diff options
Diffstat (limited to 'kernel')
| -rw-r--r-- | kernel/cgroup_freezer.c | 5 | ||||
| -rw-r--r-- | kernel/perf_event.c | 56 | ||||
| -rw-r--r-- | kernel/sched.c | 18 | ||||
| -rw-r--r-- | kernel/trace/trace_events_filter.c | 2 | ||||
| -rw-r--r-- | kernel/workqueue.c | 2 |
5 files changed, 60 insertions, 23 deletions
diff --git a/kernel/cgroup_freezer.c b/kernel/cgroup_freezer.c index da5e13975531..e5c0244962b0 100644 --- a/kernel/cgroup_freezer.c +++ b/kernel/cgroup_freezer.c @@ -205,9 +205,12 @@ static void freezer_fork(struct cgroup_subsys *ss, struct task_struct *task) * No lock is needed, since the task isn't on tasklist yet, * so it can't be moved to another cgroup, which means the * freezer won't be removed and will be valid during this - * function call. + * function call. Nevertheless, apply RCU read-side critical + * section to suppress RCU lockdep false positives. */ + rcu_read_lock(); freezer = task_freezer(task); + rcu_read_unlock(); /* * The root cgroup is non-freezable, so we can skip the diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 9dbe8cdaf145..bb06382f98e7 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -83,14 +83,6 @@ extern __weak const struct pmu *hw_perf_event_init(struct perf_event *event) void __weak hw_perf_disable(void) { barrier(); } void __weak hw_perf_enable(void) { barrier(); } -int __weak -hw_perf_group_sched_in(struct perf_event *group_leader, - struct perf_cpu_context *cpuctx, - struct perf_event_context *ctx) -{ - return 0; -} - void __weak perf_event_print_debug(void) { } static DEFINE_PER_CPU(int, perf_disable_count); @@ -342,6 +334,9 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx) if (event->state > PERF_EVENT_STATE_OFF) event->state = PERF_EVENT_STATE_OFF; + if (event->state > PERF_EVENT_STATE_FREE) + return; + /* * If this was a group event with sibling events then * upgrade the siblings to singleton events by adding them @@ -641,15 +636,20 @@ group_sched_in(struct perf_event *group_event, struct perf_cpu_context *cpuctx, struct perf_event_context *ctx) { - struct perf_event *event, *partial_group; + struct perf_event *event, *partial_group = NULL; + const struct pmu *pmu = group_event->pmu; + bool txn = false; int ret; if (group_event->state == PERF_EVENT_STATE_OFF) return 0; - ret = hw_perf_group_sched_in(group_event, cpuctx, ctx); - if (ret) - return ret < 0 ? ret : 0; + /* Check if group transaction availabe */ + if (pmu->start_txn) + txn = true; + + if (txn) + pmu->start_txn(pmu); if (event_sched_in(group_event, cpuctx, ctx)) return -EAGAIN; @@ -664,9 +664,19 @@ group_sched_in(struct perf_event *group_event, } } - return 0; + if (txn) { + ret = pmu->commit_txn(pmu); + if (!ret) { + pmu->cancel_txn(pmu); + + return 0; + } + } group_error: + if (txn) + pmu->cancel_txn(pmu); + /* * Groups can be scheduled in as one unit only, so undo any * partial group before returning: @@ -1861,8 +1871,22 @@ int perf_event_release_kernel(struct perf_event *event) { struct perf_event_context *ctx = event->ctx; + event->state = PERF_EVENT_STATE_FREE; + WARN_ON_ONCE(ctx->parent_ctx); - mutex_lock(&ctx->mutex); + /* + * There are two ways this annotation is useful: + * + * 1) there is a lock recursion from perf_event_exit_task + * see the comment there. + * + * 2) there is a lock-inversion with mmap_sem through + * perf_event_read_group(), which takes faults while + * holding ctx->mutex, however this is called after + * the last filedesc died, so there is no possibility + * to trigger the AB-BA case. + */ + mutex_lock_nested(&ctx->mutex, SINGLE_DEPTH_NESTING); perf_event_remove_from_context(event); mutex_unlock(&ctx->mutex); @@ -5021,7 +5045,7 @@ err_fput_free_put_context: err_free_put_context: if (err < 0) - kfree(event); + free_event(event); err_put_context: if (err < 0) @@ -5300,7 +5324,7 @@ void perf_event_exit_task(struct task_struct *child) * * But since its the parent context it won't be the same instance. */ - mutex_lock_nested(&child_ctx->mutex, SINGLE_DEPTH_NESTING); + mutex_lock(&child_ctx->mutex); again: list_for_each_entry_safe(child_event, tmp, &child_ctx->pinned_groups, diff --git a/kernel/sched.c b/kernel/sched.c index b0bbadc24955..b11b80a3eed3 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -323,6 +323,15 @@ static inline struct task_group *task_group(struct task_struct *p) /* Change a task's cfs_rq and parent entity if it moves across CPUs/groups */ static inline void set_task_rq(struct task_struct *p, unsigned int cpu) { + /* + * Strictly speaking this rcu_read_lock() is not needed since the + * task_group is tied to the cgroup, which in turn can never go away + * as long as there are tasks attached to it. + * + * However since task_group() uses task_subsys_state() which is an + * rcu_dereference() user, this quiets CONFIG_PROVE_RCU. + */ + rcu_read_lock(); #ifdef CONFIG_FAIR_GROUP_SCHED p->se.cfs_rq = task_group(p)->cfs_rq[cpu]; p->se.parent = task_group(p)->se[cpu]; @@ -332,6 +341,7 @@ static inline void set_task_rq(struct task_struct *p, unsigned int cpu) p->rt.rt_rq = task_group(p)->rt_rq[cpu]; p->rt.parent = task_group(p)->rt_se[cpu]; #endif + rcu_read_unlock(); } #else @@ -3737,7 +3747,7 @@ int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner) * the mutex owner just released it and exited. */ if (probe_kernel_address(&owner->cpu, cpu)) - goto out; + return 0; #else cpu = owner->cpu; #endif @@ -3747,14 +3757,14 @@ int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner) * the cpu field may no longer be valid. */ if (cpu >= nr_cpumask_bits) - goto out; + return 0; /* * We need to validate that we can do a * get_cpu() and that we have the percpu area. */ if (!cpu_online(cpu)) - goto out; + return 0; rq = cpu_rq(cpu); @@ -3773,7 +3783,7 @@ int mutex_spin_on_owner(struct mutex *lock, struct thread_info *owner) cpu_relax(); } -out: + return 1; } #endif diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 88c0b6dbd7fe..58092d844a1f 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -1398,7 +1398,7 @@ int ftrace_profile_set_filter(struct perf_event *event, int event_id, } err = -EINVAL; - if (!call) + if (&call->list == &ftrace_events) goto out_unlock; err = -EEXIST; diff --git a/kernel/workqueue.c b/kernel/workqueue.c index dee48658805c..5bfb213984b2 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -774,7 +774,7 @@ void flush_delayed_work(struct delayed_work *dwork) { if (del_timer_sync(&dwork->timer)) { struct cpu_workqueue_struct *cwq; - cwq = wq_per_cpu(keventd_wq, get_cpu()); + cwq = wq_per_cpu(get_wq_data(&dwork->work)->wq, get_cpu()); __queue_work(cwq, &dwork->work); put_cpu(); } |