summaryrefslogtreecommitdiffstats
path: root/kernel/sched
diff options
context:
space:
mode:
Diffstat (limited to 'kernel/sched')
-rw-r--r--kernel/sched/core.c24
-rw-r--r--kernel/sched/fair.c66
-rw-r--r--kernel/sched/psi.c71
-rw-r--r--kernel/sched/rt.c2
-rw-r--r--kernel/sched/sched.h4
-rw-r--r--kernel/sched/stats.h8
-rw-r--r--kernel/sched/topology.c2
7 files changed, 114 insertions, 63 deletions
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index f12225f26b70..6fedf3a98581 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -5738,15 +5738,10 @@ int sched_cpu_activate(unsigned int cpu)
#ifdef CONFIG_SCHED_SMT
/*
- * The sched_smt_present static key needs to be evaluated on every
- * hotplug event because at boot time SMT might be disabled when
- * the number of booted CPUs is limited.
- *
- * If then later a sibling gets hotplugged, then the key would stay
- * off and SMT scheduling would never be functional.
+ * When going up, increment the number of cores with SMT present.
*/
- if (cpumask_weight(cpu_smt_mask(cpu)) > 1)
- static_branch_enable_cpuslocked(&sched_smt_present);
+ if (cpumask_weight(cpu_smt_mask(cpu)) == 2)
+ static_branch_inc_cpuslocked(&sched_smt_present);
#endif
set_cpu_active(cpu, true);
@@ -5790,6 +5785,14 @@ int sched_cpu_deactivate(unsigned int cpu)
*/
synchronize_rcu_mult(call_rcu, call_rcu_sched);
+#ifdef CONFIG_SCHED_SMT
+ /*
+ * When going down, decrement the number of cores with SMT present.
+ */
+ if (cpumask_weight(cpu_smt_mask(cpu)) == 2)
+ static_branch_dec_cpuslocked(&sched_smt_present);
+#endif
+
if (!sched_smp_initialized)
return 0;
@@ -5851,11 +5854,14 @@ void __init sched_init_smp(void)
/*
* There's no userspace yet to cause hotplug operations; hence all the
* CPU masks are stable and all blatant races in the below code cannot
- * happen.
+ * happen. The hotplug lock is nevertheless taken to satisfy lockdep,
+ * but there won't be any contention on it.
*/
+ cpus_read_lock();
mutex_lock(&sched_domains_mutex);
sched_init_domains(cpu_active_mask);
mutex_unlock(&sched_domains_mutex);
+ cpus_read_unlock();
/* Move init over to a non-isolated CPU */
if (set_cpus_allowed_ptr(current, housekeeping_cpumask(HK_FLAG_DOMAIN)) < 0)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index ee271bb661cc..ac855b2f4774 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -2400,8 +2400,8 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags)
local = 1;
/*
- * Retry task to preferred node migration periodically, in case it
- * case it previously failed, or the scheduler moved us.
+ * Retry to migrate task to preferred node periodically, in case it
+ * previously failed, or the scheduler moved us.
*/
if (time_after(jiffies, p->numa_migrate_retry)) {
task_numa_placement(p);
@@ -5674,11 +5674,11 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p,
return target;
}
-static unsigned long cpu_util_wake(int cpu, struct task_struct *p);
+static unsigned long cpu_util_without(int cpu, struct task_struct *p);
-static unsigned long capacity_spare_wake(int cpu, struct task_struct *p)
+static unsigned long capacity_spare_without(int cpu, struct task_struct *p)
{
- return max_t(long, capacity_of(cpu) - cpu_util_wake(cpu, p), 0);
+ return max_t(long, capacity_of(cpu) - cpu_util_without(cpu, p), 0);
}
/*
@@ -5738,7 +5738,7 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p,
avg_load += cfs_rq_load_avg(&cpu_rq(i)->cfs);
- spare_cap = capacity_spare_wake(i, p);
+ spare_cap = capacity_spare_without(i, p);
if (spare_cap > max_spare_cap)
max_spare_cap = spare_cap;
@@ -5889,8 +5889,8 @@ static inline int find_idlest_cpu(struct sched_domain *sd, struct task_struct *p
return prev_cpu;
/*
- * We need task's util for capacity_spare_wake, sync it up to prev_cpu's
- * last_update_time.
+ * We need task's util for capacity_spare_without, sync it up to
+ * prev_cpu's last_update_time.
*/
if (!(sd_flag & SD_BALANCE_FORK))
sync_entity_load_avg(&p->se);
@@ -6216,10 +6216,19 @@ static inline unsigned long cpu_util(int cpu)
}
/*
- * cpu_util_wake: Compute CPU utilization with any contributions from
- * the waking task p removed.
+ * cpu_util_without: compute cpu utilization without any contributions from *p
+ * @cpu: the CPU which utilization is requested
+ * @p: the task which utilization should be discounted
+ *
+ * The utilization of a CPU is defined by the utilization of tasks currently
+ * enqueued on that CPU as well as tasks which are currently sleeping after an
+ * execution on that CPU.
+ *
+ * This method returns the utilization of the specified CPU by discounting the
+ * utilization of the specified task, whenever the task is currently
+ * contributing to the CPU utilization.
*/
-static unsigned long cpu_util_wake(int cpu, struct task_struct *p)
+static unsigned long cpu_util_without(int cpu, struct task_struct *p)
{
struct cfs_rq *cfs_rq;
unsigned int util;
@@ -6231,7 +6240,7 @@ static unsigned long cpu_util_wake(int cpu, struct task_struct *p)
cfs_rq = &cpu_rq(cpu)->cfs;
util = READ_ONCE(cfs_rq->avg.util_avg);
- /* Discount task's blocked util from CPU's util */
+ /* Discount task's util from CPU's util */
util -= min_t(unsigned int, util, task_util(p));
/*
@@ -6240,14 +6249,14 @@ static unsigned long cpu_util_wake(int cpu, struct task_struct *p)
* a) if *p is the only task sleeping on this CPU, then:
* cpu_util (== task_util) > util_est (== 0)
* and thus we return:
- * cpu_util_wake = (cpu_util - task_util) = 0
+ * cpu_util_without = (cpu_util - task_util) = 0
*
* b) if other tasks are SLEEPING on this CPU, which is now exiting
* IDLE, then:
* cpu_util >= task_util
* cpu_util > util_est (== 0)
* and thus we discount *p's blocked utilization to return:
- * cpu_util_wake = (cpu_util - task_util) >= 0
+ * cpu_util_without = (cpu_util - task_util) >= 0
*
* c) if other tasks are RUNNABLE on that CPU and
* util_est > cpu_util
@@ -6260,8 +6269,33 @@ static unsigned long cpu_util_wake(int cpu, struct task_struct *p)
* covered by the following code when estimated utilization is
* enabled.
*/
- if (sched_feat(UTIL_EST))
- util = max(util, READ_ONCE(cfs_rq->avg.util_est.enqueued));
+ if (sched_feat(UTIL_EST)) {
+ unsigned int estimated =
+ READ_ONCE(cfs_rq->avg.util_est.enqueued);
+
+ /*
+ * Despite the following checks we still have a small window
+ * for a possible race, when an execl's select_task_rq_fair()
+ * races with LB's detach_task():
+ *
+ * detach_task()
+ * p->on_rq = TASK_ON_RQ_MIGRATING;
+ * ---------------------------------- A
+ * deactivate_task() \
+ * dequeue_task() + RaceTime
+ * util_est_dequeue() /
+ * ---------------------------------- B
+ *
+ * The additional check on "current == p" it's required to
+ * properly fix the execl regression and it helps in further
+ * reducing the chances for the above race.
+ */
+ if (unlikely(task_on_rq_queued(p) || current == p)) {
+ estimated -= min_t(unsigned int, estimated,
+ (_task_util_est(p) | UTIL_AVG_UNCHANGED));
+ }
+ util = max(util, estimated);
+ }
/*
* Utilization (estimated) can exceed the CPU capacity, thus let's
diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c
index 7cdecfc010af..fe24de3fbc93 100644
--- a/kernel/sched/psi.c
+++ b/kernel/sched/psi.c
@@ -136,8 +136,18 @@
static int psi_bug __read_mostly;
-bool psi_disabled __read_mostly;
-core_param(psi_disabled, psi_disabled, bool, 0644);
+DEFINE_STATIC_KEY_FALSE(psi_disabled);
+
+#ifdef CONFIG_PSI_DEFAULT_DISABLED
+bool psi_enable;
+#else
+bool psi_enable = true;
+#endif
+static int __init setup_psi(char *str)
+{
+ return kstrtobool(str, &psi_enable) == 0;
+}
+__setup("psi=", setup_psi);
/* Running averages - we need to be higher-res than loadavg */
#define PSI_FREQ (2*HZ+1) /* 2 sec intervals */
@@ -169,8 +179,10 @@ static void group_init(struct psi_group *group)
void __init psi_init(void)
{
- if (psi_disabled)
+ if (!psi_enable) {
+ static_branch_enable(&psi_disabled);
return;
+ }
psi_period = jiffies_to_nsecs(PSI_FREQ);
group_init(&psi_system);
@@ -549,7 +561,7 @@ void psi_memstall_enter(unsigned long *flags)
struct rq_flags rf;
struct rq *rq;
- if (psi_disabled)
+ if (static_branch_likely(&psi_disabled))
return;
*flags = current->flags & PF_MEMSTALL;
@@ -579,7 +591,7 @@ void psi_memstall_leave(unsigned long *flags)
struct rq_flags rf;
struct rq *rq;
- if (psi_disabled)
+ if (static_branch_likely(&psi_disabled))
return;
if (*flags)
@@ -600,7 +612,7 @@ void psi_memstall_leave(unsigned long *flags)
#ifdef CONFIG_CGROUPS
int psi_cgroup_alloc(struct cgroup *cgroup)
{
- if (psi_disabled)
+ if (static_branch_likely(&psi_disabled))
return 0;
cgroup->psi.pcpu = alloc_percpu(struct psi_group_cpu);
@@ -612,7 +624,7 @@ int psi_cgroup_alloc(struct cgroup *cgroup)
void psi_cgroup_free(struct cgroup *cgroup)
{
- if (psi_disabled)
+ if (static_branch_likely(&psi_disabled))
return;
cancel_delayed_work_sync(&cgroup->psi.clock_work);
@@ -633,38 +645,39 @@ void psi_cgroup_free(struct cgroup *cgroup)
*/
void cgroup_move_task(struct task_struct *task, struct css_set *to)
{
- bool move_psi = !psi_disabled;
unsigned int task_flags = 0;
struct rq_flags rf;
struct rq *rq;
- if (move_psi) {
- rq = task_rq_lock(task, &rf);
+ if (static_branch_likely(&psi_disabled)) {
+ /*
+ * Lame to do this here, but the scheduler cannot be locked
+ * from the outside, so we move cgroups from inside sched/.
+ */
+ rcu_assign_pointer(task->cgroups, to);
+ return;
+ }
- if (task_on_rq_queued(task))
- task_flags = TSK_RUNNING;
- else if (task->in_iowait)
- task_flags = TSK_IOWAIT;
+ rq = task_rq_lock(task, &rf);
- if (task->flags & PF_MEMSTALL)
- task_flags |= TSK_MEMSTALL;
+ if (task_on_rq_queued(task))
+ task_flags = TSK_RUNNING;
+ else if (task->in_iowait)
+ task_flags = TSK_IOWAIT;
- if (task_flags)
- psi_task_change(task, task_flags, 0);
- }
+ if (task->flags & PF_MEMSTALL)
+ task_flags |= TSK_MEMSTALL;
- /*
- * Lame to do this here, but the scheduler cannot be locked
- * from the outside, so we move cgroups from inside sched/.
- */
+ if (task_flags)
+ psi_task_change(task, task_flags, 0);
+
+ /* See comment above */
rcu_assign_pointer(task->cgroups, to);
- if (move_psi) {
- if (task_flags)
- psi_task_change(task, 0, task_flags);
+ if (task_flags)
+ psi_task_change(task, 0, task_flags);
- task_rq_unlock(rq, task, &rf);
- }
+ task_rq_unlock(rq, task, &rf);
}
#endif /* CONFIG_CGROUPS */
@@ -672,7 +685,7 @@ int psi_show(struct seq_file *m, struct psi_group *group, enum psi_res res)
{
int full;
- if (psi_disabled)
+ if (static_branch_likely(&psi_disabled))
return -EOPNOTSUPP;
update_stats(group);
diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index 2e2955a8cf8f..a21ea6021929 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -1561,7 +1561,7 @@ pick_next_task_rt(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
/*
* We may dequeue prev's rt_rq in put_prev_task().
- * So, we update time before rt_nr_running check.
+ * So, we update time before rt_queued check.
*/
if (prev->sched_class == &rt_sched_class)
update_curr_rt(rq);
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 618577fc9aa8..4e524ab589c9 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -23,6 +23,7 @@
#include <linux/sched/prio.h>
#include <linux/sched/rt.h>
#include <linux/sched/signal.h>
+#include <linux/sched/smt.h>
#include <linux/sched/stat.h>
#include <linux/sched/sysctl.h>
#include <linux/sched/task.h>
@@ -936,9 +937,6 @@ static inline int cpu_of(struct rq *rq)
#ifdef CONFIG_SCHED_SMT
-
-extern struct static_key_false sched_smt_present;
-
extern void __update_idle_core(struct rq *rq);
static inline void update_idle_core(struct rq *rq)
diff --git a/kernel/sched/stats.h b/kernel/sched/stats.h
index 4904c4677000..aa0de240fb41 100644
--- a/kernel/sched/stats.h
+++ b/kernel/sched/stats.h
@@ -66,7 +66,7 @@ static inline void psi_enqueue(struct task_struct *p, bool wakeup)
{
int clear = 0, set = TSK_RUNNING;
- if (psi_disabled)
+ if (static_branch_likely(&psi_disabled))
return;
if (!wakeup || p->sched_psi_wake_requeue) {
@@ -86,7 +86,7 @@ static inline void psi_dequeue(struct task_struct *p, bool sleep)
{
int clear = TSK_RUNNING, set = 0;
- if (psi_disabled)
+ if (static_branch_likely(&psi_disabled))
return;
if (!sleep) {
@@ -102,7 +102,7 @@ static inline void psi_dequeue(struct task_struct *p, bool sleep)
static inline void psi_ttwu_dequeue(struct task_struct *p)
{
- if (psi_disabled)
+ if (static_branch_likely(&psi_disabled))
return;
/*
* Is the task being migrated during a wakeup? Make sure to
@@ -128,7 +128,7 @@ static inline void psi_ttwu_dequeue(struct task_struct *p)
static inline void psi_task_tick(struct rq *rq)
{
- if (psi_disabled)
+ if (static_branch_likely(&psi_disabled))
return;
if (unlikely(rq->curr->flags & PF_MEMSTALL))
diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c
index 9d74371e4aad..8d7f15ba5916 100644
--- a/kernel/sched/topology.c
+++ b/kernel/sched/topology.c
@@ -1337,7 +1337,7 @@ void sched_init_numa(void)
int level = 0;
int i, j, k;
- sched_domains_numa_distance = kzalloc(sizeof(int) * nr_node_ids, GFP_KERNEL);
+ sched_domains_numa_distance = kzalloc(sizeof(int) * (nr_node_ids + 1), GFP_KERNEL);
if (!sched_domains_numa_distance)
return;
generated by cgit v1.2.3 (git 2.39.1) at 2025-03-30 21:24:42 +0000