From e162b39a368f0401e41b558f430c354d12a85b37 Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Thu, 15 Jan 2009 11:08:40 -0800 Subject: softlockup: decouple hung tasks check from softlockup detection Decoupling allows: * hung tasks check to happen at very low priority * hung tasks check and softlockup to be enabled/disabled independently at compile and/or run-time * individual panic settings to be enabled disabled independently at compile and/or run-time * softlockup threshold to be reduced without increasing hung tasks poll frequency (hung task check is expensive relative to softlock watchdog) * hung task check to be zero over-head when disabled at run-time Signed-off-by: Mandeep Singh Baines Signed-off-by: Ingo Molnar --- include/linux/sched.h | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) (limited to 'include/linux/sched.h') diff --git a/include/linux/sched.h b/include/linux/sched.h index 54cbabf3b87..f2f94d53230 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -297,9 +297,6 @@ extern int proc_dosoftlockup_thresh(struct ctl_table *table, int write, struct file *filp, void __user *buffer, size_t *lenp, loff_t *ppos); extern unsigned int softlockup_panic; -extern unsigned long sysctl_hung_task_check_count; -extern unsigned long sysctl_hung_task_timeout_secs; -extern unsigned long sysctl_hung_task_warnings; extern int softlockup_thresh; #else static inline void softlockup_tick(void) @@ -316,6 +313,15 @@ static inline void touch_all_softlockup_watchdogs(void) } #endif +#ifdef CONFIG_DETECT_HUNG_TASK +extern unsigned int sysctl_hung_task_panic; +extern unsigned long sysctl_hung_task_check_count; +extern unsigned long sysctl_hung_task_timeout_secs; +extern unsigned long sysctl_hung_task_warnings; +extern int proc_dohung_task_timeout_secs(struct ctl_table *table, int write, + struct file *filp, void __user *buffer, + size_t *lenp, loff_t *ppos); +#endif /* Attach to any functions which should be ignored in wchan output. */ #define __sched __attribute__((__section__(".sched.text"))) @@ -1236,7 +1242,7 @@ struct task_struct { /* ipc stuff */ struct sysv_sem sysvsem; #endif -#ifdef CONFIG_DETECT_SOFTLOCKUP +#ifdef CONFIG_DETECT_HUNG_TASK /* hung task detection */ unsigned long last_switch_timestamp; unsigned long last_switch_count; -- cgit v1.2.3-70-g09d2 From 5e54f5986a579b8445aa1d5ad3435c2cf7568bed Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Fri, 30 Jan 2009 15:29:54 -0800 Subject: softlockup: remove unused definition for spawn_softlockup_task The definition of spawn_softlockup_task in sched.h became unnecessary once it was converted to the early_initcall() interface. Signed-off-by: Mandeep Singh Baines Signed-off-by: Ingo Molnar --- include/linux/sched.h | 3 --- 1 file changed, 3 deletions(-) (limited to 'include/linux/sched.h') diff --git a/include/linux/sched.h b/include/linux/sched.h index f2f94d53230..2a2811c6239 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -302,9 +302,6 @@ extern int softlockup_thresh; static inline void softlockup_tick(void) { } -static inline void spawn_softlockup_task(void) -{ -} static inline void touch_softlockup_watchdog(void) { } -- cgit v1.2.3-70-g09d2 From 17406b82d621930cca8ccc1272cdac9a7dae8e40 Mon Sep 17 00:00:00 2001 From: Mandeep Singh Baines Date: Fri, 6 Feb 2009 15:37:47 -0800 Subject: softlockup: remove timestamp checking from hung_task Impact: saves sizeof(long) bytes per task_struct By guaranteeing that sysctl_hung_task_timeout_secs have elapsed between tasklist scans we can avoid using timestamps. Signed-off-by: Mandeep Singh Baines Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 - kernel/fork.c | 8 +++----- kernel/hung_task.c | 48 +++++++++--------------------------------------- 3 files changed, 12 insertions(+), 45 deletions(-) (limited to 'include/linux/sched.h') diff --git a/include/linux/sched.h b/include/linux/sched.h index 2a2811c6239..e0d723fea9f 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1241,7 +1241,6 @@ struct task_struct { #endif #ifdef CONFIG_DETECT_HUNG_TASK /* hung task detection */ - unsigned long last_switch_timestamp; unsigned long last_switch_count; #endif /* CPU-specific state of this task */ diff --git a/kernel/fork.c b/kernel/fork.c index fb944428283..bf582f75014 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -639,6 +639,9 @@ static int copy_mm(unsigned long clone_flags, struct task_struct * tsk) tsk->min_flt = tsk->maj_flt = 0; tsk->nvcsw = tsk->nivcsw = 0; +#ifdef CONFIG_DETECT_HUNG_TASK + tsk->last_switch_count = tsk->nvcsw + tsk->nivcsw; +#endif tsk->mm = NULL; tsk->active_mm = NULL; @@ -1041,11 +1044,6 @@ static struct task_struct *copy_process(unsigned long clone_flags, p->default_timer_slack_ns = current->timer_slack_ns; -#ifdef CONFIG_DETECT_HUNG_TASK - p->last_switch_count = 0; - p->last_switch_timestamp = 0; -#endif - task_io_accounting_init(&p->ioac); acct_clear_integrals(p); diff --git a/kernel/hung_task.c b/kernel/hung_task.c index 3951a80e7cb..0c924de58cb 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -34,7 +34,6 @@ unsigned long __read_mostly sysctl_hung_task_check_count = PID_MAX_LIMIT; * Zero means infinite timeout - no checking done: */ unsigned long __read_mostly sysctl_hung_task_timeout_secs = 120; -static unsigned long __read_mostly hung_task_poll_jiffies; unsigned long __read_mostly sysctl_hung_task_warnings = 10; @@ -69,33 +68,17 @@ static struct notifier_block panic_block = { .notifier_call = hung_task_panic, }; -/* - * Returns seconds, approximately. We don't need nanosecond - * resolution, and we don't need to waste time with a big divide when - * 2^30ns == 1.074s. - */ -static unsigned long get_timestamp(void) -{ - int this_cpu = raw_smp_processor_id(); - - return cpu_clock(this_cpu) >> 30LL; /* 2^30 ~= 10^9 */ -} - -static void check_hung_task(struct task_struct *t, unsigned long now, - unsigned long timeout) +static void check_hung_task(struct task_struct *t, unsigned long timeout) { 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) { + if (switch_count != t->last_switch_count) { t->last_switch_count = switch_count; - t->last_switch_timestamp = now; return; } - if ((long)(now - t->last_switch_timestamp) < timeout) - return; if (!sysctl_hung_task_warnings) return; sysctl_hung_task_warnings--; @@ -111,7 +94,6 @@ static void check_hung_task(struct task_struct *t, unsigned long now, sched_show_task(t); __debug_show_held_locks(t); - t->last_switch_timestamp = now; touch_nmi_watchdog(); if (sysctl_hung_task_panic) @@ -145,7 +127,6 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout) { int max_count = sysctl_hung_task_check_count; int batch_count = HUNG_TASK_BATCHING; - unsigned long now = get_timestamp(); struct task_struct *g, *t; /* @@ -168,19 +149,16 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout) } /* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */ if (t->state == TASK_UNINTERRUPTIBLE) - check_hung_task(t, now, timeout); + check_hung_task(t, timeout); } while_each_thread(g, t); unlock: rcu_read_unlock(); } -static void update_poll_jiffies(void) +static unsigned long timeout_jiffies(unsigned long timeout) { /* timeout of 0 will disable the watchdog */ - if (sysctl_hung_task_timeout_secs == 0) - hung_task_poll_jiffies = MAX_SCHEDULE_TIMEOUT; - else - hung_task_poll_jiffies = sysctl_hung_task_timeout_secs * HZ / 2; + return timeout ? timeout * HZ : MAX_SCHEDULE_TIMEOUT; } /* @@ -197,8 +175,6 @@ int proc_dohung_task_timeout_secs(struct ctl_table *table, int write, if (ret || !write) goto out; - update_poll_jiffies(); - wake_up_process(watchdog_task); out: @@ -211,20 +187,14 @@ int proc_dohung_task_timeout_secs(struct ctl_table *table, int write, static int watchdog(void *dummy) { set_user_nice(current, 0); - update_poll_jiffies(); for ( ; ; ) { - unsigned long timeout; + unsigned long timeout = sysctl_hung_task_timeout_secs; - while (schedule_timeout_interruptible(hung_task_poll_jiffies)); + while (schedule_timeout_interruptible(timeout_jiffies(timeout))) + timeout = sysctl_hung_task_timeout_secs; - /* - * Need to cache timeout here to avoid timeout being set - * to 0 via sysctl while inside check_hung_*_tasks(). - */ - timeout = sysctl_hung_task_timeout_secs; - if (timeout) - check_hung_uninterruptible_tasks(timeout); + check_hung_uninterruptible_tasks(timeout); } return 0; -- cgit v1.2.3-70-g09d2 From 3aa551c9b4c40018f0e261a178e3d25478dc04a9 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 23 Mar 2009 18:28:15 +0100 Subject: genirq: add threaded interrupt handler support Add support for threaded interrupt handlers: A device driver can request that its main interrupt handler runs in a thread. To achive this the device driver requests the interrupt with request_threaded_irq() and provides additionally to the handler a thread function. The handler function is called in hard interrupt context and needs to check whether the interrupt originated from the device. If the interrupt originated from the device then the handler can either return IRQ_HANDLED or IRQ_WAKE_THREAD. IRQ_HANDLED is returned when no further action is required. IRQ_WAKE_THREAD causes the genirq code to invoke the threaded (main) handler. When IRQ_WAKE_THREAD is returned handler must have disabled the interrupt on the device level. This is mandatory for shared interrupt handlers, but we need to do it as well for obscure x86 hardware where disabling an interrupt on the IO_APIC level redirects the interrupt to the legacy PIC interrupt lines. Signed-off-by: Thomas Gleixner Reviewed-by: Ingo Molnar --- include/linux/hardirq.h | 2 +- include/linux/interrupt.h | 37 ++++++++- include/linux/irq.h | 5 ++ include/linux/irqreturn.h | 2 + include/linux/sched.h | 5 ++ kernel/exit.c | 2 + kernel/irq/handle.c | 31 +++++++- kernel/irq/manage.c | 192 ++++++++++++++++++++++++++++++++++++++++++---- 8 files changed, 259 insertions(+), 17 deletions(-) (limited to 'include/linux/sched.h') diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h index f83288347dd..2dfaadbdb2a 100644 --- a/include/linux/hardirq.h +++ b/include/linux/hardirq.h @@ -105,7 +105,7 @@ # define IRQ_EXIT_OFFSET HARDIRQ_OFFSET #endif -#ifdef CONFIG_SMP +#if defined(CONFIG_SMP) || defined(CONFIG_GENERIC_HARDIRQS) extern void synchronize_irq(unsigned int irq); #else # define synchronize_irq(irq) barrier() diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index 0c9cb63e689..6fc2b720c23 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -59,6 +59,16 @@ #define IRQF_NOBALANCING 0x00000800 #define IRQF_IRQPOLL 0x00001000 +/* + * Bits used by threaded handlers: + * IRQTF_RUNTHREAD - signals that the interrupt handler thread should run + * IRQTF_DIED - handler thread died + */ +enum { + IRQTF_RUNTHREAD, + IRQTF_DIED, +}; + typedef irqreturn_t (*irq_handler_t)(int, void *); /** @@ -71,6 +81,9 @@ typedef irqreturn_t (*irq_handler_t)(int, void *); * @next: pointer to the next irqaction for shared interrupts * @irq: interrupt number * @dir: pointer to the proc/irq/NN/name entry + * @thread_fn: interupt handler function for threaded interrupts + * @thread: thread pointer for threaded interrupts + * @thread_flags: flags related to @thread */ struct irqaction { irq_handler_t handler; @@ -81,11 +94,31 @@ struct irqaction { struct irqaction *next; int irq; struct proc_dir_entry *dir; + irq_handler_t thread_fn; + struct task_struct *thread; + unsigned long thread_flags; }; extern irqreturn_t no_action(int cpl, void *dev_id); -extern int __must_check request_irq(unsigned int, irq_handler_t handler, - unsigned long, const char *, void *); + +extern int __must_check +request_threaded_irq(unsigned int irq, irq_handler_t handler, + irq_handler_t thread_fn, + unsigned long flags, const char *name, void *dev); + +static inline int __must_check +request_irq(unsigned int irq, irq_handler_t handler, unsigned long flags, + const char *name, void *dev) +{ + return request_threaded_irq(irq, handler, NULL, flags, name, dev); +} + +#ifdef CONFIG_GENERIC_HARDIRQS +extern void exit_irq_thread(void); +#else +static inline void exit_irq_thread(void) { } +#endif + extern void free_irq(unsigned int, void *); struct device; diff --git a/include/linux/irq.h b/include/linux/irq.h index 873e4ac11b8..8b1cf063021 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -20,6 +20,7 @@ #include #include #include +#include #include #include @@ -155,6 +156,8 @@ struct irq_2_iommu; * @affinity: IRQ affinity on SMP * @cpu: cpu index useful for balancing * @pending_mask: pending rebalanced interrupts + * @threads_active: number of irqaction threads currently running + * @wait_for_threads: wait queue for sync_irq to wait for threaded handlers * @dir: /proc/irq/ procfs entry * @name: flow handler name for /proc/interrupts output */ @@ -186,6 +189,8 @@ struct irq_desc { cpumask_var_t pending_mask; #endif #endif + atomic_t threads_active; + wait_queue_head_t wait_for_threads; #ifdef CONFIG_PROC_FS struct proc_dir_entry *dir; #endif diff --git a/include/linux/irqreturn.h b/include/linux/irqreturn.h index c5584ca5b8c..819acaaac3f 100644 --- a/include/linux/irqreturn.h +++ b/include/linux/irqreturn.h @@ -5,10 +5,12 @@ * enum irqreturn * @IRQ_NONE interrupt was not from this device * @IRQ_HANDLED interrupt was handled by this device + * @IRQ_WAKE_THREAD handler requests to wake the handler thread */ enum irqreturn { IRQ_NONE, IRQ_HANDLED, + IRQ_WAKE_THREAD, }; typedef enum irqreturn irqreturn_t; diff --git a/include/linux/sched.h b/include/linux/sched.h index 46d680643f8..38b77b0f56e 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1292,6 +1292,11 @@ struct task_struct { /* Protection of (de-)allocation: mm, files, fs, tty, keyrings */ spinlock_t alloc_lock; +#ifdef CONFIG_GENERIC_HARDIRQS + /* IRQ handler threads */ + struct irqaction *irqaction; +#endif + /* Protection of the PI data structures: */ spinlock_t pi_lock; diff --git a/kernel/exit.c b/kernel/exit.c index 167e1e3ad7c..ca0b3488c4a 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -1037,6 +1037,8 @@ NORET_TYPE void do_exit(long code) schedule(); } + exit_irq_thread(); + exit_signals(tsk); /* sets PF_EXITING */ /* * tsk->flags are checked in the futex code to protect against diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 9ebf7796887..fe8f45374e8 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -357,8 +357,37 @@ irqreturn_t handle_IRQ_event(unsigned int irq, struct irqaction *action) do { ret = action->handler(irq, action->dev_id); - if (ret == IRQ_HANDLED) + + switch (ret) { + case IRQ_WAKE_THREAD: + /* + * Wake up the handler thread for this + * action. In case the thread crashed and was + * killed we just pretend that we handled the + * interrupt. The hardirq handler above has + * disabled the device interrupt, so no irq + * storm is lurking. + */ + if (likely(!test_bit(IRQTF_DIED, + &action->thread_flags))) { + set_bit(IRQTF_RUNTHREAD, &action->thread_flags); + wake_up_process(action->thread); + } + + /* + * Set it to handled so the spurious check + * does not trigger. + */ + ret = IRQ_HANDLED; + /* Fall through to add to randomness */ + case IRQ_HANDLED: status |= action->flags; + break; + + default: + break; + } + retval |= ret; action = action->next; } while (action); diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 6458e99984c..a4c1ab86cd2 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -8,16 +8,15 @@ */ #include +#include #include #include #include #include +#include #include "internals.h" -#if defined(CONFIG_SMP) && defined(CONFIG_GENERIC_HARDIRQS) -cpumask_var_t irq_default_affinity; - /** * synchronize_irq - wait for pending IRQ handlers (on other CPUs) * @irq: interrupt number to wait for @@ -53,9 +52,18 @@ void synchronize_irq(unsigned int irq) /* Oops, that failed? */ } while (status & IRQ_INPROGRESS); + + /* + * We made sure that no hardirq handler is running. Now verify + * that no threaded handlers are active. + */ + wait_event(desc->wait_for_threads, !atomic_read(&desc->threads_active)); } EXPORT_SYMBOL(synchronize_irq); +#ifdef CONFIG_SMP +cpumask_var_t irq_default_affinity; + /** * irq_can_set_affinity - Check if the affinity of a given irq can be set * @irq: Interrupt to check @@ -72,6 +80,18 @@ int irq_can_set_affinity(unsigned int irq) return 1; } +static void +irq_set_thread_affinity(struct irq_desc *desc, const struct cpumask *cpumask) +{ + struct irqaction *action = desc->action; + + while (action) { + if (action->thread) + set_cpus_allowed_ptr(action->thread, cpumask); + action = action->next; + } +} + /** * irq_set_affinity - Set the irq affinity of a given irq * @irq: Interrupt to set affinity @@ -100,6 +120,7 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask) cpumask_copy(desc->affinity, cpumask); desc->chip->set_affinity(irq, cpumask); #endif + irq_set_thread_affinity(desc, cpumask); desc->status |= IRQ_AFFINITY_SET; spin_unlock_irqrestore(&desc->lock, flags); return 0; @@ -150,6 +171,8 @@ int irq_select_affinity_usr(unsigned int irq) spin_lock_irqsave(&desc->lock, flags); ret = setup_affinity(irq, desc); + if (!ret) + irq_set_thread_affinity(desc, desc->affinity); spin_unlock_irqrestore(&desc->lock, flags); return ret; @@ -384,6 +407,93 @@ int __irq_set_trigger(struct irq_desc *desc, unsigned int irq, return ret; } +static inline int irq_thread_should_run(struct irqaction *action) +{ + return test_and_clear_bit(IRQTF_RUNTHREAD, &action->thread_flags); +} + +static int irq_wait_for_interrupt(struct irqaction *action) +{ + while (!kthread_should_stop()) { + set_current_state(TASK_INTERRUPTIBLE); + if (irq_thread_should_run(action)) { + __set_current_state(TASK_RUNNING); + return 0; + } else + schedule(); + } + return -1; +} + +/* + * Interrupt handler thread + */ +static int irq_thread(void *data) +{ + struct sched_param param = { .sched_priority = MAX_USER_RT_PRIO/2, }; + struct irqaction *action = data; + struct irq_desc *desc = irq_to_desc(action->irq); + int wake; + + sched_setscheduler(current, SCHED_FIFO, ¶m); + current->irqaction = action; + + while (!irq_wait_for_interrupt(action)) { + + atomic_inc(&desc->threads_active); + + spin_lock_irq(&desc->lock); + if (unlikely(desc->status & IRQ_DISABLED)) { + /* + * CHECKME: We might need a dedicated + * IRQ_THREAD_PENDING flag here, which + * retriggers the thread in check_irq_resend() + * but AFAICT IRQ_PENDING should be fine as it + * retriggers the interrupt itself --- tglx + */ + desc->status |= IRQ_PENDING; + spin_unlock_irq(&desc->lock); + } else { + spin_unlock_irq(&desc->lock); + + action->thread_fn(action->irq, action->dev_id); + } + + wake = atomic_dec_and_test(&desc->threads_active); + + if (wake && waitqueue_active(&desc->wait_for_threads)) + wake_up(&desc->wait_for_threads); + } + + /* + * Clear irqaction. Otherwise exit_irq_thread() would make + * fuzz about an active irq thread going into nirvana. + */ + current->irqaction = NULL; + return 0; +} + +/* + * Called from do_exit() + */ +void exit_irq_thread(void) +{ + struct task_struct *tsk = current; + + if (!tsk->irqaction) + return; + + printk(KERN_ERR + "exiting task \"%s\" (%d) is an active IRQ thread (irq %d)\n", + tsk->comm ? tsk->comm : "", tsk->pid, tsk->irqaction->irq); + + /* + * Set the THREAD DIED flag to prevent further wakeups of the + * soon to be gone threaded handler. + */ + set_bit(IRQTF_DIED, &tsk->irqaction->flags); +} + /* * Internal function to register an irqaction - typically used to * allocate special interrupts that are part of the architecture. @@ -419,6 +529,26 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) rand_initialize_irq(irq); } + /* + * Threaded handler ? + */ + if (new->thread_fn) { + struct task_struct *t; + + t = kthread_create(irq_thread, new, "irq/%d-%s", irq, + new->name); + if (IS_ERR(t)) + return PTR_ERR(t); + /* + * We keep the reference to the task struct even if + * the thread dies to avoid that the interrupt code + * references an already freed task_struct. + */ + get_task_struct(t); + new->thread = t; + wake_up_process(t); + } + /* * The following block of code has to be executed atomically */ @@ -456,15 +586,15 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new) if (!shared) { irq_chip_set_defaults(desc->chip); + init_waitqueue_head(&desc->wait_for_threads); + /* Setup the type (level, edge polarity) if configured: */ if (new->flags & IRQF_TRIGGER_MASK) { ret = __irq_set_trigger(desc, irq, new->flags & IRQF_TRIGGER_MASK); - if (ret) { - spin_unlock_irqrestore(&desc->lock, flags); - return ret; - } + if (ret) + goto out_thread; } else compat_irq_chip_set_default_handler(desc); #if defined(CONFIG_IRQ_PER_CPU) @@ -532,8 +662,19 @@ mismatch: dump_stack(); } #endif + ret = -EBUSY; + +out_thread: spin_unlock_irqrestore(&desc->lock, flags); - return -EBUSY; + if (new->thread) { + struct task_struct *t = new->thread; + + new->thread = NULL; + if (likely(!test_bit(IRQTF_DIED, &new->thread_flags))) + kthread_stop(t); + put_task_struct(t); + } + return ret; } /** @@ -559,6 +700,7 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) { struct irq_desc *desc = irq_to_desc(irq); struct irqaction *action, **action_ptr; + struct task_struct *irqthread; unsigned long flags; WARN(in_interrupt(), "Trying to free IRQ %d from IRQ context!\n", irq); @@ -605,6 +747,10 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) else desc->chip->disable(irq); } + + irqthread = action->thread; + action->thread = NULL; + spin_unlock_irqrestore(&desc->lock, flags); unregister_handler_proc(irq, action); @@ -612,6 +758,12 @@ static struct irqaction *__free_irq(unsigned int irq, void *dev_id) /* Make sure it's not being used on another CPU: */ synchronize_irq(irq); + if (irqthread) { + if (!test_bit(IRQTF_DIED, &action->thread_flags)) + kthread_stop(irqthread); + put_task_struct(irqthread); + } + #ifdef CONFIG_DEBUG_SHIRQ /* * It's a shared IRQ -- the driver ought to be prepared for an IRQ @@ -664,9 +816,12 @@ void free_irq(unsigned int irq, void *dev_id) EXPORT_SYMBOL(free_irq); /** - * request_irq - allocate an interrupt line + * request_threaded_irq - allocate an interrupt line * @irq: Interrupt line to allocate - * @handler: Function to be called when the IRQ occurs + * @handler: Function to be called when the IRQ occurs. + * Primary handler for threaded interrupts + * @thread_fn: Function called from the irq handler thread + * If NULL, no irq thread is created * @irqflags: Interrupt type flags * @devname: An ascii name for the claiming device * @dev_id: A cookie passed back to the handler function @@ -678,6 +833,15 @@ EXPORT_SYMBOL(free_irq); * raises, you must take care both to initialise your hardware * and to set up the interrupt handler in the right order. * + * If you want to set up a threaded irq handler for your device + * then you need to supply @handler and @thread_fn. @handler ist + * still called in hard interrupt context and has to check + * whether the interrupt originates from the device. If yes it + * needs to disable the interrupt on the device and return + * IRQ_THREAD_WAKE which will wake up the handler thread and run + * @thread_fn. This split handler design is necessary to support + * shared interrupts. + * * Dev_id must be globally unique. Normally the address of the * device data structure is used as the cookie. Since the handler * receives this value it makes sense to use it. @@ -693,8 +857,9 @@ EXPORT_SYMBOL(free_irq); * IRQF_TRIGGER_* Specify active edge(s) or level * */ -int request_irq(unsigned int irq, irq_handler_t handler, - unsigned long irqflags, const char *devname, void *dev_id) +int request_threaded_irq(unsigned int irq, irq_handler_t handler, + irq_handler_t thread_fn, unsigned long irqflags, + const char *devname, void *dev_id) { struct irqaction *action; struct irq_desc *desc; @@ -742,6 +907,7 @@ int request_irq(unsigned int irq, irq_handler_t handler, return -ENOMEM; action->handler = handler; + action->thread_fn = thread_fn; action->flags = irqflags; action->name = devname; action->dev_id = dev_id; @@ -771,4 +937,4 @@ int request_irq(unsigned int irq, irq_handler_t handler, #endif return retval; } -EXPORT_SYMBOL(request_irq); +EXPORT_SYMBOL(request_threaded_irq); -- cgit v1.2.3-70-g09d2 From e3c8ca8336707062f3f7cb1cd7e6b3c753baccdd Mon Sep 17 00:00:00 2001 From: Nathan Lynch Date: Wed, 8 Apr 2009 19:45:12 -0500 Subject: sched: do not count frozen tasks toward load Freezing tasks via the cgroup freezer causes the load average to climb because the freezer's current implementation puts frozen tasks in uninterruptible sleep (D state). Some applications which perform job-scheduling functions consult the load average when making decisions. If a cgroup is frozen, the load average does not provide a useful measure of the system's utilization to such applications. This is especially inconvenient if the job scheduler employs the cgroup freezer as a mechanism for preempting low priority jobs. Contrast this with using SIGSTOP for the same purpose: the stopped tasks do not count toward system load. Change task_contributes_to_load() to return false if the task is frozen. This results in /proc/loadavg behavior that better meets users' expectations. Signed-off-by: Nathan Lynch Acked-by: Andrew Morton Acked-by: Nigel Cunningham Tested-by: Nigel Cunningham Cc: Cc: containers@lists.linux-foundation.org Cc: linux-pm@lists.linux-foundation.org Cc: Matt Helsley LKML-Reference: <20090408194512.47a99b95@manatee.lan> Signed-off-by: Ingo Molnar --- include/linux/sched.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux/sched.h') diff --git a/include/linux/sched.h b/include/linux/sched.h index 98e1fe51601..b4c38bc8049 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -205,7 +205,8 @@ extern unsigned long long time_sync_thresh; #define task_is_stopped_or_traced(task) \ ((task->state & (__TASK_STOPPED | __TASK_TRACED)) != 0) #define task_contributes_to_load(task) \ - ((task->state & TASK_UNINTERRUPTIBLE) != 0) + ((task->state & TASK_UNINTERRUPTIBLE) != 0 && \ + (task->flags & PF_FROZEN) == 0) #define __set_task_state(tsk, state_value) \ do { (tsk)->state = (state_value); } while (0) -- cgit v1.2.3-70-g09d2