#define TVN_MASK (TVN_SIZE - 1)
#define TVR_MASK (TVR_SIZE - 1)
+struct timer_base_s {
+ spinlock_t lock;
+ struct timer_list *running_timer;
+};
+
typedef struct tvec_s {
struct list_head vec[TVN_SIZE];
} tvec_t;
} tvec_root_t;
struct tvec_t_base_s {
- spinlock_t lock;
+ struct timer_base_s t_base;
unsigned long timer_jiffies;
- struct timer_list *running_timer;
tvec_root_t tv1;
tvec_t tv2;
tvec_t tv3;
} ____cacheline_aligned_in_smp;
typedef struct tvec_t_base_s tvec_base_t;
+static DEFINE_PER_CPU(tvec_base_t, tvec_bases);
static inline void set_running_timer(tvec_base_t *base,
struct timer_list *timer)
{
#ifdef CONFIG_SMP
- base->running_timer = timer;
+ base->t_base.running_timer = timer;
#endif
}
-/* Fake initialization */
-static DEFINE_PER_CPU(tvec_base_t, tvec_bases) = { SPIN_LOCK_UNLOCKED };
-
static void check_timer_failed(struct timer_list *timer)
{
static int whine_count;
/*
* Now fix it up
*/
- spin_lock_init(&timer->lock);
timer->magic = TIMER_MAGIC;
}
list_add_tail(&timer->entry, vec);
}
+typedef struct timer_base_s timer_base_t;
+/*
+ * Used by TIMER_INITIALIZER, we can't use per_cpu(tvec_bases)
+ * at compile time, and we need timer->base to lock the timer.
+ */
+timer_base_t __init_timer_base
+ ____cacheline_aligned_in_smp = { .lock = SPIN_LOCK_UNLOCKED };
+EXPORT_SYMBOL(__init_timer_base);
+
+/***
+ * init_timer - initialize a timer.
+ * @timer: the timer to be initialized
+ *
+ * init_timer() must be done to a timer prior calling *any* of the
+ * other timer functions.
+ */
+void fastcall init_timer(struct timer_list *timer)
+{
+ timer->entry.next = NULL;
+ timer->base = &per_cpu(tvec_bases, raw_smp_processor_id()).t_base;
+ timer->magic = TIMER_MAGIC;
+}
+EXPORT_SYMBOL(init_timer);
+
+static inline void detach_timer(struct timer_list *timer,
+ int clear_pending)
+{
+ struct list_head *entry = &timer->entry;
+
+ __list_del(entry->prev, entry->next);
+ if (clear_pending)
+ entry->next = NULL;
+ entry->prev = LIST_POISON2;
+}
+
+/*
+ * We are using hashed locking: holding per_cpu(tvec_bases).t_base.lock
+ * means that all timers which are tied to this base via timer->base are
+ * locked, and the base itself is locked too.
+ *
+ * So __run_timers/migrate_timers can safely modify all timers which could
+ * be found on ->tvX lists.
+ *
+ * When the timer's base is locked, and the timer removed from list, it is
+ * possible to set timer->base = NULL and drop the lock: the timer remains
+ * locked.
+ */
+static timer_base_t *lock_timer_base(struct timer_list *timer,
+ unsigned long *flags)
+{
+ timer_base_t *base;
+
+ for (;;) {
+ base = timer->base;
+ if (likely(base != NULL)) {
+ spin_lock_irqsave(&base->lock, *flags);
+ if (likely(base == timer->base))
+ return base;
+ /* The timer has migrated to another CPU */
+ spin_unlock_irqrestore(&base->lock, *flags);
+ }
+ cpu_relax();
+ }
+}
+
int __mod_timer(struct timer_list *timer, unsigned long expires)
{
- tvec_base_t *old_base, *new_base;
+ timer_base_t *base;
+ tvec_base_t *new_base;
unsigned long flags;
int ret = 0;
BUG_ON(!timer->function);
-
check_timer(timer);
- spin_lock_irqsave(&timer->lock, flags);
+ base = lock_timer_base(timer, &flags);
+
+ if (timer_pending(timer)) {
+ detach_timer(timer, 0);
+ ret = 1;
+ }
+
new_base = &__get_cpu_var(tvec_bases);
-repeat:
- old_base = timer->base;
- /*
- * Prevent deadlocks via ordering by old_base < new_base.
- */
- if (old_base && (new_base != old_base)) {
- if (old_base < new_base) {
- spin_lock(&new_base->lock);
- spin_lock(&old_base->lock);
- } else {
- spin_lock(&old_base->lock);
- spin_lock(&new_base->lock);
- }
+ if (base != &new_base->t_base) {
/*
- * The timer base might have been cancelled while we were
- * trying to take the lock(s):
+ * We are trying to schedule the timer on the local CPU.
+ * However we can't change timer's base while it is running,
+ * otherwise del_timer_sync() can't detect that the timer's
+ * handler yet has not finished. This also guarantees that
+ * the timer is serialized wrt itself.
*/
- if (timer->base != old_base) {
- spin_unlock(&new_base->lock);
- spin_unlock(&old_base->lock);
- goto repeat;
- }
- } else {
- spin_lock(&new_base->lock);
- if (timer->base != old_base) {
- spin_unlock(&new_base->lock);
- goto repeat;
+ if (unlikely(base->running_timer == timer)) {
+ /* The timer remains on a former base */
+ new_base = container_of(base, tvec_base_t, t_base);
+ } else {
+ /* See the comment in lock_timer_base() */
+ timer->base = NULL;
+ spin_unlock(&base->lock);
+ spin_lock(&new_base->t_base.lock);
+ timer->base = &new_base->t_base;
}
}
- /*
- * Delete the previous timeout (if there was any), and install
- * the new one:
- */
- if (old_base) {
- list_del(&timer->entry);
- ret = 1;
- }
timer->expires = expires;
internal_add_timer(new_base, timer);
- timer->base = new_base;
-
- if (old_base && (new_base != old_base))
- spin_unlock(&old_base->lock);
- spin_unlock(&new_base->lock);
- spin_unlock_irqrestore(&timer->lock, flags);
+ spin_unlock_irqrestore(&new_base->t_base.lock, flags);
return ret;
}
{
tvec_base_t *base = &per_cpu(tvec_bases, cpu);
unsigned long flags;
-
+
BUG_ON(timer_pending(timer) || !timer->function);
check_timer(timer);
- spin_lock_irqsave(&base->lock, flags);
+ spin_lock_irqsave(&base->t_base.lock, flags);
+ timer->base = &base->t_base;
internal_add_timer(base, timer);
- timer->base = base;
- spin_unlock_irqrestore(&base->lock, flags);
+ spin_unlock_irqrestore(&base->t_base.lock, flags);
}
*/
int del_timer(struct timer_list *timer)
{
+ timer_base_t *base;
unsigned long flags;
- tvec_base_t *base;
+ int ret = 0;
check_timer(timer);
-repeat:
- base = timer->base;
- if (!base)
- return 0;
- spin_lock_irqsave(&base->lock, flags);
- if (base != timer->base) {
+ if (timer_pending(timer)) {
+ base = lock_timer_base(timer, &flags);
+ if (timer_pending(timer)) {
+ detach_timer(timer, 1);
+ ret = 1;
+ }
spin_unlock_irqrestore(&base->lock, flags);
- goto repeat;
}
- list_del(&timer->entry);
- /* Need to make sure that anybody who sees a NULL base also sees the list ops */
- smp_wmb();
- timer->base = NULL;
- spin_unlock_irqrestore(&base->lock, flags);
- return 1;
+ return ret;
}
EXPORT_SYMBOL(del_timer);
#ifdef CONFIG_SMP
-/***
- * del_timer_sync - deactivate a timer and wait for the handler to finish.
- * @timer: the timer to be deactivated
- *
- * This function only differs from del_timer() on SMP: besides deactivating
- * the timer it also makes sure the handler has finished executing on other
- * CPUs.
- *
- * Synchronization rules: callers must prevent restarting of the timer,
- * otherwise this function is meaningless. It must not be called from
- * interrupt contexts. The caller must not hold locks which would prevent
- * completion of the timer's handler. Upon exit the timer is not queued and
- * the handler is not running on any CPU.
- *
- * The function returns whether it has deactivated a pending timer or not.
+/*
+ * This function tries to deactivate a timer. Upon successful (ret >= 0)
+ * exit the timer is not queued and the handler is not running on any CPU.
*
- * del_timer_sync() is slow and complicated because it copes with timer
- * handlers which re-arm the timer (periodic timers). If the timer handler
- * is known to not do this (a single shot timer) then use
- * del_singleshot_timer_sync() instead.
+ * It must not be called from interrupt contexts.
*/
-int del_timer_sync(struct timer_list *timer)
+int try_to_del_timer_sync(struct timer_list *timer)
{
- tvec_base_t *base;
- int i, ret = 0;
+ timer_base_t *base;
+ unsigned long flags;
+ int ret = -1;
- check_timer(timer);
+ base = lock_timer_base(timer, &flags);
-del_again:
- ret += del_timer(timer);
+ if (base->running_timer == timer)
+ goto out;
- for_each_online_cpu(i) {
- base = &per_cpu(tvec_bases, i);
- if (base->running_timer == timer) {
- while (base->running_timer == timer) {
- cpu_relax();
- preempt_check_resched();
- }
- break;
- }
+ ret = 0;
+ if (timer_pending(timer)) {
+ detach_timer(timer, 1);
+ ret = 1;
}
- smp_rmb();
- if (timer_pending(timer))
- goto del_again;
+out:
+ spin_unlock_irqrestore(&base->lock, flags);
return ret;
}
-EXPORT_SYMBOL(del_timer_sync);
/***
- * del_singleshot_timer_sync - deactivate a non-recursive timer
+ * del_timer_sync - deactivate a timer and wait for the handler to finish.
* @timer: the timer to be deactivated
*
- * This function is an optimization of del_timer_sync for the case where the
- * caller can guarantee the timer does not reschedule itself in its timer
- * function.
+ * This function only differs from del_timer() on SMP: besides deactivating
+ * the timer it also makes sure the handler has finished executing on other
+ * CPUs.
*
* Synchronization rules: callers must prevent restarting of the timer,
* otherwise this function is meaningless. It must not be called from
- * interrupt contexts. The caller must not hold locks which wold prevent
- * completion of the timer's handler. Upon exit the timer is not queued and
- * the handler is not running on any CPU.
+ * interrupt contexts. The caller must not hold locks which would prevent
+ * completion of the timer's handler. The timer's handler must not call
+ * add_timer_on(). Upon exit the timer is not queued and the handler is
+ * not running on any CPU.
*
* The function returns whether it has deactivated a pending timer or not.
*/
-int del_singleshot_timer_sync(struct timer_list *timer)
+int del_timer_sync(struct timer_list *timer)
{
- int ret = del_timer(timer);
+ check_timer(timer);
- if (!ret) {
- ret = del_timer_sync(timer);
- BUG_ON(ret);
+ for (;;) {
+ int ret = try_to_del_timer_sync(timer);
+ if (ret >= 0)
+ return ret;
}
-
- return ret;
}
-EXPORT_SYMBOL(del_singleshot_timer_sync);
+
+EXPORT_SYMBOL(del_timer_sync);
#endif
static int cascade(tvec_base_t *base, tvec_t *tv, int index)
struct timer_list *tmp;
tmp = list_entry(curr, struct timer_list, entry);
- BUG_ON(tmp->base != base);
+ BUG_ON(tmp->base != &base->t_base);
curr = curr->next;
internal_add_timer(base, tmp);
}
{
struct timer_list *timer;
- spin_lock_irq(&base->lock);
+ spin_lock_irq(&base->t_base.lock);
while (time_after_eq(jiffies, base->timer_jiffies)) {
struct list_head work_list = LIST_HEAD_INIT(work_list);
struct list_head *head = &work_list;
cascade(base, &base->tv5, INDEX(3));
++base->timer_jiffies;
list_splice_init(base->tv1.vec + index, &work_list);
-repeat:
- if (!list_empty(head)) {
+ while (!list_empty(head)) {
void (*fn)(unsigned long);
unsigned long data;
fn = timer->function;
data = timer->data;
- list_del(&timer->entry);
set_running_timer(base, timer);
- smp_wmb();
- timer->base = NULL;
- spin_unlock_irq(&base->lock);
+ detach_timer(timer, 1);
+ spin_unlock_irq(&base->t_base.lock);
{
- u32 preempt_count = preempt_count();
+ int preempt_count = preempt_count();
fn(data);
if (preempt_count != preempt_count()) {
- printk("huh, entered %p with %08x, exited with %08x?\n", fn, preempt_count, preempt_count());
+ printk(KERN_WARNING "huh, entered %p "
+ "with preempt_count %08x, exited"
+ " with %08x?\n",
+ fn, preempt_count,
+ preempt_count());
BUG();
}
}
- spin_lock_irq(&base->lock);
- goto repeat;
+ spin_lock_irq(&base->t_base.lock);
}
}
set_running_timer(base, NULL);
- spin_unlock_irq(&base->lock);
+ spin_unlock_irq(&base->t_base.lock);
}
#ifdef CONFIG_NO_IDLE_HZ
int i, j;
base = &__get_cpu_var(tvec_bases);
- spin_lock(&base->lock);
+ spin_lock(&base->t_base.lock);
expires = base->timer_jiffies + (LONG_MAX >> 1);
list = 0;
expires = nte->expires;
}
}
- spin_unlock(&base->lock);
+ spin_unlock(&base->t_base.lock);
return expires;
}
#endif
{
jiffies_64++;
update_times();
+ softlockup_tick(regs);
}
#ifdef __ARCH_WANT_SYS_ALARM
parent = me->group_leader->real_parent;
for (;;) {
pid = parent->tgid;
-#ifdef CONFIG_SMP
+#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT)
{
struct task_struct *old = parent;
{
int j;
tvec_base_t *base;
-
+
base = &per_cpu(tvec_bases, cpu);
- spin_lock_init(&base->lock);
+ spin_lock_init(&base->t_base.lock);
for (j = 0; j < TVN_SIZE; j++) {
INIT_LIST_HEAD(base->tv5.vec + j);
INIT_LIST_HEAD(base->tv4.vec + j);
}
#ifdef CONFIG_HOTPLUG_CPU
-static int migrate_timer_list(tvec_base_t *new_base, struct list_head *head)
+static void migrate_timer_list(tvec_base_t *new_base, struct list_head *head)
{
struct timer_list *timer;
while (!list_empty(head)) {
timer = list_entry(head->next, struct timer_list, entry);
- /* We're locking backwards from __mod_timer order here,
- beware deadlock. */
- if (!spin_trylock(&timer->lock))
- return 0;
- list_del(&timer->entry);
+ detach_timer(timer, 0);
+ timer->base = &new_base->t_base;
internal_add_timer(new_base, timer);
- timer->base = new_base;
- spin_unlock(&timer->lock);
}
- return 1;
}
static void __devinit migrate_timers(int cpu)
new_base = &get_cpu_var(tvec_bases);
local_irq_disable();
-again:
- /* Prevent deadlocks via ordering by old_base < new_base. */
- if (old_base < new_base) {
- spin_lock(&new_base->lock);
- spin_lock(&old_base->lock);
- } else {
- spin_lock(&old_base->lock);
- spin_lock(&new_base->lock);
- }
+ spin_lock(&new_base->t_base.lock);
+ spin_lock(&old_base->t_base.lock);
- if (old_base->running_timer)
+ if (old_base->t_base.running_timer)
BUG();
for (i = 0; i < TVR_SIZE; i++)
- if (!migrate_timer_list(new_base, old_base->tv1.vec + i))
- goto unlock_again;
- for (i = 0; i < TVN_SIZE; i++)
- if (!migrate_timer_list(new_base, old_base->tv2.vec + i)
- || !migrate_timer_list(new_base, old_base->tv3.vec + i)
- || !migrate_timer_list(new_base, old_base->tv4.vec + i)
- || !migrate_timer_list(new_base, old_base->tv5.vec + i))
- goto unlock_again;
- spin_unlock(&old_base->lock);
- spin_unlock(&new_base->lock);
+ migrate_timer_list(new_base, old_base->tv1.vec + i);
+ for (i = 0; i < TVN_SIZE; i++) {
+ migrate_timer_list(new_base, old_base->tv2.vec + i);
+ migrate_timer_list(new_base, old_base->tv3.vec + i);
+ migrate_timer_list(new_base, old_base->tv4.vec + i);
+ migrate_timer_list(new_base, old_base->tv5.vec + i);
+ }
+
+ spin_unlock(&old_base->t_base.lock);
+ spin_unlock(&new_base->t_base.lock);
local_irq_enable();
put_cpu_var(tvec_bases);
- return;
-
-unlock_again:
- /* Avoid deadlock with __mod_timer, by backing off. */
- spin_unlock(&old_base->lock);
- spin_unlock(&new_base->lock);
- cpu_relax();
- goto again;
}
#endif /* CONFIG_HOTPLUG_CPU */
}
}
-static inline u64 time_interpolator_get_counter(void)
+static inline u64 time_interpolator_get_counter(int writelock)
{
unsigned int src = time_interpolator->source;
now = time_interpolator_get_cycles(src);
if (lcycle && time_after(lcycle, now))
return lcycle;
+
+ /* When holding the xtime write lock, there's no need
+ * to add the overhead of the cmpxchg. Readers are
+ * force to retry until the write lock is released.
+ */
+ if (writelock) {
+ time_interpolator->last_cycle = now;
+ return now;
+ }
/* Keep track of the last timer value returned. The use of cmpxchg here
* will cause contention in an SMP environment.
*/
void time_interpolator_reset(void)
{
time_interpolator->offset = 0;
- time_interpolator->last_counter = time_interpolator_get_counter();
+ time_interpolator->last_counter = time_interpolator_get_counter(1);
}
#define GET_TI_NSECS(count,i) (((((count) - i->last_counter) & (i)->mask) * (i)->nsec_per_cyc) >> (i)->shift)
return 0;
return time_interpolator->offset +
- GET_TI_NSECS(time_interpolator_get_counter(), time_interpolator);
+ GET_TI_NSECS(time_interpolator_get_counter(0), time_interpolator);
}
#define INTERPOLATOR_ADJUST 65536
* and the tuning logic insures that.
*/
- counter = time_interpolator_get_counter();
+ counter = time_interpolator_get_counter(1);
offset = time_interpolator->offset + GET_TI_NSECS(counter, time_interpolator);
if (delta_nsec < 0 || (unsigned long) delta_nsec < offset)
EXPORT_SYMBOL(msleep);
/**
- * msleep_interruptible - sleep waiting for waitqueue interruptions
+ * msleep_interruptible - sleep waiting for signals
* @msecs: Time in milliseconds to sleep for
*/
unsigned long msleep_interruptible(unsigned int msecs)
git.openpandora.org / pandora-kernel.git / blobdiff