Merge branch 'master'

4 files changed, 70 insertions, 74 deletions

diff --git a/kernel/exit.c b/kernel/exit.c
index 43077732619..4897977a1f4 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -825,14 +825,6 @@ fastcall NORET_TYPE void do_exit(long code)
 
 	tsk->flags |= PF_EXITING;
 
-	/*
-	 * Make sure we don't try to process any timer firings
-	 * while we are already exiting.
-	 */
- 	tsk->it_virt_expires = cputime_zero;
- 	tsk->it_prof_expires = cputime_zero;
-	tsk->it_sched_expires = 0;
-
 	if (unlikely(in_atomic()))
 		printk(KERN_INFO "note: %s[%d] exited with preempt_count %d\n",
 				current->comm, current->pid,
@@ -843,6 +835,7 @@ fastcall NORET_TYPE void do_exit(long code)
 	group_dead = atomic_dec_and_test(&tsk->signal->live);
 	if (group_dead) {
  		del_timer_sync(&tsk->signal->real_timer);
+		exit_itimers(tsk->signal);
 		acct_process(code);
 	}
 	exit_mm(tsk);
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c
index b3f3edc475d..b15462b17a5 100644
--- a/kernel/posix-cpu-timers.c
+++ b/kernel/posix-cpu-timers.c
@@ -380,28 +380,31 @@ int posix_cpu_timer_create(struct k_itimer *new_timer)
 int posix_cpu_timer_del(struct k_itimer *timer)
 {
 	struct task_struct *p = timer->it.cpu.task;
+	int ret = 0;
 
-	if (timer->it.cpu.firing)
-		return TIMER_RETRY;
-
-	if (unlikely(p == NULL))
-		return 0;
+	if (likely(p != NULL)) {
+		read_lock(&tasklist_lock);
+		if (unlikely(p->signal == NULL)) {
+			/*
+			 * We raced with the reaping of the task.
+			 * The deletion should have cleared us off the list.
+			 */
+			BUG_ON(!list_empty(&timer->it.cpu.entry));
+		} else {
+			spin_lock(&p->sighand->siglock);
+			if (timer->it.cpu.firing)
+				ret = TIMER_RETRY;
+			else
+				list_del(&timer->it.cpu.entry);
+			spin_unlock(&p->sighand->siglock);
+		}
+		read_unlock(&tasklist_lock);
 
-	spin_lock(&p->sighand->siglock);
-	if (!list_empty(&timer->it.cpu.entry)) {
-		/*
-		 * Take us off the task's timer list.  We don't need to
-		 * take tasklist_lock and check for the task being reaped.
-		 * If it was reaped, it already called posix_cpu_timers_exit
-		 * and posix_cpu_timers_exit_group to clear all the timers
-		 * that pointed to it.
-		 */
-		list_del(&timer->it.cpu.entry);
-		put_task_struct(p);
+		if (!ret)
+			put_task_struct(p);
 	}
-	spin_unlock(&p->sighand->siglock);
 
-	return 0;
+	return ret;
 }
 
 /*
@@ -418,8 +421,6 @@ static void cleanup_timers(struct list_head *head,
 	cputime_t ptime = cputime_add(utime, stime);
 
 	list_for_each_entry_safe(timer, next, head, entry) {
-		put_task_struct(timer->task);
-		timer->task = NULL;
 		list_del_init(&timer->entry);
 		if (cputime_lt(timer->expires.cpu, ptime)) {
 			timer->expires.cpu = cputime_zero;
@@ -431,8 +432,6 @@ static void cleanup_timers(struct list_head *head,
 
 	++head;
 	list_for_each_entry_safe(timer, next, head, entry) {
-		put_task_struct(timer->task);
-		timer->task = NULL;
 		list_del_init(&timer->entry);
 		if (cputime_lt(timer->expires.cpu, utime)) {
 			timer->expires.cpu = cputime_zero;
@@ -444,8 +443,6 @@ static void cleanup_timers(struct list_head *head,
 
 	++head;
 	list_for_each_entry_safe(timer, next, head, entry) {
-		put_task_struct(timer->task);
-		timer->task = NULL;
 		list_del_init(&timer->entry);
 		if (timer->expires.sched < sched_time) {
 			timer->expires.sched = 0;
@@ -489,6 +486,9 @@ static void process_timer_rebalance(struct task_struct *p,
  	struct task_struct *t = p;
 	unsigned int nthreads = atomic_read(&p->signal->live);
 
+	if (!nthreads)
+		return;
+
 	switch (clock_idx) {
 	default:
 		BUG();
@@ -730,9 +730,15 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags,
 	 * Disarm any old timer after extracting its expiry time.
 	 */
 	BUG_ON(!irqs_disabled());
+
+	ret = 0;
 	spin_lock(&p->sighand->siglock);
 	old_expires = timer->it.cpu.expires;
-	list_del_init(&timer->it.cpu.entry);
+	if (unlikely(timer->it.cpu.firing)) {
+		timer->it.cpu.firing = -1;
+		ret = TIMER_RETRY;
+	} else
+		list_del_init(&timer->it.cpu.entry);
 	spin_unlock(&p->sighand->siglock);
 
 	/*
@@ -780,7 +786,7 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags,
 		}
 	}
 
-	if (unlikely(timer->it.cpu.firing)) {
+	if (unlikely(ret)) {
 		/*
 		 * We are colliding with the timer actually firing.
 		 * Punt after filling in the timer's old value, and
@@ -788,8 +794,6 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags,
 		 * it as an overrun (thanks to bump_cpu_timer above).
 		 */
 		read_unlock(&tasklist_lock);
-		timer->it.cpu.firing = -1;
-		ret = TIMER_RETRY;
 		goto out;
 	}
 
@@ -955,14 +959,16 @@ void posix_cpu_timer_get(struct k_itimer *timer, struct itimerspec *itp)
 static void check_thread_timers(struct task_struct *tsk,
 				struct list_head *firing)
 {
+	int maxfire;
 	struct list_head *timers = tsk->cpu_timers;
 
+	maxfire = 20;
 	tsk->it_prof_expires = cputime_zero;
 	while (!list_empty(timers)) {
 		struct cpu_timer_list *t = list_entry(timers->next,
 						      struct cpu_timer_list,
 						      entry);
-		if (cputime_lt(prof_ticks(tsk), t->expires.cpu)) {
+		if (!--maxfire || cputime_lt(prof_ticks(tsk), t->expires.cpu)) {
 			tsk->it_prof_expires = t->expires.cpu;
 			break;
 		}
@@ -971,12 +977,13 @@ static void check_thread_timers(struct task_struct *tsk,
 	}
 
 	++timers;
+	maxfire = 20;
 	tsk->it_virt_expires = cputime_zero;
 	while (!list_empty(timers)) {
 		struct cpu_timer_list *t = list_entry(timers->next,
 						      struct cpu_timer_list,
 						      entry);
-		if (cputime_lt(virt_ticks(tsk), t->expires.cpu)) {
+		if (!--maxfire || cputime_lt(virt_ticks(tsk), t->expires.cpu)) {
 			tsk->it_virt_expires = t->expires.cpu;
 			break;
 		}
@@ -985,12 +992,13 @@ static void check_thread_timers(struct task_struct *tsk,
 	}
 
 	++timers;
+	maxfire = 20;
 	tsk->it_sched_expires = 0;
 	while (!list_empty(timers)) {
 		struct cpu_timer_list *t = list_entry(timers->next,
 						      struct cpu_timer_list,
 						      entry);
-		if (tsk->sched_time < t->expires.sched) {
+		if (!--maxfire || tsk->sched_time < t->expires.sched) {
 			tsk->it_sched_expires = t->expires.sched;
 			break;
 		}
@@ -1007,6 +1015,7 @@ static void check_thread_timers(struct task_struct *tsk,
 static void check_process_timers(struct task_struct *tsk,
 				 struct list_head *firing)
 {
+	int maxfire;
 	struct signal_struct *const sig = tsk->signal;
 	cputime_t utime, stime, ptime, virt_expires, prof_expires;
 	unsigned long long sched_time, sched_expires;
@@ -1039,12 +1048,13 @@ static void check_process_timers(struct task_struct *tsk,
 	} while (t != tsk);
 	ptime = cputime_add(utime, stime);
 
+	maxfire = 20;
 	prof_expires = cputime_zero;
 	while (!list_empty(timers)) {
 		struct cpu_timer_list *t = list_entry(timers->next,
 						      struct cpu_timer_list,
 						      entry);
-		if (cputime_lt(ptime, t->expires.cpu)) {
+		if (!--maxfire || cputime_lt(ptime, t->expires.cpu)) {
 			prof_expires = t->expires.cpu;
 			break;
 		}
@@ -1053,12 +1063,13 @@ static void check_process_timers(struct task_struct *tsk,
 	}
 
 	++timers;
+	maxfire = 20;
 	virt_expires = cputime_zero;
 	while (!list_empty(timers)) {
 		struct cpu_timer_list *t = list_entry(timers->next,
 						      struct cpu_timer_list,
 						      entry);
-		if (cputime_lt(utime, t->expires.cpu)) {
+		if (!--maxfire || cputime_lt(utime, t->expires.cpu)) {
 			virt_expires = t->expires.cpu;
 			break;
 		}
@@ -1067,12 +1078,13 @@ static void check_process_timers(struct task_struct *tsk,
 	}
 
 	++timers;
+	maxfire = 20;
 	sched_expires = 0;
 	while (!list_empty(timers)) {
 		struct cpu_timer_list *t = list_entry(timers->next,
 						      struct cpu_timer_list,
 						      entry);
-		if (sched_time < t->expires.sched) {
+		if (!--maxfire || sched_time < t->expires.sched) {
 			sched_expires = t->expires.sched;
 			break;
 		}
@@ -1155,6 +1167,9 @@ static void check_process_timers(struct task_struct *tsk,
 		unsigned long long sched_left, sched;
 		const unsigned int nthreads = atomic_read(&sig->live);
 
+		if (!nthreads)
+			return;
+
 		prof_left = cputime_sub(prof_expires, utime);
 		prof_left = cputime_sub(prof_left, stime);
 		prof_left = cputime_div(prof_left, nthreads);
@@ -1280,30 +1295,30 @@ void run_posix_cpu_timers(struct task_struct *tsk)
 
 #undef	UNEXPIRED
 
-	BUG_ON(tsk->exit_state);
-
 	/*
 	 * Double-check with locks held.
 	 */
 	read_lock(&tasklist_lock);
-	spin_lock(&tsk->sighand->siglock);
+	if (likely(tsk->signal != NULL)) {
+		spin_lock(&tsk->sighand->siglock);
 
-	/*
-	 * Here we take off tsk->cpu_timers[N] and tsk->signal->cpu_timers[N]
-	 * all the timers that are firing, and put them on the firing list.
-	 */
-	check_thread_timers(tsk, &firing);
-	check_process_timers(tsk, &firing);
+		/*
+		 * Here we take off tsk->cpu_timers[N] and tsk->signal->cpu_timers[N]
+		 * all the timers that are firing, and put them on the firing list.
+		 */
+		check_thread_timers(tsk, &firing);
+		check_process_timers(tsk, &firing);
 
-	/*
-	 * We must release these locks before taking any timer's lock.
-	 * There is a potential race with timer deletion here, as the
-	 * siglock now protects our private firing list.  We have set
-	 * the firing flag in each timer, so that a deletion attempt
-	 * that gets the timer lock before we do will give it up and
-	 * spin until we've taken care of that timer below.
-	 */
-	spin_unlock(&tsk->sighand->siglock);
+		/*
+		 * We must release these locks before taking any timer's lock.
+		 * There is a potential race with timer deletion here, as the
+		 * siglock now protects our private firing list.  We have set
+		 * the firing flag in each timer, so that a deletion attempt
+		 * that gets the timer lock before we do will give it up and
+		 * spin until we've taken care of that timer below.
+		 */
+		spin_unlock(&tsk->sighand->siglock);
+	}
 	read_unlock(&tasklist_lock);
 
 	/*
diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c
index b7b532acd9f..dda3cda73c7 100644
--- a/kernel/posix-timers.c
+++ b/kernel/posix-timers.c
@@ -1157,7 +1157,7 @@ retry_delete:
 }
 
 /*
- * This is called by __exit_signal, only when there are no more
+ * This is called by do_exit or de_thread, only when there are no more
  * references to the shared signal_struct.
  */
 void exit_itimers(struct signal_struct *sig)
diff --git a/kernel/signal.c b/kernel/signal.c
index 50c99264377..f2b96b08fb4 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -397,20 +397,8 @@ void __exit_signal(struct task_struct *tsk)
 	flush_sigqueue(&tsk->pending);
 	if (sig) {
 		/*
-		 * We are cleaning up the signal_struct here.  We delayed
-		 * calling exit_itimers until after flush_sigqueue, just in
-		 * case our thread-local pending queue contained a queued
-		 * timer signal that would have been cleared in
-		 * exit_itimers.  When that called sigqueue_free, it would
-		 * attempt to re-take the tasklist_lock and deadlock.  This
-		 * can never happen if we ensure that all queues the
-		 * timer's signal might be queued on have been flushed
-		 * first.  The shared_pending queue, and our own pending
-		 * queue are the only queues the timer could be on, since
-		 * there are no other threads left in the group and timer
-		 * signals are constrained to threads inside the group.
+		 * We are cleaning up the signal_struct here.
 		 */
-		exit_itimers(sig);
 		exit_thread_group_keys(sig);
 		kmem_cache_free(signal_cachep, sig);
 	}