summaryrefslogtreecommitdiff
path: root/kernel
diff options
context:
space:
mode:
Diffstat (limited to 'kernel')
-rw-r--r--kernel/bpf/stackmap.c5
-rw-r--r--kernel/cgroup/cgroup.c11
-rw-r--r--kernel/events/core.c246
-rw-r--r--kernel/power/snapshot.c21
-rw-r--r--kernel/power/wakelock.c11
-rw-r--r--kernel/rcu/tasks.h12
-rw-r--r--kernel/sched/core.c15
-rw-r--r--kernel/sched/core_sched.c2
-rw-r--r--kernel/sched/fair.c118
-rw-r--r--kernel/sched/pelt.h4
-rw-r--r--kernel/sched/psi.c66
-rw-r--r--kernel/trace/Kconfig8
-rw-r--r--kernel/trace/trace.c3
-rw-r--r--kernel/trace/trace_events_hist.c10
-rw-r--r--kernel/ucount.c2
15 files changed, 309 insertions, 225 deletions
diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c
index 49e567209c6b..22c8ae94e4c1 100644
--- a/kernel/bpf/stackmap.c
+++ b/kernel/bpf/stackmap.c
@@ -472,13 +472,14 @@ BPF_CALL_4(bpf_get_task_stack, struct task_struct *, task, void *, buf,
u32, size, u64, flags)
{
struct pt_regs *regs;
- long res;
+ long res = -EINVAL;
if (!try_get_task_stack(task))
return -EFAULT;
regs = task_pt_regs(task);
- res = __bpf_get_stack(regs, task, NULL, buf, size, flags);
+ if (regs)
+ res = __bpf_get_stack(regs, task, NULL, buf, size, flags);
put_task_stack(task);
return res;
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index b31e1465868a..9d05c3ca2d5e 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -3643,6 +3643,12 @@ static ssize_t cgroup_pressure_write(struct kernfs_open_file *of, char *buf,
cgroup_get(cgrp);
cgroup_kn_unlock(of->kn);
+ /* Allow only one trigger per file descriptor */
+ if (ctx->psi.trigger) {
+ cgroup_put(cgrp);
+ return -EBUSY;
+ }
+
psi = cgroup_ino(cgrp) == 1 ? &psi_system : &cgrp->psi;
new = psi_trigger_create(psi, buf, nbytes, res);
if (IS_ERR(new)) {
@@ -3650,8 +3656,7 @@ static ssize_t cgroup_pressure_write(struct kernfs_open_file *of, char *buf,
return PTR_ERR(new);
}
- psi_trigger_replace(&ctx->psi.trigger, new);
-
+ smp_store_release(&ctx->psi.trigger, new);
cgroup_put(cgrp);
return nbytes;
@@ -3690,7 +3695,7 @@ static void cgroup_pressure_release(struct kernfs_open_file *of)
{
struct cgroup_file_ctx *ctx = of->priv;
- psi_trigger_replace(&ctx->psi.trigger, NULL);
+ psi_trigger_destroy(ctx->psi.trigger);
}
bool cgroup_psi_enabled(void)
diff --git a/kernel/events/core.c b/kernel/events/core.c
index fc18664f49b0..479c9e672ec4 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -674,6 +674,23 @@ perf_event_set_state(struct perf_event *event, enum perf_event_state state)
WRITE_ONCE(event->state, state);
}
+/*
+ * UP store-release, load-acquire
+ */
+
+#define __store_release(ptr, val) \
+do { \
+ barrier(); \
+ WRITE_ONCE(*(ptr), (val)); \
+} while (0)
+
+#define __load_acquire(ptr) \
+({ \
+ __unqual_scalar_typeof(*(ptr)) ___p = READ_ONCE(*(ptr)); \
+ barrier(); \
+ ___p; \
+})
+
#ifdef CONFIG_CGROUP_PERF
static inline bool
@@ -719,34 +736,51 @@ static inline u64 perf_cgroup_event_time(struct perf_event *event)
return t->time;
}
-static inline void __update_cgrp_time(struct perf_cgroup *cgrp)
+static inline u64 perf_cgroup_event_time_now(struct perf_event *event, u64 now)
{
- struct perf_cgroup_info *info;
- u64 now;
-
- now = perf_clock();
+ struct perf_cgroup_info *t;
- info = this_cpu_ptr(cgrp->info);
+ t = per_cpu_ptr(event->cgrp->info, event->cpu);
+ if (!__load_acquire(&t->active))
+ return t->time;
+ now += READ_ONCE(t->timeoffset);
+ return now;
+}
- info->time += now - info->timestamp;
+static inline void __update_cgrp_time(struct perf_cgroup_info *info, u64 now, bool adv)
+{
+ if (adv)
+ info->time += now - info->timestamp;
info->timestamp = now;
+ /*
+ * see update_context_time()
+ */
+ WRITE_ONCE(info->timeoffset, info->time - info->timestamp);
}
-static inline void update_cgrp_time_from_cpuctx(struct perf_cpu_context *cpuctx)
+static inline void update_cgrp_time_from_cpuctx(struct perf_cpu_context *cpuctx, bool final)
{
struct perf_cgroup *cgrp = cpuctx->cgrp;
struct cgroup_subsys_state *css;
+ struct perf_cgroup_info *info;
if (cgrp) {
+ u64 now = perf_clock();
+
for (css = &cgrp->css; css; css = css->parent) {
cgrp = container_of(css, struct perf_cgroup, css);
- __update_cgrp_time(cgrp);
+ info = this_cpu_ptr(cgrp->info);
+
+ __update_cgrp_time(info, now, true);
+ if (final)
+ __store_release(&info->active, 0);
}
}
}
static inline void update_cgrp_time_from_event(struct perf_event *event)
{
+ struct perf_cgroup_info *info;
struct perf_cgroup *cgrp;
/*
@@ -760,8 +794,10 @@ static inline void update_cgrp_time_from_event(struct perf_event *event)
/*
* Do not update time when cgroup is not active
*/
- if (cgroup_is_descendant(cgrp->css.cgroup, event->cgrp->css.cgroup))
- __update_cgrp_time(event->cgrp);
+ if (cgroup_is_descendant(cgrp->css.cgroup, event->cgrp->css.cgroup)) {
+ info = this_cpu_ptr(event->cgrp->info);
+ __update_cgrp_time(info, perf_clock(), true);
+ }
}
static inline void
@@ -785,7 +821,8 @@ perf_cgroup_set_timestamp(struct task_struct *task,
for (css = &cgrp->css; css; css = css->parent) {
cgrp = container_of(css, struct perf_cgroup, css);
info = this_cpu_ptr(cgrp->info);
- info->timestamp = ctx->timestamp;
+ __update_cgrp_time(info, ctx->timestamp, false);
+ __store_release(&info->active, 1);
}
}
@@ -982,14 +1019,6 @@ out:
}
static inline void
-perf_cgroup_set_shadow_time(struct perf_event *event, u64 now)
-{
- struct perf_cgroup_info *t;
- t = per_cpu_ptr(event->cgrp->info, event->cpu);
- event->shadow_ctx_time = now - t->timestamp;
-}
-
-static inline void
perf_cgroup_event_enable(struct perf_event *event, struct perf_event_context *ctx)
{
struct perf_cpu_context *cpuctx;
@@ -1066,7 +1095,8 @@ static inline void update_cgrp_time_from_event(struct perf_event *event)
{
}
-static inline void update_cgrp_time_from_cpuctx(struct perf_cpu_context *cpuctx)
+static inline void update_cgrp_time_from_cpuctx(struct perf_cpu_context *cpuctx,
+ bool final)
{
}
@@ -1098,12 +1128,12 @@ perf_cgroup_switch(struct task_struct *task, struct task_struct *next)
{
}
-static inline void
-perf_cgroup_set_shadow_time(struct perf_event *event, u64 now)
+static inline u64 perf_cgroup_event_time(struct perf_event *event)
{
+ return 0;
}
-static inline u64 perf_cgroup_event_time(struct perf_event *event)
+static inline u64 perf_cgroup_event_time_now(struct perf_event *event, u64 now)
{
return 0;
}
@@ -1525,22 +1555,59 @@ static void perf_unpin_context(struct perf_event_context *ctx)
/*
* Update the record of the current time in a context.
*/
-static void update_context_time(struct perf_event_context *ctx)
+static void __update_context_time(struct perf_event_context *ctx, bool adv)
{
u64 now = perf_clock();
- ctx->time += now - ctx->timestamp;
+ if (adv)
+ ctx->time += now - ctx->timestamp;
ctx->timestamp = now;
+
+ /*
+ * The above: time' = time + (now - timestamp), can be re-arranged
+ * into: time` = now + (time - timestamp), which gives a single value
+ * offset to compute future time without locks on.
+ *
+ * See perf_event_time_now(), which can be used from NMI context where
+ * it's (obviously) not possible to acquire ctx->lock in order to read
+ * both the above values in a consistent manner.
+ */
+ WRITE_ONCE(ctx->timeoffset, ctx->time - ctx->timestamp);
+}
+
+static void update_context_time(struct perf_event_context *ctx)
+{
+ __update_context_time(ctx, true);
}
static u64 perf_event_time(struct perf_event *event)
{
struct perf_event_context *ctx = event->ctx;
+ if (unlikely(!ctx))
+ return 0;
+
if (is_cgroup_event(event))
return perf_cgroup_event_time(event);
- return ctx ? ctx->time : 0;
+ return ctx->time;
+}
+
+static u64 perf_event_time_now(struct perf_event *event, u64 now)
+{
+ struct perf_event_context *ctx = event->ctx;
+
+ if (unlikely(!ctx))
+ return 0;
+
+ if (is_cgroup_event(event))
+ return perf_cgroup_event_time_now(event, now);
+
+ if (!(__load_acquire(&ctx->is_active) & EVENT_TIME))
+ return ctx->time;
+
+ now += READ_ONCE(ctx->timeoffset);
+ return now;
}
static enum event_type_t get_event_type(struct perf_event *event)
@@ -2350,7 +2417,7 @@ __perf_remove_from_context(struct perf_event *event,
if (ctx->is_active & EVENT_TIME) {
update_context_time(ctx);
- update_cgrp_time_from_cpuctx(cpuctx);
+ update_cgrp_time_from_cpuctx(cpuctx, false);
}
event_sched_out(event, cpuctx, ctx);
@@ -2361,6 +2428,9 @@ __perf_remove_from_context(struct perf_event *event,
list_del_event(event, ctx);
if (!ctx->nr_events && ctx->is_active) {
+ if (ctx == &cpuctx->ctx)
+ update_cgrp_time_from_cpuctx(cpuctx, true);
+
ctx->is_active = 0;
ctx->rotate_necessary = 0;
if (ctx->task) {
@@ -2482,40 +2552,6 @@ void perf_event_disable_inatomic(struct perf_event *event)
irq_work_queue(&event->pending);
}
-static void perf_set_shadow_time(struct perf_event *event,
- struct perf_event_context *ctx)
-{
- /*
- * use the correct time source for the time snapshot
- *
- * We could get by without this by leveraging the
- * fact that to get to this function, the caller
- * has most likely already called update_context_time()
- * and update_cgrp_time_xx() and thus both timestamp
- * are identical (or very close). Given that tstamp is,
- * already adjusted for cgroup, we could say that:
- * tstamp - ctx->timestamp
- * is equivalent to
- * tstamp - cgrp->timestamp.
- *
- * Then, in perf_output_read(), the calculation would
- * work with no changes because:
- * - event is guaranteed scheduled in
- * - no scheduled out in between
- * - thus the timestamp would be the same
- *
- * But this is a bit hairy.
- *
- * So instead, we have an explicit cgroup call to remain
- * within the time source all along. We believe it
- * is cleaner and simpler to understand.
- */
- if (is_cgroup_event(event))
- perf_cgroup_set_shadow_time(event, event->tstamp);
- else
- event->shadow_ctx_time = event->tstamp - ctx->timestamp;
-}
-
#define MAX_INTERRUPTS (~0ULL)
static void perf_log_throttle(struct perf_event *event, int enable);
@@ -2556,8 +2592,6 @@ event_sched_in(struct perf_event *event,
perf_pmu_disable(event->pmu);
- perf_set_shadow_time(event, ctx);
-
perf_log_itrace_start(event);
if (event->pmu->add(event, PERF_EF_START)) {
@@ -3251,16 +3285,6 @@ static void ctx_sched_out(struct perf_event_context *ctx,
return;
}
- ctx->is_active &= ~event_type;
- if (!(ctx->is_active & EVENT_ALL))
- ctx->is_active = 0;
-
- if (ctx->task) {
- WARN_ON_ONCE(cpuctx->task_ctx != ctx);
- if (!ctx->is_active)
- cpuctx->task_ctx = NULL;
- }
-
/*
* Always update time if it was set; not only when it changes.
* Otherwise we can 'forget' to update time for any but the last
@@ -3274,7 +3298,22 @@ static void ctx_sched_out(struct perf_event_context *ctx,
if (is_active & EVENT_TIME) {
/* update (and stop) ctx time */
update_context_time(ctx);
- update_cgrp_time_from_cpuctx(cpuctx);
+ update_cgrp_time_from_cpuctx(cpuctx, ctx == &cpuctx->ctx);
+ /*
+ * CPU-release for the below ->is_active store,
+ * see __load_acquire() in perf_event_time_now()
+ */
+ barrier();
+ }
+
+ ctx->is_active &= ~event_type;
+ if (!(ctx->is_active & EVENT_ALL))
+ ctx->is_active = 0;
+
+ if (ctx->task) {
+ WARN_ON_ONCE(cpuctx->task_ctx != ctx);
+ if (!ctx->is_active)
+ cpuctx->task_ctx = NULL;
}
is_active ^= ctx->is_active; /* changed bits */
@@ -3711,13 +3750,19 @@ static noinline int visit_groups_merge(struct perf_cpu_context *cpuctx,
return 0;
}
+/*
+ * Because the userpage is strictly per-event (there is no concept of context,
+ * so there cannot be a context indirection), every userpage must be updated
+ * when context time starts :-(
+ *
+ * IOW, we must not miss EVENT_TIME edges.
+ */
static inline bool event_update_userpage(struct perf_event *event)
{
if (likely(!atomic_read(&event->mmap_count)))
return false;
perf_event_update_time(event);
- perf_set_shadow_time(event, event->ctx);
perf_event_update_userpage(event);
return true;
@@ -3801,13 +3846,23 @@ ctx_sched_in(struct perf_event_context *ctx,
struct task_struct *task)
{
int is_active = ctx->is_active;
- u64 now;
lockdep_assert_held(&ctx->lock);
if (likely(!ctx->nr_events))
return;
+ if (is_active ^ EVENT_TIME) {
+ /* start ctx time */
+ __update_context_time(ctx, false);
+ perf_cgroup_set_timestamp(task, ctx);
+ /*
+ * CPU-release for the below ->is_active store,
+ * see __load_acquire() in perf_event_time_now()
+ */
+ barrier();
+ }
+
ctx->is_active |= (event_type | EVENT_TIME);
if (ctx->task) {
if (!is_active)
@@ -3818,13 +3873,6 @@ ctx_sched_in(struct perf_event_context *ctx,
is_active ^= ctx->is_active; /* changed bits */
- if (is_active & EVENT_TIME) {
- /* start ctx time */
- now = perf_clock();
- ctx->timestamp = now;
- perf_cgroup_set_timestamp(task, ctx);
- }
-
/*
* First go through the list and put on any pinned groups
* in order to give them the best chance of going on.
@@ -4418,6 +4466,18 @@ static inline u64 perf_event_count(struct perf_event *event)
return local64_read(&event->count) + atomic64_read(&event->child_count);
}
+static void calc_timer_values(struct perf_event *event,
+ u64 *now,
+ u64 *enabled,
+ u64 *running)
+{
+ u64 ctx_time;
+
+ *now = perf_clock();
+ ctx_time = perf_event_time_now(event, *now);
+ __perf_update_times(event, ctx_time, enabled, running);
+}
+
/*
* NMI-safe method to read a local event, that is an event that
* is:
@@ -4477,10 +4537,9 @@ int perf_event_read_local(struct perf_event *event, u64 *value,
*value = local64_read(&event->count);
if (enabled || running) {
- u64 now = event->shadow_ctx_time + perf_clock();
- u64 __enabled, __running;
+ u64 __enabled, __running, __now;;
- __perf_update_times(event, now, &__enabled, &__running);
+ calc_timer_values(event, &__now, &__enabled, &__running);
if (enabled)
*enabled = __enabled;
if (running)
@@ -5802,18 +5861,6 @@ static int perf_event_index(struct perf_event *event)
return event->pmu->event_idx(event);
}
-static void calc_timer_values(struct perf_event *event,
- u64 *now,
- u64 *enabled,
- u64 *running)
-{
- u64 ctx_time;
-
- *now = perf_clock();
- ctx_time = event->shadow_ctx_time + *now;
- __perf_update_times(event, ctx_time, enabled, running);
-}
-
static void perf_event_init_userpage(struct perf_event *event)
{
struct perf_event_mmap_page *userpg;
@@ -6353,7 +6400,6 @@ accounting:
ring_buffer_attach(event, rb);
perf_event_update_time(event);
- perf_set_shadow_time(event, event->ctx);
perf_event_init_userpage(event);
perf_event_update_userpage(event);
} else {
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index f7a986078213..330d49937692 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -978,8 +978,7 @@ static void memory_bm_recycle(struct memory_bitmap *bm)
* Register a range of page frames the contents of which should not be saved
* during hibernation (to be used in the early initialization code).
*/
-void __init __register_nosave_region(unsigned long start_pfn,
- unsigned long end_pfn, int use_kmalloc)
+void __init register_nosave_region(unsigned long start_pfn, unsigned long end_pfn)
{
struct nosave_region *region;
@@ -995,18 +994,12 @@ void __init __register_nosave_region(unsigned long start_pfn,
goto Report;
}
}
- if (use_kmalloc) {
- /* During init, this shouldn't fail */
- region = kmalloc(sizeof(struct nosave_region), GFP_KERNEL);
- BUG_ON(!region);
- } else {
- /* This allocation cannot fail */
- region = memblock_alloc(sizeof(struct nosave_region),
- SMP_CACHE_BYTES);
- if (!region)
- panic("%s: Failed to allocate %zu bytes\n", __func__,
- sizeof(struct nosave_region));
- }
+ /* This allocation cannot fail */
+ region = memblock_alloc(sizeof(struct nosave_region),
+ SMP_CACHE_BYTES);
+ if (!region)
+ panic("%s: Failed to allocate %zu bytes\n", __func__,
+ sizeof(struct nosave_region));
region->start_pfn = start_pfn;
region->end_pfn = end_pfn;
list_add_tail(&region->list, &nosave_regions);
diff --git a/kernel/power/wakelock.c b/kernel/power/wakelock.c
index 105df4dfc783..52571dcad768 100644
--- a/kernel/power/wakelock.c
+++ b/kernel/power/wakelock.c
@@ -39,23 +39,20 @@ ssize_t pm_show_wakelocks(char *buf, bool show_active)
{
struct rb_node *node;
struct wakelock *wl;
- char *str = buf;
- char *end = buf + PAGE_SIZE;
+ int len = 0;
mutex_lock(&wakelocks_lock);
for (node = rb_first(&wakelocks_tree); node; node = rb_next(node)) {
wl = rb_entry(node, struct wakelock, node);
if (wl->ws->active == show_active)
- str += scnprintf(str, end - str, "%s ", wl->name);
+ len += sysfs_emit_at(buf, len, "%s ", wl->name);
}
- if (str > buf)
- str--;
- str += scnprintf(str, end - str, "\n");
+ len += sysfs_emit_at(buf, len, "\n");
mutex_unlock(&wakelocks_lock);
- return (str - buf);
+ return len;
}
#if CONFIG_PM_WAKELOCKS_LIMIT > 0
diff --git a/kernel/rcu/tasks.h b/kernel/rcu/tasks.h
index 84f1d91604cc..d64f0b1d8cd3 100644
--- a/kernel/rcu/tasks.h
+++ b/kernel/rcu/tasks.h
@@ -123,7 +123,7 @@ static struct rcu_tasks rt_name = \
.call_func = call, \
.rtpcpu = &rt_name ## __percpu, \
.name = n, \
- .percpu_enqueue_shift = ilog2(CONFIG_NR_CPUS), \
+ .percpu_enqueue_shift = ilog2(CONFIG_NR_CPUS) + 1, \
.percpu_enqueue_lim = 1, \
.percpu_dequeue_lim = 1, \
.barrier_q_mutex = __MUTEX_INITIALIZER(rt_name.barrier_q_mutex), \
@@ -216,6 +216,7 @@ static void cblist_init_generic(struct rcu_tasks *rtp)
int cpu;
unsigned long flags;
int lim;
+ int shift;
raw_spin_lock_irqsave(&rtp->cbs_gbl_lock, flags);
if (rcu_task_enqueue_lim < 0) {
@@ -229,7 +230,10 @@ static void cblist_init_generic(struct rcu_tasks *rtp)
if (lim > nr_cpu_ids)
lim = nr_cpu_ids;
- WRITE_ONCE(rtp->percpu_enqueue_shift, ilog2(nr_cpu_ids / lim));
+ shift = ilog2(nr_cpu_ids / lim);
+ if (((nr_cpu_ids - 1) >> shift) >= lim)
+ shift++;
+ WRITE_ONCE(rtp->percpu_enqueue_shift, shift);
WRITE_ONCE(rtp->percpu_dequeue_lim, lim);
smp_store_release(&rtp->percpu_enqueue_lim, lim);
for_each_possible_cpu(cpu) {
@@ -298,7 +302,7 @@ static void call_rcu_tasks_generic(struct rcu_head *rhp, rcu_callback_t func,
if (unlikely(needadjust)) {
raw_spin_lock_irqsave(&rtp->cbs_gbl_lock, flags);
if (rtp->percpu_enqueue_lim != nr_cpu_ids) {
- WRITE_ONCE(rtp->percpu_enqueue_shift, ilog2(nr_cpu_ids));
+ WRITE_ONCE(rtp->percpu_enqueue_shift, ilog2(nr_cpu_ids) + 1);
WRITE_ONCE(rtp->percpu_dequeue_lim, nr_cpu_ids);
smp_store_release(&rtp->percpu_enqueue_lim, nr_cpu_ids);
pr_info("Switching %s to per-CPU callback queuing.\n", rtp->name);
@@ -413,7 +417,7 @@ static int rcu_tasks_need_gpcb(struct rcu_tasks *rtp)
if (rcu_task_cb_adjust && ncbs <= rcu_task_collapse_lim) {
raw_spin_lock_irqsave(&rtp->cbs_gbl_lock, flags);
if (rtp->percpu_enqueue_lim > 1) {
- WRITE_ONCE(rtp->percpu_enqueue_shift, ilog2(nr_cpu_ids));
+ WRITE_ONCE(rtp->percpu_enqueue_shift, ilog2(nr_cpu_ids) + 1);
smp_store_release(&rtp->percpu_enqueue_lim, 1);
rtp->percpu_dequeue_gpseq = get_state_synchronize_rcu();
pr_info("Starting switch %s to CPU-0 callback queuing.\n", rtp->name);
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 2e4ae00e52d1..848eaa0efe0e 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -5822,8 +5822,7 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
}
if (schedstat_enabled() && rq->core->core_forceidle_count) {
- if (cookie)
- rq->core->core_forceidle_start = rq_clock(rq->core);
+ rq->core->core_forceidle_start = rq_clock(rq->core);
rq->core->core_forceidle_occupation = occ;
}
@@ -8219,9 +8218,7 @@ int __cond_resched_lock(spinlock_t *lock)
if (spin_needbreak(lock) || resched) {
spin_unlock(lock);
- if (resched)
- preempt_schedule_common();
- else
+ if (!_cond_resched())
cpu_relax();
ret = 1;
spin_lock(lock);
@@ -8239,9 +8236,7 @@ int __cond_resched_rwlock_read(rwlock_t *lock)
if (rwlock_needbreak(lock) || resched) {
read_unlock(lock);
- if (resched)
- preempt_schedule_common();
- else
+ if (!_cond_resched())
cpu_relax();
ret = 1;
read_lock(lock);
@@ -8259,9 +8254,7 @@ int __cond_resched_rwlock_write(rwlock_t *lock)
if (rwlock_needbreak(lock) || resched) {
write_unlock(lock);
- if (resched)
- preempt_schedule_common();
- else
+ if (!_cond_resched())
cpu_relax();
ret = 1;
write_lock(lock);
diff --git a/kernel/sched/core_sched.c b/kernel/sched/core_sched.c
index 1fb45672ec85..c8746a9a7ada 100644
--- a/kernel/sched/core_sched.c
+++ b/kernel/sched/core_sched.c
@@ -277,7 +277,7 @@ void __sched_core_account_forceidle(struct rq *rq)
rq_i = cpu_rq(i);
p = rq_i->core_pick ?: rq_i->curr;
- if (!p->core_cookie)
+ if (p == rq_i->idle)
continue;
__schedstat_add(p->stats.core_forceidle_sum, delta);
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 095b0aa378df..5146163bfabb 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -3028,9 +3028,11 @@ enqueue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
static inline void
dequeue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
{
- u32 divider = get_pelt_divider(&se->avg);
sub_positive(&cfs_rq->avg.load_avg, se->avg.load_avg);
- cfs_rq->avg.load_sum = cfs_rq->avg.load_avg * divider;
+ sub_positive(&cfs_rq->avg.load_sum, se_weight(se) * se->avg.load_sum);
+ /* See update_cfs_rq_load_avg() */
+ cfs_rq->avg.load_sum = max_t(u32, cfs_rq->avg.load_sum,
+ cfs_rq->avg.load_avg * PELT_MIN_DIVIDER);
}
#else
static inline void
@@ -3381,7 +3383,6 @@ void set_task_rq_fair(struct sched_entity *se,
se->avg.last_update_time = n_last_update_time;
}
-
/*
* When on migration a sched_entity joins/leaves the PELT hierarchy, we need to
* propagate its contribution. The key to this propagation is the invariant
@@ -3449,15 +3450,14 @@ void set_task_rq_fair(struct sched_entity *se,
* XXX: only do this for the part of runnable > running ?
*
*/
-
static inline void
update_tg_cfs_util(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq *gcfs_rq)
{
- long delta = gcfs_rq->avg.util_avg - se->avg.util_avg;
- u32 divider;
+ long delta_sum, delta_avg = gcfs_rq->avg.util_avg - se->avg.util_avg;
+ u32 new_sum, divider;
/* Nothing to update */
- if (!delta)
+ if (!delta_avg)
return;
/*
@@ -3466,23 +3466,30 @@ update_tg_cfs_util(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq
*/
divider = get_pelt_divider(&cfs_rq->avg);
+
/* Set new sched_entity's utilization */
se->avg.util_avg = gcfs_rq->avg.util_avg;
- se->avg.util_sum = se->avg.util_avg * divider;
+ new_sum = se->avg.util_avg * divider;
+ delta_sum = (long)new_sum - (long)se->avg.util_sum;
+ se->avg.util_sum = new_sum;
/* Update parent cfs_rq utilization */
- add_positive(&cfs_rq->avg.util_avg, delta);
- cfs_rq->avg.util_sum = cfs_rq->avg.util_avg * divider;
+ add_positive(&cfs_rq->avg.util_avg, delta_avg);
+ add_positive(&cfs_rq->avg.util_sum, delta_sum);
+
+ /* See update_cfs_rq_load_avg() */
+ cfs_rq->avg.util_sum = max_t(u32, cfs_rq->avg.util_sum,
+ cfs_rq->avg.util_avg * PELT_MIN_DIVIDER);
}
static inline void
update_tg_cfs_runnable(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq *gcfs_rq)
{
- long delta = gcfs_rq->avg.runnable_avg - se->avg.runnable_avg;
- u32 divider;
+ long delta_sum, delta_avg = gcfs_rq->avg.runnable_avg - se->avg.runnable_avg;
+ u32 new_sum, divider;
/* Nothing to update */
- if (!delta)
+ if (!delta_avg)
return;
/*
@@ -3493,19 +3500,25 @@ update_tg_cfs_runnable(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cf
/* Set new sched_entity's runnable */
se->avg.runnable_avg = gcfs_rq->avg.runnable_avg;
- se->avg.runnable_sum = se->avg.runnable_avg * divider;
+ new_sum = se->avg.runnable_avg * divider;
+ delta_sum = (long)new_sum - (long)se->avg.runnable_sum;
+ se->avg.runnable_sum = new_sum;
/* Update parent cfs_rq runnable */
- add_positive(&cfs_rq->avg.runnable_avg, delta);
- cfs_rq->avg.runnable_sum = cfs_rq->avg.runnable_avg * divider;
+ add_positive(&cfs_rq->avg.runnable_avg, delta_avg);
+ add_positive(&cfs_rq->avg.runnable_sum, delta_sum);
+ /* See update_cfs_rq_load_avg() */
+ cfs_rq->avg.runnable_sum = max_t(u32, cfs_rq->avg.runnable_sum,
+ cfs_rq->avg.runnable_avg * PELT_MIN_DIVIDER);
}
static inline void
update_tg_cfs_load(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq *gcfs_rq)
{
- long delta, running_sum, runnable_sum = gcfs_rq->prop_runnable_sum;
+ long delta_avg, running_sum, runnable_sum = gcfs_rq->prop_runnable_sum;
unsigned long load_avg;
u64 load_sum = 0;
+ s64 delta_sum;
u32 divider;
if (!runnable_sum)
@@ -3532,7 +3545,7 @@ update_tg_cfs_load(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq
* assuming all tasks are equally runnable.
*/
if (scale_load_down(gcfs_rq->load.weight)) {
- load_sum = div_s64(gcfs_rq->avg.load_sum,
+ load_sum = div_u64(gcfs_rq->avg.load_sum,
scale_load_down(gcfs_rq->load.weight));
}
@@ -3549,19 +3562,22 @@ update_tg_cfs_load(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq
running_sum = se->avg.util_sum >> SCHED_CAPACITY_SHIFT;
runnable_sum = max(runnable_sum, running_sum);
- load_sum = (s64)se_weight(se) * runnable_sum;
- load_avg = div_s64(load_sum, divider);
-
- se->avg.load_sum = runnable_sum;
+ load_sum = se_weight(se) * runnable_sum;
+ load_avg = div_u64(load_sum, divider);
- delta = load_avg - se->avg.load_avg;
- if (!delta)
+ delta_avg = load_avg - se->avg.load_avg;
+ if (!delta_avg)
return;
- se->avg.load_avg = load_avg;
+ delta_sum = load_sum - (s64)se_weight(se) * se->avg.load_sum;
- add_positive(&cfs_rq->avg.load_avg, delta);
- cfs_rq->avg.load_sum = cfs_rq->avg.load_avg * divider;
+ se->avg.load_sum = runnable_sum;
+ se->avg.load_avg = load_avg;
+ add_positive(&cfs_rq->avg.load_avg, delta_avg);
+ add_positive(&cfs_rq->avg.load_sum, delta_sum);
+ /* See update_cfs_rq_load_avg() */
+ cfs_rq->avg.load_sum = max_t(u32, cfs_rq->avg.load_sum,
+ cfs_rq->avg.load_avg * PELT_MIN_DIVIDER);
}
static inline void add_tg_cfs_propagate(struct cfs_rq *cfs_rq, long runnable_sum)
@@ -3652,7 +3668,7 @@ static inline void add_tg_cfs_propagate(struct cfs_rq *cfs_rq, long runnable_sum
*
* cfs_rq->avg is used for task_h_load() and update_cfs_share() for example.
*
- * Returns true if the load decayed or we removed load.
+ * Return: 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.
@@ -3677,15 +3693,32 @@ update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq)
r = removed_load;
sub_positive(&sa->load_avg, r);
- sa->load_sum = sa->load_avg * divider;
+ sub_positive(&sa->load_sum, r * divider);
+ /* See sa->util_sum below */
+ sa->load_sum = max_t(u32, sa->load_sum, sa->load_avg * PELT_MIN_DIVIDER);
r = removed_util;
sub_positive(&sa->util_avg, r);
- sa->util_sum = sa->util_avg * divider;
+ sub_positive(&sa->util_sum, r * divider);
+ /*
+ * Because of rounding, se->util_sum might ends up being +1 more than
+ * cfs->util_sum. Although this is not a problem by itself, detaching
+ * a lot of tasks with the rounding problem between 2 updates of
+ * util_avg (~1ms) can make cfs->util_sum becoming null whereas
+ * cfs_util_avg is not.
+ * Check that util_sum is still above its lower bound for the new
+ * util_avg. Given that period_contrib might have moved since the last
+ * sync, we are only sure that util_sum must be above or equal to
+ * util_avg * minimum possible divider
+ */
+ sa->util_sum = max_t(u32, sa->util_sum, sa->util_avg * PELT_MIN_DIVIDER);
r = removed_runnable;
sub_positive(&sa->runnable_avg, r);
- sa->runnable_sum = sa->runnable_avg * divider;
+ sub_positive(&sa->runnable_sum, r * divider);
+ /* See sa->util_sum above */
+ sa->runnable_sum = max_t(u32, sa->runnable_sum,
+ sa->runnable_avg * PELT_MIN_DIVIDER);
/*
* removed_runnable is the unweighted version of removed_load so we
@@ -3772,17 +3805,18 @@ static void attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s
*/
static void detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
{
- /*
- * cfs_rq->avg.period_contrib can be used for both cfs_rq and se.
- * See ___update_load_avg() for details.
- */
- u32 divider = get_pelt_divider(&cfs_rq->avg);
-
dequeue_load_avg(cfs_rq, se);
sub_positive(&cfs_rq->avg.util_avg, se->avg.util_avg);
- cfs_rq->avg.util_sum = cfs_rq->avg.util_avg * divider;
+ sub_positive(&cfs_rq->avg.util_sum, se->avg.util_sum);
+ /* See update_cfs_rq_load_avg() */
+ cfs_rq->avg.util_sum = max_t(u32, cfs_rq->avg.util_sum,
+ cfs_rq->avg.util_avg * PELT_MIN_DIVIDER);
+
sub_positive(&cfs_rq->avg.runnable_avg, se->avg.runnable_avg);
- cfs_rq->avg.runnable_sum = cfs_rq->avg.runnable_avg * divider;
+ sub_positive(&cfs_rq->avg.runnable_sum, se->avg.runnable_sum);
+ /* See update_cfs_rq_load_avg() */
+ cfs_rq->avg.runnable_sum = max_t(u32, cfs_rq->avg.runnable_sum,
+ cfs_rq->avg.runnable_avg * PELT_MIN_DIVIDER);
add_tg_cfs_propagate(cfs_rq, -se->avg.load_sum);
@@ -8539,6 +8573,8 @@ group_type group_classify(unsigned int imbalance_pct,
*
* If @sg does not have SMT siblings, only pull tasks if all of the SMT siblings
* of @dst_cpu are idle and @sg has lower priority.
+ *
+ * Return: true if @dst_cpu can pull tasks, false otherwise.
*/
static bool asym_smt_can_pull_tasks(int dst_cpu, struct sd_lb_stats *sds,
struct sg_lb_stats *sgs,
@@ -8614,6 +8650,7 @@ sched_asym(struct lb_env *env, struct sd_lb_stats *sds, struct sg_lb_stats *sgs
/**
* update_sg_lb_stats - Update sched_group's statistics for load balancing.
* @env: The load balancing environment.
+ * @sds: Load-balancing data with statistics of the local group.
* @group: sched_group whose statistics are to be updated.
* @sgs: variable to hold the statistics for this group.
* @sg_status: Holds flag indicating the status of the sched_group
@@ -9421,12 +9458,11 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
/**
* find_busiest_group - Returns the busiest group within the sched_domain
* if there is an imbalance.
+ * @env: The load balancing environment.
*
* Also calculates the amount of runnable load which should be moved
* to restore balance.
*
- * @env: The load balancing environment.
- *
* Return: - The busiest group if imbalance exists.
*/
static struct sched_group *find_busiest_group(struct lb_env *env)
diff --git a/kernel/sched/pelt.h b/kernel/sched/pelt.h
index e06071bf3472..c336f5f481bc 100644
--- a/kernel/sched/pelt.h
+++ b/kernel/sched/pelt.h
@@ -37,9 +37,11 @@ update_irq_load_avg(struct rq *rq, u64 running)
}
#endif
+#define PELT_MIN_DIVIDER (LOAD_AVG_MAX - 1024)
+
static inline u32 get_pelt_divider(struct sched_avg *avg)
{
- return LOAD_AVG_MAX - 1024 + avg->period_contrib;
+ return PELT_MIN_DIVIDER + avg->period_contrib;
}
static inline void cfs_se_util_change(struct sched_avg *avg)
diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c
index a679613a7cb7..c137c4d6983e 100644
--- a/kernel/sched/psi.c
+++ b/kernel/sched/psi.c
@@ -1162,7 +1162,6 @@ struct psi_trigger *psi_trigger_create(struct psi_group *group,
t->event = 0;
t->last_event_time = 0;
init_waitqueue_head(&t->event_wait);
- kref_init(&t->refcount);
mutex_lock(&group->trigger_lock);
@@ -1191,15 +1190,19 @@ struct psi_trigger *psi_trigger_create(struct psi_group *group,
return t;
}
-static void psi_trigger_destroy(struct kref *ref)
+void psi_trigger_destroy(struct psi_trigger *t)
{
- struct psi_trigger *t = container_of(ref, struct psi_trigger, refcount);
- struct psi_group *group = t->group;
+ struct psi_group *group;
struct task_struct *task_to_destroy = NULL;
- if (static_branch_likely(&psi_disabled))
+ /*
+ * We do not check psi_disabled since it might have been disabled after
+ * the trigger got created.
+ */
+ if (!t)
return;
+ group = t->group;
/*
* Wakeup waiters to stop polling. Can happen if cgroup is deleted
* from under a polling process.
@@ -1235,9 +1238,9 @@ static void psi_trigger_destroy(struct kref *ref)
mutex_unlock(&group->trigger_lock);
/*
- * Wait for both *trigger_ptr from psi_trigger_replace and
- * poll_task RCUs to complete their read-side critical sections
- * before destroying the trigger and optionally the poll_task
+ * Wait for psi_schedule_poll_work RCU to complete its read-side
+ * critical section before destroying the trigger and optionally the
+ * poll_task.
*/
synchronize_rcu();
/*
@@ -1254,18 +1257,6 @@ static void psi_trigger_destroy(struct kref *ref)
kfree(t);
}
-void psi_trigger_replace(void **trigger_ptr, struct psi_trigger *new)
-{
- struct psi_trigger *old = *trigger_ptr;
-
- if (static_branch_likely(&psi_disabled))
- return;
-
- rcu_assign_pointer(*trigger_ptr, new);
- if (old)
- kref_put(&old->refcount, psi_trigger_destroy);
-}
-
__poll_t psi_trigger_poll(void **trigger_ptr,
struct file *file, poll_table *wait)
{
@@ -1275,24 +1266,15 @@ __poll_t psi_trigger_poll(void **trigger_ptr,
if (static_branch_likely(&psi_disabled))
return DEFAULT_POLLMASK | EPOLLERR | EPOLLPRI;
- rcu_read_lock();
-
- t = rcu_dereference(*(void __rcu __force **)trigger_ptr);
- if (!t) {
- rcu_read_unlock();
+ t = smp_load_acquire(trigger_ptr);
+ if (!t)
return DEFAULT_POLLMASK | EPOLLERR | EPOLLPRI;
- }
- kref_get(&t->refcount);
-
- rcu_read_unlock();
poll_wait(file, &t->event_wait, wait);
if (cmpxchg(&t->event, 1, 0) == 1)
ret |= EPOLLPRI;
- kref_put(&t->refcount, psi_trigger_destroy);
-
return ret;
}
@@ -1316,14 +1298,24 @@ static ssize_t psi_write(struct file *file, const char __user *user_buf,
buf[buf_size - 1] = '\0';
- new = psi_trigger_create(&psi_system, buf, nbytes, res);
- if (IS_ERR(new))
- return PTR_ERR(new);
-
seq = file->private_data;
+
/* Take seq->lock to protect seq->private from concurrent writes */
mutex_lock(&seq->lock);
- psi_trigger_replace(&seq->private, new);
+
+ /* Allow only one trigger per file descriptor */
+ if (seq->private) {
+ mutex_unlock(&seq->lock);
+ return -EBUSY;
+ }
+
+ new = psi_trigger_create(&psi_system, buf, nbytes, res);
+ if (IS_ERR(new)) {
+ mutex_unlock(&seq->lock);
+ return PTR_ERR(new);
+ }
+
+ smp_store_release(&seq->private, new);
mutex_unlock(&seq->lock);
return nbytes;
@@ -1358,7 +1350,7 @@ static int psi_fop_release(struct inode *inode, struct file *file)
{
struct seq_file *seq = file->private_data;
- psi_trigger_replace(&seq->private, NULL);
+ psi_trigger_destroy(seq->private);
return single_release(inode, file);
}
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index 752ed89a293b..a5eb5e7fd624 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -70,10 +70,16 @@ config HAVE_C_RECORDMCOUNT
help
C version of recordmcount available?
+config HAVE_BUILDTIME_MCOUNT_SORT
+ bool
+ help
+ An architecture selects this if it sorts the mcount_loc section
+ at build time.
+
config BUILDTIME_MCOUNT_SORT
bool
default y
- depends on BUILDTIME_TABLE_SORT && !S390
+ depends on HAVE_BUILDTIME_MCOUNT_SORT && DYNAMIC_FTRACE
help
Sort the mcount_loc section at build time.
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index a569a0cb81ee..c860f582b078 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -7740,7 +7740,8 @@ static struct tracing_log_err *get_tracing_log_err(struct trace_array *tr)
err = kzalloc(sizeof(*err), GFP_KERNEL);
if (!err)
err = ERR_PTR(-ENOMEM);
- tr->n_err_log_entries++;
+ else
+ tr->n_err_log_entries++;
return err;
}
diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c
index 5e6a988a8a51..ada87bfb5bb8 100644
--- a/kernel/trace/trace_events_hist.c
+++ b/kernel/trace/trace_events_hist.c
@@ -2503,6 +2503,8 @@ static struct hist_field *parse_unary(struct hist_trigger_data *hist_data,
(HIST_FIELD_FL_TIMESTAMP | HIST_FIELD_FL_TIMESTAMP_USECS);
expr->fn = hist_field_unary_minus;
expr->operands[0] = operand1;
+ expr->size = operand1->size;
+ expr->is_signed = operand1->is_signed;
expr->operator = FIELD_OP_UNARY_MINUS;
expr->name = expr_str(expr, 0);
expr->type = kstrdup_const(operand1->type, GFP_KERNEL);
@@ -2719,6 +2721,7 @@ static struct hist_field *parse_expr(struct hist_trigger_data *hist_data,
/* The operand sizes should be the same, so just pick one */
expr->size = operand1->size;
+ expr->is_signed = operand1->is_signed;
expr->operator = field_op;
expr->type = kstrdup_const(operand1->type, GFP_KERNEL);
@@ -3935,6 +3938,7 @@ static int trace_action_create(struct hist_trigger_data *hist_data,
var_ref_idx = find_var_ref_idx(hist_data, var_ref);
if (WARN_ON(var_ref_idx < 0)) {
+ kfree(p);
ret = var_ref_idx;
goto err;
}
@@ -6163,7 +6167,9 @@ static int event_hist_trigger_parse(struct event_command *cmd_ops,
lockdep_assert_held(&event_mutex);
- if (glob && strlen(glob)) {
+ WARN_ON(!glob);
+
+ if (strlen(glob)) {
hist_err_clear();
last_cmd_set(file, param);
}
@@ -6196,7 +6202,7 @@ static int event_hist_trigger_parse(struct event_command *cmd_ops,
continue;
}
break;
- } while (p);
+ } while (1);
if (!p)
param = NULL;
diff --git a/kernel/ucount.c b/kernel/ucount.c
index 7b32c356ebc5..65b597431c86 100644
--- a/kernel/ucount.c
+++ b/kernel/ucount.c
@@ -190,6 +190,7 @@ struct ucounts *alloc_ucounts(struct user_namespace *ns, kuid_t uid)
kfree(new);
} else {
hlist_add_head(&new->node, hashent);
+ get_user_ns(new->ns);
spin_unlock_irq(&ucounts_lock);
return new;
}
@@ -210,6 +211,7 @@ void put_ucounts(struct ucounts *ucounts)
if (atomic_dec_and_lock_irqsave(&ucounts->count, &ucounts_lock, flags)) {
hlist_del_init(&ucounts->node);
spin_unlock_irqrestore(&ucounts_lock, flags);
+ put_user_ns(ucounts->ns);
kfree(ucounts);
}
}