return sysctl_perf_event_paranoid > 1;
}
-int sysctl_perf_event_mlock __read_mostly = 512; /* 'free' kb per user */
+/* Minimum for 128 pages + 1 for the user control page */
+int sysctl_perf_event_mlock __read_mostly = 516; /* 'free' kb per user */
/*
* max perf event sample rate
if (event->state != PERF_EVENT_STATE_ACTIVE)
continue;
+ if (event->cpu != -1 && event->cpu != smp_processor_id())
+ continue;
+
hwc = &event->hw;
interrupts = hwc->interrupts;
if (perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN))
return ERR_PTR(-EACCES);
- if (cpu < 0 || cpu > num_possible_cpus())
+ if (cpu < 0 || cpu >= nr_cpumask_bits)
return ERR_PTR(-EINVAL);
/*
perf_mmap_free_page((unsigned long)data->user_page);
for (i = 0; i < data->nr_pages; i++)
perf_mmap_free_page((unsigned long)data->data_pages[i]);
+ kfree(data);
}
#else
perf_mmap_unmark_page(base + (i * PAGE_SIZE));
vfree(base);
+ kfree(data);
}
static void perf_mmap_data_free(struct perf_mmap_data *data)
data = container_of(rcu_head, struct perf_mmap_data, rcu_head);
perf_mmap_data_free(data);
- kfree(data);
}
static void perf_mmap_data_release(struct perf_event *event)
static int perf_event_task_match(struct perf_event *event)
{
+ if (event->state != PERF_EVENT_STATE_ACTIVE)
+ return 0;
+
+ if (event->cpu != -1 && event->cpu != smp_processor_id())
+ return 0;
+
if (event->attr.comm || event->attr.mmap || event->attr.task)
return 1;
cpuctx = &get_cpu_var(perf_cpu_context);
perf_event_task_ctx(&cpuctx->ctx, task_event);
- put_cpu_var(perf_cpu_context);
rcu_read_lock();
if (!ctx)
ctx = rcu_dereference(task_event->task->perf_event_ctxp);
if (ctx)
perf_event_task_ctx(ctx, task_event);
+ put_cpu_var(perf_cpu_context);
rcu_read_unlock();
}
static int perf_event_comm_match(struct perf_event *event)
{
+ if (event->state != PERF_EVENT_STATE_ACTIVE)
+ return 0;
+
+ if (event->cpu != -1 && event->cpu != smp_processor_id())
+ return 0;
+
if (event->attr.comm)
return 1;
cpuctx = &get_cpu_var(perf_cpu_context);
perf_event_comm_ctx(&cpuctx->ctx, comm_event);
- put_cpu_var(perf_cpu_context);
rcu_read_lock();
/*
ctx = rcu_dereference(current->perf_event_ctxp);
if (ctx)
perf_event_comm_ctx(ctx, comm_event);
+ put_cpu_var(perf_cpu_context);
rcu_read_unlock();
}
static int perf_event_mmap_match(struct perf_event *event,
struct perf_mmap_event *mmap_event)
{
+ if (event->state != PERF_EVENT_STATE_ACTIVE)
+ return 0;
+
+ if (event->cpu != -1 && event->cpu != smp_processor_id())
+ return 0;
+
if (event->attr.mmap)
return 1;
cpuctx = &get_cpu_var(perf_cpu_context);
perf_event_mmap_ctx(&cpuctx->ctx, mmap_event);
- put_cpu_var(perf_cpu_context);
rcu_read_lock();
/*
ctx = rcu_dereference(current->perf_event_ctxp);
if (ctx)
perf_event_mmap_ctx(ctx, mmap_event);
+ put_cpu_var(perf_cpu_context);
rcu_read_unlock();
kfree(buf);
enum perf_type_id type,
u32 event_id, struct pt_regs *regs)
{
+ if (event->cpu != -1 && event->cpu != smp_processor_id())
+ return 0;
+
if (!perf_swevent_is_counting(event))
return 0;
event->pmu->read(event);
data.addr = 0;
+ data.period = event->hw.last_period;
regs = get_irq_regs();
/*
* In case we exclude kernel IPs or are somehow not in interrupt
regs = task_pt_regs(current);
if (regs) {
- if (perf_event_overflow(event, 0, &data, regs))
- ret = HRTIMER_NORESTART;
+ if (!(event->attr.exclude_idle && current->pid == 0))
+ if (perf_event_overflow(event, 0, &data, regs))
+ ret = HRTIMER_NORESTART;
}
period = max_t(u64, 10000, event->hw.sample_period);
return ret;
}
+static void perf_swevent_start_hrtimer(struct perf_event *event)
+{
+ struct hw_perf_event *hwc = &event->hw;
+
+ hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+ hwc->hrtimer.function = perf_swevent_hrtimer;
+ if (hwc->sample_period) {
+ u64 period;
+
+ if (hwc->remaining) {
+ if (hwc->remaining < 0)
+ period = 10000;
+ else
+ period = hwc->remaining;
+ hwc->remaining = 0;
+ } else {
+ period = max_t(u64, 10000, hwc->sample_period);
+ }
+ __hrtimer_start_range_ns(&hwc->hrtimer,
+ ns_to_ktime(period), 0,
+ HRTIMER_MODE_REL, 0);
+ }
+}
+
+static void perf_swevent_cancel_hrtimer(struct perf_event *event)
+{
+ struct hw_perf_event *hwc = &event->hw;
+
+ if (hwc->sample_period) {
+ ktime_t remaining = hrtimer_get_remaining(&hwc->hrtimer);
+ hwc->remaining = ktime_to_ns(remaining);
+
+ hrtimer_cancel(&hwc->hrtimer);
+ }
+}
+
/*
* Software event: cpu wall time clock
*/
int cpu = raw_smp_processor_id();
atomic64_set(&hwc->prev_count, cpu_clock(cpu));
- hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
- hwc->hrtimer.function = perf_swevent_hrtimer;
- if (hwc->sample_period) {
- u64 period = max_t(u64, 10000, hwc->sample_period);
- __hrtimer_start_range_ns(&hwc->hrtimer,
- ns_to_ktime(period), 0,
- HRTIMER_MODE_REL, 0);
- }
+ perf_swevent_start_hrtimer(event);
return 0;
}
static void cpu_clock_perf_event_disable(struct perf_event *event)
{
- if (event->hw.sample_period)
- hrtimer_cancel(&event->hw.hrtimer);
+ perf_swevent_cancel_hrtimer(event);
cpu_clock_perf_event_update(event);
}
now = event->ctx->time;
atomic64_set(&hwc->prev_count, now);
- hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
- hwc->hrtimer.function = perf_swevent_hrtimer;
- if (hwc->sample_period) {
- u64 period = max_t(u64, 10000, hwc->sample_period);
- __hrtimer_start_range_ns(&hwc->hrtimer,
- ns_to_ktime(period), 0,
- HRTIMER_MODE_REL, 0);
- }
+
+ perf_swevent_start_hrtimer(event);
return 0;
}
static void task_clock_perf_event_disable(struct perf_event *event)
{
- if (event->hw.sample_period)
- hrtimer_cancel(&event->hw.hrtimer);
+ perf_swevent_cancel_hrtimer(event);
task_clock_perf_event_update(event, event->ctx->time);
}
struct perf_event_context *ctx;
struct file *event_file = NULL;
struct file *group_file = NULL;
+ int event_fd;
int fput_needed = 0;
- int fput_needed2 = 0;
int err;
/* for future expandability... */
return -EINVAL;
}
+ event_fd = get_unused_fd_flags(O_RDWR);
+ if (event_fd < 0)
+ return event_fd;
+
/*
* Get the target context (task or percpu):
*/
ctx = find_get_context(pid, cpu);
- if (IS_ERR(ctx))
- return PTR_ERR(ctx);
+ if (IS_ERR(ctx)) {
+ err = PTR_ERR(ctx);
+ goto err_fd;
+ }
/*
* Look up the group leader (we will attach this event to it):
if (IS_ERR(event))
goto err_put_context;
- err = anon_inode_getfd("[perf_event]", &perf_fops, event, 0);
- if (err < 0)
- goto err_free_put_context;
-
- event_file = fget_light(err, &fput_needed2);
- if (!event_file)
+ event_file = anon_inode_getfile("[perf_event]", &perf_fops, event, O_RDWR);
+ if (IS_ERR(event_file)) {
+ err = PTR_ERR(event_file);
goto err_free_put_context;
+ }
if (flags & PERF_FLAG_FD_OUTPUT) {
err = perf_event_set_output(event, group_fd);
list_add_tail(&event->owner_entry, ¤t->perf_event_list);
mutex_unlock(¤t->perf_event_mutex);
-err_fput_free_put_context:
- fput_light(event_file, fput_needed2);
+ fput_light(group_file, fput_needed);
+ fd_install(event_fd, event_file);
+ return event_fd;
+err_fput_free_put_context:
+ fput(event_file);
err_free_put_context:
- if (err < 0)
- kfree(event);
-
+ free_event(event);
err_put_context:
- if (err < 0)
- put_ctx(ctx);
-
fput_light(group_file, fput_needed);
-
+ put_ctx(ctx);
+err_fd:
+ put_unused_fd(event_fd);
return err;
}
return ret;
}
+static void __init perf_event_init_all_cpus(void)
+{
+ int cpu;
+ struct perf_cpu_context *cpuctx;
+
+ for_each_possible_cpu(cpu) {
+ cpuctx = &per_cpu(perf_cpu_context, cpu);
+ __perf_event_init_context(&cpuctx->ctx, NULL);
+ }
+}
+
static void __cpuinit perf_event_init_cpu(int cpu)
{
struct perf_cpu_context *cpuctx;
cpuctx = &per_cpu(perf_cpu_context, cpu);
- __perf_event_init_context(&cpuctx->ctx, NULL);
spin_lock(&perf_resource_lock);
cpuctx->max_pertask = perf_max_events - perf_reserved_percpu;
void __init perf_event_init(void)
{
+ perf_event_init_all_cpus();
perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_UP_PREPARE,
(void *)(long)smp_processor_id());
perf_cpu_notify(&perf_cpu_nb, (unsigned long)CPU_ONLINE,