriv; - - return 0; -} - -static void host1x_drm_context_free(struct host1x_drm_context *context) -{ - context->client->ops->close_channel(context); - kfree(context); -} - -static void tegra_drm_lastclose(struct drm_device *drm) -{ - struct host1x_drm *host1x = drm->dev_private; - - tegra_fbdev_restore_mode(host1x->fbdev); -} - -#ifdef CONFIG_DRM_TEGRA_STAGING -static bool host1x_drm_file_owns_context(struct host1x_drm_file *file, - struct host1x_drm_context *context) -{ - struct host1x_drm_context *ctx; - - list_for_each_entry(ctx, &file->contexts, list) - if (ctx == context) - return true; - - return false; -} - -static int tegra_gem_create(struct drm_device *drm, void *data, - struct drm_file *file) -{ - struct drm_tegra_gem_create *args = data; - struct tegra_bo *bo; - - bo = tegra_bo_create_with_handle(file, drm, args->size, - &args->handle); - if (IS_ERR(bo)) - return PTR_ERR(bo); - - return 0; -} - -static int tegra_gem_mmap(struct drm_device *drm, void *data, - struct drm_file *file) -{ - struct drm_tegra_gem_mmap *args = data; - struct drm_gem_object *gem; - struct tegra_bo *bo; - - gem = drm_gem_object_lookup(drm, file, args->handle); - if (!gem) - return -EINVAL; - - bo = to_tegra_bo(gem); - - args->offset = tegra_bo_get_mmap_offset(bo); - - drm_gem_object_unreference(gem); - - return 0; -} - -static int tegra_syncpt_read(struct drm_device *drm, void *data, - struct drm_file *file) -{ - struct drm_tegra_syncpt_read *args = data; - struct host1x *host = dev_get_drvdata(drm->dev); - struct host1x_syncpt *sp = host1x_syncpt_get(host, args->id); - - if (!sp) - return -EINVAL; - - args->value = host1x_syncpt_read_min(sp); - return 0; -} - -static int tegra_syncpt_incr(struct drm_device *drm, void *data, - struct drm_file *file) -{ - struct drm_tegra_syncpt_incr *args = data; - struct host1x *host = dev_get_drvdata(drm->dev); - struct host1x_syncpt *sp = host1x_syncpt_get(host, args->id); - - if (!sp) - return -EINVAL; - - return host1x_syncpt_incr(sp); -} - -static int tegra_syncpt_wait(struct drm_device *drm, void *data, - struct drm_file *file) -{ - struct drm_tegra_syncpt_wait *args = data; - struct host1x *host = dev_get_drvdata(drm->dev); - struct host1x_syncpt *sp = host1x_syncpt_get(host, args->id); - - if (!sp) - return -EINVAL; - - return host1x_syncpt_wait(sp, args->thresh, args->timeout, - &args->value); -} - -static int tegra_open_channel(struct drm_device *drm, void *data, - struct drm_file *file) -{ - struct drm_tegra_open_channel *args = data; - struct host1x_client *client; - struct host1x_drm_context *context; - struct host1x_drm_file *fpriv = file->driver_priv; - struct host1x_drm *host1x = drm->dev_private; - int err = -ENODEV; - - context = kzalloc(sizeof(*context), GFP_KERNEL); - if (!context) - return -ENOMEM; - - list_for_each_entry(client, &host1x->clients, list) - if (client->class == args->client) { - err = client->ops->open_channel(client, context); - if (err) - break; - - context->client = client; - list_add(&context->list, &fpriv->contexts); - args->context = (uintptr_t)context; - return 0; - } - - kfree(context); - return err; -} - -static int tegra_close_channel(struct drm_device *drm, void *data, - struct drm_file *file) -{ - struct drm_tegra_close_channel *args = data; - struct host1x_drm_file *fpriv = file->driver_priv; - struct host1x_drm_context *context = - (struct host1x_drm_context *)(uintptr_t)args->context; - - if (!host1x_drm_file_owns_context(fpriv, context)) - return -EINVAL; - - list_del(&context->list); - host1x_drm_context_free(context); - - return 0; -} - -static int tegra_get_syncpt(struct drm_device *drm, void *data, - struct drm_file *file) -{ - struct drm_tegra_get_syncpt *args = data; - struct host1x_drm_file *fpriv = file->driver_priv; - struct host1x_drm_context *context = - (struct host1x_drm_context *)(uintptr_t)args->context; - struct host1x_syncpt *syncpt; - - if (!host1x_drm_file_owns_context(fpriv, context)) - return -ENODEV; - - if (args->index >= context->client->num_syncpts) - return -EINVAL; - - syncpt = context->client->syncpts[args->index]; - args->id = host1x_syncpt_id(syncpt); - - return 0; -} - -static int tegra_submit(struct drm_device *drm, void *data, - struct drm_file *file) -{ - struct drm_tegra_submit *args = data; - struct host1x_drm_file *fpriv = file->driver_priv; - struct host1x_drm_context *context = - (struct host1x_drm_context *)(uintptr_t)args->context; - - if (!host1x_drm_file_owns_context(fpriv, context)) - return -ENODEV; - - return context->client->ops->submit(context, args, drm, file); -} -#endif - -static struct drm_ioctl_desc tegra_drm_ioctls[] = { -#ifdef CONFIG_DRM_TEGRA_STAGING - DRM_IOCTL_DEF_DRV(TEGRA_GEM_CREATE, tegra_gem_create, DRM_UNLOCKED | DRM_AUTH), - DRM_IOCTL_DEF_DRV(TEGRA_GEM_MMAP, tegra_gem_mmap, DRM_UNLOCKED), - DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_READ, tegra_syncpt_read, DRM_UNLOCKED), - DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_INCR, tegra_syncpt_incr, DRM_UNLOCKED), - DRM_IOCTL_DEF_DRV(TEGRA_SYNCPT_WAIT, tegra_syncpt_wait, DRM_UNLOCKED), - DRM_IOCTL_DEF_DRV(TEGRA_OPEN_CHANNEL, tegra_open_channel, DRM_UNLOCKED), - DRM_IOCTL_DEF_DRV(TEGRA_CLOSE_CHANNEL, tegra_close_channel, DRM_UNLOCKED), - DRM_IOCTL_DEF_DRV(TEGRA_GET_SYNCPT, tegra_get_syncpt, DRM_UNLOCKED), - DRM_IOCTL_DEF_DRV(TEGRA_SUBMIT, tegra_submit, DRM_UNLOCKED), -#endif -}; - -static const struct file_operations tegra_drm_fops = { - .owner = THIS_MODULE, - .open = drm_open, - .release = drm_release, - .unlocked_ioctl = drm_ioctl, - .mmap = tegra_drm_mmap, - .poll = drm_poll, - .fasync = drm_fasync, - .read = drm_read, -#ifdef CONFIG_COMPAT - .compat_ioctl = drm_compat_ioctl, -#endif - .llseek = noop_llseek, -}; - -static struct drm_crtc *tegra_crtc_from_pipe(struct drm_device *drm, int pipe) -{ - struct drm_crtc *crtc; - - list_for_each_entry(crtc, &drm->mode_config.crtc_list, head) { - struct tegra_dc *dc = to_tegra_dc(crtc); - - if (dc->pipe == pipe) - return crtc; - } - - return NULL; -} - -static u32 tegra_drm_get_vblank_counter(struct drm_device *dev, int crtc) -{ - /* TODO: implement real hardware counter using syncpoints */ - return drm_vblank_count(dev, crtc); -} - -static int tegra_drm_enable_vblank(struct drm_device *drm, int pipe) -{ - struct drm_crtc *crtc = tegra_crtc_from_pipe(drm, pipe); - struct tegra_dc *dc = to_tegra_dc(crtc); - - if (!crtc) - return -ENODEV; - - tegra_dc_enable_vblank(dc); - - return 0; -} - -static void tegra_drm_disable_vblank(struct drm_device *drm, int pipe) -{ - struct drm_crtc *crtc = tegra_crtc_from_pipe(drm, pipe); - struct tegra_dc *dc = to_tegra_dc(crtc); - - if (crtc) - tegra_dc_disable_vblank(dc); -} - -static void tegra_drm_preclose(struct drm_device *drm, struct drm_file *file) -{ - struct host1x_drm_file *fpriv = file->driver_priv; - struct host1x_drm_context *context, *tmp; - struct drm_crtc *crtc; - - list_for_each_entry(crtc, &drm->mode_config.crtc_list, head) - tegra_dc_cancel_page_flip(crtc, file); - - list_for_each_entry_safe(context, tmp, &fpriv->contexts, list) - host1x_drm_context_free(context); - - kfree(fpriv); -} - -#ifdef CONFIG_DEBUG_FS -static int tegra_debugfs_framebuffers(struct seq_file *s, void *data) -{ - struct drm_info_node *node = (struct drm_info_node *)s->private; - struct drm_device *drm = node->minor->dev; - struct drm_framebuffer *fb; - - mutex_lock(&drm->mode_config.fb_lock); - - list_for_each_entry(fb, &drm->mode_config.fb_list, head) { - seq_printf(s, "%3d: user size: %d x %d, depth %d, %d bpp, refcount %d\n", - fb->base.id, fb->width, fb->height, fb->depth, - fb->bits_per_pixel, - atomic_read(&fb->refcount.refcount)); - } - - mutex_unlock(&drm->mode_config.fb_lock); - - return 0; -} - -static struct drm_info_list tegra_debugfs_list[] = { - { "framebuffers", tegra_debugfs_framebuffers, 0 }, -}; - -static int tegra_debugfs_init(struct drm_minor *minor) -{ - return drm_debugfs_create_files(tegra_debugfs_list, - ARRAY_SIZE(tegra_debugfs_list), - minor->debugfs_root, minor); -} - -static void tegra_debugfs_cleanup(struct drm_minor *minor) -{ - drm_debugfs_remove_files(tegra_debugfs_list, - ARRAY_SIZE(tegra_debugfs_list), minor); -} -#endif - -struct drm_driver tegra_drm_driver = { - .driver_features = DRIVER_MODESET | DRIVER_GEM, - .load = tegra_drm_load, - .unload = tegra_drm_unload, - .open = tegra_drm_open, - .preclose = tegra_drm_preclose, - .lastclose = tegra_drm_lastclose, - - .get_vblank_counter = tegra_drm_get_vblank_counter, - .enable_vblank = tegra_drm_enable_vblank, - .disable_vblank = tegra_drm_disable_vblank, - -#if defined(CONFIG_DEBUG_FS) - .debugfs_init = tegra_debugfs_init, - .debugfs_cleanup = tegra_debugfs_cleanup, -#endif - - .gem_free_object = tegra_bo_free_object, - .gem_vm_ops = &tegra_bo_vm_ops, - .dumb_create = tegra_bo_dumb_create, - .dumb_map_offset = tegra_bo_dumb_map_offset, - .dumb_destroy = tegra_bo_dumb_destroy, - - .ioctls = tegra_drm_ioctls, - .num_ioctls = ARRAY_SIZE(tegra_drm_ioctls), - .fops = &tegra_drm_fops, - - .name = DRIVER_NAME, - .desc = DRIVER_DESC, - .date = DRIVER_DATE, - .major = DRIVER_MAJOR, - .minor = DRIVER_MINOR, - .patchlevel = DRIVER_PATCHLEVEL, -}; diff --git a/drivers/gpu/host1x/drm/drm.h b/drivers/gpu/host1x/drm/drm.h deleted file mode 100644 index 02ce020f2575..000000000000 --- a/drivers/gpu/host1x/drm/drm.h +++ /dev/null @@ -1,271 +0,0 @@ -/* - * Copyright (C) 2012 Avionic Design GmbH - * Copyright (C) 2012-2013 NVIDIA CORPORATION. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#ifndef HOST1X_DRM_H -#define HOST1X_DRM_H 1 - -#include <drm/drmP.h> -#include <drm/drm_crtc_helper.h> -#include <drm/drm_edid.h> -#include <drm/drm_fb_helper.h> -#include <drm/drm_fixed.h> -#include <uapi/drm/tegra_drm.h> - -#include "host1x.h" - -struct tegra_fb { - struct drm_framebuffer base; - struct tegra_bo **planes; - unsigned int num_planes; -}; - -struct tegra_fbdev { - struct drm_fb_helper base; - struct tegra_fb *fb; -}; - -struct host1x_drm { - struct drm_device *drm; - struct device *dev; - void __iomem *regs; - struct clk *clk; - int syncpt; - int irq; - - struct mutex drm_clients_lock; - struct list_head drm_clients; - struct list_head drm_active; - - struct mutex clients_lock; - struct list_head clients; - - struct tegra_fbdev *fbdev; -}; - -struct host1x_client; - -struct host1x_drm_context { - struct host1x_client *client; - struct host1x_channel *channel; - struct list_head list; -}; - -struct host1x_client_ops { - int (*drm_init)(struct host1x_client *client, struct drm_device *drm); - int (*drm_exit)(struct host1x_client *client); - int (*open_channel)(struct host1x_client *client, - struct host1x_drm_context *context); - void (*close_channel)(struct host1x_drm_context *context); - int (*submit)(struct host1x_drm_context *context, - struct drm_tegra_submit *args, struct drm_device *drm, - struct drm_file *file); -}; - -struct host1x_drm_file { - struct list_head contexts; -}; - -struct host1x_client { - struct host1x_drm *host1x; - struct device *dev; - - const struct host1x_client_ops *ops; - - enum host1x_class class; - struct host1x_channel *channel; - - struct host1x_syncpt **syncpts; - unsigned int num_syncpts; - - struct list_head list; -}; - -extern int host1x_drm_init(struct host1x_drm *host1x, struct drm_device *drm); -extern int host1x_drm_exit(struct host1x_drm *host1x); - -extern int host1x_register_client(struct host1x_drm *host1x, - struct host1x_client *client); -extern int host1x_unregister_client(struct host1x_drm *host1x, - struct host1x_client *client); - -struct tegra_output; - -struct tegra_dc { - struct host1x_client client; - spinlock_t lock; - - struct host1x_drm *host1x; - struct device *dev; - - struct drm_crtc base; - int pipe; - - struct clk *clk; - - void __iomem *regs; - int irq; - - struct tegra_output *rgb; - - struct list_head list; - - struct drm_info_list *debugfs_files; - struct drm_minor *minor; - struct dentry *debugfs; - - /* page-flip handling */ - struct drm_pending_vblank_event *event; -}; - -static inline struct tegra_dc *host1x_client_to_dc(struct host1x_client *client) -{ - return container_of(client, struct tegra_dc, client); -} - -static inline struct tegra_dc *to_tegra_dc(struct drm_crtc *crtc) -{ - return container_of(crtc, struct tegra_dc, base); -} - -static inline void tegra_dc_writel(struct tegra_dc *dc, unsigned long value, - unsigned long reg) -{ - writel(value, dc->regs + (reg << 2)); -} - -static inline unsigned long tegra_dc_readl(struct tegra_dc *dc, - unsigned long reg) -{ - return readl(dc->regs + (reg << 2)); -} - -struct tegra_dc_window { - struct { - unsigned int x; - unsigned int y; - unsigned int w; - unsigned int h; - } src; - struct { - unsigned int x; - unsigned int y; - unsigned int w; - unsigned int h; - } dst; - unsigned int bits_per_pixel; - unsigned int format; - unsigned int stride[2]; - unsigned long base[3]; -}; - -/* from dc.c */ -extern unsigned int tegra_dc_format(uint32_t format); -extern int tegra_dc_setup_window(struct tegra_dc *dc, unsigned int index, - const struct tegra_dc_window *window); -extern void tegra_dc_enable_vblank(struct tegra_dc *dc); -extern void tegra_dc_disable_vblank(struct tegra_dc *dc); -extern void tegra_dc_cancel_page_flip(struct drm_crtc *crtc, - struct drm_file *file); - -struct tegra_output_ops { - int (*enable)(struct tegra_output *output); - int (*disable)(struct tegra_output *output); - int (*setup_clock)(struct tegra_output *output, struct clk *clk, - unsigned long pclk); - int (*check_mode)(struct tegra_output *output, - struct drm_display_mode *mode, - enum drm_mode_status *status); -}; - -enum tegra_output_type { - TEGRA_OUTPUT_RGB, - TEGRA_OUTPUT_HDMI, -}; - -struct tegra_output { - struct device_node *of_node; - struct device *dev; - - const struct tegra_output_ops *ops; - enum tegra_output_type type; - - struct i2c_adapter *ddc; - const struct edid *edid; - unsigned int hpd_irq; - int hpd_gpio; - - struct drm_encoder encoder; - struct drm_connector connector; -}; - -static inline struct tegra_output *encoder_to_output(struct drm_encoder *e) -{ - return container_of(e, struct tegra_output, encoder); -} - -static inline struct tegra_output *connector_to_output(struct drm_connector *c) -{ - return container_of(c, struct tegra_output, connector); -} - -static inline int tegra_output_enable(struct tegra_output *output) -{ - if (output && output->ops && output->ops->enable) - return output->ops->enable(output); - - return output ? -ENOSYS : -EINVAL; -} - -static inline int tegra_output_disable(struct tegra_output *output) -{ - if (output && output->ops && output->ops->disable) - return output->ops->disable(output); - - return output ? -ENOSYS : -EINVAL; -} - -static inline int tegra_output_setup_clock(struct tegra_output *output, - struct clk *clk, unsigned long pclk) -{ - if (output && output->ops && output->ops->setup_clock) - return output->ops->setup_clock(output, clk, pclk); - - return output ? -ENOSYS : -EINVAL; -} - -static inline int tegra_output_check_mode(struct tegra_output *output, - struct drm_display_mode *mode, - enum drm_mode_status *status) -{ - if (output && output->ops && output->ops->check_mode) - return output->ops->check_mode(output, mode, status); - - return output ? -ENOSYS : -EINVAL; -} - -/* from rgb.c */ -extern int tegra_dc_rgb_probe(struct tegra_dc *dc); -extern int tegra_dc_rgb_init(struct drm_device *drm, struct tegra_dc *dc); -extern int tegra_dc_rgb_exit(struct tegra_dc *dc); - -/* from output.c */ -extern int tegra_output_parse_dt(struct tegra_output *output); -extern int tegra_output_init(struct drm_device *drm, struct tegra_output *output); -extern int tegra_output_exit(struct tegra_output *output); - -/* from fb.c */ -struct tegra_bo *tegra_fb_get_plane(struct drm_framebuffer *framebuffer, - unsigned int index); -extern int tegra_drm_fb_init(struct drm_device *drm); -extern void tegra_drm_fb_exit(struct drm_device *drm); -extern void tegra_fbdev_restore_mode(struct tegra_fbdev *fbdev); - -extern struct drm_driver tegra_drm_driver; - -#endif /* HOST1X_DRM_H */ diff --git a/drivers/gpu/host1x/drm/fb.c b/drivers/gpu/host1x/drm/fb.c deleted file mode 100644 index 979a3e32b78b..000000000000 --- a/drivers/gpu/host1x/drm/fb.c +++ /dev/null @@ -1,374 +0,0 @@ -/* - * Copyright (C) 2012-2013 Avionic Design GmbH - * Copyright (C) 2012 NVIDIA CORPORATION. All rights reserved. - * - * Based on the KMS/FB CMA helpers - * Copyright (C) 2012 Analog Device Inc. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#include <linux/module.h> - -#include "drm.h" -#include "gem.h" - -static inline struct tegra_fb *to_tegra_fb(struct drm_framebuffer *fb) -{ - return container_of(fb, struct tegra_fb, base); -} - -static inline struct tegra_fbdev *to_tegra_fbdev(struct drm_fb_helper *helper) -{ - return container_of(helper, struct tegra_fbdev, base); -} - -struct tegra_bo *tegra_fb_get_plane(struct drm_framebuffer *framebuffer, - unsigned int index) -{ - struct tegra_fb *fb = to_tegra_fb(framebuffer); - - if (index >= drm_format_num_planes(framebuffer->pixel_format)) - return NULL; - - return fb->planes[index]; -} - -static void tegra_fb_destroy(struct drm_framebuffer *framebuffer) -{ - struct tegra_fb *fb = to_tegra_fb(framebuffer); - unsigned int i; - - for (i = 0; i < fb->num_planes; i++) { - struct tegra_bo *bo = fb->planes[i]; - - if (bo) - drm_gem_object_unreference_unlocked(&bo->gem); - } - - drm_framebuffer_cleanup(framebuffer); - kfree(fb->planes); - kfree(fb); -} - -static int tegra_fb_create_handle(struct drm_framebuffer *framebuffer, - struct drm_file *file, unsigned int *handle) -{ - struct tegra_fb *fb = to_tegra_fb(framebuffer); - - return drm_gem_handle_create(file, &fb->planes[0]->gem, handle); -} - -static struct drm_framebuffer_funcs tegra_fb_funcs = { - .destroy = tegra_fb_destroy, - .create_handle = tegra_fb_create_handle, -}; - -static struct tegra_fb *tegra_fb_alloc(struct drm_device *drm, - struct drm_mode_fb_cmd2 *mode_cmd, - struct tegra_bo **planes, - unsigned int num_planes) -{ - struct tegra_fb *fb; - unsigned int i; - int err; - - fb = kzalloc(sizeof(*fb), GFP_KERNEL); - if (!fb) - return ERR_PTR(-ENOMEM); - - fb->planes = kzalloc(num_planes * sizeof(*planes), GFP_KERNEL); - if (!fb->planes) - return ERR_PTR(-ENOMEM); - - fb->num_planes = num_planes; - - drm_helper_mode_fill_fb_struct(&fb->base, mode_cmd); - - for (i = 0; i < fb->num_planes; i++) - fb->planes[i] = planes[i]; - - err = drm_framebuffer_init(drm, &fb->base, &tegra_fb_funcs); - if (err < 0) { - dev_err(drm->dev, "failed to initialize framebuffer: %d\n", - err); - kfree(fb->planes); - kfree(fb); - return ERR_PTR(err); - } - - return fb; -} - -static struct drm_framebuffer *tegra_fb_create(struct drm_device *drm, - struct drm_file *file, - struct drm_mode_fb_cmd2 *cmd) -{ - unsigned int hsub, vsub, i; - struct tegra_bo *planes[4]; - struct drm_gem_object *gem; - struct tegra_fb *fb; - int err; - - hsub = drm_format_horz_chroma_subsampling(cmd->pixel_format); - vsub = drm_format_vert_chroma_subsampling(cmd->pixel_format); - - for (i = 0; i < drm_format_num_planes(cmd->pixel_format); i++) { - unsigned int width = cmd->width / (i ? hsub : 1); - unsigned int height = cmd->height / (i ? vsub : 1); - unsigned int size, bpp; - - gem = drm_gem_object_lookup(drm, file, cmd->handles[i]); - if (!gem) { - err = -ENXIO; - goto unreference; - } - - bpp = drm_format_plane_cpp(cmd->pixel_format, i); - - size = (height - 1) * cmd->pitches[i] + - width * bpp + cmd->offsets[i]; - - if (gem->size < size) { - err = -EINVAL; - goto unreference; - } - - planes[i] = to_tegra_bo(gem); - } - - fb = tegra_fb_alloc(drm, cmd, planes, i); - if (IS_ERR(fb)) { - err = PTR_ERR(fb); - goto unreference; - } - - return &fb->base; - -unreference: - while (i--) - drm_gem_object_unreference_unlocked(&planes[i]->gem); - - return ERR_PTR(err); -} - -static struct fb_ops tegra_fb_ops = { - .owner = THIS_MODULE, - .fb_fillrect = sys_fillrect, - .fb_copyarea = sys_copyarea, - .fb_imageblit = sys_imageblit, - .fb_check_var = drm_fb_helper_check_var, - .fb_set_par = drm_fb_helper_set_par, - .fb_blank = drm_fb_helper_blank, - .fb_pan_display = drm_fb_helper_pan_display, - .fb_setcmap = drm_fb_helper_setcmap, -}; - -static int tegra_fbdev_probe(struct drm_fb_helper *helper, - struct drm_fb_helper_surface_size *sizes) -{ - struct tegra_fbdev *fbdev = to_tegra_fbdev(helper); - struct drm_device *drm = helper->dev; - struct drm_mode_fb_cmd2 cmd = { 0 }; - unsigned int bytes_per_pixel; - struct drm_framebuffer *fb; - unsigned long offset; - struct fb_info *info; - struct tegra_bo *bo; - size_t size; - int err; - - bytes_per_pixel = DIV_ROUND_UP(sizes->surface_bpp, 8); - - cmd.width = sizes->surface_width; - cmd.height = sizes->surface_height; - cmd.pitches[0] = sizes->surface_width * bytes_per_pixel; - cmd.pixel_format = drm_mode_legacy_fb_format(sizes->surface_bpp, - sizes->surface_depth); - - size = cmd.pitches[0] * cmd.height; - - bo = tegra_bo_create(drm, size); - if (IS_ERR(bo)) - return PTR_ERR(bo); - - info = framebuffer_alloc(0, drm->dev); - if (!info) { - dev_err(drm->dev, "failed to allocate framebuffer info\n"); - tegra_bo_free_object(&bo->gem); - return -ENOMEM; - } - - fbdev->fb = tegra_fb_alloc(drm, &cmd, &bo, 1); - if (IS_ERR(fbdev->fb)) { - dev_err(drm->dev, "failed to allocate DRM framebuffer\n"); - err = PTR_ERR(fbdev->fb); - goto release; - } - - fb = &fbdev->fb->base; - helper->fb = fb; - helper->fbdev = info; - - info->par = helper; - info->flags = FBINFO_FLAG_DEFAULT; - info->fbops = &tegra_fb_ops; - - err = fb_alloc_cmap(&info->cmap, 256, 0); - if (err < 0) { - dev_err(drm->dev, "failed to allocate color map: %d\n", err); - goto destroy; - } - - drm_fb_helper_fill_fix(info, fb->pitches[0], fb->depth); - drm_fb_helper_fill_var(info, helper, fb->width, fb->height); - - offset = info->var.xoffset * bytes_per_pixel + - info->var.yoffset * fb->pitches[0]; - - drm->mode_config.fb_base = (resource_size_t)bo->paddr; - info->screen_base = bo->vaddr + offset; - info->screen_size = size; - info->fix.smem_start = (unsigned long)(bo->paddr + offset); - info->fix.smem_len = size; - - return 0; - -destroy: - drm_framebuffer_unregister_private(fb); - tegra_fb_destroy(fb); -release: - framebuffer_release(info); - return err; -} - -static struct drm_fb_helper_funcs tegra_fb_helper_funcs = { - .fb_probe = tegra_fbdev_probe, -}; - -static struct tegra_fbdev *tegra_fbdev_create(struct drm_device *drm, - unsigned int preferred_bpp, - unsigned int num_crtc, - unsigned int max_connectors) -{ - struct drm_fb_helper *helper; - struct tegra_fbdev *fbdev; - int err; - - fbdev = kzalloc(sizeof(*fbdev), GFP_KERNEL); - if (!fbdev) { - dev_err(drm->dev, "failed to allocate DRM fbdev\n"); - return ERR_PTR(-ENOMEM); - } - - fbdev->base.funcs = &tegra_fb_helper_funcs; - helper = &fbdev->base; - - err = drm_fb_helper_init(drm, &fbdev->base, num_crtc, max_connectors); - if (err < 0) { - dev_err(drm->dev, "failed to initialize DRM FB helper\n"); - goto free; - } - - err = drm_fb_helper_single_add_all_connectors(&fbdev->base); - if (err < 0) { - dev_err(drm->dev, "failed to add connectors\n"); - goto fini; - } - - drm_helper_disable_unused_functions(drm); - - err = drm_fb_helper_initial_config(&fbdev->base, preferred_bpp); - if (err < 0) { - dev_err(drm->dev, "failed to set initial configuration\n"); - goto fini; - } - - return fbdev; - -fini: - drm_fb_helper_fini(&fbdev->base); -free: - kfree(fbdev); - return ERR_PTR(err); -} - -static void tegra_fbdev_free(struct tegra_fbdev *fbdev) -{ - struct fb_info *info = fbdev->base.fbdev; - - if (info) { - int err; - - err = unregister_framebuffer(info); - if (err < 0) - DRM_DEBUG_KMS("failed to unregister framebuffer\n"); - - if (info->cmap.len) - fb_dealloc_cmap(&info->cmap); - - framebuffer_release(info); - } - - if (fbdev->fb) { - drm_framebuffer_unregister_private(&fbdev->fb->base); - tegra_fb_destroy(&fbdev->fb->base); - } - - drm_fb_helper_fini(&fbdev->base); - kfree(fbdev); -} - -static void tegra_fb_output_poll_changed(struct drm_device *drm) -{ - struct host1x_drm *host1x = drm->dev_private; - - if (host1x->fbdev) - drm_fb_helper_hotplug_event(&host1x->fbdev->base); -} - -static const struct drm_mode_config_funcs tegra_drm_mode_funcs = { - .fb_create = tegra_fb_create, - .output_poll_changed = tegra_fb_output_poll_changed, -}; - -int tegra_drm_fb_init(struct drm_device *drm) -{ - struct host1x_drm *host1x = drm->dev_private; - struct tegra_fbdev *fbdev; - - drm->mode_config.min_width = 0; - drm->mode_config.min_height = 0; - - drm->mode_config.max_width = 4096; - drm->mode_config.max_height = 4096; - - drm->mode_config.funcs = &tegra_drm_mode_funcs; - - fbdev = tegra_fbdev_create(drm, 32, drm->mode_config.num_crtc, - drm->mode_config.num_connector); - if (IS_ERR(fbdev)) - return PTR_ERR(fbdev); - - host1x->fbdev = fbdev; - - return 0; -} - -void tegra_drm_fb_exit(struct drm_device *drm) -{ - struct host1x_drm *host1x = drm->dev_private; - - tegra_fbdev_free(host1x->fbdev); -} - -void tegra_fbdev_restore_mode(struct tegra_fbdev *fbdev) -{ - if (fbdev) { - drm_modeset_lock_all(fbdev->base.dev); - drm_fb_helper_restore_fbdev_mode(&fbdev->base); - drm_modeset_unlock_all(fbdev->base.dev); - } -} diff --git a/drivers/gpu/host1x/drm/gem.c b/drivers/gpu/host1x/drm/gem.c deleted file mode 100644 index c5e9a9b494c2..000000000000 --- a/drivers/gpu/host1x/drm/gem.c +++ /dev/null @@ -1,270 +0,0 @@ -/* - * NVIDIA Tegra DRM GEM helper functions - * - * Copyright (C) 2012 Sascha Hauer, Pengutronix - * Copyright (C) 2013 NVIDIA CORPORATION, All rights reserved. - * - * Based on the GEM/CMA helpers - * - * Copyright (c) 2011 Samsung Electronics Co., Ltd. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - */ - -#include <linux/mm.h> -#include <linux/slab.h> -#include <linux/mutex.h> -#include <linux/export.h> -#include <linux/dma-mapping.h> - -#include <drm/drmP.h> -#include <drm/drm.h> - -#include "gem.h" - -static inline struct tegra_bo *host1x_to_drm_bo(struct host1x_bo *bo) -{ - return container_of(bo, struct tegra_bo, base); -} - -static void tegra_bo_put(struct host1x_bo *bo) -{ - struct tegra_bo *obj = host1x_to_drm_bo(bo); - struct drm_device *drm = obj->gem.dev; - - mutex_lock(&drm->struct_mutex); - drm_gem_object_unreference(&obj->gem); - mutex_unlock(&drm->struct_mutex); -} - -static dma_addr_t tegra_bo_pin(struct host1x_bo *bo, struct sg_table **sgt) -{ - struct tegra_bo *obj = host1x_to_drm_bo(bo); - - return obj->paddr; -} - -static void tegra_bo_unpin(struct host1x_bo *bo, struct sg_table *sgt) -{ -} - -static void *tegra_bo_mmap(struct host1x_bo *bo) -{ - struct tegra_bo *obj = host1x_to_drm_bo(bo); - - return obj->vaddr; -} - -static void tegra_bo_munmap(struct host1x_bo *bo, void *addr) -{ -} - -static void *tegra_bo_kmap(struct host1x_bo *bo, unsigned int page) -{ - struct tegra_bo *obj = host1x_to_drm_bo(bo); - - return obj->vaddr + page * PAGE_SIZE; -} - -static void tegra_bo_kunmap(struct host1x_bo *bo, unsigned int page, - void *addr) -{ -} - -static struct host1x_bo *tegra_bo_get(struct host1x_bo *bo) -{ - struct tegra_bo *obj = host1x_to_drm_bo(bo); - struct drm_device *drm = obj->gem.dev; - - mutex_lock(&drm->struct_mutex); - drm_gem_object_reference(&obj->gem); - mutex_unlock(&drm->struct_mutex); - - return bo; -} - -const struct host1x_bo_ops tegra_bo_ops = { - .get = tegra_bo_get, - .put = tegra_bo_put, - .pin = tegra_bo_pin, - .unpin = tegra_bo_unpin, - .mmap = tegra_bo_mmap, - .munmap = tegra_bo_munmap, - .kmap = tegra_bo_kmap, - .kunmap = tegra_bo_kunmap, -}; - -static void tegra_bo_destroy(struct drm_device *drm, struct tegra_bo *bo) -{ - dma_free_writecombine(drm->dev, bo->gem.size, bo->vaddr, bo->paddr); -} - -unsigned int tegra_bo_get_mmap_offset(struct tegra_bo *bo) -{ - return (unsigned int)bo->gem.map_list.hash.key << PAGE_SHIFT; -} - -struct tegra_bo *tegra_bo_create(struct drm_device *drm, unsigned int size) -{ - struct tegra_bo *bo; - int err; - - bo = kzalloc(sizeof(*bo), GFP_KERNEL); - if (!bo) - return ERR_PTR(-ENOMEM); - - host1x_bo_init(&bo->base, &tegra_bo_ops); - size = round_up(size, PAGE_SIZE); - - bo->vaddr = dma_alloc_writecombine(drm->dev, size, &bo->paddr, - GFP_KERNEL | __GFP_NOWARN); - if (!bo->vaddr) { - dev_err(drm->dev, "failed to allocate buffer with size %u\n", - size); - err = -ENOMEM; - goto err_dma; - } - - err = drm_gem_object_init(drm, &bo->gem, size); - if (err) - goto err_init; - - err = drm_gem_create_mmap_offset(&bo->gem); - if (err) - goto err_mmap; - - return bo; - -err_mmap: - drm_gem_object_release(&bo->gem); -err_init: - tegra_bo_destroy(drm, bo); -err_dma: - kfree(bo); - - return ERR_PTR(err); - -} - -struct tegra_bo *tegra_bo_create_with_handle(struct drm_file *file, - struct drm_device *drm, - unsigned int size, - unsigned int *handle) -{ - struct tegra_bo *bo; - int ret; - - bo = tegra_bo_create(drm, size); - if (IS_ERR(bo)) - return bo; - - ret = drm_gem_handle_create(file, &bo->gem, handle); - if (ret) - goto err; - - drm_gem_object_unreference_unlocked(&bo->gem); - - return bo; - -err: - tegra_bo_free_object(&bo->gem); - return ERR_PTR(ret); -} - -void tegra_bo_free_object(struct drm_gem_object *gem) -{ - struct tegra_bo *bo = to_tegra_bo(gem); - - if (gem->map_list.map) - drm_gem_free_mmap_offset(gem); - - drm_gem_object_release(gem); - tegra_bo_destroy(gem->dev, bo); - - kfree(bo); -} - -int tegra_bo_dumb_create(struct drm_file *file, struct drm_device *drm, - struct drm_mode_create_dumb *args) -{ - int min_pitch = DIV_ROUND_UP(args->width * args->bpp, 8); - struct tegra_bo *bo; - - if (args->pitch < min_pitch) - args->pitch = min_pitch; - - if (args->size < args->pitch * args->height) - args->size = args->pitch * args->height; - - bo = tegra_bo_create_with_handle(file, drm, args->size, - &args->handle); - if (IS_ERR(bo)) - return PTR_ERR(bo); - - return 0; -} - -int tegra_bo_dumb_map_offset(struct drm_file *file, struct drm_device *drm, - uint32_t handle, uint64_t *offset) -{ - struct drm_gem_object *gem; - struct tegra_bo *bo; - - mutex_lock(&drm->struct_mutex); - - gem = drm_gem_object_lookup(drm, file, handle); - if (!gem) { - dev_err(drm->dev, "failed to lookup GEM object\n"); - mutex_unlock(&drm->struct_mutex); - return -EINVAL; - } - - bo = to_tegra_bo(gem); - - *offset = tegra_bo_get_mmap_offset(bo); - - drm_gem_object_unreference(gem); - - mutex_unlock(&drm->struct_mutex); - - return 0; -} - -const struct vm_operations_struct tegra_bo_vm_ops = { - .open = drm_gem_vm_open, - .close = drm_gem_vm_close, -}; - -int tegra_drm_mmap(struct file *file, struct vm_area_struct *vma) -{ - struct drm_gem_object *gem; - struct tegra_bo *bo; - int ret; - - ret = drm_gem_mmap(file, vma); - if (ret) - return ret; - - gem = vma->vm_private_data; - bo = to_tegra_bo(gem); - - ret = remap_pfn_range(vma, vma->vm_start, bo->paddr >> PAGE_SHIFT, - vma->vm_end - vma->vm_start, vma->vm_page_prot); - if (ret) - drm_gem_vm_close(vma); - - return ret; -} - -int tegra_bo_dumb_destroy(struct drm_file *file, struct drm_device *drm, - unsigned int handle) -{ - return drm_gem_handle_delete(file, handle); -} diff --git a/drivers/gpu/host1x/drm/gem.h b/drivers/gpu/host1x/drm/gem.h deleted file mode 100644 index 34de2b486eb7..000000000000 --- a/drivers/gpu/host1x/drm/gem.h +++ /dev/null @@ -1,59 +0,0 @@ -/* - * Tegra host1x GEM implementation - * - * Copyright (c) 2012-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. - */ - -#ifndef __HOST1X_GEM_H -#define __HOST1X_GEM_H - -#include <drm/drm.h> -#include <drm/drmP.h> - -#include "host1x_bo.h" - -struct tegra_bo { - struct drm_gem_object gem; - struct host1x_bo base; - dma_addr_t paddr; - void *vaddr; -}; - -static inline struct tegra_bo *to_tegra_bo(struct drm_gem_object *gem) -{ - return container_of(gem, struct tegra_bo, gem); -} - -extern const struct host1x_bo_ops tegra_bo_ops; - -struct tegra_bo *tegra_bo_create(struct drm_device *drm, unsigned int size); -struct tegra_bo *tegra_bo_create_with_handle(struct drm_file *file, - struct drm_device *drm, - unsigned int size, - unsigned int *handle); -void tegra_bo_free_object(struct drm_gem_object *gem); -unsigned int tegra_bo_get_mmap_offset(struct tegra_bo *bo); -int tegra_bo_dumb_create(struct drm_file *file, struct drm_device *drm, - struct drm_mode_create_dumb *args); -int tegra_bo_dumb_map_offset(struct drm_file *file, struct drm_device *drm, - uint32_t handle, uint64_t *offset); -int tegra_bo_dumb_destroy(struct drm_file *file, struct drm_device *drm, - unsigned int handle); - -int tegra_drm_mmap(struct file *file, struct vm_area_struct *vma); - -extern const struct vm_operations_struct tegra_bo_vm_ops; - -#endif diff --git a/drivers/gpu/host1x/drm/gr2d.c b/drivers/gpu/host1x/drm/gr2d.c deleted file mode 100644 index 27ffcf15a4b4..000000000000 --- a/drivers/gpu/host1x/drm/gr2d.c +++ /dev/null @@ -1,343 +0,0 @@ -/* - * drivers/video/tegra/host/gr2d/gr2d.c - * - * Tegra Graphics 2D - * - * Copyright (c) 2012-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. - */ - -#include <linux/export.h> -#include <linux/of.h> -#include <linux/of_device.h> -#include <linux/clk.h> - -#include "channel.h" -#include "drm.h" -#include "gem.h" -#include "job.h" -#include "host1x.h" -#include "host1x_bo.h" -#include "host1x_client.h" -#include "syncpt.h" - -struct gr2d { - struct host1x_client client; - struct clk *clk; - struct host1x_channel *channel; - unsigned long *addr_regs; -}; - -static inline struct gr2d *to_gr2d(struct host1x_client *client) -{ - return container_of(client, struct gr2d, client); -} - -static int gr2d_is_addr_reg(struct device *dev, u32 class, u32 reg); - -static int gr2d_client_init(struct host1x_client *client, - struct drm_device *drm) -{ - return 0; -} - -static int gr2d_client_exit(struct host1x_client *client) -{ - return 0; -} - -static int gr2d_open_channel(struct host1x_client *client, - struct host1x_drm_context *context) -{ - struct gr2d *gr2d = to_gr2d(client); - - context->channel = host1x_channel_get(gr2d->channel); - - if (!context->channel) - return -ENOMEM; - - return 0; -} - -static void gr2d_close_channel(struct host1x_drm_context *context) -{ - host1x_channel_put(context->channel); -} - -static struct host1x_bo *host1x_bo_lookup(struct drm_device *drm, - struct drm_file *file, - u32 handle) -{ - struct drm_gem_object *gem; - struct tegra_bo *bo; - - gem = drm_gem_object_lookup(drm, file, handle); - if (!gem) - return NULL; - - mutex_lock(&drm->struct_mutex); - drm_gem_object_unreference(gem); - mutex_unlock(&drm->struct_mutex); - - bo = to_tegra_bo(gem); - return &bo->base; -} - -static int gr2d_submit(struct host1x_drm_context *context, - struct drm_tegra_submit *args, struct drm_device *drm, - struct drm_file *file) -{ - struct host1x_job *job; - unsigned int num_cmdbufs = args->num_cmdbufs; - unsigned int num_relocs = args->num_relocs; - unsigned int num_waitchks = args->num_waitchks; - struct drm_tegra_cmdbuf __user *cmdbufs = - (void * __user)(uintptr_t)args->cmdbufs; - struct drm_tegra_reloc __user *relocs = - (void * __user)(uintptr_t)args->relocs; - struct drm_tegra_waitchk __user *waitchks = - (void * __user)(uintptr_t)args->waitchks; - struct drm_tegra_syncpt syncpt; - int err; - - /* We don't yet support other than one syncpt_incr struct per submit */ - if (args->num_syncpts != 1) - return -EINVAL; - - job = host1x_job_alloc(context->channel, args->num_cmdbufs, - args->num_relocs, args->num_waitchks); - if (!job) - return -ENOMEM; - - job->num_relocs = args->num_relocs; - job->num_waitchk = args->num_waitchks; - job->client = (u32)args->context; - job->class = context->client->class; - job->serialize = true; - - while (num_cmdbufs) { - struct drm_tegra_cmdbuf cmdbuf; - struct host1x_bo *bo; - - err = copy_from_user(&cmdbuf, cmdbufs, sizeof(cmdbuf)); - if (err) - goto fail; - - bo = host1x_bo_lookup(drm, file, cmdbuf.handle); - if (!bo) { - err = -ENOENT; - goto fail; - } - - host1x_job_add_gather(job, bo, cmdbuf.words, cmdbuf.offset); - num_cmdbufs--; - cmdbufs++; - } - - err = copy_from_user(job->relocarray, relocs, - sizeof(*relocs) * num_relocs); - if (err) - goto fail; - - while (num_relocs--) { - struct host1x_reloc *reloc = &job->relocarray[num_relocs]; - struct host1x_bo *cmdbuf, *target; - - cmdbuf = host1x_bo_lookup(drm, file, (u32)reloc->cmdbuf); - target = host1x_bo_lookup(drm, file, (u32)reloc->target); - - reloc->cmdbuf = cmdbuf; - reloc->target = target; - - if (!reloc->target || !reloc->cmdbuf) { - err = -ENOENT; - goto fail; - } - } - - err = copy_from_user(job->waitchk, waitchks, - sizeof(*waitchks) * num_waitchks); - if (err) - goto fail; - - err = copy_from_user(&syncpt, (void * __user)(uintptr_t)args->syncpts, - sizeof(syncpt)); - if (err) - goto fail; - - job->syncpt_id = syncpt.id; - job->syncpt_incrs = syncpt.incrs; - job->timeout = 10000; - job->is_addr_reg = gr2d_is_addr_reg; - - if (args->timeout && args->timeout < 10000) - job->timeout = args->timeout; - - err = host1x_job_pin(job, context->client->dev); - if (err) - goto fail; - - err = host1x_job_submit(job); - if (err) - goto fail_submit; - - args->fence = job->syncpt_end; - - host1x_job_put(job); - return 0; - -fail_submit: - host1x_job_unpin(job); -fail: - host1x_job_put(job); - return err; -} - -static struct host1x_client_ops gr2d_client_ops = { - .drm_init = gr2d_client_init, - .drm_exit = gr2d_client_exit, - .open_channel = gr2d_open_channel, - .close_channel = gr2d_close_channel, - .submit = gr2d_submit, -}; - -static void gr2d_init_addr_reg_map(struct device *dev, struct gr2d *gr2d) -{ - const u32 gr2d_addr_regs[] = {0x1a, 0x1b, 0x26, 0x2b, 0x2c, 0x2d, 0x31, - 0x32, 0x48, 0x49, 0x4a, 0x4b, 0x4c}; - unsigned long *bitmap; - int i; - - bitmap = devm_kzalloc(dev, DIV_ROUND_UP(256, BITS_PER_BYTE), - GFP_KERNEL); - - for (i = 0; i < ARRAY_SIZE(gr2d_addr_regs); ++i) { - u32 reg = gr2d_addr_regs[i]; - bitmap[BIT_WORD(reg)] |= BIT_MASK(reg); - } - - gr2d->addr_regs = bitmap; -} - -static int gr2d_is_addr_reg(struct device *dev, u32 class, u32 reg) -{ - struct gr2d *gr2d = dev_get_drvdata(dev); - - switch (class) { - case HOST1X_CLASS_HOST1X: - return reg == 0x2b; - case HOST1X_CLASS_GR2D: - case HOST1X_CLASS_GR2D_SB: - reg &= 0xff; - if (gr2d->addr_regs[BIT_WORD(reg)] & BIT_MASK(reg)) - return 1; - default: - return 0; - } -} - -static const struct of_device_id gr2d_match[] = { - { .compatible = "nvidia,tegra30-gr2d" }, - { .compatible = "nvidia,tegra20-gr2d" }, - { }, -}; - -static int gr2d_probe(struct platform_device *pdev) -{ - struct device *dev = &pdev->dev; - struct host1x_drm *host1x = host1x_get_drm_data(dev->parent); - int err; - struct gr2d *gr2d = NULL; - struct host1x_syncpt **syncpts; - - gr2d = devm_kzalloc(dev, sizeof(*gr2d), GFP_KERNEL); - if (!gr2d) - return -ENOMEM; - - syncpts = devm_kzalloc(dev, sizeof(*syncpts), GFP_KERNEL); - if (!syncpts) - return -ENOMEM; - - gr2d->clk = devm_clk_get(dev, NULL); - if (IS_ERR(gr2d->clk)) { - dev_err(dev, "cannot get clock\n"); - return PTR_ERR(gr2d->clk); - } - - err = clk_prepare_enable(gr2d->clk); - if (err) { - dev_err(dev, "cannot turn on clock\n"); - return err; - } - - gr2d->channel = host1x_channel_request(dev); - if (!gr2d->channel) - return -ENOMEM; - - *syncpts = host1x_syncpt_request(dev, false); - if (!(*syncpts)) { - host1x_channel_free(gr2d->channel); - return -ENOMEM; - } - - gr2d->client.ops = &gr2d_client_ops; - gr2d->client.dev = dev; - gr2d->client.class = HOST1X_CLASS_GR2D; - gr2d->client.syncpts = syncpts; - gr2d->client.num_syncpts = 1; - - err = host1x_register_client(host1x, &gr2d->client); - if (err < 0) { - dev_err(dev, "failed to register host1x client: %d\n", err); - return err; - } - - gr2d_init_addr_reg_map(dev, gr2d); - - platform_set_drvdata(pdev, gr2d); - - return 0; -} - -static int __exit gr2d_remove(struct platform_device *pdev) -{ - struct host1x_drm *host1x = host1x_get_drm_data(pdev->dev.parent); - struct gr2d *gr2d = platform_get_drvdata(pdev); - unsigned int i; - int err; - - err = host1x_unregister_client(host1x, &gr2d->client); - if (err < 0) { - dev_err(&pdev->dev, "failed to unregister client: %d\n", err); - return err; - } - - for (i = 0; i < gr2d->client.num_syncpts; i++) - host1x_syncpt_free(gr2d->client.syncpts[i]); - - host1x_channel_free(gr2d->channel); - clk_disable_unprepare(gr2d->clk); - - return 0; -} - -struct platform_driver tegra_gr2d_driver = { - .probe = gr2d_probe, - .remove = __exit_p(gr2d_remove), - .driver = { - .owner = THIS_MODULE, - .name = "gr2d", - .of_match_table = gr2d_match, - } -}; diff --git a/drivers/gpu/host1x/drm/hdmi.c b/drivers/gpu/host1x/drm/hdmi.c deleted file mode 100644 index 01097da09f7f..000000000000 --- a/drivers/gpu/host1x/drm/hdmi.c +++ /dev/null @@ -1,1313 +0,0 @@ -/* - * Copyright (C) 2012 Avionic Design GmbH - * Copyright (C) 2012 NVIDIA CORPORATION. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#include <linux/clk.h> -#include <linux/debugfs.h> -#include <linux/gpio.h> -#include <linux/hdmi.h> -#include <linux/module.h> -#include <linux/of.h> -#include <linux/platform_device.h> -#include <linux/regulator/consumer.h> -#include <linux/clk/tegra.h> - -#include <drm/drm_edid.h> - -#include "hdmi.h" -#include "drm.h" -#include "dc.h" -#include "host1x_client.h" - -struct tegra_hdmi { - struct host1x_client client; - struct tegra_output output; - struct device *dev; - - struct regulator *vdd; - struct regulator *pll; - - void __iomem *regs; - unsigned int irq; - - struct clk *clk_parent; - struct clk *clk; - - unsigned int audio_source; - unsigned int audio_freq; - bool stereo; - bool dvi; - - struct drm_info_list *debugfs_files; - struct drm_minor *minor; - struct dentry *debugfs; -}; - -static inline struct tegra_hdmi * -host1x_client_to_hdmi(struct host1x_client *client) -{ - return container_of(client, struct tegra_hdmi, client); -} - -static inline struct tegra_hdmi *to_hdmi(struct tegra_output *output) -{ - return container_of(output, struct tegra_hdmi, output); -} - -#define HDMI_AUDIOCLK_FREQ 216000000 -#define HDMI_REKEY_DEFAULT 56 - -enum { - AUTO = 0, - SPDIF, - HDA, -}; - -static inline unsigned long tegra_hdmi_readl(struct tegra_hdmi *hdmi, - unsigned long reg) -{ - return readl(hdmi->regs + (reg << 2)); -} - -static inline void tegra_hdmi_writel(struct tegra_hdmi *hdmi, unsigned long val, - unsigned long reg) -{ - writel(val, hdmi->regs + (reg << 2)); -} - -struct tegra_hdmi_audio_config { - unsigned int pclk; - unsigned int n; - unsigned int cts; - unsigned int aval; -}; - -static const struct tegra_hdmi_audio_config tegra_hdmi_audio_32k[] = { - { 25200000, 4096, 25200, 24000 }, - { 27000000, 4096, 27000, 24000 }, - { 74250000, 4096, 74250, 24000 }, - { 148500000, 4096, 148500, 24000 }, - { 0, 0, 0, 0 }, -}; - -static const struct tegra_hdmi_audio_config tegra_hdmi_audio_44_1k[] = { - { 25200000, 5880, 26250, 25000 }, - { 27000000, 5880, 28125, 25000 }, - { 74250000, 4704, 61875, 20000 }, - { 148500000, 4704, 123750, 20000 }, - { 0, 0, 0, 0 }, -}; - -static const struct tegra_hdmi_audio_config tegra_hdmi_audio_48k[] = { - { 25200000, 6144, 25200, 24000 }, - { 27000000, 6144, 27000, 24000 }, - { 74250000, 6144, 74250, 24000 }, - { 148500000, 6144, 148500, 24000 }, - { 0, 0, 0, 0 }, -}; - -static const struct tegra_hdmi_audio_config tegra_hdmi_audio_88_2k[] = { - { 25200000, 11760, 26250, 25000 }, - { 27000000, 11760, 28125, 25000 }, - { 74250000, 9408, 61875, 20000 }, - { 148500000, 9408, 123750, 20000 }, - { 0, 0, 0, 0 }, -}; - -static const struct tegra_hdmi_audio_config tegra_hdmi_audio_96k[] = { - { 25200000, 12288, 25200, 24000 }, - { 27000000, 12288, 27000, 24000 }, - { 74250000, 12288, 74250, 24000 }, - { 148500000, 12288, 148500, 24000 }, - { 0, 0, 0, 0 }, -}; - -static const struct tegra_hdmi_audio_config tegra_hdmi_audio_176_4k[] = { - { 25200000, 23520, 26250, 25000 }, - { 27000000, 23520, 28125, 25000 }, - { 74250000, 18816, 61875, 20000 }, - { 148500000, 18816, 123750, 20000 }, - { 0, 0, 0, 0 }, -}; - -static const struct tegra_hdmi_audio_config tegra_hdmi_audio_192k[] = { - { 25200000, 24576, 25200, 24000 }, - { 27000000, 24576, 27000, 24000 }, - { 74250000, 24576, 74250, 24000 }, - { 148500000, 24576, 148500, 24000 }, - { 0, 0, 0, 0 }, -}; - -struct tmds_config { - unsigned int pclk; - u32 pll0; - u32 pll1; - u32 pe_current; - u32 drive_current; -}; - -static const struct tmds_config tegra2_tmds_config[] = { - { /* slow pixel clock modes */ - .pclk = 27000000, - .pll0 = SOR_PLL_BG_V17_S(3) | SOR_PLL_ICHPMP(1) | - SOR_PLL_RESISTORSEL | SOR_PLL_VCOCAP(0) | - SOR_PLL_TX_REG_LOAD(3), - .pll1 = SOR_PLL_TMDS_TERM_ENABLE, - .pe_current = PE_CURRENT0(PE_CURRENT_0_0_mA) | - PE_CURRENT1(PE_CURRENT_0_0_mA) | - PE_CURRENT2(PE_CURRENT_0_0_mA) | - PE_CURRENT3(PE_CURRENT_0_0_mA), - .drive_current = DRIVE_CURRENT_LANE0(DRIVE_CURRENT_7_125_mA) | - DRIVE_CURRENT_LANE1(DRIVE_CURRENT_7_125_mA) | - DRIVE_CURRENT_LANE2(DRIVE_CURRENT_7_125_mA) | - DRIVE_CURRENT_LANE3(DRIVE_CURRENT_7_125_mA), - }, - { /* high pixel clock modes */ - .pclk = UINT_MAX, - .pll0 = SOR_PLL_BG_V17_S(3) | SOR_PLL_ICHPMP(1) | - SOR_PLL_RESISTORSEL | SOR_PLL_VCOCAP(1) | - SOR_PLL_TX_REG_LOAD(3), - .pll1 = SOR_PLL_TMDS_TERM_ENABLE | SOR_PLL_PE_EN, - .pe_current = PE_CURRENT0(PE_CURRENT_6_0_mA) | - PE_CURRENT1(PE_CURRENT_6_0_mA) | - PE_CURRENT2(PE_CURRENT_6_0_mA) | - PE_CURRENT3(PE_CURRENT_6_0_mA), - .drive_current = DRIVE_CURRENT_LANE0(DRIVE_CURRENT_7_125_mA) | - DRIVE_CURRENT_LANE1(DRIVE_CURRENT_7_125_mA) | - DRIVE_CURRENT_LANE2(DRIVE_CURRENT_7_125_mA) | - DRIVE_CURRENT_LANE3(DRIVE_CURRENT_7_125_mA), - }, -}; - -static const struct tmds_config tegra3_tmds_config[] = { - { /* 480p modes */ - .pclk = 27000000, - .pll0 = SOR_PLL_BG_V17_S(3) | SOR_PLL_ICHPMP(1) | - SOR_PLL_RESISTORSEL | SOR_PLL_VCOCAP(0) | - SOR_PLL_TX_REG_LOAD(0), - .pll1 = SOR_PLL_TMDS_TERM_ENABLE, - .pe_current = PE_CURRENT0(PE_CURRENT_0_0_mA) | - PE_CURRENT1(PE_CURRENT_0_0_mA) | - PE_CURRENT2(PE_CURRENT_0_0_mA) | - PE_CURRENT3(PE_CURRENT_0_0_mA), - .drive_current = DRIVE_CURRENT_LANE0(DRIVE_CURRENT_5_250_mA) | - DRIVE_CURRENT_LANE1(DRIVE_CURRENT_5_250_mA) | - DRIVE_CURRENT_LANE2(DRIVE_CURRENT_5_250_mA) | - DRIVE_CURRENT_LANE3(DRIVE_CURRENT_5_250_mA), - }, { /* 720p modes */ - .pclk = 74250000, - .pll0 = SOR_PLL_BG_V17_S(3) | SOR_PLL_ICHPMP(1) | - SOR_PLL_RESISTORSEL | SOR_PLL_VCOCAP(1) | - SOR_PLL_TX_REG_LOAD(0), - .pll1 = SOR_PLL_TMDS_TERM_ENABLE | SOR_PLL_PE_EN, - .pe_current = PE_CURRENT0(PE_CURRENT_5_0_mA) | - PE_CURRENT1(PE_CURRENT_5_0_mA) | - PE_CURRENT2(PE_CURRENT_5_0_mA) | - PE_CURRENT3(PE_CURRENT_5_0_mA), - .drive_current = DRIVE_CURRENT_LANE0(DRIVE_CURRENT_5_250_mA) | - DRIVE_CURRENT_LANE1(DRIVE_CURRENT_5_250_mA) | - DRIVE_CURRENT_LANE2(DRIVE_CURRENT_5_250_mA) | - DRIVE_CURRENT_LANE3(DRIVE_CURRENT_5_250_mA), - }, { /* 1080p modes */ - .pclk = UINT_MAX, - .pll0 = SOR_PLL_BG_V17_S(3) | SOR_PLL_ICHPMP(1) | - SOR_PLL_RESISTORSEL | SOR_PLL_VCOCAP(3) | - SOR_PLL_TX_REG_LOAD(0), - .pll1 = SOR_PLL_TMDS_TERM_ENABLE | SOR_PLL_PE_EN, - .pe_current = PE_CURRENT0(PE_CURRENT_5_0_mA) | - PE_CURRENT1(PE_CURRENT_5_0_mA) | - PE_CURRENT2(PE_CURRENT_5_0_mA) | - PE_CURRENT3(PE_CURRENT_5_0_mA), - .drive_current = DRIVE_CURRENT_LANE0(DRIVE_CURRENT_5_250_mA) | - DRIVE_CURRENT_LANE1(DRIVE_CURRENT_5_250_mA) | - DRIVE_CURRENT_LANE2(DRIVE_CURRENT_5_250_mA) | - DRIVE_CURRENT_LANE3(DRIVE_CURRENT_5_250_mA), - }, -}; - -static const struct tegra_hdmi_audio_config * -tegra_hdmi_get_audio_config(unsigned int audio_freq, unsigned int pclk) -{ - const struct tegra_hdmi_audio_config *table; - - switch (audio_freq) { - case 32000: - table = tegra_hdmi_audio_32k; - break; - - case 44100: - table = tegra_hdmi_audio_44_1k; - break; - - case 48000: - table = tegra_hdmi_audio_48k; - break; - - case 88200: - table = tegra_hdmi_audio_88_2k; - break; - - case 96000: - table = tegra_hdmi_audio_96k; - break; - - case 176400: - table = tegra_hdmi_audio_176_4k; - break; - - case 192000: - table = tegra_hdmi_audio_192k; - break; - - default: - return NULL; - } - - while (table->pclk) { - if (table->pclk == pclk) - return table; - - table++; - } - - return NULL; -} - -static void tegra_hdmi_setup_audio_fs_tables(struct tegra_hdmi *hdmi) -{ - const unsigned int freqs[] = { - 32000, 44100, 48000, 88200, 96000, 176400, 192000 - }; - unsigned int i; - - for (i = 0; i < ARRAY_SIZE(freqs); i++) { - unsigned int f = freqs[i]; - unsigned int eight_half; - unsigned long value; - unsigned int delta; - - if (f > 96000) - delta = 2; - else if (f > 480000) - delta = 6; - else - delta = 9; - - eight_half = (8 * HDMI_AUDIOCLK_FREQ) / (f * 128); - value = AUDIO_FS_LOW(eight_half - delta) | - AUDIO_FS_HIGH(eight_half + delta); - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_AUDIO_FS(i)); - } -} - -static int tegra_hdmi_setup_audio(struct tegra_hdmi *hdmi, unsigned int pclk) -{ - struct device_node *node = hdmi->dev->of_node; - const struct tegra_hdmi_audio_config *config; - unsigned int offset = 0; - unsigned long value; - - switch (hdmi->audio_source) { - case HDA: - value = AUDIO_CNTRL0_SOURCE_SELECT_HDAL; - break; - - case SPDIF: - value = AUDIO_CNTRL0_SOURCE_SELECT_SPDIF; - break; - - default: - value = AUDIO_CNTRL0_SOURCE_SELECT_AUTO; - break; - } - - if (of_device_is_compatible(node, "nvidia,tegra30-hdmi")) { - value |= AUDIO_CNTRL0_ERROR_TOLERANCE(6) | - AUDIO_CNTRL0_FRAMES_PER_BLOCK(0xc0); - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_AUDIO_CNTRL0); - } else { - value |= AUDIO_CNTRL0_INJECT_NULLSMPL; - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_AUDIO_CNTRL0); - - value = AUDIO_CNTRL0_ERROR_TOLERANCE(6) | - AUDIO_CNTRL0_FRAMES_PER_BLOCK(0xc0); - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_AUDIO_CNTRL0); - } - - config = tegra_hdmi_get_audio_config(hdmi->audio_freq, pclk); - if (!config) { - dev_err(hdmi->dev, "cannot set audio to %u at %u pclk\n", - hdmi->audio_freq, pclk); - return -EINVAL; - } - - tegra_hdmi_writel(hdmi, 0, HDMI_NV_PDISP_HDMI_ACR_CTRL); - - value = AUDIO_N_RESETF | AUDIO_N_GENERATE_ALTERNATE | - AUDIO_N_VALUE(config->n - 1); - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_AUDIO_N); - - tegra_hdmi_writel(hdmi, ACR_SUBPACK_N(config->n) | ACR_ENABLE, - HDMI_NV_PDISP_HDMI_ACR_0441_SUBPACK_HIGH); - - value = ACR_SUBPACK_CTS(config->cts); - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_HDMI_ACR_0441_SUBPACK_LOW); - - value = SPARE_HW_CTS | SPARE_FORCE_SW_CTS | SPARE_CTS_RESET_VAL(1); - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_HDMI_SPARE); - - value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_AUDIO_N); - value &= ~AUDIO_N_RESETF; - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_AUDIO_N); - - if (of_device_is_compatible(node, "nvidia,tegra30-hdmi")) { - switch (hdmi->audio_freq) { - case 32000: - offset = HDMI_NV_PDISP_SOR_AUDIO_AVAL_0320; - break; - - case 44100: - offset = HDMI_NV_PDISP_SOR_AUDIO_AVAL_0441; - break; - - case 48000: - offset = HDMI_NV_PDISP_SOR_AUDIO_AVAL_0480; - break; - - case 88200: - offset = HDMI_NV_PDISP_SOR_AUDIO_AVAL_0882; - break; - - case 96000: - offset = HDMI_NV_PDISP_SOR_AUDIO_AVAL_0960; - break; - - case 176400: - offset = HDMI_NV_PDISP_SOR_AUDIO_AVAL_1764; - break; - - case 192000: - offset = HDMI_NV_PDISP_SOR_AUDIO_AVAL_1920; - break; - } - - tegra_hdmi_writel(hdmi, config->aval, offset); - } - - tegra_hdmi_setup_audio_fs_tables(hdmi); - - return 0; -} - -static inline unsigned long tegra_hdmi_subpack(const u8 *ptr, size_t size) -{ - unsigned long value = 0; - size_t i; - - for (i = size; i > 0; i--) - value = (value << 8) | ptr[i - 1]; - - return value; -} - -static void tegra_hdmi_write_infopack(struct tegra_hdmi *hdmi, const void *data, - size_t size) -{ - const u8 *ptr = data; - unsigned long offset; - unsigned long value; - size_t i, j; - - switch (ptr[0]) { - case HDMI_INFOFRAME_TYPE_AVI: - offset = HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_HEADER; - break; - - case HDMI_INFOFRAME_TYPE_AUDIO: - offset = HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_HEADER; - break; - - case HDMI_INFOFRAME_TYPE_VENDOR: - offset = HDMI_NV_PDISP_HDMI_GENERIC_HEADER; - break; - - default: - dev_err(hdmi->dev, "unsupported infoframe type: %02x\n", - ptr[0]); - return; - } - - value = INFOFRAME_HEADER_TYPE(ptr[0]) | - INFOFRAME_HEADER_VERSION(ptr[1]) | - INFOFRAME_HEADER_LEN(ptr[2]); - tegra_hdmi_writel(hdmi, value, offset); - offset++; - - /* - * Each subpack contains 7 bytes, divided into: - * - subpack_low: bytes 0 - 3 - * - subpack_high: bytes 4 - 6 (with byte 7 padded to 0x00) - */ - for (i = 3, j = 0; i < size; i += 7, j += 8) { - size_t rem = size - i, num = min_t(size_t, rem, 4); - - value = tegra_hdmi_subpack(&ptr[i], num); - tegra_hdmi_writel(hdmi, value, offset++); - - num = min_t(size_t, rem - num, 3); - - value = tegra_hdmi_subpack(&ptr[i + 4], num); - tegra_hdmi_writel(hdmi, value, offset++); - } -} - -static void tegra_hdmi_setup_avi_infoframe(struct tegra_hdmi *hdmi, - struct drm_display_mode *mode) -{ - struct hdmi_avi_infoframe frame; - u8 buffer[17]; - ssize_t err; - - if (hdmi->dvi) { - tegra_hdmi_writel(hdmi, 0, - HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_CTRL); - return; - } - - err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode); - if (err < 0) { - dev_err(hdmi->dev, "failed to setup AVI infoframe: %zd\n", err); - return; - } - - err = hdmi_avi_infoframe_pack(&frame, buffer, sizeof(buffer)); - if (err < 0) { - dev_err(hdmi->dev, "failed to pack AVI infoframe: %zd\n", err); - return; - } - - tegra_hdmi_write_infopack(hdmi, buffer, err); - - tegra_hdmi_writel(hdmi, INFOFRAME_CTRL_ENABLE, - HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_CTRL); -} - -static void tegra_hdmi_setup_audio_infoframe(struct tegra_hdmi *hdmi) -{ - struct hdmi_audio_infoframe frame; - u8 buffer[14]; - ssize_t err; - - if (hdmi->dvi) { - tegra_hdmi_writel(hdmi, 0, - HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_CTRL); - return; - } - - err = hdmi_audio_infoframe_init(&frame); - if (err < 0) { - dev_err(hdmi->dev, "failed to initialize audio infoframe: %d\n", - err); - return; - } - - frame.channels = 2; - - err = hdmi_audio_infoframe_pack(&frame, buffer, sizeof(buffer)); - if (err < 0) { - dev_err(hdmi->dev, "failed to pack audio infoframe: %zd\n", - err); - return; - } - - /* - * The audio infoframe has only one set of subpack registers, so the - * infoframe needs to be truncated. One set of subpack registers can - * contain 7 bytes. Including the 3 byte header only the first 10 - * bytes can be programmed. - */ - tegra_hdmi_write_infopack(hdmi, buffer, min(10, err)); - - tegra_hdmi_writel(hdmi, INFOFRAME_CTRL_ENABLE, - HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_CTRL); -} - -static void tegra_hdmi_setup_stereo_infoframe(struct tegra_hdmi *hdmi) -{ - struct hdmi_vendor_infoframe frame; - unsigned long value; - u8 buffer[10]; - ssize_t err; - - if (!hdmi->stereo) { - value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_HDMI_GENERIC_CTRL); - value &= ~GENERIC_CTRL_ENABLE; - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_HDMI_GENERIC_CTRL); - return; - } - - memset(&frame, 0, sizeof(frame)); - - frame.type = HDMI_INFOFRAME_TYPE_VENDOR; - frame.version = 0x01; - frame.length = 6; - - frame.data[0] = 0x03; /* regid0 */ - frame.data[1] = 0x0c; /* regid1 */ - frame.data[2] = 0x00; /* regid2 */ - frame.data[3] = 0x02 << 5; /* video format */ - - /* TODO: 74 MHz limit? */ - if (1) { - frame.data[4] = 0x00 << 4; /* 3D structure */ - } else { - frame.data[4] = 0x08 << 4; /* 3D structure */ - frame.data[5] = 0x00 << 4; /* 3D ext. data */ - } - - err = hdmi_vendor_infoframe_pack(&frame, buffer, sizeof(buffer)); - if (err < 0) { - dev_err(hdmi->dev, "failed to pack vendor infoframe: %zd\n", - err); - return; - } - - tegra_hdmi_write_infopack(hdmi, buffer, err); - - value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_HDMI_GENERIC_CTRL); - value |= GENERIC_CTRL_ENABLE; - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_HDMI_GENERIC_CTRL); -} - -static void tegra_hdmi_setup_tmds(struct tegra_hdmi *hdmi, - const struct tmds_config *tmds) -{ - unsigned long value; - - tegra_hdmi_writel(hdmi, tmds->pll0, HDMI_NV_PDISP_SOR_PLL0); - tegra_hdmi_writel(hdmi, tmds->pll1, HDMI_NV_PDISP_SOR_PLL1); - tegra_hdmi_writel(hdmi, tmds->pe_current, HDMI_NV_PDISP_PE_CURRENT); - - value = tmds->drive_current | DRIVE_CURRENT_FUSE_OVERRIDE; - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_LANE_DRIVE_CURRENT); -} - -static int tegra_output_hdmi_enable(struct tegra_output *output) -{ - unsigned int h_sync_width, h_front_porch, h_back_porch, i, rekey; - struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc); - struct drm_display_mode *mode = &dc->base.mode; - struct tegra_hdmi *hdmi = to_hdmi(output); - struct device_node *node = hdmi->dev->of_node; - unsigned int pulse_start, div82, pclk; - const struct tmds_config *tmds; - unsigned int num_tmds; - unsigned long value; - int retries = 1000; - int err; - - pclk = mode->clock * 1000; - h_sync_width = mode->hsync_end - mode->hsync_start; - h_back_porch = mode->htotal - mode->hsync_end; - h_front_porch = mode->hsync_start - mode->hdisplay; - - err = regulator_enable(hdmi->vdd); - if (err < 0) { - dev_err(hdmi->dev, "failed to enable VDD regulator: %d\n", err); - return err; - } - - err = regulator_enable(hdmi->pll); - if (err < 0) { - dev_err(hdmi->dev, "failed to enable PLL regulator: %d\n", err); - return err; - } - - /* - * This assumes that the display controller will divide its parent - * clock by 2 to generate the pixel clock. - */ - err = tegra_output_setup_clock(output, hdmi->clk, pclk * 2); - if (err < 0) { - dev_err(hdmi->dev, "failed to setup clock: %d\n", err); - return err; - } - - err = clk_set_rate(hdmi->clk, pclk); - if (err < 0) - return err; - - err = clk_enable(hdmi->clk); - if (err < 0) { - dev_err(hdmi->dev, "failed to enable clock: %d\n", err); - return err; - } - - tegra_periph_reset_assert(hdmi->clk); - usleep_range(1000, 2000); - tegra_periph_reset_deassert(hdmi->clk); - - tegra_dc_writel(dc, VSYNC_H_POSITION(1), - DC_DISP_DISP_TIMING_OPTIONS); - tegra_dc_writel(dc, DITHER_CONTROL_DISABLE | BASE_COLOR_SIZE888, - DC_DISP_DISP_COLOR_CONTROL); - - /* video_preamble uses h_pulse2 */ - pulse_start = 1 + h_sync_width + h_back_porch - 10; - - tegra_dc_writel(dc, H_PULSE_2_ENABLE, DC_DISP_DISP_SIGNAL_OPTIONS0); - - value = PULSE_MODE_NORMAL | PULSE_POLARITY_HIGH | PULSE_QUAL_VACTIVE | - PULSE_LAST_END_A; - tegra_dc_writel(dc, value, DC_DISP_H_PULSE2_CONTROL); - - value = PULSE_START(pulse_start) | PULSE_END(pulse_start + 8); - tegra_dc_writel(dc, value, DC_DISP_H_PULSE2_POSITION_A); - - value = VSYNC_WINDOW_END(0x210) | VSYNC_WINDOW_START(0x200) | - VSYNC_WINDOW_ENABLE; - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_HDMI_VSYNC_WINDOW); - - if (dc->pipe) - value = HDMI_SRC_DISPLAYB; - else - value = HDMI_SRC_DISPLAYA; - - if ((mode->hdisplay == 720) && ((mode->vdisplay == 480) || - (mode->vdisplay == 576))) - tegra_hdmi_writel(hdmi, - value | ARM_VIDEO_RANGE_FULL, - HDMI_NV_PDISP_INPUT_CONTROL); - else - tegra_hdmi_writel(hdmi, - value | ARM_VIDEO_RANGE_LIMITED, - HDMI_NV_PDISP_INPUT_CONTROL); - - div82 = clk_get_rate(hdmi->clk) / 1000000 * 4; - value = SOR_REFCLK_DIV_INT(div82 >> 2) | SOR_REFCLK_DIV_FRAC(div82); - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_REFCLK); - - if (!hdmi->dvi) { - err = tegra_hdmi_setup_audio(hdmi, pclk); - if (err < 0) - hdmi->dvi = true; - } - - if (of_device_is_compatible(node, "nvidia,tegra20-hdmi")) { - /* - * TODO: add ELD support - */ - } - - rekey = HDMI_REKEY_DEFAULT; - value = HDMI_CTRL_REKEY(rekey); - value |= HDMI_CTRL_MAX_AC_PACKET((h_sync_width + h_back_porch + - h_front_porch - rekey - 18) / 32); - - if (!hdmi->dvi) - value |= HDMI_CTRL_ENABLE; - - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_HDMI_CTRL); - - if (hdmi->dvi) - tegra_hdmi_writel(hdmi, 0x0, - HDMI_NV_PDISP_HDMI_GENERIC_CTRL); - else - tegra_hdmi_writel(hdmi, GENERIC_CTRL_AUDIO, - HDMI_NV_PDISP_HDMI_GENERIC_CTRL); - - tegra_hdmi_setup_avi_infoframe(hdmi, mode); - tegra_hdmi_setup_audio_infoframe(hdmi); - tegra_hdmi_setup_stereo_infoframe(hdmi); - - /* TMDS CONFIG */ - if (of_device_is_compatible(node, "nvidia,tegra30-hdmi")) { - num_tmds = ARRAY_SIZE(tegra3_tmds_config); - tmds = tegra3_tmds_config; - } else { - num_tmds = ARRAY_SIZE(tegra2_tmds_config); - tmds = tegra2_tmds_config; - } - - for (i = 0; i < num_tmds; i++) { - if (pclk <= tmds[i].pclk) { - tegra_hdmi_setup_tmds(hdmi, &tmds[i]); - break; - } - } - - tegra_hdmi_writel(hdmi, - SOR_SEQ_CTL_PU_PC(0) | - SOR_SEQ_PU_PC_ALT(0) | - SOR_SEQ_PD_PC(8) | - SOR_SEQ_PD_PC_ALT(8), - HDMI_NV_PDISP_SOR_SEQ_CTL); - - value = SOR_SEQ_INST_WAIT_TIME(1) | - SOR_SEQ_INST_WAIT_UNITS_VSYNC | - SOR_SEQ_INST_HALT | - SOR_SEQ_INST_PIN_A_LOW | - SOR_SEQ_INST_PIN_B_LOW | - SOR_SEQ_INST_DRIVE_PWM_OUT_LO; - - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_SEQ_INST(0)); - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_SEQ_INST(8)); - - value = 0x1c800; - value &= ~SOR_CSTM_ROTCLK(~0); - value |= SOR_CSTM_ROTCLK(2); - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_CSTM); - - tegra_dc_writel(dc, DISP_CTRL_MODE_STOP, DC_CMD_DISPLAY_COMMAND); - tegra_dc_writel(dc, GENERAL_ACT_REQ << 8, DC_CMD_STATE_CONTROL); - tegra_dc_writel(dc, GENERAL_ACT_REQ, DC_CMD_STATE_CONTROL); - - /* start SOR */ - tegra_hdmi_writel(hdmi, - SOR_PWR_NORMAL_STATE_PU | - SOR_PWR_NORMAL_START_NORMAL | - SOR_PWR_SAFE_STATE_PD | - SOR_PWR_SETTING_NEW_TRIGGER, - HDMI_NV_PDISP_SOR_PWR); - tegra_hdmi_writel(hdmi, - SOR_PWR_NORMAL_STATE_PU | - SOR_PWR_NORMAL_START_NORMAL | - SOR_PWR_SAFE_STATE_PD | - SOR_PWR_SETTING_NEW_DONE, - HDMI_NV_PDISP_SOR_PWR); - - do { - BUG_ON(--retries < 0); - value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_SOR_PWR); - } while (value & SOR_PWR_SETTING_NEW_PENDING); - - value = SOR_STATE_ASY_CRCMODE_COMPLETE | - SOR_STATE_ASY_OWNER_HEAD0 | - SOR_STATE_ASY_SUBOWNER_BOTH | - SOR_STATE_ASY_PROTOCOL_SINGLE_TMDS_A | - SOR_STATE_ASY_DEPOL_POS; - - /* setup sync polarities */ - if (mode->flags & DRM_MODE_FLAG_PHSYNC) - value |= SOR_STATE_ASY_HSYNCPOL_POS; - - if (mode->flags & DRM_MODE_FLAG_NHSYNC) - value |= SOR_STATE_ASY_HSYNCPOL_NEG; - - if (mode->flags & DRM_MODE_FLAG_PVSYNC) - value |= SOR_STATE_ASY_VSYNCPOL_POS; - - if (mode->flags & DRM_MODE_FLAG_NVSYNC) - value |= SOR_STATE_ASY_VSYNCPOL_NEG; - - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_STATE2); - - value = SOR_STATE_ASY_HEAD_OPMODE_AWAKE | SOR_STATE_ASY_ORMODE_NORMAL; - tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_SOR_STATE1); - - tegra_hdmi_writel(hdmi, 0, HDMI_NV_PDISP_SOR_STATE0); - tegra_hdmi_writel(hdmi, SOR_STATE_UPDATE, HDMI_NV_PDISP_SOR_STATE0); - tegra_hdmi_writel(hdmi, value | SOR_STATE_ATTACHED, - HDMI_NV_PDISP_SOR_STATE1); - tegra_hdmi_writel(hdmi, 0, HDMI_NV_PDISP_SOR_STATE0); - - tegra_dc_writel(dc, HDMI_ENABLE, DC_DISP_DISP_WIN_OPTIONS); - - value = PW0_ENABLE | PW1_ENABLE | PW2_ENABLE | PW3_ENABLE | - PW4_ENABLE | PM0_ENABLE | PM1_ENABLE; - tegra_dc_writel(dc, value, DC_CMD_DISPLAY_POWER_CONTROL); - - value = DISP_CTRL_MODE_C_DISPLAY; - tegra_dc_writel(dc, value, DC_CMD_DISPLAY_COMMAND); - - tegra_dc_writel(dc, GENERAL_ACT_REQ << 8, DC_CMD_STATE_CONTROL); - tegra_dc_writel(dc, GENERAL_ACT_REQ, DC_CMD_STATE_CONTROL); - - /* TODO: add HDCP support */ - - return 0; -} - -static int tegra_output_hdmi_disable(struct tegra_output *output) -{ - struct tegra_hdmi *hdmi = to_hdmi(output); - - tegra_periph_reset_assert(hdmi->clk); - clk_disable(hdmi->clk); - regulator_disable(hdmi->pll); - regulator_disable(hdmi->vdd); - - return 0; -} - -static int tegra_output_hdmi_setup_clock(struct tegra_output *output, - struct clk *clk, unsigned long pclk) -{ - struct tegra_hdmi *hdmi = to_hdmi(output); - struct clk *base; - int err; - - err = clk_set_parent(clk, hdmi->clk_parent); - if (err < 0) { - dev_err(output->dev, "failed to set parent: %d\n", err); - return err; - } - - base = clk_get_parent(hdmi->clk_parent); - - /* - * This assumes that the parent clock is pll_d_out0 or pll_d2_out - * respectively, each of which divides the base pll_d by 2. - */ - err = clk_set_rate(base, pclk * 2); - if (err < 0) - dev_err(output->dev, - "failed to set base clock rate to %lu Hz\n", - pclk * 2); - - return 0; -} - -static int tegra_output_hdmi_check_mode(struct tegra_output *output, - struct drm_display_mode *mode, - enum drm_mode_status *status) -{ - struct tegra_hdmi *hdmi = to_hdmi(output); - unsigned long pclk = mode->clock * 1000; - struct clk *parent; - long err; - - parent = clk_get_parent(hdmi->clk_parent); - - err = clk_round_rate(parent, pclk * 4); - if (err < 0) - *status = MODE_NOCLOCK; - else - *status = MODE_OK; - - return 0; -} - -static const struct tegra_output_ops hdmi_ops = { - .enable = tegra_output_hdmi_enable, - .disable = tegra_output_hdmi_disable, - .setup_clock = tegra_output_hdmi_setup_clock, - .check_mode = tegra_output_hdmi_check_mode, -}; - -static int tegra_hdmi_show_regs(struct seq_file *s, void *data) -{ - struct drm_info_node *node = s->private; - struct tegra_hdmi *hdmi = node->info_ent->data; - -#define DUMP_REG(name) \ - seq_printf(s, "%-56s %#05x %08lx\n", #name, name, \ - tegra_hdmi_readl(hdmi, name)) - - DUMP_REG(HDMI_CTXSW); - DUMP_REG(HDMI_NV_PDISP_SOR_STATE0); - DUMP_REG(HDMI_NV_PDISP_SOR_STATE1); - DUMP_REG(HDMI_NV_PDISP_SOR_STATE2); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_AN_MSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_AN_LSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_CN_MSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_CN_LSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_AKSV_MSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_AKSV_LSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_BKSV_MSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_BKSV_LSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_CKSV_MSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_CKSV_LSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_DKSV_MSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_DKSV_LSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_CTRL); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_CMODE); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_MPRIME_MSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_MPRIME_LSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_SPRIME_MSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_SPRIME_LSB2); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_SPRIME_LSB1); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_RI); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_CS_MSB); - DUMP_REG(HDMI_NV_PDISP_RG_HDCP_CS_LSB); - DUMP_REG(HDMI_NV_PDISP_HDMI_AUDIO_EMU0); - DUMP_REG(HDMI_NV_PDISP_HDMI_AUDIO_EMU_RDATA0); - DUMP_REG(HDMI_NV_PDISP_HDMI_AUDIO_EMU1); - DUMP_REG(HDMI_NV_PDISP_HDMI_AUDIO_EMU2); - DUMP_REG(HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_CTRL); - DUMP_REG(HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_STATUS); - DUMP_REG(HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_HEADER); - DUMP_REG(HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_SUBPACK0_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_SUBPACK0_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_CTRL); - DUMP_REG(HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_STATUS); - DUMP_REG(HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_HEADER); - DUMP_REG(HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_SUBPACK0_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_SUBPACK0_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_SUBPACK1_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_SUBPACK1_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_GENERIC_CTRL); - DUMP_REG(HDMI_NV_PDISP_HDMI_GENERIC_STATUS); - DUMP_REG(HDMI_NV_PDISP_HDMI_GENERIC_HEADER); - DUMP_REG(HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK0_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK0_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK1_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK1_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK2_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK2_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK3_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK3_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_CTRL); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_0320_SUBPACK_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_0320_SUBPACK_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_0441_SUBPACK_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_0441_SUBPACK_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_0882_SUBPACK_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_0882_SUBPACK_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_1764_SUBPACK_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_1764_SUBPACK_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_0480_SUBPACK_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_0480_SUBPACK_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_0960_SUBPACK_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_0960_SUBPACK_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_1920_SUBPACK_LOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_ACR_1920_SUBPACK_HIGH); - DUMP_REG(HDMI_NV_PDISP_HDMI_CTRL); - DUMP_REG(HDMI_NV_PDISP_HDMI_VSYNC_KEEPOUT); - DUMP_REG(HDMI_NV_PDISP_HDMI_VSYNC_WINDOW); - DUMP_REG(HDMI_NV_PDISP_HDMI_GCP_CTRL); - DUMP_REG(HDMI_NV_PDISP_HDMI_GCP_STATUS); - DUMP_REG(HDMI_NV_PDISP_HDMI_GCP_SUBPACK); - DUMP_REG(HDMI_NV_PDISP_HDMI_CHANNEL_STATUS1); - DUMP_REG(HDMI_NV_PDISP_HDMI_CHANNEL_STATUS2); - DUMP_REG(HDMI_NV_PDISP_HDMI_EMU0); - DUMP_REG(HDMI_NV_PDISP_HDMI_EMU1); - DUMP_REG(HDMI_NV_PDISP_HDMI_EMU1_RDATA); - DUMP_REG(HDMI_NV_PDISP_HDMI_SPARE); - DUMP_REG(HDMI_NV_PDISP_HDMI_SPDIF_CHN_STATUS1); - DUMP_REG(HDMI_NV_PDISP_HDMI_SPDIF_CHN_STATUS2); - DUMP_REG(HDMI_NV_PDISP_HDMI_HDCPRIF_ROM_CTRL); - DUMP_REG(HDMI_NV_PDISP_SOR_CAP); - DUMP_REG(HDMI_NV_PDISP_SOR_PWR); - DUMP_REG(HDMI_NV_PDISP_SOR_TEST); - DUMP_REG(HDMI_NV_PDISP_SOR_PLL0); - DUMP_REG(HDMI_NV_PDISP_SOR_PLL1); - DUMP_REG(HDMI_NV_PDISP_SOR_PLL2); - DUMP_REG(HDMI_NV_PDISP_SOR_CSTM); - DUMP_REG(HDMI_NV_PDISP_SOR_LVDS); - DUMP_REG(HDMI_NV_PDISP_SOR_CRCA); - DUMP_REG(HDMI_NV_PDISP_SOR_CRCB); - DUMP_REG(HDMI_NV_PDISP_SOR_BLANK); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_CTL); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(0)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(1)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(2)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(3)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(4)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(5)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(6)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(7)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(8)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(9)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(10)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(11)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(12)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(13)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(14)); - DUMP_REG(HDMI_NV_PDISP_SOR_SEQ_INST(15)); - DUMP_REG(HDMI_NV_PDISP_SOR_VCRCA0); - DUMP_REG(HDMI_NV_PDISP_SOR_VCRCA1); - DUMP_REG(HDMI_NV_PDISP_SOR_CCRCA0); - DUMP_REG(HDMI_NV_PDISP_SOR_CCRCA1); - DUMP_REG(HDMI_NV_PDISP_SOR_EDATAA0); - DUMP_REG(HDMI_NV_PDISP_SOR_EDATAA1); - DUMP_REG(HDMI_NV_PDISP_SOR_COUNTA0); - DUMP_REG(HDMI_NV_PDISP_SOR_COUNTA1); - DUMP_REG(HDMI_NV_PDISP_SOR_DEBUGA0); - DUMP_REG(HDMI_NV_PDISP_SOR_DEBUGA1); - DUMP_REG(HDMI_NV_PDISP_SOR_TRIG); - DUMP_REG(HDMI_NV_PDISP_SOR_MSCHECK); - DUMP_REG(HDMI_NV_PDISP_SOR_LANE_DRIVE_CURRENT); - DUMP_REG(HDMI_NV_PDISP_AUDIO_DEBUG0); - DUMP_REG(HDMI_NV_PDISP_AUDIO_DEBUG1); - DUMP_REG(HDMI_NV_PDISP_AUDIO_DEBUG2); - DUMP_REG(HDMI_NV_PDISP_AUDIO_FS(0)); - DUMP_REG(HDMI_NV_PDISP_AUDIO_FS(1)); - DUMP_REG(HDMI_NV_PDISP_AUDIO_FS(2)); - DUMP_REG(HDMI_NV_PDISP_AUDIO_FS(3)); - DUMP_REG(HDMI_NV_PDISP_AUDIO_FS(4)); - DUMP_REG(HDMI_NV_PDISP_AUDIO_FS(5)); - DUMP_REG(HDMI_NV_PDISP_AUDIO_FS(6)); - DUMP_REG(HDMI_NV_PDISP_AUDIO_PULSE_WIDTH); - DUMP_REG(HDMI_NV_PDISP_AUDIO_THRESHOLD); - DUMP_REG(HDMI_NV_PDISP_AUDIO_CNTRL0); - DUMP_REG(HDMI_NV_PDISP_AUDIO_N); - DUMP_REG(HDMI_NV_PDISP_HDCPRIF_ROM_TIMING); - DUMP_REG(HDMI_NV_PDISP_SOR_REFCLK); - DUMP_REG(HDMI_NV_PDISP_CRC_CONTROL); - DUMP_REG(HDMI_NV_PDISP_INPUT_CONTROL); - DUMP_REG(HDMI_NV_PDISP_SCRATCH); - DUMP_REG(HDMI_NV_PDISP_PE_CURRENT); - DUMP_REG(HDMI_NV_PDISP_KEY_CTRL); - DUMP_REG(HDMI_NV_PDISP_KEY_DEBUG0); - DUMP_REG(HDMI_NV_PDISP_KEY_DEBUG1); - DUMP_REG(HDMI_NV_PDISP_KEY_DEBUG2); - DUMP_REG(HDMI_NV_PDISP_KEY_HDCP_KEY_0); - DUMP_REG(HDMI_NV_PDISP_KEY_HDCP_KEY_1); - DUMP_REG(HDMI_NV_PDISP_KEY_HDCP_KEY_2); - DUMP_REG(HDMI_NV_PDISP_KEY_HDCP_KEY_3); - DUMP_REG(HDMI_NV_PDISP_KEY_HDCP_KEY_TRIG); - DUMP_REG(HDMI_NV_PDISP_KEY_SKEY_INDEX); - DUMP_REG(HDMI_NV_PDISP_SOR_AUDIO_CNTRL0); - DUMP_REG(HDMI_NV_PDISP_SOR_AUDIO_HDA_ELD_BUFWR); - DUMP_REG(HDMI_NV_PDISP_SOR_AUDIO_HDA_PRESENSE); - -#undef DUMP_REG - - return 0; -} - -static struct drm_info_list debugfs_files[] = { - { "regs", tegra_hdmi_show_regs, 0, NULL }, -}; - -static int tegra_hdmi_debugfs_init(struct tegra_hdmi *hdmi, - struct drm_minor *minor) -{ - unsigned int i; - int err; - - hdmi->debugfs = debugfs_create_dir("hdmi", minor->debugfs_root); - if (!hdmi->debugfs) - return -ENOMEM; - - hdmi->debugfs_files = kmemdup(debugfs_files, sizeof(debugfs_files), - GFP_KERNEL); - if (!hdmi->debugfs_files) { - err = -ENOMEM; - goto remove; - } - - for (i = 0; i < ARRAY_SIZE(debugfs_files); i++) - hdmi->debugfs_files[i].data = hdmi; - - err = drm_debugfs_create_files(hdmi->debugfs_files, - ARRAY_SIZE(debugfs_files), - hdmi->debugfs, minor); - if (err < 0) - goto free; - - hdmi->minor = minor; - - return 0; - -free: - kfree(hdmi->debugfs_files); - hdmi->debugfs_files = NULL; -remove: - debugfs_remove(hdmi->debugfs); - hdmi->debugfs = NULL; - - return err; -} - -static int tegra_hdmi_debugfs_exit(struct tegra_hdmi *hdmi) -{ - drm_debugfs_remove_files(hdmi->debugfs_files, ARRAY_SIZE(debugfs_files), - hdmi->minor); - hdmi->minor = NULL; - - kfree(hdmi->debugfs_files); - hdmi->debugfs_files = NULL; - - debugfs_remove(hdmi->debugfs); - hdmi->debugfs = NULL; - - return 0; -} - -static int tegra_hdmi_drm_init(struct host1x_client *client, - struct drm_device *drm) -{ - struct tegra_hdmi *hdmi = host1x_client_to_hdmi(client); - int err; - - hdmi->output.type = TEGRA_OUTPUT_HDMI; - hdmi->output.dev = client->dev; - hdmi->output.ops = &hdmi_ops; - - err = tegra_output_init(drm, &hdmi->output); - if (err < 0) { - dev_err(client->dev, "output setup failed: %d\n", err); - return err; - } - - if (IS_ENABLED(CONFIG_DEBUG_FS)) { - err = tegra_hdmi_debugfs_init(hdmi, drm->primary); - if (err < 0) - dev_err(client->dev, "debugfs setup failed: %d\n", err); - } - - return 0; -} - -static int tegra_hdmi_drm_exit(struct host1x_client *client) -{ - struct tegra_hdmi *hdmi = host1x_client_to_hdmi(client); - int err; - - if (IS_ENABLED(CONFIG_DEBUG_FS)) { - err = tegra_hdmi_debugfs_exit(hdmi); - if (err < 0) - dev_err(client->dev, "debugfs cleanup failed: %d\n", - err); - } - - err = tegra_output_disable(&hdmi->output); - if (err < 0) { - dev_err(client->dev, "output failed to disable: %d\n", err); - return err; - } - - err = tegra_output_exit(&hdmi->output); - if (err < 0) { - dev_err(client->dev, "output cleanup failed: %d\n", err); - return err; - } - - return 0; -} - -static const struct host1x_client_ops hdmi_client_ops = { - .drm_init = tegra_hdmi_drm_init, - .drm_exit = tegra_hdmi_drm_exit, -}; - -static int tegra_hdmi_probe(struct platform_device *pdev) -{ - struct host1x_drm *host1x = host1x_get_drm_data(pdev->dev.parent); - struct tegra_hdmi *hdmi; - struct resource *regs; - int err; - - hdmi = devm_kzalloc(&pdev->dev, sizeof(*hdmi), GFP_KERNEL); - if (!hdmi) - return -ENOMEM; - - hdmi->dev = &pdev->dev; - hdmi->audio_source = AUTO; - hdmi->audio_freq = 44100; - hdmi->stereo = false; - hdmi->dvi = false; - - hdmi->clk = devm_clk_get(&pdev->dev, NULL); - if (IS_ERR(hdmi->clk)) { - dev_err(&pdev->dev, "failed to get clock\n"); - return PTR_ERR(hdmi->clk); - } - - err = clk_prepare(hdmi->clk); - if (err < 0) - return err; - - hdmi->clk_parent = devm_clk_get(&pdev->dev, "parent"); - if (IS_ERR(hdmi->clk_parent)) - return PTR_ERR(hdmi->clk_parent); - - err = clk_prepare(hdmi->clk_parent); - if (err < 0) - return err; - - err = clk_set_parent(hdmi->clk, hdmi->clk_parent); - if (err < 0) { - dev_err(&pdev->dev, "failed to setup clocks: %d\n", err); - return err; - } - - hdmi->vdd = devm_regulator_get(&pdev->dev, "vdd"); - if (IS_ERR(hdmi->vdd)) { - dev_err(&pdev->dev, "failed to get VDD regulator\n"); - return PTR_ERR(hdmi->vdd); - } - - hdmi->pll = devm_regulator_get(&pdev->dev, "pll"); - if (IS_ERR(hdmi->pll)) { - dev_err(&pdev->dev, "failed to get PLL regulator\n"); - return PTR_ERR(hdmi->pll); - } - - hdmi->output.dev = &pdev->dev; - - err = tegra_output_parse_dt(&hdmi->output); - if (err < 0) - return err; - - regs = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (!regs) - return -ENXIO; - - hdmi->regs = devm_ioremap_resource(&pdev->dev, regs); - if (IS_ERR(hdmi->regs)) - return PTR_ERR(hdmi->regs); - - err = platform_get_irq(pdev, 0); - if (err < 0) - return err; - - hdmi->irq = err; - - hdmi->client.ops = &hdmi_client_ops; - INIT_LIST_HEAD(&hdmi->client.list); - hdmi->client.dev = &pdev->dev; - - err = host1x_register_client(host1x, &hdmi->client); - if (err < 0) { - dev_err(&pdev->dev, "failed to register host1x client: %d\n", - err); - return err; - } - - platform_set_drvdata(pdev, hdmi); - - return 0; -} - -static int tegra_hdmi_remove(struct platform_device *pdev) -{ - struct host1x_drm *host1x = host1x_get_drm_data(pdev->dev.parent); - struct tegra_hdmi *hdmi = platform_get_drvdata(pdev); - int err; - - err = host1x_unregister_client(host1x, &hdmi->client); - if (err < 0) { - dev_err(&pdev->dev, "failed to unregister host1x client: %d\n", - err); - return err; - } - - clk_unprepare(hdmi->clk_parent); - clk_unprepare(hdmi->clk); - - return 0; -} - -static struct of_device_id tegra_hdmi_of_match[] = { - { .compatible = "nvidia,tegra30-hdmi", }, - { .compatible = "nvidia,tegra20-hdmi", }, - { }, -}; - -struct platform_driver tegra_hdmi_driver = { - .driver = { - .name = "tegra-hdmi", - .owner = THIS_MODULE, - .of_match_table = tegra_hdmi_of_match, - }, - .probe = tegra_hdmi_probe, - .remove = tegra_hdmi_remove, -}; diff --git a/drivers/gpu/host1x/drm/hdmi.h b/drivers/gpu/host1x/drm/hdmi.h deleted file mode 100644 index 52ac36e08ccb..000000000000 --- a/drivers/gpu/host1x/drm/hdmi.h +++ /dev/null @@ -1,386 +0,0 @@ -/* - * Copyright (C) 2012 Avionic Design GmbH - * Copyright (C) 2012 NVIDIA CORPORATION. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#ifndef TEGRA_HDMI_H -#define TEGRA_HDMI_H 1 - -/* register definitions */ -#define HDMI_CTXSW 0x00 - -#define HDMI_NV_PDISP_SOR_STATE0 0x01 -#define SOR_STATE_UPDATE (1 << 0) - -#define HDMI_NV_PDISP_SOR_STATE1 0x02 -#define SOR_STATE_ASY_HEAD_OPMODE_AWAKE (2 << 0) -#define SOR_STATE_ASY_ORMODE_NORMAL (1 << 2) -#define SOR_STATE_ATTACHED (1 << 3) - -#define HDMI_NV_PDISP_SOR_STATE2 0x03 -#define SOR_STATE_ASY_OWNER_NONE (0 << 0) -#define SOR_STATE_ASY_OWNER_HEAD0 (1 << 0) -#define SOR_STATE_ASY_SUBOWNER_NONE (0 << 4) -#define SOR_STATE_ASY_SUBOWNER_SUBHEAD0 (1 << 4) -#define SOR_STATE_ASY_SUBOWNER_SUBHEAD1 (2 << 4) -#define SOR_STATE_ASY_SUBOWNER_BOTH (3 << 4) -#define SOR_STATE_ASY_CRCMODE_ACTIVE (0 << 6) -#define SOR_STATE_ASY_CRCMODE_COMPLETE (1 << 6) -#define SOR_STATE_ASY_CRCMODE_NON_ACTIVE (2 << 6) -#define SOR_STATE_ASY_PROTOCOL_SINGLE_TMDS_A (1 << 8) -#define SOR_STATE_ASY_PROTOCOL_CUSTOM (15 << 8) -#define SOR_STATE_ASY_HSYNCPOL_POS (0 << 12) -#define SOR_STATE_ASY_HSYNCPOL_NEG (1 << 12) -#define SOR_STATE_ASY_VSYNCPOL_POS (0 << 13) -#define SOR_STATE_ASY_VSYNCPOL_NEG (1 << 13) -#define SOR_STATE_ASY_DEPOL_POS (0 << 14) -#define SOR_STATE_ASY_DEPOL_NEG (1 << 14) - -#define HDMI_NV_PDISP_RG_HDCP_AN_MSB 0x04 -#define HDMI_NV_PDISP_RG_HDCP_AN_LSB 0x05 -#define HDMI_NV_PDISP_RG_HDCP_CN_MSB 0x06 -#define HDMI_NV_PDISP_RG_HDCP_CN_LSB 0x07 -#define HDMI_NV_PDISP_RG_HDCP_AKSV_MSB 0x08 -#define HDMI_NV_PDISP_RG_HDCP_AKSV_LSB 0x09 -#define HDMI_NV_PDISP_RG_HDCP_BKSV_MSB 0x0a -#define HDMI_NV_PDISP_RG_HDCP_BKSV_LSB 0x0b -#define HDMI_NV_PDISP_RG_HDCP_CKSV_MSB 0x0c -#define HDMI_NV_PDISP_RG_HDCP_CKSV_LSB 0x0d -#define HDMI_NV_PDISP_RG_HDCP_DKSV_MSB 0x0e -#define HDMI_NV_PDISP_RG_HDCP_DKSV_LSB 0x0f -#define HDMI_NV_PDISP_RG_HDCP_CTRL 0x10 -#define HDMI_NV_PDISP_RG_HDCP_CMODE 0x11 -#define HDMI_NV_PDISP_RG_HDCP_MPRIME_MSB 0x12 -#define HDMI_NV_PDISP_RG_HDCP_MPRIME_LSB 0x13 -#define HDMI_NV_PDISP_RG_HDCP_SPRIME_MSB 0x14 -#define HDMI_NV_PDISP_RG_HDCP_SPRIME_LSB2 0x15 -#define HDMI_NV_PDISP_RG_HDCP_SPRIME_LSB1 0x16 -#define HDMI_NV_PDISP_RG_HDCP_RI 0x17 -#define HDMI_NV_PDISP_RG_HDCP_CS_MSB 0x18 -#define HDMI_NV_PDISP_RG_HDCP_CS_LSB 0x19 -#define HDMI_NV_PDISP_HDMI_AUDIO_EMU0 0x1a -#define HDMI_NV_PDISP_HDMI_AUDIO_EMU_RDATA0 0x1b -#define HDMI_NV_PDISP_HDMI_AUDIO_EMU1 0x1c -#define HDMI_NV_PDISP_HDMI_AUDIO_EMU2 0x1d - -#define HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_CTRL 0x1e -#define HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_STATUS 0x1f -#define HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_HEADER 0x20 -#define HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_SUBPACK0_LOW 0x21 -#define HDMI_NV_PDISP_HDMI_AUDIO_INFOFRAME_SUBPACK0_HIGH 0x22 -#define HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_CTRL 0x23 -#define HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_STATUS 0x24 -#define HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_HEADER 0x25 -#define HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_SUBPACK0_LOW 0x26 -#define HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_SUBPACK0_HIGH 0x27 -#define HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_SUBPACK1_LOW 0x28 -#define HDMI_NV_PDISP_HDMI_AVI_INFOFRAME_SUBPACK1_HIGH 0x29 - -#define INFOFRAME_CTRL_ENABLE (1 << 0) - -#define INFOFRAME_HEADER_TYPE(x) (((x) & 0xff) << 0) -#define INFOFRAME_HEADER_VERSION(x) (((x) & 0xff) << 8) -#define INFOFRAME_HEADER_LEN(x) (((x) & 0x0f) << 16) - -#define HDMI_NV_PDISP_HDMI_GENERIC_CTRL 0x2a -#define GENERIC_CTRL_ENABLE (1 << 0) -#define GENERIC_CTRL_OTHER (1 << 4) -#define GENERIC_CTRL_SINGLE (1 << 8) -#define GENERIC_CTRL_HBLANK (1 << 12) -#define GENERIC_CTRL_AUDIO (1 << 16) - -#define HDMI_NV_PDISP_HDMI_GENERIC_STATUS 0x2b -#define HDMI_NV_PDISP_HDMI_GENERIC_HEADER 0x2c -#define HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK0_LOW 0x2d -#define HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK0_HIGH 0x2e -#define HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK1_LOW 0x2f -#define HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK1_HIGH 0x30 -#define HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK2_LOW 0x31 -#define HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK2_HIGH 0x32 -#define HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK3_LOW 0x33 -#define HDMI_NV_PDISP_HDMI_GENERIC_SUBPACK3_HIGH 0x34 - -#define HDMI_NV_PDISP_HDMI_ACR_CTRL 0x35 -#define HDMI_NV_PDISP_HDMI_ACR_0320_SUBPACK_LOW 0x36 -#define HDMI_NV_PDISP_HDMI_ACR_0320_SUBPACK_HIGH 0x37 -#define HDMI_NV_PDISP_HDMI_ACR_0441_SUBPACK_LOW 0x38 -#define HDMI_NV_PDISP_HDMI_ACR_0441_SUBPACK_HIGH 0x39 -#define HDMI_NV_PDISP_HDMI_ACR_0882_SUBPACK_LOW 0x3a -#define HDMI_NV_PDISP_HDMI_ACR_0882_SUBPACK_HIGH 0x3b -#define HDMI_NV_PDISP_HDMI_ACR_1764_SUBPACK_LOW 0x3c -#define HDMI_NV_PDISP_HDMI_ACR_1764_SUBPACK_HIGH 0x3d -#define HDMI_NV_PDISP_HDMI_ACR_0480_SUBPACK_LOW 0x3e -#define HDMI_NV_PDISP_HDMI_ACR_0480_SUBPACK_HIGH 0x3f -#define HDMI_NV_PDISP_HDMI_ACR_0960_SUBPACK_LOW 0x40 -#define HDMI_NV_PDISP_HDMI_ACR_0960_SUBPACK_HIGH 0x41 -#define HDMI_NV_PDISP_HDMI_ACR_1920_SUBPACK_LOW 0x42 -#define HDMI_NV_PDISP_HDMI_ACR_1920_SUBPACK_HIGH 0x43 - -#define ACR_SUBPACK_CTS(x) (((x) & 0xffffff) << 8) -#define ACR_SUBPACK_N(x) (((x) & 0xffffff) << 0) -#define ACR_ENABLE (1 << 31) - -#define HDMI_NV_PDISP_HDMI_CTRL 0x44 -#define HDMI_CTRL_REKEY(x) (((x) & 0x7f) << 0) -#define HDMI_CTRL_MAX_AC_PACKET(x) (((x) & 0x1f) << 16) -#define HDMI_CTRL_ENABLE (1 << 30) - -#define HDMI_NV_PDISP_HDMI_VSYNC_KEEPOUT 0x45 -#define HDMI_NV_PDISP_HDMI_VSYNC_WINDOW 0x46 -#define VSYNC_WINDOW_END(x) (((x) & 0x3ff) << 0) -#define VSYNC_WINDOW_START(x) (((x) & 0x3ff) << 16) -#define VSYNC_WINDOW_ENABLE (1 << 31) - -#define HDMI_NV_PDISP_HDMI_GCP_CTRL 0x47 -#define HDMI_NV_PDISP_HDMI_GCP_STATUS 0x48 -#define HDMI_NV_PDISP_HDMI_GCP_SUBPACK 0x49 -#define HDMI_NV_PDISP_HDMI_CHANNEL_STATUS1 0x4a -#define HDMI_NV_PDISP_HDMI_CHANNEL_STATUS2 0x4b -#define HDMI_NV_PDISP_HDMI_EMU0 0x4c -#define HDMI_NV_PDISP_HDMI_EMU1 0x4d -#define HDMI_NV_PDISP_HDMI_EMU1_RDATA 0x4e - -#define HDMI_NV_PDISP_HDMI_SPARE 0x4f -#define SPARE_HW_CTS (1 << 0) -#define SPARE_FORCE_SW_CTS (1 << 1) -#define SPARE_CTS_RESET_VAL(x) (((x) & 0x7) << 16) - -#define HDMI_NV_PDISP_HDMI_SPDIF_CHN_STATUS1 0x50 -#define HDMI_NV_PDISP_HDMI_SPDIF_CHN_STATUS2 0x51 -#define HDMI_NV_PDISP_HDMI_HDCPRIF_ROM_CTRL 0x53 -#define HDMI_NV_PDISP_SOR_CAP 0x54 -#define HDMI_NV_PDISP_SOR_PWR 0x55 -#define SOR_PWR_NORMAL_STATE_PD (0 << 0) -#define SOR_PWR_NORMAL_STATE_PU (1 << 0) -#define SOR_PWR_NORMAL_START_NORMAL (0 << 1) -#define SOR_PWR_NORMAL_START_ALT (1 << 1) -#define SOR_PWR_SAFE_STATE_PD (0 << 16) -#define SOR_PWR_SAFE_STATE_PU (1 << 16) -#define SOR_PWR_SETTING_NEW_DONE (0 << 31) -#define SOR_PWR_SETTING_NEW_PENDING (1 << 31) -#define SOR_PWR_SETTING_NEW_TRIGGER (1 << 31) - -#define HDMI_NV_PDISP_SOR_TEST 0x56 -#define HDMI_NV_PDISP_SOR_PLL0 0x57 -#define SOR_PLL_PWR (1 << 0) -#define SOR_PLL_PDBG (1 << 1) -#define SOR_PLL_VCAPD (1 << 2) -#define SOR_PLL_PDPORT (1 << 3) -#define SOR_PLL_RESISTORSEL (1 << 4) -#define SOR_PLL_PULLDOWN (1 << 5) -#define SOR_PLL_VCOCAP(x) (((x) & 0xf) << 8) -#define SOR_PLL_BG_V17_S(x) (((x) & 0xf) << 12) -#define SOR_PLL_FILTER(x) (((x) & 0xf) << 16) -#define SOR_PLL_ICHPMP(x) (((x) & 0xf) << 24) -#define SOR_PLL_TX_REG_LOAD(x) (((x) & 0xf) << 28) - -#define HDMI_NV_PDISP_SOR_PLL1 0x58 -#define SOR_PLL_TMDS_TERM_ENABLE (1 << 8) -#define SOR_PLL_TMDS_TERMADJ(x) (((x) & 0xf) << 9) -#define SOR_PLL_LOADADJ(x) (((x) & 0xf) << 20) -#define SOR_PLL_PE_EN (1 << 28) -#define SOR_PLL_HALF_FULL_PE (1 << 29) -#define SOR_PLL_S_D_PIN_PE (1 << 30) - -#define HDMI_NV_PDISP_SOR_PLL2 0x59 - -#define HDMI_NV_PDISP_SOR_CSTM 0x5a -#define SOR_CSTM_ROTCLK(x) (((x) & 0xf) << 24) - -#define HDMI_NV_PDISP_SOR_LVDS 0x5b -#define HDMI_NV_PDISP_SOR_CRCA 0x5c -#define HDMI_NV_PDISP_SOR_CRCB 0x5d -#define HDMI_NV_PDISP_SOR_BLANK 0x5e -#define HDMI_NV_PDISP_SOR_SEQ_CTL 0x5f -#define SOR_SEQ_CTL_PU_PC(x) (((x) & 0xf) << 0) -#define SOR_SEQ_PU_PC_ALT(x) (((x) & 0xf) << 4) -#define SOR_SEQ_PD_PC(x) (((x) & 0xf) << 8) -#define SOR_SEQ_PD_PC_ALT(x) (((x) & 0xf) << 12) -#define SOR_SEQ_PC(x) (((x) & 0xf) << 16) -#define SOR_SEQ_STATUS (1 << 28) -#define SOR_SEQ_SWITCH (1 << 30) - -#define HDMI_NV_PDISP_SOR_SEQ_INST(x) (0x60 + (x)) - -#define SOR_SEQ_INST_WAIT_TIME(x) (((x) & 0x3ff) << 0) -#define SOR_SEQ_INST_WAIT_UNITS_VSYNC (2 << 12) -#define SOR_SEQ_INST_HALT (1 << 15) -#define SOR_SEQ_INST_PIN_A_LOW (0 << 21) -#define SOR_SEQ_INST_PIN_A_HIGH (1 << 21) -#define SOR_SEQ_INST_PIN_B_LOW (0 << 22) -#define SOR_SEQ_INST_PIN_B_HIGH (1 << 22) -#define SOR_SEQ_INST_DRIVE_PWM_OUT_LO (1 << 23) - -#define HDMI_NV_PDISP_SOR_VCRCA0 0x72 -#define HDMI_NV_PDISP_SOR_VCRCA1 0x73 -#define HDMI_NV_PDISP_SOR_CCRCA0 0x74 -#define HDMI_NV_PDISP_SOR_CCRCA1 0x75 -#define HDMI_NV_PDISP_SOR_EDATAA0 0x76 -#define HDMI_NV_PDISP_SOR_EDATAA1 0x77 -#define HDMI_NV_PDISP_SOR_COUNTA0 0x78 -#define HDMI_NV_PDISP_SOR_COUNTA1 0x79 -#define HDMI_NV_PDISP_SOR_DEBUGA0 0x7a -#define HDMI_NV_PDISP_SOR_DEBUGA1 0x7b -#define HDMI_NV_PDISP_SOR_TRIG 0x7c -#define HDMI_NV_PDISP_SOR_MSCHECK 0x7d - -#define HDMI_NV_PDISP_SOR_LANE_DRIVE_CURRENT 0x7e -#define DRIVE_CURRENT_LANE0(x) (((x) & 0x3f) << 0) -#define DRIVE_CURRENT_LANE1(x) (((x) & 0x3f) << 8) -#define DRIVE_CURRENT_LANE2(x) (((x) & 0x3f) << 16) -#define DRIVE_CURRENT_LANE3(x) (((x) & 0x3f) << 24) -#define DRIVE_CURRENT_FUSE_OVERRIDE (1 << 31) - -#define DRIVE_CURRENT_1_500_mA 0x00 -#define DRIVE_CURRENT_1_875_mA 0x01 -#define DRIVE_CURRENT_2_250_mA 0x02 -#define DRIVE_CURRENT_2_625_mA 0x03 -#define DRIVE_CURRENT_3_000_mA 0x04 -#define DRIVE_CURRENT_3_375_mA 0x05 -#define DRIVE_CURRENT_3_750_mA 0x06 -#define DRIVE_CURRENT_4_125_mA 0x07 -#define DRIVE_CURRENT_4_500_mA 0x08 -#define DRIVE_CURRENT_4_875_mA 0x09 -#define DRIVE_CURRENT_5_250_mA 0x0a -#define DRIVE_CURRENT_5_625_mA 0x0b -#define DRIVE_CURRENT_6_000_mA 0x0c -#define DRIVE_CURRENT_6_375_mA 0x0d -#define DRIVE_CURRENT_6_750_mA 0x0e -#define DRIVE_CURRENT_7_125_mA 0x0f -#define DRIVE_CURRENT_7_500_mA 0x10 -#define DRIVE_CURRENT_7_875_mA 0x11 -#define DRIVE_CURRENT_8_250_mA 0x12 -#define DRIVE_CURRENT_8_625_mA 0x13 -#define DRIVE_CURRENT_9_000_mA 0x14 -#define DRIVE_CURRENT_9_375_mA 0x15 -#define DRIVE_CURRENT_9_750_mA 0x16 -#define DRIVE_CURRENT_10_125_mA 0x17 -#define DRIVE_CURRENT_10_500_mA 0x18 -#define DRIVE_CURRENT_10_875_mA 0x19 -#define DRIVE_CURRENT_11_250_mA 0x1a -#define DRIVE_CURRENT_11_625_mA 0x1b -#define DRIVE_CURRENT_12_000_mA 0x1c -#define DRIVE_CURRENT_12_375_mA 0x1d -#define DRIVE_CURRENT_12_750_mA 0x1e -#define DRIVE_CURRENT_13_125_mA 0x1f -#define DRIVE_CURRENT_13_500_mA 0x20 -#define DRIVE_CURRENT_13_875_mA 0x21 -#define DRIVE_CURRENT_14_250_mA 0x22 -#define DRIVE_CURRENT_14_625_mA 0x23 -#define DRIVE_CURRENT_15_000_mA 0x24 -#define DRIVE_CURRENT_15_375_mA 0x25 -#define DRIVE_CURRENT_15_750_mA 0x26 -#define DRIVE_CURRENT_16_125_mA 0x27 -#define DRIVE_CURRENT_16_500_mA 0x28 -#define DRIVE_CURRENT_16_875_mA 0x29 -#define DRIVE_CURRENT_17_250_mA 0x2a -#define DRIVE_CURRENT_17_625_mA 0x2b -#define DRIVE_CURRENT_18_000_mA 0x2c -#define DRIVE_CURRENT_18_375_mA 0x2d -#define DRIVE_CURRENT_18_750_mA 0x2e -#define DRIVE_CURRENT_19_125_mA 0x2f -#define DRIVE_CURRENT_19_500_mA 0x30 -#define DRIVE_CURRENT_19_875_mA 0x31 -#define DRIVE_CURRENT_20_250_mA 0x32 -#define DRIVE_CURRENT_20_625_mA 0x33 -#define DRIVE_CURRENT_21_000_mA 0x34 -#define DRIVE_CURRENT_21_375_mA 0x35 -#define DRIVE_CURRENT_21_750_mA 0x36 -#define DRIVE_CURRENT_22_125_mA 0x37 -#define DRIVE_CURRENT_22_500_mA 0x38 -#define DRIVE_CURRENT_22_875_mA 0x39 -#define DRIVE_CURRENT_23_250_mA 0x3a -#define DRIVE_CURRENT_23_625_mA 0x3b -#define DRIVE_CURRENT_24_000_mA 0x3c -#define DRIVE_CURRENT_24_375_mA 0x3d -#define DRIVE_CURRENT_24_750_mA 0x3e - -#define HDMI_NV_PDISP_AUDIO_DEBUG0 0x7f -#define HDMI_NV_PDISP_AUDIO_DEBUG1 0x80 -#define HDMI_NV_PDISP_AUDIO_DEBUG2 0x81 - -#define HDMI_NV_PDISP_AUDIO_FS(x) (0x82 + (x)) -#define AUDIO_FS_LOW(x) (((x) & 0xfff) << 0) -#define AUDIO_FS_HIGH(x) (((x) & 0xfff) << 16) - -#define HDMI_NV_PDISP_AUDIO_PULSE_WIDTH 0x89 -#define HDMI_NV_PDISP_AUDIO_THRESHOLD 0x8a -#define HDMI_NV_PDISP_AUDIO_CNTRL0 0x8b -#define AUDIO_CNTRL0_ERROR_TOLERANCE(x) (((x) & 0xff) << 0) -#define AUDIO_CNTRL0_SOURCE_SELECT_AUTO (0 << 20) -#define AUDIO_CNTRL0_SOURCE_SELECT_SPDIF (1 << 20) -#define AUDIO_CNTRL0_SOURCE_SELECT_HDAL (2 << 20) -#define AUDIO_CNTRL0_FRAMES_PER_BLOCK(x) (((x) & 0xff) << 24) - -#define HDMI_NV_PDISP_AUDIO_N 0x8c -#define AUDIO_N_VALUE(x) (((x) & 0xfffff) << 0) -#define AUDIO_N_RESETF (1 << 20) -#define AUDIO_N_GENERATE_NORMAL (0 << 24) -#define AUDIO_N_GENERATE_ALTERNATE (1 << 24) - -#define HDMI_NV_PDISP_HDCPRIF_ROM_TIMING 0x94 -#define HDMI_NV_PDISP_SOR_REFCLK 0x95 -#define SOR_REFCLK_DIV_INT(x) (((x) & 0xff) << 8) -#define SOR_REFCLK_DIV_FRAC(x) (((x) & 0x03) << 6) - -#define HDMI_NV_PDISP_CRC_CONTROL 0x96 -#define HDMI_NV_PDISP_INPUT_CONTROL 0x97 -#define HDMI_SRC_DISPLAYA (0 << 0) -#define HDMI_SRC_DISPLAYB (1 << 0) -#define ARM_VIDEO_RANGE_FULL (0 << 1) -#define ARM_VIDEO_RANGE_LIMITED (1 << 1) - -#define HDMI_NV_PDISP_SCRATCH 0x98 -#define HDMI_NV_PDISP_PE_CURRENT 0x99 -#define PE_CURRENT0(x) (((x) & 0xf) << 0) -#define PE_CURRENT1(x) (((x) & 0xf) << 8) -#define PE_CURRENT2(x) (((x) & 0xf) << 16) -#define PE_CURRENT3(x) (((x) & 0xf) << 24) - -#define PE_CURRENT_0_0_mA 0x0 -#define PE_CURRENT_0_5_mA 0x1 -#define PE_CURRENT_1_0_mA 0x2 -#define PE_CURRENT_1_5_mA 0x3 -#define PE_CURRENT_2_0_mA 0x4 -#define PE_CURRENT_2_5_mA 0x5 -#define PE_CURRENT_3_0_mA 0x6 -#define PE_CURRENT_3_5_mA 0x7 -#define PE_CURRENT_4_0_mA 0x8 -#define PE_CURRENT_4_5_mA 0x9 -#define PE_CURRENT_5_0_mA 0xa -#define PE_CURRENT_5_5_mA 0xb -#define PE_CURRENT_6_0_mA 0xc -#define PE_CURRENT_6_5_mA 0xd -#define PE_CURRENT_7_0_mA 0xe -#define PE_CURRENT_7_5_mA 0xf - -#define HDMI_NV_PDISP_KEY_CTRL 0x9a -#define HDMI_NV_PDISP_KEY_DEBUG0 0x9b -#define HDMI_NV_PDISP_KEY_DEBUG1 0x9c -#define HDMI_NV_PDISP_KEY_DEBUG2 0x9d -#define HDMI_NV_PDISP_KEY_HDCP_KEY_0 0x9e -#define HDMI_NV_PDISP_KEY_HDCP_KEY_1 0x9f -#define HDMI_NV_PDISP_KEY_HDCP_KEY_2 0xa0 -#define HDMI_NV_PDISP_KEY_HDCP_KEY_3 0xa1 -#define HDMI_NV_PDISP_KEY_HDCP_KEY_TRIG 0xa2 -#define HDMI_NV_PDISP_KEY_SKEY_INDEX 0xa3 - -#define HDMI_NV_PDISP_SOR_AUDIO_CNTRL0 0xac -#define AUDIO_CNTRL0_INJECT_NULLSMPL (1 << 29) -#define HDMI_NV_PDISP_SOR_AUDIO_HDA_ELD_BUFWR 0xbc -#define HDMI_NV_PDISP_SOR_AUDIO_HDA_PRESENSE 0xbd - -#define HDMI_NV_PDISP_SOR_AUDIO_AVAL_0320 0xbf -#define HDMI_NV_PDISP_SOR_AUDIO_AVAL_0441 0xc0 -#define HDMI_NV_PDISP_SOR_AUDIO_AVAL_0882 0xc1 -#define HDMI_NV_PDISP_SOR_AUDIO_AVAL_1764 0xc2 -#define HDMI_NV_PDISP_SOR_AUDIO_AVAL_0480 0xc3 -#define HDMI_NV_PDISP_SOR_AUDIO_AVAL_0960 0xc4 -#define HDMI_NV_PDISP_SOR_AUDIO_AVAL_1920 0xc5 -#define HDMI_NV_PDISP_SOR_AUDIO_AVAL_DEFAULT 0xc5 - -#endif /* TEGRA_HDMI_H */ diff --git a/drivers/gpu/host1x/drm/output.c b/drivers/gpu/host1x/drm/output.c deleted file mode 100644 index 8140fc6c34d8..000000000000 --- a/drivers/gpu/host1x/drm/output.c +++ /dev/null @@ -1,272 +0,0 @@ -/* - * Copyright (C) 2012 Avionic Design GmbH - * Copyright (C) 2012 NVIDIA CORPORATION. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#include <linux/module.h> -#include <linux/of_gpio.h> -#include <linux/of_i2c.h> - -#include "drm.h" - -static int tegra_connector_get_modes(struct drm_connector *connector) -{ - struct tegra_output *output = connector_to_output(connector); - struct edid *edid = NULL; - int err = 0; - - if (output->edid) - edid = kmemdup(output->edid, sizeof(*edid), GFP_KERNEL); - else if (output->ddc) - edid = drm_get_edid(connector, output->ddc); - - drm_mode_connector_update_edid_property(connector, edid); - - if (edid) { - err = drm_add_edid_modes(connector, edid); - kfree(edid); - } - - return err; -} - -static int tegra_connector_mode_valid(struct drm_connector *connector, - struct drm_display_mode *mode) -{ - struct tegra_output *output = connector_to_output(connector); - enum drm_mode_status status = MODE_OK; - int err; - - err = tegra_output_check_mode(output, mode, &status); - if (err < 0) - return MODE_ERROR; - - return status; -} - -static struct drm_encoder * -tegra_connector_best_encoder(struct drm_connector *connector) -{ - struct tegra_output *output = connector_to_output(connector); - - return &output->encoder; -} - -static const struct drm_connector_helper_funcs connector_helper_funcs = { - .get_modes = tegra_connector_get_modes, - .mode_valid = tegra_connector_mode_valid, - .best_encoder = tegra_connector_best_encoder, -}; - -static enum drm_connector_status -tegra_connector_detect(struct drm_connector *connector, bool force) -{ - struct tegra_output *output = connector_to_output(connector); - enum drm_connector_status status = connector_status_unknown; - - if (gpio_is_valid(output->hpd_gpio)) { - if (gpio_get_value(output->hpd_gpio) == 0) - status = connector_status_disconnected; - else - status = connector_status_connected; - } else { - if (connector->connector_type == DRM_MODE_CONNECTOR_LVDS) - status = connector_status_connected; - } - - return status; -} - -static void tegra_connector_destroy(struct drm_connector *connector) -{ - drm_sysfs_connector_remove(connector); - drm_connector_cleanup(connector); -} - -static const struct drm_connector_funcs connector_funcs = { - .dpms = drm_helper_connector_dpms, - .detect = tegra_connector_detect, - .fill_modes = drm_helper_probe_single_connector_modes, - .destroy = tegra_connector_destroy, -}; - -static void tegra_encoder_destroy(struct drm_encoder *encoder) -{ - drm_encoder_cleanup(encoder); -} - -static const struct drm_encoder_funcs encoder_funcs = { - .destroy = tegra_encoder_destroy, -}; - -static void tegra_encoder_dpms(struct drm_encoder *encoder, int mode) -{ -} - -static bool tegra_encoder_mode_fixup(struct drm_encoder *encoder, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted) -{ - return true; -} - -static void tegra_encoder_prepare(struct drm_encoder *encoder) -{ -} - -static void tegra_encoder_commit(struct drm_encoder *encoder) -{ -} - -static void tegra_encoder_mode_set(struct drm_encoder *encoder, - struct drm_display_mode *mode, - struct drm_display_mode *adjusted) -{ - struct tegra_output *output = encoder_to_output(encoder); - int err; - - err = tegra_output_enable(output); - if (err < 0) - dev_err(encoder->dev->dev, "tegra_output_enable(): %d\n", err); -} - -static const struct drm_encoder_helper_funcs encoder_helper_funcs = { - .dpms = tegra_encoder_dpms, - .mode_fixup = tegra_encoder_mode_fixup, - .prepare = tegra_encoder_prepare, - .commit = tegra_encoder_commit, - .mode_set = tegra_encoder_mode_set, -}; - -static irqreturn_t hpd_irq(int irq, void *data) -{ - struct tegra_output *output = data; - - drm_helper_hpd_irq_event(output->connector.dev); - - return IRQ_HANDLED; -} - -int tegra_output_parse_dt(struct tegra_output *output) -{ - enum of_gpio_flags flags; - struct device_node *ddc; - size_t size; - int err; - - if (!output->of_node) - output->of_node = output->dev->of_node; - - output->edid = of_get_property(output->of_node, "nvidia,edid", &size); - - ddc = of_parse_phandle(output->of_node, "nvidia,ddc-i2c-bus", 0); - if (ddc) { - output->ddc = of_find_i2c_adapter_by_node(ddc); - if (!output->ddc) { - err = -EPROBE_DEFER; - of_node_put(ddc); - return err; - } - - of_node_put(ddc); - } - - if (!output->edid && !output->ddc) - return -ENODEV; - - output->hpd_gpio = of_get_named_gpio_flags(output->of_node, - "nvidia,hpd-gpio", 0, - &flags); - - return 0; -} - -int tegra_output_init(struct drm_device *drm, struct tegra_output *output) -{ - int connector, encoder, err; - - if (gpio_is_valid(output->hpd_gpio)) { - unsigned long flags; - - err = gpio_request_one(output->hpd_gpio, GPIOF_DIR_IN, - "HDMI hotplug detect"); - if (err < 0) { - dev_err(output->dev, "gpio_request_one(): %d\n", err); - return err; - } - - err = gpio_to_irq(output->hpd_gpio); - if (err < 0) { - dev_err(output->dev, "gpio_to_irq(): %d\n", err); - goto free_hpd; - } - - output->hpd_irq = err; - - flags = IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING | - IRQF_ONESHOT; - - err = request_threaded_irq(output->hpd_irq, NULL, hpd_irq, - flags, "hpd", output); - if (err < 0) { - dev_err(output->dev, "failed to request IRQ#%u: %d\n", - output->hpd_irq, err); - goto free_hpd; - } - - output->connector.polled = DRM_CONNECTOR_POLL_HPD; - } - - switch (output->type) { - case TEGRA_OUTPUT_RGB: - connector = DRM_MODE_CONNECTOR_LVDS; - encoder = DRM_MODE_ENCODER_LVDS; - break; - - case TEGRA_OUTPUT_HDMI: - connector = DRM_MODE_CONNECTOR_HDMIA; - encoder = DRM_MODE_ENCODER_TMDS; - break; - - default: - connector = DRM_MODE_CONNECTOR_Unknown; - encoder = DRM_MODE_ENCODER_NONE; - break; - } - - drm_connector_init(drm, &output->connector, &connector_funcs, - connector); - drm_connector_helper_add(&output->connector, &connector_helper_funcs); - - drm_encoder_init(drm, &output->encoder, &encoder_funcs, encoder); - drm_encoder_helper_add(&output->encoder, &encoder_helper_funcs); - - drm_mode_connector_attach_encoder(&output->connector, &output->encoder); - drm_sysfs_connector_add(&output->connector); - - output->encoder.possible_crtcs = 0x3; - - return 0; - -free_hpd: - gpio_free(output->hpd_gpio); - - return err; -} - -int tegra_output_exit(struct tegra_output *output) -{ - if (gpio_is_valid(output->hpd_gpio)) { - free_irq(output->hpd_irq, output); - gpio_free(output->hpd_gpio); - } - - if (output->ddc) - put_device(&output->ddc->dev); - - return 0; -} diff --git a/drivers/gpu/host1x/drm/rgb.c b/drivers/gpu/host1x/drm/rgb.c deleted file mode 100644 index ed4416f20260..000000000000 --- a/drivers/gpu/host1x/drm/rgb.c +++ /dev/null @@ -1,228 +0,0 @@ -/* - * Copyright (C) 2012 Avionic Design GmbH - * Copyright (C) 2012 NVIDIA CORPORATION. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#include <linux/clk.h> -#include <linux/module.h> -#include <linux/of.h> -#include <linux/platform_device.h> - -#include "drm.h" -#include "dc.h" - -struct tegra_rgb { - struct tegra_output output; - struct clk *clk_parent; - struct clk *clk; -}; - -static inline struct tegra_rgb *to_rgb(struct tegra_output *output) -{ - return container_of(output, struct tegra_rgb, output); -} - -struct reg_entry { - unsigned long offset; - unsigned long value; -}; - -static const struct reg_entry rgb_enable[] = { - { DC_COM_PIN_OUTPUT_ENABLE(0), 0x00000000 }, - { DC_COM_PIN_OUTPUT_ENABLE(1), 0x00000000 }, - { DC_COM_PIN_OUTPUT_ENABLE(2), 0x00000000 }, - { DC_COM_PIN_OUTPUT_ENABLE(3), 0x00000000 }, - { DC_COM_PIN_OUTPUT_POLARITY(0), 0x00000000 }, - { DC_COM_PIN_OUTPUT_POLARITY(1), 0x01000000 }, - { DC_COM_PIN_OUTPUT_POLARITY(2), 0x00000000 }, - { DC_COM_PIN_OUTPUT_POLARITY(3), 0x00000000 }, - { DC_COM_PIN_OUTPUT_DATA(0), 0x00000000 }, - { DC_COM_PIN_OUTPUT_DATA(1), 0x00000000 }, - { DC_COM_PIN_OUTPUT_DATA(2), 0x00000000 }, - { DC_COM_PIN_OUTPUT_DATA(3), 0x00000000 }, - { DC_COM_PIN_OUTPUT_SELECT(0), 0x00000000 }, - { DC_COM_PIN_OUTPUT_SELECT(1), 0x00000000 }, - { DC_COM_PIN_OUTPUT_SELECT(2), 0x00000000 }, - { DC_COM_PIN_OUTPUT_SELECT(3), 0x00000000 }, - { DC_COM_PIN_OUTPUT_SELECT(4), 0x00210222 }, - { DC_COM_PIN_OUTPUT_SELECT(5), 0x00002200 }, - { DC_COM_PIN_OUTPUT_SELECT(6), 0x00020000 }, -}; - -static const struct reg_entry rgb_disable[] = { - { DC_COM_PIN_OUTPUT_SELECT(6), 0x00000000 }, - { DC_COM_PIN_OUTPUT_SELECT(5), 0x00000000 }, - { DC_COM_PIN_OUTPUT_SELECT(4), 0x00000000 }, - { DC_COM_PIN_OUTPUT_SELECT(3), 0x00000000 }, - { DC_COM_PIN_OUTPUT_SELECT(2), 0x00000000 }, - { DC_COM_PIN_OUTPUT_SELECT(1), 0x00000000 }, - { DC_COM_PIN_OUTPUT_SELECT(0), 0x00000000 }, - { DC_COM_PIN_OUTPUT_DATA(3), 0xaaaaaaaa }, - { DC_COM_PIN_OUTPUT_DATA(2), 0xaaaaaaaa }, - { DC_COM_PIN_OUTPUT_DATA(1), 0xaaaaaaaa }, - { DC_COM_PIN_OUTPUT_DATA(0), 0xaaaaaaaa }, - { DC_COM_PIN_OUTPUT_POLARITY(3), 0x00000000 }, - { DC_COM_PIN_OUTPUT_POLARITY(2), 0x00000000 }, - { DC_COM_PIN_OUTPUT_POLARITY(1), 0x00000000 }, - { DC_COM_PIN_OUTPUT_POLARITY(0), 0x00000000 }, - { DC_COM_PIN_OUTPUT_ENABLE(3), 0x55555555 }, - { DC_COM_PIN_OUTPUT_ENABLE(2), 0x55555555 }, - { DC_COM_PIN_OUTPUT_ENABLE(1), 0x55150005 }, - { DC_COM_PIN_OUTPUT_ENABLE(0), 0x55555555 }, -}; - -static void tegra_dc_write_regs(struct tegra_dc *dc, - const struct reg_entry *table, - unsigned int num) -{ - unsigned int i; - - for (i = 0; i < num; i++) - tegra_dc_writel(dc, table[i].value, table[i].offset); -} - -static int tegra_output_rgb_enable(struct tegra_output *output) -{ - struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc); - - tegra_dc_write_regs(dc, rgb_enable, ARRAY_SIZE(rgb_enable)); - - return 0; -} - -static int tegra_output_rgb_disable(struct tegra_output *output) -{ - struct tegra_dc *dc = to_tegra_dc(output->encoder.crtc); - - tegra_dc_write_regs(dc, rgb_disable, ARRAY_SIZE(rgb_disable)); - - return 0; -} - -static int tegra_output_rgb_setup_clock(struct tegra_output *output, - struct clk *clk, unsigned long pclk) -{ - struct tegra_rgb *rgb = to_rgb(output); - - return clk_set_parent(clk, rgb->clk_parent); -} - -static int tegra_output_rgb_check_mode(struct tegra_output *output, - struct drm_display_mode *mode, - enum drm_mode_status *status) -{ - /* - * FIXME: For now, always assume that the mode is okay. There are - * unresolved issues with clk_round_rate(), which doesn't always - * reliably report whether a frequency can be set or not. - */ - - *status = MODE_OK; - - return 0; -} - -static const struct tegra_output_ops rgb_ops = { - .enable = tegra_output_rgb_enable, - .disable = tegra_output_rgb_disable, - .setup_clock = tegra_output_rgb_setup_clock, - .check_mode = tegra_output_rgb_check_mode, -}; - -int tegra_dc_rgb_probe(struct tegra_dc *dc) -{ - struct device_node *np; - struct tegra_rgb *rgb; - int err; - - np = of_get_child_by_name(dc->dev->of_node, "rgb"); - if (!np || !of_device_is_available(np)) - return -ENODEV; - - rgb = devm_kzalloc(dc->dev, sizeof(*rgb), GFP_KERNEL); - if (!rgb) - return -ENOMEM; - - rgb->clk = devm_clk_get(dc->dev, NULL); - if (IS_ERR(rgb->clk)) { - dev_err(dc->dev, "failed to get clock\n"); - return PTR_ERR(rgb->clk); - } - - rgb->clk_parent = devm_clk_get(dc->dev, "parent"); - if (IS_ERR(rgb->clk_parent)) { - dev_err(dc->dev, "failed to get parent clock\n"); - return PTR_ERR(rgb->clk_parent); - } - - err = clk_set_parent(rgb->clk, rgb->clk_parent); - if (err < 0) { - dev_err(dc->dev, "failed to set parent clock: %d\n", err); - return err; - } - - rgb->output.dev = dc->dev; - rgb->output.of_node = np; - - err = tegra_output_parse_dt(&rgb->output); - if (err < 0) - return err; - - dc->rgb = &rgb->output; - - return 0; -} - -int tegra_dc_rgb_init(struct drm_device *drm, struct tegra_dc *dc) -{ - struct tegra_rgb *rgb = to_rgb(dc->rgb); - int err; - - if (!dc->rgb) - return -ENODEV; - - rgb->output.type = TEGRA_OUTPUT_RGB; - rgb->output.ops = &rgb_ops; - - err = tegra_output_init(dc->base.dev, &rgb->output); - if (err < 0) { - dev_err(dc->dev, "output setup failed: %d\n", err); - return err; - } - - /* - * By default, outputs can be associated with each display controller. - * RGB outputs are an exception, so we make sure they can be attached - * to only their parent display controller. - */ - rgb->output.encoder.possible_crtcs = 1 << dc->pipe; - - return 0; -} - -int tegra_dc_rgb_exit(struct tegra_dc *dc) -{ - if (dc->rgb) { - int err; - - err = tegra_output_disable(dc->rgb); - if (err < 0) { - dev_err(dc->dev, "output failed to disable: %d\n", err); - return err; - } - - err = tegra_output_exit(dc->rgb); - if (err < 0) { - dev_err(dc->dev, "output cleanup failed: %d\n", err); - return err; - } - - dc->rgb = NULL; - } - - return 0; -} diff --git a/drivers/gpu/host1x/fence.c b/drivers/gpu/host1x/fence.c new file mode 100644 index 000000000000..139ad1afd935 --- /dev/null +++ b/drivers/gpu/host1x/fence.c @@ -0,0 +1,154 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Syncpoint dma_fence implementation + * + * Copyright (c) 2020, NVIDIA Corporation. + */ + +#include <linux/dma-fence.h> +#include <linux/file.h> +#include <linux/fs.h> +#include <linux/slab.h> +#include <linux/sync_file.h> + +#include "fence.h" +#include "intr.h" +#include "syncpt.h" + +static const char *host1x_syncpt_fence_get_driver_name(struct dma_fence *f) +{ + return "host1x"; +} + +static const char *host1x_syncpt_fence_get_timeline_name(struct dma_fence *f) +{ + return "syncpoint"; +} + +static struct host1x_syncpt_fence *to_host1x_fence(struct dma_fence *f) +{ + return container_of(f, struct host1x_syncpt_fence, base); +} + +static bool host1x_syncpt_fence_enable_signaling(struct dma_fence *f) +{ + struct host1x_syncpt_fence *sf = to_host1x_fence(f); + + if (host1x_syncpt_is_expired(sf->sp, sf->threshold)) + return false; + + /* Reference for interrupt path. */ + dma_fence_get(f); + + /* + * The dma_fence framework requires the fence driver to keep a + * reference to any fences for which 'enable_signaling' has been + * called (and that have not been signalled). + * + * We cannot currently always guarantee that all fences get signalled + * or cancelled. As such, for such situations, set up a timeout, so + * that long-lasting fences will get reaped eventually. + */ + if (sf->timeout) { + /* Reference for timeout path. */ + dma_fence_get(f); + schedule_delayed_work(&sf->timeout_work, msecs_to_jiffies(30000)); + } + + host1x_intr_add_fence_locked(sf->sp->host, sf); + + /* + * The fence may get signalled at any time after the above call, + * so we need to initialize all state used by signalling + * before it. + */ + + return true; +} + +static const struct dma_fence_ops host1x_syncpt_fence_ops = { + .get_driver_name = host1x_syncpt_fence_get_driver_name, + .get_timeline_name = host1x_syncpt_fence_get_timeline_name, + .enable_signaling = host1x_syncpt_fence_enable_signaling, +}; + +void host1x_fence_signal(struct host1x_syncpt_fence *f) +{ + if (atomic_xchg(&f->signaling, 1)) { + /* + * Already on timeout path, but we removed the fence before + * timeout path could, so drop interrupt path reference. + */ + dma_fence_put(&f->base); + return; + } + + if (f->timeout && cancel_delayed_work(&f->timeout_work)) { + /* + * We know that the timeout path will not be entered. + * Safe to drop the timeout path's reference now. + */ + dma_fence_put(&f->base); + } + + dma_fence_signal_locked(&f->base); + dma_fence_put(&f->base); +} + +static void do_fence_timeout(struct work_struct *work) +{ + struct delayed_work *dwork = (struct delayed_work *)work; + struct host1x_syncpt_fence *f = + container_of(dwork, struct host1x_syncpt_fence, timeout_work); + + if (atomic_xchg(&f->signaling, 1)) { + /* Already on interrupt path, drop timeout path reference if any. */ + if (f->timeout) + dma_fence_put(&f->base); + return; + } + + if (host1x_intr_remove_fence(f->sp->host, f)) { + /* + * Managed to remove fence from queue, so it's safe to drop + * the interrupt path's reference. + */ + dma_fence_put(&f->base); + } + + dma_fence_set_error(&f->base, -ETIMEDOUT); + dma_fence_signal(&f->base); + if (f->timeout) + dma_fence_put(&f->base); +} + +struct dma_fence *host1x_fence_create(struct host1x_syncpt *sp, u32 threshold, + bool timeout) +{ + struct host1x_syncpt_fence *fence; + + fence = kzalloc(sizeof(*fence), GFP_KERNEL); + if (!fence) + return ERR_PTR(-ENOMEM); + + fence->sp = sp; + fence->threshold = threshold; + fence->timeout = timeout; + + dma_fence_init(&fence->base, &host1x_syncpt_fence_ops, &sp->fences.lock, + dma_fence_context_alloc(1), 0); + + INIT_DELAYED_WORK(&fence->timeout_work, do_fence_timeout); + + return &fence->base; +} +EXPORT_SYMBOL(host1x_fence_create); + +void host1x_fence_cancel(struct dma_fence *f) +{ + struct host1x_syncpt_fence *sf = to_host1x_fence(f); + + schedule_delayed_work(&sf->timeout_work, 0); + flush_delayed_work(&sf->timeout_work); +} +EXPORT_SYMBOL(host1x_fence_cancel); diff --git a/drivers/gpu/host1x/fence.h b/drivers/gpu/host1x/fence.h new file mode 100644 index 000000000000..f3c644c73cad --- /dev/null +++ b/drivers/gpu/host1x/fence.h @@ -0,0 +1,30 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2020, NVIDIA Corporation. + */ + +#ifndef HOST1X_FENCE_H +#define HOST1X_FENCE_H + +struct host1x_syncpt_fence { + struct dma_fence base; + + atomic_t signaling; + + struct host1x_syncpt *sp; + u32 threshold; + bool timeout; + + struct delayed_work timeout_work; + + struct list_head list; +}; + +struct host1x_fence_list { + spinlock_t lock; + struct list_head list; +}; + +void host1x_fence_signal(struct host1x_syncpt_fence *fence); + +#endif diff --git a/drivers/gpu/host1x/host1x.h b/drivers/gpu/host1x/host1x.h deleted file mode 100644 index a2bc1e65e972..000000000000 --- a/drivers/gpu/host1x/host1x.h +++ /dev/null @@ -1,30 +0,0 @@ -/* - * Tegra host1x driver - * - * Copyright (c) 2009-2013, NVIDIA Corporation. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write to the Free Software Foundation, Inc., - * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. - */ - -#ifndef __LINUX_HOST1X_H -#define __LINUX_HOST1X_H - -enum host1x_class { - HOST1X_CLASS_HOST1X = 0x1, - HOST1X_CLASS_GR2D = 0x51, - HOST1X_CLASS_GR2D_SB = 0x52 -}; - -#endif diff --git a/drivers/gpu/host1x/host1x_bo.h b/drivers/gpu/host1x/host1x_bo.h deleted file mode 100644 index 4c1f10bd773d..000000000000 --- a/drivers/gpu/host1x/host1x_bo.h +++ /dev/null @@ -1,87 +0,0 @@ -/* - * Tegra host1x Memory Management Abstraction header - * - * Copyright (c) 2012-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. - */ - -#ifndef _HOST1X_BO_H -#define _HOST1X_BO_H - -struct host1x_bo; - -struct host1x_bo_ops { - struct host1x_bo *(*get)(struct host1x_bo *bo); - void (*put)(struct host1x_bo *bo); - dma_addr_t (*pin)(struct host1x_bo *bo, struct sg_table **sgt); - void (*unpin)(struct host1x_bo *bo, struct sg_table *sgt); - void *(*mmap)(struct host1x_bo *bo); - void (*munmap)(struct host1x_bo *bo, void *addr); - void *(*kmap)(struct host1x_bo *bo, unsigned int pagenum); - void (*kunmap)(struct host1x_bo *bo, unsigned int pagenum, void *addr); -}; - -struct host1x_bo { - const struct host1x_bo_ops *ops; -}; - -static inline void host1x_bo_init(struct host1x_bo *bo, - const struct host1x_bo_ops *ops) -{ - bo->ops = ops; -} - -static inline struct host1x_bo *host1x_bo_get(struct host1x_bo *bo) -{ - return bo->ops->get(bo); -} - -static inline void host1x_bo_put(struct host1x_bo *bo) -{ - bo->ops->put(bo); -} - -static inline dma_addr_t host1x_bo_pin(struct host1x_bo *bo, - struct sg_table **sgt) -{ - return bo->ops->pin(bo, sgt); -} - -static inline void host1x_bo_unpin(struct host1x_bo *bo, struct sg_table *sgt) -{ - bo->ops->unpin(bo, sgt); -} - -static inline void *host1x_bo_mmap(struct host1x_bo *bo) -{ - return bo->ops->mmap(bo); -} - -static inline void host1x_bo_munmap(struct host1x_bo *bo, void *addr) -{ - bo->ops->munmap(bo, addr); -} - -static inline void *host1x_bo_kmap(struct host1x_bo *bo, unsigned int pagenum) -{ - return bo->ops->kmap(bo, pagenum); -} - -static inline void host1x_bo_kunmap(struct host1x_bo *bo, - unsigned int pagenum, void *addr) -{ - bo->ops->kunmap(bo, pagenum, addr); -} - -#endif diff --git a/drivers/gpu/host1x/host1x_client.h b/drivers/gpu/host1x/host1x_client.h deleted file mode 100644 index 9b85f10f4a44..000000000000 --- a/drivers/gpu/host1x/host1x_client.h +++ /dev/null @@ -1,35 +0,0 @@ -/* - * Copyright (c) 2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. - */ - -#ifndef HOST1X_CLIENT_H -#define HOST1X_CLIENT_H - -struct device; -struct platform_device; - -#ifdef CONFIG_DRM_TEGRA -int host1x_drm_alloc(struct platform_device *pdev); -#else -static inline int host1x_drm_alloc(struct platform_device *pdev) -{ - return 0; -} -#endif - -void host1x_set_drm_data(struct device *dev, void *data); -void *host1x_get_drm_data(struct device *dev); - -#endif diff --git a/drivers/gpu/host1x/hw/Makefile b/drivers/gpu/host1x/hw/Makefile deleted file mode 100644 index 9b50863a2236..000000000000 --- a/drivers/gpu/host1x/hw/Makefile +++ /dev/null @@ -1,6 +0,0 @@ -ccflags-y = -Idrivers/gpu/host1x - -host1x-hw-objs = \ - host1x01.o - -obj-$(CONFIG_TEGRA_HOST1X) += host1x-hw.o diff --git a/drivers/gpu/host1x/hw/cdma_hw.c b/drivers/gpu/host1x/hw/cdma_hw.c index 2ee4ad55c4db..3f3f0018eee0 100644 --- a/drivers/gpu/host1x/hw/cdma_hw.c +++ b/drivers/gpu/host1x/hw/cdma_hw.c @@ -1,36 +1,25 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Tegra host1x Command DMA * * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ #include <linux/slab.h> #include <linux/scatterlist.h> #include <linux/dma-mapping.h> -#include "cdma.h" -#include "channel.h" -#include "dev.h" -#include "debug.h" +#include "../cdma.h" +#include "../channel.h" +#include "../dev.h" +#include "../debug.h" /* - * Put the restart at the end of pushbuffer memor + * Put the restart at the end of pushbuffer memory */ static void push_buffer_init(struct push_buffer *pb) { - *(pb->mapped + (pb->size_bytes >> 2)) = host1x_opcode_restart(0); + *(u32 *)(pb->mapped + pb->size) = host1x_opcode_restart(0); } /* @@ -39,26 +28,13 @@ static void push_buffer_init(struct push_buffer *pb) static void cdma_timeout_cpu_incr(struct host1x_cdma *cdma, u32 getptr, u32 syncpt_incrs, u32 syncval, u32 nr_slots) { - struct host1x *host1x = cdma_to_host1x(cdma); - struct push_buffer *pb = &cdma->push_buffer; - u32 i; + unsigned int i; for (i = 0; i < syncpt_incrs; i++) host1x_syncpt_incr(cdma->timeout.syncpt); /* after CPU incr, ensure shadow is up to date */ host1x_syncpt_load(cdma->timeout.syncpt); - - /* NOP all the PB slots */ - while (nr_slots--) { - u32 *p = (u32 *)((u32)pb->mapped + getptr); - *(p++) = HOST1X_OPCODE_NOP; - *(p++) = HOST1X_OPCODE_NOP; - dev_dbg(host1x->dev, "%s: NOP at 0x%x\n", __func__, - pb->phys + getptr); - getptr = (getptr + 8) & (pb->size_bytes - 1); - } - wmb(); } /* @@ -67,21 +43,31 @@ static void cdma_timeout_cpu_incr(struct host1x_cdma *cdma, u32 getptr, static void cdma_start(struct host1x_cdma *cdma) { struct host1x_channel *ch = cdma_to_channel(cdma); + u64 start, end; if (cdma->running) return; cdma->last_pos = cdma->push_buffer.pos; + start = cdma->push_buffer.dma; + end = cdma->push_buffer.size + 4; host1x_ch_writel(ch, HOST1X_CHANNEL_DMACTRL_DMASTOP, HOST1X_CHANNEL_DMACTRL); /* set base, put and end pointer */ - host1x_ch_writel(ch, cdma->push_buffer.phys, HOST1X_CHANNEL_DMASTART); + host1x_ch_writel(ch, lower_32_bits(start), HOST1X_CHANNEL_DMASTART); +#if HOST1X_HW >= 6 + host1x_ch_writel(ch, upper_32_bits(start), HOST1X_CHANNEL_DMASTART_HI); +#endif host1x_ch_writel(ch, cdma->push_buffer.pos, HOST1X_CHANNEL_DMAPUT); - host1x_ch_writel(ch, cdma->push_buffer.phys + - cdma->push_buffer.size_bytes + 4, - HOST1X_CHANNEL_DMAEND); +#if HOST1X_HW >= 6 + host1x_ch_writel(ch, 0, HOST1X_CHANNEL_DMAPUT_HI); +#endif + host1x_ch_writel(ch, lower_32_bits(end), HOST1X_CHANNEL_DMAEND); +#if HOST1X_HW >= 6 + host1x_ch_writel(ch, upper_32_bits(end), HOST1X_CHANNEL_DMAEND_HI); +#endif /* reset GET */ host1x_ch_writel(ch, HOST1X_CHANNEL_DMACTRL_DMASTOP | @@ -104,6 +90,7 @@ static void cdma_timeout_restart(struct host1x_cdma *cdma, u32 getptr) { struct host1x *host1x = cdma_to_host1x(cdma); struct host1x_channel *ch = cdma_to_channel(cdma); + u64 start, end; if (cdma->running) return; @@ -113,11 +100,18 @@ static void cdma_timeout_restart(struct host1x_cdma *cdma, u32 getptr) host1x_ch_writel(ch, HOST1X_CHANNEL_DMACTRL_DMASTOP, HOST1X_CHANNEL_DMACTRL); + start = cdma->push_buffer.dma; + end = cdma->push_buffer.size + 4; + /* set base, end pointer (all of memory) */ - host1x_ch_writel(ch, cdma->push_buffer.phys, HOST1X_CHANNEL_DMASTART); - host1x_ch_writel(ch, cdma->push_buffer.phys + - cdma->push_buffer.size_bytes, - HOST1X_CHANNEL_DMAEND); + host1x_ch_writel(ch, lower_32_bits(start), HOST1X_CHANNEL_DMASTART); +#if HOST1X_HW >= 6 + host1x_ch_writel(ch, upper_32_bits(start), HOST1X_CHANNEL_DMASTART_HI); +#endif + host1x_ch_writel(ch, lower_32_bits(end), HOST1X_CHANNEL_DMAEND); +#if HOST1X_HW >= 6 + host1x_ch_writel(ch, upper_32_bits(end), HOST1X_CHANNEL_DMAEND_HI); +#endif /* set GET, by loading the value in PUT (then reset GET) */ host1x_ch_writel(ch, getptr, HOST1X_CHANNEL_DMAPUT); @@ -162,15 +156,41 @@ static void cdma_stop(struct host1x_cdma *cdma) struct host1x_channel *ch = cdma_to_channel(cdma); mutex_lock(&cdma->lock); + if (cdma->running) { host1x_cdma_wait_locked(cdma, CDMA_EVENT_SYNC_QUEUE_EMPTY); host1x_ch_writel(ch, HOST1X_CHANNEL_DMACTRL_DMASTOP, HOST1X_CHANNEL_DMACTRL); cdma->running = false; } + mutex_unlock(&cdma->lock); } +static void cdma_hw_cmdproc_stop(struct host1x *host, struct host1x_channel *ch, + bool stop) +{ +#if HOST1X_HW >= 6 + host1x_ch_writel(ch, stop ? 0x1 : 0x0, HOST1X_CHANNEL_CMDPROC_STOP); +#else + u32 cmdproc_stop = host1x_sync_readl(host, HOST1X_SYNC_CMDPROC_STOP); + if (stop) + cmdproc_stop |= BIT(ch->id); + else + cmdproc_stop &= ~BIT(ch->id); + host1x_sync_writel(host, cmdproc_stop, HOST1X_SYNC_CMDPROC_STOP); +#endif +} + +static void cdma_hw_teardown(struct host1x *host, struct host1x_channel *ch) +{ +#if HOST1X_HW >= 6 + host1x_ch_writel(ch, 0x1, HOST1X_CHANNEL_TEARDOWN); +#else + host1x_sync_writel(host, BIT(ch->id), HOST1X_SYNC_CH_TEARDOWN); +#endif +} + /* * Stops both channel's command processor and CDMA immediately. * Also, tears down the channel and resets corresponding module. @@ -179,7 +199,6 @@ static void cdma_freeze(struct host1x_cdma *cdma) { struct host1x *host = cdma_to_host1x(cdma); struct host1x_channel *ch = cdma_to_channel(cdma); - u32 cmdproc_stop; if (cdma->torndown && !cdma->running) { dev_warn(host->dev, "Already torn down\n"); @@ -188,9 +207,7 @@ static void cdma_freeze(struct host1x_cdma *cdma) dev_dbg(host->dev, "freezing channel (id %d)\n", ch->id); - cmdproc_stop = host1x_sync_readl(host, HOST1X_SYNC_CMDPROC_STOP); - cmdproc_stop |= BIT(ch->id); - host1x_sync_writel(host, cmdproc_stop, HOST1X_SYNC_CMDPROC_STOP); + cdma_hw_cmdproc_stop(host, ch, true); dev_dbg(host->dev, "%s: DMA GET 0x%x, PUT HW 0x%x / shadow 0x%x\n", __func__, host1x_ch_readl(ch, HOST1X_CHANNEL_DMAGET), @@ -200,7 +217,7 @@ static void cdma_freeze(struct host1x_cdma *cdma) host1x_ch_writel(ch, HOST1X_CHANNEL_DMACTRL_DMASTOP, HOST1X_CHANNEL_DMACTRL); - host1x_sync_writel(host, BIT(ch->id), HOST1X_SYNC_CH_TEARDOWN); + cdma_hw_teardown(host, ch); cdma->running = false; cdma->torndown = true; @@ -210,20 +227,60 @@ static void cdma_resume(struct host1x_cdma *cdma, u32 getptr) { struct host1x *host1x = cdma_to_host1x(cdma); struct host1x_channel *ch = cdma_to_channel(cdma); - u32 cmdproc_stop; dev_dbg(host1x->dev, - "resuming channel (id %d, DMAGET restart = 0x%x)\n", + "resuming channel (id %u, DMAGET restart = 0x%x)\n", ch->id, getptr); - cmdproc_stop = host1x_sync_readl(host1x, HOST1X_SYNC_CMDPROC_STOP); - cmdproc_stop &= ~(BIT(ch->id)); - host1x_sync_writel(host1x, cmdproc_stop, HOST1X_SYNC_CMDPROC_STOP); + cdma_hw_cmdproc_stop(host1x, ch, false); cdma->torndown = false; cdma_timeout_restart(cdma, getptr); } +static void timeout_release_mlock(struct host1x_cdma *cdma) +{ +#if HOST1X_HW >= 8 + /* Tegra186 and Tegra194 require a more complicated MLOCK release + * sequence. Furthermore, those chips by default don't enforce MLOCKs, + * so it turns out that if we don't /actually/ need MLOCKs, we can just + * ignore them. + * + * As such, for now just implement this on Tegra234 where things are + * stricter but also easy to implement. + */ + struct host1x_channel *ch = cdma_to_channel(cdma); + struct host1x *host1x = cdma_to_host1x(cdma); + u32 offset; + + switch (ch->client->class) { + case HOST1X_CLASS_NVJPG1: + offset = HOST1X_COMMON_NVJPG1_MLOCK; + break; + case HOST1X_CLASS_NVENC: + offset = HOST1X_COMMON_NVENC_MLOCK; + break; + case HOST1X_CLASS_VIC: + offset = HOST1X_COMMON_VIC_MLOCK; + break; + case HOST1X_CLASS_NVJPG: + offset = HOST1X_COMMON_NVJPG_MLOCK; + break; + case HOST1X_CLASS_NVDEC: + offset = HOST1X_COMMON_NVDEC_MLOCK; + break; + case HOST1X_CLASS_OFA: + offset = HOST1X_COMMON_OFA_MLOCK; + break; + default: + WARN(1, "%s was not updated for class %u", __func__, ch->client->class); + return; + } + + host1x_common_writel(host1x, 0x0, offset); +#endif +} + /* * If this timeout fires, it indicates the current sync_queue entry has * exceeded its TTL and the userctx should be timed out and remaining @@ -231,14 +288,11 @@ static void cdma_resume(struct host1x_cdma *cdma, u32 getptr) */ static void cdma_timeout_handler(struct work_struct *work) { + u32 syncpt_val; struct host1x_cdma *cdma; struct host1x *host1x; struct host1x_channel *ch; - u32 syncpt_val; - - u32 prev_cmdproc, cmdproc_stop; - cdma = container_of(to_delayed_work(work), struct host1x_cdma, timeout.wq); host1x = cdma_to_host1x(cdma); @@ -256,12 +310,7 @@ static void cdma_timeout_handler(struct work_struct *work) } /* stop processing to get a clean snapshot */ - prev_cmdproc = host1x_sync_readl(host1x, HOST1X_SYNC_CMDPROC_STOP); - cmdproc_stop = prev_cmdproc | BIT(ch->id); - host1x_sync_writel(host1x, cmdproc_stop, HOST1X_SYNC_CMDPROC_STOP); - - dev_dbg(host1x->dev, "cdma_timeout: cmdproc was 0x%x is 0x%x\n", - prev_cmdproc, cmdproc_stop); + cdma_hw_cmdproc_stop(host1x, ch, true); syncpt_val = host1x_syncpt_load(cdma->timeout.syncpt); @@ -270,20 +319,21 @@ static void cdma_timeout_handler(struct work_struct *work) dev_dbg(host1x->dev, "cdma_timeout: expired, but buffer had completed\n"); /* restore */ - cmdproc_stop = prev_cmdproc & ~(BIT(ch->id)); - host1x_sync_writel(host1x, cmdproc_stop, - HOST1X_SYNC_CMDPROC_STOP); + cdma_hw_cmdproc_stop(host1x, ch, false); mutex_unlock(&cdma->lock); return; } - dev_warn(host1x->dev, "%s: timeout: %d (%s), HW thresh %d, done %d\n", - __func__, cdma->timeout.syncpt->id, cdma->timeout.syncpt->name, - syncpt_val, cdma->timeout.syncpt_val); + dev_warn(host1x->dev, "%s: timeout: %u (%s), HW thresh %d, done %d\n", + __func__, cdma->timeout.syncpt->id, cdma->timeout.syncpt->name, + syncpt_val, cdma->timeout.syncpt_val); /* stop HW, resetting channel/module */ host1x_hw_cdma_freeze(host1x, cdma); + /* release any held MLOCK */ + timeout_release_mlock(cdma); + host1x_cdma_update_sync_queue(cdma, ch->dev); mutex_unlock(&cdma->lock); } @@ -291,7 +341,7 @@ static void cdma_timeout_handler(struct work_struct *work) /* * Init timeout resources */ -static int cdma_timeout_init(struct host1x_cdma *cdma, u32 syncpt_id) +static int cdma_timeout_init(struct host1x_cdma *cdma) { INIT_DELAYED_WORK(&cdma->timeout.wq, cdma_timeout_handler); cdma->timeout.initialized = true; @@ -306,6 +356,7 @@ static void cdma_timeout_destroy(struct host1x_cdma *cdma) { if (cdma->timeout.initialized) cancel_delayed_work(&cdma->timeout.wq); + cdma->timeout.initialized = false; } diff --git a/drivers/gpu/host1x/hw/channel_hw.c b/drivers/gpu/host1x/hw/channel_hw.c index ee199623e365..2df6a16d484e 100644 --- a/drivers/gpu/host1x/hw/channel_hw.c +++ b/drivers/gpu/host1x/hw/channel_hw.c @@ -1,37 +1,27 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Tegra host1x Channel * * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ +#include <linux/host1x.h> +#include <linux/iommu.h> #include <linux/slab.h> + #include <trace/events/host1x.h> -#include "host1x.h" -#include "host1x_bo.h" -#include "channel.h" -#include "dev.h" -#include "intr.h" -#include "job.h" +#include "../channel.h" +#include "../dev.h" +#include "../intr.h" +#include "../job.h" -#define HOST1X_CHANNEL_SIZE 16384 #define TRACE_MAX_LENGTH 128U static void trace_write_gather(struct host1x_cdma *cdma, struct host1x_bo *bo, u32 offset, u32 words) { + struct device *dev = cdma_to_channel(cdma)->dev; void *mem = NULL; if (host1x_debug_trace_cmdbuf) @@ -44,121 +34,352 @@ static void trace_write_gather(struct host1x_cdma *cdma, struct host1x_bo *bo, * of how much you can output to ftrace at once. */ for (i = 0; i < words; i += TRACE_MAX_LENGTH) { - trace_host1x_cdma_push_gather( - dev_name(cdma_to_channel(cdma)->dev), - (u32)bo, min(words - i, TRACE_MAX_LENGTH), - offset + i * sizeof(u32), mem); + u32 num_words = min(words - i, TRACE_MAX_LENGTH); + + offset += i * sizeof(u32); + + trace_host1x_cdma_push_gather(dev_name(dev), bo, + num_words, offset, + mem); } + host1x_bo_munmap(bo, mem); } } -static void submit_gathers(struct host1x_job *job) +static void submit_wait(struct host1x_job *job, u32 id, u32 threshold) { struct host1x_cdma *cdma = &job->channel->cdma; + +#if HOST1X_HW >= 2 + host1x_cdma_push_wide(cdma, + host1x_opcode_setclass( + HOST1X_CLASS_HOST1X, + HOST1X_UCLASS_LOAD_SYNCPT_PAYLOAD_32, + /* WAIT_SYNCPT_32 is at SYNCPT_PAYLOAD_32+2 */ + BIT(0) | BIT(2) + ), + threshold, + id, + HOST1X_OPCODE_NOP + ); +#else + /* TODO add waitchk or use waitbases or other mitigation */ + host1x_cdma_push(cdma, + host1x_opcode_setclass( + HOST1X_CLASS_HOST1X, + host1x_uclass_wait_syncpt_r(), + BIT(0) + ), + host1x_class_host_wait_syncpt(id, threshold) + ); +#endif +} + +static void submit_setclass(struct host1x_job *job, u32 next_class) +{ + struct host1x_cdma *cdma = &job->channel->cdma; + +#if HOST1X_HW >= 6 + u32 stream_id; + + /* + * If a memory context has been set, use it. Otherwise + * (if context isolation is disabled) use the engine's + * firmware stream ID. + */ + if (job->memory_context) + stream_id = job->memory_context->stream_id; + else + stream_id = job->engine_fallback_streamid; + + host1x_cdma_push_wide(cdma, + host1x_opcode_setclass(next_class, 0, 0), + host1x_opcode_setpayload(stream_id), + host1x_opcode_setstreamid(job->engine_streamid_offset / 4), + HOST1X_OPCODE_NOP); +#else + host1x_cdma_push(cdma, + host1x_opcode_setclass(next_class, 0, 0), + HOST1X_OPCODE_NOP + ); +#endif +} + +static void submit_gathers(struct host1x_job *job, struct host1x_job_cmd *cmds, u32 num_cmds, + u32 job_syncpt_base) +{ + struct host1x_cdma *cdma = &job->channel->cdma; +#if HOST1X_HW < 6 + struct device *dev = job->channel->dev; +#endif unsigned int i; + u32 threshold; + + for (i = 0; i < num_cmds; i++) { + struct host1x_job_cmd *cmd = &cmds[i]; + + if (cmd->is_wait) { + if (cmd->wait.relative) + threshold = job_syncpt_base + cmd->wait.threshold; + else + threshold = cmd->wait.threshold; + + submit_wait(job, cmd->wait.id, threshold); + submit_setclass(job, cmd->wait.next_class); + } else { + struct host1x_job_gather *g = &cmd->gather; + + dma_addr_t addr = g->base + g->offset; + u32 op2, op3; + + op2 = lower_32_bits(addr); + op3 = upper_32_bits(addr); + + trace_write_gather(cdma, g->bo, g->offset, g->words); + + if (op3 != 0) { +#if HOST1X_HW >= 6 + u32 op1 = host1x_opcode_gather_wide(g->words); + u32 op4 = HOST1X_OPCODE_NOP; + + host1x_cdma_push_wide(cdma, op1, op2, op3, op4); +#else + dev_err(dev, "invalid gather for push buffer %pad\n", + &addr); + continue; +#endif + } else { + u32 op1 = host1x_opcode_gather(g->words); - for (i = 0; i < job->num_gathers; i++) { - struct host1x_job_gather *g = &job->gathers[i]; - u32 op1 = host1x_opcode_gather(g->words); - u32 op2 = g->base + g->offset; - trace_write_gather(cdma, g->bo, g->offset, op1 & 0xffff); - host1x_cdma_push(cdma, op1, op2); + host1x_cdma_push(cdma, op1, op2); + } + } } } -static int channel_submit(struct host1x_job *job) +static inline void synchronize_syncpt_base(struct host1x_job *job) { - struct host1x_channel *ch = job->channel; - struct host1x_syncpt *sp; - u32 user_syncpt_incrs = job->syncpt_incrs; - u32 prev_max = 0; - u32 syncval; - int err; - struct host1x_waitlist *completed_waiter = NULL; + struct host1x_syncpt *sp = job->syncpt; + unsigned int id; + u32 value; + + value = host1x_syncpt_read_max(sp); + id = sp->base->id; + + host1x_cdma_push(&job->channel->cdma, + host1x_opcode_setclass(HOST1X_CLASS_HOST1X, + HOST1X_UCLASS_LOAD_SYNCPT_BASE, 1), + HOST1X_UCLASS_LOAD_SYNCPT_BASE_BASE_INDX_F(id) | + HOST1X_UCLASS_LOAD_SYNCPT_BASE_VALUE_F(value)); +} + +static void host1x_channel_set_streamid(struct host1x_channel *channel) +{ +#if HOST1X_HW >= 6 + u32 stream_id; + + if (!tegra_dev_iommu_get_stream_id(channel->dev->parent, &stream_id)) + stream_id = TEGRA_STREAM_ID_BYPASS; + + host1x_ch_writel(channel, stream_id, HOST1X_CHANNEL_SMMU_STREAMID); +#endif +} + +static void host1x_enable_gather_filter(struct host1x_channel *ch) +{ +#if HOST1X_HW >= 6 struct host1x *host = dev_get_drvdata(ch->dev->parent); + u32 val; - sp = host->syncpt + job->syncpt_id; - trace_host1x_channel_submit(dev_name(ch->dev), - job->num_gathers, job->num_relocs, - job->num_waitchk, job->syncpt_id, - job->syncpt_incrs); + if (!host->hv_regs) + return; - /* before error checks, return current max */ - prev_max = job->syncpt_end = host1x_syncpt_read_max(sp); + val = host1x_hypervisor_readl( + host, HOST1X_HV_CH_KERNEL_FILTER_GBUFFER(ch->id / 32)); + val |= BIT(ch->id % 32); + host1x_hypervisor_writel( + host, val, HOST1X_HV_CH_KERNEL_FILTER_GBUFFER(ch->id / 32)); +#elif HOST1X_HW >= 4 + host1x_ch_writel(ch, + HOST1X_CHANNEL_CHANNELCTRL_KERNEL_FILTER_GBUFFER(1), + HOST1X_CHANNEL_CHANNELCTRL); +#endif +} - /* get submit lock */ - err = mutex_lock_interruptible(&ch->submitlock); - if (err) - goto error; +static void channel_program_cdma(struct host1x_job *job) +{ + struct host1x_cdma *cdma = &job->channel->cdma; + struct host1x_syncpt *sp = job->syncpt; - completed_waiter = kzalloc(sizeof(*completed_waiter), GFP_KERNEL); - if (!completed_waiter) { - mutex_unlock(&ch->submitlock); - err = -ENOMEM; - goto error; - } +#if HOST1X_HW >= 6 + u32 fence; + int i = 0; - /* begin a CDMA submit */ - err = host1x_cdma_begin(&ch->cdma, job); - if (err) { - mutex_unlock(&ch->submitlock); - goto error; + if (job->num_cmds == 0) + goto prefences_done; + if (!job->cmds[0].is_wait || job->cmds[0].wait.relative) + goto prefences_done; + + /* Enter host1x class with invalid stream ID for prefence waits. */ + host1x_cdma_push_wide(cdma, + host1x_opcode_acquire_mlock(1), + host1x_opcode_setclass(1, 0, 0), + host1x_opcode_setpayload(0), + host1x_opcode_setstreamid(0x1fffff)); + + for (i = 0; i < job->num_cmds; i++) { + struct host1x_job_cmd *cmd = &job->cmds[i]; + + if (!cmd->is_wait || cmd->wait.relative) + break; + + submit_wait(job, cmd->wait.id, cmd->wait.threshold); } + host1x_cdma_push(cdma, + HOST1X_OPCODE_NOP, + host1x_opcode_release_mlock(1)); + +prefences_done: + /* Enter engine class with invalid stream ID. */ + host1x_cdma_push_wide(cdma, + host1x_opcode_acquire_mlock(job->class), + host1x_opcode_setclass(job->class, 0, 0), + host1x_opcode_setpayload(0), + host1x_opcode_setstreamid(job->engine_streamid_offset / 4)); + + /* Before switching stream ID to real stream ID, ensure engine is idle. */ + fence = host1x_syncpt_incr_max(sp, 1); + host1x_cdma_push(&job->channel->cdma, + host1x_opcode_nonincr(HOST1X_UCLASS_INCR_SYNCPT, 1), + HOST1X_UCLASS_INCR_SYNCPT_INDX_F(job->syncpt->id) | + HOST1X_UCLASS_INCR_SYNCPT_COND_F(4)); + submit_wait(job, job->syncpt->id, fence); + submit_setclass(job, job->class); + + /* Submit work. */ + job->syncpt_end = host1x_syncpt_incr_max(sp, job->syncpt_incrs); + submit_gathers(job, job->cmds + i, job->num_cmds - i, job->syncpt_end - job->syncpt_incrs); + + /* Before releasing MLOCK, ensure engine is idle again. */ + fence = host1x_syncpt_incr_max(sp, 1); + host1x_cdma_push(&job->channel->cdma, + host1x_opcode_nonincr(HOST1X_UCLASS_INCR_SYNCPT, 1), + HOST1X_UCLASS_INCR_SYNCPT_INDX_F(job->syncpt->id) | + HOST1X_UCLASS_INCR_SYNCPT_COND_F(4)); + submit_wait(job, job->syncpt->id, fence); + + /* Release MLOCK. */ + host1x_cdma_push(cdma, + HOST1X_OPCODE_NOP, host1x_opcode_release_mlock(job->class)); +#else if (job->serialize) { /* * Force serialization by inserting a host wait for the * previous job to finish before this one can commence. */ - host1x_cdma_push(&ch->cdma, + host1x_cdma_push(cdma, host1x_opcode_setclass(HOST1X_CLASS_HOST1X, host1x_uclass_wait_syncpt_r(), 1), - host1x_class_host_wait_syncpt(job->syncpt_id, + host1x_class_host_wait_syncpt(job->syncpt->id, host1x_syncpt_read_max(sp))); } - syncval = host1x_syncpt_incr_max(sp, user_syncpt_incrs); - - job->syncpt_end = syncval; + /* Synchronize base register to allow using it for relative waiting */ + if (sp->base) + synchronize_syncpt_base(job); /* add a setclass for modules that require it */ if (job->class) - host1x_cdma_push(&ch->cdma, + host1x_cdma_push(cdma, host1x_opcode_setclass(job->class, 0, 0), HOST1X_OPCODE_NOP); - submit_gathers(job); + job->syncpt_end = host1x_syncpt_incr_max(sp, job->syncpt_incrs); + + submit_gathers(job, job->cmds, job->num_cmds, job->syncpt_end - job->syncpt_incrs); +#endif +} + +static void job_complete_callback(struct dma_fence *fence, struct dma_fence_cb *cb) +{ + struct host1x_job *job = container_of(cb, struct host1x_job, fence_cb); + + /* Schedules CDMA update. */ + host1x_cdma_update(&job->channel->cdma); +} + +static int channel_submit(struct host1x_job *job) +{ + struct host1x_channel *ch = job->channel; + struct host1x_syncpt *sp = job->syncpt; + u32 prev_max = 0; + u32 syncval; + int err; + struct host1x *host = dev_get_drvdata(ch->dev->parent); + + trace_host1x_channel_submit(dev_name(ch->dev), + job->num_cmds, job->num_relocs, + job->syncpt->id, job->syncpt_incrs); + + /* before error checks, return current max */ + prev_max = job->syncpt_end = host1x_syncpt_read_max(sp); + + /* get submit lock */ + err = mutex_lock_interruptible(&ch->submitlock); + if (err) + return err; + + host1x_channel_set_streamid(ch); + host1x_enable_gather_filter(ch); + host1x_hw_syncpt_assign_to_channel(host, sp, ch); + + /* begin a CDMA submit */ + err = host1x_cdma_begin(&ch->cdma, job); + if (err) { + mutex_unlock(&ch->submitlock); + return err; + } + + channel_program_cdma(job); + syncval = host1x_syncpt_read_max(sp); + + /* + * Create fence before submitting job to HW to avoid job completing + * before the fence is set up. + */ + job->fence = host1x_fence_create(sp, syncval, true); + if (WARN(IS_ERR(job->fence), "Failed to create submit complete fence")) { + job->fence = NULL; + } else { + err = dma_fence_add_callback(job->fence, &job->fence_cb, + job_complete_callback); + } /* end CDMA submit & stash pinned hMems into sync queue */ host1x_cdma_end(&ch->cdma, job); trace_host1x_channel_submitted(dev_name(ch->dev), prev_max, syncval); - /* schedule a submit complete interrupt */ - err = host1x_intr_add_action(host, job->syncpt_id, syncval, - HOST1X_INTR_ACTION_SUBMIT_COMPLETE, ch, - completed_waiter, NULL); - completed_waiter = NULL; - WARN(err, "Failed to set submit complete interrupt"); - mutex_unlock(&ch->submitlock); - return 0; + if (err == -ENOENT) + host1x_cdma_update(&ch->cdma); + else + WARN(err, "Failed to set submit complete interrupt"); -error: - kfree(completed_waiter); - return err; + return 0; } static int host1x_channel_init(struct host1x_channel *ch, struct host1x *dev, unsigned int index) { - ch->id = index; - mutex_init(&ch->reflock); - mutex_init(&ch->submitlock); - - ch->regs = dev->regs + index * HOST1X_CHANNEL_SIZE; +#if HOST1X_HW < 6 + ch->regs = dev->regs + index * 0x4000; +#else + ch->regs = dev->regs + index * 0x100; +#endif return 0; } diff --git a/drivers/gpu/host1x/hw/debug_hw.c b/drivers/gpu/host1x/hw/debug_hw.c index 334c038052f5..4c32aa1b95e8 100644 --- a/drivers/gpu/host1x/hw/debug_hw.c +++ b/drivers/gpu/host1x/hw/debug_hw.c @@ -1,32 +1,15 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Copyright (C) 2010 Google, Inc. * Author: Erik Gilling <konkers@android.com> * * Copyright (C) 2011-2013 NVIDIA Corporation - * - * This software is licensed under the terms of the GNU General Public - * License version 2, as published by the Free Software Foundation, and - * may be copied, distributed, and modified under those terms. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * */ -#include <linux/debugfs.h> -#include <linux/seq_file.h> -#include <linux/mm.h> -#include <linux/scatterlist.h> - -#include <linux/io.h> - -#include "dev.h" -#include "debug.h" -#include "cdma.h" -#include "channel.h" -#include "host1x_bo.h" +#include "../dev.h" +#include "../debug.h" +#include "../cdma.h" +#include "../channel.h" #define HOST1X_DEBUG_MAX_PAGE_OFFSET 102400 @@ -38,6 +21,13 @@ enum { HOST1X_OPCODE_IMM = 0x04, HOST1X_OPCODE_RESTART = 0x05, HOST1X_OPCODE_GATHER = 0x06, + HOST1X_OPCODE_SETSTRMID = 0x07, + HOST1X_OPCODE_SETAPPID = 0x08, + HOST1X_OPCODE_SETPYLD = 0x09, + HOST1X_OPCODE_INCR_W = 0x0a, + HOST1X_OPCODE_NONINCR_W = 0x0b, + HOST1X_OPCODE_GATHER_W = 0x0c, + HOST1X_OPCODE_RESTART_W = 0x0d, HOST1X_OPCODE_EXTEND = 0x0e, }; @@ -46,80 +36,134 @@ enum { HOST1X_OPCODE_EXTEND_RELEASE_MLOCK = 0x01, }; -static unsigned int show_channel_command(struct output *o, u32 val) +#define INVALID_PAYLOAD 0xffffffff + +static unsigned int show_channel_command(struct output *o, u32 val, + u32 *payload) { - unsigned mask; - unsigned subop; + unsigned int mask, subop, num, opcode; - switch (val >> 28) { + opcode = val >> 28; + + switch (opcode) { case HOST1X_OPCODE_SETCLASS: mask = val & 0x3f; if (mask) { - host1x_debug_output(o, "SETCL(class=%03x, offset=%03x, mask=%02x, [", + host1x_debug_cont(o, "SETCL(class=%03x, offset=%03x, mask=%02x, [", val >> 6 & 0x3ff, val >> 16 & 0xfff, mask); return hweight8(mask); - } else { - host1x_debug_output(o, "SETCL(class=%03x)\n", - val >> 6 & 0x3ff); - return 0; } + host1x_debug_cont(o, "SETCL(class=%03x)\n", val >> 6 & 0x3ff); + return 0; + case HOST1X_OPCODE_INCR: - host1x_debug_output(o, "INCR(offset=%03x, [", + num = val & 0xffff; + host1x_debug_cont(o, "INCR(offset=%03x, [", val >> 16 & 0xfff); - return val & 0xffff; + if (!num) + host1x_debug_cont(o, "])\n"); + + return num; case HOST1X_OPCODE_NONINCR: - host1x_debug_output(o, "NONINCR(offset=%03x, [", + num = val & 0xffff; + host1x_debug_cont(o, "NONINCR(offset=%03x, [", val >> 16 & 0xfff); - return val & 0xffff; + if (!num) + host1x_debug_cont(o, "])\n"); + + return num; case HOST1X_OPCODE_MASK: mask = val & 0xffff; - host1x_debug_output(o, "MASK(offset=%03x, mask=%03x, [", + host1x_debug_cont(o, "MASK(offset=%03x, mask=%03x, [", val >> 16 & 0xfff, mask); + if (!mask) + host1x_debug_cont(o, "])\n"); + return hweight16(mask); case HOST1X_OPCODE_IMM: - host1x_debug_output(o, "IMM(offset=%03x, data=%03x)\n", + host1x_debug_cont(o, "IMM(offset=%03x, data=%03x)\n", val >> 16 & 0xfff, val & 0xffff); return 0; case HOST1X_OPCODE_RESTART: - host1x_debug_output(o, "RESTART(offset=%08x)\n", val << 4); + host1x_debug_cont(o, "RESTART(offset=%08x)\n", val << 4); return 0; case HOST1X_OPCODE_GATHER: - host1x_debug_output(o, "GATHER(offset=%03x, insert=%d, type=%d, count=%04x, addr=[", + host1x_debug_cont(o, "GATHER(offset=%03x, insert=%d, type=%d, count=%04x, addr=[", val >> 16 & 0xfff, val >> 15 & 0x1, val >> 14 & 0x1, val & 0x3fff); return 1; +#if HOST1X_HW >= 6 + case HOST1X_OPCODE_SETSTRMID: + host1x_debug_cont(o, "SETSTRMID(offset=%06x)\n", + val & 0x3fffff); + return 0; + + case HOST1X_OPCODE_SETAPPID: + host1x_debug_cont(o, "SETAPPID(appid=%02x)\n", val & 0xff); + return 0; + + case HOST1X_OPCODE_SETPYLD: + *payload = val & 0xffff; + host1x_debug_cont(o, "SETPYLD(data=%04x)\n", *payload); + return 0; + + case HOST1X_OPCODE_INCR_W: + case HOST1X_OPCODE_NONINCR_W: + host1x_debug_cont(o, "%s(offset=%06x, ", + opcode == HOST1X_OPCODE_INCR_W ? + "INCR_W" : "NONINCR_W", + val & 0x3fffff); + if (*payload == 0) { + host1x_debug_cont(o, "[])\n"); + return 0; + } else if (*payload == INVALID_PAYLOAD) { + host1x_debug_cont(o, "unknown)\n"); + return 0; + } else { + host1x_debug_cont(o, "["); + return *payload; + } + + case HOST1X_OPCODE_GATHER_W: + host1x_debug_cont(o, "GATHER_W(count=%04x, addr=[", + val & 0x3fff); + return 2; +#endif + case HOST1X_OPCODE_EXTEND: subop = val >> 24 & 0xf; if (subop == HOST1X_OPCODE_EXTEND_ACQUIRE_MLOCK) - host1x_debug_output(o, "ACQUIRE_MLOCK(index=%d)\n", + host1x_debug_cont(o, "ACQUIRE_MLOCK(index=%d)\n", val & 0xff); else if (subop == HOST1X_OPCODE_EXTEND_RELEASE_MLOCK) - host1x_debug_output(o, "RELEASE_MLOCK(index=%d)\n", + host1x_debug_cont(o, "RELEASE_MLOCK(index=%d)\n", val & 0xff); else - host1x_debug_output(o, "EXTEND_UNKNOWN(%08x)\n", val); + host1x_debug_cont(o, "EXTEND_UNKNOWN(%08x)\n", val); return 0; default: + host1x_debug_cont(o, "UNKNOWN\n"); return 0; } } -static void show_gather(struct output *o, phys_addr_t phys_addr, +static void show_gather(struct output *o, dma_addr_t phys_addr, unsigned int words, struct host1x_cdma *cdma, - phys_addr_t pin_addr, u32 *map_addr) + dma_addr_t pin_addr, u32 *map_addr) { /* Map dmaget cursor to corresponding mem handle */ u32 offset = phys_addr - pin_addr; unsigned int data_count = 0, i; + u32 payload = INVALID_PAYLOAD; /* * Sometimes we're given different hardware address to the same @@ -132,15 +176,24 @@ static void show_gather(struct output *o, phys_addr_t phys_addr, } for (i = 0; i < words; i++) { - u32 addr = phys_addr + i * 4; - u32 val = *(map_addr + offset / 4 + i); + dma_addr_t addr = phys_addr + i * 4; + u32 voffset = offset + i * 4; + u32 val; + + /* If we reach the RESTART opcode, continue at the beginning of pushbuffer */ + if (cdma && voffset >= cdma->push_buffer.size) { + addr -= cdma->push_buffer.size; + voffset -= cdma->push_buffer.size; + } + + val = *(map_addr + voffset / 4); if (!data_count) { - host1x_debug_output(o, "%08x: %08x:", addr, val); - data_count = show_channel_command(o, val); + host1x_debug_output(o, " %pad: %08x: ", &addr, val); + data_count = show_channel_command(o, val, &payload); } else { - host1x_debug_output(o, "%08x%s", val, - data_count > 0 ? ", " : "])\n"); + host1x_debug_cont(o, "%08x%s", val, + data_count > 1 ? ", " : "])\n"); data_count--; } } @@ -148,19 +201,28 @@ static void show_gather(struct output *o, phys_addr_t phys_addr, static void show_channel_gathers(struct output *o, struct host1x_cdma *cdma) { + struct push_buffer *pb = &cdma->push_buffer; struct host1x_job *job; list_for_each_entry(job, &cdma->sync_queue, list) { - int i; - host1x_debug_output(o, "\n%p: JOB, syncpt_id=%d, syncpt_val=%d, first_get=%08x, timeout=%d num_slots=%d, num_handles=%d\n", - job, job->syncpt_id, job->syncpt_end, - job->first_get, job->timeout, + unsigned int i; + + host1x_debug_output(o, "JOB, syncpt %u: %u timeout: %u num_slots: %u num_handles: %u\n", + job->syncpt->id, job->syncpt_end, job->timeout, job->num_slots, job->num_unpins); - for (i = 0; i < job->num_gathers; i++) { - struct host1x_job_gather *g = &job->gathers[i]; + show_gather(o, pb->dma + job->first_get, job->num_slots * 2, cdma, + pb->dma, pb->mapped); + + for (i = 0; i < job->num_cmds; i++) { + struct host1x_job_gather *g; u32 *mapped; + if (job->cmds[i].is_wait) + continue; + + g = &job->cmds[i].gather; + if (job->gather_copy_mapped) mapped = (u32 *)job->gather_copy_mapped; else @@ -171,10 +233,10 @@ static void show_channel_gathers(struct output *o, struct host1x_cdma *cdma) continue; } - host1x_debug_output(o, " GATHER at %08x+%04x, %d words\n", - g->base, g->offset, g->words); + host1x_debug_output(o, " GATHER at %pad+%#x, %d words\n", + &g->base, g->offset, g->words); - show_gather(o, g->base + g->offset, g->words, cdma, + show_gather(o, g->base + g->offset, g->words, NULL, g->base, mapped); if (!job->gather_copy_mapped) @@ -183,137 +245,11 @@ static void show_channel_gathers(struct output *o, struct host1x_cdma *cdma) } } -static void host1x_debug_show_channel_cdma(struct host1x *host, - struct host1x_channel *ch, - struct output *o) -{ - struct host1x_cdma *cdma = &ch->cdma; - u32 dmaput, dmaget, dmactrl; - u32 cbstat, cbread; - u32 val, base, baseval; - - dmaput = host1x_ch_readl(ch, HOST1X_CHANNEL_DMAPUT); - dmaget = host1x_ch_readl(ch, HOST1X_CHANNEL_DMAGET); - dmactrl = host1x_ch_readl(ch, HOST1X_CHANNEL_DMACTRL); - cbread = host1x_sync_readl(host, HOST1X_SYNC_CBREAD(ch->id)); - cbstat = host1x_sync_readl(host, HOST1X_SYNC_CBSTAT(ch->id)); - - host1x_debug_output(o, "%d-%s: ", ch->id, dev_name(ch->dev)); - - if (HOST1X_CHANNEL_DMACTRL_DMASTOP_V(dmactrl) || - !ch->cdma.push_buffer.mapped) { - host1x_debug_output(o, "inactive\n\n"); - return; - } - - if (HOST1X_SYNC_CBSTAT_CBCLASS_V(cbstat) == HOST1X_CLASS_HOST1X && - HOST1X_SYNC_CBSTAT_CBOFFSET_V(cbstat) == - HOST1X_UCLASS_WAIT_SYNCPT) - host1x_debug_output(o, "waiting on syncpt %d val %d\n", - cbread >> 24, cbread & 0xffffff); - else if (HOST1X_SYNC_CBSTAT_CBCLASS_V(cbstat) == - HOST1X_CLASS_HOST1X && - HOST1X_SYNC_CBSTAT_CBOFFSET_V(cbstat) == - HOST1X_UCLASS_WAIT_SYNCPT_BASE) { - - base = (cbread >> 16) & 0xff; - baseval = - host1x_sync_readl(host, HOST1X_SYNC_SYNCPT_BASE(base)); - val = cbread & 0xffff; - host1x_debug_output(o, "waiting on syncpt %d val %d (base %d = %d; offset = %d)\n", - cbread >> 24, baseval + val, base, - baseval, val); - } else - host1x_debug_output(o, "active class %02x, offset %04x, val %08x\n", - HOST1X_SYNC_CBSTAT_CBCLASS_V(cbstat), - HOST1X_SYNC_CBSTAT_CBOFFSET_V(cbstat), - cbread); - - host1x_debug_output(o, "DMAPUT %08x, DMAGET %08x, DMACTL %08x\n", - dmaput, dmaget, dmactrl); - host1x_debug_output(o, "CBREAD %08x, CBSTAT %08x\n", cbread, cbstat); - - show_channel_gathers(o, cdma); - host1x_debug_output(o, "\n"); -} - -static void host1x_debug_show_channel_fifo(struct host1x *host, - struct host1x_channel *ch, - struct output *o) -{ - u32 val, rd_ptr, wr_ptr, start, end; - unsigned int data_count = 0; - - host1x_debug_output(o, "%d: fifo:\n", ch->id); - - val = host1x_ch_readl(ch, HOST1X_CHANNEL_FIFOSTAT); - host1x_debug_output(o, "FIFOSTAT %08x\n", val); - if (HOST1X_CHANNEL_FIFOSTAT_CFEMPTY_V(val)) { - host1x_debug_output(o, "[empty]\n"); - return; - } - - host1x_sync_writel(host, 0x0, HOST1X_SYNC_CFPEEK_CTRL); - host1x_sync_writel(host, HOST1X_SYNC_CFPEEK_CTRL_ENA_F(1) | - HOST1X_SYNC_CFPEEK_CTRL_CHANNR_F(ch->id), - HOST1X_SYNC_CFPEEK_CTRL); - - val = host1x_sync_readl(host, HOST1X_SYNC_CFPEEK_PTRS); - rd_ptr = HOST1X_SYNC_CFPEEK_PTRS_CF_RD_PTR_V(val); - wr_ptr = HOST1X_SYNC_CFPEEK_PTRS_CF_WR_PTR_V(val); - - val = host1x_sync_readl(host, HOST1X_SYNC_CF_SETUP(ch->id)); - start = HOST1X_SYNC_CF_SETUP_BASE_V(val); - end = HOST1X_SYNC_CF_SETUP_LIMIT_V(val); - - do { - host1x_sync_writel(host, 0x0, HOST1X_SYNC_CFPEEK_CTRL); - host1x_sync_writel(host, HOST1X_SYNC_CFPEEK_CTRL_ENA_F(1) | - HOST1X_SYNC_CFPEEK_CTRL_CHANNR_F(ch->id) | - HOST1X_SYNC_CFPEEK_CTRL_ADDR_F(rd_ptr), - HOST1X_SYNC_CFPEEK_CTRL); - val = host1x_sync_readl(host, HOST1X_SYNC_CFPEEK_READ); - - if (!data_count) { - host1x_debug_output(o, "%08x:", val); - data_count = show_channel_command(o, val); - } else { - host1x_debug_output(o, "%08x%s", val, - data_count > 0 ? ", " : "])\n"); - data_count--; - } - - if (rd_ptr == end) - rd_ptr = start; - else - rd_ptr++; - } while (rd_ptr != wr_ptr); - - if (data_count) - host1x_debug_output(o, ", ...])\n"); - host1x_debug_output(o, "\n"); - - host1x_sync_writel(host, 0x0, HOST1X_SYNC_CFPEEK_CTRL); -} - -static void host1x_debug_show_mlocks(struct host1x *host, struct output *o) -{ - int i; - - host1x_debug_output(o, "---- mlocks ----\n"); - for (i = 0; i < host1x_syncpt_nb_mlocks(host); i++) { - u32 owner = - host1x_sync_readl(host, HOST1X_SYNC_MLOCK_OWNER(i)); - if (HOST1X_SYNC_MLOCK_OWNER_CH_OWNS_V(owner)) - host1x_debug_output(o, "%d: locked by channel %d\n", - i, HOST1X_SYNC_MLOCK_OWNER_CHID_F(owner)); - else if (HOST1X_SYNC_MLOCK_OWNER_CPU_OWNS_V(owner)) - host1x_debug_output(o, "%d: locked by cpu\n", i); - else - host1x_debug_output(o, "%d: unlocked\n", i); - } - host1x_debug_output(o, "\n"); -} +#if HOST1X_HW >= 6 +#include "debug_hw_1x06.c" +#else +#include "debug_hw_1x01.c" +#endif static const struct host1x_debug_ops host1x_debug_ops = { .show_channel_cdma = host1x_debug_show_channel_cdma, diff --git a/drivers/gpu/host1x/hw/debug_hw_1x01.c b/drivers/gpu/host1x/hw/debug_hw_1x01.c new file mode 100644 index 000000000000..85242a59fa6a --- /dev/null +++ b/drivers/gpu/host1x/hw/debug_hw_1x01.c @@ -0,0 +1,149 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2010 Google, Inc. + * Author: Erik Gilling <konkers@android.com> + * + * Copyright (C) 2011-2013 NVIDIA Corporation + */ + +#include "../dev.h" +#include "../debug.h" +#include "../cdma.h" +#include "../channel.h" + +static void host1x_debug_show_channel_cdma(struct host1x *host, + struct host1x_channel *ch, + struct output *o) +{ + struct host1x_cdma *cdma = &ch->cdma; + dma_addr_t dmastart, dmaend; + u32 dmaput, dmaget, dmactrl; + u32 cbstat, cbread; + u32 val, base, baseval; + + dmastart = host1x_ch_readl(ch, HOST1X_CHANNEL_DMASTART); + dmaend = host1x_ch_readl(ch, HOST1X_CHANNEL_DMAEND); + dmaput = host1x_ch_readl(ch, HOST1X_CHANNEL_DMAPUT); + dmaget = host1x_ch_readl(ch, HOST1X_CHANNEL_DMAGET); + dmactrl = host1x_ch_readl(ch, HOST1X_CHANNEL_DMACTRL); + cbread = host1x_sync_readl(host, HOST1X_SYNC_CBREAD(ch->id)); + cbstat = host1x_sync_readl(host, HOST1X_SYNC_CBSTAT(ch->id)); + + host1x_debug_output(o, "%u-%s: ", ch->id, dev_name(ch->dev)); + + if (HOST1X_CHANNEL_DMACTRL_DMASTOP_V(dmactrl) || + !ch->cdma.push_buffer.mapped) { + host1x_debug_output(o, "inactive\n\n"); + return; + } + + if (HOST1X_SYNC_CBSTAT_CBCLASS_V(cbstat) == HOST1X_CLASS_HOST1X && + HOST1X_SYNC_CBSTAT_CBOFFSET_V(cbstat) == + HOST1X_UCLASS_WAIT_SYNCPT) + host1x_debug_output(o, "waiting on syncpt %d val %d\n", + cbread >> 24, cbread & 0xffffff); + else if (HOST1X_SYNC_CBSTAT_CBCLASS_V(cbstat) == + HOST1X_CLASS_HOST1X && + HOST1X_SYNC_CBSTAT_CBOFFSET_V(cbstat) == + HOST1X_UCLASS_WAIT_SYNCPT_BASE) { + base = (cbread >> 16) & 0xff; + baseval = + host1x_sync_readl(host, HOST1X_SYNC_SYNCPT_BASE(base)); + val = cbread & 0xffff; + host1x_debug_output(o, "waiting on syncpt %d val %d (base %d = %d; offset = %d)\n", + cbread >> 24, baseval + val, base, + baseval, val); + } else + host1x_debug_output(o, "active class %02x, offset %04x, val %08x\n", + HOST1X_SYNC_CBSTAT_CBCLASS_V(cbstat), + HOST1X_SYNC_CBSTAT_CBOFFSET_V(cbstat), + cbread); + + host1x_debug_output(o, "DMASTART %pad, DMAEND %pad\n", &dmastart, &dmaend); + host1x_debug_output(o, "DMAPUT %08x DMAGET %08x DMACTL %08x\n", + dmaput, dmaget, dmactrl); + host1x_debug_output(o, "CBREAD %08x CBSTAT %08x\n", cbread, cbstat); + + show_channel_gathers(o, cdma); + host1x_debug_output(o, "\n"); +} + +static void host1x_debug_show_channel_fifo(struct host1x *host, + struct host1x_channel *ch, + struct output *o) +{ + u32 val, rd_ptr, wr_ptr, start, end; + unsigned int data_count = 0; + + host1x_debug_output(o, "%u: fifo:\n", ch->id); + + val = host1x_ch_readl(ch, HOST1X_CHANNEL_FIFOSTAT); + host1x_debug_output(o, "FIFOSTAT %08x\n", val); + if (HOST1X_CHANNEL_FIFOSTAT_CFEMPTY_V(val)) { + host1x_debug_output(o, "[empty]\n"); + return; + } + + host1x_sync_writel(host, 0x0, HOST1X_SYNC_CFPEEK_CTRL); + host1x_sync_writel(host, HOST1X_SYNC_CFPEEK_CTRL_ENA_F(1) | + HOST1X_SYNC_CFPEEK_CTRL_CHANNR_F(ch->id), + HOST1X_SYNC_CFPEEK_CTRL); + + val = host1x_sync_readl(host, HOST1X_SYNC_CFPEEK_PTRS); + rd_ptr = HOST1X_SYNC_CFPEEK_PTRS_CF_RD_PTR_V(val); + wr_ptr = HOST1X_SYNC_CFPEEK_PTRS_CF_WR_PTR_V(val); + + val = host1x_sync_readl(host, HOST1X_SYNC_CF_SETUP(ch->id)); + start = HOST1X_SYNC_CF_SETUP_BASE_V(val); + end = HOST1X_SYNC_CF_SETUP_LIMIT_V(val); + + do { + host1x_sync_writel(host, 0x0, HOST1X_SYNC_CFPEEK_CTRL); + host1x_sync_writel(host, HOST1X_SYNC_CFPEEK_CTRL_ENA_F(1) | + HOST1X_SYNC_CFPEEK_CTRL_CHANNR_F(ch->id) | + HOST1X_SYNC_CFPEEK_CTRL_ADDR_F(rd_ptr), + HOST1X_SYNC_CFPEEK_CTRL); + val = host1x_sync_readl(host, HOST1X_SYNC_CFPEEK_READ); + + if (!data_count) { + host1x_debug_output(o, "%08x: ", val); + data_count = show_channel_command(o, val, NULL); + } else { + host1x_debug_cont(o, "%08x%s", val, + data_count > 1 ? ", " : "])\n"); + data_count--; + } + + if (rd_ptr == end) + rd_ptr = start; + else + rd_ptr++; + } while (rd_ptr != wr_ptr); + + if (data_count) + host1x_debug_cont(o, ", ...])\n"); + host1x_debug_output(o, "\n"); + + host1x_sync_writel(host, 0x0, HOST1X_SYNC_CFPEEK_CTRL); +} + +static void host1x_debug_show_mlocks(struct host1x *host, struct output *o) +{ + unsigned int i; + + host1x_debug_output(o, "---- mlocks ----\n"); + + for (i = 0; i < host1x_syncpt_nb_mlocks(host); i++) { + u32 owner = + host1x_sync_readl(host, HOST1X_SYNC_MLOCK_OWNER(i)); + if (HOST1X_SYNC_MLOCK_OWNER_CH_OWNS_V(owner)) + host1x_debug_output(o, "%u: locked by channel %u\n", + i, HOST1X_SYNC_MLOCK_OWNER_CHID_V(owner)); + else if (HOST1X_SYNC_MLOCK_OWNER_CPU_OWNS_V(owner)) + host1x_debug_output(o, "%u: locked by cpu\n", i); + else + host1x_debug_output(o, "%u: unlocked\n", i); + } + + host1x_debug_output(o, "\n"); +} diff --git a/drivers/gpu/host1x/hw/debug_hw_1x06.c b/drivers/gpu/host1x/hw/debug_hw_1x06.c new file mode 100644 index 000000000000..9d0667879a19 --- /dev/null +++ b/drivers/gpu/host1x/hw/debug_hw_1x06.c @@ -0,0 +1,145 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2010 Google, Inc. + * Author: Erik Gilling <konkers@android.com> + * + * Copyright (C) 2011-2017 NVIDIA Corporation + */ + +#include "../dev.h" +#include "../debug.h" +#include "../cdma.h" +#include "../channel.h" + +static void host1x_debug_show_channel_cdma(struct host1x *host, + struct host1x_channel *ch, + struct output *o) +{ + struct host1x_cdma *cdma = &ch->cdma; + dma_addr_t dmastart = 0, dmaend = 0; + u32 dmaput, dmaget, dmactrl; + u32 offset, class; + u32 ch_stat; + +#if defined(CONFIG_ARCH_DMA_ADDR_T_64BIT) && HOST1X_HW >= 6 + dmastart = host1x_ch_readl(ch, HOST1X_CHANNEL_DMASTART_HI); + dmastart <<= 32; +#endif + dmastart |= host1x_ch_readl(ch, HOST1X_CHANNEL_DMASTART); + +#if defined(CONFIG_ARCH_DMA_ADDR_T_64BIT) && HOST1X_HW >= 6 + dmaend = host1x_ch_readl(ch, HOST1X_CHANNEL_DMAEND_HI); + dmaend <<= 32; +#endif + dmaend |= host1x_ch_readl(ch, HOST1X_CHANNEL_DMAEND); + + dmaput = host1x_ch_readl(ch, HOST1X_CHANNEL_DMAPUT); + dmaget = host1x_ch_readl(ch, HOST1X_CHANNEL_DMAGET); + dmactrl = host1x_ch_readl(ch, HOST1X_CHANNEL_DMACTRL); + offset = host1x_ch_readl(ch, HOST1X_CHANNEL_CMDP_OFFSET); + class = host1x_ch_readl(ch, HOST1X_CHANNEL_CMDP_CLASS); + ch_stat = host1x_ch_readl(ch, HOST1X_CHANNEL_CHANNELSTAT); + + host1x_debug_output(o, "%u-%s: ", ch->id, dev_name(ch->dev)); + + if (dmactrl & HOST1X_CHANNEL_DMACTRL_DMASTOP || + !ch->cdma.push_buffer.mapped) { + host1x_debug_output(o, "inactive\n\n"); + return; + } + + if (class == HOST1X_CLASS_HOST1X && offset == HOST1X_UCLASS_WAIT_SYNCPT) + host1x_debug_output(o, "waiting on syncpt\n"); + else + host1x_debug_output(o, "active class %02x, offset %04x\n", + class, offset); + + host1x_debug_output(o, "DMASTART %pad, DMAEND %pad\n", &dmastart, &dmaend); + host1x_debug_output(o, "DMAPUT %08x DMAGET %08x DMACTL %08x\n", + dmaput, dmaget, dmactrl); + host1x_debug_output(o, "CHANNELSTAT %02x\n", ch_stat); + + show_channel_gathers(o, cdma); + host1x_debug_output(o, "\n"); +} + +static void host1x_debug_show_channel_fifo(struct host1x *host, + struct host1x_channel *ch, + struct output *o) +{ +#if HOST1X_HW <= 6 + u32 rd_ptr, wr_ptr, start, end; + u32 payload = INVALID_PAYLOAD; + unsigned int data_count = 0; +#endif + u32 val; + + host1x_debug_output(o, "%u: fifo:\n", ch->id); + + val = host1x_ch_readl(ch, HOST1X_CHANNEL_CMDFIFO_STAT); + host1x_debug_output(o, "CMDFIFO_STAT %08x\n", val); + if (val & HOST1X_CHANNEL_CMDFIFO_STAT_EMPTY) { + host1x_debug_output(o, "[empty]\n"); + return; + } + + val = host1x_ch_readl(ch, HOST1X_CHANNEL_CMDFIFO_RDATA); + host1x_debug_output(o, "CMDFIFO_RDATA %08x\n", val); + +#if HOST1X_HW <= 6 + /* Peek pointer values are invalid during SLCG, so disable it */ + host1x_hypervisor_writel(host, 0x1, HOST1X_HV_ICG_EN_OVERRIDE); + + val = 0; + val |= HOST1X_HV_CMDFIFO_PEEK_CTRL_ENABLE; + val |= HOST1X_HV_CMDFIFO_PEEK_CTRL_CHANNEL(ch->id); + host1x_hypervisor_writel(host, val, HOST1X_HV_CMDFIFO_PEEK_CTRL); + + val = host1x_hypervisor_readl(host, HOST1X_HV_CMDFIFO_PEEK_PTRS); + rd_ptr = HOST1X_HV_CMDFIFO_PEEK_PTRS_RD_PTR_V(val); + wr_ptr = HOST1X_HV_CMDFIFO_PEEK_PTRS_WR_PTR_V(val); + + val = host1x_hypervisor_readl(host, HOST1X_HV_CMDFIFO_SETUP(ch->id)); + start = HOST1X_HV_CMDFIFO_SETUP_BASE_V(val); + end = HOST1X_HV_CMDFIFO_SETUP_LIMIT_V(val); + + do { + val = 0; + val |= HOST1X_HV_CMDFIFO_PEEK_CTRL_ENABLE; + val |= HOST1X_HV_CMDFIFO_PEEK_CTRL_CHANNEL(ch->id); + val |= HOST1X_HV_CMDFIFO_PEEK_CTRL_ADDR(rd_ptr); + host1x_hypervisor_writel(host, val, + HOST1X_HV_CMDFIFO_PEEK_CTRL); + + val = host1x_hypervisor_readl(host, + HOST1X_HV_CMDFIFO_PEEK_READ); + + if (!data_count) { + host1x_debug_output(o, "%03x 0x%08x: ", + rd_ptr - start, val); + data_count = show_channel_command(o, val, &payload); + } else { + host1x_debug_cont(o, "%08x%s", val, + data_count > 1 ? ", " : "])\n"); + data_count--; + } + + if (rd_ptr == end) + rd_ptr = start; + else + rd_ptr++; + } while (rd_ptr != wr_ptr); + + if (data_count) + host1x_debug_cont(o, ", ...])\n"); + host1x_debug_output(o, "\n"); + + host1x_hypervisor_writel(host, 0x0, HOST1X_HV_CMDFIFO_PEEK_CTRL); + host1x_hypervisor_writel(host, 0x0, HOST1X_HV_ICG_EN_OVERRIDE); +#endif +} + +static void host1x_debug_show_mlocks(struct host1x *host, struct output *o) +{ + /* TODO */ +} diff --git a/drivers/gpu/host1x/hw/host1x01.c b/drivers/gpu/host1x/hw/host1x01.c index a14e91cd1e58..8d8a117a5153 100644 --- a/drivers/gpu/host1x/hw/host1x01.c +++ b/drivers/gpu/host1x/hw/host1x01.c @@ -1,33 +1,24 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Host1x init for T20 and T30 Architecture Chips * * Copyright (c) 2011-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ /* include hw specification */ -#include "hw/host1x01.h" -#include "hw/host1x01_hardware.h" +#include "host1x01.h" +#include "host1x01_hardware.h" /* include code */ -#include "hw/cdma_hw.c" -#include "hw/channel_hw.c" -#include "hw/debug_hw.c" -#include "hw/intr_hw.c" -#include "hw/syncpt_hw.c" +#define HOST1X_HW 1 + +#include "cdma_hw.c" +#include "channel_hw.c" +#include "debug_hw.c" +#include "intr_hw.c" +#include "syncpt_hw.c" -#include "dev.h" +#include "../dev.h" int host1x01_init(struct host1x *host) { diff --git a/drivers/gpu/host1x/hw/host1x01.h b/drivers/gpu/host1x/hw/host1x01.h index 2706b6743250..6516c3f1edf5 100644 --- a/drivers/gpu/host1x/hw/host1x01.h +++ b/drivers/gpu/host1x/hw/host1x01.h @@ -1,19 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ /* * Host1x init for T20 and T30 Architecture Chips * * Copyright (c) 2011-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ #ifndef HOST1X_HOST1X01_H #define HOST1X_HOST1X01_H diff --git a/drivers/gpu/host1x/hw/host1x01_hardware.h b/drivers/gpu/host1x/hw/host1x01_hardware.h index 5f0fb866efa8..cb93d7c1808c 100644 --- a/drivers/gpu/host1x/hw/host1x01_hardware.h +++ b/drivers/gpu/host1x/hw/host1x01_hardware.h @@ -1,19 +1,8 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ /* * Tegra host1x Register Offsets for Tegra20 and Tegra30 * * Copyright (c) 2010-2013 NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ #ifndef __HOST1X_HOST1X01_HARDWARE_H @@ -26,118 +15,6 @@ #include "hw_host1x01_sync.h" #include "hw_host1x01_uclass.h" -static inline u32 host1x_class_host_wait_syncpt( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_wait_syncpt_indx_f(indx) - | host1x_uclass_wait_syncpt_thresh_f(threshold); -} - -static inline u32 host1x_class_host_load_syncpt_base( - unsigned indx, unsigned threshold) -{ - return host1x_uclass_load_syncpt_base_base_indx_f(indx) - | host1x_uclass_load_syncpt_base_value_f(threshold); -} - -static inline u32 host1x_class_host_wait_syncpt_base( - unsigned indx, unsigned base_indx, unsigned offset) -{ - return host1x_uclass_wait_syncpt_base_indx_f(indx) - | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_wait_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt_base( - unsigned base_indx, unsigned offset) -{ - return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) - | host1x_uclass_incr_syncpt_base_offset_f(offset); -} - -static inline u32 host1x_class_host_incr_syncpt( - unsigned cond, unsigned indx) -{ - return host1x_uclass_incr_syncpt_cond_f(cond) - | host1x_uclass_incr_syncpt_indx_f(indx); -} - -static inline u32 host1x_class_host_indoff_reg_write( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indbe_f(0xf) - | host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - -static inline u32 host1x_class_host_indoff_reg_read( - unsigned mod_id, unsigned offset, bool auto_inc) -{ - u32 v = host1x_uclass_indoff_indmodid_f(mod_id) - | host1x_uclass_indoff_indroffset_f(offset) - | host1x_uclass_indoff_rwn_read_v(); - if (auto_inc) - v |= host1x_uclass_indoff_autoinc_f(1); - return v; -} - - -/* cdma opcodes */ -static inline u32 host1x_opcode_setclass( - unsigned class_id, unsigned offset, unsigned mask) -{ - return (0 << 28) | (offset << 16) | (class_id << 6) | mask; -} - -static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) -{ - return (1 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) -{ - return (2 << 28) | (offset << 16) | count; -} - -static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) -{ - return (3 << 28) | (offset << 16) | mask; -} - -static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) -{ - return (4 << 28) | (offset << 16) | value; -} - -static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) -{ - return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), - host1x_class_host_incr_syncpt(cond, indx)); -} - -static inline u32 host1x_opcode_restart(unsigned address) -{ - return (5 << 28) | (address >> 4); -} - -static inline u32 host1x_opcode_gather(unsigned count) -{ - return (6 << 28) | count; -} - -static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | count; -} - -static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) -{ - return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; -} - -#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) +#include "opcodes.h" #endif diff --git a/drivers/gpu/host1x/hw/host1x02.c b/drivers/gpu/host1x/hw/host1x02.c new file mode 100644 index 000000000000..583b33c04884 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x02.c @@ -0,0 +1,33 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Host1x init for Tegra114 SoCs + * + * Copyright (c) 2013 NVIDIA Corporation. + */ + +/* include hw specification */ +#include "host1x02.h" +#include "host1x02_hardware.h" + +/* include code */ +#define HOST1X_HW 2 + +#include "cdma_hw.c" +#include "channel_hw.c" +#include "debug_hw.c" +#include "intr_hw.c" +#include "syncpt_hw.c" + +#include "../dev.h" + +int host1x02_init(struct host1x *host) +{ + host->channel_op = &host1x_channel_ops; + host->cdma_op = &host1x_cdma_ops; + host->cdma_pb_op = &host1x_pushbuffer_ops; + host->syncpt_op = &host1x_syncpt_ops; + host->intr_op = &host1x_intr_ops; + host->debug_op = &host1x_debug_ops; + + return 0; +} diff --git a/drivers/gpu/host1x/hw/host1x02.h b/drivers/gpu/host1x/hw/host1x02.h new file mode 100644 index 000000000000..7e5c3e4700d2 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x02.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Host1x init for Tegra114 SoCs + * + * Copyright (c) 2013 NVIDIA Corporation. + */ + +#ifndef HOST1X_HOST1X02_H +#define HOST1X_HOST1X02_H + +struct host1x; + +int host1x02_init(struct host1x *host); + +#endif diff --git a/drivers/gpu/host1x/hw/host1x02_hardware.h b/drivers/gpu/host1x/hw/host1x02_hardware.h new file mode 100644 index 000000000000..2d1282b9bc33 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x02_hardware.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Tegra host1x Register Offsets for Tegra114 + * + * Copyright (c) 2010-2013 NVIDIA Corporation. + */ + +#ifndef __HOST1X_HOST1X02_HARDWARE_H +#define __HOST1X_HOST1X02_HARDWARE_H + +#include <linux/types.h> +#include <linux/bitops.h> + +#include "hw_host1x02_channel.h" +#include "hw_host1x02_sync.h" +#include "hw_host1x02_uclass.h" + +#include "opcodes.h" + +#endif diff --git a/drivers/gpu/host1x/hw/host1x04.c b/drivers/gpu/host1x/hw/host1x04.c new file mode 100644 index 000000000000..26b459eb2d3e --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x04.c @@ -0,0 +1,33 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Host1x init for Tegra124 SoCs + * + * Copyright (c) 2013 NVIDIA Corporation. + */ + +/* include hw specification */ +#include "host1x04.h" +#include "host1x04_hardware.h" + +/* include code */ +#define HOST1X_HW 4 + +#include "cdma_hw.c" +#include "channel_hw.c" +#include "debug_hw.c" +#include "intr_hw.c" +#include "syncpt_hw.c" + +#include "../dev.h" + +int host1x04_init(struct host1x *host) +{ + host->channel_op = &host1x_channel_ops; + host->cdma_op = &host1x_cdma_ops; + host->cdma_pb_op = &host1x_pushbuffer_ops; + host->syncpt_op = &host1x_syncpt_ops; + host->intr_op = &host1x_intr_ops; + host->debug_op = &host1x_debug_ops; + + return 0; +} diff --git a/drivers/gpu/host1x/hw/host1x04.h b/drivers/gpu/host1x/hw/host1x04.h new file mode 100644 index 000000000000..2a1e8153c5fe --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x04.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Host1x init for Tegra124 SoCs + * + * Copyright (c) 2013 NVIDIA Corporation. + */ + +#ifndef HOST1X_HOST1X04_H +#define HOST1X_HOST1X04_H + +struct host1x; + +int host1x04_init(struct host1x *host); + +#endif diff --git a/drivers/gpu/host1x/hw/host1x04_hardware.h b/drivers/gpu/host1x/hw/host1x04_hardware.h new file mode 100644 index 000000000000..84d244e8af30 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x04_hardware.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Tegra host1x Register Offsets for Tegra124 + * + * Copyright (c) 2010-2013 NVIDIA Corporation. + */ + +#ifndef __HOST1X_HOST1X04_HARDWARE_H +#define __HOST1X_HOST1X04_HARDWARE_H + +#include <linux/types.h> +#include <linux/bitops.h> + +#include "hw_host1x04_channel.h" +#include "hw_host1x04_sync.h" +#include "hw_host1x04_uclass.h" + +#include "opcodes.h" + +#endif diff --git a/drivers/gpu/host1x/hw/host1x05.c b/drivers/gpu/host1x/hw/host1x05.c new file mode 100644 index 000000000000..6d9803343aae --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x05.c @@ -0,0 +1,33 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Host1x init for Tegra210 SoCs + * + * Copyright (c) 2015 NVIDIA Corporation. + */ + +/* include hw specification */ +#include "host1x05.h" +#include "host1x05_hardware.h" + +/* include code */ +#define HOST1X_HW 5 + +#include "cdma_hw.c" +#include "channel_hw.c" +#include "debug_hw.c" +#include "intr_hw.c" +#include "syncpt_hw.c" + +#include "../dev.h" + +int host1x05_init(struct host1x *host) +{ + host->channel_op = &host1x_channel_ops; + host->cdma_op = &host1x_cdma_ops; + host->cdma_pb_op = &host1x_pushbuffer_ops; + host->syncpt_op = &host1x_syncpt_ops; + host->intr_op = &host1x_intr_ops; + host->debug_op = &host1x_debug_ops; + + return 0; +} diff --git a/drivers/gpu/host1x/hw/host1x05.h b/drivers/gpu/host1x/hw/host1x05.h new file mode 100644 index 000000000000..addfd41e7ef3 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x05.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Host1x init for Tegra210 SoCs + * + * Copyright (c) 2015 NVIDIA Corporation. + */ + +#ifndef HOST1X_HOST1X05_H +#define HOST1X_HOST1X05_H + +struct host1x; + +int host1x05_init(struct host1x *host); + +#endif diff --git a/drivers/gpu/host1x/hw/host1x05_hardware.h b/drivers/gpu/host1x/hw/host1x05_hardware.h new file mode 100644 index 000000000000..1dcde6ec7909 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x05_hardware.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Tegra host1x Register Offsets for Tegra210 + * + * Copyright (c) 2015 NVIDIA Corporation. + */ + +#ifndef __HOST1X_HOST1X05_HARDWARE_H +#define __HOST1X_HOST1X05_HARDWARE_H + +#include <linux/types.h> +#include <linux/bitops.h> + +#include "hw_host1x05_channel.h" +#include "hw_host1x05_sync.h" +#include "hw_host1x05_uclass.h" + +#include "opcodes.h" + +#endif diff --git a/drivers/gpu/host1x/hw/host1x06.c b/drivers/gpu/host1x/hw/host1x06.c new file mode 100644 index 000000000000..844f81ae2d7e --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x06.c @@ -0,0 +1,33 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Host1x init for Tegra186 SoCs + * + * Copyright (c) 2017 NVIDIA Corporation. + */ + +/* include hw specification */ +#include "host1x06.h" +#include "host1x06_hardware.h" + +/* include code */ +#define HOST1X_HW 6 + +#include "cdma_hw.c" +#include "channel_hw.c" +#include "debug_hw.c" +#include "intr_hw.c" +#include "syncpt_hw.c" + +#include "../dev.h" + +int host1x06_init(struct host1x *host) +{ + host->channel_op = &host1x_channel_ops; + host->cdma_op = &host1x_cdma_ops; + host->cdma_pb_op = &host1x_pushbuffer_ops; + host->syncpt_op = &host1x_syncpt_ops; + host->intr_op = &host1x_intr_ops; + host->debug_op = &host1x_debug_ops; + + return 0; +} diff --git a/drivers/gpu/host1x/hw/host1x06.h b/drivers/gpu/host1x/hw/host1x06.h new file mode 100644 index 000000000000..4ea756895ca5 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x06.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Host1x init for Tegra186 SoCs + * + * Copyright (c) 2017 NVIDIA Corporation. + */ + +#ifndef HOST1X_HOST1X06_H +#define HOST1X_HOST1X06_H + +struct host1x; + +int host1x06_init(struct host1x *host); + +#endif diff --git a/drivers/gpu/host1x/hw/host1x06_hardware.h b/drivers/gpu/host1x/hw/host1x06_hardware.h new file mode 100644 index 000000000000..c05cfa7e3090 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x06_hardware.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Tegra host1x Register Offsets for Tegra186 + * + * Copyright (c) 2017 NVIDIA Corporation. + */ + +#ifndef __HOST1X_HOST1X06_HARDWARE_H +#define __HOST1X_HOST1X06_HARDWARE_H + +#include <linux/types.h> +#include <linux/bitops.h> + +#include "hw_host1x06_channel.h" +#include "hw_host1x06_uclass.h" +#include "hw_host1x06_vm.h" +#include "hw_host1x06_hypervisor.h" + +#include "opcodes.h" + +#endif diff --git a/drivers/gpu/host1x/hw/host1x07.c b/drivers/gpu/host1x/hw/host1x07.c new file mode 100644 index 000000000000..0c6f14f7ec80 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x07.c @@ -0,0 +1,33 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Host1x init for Tegra194 SoCs + * + * Copyright (c) 2018 NVIDIA Corporation. + */ + +/* include hw specification */ +#include "host1x07.h" +#include "host1x07_hardware.h" + +/* include code */ +#define HOST1X_HW 7 + +#include "cdma_hw.c" +#include "channel_hw.c" +#include "debug_hw.c" +#include "intr_hw.c" +#include "syncpt_hw.c" + +#include "../dev.h" + +int host1x07_init(struct host1x *host) +{ + host->channel_op = &host1x_channel_ops; + host->cdma_op = &host1x_cdma_ops; + host->cdma_pb_op = &host1x_pushbuffer_ops; + host->syncpt_op = &host1x_syncpt_ops; + host->intr_op = &host1x_intr_ops; + host->debug_op = &host1x_debug_ops; + + return 0; +} diff --git a/drivers/gpu/host1x/hw/host1x07.h b/drivers/gpu/host1x/hw/host1x07.h new file mode 100644 index 000000000000..419b6eaad3d8 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x07.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Host1x init for Tegra194 SoCs + * + * Copyright (c) 2018 NVIDIA Corporation. + */ + +#ifndef HOST1X_HOST1X07_H +#define HOST1X_HOST1X07_H + +struct host1x; + +int host1x07_init(struct host1x *host); + +#endif diff --git a/drivers/gpu/host1x/hw/host1x07_hardware.h b/drivers/gpu/host1x/hw/host1x07_hardware.h new file mode 100644 index 000000000000..d67364e03956 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x07_hardware.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Tegra host1x Register Offsets for Tegra194 + * + * Copyright (c) 2018 NVIDIA Corporation. + */ + +#ifndef __HOST1X_HOST1X07_HARDWARE_H +#define __HOST1X_HOST1X07_HARDWARE_H + +#include <linux/types.h> +#include <linux/bitops.h> + +#include "hw_host1x07_channel.h" +#include "hw_host1x07_uclass.h" +#include "hw_host1x07_vm.h" +#include "hw_host1x07_hypervisor.h" + +#include "opcodes.h" + +#endif diff --git a/drivers/gpu/host1x/hw/host1x08.c b/drivers/gpu/host1x/hw/host1x08.c new file mode 100644 index 000000000000..754890c34c74 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x08.c @@ -0,0 +1,33 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Host1x init for Tegra234 SoCs + * + * Copyright (c) 2022 NVIDIA Corporation. + */ + +/* include hw specification */ +#include "host1x08.h" +#include "host1x08_hardware.h" + +/* include code */ +#define HOST1X_HW 8 + +#include "cdma_hw.c" +#include "channel_hw.c" +#include "debug_hw.c" +#include "intr_hw.c" +#include "syncpt_hw.c" + +#include "../dev.h" + +int host1x08_init(struct host1x *host) +{ + host->channel_op = &host1x_channel_ops; + host->cdma_op = &host1x_cdma_ops; + host->cdma_pb_op = &host1x_pushbuffer_ops; + host->syncpt_op = &host1x_syncpt_ops; + host->intr_op = &host1x_intr_ops; + host->debug_op = &host1x_debug_ops; + + return 0; +} diff --git a/drivers/gpu/host1x/hw/host1x08.h b/drivers/gpu/host1x/hw/host1x08.h new file mode 100644 index 000000000000..a6bad56e44cf --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x08.h @@ -0,0 +1,15 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Host1x init for Tegra234 SoCs + * + * Copyright (c) 2018 NVIDIA Corporation. + */ + +#ifndef HOST1X_HOST1X08_H +#define HOST1X_HOST1X08_H + +struct host1x; + +int host1x08_init(struct host1x *host); + +#endif diff --git a/drivers/gpu/host1x/hw/host1x08_hardware.h b/drivers/gpu/host1x/hw/host1x08_hardware.h new file mode 100644 index 000000000000..936243060bff --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x08_hardware.h @@ -0,0 +1,21 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Tegra host1x Register Offsets for Tegra234 + * + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#ifndef __HOST1X_HOST1X08_HARDWARE_H +#define __HOST1X_HOST1X08_HARDWARE_H + +#include <linux/types.h> +#include <linux/bitops.h> + +#include "hw_host1x08_uclass.h" +#include "hw_host1x08_vm.h" +#include "hw_host1x08_hypervisor.h" +#include "hw_host1x08_common.h" + +#include "opcodes.h" + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x01_channel.h b/drivers/gpu/host1x/hw/hw_host1x01_channel.h index b4bc7ca4e051..8da43eaba2c8 100644 --- a/drivers/gpu/host1x/hw/hw_host1x01_channel.h +++ b/drivers/gpu/host1x/hw/hw_host1x01_channel.h @@ -1,18 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ /* * Copyright (c) 2012-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. - * */ /* diff --git a/drivers/gpu/host1x/hw/hw_host1x01_sync.h b/drivers/gpu/host1x/hw/hw_host1x01_sync.h index ac704e579977..ec95e7ae7ca5 100644 --- a/drivers/gpu/host1x/hw/hw_host1x01_sync.h +++ b/drivers/gpu/host1x/hw/hw_host1x01_sync.h @@ -1,18 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ /* * Copyright (c) 2012-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. - * */ /* @@ -131,12 +119,12 @@ static inline u32 host1x_sync_mlock_owner_r(unsigned int id) } #define HOST1X_SYNC_MLOCK_OWNER(id) \ host1x_sync_mlock_owner_r(id) -static inline u32 host1x_sync_mlock_owner_chid_f(u32 v) +static inline u32 host1x_sync_mlock_owner_chid_v(u32 v) { - return (v & 0xf) << 8; + return (v >> 8) & 0xf; } -#define HOST1X_SYNC_MLOCK_OWNER_CHID_F(v) \ - host1x_sync_mlock_owner_chid_f(v) +#define HOST1X_SYNC_MLOCK_OWNER_CHID_V(v) \ + host1x_sync_mlock_owner_chid_v(v) static inline u32 host1x_sync_mlock_owner_cpu_owns_v(u32 r) { return (r >> 1) & 0x1; diff --git a/drivers/gpu/host1x/hw/hw_host1x01_uclass.h b/drivers/gpu/host1x/hw/hw_host1x01_uclass.h index 42f3ce19ca32..1239bfd46a5e 100644 --- a/drivers/gpu/host1x/hw/hw_host1x01_uclass.h +++ b/drivers/gpu/host1x/hw/hw_host1x01_uclass.h @@ -1,18 +1,6 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ /* * Copyright (c) 2012-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. - * */ /* @@ -111,6 +99,12 @@ static inline u32 host1x_uclass_wait_syncpt_base_offset_f(u32 v) } #define HOST1X_UCLASS_WAIT_SYNCPT_BASE_OFFSET_F(v) \ host1x_uclass_wait_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_load_syncpt_base_r(void) +{ + return 0xb; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE \ + host1x_uclass_load_syncpt_base_r() static inline u32 host1x_uclass_load_syncpt_base_base_indx_f(u32 v) { return (v & 0xff) << 24; diff --git a/drivers/gpu/host1x/hw/hw_host1x02_channel.h b/drivers/gpu/host1x/hw/hw_host1x02_channel.h new file mode 100644 index 000000000000..210d317ad2b7 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x02_channel.h @@ -0,0 +1,109 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2013 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X02_CHANNEL_H +#define HOST1X_HW_HOST1X02_CHANNEL_H + +static inline u32 host1x_channel_fifostat_r(void) +{ + return 0x0; +} +#define HOST1X_CHANNEL_FIFOSTAT \ + host1x_channel_fifostat_r() +static inline u32 host1x_channel_fifostat_cfempty_v(u32 r) +{ + return (r >> 11) & 0x1; +} +#define HOST1X_CHANNEL_FIFOSTAT_CFEMPTY_V(r) \ + host1x_channel_fifostat_cfempty_v(r) +static inline u32 host1x_channel_dmastart_r(void) +{ + return 0x14; +} +#define HOST1X_CHANNEL_DMASTART \ + host1x_channel_dmastart_r() +static inline u32 host1x_channel_dmaput_r(void) +{ + return 0x18; +} +#define HOST1X_CHANNEL_DMAPUT \ + host1x_channel_dmaput_r() +static inline u32 host1x_channel_dmaget_r(void) +{ + return 0x1c; +} +#define HOST1X_CHANNEL_DMAGET \ + host1x_channel_dmaget_r() +static inline u32 host1x_channel_dmaend_r(void) +{ + return 0x20; +} +#define HOST1X_CHANNEL_DMAEND \ + host1x_channel_dmaend_r() +static inline u32 host1x_channel_dmactrl_r(void) +{ + return 0x24; +} +#define HOST1X_CHANNEL_DMACTRL \ + host1x_channel_dmactrl_r() +static inline u32 host1x_channel_dmactrl_dmastop(void) +{ + return 1 << 0; +} +#define HOST1X_CHANNEL_DMACTRL_DMASTOP \ + host1x_channel_dmactrl_dmastop() +static inline u32 host1x_channel_dmactrl_dmastop_v(u32 r) +{ + return (r >> 0) & 0x1; +} +#define HOST1X_CHANNEL_DMACTRL_DMASTOP_V(r) \ + host1x_channel_dmactrl_dmastop_v(r) +static inline u32 host1x_channel_dmactrl_dmagetrst(void) +{ + return 1 << 1; +} +#define HOST1X_CHANNEL_DMACTRL_DMAGETRST \ + host1x_channel_dmactrl_dmagetrst() +static inline u32 host1x_channel_dmactrl_dmainitget(void) +{ + return 1 << 2; +} +#define HOST1X_CHANNEL_DMACTRL_DMAINITGET \ + host1x_channel_dmactrl_dmainitget() + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x02_sync.h b/drivers/gpu/host1x/hw/hw_host1x02_sync.h new file mode 100644 index 000000000000..44b4f8379732 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x02_sync.h @@ -0,0 +1,231 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2013 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X02_SYNC_H +#define HOST1X_HW_HOST1X02_SYNC_H + +#define REGISTER_STRIDE 4 + +static inline u32 host1x_sync_syncpt_r(unsigned int id) +{ + return 0x400 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT(id) \ + host1x_sync_syncpt_r(id) +static inline u32 host1x_sync_syncpt_thresh_cpu0_int_status_r(unsigned int id) +{ + return 0x40 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(id) \ + host1x_sync_syncpt_thresh_cpu0_int_status_r(id) +static inline u32 host1x_sync_syncpt_thresh_int_disable_r(unsigned int id) +{ + return 0x60 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(id) \ + host1x_sync_syncpt_thresh_int_disable_r(id) +static inline u32 host1x_sync_syncpt_thresh_int_enable_cpu0_r(unsigned int id) +{ + return 0x68 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(id) \ + host1x_sync_syncpt_thresh_int_enable_cpu0_r(id) +static inline u32 host1x_sync_cf_setup_r(unsigned int channel) +{ + return 0x80 + channel * REGISTER_STRIDE; +} +#define HOST1X_SYNC_CF_SETUP(channel) \ + host1x_sync_cf_setup_r(channel) +static inline u32 host1x_sync_cf_setup_base_v(u32 r) +{ + return (r >> 0) & 0x3ff; +} +#define HOST1X_SYNC_CF_SETUP_BASE_V(r) \ + host1x_sync_cf_setup_base_v(r) +static inline u32 host1x_sync_cf_setup_limit_v(u32 r) +{ + return (r >> 16) & 0x3ff; +} +#define HOST1X_SYNC_CF_SETUP_LIMIT_V(r) \ + host1x_sync_cf_setup_limit_v(r) +static inline u32 host1x_sync_cmdproc_stop_r(void) +{ + return 0xac; +} +#define HOST1X_SYNC_CMDPROC_STOP \ + host1x_sync_cmdproc_stop_r() +static inline u32 host1x_sync_ch_teardown_r(void) +{ + return 0xb0; +} +#define HOST1X_SYNC_CH_TEARDOWN \ + host1x_sync_ch_teardown_r() +static inline u32 host1x_sync_usec_clk_r(void) +{ + return 0x1a4; +} +#define HOST1X_SYNC_USEC_CLK \ + host1x_sync_usec_clk_r() +static inline u32 host1x_sync_ctxsw_timeout_cfg_r(void) +{ + return 0x1a8; +} +#define HOST1X_SYNC_CTXSW_TIMEOUT_CFG \ + host1x_sync_ctxsw_timeout_cfg_r() +static inline u32 host1x_sync_ip_busy_timeout_r(void) +{ + return 0x1bc; +} +#define HOST1X_SYNC_IP_BUSY_TIMEOUT \ + host1x_sync_ip_busy_timeout_r() +static inline u32 host1x_sync_mlock_owner_r(unsigned int id) +{ + return 0x340 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_MLOCK_OWNER(id) \ + host1x_sync_mlock_owner_r(id) +static inline u32 host1x_sync_mlock_owner_chid_v(u32 v) +{ + return (v >> 8) & 0xf; +} +#define HOST1X_SYNC_MLOCK_OWNER_CHID_V(v) \ + host1x_sync_mlock_owner_chid_v(v) +static inline u32 host1x_sync_mlock_owner_cpu_owns_v(u32 r) +{ + return (r >> 1) & 0x1; +} +#define HOST1X_SYNC_MLOCK_OWNER_CPU_OWNS_V(r) \ + host1x_sync_mlock_owner_cpu_owns_v(r) +static inline u32 host1x_sync_mlock_owner_ch_owns_v(u32 r) +{ + return (r >> 0) & 0x1; +} +#define HOST1X_SYNC_MLOCK_OWNER_CH_OWNS_V(r) \ + host1x_sync_mlock_owner_ch_owns_v(r) +static inline u32 host1x_sync_syncpt_int_thresh_r(unsigned int id) +{ + return 0x500 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_INT_THRESH(id) \ + host1x_sync_syncpt_int_thresh_r(id) +static inline u32 host1x_sync_syncpt_base_r(unsigned int id) +{ + return 0x600 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_BASE(id) \ + host1x_sync_syncpt_base_r(id) +static inline u32 host1x_sync_syncpt_cpu_incr_r(unsigned int id) +{ + return 0x700 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_CPU_INCR(id) \ + host1x_sync_syncpt_cpu_incr_r(id) +static inline u32 host1x_sync_cbread_r(unsigned int channel) +{ + return 0x720 + channel * REGISTER_STRIDE; +} +#define HOST1X_SYNC_CBREAD(channel) \ + host1x_sync_cbread_r(channel) +static inline u32 host1x_sync_cfpeek_ctrl_r(void) +{ + return 0x74c; +} +#define HOST1X_SYNC_CFPEEK_CTRL \ + host1x_sync_cfpeek_ctrl_r() +static inline u32 host1x_sync_cfpeek_ctrl_addr_f(u32 v) +{ + return (v & 0x3ff) << 0; +} +#define HOST1X_SYNC_CFPEEK_CTRL_ADDR_F(v) \ + host1x_sync_cfpeek_ctrl_addr_f(v) +static inline u32 host1x_sync_cfpeek_ctrl_channr_f(u32 v) +{ + return (v & 0xf) << 16; +} +#define HOST1X_SYNC_CFPEEK_CTRL_CHANNR_F(v) \ + host1x_sync_cfpeek_ctrl_channr_f(v) +static inline u32 host1x_sync_cfpeek_ctrl_ena_f(u32 v) +{ + return (v & 0x1) << 31; +} +#define HOST1X_SYNC_CFPEEK_CTRL_ENA_F(v) \ + host1x_sync_cfpeek_ctrl_ena_f(v) +static inline u32 host1x_sync_cfpeek_read_r(void) +{ + return 0x750; +} +#define HOST1X_SYNC_CFPEEK_READ \ + host1x_sync_cfpeek_read_r() +static inline u32 host1x_sync_cfpeek_ptrs_r(void) +{ + return 0x754; +} +#define HOST1X_SYNC_CFPEEK_PTRS \ + host1x_sync_cfpeek_ptrs_r() +static inline u32 host1x_sync_cfpeek_ptrs_cf_rd_ptr_v(u32 r) +{ + return (r >> 0) & 0x3ff; +} +#define HOST1X_SYNC_CFPEEK_PTRS_CF_RD_PTR_V(r) \ + host1x_sync_cfpeek_ptrs_cf_rd_ptr_v(r) +static inline u32 host1x_sync_cfpeek_ptrs_cf_wr_ptr_v(u32 r) +{ + return (r >> 16) & 0x3ff; +} +#define HOST1X_SYNC_CFPEEK_PTRS_CF_WR_PTR_V(r) \ + host1x_sync_cfpeek_ptrs_cf_wr_ptr_v(r) +static inline u32 host1x_sync_cbstat_r(unsigned int channel) +{ + return 0x758 + channel * REGISTER_STRIDE; +} +#define HOST1X_SYNC_CBSTAT(channel) \ + host1x_sync_cbstat_r(channel) +static inline u32 host1x_sync_cbstat_cboffset_v(u32 r) +{ + return (r >> 0) & 0xffff; +} +#define HOST1X_SYNC_CBSTAT_CBOFFSET_V(r) \ + host1x_sync_cbstat_cboffset_v(r) +static inline u32 host1x_sync_cbstat_cbclass_v(u32 r) +{ + return (r >> 16) & 0x3ff; +} +#define HOST1X_SYNC_CBSTAT_CBCLASS_V(r) \ + host1x_sync_cbstat_cbclass_v(r) + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x02_uclass.h b/drivers/gpu/host1x/hw/hw_host1x02_uclass.h new file mode 100644 index 000000000000..0a2ab8f1da6f --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x02_uclass.h @@ -0,0 +1,181 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2013 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X02_UCLASS_H +#define HOST1X_HW_HOST1X02_UCLASS_H + +static inline u32 host1x_uclass_incr_syncpt_r(void) +{ + return 0x0; +} +#define HOST1X_UCLASS_INCR_SYNCPT \ + host1x_uclass_incr_syncpt_r() +static inline u32 host1x_uclass_incr_syncpt_cond_f(u32 v) +{ + return (v & 0xff) << 8; +} +#define HOST1X_UCLASS_INCR_SYNCPT_COND_F(v) \ + host1x_uclass_incr_syncpt_cond_f(v) +static inline u32 host1x_uclass_incr_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_INDX_F(v) \ + host1x_uclass_incr_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_r(void) +{ + return 0x8; +} +#define HOST1X_UCLASS_WAIT_SYNCPT \ + host1x_uclass_wait_syncpt_r() +static inline u32 host1x_uclass_wait_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_INDX_F(v) \ + host1x_uclass_wait_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_thresh_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_THRESH_F(v) \ + host1x_uclass_wait_syncpt_thresh_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_r(void) +{ + return 0x9; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE \ + host1x_uclass_wait_syncpt_base_r() +static inline u32 host1x_uclass_wait_syncpt_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 16; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_wait_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_load_syncpt_base_r(void) +{ + return 0xb; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE \ + host1x_uclass_load_syncpt_base_r() +static inline u32 host1x_uclass_load_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_load_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_load_syncpt_base_value_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_VALUE_F(v) \ + host1x_uclass_load_syncpt_base_value_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_incr_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_incr_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_indoff_r(void) +{ + return 0x2d; +} +#define HOST1X_UCLASS_INDOFF \ + host1x_uclass_indoff_r() +static inline u32 host1x_uclass_indoff_indbe_f(u32 v) +{ + return (v & 0xf) << 28; +} +#define HOST1X_UCLASS_INDOFF_INDBE_F(v) \ + host1x_uclass_indoff_indbe_f(v) +static inline u32 host1x_uclass_indoff_autoinc_f(u32 v) +{ + return (v & 0x1) << 27; +} +#define HOST1X_UCLASS_INDOFF_AUTOINC_F(v) \ + host1x_uclass_indoff_autoinc_f(v) +static inline u32 host1x_uclass_indoff_indmodid_f(u32 v) +{ + return (v & 0xff) << 18; +} +#define HOST1X_UCLASS_INDOFF_INDMODID_F(v) \ + host1x_uclass_indoff_indmodid_f(v) +static inline u32 host1x_uclass_indoff_indroffset_f(u32 v) +{ + return (v & 0xffff) << 2; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_indoff_rwn_read_v(void) +{ + return 1; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_load_syncpt_payload_32_r(void) +{ + return 0x4e; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_PAYLOAD_32 \ + host1x_uclass_load_syncpt_payload_32_r() +static inline u32 host1x_uclass_wait_syncpt_32_r(void) +{ + return 0x50; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_32 \ + host1x_uclass_wait_syncpt_32_r() + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x04_channel.h b/drivers/gpu/host1x/hw/hw_host1x04_channel.h new file mode 100644 index 000000000000..38d110645ee0 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x04_channel.h @@ -0,0 +1,121 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2013 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X04_CHANNEL_H +#define HOST1X_HW_HOST1X04_CHANNEL_H + +static inline u32 host1x_channel_fifostat_r(void) +{ + return 0x0; +} +#define HOST1X_CHANNEL_FIFOSTAT \ + host1x_channel_fifostat_r() +static inline u32 host1x_channel_fifostat_cfempty_v(u32 r) +{ + return (r >> 11) & 0x1; +} +#define HOST1X_CHANNEL_FIFOSTAT_CFEMPTY_V(r) \ + host1x_channel_fifostat_cfempty_v(r) +static inline u32 host1x_channel_dmastart_r(void) +{ + return 0x14; +} +#define HOST1X_CHANNEL_DMASTART \ + host1x_channel_dmastart_r() +static inline u32 host1x_channel_dmaput_r(void) +{ + return 0x18; +} +#define HOST1X_CHANNEL_DMAPUT \ + host1x_channel_dmaput_r() +static inline u32 host1x_channel_dmaget_r(void) +{ + return 0x1c; +} +#define HOST1X_CHANNEL_DMAGET \ + host1x_channel_dmaget_r() +static inline u32 host1x_channel_dmaend_r(void) +{ + return 0x20; +} +#define HOST1X_CHANNEL_DMAEND \ + host1x_channel_dmaend_r() +static inline u32 host1x_channel_dmactrl_r(void) +{ + return 0x24; +} +#define HOST1X_CHANNEL_DMACTRL \ + host1x_channel_dmactrl_r() +static inline u32 host1x_channel_dmactrl_dmastop(void) +{ + return 1 << 0; +} +#define HOST1X_CHANNEL_DMACTRL_DMASTOP \ + host1x_channel_dmactrl_dmastop() +static inline u32 host1x_channel_dmactrl_dmastop_v(u32 r) +{ + return (r >> 0) & 0x1; +} +#define HOST1X_CHANNEL_DMACTRL_DMASTOP_V(r) \ + host1x_channel_dmactrl_dmastop_v(r) +static inline u32 host1x_channel_dmactrl_dmagetrst(void) +{ + return 1 << 1; +} +#define HOST1X_CHANNEL_DMACTRL_DMAGETRST \ + host1x_channel_dmactrl_dmagetrst() +static inline u32 host1x_channel_dmactrl_dmainitget(void) +{ + return 1 << 2; +} +#define HOST1X_CHANNEL_DMACTRL_DMAINITGET \ + host1x_channel_dmactrl_dmainitget() +static inline u32 host1x_channel_channelctrl_r(void) +{ + return 0x98; +} +#define HOST1X_CHANNEL_CHANNELCTRL \ + host1x_channel_channelctrl_r() +static inline u32 host1x_channel_channelctrl_kernel_filter_gbuffer_f(u32 v) +{ + return (v & 0x1) << 2; +} +#define HOST1X_CHANNEL_CHANNELCTRL_KERNEL_FILTER_GBUFFER(v) \ + host1x_channel_channelctrl_kernel_filter_gbuffer_f(v) + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x04_sync.h b/drivers/gpu/host1x/hw/hw_host1x04_sync.h new file mode 100644 index 000000000000..0be98562c201 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x04_sync.h @@ -0,0 +1,231 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2013 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X04_SYNC_H +#define HOST1X_HW_HOST1X04_SYNC_H + +#define REGISTER_STRIDE 4 + +static inline u32 host1x_sync_syncpt_r(unsigned int id) +{ + return 0xf80 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT(id) \ + host1x_sync_syncpt_r(id) +static inline u32 host1x_sync_syncpt_thresh_cpu0_int_status_r(unsigned int id) +{ + return 0xe80 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(id) \ + host1x_sync_syncpt_thresh_cpu0_int_status_r(id) +static inline u32 host1x_sync_syncpt_thresh_int_disable_r(unsigned int id) +{ + return 0xf00 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(id) \ + host1x_sync_syncpt_thresh_int_disable_r(id) +static inline u32 host1x_sync_syncpt_thresh_int_enable_cpu0_r(unsigned int id) +{ + return 0xf20 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(id) \ + host1x_sync_syncpt_thresh_int_enable_cpu0_r(id) +static inline u32 host1x_sync_cf_setup_r(unsigned int channel) +{ + return 0xc00 + channel * REGISTER_STRIDE; +} +#define HOST1X_SYNC_CF_SETUP(channel) \ + host1x_sync_cf_setup_r(channel) +static inline u32 host1x_sync_cf_setup_base_v(u32 r) +{ + return (r >> 0) & 0x3ff; +} +#define HOST1X_SYNC_CF_SETUP_BASE_V(r) \ + host1x_sync_cf_setup_base_v(r) +static inline u32 host1x_sync_cf_setup_limit_v(u32 r) +{ + return (r >> 16) & 0x3ff; +} +#define HOST1X_SYNC_CF_SETUP_LIMIT_V(r) \ + host1x_sync_cf_setup_limit_v(r) +static inline u32 host1x_sync_cmdproc_stop_r(void) +{ + return 0xac; +} +#define HOST1X_SYNC_CMDPROC_STOP \ + host1x_sync_cmdproc_stop_r() +static inline u32 host1x_sync_ch_teardown_r(void) +{ + return 0xb0; +} +#define HOST1X_SYNC_CH_TEARDOWN \ + host1x_sync_ch_teardown_r() +static inline u32 host1x_sync_usec_clk_r(void) +{ + return 0x1a4; +} +#define HOST1X_SYNC_USEC_CLK \ + host1x_sync_usec_clk_r() +static inline u32 host1x_sync_ctxsw_timeout_cfg_r(void) +{ + return 0x1a8; +} +#define HOST1X_SYNC_CTXSW_TIMEOUT_CFG \ + host1x_sync_ctxsw_timeout_cfg_r() +static inline u32 host1x_sync_ip_busy_timeout_r(void) +{ + return 0x1bc; +} +#define HOST1X_SYNC_IP_BUSY_TIMEOUT \ + host1x_sync_ip_busy_timeout_r() +static inline u32 host1x_sync_mlock_owner_r(unsigned int id) +{ + return 0x340 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_MLOCK_OWNER(id) \ + host1x_sync_mlock_owner_r(id) +static inline u32 host1x_sync_mlock_owner_chid_v(u32 v) +{ + return (v >> 8) & 0xf; +} +#define HOST1X_SYNC_MLOCK_OWNER_CHID_V(v) \ + host1x_sync_mlock_owner_chid_v(v) +static inline u32 host1x_sync_mlock_owner_cpu_owns_v(u32 r) +{ + return (r >> 1) & 0x1; +} +#define HOST1X_SYNC_MLOCK_OWNER_CPU_OWNS_V(r) \ + host1x_sync_mlock_owner_cpu_owns_v(r) +static inline u32 host1x_sync_mlock_owner_ch_owns_v(u32 r) +{ + return (r >> 0) & 0x1; +} +#define HOST1X_SYNC_MLOCK_OWNER_CH_OWNS_V(r) \ + host1x_sync_mlock_owner_ch_owns_v(r) +static inline u32 host1x_sync_syncpt_int_thresh_r(unsigned int id) +{ + return 0x1380 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_INT_THRESH(id) \ + host1x_sync_syncpt_int_thresh_r(id) +static inline u32 host1x_sync_syncpt_base_r(unsigned int id) +{ + return 0x600 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_BASE(id) \ + host1x_sync_syncpt_base_r(id) +static inline u32 host1x_sync_syncpt_cpu_incr_r(unsigned int id) +{ + return 0xf60 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_CPU_INCR(id) \ + host1x_sync_syncpt_cpu_incr_r(id) +static inline u32 host1x_sync_cbread_r(unsigned int channel) +{ + return 0xc80 + channel * REGISTER_STRIDE; +} +#define HOST1X_SYNC_CBREAD(channel) \ + host1x_sync_cbread_r(channel) +static inline u32 host1x_sync_cfpeek_ctrl_r(void) +{ + return 0x74c; +} +#define HOST1X_SYNC_CFPEEK_CTRL \ + host1x_sync_cfpeek_ctrl_r() +static inline u32 host1x_sync_cfpeek_ctrl_addr_f(u32 v) +{ + return (v & 0x3ff) << 0; +} +#define HOST1X_SYNC_CFPEEK_CTRL_ADDR_F(v) \ + host1x_sync_cfpeek_ctrl_addr_f(v) +static inline u32 host1x_sync_cfpeek_ctrl_channr_f(u32 v) +{ + return (v & 0xf) << 16; +} +#define HOST1X_SYNC_CFPEEK_CTRL_CHANNR_F(v) \ + host1x_sync_cfpeek_ctrl_channr_f(v) +static inline u32 host1x_sync_cfpeek_ctrl_ena_f(u32 v) +{ + return (v & 0x1) << 31; +} +#define HOST1X_SYNC_CFPEEK_CTRL_ENA_F(v) \ + host1x_sync_cfpeek_ctrl_ena_f(v) +static inline u32 host1x_sync_cfpeek_read_r(void) +{ + return 0x750; +} +#define HOST1X_SYNC_CFPEEK_READ \ + host1x_sync_cfpeek_read_r() +static inline u32 host1x_sync_cfpeek_ptrs_r(void) +{ + return 0x754; +} +#define HOST1X_SYNC_CFPEEK_PTRS \ + host1x_sync_cfpeek_ptrs_r() +static inline u32 host1x_sync_cfpeek_ptrs_cf_rd_ptr_v(u32 r) +{ + return (r >> 0) & 0x3ff; +} +#define HOST1X_SYNC_CFPEEK_PTRS_CF_RD_PTR_V(r) \ + host1x_sync_cfpeek_ptrs_cf_rd_ptr_v(r) +static inline u32 host1x_sync_cfpeek_ptrs_cf_wr_ptr_v(u32 r) +{ + return (r >> 16) & 0x3ff; +} +#define HOST1X_SYNC_CFPEEK_PTRS_CF_WR_PTR_V(r) \ + host1x_sync_cfpeek_ptrs_cf_wr_ptr_v(r) +static inline u32 host1x_sync_cbstat_r(unsigned int channel) +{ + return 0xcc0 + channel * REGISTER_STRIDE; +} +#define HOST1X_SYNC_CBSTAT(channel) \ + host1x_sync_cbstat_r(channel) +static inline u32 host1x_sync_cbstat_cboffset_v(u32 r) +{ + return (r >> 0) & 0xffff; +} +#define HOST1X_SYNC_CBSTAT_CBOFFSET_V(r) \ + host1x_sync_cbstat_cboffset_v(r) +static inline u32 host1x_sync_cbstat_cbclass_v(u32 r) +{ + return (r >> 16) & 0x3ff; +} +#define HOST1X_SYNC_CBSTAT_CBCLASS_V(r) \ + host1x_sync_cbstat_cbclass_v(r) + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x04_uclass.h b/drivers/gpu/host1x/hw/hw_host1x04_uclass.h new file mode 100644 index 000000000000..60c692b92955 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x04_uclass.h @@ -0,0 +1,181 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2013 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X04_UCLASS_H +#define HOST1X_HW_HOST1X04_UCLASS_H + +static inline u32 host1x_uclass_incr_syncpt_r(void) +{ + return 0x0; +} +#define HOST1X_UCLASS_INCR_SYNCPT \ + host1x_uclass_incr_syncpt_r() +static inline u32 host1x_uclass_incr_syncpt_cond_f(u32 v) +{ + return (v & 0xff) << 8; +} +#define HOST1X_UCLASS_INCR_SYNCPT_COND_F(v) \ + host1x_uclass_incr_syncpt_cond_f(v) +static inline u32 host1x_uclass_incr_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_INDX_F(v) \ + host1x_uclass_incr_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_r(void) +{ + return 0x8; +} +#define HOST1X_UCLASS_WAIT_SYNCPT \ + host1x_uclass_wait_syncpt_r() +static inline u32 host1x_uclass_wait_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_INDX_F(v) \ + host1x_uclass_wait_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_thresh_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_THRESH_F(v) \ + host1x_uclass_wait_syncpt_thresh_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_r(void) +{ + return 0x9; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE \ + host1x_uclass_wait_syncpt_base_r() +static inline u32 host1x_uclass_wait_syncpt_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 16; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_wait_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_load_syncpt_base_r(void) +{ + return 0xb; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE \ + host1x_uclass_load_syncpt_base_r() +static inline u32 host1x_uclass_load_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_load_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_load_syncpt_base_value_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_VALUE_F(v) \ + host1x_uclass_load_syncpt_base_value_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_incr_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_incr_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_indoff_r(void) +{ + return 0x2d; +} +#define HOST1X_UCLASS_INDOFF \ + host1x_uclass_indoff_r() +static inline u32 host1x_uclass_indoff_indbe_f(u32 v) +{ + return (v & 0xf) << 28; +} +#define HOST1X_UCLASS_INDOFF_INDBE_F(v) \ + host1x_uclass_indoff_indbe_f(v) +static inline u32 host1x_uclass_indoff_autoinc_f(u32 v) +{ + return (v & 0x1) << 27; +} +#define HOST1X_UCLASS_INDOFF_AUTOINC_F(v) \ + host1x_uclass_indoff_autoinc_f(v) +static inline u32 host1x_uclass_indoff_indmodid_f(u32 v) +{ + return (v & 0xff) << 18; +} +#define HOST1X_UCLASS_INDOFF_INDMODID_F(v) \ + host1x_uclass_indoff_indmodid_f(v) +static inline u32 host1x_uclass_indoff_indroffset_f(u32 v) +{ + return (v & 0xffff) << 2; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_indoff_rwn_read_v(void) +{ + return 1; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_load_syncpt_payload_32_r(void) +{ + return 0x4e; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_PAYLOAD_32 \ + host1x_uclass_load_syncpt_payload_32_r() +static inline u32 host1x_uclass_wait_syncpt_32_r(void) +{ + return 0x50; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_32 \ + host1x_uclass_wait_syncpt_32_r() + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x05_channel.h b/drivers/gpu/host1x/hw/hw_host1x05_channel.h new file mode 100644 index 000000000000..7e628ef58c49 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x05_channel.h @@ -0,0 +1,121 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2015 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X05_CHANNEL_H +#define HOST1X_HW_HOST1X05_CHANNEL_H + +static inline u32 host1x_channel_fifostat_r(void) +{ + return 0x0; +} +#define HOST1X_CHANNEL_FIFOSTAT \ + host1x_channel_fifostat_r() +static inline u32 host1x_channel_fifostat_cfempty_v(u32 r) +{ + return (r >> 11) & 0x1; +} +#define HOST1X_CHANNEL_FIFOSTAT_CFEMPTY_V(r) \ + host1x_channel_fifostat_cfempty_v(r) +static inline u32 host1x_channel_dmastart_r(void) +{ + return 0x14; +} +#define HOST1X_CHANNEL_DMASTART \ + host1x_channel_dmastart_r() +static inline u32 host1x_channel_dmaput_r(void) +{ + return 0x18; +} +#define HOST1X_CHANNEL_DMAPUT \ + host1x_channel_dmaput_r() +static inline u32 host1x_channel_dmaget_r(void) +{ + return 0x1c; +} +#define HOST1X_CHANNEL_DMAGET \ + host1x_channel_dmaget_r() +static inline u32 host1x_channel_dmaend_r(void) +{ + return 0x20; +} +#define HOST1X_CHANNEL_DMAEND \ + host1x_channel_dmaend_r() +static inline u32 host1x_channel_dmactrl_r(void) +{ + return 0x24; +} +#define HOST1X_CHANNEL_DMACTRL \ + host1x_channel_dmactrl_r() +static inline u32 host1x_channel_dmactrl_dmastop(void) +{ + return 1 << 0; +} +#define HOST1X_CHANNEL_DMACTRL_DMASTOP \ + host1x_channel_dmactrl_dmastop() +static inline u32 host1x_channel_dmactrl_dmastop_v(u32 r) +{ + return (r >> 0) & 0x1; +} +#define HOST1X_CHANNEL_DMACTRL_DMASTOP_V(r) \ + host1x_channel_dmactrl_dmastop_v(r) +static inline u32 host1x_channel_dmactrl_dmagetrst(void) +{ + return 1 << 1; +} +#define HOST1X_CHANNEL_DMACTRL_DMAGETRST \ + host1x_channel_dmactrl_dmagetrst() +static inline u32 host1x_channel_dmactrl_dmainitget(void) +{ + return 1 << 2; +} +#define HOST1X_CHANNEL_DMACTRL_DMAINITGET \ + host1x_channel_dmactrl_dmainitget() +static inline u32 host1x_channel_channelctrl_r(void) +{ + return 0x98; +} +#define HOST1X_CHANNEL_CHANNELCTRL \ + host1x_channel_channelctrl_r() +static inline u32 host1x_channel_channelctrl_kernel_filter_gbuffer_f(u32 v) +{ + return (v & 0x1) << 2; +} +#define HOST1X_CHANNEL_CHANNELCTRL_KERNEL_FILTER_GBUFFER(v) \ + host1x_channel_channelctrl_kernel_filter_gbuffer_f(v) + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x05_sync.h b/drivers/gpu/host1x/hw/hw_host1x05_sync.h new file mode 100644 index 000000000000..1a85c793bd3e --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x05_sync.h @@ -0,0 +1,231 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2015 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X05_SYNC_H +#define HOST1X_HW_HOST1X05_SYNC_H + +#define REGISTER_STRIDE 4 + +static inline u32 host1x_sync_syncpt_r(unsigned int id) +{ + return 0xf80 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT(id) \ + host1x_sync_syncpt_r(id) +static inline u32 host1x_sync_syncpt_thresh_cpu0_int_status_r(unsigned int id) +{ + return 0xe80 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(id) \ + host1x_sync_syncpt_thresh_cpu0_int_status_r(id) +static inline u32 host1x_sync_syncpt_thresh_int_disable_r(unsigned int id) +{ + return 0xf00 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(id) \ + host1x_sync_syncpt_thresh_int_disable_r(id) +static inline u32 host1x_sync_syncpt_thresh_int_enable_cpu0_r(unsigned int id) +{ + return 0xf20 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(id) \ + host1x_sync_syncpt_thresh_int_enable_cpu0_r(id) +static inline u32 host1x_sync_cf_setup_r(unsigned int channel) +{ + return 0xc00 + channel * REGISTER_STRIDE; +} +#define HOST1X_SYNC_CF_SETUP(channel) \ + host1x_sync_cf_setup_r(channel) +static inline u32 host1x_sync_cf_setup_base_v(u32 r) +{ + return (r >> 0) & 0x3ff; +} +#define HOST1X_SYNC_CF_SETUP_BASE_V(r) \ + host1x_sync_cf_setup_base_v(r) +static inline u32 host1x_sync_cf_setup_limit_v(u32 r) +{ + return (r >> 16) & 0x3ff; +} +#define HOST1X_SYNC_CF_SETUP_LIMIT_V(r) \ + host1x_sync_cf_setup_limit_v(r) +static inline u32 host1x_sync_cmdproc_stop_r(void) +{ + return 0xac; +} +#define HOST1X_SYNC_CMDPROC_STOP \ + host1x_sync_cmdproc_stop_r() +static inline u32 host1x_sync_ch_teardown_r(void) +{ + return 0xb0; +} +#define HOST1X_SYNC_CH_TEARDOWN \ + host1x_sync_ch_teardown_r() +static inline u32 host1x_sync_usec_clk_r(void) +{ + return 0x1a4; +} +#define HOST1X_SYNC_USEC_CLK \ + host1x_sync_usec_clk_r() +static inline u32 host1x_sync_ctxsw_timeout_cfg_r(void) +{ + return 0x1a8; +} +#define HOST1X_SYNC_CTXSW_TIMEOUT_CFG \ + host1x_sync_ctxsw_timeout_cfg_r() +static inline u32 host1x_sync_ip_busy_timeout_r(void) +{ + return 0x1bc; +} +#define HOST1X_SYNC_IP_BUSY_TIMEOUT \ + host1x_sync_ip_busy_timeout_r() +static inline u32 host1x_sync_mlock_owner_r(unsigned int id) +{ + return 0x340 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_MLOCK_OWNER(id) \ + host1x_sync_mlock_owner_r(id) +static inline u32 host1x_sync_mlock_owner_chid_v(u32 r) +{ + return (r >> 8) & 0xf; +} +#define HOST1X_SYNC_MLOCK_OWNER_CHID_V(v) \ + host1x_sync_mlock_owner_chid_v(v) +static inline u32 host1x_sync_mlock_owner_cpu_owns_v(u32 r) +{ + return (r >> 1) & 0x1; +} +#define HOST1X_SYNC_MLOCK_OWNER_CPU_OWNS_V(r) \ + host1x_sync_mlock_owner_cpu_owns_v(r) +static inline u32 host1x_sync_mlock_owner_ch_owns_v(u32 r) +{ + return (r >> 0) & 0x1; +} +#define HOST1X_SYNC_MLOCK_OWNER_CH_OWNS_V(r) \ + host1x_sync_mlock_owner_ch_owns_v(r) +static inline u32 host1x_sync_syncpt_int_thresh_r(unsigned int id) +{ + return 0x1380 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_INT_THRESH(id) \ + host1x_sync_syncpt_int_thresh_r(id) +static inline u32 host1x_sync_syncpt_base_r(unsigned int id) +{ + return 0x600 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_BASE(id) \ + host1x_sync_syncpt_base_r(id) +static inline u32 host1x_sync_syncpt_cpu_incr_r(unsigned int id) +{ + return 0xf60 + id * REGISTER_STRIDE; +} +#define HOST1X_SYNC_SYNCPT_CPU_INCR(id) \ + host1x_sync_syncpt_cpu_incr_r(id) +static inline u32 host1x_sync_cbread_r(unsigned int channel) +{ + return 0xc80 + channel * REGISTER_STRIDE; +} +#define HOST1X_SYNC_CBREAD(channel) \ + host1x_sync_cbread_r(channel) +static inline u32 host1x_sync_cfpeek_ctrl_r(void) +{ + return 0x74c; +} +#define HOST1X_SYNC_CFPEEK_CTRL \ + host1x_sync_cfpeek_ctrl_r() +static inline u32 host1x_sync_cfpeek_ctrl_addr_f(u32 v) +{ + return (v & 0x3ff) << 0; +} +#define HOST1X_SYNC_CFPEEK_CTRL_ADDR_F(v) \ + host1x_sync_cfpeek_ctrl_addr_f(v) +static inline u32 host1x_sync_cfpeek_ctrl_channr_f(u32 v) +{ + return (v & 0xf) << 16; +} +#define HOST1X_SYNC_CFPEEK_CTRL_CHANNR_F(v) \ + host1x_sync_cfpeek_ctrl_channr_f(v) +static inline u32 host1x_sync_cfpeek_ctrl_ena_f(u32 v) +{ + return (v & 0x1) << 31; +} +#define HOST1X_SYNC_CFPEEK_CTRL_ENA_F(v) \ + host1x_sync_cfpeek_ctrl_ena_f(v) +static inline u32 host1x_sync_cfpeek_read_r(void) +{ + return 0x750; +} +#define HOST1X_SYNC_CFPEEK_READ \ + host1x_sync_cfpeek_read_r() +static inline u32 host1x_sync_cfpeek_ptrs_r(void) +{ + return 0x754; +} +#define HOST1X_SYNC_CFPEEK_PTRS \ + host1x_sync_cfpeek_ptrs_r() +static inline u32 host1x_sync_cfpeek_ptrs_cf_rd_ptr_v(u32 r) +{ + return (r >> 0) & 0x3ff; +} +#define HOST1X_SYNC_CFPEEK_PTRS_CF_RD_PTR_V(r) \ + host1x_sync_cfpeek_ptrs_cf_rd_ptr_v(r) +static inline u32 host1x_sync_cfpeek_ptrs_cf_wr_ptr_v(u32 r) +{ + return (r >> 16) & 0x3ff; +} +#define HOST1X_SYNC_CFPEEK_PTRS_CF_WR_PTR_V(r) \ + host1x_sync_cfpeek_ptrs_cf_wr_ptr_v(r) +static inline u32 host1x_sync_cbstat_r(unsigned int channel) +{ + return 0xcc0 + channel * REGISTER_STRIDE; +} +#define HOST1X_SYNC_CBSTAT(channel) \ + host1x_sync_cbstat_r(channel) +static inline u32 host1x_sync_cbstat_cboffset_v(u32 r) +{ + return (r >> 0) & 0xffff; +} +#define HOST1X_SYNC_CBSTAT_CBOFFSET_V(r) \ + host1x_sync_cbstat_cboffset_v(r) +static inline u32 host1x_sync_cbstat_cbclass_v(u32 r) +{ + return (r >> 16) & 0x3ff; +} +#define HOST1X_SYNC_CBSTAT_CBCLASS_V(r) \ + host1x_sync_cbstat_cbclass_v(r) + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x05_uclass.h b/drivers/gpu/host1x/hw/hw_host1x05_uclass.h new file mode 100644 index 000000000000..2fcc9a2ad3ef --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x05_uclass.h @@ -0,0 +1,181 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2015 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X05_UCLASS_H +#define HOST1X_HW_HOST1X05_UCLASS_H + +static inline u32 host1x_uclass_incr_syncpt_r(void) +{ + return 0x0; +} +#define HOST1X_UCLASS_INCR_SYNCPT \ + host1x_uclass_incr_syncpt_r() +static inline u32 host1x_uclass_incr_syncpt_cond_f(u32 v) +{ + return (v & 0xff) << 8; +} +#define HOST1X_UCLASS_INCR_SYNCPT_COND_F(v) \ + host1x_uclass_incr_syncpt_cond_f(v) +static inline u32 host1x_uclass_incr_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_INDX_F(v) \ + host1x_uclass_incr_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_r(void) +{ + return 0x8; +} +#define HOST1X_UCLASS_WAIT_SYNCPT \ + host1x_uclass_wait_syncpt_r() +static inline u32 host1x_uclass_wait_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_INDX_F(v) \ + host1x_uclass_wait_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_thresh_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_THRESH_F(v) \ + host1x_uclass_wait_syncpt_thresh_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_r(void) +{ + return 0x9; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE \ + host1x_uclass_wait_syncpt_base_r() +static inline u32 host1x_uclass_wait_syncpt_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 16; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_wait_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_load_syncpt_base_r(void) +{ + return 0xb; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE \ + host1x_uclass_load_syncpt_base_r() +static inline u32 host1x_uclass_load_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_load_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_load_syncpt_base_value_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_VALUE_F(v) \ + host1x_uclass_load_syncpt_base_value_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_incr_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_incr_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_indoff_r(void) +{ + return 0x2d; +} +#define HOST1X_UCLASS_INDOFF \ + host1x_uclass_indoff_r() +static inline u32 host1x_uclass_indoff_indbe_f(u32 v) +{ + return (v & 0xf) << 28; +} +#define HOST1X_UCLASS_INDOFF_INDBE_F(v) \ + host1x_uclass_indoff_indbe_f(v) +static inline u32 host1x_uclass_indoff_autoinc_f(u32 v) +{ + return (v & 0x1) << 27; +} +#define HOST1X_UCLASS_INDOFF_AUTOINC_F(v) \ + host1x_uclass_indoff_autoinc_f(v) +static inline u32 host1x_uclass_indoff_indmodid_f(u32 v) +{ + return (v & 0xff) << 18; +} +#define HOST1X_UCLASS_INDOFF_INDMODID_F(v) \ + host1x_uclass_indoff_indmodid_f(v) +static inline u32 host1x_uclass_indoff_indroffset_f(u32 v) +{ + return (v & 0xffff) << 2; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_indoff_rwn_read_v(void) +{ + return 1; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_load_syncpt_payload_32_r(void) +{ + return 0x4e; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_PAYLOAD_32 \ + host1x_uclass_load_syncpt_payload_32_r() +static inline u32 host1x_uclass_wait_syncpt_32_r(void) +{ + return 0x50; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_32 \ + host1x_uclass_wait_syncpt_32_r() + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x06_channel.h b/drivers/gpu/host1x/hw/hw_host1x06_channel.h new file mode 100644 index 000000000000..18ae1c57bbea --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x06_channel.h @@ -0,0 +1,11 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2019 NVIDIA Corporation. + */ + +#ifndef HOST1X_HW_HOST1X06_CHANNEL_H +#define HOST1X_HW_HOST1X06_CHANNEL_H + +#define HOST1X_CHANNEL_SMMU_STREAMID 0x084 + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x06_hypervisor.h b/drivers/gpu/host1x/hw/hw_host1x06_hypervisor.h new file mode 100644 index 000000000000..a7fc9ec4bc3e --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x06_hypervisor.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2017 NVIDIA Corporation. + */ + +#define HOST1X_HV_SYNCPT_PROT_EN 0x1ac4 +#define HOST1X_HV_SYNCPT_PROT_EN_CH_EN BIT(1) +#define HOST1X_HV_CH_KERNEL_FILTER_GBUFFER(x) (0x2020 + (x * 4)) +#define HOST1X_HV_CMDFIFO_PEEK_CTRL 0x233c +#define HOST1X_HV_CMDFIFO_PEEK_CTRL_ADDR(x) (x) +#define HOST1X_HV_CMDFIFO_PEEK_CTRL_CHANNEL(x) ((x) << 16) +#define HOST1X_HV_CMDFIFO_PEEK_CTRL_ENABLE BIT(31) +#define HOST1X_HV_CMDFIFO_PEEK_READ 0x2340 +#define HOST1X_HV_CMDFIFO_PEEK_PTRS 0x2344 +#define HOST1X_HV_CMDFIFO_PEEK_PTRS_WR_PTR_V(x) (((x) >> 16) & 0xfff) +#define HOST1X_HV_CMDFIFO_PEEK_PTRS_RD_PTR_V(x) ((x) & 0xfff) +#define HOST1X_HV_CMDFIFO_SETUP(x) (0x2588 + (x * 4)) +#define HOST1X_HV_CMDFIFO_SETUP_LIMIT_V(x) (((x) >> 16) & 0xfff) +#define HOST1X_HV_CMDFIFO_SETUP_BASE_V(x) ((x) & 0xfff) +#define HOST1X_HV_ICG_EN_OVERRIDE 0x2aa8 diff --git a/drivers/gpu/host1x/hw/hw_host1x06_uclass.h b/drivers/gpu/host1x/hw/hw_host1x06_uclass.h new file mode 100644 index 000000000000..50c32de452fb --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x06_uclass.h @@ -0,0 +1,181 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2017 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X06_UCLASS_H +#define HOST1X_HW_HOST1X06_UCLASS_H + +static inline u32 host1x_uclass_incr_syncpt_r(void) +{ + return 0x0; +} +#define HOST1X_UCLASS_INCR_SYNCPT \ + host1x_uclass_incr_syncpt_r() +static inline u32 host1x_uclass_incr_syncpt_cond_f(u32 v) +{ + return (v & 0xff) << 10; +} +#define HOST1X_UCLASS_INCR_SYNCPT_COND_F(v) \ + host1x_uclass_incr_syncpt_cond_f(v) +static inline u32 host1x_uclass_incr_syncpt_indx_f(u32 v) +{ + return (v & 0x3ff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_INDX_F(v) \ + host1x_uclass_incr_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_r(void) +{ + return 0x8; +} +#define HOST1X_UCLASS_WAIT_SYNCPT \ + host1x_uclass_wait_syncpt_r() +static inline u32 host1x_uclass_wait_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_INDX_F(v) \ + host1x_uclass_wait_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_thresh_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_THRESH_F(v) \ + host1x_uclass_wait_syncpt_thresh_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_r(void) +{ + return 0x9; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE \ + host1x_uclass_wait_syncpt_base_r() +static inline u32 host1x_uclass_wait_syncpt_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 16; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_wait_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_load_syncpt_base_r(void) +{ + return 0xb; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE \ + host1x_uclass_load_syncpt_base_r() +static inline u32 host1x_uclass_load_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_load_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_load_syncpt_base_value_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_VALUE_F(v) \ + host1x_uclass_load_syncpt_base_value_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_incr_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_incr_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_indoff_r(void) +{ + return 0x2d; +} +#define HOST1X_UCLASS_INDOFF \ + host1x_uclass_indoff_r() +static inline u32 host1x_uclass_indoff_indbe_f(u32 v) +{ + return (v & 0xf) << 28; +} +#define HOST1X_UCLASS_INDOFF_INDBE_F(v) \ + host1x_uclass_indoff_indbe_f(v) +static inline u32 host1x_uclass_indoff_autoinc_f(u32 v) +{ + return (v & 0x1) << 27; +} +#define HOST1X_UCLASS_INDOFF_AUTOINC_F(v) \ + host1x_uclass_indoff_autoinc_f(v) +static inline u32 host1x_uclass_indoff_indmodid_f(u32 v) +{ + return (v & 0xff) << 18; +} +#define HOST1X_UCLASS_INDOFF_INDMODID_F(v) \ + host1x_uclass_indoff_indmodid_f(v) +static inline u32 host1x_uclass_indoff_indroffset_f(u32 v) +{ + return (v & 0xffff) << 2; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_indoff_rwn_read_v(void) +{ + return 1; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_load_syncpt_payload_32_r(void) +{ + return 0x4e; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_PAYLOAD_32 \ + host1x_uclass_load_syncpt_payload_32_r() +static inline u32 host1x_uclass_wait_syncpt_32_r(void) +{ + return 0x50; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_32 \ + host1x_uclass_wait_syncpt_32_r() + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x06_vm.h b/drivers/gpu/host1x/hw/hw_host1x06_vm.h new file mode 100644 index 000000000000..818564a76bc6 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x06_vm.h @@ -0,0 +1,35 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2017 NVIDIA Corporation. + */ + +#define HOST1X_CHANNEL_DMASTART 0x0000 +#define HOST1X_CHANNEL_DMASTART_HI 0x0004 +#define HOST1X_CHANNEL_DMAPUT 0x0008 +#define HOST1X_CHANNEL_DMAPUT_HI 0x000c +#define HOST1X_CHANNEL_DMAGET 0x0010 +#define HOST1X_CHANNEL_DMAGET_HI 0x0014 +#define HOST1X_CHANNEL_DMAEND 0x0018 +#define HOST1X_CHANNEL_DMAEND_HI 0x001c +#define HOST1X_CHANNEL_DMACTRL 0x0020 +#define HOST1X_CHANNEL_DMACTRL_DMASTOP BIT(0) +#define HOST1X_CHANNEL_DMACTRL_DMAGETRST BIT(1) +#define HOST1X_CHANNEL_DMACTRL_DMAINITGET BIT(2) +#define HOST1X_CHANNEL_CMDFIFO_STAT 0x0024 +#define HOST1X_CHANNEL_CMDFIFO_STAT_EMPTY BIT(13) +#define HOST1X_CHANNEL_CMDFIFO_RDATA 0x0028 +#define HOST1X_CHANNEL_CMDP_OFFSET 0x0030 +#define HOST1X_CHANNEL_CMDP_CLASS 0x0034 +#define HOST1X_CHANNEL_CHANNELSTAT 0x0038 +#define HOST1X_CHANNEL_CMDPROC_STOP 0x0048 +#define HOST1X_CHANNEL_TEARDOWN 0x004c + +#define HOST1X_SYNC_SYNCPT_CPU_INCR(x) (0x6400 + 4*(x)) +#define HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(x) (0x6464 + 4*(x)) +#define HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(x) (0x652c + 4*(x)) +#define HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(x) (0x6590 + 4*(x)) +#define HOST1X_SYNC_SYNCPT_BASE(x) (0x8000 + 4*(x)) +#define HOST1X_SYNC_SYNCPT(x) (0x8080 + 4*(x)) +#define HOST1X_SYNC_SYNCPT_INT_THRESH(x) (0x8a00 + 4*(x)) +#define HOST1X_SYNC_SYNCPT_CH_APP(x) (0x9384 + 4*(x)) +#define HOST1X_SYNC_SYNCPT_CH_APP_CH(v) (((v) & 0x3f) << 8) diff --git a/drivers/gpu/host1x/hw/hw_host1x07_channel.h b/drivers/gpu/host1x/hw/hw_host1x07_channel.h new file mode 100644 index 000000000000..96fa72bbd7ab --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x07_channel.h @@ -0,0 +1,11 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (C) 2019 NVIDIA Corporation. + */ + +#ifndef HOST1X_HW_HOST1X07_CHANNEL_H +#define HOST1X_HW_HOST1X07_CHANNEL_H + +#define HOST1X_CHANNEL_SMMU_STREAMID 0x084 + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x07_hypervisor.h b/drivers/gpu/host1x/hw/hw_host1x07_hypervisor.h new file mode 100644 index 000000000000..52141d53954a --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x07_hypervisor.h @@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2018 NVIDIA Corporation. + */ + +#define HOST1X_HV_SYNCPT_PROT_EN 0x1ac4 +#define HOST1X_HV_SYNCPT_PROT_EN_CH_EN BIT(1) +#define HOST1X_HV_CH_KERNEL_FILTER_GBUFFER(x) (0x2020 + (x * 4)) +#define HOST1X_HV_CMDFIFO_PEEK_CTRL 0x233c +#define HOST1X_HV_CMDFIFO_PEEK_CTRL_ADDR(x) (x) +#define HOST1X_HV_CMDFIFO_PEEK_CTRL_CHANNEL(x) ((x) << 16) +#define HOST1X_HV_CMDFIFO_PEEK_CTRL_ENABLE BIT(31) +#define HOST1X_HV_CMDFIFO_PEEK_READ 0x2340 +#define HOST1X_HV_CMDFIFO_PEEK_PTRS 0x2344 +#define HOST1X_HV_CMDFIFO_PEEK_PTRS_WR_PTR_V(x) (((x) >> 16) & 0xfff) +#define HOST1X_HV_CMDFIFO_PEEK_PTRS_RD_PTR_V(x) ((x) & 0xfff) +#define HOST1X_HV_CMDFIFO_SETUP(x) (0x2588 + (x * 4)) +#define HOST1X_HV_CMDFIFO_SETUP_LIMIT_V(x) (((x) >> 16) & 0xfff) +#define HOST1X_HV_CMDFIFO_SETUP_BASE_V(x) ((x) & 0xfff) +#define HOST1X_HV_ICG_EN_OVERRIDE 0x2aa8 diff --git a/drivers/gpu/host1x/hw/hw_host1x07_uclass.h b/drivers/gpu/host1x/hw/hw_host1x07_uclass.h new file mode 100644 index 000000000000..887b878f92f7 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x07_uclass.h @@ -0,0 +1,181 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2018 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X07_UCLASS_H +#define HOST1X_HW_HOST1X07_UCLASS_H + +static inline u32 host1x_uclass_incr_syncpt_r(void) +{ + return 0x0; +} +#define HOST1X_UCLASS_INCR_SYNCPT \ + host1x_uclass_incr_syncpt_r() +static inline u32 host1x_uclass_incr_syncpt_cond_f(u32 v) +{ + return (v & 0xff) << 10; +} +#define HOST1X_UCLASS_INCR_SYNCPT_COND_F(v) \ + host1x_uclass_incr_syncpt_cond_f(v) +static inline u32 host1x_uclass_incr_syncpt_indx_f(u32 v) +{ + return (v & 0x3ff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_INDX_F(v) \ + host1x_uclass_incr_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_r(void) +{ + return 0x8; +} +#define HOST1X_UCLASS_WAIT_SYNCPT \ + host1x_uclass_wait_syncpt_r() +static inline u32 host1x_uclass_wait_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_INDX_F(v) \ + host1x_uclass_wait_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_thresh_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_THRESH_F(v) \ + host1x_uclass_wait_syncpt_thresh_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_r(void) +{ + return 0x9; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE \ + host1x_uclass_wait_syncpt_base_r() +static inline u32 host1x_uclass_wait_syncpt_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 16; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_wait_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_load_syncpt_base_r(void) +{ + return 0xb; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE \ + host1x_uclass_load_syncpt_base_r() +static inline u32 host1x_uclass_load_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_load_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_load_syncpt_base_value_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_VALUE_F(v) \ + host1x_uclass_load_syncpt_base_value_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_incr_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_incr_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_indoff_r(void) +{ + return 0x2d; +} +#define HOST1X_UCLASS_INDOFF \ + host1x_uclass_indoff_r() +static inline u32 host1x_uclass_indoff_indbe_f(u32 v) +{ + return (v & 0xf) << 28; +} +#define HOST1X_UCLASS_INDOFF_INDBE_F(v) \ + host1x_uclass_indoff_indbe_f(v) +static inline u32 host1x_uclass_indoff_autoinc_f(u32 v) +{ + return (v & 0x1) << 27; +} +#define HOST1X_UCLASS_INDOFF_AUTOINC_F(v) \ + host1x_uclass_indoff_autoinc_f(v) +static inline u32 host1x_uclass_indoff_indmodid_f(u32 v) +{ + return (v & 0xff) << 18; +} +#define HOST1X_UCLASS_INDOFF_INDMODID_F(v) \ + host1x_uclass_indoff_indmodid_f(v) +static inline u32 host1x_uclass_indoff_indroffset_f(u32 v) +{ + return (v & 0xffff) << 2; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_indoff_rwn_read_v(void) +{ + return 1; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_load_syncpt_payload_32_r(void) +{ + return 0x4e; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_PAYLOAD_32 \ + host1x_uclass_load_syncpt_payload_32_r() +static inline u32 host1x_uclass_wait_syncpt_32_r(void) +{ + return 0x50; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_32 \ + host1x_uclass_wait_syncpt_32_r() + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x07_vm.h b/drivers/gpu/host1x/hw/hw_host1x07_vm.h new file mode 100644 index 000000000000..b766851d5b83 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x07_vm.h @@ -0,0 +1,34 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2018 NVIDIA Corporation. + */ + +#define HOST1X_CHANNEL_DMASTART 0x0000 +#define HOST1X_CHANNEL_DMASTART_HI 0x0004 +#define HOST1X_CHANNEL_DMAPUT 0x0008 +#define HOST1X_CHANNEL_DMAPUT_HI 0x000c +#define HOST1X_CHANNEL_DMAGET 0x0010 +#define HOST1X_CHANNEL_DMAGET_HI 0x0014 +#define HOST1X_CHANNEL_DMAEND 0x0018 +#define HOST1X_CHANNEL_DMAEND_HI 0x001c +#define HOST1X_CHANNEL_DMACTRL 0x0020 +#define HOST1X_CHANNEL_DMACTRL_DMASTOP BIT(0) +#define HOST1X_CHANNEL_DMACTRL_DMAGETRST BIT(1) +#define HOST1X_CHANNEL_DMACTRL_DMAINITGET BIT(2) +#define HOST1X_CHANNEL_CMDFIFO_STAT 0x0024 +#define HOST1X_CHANNEL_CMDFIFO_STAT_EMPTY BIT(13) +#define HOST1X_CHANNEL_CMDFIFO_RDATA 0x0028 +#define HOST1X_CHANNEL_CMDP_OFFSET 0x0030 +#define HOST1X_CHANNEL_CMDP_CLASS 0x0034 +#define HOST1X_CHANNEL_CHANNELSTAT 0x0038 +#define HOST1X_CHANNEL_CMDPROC_STOP 0x0048 +#define HOST1X_CHANNEL_TEARDOWN 0x004c + +#define HOST1X_SYNC_SYNCPT_CPU_INCR(x) (0x6400 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(x) (0x6464 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(x) (0x652c + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(x) (0x6590 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT(x) (0x8080 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_INT_THRESH(x) (0x9980 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_CH_APP(x) (0xa604 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_CH_APP_CH(v) (((v) & 0x3f) << 8) diff --git a/drivers/gpu/host1x/hw/hw_host1x08_channel.h b/drivers/gpu/host1x/hw/hw_host1x08_channel.h new file mode 100644 index 000000000000..c9272d2ab14a --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_channel.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#ifndef HOST1X_HW_HOST1X08_CHANNEL_H +#define HOST1X_HW_HOST1X08_CHANNEL_H + +#define HOST1X_CHANNEL_SMMU_STREAMID 0x084 + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x08_common.h b/drivers/gpu/host1x/hw/hw_host1x08_common.h new file mode 100644 index 000000000000..8e0c99150ec2 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_common.h @@ -0,0 +1,11 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#define HOST1X_COMMON_OFA_MLOCK 0x4050 +#define HOST1X_COMMON_NVJPG1_MLOCK 0x4070 +#define HOST1X_COMMON_VIC_MLOCK 0x4078 +#define HOST1X_COMMON_NVENC_MLOCK 0x407c +#define HOST1X_COMMON_NVDEC_MLOCK 0x4080 +#define HOST1X_COMMON_NVJPG_MLOCK 0x4084 diff --git a/drivers/gpu/host1x/hw/hw_host1x08_hypervisor.h b/drivers/gpu/host1x/hw/hw_host1x08_hypervisor.h new file mode 100644 index 000000000000..22964324c914 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_hypervisor.h @@ -0,0 +1,9 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#define HOST1X_HV_SYNCPT_PROT_EN 0x1724 +#define HOST1X_HV_SYNCPT_PROT_EN_CH_EN BIT(1) +#define HOST1X_HV_CH_MLOCK_EN(x) (0x1700 + (x * 4)) +#define HOST1X_HV_CH_KERNEL_FILTER_GBUFFER(x) (0x1710 + (x * 4)) diff --git a/drivers/gpu/host1x/hw/hw_host1x08_uclass.h b/drivers/gpu/host1x/hw/hw_host1x08_uclass.h new file mode 100644 index 000000000000..4fb1d090edae --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_uclass.h @@ -0,0 +1,181 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2018 NVIDIA Corporation. + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X08_UCLASS_H +#define HOST1X_HW_HOST1X08_UCLASS_H + +static inline u32 host1x_uclass_incr_syncpt_r(void) +{ + return 0x0; +} +#define HOST1X_UCLASS_INCR_SYNCPT \ + host1x_uclass_incr_syncpt_r() +static inline u32 host1x_uclass_incr_syncpt_cond_f(u32 v) +{ + return (v & 0xff) << 10; +} +#define HOST1X_UCLASS_INCR_SYNCPT_COND_F(v) \ + host1x_uclass_incr_syncpt_cond_f(v) +static inline u32 host1x_uclass_incr_syncpt_indx_f(u32 v) +{ + return (v & 0x3ff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_INDX_F(v) \ + host1x_uclass_incr_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_r(void) +{ + return 0x8; +} +#define HOST1X_UCLASS_WAIT_SYNCPT \ + host1x_uclass_wait_syncpt_r() +static inline u32 host1x_uclass_wait_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_INDX_F(v) \ + host1x_uclass_wait_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_thresh_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_THRESH_F(v) \ + host1x_uclass_wait_syncpt_thresh_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_r(void) +{ + return 0x9; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE \ + host1x_uclass_wait_syncpt_base_r() +static inline u32 host1x_uclass_wait_syncpt_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 16; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_wait_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_load_syncpt_base_r(void) +{ + return 0xb; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE \ + host1x_uclass_load_syncpt_base_r() +static inline u32 host1x_uclass_load_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_load_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_load_syncpt_base_value_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_VALUE_F(v) \ + host1x_uclass_load_syncpt_base_value_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_incr_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_incr_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_indoff_r(void) +{ + return 0x2d; +} +#define HOST1X_UCLASS_INDOFF \ + host1x_uclass_indoff_r() +static inline u32 host1x_uclass_indoff_indbe_f(u32 v) +{ + return (v & 0xf) << 28; +} +#define HOST1X_UCLASS_INDOFF_INDBE_F(v) \ + host1x_uclass_indoff_indbe_f(v) +static inline u32 host1x_uclass_indoff_autoinc_f(u32 v) +{ + return (v & 0x1) << 27; +} +#define HOST1X_UCLASS_INDOFF_AUTOINC_F(v) \ + host1x_uclass_indoff_autoinc_f(v) +static inline u32 host1x_uclass_indoff_indmodid_f(u32 v) +{ + return (v & 0xff) << 18; +} +#define HOST1X_UCLASS_INDOFF_INDMODID_F(v) \ + host1x_uclass_indoff_indmodid_f(v) +static inline u32 host1x_uclass_indoff_indroffset_f(u32 v) +{ + return (v & 0xffff) << 2; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_indoff_rwn_read_v(void) +{ + return 1; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_load_syncpt_payload_32_r(void) +{ + return 0x4e; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_PAYLOAD_32 \ + host1x_uclass_load_syncpt_payload_32_r() +static inline u32 host1x_uclass_wait_syncpt_32_r(void) +{ + return 0x50; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_32 \ + host1x_uclass_wait_syncpt_32_r() + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x08_vm.h b/drivers/gpu/host1x/hw/hw_host1x08_vm.h new file mode 100644 index 000000000000..1455a4670bf8 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x08_vm.h @@ -0,0 +1,36 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#define HOST1X_CHANNEL_DMASTART 0x0000 +#define HOST1X_CHANNEL_DMASTART_HI 0x0004 +#define HOST1X_CHANNEL_DMAPUT 0x0008 +#define HOST1X_CHANNEL_DMAPUT_HI 0x000c +#define HOST1X_CHANNEL_DMAGET 0x0010 +#define HOST1X_CHANNEL_DMAGET_HI 0x0014 +#define HOST1X_CHANNEL_DMAEND 0x0018 +#define HOST1X_CHANNEL_DMAEND_HI 0x001c +#define HOST1X_CHANNEL_DMACTRL 0x0020 +#define HOST1X_CHANNEL_DMACTRL_DMASTOP BIT(0) +#define HOST1X_CHANNEL_DMACTRL_DMAGETRST BIT(1) +#define HOST1X_CHANNEL_DMACTRL_DMAINITGET BIT(2) +#define HOST1X_CHANNEL_CMDFIFO_STAT 0x0024 +#define HOST1X_CHANNEL_CMDFIFO_STAT_EMPTY BIT(13) +#define HOST1X_CHANNEL_CMDFIFO_RDATA 0x0028 +#define HOST1X_CHANNEL_CMDP_OFFSET 0x0030 +#define HOST1X_CHANNEL_CMDP_CLASS 0x0034 +#define HOST1X_CHANNEL_CHANNELSTAT 0x0038 +#define HOST1X_CHANNEL_CMDPROC_STOP 0x0048 +#define HOST1X_CHANNEL_TEARDOWN 0x004c +#define HOST1X_CHANNEL_SMMU_STREAMID 0x0084 + +#define HOST1X_SYNC_SYNCPT_CPU_INCR(x) (0x6400 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(x) (0x6600 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_INTR_DEST(x) (0x6684 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(x) (0x770c + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(x) (0x7790 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT(x) (0x8080 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_INT_THRESH(x) (0xa088 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_CH_APP(x) (0xb090 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_CH_APP_CH(v) (((v) & 0x3f) << 8) diff --git a/drivers/gpu/host1x/hw/intr_hw.c b/drivers/gpu/host1x/hw/intr_hw.c index b592eef1efcb..bd5b5ef62f35 100644 --- a/drivers/gpu/host1x/hw/intr_hw.c +++ b/drivers/gpu/host1x/hw/intr_hw.c @@ -1,71 +1,83 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Tegra host1x Interrupt Management * * Copyright (C) 2010 Google, Inc. * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ -#include <linux/interrupt.h> -#include <linux/irq.h> #include <linux/io.h> -#include <asm/mach/irq.h> -#include "intr.h" -#include "dev.h" +#include "../intr.h" +#include "../dev.h" -/* - * Sync point threshold interrupt service function - * Handles sync point threshold triggers, in interrupt context - */ -static void host1x_intr_syncpt_handle(struct host1x_syncpt *syncpt) +static void process_32_syncpts(struct host1x *host, unsigned long val, u32 reg_offset) { - unsigned int id = syncpt->id; - struct host1x *host = syncpt->host; + unsigned int id; + + if (!val) + return; - host1x_sync_writel(host, BIT_MASK(id), - HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(BIT_WORD(id))); - host1x_sync_writel(host, BIT_MASK(id), - HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(BIT_WORD(id))); + host1x_sync_writel(host, val, HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(reg_offset)); + host1x_sync_writel(host, val, HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(reg_offset)); - queue_work(host->intr_wq, &syncpt->intr.work); + for_each_set_bit(id, &val, 32) + host1x_intr_handle_interrupt(host, reg_offset * 32 + id); } static irqreturn_t syncpt_thresh_isr(int irq, void *dev_id) { - struct host1x *host = dev_id; + struct host1x_intr_irq_data *irq_data = dev_id; + struct host1x *host = irq_data->host; unsigned long reg; - int i, id; + unsigned int i; - for (i = 0; i <= BIT_WORD(host->info->nb_pts); i++) { +#if !defined(CONFIG_64BIT) + for (i = irq_data->offset; i < DIV_ROUND_UP(host->info->nb_pts, 32); + i += host->num_syncpt_irqs) { reg = host1x_sync_readl(host, HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(i)); - for_each_set_bit(id, ®, BITS_PER_LONG) { - struct host1x_syncpt *syncpt = - host->syncpt + (i * BITS_PER_LONG + id); - host1x_intr_syncpt_handle(syncpt); - } + + process_32_syncpts(host, reg, i); + } +#elif HOST1X_HW == 6 || HOST1X_HW == 7 + /* + * Tegra186 and Tegra194 have the first INT_STATUS register not 64-bit aligned, + * and only have one interrupt line. + */ + reg = host1x_sync_readl(host, HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(0)); + process_32_syncpts(host, reg, 0); + + for (i = 1; i < (host->info->nb_pts / 32) - 1; i += 2) { + reg = host1x_sync_readq(host, + HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(i)); + + process_32_syncpts(host, lower_32_bits(reg), i); + process_32_syncpts(host, upper_32_bits(reg), i + 1); + } + + reg = host1x_sync_readl(host, HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(i)); + process_32_syncpts(host, reg, i); +#else + /* All 64-bit capable SoCs have number of syncpoints divisible by 64 */ + for (i = irq_data->offset; i < DIV_ROUND_UP(host->info->nb_pts, 64); + i += host->num_syncpt_irqs) { + reg = host1x_sync_readq(host, + HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(i * 2)); + + process_32_syncpts(host, lower_32_bits(reg), i * 2 + 0); + process_32_syncpts(host, upper_32_bits(reg), i * 2 + 1); } +#endif return IRQ_HANDLED; } -static void _host1x_intr_disable_all_syncpt_intrs(struct host1x *host) +static void host1x_intr_disable_all_syncpt_intrs(struct host1x *host) { - u32 i; + unsigned int i; - for (i = 0; i <= BIT_WORD(host->info->nb_pts); ++i) { + for (i = 0; i < DIV_ROUND_UP(host->info->nb_pts, 32); ++i) { host1x_sync_writel(host, 0xffffffffu, HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(i)); host1x_sync_writel(host, 0xffffffffu, @@ -73,24 +85,10 @@ static void _host1x_intr_disable_all_syncpt_intrs(struct host1x *host) } } -static int _host1x_intr_init_host_sync(struct host1x *host, u32 cpm, - void (*syncpt_thresh_work)(struct work_struct *)) +static int +host1x_intr_init_host_sync(struct host1x *host, u32 cpm) { - int i, err; - - host1x_hw_intr_disable_all_syncpt_intrs(host); - - for (i = 0; i < host->info->nb_pts; i++) - INIT_WORK(&host->syncpt[i].intr.work, syncpt_thresh_work); - - err = devm_request_irq(host->dev, host->intr_syncpt_irq, - syncpt_thresh_isr, IRQF_SHARED, - "host1x_syncpt", host); - if (IS_ERR_VALUE(err)) { - WARN_ON(1); - return err; - } - +#if HOST1X_HW < 6 /* disable the ip_busy_timeout. this prevents write drops */ host1x_sync_writel(host, 0, HOST1X_SYNC_IP_BUSY_TIMEOUT); @@ -102,42 +100,55 @@ static int _host1x_intr_init_host_sync(struct host1x *host, u32 cpm, /* update host clocks per usec */ host1x_sync_writel(host, cpm, HOST1X_SYNC_USEC_CLK); +#endif +#if HOST1X_HW >= 8 + u32 id; + + /* + * Program threshold interrupt destination among 8 lines per VM, + * per syncpoint. For each group of 64 syncpoints (corresponding to two + * interrupt status registers), direct to one interrupt line, going + * around in a round robin fashion. + */ + for (id = 0; id < host->info->nb_pts; id++) { + u32 reg_offset = id / 64; + u32 irq_index = reg_offset % host->num_syncpt_irqs; + + host1x_sync_writel(host, irq_index, HOST1X_SYNC_SYNCPT_INTR_DEST(id)); + } +#endif return 0; } -static void _host1x_intr_set_syncpt_threshold(struct host1x *host, - u32 id, u32 thresh) +static void host1x_intr_set_syncpt_threshold(struct host1x *host, + unsigned int id, + u32 thresh) { host1x_sync_writel(host, thresh, HOST1X_SYNC_SYNCPT_INT_THRESH(id)); } -static void _host1x_intr_enable_syncpt_intr(struct host1x *host, u32 id) +static void host1x_intr_enable_syncpt_intr(struct host1x *host, + unsigned int id) { - host1x_sync_writel(host, BIT_MASK(id), - HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(BIT_WORD(id))); + host1x_sync_writel(host, BIT(id % 32), + HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(id / 32)); } -static void _host1x_intr_disable_syncpt_intr(struct host1x *host, u32 id) +static void host1x_intr_disable_syncpt_intr(struct host1x *host, + unsigned int id) { - host1x_sync_writel(host, BIT_MASK(id), - HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(BIT_WORD(id))); - host1x_sync_writel(host, BIT_MASK(id), - HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(BIT_WORD(id))); -} - -static int _host1x_free_syncpt_irq(struct host1x *host) -{ - devm_free_irq(host->dev, host->intr_syncpt_irq, host); - flush_workqueue(host->intr_wq); - return 0; + host1x_sync_writel(host, BIT(id % 32), + HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(id / 32)); + host1x_sync_writel(host, BIT(id % 32), + HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(id / 32)); } static const struct host1x_intr_ops host1x_intr_ops = { - .init_host_sync = _host1x_intr_init_host_sync, - .set_syncpt_threshold = _host1x_intr_set_syncpt_threshold, - .enable_syncpt_intr = _host1x_intr_enable_syncpt_intr, - .disable_syncpt_intr = _host1x_intr_disable_syncpt_intr, - .disable_all_syncpt_intrs = _host1x_intr_disable_all_syncpt_intrs, - .free_syncpt_irq = _host1x_free_syncpt_irq, + .init_host_sync = host1x_intr_init_host_sync, + .set_syncpt_threshold = host1x_intr_set_syncpt_threshold, + .enable_syncpt_intr = host1x_intr_enable_syncpt_intr, + .disable_syncpt_intr = host1x_intr_disable_syncpt_intr, + .disable_all_syncpt_intrs = host1x_intr_disable_all_syncpt_intrs, + .isr = syncpt_thresh_isr, }; diff --git a/drivers/gpu/host1x/hw/opcodes.h b/drivers/gpu/host1x/hw/opcodes.h new file mode 100644 index 000000000000..649614499b04 --- /dev/null +++ b/drivers/gpu/host1x/hw/opcodes.h @@ -0,0 +1,150 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Tegra host1x opcodes + * + * Copyright (c) 2022 NVIDIA Corporation. + */ + +#ifndef __HOST1X_OPCODES_H +#define __HOST1X_OPCODES_H + +#include <linux/types.h> + +static inline u32 host1x_class_host_wait_syncpt( + unsigned indx, unsigned threshold) +{ + return host1x_uclass_wait_syncpt_indx_f(indx) + | host1x_uclass_wait_syncpt_thresh_f(threshold); +} + +static inline u32 host1x_class_host_load_syncpt_base( + unsigned indx, unsigned threshold) +{ + return host1x_uclass_load_syncpt_base_base_indx_f(indx) + | host1x_uclass_load_syncpt_base_value_f(threshold); +} + +static inline u32 host1x_class_host_wait_syncpt_base( + unsigned indx, unsigned base_indx, unsigned offset) +{ + return host1x_uclass_wait_syncpt_base_indx_f(indx) + | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) + | host1x_uclass_wait_syncpt_base_offset_f(offset); +} + +static inline u32 host1x_class_host_incr_syncpt_base( + unsigned base_indx, unsigned offset) +{ + return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) + | host1x_uclass_incr_syncpt_base_offset_f(offset); +} + +static inline u32 host1x_class_host_incr_syncpt( + unsigned cond, unsigned indx) +{ + return host1x_uclass_incr_syncpt_cond_f(cond) + | host1x_uclass_incr_syncpt_indx_f(indx); +} + +static inline u32 host1x_class_host_indoff_reg_write( + unsigned mod_id, unsigned offset, bool auto_inc) +{ + u32 v = host1x_uclass_indoff_indbe_f(0xf) + | host1x_uclass_indoff_indmodid_f(mod_id) + | host1x_uclass_indoff_indroffset_f(offset); + if (auto_inc) + v |= host1x_uclass_indoff_autoinc_f(1); + return v; +} + +static inline u32 host1x_class_host_indoff_reg_read( + unsigned mod_id, unsigned offset, bool auto_inc) +{ + u32 v = host1x_uclass_indoff_indmodid_f(mod_id) + | host1x_uclass_indoff_indroffset_f(offset) + | host1x_uclass_indoff_rwn_read_v(); + if (auto_inc) + v |= host1x_uclass_indoff_autoinc_f(1); + return v; +} + +static inline u32 host1x_opcode_setclass( + unsigned class_id, unsigned offset, unsigned mask) +{ + return (0 << 28) | (offset << 16) | (class_id << 6) | mask; +} + +static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) +{ + return (1 << 28) | (offset << 16) | count; +} + +static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) +{ + return (2 << 28) | (offset << 16) | count; +} + +static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) +{ + return (3 << 28) | (offset << 16) | mask; +} + +static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) +{ + return (4 << 28) | (offset << 16) | value; +} + +static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) +{ + return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), + host1x_class_host_incr_syncpt(cond, indx)); +} + +static inline u32 host1x_opcode_restart(unsigned address) +{ + return (5 << 28) | (address >> 4); +} + +static inline u32 host1x_opcode_gather(unsigned count) +{ + return (6 << 28) | count; +} + +static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) +{ + return (6 << 28) | (offset << 16) | BIT(15) | count; +} + +static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) +{ + return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; +} + +static inline u32 host1x_opcode_setstreamid(unsigned streamid) +{ + return (7 << 28) | streamid; +} + +static inline u32 host1x_opcode_setpayload(unsigned payload) +{ + return (9 << 28) | payload; +} + +static inline u32 host1x_opcode_gather_wide(unsigned count) +{ + return (12 << 28) | count; +} + +static inline u32 host1x_opcode_acquire_mlock(unsigned mlock) +{ + return (14 << 28) | (0 << 24) | mlock; +} + +static inline u32 host1x_opcode_release_mlock(unsigned mlock) +{ + return (14 << 28) | (1 << 24) | mlock; +} + +#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) + +#endif diff --git a/drivers/gpu/host1x/hw/syncpt_hw.c b/drivers/gpu/host1x/hw/syncpt_hw.c index 0cf6095d3367..8cf35b2eff3d 100644 --- a/drivers/gpu/host1x/hw/syncpt_hw.c +++ b/drivers/gpu/host1x/hw/syncpt_hw.c @@ -1,33 +1,23 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Tegra host1x Syncpoints * * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ #include <linux/io.h> -#include "dev.h" -#include "syncpt.h" +#include "../dev.h" +#include "../syncpt.h" /* * Write the current syncpoint value back to hw. */ static void syncpt_restore(struct host1x_syncpt *sp) { + u32 min = host1x_syncpt_read_min(sp); struct host1x *host = sp->host; - int min = host1x_syncpt_read_min(sp); + host1x_sync_writel(host, min, HOST1X_SYNC_SYNCPT(sp->id)); } @@ -36,9 +26,12 @@ static void syncpt_restore(struct host1x_syncpt *sp) */ static void syncpt_restore_wait_base(struct host1x_syncpt *sp) { +#if HOST1X_HW < 7 struct host1x *host = sp->host; + host1x_sync_writel(host, sp->base_val, HOST1X_SYNC_SYNCPT_BASE(sp->id)); +#endif } /* @@ -46,9 +39,12 @@ static void syncpt_restore_wait_base(struct host1x_syncpt *sp) */ static void syncpt_read_wait_base(struct host1x_syncpt *sp) { +#if HOST1X_HW < 7 struct host1x *host = sp->host; + sp->base_val = host1x_sync_readl(host, HOST1X_SYNC_SYNCPT_BASE(sp->id)); +#endif } /* @@ -85,21 +81,53 @@ static int syncpt_cpu_incr(struct host1x_syncpt *sp) if (!host1x_syncpt_client_managed(sp) && host1x_syncpt_idle(sp)) return -EINVAL; - host1x_sync_writel(host, BIT_MASK(sp->id), + + host1x_sync_writel(host, BIT(sp->id % 32), HOST1X_SYNC_SYNCPT_CPU_INCR(reg_offset)); wmb(); return 0; } -/* remove a wait pointed to by patch_addr */ -static int syncpt_patch_wait(struct host1x_syncpt *sp, void *patch_addr) +/** + * syncpt_assign_to_channel() - Assign syncpoint to channel + * @sp: syncpoint + * @ch: channel + * + * On chips with the syncpoint protection feature (Tegra186+), assign @sp to + * @ch, preventing other channels from incrementing the syncpoints. If @ch is + * NULL, unassigns the syncpoint. + * + * On older chips, do nothing. + */ +static void syncpt_assign_to_channel(struct host1x_syncpt *sp, + struct host1x_channel *ch) +{ +#if HOST1X_HW >= 6 + struct host1x *host = sp->host; + + host1x_sync_writel(host, + HOST1X_SYNC_SYNCPT_CH_APP_CH(ch ? ch->id : 0xff), + HOST1X_SYNC_SYNCPT_CH_APP(sp->id)); +#endif +} + +/** + * syncpt_enable_protection() - Enable syncpoint protection + * @host: host1x instance + * + * On chips with the syncpoint protection feature (Tegra186+), enable this + * feature. On older chips, do nothing. + */ +static void syncpt_enable_protection(struct host1x *host) { - u32 override = host1x_class_host_wait_syncpt( - HOST1X_SYNCPT_RESERVED, 0); +#if HOST1X_HW >= 6 + if (!host->hv_regs) + return; - *((u32 *)patch_addr) = override; - return 0; + host1x_hypervisor_writel(host, HOST1X_HV_SYNCPT_PROT_EN_CH_EN, + HOST1X_HV_SYNCPT_PROT_EN); +#endif } static const struct host1x_syncpt_ops host1x_syncpt_ops = { @@ -108,5 +136,6 @@ static const struct host1x_syncpt_ops host1x_syncpt_ops = { .load_wait_base = syncpt_read_wait_base, .load = syncpt_load, .cpu_incr = syncpt_cpu_incr, - .patch_wait = syncpt_patch_wait, + .assign_to_channel = syncpt_assign_to_channel, + .enable_protection = syncpt_enable_protection, }; diff --git a/drivers/gpu/host1x/intr.c b/drivers/gpu/host1x/intr.c index 2491bf82e30c..f77a678949e9 100644 --- a/drivers/gpu/host1x/intr.c +++ b/drivers/gpu/host1x/intr.c @@ -1,305 +1,138 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Tegra host1x Interrupt Management * - * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (c) 2010-2021, NVIDIA Corporation. */ #include <linux/clk.h> #include <linux/interrupt.h> -#include <linux/slab.h> -#include <linux/irq.h> - -#include <trace/events/host1x.h> -#include "channel.h" #include "dev.h" +#include "fence.h" #include "intr.h" -/* Wait list management */ - -enum waitlist_state { - WLS_PENDING, - WLS_REMOVED, - WLS_CANCELLED, - WLS_HANDLED -}; - -static void waiter_release(struct kref *kref) -{ - kfree(container_of(kref, struct host1x_waitlist, refcount)); -} - -/* - * add a waiter to a waiter queue, sorted by threshold - * returns true if it was added at the head of the queue - */ -static bool add_waiter_to_queue(struct host1x_waitlist *waiter, - struct list_head *queue) +static void host1x_intr_add_fence_to_list(struct host1x_fence_list *list, + struct host1x_syncpt_fence *fence) { - struct host1x_waitlist *pos; - u32 thresh = waiter->thresh; + struct host1x_syncpt_fence *fence_in_list; - list_for_each_entry_reverse(pos, queue, list) - if ((s32)(pos->thresh - thresh) <= 0) { - list_add(&waiter->list, &pos->list); - return false; + list_for_each_entry_reverse(fence_in_list, &list->list, list) { + if ((s32)(fence_in_list->threshold - fence->threshold) <= 0) { + /* Fence in list is before us, we can insert here */ + list_add(&fence->list, &fence_in_list->list); + return; } + } - list_add(&waiter->list, queue); - return true; + /* Add as first in list */ + list_add(&fence->list, &list->list); } -/* - * run through a waiter queue for a single sync point ID - * and gather all completed waiters into lists by actions - */ -static void remove_completed_waiters(struct list_head *head, u32 sync, - struct list_head completed[HOST1X_INTR_ACTION_COUNT]) +static void host1x_intr_update_hw_state(struct host1x *host, struct host1x_syncpt *sp) { - struct list_head *dest; - struct host1x_waitlist *waiter, *next, *prev; + struct host1x_syncpt_fence *fence; - list_for_each_entry_safe(waiter, next, head, list) { - if ((s32)(waiter->thresh - sync) > 0) - break; + if (!list_empty(&sp->fences.list)) { + fence = list_first_entry(&sp->fences.list, struct host1x_syncpt_fence, list); - dest = completed + waiter->action; - - /* consolidate submit cleanups */ - if (waiter->action == HOST1X_INTR_ACTION_SUBMIT_COMPLETE && - !list_empty(dest)) { - prev = list_entry(dest->prev, - struct host1x_waitlist, list); - if (prev->data == waiter->data) { - prev->count++; - dest = NULL; - } - } - - /* PENDING->REMOVED or CANCELLED->HANDLED */ - if (atomic_inc_return(&waiter->state) == WLS_HANDLED || !dest) { - list_del(&waiter->list); - kref_put(&waiter->refcount, waiter_release); - } else - list_move_tail(&waiter->list, dest); + host1x_hw_intr_set_syncpt_threshold(host, sp->id, fence->threshold); + host1x_hw_intr_enable_syncpt_intr(host, sp->id); + } else { + host1x_hw_intr_disable_syncpt_intr(host, sp->id); } } -static void reset_threshold_interrupt(struct host1x *host, - struct list_head *head, - unsigned int id) +void host1x_intr_add_fence_locked(struct host1x *host, struct host1x_syncpt_fence *fence) { - u32 thresh = - list_first_entry(head, struct host1x_waitlist, list)->thresh; + struct host1x_fence_list *fence_list = &fence->sp->fences; - host1x_hw_intr_set_syncpt_threshold(host, id, thresh); - host1x_hw_intr_enable_syncpt_intr(host, id); -} + INIT_LIST_HEAD(&fence->list); -static void action_submit_complete(struct host1x_waitlist *waiter) -{ - struct host1x_channel *channel = waiter->data; - - host1x_cdma_update(&channel->cdma); - - /* Add nr_completed to trace */ - trace_host1x_channel_submit_complete(dev_name(channel->dev), - waiter->count, waiter->thresh); - -} - -static void action_wakeup(struct host1x_waitlist *waiter) -{ - wait_queue_head_t *wq = waiter->data; - wake_up(wq); + host1x_intr_add_fence_to_list(fence_list, fence); + host1x_intr_update_hw_state(host, fence->sp); } -static void action_wakeup_interruptible(struct host1x_waitlist *waiter) +bool host1x_intr_remove_fence(struct host1x *host, struct host1x_syncpt_fence *fence) { - wait_queue_head_t *wq = waiter->data; - wake_up_interruptible(wq); -} + struct host1x_fence_list *fence_list = &fence->sp->fences; + unsigned long irqflags; -typedef void (*action_handler)(struct host1x_waitlist *waiter); + spin_lock_irqsave(&fence_list->lock, irqflags); -static action_handler action_handlers[HOST1X_INTR_ACTION_COUNT] = { - action_submit_complete, - action_wakeup, - action_wakeup_interruptible, -}; - -static void run_handlers(struct list_head completed[HOST1X_INTR_ACTION_COUNT]) -{ - struct list_head *head = completed; - int i; - - for (i = 0; i < HOST1X_INTR_ACTION_COUNT; ++i, ++head) { - action_handler handler = action_handlers[i]; - struct host1x_waitlist *waiter, *next; - - list_for_each_entry_safe(waiter, next, head, list) { - list_del(&waiter->list); - handler(waiter); - WARN_ON(atomic_xchg(&waiter->state, WLS_HANDLED) != - WLS_REMOVED); - kref_put(&waiter->refcount, waiter_release); - } + if (list_empty(&fence->list)) { + spin_unlock_irqrestore(&fence_list->lock, irqflags); + return false; } -} - -/* - * Remove & handle all waiters that have completed for the given syncpt - */ -static int process_wait_list(struct host1x *host, - struct host1x_syncpt *syncpt, - u32 threshold) -{ - struct list_head completed[HOST1X_INTR_ACTION_COUNT]; - unsigned int i; - int empty; - - for (i = 0; i < HOST1X_INTR_ACTION_COUNT; ++i) - INIT_LIST_HEAD(completed + i); - - spin_lock(&syncpt->intr.lock); - - remove_completed_waiters(&syncpt->intr.wait_head, threshold, - completed); - empty = list_empty(&syncpt->intr.wait_head); - if (empty) - host1x_hw_intr_disable_syncpt_intr(host, syncpt->id); - else - reset_threshold_interrupt(host, &syncpt->intr.wait_head, - syncpt->id); + list_del_init(&fence->list); + host1x_intr_update_hw_state(host, fence->sp); - spin_unlock(&syncpt->intr.lock); + spin_unlock_irqrestore(&fence_list->lock, irqflags); - run_handlers(completed); - - return empty; -} - -/* - * Sync point threshold interrupt service thread function - * Handles sync point threshold triggers, in thread context - */ - -static void syncpt_thresh_work(struct work_struct *work) -{ - struct host1x_syncpt_intr *syncpt_intr = - container_of(work, struct host1x_syncpt_intr, work); - struct host1x_syncpt *syncpt = - container_of(syncpt_intr, struct host1x_syncpt, intr); - unsigned int id = syncpt->id; - struct host1x *host = syncpt->host; - - (void)process_wait_list(host, syncpt, - host1x_syncpt_load(host->syncpt + id)); + return true; } -int host1x_intr_add_action(struct host1x *host, u32 id, u32 thresh, - enum host1x_intr_action action, void *data, - struct host1x_waitlist *waiter, void **ref) +void host1x_intr_handle_interrupt(struct host1x *host, unsigned int id) { - struct host1x_syncpt *syncpt; - int queue_was_empty; - - if (waiter == NULL) { - pr_warn("%s: NULL waiter\n", __func__); - return -EINVAL; - } + struct host1x_syncpt *sp = &host->syncpt[id]; + struct host1x_syncpt_fence *fence, *tmp; + unsigned int value; - /* initialize a new waiter */ - INIT_LIST_HEAD(&waiter->list); - kref_init(&waiter->refcount); - if (ref) - kref_get(&waiter->refcount); - waiter->thresh = thresh; - waiter->action = action; - atomic_set(&waiter->state, WLS_PENDING); - waiter->data = data; - waiter->count = 1; + value = host1x_syncpt_load(sp); - syncpt = host->syncpt + id; + spin_lock(&sp->fences.lock); - spin_lock(&syncpt->intr.lock); - - queue_was_empty = list_empty(&syncpt->intr.wait_head); - - if (add_waiter_to_queue(waiter, &syncpt->intr.wait_head)) { - /* added at head of list - new threshold value */ - host1x_hw_intr_set_syncpt_threshold(host, id, thresh); + list_for_each_entry_safe(fence, tmp, &sp->fences.list, list) { + if (((value - fence->threshold) & 0x80000000U) != 0U) { + /* Fence is not yet expired, we are done */ + break; + } - /* added as first waiter - enable interrupt */ - if (queue_was_empty) - host1x_hw_intr_enable_syncpt_intr(host, id); + list_del_init(&fence->list); + host1x_fence_signal(fence); } - spin_unlock(&syncpt->intr.lock); + /* Re-enable interrupt if necessary */ + host1x_intr_update_hw_state(host, sp); - if (ref) - *ref = waiter; - return 0; + spin_unlock(&sp->fences.lock); } -void host1x_intr_put_ref(struct host1x *host, u32 id, void *ref) +int host1x_intr_init(struct host1x *host) { - struct host1x_waitlist *waiter = ref; - struct host1x_syncpt *syncpt; - - while (atomic_cmpxchg(&waiter->state, WLS_PENDING, WLS_CANCELLED) == - WLS_REMOVED) - schedule(); + struct host1x_intr_irq_data *irq_data; + unsigned int id; + int i, err; - syncpt = host->syncpt + id; - (void)process_wait_list(host, syncpt, - host1x_syncpt_load(host->syncpt + id)); + for (id = 0; id < host1x_syncpt_nb_pts(host); ++id) { + struct host1x_syncpt *syncpt = &host->syncpt[id]; - kref_put(&waiter->refcount, waiter_release); -} - -int host1x_intr_init(struct host1x *host, unsigned int irq_sync) -{ - unsigned int id; - u32 nb_pts = host1x_syncpt_nb_pts(host); + spin_lock_init(&syncpt->fences.lock); + INIT_LIST_HEAD(&syncpt->fences.list); + } - mutex_init(&host->intr_mutex); - host->intr_syncpt_irq = irq_sync; - host->intr_wq = create_workqueue("host_syncpt"); - if (!host->intr_wq) + irq_data = devm_kcalloc(host->dev, host->num_syncpt_irqs, sizeof(irq_data[0]), GFP_KERNEL); + if (!irq_data) return -ENOMEM; - for (id = 0; id < nb_pts; ++id) { - struct host1x_syncpt *syncpt = host->syncpt + id; + host1x_hw_intr_disable_all_syncpt_intrs(host); - spin_lock_init(&syncpt->intr.lock); - INIT_LIST_HEAD(&syncpt->intr.wait_head); - snprintf(syncpt->intr.thresh_irq_name, - sizeof(syncpt->intr.thresh_irq_name), - "host1x_sp_%02d", id); - } + for (i = 0; i < host->num_syncpt_irqs; i++) { + irq_data[i].host = host; + irq_data[i].offset = i; - host1x_intr_start(host); + err = devm_request_irq(host->dev, host->syncpt_irqs[i], + host->intr_op->isr, IRQF_SHARED, + "host1x_syncpt", &irq_data[i]); + if (err < 0) + return err; + } return 0; } void host1x_intr_deinit(struct host1x *host) { - host1x_intr_stop(host); - destroy_workqueue(host->intr_wq); } void host1x_intr_start(struct host1x *host) @@ -308,8 +141,7 @@ void host1x_intr_start(struct host1x *host) int err; mutex_lock(&host->intr_mutex); - err = host1x_hw_intr_init_host_sync(host, DIV_ROUND_UP(hz, 1000000), - syncpt_thresh_work); + err = host1x_hw_intr_init_host_sync(host, DIV_ROUND_UP(hz, 1000000)); if (err) { mutex_unlock(&host->intr_mutex); return; @@ -319,36 +151,5 @@ void host1x_intr_start(struct host1x *host) void host1x_intr_stop(struct host1x *host) { - unsigned int id; - struct host1x_syncpt *syncpt = host->syncpt; - u32 nb_pts = host1x_syncpt_nb_pts(host); - - mutex_lock(&host->intr_mutex); - host1x_hw_intr_disable_all_syncpt_intrs(host); - - for (id = 0; id < nb_pts; ++id) { - struct host1x_waitlist *waiter, *next; - - list_for_each_entry_safe(waiter, next, - &syncpt[id].intr.wait_head, list) { - if (atomic_cmpxchg(&waiter->state, - WLS_CANCELLED, WLS_HANDLED) == WLS_CANCELLED) { - list_del(&waiter->list); - kref_put(&waiter->refcount, waiter_release); - } - } - - if (!list_empty(&syncpt[id].intr.wait_head)) { - /* output diagnostics */ - mutex_unlock(&host->intr_mutex); - pr_warn("%s cannot stop syncpt intr id=%d\n", - __func__, id); - return; - } - } - - host1x_hw_intr_free_syncpt_irq(host); - - mutex_unlock(&host->intr_mutex); } diff --git a/drivers/gpu/host1x/intr.h b/drivers/gpu/host1x/intr.h index 2b8adf016a05..11cdf13e32fe 100644 --- a/drivers/gpu/host1x/intr.h +++ b/drivers/gpu/host1x/intr.h @@ -1,93 +1,23 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ /* * Tegra host1x Interrupt Management * - * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (c) 2010-2021, NVIDIA Corporation. */ #ifndef __HOST1X_INTR_H #define __HOST1X_INTR_H -#include <linux/interrupt.h> -#include <linux/workqueue.h> - struct host1x; +struct host1x_syncpt_fence; -enum host1x_intr_action { - /* - * Perform cleanup after a submit has completed. - * 'data' points to a channel - */ - HOST1X_INTR_ACTION_SUBMIT_COMPLETE = 0, - - /* - * Wake up a task. - * 'data' points to a wait_queue_head_t - */ - HOST1X_INTR_ACTION_WAKEUP, - - /* - * Wake up a interruptible task. - * 'data' points to a wait_queue_head_t - */ - HOST1X_INTR_ACTION_WAKEUP_INTERRUPTIBLE, - - HOST1X_INTR_ACTION_COUNT -}; - -struct host1x_syncpt_intr { - spinlock_t lock; - struct list_head wait_head; - char thresh_irq_name[12]; - struct work_struct work; -}; - -struct host1x_waitlist { - struct list_head list; - struct kref refcount; - u32 thresh; - enum host1x_intr_action action; - atomic_t state; - void *data; - int count; +struct host1x_intr_irq_data { + struct host1x *host; + u32 offset; }; -/* - * Schedule an action to be taken when a sync point reaches the given threshold. - * - * @id the sync point - * @thresh the threshold - * @action the action to take - * @data a pointer to extra data depending on action, see above - * @waiter waiter structure - assumes ownership - * @ref must be passed if cancellation is possible, else NULL - * - * This is a non-blocking api. - */ -int host1x_intr_add_action(struct host1x *host, u32 id, u32 thresh, - enum host1x_intr_action action, void *data, - struct host1x_waitlist *waiter, void **ref); - -/* - * Unreference an action submitted to host1x_intr_add_action(). - * You must call this if you passed non-NULL as ref. - * @ref the ref returned from host1x_intr_add_action() - */ -void host1x_intr_put_ref(struct host1x *host, u32 id, void *ref); - /* Initialize host1x sync point interrupt */ -int host1x_intr_init(struct host1x *host, unsigned int irq_sync); +int host1x_intr_init(struct host1x *host); /* Deinitialize host1x sync point interrupt */ void host1x_intr_deinit(struct host1x *host); @@ -98,5 +28,10 @@ void host1x_intr_start(struct host1x *host); /* Disable host1x sync point interrupt */ void host1x_intr_stop(struct host1x *host); -irqreturn_t host1x_syncpt_thresh_fn(void *dev_id); +void host1x_intr_handle_interrupt(struct host1x *host, unsigned int id); + +void host1x_intr_add_fence_locked(struct host1x *host, struct host1x_syncpt_fence *fence); + +bool host1x_intr_remove_fence(struct host1x *host, struct host1x_syncpt_fence *fence); + #endif diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c index cc807667d8f1..3ed49e1fd933 100644 --- a/drivers/gpu/host1x/job.c +++ b/drivers/gpu/host1x/job.c @@ -1,23 +1,14 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Tegra host1x Job * - * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (c) 2010-2015, NVIDIA Corporation. */ #include <linux/dma-mapping.h> #include <linux/err.h> +#include <linux/host1x.h> +#include <linux/iommu.h> #include <linux/kref.h> #include <linux/module.h> #include <linux/scatterlist.h> @@ -27,27 +18,33 @@ #include "channel.h" #include "dev.h" -#include "host1x_bo.h" #include "job.h" #include "syncpt.h" +#define HOST1X_WAIT_SYNCPT_OFFSET 0x8 + struct host1x_job *host1x_job_alloc(struct host1x_channel *ch, u32 num_cmdbufs, u32 num_relocs, - u32 num_waitchks) + bool skip_firewall) { struct host1x_job *job = NULL; - unsigned int num_unpins = num_cmdbufs + num_relocs; + unsigned int num_unpins = num_relocs; + bool enable_firewall; u64 total; void *mem; + enable_firewall = IS_ENABLED(CONFIG_TEGRA_HOST1X_FIREWALL) && !skip_firewall; + + if (!enable_firewall) + num_unpins += num_cmdbufs; + /* Check that we're not going to overflow */ total = sizeof(struct host1x_job) + - num_relocs * sizeof(struct host1x_reloc) + - num_unpins * sizeof(struct host1x_job_unpin_data) + - num_waitchks * sizeof(struct host1x_waitchk) + - num_cmdbufs * sizeof(struct host1x_job_gather) + - num_unpins * sizeof(dma_addr_t) + - num_unpins * sizeof(u32 *); + (u64)num_relocs * sizeof(struct host1x_reloc) + + (u64)num_unpins * sizeof(struct host1x_job_unpin_data) + + (u64)num_cmdbufs * sizeof(struct host1x_job_cmd) + + (u64)num_unpins * sizeof(dma_addr_t) + + (u64)num_unpins * sizeof(u32 *); if (total > ULONG_MAX) return NULL; @@ -55,19 +52,19 @@ struct host1x_job *host1x_job_alloc(struct host1x_channel *ch, if (!job) return NULL; + job->enable_firewall = enable_firewall; + kref_init(&job->ref); job->channel = ch; /* Redistribute memory to the structs */ mem += sizeof(struct host1x_job); - job->relocarray = num_relocs ? mem : NULL; + job->relocs = num_relocs ? mem : NULL; mem += num_relocs * sizeof(struct host1x_reloc); job->unpins = num_unpins ? mem : NULL; mem += num_unpins * sizeof(struct host1x_job_unpin_data); - job->waitchk = num_waitchks ? mem : NULL; - mem += num_waitchks * sizeof(struct host1x_waitchk); - job->gathers = num_cmdbufs ? mem : NULL; - mem += num_cmdbufs * sizeof(struct host1x_job_gather); + job->cmds = num_cmdbufs ? mem : NULL; + mem += num_cmdbufs * sizeof(struct host1x_job_cmd); job->addr_phys = num_unpins ? mem : NULL; job->reloc_addr_phys = job->addr_phys; @@ -75,17 +72,35 @@ struct host1x_job *host1x_job_alloc(struct host1x_channel *ch, return job; } +EXPORT_SYMBOL(host1x_job_alloc); struct host1x_job *host1x_job_get(struct host1x_job *job) { kref_get(&job->ref); return job; } +EXPORT_SYMBOL(host1x_job_get); static void job_free(struct kref *ref) { struct host1x_job *job = container_of(ref, struct host1x_job, ref); + if (job->release) + job->release(job); + + if (job->fence) { + /* + * remove_callback is atomic w.r.t. fence signaling, so + * after the call returns, we know that the callback is not + * in execution, and the fence can be safely freed. + */ + dma_fence_remove_callback(job->fence, &job->fence_cb); + dma_fence_put(job->fence); + } + + if (job->syncpt) + host1x_syncpt_put(job->syncpt); + kfree(job); } @@ -93,182 +108,228 @@ void host1x_job_put(struct host1x_job *job) { kref_put(&job->ref, job_free); } +EXPORT_SYMBOL(host1x_job_put); void host1x_job_add_gather(struct host1x_job *job, struct host1x_bo *bo, - u32 words, u32 offset) + unsigned int words, unsigned int offset) { - struct host1x_job_gather *cur_gather = &job->gathers[job->num_gathers]; + struct host1x_job_gather *gather = &job->cmds[job->num_cmds].gather; - cur_gather->words = words; - cur_gather->bo = bo; - cur_gather->offset = offset; - job->num_gathers++; -} + gather->words = words; + gather->bo = bo; + gather->offset = offset; -/* - * NULL an already satisfied WAIT_SYNCPT host method, by patching its - * args in the command stream. The method data is changed to reference - * a reserved (never given out or incr) HOST1X_SYNCPT_RESERVED syncpt - * with a matching threshold value of 0, so is guaranteed to be popped - * by the host HW. - */ -static void host1x_syncpt_patch_offset(struct host1x_syncpt *sp, - struct host1x_bo *h, u32 offset) -{ - void *patch_addr = NULL; - - /* patch the wait */ - patch_addr = host1x_bo_kmap(h, offset >> PAGE_SHIFT); - if (patch_addr) { - host1x_syncpt_patch_wait(sp, - patch_addr + (offset & ~PAGE_MASK)); - host1x_bo_kunmap(h, offset >> PAGE_SHIFT, patch_addr); - } else - pr_err("Could not map cmdbuf for wait check\n"); + job->num_cmds++; } +EXPORT_SYMBOL(host1x_job_add_gather); -/* - * Check driver supplied waitchk structs for syncpt thresholds - * that have already been satisfied and NULL the comparison (to - * avoid a wrap condition in the HW). - */ -static int do_waitchks(struct host1x_job *job, struct host1x *host, - struct host1x_bo *patch) +void host1x_job_add_wait(struct host1x_job *job, u32 id, u32 thresh, + bool relative, u32 next_class) { - int i; + struct host1x_job_cmd *cmd = &job->cmds[job->num_cmds]; - /* compare syncpt vs wait threshold */ - for (i = 0; i < job->num_waitchk; i++) { - struct host1x_waitchk *wait = &job->waitchk[i]; - struct host1x_syncpt *sp = - host1x_syncpt_get(host, wait->syncpt_id); + cmd->is_wait = true; + cmd->wait.id = id; + cmd->wait.threshold = thresh; + cmd->wait.next_class = next_class; + cmd->wait.relative = relative; - /* validate syncpt id */ - if (wait->syncpt_id > host1x_syncpt_nb_pts(host)) - continue; + job->num_cmds++; +} +EXPORT_SYMBOL(host1x_job_add_wait); - /* skip all other gathers */ - if (patch != wait->bo) - continue; +static unsigned int pin_job(struct host1x *host, struct host1x_job *job) +{ + unsigned long mask = HOST1X_RELOC_READ | HOST1X_RELOC_WRITE; + struct host1x_client *client = job->client; + struct device *dev = client->dev; + struct host1x_job_gather *g; + unsigned int i; + int err; - trace_host1x_syncpt_wait_check(wait->bo, wait->offset, - wait->syncpt_id, wait->thresh, - host1x_syncpt_read_min(sp)); + job->num_unpins = 0; - if (host1x_syncpt_is_expired(sp, wait->thresh)) { - dev_dbg(host->dev, - "drop WAIT id %d (%s) thresh 0x%x, min 0x%x\n", - wait->syncpt_id, sp->name, wait->thresh, - host1x_syncpt_read_min(sp)); + for (i = 0; i < job->num_relocs; i++) { + struct host1x_reloc *reloc = &job->relocs[i]; + enum dma_data_direction direction; + struct host1x_bo_mapping *map; + struct host1x_bo *bo; - host1x_syncpt_patch_offset(sp, patch, wait->offset); + reloc->target.bo = host1x_bo_get(reloc->target.bo); + if (!reloc->target.bo) { + err = -EINVAL; + goto unpin; } - wait->bo = NULL; - } + bo = reloc->target.bo; - return 0; -} + switch (reloc->flags & mask) { + case HOST1X_RELOC_READ: + direction = DMA_TO_DEVICE; + break; -static unsigned int pin_job(struct host1x_job *job) -{ - unsigned int i; + case HOST1X_RELOC_WRITE: + direction = DMA_FROM_DEVICE; + break; - job->num_unpins = 0; + case HOST1X_RELOC_READ | HOST1X_RELOC_WRITE: + direction = DMA_BIDIRECTIONAL; + break; - for (i = 0; i < job->num_relocs; i++) { - struct host1x_reloc *reloc = &job->relocarray[i]; - struct sg_table *sgt; - dma_addr_t phys_addr; + default: + err = -EINVAL; + goto unpin; + } - reloc->target = host1x_bo_get(reloc->target); - if (!reloc->target) + map = host1x_bo_pin(dev, bo, direction, NULL); + if (IS_ERR(map)) { + err = PTR_ERR(map); goto unpin; + } - phys_addr = host1x_bo_pin(reloc->target, &sgt); - if (!phys_addr) + /* + * host1x clients are generally not able to do scatter-gather themselves, so fail + * if the buffer is discontiguous and we fail to map its SG table to a single + * contiguous chunk of I/O virtual memory. + */ + if (map->chunks > 1) { + err = -EINVAL; goto unpin; + } - job->addr_phys[job->num_unpins] = phys_addr; - job->unpins[job->num_unpins].bo = reloc->target; - job->unpins[job->num_unpins].sgt = sgt; + job->addr_phys[job->num_unpins] = map->phys; + job->unpins[job->num_unpins].map = map; job->num_unpins++; } - for (i = 0; i < job->num_gathers; i++) { - struct host1x_job_gather *g = &job->gathers[i]; - struct sg_table *sgt; - dma_addr_t phys_addr; + /* + * We will copy gathers BO content later, so there is no need to + * hold and pin them. + */ + if (job->enable_firewall) + return 0; + + for (i = 0; i < job->num_cmds; i++) { + struct host1x_bo_mapping *map; + size_t gather_size = 0; + struct scatterlist *sg; + unsigned long shift; + struct iova *alloc; + unsigned int j; + + if (job->cmds[i].is_wait) + continue; + + g = &job->cmds[i].gather; g->bo = host1x_bo_get(g->bo); - if (!g->bo) + if (!g->bo) { + err = -EINVAL; goto unpin; + } - phys_addr = host1x_bo_pin(g->bo, &sgt); - if (!phys_addr) + map = host1x_bo_pin(host->dev, g->bo, DMA_TO_DEVICE, NULL); + if (IS_ERR(map)) { + err = PTR_ERR(map); goto unpin; + } + + if (host->domain) { + for_each_sgtable_sg(map->sgt, sg, j) + gather_size += sg->length; + + gather_size = iova_align(&host->iova, gather_size); + + shift = iova_shift(&host->iova); + alloc = alloc_iova(&host->iova, gather_size >> shift, + host->iova_end >> shift, true); + if (!alloc) { + err = -ENOMEM; + goto put; + } + + err = iommu_map_sgtable(host->domain, iova_dma_addr(&host->iova, alloc), + map->sgt, IOMMU_READ); + if (err == 0) { + __free_iova(&host->iova, alloc); + err = -EINVAL; + goto put; + } - job->addr_phys[job->num_unpins] = phys_addr; - job->unpins[job->num_unpins].bo = g->bo; - job->unpins[job->num_unpins].sgt = sgt; + map->phys = iova_dma_addr(&host->iova, alloc); + map->size = gather_size; + } + + job->addr_phys[job->num_unpins] = map->phys; + job->unpins[job->num_unpins].map = map; job->num_unpins++; + + job->gather_addr_phys[i] = map->phys; } - return job->num_unpins; + return 0; +put: + host1x_bo_put(g->bo); unpin: host1x_job_unpin(job); - return 0; + return err; } -static unsigned int do_relocs(struct host1x_job *job, struct host1x_bo *cmdbuf) +static int do_relocs(struct host1x_job *job, struct host1x_job_gather *g) { - int i = 0; - u32 last_page = ~0; - void *cmdbuf_page_addr = NULL; + void *cmdbuf_addr = NULL; + struct host1x_bo *cmdbuf = g->bo; + unsigned int i; /* pin & patch the relocs for one gather */ for (i = 0; i < job->num_relocs; i++) { - struct host1x_reloc *reloc = &job->relocarray[i]; + struct host1x_reloc *reloc = &job->relocs[i]; u32 reloc_addr = (job->reloc_addr_phys[i] + - reloc->target_offset) >> reloc->shift; + reloc->target.offset) >> reloc->shift; u32 *target; /* skip all other gathers */ - if (cmdbuf != reloc->cmdbuf) + if (cmdbuf != reloc->cmdbuf.bo) continue; - if (last_page != reloc->cmdbuf_offset >> PAGE_SHIFT) { - if (cmdbuf_page_addr) - host1x_bo_kunmap(cmdbuf, last_page, - cmdbuf_page_addr); + if (job->enable_firewall) { + target = (u32 *)job->gather_copy_mapped + + reloc->cmdbuf.offset / sizeof(u32) + + g->offset / sizeof(u32); + goto patch_reloc; + } - cmdbuf_page_addr = host1x_bo_kmap(cmdbuf, - reloc->cmdbuf_offset >> PAGE_SHIFT); - last_page = reloc->cmdbuf_offset >> PAGE_SHIFT; + if (!cmdbuf_addr) { + cmdbuf_addr = host1x_bo_mmap(cmdbuf); - if (unlikely(!cmdbuf_page_addr)) { + if (unlikely(!cmdbuf_addr)) { pr_err("Could not map cmdbuf for relocation\n"); return -ENOMEM; } } - target = cmdbuf_page_addr + (reloc->cmdbuf_offset & ~PAGE_MASK); + target = cmdbuf_addr + reloc->cmdbuf.offset; +patch_reloc: *target = reloc_addr; } - if (cmdbuf_page_addr) - host1x_bo_kunmap(cmdbuf, last_page, cmdbuf_page_addr); + if (cmdbuf_addr) + host1x_bo_munmap(cmdbuf, cmdbuf_addr); return 0; } static bool check_reloc(struct host1x_reloc *reloc, struct host1x_bo *cmdbuf, - unsigned int offset) + unsigned int offset) { offset *= sizeof(u32); - if (reloc->cmdbuf != cmdbuf || reloc->cmdbuf_offset != offset) + if (reloc->cmdbuf.bo != cmdbuf || reloc->cmdbuf.offset != offset) + return false; + + /* relocation shift value validation isn't implemented yet */ + if (reloc->shift) return false; return true; @@ -281,7 +342,7 @@ struct host1x_firewall { unsigned int num_relocs; struct host1x_reloc *reloc; - struct host1x_bo *cmdbuf_id; + struct host1x_bo *cmdbuf; unsigned int offset; u32 words; @@ -291,25 +352,53 @@ struct host1x_firewall { u32 count; }; +static int check_register(struct host1x_firewall *fw, unsigned long offset) +{ + if (!fw->job->is_addr_reg) + return 0; + + if (fw->job->is_addr_reg(fw->dev, fw->class, offset)) { + if (!fw->num_relocs) + return -EINVAL; + + if (!check_reloc(fw->reloc, fw->cmdbuf, fw->offset)) + return -EINVAL; + + fw->num_relocs--; + fw->reloc++; + } + + return 0; +} + +static int check_class(struct host1x_firewall *fw, u32 class) +{ + if (!fw->job->is_valid_class) { + if (fw->class != class) + return -EINVAL; + } else { + if (!fw->job->is_valid_class(fw->class)) + return -EINVAL; + } + + return 0; +} + static int check_mask(struct host1x_firewall *fw) { u32 mask = fw->mask; u32 reg = fw->reg; + int ret; while (mask) { if (fw->words == 0) return -EINVAL; if (mask & 1) { - if (fw->job->is_addr_reg(fw->dev, fw->class, reg)) { - if (!fw->num_relocs) - return -EINVAL; - if (!check_reloc(fw->reloc, fw->cmdbuf_id, - fw->offset)) - return -EINVAL; - fw->reloc++; - fw->num_relocs--; - } + ret = check_register(fw, reg); + if (ret < 0) + return ret; + fw->words--; fw->offset++; } @@ -324,19 +413,16 @@ static int check_incr(struct host1x_firewall *fw) { u32 count = fw->count; u32 reg = fw->reg; + int ret; while (count) { if (fw->words == 0) return -EINVAL; - if (fw->job->is_addr_reg(fw->dev, fw->class, reg)) { - if (!fw->num_relocs) - return -EINVAL; - if (!check_reloc(fw->reloc, fw->cmdbuf_id, fw->offset)) - return -EINVAL; - fw->reloc++; - fw->num_relocs--; - } + ret = check_register(fw, reg); + if (ret < 0) + return ret; + reg++; fw->words--; fw->offset++; @@ -348,21 +434,17 @@ static int check_incr(struct host1x_firewall *fw) static int check_nonincr(struct host1x_firewall *fw) { - int is_addr_reg = fw->job->is_addr_reg(fw->dev, fw->class, fw->reg); u32 count = fw->count; + int ret; while (count) { if (fw->words == 0) return -EINVAL; - if (is_addr_reg) { - if (!fw->num_relocs) - return -EINVAL; - if (!check_reloc(fw->reloc, fw->cmdbuf_id, fw->offset)) - return -EINVAL; - fw->reloc++; - fw->num_relocs--; - } + ret = check_register(fw, fw->reg); + if (ret < 0) + return ret; + fw->words--; fw->offset++; count--; @@ -375,13 +457,11 @@ static int validate(struct host1x_firewall *fw, struct host1x_job_gather *g) { u32 *cmdbuf_base = (u32 *)fw->job->gather_copy_mapped + (g->offset / sizeof(u32)); + u32 job_class = fw->class; int err = 0; - if (!fw->job->is_addr_reg) - return 0; - fw->words = g->words; - fw->cmdbuf_id = g->bo; + fw->cmdbuf = g->bo; fw->offset = 0; while (fw->words && !err) { @@ -399,7 +479,9 @@ static int validate(struct host1x_firewall *fw, struct host1x_job_gather *g) fw->class = word >> 6 & 0x3ff; fw->mask = word & 0x3f; fw->reg = word >> 16 & 0xfff; - err = check_mask(fw); + err = check_class(fw, job_class); + if (!err) + err = check_mask(fw); if (err) goto out; break; @@ -427,7 +509,6 @@ static int validate(struct host1x_firewall *fw, struct host1x_job_gather *g) goto out; break; case 4: - case 5: case 14: break; default: @@ -436,47 +517,60 @@ static int validate(struct host1x_firewall *fw, struct host1x_job_gather *g) } } - /* No relocs should remain at this point */ - if (fw->num_relocs) - err = -EINVAL; - out: return err; } -static inline int copy_gathers(struct host1x_job *job, struct device *dev) +static inline int copy_gathers(struct device *host, struct host1x_job *job, + struct device *dev) { struct host1x_firewall fw; size_t size = 0; size_t offset = 0; - int i; + unsigned int i; fw.job = job; fw.dev = dev; - fw.reloc = job->relocarray; + fw.reloc = job->relocs; fw.num_relocs = job->num_relocs; - fw.class = 0; + fw.class = job->class; + + for (i = 0; i < job->num_cmds; i++) { + struct host1x_job_gather *g; + + if (job->cmds[i].is_wait) + continue; + + g = &job->cmds[i].gather; - for (i = 0; i < job->num_gathers; i++) { - struct host1x_job_gather *g = &job->gathers[i]; size += g->words * sizeof(u32); } - job->gather_copy_mapped = dma_alloc_writecombine(dev, size, - &job->gather_copy, - GFP_KERNEL); - if (!job->gather_copy_mapped) { - int err = PTR_ERR(job->gather_copy_mapped); - job->gather_copy_mapped = NULL; - return err; - } + /* + * Try a non-blocking allocation from a higher priority pools first, + * as awaiting for the allocation here is a major performance hit. + */ + job->gather_copy_mapped = dma_alloc_wc(host, size, &job->gather_copy, + GFP_NOWAIT); + + /* the higher priority allocation failed, try the generic-blocking */ + if (!job->gather_copy_mapped) + job->gather_copy_mapped = dma_alloc_wc(host, size, + &job->gather_copy, + GFP_KERNEL); + if (!job->gather_copy_mapped) + return -ENOMEM; job->gather_copy_size = size; - for (i = 0; i < job->num_gathers; i++) { - struct host1x_job_gather *g = &job->gathers[i]; + for (i = 0; i < job->num_cmds; i++) { + struct host1x_job_gather *g; void *gather; + if (job->cmds[i].is_wait) + continue; + g = &job->cmds[i].gather; + /* Copy the gather */ gather = host1x_bo_mmap(g->bo); memcpy(job->gather_copy_mapped + offset, gather + g->offset, @@ -494,6 +588,10 @@ static inline int copy_gathers(struct host1x_job *job, struct device *dev) offset += g->words * sizeof(u32); } + /* No relocs should remain at this point */ + if (fw.num_relocs) + return -EINVAL; + return 0; } @@ -502,84 +600,88 @@ int host1x_job_pin(struct host1x_job *job, struct device *dev) int err; unsigned int i, j; struct host1x *host = dev_get_drvdata(dev->parent); - DECLARE_BITMAP(waitchk_mask, host1x_syncpt_nb_pts(host)); - - bitmap_zero(waitchk_mask, host1x_syncpt_nb_pts(host)); - for (i = 0; i < job->num_waitchk; i++) { - u32 syncpt_id = job->waitchk[i].syncpt_id; - if (syncpt_id < host1x_syncpt_nb_pts(host)) - set_bit(syncpt_id, waitchk_mask); - } - - /* get current syncpt values for waitchk */ - for_each_set_bit(i, waitchk_mask, host1x_syncpt_nb_pts(host)) - host1x_syncpt_load(host->syncpt + i); /* pin memory */ - err = pin_job(job); - if (!err) + err = pin_job(host, job); + if (err) goto out; + if (job->enable_firewall) { + err = copy_gathers(host->dev, job, dev); + if (err) + goto out; + } + /* patch gathers */ - for (i = 0; i < job->num_gathers; i++) { - struct host1x_job_gather *g = &job->gathers[i]; + for (i = 0; i < job->num_cmds; i++) { + struct host1x_job_gather *g; + + if (job->cmds[i].is_wait) + continue; + g = &job->cmds[i].gather; /* process each gather mem only once */ if (g->handled) continue; - g->base = job->gather_addr_phys[i]; + /* copy_gathers() sets gathers base if firewall is enabled */ + if (!job->enable_firewall) + g->base = job->gather_addr_phys[i]; - for (j = 0; j < job->num_gathers; j++) - if (job->gathers[j].bo == g->bo) - job->gathers[j].handled = true; - - err = do_relocs(job, g->bo); - if (err) - break; + for (j = i + 1; j < job->num_cmds; j++) { + if (!job->cmds[j].is_wait && + job->cmds[j].gather.bo == g->bo) { + job->cmds[j].gather.handled = true; + job->cmds[j].gather.base = g->base; + } + } - err = do_waitchks(job, host, g->bo); + err = do_relocs(job, g); if (err) break; } - if (IS_ENABLED(CONFIG_TEGRA_HOST1X_FIREWALL) && !err) { - err = copy_gathers(job, dev); - if (err) { - host1x_job_unpin(job); - return err; - } - } - out: + if (err) + host1x_job_unpin(job); wmb(); return err; } +EXPORT_SYMBOL(host1x_job_pin); void host1x_job_unpin(struct host1x_job *job) { + struct host1x *host = dev_get_drvdata(job->channel->dev->parent); unsigned int i; for (i = 0; i < job->num_unpins; i++) { - struct host1x_job_unpin_data *unpin = &job->unpins[i]; - host1x_bo_unpin(unpin->bo, unpin->sgt); - host1x_bo_put(unpin->bo); + struct host1x_bo_mapping *map = job->unpins[i].map; + struct host1x_bo *bo = map->bo; + + if (!job->enable_firewall && map->size && host->domain) { + iommu_unmap(host->domain, job->addr_phys[i], map->size); + free_iova(&host->iova, iova_pfn(&host->iova, job->addr_phys[i])); + } + + host1x_bo_unpin(map); + host1x_bo_put(bo); } + job->num_unpins = 0; if (job->gather_copy_size) - dma_free_writecombine(job->channel->dev, job->gather_copy_size, - job->gather_copy_mapped, - job->gather_copy); + dma_free_wc(host->dev, job->gather_copy_size, + job->gather_copy_mapped, job->gather_copy); } +EXPORT_SYMBOL(host1x_job_unpin); /* * Debug routine used to dump job entries */ void host1x_job_dump(struct device *dev, struct host1x_job *job) { - dev_dbg(dev, " SYNCPT_ID %d\n", job->syncpt_id); + dev_dbg(dev, " SYNCPT_ID %d\n", job->syncpt->id); dev_dbg(dev, " SYNCPT_VAL %d\n", job->syncpt_end); dev_dbg(dev, " FIRST_GET 0x%x\n", job->first_get); dev_dbg(dev, " TIMEOUT %d\n", job->timeout); diff --git a/drivers/gpu/host1x/job.h b/drivers/gpu/host1x/job.h index fba45f20458e..dad5a1946693 100644 --- a/drivers/gpu/host1x/job.h +++ b/drivers/gpu/host1x/job.h @@ -1,160 +1,44 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ /* * Tegra host1x Job * * Copyright (c) 2011-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ #ifndef __HOST1X_JOB_H #define __HOST1X_JOB_H +#include <linux/dma-direction.h> + struct host1x_job_gather { - u32 words; + unsigned int words; dma_addr_t base; struct host1x_bo *bo; - int offset; + unsigned int offset; bool handled; }; -struct host1x_cmdbuf { - u32 handle; - u32 offset; - u32 words; - u32 pad; +struct host1x_job_wait { + u32 id; + u32 threshold; + u32 next_class; + bool relative; }; -struct host1x_reloc { - struct host1x_bo *cmdbuf; - u32 cmdbuf_offset; - struct host1x_bo *target; - u32 target_offset; - u32 shift; - u32 pad; -}; +struct host1x_job_cmd { + bool is_wait; -struct host1x_waitchk { - struct host1x_bo *bo; - u32 offset; - u32 syncpt_id; - u32 thresh; + union { + struct host1x_job_gather gather; + struct host1x_job_wait wait; + }; }; struct host1x_job_unpin_data { - struct host1x_bo *bo; - struct sg_table *sgt; + struct host1x_bo_mapping *map; }; /* - * Each submit is tracked as a host1x_job. - */ -struct host1x_job { - /* When refcount goes to zero, job can be freed */ - struct kref ref; - - /* List entry */ - struct list_head list; - - /* Channel where job is submitted to */ - struct host1x_channel *channel; - - u32 client; - - /* Gathers and their memory */ - struct host1x_job_gather *gathers; - unsigned int num_gathers; - - /* Wait checks to be processed at submit time */ - struct host1x_waitchk *waitchk; - unsigned int num_waitchk; - u32 waitchk_mask; - - /* Array of handles to be pinned & unpinned */ - struct host1x_reloc *relocarray; - unsigned int num_relocs; - struct host1x_job_unpin_data *unpins; - unsigned int num_unpins; - - dma_addr_t *addr_phys; - dma_addr_t *gather_addr_phys; - dma_addr_t *reloc_addr_phys; - - /* Sync point id, number of increments and end related to the submit */ - u32 syncpt_id; - u32 syncpt_incrs; - u32 syncpt_end; - - /* Maximum time to wait for this job */ - unsigned int timeout; - - /* Index and number of slots used in the push buffer */ - unsigned int first_get; - unsigned int num_slots; - - /* Copy of gathers */ - size_t gather_copy_size; - dma_addr_t gather_copy; - u8 *gather_copy_mapped; - - /* Check if register is marked as an address reg */ - int (*is_addr_reg)(struct device *dev, u32 reg, u32 class); - - /* Request a SETCLASS to this class */ - u32 class; - - /* Add a channel wait for previous ops to complete */ - bool serialize; -}; -/* - * Allocate memory for a job. Just enough memory will be allocated to - * accomodate the submit. - */ -struct host1x_job *host1x_job_alloc(struct host1x_channel *ch, - u32 num_cmdbufs, u32 num_relocs, - u32 num_waitchks); - -/* - * Add a gather to a job. - */ -void host1x_job_add_gather(struct host1x_job *job, struct host1x_bo *mem_id, - u32 words, u32 offset); - -/* - * Increment reference going to host1x_job. - */ -struct host1x_job *host1x_job_get(struct host1x_job *job); - -/* - * Decrement reference job, free if goes to zero. - */ -void host1x_job_put(struct host1x_job *job); - -/* - * Pin memory related to job. This handles relocation of addresses to the - * host1x address space. Handles both the gather memory and any other memory - * referred to from the gather buffers. - * - * Handles also patching out host waits that would wait for an expired sync - * point value. - */ -int host1x_job_pin(struct host1x_job *job, struct device *dev); - -/* - * Unpin memory related to job. - */ -void host1x_job_unpin(struct host1x_job *job); - -/* * Dump contents of job to debug output. */ void host1x_job_dump(struct device *dev, struct host1x_job *job); diff --git a/drivers/gpu/host1x/mipi.c b/drivers/gpu/host1x/mipi.c new file mode 100644 index 000000000000..e51b43dd15a3 --- /dev/null +++ b/drivers/gpu/host1x/mipi.c @@ -0,0 +1,539 @@ +/* + * Copyright (C) 2013 NVIDIA Corporation + * + * Permission to use, copy, modify, distribute, and sell this software and its + * documentation for any purpose is hereby granted without fee, provided that + * the above copyright notice appear in all copies and that both that copyright + * notice and this permission notice appear in supporting documentation, and + * that the name of the copyright holders not be used in advertising or + * publicity pertaining to distribution of the software without specific, + * written prior permission. The copyright holders make no representations + * about the suitability of this software for any purpose. It is provided "as + * is" without express or implied warranty. + * + * THE COPYRIGHT HOLDERS DISCLAIM ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, + * INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO + * EVENT SHALL THE COPYRIGHT HOLDERS BE LIABLE FOR ANY SPECIAL, INDIRECT OR + * CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, + * DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER + * TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE + * OF THIS SOFTWARE. + */ + +#include <linux/clk.h> +#include <linux/host1x.h> +#include <linux/io.h> +#include <linux/iopoll.h> +#include <linux/of_platform.h> +#include <linux/platform_device.h> +#include <linux/slab.h> + +#include "dev.h" + +#define MIPI_CAL_CTRL 0x00 +#define MIPI_CAL_CTRL_NOISE_FILTER(x) (((x) & 0xf) << 26) +#define MIPI_CAL_CTRL_PRESCALE(x) (((x) & 0x3) << 24) +#define MIPI_CAL_CTRL_CLKEN_OVR (1 << 4) +#define MIPI_CAL_CTRL_START (1 << 0) + +#define MIPI_CAL_AUTOCAL_CTRL 0x01 + +#define MIPI_CAL_STATUS 0x02 +#define MIPI_CAL_STATUS_DONE (1 << 16) +#define MIPI_CAL_STATUS_ACTIVE (1 << 0) + +#define MIPI_CAL_CONFIG_CSIA 0x05 +#define MIPI_CAL_CONFIG_CSIB 0x06 +#define MIPI_CAL_CONFIG_CSIC 0x07 +#define MIPI_CAL_CONFIG_CSID 0x08 +#define MIPI_CAL_CONFIG_CSIE 0x09 +#define MIPI_CAL_CONFIG_CSIF 0x0a +#define MIPI_CAL_CONFIG_DSIA 0x0e +#define MIPI_CAL_CONFIG_DSIB 0x0f +#define MIPI_CAL_CONFIG_DSIC 0x10 +#define MIPI_CAL_CONFIG_DSID 0x11 + +#define MIPI_CAL_CONFIG_DSIA_CLK 0x19 +#define MIPI_CAL_CONFIG_DSIB_CLK 0x1a +#define MIPI_CAL_CONFIG_CSIAB_CLK 0x1b +#define MIPI_CAL_CONFIG_DSIC_CLK 0x1c +#define MIPI_CAL_CONFIG_CSICD_CLK 0x1c +#define MIPI_CAL_CONFIG_DSID_CLK 0x1d +#define MIPI_CAL_CONFIG_CSIE_CLK 0x1d + +/* for data and clock lanes */ +#define MIPI_CAL_CONFIG_SELECT (1 << 21) + +/* for data lanes */ +#define MIPI_CAL_CONFIG_HSPDOS(x) (((x) & 0x1f) << 16) +#define MIPI_CAL_CONFIG_HSPUOS(x) (((x) & 0x1f) << 8) +#define MIPI_CAL_CONFIG_TERMOS(x) (((x) & 0x1f) << 0) + +/* for clock lanes */ +#define MIPI_CAL_CONFIG_HSCLKPDOSD(x) (((x) & 0x1f) << 8) +#define MIPI_CAL_CONFIG_HSCLKPUOSD(x) (((x) & 0x1f) << 0) + +#define MIPI_CAL_BIAS_PAD_CFG0 0x16 +#define MIPI_CAL_BIAS_PAD_PDVCLAMP (1 << 1) +#define MIPI_CAL_BIAS_PAD_E_VCLAMP_REF (1 << 0) + +#define MIPI_CAL_BIAS_PAD_CFG1 0x17 +#define MIPI_CAL_BIAS_PAD_DRV_DN_REF(x) (((x) & 0x7) << 16) +#define MIPI_CAL_BIAS_PAD_DRV_UP_REF(x) (((x) & 0x7) << 8) + +#define MIPI_CAL_BIAS_PAD_CFG2 0x18 +#define MIPI_CAL_BIAS_PAD_VCLAMP(x) (((x) & 0x7) << 16) +#define MIPI_CAL_BIAS_PAD_VAUXP(x) (((x) & 0x7) << 4) +#define MIPI_CAL_BIAS_PAD_PDVREG (1 << 1) + +struct tegra_mipi_pad { + unsigned long data; + unsigned long clk; +}; + +struct tegra_mipi_soc { + bool has_clk_lane; + const struct tegra_mipi_pad *pads; + unsigned int num_pads; + + bool clock_enable_override; + bool needs_vclamp_ref; + + /* bias pad configuration settings */ + u8 pad_drive_down_ref; + u8 pad_drive_up_ref; + + u8 pad_vclamp_level; + u8 pad_vauxp_level; + + /* calibration settings for data lanes */ + u8 hspdos; + u8 hspuos; + u8 termos; + + /* calibration settings for clock lanes */ + u8 hsclkpdos; + u8 hsclkpuos; +}; + +struct tegra_mipi { + const struct tegra_mipi_soc *soc; + struct device *dev; + void __iomem *regs; + struct mutex lock; + struct clk *clk; + + unsigned long usage_count; +}; + +struct tegra_mipi_device { + struct platform_device *pdev; + struct tegra_mipi *mipi; + struct device *device; + unsigned long pads; +}; + +static inline u32 tegra_mipi_readl(struct tegra_mipi *mipi, + unsigned long offset) +{ + return readl(mipi->regs + (offset << 2)); +} + +static inline void tegra_mipi_writel(struct tegra_mipi *mipi, u32 value, + unsigned long offset) +{ + writel(value, mipi->regs + (offset << 2)); +} + +static int tegra_mipi_power_up(struct tegra_mipi *mipi) +{ + u32 value; + int err; + + err = clk_enable(mipi->clk); + if (err < 0) + return err; + + value = tegra_mipi_readl(mipi, MIPI_CAL_BIAS_PAD_CFG0); + value &= ~MIPI_CAL_BIAS_PAD_PDVCLAMP; + + if (mipi->soc->needs_vclamp_ref) + value |= MIPI_CAL_BIAS_PAD_E_VCLAMP_REF; + + tegra_mipi_writel(mipi, value, MIPI_CAL_BIAS_PAD_CFG0); + + value = tegra_mipi_readl(mipi, MIPI_CAL_BIAS_PAD_CFG2); + value &= ~MIPI_CAL_BIAS_PAD_PDVREG; + tegra_mipi_writel(mipi, value, MIPI_CAL_BIAS_PAD_CFG2); + + clk_disable(mipi->clk); + + return 0; +} + +static int tegra_mipi_power_down(struct tegra_mipi *mipi) +{ + u32 value; + int err; + + err = clk_enable(mipi->clk); + if (err < 0) + return err; + + /* + * The MIPI_CAL_BIAS_PAD_PDVREG controls a voltage regulator that + * supplies the DSI pads. This must be kept enabled until none of the + * DSI lanes are used anymore. + */ + value = tegra_mipi_readl(mipi, MIPI_CAL_BIAS_PAD_CFG2); + value |= MIPI_CAL_BIAS_PAD_PDVREG; + tegra_mipi_writel(mipi, value, MIPI_CAL_BIAS_PAD_CFG2); + + /* + * MIPI_CAL_BIAS_PAD_PDVCLAMP and MIPI_CAL_BIAS_PAD_E_VCLAMP_REF + * control a regulator that supplies current to the pre-driver logic. + * Powering down this regulator causes DSI to fail, so it must remain + * powered on until none of the DSI lanes are used anymore. + */ + value = tegra_mipi_readl(mipi, MIPI_CAL_BIAS_PAD_CFG0); + + if (mipi->soc->needs_vclamp_ref) + value &= ~MIPI_CAL_BIAS_PAD_E_VCLAMP_REF; + + value |= MIPI_CAL_BIAS_PAD_PDVCLAMP; + tegra_mipi_writel(mipi, value, MIPI_CAL_BIAS_PAD_CFG0); + + return 0; +} + +struct tegra_mipi_device *tegra_mipi_request(struct device *device, + struct device_node *np) +{ + struct tegra_mipi_device *dev; + struct of_phandle_args args; + int err; + + err = of_parse_phandle_with_args(np, "nvidia,mipi-calibrate", + "#nvidia,mipi-calibrate-cells", 0, + &args); + if (err < 0) + return ERR_PTR(err); + + dev = kzalloc(sizeof(*dev), GFP_KERNEL); + if (!dev) { + err = -ENOMEM; + goto out; + } + + dev->pdev = of_find_device_by_node(args.np); + if (!dev->pdev) { + err = -ENODEV; + goto free; + } + + dev->mipi = platform_get_drvdata(dev->pdev); + if (!dev->mipi) { + err = -EPROBE_DEFER; + goto put; + } + + of_node_put(args.np); + + dev->pads = args.args[0]; + dev->device = device; + + return dev; + +put: + platform_device_put(dev->pdev); +free: + kfree(dev); +out: + of_node_put(args.np); + return ERR_PTR(err); +} +EXPORT_SYMBOL(tegra_mipi_request); + +void tegra_mipi_free(struct tegra_mipi_device *device) +{ + platform_device_put(device->pdev); + kfree(device); +} +EXPORT_SYMBOL(tegra_mipi_free); + +int tegra_mipi_enable(struct tegra_mipi_device *dev) +{ + int err = 0; + + mutex_lock(&dev->mipi->lock); + + if (dev->mipi->usage_count++ == 0) + err = tegra_mipi_power_up(dev->mipi); + + mutex_unlock(&dev->mipi->lock); + + return err; + +} +EXPORT_SYMBOL(tegra_mipi_enable); + +int tegra_mipi_disable(struct tegra_mipi_device *dev) +{ + int err = 0; + + mutex_lock(&dev->mipi->lock); + + if (--dev->mipi->usage_count == 0) + err = tegra_mipi_power_down(dev->mipi); + + mutex_unlock(&dev->mipi->lock); + + return err; + +} +EXPORT_SYMBOL(tegra_mipi_disable); + +int tegra_mipi_finish_calibration(struct tegra_mipi_device *device) +{ + struct tegra_mipi *mipi = device->mipi; + void __iomem *status_reg = mipi->regs + (MIPI_CAL_STATUS << 2); + u32 value; + int err; + + err = readl_relaxed_poll_timeout(status_reg, value, + !(value & MIPI_CAL_STATUS_ACTIVE) && + (value & MIPI_CAL_STATUS_DONE), 50, + 250000); + mutex_unlock(&device->mipi->lock); + clk_disable(device->mipi->clk); + + return err; +} +EXPORT_SYMBOL(tegra_mipi_finish_calibration); + +int tegra_mipi_start_calibration(struct tegra_mipi_device *device) +{ + const struct tegra_mipi_soc *soc = device->mipi->soc; + unsigned int i; + u32 value; + int err; + + err = clk_enable(device->mipi->clk); + if (err < 0) + return err; + + mutex_lock(&device->mipi->lock); + + value = MIPI_CAL_BIAS_PAD_DRV_DN_REF(soc->pad_drive_down_ref) | + MIPI_CAL_BIAS_PAD_DRV_UP_REF(soc->pad_drive_up_ref); + tegra_mipi_writel(device->mipi, value, MIPI_CAL_BIAS_PAD_CFG1); + + value = tegra_mipi_readl(device->mipi, MIPI_CAL_BIAS_PAD_CFG2); + value &= ~MIPI_CAL_BIAS_PAD_VCLAMP(0x7); + value &= ~MIPI_CAL_BIAS_PAD_VAUXP(0x7); + value |= MIPI_CAL_BIAS_PAD_VCLAMP(soc->pad_vclamp_level); + value |= MIPI_CAL_BIAS_PAD_VAUXP(soc->pad_vauxp_level); + tegra_mipi_writel(device->mipi, value, MIPI_CAL_BIAS_PAD_CFG2); + + for (i = 0; i < soc->num_pads; i++) { + u32 clk = 0, data = 0; + + if (device->pads & BIT(i)) { + data = MIPI_CAL_CONFIG_SELECT | + MIPI_CAL_CONFIG_HSPDOS(soc->hspdos) | + MIPI_CAL_CONFIG_HSPUOS(soc->hspuos) | + MIPI_CAL_CONFIG_TERMOS(soc->termos); + clk = MIPI_CAL_CONFIG_SELECT | + MIPI_CAL_CONFIG_HSCLKPDOSD(soc->hsclkpdos) | + MIPI_CAL_CONFIG_HSCLKPUOSD(soc->hsclkpuos); + } + + tegra_mipi_writel(device->mipi, data, soc->pads[i].data); + + if (soc->has_clk_lane && soc->pads[i].clk != 0) + tegra_mipi_writel(device->mipi, clk, soc->pads[i].clk); + } + + value = tegra_mipi_readl(device->mipi, MIPI_CAL_CTRL); + value &= ~MIPI_CAL_CTRL_NOISE_FILTER(0xf); + value &= ~MIPI_CAL_CTRL_PRESCALE(0x3); + value |= MIPI_CAL_CTRL_NOISE_FILTER(0xa); + value |= MIPI_CAL_CTRL_PRESCALE(0x2); + + if (!soc->clock_enable_override) + value &= ~MIPI_CAL_CTRL_CLKEN_OVR; + else + value |= MIPI_CAL_CTRL_CLKEN_OVR; + + tegra_mipi_writel(device->mipi, value, MIPI_CAL_CTRL); + + /* clear any pending status bits */ + value = tegra_mipi_readl(device->mipi, MIPI_CAL_STATUS); + tegra_mipi_writel(device->mipi, value, MIPI_CAL_STATUS); + + value = tegra_mipi_readl(device->mipi, MIPI_CAL_CTRL); + value |= MIPI_CAL_CTRL_START; + tegra_mipi_writel(device->mipi, value, MIPI_CAL_CTRL); + + /* + * Wait for min 72uS to let calibration logic finish calibration + * sequence codes before waiting for pads idle state to apply the + * results. + */ + usleep_range(75, 80); + + return 0; +} +EXPORT_SYMBOL(tegra_mipi_start_calibration); + +static const struct tegra_mipi_pad tegra114_mipi_pads[] = { + { .data = MIPI_CAL_CONFIG_CSIA }, + { .data = MIPI_CAL_CONFIG_CSIB }, + { .data = MIPI_CAL_CONFIG_CSIC }, + { .data = MIPI_CAL_CONFIG_CSID }, + { .data = MIPI_CAL_CONFIG_CSIE }, + { .data = MIPI_CAL_CONFIG_DSIA }, + { .data = MIPI_CAL_CONFIG_DSIB }, + { .data = MIPI_CAL_CONFIG_DSIC }, + { .data = MIPI_CAL_CONFIG_DSID }, +}; + +static const struct tegra_mipi_soc tegra114_mipi_soc = { + .has_clk_lane = false, + .pads = tegra114_mipi_pads, + .num_pads = ARRAY_SIZE(tegra114_mipi_pads), + .clock_enable_override = true, + .needs_vclamp_ref = true, + .pad_drive_down_ref = 0x2, + .pad_drive_up_ref = 0x0, + .pad_vclamp_level = 0x0, + .pad_vauxp_level = 0x0, + .hspdos = 0x0, + .hspuos = 0x4, + .termos = 0x5, + .hsclkpdos = 0x0, + .hsclkpuos = 0x4, +}; + +static const struct tegra_mipi_pad tegra124_mipi_pads[] = { + { .data = MIPI_CAL_CONFIG_CSIA, .clk = MIPI_CAL_CONFIG_CSIAB_CLK }, + { .data = MIPI_CAL_CONFIG_CSIB, .clk = MIPI_CAL_CONFIG_CSIAB_CLK }, + { .data = MIPI_CAL_CONFIG_CSIC, .clk = MIPI_CAL_CONFIG_CSICD_CLK }, + { .data = MIPI_CAL_CONFIG_CSID, .clk = MIPI_CAL_CONFIG_CSICD_CLK }, + { .data = MIPI_CAL_CONFIG_CSIE, .clk = MIPI_CAL_CONFIG_CSIE_CLK }, + { .data = MIPI_CAL_CONFIG_DSIA, .clk = MIPI_CAL_CONFIG_DSIA_CLK }, + { .data = MIPI_CAL_CONFIG_DSIB, .clk = MIPI_CAL_CONFIG_DSIB_CLK }, +}; + +static const struct tegra_mipi_soc tegra124_mipi_soc = { + .has_clk_lane = true, + .pads = tegra124_mipi_pads, + .num_pads = ARRAY_SIZE(tegra124_mipi_pads), + .clock_enable_override = true, + .needs_vclamp_ref = true, + .pad_drive_down_ref = 0x2, + .pad_drive_up_ref = 0x0, + .pad_vclamp_level = 0x0, + .pad_vauxp_level = 0x0, + .hspdos = 0x0, + .hspuos = 0x0, + .termos = 0x0, + .hsclkpdos = 0x1, + .hsclkpuos = 0x2, +}; + +static const struct tegra_mipi_soc tegra132_mipi_soc = { + .has_clk_lane = true, + .pads = tegra124_mipi_pads, + .num_pads = ARRAY_SIZE(tegra124_mipi_pads), + .clock_enable_override = false, + .needs_vclamp_ref = false, + .pad_drive_down_ref = 0x0, + .pad_drive_up_ref = 0x3, + .pad_vclamp_level = 0x0, + .pad_vauxp_level = 0x0, + .hspdos = 0x0, + .hspuos = 0x0, + .termos = 0x0, + .hsclkpdos = 0x3, + .hsclkpuos = 0x2, +}; + +static const struct tegra_mipi_pad tegra210_mipi_pads[] = { + { .data = MIPI_CAL_CONFIG_CSIA, .clk = 0 }, + { .data = MIPI_CAL_CONFIG_CSIB, .clk = 0 }, + { .data = MIPI_CAL_CONFIG_CSIC, .clk = 0 }, + { .data = MIPI_CAL_CONFIG_CSID, .clk = 0 }, + { .data = MIPI_CAL_CONFIG_CSIE, .clk = 0 }, + { .data = MIPI_CAL_CONFIG_CSIF, .clk = 0 }, + { .data = MIPI_CAL_CONFIG_DSIA, .clk = MIPI_CAL_CONFIG_DSIA_CLK }, + { .data = MIPI_CAL_CONFIG_DSIB, .clk = MIPI_CAL_CONFIG_DSIB_CLK }, + { .data = MIPI_CAL_CONFIG_DSIC, .clk = MIPI_CAL_CONFIG_DSIC_CLK }, + { .data = MIPI_CAL_CONFIG_DSID, .clk = MIPI_CAL_CONFIG_DSID_CLK }, +}; + +static const struct tegra_mipi_soc tegra210_mipi_soc = { + .has_clk_lane = true, + .pads = tegra210_mipi_pads, + .num_pads = ARRAY_SIZE(tegra210_mipi_pads), + .clock_enable_override = true, + .needs_vclamp_ref = false, + .pad_drive_down_ref = 0x0, + .pad_drive_up_ref = 0x3, + .pad_vclamp_level = 0x1, + .pad_vauxp_level = 0x1, + .hspdos = 0x0, + .hspuos = 0x2, + .termos = 0x0, + .hsclkpdos = 0x0, + .hsclkpuos = 0x2, +}; + +static const struct of_device_id tegra_mipi_of_match[] = { + { .compatible = "nvidia,tegra114-mipi", .data = &tegra114_mipi_soc }, + { .compatible = "nvidia,tegra124-mipi", .data = &tegra124_mipi_soc }, + { .compatible = "nvidia,tegra132-mipi", .data = &tegra132_mipi_soc }, + { .compatible = "nvidia,tegra210-mipi", .data = &tegra210_mipi_soc }, + { }, +}; + +static int tegra_mipi_probe(struct platform_device *pdev) +{ + const struct of_device_id *match; + struct tegra_mipi *mipi; + + match = of_match_node(tegra_mipi_of_match, pdev->dev.of_node); + if (!match) + return -ENODEV; + + mipi = devm_kzalloc(&pdev->dev, sizeof(*mipi), GFP_KERNEL); + if (!mipi) + return -ENOMEM; + + mipi->soc = match->data; + mipi->dev = &pdev->dev; + + mipi->regs = devm_platform_get_and_ioremap_resource(pdev, 0, NULL); + if (IS_ERR(mipi->regs)) + return PTR_ERR(mipi->regs); + + mutex_init(&mipi->lock); + + mipi->clk = devm_clk_get_prepared(&pdev->dev, NULL); + if (IS_ERR(mipi->clk)) { + dev_err(&pdev->dev, "failed to get clock\n"); + return PTR_ERR(mipi->clk); + } + + platform_set_drvdata(pdev, mipi); + + return 0; +} + +struct platform_driver tegra_mipi_driver = { + .driver = { + .name = "tegra-mipi", + .of_match_table = tegra_mipi_of_match, + }, + .probe = tegra_mipi_probe, +}; diff --git a/drivers/gpu/host1x/syncpt.c b/drivers/gpu/host1x/syncpt.c index 409745b949db..acc7d82e0585 100644 --- a/drivers/gpu/host1x/syncpt.c +++ b/drivers/gpu/host1x/syncpt.c @@ -1,23 +1,13 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * Tegra host1x Syncpoints * - * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. + * Copyright (c) 2010-2015, NVIDIA Corporation. */ #include <linux/module.h> #include <linux/device.h> +#include <linux/dma-fence.h> #include <linux/slab.h> #include <trace/events/host1x.h> @@ -30,44 +20,115 @@ #define SYNCPT_CHECK_PERIOD (2 * HZ) #define MAX_STUCK_CHECK_COUNT 15 -static struct host1x_syncpt *_host1x_syncpt_alloc(struct host1x *host, - struct device *dev, - bool client_managed) +static struct host1x_syncpt_base * +host1x_syncpt_base_request(struct host1x *host) { - int i; - struct host1x_syncpt *sp = host->syncpt; - char *name; + struct host1x_syncpt_base *bases = host->bases; + unsigned int i; - for (i = 0; i < host->info->nb_pts && sp->name; i++, sp++) - ; + for (i = 0; i < host->info->nb_bases; i++) + if (!bases[i].requested) + break; - if (i >= host->info->nb_pts) + if (i >= host->info->nb_bases) return NULL; - name = kasprintf(GFP_KERNEL, "%02d-%s", sp->id, - dev ? dev_name(dev) : NULL); + bases[i].requested = true; + return &bases[i]; +} + +static void host1x_syncpt_base_free(struct host1x_syncpt_base *base) +{ + if (base) + base->requested = false; +} + +/** + * host1x_syncpt_alloc() - allocate a syncpoint + * @host: host1x device data + * @flags: bitfield of HOST1X_SYNCPT_* flags + * @name: name for the syncpoint for use in debug prints + * + * Allocates a hardware syncpoint for the caller's use. The caller then has + * the sole authority to mutate the syncpoint's value until it is freed again. + * + * If no free syncpoints are available, or a NULL name was specified, returns + * NULL. + */ +struct host1x_syncpt *host1x_syncpt_alloc(struct host1x *host, + unsigned long flags, + const char *name) +{ + struct host1x_syncpt *sp = host->syncpt; + char *full_name; + unsigned int i; + if (!name) return NULL; - sp->dev = dev; - sp->name = name; - sp->client_managed = client_managed; + mutex_lock(&host->syncpt_mutex); + + for (i = 0; i < host->info->nb_pts && kref_read(&sp->ref); i++, sp++) + ; + if (i >= host->info->nb_pts) + goto unlock; + + if (flags & HOST1X_SYNCPT_HAS_BASE) { + sp->base = host1x_syncpt_base_request(host); + if (!sp->base) + goto unlock; + } + + full_name = kasprintf(GFP_KERNEL, "%u-%s", sp->id, name); + if (!full_name) + goto free_base; + + sp->name = full_name; + + if (flags & HOST1X_SYNCPT_CLIENT_MANAGED) + sp->client_managed = true; + else + sp->client_managed = false; + + kref_init(&sp->ref); + + mutex_unlock(&host->syncpt_mutex); return sp; + +free_base: + host1x_syncpt_base_free(sp->base); + sp->base = NULL; +unlock: + mutex_unlock(&host->syncpt_mutex); + return NULL; } +EXPORT_SYMBOL(host1x_syncpt_alloc); +/** + * host1x_syncpt_id() - retrieve syncpoint ID + * @sp: host1x syncpoint + * + * Given a pointer to a struct host1x_syncpt, retrieves its ID. This ID is + * often used as a value to program into registers that control how hardware + * blocks interact with syncpoints. + */ u32 host1x_syncpt_id(struct host1x_syncpt *sp) { return sp->id; } +EXPORT_SYMBOL(host1x_syncpt_id); -/* - * Updates the value sent to hardware. +/** + * host1x_syncpt_incr_max() - update the value sent to hardware + * @sp: host1x syncpoint + * @incrs: number of increments */ u32 host1x_syncpt_incr_max(struct host1x_syncpt *sp, u32 incrs) { return (u32)atomic_add_return(incrs, &sp->max_val); } +EXPORT_SYMBOL(host1x_syncpt_incr_max); /* * Write cached syncpoint and waitbase values to hardware. @@ -75,12 +136,23 @@ u32 host1x_syncpt_incr_max(struct host1x_syncpt *sp, u32 incrs) void host1x_syncpt_restore(struct host1x *host) { struct host1x_syncpt *sp_base = host->syncpt; - u32 i; + unsigned int i; - for (i = 0; i < host1x_syncpt_nb_pts(host); i++) + for (i = 0; i < host1x_syncpt_nb_pts(host); i++) { + /* + * Unassign syncpt from channels for purposes of Tegra186 + * syncpoint protection. This prevents any channel from + * accessing it until it is reassigned. + */ + host1x_hw_syncpt_assign_to_channel(host, sp_base + i, NULL); host1x_hw_syncpt_restore(host, sp_base + i); + } + for (i = 0; i < host1x_syncpt_nb_bases(host); i++) host1x_hw_syncpt_restore_wait_base(host, sp_base + i); + + host1x_hw_syncpt_enable_protection(host); + wmb(); } @@ -91,7 +163,7 @@ void host1x_syncpt_restore(struct host1x *host) void host1x_syncpt_save(struct host1x *host) { struct host1x_syncpt *sp_base = host->syncpt; - u32 i; + unsigned int i; for (i = 0; i < host1x_syncpt_nb_pts(host); i++) { if (host1x_syncpt_client_managed(sp_base + i)) @@ -111,6 +183,7 @@ void host1x_syncpt_save(struct host1x *host) u32 host1x_syncpt_load(struct host1x_syncpt *sp) { u32 val; + val = host1x_hw_syncpt_load(sp->host, sp); trace_host1x_syncpt_load_min(sp->id, val); @@ -122,117 +195,71 @@ u32 host1x_syncpt_load(struct host1x_syncpt *sp) */ u32 host1x_syncpt_load_wait_base(struct host1x_syncpt *sp) { - u32 val; host1x_hw_syncpt_load_wait_base(sp->host, sp); - val = sp->base_val; - return val; + + return sp->base_val; } -/* - * Increment syncpoint value from cpu, updating cache +/** + * host1x_syncpt_incr() - increment syncpoint value from CPU, updating cache + * @sp: host1x syncpoint */ int host1x_syncpt_incr(struct host1x_syncpt *sp) { return host1x_hw_syncpt_cpu_incr(sp->host, sp); } - -/* - * Updated sync point form hardware, and returns true if syncpoint is expired, - * false if we may need to wait - */ -static bool syncpt_load_min_is_expired(struct host1x_syncpt *sp, u32 thresh) -{ - host1x_hw_syncpt_load(sp->host, sp); - return host1x_syncpt_is_expired(sp, thresh); -} - -/* - * Main entrypoint for syncpoint value waits. +EXPORT_SYMBOL(host1x_syncpt_incr); + +/** + * host1x_syncpt_wait() - wait for a syncpoint to reach a given value + * @sp: host1x syncpoint + * @thresh: threshold + * @timeout: maximum time to wait for the syncpoint to reach the given value + * @value: return location for the syncpoint value */ int host1x_syncpt_wait(struct host1x_syncpt *sp, u32 thresh, long timeout, - u32 *value) + u32 *value) { - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); - void *ref; - struct host1x_waitlist *waiter; - int err = 0, check_count = 0; - u32 val; + struct dma_fence *fence; + long wait_err; + + host1x_hw_syncpt_load(sp->host, sp); if (value) - *value = 0; + *value = host1x_syncpt_load(sp); - /* first check cache */ - if (host1x_syncpt_is_expired(sp, thresh)) { - if (value) - *value = host1x_syncpt_load(sp); + if (host1x_syncpt_is_expired(sp, thresh)) return 0; - } - /* try to read from register */ - val = host1x_hw_syncpt_load(sp->host, sp); - if (host1x_syncpt_is_expired(sp, thresh)) { - if (value) - *value = val; - goto done; - } - - if (!timeout) { - err = -EAGAIN; - goto done; - } - - /* allocate a waiter */ - waiter = kzalloc(sizeof(*waiter), GFP_KERNEL); - if (!waiter) { - err = -ENOMEM; - goto done; - } - - /* schedule a wakeup when the syncpoint value is reached */ - err = host1x_intr_add_action(sp->host, sp->id, thresh, - HOST1X_INTR_ACTION_WAKEUP_INTERRUPTIBLE, - &wq, waiter, &ref); - if (err) - goto done; - - err = -EAGAIN; - /* Caller-specified timeout may be impractically low */ if (timeout < 0) timeout = LONG_MAX; + else if (timeout == 0) + return -EAGAIN; - /* wait for the syncpoint, or timeout, or signal */ - while (timeout) { - long check = min_t(long, SYNCPT_CHECK_PERIOD, timeout); - int remain = wait_event_interruptible_timeout(wq, - syncpt_load_min_is_expired(sp, thresh), - check); - if (remain > 0 || host1x_syncpt_is_expired(sp, thresh)) { - if (value) - *value = host1x_syncpt_load(sp); - err = 0; - break; - } - if (remain < 0) { - err = remain; - break; - } - timeout -= check; - if (timeout && check_count <= MAX_STUCK_CHECK_COUNT) { - dev_warn(sp->host->dev, - "%s: syncpoint id %d (%s) stuck waiting %d, timeout=%ld\n", - current->comm, sp->id, sp->name, - thresh, timeout); - - host1x_debug_dump_syncpts(sp->host); - if (check_count == MAX_STUCK_CHECK_COUNT) - host1x_debug_dump(sp->host); - check_count++; - } - } - host1x_intr_put_ref(sp->host, sp->id, ref); + fence = host1x_fence_create(sp, thresh, false); + if (IS_ERR(fence)) + return PTR_ERR(fence); + + wait_err = dma_fence_wait_timeout(fence, true, timeout); + if (wait_err == 0) + host1x_fence_cancel(fence); + dma_fence_put(fence); + + if (value) + *value = host1x_syncpt_load(sp); -done: - return err; + /* + * Don't rely on dma_fence_wait_timeout return value, + * since it returns zero both on timeout and if the + * wait completed with 0 jiffies left. + */ + host1x_hw_syncpt_load(sp->host, sp); + if (wait_err == 0 && !host1x_syncpt_is_expired(sp, thresh)) + return -EAGAIN; + else if (wait_err < 0) + return wait_err; + else + return 0; } EXPORT_SYMBOL(host1x_syncpt_wait); @@ -242,136 +269,262 @@ EXPORT_SYMBOL(host1x_syncpt_wait); bool host1x_syncpt_is_expired(struct host1x_syncpt *sp, u32 thresh) { u32 current_val; - u32 future_val; + smp_rmb(); + current_val = (u32)atomic_read(&sp->min_val); - future_val = (u32)atomic_read(&sp->max_val); - - /* Note the use of unsigned arithmetic here (mod 1<<32). - * - * c = current_val = min_val = the current value of the syncpoint. - * t = thresh = the value we are checking - * f = future_val = max_val = the value c will reach when all - * outstanding increments have completed. - * - * Note that c always chases f until it reaches f. - * - * Dtf = (f - t) - * Dtc = (c - t) - * - * Consider all cases: - * - * A) .....c..t..f..... Dtf < Dtc need to wait - * B) .....c.....f..t.. Dtf > Dtc expired - * C) ..t..c.....f..... Dtf > Dtc expired (Dct very large) - * - * Any case where f==c: always expired (for any t). Dtf == Dcf - * Any case where t==c: always expired (for any f). Dtf >= Dtc (because Dtc==0) - * Any case where t==f!=c: always wait. Dtf < Dtc (because Dtf==0, - * Dtc!=0) - * - * Other cases: - * - * A) .....t..f..c..... Dtf < Dtc need to wait - * A) .....f..c..t..... Dtf < Dtc need to wait - * A) .....f..t..c..... Dtf > Dtc expired - * - * So: - * Dtf >= Dtc implies EXPIRED (return true) - * Dtf < Dtc implies WAIT (return false) - * - * Note: If t is expired then we *cannot* wait on it. We would wait - * forever (hang the system). - * - * Note: do NOT get clever and remove the -thresh from both sides. It - * is NOT the same. - * - * If future valueis zero, we have a client managed sync point. In that - * case we do a direct comparison. - */ - if (!host1x_syncpt_client_managed(sp)) - return future_val - thresh >= current_val - thresh; - else - return (s32)(current_val - thresh) >= 0; -} -/* remove a wait pointed to by patch_addr */ -int host1x_syncpt_patch_wait(struct host1x_syncpt *sp, void *patch_addr) -{ - return host1x_hw_syncpt_patch_wait(sp->host, sp, patch_addr); + return ((current_val - thresh) & 0x80000000U) == 0U; } int host1x_syncpt_init(struct host1x *host) { + struct host1x_syncpt_base *bases; struct host1x_syncpt *syncpt; - int i; + unsigned int i; - syncpt = devm_kzalloc(host->dev, sizeof(*syncpt) * host->info->nb_pts, - GFP_KERNEL); + syncpt = devm_kcalloc(host->dev, host->info->nb_pts, sizeof(*syncpt), + GFP_KERNEL); if (!syncpt) return -ENOMEM; - for (i = 0; i < host->info->nb_pts; ++i) { + bases = devm_kcalloc(host->dev, host->info->nb_bases, sizeof(*bases), + GFP_KERNEL); + if (!bases) + return -ENOMEM; + + for (i = 0; i < host->info->nb_pts; i++) { syncpt[i].id = i; syncpt[i].host = host; } - host->syncpt = syncpt; + for (i = 0; i < host->info->nb_bases; i++) + bases[i].id = i; - host1x_syncpt_restore(host); + mutex_init(&host->syncpt_mutex); + host->syncpt = syncpt; + host->bases = bases; /* Allocate sync point to use for clearing waits for expired fences */ - host->nop_sp = _host1x_syncpt_alloc(host, NULL, false); + host->nop_sp = host1x_syncpt_alloc(host, 0, "reserved-nop"); if (!host->nop_sp) return -ENOMEM; + if (host->info->reserve_vblank_syncpts) { + kref_init(&host->syncpt[26].ref); + kref_init(&host->syncpt[27].ref); + } + return 0; } -struct host1x_syncpt *host1x_syncpt_request(struct device *dev, - bool client_managed) +/** + * host1x_syncpt_request() - request a syncpoint + * @client: client requesting the syncpoint + * @flags: flags + * + * host1x client drivers can use this function to allocate a syncpoint for + * subsequent use. A syncpoint returned by this function will be reserved for + * use by the client exclusively. When no longer using a syncpoint, a host1x + * client driver needs to release it using host1x_syncpt_put(). + */ +struct host1x_syncpt *host1x_syncpt_request(struct host1x_client *client, + unsigned long flags) { - struct host1x *host = dev_get_drvdata(dev->parent); - return _host1x_syncpt_alloc(host, dev, client_managed); + struct host1x *host = dev_get_drvdata(client->host->parent); + + return host1x_syncpt_alloc(host, flags, dev_name(client->dev)); } +EXPORT_SYMBOL(host1x_syncpt_request); -void host1x_syncpt_free(struct host1x_syncpt *sp) +static void syncpt_release(struct kref *ref) { - if (!sp) - return; + struct host1x_syncpt *sp = container_of(ref, struct host1x_syncpt, ref); + + atomic_set(&sp->max_val, host1x_syncpt_read(sp)); + sp->locked = false; + + host1x_syncpt_base_free(sp->base); kfree(sp->name); - sp->dev = NULL; + sp->base = NULL; sp->name = NULL; sp->client_managed = false; + + mutex_unlock(&sp->host->syncpt_mutex); } +/** + * host1x_syncpt_put() - free a requested syncpoint + * @sp: host1x syncpoint + * + * Release a syncpoint previously allocated using host1x_syncpt_request(). A + * host1x client driver should call this when the syncpoint is no longer in + * use. + */ +void host1x_syncpt_put(struct host1x_syncpt *sp) +{ + if (!sp) + return; + + kref_put_mutex(&sp->ref, syncpt_release, &sp->host->syncpt_mutex); +} +EXPORT_SYMBOL(host1x_syncpt_put); + void host1x_syncpt_deinit(struct host1x *host) { - int i; struct host1x_syncpt *sp = host->syncpt; + unsigned int i; + for (i = 0; i < host->info->nb_pts; i++, sp++) kfree(sp->name); } -int host1x_syncpt_nb_pts(struct host1x *host) +/** + * host1x_syncpt_read_max() - read maximum syncpoint value + * @sp: host1x syncpoint + * + * The maximum syncpoint value indicates how many operations there are in + * queue, either in channel or in a software thread. + */ +u32 host1x_syncpt_read_max(struct host1x_syncpt *sp) +{ + smp_rmb(); + + return (u32)atomic_read(&sp->max_val); +} +EXPORT_SYMBOL(host1x_syncpt_read_max); + +/** + * host1x_syncpt_read_min() - read minimum syncpoint value + * @sp: host1x syncpoint + * + * The minimum syncpoint value is a shadow of the current sync point value in + * hardware. + */ +u32 host1x_syncpt_read_min(struct host1x_syncpt *sp) +{ + smp_rmb(); + + return (u32)atomic_read(&sp->min_val); +} +EXPORT_SYMBOL(host1x_syncpt_read_min); + +/** + * host1x_syncpt_read() - read the current syncpoint value + * @sp: host1x syncpoint + */ +u32 host1x_syncpt_read(struct host1x_syncpt *sp) +{ + return host1x_syncpt_load(sp); +} +EXPORT_SYMBOL(host1x_syncpt_read); + +unsigned int host1x_syncpt_nb_pts(struct host1x *host) { return host->info->nb_pts; } -int host1x_syncpt_nb_bases(struct host1x *host) +unsigned int host1x_syncpt_nb_bases(struct host1x *host) { return host->info->nb_bases; } -int host1x_syncpt_nb_mlocks(struct host1x *host) +unsigned int host1x_syncpt_nb_mlocks(struct host1x *host) { return host->info->nb_mlocks; } -struct host1x_syncpt *host1x_syncpt_get(struct host1x *host, u32 id) +/** + * host1x_syncpt_get_by_id() - obtain a syncpoint by ID + * @host: host1x controller + * @id: syncpoint ID + */ +struct host1x_syncpt *host1x_syncpt_get_by_id(struct host1x *host, + unsigned int id) +{ + if (id >= host->info->nb_pts) + return NULL; + + if (kref_get_unless_zero(&host->syncpt[id].ref)) + return &host->syncpt[id]; + else + return NULL; +} +EXPORT_SYMBOL(host1x_syncpt_get_by_id); + +/** + * host1x_syncpt_get_by_id_noref() - obtain a syncpoint by ID but don't + * increase the refcount. + * @host: host1x controller + * @id: syncpoint ID + */ +struct host1x_syncpt *host1x_syncpt_get_by_id_noref(struct host1x *host, + unsigned int id) { - if (host->info->nb_pts < id) + if (id >= host->info->nb_pts) return NULL; - return host->syncpt + id; + + return &host->syncpt[id]; +} +EXPORT_SYMBOL(host1x_syncpt_get_by_id_noref); + +/** + * host1x_syncpt_get() - increment syncpoint refcount + * @sp: syncpoint + */ +struct host1x_syncpt *host1x_syncpt_get(struct host1x_syncpt *sp) +{ + kref_get(&sp->ref); + + return sp; +} +EXPORT_SYMBOL(host1x_syncpt_get); + +/** + * host1x_syncpt_get_base() - obtain the wait base associated with a syncpoint + * @sp: host1x syncpoint + */ +struct host1x_syncpt_base *host1x_syncpt_get_base(struct host1x_syncpt *sp) +{ + return sp ? sp->base : NULL; +} +EXPORT_SYMBOL(host1x_syncpt_get_base); + +/** + * host1x_syncpt_base_id() - retrieve the ID of a syncpoint wait base + * @base: host1x syncpoint wait base + */ +u32 host1x_syncpt_base_id(struct host1x_syncpt_base *base) +{ + return base->id; +} +EXPORT_SYMBOL(host1x_syncpt_base_id); + +static void do_nothing(struct kref *ref) +{ +} + +/** + * host1x_syncpt_release_vblank_reservation() - Make VBLANK syncpoint + * available for allocation + * + * @client: host1x bus client + * @syncpt_id: syncpoint ID to make available + * + * Makes VBLANK<i> syncpoint available for allocatation if it was + * reserved at initialization time. This should be called by the display + * driver after it has ensured that any VBLANK increment programming configured + * by the boot chain has been disabled. + */ +void host1x_syncpt_release_vblank_reservation(struct host1x_client *client, + u32 syncpt_id) +{ + struct host1x *host = dev_get_drvdata(client->host->parent); + + if (!host->info->reserve_vblank_syncpts) + return; + + kref_put(&host->syncpt[syncpt_id].ref, do_nothing); } +EXPORT_SYMBOL(host1x_syncpt_release_vblank_reservation); diff --git a/drivers/gpu/host1x/syncpt.h b/drivers/gpu/host1x/syncpt.h index 267c0b9d3647..4c3f3b2f0e9c 100644 --- a/drivers/gpu/host1x/syncpt.h +++ b/drivers/gpu/host1x/syncpt.h @@ -1,28 +1,20 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ /* * Tegra host1x Syncpoints * * Copyright (c) 2010-2013, NVIDIA Corporation. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License - * along with this program. If not, see <http://www.gnu.org/licenses/>. */ #ifndef __HOST1X_SYNCPT_H #define __HOST1X_SYNCPT_H #include <linux/atomic.h> +#include <linux/host1x.h> #include <linux/kernel.h> +#include <linux/kref.h> #include <linux/sched.h> +#include "fence.h" #include "intr.h" struct host1x; @@ -30,18 +22,32 @@ struct host1x; /* Reserved for replacing an expired wait with a NOP */ #define HOST1X_SYNCPT_RESERVED 0 +struct host1x_syncpt_base { + unsigned int id; + bool requested; +}; + struct host1x_syncpt { - int id; + struct kref ref; + + unsigned int id; atomic_t min_val; atomic_t max_val; u32 base_val; const char *name; bool client_managed; struct host1x *host; - struct device *dev; + struct host1x_syncpt_base *base; /* interrupt data */ - struct host1x_syncpt_intr intr; + struct host1x_fence_list fences; + + /* + * If a submission incrementing this syncpoint fails, lock it so that + * further submission cannot be made until application has handled the + * failure. + */ + bool locked; }; /* Initialize sync point array */ @@ -50,33 +56,14 @@ int host1x_syncpt_init(struct host1x *host); /* Free sync point array */ void host1x_syncpt_deinit(struct host1x *host); -/* - * Read max. It indicates how many operations there are in queue, either in - * channel or in a software thread. - * */ -static inline u32 host1x_syncpt_read_max(struct host1x_syncpt *sp) -{ - smp_rmb(); - return (u32)atomic_read(&sp->max_val); -} - -/* - * Read min, which is a shadow of the current sync point value in hardware. - */ -static inline u32 host1x_syncpt_read_min(struct host1x_syncpt *sp) -{ - smp_rmb(); - return (u32)atomic_read(&sp->min_val); -} - /* Return number of sync point supported. */ -int host1x_syncpt_nb_pts(struct host1x *host); +unsigned int host1x_syncpt_nb_pts(struct host1x *host); /* Return number of wait bases supported. */ -int host1x_syncpt_nb_bases(struct host1x *host); +unsigned int host1x_syncpt_nb_bases(struct host1x *host); /* Return number of mlocks supported. */ -int host1x_syncpt_nb_mlocks(struct host1x *host); +unsigned int host1x_syncpt_nb_mlocks(struct host1x *host); /* * Check sync point sanity. If max is larger than min, there have too many @@ -112,9 +99,6 @@ static inline bool host1x_syncpt_idle(struct host1x_syncpt *sp) return (min == max); } -/* Return pointer to struct denoting sync point id. */ -struct host1x_syncpt *host1x_syncpt_get(struct host1x *host, u32 id); - /* Load current value from hardware to the shadow register. */ u32 host1x_syncpt_load(struct host1x_syncpt *sp); @@ -130,33 +114,18 @@ void host1x_syncpt_restore(struct host1x *host); /* Read current wait base value into shadow register and return it. */ u32 host1x_syncpt_load_wait_base(struct host1x_syncpt *sp); -/* Request incrementing a sync point. */ -int host1x_syncpt_incr(struct host1x_syncpt *sp); - /* Indicate future operations by incrementing the sync point max. */ u32 host1x_syncpt_incr_max(struct host1x_syncpt *sp, u32 incrs); -/* Wait until sync point reaches a threshold value, or a timeout. */ -int host1x_syncpt_wait(struct host1x_syncpt *sp, u32 thresh, - long timeout, u32 *value); - /* Check if sync point id is valid. */ static inline int host1x_syncpt_is_valid(struct host1x_syncpt *sp) { return sp->id < host1x_syncpt_nb_pts(sp->host); } -/* Patch a wait by replacing it with a wait for syncpt 0 value 0 */ -int host1x_syncpt_patch_wait(struct host1x_syncpt *sp, void *patch_addr); - -/* Return id of the sync point */ -u32 host1x_syncpt_id(struct host1x_syncpt *sp); - -/* Allocate a sync point for a device. */ -struct host1x_syncpt *host1x_syncpt_request(struct device *dev, - bool client_managed); - -/* Free a sync point. */ -void host1x_syncpt_free(struct host1x_syncpt *sp); +static inline void host1x_syncpt_set_locked(struct host1x_syncpt *sp) +{ + sp->locked = true; +} #endif |
