Merge branch 'quilt/rr'

6 files changed, 143 insertions, 202 deletions

diff --git a/kernel/cpu.c b/kernel/cpu.c
index d26d0b095b3b..472149ca920e 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -214,7 +214,6 @@ static int __ref take_cpu_down(void *_param)
 static int __ref _cpu_down(unsigned int cpu, int tasks_frozen)
 {
 	int err, nr_calls = 0;
-	struct task_struct *p;
 	cpumask_t old_allowed, tmp;
 	void *hcpu = (void *)(long)cpu;
 	unsigned long mod = tasks_frozen ? CPU_TASKS_FROZEN : 0;
@@ -248,19 +247,15 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen)
 	cpu_clear(cpu, tmp);
 	set_cpus_allowed_ptr(current, &tmp);
 
-	p = __stop_machine_run(take_cpu_down, &tcd_param, cpu);
+	err = __stop_machine_run(take_cpu_down, &tcd_param, cpu);
 
-	if (IS_ERR(p) || cpu_online(cpu)) {
+	if (err || cpu_online(cpu)) {
 		/* CPU didn't die: tell everyone.  Can't complain. */
 		if (raw_notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED | mod,
 					    hcpu) == NOTIFY_BAD)
 			BUG();
 
-		if (IS_ERR(p)) {
-			err = PTR_ERR(p);
-			goto out_allowed;
-		}
-		goto out_thread;
+		goto out_allowed;
 	}
 
 	/* Wait for it to sleep (leaving idle task). */
@@ -277,8 +272,6 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen)
 
 	check_for_tasks(cpu);
 
-out_thread:
-	err = kthread_stop(p);
 out_allowed:
 	set_cpus_allowed_ptr(current, &old_allowed);
 out_release:
diff --git a/kernel/kthread.c b/kernel/kthread.c
index 97747cdd37c9..27ab8afc6996 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -111,29 +111,10 @@ static void create_kthread(struct kthread_create_info *create)
 	complete(&create->done);
 }
 
-/**
- * kthread_create - create a kthread.
- * @threadfn: the function to run until signal_pending(current).
- * @data: data ptr for @threadfn.
- * @namefmt: printf-style name for the thread.
- *
- * Description: This helper function creates and names a kernel
- * thread.  The thread will be stopped: use wake_up_process() to start
- * it.  See also kthread_run(), kthread_create_on_cpu().
- *
- * When woken, the thread will run @threadfn() with @data as its
- * argument. @threadfn() can either call do_exit() directly if it is a
- * standalone thread for which noone will call kthread_stop(), or
- * return when 'kthread_should_stop()' is true (which means
- * kthread_stop() has been called).  The return value should be zero
- * or a negative error number; it will be passed to kthread_stop().
- *
- * Returns a task_struct or ERR_PTR(-ENOMEM).
- */
-struct task_struct *kthread_create(int (*threadfn)(void *data),
-				   void *data,
-				   const char namefmt[],
-				   ...)
+struct task_struct *__kthread_create(int (*threadfn)(void *data),
+				     void *data,
+				     const char namefmt[],
+				     ...)
 {
 	struct kthread_create_info create;
 
@@ -158,7 +139,7 @@ struct task_struct *kthread_create(int (*threadfn)(void *data),
 	}
 	return create.result;
 }
-EXPORT_SYMBOL(kthread_create);
+EXPORT_SYMBOL(__kthread_create);
 
 /**
  * kthread_bind - bind a just-created kthread to a cpu.
diff --git a/kernel/mutex.c b/kernel/mutex.c
index bcdc9ac8ef60..7502055dc6d0 100644
--- a/kernel/mutex.c
+++ b/kernel/mutex.c
@@ -370,8 +370,8 @@ static inline int __mutex_trylock_slowpath(atomic_t *lock_count)
  * Try to acquire the mutex atomically. Returns 1 if the mutex
  * has been acquired successfully, and 0 on contention.
  *
- * NOTE: this function follows the spin_trylock() convention, so
- * it is negated to the down_trylock() return values! Be careful
+ * NOTE: this function follows the spin_trylock()/down_try() convention,
+ * so it is negated to the old down_trylock() return values! Be careful
  * about this when converting semaphore users to mutexes.
  *
  * This function must not be used in interrupt context. The
diff --git a/kernel/printk.c b/kernel/printk.c
index 84d640530bfe..9544f1b7a003 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -969,7 +969,7 @@ EXPORT_SYMBOL(acquire_console_sem);
 
 int try_acquire_console_sem(void)
 {
-	if (down_trylock(&console_sem))
+	if (!down_try(&console_sem))
 		return -1;
 	console_locked = 1;
 	console_may_schedule = 0;
@@ -1068,7 +1068,7 @@ void console_unblank(void)
 	 * oops_in_progress is set to 1..
 	 */
 	if (oops_in_progress) {
-		if (down_trylock(&console_sem) != 0)
+		if (!down_try(&console_sem))
 			return;
 	} else
 		acquire_console_sem();
diff --git a/kernel/semaphore.c b/kernel/semaphore.c
index 1a064adab658..bbab232ee185 100644
--- a/kernel/semaphore.c
+++ b/kernel/semaphore.c
@@ -14,7 +14,7 @@
  * Some notes on the implementation:
  *
  * The spinlock controls access to the other members of the semaphore.
- * down_trylock() and up() can be called from interrupt context, so we
+ * down_try() and up() can be called from interrupt context, so we
  * have to disable interrupts when taking the lock.  It turns out various
  * parts of the kernel expect to be able to use down() on a semaphore in
  * interrupt context when they know it will succeed, so we have to use
@@ -116,19 +116,18 @@ int down_killable(struct semaphore *sem)
 EXPORT_SYMBOL(down_killable);
 
 /**
- * down_trylock - try to acquire the semaphore, without waiting
+ * down_try - try to acquire the semaphore, without waiting
  * @sem: the semaphore to be acquired
  *
- * Try to acquire the semaphore atomically.  Returns 0 if the mutex has
- * been acquired successfully or 1 if it it cannot be acquired.
+ * Try to acquire the semaphore atomically.  Returns true if the mutex has
+ * been acquired successfully or 0 if it it cannot be acquired.
  *
- * NOTE: This return value is inverted from both spin_trylock and
- * mutex_trylock!  Be careful about this when converting code.
+ * NOTE: This replaces down_trylock() which returned the reverse.
  *
  * Unlike mutex_trylock, this function can be used from interrupt context,
  * and the semaphore can be released by any task or interrupt.
  */
-int down_trylock(struct semaphore *sem)
+int down_try(struct semaphore *sem)
 {
 	unsigned long flags;
 	int count;
@@ -139,9 +138,9 @@ int down_trylock(struct semaphore *sem)
 		sem->count = count;
 	spin_unlock_irqrestore(&sem->lock, flags);
 
-	return (count < 0);
+	return (count >= 0);
 }
-EXPORT_SYMBOL(down_trylock);
+EXPORT_SYMBOL(down_try);
 
 /**
  * down_timeout - acquire the semaphore within a specified time
diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
index b7350bbfb076..36e166def7c7 100644
--- a/kernel/stop_machine.c
+++ b/kernel/stop_machine.c
@@ -13,203 +13,171 @@
 #include <asm/atomic.h>
 #include <asm/uaccess.h>
 
-/* Since we effect priority and affinity (both of which are visible
- * to, and settable by outside processes) we do indirection via a
- * kthread. */
-
-/* Thread to stop each CPU in user context. */
+/* This controls the threads on each CPU. */
 enum stopmachine_state {
-	STOPMACHINE_WAIT,
-	STOPMACHINE_PREPARE,
+	/* Dummy starting state for thread. */
+	STOPMACHINE_NONE,
+	/* Disable interrupts. */
 	STOPMACHINE_DISABLE_IRQ,
+	/* Run the function */
+	STOPMACHINE_RUN,
+	/* Exit */
 	STOPMACHINE_EXIT,
+	/* Everyone exited. */
+	STOPMACHINE_COMPLETE,
 };
+static enum stopmachine_state state;
 
-static enum stopmachine_state stopmachine_state;
-static unsigned int stopmachine_num_threads;
-static atomic_t stopmachine_thread_ack;
+struct stop_machine_data {
+	int (*fn)(void *);
+	void *data;
+	int fnret;
+};
+
+/* Like num_online_cpus(), but hotplug cpu uses us, so we need this. */
+static unsigned int num_threads;
+static atomic_t thread_ack;
+static struct completion finished;
 
-static int stopmachine(void *cpu)
+static void set_state(enum stopmachine_state newstate)
 {
-	int irqs_disabled = 0;
-	int prepared = 0;
+	/* Reset ack counter. */
+	atomic_set(&thread_ack, num_threads);
+	smp_wmb();
+	state = newstate;
+}
 
-	set_cpus_allowed_ptr(current, &cpumask_of_cpu((int)(long)cpu));
+/* Last one to ack a state moves to the next state. */
+static void ack_state(void)
+{
+	if (atomic_dec_and_test(&thread_ack)) {
+		set_state(state + 1);
+		if (state == STOPMACHINE_COMPLETE)
+			complete(&finished);
+	}
+}
 
-	/* Ack: we are alive */
-	smp_mb(); /* Theoretically the ack = 0 might not be on this CPU yet. */
-	atomic_inc(&stopmachine_thread_ack);
+/* This is the actual thread which stops the CPU.  It exits by itself rather
+ * than waiting for kthread_stop(), because it's easier for hotplug CPU. */
+static int stop_cpu(struct stop_machine_data *smdata)
+{
+	enum stopmachine_state curstate = STOPMACHINE_NONE;
+	int uninitialized_var(ret);
 
 	/* Simple state machine */
-	while (stopmachine_state != STOPMACHINE_EXIT) {
-		if (stopmachine_state == STOPMACHINE_DISABLE_IRQ 
-		    && !irqs_disabled) {
-			local_irq_disable();
-			hard_irq_disable();
-			irqs_disabled = 1;
-			/* Ack: irqs disabled. */
-			smp_mb(); /* Must read state first. */
-			atomic_inc(&stopmachine_thread_ack);
-		} else if (stopmachine_state == STOPMACHINE_PREPARE
-			   && !prepared) {
-			/* Everyone is in place, hold CPU. */
-			preempt_disable();
-			prepared = 1;
-			smp_mb(); /* Must read state first. */
-			atomic_inc(&stopmachine_thread_ack);
-		}
-		/* Yield in first stage: migration threads need to
-		 * help our sisters onto their CPUs. */
-		if (!prepared && !irqs_disabled)
-			yield();
+	do {
+		/* Chill out and ensure we re-read stopmachine_state. */
 		cpu_relax();
-	}
-
-	/* Ack: we are exiting. */
-	smp_mb(); /* Must read state first. */
-	atomic_inc(&stopmachine_thread_ack);
-
-	if (irqs_disabled)
-		local_irq_enable();
-	if (prepared)
-		preempt_enable();
+		if (state != curstate) {
+			curstate = state;
+			switch (curstate) {
+			case STOPMACHINE_DISABLE_IRQ:
+				local_irq_disable();
+				hard_irq_disable();
+				break;
+			case STOPMACHINE_RUN:
+				/* |= allows error detection if functions on
+				 * multiple CPUs. */
+				smdata->fnret |= smdata->fn(smdata->data);
+				break;
+			default:
+				break;
+			}
+			ack_state();
+		}
+	} while (curstate < STOPMACHINE_EXIT);
 
-	return 0;
+	local_irq_enable();
+	do_exit(0);
 }
 
-/* Change the thread state */
-static void stopmachine_set_state(enum stopmachine_state state)
+/* Callback for CPUs which aren't supposed to do anything. */
+static int chill(void *unused)
 {
-	atomic_set(&stopmachine_thread_ack, 0);
-	smp_wmb();
-	stopmachine_state = state;
-	while (atomic_read(&stopmachine_thread_ack) != stopmachine_num_threads)
-		cpu_relax();
+	return 0;
 }
 
-static int stop_machine(void)
+int __stop_machine_run(int (*fn)(void *), void *data, unsigned int cpu)
 {
-	int i, ret = 0;
-
-	atomic_set(&stopmachine_thread_ack, 0);
-	stopmachine_num_threads = 0;
-	stopmachine_state = STOPMACHINE_WAIT;
-
-	for_each_online_cpu(i) {
-		if (i == raw_smp_processor_id())
-			continue;
-		ret = kernel_thread(stopmachine, (void *)(long)i,CLONE_KERNEL);
-		if (ret < 0)
-			break;
-		stopmachine_num_threads++;
-	}
-
-	/* Wait for them all to come to life. */
-	while (atomic_read(&stopmachine_thread_ack) != stopmachine_num_threads) {
-		yield();
-		cpu_relax();
-	}
-
-	/* If some failed, kill them all. */
-	if (ret < 0) {
-		stopmachine_set_state(STOPMACHINE_EXIT);
-		return ret;
-	}
+	int i, err;
+	struct stop_machine_data active, idle;
+	struct task_struct **threads;
 
-	/* Now they are all started, make them hold the CPUs, ready. */
-	preempt_disable();
-	stopmachine_set_state(STOPMACHINE_PREPARE);
+	active.fn = fn;
+	active.data = data;
+	active.fnret = 0;
+	idle.fn = chill;
+	idle.data = NULL;
 
-	/* Make them disable irqs. */
-	local_irq_disable();
-	hard_irq_disable();
-	stopmachine_set_state(STOPMACHINE_DISABLE_IRQ);
+	/* If they don't care which cpu fn runs on, just pick one. */
+	if (cpu == NR_CPUS)
+		cpu = any_online_cpu(cpu_online_map);
 
-	return 0;
-}
+	/* This could be too big for stack on large machines. */
+	threads = kcalloc(NR_CPUS, sizeof(threads[0]), GFP_KERNEL);
+	if (!threads)
+		return -ENOMEM;
 
-static void restart_machine(void)
-{
-	stopmachine_set_state(STOPMACHINE_EXIT);
-	local_irq_enable();
-	preempt_enable_no_resched();
-}
+	/* Set up initial state. */
+	init_completion(&finished);
+	num_threads = num_online_cpus();
+	set_state(STOPMACHINE_DISABLE_IRQ);
 
-struct stop_machine_data {
-	int (*fn)(void *);
-	void *data;
-	struct completion done;
-};
+	for_each_online_cpu(i) {
+		struct stop_machine_data *smdata;
+		struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 };
 
-static int do_stop(void *_smdata)
-{
-	struct stop_machine_data *smdata = _smdata;
-	int ret;
+		if (cpu == ALL_CPUS || i == cpu)
+			smdata = &active;
+		else
+			smdata = &idle;
 
-	ret = stop_machine();
-	if (ret == 0) {
-		ret = smdata->fn(smdata->data);
-		restart_machine();
-	}
+		threads[i] = kthread_create(stop_cpu, smdata, "kstop%u", i);
+		if (IS_ERR(threads[i])) {
+			err = PTR_ERR(threads[i]);
+			threads[i] = NULL;
+			goto kill_threads;
+		}
 
-	/* We're done: you can kthread_stop us now */
-	complete(&smdata->done);
+		/* Place it onto correct cpu. */
+		kthread_bind(threads[i], i);
 
-	/* Wait for kthread_stop */
-	set_current_state(TASK_INTERRUPTIBLE);
-	while (!kthread_should_stop()) {
-		schedule();
-		set_current_state(TASK_INTERRUPTIBLE);
+		/* Make it highest prio. */
+		if (sched_setscheduler(threads[i], SCHED_FIFO, &param) != 0)
+			BUG();
 	}
-	__set_current_state(TASK_RUNNING);
-	return ret;
-}
-
-struct task_struct *__stop_machine_run(int (*fn)(void *), void *data,
-				       unsigned int cpu)
-{
-	static DEFINE_MUTEX(stopmachine_mutex);
-	struct stop_machine_data smdata;
-	struct task_struct *p;
 
-	smdata.fn = fn;
-	smdata.data = data;
-	init_completion(&smdata.done);
+	/* We've created all the threads.  Wake them all: hold this CPU so one
+	 * doesn't hit this CPU until we're ready. */
+	cpu = get_cpu();
+	for_each_online_cpu(i)
+		wake_up_process(threads[i]);
 
-	mutex_lock(&stopmachine_mutex);
+	/* This will release the thread on our CPU. */
+	put_cpu();
+	wait_for_completion(&finished);
 
-	/* If they don't care which CPU fn runs on, bind to any online one. */
-	if (cpu == NR_CPUS)
-		cpu = raw_smp_processor_id();
+	kfree(threads);
 
-	p = kthread_create(do_stop, &smdata, "kstopmachine");
-	if (!IS_ERR(p)) {
-		struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 };
+	return active.fnret;
 
-		/* One high-prio thread per cpu.  We'll do this one. */
-		sched_setscheduler(p, SCHED_FIFO, &param);
-		kthread_bind(p, cpu);
-		wake_up_process(p);
-		wait_for_completion(&smdata.done);
-	}
-	mutex_unlock(&stopmachine_mutex);
-	return p;
+kill_threads:
+	for_each_online_cpu(i)
+		if (threads[i])
+			kthread_stop(threads[i]);
+	kfree(threads);
+	return err;
 }
 
-int stop_machine_run(int (*fn)(void *), void *data, unsigned int cpu)
+int stop_machine_run_notype(int (*fn)(void *), void *data, unsigned int cpu)
 {
-	struct task_struct *p;
 	int ret;
 
 	/* No CPUs can come up or down during this. */
 	get_online_cpus();
-	p = __stop_machine_run(fn, data, cpu);
-	if (!IS_ERR(p))
-		ret = kthread_stop(p);
-	else
-		ret = PTR_ERR(p);
+	ret = __stop_machine_run(fn, data, cpu);
 	put_online_cpus();
 
 	return ret;
 }
-EXPORT_SYMBOL_GPL(stop_machine_run);
+EXPORT_SYMBOL_GPL(stop_machine_run_notype);