linux/kernel/irq_work.c

328 lines
7.7 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* Copyright (C) 2010 Red Hat, Inc., Peter Zijlstra
*
* Provides a framework for enqueueing and running callbacks from hardirq
* context. The enqueueing is NMI-safe.
*/
#include <linux/bug.h>
#include <linux/kernel.h>
#include <linux/export.h>
#include <linux/irq_work.h>
#include <linux/percpu.h>
#include <linux/hardirq.h>
#include <linux/irqflags.h>
#include <linux/sched.h>
#include <linux/tick.h>
#include <linux/cpu.h>
#include <linux/notifier.h>
#include <linux/smp.h>
irq_work: Handle some irq_work in a per-CPU thread on PREEMPT_RT The irq_work callback is invoked in hard IRQ context. By default all callbacks are scheduled for invocation right away (given supported by the architecture) except for the ones marked IRQ_WORK_LAZY which are delayed until the next timer-tick. While looking over the callbacks, some of them may acquire locks (spinlock_t, rwlock_t) which are transformed into sleeping locks on PREEMPT_RT and must not be acquired in hard IRQ context. Changing the locks into locks which could be acquired in this context will lead to other problems such as increased latencies if everything in the chain has IRQ-off locks. This will not solve all the issues as one callback has been noticed which invoked kref_put() and its callback invokes kfree() and this can not be invoked in hardirq context. Some callbacks are required to be invoked in hardirq context even on PREEMPT_RT to work properly. This includes for instance the NO_HZ callback which needs to be able to observe the idle context. The callbacks which require to be run in hardirq have already been marked. Use this information to split the callbacks onto the two lists on PREEMPT_RT: - lazy_list Work items which are not marked with IRQ_WORK_HARD_IRQ will be added to this list. Callbacks on this list will be invoked from a per-CPU thread. The handler here may acquire sleeping locks such as spinlock_t and invoke kfree(). - raised_list Work items which are marked with IRQ_WORK_HARD_IRQ will be added to this list. They will be invoked in hardirq context and must not acquire any sleeping locks. The wake up of the per-CPU thread occurs from irq_work handler/ hardirq context. The thread runs with lowest RT priority to ensure it runs before any SCHED_OTHER tasks do. [bigeasy: melt tglx's irq_work_tick_soft() which splits irq_work_tick() into a hard and soft variant. Collected fixes over time from Steven Rostedt and Mike Galbraith. Move to per-CPU threads instead of softirq as suggested by PeterZ.] Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20211007092646.uhshe3ut2wkrcfzv@linutronix.de
2021-10-07 09:26:46 +00:00
#include <linux/smpboot.h>
#include <asm/processor.h>
#include <linux/kasan.h>
#include <trace/events/ipi.h>
irq_work: Split raised and lazy lists An irq work can be handled from two places: from the tick if the work carries the "lazy" flag and the tick is periodic, or from a self IPI. We merge all these works in a single list and we use some per cpu latch to avoid raising a self-IPI when one is already pending. Now we could do away with this ugly latch if only the list was only made of non-lazy works. Just enqueueing a work on the empty list would be enough to know if we need to raise an IPI or not. Also we are going to implement remote irq work queuing. Then the per CPU latch will need to become atomic in the global scope. That's too bad because, here as well, just enqueueing a work on an empty list of non-lazy works would be enough to know if we need to raise an IPI or not. So lets take a way out of this: split the works in two distinct lists, one for the works that can be handled by the next tick and another one for those handled by the IPI. Just checking if the latter is empty when we queue a new work is enough to know if we need to raise an IPI. Suggested-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Peter Zijlstra <peterz@infradead.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Kevin Hilman <khilman@linaro.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2014-05-23 16:10:21 +00:00
static DEFINE_PER_CPU(struct llist_head, raised_list);
static DEFINE_PER_CPU(struct llist_head, lazy_list);
irq_work: Handle some irq_work in a per-CPU thread on PREEMPT_RT The irq_work callback is invoked in hard IRQ context. By default all callbacks are scheduled for invocation right away (given supported by the architecture) except for the ones marked IRQ_WORK_LAZY which are delayed until the next timer-tick. While looking over the callbacks, some of them may acquire locks (spinlock_t, rwlock_t) which are transformed into sleeping locks on PREEMPT_RT and must not be acquired in hard IRQ context. Changing the locks into locks which could be acquired in this context will lead to other problems such as increased latencies if everything in the chain has IRQ-off locks. This will not solve all the issues as one callback has been noticed which invoked kref_put() and its callback invokes kfree() and this can not be invoked in hardirq context. Some callbacks are required to be invoked in hardirq context even on PREEMPT_RT to work properly. This includes for instance the NO_HZ callback which needs to be able to observe the idle context. The callbacks which require to be run in hardirq have already been marked. Use this information to split the callbacks onto the two lists on PREEMPT_RT: - lazy_list Work items which are not marked with IRQ_WORK_HARD_IRQ will be added to this list. Callbacks on this list will be invoked from a per-CPU thread. The handler here may acquire sleeping locks such as spinlock_t and invoke kfree(). - raised_list Work items which are marked with IRQ_WORK_HARD_IRQ will be added to this list. They will be invoked in hardirq context and must not acquire any sleeping locks. The wake up of the per-CPU thread occurs from irq_work handler/ hardirq context. The thread runs with lowest RT priority to ensure it runs before any SCHED_OTHER tasks do. [bigeasy: melt tglx's irq_work_tick_soft() which splits irq_work_tick() into a hard and soft variant. Collected fixes over time from Steven Rostedt and Mike Galbraith. Move to per-CPU threads instead of softirq as suggested by PeterZ.] Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20211007092646.uhshe3ut2wkrcfzv@linutronix.de
2021-10-07 09:26:46 +00:00
static DEFINE_PER_CPU(struct task_struct *, irq_workd);
static void wake_irq_workd(void)
{
struct task_struct *tsk = __this_cpu_read(irq_workd);
if (!llist_empty(this_cpu_ptr(&lazy_list)) && tsk)
wake_up_process(tsk);
}
#ifdef CONFIG_SMP
static void irq_work_wake(struct irq_work *entry)
{
wake_irq_workd();
}
static DEFINE_PER_CPU(struct irq_work, irq_work_wakeup) =
IRQ_WORK_INIT_HARD(irq_work_wake);
#endif
static int irq_workd_should_run(unsigned int cpu)
{
return !llist_empty(this_cpu_ptr(&lazy_list));
}
/*
* Claim the entry so that no one else will poke at it.
*/
static bool irq_work_claim(struct irq_work *work)
{
int oflags;
oflags = atomic_fetch_or(IRQ_WORK_CLAIMED | CSD_TYPE_IRQ_WORK, &work->node.a_flags);
irq_work: Fix racy check on work pending flag Work claiming wants to be SMP-safe. And by the time we try to claim a work, if it is already executing concurrently on another CPU, we want to succeed the claiming and queue the work again because the other CPU may have missed the data we wanted to handle in our work if it's about to complete there. This scenario is summarized below: CPU 1 CPU 2 ----- ----- (flags = 0) cmpxchg(flags, 0, IRQ_WORK_FLAGS) (flags = 3) [...] xchg(flags, IRQ_WORK_BUSY) (flags = 2) func() if (flags & IRQ_WORK_PENDING) (not true) cmpxchg(flags, flags, IRQ_WORK_FLAGS) (flags = 3) [...] cmpxchg(flags, IRQ_WORK_BUSY, 0); (fail, pending on CPU 2) This state machine is synchronized using [cmp]xchg() on the flags. As such, the early IRQ_WORK_PENDING check in CPU 2 above is racy. By the time we check it, we may be dealing with a stale value because we aren't using an atomic accessor. As a result, CPU 2 may "see" that the work is still pending on another CPU while it may be actually completing the work function exection already, leaving our data unprocessed. To fix this, we start by speculating about the value we wish to be in the work->flags but we only make any conclusion after the value returned by the cmpxchg() call that either claims the work or let the current owner handle the pending work for us. Changelog-heavily-inspired-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Acked-by: Steven Rostedt <rostedt@goodmis.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Paul Gortmaker <paul.gortmaker@windriver.com> Cc: Anish Kumar <anish198519851985@gmail.com>
2012-10-27 13:21:36 +00:00
/*
* If the work is already pending, no need to raise the IPI.
* The pairing smp_mb() in irq_work_single() makes sure
* everything we did before is visible.
irq_work: Fix racy check on work pending flag Work claiming wants to be SMP-safe. And by the time we try to claim a work, if it is already executing concurrently on another CPU, we want to succeed the claiming and queue the work again because the other CPU may have missed the data we wanted to handle in our work if it's about to complete there. This scenario is summarized below: CPU 1 CPU 2 ----- ----- (flags = 0) cmpxchg(flags, 0, IRQ_WORK_FLAGS) (flags = 3) [...] xchg(flags, IRQ_WORK_BUSY) (flags = 2) func() if (flags & IRQ_WORK_PENDING) (not true) cmpxchg(flags, flags, IRQ_WORK_FLAGS) (flags = 3) [...] cmpxchg(flags, IRQ_WORK_BUSY, 0); (fail, pending on CPU 2) This state machine is synchronized using [cmp]xchg() on the flags. As such, the early IRQ_WORK_PENDING check in CPU 2 above is racy. By the time we check it, we may be dealing with a stale value because we aren't using an atomic accessor. As a result, CPU 2 may "see" that the work is still pending on another CPU while it may be actually completing the work function exection already, leaving our data unprocessed. To fix this, we start by speculating about the value we wish to be in the work->flags but we only make any conclusion after the value returned by the cmpxchg() call that either claims the work or let the current owner handle the pending work for us. Changelog-heavily-inspired-by: Steven Rostedt <rostedt@goodmis.org> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Acked-by: Steven Rostedt <rostedt@goodmis.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Paul Gortmaker <paul.gortmaker@windriver.com> Cc: Anish Kumar <anish198519851985@gmail.com>
2012-10-27 13:21:36 +00:00
*/
if (oflags & IRQ_WORK_PENDING)
return false;
return true;
}
void __weak arch_irq_work_raise(void)
{
/*
* Lame architectures will get the timer tick callback
*/
}
static __always_inline void irq_work_raise(struct irq_work *work)
{
if (trace_ipi_send_cpu_enabled() && arch_irq_work_has_interrupt())
trace_ipi_send_cpu(smp_processor_id(), _RET_IP_, work->func);
arch_irq_work_raise();
}
/* Enqueue on current CPU, work must already be claimed and preempt disabled */
static void __irq_work_queue_local(struct irq_work *work)
{
irq_work: Handle some irq_work in a per-CPU thread on PREEMPT_RT The irq_work callback is invoked in hard IRQ context. By default all callbacks are scheduled for invocation right away (given supported by the architecture) except for the ones marked IRQ_WORK_LAZY which are delayed until the next timer-tick. While looking over the callbacks, some of them may acquire locks (spinlock_t, rwlock_t) which are transformed into sleeping locks on PREEMPT_RT and must not be acquired in hard IRQ context. Changing the locks into locks which could be acquired in this context will lead to other problems such as increased latencies if everything in the chain has IRQ-off locks. This will not solve all the issues as one callback has been noticed which invoked kref_put() and its callback invokes kfree() and this can not be invoked in hardirq context. Some callbacks are required to be invoked in hardirq context even on PREEMPT_RT to work properly. This includes for instance the NO_HZ callback which needs to be able to observe the idle context. The callbacks which require to be run in hardirq have already been marked. Use this information to split the callbacks onto the two lists on PREEMPT_RT: - lazy_list Work items which are not marked with IRQ_WORK_HARD_IRQ will be added to this list. Callbacks on this list will be invoked from a per-CPU thread. The handler here may acquire sleeping locks such as spinlock_t and invoke kfree(). - raised_list Work items which are marked with IRQ_WORK_HARD_IRQ will be added to this list. They will be invoked in hardirq context and must not acquire any sleeping locks. The wake up of the per-CPU thread occurs from irq_work handler/ hardirq context. The thread runs with lowest RT priority to ensure it runs before any SCHED_OTHER tasks do. [bigeasy: melt tglx's irq_work_tick_soft() which splits irq_work_tick() into a hard and soft variant. Collected fixes over time from Steven Rostedt and Mike Galbraith. Move to per-CPU threads instead of softirq as suggested by PeterZ.] Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20211007092646.uhshe3ut2wkrcfzv@linutronix.de
2021-10-07 09:26:46 +00:00
struct llist_head *list;
bool rt_lazy_work = false;
bool lazy_work = false;
int work_flags;
work_flags = atomic_read(&work->node.a_flags);
if (work_flags & IRQ_WORK_LAZY)
lazy_work = true;
else if (IS_ENABLED(CONFIG_PREEMPT_RT) &&
!(work_flags & IRQ_WORK_HARD_IRQ))
rt_lazy_work = true;
if (lazy_work || rt_lazy_work)
list = this_cpu_ptr(&lazy_list);
else
list = this_cpu_ptr(&raised_list);
if (!llist_add(&work->node.llist, list))
return;
/* If the work is "lazy", handle it from next tick if any */
irq_work: Handle some irq_work in a per-CPU thread on PREEMPT_RT The irq_work callback is invoked in hard IRQ context. By default all callbacks are scheduled for invocation right away (given supported by the architecture) except for the ones marked IRQ_WORK_LAZY which are delayed until the next timer-tick. While looking over the callbacks, some of them may acquire locks (spinlock_t, rwlock_t) which are transformed into sleeping locks on PREEMPT_RT and must not be acquired in hard IRQ context. Changing the locks into locks which could be acquired in this context will lead to other problems such as increased latencies if everything in the chain has IRQ-off locks. This will not solve all the issues as one callback has been noticed which invoked kref_put() and its callback invokes kfree() and this can not be invoked in hardirq context. Some callbacks are required to be invoked in hardirq context even on PREEMPT_RT to work properly. This includes for instance the NO_HZ callback which needs to be able to observe the idle context. The callbacks which require to be run in hardirq have already been marked. Use this information to split the callbacks onto the two lists on PREEMPT_RT: - lazy_list Work items which are not marked with IRQ_WORK_HARD_IRQ will be added to this list. Callbacks on this list will be invoked from a per-CPU thread. The handler here may acquire sleeping locks such as spinlock_t and invoke kfree(). - raised_list Work items which are marked with IRQ_WORK_HARD_IRQ will be added to this list. They will be invoked in hardirq context and must not acquire any sleeping locks. The wake up of the per-CPU thread occurs from irq_work handler/ hardirq context. The thread runs with lowest RT priority to ensure it runs before any SCHED_OTHER tasks do. [bigeasy: melt tglx's irq_work_tick_soft() which splits irq_work_tick() into a hard and soft variant. Collected fixes over time from Steven Rostedt and Mike Galbraith. Move to per-CPU threads instead of softirq as suggested by PeterZ.] Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20211007092646.uhshe3ut2wkrcfzv@linutronix.de
2021-10-07 09:26:46 +00:00
if (!lazy_work || tick_nohz_tick_stopped())
irq_work_raise(work);
}
/* Enqueue the irq work @work on the current CPU */
bool irq_work_queue(struct irq_work *work)
{
/* Only queue if not already pending */
if (!irq_work_claim(work))
return false;
/* Queue the entry and raise the IPI if needed. */
preempt_disable();
__irq_work_queue_local(work);
preempt_enable();
return true;
}
EXPORT_SYMBOL_GPL(irq_work_queue);
/*
* Enqueue the irq_work @work on @cpu unless it's already pending
* somewhere.
*
* Can be re-enqueued while the callback is still in progress.
*/
bool irq_work_queue_on(struct irq_work *work, int cpu)
{
#ifndef CONFIG_SMP
return irq_work_queue(work);
#else /* CONFIG_SMP: */
/* All work should have been flushed before going offline */
WARN_ON_ONCE(cpu_is_offline(cpu));
/* Only queue if not already pending */
if (!irq_work_claim(work))
return false;
irq_work: use kasan_record_aux_stack_noalloc() record callstack On PREEMPT_RT kernel and KASAN is enabled. the kasan_record_aux_stack() may call alloc_pages(), and the rt-spinlock will be acquired, if currently in atomic context, will trigger warning: BUG: sleeping function called from invalid context at kernel/locking/spinlock_rt.c:46 in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 239, name: bootlogd Preemption disabled at: [<ffffffffbab1a531>] rt_mutex_slowunlock+0xa1/0x4e0 CPU: 3 PID: 239 Comm: bootlogd Tainted: G W 5.17.1-rt17-yocto-preempt-rt+ #105 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.15.0-0-g2dd4b9b3f840-prebuilt.qemu.org 04/01/2014 Call Trace: __might_resched.cold+0x13b/0x173 rt_spin_lock+0x5b/0xf0 get_page_from_freelist+0x20c/0x1610 __alloc_pages+0x25e/0x5e0 __stack_depot_save+0x3c0/0x4a0 kasan_save_stack+0x3a/0x50 __kasan_record_aux_stack+0xb6/0xc0 kasan_record_aux_stack+0xe/0x10 irq_work_queue_on+0x6a/0x1c0 pull_rt_task+0x631/0x6b0 do_balance_callbacks+0x56/0x80 __balance_callbacks+0x63/0x90 rt_mutex_setprio+0x349/0x880 rt_mutex_slowunlock+0x22a/0x4e0 rt_spin_unlock+0x49/0x80 uart_write+0x186/0x2b0 do_output_char+0x2e9/0x3a0 n_tty_write+0x306/0x800 file_tty_write.isra.0+0x2af/0x450 tty_write+0x22/0x30 new_sync_write+0x27c/0x3a0 vfs_write+0x3f7/0x5d0 ksys_write+0xd9/0x180 __x64_sys_write+0x43/0x50 do_syscall_64+0x44/0x90 entry_SYSCALL_64_after_hwframe+0x44/0xae Fix it by using kasan_record_aux_stack_noalloc() to avoid the call to alloc_pages(). Link: https://lkml.kernel.org/r/20220402142555.2699582-1-qiang1.zhang@intel.com Signed-off-by: Zqiang <qiang1.zhang@intel.com> Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com> Cc: Alexander Potapenko <glider@google.com> Cc: Andrey Konovalov <andreyknvl@gmail.com> Cc: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2022-04-15 02:13:34 +00:00
kasan_record_aux_stack_noalloc(work);
preempt_disable();
if (cpu != smp_processor_id()) {
/* Arch remote IPI send/receive backend aren't NMI safe */
WARN_ON_ONCE(in_nmi());
irq_work: Handle some irq_work in a per-CPU thread on PREEMPT_RT The irq_work callback is invoked in hard IRQ context. By default all callbacks are scheduled for invocation right away (given supported by the architecture) except for the ones marked IRQ_WORK_LAZY which are delayed until the next timer-tick. While looking over the callbacks, some of them may acquire locks (spinlock_t, rwlock_t) which are transformed into sleeping locks on PREEMPT_RT and must not be acquired in hard IRQ context. Changing the locks into locks which could be acquired in this context will lead to other problems such as increased latencies if everything in the chain has IRQ-off locks. This will not solve all the issues as one callback has been noticed which invoked kref_put() and its callback invokes kfree() and this can not be invoked in hardirq context. Some callbacks are required to be invoked in hardirq context even on PREEMPT_RT to work properly. This includes for instance the NO_HZ callback which needs to be able to observe the idle context. The callbacks which require to be run in hardirq have already been marked. Use this information to split the callbacks onto the two lists on PREEMPT_RT: - lazy_list Work items which are not marked with IRQ_WORK_HARD_IRQ will be added to this list. Callbacks on this list will be invoked from a per-CPU thread. The handler here may acquire sleeping locks such as spinlock_t and invoke kfree(). - raised_list Work items which are marked with IRQ_WORK_HARD_IRQ will be added to this list. They will be invoked in hardirq context and must not acquire any sleeping locks. The wake up of the per-CPU thread occurs from irq_work handler/ hardirq context. The thread runs with lowest RT priority to ensure it runs before any SCHED_OTHER tasks do. [bigeasy: melt tglx's irq_work_tick_soft() which splits irq_work_tick() into a hard and soft variant. Collected fixes over time from Steven Rostedt and Mike Galbraith. Move to per-CPU threads instead of softirq as suggested by PeterZ.] Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20211007092646.uhshe3ut2wkrcfzv@linutronix.de
2021-10-07 09:26:46 +00:00
/*
* On PREEMPT_RT the items which are not marked as
* IRQ_WORK_HARD_IRQ are added to the lazy list and a HARD work
* item is used on the remote CPU to wake the thread.
*/
if (IS_ENABLED(CONFIG_PREEMPT_RT) &&
!(atomic_read(&work->node.a_flags) & IRQ_WORK_HARD_IRQ)) {
if (!llist_add(&work->node.llist, &per_cpu(lazy_list, cpu)))
goto out;
work = &per_cpu(irq_work_wakeup, cpu);
if (!irq_work_claim(work))
goto out;
}
__smp_call_single_queue(cpu, &work->node.llist);
irq_work: Split raised and lazy lists An irq work can be handled from two places: from the tick if the work carries the "lazy" flag and the tick is periodic, or from a self IPI. We merge all these works in a single list and we use some per cpu latch to avoid raising a self-IPI when one is already pending. Now we could do away with this ugly latch if only the list was only made of non-lazy works. Just enqueueing a work on the empty list would be enough to know if we need to raise an IPI or not. Also we are going to implement remote irq work queuing. Then the per CPU latch will need to become atomic in the global scope. That's too bad because, here as well, just enqueueing a work on an empty list of non-lazy works would be enough to know if we need to raise an IPI or not. So lets take a way out of this: split the works in two distinct lists, one for the works that can be handled by the next tick and another one for those handled by the IPI. Just checking if the latter is empty when we queue a new work is enough to know if we need to raise an IPI. Suggested-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Peter Zijlstra <peterz@infradead.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Kevin Hilman <khilman@linaro.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2014-05-23 16:10:21 +00:00
} else {
__irq_work_queue_local(work);
}
irq_work: Handle some irq_work in a per-CPU thread on PREEMPT_RT The irq_work callback is invoked in hard IRQ context. By default all callbacks are scheduled for invocation right away (given supported by the architecture) except for the ones marked IRQ_WORK_LAZY which are delayed until the next timer-tick. While looking over the callbacks, some of them may acquire locks (spinlock_t, rwlock_t) which are transformed into sleeping locks on PREEMPT_RT and must not be acquired in hard IRQ context. Changing the locks into locks which could be acquired in this context will lead to other problems such as increased latencies if everything in the chain has IRQ-off locks. This will not solve all the issues as one callback has been noticed which invoked kref_put() and its callback invokes kfree() and this can not be invoked in hardirq context. Some callbacks are required to be invoked in hardirq context even on PREEMPT_RT to work properly. This includes for instance the NO_HZ callback which needs to be able to observe the idle context. The callbacks which require to be run in hardirq have already been marked. Use this information to split the callbacks onto the two lists on PREEMPT_RT: - lazy_list Work items which are not marked with IRQ_WORK_HARD_IRQ will be added to this list. Callbacks on this list will be invoked from a per-CPU thread. The handler here may acquire sleeping locks such as spinlock_t and invoke kfree(). - raised_list Work items which are marked with IRQ_WORK_HARD_IRQ will be added to this list. They will be invoked in hardirq context and must not acquire any sleeping locks. The wake up of the per-CPU thread occurs from irq_work handler/ hardirq context. The thread runs with lowest RT priority to ensure it runs before any SCHED_OTHER tasks do. [bigeasy: melt tglx's irq_work_tick_soft() which splits irq_work_tick() into a hard and soft variant. Collected fixes over time from Steven Rostedt and Mike Galbraith. Move to per-CPU threads instead of softirq as suggested by PeterZ.] Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20211007092646.uhshe3ut2wkrcfzv@linutronix.de
2021-10-07 09:26:46 +00:00
out:
preempt_enable();
return true;
#endif /* CONFIG_SMP */
}
bool irq_work_needs_cpu(void)
{
irq_work: Split raised and lazy lists An irq work can be handled from two places: from the tick if the work carries the "lazy" flag and the tick is periodic, or from a self IPI. We merge all these works in a single list and we use some per cpu latch to avoid raising a self-IPI when one is already pending. Now we could do away with this ugly latch if only the list was only made of non-lazy works. Just enqueueing a work on the empty list would be enough to know if we need to raise an IPI or not. Also we are going to implement remote irq work queuing. Then the per CPU latch will need to become atomic in the global scope. That's too bad because, here as well, just enqueueing a work on an empty list of non-lazy works would be enough to know if we need to raise an IPI or not. So lets take a way out of this: split the works in two distinct lists, one for the works that can be handled by the next tick and another one for those handled by the IPI. Just checking if the latter is empty when we queue a new work is enough to know if we need to raise an IPI. Suggested-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Peter Zijlstra <peterz@infradead.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Kevin Hilman <khilman@linaro.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2014-05-23 16:10:21 +00:00
struct llist_head *raised, *lazy;
raised = this_cpu_ptr(&raised_list);
lazy = this_cpu_ptr(&lazy_list);
irq_work: Force raised irq work to run on irq work interrupt The nohz full kick, which restarts the tick when any resource depend on it, can't be executed anywhere given the operation it does on timers. If it is called from the scheduler or timers code, chances are that we run into a deadlock. This is why we run the nohz full kick from an irq work. That way we make sure that the kick runs on a virgin context. However if that's the case when irq work runs in its own dedicated self-ipi, things are different for the big bunch of archs that don't support the self triggered way. In order to support them, irq works are also handled by the timer interrupt as fallback. Now when irq works run on the timer interrupt, the context isn't blank. More precisely, they can run in the context of the hrtimer that runs the tick. But the nohz kick cancels and restarts this hrtimer and cancelling an hrtimer from itself isn't allowed. This is why we run in an endless loop: Kernel panic - not syncing: Watchdog detected hard LOCKUP on cpu 2 CPU: 2 PID: 7538 Comm: kworker/u8:8 Not tainted 3.16.0+ #34 Workqueue: btrfs-endio-write normal_work_helper [btrfs] ffff880244c06c88 000000001b486fe1 ffff880244c06bf0 ffffffff8a7f1e37 ffffffff8ac52a18 ffff880244c06c78 ffffffff8a7ef928 0000000000000010 ffff880244c06c88 ffff880244c06c20 000000001b486fe1 0000000000000000 Call Trace: <NMI[<ffffffff8a7f1e37>] dump_stack+0x4e/0x7a [<ffffffff8a7ef928>] panic+0xd4/0x207 [<ffffffff8a1450e8>] watchdog_overflow_callback+0x118/0x120 [<ffffffff8a186b0e>] __perf_event_overflow+0xae/0x350 [<ffffffff8a184f80>] ? perf_event_task_disable+0xa0/0xa0 [<ffffffff8a01a4cf>] ? x86_perf_event_set_period+0xbf/0x150 [<ffffffff8a187934>] perf_event_overflow+0x14/0x20 [<ffffffff8a020386>] intel_pmu_handle_irq+0x206/0x410 [<ffffffff8a01937b>] perf_event_nmi_handler+0x2b/0x50 [<ffffffff8a007b72>] nmi_handle+0xd2/0x390 [<ffffffff8a007aa5>] ? nmi_handle+0x5/0x390 [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 [<ffffffff8a008062>] default_do_nmi+0x72/0x1c0 [<ffffffff8a008268>] do_nmi+0xb8/0x100 [<ffffffff8a7ff66a>] end_repeat_nmi+0x1e/0x2e [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 <<EOE><IRQ[<ffffffff8a0ccd2f>] lock_acquired+0xaf/0x450 [<ffffffff8a0f74c5>] ? lock_hrtimer_base.isra.20+0x25/0x50 [<ffffffff8a7fc678>] _raw_spin_lock_irqsave+0x78/0x90 [<ffffffff8a0f74c5>] ? lock_hrtimer_base.isra.20+0x25/0x50 [<ffffffff8a0f74c5>] lock_hrtimer_base.isra.20+0x25/0x50 [<ffffffff8a0f7723>] hrtimer_try_to_cancel+0x33/0x1e0 [<ffffffff8a0f78ea>] hrtimer_cancel+0x1a/0x30 [<ffffffff8a109237>] tick_nohz_restart+0x17/0x90 [<ffffffff8a10a213>] __tick_nohz_full_check+0xc3/0x100 [<ffffffff8a10a25e>] nohz_full_kick_work_func+0xe/0x10 [<ffffffff8a17c884>] irq_work_run_list+0x44/0x70 [<ffffffff8a17c8da>] irq_work_run+0x2a/0x50 [<ffffffff8a0f700b>] update_process_times+0x5b/0x70 [<ffffffff8a109005>] tick_sched_handle.isra.21+0x25/0x60 [<ffffffff8a109b81>] tick_sched_timer+0x41/0x60 [<ffffffff8a0f7aa2>] __run_hrtimer+0x72/0x470 [<ffffffff8a109b40>] ? tick_sched_do_timer+0xb0/0xb0 [<ffffffff8a0f8707>] hrtimer_interrupt+0x117/0x270 [<ffffffff8a034357>] local_apic_timer_interrupt+0x37/0x60 [<ffffffff8a80010f>] smp_apic_timer_interrupt+0x3f/0x50 [<ffffffff8a7fe52f>] apic_timer_interrupt+0x6f/0x80 To fix this we force non-lazy irq works to run on irq work self-IPIs when available. That ability of the arch to trigger irq work self IPIs is available with arch_irq_work_has_interrupt(). Reported-by: Catalin Iacob <iacobcatalin@gmail.com> Reported-by: Dave Jones <davej@redhat.com> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2014-08-16 16:37:19 +00:00
if (llist_empty(raised) || arch_irq_work_has_interrupt())
if (llist_empty(lazy))
return false;
/* All work should have been flushed before going offline */
WARN_ON_ONCE(cpu_is_offline(smp_processor_id()));
return true;
}
void irq_work_single(void *arg)
{
struct irq_work *work = arg;
int flags;
/*
* Clear the PENDING bit, after this point the @work can be re-used.
* The PENDING bit acts as a lock, and we own it, so we can clear it
* without atomic ops.
*/
flags = atomic_read(&work->node.a_flags);
flags &= ~IRQ_WORK_PENDING;
atomic_set(&work->node.a_flags, flags);
/*
* See irq_work_claim().
*/
smp_mb();
lockdep_irq_work_enter(flags);
work->func(work);
lockdep_irq_work_exit(flags);
/*
* Clear the BUSY bit, if set, and return to the free state if no-one
* else claimed it meanwhile.
*/
(void)atomic_cmpxchg(&work->node.a_flags, flags, flags & ~IRQ_WORK_BUSY);
if ((IS_ENABLED(CONFIG_PREEMPT_RT) && !irq_work_is_hard(work)) ||
!arch_irq_work_has_interrupt())
rcuwait_wake_up(&work->irqwait);
}
irq_work: Split raised and lazy lists An irq work can be handled from two places: from the tick if the work carries the "lazy" flag and the tick is periodic, or from a self IPI. We merge all these works in a single list and we use some per cpu latch to avoid raising a self-IPI when one is already pending. Now we could do away with this ugly latch if only the list was only made of non-lazy works. Just enqueueing a work on the empty list would be enough to know if we need to raise an IPI or not. Also we are going to implement remote irq work queuing. Then the per CPU latch will need to become atomic in the global scope. That's too bad because, here as well, just enqueueing a work on an empty list of non-lazy works would be enough to know if we need to raise an IPI or not. So lets take a way out of this: split the works in two distinct lists, one for the works that can be handled by the next tick and another one for those handled by the IPI. Just checking if the latter is empty when we queue a new work is enough to know if we need to raise an IPI. Suggested-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Peter Zijlstra <peterz@infradead.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Kevin Hilman <khilman@linaro.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2014-05-23 16:10:21 +00:00
static void irq_work_run_list(struct llist_head *list)
{
struct irq_work *work, *tmp;
struct llist_node *llnode;
irq_work: Handle some irq_work in a per-CPU thread on PREEMPT_RT The irq_work callback is invoked in hard IRQ context. By default all callbacks are scheduled for invocation right away (given supported by the architecture) except for the ones marked IRQ_WORK_LAZY which are delayed until the next timer-tick. While looking over the callbacks, some of them may acquire locks (spinlock_t, rwlock_t) which are transformed into sleeping locks on PREEMPT_RT and must not be acquired in hard IRQ context. Changing the locks into locks which could be acquired in this context will lead to other problems such as increased latencies if everything in the chain has IRQ-off locks. This will not solve all the issues as one callback has been noticed which invoked kref_put() and its callback invokes kfree() and this can not be invoked in hardirq context. Some callbacks are required to be invoked in hardirq context even on PREEMPT_RT to work properly. This includes for instance the NO_HZ callback which needs to be able to observe the idle context. The callbacks which require to be run in hardirq have already been marked. Use this information to split the callbacks onto the two lists on PREEMPT_RT: - lazy_list Work items which are not marked with IRQ_WORK_HARD_IRQ will be added to this list. Callbacks on this list will be invoked from a per-CPU thread. The handler here may acquire sleeping locks such as spinlock_t and invoke kfree(). - raised_list Work items which are marked with IRQ_WORK_HARD_IRQ will be added to this list. They will be invoked in hardirq context and must not acquire any sleeping locks. The wake up of the per-CPU thread occurs from irq_work handler/ hardirq context. The thread runs with lowest RT priority to ensure it runs before any SCHED_OTHER tasks do. [bigeasy: melt tglx's irq_work_tick_soft() which splits irq_work_tick() into a hard and soft variant. Collected fixes over time from Steven Rostedt and Mike Galbraith. Move to per-CPU threads instead of softirq as suggested by PeterZ.] Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20211007092646.uhshe3ut2wkrcfzv@linutronix.de
2021-10-07 09:26:46 +00:00
/*
* On PREEMPT_RT IRQ-work which is not marked as HARD will be processed
* in a per-CPU thread in preemptible context. Only the items which are
* marked as IRQ_WORK_HARD_IRQ will be processed in hardirq context.
*/
BUG_ON(!irqs_disabled() && !IS_ENABLED(CONFIG_PREEMPT_RT));
irq_work: Split raised and lazy lists An irq work can be handled from two places: from the tick if the work carries the "lazy" flag and the tick is periodic, or from a self IPI. We merge all these works in a single list and we use some per cpu latch to avoid raising a self-IPI when one is already pending. Now we could do away with this ugly latch if only the list was only made of non-lazy works. Just enqueueing a work on the empty list would be enough to know if we need to raise an IPI or not. Also we are going to implement remote irq work queuing. Then the per CPU latch will need to become atomic in the global scope. That's too bad because, here as well, just enqueueing a work on an empty list of non-lazy works would be enough to know if we need to raise an IPI or not. So lets take a way out of this: split the works in two distinct lists, one for the works that can be handled by the next tick and another one for those handled by the IPI. Just checking if the latter is empty when we queue a new work is enough to know if we need to raise an IPI. Suggested-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Peter Zijlstra <peterz@infradead.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Kevin Hilman <khilman@linaro.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2014-05-23 16:10:21 +00:00
if (llist_empty(list))
return;
irq_work: Split raised and lazy lists An irq work can be handled from two places: from the tick if the work carries the "lazy" flag and the tick is periodic, or from a self IPI. We merge all these works in a single list and we use some per cpu latch to avoid raising a self-IPI when one is already pending. Now we could do away with this ugly latch if only the list was only made of non-lazy works. Just enqueueing a work on the empty list would be enough to know if we need to raise an IPI or not. Also we are going to implement remote irq work queuing. Then the per CPU latch will need to become atomic in the global scope. That's too bad because, here as well, just enqueueing a work on an empty list of non-lazy works would be enough to know if we need to raise an IPI or not. So lets take a way out of this: split the works in two distinct lists, one for the works that can be handled by the next tick and another one for those handled by the IPI. Just checking if the latter is empty when we queue a new work is enough to know if we need to raise an IPI. Suggested-by: Peter Zijlstra <peterz@infradead.org> Acked-by: Peter Zijlstra <peterz@infradead.org> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Kevin Hilman <khilman@linaro.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Viresh Kumar <viresh.kumar@linaro.org> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2014-05-23 16:10:21 +00:00
llnode = llist_del_all(list);
llist_for_each_entry_safe(work, tmp, llnode, node.llist)
irq_work_single(work);
}
/*
* hotplug calls this through:
* hotplug_cfd() -> flush_smp_call_function_queue()
*/
void irq_work_run(void)
{
irq_work_run_list(this_cpu_ptr(&raised_list));
irq_work: Handle some irq_work in a per-CPU thread on PREEMPT_RT The irq_work callback is invoked in hard IRQ context. By default all callbacks are scheduled for invocation right away (given supported by the architecture) except for the ones marked IRQ_WORK_LAZY which are delayed until the next timer-tick. While looking over the callbacks, some of them may acquire locks (spinlock_t, rwlock_t) which are transformed into sleeping locks on PREEMPT_RT and must not be acquired in hard IRQ context. Changing the locks into locks which could be acquired in this context will lead to other problems such as increased latencies if everything in the chain has IRQ-off locks. This will not solve all the issues as one callback has been noticed which invoked kref_put() and its callback invokes kfree() and this can not be invoked in hardirq context. Some callbacks are required to be invoked in hardirq context even on PREEMPT_RT to work properly. This includes for instance the NO_HZ callback which needs to be able to observe the idle context. The callbacks which require to be run in hardirq have already been marked. Use this information to split the callbacks onto the two lists on PREEMPT_RT: - lazy_list Work items which are not marked with IRQ_WORK_HARD_IRQ will be added to this list. Callbacks on this list will be invoked from a per-CPU thread. The handler here may acquire sleeping locks such as spinlock_t and invoke kfree(). - raised_list Work items which are marked with IRQ_WORK_HARD_IRQ will be added to this list. They will be invoked in hardirq context and must not acquire any sleeping locks. The wake up of the per-CPU thread occurs from irq_work handler/ hardirq context. The thread runs with lowest RT priority to ensure it runs before any SCHED_OTHER tasks do. [bigeasy: melt tglx's irq_work_tick_soft() which splits irq_work_tick() into a hard and soft variant. Collected fixes over time from Steven Rostedt and Mike Galbraith. Move to per-CPU threads instead of softirq as suggested by PeterZ.] Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20211007092646.uhshe3ut2wkrcfzv@linutronix.de
2021-10-07 09:26:46 +00:00
if (!IS_ENABLED(CONFIG_PREEMPT_RT))
irq_work_run_list(this_cpu_ptr(&lazy_list));
else
wake_irq_workd();
}
EXPORT_SYMBOL_GPL(irq_work_run);
irq_work: Force raised irq work to run on irq work interrupt The nohz full kick, which restarts the tick when any resource depend on it, can't be executed anywhere given the operation it does on timers. If it is called from the scheduler or timers code, chances are that we run into a deadlock. This is why we run the nohz full kick from an irq work. That way we make sure that the kick runs on a virgin context. However if that's the case when irq work runs in its own dedicated self-ipi, things are different for the big bunch of archs that don't support the self triggered way. In order to support them, irq works are also handled by the timer interrupt as fallback. Now when irq works run on the timer interrupt, the context isn't blank. More precisely, they can run in the context of the hrtimer that runs the tick. But the nohz kick cancels and restarts this hrtimer and cancelling an hrtimer from itself isn't allowed. This is why we run in an endless loop: Kernel panic - not syncing: Watchdog detected hard LOCKUP on cpu 2 CPU: 2 PID: 7538 Comm: kworker/u8:8 Not tainted 3.16.0+ #34 Workqueue: btrfs-endio-write normal_work_helper [btrfs] ffff880244c06c88 000000001b486fe1 ffff880244c06bf0 ffffffff8a7f1e37 ffffffff8ac52a18 ffff880244c06c78 ffffffff8a7ef928 0000000000000010 ffff880244c06c88 ffff880244c06c20 000000001b486fe1 0000000000000000 Call Trace: <NMI[<ffffffff8a7f1e37>] dump_stack+0x4e/0x7a [<ffffffff8a7ef928>] panic+0xd4/0x207 [<ffffffff8a1450e8>] watchdog_overflow_callback+0x118/0x120 [<ffffffff8a186b0e>] __perf_event_overflow+0xae/0x350 [<ffffffff8a184f80>] ? perf_event_task_disable+0xa0/0xa0 [<ffffffff8a01a4cf>] ? x86_perf_event_set_period+0xbf/0x150 [<ffffffff8a187934>] perf_event_overflow+0x14/0x20 [<ffffffff8a020386>] intel_pmu_handle_irq+0x206/0x410 [<ffffffff8a01937b>] perf_event_nmi_handler+0x2b/0x50 [<ffffffff8a007b72>] nmi_handle+0xd2/0x390 [<ffffffff8a007aa5>] ? nmi_handle+0x5/0x390 [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 [<ffffffff8a008062>] default_do_nmi+0x72/0x1c0 [<ffffffff8a008268>] do_nmi+0xb8/0x100 [<ffffffff8a7ff66a>] end_repeat_nmi+0x1e/0x2e [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 <<EOE><IRQ[<ffffffff8a0ccd2f>] lock_acquired+0xaf/0x450 [<ffffffff8a0f74c5>] ? lock_hrtimer_base.isra.20+0x25/0x50 [<ffffffff8a7fc678>] _raw_spin_lock_irqsave+0x78/0x90 [<ffffffff8a0f74c5>] ? lock_hrtimer_base.isra.20+0x25/0x50 [<ffffffff8a0f74c5>] lock_hrtimer_base.isra.20+0x25/0x50 [<ffffffff8a0f7723>] hrtimer_try_to_cancel+0x33/0x1e0 [<ffffffff8a0f78ea>] hrtimer_cancel+0x1a/0x30 [<ffffffff8a109237>] tick_nohz_restart+0x17/0x90 [<ffffffff8a10a213>] __tick_nohz_full_check+0xc3/0x100 [<ffffffff8a10a25e>] nohz_full_kick_work_func+0xe/0x10 [<ffffffff8a17c884>] irq_work_run_list+0x44/0x70 [<ffffffff8a17c8da>] irq_work_run+0x2a/0x50 [<ffffffff8a0f700b>] update_process_times+0x5b/0x70 [<ffffffff8a109005>] tick_sched_handle.isra.21+0x25/0x60 [<ffffffff8a109b81>] tick_sched_timer+0x41/0x60 [<ffffffff8a0f7aa2>] __run_hrtimer+0x72/0x470 [<ffffffff8a109b40>] ? tick_sched_do_timer+0xb0/0xb0 [<ffffffff8a0f8707>] hrtimer_interrupt+0x117/0x270 [<ffffffff8a034357>] local_apic_timer_interrupt+0x37/0x60 [<ffffffff8a80010f>] smp_apic_timer_interrupt+0x3f/0x50 [<ffffffff8a7fe52f>] apic_timer_interrupt+0x6f/0x80 To fix this we force non-lazy irq works to run on irq work self-IPIs when available. That ability of the arch to trigger irq work self IPIs is available with arch_irq_work_has_interrupt(). Reported-by: Catalin Iacob <iacobcatalin@gmail.com> Reported-by: Dave Jones <davej@redhat.com> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2014-08-16 16:37:19 +00:00
void irq_work_tick(void)
{
struct llist_head *raised = this_cpu_ptr(&raised_list);
irq_work: Force raised irq work to run on irq work interrupt The nohz full kick, which restarts the tick when any resource depend on it, can't be executed anywhere given the operation it does on timers. If it is called from the scheduler or timers code, chances are that we run into a deadlock. This is why we run the nohz full kick from an irq work. That way we make sure that the kick runs on a virgin context. However if that's the case when irq work runs in its own dedicated self-ipi, things are different for the big bunch of archs that don't support the self triggered way. In order to support them, irq works are also handled by the timer interrupt as fallback. Now when irq works run on the timer interrupt, the context isn't blank. More precisely, they can run in the context of the hrtimer that runs the tick. But the nohz kick cancels and restarts this hrtimer and cancelling an hrtimer from itself isn't allowed. This is why we run in an endless loop: Kernel panic - not syncing: Watchdog detected hard LOCKUP on cpu 2 CPU: 2 PID: 7538 Comm: kworker/u8:8 Not tainted 3.16.0+ #34 Workqueue: btrfs-endio-write normal_work_helper [btrfs] ffff880244c06c88 000000001b486fe1 ffff880244c06bf0 ffffffff8a7f1e37 ffffffff8ac52a18 ffff880244c06c78 ffffffff8a7ef928 0000000000000010 ffff880244c06c88 ffff880244c06c20 000000001b486fe1 0000000000000000 Call Trace: <NMI[<ffffffff8a7f1e37>] dump_stack+0x4e/0x7a [<ffffffff8a7ef928>] panic+0xd4/0x207 [<ffffffff8a1450e8>] watchdog_overflow_callback+0x118/0x120 [<ffffffff8a186b0e>] __perf_event_overflow+0xae/0x350 [<ffffffff8a184f80>] ? perf_event_task_disable+0xa0/0xa0 [<ffffffff8a01a4cf>] ? x86_perf_event_set_period+0xbf/0x150 [<ffffffff8a187934>] perf_event_overflow+0x14/0x20 [<ffffffff8a020386>] intel_pmu_handle_irq+0x206/0x410 [<ffffffff8a01937b>] perf_event_nmi_handler+0x2b/0x50 [<ffffffff8a007b72>] nmi_handle+0xd2/0x390 [<ffffffff8a007aa5>] ? nmi_handle+0x5/0x390 [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 [<ffffffff8a008062>] default_do_nmi+0x72/0x1c0 [<ffffffff8a008268>] do_nmi+0xb8/0x100 [<ffffffff8a7ff66a>] end_repeat_nmi+0x1e/0x2e [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 <<EOE><IRQ[<ffffffff8a0ccd2f>] lock_acquired+0xaf/0x450 [<ffffffff8a0f74c5>] ? lock_hrtimer_base.isra.20+0x25/0x50 [<ffffffff8a7fc678>] _raw_spin_lock_irqsave+0x78/0x90 [<ffffffff8a0f74c5>] ? lock_hrtimer_base.isra.20+0x25/0x50 [<ffffffff8a0f74c5>] lock_hrtimer_base.isra.20+0x25/0x50 [<ffffffff8a0f7723>] hrtimer_try_to_cancel+0x33/0x1e0 [<ffffffff8a0f78ea>] hrtimer_cancel+0x1a/0x30 [<ffffffff8a109237>] tick_nohz_restart+0x17/0x90 [<ffffffff8a10a213>] __tick_nohz_full_check+0xc3/0x100 [<ffffffff8a10a25e>] nohz_full_kick_work_func+0xe/0x10 [<ffffffff8a17c884>] irq_work_run_list+0x44/0x70 [<ffffffff8a17c8da>] irq_work_run+0x2a/0x50 [<ffffffff8a0f700b>] update_process_times+0x5b/0x70 [<ffffffff8a109005>] tick_sched_handle.isra.21+0x25/0x60 [<ffffffff8a109b81>] tick_sched_timer+0x41/0x60 [<ffffffff8a0f7aa2>] __run_hrtimer+0x72/0x470 [<ffffffff8a109b40>] ? tick_sched_do_timer+0xb0/0xb0 [<ffffffff8a0f8707>] hrtimer_interrupt+0x117/0x270 [<ffffffff8a034357>] local_apic_timer_interrupt+0x37/0x60 [<ffffffff8a80010f>] smp_apic_timer_interrupt+0x3f/0x50 [<ffffffff8a7fe52f>] apic_timer_interrupt+0x6f/0x80 To fix this we force non-lazy irq works to run on irq work self-IPIs when available. That ability of the arch to trigger irq work self IPIs is available with arch_irq_work_has_interrupt(). Reported-by: Catalin Iacob <iacobcatalin@gmail.com> Reported-by: Dave Jones <davej@redhat.com> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2014-08-16 16:37:19 +00:00
if (!llist_empty(raised) && !arch_irq_work_has_interrupt())
irq_work_run_list(raised);
irq_work: Handle some irq_work in a per-CPU thread on PREEMPT_RT The irq_work callback is invoked in hard IRQ context. By default all callbacks are scheduled for invocation right away (given supported by the architecture) except for the ones marked IRQ_WORK_LAZY which are delayed until the next timer-tick. While looking over the callbacks, some of them may acquire locks (spinlock_t, rwlock_t) which are transformed into sleeping locks on PREEMPT_RT and must not be acquired in hard IRQ context. Changing the locks into locks which could be acquired in this context will lead to other problems such as increased latencies if everything in the chain has IRQ-off locks. This will not solve all the issues as one callback has been noticed which invoked kref_put() and its callback invokes kfree() and this can not be invoked in hardirq context. Some callbacks are required to be invoked in hardirq context even on PREEMPT_RT to work properly. This includes for instance the NO_HZ callback which needs to be able to observe the idle context. The callbacks which require to be run in hardirq have already been marked. Use this information to split the callbacks onto the two lists on PREEMPT_RT: - lazy_list Work items which are not marked with IRQ_WORK_HARD_IRQ will be added to this list. Callbacks on this list will be invoked from a per-CPU thread. The handler here may acquire sleeping locks such as spinlock_t and invoke kfree(). - raised_list Work items which are marked with IRQ_WORK_HARD_IRQ will be added to this list. They will be invoked in hardirq context and must not acquire any sleeping locks. The wake up of the per-CPU thread occurs from irq_work handler/ hardirq context. The thread runs with lowest RT priority to ensure it runs before any SCHED_OTHER tasks do. [bigeasy: melt tglx's irq_work_tick_soft() which splits irq_work_tick() into a hard and soft variant. Collected fixes over time from Steven Rostedt and Mike Galbraith. Move to per-CPU threads instead of softirq as suggested by PeterZ.] Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20211007092646.uhshe3ut2wkrcfzv@linutronix.de
2021-10-07 09:26:46 +00:00
if (!IS_ENABLED(CONFIG_PREEMPT_RT))
irq_work_run_list(this_cpu_ptr(&lazy_list));
else
wake_irq_workd();
irq_work: Force raised irq work to run on irq work interrupt The nohz full kick, which restarts the tick when any resource depend on it, can't be executed anywhere given the operation it does on timers. If it is called from the scheduler or timers code, chances are that we run into a deadlock. This is why we run the nohz full kick from an irq work. That way we make sure that the kick runs on a virgin context. However if that's the case when irq work runs in its own dedicated self-ipi, things are different for the big bunch of archs that don't support the self triggered way. In order to support them, irq works are also handled by the timer interrupt as fallback. Now when irq works run on the timer interrupt, the context isn't blank. More precisely, they can run in the context of the hrtimer that runs the tick. But the nohz kick cancels and restarts this hrtimer and cancelling an hrtimer from itself isn't allowed. This is why we run in an endless loop: Kernel panic - not syncing: Watchdog detected hard LOCKUP on cpu 2 CPU: 2 PID: 7538 Comm: kworker/u8:8 Not tainted 3.16.0+ #34 Workqueue: btrfs-endio-write normal_work_helper [btrfs] ffff880244c06c88 000000001b486fe1 ffff880244c06bf0 ffffffff8a7f1e37 ffffffff8ac52a18 ffff880244c06c78 ffffffff8a7ef928 0000000000000010 ffff880244c06c88 ffff880244c06c20 000000001b486fe1 0000000000000000 Call Trace: <NMI[<ffffffff8a7f1e37>] dump_stack+0x4e/0x7a [<ffffffff8a7ef928>] panic+0xd4/0x207 [<ffffffff8a1450e8>] watchdog_overflow_callback+0x118/0x120 [<ffffffff8a186b0e>] __perf_event_overflow+0xae/0x350 [<ffffffff8a184f80>] ? perf_event_task_disable+0xa0/0xa0 [<ffffffff8a01a4cf>] ? x86_perf_event_set_period+0xbf/0x150 [<ffffffff8a187934>] perf_event_overflow+0x14/0x20 [<ffffffff8a020386>] intel_pmu_handle_irq+0x206/0x410 [<ffffffff8a01937b>] perf_event_nmi_handler+0x2b/0x50 [<ffffffff8a007b72>] nmi_handle+0xd2/0x390 [<ffffffff8a007aa5>] ? nmi_handle+0x5/0x390 [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 [<ffffffff8a008062>] default_do_nmi+0x72/0x1c0 [<ffffffff8a008268>] do_nmi+0xb8/0x100 [<ffffffff8a7ff66a>] end_repeat_nmi+0x1e/0x2e [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 [<ffffffff8a0cb7f8>] ? match_held_lock+0x8/0x1b0 <<EOE><IRQ[<ffffffff8a0ccd2f>] lock_acquired+0xaf/0x450 [<ffffffff8a0f74c5>] ? lock_hrtimer_base.isra.20+0x25/0x50 [<ffffffff8a7fc678>] _raw_spin_lock_irqsave+0x78/0x90 [<ffffffff8a0f74c5>] ? lock_hrtimer_base.isra.20+0x25/0x50 [<ffffffff8a0f74c5>] lock_hrtimer_base.isra.20+0x25/0x50 [<ffffffff8a0f7723>] hrtimer_try_to_cancel+0x33/0x1e0 [<ffffffff8a0f78ea>] hrtimer_cancel+0x1a/0x30 [<ffffffff8a109237>] tick_nohz_restart+0x17/0x90 [<ffffffff8a10a213>] __tick_nohz_full_check+0xc3/0x100 [<ffffffff8a10a25e>] nohz_full_kick_work_func+0xe/0x10 [<ffffffff8a17c884>] irq_work_run_list+0x44/0x70 [<ffffffff8a17c8da>] irq_work_run+0x2a/0x50 [<ffffffff8a0f700b>] update_process_times+0x5b/0x70 [<ffffffff8a109005>] tick_sched_handle.isra.21+0x25/0x60 [<ffffffff8a109b81>] tick_sched_timer+0x41/0x60 [<ffffffff8a0f7aa2>] __run_hrtimer+0x72/0x470 [<ffffffff8a109b40>] ? tick_sched_do_timer+0xb0/0xb0 [<ffffffff8a0f8707>] hrtimer_interrupt+0x117/0x270 [<ffffffff8a034357>] local_apic_timer_interrupt+0x37/0x60 [<ffffffff8a80010f>] smp_apic_timer_interrupt+0x3f/0x50 [<ffffffff8a7fe52f>] apic_timer_interrupt+0x6f/0x80 To fix this we force non-lazy irq works to run on irq work self-IPIs when available. That ability of the arch to trigger irq work self IPIs is available with arch_irq_work_has_interrupt(). Reported-by: Catalin Iacob <iacobcatalin@gmail.com> Reported-by: Dave Jones <davej@redhat.com> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Ingo Molnar <mingo@kernel.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
2014-08-16 16:37:19 +00:00
}
/*
* Synchronize against the irq_work @entry, ensures the entry is not
* currently in use.
*/
void irq_work_sync(struct irq_work *work)
{
lockdep_assert_irqs_enabled();
might_sleep();
if ((IS_ENABLED(CONFIG_PREEMPT_RT) && !irq_work_is_hard(work)) ||
!arch_irq_work_has_interrupt()) {
rcuwait_wait_event(&work->irqwait, !irq_work_is_busy(work),
TASK_UNINTERRUPTIBLE);
return;
}
while (irq_work_is_busy(work))
cpu_relax();
}
EXPORT_SYMBOL_GPL(irq_work_sync);
irq_work: Handle some irq_work in a per-CPU thread on PREEMPT_RT The irq_work callback is invoked in hard IRQ context. By default all callbacks are scheduled for invocation right away (given supported by the architecture) except for the ones marked IRQ_WORK_LAZY which are delayed until the next timer-tick. While looking over the callbacks, some of them may acquire locks (spinlock_t, rwlock_t) which are transformed into sleeping locks on PREEMPT_RT and must not be acquired in hard IRQ context. Changing the locks into locks which could be acquired in this context will lead to other problems such as increased latencies if everything in the chain has IRQ-off locks. This will not solve all the issues as one callback has been noticed which invoked kref_put() and its callback invokes kfree() and this can not be invoked in hardirq context. Some callbacks are required to be invoked in hardirq context even on PREEMPT_RT to work properly. This includes for instance the NO_HZ callback which needs to be able to observe the idle context. The callbacks which require to be run in hardirq have already been marked. Use this information to split the callbacks onto the two lists on PREEMPT_RT: - lazy_list Work items which are not marked with IRQ_WORK_HARD_IRQ will be added to this list. Callbacks on this list will be invoked from a per-CPU thread. The handler here may acquire sleeping locks such as spinlock_t and invoke kfree(). - raised_list Work items which are marked with IRQ_WORK_HARD_IRQ will be added to this list. They will be invoked in hardirq context and must not acquire any sleeping locks. The wake up of the per-CPU thread occurs from irq_work handler/ hardirq context. The thread runs with lowest RT priority to ensure it runs before any SCHED_OTHER tasks do. [bigeasy: melt tglx's irq_work_tick_soft() which splits irq_work_tick() into a hard and soft variant. Collected fixes over time from Steven Rostedt and Mike Galbraith. Move to per-CPU threads instead of softirq as suggested by PeterZ.] Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20211007092646.uhshe3ut2wkrcfzv@linutronix.de
2021-10-07 09:26:46 +00:00
static void run_irq_workd(unsigned int cpu)
{
irq_work_run_list(this_cpu_ptr(&lazy_list));
}
static void irq_workd_setup(unsigned int cpu)
{
sched_set_fifo_low(current);
}
static struct smp_hotplug_thread irqwork_threads = {
.store = &irq_workd,
.setup = irq_workd_setup,
.thread_should_run = irq_workd_should_run,
.thread_fn = run_irq_workd,
.thread_comm = "irq_work/%u",
};
static __init int irq_work_init_threads(void)
{
if (IS_ENABLED(CONFIG_PREEMPT_RT))
BUG_ON(smpboot_register_percpu_thread(&irqwork_threads));
return 0;
}
early_initcall(irq_work_init_threads);