1 files changed, 99 insertions, 98 deletions

diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index b93c8c3dc05a..475bb5998295 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -91,12 +91,6 @@ struct cpuidle_state;
 #include "cpupri.h"
 #include "cpudeadline.h"
 
-#ifdef CONFIG_SCHED_DEBUG
-# define SCHED_WARN_ON(x)      WARN_ONCE(x, #x)
-#else
-# define SCHED_WARN_ON(x)      ({ (void)(x), 0; })
-#endif
-
 /* task_struct::on_rq states: */
 #define TASK_ON_RQ_QUEUED	1
 #define TASK_ON_RQ_MIGRATING	2
@@ -819,15 +813,17 @@ struct rt_rq {
 
 #ifdef CONFIG_RT_GROUP_SCHED
 	int			rt_throttled;
-	u64			rt_time;
-	u64			rt_runtime;
+	u64			rt_time; /* consumed RT time, goes up in update_curr_rt */
+	u64			rt_runtime; /* allotted RT time, "slice" from rt_bandwidth, RT sharing/balancing */
 	/* Nests inside the rq lock: */
 	raw_spinlock_t		rt_runtime_lock;
 
 	unsigned int		rt_nr_boosted;
 
-	struct rq		*rq;
-	struct task_group	*tg;
+	struct rq		*rq; /* this is always top-level rq, cache? */
+#endif
+#ifdef CONFIG_CGROUP_SCHED
+	struct task_group	*tg; /* this tg has "this" rt_rq on given CPU for runnable entities */
 #endif
 };
 
@@ -998,7 +994,7 @@ struct root_domain {
 	 * Also, some corner cases, like 'wrap around' is dangerous, but given
 	 * that u64 is 'big enough'. So that shouldn't be a concern.
 	 */
-	u64 visit_gen;
+	u64 visit_cookie;
 
 #ifdef HAVE_RT_PUSH_IPI
 	/*
@@ -1180,10 +1176,8 @@ struct rq {
 
 	atomic_t		nr_iowait;
 
-#ifdef CONFIG_SCHED_DEBUG
 	u64 last_seen_need_resched_ns;
 	int ticks_without_resched;
-#endif
 
 #ifdef CONFIG_MEMBARRIER
 	int membarrier_state;
@@ -1506,6 +1500,23 @@ static inline bool sched_group_cookie_match(struct rq *rq,
 }
 
 #endif /* !CONFIG_SCHED_CORE */
+#ifdef CONFIG_RT_GROUP_SCHED
+# ifdef CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED
+DECLARE_STATIC_KEY_FALSE(rt_group_sched);
+static inline bool rt_group_sched_enabled(void)
+{
+	return static_branch_unlikely(&rt_group_sched);
+}
+# else
+DECLARE_STATIC_KEY_TRUE(rt_group_sched);
+static inline bool rt_group_sched_enabled(void)
+{
+	return static_branch_likely(&rt_group_sched);
+}
+# endif /* CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED */
+#else
+# define rt_group_sched_enabled()	false
+#endif /* CONFIG_RT_GROUP_SCHED */
 
 static inline void lockdep_assert_rq_held(struct rq *rq)
 {
@@ -1571,7 +1582,7 @@ static inline void update_idle_core(struct rq *rq) { }
 
 static inline struct task_struct *task_of(struct sched_entity *se)
 {
-	SCHED_WARN_ON(!entity_is_task(se));
+	WARN_ON_ONCE(!entity_is_task(se));
 	return container_of(se, struct task_struct, se);
 }
 
@@ -1652,7 +1663,7 @@ static inline void assert_clock_updated(struct rq *rq)
 	 * The only reason for not seeing a clock update since the
 	 * last rq_pin_lock() is if we're currently skipping updates.
 	 */
-	SCHED_WARN_ON(rq->clock_update_flags < RQCF_ACT_SKIP);
+	WARN_ON_ONCE(rq->clock_update_flags < RQCF_ACT_SKIP);
 }
 
 static inline u64 rq_clock(struct rq *rq)
@@ -1699,7 +1710,7 @@ static inline void rq_clock_cancel_skipupdate(struct rq *rq)
 static inline void rq_clock_start_loop_update(struct rq *rq)
 {
 	lockdep_assert_rq_held(rq);
-	SCHED_WARN_ON(rq->clock_update_flags & RQCF_ACT_SKIP);
+	WARN_ON_ONCE(rq->clock_update_flags & RQCF_ACT_SKIP);
 	rq->clock_update_flags |= RQCF_ACT_SKIP;
 }
 
@@ -1712,14 +1723,12 @@ static inline void rq_clock_stop_loop_update(struct rq *rq)
 struct rq_flags {
 	unsigned long flags;
 	struct pin_cookie cookie;
-#ifdef CONFIG_SCHED_DEBUG
 	/*
 	 * A copy of (rq::clock_update_flags & RQCF_UPDATED) for the
 	 * current pin context is stashed here in case it needs to be
 	 * restored in rq_repin_lock().
 	 */
 	unsigned int clock_update_flags;
-#endif
 };
 
 extern struct balance_callback balance_push_callback;
@@ -1727,10 +1736,10 @@ extern struct balance_callback balance_push_callback;
 #ifdef CONFIG_SCHED_CLASS_EXT
 extern const struct sched_class ext_sched_class;
 
-DECLARE_STATIC_KEY_FALSE(__scx_ops_enabled);	/* SCX BPF scheduler loaded */
+DECLARE_STATIC_KEY_FALSE(__scx_enabled);	/* SCX BPF scheduler loaded */
 DECLARE_STATIC_KEY_FALSE(__scx_switched_all);	/* all fair class tasks on SCX */
 
-#define scx_enabled()		static_branch_unlikely(&__scx_ops_enabled)
+#define scx_enabled()		static_branch_unlikely(&__scx_enabled)
 #define scx_switched_all()	static_branch_unlikely(&__scx_switched_all)
 
 static inline void scx_rq_clock_update(struct rq *rq, u64 clock)
@@ -1770,21 +1779,18 @@ static inline void rq_pin_lock(struct rq *rq, struct rq_flags *rf)
 {
 	rf->cookie = lockdep_pin_lock(__rq_lockp(rq));
 
-#ifdef CONFIG_SCHED_DEBUG
 	rq->clock_update_flags &= (RQCF_REQ_SKIP|RQCF_ACT_SKIP);
 	rf->clock_update_flags = 0;
-# ifdef CONFIG_SMP
-	SCHED_WARN_ON(rq->balance_callback && rq->balance_callback != &balance_push_callback);
-# endif
+#ifdef CONFIG_SMP
+	WARN_ON_ONCE(rq->balance_callback && rq->balance_callback != &balance_push_callback);
 #endif
 }
 
 static inline void rq_unpin_lock(struct rq *rq, struct rq_flags *rf)
 {
-#ifdef CONFIG_SCHED_DEBUG
 	if (rq->clock_update_flags > RQCF_ACT_SKIP)
 		rf->clock_update_flags = RQCF_UPDATED;
-#endif
+
 	scx_rq_clock_invalidate(rq);
 	lockdep_unpin_lock(__rq_lockp(rq), rf->cookie);
 }
@@ -1793,12 +1799,10 @@ static inline void rq_repin_lock(struct rq *rq, struct rq_flags *rf)
 {
 	lockdep_repin_lock(__rq_lockp(rq), rf->cookie);
 
-#ifdef CONFIG_SCHED_DEBUG
 	/*
 	 * Restore the value we stashed in @rf for this pin context.
 	 */
 	rq->clock_update_flags |= rf->clock_update_flags;
-#endif
 }
 
 extern
@@ -2072,9 +2076,7 @@ struct sched_group_capacity {
 	unsigned long		next_update;
 	int			imbalance;		/* XXX unrelated to capacity but shared group state */
 
-#ifdef CONFIG_SCHED_DEBUG
 	int			id;
-#endif
 
 	unsigned long		cpumask[];		/* Balance mask */
 };
@@ -2114,13 +2116,8 @@ static inline struct cpumask *group_balance_mask(struct sched_group *sg)
 
 extern int group_balance_cpu(struct sched_group *sg);
 
-#ifdef CONFIG_SCHED_DEBUG
 extern void update_sched_domain_debugfs(void);
 extern void dirty_sched_domain_sysctl(int cpu);
-#else
-static inline void update_sched_domain_debugfs(void) { }
-static inline void dirty_sched_domain_sysctl(int cpu) { }
-#endif
 
 extern int sched_update_scaling(void);
 
@@ -2168,6 +2165,13 @@ static inline void set_task_rq(struct task_struct *p, unsigned int cpu)
 #endif
 
 #ifdef CONFIG_RT_GROUP_SCHED
+	/*
+	 * p->rt.rt_rq is NULL initially and it is easier to assign
+	 * root_task_group's rt_rq than switching in rt_rq_of_se()
+	 * Clobbers tg(!)
+	 */
+	if (!rt_group_sched_enabled())
+		tg = &root_task_group;
 	p->rt.rt_rq  = tg->rt_rq[cpu];
 	p->rt.parent = tg->rt_se[cpu];
 #endif
@@ -2200,13 +2204,8 @@ static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
 }
 
 /*
- * Tunables that become constants when CONFIG_SCHED_DEBUG is off:
+ * Tunables:
  */
-#ifdef CONFIG_SCHED_DEBUG
-# define const_debug __read_mostly
-#else
-# define const_debug const
-#endif
 
 #define SCHED_FEAT(name, enabled)	\
 	__SCHED_FEAT_##name ,
@@ -2218,13 +2217,11 @@ enum {
 
 #undef SCHED_FEAT
 
-#ifdef CONFIG_SCHED_DEBUG
-
 /*
  * To support run-time toggling of sched features, all the translation units
  * (but core.c) reference the sysctl_sched_features defined in core.c.
  */
-extern const_debug unsigned int sysctl_sched_features;
+extern __read_mostly unsigned int sysctl_sched_features;
 
 #ifdef CONFIG_JUMP_LABEL
 
@@ -2246,24 +2243,6 @@ extern struct static_key sched_feat_keys[__SCHED_FEAT_NR];
 
 #endif /* !CONFIG_JUMP_LABEL */
 
-#else /* !SCHED_DEBUG: */
-
-/*
- * Each translation unit has its own copy of sysctl_sched_features to allow
- * constants propagation at compile time and compiler optimization based on
- * features default.
- */
-#define SCHED_FEAT(name, enabled)	\
-	(1UL << __SCHED_FEAT_##name) * enabled |
-static const_debug __maybe_unused unsigned int sysctl_sched_features =
-#include "features.h"
-	0;
-#undef SCHED_FEAT
-
-#define sched_feat(x) !!(sysctl_sched_features & (1UL << __SCHED_FEAT_##x))
-
-#endif /* !SCHED_DEBUG */
-
 extern struct static_key_false sched_numa_balancing;
 extern struct static_key_false sched_schedstats;
 
@@ -2685,7 +2664,7 @@ static inline void idle_set_state(struct rq *rq,
 
 static inline struct cpuidle_state *idle_get_state(struct rq *rq)
 {
-	SCHED_WARN_ON(!rcu_read_lock_held());
+	WARN_ON_ONCE(!rcu_read_lock_held());
 
 	return rq->idle_state;
 }
@@ -2843,12 +2822,11 @@ extern void wakeup_preempt(struct rq *rq, struct task_struct *p, int flags);
 # define SCHED_NR_MIGRATE_BREAK 32
 #endif
 
-extern const_debug unsigned int sysctl_sched_nr_migrate;
-extern const_debug unsigned int sysctl_sched_migration_cost;
+extern __read_mostly unsigned int sysctl_sched_nr_migrate;
+extern __read_mostly unsigned int sysctl_sched_migration_cost;
 
 extern unsigned int sysctl_sched_base_slice;
 
-#ifdef CONFIG_SCHED_DEBUG
 extern int sysctl_resched_latency_warn_ms;
 extern int sysctl_resched_latency_warn_once;
 
@@ -2859,7 +2837,6 @@ extern unsigned int sysctl_numa_balancing_scan_period_min;
 extern unsigned int sysctl_numa_balancing_scan_period_max;
 extern unsigned int sysctl_numa_balancing_scan_size;
 extern unsigned int sysctl_numa_balancing_hot_threshold;
-#endif
 
 #ifdef CONFIG_SCHED_HRTICK
 
@@ -2932,7 +2909,6 @@ unsigned long arch_scale_freq_capacity(int cpu)
 }
 #endif
 
-#ifdef CONFIG_SCHED_DEBUG
 /*
  * In double_lock_balance()/double_rq_lock(), we use raw_spin_rq_lock() to
  * acquire rq lock instead of rq_lock(). So at the end of these two functions
@@ -2947,9 +2923,6 @@ static inline void double_rq_clock_clear_update(struct rq *rq1, struct rq *rq2)
 	rq2->clock_update_flags &= (RQCF_REQ_SKIP|RQCF_ACT_SKIP);
 #endif
 }
-#else
-static inline void double_rq_clock_clear_update(struct rq *rq1, struct rq *rq2) { }
-#endif
 
 #define DEFINE_LOCK_GUARD_2(name, type, _lock, _unlock, ...)				\
 __DEFINE_UNLOCK_GUARD(name, type, _unlock, type *lock2; __VA_ARGS__)			\
@@ -3162,7 +3135,6 @@ extern struct sched_entity *__pick_root_entity(struct cfs_rq *cfs_rq);
 extern struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq);
 extern struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq);
 
-#ifdef	CONFIG_SCHED_DEBUG
 extern bool sched_debug_verbose;
 
 extern void print_cfs_stats(struct seq_file *m, int cpu);
@@ -3173,15 +3145,12 @@ extern void print_rt_rq(struct seq_file *m, int cpu, struct rt_rq *rt_rq);
 extern void print_dl_rq(struct seq_file *m, int cpu, struct dl_rq *dl_rq);
 
 extern void resched_latency_warn(int cpu, u64 latency);
-# ifdef CONFIG_NUMA_BALANCING
+#ifdef CONFIG_NUMA_BALANCING
 extern void show_numa_stats(struct task_struct *p, struct seq_file *m);
 extern void
 print_numa_stats(struct seq_file *m, int node, unsigned long tsf,
 		 unsigned long tpf, unsigned long gsf, unsigned long gpf);
-# endif /* CONFIG_NUMA_BALANCING */
-#else /* !CONFIG_SCHED_DEBUG: */
-static inline void resched_latency_warn(int cpu, u64 latency) { }
-#endif /* !CONFIG_SCHED_DEBUG */
+#endif /* CONFIG_NUMA_BALANCING */
 
 extern void init_cfs_rq(struct cfs_rq *cfs_rq);
 extern void init_rt_rq(struct rt_rq *rt_rq);
@@ -3259,11 +3228,11 @@ struct irqtime {
 };
 
 DECLARE_PER_CPU(struct irqtime, cpu_irqtime);
-DECLARE_STATIC_KEY_FALSE(sched_clock_irqtime);
+extern int sched_clock_irqtime;
 
 static inline int irqtime_enabled(void)
 {
-	return static_branch_likely(&sched_clock_irqtime);
+	return sched_clock_irqtime;
 }
 
 /*
@@ -3394,6 +3363,31 @@ static inline bool update_other_load_avgs(struct rq *rq) { return false; }
 
 unsigned long uclamp_eff_value(struct task_struct *p, enum uclamp_id clamp_id);
 
+/*
+ * When uclamp is compiled in, the aggregation at rq level is 'turned off'
+ * by default in the fast path and only gets turned on once userspace performs
+ * an operation that requires it.
+ *
+ * Returns true if userspace opted-in to use uclamp and aggregation at rq level
+ * hence is active.
+ */
+static inline bool uclamp_is_used(void)
+{
+	return static_branch_likely(&sched_uclamp_used);
+}
+
+/*
+ * Enabling static branches would get the cpus_read_lock(),
+ * check whether uclamp_is_used before enable it to avoid always
+ * calling cpus_read_lock(). Because we never disable this
+ * static key once enable it.
+ */
+static inline void sched_uclamp_enable(void)
+{
+	if (!uclamp_is_used())
+		static_branch_enable(&sched_uclamp_used);
+}
+
 static inline unsigned long uclamp_rq_get(struct rq *rq,
 					  enum uclamp_id clamp_id)
 {
@@ -3417,7 +3411,7 @@ static inline bool uclamp_rq_is_capped(struct rq *rq)
 	unsigned long rq_util;
 	unsigned long max_util;
 
-	if (!static_branch_likely(&sched_uclamp_used))
+	if (!uclamp_is_used())
 		return false;
 
 	rq_util = cpu_util_cfs(cpu_of(rq)) + cpu_util_rt(rq);
@@ -3426,19 +3420,6 @@ static inline bool uclamp_rq_is_capped(struct rq *rq)
 	return max_util != SCHED_CAPACITY_SCALE && rq_util >= max_util;
 }
 
-/*
- * When uclamp is compiled in, the aggregation at rq level is 'turned off'
- * by default in the fast path and only gets turned on once userspace performs
- * an operation that requires it.
- *
- * Returns true if userspace opted-in to use uclamp and aggregation at rq level
- * hence is active.
- */
-static inline bool uclamp_is_used(void)
-{
-	return static_branch_likely(&sched_uclamp_used);
-}
-
 #define for_each_clamp_id(clamp_id) \
 	for ((clamp_id) = 0; (clamp_id) < UCLAMP_CNT; (clamp_id)++)
 
@@ -3486,6 +3467,8 @@ static inline bool uclamp_is_used(void)
 	return false;
 }
 
+static inline void sched_uclamp_enable(void) {}
+
 static inline unsigned long
 uclamp_rq_get(struct rq *rq, enum uclamp_id clamp_id)
 {
@@ -3552,8 +3535,6 @@ static inline bool sched_energy_enabled(void)
 	return static_branch_unlikely(&sched_energy_present);
 }
 
-extern struct cpufreq_governor schedutil_gov;
-
 #else /* ! (CONFIG_ENERGY_MODEL && CONFIG_CPU_FREQ_GOV_SCHEDUTIL) */
 
 #define perf_domain_span(pd) NULL
@@ -3619,6 +3600,7 @@ extern int preempt_dynamic_mode;
 extern int sched_dynamic_mode(const char *str);
 extern void sched_dynamic_update(int mode);
 #endif
+extern const char *preempt_modes[];
 
 #ifdef CONFIG_SCHED_MM_CID
 
@@ -3698,10 +3680,28 @@ static inline int __mm_cid_try_get(struct task_struct *t, struct mm_struct *mm)
 {
 	struct cpumask *cidmask = mm_cidmask(mm);
 	struct mm_cid __percpu *pcpu_cid = mm->pcpu_cid;
-	int cid = __this_cpu_read(pcpu_cid->recent_cid);
+	int cid, max_nr_cid, allowed_max_nr_cid;
 
+	/*
+	 * After shrinking the number of threads or reducing the number
+	 * of allowed cpus, reduce the value of max_nr_cid so expansion
+	 * of cid allocation will preserve cache locality if the number
+	 * of threads or allowed cpus increase again.
+	 */
+	max_nr_cid = atomic_read(&mm->max_nr_cid);
+	while ((allowed_max_nr_cid = min_t(int, READ_ONCE(mm->nr_cpus_allowed),
+					   atomic_read(&mm->mm_users))),
+	       max_nr_cid > allowed_max_nr_cid) {
+		/* atomic_try_cmpxchg loads previous mm->max_nr_cid into max_nr_cid. */
+		if (atomic_try_cmpxchg(&mm->max_nr_cid, &max_nr_cid, allowed_max_nr_cid)) {
+			max_nr_cid = allowed_max_nr_cid;
+			break;
+		}
+	}
 	/* Try to re-use recent cid. This improves cache locality. */
-	if (!mm_cid_is_unset(cid) && !cpumask_test_and_set_cpu(cid, cidmask))
+	cid = __this_cpu_read(pcpu_cid->recent_cid);
+	if (!mm_cid_is_unset(cid) && cid < max_nr_cid &&
+	    !cpumask_test_and_set_cpu(cid, cidmask))
 		return cid;
 	/*
 	 * Expand cid allocation if the maximum number of concurrency
@@ -3709,8 +3709,9 @@ static inline int __mm_cid_try_get(struct task_struct *t, struct mm_struct *mm)
 	 * and number of threads. Expanding cid allocation as much as
 	 * possible improves cache locality.
 	 */
-	cid = atomic_read(&mm->max_nr_cid);
+	cid = max_nr_cid;
 	while (cid < READ_ONCE(mm->nr_cpus_allowed) && cid < atomic_read(&mm->mm_users)) {
+		/* atomic_try_cmpxchg loads previous mm->max_nr_cid into cid. */
 		if (!atomic_try_cmpxchg(&mm->max_nr_cid, &cid, cid + 1))
 			continue;
 		if (!cpumask_test_and_set_cpu(cid, cidmask))