#define SD_BALANCE_FORK 0x0008 /* Balance on fork, clone */
#define SD_BALANCE_WAKE 0x0010 /* Balance on wakeup */
#define SD_WAKE_AFFINE 0x0020 /* Wake task to waking CPU */
-#define SD_SHARE_CPUCAPACITY 0x0080 /* Domain members share cpu power */
+#define SD_ASYM_CPUCAPACITY 0x0040 /* Groups have different max cpu capacities */
+#define SD_SHARE_CPUCAPACITY 0x0080 /* Domain members share cpu capacity */
#define SD_SHARE_POWERDOMAIN 0x0100 /* Domain members share power domain */
#define SD_SHARE_PKG_RESOURCES 0x0200 /* Domain members share cpu pkg resources */
#define SD_SERIALIZE 0x0400 /* Only a single load balancing instance */
#endif
}
+static inline unsigned long get_preempt_disable_ip(struct task_struct *p)
+{
+#ifdef CONFIG_DEBUG_PREEMPT
+ return p->preempt_disable_ip;
+#else
+ return 0;
+#endif
+}
+
/*
* Does a critical section need to be broken due to another
* task waiting?: (technically does not depend on CONFIG_PREEMPT,
/*
* Task isn't running anymore; make it appear like we migrated
* it before it went to sleep. This means on wakeup we make the
- * previous cpu our targer instead of where it really is.
+ * previous cpu our target instead of where it really is.
*/
p->wake_cpu = cpu;
}
/**
* try_to_wake_up_local - try to wake up a local task with rq lock held
* @p: the thread to be awakened
+ * @cookie: context's cookie for pinning
*
* Put @p on the run-queue if it's not already there. The caller must
* ensure that this_rq() is locked, @p is bound to this_rq() and not
*/
static noinline void __schedule_bug(struct task_struct *prev)
{
+ /* Save this before calling printk(), since that will clobber it */
+ unsigned long preempt_disable_ip = get_preempt_disable_ip(current);
+
if (oops_in_progress)
return;
print_modules();
if (irqs_disabled())
print_irqtrace_events(prev);
-#ifdef CONFIG_DEBUG_PREEMPT
- if (in_atomic_preempt_off()) {
+ if (IS_ENABLED(CONFIG_DEBUG_PREEMPT)
+ && in_atomic_preempt_off()) {
pr_err("Preemption disabled at:");
- print_ip_sym(current->preempt_disable_ip);
+ print_ip_sym(preempt_disable_ip);
pr_cont("\n");
}
-#endif
if (panic_on_warn)
panic("scheduling while atomic\n");
SD_BALANCE_FORK |
SD_BALANCE_EXEC |
SD_SHARE_CPUCAPACITY |
+ SD_ASYM_CPUCAPACITY |
SD_SHARE_PKG_RESOURCES |
SD_SHARE_POWERDOMAIN)) {
if (sd->groups != sd->groups->next)
SD_BALANCE_NEWIDLE |
SD_BALANCE_FORK |
SD_BALANCE_EXEC |
+ SD_ASYM_CPUCAPACITY |
SD_SHARE_CPUCAPACITY |
SD_SHARE_PKG_RESOURCES |
SD_PREFER_SIBLING |
/*
* SD_flags allowed in topology descriptions.
*
- * SD_SHARE_CPUCAPACITY - describes SMT topologies
- * SD_SHARE_PKG_RESOURCES - describes shared caches
- * SD_NUMA - describes NUMA topologies
- * SD_SHARE_POWERDOMAIN - describes shared power domain
+ * These flags are purely descriptive of the topology and do not prescribe
+ * behaviour. Behaviour is artificial and mapped in the below sd_init()
+ * function:
+ *
+ * SD_SHARE_CPUCAPACITY - describes SMT topologies
+ * SD_SHARE_PKG_RESOURCES - describes shared caches
+ * SD_NUMA - describes NUMA topologies
+ * SD_SHARE_POWERDOMAIN - describes shared power domain
+ * SD_ASYM_CPUCAPACITY - describes mixed capacity topologies
+ *
+ * Odd one out, which beside describing the topology has a quirk also
+ * prescribes the desired behaviour that goes along with it:
*
- * Odd one out:
- * SD_ASYM_PACKING - describes SMT quirks
+ * SD_ASYM_PACKING - describes SMT quirks
*/
#define TOPOLOGY_SD_FLAGS \
(SD_SHARE_CPUCAPACITY | \
SD_SHARE_PKG_RESOURCES | \
SD_NUMA | \
SD_ASYM_PACKING | \
+ SD_ASYM_CPUCAPACITY | \
SD_SHARE_POWERDOMAIN)
static struct sched_domain *
-sd_init(struct sched_domain_topology_level *tl, int cpu)
+sd_init(struct sched_domain_topology_level *tl,
+ struct sched_domain *child, int cpu)
{
struct sched_domain *sd = *per_cpu_ptr(tl->data.sd, cpu);
int sd_weight, sd_flags = 0;
.smt_gain = 0,
.max_newidle_lb_cost = 0,
.next_decay_max_lb_cost = jiffies,
+ .child = child,
#ifdef CONFIG_SCHED_DEBUG
.name = tl->name,
#endif
* Convert topological properties into behaviour.
*/
+ if (sd->flags & SD_ASYM_CPUCAPACITY) {
+ struct sched_domain *t = sd;
+
+ for_each_lower_domain(t)
+ t->flags |= SD_BALANCE_WAKE;
+ }
+
if (sd->flags & SD_SHARE_CPUCAPACITY) {
sd->flags |= SD_PREFER_SIBLING;
sd->imbalance_pct = 110;
const struct cpumask *cpu_map, struct sched_domain_attr *attr,
struct sched_domain *child, int cpu)
{
- struct sched_domain *sd = sd_init(tl, cpu);
- if (!sd)
- return child;
+ struct sched_domain *sd = sd_init(tl, child, cpu);
cpumask_and(sched_domain_span(sd), cpu_map, tl->mask(cpu));
if (child) {
sd->level = child->level + 1;
sched_domain_level_max = max(sched_domain_level_max, sd->level);
child->parent = sd;
- sd->child = child;
if (!cpumask_subset(sched_domain_span(child),
sched_domain_span(sd))) {
enum s_alloc alloc_state;
struct sched_domain *sd;
struct s_data d;
+ struct rq *rq = NULL;
int i, ret = -ENOMEM;
alloc_state = __visit_domain_allocation_hell(&d, cpu_map);
/* Attach the domains */
rcu_read_lock();
for_each_cpu(i, cpu_map) {
+ rq = cpu_rq(i);
sd = *per_cpu_ptr(d.sd, i);
+
+ /* Use READ_ONCE()/WRITE_ONCE() to avoid load/store tearing: */
+ if (rq->cpu_capacity_orig > READ_ONCE(d.rd->max_cpu_capacity))
+ WRITE_ONCE(d.rd->max_cpu_capacity, rq->cpu_capacity_orig);
+
cpu_attach_domain(sd, d.rd, i);
}
rcu_read_unlock();
+ if (rq) {
+ pr_info("span: %*pbl (max cpu_capacity = %lu)\n",
+ cpumask_pr_args(cpu_map), rq->rd->max_cpu_capacity);
+ }
+
ret = 0;
error:
__free_domain_allocs(&d, alloc_state, cpu_map);
void ___might_sleep(const char *file, int line, int preempt_offset)
{
static unsigned long prev_jiffy; /* ratelimiting */
+ unsigned long preempt_disable_ip;
rcu_sleep_check(); /* WARN_ON_ONCE() by default, no rate limit reqd. */
if ((preempt_count_equals(preempt_offset) && !irqs_disabled() &&
return;
prev_jiffy = jiffies;
+ /* Save this before calling printk(), since that will clobber it */
+ preempt_disable_ip = get_preempt_disable_ip(current);
+
printk(KERN_ERR
"BUG: sleeping function called from invalid context at %s:%d\n",
file, line);
debug_show_held_locks(current);
if (irqs_disabled())
print_irqtrace_events(current);
-#ifdef CONFIG_DEBUG_PREEMPT
- if (!preempt_count_equals(preempt_offset)) {
+ if (IS_ENABLED(CONFIG_DEBUG_PREEMPT)
+ && !preempt_count_equals(preempt_offset)) {
pr_err("Preemption disabled at:");
- print_ip_sym(current->preempt_disable_ip);
+ print_ip_sym(preempt_disable_ip);
pr_cont("\n");
}
-#endif
dump_stack();
+ add_taint(TAINT_WARN, LOCKDEP_STILL_OK);
}
EXPORT_SYMBOL(___might_sleep);
#endif
return accounted;
}
+#ifdef CONFIG_64BIT
+static inline u64 read_sum_exec_runtime(struct task_struct *t)
+{
+ return t->se.sum_exec_runtime;
+}
+#else
+static u64 read_sum_exec_runtime(struct task_struct *t)
+{
+ u64 ns;
+ struct rq_flags rf;
+ struct rq *rq;
+
+ rq = task_rq_lock(t, &rf);
+ ns = t->se.sum_exec_runtime;
+ task_rq_unlock(rq, t, &rf);
+
+ return ns;
+}
+#endif
+
/*
* Accumulate raw cputime values of dead tasks (sig->[us]time) and live
* tasks (sum on group iteration) belonging to @tsk's group.
unsigned int seq, nextseq;
unsigned long flags;
+ /*
+ * Update current task runtime to account pending time since last
+ * scheduler action or thread_group_cputime() call. This thread group
+ * might have other running tasks on different CPUs, but updating
+ * their runtime can affect syscall performance, so we skip account
+ * those pending times and rely only on values updated on tick or
+ * other scheduler action.
+ */
+ if (same_thread_group(current, tsk))
+ (void) task_sched_runtime(current);
+
rcu_read_lock();
/* Attempt a lockless read on the first round. */
nextseq = 0;
task_cputime(t, &utime, &stime);
times->utime += utime;
times->stime += stime;
- times->sum_exec_runtime += task_sched_runtime(t);
+ times->sum_exec_runtime += read_sum_exec_runtime(t);
}
/* If lockless access failed, take the lock. */
nextseq = 1;
* one, and to (try to!) reconcile itself with its own scheduling
* parameters.
*/
-static inline void setup_new_dl_entity(struct sched_dl_entity *dl_se,
- struct sched_dl_entity *pi_se)
+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(dl_time_before(rq_clock(rq), dl_se->deadline));
/*
* future; in fact, we must consider execution overheads (time
* spent on hardirq context, etc.).
*/
- dl_se->deadline = rq_clock(rq) + pi_se->dl_deadline;
- dl_se->runtime = pi_se->dl_runtime;
+ dl_se->deadline = rq_clock(rq) + dl_se->dl_deadline;
+ dl_se->runtime = dl_se->dl_runtime;
}
/*
*/
static void switched_to_dl(struct rq *rq, struct task_struct *p)
{
+
+ /* If p is not queued we will update its parameters at next wakeup. */
+ if (!task_on_rq_queued(p))
+ return;
+
+ /*
+ * If p is boosted we already updated its params in
+ * rt_mutex_setprio()->enqueue_task(..., ENQUEUE_REPLENISH),
+ * p's deadline being now already after rq_clock(rq).
+ */
if (dl_time_before(p->dl.deadline, rq_clock(rq)))
- setup_new_dl_entity(&p->dl, &p->dl);
+ setup_new_dl_entity(&p->dl);
- if (task_on_rq_queued(p) && rq->curr != p) {
+ if (rq->curr != p) {
#ifdef CONFIG_SMP
if (tsk_nr_cpus_allowed(p) > 1 && rq->dl.overloaded)
queue_push_tasks(rq);
unsigned int sysctl_sched_cfs_bandwidth_slice = 5000UL;
#endif
+/*
+ * The margin used when comparing utilization with CPU capacity:
+ * util * 1024 < capacity * margin
+ */
+unsigned int capacity_margin = 1280; /* ~20% */
+
static inline void update_load_add(struct load_weight *lw, unsigned long inc)
{
lw->weight += inc;
}
#ifdef CONFIG_SMP
-static int select_idle_sibling(struct task_struct *p, int cpu);
+static int select_idle_sibling(struct task_struct *p, int prev_cpu, int cpu);
static unsigned long task_h_load(struct task_struct *p);
/*
struct sched_avg *sa = &se->avg;
long cap = (long)(SCHED_CAPACITY_SCALE - cfs_rq->avg.util_avg) / 2;
u64 now = cfs_rq_clock_task(cfs_rq);
- int tg_update;
if (cap > 0) {
if (cfs_rq->avg.util_avg != 0) {
}
}
- tg_update = update_cfs_rq_load_avg(now, cfs_rq, false);
+ update_cfs_rq_load_avg(now, cfs_rq, false);
attach_entity_load_avg(cfs_rq, se);
- if (tg_update)
- update_tg_load_avg(cfs_rq, false);
+ update_tg_load_avg(cfs_rq, false);
}
#else /* !CONFIG_SMP */
* Call select_idle_sibling to maybe find a better one.
*/
if (!cur)
- env->dst_cpu = select_idle_sibling(env->p, env->dst_cpu);
+ env->dst_cpu = select_idle_sibling(env->p, env->src_cpu,
+ env->dst_cpu);
assign:
task_numa_assign(env, cur, imp);
}
#ifdef CONFIG_FAIR_GROUP_SCHED
-/*
- * Updating tg's load_avg is necessary before update_cfs_share (which is done)
- * and effective_load (which is not done because it is too costly).
+/**
+ * update_tg_load_avg - update the tg's load avg
+ * @cfs_rq: the cfs_rq whose avg changed
+ * @force: update regardless of how small the difference
+ *
+ * This function 'ensures': tg->load_avg := \Sum tg->cfs_rq[]->avg.load.
+ * However, because tg->load_avg is a global value there are performance
+ * considerations.
+ *
+ * In order to avoid having to look at the other cfs_rq's, we use a
+ * differential update where we store the last value we propagated. This in
+ * turn allows skipping updates if the differential is 'small'.
+ *
+ * Updating tg's load_avg is necessary before update_cfs_share() (which is
+ * done) and effective_load() (which is not done because it is too costly).
*/
static inline void update_tg_load_avg(struct cfs_rq *cfs_rq, int force)
{
*
* cfs_rq->avg is used for task_h_load() and update_cfs_share() for example.
*
- * Returns true if the load decayed or we removed utilization. It is expected
- * that one calls update_tg_load_avg() on this condition, but after you've
- * modified the cfs_rq avg (attach/detach), such that we propagate the new
- * avg up.
+ * Returns true if the load decayed or we removed load.
+ *
+ * Since both these conditions indicate a changed cfs_rq->avg.load we should
+ * call update_tg_load_avg() when this function returns true.
*/
static inline int
update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq, bool update_freq)
return 1;
}
-static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)
+static int wake_affine(struct sched_domain *sd, struct task_struct *p,
+ int prev_cpu, int sync)
{
s64 this_load, load;
s64 this_eff_load, prev_eff_load;
- int idx, this_cpu, prev_cpu;
+ int idx, this_cpu;
struct task_group *tg;
unsigned long weight;
int balanced;
idx = sd->wake_idx;
this_cpu = smp_processor_id();
- prev_cpu = task_cpu(p);
load = source_load(prev_cpu, idx);
this_load = target_load(this_cpu, idx);
int shallowest_idle_cpu = -1;
int i;
+ /* Check if we have any choice: */
+ if (group->group_weight == 1)
+ return cpumask_first(sched_group_cpus(group));
+
/* Traverse only the allowed CPUs */
for_each_cpu_and(i, sched_group_cpus(group), tsk_cpus_allowed(p)) {
if (idle_cpu(i)) {
/*
* Try and locate an idle CPU in the sched_domain.
*/
-static int select_idle_sibling(struct task_struct *p, int target)
+static int select_idle_sibling(struct task_struct *p, int prev, int target)
{
struct sched_domain *sd;
struct sched_group *sg;
- int i = task_cpu(p);
if (idle_cpu(target))
return target;
/*
* If the prevous cpu is cache affine and idle, don't be stupid.
*/
- if (i != target && cpus_share_cache(i, target) && idle_cpu(i))
- return i;
+ if (prev != target && cpus_share_cache(prev, target) && idle_cpu(prev))
+ return prev;
/*
* Otherwise, iterate the domains and find an eligible idle cpu.
for_each_lower_domain(sd) {
sg = sd->groups;
do {
+ int i;
+
if (!cpumask_intersects(sched_group_cpus(sg),
tsk_cpus_allowed(p)))
goto next;
return (util >= capacity) ? capacity : util;
}
+static inline int task_util(struct task_struct *p)
+{
+ return p->se.avg.util_avg;
+}
+
+/*
+ * Disable WAKE_AFFINE in the case where task @p doesn't fit in the
+ * capacity of either the waking CPU @cpu or the previous CPU @prev_cpu.
+ *
+ * In that case WAKE_AFFINE doesn't make sense and we'll let
+ * BALANCE_WAKE sort things out.
+ */
+static int wake_cap(struct task_struct *p, int cpu, int prev_cpu)
+{
+ long min_cap, max_cap;
+
+ min_cap = min(capacity_orig_of(prev_cpu), capacity_orig_of(cpu));
+ max_cap = cpu_rq(cpu)->rd->max_cpu_capacity;
+
+ /* Minimum capacity is close to max, no need to abort wake_affine */
+ if (max_cap - min_cap < max_cap >> 3)
+ return 0;
+
+ return min_cap * 1024 < task_util(p) * capacity_margin;
+}
+
/*
* select_task_rq_fair: Select target runqueue for the waking task in domains
* that have the 'sd_flag' flag set. In practice, this is SD_BALANCE_WAKE,
if (sd_flag & SD_BALANCE_WAKE) {
record_wakee(p);
- want_affine = !wake_wide(p) && cpumask_test_cpu(cpu, tsk_cpus_allowed(p));
+ want_affine = !wake_wide(p) && !wake_cap(p, cpu, prev_cpu)
+ && cpumask_test_cpu(cpu, tsk_cpus_allowed(p));
}
rcu_read_lock();
if (affine_sd) {
sd = NULL; /* Prefer wake_affine over balance flags */
- if (cpu != prev_cpu && wake_affine(affine_sd, p, sync))
+ if (cpu != prev_cpu && wake_affine(affine_sd, p, prev_cpu, sync))
new_cpu = cpu;
}
if (!sd) {
if (sd_flag & SD_BALANCE_WAKE) /* XXX always ? */
- new_cpu = select_idle_sibling(p, new_cpu);
+ new_cpu = select_idle_sibling(p, prev_cpu, new_cpu);
} else while (sd) {
struct sched_group *group;
}
static inline void
-update_next_balance(struct sched_domain *sd, int cpu_busy, unsigned long *next_balance)
+update_next_balance(struct sched_domain *sd, unsigned long *next_balance)
{
unsigned long interval, next;
- interval = get_sd_balance_interval(sd, cpu_busy);
+ /* used by idle balance, so cpu_busy = 0 */
+ interval = get_sd_balance_interval(sd, 0);
next = sd->last_balance + interval;
if (time_after(*next_balance, next))
rcu_read_lock();
sd = rcu_dereference_check_sched_domain(this_rq->sd);
if (sd)
- update_next_balance(sd, 0, &next_balance);
+ update_next_balance(sd, &next_balance);
rcu_read_unlock();
goto out;
continue;
if (this_rq->avg_idle < curr_cost + sd->max_newidle_lb_cost) {
- update_next_balance(sd, 0, &next_balance);
+ update_next_balance(sd, &next_balance);
break;
}
curr_cost += domain_cost;
}
- update_next_balance(sd, 0, &next_balance);
+ update_next_balance(sd, &next_balance);
/*
* Stop searching for tasks to pull if there are
struct sched_entity *se = &p->se;
struct cfs_rq *cfs_rq = cfs_rq_of(se);
u64 now = cfs_rq_clock_task(cfs_rq);
- int tg_update;
if (!vruntime_normalized(p)) {
/*
}
/* Catch up with the cfs_rq and remove our load when we leave */
- tg_update = update_cfs_rq_load_avg(now, cfs_rq, false);
+ update_cfs_rq_load_avg(now, cfs_rq, false);
detach_entity_load_avg(cfs_rq, se);
- if (tg_update)
- update_tg_load_avg(cfs_rq, false);
+ update_tg_load_avg(cfs_rq, false);
}
static void attach_task_cfs_rq(struct task_struct *p)
struct sched_entity *se = &p->se;
struct cfs_rq *cfs_rq = cfs_rq_of(se);
u64 now = cfs_rq_clock_task(cfs_rq);
- int tg_update;
#ifdef CONFIG_FAIR_GROUP_SCHED
/*
#endif
/* Synchronize task with its cfs_rq */
- tg_update = update_cfs_rq_load_avg(now, cfs_rq, false);
+ update_cfs_rq_load_avg(now, cfs_rq, false);
attach_entity_load_avg(cfs_rq, se);
- if (tg_update)
- update_tg_load_avg(cfs_rq, false);
+ update_tg_load_avg(cfs_rq, false);
if (!vruntime_normalized(p))
se->vruntime += cfs_rq->min_vruntime;
*/
cpumask_var_t rto_mask;
struct cpupri cpupri;
+
+ unsigned long max_cpu_capacity;
};
extern struct root_domain def_root_domain;
#ifdef CONFIG_SMP
unsigned long last_load_update_tick;
#endif /* CONFIG_SMP */
- u64 nohz_stamp;
unsigned long nohz_flags;
#endif /* CONFIG_NO_HZ_COMMON */
#ifdef CONFIG_NO_HZ_FULL