[PATCH 2/3] utrace_do_stop: cleanup the usage of ->siglock

[Oleg Nesterov]

The unconditional spin_lock(->siglock) looks a bit ugly. We can
microoptimize the code and take it only when neede. Imho, also
makes the code a little bit more understandable.

Signed-off-by: Oleg Nesterov <o...@redhat.com>

--- __UTRACE/kernel/utrace.c~6_DO_STOP_SIGLOCK  2009-08-26 16:53:49.000000000 
+0200
+++ __UTRACE/kernel/utrace.c    2009-08-26 18:03:36.000000000 +0200
@@ -660,7 +660,6 @@ static bool utrace_do_stop(struct task_s
 {
        bool stopped = false;
 
-       spin_lock_irq(&target->sighand->siglock);
        if (unlikely(target->exit_state)) {
                /*
                 * On the exit path, it's only truly quiescent
@@ -674,13 +673,16 @@ static bool utrace_do_stop(struct task_s
                 * Stopped is considered quiescent; when it wakes up, it will
                 * go through utrace_finish_jctl() before doing anything else.
                 */
-               __set_task_state(target, TASK_TRACED);
-               utrace->stopped = stopped = true;
+               spin_lock_irq(&target->sighand->siglock);
+               if (likely(task_is_stopped(target))) {
+                       __set_task_state(target, TASK_TRACED);
+                       utrace->stopped = stopped = true;
+               }
+               spin_unlock_irq(&target->sighand->siglock);
        } else if (!utrace->report && !utrace->interrupt) {
                utrace->report = 1;
                set_notify_resume(target);
        }
-       spin_unlock_irq(&target->sighand->siglock);
 
        return stopped;
 }