1 files changed, 188 insertions, 35 deletions

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 9289da7f0ac4..889b3632d56d 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -74,7 +74,11 @@ __read_mostly int sysctl_resched_latency_warn_once = 1;
  * Number of tasks to iterate in a single balance run.
  * Limited because this is done with IRQs disabled.
  */
+#ifdef CONFIG_PREEMPT_RT
+const_debug unsigned int sysctl_sched_nr_migrate = 8;
+#else
 const_debug unsigned int sysctl_sched_nr_migrate = 32;
+#endif
 
 /*
  * period over which we measure -rt task CPU usage in us.
@@ -982,6 +986,46 @@ void resched_curr(struct rq *rq)
 		trace_sched_wake_idle_without_ipi(cpu);
 }
 
+#ifdef CONFIG_PREEMPT_LAZY
+
+static int tsk_is_polling(struct task_struct *p)
+{
+#ifdef TIF_POLLING_NRFLAG
+	return test_tsk_thread_flag(p, TIF_POLLING_NRFLAG);
+#else
+	return 0;
+#endif
+}
+
+void resched_curr_lazy(struct rq *rq)
+{
+	struct task_struct *curr = rq->curr;
+	int cpu;
+
+	if (!sched_feat(PREEMPT_LAZY)) {
+		resched_curr(rq);
+		return;
+	}
+
+	if (test_tsk_need_resched(curr))
+		return;
+
+	if (test_tsk_need_resched_lazy(curr))
+		return;
+
+	set_tsk_need_resched_lazy(curr);
+
+	cpu = cpu_of(rq);
+	if (cpu == smp_processor_id())
+		return;
+
+	/* NEED_RESCHED_LAZY must be visible before we test polling */
+	smp_mb();
+	if (!tsk_is_polling(curr))
+		smp_send_reschedule(cpu);
+}
+#endif
+
 void resched_cpu(int cpu)
 {
 	struct rq *rq = cpu_rq(cpu);
@@ -2135,6 +2179,7 @@ void migrate_disable(void)
 	preempt_disable();
 	this_rq()->nr_pinned++;
 	p->migration_disabled = 1;
+	preempt_lazy_disable();
 	preempt_enable();
 }
 EXPORT_SYMBOL_GPL(migrate_disable);
@@ -2146,6 +2191,8 @@ void migrate_enable(void)
 	if (p->migration_disabled > 1) {
 		p->migration_disabled--;
 		return;
+	} else if (WARN_ON_ONCE(p->migration_disabled == 0)) {
+		return;
 	}
 
 	/*
@@ -2163,6 +2210,7 @@ void migrate_enable(void)
 	barrier();
 	p->migration_disabled = 0;
 	this_rq()->nr_pinned--;
+	preempt_lazy_enable();
 	preempt_enable();
 }
 EXPORT_SYMBOL_GPL(migrate_enable);
@@ -3011,7 +3059,7 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state
 		 * is actually now running somewhere else!
 		 */
 		while (task_running(rq, p)) {
-			if (match_state && unlikely(READ_ONCE(p->__state) != match_state))
+			if (match_state && !task_match_state_lock(p, match_state))
 				return 0;
 			cpu_relax();
 		}
@@ -3026,7 +3074,7 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state
 		running = task_running(rq, p);
 		queued = task_on_rq_queued(p);
 		ncsw = 0;
-		if (!match_state || READ_ONCE(p->__state) == match_state)
+		if (!match_state || task_match_state_or_saved(p, match_state))
 			ncsw = p->nvcsw | LONG_MIN; /* sets MSB */
 		task_rq_unlock(rq, p, &rf);
 
@@ -3060,7 +3108,7 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state
 			ktime_t to = NSEC_PER_SEC / HZ;
 
 			set_current_state(TASK_UNINTERRUPTIBLE);
-			schedule_hrtimeout(&to, HRTIMER_MODE_REL);
+			schedule_hrtimeout(&to, HRTIMER_MODE_REL_HARD);
 			continue;
 		}
 
@@ -3189,8 +3237,8 @@ out:
 		 * leave kernel.
 		 */
 		if (p->mm && printk_ratelimit()) {
-			printk_deferred("process %d (%s) no longer affine to cpu%d\n",
-					task_pid_nr(p), p->comm, cpu);
+			printk("process %d (%s) no longer affine to cpu%d\n",
+			       task_pid_nr(p), p->comm, cpu);
 		}
 	}
 
@@ -3589,6 +3637,55 @@ static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags)
 }
 
 /*
+ * Invoked from try_to_wake_up() to check whether the task can be woken up.
+ *
+ * The caller holds p::pi_lock if p != current or has preemption
+ * disabled when p == current.
+ *
+ * The rules of PREEMPT_RT saved_state:
+ *
+ *   The related locking code always holds p::pi_lock when updating
+ *   p::saved_state, which means the code is fully serialized in both cases.
+ *
+ *   The lock wait and lock wakeups happen via TASK_RTLOCK_WAIT. No other
+ *   bits set. This allows to distinguish all wakeup scenarios.
+ */
+static __always_inline
+bool ttwu_state_match(struct task_struct *p, unsigned int state, int *success)
+{
+	if (IS_ENABLED(CONFIG_DEBUG_PREEMPT)) {
+		WARN_ON_ONCE((state & TASK_RTLOCK_WAIT) &&
+			     state != TASK_RTLOCK_WAIT);
+	}
+
+	if (READ_ONCE(p->__state) & state) {
+		*success = 1;
+		return true;
+	}
+
+#ifdef CONFIG_PREEMPT_RT
+	/*
+	 * Saved state preserves the task state across blocking on
+	 * an RT lock.  If the state matches, set p::saved_state to
+	 * TASK_RUNNING, but do not wake the task because it waits
+	 * for a lock wakeup. Also indicate success because from
+	 * the regular waker's point of view this has succeeded.
+	 *
+	 * After acquiring the lock the task will restore p::__state
+	 * from p::saved_state which ensures that the regular
+	 * wakeup is not lost. The restore will also set
+	 * p::saved_state to TASK_RUNNING so any further tests will
+	 * not result in false positives vs. @success
+	 */
+	if (p->saved_state & state) {
+		p->saved_state = TASK_RUNNING;
+		*success = 1;
+	}
+#endif
+	return false;
+}
+
+/*
  * Notes on Program-Order guarantees on SMP systems.
  *
  *  MIGRATION
@@ -3727,10 +3824,9 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
 		 *  - we're serialized against set_special_state() by virtue of
 		 *    it disabling IRQs (this allows not taking ->pi_lock).
 		 */
-		if (!(READ_ONCE(p->__state) & state))
+		if (!ttwu_state_match(p, state, &success))
 			goto out;
 
-		success = 1;
 		trace_sched_waking(p);
 		WRITE_ONCE(p->__state, TASK_RUNNING);
 		trace_sched_wakeup(p);
@@ -3745,14 +3841,11 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
 	 */
 	raw_spin_lock_irqsave(&p->pi_lock, flags);
 	smp_mb__after_spinlock();
-	if (!(READ_ONCE(p->__state) & state))
+	if (!ttwu_state_match(p, state, &success))
 		goto unlock;
 
 	trace_sched_waking(p);
 
-	/* We're going to change ->state: */
-	success = 1;
-
 	/*
 	 * Ensure we load p->on_rq _after_ p->state, otherwise it would
 	 * be possible to, falsely, observe p->on_rq == 0 and get stuck
@@ -4149,6 +4242,9 @@ int sched_fork(unsigned long clone_flags, struct task_struct *p)
 	p->on_cpu = 0;
 #endif
 	init_task_preempt_count(p);
+#ifdef CONFIG_HAVE_PREEMPT_LAZY
+	task_thread_info(p)->preempt_lazy_count = 0;
+#endif
 #ifdef CONFIG_SMP
 	plist_node_init(&p->pushable_tasks, MAX_PRIO);
 	RB_CLEAR_NODE(&p->pushable_dl_tasks);
@@ -4603,23 +4699,18 @@ static struct rq *finish_task_switch(struct task_struct *prev)
 	 *   provided by mmdrop(),
 	 * - a sync_core for SYNC_CORE.
 	 */
+	/*
+	 * We use mmdrop_delayed() here so we don't have to do the
+	 * full __mmdrop() when we are the last user.
+	 */
 	if (mm) {
 		membarrier_mm_sync_core_before_usermode(mm);
-		mmdrop(mm);
+		mmdrop_delayed(mm);
 	}
 	if (unlikely(prev_state == TASK_DEAD)) {
 		if (prev->sched_class->task_dead)
 			prev->sched_class->task_dead(prev);
 
-		/*
-		 * Remove function-return probe instances associated with this
-		 * task and put them back on the free list.
-		 */
-		kprobe_flush_task(prev);
-
-		/* Task is done with its stack. */
-		put_task_stack(prev);
-
 		put_task_struct_rcu_user(prev);
 	}
 
@@ -5877,6 +5968,24 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 #endif /* CONFIG_SCHED_CORE */
 
 /*
+ * Constants for the sched_mode argument of __schedule().
+ *
+ * The mode argument allows RT enabled kernels to differentiate a
+ * preemption from blocking on an 'sleeping' spin/rwlock. Note that
+ * SM_MASK_PREEMPT for !RT has all bits set, which allows the compiler to
+ * optimize the AND operation out and just check for zero.
+ */
+#define SM_NONE			0x0
+#define SM_PREEMPT		0x1
+#define SM_RTLOCK_WAIT		0x2
+
+#ifndef CONFIG_PREEMPT_RT
+# define SM_MASK_PREEMPT	(~0U)
+#else
+# define SM_MASK_PREEMPT	SM_PREEMPT
+#endif
+
+/*
  * __schedule() is the main scheduler function.
  *
  * The main means of driving the scheduler and thus entering this function are:
@@ -5915,7 +6024,7 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
  *
  * WARNING: must be called with preemption disabled!
  */
-static void __sched notrace __schedule(bool preempt)
+static void __sched notrace __schedule(unsigned int sched_mode)
 {
 	struct task_struct *prev, *next;
 	unsigned long *switch_count;
@@ -5928,13 +6037,13 @@ static void __sched notrace __schedule(bool preempt)
 	rq = cpu_rq(cpu);
 	prev = rq->curr;
 
-	schedule_debug(prev, preempt);
+	schedule_debug(prev, !!sched_mode);
 
 	if (sched_feat(HRTICK) || sched_feat(HRTICK_DL))
 		hrtick_clear(rq);
 
 	local_irq_disable();
-	rcu_note_context_switch(preempt);
+	rcu_note_context_switch(!!sched_mode);
 
 	/*
 	 * Make sure that signal_pending_state()->signal_pending() below
@@ -5968,7 +6077,7 @@ static void __sched notrace __schedule(bool preempt)
 	 *  - ptrace_{,un}freeze_traced() can change ->state underneath us.
 	 */
 	prev_state = READ_ONCE(prev->__state);
-	if (!preempt && prev_state) {
+	if (!(sched_mode & SM_MASK_PREEMPT) && prev_state) {
 		if (signal_pending_state(prev_state, prev)) {
 			WRITE_ONCE(prev->__state, TASK_RUNNING);
 		} else {
@@ -6003,6 +6112,7 @@ static void __sched notrace __schedule(bool preempt)
 
 	next = pick_next_task(rq, prev, &rf);
 	clear_tsk_need_resched(prev);
+	clear_tsk_need_resched_lazy(prev);
 	clear_preempt_need_resched();
 #ifdef CONFIG_SCHED_DEBUG
 	rq->last_seen_need_resched_ns = 0;
@@ -6034,7 +6144,7 @@ static void __sched notrace __schedule(bool preempt)
 		migrate_disable_switch(rq, prev);
 		psi_sched_switch(prev, next, !task_on_rq_queued(prev));
 
-		trace_sched_switch(preempt, prev, next);
+		trace_sched_switch(sched_mode & SM_MASK_PREEMPT, prev, next);
 
 		/* Also unlocks the rq: */
 		rq = context_switch(rq, prev, next, &rf);
@@ -6055,7 +6165,7 @@ void __noreturn do_task_dead(void)
 	/* Tell freezer to ignore us: */
 	current->flags |= PF_NOFREEZE;
 
-	__schedule(false);
+	__schedule(SM_NONE);
 	BUG();
 
 	/* Avoid "noreturn function does return" - but don't continue if BUG() is a NOP: */
@@ -6116,7 +6226,7 @@ asmlinkage __visible void __sched schedule(void)
 	sched_submit_work(tsk);
 	do {
 		preempt_disable();
-		__schedule(false);
+		__schedule(SM_NONE);
 		sched_preempt_enable_no_resched();
 	} while (need_resched());
 	sched_update_worker(tsk);
@@ -6144,7 +6254,7 @@ void __sched schedule_idle(void)
 	 */
 	WARN_ON_ONCE(current->__state);
 	do {
-		__schedule(false);
+		__schedule(SM_NONE);
 	} while (need_resched());
 }
 
@@ -6179,6 +6289,18 @@ void __sched schedule_preempt_disabled(void)
 	preempt_disable();
 }
 
+#ifdef CONFIG_PREEMPT_RT
+void __sched notrace schedule_rtlock(void)
+{
+	do {
+		preempt_disable();
+		__schedule(SM_RTLOCK_WAIT);
+		sched_preempt_enable_no_resched();
+	} while (need_resched());
+}
+NOKPROBE_SYMBOL(schedule_rtlock);
+#endif
+
 static void __sched notrace preempt_schedule_common(void)
 {
 	do {
@@ -6197,7 +6319,7 @@ static void __sched notrace preempt_schedule_common(void)
 		 */
 		preempt_disable_notrace();
 		preempt_latency_start(1);
-		__schedule(true);
+		__schedule(SM_PREEMPT);
 		preempt_latency_stop(1);
 		preempt_enable_no_resched_notrace();
 
@@ -6208,6 +6330,30 @@ static void __sched notrace preempt_schedule_common(void)
 	} while (need_resched());
 }
 
+#ifdef CONFIG_PREEMPT_LAZY
+/*
+ * If TIF_NEED_RESCHED is then we allow to be scheduled away since this is
+ * set by a RT task. Oterwise we try to avoid beeing scheduled out as long as
+ * preempt_lazy_count counter >0.
+ */
+static __always_inline int preemptible_lazy(void)
+{
+	if (test_thread_flag(TIF_NEED_RESCHED))
+		return 1;
+	if (current_thread_info()->preempt_lazy_count)
+		return 0;
+	return 1;
+}
+
+#else
+
+static inline int preemptible_lazy(void)
+{
+	return 1;
+}
+
+#endif
+
 #ifdef CONFIG_PREEMPTION
 /*
  * This is the entry point to schedule() from in-kernel preemption
@@ -6221,7 +6367,8 @@ asmlinkage __visible void __sched notrace preempt_schedule(void)
 	 */
 	if (likely(!preemptible()))
 		return;
-
+	if (!preemptible_lazy())
+		return;
 	preempt_schedule_common();
 }
 NOKPROBE_SYMBOL(preempt_schedule);
@@ -6254,6 +6401,9 @@ asmlinkage __visible void __sched notrace preempt_schedule_notrace(void)
 	if (likely(!preemptible()))
 		return;
 
+	if (!preemptible_lazy())
+		return;
+
 	do {
 		/*
 		 * Because the function tracer can trace preempt_count_sub()
@@ -6276,7 +6426,7 @@ asmlinkage __visible void __sched notrace preempt_schedule_notrace(void)
 		 * an infinite recursion.
 		 */
 		prev_ctx = exception_enter();
-		__schedule(true);
+		__schedule(SM_PREEMPT);
 		exception_exit(prev_ctx);
 
 		preempt_latency_stop(1);
@@ -6425,7 +6575,7 @@ asmlinkage __visible void __sched preempt_schedule_irq(void)
 	do {
 		preempt_disable();
 		local_irq_enable();
-		__schedule(true);
+		__schedule(SM_PREEMPT);
 		local_irq_disable();
 		sched_preempt_enable_no_resched();
 	} while (need_resched());
@@ -8375,7 +8525,9 @@ void __init init_idle(struct task_struct *idle, int cpu)
 
 	/* Set the preempt count _outside_ the spinlocks! */
 	init_idle_preempt_count(idle, cpu);
-
+#ifdef CONFIG_HAVE_PREEMPT_LAZY
+	task_thread_info(idle)->preempt_lazy_count = 0;
+#endif
 	/*
 	 * The idle tasks have their own, simple scheduling class:
 	 */
@@ -8480,6 +8632,7 @@ void sched_setnuma(struct task_struct *p, int nid)
 #endif /* CONFIG_NUMA_BALANCING */
 
 #ifdef CONFIG_HOTPLUG_CPU
+
 /*
  * Ensure that the idle task is using init_mm right before its CPU goes
  * offline.
@@ -9172,7 +9325,7 @@ void __init sched_init(void)
 #ifdef CONFIG_DEBUG_ATOMIC_SLEEP
 static inline int preempt_count_equals(int preempt_offset)
 {
-	int nested = preempt_count() + rcu_preempt_depth();
+	int nested = preempt_count() + sched_rcu_preempt_depth();
 
 	return (nested == preempt_offset);
 }