TOMOYO Linux Cross Reference
Linux/kernel/events/core.c

   // SPDX-License-Identifier: GPL-2.0
   /*
    * Performance events core code:
    *
    *  Copyright (C) 2008 Thomas Gleixner <tglx@linutronix.de>
    *  Copyright (C) 2008-2011 Red Hat, Inc., Ingo Molnar
    *  Copyright (C) 2008-2011 Red Hat, Inc., Peter Zijlstra
    *  Copyright  ©  2009 Paul Mackerras, IBM Corp. <paulus@au1.ibm.com>
    */
  
  #include <linux/fs.h>
  #include <linux/mm.h>
  #include <linux/cpu.h>
  #include <linux/smp.h>
  #include <linux/idr.h>
  #include <linux/file.h>
  #include <linux/poll.h>
  #include <linux/slab.h>
  #include <linux/hash.h>
  #include <linux/tick.h>
  #include <linux/sysfs.h>
  #include <linux/dcache.h>
  #include <linux/percpu.h>
  #include <linux/ptrace.h>
  #include <linux/reboot.h>
  #include <linux/vmstat.h>
  #include <linux/device.h>
  #include <linux/export.h>
  #include <linux/vmalloc.h>
  #include <linux/hardirq.h>
  #include <linux/hugetlb.h>
  #include <linux/rculist.h>
  #include <linux/uaccess.h>
  #include <linux/syscalls.h>
  #include <linux/anon_inodes.h>
  #include <linux/kernel_stat.h>
  #include <linux/cgroup.h>
  #include <linux/perf_event.h>
  #include <linux/trace_events.h>
  #include <linux/hw_breakpoint.h>
  #include <linux/mm_types.h>
  #include <linux/module.h>
  #include <linux/mman.h>
  #include <linux/compat.h>
  #include <linux/bpf.h>
  #include <linux/filter.h>
  #include <linux/namei.h>
  #include <linux/parser.h>
  #include <linux/sched/clock.h>
  #include <linux/sched/mm.h>
  #include <linux/proc_ns.h>
  #include <linux/mount.h>
  #include <linux/min_heap.h>
  #include <linux/highmem.h>
  #include <linux/pgtable.h>
  #include <linux/buildid.h>
  
  #include "internal.h"
  
  #include <asm/irq_regs.h>
  
  typedef int (*remote_function_f)(void *);
  
  struct remote_function_call {
          struct task_struct      *p;
          remote_function_f       func;
          void                    *info;
          int                     ret;
  };
  
  static void remote_function(void *data)
  {
          struct remote_function_call *tfc = data;
          struct task_struct *p = tfc->p;
  
          if (p) {
                  /* -EAGAIN */
                  if (task_cpu(p) != smp_processor_id())
                          return;
  
                  /*
                   * Now that we're on right CPU with IRQs disabled, we can test
                   * if we hit the right task without races.
                   */
  
                  tfc->ret = -ESRCH; /* No such (running) process */
                  if (p != current)
                          return;
          }
  
          tfc->ret = tfc->func(tfc->info);
  }
  
  /**
   * task_function_call - call a function on the cpu on which a task runs
   * @p:          the task to evaluate
   * @func:       the function to be called
   * @info:       the function call argument
   *
  * Calls the function @func when the task is currently running. This might
  * be on the current CPU, which just calls the function directly.  This will
  * retry due to any failures in smp_call_function_single(), such as if the
  * task_cpu() goes offline concurrently.
  *
  * returns @func return value or -ESRCH or -ENXIO when the process isn't running
  */
 static int
 task_function_call(struct task_struct *p, remote_function_f func, void *info)
 {
         struct remote_function_call data = {
                 .p      = p,
                 .func   = func,
                 .info   = info,
                 .ret    = -EAGAIN,
         };
         int ret;
 
         for (;;) {
                 ret = smp_call_function_single(task_cpu(p), remote_function,
                                                &data, 1);
                 if (!ret)
                         ret = data.ret;
 
                 if (ret != -EAGAIN)
                         break;
 
                 cond_resched();
         }
 
         return ret;
 }
 
 /**
  * cpu_function_call - call a function on the cpu
  * @cpu:        target cpu to queue this function
  * @func:       the function to be called
  * @info:       the function call argument
  *
  * Calls the function @func on the remote cpu.
  *
  * returns: @func return value or -ENXIO when the cpu is offline
  */
 static int cpu_function_call(int cpu, remote_function_f func, void *info)
 {
         struct remote_function_call data = {
                 .p      = NULL,
                 .func   = func,
                 .info   = info,
                 .ret    = -ENXIO, /* No such CPU */
         };
 
         smp_call_function_single(cpu, remote_function, &data, 1);
 
         return data.ret;
 }
 
 static inline struct perf_cpu_context *
 __get_cpu_context(struct perf_event_context *ctx)
 {
         return this_cpu_ptr(ctx->pmu->pmu_cpu_context);
 }
 
 static void perf_ctx_lock(struct perf_cpu_context *cpuctx,
                           struct perf_event_context *ctx)
 {
         raw_spin_lock(&cpuctx->ctx.lock);
         if (ctx)
                 raw_spin_lock(&ctx->lock);
 }
 
 static void perf_ctx_unlock(struct perf_cpu_context *cpuctx,
                             struct perf_event_context *ctx)
 {
         if (ctx)
                 raw_spin_unlock(&ctx->lock);
         raw_spin_unlock(&cpuctx->ctx.lock);
 }
 
 #define TASK_TOMBSTONE ((void *)-1L)
 
 static bool is_kernel_event(struct perf_event *event)
 {
         return READ_ONCE(event->owner) == TASK_TOMBSTONE;
 }
 
 /*
  * On task ctx scheduling...
  *
  * When !ctx->nr_events a task context will not be scheduled. This means
  * we can disable the scheduler hooks (for performance) without leaving
  * pending task ctx state.
  *
  * This however results in two special cases:
  *
  *  - removing the last event from a task ctx; this is relatively straight
  *    forward and is done in __perf_remove_from_context.
  *
  *  - adding the first event to a task ctx; this is tricky because we cannot
  *    rely on ctx->is_active and therefore cannot use event_function_call().
  *    See perf_install_in_context().
  *
  * If ctx->nr_events, then ctx->is_active and cpuctx->task_ctx are set.
  */
 
 typedef void (*event_f)(struct perf_event *, struct perf_cpu_context *,
                         struct perf_event_context *, void *);
 
 struct event_function_struct {
         struct perf_event *event;
         event_f func;
         void *data;
 };
 
 static int event_function(void *info)
 {
         struct event_function_struct *efs = info;
         struct perf_event *event = efs->event;
         struct perf_event_context *ctx = event->ctx;
         struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
         struct perf_event_context *task_ctx = cpuctx->task_ctx;
         int ret = 0;
 
         lockdep_assert_irqs_disabled();
 
         perf_ctx_lock(cpuctx, task_ctx);
         /*
          * Since we do the IPI call without holding ctx->lock things can have
          * changed, double check we hit the task we set out to hit.
          */
         if (ctx->task) {
                 if (ctx->task != current) {
                         ret = -ESRCH;
                         goto unlock;
                 }
 
                 /*
                  * We only use event_function_call() on established contexts,
                  * and event_function() is only ever called when active (or
                  * rather, we'll have bailed in task_function_call() or the
                  * above ctx->task != current test), therefore we must have
                  * ctx->is_active here.
                  */
                 WARN_ON_ONCE(!ctx->is_active);
                 /*
                  * And since we have ctx->is_active, cpuctx->task_ctx must
                  * match.
                  */
                 WARN_ON_ONCE(task_ctx != ctx);
         } else {
                 WARN_ON_ONCE(&cpuctx->ctx != ctx);
         }
 
         efs->func(event, cpuctx, ctx, efs->data);
 unlock:
         perf_ctx_unlock(cpuctx, task_ctx);
 
         return ret;
 }
 
 static void event_function_call(struct perf_event *event, event_f func, void *data)
 {
         struct perf_event_context *ctx = event->ctx;
         struct task_struct *task = READ_ONCE(ctx->task); /* verified in event_function */
         struct event_function_struct efs = {
                 .event = event,
                 .func = func,
                 .data = data,
         };
 
         if (!event->parent) {
                 /*
                  * If this is a !child event, we must hold ctx::mutex to
                  * stabilize the event->ctx relation. See
                  * perf_event_ctx_lock().
                  */
                 lockdep_assert_held(&ctx->mutex);
         }
 
         if (!task) {
                 cpu_function_call(event->cpu, event_function, &efs);
                 return;
         }
 
         if (task == TASK_TOMBSTONE)
                 return;
 
 again:
         if (!task_function_call(task, event_function, &efs))
                 return;
 
         raw_spin_lock_irq(&ctx->lock);
         /*
          * Reload the task pointer, it might have been changed by
          * a concurrent perf_event_context_sched_out().
          */
         task = ctx->task;
         if (task == TASK_TOMBSTONE) {
                 raw_spin_unlock_irq(&ctx->lock);
                 return;
         }
         if (ctx->is_active) {
                 raw_spin_unlock_irq(&ctx->lock);
                 goto again;
         }
         func(event, NULL, ctx, data);
         raw_spin_unlock_irq(&ctx->lock);
 }
 
 /*
  * Similar to event_function_call() + event_function(), but hard assumes IRQs
  * are already disabled and we're on the right CPU.
  */
 static void event_function_local(struct perf_event *event, event_f func, void *data)
 {
         struct perf_event_context *ctx = event->ctx;
         struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
         struct task_struct *task = READ_ONCE(ctx->task);
         struct perf_event_context *task_ctx = NULL;
 
         lockdep_assert_irqs_disabled();
 
         if (task) {
                 if (task == TASK_TOMBSTONE)
                         return;
 
                 task_ctx = ctx;
         }
 
         perf_ctx_lock(cpuctx, task_ctx);
 
         task = ctx->task;
         if (task == TASK_TOMBSTONE)
                 goto unlock;
 
         if (task) {
                 /*
                  * We must be either inactive or active and the right task,
                  * otherwise we're screwed, since we cannot IPI to somewhere
                  * else.
                  */
                 if (ctx->is_active) {
                         if (WARN_ON_ONCE(task != current))
                                 goto unlock;
 
                         if (WARN_ON_ONCE(cpuctx->task_ctx != ctx))
                                 goto unlock;
                 }
         } else {
                 WARN_ON_ONCE(&cpuctx->ctx != ctx);
         }
 
         func(event, cpuctx, ctx, data);
 unlock:
         perf_ctx_unlock(cpuctx, task_ctx);
 }
 
 #define PERF_FLAG_ALL (PERF_FLAG_FD_NO_GROUP |\
                        PERF_FLAG_FD_OUTPUT  |\
                        PERF_FLAG_PID_CGROUP |\
                        PERF_FLAG_FD_CLOEXEC)
 
 /*
  * branch priv levels that need permission checks
  */
 #define PERF_SAMPLE_BRANCH_PERM_PLM \
         (PERF_SAMPLE_BRANCH_KERNEL |\
          PERF_SAMPLE_BRANCH_HV)
 
 enum event_type_t {
         EVENT_FLEXIBLE = 0x1,
         EVENT_PINNED = 0x2,
         EVENT_TIME = 0x4,
         /* see ctx_resched() for details */
         EVENT_CPU = 0x8,
         EVENT_ALL = EVENT_FLEXIBLE | EVENT_PINNED,
 };
 
 /*
  * perf_sched_events : >0 events exist
  * perf_cgroup_events: >0 per-cpu cgroup events exist on this cpu
  */
 
 static void perf_sched_delayed(struct work_struct *work);
 DEFINE_STATIC_KEY_FALSE(perf_sched_events);
 static DECLARE_DELAYED_WORK(perf_sched_work, perf_sched_delayed);
 static DEFINE_MUTEX(perf_sched_mutex);
 static atomic_t perf_sched_count;
 
 static DEFINE_PER_CPU(atomic_t, perf_cgroup_events);
 static DEFINE_PER_CPU(int, perf_sched_cb_usages);
 static DEFINE_PER_CPU(struct pmu_event_list, pmu_sb_events);
 
 static atomic_t nr_mmap_events __read_mostly;
 static atomic_t nr_comm_events __read_mostly;
 static atomic_t nr_namespaces_events __read_mostly;
 static atomic_t nr_task_events __read_mostly;
 static atomic_t nr_freq_events __read_mostly;
 static atomic_t nr_switch_events __read_mostly;
 static atomic_t nr_ksymbol_events __read_mostly;
 static atomic_t nr_bpf_events __read_mostly;
 static atomic_t nr_cgroup_events __read_mostly;
 static atomic_t nr_text_poke_events __read_mostly;
 static atomic_t nr_build_id_events __read_mostly;
 
 static LIST_HEAD(pmus);
 static DEFINE_MUTEX(pmus_lock);
 static struct srcu_struct pmus_srcu;
 static cpumask_var_t perf_online_mask;
 static struct kmem_cache *perf_event_cache;
 
 /*
  * perf event paranoia level:
  *  -1 - not paranoid at all
  *   0 - disallow raw tracepoint access for unpriv
  *   1 - disallow cpu events for unpriv
  *   2 - disallow kernel profiling for unpriv
  */
 int sysctl_perf_event_paranoid __read_mostly = 2;
 
 /* Minimum for 512 kiB + 1 user control page */
 int sysctl_perf_event_mlock __read_mostly = 512 + (PAGE_SIZE / 1024); /* 'free' kiB per user */
 
 /*
  * max perf event sample rate
  */
 #define DEFAULT_MAX_SAMPLE_RATE         100000
 #define DEFAULT_SAMPLE_PERIOD_NS        (NSEC_PER_SEC / DEFAULT_MAX_SAMPLE_RATE)
 #define DEFAULT_CPU_TIME_MAX_PERCENT    25
 
 int sysctl_perf_event_sample_rate __read_mostly = DEFAULT_MAX_SAMPLE_RATE;
 
 static int max_samples_per_tick __read_mostly   = DIV_ROUND_UP(DEFAULT_MAX_SAMPLE_RATE, HZ);
 static int perf_sample_period_ns __read_mostly  = DEFAULT_SAMPLE_PERIOD_NS;
 
 static int perf_sample_allowed_ns __read_mostly =
         DEFAULT_SAMPLE_PERIOD_NS * DEFAULT_CPU_TIME_MAX_PERCENT / 100;
 
 static void update_perf_cpu_limits(void)
 {
         u64 tmp = perf_sample_period_ns;
 
         tmp *= sysctl_perf_cpu_time_max_percent;
         tmp = div_u64(tmp, 100);
         if (!tmp)
                 tmp = 1;
 
         WRITE_ONCE(perf_sample_allowed_ns, tmp);
 }
 
 static bool perf_rotate_context(struct perf_cpu_context *cpuctx);
 
 int perf_proc_update_handler(struct ctl_table *table, int write,
                 void *buffer, size_t *lenp, loff_t *ppos)
 {
         int ret;
         int perf_cpu = sysctl_perf_cpu_time_max_percent;
         /*
          * If throttling is disabled don't allow the write:
          */
         if (write && (perf_cpu == 100 || perf_cpu == 0))
                 return -EINVAL;
 
         ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
         if (ret || !write)
                 return ret;
 
         max_samples_per_tick = DIV_ROUND_UP(sysctl_perf_event_sample_rate, HZ);
         perf_sample_period_ns = NSEC_PER_SEC / sysctl_perf_event_sample_rate;
         update_perf_cpu_limits();
 
         return 0;
 }
 
 int sysctl_perf_cpu_time_max_percent __read_mostly = DEFAULT_CPU_TIME_MAX_PERCENT;
 
 int perf_cpu_time_max_percent_handler(struct ctl_table *table, int write,
                 void *buffer, size_t *lenp, loff_t *ppos)
 {
         int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
 
         if (ret || !write)
                 return ret;
 
         if (sysctl_perf_cpu_time_max_percent == 100 ||
             sysctl_perf_cpu_time_max_percent == 0) {
                 printk(KERN_WARNING
                        "perf: Dynamic interrupt throttling disabled, can hang your system!\n");
                 WRITE_ONCE(perf_sample_allowed_ns, 0);
         } else {
                 update_perf_cpu_limits();
         }
 
         return 0;
 }
 
 /*
  * perf samples are done in some very critical code paths (NMIs).
  * If they take too much CPU time, the system can lock up and not
  * get any real work done.  This will drop the sample rate when
  * we detect that events are taking too long.
  */
 #define NR_ACCUMULATED_SAMPLES 128
 static DEFINE_PER_CPU(u64, running_sample_length);
 
 static u64 __report_avg;
 static u64 __report_allowed;
 
 static void perf_duration_warn(struct irq_work *w)
 {
         printk_ratelimited(KERN_INFO
                 "perf: interrupt took too long (%lld > %lld), lowering "
                 "kernel.perf_event_max_sample_rate to %d\n",
                 __report_avg, __report_allowed,
                 sysctl_perf_event_sample_rate);
 }
 
 static DEFINE_IRQ_WORK(perf_duration_work, perf_duration_warn);
 
 void perf_sample_event_took(u64 sample_len_ns)
 {
         u64 max_len = READ_ONCE(perf_sample_allowed_ns);
         u64 running_len;
         u64 avg_len;
         u32 max;
 
         if (max_len == 0)
                 return;
 
         /* Decay the counter by 1 average sample. */
         running_len = __this_cpu_read(running_sample_length);
         running_len -= running_len/NR_ACCUMULATED_SAMPLES;
         running_len += sample_len_ns;
         __this_cpu_write(running_sample_length, running_len);
 
         /*
          * Note: this will be biased artifically low until we have
          * seen NR_ACCUMULATED_SAMPLES. Doing it this way keeps us
          * from having to maintain a count.
          */
         avg_len = running_len/NR_ACCUMULATED_SAMPLES;
         if (avg_len <= max_len)
                 return;
 
         __report_avg = avg_len;
         __report_allowed = max_len;
 
         /*
          * Compute a throttle threshold 25% below the current duration.
          */
         avg_len += avg_len / 4;
         max = (TICK_NSEC / 100) * sysctl_perf_cpu_time_max_percent;
         if (avg_len < max)
                 max /= (u32)avg_len;
         else
                 max = 1;
 
         WRITE_ONCE(perf_sample_allowed_ns, avg_len);
         WRITE_ONCE(max_samples_per_tick, max);
 
         sysctl_perf_event_sample_rate = max * HZ;
         perf_sample_period_ns = NSEC_PER_SEC / sysctl_perf_event_sample_rate;
 
         if (!irq_work_queue(&perf_duration_work)) {
                 early_printk("perf: interrupt took too long (%lld > %lld), lowering "
                              "kernel.perf_event_max_sample_rate to %d\n",
                              __report_avg, __report_allowed,
                              sysctl_perf_event_sample_rate);
         }
 }
 
 static atomic64_t perf_event_id;
 
 static void cpu_ctx_sched_out(struct perf_cpu_context *cpuctx,
                               enum event_type_t event_type);
 
 static void cpu_ctx_sched_in(struct perf_cpu_context *cpuctx,
                              enum event_type_t event_type);
 
 static void update_context_time(struct perf_event_context *ctx);
 static u64 perf_event_time(struct perf_event *event);
 
 void __weak perf_event_print_debug(void)        { }
 
 static inline u64 perf_clock(void)
 {
         return local_clock();
 }
 
 static inline u64 perf_event_clock(struct perf_event *event)
 {
         return event->clock();
 }
 
 /*
  * State based event timekeeping...
  *
  * The basic idea is to use event->state to determine which (if any) time
  * fields to increment with the current delta. This means we only need to
  * update timestamps when we change state or when they are explicitly requested
  * (read).
  *
  * Event groups make things a little more complicated, but not terribly so. The
  * rules for a group are that if the group leader is OFF the entire group is
  * OFF, irrespecive of what the group member states are. This results in
  * __perf_effective_state().
  *
  * A futher ramification is that when a group leader flips between OFF and
  * !OFF, we need to update all group member times.
  *
  *
  * NOTE: perf_event_time() is based on the (cgroup) context time, and thus we
  * need to make sure the relevant context time is updated before we try and
  * update our timestamps.
  */
 
 static __always_inline enum perf_event_state
 __perf_effective_state(struct perf_event *event)
 {
         struct perf_event *leader = event->group_leader;
 
         if (leader->state <= PERF_EVENT_STATE_OFF)
                 return leader->state;
 
         return event->state;
 }
 
 static __always_inline void
 __perf_update_times(struct perf_event *event, u64 now, u64 *enabled, u64 *running)
 {
         enum perf_event_state state = __perf_effective_state(event);
         u64 delta = now - event->tstamp;
 
         *enabled = event->total_time_enabled;
         if (state >= PERF_EVENT_STATE_INACTIVE)
                 *enabled += delta;
 
         *running = event->total_time_running;
         if (state >= PERF_EVENT_STATE_ACTIVE)
                 *running += delta;
 }
 
 static void perf_event_update_time(struct perf_event *event)
 {
         u64 now = perf_event_time(event);
 
         __perf_update_times(event, now, &event->total_time_enabled,
                                         &event->total_time_running);
         event->tstamp = now;
 }
 
 static void perf_event_update_sibling_time(struct perf_event *leader)
 {
         struct perf_event *sibling;
 
         for_each_sibling_event(sibling, leader)
                 perf_event_update_time(sibling);
 }
 
 static void
 perf_event_set_state(struct perf_event *event, enum perf_event_state state)
 {
         if (event->state == state)
                 return;
 
         perf_event_update_time(event);
         /*
          * If a group leader gets enabled/disabled all its siblings
          * are affected too.
          */
         if ((event->state < 0) ^ (state < 0))
                 perf_event_update_sibling_time(event);
 
         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
 perf_cgroup_match(struct perf_event *event)
 {
         struct perf_event_context *ctx = event->ctx;
         struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
         /* @event doesn't care about cgroup */
         if (!event->cgrp)
                 return true;
 
         /* wants specific cgroup scope but @cpuctx isn't associated with any */
         if (!cpuctx->cgrp)
                 return false;
 
         /*
          * Cgroup scoping is recursive.  An event enabled for a cgroup is
          * also enabled for all its descendant cgroups.  If @cpuctx's
          * cgroup is a descendant of @event's (the test covers identity
          * case), it's a match.
          */
         return cgroup_is_descendant(cpuctx->cgrp->css.cgroup,
                                     event->cgrp->css.cgroup);
 }
 
 static inline void perf_detach_cgroup(struct perf_event *event)
 {
         css_put(&event->cgrp->css);
         event->cgrp = NULL;
 }
 
 static inline int is_cgroup_event(struct perf_event *event)
 {
         return event->cgrp != NULL;
 }
 
 static inline u64 perf_cgroup_event_time(struct perf_event *event)
 {
         struct perf_cgroup_info *t;
 
         t = per_cpu_ptr(event->cgrp->info, event->cpu);
         return t->time;
 }
 
 static inline u64 perf_cgroup_event_time_now(struct perf_event *event, u64 now)
 {
         struct perf_cgroup_info *t;
 
         t = per_cpu_ptr(event->cgrp->info, event->cpu);
         if (!__load_acquire(&t->active))
                 return t->time;
         now += READ_ONCE(t->timeoffset);
         return now;
 }
 
 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, 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);
                         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;
 
         /*
          * ensure we access cgroup data only when needed and
          * when we know the cgroup is pinned (css_get)
          */
         if (!is_cgroup_event(event))
                 return;
 
         info = this_cpu_ptr(event->cgrp->info);
         /*
          * Do not update time when cgroup is not active
          */
         if (info->active)
                 __update_cgrp_time(info, perf_clock(), true);
 }
 
 static inline void
 perf_cgroup_set_timestamp(struct perf_cpu_context *cpuctx)
 {
         struct perf_event_context *ctx = &cpuctx->ctx;
         struct perf_cgroup *cgrp = cpuctx->cgrp;
         struct perf_cgroup_info *info;
         struct cgroup_subsys_state *css;
 
         /*
          * ctx->lock held by caller
          * ensure we do not access cgroup data
          * unless we have the cgroup pinned (css_get)
          */
         if (!cgrp)
                 return;
 
         WARN_ON_ONCE(!ctx->nr_cgroups);
 
         for (css = &cgrp->css; css; css = css->parent) {
                 cgrp = container_of(css, struct perf_cgroup, css);
                 info = this_cpu_ptr(cgrp->info);
                 __update_cgrp_time(info, ctx->timestamp, false);
                 __store_release(&info->active, 1);
         }
 }
 
 static DEFINE_PER_CPU(struct list_head, cgrp_cpuctx_list);
 
 /*
  * reschedule events based on the cgroup constraint of task.
  */
 static void perf_cgroup_switch(struct task_struct *task)
 {
         struct perf_cgroup *cgrp;
         struct perf_cpu_context *cpuctx, *tmp;
         struct list_head *list;
         unsigned long flags;
 
         /*
          * Disable interrupts and preemption to avoid this CPU's
          * cgrp_cpuctx_entry to change under us.
          */
         local_irq_save(flags);
 
         cgrp = perf_cgroup_from_task(task, NULL);
 
         list = this_cpu_ptr(&cgrp_cpuctx_list);
         list_for_each_entry_safe(cpuctx, tmp, list, cgrp_cpuctx_entry) {
                 WARN_ON_ONCE(cpuctx->ctx.nr_cgroups == 0);
                 if (READ_ONCE(cpuctx->cgrp) == cgrp)
                         continue;
 
                 perf_ctx_lock(cpuctx, cpuctx->task_ctx);
                 perf_pmu_disable(cpuctx->ctx.pmu);
 
                 cpu_ctx_sched_out(cpuctx, EVENT_ALL);
                 /*
                  * must not be done before ctxswout due
                  * to update_cgrp_time_from_cpuctx() in
                  * ctx_sched_out()
                  */
                 cpuctx->cgrp = cgrp;
                 /*
                  * set cgrp before ctxsw in to allow
                  * perf_cgroup_set_timestamp() in ctx_sched_in()
                  * to not have to pass task around
                  */
                 cpu_ctx_sched_in(cpuctx, EVENT_ALL);
 
                 perf_pmu_enable(cpuctx->ctx.pmu);
                 perf_ctx_unlock(cpuctx, cpuctx->task_ctx);
         }
 
         local_irq_restore(flags);
 }
 
 static int perf_cgroup_ensure_storage(struct perf_event *event,
                                 struct cgroup_subsys_state *css)
 {
         struct perf_cpu_context *cpuctx;
         struct perf_event **storage;
         int cpu, heap_size, ret = 0;
 
         /*
          * Allow storage to have sufficent space for an iterator for each
          * possibly nested cgroup plus an iterator for events with no cgroup.
          */
         for (heap_size = 1; css; css = css->parent)
                 heap_size++;
 
         for_each_possible_cpu(cpu) {
                 cpuctx = per_cpu_ptr(event->pmu->pmu_cpu_context, cpu);
                 if (heap_size <= cpuctx->heap_size)
                         continue;
 
                 storage = kmalloc_node(heap_size * sizeof(struct perf_event *),
                                        GFP_KERNEL, cpu_to_node(cpu));
                 if (!storage) {
                         ret = -ENOMEM;
                         break;
                 }
 
                 raw_spin_lock_irq(&cpuctx->ctx.lock);
                 if (cpuctx->heap_size < heap_size) {
                         swap(cpuctx->heap, storage);
                         if (storage == cpuctx->heap_default)
                                 storage = NULL;
                         cpuctx->heap_size = heap_size;
                 }
                 raw_spin_unlock_irq(&cpuctx->ctx.lock);
 
                 kfree(storage);
         }
 
         return ret;
 }
 
 static inline int perf_cgroup_connect(int fd, struct perf_event *event,
                                       struct perf_event_attr *attr,
                                       struct perf_event *group_leader)
 {
         struct perf_cgroup *cgrp;
         struct cgroup_subsys_state *css;
         struct fd f = fdget(fd);
         int ret = 0;
 
         if (!f.file)
                 return -EBADF;
 
         css = css_tryget_online_from_dir(f.file->f_path.dentry,
                                          &perf_event_cgrp_subsys);
         if (IS_ERR(css)) {
                 ret = PTR_ERR(css);
                 goto out;
         }
 
         ret = perf_cgroup_ensure_storage(event, css);
         if (ret)
                 goto out;
 
         cgrp = container_of(css, struct perf_cgroup, css);
         event->cgrp = cgrp;
 
         /*
          * all events in a group must monitor
          * the same cgroup because a task belongs
          * to only one perf cgroup at a time
          */
         if (group_leader && group_leader->cgrp != cgrp) {
                 perf_detach_cgroup(event);
                 ret = -EINVAL;
         }
 out:
         fdput(f);
         return ret;
 }
 
 static inline void
 perf_cgroup_event_enable(struct perf_event *event, struct perf_event_context *ctx)
 {
         struct perf_cpu_context *cpuctx;
 
         if (!is_cgroup_event(event))
                 return;
 
         /*
          * Because cgroup events are always per-cpu events,
          * @ctx == &cpuctx->ctx.
          */
         cpuctx = container_of(ctx, struct perf_cpu_context, ctx);
 
         if (ctx->nr_cgroups++)
                 return;
 
         cpuctx->cgrp = perf_cgroup_from_task(current, ctx);
         list_add(&cpuctx->cgrp_cpuctx_entry,
                         per_cpu_ptr(&cgrp_cpuctx_list, event->cpu));
 }
 
 static inline void
 perf_cgroup_event_disable(struct perf_event *event, struct perf_event_context *ctx)
 {
         struct perf_cpu_context *cpuctx;
 
         if (!is_cgroup_event(event))
                 return;
 
         /*
          * Because cgroup events are always per-cpu events,
          * @ctx == &cpuctx->ctx.
          */
         cpuctx = container_of(ctx, struct perf_cpu_context, ctx);
 
         if (--ctx->nr_cgroups)
                 return;
 
         cpuctx->cgrp = NULL;
         list_del(&cpuctx->cgrp_cpuctx_entry);
 }
 
 #else /* !CONFIG_CGROUP_PERF */
 
 static inline bool
 perf_cgroup_match(struct perf_event *event)
 {
         return true;
 }
 
 static inline void perf_detach_cgroup(struct perf_event *event)
 {}
 
 static inline int is_cgroup_event(struct perf_event *event)
 {
         return 0;
 }
 
 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,
                                                 bool final)
 {
 }
 
 static inline int perf_cgroup_connect(pid_t pid, struct perf_event *event,
                                       struct perf_event_attr *attr,
                                       struct perf_event *group_leader)
 {
         return -EINVAL;
 }
 
 static inline void
 perf_cgroup_set_timestamp(struct perf_cpu_context *cpuctx)
 {
 }
 
 static inline u64 perf_cgroup_event_time(struct perf_event *event)
 {
         return 0;
 }
 
 static inline u64 perf_cgroup_event_time_now(struct perf_event *event, u64 now)
 {
         return 0;
 }
 
 static inline void
 perf_cgroup_event_enable(struct perf_event *event, struct perf_event_context *ctx)
 {
 }
 
 static inline void
 perf_cgroup_event_disable(struct perf_event *event, struct perf_event_context *ctx)
 {
 }
 
 static void perf_cgroup_switch(struct task_struct *task)
 {
 }
 #endif
 
 /*
  * set default to be dependent on timer tick just
  * like original code
  */
 #define PERF_CPU_HRTIMER (1000 / HZ)
 /*
  * function must be called with interrupts disabled
  */
 static enum hrtimer_restart perf_mux_hrtimer_handler(struct hrtimer *hr)
 {
         struct perf_cpu_context *cpuctx;
         bool rotations;
 
         lockdep_assert_irqs_disabled();
 
         cpuctx = container_of(hr, struct perf_cpu_context, hrtimer);
         rotations = perf_rotate_context(cpuctx);
 
         raw_spin_lock(&cpuctx->hrtimer_lock);
         if (rotations)
                 hrtimer_forward_now(hr, cpuctx->hrtimer_interval);
         else
                 cpuctx->hrtimer_active = 0;
         raw_spin_unlock(&cpuctx->hrtimer_lock);
 
         return rotations ? HRTIMER_RESTART : HRTIMER_NORESTART;
 }
 
 static void __perf_mux_hrtimer_init(struct perf_cpu_context *cpuctx, int cpu)
 {
         struct hrtimer *timer = &cpuctx->hrtimer;
         struct pmu *pmu = cpuctx->ctx.pmu;
         u64 interval;
 
         /* no multiplexing needed for SW PMU */
         if (pmu->task_ctx_nr == perf_sw_context)
                 return;
 
         /*
          * check default is sane, if not set then force to
          * default interval (1/tick)
          */
         interval = pmu->hrtimer_interval_ms;
         if (interval < 1)
                 interval = pmu->hrtimer_interval_ms = PERF_CPU_HRTIMER;
 
         cpuctx->hrtimer_interval = ns_to_ktime(NSEC_PER_MSEC * interval);
 
         raw_spin_lock_init(&cpuctx->hrtimer_lock);
         hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED_HARD);
         timer->function = perf_mux_hrtimer_handler;
 }
 
 static int perf_mux_hrtimer_restart(struct perf_cpu_context *cpuctx)
 {
         struct hrtimer *timer = &cpuctx->hrtimer;
         struct pmu *pmu = cpuctx->ctx.pmu;
         unsigned long flags;
 
         /* not for SW PMU */
         if (pmu->task_ctx_nr == perf_sw_context)
                 return 0;
 
         raw_spin_lock_irqsave(&cpuctx->hrtimer_lock, flags);
         if (!cpuctx->hrtimer_active) {
                 cpuctx->hrtimer_active = 1;
                 hrtimer_forward_now(timer, cpuctx->hrtimer_interval);
                 hrtimer_start_expires(timer, HRTIMER_MODE_ABS_PINNED_HARD);
         }
         raw_spin_unlock_irqrestore(&cpuctx->hrtimer_lock, flags);
 
         return 0;
 }
 
 void perf_pmu_disable(struct pmu *pmu)
 {
         int *count = this_cpu_ptr(pmu->pmu_disable_count);
         if (!(*count)++)
                 pmu->pmu_disable(pmu);
 }
 
 void perf_pmu_enable(struct pmu *pmu)
 {
         int *count = this_cpu_ptr(pmu->pmu_disable_count);
         if (!--(*count))
                 pmu->pmu_enable(pmu);
 }
 
 static DEFINE_PER_CPU(struct list_head, active_ctx_list);
 
 /*
  * perf_event_ctx_activate(), perf_event_ctx_deactivate(), and
  * perf_event_task_tick() are fully serialized because they're strictly cpu
  * affine and perf_event_ctx{activate,deactivate} are called with IRQs
  * disabled, while perf_event_task_tick is called from IRQ context.
  */
 static void perf_event_ctx_activate(struct perf_event_context *ctx)
 {
         struct list_head *head = this_cpu_ptr(&active_ctx_list);
 
         lockdep_assert_irqs_disabled();
 
         WARN_ON(!list_empty(&ctx->active_ctx_list));
 
         list_add(&ctx->active_ctx_list, head);
 }
 
 static void perf_event_ctx_deactivate(struct perf_event_context *ctx)
 {
         lockdep_assert_irqs_disabled();
 
         WARN_ON(list_empty(&ctx->active_ctx_list));
 
         list_del_init(&ctx->active_ctx_list);
 }
 
 static void get_ctx(struct perf_event_context *ctx)
 {
         refcount_inc(&ctx->refcount);
 }
 
 static void *alloc_task_ctx_data(struct pmu *pmu)
 {
         if (pmu->task_ctx_cache)
                 return kmem_cache_zalloc(pmu->task_ctx_cache, GFP_KERNEL);
 
         return NULL;
 }
 
 static void free_task_ctx_data(struct pmu *pmu, void *task_ctx_data)
 {
         if (pmu->task_ctx_cache && task_ctx_data)
                 kmem_cache_free(pmu->task_ctx_cache, task_ctx_data);
 }
 
 static void free_ctx(struct rcu_head *head)
 {
         struct perf_event_context *ctx;
 
         ctx = container_of(head, struct perf_event_context, rcu_head);
         free_task_ctx_data(ctx->pmu, ctx->task_ctx_data);
         kfree(ctx);
 }
 
 static void put_ctx(struct perf_event_context *ctx)
 {
         if (refcount_dec_and_test(&ctx->refcount)) {
                 if (ctx->parent_ctx)
                         put_ctx(ctx->parent_ctx);
                 if (ctx->task && ctx->task != TASK_TOMBSTONE)
                         put_task_struct(ctx->task);
                 call_rcu(&ctx->rcu_head, free_ctx);
         }
 }
 
 /*
  * Because of perf_event::ctx migration in sys_perf_event_open::move_group and
  * perf_pmu_migrate_context() we need some magic.
  *
  * Those places that change perf_event::ctx will hold both
  * perf_event_ctx::mutex of the 'old' and 'new' ctx value.
  *
  * Lock ordering is by mutex address. There are two other sites where
  * perf_event_context::mutex nests and those are:
  *
  *  - perf_event_exit_task_context()    [ child , 0 ]
  *      perf_event_exit_event()
  *        put_event()                   [ parent, 1 ]
  *
  *  - perf_event_init_context()         [ parent, 0 ]
  *      inherit_task_group()
  *        inherit_group()
  *          inherit_event()
  *            perf_event_alloc()
  *              perf_init_event()
  *                perf_try_init_event() [ child , 1 ]
  *
  * While it appears there is an obvious deadlock here -- the parent and child
  * nesting levels are inverted between the two. This is in fact safe because
  * life-time rules separate them. That is an exiting task cannot fork, and a
  * spawning task cannot (yet) exit.
  *
  * But remember that these are parent<->child context relations, and
  * migration does not affect children, therefore these two orderings should not
  * interact.
  *
  * The change in perf_event::ctx does not affect children (as claimed above)
  * because the sys_perf_event_open() case will install a new event and break
  * the ctx parent<->child relation, and perf_pmu_migrate_context() is only
  * concerned with cpuctx and that doesn't have children.
  *
  * The places that change perf_event::ctx will issue:
  *
  *   perf_remove_from_context();
  *   synchronize_rcu();
  *   perf_install_in_context();
  *
  * to affect the change. The remove_from_context() + synchronize_rcu() should
  * quiesce the event, after which we can install it in the new location. This
  * means that only external vectors (perf_fops, prctl) can perturb the event
  * while in transit. Therefore all such accessors should also acquire
  * perf_event_context::mutex to serialize against this.
  *
  * However; because event->ctx can change while we're waiting to acquire
  * ctx->mutex we must be careful and use the below perf_event_ctx_lock()
  * function.
  *
  * Lock order:
  *    exec_update_lock
  *      task_struct::perf_event_mutex
  *        perf_event_context::mutex
  *          perf_event::child_mutex;
  *            perf_event_context::lock
  *          perf_event::mmap_mutex
  *          mmap_lock
  *            perf_addr_filters_head::lock
  *
  *    cpu_hotplug_lock
  *      pmus_lock
  *        cpuctx->mutex / perf_event_context::mutex
  */
 static struct perf_event_context *
 perf_event_ctx_lock_nested(struct perf_event *event, int nesting)
 {
         struct perf_event_context *ctx;
 
 again:
         rcu_read_lock();
         ctx = READ_ONCE(event->ctx);
         if (!refcount_inc_not_zero(&ctx->refcount)) {
                 rcu_read_unlock();
                 goto again;
         }
         rcu_read_unlock();
 
         mutex_lock_nested(&ctx->mutex, nesting);
         if (event->ctx != ctx) {
                 mutex_unlock(&ctx->mutex);
                 put_ctx(ctx);
                 goto again;
         }
 
         return ctx;
 }
 
 static inline struct perf_event_context *
 perf_event_ctx_lock(struct perf_event *event)
 {
         return perf_event_ctx_lock_nested(event, 0);
 }
 
 static void perf_event_ctx_unlock(struct perf_event *event,
                                   struct perf_event_context *ctx)
 {
         mutex_unlock(&ctx->mutex);
         put_ctx(ctx);
 }
 
 /*
  * This must be done under the ctx->lock, such as to serialize against
  * context_equiv(), therefore we cannot call put_ctx() since that might end up
  * calling scheduler related locks and ctx->lock nests inside those.
  */
 static __must_check struct perf_event_context *
 unclone_ctx(struct perf_event_context *ctx)
 {
         struct perf_event_context *parent_ctx = ctx->parent_ctx;
 
         lockdep_assert_held(&ctx->lock);
 
         if (parent_ctx)
                 ctx->parent_ctx = NULL;
         ctx->generation++;
 
         return parent_ctx;
 }
 
 static u32 perf_event_pid_type(struct perf_event *event, struct task_struct *p,
                                 enum pid_type type)
 {
         u32 nr;
         /*
          * only top level events have the pid namespace they were created in
          */
         if (event->parent)
                 event = event->parent;
 
         nr = __task_pid_nr_ns(p, type, event->ns);
         /* avoid -1 if it is idle thread or runs in another ns */
         if (!nr && !pid_alive(p))
                 nr = -1;
         return nr;
 }
 
 static u32 perf_event_pid(struct perf_event *event, struct task_struct *p)
 {
         return perf_event_pid_type(event, p, PIDTYPE_TGID);
 }
 
 static u32 perf_event_tid(struct perf_event *event, struct task_struct *p)
 {
         return perf_event_pid_type(event, p, PIDTYPE_PID);
 }
 
 /*
  * If we inherit events we want to return the parent event id
  * to userspace.
  */
 static u64 primary_event_id(struct perf_event *event)
 {
         u64 id = event->id;
 
         if (event->parent)
                 id = event->parent->id;
 
         return id;
 }
 
 /*
  * Get the perf_event_context for a task and lock it.
  *
  * This has to cope with the fact that until it is locked,
  * the context could get moved to another task.
  */
 static struct perf_event_context *
 perf_lock_task_context(struct task_struct *task, int ctxn, unsigned long *flags)
 {
         struct perf_event_context *ctx;
 
 retry:
         /*
          * One of the few rules of preemptible RCU is that one cannot do
          * rcu_read_unlock() while holding a scheduler (or nested) lock when
          * part of the read side critical section was irqs-enabled -- see
          * rcu_read_unlock_special().
          *
          * Since ctx->lock nests under rq->lock we must ensure the entire read
          * side critical section has interrupts disabled.
          */
         local_irq_save(*flags);
         rcu_read_lock();
         ctx = rcu_dereference(task->perf_event_ctxp[ctxn]);
         if (ctx) {
                 /*
                  * If this context is a clone of another, it might
                  * get swapped for another underneath us by
                  * perf_event_task_sched_out, though the
                  * rcu_read_lock() protects us from any context
                  * getting freed.  Lock the context and check if it
                  * got swapped before we could get the lock, and retry
                  * if so.  If we locked the right context, then it
                  * can't get swapped on us any more.
                  */
                 raw_spin_lock(&ctx->lock);
                 if (ctx != rcu_dereference(task->perf_event_ctxp[ctxn])) {
                         raw_spin_unlock(&ctx->lock);
                         rcu_read_unlock();
                         local_irq_restore(*flags);
                         goto retry;
                 }
 
                 if (ctx->task == TASK_TOMBSTONE ||
                     !refcount_inc_not_zero(&ctx->refcount)) {
                         raw_spin_unlock(&ctx->lock);
                         ctx = NULL;
                 } else {
                         WARN_ON_ONCE(ctx->task != task);
                 }
         }
         rcu_read_unlock();
         if (!ctx)
                 local_irq_restore(*flags);
         return ctx;
 }
 
 /*
  * Get the context for a task and increment its pin_count so it
  * can't get swapped to another task.  This also increments its
  * reference count so that the context can't get freed.
  */
 static struct perf_event_context *
 perf_pin_task_context(struct task_struct *task, int ctxn)
 {
         struct perf_event_context *ctx;
         unsigned long flags;
 
         ctx = perf_lock_task_context(task, ctxn, &flags);
         if (ctx) {
                 ++ctx->pin_count;
                 raw_spin_unlock_irqrestore(&ctx->lock, flags);
         }
         return ctx;
 }
 
 static void perf_unpin_context(struct perf_event_context *ctx)
 {
         unsigned long flags;
 
         raw_spin_lock_irqsave(&ctx->lock, flags);
         --ctx->pin_count;
         raw_spin_unlock_irqrestore(&ctx->lock, flags);
 }
 
 /*
  * Update the record of the current time in a context.
  */
 static void __update_context_time(struct perf_event_context *ctx, bool adv)
 {
         u64 now = perf_clock();
 
         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->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)
 {
         struct perf_event_context *ctx = event->ctx;
         enum event_type_t event_type;
 
         lockdep_assert_held(&ctx->lock);
 
         /*
          * It's 'group type', really, because if our group leader is
          * pinned, so are we.
          */
         if (event->group_leader != event)
                 event = event->group_leader;
 
         event_type = event->attr.pinned ? EVENT_PINNED : EVENT_FLEXIBLE;
         if (!ctx->task)
                 event_type |= EVENT_CPU;
 
         return event_type;
 }
 
 /*
  * Helper function to initialize event group nodes.
  */
 static void init_event_group(struct perf_event *event)
 {
         RB_CLEAR_NODE(&event->group_node);
         event->group_index = 0;
 }
 
 /*
  * Extract pinned or flexible groups from the context
  * based on event attrs bits.
  */
 static struct perf_event_groups *
 get_event_groups(struct perf_event *event, struct perf_event_context *ctx)
 {
         if (event->attr.pinned)
                 return &ctx->pinned_groups;
         else
                 return &ctx->flexible_groups;
 }
 
 /*
  * Helper function to initializes perf_event_group trees.
  */
 static void perf_event_groups_init(struct perf_event_groups *groups)
 {
         groups->tree = RB_ROOT;
         groups->index = 0;
 }
 
 static inline struct cgroup *event_cgroup(const struct perf_event *event)
 {
         struct cgroup *cgroup = NULL;
 
 #ifdef CONFIG_CGROUP_PERF
         if (event->cgrp)
                 cgroup = event->cgrp->css.cgroup;
 #endif
 
         return cgroup;
 }
 
 /*
  * Compare function for event groups;
  *
  * Implements complex key that first sorts by CPU and then by virtual index
  * which provides ordering when rotating groups for the same CPU.
  */
 static __always_inline int
 perf_event_groups_cmp(const int left_cpu, const struct cgroup *left_cgroup,
                       const u64 left_group_index, const struct perf_event *right)
 {
         if (left_cpu < right->cpu)
                 return -1;
         if (left_cpu > right->cpu)
                 return 1;
 
 #ifdef CONFIG_CGROUP_PERF
         {
                 const struct cgroup *right_cgroup = event_cgroup(right);
 
                 if (left_cgroup != right_cgroup) {
                         if (!left_cgroup) {
                                 /*
                                  * Left has no cgroup but right does, no
                                  * cgroups come first.
                                  */
                                 return -1;
                         }
                         if (!right_cgroup) {
                                 /*
                                  * Right has no cgroup but left does, no
                                  * cgroups come first.
                                  */
                                 return 1;
                         }
                         /* Two dissimilar cgroups, order by id. */
                         if (cgroup_id(left_cgroup) < cgroup_id(right_cgroup))
                                 return -1;
 
                         return 1;
                 }
         }
 #endif
 
         if (left_group_index < right->group_index)
                 return -1;
         if (left_group_index > right->group_index)
                 return 1;
 
         return 0;
 }
 
 #define __node_2_pe(node) \
         rb_entry((node), struct perf_event, group_node)
 
 static inline bool __group_less(struct rb_node *a, const struct rb_node *b)
 {
         struct perf_event *e = __node_2_pe(a);
         return perf_event_groups_cmp(e->cpu, event_cgroup(e), e->group_index,
                                      __node_2_pe(b)) < 0;
 }
 
 struct __group_key {
         int cpu;
         struct cgroup *cgroup;
 };
 
 static inline int __group_cmp(const void *key, const struct rb_node *node)
 {
         const struct __group_key *a = key;
         const struct perf_event *b = __node_2_pe(node);
 
         /* partial/subtree match: @cpu, @cgroup; ignore: @group_index */
         return perf_event_groups_cmp(a->cpu, a->cgroup, b->group_index, b);
 }
 
 /*
  * Insert @event into @groups' tree; using {@event->cpu, ++@groups->index} for
  * key (see perf_event_groups_less). This places it last inside the CPU
  * subtree.
  */
 static void
 perf_event_groups_insert(struct perf_event_groups *groups,
                          struct perf_event *event)
 {
         event->group_index = ++groups->index;
 
         rb_add(&event->group_node, &groups->tree, __group_less);
 }
 
 /*
  * Helper function to insert event into the pinned or flexible groups.
  */
 static void
 add_event_to_groups(struct perf_event *event, struct perf_event_context *ctx)
 {
         struct perf_event_groups *groups;
 
         groups = get_event_groups(event, ctx);
         perf_event_groups_insert(groups, event);
 }
 
 /*
  * Delete a group from a tree.
  */
 static void
 perf_event_groups_delete(struct perf_event_groups *groups,
                          struct perf_event *event)
 {
         WARN_ON_ONCE(RB_EMPTY_NODE(&event->group_node) ||
                      RB_EMPTY_ROOT(&groups->tree));
 
         rb_erase(&event->group_node, &groups->tree);
         init_event_group(event);
 }
 
 /*
  * Helper function to delete event from its groups.
  */
 static void
 del_event_from_groups(struct perf_event *event, struct perf_event_context *ctx)
 {
         struct perf_event_groups *groups;
 
         groups = get_event_groups(event, ctx);
         perf_event_groups_delete(groups, event);
 }
 
 /*
  * Get the leftmost event in the cpu/cgroup subtree.
  */
 static struct perf_event *
 perf_event_groups_first(struct perf_event_groups *groups, int cpu,
                         struct cgroup *cgrp)
 {
         struct __group_key key = {
                 .cpu = cpu,
                 .cgroup = cgrp,
         };
         struct rb_node *node;
 
         node = rb_find_first(&key, &groups->tree, __group_cmp);
         if (node)
                 return __node_2_pe(node);
 
         return NULL;
 }
 
 /*
  * Like rb_entry_next_safe() for the @cpu subtree.
  */
 static struct perf_event *
 perf_event_groups_next(struct perf_event *event)
 {
         struct __group_key key = {
                 .cpu = event->cpu,
                 .cgroup = event_cgroup(event),
         };
         struct rb_node *next;
 
         next = rb_next_match(&key, &event->group_node, __group_cmp);
         if (next)
                 return __node_2_pe(next);
 
         return NULL;
 }
 
 /*
  * Iterate through the whole groups tree.
  */
 #define perf_event_groups_for_each(event, groups)                       \
         for (event = rb_entry_safe(rb_first(&((groups)->tree)),         \
                                 typeof(*event), group_node); event;     \
                 event = rb_entry_safe(rb_next(&event->group_node),      \
                                 typeof(*event), group_node))
 
 /*
  * Add an event from the lists for its context.
  * Must be called with ctx->mutex and ctx->lock held.
  */
 static void
 list_add_event(struct perf_event *event, struct perf_event_context *ctx)
 {
         lockdep_assert_held(&ctx->lock);
 
         WARN_ON_ONCE(event->attach_state & PERF_ATTACH_CONTEXT);
         event->attach_state |= PERF_ATTACH_CONTEXT;
 
         event->tstamp = perf_event_time(event);
 
         /*
          * If we're a stand alone event or group leader, we go to the context
          * list, group events are kept attached to the group so that
          * perf_group_detach can, at all times, locate all siblings.
          */
         if (event->group_leader == event) {
                 event->group_caps = event->event_caps;
                 add_event_to_groups(event, ctx);
         }
 
         list_add_rcu(&event->event_entry, &ctx->event_list);
         ctx->nr_events++;
         if (event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT)
                 ctx->nr_user++;
         if (event->attr.inherit_stat)
                 ctx->nr_stat++;
 
         if (event->state > PERF_EVENT_STATE_OFF)
                 perf_cgroup_event_enable(event, ctx);
 
         ctx->generation++;
 }
 
 /*
  * Initialize event state based on the perf_event_attr::disabled.
  */
 static inline void perf_event__state_init(struct perf_event *event)
 {
         event->state = event->attr.disabled ? PERF_EVENT_STATE_OFF :
                                               PERF_EVENT_STATE_INACTIVE;
 }
 
 static void __perf_event_read_size(struct perf_event *event, int nr_siblings)
 {
         int entry = sizeof(u64); /* value */
         int size = 0;
         int nr = 1;
 
         if (event->attr.read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
                 size += sizeof(u64);
 
         if (event->attr.read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
                 size += sizeof(u64);
 
         if (event->attr.read_format & PERF_FORMAT_ID)
                 entry += sizeof(u64);
 
         if (event->attr.read_format & PERF_FORMAT_GROUP) {
                 nr += nr_siblings;
                 size += sizeof(u64);
         }
 
         size += entry * nr;
         event->read_size = size;
 }
 
 static void __perf_event_header_size(struct perf_event *event, u64 sample_type)
 {
         struct perf_sample_data *data;
         u16 size = 0;
 
         if (sample_type & PERF_SAMPLE_IP)
                 size += sizeof(data->ip);
 
         if (sample_type & PERF_SAMPLE_ADDR)
                 size += sizeof(data->addr);
 
         if (sample_type & PERF_SAMPLE_PERIOD)
                 size += sizeof(data->period);
 
         if (sample_type & PERF_SAMPLE_WEIGHT_TYPE)
                 size += sizeof(data->weight.full);
 
         if (sample_type & PERF_SAMPLE_READ)
                 size += event->read_size;
 
         if (sample_type & PERF_SAMPLE_DATA_SRC)
                 size += sizeof(data->data_src.val);
 
         if (sample_type & PERF_SAMPLE_TRANSACTION)
                 size += sizeof(data->txn);
 
         if (sample_type & PERF_SAMPLE_PHYS_ADDR)
                 size += sizeof(data->phys_addr);
 
         if (sample_type & PERF_SAMPLE_CGROUP)
                 size += sizeof(data->cgroup);
 
         if (sample_type & PERF_SAMPLE_DATA_PAGE_SIZE)
                 size += sizeof(data->data_page_size);
 
         if (sample_type & PERF_SAMPLE_CODE_PAGE_SIZE)
                 size += sizeof(data->code_page_size);
 
         event->header_size = size;
 }
 
 /*
  * Called at perf_event creation and when events are attached/detached from a
  * group.
  */
 static void perf_event__header_size(struct perf_event *event)
 {
         __perf_event_read_size(event,
                                event->group_leader->nr_siblings);
         __perf_event_header_size(event, event->attr.sample_type);
 }
 
 static void perf_event__id_header_size(struct perf_event *event)
 {
         struct perf_sample_data *data;
         u64 sample_type = event->attr.sample_type;
         u16 size = 0;
 
         if (sample_type & PERF_SAMPLE_TID)
                 size += sizeof(data->tid_entry);
 
         if (sample_type & PERF_SAMPLE_TIME)
                 size += sizeof(data->time);
 
         if (sample_type & PERF_SAMPLE_IDENTIFIER)
                 size += sizeof(data->id);
 
         if (sample_type & PERF_SAMPLE_ID)
                 size += sizeof(data->id);
 
         if (sample_type & PERF_SAMPLE_STREAM_ID)
                 size += sizeof(data->stream_id);
 
         if (sample_type & PERF_SAMPLE_CPU)
                 size += sizeof(data->cpu_entry);
 
         event->id_header_size = size;
 }
 
 static bool perf_event_validate_size(struct perf_event *event)
 {
         /*
          * The values computed here will be over-written when we actually
          * attach the event.
          */
         __perf_event_read_size(event, event->group_leader->nr_siblings + 1);
         __perf_event_header_size(event, event->attr.sample_type & ~PERF_SAMPLE_READ);
         perf_event__id_header_size(event);
 
         /*
          * Sum the lot; should not exceed the 64k limit we have on records.
          * Conservative limit to allow for callchains and other variable fields.
          */
         if (event->read_size + event->header_size +
             event->id_header_size + sizeof(struct perf_event_header) >= 16*1024)
                 return false;
 
         return true;
 }
 
 static void perf_group_attach(struct perf_event *event)
 {
         struct perf_event *group_leader = event->group_leader, *pos;
 
         lockdep_assert_held(&event->ctx->lock);
 
         /*
          * We can have double attach due to group movement in perf_event_open.
          */
         if (event->attach_state & PERF_ATTACH_GROUP)
                 return;
 
         event->attach_state |= PERF_ATTACH_GROUP;
 
         if (group_leader == event)
                 return;
 
         WARN_ON_ONCE(group_leader->ctx != event->ctx);
 
         group_leader->group_caps &= event->event_caps;
 
         list_add_tail(&event->sibling_list, &group_leader->sibling_list);
         group_leader->nr_siblings++;
 
         perf_event__header_size(group_leader);
 
         for_each_sibling_event(pos, group_leader)
                 perf_event__header_size(pos);
 }
 
 /*
  * Remove an event from the lists for its context.
  * Must be called with ctx->mutex and ctx->lock held.
  */
 static void
 list_del_event(struct perf_event *event, struct perf_event_context *ctx)
 {
         WARN_ON_ONCE(event->ctx != ctx);
         lockdep_assert_held(&ctx->lock);
 
         /*
          * We can have double detach due to exit/hot-unplug + close.
          */
         if (!(event->attach_state & PERF_ATTACH_CONTEXT))
                 return;
 
         event->attach_state &= ~PERF_ATTACH_CONTEXT;
 
         ctx->nr_events--;
         if (event->hw.flags & PERF_EVENT_FLAG_USER_READ_CNT)
                 ctx->nr_user--;
         if (event->attr.inherit_stat)
                 ctx->nr_stat--;
 
         list_del_rcu(&event->event_entry);
 
         if (event->group_leader == event)
                 del_event_from_groups(event, ctx);
 
         /*
          * If event was in error state, then keep it
          * that way, otherwise bogus counts will be
          * returned on read(). The only way to get out
          * of error state is by explicit re-enabling
          * of the event
          */
         if (event->state > PERF_EVENT_STATE_OFF) {
                 perf_cgroup_event_disable(event, ctx);
                 perf_event_set_state(event, PERF_EVENT_STATE_OFF);
         }
 
         ctx->generation++;
 }
 
 static int
 perf_aux_output_match(struct perf_event *event, struct perf_event *aux_event)
 {
         if (!has_aux(aux_event))
                 return 0;
 
         if (!event->pmu->aux_output_match)
                 return 0;
 
         return event->pmu->aux_output_match(aux_event);
 }
 
 static void put_event(struct perf_event *event);
 static void event_sched_out(struct perf_event *event,
                             struct perf_cpu_context *cpuctx,
                             struct perf_event_context *ctx);
 
 static void perf_put_aux_event(struct perf_event *event)
 {
         struct perf_event_context *ctx = event->ctx;
         struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
         struct perf_event *iter;
 
         /*
          * If event uses aux_event tear down the link
          */
         if (event->aux_event) {
                 iter = event->aux_event;
                 event->aux_event = NULL;
                 put_event(iter);
                 return;
         }
 
         /*
          * If the event is an aux_event, tear down all links to
          * it from other events.
          */
         for_each_sibling_event(iter, event->group_leader) {
                 if (iter->aux_event != event)
                         continue;
 
                 iter->aux_event = NULL;
                 put_event(event);
 
                 /*
                  * If it's ACTIVE, schedule it out and put it into ERROR
                  * state so that we don't try to schedule it again. Note
                  * that perf_event_enable() will clear the ERROR status.
                  */
                 event_sched_out(iter, cpuctx, ctx);
                 perf_event_set_state(event, PERF_EVENT_STATE_ERROR);
         }
 }
 
 static bool perf_need_aux_event(struct perf_event *event)
 {
         return !!event->attr.aux_output || !!event->attr.aux_sample_size;
 }
 
 static int perf_get_aux_event(struct perf_event *event,
                               struct perf_event *group_leader)
 {
         /*
          * Our group leader must be an aux event if we want to be
          * an aux_output. This way, the aux event will precede its
          * aux_output events in the group, and therefore will always
          * schedule first.
          */
         if (!group_leader)
                 return 0;
 
         /*
          * aux_output and aux_sample_size are mutually exclusive.
          */
         if (event->attr.aux_output && event->attr.aux_sample_size)
                 return 0;
 
         if (event->attr.aux_output &&
             !perf_aux_output_match(event, group_leader))
                 return 0;
 
         if (event->attr.aux_sample_size && !group_leader->pmu->snapshot_aux)
                 return 0;
 
         if (!atomic_long_inc_not_zero(&group_leader->refcount))
                 return 0;
 
         /*
          * Link aux_outputs to their aux event; this is undone in
          * perf_group_detach() by perf_put_aux_event(). When the
          * group in torn down, the aux_output events loose their
          * link to the aux_event and can't schedule any more.
          */
         event->aux_event = group_leader;
 
         return 1;
 }
 
 static inline struct list_head *get_event_list(struct perf_event *event)
 {
         struct perf_event_context *ctx = event->ctx;
         return event->attr.pinned ? &ctx->pinned_active : &ctx->flexible_active;
 }
 
 /*
  * Events that have PERF_EV_CAP_SIBLING require being part of a group and
  * cannot exist on their own, schedule them out and move them into the ERROR
  * state. Also see _perf_event_enable(), it will not be able to recover
  * this ERROR state.
  */
 static inline void perf_remove_sibling_event(struct perf_event *event)
 {
         struct perf_event_context *ctx = event->ctx;
         struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
         event_sched_out(event, cpuctx, ctx);
         perf_event_set_state(event, PERF_EVENT_STATE_ERROR);
 }
 
 static void perf_group_detach(struct perf_event *event)
 {
         struct perf_event *leader = event->group_leader;
         struct perf_event *sibling, *tmp;
         struct perf_event_context *ctx = event->ctx;
 
         lockdep_assert_held(&ctx->lock);
 
         /*
          * We can have double detach due to exit/hot-unplug + close.
          */
         if (!(event->attach_state & PERF_ATTACH_GROUP))
                 return;
 
         event->attach_state &= ~PERF_ATTACH_GROUP;
 
         perf_put_aux_event(event);
 
         /*
          * If this is a sibling, remove it from its group.
          */
         if (leader != event) {
                 list_del_init(&event->sibling_list);
                 event->group_leader->nr_siblings--;
                 goto out;
         }
 
         /*
          * If this was a group event with sibling events then
          * upgrade the siblings to singleton events by adding them
          * to whatever list we are on.
          */
         list_for_each_entry_safe(sibling, tmp, &event->sibling_list, sibling_list) {
 
                 if (sibling->event_caps & PERF_EV_CAP_SIBLING)
                         perf_remove_sibling_event(sibling);
 
                 sibling->group_leader = sibling;
                 list_del_init(&sibling->sibling_list);
 
                 /* Inherit group flags from the previous leader */
                 sibling->group_caps = event->group_caps;
 
                 if (!RB_EMPTY_NODE(&event->group_node)) {
                         add_event_to_groups(sibling, event->ctx);
 
                         if (sibling->state == PERF_EVENT_STATE_ACTIVE)
                                 list_add_tail(&sibling->active_list, get_event_list(sibling));
                 }
 
                 WARN_ON_ONCE(sibling->ctx != event->ctx);
         }
 
 out:
         for_each_sibling_event(tmp, leader)
                 perf_event__header_size(tmp);
 
         perf_event__header_size(leader);
 }
 
 static void sync_child_event(struct perf_event *child_event);
 
 static void perf_child_detach(struct perf_event *event)
 {
         struct perf_event *parent_event = event->parent;
 
         if (!(event->attach_state & PERF_ATTACH_CHILD))
                 return;
 
         event->attach_state &= ~PERF_ATTACH_CHILD;
 
         if (WARN_ON_ONCE(!parent_event))
                 return;
 
         lockdep_assert_held(&parent_event->child_mutex);
 
         sync_child_event(event);
         list_del_init(&event->child_list);
 }
 
 static bool is_orphaned_event(struct perf_event *event)
 {
         return event->state == PERF_EVENT_STATE_DEAD;
 }
 
 static inline int __pmu_filter_match(struct perf_event *event)
 {
         struct pmu *pmu = event->pmu;
         return pmu->filter_match ? pmu->filter_match(event) : 1;
 }
 
 /*
  * Check whether we should attempt to schedule an event group based on
  * PMU-specific filtering. An event group can consist of HW and SW events,
  * potentially with a SW leader, so we must check all the filters, to
  * determine whether a group is schedulable:
  */
 static inline int pmu_filter_match(struct perf_event *event)
 {
         struct perf_event *sibling;
 
         if (!__pmu_filter_match(event))
                 return 0;
 
         for_each_sibling_event(sibling, event) {
                 if (!__pmu_filter_match(sibling))
                         return 0;
         }
 
         return 1;
 }
 
 static inline int
 event_filter_match(struct perf_event *event)
 {
         return (event->cpu == -1 || event->cpu == smp_processor_id()) &&
                perf_cgroup_match(event) && pmu_filter_match(event);
 }
 
 static void
 event_sched_out(struct perf_event *event,
                   struct perf_cpu_context *cpuctx,
                   struct perf_event_context *ctx)
 {
         enum perf_event_state state = PERF_EVENT_STATE_INACTIVE;
 
         WARN_ON_ONCE(event->ctx != ctx);
         lockdep_assert_held(&ctx->lock);
 
         if (event->state != PERF_EVENT_STATE_ACTIVE)
                 return;
 
         /*
          * Asymmetry; we only schedule events _IN_ through ctx_sched_in(), but
          * we can schedule events _OUT_ individually through things like
          * __perf_remove_from_context().
          */
         list_del_init(&event->active_list);
 
         perf_pmu_disable(event->pmu);
 
         event->pmu->del(event, 0);
         event->oncpu = -1;
 
         if (READ_ONCE(event->pending_disable) >= 0) {
                 WRITE_ONCE(event->pending_disable, -1);
                 perf_cgroup_event_disable(event, ctx);
                 state = PERF_EVENT_STATE_OFF;
         }
         perf_event_set_state(event, state);
 
         if (!is_software_event(event))
                 cpuctx->active_oncpu--;
         if (!--ctx->nr_active)
                 perf_event_ctx_deactivate(ctx);
         if (event->attr.freq && event->attr.sample_freq)
                 ctx->nr_freq--;
         if (event->attr.exclusive || !cpuctx->active_oncpu)
                 cpuctx->exclusive = 0;
 
         perf_pmu_enable(event->pmu);
 }
 
 static void
 group_sched_out(struct perf_event *group_event,
                 struct perf_cpu_context *cpuctx,
                 struct perf_event_context *ctx)
 {
         struct perf_event *event;
 
         if (group_event->state != PERF_EVENT_STATE_ACTIVE)
                 return;
 
         perf_pmu_disable(ctx->pmu);
 
         event_sched_out(group_event, cpuctx, ctx);
 
         /*
          * Schedule out siblings (if any):
          */
         for_each_sibling_event(event, group_event)
                 event_sched_out(event, cpuctx, ctx);
 
         perf_pmu_enable(ctx->pmu);
 }
 
 #define DETACH_GROUP    0x01UL
 #define DETACH_CHILD    0x02UL
 
 /*
  * Cross CPU call to remove a performance event
  *
  * We disable the event on the hardware level first. After that we
  * remove it from the context list.
  */
 static void
 __perf_remove_from_context(struct perf_event *event,
                            struct perf_cpu_context *cpuctx,
                            struct perf_event_context *ctx,
                            void *info)
 {
         unsigned long flags = (unsigned long)info;
 
         if (ctx->is_active & EVENT_TIME) {
                 update_context_time(ctx);
                 update_cgrp_time_from_cpuctx(cpuctx, false);
         }
 
         event_sched_out(event, cpuctx, ctx);
         if (flags & DETACH_GROUP)
                 perf_group_detach(event);
         if (flags & DETACH_CHILD)
                 perf_child_detach(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) {
                         WARN_ON_ONCE(cpuctx->task_ctx != ctx);
                         cpuctx->task_ctx = NULL;
                 }
         }
 }
 
 /*
  * Remove the event from a task's (or a CPU's) list of events.
  *
  * If event->ctx is a cloned context, callers must make sure that
  * every task struct that event->ctx->task could possibly point to
  * remains valid.  This is OK when called from perf_release since
  * that only calls us on the top-level context, which can't be a clone.
  * When called from perf_event_exit_task, it's OK because the
  * context has been detached from its task.
  */
 static void perf_remove_from_context(struct perf_event *event, unsigned long flags)
 {
         struct perf_event_context *ctx = event->ctx;
 
         lockdep_assert_held(&ctx->mutex);
 
         /*
          * Because of perf_event_exit_task(), perf_remove_from_context() ought
          * to work in the face of TASK_TOMBSTONE, unlike every other
          * event_function_call() user.
          */
         raw_spin_lock_irq(&ctx->lock);
         /*
          * Cgroup events are per-cpu events, and must IPI because of
          * cgrp_cpuctx_list.
          */
         if (!ctx->is_active && !is_cgroup_event(event)) {
                 __perf_remove_from_context(event, __get_cpu_context(ctx),
                                            ctx, (void *)flags);
                 raw_spin_unlock_irq(&ctx->lock);
                 return;
         }
         raw_spin_unlock_irq(&ctx->lock);
 
         event_function_call(event, __perf_remove_from_context, (void *)flags);
 }
 
 /*
  * Cross CPU call to disable a performance event
  */
 static void __perf_event_disable(struct perf_event *event,
                                  struct perf_cpu_context *cpuctx,
                                  struct perf_event_context *ctx,
                                  void *info)
 {
         if (event->state < PERF_EVENT_STATE_INACTIVE)
                 return;
 
         if (ctx->is_active & EVENT_TIME) {
                 update_context_time(ctx);
                 update_cgrp_time_from_event(event);
         }
 
         if (event == event->group_leader)
                 group_sched_out(event, cpuctx, ctx);
         else
                 event_sched_out(event, cpuctx, ctx);
 
         perf_event_set_state(event, PERF_EVENT_STATE_OFF);
         perf_cgroup_event_disable(event, ctx);
 }
 
 /*
  * Disable an event.
  *
  * If event->ctx is a cloned context, callers must make sure that
  * every task struct that event->ctx->task could possibly point to
  * remains valid.  This condition is satisfied when called through
  * perf_event_for_each_child or perf_event_for_each because they
  * hold the top-level event's child_mutex, so any descendant that
  * goes to exit will block in perf_event_exit_event().
  *
  * When called from perf_pending_event it's OK because event->ctx
  * is the current context on this CPU and preemption is disabled,
  * hence we can't get into perf_event_task_sched_out for this context.
  */
 static void _perf_event_disable(struct perf_event *event)
 {
         struct perf_event_context *ctx = event->ctx;
 
         raw_spin_lock_irq(&ctx->lock);
         if (event->state <= PERF_EVENT_STATE_OFF) {
                 raw_spin_unlock_irq(&ctx->lock);
                 return;
         }
         raw_spin_unlock_irq(&ctx->lock);
 
         event_function_call(event, __perf_event_disable, NULL);
 }
 
 void perf_event_disable_local(struct perf_event *event)
 {
         event_function_local(event, __perf_event_disable, NULL);
 }
 
 /*
  * Strictly speaking kernel users cannot create groups and therefore this
  * interface does not need the perf_event_ctx_lock() magic.
  */
 void perf_event_disable(struct perf_event *event)
 {
         struct perf_event_context *ctx;
 
         ctx = perf_event_ctx_lock(event);
         _perf_event_disable(event);
         perf_event_ctx_unlock(event, ctx);
 }
 EXPORT_SYMBOL_GPL(perf_event_disable);
 
 void perf_event_disable_inatomic(struct perf_event *event)
 {
         WRITE_ONCE(event->pending_disable, smp_processor_id());
         /* can fail, see perf_pending_event_disable() */
         irq_work_queue(&event->pending);
 }
 
 #define MAX_INTERRUPTS (~0ULL)
 
 static void perf_log_throttle(struct perf_event *event, int enable);
 static void perf_log_itrace_start(struct perf_event *event);
 
 static int
 event_sched_in(struct perf_event *event,
                  struct perf_cpu_context *cpuctx,
                  struct perf_event_context *ctx)
 {
         int ret = 0;
 
         WARN_ON_ONCE(event->ctx != ctx);
 
         lockdep_assert_held(&ctx->lock);
 
         if (event->state <= PERF_EVENT_STATE_OFF)
                 return 0;
 
         WRITE_ONCE(event->oncpu, smp_processor_id());
         /*
          * Order event::oncpu write to happen before the ACTIVE state is
          * visible. This allows perf_event_{stop,read}() to observe the correct
          * ->oncpu if it sees ACTIVE.
          */
         smp_wmb();
         perf_event_set_state(event, PERF_EVENT_STATE_ACTIVE);
 
         /*
          * Unthrottle events, since we scheduled we might have missed several
          * ticks already, also for a heavily scheduling task there is little
          * guarantee it'll get a tick in a timely manner.
          */
         if (unlikely(event->hw.interrupts == MAX_INTERRUPTS)) {
                 perf_log_throttle(event, 1);
                 event->hw.interrupts = 0;
         }
 
         perf_pmu_disable(event->pmu);
 
         perf_log_itrace_start(event);
 
         if (event->pmu->add(event, PERF_EF_START)) {
                 perf_event_set_state(event, PERF_EVENT_STATE_INACTIVE);
                 event->oncpu = -1;
                 ret = -EAGAIN;
                 goto out;
         }
 
         if (!is_software_event(event))
                 cpuctx->active_oncpu++;
         if (!ctx->nr_active++)
                 perf_event_ctx_activate(ctx);
         if (event->attr.freq && event->attr.sample_freq)
                 ctx->nr_freq++;
 
         if (event->attr.exclusive)
                 cpuctx->exclusive = 1;
 
 out:
         perf_pmu_enable(event->pmu);
 
         return ret;
 }
 
 static int
 group_sched_in(struct perf_event *group_event,
                struct perf_cpu_context *cpuctx,
                struct perf_event_context *ctx)
 {
         struct perf_event *event, *partial_group = NULL;
         struct pmu *pmu = ctx->pmu;
 
         if (group_event->state == PERF_EVENT_STATE_OFF)
                 return 0;
 
         pmu->start_txn(pmu, PERF_PMU_TXN_ADD);
 
         if (event_sched_in(group_event, cpuctx, ctx))
                 goto error;
 
         /*
          * Schedule in siblings as one group (if any):
          */
         for_each_sibling_event(event, group_event) {
                 if (event_sched_in(event, cpuctx, ctx)) {
                         partial_group = event;
                         goto group_error;
                 }
         }
 
         if (!pmu->commit_txn(pmu))
                 return 0;
 
 group_error:
         /*
          * Groups can be scheduled in as one unit only, so undo any
          * partial group before returning:
          * The events up to the failed event are scheduled out normally.
          */
         for_each_sibling_event(event, group_event) {
                 if (event == partial_group)
                         break;
 
                 event_sched_out(event, cpuctx, ctx);
         }
         event_sched_out(group_event, cpuctx, ctx);
 
 error:
         pmu->cancel_txn(pmu);
         return -EAGAIN;
 }
 
 /*
  * Work out whether we can put this event group on the CPU now.
  */
 static int group_can_go_on(struct perf_event *event,
                            struct perf_cpu_context *cpuctx,
                            int can_add_hw)
 {
         /*
          * Groups consisting entirely of software events can always go on.
          */
         if (event->group_caps & PERF_EV_CAP_SOFTWARE)
                 return 1;
         /*
          * If an exclusive group is already on, no other hardware
          * events can go on.
          */
         if (cpuctx->exclusive)
                 return 0;
         /*
          * If this group is exclusive and there are already
          * events on the CPU, it can't go on.
          */
         if (event->attr.exclusive && !list_empty(get_event_list(event)))
                 return 0;
         /*
          * Otherwise, try to add it if all previous groups were able
          * to go on.
          */
         return can_add_hw;
 }
 
 static void add_event_to_ctx(struct perf_event *event,
                                struct perf_event_context *ctx)
 {
         list_add_event(event, ctx);
         perf_group_attach(event);
 }
 
 static void ctx_sched_out(struct perf_event_context *ctx,
                           struct perf_cpu_context *cpuctx,
                           enum event_type_t event_type);
 static void
 ctx_sched_in(struct perf_event_context *ctx,
              struct perf_cpu_context *cpuctx,
              enum event_type_t event_type);
 
 static void task_ctx_sched_out(struct perf_cpu_context *cpuctx,
                                struct perf_event_context *ctx,
                                enum event_type_t event_type)
 {
         if (!cpuctx->task_ctx)
                 return;
 
         if (WARN_ON_ONCE(ctx != cpuctx->task_ctx))
                 return;
 
         ctx_sched_out(ctx, cpuctx, event_type);
 }
 
 static void perf_event_sched_in(struct perf_cpu_context *cpuctx,
                                 struct perf_event_context *ctx)
 {
         cpu_ctx_sched_in(cpuctx, EVENT_PINNED);
         if (ctx)
                 ctx_sched_in(ctx, cpuctx, EVENT_PINNED);
         cpu_ctx_sched_in(cpuctx, EVENT_FLEXIBLE);
         if (ctx)
                 ctx_sched_in(ctx, cpuctx, EVENT_FLEXIBLE);
 }
 
 /*
  * We want to maintain the following priority of scheduling:
  *  - CPU pinned (EVENT_CPU | EVENT_PINNED)
  *  - task pinned (EVENT_PINNED)
  *  - CPU flexible (EVENT_CPU | EVENT_FLEXIBLE)
  *  - task flexible (EVENT_FLEXIBLE).
  *
  * In order to avoid unscheduling and scheduling back in everything every
  * time an event is added, only do it for the groups of equal priority and
  * below.
  *
  * This can be called after a batch operation on task events, in which case
  * event_type is a bit mask of the types of events involved. For CPU events,
  * event_type is only either EVENT_PINNED or EVENT_FLEXIBLE.
  */
 static void ctx_resched(struct perf_cpu_context *cpuctx,
                         struct perf_event_context *task_ctx,
                         enum event_type_t event_type)
 {
         enum event_type_t ctx_event_type;
         bool cpu_event = !!(event_type & EVENT_CPU);
 
         /*
          * If pinned groups are involved, flexible groups also need to be
          * scheduled out.
          */
         if (event_type & EVENT_PINNED)
                 event_type |= EVENT_FLEXIBLE;
 
         ctx_event_type = event_type & EVENT_ALL;
 
         perf_pmu_disable(cpuctx->ctx.pmu);
         if (task_ctx)
                 task_ctx_sched_out(cpuctx, task_ctx, event_type);
 
         /*
          * Decide which cpu ctx groups to schedule out based on the types
          * of events that caused rescheduling:
          *  - EVENT_CPU: schedule out corresponding groups;
          *  - EVENT_PINNED task events: schedule out EVENT_FLEXIBLE groups;
          *  - otherwise, do nothing more.
          */
         if (cpu_event)
                 cpu_ctx_sched_out(cpuctx, ctx_event_type);
         else if (ctx_event_type & EVENT_PINNED)
                 cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE);
 
         perf_event_sched_in(cpuctx, task_ctx);
         perf_pmu_enable(cpuctx->ctx.pmu);
 }
 
 void perf_pmu_resched(struct pmu *pmu)
 {
         struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
         struct perf_event_context *task_ctx = cpuctx->task_ctx;
 
         perf_ctx_lock(cpuctx, task_ctx);
         ctx_resched(cpuctx, task_ctx, EVENT_ALL|EVENT_CPU);
         perf_ctx_unlock(cpuctx, task_ctx);
 }
 
 /*
  * Cross CPU call to install and enable a performance event
  *
  * Very similar to remote_function() + event_function() but cannot assume that
  * things like ctx->is_active and cpuctx->task_ctx are set.
  */
 static int  __perf_install_in_context(void *info)
 {
         struct perf_event *event = info;
         struct perf_event_context *ctx = event->ctx;
         struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
         struct perf_event_context *task_ctx = cpuctx->task_ctx;
         bool reprogram = true;
         int ret = 0;
 
         raw_spin_lock(&cpuctx->ctx.lock);
         if (ctx->task) {
                 raw_spin_lock(&ctx->lock);
                 task_ctx = ctx;
 
                 reprogram = (ctx->task == current);
 
                 /*
                  * If the task is running, it must be running on this CPU,
                  * otherwise we cannot reprogram things.
                  *
                  * If its not running, we don't care, ctx->lock will
                  * serialize against it becoming runnable.
                  */
                 if (task_curr(ctx->task) && !reprogram) {
                         ret = -ESRCH;
                         goto unlock;
                 }
 
                 WARN_ON_ONCE(reprogram && cpuctx->task_ctx && cpuctx->task_ctx != ctx);
         } else if (task_ctx) {
                 raw_spin_lock(&task_ctx->lock);
         }
 
 #ifdef CONFIG_CGROUP_PERF
         if (event->state > PERF_EVENT_STATE_OFF && is_cgroup_event(event)) {
                 /*
                  * If the current cgroup doesn't match the event's
                  * cgroup, we should not try to schedule it.
                  */
                 struct perf_cgroup *cgrp = perf_cgroup_from_task(current, ctx);
                 reprogram = cgroup_is_descendant(cgrp->css.cgroup,
                                         event->cgrp->css.cgroup);
         }
 #endif
 
         if (reprogram) {
                 ctx_sched_out(ctx, cpuctx, EVENT_TIME);
                 add_event_to_ctx(event, ctx);
                 ctx_resched(cpuctx, task_ctx, get_event_type(event));
         } else {
                 add_event_to_ctx(event, ctx);
         }
 
 unlock:
         perf_ctx_unlock(cpuctx, task_ctx);
 
         return ret;
 }
 
 static bool exclusive_event_installable(struct perf_event *event,
                                         struct perf_event_context *ctx);
 
 /*
  * Attach a performance event to a context.
  *
  * Very similar to event_function_call, see comment there.
  */
 static void
 perf_install_in_context(struct perf_event_context *ctx,
                         struct perf_event *event,
                         int cpu)
 {
         struct task_struct *task = READ_ONCE(ctx->task);
 
         lockdep_assert_held(&ctx->mutex);
 
         WARN_ON_ONCE(!exclusive_event_installable(event, ctx));
 
         if (event->cpu != -1)
                 event->cpu = cpu;
 
         /*
          * Ensures that if we can observe event->ctx, both the event and ctx
          * will be 'complete'. See perf_iterate_sb_cpu().
          */
         smp_store_release(&event->ctx, ctx);
 
         /*
          * perf_event_attr::disabled events will not run and can be initialized
          * without IPI. Except when this is the first event for the context, in
          * that case we need the magic of the IPI to set ctx->is_active.
          * Similarly, cgroup events for the context also needs the IPI to
          * manipulate the cgrp_cpuctx_list.
          *
          * The IOC_ENABLE that is sure to follow the creation of a disabled
          * event will issue the IPI and reprogram the hardware.
          */
         if (__perf_effective_state(event) == PERF_EVENT_STATE_OFF &&
             ctx->nr_events && !is_cgroup_event(event)) {
                 raw_spin_lock_irq(&ctx->lock);
                 if (ctx->task == TASK_TOMBSTONE) {
                         raw_spin_unlock_irq(&ctx->lock);
                         return;
                 }
                 add_event_to_ctx(event, ctx);
                 raw_spin_unlock_irq(&ctx->lock);
                 return;
         }
 
         if (!task) {
                 cpu_function_call(cpu, __perf_install_in_context, event);
                 return;
         }
 
         /*
          * Should not happen, we validate the ctx is still alive before calling.
          */
         if (WARN_ON_ONCE(task == TASK_TOMBSTONE))
                 return;
 
         /*
          * Installing events is tricky because we cannot rely on ctx->is_active
          * to be set in case this is the nr_events 0 -> 1 transition.
          *
          * Instead we use task_curr(), which tells us if the task is running.
          * However, since we use task_curr() outside of rq::lock, we can race
          * against the actual state. This means the result can be wrong.
          *
          * If we get a false positive, we retry, this is harmless.
          *
          * If we get a false negative, things are complicated. If we are after
          * perf_event_context_sched_in() ctx::lock will serialize us, and the
          * value must be correct. If we're before, it doesn't matter since
          * perf_event_context_sched_in() will program the counter.
          *
          * However, this hinges on the remote context switch having observed
          * our task->perf_event_ctxp[] store, such that it will in fact take
          * ctx::lock in perf_event_context_sched_in().
          *
          * We do this by task_function_call(), if the IPI fails to hit the task
          * we know any future context switch of task must see the
          * perf_event_ctpx[] store.
          */
 
         /*
          * This smp_mb() orders the task->perf_event_ctxp[] store with the
          * task_cpu() load, such that if the IPI then does not find the task
          * running, a future context switch of that task must observe the
          * store.
          */
         smp_mb();
 again:
         if (!task_function_call(task, __perf_install_in_context, event))
                 return;
 
         raw_spin_lock_irq(&ctx->lock);
         task = ctx->task;
         if (WARN_ON_ONCE(task == TASK_TOMBSTONE)) {
                 /*
                  * Cannot happen because we already checked above (which also
                  * cannot happen), and we hold ctx->mutex, which serializes us
                  * against perf_event_exit_task_context().
                  */
                 raw_spin_unlock_irq(&ctx->lock);
                 return;
         }
         /*
          * If the task is not running, ctx->lock will avoid it becoming so,
          * thus we can safely install the event.
          */
         if (task_curr(task)) {
                 raw_spin_unlock_irq(&ctx->lock);
                 goto again;
         }
         add_event_to_ctx(event, ctx);
         raw_spin_unlock_irq(&ctx->lock);
 }
 
 /*
  * Cross CPU call to enable a performance event
  */
 static void __perf_event_enable(struct perf_event *event,
                                 struct perf_cpu_context *cpuctx,
                                 struct perf_event_context *ctx,
                                 void *info)
 {
         struct perf_event *leader = event->group_leader;
         struct perf_event_context *task_ctx;
 
         if (event->state >= PERF_EVENT_STATE_INACTIVE ||
             event->state <= PERF_EVENT_STATE_ERROR)
                 return;
 
         if (ctx->is_active)
                 ctx_sched_out(ctx, cpuctx, EVENT_TIME);
 
         perf_event_set_state(event, PERF_EVENT_STATE_INACTIVE);
         perf_cgroup_event_enable(event, ctx);
 
         if (!ctx->is_active)
                 return;
 
         if (!event_filter_match(event)) {
                 ctx_sched_in(ctx, cpuctx, EVENT_TIME);
                 return;
         }
 
         /*
          * If the event is in a group and isn't the group leader,
          * then don't put it on unless the group is on.
          */
         if (leader != event && leader->state != PERF_EVENT_STATE_ACTIVE) {
                 ctx_sched_in(ctx, cpuctx, EVENT_TIME);
                 return;
         }
 
         task_ctx = cpuctx->task_ctx;
         if (ctx->task)
                 WARN_ON_ONCE(task_ctx != ctx);
 
         ctx_resched(cpuctx, task_ctx, get_event_type(event));
 }
 
 /*
  * Enable an event.
  *
  * If event->ctx is a cloned context, callers must make sure that
  * every task struct that event->ctx->task could possibly point to
  * remains valid.  This condition is satisfied when called through
  * perf_event_for_each_child or perf_event_for_each as described
  * for perf_event_disable.
  */
 static void _perf_event_enable(struct perf_event *event)
 {
         struct perf_event_context *ctx = event->ctx;
 
         raw_spin_lock_irq(&ctx->lock);
         if (event->state >= PERF_EVENT_STATE_INACTIVE ||
             event->state <  PERF_EVENT_STATE_ERROR) {
 out:
                 raw_spin_unlock_irq(&ctx->lock);
                 return;
         }
 
         /*
          * If the event is in error state, clear that first.
          *
          * That way, if we see the event in error state below, we know that it
          * has gone back into error state, as distinct from the task having
          * been scheduled away before the cross-call arrived.
          */
         if (event->state == PERF_EVENT_STATE_ERROR) {
                 /*
                  * Detached SIBLING events cannot leave ERROR state.
                  */
                 if (event->event_caps & PERF_EV_CAP_SIBLING &&
                     event->group_leader == event)
                         goto out;
 
                 event->state = PERF_EVENT_STATE_OFF;
         }
         raw_spin_unlock_irq(&ctx->lock);
 
         event_function_call(event, __perf_event_enable, NULL);
 }
 
 /*
  * See perf_event_disable();
  */
 void perf_event_enable(struct perf_event *event)
 {
         struct perf_event_context *ctx;
 
         ctx = perf_event_ctx_lock(event);
         _perf_event_enable(event);
         perf_event_ctx_unlock(event, ctx);
 }
 EXPORT_SYMBOL_GPL(perf_event_enable);
 
 struct stop_event_data {
         struct perf_event       *event;
         unsigned int            restart;
 };
 
 static int __perf_event_stop(void *info)
 {
         struct stop_event_data *sd = info;
         struct perf_event *event = sd->event;
 
         /* if it's already INACTIVE, do nothing */
         if (READ_ONCE(event->state) != PERF_EVENT_STATE_ACTIVE)
                 return 0;
 
         /* matches smp_wmb() in event_sched_in() */
         smp_rmb();
 
         /*
          * There is a window with interrupts enabled before we get here,
          * so we need to check again lest we try to stop another CPU's event.
          */
         if (READ_ONCE(event->oncpu) != smp_processor_id())
                 return -EAGAIN;
 
         event->pmu->stop(event, PERF_EF_UPDATE);
 
         /*
          * May race with the actual stop (through perf_pmu_output_stop()),
          * but it is only used for events with AUX ring buffer, and such
          * events will refuse to restart because of rb::aux_mmap_count==0,
          * see comments in perf_aux_output_begin().
          *
          * Since this is happening on an event-local CPU, no trace is lost
          * while restarting.
          */
         if (sd->restart)
                 event->pmu->start(event, 0);
 
         return 0;
 }
 
 static int perf_event_stop(struct perf_event *event, int restart)
 {
         struct stop_event_data sd = {
                 .event          = event,
                 .restart        = restart,
         };
         int ret = 0;
 
         do {
                 if (READ_ONCE(event->state) != PERF_EVENT_STATE_ACTIVE)
                         return 0;
 
                 /* matches smp_wmb() in event_sched_in() */
                 smp_rmb();
 
                 /*
                  * We only want to restart ACTIVE events, so if the event goes
                  * inactive here (event->oncpu==-1), there's nothing more to do;
                  * fall through with ret==-ENXIO.
                  */
                 ret = cpu_function_call(READ_ONCE(event->oncpu),
                                         __perf_event_stop, &sd);
         } while (ret == -EAGAIN);
 
         return ret;
 }
 
 /*
  * In order to contain the amount of racy and tricky in the address filter
  * configuration management, it is a two part process:
  *
  * (p1) when userspace mappings change as a result of (1) or (2) or (3) below,
  *      we update the addresses of corresponding vmas in
  *      event::addr_filter_ranges array and bump the event::addr_filters_gen;
  * (p2) when an event is scheduled in (pmu::add), it calls
  *      perf_event_addr_filters_sync() which calls pmu::addr_filters_sync()
  *      if the generation has changed since the previous call.
  *
  * If (p1) happens while the event is active, we restart it to force (p2).
  *
  * (1) perf_addr_filters_apply(): adjusting filters' offsets based on
  *     pre-existing mappings, called once when new filters arrive via SET_FILTER
  *     ioctl;
  * (2) perf_addr_filters_adjust(): adjusting filters' offsets based on newly
  *     registered mapping, called for every new mmap(), with mm::mmap_lock down
  *     for reading;
  * (3) perf_event_addr_filters_exec(): clearing filters' offsets in the process
  *     of exec.
  */
 void perf_event_addr_filters_sync(struct perf_event *event)
 {
         struct perf_addr_filters_head *ifh = perf_event_addr_filters(event);
 
         if (!has_addr_filter(event))
                 return;
 
         raw_spin_lock(&ifh->lock);
         if (event->addr_filters_gen != event->hw.addr_filters_gen) {
                 event->pmu->addr_filters_sync(event);
                 event->hw.addr_filters_gen = event->addr_filters_gen;
         }
         raw_spin_unlock(&ifh->lock);
 }
 EXPORT_SYMBOL_GPL(perf_event_addr_filters_sync);
 
 static int _perf_event_refresh(struct perf_event *event, int refresh)
 {
         /*
          * not supported on inherited events
          */
         if (event->attr.inherit || !is_sampling_event(event))
                 return -EINVAL;
 
         atomic_add(refresh, &event->event_limit);
         _perf_event_enable(event);
 
         return 0;
 }
 
 /*
  * See perf_event_disable()
  */
 int perf_event_refresh(struct perf_event *event, int refresh)
 {
         struct perf_event_context *ctx;
         int ret;
 
         ctx = perf_event_ctx_lock(event);
         ret = _perf_event_refresh(event, refresh);
         perf_event_ctx_unlock(event, ctx);
 
         return ret;
 }
 EXPORT_SYMBOL_GPL(perf_event_refresh);
 
 static int perf_event_modify_breakpoint(struct perf_event *bp,
                                          struct perf_event_attr *attr)
 {
         int err;
 
         _perf_event_disable(bp);
 
         err = modify_user_hw_breakpoint_check(bp, attr, true);
 
         if (!bp->attr.disabled)
                 _perf_event_enable(bp);
 
         return err;
 }
 
 /*
  * Copy event-type-independent attributes that may be modified.
  */
 static void perf_event_modify_copy_attr(struct perf_event_attr *to,
                                         const struct perf_event_attr *from)
 {
         to->sig_data = from->sig_data;
 }
 
 static int perf_event_modify_attr(struct perf_event *event,
                                   struct perf_event_attr *attr)
 {
         int (*func)(struct perf_event *, struct perf_event_attr *);
         struct perf_event *child;
         int err;
 
         if (event->attr.type != attr->type)
                 return -EINVAL;
 
         switch (event->attr.type) {
         case PERF_TYPE_BREAKPOINT:
                 func = perf_event_modify_breakpoint;
                 break;
         default:
                 /* Place holder for future additions. */
                 return -EOPNOTSUPP;
         }
 
         WARN_ON_ONCE(event->ctx->parent_ctx);
 
         mutex_lock(&event->child_mutex);
         /*
          * Event-type-independent attributes must be copied before event-type
          * modification, which will validate that final attributes match the
          * source attributes after all relevant attributes have been copied.
          */
         perf_event_modify_copy_attr(&event->attr, attr);
         err = func(event, attr);
         if (err)
                 goto out;
         list_for_each_entry(child, &event->child_list, child_list) {
                 perf_event_modify_copy_attr(&child->attr, attr);
                 err = func(child, attr);
                 if (err)
                         goto out;
         }
 out:
         mutex_unlock(&event->child_mutex);
         return err;
 }
 
 static void ctx_sched_out(struct perf_event_context *ctx,
                           struct perf_cpu_context *cpuctx,
                           enum event_type_t event_type)
 {
         struct perf_event *event, *tmp;
         int is_active = ctx->is_active;
 
         lockdep_assert_held(&ctx->lock);
 
         if (likely(!ctx->nr_events)) {
                 /*
                  * See __perf_remove_from_context().
                  */
                 WARN_ON_ONCE(ctx->is_active);
                 if (ctx->task)
                         WARN_ON_ONCE(cpuctx->task_ctx);
                 return;
         }
 
         /*
          * Always update time if it was set; not only when it changes.
          * Otherwise we can 'forget' to update time for any but the last
          * context we sched out. For example:
          *
          *   ctx_sched_out(.event_type = EVENT_FLEXIBLE)
          *   ctx_sched_out(.event_type = EVENT_PINNED)
          *
          * would only update time for the pinned events.
          */
         if (is_active & EVENT_TIME) {
                 /* update (and stop) ctx time */
                 update_context_time(ctx);
                 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 */
 
         if (!ctx->nr_active || !(is_active & EVENT_ALL))
                 return;
 
         perf_pmu_disable(ctx->pmu);
         if (is_active & EVENT_PINNED) {
                 list_for_each_entry_safe(event, tmp, &ctx->pinned_active, active_list)
                         group_sched_out(event, cpuctx, ctx);
         }
 
         if (is_active & EVENT_FLEXIBLE) {
                 list_for_each_entry_safe(event, tmp, &ctx->flexible_active, active_list)
                         group_sched_out(event, cpuctx, ctx);
 
                 /*
                  * Since we cleared EVENT_FLEXIBLE, also clear
                  * rotate_necessary, is will be reset by
                  * ctx_flexible_sched_in() when needed.
                  */
                 ctx->rotate_necessary = 0;
         }
         perf_pmu_enable(ctx->pmu);
 }
 
 /*
  * Test whether two contexts are equivalent, i.e. whether they have both been
  * cloned from the same version of the same context.
  *
  * Equivalence is measured using a generation number in the context that is
  * incremented on each modification to it; see unclone_ctx(), list_add_event()
  * and list_del_event().
  */
 static int context_equiv(struct perf_event_context *ctx1,
                          struct perf_event_context *ctx2)
 {
         lockdep_assert_held(&ctx1->lock);
         lockdep_assert_held(&ctx2->lock);
 
         /* Pinning disables the swap optimization */
         if (ctx1->pin_count || ctx2->pin_count)
                 return 0;
 
         /* If ctx1 is the parent of ctx2 */
         if (ctx1 == ctx2->parent_ctx && ctx1->generation == ctx2->parent_gen)
                 return 1;
 
         /* If ctx2 is the parent of ctx1 */
         if (ctx1->parent_ctx == ctx2 && ctx1->parent_gen == ctx2->generation)
                 return 1;
 
         /*
          * If ctx1 and ctx2 have the same parent; we flatten the parent
          * hierarchy, see perf_event_init_context().
          */
         if (ctx1->parent_ctx && ctx1->parent_ctx == ctx2->parent_ctx &&
                         ctx1->parent_gen == ctx2->parent_gen)
                 return 1;
 
         /* Unmatched */
         return 0;
 }
 
 static void __perf_event_sync_stat(struct perf_event *event,
                                      struct perf_event *next_event)
 {
         u64 value;
 
         if (!event->attr.inherit_stat)
                 return;
 
         /*
          * Update the event value, we cannot use perf_event_read()
          * because we're in the middle of a context switch and have IRQs
          * disabled, which upsets smp_call_function_single(), however
          * we know the event must be on the current CPU, therefore we
          * don't need to use it.
          */
         if (event->state == PERF_EVENT_STATE_ACTIVE)
                 event->pmu->read(event);
 
         perf_event_update_time(event);
 
         /*
          * In order to keep per-task stats reliable we need to flip the event
          * values when we flip the contexts.
          */
         value = local64_read(&next_event->count);
         value = local64_xchg(&event->count, value);
         local64_set(&next_event->count, value);
 
         swap(event->total_time_enabled, next_event->total_time_enabled);
         swap(event->total_time_running, next_event->total_time_running);
 
         /*
          * Since we swizzled the values, update the user visible data too.
          */
         perf_event_update_userpage(event);
         perf_event_update_userpage(next_event);
 }
 
 static void perf_event_sync_stat(struct perf_event_context *ctx,
                                    struct perf_event_context *next_ctx)
 {
         struct perf_event *event, *next_event;
 
         if (!ctx->nr_stat)
                 return;
 
         update_context_time(ctx);
 
         event = list_first_entry(&ctx->event_list,
                                    struct perf_event, event_entry);
 
         next_event = list_first_entry(&next_ctx->event_list,
                                         struct perf_event, event_entry);
 
         while (&event->event_entry != &ctx->event_list &&
                &next_event->event_entry != &next_ctx->event_list) {
 
                 __perf_event_sync_stat(event, next_event);
 
                 event = list_next_entry(event, event_entry);
                 next_event = list_next_entry(next_event, event_entry);
         }
 }
 
 static void perf_event_context_sched_out(struct task_struct *task, int ctxn,
                                          struct task_struct *next)
 {
         struct perf_event_context *ctx = task->perf_event_ctxp[ctxn];
         struct perf_event_context *next_ctx;
         struct perf_event_context *parent, *next_parent;
         struct perf_cpu_context *cpuctx;
         int do_switch = 1;
         struct pmu *pmu;
 
         if (likely(!ctx))
                 return;
 
         pmu = ctx->pmu;
         cpuctx = __get_cpu_context(ctx);
         if (!cpuctx->task_ctx)
                 return;
 
         rcu_read_lock();
         next_ctx = next->perf_event_ctxp[ctxn];
         if (!next_ctx)
                 goto unlock;
 
         parent = rcu_dereference(ctx->parent_ctx);
         next_parent = rcu_dereference(next_ctx->parent_ctx);
 
         /* If neither context have a parent context; they cannot be clones. */
         if (!parent && !next_parent)
                 goto unlock;
 
         if (next_parent == ctx || next_ctx == parent || next_parent == parent) {
                 /*
                  * Looks like the two contexts are clones, so we might be
                  * able to optimize the context switch.  We lock both
                  * contexts and check that they are clones under the
                  * lock (including re-checking that neither has been
                  * uncloned in the meantime).  It doesn't matter which
                  * order we take the locks because no other cpu could
                  * be trying to lock both of these tasks.
                  */
                 raw_spin_lock(&ctx->lock);
                 raw_spin_lock_nested(&next_ctx->lock, SINGLE_DEPTH_NESTING);
                 if (context_equiv(ctx, next_ctx)) {
 
                         WRITE_ONCE(ctx->task, next);
                         WRITE_ONCE(next_ctx->task, task);
 
                         perf_pmu_disable(pmu);
 
                         if (cpuctx->sched_cb_usage && pmu->sched_task)
                                 pmu->sched_task(ctx, false);
 
                         /*
                          * PMU specific parts of task perf context can require
                          * additional synchronization. As an example of such
                          * synchronization see implementation details of Intel
                          * LBR call stack data profiling;
                          */
                         if (pmu->swap_task_ctx)
                                 pmu->swap_task_ctx(ctx, next_ctx);
                         else
                                 swap(ctx->task_ctx_data, next_ctx->task_ctx_data);
 
                         perf_pmu_enable(pmu);
 
                         /*
                          * RCU_INIT_POINTER here is safe because we've not
                          * modified the ctx and the above modification of
                          * ctx->task and ctx->task_ctx_data are immaterial
                          * since those values are always verified under
                          * ctx->lock which we're now holding.
                          */
                         RCU_INIT_POINTER(task->perf_event_ctxp[ctxn], next_ctx);
                         RCU_INIT_POINTER(next->perf_event_ctxp[ctxn], ctx);
 
                         do_switch = 0;
 
                         perf_event_sync_stat(ctx, next_ctx);
                 }
                 raw_spin_unlock(&next_ctx->lock);
                 raw_spin_unlock(&ctx->lock);
         }
 unlock:
         rcu_read_unlock();
 
         if (do_switch) {
                 raw_spin_lock(&ctx->lock);
                 perf_pmu_disable(pmu);
 
                 if (cpuctx->sched_cb_usage && pmu->sched_task)
                         pmu->sched_task(ctx, false);
                 task_ctx_sched_out(cpuctx, ctx, EVENT_ALL);
 
                 perf_pmu_enable(pmu);
                 raw_spin_unlock(&ctx->lock);
         }
 }
 
 static DEFINE_PER_CPU(struct list_head, sched_cb_list);
 
 void perf_sched_cb_dec(struct pmu *pmu)
 {
         struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
 
         this_cpu_dec(perf_sched_cb_usages);
 
         if (!--cpuctx->sched_cb_usage)
                 list_del(&cpuctx->sched_cb_entry);
 }
 
 
 void perf_sched_cb_inc(struct pmu *pmu)
 {
         struct perf_cpu_context *cpuctx = this_cpu_ptr(pmu->pmu_cpu_context);
 
         if (!cpuctx->sched_cb_usage++)
                 list_add(&cpuctx->sched_cb_entry, this_cpu_ptr(&sched_cb_list));
 
         this_cpu_inc(perf_sched_cb_usages);
 }
 
 /*
  * This function provides the context switch callback to the lower code
  * layer. It is invoked ONLY when the context switch callback is enabled.
  *
  * This callback is relevant even to per-cpu events; for example multi event
  * PEBS requires this to provide PID/TID information. This requires we flush
  * all queued PEBS records before we context switch to a new task.
  */
 static void __perf_pmu_sched_task(struct perf_cpu_context *cpuctx, bool sched_in)
 {
         struct pmu *pmu;
 
         pmu = cpuctx->ctx.pmu; /* software PMUs will not have sched_task */
 
         if (WARN_ON_ONCE(!pmu->sched_task))
                 return;
 
         perf_ctx_lock(cpuctx, cpuctx->task_ctx);
         perf_pmu_disable(pmu);
 
         pmu->sched_task(cpuctx->task_ctx, sched_in);
 
         perf_pmu_enable(pmu);
         perf_ctx_unlock(cpuctx, cpuctx->task_ctx);
 }
 
 static void perf_pmu_sched_task(struct task_struct *prev,
                                 struct task_struct *next,
                                 bool sched_in)
 {
         struct perf_cpu_context *cpuctx;
 
         if (prev == next)
                 return;
 
         list_for_each_entry(cpuctx, this_cpu_ptr(&sched_cb_list), sched_cb_entry) {
                 /* will be handled in perf_event_context_sched_in/out */
                 if (cpuctx->task_ctx)
                         continue;
 
                 __perf_pmu_sched_task(cpuctx, sched_in);
         }
 }
 
 static void perf_event_switch(struct task_struct *task,
                               struct task_struct *next_prev, bool sched_in);
 
 #define for_each_task_context_nr(ctxn)                                  \
         for ((ctxn) = 0; (ctxn) < perf_nr_task_contexts; (ctxn)++)
 
 /*
  * Called from scheduler to remove the events of the current task,
  * with interrupts disabled.
  *
  * We stop each event and update the event value in event->count.
  *
  * This does not protect us against NMI, but disable()
  * sets the disabled bit in the control field of event _before_
  * accessing the event control register. If a NMI hits, then it will
  * not restart the event.
  */
 void __perf_event_task_sched_out(struct task_struct *task,
                                  struct task_struct *next)
 {
         int ctxn;
 
         if (__this_cpu_read(perf_sched_cb_usages))
                 perf_pmu_sched_task(task, next, false);
 
         if (atomic_read(&nr_switch_events))
                 perf_event_switch(task, next, false);
 
         for_each_task_context_nr(ctxn)
                 perf_event_context_sched_out(task, ctxn, next);
 
         /*
          * if cgroup events exist on this CPU, then we need
          * to check if we have to switch out PMU state.
          * cgroup event are system-wide mode only
          */
         if (atomic_read(this_cpu_ptr(&perf_cgroup_events)))
                 perf_cgroup_switch(next);
 }
 
 /*
  * Called with IRQs disabled
  */
 static void cpu_ctx_sched_out(struct perf_cpu_context *cpuctx,
                               enum event_type_t event_type)
 {
         ctx_sched_out(&cpuctx->ctx, cpuctx, event_type);
 }
 
 static bool perf_less_group_idx(const void *l, const void *r)
 {
         const struct perf_event *le = *(const struct perf_event **)l;
         const struct perf_event *re = *(const struct perf_event **)r;
 
         return le->group_index < re->group_index;
 }
 
 static void swap_ptr(void *l, void *r)
 {
         void **lp = l, **rp = r;
 
         swap(*lp, *rp);
 }
 
 static const struct min_heap_callbacks perf_min_heap = {
         .elem_size = sizeof(struct perf_event *),
         .less = perf_less_group_idx,
         .swp = swap_ptr,
 };
 
 static void __heap_add(struct min_heap *heap, struct perf_event *event)
 {
         struct perf_event **itrs = heap->data;
 
         if (event) {
                 itrs[heap->nr] = event;
                 heap->nr++;
         }
 }
 
 static noinline int visit_groups_merge(struct perf_cpu_context *cpuctx,
                                 struct perf_event_groups *groups, int cpu,
                                 int (*func)(struct perf_event *, void *),
                                 void *data)
 {
 #ifdef CONFIG_CGROUP_PERF
         struct cgroup_subsys_state *css = NULL;
 #endif
         /* Space for per CPU and/or any CPU event iterators. */
         struct perf_event *itrs[2];
         struct min_heap event_heap;
         struct perf_event **evt;
         int ret;
 
         if (cpuctx) {
                 event_heap = (struct min_heap){
                         .data = cpuctx->heap,
                         .nr = 0,
                         .size = cpuctx->heap_size,
                 };
 
                 lockdep_assert_held(&cpuctx->ctx.lock);
 
 #ifdef CONFIG_CGROUP_PERF
                 if (cpuctx->cgrp)
                         css = &cpuctx->cgrp->css;
 #endif
         } else {
                 event_heap = (struct min_heap){
                         .data = itrs,
                         .nr = 0,
                         .size = ARRAY_SIZE(itrs),
                 };
                 /* Events not within a CPU context may be on any CPU. */
                 __heap_add(&event_heap, perf_event_groups_first(groups, -1, NULL));
         }
         evt = event_heap.data;
 
         __heap_add(&event_heap, perf_event_groups_first(groups, cpu, NULL));
 
 #ifdef CONFIG_CGROUP_PERF
         for (; css; css = css->parent)
                 __heap_add(&event_heap, perf_event_groups_first(groups, cpu, css->cgroup));
 #endif
 
         min_heapify_all(&event_heap, &perf_min_heap);
 
         while (event_heap.nr) {
                 ret = func(*evt, data);
                 if (ret)
                         return ret;
 
                 *evt = perf_event_groups_next(*evt);
                 if (*evt)
                         min_heapify(&event_heap, 0, &perf_min_heap);
                 else
                         min_heap_pop(&event_heap, &perf_min_heap);
         }
 
         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_event_update_userpage(event);
 
         return true;
 }
 
 static inline void group_update_userpage(struct perf_event *group_event)
 {
         struct perf_event *event;
 
         if (!event_update_userpage(group_event))
                 return;
 
         for_each_sibling_event(event, group_event)
                 event_update_userpage(event);
 }
 
 static int merge_sched_in(struct perf_event *event, void *data)
 {
         struct perf_event_context *ctx = event->ctx;
         struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
         int *can_add_hw = data;
 
         if (event->state <= PERF_EVENT_STATE_OFF)
                 return 0;
 
         if (!event_filter_match(event))
                 return 0;
 
         if (group_can_go_on(event, cpuctx, *can_add_hw)) {
                 if (!group_sched_in(event, cpuctx, ctx))
                         list_add_tail(&event->active_list, get_event_list(event));
         }
 
         if (event->state == PERF_EVENT_STATE_INACTIVE) {
                 *can_add_hw = 0;
                 if (event->attr.pinned) {
                         perf_cgroup_event_disable(event, ctx);
                         perf_event_set_state(event, PERF_EVENT_STATE_ERROR);
                 } else {
                         ctx->rotate_necessary = 1;
                         perf_mux_hrtimer_restart(cpuctx);
                         group_update_userpage(event);
                 }
         }
 
         return 0;
 }
 
 static void
 ctx_pinned_sched_in(struct perf_event_context *ctx,
                     struct perf_cpu_context *cpuctx)
 {
         int can_add_hw = 1;
 
         if (ctx != &cpuctx->ctx)
                 cpuctx = NULL;
 
         visit_groups_merge(cpuctx, &ctx->pinned_groups,
                            smp_processor_id(),
                            merge_sched_in, &can_add_hw);
 }
 
 static void
 ctx_flexible_sched_in(struct perf_event_context *ctx,
                       struct perf_cpu_context *cpuctx)
 {
         int can_add_hw = 1;
 
         if (ctx != &cpuctx->ctx)
                 cpuctx = NULL;
 
         visit_groups_merge(cpuctx, &ctx->flexible_groups,
                            smp_processor_id(),
                            merge_sched_in, &can_add_hw);
 }
 
 static void
 ctx_sched_in(struct perf_event_context *ctx,
              struct perf_cpu_context *cpuctx,
              enum event_type_t event_type)
 {
         int is_active = ctx->is_active;
 
         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(cpuctx);
                 /*
                  * 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)
                         cpuctx->task_ctx = ctx;
                 else
                         WARN_ON_ONCE(cpuctx->task_ctx != ctx);
         }
 
         is_active ^= ctx->is_active; /* changed bits */
 
         /*
          * First go through the list and put on any pinned groups
          * in order to give them the best chance of going on.
          */
         if (is_active & EVENT_PINNED)
                 ctx_pinned_sched_in(ctx, cpuctx);
 
         /* Then walk through the lower prio flexible groups */
         if (is_active & EVENT_FLEXIBLE)
                 ctx_flexible_sched_in(ctx, cpuctx);
 }
 
 static void cpu_ctx_sched_in(struct perf_cpu_context *cpuctx,
                              enum event_type_t event_type)
 {
         struct perf_event_context *ctx = &cpuctx->ctx;
 
         ctx_sched_in(ctx, cpuctx, event_type);
 }
 
 static void perf_event_context_sched_in(struct perf_event_context *ctx,
                                         struct task_struct *task)
 {
         struct perf_cpu_context *cpuctx;
         struct pmu *pmu;
 
         cpuctx = __get_cpu_context(ctx);
 
         /*
          * HACK: for HETEROGENEOUS the task context might have switched to a
          * different PMU, force (re)set the context,
          */
         pmu = ctx->pmu = cpuctx->ctx.pmu;
 
         if (cpuctx->task_ctx == ctx) {
                 if (cpuctx->sched_cb_usage)
                         __perf_pmu_sched_task(cpuctx, true);
                 return;
         }
 
         perf_ctx_lock(cpuctx, ctx);
         /*
          * We must check ctx->nr_events while holding ctx->lock, such
          * that we serialize against perf_install_in_context().
          */
         if (!ctx->nr_events)
                 goto unlock;
 
         perf_pmu_disable(pmu);
         /*
          * We want to keep the following priority order:
          * cpu pinned (that don't need to move), task pinned,
          * cpu flexible, task flexible.
          *
          * However, if task's ctx is not carrying any pinned
          * events, no need to flip the cpuctx's events around.
          */
         if (!RB_EMPTY_ROOT(&ctx->pinned_groups.tree))
                 cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE);
         perf_event_sched_in(cpuctx, ctx);
 
         if (cpuctx->sched_cb_usage && pmu->sched_task)
                 pmu->sched_task(cpuctx->task_ctx, true);
 
         perf_pmu_enable(pmu);
 
 unlock:
         perf_ctx_unlock(cpuctx, ctx);
 }
 
 /*
  * Called from scheduler to add the events of the current task
  * with interrupts disabled.
  *
  * We restore the event value and then enable it.
  *
  * This does not protect us against NMI, but enable()
  * sets the enabled bit in the control field of event _before_
  * accessing the event control register. If a NMI hits, then it will
  * keep the event running.
  */
 void __perf_event_task_sched_in(struct task_struct *prev,
                                 struct task_struct *task)
 {
         struct perf_event_context *ctx;
         int ctxn;
 
         for_each_task_context_nr(ctxn) {
                 ctx = task->perf_event_ctxp[ctxn];
                 if (likely(!ctx))
                         continue;
 
                 perf_event_context_sched_in(ctx, task);
         }
 
         if (atomic_read(&nr_switch_events))
                 perf_event_switch(task, prev, true);
 
         if (__this_cpu_read(perf_sched_cb_usages))
                 perf_pmu_sched_task(prev, task, true);
 }
 
 static u64 perf_calculate_period(struct perf_event *event, u64 nsec, u64 count)
 {
         u64 frequency = event->attr.sample_freq;
         u64 sec = NSEC_PER_SEC;
         u64 divisor, dividend;
 
         int count_fls, nsec_fls, frequency_fls, sec_fls;
 
         count_fls = fls64(count);
         nsec_fls = fls64(nsec);
         frequency_fls = fls64(frequency);
         sec_fls = 30;
 
         /*
          * We got @count in @nsec, with a target of sample_freq HZ
          * the target period becomes:
          *
          *             @count * 10^9
          * period = -------------------
          *          @nsec * sample_freq
          *
          */
 
         /*
          * Reduce accuracy by one bit such that @a and @b converge
          * to a similar magnitude.
          */
 #define REDUCE_FLS(a, b)                \
 do {                                    \
         if (a##_fls > b##_fls) {        \
                 a >>= 1;                \
                 a##_fls--;              \
         } else {                        \
                 b >>= 1;                \
                 b##_fls--;              \
         }                               \
 } while (0)
 
         /*
          * Reduce accuracy until either term fits in a u64, then proceed with
          * the other, so that finally we can do a u64/u64 division.
          */
         while (count_fls + sec_fls > 64 && nsec_fls + frequency_fls > 64) {
                 REDUCE_FLS(nsec, frequency);
                 REDUCE_FLS(sec, count);
         }
 
         if (count_fls + sec_fls > 64) {
                 divisor = nsec * frequency;
 
                 while (count_fls + sec_fls > 64) {
                         REDUCE_FLS(count, sec);
                         divisor >>= 1;
                 }
 
                 dividend = count * sec;
         } else {
                 dividend = count * sec;
 
                 while (nsec_fls + frequency_fls > 64) {
                         REDUCE_FLS(nsec, frequency);
                         dividend >>= 1;
                 }
 
                 divisor = nsec * frequency;
         }
 
         if (!divisor)
                 return dividend;
 
         return div64_u64(dividend, divisor);
 }
 
 static DEFINE_PER_CPU(int, perf_throttled_count);
 static DEFINE_PER_CPU(u64, perf_throttled_seq);
 
 static void perf_adjust_period(struct perf_event *event, u64 nsec, u64 count, bool disable)
 {
         struct hw_perf_event *hwc = &event->hw;
         s64 period, sample_period;
         s64 delta;
 
         period = perf_calculate_period(event, nsec, count);
 
         delta = (s64)(period - hwc->sample_period);
         delta = (delta + 7) / 8; /* low pass filter */
 
         sample_period = hwc->sample_period + delta;
 
         if (!sample_period)
                 sample_period = 1;
 
         hwc->sample_period = sample_period;
 
         if (local64_read(&hwc->period_left) > 8*sample_period) {
                 if (disable)
                         event->pmu->stop(event, PERF_EF_UPDATE);
 
                 local64_set(&hwc->period_left, 0);
 
                 if (disable)
                         event->pmu->start(event, PERF_EF_RELOAD);
         }
 }
 
 /*
  * combine freq adjustment with unthrottling to avoid two passes over the
  * events. At the same time, make sure, having freq events does not change
  * the rate of unthrottling as that would introduce bias.
  */
 static void perf_adjust_freq_unthr_context(struct perf_event_context *ctx,
                                            int needs_unthr)
 {
         struct perf_event *event;
         struct hw_perf_event *hwc;
         u64 now, period = TICK_NSEC;
         s64 delta;
 
         /*
          * only need to iterate over all events iff:
          * - context have events in frequency mode (needs freq adjust)
          * - there are events to unthrottle on this cpu
          */
         if (!(ctx->nr_freq || needs_unthr))
                 return;
 
         raw_spin_lock(&ctx->lock);
         perf_pmu_disable(ctx->pmu);
 
         list_for_each_entry_rcu(event, &ctx->event_list, event_entry) {
                 if (event->state != PERF_EVENT_STATE_ACTIVE)
                         continue;
 
                 if (!event_filter_match(event))
                         continue;
 
                 perf_pmu_disable(event->pmu);
 
                 hwc = &event->hw;
 
                 if (hwc->interrupts == MAX_INTERRUPTS) {
                         hwc->interrupts = 0;
                         perf_log_throttle(event, 1);
                         event->pmu->start(event, 0);
                 }
 
                 if (!event->attr.freq || !event->attr.sample_freq)
                         goto next;
 
                 /*
                  * stop the event and update event->count
                  */
                 event->pmu->stop(event, PERF_EF_UPDATE);
 
                 now = local64_read(&event->count);
                 delta = now - hwc->freq_count_stamp;
                 hwc->freq_count_stamp = now;
 
                 /*
                  * restart the event
                  * reload only if value has changed
                  * we have stopped the event so tell that
                  * to perf_adjust_period() to avoid stopping it
                  * twice.
                  */
                 if (delta > 0)
                         perf_adjust_period(event, period, delta, false);
 
                 event->pmu->start(event, delta > 0 ? PERF_EF_RELOAD : 0);
         next:
                 perf_pmu_enable(event->pmu);
         }
 
         perf_pmu_enable(ctx->pmu);
         raw_spin_unlock(&ctx->lock);
 }
 
 /*
  * Move @event to the tail of the @ctx's elegible events.
  */
 static void rotate_ctx(struct perf_event_context *ctx, struct perf_event *event)
 {
         /*
          * Rotate the first entry last of non-pinned groups. Rotation might be
          * disabled by the inheritance code.
          */
         if (ctx->rotate_disable)
                 return;
 
         perf_event_groups_delete(&ctx->flexible_groups, event);
         perf_event_groups_insert(&ctx->flexible_groups, event);
 }
 
 /* pick an event from the flexible_groups to rotate */
 static inline struct perf_event *
 ctx_event_to_rotate(struct perf_event_context *ctx)
 {
         struct perf_event *event;
 
         /* pick the first active flexible event */
         event = list_first_entry_or_null(&ctx->flexible_active,
                                          struct perf_event, active_list);
 
         /* if no active flexible event, pick the first event */
         if (!event) {
                 event = rb_entry_safe(rb_first(&ctx->flexible_groups.tree),
                                       typeof(*event), group_node);
         }
 
         /*
          * Unconditionally clear rotate_necessary; if ctx_flexible_sched_in()
          * finds there are unschedulable events, it will set it again.
          */
         ctx->rotate_necessary = 0;
 
         return event;
 }
 
 static bool perf_rotate_context(struct perf_cpu_context *cpuctx)
 {
         struct perf_event *cpu_event = NULL, *task_event = NULL;
         struct perf_event_context *task_ctx = NULL;
         int cpu_rotate, task_rotate;
 
         /*
          * Since we run this from IRQ context, nobody can install new
          * events, thus the event count values are stable.
          */
 
         cpu_rotate = cpuctx->ctx.rotate_necessary;
         task_ctx = cpuctx->task_ctx;
         task_rotate = task_ctx ? task_ctx->rotate_necessary : 0;
 
         if (!(cpu_rotate || task_rotate))
                 return false;
 
         perf_ctx_lock(cpuctx, cpuctx->task_ctx);
         perf_pmu_disable(cpuctx->ctx.pmu);
 
         if (task_rotate)
                 task_event = ctx_event_to_rotate(task_ctx);
         if (cpu_rotate)
                 cpu_event = ctx_event_to_rotate(&cpuctx->ctx);
 
         /*
          * As per the order given at ctx_resched() first 'pop' task flexible
          * and then, if needed CPU flexible.
          */
         if (task_event || (task_ctx && cpu_event))
                 ctx_sched_out(task_ctx, cpuctx, EVENT_FLEXIBLE);
         if (cpu_event)
                 cpu_ctx_sched_out(cpuctx, EVENT_FLEXIBLE);
 
         if (task_event)
                 rotate_ctx(task_ctx, task_event);
         if (cpu_event)
                 rotate_ctx(&cpuctx->ctx, cpu_event);
 
         perf_event_sched_in(cpuctx, task_ctx);
 
         perf_pmu_enable(cpuctx->ctx.pmu);
         perf_ctx_unlock(cpuctx, cpuctx->task_ctx);
 
         return true;
 }
 
 void perf_event_task_tick(void)
 {
         struct list_head *head = this_cpu_ptr(&active_ctx_list);
         struct perf_event_context *ctx, *tmp;
         int throttled;
 
         lockdep_assert_irqs_disabled();
 
         __this_cpu_inc(perf_throttled_seq);
         throttled = __this_cpu_xchg(perf_throttled_count, 0);
         tick_dep_clear_cpu(smp_processor_id(), TICK_DEP_BIT_PERF_EVENTS);
 
         list_for_each_entry_safe(ctx, tmp, head, active_ctx_list)
                 perf_adjust_freq_unthr_context(ctx, throttled);
 }
 
 static int event_enable_on_exec(struct perf_event *event,
                                 struct perf_event_context *ctx)
 {
         if (!event->attr.enable_on_exec)
                 return 0;
 
         event->attr.enable_on_exec = 0;
         if (event->state >= PERF_EVENT_STATE_INACTIVE)
                 return 0;
 
         perf_event_set_state(event, PERF_EVENT_STATE_INACTIVE);
 
         return 1;
 }
 
 /*
  * Enable all of a task's events that have been marked enable-on-exec.
  * This expects task == current.
  */
 static void perf_event_enable_on_exec(int ctxn)
 {
         struct perf_event_context *ctx, *clone_ctx = NULL;
         enum event_type_t event_type = 0;
         struct perf_cpu_context *cpuctx;
         struct perf_event *event;
         unsigned long flags;
         int enabled = 0;
 
         local_irq_save(flags);
         ctx = current->perf_event_ctxp[ctxn];
         if (!ctx || !ctx->nr_events)
                 goto out;
 
         cpuctx = __get_cpu_context(ctx);
         perf_ctx_lock(cpuctx, ctx);
         ctx_sched_out(ctx, cpuctx, EVENT_TIME);
         list_for_each_entry(event, &ctx->event_list, event_entry) {
                 enabled |= event_enable_on_exec(event, ctx);
                 event_type |= get_event_type(event);
         }
 
         /*
          * Unclone and reschedule this context if we enabled any event.
          */
         if (enabled) {
                 clone_ctx = unclone_ctx(ctx);
                 ctx_resched(cpuctx, ctx, event_type);
         } else {
                 ctx_sched_in(ctx, cpuctx, EVENT_TIME);
         }
         perf_ctx_unlock(cpuctx, ctx);
 
 out:
         local_irq_restore(flags);
 
         if (clone_ctx)
                 put_ctx(clone_ctx);
 }
 
 static void perf_remove_from_owner(struct perf_event *event);
 static void perf_event_exit_event(struct perf_event *event,
                                   struct perf_event_context *ctx);
 
 /*
  * Removes all events from the current task that have been marked
  * remove-on-exec, and feeds their values back to parent events.
  */
 static void perf_event_remove_on_exec(int ctxn)
 {
         struct perf_event_context *ctx, *clone_ctx = NULL;
         struct perf_event *event, *next;
         LIST_HEAD(free_list);
         unsigned long flags;
         bool modified = false;
 
         ctx = perf_pin_task_context(current, ctxn);
         if (!ctx)
                 return;
 
         mutex_lock(&ctx->mutex);
 
         if (WARN_ON_ONCE(ctx->task != current))
                 goto unlock;
 
         list_for_each_entry_safe(event, next, &ctx->event_list, event_entry) {
                 if (!event->attr.remove_on_exec)
                         continue;
 
                 if (!is_kernel_event(event))
                         perf_remove_from_owner(event);
 
                 modified = true;
 
                 perf_event_exit_event(event, ctx);
         }
 
         raw_spin_lock_irqsave(&ctx->lock, flags);
         if (modified)
                 clone_ctx = unclone_ctx(ctx);
         --ctx->pin_count;
         raw_spin_unlock_irqrestore(&ctx->lock, flags);
 
 unlock:
         mutex_unlock(&ctx->mutex);
 
         put_ctx(ctx);
         if (clone_ctx)
                 put_ctx(clone_ctx);
 }
 
 struct perf_read_data {
         struct perf_event *event;
         bool group;
         int ret;
 };
 
 static int __perf_event_read_cpu(struct perf_event *event, int event_cpu)
 {
         u16 local_pkg, event_pkg;
 
         if (event->group_caps & PERF_EV_CAP_READ_ACTIVE_PKG) {
                 int local_cpu = smp_processor_id();
 
                 event_pkg = topology_physical_package_id(event_cpu);
                 local_pkg = topology_physical_package_id(local_cpu);
 
                 if (event_pkg == local_pkg)
                         return local_cpu;
         }
 
         return event_cpu;
 }
 
 /*
  * Cross CPU call to read the hardware event
  */
 static void __perf_event_read(void *info)
 {
         struct perf_read_data *data = info;
         struct perf_event *sub, *event = data->event;
         struct perf_event_context *ctx = event->ctx;
         struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
         struct pmu *pmu = event->pmu;
 
         /*
          * If this is a task context, we need to check whether it is
          * the current task context of this cpu.  If not it has been
          * scheduled out before the smp call arrived.  In that case
          * event->count would have been updated to a recent sample
          * when the event was scheduled out.
          */
         if (ctx->task && cpuctx->task_ctx != ctx)
                 return;
 
         raw_spin_lock(&ctx->lock);
         if (ctx->is_active & EVENT_TIME) {
                 update_context_time(ctx);
                 update_cgrp_time_from_event(event);
         }
 
         perf_event_update_time(event);
         if (data->group)
                 perf_event_update_sibling_time(event);
 
         if (event->state != PERF_EVENT_STATE_ACTIVE)
                 goto unlock;
 
         if (!data->group) {
                 pmu->read(event);
                 data->ret = 0;
                 goto unlock;
         }
 
         pmu->start_txn(pmu, PERF_PMU_TXN_READ);
 
         pmu->read(event);
 
         for_each_sibling_event(sub, event) {
                 if (sub->state == PERF_EVENT_STATE_ACTIVE) {
                         /*
                          * Use sibling's PMU rather than @event's since
                          * sibling could be on different (eg: software) PMU.
                          */
                         sub->pmu->read(sub);
                 }
         }
 
         data->ret = pmu->commit_txn(pmu);
 
 unlock:
         raw_spin_unlock(&ctx->lock);
 }
 
 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:
  *   - either for the current task, or for this CPU
  *   - does not have inherit set, for inherited task events
  *     will not be local and we cannot read them atomically
  *   - must not have a pmu::count method
  */
 int perf_event_read_local(struct perf_event *event, u64 *value,
                           u64 *enabled, u64 *running)
 {
         unsigned long flags;
         int ret = 0;
 
         /*
          * Disabling interrupts avoids all counter scheduling (context
          * switches, timer based rotation and IPIs).
          */
         local_irq_save(flags);
 
         /*
          * It must not be an event with inherit set, we cannot read
          * all child counters from atomic context.
          */
         if (event->attr.inherit) {
                 ret = -EOPNOTSUPP;
                 goto out;
         }
 
         /* If this is a per-task event, it must be for current */
         if ((event->attach_state & PERF_ATTACH_TASK) &&
             event->hw.target != current) {
                 ret = -EINVAL;
                 goto out;
         }
 
         /* If this is a per-CPU event, it must be for this CPU */
         if (!(event->attach_state & PERF_ATTACH_TASK) &&
             event->cpu != smp_processor_id()) {
                 ret = -EINVAL;
                 goto out;
         }
 
         /* If this is a pinned event it must be running on this CPU */
         if (event->attr.pinned && event->oncpu != smp_processor_id()) {
                 ret = -EBUSY;
                 goto out;
         }
 
         /*
          * If the event is currently on this CPU, its either a per-task event,
          * or local to this CPU. Furthermore it means its ACTIVE (otherwise
          * oncpu == -1).
          */
         if (event->oncpu == smp_processor_id())
                 event->pmu->read(event);
 
         *value = local64_read(&event->count);
         if (enabled || running) {
                 u64 __enabled, __running, __now;;
 
                 calc_timer_values(event, &__now, &__enabled, &__running);
                 if (enabled)
                         *enabled = __enabled;
                 if (running)
                         *running = __running;
         }
 out:
         local_irq_restore(flags);
 
         return ret;
 }
 
 static int perf_event_read(struct perf_event *event, bool group)
 {
         enum perf_event_state state = READ_ONCE(event->state);
         int event_cpu, ret = 0;
 
         /*
          * If event is enabled and currently active on a CPU, update the
          * value in the event structure:
          */
 again:
         if (state == PERF_EVENT_STATE_ACTIVE) {
                 struct perf_read_data data;
 
                 /*
                  * Orders the ->state and ->oncpu loads such that if we see
                  * ACTIVE we must also see the right ->oncpu.
                  *
                  * Matches the smp_wmb() from event_sched_in().
                  */
                 smp_rmb();
 
                 event_cpu = READ_ONCE(event->oncpu);
                 if ((unsigned)event_cpu >= nr_cpu_ids)
                         return 0;
 
                 data = (struct perf_read_data){
                         .event = event,
                         .group = group,
                         .ret = 0,
                 };
 
                 preempt_disable();
                 event_cpu = __perf_event_read_cpu(event, event_cpu);
 
                 /*
                  * Purposely ignore the smp_call_function_single() return
                  * value.
                  *
                  * If event_cpu isn't a valid CPU it means the event got
                  * scheduled out and that will have updated the event count.
                  *
                  * Therefore, either way, we'll have an up-to-date event count
                  * after this.
                  */
                 (void)smp_call_function_single(event_cpu, __perf_event_read, &data, 1);
                 preempt_enable();
                 ret = data.ret;
 
         } else if (state == PERF_EVENT_STATE_INACTIVE) {
                 struct perf_event_context *ctx = event->ctx;
                 unsigned long flags;
 
                 raw_spin_lock_irqsave(&ctx->lock, flags);
                 state = event->state;
                 if (state != PERF_EVENT_STATE_INACTIVE) {
                         raw_spin_unlock_irqrestore(&ctx->lock, flags);
                         goto again;
                 }
 
                 /*
                  * May read while context is not active (e.g., thread is
                  * blocked), in that case we cannot update context time
                  */
                 if (ctx->is_active & EVENT_TIME) {
                         update_context_time(ctx);
                         update_cgrp_time_from_event(event);
                 }
 
                 perf_event_update_time(event);
                 if (group)
                         perf_event_update_sibling_time(event);
                 raw_spin_unlock_irqrestore(&ctx->lock, flags);
         }
 
         return ret;
 }
 
 /*
  * Initialize the perf_event context in a task_struct:
  */
 static void __perf_event_init_context(struct perf_event_context *ctx)
 {
         raw_spin_lock_init(&ctx->lock);
         mutex_init(&ctx->mutex);
         INIT_LIST_HEAD(&ctx->active_ctx_list);
         perf_event_groups_init(&ctx->pinned_groups);
         perf_event_groups_init(&ctx->flexible_groups);
         INIT_LIST_HEAD(&ctx->event_list);
         INIT_LIST_HEAD(&ctx->pinned_active);
         INIT_LIST_HEAD(&ctx->flexible_active);
         refcount_set(&ctx->refcount, 1);
 }
 
 static struct perf_event_context *
 alloc_perf_context(struct pmu *pmu, struct task_struct *task)
 {
         struct perf_event_context *ctx;
 
         ctx = kzalloc(sizeof(struct perf_event_context), GFP_KERNEL);
         if (!ctx)
                 return NULL;
 
         __perf_event_init_context(ctx);
         if (task)
                 ctx->task = get_task_struct(task);
         ctx->pmu = pmu;
 
         return ctx;
 }
 
 static struct task_struct *
 find_lively_task_by_vpid(pid_t vpid)
 {
         struct task_struct *task;
 
         rcu_read_lock();
         if (!vpid)
                 task = current;
         else
                 task = find_task_by_vpid(vpid);
         if (task)
                 get_task_struct(task);
         rcu_read_unlock();
 
         if (!task)
                 return ERR_PTR(-ESRCH);
 
         return task;
 }
 
 /*
  * Returns a matching context with refcount and pincount.
  */
 static struct perf_event_context *
 find_get_context(struct pmu *pmu, struct task_struct *task,
                 struct perf_event *event)
 {
         struct perf_event_context *ctx, *clone_ctx = NULL;
         struct perf_cpu_context *cpuctx;
         void *task_ctx_data = NULL;
         unsigned long flags;
         int ctxn, err;
         int cpu = event->cpu;
 
         if (!task) {
                 /* Must be root to operate on a CPU event: */
                 err = perf_allow_cpu(&event->attr);
                 if (err)
                         return ERR_PTR(err);
 
                 cpuctx = per_cpu_ptr(pmu->pmu_cpu_context, cpu);
                 ctx = &cpuctx->ctx;
                 get_ctx(ctx);
                 raw_spin_lock_irqsave(&ctx->lock, flags);
                 ++ctx->pin_count;
                 raw_spin_unlock_irqrestore(&ctx->lock, flags);
 
                 return ctx;
         }
 
         err = -EINVAL;
         ctxn = pmu->task_ctx_nr;
         if (ctxn < 0)
                 goto errout;
 
         if (event->attach_state & PERF_ATTACH_TASK_DATA) {
                 task_ctx_data = alloc_task_ctx_data(pmu);
                 if (!task_ctx_data) {
                         err = -ENOMEM;
                         goto errout;
                 }
         }
 
 retry:
         ctx = perf_lock_task_context(task, ctxn, &flags);
         if (ctx) {
                 clone_ctx = unclone_ctx(ctx);
                 ++ctx->pin_count;
 
                 if (task_ctx_data && !ctx->task_ctx_data) {
                         ctx->task_ctx_data = task_ctx_data;
                         task_ctx_data = NULL;
                 }
                 raw_spin_unlock_irqrestore(&ctx->lock, flags);
 
                 if (clone_ctx)
                         put_ctx(clone_ctx);
         } else {
                 ctx = alloc_perf_context(pmu, task);
                 err = -ENOMEM;
                 if (!ctx)
                         goto errout;
 
                 if (task_ctx_data) {
                         ctx->task_ctx_data = task_ctx_data;
                         task_ctx_data = NULL;
                 }
 
                 err = 0;
                 mutex_lock(&task->perf_event_mutex);
                 /*
                  * If it has already passed perf_event_exit_task().
                  * we must see PF_EXITING, it takes this mutex too.
                  */
                 if (task->flags & PF_EXITING)
                         err = -ESRCH;
                 else if (task->perf_event_ctxp[ctxn])
                         err = -EAGAIN;
                 else {
                         get_ctx(ctx);
                         ++ctx->pin_count;
                         rcu_assign_pointer(task->perf_event_ctxp[ctxn], ctx);
                 }
                 mutex_unlock(&task->perf_event_mutex);
 
                 if (unlikely(err)) {
                         put_ctx(ctx);
 
                         if (err == -EAGAIN)
                                 goto retry;
                         goto errout;
                 }
         }
 
         free_task_ctx_data(pmu, task_ctx_data);
         return ctx;
 
 errout:
         free_task_ctx_data(pmu, task_ctx_data);
         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);
         kmem_cache_free(perf_event_cache, event);
 }
 
 static void ring_buffer_attach(struct perf_event *event,
                                struct perf_buffer *rb);
 
 static void detach_sb_event(struct perf_event *event)
 {
         struct pmu_event_list *pel = per_cpu_ptr(&pmu_sb_events, event->cpu);
 
         raw_spin_lock(&pel->lock);
         list_del_rcu(&event->sb_list);
         raw_spin_unlock(&pel->lock);
 }
 
 static bool is_sb_event(struct perf_event *event)
 {
         struct perf_event_attr *attr = &event->attr;
 
         if (event->parent)
                 return false;
 
         if (event->attach_state & PERF_ATTACH_TASK)
                 return false;
 
         if (attr->mmap || attr->mmap_data || attr->mmap2 ||
             attr->comm || attr->comm_exec ||
             attr->task || attr->ksymbol ||
             attr->context_switch || attr->text_poke ||
             attr->bpf_event)
                 return true;
         return false;
 }
 
 static void unaccount_pmu_sb_event(struct perf_event *event)
 {
         if (is_sb_event(event))
                 detach_sb_event(event);
 }
 
 static void unaccount_event_cpu(struct perf_event *event, int cpu)
 {
         if (event->parent)
                 return;
 
         if (is_cgroup_event(event))
                 atomic_dec(&per_cpu(perf_cgroup_events, cpu));
 }
 
 #ifdef CONFIG_NO_HZ_FULL
 static DEFINE_SPINLOCK(nr_freq_lock);
 #endif
 
 static void unaccount_freq_event_nohz(void)
 {
 #ifdef CONFIG_NO_HZ_FULL
         spin_lock(&nr_freq_lock);
         if (atomic_dec_and_test(&nr_freq_events))
                 tick_nohz_dep_clear(TICK_DEP_BIT_PERF_EVENTS);
         spin_unlock(&nr_freq_lock);
 #endif
 }
 
 static void unaccount_freq_event(void)
 {
         if (tick_nohz_full_enabled())
                 unaccount_freq_event_nohz();
         else
                 atomic_dec(&nr_freq_events);
 }
 
 static void unaccount_event(struct perf_event *event)
 {
         bool dec = false;
 
         if (event->parent)
                 return;
 
         if (event->attach_state & (PERF_ATTACH_TASK | PERF_ATTACH_SCHED_CB))
                 dec = true;
         if (event->attr.mmap || event->attr.mmap_data)
                 atomic_dec(&nr_mmap_events);
         if (event->attr.build_id)
                 atomic_dec(&nr_build_id_events);
         if (event->attr.comm)
                 atomic_dec(&nr_comm_events);
         if (event->attr.namespaces)
                 atomic_dec(&nr_namespaces_events);
         if (event->attr.cgroup)
                 atomic_dec(&nr_cgroup_events);
         if (event->attr.task)
                 atomic_dec(&nr_task_events);
         if (event->attr.freq)
                 unaccount_freq_event();
         if (event->attr.context_switch) {
                 dec = true;
                 atomic_dec(&nr_switch_events);
         }
         if (is_cgroup_event(event))
                 dec = true;
         if (has_branch_stack(event))
                 dec = true;
         if (event->attr.ksymbol)
                 atomic_dec(&nr_ksymbol_events);
         if (event->attr.bpf_event)
                 atomic_dec(&nr_bpf_events);
         if (event->attr.text_poke)
                 atomic_dec(&nr_text_poke_events);
 
         if (dec) {
                 if (!atomic_add_unless(&perf_sched_count, -1, 1))
                         schedule_delayed_work(&perf_sched_work, HZ);
         }
 
         unaccount_event_cpu(event, event->cpu);
 
         unaccount_pmu_sb_event(event);
 }
 
 static void perf_sched_delayed(struct work_struct *work)
 {
         mutex_lock(&perf_sched_mutex);
         if (atomic_dec_and_test(&perf_sched_count))
                 static_branch_disable(&perf_sched_events);
         mutex_unlock(&perf_sched_mutex);
 }
 
 /*
  * The following implement mutual exclusion of events on "exclusive" pmus
  * (PERF_PMU_CAP_EXCLUSIVE). Such pmus can only have one event scheduled
  * at a time, so we disallow creating events that might conflict, namely:
  *
  *  1) cpu-wide events in the presence of per-task events,
  *  2) per-task events in the presence of cpu-wide events,
  *  3) two matching events on the same context.
  *
  * The former two cases are handled in the allocation path (perf_event_alloc(),
  * _free_event()), the latter -- before the first perf_install_in_context().
  */
 static int exclusive_event_init(struct perf_event *event)
 {
         struct pmu *pmu = event->pmu;
 
         if (!is_exclusive_pmu(pmu))
                 return 0;
 
         /*
          * Prevent co-existence of per-task and cpu-wide events on the
          * same exclusive pmu.
          *
          * Negative pmu::exclusive_cnt means there are cpu-wide
          * events on this "exclusive" pmu, positive means there are
          * per-task events.
          *
          * Since this is called in perf_event_alloc() path, event::ctx
          * doesn't exist yet; it is, however, safe to use PERF_ATTACH_TASK
          * to mean "per-task event", because unlike other attach states it
          * never gets cleared.
          */
         if (event->attach_state & PERF_ATTACH_TASK) {
                 if (!atomic_inc_unless_negative(&pmu->exclusive_cnt))
                         return -EBUSY;
         } else {
                 if (!atomic_dec_unless_positive(&pmu->exclusive_cnt))
                         return -EBUSY;
         }
 
         return 0;
 }
 
 static void exclusive_event_destroy(struct perf_event *event)
 {
         struct pmu *pmu = event->pmu;
 
         if (!is_exclusive_pmu(pmu))
                 return;
 
         /* see comment in exclusive_event_init() */
         if (event->attach_state & PERF_ATTACH_TASK)
                 atomic_dec(&pmu->exclusive_cnt);
         else
                 atomic_inc(&pmu->exclusive_cnt);
 }
 
 static bool exclusive_event_match(struct perf_event *e1, struct perf_event *e2)
 {
         if ((e1->pmu == e2->pmu) &&
             (e1->cpu == e2->cpu ||
              e1->cpu == -1 ||
              e2->cpu == -1))
                 return true;
         return false;
 }
 
 static bool exclusive_event_installable(struct perf_event *event,
                                         struct perf_event_context *ctx)
 {
         struct perf_event *iter_event;
         struct pmu *pmu = event->pmu;
 
         lockdep_assert_held(&ctx->mutex);
 
         if (!is_exclusive_pmu(pmu))
                 return true;
 
         list_for_each_entry(iter_event, &ctx->event_list, event_entry) {
                 if (exclusive_event_match(iter_event, event))
                         return false;
         }
 
         return true;
 }
 
 static void perf_addr_filters_splice(struct perf_event *event,
                                        struct list_head *head);
 
 static void _free_event(struct perf_event *event)
 {
         irq_work_sync(&event->pending);
 
         unaccount_event(event);
 
         security_perf_event_free(event);
 
         if (event->rb) {
                 /*
                  * Can happen when we close an event with re-directed output.
                  *
                  * Since we have a 0 refcount, perf_mmap_close() will skip
                  * over us; possibly making our ring_buffer_put() the last.
                  */
                 mutex_lock(&event->mmap_mutex);
                 ring_buffer_attach(event, NULL);
                 mutex_unlock(&event->mmap_mutex);
         }
 
         if (is_cgroup_event(event))
                 perf_detach_cgroup(event);
 
         if (!event->parent) {
                 if (event->attr.sample_type & PERF_SAMPLE_CALLCHAIN)
                         put_callchain_buffers();
         }
 
         perf_event_free_bpf_prog(event);
         perf_addr_filters_splice(event, NULL);
         kfree(event->addr_filter_ranges);
 
         if (event->destroy)
                 event->destroy(event);
 
         /*
          * Must be after ->destroy(), due to uprobe_perf_close() using
          * hw.target.
          */
         if (event->hw.target)
                 put_task_struct(event->hw.target);
 
         /*
          * perf_event_free_task() relies on put_ctx() being 'last', in particular
          * all task references must be cleaned up.
          */
         if (event->ctx)
                 put_ctx(event->ctx);
 
         exclusive_event_destroy(event);
         module_put(event->pmu->module);
 
         call_rcu(&event->rcu_head, free_event_rcu);
 }
 
 /*
  * Used to free events which have a known refcount of 1, such as in error paths
  * where the event isn't exposed yet and inherited events.
  */
 static void free_event(struct perf_event *event)
 {
         if (WARN(atomic_long_cmpxchg(&event->refcount, 1, 0) != 1,
                                 "unexpected event refcount: %ld; ptr=%p\n",
                                 atomic_long_read(&event->refcount), event)) {
                 /* leak to avoid use-after-free */
                 return;
         }
 
         _free_event(event);
 }
 
 /*
  * Remove user event from the owner task.
  */
 static void perf_remove_from_owner(struct perf_event *event)
 {
         struct task_struct *owner;
 
         rcu_read_lock();
         /*
          * Matches the smp_store_release() in perf_event_exit_task(). If we
          * observe !owner it means the list deletion is complete and we can
          * indeed free this event, otherwise we need to serialize on
          * owner->perf_event_mutex.
          */
         owner = READ_ONCE(event->owner);
         if (owner) {
                 /*
                  * Since delayed_put_task_struct() also drops the last
                  * task reference we can safely take a new reference
                  * while holding the rcu_read_lock().
                  */
                 get_task_struct(owner);
         }
         rcu_read_unlock();
 
         if (owner) {
                 /*
                  * If we're here through perf_event_exit_task() we're already
                  * holding ctx->mutex which would be an inversion wrt. the
                  * normal lock order.
                  *
                  * However we can safely take this lock because its the child
                  * ctx->mutex.
                  */
                 mutex_lock_nested(&owner->perf_event_mutex, SINGLE_DEPTH_NESTING);
 
                 /*
                  * We have to re-check the event->owner field, if it is cleared
                  * we raced with perf_event_exit_task(), acquiring the mutex
                  * ensured they're done, and we can proceed with freeing the
                  * event.
                  */
                 if (event->owner) {
                         list_del_init(&event->owner_entry);
                         smp_store_release(&event->owner, NULL);
                 }
                 mutex_unlock(&owner->perf_event_mutex);
                 put_task_struct(owner);
         }
 }
 
 static void put_event(struct perf_event *event)
 {
         if (!atomic_long_dec_and_test(&event->refcount))
                 return;
 
         _free_event(event);
 }
 
 /*
  * Kill an event dead; while event:refcount will preserve the event
  * object, it will not preserve its functionality. Once the last 'user'
  * gives up the object, we'll destroy the thing.
  */
 int perf_event_release_kernel(struct perf_event *event)
 {
         struct perf_event_context *ctx = event->ctx;
         struct perf_event *child, *tmp;
         LIST_HEAD(free_list);
 
         /*
          * If we got here through err_file: fput(event_file); we will not have
          * attached to a context yet.
          */
         if (!ctx) {
                 WARN_ON_ONCE(event->attach_state &
                                 (PERF_ATTACH_CONTEXT|PERF_ATTACH_GROUP));
                 goto no_ctx;
         }
 
         if (!is_kernel_event(event))
                 perf_remove_from_owner(event);
 
         ctx = perf_event_ctx_lock(event);
         WARN_ON_ONCE(ctx->parent_ctx);
         perf_remove_from_context(event, DETACH_GROUP);
 
         raw_spin_lock_irq(&ctx->lock);
         /*
          * Mark this event as STATE_DEAD, there is no external reference to it
          * anymore.
          *
          * Anybody acquiring event->child_mutex after the below loop _must_
          * also see this, most importantly inherit_event() which will avoid
          * placing more children on the list.
          *
          * Thus this guarantees that we will in fact observe and kill _ALL_
          * child events.
          */
         event->state = PERF_EVENT_STATE_DEAD;
         raw_spin_unlock_irq(&ctx->lock);
 
         perf_event_ctx_unlock(event, ctx);
 
 again:
         mutex_lock(&event->child_mutex);
         list_for_each_entry(child, &event->child_list, child_list) {
 
                 /*
                  * Cannot change, child events are not migrated, see the
                  * comment with perf_event_ctx_lock_nested().
                  */
                 ctx = READ_ONCE(child->ctx);
                 /*
                  * Since child_mutex nests inside ctx::mutex, we must jump
                  * through hoops. We start by grabbing a reference on the ctx.
                  *
                  * Since the event cannot get freed while we hold the
                  * child_mutex, the context must also exist and have a !0
                  * reference count.
                  */
                 get_ctx(ctx);
 
                 /*
                  * Now that we have a ctx ref, we can drop child_mutex, and
                  * acquire ctx::mutex without fear of it going away. Then we
                  * can re-acquire child_mutex.
                  */
                 mutex_unlock(&event->child_mutex);
                 mutex_lock(&ctx->mutex);
                 mutex_lock(&event->child_mutex);
 
                 /*
                  * Now that we hold ctx::mutex and child_mutex, revalidate our
                  * state, if child is still the first entry, it didn't get freed
                  * and we can continue doing so.
                  */
                 tmp = list_first_entry_or_null(&event->child_list,
                                                struct perf_event, child_list);
                 if (tmp == child) {
                         perf_remove_from_context(child, DETACH_GROUP);
                         list_move(&child->child_list, &free_list);
                         /*
                          * This matches the refcount bump in inherit_event();
                          * this can't be the last reference.
                          */
                         put_event(event);
                 }
 
                 mutex_unlock(&event->child_mutex);
                 mutex_unlock(&ctx->mutex);
                 put_ctx(ctx);
                 goto again;
         }
         mutex_unlock(&event->child_mutex);
 
         list_for_each_entry_safe(child, tmp, &free_list, child_list) {
                 void *var = &child->ctx->refcount;
 
                 list_del(&child->child_list);
                 free_event(child);
 
                 /*
                  * Wake any perf_event_free_task() waiting for this event to be
                  * freed.
                  */
                 smp_mb(); /* pairs with wait_var_event() */
                 wake_up_var(var);
         }
 
 no_ctx:
         put_event(event); /* Must be the 'last' reference */
         return 0;
 }
 EXPORT_SYMBOL_GPL(perf_event_release_kernel);
 
 /*
  * Called when the last reference to the file is gone.
  */
 static int perf_release(struct inode *inode, struct file *file)
 {
         perf_event_release_kernel(file->private_data);
         return 0;
 }
 
 static u64 __perf_event_read_value(struct perf_event *event, u64 *enabled, u64 *running)
 {
         struct perf_event *child;
         u64 total = 0;
 
         *enabled = 0;
         *running = 0;
 
         mutex_lock(&event->child_mutex);
 
         (void)perf_event_read(event, false);
         total += perf_event_count(event);
 
         *enabled += event->total_time_enabled +
                         atomic64_read(&event->child_total_time_enabled);
         *running += event->total_time_running +
                         atomic64_read(&event->child_total_time_running);
 
         list_for_each_entry(child, &event->child_list, child_list) {
                 (void)perf_event_read(child, false);
                 total += perf_event_count(child);
                 *enabled += child->total_time_enabled;
                 *running += child->total_time_running;
         }
         mutex_unlock(&event->child_mutex);
 
         return total;
 }
 
 u64 perf_event_read_value(struct perf_event *event, u64 *enabled, u64 *running)
 {
         struct perf_event_context *ctx;
         u64 count;
 
         ctx = perf_event_ctx_lock(event);
         count = __perf_event_read_value(event, enabled, running);
         perf_event_ctx_unlock(event, ctx);
 
         return count;
 }
 EXPORT_SYMBOL_GPL(perf_event_read_value);
 
 static int __perf_read_group_add(struct perf_event *leader,
                                         u64 read_format, u64 *values)
 {
         struct perf_event_context *ctx = leader->ctx;
         struct perf_event *sub;
         unsigned long flags;
         int n = 1; /* skip @nr */
         int ret;
 
         ret = perf_event_read(leader, true);
         if (ret)
                 return ret;
 
         raw_spin_lock_irqsave(&ctx->lock, flags);
 
         /*
          * Since we co-schedule groups, {enabled,running} times of siblings
          * will be identical to those of the leader, so we only publish one
          * set.
          */
         if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED) {
                 values[n++] += leader->total_time_enabled +
                         atomic64_read(&leader->child_total_time_enabled);
         }
 
         if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING) {
                 values[n++] += leader->total_time_running +
                         atomic64_read(&leader->child_total_time_running);
         }
 
         /*
          * Write {count,id} tuples for every sibling.
          */
         values[n++] += perf_event_count(leader);
         if (read_format & PERF_FORMAT_ID)
                 values[n++] = primary_event_id(leader);
 
         for_each_sibling_event(sub, leader) {
                 values[n++] += perf_event_count(sub);
                 if (read_format & PERF_FORMAT_ID)
                         values[n++] = primary_event_id(sub);
         }
 
         raw_spin_unlock_irqrestore(&ctx->lock, flags);
         return 0;
 }
 
 static int perf_read_group(struct perf_event *event,
                                    u64 read_format, char __user *buf)
 {
         struct perf_event *leader = event->group_leader, *child;
         struct perf_event_context *ctx = leader->ctx;
         int ret;
         u64 *values;
 
         lockdep_assert_held(&ctx->mutex);
 
         values = kzalloc(event->read_size, GFP_KERNEL);
         if (!values)
                 return -ENOMEM;
 
         values[0] = 1 + leader->nr_siblings;
 
         /*
          * By locking the child_mutex of the leader we effectively
          * lock the child list of all siblings.. XXX explain how.
          */
         mutex_lock(&leader->child_mutex);
 
         ret = __perf_read_group_add(leader, read_format, values);
         if (ret)
                 goto unlock;
 
         list_for_each_entry(child, &leader->child_list, child_list) {
                 ret = __perf_read_group_add(child, read_format, values);
                 if (ret)
                         goto unlock;
         }
 
         mutex_unlock(&leader->child_mutex);
 
         ret = event->read_size;
         if (copy_to_user(buf, values, event->read_size))
                 ret = -EFAULT;
         goto out;
 
 unlock:
         mutex_unlock(&leader->child_mutex);
 out:
         kfree(values);
         return ret;
 }
 
 static int perf_read_one(struct perf_event *event,
                                  u64 read_format, char __user *buf)
 {
         u64 enabled, running;
         u64 values[4];
         int n = 0;
 
         values[n++] = __perf_event_read_value(event, &enabled, &running);
         if (read_format & PERF_FORMAT_TOTAL_TIME_ENABLED)
                 values[n++] = enabled;
         if (read_format & PERF_FORMAT_TOTAL_TIME_RUNNING)
                 values[n++] = running;
         if (read_format & PERF_FORMAT_ID)
                 values[n++] = primary_event_id(event);
 
         if (copy_to_user(buf, values, n * sizeof(u64)))
                 return -EFAULT;
 
         return n * sizeof(u64);
 }
 
 static bool is_event_hup(struct perf_event *event)
 {
         bool no_children;
 
         if (event->state > PERF_EVENT_STATE_EXIT)
                 return false;
 
         mutex_lock(&event->child_mutex);
         no_children = list_empty(&event->child_list);
         mutex_unlock(&event->child_mutex);
         return no_children;
 }
 
 /*
  * Read the performance event - simple non blocking version for now
  */
 static ssize_t
 __perf_read(struct perf_event *event, char __user *buf, size_t count)
 {
         u64 read_format = event->attr.read_format;
         int ret;
 
         /*
          * Return end-of-file for a read on an event that is in
          * error state (i.e. because it was pinned but it couldn't be
          * scheduled on to the CPU at some point).
          */
         if (event->state == PERF_EVENT_STATE_ERROR)
                 return 0;
 
         if (count < event->read_size)
                 return -ENOSPC;
 
         WARN_ON_ONCE(event->ctx->parent_ctx);
         if (read_format & PERF_FORMAT_GROUP)
                 ret = perf_read_group(event, read_format, buf);
         else
                 ret = perf_read_one(event, read_format, buf);
 
         return ret;
 }
 
 static ssize_t
 perf_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 {
         struct perf_event *event = file->private_data;
         struct perf_event_context *ctx;
         int ret;
 
         ret = security_perf_event_read(event);
         if (ret)
                 return ret;
 
         ctx = perf_event_ctx_lock(event);
         ret = __perf_read(event, buf, count);
         perf_event_ctx_unlock(event, ctx);
 
         return ret;
 }
 
 static __poll_t perf_poll(struct file *file, poll_table *wait)
 {
         struct perf_event *event = file->private_data;
         struct perf_buffer *rb;
         __poll_t events = EPOLLHUP;
 
         poll_wait(file, &event->waitq, wait);
 
         if (is_event_hup(event))
                 return events;
 
         /*
          * Pin the event->rb by taking event->mmap_mutex; otherwise
          * perf_event_set_output() can swizzle our rb and make us miss wakeups.
          */
         mutex_lock(&event->mmap_mutex);
         rb = event->rb;
         if (rb)
                 events = atomic_xchg(&rb->poll, 0);
         mutex_unlock(&event->mmap_mutex);
         return events;
 }
 
 static void _perf_event_reset(struct perf_event *event)
 {
         (void)perf_event_read(event, false);
         local64_set(&event->count, 0);
         perf_event_update_userpage(event);
 }
 
 /* Assume it's not an event with inherit set. */
 u64 perf_event_pause(struct perf_event *event, bool reset)
 {
         struct perf_event_context *ctx;
         u64 count;
 
         ctx = perf_event_ctx_lock(event);
         WARN_ON_ONCE(event->attr.inherit);
         _perf_event_disable(event);
         count = local64_read(&event->count);
         if (reset)
                 local64_set(&event->count, 0);
         perf_event_ctx_unlock(event, ctx);
 
         return count;
 }
 EXPORT_SYMBOL_GPL(perf_event_pause);
 
 /*
  * Holding the top-level event's child_mutex means that any
  * descendant process that has inherited this event will block
  * in perf_event_exit_event() if it goes to exit, thus satisfying the
  * task existence requirements of perf_event_enable/disable.
  */
 static void perf_event_for_each_child(struct perf_event *event,
                                         void (*func)(struct perf_event *))
 {
         struct perf_event *child;
 
         WARN_ON_ONCE(event->ctx->parent_ctx);
 
         mutex_lock(&event->child_mutex);
         func(event);
         list_for_each_entry(child, &event->child_list, child_list)
                 func(child);
         mutex_unlock(&event->child_mutex);
 }
 
 static void perf_event_for_each(struct perf_event *event,
                                   void (*func)(struct perf_event *))
 {
         struct perf_event_context *ctx = event->ctx;
         struct perf_event *sibling;
 
         lockdep_assert_held(&ctx->mutex);
 
         event = event->group_leader;
 
         perf_event_for_each_child(event, func);
         for_each_sibling_event(sibling, event)
                 perf_event_for_each_child(sibling, func);
 }
 
 static void __perf_event_period(struct perf_event *event,
                                 struct perf_cpu_context *cpuctx,
                                 struct perf_event_context *ctx,
                                 void *info)
 {
         u64 value = *((u64 *)info);
         bool active;
 
         if (event->attr.freq) {
                 event->attr.sample_freq = value;
         } else {
                 event->attr.sample_period = value;
                 event->hw.sample_period = value;
         }
 
         active = (event->state == PERF_EVENT_STATE_ACTIVE);
         if (active) {
                 perf_pmu_disable(ctx->pmu);
                 /*
                  * We could be throttled; unthrottle now to avoid the tick
                  * trying to unthrottle while we already re-started the event.
                  */
                 if (event->hw.interrupts == MAX_INTERRUPTS) {
                         event->hw.interrupts = 0;
                         perf_log_throttle(event, 1);
                 }
                 event->pmu->stop(event, PERF_EF_UPDATE);
         }
 
         local64_set(&event->hw.period_left, 0);
 
         if (active) {
                 event->pmu->start(event, PERF_EF_RELOAD);
                 perf_pmu_enable(ctx->pmu);
         }
 }
 
 static int perf_event_check_period(struct perf_event *event, u64 value)
 {
         return event->pmu->check_period(event, value);
 }
 
 static int _perf_event_period(struct perf_event *event, u64 value)
 {
         if (!is_sampling_event(event))
                 return -EINVAL;
 
         if (!value)
                 return -EINVAL;
 
         if (event->attr.freq && value > sysctl_perf_event_sample_rate)
                 return -EINVAL;
 
         if (perf_event_check_period(event, value))
                 return -EINVAL;
 
         if (!event->attr.freq && (value & (1ULL << 63)))
                 return -EINVAL;
 
         event_function_call(event, __perf_event_period, &value);
 
         return 0;
 }
 
 int perf_event_period(struct perf_event *event, u64 value)
 {
         struct perf_event_context *ctx;
         int ret;
 
         ctx = perf_event_ctx_lock(event);
         ret = _perf_event_period(event, value);
         perf_event_ctx_unlock(event, ctx);
 
         return ret;
 }
 EXPORT_SYMBOL_GPL(perf_event_period);
 
 static const struct file_operations perf_fops;
 
 static inline int perf_fget_light(int fd, struct fd *p)
 {
         struct fd f = fdget(fd);
         if (!f.file)
                 return -EBADF;
 
         if (f.file->f_op != &perf_fops) {
                 fdput(f);
                 return -EBADF;
         }
         *p = f;
         return 0;
 }
 
 static int perf_event_set_output(struct perf_event *event,
                                  struct perf_event *output_event);
 static int perf_event_set_filter(struct perf_event *event, void __user *arg);
 static int perf_copy_attr(struct perf_event_attr __user *uattr,
                           struct perf_event_attr *attr);
 
 static long _perf_ioctl(struct perf_event *event, unsigned int cmd, unsigned long arg)
 {
         void (*func)(struct perf_event *);
         u32 flags = arg;
 
         switch (cmd) {
         case PERF_EVENT_IOC_ENABLE:
                 func = _perf_event_enable;
                 break;
         case PERF_EVENT_IOC_DISABLE:
                 func = _perf_event_disable;
                 break;
         case PERF_EVENT_IOC_RESET:
                 func = _perf_event_reset;
                 break;
 
         case PERF_EVENT_IOC_REFRESH:
                 return _perf_event_refresh(event, arg);
 
         case PERF_EVENT_IOC_PERIOD:
         {
                 u64 value;
 
                 if (copy_from_user(&value, (u64 __user *)arg, sizeof(value)))
                         return -EFAULT;
 
                 return _perf_event_period(event, value);
         }
         case PERF_EVENT_IOC_ID:
         {
                 u64 id = primary_event_id(event);
 
                 if (copy_to_user((void __user *)arg, &id, sizeof(id)))
                         return -EFAULT;
                 return 0;
         }
 
         case PERF_EVENT_IOC_SET_OUTPUT:
         {
                 int ret;
                 if (arg != -1) {
                         struct perf_event *output_event;
                         struct fd output;
                         ret = perf_fget_light(arg, &output);
                         if (ret)
                                 return ret;
                         output_event = output.file->private_data;
                         ret = perf_event_set_output(event, output_event);
                         fdput(output);
                 } else {
                         ret = perf_event_set_output(event, NULL);
                 }
                 return ret;
         }
 
         case PERF_EVENT_IOC_SET_FILTER:
                 return perf_event_set_filter(event, (void __user *)arg);
 
         case PERF_EVENT_IOC_SET_BPF:
         {
                 struct bpf_prog *prog;
                 int err;
 
                 prog = bpf_prog_get(arg);
                 if (IS_ERR(prog))
                         return PTR_ERR(prog);
 
                 err = perf_event_set_bpf_prog(event, prog, 0);
                 if (err) {
                         bpf_prog_put(prog);
                         return err;
                 }
 
                 return 0;
         }
 
         case PERF_EVENT_IOC_PAUSE_OUTPUT: {
                 struct perf_buffer *rb;
 
                 rcu_read_lock();
                 rb = rcu_dereference(event->rb);
                 if (!rb || !rb->nr_pages) {
                         rcu_read_unlock();
                         return -EINVAL;
                 }
                 rb_toggle_paused(rb, !!arg);
                 rcu_read_unlock();
                 return 0;
         }
 
         case PERF_EVENT_IOC_QUERY_BPF:
                 return perf_event_query_prog_array(event, (void __user *)arg);
 
         case PERF_EVENT_IOC_MODIFY_ATTRIBUTES: {
                 struct perf_event_attr new_attr;
                 int err = perf_copy_attr((struct perf_event_attr __user *)arg,
                                          &new_attr);
 
                 if (err)
                         return err;
 
                 return perf_event_modify_attr(event,  &new_attr);
         }
         default:
                 return -ENOTTY;
         }
 
         if (flags & PERF_IOC_FLAG_GROUP)
                 perf_event_for_each(event, func);
         else
                 perf_event_for_each_child(event, func);
 
         return 0;
 }
 
 static long perf_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
         struct perf_event *event = file->private_data;
         struct perf_event_context *ctx;
         long ret;
 
         /* Treat ioctl like writes as it is likely a mutating operation. */
         ret = security_perf_event_write(event);
         if (ret)
                 return ret;
 
         ctx = perf_event_ctx_lock(event);
         ret = _perf_ioctl(event, cmd, arg);
         perf_event_ctx_unlock(event, ctx);
 
         return ret;
 }
 
 #ifdef CONFIG_COMPAT
 static long perf_compat_ioctl(struct file *file, unsigned int cmd,
                                 unsigned long arg)
 {
         switch (_IOC_NR(cmd)) {
         case _IOC_NR(PERF_EVENT_IOC_SET_FILTER):
         case _IOC_NR(PERF_EVENT_IOC_ID):
         case _IOC_NR(PERF_EVENT_IOC_QUERY_BPF):
         case _IOC_NR(PERF_EVENT_IOC_MODIFY_ATTRIBUTES):
                 /* Fix up pointer size (usually 4 -> 8 in 32-on-64-bit case */
                 if (_IOC_SIZE(cmd) == sizeof(compat_uptr_t)) {
                         cmd &= ~IOCSIZE_MASK;
                         cmd |= sizeof(void *) << IOCSIZE_SHIFT;
                 }
                 break;
         }
         return perf_ioctl(file, cmd, arg);
 }
 #else
 # define perf_compat_ioctl NULL
 #endif
 
 int perf_event_task_enable(void)
 {
         struct perf_event_context *ctx;
         struct perf_event *event;
 
         mutex_lock(&current->perf_event_mutex);
         list_for_each_entry(event, &current->perf_event_list, owner_entry) {
                 ctx = perf_event_ctx_lock(event);
                 perf_event_for_each_child(event, _perf_event_enable);
                 perf_event_ctx_unlock(event, ctx);
         }
         mutex_unlock(&current->perf_event_mutex);
 
         return 0;
 }
 
 int perf_event_task_disable(void)
 {
         struct perf_event_context *ctx;
         struct perf_event *event;
 
         mutex_lock(&current->perf_event_mutex);
         list_for_each_entry(event, &current->perf_event_list, owner_entry) {
                 ctx = perf_event_ctx_lock(event);
                 perf_event_for_each_child(event, _perf_event_disable);
                 perf_event_ctx_unlock(event, ctx);
         }
         mutex_unlock(&current->perf_event_mutex);
 
         return 0;
 }
 
 static int perf_event_index(struct perf_event *event)
 {
         if (event->hw.state & PERF_HES_STOPPED)
                 return 0;
 
         if (event->state != PERF_EVENT_STATE_ACTIVE)
                 return 0;
 
         return event->pmu->event_idx(event);
 }
 
 static void perf_event_init_userpage(struct perf_event *event)
 {
         struct perf_event_mmap_page *userpg;
         struct perf_buffer *rb;
 
         rcu_read_lock();
         rb = rcu_dereference(event->rb);
         if (!rb)
                 goto unlock;
 
         userpg = rb->user_page;
 
         /* Allow new userspace to detect that bit 0 is deprecated */
         userpg->cap_bit0_is_deprecated = 1;
         userpg->size = offsetof(struct perf_event_mmap_page, __reserved);
         userpg->data_offset = PAGE_SIZE;
         userpg->data_size = perf_data_size(rb);
 
 unlock:
         rcu_read_unlock();
 }
 
 void __weak arch_perf_update_userpage(
         struct perf_event *event, struct perf_event_mmap_page *userpg, u64 now)
 {
 }
 
 /*
  * Callers need to ensure there can be no nesting of this function, otherwise
  * the seqlock logic goes bad. We can not serialize this because the arch
  * code calls this from NMI context.
  */
 void perf_event_update_userpage(struct perf_event *event)
 {
         struct perf_event_mmap_page *userpg;
         struct perf_buffer *rb;
         u64 enabled, running, now;
 
         rcu_read_lock();
         rb = rcu_dereference(event->rb);
         if (!rb)
                 goto unlock;
 
         /*
          * compute total_time_enabled, total_time_running
          * based on snapshot values taken when the event
          * was last scheduled in.
          *
          * we cannot simply called update_context_time()
          * because of locking issue as we can be called in
          * NMI context
          */
         calc_timer_values(event, &now, &enabled, &running);
 
         userpg = rb->user_page;
         /*
          * Disable preemption to guarantee consistent time stamps are stored to
          * the user page.
          */
         preempt_disable();
         ++userpg->lock;
         barrier();
         userpg->index = perf_event_index(event);
         userpg->offset = perf_event_count(event);
         if (userpg->index)
                 userpg->offset -= local64_read(&event->hw.prev_count);
 
         userpg->time_enabled = enabled +
                         atomic64_read(&event->child_total_time_enabled);
 
         userpg->time_running = running +
                         atomic64_read(&event->child_total_time_running);
 
         arch_perf_update_userpage(event, userpg, now);
 
         barrier();
         ++userpg->lock;
         preempt_enable();
 unlock:
         rcu_read_unlock();
 }
 EXPORT_SYMBOL_GPL(perf_event_update_userpage);
 
 static vm_fault_t perf_mmap_fault(struct vm_fault *vmf)
 {
         struct perf_event *event = vmf->vma->vm_file->private_data;
         struct perf_buffer *rb;
         vm_fault_t ret = VM_FAULT_SIGBUS;
 
         if (vmf->flags & FAULT_FLAG_MKWRITE) {
                 if (vmf->pgoff == 0)
                         ret = 0;
                 return ret;
         }
 
         rcu_read_lock();
         rb = rcu_dereference(event->rb);
         if (!rb)
                 goto unlock;
 
5882<