11 files changed, 805 insertions, 78 deletions

diff --git a/kernel/Makefile b/kernel/Makefile
index 864ff75d65f2..6aebdeb2aa34 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -100,6 +100,7 @@ obj-$(CONFIG_SLOW_WORK_DEBUG) += slow-work-debugfs.o
 obj-$(CONFIG_PERF_EVENTS) += perf_event.o
 obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o
 obj-$(CONFIG_USER_RETURN_NOTIFIER) += user-return-notifier.o
+obj-$(CONFIG_PADATA) += padata.o
 
 ifneq ($(CONFIG_SCHED_OMIT_FRAME_POINTER),y)
 # According to Alan Modra <alan@linuxcare.com.au>, the -fno-omit-frame-pointer is
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index b7df302a0204..c4b43430d393 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -93,6 +93,7 @@ static struct kprobe_blackpoint kprobe_blacklist[] = {
 	{"native_get_debugreg",},
 	{"irq_entries_start",},
 	{"common_interrupt",},
+	{"mcount",},	/* mcount can be called from everywhere */
 	{NULL}    /* Terminator */
 };
 
diff --git a/kernel/padata.c b/kernel/padata.c
new file mode 100644
index 000000000000..6f9bcb8313d6
--- /dev/null
+++ b/kernel/padata.c
@@ -0,0 +1,690 @@
+/*
+ * padata.c - generic interface to process data streams in parallel
+ *
+ * Copyright (C) 2008, 2009 secunet Security Networks AG
+ * Copyright (C) 2008, 2009 Steffen Klassert <steffen.klassert@secunet.com>
+ *
+ * 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, write to the Free Software Foundation, Inc.,
+ * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA.
+ */
+
+#include <linux/module.h>
+#include <linux/cpumask.h>
+#include <linux/err.h>
+#include <linux/cpu.h>
+#include <linux/padata.h>
+#include <linux/mutex.h>
+#include <linux/sched.h>
+#include <linux/rcupdate.h>
+
+#define MAX_SEQ_NR INT_MAX - NR_CPUS
+#define MAX_OBJ_NUM 10000 * NR_CPUS
+
+static int padata_index_to_cpu(struct parallel_data *pd, int cpu_index)
+{
+	int cpu, target_cpu;
+
+	target_cpu = cpumask_first(pd->cpumask);
+	for (cpu = 0; cpu < cpu_index; cpu++)
+		target_cpu = cpumask_next(target_cpu, pd->cpumask);
+
+	return target_cpu;
+}
+
+static int padata_cpu_hash(struct padata_priv *padata)
+{
+	int cpu_index;
+	struct parallel_data *pd;
+
+	pd =  padata->pd;
+
+	/*
+	 * Hash the sequence numbers to the cpus by taking
+	 * seq_nr mod. number of cpus in use.
+	 */
+	cpu_index =  padata->seq_nr % cpumask_weight(pd->cpumask);
+
+	return padata_index_to_cpu(pd, cpu_index);
+}
+
+static void padata_parallel_worker(struct work_struct *work)
+{
+	struct padata_queue *queue;
+	struct parallel_data *pd;
+	struct padata_instance *pinst;
+	LIST_HEAD(local_list);
+
+	local_bh_disable();
+	queue = container_of(work, struct padata_queue, pwork);
+	pd = queue->pd;
+	pinst = pd->pinst;
+
+	spin_lock(&queue->parallel.lock);
+	list_replace_init(&queue->parallel.list, &local_list);
+	spin_unlock(&queue->parallel.lock);
+
+	while (!list_empty(&local_list)) {
+		struct padata_priv *padata;
+
+		padata = list_entry(local_list.next,
+				    struct padata_priv, list);
+
+		list_del_init(&padata->list);
+
+		padata->parallel(padata);
+	}
+
+	local_bh_enable();
+}
+
+/*
+ * padata_do_parallel - padata parallelization function
+ *
+ * @pinst: padata instance
+ * @padata: object to be parallelized
+ * @cb_cpu: cpu the serialization callback function will run on,
+ *          must be in the cpumask of padata.
+ *
+ * The parallelization callback function will run with BHs off.
+ * Note: Every object which is parallelized by padata_do_parallel
+ * must be seen by padata_do_serial.
+ */
+int padata_do_parallel(struct padata_instance *pinst,
+		       struct padata_priv *padata, int cb_cpu)
+{
+	int target_cpu, err;
+	struct padata_queue *queue;
+	struct parallel_data *pd;
+
+	rcu_read_lock_bh();
+
+	pd = rcu_dereference(pinst->pd);
+
+	err = 0;
+	if (!(pinst->flags & PADATA_INIT))
+		goto out;
+
+	err =  -EBUSY;
+	if ((pinst->flags & PADATA_RESET))
+		goto out;
+
+	if (atomic_read(&pd->refcnt) >= MAX_OBJ_NUM)
+		goto out;
+
+	err = -EINVAL;
+	if (!cpumask_test_cpu(cb_cpu, pd->cpumask))
+		goto out;
+
+	err = -EINPROGRESS;
+	atomic_inc(&pd->refcnt);
+	padata->pd = pd;
+	padata->cb_cpu = cb_cpu;
+
+	if (unlikely(atomic_read(&pd->seq_nr) == pd->max_seq_nr))
+		atomic_set(&pd->seq_nr, -1);
+
+	padata->seq_nr = atomic_inc_return(&pd->seq_nr);
+
+	target_cpu = padata_cpu_hash(padata);
+	queue = per_cpu_ptr(pd->queue, target_cpu);
+
+	spin_lock(&queue->parallel.lock);
+	list_add_tail(&padata->list, &queue->parallel.list);
+	spin_unlock(&queue->parallel.lock);
+
+	queue_work_on(target_cpu, pinst->wq, &queue->pwork);
+
+out:
+	rcu_read_unlock_bh();
+
+	return err;
+}
+EXPORT_SYMBOL(padata_do_parallel);
+
+static struct padata_priv *padata_get_next(struct parallel_data *pd)
+{
+	int cpu, num_cpus, empty, calc_seq_nr;
+	int seq_nr, next_nr, overrun, next_overrun;
+	struct padata_queue *queue, *next_queue;
+	struct padata_priv *padata;
+	struct padata_list *reorder;
+
+	empty = 0;
+	next_nr = -1;
+	next_overrun = 0;
+	next_queue = NULL;
+
+	num_cpus = cpumask_weight(pd->cpumask);
+
+	for_each_cpu(cpu, pd->cpumask) {
+		queue = per_cpu_ptr(pd->queue, cpu);
+		reorder = &queue->reorder;
+
+		/*
+		 * Calculate the seq_nr of the object that should be
+		 * next in this queue.
+		 */
+		overrun = 0;
+		calc_seq_nr = (atomic_read(&queue->num_obj) * num_cpus)
+			       + queue->cpu_index;
+
+		if (unlikely(calc_seq_nr > pd->max_seq_nr)) {
+			calc_seq_nr = calc_seq_nr - pd->max_seq_nr - 1;
+			overrun = 1;
+		}
+
+		if (!list_empty(&reorder->list)) {
+			padata = list_entry(reorder->list.next,
+					    struct padata_priv, list);
+
+			seq_nr  = padata->seq_nr;
+			BUG_ON(calc_seq_nr != seq_nr);
+		} else {
+			seq_nr = calc_seq_nr;
+			empty++;
+		}
+
+		if (next_nr < 0 || seq_nr < next_nr
+		    || (next_overrun && !overrun)) {
+			next_nr = seq_nr;
+			next_overrun = overrun;
+			next_queue = queue;
+		}
+	}
+
+	padata = NULL;
+
+	if (empty == num_cpus)
+		goto out;
+
+	reorder = &next_queue->reorder;
+
+	if (!list_empty(&reorder->list)) {
+		padata = list_entry(reorder->list.next,
+				    struct padata_priv, list);
+
+		if (unlikely(next_overrun)) {
+			for_each_cpu(cpu, pd->cpumask) {
+				queue = per_cpu_ptr(pd->queue, cpu);
+				atomic_set(&queue->num_obj, 0);
+			}
+		}
+
+		spin_lock(&reorder->lock);
+		list_del_init(&padata->list);
+		atomic_dec(&pd->reorder_objects);
+		spin_unlock(&reorder->lock);
+
+		atomic_inc(&next_queue->num_obj);
+
+		goto out;
+	}
+
+	if (next_nr % num_cpus == next_queue->cpu_index) {
+		padata = ERR_PTR(-ENODATA);
+		goto out;
+	}
+
+	padata = ERR_PTR(-EINPROGRESS);
+out:
+	return padata;
+}
+
+static void padata_reorder(struct parallel_data *pd)
+{
+	struct padata_priv *padata;
+	struct padata_queue *queue;
+	struct padata_instance *pinst = pd->pinst;
+
+try_again:
+	if (!spin_trylock_bh(&pd->lock))
+		goto out;
+
+	while (1) {
+		padata = padata_get_next(pd);
+
+		if (!padata || PTR_ERR(padata) == -EINPROGRESS)
+			break;
+
+		if (PTR_ERR(padata) == -ENODATA) {
+			spin_unlock_bh(&pd->lock);
+			goto out;
+		}
+
+		queue = per_cpu_ptr(pd->queue, padata->cb_cpu);
+
+		spin_lock(&queue->serial.lock);
+		list_add_tail(&padata->list, &queue->serial.list);
+		spin_unlock(&queue->serial.lock);
+
+		queue_work_on(padata->cb_cpu, pinst->wq, &queue->swork);
+	}
+
+	spin_unlock_bh(&pd->lock);
+
+	if (atomic_read(&pd->reorder_objects))
+		goto try_again;
+
+out:
+	return;
+}
+
+static void padata_serial_worker(struct work_struct *work)
+{
+	struct padata_queue *queue;
+	struct parallel_data *pd;
+	LIST_HEAD(local_list);
+
+	local_bh_disable();
+	queue = container_of(work, struct padata_queue, swork);
+	pd = queue->pd;
+
+	spin_lock(&queue->serial.lock);
+	list_replace_init(&queue->serial.list, &local_list);
+	spin_unlock(&queue->serial.lock);
+
+	while (!list_empty(&local_list)) {
+		struct padata_priv *padata;
+
+		padata = list_entry(local_list.next,
+				    struct padata_priv, list);
+
+		list_del_init(&padata->list);
+
+		padata->serial(padata);
+		atomic_dec(&pd->refcnt);
+	}
+	local_bh_enable();
+}
+
+/*
+ * padata_do_serial - padata serialization function
+ *
+ * @padata: object to be serialized.
+ *
+ * padata_do_serial must be called for every parallelized object.
+ * The serialization callback function will run with BHs off.
+ */
+void padata_do_serial(struct padata_priv *padata)
+{
+	int cpu;
+	struct padata_queue *queue;
+	struct parallel_data *pd;
+
+	pd = padata->pd;
+
+	cpu = get_cpu();
+	queue = per_cpu_ptr(pd->queue, cpu);
+
+	spin_lock(&queue->reorder.lock);
+	atomic_inc(&pd->reorder_objects);
+	list_add_tail(&padata->list, &queue->reorder.list);
+	spin_unlock(&queue->reorder.lock);
+
+	put_cpu();
+
+	padata_reorder(pd);
+}
+EXPORT_SYMBOL(padata_do_serial);
+
+static struct parallel_data *padata_alloc_pd(struct padata_instance *pinst,
+					     const struct cpumask *cpumask)
+{
+	int cpu, cpu_index, num_cpus;
+	struct padata_queue *queue;
+	struct parallel_data *pd;
+
+	cpu_index = 0;
+
+	pd = kzalloc(sizeof(struct parallel_data), GFP_KERNEL);
+	if (!pd)
+		goto err;
+
+	pd->queue = alloc_percpu(struct padata_queue);
+	if (!pd->queue)
+		goto err_free_pd;
+
+	if (!alloc_cpumask_var(&pd->cpumask, GFP_KERNEL))
+		goto err_free_queue;
+
+	for_each_possible_cpu(cpu) {
+		queue = per_cpu_ptr(pd->queue, cpu);
+
+		queue->pd = pd;
+
+		if (cpumask_test_cpu(cpu, cpumask)
+		    && cpumask_test_cpu(cpu, cpu_active_mask)) {
+			queue->cpu_index = cpu_index;
+			cpu_index++;
+		} else
+			queue->cpu_index = -1;
+
+		INIT_LIST_HEAD(&queue->reorder.list);
+		INIT_LIST_HEAD(&queue->parallel.list);
+		INIT_LIST_HEAD(&queue->serial.list);
+		spin_lock_init(&queue->reorder.lock);
+		spin_lock_init(&queue->parallel.lock);
+		spin_lock_init(&queue->serial.lock);
+
+		INIT_WORK(&queue->pwork, padata_parallel_worker);
+		INIT_WORK(&queue->swork, padata_serial_worker);
+		atomic_set(&queue->num_obj, 0);
+	}
+
+	cpumask_and(pd->cpumask, cpumask, cpu_active_mask);
+
+	num_cpus = cpumask_weight(pd->cpumask);
+	pd->max_seq_nr = (MAX_SEQ_NR / num_cpus) * num_cpus - 1;
+
+	atomic_set(&pd->seq_nr, -1);
+	atomic_set(&pd->reorder_objects, 0);
+	atomic_set(&pd->refcnt, 0);
+	pd->pinst = pinst;
+	spin_lock_init(&pd->lock);
+
+	return pd;
+
+err_free_queue:
+	free_percpu(pd->queue);
+err_free_pd:
+	kfree(pd);
+err:
+	return NULL;
+}
+
+static void padata_free_pd(struct parallel_data *pd)
+{
+	free_cpumask_var(pd->cpumask);
+	free_percpu(pd->queue);
+	kfree(pd);
+}
+
+static void padata_replace(struct padata_instance *pinst,
+			   struct parallel_data *pd_new)
+{
+	struct parallel_data *pd_old = pinst->pd;
+
+	pinst->flags |= PADATA_RESET;
+
+	rcu_assign_pointer(pinst->pd, pd_new);
+
+	synchronize_rcu();
+
+	while (atomic_read(&pd_old->refcnt) != 0)
+		yield();
+
+	flush_workqueue(pinst->wq);
+
+	padata_free_pd(pd_old);
+
+	pinst->flags &= ~PADATA_RESET;
+}
+
+/*
+ * padata_set_cpumask - set the cpumask that padata should use
+ *
+ * @pinst: padata instance
+ * @cpumask: the cpumask to use
+ */
+int padata_set_cpumask(struct padata_instance *pinst,
+			cpumask_var_t cpumask)
+{
+	struct parallel_data *pd;
+	int err = 0;
+
+	might_sleep();
+
+	mutex_lock(&pinst->lock);
+
+	pd = padata_alloc_pd(pinst, cpumask);
+	if (!pd) {
+		err = -ENOMEM;
+		goto out;
+	}
+
+	cpumask_copy(pinst->cpumask, cpumask);
+
+	padata_replace(pinst, pd);
+
+out:
+	mutex_unlock(&pinst->lock);
+
+	return err;
+}
+EXPORT_SYMBOL(padata_set_cpumask);
+
+static int __padata_add_cpu(struct padata_instance *pinst, int cpu)
+{
+	struct parallel_data *pd;
+
+	if (cpumask_test_cpu(cpu, cpu_active_mask)) {
+		pd = padata_alloc_pd(pinst, pinst->cpumask);
+		if (!pd)
+			return -ENOMEM;
+
+		padata_replace(pinst, pd);
+	}
+
+	return 0;
+}
+
+/*
+ * padata_add_cpu - add a cpu to the padata cpumask
+ *
+ * @pinst: padata instance
+ * @cpu: cpu to add
+ */
+int padata_add_cpu(struct padata_instance *pinst, int cpu)
+{
+	int err;
+
+	might_sleep();
+
+	mutex_lock(&pinst->lock);
+
+	cpumask_set_cpu(cpu, pinst->cpumask);
+	err = __padata_add_cpu(pinst, cpu);
+
+	mutex_unlock(&pinst->lock);
+
+	return err;
+}
+EXPORT_SYMBOL(padata_add_cpu);
+
+static int __padata_remove_cpu(struct padata_instance *pinst, int cpu)
+{
+	struct parallel_data *pd;
+
+	if (cpumask_test_cpu(cpu, cpu_online_mask)) {
+		pd = padata_alloc_pd(pinst, pinst->cpumask);
+		if (!pd)
+			return -ENOMEM;
+
+		padata_replace(pinst, pd);
+	}
+
+	return 0;
+}
+
+/*
+ * padata_remove_cpu - remove a cpu from the padata cpumask
+ *
+ * @pinst: padata instance
+ * @cpu: cpu to remove
+ */
+int padata_remove_cpu(struct padata_instance *pinst, int cpu)
+{
+	int err;
+
+	might_sleep();
+
+	mutex_lock(&pinst->lock);
+
+	cpumask_clear_cpu(cpu, pinst->cpumask);
+	err = __padata_remove_cpu(pinst, cpu);
+
+	mutex_unlock(&pinst->lock);
+
+	return err;
+}
+EXPORT_SYMBOL(padata_remove_cpu);
+
+/*
+ * padata_start - start the parallel processing
+ *
+ * @pinst: padata instance to start
+ */
+void padata_start(struct padata_instance *pinst)
+{
+	might_sleep();
+
+	mutex_lock(&pinst->lock);
+	pinst->flags |= PADATA_INIT;
+	mutex_unlock(&pinst->lock);
+}
+EXPORT_SYMBOL(padata_start);
+
+/*
+ * padata_stop - stop the parallel processing
+ *
+ * @pinst: padata instance to stop
+ */
+void padata_stop(struct padata_instance *pinst)
+{
+	might_sleep();
+
+	mutex_lock(&pinst->lock);
+	pinst->flags &= ~PADATA_INIT;
+	mutex_unlock(&pinst->lock);
+}
+EXPORT_SYMBOL(padata_stop);
+
+static int __cpuinit padata_cpu_callback(struct notifier_block *nfb,
+					 unsigned long action, void *hcpu)
+{
+	int err;
+	struct padata_instance *pinst;
+	int cpu = (unsigned long)hcpu;
+
+	pinst = container_of(nfb, struct padata_instance, cpu_notifier);
+
+	switch (action) {
+	case CPU_ONLINE:
+	case CPU_ONLINE_FROZEN:
+		if (!cpumask_test_cpu(cpu, pinst->cpumask))
+			break;
+		mutex_lock(&pinst->lock);
+		err = __padata_add_cpu(pinst, cpu);
+		mutex_unlock(&pinst->lock);
+		if (err)
+			return NOTIFY_BAD;
+		break;
+
+	case CPU_DOWN_PREPARE:
+	case CPU_DOWN_PREPARE_FROZEN:
+		if (!cpumask_test_cpu(cpu, pinst->cpumask))
+			break;
+		mutex_lock(&pinst->lock);
+		err = __padata_remove_cpu(pinst, cpu);
+		mutex_unlock(&pinst->lock);
+		if (err)
+			return NOTIFY_BAD;
+		break;
+
+	case CPU_UP_CANCELED:
+	case CPU_UP_CANCELED_FROZEN:
+		if (!cpumask_test_cpu(cpu, pinst->cpumask))
+			break;
+		mutex_lock(&pinst->lock);
+		__padata_remove_cpu(pinst, cpu);
+		mutex_unlock(&pinst->lock);
+
+	case CPU_DOWN_FAILED:
+	case CPU_DOWN_FAILED_FROZEN:
+		if (!cpumask_test_cpu(cpu, pinst->cpumask))
+			break;
+		mutex_lock(&pinst->lock);
+		__padata_add_cpu(pinst, cpu);
+		mutex_unlock(&pinst->lock);
+	}
+
+	return NOTIFY_OK;
+}
+
+/*
+ * padata_alloc - allocate and initialize a padata instance
+ *
+ * @cpumask: cpumask that padata uses for parallelization
+ * @wq: workqueue to use for the allocated padata instance
+ */
+struct padata_instance *padata_alloc(const struct cpumask *cpumask,
+				     struct workqueue_struct *wq)
+{
+	int err;
+	struct padata_instance *pinst;
+	struct parallel_data *pd;
+
+	pinst = kzalloc(sizeof(struct padata_instance), GFP_KERNEL);
+	if (!pinst)
+		goto err;
+
+	pd = padata_alloc_pd(pinst, cpumask);
+	if (!pd)
+		goto err_free_inst;
+
+	rcu_assign_pointer(pinst->pd, pd);
+
+	pinst->wq = wq;
+
+	cpumask_copy(pinst->cpumask, cpumask);
+
+	pinst->flags = 0;
+
+	pinst->cpu_notifier.notifier_call = padata_cpu_callback;
+	pinst->cpu_notifier.priority = 0;
+	err = register_hotcpu_notifier(&pinst->cpu_notifier);
+	if (err)
+		goto err_free_pd;
+
+	mutex_init(&pinst->lock);
+
+	return pinst;
+
+err_free_pd:
+	padata_free_pd(pd);
+err_free_inst:
+	kfree(pinst);
+err:
+	return NULL;
+}
+EXPORT_SYMBOL(padata_alloc);
+
+/*
+ * padata_free - free a padata instance
+ *
+ * @ padata_inst: padata instance to free
+ */
+void padata_free(struct padata_instance *pinst)
+{
+	padata_stop(pinst);
+
+	synchronize_rcu();
+
+	while (atomic_read(&pinst->pd->refcnt) != 0)
+		yield();
+
+	unregister_hotcpu_notifier(&pinst->cpu_notifier);
+	padata_free_pd(pinst->pd);
+	kfree(pinst);
+}
+EXPORT_SYMBOL(padata_free);
diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig
index 91e09d3b2eb2..5c36ea9d55d2 100644
--- a/kernel/power/Kconfig
+++ b/kernel/power/Kconfig
@@ -27,6 +27,15 @@ config PM_DEBUG
 	code. This is helpful when debugging and reporting PM bugs, like
 	suspend support.
 
+config PM_ADVANCED_DEBUG
+	bool "Extra PM attributes in sysfs for low-level debugging/testing"
+	depends on PM_DEBUG
+	default n
+	---help---
+	Add extra sysfs attributes allowing one to access some Power Management
+	fields of device objects from user space.  If you are not a kernel
+	developer interested in debugging/testing Power Management, say "no".
+
 config PM_VERBOSE
 	bool "Verbose Power Management debugging"
 	depends on PM_DEBUG
@@ -85,6 +94,11 @@ config PM_SLEEP
 	depends on SUSPEND || HIBERNATION || XEN_SAVE_RESTORE
 	default y
 
+config PM_SLEEP_ADVANCED_DEBUG
+	bool
+	depends on PM_ADVANCED_DEBUG
+	default n
+
 config SUSPEND
 	bool "Suspend to RAM and standby"
 	depends on PM && ARCH_SUSPEND_POSSIBLE
@@ -222,3 +236,8 @@ config PM_RUNTIME
 	  and the bus type drivers of the buses the devices are on are
 	  responsible for the actual handling of the autosuspend requests and
 	  wake-up events.
+
+config PM_OPS
+	bool
+	depends on PM_SLEEP || PM_RUNTIME
+	default y
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 0998c7139053..b58800b21fc0 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -44,6 +44,32 @@ int pm_notifier_call_chain(unsigned long val)
 			== NOTIFY_BAD) ? -EINVAL : 0;
 }
 
+/* If set, devices may be suspended and resumed asynchronously. */
+int pm_async_enabled = 1;
+
+static ssize_t pm_async_show(struct kobject *kobj, struct kobj_attribute *attr,
+			     char *buf)
+{
+	return sprintf(buf, "%d\n", pm_async_enabled);
+}
+
+static ssize_t pm_async_store(struct kobject *kobj, struct kobj_attribute *attr,
+			      const char *buf, size_t n)
+{
+	unsigned long val;
+
+	if (strict_strtoul(buf, 10, &val))
+		return -EINVAL;
+
+	if (val > 1)
+		return -EINVAL;
+
+	pm_async_enabled = val;
+	return n;
+}
+
+power_attr(pm_async);
+
 #ifdef CONFIG_PM_DEBUG
 int pm_test_level = TEST_NONE;
 
@@ -208,9 +234,12 @@ static struct attribute * g[] = {
 #ifdef CONFIG_PM_TRACE
 	&pm_trace_attr.attr,
 #endif
-#if defined(CONFIG_PM_SLEEP) && defined(CONFIG_PM_DEBUG)
+#ifdef CONFIG_PM_SLEEP
+	&pm_async_attr.attr,
+#ifdef CONFIG_PM_DEBUG
 	&pm_test_attr.attr,
 #endif
+#endif
 	NULL,
 };
 
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 36cb168e4330..830cadecbdfc 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -1181,7 +1181,7 @@ static void free_unnecessary_pages(void)
 
 	memory_bm_position_reset(&copy_bm);
 
-	while (to_free_normal > 0 && to_free_highmem > 0) {
+	while (to_free_normal > 0 || to_free_highmem > 0) {
 		unsigned long pfn = memory_bm_next_pfn(&copy_bm);
 		struct page *page = pfn_to_page(pfn);
 
@@ -1500,7 +1500,7 @@ asmlinkage int swsusp_save(void)
 {
 	unsigned int nr_pages, nr_highmem;
 
-	printk(KERN_INFO "PM: Creating hibernation image: \n");
+	printk(KERN_INFO "PM: Creating hibernation image:\n");
 
 	drain_local_pages(NULL);
 	nr_pages = count_data_pages();
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
index 09b2b0ae9e9d..1d575733d4e1 100644
--- a/kernel/power/swap.c
+++ b/kernel/power/swap.c
@@ -657,10 +657,6 @@ int swsusp_read(unsigned int *flags_p)
 	struct swsusp_info *header;
 
 	*flags_p = swsusp_header->flags;
-	if (IS_ERR(resume_bdev)) {
-		pr_debug("PM: Image device not initialised\n");
-		return PTR_ERR(resume_bdev);
-	}
 
 	memset(&snapshot, 0, sizeof(struct snapshot_handle));
 	error = snapshot_write_next(&snapshot, PAGE_SIZE);
diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c
deleted file mode 100644
index 5b3601bd1893..000000000000
--- a/kernel/power/swsusp.c
+++ /dev/null
@@ -1,58 +0,0 @@
-/*
- * linux/kernel/power/swsusp.c
- *
- * This file provides code to write suspend image to swap and read it back.
- *
- * Copyright (C) 1998-2001 Gabor Kuti <seasons@fornax.hu>
- * Copyright (C) 1998,2001-2005 Pavel Machek <pavel@suse.cz>
- *
- * This file is released under the GPLv2.
- *
- * I'd like to thank the following people for their work:
- *
- * Pavel Machek <pavel@ucw.cz>:
- * Modifications, defectiveness pointing, being with me at the very beginning,
- * suspend to swap space, stop all tasks. Port to 2.4.18-ac and 2.5.17.
- *
- * Steve Doddi <dirk@loth.demon.co.uk>:
- * Support the possibility of hardware state restoring.
- *
- * Raph <grey.havens@earthling.net>:
- * Support for preserving states of network devices and virtual console
- * (including X and svgatextmode)
- *
- * Kurt Garloff <garloff@suse.de>:
- * Straightened the critical function in order to prevent compilers from
- * playing tricks with local variables.
- *
- * Andreas Mohr <a.mohr@mailto.de>
- *
- * Alex Badea <vampire@go.ro>:
- * Fixed runaway init
- *
- * Rafael J. Wysocki <rjw@sisk.pl>
- * Reworked the freeing of memory and the handling of swap
- *
- * More state savers are welcome. Especially for the scsi layer...
- *
- * For TODOs,FIXMEs also look in Documentation/power/swsusp.txt
- */
-
-#include <linux/mm.h>
-#include <linux/suspend.h>
-#include <linux/spinlock.h>
-#include <linux/kernel.h>
-#include <linux/major.h>
-#include <linux/swap.h>
-#include <linux/pm.h>
-#include <linux/swapops.h>
-#include <linux/bootmem.h>
-#include <linux/syscalls.h>
-#include <linux/highmem.h>
-#include <linux/time.h>
-#include <linux/rbtree.h>
-#include <linux/io.h>
-
-#include "power.h"
-
-int in_suspend __nosavedata = 0;
diff --git a/kernel/power/user.c b/kernel/power/user.c
index bf0014d6a5f0..4d2289626a84 100644
--- a/kernel/power/user.c
+++ b/kernel/power/user.c
@@ -195,6 +195,15 @@ static ssize_t snapshot_write(struct file *filp, const char __user *buf,
 	return res;
 }
 
+static void snapshot_deprecated_ioctl(unsigned int cmd)
+{
+	if (printk_ratelimit())
+		printk(KERN_NOTICE "%pf: ioctl '%.8x' is deprecated and will "
+				"be removed soon, update your suspend-to-disk "
+				"utilities\n",
+				__builtin_return_address(0), cmd);
+}
+
 static long snapshot_ioctl(struct file *filp, unsigned int cmd,
 							unsigned long arg)
 {
@@ -246,8 +255,9 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd,
 		data->frozen = 0;
 		break;
 
-	case SNAPSHOT_CREATE_IMAGE:
 	case SNAPSHOT_ATOMIC_SNAPSHOT:
+		snapshot_deprecated_ioctl(cmd);
+	case SNAPSHOT_CREATE_IMAGE:
 		if (data->mode != O_RDONLY || !data->frozen  || data->ready) {
 			error = -EPERM;
 			break;
@@ -275,8 +285,9 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd,
 		data->ready = 0;
 		break;
 
-	case SNAPSHOT_PREF_IMAGE_SIZE:
 	case SNAPSHOT_SET_IMAGE_SIZE:
+		snapshot_deprecated_ioctl(cmd);
+	case SNAPSHOT_PREF_IMAGE_SIZE:
 		image_size = arg;
 		break;
 
@@ -290,15 +301,17 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd,
 		error = put_user(size, (loff_t __user *)arg);
 		break;
 
-	case SNAPSHOT_AVAIL_SWAP_SIZE:
 	case SNAPSHOT_AVAIL_SWAP:
+		snapshot_deprecated_ioctl(cmd);
+	case SNAPSHOT_AVAIL_SWAP_SIZE:
 		size = count_swap_pages(data->swap, 1);
 		size <<= PAGE_SHIFT;
 		error = put_user(size, (loff_t __user *)arg);
 		break;
 
-	case SNAPSHOT_ALLOC_SWAP_PAGE:
 	case SNAPSHOT_GET_SWAP_PAGE:
+		snapshot_deprecated_ioctl(cmd);
+	case SNAPSHOT_ALLOC_SWAP_PAGE:
 		if (data->swap < 0 || data->swap >= MAX_SWAPFILES) {
 			error = -ENODEV;
 			break;
@@ -321,6 +334,7 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd,
 		break;
 
 	case SNAPSHOT_SET_SWAP_FILE: /* This ioctl is deprecated */
+		snapshot_deprecated_ioctl(cmd);
 		if (!swsusp_swap_in_use()) {
 			/*
 			 * User space encodes device types as two-byte values,
@@ -362,6 +376,7 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd,
 		break;
 
 	case SNAPSHOT_PMOPS: /* This ioctl is deprecated */
+		snapshot_deprecated_ioctl(cmd);
 		error = -EINVAL;
 
 		switch (arg) {
diff --git a/kernel/resource.c b/kernel/resource.c
index af96c1e4b54b..24e9e60c1459 100644
--- a/kernel/resource.c
+++ b/kernel/resource.c
@@ -188,6 +188,36 @@ static int __release_resource(struct resource *old)
 	return -EINVAL;
 }
 
+static void __release_child_resources(struct resource *r)
+{
+	struct resource *tmp, *p;
+	resource_size_t size;
+
+	p = r->child;
+	r->child = NULL;
+	while (p) {
+		tmp = p;
+		p = p->sibling;
+
+		tmp->parent = NULL;
+		tmp->sibling = NULL;
+		__release_child_resources(tmp);
+
+		printk(KERN_DEBUG "release child resource %pR\n", tmp);
+		/* need to restore size, and keep flags */
+		size = resource_size(tmp);
+		tmp->start = 0;
+		tmp->end = size - 1;
+	}
+}
+
+void release_child_resources(struct resource *r)
+{
+	write_lock(&resource_lock);
+	__release_child_resources(r);
+	write_unlock(&resource_lock);
+}
+
 /**
  * request_resource - request and reserve an I/O or memory resource
  * @root: root resource descriptor
@@ -303,8 +333,10 @@ int walk_system_ram_range(unsigned long start_pfn, unsigned long nr_pages,
 static int find_resource(struct resource *root, struct resource *new,
 			 resource_size_t size, resource_size_t min,
 			 resource_size_t max, resource_size_t align,
-			 void (*alignf)(void *, struct resource *,
-					resource_size_t, resource_size_t),
+			 resource_size_t (*alignf)(void *,
+						   const struct resource *,
+						   resource_size_t,
+						   resource_size_t),
 			 void *alignf_data)
 {
 	struct resource *this = root->child;
@@ -330,7 +362,7 @@ static int find_resource(struct resource *root, struct resource *new,
 			tmp.end = max;
 		tmp.start = ALIGN(tmp.start, align);
 		if (alignf)
-			alignf(alignf_data, &tmp, size, align);
+			tmp.start = alignf(alignf_data, &tmp, size, align);
 		if (tmp.start < tmp.end && tmp.end - tmp.start >= size - 1) {
 			new->start = tmp.start;
 			new->end = tmp.start + size - 1;
@@ -358,8 +390,10 @@ static int find_resource(struct resource *root, struct resource *new,
 int allocate_resource(struct resource *root, struct resource *new,
 		      resource_size_t size, resource_size_t min,
 		      resource_size_t max, resource_size_t align,
-		      void (*alignf)(void *, struct resource *,
-				     resource_size_t, resource_size_t),
+		      resource_size_t (*alignf)(void *,
+						const struct resource *,
+						resource_size_t,
+						resource_size_t),
 		      void *alignf_data)
 {
 	int err;
diff --git a/kernel/smp.c b/kernel/smp.c
index f10408422444..9867b6bfefce 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -12,8 +12,6 @@
 #include <linux/smp.h>
 #include <linux/cpu.h>
 
-static DEFINE_PER_CPU(struct call_single_queue, call_single_queue);
-
 static struct {
 	struct list_head	queue;
 	raw_spinlock_t		lock;
@@ -33,12 +31,14 @@ struct call_function_data {
 	cpumask_var_t		cpumask;
 };
 
+static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_function_data, cfd_data);
+
 struct call_single_queue {
 	struct list_head	list;
 	raw_spinlock_t		lock;
 };
 
-static DEFINE_PER_CPU(struct call_function_data, cfd_data);
+static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_single_queue, call_single_queue);
 
 static int
 hotplug_cfd(struct notifier_block *nfb, unsigned long action, void *hcpu)
@@ -256,7 +256,7 @@ void generic_smp_call_function_single_interrupt(void)
 	}
 }
 
-static DEFINE_PER_CPU(struct call_single_data, csd_data);
+static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_single_data, csd_data);
 
 /*
  * smp_call_function_single - Run a function on a specific CPU