*
* (C) Jens Axboe <jens.axboe@oracle.com> 2008
*/
+#include <linux/irq_work.h>
#include <linux/rcupdate.h>
#include <linux/rculist.h>
#include <linux/kernel.h>
#include <linux/gfp.h>
#include <linux/smp.h>
#include <linux/cpu.h>
+#include <linux/sched.h>
#include "smpboot.h"
-#ifdef CONFIG_USE_GENERIC_SMP_HELPERS
enum {
CSD_FLAG_LOCK = 0x01,
+ CSD_FLAG_SYNCHRONOUS = 0x02,
};
struct call_function_data {
struct call_single_data __percpu *csd;
cpumask_var_t cpumask;
- cpumask_var_t cpumask_ipi;
};
static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_function_data, cfd_data);
-struct call_single_queue {
- struct list_head list;
- raw_spinlock_t lock;
-};
+static DEFINE_PER_CPU_SHARED_ALIGNED(struct llist_head, call_single_queue);
-static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_single_queue, call_single_queue);
+static void flush_smp_call_function_queue(bool warn_cpu_offline);
static int
hotplug_cfd(struct notifier_block *nfb, unsigned long action, void *hcpu)
if (!zalloc_cpumask_var_node(&cfd->cpumask, GFP_KERNEL,
cpu_to_node(cpu)))
return notifier_from_errno(-ENOMEM);
- if (!zalloc_cpumask_var_node(&cfd->cpumask_ipi, GFP_KERNEL,
- cpu_to_node(cpu)))
- return notifier_from_errno(-ENOMEM);
cfd->csd = alloc_percpu(struct call_single_data);
if (!cfd->csd) {
free_cpumask_var(cfd->cpumask);
#ifdef CONFIG_HOTPLUG_CPU
case CPU_UP_CANCELED:
case CPU_UP_CANCELED_FROZEN:
+ /* Fall-through to the CPU_DEAD[_FROZEN] case. */
case CPU_DEAD:
case CPU_DEAD_FROZEN:
free_cpumask_var(cfd->cpumask);
- free_cpumask_var(cfd->cpumask_ipi);
free_percpu(cfd->csd);
break;
+
+ case CPU_DYING:
+ case CPU_DYING_FROZEN:
+ /*
+ * The IPIs for the smp-call-function callbacks queued by other
+ * CPUs might arrive late, either due to hardware latencies or
+ * because this CPU disabled interrupts (inside stop-machine)
+ * before the IPIs were sent. So flush out any pending callbacks
+ * explicitly (without waiting for the IPIs to arrive), to
+ * ensure that the outgoing CPU doesn't go offline with work
+ * still pending.
+ */
+ flush_smp_call_function_queue(false);
+ break;
#endif
};
return NOTIFY_OK;
}
-static struct notifier_block __cpuinitdata hotplug_cfd_notifier = {
+static struct notifier_block hotplug_cfd_notifier = {
.notifier_call = hotplug_cfd,
};
void *cpu = (void *)(long)smp_processor_id();
int i;
- for_each_possible_cpu(i) {
- struct call_single_queue *q = &per_cpu(call_single_queue, i);
-
- raw_spin_lock_init(&q->lock);
- INIT_LIST_HEAD(&q->list);
- }
+ for_each_possible_cpu(i)
+ init_llist_head(&per_cpu(call_single_queue, i));
hotplug_cfd(&hotplug_cfd_notifier, CPU_UP_PREPARE, cpu);
register_cpu_notifier(&hotplug_cfd_notifier);
*/
static void csd_lock_wait(struct call_single_data *csd)
{
- while (csd->flags & CSD_FLAG_LOCK)
+ while (smp_load_acquire(&csd->flags) & CSD_FLAG_LOCK)
cpu_relax();
}
* to ->flags with any subsequent assignments to other
* fields of the specified call_single_data structure:
*/
- smp_mb();
+ smp_wmb();
}
static void csd_unlock(struct call_single_data *csd)
/*
* ensure we're all done before releasing data:
*/
- smp_mb();
-
- csd->flags &= ~CSD_FLAG_LOCK;
+ smp_store_release(&csd->flags, 0);
}
+static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_single_data, csd_data);
+
/*
* Insert a previously allocated call_single_data element
* for execution on the given CPU. data must already have
* ->func, ->info, and ->flags set.
*/
-static
-void generic_exec_single(int cpu, struct call_single_data *csd, int wait)
+static int generic_exec_single(int cpu, struct call_single_data *csd,
+ smp_call_func_t func, void *info)
{
- struct call_single_queue *dst = &per_cpu(call_single_queue, cpu);
- unsigned long flags;
- int ipi;
+ if (cpu == smp_processor_id()) {
+ unsigned long flags;
+
+ /*
+ * We can unlock early even for the synchronous on-stack case,
+ * since we're doing this from the same CPU..
+ */
+ csd_unlock(csd);
+ local_irq_save(flags);
+ func(info);
+ local_irq_restore(flags);
+ return 0;
+ }
+
+
+ if ((unsigned)cpu >= nr_cpu_ids || !cpu_online(cpu)) {
+ csd_unlock(csd);
+ return -ENXIO;
+ }
- raw_spin_lock_irqsave(&dst->lock, flags);
- ipi = list_empty(&dst->list);
- list_add_tail(&csd->list, &dst->list);
- raw_spin_unlock_irqrestore(&dst->lock, flags);
+ csd->func = func;
+ csd->info = info;
/*
* The list addition should be visible before sending the IPI
* locking and barrier primitives. Generic code isn't really
* equipped to do the right thing...
*/
- if (ipi)
+ if (llist_add(&csd->llist, &per_cpu(call_single_queue, cpu)))
arch_send_call_function_single_ipi(cpu);
- if (wait)
- csd_lock_wait(csd);
+ return 0;
}
-/*
- * Invoked by arch to handle an IPI for call function single. Must be
- * called from the arch with interrupts disabled.
+/**
+ * generic_smp_call_function_single_interrupt - Execute SMP IPI callbacks
+ *
+ * Invoked by arch to handle an IPI for call function single.
+ * Must be called with interrupts disabled.
*/
void generic_smp_call_function_single_interrupt(void)
{
- struct call_single_queue *q = &__get_cpu_var(call_single_queue);
- LIST_HEAD(list);
+ flush_smp_call_function_queue(true);
+}
- /*
- * Shouldn't receive this interrupt on a cpu that is not yet online.
- */
- WARN_ON_ONCE(!cpu_online(smp_processor_id()));
+/**
+ * flush_smp_call_function_queue - Flush pending smp-call-function callbacks
+ *
+ * @warn_cpu_offline: If set to 'true', warn if callbacks were queued on an
+ * offline CPU. Skip this check if set to 'false'.
+ *
+ * Flush any pending smp-call-function callbacks queued on this CPU. This is
+ * invoked by the generic IPI handler, as well as by a CPU about to go offline,
+ * to ensure that all pending IPI callbacks are run before it goes completely
+ * offline.
+ *
+ * Loop through the call_single_queue and run all the queued callbacks.
+ * Must be called with interrupts disabled.
+ */
+static void flush_smp_call_function_queue(bool warn_cpu_offline)
+{
+ struct llist_head *head;
+ struct llist_node *entry;
+ struct call_single_data *csd, *csd_next;
+ static bool warned;
- raw_spin_lock(&q->lock);
- list_replace_init(&q->list, &list);
- raw_spin_unlock(&q->lock);
+ WARN_ON(!irqs_disabled());
- while (!list_empty(&list)) {
- struct call_single_data *csd;
- unsigned int csd_flags;
+ head = this_cpu_ptr(&call_single_queue);
+ entry = llist_del_all(head);
+ entry = llist_reverse_order(entry);
- csd = list_entry(list.next, struct call_single_data, list);
- list_del(&csd->list);
+ /* There shouldn't be any pending callbacks on an offline CPU. */
+ if (unlikely(warn_cpu_offline && !cpu_online(smp_processor_id()) &&
+ !warned && !llist_empty(head))) {
+ warned = true;
+ WARN(1, "IPI on offline CPU %d\n", smp_processor_id());
/*
- * 'csd' can be invalid after this call if flags == 0
- * (when called through generic_exec_single()),
- * so save them away before making the call:
+ * We don't have to use the _safe() variant here
+ * because we are not invoking the IPI handlers yet.
*/
- csd_flags = csd->flags;
+ llist_for_each_entry(csd, entry, llist)
+ pr_warn("IPI callback %pS sent to offline CPU\n",
+ csd->func);
+ }
- csd->func(csd->info);
+ llist_for_each_entry_safe(csd, csd_next, entry, llist) {
+ smp_call_func_t func = csd->func;
+ void *info = csd->info;
- /*
- * Unlocked CSDs are valid through generic_exec_single():
- */
- if (csd_flags & CSD_FLAG_LOCK)
+ /* Do we wait until *after* callback? */
+ if (csd->flags & CSD_FLAG_SYNCHRONOUS) {
+ func(info);
+ csd_unlock(csd);
+ } else {
csd_unlock(csd);
+ func(info);
+ }
}
-}
-static DEFINE_PER_CPU_SHARED_ALIGNED(struct call_single_data, csd_data);
+ /*
+ * Handle irq works queued remotely by irq_work_queue_on().
+ * Smp functions above are typically synchronous so they
+ * better run first since some other CPUs may be busy waiting
+ * for them.
+ */
+ irq_work_run();
+}
/*
* smp_call_function_single - Run a function on a specific CPU
int smp_call_function_single(int cpu, smp_call_func_t func, void *info,
int wait)
{
- struct call_single_data d = {
- .flags = 0,
- };
- unsigned long flags;
+ struct call_single_data *csd;
+ struct call_single_data csd_stack = { .flags = CSD_FLAG_LOCK | CSD_FLAG_SYNCHRONOUS };
int this_cpu;
- int err = 0;
+ int err;
/*
* prevent preemption and reschedule on another processor,
WARN_ON_ONCE(cpu_online(this_cpu) && irqs_disabled()
&& !oops_in_progress);
- if (cpu == this_cpu) {
- local_irq_save(flags);
- func(info);
- local_irq_restore(flags);
- } else {
- if ((unsigned)cpu < nr_cpu_ids && cpu_online(cpu)) {
- struct call_single_data *csd = &d;
-
- if (!wait)
- csd = &__get_cpu_var(csd_data);
+ csd = &csd_stack;
+ if (!wait) {
+ csd = this_cpu_ptr(&csd_data);
+ csd_lock(csd);
+ }
- csd_lock(csd);
+ err = generic_exec_single(cpu, csd, func, info);
- csd->func = func;
- csd->info = info;
- generic_exec_single(cpu, csd, wait);
- } else {
- err = -ENXIO; /* CPU not online */
- }
- }
+ if (wait)
+ csd_lock_wait(csd);
put_cpu();
}
EXPORT_SYMBOL(smp_call_function_single);
+/**
+ * smp_call_function_single_async(): Run an asynchronous function on a
+ * specific CPU.
+ * @cpu: The CPU to run on.
+ * @csd: Pre-allocated and setup data structure
+ *
+ * Like smp_call_function_single(), but the call is asynchonous and
+ * can thus be done from contexts with disabled interrupts.
+ *
+ * The caller passes his own pre-allocated data structure
+ * (ie: embedded in an object) and is responsible for synchronizing it
+ * such that the IPIs performed on the @csd are strictly serialized.
+ *
+ * NOTE: Be careful, there is unfortunately no current debugging facility to
+ * validate the correctness of this serialization.
+ */
+int smp_call_function_single_async(int cpu, struct call_single_data *csd)
+{
+ int err = 0;
+
+ preempt_disable();
+
+ /* We could deadlock if we have to wait here with interrupts disabled! */
+ if (WARN_ON_ONCE(csd->flags & CSD_FLAG_LOCK))
+ csd_lock_wait(csd);
+
+ csd->flags = CSD_FLAG_LOCK;
+ smp_wmb();
+
+ err = generic_exec_single(cpu, csd, csd->func, csd->info);
+ preempt_enable();
+
+ return err;
+}
+EXPORT_SYMBOL_GPL(smp_call_function_single_async);
+
/*
* smp_call_function_any - Run a function on any of the given cpus
* @mask: The mask of cpus it can run on.
* @wait: If true, wait until function has completed.
*
* Returns 0 on success, else a negative status code (if no cpus were online).
- * Note that @wait will be implicitly turned on in case of allocation failures,
- * since we fall back to on-stack allocation.
*
* Selection preference:
* 1) current cpu if in @mask
}
EXPORT_SYMBOL_GPL(smp_call_function_any);
-/**
- * __smp_call_function_single(): Run a function on a specific CPU
- * @cpu: The CPU to run on.
- * @data: Pre-allocated and setup data structure
- * @wait: If true, wait until function has completed on specified CPU.
- *
- * Like smp_call_function_single(), but allow caller to pass in a
- * pre-allocated data structure. Useful for embedding @data inside
- * other structures, for instance.
- */
-void __smp_call_function_single(int cpu, struct call_single_data *csd,
- int wait)
-{
- unsigned int this_cpu;
- unsigned long flags;
-
- this_cpu = get_cpu();
- /*
- * Can deadlock when called with interrupts disabled.
- * We allow cpu's that are not yet online though, as no one else can
- * send smp call function interrupt to this cpu and as such deadlocks
- * can't happen.
- */
- WARN_ON_ONCE(cpu_online(smp_processor_id()) && wait && irqs_disabled()
- && !oops_in_progress);
-
- if (cpu == this_cpu) {
- local_irq_save(flags);
- csd->func(csd->info);
- local_irq_restore(flags);
- } else {
- csd_lock(csd);
- generic_exec_single(cpu, csd, wait);
- }
- put_cpu();
-}
-
/**
* smp_call_function_many(): Run a function on a set of other CPUs.
* @mask: The set of cpus to run on (only runs on online subset).
return;
}
- cfd = &__get_cpu_var(cfd_data);
+ cfd = this_cpu_ptr(&cfd_data);
cpumask_and(cfd->cpumask, mask, cpu_online_mask);
cpumask_clear_cpu(this_cpu, cfd->cpumask);
if (unlikely(!cpumask_weight(cfd->cpumask)))
return;
- /*
- * After we put an entry into the list, cfd->cpumask may be cleared
- * again when another CPU sends another IPI for a SMP function call, so
- * cfd->cpumask will be zero.
- */
- cpumask_copy(cfd->cpumask_ipi, cfd->cpumask);
-
for_each_cpu(cpu, cfd->cpumask) {
struct call_single_data *csd = per_cpu_ptr(cfd->csd, cpu);
- struct call_single_queue *dst =
- &per_cpu(call_single_queue, cpu);
- unsigned long flags;
csd_lock(csd);
+ if (wait)
+ csd->flags |= CSD_FLAG_SYNCHRONOUS;
csd->func = func;
csd->info = info;
-
- raw_spin_lock_irqsave(&dst->lock, flags);
- list_add_tail(&csd->list, &dst->list);
- raw_spin_unlock_irqrestore(&dst->lock, flags);
+ llist_add(&csd->llist, &per_cpu(call_single_queue, cpu));
}
/* Send a message to all CPUs in the map */
- arch_send_call_function_ipi_mask(cfd->cpumask_ipi);
+ arch_send_call_function_ipi_mask(cfd->cpumask);
if (wait) {
for_each_cpu(cpu, cfd->cpumask) {
return 0;
}
EXPORT_SYMBOL(smp_call_function);
-#endif /* USE_GENERIC_SMP_HELPERS */
/* Setup configured maximum number of CPUs to activate */
unsigned int setup_max_cpus = NR_CPUS;
nr_cpu_ids = find_last_bit(cpumask_bits(cpu_possible_mask),NR_CPUS) + 1;
}
+void __weak smp_announce(void)
+{
+ printk(KERN_INFO "Brought up %d CPUs\n", num_online_cpus());
+}
+
/* Called by boot processor to activate the rest. */
void __init smp_init(void)
{
}
/* Any cleanup work */
- printk(KERN_INFO "Brought up %ld CPUs\n", (long)num_online_cpus());
+ smp_announce();
smp_cpus_done(setup_max_cpus);
}
*
* If @wait is true, then returns once @func has returned.
*
- * You must not call this function with disabled interrupts or
- * from a hardware interrupt handler or from a bottom half handler.
+ * You must not call this function with disabled interrupts or from a
+ * hardware interrupt handler or from a bottom half handler. The
+ * exception is that it may be used during early boot while
+ * early_boot_irqs_disabled is set.
*/
void on_each_cpu_mask(const struct cpumask *mask, smp_call_func_t func,
void *info, bool wait)
smp_call_function_many(mask, func, info, wait);
if (cpumask_test_cpu(cpu, mask)) {
- local_irq_disable();
+ unsigned long flags;
+ local_irq_save(flags);
func(info);
- local_irq_enable();
+ local_irq_restore(flags);
}
put_cpu();
}
cpumask_var_t cpus;
int cpu, ret;
- might_sleep_if(gfp_flags & __GFP_WAIT);
+ might_sleep_if(gfpflags_allow_blocking(gfp_flags));
if (likely(zalloc_cpumask_var(&cpus, (gfp_flags|__GFP_NOWARN)))) {
preempt_disable();
if (cond_func(cpu, info)) {
ret = smp_call_function_single(cpu, func,
info, wait);
- WARN_ON_ONCE(!ret);
+ WARN_ON_ONCE(ret);
}
preempt_enable();
}
smp_call_function(do_nothing, NULL, 1);
}
EXPORT_SYMBOL_GPL(kick_all_cpus_sync);
+
+/**
+ * wake_up_all_idle_cpus - break all cpus out of idle
+ * wake_up_all_idle_cpus try to break all cpus which is in idle state even
+ * including idle polling cpus, for non-idle cpus, we will do nothing
+ * for them.
+ */
+void wake_up_all_idle_cpus(void)
+{
+ int cpu;
+
+ preempt_disable();
+ for_each_online_cpu(cpu) {
+ if (cpu == smp_processor_id())
+ continue;
+
+ wake_up_if_idle(cpu);
+ }
+ preempt_enable();
+}
+EXPORT_SYMBOL_GPL(wake_up_all_idle_cpus);