[GIT pull] scheduler fixes for 4.17

From: Thomas Gleixner
Date: Sun Jun 03 2018 - 05:23:11 EST


Linus,

please pull the latest sched-urgent-for-linus git tree from:

git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git sched-urgent-for-linus

A set of scheduler fixes:

- Two patches addressing the problem that the scheduler allows under
certain conditions user space tasks to be scheduled on CPUs which are
not yet fully booted which causes a few subtle and hard to debug issue.

- Add a missing runqueue clock update in the deadline scheduler which
triggers a warning under certain circumstances.

- Fix a silly typo in the scheduler header file.

Thanks,

tglx

------------------>
Davidlohr Bueso (1):
sched/headers: Fix typo

Juri Lelli (1):
sched/deadline: Fix missing clock update

Paul Burton (1):
sched/core: Require cpu_active() in select_task_rq(), for user tasks

Peter Zijlstra (1):
sched/core: Fix rules for running on online && !active CPUs


kernel/sched/core.c | 45 +++++++++++++++++++++++++++++++--------------
kernel/sched/deadline.c | 6 +++---
kernel/sched/sched.h | 2 +-
3 files changed, 35 insertions(+), 18 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 092f7c4de903..211890edf37e 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -881,6 +881,33 @@ void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags)
}

#ifdef CONFIG_SMP
+
+static inline bool is_per_cpu_kthread(struct task_struct *p)
+{
+ if (!(p->flags & PF_KTHREAD))
+ return false;
+
+ if (p->nr_cpus_allowed != 1)
+ return false;
+
+ return true;
+}
+
+/*
+ * Per-CPU kthreads are allowed to run on !actie && online CPUs, see
+ * __set_cpus_allowed_ptr() and select_fallback_rq().
+ */
+static inline bool is_cpu_allowed(struct task_struct *p, int cpu)
+{
+ if (!cpumask_test_cpu(cpu, &p->cpus_allowed))
+ return false;
+
+ if (is_per_cpu_kthread(p))
+ return cpu_online(cpu);
+
+ return cpu_active(cpu);
+}
+
/*
* This is how migration works:
*
@@ -938,16 +965,8 @@ struct migration_arg {
static struct rq *__migrate_task(struct rq *rq, struct rq_flags *rf,
struct task_struct *p, int dest_cpu)
{
- if (p->flags & PF_KTHREAD) {
- if (unlikely(!cpu_online(dest_cpu)))
- return rq;
- } else {
- if (unlikely(!cpu_active(dest_cpu)))
- return rq;
- }
-
/* Affinity changed (again). */
- if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed))
+ if (!is_cpu_allowed(p, dest_cpu))
return rq;

update_rq_clock(rq);
@@ -1476,10 +1495,9 @@ static int select_fallback_rq(int cpu, struct task_struct *p)
for (;;) {
/* Any allowed, online CPU? */
for_each_cpu(dest_cpu, &p->cpus_allowed) {
- if (!(p->flags & PF_KTHREAD) && !cpu_active(dest_cpu))
- continue;
- if (!cpu_online(dest_cpu))
+ if (!is_cpu_allowed(p, dest_cpu))
continue;
+
goto out;
}

@@ -1542,8 +1560,7 @@ int select_task_rq(struct task_struct *p, int cpu, int sd_flags, int wake_flags)
* [ this allows ->select_task() to simply return task_cpu(p) and
* not worry about this generic constraint ]
*/
- if (unlikely(!cpumask_test_cpu(cpu, &p->cpus_allowed) ||
- !cpu_online(cpu)))
+ if (unlikely(!is_cpu_allowed(p, cpu)))
cpu = select_fallback_rq(task_cpu(p), p);

return cpu;
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index 1356afd1eeb6..fbfc3f1d368a 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -1259,6 +1259,9 @@ static enum hrtimer_restart inactive_task_timer(struct hrtimer *timer)

rq = task_rq_lock(p, &rf);

+ sched_clock_tick();
+ update_rq_clock(rq);
+
if (!dl_task(p) || p->state == TASK_DEAD) {
struct dl_bw *dl_b = dl_bw_of(task_cpu(p));

@@ -1278,9 +1281,6 @@ static enum hrtimer_restart inactive_task_timer(struct hrtimer *timer)
if (dl_se->dl_non_contending == 0)
goto unlock;

- sched_clock_tick();
- update_rq_clock(rq);
-
sub_running_bw(dl_se, &rq->dl);
dl_se->dl_non_contending = 0;
unlock:
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 1f0a4bc6a39d..cb467c221b15 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -983,7 +983,7 @@ static inline void rq_clock_skip_update(struct rq *rq)
}

/*
- * See rt task throttoling, which is the only time a skip
+ * See rt task throttling, which is the only time a skip
* request is cancelled.
*/
static inline void rq_clock_cancel_skipupdate(struct rq *rq)