linux/debian/patches-rt/0101-hrtimer-by-timers-by-default-into-the-softirq-contex.patch

From e9a7b277a228256c44836983088b956e32bf3dda Mon Sep 17 00:00:00 2001
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Date: Fri, 3 Jul 2009 08:44:31 -0500
Subject: [PATCH 101/291] hrtimer: by timers by default into the softirq
 context
Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.19/older/patches-4.19.90-rt35.tar.xz

We can't have hrtimers callbacks running in hardirq context on RT. Therefore
the timers are deferred to the softirq context by default.
There are few timers which expect to be run in hardirq context even on RT.
Those are:
- very short running where low latency is critical (kvm lapic)
- timers which take raw locks and need run in hard-irq context (perf, sched)
- wake up related timer (kernel side of clock_nanosleep() and so on)

Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
---
 arch/x86/kvm/lapic.c                 |  2 +-
 include/linux/hrtimer.h              |  6 ++++++
 kernel/events/core.c                 |  4 ++--
 kernel/sched/core.c                  |  2 +-
 kernel/sched/deadline.c              |  2 +-
 kernel/sched/fair.c                  |  4 ++--
 kernel/sched/rt.c                    |  4 ++--
 kernel/time/hrtimer.c                | 21 +++++++++++++++++++--
 kernel/time/tick-broadcast-hrtimer.c |  2 +-
 kernel/time/tick-sched.c             |  2 +-
 kernel/watchdog.c                    |  2 +-
 11 files changed, 37 insertions(+), 14 deletions(-)

diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 262e49301cae..c2f51b6e8974 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -2257,7 +2257,7 @@ int kvm_create_lapic(struct kvm_vcpu *vcpu)
 	apic->vcpu = vcpu;
 
 	hrtimer_init(&apic->lapic_timer.timer, CLOCK_MONOTONIC,
-		     HRTIMER_MODE_ABS_PINNED);
+		     HRTIMER_MODE_ABS_PINNED_HARD);
 	apic->lapic_timer.timer.function = apic_timer_fn;
 
 	/*
diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index 73ad7309436a..2bdb047c7656 100644
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -42,6 +42,7 @@ enum hrtimer_mode {
 	HRTIMER_MODE_REL	= 0x01,
 	HRTIMER_MODE_PINNED	= 0x02,
 	HRTIMER_MODE_SOFT	= 0x04,
+	HRTIMER_MODE_HARD	= 0x08,
 
 	HRTIMER_MODE_ABS_PINNED = HRTIMER_MODE_ABS | HRTIMER_MODE_PINNED,
 	HRTIMER_MODE_REL_PINNED = HRTIMER_MODE_REL | HRTIMER_MODE_PINNED,
@@ -52,6 +53,11 @@ enum hrtimer_mode {
 	HRTIMER_MODE_ABS_PINNED_SOFT = HRTIMER_MODE_ABS_PINNED | HRTIMER_MODE_SOFT,
 	HRTIMER_MODE_REL_PINNED_SOFT = HRTIMER_MODE_REL_PINNED | HRTIMER_MODE_SOFT,
 
+	HRTIMER_MODE_ABS_HARD	= HRTIMER_MODE_ABS | HRTIMER_MODE_HARD,
+	HRTIMER_MODE_REL_HARD	= HRTIMER_MODE_REL | HRTIMER_MODE_HARD,
+
+	HRTIMER_MODE_ABS_PINNED_HARD = HRTIMER_MODE_ABS_PINNED | HRTIMER_MODE_HARD,
+	HRTIMER_MODE_REL_PINNED_HARD = HRTIMER_MODE_REL_PINNED | HRTIMER_MODE_HARD,
 };
 
 /*
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 460d5fd3ec4e..60a883088961 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -1102,7 +1102,7 @@ static void __perf_mux_hrtimer_init(struct perf_cpu_context *cpuctx, int cpu)
 	cpuctx->hrtimer_interval = ns_to_ktime(NSEC_PER_MSEC * interval);
 
 	raw_spin_lock_init(&cpuctx->hrtimer_lock);
-	hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED);
+	hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED_HARD);
 	timer->function = perf_mux_hrtimer_handler;
 }
 
@@ -9239,7 +9239,7 @@ static void perf_swevent_init_hrtimer(struct perf_event *event)
 	if (!is_sampling_event(event))
 		return;
 
-	hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD);
 	hwc->hrtimer.function = perf_swevent_hrtimer;
 
 	/*
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 584978640512..2f6b4365d070 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -315,7 +315,7 @@ static void hrtick_rq_init(struct rq *rq)
 	rq->hrtick_csd.info = rq;
 #endif
 
-	hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD);
 	rq->hrtick_timer.function = hrtick;
 }
 #else	/* CONFIG_SCHED_HRTICK */
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index 4b13df38c069..974a8f9b615a 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -1086,7 +1086,7 @@ void init_dl_task_timer(struct sched_dl_entity *dl_se)
 {
 	struct hrtimer *timer = &dl_se->dl_timer;
 
-	hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD);
 	timer->function = dl_task_timer;
 }
 
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 6e6d9e999814..21cf60c360e8 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -4914,9 +4914,9 @@ void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b)
 	cfs_b->period = ns_to_ktime(default_cfs_period());
 
 	INIT_LIST_HEAD(&cfs_b->throttled_cfs_rq);
-	hrtimer_init(&cfs_b->period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED);
+	hrtimer_init(&cfs_b->period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED_HARD);
 	cfs_b->period_timer.function = sched_cfs_period_timer;
-	hrtimer_init(&cfs_b->slack_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	hrtimer_init(&cfs_b->slack_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD);
 	cfs_b->slack_timer.function = sched_cfs_slack_timer;
 	cfs_b->distribute_running = 0;
 }
diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index b6ca4a630050..aeb99395c03b 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -45,8 +45,8 @@ void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime)
 
 	raw_spin_lock_init(&rt_b->rt_runtime_lock);
 
-	hrtimer_init(&rt_b->rt_period_timer,
-			CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	hrtimer_init(&rt_b->rt_period_timer, CLOCK_MONOTONIC,
+		     HRTIMER_MODE_REL_HARD);
 	rt_b->rt_period_timer.function = sched_rt_period_timer;
 }
 
diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
index 923a650e5c35..abf24e60b6e8 100644
--- a/kernel/time/hrtimer.c
+++ b/kernel/time/hrtimer.c
@@ -1135,7 +1135,9 @@ void hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim,
 	 * Check whether the HRTIMER_MODE_SOFT bit and hrtimer.is_soft
 	 * match.
 	 */
+#ifndef CONFIG_PREEMPT_RT_BASE
 	WARN_ON_ONCE(!(mode & HRTIMER_MODE_SOFT) ^ !timer->is_soft);
+#endif
 
 	base = lock_hrtimer_base(timer, &flags);
 
@@ -1295,10 +1297,17 @@ static inline int hrtimer_clockid_to_base(clockid_t clock_id)
 static void __hrtimer_init(struct hrtimer *timer, clockid_t clock_id,
 			   enum hrtimer_mode mode)
 {
-	bool softtimer = !!(mode & HRTIMER_MODE_SOFT);
-	int base = softtimer ? HRTIMER_MAX_CLOCK_BASES / 2 : 0;
+	bool softtimer;
+	int base;
 	struct hrtimer_cpu_base *cpu_base;
 
+	softtimer = !!(mode & HRTIMER_MODE_SOFT);
+#ifdef CONFIG_PREEMPT_RT_FULL
+	if (!softtimer && !(mode & HRTIMER_MODE_HARD))
+		softtimer = true;
+#endif
+	base = softtimer ? HRTIMER_MAX_CLOCK_BASES / 2 : 0;
+
 	memset(timer, 0, sizeof(struct hrtimer));
 
 	cpu_base = raw_cpu_ptr(&hrtimer_bases);
@@ -1681,6 +1690,14 @@ static void __hrtimer_init_sleeper(struct hrtimer_sleeper *sl,
 				   enum hrtimer_mode mode,
 				   struct task_struct *task)
 {
+#ifdef CONFIG_PREEMPT_RT_FULL
+	if (!(mode & (HRTIMER_MODE_SOFT | HRTIMER_MODE_HARD))) {
+		if (task_is_realtime(current) || system_state != SYSTEM_RUNNING)
+			mode |= HRTIMER_MODE_HARD;
+		else
+			mode |= HRTIMER_MODE_SOFT;
+	}
+#endif
 	__hrtimer_init(&sl->timer, clock_id, mode);
 	sl->timer.function = hrtimer_wakeup;
 	sl->task = task;
diff --git a/kernel/time/tick-broadcast-hrtimer.c b/kernel/time/tick-broadcast-hrtimer.c
index a836efd34589..c50e8f3262de 100644
--- a/kernel/time/tick-broadcast-hrtimer.c
+++ b/kernel/time/tick-broadcast-hrtimer.c
@@ -107,7 +107,7 @@ static enum hrtimer_restart bc_handler(struct hrtimer *t)
 
 void tick_setup_hrtimer_broadcast(void)
 {
-	hrtimer_init(&bctimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
+	hrtimer_init(&bctimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_HARD);
 	bctimer.function = bc_handler;
 	clockevents_register_device(&ce_broadcast_hrtimer);
 }
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 54fd344ef973..c217af74dddf 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -1310,7 +1310,7 @@ void tick_setup_sched_timer(void)
 	/*
 	 * Emulate tick processing via per-CPU hrtimers:
 	 */
-	hrtimer_init(&ts->sched_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
+	hrtimer_init(&ts->sched_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_HARD);
 	ts->sched_timer.function = tick_sched_timer;
 
 	/* Get the next period (per-CPU) */
diff --git a/kernel/watchdog.c b/kernel/watchdog.c
index bbc4940f21af..defd493ba967 100644
--- a/kernel/watchdog.c
+++ b/kernel/watchdog.c
@@ -483,7 +483,7 @@ static void watchdog_enable(unsigned int cpu)
 	 * Start the timer first to prevent the NMI watchdog triggering
 	 * before the timer has a chance to fire.
 	 */
-	hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD);
 	hrtimer->function = watchdog_timer_fn;
 	hrtimer_start(hrtimer, ns_to_ktime(sample_period),
 		      HRTIMER_MODE_REL_PINNED);
-- 
2.24.0