Merge branch 'linux-2.6'

57 files changed, 6354 insertions, 1833 deletions

diff --git a/kernel/Kconfig.hz b/kernel/Kconfig.hz
index 4af15802ccd..526128a2e62 100644
--- a/kernel/Kconfig.hz
+++ b/kernel/Kconfig.hz
@@ -54,3 +54,5 @@ config HZ
 	default 300 if HZ_300
 	default 1000 if HZ_1000
 
+config SCHED_HRTICK
+	def_bool HIGH_RES_TIMERS && X86
diff --git a/kernel/Kconfig.preempt b/kernel/Kconfig.preempt
index c64ce9c1420..0669b70fa6a 100644
--- a/kernel/Kconfig.preempt
+++ b/kernel/Kconfig.preempt
@@ -52,14 +52,13 @@ config PREEMPT
 
 endchoice
 
-config PREEMPT_BKL
-	bool "Preempt The Big Kernel Lock"
-	depends on SMP || PREEMPT
+config RCU_TRACE
+	bool "Enable tracing for RCU - currently stats in debugfs"
+	select DEBUG_FS
 	default y
 	help
-	  This option reduces the latency of the kernel by making the
-	  big kernel lock preemptible.
+	  This option provides tracing in RCU which presents stats
+	  in debugfs for debugging RCU implementation.
 
-	  Say Y here if you are building a kernel for a desktop system.
+	  Say Y here if you want to enable RCU tracing
 	  Say N if you are unsure.
-
diff --git a/kernel/Makefile b/kernel/Makefile
index dfa96956dae..8885627ea02 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -36,6 +36,7 @@ obj-$(CONFIG_KALLSYMS) += kallsyms.o
 obj-$(CONFIG_PM) += power/
 obj-$(CONFIG_BSD_PROCESS_ACCT) += acct.o
 obj-$(CONFIG_KEXEC) += kexec.o
+obj-$(CONFIG_BACKTRACE_SELF_TEST) += backtracetest.o
 obj-$(CONFIG_COMPAT) += compat.o
 obj-$(CONFIG_CGROUPS) += cgroup.o
 obj-$(CONFIG_CGROUP_DEBUG) += cgroup_debug.o
@@ -43,6 +44,7 @@ obj-$(CONFIG_CPUSETS) += cpuset.o
 obj-$(CONFIG_CGROUP_NS) += ns_cgroup.o
 obj-$(CONFIG_IKCONFIG) += configs.o
 obj-$(CONFIG_STOP_MACHINE) += stop_machine.o
+obj-$(CONFIG_KPROBES_SANITY_TEST) += test_kprobes.o
 obj-$(CONFIG_AUDIT) += audit.o auditfilter.o
 obj-$(CONFIG_AUDITSYSCALL) += auditsc.o
 obj-$(CONFIG_AUDIT_TREE) += audit_tree.o
@@ -52,11 +54,17 @@ obj-$(CONFIG_DETECT_SOFTLOCKUP) += softlockup.o
 obj-$(CONFIG_GENERIC_HARDIRQS) += irq/
 obj-$(CONFIG_SECCOMP) += seccomp.o
 obj-$(CONFIG_RCU_TORTURE_TEST) += rcutorture.o
+obj-$(CONFIG_CLASSIC_RCU) += rcuclassic.o
+obj-$(CONFIG_PREEMPT_RCU) += rcupreempt.o
+ifeq ($(CONFIG_PREEMPT_RCU),y)
+obj-$(CONFIG_RCU_TRACE) += rcupreempt_trace.o
+endif
 obj-$(CONFIG_RELAY) += relay.o
 obj-$(CONFIG_SYSCTL) += utsname_sysctl.o
 obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o
 obj-$(CONFIG_TASKSTATS) += taskstats.o tsacct.o
 obj-$(CONFIG_MARKERS) += marker.o
+obj-$(CONFIG_LATENCYTOP) += latencytop.o
 
 ifneq ($(CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER),y)
 # According to Alan Modra <alan@linuxcare.com.au>, the -fno-omit-frame-pointer is
diff --git a/kernel/backtracetest.c b/kernel/backtracetest.c
new file mode 100644
index 00000000000..d1a7605c5b8
--- /dev/null
+++ b/kernel/backtracetest.c
@@ -0,0 +1,48 @@
+/*
+ * Simple stack backtrace regression test module
+ *
+ * (C) Copyright 2008 Intel Corporation
+ * Author: Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * 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; version 2
+ * of the License.
+ */
+
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/delay.h>
+
+static struct timer_list backtrace_timer;
+
+static void backtrace_test_timer(unsigned long data)
+{
+	printk("Testing a backtrace from irq context.\n");
+	printk("The following trace is a kernel self test and not a bug!\n");
+	dump_stack();
+}
+static int backtrace_regression_test(void)
+{
+	printk("====[ backtrace testing ]===========\n");
+	printk("Testing a backtrace from process context.\n");
+	printk("The following trace is a kernel self test and not a bug!\n");
+	dump_stack();
+
+	init_timer(&backtrace_timer);
+	backtrace_timer.function = backtrace_test_timer;
+	mod_timer(&backtrace_timer, jiffies + 10);
+
+	msleep(10);
+	printk("====[ end of backtrace testing ]====\n");
+	return 0;
+}
+
+static void exitf(void)
+{
+}
+
+module_init(backtrace_regression_test);
+module_exit(exitf);
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Arjan van de Ven <arjan@linux.intel.com>");
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 6b3a0c15144..e0d3a4f56ec 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -15,9 +15,8 @@
 #include <linux/stop_machine.h>
 #include <linux/mutex.h>
 
-/* This protects CPUs going up and down... */
+/* Serializes the updates to cpu_online_map, cpu_present_map */
 static DEFINE_MUTEX(cpu_add_remove_lock);
-static DEFINE_MUTEX(cpu_bitmask_lock);
 
 static __cpuinitdata RAW_NOTIFIER_HEAD(cpu_chain);
 
@@ -26,52 +25,123 @@ static __cpuinitdata RAW_NOTIFIER_HEAD(cpu_chain);
  */
 static int cpu_hotplug_disabled;
 
-#ifdef CONFIG_HOTPLUG_CPU
+static struct {
+	struct task_struct *active_writer;
+	struct mutex lock; /* Synchronizes accesses to refcount, */
+	/*
+	 * Also blocks the new readers during
+	 * an ongoing cpu hotplug operation.
+	 */
+	int refcount;
+	wait_queue_head_t writer_queue;
+} cpu_hotplug;
 
-/* Crappy recursive lock-takers in cpufreq! Complain loudly about idiots */
-static struct task_struct *recursive;
-static int recursive_depth;
+#define writer_exists() (cpu_hotplug.active_writer != NULL)
 
-void lock_cpu_hotplug(void)
+void __init cpu_hotplug_init(void)
 {
-	struct task_struct *tsk = current;
-
-	if (tsk == recursive) {
-		static int warnings = 10;
-		if (warnings) {
-			printk(KERN_ERR "Lukewarm IQ detected in hotplug locking\n");
-			WARN_ON(1);
-			warnings--;
-		}
-		recursive_depth++;
+	cpu_hotplug.active_writer = NULL;
+	mutex_init(&cpu_hotplug.lock);
+	cpu_hotplug.refcount = 0;
+	init_waitqueue_head(&cpu_hotplug.writer_queue);
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+
+void get_online_cpus(void)
+{
+	might_sleep();
+	if (cpu_hotplug.active_writer == current)
 		return;
-	}
-	mutex_lock(&cpu_bitmask_lock);
-	recursive = tsk;
+	mutex_lock(&cpu_hotplug.lock);
+	cpu_hotplug.refcount++;
+	mutex_unlock(&cpu_hotplug.lock);
+
 }
-EXPORT_SYMBOL_GPL(lock_cpu_hotplug);
+EXPORT_SYMBOL_GPL(get_online_cpus);
 
-void unlock_cpu_hotplug(void)
+void put_online_cpus(void)
 {
-	WARN_ON(recursive != current);
-	if (recursive_depth) {
-		recursive_depth--;
+	if (cpu_hotplug.active_writer == current)
 		return;
-	}
-	recursive = NULL;
-	mutex_unlock(&cpu_bitmask_lock);
+	mutex_lock(&cpu_hotplug.lock);
+	cpu_hotplug.refcount--;
+
+	if (unlikely(writer_exists()) && !cpu_hotplug.refcount)
+		wake_up(&cpu_hotplug.writer_queue);
+
+	mutex_unlock(&cpu_hotplug.lock);
+
 }
-EXPORT_SYMBOL_GPL(unlock_cpu_hotplug);
+EXPORT_SYMBOL_GPL(put_online_cpus);
 
 #endif	/* CONFIG_HOTPLUG_CPU */
 
+/*
+ * The following two API's must be used when attempting
+ * to serialize the updates to cpu_online_map, cpu_present_map.
+ */
+void cpu_maps_update_begin(void)
+{
+	mutex_lock(&cpu_add_remove_lock);
+}
+
+void cpu_maps_update_done(void)
+{
+	mutex_unlock(&cpu_add_remove_lock);
+}
+
+/*
+ * This ensures that the hotplug operation can begin only when the
+ * refcount goes to zero.
+ *
+ * Note that during a cpu-hotplug operation, the new readers, if any,
+ * will be blocked by the cpu_hotplug.lock
+ *
+ * Since cpu_maps_update_begin is always called after invoking
+ * cpu_maps_update_begin, we can be sure that only one writer is active.
+ *
+ * Note that theoretically, there is a possibility of a livelock:
+ * - Refcount goes to zero, last reader wakes up the sleeping
+ *   writer.
+ * - Last reader unlocks the cpu_hotplug.lock.
+ * - A new reader arrives at this moment, bumps up the refcount.
+ * - The writer acquires the cpu_hotplug.lock finds the refcount
+ *   non zero and goes to sleep again.
+ *
+ * However, this is very difficult to achieve in practice since
+ * get_online_cpus() not an api which is called all that often.
+ *
+ */
+static void cpu_hotplug_begin(void)
+{
+	DECLARE_WAITQUEUE(wait, current);
+
+	mutex_lock(&cpu_hotplug.lock);
+
+	cpu_hotplug.active_writer = current;
+	add_wait_queue_exclusive(&cpu_hotplug.writer_queue, &wait);
+	while (cpu_hotplug.refcount) {
+		set_current_state(TASK_UNINTERRUPTIBLE);
+		mutex_unlock(&cpu_hotplug.lock);
+		schedule();
+		mutex_lock(&cpu_hotplug.lock);
+	}
+	remove_wait_queue_locked(&cpu_hotplug.writer_queue, &wait);
+}
+
+static void cpu_hotplug_done(void)
+{
+	cpu_hotplug.active_writer = NULL;
+	mutex_unlock(&cpu_hotplug.lock);
+}
 /* Need to know about CPUs going up/down? */
 int __cpuinit register_cpu_notifier(struct notifier_block *nb)
 {
 	int ret;
-	mutex_lock(&cpu_add_remove_lock);
+	cpu_maps_update_begin();
 	ret = raw_notifier_chain_register(&cpu_chain, nb);
-	mutex_unlock(&cpu_add_remove_lock);
+	cpu_maps_update_done();
 	return ret;
 }
 
@@ -81,9 +151,9 @@ EXPORT_SYMBOL(register_cpu_notifier);
 
 void unregister_cpu_notifier(struct notifier_block *nb)
 {
-	mutex_lock(&cpu_add_remove_lock);
+	cpu_maps_update_begin();
 	raw_notifier_chain_unregister(&cpu_chain, nb);
-	mutex_unlock(&cpu_add_remove_lock);
+	cpu_maps_update_done();
 }
 EXPORT_SYMBOL(unregister_cpu_notifier);
 
@@ -147,7 +217,7 @@ static int _cpu_down(unsigned int cpu, int tasks_frozen)
 	if (!cpu_online(cpu))
 		return -EINVAL;
 
-	raw_notifier_call_chain(&cpu_chain, CPU_LOCK_ACQUIRE, hcpu);
+	cpu_hotplug_begin();
 	err = __raw_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE | mod,
 					hcpu, -1, &nr_calls);
 	if (err == NOTIFY_BAD) {
@@ -166,9 +236,7 @@ static int _cpu_down(unsigned int cpu, int tasks_frozen)
 	cpu_clear(cpu, tmp);
 	set_cpus_allowed(current, tmp);
 
-	mutex_lock(&cpu_bitmask_lock);
 	p = __stop_machine_run(take_cpu_down, &tcd_param, cpu);
-	mutex_unlock(&cpu_bitmask_lock);
 
 	if (IS_ERR(p) || cpu_online(cpu)) {
 		/* CPU didn't die: tell everyone.  Can't complain. */
@@ -202,7 +270,7 @@ out_thread:
 out_allowed:
 	set_cpus_allowed(current, old_allowed);
 out_release:
-	raw_notifier_call_chain(&cpu_chain, CPU_LOCK_RELEASE, hcpu);
+	cpu_hotplug_done();
 	return err;
 }
 
@@ -210,13 +278,13 @@ int cpu_down(unsigned int cpu)
 {
 	int err = 0;
 
-	mutex_lock(&cpu_add_remove_lock);
+	cpu_maps_update_begin();
 	if (cpu_hotplug_disabled)
 		err = -EBUSY;
 	else
 		err = _cpu_down(cpu, 0);
 
-	mutex_unlock(&cpu_add_remove_lock);
+	cpu_maps_update_done();
 	return err;
 }
 #endif /*CONFIG_HOTPLUG_CPU*/
@@ -231,7 +299,7 @@ static int __cpuinit _cpu_up(unsigned int cpu, int tasks_frozen)
 	if (cpu_online(cpu) || !cpu_present(cpu))
 		return -EINVAL;
 
-	raw_notifier_call_chain(&cpu_chain, CPU_LOCK_ACQUIRE, hcpu);
+	cpu_hotplug_begin();
 	ret = __raw_notifier_call_chain(&cpu_chain, CPU_UP_PREPARE | mod, hcpu,
 							-1, &nr_calls);
 	if (ret == NOTIFY_BAD) {
@@ -243,9 +311,7 @@ static int __cpuinit _cpu_up(unsigned int cpu, int tasks_frozen)
 	}
 
 	/* Arch-specific enabling code. */
-	mutex_lock(&cpu_bitmask_lock);
 	ret = __cpu_up(cpu);
-	mutex_unlock(&cpu_bitmask_lock);
 	if (ret != 0)
 		goto out_notify;
 	BUG_ON(!cpu_online(cpu));
@@ -257,7 +323,7 @@ out_notify:
 	if (ret != 0)
 		__raw_notifier_call_chain(&cpu_chain,
 				CPU_UP_CANCELED | mod, hcpu, nr_calls, NULL);
-	raw_notifier_call_chain(&cpu_chain, CPU_LOCK_RELEASE, hcpu);
+	cpu_hotplug_done();
 
 	return ret;
 }
@@ -275,13 +341,13 @@ int __cpuinit cpu_up(unsigned int cpu)
 		return -EINVAL;
 	}
 
-	mutex_lock(&cpu_add_remove_lock);
+	cpu_maps_update_begin();
 	if (cpu_hotplug_disabled)
 		err = -EBUSY;
 	else
 		err = _cpu_up(cpu, 0);
 
-	mutex_unlock(&cpu_add_remove_lock);
+	cpu_maps_update_done();
 	return err;
 }
 
@@ -292,7 +358,7 @@ int disable_nonboot_cpus(void)
 {
 	int cpu, first_cpu, error = 0;
 
-	mutex_lock(&cpu_add_remove_lock);
+	cpu_maps_update_begin();
 	first_cpu = first_cpu(cpu_online_map);
 	/* We take down all of the non-boot CPUs in one shot to avoid races
 	 * with the userspace trying to use the CPU hotplug at the same time
@@ -319,7 +385,7 @@ int disable_nonboot_cpus(void)
 	} else {
 		printk(KERN_ERR "Non-boot CPUs are not disabled\n");
 	}
-	mutex_unlock(&cpu_add_remove_lock);
+	cpu_maps_update_done();
 	return error;
 }
 
@@ -328,7 +394,7 @@ void enable_nonboot_cpus(void)
 	int cpu, error;
 
 	/* Allow everyone to use the CPU hotplug again */
-	mutex_lock(&cpu_add_remove_lock);
+	cpu_maps_update_begin();
 	cpu_hotplug_disabled = 0;
 	if (cpus_empty(frozen_cpus))
 		goto out;
@@ -344,6 +410,6 @@ void enable_nonboot_cpus(void)
 	}
 	cpus_clear(frozen_cpus);
 out:
-	mutex_unlock(&cpu_add_remove_lock);
+	cpu_maps_update_done();
 }
 #endif /* CONFIG_PM_SLEEP_SMP */
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 50f5dc46368..cfaf6419d81 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -537,10 +537,10 @@ static int cpusets_overlap(struct cpuset *a, struct cpuset *b)
  *
  * Call with cgroup_mutex held.  May take callback_mutex during
  * call due to the kfifo_alloc() and kmalloc() calls.  May nest
- * a call to the lock_cpu_hotplug()/unlock_cpu_hotplug() pair.
+ * a call to the get_online_cpus()/put_online_cpus() pair.
  * Must not be called holding callback_mutex, because we must not
- * call lock_cpu_hotplug() while holding callback_mutex.  Elsewhere
- * the kernel nests callback_mutex inside lock_cpu_hotplug() calls.
+ * call get_online_cpus() while holding callback_mutex.  Elsewhere
+ * the kernel nests callback_mutex inside get_online_cpus() calls.
  * So the reverse nesting would risk an ABBA deadlock.
  *
  * The three key local variables below are:
@@ -691,9 +691,9 @@ restart:
 
 rebuild:
 	/* Have scheduler rebuild sched domains */
-	lock_cpu_hotplug();
+	get_online_cpus();
 	partition_sched_domains(ndoms, doms);
-	unlock_cpu_hotplug();
+	put_online_cpus();
 
 done:
 	if (q && !IS_ERR(q))
@@ -1617,10 +1617,10 @@ static struct cgroup_subsys_state *cpuset_create(
  *
  * If the cpuset being removed has its flag 'sched_load_balance'
  * enabled, then simulate turning sched_load_balance off, which
- * will call rebuild_sched_domains().  The lock_cpu_hotplug()
+ * will call rebuild_sched_domains().  The get_online_cpus()
  * call in rebuild_sched_domains() must not be made while holding
  * callback_mutex.  Elsewhere the kernel nests callback_mutex inside
- * lock_cpu_hotplug() calls.  So the reverse nesting would risk an
+ * get_online_cpus() calls.  So the reverse nesting would risk an
  * ABBA deadlock.
  */
 
diff --git a/kernel/extable.c b/kernel/extable.c
index 7fe26285531..a26cb2e1702 100644
--- a/kernel/extable.c
+++ b/kernel/extable.c
@@ -46,7 +46,8 @@ int core_kernel_text(unsigned long addr)
 	    addr <= (unsigned long)_etext)
 		return 1;
 
-	if (addr >= (unsigned long)_sinittext &&
+	if (system_state == SYSTEM_BOOTING &&
+	    addr >= (unsigned long)_sinittext &&
 	    addr <= (unsigned long)_einittext)
 		return 1;
 	return 0;
diff --git a/kernel/fork.c b/kernel/fork.c
index 8dd8ff28100..05e0b6f4365 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -51,6 +51,7 @@
 #include <linux/random.h>
 #include <linux/tty.h>
 #include <linux/proc_fs.h>
+#include <linux/blkdev.h>
 
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
@@ -392,6 +393,7 @@ void fastcall __mmdrop(struct mm_struct *mm)
 	destroy_context(mm);
 	free_mm(mm);
 }
+EXPORT_SYMBOL_GPL(__mmdrop);
 
 /*
  * Decrement the use count and release all resources for an mm.
@@ -791,6 +793,31 @@ out:
 	return error;
 }
 
+static int copy_io(unsigned long clone_flags, struct task_struct *tsk)
+{
+#ifdef CONFIG_BLOCK
+	struct io_context *ioc = current->io_context;
+
+	if (!ioc)
+		return 0;
+	/*
+	 * Share io context with parent, if CLONE_IO is set
+	 */
+	if (clone_flags & CLONE_IO) {
+		tsk->io_context = ioc_task_link(ioc);
+		if (unlikely(!tsk->io_context))
+			return -ENOMEM;
+	} else if (ioprio_valid(ioc->ioprio)) {
+		tsk->io_context = alloc_io_context(GFP_KERNEL, -1);
+		if (unlikely(!tsk->io_context))
+			return -ENOMEM;
+
+		tsk->io_context->ioprio = ioc->ioprio;
+	}
+#endif
+	return 0;
+}
+
 /*
  *	Helper to unshare the files of the current task.
  *	We don't want to expose copy_files internals to
@@ -1045,6 +1072,10 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	copy_flags(clone_flags, p);
 	INIT_LIST_HEAD(&p->children);
 	INIT_LIST_HEAD(&p->sibling);
+#ifdef CONFIG_PREEMPT_RCU
+	p->rcu_read_lock_nesting = 0;
+	p->rcu_flipctr_idx = 0;
+#endif /* #ifdef CONFIG_PREEMPT_RCU */
 	p->vfork_done = NULL;
 	spin_lock_init(&p->alloc_lock);
 
@@ -1059,6 +1090,11 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	p->prev_utime = cputime_zero;
 	p->prev_stime = cputime_zero;
 
+#ifdef CONFIG_DETECT_SOFTLOCKUP
+	p->last_switch_count = 0;
+	p->last_switch_timestamp = 0;
+#endif
+
 #ifdef CONFIG_TASK_XACCT
 	p->rchar = 0;		/* I/O counter: bytes read */
 	p->wchar = 0;		/* I/O counter: bytes written */
@@ -1147,15 +1183,17 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 		goto bad_fork_cleanup_mm;
 	if ((retval = copy_namespaces(clone_flags, p)))
 		goto bad_fork_cleanup_keys;
+	if ((retval = copy_io(clone_flags, p)))
+		goto bad_fork_cleanup_namespaces;
 	retval = copy_thread(0, clone_flags, stack_start, stack_size, p, regs);
 	if (retval)
-		goto bad_fork_cleanup_namespaces;
+		goto bad_fork_cleanup_io;
 
 	if (pid != &init_struct_pid) {
 		retval = -ENOMEM;
 		pid = alloc_pid(task_active_pid_ns(p));
 		if (!pid)
-			goto bad_fork_cleanup_namespaces;
+			goto bad_fork_cleanup_io;
 
 		if (clone_flags & CLONE_NEWPID) {
 			retval = pid_ns_prepare_proc(task_active_pid_ns(p));
@@ -1196,6 +1234,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 #ifdef TIF_SYSCALL_EMU
 	clear_tsk_thread_flag(p, TIF_SYSCALL_EMU);
 #endif
+	clear_all_latency_tracing(p);
 
 	/* Our parent execution domain becomes current domain
 	   These must match for thread signalling to apply */
@@ -1224,9 +1263,6 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	/* Need tasklist lock for parent etc handling! */
 	write_lock_irq(&tasklist_lock);
 
-	/* for sys_ioprio_set(IOPRIO_WHO_PGRP) */
-	p->ioprio = current->ioprio;
-
 	/*
 	 * The task hasn't been attached yet, so its cpus_allowed mask will
 	 * not be changed, nor will its assigned CPU.
@@ -1237,6 +1273,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	 * parent's CPU). This avoids alot of nasty races.
 	 */
 	p->cpus_allowed = current->cpus_allowed;
+	p->rt.nr_cpus_allowed = current->rt.nr_cpus_allowed;
 	if (unlikely(!cpu_isset(task_cpu(p), p->cpus_allowed) ||
 			!cpu_online(task_cpu(p))))
 		set_task_cpu(p, smp_processor_id());
@@ -1317,6 +1354,8 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 bad_fork_free_pid:
 	if (pid != &init_struct_pid)
 		free_pid(pid);
+bad_fork_cleanup_io:
+	put_io_context(p->io_context);
 bad_fork_cleanup_namespaces:
 	exit_task_namespaces(p);
 bad_fork_cleanup_keys:
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index f994bb8065e..bd5d6b5060b 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -325,6 +325,22 @@ unsigned long ktime_divns(const ktime_t kt, s64 div)
 }
 #endif /* BITS_PER_LONG >= 64 */
 
+/*
+ * Check, whether the timer is on the callback pending list
+ */
+static inline int hrtimer_cb_pending(const struct hrtimer *timer)
+{
+	return timer->state & HRTIMER_STATE_PENDING;
+}
+
+/*
+ * Remove a timer from the callback pending list
+ */
+static inline void hrtimer_remove_cb_pending(struct hrtimer *timer)
+{
+	list_del_init(&timer->cb_entry);
+}
+
 /* High resolution timer related functions */
 #ifdef CONFIG_HIGH_RES_TIMERS
 
@@ -494,29 +510,12 @@ void hres_timers_resume(void)
 }
 
 /*
- * Check, whether the timer is on the callback pending list
- */
-static inline int hrtimer_cb_pending(const struct hrtimer *timer)
-{
-	return timer->state & HRTIMER_STATE_PENDING;
-}
-
-/*
- * Remove a timer from the callback pending list
- */
-static inline void hrtimer_remove_cb_pending(struct hrtimer *timer)
-{
-	list_del_init(&timer->cb_entry);
-}
-
-/*
  * Initialize the high resolution related parts of cpu_base
  */
 static inline void hrtimer_init_hres(struct hrtimer_cpu_base *base)
 {
 	base->expires_next.tv64 = KTIME_MAX;
 	base->hres_active = 0;
-	INIT_LIST_HEAD(&base->cb_pending);
 }
 
 /*
@@ -524,7 +523,6 @@ static inline void hrtimer_init_hres(struct hrtimer_cpu_base *base)
  */
 static inline void hrtimer_init_timer_hres(struct hrtimer *timer)
 {
-	INIT_LIST_HEAD(&timer->cb_entry);
 }
 
 /*
@@ -618,10 +616,13 @@ static inline int hrtimer_enqueue_reprogram(struct hrtimer *timer,
 {
 	return 0;
 }
-static inline int hrtimer_cb_pending(struct hrtimer *timer) { return 0; }
-static inline void hrtimer_remove_cb_pending(struct hrtimer *timer) { }
 static inline void hrtimer_init_hres(struct hrtimer_cpu_base *base) { }
 static inline void hrtimer_init_timer_hres(struct hrtimer *timer) { }
+static inline int hrtimer_reprogram(struct hrtimer *timer,
+				    struct hrtimer_clock_base *base)
+{
+	return 0;
+}
 
 #endif /* CONFIG_HIGH_RES_TIMERS */
 
@@ -1001,6 +1002,7 @@ void hrtimer_init(struct hrtimer *timer, clockid_t clock_id,
 		clock_id = CLOCK_MONOTONIC;
 
 	timer->base = &cpu_base->clock_base[clock_id];
+	INIT_LIST_HEAD(&timer->cb_entry);
 	hrtimer_init_timer_hres(timer);
 
 #ifdef CONFIG_TIMER_STATS
@@ -1030,6 +1032,85 @@ int hrtimer_get_res(const clockid_t which_clock, struct timespec *tp)
 }
 EXPORT_SYMBOL_GPL(hrtimer_get_res);
 
+static void run_hrtimer_pending(struct hrtimer_cpu_base *cpu_base)
+{
+	spin_lock_irq(&cpu_base->lock);
+
+	while (!list_empty(&cpu_base->cb_pending)) {
+		enum hrtimer_restart (*fn)(struct hrtimer *);
+		struct hrtimer *timer;
+		int restart;
+
+		timer = list_entry(cpu_base->cb_pending.next,
+				   struct hrtimer, cb_entry);
+
+		timer_stats_account_hrtimer(timer);
+
+		fn = timer->function;
+		__remove_hrtimer(timer, timer->base, HRTIMER_STATE_CALLBACK, 0);
+		spin_unlock_irq(&cpu_base->lock);
+
+		restart = fn(timer);
+
+		spin_lock_irq(&cpu_base->lock);
+
+		timer->state &= ~HRTIMER_STATE_CALLBACK;
+		if (restart == HRTIMER_RESTART) {
+			BUG_ON(hrtimer_active(timer));
+			/*
+			 * Enqueue the timer, allow reprogramming of the event
+			 * device
+			 */
+			enqueue_hrtimer(timer, timer->base, 1);
+		} else if (hrtimer_active(timer)) {
+			/*
+			 * If the timer was rearmed on another CPU, reprogram
+			 * the event device.
+			 */
+			if (timer->base->first == &timer->node)
+				hrtimer_reprogram(timer, timer->base);
+		}
+	}
+	spin_unlock_irq(&cpu_base->lock);
+}
+
+static void __run_hrtimer(struct hrtimer *timer)
+{
+	struct hrtimer_clock_base *base = timer->base;
+	struct hrtimer_cpu_base *cpu_base = base->cpu_base;
+	enum hrtimer_restart (*fn)(struct hrtimer *);
+	int restart;
+
+	__remove_hrtimer(timer, base, HRTIMER_STATE_CALLBACK, 0);
+	timer_stats_account_hrtimer(timer);
+
+	fn = timer->function;
+	if (timer->cb_mode == HRTIMER_CB_IRQSAFE_NO_SOFTIRQ) {
+		/*
+		 * Used for scheduler timers, avoid lock inversion with
+		 * rq->lock and tasklist_lock.
+		 *
+		 * These timers are required to deal with enqueue expiry
+		 * themselves and are not allowed to migrate.
+		 */
+		spin_unlock(&cpu_base->lock);
+		restart = fn(timer);
+		spin_lock(&cpu_base->lock);
+	} else
+		restart = fn(timer);
+
+	/*
+	 * Note: We clear the CALLBACK bit after enqueue_hrtimer to avoid
+	 * reprogramming of the event hardware. This happens at the end of this
+	 * function anyway.
+	 */
+	if (restart != HRTIMER_NORESTART) {
+		BUG_ON(timer->state != HRTIMER_STATE_CALLBACK);
+		enqueue_hrtimer(timer, base, 0);
+	}
+	timer->state &= ~HRTIMER_STATE_CALLBACK;
+}
+
 #ifdef CONFIG_HIGH_RES_TIMERS
 
 /*
@@ -1087,21 +1168,7 @@ void hrtimer_interrupt(struct clock_event_device *dev)
 				continue;
 			}
 
-			__remove_hrtimer(timer, base,
-					 HRTIMER_STATE_CALLBACK, 0);
-			timer_stats_account_hrtimer(timer);
-
-			/*
-			 * Note: We clear the CALLBACK bit after
-			 * enqueue_hrtimer to avoid reprogramming of
-			 * the event hardware. This happens at the end
-			 * of this function anyway.
-			 */
-			if (timer->function(timer) != HRTIMER_NORESTART) {
-				BUG_ON(timer->state != HRTIMER_STATE_CALLBACK);
-				enqueue_hrtimer(timer, base, 0);
-			}
-			timer->state &= ~HRTIMER_STATE_CALLBACK;
+			__run_hrtimer(timer);
 		}
 		spin_unlock(&cpu_base->lock);
 		base++;
@@ -1122,52 +1189,41 @@ void hrtimer_interrupt(struct clock_event_device *dev)
 
 static void run_hrtimer_softirq(struct softirq_action *h)
 {
-	struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases);
-
-	spin_lock_irq(&cpu_base->lock);
-
-	while (!list_empty(&cpu_base->cb_pending)) {
-		enum hrtimer_restart (*fn)(struct hrtimer *);
-		struct hrtimer *timer;
-		int restart;
-
-		timer = list_entry(cpu_base->cb_pending.next,
-				   struct hrtimer, cb_entry);
+	run_hrtimer_pending(&__get_cpu_var(hrtimer_bases));
+}
 
-		timer_stats_account_hrtimer(timer);
+#endif	/* CONFIG_HIGH_RES_TIMERS */
 
-		fn = timer->function;
-		__remove_hrtimer(timer, timer->base, HRTIMER_STATE_CALLBACK, 0);
-		spin_unlock_irq(&cpu_base->lock);
+/*
+ * Called from timer softirq every jiffy, expire hrtimers:
+ *
+ * For HRT its the fall back code to run the softirq in the timer
+ * softirq context in case the hrtimer initialization failed or has
+ * not been done yet.
+ */
+void hrtimer_run_pending(void)
+{
+	struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases);
 
-		restart = fn(timer);
+	if (hrtimer_hres_active())
+		return;
 
-		spin_lock_irq(&cpu_base->lock);
+	/*
+	 * This _is_ ugly: We have to check in the softirq context,
+	 * whether we can switch to highres and / or nohz mode. The
+	 * clocksource switch happens in the timer interrupt with
+	 * xtime_lock held. Notification from there only sets the
+	 * check bit in the tick_oneshot code, otherwise we might
+	 * deadlock vs. xtime_lock.
+	 */
+	if (tick_check_oneshot_change(!hrtimer_is_hres_enabled()))
+		hrtimer_switch_to_hres();
 
-		timer->state &= ~HRTIMER_STATE_CALLBACK;
-		if (restart == HRTIMER_RESTART) {
-			BUG_ON(hrtimer_active(timer));
-			/*
-			 * Enqueue the timer, allow reprogramming of the event
-			 * device
-			 */
-			enqueue_hrtimer(timer, timer->base, 1);
-		} else if (hrtimer_active(timer)) {
-			/*
-			 * If the timer was rearmed on another CPU, reprogram
-			 * the event device.
-			 */
-			if (timer->base->first == &timer->node)
-				hrtimer_reprogram(timer, timer->base);
-		}
-	}
-	spin_unlock_irq(&cpu_base->lock);
+	run_hrtimer_pending(cpu_base);
 }
 
-#endif	/* CONFIG_HIGH_RES_TIMERS */
-
 /*
- * Expire the per base hrtimer-queue:
+ * Called from hardirq context every jiffy
  */
 static inline void run_hrtimer_queue(struct hrtimer_cpu_base *cpu_base,
 				     int index)
@@ -1181,46 +1237,27 @@ static inline void run_hrtimer_queue(struct hrtimer_cpu_base *cpu_base,
 	if (base->get_softirq_time)
 		base->softirq_time = base->get_softirq_time();
 
-	spin_lock_irq(&cpu_base->lock);
+	spin_lock(&cpu_base->lock);
 
 	while ((node = base->first)) {
 		struct hrtimer *timer;
-		enum hrtimer_restart (*fn)(struct hrtimer *);
-		int restart;
 
 		timer = rb_entry(node, struct hrtimer, node);
 		if (base->softirq_time.tv64 <= timer->expires.tv64)
 			break;
 
-#ifdef CONFIG_HIGH_RES_TIMERS
-		WARN_ON_ONCE(timer->cb_mode == HRTIMER_CB_IRQSAFE_NO_SOFTIRQ);
-#endif
-		timer_stats_account_hrtimer(timer);
-
-		fn = timer->function;
-		__remove_hrtimer(timer, base, HRTIMER_STATE_CALLBACK, 0);
-		spin_unlock_irq(&cpu_base->lock);
-
-		restart = fn(timer);
-
-		spin_lock_irq(&cpu_base->lock);
-
-		timer->state &= ~HRTIMER_STATE_CALLBACK;
-		if (restart != HRTIMER_NORESTART) {
-			BUG_ON(hrtimer_active(timer));
-			enqueue_hrtimer(timer, base, 0);
+		if (timer->cb_mode == HRTIMER_CB_SOFTIRQ) {
+			__remove_hrtimer(timer, base, HRTIMER_STATE_PENDING, 0);
+			list_add_tail(&timer->cb_entry,
+					&base->cpu_base->cb_pending);
+			continue;
 		}
+
+		__run_hrtimer(timer);
 	}
-	spin_unlock_irq(&cpu_base->lock);
+	spin_unlock(&cpu_base->lock);
 }
 
-/*
- * Called from timer softirq every jiffy, expire hrtimers:
- *
- * For HRT its the fall back code to run the softirq in the timer
- * softirq context in case the hrtimer initialization failed or has
- * not been done yet.
- */
 void hrtimer_run_queues(void)
 {
 	struct hrtimer_cpu_base *cpu_base = &__get_cpu_var(hrtimer_bases);
@@ -1229,18 +1266,6 @@ void hrtimer_run_queues(void)
 	if (hrtimer_hres_active())
 		return;
 
-	/*
-	 * This _is_ ugly: We have to check in the softirq context,
-	 * whether we can switch to highres and / or nohz mode. The
-	 * clocksource switch happens in the timer interrupt with
-	 * xtime_lock held. Notification from there only sets the
-	 * check bit in the tick_oneshot code, otherwise we might
-	 * deadlock vs. xtime_lock.
-	 */
-	if (tick_check_oneshot_change(!hrtimer_is_hres_enabled()))
-		if (hrtimer_switch_to_hres())
-			return;
-
 	hrtimer_get_softirq_time(cpu_base);
 
 	for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++)
@@ -1268,7 +1293,7 @@ void hrtimer_init_sleeper(struct hrtimer_sleeper *sl, struct task_struct *task)
 	sl->timer.function = hrtimer_wakeup;
 	sl->task = task;
 #ifdef CONFIG_HIGH_RES_TIMERS
-	sl->timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_RESTART;
+	sl->timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ;
 #endif
 }
 
@@ -1279,6 +1304,8 @@ static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mod
 	do {
 		set_current_state(TASK_INTERRUPTIBLE);
 		hrtimer_start(&t->timer, t->timer.expires, mode);
+		if (!hrtimer_active(&t->timer))
+			t->task = NULL;
 
 		if (likely(t->task))
 			schedule();
@@ -1389,6 +1416,7 @@ static void __cpuinit init_hrtimers_cpu(int cpu)
 	for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++)
 		cpu_base->clock_base[i].cpu_base = cpu_base;
 
+	INIT_LIST_HEAD(&cpu_base->cb_pending);
 	hrtimer_init_hres(cpu_base);
 }
 
diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 1f314221d53..438a0146428 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -479,6 +479,9 @@ void free_irq(unsigned int irq, void *dev_id)
 			return;
 		}
 		printk(KERN_ERR "Trying to free already-free IRQ %d\n", irq);
+#ifdef CONFIG_DEBUG_SHIRQ
+		dump_stack();
+#endif
 		spin_unlock_irqrestore(&desc->lock, flags);
 		return;
 	}
diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c
index 50b81b98046..c2f2ccb0549 100644
--- a/kernel/irq/proc.c
+++ b/kernel/irq/proc.c
@@ -75,6 +75,18 @@ static int irq_affinity_write_proc(struct file *file, const char __user *buffer,
 
 #endif
 
+static int irq_spurious_read(char *page, char **start, off_t off,
+				  int count, int *eof, void *data)
+{
+	struct irq_desc *d = &irq_desc[(long) data];
+	return sprintf(page, "count %u\n"
+			     "unhandled %u\n"
+			     "last_unhandled %u ms\n",
+			d->irq_count,
+			d->irqs_unhandled,
+			jiffies_to_msecs(d->last_unhandled));
+}
+
 #define MAX_NAMELEN 128
 
 static int name_unique(unsigned int irq, struct irqaction *new_action)
@@ -118,6 +130,7 @@ void register_handler_proc(unsigned int irq, struct irqaction *action)
 void register_irq_proc(unsigned int irq)
 {
 	char name [MAX_NAMELEN];
+	struct proc_dir_entry *entry;
 
 	if (!root_irq_dir ||
 		(irq_desc[irq].chip == &no_irq_chip) ||
@@ -132,8 +145,6 @@ void register_irq_proc(unsigned int irq)
 
 #ifdef CONFIG_SMP
 	{
-		struct proc_dir_entry *entry;
-
 		/* create /proc/irq/<irq>/smp_affinity */
 		entry = create_proc_entry("smp_affinity", 0600, irq_desc[irq].dir);
 
@@ -144,6 +155,12 @@ void register_irq_proc(unsigned int irq)
 		}
 	}
 #endif
+
+	entry = create_proc_entry("spurious", 0444, irq_desc[irq].dir);
+	if (entry) {
+		entry->data = (void *)(long)irq;
+		entry->read_proc = irq_spurious_read;
+	}
 }
 
 #undef MAX_NAMELEN
diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
index 32b161972fa..a6b2bc831dd 100644
--- a/kernel/irq/spurious.c
+++ b/kernel/irq/spurious.c
@@ -10,6 +10,7 @@
 #include <linux/module.h>
 #include <linux/kallsyms.h>
 #include <linux/interrupt.h>
+#include <linux/moduleparam.h>
 
 static int irqfixup __read_mostly;
 
@@ -225,6 +226,8 @@ int noirqdebug_setup(char *str)
 }
 
 __setup("noirqdebug", noirqdebug_setup);
+module_param(noirqdebug, bool, 0644);
+MODULE_PARM_DESC(noirqdebug, "Disable irq lockup detection when true");
 
 static int __init irqfixup_setup(char *str)
 {
@@ -236,6 +239,8 @@ static int __init irqfixup_setup(char *str)
 }
 
 __setup("irqfixup", irqfixup_setup);
+module_param(irqfixup, int, 0644);
+MODULE_PARM_DESC("irqfixup", "0: No fixup, 1: irqfixup mode 2: irqpoll mode");
 
 static int __init irqpoll_setup(char *str)
 {
diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c
index 2fc25810509..7dadc71ce51 100644
--- a/kernel/kallsyms.c
+++ b/kernel/kallsyms.c
@@ -233,10 +233,11 @@ static unsigned long get_symbol_pos(unsigned long addr,
 int kallsyms_lookup_size_offset(unsigned long addr, unsigned long *symbolsize,
 				unsigned long *offset)
 {
+	char namebuf[KSYM_NAME_LEN];
 	if (is_ksym_addr(addr))
 		return !!get_symbol_pos(addr, symbolsize, offset);
 
-	return !!module_address_lookup(addr, symbolsize, offset, NULL);
+	return !!module_address_lookup(addr, symbolsize, offset, NULL, namebuf);
 }
 
 /*
@@ -251,8 +252,6 @@ const char *kallsyms_lookup(unsigned long addr,
 			    unsigned long *offset,
 			    char **modname, char *namebuf)
 {
-	const char *msym;
-
 	namebuf[KSYM_NAME_LEN - 1] = 0;
 	namebuf[0] = 0;
 
@@ -268,10 +267,8 @@ const char *kallsyms_lookup(unsigned long addr,
 	}
 
 	/* see if it's in a module */
-	msym = module_address_lookup(addr, symbolsize, offset, modname);
-	if (msym)
-		return strncpy(namebuf, msym, KSYM_NAME_LEN - 1);
-
+	return module_address_lookup(addr, symbolsize, offset, modname,
+				     namebuf);
 	return NULL;
 }
 
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index e3a5d817ac9..d0493eafea3 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -824,6 +824,8 @@ static int __init init_kprobes(void)
 	if (!err)
 		err = register_die_notifier(&kprobe_exceptions_nb);
 
+	if (!err)
+		init_test_probes();
 	return err;
 }
 
diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c
index 65daa5373ca..e53bc30e9ba 100644
--- a/kernel/ksysfs.c
+++ b/kernel/ksysfs.c
@@ -17,30 +17,34 @@
 #include <linux/sched.h>
 
 #define KERNEL_ATTR_RO(_name) \
-static struct subsys_attribute _name##_attr = __ATTR_RO(_name)
+static struct kobj_attribute _name##_attr = __ATTR_RO(_name)
 
 #define KERNEL_ATTR_RW(_name) \
-static struct subsys_attribute _name##_attr = \
+static struct kobj_attribute _name##_attr = \
 	__ATTR(_name, 0644, _name##_show, _name##_store)
 
 #if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET)
 /* current uevent sequence number */
-static ssize_t uevent_seqnum_show(struct kset *kset, char *page)
+static ssize_t uevent_seqnum_show(struct kobject *kobj,
+				  struct kobj_attribute *attr, char *buf)
 {
-	return sprintf(page, "%llu\n", (unsigned long long)uevent_seqnum);
+	return sprintf(buf, "%llu\n", (unsigned long long)uevent_seqnum);
 }
 KERNEL_ATTR_RO(uevent_seqnum);
 
 /* uevent helper program, used during early boo */
-static ssize_t uevent_helper_show(struct kset *kset, char *page)
+static ssize_t uevent_helper_show(struct kobject *kobj,
+				  struct kobj_attribute *attr, char *buf)
 {
-	return sprintf(page, "%s\n", uevent_helper);
+	return sprintf(buf, "%s\n", uevent_helper);
 }
-static ssize_t uevent_helper_store(struct kset *kset, const char *page, size_t count)
+static ssize_t uevent_helper_store(struct kobject *kobj,
+				   struct kobj_attribute *attr,
+				   const char *buf, size_t count)
 {
 	if (count+1 > UEVENT_HELPER_PATH_LEN)
 		return -ENOENT;
-	memcpy(uevent_helper, page, count);
+	memcpy(uevent_helper, buf, count);
 	uevent_helper[count] = '\0';
 	if (count && uevent_helper[count-1] == '\n')
 		uevent_helper[count-1] = '\0';
@@ -50,21 +54,24 @@ KERNEL_ATTR_RW(uevent_helper);
 #endif
 
 #ifdef CONFIG_KEXEC
-static ssize_t kexec_loaded_show(struct kset *kset, char *page)
+static ssize_t kexec_loaded_show(struct kobject *kobj,
+				 struct kobj_attribute *attr, char *buf)
 {
-	return sprintf(page, "%d\n", !!kexec_image);
+	return sprintf(buf, "%d\n", !!kexec_image);
 }
 KERNEL_ATTR_RO(kexec_loaded);
 
-static ssize_t kexec_crash_loaded_show(struct kset *kset, char *page)
+static ssize_t kexec_crash_loaded_show(struct kobject *kobj,
+				       struct kobj_attribute *attr, char *buf)
 {
-	return sprintf(page, "%d\n", !!kexec_crash_image);
+	return sprintf(buf, "%d\n", !!kexec_crash_image);
 }
 KERNEL_ATTR_RO(kexec_crash_loaded);
 
-static ssize_t vmcoreinfo_show(struct kset *kset, char *page)
+static ssize_t vmcoreinfo_show(struct kobject *kobj,
+			       struct kobj_attribute *attr, char *buf)
 {
-	return sprintf(page, "%lx %x\n",
+	return sprintf(buf, "%lx %x\n",
 		       paddr_vmcoreinfo_note(),
 		       (unsigned int)vmcoreinfo_max_size);
 }
@@ -94,8 +101,8 @@ static struct bin_attribute notes_attr = {
 	.read = &notes_read,
 };
 
-decl_subsys(kernel, NULL, NULL);
-EXPORT_SYMBOL_GPL(kernel_subsys);
+struct kobject *kernel_kobj;
+EXPORT_SYMBOL_GPL(kernel_kobj);
 
 static struct attribute * kernel_attrs[] = {
 #if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET)
@@ -116,24 +123,39 @@ static struct attribute_group kernel_attr_group = {
 
 static int __init ksysfs_init(void)
 {
-	int error = subsystem_register(&kernel_subsys);
-	if (!error)
-		error = sysfs_create_group(&kernel_subsys.kobj,
-					   &kernel_attr_group);
+	int error;
 
-	if (!error && notes_size > 0) {
-		notes_attr.size = notes_size;
-		error = sysfs_create_bin_file(&kernel_subsys.kobj,
-					      &notes_attr);
+	kernel_kobj = kobject_create_and_add("kernel", NULL);
+	if (!kernel_kobj) {
+		error = -ENOMEM;
+		goto exit;
 	}
+	error = sysfs_create_group(kernel_kobj, &kernel_attr_group);
+	if (error)
+		goto kset_exit;
 
-	/*
-	 * Create "/sys/kernel/uids" directory and corresponding root user's
-	 * directory under it.
-	 */
-	if (!error)
-		error = uids_kobject_init();
+	if (notes_size > 0) {
+		notes_attr.size = notes_size;
+		error = sysfs_create_bin_file(kernel_kobj, &notes_attr);
+		if (error)
+			goto group_exit;
+	}
 
+	/* create the /sys/kernel/uids/ directory */
+	error = uids_sysfs_init();
+	if (error)
+		goto notes_exit;
+
+	return 0;
+
+notes_exit:
+	if (notes_size > 0)
+		sysfs_remove_bin_file(kernel_kobj, &notes_attr);
+group_exit:
+	sysfs_remove_group(kernel_kobj, &kernel_attr_group);
+kset_exit:
+	kobject_put(kernel_kobj);
+exit:
 	return error;
 }
 
diff --git a/kernel/kthread.c b/kernel/kthread.c
index dcfe724300e..0ac887882f9 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -15,6 +15,8 @@
 #include <linux/mutex.h>
 #include <asm/semaphore.h>
 
+#define KTHREAD_NICE_LEVEL (-5)
+
 static DEFINE_SPINLOCK(kthread_create_lock);
 static LIST_HEAD(kthread_create_list);
 struct task_struct *kthreadd_task;
@@ -94,10 +96,18 @@ static void create_kthread(struct kthread_create_info *create)
 	if (pid < 0) {
 		create->result = ERR_PTR(pid);
 	} else {
+		struct sched_param param = { .sched_priority = 0 };
 		wait_for_completion(&create->started);
 		read_lock(&tasklist_lock);
 		create->result = find_task_by_pid(pid);
 		read_unlock(&tasklist_lock);
+		/*
+		 * root may have changed our (kthreadd's) priority or CPU mask.
+		 * The kernel thread should not inherit these properties.
+		 */
+		sched_setscheduler(create->result, SCHED_NORMAL, &param);
+		set_user_nice(create->result, KTHREAD_NICE_LEVEL);
+		set_cpus_allowed(create->result, CPU_MASK_ALL);
 	}
 	complete(&create->done);
 }
@@ -221,7 +231,7 @@ int kthreadd(void *unused)
 	/* Setup a clean context for our children to inherit. */
 	set_task_comm(tsk, "kthreadd");
 	ignore_signals(tsk);
-	set_user_nice(tsk, -5);
+	set_user_nice(tsk, KTHREAD_NICE_LEVEL);
 	set_cpus_allowed(tsk, CPU_MASK_ALL);
 
 	current->flags |= PF_NOFREEZE;
diff --git a/kernel/latencytop.c b/kernel/latencytop.c
new file mode 100644
index 00000000000..b4e3c85abe7
--- /dev/null
+++ b/kernel/latencytop.c
@@ -0,0 +1,239 @@
+/*
+ * latencytop.c: Latency display infrastructure
+ *
+ * (C) Copyright 2008 Intel Corporation
+ * Author: Arjan van de Ven <arjan@linux.intel.com>
+ *
+ * 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; version 2
+ * of the License.
+ */
+#include <linux/latencytop.h>
+#include <linux/kallsyms.h>
+#include <linux/seq_file.h>
+#include <linux/notifier.h>
+#include <linux/spinlock.h>
+#include <linux/proc_fs.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/list.h>
+#include <linux/slab.h>
+#include <linux/stacktrace.h>
+
+static DEFINE_SPINLOCK(latency_lock);
+
+#define MAXLR 128
+static struct latency_record latency_record[MAXLR];
+
+int latencytop_enabled;
+
+void clear_all_latency_tracing(struct task_struct *p)
+{
+	unsigned long flags;
+
+	if (!latencytop_enabled)
+		return;
+
+	spin_lock_irqsave(&latency_lock, flags);
+	memset(&p->latency_record, 0, sizeof(p->latency_record));
+	p->latency_record_count = 0;
+	spin_unlock_irqrestore(&latency_lock, flags);
+}
+
+static void clear_global_latency_tracing(void)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&latency_lock, flags);
+	memset(&latency_record, 0, sizeof(latency_record));
+	spin_unlock_irqrestore(&latency_lock, flags);
+}
+
+static void __sched
+account_global_scheduler_latency(struct task_struct *tsk, struct latency_record *lat)
+{
+	int firstnonnull = MAXLR + 1;
+	int i;
+
+	if (!latencytop_enabled)
+		return;
+
+	/* skip kernel threads for now */
+	if (!tsk->mm)
+		return;
+
+	for (i = 0; i < MAXLR; i++) {
+		int q;
+		int same = 1;
+		/* Nothing stored: */
+		if (!latency_record[i].backtrace[0]) {
+			if (firstnonnull > i)
+				firstnonnull = i;
+			continue;
+		}
+		for (q = 0 ; q < LT_BACKTRACEDEPTH ; q++) {
+			if (latency_record[i].backtrace[q] !=
+				lat->backtrace[q])
+				same = 0;
+			if (same && lat->backtrace[q] == 0)
+				break;
+			if (same && lat->backtrace[q] == ULONG_MAX)
+				break;
+		}
+		if (same) {
+			latency_record[i].count++;
+			latency_record[i].time += lat->time;
+			if (lat->time > latency_record[i].max)
+				latency_record[i].max = lat->time;
+			return;
+		}
+	}
+
+	i = firstnonnull;
+	if (i >= MAXLR - 1)
+		return;
+
+	/* Allocted a new one: */
+	memcpy(&latency_record[i], lat, sizeof(struct latency_record));
+}
+
+static inline void store_stacktrace(struct task_struct *tsk, struct latency_record *lat)
+{
+	struct stack_trace trace;
+
+	memset(&trace, 0, sizeof(trace));
+	trace.max_entries = LT_BACKTRACEDEPTH;
+	trace.entries = &lat->backtrace[0];
+	trace.skip = 0;
+	save_stack_trace_tsk(tsk, &trace);
+}
+
+void __sched
+account_scheduler_latency(struct task_struct *tsk, int usecs, int inter)
+{
+	unsigned long flags;
+	int i, q;
+	struct latency_record lat;
+
+	if (!latencytop_enabled)
+		return;
+
+	/* Long interruptible waits are generally user requested... */
+	if (inter && usecs > 5000)
+		return;
+
+	memset(&lat, 0, sizeof(lat));
+	lat.count = 1;
+	lat.time = usecs;
+	lat.max = usecs;
+	store_stacktrace(tsk, &lat);
+
+	spin_lock_irqsave(&latency_lock, flags);
+
+	account_global_scheduler_latency(tsk, &lat);
+
+	/*
+	 * short term hack; if we're > 32 we stop; future we recycle:
+	 */
+	tsk->latency_record_count++;
+	if (tsk->latency_record_count >= LT_SAVECOUNT)
+		goto out_unlock;
+
+	for (i = 0; i < LT_SAVECOUNT ; i++) {
+		struct latency_record *mylat;
+		int same = 1;
+		mylat = &tsk->latency_record[i];
+		for (q = 0 ; q < LT_BACKTRACEDEPTH ; q++) {
+			if (mylat->backtrace[q] !=
+				lat.backtrace[q])
+				same = 0;
+			if (same && lat.backtrace[q] == 0)
+				break;
+			if (same && lat.backtrace[q] == ULONG_MAX)
+				break;
+		}
+		if (same) {
+			mylat->count++;
+			mylat->time += lat.time;
+			if (lat.time > mylat->max)
+				mylat->max = lat.time;
+			goto out_unlock;
+		}
+	}
+
+	/* Allocated a new one: */
+	i = tsk->latency_record_count;
+	memcpy(&tsk->latency_record[i], &lat, sizeof(struct latency_record));
+
+out_unlock:
+	spin_unlock_irqrestore(&latency_lock, flags);
+}
+
+static int lstats_show(struct seq_file *m, void *v)
+{
+	int i;
+
+	seq_puts(m, "Latency Top version : v0.1\n");
+
+	for (i = 0; i < MAXLR; i++) {
+		if (latency_record[i].backtrace[0]) {
+			int q;
+			seq_printf(m, "%i %li %li ",
+				latency_record[i].count,
+				latency_record[i].time,
+				latency_record[i].max);
+			for (q = 0; q < LT_BACKTRACEDEPTH; q++) {
+				char sym[KSYM_NAME_LEN];
+				char *c;
+				if (!latency_record[i].backtrace[q])
+					break;
+				if (latency_record[i].backtrace[q] == ULONG_MAX)
+					break;
+				sprint_symbol(sym, latency_record[i].backtrace[q]);
+				c = strchr(sym, '+');
+				if (c)
+					*c = 0;
+				seq_printf(m, "%s ", sym);
+			}
+			seq_printf(m, "\n");
+		}
+	}
+	return 0;
+}
+
+static ssize_t
+lstats_write(struct file *file, const char __user *buf, size_t count,
+	     loff_t *offs)
+{
+	clear_global_latency_tracing();
+
+	return count;
+}
+
+static int lstats_open(struct inode *inode, struct file *filp)
+{
+	return single_open(filp, lstats_show, NULL);
+}
+
+static struct file_operations lstats_fops = {
+	.open		= lstats_open,
+	.read		= seq_read,
+	.write		= lstats_write,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+static int __init init_lstats_procfs(void)
+{
+	struct proc_dir_entry *pe;
+
+	pe = create_proc_entry("latency_stats", 0644, NULL);
+	if (!pe)
+		return -ENOMEM;
+
+	pe->proc_fops = &lstats_fops;
+
+	return 0;
+}
+__initcall(init_lstats_procfs);
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 4335f12a27c..3574379f4d6 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -2932,7 +2932,7 @@ static void zap_class(struct lock_class *class)
 
 }
 
-static inline int within(void *addr, void *start, unsigned long size)
+static inline int within(const void *addr, void *start, unsigned long size)
 {
 	return addr >= start && addr < start + size;
 }
@@ -2955,9 +2955,12 @@ void lockdep_free_key_range(void *start, unsigned long size)
 		head = classhash_table + i;
 		if (list_empty(head))
 			continue;
-		list_for_each_entry_safe(class, next, head, hash_entry)
+		list_for_each_entry_safe(class, next, head, hash_entry) {
 			if (within(class->key, start, size))
 				zap_class(class);
+			else if (within(class->name, start, size))
+				zap_class(class);
+		}
 	}
 
 	if (locked)
@@ -3203,7 +3206,11 @@ retry:
 
 EXPORT_SYMBOL_GPL(debug_show_all_locks);
 
-void debug_show_held_locks(struct task_struct *task)
+/*
+ * Careful: only use this function if you are sure that
+ * the task cannot run in parallel!
+ */
+void __debug_show_held_locks(struct task_struct *task)
 {
 	if (unlikely(!debug_locks)) {
 		printk("INFO: lockdep is turned off.\n");
@@ -3211,6 +3218,12 @@ void debug_show_held_locks(struct task_struct *task)
 	}
 	lockdep_print_held_locks(task);
 }
+EXPORT_SYMBOL_GPL(__debug_show_held_locks);
+
+void debug_show_held_locks(struct task_struct *task)
+{
+		__debug_show_held_locks(task);
+}
 
 EXPORT_SYMBOL_GPL(debug_show_held_locks);
 
diff --git a/kernel/module.c b/kernel/module.c
index c2e3e2e9880..bd60278ee70 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -47,8 +47,6 @@
 #include <asm/cacheflush.h>
 #include <linux/license.h>
 
-extern int module_sysfs_initialized;
-
 #if 0
 #define DEBUGP printk
 #else
@@ -67,6 +65,9 @@ extern int module_sysfs_initialized;
 static DEFINE_MUTEX(module_mutex);
 static LIST_HEAD(modules);
 
+/* Waiting for a module to finish initializing? */
+static DECLARE_WAIT_QUEUE_HEAD(module_wq);
+
 static BLOCKING_NOTIFIER_HEAD(module_notify_list);
 
 int register_module_notifier(struct notifier_block * nb)
@@ -86,8 +87,11 @@ EXPORT_SYMBOL(unregister_module_notifier);
 static inline int strong_try_module_get(struct module *mod)
 {
 	if (mod && mod->state == MODULE_STATE_COMING)
+		return -EBUSY;
+	if (try_module_get(mod))
 		return 0;
-	return try_module_get(mod);
+	else
+		return -ENOENT;
 }
 
 static inline void add_taint_module(struct module *mod, unsigned flag)
@@ -426,6 +430,14 @@ static unsigned int find_pcpusec(Elf_Ehdr *hdr,
 	return find_sec(hdr, sechdrs, secstrings, ".data.percpu");
 }
 
+static void percpu_modcopy(void *pcpudest, const void *from, unsigned long size)
+{
+	int cpu;
+
+	for_each_possible_cpu(cpu)
+		memcpy(pcpudest + per_cpu_offset(cpu), from, size);
+}
+
 static int percpu_modinit(void)
 {
 	pcpu_num_used = 2;
@@ -498,6 +510,8 @@ static struct module_attribute modinfo_##field = {                    \
 MODINFO_ATTR(version);
 MODINFO_ATTR(srcversion);
 
+static char last_unloaded_module[MODULE_NAME_LEN+1];
+
 #ifdef CONFIG_MODULE_UNLOAD
 /* Init the unload section of the module. */
 static void module_unload_init(struct module *mod)
@@ -539,11 +553,21 @@ static int already_uses(struct module *a, struct module *b)
 static int use_module(struct module *a, struct module *b)
 {
 	struct module_use *use;
-	int no_warn;
+	int no_warn, err;
 
 	if (b == NULL || already_uses(a, b)) return 1;
 
-	if (!strong_try_module_get(b))
+	/* If we're interrupted or time out, we fail. */
+	if (wait_event_interruptible_timeout(
+		    module_wq, (err = strong_try_module_get(b)) != -EBUSY,
+		    30 * HZ) <= 0) {
+		printk("%s: gave up waiting for init of module %s.\n",
+		       a->name, b->name);
+		return 0;
+	}
+
+	/* If strong_try_module_get() returned a different error, we fail. */
+	if (err)
 		return 0;
 
 	DEBUGP("Allocating new usage for %s.\n", a->name);
@@ -721,6 +745,8 @@ sys_delete_module(const char __user *name_user, unsigned int flags)
 		mod->exit();
 		mutex_lock(&module_mutex);
 	}
+	/* Store the name of the last unloaded module for diagnostic purposes */
+	strlcpy(last_unloaded_module, mod->name, sizeof(last_unloaded_module));
 	free_module(mod);
 
  out:
@@ -814,7 +840,7 @@ static inline void module_unload_free(struct module *mod)
 
 static inline int use_module(struct module *a, struct module *b)
 {
-	return strong_try_module_get(b);
+	return strong_try_module_get(b) == 0;
 }
 
 static inline void module_unload_init(struct module *mod)
@@ -1122,7 +1148,7 @@ static void add_notes_attrs(struct module *mod, unsigned int nsect,
 		++loaded;
 	}
 
-	notes_attrs->dir = kobject_add_dir(&mod->mkobj.kobj, "notes");
+	notes_attrs->dir = kobject_create_and_add("notes", &mod->mkobj.kobj);
 	if (!notes_attrs->dir)
 		goto out;
 
@@ -1212,6 +1238,7 @@ void module_remove_modinfo_attrs(struct module *mod)
 int mod_sysfs_init(struct module *mod)
 {
 	int err;
+	struct kobject *kobj;
 
 	if (!module_sysfs_initialized) {
 		printk(KERN_ERR "%s: module sysfs not initialized\n",
@@ -1219,15 +1246,25 @@ int mod_sysfs_init(struct module *mod)
 		err = -EINVAL;
 		goto out;
 	}
-	memset(&mod->mkobj.kobj, 0, sizeof(mod->mkobj.kobj));
-	err = kobject_set_name(&mod->mkobj.kobj, "%s", mod->name);
-	if (err)
+
+	kobj = kset_find_obj(module_kset, mod->name);
+	if (kobj) {
+		printk(KERN_ERR "%s: module is already loaded\n", mod->name);
+		kobject_put(kobj);
+		err = -EINVAL;
 		goto out;
-	kobj_set_kset_s(&mod->mkobj, module_subsys);
+	}
+
 	mod->mkobj.mod = mod;
 
-	kobject_init(&mod->mkobj.kobj);
+	memset(&mod->mkobj.kobj, 0, sizeof(mod->mkobj.kobj));
+	mod->mkobj.kobj.kset = module_kset;
+	err = kobject_init_and_add(&mod->mkobj.kobj, &module_ktype, NULL,
+				   "%s", mod->name);
+	if (err)
+		kobject_put(&mod->mkobj.kobj);
 
+	/* delay uevent until full sysfs population */
 out:
 	return err;
 }
@@ -1238,12 +1275,7 @@ int mod_sysfs_setup(struct module *mod,
 {
 	int err;
 
-	/* delay uevent until full sysfs population */
-	err = kobject_add(&mod->mkobj.kobj);
-	if (err)
-		goto out;
-
-	mod->holders_dir = kobject_add_dir(&mod->mkobj.kobj, "holders");
+	mod->holders_dir = kobject_create_and_add("holders", &mod->mkobj.kobj);
 	if (!mod->holders_dir) {
 		err = -ENOMEM;
 		goto out_unreg;
@@ -1263,11 +1295,9 @@ int mod_sysfs_setup(struct module *mod,
 out_unreg_param:
 	module_param_sysfs_remove(mod);
 out_unreg_holders:
-	kobject_unregister(mod->holders_dir);
+	kobject_put(mod->holders_dir);
 out_unreg:
-	kobject_del(&mod->mkobj.kobj);
 	kobject_put(&mod->mkobj.kobj);
-out:
 	return err;
 }
 #endif
@@ -1276,9 +1306,20 @@ static void mod_kobject_remove(struct module *mod)
 {
 	module_remove_modinfo_attrs(mod);
 	module_param_sysfs_remove(mod);
-	kobject_unregister(mod->mkobj.drivers_dir);
-	kobject_unregister(mod->holders_dir);
-	kobject_unregister(&mod->mkobj.kobj);
+	kobject_put(mod->mkobj.drivers_dir);
+	kobject_put(mod->holders_dir);
+	kobject_put(&mod->mkobj.kobj);
+}
+
+/*
+ * link the module with the whole machine is stopped with interrupts off
+ * - this defends against kallsyms not taking locks
+ */
+static int __link_module(void *_mod)
+{
+	struct module *mod = _mod;
+	list_add(&mod->list, &modules);
+	return 0;
 }
 
 /*
@@ -1330,7 +1371,7 @@ void *__symbol_get(const char *symbol)
 
 	preempt_disable();
 	value = __find_symbol(symbol, &owner, &crc, 1);
-	if (value && !strong_try_module_get(owner))
+	if (value && strong_try_module_get(owner) != 0)
 		value = 0;
 	preempt_enable();
 
@@ -1884,16 +1925,16 @@ static struct module *load_module(void __user *umod,
 	/* Now we've moved module, initialize linked lists, etc. */
 	module_unload_init(mod);
 
-	/* Initialize kobject, so we can reference it. */
+	/* add kobject, so we can reference it. */
 	err = mod_sysfs_init(mod);
 	if (err)
-		goto cleanup;
+		goto free_unload;
 
 	/* Set up license info based on the info section */
 	set_license(mod, get_modinfo(sechdrs, infoindex, "license"));
 
 	if (strcmp(mod->name, "ndiswrapper") == 0)
-		add_taint(TAINT_PROPRIETARY_MODULE);
+		add_taint_module(mod, TAINT_PROPRIETARY_MODULE);
 	if (strcmp(mod->name, "driverloader") == 0)
 		add_taint_module(mod, TAINT_PROPRIETARY_MODULE);
 
@@ -2023,6 +2064,11 @@ static struct module *load_module(void __user *umod,
 		printk(KERN_WARNING "%s: Ignoring obsolete parameters\n",
 		       mod->name);
 
+	/* Now sew it into the lists so we can get lockdep and oops
+         * info during argument parsing.  Noone should access us, since
+         * strong_try_module_get() will fail. */
+	stop_machine_run(__link_module, mod, NR_CPUS);
+
 	/* Size of section 0 is 0, so this works well if no params */
 	err = parse_args(mod->name, mod->args,
 			 (struct kernel_param *)
@@ -2031,7 +2077,7 @@ static struct module *load_module(void __user *umod,
 			 / sizeof(struct kernel_param),
 			 NULL);
 	if (err < 0)
-		goto arch_cleanup;
+		goto unlink;
 
 	err = mod_sysfs_setup(mod,
 			      (struct kernel_param *)
@@ -2039,7 +2085,7 @@ static struct module *load_module(void __user *umod,
 			      sechdrs[setupindex].sh_size
 			      / sizeof(struct kernel_param));
 	if (err < 0)
-		goto arch_cleanup;
+		goto unlink;
 	add_sect_attrs(mod, hdr->e_shnum, secstrings, sechdrs);
 	add_notes_attrs(mod, hdr->e_shnum, secstrings, sechdrs);
 
@@ -2054,9 +2100,13 @@ static struct module *load_module(void __user *umod,
 	/* Done! */
 	return mod;
 
- arch_cleanup:
+ unlink:
+	stop_machine_run(__unlink_module, mod, NR_CPUS);
 	module_arch_cleanup(mod);
  cleanup:
+	kobject_del(&mod->mkobj.kobj);
+	kobject_put(&mod->mkobj.kobj);
+ free_unload:
 	module_unload_free(mod);
 	module_free(mod, mod->module_init);
  free_core:
@@ -2076,17 +2126,6 @@ static struct module *load_module(void __user *umod,
 	goto free_hdr;
 }
 
-/*
- * link the module with the whole machine is stopped with interrupts off
- * - this defends against kallsyms not taking locks
- */
-static int __link_module(void *_mod)
-{
-	struct module *mod = _mod;
-	list_add(&mod->list, &modules);
-	return 0;
-}
-
 /* This is where the real work happens */
 asmlinkage long
 sys_init_module(void __user *umod,
@@ -2111,10 +2150,6 @@ sys_init_module(void __user *umod,
 		return PTR_ERR(mod);
 	}
 
-	/* Now sew it into the lists.  They won't access us, since
-           strong_try_module_get() will fail. */
-	stop_machine_run(__link_module, mod, NR_CPUS);
-
 	/* Drop lock so they can recurse */
 	mutex_unlock(&module_mutex);
 
@@ -2133,6 +2168,7 @@ sys_init_module(void __user *umod,
 		mutex_lock(&module_mutex);
 		free_module(mod);
 		mutex_unlock(&module_mutex);
+		wake_up(&module_wq);
 		return ret;
 	}
 
@@ -2147,6 +2183,7 @@ sys_init_module(void __user *umod,
 	mod->init_size = 0;
 	mod->init_text_size = 0;
 	mutex_unlock(&module_mutex);
+	wake_up(&module_wq);
 
 	return 0;
 }
@@ -2211,14 +2248,13 @@ static const char *get_ksymbol(struct module *mod,
 	return mod->strtab + mod->symtab[best].st_name;
 }
 
-/* For kallsyms to ask for address resolution.  NULL means not found.
-   We don't lock, as this is used for oops resolution and races are a
-   lesser concern. */
-/* FIXME: Risky: returns a pointer into a module w/o lock */
-const char *module_address_lookup(unsigned long addr,
-				  unsigned long *size,
-				  unsigned long *offset,
-				  char **modname)
+/* For kallsyms to ask for address resolution.  NULL means not found.  Careful
+ * not to lock to avoid deadlock on oopses, simply disable preemption. */
+char *module_address_lookup(unsigned long addr,
+			    unsigned long *size,
+			    unsigned long *offset,
+			    char **modname,
+			    char *namebuf)
 {
 	struct module *mod;
 	const char *ret = NULL;
@@ -2233,8 +2269,13 @@ const char *module_address_lookup(unsigned long addr,
 			break;
 		}
 	}
+	/* Make a copy in here where it's safe */
+	if (ret) {
+		strncpy(namebuf, ret, KSYM_NAME_LEN - 1);
+		ret = namebuf;
+	}
 	preempt_enable();
-	return ret;
+	return (char *)ret;
 }
 
 int lookup_module_symbol_name(unsigned long addr, char *symname)
@@ -2362,21 +2403,30 @@ static void m_stop(struct seq_file *m, void *p)
 	mutex_unlock(&module_mutex);
 }
 
-static char *taint_flags(unsigned int taints, char *buf)
+static char *module_flags(struct module *mod, char *buf)
 {
 	int bx = 0;
 
-	if (taints) {
+	if (mod->taints ||
+	    mod->state == MODULE_STATE_GOING ||
+	    mod->state == MODULE_STATE_COMING) {
 		buf[bx++] = '(';
-		if (taints & TAINT_PROPRIETARY_MODULE)
+		if (mod->taints & TAINT_PROPRIETARY_MODULE)
 			buf[bx++] = 'P';
-		if (taints & TAINT_FORCED_MODULE)
+		if (mod->taints & TAINT_FORCED_MODULE)
 			buf[bx++] = 'F';
 		/*
 		 * TAINT_FORCED_RMMOD: could be added.
 		 * TAINT_UNSAFE_SMP, TAINT_MACHINE_CHECK, TAINT_BAD_PAGE don't
 		 * apply to modules.
 		 */
+
+		/* Show a - for module-is-being-unloaded */
+		if (mod->state == MODULE_STATE_GOING)
+			buf[bx++] = '-';
+		/* Show a + for module-is-being-loaded */
+		if (mod->state == MODULE_STATE_COMING)
+			buf[bx++] = '+';
 		buf[bx++] = ')';
 	}
 	buf[bx] = '\0';
@@ -2403,7 +2453,7 @@ static int m_show(struct seq_file *m, void *p)
 
 	/* Taints info */
 	if (mod->taints)
-		seq_printf(m, " %s", taint_flags(mod->taints, buf));
+		seq_printf(m, " %s", module_flags(mod, buf));
 
 	seq_printf(m, "\n");
 	return 0;
@@ -2498,97 +2548,12 @@ void print_modules(void)
 
 	printk("Modules linked in:");
 	list_for_each_entry(mod, &modules, list)
-		printk(" %s%s", mod->name, taint_flags(mod->taints, buf));
+		printk(" %s%s", mod->name, module_flags(mod, buf));
+	if (last_unloaded_module[0])
+		printk(" [last unloaded: %s]", last_unloaded_module);
 	printk("\n");
 }
 
-#ifdef CONFIG_SYSFS
-static char *make_driver_name(struct device_driver *drv)
-{
-	char *driver_name;
-
-	driver_name = kmalloc(strlen(drv->name) + strlen(drv->bus->name) + 2,
-			      GFP_KERNEL);
-	if (!driver_name)
-		return NULL;
-
-	sprintf(driver_name, "%s:%s", drv->bus->name, drv->name);
-	return driver_name;
-}
-
-static void module_create_drivers_dir(struct module_kobject *mk)
-{
-	if (!mk || mk->drivers_dir)
-		return;
-
-	mk->drivers_dir = kobject_add_dir(&mk->kobj, "drivers");
-}
-
-void module_add_driver(struct module *mod, struct device_driver *drv)
-{
-	char *driver_name;
-	int no_warn;
-	struct module_kobject *mk = NULL;
-
-	if (!drv)
-		return;
-
-	if (mod)
-		mk = &mod->mkobj;
-	else if (drv->mod_name) {
-		struct kobject *mkobj;
-
-		/* Lookup built-in module entry in /sys/modules */
-		mkobj = kset_find_obj(&module_subsys, drv->mod_name);
-		if (mkobj) {
-			mk = container_of(mkobj, struct module_kobject, kobj);
-			/* remember our module structure */
-			drv->mkobj = mk;
-			/* kset_find_obj took a reference */
-			kobject_put(mkobj);
-		}
-	}
-
-	if (!mk)
-		return;
-
-	/* Don't check return codes; these calls are idempotent */
-	no_warn = sysfs_create_link(&drv->kobj, &mk->kobj, "module");
-	driver_name = make_driver_name(drv);
-	if (driver_name) {
-		module_create_drivers_dir(mk);
-		no_warn = sysfs_create_link(mk->drivers_dir, &drv->kobj,
-					    driver_name);
-		kfree(driver_name);
-	}
-}
-EXPORT_SYMBOL(module_add_driver);
-
-void module_remove_driver(struct device_driver *drv)
-{
-	struct module_kobject *mk = NULL;
-	char *driver_name;
-
-	if (!drv)
-		return;
-
-	sysfs_remove_link(&drv->kobj, "module");
-
-	if (drv->owner)
-		mk = &drv->owner->mkobj;
-	else if (drv->mkobj)
-		mk = drv->mkobj;
-	if (mk && mk->drivers_dir) {
-		driver_name = make_driver_name(drv);
-		if (driver_name) {
-			sysfs_remove_link(mk->drivers_dir, driver_name);
-			kfree(driver_name);
-		}
-	}
-}
-EXPORT_SYMBOL(module_remove_driver);
-#endif
-
 #ifdef CONFIG_MODVERSIONS
 /* Generate the signature for struct module here, too, for modversions. */
 void struct_module(struct module *mod) { return; }
diff --git a/kernel/panic.c b/kernel/panic.c
index da4d6bac270..d9e90cfe329 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -20,6 +20,7 @@
 #include <linux/kexec.h>
 #include <linux/debug_locks.h>
 #include <linux/random.h>
+#include <linux/kallsyms.h>
 
 int panic_on_oops;
 int tainted;
@@ -280,6 +281,13 @@ static int init_oops_id(void)
 }
 late_initcall(init_oops_id);
 
+static void print_oops_end_marker(void)
+{
+	init_oops_id();
+	printk(KERN_WARNING "---[ end trace %016llx ]---\n",
+		(unsigned long long)oops_id);
+}
+
 /*
  * Called when the architecture exits its oops handler, after printing
  * everything.
@@ -287,11 +295,26 @@ late_initcall(init_oops_id);
 void oops_exit(void)
 {
 	do_oops_enter_exit();
-	init_oops_id();
-	printk(KERN_WARNING "---[ end trace %016llx ]---\n",
-		(unsigned long long)oops_id);
+	print_oops_end_marker();
 }
 
+#ifdef WANT_WARN_ON_SLOWPATH
+void warn_on_slowpath(const char *file, int line)
+{
+	char function[KSYM_SYMBOL_LEN];
+	unsigned long caller = (unsigned long) __builtin_return_address(0);
+	sprint_symbol(function, caller);
+
+	printk(KERN_WARNING "------------[ cut here ]------------\n");
+	printk(KERN_WARNING "WARNING: at %s:%d %s()\n", file,
+		line, function);
+	print_modules();
+	dump_stack();
+	print_oops_end_marker();
+}
+EXPORT_SYMBOL(warn_on_slowpath);
+#endif
+
 #ifdef CONFIG_CC_STACKPROTECTOR
 /*
  * Called when gcc's -fstack-protector feature is used, and
diff --git a/kernel/params.c b/kernel/params.c
index 7686417ee00..42fe5e6126c 100644
--- a/kernel/params.c
+++ b/kernel/params.c
@@ -376,8 +376,6 @@ int param_get_string(char *buffer, struct kernel_param *kp)
 
 extern struct kernel_param __start___param[], __stop___param[];
 
-#define MAX_KBUILD_MODNAME KOBJ_NAME_LEN
-
 struct param_attribute
 {
 	struct module_attribute mattr;
@@ -472,7 +470,7 @@ param_sysfs_setup(struct module_kobject *mk,
 			sizeof(mp->grp.attrs[0]));
 	size[1] = (valid_attrs + 1) * sizeof(mp->grp.attrs[0]);
 
-	mp = kmalloc(size[0] + size[1], GFP_KERNEL);
+	mp = kzalloc(size[0] + size[1], GFP_KERNEL);
 	if (!mp)
 		return ERR_PTR(-ENOMEM);
 
@@ -560,11 +558,10 @@ static void __init kernel_param_sysfs_setup(const char *name,
 	BUG_ON(!mk);
 
 	mk->mod = THIS_MODULE;
-	kobj_set_kset_s(mk, module_subsys);
-	kobject_set_name(&mk->kobj, name);
-	kobject_init(&mk->kobj);
-	ret = kobject_add(&mk->kobj);
+	mk->kobj.kset = module_kset;
+	ret = kobject_init_and_add(&mk->kobj, &module_ktype, NULL, "%s", name);
 	if (ret) {
+		kobject_put(&mk->kobj);
 		printk(KERN_ERR "Module '%s' failed to be added to sysfs, "
 		      "error number %d\n", name, ret);
 		printk(KERN_ERR	"The system will be unstable now.\n");
@@ -588,7 +585,7 @@ static void __init param_sysfs_builtin(void)
 {
 	struct kernel_param *kp, *kp_begin = NULL;
 	unsigned int i, name_len, count = 0;
-	char modname[MAX_KBUILD_MODNAME + 1] = "";
+	char modname[MODULE_NAME_LEN + 1] = "";
 
 	for (i=0; i < __stop___param - __start___param; i++) {
 		char *dot;
@@ -596,12 +593,12 @@ static void __init param_sysfs_builtin(void)
 
 		kp = &__start___param[i];
 		max_name_len =
-			min_t(size_t, MAX_KBUILD_MODNAME, strlen(kp->name));
+			min_t(size_t, MODULE_NAME_LEN, strlen(kp->name));
 
 		dot = memchr(kp->name, '.', max_name_len);
 		if (!dot) {
 			DEBUGP("couldn't find period in first %d characters "
-			       "of %s\n", MAX_KBUILD_MODNAME, kp->name);
+			       "of %s\n", MODULE_NAME_LEN, kp->name);
 			continue;
 		}
 		name_len = dot - kp->name;
@@ -679,8 +676,6 @@ static struct sysfs_ops module_sysfs_ops = {
 	.store = module_attr_store,
 };
 
-static struct kobj_type module_ktype;
-
 static int uevent_filter(struct kset *kset, struct kobject *kobj)
 {
 	struct kobj_type *ktype = get_ktype(kobj);
@@ -694,21 +689,11 @@ static struct kset_uevent_ops module_uevent_ops = {
 	.filter = uevent_filter,
 };
 
-decl_subsys(module, &module_ktype, &module_uevent_ops);
+struct kset *module_kset;
 int module_sysfs_initialized;
 
-static void module_release(struct kobject *kobj)
-{
-	/*
-	 * Stupid empty release function to allow the memory for the kobject to
-	 * be properly cleaned up.  This will not need to be present for 2.6.25
-	 * with the upcoming kobject core rework.
-	 */
-}
-
-static struct kobj_type module_ktype = {
+struct kobj_type module_ktype = {
 	.sysfs_ops =	&module_sysfs_ops,
-	.release =	module_release,
 };
 
 /*
@@ -716,13 +701,11 @@ static struct kobj_type module_ktype = {
  */
 static int __init param_sysfs_init(void)
 {
-	int ret;
-
-	ret = subsystem_register(&module_subsys);
-	if (ret < 0) {
-		printk(KERN_WARNING "%s (%d): subsystem_register error: %d\n",
-			__FILE__, __LINE__, ret);
-		return ret;
+	module_kset = kset_create_and_add("module", &module_uevent_ops, NULL);
+	if (!module_kset) {
+		printk(KERN_WARNING "%s (%d): error creating kset\n",
+			__FILE__, __LINE__);
+		return -ENOMEM;
 	}
 	module_sysfs_initialized = 1;
 
@@ -732,14 +715,7 @@ static int __init param_sysfs_init(void)
 }
 subsys_initcall(param_sysfs_init);
 
-#else
-#if 0
-static struct sysfs_ops module_sysfs_ops = {
-	.show = NULL,
-	.store = NULL,
-};
-#endif
-#endif
+#endif /* CONFIG_SYSFS */
 
 EXPORT_SYMBOL(param_set_byte);
 EXPORT_SYMBOL(param_get_byte);
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index 68c96376e84..0b7c82ac467 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -967,6 +967,7 @@ static void check_thread_timers(struct task_struct *tsk,
 {
 	int maxfire;
 	struct list_head *timers = tsk->cpu_timers;
+	struct signal_struct *const sig = tsk->signal;
 
 	maxfire = 20;
 	tsk->it_prof_expires = cputime_zero;
@@ -1011,6 +1012,35 @@ static void check_thread_timers(struct task_struct *tsk,
 		t->firing = 1;
 		list_move_tail(&t->entry, firing);
 	}
+
+	/*
+	 * Check for the special case thread timers.
+	 */
+	if (sig->rlim[RLIMIT_RTTIME].rlim_cur != RLIM_INFINITY) {
+		unsigned long hard = sig->rlim[RLIMIT_RTTIME].rlim_max;
+		unsigned long *soft = &sig->rlim[RLIMIT_RTTIME].rlim_cur;
+
+		if (hard != RLIM_INFINITY &&
+		    tsk->rt.timeout > DIV_ROUND_UP(hard, USEC_PER_SEC/HZ)) {
+			/*
+			 * At the hard limit, we just die.
+			 * No need to calculate anything else now.
+			 */
+			__group_send_sig_info(SIGKILL, SEND_SIG_PRIV, tsk);
+			return;
+		}
+		if (tsk->rt.timeout > DIV_ROUND_UP(*soft, USEC_PER_SEC/HZ)) {
+			/*
+			 * At the soft limit, send a SIGXCPU every second.
+			 */
+			if (sig->rlim[RLIMIT_RTTIME].rlim_cur
+			    < sig->rlim[RLIMIT_RTTIME].rlim_max) {
+				sig->rlim[RLIMIT_RTTIME].rlim_cur +=
+								USEC_PER_SEC;
+			}
+			__group_send_sig_info(SIGXCPU, SEND_SIG_PRIV, tsk);
+		}
+	}
 }
 
 /*
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index 05b64790fe8..b138b431e27 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -567,7 +567,8 @@ static const char * const hibernation_modes[] = {
  *	supports it (as determined by having hibernation_ops).
  */
 
-static ssize_t disk_show(struct kset *kset, char *buf)
+static ssize_t disk_show(struct kobject *kobj, struct kobj_attribute *attr,
+			 char *buf)
 {
 	int i;
 	char *start = buf;
@@ -597,7 +598,8 @@ static ssize_t disk_show(struct kset *kset, char *buf)
 }
 
 
-static ssize_t disk_store(struct kset *kset, const char *buf, size_t n)
+static ssize_t disk_store(struct kobject *kobj, struct kobj_attribute *attr,
+			  const char *buf, size_t n)
 {
 	int error = 0;
 	int i;
@@ -642,13 +644,15 @@ static ssize_t disk_store(struct kset *kset, const char *buf, size_t n)
 
 power_attr(disk);
 
-static ssize_t resume_show(struct kset *kset, char *buf)
+static ssize_t resume_show(struct kobject *kobj, struct kobj_attribute *attr,
+			   char *buf)
 {
 	return sprintf(buf,"%d:%d\n", MAJOR(swsusp_resume_device),
 		       MINOR(swsusp_resume_device));
 }
 
-static ssize_t resume_store(struct kset *kset, const char *buf, size_t n)
+static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr,
+			    const char *buf, size_t n)
 {
 	unsigned int maj, min;
 	dev_t res;
@@ -674,12 +678,14 @@ static ssize_t resume_store(struct kset *kset, const char *buf, size_t n)
 
 power_attr(resume);
 
-static ssize_t image_size_show(struct kset *kset, char *buf)
+static ssize_t image_size_show(struct kobject *kobj, struct kobj_attribute *attr,
+			       char *buf)
 {
 	return sprintf(buf, "%lu\n", image_size);
 }
 
-static ssize_t image_size_store(struct kset *kset, const char *buf, size_t n)
+static ssize_t image_size_store(struct kobject *kobj, struct kobj_attribute *attr,
+				const char *buf, size_t n)
 {
 	unsigned long size;
 
@@ -708,7 +714,7 @@ static struct attribute_group attr_group = {
 
 static int __init pm_disk_init(void)
 {
-	return sysfs_create_group(&power_subsys.kobj, &attr_group);
+	return sysfs_create_group(power_kobj, &attr_group);
 }
 
 core_initcall(pm_disk_init);
diff --git a/kernel/power/main.c b/kernel/power/main.c
index f71c9504a5c..efc08360e62 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -276,8 +276,7 @@ EXPORT_SYMBOL(pm_suspend);
 
 #endif /* CONFIG_SUSPEND */
 
-decl_subsys(power,NULL,NULL);
-
+struct kobject *power_kobj;
 
 /**
  *	state - control system power state.
@@ -290,7 +289,8 @@ decl_subsys(power,NULL,NULL);
  *	proper enumerated value, and initiates a suspend transition.
  */
 
-static ssize_t state_show(struct kset *kset, char *buf)
+static ssize_t state_show(struct kobject *kobj, struct kobj_attribute *attr,
+			  char *buf)
 {
 	char *s = buf;
 #ifdef CONFIG_SUSPEND
@@ -311,7 +311,8 @@ static ssize_t state_show(struct kset *kset, char *buf)
 	return (s - buf);
 }
 
-static ssize_t state_store(struct kset *kset, const char *buf, size_t n)
+static ssize_t state_store(struct kobject *kobj, struct kobj_attribute *attr,
+			   const char *buf, size_t n)
 {
 #ifdef CONFIG_SUSPEND
 	suspend_state_t state = PM_SUSPEND_STANDBY;
@@ -348,13 +349,15 @@ power_attr(state);
 #ifdef CONFIG_PM_TRACE
 int pm_trace_enabled;
 
-static ssize_t pm_trace_show(struct kset *kset, char *buf)
+static ssize_t pm_trace_show(struct kobject *kobj, struct kobj_attribute *attr,
+			     char *buf)
 {
 	return sprintf(buf, "%d\n", pm_trace_enabled);
 }
 
 static ssize_t
-pm_trace_store(struct kset *kset, const char *buf, size_t n)
+pm_trace_store(struct kobject *kobj, struct kobj_attribute *attr,
+	       const char *buf, size_t n)
 {
 	int val;
 
@@ -386,10 +389,10 @@ static struct attribute_group attr_group = {
 
 static int __init pm_init(void)
 {
-	int error = subsystem_register(&power_subsys);
-	if (!error)
-		error = sysfs_create_group(&power_subsys.kobj,&attr_group);
-	return error;
+	power_kobj = kobject_create_and_add("power", NULL);
+	if (!power_kobj)
+		return -ENOMEM;
+	return sysfs_create_group(power_kobj, &attr_group);
 }
 
 core_initcall(pm_init);
diff --git a/kernel/power/power.h b/kernel/power/power.h
index 195dc461176..2093c3a9a99 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -54,7 +54,7 @@ extern int pfn_is_nosave(unsigned long);
 extern struct mutex pm_mutex;
 
 #define power_attr(_name) \
-static struct subsys_attribute _name##_attr = {	\
+static struct kobj_attribute _name##_attr = {	\
 	.attr	= {				\
 		.name = __stringify(_name),	\
 		.mode = 0644,			\
@@ -63,8 +63,6 @@ static struct subsys_attribute _name##_attr = {	\
 	.store	= _name##_store,		\
 }
 
-extern struct kset power_subsys;
-
 /* Preferred image size in bytes (default 500 MB) */
 extern unsigned long image_size;
 extern int in_suspend;
diff --git a/kernel/printk.c b/kernel/printk.c
index 89011bf8c10..58bbec68411 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -36,6 +36,13 @@
 
 #include <asm/uaccess.h>
 
+/*
+ * Architectures can override it:
+ */
+void __attribute__((weak)) early_printk(const char *fmt, ...)
+{
+}
+
 #define __LOG_BUF_LEN	(1 << CONFIG_LOG_BUF_SHIFT)
 
 /* printk's without a loglevel use this.. */
@@ -573,11 +580,6 @@ static int __init printk_time_setup(char *str)
 
 __setup("time", printk_time_setup);
 
-__attribute__((weak)) unsigned long long printk_clock(void)
-{
-	return sched_clock();
-}
-
 /* Check if we have any console registered that can be called early in boot. */
 static int have_callable_console(void)
 {
@@ -628,30 +630,57 @@ asmlinkage int printk(const char *fmt, ...)
 /* cpu currently holding logbuf_lock */
 static volatile unsigned int printk_cpu = UINT_MAX;
 
+const char printk_recursion_bug_msg [] =
+			KERN_CRIT "BUG: recent printk recursion!\n";
+static int printk_recursion_bug;
+
 asmlinkage int vprintk(const char *fmt, va_list args)
 {
+	static int log_level_unknown = 1;
+	static char printk_buf[1024];
+
 	unsigned long flags;
-	int printed_len;
+	int printed_len = 0;
+	int this_cpu;
 	char *p;
-	static char printk_buf[1024];
-	static int log_level_unknown = 1;
 
 	boot_delay_msec();
 
 	preempt_disable();
-	if (unlikely(oops_in_progress) && printk_cpu == smp_processor_id())
-		/* If a crash is occurring during printk() on this CPU,
-		 * make sure we can't deadlock */
-		zap_locks();
-
 	/* This stops the holder of console_sem just where we want him */
 	raw_local_irq_save(flags);
+	this_cpu = smp_processor_id();
+
+	/*
+	 * Ouch, printk recursed into itself!
+	 */
+	if (unlikely(printk_cpu == this_cpu)) {
+		/*
+		 * If a crash is occurring during printk() on this CPU,
+		 * then try to get the crash message out but make sure
+		 * we can't deadlock. Otherwise just return to avoid the
+		 * recursion and return - but flag the recursion so that
+		 * it can be printed at the next appropriate moment:
+		 */
+		if (!oops_in_progress) {
+			printk_recursion_bug = 1;
+			goto out_restore_irqs;
+		}
+		zap_locks();
+	}
+
 	lockdep_off();
 	spin_lock(&logbuf_lock);
-	printk_cpu = smp_processor_id();
+	printk_cpu = this_cpu;
 
+	if (printk_recursion_bug) {
+		printk_recursion_bug = 0;
+		strcpy(printk_buf, printk_recursion_bug_msg);
+		printed_len = sizeof(printk_recursion_bug_msg);
+	}
 	/* Emit the output into the temporary buffer */
-	printed_len = vscnprintf(printk_buf, sizeof(printk_buf), fmt, args);
+	printed_len += vscnprintf(printk_buf + printed_len,
+				  sizeof(printk_buf), fmt, args);
 
 	/*
 	 * Copy the output into log_buf.  If the caller didn't provide
@@ -680,7 +709,7 @@ asmlinkage int vprintk(const char *fmt, va_list args)
 					loglev_char = default_message_loglevel
 						+ '0';
 				}
-				t = printk_clock();
+				t = cpu_clock(printk_cpu);
 				nanosec_rem = do_div(t, 1000000000);
 				tlen = sprintf(tbuf,
 						"<%c>[%5lu.%06lu] ",
@@ -744,6 +773,7 @@ asmlinkage int vprintk(const char *fmt, va_list args)
 		printk_cpu = UINT_MAX;
 		spin_unlock(&logbuf_lock);
 		lockdep_on();
+out_restore_irqs:
 		raw_local_irq_restore(flags);
 	}
 
diff --git a/kernel/profile.c b/kernel/profile.c
index 5e95330e512..e64c2da11c0 100644
--- a/kernel/profile.c
+++ b/kernel/profile.c
@@ -52,7 +52,7 @@ static DEFINE_PER_CPU(int, cpu_profile_flip);
 static DEFINE_MUTEX(profile_flip_mutex);
 #endif /* CONFIG_SMP */
 
-static int __init profile_setup(char * str)
+static int __init profile_setup(char *str)
 {
 	static char __initdata schedstr[] = "schedule";
 	static char __initdata sleepstr[] = "sleep";
@@ -104,28 +104,28 @@ __setup("profile=", profile_setup);
 
 void __init profile_init(void)
 {
-	if (!prof_on) 
+	if (!prof_on)
 		return;
- 
+
 	/* only text is profiled */
 	prof_len = (_etext - _stext) >> prof_shift;
 	prof_buffer = alloc_bootmem(prof_len*sizeof(atomic_t));
 }
 
 /* Profile event notifications */
- 
+
 #ifdef CONFIG_PROFILING
- 
+
 static BLOCKING_NOTIFIER_HEAD(task_exit_notifier);
 static ATOMIC_NOTIFIER_HEAD(task_free_notifier);
 static BLOCKING_NOTIFIER_HEAD(munmap_notifier);
- 
-void profile_task_exit(struct task_struct * task)
+
+void profile_task_exit(struct task_struct *task)
 {
 	blocking_notifier_call_chain(&task_exit_notifier, 0, task);
 }
- 
-int profile_handoff_task(struct task_struct * task)
+
+int profile_handoff_task(struct task_struct *task)
 {
 	int ret;
 	ret = atomic_notifier_call_chain(&task_free_notifier, 0, task);
@@ -137,52 +137,55 @@ void profile_munmap(unsigned long addr)
 	blocking_notifier_call_chain(&munmap_notifier, 0, (void *)addr);
 }
 
-int task_handoff_register(struct notifier_block * n)
+int task_handoff_register(struct notifier_block *n)
 {
 	return atomic_notifier_chain_register(&task_free_notifier, n);
 }
+EXPORT_SYMBOL_GPL(task_handoff_register);
 
-int task_handoff_unregister(struct notifier_block * n)
+int task_handoff_unregister(struct notifier_block *n)
 {
 	return atomic_notifier_chain_unregister(&task_free_notifier, n);
 }
+EXPORT_SYMBOL_GPL(task_handoff_unregister);
 
-int profile_event_register(enum profile_type type, struct notifier_block * n)
+int profile_event_register(enum profile_type type, struct notifier_block *n)
 {
 	int err = -EINVAL;
- 
+
 	switch (type) {
-		case PROFILE_TASK_EXIT:
-			err = blocking_notifier_chain_register(
-					&task_exit_notifier, n);
-			break;
-		case PROFILE_MUNMAP:
-			err = blocking_notifier_chain_register(
-					&munmap_notifier, n);
-			break;
+	case PROFILE_TASK_EXIT:
+		err = blocking_notifier_chain_register(
+				&task_exit_notifier, n);
+		break;
+	case PROFILE_MUNMAP:
+		err = blocking_notifier_chain_register(
+				&munmap_notifier, n);
+		break;
 	}
- 
+
 	return err;
 }
+EXPORT_SYMBOL_GPL(profile_event_register);
 
- 
-int profile_event_unregister(enum profile_type type, struct notifier_block * n)
+int profile_event_unregister(enum profile_type type, struct notifier_block *n)
 {
 	int err = -EINVAL;
- 
+
 	switch (type) {
-		case PROFILE_TASK_EXIT:
-			err = blocking_notifier_chain_unregister(
-					&task_exit_notifier, n);
-			break;
-		case PROFILE_MUNMAP:
-			err = blocking_notifier_chain_unregister(
-					&munmap_notifier, n);
-			break;
+	case PROFILE_TASK_EXIT:
+		err = blocking_notifier_chain_unregister(
+				&task_exit_notifier, n);
+		break;
+	case PROFILE_MUNMAP:
+		err = blocking_notifier_chain_unregister(
+				&munmap_notifier, n);
+		break;
 	}
 
 	return err;
 }
+EXPORT_SYMBOL_GPL(profile_event_unregister);
 
 int register_timer_hook(int (*hook)(struct pt_regs *))
 {
@@ -191,6 +194,7 @@ int register_timer_hook(int (*hook)(struct pt_regs *))
 	timer_hook = hook;
 	return 0;
 }
+EXPORT_SYMBOL_GPL(register_timer_hook);
 
 void unregister_timer_hook(int (*hook)(struct pt_regs *))
 {
@@ -199,13 +203,7 @@ void unregister_timer_hook(int (*hook)(struct pt_regs *))
 	/* make sure all CPUs see the NULL hook */
 	synchronize_sched();  /* Allow ongoing interrupts to complete. */
 }
-
-EXPORT_SYMBOL_GPL(register_timer_hook);
 EXPORT_SYMBOL_GPL(unregister_timer_hook);
-EXPORT_SYMBOL_GPL(task_handoff_register);
-EXPORT_SYMBOL_GPL(task_handoff_unregister);
-EXPORT_SYMBOL_GPL(profile_event_register);
-EXPORT_SYMBOL_GPL(profile_event_unregister);
 
 #endif /* CONFIG_PROFILING */
 
@@ -366,7 +364,7 @@ static int __devinit profile_cpu_callback(struct notifier_block *info,
 			per_cpu(cpu_profile_hits, cpu)[0] = page_address(page);
 		}
 		break;
-	out_free:
+out_free:
 		page = virt_to_page(per_cpu(cpu_profile_hits, cpu)[1]);
 		per_cpu(cpu_profile_hits, cpu)[1] = NULL;
 		__free_page(page);
@@ -409,7 +407,6 @@ void profile_hits(int type, void *__pc, unsigned int nr_hits)
 	atomic_add(nr_hits, &prof_buffer[min(pc, prof_len - 1)]);
 }
 #endif /* !CONFIG_SMP */
-
 EXPORT_SYMBOL_GPL(profile_hits);
 
 void profile_tick(int type)
@@ -427,7 +424,7 @@ void profile_tick(int type)
 #include <asm/uaccess.h>
 #include <asm/ptrace.h>
 
-static int prof_cpu_mask_read_proc (char *page, char **start, off_t off,
+static int prof_cpu_mask_read_proc(char *page, char **start, off_t off,
 			int count, int *eof, void *data)
 {
 	int len = cpumask_scnprintf(page, count, *(cpumask_t *)data);
@@ -437,8 +434,8 @@ static int prof_cpu_mask_read_proc (char *page, char **start, off_t off,
 	return len;
 }
 
-static int prof_cpu_mask_write_proc (struct file *file, const char __user *buffer,
-					unsigned long count, void *data)
+static int prof_cpu_mask_write_proc(struct file *file,
+	const char __user *buffer,  unsigned long count, void *data)
 {
 	cpumask_t *mask = (cpumask_t *)data;
 	unsigned long full_count = count, err;
@@ -457,7 +454,8 @@ void create_prof_cpu_mask(struct proc_dir_entry *root_irq_dir)
 	struct proc_dir_entry *entry;
 
 	/* create /proc/irq/prof_cpu_mask */
-	if (!(entry = create_proc_entry("prof_cpu_mask", 0600, root_irq_dir)))
+	entry = create_proc_entry("prof_cpu_mask", 0600, root_irq_dir);
+	if (!entry)
 		return;
 	entry->data = (void *)&prof_cpu_mask;
 	entry->read_proc = prof_cpu_mask_read_proc;
@@ -475,7 +473,7 @@ read_profile(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 {
 	unsigned long p = *ppos;
 	ssize_t read;
-	char * pnt;
+	char *pnt;
 	unsigned int sample_step = 1 << prof_shift;
 
 	profile_flip_buffers();
@@ -486,12 +484,12 @@ read_profile(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 	read = 0;
 
 	while (p < sizeof(unsigned int) && count > 0) {
-		if (put_user(*((char *)(&sample_step)+p),buf))
+		if (put_user(*((char *)(&sample_step)+p), buf))
 			return -EFAULT;
 		buf++; p++; count--; read++;
 	}
 	pnt = (char *)prof_buffer + p - sizeof(atomic_t);
-	if (copy_to_user(buf,(void *)pnt,count))
+	if (copy_to_user(buf, (void *)pnt, count))
 		return -EFAULT;
 	read += count;
 	*ppos += read;
@@ -508,7 +506,7 @@ static ssize_t write_profile(struct file *file, const char __user *buf,
 			     size_t count, loff_t *ppos)
 {
 #ifdef CONFIG_SMP
-	extern int setup_profiling_timer (unsigned int multiplier);
+	extern int setup_profiling_timer(unsigned int multiplier);
 
 	if (count == sizeof(int)) {
 		unsigned int multiplier;
@@ -591,7 +589,8 @@ static int __init create_proc_profile(void)
 		return 0;
 	if (create_hash_tables())
 		return -1;
-	if (!(entry = create_proc_entry("profile", S_IWUSR | S_IRUGO, NULL)))
+	entry = create_proc_entry("profile", S_IWUSR | S_IRUGO, NULL);
+	if (!entry)
 		return 0;
 	entry->proc_fops = &proc_profile_operations;
 	entry->size = (1+prof_len) * sizeof(atomic_t);
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index c25db863081..e6e9b8be4b0 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -366,12 +366,73 @@ static int ptrace_setsiginfo(struct task_struct *child, siginfo_t __user * data)
 	return error;
 }
 
+
+#ifdef PTRACE_SINGLESTEP
+#define is_singlestep(request)		((request) == PTRACE_SINGLESTEP)
+#else
+#define is_singlestep(request)		0
+#endif
+
+#ifdef PTRACE_SINGLEBLOCK
+#define is_singleblock(request)		((request) == PTRACE_SINGLEBLOCK)
+#else
+#define is_singleblock(request)		0
+#endif
+
+#ifdef PTRACE_SYSEMU
+#define is_sysemu_singlestep(request)	((request) == PTRACE_SYSEMU_SINGLESTEP)
+#else
+#define is_sysemu_singlestep(request)	0
+#endif
+
+static int ptrace_resume(struct task_struct *child, long request, long data)
+{
+	if (!valid_signal(data))
+		return -EIO;
+
+	if (request == PTRACE_SYSCALL)
+		set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+	else
+		clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
+
+#ifdef TIF_SYSCALL_EMU
+	if (request == PTRACE_SYSEMU || request == PTRACE_SYSEMU_SINGLESTEP)
+		set_tsk_thread_flag(child, TIF_SYSCALL_EMU);
+	else
+		clear_tsk_thread_flag(child, TIF_SYSCALL_EMU);
+#endif
+
+	if (is_singleblock(request)) {
+		if (unlikely(!arch_has_block_step()))
+			return -EIO;
+		user_enable_block_step(child);
+	} else if (is_singlestep(request) || is_sysemu_singlestep(request)) {
+		if (unlikely(!arch_has_single_step()))
+			return -EIO;
+		user_enable_single_step(child);
+	}
+	else
+		user_disable_single_step(child);
+
+	child->exit_code = data;
+	wake_up_process(child);
+
+	return 0;
+}
+
 int ptrace_request(struct task_struct *child, long request,
 		   long addr, long data)
 {
 	int ret = -EIO;
 
 	switch (request) {
+	case PTRACE_PEEKTEXT:
+	case PTRACE_PEEKDATA:
+		return generic_ptrace_peekdata(child, addr, data);
+	case PTRACE_POKETEXT:
+	case PTRACE_POKEDATA:
+		return generic_ptrace_pokedata(child, addr, data);
+
 #ifdef PTRACE_OLDSETOPTIONS
 	case PTRACE_OLDSETOPTIONS:
 #endif
@@ -390,6 +451,26 @@ int ptrace_request(struct task_struct *child, long request,
 	case PTRACE_DETACH:	 /* detach a process that was attached. */
 		ret = ptrace_detach(child, data);
 		break;
+
+#ifdef PTRACE_SINGLESTEP
+	case PTRACE_SINGLESTEP:
+#endif
+#ifdef PTRACE_SINGLEBLOCK
+	case PTRACE_SINGLEBLOCK:
+#endif
+#ifdef PTRACE_SYSEMU
+	case PTRACE_SYSEMU:
+	case PTRACE_SYSEMU_SINGLESTEP:
+#endif
+	case PTRACE_SYSCALL:
+	case PTRACE_CONT:
+		return ptrace_resume(child, request, data);
+
+	case PTRACE_KILL:
+		if (child->exit_state)	/* already dead */
+			return 0;
+		return ptrace_resume(child, request, SIGKILL);
+
 	default:
 		break;
 	}
@@ -470,6 +551,8 @@ asmlinkage long sys_ptrace(long request, long pid, long addr, long data)
 	lock_kernel();
 	if (request == PTRACE_TRACEME) {
 		ret = ptrace_traceme();
+		if (!ret)
+			arch_ptrace_attach(current);
 		goto out;
 	}
 
@@ -524,3 +607,87 @@ int generic_ptrace_pokedata(struct task_struct *tsk, long addr, long data)
 	copied = access_process_vm(tsk, addr, &data, sizeof(data), 1);
 	return (copied == sizeof(data)) ? 0 : -EIO;
 }
+
+#ifdef CONFIG_COMPAT
+#include <linux/compat.h>
+
+int compat_ptrace_request(struct task_struct *child, compat_long_t request,
+			  compat_ulong_t addr, compat_ulong_t data)
+{
+	compat_ulong_t __user *datap = compat_ptr(data);
+	compat_ulong_t word;
+	int ret;
+
+	switch (request) {
+	case PTRACE_PEEKTEXT:
+	case PTRACE_PEEKDATA:
+		ret = access_process_vm(child, addr, &word, sizeof(word), 0);
+		if (ret != sizeof(word))
+			ret = -EIO;
+		else
+			ret = put_user(word, datap);
+		break;
+
+	case PTRACE_POKETEXT:
+	case PTRACE_POKEDATA:
+		ret = access_process_vm(child, addr, &data, sizeof(data), 1);
+		ret = (ret != sizeof(data) ? -EIO : 0);
+		break;
+
+	case PTRACE_GETEVENTMSG:
+		ret = put_user((compat_ulong_t) child->ptrace_message, datap);
+		break;
+
+	default:
+		ret = ptrace_request(child, request, addr, data);
+	}
+
+	return ret;
+}
+
+#ifdef __ARCH_WANT_COMPAT_SYS_PTRACE
+asmlinkage long compat_sys_ptrace(compat_long_t request, compat_long_t pid,
+				  compat_long_t addr, compat_long_t data)
+{
+	struct task_struct *child;
+	long ret;
+
+	/*
+	 * This lock_kernel fixes a subtle race with suid exec
+	 */
+	lock_kernel();
+	if (request == PTRACE_TRACEME) {
+		ret = ptrace_traceme();
+		goto out;
+	}
+
+	child = ptrace_get_task_struct(pid);
+	if (IS_ERR(child)) {
+		ret = PTR_ERR(child);
+		goto out;
+	}
+
+	if (request == PTRACE_ATTACH) {
+		ret = ptrace_attach(child);
+		/*
+		 * Some architectures need to do book-keeping after
+		 * a ptrace attach.
+		 */
+		if (!ret)
+			arch_ptrace_attach(child);
+		goto out_put_task_struct;
+	}
+
+	ret = ptrace_check_attach(child, request == PTRACE_KILL);
+	if (!ret)
+		ret = compat_arch_ptrace(child, request, addr, data);
+
+ out_put_task_struct:
+	put_task_struct(child);
+ out:
+	unlock_kernel();
+	return ret;
+}
+#endif /* __ARCH_WANT_COMPAT_SYS_PTRACE */
+
+#endif	/* CONFIG_COMPAT */
diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c
new file mode 100644
index 00000000000..f4ffbd0f306
--- /dev/null
+++ b/kernel/rcuclassic.c
@@ -0,0 +1,575 @@
+/*
+ * Read-Copy Update mechanism for mutual exclusion
+ *
+ * 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., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright IBM Corporation, 2001
+ *
+ * Authors: Dipankar Sarma <dipankar@in.ibm.com>
+ *	    Manfred Spraul <manfred@colorfullife.com>
+ *
+ * Based on the original work by Paul McKenney <paulmck@us.ibm.com>
+ * and inputs from Rusty Russell, Andrea Arcangeli and Andi Kleen.
+ * Papers:
+ * http://www.rdrop.com/users/paulmck/paper/rclockpdcsproof.pdf
+ * http://lse.sourceforge.net/locking/rclock_OLS.2001.05.01c.sc.pdf (OLS2001)
+ *
+ * For detailed explanation of Read-Copy Update mechanism see -
+ * 		Documentation/RCU
+ *
+ */
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/smp.h>
+#include <linux/rcupdate.h>
+#include <linux/interrupt.h>
+#include <linux/sched.h>
+#include <asm/atomic.h>
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <linux/completion.h>
+#include <linux/moduleparam.h>
+#include <linux/percpu.h>
+#include <linux/notifier.h>
+#include <linux/cpu.h>
+#include <linux/mutex.h>
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+static struct lock_class_key rcu_lock_key;
+struct lockdep_map rcu_lock_map =
+	STATIC_LOCKDEP_MAP_INIT("rcu_read_lock", &rcu_lock_key);
+EXPORT_SYMBOL_GPL(rcu_lock_map);
+#endif
+
+
+/* Definition for rcupdate control block. */
+static struct rcu_ctrlblk rcu_ctrlblk = {
+	.cur = -300,
+	.completed = -300,
+	.lock = __SPIN_LOCK_UNLOCKED(&rcu_ctrlblk.lock),
+	.cpumask = CPU_MASK_NONE,
+};
+static struct rcu_ctrlblk rcu_bh_ctrlblk = {
+	.cur = -300,
+	.completed = -300,
+	.lock = __SPIN_LOCK_UNLOCKED(&rcu_bh_ctrlblk.lock),
+	.cpumask = CPU_MASK_NONE,
+};
+
+DEFINE_PER_CPU(struct rcu_data, rcu_data) = { 0L };
+DEFINE_PER_CPU(struct rcu_data, rcu_bh_data) = { 0L };
+
+static int blimit = 10;
+static int qhimark = 10000;
+static int qlowmark = 100;
+
+#ifdef CONFIG_SMP
+static void force_quiescent_state(struct rcu_data *rdp,
+			struct rcu_ctrlblk *rcp)
+{
+	int cpu;
+	cpumask_t cpumask;
+	set_need_resched();
+	if (unlikely(!rcp->signaled)) {
+		rcp->signaled = 1;
+		/*
+		 * Don't send IPI to itself. With irqs disabled,
+		 * rdp->cpu is the current cpu.
+		 */
+		cpumask = rcp->cpumask;
+		cpu_clear(rdp->cpu, cpumask);
+		for_each_cpu_mask(cpu, cpumask)
+			smp_send_reschedule(cpu);
+	}
+}
+#else
+static inline void force_quiescent_state(struct rcu_data *rdp,
+			struct rcu_ctrlblk *rcp)
+{
+	set_need_resched();
+}
+#endif
+
+/**
+ * call_rcu - Queue an RCU callback for invocation after a grace period.
+ * @head: structure to be used for queueing the RCU updates.
+ * @func: actual update function to be invoked after the grace period
+ *
+ * The update function will be invoked some time after a full grace
+ * period elapses, in other words after all currently executing RCU
+ * read-side critical sections have completed.  RCU read-side critical
+ * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
+ * and may be nested.
+ */
+void call_rcu(struct rcu_head *head,
+				void (*func)(struct rcu_head *rcu))
+{
+	unsigned long flags;
+	struct rcu_data *rdp;
+
+	head->func = func;
+	head->next = NULL;
+	local_irq_save(flags);
+	rdp = &__get_cpu_var(rcu_data);
+	*rdp->nxttail = head;
+	rdp->nxttail = &head->next;
+	if (unlikely(++rdp->qlen > qhimark)) {
+		rdp->blimit = INT_MAX;
+		force_quiescent_state(rdp, &rcu_ctrlblk);
+	}
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL_GPL(call_rcu);
+
+/**
+ * call_rcu_bh - Queue an RCU for invocation after a quicker grace period.
+ * @head: structure to be used for queueing the RCU updates.
+ * @func: actual update function to be invoked after the grace period
+ *
+ * The update function will be invoked some time after a full grace
+ * period elapses, in other words after all currently executing RCU
+ * read-side critical sections have completed. call_rcu_bh() assumes
+ * that the read-side critical sections end on completion of a softirq
+ * handler. This means that read-side critical sections in process
+ * context must not be interrupted by softirqs. This interface is to be
+ * used when most of the read-side critical sections are in softirq context.
+ * RCU read-side critical sections are delimited by rcu_read_lock() and
+ * rcu_read_unlock(), * if in interrupt context or rcu_read_lock_bh()
+ * and rcu_read_unlock_bh(), if in process context. These may be nested.
+ */
+void call_rcu_bh(struct rcu_head *head,
+				void (*func)(struct rcu_head *rcu))
+{
+	unsigned long flags;
+	struct rcu_data *rdp;
+
+	head->func = func;
+	head->next = NULL;
+	local_irq_save(flags);
+	rdp = &__get_cpu_var(rcu_bh_data);
+	*rdp->nxttail = head;
+	rdp->nxttail = &head->next;
+
+	if (unlikely(++rdp->qlen > qhimark)) {
+		rdp->blimit = INT_MAX;
+		force_quiescent_state(rdp, &rcu_bh_ctrlblk);
+	}
+
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL_GPL(call_rcu_bh);
+
+/*
+ * Return the number of RCU batches processed thus far.  Useful
+ * for debug and statistics.
+ */
+long rcu_batches_completed(void)
+{
+	return rcu_ctrlblk.completed;
+}
+EXPORT_SYMBOL_GPL(rcu_batches_completed);
+
+/*
+ * Return the number of RCU batches processed thus far.  Useful
+ * for debug and statistics.
+ */
+long rcu_batches_completed_bh(void)
+{
+	return rcu_bh_ctrlblk.completed;
+}
+EXPORT_SYMBOL_GPL(rcu_batches_completed_bh);
+
+/* Raises the softirq for processing rcu_callbacks. */
+static inline void raise_rcu_softirq(void)
+{
+	raise_softirq(RCU_SOFTIRQ);
+	/*
+	 * The smp_mb() here is required to ensure that this cpu's
+	 * __rcu_process_callbacks() reads the most recently updated
+	 * value of rcu->cur.
+	 */
+	smp_mb();
+}
+
+/*
+ * Invoke the completed RCU callbacks. They are expected to be in
+ * a per-cpu list.
+ */
+static void rcu_do_batch(struct rcu_data *rdp)
+{
+	struct rcu_head *next, *list;
+	int count = 0;
+
+	list = rdp->donelist;
+	while (list) {
+		next = list->next;
+		prefetch(next);
+		list->func(list);
+		list = next;
+		if (++count >= rdp->blimit)
+			break;
+	}
+	rdp->donelist = list;
+
+	local_irq_disable();
+	rdp->qlen -= count;
+	local_irq_enable();
+	if (rdp->blimit == INT_MAX && rdp->qlen <= qlowmark)
+		rdp->blimit = blimit;
+
+	if (!rdp->donelist)
+		rdp->donetail = &rdp->donelist;
+	else
+		raise_rcu_softirq();
+}
+
+/*
+ * Grace period handling:
+ * The grace period handling consists out of two steps:
+ * - A new grace period is started.
+ *   This is done by rcu_start_batch. The start is not broadcasted to
+ *   all cpus, they must pick this up by comparing rcp->cur with
+ *   rdp->quiescbatch. All cpus are recorded  in the
+ *   rcu_ctrlblk.cpumask bitmap.
+ * - All cpus must go through a quiescent state.
+ *   Since the start of the grace period is not broadcasted, at least two
+ *   calls to rcu_check_quiescent_state are required:
+ *   The first call just notices that a new grace period is running. The
+ *   following calls check if there was a quiescent state since the beginning
+ *   of the grace period. If so, it updates rcu_ctrlblk.cpumask. If
+ *   the bitmap is empty, then the grace period is completed.
+ *   rcu_check_quiescent_state calls rcu_start_batch(0) to start the next grace
+ *   period (if necessary).
+ */
+/*
+ * Register a new batch of callbacks, and start it up if there is currently no
+ * active batch and the batch to be registered has not already occurred.
+ * Caller must hold rcu_ctrlblk.lock.
+ */
+static void rcu_start_batch(struct rcu_ctrlblk *rcp)
+{
+	if (rcp->next_pending &&
+			rcp->completed == rcp->cur) {
+		rcp->next_pending = 0;
+		/*
+		 * next_pending == 0 must be visible in
+		 * __rcu_process_callbacks() before it can see new value of cur.
+		 */
+		smp_wmb();
+		rcp->cur++;
+
+		/*
+		 * Accessing nohz_cpu_mask before incrementing rcp->cur needs a
+		 * Barrier  Otherwise it can cause tickless idle CPUs to be
+		 * included in rcp->cpumask, which will extend graceperiods
+		 * unnecessarily.
+		 */
+		smp_mb();
+		cpus_andnot(rcp->cpumask, cpu_online_map, nohz_cpu_mask);
+
+		rcp->signaled = 0;
+	}
+}
+
+/*
+ * cpu went through a quiescent state since the beginning of the grace period.
+ * Clear it from the cpu mask and complete the grace period if it was the last
+ * cpu. Start another grace period if someone has further entries pending
+ */
+static void cpu_quiet(int cpu, struct rcu_ctrlblk *rcp)
+{
+	cpu_clear(cpu, rcp->cpumask);
+	if (cpus_empty(rcp->cpumask)) {
+		/* batch completed ! */
+		rcp->completed = rcp->cur;
+		rcu_start_batch(rcp);
+	}
+}
+
+/*
+ * Check if the cpu has gone through a quiescent state (say context
+ * switch). If so and if it already hasn't done so in this RCU
+ * quiescent cycle, then indicate that it has done so.
+ */
+static void rcu_check_quiescent_state(struct rcu_ctrlblk *rcp,
+					struct rcu_data *rdp)
+{
+	if (rdp->quiescbatch != rcp->cur) {
+		/* start new grace period: */
+		rdp->qs_pending = 1;
+		rdp->passed_quiesc = 0;
+		rdp->quiescbatch = rcp->cur;
+		return;
+	}
+
+	/* Grace period already completed for this cpu?
+	 * qs_pending is checked instead of the actual bitmap to avoid
+	 * cacheline trashing.
+	 */
+	if (!rdp->qs_pending)
+		return;
+
+	/*
+	 * Was there a quiescent state since the beginning of the grace
+	 * period? If no, then exit and wait for the next call.
+	 */
+	if (!rdp->passed_quiesc)
+		return;
+	rdp->qs_pending = 0;
+
+	spin_lock(&rcp->lock);
+	/*
+	 * rdp->quiescbatch/rcp->cur and the cpu bitmap can come out of sync
+	 * during cpu startup. Ignore the quiescent state.
+	 */
+	if (likely(rdp->quiescbatch == rcp->cur))
+		cpu_quiet(rdp->cpu, rcp);
+
+	spin_unlock(&rcp->lock);
+}
+
+
+#ifdef CONFIG_HOTPLUG_CPU
+
+/* warning! helper for rcu_offline_cpu. do not use elsewhere without reviewing
+ * locking requirements, the list it's pulling from has to belong to a cpu
+ * which is dead and hence not processing interrupts.
+ */
+static void rcu_move_batch(struct rcu_data *this_rdp, struct rcu_head *list,
+				struct rcu_head **tail)
+{
+	local_irq_disable();
+	*this_rdp->nxttail = list;
+	if (list)
+		this_rdp->nxttail = tail;
+	local_irq_enable();
+}
+
+static void __rcu_offline_cpu(struct rcu_data *this_rdp,
+				struct rcu_ctrlblk *rcp, struct rcu_data *rdp)
+{
+	/* if the cpu going offline owns the grace period
+	 * we can block indefinitely waiting for it, so flush
+	 * it here
+	 */
+	spin_lock_bh(&rcp->lock);
+	if (rcp->cur != rcp->completed)
+		cpu_quiet(rdp->cpu, rcp);
+	spin_unlock_bh(&rcp->lock);
+	rcu_move_batch(this_rdp, rdp->donelist, rdp->donetail);
+	rcu_move_batch(this_rdp, rdp->curlist, rdp->curtail);
+	rcu_move_batch(this_rdp, rdp->nxtlist, rdp->nxttail);
+}
+
+static void rcu_offline_cpu(int cpu)
+{
+	struct rcu_data *this_rdp = &get_cpu_var(rcu_data);
+	struct rcu_data *this_bh_rdp = &get_cpu_var(rcu_bh_data);
+
+	__rcu_offline_cpu(this_rdp, &rcu_ctrlblk,
+					&per_cpu(rcu_data, cpu));
+	__rcu_offline_cpu(this_bh_rdp, &rcu_bh_ctrlblk,
+					&per_cpu(rcu_bh_data, cpu));
+	put_cpu_var(rcu_data);
+	put_cpu_var(rcu_bh_data);
+}
+
+#else
+
+static void rcu_offline_cpu(int cpu)
+{
+}
+
+#endif
+
+/*
+ * This does the RCU processing work from softirq context.
+ */
+static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp,
+					struct rcu_data *rdp)
+{
+	if (rdp->curlist && !rcu_batch_before(rcp->completed, rdp->batch)) {
+		*rdp->donetail = rdp->curlist;
+		rdp->donetail = rdp->curtail;
+		rdp->curlist = NULL;
+		rdp->curtail = &rdp->curlist;
+	}
+
+	if (rdp->nxtlist && !rdp->curlist) {
+		local_irq_disable();
+		rdp->curlist = rdp->nxtlist;
+		rdp->curtail = rdp->nxttail;
+		rdp->nxtlist = NULL;
+		rdp->nxttail = &rdp->nxtlist;
+		local_irq_enable();
+
+		/*
+		 * start the next batch of callbacks
+		 */
+
+		/* determine batch number */
+		rdp->batch = rcp->cur + 1;
+		/* see the comment and corresponding wmb() in
+		 * the rcu_start_batch()
+		 */
+		smp_rmb();
+
+		if (!rcp->next_pending) {
+			/* and start it/schedule start if it's a new batch */
+			spin_lock(&rcp->lock);
+			rcp->next_pending = 1;
+			rcu_start_batch(rcp);
+			spin_unlock(&rcp->lock);
+		}
+	}
+
+	rcu_check_quiescent_state(rcp, rdp);
+	if (rdp->donelist)
+		rcu_do_batch(rdp);
+}
+
+static void rcu_process_callbacks(struct softirq_action *unused)
+{
+	__rcu_process_callbacks(&rcu_ctrlblk, &__get_cpu_var(rcu_data));
+	__rcu_process_callbacks(&rcu_bh_ctrlblk, &__get_cpu_var(rcu_bh_data));
+}
+
+static int __rcu_pending(struct rcu_ctrlblk *rcp, struct rcu_data *rdp)
+{
+	/* This cpu has pending rcu entries and the grace period
+	 * for them has completed.
+	 */
+	if (rdp->curlist && !rcu_batch_before(rcp->completed, rdp->batch))
+		return 1;
+
+	/* This cpu has no pending entries, but there are new entries */
+	if (!rdp->curlist && rdp->nxtlist)
+		return 1;
+
+	/* This cpu has finished callbacks to invoke */
+	if (rdp->donelist)
+		return 1;
+
+	/* The rcu core waits for a quiescent state from the cpu */
+	if (rdp->quiescbatch != rcp->cur || rdp->qs_pending)
+		return 1;
+
+	/* nothing to do */
+	return 0;
+}
+
+/*
+ * Check to see if there is any immediate RCU-related work to be done
+ * by the current CPU, returning 1 if so.  This function is part of the
+ * RCU implementation; it is -not- an exported member of the RCU API.
+ */
+int rcu_pending(int cpu)
+{
+	return __rcu_pending(&rcu_ctrlblk, &per_cpu(rcu_data, cpu)) ||
+		__rcu_pending(&rcu_bh_ctrlblk, &per_cpu(rcu_bh_data, cpu));
+}
+
+/*
+ * Check to see if any future RCU-related work will need to be done
+ * by the current CPU, even if none need be done immediately, returning
+ * 1 if so.  This function is part of the RCU implementation; it is -not-
+ * an exported member of the RCU API.
+ */
+int rcu_needs_cpu(int cpu)
+{
+	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
+	struct rcu_data *rdp_bh = &per_cpu(rcu_bh_data, cpu);
+
+	return (!!rdp->curlist || !!rdp_bh->curlist || rcu_pending(cpu));
+}
+
+void rcu_check_callbacks(int cpu, int user)
+{
+	if (user ||
+	    (idle_cpu(cpu) && !in_softirq() &&
+				hardirq_count() <= (1 << HARDIRQ_SHIFT))) {
+		rcu_qsctr_inc(cpu);
+		rcu_bh_qsctr_inc(cpu);
+	} else if (!in_softirq())
+		rcu_bh_qsctr_inc(cpu);
+	raise_rcu_softirq();
+}
+
+static void rcu_init_percpu_data(int cpu, struct rcu_ctrlblk *rcp,
+						struct rcu_data *rdp)
+{
+	memset(rdp, 0, sizeof(*rdp));
+	rdp->curtail = &rdp->curlist;
+	rdp->nxttail = &rdp->nxtlist;
+	rdp->donetail = &rdp->donelist;
+	rdp->quiescbatch = rcp->completed;
+	rdp->qs_pending = 0;
+	rdp->cpu = cpu;
+	rdp->blimit = blimit;
+}
+
+static void __cpuinit rcu_online_cpu(int cpu)
+{
+	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
+	struct rcu_data *bh_rdp = &per_cpu(rcu_bh_data, cpu);
+
+	rcu_init_percpu_data(cpu, &rcu_ctrlblk, rdp);
+	rcu_init_percpu_data(cpu, &rcu_bh_ctrlblk, bh_rdp);
+	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks, NULL);
+}
+
+static int __cpuinit rcu_cpu_notify(struct notifier_block *self,
+				unsigned long action, void *hcpu)
+{
+	long cpu = (long)hcpu;
+
+	switch (action) {
+	case CPU_UP_PREPARE:
+	case CPU_UP_PREPARE_FROZEN:
+		rcu_online_cpu(cpu);
+		break;
+	case CPU_DEAD:
+	case CPU_DEAD_FROZEN:
+		rcu_offline_cpu(cpu);
+		break;
+	default:
+		break;
+	}
+	return NOTIFY_OK;
+}
+
+static struct notifier_block __cpuinitdata rcu_nb = {
+	.notifier_call	= rcu_cpu_notify,
+};
+
+/*
+ * Initializes rcu mechanism.  Assumed to be called early.
+ * That is before local timer(SMP) or jiffie timer (uniproc) is setup.
+ * Note that rcu_qsctr and friends are implicitly
+ * initialized due to the choice of ``0'' for RCU_CTR_INVALID.
+ */
+void __init __rcu_init(void)
+{
+	rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE,
+			(void *)(long)smp_processor_id());
+	/* Register notifier for non-boot CPUs */
+	register_cpu_notifier(&rcu_nb);
+}
+
+module_param(blimit, int, 0);
+module_param(qhimark, int, 0);
+module_param(qlowmark, int, 0);
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index f2c1a04e9b1..760dfc233a0 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -15,7 +15,7 @@
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
  *
- * Copyright (C) IBM Corporation, 2001
+ * Copyright IBM Corporation, 2001
  *
  * Authors: Dipankar Sarma <dipankar@in.ibm.com>
  *	    Manfred Spraul <manfred@colorfullife.com>
@@ -35,165 +35,57 @@
 #include <linux/init.h>
 #include <linux/spinlock.h>
 #include <linux/smp.h>
-#include <linux/rcupdate.h>
 #include <linux/interrupt.h>
 #include <linux/sched.h>
 #include <asm/atomic.h>
 #include <linux/bitops.h>
-#include <linux/module.h>
 #include <linux/completion.h>
-#include <linux/moduleparam.h>
 #include <linux/percpu.h>
 #include <linux/notifier.h>
 #include <linux/cpu.h>
 #include <linux/mutex.h>
+#include <linux/module.h>
 
-#ifdef CONFIG_DEBUG_LOCK_ALLOC
-static struct lock_class_key rcu_lock_key;
-struct lockdep_map rcu_lock_map =
-	STATIC_LOCKDEP_MAP_INIT("rcu_read_lock", &rcu_lock_key);
-
-EXPORT_SYMBOL_GPL(rcu_lock_map);
-#endif
-
-/* Definition for rcupdate control block. */
-static struct rcu_ctrlblk rcu_ctrlblk = {
-	.cur = -300,
-	.completed = -300,
-	.lock = __SPIN_LOCK_UNLOCKED(&rcu_ctrlblk.lock),
-	.cpumask = CPU_MASK_NONE,
-};
-static struct rcu_ctrlblk rcu_bh_ctrlblk = {
-	.cur = -300,
-	.completed = -300,
-	.lock = __SPIN_LOCK_UNLOCKED(&rcu_bh_ctrlblk.lock),
-	.cpumask = CPU_MASK_NONE,
+struct rcu_synchronize {
+	struct rcu_head head;
+	struct completion completion;
 };
 
-DEFINE_PER_CPU(struct rcu_data, rcu_data) = { 0L };
-DEFINE_PER_CPU(struct rcu_data, rcu_bh_data) = { 0L };
-
-/* Fake initialization required by compiler */
-static DEFINE_PER_CPU(struct tasklet_struct, rcu_tasklet) = {NULL};
-static int blimit = 10;
-static int qhimark = 10000;
-static int qlowmark = 100;
-
+static DEFINE_PER_CPU(struct rcu_head, rcu_barrier_head) = {NULL};
 static atomic_t rcu_barrier_cpu_count;
 static DEFINE_MUTEX(rcu_barrier_mutex);
 static struct completion rcu_barrier_completion;
 
-#ifdef CONFIG_SMP
-static void force_quiescent_state(struct rcu_data *rdp,
-			struct rcu_ctrlblk *rcp)
-{
-	int cpu;
-	cpumask_t cpumask;
-	set_need_resched();
-	if (unlikely(!rcp->signaled)) {
-		rcp->signaled = 1;
-		/*
-		 * Don't send IPI to itself. With irqs disabled,
-		 * rdp->cpu is the current cpu.
-		 */
-		cpumask = rcp->cpumask;
-		cpu_clear(rdp->cpu, cpumask);
-		for_each_cpu_mask(cpu, cpumask)
-			smp_send_reschedule(cpu);
-	}
-}
-#else
-static inline void force_quiescent_state(struct rcu_data *rdp,
-			struct rcu_ctrlblk *rcp)
+/* Because of FASTCALL declaration of complete, we use this wrapper */
+static void wakeme_after_rcu(struct rcu_head  *head)
 {
-	set_need_resched();
+	struct rcu_synchronize *rcu;
+
+	rcu = container_of(head, struct rcu_synchronize, head);
+	complete(&rcu->completion);
 }
-#endif
 
 /**
- * call_rcu - Queue an RCU callback for invocation after a grace period.
- * @head: structure to be used for queueing the RCU updates.
- * @func: actual update function to be invoked after the grace period
+ * synchronize_rcu - wait until a grace period has elapsed.
  *
- * The update function will be invoked some time after a full grace
- * period elapses, in other words after all currently executing RCU
+ * Control will return to the caller some time after a full grace
+ * period has elapsed, in other words after all currently executing RCU
  * read-side critical sections have completed.  RCU read-side critical
  * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
  * and may be nested.
  */
-void fastcall call_rcu(struct rcu_head *head,
-				void (*func)(struct rcu_head *rcu))
-{
-	unsigned long flags;
-	struct rcu_data *rdp;
-
-	head->func = func;
-	head->next = NULL;
-	local_irq_save(flags);
-	rdp = &__get_cpu_var(rcu_data);
-	*rdp->nxttail = head;
-	rdp->nxttail = &head->next;
-	if (unlikely(++rdp->qlen > qhimark)) {
-		rdp->blimit = INT_MAX;
-		force_quiescent_state(rdp, &rcu_ctrlblk);
-	}
-	local_irq_restore(flags);
-}
-
-/**
- * call_rcu_bh - Queue an RCU for invocation after a quicker grace period.
- * @head: structure to be used for queueing the RCU updates.
- * @func: actual update function to be invoked after the grace period
- *
- * The update function will be invoked some time after a full grace
- * period elapses, in other words after all currently executing RCU
- * read-side critical sections have completed. call_rcu_bh() assumes
- * that the read-side critical sections end on completion of a softirq
- * handler. This means that read-side critical sections in process
- * context must not be interrupted by softirqs. This interface is to be
- * used when most of the read-side critical sections are in softirq context.
- * RCU read-side critical sections are delimited by rcu_read_lock() and
- * rcu_read_unlock(), * if in interrupt context or rcu_read_lock_bh()
- * and rcu_read_unlock_bh(), if in process context. These may be nested.
- */
-void fastcall call_rcu_bh(struct rcu_head *head,
-				void (*func)(struct rcu_head *rcu))
+void synchronize_rcu(void)
 {
-	unsigned long flags;
-	struct rcu_data *rdp;
-
-	head->func = func;
-	head->next = NULL;
-	local_irq_save(flags);
-	rdp = &__get_cpu_var(rcu_bh_data);
-	*rdp->nxttail = head;
-	rdp->nxttail = &head->next;
-
-	if (unlikely(++rdp->qlen > qhimark)) {
-		rdp->blimit = INT_MAX;
-		force_quiescent_state(rdp, &rcu_bh_ctrlblk);
-	}
-
-	local_irq_restore(flags);
-}
+	struct rcu_synchronize rcu;
 
-/*
- * Return the number of RCU batches processed thus far.  Useful
- * for debug and statistics.
- */
-long rcu_batches_completed(void)
-{
-	return rcu_ctrlblk.completed;
-}
+	init_completion(&rcu.completion);
+	/* Will wake me after RCU finished */
+	call_rcu(&rcu.head, wakeme_after_rcu);
 
-/*
- * Return the number of RCU batches processed thus far.  Useful
- * for debug and statistics.
- */
-long rcu_batches_completed_bh(void)
-{
-	return rcu_bh_ctrlblk.completed;
+	/* Wait for it */
+	wait_for_completion(&rcu.completion);
 }
+EXPORT_SYMBOL_GPL(synchronize_rcu);
 
 static void rcu_barrier_callback(struct rcu_head *notused)
 {
@@ -207,10 +99,8 @@ static void rcu_barrier_callback(struct rcu_head *notused)
 static void rcu_barrier_func(void *notused)
 {
 	int cpu = smp_processor_id();
-	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
-	struct rcu_head *head;
+	struct rcu_head *head = &per_cpu(rcu_barrier_head, cpu);
 
-	head = &rdp->barrier;
 	atomic_inc(&rcu_barrier_cpu_count);
 	call_rcu(head, rcu_barrier_callback);
 }
@@ -225,420 +115,24 @@ void rcu_barrier(void)
 	mutex_lock(&rcu_barrier_mutex);
 	init_completion(&rcu_barrier_completion);
 	atomic_set(&rcu_barrier_cpu_count, 0);
+	/*
+	 * The queueing of callbacks in all CPUs must be atomic with
+	 * respect to RCU, otherwise one CPU may queue a callback,
+	 * wait for a grace period, decrement barrier count and call
+	 * complete(), while other CPUs have not yet queued anything.
+	 * So, we need to make sure that grace periods cannot complete
+	 * until all the callbacks are queued.
+	 */
+	rcu_read_lock();
 	on_each_cpu(rcu_barrier_func, NULL, 0, 1);
+	rcu_read_unlock();
 	wait_for_completion(&rcu_barrier_completion);
 	mutex_unlock(&rcu_barrier_mutex);
 }
 EXPORT_SYMBOL_GPL(rcu_barrier);
 
-/*
- * Invoke the completed RCU callbacks. They are expected to be in
- * a per-cpu list.
- */
-static void rcu_do_batch(struct rcu_data *rdp)
-{
-	struct rcu_head *next, *list;
-	int count = 0;
-
-	list = rdp->donelist;
-	while (list) {
-		next = list->next;
-		prefetch(next);
-		list->func(list);
-		list = next;
-		if (++count >= rdp->blimit)
-			break;
-	}
-	rdp->donelist = list;
-
-	local_irq_disable();
-	rdp->qlen -= count;
-	local_irq_enable();
-	if (rdp->blimit == INT_MAX && rdp->qlen <= qlowmark)
-		rdp->blimit = blimit;
-
-	if (!rdp->donelist)
-		rdp->donetail = &rdp->donelist;
-	else
-		tasklet_schedule(&per_cpu(rcu_tasklet, rdp->cpu));
-}
-
-/*
- * Grace period handling:
- * The grace period handling consists out of two steps:
- * - A new grace period is started.
- *   This is done by rcu_start_batch. The start is not broadcasted to
- *   all cpus, they must pick this up by comparing rcp->cur with
- *   rdp->quiescbatch. All cpus are recorded  in the
- *   rcu_ctrlblk.cpumask bitmap.
- * - All cpus must go through a quiescent state.
- *   Since the start of the grace period is not broadcasted, at least two
- *   calls to rcu_check_quiescent_state are required:
- *   The first call just notices that a new grace period is running. The
- *   following calls check if there was a quiescent state since the beginning
- *   of the grace period. If so, it updates rcu_ctrlblk.cpumask. If
- *   the bitmap is empty, then the grace period is completed.
- *   rcu_check_quiescent_state calls rcu_start_batch(0) to start the next grace
- *   period (if necessary).
- */
-/*
- * Register a new batch of callbacks, and start it up if there is currently no
- * active batch and the batch to be registered has not already occurred.
- * Caller must hold rcu_ctrlblk.lock.
- */
-static void rcu_start_batch(struct rcu_ctrlblk *rcp)
-{
-	if (rcp->next_pending &&
-			rcp->completed == rcp->cur) {
-		rcp->next_pending = 0;
-		/*
-		 * next_pending == 0 must be visible in
-		 * __rcu_process_callbacks() before it can see new value of cur.
-		 */
-		smp_wmb();
-		rcp->cur++;
-
-		/*
-		 * Accessing nohz_cpu_mask before incrementing rcp->cur needs a
-		 * Barrier  Otherwise it can cause tickless idle CPUs to be
-		 * included in rcp->cpumask, which will extend graceperiods
-		 * unnecessarily.
-		 */
-		smp_mb();
-		cpus_andnot(rcp->cpumask, cpu_online_map, nohz_cpu_mask);
-
-		rcp->signaled = 0;
-	}
-}
-
-/*
- * cpu went through a quiescent state since the beginning of the grace period.
- * Clear it from the cpu mask and complete the grace period if it was the last
- * cpu. Start another grace period if someone has further entries pending
- */
-static void cpu_quiet(int cpu, struct rcu_ctrlblk *rcp)
-{
-	cpu_clear(cpu, rcp->cpumask);
-	if (cpus_empty(rcp->cpumask)) {
-		/* batch completed ! */
-		rcp->completed = rcp->cur;
-		rcu_start_batch(rcp);
-	}
-}
-
-/*
- * Check if the cpu has gone through a quiescent state (say context
- * switch). If so and if it already hasn't done so in this RCU
- * quiescent cycle, then indicate that it has done so.
- */
-static void rcu_check_quiescent_state(struct rcu_ctrlblk *rcp,
-					struct rcu_data *rdp)
-{
-	if (rdp->quiescbatch != rcp->cur) {
-		/* start new grace period: */
-		rdp->qs_pending = 1;
-		rdp->passed_quiesc = 0;
-		rdp->quiescbatch = rcp->cur;
-		return;
-	}
-
-	/* Grace period already completed for this cpu?
-	 * qs_pending is checked instead of the actual bitmap to avoid
-	 * cacheline trashing.
-	 */
-	if (!rdp->qs_pending)
-		return;
-
-	/* 
-	 * Was there a quiescent state since the beginning of the grace
-	 * period? If no, then exit and wait for the next call.
-	 */
-	if (!rdp->passed_quiesc)
-		return;
-	rdp->qs_pending = 0;
-
-	spin_lock(&rcp->lock);
-	/*
-	 * rdp->quiescbatch/rcp->cur and the cpu bitmap can come out of sync
-	 * during cpu startup. Ignore the quiescent state.
-	 */
-	if (likely(rdp->quiescbatch == rcp->cur))
-		cpu_quiet(rdp->cpu, rcp);
-
-	spin_unlock(&rcp->lock);
-}
-
-
-#ifdef CONFIG_HOTPLUG_CPU
-
-/* warning! helper for rcu_offline_cpu. do not use elsewhere without reviewing
- * locking requirements, the list it's pulling from has to belong to a cpu
- * which is dead and hence not processing interrupts.
- */
-static void rcu_move_batch(struct rcu_data *this_rdp, struct rcu_head *list,
-				struct rcu_head **tail)
-{
-	local_irq_disable();
-	*this_rdp->nxttail = list;
-	if (list)
-		this_rdp->nxttail = tail;
-	local_irq_enable();
-}
-
-static void __rcu_offline_cpu(struct rcu_data *this_rdp,
-				struct rcu_ctrlblk *rcp, struct rcu_data *rdp)
-{
-	/* if the cpu going offline owns the grace period
-	 * we can block indefinitely waiting for it, so flush
-	 * it here
-	 */
-	spin_lock_bh(&rcp->lock);
-	if (rcp->cur != rcp->completed)
-		cpu_quiet(rdp->cpu, rcp);
-	spin_unlock_bh(&rcp->lock);
-	rcu_move_batch(this_rdp, rdp->curlist, rdp->curtail);
-	rcu_move_batch(this_rdp, rdp->nxtlist, rdp->nxttail);
-	rcu_move_batch(this_rdp, rdp->donelist, rdp->donetail);
-}
-
-static void rcu_offline_cpu(int cpu)
-{
-	struct rcu_data *this_rdp = &get_cpu_var(rcu_data);
-	struct rcu_data *this_bh_rdp = &get_cpu_var(rcu_bh_data);
-
-	__rcu_offline_cpu(this_rdp, &rcu_ctrlblk,
-					&per_cpu(rcu_data, cpu));
-	__rcu_offline_cpu(this_bh_rdp, &rcu_bh_ctrlblk,
-					&per_cpu(rcu_bh_data, cpu));
-	put_cpu_var(rcu_data);
-	put_cpu_var(rcu_bh_data);
-	tasklet_kill_immediate(&per_cpu(rcu_tasklet, cpu), cpu);
-}
-
-#else
-
-static void rcu_offline_cpu(int cpu)
-{
-}
-
-#endif
-
-/*
- * This does the RCU processing work from tasklet context. 
- */
-static void __rcu_process_callbacks(struct rcu_ctrlblk *rcp,
-					struct rcu_data *rdp)
-{
-	if (rdp->curlist && !rcu_batch_before(rcp->completed, rdp->batch)) {
-		*rdp->donetail = rdp->curlist;
-		rdp->donetail = rdp->curtail;
-		rdp->curlist = NULL;
-		rdp->curtail = &rdp->curlist;
-	}
-
-	if (rdp->nxtlist && !rdp->curlist) {
-		local_irq_disable();
-		rdp->curlist = rdp->nxtlist;
-		rdp->curtail = rdp->nxttail;
-		rdp->nxtlist = NULL;
-		rdp->nxttail = &rdp->nxtlist;
-		local_irq_enable();
-
-		/*
-		 * start the next batch of callbacks
-		 */
-
-		/* determine batch number */
-		rdp->batch = rcp->cur + 1;
-		/* see the comment and corresponding wmb() in
-		 * the rcu_start_batch()
-		 */
-		smp_rmb();
-
-		if (!rcp->next_pending) {
-			/* and start it/schedule start if it's a new batch */
-			spin_lock(&rcp->lock);
-			rcp->next_pending = 1;
-			rcu_start_batch(rcp);
-			spin_unlock(&rcp->lock);
-		}
-	}
-
-	rcu_check_quiescent_state(rcp, rdp);
-	if (rdp->donelist)
-		rcu_do_batch(rdp);
-}
-
-static void rcu_process_callbacks(unsigned long unused)
-{
-	__rcu_process_callbacks(&rcu_ctrlblk, &__get_cpu_var(rcu_data));
-	__rcu_process_callbacks(&rcu_bh_ctrlblk, &__get_cpu_var(rcu_bh_data));
-}
-
-static int __rcu_pending(struct rcu_ctrlblk *rcp, struct rcu_data *rdp)
-{
-	/* This cpu has pending rcu entries and the grace period
-	 * for them has completed.
-	 */
-	if (rdp->curlist && !rcu_batch_before(rcp->completed, rdp->batch))
-		return 1;
-
-	/* This cpu has no pending entries, but there are new entries */
-	if (!rdp->curlist && rdp->nxtlist)
-		return 1;
-
-	/* This cpu has finished callbacks to invoke */
-	if (rdp->donelist)
-		return 1;
-
-	/* The rcu core waits for a quiescent state from the cpu */
-	if (rdp->quiescbatch != rcp->cur || rdp->qs_pending)
-		return 1;
-
-	/* nothing to do */
-	return 0;
-}
-
-/*
- * Check to see if there is any immediate RCU-related work to be done
- * by the current CPU, returning 1 if so.  This function is part of the
- * RCU implementation; it is -not- an exported member of the RCU API.
- */
-int rcu_pending(int cpu)
-{
-	return __rcu_pending(&rcu_ctrlblk, &per_cpu(rcu_data, cpu)) ||
-		__rcu_pending(&rcu_bh_ctrlblk, &per_cpu(rcu_bh_data, cpu));
-}
-
-/*
- * Check to see if any future RCU-related work will need to be done
- * by the current CPU, even if none need be done immediately, returning
- * 1 if so.  This function is part of the RCU implementation; it is -not-
- * an exported member of the RCU API.
- */
-int rcu_needs_cpu(int cpu)
-{
-	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
-	struct rcu_data *rdp_bh = &per_cpu(rcu_bh_data, cpu);
-
-	return (!!rdp->curlist || !!rdp_bh->curlist || rcu_pending(cpu));
-}
-
-void rcu_check_callbacks(int cpu, int user)
-{
-	if (user || 
-	    (idle_cpu(cpu) && !in_softirq() && 
-				hardirq_count() <= (1 << HARDIRQ_SHIFT))) {
-		rcu_qsctr_inc(cpu);
-		rcu_bh_qsctr_inc(cpu);
-	} else if (!in_softirq())
-		rcu_bh_qsctr_inc(cpu);
-	tasklet_schedule(&per_cpu(rcu_tasklet, cpu));
-}
-
-static void rcu_init_percpu_data(int cpu, struct rcu_ctrlblk *rcp,
-						struct rcu_data *rdp)
-{
-	memset(rdp, 0, sizeof(*rdp));
-	rdp->curtail = &rdp->curlist;
-	rdp->nxttail = &rdp->nxtlist;
-	rdp->donetail = &rdp->donelist;
-	rdp->quiescbatch = rcp->completed;
-	rdp->qs_pending = 0;
-	rdp->cpu = cpu;
-	rdp->blimit = blimit;
-}
-
-static void __cpuinit rcu_online_cpu(int cpu)
-{
-	struct rcu_data *rdp = &per_cpu(rcu_data, cpu);
-	struct rcu_data *bh_rdp = &per_cpu(rcu_bh_data, cpu);
-
-	rcu_init_percpu_data(cpu, &rcu_ctrlblk, rdp);
-	rcu_init_percpu_data(cpu, &rcu_bh_ctrlblk, bh_rdp);
-	tasklet_init(&per_cpu(rcu_tasklet, cpu), rcu_process_callbacks, 0UL);
-}
-
-static int __cpuinit rcu_cpu_notify(struct notifier_block *self,
-				unsigned long action, void *hcpu)
-{
-	long cpu = (long)hcpu;
-	switch (action) {
-	case CPU_UP_PREPARE:
-	case CPU_UP_PREPARE_FROZEN:
-		rcu_online_cpu(cpu);
-		break;
-	case CPU_DEAD:
-	case CPU_DEAD_FROZEN:
-		rcu_offline_cpu(cpu);
-		break;
-	default:
-		break;
-	}
-	return NOTIFY_OK;
-}
-
-static struct notifier_block __cpuinitdata rcu_nb = {
-	.notifier_call	= rcu_cpu_notify,
-};
-
-/*
- * Initializes rcu mechanism.  Assumed to be called early.
- * That is before local timer(SMP) or jiffie timer (uniproc) is setup.
- * Note that rcu_qsctr and friends are implicitly
- * initialized due to the choice of ``0'' for RCU_CTR_INVALID.
- */
 void __init rcu_init(void)
 {
-	rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE,
-			(void *)(long)smp_processor_id());
-	/* Register notifier for non-boot CPUs */
-	register_cpu_notifier(&rcu_nb);
-}
-
-struct rcu_synchronize {
-	struct rcu_head head;
-	struct completion completion;
-};
-
-/* Because of FASTCALL declaration of complete, we use this wrapper */
-static void wakeme_after_rcu(struct rcu_head  *head)
-{
-	struct rcu_synchronize *rcu;
-
-	rcu = container_of(head, struct rcu_synchronize, head);
-	complete(&rcu->completion);
+	__rcu_init();
 }
 
-/**
- * synchronize_rcu - wait until a grace period has elapsed.
- *
- * Control will return to the caller some time after a full grace
- * period has elapsed, in other words after all currently executing RCU
- * read-side critical sections have completed.  RCU read-side critical
- * sections are delimited by rcu_read_lock() and rcu_read_unlock(),
- * and may be nested.
- *
- * If your read-side code is not protected by rcu_read_lock(), do -not-
- * use synchronize_rcu().
- */
-void synchronize_rcu(void)
-{
-	struct rcu_synchronize rcu;
-
-	init_completion(&rcu.completion);
-	/* Will wake me after RCU finished */
-	call_rcu(&rcu.head, wakeme_after_rcu);
-
-	/* Wait for it */
-	wait_for_completion(&rcu.completion);
-}
-
-module_param(blimit, int, 0);
-module_param(qhimark, int, 0);
-module_param(qlowmark, int, 0);
-EXPORT_SYMBOL_GPL(rcu_batches_completed);
-EXPORT_SYMBOL_GPL(rcu_batches_completed_bh);
-EXPORT_SYMBOL_GPL(call_rcu);
-EXPORT_SYMBOL_GPL(call_rcu_bh);
-EXPORT_SYMBOL_GPL(synchronize_rcu);
diff --git a/kernel/rcupreempt.c b/kernel/rcupreempt.c
new file mode 100644
index 00000000000..987cfb7ade8
--- /dev/null
+++ b/kernel/rcupreempt.c
@@ -0,0 +1,953 @@
+/*
+ * Read-Copy Update mechanism for mutual exclusion, realtime implementation
+ *
+ * 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., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright IBM Corporation, 2006
+ *
+ * Authors: Paul E. McKenney <paulmck@us.ibm.com>
+ *		With thanks to Esben Nielsen, Bill Huey, and Ingo Molnar
+ *		for pushing me away from locks and towards counters, and
+ *		to Suparna Bhattacharya for pushing me completely away
+ *		from atomic instructions on the read side.
+ *
+ * Papers:  http://www.rdrop.com/users/paulmck/RCU
+ *
+ * Design Document: http://lwn.net/Articles/253651/
+ *
+ * For detailed explanation of Read-Copy Update mechanism see -
+ * 		Documentation/RCU/ *.txt
+ *
+ */
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/smp.h>
+#include <linux/rcupdate.h>
+#include <linux/interrupt.h>
+#include <linux/sched.h>
+#include <asm/atomic.h>
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <linux/completion.h>
+#include <linux/moduleparam.h>
+#include <linux/percpu.h>
+#include <linux/notifier.h>
+#include <linux/rcupdate.h>
+#include <linux/cpu.h>
+#include <linux/random.h>
+#include <linux/delay.h>
+#include <linux/byteorder/swabb.h>
+#include <linux/cpumask.h>
+#include <linux/rcupreempt_trace.h>
+
+/*
+ * Macro that prevents the compiler from reordering accesses, but does
+ * absolutely -nothing- to prevent CPUs from reordering.  This is used
+ * only to mediate communication between mainline code and hardware
+ * interrupt and NMI handlers.
+ */
+#define ACCESS_ONCE(x) (*(volatile typeof(x) *)&(x))
+
+/*
+ * PREEMPT_RCU data structures.
+ */
+
+/*
+ * GP_STAGES specifies the number of times the state machine has
+ * to go through the all the rcu_try_flip_states (see below)
+ * in a single Grace Period.
+ *
+ * GP in GP_STAGES stands for Grace Period ;)
+ */
+#define GP_STAGES    2
+struct rcu_data {
+	spinlock_t	lock;		/* Protect rcu_data fields. */
+	long		completed;	/* Number of last completed batch. */
+	int		waitlistcount;
+	struct tasklet_struct rcu_tasklet;
+	struct rcu_head *nextlist;
+	struct rcu_head **nexttail;
+	struct rcu_head *waitlist[GP_STAGES];
+	struct rcu_head **waittail[GP_STAGES];
+	struct rcu_head *donelist;
+	struct rcu_head **donetail;
+	long rcu_flipctr[2];
+#ifdef CONFIG_RCU_TRACE
+	struct rcupreempt_trace trace;
+#endif /* #ifdef CONFIG_RCU_TRACE */
+};
+
+/*
+ * States for rcu_try_flip() and friends.
+ */
+
+enum rcu_try_flip_states {
+
+	/*
+	 * Stay here if nothing is happening. Flip the counter if somthing
+	 * starts happening. Denoted by "I"
+	 */
+	rcu_try_flip_idle_state,
+
+	/*
+	 * Wait here for all CPUs to notice that the counter has flipped. This
+	 * prevents the old set of counters from ever being incremented once
+	 * we leave this state, which in turn is necessary because we cannot
+	 * test any individual counter for zero -- we can only check the sum.
+	 * Denoted by "A".
+	 */
+	rcu_try_flip_waitack_state,
+
+	/*
+	 * Wait here for the sum of the old per-CPU counters to reach zero.
+	 * Denoted by "Z".
+	 */
+	rcu_try_flip_waitzero_state,
+
+	/*
+	 * Wait here for each of the other CPUs to execute a memory barrier.
+	 * This is necessary to ensure that these other CPUs really have
+	 * completed executing their RCU read-side critical sections, despite
+	 * their CPUs wildly reordering memory. Denoted by "M".
+	 */
+	rcu_try_flip_waitmb_state,
+};
+
+struct rcu_ctrlblk {
+	spinlock_t	fliplock;	/* Protect state-machine transitions. */
+	long		completed;	/* Number of last completed batch. */
+	enum rcu_try_flip_states rcu_try_flip_state; /* The current state of
+							the rcu state machine */
+};
+
+static DEFINE_PER_CPU(struct rcu_data, rcu_data);
+static struct rcu_ctrlblk rcu_ctrlblk = {
+	.fliplock = __SPIN_LOCK_UNLOCKED(rcu_ctrlblk.fliplock),
+	.completed = 0,
+	.rcu_try_flip_state = rcu_try_flip_idle_state,
+};
+
+
+#ifdef CONFIG_RCU_TRACE
+static char *rcu_try_flip_state_names[] =
+	{ "idle", "waitack", "waitzero", "waitmb" };
+#endif /* #ifdef CONFIG_RCU_TRACE */
+
+static cpumask_t rcu_cpu_online_map __read_mostly = CPU_MASK_NONE;
+
+/*
+ * Enum and per-CPU flag to determine when each CPU has seen
+ * the most recent counter flip.
+ */
+
+enum rcu_flip_flag_values {
+	rcu_flip_seen,		/* Steady/initial state, last flip seen. */
+				/* Only GP detector can update. */
+	rcu_flipped		/* Flip just completed, need confirmation. */
+				/* Only corresponding CPU can update. */
+};
+static DEFINE_PER_CPU_SHARED_ALIGNED(enum rcu_flip_flag_values, rcu_flip_flag)
+								= rcu_flip_seen;
+
+/*
+ * Enum and per-CPU flag to determine when each CPU has executed the
+ * needed memory barrier to fence in memory references from its last RCU
+ * read-side critical section in the just-completed grace period.
+ */
+
+enum rcu_mb_flag_values {
+	rcu_mb_done,		/* Steady/initial state, no mb()s required. */
+				/* Only GP detector can update. */
+	rcu_mb_needed		/* Flip just completed, need an mb(). */
+				/* Only corresponding CPU can update. */
+};
+static DEFINE_PER_CPU_SHARED_ALIGNED(enum rcu_mb_flag_values, rcu_mb_flag)
+								= rcu_mb_done;
+
+/*
+ * RCU_DATA_ME: find the current CPU's rcu_data structure.
+ * RCU_DATA_CPU: find the specified CPU's rcu_data structure.
+ */
+#define RCU_DATA_ME()		(&__get_cpu_var(rcu_data))
+#define RCU_DATA_CPU(cpu)	(&per_cpu(rcu_data, cpu))
+
+/*
+ * Helper macro for tracing when the appropriate rcu_data is not
+ * cached in a local variable, but where the CPU number is so cached.
+ */
+#define RCU_TRACE_CPU(f, cpu) RCU_TRACE(f, &(RCU_DATA_CPU(cpu)->trace));
+
+/*
+ * Helper macro for tracing when the appropriate rcu_data is not
+ * cached in a local variable.
+ */
+#define RCU_TRACE_ME(f) RCU_TRACE(f, &(RCU_DATA_ME()->trace));
+
+/*
+ * Helper macro for tracing when the appropriate rcu_data is pointed
+ * to by a local variable.
+ */
+#define RCU_TRACE_RDP(f, rdp) RCU_TRACE(f, &((rdp)->trace));
+
+/*
+ * Return the number of RCU batches processed thus far.  Useful
+ * for debug and statistics.
+ */
+long rcu_batches_completed(void)
+{
+	return rcu_ctrlblk.completed;
+}
+EXPORT_SYMBOL_GPL(rcu_batches_completed);
+
+EXPORT_SYMBOL_GPL(rcu_batches_completed_bh);
+
+void __rcu_read_lock(void)
+{
+	int idx;
+	struct task_struct *t = current;
+	int nesting;
+
+	nesting = ACCESS_ONCE(t->rcu_read_lock_nesting);
+	if (nesting != 0) {
+
+		/* An earlier rcu_read_lock() covers us, just count it. */
+
+		t->rcu_read_lock_nesting = nesting + 1;
+
+	} else {
+		unsigned long flags;
+
+		/*
+		 * We disable interrupts for the following reasons:
+		 * - If we get scheduling clock interrupt here, and we
+		 *   end up acking the counter flip, it's like a promise
+		 *   that we will never increment the old counter again.
+		 *   Thus we will break that promise if that
+		 *   scheduling clock interrupt happens between the time
+		 *   we pick the .completed field and the time that we
+		 *   increment our counter.
+		 *
+		 * - We don't want to be preempted out here.
+		 *
+		 * NMIs can still occur, of course, and might themselves
+		 * contain rcu_read_lock().
+		 */
+
+		local_irq_save(flags);
+
+		/*
+		 * Outermost nesting of rcu_read_lock(), so increment
+		 * the current counter for the current CPU.  Use volatile
+		 * casts to prevent the compiler from reordering.
+		 */
+
+		idx = ACCESS_ONCE(rcu_ctrlblk.completed) & 0x1;
+		ACCESS_ONCE(RCU_DATA_ME()->rcu_flipctr[idx])++;
+
+		/*
+		 * Now that the per-CPU counter has been incremented, we
+		 * are protected from races with rcu_read_lock() invoked
+		 * from NMI handlers on this CPU.  We can therefore safely
+		 * increment the nesting counter, relieving further NMIs
+		 * of the need to increment the per-CPU counter.
+		 */
+
+		ACCESS_ONCE(t->rcu_read_lock_nesting) = nesting + 1;
+
+		/*
+		 * Now that we have preventing any NMIs from storing
+		 * to the ->rcu_flipctr_idx, we can safely use it to
+		 * remember which counter to decrement in the matching
+		 * rcu_read_unlock().
+		 */
+
+		ACCESS_ONCE(t->rcu_flipctr_idx) = idx;
+		local_irq_restore(flags);
+	}
+}
+EXPORT_SYMBOL_GPL(__rcu_read_lock);
+
+void __rcu_read_unlock(void)
+{
+	int idx;
+	struct task_struct *t = current;
+	int nesting;
+
+	nesting = ACCESS_ONCE(t->rcu_read_lock_nesting);
+	if (nesting > 1) {
+
+		/*
+		 * We are still protected by the enclosing rcu_read_lock(),
+		 * so simply decrement the counter.
+		 */
+
+		t->rcu_read_lock_nesting = nesting - 1;
+
+	} else {
+		unsigned long flags;
+
+		/*
+		 * Disable local interrupts to prevent the grace-period
+		 * detection state machine from seeing us half-done.
+		 * NMIs can still occur, of course, and might themselves
+		 * contain rcu_read_lock() and rcu_read_unlock().
+		 */
+
+		local_irq_save(flags);
+
+		/*
+		 * Outermost nesting of rcu_read_unlock(), so we must
+		 * decrement the current counter for the current CPU.
+		 * This must be done carefully, because NMIs can
+		 * occur at any point in this code, and any rcu_read_lock()
+		 * and rcu_read_unlock() pairs in the NMI handlers
+		 * must interact non-destructively with this code.
+		 * Lots of volatile casts, and -very- careful ordering.
+		 *
+		 * Changes to this code, including this one, must be
+		 * inspected, validated, and tested extremely carefully!!!
+		 */
+
+		/*
+		 * First, pick up the index.
+		 */
+
+		idx = ACCESS_ONCE(t->rcu_flipctr_idx);
+
+		/*
+		 * Now that we have fetched the counter index, it is
+		 * safe to decrement the per-task RCU nesting counter.
+		 * After this, any interrupts or NMIs will increment and
+		 * decrement the per-CPU counters.
+		 */
+		ACCESS_ONCE(t->rcu_read_lock_nesting) = nesting - 1;
+
+		/*
+		 * It is now safe to decrement this task's nesting count.
+		 * NMIs that occur after this statement will route their
+		 * rcu_read_lock() calls through this "else" clause, and
+		 * will thus start incrementing the per-CPU counter on
+		 * their own.  They will also clobber ->rcu_flipctr_idx,
+		 * but that is OK, since we have already fetched it.
+		 */
+
+		ACCESS_ONCE(RCU_DATA_ME()->rcu_flipctr[idx])--;
+		local_irq_restore(flags);
+	}
+}
+EXPORT_SYMBOL_GPL(__rcu_read_unlock);
+
+/*
+ * If a global counter flip has occurred since the last time that we
+ * advanced callbacks, advance them.  Hardware interrupts must be
+ * disabled when calling this function.
+ */
+static void __rcu_advance_callbacks(struct rcu_data *rdp)
+{
+	int cpu;
+	int i;
+	int wlc = 0;
+
+	if (rdp->completed != rcu_ctrlblk.completed) {
+		if (rdp->waitlist[GP_STAGES - 1] != NULL) {
+			*rdp->donetail = rdp->waitlist[GP_STAGES - 1];
+			rdp->donetail = rdp->waittail[GP_STAGES - 1];
+			RCU_TRACE_RDP(rcupreempt_trace_move2done, rdp);
+		}
+		for (i = GP_STAGES - 2; i >= 0; i--) {
+			if (rdp->waitlist[i] != NULL) {
+				rdp->waitlist[i + 1] = rdp->waitlist[i];
+				rdp->waittail[i + 1] = rdp->waittail[i];
+				wlc++;
+			} else {
+				rdp->waitlist[i + 1] = NULL;
+				rdp->waittail[i + 1] =
+					&rdp->waitlist[i + 1];
+			}
+		}
+		if (rdp->nextlist != NULL) {
+			rdp->waitlist[0] = rdp->nextlist;
+			rdp->waittail[0] = rdp->nexttail;
+			wlc++;
+			rdp->nextlist = NULL;
+			rdp->nexttail = &rdp->nextlist;
+			RCU_TRACE_RDP(rcupreempt_trace_move2wait, rdp);
+		} else {
+			rdp->waitlist[0] = NULL;
+			rdp->waittail[0] = &rdp->waitlist[0];
+		}
+		rdp->waitlistcount = wlc;
+		rdp->completed = rcu_ctrlblk.completed;
+	}
+
+	/*
+	 * Check to see if this CPU needs to report that it has seen
+	 * the most recent counter flip, thereby declaring that all
+	 * subsequent rcu_read_lock() invocations will respect this flip.
+	 */
+
+	cpu = raw_smp_processor_id();
+	if (per_cpu(rcu_flip_flag, cpu) == rcu_flipped) {
+		smp_mb();  /* Subsequent counter accesses must see new value */
+		per_cpu(rcu_flip_flag, cpu) = rcu_flip_seen;
+		smp_mb();  /* Subsequent RCU read-side critical sections */
+			   /*  seen -after- acknowledgement. */
+	}
+}
+
+/*
+ * Get here when RCU is idle.  Decide whether we need to
+ * move out of idle state, and return non-zero if so.
+ * "Straightforward" approach for the moment, might later
+ * use callback-list lengths, grace-period duration, or
+ * some such to determine when to exit idle state.
+ * Might also need a pre-idle test that does not acquire
+ * the lock, but let's get the simple case working first...
+ */
+
+static int
+rcu_try_flip_idle(void)
+{
+	int cpu;
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_i1);
+	if (!rcu_pending(smp_processor_id())) {
+		RCU_TRACE_ME(rcupreempt_trace_try_flip_ie1);
+		return 0;
+	}
+
+	/*
+	 * Do the flip.
+	 */
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_g1);
+	rcu_ctrlblk.completed++;  /* stands in for rcu_try_flip_g2 */
+
+	/*
+	 * Need a memory barrier so that other CPUs see the new
+	 * counter value before they see the subsequent change of all
+	 * the rcu_flip_flag instances to rcu_flipped.
+	 */
+
+	smp_mb();	/* see above block comment. */
+
+	/* Now ask each CPU for acknowledgement of the flip. */
+
+	for_each_cpu_mask(cpu, rcu_cpu_online_map)
+		per_cpu(rcu_flip_flag, cpu) = rcu_flipped;
+
+	return 1;
+}
+
+/*
+ * Wait for CPUs to acknowledge the flip.
+ */
+
+static int
+rcu_try_flip_waitack(void)
+{
+	int cpu;
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_a1);
+	for_each_cpu_mask(cpu, rcu_cpu_online_map)
+		if (per_cpu(rcu_flip_flag, cpu) != rcu_flip_seen) {
+			RCU_TRACE_ME(rcupreempt_trace_try_flip_ae1);
+			return 0;
+		}
+
+	/*
+	 * Make sure our checks above don't bleed into subsequent
+	 * waiting for the sum of the counters to reach zero.
+	 */
+
+	smp_mb();	/* see above block comment. */
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_a2);
+	return 1;
+}
+
+/*
+ * Wait for collective ``last'' counter to reach zero,
+ * then tell all CPUs to do an end-of-grace-period memory barrier.
+ */
+
+static int
+rcu_try_flip_waitzero(void)
+{
+	int cpu;
+	int lastidx = !(rcu_ctrlblk.completed & 0x1);
+	int sum = 0;
+
+	/* Check to see if the sum of the "last" counters is zero. */
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_z1);
+	for_each_cpu_mask(cpu, rcu_cpu_online_map)
+		sum += RCU_DATA_CPU(cpu)->rcu_flipctr[lastidx];
+	if (sum != 0) {
+		RCU_TRACE_ME(rcupreempt_trace_try_flip_ze1);
+		return 0;
+	}
+
+	/*
+	 * This ensures that the other CPUs see the call for
+	 * memory barriers -after- the sum to zero has been
+	 * detected here
+	 */
+	smp_mb();  /*  ^^^^^^^^^^^^ */
+
+	/* Call for a memory barrier from each CPU. */
+	for_each_cpu_mask(cpu, rcu_cpu_online_map)
+		per_cpu(rcu_mb_flag, cpu) = rcu_mb_needed;
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_z2);
+	return 1;
+}
+
+/*
+ * Wait for all CPUs to do their end-of-grace-period memory barrier.
+ * Return 0 once all CPUs have done so.
+ */
+
+static int
+rcu_try_flip_waitmb(void)
+{
+	int cpu;
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_m1);
+	for_each_cpu_mask(cpu, rcu_cpu_online_map)
+		if (per_cpu(rcu_mb_flag, cpu) != rcu_mb_done) {
+			RCU_TRACE_ME(rcupreempt_trace_try_flip_me1);
+			return 0;
+		}
+
+	smp_mb(); /* Ensure that the above checks precede any following flip. */
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_m2);
+	return 1;
+}
+
+/*
+ * Attempt a single flip of the counters.  Remember, a single flip does
+ * -not- constitute a grace period.  Instead, the interval between
+ * at least GP_STAGES consecutive flips is a grace period.
+ *
+ * If anyone is nuts enough to run this CONFIG_PREEMPT_RCU implementation
+ * on a large SMP, they might want to use a hierarchical organization of
+ * the per-CPU-counter pairs.
+ */
+static void rcu_try_flip(void)
+{
+	unsigned long flags;
+
+	RCU_TRACE_ME(rcupreempt_trace_try_flip_1);
+	if (unlikely(!spin_trylock_irqsave(&rcu_ctrlblk.fliplock, flags))) {
+		RCU_TRACE_ME(rcupreempt_trace_try_flip_e1);
+		return;
+	}
+
+	/*
+	 * Take the next transition(s) through the RCU grace-period
+	 * flip-counter state machine.
+	 */
+
+	switch (rcu_ctrlblk.rcu_try_flip_state) {
+	case rcu_try_flip_idle_state:
+		if (rcu_try_flip_idle())
+			rcu_ctrlblk.rcu_try_flip_state =
+				rcu_try_flip_waitack_state;
+		break;
+	case rcu_try_flip_waitack_state:
+		if (rcu_try_flip_waitack())
+			rcu_ctrlblk.rcu_try_flip_state =
+				rcu_try_flip_waitzero_state;
+		break;
+	case rcu_try_flip_waitzero_state:
+		if (rcu_try_flip_waitzero())
+			rcu_ctrlblk.rcu_try_flip_state =
+				rcu_try_flip_waitmb_state;
+		break;
+	case rcu_try_flip_waitmb_state:
+		if (rcu_try_flip_waitmb())
+			rcu_ctrlblk.rcu_try_flip_state =
+				rcu_try_flip_idle_state;
+	}
+	spin_unlock_irqrestore(&rcu_ctrlblk.fliplock, flags);
+}
+
+/*
+ * Check to see if this CPU needs to do a memory barrier in order to
+ * ensure that any prior RCU read-side critical sections have committed
+ * their counter manipulations and critical-section memory references
+ * before declaring the grace period to be completed.
+ */
+static void rcu_check_mb(int cpu)
+{
+	if (per_cpu(rcu_mb_flag, cpu) == rcu_mb_needed) {
+		smp_mb();  /* Ensure RCU read-side accesses are visible. */
+		per_cpu(rcu_mb_flag, cpu) = rcu_mb_done;
+	}
+}
+
+void rcu_check_callbacks(int cpu, int user)
+{
+	unsigned long flags;
+	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
+
+	rcu_check_mb(cpu);
+	if (rcu_ctrlblk.completed == rdp->completed)
+		rcu_try_flip();
+	spin_lock_irqsave(&rdp->lock, flags);
+	RCU_TRACE_RDP(rcupreempt_trace_check_callbacks, rdp);
+	__rcu_advance_callbacks(rdp);
+	if (rdp->donelist == NULL) {
+		spin_unlock_irqrestore(&rdp->lock, flags);
+	} else {
+		spin_unlock_irqrestore(&rdp->lock, flags);
+		raise_softirq(RCU_SOFTIRQ);
+	}
+}
+
+/*
+ * Needed by dynticks, to make sure all RCU processing has finished
+ * when we go idle:
+ */
+void rcu_advance_callbacks(int cpu, int user)
+{
+	unsigned long flags;
+	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
+
+	if (rcu_ctrlblk.completed == rdp->completed) {
+		rcu_try_flip();
+		if (rcu_ctrlblk.completed == rdp->completed)
+			return;
+	}
+	spin_lock_irqsave(&rdp->lock, flags);
+	RCU_TRACE_RDP(rcupreempt_trace_check_callbacks, rdp);
+	__rcu_advance_callbacks(rdp);
+	spin_unlock_irqrestore(&rdp->lock, flags);
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+#define rcu_offline_cpu_enqueue(srclist, srctail, dstlist, dsttail) do { \
+		*dsttail = srclist; \
+		if (srclist != NULL) { \
+			dsttail = srctail; \
+			srclist = NULL; \
+			srctail = &srclist;\
+		} \
+	} while (0)
+
+void rcu_offline_cpu(int cpu)
+{
+	int i;
+	struct rcu_head *list = NULL;
+	unsigned long flags;
+	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
+	struct rcu_head **tail = &list;
+
+	/*
+	 * Remove all callbacks from the newly dead CPU, retaining order.
+	 * Otherwise rcu_barrier() will fail
+	 */
+
+	spin_lock_irqsave(&rdp->lock, flags);
+	rcu_offline_cpu_enqueue(rdp->donelist, rdp->donetail, list, tail);
+	for (i = GP_STAGES - 1; i >= 0; i--)
+		rcu_offline_cpu_enqueue(rdp->waitlist[i], rdp->waittail[i],
+						list, tail);
+	rcu_offline_cpu_enqueue(rdp->nextlist, rdp->nexttail, list, tail);
+	spin_unlock_irqrestore(&rdp->lock, flags);
+	rdp->waitlistcount = 0;
+
+	/* Disengage the newly dead CPU from the grace-period computation. */
+
+	spin_lock_irqsave(&rcu_ctrlblk.fliplock, flags);
+	rcu_check_mb(cpu);
+	if (per_cpu(rcu_flip_flag, cpu) == rcu_flipped) {
+		smp_mb();  /* Subsequent counter accesses must see new value */
+		per_cpu(rcu_flip_flag, cpu) = rcu_flip_seen;
+		smp_mb();  /* Subsequent RCU read-side critical sections */
+			   /*  seen -after- acknowledgement. */
+	}
+
+	RCU_DATA_ME()->rcu_flipctr[0] += RCU_DATA_CPU(cpu)->rcu_flipctr[0];
+	RCU_DATA_ME()->rcu_flipctr[1] += RCU_DATA_CPU(cpu)->rcu_flipctr[1];
+
+	RCU_DATA_CPU(cpu)->rcu_flipctr[0] = 0;
+	RCU_DATA_CPU(cpu)->rcu_flipctr[1] = 0;
+
+	cpu_clear(cpu, rcu_cpu_online_map);
+
+	spin_unlock_irqrestore(&rcu_ctrlblk.fliplock, flags);
+
+	/*
+	 * Place the removed callbacks on the current CPU's queue.
+	 * Make them all start a new grace period: simple approach,
+	 * in theory could starve a given set of callbacks, but
+	 * you would need to be doing some serious CPU hotplugging
+	 * to make this happen.  If this becomes a problem, adding
+	 * a synchronize_rcu() to the hotplug path would be a simple
+	 * fix.
+	 */
+
+	rdp = RCU_DATA_ME();
+	spin_lock_irqsave(&rdp->lock, flags);
+	*rdp->nexttail = list;
+	if (list)
+		rdp->nexttail = tail;
+	spin_unlock_irqrestore(&rdp->lock, flags);
+}
+
+void __devinit rcu_online_cpu(int cpu)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rcu_ctrlblk.fliplock, flags);
+	cpu_set(cpu, rcu_cpu_online_map);
+	spin_unlock_irqrestore(&rcu_ctrlblk.fliplock, flags);
+}
+
+#else /* #ifdef CONFIG_HOTPLUG_CPU */
+
+void rcu_offline_cpu(int cpu)
+{
+}
+
+void __devinit rcu_online_cpu(int cpu)
+{
+}
+
+#endif /* #else #ifdef CONFIG_HOTPLUG_CPU */
+
+static void rcu_process_callbacks(struct softirq_action *unused)
+{
+	unsigned long flags;
+	struct rcu_head *next, *list;
+	struct rcu_data *rdp = RCU_DATA_ME();
+
+	spin_lock_irqsave(&rdp->lock, flags);
+	list = rdp->donelist;
+	if (list == NULL) {
+		spin_unlock_irqrestore(&rdp->lock, flags);
+		return;
+	}
+	rdp->donelist = NULL;
+	rdp->donetail = &rdp->donelist;
+	RCU_TRACE_RDP(rcupreempt_trace_done_remove, rdp);
+	spin_unlock_irqrestore(&rdp->lock, flags);
+	while (list) {
+		next = list->next;
+		list->func(list);
+		list = next;
+		RCU_TRACE_ME(rcupreempt_trace_invoke);
+	}
+}
+
+void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *rcu))
+{
+	unsigned long flags;
+	struct rcu_data *rdp;
+
+	head->func = func;
+	head->next = NULL;
+	local_irq_save(flags);
+	rdp = RCU_DATA_ME();
+	spin_lock(&rdp->lock);
+	__rcu_advance_callbacks(rdp);
+	*rdp->nexttail = head;
+	rdp->nexttail = &head->next;
+	RCU_TRACE_RDP(rcupreempt_trace_next_add, rdp);
+	spin_unlock(&rdp->lock);
+	local_irq_restore(flags);
+}
+EXPORT_SYMBOL_GPL(call_rcu);
+
+/*
+ * Wait until all currently running preempt_disable() code segments
+ * (including hardware-irq-disable segments) complete.  Note that
+ * in -rt this does -not- necessarily result in all currently executing
+ * interrupt -handlers- having completed.
+ */
+void __synchronize_sched(void)
+{
+	cpumask_t oldmask;
+	int cpu;
+
+	if (sched_getaffinity(0, &oldmask) < 0)
+		oldmask = cpu_possible_map;
+	for_each_online_cpu(cpu) {
+		sched_setaffinity(0, cpumask_of_cpu(cpu));
+		schedule();
+	}
+	sched_setaffinity(0, oldmask);
+}
+EXPORT_SYMBOL_GPL(__synchronize_sched);
+
+/*
+ * Check to see if any future RCU-related work will need to be done
+ * by the current CPU, even if none need be done immediately, returning
+ * 1 if so.  Assumes that notifiers would take care of handling any
+ * outstanding requests from the RCU core.
+ *
+ * This function is part of the RCU implementation; it is -not-
+ * an exported member of the RCU API.
+ */
+int rcu_needs_cpu(int cpu)
+{
+	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
+
+	return (rdp->donelist != NULL ||
+		!!rdp->waitlistcount ||
+		rdp->nextlist != NULL);
+}
+
+int rcu_pending(int cpu)
+{
+	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
+
+	/* The CPU has at least one callback queued somewhere. */
+
+	if (rdp->donelist != NULL ||
+	    !!rdp->waitlistcount ||
+	    rdp->nextlist != NULL)
+		return 1;
+
+	/* The RCU core needs an acknowledgement from this CPU. */
+
+	if ((per_cpu(rcu_flip_flag, cpu) == rcu_flipped) ||
+	    (per_cpu(rcu_mb_flag, cpu) == rcu_mb_needed))
+		return 1;
+
+	/* This CPU has fallen behind the global grace-period number. */
+
+	if (rdp->completed != rcu_ctrlblk.completed)
+		return 1;
+
+	/* Nothing needed from this CPU. */
+
+	return 0;
+}
+
+static int __cpuinit rcu_cpu_notify(struct notifier_block *self,
+				unsigned long action, void *hcpu)
+{
+	long cpu = (long)hcpu;
+
+	switch (action) {
+	case CPU_UP_PREPARE:
+	case CPU_UP_PREPARE_FROZEN:
+		rcu_online_cpu(cpu);
+		break;
+	case CPU_UP_CANCELED:
+	case CPU_UP_CANCELED_FROZEN:
+	case CPU_DEAD:
+	case CPU_DEAD_FROZEN:
+		rcu_offline_cpu(cpu);
+		break;
+	default:
+		break;
+	}
+	return NOTIFY_OK;
+}
+
+static struct notifier_block __cpuinitdata rcu_nb = {
+	.notifier_call = rcu_cpu_notify,
+};
+
+void __init __rcu_init(void)
+{
+	int cpu;
+	int i;
+	struct rcu_data *rdp;
+
+	printk(KERN_NOTICE "Preemptible RCU implementation.\n");
+	for_each_possible_cpu(cpu) {
+		rdp = RCU_DATA_CPU(cpu);
+		spin_lock_init(&rdp->lock);
+		rdp->completed = 0;
+		rdp->waitlistcount = 0;
+		rdp->nextlist = NULL;
+		rdp->nexttail = &rdp->nextlist;
+		for (i = 0; i < GP_STAGES; i++) {
+			rdp->waitlist[i] = NULL;
+			rdp->waittail[i] = &rdp->waitlist[i];
+		}
+		rdp->donelist = NULL;
+		rdp->donetail = &rdp->donelist;
+		rdp->rcu_flipctr[0] = 0;
+		rdp->rcu_flipctr[1] = 0;
+	}
+	register_cpu_notifier(&rcu_nb);
+
+	/*
+	 * We don't need protection against CPU-Hotplug here
+	 * since
+	 * a) If a CPU comes online while we are iterating over the
+	 *    cpu_online_map below, we would only end up making a
+	 *    duplicate call to rcu_online_cpu() which sets the corresponding
+	 *    CPU's mask in the rcu_cpu_online_map.
+	 *
+	 * b) A CPU cannot go offline at this point in time since the user
+	 *    does not have access to the sysfs interface, nor do we
+	 *    suspend the system.
+	 */
+	for_each_online_cpu(cpu)
+		rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE,	(void *)(long) cpu);
+
+	open_softirq(RCU_SOFTIRQ, rcu_process_callbacks, NULL);
+}
+
+/*
+ * Deprecated, use synchronize_rcu() or synchronize_sched() instead.
+ */
+void synchronize_kernel(void)
+{
+	synchronize_rcu();
+}
+
+#ifdef CONFIG_RCU_TRACE
+long *rcupreempt_flipctr(int cpu)
+{
+	return &RCU_DATA_CPU(cpu)->rcu_flipctr[0];
+}
+EXPORT_SYMBOL_GPL(rcupreempt_flipctr);
+
+int rcupreempt_flip_flag(int cpu)
+{
+	return per_cpu(rcu_flip_flag, cpu);
+}
+EXPORT_SYMBOL_GPL(rcupreempt_flip_flag);
+
+int rcupreempt_mb_flag(int cpu)
+{
+	return per_cpu(rcu_mb_flag, cpu);
+}
+EXPORT_SYMBOL_GPL(rcupreempt_mb_flag);
+
+char *rcupreempt_try_flip_state_name(void)
+{
+	return rcu_try_flip_state_names[rcu_ctrlblk.rcu_try_flip_state];
+}
+EXPORT_SYMBOL_GPL(rcupreempt_try_flip_state_name);
+
+struct rcupreempt_trace *rcupreempt_trace_cpu(int cpu)
+{
+	struct rcu_data *rdp = RCU_DATA_CPU(cpu);
+
+	return &rdp->trace;
+}
+EXPORT_SYMBOL_GPL(rcupreempt_trace_cpu);
+
+#endif /* #ifdef RCU_TRACE */
diff --git a/kernel/rcupreempt_trace.c b/kernel/rcupreempt_trace.c
new file mode 100644
index 00000000000..49ac4947af2
--- /dev/null
+++ b/kernel/rcupreempt_trace.c
@@ -0,0 +1,330 @@
+/*
+ * Read-Copy Update tracing for realtime implementation
+ *
+ * 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., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright IBM Corporation, 2006
+ *
+ * Papers:  http://www.rdrop.com/users/paulmck/RCU
+ *
+ * For detailed explanation of Read-Copy Update mechanism see -
+ * 		Documentation/RCU/ *.txt
+ *
+ */
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/smp.h>
+#include <linux/rcupdate.h>
+#include <linux/interrupt.h>
+#include <linux/sched.h>
+#include <asm/atomic.h>
+#include <linux/bitops.h>
+#include <linux/module.h>
+#include <linux/completion.h>
+#include <linux/moduleparam.h>
+#include <linux/percpu.h>
+#include <linux/notifier.h>
+#include <linux/rcupdate.h>
+#include <linux/cpu.h>
+#include <linux/mutex.h>
+#include <linux/rcupreempt_trace.h>
+#include <linux/debugfs.h>
+
+static struct mutex rcupreempt_trace_mutex;
+static char *rcupreempt_trace_buf;
+#define RCUPREEMPT_TRACE_BUF_SIZE 4096
+
+void rcupreempt_trace_move2done(struct rcupreempt_trace *trace)
+{
+	trace->done_length += trace->wait_length;
+	trace->done_add += trace->wait_length;
+	trace->wait_length = 0;
+}
+void rcupreempt_trace_move2wait(struct rcupreempt_trace *trace)
+{
+	trace->wait_length += trace->next_length;
+	trace->wait_add += trace->next_length;
+	trace->next_length = 0;
+}
+void rcupreempt_trace_try_flip_1(struct rcupreempt_trace *trace)
+{
+	atomic_inc(&trace->rcu_try_flip_1);
+}
+void rcupreempt_trace_try_flip_e1(struct rcupreempt_trace *trace)
+{
+	atomic_inc(&trace->rcu_try_flip_e1);
+}
+void rcupreempt_trace_try_flip_i1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_i1++;
+}
+void rcupreempt_trace_try_flip_ie1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_ie1++;
+}
+void rcupreempt_trace_try_flip_g1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_g1++;
+}
+void rcupreempt_trace_try_flip_a1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_a1++;
+}
+void rcupreempt_trace_try_flip_ae1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_ae1++;
+}
+void rcupreempt_trace_try_flip_a2(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_a2++;
+}
+void rcupreempt_trace_try_flip_z1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_z1++;
+}
+void rcupreempt_trace_try_flip_ze1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_ze1++;
+}
+void rcupreempt_trace_try_flip_z2(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_z2++;
+}
+void rcupreempt_trace_try_flip_m1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_m1++;
+}
+void rcupreempt_trace_try_flip_me1(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_me1++;
+}
+void rcupreempt_trace_try_flip_m2(struct rcupreempt_trace *trace)
+{
+	trace->rcu_try_flip_m2++;
+}
+void rcupreempt_trace_check_callbacks(struct rcupreempt_trace *trace)
+{
+	trace->rcu_check_callbacks++;
+}
+void rcupreempt_trace_done_remove(struct rcupreempt_trace *trace)
+{
+	trace->done_remove += trace->done_length;
+	trace->done_length = 0;
+}
+void rcupreempt_trace_invoke(struct rcupreempt_trace *trace)
+{
+	atomic_inc(&trace->done_invoked);
+}
+void rcupreempt_trace_next_add(struct rcupreempt_trace *trace)
+{
+	trace->next_add++;
+	trace->next_length++;
+}
+
+static void rcupreempt_trace_sum(struct rcupreempt_trace *sp)
+{
+	struct rcupreempt_trace *cp;
+	int cpu;
+
+	memset(sp, 0, sizeof(*sp));
+	for_each_possible_cpu(cpu) {
+		cp = rcupreempt_trace_cpu(cpu);
+		sp->next_length += cp->next_length;
+		sp->next_add += cp->next_add;
+		sp->wait_length += cp->wait_length;
+		sp->wait_add += cp->wait_add;
+		sp->done_length += cp->done_length;
+		sp->done_add += cp->done_add;
+		sp->done_remove += cp->done_remove;
+		atomic_set(&sp->done_invoked, atomic_read(&cp->done_invoked));
+		sp->rcu_check_callbacks += cp->rcu_check_callbacks;
+		atomic_set(&sp->rcu_try_flip_1,
+			   atomic_read(&cp->rcu_try_flip_1));
+		atomic_set(&sp->rcu_try_flip_e1,
+			   atomic_read(&cp->rcu_try_flip_e1));
+		sp->rcu_try_flip_i1 += cp->rcu_try_flip_i1;
+		sp->rcu_try_flip_ie1 += cp->rcu_try_flip_ie1;
+		sp->rcu_try_flip_g1 += cp->rcu_try_flip_g1;
+		sp->rcu_try_flip_a1 += cp->rcu_try_flip_a1;
+		sp->rcu_try_flip_ae1 += cp->rcu_try_flip_ae1;
+		sp->rcu_try_flip_a2 += cp->rcu_try_flip_a2;
+		sp->rcu_try_flip_z1 += cp->rcu_try_flip_z1;
+		sp->rcu_try_flip_ze1 += cp->rcu_try_flip_ze1;
+		sp->rcu_try_flip_z2 += cp->rcu_try_flip_z2;
+		sp->rcu_try_flip_m1 += cp->rcu_try_flip_m1;
+		sp->rcu_try_flip_me1 += cp->rcu_try_flip_me1;
+		sp->rcu_try_flip_m2 += cp->rcu_try_flip_m2;
+	}
+}
+
+static ssize_t rcustats_read(struct file *filp, char __user *buffer,
+				size_t count, loff_t *ppos)
+{
+	struct rcupreempt_trace trace;
+	ssize_t bcount;
+	int cnt = 0;
+
+	rcupreempt_trace_sum(&trace);
+	mutex_lock(&rcupreempt_trace_mutex);
+	snprintf(&rcupreempt_trace_buf[cnt], RCUPREEMPT_TRACE_BUF_SIZE - cnt,
+		 "ggp=%ld rcc=%ld\n",
+		 rcu_batches_completed(),
+		 trace.rcu_check_callbacks);
+	snprintf(&rcupreempt_trace_buf[cnt], RCUPREEMPT_TRACE_BUF_SIZE - cnt,
+		 "na=%ld nl=%ld wa=%ld wl=%ld da=%ld dl=%ld dr=%ld di=%d\n"
+		 "1=%d e1=%d i1=%ld ie1=%ld g1=%ld a1=%ld ae1=%ld a2=%ld\n"
+		 "z1=%ld ze1=%ld z2=%ld m1=%ld me1=%ld m2=%ld\n",
+
+		 trace.next_add, trace.next_length,
+		 trace.wait_add, trace.wait_length,
+		 trace.done_add, trace.done_length,
+		 trace.done_remove, atomic_read(&trace.done_invoked),
+		 atomic_read(&trace.rcu_try_flip_1),
+		 atomic_read(&trace.rcu_try_flip_e1),
+		 trace.rcu_try_flip_i1, trace.rcu_try_flip_ie1,
+		 trace.rcu_try_flip_g1,
+		 trace.rcu_try_flip_a1, trace.rcu_try_flip_ae1,
+			 trace.rcu_try_flip_a2,
+		 trace.rcu_try_flip_z1, trace.rcu_try_flip_ze1,
+			 trace.rcu_try_flip_z2,
+		 trace.rcu_try_flip_m1, trace.rcu_try_flip_me1,
+			trace.rcu_try_flip_m2);
+	bcount = simple_read_from_buffer(buffer, count, ppos,
+			rcupreempt_trace_buf, strlen(rcupreempt_trace_buf));
+	mutex_unlock(&rcupreempt_trace_mutex);
+	return bcount;
+}
+
+static ssize_t rcugp_read(struct file *filp, char __user *buffer,
+				size_t count, loff_t *ppos)
+{
+	long oldgp = rcu_batches_completed();
+	ssize_t bcount;
+
+	mutex_lock(&rcupreempt_trace_mutex);
+	synchronize_rcu();
+	snprintf(rcupreempt_trace_buf, RCUPREEMPT_TRACE_BUF_SIZE,
+		"oldggp=%ld  newggp=%ld\n", oldgp, rcu_batches_completed());
+	bcount = simple_read_from_buffer(buffer, count, ppos,
+			rcupreempt_trace_buf, strlen(rcupreempt_trace_buf));
+	mutex_unlock(&rcupreempt_trace_mutex);
+	return bcount;
+}
+
+static ssize_t rcuctrs_read(struct file *filp, char __user *buffer,
+				size_t count, loff_t *ppos)
+{
+	int cnt = 0;
+	int cpu;
+	int f = rcu_batches_completed() & 0x1;
+	ssize_t bcount;
+
+	mutex_lock(&rcupreempt_trace_mutex);
+
+	cnt += snprintf(&rcupreempt_trace_buf[cnt], RCUPREEMPT_TRACE_BUF_SIZE,
+				"CPU last cur F M\n");
+	for_each_online_cpu(cpu) {
+		long *flipctr = rcupreempt_flipctr(cpu);
+		cnt += snprintf(&rcupreempt_trace_buf[cnt],
+				RCUPREEMPT_TRACE_BUF_SIZE - cnt,
+					"%3d %4ld %3ld %d %d\n",
+			       cpu,
+			       flipctr[!f],
+			       flipctr[f],
+			       rcupreempt_flip_flag(cpu),
+			       rcupreempt_mb_flag(cpu));
+	}
+	cnt += snprintf(&rcupreempt_trace_buf[cnt],
+			RCUPREEMPT_TRACE_BUF_SIZE - cnt,
+			"ggp = %ld, state = %s\n",
+			rcu_batches_completed(),
+			rcupreempt_try_flip_state_name());
+	cnt += snprintf(&rcupreempt_trace_buf[cnt],
+			RCUPREEMPT_TRACE_BUF_SIZE - cnt,
+			"\n");
+	bcount = simple_read_from_buffer(buffer, count, ppos,
+			rcupreempt_trace_buf, strlen(rcupreempt_trace_buf));
+	mutex_unlock(&rcupreempt_trace_mutex);
+	return bcount;
+}
+
+static struct file_operations rcustats_fops = {
+	.owner = THIS_MODULE,
+	.read = rcustats_read,
+};
+
+static struct file_operations rcugp_fops = {
+	.owner = THIS_MODULE,
+	.read = rcugp_read,
+};
+
+static struct file_operations rcuctrs_fops = {
+	.owner = THIS_MODULE,
+	.read = rcuctrs_read,
+};
+
+static struct dentry *rcudir, *statdir, *ctrsdir, *gpdir;
+static int rcupreempt_debugfs_init(void)
+{
+	rcudir = debugfs_create_dir("rcu", NULL);
+	if (!rcudir)
+		goto out;
+	statdir = debugfs_create_file("rcustats", 0444, rcudir,
+						NULL, &rcustats_fops);
+	if (!statdir)
+		goto free_out;
+
+	gpdir = debugfs_create_file("rcugp", 0444, rcudir, NULL, &rcugp_fops);
+	if (!gpdir)
+		goto free_out;
+
+	ctrsdir = debugfs_create_file("rcuctrs", 0444, rcudir,
+						NULL, &rcuctrs_fops);
+	if (!ctrsdir)
+		goto free_out;
+	return 0;
+free_out:
+	if (statdir)
+		debugfs_remove(statdir);
+	if (gpdir)
+		debugfs_remove(gpdir);
+	debugfs_remove(rcudir);
+out:
+	return 1;
+}
+
+static int __init rcupreempt_trace_init(void)
+{
+	mutex_init(&rcupreempt_trace_mutex);
+	rcupreempt_trace_buf = kmalloc(RCUPREEMPT_TRACE_BUF_SIZE, GFP_KERNEL);
+	if (!rcupreempt_trace_buf)
+		return 1;
+	return rcupreempt_debugfs_init();
+}
+
+static void __exit rcupreempt_trace_cleanup(void)
+{
+	debugfs_remove(statdir);
+	debugfs_remove(gpdir);
+	debugfs_remove(ctrsdir);
+	debugfs_remove(rcudir);
+	kfree(rcupreempt_trace_buf);
+}
+
+
+module_init(rcupreempt_trace_init);
+module_exit(rcupreempt_trace_cleanup);
diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c
index c3e165c2318..fd599829e72 100644
--- a/kernel/rcutorture.c
+++ b/kernel/rcutorture.c
@@ -726,11 +726,11 @@ static void rcu_torture_shuffle_tasks(void)
 	cpumask_t tmp_mask = CPU_MASK_ALL;
 	int i;
 
-	lock_cpu_hotplug();
+	get_online_cpus();
 
 	/* No point in shuffling if there is only one online CPU (ex: UP) */
 	if (num_online_cpus() == 1) {
-		unlock_cpu_hotplug();
+		put_online_cpus();
 		return;
 	}
 
@@ -762,7 +762,7 @@ static void rcu_torture_shuffle_tasks(void)
 	else
 		rcu_idle_cpu--;
 
-	unlock_cpu_hotplug();
+	put_online_cpus();
 }
 
 /* Shuffle tasks across CPUs, with the intent of allowing each CPU in the
diff --git a/kernel/rtmutex-tester.c b/kernel/rtmutex-tester.c
index e3055ba6915..092e4c620af 100644
--- a/kernel/rtmutex-tester.c
+++ b/kernel/rtmutex-tester.c
@@ -394,7 +394,7 @@ static SYSDEV_ATTR(status, 0600, sysfs_test_status, NULL);
 static SYSDEV_ATTR(command, 0600, NULL, sysfs_test_command);
 
 static struct sysdev_class rttest_sysclass = {
-	set_kset_name("rttest"),
+	.name = "rttest",
 };
 
 static int init_test_thread(int id)
diff --git a/kernel/sched.c b/kernel/sched.c
index e76b11ca6df..ba4c88088f6 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -22,6 +22,8 @@
  *              by Peter Williams
  *  2007-05-06  Interactivity improvements to CFS by Mike Galbraith
  *  2007-07-01  Group scheduling enhancements by Srivatsa Vaddagiri
+ *  2007-11-29  RT balancing improvements by Steven Rostedt, Gregory Haskins,
+ *              Thomas Gleixner, Mike Kravetz
  */
 
 #include <linux/mm.h>
@@ -63,6 +65,7 @@
 #include <linux/reciprocal_div.h>
 #include <linux/unistd.h>
 #include <linux/pagemap.h>
+#include <linux/hrtimer.h>
 
 #include <asm/tlb.h>
 #include <asm/irq_regs.h>
@@ -96,10 +99,9 @@ unsigned long long __attribute__((weak)) sched_clock(void)
 #define MAX_USER_PRIO		(USER_PRIO(MAX_PRIO))
 
 /*
- * Some helpers for converting nanosecond timing to jiffy resolution
+ * Helpers for converting nanosecond timing to jiffy resolution
  */
 #define NS_TO_JIFFIES(TIME)	((unsigned long)(TIME) / (NSEC_PER_SEC / HZ))
-#define JIFFIES_TO_NS(TIME)	((TIME) * (NSEC_PER_SEC / HZ))
 
 #define NICE_0_LOAD		SCHED_LOAD_SCALE
 #define NICE_0_SHIFT		SCHED_LOAD_SHIFT
@@ -159,6 +161,8 @@ struct rt_prio_array {
 
 struct cfs_rq;
 
+static LIST_HEAD(task_groups);
+
 /* task group related information */
 struct task_group {
 #ifdef CONFIG_FAIR_CGROUP_SCHED
@@ -168,10 +172,50 @@ struct task_group {
 	struct sched_entity **se;
 	/* runqueue "owned" by this group on each cpu */
 	struct cfs_rq **cfs_rq;
+
+	struct sched_rt_entity **rt_se;
+	struct rt_rq **rt_rq;
+
+	unsigned int rt_ratio;
+
+	/*
+	 * shares assigned to a task group governs how much of cpu bandwidth
+	 * is allocated to the group. The more shares a group has, the more is
+	 * the cpu bandwidth allocated to it.
+	 *
+	 * For ex, lets say that there are three task groups, A, B and C which
+	 * have been assigned shares 1000, 2000 and 3000 respectively. Then,
+	 * cpu bandwidth allocated by the scheduler to task groups A, B and C
+	 * should be:
+	 *
+	 *	Bw(A) = 1000/(1000+2000+3000) * 100 = 16.66%
+	 *	Bw(B) = 2000/(1000+2000+3000) * 100 = 33.33%
+	 *	Bw(C) = 3000/(1000+2000+3000) * 100 = 50%
+	 *
+	 * The weight assigned to a task group's schedulable entities on every
+	 * cpu (task_group.se[a_cpu]->load.weight) is derived from the task
+	 * group's shares. For ex: lets say that task group A has been
+	 * assigned shares of 1000 and there are two CPUs in a system. Then,
+	 *
+	 *  tg_A->se[0]->load.weight = tg_A->se[1]->load.weight = 1000;
+	 *
+	 * Note: It's not necessary that each of a task's group schedulable
+	 *	 entity have the same weight on all CPUs. If the group
+	 *	 has 2 of its tasks on CPU0 and 1 task on CPU1, then a
+	 *	 better distribution of weight could be:
+	 *
+	 *	tg_A->se[0]->load.weight = 2/3 * 2000 = 1333
+	 *	tg_A->se[1]->load.weight = 1/2 * 2000 =  667
+	 *
+	 * rebalance_shares() is responsible for distributing the shares of a
+	 * task groups like this among the group's schedulable entities across
+	 * cpus.
+	 *
+	 */
 	unsigned long shares;
-	/* spinlock to serialize modification to shares */
-	spinlock_t lock;
+
 	struct rcu_head rcu;
+	struct list_head list;
 };
 
 /* Default task group's sched entity on each cpu */
@@ -179,24 +223,51 @@ static DEFINE_PER_CPU(struct sched_entity, init_sched_entity);
 /* Default task group's cfs_rq on each cpu */
 static DEFINE_PER_CPU(struct cfs_rq, init_cfs_rq) ____cacheline_aligned_in_smp;
 
+static DEFINE_PER_CPU(struct sched_rt_entity, init_sched_rt_entity);
+static DEFINE_PER_CPU(struct rt_rq, init_rt_rq) ____cacheline_aligned_in_smp;
+
 static struct sched_entity *init_sched_entity_p[NR_CPUS];
 static struct cfs_rq *init_cfs_rq_p[NR_CPUS];
 
+static struct sched_rt_entity *init_sched_rt_entity_p[NR_CPUS];
+static struct rt_rq *init_rt_rq_p[NR_CPUS];
+
+/* task_group_mutex serializes add/remove of task groups and also changes to
+ * a task group's cpu shares.
+ */
+static DEFINE_MUTEX(task_group_mutex);
+
+/* doms_cur_mutex serializes access to doms_cur[] array */
+static DEFINE_MUTEX(doms_cur_mutex);
+
+#ifdef CONFIG_SMP
+/* kernel thread that runs rebalance_shares() periodically */
+static struct task_struct *lb_monitor_task;
+static int load_balance_monitor(void *unused);
+#endif
+
+static void set_se_shares(struct sched_entity *se, unsigned long shares);
+
 /* Default task group.
  *	Every task in system belong to this group at bootup.
  */
 struct task_group init_task_group = {
-	.se     = init_sched_entity_p,
+	.se	= init_sched_entity_p,
 	.cfs_rq = init_cfs_rq_p,
+
+	.rt_se	= init_sched_rt_entity_p,
+	.rt_rq	= init_rt_rq_p,
 };
 
 #ifdef CONFIG_FAIR_USER_SCHED
-# define INIT_TASK_GRP_LOAD	2*NICE_0_LOAD
+# define INIT_TASK_GROUP_LOAD	(2*NICE_0_LOAD)
 #else
-# define INIT_TASK_GRP_LOAD	NICE_0_LOAD
+# define INIT_TASK_GROUP_LOAD	NICE_0_LOAD
 #endif
 
-static int init_task_group_load = INIT_TASK_GRP_LOAD;
+#define MIN_GROUP_SHARES	2
+
+static int init_task_group_load = INIT_TASK_GROUP_LOAD;
 
 /* return group to which a task belongs */
 static inline struct task_group *task_group(struct task_struct *p)
@@ -215,15 +286,42 @@ static inline struct task_group *task_group(struct task_struct *p)
 }
 
 /* Change a task's cfs_rq and parent entity if it moves across CPUs/groups */
-static inline void set_task_cfs_rq(struct task_struct *p, unsigned int cpu)
+static inline void set_task_rq(struct task_struct *p, unsigned int cpu)
 {
 	p->se.cfs_rq = task_group(p)->cfs_rq[cpu];
 	p->se.parent = task_group(p)->se[cpu];
+
+	p->rt.rt_rq  = task_group(p)->rt_rq[cpu];
+	p->rt.parent = task_group(p)->rt_se[cpu];
+}
+
+static inline void lock_task_group_list(void)
+{
+	mutex_lock(&task_group_mutex);
+}
+
+static inline void unlock_task_group_list(void)
+{
+	mutex_unlock(&task_group_mutex);
+}
+
+static inline void lock_doms_cur(void)
+{
+	mutex_lock(&doms_cur_mutex);
+}
+
+static inline void unlock_doms_cur(void)
+{
+	mutex_unlock(&doms_cur_mutex);
 }
 
 #else
 
-static inline void set_task_cfs_rq(struct task_struct *p, unsigned int cpu) { }
+static inline void set_task_rq(struct task_struct *p, unsigned int cpu) { }
+static inline void lock_task_group_list(void) { }
+static inline void unlock_task_group_list(void) { }
+static inline void lock_doms_cur(void) { }
+static inline void unlock_doms_cur(void) { }
 
 #endif	/* CONFIG_FAIR_GROUP_SCHED */
 
@@ -264,11 +362,57 @@ struct cfs_rq {
 /* Real-Time classes' related field in a runqueue: */
 struct rt_rq {
 	struct rt_prio_array active;
-	int rt_load_balance_idx;
-	struct list_head *rt_load_balance_head, *rt_load_balance_curr;
+	unsigned long rt_nr_running;
+#if defined CONFIG_SMP || defined CONFIG_FAIR_GROUP_SCHED
+	int highest_prio; /* highest queued rt task prio */
+#endif
+#ifdef CONFIG_SMP
+	unsigned long rt_nr_migratory;
+	int overloaded;
+#endif
+	int rt_throttled;
+	u64 rt_time;
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	struct rq *rq;
+	struct list_head leaf_rt_rq_list;
+	struct task_group *tg;
+	struct sched_rt_entity *rt_se;
+#endif
+};
+
+#ifdef CONFIG_SMP
+
+/*
+ * We add the notion of a root-domain which will be used to define per-domain
+ * variables. Each exclusive cpuset essentially defines an island domain by
+ * fully partitioning the member cpus from any other cpuset. Whenever a new
+ * exclusive cpuset is created, we also create and attach a new root-domain
+ * object.
+ *
+ */
+struct root_domain {
+	atomic_t refcount;
+	cpumask_t span;
+	cpumask_t online;
+
+	/*
+	 * The "RT overload" flag: it gets set if a CPU has more than
+	 * one runnable RT task.
+	 */
+	cpumask_t rto_mask;
+	atomic_t rto_count;
 };
 
 /*
+ * By default the system creates a single root-domain with all cpus as
+ * members (mimicking the global state we have today).
+ */
+static struct root_domain def_root_domain;
+
+#endif
+
+/*
  * This is the main, per-CPU runqueue data structure.
  *
  * Locking rule: those places that want to lock multiple runqueues
@@ -296,11 +440,15 @@ struct rq {
 	u64 nr_switches;
 
 	struct cfs_rq cfs;
+	struct rt_rq rt;
+	u64 rt_period_expire;
+	int rt_throttled;
+
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	/* list of leaf cfs_rq on this cpu: */
 	struct list_head leaf_cfs_rq_list;
+	struct list_head leaf_rt_rq_list;
 #endif
-	struct rt_rq rt;
 
 	/*
 	 * This is part of a global counter where only the total sum
@@ -317,7 +465,7 @@ struct rq {
 	u64 clock, prev_clock_raw;
 	s64 clock_max_delta;
 
-	unsigned int clock_warps, clock_overflows;
+	unsigned int clock_warps, clock_overflows, clock_underflows;
 	u64 idle_clock;
 	unsigned int clock_deep_idle_events;
 	u64 tick_timestamp;
@@ -325,6 +473,7 @@ struct rq {
 	atomic_t nr_iowait;
 
 #ifdef CONFIG_SMP
+	struct root_domain *rd;
 	struct sched_domain *sd;
 
 	/* For active balancing */
@@ -337,6 +486,12 @@ struct rq {
 	struct list_head migration_queue;
 #endif
 
+#ifdef CONFIG_SCHED_HRTICK
+	unsigned long hrtick_flags;
+	ktime_t hrtick_expire;
+	struct hrtimer hrtick_timer;
+#endif
+
 #ifdef CONFIG_SCHEDSTATS
 	/* latency stats */
 	struct sched_info rq_sched_info;
@@ -363,7 +518,6 @@ struct rq {
 };
 
 static DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues);
-static DEFINE_MUTEX(sched_hotcpu_mutex);
 
 static inline void check_preempt_curr(struct rq *rq, struct task_struct *p)
 {
@@ -441,6 +595,23 @@ static void update_rq_clock(struct rq *rq)
 #define task_rq(p)		cpu_rq(task_cpu(p))
 #define cpu_curr(cpu)		(cpu_rq(cpu)->curr)
 
+unsigned long rt_needs_cpu(int cpu)
+{
+	struct rq *rq = cpu_rq(cpu);
+	u64 delta;
+
+	if (!rq->rt_throttled)
+		return 0;
+
+	if (rq->clock > rq->rt_period_expire)
+		return 1;
+
+	delta = rq->rt_period_expire - rq->clock;
+	do_div(delta, NSEC_PER_SEC / HZ);
+
+	return (unsigned long)delta;
+}
+
 /*
  * Tunables that become constants when CONFIG_SCHED_DEBUG is off:
  */
@@ -459,6 +630,8 @@ enum {
 	SCHED_FEAT_START_DEBIT		= 4,
 	SCHED_FEAT_TREE_AVG		= 8,
 	SCHED_FEAT_APPROX_AVG		= 16,
+	SCHED_FEAT_HRTICK		= 32,
+	SCHED_FEAT_DOUBLE_TICK		= 64,
 };
 
 const_debug unsigned int sysctl_sched_features =
@@ -466,7 +639,9 @@ const_debug unsigned int sysctl_sched_features =
 		SCHED_FEAT_WAKEUP_PREEMPT	* 1 |
 		SCHED_FEAT_START_DEBIT		* 1 |
 		SCHED_FEAT_TREE_AVG		* 0 |
-		SCHED_FEAT_APPROX_AVG		* 0;
+		SCHED_FEAT_APPROX_AVG		* 0 |
+		SCHED_FEAT_HRTICK		* 1 |
+		SCHED_FEAT_DOUBLE_TICK		* 0;
 
 #define sched_feat(x) (sysctl_sched_features & SCHED_FEAT_##x)
 
@@ -477,6 +652,21 @@ const_debug unsigned int sysctl_sched_features =
 const_debug unsigned int sysctl_sched_nr_migrate = 32;
 
 /*
+ * period over which we measure -rt task cpu usage in ms.
+ * default: 1s
+ */
+const_debug unsigned int sysctl_sched_rt_period = 1000;
+
+#define SCHED_RT_FRAC_SHIFT	16
+#define SCHED_RT_FRAC		(1UL << SCHED_RT_FRAC_SHIFT)
+
+/*
+ * ratio of time -rt tasks may consume.
+ * default: 95%
+ */
+const_debug unsigned int sysctl_sched_rt_ratio = 62259;
+
+/*
  * For kernel-internal use: high-speed (but slightly incorrect) per-cpu
  * clock constructed from sched_clock():
  */
@@ -668,7 +858,6 @@ void sched_clock_idle_wakeup_event(u64 delta_ns)
 	struct rq *rq = cpu_rq(smp_processor_id());
 	u64 now = sched_clock();
 
-	touch_softlockup_watchdog();
 	rq->idle_clock += delta_ns;
 	/*
 	 * Override the previous timestamp and ignore all
@@ -680,9 +869,177 @@ void sched_clock_idle_wakeup_event(u64 delta_ns)
 	rq->prev_clock_raw = now;
 	rq->clock += delta_ns;
 	spin_unlock(&rq->lock);
+	touch_softlockup_watchdog();
 }
 EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event);
 
+static void __resched_task(struct task_struct *p, int tif_bit);
+
+static inline void resched_task(struct task_struct *p)
+{
+	__resched_task(p, TIF_NEED_RESCHED);
+}
+
+#ifdef CONFIG_SCHED_HRTICK
+/*
+ * Use HR-timers to deliver accurate preemption points.
+ *
+ * Its all a bit involved since we cannot program an hrt while holding the
+ * rq->lock. So what we do is store a state in in rq->hrtick_* and ask for a
+ * reschedule event.
+ *
+ * When we get rescheduled we reprogram the hrtick_timer outside of the
+ * rq->lock.
+ */
+static inline void resched_hrt(struct task_struct *p)
+{
+	__resched_task(p, TIF_HRTICK_RESCHED);
+}
+
+static inline void resched_rq(struct rq *rq)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&rq->lock, flags);
+	resched_task(rq->curr);
+	spin_unlock_irqrestore(&rq->lock, flags);
+}
+
+enum {
+	HRTICK_SET,		/* re-programm hrtick_timer */
+	HRTICK_RESET,		/* not a new slice */
+};
+
+/*
+ * Use hrtick when:
+ *  - enabled by features
+ *  - hrtimer is actually high res
+ */
+static inline int hrtick_enabled(struct rq *rq)
+{
+	if (!sched_feat(HRTICK))
+		return 0;
+	return hrtimer_is_hres_active(&rq->hrtick_timer);
+}
+
+/*
+ * Called to set the hrtick timer state.
+ *
+ * called with rq->lock held and irqs disabled
+ */
+static void hrtick_start(struct rq *rq, u64 delay, int reset)
+{
+	assert_spin_locked(&rq->lock);
+
+	/*
+	 * preempt at: now + delay
+	 */
+	rq->hrtick_expire =
+		ktime_add_ns(rq->hrtick_timer.base->get_time(), delay);
+	/*
+	 * indicate we need to program the timer
+	 */
+	__set_bit(HRTICK_SET, &rq->hrtick_flags);
+	if (reset)
+		__set_bit(HRTICK_RESET, &rq->hrtick_flags);
+
+	/*
+	 * New slices are called from the schedule path and don't need a
+	 * forced reschedule.
+	 */
+	if (reset)
+		resched_hrt(rq->curr);
+}
+
+static void hrtick_clear(struct rq *rq)
+{
+	if (hrtimer_active(&rq->hrtick_timer))
+		hrtimer_cancel(&rq->hrtick_timer);
+}
+
+/*
+ * Update the timer from the possible pending state.
+ */
+static void hrtick_set(struct rq *rq)
+{
+	ktime_t time;
+	int set, reset;
+	unsigned long flags;
+
+	WARN_ON_ONCE(cpu_of(rq) != smp_processor_id());
+
+	spin_lock_irqsave(&rq->lock, flags);
+	set = __test_and_clear_bit(HRTICK_SET, &rq->hrtick_flags);
+	reset = __test_and_clear_bit(HRTICK_RESET, &rq->hrtick_flags);
+	time = rq->hrtick_expire;
+	clear_thread_flag(TIF_HRTICK_RESCHED);
+	spin_unlock_irqrestore(&rq->lock, flags);
+
+	if (set) {
+		hrtimer_start(&rq->hrtick_timer, time, HRTIMER_MODE_ABS);
+		if (reset && !hrtimer_active(&rq->hrtick_timer))
+			resched_rq(rq);
+	} else
+		hrtick_clear(rq);
+}
+
+/*
+ * High-resolution timer tick.
+ * Runs from hardirq context with interrupts disabled.
+ */
+static enum hrtimer_restart hrtick(struct hrtimer *timer)
+{
+	struct rq *rq = container_of(timer, struct rq, hrtick_timer);
+
+	WARN_ON_ONCE(cpu_of(rq) != smp_processor_id());
+
+	spin_lock(&rq->lock);
+	__update_rq_clock(rq);
+	rq->curr->sched_class->task_tick(rq, rq->curr, 1);
+	spin_unlock(&rq->lock);
+
+	return HRTIMER_NORESTART;
+}
+
+static inline void init_rq_hrtick(struct rq *rq)
+{
+	rq->hrtick_flags = 0;
+	hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	rq->hrtick_timer.function = hrtick;
+	rq->hrtick_timer.cb_mode = HRTIMER_CB_IRQSAFE_NO_SOFTIRQ;
+}
+
+void hrtick_resched(void)
+{
+	struct rq *rq;
+	unsigned long flags;
+
+	if (!test_thread_flag(TIF_HRTICK_RESCHED))
+		return;
+
+	local_irq_save(flags);
+	rq = cpu_rq(smp_processor_id());
+	hrtick_set(rq);
+	local_irq_restore(flags);
+}
+#else
+static inline void hrtick_clear(struct rq *rq)
+{
+}
+
+static inline void hrtick_set(struct rq *rq)
+{
+}
+
+static inline void init_rq_hrtick(struct rq *rq)
+{
+}
+
+void hrtick_resched(void)
+{
+}
+#endif
+
 /*
  * resched_task - mark a task 'to be rescheduled now'.
  *
@@ -696,16 +1053,16 @@ EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event);
 #define tsk_is_polling(t) test_tsk_thread_flag(t, TIF_POLLING_NRFLAG)
 #endif
 
-static void resched_task(struct task_struct *p)
+static void __resched_task(struct task_struct *p, int tif_bit)
 {
 	int cpu;
 
 	assert_spin_locked(&task_rq(p)->lock);
 
-	if (unlikely(test_tsk_thread_flag(p, TIF_NEED_RESCHED)))
+	if (unlikely(test_tsk_thread_flag(p, tif_bit)))
 		return;
 
-	set_tsk_thread_flag(p, TIF_NEED_RESCHED);
+	set_tsk_thread_flag(p, tif_bit);
 
 	cpu = task_cpu(p);
 	if (cpu == smp_processor_id())
@@ -728,10 +1085,10 @@ static void resched_cpu(int cpu)
 	spin_unlock_irqrestore(&rq->lock, flags);
 }
 #else
-static inline void resched_task(struct task_struct *p)
+static void __resched_task(struct task_struct *p, int tif_bit)
 {
 	assert_spin_locked(&task_rq(p)->lock);
-	set_tsk_need_resched(p);
+	set_tsk_thread_flag(p, tif_bit);
 }
 #endif
 
@@ -871,6 +1228,23 @@ static void cpuacct_charge(struct task_struct *tsk, u64 cputime);
 static inline void cpuacct_charge(struct task_struct *tsk, u64 cputime) {}
 #endif
 
+static inline void inc_cpu_load(struct rq *rq, unsigned long load)
+{
+	update_load_add(&rq->load, load);
+}
+
+static inline void dec_cpu_load(struct rq *rq, unsigned long load)
+{
+	update_load_sub(&rq->load, load);
+}
+
+#ifdef CONFIG_SMP
+static unsigned long source_load(int cpu, int type);
+static unsigned long target_load(int cpu, int type);
+static unsigned long cpu_avg_load_per_task(int cpu);
+static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd);
+#endif /* CONFIG_SMP */
+
 #include "sched_stats.h"
 #include "sched_idletask.c"
 #include "sched_fair.c"
@@ -881,41 +1255,14 @@ static inline void cpuacct_charge(struct task_struct *tsk, u64 cputime) {}
 
 #define sched_class_highest (&rt_sched_class)
 
-/*
- * Update delta_exec, delta_fair fields for rq.
- *
- * delta_fair clock advances at a rate inversely proportional to
- * total load (rq->load.weight) on the runqueue, while
- * delta_exec advances at the same rate as wall-clock (provided
- * cpu is not idle).
- *
- * delta_exec / delta_fair is a measure of the (smoothened) load on this
- * runqueue over any given interval. This (smoothened) load is used
- * during load balance.
- *
- * This function is called /before/ updating rq->load
- * and when switching tasks.
- */
-static inline void inc_load(struct rq *rq, const struct task_struct *p)
-{
-	update_load_add(&rq->load, p->se.load.weight);
-}
-
-static inline void dec_load(struct rq *rq, const struct task_struct *p)
-{
-	update_load_sub(&rq->load, p->se.load.weight);
-}
-
 static void inc_nr_running(struct task_struct *p, struct rq *rq)
 {
 	rq->nr_running++;
-	inc_load(rq, p);
 }
 
 static void dec_nr_running(struct task_struct *p, struct rq *rq)
 {
 	rq->nr_running--;
-	dec_load(rq, p);
 }
 
 static void set_load_weight(struct task_struct *p)
@@ -1039,7 +1386,7 @@ unsigned long weighted_cpuload(const int cpu)
 
 static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
 {
-	set_task_cfs_rq(p, cpu);
+	set_task_rq(p, cpu);
 #ifdef CONFIG_SMP
 	/*
 	 * After ->cpu is set up to a new value, task_rq_lock(p, ...) can be
@@ -1051,12 +1398,24 @@ static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
 #endif
 }
 
+static inline void check_class_changed(struct rq *rq, struct task_struct *p,
+				       const struct sched_class *prev_class,
+				       int oldprio, int running)
+{
+	if (prev_class != p->sched_class) {
+		if (prev_class->switched_from)
+			prev_class->switched_from(rq, p, running);
+		p->sched_class->switched_to(rq, p, running);
+	} else
+		p->sched_class->prio_changed(rq, p, oldprio, running);
+}
+
 #ifdef CONFIG_SMP
 
 /*
  * Is this task likely cache-hot:
  */
-static inline int
+static int
 task_hot(struct task_struct *p, u64 now, struct sched_domain *sd)
 {
 	s64 delta;
@@ -1281,7 +1640,7 @@ static unsigned long target_load(int cpu, int type)
 /*
  * Return the average load per task on the cpu's run queue
  */
-static inline unsigned long cpu_avg_load_per_task(int cpu)
+static unsigned long cpu_avg_load_per_task(int cpu)
 {
 	struct rq *rq = cpu_rq(cpu);
 	unsigned long total = weighted_cpuload(cpu);
@@ -1438,58 +1797,6 @@ static int sched_balance_self(int cpu, int flag)
 
 #endif /* CONFIG_SMP */
 
-/*
- * wake_idle() will wake a task on an idle cpu if task->cpu is
- * not idle and an idle cpu is available.  The span of cpus to
- * search starts with cpus closest then further out as needed,
- * so we always favor a closer, idle cpu.
- *
- * Returns the CPU we should wake onto.
- */
-#if defined(ARCH_HAS_SCHED_WAKE_IDLE)
-static int wake_idle(int cpu, struct task_struct *p)
-{
-	cpumask_t tmp;
-	struct sched_domain *sd;
-	int i;
-
-	/*
-	 * If it is idle, then it is the best cpu to run this task.
-	 *
-	 * This cpu is also the best, if it has more than one task already.
-	 * Siblings must be also busy(in most cases) as they didn't already
-	 * pickup the extra load from this cpu and hence we need not check
-	 * sibling runqueue info. This will avoid the checks and cache miss
-	 * penalities associated with that.
-	 */
-	if (idle_cpu(cpu) || cpu_rq(cpu)->nr_running > 1)
-		return cpu;
-
-	for_each_domain(cpu, sd) {
-		if (sd->flags & SD_WAKE_IDLE) {
-			cpus_and(tmp, sd->span, p->cpus_allowed);
-			for_each_cpu_mask(i, tmp) {
-				if (idle_cpu(i)) {
-					if (i != task_cpu(p)) {
-						schedstat_inc(p,
-							se.nr_wakeups_idle);
-					}
-					return i;
-				}
-			}
-		} else {
-			break;
-		}
-	}
-	return cpu;
-}
-#else
-static inline int wake_idle(int cpu, struct task_struct *p)
-{
-	return cpu;
-}
-#endif
-
 /***
  * try_to_wake_up - wake up a thread
  * @p: the to-be-woken-up thread
@@ -1510,11 +1817,6 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 	unsigned long flags;
 	long old_state;
 	struct rq *rq;
-#ifdef CONFIG_SMP
-	struct sched_domain *sd, *this_sd = NULL;
-	unsigned long load, this_load;
-	int new_cpu;
-#endif
 
 	rq = task_rq_lock(p, &flags);
 	old_state = p->state;
@@ -1532,92 +1834,9 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync)
 	if (unlikely(task_running(rq, p)))
 		goto out_activate;
 
-	new_cpu = cpu;
-
-	schedstat_inc(rq, ttwu_count);
-	if (cpu == this_cpu) {
-		schedstat_inc(rq, ttwu_local);
-		goto out_set_cpu;
-	}
-
-	for_each_domain(this_cpu, sd) {
-		if (cpu_isset(cpu, sd->span)) {
-			schedstat_inc(sd, ttwu_wake_remote);
-			this_sd = sd;
-			break;
-		}
-	}
-
-	if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed)))
-		goto out_set_cpu;
-
-	/*
-	 * Check for affine wakeup and passive balancing possibilities.
-	 */
-	if (this_sd) {
-		int idx = this_sd->wake_idx;
-		unsigned int imbalance;
-
-		imbalance = 100 + (this_sd->imbalance_pct - 100) / 2;
-
-		load = source_load(cpu, idx);
-		this_load = target_load(this_cpu, idx);
-
-		new_cpu = this_cpu; /* Wake to this CPU if we can */
-
-		if (this_sd->flags & SD_WAKE_AFFINE) {
-			unsigned long tl = this_load;
-			unsigned long tl_per_task;
-
-			/*
-			 * Attract cache-cold tasks on sync wakeups:
-			 */
-			if (sync && !task_hot(p, rq->clock, this_sd))
-				goto out_set_cpu;
-
-			schedstat_inc(p, se.nr_wakeups_affine_attempts);
-			tl_per_task = cpu_avg_load_per_task(this_cpu);
-
-			/*
-			 * If sync wakeup then subtract the (maximum possible)
-			 * effect of the currently running task from the load
-			 * of the current CPU:
-			 */
-			if (sync)
-				tl -= current->se.load.weight;
-
-			if ((tl <= load &&
-				tl + target_load(cpu, idx) <= tl_per_task) ||
-			       100*(tl + p->se.load.weight) <= imbalance*load) {
-				/*
-				 * This domain has SD_WAKE_AFFINE and
-				 * p is cache cold in this domain, and
-				 * there is no bad imbalance.
-				 */
-				schedstat_inc(this_sd, ttwu_move_affine);
-				schedstat_inc(p, se.nr_wakeups_affine);
-				goto out_set_cpu;
-			}
-		}
-
-		/*
-		 * Start passive balancing when half the imbalance_pct
-		 * limit is reached.
-		 */
-		if (this_sd->flags & SD_WAKE_BALANCE) {
-			if (imbalance*this_load <= 100*load) {
-				schedstat_inc(this_sd, ttwu_move_balance);
-				schedstat_inc(p, se.nr_wakeups_passive);
-				goto out_set_cpu;
-			}
-		}
-	}
-
-	new_cpu = cpu; /* Could not wake to this_cpu. Wake to cpu instead */
-out_set_cpu:
-	new_cpu = wake_idle(new_cpu, p);
-	if (new_cpu != cpu) {
-		set_task_cpu(p, new_cpu);
+	cpu = p->sched_class->select_task_rq(p, sync);
+	if (cpu != orig_cpu) {
+		set_task_cpu(p, cpu);
 		task_rq_unlock(rq, &flags);
 		/* might preempt at this point */
 		rq = task_rq_lock(p, &flags);
@@ -1631,6 +1850,21 @@ out_set_cpu:
 		cpu = task_cpu(p);
 	}
 
+#ifdef CONFIG_SCHEDSTATS
+	schedstat_inc(rq, ttwu_count);
+	if (cpu == this_cpu)
+		schedstat_inc(rq, ttwu_local);
+	else {
+		struct sched_domain *sd;
+		for_each_domain(this_cpu, sd) {
+			if (cpu_isset(cpu, sd->span)) {
+				schedstat_inc(sd, ttwu_wake_remote);
+				break;
+			}
+		}
+	}
+#endif
+
 out_activate:
 #endif /* CONFIG_SMP */
 	schedstat_inc(p, se.nr_wakeups);
@@ -1649,6 +1883,10 @@ out_activate:
 
 out_running:
 	p->state = TASK_RUNNING;
+#ifdef CONFIG_SMP
+	if (p->sched_class->task_wake_up)
+		p->sched_class->task_wake_up(rq, p);
+#endif
 out:
 	task_rq_unlock(rq, &flags);
 
@@ -1691,7 +1929,7 @@ static void __sched_fork(struct task_struct *p)
 	p->se.wait_max			= 0;
 #endif
 
-	INIT_LIST_HEAD(&p->run_list);
+	INIT_LIST_HEAD(&p->rt.run_list);
 	p->se.on_rq = 0;
 
 #ifdef CONFIG_PREEMPT_NOTIFIERS
@@ -1771,6 +2009,10 @@ void fastcall wake_up_new_task(struct task_struct *p, unsigned long clone_flags)
 		inc_nr_running(p, rq);
 	}
 	check_preempt_curr(rq, p);
+#ifdef CONFIG_SMP
+	if (p->sched_class->task_wake_up)
+		p->sched_class->task_wake_up(rq, p);
+#endif
 	task_rq_unlock(rq, &flags);
 }
 
@@ -1891,6 +2133,11 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev)
 	prev_state = prev->state;
 	finish_arch_switch(prev);
 	finish_lock_switch(rq, prev);
+#ifdef CONFIG_SMP
+	if (current->sched_class->post_schedule)
+		current->sched_class->post_schedule(rq);
+#endif
+
 	fire_sched_in_preempt_notifiers(current);
 	if (mm)
 		mmdrop(mm);
@@ -2124,11 +2371,13 @@ static void double_rq_unlock(struct rq *rq1, struct rq *rq2)
 /*
  * double_lock_balance - lock the busiest runqueue, this_rq is locked already.
  */
-static void double_lock_balance(struct rq *this_rq, struct rq *busiest)
+static int double_lock_balance(struct rq *this_rq, struct rq *busiest)
 	__releases(this_rq->lock)
 	__acquires(busiest->lock)
 	__acquires(this_rq->lock)
 {
+	int ret = 0;
+
 	if (unlikely(!irqs_disabled())) {
 		/* printk() doesn't work good under rq->lock */
 		spin_unlock(&this_rq->lock);
@@ -2139,9 +2388,11 @@ static void double_lock_balance(struct rq *this_rq, struct rq *busiest)
 			spin_unlock(&this_rq->lock);
 			spin_lock(&busiest->lock);
 			spin_lock(&this_rq->lock);
+			ret = 1;
 		} else
 			spin_lock(&busiest->lock);
 	}
+	return ret;
 }
 
 /*
@@ -3485,12 +3736,14 @@ void scheduler_tick(void)
 	/*
 	 * Let rq->clock advance by at least TICK_NSEC:
 	 */
-	if (unlikely(rq->clock < next_tick))
+	if (unlikely(rq->clock < next_tick)) {
 		rq->clock = next_tick;
+		rq->clock_underflows++;
+	}
 	rq->tick_timestamp = rq->clock;
 	update_cpu_load(rq);
-	if (curr != rq->idle) /* FIXME: needed? */
-		curr->sched_class->task_tick(rq, curr);
+	curr->sched_class->task_tick(rq, curr, 0);
+	update_sched_rt_period(rq);
 	spin_unlock(&rq->lock);
 
 #ifdef CONFIG_SMP
@@ -3636,6 +3889,8 @@ need_resched_nonpreemptible:
 
 	schedule_debug(prev);
 
+	hrtick_clear(rq);
+
 	/*
 	 * Do the rq-clock update outside the rq lock:
 	 */
@@ -3654,6 +3909,11 @@ need_resched_nonpreemptible:
 		switch_count = &prev->nvcsw;
 	}
 
+#ifdef CONFIG_SMP
+	if (prev->sched_class->pre_schedule)
+		prev->sched_class->pre_schedule(rq, prev);
+#endif
+
 	if (unlikely(!rq->nr_running))
 		idle_balance(cpu, rq);
 
@@ -3668,14 +3928,20 @@ need_resched_nonpreemptible:
 		++*switch_count;
 
 		context_switch(rq, prev, next); /* unlocks the rq */
+		/*
+		 * the context switch might have flipped the stack from under
+		 * us, hence refresh the local variables.
+		 */
+		cpu = smp_processor_id();
+		rq = cpu_rq(cpu);
 	} else
 		spin_unlock_irq(&rq->lock);
 
-	if (unlikely(reacquire_kernel_lock(current) < 0)) {
-		cpu = smp_processor_id();
-		rq = cpu_rq(cpu);
+	hrtick_set(rq);
+
+	if (unlikely(reacquire_kernel_lock(current) < 0))
 		goto need_resched_nonpreemptible;
-	}
+
 	preempt_enable_no_resched();
 	if (unlikely(test_thread_flag(TIF_NEED_RESCHED)))
 		goto need_resched;
@@ -3691,10 +3957,9 @@ EXPORT_SYMBOL(schedule);
 asmlinkage void __sched preempt_schedule(void)
 {
 	struct thread_info *ti = current_thread_info();
-#ifdef CONFIG_PREEMPT_BKL
 	struct task_struct *task = current;
 	int saved_lock_depth;
-#endif
+
 	/*
 	 * If there is a non-zero preempt_count or interrupts are disabled,
 	 * we do not want to preempt the current task. Just return..
@@ -3710,14 +3975,10 @@ asmlinkage void __sched preempt_schedule(void)
 		 * clear ->lock_depth so that schedule() doesnt
 		 * auto-release the semaphore:
 		 */
-#ifdef CONFIG_PREEMPT_BKL
 		saved_lock_depth = task->lock_depth;
 		task->lock_depth = -1;
-#endif
 		schedule();
-#ifdef CONFIG_PREEMPT_BKL
 		task->lock_depth = saved_lock_depth;
-#endif
 		sub_preempt_count(PREEMPT_ACTIVE);
 
 		/*
@@ -3738,10 +3999,9 @@ EXPORT_SYMBOL(preempt_schedule);
 asmlinkage void __sched preempt_schedule_irq(void)
 {
 	struct thread_info *ti = current_thread_info();
-#ifdef CONFIG_PREEMPT_BKL
 	struct task_struct *task = current;
 	int saved_lock_depth;
-#endif
+
 	/* Catch callers which need to be fixed */
 	BUG_ON(ti->preempt_count || !irqs_disabled());
 
@@ -3753,16 +4013,12 @@ asmlinkage void __sched preempt_schedule_irq(void)
 		 * clear ->lock_depth so that schedule() doesnt
 		 * auto-release the semaphore:
 		 */
-#ifdef CONFIG_PREEMPT_BKL
 		saved_lock_depth = task->lock_depth;
 		task->lock_depth = -1;
-#endif
 		local_irq_enable();
 		schedule();
 		local_irq_disable();
-#ifdef CONFIG_PREEMPT_BKL
 		task->lock_depth = saved_lock_depth;
-#endif
 		sub_preempt_count(PREEMPT_ACTIVE);
 
 		/*
@@ -4019,6 +4275,7 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
 	unsigned long flags;
 	int oldprio, on_rq, running;
 	struct rq *rq;
+	const struct sched_class *prev_class = p->sched_class;
 
 	BUG_ON(prio < 0 || prio > MAX_PRIO);
 
@@ -4044,18 +4301,10 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
 	if (on_rq) {
 		if (running)
 			p->sched_class->set_curr_task(rq);
+
 		enqueue_task(rq, p, 0);
-		/*
-		 * Reschedule if we are currently running on this runqueue and
-		 * our priority decreased, or if we are not currently running on
-		 * this runqueue and our priority is higher than the current's
-		 */
-		if (running) {
-			if (p->prio > oldprio)
-				resched_task(rq->curr);
-		} else {
-			check_preempt_curr(rq, p);
-		}
+
+		check_class_changed(rq, p, prev_class, oldprio, running);
 	}
 	task_rq_unlock(rq, &flags);
 }
@@ -4087,10 +4336,8 @@ void set_user_nice(struct task_struct *p, long nice)
 		goto out_unlock;
 	}
 	on_rq = p->se.on_rq;
-	if (on_rq) {
+	if (on_rq)
 		dequeue_task(rq, p, 0);
-		dec_load(rq, p);
-	}
 
 	p->static_prio = NICE_TO_PRIO(nice);
 	set_load_weight(p);
@@ -4100,7 +4347,6 @@ void set_user_nice(struct task_struct *p, long nice)
 
 	if (on_rq) {
 		enqueue_task(rq, p, 0);
-		inc_load(rq, p);
 		/*
 		 * If the task increased its priority or is running and
 		 * lowered its priority, then reschedule its CPU:
@@ -4258,6 +4504,7 @@ int sched_setscheduler(struct task_struct *p, int policy,
 {
 	int retval, oldprio, oldpolicy = -1, on_rq, running;
 	unsigned long flags;
+	const struct sched_class *prev_class = p->sched_class;
 	struct rq *rq;
 
 	/* may grab non-irq protected spin_locks */
@@ -4351,18 +4598,10 @@ recheck:
 	if (on_rq) {
 		if (running)
 			p->sched_class->set_curr_task(rq);
+
 		activate_task(rq, p, 0);
-		/*
-		 * Reschedule if we are currently running on this runqueue and
-		 * our priority decreased, or if we are not currently running on
-		 * this runqueue and our priority is higher than the current's
-		 */
-		if (running) {
-			if (p->prio > oldprio)
-				resched_task(rq->curr);
-		} else {
-			check_preempt_curr(rq, p);
-		}
+
+		check_class_changed(rq, p, prev_class, oldprio, running);
 	}
 	__task_rq_unlock(rq);
 	spin_unlock_irqrestore(&p->pi_lock, flags);
@@ -4490,13 +4729,13 @@ long sched_setaffinity(pid_t pid, cpumask_t new_mask)
 	struct task_struct *p;
 	int retval;
 
-	mutex_lock(&sched_hotcpu_mutex);
+	get_online_cpus();
 	read_lock(&tasklist_lock);
 
 	p = find_process_by_pid(pid);
 	if (!p) {
 		read_unlock(&tasklist_lock);
-		mutex_unlock(&sched_hotcpu_mutex);
+		put_online_cpus();
 		return -ESRCH;
 	}
 
@@ -4536,7 +4775,7 @@ long sched_setaffinity(pid_t pid, cpumask_t new_mask)
 	}
 out_unlock:
 	put_task_struct(p);
-	mutex_unlock(&sched_hotcpu_mutex);
+	put_online_cpus();
 	return retval;
 }
 
@@ -4593,7 +4832,7 @@ long sched_getaffinity(pid_t pid, cpumask_t *mask)
 	struct task_struct *p;
 	int retval;
 
-	mutex_lock(&sched_hotcpu_mutex);
+	get_online_cpus();
 	read_lock(&tasklist_lock);
 
 	retval = -ESRCH;
@@ -4609,7 +4848,7 @@ long sched_getaffinity(pid_t pid, cpumask_t *mask)
 
 out_unlock:
 	read_unlock(&tasklist_lock);
-	mutex_unlock(&sched_hotcpu_mutex);
+	put_online_cpus();
 
 	return retval;
 }
@@ -4683,7 +4922,8 @@ static void __cond_resched(void)
 	} while (need_resched());
 }
 
-int __sched cond_resched(void)
+#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PREEMPT_VOLUNTARY)
+int __sched _cond_resched(void)
 {
 	if (need_resched() && !(preempt_count() & PREEMPT_ACTIVE) &&
 					system_state == SYSTEM_RUNNING) {
@@ -4692,7 +4932,8 @@ int __sched cond_resched(void)
 	}
 	return 0;
 }
-EXPORT_SYMBOL(cond_resched);
+EXPORT_SYMBOL(_cond_resched);
+#endif
 
 /*
  * cond_resched_lock() - if a reschedule is pending, drop the given lock,
@@ -4704,19 +4945,15 @@ EXPORT_SYMBOL(cond_resched);
  */
 int cond_resched_lock(spinlock_t *lock)
 {
+	int resched = need_resched() && system_state == SYSTEM_RUNNING;
 	int ret = 0;
 
-	if (need_lockbreak(lock)) {
+	if (spin_needbreak(lock) || resched) {
 		spin_unlock(lock);
-		cpu_relax();
-		ret = 1;
-		spin_lock(lock);
-	}
-	if (need_resched() && system_state == SYSTEM_RUNNING) {
-		spin_release(&lock->dep_map, 1, _THIS_IP_);
-		_raw_spin_unlock(lock);
-		preempt_enable_no_resched();
-		__cond_resched();
+		if (resched && need_resched())
+			__cond_resched();
+		else
+			cpu_relax();
 		ret = 1;
 		spin_lock(lock);
 	}
@@ -4890,7 +5127,7 @@ out_unlock:
 
 static const char stat_nam[] = "RSDTtZX";
 
-static void show_task(struct task_struct *p)
+void sched_show_task(struct task_struct *p)
 {
 	unsigned long free = 0;
 	unsigned state;
@@ -4920,8 +5157,7 @@ static void show_task(struct task_struct *p)
 	printk(KERN_CONT "%5lu %5d %6d\n", free,
 		task_pid_nr(p), task_pid_nr(p->real_parent));
 
-	if (state != TASK_RUNNING)
-		show_stack(p, NULL);
+	show_stack(p, NULL);
 }
 
 void show_state_filter(unsigned long state_filter)
@@ -4943,7 +5179,7 @@ void show_state_filter(unsigned long state_filter)
 		 */
 		touch_nmi_watchdog();
 		if (!state_filter || (p->state & state_filter))
-			show_task(p);
+			sched_show_task(p);
 	} while_each_thread(g, p);
 
 	touch_all_softlockup_watchdogs();
@@ -4992,11 +5228,8 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu)
 	spin_unlock_irqrestore(&rq->lock, flags);
 
 	/* Set the preempt count _outside_ the spinlocks! */
-#if defined(CONFIG_PREEMPT) && !defined(CONFIG_PREEMPT_BKL)
-	task_thread_info(idle)->preempt_count = (idle->lock_depth >= 0);
-#else
 	task_thread_info(idle)->preempt_count = 0;
-#endif
+
 	/*
 	 * The idle tasks have their own, simple scheduling class:
 	 */
@@ -5077,7 +5310,13 @@ int set_cpus_allowed(struct task_struct *p, cpumask_t new_mask)
 		goto out;
 	}
 
-	p->cpus_allowed = new_mask;
+	if (p->sched_class->set_cpus_allowed)
+		p->sched_class->set_cpus_allowed(p, &new_mask);
+	else {
+		p->cpus_allowed = new_mask;
+		p->rt.nr_cpus_allowed = cpus_weight(new_mask);
+	}
+
 	/* Can the task run on the task's current CPU? If so, we're done */
 	if (cpu_isset(task_cpu(p), new_mask))
 		goto out;
@@ -5569,9 +5808,6 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 	struct rq *rq;
 
 	switch (action) {
-	case CPU_LOCK_ACQUIRE:
-		mutex_lock(&sched_hotcpu_mutex);
-		break;
 
 	case CPU_UP_PREPARE:
 	case CPU_UP_PREPARE_FROZEN:
@@ -5590,6 +5826,15 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 	case CPU_ONLINE_FROZEN:
 		/* Strictly unnecessary, as first user will wake it. */
 		wake_up_process(cpu_rq(cpu)->migration_thread);
+
+		/* Update our root-domain */
+		rq = cpu_rq(cpu);
+		spin_lock_irqsave(&rq->lock, flags);
+		if (rq->rd) {
+			BUG_ON(!cpu_isset(cpu, rq->rd->span));
+			cpu_set(cpu, rq->rd->online);
+		}
+		spin_unlock_irqrestore(&rq->lock, flags);
 		break;
 
 #ifdef CONFIG_HOTPLUG_CPU
@@ -5640,10 +5885,18 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 		}
 		spin_unlock_irq(&rq->lock);
 		break;
-#endif
-	case CPU_LOCK_RELEASE:
-		mutex_unlock(&sched_hotcpu_mutex);
+
+	case CPU_DOWN_PREPARE:
+		/* Update our root-domain */
+		rq = cpu_rq(cpu);
+		spin_lock_irqsave(&rq->lock, flags);
+		if (rq->rd) {
+			BUG_ON(!cpu_isset(cpu, rq->rd->span));
+			cpu_clear(cpu, rq->rd->online);
+		}
+		spin_unlock_irqrestore(&rq->lock, flags);
 		break;
+#endif
 	}
 	return NOTIFY_OK;
 }
@@ -5831,11 +6084,76 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent)
 	return 1;
 }
 
+static void rq_attach_root(struct rq *rq, struct root_domain *rd)
+{
+	unsigned long flags;
+	const struct sched_class *class;
+
+	spin_lock_irqsave(&rq->lock, flags);
+
+	if (rq->rd) {
+		struct root_domain *old_rd = rq->rd;
+
+		for (class = sched_class_highest; class; class = class->next) {
+			if (class->leave_domain)
+				class->leave_domain(rq);
+		}
+
+		cpu_clear(rq->cpu, old_rd->span);
+		cpu_clear(rq->cpu, old_rd->online);
+
+		if (atomic_dec_and_test(&old_rd->refcount))
+			kfree(old_rd);
+	}
+
+	atomic_inc(&rd->refcount);
+	rq->rd = rd;
+
+	cpu_set(rq->cpu, rd->span);
+	if (cpu_isset(rq->cpu, cpu_online_map))
+		cpu_set(rq->cpu, rd->online);
+
+	for (class = sched_class_highest; class; class = class->next) {
+		if (class->join_domain)
+			class->join_domain(rq);
+	}
+
+	spin_unlock_irqrestore(&rq->lock, flags);
+}
+
+static void init_rootdomain(struct root_domain *rd)
+{
+	memset(rd, 0, sizeof(*rd));
+
+	cpus_clear(rd->span);
+	cpus_clear(rd->online);
+}
+
+static void init_defrootdomain(void)
+{
+	init_rootdomain(&def_root_domain);
+	atomic_set(&def_root_domain.refcount, 1);
+}
+
+static struct root_domain *alloc_rootdomain(void)
+{
+	struct root_domain *rd;
+
+	rd = kmalloc(sizeof(*rd), GFP_KERNEL);
+	if (!rd)
+		return NULL;
+
+	init_rootdomain(rd);
+
+	return rd;
+}
+
 /*
- * Attach the domain 'sd' to 'cpu' as its base domain.  Callers must
+ * Attach the domain 'sd' to 'cpu' as its base domain. Callers must
  * hold the hotplug lock.
  */
-static void cpu_attach_domain(struct sched_domain *sd, int cpu)
+static void
+cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu)
 {
 	struct rq *rq = cpu_rq(cpu);
 	struct sched_domain *tmp;
@@ -5860,6 +6178,7 @@ static void cpu_attach_domain(struct sched_domain *sd, int cpu)
 
 	sched_domain_debug(sd, cpu);
 
+	rq_attach_root(rq, rd);
 	rcu_assign_pointer(rq->sd, sd);
 }
 
@@ -6228,6 +6547,7 @@ static void init_sched_groups_power(int cpu, struct sched_domain *sd)
 static int build_sched_domains(const cpumask_t *cpu_map)
 {
 	int i;
+	struct root_domain *rd;
 #ifdef CONFIG_NUMA
 	struct sched_group **sched_group_nodes = NULL;
 	int sd_allnodes = 0;
@@ -6244,6 +6564,12 @@ static int build_sched_domains(const cpumask_t *cpu_map)
 	sched_group_nodes_bycpu[first_cpu(*cpu_map)] = sched_group_nodes;
 #endif
 
+	rd = alloc_rootdomain();
+	if (!rd) {
+		printk(KERN_WARNING "Cannot alloc root domain\n");
+		return -ENOMEM;
+	}
+
 	/*
 	 * Set up domains for cpus specified by the cpu_map.
 	 */
@@ -6460,7 +6786,7 @@ static int build_sched_domains(const cpumask_t *cpu_map)
 #else
 		sd = &per_cpu(phys_domains, i);
 #endif
-		cpu_attach_domain(sd, i);
+		cpu_attach_domain(sd, rd, i);
 	}
 
 	return 0;
@@ -6518,7 +6844,7 @@ static void detach_destroy_domains(const cpumask_t *cpu_map)
 	unregister_sched_domain_sysctl();
 
 	for_each_cpu_mask(i, *cpu_map)
-		cpu_attach_domain(NULL, i);
+		cpu_attach_domain(NULL, &def_root_domain, i);
 	synchronize_sched();
 	arch_destroy_sched_domains(cpu_map);
 }
@@ -6548,6 +6874,8 @@ void partition_sched_domains(int ndoms_new, cpumask_t *doms_new)
 {
 	int i, j;
 
+	lock_doms_cur();
+
 	/* always unregister in case we don't destroy any domains */
 	unregister_sched_domain_sysctl();
 
@@ -6588,6 +6916,8 @@ match2:
 	ndoms_cur = ndoms_new;
 
 	register_sched_domain_sysctl();
+
+	unlock_doms_cur();
 }
 
 #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
@@ -6595,10 +6925,10 @@ static int arch_reinit_sched_domains(void)
 {
 	int err;
 
-	mutex_lock(&sched_hotcpu_mutex);
+	get_online_cpus();
 	detach_destroy_domains(&cpu_online_map);
 	err = arch_init_sched_domains(&cpu_online_map);
-	mutex_unlock(&sched_hotcpu_mutex);
+	put_online_cpus();
 
 	return err;
 }
@@ -6709,12 +7039,12 @@ void __init sched_init_smp(void)
 {
 	cpumask_t non_isolated_cpus;
 
-	mutex_lock(&sched_hotcpu_mutex);
+	get_online_cpus();
 	arch_init_sched_domains(&cpu_online_map);
 	cpus_andnot(non_isolated_cpus, cpu_possible_map, cpu_isolated_map);
 	if (cpus_empty(non_isolated_cpus))
 		cpu_set(smp_processor_id(), non_isolated_cpus);
-	mutex_unlock(&sched_hotcpu_mutex);
+	put_online_cpus();
 	/* XXX: Theoretical race here - CPU may be hotplugged now */
 	hotcpu_notifier(update_sched_domains, 0);
 
@@ -6722,6 +7052,21 @@ void __init sched_init_smp(void)
 	if (set_cpus_allowed(current, non_isolated_cpus) < 0)
 		BUG();
 	sched_init_granularity();
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	if (nr_cpu_ids == 1)
+		return;
+
+	lb_monitor_task = kthread_create(load_balance_monitor, NULL,
+					 "group_balance");
+	if (!IS_ERR(lb_monitor_task)) {
+		lb_monitor_task->flags |= PF_NOFREEZE;
+		wake_up_process(lb_monitor_task);
+	} else {
+		printk(KERN_ERR "Could not create load balance monitor thread"
+			"(error = %ld) \n", PTR_ERR(lb_monitor_task));
+	}
+#endif
 }
 #else
 void __init sched_init_smp(void)
@@ -6746,13 +7091,87 @@ static void init_cfs_rq(struct cfs_rq *cfs_rq, struct rq *rq)
 	cfs_rq->min_vruntime = (u64)(-(1LL << 20));
 }
 
+static void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq)
+{
+	struct rt_prio_array *array;
+	int i;
+
+	array = &rt_rq->active;
+	for (i = 0; i < MAX_RT_PRIO; i++) {
+		INIT_LIST_HEAD(array->queue + i);
+		__clear_bit(i, array->bitmap);
+	}
+	/* delimiter for bitsearch: */
+	__set_bit(MAX_RT_PRIO, array->bitmap);
+
+#if defined CONFIG_SMP || defined CONFIG_FAIR_GROUP_SCHED
+	rt_rq->highest_prio = MAX_RT_PRIO;
+#endif
+#ifdef CONFIG_SMP
+	rt_rq->rt_nr_migratory = 0;
+	rt_rq->overloaded = 0;
+#endif
+
+	rt_rq->rt_time = 0;
+	rt_rq->rt_throttled = 0;
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	rt_rq->rq = rq;
+#endif
+}
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+static void init_tg_cfs_entry(struct rq *rq, struct task_group *tg,
+		struct cfs_rq *cfs_rq, struct sched_entity *se,
+		int cpu, int add)
+{
+	tg->cfs_rq[cpu] = cfs_rq;
+	init_cfs_rq(cfs_rq, rq);
+	cfs_rq->tg = tg;
+	if (add)
+		list_add(&cfs_rq->leaf_cfs_rq_list, &rq->leaf_cfs_rq_list);
+
+	tg->se[cpu] = se;
+	se->cfs_rq = &rq->cfs;
+	se->my_q = cfs_rq;
+	se->load.weight = tg->shares;
+	se->load.inv_weight = div64_64(1ULL<<32, se->load.weight);
+	se->parent = NULL;
+}
+
+static void init_tg_rt_entry(struct rq *rq, struct task_group *tg,
+		struct rt_rq *rt_rq, struct sched_rt_entity *rt_se,
+		int cpu, int add)
+{
+	tg->rt_rq[cpu] = rt_rq;
+	init_rt_rq(rt_rq, rq);
+	rt_rq->tg = tg;
+	rt_rq->rt_se = rt_se;
+	if (add)
+		list_add(&rt_rq->leaf_rt_rq_list, &rq->leaf_rt_rq_list);
+
+	tg->rt_se[cpu] = rt_se;
+	rt_se->rt_rq = &rq->rt;
+	rt_se->my_q = rt_rq;
+	rt_se->parent = NULL;
+	INIT_LIST_HEAD(&rt_se->run_list);
+}
+#endif
+
 void __init sched_init(void)
 {
 	int highest_cpu = 0;
 	int i, j;
 
+#ifdef CONFIG_SMP
+	init_defrootdomain();
+#endif
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	list_add(&init_task_group.list, &task_groups);
+#endif
+
 	for_each_possible_cpu(i) {
-		struct rt_prio_array *array;
 		struct rq *rq;
 
 		rq = cpu_rq(i);
@@ -6761,52 +7180,39 @@ void __init sched_init(void)
 		rq->nr_running = 0;
 		rq->clock = 1;
 		init_cfs_rq(&rq->cfs, rq);
+		init_rt_rq(&rq->rt, rq);
 #ifdef CONFIG_FAIR_GROUP_SCHED
-		INIT_LIST_HEAD(&rq->leaf_cfs_rq_list);
-		{
-			struct cfs_rq *cfs_rq = &per_cpu(init_cfs_rq, i);
-			struct sched_entity *se =
-					 &per_cpu(init_sched_entity, i);
-
-			init_cfs_rq_p[i] = cfs_rq;
-			init_cfs_rq(cfs_rq, rq);
-			cfs_rq->tg = &init_task_group;
-			list_add(&cfs_rq->leaf_cfs_rq_list,
-							 &rq->leaf_cfs_rq_list);
-
-			init_sched_entity_p[i] = se;
-			se->cfs_rq = &rq->cfs;
-			se->my_q = cfs_rq;
-			se->load.weight = init_task_group_load;
-			se->load.inv_weight =
-				 div64_64(1ULL<<32, init_task_group_load);
-			se->parent = NULL;
-		}
 		init_task_group.shares = init_task_group_load;
-		spin_lock_init(&init_task_group.lock);
+		INIT_LIST_HEAD(&rq->leaf_cfs_rq_list);
+		init_tg_cfs_entry(rq, &init_task_group,
+				&per_cpu(init_cfs_rq, i),
+				&per_cpu(init_sched_entity, i), i, 1);
+
+		init_task_group.rt_ratio = sysctl_sched_rt_ratio; /* XXX */
+		INIT_LIST_HEAD(&rq->leaf_rt_rq_list);
+		init_tg_rt_entry(rq, &init_task_group,
+				&per_cpu(init_rt_rq, i),
+				&per_cpu(init_sched_rt_entity, i), i, 1);
 #endif
+		rq->rt_period_expire = 0;
+		rq->rt_throttled = 0;
 
 		for (j = 0; j < CPU_LOAD_IDX_MAX; j++)
 			rq->cpu_load[j] = 0;
 #ifdef CONFIG_SMP
 		rq->sd = NULL;
+		rq->rd = NULL;
 		rq->active_balance = 0;
 		rq->next_balance = jiffies;
 		rq->push_cpu = 0;
 		rq->cpu = i;
 		rq->migration_thread = NULL;
 		INIT_LIST_HEAD(&rq->migration_queue);
+		rq_attach_root(rq, &def_root_domain);
 #endif
+		init_rq_hrtick(rq);
 		atomic_set(&rq->nr_iowait, 0);
-
-		array = &rq->rt.active;
-		for (j = 0; j < MAX_RT_PRIO; j++) {
-			INIT_LIST_HEAD(array->queue + j);
-			__clear_bit(j, array->bitmap);
-		}
 		highest_cpu = i;
-		/* delimiter for bitsearch: */
-		__set_bit(MAX_RT_PRIO, array->bitmap);
 	}
 
 	set_load_weight(&init_task);
@@ -6975,12 +7381,187 @@ void set_curr_task(int cpu, struct task_struct *p)
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
 
+#ifdef CONFIG_SMP
+/*
+ * distribute shares of all task groups among their schedulable entities,
+ * to reflect load distribution across cpus.
+ */
+static int rebalance_shares(struct sched_domain *sd, int this_cpu)
+{
+	struct cfs_rq *cfs_rq;
+	struct rq *rq = cpu_rq(this_cpu);
+	cpumask_t sdspan = sd->span;
+	int balanced = 1;
+
+	/* Walk thr' all the task groups that we have */
+	for_each_leaf_cfs_rq(rq, cfs_rq) {
+		int i;
+		unsigned long total_load = 0, total_shares;
+		struct task_group *tg = cfs_rq->tg;
+
+		/* Gather total task load of this group across cpus */
+		for_each_cpu_mask(i, sdspan)
+			total_load += tg->cfs_rq[i]->load.weight;
+
+		/* Nothing to do if this group has no load */
+		if (!total_load)
+			continue;
+
+		/*
+		 * tg->shares represents the number of cpu shares the task group
+		 * is eligible to hold on a single cpu. On N cpus, it is
+		 * eligible to hold (N * tg->shares) number of cpu shares.
+		 */
+		total_shares = tg->shares * cpus_weight(sdspan);
+
+		/*
+		 * redistribute total_shares across cpus as per the task load
+		 * distribution.
+		 */
+		for_each_cpu_mask(i, sdspan) {
+			unsigned long local_load, local_shares;
+
+			local_load = tg->cfs_rq[i]->load.weight;
+			local_shares = (local_load * total_shares) / total_load;
+			if (!local_shares)
+				local_shares = MIN_GROUP_SHARES;
+			if (local_shares == tg->se[i]->load.weight)
+				continue;
+
+			spin_lock_irq(&cpu_rq(i)->lock);
+			set_se_shares(tg->se[i], local_shares);
+			spin_unlock_irq(&cpu_rq(i)->lock);
+			balanced = 0;
+		}
+	}
+
+	return balanced;
+}
+
+/*
+ * How frequently should we rebalance_shares() across cpus?
+ *
+ * The more frequently we rebalance shares, the more accurate is the fairness
+ * of cpu bandwidth distribution between task groups. However higher frequency
+ * also implies increased scheduling overhead.
+ *
+ * sysctl_sched_min_bal_int_shares represents the minimum interval between
+ * consecutive calls to rebalance_shares() in the same sched domain.
+ *
+ * sysctl_sched_max_bal_int_shares represents the maximum interval between
+ * consecutive calls to rebalance_shares() in the same sched domain.
+ *
+ * These settings allows for the appropriate trade-off between accuracy of
+ * fairness and the associated overhead.
+ *
+ */
+
+/* default: 8ms, units: milliseconds */
+const_debug unsigned int sysctl_sched_min_bal_int_shares = 8;
+
+/* default: 128ms, units: milliseconds */
+const_debug unsigned int sysctl_sched_max_bal_int_shares = 128;
+
+/* kernel thread that runs rebalance_shares() periodically */
+static int load_balance_monitor(void *unused)
+{
+	unsigned int timeout = sysctl_sched_min_bal_int_shares;
+	struct sched_param schedparm;
+	int ret;
+
+	/*
+	 * We don't want this thread's execution to be limited by the shares
+	 * assigned to default group (init_task_group). Hence make it run
+	 * as a SCHED_RR RT task at the lowest priority.
+	 */
+	schedparm.sched_priority = 1;
+	ret = sched_setscheduler(current, SCHED_RR, &schedparm);
+	if (ret)
+		printk(KERN_ERR "Couldn't set SCHED_RR policy for load balance"
+				" monitor thread (error = %d) \n", ret);
+
+	while (!kthread_should_stop()) {
+		int i, cpu, balanced = 1;
+
+		/* Prevent cpus going down or coming up */
+		get_online_cpus();
+		/* lockout changes to doms_cur[] array */
+		lock_doms_cur();
+		/*
+		 * Enter a rcu read-side critical section to safely walk rq->sd
+		 * chain on various cpus and to walk task group list
+		 * (rq->leaf_cfs_rq_list) in rebalance_shares().
+		 */
+		rcu_read_lock();
+
+		for (i = 0; i < ndoms_cur; i++) {
+			cpumask_t cpumap = doms_cur[i];
+			struct sched_domain *sd = NULL, *sd_prev = NULL;
+
+			cpu = first_cpu(cpumap);
+
+			/* Find the highest domain at which to balance shares */
+			for_each_domain(cpu, sd) {
+				if (!(sd->flags & SD_LOAD_BALANCE))
+					continue;
+				sd_prev = sd;
+			}
+
+			sd = sd_prev;
+			/* sd == NULL? No load balance reqd in this domain */
+			if (!sd)
+				continue;
+
+			balanced &= rebalance_shares(sd, cpu);
+		}
+
+		rcu_read_unlock();
+
+		unlock_doms_cur();
+		put_online_cpus();
+
+		if (!balanced)
+			timeout = sysctl_sched_min_bal_int_shares;
+		else if (timeout < sysctl_sched_max_bal_int_shares)
+			timeout *= 2;
+
+		msleep_interruptible(timeout);
+	}
+
+	return 0;
+}
+#endif	/* CONFIG_SMP */
+
+static void free_sched_group(struct task_group *tg)
+{
+	int i;
+
+	for_each_possible_cpu(i) {
+		if (tg->cfs_rq)
+			kfree(tg->cfs_rq[i]);
+		if (tg->se)
+			kfree(tg->se[i]);
+		if (tg->rt_rq)
+			kfree(tg->rt_rq[i]);
+		if (tg->rt_se)
+			kfree(tg->rt_se[i]);
+	}
+
+	kfree(tg->cfs_rq);
+	kfree(tg->se);
+	kfree(tg->rt_rq);
+	kfree(tg->rt_se);
+	kfree(tg);
+}
+
 /* allocate runqueue etc for a new task group */
 struct task_group *sched_create_group(void)
 {
 	struct task_group *tg;
 	struct cfs_rq *cfs_rq;
 	struct sched_entity *se;
+	struct rt_rq *rt_rq;
+	struct sched_rt_entity *rt_se;
 	struct rq *rq;
 	int i;
 
@@ -6994,97 +7575,89 @@ struct task_group *sched_create_group(void)
 	tg->se = kzalloc(sizeof(se) * NR_CPUS, GFP_KERNEL);
 	if (!tg->se)
 		goto err;
+	tg->rt_rq = kzalloc(sizeof(rt_rq) * NR_CPUS, GFP_KERNEL);
+	if (!tg->rt_rq)
+		goto err;
+	tg->rt_se = kzalloc(sizeof(rt_se) * NR_CPUS, GFP_KERNEL);
+	if (!tg->rt_se)
+		goto err;
+
+	tg->shares = NICE_0_LOAD;
+	tg->rt_ratio = 0; /* XXX */
 
 	for_each_possible_cpu(i) {
 		rq = cpu_rq(i);
 
-		cfs_rq = kmalloc_node(sizeof(struct cfs_rq), GFP_KERNEL,
-							 cpu_to_node(i));
+		cfs_rq = kmalloc_node(sizeof(struct cfs_rq),
+				GFP_KERNEL|__GFP_ZERO, cpu_to_node(i));
 		if (!cfs_rq)
 			goto err;
 
-		se = kmalloc_node(sizeof(struct sched_entity), GFP_KERNEL,
-							cpu_to_node(i));
+		se = kmalloc_node(sizeof(struct sched_entity),
+				GFP_KERNEL|__GFP_ZERO, cpu_to_node(i));
 		if (!se)
 			goto err;
 
-		memset(cfs_rq, 0, sizeof(struct cfs_rq));
-		memset(se, 0, sizeof(struct sched_entity));
+		rt_rq = kmalloc_node(sizeof(struct rt_rq),
+				GFP_KERNEL|__GFP_ZERO, cpu_to_node(i));
+		if (!rt_rq)
+			goto err;
 
-		tg->cfs_rq[i] = cfs_rq;
-		init_cfs_rq(cfs_rq, rq);
-		cfs_rq->tg = tg;
+		rt_se = kmalloc_node(sizeof(struct sched_rt_entity),
+				GFP_KERNEL|__GFP_ZERO, cpu_to_node(i));
+		if (!rt_se)
+			goto err;
 
-		tg->se[i] = se;
-		se->cfs_rq = &rq->cfs;
-		se->my_q = cfs_rq;
-		se->load.weight = NICE_0_LOAD;
-		se->load.inv_weight = div64_64(1ULL<<32, NICE_0_LOAD);
-		se->parent = NULL;
+		init_tg_cfs_entry(rq, tg, cfs_rq, se, i, 0);
+		init_tg_rt_entry(rq, tg, rt_rq, rt_se, i, 0);
 	}
 
+	lock_task_group_list();
 	for_each_possible_cpu(i) {
 		rq = cpu_rq(i);
 		cfs_rq = tg->cfs_rq[i];
 		list_add_rcu(&cfs_rq->leaf_cfs_rq_list, &rq->leaf_cfs_rq_list);
+		rt_rq = tg->rt_rq[i];
+		list_add_rcu(&rt_rq->leaf_rt_rq_list, &rq->leaf_rt_rq_list);
 	}
-
-	tg->shares = NICE_0_LOAD;
-	spin_lock_init(&tg->lock);
+	list_add_rcu(&tg->list, &task_groups);
+	unlock_task_group_list();
 
 	return tg;
 
 err:
-	for_each_possible_cpu(i) {
-		if (tg->cfs_rq)
-			kfree(tg->cfs_rq[i]);
-		if (tg->se)
-			kfree(tg->se[i]);
-	}
-	kfree(tg->cfs_rq);
-	kfree(tg->se);
-	kfree(tg);
-
+	free_sched_group(tg);
 	return ERR_PTR(-ENOMEM);
 }
 
 /* rcu callback to free various structures associated with a task group */
-static void free_sched_group(struct rcu_head *rhp)
+static void free_sched_group_rcu(struct rcu_head *rhp)
 {
-	struct task_group *tg = container_of(rhp, struct task_group, rcu);
-	struct cfs_rq *cfs_rq;
-	struct sched_entity *se;
-	int i;
-
 	/* now it should be safe to free those cfs_rqs */
-	for_each_possible_cpu(i) {
-		cfs_rq = tg->cfs_rq[i];
-		kfree(cfs_rq);
-
-		se = tg->se[i];
-		kfree(se);
-	}
-
-	kfree(tg->cfs_rq);
-	kfree(tg->se);
-	kfree(tg);
+	free_sched_group(container_of(rhp, struct task_group, rcu));
 }
 
 /* Destroy runqueue etc associated with a task group */
 void sched_destroy_group(struct task_group *tg)
 {
 	struct cfs_rq *cfs_rq = NULL;
+	struct rt_rq *rt_rq = NULL;
 	int i;
 
+	lock_task_group_list();
 	for_each_possible_cpu(i) {
 		cfs_rq = tg->cfs_rq[i];
 		list_del_rcu(&cfs_rq->leaf_cfs_rq_list);
+		rt_rq = tg->rt_rq[i];
+		list_del_rcu(&rt_rq->leaf_rt_rq_list);
 	}
+	list_del_rcu(&tg->list);
+	unlock_task_group_list();
 
 	BUG_ON(!cfs_rq);
 
 	/* wait for possible concurrent references to cfs_rqs complete */
-	call_rcu(&tg->rcu, free_sched_group);
+	call_rcu(&tg->rcu, free_sched_group_rcu);
 }
 
 /* change task's runqueue when it moves between groups.
@@ -7100,11 +7673,6 @@ void sched_move_task(struct task_struct *tsk)
 
 	rq = task_rq_lock(tsk, &flags);
 
-	if (tsk->sched_class != &fair_sched_class) {
-		set_task_cfs_rq(tsk, task_cpu(tsk));
-		goto done;
-	}
-
 	update_rq_clock(rq);
 
 	running = task_current(rq, tsk);
@@ -7116,7 +7684,7 @@ void sched_move_task(struct task_struct *tsk)
 			tsk->sched_class->put_prev_task(rq, tsk);
 	}
 
-	set_task_cfs_rq(tsk, task_cpu(tsk));
+	set_task_rq(tsk, task_cpu(tsk));
 
 	if (on_rq) {
 		if (unlikely(running))
@@ -7124,53 +7692,82 @@ void sched_move_task(struct task_struct *tsk)
 		enqueue_task(rq, tsk, 0);
 	}
 
-done:
 	task_rq_unlock(rq, &flags);
 }
 
+/* rq->lock to be locked by caller */
 static void set_se_shares(struct sched_entity *se, unsigned long shares)
 {
 	struct cfs_rq *cfs_rq = se->cfs_rq;
 	struct rq *rq = cfs_rq->rq;
 	int on_rq;
 
-	spin_lock_irq(&rq->lock);
+	if (!shares)
+		shares = MIN_GROUP_SHARES;
 
 	on_rq = se->on_rq;
-	if (on_rq)
+	if (on_rq) {
 		dequeue_entity(cfs_rq, se, 0);
+		dec_cpu_load(rq, se->load.weight);
+	}
 
 	se->load.weight = shares;
 	se->load.inv_weight = div64_64((1ULL<<32), shares);
 
-	if (on_rq)
+	if (on_rq) {
 		enqueue_entity(cfs_rq, se, 0);
-
-	spin_unlock_irq(&rq->lock);
+		inc_cpu_load(rq, se->load.weight);
+	}
 }
 
 int sched_group_set_shares(struct task_group *tg, unsigned long shares)
 {
 	int i;
+	struct cfs_rq *cfs_rq;
+	struct rq *rq;
+
+	lock_task_group_list();
+	if (tg->shares == shares)
+		goto done;
+
+	if (shares < MIN_GROUP_SHARES)
+		shares = MIN_GROUP_SHARES;
 
 	/*
-	 * A weight of 0 or 1 can cause arithmetics problems.
-	 * (The default weight is 1024 - so there's no practical
-	 *  limitation from this.)
+	 * Prevent any load balance activity (rebalance_shares,
+	 * load_balance_fair) from referring to this group first,
+	 * by taking it off the rq->leaf_cfs_rq_list on each cpu.
 	 */
-	if (shares < 2)
-		shares = 2;
+	for_each_possible_cpu(i) {
+		cfs_rq = tg->cfs_rq[i];
+		list_del_rcu(&cfs_rq->leaf_cfs_rq_list);
+	}
 
-	spin_lock(&tg->lock);
-	if (tg->shares == shares)
-		goto done;
+	/* wait for any ongoing reference to this group to finish */
+	synchronize_sched();
 
+	/*
+	 * Now we are free to modify the group's share on each cpu
+	 * w/o tripping rebalance_share or load_balance_fair.
+	 */
 	tg->shares = shares;
-	for_each_possible_cpu(i)
+	for_each_possible_cpu(i) {
+		spin_lock_irq(&cpu_rq(i)->lock);
 		set_se_shares(tg->se[i], shares);
+		spin_unlock_irq(&cpu_rq(i)->lock);
+	}
 
+	/*
+	 * Enable load balance activity on this group, by inserting it back on
+	 * each cpu's rq->leaf_cfs_rq_list.
+	 */
+	for_each_possible_cpu(i) {
+		rq = cpu_rq(i);
+		cfs_rq = tg->cfs_rq[i];
+		list_add_rcu(&cfs_rq->leaf_cfs_rq_list, &rq->leaf_cfs_rq_list);
+	}
 done:
-	spin_unlock(&tg->lock);
+	unlock_task_group_list();
 	return 0;
 }
 
@@ -7179,6 +7776,31 @@ unsigned long sched_group_shares(struct task_group *tg)
 	return tg->shares;
 }
 
+/*
+ * Ensure the total rt_ratio <= sysctl_sched_rt_ratio
+ */
+int sched_group_set_rt_ratio(struct task_group *tg, unsigned long rt_ratio)
+{
+	struct task_group *tgi;
+	unsigned long total = 0;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(tgi, &task_groups, list)
+		total += tgi->rt_ratio;
+	rcu_read_unlock();
+
+	if (total + rt_ratio - tg->rt_ratio > sysctl_sched_rt_ratio)
+		return -EINVAL;
+
+	tg->rt_ratio = rt_ratio;
+	return 0;
+}
+
+unsigned long sched_group_rt_ratio(struct task_group *tg)
+{
+	return tg->rt_ratio;
+}
+
 #endif	/* CONFIG_FAIR_GROUP_SCHED */
 
 #ifdef CONFIG_FAIR_CGROUP_SCHED
@@ -7254,12 +7876,30 @@ static u64 cpu_shares_read_uint(struct cgroup *cgrp, struct cftype *cft)
 	return (u64) tg->shares;
 }
 
+static int cpu_rt_ratio_write_uint(struct cgroup *cgrp, struct cftype *cftype,
+		u64 rt_ratio_val)
+{
+	return sched_group_set_rt_ratio(cgroup_tg(cgrp), rt_ratio_val);
+}
+
+static u64 cpu_rt_ratio_read_uint(struct cgroup *cgrp, struct cftype *cft)
+{
+	struct task_group *tg = cgroup_tg(cgrp);
+
+	return (u64) tg->rt_ratio;
+}
+
 static struct cftype cpu_files[] = {
 	{
 		.name = "shares",
 		.read_uint = cpu_shares_read_uint,
 		.write_uint = cpu_shares_write_uint,
 	},
+	{
+		.name = "rt_ratio",
+		.read_uint = cpu_rt_ratio_read_uint,
+		.write_uint = cpu_rt_ratio_write_uint,
+	},
 };
 
 static int cpu_cgroup_populate(struct cgroup_subsys *ss, struct cgroup *cont)
diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c
index 80fbbfc0429..4b5e24cf2f4 100644
--- a/kernel/sched_debug.c
+++ b/kernel/sched_debug.c
@@ -179,6 +179,7 @@ static void print_cpu(struct seq_file *m, int cpu)
 	PN(prev_clock_raw);
 	P(clock_warps);
 	P(clock_overflows);
+	P(clock_underflows);
 	P(clock_deep_idle_events);
 	PN(clock_max_delta);
 	P(cpu_load[0]);
@@ -299,6 +300,8 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m)
 	PN(se.exec_max);
 	PN(se.slice_max);
 	PN(se.wait_max);
+	PN(se.wait_sum);
+	P(se.wait_count);
 	P(sched_info.bkl_count);
 	P(se.nr_migrations);
 	P(se.nr_migrations_cold);
@@ -366,6 +369,8 @@ void proc_sched_set_task(struct task_struct *p)
 {
 #ifdef CONFIG_SCHEDSTATS
 	p->se.wait_max				= 0;
+	p->se.wait_sum				= 0;
+	p->se.wait_count			= 0;
 	p->se.sleep_max				= 0;
 	p->se.sum_sleep_runtime			= 0;
 	p->se.block_max				= 0;
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index da7c061e720..72e25c7a3a1 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -20,6 +20,8 @@
  *  Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
  */
 
+#include <linux/latencytop.h>
+
 /*
  * Targeted preemption latency for CPU-bound tasks:
  * (default: 20ms * (1 + ilog(ncpus)), units: nanoseconds)
@@ -248,8 +250,8 @@ static u64 __sched_period(unsigned long nr_running)
 	unsigned long nr_latency = sched_nr_latency;
 
 	if (unlikely(nr_running > nr_latency)) {
+		period = sysctl_sched_min_granularity;
 		period *= nr_running;
-		do_div(period, nr_latency);
 	}
 
 	return period;
@@ -383,6 +385,9 @@ update_stats_wait_end(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
 	schedstat_set(se->wait_max, max(se->wait_max,
 			rq_of(cfs_rq)->clock - se->wait_start));
+	schedstat_set(se->wait_count, se->wait_count + 1);
+	schedstat_set(se->wait_sum, se->wait_sum +
+			rq_of(cfs_rq)->clock - se->wait_start);
 	schedstat_set(se->wait_start, 0);
 }
 
@@ -434,6 +439,7 @@ static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
 #ifdef CONFIG_SCHEDSTATS
 	if (se->sleep_start) {
 		u64 delta = rq_of(cfs_rq)->clock - se->sleep_start;
+		struct task_struct *tsk = task_of(se);
 
 		if ((s64)delta < 0)
 			delta = 0;
@@ -443,9 +449,12 @@ static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
 
 		se->sleep_start = 0;
 		se->sum_sleep_runtime += delta;
+
+		account_scheduler_latency(tsk, delta >> 10, 1);
 	}
 	if (se->block_start) {
 		u64 delta = rq_of(cfs_rq)->clock - se->block_start;
+		struct task_struct *tsk = task_of(se);
 
 		if ((s64)delta < 0)
 			delta = 0;
@@ -462,11 +471,11 @@ static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
 		 * time that the task spent sleeping:
 		 */
 		if (unlikely(prof_on == SLEEP_PROFILING)) {
-			struct task_struct *tsk = task_of(se);
 
 			profile_hits(SLEEP_PROFILING, (void *)get_wchan(tsk),
 				     delta >> 20);
 		}
+		account_scheduler_latency(tsk, delta >> 10, 0);
 	}
 #endif
 }
@@ -642,13 +651,29 @@ static void put_prev_entity(struct cfs_rq *cfs_rq, struct sched_entity *prev)
 	cfs_rq->curr = NULL;
 }
 
-static void entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr)
+static void
+entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued)
 {
 	/*
 	 * Update run-time statistics of the 'current'.
 	 */
 	update_curr(cfs_rq);
 
+#ifdef CONFIG_SCHED_HRTICK
+	/*
+	 * queued ticks are scheduled to match the slice, so don't bother
+	 * validating it and just reschedule.
+	 */
+	if (queued)
+		return resched_task(rq_of(cfs_rq)->curr);
+	/*
+	 * don't let the period tick interfere with the hrtick preemption
+	 */
+	if (!sched_feat(DOUBLE_TICK) &&
+			hrtimer_active(&rq_of(cfs_rq)->hrtick_timer))
+		return;
+#endif
+
 	if (cfs_rq->nr_running > 1 || !sched_feat(WAKEUP_PREEMPT))
 		check_preempt_tick(cfs_rq, curr);
 }
@@ -690,7 +715,7 @@ static inline struct cfs_rq *cpu_cfs_rq(struct cfs_rq *cfs_rq, int this_cpu)
 
 /* Iterate thr' all leaf cfs_rq's on a runqueue */
 #define for_each_leaf_cfs_rq(rq, cfs_rq) \
-	list_for_each_entry(cfs_rq, &rq->leaf_cfs_rq_list, leaf_cfs_rq_list)
+	list_for_each_entry_rcu(cfs_rq, &rq->leaf_cfs_rq_list, leaf_cfs_rq_list)
 
 /* Do the two (enqueued) entities belong to the same group ? */
 static inline int
@@ -707,6 +732,8 @@ static inline struct sched_entity *parent_entity(struct sched_entity *se)
 	return se->parent;
 }
 
+#define GROUP_IMBALANCE_PCT	20
+
 #else	/* CONFIG_FAIR_GROUP_SCHED */
 
 #define for_each_sched_entity(se) \
@@ -752,6 +779,43 @@ static inline struct sched_entity *parent_entity(struct sched_entity *se)
 
 #endif	/* CONFIG_FAIR_GROUP_SCHED */
 
+#ifdef CONFIG_SCHED_HRTICK
+static void hrtick_start_fair(struct rq *rq, struct task_struct *p)
+{
+	int requeue = rq->curr == p;
+	struct sched_entity *se = &p->se;
+	struct cfs_rq *cfs_rq = cfs_rq_of(se);
+
+	WARN_ON(task_rq(p) != rq);
+
+	if (hrtick_enabled(rq) && cfs_rq->nr_running > 1) {
+		u64 slice = sched_slice(cfs_rq, se);
+		u64 ran = se->sum_exec_runtime - se->prev_sum_exec_runtime;
+		s64 delta = slice - ran;
+
+		if (delta < 0) {
+			if (rq->curr == p)
+				resched_task(p);
+			return;
+		}
+
+		/*
+		 * Don't schedule slices shorter than 10000ns, that just
+		 * doesn't make sense. Rely on vruntime for fairness.
+		 */
+		if (!requeue)
+			delta = max(10000LL, delta);
+
+		hrtick_start(rq, delta, requeue);
+	}
+}
+#else
+static inline void
+hrtick_start_fair(struct rq *rq, struct task_struct *p)
+{
+}
+#endif
+
 /*
  * The enqueue_task method is called before nr_running is
  * increased. Here we update the fair scheduling stats and
@@ -760,15 +824,28 @@ static inline struct sched_entity *parent_entity(struct sched_entity *se)
 static void enqueue_task_fair(struct rq *rq, struct task_struct *p, int wakeup)
 {
 	struct cfs_rq *cfs_rq;
-	struct sched_entity *se = &p->se;
+	struct sched_entity *se = &p->se,
+			    *topse = NULL;	/* Highest schedulable entity */
+	int incload = 1;
 
 	for_each_sched_entity(se) {
-		if (se->on_rq)
+		topse = se;
+		if (se->on_rq) {
+			incload = 0;
 			break;
+		}
 		cfs_rq = cfs_rq_of(se);
 		enqueue_entity(cfs_rq, se, wakeup);
 		wakeup = 1;
 	}
+	/* Increment cpu load if we just enqueued the first task of a group on
+	 * 'rq->cpu'. 'topse' represents the group to which task 'p' belongs
+	 * at the highest grouping level.
+	 */
+	if (incload)
+		inc_cpu_load(rq, topse->load.weight);
+
+	hrtick_start_fair(rq, rq->curr);
 }
 
 /*
@@ -779,16 +856,30 @@ static void enqueue_task_fair(struct rq *rq, struct task_struct *p, int wakeup)
 static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int sleep)
 {
 	struct cfs_rq *cfs_rq;
-	struct sched_entity *se = &p->se;
+	struct sched_entity *se = &p->se,
+			    *topse = NULL; 	/* Highest schedulable entity */
+	int decload = 1;
 
 	for_each_sched_entity(se) {
+		topse = se;
 		cfs_rq = cfs_rq_of(se);
 		dequeue_entity(cfs_rq, se, sleep);
 		/* Don't dequeue parent if it has other entities besides us */
-		if (cfs_rq->load.weight)
+		if (cfs_rq->load.weight) {
+			if (parent_entity(se))
+				decload = 0;
 			break;
+		}
 		sleep = 1;
 	}
+	/* Decrement cpu load if we just dequeued the last task of a group on
+	 * 'rq->cpu'. 'topse' represents the group to which task 'p' belongs
+	 * at the highest grouping level.
+	 */
+	if (decload)
+		dec_cpu_load(rq, topse->load.weight);
+
+	hrtick_start_fair(rq, rq->curr);
 }
 
 /*
@@ -836,6 +927,154 @@ static void yield_task_fair(struct rq *rq)
 }
 
 /*
+ * wake_idle() will wake a task on an idle cpu if task->cpu is
+ * not idle and an idle cpu is available.  The span of cpus to
+ * search starts with cpus closest then further out as needed,
+ * so we always favor a closer, idle cpu.
+ *
+ * Returns the CPU we should wake onto.
+ */
+#if defined(ARCH_HAS_SCHED_WAKE_IDLE)
+static int wake_idle(int cpu, struct task_struct *p)
+{
+	cpumask_t tmp;
+	struct sched_domain *sd;
+	int i;
+
+	/*
+	 * If it is idle, then it is the best cpu to run this task.
+	 *
+	 * This cpu is also the best, if it has more than one task already.
+	 * Siblings must be also busy(in most cases) as they didn't already
+	 * pickup the extra load from this cpu and hence we need not check
+	 * sibling runqueue info. This will avoid the checks and cache miss
+	 * penalities associated with that.
+	 */
+	if (idle_cpu(cpu) || cpu_rq(cpu)->nr_running > 1)
+		return cpu;
+
+	for_each_domain(cpu, sd) {
+		if (sd->flags & SD_WAKE_IDLE) {
+			cpus_and(tmp, sd->span, p->cpus_allowed);
+			for_each_cpu_mask(i, tmp) {
+				if (idle_cpu(i)) {
+					if (i != task_cpu(p)) {
+						schedstat_inc(p,
+						       se.nr_wakeups_idle);
+					}
+					return i;
+				}
+			}
+		} else {
+			break;
+		}
+	}
+	return cpu;
+}
+#else
+static inline int wake_idle(int cpu, struct task_struct *p)
+{
+	return cpu;
+}
+#endif
+
+#ifdef CONFIG_SMP
+static int select_task_rq_fair(struct task_struct *p, int sync)
+{
+	int cpu, this_cpu;
+	struct rq *rq;
+	struct sched_domain *sd, *this_sd = NULL;
+	int new_cpu;
+
+	cpu      = task_cpu(p);
+	rq       = task_rq(p);
+	this_cpu = smp_processor_id();
+	new_cpu  = cpu;
+
+	if (cpu == this_cpu)
+		goto out_set_cpu;
+
+	for_each_domain(this_cpu, sd) {
+		if (cpu_isset(cpu, sd->span)) {
+			this_sd = sd;
+			break;
+		}
+	}
+
+	if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed)))
+		goto out_set_cpu;
+
+	/*
+	 * Check for affine wakeup and passive balancing possibilities.
+	 */
+	if (this_sd) {
+		int idx = this_sd->wake_idx;
+		unsigned int imbalance;
+		unsigned long load, this_load;
+
+		imbalance = 100 + (this_sd->imbalance_pct - 100) / 2;
+
+		load = source_load(cpu, idx);
+		this_load = target_load(this_cpu, idx);
+
+		new_cpu = this_cpu; /* Wake to this CPU if we can */
+
+		if (this_sd->flags & SD_WAKE_AFFINE) {
+			unsigned long tl = this_load;
+			unsigned long tl_per_task;
+
+			/*
+			 * Attract cache-cold tasks on sync wakeups:
+			 */
+			if (sync && !task_hot(p, rq->clock, this_sd))
+				goto out_set_cpu;
+
+			schedstat_inc(p, se.nr_wakeups_affine_attempts);
+			tl_per_task = cpu_avg_load_per_task(this_cpu);
+
+			/*
+			 * If sync wakeup then subtract the (maximum possible)
+			 * effect of the currently running task from the load
+			 * of the current CPU:
+			 */
+			if (sync)
+				tl -= current->se.load.weight;
+
+			if ((tl <= load &&
+				tl + target_load(cpu, idx) <= tl_per_task) ||
+			       100*(tl + p->se.load.weight) <= imbalance*load) {
+				/*
+				 * This domain has SD_WAKE_AFFINE and
+				 * p is cache cold in this domain, and
+				 * there is no bad imbalance.
+				 */
+				schedstat_inc(this_sd, ttwu_move_affine);
+				schedstat_inc(p, se.nr_wakeups_affine);
+				goto out_set_cpu;
+			}
+		}
+
+		/*
+		 * Start passive balancing when half the imbalance_pct
+		 * limit is reached.
+		 */
+		if (this_sd->flags & SD_WAKE_BALANCE) {
+			if (imbalance*this_load <= 100*load) {
+				schedstat_inc(this_sd, ttwu_move_balance);
+				schedstat_inc(p, se.nr_wakeups_passive);
+				goto out_set_cpu;
+			}
+		}
+	}
+
+	new_cpu = cpu; /* Could not wake to this_cpu. Wake to cpu instead */
+out_set_cpu:
+	return wake_idle(new_cpu, p);
+}
+#endif /* CONFIG_SMP */
+
+
+/*
  * Preempt the current task with a newly woken task if needed:
  */
 static void check_preempt_wakeup(struct rq *rq, struct task_struct *p)
@@ -876,6 +1115,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p)
 
 static struct task_struct *pick_next_task_fair(struct rq *rq)
 {
+	struct task_struct *p;
 	struct cfs_rq *cfs_rq = &rq->cfs;
 	struct sched_entity *se;
 
@@ -887,7 +1127,10 @@ static struct task_struct *pick_next_task_fair(struct rq *rq)
 		cfs_rq = group_cfs_rq(se);
 	} while (cfs_rq);
 
-	return task_of(se);
+	p = task_of(se);
+	hrtick_start_fair(rq, p);
+
+	return p;
 }
 
 /*
@@ -944,25 +1187,6 @@ static struct task_struct *load_balance_next_fair(void *arg)
 	return __load_balance_iterator(cfs_rq, cfs_rq->rb_load_balance_curr);
 }
 
-#ifdef CONFIG_FAIR_GROUP_SCHED
-static int cfs_rq_best_prio(struct cfs_rq *cfs_rq)
-{
-	struct sched_entity *curr;
-	struct task_struct *p;
-
-	if (!cfs_rq->nr_running)
-		return MAX_PRIO;
-
-	curr = cfs_rq->curr;
-	if (!curr)
-		curr = __pick_next_entity(cfs_rq);
-
-	p = task_of(curr);
-
-	return p->prio;
-}
-#endif
-
 static unsigned long
 load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		  unsigned long max_load_move,
@@ -972,28 +1196,45 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 	struct cfs_rq *busy_cfs_rq;
 	long rem_load_move = max_load_move;
 	struct rq_iterator cfs_rq_iterator;
+	unsigned long load_moved;
 
 	cfs_rq_iterator.start = load_balance_start_fair;
 	cfs_rq_iterator.next = load_balance_next_fair;
 
 	for_each_leaf_cfs_rq(busiest, busy_cfs_rq) {
 #ifdef CONFIG_FAIR_GROUP_SCHED
-		struct cfs_rq *this_cfs_rq;
-		long imbalance;
-		unsigned long maxload;
+		struct cfs_rq *this_cfs_rq = busy_cfs_rq->tg->cfs_rq[this_cpu];
+		unsigned long maxload, task_load, group_weight;
+		unsigned long thisload, per_task_load;
+		struct sched_entity *se = busy_cfs_rq->tg->se[busiest->cpu];
+
+		task_load = busy_cfs_rq->load.weight;
+		group_weight = se->load.weight;
 
-		this_cfs_rq = cpu_cfs_rq(busy_cfs_rq, this_cpu);
+		/*
+		 * 'group_weight' is contributed by tasks of total weight
+		 * 'task_load'. To move 'rem_load_move' worth of weight only,
+		 * we need to move a maximum task load of:
+		 *
+		 * 	maxload = (remload / group_weight) * task_load;
+		 */
+		maxload = (rem_load_move * task_load) / group_weight;
 
-		imbalance = busy_cfs_rq->load.weight - this_cfs_rq->load.weight;
-		/* Don't pull if this_cfs_rq has more load than busy_cfs_rq */
-		if (imbalance <= 0)
+		if (!maxload || !task_load)
 			continue;
 
-		/* Don't pull more than imbalance/2 */
-		imbalance /= 2;
-		maxload = min(rem_load_move, imbalance);
+		per_task_load = task_load / busy_cfs_rq->nr_running;
+		/*
+		 * balance_tasks will try to forcibly move atleast one task if
+		 * possible (because of SCHED_LOAD_SCALE_FUZZ). Avoid that if
+		 * maxload is less than GROUP_IMBALANCE_FUZZ% the per_task_load.
+		 */
+		 if (100 * maxload < GROUP_IMBALANCE_PCT * per_task_load)
+			continue;
 
-		*this_best_prio = cfs_rq_best_prio(this_cfs_rq);
+		/* Disable priority-based load balance */
+		*this_best_prio = 0;
+		thisload = this_cfs_rq->load.weight;
 #else
 # define maxload rem_load_move
 #endif
@@ -1002,11 +1243,33 @@ load_balance_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		 * load_balance_[start|next]_fair iterators
 		 */
 		cfs_rq_iterator.arg = busy_cfs_rq;
-		rem_load_move -= balance_tasks(this_rq, this_cpu, busiest,
+		load_moved = balance_tasks(this_rq, this_cpu, busiest,
 					       maxload, sd, idle, all_pinned,
 					       this_best_prio,
 					       &cfs_rq_iterator);
 
+#ifdef CONFIG_FAIR_GROUP_SCHED
+		/*
+		 * load_moved holds the task load that was moved. The
+		 * effective (group) weight moved would be:
+		 * 	load_moved_eff = load_moved/task_load * group_weight;
+		 */
+		load_moved = (group_weight * load_moved) / task_load;
+
+		/* Adjust shares on both cpus to reflect load_moved */
+		group_weight -= load_moved;
+		set_se_shares(se, group_weight);
+
+		se = busy_cfs_rq->tg->se[this_cpu];
+		if (!thisload)
+			group_weight = load_moved;
+		else
+			group_weight = se->load.weight + load_moved;
+		set_se_shares(se, group_weight);
+#endif
+
+		rem_load_move -= load_moved;
+
 		if (rem_load_move <= 0)
 			break;
 	}
@@ -1042,14 +1305,14 @@ move_one_task_fair(struct rq *this_rq, int this_cpu, struct rq *busiest,
 /*
  * scheduler tick hitting a task of our scheduling class:
  */
-static void task_tick_fair(struct rq *rq, struct task_struct *curr)
+static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued)
 {
 	struct cfs_rq *cfs_rq;
 	struct sched_entity *se = &curr->se;
 
 	for_each_sched_entity(se) {
 		cfs_rq = cfs_rq_of(se);
-		entity_tick(cfs_rq, se);
+		entity_tick(cfs_rq, se, queued);
 	}
 }
 
@@ -1087,6 +1350,42 @@ static void task_new_fair(struct rq *rq, struct task_struct *p)
 	resched_task(rq->curr);
 }
 
+/*
+ * Priority of the task has changed. Check to see if we preempt
+ * the current task.
+ */
+static void prio_changed_fair(struct rq *rq, struct task_struct *p,
+			      int oldprio, int running)
+{
+	/*
+	 * Reschedule if we are currently running on this runqueue and
+	 * our priority decreased, or if we are not currently running on
+	 * this runqueue and our priority is higher than the current's
+	 */
+	if (running) {
+		if (p->prio > oldprio)
+			resched_task(rq->curr);
+	} else
+		check_preempt_curr(rq, p);
+}
+
+/*
+ * We switched to the sched_fair class.
+ */
+static void switched_to_fair(struct rq *rq, struct task_struct *p,
+			     int running)
+{
+	/*
+	 * We were most likely switched from sched_rt, so
+	 * kick off the schedule if running, otherwise just see
+	 * if we can still preempt the current task.
+	 */
+	if (running)
+		resched_task(rq->curr);
+	else
+		check_preempt_curr(rq, p);
+}
+
 /* Account for a task changing its policy or group.
  *
  * This routine is mostly called to set cfs_rq->curr field when a task
@@ -1108,6 +1407,9 @@ static const struct sched_class fair_sched_class = {
 	.enqueue_task		= enqueue_task_fair,
 	.dequeue_task		= dequeue_task_fair,
 	.yield_task		= yield_task_fair,
+#ifdef CONFIG_SMP
+	.select_task_rq		= select_task_rq_fair,
+#endif /* CONFIG_SMP */
 
 	.check_preempt_curr	= check_preempt_wakeup,
 
@@ -1122,6 +1424,9 @@ static const struct sched_class fair_sched_class = {
 	.set_curr_task          = set_curr_task_fair,
 	.task_tick		= task_tick_fair,
 	.task_new		= task_new_fair,
+
+	.prio_changed		= prio_changed_fair,
+	.switched_to		= switched_to_fair,
 };
 
 #ifdef CONFIG_SCHED_DEBUG
@@ -1132,7 +1437,9 @@ static void print_cfs_stats(struct seq_file *m, int cpu)
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	print_cfs_rq(m, cpu, &cpu_rq(cpu)->cfs);
 #endif
+	rcu_read_lock();
 	for_each_leaf_cfs_rq(cpu_rq(cpu), cfs_rq)
 		print_cfs_rq(m, cpu, cfs_rq);
+	rcu_read_unlock();
 }
 #endif
diff --git a/kernel/sched_idletask.c b/kernel/sched_idletask.c
index bf9c25c15b8..2bcafa37563 100644
--- a/kernel/sched_idletask.c
+++ b/kernel/sched_idletask.c
@@ -5,6 +5,12 @@
  *  handled in sched_fair.c)
  */
 
+#ifdef CONFIG_SMP
+static int select_task_rq_idle(struct task_struct *p, int sync)
+{
+	return task_cpu(p); /* IDLE tasks as never migrated */
+}
+#endif /* CONFIG_SMP */
 /*
  * Idle tasks are unconditionally rescheduled:
  */
@@ -55,7 +61,7 @@ move_one_task_idle(struct rq *this_rq, int this_cpu, struct rq *busiest,
 }
 #endif
 
-static void task_tick_idle(struct rq *rq, struct task_struct *curr)
+static void task_tick_idle(struct rq *rq, struct task_struct *curr, int queued)
 {
 }
 
@@ -63,6 +69,33 @@ static void set_curr_task_idle(struct rq *rq)
 {
 }
 
+static void switched_to_idle(struct rq *rq, struct task_struct *p,
+			     int running)
+{
+	/* Can this actually happen?? */
+	if (running)
+		resched_task(rq->curr);
+	else
+		check_preempt_curr(rq, p);
+}
+
+static void prio_changed_idle(struct rq *rq, struct task_struct *p,
+			      int oldprio, int running)
+{
+	/* This can happen for hot plug CPUS */
+
+	/*
+	 * Reschedule if we are currently running on this runqueue and
+	 * our priority decreased, or if we are not currently running on
+	 * this runqueue and our priority is higher than the current's
+	 */
+	if (running) {
+		if (p->prio > oldprio)
+			resched_task(rq->curr);
+	} else
+		check_preempt_curr(rq, p);
+}
+
 /*
  * Simple, special scheduling class for the per-CPU idle tasks:
  */
@@ -72,6 +105,9 @@ const struct sched_class idle_sched_class = {
 
 	/* dequeue is not valid, we print a debug message there: */
 	.dequeue_task		= dequeue_task_idle,
+#ifdef CONFIG_SMP
+	.select_task_rq		= select_task_rq_idle,
+#endif /* CONFIG_SMP */
 
 	.check_preempt_curr	= check_preempt_curr_idle,
 
@@ -85,5 +121,9 @@ const struct sched_class idle_sched_class = {
 
 	.set_curr_task          = set_curr_task_idle,
 	.task_tick		= task_tick_idle,
+
+	.prio_changed		= prio_changed_idle,
+	.switched_to		= switched_to_idle,
+
 	/* no .task_new for idle tasks */
 };
diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 9ba3daa0347..274b40d7bef 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -3,6 +3,217 @@
  * policies)
  */
 
+#ifdef CONFIG_SMP
+
+static inline int rt_overloaded(struct rq *rq)
+{
+	return atomic_read(&rq->rd->rto_count);
+}
+
+static inline void rt_set_overload(struct rq *rq)
+{
+	cpu_set(rq->cpu, rq->rd->rto_mask);
+	/*
+	 * Make sure the mask is visible before we set
+	 * the overload count. That is checked to determine
+	 * if we should look at the mask. It would be a shame
+	 * if we looked at the mask, but the mask was not
+	 * updated yet.
+	 */
+	wmb();
+	atomic_inc(&rq->rd->rto_count);
+}
+
+static inline void rt_clear_overload(struct rq *rq)
+{
+	/* the order here really doesn't matter */
+	atomic_dec(&rq->rd->rto_count);
+	cpu_clear(rq->cpu, rq->rd->rto_mask);
+}
+
+static void update_rt_migration(struct rq *rq)
+{
+	if (rq->rt.rt_nr_migratory && (rq->rt.rt_nr_running > 1)) {
+		if (!rq->rt.overloaded) {
+			rt_set_overload(rq);
+			rq->rt.overloaded = 1;
+		}
+	} else if (rq->rt.overloaded) {
+		rt_clear_overload(rq);
+		rq->rt.overloaded = 0;
+	}
+}
+#endif /* CONFIG_SMP */
+
+static inline struct task_struct *rt_task_of(struct sched_rt_entity *rt_se)
+{
+	return container_of(rt_se, struct task_struct, rt);
+}
+
+static inline int on_rt_rq(struct sched_rt_entity *rt_se)
+{
+	return !list_empty(&rt_se->run_list);
+}
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+
+static inline unsigned int sched_rt_ratio(struct rt_rq *rt_rq)
+{
+	if (!rt_rq->tg)
+		return SCHED_RT_FRAC;
+
+	return rt_rq->tg->rt_ratio;
+}
+
+#define for_each_leaf_rt_rq(rt_rq, rq) \
+	list_for_each_entry(rt_rq, &rq->leaf_rt_rq_list, leaf_rt_rq_list)
+
+static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq)
+{
+	return rt_rq->rq;
+}
+
+static inline struct rt_rq *rt_rq_of_se(struct sched_rt_entity *rt_se)
+{
+	return rt_se->rt_rq;
+}
+
+#define for_each_sched_rt_entity(rt_se) \
+	for (; rt_se; rt_se = rt_se->parent)
+
+static inline struct rt_rq *group_rt_rq(struct sched_rt_entity *rt_se)
+{
+	return rt_se->my_q;
+}
+
+static void enqueue_rt_entity(struct sched_rt_entity *rt_se);
+static void dequeue_rt_entity(struct sched_rt_entity *rt_se);
+
+static void sched_rt_ratio_enqueue(struct rt_rq *rt_rq)
+{
+	struct sched_rt_entity *rt_se = rt_rq->rt_se;
+
+	if (rt_se && !on_rt_rq(rt_se) && rt_rq->rt_nr_running) {
+		struct task_struct *curr = rq_of_rt_rq(rt_rq)->curr;
+
+		enqueue_rt_entity(rt_se);
+		if (rt_rq->highest_prio < curr->prio)
+			resched_task(curr);
+	}
+}
+
+static void sched_rt_ratio_dequeue(struct rt_rq *rt_rq)
+{
+	struct sched_rt_entity *rt_se = rt_rq->rt_se;
+
+	if (rt_se && on_rt_rq(rt_se))
+		dequeue_rt_entity(rt_se);
+}
+
+#else
+
+static inline unsigned int sched_rt_ratio(struct rt_rq *rt_rq)
+{
+	return sysctl_sched_rt_ratio;
+}
+
+#define for_each_leaf_rt_rq(rt_rq, rq) \
+	for (rt_rq = &rq->rt; rt_rq; rt_rq = NULL)
+
+static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq)
+{
+	return container_of(rt_rq, struct rq, rt);
+}
+
+static inline struct rt_rq *rt_rq_of_se(struct sched_rt_entity *rt_se)
+{
+	struct task_struct *p = rt_task_of(rt_se);
+	struct rq *rq = task_rq(p);
+
+	return &rq->rt;
+}
+
+#define for_each_sched_rt_entity(rt_se) \
+	for (; rt_se; rt_se = NULL)
+
+static inline struct rt_rq *group_rt_rq(struct sched_rt_entity *rt_se)
+{
+	return NULL;
+}
+
+static inline void sched_rt_ratio_enqueue(struct rt_rq *rt_rq)
+{
+}
+
+static inline void sched_rt_ratio_dequeue(struct rt_rq *rt_rq)
+{
+}
+
+#endif
+
+static inline int rt_se_prio(struct sched_rt_entity *rt_se)
+{
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	struct rt_rq *rt_rq = group_rt_rq(rt_se);
+
+	if (rt_rq)
+		return rt_rq->highest_prio;
+#endif
+
+	return rt_task_of(rt_se)->prio;
+}
+
+static int sched_rt_ratio_exceeded(struct rt_rq *rt_rq)
+{
+	unsigned int rt_ratio = sched_rt_ratio(rt_rq);
+	u64 period, ratio;
+
+	if (rt_ratio == SCHED_RT_FRAC)
+		return 0;
+
+	if (rt_rq->rt_throttled)
+		return 1;
+
+	period = (u64)sysctl_sched_rt_period * NSEC_PER_MSEC;
+	ratio = (period * rt_ratio) >> SCHED_RT_FRAC_SHIFT;
+
+	if (rt_rq->rt_time > ratio) {
+		struct rq *rq = rq_of_rt_rq(rt_rq);
+
+		rq->rt_throttled = 1;
+		rt_rq->rt_throttled = 1;
+
+		sched_rt_ratio_dequeue(rt_rq);
+		return 1;
+	}
+
+	return 0;
+}
+
+static void update_sched_rt_period(struct rq *rq)
+{
+	struct rt_rq *rt_rq;
+	u64 period;
+
+	while (rq->clock > rq->rt_period_expire) {
+		period = (u64)sysctl_sched_rt_period * NSEC_PER_MSEC;
+		rq->rt_period_expire += period;
+
+		for_each_leaf_rt_rq(rt_rq, rq) {
+			unsigned long rt_ratio = sched_rt_ratio(rt_rq);
+			u64 ratio = (period * rt_ratio) >> SCHED_RT_FRAC_SHIFT;
+
+			rt_rq->rt_time -= min(rt_rq->rt_time, ratio);
+			if (rt_rq->rt_throttled) {
+				rt_rq->rt_throttled = 0;
+				sched_rt_ratio_enqueue(rt_rq);
+			}
+		}
+
+		rq->rt_throttled = 0;
+	}
+}
+
 /*
  * Update the current task's runtime statistics. Skip current tasks that
  * are not in our scheduling class.
@@ -10,6 +221,8 @@
 static void update_curr_rt(struct rq *rq)
 {
 	struct task_struct *curr = rq->curr;
+	struct sched_rt_entity *rt_se = &curr->rt;
+	struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
 	u64 delta_exec;
 
 	if (!task_has_rt_policy(curr))
@@ -24,47 +237,228 @@ static void update_curr_rt(struct rq *rq)
 	curr->se.sum_exec_runtime += delta_exec;
 	curr->se.exec_start = rq->clock;
 	cpuacct_charge(curr, delta_exec);
+
+	rt_rq->rt_time += delta_exec;
+	/*
+	 * might make it a tad more accurate:
+	 *
+	 * update_sched_rt_period(rq);
+	 */
+	if (sched_rt_ratio_exceeded(rt_rq))
+		resched_task(curr);
 }
 
-static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup)
+static inline
+void inc_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
+{
+	WARN_ON(!rt_prio(rt_se_prio(rt_se)));
+	rt_rq->rt_nr_running++;
+#if defined CONFIG_SMP || defined CONFIG_FAIR_GROUP_SCHED
+	if (rt_se_prio(rt_se) < rt_rq->highest_prio)
+		rt_rq->highest_prio = rt_se_prio(rt_se);
+#endif
+#ifdef CONFIG_SMP
+	if (rt_se->nr_cpus_allowed > 1) {
+		struct rq *rq = rq_of_rt_rq(rt_rq);
+		rq->rt.rt_nr_migratory++;
+	}
+
+	update_rt_migration(rq_of_rt_rq(rt_rq));
+#endif
+}
+
+static inline
+void dec_rt_tasks(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq)
+{
+	WARN_ON(!rt_prio(rt_se_prio(rt_se)));
+	WARN_ON(!rt_rq->rt_nr_running);
+	rt_rq->rt_nr_running--;
+#if defined CONFIG_SMP || defined CONFIG_FAIR_GROUP_SCHED
+	if (rt_rq->rt_nr_running) {
+		struct rt_prio_array *array;
+
+		WARN_ON(rt_se_prio(rt_se) < rt_rq->highest_prio);
+		if (rt_se_prio(rt_se) == rt_rq->highest_prio) {
+			/* recalculate */
+			array = &rt_rq->active;
+			rt_rq->highest_prio =
+				sched_find_first_bit(array->bitmap);
+		} /* otherwise leave rq->highest prio alone */
+	} else
+		rt_rq->highest_prio = MAX_RT_PRIO;
+#endif
+#ifdef CONFIG_SMP
+	if (rt_se->nr_cpus_allowed > 1) {
+		struct rq *rq = rq_of_rt_rq(rt_rq);
+		rq->rt.rt_nr_migratory--;
+	}
+
+	update_rt_migration(rq_of_rt_rq(rt_rq));
+#endif /* CONFIG_SMP */
+}
+
+static void enqueue_rt_entity(struct sched_rt_entity *rt_se)
+{
+	struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
+	struct rt_prio_array *array = &rt_rq->active;
+	struct rt_rq *group_rq = group_rt_rq(rt_se);
+
+	if (group_rq && group_rq->rt_throttled)
+		return;
+
+	list_add_tail(&rt_se->run_list, array->queue + rt_se_prio(rt_se));
+	__set_bit(rt_se_prio(rt_se), array->bitmap);
+
+	inc_rt_tasks(rt_se, rt_rq);
+}
+
+static void dequeue_rt_entity(struct sched_rt_entity *rt_se)
 {
-	struct rt_prio_array *array = &rq->rt.active;
+	struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
+	struct rt_prio_array *array = &rt_rq->active;
+
+	list_del_init(&rt_se->run_list);
+	if (list_empty(array->queue + rt_se_prio(rt_se)))
+		__clear_bit(rt_se_prio(rt_se), array->bitmap);
 
-	list_add_tail(&p->run_list, array->queue + p->prio);
-	__set_bit(p->prio, array->bitmap);
+	dec_rt_tasks(rt_se, rt_rq);
+}
+
+/*
+ * Because the prio of an upper entry depends on the lower
+ * entries, we must remove entries top - down.
+ *
+ * XXX: O(1/2 h^2) because we can only walk up, not down the chain.
+ *      doesn't matter much for now, as h=2 for GROUP_SCHED.
+ */
+static void dequeue_rt_stack(struct task_struct *p)
+{
+	struct sched_rt_entity *rt_se, *top_se;
+
+	/*
+	 * dequeue all, top - down.
+	 */
+	do {
+		rt_se = &p->rt;
+		top_se = NULL;
+		for_each_sched_rt_entity(rt_se) {
+			if (on_rt_rq(rt_se))
+				top_se = rt_se;
+		}
+		if (top_se)
+			dequeue_rt_entity(top_se);
+	} while (top_se);
 }
 
 /*
  * Adding/removing a task to/from a priority array:
  */
+static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup)
+{
+	struct sched_rt_entity *rt_se = &p->rt;
+
+	if (wakeup)
+		rt_se->timeout = 0;
+
+	dequeue_rt_stack(p);
+
+	/*
+	 * enqueue everybody, bottom - up.
+	 */
+	for_each_sched_rt_entity(rt_se)
+		enqueue_rt_entity(rt_se);
+
+	inc_cpu_load(rq, p->se.load.weight);
+}
+
 static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep)
 {
-	struct rt_prio_array *array = &rq->rt.active;
+	struct sched_rt_entity *rt_se = &p->rt;
+	struct rt_rq *rt_rq;
 
 	update_curr_rt(rq);
 
-	list_del(&p->run_list);
-	if (list_empty(array->queue + p->prio))
-		__clear_bit(p->prio, array->bitmap);
+	dequeue_rt_stack(p);
+
+	/*
+	 * re-enqueue all non-empty rt_rq entities.
+	 */
+	for_each_sched_rt_entity(rt_se) {
+		rt_rq = group_rt_rq(rt_se);
+		if (rt_rq && rt_rq->rt_nr_running)
+			enqueue_rt_entity(rt_se);
+	}
+
+	dec_cpu_load(rq, p->se.load.weight);
 }
 
 /*
  * Put task to the end of the run list without the overhead of dequeue
  * followed by enqueue.
  */
+static
+void requeue_rt_entity(struct rt_rq *rt_rq, struct sched_rt_entity *rt_se)
+{
+	struct rt_prio_array *array = &rt_rq->active;
+
+	list_move_tail(&rt_se->run_list, array->queue + rt_se_prio(rt_se));
+}
+
 static void requeue_task_rt(struct rq *rq, struct task_struct *p)
 {
-	struct rt_prio_array *array = &rq->rt.active;
+	struct sched_rt_entity *rt_se = &p->rt;
+	struct rt_rq *rt_rq;
 
-	list_move_tail(&p->run_list, array->queue + p->prio);
+	for_each_sched_rt_entity(rt_se) {
+		rt_rq = rt_rq_of_se(rt_se);
+		requeue_rt_entity(rt_rq, rt_se);
+	}
 }
 
-static void
-yield_task_rt(struct rq *rq)
+static void yield_task_rt(struct rq *rq)
 {
 	requeue_task_rt(rq, rq->curr);
 }
 
+#ifdef CONFIG_SMP
+static int find_lowest_rq(struct task_struct *task);
+
+static int select_task_rq_rt(struct task_struct *p, int sync)
+{
+	struct rq *rq = task_rq(p);
+
+	/*
+	 * If the current task is an RT task, then
+	 * try to see if we can wake this RT task up on another
+	 * runqueue. Otherwise simply start this RT task
+	 * on its current runqueue.
+	 *
+	 * We want to avoid overloading runqueues. Even if
+	 * the RT task is of higher priority than the current RT task.
+	 * RT tasks behave differently than other tasks. If
+	 * one gets preempted, we try to push it off to another queue.
+	 * So trying to keep a preempting RT task on the same
+	 * cache hot CPU will force the running RT task to
+	 * a cold CPU. So we waste all the cache for the lower
+	 * RT task in hopes of saving some of a RT task
+	 * that is just being woken and probably will have
+	 * cold cache anyway.
+	 */
+	if (unlikely(rt_task(rq->curr)) &&
+	    (p->rt.nr_cpus_allowed > 1)) {
+		int cpu = find_lowest_rq(p);
+
+		return (cpu == -1) ? task_cpu(p) : cpu;
+	}
+
+	/*
+	 * Otherwise, just let it ride on the affined RQ and the
+	 * post-schedule router will push the preempted task away
+	 */
+	return task_cpu(p);
+}
+#endif /* CONFIG_SMP */
+
 /*
  * Preempt the current task with a newly woken task if needed:
  */
@@ -74,25 +468,48 @@ static void check_preempt_curr_rt(struct rq *rq, struct task_struct *p)
 		resched_task(rq->curr);
 }
 
-static struct task_struct *pick_next_task_rt(struct rq *rq)
+static struct sched_rt_entity *pick_next_rt_entity(struct rq *rq,
+						   struct rt_rq *rt_rq)
 {
-	struct rt_prio_array *array = &rq->rt.active;
-	struct task_struct *next;
+	struct rt_prio_array *array = &rt_rq->active;
+	struct sched_rt_entity *next = NULL;
 	struct list_head *queue;
 	int idx;
 
 	idx = sched_find_first_bit(array->bitmap);
-	if (idx >= MAX_RT_PRIO)
-		return NULL;
+	BUG_ON(idx >= MAX_RT_PRIO);
 
 	queue = array->queue + idx;
-	next = list_entry(queue->next, struct task_struct, run_list);
-
-	next->se.exec_start = rq->clock;
+	next = list_entry(queue->next, struct sched_rt_entity, run_list);
 
 	return next;
 }
 
+static struct task_struct *pick_next_task_rt(struct rq *rq)
+{
+	struct sched_rt_entity *rt_se;
+	struct task_struct *p;
+	struct rt_rq *rt_rq;
+
+	rt_rq = &rq->rt;
+
+	if (unlikely(!rt_rq->rt_nr_running))
+		return NULL;
+
+	if (sched_rt_ratio_exceeded(rt_rq))
+		return NULL;
+
+	do {
+		rt_se = pick_next_rt_entity(rq, rt_rq);
+		BUG_ON(!rt_se);
+		rt_rq = group_rt_rq(rt_se);
+	} while (rt_rq);
+
+	p = rt_task_of(rt_se);
+	p->se.exec_start = rq->clock;
+	return p;
+}
+
 static void put_prev_task_rt(struct rq *rq, struct task_struct *p)
 {
 	update_curr_rt(rq);
@@ -100,76 +517,448 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p)
 }
 
 #ifdef CONFIG_SMP
-/*
- * Load-balancing iterator. Note: while the runqueue stays locked
- * during the whole iteration, the current task might be
- * dequeued so the iterator has to be dequeue-safe. Here we
- * achieve that by always pre-iterating before returning
- * the current task:
- */
-static struct task_struct *load_balance_start_rt(void *arg)
+
+/* Only try algorithms three times */
+#define RT_MAX_TRIES 3
+
+static int double_lock_balance(struct rq *this_rq, struct rq *busiest);
+static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep);
+
+static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu)
 {
-	struct rq *rq = arg;
-	struct rt_prio_array *array = &rq->rt.active;
-	struct list_head *head, *curr;
-	struct task_struct *p;
+	if (!task_running(rq, p) &&
+	    (cpu < 0 || cpu_isset(cpu, p->cpus_allowed)) &&
+	    (p->rt.nr_cpus_allowed > 1))
+		return 1;
+	return 0;
+}
+
+/* Return the second highest RT task, NULL otherwise */
+static struct task_struct *pick_next_highest_task_rt(struct rq *rq, int cpu)
+{
+	struct task_struct *next = NULL;
+	struct sched_rt_entity *rt_se;
+	struct rt_prio_array *array;
+	struct rt_rq *rt_rq;
 	int idx;
 
-	idx = sched_find_first_bit(array->bitmap);
-	if (idx >= MAX_RT_PRIO)
-		return NULL;
+	for_each_leaf_rt_rq(rt_rq, rq) {
+		array = &rt_rq->active;
+		idx = sched_find_first_bit(array->bitmap);
+ next_idx:
+		if (idx >= MAX_RT_PRIO)
+			continue;
+		if (next && next->prio < idx)
+			continue;
+		list_for_each_entry(rt_se, array->queue + idx, run_list) {
+			struct task_struct *p = rt_task_of(rt_se);
+			if (pick_rt_task(rq, p, cpu)) {
+				next = p;
+				break;
+			}
+		}
+		if (!next) {
+			idx = find_next_bit(array->bitmap, MAX_RT_PRIO, idx+1);
+			goto next_idx;
+		}
+	}
 
-	head = array->queue + idx;
-	curr = head->prev;
+	return next;
+}
 
-	p = list_entry(curr, struct task_struct, run_list);
+static DEFINE_PER_CPU(cpumask_t, local_cpu_mask);
 
-	curr = curr->prev;
+static int find_lowest_cpus(struct task_struct *task, cpumask_t *lowest_mask)
+{
+	int       lowest_prio = -1;
+	int       lowest_cpu  = -1;
+	int       count       = 0;
+	int       cpu;
 
-	rq->rt.rt_load_balance_idx = idx;
-	rq->rt.rt_load_balance_head = head;
-	rq->rt.rt_load_balance_curr = curr;
+	cpus_and(*lowest_mask, task_rq(task)->rd->online, task->cpus_allowed);
 
-	return p;
+	/*
+	 * Scan each rq for the lowest prio.
+	 */
+	for_each_cpu_mask(cpu, *lowest_mask) {
+		struct rq *rq = cpu_rq(cpu);
+
+		/* We look for lowest RT prio or non-rt CPU */
+		if (rq->rt.highest_prio >= MAX_RT_PRIO) {
+			/*
+			 * if we already found a low RT queue
+			 * and now we found this non-rt queue
+			 * clear the mask and set our bit.
+			 * Otherwise just return the queue as is
+			 * and the count==1 will cause the algorithm
+			 * to use the first bit found.
+			 */
+			if (lowest_cpu != -1) {
+				cpus_clear(*lowest_mask);
+				cpu_set(rq->cpu, *lowest_mask);
+			}
+			return 1;
+		}
+
+		/* no locking for now */
+		if ((rq->rt.highest_prio > task->prio)
+		    && (rq->rt.highest_prio >= lowest_prio)) {
+			if (rq->rt.highest_prio > lowest_prio) {
+				/* new low - clear old data */
+				lowest_prio = rq->rt.highest_prio;
+				lowest_cpu = cpu;
+				count = 0;
+			}
+			count++;
+		} else
+			cpu_clear(cpu, *lowest_mask);
+	}
+
+	/*
+	 * Clear out all the set bits that represent
+	 * runqueues that were of higher prio than
+	 * the lowest_prio.
+	 */
+	if (lowest_cpu > 0) {
+		/*
+		 * Perhaps we could add another cpumask op to
+		 * zero out bits. Like cpu_zero_bits(cpumask, nrbits);
+		 * Then that could be optimized to use memset and such.
+		 */
+		for_each_cpu_mask(cpu, *lowest_mask) {
+			if (cpu >= lowest_cpu)
+				break;
+			cpu_clear(cpu, *lowest_mask);
+		}
+	}
+
+	return count;
 }
 
-static struct task_struct *load_balance_next_rt(void *arg)
+static inline int pick_optimal_cpu(int this_cpu, cpumask_t *mask)
 {
-	struct rq *rq = arg;
-	struct rt_prio_array *array = &rq->rt.active;
-	struct list_head *head, *curr;
-	struct task_struct *p;
-	int idx;
+	int first;
+
+	/* "this_cpu" is cheaper to preempt than a remote processor */
+	if ((this_cpu != -1) && cpu_isset(this_cpu, *mask))
+		return this_cpu;
+
+	first = first_cpu(*mask);
+	if (first != NR_CPUS)
+		return first;
+
+	return -1;
+}
+
+static int find_lowest_rq(struct task_struct *task)
+{
+	struct sched_domain *sd;
+	cpumask_t *lowest_mask = &__get_cpu_var(local_cpu_mask);
+	int this_cpu = smp_processor_id();
+	int cpu      = task_cpu(task);
+	int count    = find_lowest_cpus(task, lowest_mask);
 
-	idx = rq->rt.rt_load_balance_idx;
-	head = rq->rt.rt_load_balance_head;
-	curr = rq->rt.rt_load_balance_curr;
+	if (!count)
+		return -1; /* No targets found */
 
 	/*
-	 * If we arrived back to the head again then
-	 * iterate to the next queue (if any):
+	 * There is no sense in performing an optimal search if only one
+	 * target is found.
 	 */
-	if (unlikely(head == curr)) {
-		int next_idx = find_next_bit(array->bitmap, MAX_RT_PRIO, idx+1);
+	if (count == 1)
+		return first_cpu(*lowest_mask);
 
-		if (next_idx >= MAX_RT_PRIO)
-			return NULL;
+	/*
+	 * At this point we have built a mask of cpus representing the
+	 * lowest priority tasks in the system.  Now we want to elect
+	 * the best one based on our affinity and topology.
+	 *
+	 * We prioritize the last cpu that the task executed on since
+	 * it is most likely cache-hot in that location.
+	 */
+	if (cpu_isset(cpu, *lowest_mask))
+		return cpu;
+
+	/*
+	 * Otherwise, we consult the sched_domains span maps to figure
+	 * out which cpu is logically closest to our hot cache data.
+	 */
+	if (this_cpu == cpu)
+		this_cpu = -1; /* Skip this_cpu opt if the same */
+
+	for_each_domain(cpu, sd) {
+		if (sd->flags & SD_WAKE_AFFINE) {
+			cpumask_t domain_mask;
+			int       best_cpu;
 
-		idx = next_idx;
-		head = array->queue + idx;
-		curr = head->prev;
+			cpus_and(domain_mask, sd->span, *lowest_mask);
 
-		rq->rt.rt_load_balance_idx = idx;
-		rq->rt.rt_load_balance_head = head;
+			best_cpu = pick_optimal_cpu(this_cpu,
+						    &domain_mask);
+			if (best_cpu != -1)
+				return best_cpu;
+		}
 	}
 
-	p = list_entry(curr, struct task_struct, run_list);
+	/*
+	 * And finally, if there were no matches within the domains
+	 * just give the caller *something* to work with from the compatible
+	 * locations.
+	 */
+	return pick_optimal_cpu(this_cpu, lowest_mask);
+}
 
-	curr = curr->prev;
+/* Will lock the rq it finds */
+static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq)
+{
+	struct rq *lowest_rq = NULL;
+	int tries;
+	int cpu;
 
-	rq->rt.rt_load_balance_curr = curr;
+	for (tries = 0; tries < RT_MAX_TRIES; tries++) {
+		cpu = find_lowest_rq(task);
 
-	return p;
+		if ((cpu == -1) || (cpu == rq->cpu))
+			break;
+
+		lowest_rq = cpu_rq(cpu);
+
+		/* if the prio of this runqueue changed, try again */
+		if (double_lock_balance(rq, lowest_rq)) {
+			/*
+			 * We had to unlock the run queue. In
+			 * the mean time, task could have
+			 * migrated already or had its affinity changed.
+			 * Also make sure that it wasn't scheduled on its rq.
+			 */
+			if (unlikely(task_rq(task) != rq ||
+				     !cpu_isset(lowest_rq->cpu,
+						task->cpus_allowed) ||
+				     task_running(rq, task) ||
+				     !task->se.on_rq)) {
+
+				spin_unlock(&lowest_rq->lock);
+				lowest_rq = NULL;
+				break;
+			}
+		}
+
+		/* If this rq is still suitable use it. */
+		if (lowest_rq->rt.highest_prio > task->prio)
+			break;
+
+		/* try again */
+		spin_unlock(&lowest_rq->lock);
+		lowest_rq = NULL;
+	}
+
+	return lowest_rq;
+}
+
+/*
+ * If the current CPU has more than one RT task, see if the non
+ * running task can migrate over to a CPU that is running a task
+ * of lesser priority.
+ */
+static int push_rt_task(struct rq *rq)
+{
+	struct task_struct *next_task;
+	struct rq *lowest_rq;
+	int ret = 0;
+	int paranoid = RT_MAX_TRIES;
+
+	if (!rq->rt.overloaded)
+		return 0;
+
+	next_task = pick_next_highest_task_rt(rq, -1);
+	if (!next_task)
+		return 0;
+
+ retry:
+	if (unlikely(next_task == rq->curr)) {
+		WARN_ON(1);
+		return 0;
+	}
+
+	/*
+	 * It's possible that the next_task slipped in of
+	 * higher priority than current. If that's the case
+	 * just reschedule current.
+	 */
+	if (unlikely(next_task->prio < rq->curr->prio)) {
+		resched_task(rq->curr);
+		return 0;
+	}
+
+	/* We might release rq lock */
+	get_task_struct(next_task);
+
+	/* find_lock_lowest_rq locks the rq if found */
+	lowest_rq = find_lock_lowest_rq(next_task, rq);
+	if (!lowest_rq) {
+		struct task_struct *task;
+		/*
+		 * find lock_lowest_rq releases rq->lock
+		 * so it is possible that next_task has changed.
+		 * If it has, then try again.
+		 */
+		task = pick_next_highest_task_rt(rq, -1);
+		if (unlikely(task != next_task) && task && paranoid--) {
+			put_task_struct(next_task);
+			next_task = task;
+			goto retry;
+		}
+		goto out;
+	}
+
+	deactivate_task(rq, next_task, 0);
+	set_task_cpu(next_task, lowest_rq->cpu);
+	activate_task(lowest_rq, next_task, 0);
+
+	resched_task(lowest_rq->curr);
+
+	spin_unlock(&lowest_rq->lock);
+
+	ret = 1;
+out:
+	put_task_struct(next_task);
+
+	return ret;
+}
+
+/*
+ * TODO: Currently we just use the second highest prio task on
+ *       the queue, and stop when it can't migrate (or there's
+ *       no more RT tasks).  There may be a case where a lower
+ *       priority RT task has a different affinity than the
+ *       higher RT task. In this case the lower RT task could
+ *       possibly be able to migrate where as the higher priority
+ *       RT task could not.  We currently ignore this issue.
+ *       Enhancements are welcome!
+ */
+static void push_rt_tasks(struct rq *rq)
+{
+	/* push_rt_task will return true if it moved an RT */
+	while (push_rt_task(rq))
+		;
+}
+
+static int pull_rt_task(struct rq *this_rq)
+{
+	int this_cpu = this_rq->cpu, ret = 0, cpu;
+	struct task_struct *p, *next;
+	struct rq *src_rq;
+
+	if (likely(!rt_overloaded(this_rq)))
+		return 0;
+
+	next = pick_next_task_rt(this_rq);
+
+	for_each_cpu_mask(cpu, this_rq->rd->rto_mask) {
+		if (this_cpu == cpu)
+			continue;
+
+		src_rq = cpu_rq(cpu);
+		/*
+		 * We can potentially drop this_rq's lock in
+		 * double_lock_balance, and another CPU could
+		 * steal our next task - hence we must cause
+		 * the caller to recalculate the next task
+		 * in that case:
+		 */
+		if (double_lock_balance(this_rq, src_rq)) {
+			struct task_struct *old_next = next;
+
+			next = pick_next_task_rt(this_rq);
+			if (next != old_next)
+				ret = 1;
+		}
+
+		/*
+		 * Are there still pullable RT tasks?
+		 */
+		if (src_rq->rt.rt_nr_running <= 1)
+			goto skip;
+
+		p = pick_next_highest_task_rt(src_rq, this_cpu);
+
+		/*
+		 * Do we have an RT task that preempts
+		 * the to-be-scheduled task?
+		 */
+		if (p && (!next || (p->prio < next->prio))) {
+			WARN_ON(p == src_rq->curr);
+			WARN_ON(!p->se.on_rq);
+
+			/*
+			 * There's a chance that p is higher in priority
+			 * than what's currently running on its cpu.
+			 * This is just that p is wakeing up and hasn't
+			 * had a chance to schedule. We only pull
+			 * p if it is lower in priority than the
+			 * current task on the run queue or
+			 * this_rq next task is lower in prio than
+			 * the current task on that rq.
+			 */
+			if (p->prio < src_rq->curr->prio ||
+			    (next && next->prio < src_rq->curr->prio))
+				goto skip;
+
+			ret = 1;
+
+			deactivate_task(src_rq, p, 0);
+			set_task_cpu(p, this_cpu);
+			activate_task(this_rq, p, 0);
+			/*
+			 * We continue with the search, just in
+			 * case there's an even higher prio task
+			 * in another runqueue. (low likelyhood
+			 * but possible)
+			 *
+			 * Update next so that we won't pick a task
+			 * on another cpu with a priority lower (or equal)
+			 * than the one we just picked.
+			 */
+			next = p;
+
+		}
+ skip:
+		spin_unlock(&src_rq->lock);
+	}
+
+	return ret;
+}
+
+static void pre_schedule_rt(struct rq *rq, struct task_struct *prev)
+{
+	/* Try to pull RT tasks here if we lower this rq's prio */
+	if (unlikely(rt_task(prev)) && rq->rt.highest_prio > prev->prio)
+		pull_rt_task(rq);
+}
+
+static void post_schedule_rt(struct rq *rq)
+{
+	/*
+	 * If we have more than one rt_task queued, then
+	 * see if we can push the other rt_tasks off to other CPUS.
+	 * Note we may release the rq lock, and since
+	 * the lock was owned by prev, we need to release it
+	 * first via finish_lock_switch and then reaquire it here.
+	 */
+	if (unlikely(rq->rt.overloaded)) {
+		spin_lock_irq(&rq->lock);
+		push_rt_tasks(rq);
+		spin_unlock_irq(&rq->lock);
+	}
+}
+
+
+static void task_wake_up_rt(struct rq *rq, struct task_struct *p)
+{
+	if (!task_running(rq, p) &&
+	    (p->prio >= rq->rt.highest_prio) &&
+	    rq->rt.overloaded)
+		push_rt_tasks(rq);
 }
 
 static unsigned long
@@ -178,38 +967,170 @@ load_balance_rt(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		struct sched_domain *sd, enum cpu_idle_type idle,
 		int *all_pinned, int *this_best_prio)
 {
-	struct rq_iterator rt_rq_iterator;
-
-	rt_rq_iterator.start = load_balance_start_rt;
-	rt_rq_iterator.next = load_balance_next_rt;
-	/* pass 'busiest' rq argument into
-	 * load_balance_[start|next]_rt iterators
-	 */
-	rt_rq_iterator.arg = busiest;
-
-	return balance_tasks(this_rq, this_cpu, busiest, max_load_move, sd,
-			     idle, all_pinned, this_best_prio, &rt_rq_iterator);
+	/* don't touch RT tasks */
+	return 0;
 }
 
 static int
 move_one_task_rt(struct rq *this_rq, int this_cpu, struct rq *busiest,
 		 struct sched_domain *sd, enum cpu_idle_type idle)
 {
-	struct rq_iterator rt_rq_iterator;
+	/* don't touch RT tasks */
+	return 0;
+}
+
+static void set_cpus_allowed_rt(struct task_struct *p, cpumask_t *new_mask)
+{
+	int weight = cpus_weight(*new_mask);
+
+	BUG_ON(!rt_task(p));
 
-	rt_rq_iterator.start = load_balance_start_rt;
-	rt_rq_iterator.next = load_balance_next_rt;
-	rt_rq_iterator.arg = busiest;
+	/*
+	 * Update the migration status of the RQ if we have an RT task
+	 * which is running AND changing its weight value.
+	 */
+	if (p->se.on_rq && (weight != p->rt.nr_cpus_allowed)) {
+		struct rq *rq = task_rq(p);
+
+		if ((p->rt.nr_cpus_allowed <= 1) && (weight > 1)) {
+			rq->rt.rt_nr_migratory++;
+		} else if ((p->rt.nr_cpus_allowed > 1) && (weight <= 1)) {
+			BUG_ON(!rq->rt.rt_nr_migratory);
+			rq->rt.rt_nr_migratory--;
+		}
+
+		update_rt_migration(rq);
+	}
 
-	return iter_move_one_task(this_rq, this_cpu, busiest, sd, idle,
-				  &rt_rq_iterator);
+	p->cpus_allowed    = *new_mask;
+	p->rt.nr_cpus_allowed = weight;
 }
-#endif
 
-static void task_tick_rt(struct rq *rq, struct task_struct *p)
+/* Assumes rq->lock is held */
+static void join_domain_rt(struct rq *rq)
+{
+	if (rq->rt.overloaded)
+		rt_set_overload(rq);
+}
+
+/* Assumes rq->lock is held */
+static void leave_domain_rt(struct rq *rq)
+{
+	if (rq->rt.overloaded)
+		rt_clear_overload(rq);
+}
+
+/*
+ * When switch from the rt queue, we bring ourselves to a position
+ * that we might want to pull RT tasks from other runqueues.
+ */
+static void switched_from_rt(struct rq *rq, struct task_struct *p,
+			   int running)
+{
+	/*
+	 * If there are other RT tasks then we will reschedule
+	 * and the scheduling of the other RT tasks will handle
+	 * the balancing. But if we are the last RT task
+	 * we may need to handle the pulling of RT tasks
+	 * now.
+	 */
+	if (!rq->rt.rt_nr_running)
+		pull_rt_task(rq);
+}
+#endif /* CONFIG_SMP */
+
+/*
+ * When switching a task to RT, we may overload the runqueue
+ * with RT tasks. In this case we try to push them off to
+ * other runqueues.
+ */
+static void switched_to_rt(struct rq *rq, struct task_struct *p,
+			   int running)
+{
+	int check_resched = 1;
+
+	/*
+	 * If we are already running, then there's nothing
+	 * that needs to be done. But if we are not running
+	 * we may need to preempt the current running task.
+	 * If that current running task is also an RT task
+	 * then see if we can move to another run queue.
+	 */
+	if (!running) {
+#ifdef CONFIG_SMP
+		if (rq->rt.overloaded && push_rt_task(rq) &&
+		    /* Don't resched if we changed runqueues */
+		    rq != task_rq(p))
+			check_resched = 0;
+#endif /* CONFIG_SMP */
+		if (check_resched && p->prio < rq->curr->prio)
+			resched_task(rq->curr);
+	}
+}
+
+/*
+ * Priority of the task has changed. This may cause
+ * us to initiate a push or pull.
+ */
+static void prio_changed_rt(struct rq *rq, struct task_struct *p,
+			    int oldprio, int running)
+{
+	if (running) {
+#ifdef CONFIG_SMP
+		/*
+		 * If our priority decreases while running, we
+		 * may need to pull tasks to this runqueue.
+		 */
+		if (oldprio < p->prio)
+			pull_rt_task(rq);
+		/*
+		 * If there's a higher priority task waiting to run
+		 * then reschedule.
+		 */
+		if (p->prio > rq->rt.highest_prio)
+			resched_task(p);
+#else
+		/* For UP simply resched on drop of prio */
+		if (oldprio < p->prio)
+			resched_task(p);
+#endif /* CONFIG_SMP */
+	} else {
+		/*
+		 * This task is not running, but if it is
+		 * greater than the current running task
+		 * then reschedule.
+		 */
+		if (p->prio < rq->curr->prio)
+			resched_task(rq->curr);
+	}
+}
+
+static void watchdog(struct rq *rq, struct task_struct *p)
+{
+	unsigned long soft, hard;
+
+	if (!p->signal)
+		return;
+
+	soft = p->signal->rlim[RLIMIT_RTTIME].rlim_cur;
+	hard = p->signal->rlim[RLIMIT_RTTIME].rlim_max;
+
+	if (soft != RLIM_INFINITY) {
+		unsigned long next;
+
+		p->rt.timeout++;
+		next = DIV_ROUND_UP(min(soft, hard), USEC_PER_SEC/HZ);
+		if (p->rt.timeout > next)
+			p->it_sched_expires = p->se.sum_exec_runtime;
+	}
+}
+
+static void task_tick_rt(struct rq *rq, struct task_struct *p, int queued)
 {
 	update_curr_rt(rq);
 
+	watchdog(rq, p);
+
 	/*
 	 * RR tasks need a special form of timeslice management.
 	 * FIFO tasks have no timeslices.
@@ -217,16 +1138,16 @@ static void task_tick_rt(struct rq *rq, struct task_struct *p)
 	if (p->policy != SCHED_RR)
 		return;
 
-	if (--p->time_slice)
+	if (--p->rt.time_slice)
 		return;
 
-	p->time_slice = DEF_TIMESLICE;
+	p->rt.time_slice = DEF_TIMESLICE;
 
 	/*
 	 * Requeue to the end of queue if we are not the only element
 	 * on the queue:
 	 */
-	if (p->run_list.prev != p->run_list.next) {
+	if (p->rt.run_list.prev != p->rt.run_list.next) {
 		requeue_task_rt(rq, p);
 		set_tsk_need_resched(p);
 	}
@@ -244,6 +1165,9 @@ const struct sched_class rt_sched_class = {
 	.enqueue_task		= enqueue_task_rt,
 	.dequeue_task		= dequeue_task_rt,
 	.yield_task		= yield_task_rt,
+#ifdef CONFIG_SMP
+	.select_task_rq		= select_task_rq_rt,
+#endif /* CONFIG_SMP */
 
 	.check_preempt_curr	= check_preempt_curr_rt,
 
@@ -253,8 +1177,18 @@ const struct sched_class rt_sched_class = {
 #ifdef CONFIG_SMP
 	.load_balance		= load_balance_rt,
 	.move_one_task		= move_one_task_rt,
+	.set_cpus_allowed       = set_cpus_allowed_rt,
+	.join_domain            = join_domain_rt,
+	.leave_domain           = leave_domain_rt,
+	.pre_schedule		= pre_schedule_rt,
+	.post_schedule		= post_schedule_rt,
+	.task_wake_up		= task_wake_up_rt,
+	.switched_from		= switched_from_rt,
 #endif
 
 	.set_curr_task          = set_curr_task_rt,
 	.task_tick		= task_tick_rt,
+
+	.prio_changed		= prio_changed_rt,
+	.switched_to		= switched_to_rt,
 };
diff --git a/kernel/signal.c b/kernel/signal.c
index afa4f781f92..bf49ce6f016 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -733,13 +733,13 @@ static void print_fatal_signal(struct pt_regs *regs, int signr)
 		current->comm, task_pid_nr(current), signr);
 
 #if defined(__i386__) && !defined(__arch_um__)
-	printk("code at %08lx: ", regs->eip);
+	printk("code at %08lx: ", regs->ip);
 	{
 		int i;
 		for (i = 0; i < 16; i++) {
 			unsigned char insn;
 
-			__get_user(insn, (unsigned char *)(regs->eip + i));
+			__get_user(insn, (unsigned char *)(regs->ip + i));
 			printk("%02x ", insn);
 		}
 	}
diff --git a/kernel/softirq.c b/kernel/softirq.c
index bd89bc4eb0b..d7837d45419 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -3,7 +3,9 @@
  *
  *	Copyright (C) 1992 Linus Torvalds
  *
- * Rewritten. Old one was good in 2.2, but in 2.3 it was immoral. --ANK (990903)
+ *	Distribute under GPLv2.
+ *
+ *	Rewritten. Old one was good in 2.2, but in 2.3 it was immoral. --ANK (990903)
  */
 
 #include <linux/module.h>
@@ -278,9 +280,14 @@ asmlinkage void do_softirq(void)
  */
 void irq_enter(void)
 {
+#ifdef CONFIG_NO_HZ
+	int cpu = smp_processor_id();
+	if (idle_cpu(cpu) && !in_interrupt())
+		tick_nohz_stop_idle(cpu);
+#endif
 	__irq_enter();
 #ifdef CONFIG_NO_HZ
-	if (idle_cpu(smp_processor_id()))
+	if (idle_cpu(cpu))
 		tick_nohz_update_jiffies();
 #endif
 }
diff --git a/kernel/softlockup.c b/kernel/softlockup.c
index 11df812263c..c1d76552446 100644
--- a/kernel/softlockup.c
+++ b/kernel/softlockup.c
@@ -8,6 +8,7 @@
  */
 #include <linux/mm.h>
 #include <linux/cpu.h>
+#include <linux/nmi.h>
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <linux/freezer.h>
@@ -23,8 +24,8 @@ static DEFINE_PER_CPU(unsigned long, touch_timestamp);
 static DEFINE_PER_CPU(unsigned long, print_timestamp);
 static DEFINE_PER_CPU(struct task_struct *, watchdog_task);
 
-static int did_panic;
-int softlockup_thresh = 10;
+static int __read_mostly did_panic;
+unsigned long __read_mostly softlockup_thresh = 60;
 
 static int
 softlock_panic(struct notifier_block *this, unsigned long event, void *ptr)
@@ -45,7 +46,7 @@ static struct notifier_block panic_block = {
  */
 static unsigned long get_timestamp(int this_cpu)
 {
-	return cpu_clock(this_cpu) >> 30;  /* 2^30 ~= 10^9 */
+	return cpu_clock(this_cpu) >> 30LL;  /* 2^30 ~= 10^9 */
 }
 
 void touch_softlockup_watchdog(void)
@@ -100,11 +101,7 @@ void softlockup_tick(void)
 
 	now = get_timestamp(this_cpu);
 
-	/* Wake up the high-prio watchdog task every second: */
-	if (now > (touch_timestamp + 1))
-		wake_up_process(per_cpu(watchdog_task, this_cpu));
-
-	/* Warn about unreasonable 10+ seconds delays: */
+	/* Warn about unreasonable delays: */
 	if (now <= (touch_timestamp + softlockup_thresh))
 		return;
 
@@ -122,11 +119,93 @@ void softlockup_tick(void)
 }
 
 /*
+ * Have a reasonable limit on the number of tasks checked:
+ */
+unsigned long __read_mostly sysctl_hung_task_check_count = 1024;
+
+/*
+ * Zero means infinite timeout - no checking done:
+ */
+unsigned long __read_mostly sysctl_hung_task_timeout_secs = 120;
+
+unsigned long __read_mostly sysctl_hung_task_warnings = 10;
+
+/*
+ * Only do the hung-tasks check on one CPU:
+ */
+static int check_cpu __read_mostly = -1;
+
+static void check_hung_task(struct task_struct *t, unsigned long now)
+{
+	unsigned long switch_count = t->nvcsw + t->nivcsw;
+
+	if (t->flags & PF_FROZEN)
+		return;
+
+	if (switch_count != t->last_switch_count || !t->last_switch_timestamp) {
+		t->last_switch_count = switch_count;
+		t->last_switch_timestamp = now;
+		return;
+	}
+	if ((long)(now - t->last_switch_timestamp) <
+					sysctl_hung_task_timeout_secs)
+		return;
+	if (sysctl_hung_task_warnings < 0)
+		return;
+	sysctl_hung_task_warnings--;
+
+	/*
+	 * Ok, the task did not get scheduled for more than 2 minutes,
+	 * complain:
+	 */
+	printk(KERN_ERR "INFO: task %s:%d blocked for more than "
+			"%ld seconds.\n", t->comm, t->pid,
+			sysctl_hung_task_timeout_secs);
+	printk(KERN_ERR "\"echo 0 > /proc/sys/kernel/hung_task_timeout_secs\""
+			" disables this message.\n");
+	sched_show_task(t);
+	__debug_show_held_locks(t);
+
+	t->last_switch_timestamp = now;
+	touch_nmi_watchdog();
+}
+
+/*
+ * Check whether a TASK_UNINTERRUPTIBLE does not get woken up for
+ * a really long time (120 seconds). If that happens, print out
+ * a warning.
+ */
+static void check_hung_uninterruptible_tasks(int this_cpu)
+{
+	int max_count = sysctl_hung_task_check_count;
+	unsigned long now = get_timestamp(this_cpu);
+	struct task_struct *g, *t;
+
+	/*
+	 * If the system crashed already then all bets are off,
+	 * do not report extra hung tasks:
+	 */
+	if ((tainted & TAINT_DIE) || did_panic)
+		return;
+
+	read_lock(&tasklist_lock);
+	do_each_thread(g, t) {
+		if (!--max_count)
+			break;
+		if (t->state & TASK_UNINTERRUPTIBLE)
+			check_hung_task(t, now);
+	} while_each_thread(g, t);
+
+	read_unlock(&tasklist_lock);
+}
+
+/*
  * The watchdog thread - runs every second and touches the timestamp.
  */
 static int watchdog(void *__bind_cpu)
 {
 	struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 };
+	int this_cpu = (long)__bind_cpu;
 
 	sched_setscheduler(current, SCHED_FIFO, &param);
 
@@ -135,13 +214,18 @@ static int watchdog(void *__bind_cpu)
 
 	/*
 	 * Run briefly once per second to reset the softlockup timestamp.
-	 * If this gets delayed for more than 10 seconds then the
+	 * If this gets delayed for more than 60 seconds then the
 	 * debug-printout triggers in softlockup_tick().
 	 */
 	while (!kthread_should_stop()) {
-		set_current_state(TASK_INTERRUPTIBLE);
 		touch_softlockup_watchdog();
-		schedule();
+		msleep_interruptible(10000);
+
+		if (this_cpu != check_cpu)
+			continue;
+
+		if (sysctl_hung_task_timeout_secs)
+			check_hung_uninterruptible_tasks(this_cpu);
 	}
 
 	return 0;
@@ -171,6 +255,7 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
 		break;
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
+		check_cpu = any_online_cpu(cpu_online_map);
 		wake_up_process(per_cpu(watchdog_task, hotcpu));
 		break;
 #ifdef CONFIG_HOTPLUG_CPU
@@ -181,6 +266,15 @@ cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
 		/* Unbind so it can run.  Fall thru. */
 		kthread_bind(per_cpu(watchdog_task, hotcpu),
 			     any_online_cpu(cpu_online_map));
+	case CPU_DOWN_PREPARE:
+	case CPU_DOWN_PREPARE_FROZEN:
+		if (hotcpu == check_cpu) {
+			cpumask_t temp_cpu_online_map = cpu_online_map;
+
+			cpu_clear(hotcpu, temp_cpu_online_map);
+			check_cpu = any_online_cpu(temp_cpu_online_map);
+		}
+		break;
 	case CPU_DEAD:
 	case CPU_DEAD_FROZEN:
 		p = per_cpu(watchdog_task, hotcpu);
diff --git a/kernel/spinlock.c b/kernel/spinlock.c
index cd72424c266..ae28c824512 100644
--- a/kernel/spinlock.c
+++ b/kernel/spinlock.c
@@ -65,8 +65,7 @@ EXPORT_SYMBOL(_write_trylock);
  * even on CONFIG_PREEMPT, because lockdep assumes that interrupts are
  * not re-enabled during lock-acquire (which the preempt-spin-ops do):
  */
-#if !defined(CONFIG_PREEMPT) || !defined(CONFIG_SMP) || \
-	defined(CONFIG_DEBUG_LOCK_ALLOC)
+#if !defined(CONFIG_GENERIC_LOCKBREAK) || defined(CONFIG_DEBUG_LOCK_ALLOC)
 
 void __lockfunc _read_lock(rwlock_t *lock)
 {
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index 319821ef78a..51b5ee53571 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -203,13 +203,13 @@ int stop_machine_run(int (*fn)(void *), void *data, unsigned int cpu)
 	int ret;
 
 	/* No CPUs can come up or down during this. */
-	lock_cpu_hotplug();
+	get_online_cpus();
 	p = __stop_machine_run(fn, data, cpu);
 	if (!IS_ERR(p))
 		ret = kthread_stop(p);
 	else
 		ret = PTR_ERR(p);
-	unlock_cpu_hotplug();
+	put_online_cpus();
 
 	return ret;
 }
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index c68f68dcc60..357b68ba23e 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -53,6 +53,7 @@
 #ifdef CONFIG_X86
 #include <asm/nmi.h>
 #include <asm/stacktrace.h>
+#include <asm/io.h>
 #endif
 
 static int deprecated_sysctl_warning(struct __sysctl_args *args);
@@ -81,6 +82,7 @@ extern int compat_log;
 extern int maps_protect;
 extern int sysctl_stat_interval;
 extern int audit_argv_kb;
+extern int latencytop_enabled;
 
 /* Constants used for minimum and  maximum */
 #ifdef CONFIG_DETECT_SOFTLOCKUP
@@ -156,8 +158,16 @@ static int proc_dointvec_taint(struct ctl_table *table, int write, struct file *
 #endif
 
 static struct ctl_table root_table[];
-static struct ctl_table_header root_table_header =
-	{ root_table, LIST_HEAD_INIT(root_table_header.ctl_entry) };
+static struct ctl_table_root sysctl_table_root;
+static struct ctl_table_header root_table_header = {
+	.ctl_table = root_table,
+	.ctl_entry = LIST_HEAD_INIT(sysctl_table_root.header_list),
+	.root = &sysctl_table_root,
+};
+static struct ctl_table_root sysctl_table_root = {
+	.root_list = LIST_HEAD_INIT(sysctl_table_root.root_list),
+	.header_list = LIST_HEAD_INIT(root_table_header.ctl_entry),
+};
 
 static struct ctl_table kern_table[];
 static struct ctl_table vm_table[];
@@ -191,14 +201,6 @@ static struct ctl_table root_table[] = {
 		.mode		= 0555,
 		.child		= vm_table,
 	},
-#ifdef CONFIG_NET
-	{
-		.ctl_name	= CTL_NET,
-		.procname	= "net",
-		.mode		= 0555,
-		.child		= net_table,
-	},
-#endif
 	{
 		.ctl_name	= CTL_FS,
 		.procname	= "fs",
@@ -306,9 +308,43 @@ static struct ctl_table kern_table[] = {
 		.procname	= "sched_nr_migrate",
 		.data		= &sysctl_sched_nr_migrate,
 		.maxlen		= sizeof(unsigned int),
-		.mode		= 644,
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "sched_rt_period_ms",
+		.data		= &sysctl_sched_rt_period,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "sched_rt_ratio",
+		.data		= &sysctl_sched_rt_ratio,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
+#if defined(CONFIG_FAIR_GROUP_SCHED) && defined(CONFIG_SMP)
+	{
+		.ctl_name       = CTL_UNNUMBERED,
+		.procname       = "sched_min_bal_int_shares",
+		.data           = &sysctl_sched_min_bal_int_shares,
+		.maxlen         = sizeof(unsigned int),
+		.mode           = 0644,
+		.proc_handler   = &proc_dointvec,
+	},
+	{
+		.ctl_name       = CTL_UNNUMBERED,
+		.procname       = "sched_max_bal_int_shares",
+		.data           = &sysctl_sched_max_bal_int_shares,
+		.maxlen         = sizeof(unsigned int),
+		.mode           = 0644,
+		.proc_handler   = &proc_dointvec,
+	},
+#endif
 #endif
 	{
 		.ctl_name	= CTL_UNNUMBERED,
@@ -382,6 +418,15 @@ static struct ctl_table kern_table[] = {
 		.proc_handler	= &proc_dointvec_taint,
 	},
 #endif
+#ifdef CONFIG_LATENCYTOP
+	{
+		.procname	= "latencytop",
+		.data		= &latencytop_enabled,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
+#endif
 #ifdef CONFIG_SECURITY_CAPABILITIES
 	{
 		.procname	= "cap-bound",
@@ -683,6 +728,14 @@ static struct ctl_table kern_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "io_delay_type",
+		.data		= &io_delay_type,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
 #endif
 #if defined(CONFIG_MMU)
 	{
@@ -728,13 +781,40 @@ static struct ctl_table kern_table[] = {
 		.ctl_name	= CTL_UNNUMBERED,
 		.procname	= "softlockup_thresh",
 		.data		= &softlockup_thresh,
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0644,
-		.proc_handler	= &proc_dointvec_minmax,
+		.proc_handler	= &proc_doulongvec_minmax,
 		.strategy	= &sysctl_intvec,
 		.extra1		= &one,
 		.extra2		= &sixty,
 	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "hung_task_check_count",
+		.data		= &sysctl_hung_task_check_count,
+		.maxlen		= sizeof(unsigned long),
+		.mode		= 0644,
+		.proc_handler	= &proc_doulongvec_minmax,
+		.strategy	= &sysctl_intvec,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "hung_task_timeout_secs",
+		.data		= &sysctl_hung_task_timeout_secs,
+		.maxlen		= sizeof(unsigned long),
+		.mode		= 0644,
+		.proc_handler	= &proc_doulongvec_minmax,
+		.strategy	= &sysctl_intvec,
+	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "hung_task_warnings",
+		.data		= &sysctl_hung_task_warnings,
+		.maxlen		= sizeof(unsigned long),
+		.mode		= 0644,
+		.proc_handler	= &proc_doulongvec_minmax,
+		.strategy	= &sysctl_intvec,
+	},
 #endif
 #ifdef CONFIG_COMPAT
 	{
@@ -1300,12 +1380,27 @@ void sysctl_head_finish(struct ctl_table_header *head)
 	spin_unlock(&sysctl_lock);
 }
 
-struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev)
+static struct list_head *
+lookup_header_list(struct ctl_table_root *root, struct nsproxy *namespaces)
 {
+	struct list_head *header_list;
+	header_list = &root->header_list;
+	if (root->lookup)
+		header_list = root->lookup(root, namespaces);
+	return header_list;
+}
+
+struct ctl_table_header *__sysctl_head_next(struct nsproxy *namespaces,
+					    struct ctl_table_header *prev)
+{
+	struct ctl_table_root *root;
+	struct list_head *header_list;
 	struct ctl_table_header *head;
 	struct list_head *tmp;
+
 	spin_lock(&sysctl_lock);
 	if (prev) {
+		head = prev;
 		tmp = &prev->ctl_entry;
 		unuse_table(prev);
 		goto next;
@@ -1319,14 +1414,38 @@ struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev)
 		spin_unlock(&sysctl_lock);
 		return head;
 	next:
+		root = head->root;
 		tmp = tmp->next;
-		if (tmp == &root_table_header.ctl_entry)
-			break;
+		header_list = lookup_header_list(root, namespaces);
+		if (tmp != header_list)
+			continue;
+
+		do {
+			root = list_entry(root->root_list.next,
+					struct ctl_table_root, root_list);
+			if (root == &sysctl_table_root)
+				goto out;
+			header_list = lookup_header_list(root, namespaces);
+		} while (list_empty(header_list));
+		tmp = header_list->next;
 	}
+out:
 	spin_unlock(&sysctl_lock);
 	return NULL;
 }
 
+struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev)
+{
+	return __sysctl_head_next(current->nsproxy, prev);
+}
+
+void register_sysctl_root(struct ctl_table_root *root)
+{
+	spin_lock(&sysctl_lock);
+	list_add_tail(&root->root_list, &sysctl_table_root.root_list);
+	spin_unlock(&sysctl_lock);
+}
+
 #ifdef CONFIG_SYSCTL_SYSCALL
 int do_sysctl(int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp,
 	       void __user *newval, size_t newlen)
@@ -1483,18 +1602,21 @@ static __init int sysctl_init(void)
 {
 	int err;
 	sysctl_set_parent(NULL, root_table);
-	err = sysctl_check_table(root_table);
+	err = sysctl_check_table(current->nsproxy, root_table);
 	return 0;
 }
 
 core_initcall(sysctl_init);
 
 /**
- * register_sysctl_table - register a sysctl hierarchy
+ * __register_sysctl_paths - register a sysctl hierarchy
+ * @root: List of sysctl headers to register on
+ * @namespaces: Data to compute which lists of sysctl entries are visible
+ * @path: The path to the directory the sysctl table is in.
  * @table: the top-level table structure
  *
  * Register a sysctl table hierarchy. @table should be a filled in ctl_table
- * array. An entry with a ctl_name of 0 terminates the table. 
+ * array. A completely 0 filled entry terminates the table.
  *
  * The members of the &struct ctl_table structure are used as follows:
  *
@@ -1557,25 +1679,99 @@ core_initcall(sysctl_init);
  * This routine returns %NULL on a failure to register, and a pointer
  * to the table header on success.
  */
-struct ctl_table_header *register_sysctl_table(struct ctl_table * table)
+struct ctl_table_header *__register_sysctl_paths(
+	struct ctl_table_root *root,
+	struct nsproxy *namespaces,
+	const struct ctl_path *path, struct ctl_table *table)
 {
-	struct ctl_table_header *tmp;
-	tmp = kmalloc(sizeof(struct ctl_table_header), GFP_KERNEL);
-	if (!tmp)
+	struct list_head *header_list;
+	struct ctl_table_header *header;
+	struct ctl_table *new, **prevp;
+	unsigned int n, npath;
+
+	/* Count the path components */
+	for (npath = 0; path[npath].ctl_name || path[npath].procname; ++npath)
+		;
+
+	/*
+	 * For each path component, allocate a 2-element ctl_table array.
+	 * The first array element will be filled with the sysctl entry
+	 * for this, the second will be the sentinel (ctl_name == 0).
+	 *
+	 * We allocate everything in one go so that we don't have to
+	 * worry about freeing additional memory in unregister_sysctl_table.
+	 */
+	header = kzalloc(sizeof(struct ctl_table_header) +
+			 (2 * npath * sizeof(struct ctl_table)), GFP_KERNEL);
+	if (!header)
 		return NULL;
-	tmp->ctl_table = table;
-	INIT_LIST_HEAD(&tmp->ctl_entry);
-	tmp->used = 0;
-	tmp->unregistering = NULL;
-	sysctl_set_parent(NULL, table);
-	if (sysctl_check_table(tmp->ctl_table)) {
-		kfree(tmp);
+
+	new = (struct ctl_table *) (header + 1);
+
+	/* Now connect the dots */
+	prevp = &header->ctl_table;
+	for (n = 0; n < npath; ++n, ++path) {
+		/* Copy the procname */
+		new->procname = path->procname;
+		new->ctl_name = path->ctl_name;
+		new->mode     = 0555;
+
+		*prevp = new;
+		prevp = &new->child;
+
+		new += 2;
+	}
+	*prevp = table;
+	header->ctl_table_arg = table;
+
+	INIT_LIST_HEAD(&header->ctl_entry);
+	header->used = 0;
+	header->unregistering = NULL;
+	header->root = root;
+	sysctl_set_parent(NULL, header->ctl_table);
+	if (sysctl_check_table(namespaces, header->ctl_table)) {
+		kfree(header);
 		return NULL;
 	}
 	spin_lock(&sysctl_lock);
-	list_add_tail(&tmp->ctl_entry, &root_table_header.ctl_entry);
+	header_list = lookup_header_list(root, namespaces);
+	list_add_tail(&header->ctl_entry, header_list);
 	spin_unlock(&sysctl_lock);
-	return tmp;
+
+	return header;
+}
+
+/**
+ * register_sysctl_table_path - register a sysctl table hierarchy
+ * @path: The path to the directory the sysctl table is in.
+ * @table: the top-level table structure
+ *
+ * Register a sysctl table hierarchy. @table should be a filled in ctl_table
+ * array. A completely 0 filled entry terminates the table.
+ *
+ * See __register_sysctl_paths for more details.
+ */
+struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
+						struct ctl_table *table)
+{
+	return __register_sysctl_paths(&sysctl_table_root, current->nsproxy,
+					path, table);
+}
+
+/**
+ * register_sysctl_table - register a sysctl table hierarchy
+ * @table: the top-level table structure
+ *
+ * Register a sysctl table hierarchy. @table should be a filled in ctl_table
+ * array. A completely 0 filled entry terminates the table.
+ *
+ * See register_sysctl_paths for more details.
+ */
+struct ctl_table_header *register_sysctl_table(struct ctl_table *table)
+{
+	static const struct ctl_path null_path[] = { {} };
+
+	return register_sysctl_paths(null_path, table);
 }
 
 /**
@@ -1604,6 +1800,12 @@ struct ctl_table_header *register_sysctl_table(struct ctl_table * table)
 	return NULL;
 }
 
+struct ctl_table_header *register_sysctl_paths(const struct ctl_path *path,
+						    struct ctl_table *table)
+{
+	return NULL;
+}
+
 void unregister_sysctl_table(struct ctl_table_header * table)
 {
 }
@@ -2662,6 +2864,7 @@ EXPORT_SYMBOL(proc_dostring);
 EXPORT_SYMBOL(proc_doulongvec_minmax);
 EXPORT_SYMBOL(proc_doulongvec_ms_jiffies_minmax);
 EXPORT_SYMBOL(register_sysctl_table);
+EXPORT_SYMBOL(register_sysctl_paths);
 EXPORT_SYMBOL(sysctl_intvec);
 EXPORT_SYMBOL(sysctl_jiffies);
 EXPORT_SYMBOL(sysctl_ms_jiffies);
diff --git a/kernel/sysctl_check.c b/kernel/sysctl_check.c
index a68425a5cc1..c3206fa5004 100644
--- a/kernel/sysctl_check.c
+++ b/kernel/sysctl_check.c
@@ -1,6 +1,5 @@
 #include <linux/stat.h>
 #include <linux/sysctl.h>
-#include "../arch/s390/appldata/appldata.h"
 #include "../fs/xfs/linux-2.6/xfs_sysctl.h"
 #include <linux/sunrpc/debug.h>
 #include <linux/string.h>
@@ -1343,7 +1342,8 @@ static void sysctl_repair_table(struct ctl_table *table)
 	}
 }
 
-static struct ctl_table *sysctl_check_lookup(struct ctl_table *table)
+static struct ctl_table *sysctl_check_lookup(struct nsproxy *namespaces,
+						struct ctl_table *table)
 {
 	struct ctl_table_header *head;
 	struct ctl_table *ref, *test;
@@ -1351,8 +1351,8 @@ static struct ctl_table *sysctl_check_lookup(struct ctl_table *table)
 
 	depth = sysctl_depth(table);
 
-	for (head = sysctl_head_next(NULL); head;
-	     head = sysctl_head_next(head)) {
+	for (head = __sysctl_head_next(namespaces, NULL); head;
+	     head = __sysctl_head_next(namespaces, head)) {
 		cur_depth = depth;
 		ref = head->ctl_table;
 repeat:
@@ -1397,13 +1397,14 @@ static void set_fail(const char **fail, struct ctl_table *table, const char *str
 	*fail = str;
 }
 
-static int sysctl_check_dir(struct ctl_table *table)
+static int sysctl_check_dir(struct nsproxy *namespaces,
+				struct ctl_table *table)
 {
 	struct ctl_table *ref;
 	int error;
 
 	error = 0;
-	ref = sysctl_check_lookup(table);
+	ref = sysctl_check_lookup(namespaces, table);
 	if (ref) {
 		int match = 0;
 		if ((!table->procname && !ref->procname) ||
@@ -1428,11 +1429,12 @@ static int sysctl_check_dir(struct ctl_table *table)
 	return error;
 }
 
-static void sysctl_check_leaf(struct ctl_table *table, const char **fail)
+static void sysctl_check_leaf(struct nsproxy *namespaces,
+				struct ctl_table *table, const char **fail)
 {
 	struct ctl_table *ref;
 
-	ref = sysctl_check_lookup(table);
+	ref = sysctl_check_lookup(namespaces, table);
 	if (ref && (ref != table))
 		set_fail(fail, table, "Sysctl already exists");
 }
@@ -1456,7 +1458,7 @@ static void sysctl_check_bin_path(struct ctl_table *table, const char **fail)
 	}
 }
 
-int sysctl_check_table(struct ctl_table *table)
+int sysctl_check_table(struct nsproxy *namespaces, struct ctl_table *table)
 {
 	int error = 0;
 	for (; table->ctl_name || table->procname; table++) {
@@ -1486,7 +1488,7 @@ int sysctl_check_table(struct ctl_table *table)
 				set_fail(&fail, table, "Directory with extra1");
 			if (table->extra2)
 				set_fail(&fail, table, "Directory with extra2");
-			if (sysctl_check_dir(table))
+			if (sysctl_check_dir(namespaces, table))
 				set_fail(&fail, table, "Inconsistent directory names");
 		} else {
 			if ((table->strategy == sysctl_data) ||
@@ -1535,7 +1537,7 @@ int sysctl_check_table(struct ctl_table *table)
 			if (!table->procname && table->proc_handler)
 				set_fail(&fail, table, "proc_handler without procname");
 #endif
-			sysctl_check_leaf(table, &fail);
+			sysctl_check_leaf(namespaces, table, &fail);
 		}
 		sysctl_check_bin_path(table, &fail);
 		if (fail) {
@@ -1543,7 +1545,7 @@ int sysctl_check_table(struct ctl_table *table)
 			error = -EINVAL;
 		}
 		if (table->child)
-			error |= sysctl_check_table(table->child);
+			error |= sysctl_check_table(namespaces, table->child);
 	}
 	return error;
 }
diff --git a/kernel/test_kprobes.c b/kernel/test_kprobes.c
new file mode 100644
index 00000000000..88cdb109e13
--- /dev/null
+++ b/kernel/test_kprobes.c
@@ -0,0 +1,216 @@
+/*
+ * test_kprobes.c - simple sanity test for *probes
+ *
+ * Copyright IBM Corp. 2008
+ *
+ * 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 would 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/kernel.h>
+#include <linux/kprobes.h>
+#include <linux/random.h>
+
+#define div_factor 3
+
+static u32 rand1, preh_val, posth_val, jph_val;
+static int errors, handler_errors, num_tests;
+
+static noinline u32 kprobe_target(u32 value)
+{
+	/*
+	 * gcc ignores noinline on some architectures unless we stuff
+	 * sufficient lard into the function. The get_kprobe() here is
+	 * just for that.
+	 *
+	 * NOTE: We aren't concerned about the correctness of get_kprobe()
+	 * here; hence, this call is neither under !preempt nor with the
+	 * kprobe_mutex held. This is fine(tm)
+	 */
+	if (get_kprobe((void *)0xdeadbeef))
+		printk(KERN_INFO "Kprobe smoke test: probe on 0xdeadbeef!\n");
+
+	return (value / div_factor);
+}
+
+static int kp_pre_handler(struct kprobe *p, struct pt_regs *regs)
+{
+	preh_val = (rand1 / div_factor);
+	return 0;
+}
+
+static void kp_post_handler(struct kprobe *p, struct pt_regs *regs,
+		unsigned long flags)
+{
+	if (preh_val != (rand1 / div_factor)) {
+		handler_errors++;
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"incorrect value in post_handler\n");
+	}
+	posth_val = preh_val + div_factor;
+}
+
+static struct kprobe kp = {
+	.symbol_name = "kprobe_target",
+	.pre_handler = kp_pre_handler,
+	.post_handler = kp_post_handler
+};
+
+static int test_kprobe(void)
+{
+	int ret;
+
+	ret = register_kprobe(&kp);
+	if (ret < 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"register_kprobe returned %d\n", ret);
+		return ret;
+	}
+
+	ret = kprobe_target(rand1);
+	unregister_kprobe(&kp);
+
+	if (preh_val == 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"kprobe pre_handler not called\n");
+		handler_errors++;
+	}
+
+	if (posth_val == 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"kprobe post_handler not called\n");
+		handler_errors++;
+	}
+
+	return 0;
+}
+
+static u32 j_kprobe_target(u32 value)
+{
+	if (value != rand1) {
+		handler_errors++;
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"incorrect value in jprobe handler\n");
+	}
+
+	jph_val = rand1;
+	jprobe_return();
+	return 0;
+}
+
+static struct jprobe jp = {
+	.entry		= j_kprobe_target,
+	.kp.symbol_name = "kprobe_target"
+};
+
+static int test_jprobe(void)
+{
+	int ret;
+
+	ret = register_jprobe(&jp);
+	if (ret < 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"register_jprobe returned %d\n", ret);
+		return ret;
+	}
+
+	ret = kprobe_target(rand1);
+	unregister_jprobe(&jp);
+	if (jph_val == 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"jprobe handler not called\n");
+		handler_errors++;
+	}
+
+	return 0;
+}
+
+#ifdef CONFIG_KRETPROBES
+static u32 krph_val;
+
+static int return_handler(struct kretprobe_instance *ri, struct pt_regs *regs)
+{
+	unsigned long ret = regs_return_value(regs);
+
+	if (ret != (rand1 / div_factor)) {
+		handler_errors++;
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"incorrect value in kretprobe handler\n");
+	}
+
+	krph_val = (rand1 / div_factor);
+	return 0;
+}
+
+static struct kretprobe rp = {
+	.handler	= return_handler,
+	.kp.symbol_name = "kprobe_target"
+};
+
+static int test_kretprobe(void)
+{
+	int ret;
+
+	ret = register_kretprobe(&rp);
+	if (ret < 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"register_kretprobe returned %d\n", ret);
+		return ret;
+	}
+
+	ret = kprobe_target(rand1);
+	unregister_kretprobe(&rp);
+	if (krph_val == 0) {
+		printk(KERN_ERR "Kprobe smoke test failed: "
+				"kretprobe handler not called\n");
+		handler_errors++;
+	}
+
+	return 0;
+}
+#endif /* CONFIG_KRETPROBES */
+
+int init_test_probes(void)
+{
+	int ret;
+
+	do {
+		rand1 = random32();
+	} while (rand1 <= div_factor);
+
+	printk(KERN_INFO "Kprobe smoke test started\n");
+	num_tests++;
+	ret = test_kprobe();
+	if (ret < 0)
+		errors++;
+
+	num_tests++;
+	ret = test_jprobe();
+	if (ret < 0)
+		errors++;
+
+#ifdef CONFIG_KRETPROBES
+	num_tests++;
+	ret = test_kretprobe();
+	if (ret < 0)
+		errors++;
+#endif /* CONFIG_KRETPROBES */
+
+	if (errors)
+		printk(KERN_ERR "BUG: Kprobe smoke test: %d out of "
+				"%d tests failed\n", errors, num_tests);
+	else if (handler_errors)
+		printk(KERN_ERR "BUG: Kprobe smoke test: %d error(s) "
+				"running handlers\n", handler_errors);
+	else
+		printk(KERN_INFO "Kprobe smoke test passed successfully\n");
+
+	return 0;
+}
diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c
index 5fb139fef9f..3e59fce6dd4 100644
--- a/kernel/time/clockevents.c
+++ b/kernel/time/clockevents.c
@@ -41,6 +41,11 @@ unsigned long clockevent_delta2ns(unsigned long latch,
 {
 	u64 clc = ((u64) latch << evt->shift);
 
+	if (unlikely(!evt->mult)) {
+		evt->mult = 1;
+		WARN_ON(1);
+	}
+
 	do_div(clc, evt->mult);
 	if (clc < 1000)
 		clc = 1000;
@@ -151,6 +156,14 @@ static void clockevents_notify_released(void)
 void clockevents_register_device(struct clock_event_device *dev)
 {
 	BUG_ON(dev->mode != CLOCK_EVT_MODE_UNUSED);
+	/*
+	 * A nsec2cyc multiplicator of 0 is invalid and we'd crash
+	 * on it, so fix it up and emit a warning:
+	 */
+	if (unlikely(!dev->mult)) {
+		dev->mult = 1;
+		WARN_ON(1);
+	}
 
 	spin_lock(&clockevents_lock);
 
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index c8a9d13874d..6e9259a5d50 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -142,8 +142,13 @@ static void clocksource_watchdog(unsigned long data)
 	}
 
 	if (!list_empty(&watchdog_list)) {
-		__mod_timer(&watchdog_timer,
-			    watchdog_timer.expires + WATCHDOG_INTERVAL);
+		/* Cycle through CPUs to check if the CPUs stay synchronized to
+		 * each other. */
+		int next_cpu = next_cpu(raw_smp_processor_id(), cpu_online_map);
+		if (next_cpu >= NR_CPUS)
+			next_cpu = first_cpu(cpu_online_map);
+		watchdog_timer.expires += WATCHDOG_INTERVAL;
+		add_timer_on(&watchdog_timer, next_cpu);
 	}
 	spin_unlock(&watchdog_lock);
 }
@@ -165,7 +170,7 @@ static void clocksource_check_watchdog(struct clocksource *cs)
 		if (!started && watchdog) {
 			watchdog_last = watchdog->read();
 			watchdog_timer.expires = jiffies + WATCHDOG_INTERVAL;
-			add_timer(&watchdog_timer);
+			add_timer_on(&watchdog_timer, first_cpu(cpu_online_map));
 		}
 	} else {
 		if (cs->flags & CLOCK_SOURCE_IS_CONTINUOUS)
@@ -175,7 +180,7 @@ static void clocksource_check_watchdog(struct clocksource *cs)
 			if (watchdog)
 				del_timer(&watchdog_timer);
 			watchdog = cs;
-			init_timer(&watchdog_timer);
+			init_timer_deferrable(&watchdog_timer);
 			watchdog_timer.function = clocksource_watchdog;
 
 			/* Reset watchdog cycles */
@@ -186,7 +191,8 @@ static void clocksource_check_watchdog(struct clocksource *cs)
 				watchdog_last = watchdog->read();
 				watchdog_timer.expires =
 					jiffies + WATCHDOG_INTERVAL;
-				add_timer(&watchdog_timer);
+				add_timer_on(&watchdog_timer,
+						first_cpu(cpu_online_map));
 			}
 		}
 	}
@@ -331,6 +337,21 @@ void clocksource_change_rating(struct clocksource *cs, int rating)
 	spin_unlock_irqrestore(&clocksource_lock, flags);
 }
 
+/**
+ * clocksource_unregister - remove a registered clocksource
+ */
+void clocksource_unregister(struct clocksource *cs)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&clocksource_lock, flags);
+	list_del(&cs->list);
+	if (clocksource_override == cs)
+		clocksource_override = NULL;
+	next_clocksource = select_clocksource();
+	spin_unlock_irqrestore(&clocksource_lock, flags);
+}
+
 #ifdef CONFIG_SYSFS
 /**
  * sysfs_show_current_clocksources - sysfs interface for current clocksource
@@ -441,7 +462,7 @@ static SYSDEV_ATTR(available_clocksource, 0600,
 		   sysfs_show_available_clocksources, NULL);
 
 static struct sysdev_class clocksource_sysclass = {
-	set_kset_name("clocksource"),
+	.name = "clocksource",
 };
 
 static struct sys_device device_clocksource = {
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c
index 5b86698faa0..e1bd50cbbf5 100644
--- a/kernel/time/tick-broadcast.c
+++ b/kernel/time/tick-broadcast.c
@@ -126,9 +126,9 @@ int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu)
 /*
  * Broadcast the event to the cpus, which are set in the mask
  */
-int tick_do_broadcast(cpumask_t mask)
+static void tick_do_broadcast(cpumask_t mask)
 {
-	int ret = 0, cpu = smp_processor_id();
+	int cpu = smp_processor_id();
 	struct tick_device *td;
 
 	/*
@@ -138,7 +138,6 @@ int tick_do_broadcast(cpumask_t mask)
 		cpu_clear(cpu, mask);
 		td = &per_cpu(tick_cpu_device, cpu);
 		td->evtdev->event_handler(td->evtdev);
-		ret = 1;
 	}
 
 	if (!cpus_empty(mask)) {
@@ -151,9 +150,7 @@ int tick_do_broadcast(cpumask_t mask)
 		cpu = first_cpu(mask);
 		td = &per_cpu(tick_cpu_device, cpu);
 		td->evtdev->broadcast(mask);
-		ret = 1;
 	}
-	return ret;
 }
 
 /*
diff --git a/kernel/time/tick-internal.h b/kernel/time/tick-internal.h
index bb13f272490..f13f2b7f4fd 100644
--- a/kernel/time/tick-internal.h
+++ b/kernel/time/tick-internal.h
@@ -70,8 +70,6 @@ static inline int tick_resume_broadcast_oneshot(struct clock_event_device *bc)
  * Broadcasting support
  */
 #ifdef CONFIG_GENERIC_CLOCKEVENTS_BROADCAST
-extern int tick_do_broadcast(cpumask_t mask);
-
 extern int tick_device_uses_broadcast(struct clock_event_device *dev, int cpu);
 extern int tick_check_broadcast_device(struct clock_event_device *dev);
 extern int tick_is_broadcast_device(struct clock_event_device *dev);
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index cb89fa8db11..63f24b55069 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -9,7 +9,7 @@
  *
  *  Started by: Thomas Gleixner and Ingo Molnar
  *
- *  For licencing details see kernel-base/COPYING
+ *  Distribute under GPLv2.
  */
 #include <linux/cpu.h>
 #include <linux/err.h>
@@ -143,6 +143,44 @@ void tick_nohz_update_jiffies(void)
 	local_irq_restore(flags);
 }
 
+void tick_nohz_stop_idle(int cpu)
+{
+	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
+
+	if (ts->idle_active) {
+		ktime_t now, delta;
+		now = ktime_get();
+		delta = ktime_sub(now, ts->idle_entrytime);
+		ts->idle_lastupdate = now;
+		ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta);
+		ts->idle_active = 0;
+	}
+}
+
+static ktime_t tick_nohz_start_idle(int cpu)
+{
+	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
+	ktime_t now, delta;
+
+	now = ktime_get();
+	if (ts->idle_active) {
+		delta = ktime_sub(now, ts->idle_entrytime);
+		ts->idle_lastupdate = now;
+		ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta);
+	}
+	ts->idle_entrytime = now;
+	ts->idle_active = 1;
+	return now;
+}
+
+u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time)
+{
+	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
+
+	*last_update_time = ktime_to_us(ts->idle_lastupdate);
+	return ktime_to_us(ts->idle_sleeptime);
+}
+
 /**
  * tick_nohz_stop_sched_tick - stop the idle tick from the idle task
  *
@@ -153,14 +191,16 @@ void tick_nohz_update_jiffies(void)
 void tick_nohz_stop_sched_tick(void)
 {
 	unsigned long seq, last_jiffies, next_jiffies, delta_jiffies, flags;
+	unsigned long rt_jiffies;
 	struct tick_sched *ts;
-	ktime_t last_update, expires, now, delta;
+	ktime_t last_update, expires, now;
 	struct clock_event_device *dev = __get_cpu_var(tick_cpu_device).evtdev;
 	int cpu;
 
 	local_irq_save(flags);
 
 	cpu = smp_processor_id();
+	now = tick_nohz_start_idle(cpu);
 	ts = &per_cpu(tick_cpu_sched, cpu);
 
 	/*
@@ -192,19 +232,7 @@ void tick_nohz_stop_sched_tick(void)
 		}
 	}
 
-	now = ktime_get();
-	/*
-	 * When called from irq_exit we need to account the idle sleep time
-	 * correctly.
-	 */
-	if (ts->tick_stopped) {
-		delta = ktime_sub(now, ts->idle_entrytime);
-		ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta);
-	}
-
-	ts->idle_entrytime = now;
 	ts->idle_calls++;
-
 	/* Read jiffies and the time when jiffies were updated last */
 	do {
 		seq = read_seqbegin(&xtime_lock);
@@ -216,6 +244,10 @@ void tick_nohz_stop_sched_tick(void)
 	next_jiffies = get_next_timer_interrupt(last_jiffies);
 	delta_jiffies = next_jiffies - last_jiffies;
 
+	rt_jiffies = rt_needs_cpu(cpu);
+	if (rt_jiffies && rt_jiffies < delta_jiffies)
+		delta_jiffies = rt_jiffies;
+
 	if (rcu_needs_cpu(cpu))
 		delta_jiffies = 1;
 	/*
@@ -291,7 +323,7 @@ void tick_nohz_stop_sched_tick(void)
 			/* Check, if the timer was already in the past */
 			if (hrtimer_active(&ts->sched_timer))
 				goto out;
-		} else if(!tick_program_event(expires, 0))
+		} else if (!tick_program_event(expires, 0))
 				goto out;
 		/*
 		 * We are past the event already. So we crossed a
@@ -332,23 +364,22 @@ void tick_nohz_restart_sched_tick(void)
 	int cpu = smp_processor_id();
 	struct tick_sched *ts = &per_cpu(tick_cpu_sched, cpu);
 	unsigned long ticks;
-	ktime_t now, delta;
+	ktime_t now;
 
-	if (!ts->tick_stopped)
+	local_irq_disable();
+	tick_nohz_stop_idle(cpu);
+
+	if (!ts->tick_stopped) {
+		local_irq_enable();
 		return;
+	}
 
 	/* Update jiffies first */
-	now = ktime_get();
-
-	local_irq_disable();
 	select_nohz_load_balancer(0);
+	now = ktime_get();
 	tick_do_update_jiffies64(now);
 	cpu_clear(cpu, nohz_cpu_mask);
 
-	/* Account the idle time */
-	delta = ktime_sub(now, ts->idle_entrytime);
-	ts->idle_sleeptime = ktime_add(ts->idle_sleeptime, delta);
-
 	/*
 	 * We stopped the tick in idle. Update process times would miss the
 	 * time we slept as update_process_times does only a 1 tick
@@ -502,14 +533,13 @@ static inline void tick_nohz_switch_to_nohz(void) { }
  */
 #ifdef CONFIG_HIGH_RES_TIMERS
 /*
- * We rearm the timer until we get disabled by the idle code
+ * We rearm the timer until we get disabled by the idle code.
  * Called with interrupts disabled and timer->base->cpu_base->lock held.
  */
 static enum hrtimer_restart tick_sched_timer(struct hrtimer *timer)
 {
 	struct tick_sched *ts =
 		container_of(timer, struct tick_sched, sched_timer);
-	struct hrtimer_cpu_base *base = timer->base->cpu_base;
 	struct pt_regs *regs = get_irq_regs();
 	ktime_t now = ktime_get();
 	int cpu = smp_processor_id();
@@ -547,15 +577,8 @@ static enum hrtimer_restart tick_sched_timer(struct hrtimer *timer)
 			touch_softlockup_watchdog();
 			ts->idle_jiffies++;
 		}
-		/*
-		 * update_process_times() might take tasklist_lock, hence
-		 * drop the base lock. sched-tick hrtimers are per-CPU and
-		 * never accessible by userspace APIs, so this is safe to do.
-		 */
-		spin_unlock(&base->lock);
 		update_process_times(user_mode(regs));
 		profile_tick(CPU_PROFILING);
-		spin_lock(&base->lock);
 	}
 
 	/* Do not restart, when we are in the idle loop */
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index e5e466b2759..092a2366b5a 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -82,13 +82,12 @@ static inline s64 __get_nsec_offset(void)
 }
 
 /**
- * __get_realtime_clock_ts - Returns the time of day in a timespec
+ * getnstimeofday - Returns the time of day in a timespec
  * @ts:		pointer to the timespec to be set
  *
- * Returns the time of day in a timespec. Used by
- * do_gettimeofday() and get_realtime_clock_ts().
+ * Returns the time of day in a timespec.
  */
-static inline void __get_realtime_clock_ts(struct timespec *ts)
+void getnstimeofday(struct timespec *ts)
 {
 	unsigned long seq;
 	s64 nsecs;
@@ -104,30 +103,19 @@ static inline void __get_realtime_clock_ts(struct timespec *ts)
 	timespec_add_ns(ts, nsecs);
 }
 
-/**
- * getnstimeofday - Returns the time of day in a timespec
- * @ts:		pointer to the timespec to be set
- *
- * Returns the time of day in a timespec.
- */
-void getnstimeofday(struct timespec *ts)
-{
-	__get_realtime_clock_ts(ts);
-}
-
 EXPORT_SYMBOL(getnstimeofday);
 
 /**
  * do_gettimeofday - Returns the time of day in a timeval
  * @tv:		pointer to the timeval to be set
  *
- * NOTE: Users should be converted to using get_realtime_clock_ts()
+ * NOTE: Users should be converted to using getnstimeofday()
  */
 void do_gettimeofday(struct timeval *tv)
 {
 	struct timespec now;
 
-	__get_realtime_clock_ts(&now);
+	getnstimeofday(&now);
 	tv->tv_sec = now.tv_sec;
 	tv->tv_usec = now.tv_nsec/1000;
 }
@@ -198,7 +186,8 @@ static void change_clocksource(void)
 
 	clock->error = 0;
 	clock->xtime_nsec = 0;
-	clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH);
+	clocksource_calculate_interval(clock,
+		(unsigned long)(current_tick_length()>>TICK_LENGTH_SHIFT));
 
 	tick_clock_notify();
 
@@ -255,7 +244,8 @@ void __init timekeeping_init(void)
 	ntp_clear();
 
 	clock = clocksource_get_next();
-	clocksource_calculate_interval(clock, NTP_INTERVAL_LENGTH);
+	clocksource_calculate_interval(clock,
+		(unsigned long)(current_tick_length()>>TICK_LENGTH_SHIFT));
 	clock->cycle_last = clocksource_read(clock);
 
 	xtime.tv_sec = sec;
@@ -335,9 +325,9 @@ static int timekeeping_suspend(struct sys_device *dev, pm_message_t state)
 
 /* sysfs resume/suspend bits for timekeeping */
 static struct sysdev_class timekeeping_sysclass = {
+	.name		= "timekeeping",
 	.resume		= timekeeping_resume,
 	.suspend	= timekeeping_suspend,
-	set_kset_name("timekeeping"),
 };
 
 static struct sys_device device_timer = {
diff --git a/kernel/time/timer_stats.c b/kernel/time/timer_stats.c
index c36bb7ed030..417da8c5bc7 100644
--- a/kernel/time/timer_stats.c
+++ b/kernel/time/timer_stats.c
@@ -26,7 +26,7 @@
  * the pid and cmdline from the owner process if applicable.
  *
  * Start/stop data collection:
- * # echo 1[0] >/proc/timer_stats
+ * # echo [1|0] >/proc/timer_stats
  *
  * Display the information collected so far:
  * # cat /proc/timer_stats
diff --git a/kernel/timer.c b/kernel/timer.c
index 2a00c22203f..23f7ead78fa 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -58,59 +58,57 @@ EXPORT_SYMBOL(jiffies_64);
 #define TVN_MASK (TVN_SIZE - 1)
 #define TVR_MASK (TVR_SIZE - 1)
 
-typedef struct tvec_s {
+struct tvec {
 	struct list_head vec[TVN_SIZE];
-} tvec_t;
+};
 
-typedef struct tvec_root_s {
+struct tvec_root {
 	struct list_head vec[TVR_SIZE];
-} tvec_root_t;
+};
 
-struct tvec_t_base_s {
+struct tvec_base {
 	spinlock_t lock;
 	struct timer_list *running_timer;
 	unsigned long timer_jiffies;
-	tvec_root_t tv1;
-	tvec_t tv2;
-	tvec_t tv3;
-	tvec_t tv4;
-	tvec_t tv5;
+	struct tvec_root tv1;
+	struct tvec tv2;
+	struct tvec tv3;
+	struct tvec tv4;
+	struct tvec tv5;
 } ____cacheline_aligned;
 
-typedef struct tvec_t_base_s tvec_base_t;
-
-tvec_base_t boot_tvec_bases;
+struct tvec_base boot_tvec_bases;
 EXPORT_SYMBOL(boot_tvec_bases);
-static DEFINE_PER_CPU(tvec_base_t *, tvec_bases) = &boot_tvec_bases;
+static DEFINE_PER_CPU(struct tvec_base *, tvec_bases) = &boot_tvec_bases;
 
 /*
- * Note that all tvec_bases is 2 byte aligned and lower bit of
+ * Note that all tvec_bases are 2 byte aligned and lower bit of
  * base in timer_list is guaranteed to be zero. Use the LSB for
  * the new flag to indicate whether the timer is deferrable
  */
 #define TBASE_DEFERRABLE_FLAG		(0x1)
 
 /* Functions below help us manage 'deferrable' flag */
-static inline unsigned int tbase_get_deferrable(tvec_base_t *base)
+static inline unsigned int tbase_get_deferrable(struct tvec_base *base)
 {
 	return ((unsigned int)(unsigned long)base & TBASE_DEFERRABLE_FLAG);
 }
 
-static inline tvec_base_t *tbase_get_base(tvec_base_t *base)
+static inline struct tvec_base *tbase_get_base(struct tvec_base *base)
 {
-	return ((tvec_base_t *)((unsigned long)base & ~TBASE_DEFERRABLE_FLAG));
+	return ((struct tvec_base *)((unsigned long)base & ~TBASE_DEFERRABLE_FLAG));
 }
 
 static inline void timer_set_deferrable(struct timer_list *timer)
 {
-	timer->base = ((tvec_base_t *)((unsigned long)(timer->base) |
+	timer->base = ((struct tvec_base *)((unsigned long)(timer->base) |
 				       TBASE_DEFERRABLE_FLAG));
 }
 
 static inline void
-timer_set_base(struct timer_list *timer, tvec_base_t *new_base)
+timer_set_base(struct timer_list *timer, struct tvec_base *new_base)
 {
-	timer->base = (tvec_base_t *)((unsigned long)(new_base) |
+	timer->base = (struct tvec_base *)((unsigned long)(new_base) |
 				      tbase_get_deferrable(timer->base));
 }
 
@@ -246,7 +244,7 @@ unsigned long round_jiffies_relative(unsigned long j)
 EXPORT_SYMBOL_GPL(round_jiffies_relative);
 
 
-static inline void set_running_timer(tvec_base_t *base,
+static inline void set_running_timer(struct tvec_base *base,
 					struct timer_list *timer)
 {
 #ifdef CONFIG_SMP
@@ -254,7 +252,7 @@ static inline void set_running_timer(tvec_base_t *base,
 #endif
 }
 
-static void internal_add_timer(tvec_base_t *base, struct timer_list *timer)
+static void internal_add_timer(struct tvec_base *base, struct timer_list *timer)
 {
 	unsigned long expires = timer->expires;
 	unsigned long idx = expires - base->timer_jiffies;
@@ -371,14 +369,14 @@ static inline void detach_timer(struct timer_list *timer,
  * possible to set timer->base = NULL and drop the lock: the timer remains
  * locked.
  */
-static tvec_base_t *lock_timer_base(struct timer_list *timer,
+static struct tvec_base *lock_timer_base(struct timer_list *timer,
 					unsigned long *flags)
 	__acquires(timer->base->lock)
 {
-	tvec_base_t *base;
+	struct tvec_base *base;
 
 	for (;;) {
-		tvec_base_t *prelock_base = timer->base;
+		struct tvec_base *prelock_base = timer->base;
 		base = tbase_get_base(prelock_base);
 		if (likely(base != NULL)) {
 			spin_lock_irqsave(&base->lock, *flags);
@@ -393,7 +391,7 @@ static tvec_base_t *lock_timer_base(struct timer_list *timer,
 
 int __mod_timer(struct timer_list *timer, unsigned long expires)
 {
-	tvec_base_t *base, *new_base;
+	struct tvec_base *base, *new_base;
 	unsigned long flags;
 	int ret = 0;
 
@@ -445,7 +443,7 @@ EXPORT_SYMBOL(__mod_timer);
  */
 void add_timer_on(struct timer_list *timer, int cpu)
 {
-	tvec_base_t *base = per_cpu(tvec_bases, cpu);
+	struct tvec_base *base = per_cpu(tvec_bases, cpu);
 	unsigned long flags;
 
 	timer_stats_timer_set_start_info(timer);
@@ -508,7 +506,7 @@ EXPORT_SYMBOL(mod_timer);
  */
 int del_timer(struct timer_list *timer)
 {
-	tvec_base_t *base;
+	struct tvec_base *base;
 	unsigned long flags;
 	int ret = 0;
 
@@ -539,7 +537,7 @@ EXPORT_SYMBOL(del_timer);
  */
 int try_to_del_timer_sync(struct timer_list *timer)
 {
-	tvec_base_t *base;
+	struct tvec_base *base;
 	unsigned long flags;
 	int ret = -1;
 
@@ -591,7 +589,7 @@ int del_timer_sync(struct timer_list *timer)
 EXPORT_SYMBOL(del_timer_sync);
 #endif
 
-static int cascade(tvec_base_t *base, tvec_t *tv, int index)
+static int cascade(struct tvec_base *base, struct tvec *tv, int index)
 {
 	/* cascade all the timers from tv up one level */
 	struct timer_list *timer, *tmp;
@@ -620,7 +618,7 @@ static int cascade(tvec_base_t *base, tvec_t *tv, int index)
  * This function cascades all vectors and executes all expired timer
  * vectors.
  */
-static inline void __run_timers(tvec_base_t *base)
+static inline void __run_timers(struct tvec_base *base)
 {
 	struct timer_list *timer;
 
@@ -657,7 +655,7 @@ static inline void __run_timers(tvec_base_t *base)
 				int preempt_count = preempt_count();
 				fn(data);
 				if (preempt_count != preempt_count()) {
-					printk(KERN_WARNING "huh, entered %p "
+					printk(KERN_ERR "huh, entered %p "
 					       "with preempt_count %08x, exited"
 					       " with %08x?\n",
 					       fn, preempt_count,
@@ -678,13 +676,13 @@ static inline void __run_timers(tvec_base_t *base)
  * is used on S/390 to stop all activity when a cpus is idle.
  * This functions needs to be called disabled.
  */
-static unsigned long __next_timer_interrupt(tvec_base_t *base)
+static unsigned long __next_timer_interrupt(struct tvec_base *base)
 {
 	unsigned long timer_jiffies = base->timer_jiffies;
 	unsigned long expires = timer_jiffies + NEXT_TIMER_MAX_DELTA;
 	int index, slot, array, found = 0;
 	struct timer_list *nte;
-	tvec_t *varray[4];
+	struct tvec *varray[4];
 
 	/* Look for timer events in tv1. */
 	index = slot = timer_jiffies & TVR_MASK;
@@ -716,7 +714,7 @@ cascade:
 	varray[3] = &base->tv5;
 
 	for (array = 0; array < 4; array++) {
-		tvec_t *varp = varray[array];
+		struct tvec *varp = varray[array];
 
 		index = slot = timer_jiffies & TVN_MASK;
 		do {
@@ -795,7 +793,7 @@ static unsigned long cmp_next_hrtimer_event(unsigned long now,
  */
 unsigned long get_next_timer_interrupt(unsigned long now)
 {
-	tvec_base_t *base = __get_cpu_var(tvec_bases);
+	struct tvec_base *base = __get_cpu_var(tvec_bases);
 	unsigned long expires;
 
 	spin_lock(&base->lock);
@@ -894,9 +892,9 @@ static inline void calc_load(unsigned long ticks)
  */
 static void run_timer_softirq(struct softirq_action *h)
 {
-	tvec_base_t *base = __get_cpu_var(tvec_bases);
+	struct tvec_base *base = __get_cpu_var(tvec_bases);
 
-	hrtimer_run_queues();
+	hrtimer_run_pending();
 
 	if (time_after_eq(jiffies, base->timer_jiffies))
 		__run_timers(base);
@@ -907,6 +905,7 @@ static void run_timer_softirq(struct softirq_action *h)
  */
 void run_local_timers(void)
 {
+	hrtimer_run_queues();
 	raise_softirq(TIMER_SOFTIRQ);
 	softlockup_tick();
 }
@@ -1222,7 +1221,7 @@ static struct lock_class_key base_lock_keys[NR_CPUS];
 static int __cpuinit init_timers_cpu(int cpu)
 {
 	int j;
-	tvec_base_t *base;
+	struct tvec_base *base;
 	static char __cpuinitdata tvec_base_done[NR_CPUS];
 
 	if (!tvec_base_done[cpu]) {
@@ -1277,7 +1276,7 @@ static int __cpuinit init_timers_cpu(int cpu)
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
-static void migrate_timer_list(tvec_base_t *new_base, struct list_head *head)
+static void migrate_timer_list(struct tvec_base *new_base, struct list_head *head)
 {
 	struct timer_list *timer;
 
@@ -1291,8 +1290,8 @@ static void migrate_timer_list(tvec_base_t *new_base, struct list_head *head)
 
 static void __cpuinit migrate_timers(int cpu)
 {
-	tvec_base_t *old_base;
-	tvec_base_t *new_base;
+	struct tvec_base *old_base;
+	struct tvec_base *new_base;
 	int i;
 
 	BUG_ON(cpu_online(cpu));
diff --git a/kernel/user.c b/kernel/user.c
index 8320a87f3e5..bc1c48d35cb 100644
--- a/kernel/user.c
+++ b/kernel/user.c
@@ -115,7 +115,7 @@ static void sched_switch_user(struct task_struct *p) { }
 
 #if defined(CONFIG_FAIR_USER_SCHED) && defined(CONFIG_SYSFS)
 
-static struct kobject uids_kobject; /* represents /sys/kernel/uids directory */
+static struct kset *uids_kset; /* represents the /sys/kernel/uids/ directory */
 static DEFINE_MUTEX(uids_mutex);
 
 static inline void uids_mutex_lock(void)
@@ -128,86 +128,83 @@ static inline void uids_mutex_unlock(void)
 	mutex_unlock(&uids_mutex);
 }
 
-/* return cpu shares held by the user */
-static ssize_t cpu_shares_show(struct kset *kset, char *buffer)
+/* uid directory attributes */
+static ssize_t cpu_shares_show(struct kobject *kobj,
+			       struct kobj_attribute *attr,
+			       char *buf)
 {
-	struct user_struct *up = container_of(kset, struct user_struct, kset);
+	struct user_struct *up = container_of(kobj, struct user_struct, kobj);
 
-	return sprintf(buffer, "%lu\n", sched_group_shares(up->tg));
+	return sprintf(buf, "%lu\n", sched_group_shares(up->tg));
 }
 
-/* modify cpu shares held by the user */
-static ssize_t cpu_shares_store(struct kset *kset, const char *buffer,
-				size_t size)
+static ssize_t cpu_shares_store(struct kobject *kobj,
+				struct kobj_attribute *attr,
+				const char *buf, size_t size)
 {
-	struct user_struct *up = container_of(kset, struct user_struct, kset);
+	struct user_struct *up = container_of(kobj, struct user_struct, kobj);
 	unsigned long shares;
 	int rc;
 
-	sscanf(buffer, "%lu", &shares);
+	sscanf(buf, "%lu", &shares);
 
 	rc = sched_group_set_shares(up->tg, shares);
 
 	return (rc ? rc : size);
 }
 
-static void user_attr_init(struct subsys_attribute *sa, char *name, int mode)
+static struct kobj_attribute cpu_share_attr =
+	__ATTR(cpu_share, 0644, cpu_shares_show, cpu_shares_store);
+
+/* default attributes per uid directory */
+static struct attribute *uids_attributes[] = {
+	&cpu_share_attr.attr,
+	NULL
+};
+
+/* the lifetime of user_struct is not managed by the core (now) */
+static void uids_release(struct kobject *kobj)
 {
-	sa->attr.name = name;
-	sa->attr.mode = mode;
-	sa->show = cpu_shares_show;
-	sa->store = cpu_shares_store;
+	return;
 }
 
-/* Create "/sys/kernel/uids/<uid>" directory and
- *  "/sys/kernel/uids/<uid>/cpu_share" file for this user.
- */
-static int user_kobject_create(struct user_struct *up)
+static struct kobj_type uids_ktype = {
+	.sysfs_ops = &kobj_sysfs_ops,
+	.default_attrs = uids_attributes,
+	.release = uids_release,
+};
+
+/* create /sys/kernel/uids/<uid>/cpu_share file for this user */
+static int uids_user_create(struct user_struct *up)
 {
-	struct kset *kset = &up->kset;
-	struct kobject *kobj = &kset->kobj;
+	struct kobject *kobj = &up->kobj;
 	int error;
 
-	memset(kset, 0, sizeof(struct kset));
-	kobj->parent = &uids_kobject;	/* create under /sys/kernel/uids dir */
-	kobject_set_name(kobj, "%d", up->uid);
-	kset_init(kset);
-	user_attr_init(&up->user_attr, "cpu_share", 0644);
-
-	error = kobject_add(kobj);
-	if (error)
+	memset(kobj, 0, sizeof(struct kobject));
+	kobj->kset = uids_kset;
+	error = kobject_init_and_add(kobj, &uids_ktype, NULL, "%d", up->uid);
+	if (error) {
+		kobject_put(kobj);
 		goto done;
-
-	error = sysfs_create_file(kobj, &up->user_attr.attr);
-	if (error)
-		kobject_del(kobj);
+	}
 
 	kobject_uevent(kobj, KOBJ_ADD);
-
 done:
 	return error;
 }
 
-/* create these in sysfs filesystem:
+/* create these entries in sysfs:
  * 	"/sys/kernel/uids" directory
  * 	"/sys/kernel/uids/0" directory (for root user)
  * 	"/sys/kernel/uids/0/cpu_share" file (for root user)
  */
-int __init uids_kobject_init(void)
+int __init uids_sysfs_init(void)
 {
-	int error;
-
-	/* create under /sys/kernel dir */
-	uids_kobject.parent = &kernel_subsys.kobj;
-	uids_kobject.kset = &kernel_subsys;
-	kobject_set_name(&uids_kobject, "uids");
-	kobject_init(&uids_kobject);
+	uids_kset = kset_create_and_add("uids", NULL, kernel_kobj);
+	if (!uids_kset)
+		return -ENOMEM;
 
-	error = kobject_add(&uids_kobject);
-	if (!error)
-		error = user_kobject_create(&root_user);
-
-	return error;
+	return uids_user_create(&root_user);
 }
 
 /* work function to remove sysfs directory for a user and free up
@@ -216,7 +213,6 @@ int __init uids_kobject_init(void)
 static void remove_user_sysfs_dir(struct work_struct *w)
 {
 	struct user_struct *up = container_of(w, struct user_struct, work);
-	struct kobject *kobj = &up->kset.kobj;
 	unsigned long flags;
 	int remove_user = 0;
 
@@ -238,9 +234,9 @@ static void remove_user_sysfs_dir(struct work_struct *w)
 	if (!remove_user)
 		goto done;
 
-	sysfs_remove_file(kobj, &up->user_attr.attr);
-	kobject_uevent(kobj, KOBJ_REMOVE);
-	kobject_del(kobj);
+	kobject_uevent(&up->kobj, KOBJ_REMOVE);
+	kobject_del(&up->kobj);
+	kobject_put(&up->kobj);
 
 	sched_destroy_user(up);
 	key_put(up->uid_keyring);
@@ -267,7 +263,8 @@ static inline void free_user(struct user_struct *up, unsigned long flags)
 
 #else	/* CONFIG_FAIR_USER_SCHED && CONFIG_SYSFS */
 
-static inline int user_kobject_create(struct user_struct *up) { return 0; }
+int uids_sysfs_init(void) { return 0; }
+static inline int uids_user_create(struct user_struct *up) { return 0; }
 static inline void uids_mutex_lock(void) { }
 static inline void uids_mutex_unlock(void) { }
 
@@ -322,9 +319,9 @@ void free_uid(struct user_struct *up)
 struct user_struct * alloc_uid(struct user_namespace *ns, uid_t uid)
 {
 	struct hlist_head *hashent = uidhashentry(ns, uid);
-	struct user_struct *up;
+	struct user_struct *up, *new;
 
-	/* Make uid_hash_find() + user_kobject_create() + uid_hash_insert()
+	/* Make uid_hash_find() + uids_user_create() + uid_hash_insert()
 	 * atomic.
 	 */
 	uids_mutex_lock();
@@ -334,13 +331,9 @@ struct user_struct * alloc_uid(struct user_namespace *ns, uid_t uid)
 	spin_unlock_irq(&uidhash_lock);
 
 	if (!up) {
-		struct user_struct *new;
-
 		new = kmem_cache_alloc(uid_cachep, GFP_KERNEL);
-		if (!new) {
-			uids_mutex_unlock();
-			return NULL;
-		}
+		if (!new)
+			goto out_unlock;
 
 		new->uid = uid;
 		atomic_set(&new->__count, 1);
@@ -356,28 +349,14 @@ struct user_struct * alloc_uid(struct user_namespace *ns, uid_t uid)
 #endif
 		new->locked_shm = 0;
 
-		if (alloc_uid_keyring(new, current) < 0) {
-			kmem_cache_free(uid_cachep, new);
-			uids_mutex_unlock();
-			return NULL;
-		}
+		if (alloc_uid_keyring(new, current) < 0)
+			goto out_free_user;
 
-		if (sched_create_user(new) < 0) {
-			key_put(new->uid_keyring);
-			key_put(new->session_keyring);
-			kmem_cache_free(uid_cachep, new);
-			uids_mutex_unlock();
-			return NULL;
-		}
+		if (sched_create_user(new) < 0)
+			goto out_put_keys;
 
-		if (user_kobject_create(new)) {
-			sched_destroy_user(new);
-			key_put(new->uid_keyring);
-			key_put(new->session_keyring);
-			kmem_cache_free(uid_cachep, new);
-			uids_mutex_unlock();
-			return NULL;
-		}
+		if (uids_user_create(new))
+			goto out_destoy_sched;
 
 		/*
 		 * Before adding this, check whether we raced
@@ -405,6 +384,17 @@ struct user_struct * alloc_uid(struct user_namespace *ns, uid_t uid)
 	uids_mutex_unlock();
 
 	return up;
+
+out_destoy_sched:
+	sched_destroy_user(new);
+out_put_keys:
+	key_put(new->uid_keyring);
+	key_put(new->session_keyring);
+out_free_user:
+	kmem_cache_free(uid_cachep, new);
+out_unlock:
+	uids_mutex_unlock();
+	return NULL;
 }
 
 void switch_uid(struct user_struct *new_user)
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 8db0b597509..52db48e7f6e 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -67,9 +67,8 @@ struct workqueue_struct {
 #endif
 };
 
-/* All the per-cpu workqueues on the system, for hotplug cpu to add/remove
-   threads to each one as cpus come/go. */
-static DEFINE_MUTEX(workqueue_mutex);
+/* Serializes the accesses to the list of workqueues. */
+static DEFINE_SPINLOCK(workqueue_lock);
 static LIST_HEAD(workqueues);
 
 static int singlethread_cpu __read_mostly;
@@ -592,8 +591,6 @@ EXPORT_SYMBOL(schedule_delayed_work_on);
  * Returns zero on success.
  * Returns -ve errno on failure.
  *
- * Appears to be racy against CPU hotplug.
- *
  * schedule_on_each_cpu() is very slow.
  */
 int schedule_on_each_cpu(work_func_t func)
@@ -605,7 +602,7 @@ int schedule_on_each_cpu(work_func_t func)
 	if (!works)
 		return -ENOMEM;
 
-	preempt_disable();		/* CPU hotplug */
+	get_online_cpus();
 	for_each_online_cpu(cpu) {
 		struct work_struct *work = per_cpu_ptr(works, cpu);
 
@@ -613,8 +610,8 @@ int schedule_on_each_cpu(work_func_t func)
 		set_bit(WORK_STRUCT_PENDING, work_data_bits(work));
 		__queue_work(per_cpu_ptr(keventd_wq->cpu_wq, cpu), work);
 	}
-	preempt_enable();
 	flush_workqueue(keventd_wq);
+	put_online_cpus();
 	free_percpu(works);
 	return 0;
 }
@@ -750,8 +747,10 @@ struct workqueue_struct *__create_workqueue_key(const char *name,
 		err = create_workqueue_thread(cwq, singlethread_cpu);
 		start_workqueue_thread(cwq, -1);
 	} else {
-		mutex_lock(&workqueue_mutex);
+		get_online_cpus();
+		spin_lock(&workqueue_lock);
 		list_add(&wq->list, &workqueues);
+		spin_unlock(&workqueue_lock);
 
 		for_each_possible_cpu(cpu) {
 			cwq = init_cpu_workqueue(wq, cpu);
@@ -760,7 +759,7 @@ struct workqueue_struct *__create_workqueue_key(const char *name,
 			err = create_workqueue_thread(cwq, cpu);
 			start_workqueue_thread(cwq, cpu);
 		}
-		mutex_unlock(&workqueue_mutex);
+		put_online_cpus();
 	}
 
 	if (err) {
@@ -775,7 +774,7 @@ static void cleanup_workqueue_thread(struct cpu_workqueue_struct *cwq, int cpu)
 {
 	/*
 	 * Our caller is either destroy_workqueue() or CPU_DEAD,
-	 * workqueue_mutex protects cwq->thread
+	 * get_online_cpus() protects cwq->thread.
 	 */
 	if (cwq->thread == NULL)
 		return;
@@ -810,9 +809,11 @@ void destroy_workqueue(struct workqueue_struct *wq)
 	struct cpu_workqueue_struct *cwq;
 	int cpu;
 
-	mutex_lock(&workqueue_mutex);
+	get_online_cpus();
+	spin_lock(&workqueue_lock);
 	list_del(&wq->list);
-	mutex_unlock(&workqueue_mutex);
+	spin_unlock(&workqueue_lock);
+	put_online_cpus();
 
 	for_each_cpu_mask(cpu, *cpu_map) {
 		cwq = per_cpu_ptr(wq->cpu_wq, cpu);
@@ -835,13 +836,6 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb,
 	action &= ~CPU_TASKS_FROZEN;
 
 	switch (action) {
-	case CPU_LOCK_ACQUIRE:
-		mutex_lock(&workqueue_mutex);
-		return NOTIFY_OK;
-
-	case CPU_LOCK_RELEASE:
-		mutex_unlock(&workqueue_mutex);
-		return NOTIFY_OK;
 
 	case CPU_UP_PREPARE:
 		cpu_set(cpu, cpu_populated_map);
@@ -854,7 +848,8 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb,
 		case CPU_UP_PREPARE:
 			if (!create_workqueue_thread(cwq, cpu))
 				break;
-			printk(KERN_ERR "workqueue for %i failed\n", cpu);
+			printk(KERN_ERR "workqueue [%s] for %i failed\n",
+				wq->name, cpu);
 			return NOTIFY_BAD;
 
 		case CPU_ONLINE: