aboutsummaryrefslogtreecommitdiff
path: root/kernel/sched
diff options
context:
space:
mode:
authorThomas Gleixner2020-12-15 10:48:07 +0100
committerThomas Gleixner2020-12-15 10:48:07 +0100
commit3c41e57a1e168d879e923c5583adeae47eec9f64 (patch)
treee6272012c4b766189be2821316a3d23d115f5195 /kernel/sched
parentd14ce74f1fb376ccbbc0b05ded477ada51253729 (diff)
parent2f5fbc4305d07725bfebaedb09e57271315691ef (diff)
Merge tag 'irqchip-5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/maz/arm-platforms into irq/core
Pull irqchip updates for 5.11 from Marc Zyngier: - Preliminary support for managed interrupts on platform devices - Correctly identify allocation of MSIs proxyied by another device - Remove the fasteoi IPI flow which has been proved useless - Generalise the Ocelot support to new SoCs - Improve GICv4.1 vcpu entry, matching the corresponding KVM optimisation - Work around spurious interrupts on Qualcomm PDC - Random fixes and cleanups Link: https://lore.kernel.org/r/20201212135626.1479884-1-maz@kernel.org
Diffstat (limited to 'kernel/sched')
-rw-r--r--kernel/sched/core.c26
-rw-r--r--kernel/sched/cpufreq_schedutil.c24
-rw-r--r--kernel/sched/deadline.c97
-rw-r--r--kernel/sched/debug.c12
-rw-r--r--kernel/sched/fair.c73
-rw-r--r--kernel/sched/idle.c28
6 files changed, 160 insertions, 100 deletions
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index d2003a7d5ab5..e7e453492cff 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2501,7 +2501,12 @@ ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags,
#ifdef CONFIG_SMP
if (wake_flags & WF_MIGRATED)
en_flags |= ENQUEUE_MIGRATED;
+ else
#endif
+ if (p->in_iowait) {
+ delayacct_blkio_end(p);
+ atomic_dec(&task_rq(p)->nr_iowait);
+ }
activate_task(rq, p, en_flags);
ttwu_do_wakeup(rq, p, wake_flags, rf);
@@ -2888,11 +2893,6 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
if (READ_ONCE(p->on_rq) && ttwu_runnable(p, wake_flags))
goto unlock;
- if (p->in_iowait) {
- delayacct_blkio_end(p);
- atomic_dec(&task_rq(p)->nr_iowait);
- }
-
#ifdef CONFIG_SMP
/*
* Ensure we load p->on_cpu _after_ p->on_rq, otherwise it would be
@@ -2963,6 +2963,11 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
cpu = select_task_rq(p, p->wake_cpu, SD_BALANCE_WAKE, wake_flags);
if (task_cpu(p) != cpu) {
+ if (p->in_iowait) {
+ delayacct_blkio_end(p);
+ atomic_dec(&task_rq(p)->nr_iowait);
+ }
+
wake_flags |= WF_MIGRATED;
psi_ttwu_dequeue(p);
set_task_cpu(p, cpu);
@@ -4907,20 +4912,21 @@ void rt_mutex_setprio(struct task_struct *p, struct task_struct *pi_task)
if (!dl_prio(p->normal_prio) ||
(pi_task && dl_prio(pi_task->prio) &&
dl_entity_preempt(&pi_task->dl, &p->dl))) {
- p->dl.dl_boosted = 1;
+ p->dl.pi_se = pi_task->dl.pi_se;
queue_flag |= ENQUEUE_REPLENISH;
- } else
- p->dl.dl_boosted = 0;
+ } else {
+ p->dl.pi_se = &p->dl;
+ }
p->sched_class = &dl_sched_class;
} else if (rt_prio(prio)) {
if (dl_prio(oldprio))
- p->dl.dl_boosted = 0;
+ p->dl.pi_se = &p->dl;
if (oldprio < prio)
queue_flag |= ENQUEUE_HEAD;
p->sched_class = &rt_sched_class;
} else {
if (dl_prio(oldprio))
- p->dl.dl_boosted = 0;
+ p->dl.pi_se = &p->dl;
if (rt_prio(oldprio))
p->rt.timeout = 0;
p->sched_class = &fair_sched_class;
diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
index c03a5775d019..97d318b0cd0c 100644
--- a/kernel/sched/cpufreq_schedutil.c
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -102,9 +102,12 @@ static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 time)
static bool sugov_update_next_freq(struct sugov_policy *sg_policy, u64 time,
unsigned int next_freq)
{
- if (sg_policy->next_freq == next_freq &&
- !cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS))
- return false;
+ if (!sg_policy->need_freq_update) {
+ if (sg_policy->next_freq == next_freq)
+ return false;
+ } else {
+ sg_policy->need_freq_update = cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS);
+ }
sg_policy->next_freq = next_freq;
sg_policy->last_freq_update_time = time;
@@ -162,11 +165,9 @@ static unsigned int get_next_freq(struct sugov_policy *sg_policy,
freq = map_util_freq(util, freq, max);
- if (freq == sg_policy->cached_raw_freq && !sg_policy->need_freq_update &&
- !cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS))
+ if (freq == sg_policy->cached_raw_freq && !sg_policy->need_freq_update)
return sg_policy->next_freq;
- sg_policy->need_freq_update = false;
sg_policy->cached_raw_freq = freq;
return cpufreq_driver_resolve_freq(policy, freq);
}
@@ -442,7 +443,6 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
struct sugov_policy *sg_policy = sg_cpu->sg_policy;
unsigned long util, max;
unsigned int next_f;
- bool busy;
unsigned int cached_freq = sg_policy->cached_raw_freq;
sugov_iowait_boost(sg_cpu, time, flags);
@@ -453,9 +453,6 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
if (!sugov_should_update_freq(sg_policy, time))
return;
- /* 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;
util = sugov_iowait_apply(sg_cpu, time, util, max);
@@ -464,7 +461,7 @@ static void sugov_update_single(struct update_util_data *hook, u64 time,
* Do not reduce the frequency if the CPU has not been idle
* recently, as the reduction is likely to be premature then.
*/
- if (busy && next_f < sg_policy->next_freq) {
+ if (sugov_cpu_is_busy(sg_cpu) && next_f < sg_policy->next_freq) {
next_f = sg_policy->next_freq;
/* Restore cached freq as next_freq has changed */
@@ -829,9 +826,10 @@ static int sugov_start(struct cpufreq_policy *policy)
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;
+ sg_policy->need_freq_update = cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS);
+
for_each_cpu(cpu, policy->cpus) {
struct sugov_cpu *sg_cpu = &per_cpu(sugov_cpu, cpu);
@@ -883,7 +881,7 @@ static void sugov_limits(struct cpufreq_policy *policy)
struct cpufreq_governor schedutil_gov = {
.name = "schedutil",
.owner = THIS_MODULE,
- .dynamic_switching = true,
+ .flags = CPUFREQ_GOV_DYNAMIC_SWITCHING,
.init = sugov_init,
.exit = sugov_exit,
.start = sugov_start,
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index f232305dcefe..1d3c97268ec0 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -43,6 +43,28 @@ static inline int on_dl_rq(struct sched_dl_entity *dl_se)
return !RB_EMPTY_NODE(&dl_se->rb_node);
}
+#ifdef CONFIG_RT_MUTEXES
+static inline struct sched_dl_entity *pi_of(struct sched_dl_entity *dl_se)
+{
+ return dl_se->pi_se;
+}
+
+static inline bool is_dl_boosted(struct sched_dl_entity *dl_se)
+{
+ return pi_of(dl_se) != dl_se;
+}
+#else
+static inline struct sched_dl_entity *pi_of(struct sched_dl_entity *dl_se)
+{
+ return dl_se;
+}
+
+static inline bool is_dl_boosted(struct sched_dl_entity *dl_se)
+{
+ return false;
+}
+#endif
+
#ifdef CONFIG_SMP
static inline struct dl_bw *dl_bw_of(int i)
{
@@ -698,7 +720,7 @@ static inline void setup_new_dl_entity(struct sched_dl_entity *dl_se)
struct dl_rq *dl_rq = dl_rq_of_se(dl_se);
struct rq *rq = rq_of_dl_rq(dl_rq);
- WARN_ON(dl_se->dl_boosted);
+ WARN_ON(is_dl_boosted(dl_se));
WARN_ON(dl_time_before(rq_clock(rq), dl_se->deadline));
/*
@@ -736,21 +758,20 @@ static inline void setup_new_dl_entity(struct sched_dl_entity *dl_se)
* could happen are, typically, a entity voluntarily trying to overcome its
* runtime, or it just underestimated it during sched_setattr().
*/
-static void replenish_dl_entity(struct sched_dl_entity *dl_se,
- struct sched_dl_entity *pi_se)
+static void replenish_dl_entity(struct sched_dl_entity *dl_se)
{
struct dl_rq *dl_rq = dl_rq_of_se(dl_se);
struct rq *rq = rq_of_dl_rq(dl_rq);
- BUG_ON(pi_se->dl_runtime <= 0);
+ BUG_ON(pi_of(dl_se)->dl_runtime <= 0);
/*
* This could be the case for a !-dl task that is boosted.
* Just go with full inherited parameters.
*/
if (dl_se->dl_deadline == 0) {
- dl_se->deadline = rq_clock(rq) + pi_se->dl_deadline;
- dl_se->runtime = pi_se->dl_runtime;
+ dl_se->deadline = rq_clock(rq) + pi_of(dl_se)->dl_deadline;
+ dl_se->runtime = pi_of(dl_se)->dl_runtime;
}
if (dl_se->dl_yielded && dl_se->runtime > 0)
@@ -763,8 +784,8 @@ static void replenish_dl_entity(struct sched_dl_entity *dl_se,
* arbitrary large.
*/
while (dl_se->runtime <= 0) {
- dl_se->deadline += pi_se->dl_period;
- dl_se->runtime += pi_se->dl_runtime;
+ dl_se->deadline += pi_of(dl_se)->dl_period;
+ dl_se->runtime += pi_of(dl_se)->dl_runtime;
}
/*
@@ -778,8 +799,8 @@ static void replenish_dl_entity(struct sched_dl_entity *dl_se,
*/
if (dl_time_before(dl_se->deadline, rq_clock(rq))) {
printk_deferred_once("sched: DL replenish lagged too much\n");
- dl_se->deadline = rq_clock(rq) + pi_se->dl_deadline;
- dl_se->runtime = pi_se->dl_runtime;
+ dl_se->deadline = rq_clock(rq) + pi_of(dl_se)->dl_deadline;
+ dl_se->runtime = pi_of(dl_se)->dl_runtime;
}
if (dl_se->dl_yielded)
@@ -812,8 +833,7 @@ static void replenish_dl_entity(struct sched_dl_entity *dl_se,
* task with deadline equal to period this is the same of using
* dl_period instead of dl_deadline in the equation above.
*/
-static bool dl_entity_overflow(struct sched_dl_entity *dl_se,
- struct sched_dl_entity *pi_se, u64 t)
+static bool dl_entity_overflow(struct sched_dl_entity *dl_se, u64 t)
{
u64 left, right;
@@ -835,9 +855,9 @@ static bool dl_entity_overflow(struct sched_dl_entity *dl_se,
* of anything below microseconds resolution is actually fiction
* (but still we want to give the user that illusion >;).
*/
- left = (pi_se->dl_deadline >> DL_SCALE) * (dl_se->runtime >> DL_SCALE);
+ left = (pi_of(dl_se)->dl_deadline >> DL_SCALE) * (dl_se->runtime >> DL_SCALE);
right = ((dl_se->deadline - t) >> DL_SCALE) *
- (pi_se->dl_runtime >> DL_SCALE);
+ (pi_of(dl_se)->dl_runtime >> DL_SCALE);
return dl_time_before(right, left);
}
@@ -922,24 +942,23 @@ static inline bool dl_is_implicit(struct sched_dl_entity *dl_se)
* Please refer to the comments update_dl_revised_wakeup() function to find
* more about the Revised CBS rule.
*/
-static void update_dl_entity(struct sched_dl_entity *dl_se,
- struct sched_dl_entity *pi_se)
+static void update_dl_entity(struct sched_dl_entity *dl_se)
{
struct dl_rq *dl_rq = dl_rq_of_se(dl_se);
struct rq *rq = rq_of_dl_rq(dl_rq);
if (dl_time_before(dl_se->deadline, rq_clock(rq)) ||
- dl_entity_overflow(dl_se, pi_se, rq_clock(rq))) {
+ dl_entity_overflow(dl_se, rq_clock(rq))) {
if (unlikely(!dl_is_implicit(dl_se) &&
!dl_time_before(dl_se->deadline, rq_clock(rq)) &&
- !dl_se->dl_boosted)){
+ !is_dl_boosted(dl_se))) {
update_dl_revised_wakeup(dl_se, rq);
return;
}
- dl_se->deadline = rq_clock(rq) + pi_se->dl_deadline;
- dl_se->runtime = pi_se->dl_runtime;
+ dl_se->deadline = rq_clock(rq) + pi_of(dl_se)->dl_deadline;
+ dl_se->runtime = pi_of(dl_se)->dl_runtime;
}
}
@@ -1038,7 +1057,7 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer)
* The task might have been boosted by someone else and might be in the
* boosting/deboosting path, its not throttled.
*/
- if (dl_se->dl_boosted)
+ if (is_dl_boosted(dl_se))
goto unlock;
/*
@@ -1066,7 +1085,7 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer)
* but do not enqueue -- wait for our wakeup to do that.
*/
if (!task_on_rq_queued(p)) {
- replenish_dl_entity(dl_se, dl_se);
+ replenish_dl_entity(dl_se);
goto unlock;
}
@@ -1156,7 +1175,7 @@ static inline void dl_check_constrained_dl(struct sched_dl_entity *dl_se)
if (dl_time_before(dl_se->deadline, rq_clock(rq)) &&
dl_time_before(rq_clock(rq), dl_next_period(dl_se))) {
- if (unlikely(dl_se->dl_boosted || !start_dl_timer(p)))
+ if (unlikely(is_dl_boosted(dl_se) || !start_dl_timer(p)))
return;
dl_se->dl_throttled = 1;
if (dl_se->runtime > 0)
@@ -1287,7 +1306,7 @@ throttle:
dl_se->dl_overrun = 1;
__dequeue_task_dl(rq, curr, 0);
- if (unlikely(dl_se->dl_boosted || !start_dl_timer(curr)))
+ if (unlikely(is_dl_boosted(dl_se) || !start_dl_timer(curr)))
enqueue_task_dl(rq, curr, ENQUEUE_REPLENISH);
if (!is_leftmost(curr, &rq->dl))
@@ -1481,8 +1500,7 @@ static void __dequeue_dl_entity(struct sched_dl_entity *dl_se)
}
static void
-enqueue_dl_entity(struct sched_dl_entity *dl_se,
- struct sched_dl_entity *pi_se, int flags)
+enqueue_dl_entity(struct sched_dl_entity *dl_se, int flags)
{
BUG_ON(on_dl_rq(dl_se));
@@ -1493,9 +1511,9 @@ enqueue_dl_entity(struct sched_dl_entity *dl_se,
*/
if (flags & ENQUEUE_WAKEUP) {
task_contending(dl_se, flags);
- update_dl_entity(dl_se, pi_se);
+ update_dl_entity(dl_se);
} else if (flags & ENQUEUE_REPLENISH) {
- replenish_dl_entity(dl_se, pi_se);
+ replenish_dl_entity(dl_se);
} else if ((flags & ENQUEUE_RESTORE) &&
dl_time_before(dl_se->deadline,
rq_clock(rq_of_dl_rq(dl_rq_of_se(dl_se))))) {
@@ -1512,19 +1530,7 @@ static void dequeue_dl_entity(struct sched_dl_entity *dl_se)
static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags)
{
- struct task_struct *pi_task = rt_mutex_get_top_task(p);
- struct sched_dl_entity *pi_se = &p->dl;
-
- /*
- * Use the scheduling parameters of the top pi-waiter task if:
- * - we have a top pi-waiter which is a SCHED_DEADLINE task AND
- * - our dl_boosted is set (i.e. the pi-waiter's (absolute) deadline is
- * smaller than our deadline OR we are a !SCHED_DEADLINE task getting
- * boosted due to a SCHED_DEADLINE pi-waiter).
- * Otherwise we keep our runtime and deadline.
- */
- if (pi_task && dl_prio(pi_task->normal_prio) && p->dl.dl_boosted) {
- pi_se = &pi_task->dl;
+ if (is_dl_boosted(&p->dl)) {
/*
* Because of delays in the detection of the overrun of a
* thread's runtime, it might be the case that a thread
@@ -1557,7 +1563,7 @@ static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags)
* the throttle.
*/
p->dl.dl_throttled = 0;
- BUG_ON(!p->dl.dl_boosted || flags != ENQUEUE_REPLENISH);
+ BUG_ON(!is_dl_boosted(&p->dl) || flags != ENQUEUE_REPLENISH);
return;
}
@@ -1594,7 +1600,7 @@ static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags)
return;
}
- enqueue_dl_entity(&p->dl, pi_se, flags);
+ enqueue_dl_entity(&p->dl, flags);
if (!task_current(rq, p) && p->nr_cpus_allowed > 1)
enqueue_pushable_dl_task(rq, p);
@@ -2787,11 +2793,14 @@ void __dl_clear_params(struct task_struct *p)
dl_se->dl_bw = 0;
dl_se->dl_density = 0;
- dl_se->dl_boosted = 0;
dl_se->dl_throttled = 0;
dl_se->dl_yielded = 0;
dl_se->dl_non_contending = 0;
dl_se->dl_overrun = 0;
+
+#ifdef CONFIG_RT_MUTEXES
+ dl_se->pi_se = dl_se;
+#endif
}
bool dl_param_changed(struct task_struct *p, const struct sched_attr *attr)
diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
index 0655524700d2..2357921580f9 100644
--- a/kernel/sched/debug.c
+++ b/kernel/sched/debug.c
@@ -251,7 +251,7 @@ static int sd_ctl_doflags(struct ctl_table *table, int write,
unsigned long flags = *(unsigned long *)table->data;
size_t data_size = 0;
size_t len = 0;
- char *tmp;
+ char *tmp, *buf;
int idx;
if (write)
@@ -269,17 +269,17 @@ static int sd_ctl_doflags(struct ctl_table *table, int write,
return 0;
}
- tmp = kcalloc(data_size + 1, sizeof(*tmp), GFP_KERNEL);
- if (!tmp)
+ buf = kcalloc(data_size + 1, sizeof(*buf), GFP_KERNEL);
+ if (!buf)
return -ENOMEM;
for_each_set_bit(idx, &flags, __SD_FLAG_CNT) {
char *name = sd_flag_debug[idx].name;
- len += snprintf(tmp + len, strlen(name) + 2, "%s ", name);
+ len += snprintf(buf + len, strlen(name) + 2, "%s ", name);
}
- tmp += *ppos;
+ tmp = buf + *ppos;
len -= *ppos;
if (len > *lenp)
@@ -294,7 +294,7 @@ static int sd_ctl_doflags(struct ctl_table *table, int write,
*lenp = len;
*ppos += len;
- kfree(tmp);
+ kfree(buf);
return 0;
}
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 290f9e38378c..ae7ceba8fd4f 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5477,6 +5477,7 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
struct cfs_rq *cfs_rq;
struct sched_entity *se = &p->se;
int idle_h_nr_running = task_has_idle_policy(p);
+ int task_new = !(flags & ENQUEUE_WAKEUP);
/*
* The code below (indirectly) updates schedutil which looks at
@@ -5549,7 +5550,7 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
* into account, but that is not straightforward to implement,
* and the following generally works well enough in practice.
*/
- if (flags & ENQUEUE_WAKEUP)
+ if (!task_new)
update_overutilized_status(rq);
enqueue_throttle:
@@ -6172,21 +6173,21 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t
static int
select_idle_capacity(struct task_struct *p, struct sched_domain *sd, int target)
{
- unsigned long best_cap = 0;
+ unsigned long task_util, best_cap = 0;
int cpu, best_cpu = -1;
struct cpumask *cpus;
- sync_entity_load_avg(&p->se);
-
cpus = this_cpu_cpumask_var_ptr(select_idle_mask);
cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr);
+ task_util = uclamp_task_util(p);
+
for_each_cpu_wrap(cpu, cpus, target) {
unsigned long cpu_cap = capacity_of(cpu);
if (!available_idle_cpu(cpu) && !sched_idle_cpu(cpu))
continue;
- if (task_fits_capacity(p, cpu_cap))
+ if (fits_capacity(task_util, cpu_cap))
return cpu;
if (cpu_cap > best_cap) {
@@ -6198,44 +6199,42 @@ select_idle_capacity(struct task_struct *p, struct sched_domain *sd, int target)
return best_cpu;
}
+static inline bool asym_fits_capacity(int task_util, int cpu)
+{
+ if (static_branch_unlikely(&sched_asym_cpucapacity))
+ return fits_capacity(task_util, capacity_of(cpu));
+
+ return true;
+}
+
/*
* Try and locate an idle core/thread in the LLC cache domain.
*/
static int select_idle_sibling(struct task_struct *p, int prev, int target)
{
struct sched_domain *sd;
+ unsigned long task_util;
int i, recent_used_cpu;
/*
- * For asymmetric CPU capacity systems, our domain of interest is
- * sd_asym_cpucapacity rather than sd_llc.
+ * On asymmetric system, update task utilization because we will check
+ * that the task fits with cpu's capacity.
*/
if (static_branch_unlikely(&sched_asym_cpucapacity)) {
- sd = rcu_dereference(per_cpu(sd_asym_cpucapacity, target));
- /*
- * On an asymmetric CPU capacity system where an exclusive
- * cpuset defines a symmetric island (i.e. one unique
- * capacity_orig value through the cpuset), the key will be set
- * but the CPUs within that cpuset will not have a domain with
- * SD_ASYM_CPUCAPACITY. These should follow the usual symmetric
- * capacity path.
- */
- if (!sd)
- goto symmetric;
-
- i = select_idle_capacity(p, sd, target);
- return ((unsigned)i < nr_cpumask_bits) ? i : target;
+ sync_entity_load_avg(&p->se);
+ task_util = uclamp_task_util(p);
}
-symmetric:
- if (available_idle_cpu(target) || sched_idle_cpu(target))
+ if ((available_idle_cpu(target) || sched_idle_cpu(target)) &&
+ asym_fits_capacity(task_util, target))
return target;
/*
* If the previous CPU is cache affine and idle, don't be stupid:
*/
if (prev != target && cpus_share_cache(prev, target) &&
- (available_idle_cpu(prev) || sched_idle_cpu(prev)))
+ (available_idle_cpu(prev) || sched_idle_cpu(prev)) &&
+ asym_fits_capacity(task_util, prev))
return prev;
/*
@@ -6258,7 +6257,8 @@ symmetric:
recent_used_cpu != target &&
cpus_share_cache(recent_used_cpu, target) &&
(available_idle_cpu(recent_used_cpu) || sched_idle_cpu(recent_used_cpu)) &&
- cpumask_test_cpu(p->recent_used_cpu, p->cpus_ptr)) {
+ cpumask_test_cpu(p->recent_used_cpu, p->cpus_ptr) &&
+ asym_fits_capacity(task_util, recent_used_cpu)) {
/*
* Replace recent_used_cpu with prev as it is a potential
* candidate for the next wake:
@@ -6267,6 +6267,26 @@ symmetric:
return recent_used_cpu;
}
+ /*
+ * For asymmetric CPU capacity systems, our domain of interest is
+ * sd_asym_cpucapacity rather than sd_llc.
+ */
+ if (static_branch_unlikely(&sched_asym_cpucapacity)) {
+ sd = rcu_dereference(per_cpu(sd_asym_cpucapacity, target));
+ /*
+ * On an asymmetric CPU capacity system where an exclusive
+ * cpuset defines a symmetric island (i.e. one unique
+ * capacity_orig value through the cpuset), the key will be set
+ * but the CPUs within that cpuset will not have a domain with
+ * SD_ASYM_CPUCAPACITY. These should follow the usual symmetric
+ * capacity path.
+ */
+ if (sd) {
+ i = select_idle_capacity(p, sd, target);
+ return ((unsigned)i < nr_cpumask_bits) ? i : target;
+ }
+ }
+
sd = rcu_dereference(per_cpu(sd_llc, target));
if (!sd)
return target;
@@ -9031,7 +9051,8 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
* emptying busiest.
*/
if (local->group_type == group_has_spare) {
- if (busiest->group_type > group_fully_busy) {
+ if ((busiest->group_type > group_fully_busy) &&
+ !(env->sd->flags & SD_SHARE_PKG_RESOURCES)) {
/*
* If busiest is overloaded, try to fill spare
* capacity. This might end up creating spare capacity
diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c
index 24d0ee26377d..c6932b8f4467 100644
--- a/kernel/sched/idle.c
+++ b/kernel/sched/idle.c
@@ -78,7 +78,7 @@ void __weak arch_cpu_idle_dead(void) { }
void __weak arch_cpu_idle(void)
{
cpu_idle_force_poll = 1;
- local_irq_enable();
+ raw_local_irq_enable();
}
/**
@@ -94,9 +94,35 @@ void __cpuidle default_idle_call(void)
trace_cpu_idle(1, smp_processor_id());
stop_critical_timings();
+
+ /*
+ * arch_cpu_idle() is supposed to enable IRQs, however
+ * we can't do that because of RCU and tracing.
+ *
+ * Trace IRQs enable here, then switch off RCU, and have
+ * arch_cpu_idle() use raw_local_irq_enable(). Note that
+ * rcu_idle_enter() relies on lockdep IRQ state, so switch that
+ * last -- this is very similar to the entry code.
+ */
+ trace_hardirqs_on_prepare();
+ lockdep_hardirqs_on_prepare(_THIS_IP_);
rcu_idle_enter();
+ lockdep_hardirqs_on(_THIS_IP_);
+
arch_cpu_idle();
+
+ /*
+ * OK, so IRQs are enabled here, but RCU needs them disabled to
+ * turn itself back on.. funny thing is that disabling IRQs
+ * will cause tracing, which needs RCU. Jump through hoops to
+ * make it 'work'.
+ */
+ raw_local_irq_disable();
+ lockdep_hardirqs_off(_THIS_IP_);
rcu_idle_exit();
+ lockdep_hardirqs_on(_THIS_IP_);
+ raw_local_irq_enable();
+
start_critical_timings();
trace_cpu_idle(PWR_EVENT_EXIT, smp_processor_id());
}