#include <linux/anon_inodes.h>
#include <linux/kernel_stat.h>
#include <linux/perf_event.h>
+#include <linux/ftrace_event.h>
+#include <linux/hw_breakpoint.h>
#include <asm/irq_regs.h>
&& !ctx1->pin_count && !ctx2->pin_count;
}
-static void __perf_event_read(void *event);
-
static void __perf_event_sync_stat(struct perf_event *event,
struct perf_event *next_event)
{
*/
switch (event->state) {
case PERF_EVENT_STATE_ACTIVE:
- __perf_event_read(event);
- break;
+ event->pmu->read(event);
+ /* fall-through */
case PERF_EVENT_STATE_INACTIVE:
update_event_times(event);
if (!ctx->nr_stat)
return;
+ update_context_time(ctx);
+
event = list_first_entry(&ctx->event_list,
struct perf_event, event_entry);
if (likely(!ctx || !cpuctx->task_ctx))
return;
- update_context_time(ctx);
-
rcu_read_lock();
parent = rcu_dereference(ctx->parent_ctx);
next_ctx = next->perf_event_ctxp;
struct perf_cpu_context *cpuctx = &__get_cpu_var(perf_cpu_context);
struct perf_event *event = info;
struct perf_event_context *ctx = event->ctx;
- unsigned long flags;
/*
* If this is a task context, we need to check whether it is
if (ctx->task && cpuctx->task_ctx != ctx)
return;
- local_irq_save(flags);
if (ctx->is_active)
update_context_time(ctx);
event->pmu->read(event);
update_event_times(event);
- local_irq_restore(flags);
}
static u64 perf_event_read(struct perf_event *event)
return ERR_PTR(err);
}
+static void perf_event_free_filter(struct perf_event *event);
+
static void free_event_rcu(struct rcu_head *head)
{
struct perf_event *event;
event = container_of(head, struct perf_event, rcu_head);
if (event->ns)
put_pid_ns(event->ns);
+ perf_event_free_filter(event);
kfree(event);
}
return 0;
}
+int perf_event_release_kernel(struct perf_event *event)
+{
+ struct perf_event_context *ctx = event->ctx;
+
+ WARN_ON_ONCE(ctx->parent_ctx);
+ mutex_lock(&ctx->mutex);
+ perf_event_remove_from_context(event);
+ mutex_unlock(&ctx->mutex);
+
+ mutex_lock(&event->owner->perf_event_mutex);
+ list_del_init(&event->owner_entry);
+ mutex_unlock(&event->owner->perf_event_mutex);
+ put_task_struct(event->owner);
+
+ free_event(event);
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(perf_event_release_kernel);
+
static int perf_event_read_size(struct perf_event *event)
{
int entry = sizeof(u64); /* value */
return size;
}
-static u64 perf_event_read_value(struct perf_event *event)
+u64 perf_event_read_value(struct perf_event *event)
{
struct perf_event *child;
u64 total = 0;
return total;
}
-
-static int perf_event_read_entry(struct perf_event *event,
- u64 read_format, char __user *buf)
-{
- int n = 0, count = 0;
- u64 values[2];
-
- values[n++] = perf_event_read_value(event);
- if (read_format & PERF_FORMAT_ID)
- values[n++] = primary_event_id(event);
-
- count = n * sizeof(u64);
-
- if (copy_to_user(buf, values, count))
- return -EFAULT;
-
- return count;
-}
+EXPORT_SYMBOL_GPL(perf_event_read_value);
static int perf_event_read_group(struct perf_event *event,
u64 read_format, char __user *buf)
{
struct perf_event *leader = event->group_leader, *sub;
- int n = 0, size = 0, err = -EFAULT;
- u64 values[3];
+ int n = 0, size = 0, ret = 0;
+ u64 values[5];
+ u64 count;
+
+ count = perf_event_read_value(leader);
values[n++] = 1 + leader->nr_siblings;
if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) {
values[n++] = leader->total_time_running +
atomic64_read(&leader->child_total_time_running);
}
+ values[n++] = count;
+ if (read_format & PERF_FORMAT_ID)
+ values[n++] = primary_event_id(leader);
size = n * sizeof(u64);
if (copy_to_user(buf, values, size))
return -EFAULT;
- err = perf_event_read_entry(leader, read_format, buf + size);
- if (err < 0)
- return err;
-
- size += err;
+ ret += size;
list_for_each_entry(sub, &leader->sibling_list, group_entry) {
- err = perf_event_read_entry(sub, read_format,
- buf + size);
- if (err < 0)
- return err;
+ n = 0;
+
+ values[n++] = perf_event_read_value(sub);
+ if (read_format & PERF_FORMAT_ID)
+ values[n++] = primary_event_id(sub);
+
+ size = n * sizeof(u64);
+
+ if (copy_to_user(buf + size, values, size))
+ return -EFAULT;
- size += err;
+ ret += size;
}
- return size;
+ return ret;
}
static int perf_event_read_one(struct perf_event *event,
return ret;
}
-int perf_event_set_output(struct perf_event *event, int output_fd);
+static int perf_event_set_output(struct perf_event *event, int output_fd);
+static int perf_event_set_filter(struct perf_event *event, void __user *arg);
static long perf_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
{
case PERF_EVENT_IOC_SET_OUTPUT:
return perf_event_set_output(event, arg);
+ case PERF_EVENT_IOC_SET_FILTER:
+ return perf_event_set_filter(event, (void __user *)arg);
+
default:
return -ENOTTY;
}
static void perf_output_lock(struct perf_output_handle *handle)
{
struct perf_mmap_data *data = handle->data;
- int cpu;
+ int cur, cpu = get_cpu();
handle->locked = 0;
- local_irq_save(handle->flags);
- cpu = smp_processor_id();
-
- if (in_nmi() && atomic_read(&data->lock) == cpu)
- return;
+ for (;;) {
+ cur = atomic_cmpxchg(&data->lock, -1, cpu);
+ if (cur == -1) {
+ handle->locked = 1;
+ break;
+ }
+ if (cur == cpu)
+ break;
- while (atomic_cmpxchg(&data->lock, -1, cpu) != -1)
cpu_relax();
-
- handle->locked = 1;
+ }
}
static void perf_output_unlock(struct perf_output_handle *handle)
if (atomic_xchg(&data->wakeup, 0))
perf_output_wakeup(handle);
out:
- local_irq_restore(handle->flags);
+ put_cpu();
}
void perf_output_copy(struct perf_output_handle *handle,
{
struct perf_event *event;
- if (system_state != SYSTEM_RUNNING || list_empty(&ctx->event_list))
- return;
-
- rcu_read_lock();
list_for_each_entry_rcu(event, &ctx->event_list, event_entry) {
if (perf_event_task_match(event))
perf_event_task_output(event, task_event);
}
- rcu_read_unlock();
}
static void perf_event_task_event(struct perf_task_event *task_event)
struct perf_cpu_context *cpuctx;
struct perf_event_context *ctx = task_event->task_ctx;
+ rcu_read_lock();
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)
{
struct perf_event *event;
- if (system_state != SYSTEM_RUNNING || list_empty(&ctx->event_list))
- return;
-
- rcu_read_lock();
list_for_each_entry_rcu(event, &ctx->event_list, event_entry) {
if (perf_event_comm_match(event))
perf_event_comm_output(event, comm_event);
}
- rcu_read_unlock();
}
static void perf_event_comm_event(struct perf_comm_event *comm_event)
comm_event->event_id.header.size = sizeof(comm_event->event_id) + size;
+ rcu_read_lock();
cpuctx = &get_cpu_var(perf_cpu_context);
perf_event_comm_ctx(&cpuctx->ctx, comm_event);
put_cpu_var(perf_cpu_context);
- rcu_read_lock();
/*
* doesn't really matter which of the child contexts the
* events ends up in.
{
struct perf_event *event;
- if (system_state != SYSTEM_RUNNING || list_empty(&ctx->event_list))
- return;
-
- rcu_read_lock();
list_for_each_entry_rcu(event, &ctx->event_list, event_entry) {
if (perf_event_mmap_match(event, mmap_event))
perf_event_mmap_output(event, mmap_event);
}
- rcu_read_unlock();
}
static void perf_event_mmap_event(struct perf_mmap_event *mmap_event)
mmap_event->event_id.header.size = sizeof(mmap_event->event_id) + size;
+ rcu_read_lock();
cpuctx = &get_cpu_var(perf_cpu_context);
perf_event_mmap_ctx(&cpuctx->ctx, mmap_event);
put_cpu_var(perf_cpu_context);
- rcu_read_lock();
/*
* doesn't really matter which of the child contexts the
* events ends up in.
perf_event_disable(event);
}
- perf_event_output(event, nmi, data, regs);
+ if (event->overflow_handler)
+ event->overflow_handler(event, nmi, data, regs);
+ else
+ perf_event_output(event, nmi, data, regs);
+
return ret;
}
return nr;
}
-static void perf_swevent_overflow(struct perf_event *event,
+static void perf_swevent_overflow(struct perf_event *event, u64 overflow,
int nmi, struct perf_sample_data *data,
struct pt_regs *regs)
{
struct hw_perf_event *hwc = &event->hw;
int throttle = 0;
- u64 overflow;
data->period = event->hw.last_period;
- overflow = perf_swevent_set_period(event);
+ if (!overflow)
+ overflow = perf_swevent_set_period(event);
if (hwc->interrupts == MAX_INTERRUPTS)
return;
atomic64_add(nr, &event->count);
+ if (!regs)
+ return;
+
if (!hwc->sample_period)
return;
- if (!regs)
+ if (nr == 1 && hwc->sample_period == 1 && !event->attr.freq)
+ return perf_swevent_overflow(event, 1, nmi, data, regs);
+
+ if (atomic64_add_negative(nr, &hwc->period_left))
return;
- if (!atomic64_add_negative(nr, &hwc->period_left))
- perf_swevent_overflow(event, nmi, data, regs);
+ perf_swevent_overflow(event, 0, nmi, data, regs);
}
static int perf_swevent_is_counting(struct perf_event *event)
return 1;
}
+static int perf_tp_event_match(struct perf_event *event,
+ struct perf_sample_data *data);
+
static int perf_swevent_match(struct perf_event *event,
enum perf_type_id type,
- u32 event_id, struct pt_regs *regs)
+ u32 event_id,
+ struct perf_sample_data *data,
+ struct pt_regs *regs)
{
if (!perf_swevent_is_counting(event))
return 0;
return 0;
}
+ if (event->attr.type == PERF_TYPE_TRACEPOINT &&
+ !perf_tp_event_match(event, data))
+ return 0;
+
return 1;
}
{
struct perf_event *event;
- if (system_state != SYSTEM_RUNNING || list_empty(&ctx->event_list))
- return;
-
- rcu_read_lock();
list_for_each_entry_rcu(event, &ctx->event_list, event_entry) {
- if (perf_swevent_match(event, type, event_id, regs))
+ if (perf_swevent_match(event, type, event_id, data, regs))
perf_swevent_add(event, nr, nmi, data, regs);
}
- rcu_read_unlock();
}
static int *perf_swevent_recursion_context(struct perf_cpu_context *cpuctx)
(*recursion)++;
barrier();
+ rcu_read_lock();
perf_swevent_ctx_event(&cpuctx->ctx, type, event_id,
nr, nmi, data, regs);
- rcu_read_lock();
/*
* doesn't really matter which of the child contexts the
* events ends up in.
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);
}
};
#ifdef CONFIG_EVENT_PROFILE
+
void perf_tp_event(int event_id, u64 addr, u64 count, void *record,
int entry_size)
{
}
EXPORT_SYMBOL_GPL(perf_tp_event);
-extern int ftrace_profile_enable(int);
-extern void ftrace_profile_disable(int);
+static int perf_tp_event_match(struct perf_event *event,
+ struct perf_sample_data *data)
+{
+ void *record = data->raw->data;
+
+ if (likely(!event->filter) || filter_match_preds(event->filter, record))
+ return 1;
+ return 0;
+}
static void tp_perf_event_destroy(struct perf_event *event)
{
return &perf_ops_generic;
}
+
+static int perf_event_set_filter(struct perf_event *event, void __user *arg)
+{
+ char *filter_str;
+ int ret;
+
+ if (event->attr.type != PERF_TYPE_TRACEPOINT)
+ return -EINVAL;
+
+ filter_str = strndup_user(arg, PAGE_SIZE);
+ if (IS_ERR(filter_str))
+ return PTR_ERR(filter_str);
+
+ ret = ftrace_profile_set_filter(event, event->attr.config, filter_str);
+
+ kfree(filter_str);
+ return ret;
+}
+
+static void perf_event_free_filter(struct perf_event *event)
+{
+ ftrace_profile_free_filter(event);
+}
+
#else
+
+static int perf_tp_event_match(struct perf_event *event,
+ struct perf_sample_data *data)
+{
+ return 1;
+}
+
static const struct pmu *tp_perf_event_init(struct perf_event *event)
{
return NULL;
}
+
+static int perf_event_set_filter(struct perf_event *event, void __user *arg)
+{
+ return -ENOENT;
+}
+
+static void perf_event_free_filter(struct perf_event *event)
+{
+}
+
+#endif /* CONFIG_EVENT_PROFILE */
+
+#ifdef CONFIG_HAVE_HW_BREAKPOINT
+static void bp_perf_event_destroy(struct perf_event *event)
+{
+ release_bp_slot(event);
+}
+
+static const struct pmu *bp_perf_event_init(struct perf_event *bp)
+{
+ int err;
+ /*
+ * The breakpoint is already filled if we haven't created the counter
+ * through perf syscall
+ * FIXME: manage to get trigerred to NULL if it comes from syscalls
+ */
+ if (!bp->callback)
+ err = register_perf_hw_breakpoint(bp);
+ else
+ err = __register_perf_hw_breakpoint(bp);
+ if (err)
+ return ERR_PTR(err);
+
+ bp->destroy = bp_perf_event_destroy;
+
+ return &perf_ops_bp;
+}
+
+void perf_bp_event(struct perf_event *bp, void *regs)
+{
+ /* TODO */
+}
+#else
+static void bp_perf_event_destroy(struct perf_event *event)
+{
+}
+
+static const struct pmu *bp_perf_event_init(struct perf_event *bp)
+{
+ return NULL;
+}
+
+void perf_bp_event(struct perf_event *bp, void *regs)
+{
+}
#endif
atomic_t perf_swevent_enabled[PERF_COUNT_SW_MAX];
case PERF_COUNT_SW_PAGE_FAULTS_MAJ:
case PERF_COUNT_SW_CONTEXT_SWITCHES:
case PERF_COUNT_SW_CPU_MIGRATIONS:
+ case PERF_COUNT_SW_ALIGNMENT_FAULTS:
+ case PERF_COUNT_SW_EMULATION_FAULTS:
if (!event->parent) {
atomic_inc(&perf_swevent_enabled[event_id]);
event->destroy = sw_perf_event_destroy;
struct perf_event_context *ctx,
struct perf_event *group_leader,
struct perf_event *parent_event,
+ perf_callback_t callback,
gfp_t gfpflags)
{
const struct pmu *pmu;
event->state = PERF_EVENT_STATE_INACTIVE;
+ if (!callback && parent_event)
+ callback = parent_event->callback;
+
+ event->callback = callback;
+
if (attr->disabled)
event->state = PERF_EVENT_STATE_OFF;
pmu = tp_perf_event_init(event);
break;
+ case PERF_TYPE_BREAKPOINT:
+ pmu = bp_perf_event_init(event);
+ break;
+
+
default:
break;
}
goto out;
}
-int perf_event_set_output(struct perf_event *event, int output_fd)
+static int perf_event_set_output(struct perf_event *event, int output_fd)
{
struct perf_event *output_event = NULL;
struct file *output_file = NULL;
}
event = perf_event_alloc(&attr, cpu, ctx, group_leader,
- NULL, GFP_KERNEL);
+ NULL, NULL, GFP_KERNEL);
err = PTR_ERR(event);
if (IS_ERR(event))
goto err_put_context;
return err;
}
+/**
+ * perf_event_create_kernel_counter
+ *
+ * @attr: attributes of the counter to create
+ * @cpu: cpu in which the counter is bound
+ * @pid: task to profile
+ */
+struct perf_event *
+perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
+ pid_t pid, perf_callback_t callback)
+{
+ struct perf_event *event;
+ struct perf_event_context *ctx;
+ int err;
+
+ /*
+ * Get the target context (task or percpu):
+ */
+
+ ctx = find_get_context(pid, cpu);
+ if (IS_ERR(ctx))
+ return NULL;
+
+ event = perf_event_alloc(attr, cpu, ctx, NULL,
+ NULL, callback, GFP_KERNEL);
+ err = PTR_ERR(event);
+ if (IS_ERR(event))
+ goto err_put_context;
+
+ event->filp = NULL;
+ WARN_ON_ONCE(ctx->parent_ctx);
+ mutex_lock(&ctx->mutex);
+ perf_install_in_context(ctx, event, cpu);
+ ++ctx->generation;
+ mutex_unlock(&ctx->mutex);
+
+ event->owner = current;
+ get_task_struct(current);
+ mutex_lock(¤t->perf_event_mutex);
+ list_add_tail(&event->owner_entry, ¤t->perf_event_list);
+ mutex_unlock(¤t->perf_event_mutex);
+
+ return event;
+
+err_put_context:
+ if (err < 0)
+ put_ctx(ctx);
+
+ return NULL;
+}
+EXPORT_SYMBOL_GPL(perf_event_create_kernel_counter);
+
/*
* inherit a event from parent task to child task:
*/
child_event = perf_event_alloc(&parent_event->attr,
parent_event->cpu, child_ctx,
group_leader, parent_event,
- GFP_KERNEL);
+ NULL, GFP_KERNEL);
if (IS_ERR(child_event))
return child_event;
get_ctx(child_ctx);
if (parent_event->attr.freq)
child_event->hw.sample_period = parent_event->hw.sample_period;
+ child_event->overflow_handler = parent_event->overflow_handler;
+
/*
* Link it up in the child's context:
*/