Merge branch 'timers/nohz-posix-timers-v2' of git://git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-dynticks into timers/nohz

Pull posix cpu timers handling on full dynticks from Frederic Weisbecker.

Signed-off-by: Ingo Molnar <mingo@kernel.org>

3 files changed, 112 insertions, 16 deletions

diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index 8fd709c..84d5cb3 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -10,6 +10,8 @@
 #include <linux/kernel_stat.h>
 #include <trace/events/timer.h>
 #include <linux/random.h>
+#include <linux/tick.h>
+#include <linux/workqueue.h>
 
 /*
  * Called after updating RLIMIT_CPU to run cpu timer and update
@@ -153,6 +155,21 @@ static void bump_cpu_timer(struct k_itimer *timer,
 	}
 }
 
+/**
+ * task_cputime_zero - Check a task_cputime struct for all zero fields.
+ *
+ * @cputime:	The struct to compare.
+ *
+ * Checks @cputime to see if all fields are zero.  Returns true if all fields
+ * are zero, false if any field is nonzero.
+ */
+static inline int task_cputime_zero(const struct task_cputime *cputime)
+{
+	if (!cputime->utime && !cputime->stime && !cputime->sum_exec_runtime)
+		return 1;
+	return 0;
+}
+
 static inline cputime_t prof_ticks(struct task_struct *p)
 {
 	cputime_t utime, stime;
@@ -636,6 +653,37 @@ static int cpu_timer_sample_group(const clockid_t which_clock,
 	return 0;
 }
 
+#ifdef CONFIG_NO_HZ_FULL
+static void nohz_kick_work_fn(struct work_struct *work)
+{
+	tick_nohz_full_kick_all();
+}
+
+static DECLARE_WORK(nohz_kick_work, nohz_kick_work_fn);
+
+/*
+ * We need the IPIs to be sent from sane process context.
+ * The posix cpu timers are always set with irqs disabled.
+ */
+static void posix_cpu_timer_kick_nohz(void)
+{
+	schedule_work(&nohz_kick_work);
+}
+
+bool posix_cpu_timers_can_stop_tick(struct task_struct *tsk)
+{
+	if (!task_cputime_zero(&tsk->cputime_expires))
+		return true;
+
+	if (tsk->signal->cputimer.running)
+		return true;
+
+	return false;
+}
+#else
+static inline void posix_cpu_timer_kick_nohz(void) { }
+#endif
+
 /*
  * Guts of sys_timer_settime for CPU timers.
  * This is called with the timer locked and interrupts disabled.
@@ -794,6 +842,8 @@ static int posix_cpu_timer_set(struct k_itimer *timer, int flags,
 		sample_to_timespec(timer->it_clock,
 				   old_incr, &old->it_interval);
 	}
+	if (!ret)
+		posix_cpu_timer_kick_nohz();
 	return ret;
 }
 
@@ -1008,21 +1058,6 @@ static void check_cpu_itimer(struct task_struct *tsk, struct cpu_itimer *it,
 	}
 }
 
-/**
- * task_cputime_zero - Check a task_cputime struct for all zero fields.
- *
- * @cputime:	The struct to compare.
- *
- * Checks @cputime to see if all fields are zero.  Returns true if all fields
- * are zero, false if any field is nonzero.
- */
-static inline int task_cputime_zero(const struct task_cputime *cputime)
-{
-	if (!cputime->utime && !cputime->stime && !cputime->sum_exec_runtime)
-		return 1;
-	return 0;
-}
-
 /*
  * Check for any per-thread CPU timers that have fired and move them
  * off the tsk->*_timers list onto the firing list.  Per-thread timers
@@ -1336,6 +1371,13 @@ void run_posix_cpu_timers(struct task_struct *tsk)
 			cpu_timer_fire(timer);
 		spin_unlock(&timer->it_lock);
 	}
+
+	/*
+	 * In case some timers were rescheduled after the queue got emptied,
+	 * wake up full dynticks CPUs.
+	 */
+	if (tsk->signal->cputimer.running)
+		posix_cpu_timer_kick_nohz();
 }
 
 /*
@@ -1366,7 +1408,7 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx,
 		}
 
 		if (!*newval)
-			return;
+			goto out;
 		*newval += now.cpu;
 	}
 
@@ -1384,6 +1426,8 @@ void set_process_cpu_timer(struct task_struct *tsk, unsigned int clock_idx,
 			tsk->signal->cputime_expires.virt_exp = *newval;
 		break;
 	}
+out:
+	posix_cpu_timer_kick_nohz();
 }
 
 static int do_cpu_nanosleep(const clockid_t which_clock, int flags,
diff --git a/kernel/time/Kconfig b/kernel/time/Kconfig
index 99c3f13..f6a792a 100644
--- a/kernel/time/Kconfig
+++ b/kernel/time/Kconfig
@@ -111,6 +111,7 @@ config NO_HZ_FULL
 	select RCU_USER_QS
 	select RCU_NOCB_CPU
 	select CONTEXT_TRACKING_FORCE
+	select IRQ_WORK
 	help
 	 Adaptively try to shutdown the tick whenever possible, even when
 	 the CPU is running tasks. Typically this requires running a single
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index a76e090..884a9f3 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -147,6 +147,57 @@ static void tick_sched_handle(struct tick_sched *ts, struct pt_regs *regs)
 static cpumask_var_t nohz_full_mask;
 bool have_nohz_full_mask;
 
+/*
+ * Re-evaluate the need for the tick on the current CPU
+ * and restart it if necessary.
+ */
+static void tick_nohz_full_check(void)
+{
+	/*
+	 * STUB for now, will be filled with the full tick stop/restart
+	 * infrastructure patches
+	 */
+}
+
+static void nohz_full_kick_work_func(struct irq_work *work)
+{
+	tick_nohz_full_check();
+}
+
+static DEFINE_PER_CPU(struct irq_work, nohz_full_kick_work) = {
+	.func = nohz_full_kick_work_func,
+};
+
+/*
+ * Kick the current CPU if it's full dynticks in order to force it to
+ * re-evaluate its dependency on the tick and restart it if necessary.
+ */
+void tick_nohz_full_kick(void)
+{
+	if (tick_nohz_full_cpu(smp_processor_id()))
+		irq_work_queue(&__get_cpu_var(nohz_full_kick_work));
+}
+
+static void nohz_full_kick_ipi(void *info)
+{
+	tick_nohz_full_check();
+}
+
+/*
+ * Kick all full dynticks CPUs in order to force these to re-evaluate
+ * their dependency on the tick and restart it if necessary.
+ */
+void tick_nohz_full_kick_all(void)
+{
+	if (!have_nohz_full_mask)
+		return;
+
+	preempt_disable();
+	smp_call_function_many(nohz_full_mask,
+			       nohz_full_kick_ipi, NULL, false);
+	preempt_enable();
+}
+
 int tick_nohz_full_cpu(int cpu)
 {
 	if (!have_nohz_full_mask)