perf_counter: fix counter freeing logic

Fix counter lifetime bugs which explain the crashes reported by
Marcelo Tosatti and Arnaldo Carvalho de Melo.

The new rule is: flushing + freeing is only done for a task's
own counters, never for other tasks.

[ Impact: fix crashes/lockups with inherited counters ]

Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Reported-by: Marcelo Tosatti <mtosatti@redhat.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

2 files changed, 9 insertions, 12 deletions

diff --git a/kernel/exit.c b/kernel/exit.c
index 73affd35e76d..f9dfedd94af0 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -178,12 +178,6 @@ repeat:
 
         proc_flush_task(p);
 
-        /*
-         * Flush inherited counters to the parent - before the parent
-         * gets woken up by child-exit notifications.
-         */
-        perf_counter_exit_task(p);
-
         write_lock_irq(&tasklist_lock);
         tracehook_finish_release_task(p);
         __exit_signal(p);
@@ -985,6 +979,13 @@ NORET_TYPE void do_exit(long code)
                 module_put(tsk->binfmt->module);
 
         proc_exit_connector(tsk);
+
+        /*
+         * Flush inherited counters to the parent - before the parent
+         * gets woken up by child-exit notifications.
+         */
+        perf_counter_exit_task(tsk);
+
         exit_notify(tsk, group_dead);
 #ifdef CONFIG_NUMA
         mpol_put(tsk->mempolicy);
@@ -1257,12 +1258,6 @@ static int wait_task_zombie(struct task_struct *p, int options,
          */
         read_unlock(&tasklist_lock);
 
-        /*
-         * Flush inherited counters to the parent - before the parent
-         * gets woken up by child-exit notifications.
-         */
-        perf_counter_exit_task(p);
-
         retval = ru ? getrusage(p, RUSAGE_BOTH, ru) : 0;
         status = (p->signal->flags & SIGNAL_GROUP_EXIT)
                 ? p->signal->group_exit_code : p->exit_code;
diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index 59a926d04baf..7af16d1c480f 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -3299,6 +3299,8 @@ void perf_counter_exit_task(struct task_struct *child)
         struct perf_counter *child_counter, *tmp;
         struct perf_counter_context *child_ctx;
 
+        WARN_ON_ONCE(child != current);
+
         child_ctx = &child->perf_counter_ctx;
 
         if (likely(!child_ctx->nr_counters))