Merge tag 'driver-core-4.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core

Pull driver core updates from Greg KH:
 "Here is the driver core / firmware changes for 4.2-rc1.

  A number of small changes all over the place in the driver core, and
  in the firmware subsystem.  Nothing really major, full details in the
  shortlog.  Some of it is a bit of churn, given that the platform
  driver probing changes was found to not work well, so they were
  reverted.

  All of these have been in linux-next for a while with no reported
  issues"

* tag 'driver-core-4.2-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core: (31 commits)
  Revert "base/platform: Only insert MEM and IO resources"
  Revert "base/platform: Continue on insert_resource() error"
  Revert "of/platform: Use platform_device interface"
  Revert "base/platform: Remove code duplication"
  firmware: add missing kfree for work on async call
  fs: sysfs: don't pass count == 0 to bin file readers
  base:dd - Fix for typo in comment to function driver_deferred_probe_trigger().
  base/platform: Remove code duplication
  of/platform: Use platform_device interface
  base/platform: Continue on insert_resource() error
  base/platform: Only insert MEM and IO resources
  firmware: use const for remaining firmware names
  firmware: fix possible use after free on name on asynchronous request
  firmware: check for file truncation on direct firmware loading
  firmware: fix __getname() missing failure check
  drivers: of/base: move of_init to driver_init
  drivers/base: cacheinfo: fix annoying typo when DT nodes are absent
  sysfs: disambiguate between "error code" and "failure" in comments
  driver-core: fix build for !CONFIG_MODULES
  driver-core: make __device_attach() static
  ...

Documentation/ABI/testing/sysfs-devices-system-cpu

@@ -243,7 +243,7 @@ Description:	Parameters for the CPU cache attributes
 		coherency_line_size: the minimum amount of data in bytes that gets
 				     transferred from memory to cache
 
-		level: the cache hierarcy in the multi-level cache configuration
+		level: the cache hierarchy in the multi-level cache configuration
 
 		number_of_sets: total number of sets in the cache, a set is a
 				collection of cache lines with the same cache index

Documentation/kernel-parameters.txt

@@ -953,6 +953,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 			auto	selects the default scheme, which automatically
 				enables eagerfpu restore for xsaveopt.
 
+	module.async_probe [KNL]
+			Enable asynchronous probe on this module.
+
 	early_ioremap_debug [KNL]
 			Enable debug messages in early_ioremap support. This
 			is useful for tracking down temporary early mappings

MAINTAINERS

@@ -3450,16 +3450,17 @@ F:	drivers/block/drbd/
 F:	lib/lru_cache.c
 F:	Documentation/blockdev/drbd/
 
-DRIVER CORE, KOBJECTS, DEBUGFS AND SYSFS
+DRIVER CORE, KOBJECTS, DEBUGFS, KERNFS AND SYSFS
 M:	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core.git
 S:	Supported
 F:	Documentation/kobject.txt
 F:	drivers/base/
-F:	fs/sysfs/
 F:	fs/debugfs/
-F:	include/linux/kobj*
+F:	fs/kernfs/
+F:	fs/sysfs/
 F:	include/linux/debugfs.h
+F:	include/linux/kobj*
 F:	lib/kobj*
 
 DRM DRIVERS

arch/powerpc/mm/hugetlbpage.c

@@ -336,7 +336,7 @@ int alloc_bootmem_huge_page(struct hstate *hstate)
 unsigned long gpage_npages[MMU_PAGE_COUNT];
 
 static int __init do_gpage_early_setup(char *param, char *val,
-				       const char *unused)
+				       const char *unused, void *arg)
 {
 	static phys_addr_t size;
 	unsigned long npages;
@@ -385,7 +385,7 @@ void __init reserve_hugetlb_gpages(void)
 
 	strlcpy(cmdline, boot_command_line, COMMAND_LINE_SIZE);
 	parse_args("hugetlb gpages", cmdline, NULL, 0, 0, 0,
-			&do_gpage_early_setup);
+			NULL, &do_gpage_early_setup);
 
 	/*
 	 * Walk gpage list in reverse, allocating larger page sizes first.

drivers/base/base.h

@@ -116,6 +116,7 @@ static inline int driver_match_device(struct device_driver *drv,
 {
 	return drv->bus->match ? drv->bus->match(dev, drv) : 1;
 }
+extern bool driver_allows_async_probing(struct device_driver *drv);
 
 extern int driver_add_groups(struct device_driver *drv,
 			     const struct attribute_group **groups);

drivers/base/bus.c

@@ -10,6 +10,7 @@
  *
  */
 
+#include <linux/async.h>
 #include <linux/device.h>
 #include <linux/module.h>
 #include <linux/errno.h>
@@ -549,15 +550,12 @@ void bus_probe_device(struct device *dev)
 {
 	struct bus_type *bus = dev->bus;
 	struct subsys_interface *sif;
-	int ret;
 
 	if (!bus)
 		return;
 
-	if (bus->p->drivers_autoprobe) {
-		ret = device_attach(dev);
-		WARN_ON(ret < 0);
-	}
+	if (bus->p->drivers_autoprobe)
+		device_initial_probe(dev);
 
 	mutex_lock(&bus->p->mutex);
 	list_for_each_entry(sif, &bus->p->interfaces, node)
@@ -659,6 +657,17 @@ static ssize_t uevent_store(struct device_driver *drv, const char *buf,
 }
 static DRIVER_ATTR_WO(uevent);
 
+static void driver_attach_async(void *_drv, async_cookie_t cookie)
+{
+	struct device_driver *drv = _drv;
+	int ret;
+
+	ret = driver_attach(drv);
+
+	pr_debug("bus: '%s': driver %s async attach completed: %d\n",
+		 drv->bus->name, drv->name, ret);
+}
+
 /**
  * bus_add_driver - Add a driver to the bus.
  * @drv: driver.
@@ -691,10 +700,16 @@ int bus_add_driver(struct device_driver *drv)
 
 	klist_add_tail(&priv->knode_bus, &bus->p->klist_drivers);
 	if (drv->bus->p->drivers_autoprobe) {
+		if (driver_allows_async_probing(drv)) {
+			pr_debug("bus: '%s': probing driver %s asynchronously\n",
+				drv->bus->name, drv->name);
+			async_schedule(driver_attach_async, drv);
+		} else {
 			error = driver_attach(drv);
 			if (error)
 				goto out_unregister;
 		}
+	}
 	module_add_driver(drv->owner, drv);
 
 	error = driver_create_file(drv, &driver_attr_uevent);

drivers/base/cacheinfo.c

@@ -191,12 +191,12 @@ static int detect_cache_attributes(unsigned int cpu)
 	if (ret)
 		goto free_ci;
 	/*
-	 * For systems using DT for cache hierarcy, of_node and shared_cpu_map
+	 * For systems using DT for cache hierarchy, of_node and shared_cpu_map
 	 * will be set up here only if they are not populated already
 	 */
 	ret = cache_shared_cpu_map_setup(cpu);
 	if (ret) {
-		pr_warn("Unable to detect cache hierarcy from DT for CPU %d\n",
+		pr_warn("Unable to detect cache hierarchy from DT for CPU %d\n",
 			cpu);
 		goto free_ci;
 	}

drivers/base/cpu.c

@@ -16,6 +16,7 @@
 #include <linux/acpi.h>
 #include <linux/of.h>
 #include <linux/cpufeature.h>
+#include <linux/tick.h>
 
 #include "base.h"
 
@@ -265,6 +266,30 @@ static ssize_t print_cpus_offline(struct device *dev,
 }
 static DEVICE_ATTR(offline, 0444, print_cpus_offline, NULL);
 
+static ssize_t print_cpus_isolated(struct device *dev,
+				  struct device_attribute *attr, char *buf)
+{
+	int n = 0, len = PAGE_SIZE-2;
+
+	n = scnprintf(buf, len, "%*pbl\n", cpumask_pr_args(cpu_isolated_map));
+
+	return n;
+}
+static DEVICE_ATTR(isolated, 0444, print_cpus_isolated, NULL);
+
+#ifdef CONFIG_NO_HZ_FULL
+static ssize_t print_cpus_nohz_full(struct device *dev,
+				  struct device_attribute *attr, char *buf)
+{
+	int n = 0, len = PAGE_SIZE-2;
+
+	n = scnprintf(buf, len, "%*pbl\n", cpumask_pr_args(tick_nohz_full_mask));
+
+	return n;
+}
+static DEVICE_ATTR(nohz_full, 0444, print_cpus_nohz_full, NULL);
+#endif
+
 static void cpu_device_release(struct device *dev)
 {
 	/*
@@ -431,6 +456,10 @@ static struct attribute *cpu_root_attrs[] = {
 	&cpu_attrs[2].attr.attr,
 	&dev_attr_kernel_max.attr,
 	&dev_attr_offline.attr,
+	&dev_attr_isolated.attr,
+#ifdef CONFIG_NO_HZ_FULL
+	&dev_attr_nohz_full.attr,
+#endif
 #ifdef CONFIG_GENERIC_CPU_AUTOPROBE
 	&dev_attr_modalias.attr,
 #endif

drivers/base/dd.c

@@ -141,7 +141,7 @@ static bool driver_deferred_probe_enable = false;
  * more than one device is probing at the same time, it is possible for one
  * probe to complete successfully while another is about to defer. If the second
  * depends on the first, then it will get put on the pending list after the
- * trigger event has already occured and will be stuck there.
+ * trigger event has already occurred and will be stuck there.
  *
  * The atomic 'deferred_trigger_count' is used to determine if a successful
  * trigger has occurred in the midst of probing a driver. If the trigger count
@@ -417,31 +417,107 @@ int driver_probe_device(struct device_driver *drv, struct device *dev)
 	return ret;
 }
 
-static int __device_attach(struct device_driver *drv, void *data)
+bool driver_allows_async_probing(struct device_driver *drv)
 {
-	struct device *dev = data;
+	switch (drv->probe_type) {
+	case PROBE_PREFER_ASYNCHRONOUS:
+		return true;
+
+	case PROBE_FORCE_SYNCHRONOUS:
+		return false;
+
+	default:
+		if (module_requested_async_probing(drv->owner))
+			return true;
+
+		return false;
+	}
+}
+
+struct device_attach_data {
+	struct device *dev;
+
+	/*
+	 * Indicates whether we are are considering asynchronous probing or
+	 * not. Only initial binding after device or driver registration
+	 * (including deferral processing) may be done asynchronously, the
+	 * rest is always synchronous, as we expect it is being done by
+	 * request from userspace.
+	 */
+	bool check_async;
+
+	/*
+	 * Indicates if we are binding synchronous or asynchronous drivers.
+	 * When asynchronous probing is enabled we'll execute 2 passes
+	 * over drivers: first pass doing synchronous probing and second
+	 * doing asynchronous probing (if synchronous did not succeed -
+	 * most likely because there was no driver requiring synchronous
+	 * probing - and we found asynchronous driver during first pass).
+	 * The 2 passes are done because we can't shoot asynchronous
+	 * probe for given device and driver from bus_for_each_drv() since
+	 * driver pointer is not guaranteed to stay valid once
+	 * bus_for_each_drv() iterates to the next driver on the bus.
+	 */
+	bool want_async;
+
+	/*
+	 * We'll set have_async to 'true' if, while scanning for matching
+	 * driver, we'll encounter one that requests asynchronous probing.
+	 */
+	bool have_async;
+};
+
+static int __device_attach_driver(struct device_driver *drv, void *_data)
+{
+	struct device_attach_data *data = _data;
+	struct device *dev = data->dev;
+	bool async_allowed;
+
+	/*
+	 * Check if device has already been claimed. This may
+	 * happen with driver loading, device discovery/registration,
+	 * and deferred probe processing happens all at once with
+	 * multiple threads.
+	 */
+	if (dev->driver)
+		return -EBUSY;
 
 	if (!driver_match_device(drv, dev))
 		return 0;
 
+	async_allowed = driver_allows_async_probing(drv);
+
+	if (async_allowed)
+		data->have_async = true;
+
+	if (data->check_async && async_allowed != data->want_async)
+		return 0;
+
 	return driver_probe_device(drv, dev);
 }
 
-/**
- * device_attach - try to attach device to a driver.
- * @dev: device.
- *
- * Walk the list of drivers that the bus has and call
- * driver_probe_device() for each pair. If a compatible
- * pair is found, break out and return.
- *
- * Returns 1 if the device was bound to a driver;
- * 0 if no matching driver was found;
- * -ENODEV if the device is not registered.
- *
- * When called for a USB interface, @dev->parent lock must be held.
- */
-int device_attach(struct device *dev)
+static void __device_attach_async_helper(void *_dev, async_cookie_t cookie)
+{
+	struct device *dev = _dev;
+	struct device_attach_data data = {
+		.dev		= dev,
+		.check_async	= true,
+		.want_async	= true,
+	};
+
+	device_lock(dev);
+
+	bus_for_each_drv(dev->bus, NULL, &data, __device_attach_driver);
+	dev_dbg(dev, "async probe completed\n");
+
+	pm_request_idle(dev);
+
+	device_unlock(dev);
+
+	put_device(dev);
+}
+
+static int __device_attach(struct device *dev, bool allow_async)
 {
 	int ret = 0;
 
@@ -459,15 +535,59 @@ int device_attach(struct device *dev)
 			ret = 0;
 		}
 	} else {
-		ret = bus_for_each_drv(dev->bus, NULL, dev, __device_attach);
+		struct device_attach_data data = {
+			.dev = dev,
+			.check_async = allow_async,
+			.want_async = false,
+		};
+
+		ret = bus_for_each_drv(dev->bus, NULL, &data,
+					__device_attach_driver);
+		if (!ret && allow_async && data.have_async) {
+			/*
+			 * If we could not find appropriate driver
+			 * synchronously and we are allowed to do
+			 * async probes and there are drivers that
+			 * want to probe asynchronously, we'll
+			 * try them.
+			 */
+			dev_dbg(dev, "scheduling asynchronous probe\n");
+			get_device(dev);
+			async_schedule(__device_attach_async_helper, dev);
+		} else {
 			pm_request_idle(dev);
 		}
+	}
 out_unlock:
 	device_unlock(dev);
 	return ret;
 }
+
+/**
+ * device_attach - try to attach device to a driver.
+ * @dev: device.
+ *
+ * Walk the list of drivers that the bus has and call
+ * driver_probe_device() for each pair. If a compatible
+ * pair is found, break out and return.
+ *
+ * Returns 1 if the device was bound to a driver;
+ * 0 if no matching driver was found;
+ * -ENODEV if the device is not registered.
+ *
+ * When called for a USB interface, @dev->parent lock must be held.
+ */
+int device_attach(struct device *dev)
+{
+	return __device_attach(dev, false);
+}
 EXPORT_SYMBOL_GPL(device_attach);
 
+void device_initial_probe(struct device *dev)
+{
+	__device_attach(dev, true);
+}
+
 static int __driver_attach(struct device *dev, void *data)
 {
 	struct device_driver *drv = data;
@@ -522,6 +642,9 @@ static void __device_release_driver(struct device *dev)
 
 	drv = dev->driver;
 	if (drv) {
+		if (driver_allows_async_probing(drv))
+			async_synchronize_full();
+
 		pm_runtime_get_sync(dev);
 
 		driver_sysfs_remove(dev);

drivers/base/firmware_class.c

@@ -150,17 +150,17 @@ struct firmware_buf {
 	int page_array_size;
 	struct list_head pending_list;
 #endif
-	char fw_id[];
+	const char *fw_id;
 };
 
 struct fw_cache_entry {
 	struct list_head list;
-	char name[];
+	const char *name;
 };
 
 struct fw_name_devm {
 	unsigned long magic;
-	char name[];
+	const char *name;
 };
 
 #define to_fwbuf(d) container_of(d, struct firmware_buf, ref)
@@ -181,13 +181,17 @@ static struct firmware_buf *__allocate_fw_buf(const char *fw_name,
 {
 	struct firmware_buf *buf;
 
-	buf = kzalloc(sizeof(*buf) + strlen(fw_name) + 1, GFP_ATOMIC);
-
+	buf = kzalloc(sizeof(*buf), GFP_ATOMIC);
 	if (!buf)
-		return buf;
+		return NULL;
+
+	buf->fw_id = kstrdup_const(fw_name, GFP_ATOMIC);
+	if (!buf->fw_id) {
+		kfree(buf);
+		return NULL;
+	}
 
 	kref_init(&buf->ref);
-	strcpy(buf->fw_id, fw_name);
 	buf->fwc = fwc;
 	init_completion(&buf->completion);
 #ifdef CONFIG_FW_LOADER_USER_HELPER
@@ -257,6 +261,7 @@ static void __fw_free_buf(struct kref *ref)
 	} else
 #endif
 		vfree(buf->data);
+	kfree_const(buf->fw_id);
 	kfree(buf);
 }
 
@@ -320,9 +325,13 @@ static int fw_read_file_contents(struct file *file, struct firmware_buf *fw_buf)
 static int fw_get_filesystem_firmware(struct device *device,
 				       struct firmware_buf *buf)
 {
-	int i;
+	int i, len;
 	int rc = -ENOENT;
-	char *path = __getname();
+	char *path;
+
+	path = __getname();
+	if (!path)
+		return -ENOMEM;
 
 	for (i = 0; i < ARRAY_SIZE(fw_path); i++) {
 		struct file *file;
@@ -331,7 +340,12 @@ static int fw_get_filesystem_firmware(struct device *device,
 		if (!fw_path[i][0])
 			continue;
 
-		snprintf(path, PATH_MAX, "%s/%s", fw_path[i], buf->fw_id);
+		len = snprintf(path, PATH_MAX, "%s/%s",
+			       fw_path[i], buf->fw_id);
+		if (len >= PATH_MAX) {
+			rc = -ENAMETOOLONG;
+			break;
+		}
 
 		file = filp_open(path, O_RDONLY, 0);
 		if (IS_ERR(file))
@@ -392,6 +406,7 @@ static void fw_name_devm_release(struct device *dev, void *res)
 	if (fwn->magic == (unsigned long)&fw_cache)
 		pr_debug("%s: fw_name-%s devm-%p released\n",
 				__func__, fwn->name, res);
+	kfree_const(fwn->name);
 }
 
 static int fw_devm_match(struct device *dev, void *res,
@@ -422,13 +437,17 @@ static int fw_add_devm_name(struct device *dev, const char *name)
 	if (fwn)
 		return 1;
 
-	fwn = devres_alloc(fw_name_devm_release, sizeof(struct fw_name_devm) +
-			   strlen(name) + 1, GFP_KERNEL);
+	fwn = devres_alloc(fw_name_devm_release, sizeof(struct fw_name_devm),
+			   GFP_KERNEL);
 	if (!fwn)
 		return -ENOMEM;
+	fwn->name = kstrdup_const(name, GFP_KERNEL);
+	if (!fwn->name) {
+		kfree(fwn);
+		return -ENOMEM;
+	}
 
 	fwn->magic = (unsigned long)&fw_cache;
-	strcpy(fwn->name, name);
 	devres_add(dev, fwn);
 
 	return 0;
@@ -1247,6 +1266,7 @@ static void request_firmware_work_func(struct work_struct *work)
 	put_device(fw_work->device); /* taken in request_firmware_nowait() */
 
 	module_put(fw_work->module);
+	kfree_const(fw_work->name);
 	kfree(fw_work);
 }
 
@@ -1286,7 +1306,11 @@ request_firmware_nowait(
 		return -ENOMEM;
 
 	fw_work->module = module;
-	fw_work->name = name;
+	fw_work->name = kstrdup_const(name, gfp);
+	if (!fw_work->name) {
+		kfree(fw_work);
+		return -ENOMEM;
+	}
 	fw_work->device = device;
 	fw_work->context = context;
 	fw_work->cont = cont;
@@ -1294,6 +1318,7 @@ request_firmware_nowait(
 		(uevent ? FW_OPT_UEVENT : FW_OPT_USERHELPER);
 
 	if (!try_module_get(module)) {
+		kfree_const(fw_work->name);
 		kfree(fw_work);
 		return -EFAULT;
 	}
@@ -1384,11 +1409,16 @@ static struct fw_cache_entry *alloc_fw_cache_entry(const char *name)
 {
 	struct fw_cache_entry *fce;
 
-	fce = kzalloc(sizeof(*fce) + strlen(name) + 1, GFP_ATOMIC);
+	fce = kzalloc(sizeof(*fce), GFP_ATOMIC);
 	if (!fce)
 		goto exit;
 
-	strcpy(fce->name, name);
+	fce->name = kstrdup_const(name, GFP_ATOMIC);
+	if (!fce->name) {
+		kfree(fce);
+		fce = NULL;
+		goto exit;
+	}
 exit:
 	return fce;
 }
@@ -1428,6 +1458,7 @@ static int fw_cache_piggyback_on_request(const char *name)
 
 static void free_fw_cache_entry(struct fw_cache_entry *fce)
 {
+	kfree_const(fce->name);
 	kfree(fce);
 }
 

drivers/base/platform.c

@@ -613,6 +613,19 @@ int __init_or_module __platform_driver_probe(struct platform_driver *drv,
 {
 	int retval, code;
 
+	if (drv->driver.probe_type == PROBE_PREFER_ASYNCHRONOUS) {
+		pr_err("%s: drivers registered with %s can not be probed asynchronously\n",
+			 drv->driver.name, __func__);
+		return -EINVAL;
+	}
+
+	/*
+	 * We have to run our probes synchronously because we check if
+	 * we find any devices to bind to and exit with error if there
+	 * are any.
+	 */
+	drv->driver.probe_type = PROBE_FORCE_SYNCHRONOUS;
+
 	/*
 	 * Prevent driver from requesting probe deferral to avoid further
 	 * futile probe attempts.

drivers/edac/amd64_edac.c

@@ -2964,6 +2964,7 @@ static struct pci_driver amd64_pci_driver = {
 	.probe		= probe_one_instance,
 	.remove		= remove_one_instance,
 	.id_table	= amd64_pci_table,
+	.driver.probe_type = PROBE_FORCE_SYNCHRONOUS,
 };
 
 static void setup_pci_device(void)

fs/kernfs/file.c

@@ -785,7 +785,6 @@ static unsigned int kernfs_fop_poll(struct file *filp, poll_table *wait)
 	struct kernfs_node *kn = filp->f_path.dentry->d_fsdata;
 	struct kernfs_open_node *on = kn->attr.open;
 
-	/* need parent for the kobj, grab both */
 	if (!kernfs_get_active(kn))
 		goto trigger;
 

fs/sysfs/file.c

@@ -90,7 +90,7 @@ static ssize_t sysfs_kf_bin_read(struct kernfs_open_file *of, char *buf,
 		return 0;
 
 	if (size) {
-		if (pos > size)
+		if (pos >= size)
 			return 0;
 		if (pos + count > size)
 			count = size - pos;

fs/sysfs/group.c

@@ -135,7 +135,7 @@ static int internal_create_group(struct kobject *kobj, int update,
  * This function creates a group for the first time.  It will explicitly
  * warn and error if any of the attribute files being created already exist.
  *
- * Returns 0 on success or error.
+ * Returns 0 on success or error code on failure.
  */
 int sysfs_create_group(struct kobject *kobj,
 		       const struct attribute_group *grp)
@@ -155,7 +155,7 @@ EXPORT_SYMBOL_GPL(sysfs_create_group);
  * It will explicitly warn and error if any of the attribute files being
  * created already exist.
  *
- * Returns 0 on success or error code from sysfs_create_group on error.
+ * Returns 0 on success or error code from sysfs_create_group on failure.
  */
 int sysfs_create_groups(struct kobject *kobj,
 			const struct attribute_group **groups)
@@ -193,7 +193,7 @@ EXPORT_SYMBOL_GPL(sysfs_create_groups);
  * The primary use for this function is to call it after making a change
  * that affects group visibility.
  *
- * Returns 0 on success or error.
+ * Returns 0 on success or error code on failure.
  */
 int sysfs_update_group(struct kobject *kobj,
 		       const struct attribute_group *grp)

include/linux/cacheinfo.h

@@ -19,7 +19,7 @@ enum cache_type {
 /**
  * struct cacheinfo - represent a cache leaf node
  * @type: type of the cache - data, inst or unified
- * @level: represents the hierarcy in the multi-level cache
+ * @level: represents the hierarchy in the multi-level cache
  * @coherency_line_size: size of each cache line usually representing
  *	the minimum amount of data that gets transferred from memory
  * @number_of_sets: total number of sets, a set is a collection of cache

include/linux/device.h

@@ -195,6 +195,34 @@ extern int bus_unregister_notifier(struct bus_type *bus,
 extern struct kset *bus_get_kset(struct bus_type *bus);
 extern struct klist *bus_get_device_klist(struct bus_type *bus);
 
+/**
+ * enum probe_type - device driver probe type to try
+ *	Device drivers may opt in for special handling of their
+ *	respective probe routines. This tells the core what to
+ *	expect and prefer.
+ *
+ * @PROBE_DEFAULT_STRATEGY: Used by drivers that work equally well
+ *	whether probed synchronously or asynchronously.
+ * @PROBE_PREFER_ASYNCHRONOUS: Drivers for "slow" devices which
+ *	probing order is not essential for booting the system may
+ *	opt into executing their probes asynchronously.
+ * @PROBE_FORCE_SYNCHRONOUS: Use this to annotate drivers that need
+ *	their probe routines to run synchronously with driver and
+ *	device registration (with the exception of -EPROBE_DEFER
+ *	handling - re-probing always ends up being done asynchronously).
+ *
+ * Note that the end goal is to switch the kernel to use asynchronous
+ * probing by default, so annotating drivers with
+ * %PROBE_PREFER_ASYNCHRONOUS is a temporary measure that allows us
+ * to speed up boot process while we are validating the rest of the
+ * drivers.
+ */
+enum probe_type {
+	PROBE_DEFAULT_STRATEGY,
+	PROBE_PREFER_ASYNCHRONOUS,
+	PROBE_FORCE_SYNCHRONOUS,
+};
+
 /**
  * struct device_driver - The basic device driver structure
  * @name:	Name of the device driver.
@@ -202,6 +230,7 @@ extern struct klist *bus_get_device_klist(struct bus_type *bus);
  * @owner:	The module owner.
  * @mod_name:	Used for built-in modules.
  * @suppress_bind_attrs: Disables bind/unbind via sysfs.
+ * @probe_type:	Type of the probe (synchronous or asynchronous) to use.
  * @of_match_table: The open firmware table.
  * @acpi_match_table: The ACPI match table.
  * @probe:	Called to query the existence of a specific device,
@@ -235,6 +264,7 @@ struct device_driver {
 	const char		*mod_name;	/* used for built-in modules */
 
 	bool suppress_bind_attrs;	/* disables bind/unbind via sysfs */
+	enum probe_type probe_type;
 
 	const struct of_device_id	*of_match_table;
 	const struct acpi_device_id	*acpi_match_table;
@@ -975,6 +1005,7 @@ extern int __must_check device_bind_driver(struct device *dev);
 extern void device_release_driver(struct device *dev);
 extern int  __must_check device_attach(struct device *dev);
 extern int __must_check driver_attach(struct device_driver *drv);
+extern void device_initial_probe(struct device *dev);
 extern int __must_check device_reprobe(struct device *dev);
 
 /*

include/linux/module.h

@@ -257,6 +257,8 @@ struct module {
 	bool sig_ok;
 #endif
 
+	bool async_probe_requested;
+
 	/* symbols that will be GPL-only in the near future. */
 	const struct kernel_symbol *gpl_future_syms;
 	const unsigned long *gpl_future_crcs;
@@ -508,6 +510,11 @@ int unregister_module_notifier(struct notifier_block *nb);
 
 extern void print_modules(void);
 
+static inline bool module_requested_async_probing(struct module *module)
+{
+	return module && module->async_probe_requested;
+}
+
 #else /* !CONFIG_MODULES... */
 
 /* Given an address, look for it in the exception tables. */
@@ -618,6 +625,12 @@ static inline int unregister_module_notifier(struct notifier_block *nb)
 static inline void print_modules(void)
 {
 }
+
+static inline bool module_requested_async_probing(struct module *module)
+{
+	return false;
+}
+
 #endif /* CONFIG_MODULES */
 
 #ifdef CONFIG_SYSFS

include/linux/moduleparam.h

@@ -310,6 +310,15 @@ static inline void __kernel_param_unlock(void)
 #define core_param(name, var, type, perm)				\
 	param_check_##type(name, &(var));				\
 	__module_param_call("", name, &param_ops_##type, &var, perm, -1, 0)
+
+/**
+ * core_param_unsafe - same as core_param but taints kernel
+ */
+#define core_param_unsafe(name, var, type, perm)		\
+	param_check_##type(name, &(var));				\
+	__module_param_call("", name, &param_ops_##type, &var, perm,	\
+			    -1, KERNEL_PARAM_FL_UNSAFE)
+
 #endif /* !MODULE */
 
 /**
@@ -357,8 +366,9 @@ extern char *parse_args(const char *name,
 		      unsigned num,
 		      s16 level_min,
 		      s16 level_max,
+		      void *arg,
 		      int (*unknown)(char *param, char *val,
-			      const char *doing));
+				     const char *doing, void *arg));
 
 /* Called by module remove. */
 #ifdef CONFIG_SYSFS

init/main.c

@@ -235,7 +235,8 @@ static int __init loglevel(char *str)
 early_param("loglevel", loglevel);
 
 /* Change NUL term back to "=", to make "param" the whole string. */
-static int __init repair_env_string(char *param, char *val, const char *unused)
+static int __init repair_env_string(char *param, char *val,
+				    const char *unused, void *arg)
 {
 	if (val) {
 		/* param=val or param="val"? */
@@ -252,14 +253,15 @@ static int __init repair_env_string(char *param, char *val, const char *unused)
 }
 
 /* Anything after -- gets handed straight to init. */
-static int __init set_init_arg(char *param, char *val, const char *unused)
+static int __init set_init_arg(char *param, char *val,
+			       const char *unused, void *arg)
 {
 	unsigned int i;
 
 	if (panic_later)
 		return 0;
 
-	repair_env_string(param, val, unused);
+	repair_env_string(param, val, unused, NULL);
 
 	for (i = 0; argv_init[i]; i++) {
 		if (i == MAX_INIT_ARGS) {
@@ -276,9 +278,10 @@ static int __init set_init_arg(char *param, char *val, const char *unused)
  * Unknown boot options get handed to init, unless they look like
  * unused parameters (modprobe will find them in /proc/cmdline).
  */
-static int __init unknown_bootoption(char *param, char *val, const char *unused)
+static int __init unknown_bootoption(char *param, char *val,
+				     const char *unused, void *arg)
 {
-	repair_env_string(param, val, unused);
+	repair_env_string(param, val, unused, NULL);
 
 	/* Handle obsolete-style parameters */
 	if (obsolete_checksetup(param))
@@ -410,7 +413,8 @@ static noinline void __init_refok rest_init(void)
 }
 
 /* Check for early params. */
-static int __init do_early_param(char *param, char *val, const char *unused)
+static int __init do_early_param(char *param, char *val,
+				 const char *unused, void *arg)
 {
 	const struct obs_kernel_param *p;
 
@@ -429,7 +433,8 @@ static int __init do_early_param(char *param, char *val, const char *unused)
 
 void __init parse_early_options(char *cmdline)
 {
-	parse_args("early options", cmdline, NULL, 0, 0, 0, do_early_param);
+	parse_args("early options", cmdline, NULL, 0, 0, 0, NULL,
+		   do_early_param);
 }
 
 /* Arch code calls this early on, or if not, just before other parsing. */
@@ -535,10 +540,10 @@ asmlinkage __visible void __init start_kernel(void)
 	after_dashes = parse_args("Booting kernel",
 				  static_command_line, __start___param,
 				  __stop___param - __start___param,
-				  -1, -1, &unknown_bootoption);
+				  -1, -1, NULL, &unknown_bootoption);
 	if (!IS_ERR_OR_NULL(after_dashes))
 		parse_args("Setting init args", after_dashes, NULL, 0, -1, -1,
-			   set_init_arg);
+			   NULL, set_init_arg);
 
 	jump_label_init();
 
@@ -848,7 +853,7 @@ static void __init do_initcall_level(int level)
 		   initcall_command_line, __start___param,
 		   __stop___param - __start___param,
 		   level, level,
-		   &repair_env_string);
+		   NULL, &repair_env_string);
 
 	for (fn = initcall_levels[level]; fn < initcall_levels[level+1]; fn++)
 		do_one_initcall(*fn);

kernel/module.c

@@ -3107,7 +3107,7 @@ static noinline int do_init_module(struct module *mod)
 	 *
 	 * http://thread.gmane.org/gmane.linux.kernel/1420814
 	 */
-	if (current->flags & PF_USED_ASYNC)
+	if (!mod->async_probe_requested && (current->flags & PF_USED_ASYNC))
 		async_synchronize_full();
 
 	mutex_lock(&module_mutex);
@@ -3237,10 +3237,19 @@ static int complete_formation(struct module *mod, struct load_info *info)
 	return err;
 }
 
-static int unknown_module_param_cb(char *param, char *val, const char *modname)
+static int unknown_module_param_cb(char *param, char *val, const char *modname,
+				   void *arg)
 {
+	struct module *mod = arg;
+	int ret;
+
+	if (strcmp(param, "async_probe") == 0) {
+		mod->async_probe_requested = true;
+		return 0;
+	}
+
 	/* Check for magic 'dyndbg' arg */
-	int ret = ddebug_dyndbg_module_param_cb(param, val, modname);
+	ret = ddebug_dyndbg_module_param_cb(param, val, modname);
 	if (ret != 0)
 		pr_warn("%s: unknown parameter '%s' ignored\n", modname, param);
 	return 0;
@@ -3342,7 +3351,8 @@ static int load_module(struct load_info *info, const char __user *uargs,
 
 	/* Module is ready to execute: parsing args may do that. */
 	after_dashes = parse_args(mod->name, mod->args, mod->kp, mod->num_kp,
-				  -32768, 32767, unknown_module_param_cb);
+				  -32768, 32767, NULL,
+				  unknown_module_param_cb);
 	if (IS_ERR(after_dashes)) {
 		err = PTR_ERR(after_dashes);
 		goto bug_cleanup;

kernel/params.c

@@ -100,8 +100,9 @@ static int parse_one(char *param,
 		     unsigned num_params,
 		     s16 min_level,
 		     s16 max_level,
+		     void *arg,
 		     int (*handle_unknown)(char *param, char *val,
-				     const char *doing))
+				     const char *doing, void *arg))
 {
 	unsigned int i;
 	int err;
@@ -128,7 +129,7 @@ static int parse_one(char *param,
 
 	if (handle_unknown) {
 		pr_debug("doing %s: %s='%s'\n", doing, param, val);
-		return handle_unknown(param, val, doing);
+		return handle_unknown(param, val, doing, arg);
 	}
 
 	pr_debug("Unknown argument '%s'\n", param);
@@ -194,7 +195,9 @@ char *parse_args(const char *doing,
 		 unsigned num,
 		 s16 min_level,
 		 s16 max_level,
-		 int (*unknown)(char *param, char *val, const char *doing))
+		 void *arg,
+		 int (*unknown)(char *param, char *val,
+				const char *doing, void *arg))
 {
 	char *param, *val;
 
@@ -214,7 +217,7 @@ char *parse_args(const char *doing,
 			return args;
 		irq_was_disabled = irqs_disabled();
 		ret = parse_one(param, val, doing, params, num,
-				min_level, max_level, unknown);
+				min_level, max_level, arg, unknown);
 		if (irq_was_disabled && !irqs_disabled())
 			pr_warn("%s: option '%s' enabled irq's!\n",
 				doing, param);

lib/dynamic_debug.c

@@ -887,7 +887,7 @@ static int ddebug_dyndbg_param_cb(char *param, char *val,
 
 /* handle both dyndbg and $module.dyndbg params at boot */
 static int ddebug_dyndbg_boot_param_cb(char *param, char *val,
-				const char *unused)
+				const char *unused, void *arg)
 {
 	vpr_info("%s=\"%s\"\n", param, val);
 	return ddebug_dyndbg_param_cb(param, val, NULL, 0);
@@ -1028,7 +1028,7 @@ static int __init dynamic_debug_init(void)
 	 */
 	cmdline = kstrdup(saved_command_line, GFP_KERNEL);
 	parse_args("dyndbg params", cmdline, NULL,
-		   0, 0, 0, &ddebug_dyndbg_boot_param_cb);
+		   0, 0, 0, NULL, &ddebug_dyndbg_boot_param_cb);
 	kfree(cmdline);
 	return 0;