6 files changed, 221 insertions, 107 deletions

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 2b037f195473..7fa8e74ad2ab 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -3486,8 +3486,36 @@ void scheduler_tick(void)
 
 struct tick_work {
 	int			cpu;
+	atomic_t		state;
 	struct delayed_work	work;
 };
+/* Values for ->state, see diagram below. */
+#define TICK_SCHED_REMOTE_OFFLINE	0
+#define TICK_SCHED_REMOTE_OFFLINING	1
+#define TICK_SCHED_REMOTE_RUNNING	2
+
+/*
+ * State diagram for ->state:
+ *
+ *
+ *          TICK_SCHED_REMOTE_OFFLINE
+ *                    |   ^
+ *                    |   |
+ *                    |   | sched_tick_remote()
+ *                    |   |
+ *                    |   |
+ *                    +--TICK_SCHED_REMOTE_OFFLINING
+ *                    |   ^
+ *                    |   |
+ * sched_tick_start() |   | sched_tick_stop()
+ *                    |   |
+ *                    V   |
+ *          TICK_SCHED_REMOTE_RUNNING
+ *
+ *
+ * Other transitions get WARN_ON_ONCE(), except that sched_tick_remote()
+ * and sched_tick_start() are happy to leave the state in RUNNING.
+ */
 
 static struct tick_work __percpu *tick_work_cpu;
 
@@ -3500,6 +3528,7 @@ static void sched_tick_remote(struct work_struct *work)
 	struct task_struct *curr;
 	struct rq_flags rf;
 	u64 delta;
+	int os;
 
 	/*
 	 * Handle the tick only if it appears the remote CPU is running in full
@@ -3513,7 +3542,7 @@ static void sched_tick_remote(struct work_struct *work)
 
 	rq_lock_irq(rq, &rf);
 	curr = rq->curr;
-	if (is_idle_task(curr))
+	if (is_idle_task(curr) || cpu_is_offline(cpu))
 		goto out_unlock;
 
 	update_rq_clock(rq);
@@ -3533,13 +3562,18 @@ out_requeue:
 	/*
 	 * Run the remote tick once per second (1Hz). This arbitrary
 	 * frequency is large enough to avoid overload but short enough
-	 * to keep scheduler internal stats reasonably up to date.
+	 * to keep scheduler internal stats reasonably up to date.  But
+	 * first update state to reflect hotplug activity if required.
 	 */
-	queue_delayed_work(system_unbound_wq, dwork, HZ);
+	os = atomic_fetch_add_unless(&twork->state, -1, TICK_SCHED_REMOTE_RUNNING);
+	WARN_ON_ONCE(os == TICK_SCHED_REMOTE_OFFLINE);
+	if (os == TICK_SCHED_REMOTE_RUNNING)
+		queue_delayed_work(system_unbound_wq, dwork, HZ);
 }
 
 static void sched_tick_start(int cpu)
 {
+	int os;
 	struct tick_work *twork;
 
 	if (housekeeping_cpu(cpu, HK_FLAG_TICK))
@@ -3548,15 +3582,20 @@ static void sched_tick_start(int cpu)
 	WARN_ON_ONCE(!tick_work_cpu);
 
 	twork = per_cpu_ptr(tick_work_cpu, cpu);
-	twork->cpu = cpu;
-	INIT_DELAYED_WORK(&twork->work, sched_tick_remote);
-	queue_delayed_work(system_unbound_wq, &twork->work, HZ);
+	os = atomic_xchg(&twork->state, TICK_SCHED_REMOTE_RUNNING);
+	WARN_ON_ONCE(os == TICK_SCHED_REMOTE_RUNNING);
+	if (os == TICK_SCHED_REMOTE_OFFLINE) {
+		twork->cpu = cpu;
+		INIT_DELAYED_WORK(&twork->work, sched_tick_remote);
+		queue_delayed_work(system_unbound_wq, &twork->work, HZ);
+	}
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
 static void sched_tick_stop(int cpu)
 {
 	struct tick_work *twork;
+	int os;
 
 	if (housekeeping_cpu(cpu, HK_FLAG_TICK))
 		return;
@@ -3564,7 +3603,10 @@ static void sched_tick_stop(int cpu)
 	WARN_ON_ONCE(!tick_work_cpu);
 
 	twork = per_cpu_ptr(tick_work_cpu, cpu);
-	cancel_delayed_work_sync(&twork->work);
+	/* There cannot be competing actions, but don't rely on stop-machine. */
+	os = atomic_xchg(&twork->state, TICK_SCHED_REMOTE_OFFLINING);
+	WARN_ON_ONCE(os != TICK_SCHED_REMOTE_RUNNING);
+	/* Don't cancel, as this would mess up the state machine. */
 }
 #endif /* CONFIG_HOTPLUG_CPU */
 
@@ -3572,7 +3614,6 @@ int __init sched_tick_offload_init(void)
 {
 	tick_work_cpu = alloc_percpu(struct tick_work);
 	BUG_ON(!tick_work_cpu);
-
 	return 0;
 }
 
@@ -3904,7 +3945,7 @@ void __noreturn do_task_dead(void)
 
 static inline void sched_submit_work(struct task_struct *tsk)
 {
-	if (!tsk->state || tsk_is_pi_blocked(tsk))
+	if (!tsk->state)
 		return;
 
 	/*
@@ -3920,6 +3961,9 @@ static inline void sched_submit_work(struct task_struct *tsk)
 		preempt_enable_no_resched();
 	}
 
+	if (tsk_is_pi_blocked(tsk))
+		return;
+
 	/*
 	 * If we are going to sleep and we have plugged IO queued,
 	 * make sure to submit it to avoid deadlocks.
@@ -5102,37 +5146,40 @@ out_unlock:
 	return retval;
 }
 
-static int sched_read_attr(struct sched_attr __user *uattr,
-			   struct sched_attr *attr,
-			   unsigned int usize)
+/*
+ * Copy the kernel size attribute structure (which might be larger
+ * than what user-space knows about) to user-space.
+ *
+ * Note that all cases are valid: user-space buffer can be larger or
+ * smaller than the kernel-space buffer. The usual case is that both
+ * have the same size.
+ */
+static int
+sched_attr_copy_to_user(struct sched_attr __user *uattr,
+			struct sched_attr *kattr,
+			unsigned int usize)
 {
-	int ret;
+	unsigned int ksize = sizeof(*kattr);
 
 	if (!access_ok(uattr, usize))
 		return -EFAULT;
 
 	/*
-	 * If we're handed a smaller struct than we know of,
-	 * ensure all the unknown bits are 0 - i.e. old
-	 * user-space does not get uncomplete information.
+	 * sched_getattr() ABI forwards and backwards compatibility:
+	 *
+	 * If usize == ksize then we just copy everything to user-space and all is good.
+	 *
+	 * If usize < ksize then we only copy as much as user-space has space for,
+	 * this keeps ABI compatibility as well. We skip the rest.
+	 *
+	 * If usize > ksize then user-space is using a newer version of the ABI,
+	 * which part the kernel doesn't know about. Just ignore it - tooling can
+	 * detect the kernel's knowledge of attributes from the attr->size value
+	 * which is set to ksize in this case.
 	 */
-	if (usize < sizeof(*attr)) {
-		unsigned char *addr;
-		unsigned char *end;
+	kattr->size = min(usize, ksize);
 
-		addr = (void *)attr + usize;
-		end  = (void *)attr + sizeof(*attr);
-
-		for (; addr < end; addr++) {
-			if (*addr)
-				return -EFBIG;
-		}
-
-		attr->size = usize;
-	}
-
-	ret = copy_to_user(uattr, attr, attr->size);
-	if (ret)
+	if (copy_to_user(uattr, kattr, kattr->size))
 		return -EFAULT;
 
 	return 0;
@@ -5142,20 +5189,18 @@ static int sched_read_attr(struct sched_attr __user *uattr,
  * sys_sched_getattr - similar to sched_getparam, but with sched_attr
  * @pid: the pid in question.
  * @uattr: structure containing the extended parameters.
- * @size: sizeof(attr) for fwd/bwd comp.
+ * @usize: sizeof(attr) that user-space knows about, for forwards and backwards compatibility.
  * @flags: for future extension.
  */
 SYSCALL_DEFINE4(sched_getattr, pid_t, pid, struct sched_attr __user *, uattr,
-		unsigned int, size, unsigned int, flags)
+		unsigned int, usize, unsigned int, flags)
 {
-	struct sched_attr attr = {
-		.size = sizeof(struct sched_attr),
-	};
+	struct sched_attr kattr = { };
 	struct task_struct *p;
 	int retval;
 
-	if (!uattr || pid < 0 || size > PAGE_SIZE ||
-	    size < SCHED_ATTR_SIZE_VER0 || flags)
+	if (!uattr || pid < 0 || usize > PAGE_SIZE ||
+	    usize < SCHED_ATTR_SIZE_VER0 || flags)
 		return -EINVAL;
 
 	rcu_read_lock();
@@ -5168,25 +5213,24 @@ SYSCALL_DEFINE4(sched_getattr, pid_t, pid, struct sched_attr __user *, uattr,
 	if (retval)
 		goto out_unlock;
 
-	attr.sched_policy = p->policy;
+	kattr.sched_policy = p->policy;
 	if (p->sched_reset_on_fork)
-		attr.sched_flags |= SCHED_FLAG_RESET_ON_FORK;
+		kattr.sched_flags |= SCHED_FLAG_RESET_ON_FORK;
 	if (task_has_dl_policy(p))
-		__getparam_dl(p, &attr);
+		__getparam_dl(p, &kattr);
 	else if (task_has_rt_policy(p))
-		attr.sched_priority = p->rt_priority;
+		kattr.sched_priority = p->rt_priority;
 	else
-		attr.sched_nice = task_nice(p);
+		kattr.sched_nice = task_nice(p);
 
 #ifdef CONFIG_UCLAMP_TASK
-	attr.sched_util_min = p->uclamp_req[UCLAMP_MIN].value;
-	attr.sched_util_max = p->uclamp_req[UCLAMP_MAX].value;
+	kattr.sched_util_min = p->uclamp_req[UCLAMP_MIN].value;
+	kattr.sched_util_max = p->uclamp_req[UCLAMP_MAX].value;
 #endif
 
 	rcu_read_unlock();
 
-	retval = sched_read_attr(uattr, &attr, size);
-	return retval;
+	return sched_attr_copy_to_user(uattr, &kattr, usize);
 
 out_unlock:
 	rcu_read_unlock();
diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index 636ca6f88c8e..867b4bb6d4be 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -40,6 +40,7 @@ struct sugov_policy {
 	struct task_struct	*thread;
 	bool			work_in_progress;
 
+	bool			limits_changed;
 	bool			need_freq_update;
 };
 
@@ -89,8 +90,11 @@ static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 time)
 	    !cpufreq_this_cpu_can_update(sg_policy->policy))
 		return false;
 
-	if (unlikely(sg_policy->need_freq_update))
+	if (unlikely(sg_policy->limits_changed)) {
+		sg_policy->limits_changed = false;
+		sg_policy->need_freq_update = true;
 		return true;
+	}
 
 	delta_ns = time - sg_policy->last_freq_update_time;
 
@@ -437,7 +441,7 @@ static inline bool sugov_cpu_is_busy(struct sugov_cpu *sg_cpu) { return false; }
 static inline void ignore_dl_rate_limit(struct sugov_cpu *sg_cpu, struct sugov_policy *sg_policy)
 {
 	if (cpu_bw_dl(cpu_rq(sg_cpu->cpu)) > sg_cpu->bw_dl)
-		sg_policy->need_freq_update = true;
+		sg_policy->limits_changed = true;
 }
 
 static void sugov_update_single(struct update_util_data *hook, u64 time,
@@ -457,7 +461,8 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
 	if (!sugov_should_update_freq(sg_policy, time))
 		return;
 
-	busy = sugov_cpu_is_busy(sg_cpu);
+	/* Limits may have changed, don't skip frequency update */
+	busy = !sg_policy->need_freq_update && sugov_cpu_is_busy(sg_cpu);
 
 	util = sugov_get_util(sg_cpu);
 	max = sg_cpu->max;
@@ -831,6 +836,7 @@ static int sugov_start(struct cpufreq_policy *policy)
 	sg_policy->last_freq_update_time	= 0;
 	sg_policy->next_freq			= 0;
 	sg_policy->work_in_progress		= false;
+	sg_policy->limits_changed		= false;
 	sg_policy->need_freq_update		= false;
 	sg_policy->cached_raw_freq		= 0;
 
@@ -879,7 +885,7 @@ static void sugov_limits(struct cpufreq_policy *policy)
 		mutex_unlock(&sg_policy->work_lock);
 	}
 
-	sg_policy->need_freq_update = true;
+	sg_policy->limits_changed = true;
 }
 
 struct cpufreq_governor schedutil_gov = {
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index ef5b9f6b1d42..46122edd8552 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -2088,17 +2088,13 @@ retry:
 	}
 
 	deactivate_task(rq, next_task, 0);
-	sub_running_bw(&next_task->dl, &rq->dl);
-	sub_rq_bw(&next_task->dl, &rq->dl);
 	set_task_cpu(next_task, later_rq->cpu);
-	add_rq_bw(&next_task->dl, &later_rq->dl);
 
 	/*
 	 * Update the later_rq clock here, because the clock is used
 	 * by the cpufreq_update_util() inside __add_running_bw().
 	 */
 	update_rq_clock(later_rq);
-	add_running_bw(&next_task->dl, &later_rq->dl);
 	activate_task(later_rq, next_task, ENQUEUE_NOCLOCK);
 	ret = 1;
 
@@ -2186,11 +2182,7 @@ static void pull_dl_task(struct rq *this_rq)
 			resched = true;
 
 			deactivate_task(src_rq, p, 0);
-			sub_running_bw(&p->dl, &src_rq->dl);
-			sub_rq_bw(&p->dl, &src_rq->dl);
 			set_task_cpu(p, this_cpu);
-			add_rq_bw(&p->dl, &this_rq->dl);
-			add_running_bw(&p->dl, &this_rq->dl);
 			activate_task(this_rq, p, 0);
 			dmin = p->dl.deadline;
 
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 036be95a87e9..500f5db0de0b 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -1086,6 +1086,21 @@ struct numa_group {
 	unsigned long faults[0];
 };
 
+/*
+ * For functions that can be called in multiple contexts that permit reading
+ * ->numa_group (see struct task_struct for locking rules).
+ */
+static struct numa_group *deref_task_numa_group(struct task_struct *p)
+{
+	return rcu_dereference_check(p->numa_group, p == current ||
+		(lockdep_is_held(&task_rq(p)->lock) && !READ_ONCE(p->on_cpu)));
+}
+
+static struct numa_group *deref_curr_numa_group(struct task_struct *p)
+{
+	return rcu_dereference_protected(p->numa_group, p == current);
+}
+
 static inline unsigned long group_faults_priv(struct numa_group *ng);
 static inline unsigned long group_faults_shared(struct numa_group *ng);
 
@@ -1129,10 +1144,12 @@ static unsigned int task_scan_start(struct task_struct *p)
 {
 	unsigned long smin = task_scan_min(p);
 	unsigned long period = smin;
+	struct numa_group *ng;
 
 	/* Scale the maximum scan period with the amount of shared memory. */
-	if (p->numa_group) {
-		struct numa_group *ng = p->numa_group;
+	rcu_read_lock();
+	ng = rcu_dereference(p->numa_group);
+	if (ng) {
 		unsigned long shared = group_faults_shared(ng);
 		unsigned long private = group_faults_priv(ng);
 
@@ -1140,6 +1157,7 @@ static unsigned int task_scan_start(struct task_struct *p)
 		period *= shared + 1;
 		period /= private + shared + 1;
 	}
+	rcu_read_unlock();
 
 	return max(smin, period);
 }
@@ -1148,13 +1166,14 @@ static unsigned int task_scan_max(struct task_struct *p)
 {
 	unsigned long smin = task_scan_min(p);
 	unsigned long smax;
+	struct numa_group *ng;
 
 	/* Watch for min being lower than max due to floor calculations */
 	smax = sysctl_numa_balancing_scan_period_max / task_nr_scan_windows(p);
 
 	/* Scale the maximum scan period with the amount of shared memory. */
-	if (p->numa_group) {
-		struct numa_group *ng = p->numa_group;
+	ng = deref_curr_numa_group(p);
+	if (ng) {
 		unsigned long shared = group_faults_shared(ng);
 		unsigned long private = group_faults_priv(ng);
 		unsigned long period = smax;
@@ -1186,7 +1205,7 @@ void init_numa_balancing(unsigned long clone_flags, struct task_struct *p)
 	p->numa_scan_period		= sysctl_numa_balancing_scan_delay;
 	p->numa_work.next		= &p->numa_work;
 	p->numa_faults			= NULL;
-	p->numa_group			= NULL;
+	RCU_INIT_POINTER(p->numa_group, NULL);
 	p->last_task_numa_placement	= 0;
 	p->last_sum_exec_runtime	= 0;
 
@@ -1233,7 +1252,16 @@ static void account_numa_dequeue(struct rq *rq, struct task_struct *p)
 
 pid_t task_numa_group_id(struct task_struct *p)
 {
-	return p->numa_group ? p->numa_group->gid : 0;
+	struct numa_group *ng;
+	pid_t gid = 0;
+
+	rcu_read_lock();
+	ng = rcu_dereference(p->numa_group);
+	if (ng)
+		gid = ng->gid;
+	rcu_read_unlock();
+
+	return gid;
 }
 
 /*
@@ -1258,11 +1286,13 @@ static inline unsigned long task_faults(struct task_struct *p, int nid)
 
 static inline unsigned long group_faults(struct task_struct *p, int nid)
 {
-	if (!p->numa_group)
+	struct numa_group *ng = deref_task_numa_group(p);
+
+	if (!ng)
 		return 0;
 
-	return p->numa_group->faults[task_faults_idx(NUMA_MEM, nid, 0)] +
-		p->numa_group->faults[task_faults_idx(NUMA_MEM, nid, 1)];
+	return ng->faults[task_faults_idx(NUMA_MEM, nid, 0)] +
+		ng->faults[task_faults_idx(NUMA_MEM, nid, 1)];
 }
 
 static inline unsigned long group_faults_cpu(struct numa_group *group, int nid)
@@ -1400,12 +1430,13 @@ static inline unsigned long task_weight(struct task_struct *p, int nid,
 static inline unsigned long group_weight(struct task_struct *p, int nid,
 					 int dist)
 {
+	struct numa_group *ng = deref_task_numa_group(p);
 	unsigned long faults, total_faults;
 
-	if (!p->numa_group)
+	if (!ng)
 		return 0;
 
-	total_faults = p->numa_group->total_faults;
+	total_faults = ng->total_faults;
 
 	if (!total_faults)
 		return 0;
@@ -1419,7 +1450,7 @@ static inline unsigned long group_weight(struct task_struct *p, int nid,
 bool should_numa_migrate_memory(struct task_struct *p, struct page * page,
 				int src_nid, int dst_cpu)
 {
-	struct numa_group *ng = p->numa_group;
+	struct numa_group *ng = deref_curr_numa_group(p);
 	int dst_nid = cpu_to_node(dst_cpu);
 	int last_cpupid, this_cpupid;
 
@@ -1600,13 +1631,14 @@ static bool load_too_imbalanced(long src_load, long dst_load,
 static void task_numa_compare(struct task_numa_env *env,
 			      long taskimp, long groupimp, bool maymove)
 {
+	struct numa_group *cur_ng, *p_ng = deref_curr_numa_group(env->p);
 	struct rq *dst_rq = cpu_rq(env->dst_cpu);
+	long imp = p_ng ? groupimp : taskimp;
 	struct task_struct *cur;
 	long src_load, dst_load;
-	long load;
-	long imp = env->p->numa_group ? groupimp : taskimp;
-	long moveimp = imp;
 	int dist = env->dist;
+	long moveimp = imp;
+	long load;
 
 	if (READ_ONCE(dst_rq->numa_migrate_on))
 		return;
@@ -1645,21 +1677,22 @@ static void task_numa_compare(struct task_numa_env *env,
 	 * If dst and source tasks are in the same NUMA group, or not
 	 * in any group then look only at task weights.
 	 */
-	if (cur->numa_group == env->p->numa_group) {
+	cur_ng = rcu_dereference(cur->numa_group);
+	if (cur_ng == p_ng) {
 		imp = taskimp + task_weight(cur, env->src_nid, dist) -
 		      task_weight(cur, env->dst_nid, dist);
 		/*
 		 * Add some hysteresis to prevent swapping the
 		 * tasks within a group over tiny differences.
 		 */
-		if (cur->numa_group)
+		if (cur_ng)
 			imp -= imp / 16;
 	} else {
 		/*
 		 * Compare the group weights. If a task is all by itself
 		 * (not part of a group), use the task weight instead.
 		 */
-		if (cur->numa_group && env->p->numa_group)
+		if (cur_ng && p_ng)
 			imp += group_weight(cur, env->src_nid, dist) -
 			       group_weight(cur, env->dst_nid, dist);
 		else
@@ -1757,11 +1790,12 @@ static int task_numa_migrate(struct task_struct *p)
 		.best_imp = 0,
 		.best_cpu = -1,
 	};
+	unsigned long taskweight, groupweight;
 	struct sched_domain *sd;
+	long taskimp, groupimp;
+	struct numa_group *ng;
 	struct rq *best_rq;
-	unsigned long taskweight, groupweight;
 	int nid, ret, dist;
-	long taskimp, groupimp;
 
 	/*
 	 * Pick the lowest SD_NUMA domain, as that would have the smallest
@@ -1807,7 +1841,8 @@ static int task_numa_migrate(struct task_struct *p)
 	 *   multiple NUMA nodes; in order to better consolidate the group,
 	 *   we need to check other locations.
 	 */
-	if (env.best_cpu == -1 || (p->numa_group && p->numa_group->active_nodes > 1)) {
+	ng = deref_curr_numa_group(p);
+	if (env.best_cpu == -1 || (ng && ng->active_nodes > 1)) {
 		for_each_online_node(nid) {
 			if (nid == env.src_nid || nid == p->numa_preferred_nid)
 				continue;
@@ -1840,7 +1875,7 @@ static int task_numa_migrate(struct task_struct *p)
 	 * A task that migrated to a second choice node will be better off
 	 * trying for a better one later. Do not set the preferred node here.
 	 */
-	if (p->numa_group) {
+	if (ng) {
 		if (env.best_cpu == -1)
 			nid = env.src_nid;
 		else
@@ -2135,6 +2170,7 @@ static void task_numa_placement(struct task_struct *p)
 	unsigned long total_faults;
 	u64 runtime, period;
 	spinlock_t *group_lock = NULL;
+	struct numa_group *ng;
 
 	/*
 	 * The p->mm->numa_scan_seq field gets updated without
@@ -2152,8 +2188,9 @@ static void task_numa_placement(struct task_struct *p)
 	runtime = numa_get_avg_runtime(p, &period);
 
 	/* If the task is part of a group prevent parallel updates to group stats */
-	if (p->numa_group) {
-		group_lock = &p->numa_group->lock;
+	ng = deref_curr_numa_group(p);
+	if (ng) {
+		group_lock = &ng->lock;
 		spin_lock_irq(group_lock);
 	}
 
@@ -2194,7 +2231,7 @@ static void task_numa_placement(struct task_struct *p)
 			p->numa_faults[cpu_idx] += f_diff;
 			faults += p->numa_faults[mem_idx];
 			p->total_numa_faults += diff;
-			if (p->numa_group) {
+			if (ng) {
 				/*
 				 * safe because we can only change our own group
 				 *
@@ -2202,14 +2239,14 @@ static void task_numa_placement(struct task_struct *p)
 				 * nid and priv in a specific region because it
 				 * is at the beginning of the numa_faults array.
 				 */
-				p->numa_group->faults[mem_idx] += diff;
-				p->numa_group->faults_cpu[mem_idx] += f_diff;
-				p->numa_group->total_faults += diff;
-				group_faults += p->numa_group->faults[mem_idx];
+				ng->faults[mem_idx] += diff;
+				ng->faults_cpu[mem_idx] += f_diff;
+				ng->total_faults += diff;
+				group_faults += ng->faults[mem_idx];
 			}
 		}
 
-		if (!p->numa_group) {
+		if (!ng) {
 			if (faults > max_faults) {
 				max_faults = faults;
 				max_nid = nid;
@@ -2220,8 +2257,8 @@ static void task_numa_placement(struct task_struct *p)
 		}
 	}
 
-	if (p->numa_group) {
-		numa_group_count_active_nodes(p->numa_group);
+	if (ng) {
+		numa_group_count_active_nodes(ng);
 		spin_unlock_irq(group_lock);
 		max_nid = preferred_group_nid(p, max_nid);
 	}
@@ -2255,7 +2292,7 @@ static void task_numa_group(struct task_struct *p, int cpupid, int flags,
 	int cpu = cpupid_to_cpu(cpupid);
 	int i;
 
-	if (unlikely(!p->numa_group)) {
+	if (unlikely(!deref_curr_numa_group(p))) {
 		unsigned int size = sizeof(struct numa_group) +
 				    4*nr_node_ids*sizeof(unsigned long);
 
@@ -2291,7 +2328,7 @@ static void task_numa_group(struct task_struct *p, int cpupid, int flags,
 	if (!grp)
 		goto no_join;
 
-	my_grp = p->numa_group;
+	my_grp = deref_curr_numa_group(p);
 	if (grp == my_grp)
 		goto no_join;
 
@@ -2353,13 +2390,24 @@ no_join:
 	return;
 }
 
-void task_numa_free(struct task_struct *p)
+/*
+ * Get rid of NUMA staticstics associated with a task (either current or dead).
+ * If @final is set, the task is dead and has reached refcount zero, so we can
+ * safely free all relevant data structures. Otherwise, there might be
+ * concurrent reads from places like load balancing and procfs, and we should
+ * reset the data back to default state without freeing ->numa_faults.
+ */
+void task_numa_free(struct task_struct *p, bool final)
 {
-	struct numa_group *grp = p->numa_group;
-	void *numa_faults = p->numa_faults;
+	/* safe: p either is current or is being freed by current */
+	struct numa_group *grp = rcu_dereference_raw(p->numa_group);
+	unsigned long *numa_faults = p->numa_faults;
 	unsigned long flags;
 	int i;
 
+	if (!numa_faults)
+		return;
+
 	if (grp) {
 		spin_lock_irqsave(&grp->lock, flags);
 		for (i = 0; i < NR_NUMA_HINT_FAULT_STATS * nr_node_ids; i++)
@@ -2372,8 +2420,14 @@ void task_numa_free(struct task_struct *p)
 		put_numa_group(grp);
 	}
 
-	p->numa_faults = NULL;
-	kfree(numa_faults);
+	if (final) {
+		p->numa_faults = NULL;
+		kfree(numa_faults);
+	} else {
+		p->total_numa_faults = 0;
+		for (i = 0; i < NR_NUMA_HINT_FAULT_STATS * nr_node_ids; i++)
+			numa_faults[i] = 0;
+	}
 }
 
 /*
@@ -2426,7 +2480,7 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags)
 	 * actively using should be counted as local. This allows the
 	 * scan rate to slow down when a workload has settled down.
 	 */
-	ng = p->numa_group;
+	ng = deref_curr_numa_group(p);
 	if (!priv && !local && ng && ng->active_nodes > 1 &&
 				numa_is_active_node(cpu_node, ng) &&
 				numa_is_active_node(mem_node, ng))
@@ -4416,6 +4470,8 @@ static void __account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec)
 	if (likely(cfs_rq->runtime_remaining > 0))
 		return;
 
+	if (cfs_rq->throttled)
+		return;
 	/*
 	 * if we're unable to extend our runtime we resched so that the active
 	 * hierarchy can be throttled
@@ -4619,6 +4675,9 @@ static u64 distribute_cfs_runtime(struct cfs_bandwidth *cfs_b,
 		if (!cfs_rq_throttled(cfs_rq))
 			goto next;
 
+		/* By the above check, this should never be true */
+		SCHED_WARN_ON(cfs_rq->runtime_remaining > 0);
+
 		runtime = -cfs_rq->runtime_remaining + 1;
 		if (runtime > remaining)
 			runtime = remaining;
@@ -10444,18 +10503,22 @@ void show_numa_stats(struct task_struct *p, struct seq_file *m)
 {
 	int node;
 	unsigned long tsf = 0, tpf = 0, gsf = 0, gpf = 0;
+	struct numa_group *ng;
 
+	rcu_read_lock();
+	ng = rcu_dereference(p->numa_group);
 	for_each_online_node(node) {
 		if (p->numa_faults) {
 			tsf = p->numa_faults[task_faults_idx(NUMA_MEM, node, 0)];
 			tpf = p->numa_faults[task_faults_idx(NUMA_MEM, node, 1)];
 		}
-		if (p->numa_group) {
-			gsf = p->numa_group->faults[task_faults_idx(NUMA_MEM, node, 0)],
-			gpf = p->numa_group->faults[task_faults_idx(NUMA_MEM, node, 1)];
+		if (ng) {
+			gsf = ng->faults[task_faults_idx(NUMA_MEM, node, 0)],
+			gpf = ng->faults[task_faults_idx(NUMA_MEM, node, 1)];
 		}
 		print_numa_stats(m, node, tsf, tpf, gsf, gpf);
 	}
+	rcu_read_unlock();
 }
 #endif /* CONFIG_NUMA_BALANCING */
 #endif /* CONFIG_SCHED_DEBUG */
diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c
index 80940939b733..e4bc4aa739b8 100644
--- a/kernel/sched/idle.c
+++ b/kernel/sched/idle.c
@@ -241,13 +241,14 @@ static void do_idle(void)
 		check_pgt_cache();
 		rmb();
 
+		local_irq_disable();
+
 		if (cpu_is_offline(cpu)) {
-			tick_nohz_idle_stop_tick_protected();
+			tick_nohz_idle_stop_tick();
 			cpuhp_report_idle_dead();
 			arch_cpu_idle_dead();
 		}
 
-		local_irq_disable();
 		arch_cpu_idle_enter();
 
 		/*
diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c
index 7acc632c3b82..6e52b67b420e 100644
--- a/kernel/sched/psi.c
+++ b/kernel/sched/psi.c
@@ -1051,7 +1051,7 @@ struct psi_trigger *psi_trigger_create(struct psi_group *group,
 
 	if (!rcu_access_pointer(group->poll_kworker)) {
 		struct sched_param param = {
-			.sched_priority = MAX_RT_PRIO - 1,
+			.sched_priority = 1,
 		};
 		struct kthread_worker *kworker;
 
@@ -1061,7 +1061,7 @@ struct psi_trigger *psi_trigger_create(struct psi_group *group,
 			mutex_unlock(&group->trigger_lock);
 			return ERR_CAST(kworker);
 		}
-		sched_setscheduler(kworker->task, SCHED_FIFO, &param);
+		sched_setscheduler_nocheck(kworker->task, SCHED_FIFO, &param);
 		kthread_init_delayed_work(&group->poll_work,
 				psi_poll_work);
 		rcu_assign_pointer(group->poll_kworker, kworker);
@@ -1131,7 +1131,15 @@ static void psi_trigger_destroy(struct kref *ref)
 	 * deadlock while waiting for psi_poll_work to acquire trigger_lock
 	 */
 	if (kworker_to_destroy) {
+		/*
+		 * After the RCU grace period has expired, the worker
+		 * can no longer be found through group->poll_kworker.
+		 * But it might have been already scheduled before
+		 * that - deschedule it cleanly before destroying it.
+		 */
 		kthread_cancel_delayed_work_sync(&group->poll_work);
+		atomic_set(&group->poll_scheduled, 0);
+
 		kthread_destroy_worker(kworker_to_destroy);
 	}
 	kfree(t);