diff -u b/fs/exec.c b/fs/exec.c
--- b/fs/exec.c
+++ b/fs/exec.c
@@ -854,7 +854,6 @@
 	struct signal_struct *sig = tsk->signal;
 	struct sighand_struct *oldsighand = tsk->sighand;
 	spinlock_t *lock = &oldsighand->siglock;
-	int count;
 
 	if (thread_group_empty(tsk))
 		goto no_thread_group;
@@ -871,13 +870,13 @@
 		spin_unlock_irq(lock);
 		return -EAGAIN;
 	}
+
 	sig->group_exit_task = tsk;
-	zap_other_threads(tsk);
+	sig->notify_count = zap_other_threads(tsk);
+	if (!thread_group_leader(tsk))
+		sig->notify_count--;
 
-	/* Account for the thread group leader hanging around: */
-	count = thread_group_leader(tsk) ? 1 : 2;
-	sig->notify_count = count;
-	while (atomic_read(&sig->count) > count) {
+	while (sig->notify_count) {
 		__set_current_state(TASK_UNINTERRUPTIBLE);
 		spin_unlock_irq(lock);
 		schedule();
diff -u b/fs/proc/array.c b/fs/proc/array.c
--- b/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -264,7 +264,7 @@
 		shpending = p->signal->shared_pending.signal;
 		blocked = p->blocked;
 		collect_sigign_sigcatch(p, &ignored, &caught);
-		num_threads = atomic_read(&p->signal->count);
+		num_threads = get_nr_threads(p);
 		qsize = atomic_read(&__task_cred(p)->user->sigpending);
 		qlim = p->signal->rlim[RLIMIT_SIGPENDING].rlim_cur;
 		unlock_task_sighand(p, &flags);
@@ -430,7 +430,7 @@
 			tty_nr = new_encode_dev(tty_devnum(sig->tty));
 		}
 
-		num_threads = atomic_read(&sig->count);
+		num_threads = get_nr_threads(task);
 		collect_sigign_sigcatch(task, &sigign, &sigcatch);
 
 		cmin_flt = sig->cmin_flt;
@@ -600,13 +600,5 @@
 int proc_pid_ipaddr(struct task_struct *task, char *buffer)
 {
-	u32 curr_ip = 0;
-	unsigned long flags;
-
-	if (lock_task_sighand(task, &flags)) {
-		curr_ip = task->signal->curr_ip;
-		unlock_task_sighand(task, &flags);
-	}
-
-	return sprintf(buffer, "%pI4\n", &curr_ip);
+	return sprintf(buffer, "%pI4\n", &task->signal->curr_ip);
 }
 #endif
diff -u b/fs/proc/base.c b/fs/proc/base.c
--- b/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -181,18 +181,6 @@
 	return result;
 }
 
-static int get_nr_threads(struct task_struct *tsk)
-{
-	unsigned long flags;
-	int count = 0;
-
-	if (lock_task_sighand(tsk, &flags)) {
-		count = atomic_read(&tsk->signal->count);
-		unlock_task_sighand(tsk, &flags);
-	}
-	return count;
-}
-
 static int proc_cwd_link(struct inode *inode, struct path *path)
 {
 	struct task_struct *task = get_proc_task(inode);
diff -u b/include/linux/init_task.h b/include/linux/init_task.h
--- b/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -16,7 +16,7 @@
 extern struct fs_struct init_fs;
 
 #define INIT_SIGNALS(sig) {						\
-	.count		= ATOMIC_INIT(1), 				\
+	.nr_threads	= 1,						\
 	.wait_chldexit	= __WAIT_QUEUE_HEAD_INITIALIZER(sig.wait_chldexit),\
 	.shared_pending	= { 						\
 		.list = LIST_HEAD_INIT(sig.shared_pending.list),	\
diff -u b/include/linux/sched.h b/include/linux/sched.h
--- b/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -564,8 +564,9 @@
  * the locking of signal_struct.
  */
 struct signal_struct {
-	atomic_t		count;
+	atomic_t		sigcnt;
 	atomic_t		live;
+	int			nr_threads;
 
 	wait_queue_head_t	wait_chldexit;	/* for wait4() */
 
@@ -2179,7 +2180,7 @@
 extern void force_sig(int, struct task_struct *);
 extern void force_sig_specific(int, struct task_struct *);
 extern int send_sig(int, struct task_struct *, int);
-extern void zap_other_threads(struct task_struct *p);
+extern int zap_other_threads(struct task_struct *p);
 extern struct sigqueue *sigqueue_alloc(void);
 extern void sigqueue_free(struct sigqueue *);
 extern int send_sigqueue(struct sigqueue *,  struct task_struct *, int group);
@@ -2249,7 +2250,6 @@
 extern void exit_thread(void);
 
 extern void exit_files(struct task_struct *);
-extern void __cleanup_signal(struct signal_struct *);
 extern void __cleanup_sighand(struct sighand_struct *);
 
 extern void exit_itimers(struct signal_struct *);
@@ -2298,6 +2298,11 @@
 #define while_each_thread(g, t) \
 	while ((t = next_thread(t)) != g)
 
+static inline int get_nr_threads(struct task_struct *tsk)
+{
+	return tsk->signal->nr_threads;
+}
+
 /* de_thread depends on thread_group_leader not being a pid based check */
 #define thread_group_leader(p)	(p == p->group_leader)
 
diff -u b/kernel/exit.c b/kernel/exit.c
--- b/kernel/exit.c
+++ b/kernel/exit.c
@@ -61,11 +61,11 @@
 
 static void exit_mm(struct task_struct * tsk);
 
-static void __unhash_process(struct task_struct *p)
+static void __unhash_process(struct task_struct *p, bool group_dead)
 {
 	nr_threads--;
 	detach_pid(p, PIDTYPE_PID);
-	if (thread_group_leader(p)) {
+	if (group_dead) {
 		detach_pid(p, PIDTYPE_PGID);
 		detach_pid(p, PIDTYPE_SID);
 
@@ -82,18 +82,19 @@
 static void __exit_signal(struct task_struct *tsk)
 {
 	struct signal_struct *sig = tsk->signal;
+	bool group_dead = thread_group_leader(tsk);
 	struct sighand_struct *sighand;
-
-	BUG_ON(!sig);
-	BUG_ON(!atomic_read(&sig->count));
+	struct tty_struct *uninitialized_var(tty);
 
 	sighand = rcu_dereference(tsk->sighand);
 	spin_lock(&sighand->siglock);
 
 	posix_cpu_timers_exit(tsk);
-	if (atomic_dec_and_test(&sig->count))
+	if (group_dead) {
 		posix_cpu_timers_exit_group(tsk);
-	else {
+		tty = sig->tty;
+		sig->tty = NULL;
+	} else {
 		/*
 		 * This can only happen if the caller is de_thread().
 		 * FIXME: this is the temporary hack, we should teach
@@ -106,7 +107,7 @@
 		 * If there is any task waiting for the group exit
 		 * then notify it:
 		 */
-		if (sig->group_exit_task && atomic_read(&sig->count) == sig->notify_count)
+		if (sig->notify_count > 0 && !--sig->notify_count)
 			wake_up_process(sig->group_exit_task);
 
 		if (tsk == sig->curr_target)
@@ -132,32 +133,24 @@
 		sig->oublock += task_io_get_oublock(tsk);
 		task_io_accounting_add(&sig->ioac, &tsk->ioac);
 		sig->sum_sched_runtime += tsk->se.sum_exec_runtime;
-		sig = NULL; /* Marker for below. */
 	}
 
-	__unhash_process(tsk);
+	sig->nr_threads--;
+	__unhash_process(tsk, group_dead);
 
 	/*
 	 * Do this under ->siglock, we can race with another thread
 	 * doing sigqueue_free() if we have SIGQUEUE_PREALLOC signals.
 	 */
 	flush_sigqueue(&tsk->pending);
-
-	tsk->signal = NULL;
 	tsk->sighand = NULL;
 	spin_unlock(&sighand->siglock);
 
 	__cleanup_sighand(sighand);
 	clear_tsk_thread_flag(tsk,TIF_SIGPENDING);
-	if (sig) {
+	if (group_dead) {
 		flush_sigqueue(&sig->shared_pending);
-		taskstats_tgid_free(sig);
-		/*
-		 * Make sure ->signal can't go away under rq->lock,
-		 * see account_group_exec_runtime().
-		 */
-		task_rq_unlock_wait(tsk);
-		__cleanup_signal(sig);
+		tty_kref_put(tty);
 	}
 }
 
@@ -871,12 +864,9 @@
 
 	tsk->exit_state = signal == DEATH_REAP ? EXIT_DEAD : EXIT_ZOMBIE;
 
-	/* mt-exec, de_thread() is waiting for us */
-	if (thread_group_leader(tsk) &&
-	    tsk->signal->group_exit_task &&
-	    tsk->signal->notify_count < 0)
+	/* mt-exec, de_thread() is waiting for group leader */
+	if (unlikely(tsk->signal->notify_count < 0))
 		wake_up_process(tsk->signal->group_exit_task);
-
 	write_unlock_irq(&tasklist_lock);
 
 	tracehook_report_death(tsk, signal, cookie, group_dead);
diff -u b/kernel/fork.c b/kernel/fork.c
--- b/kernel/fork.c
+++ b/kernel/fork.c
@@ -156,6 +156,18 @@
 }
 EXPORT_SYMBOL(free_task);
 
+static inline void free_signal_struct(struct signal_struct *sig)
+{
+	taskstats_tgid_free(sig);
+	kmem_cache_free(signal_cachep, sig);
+}
+
+static inline void put_signal_struct(struct signal_struct *sig)
+{
+	if (atomic_dec_and_test(&sig->sigcnt))
+		free_signal_struct(sig);
+}
+
 void __put_task_struct(struct task_struct *tsk)
 {
 	WARN_ON(!tsk->exit_state);
@@ -164,6 +176,7 @@
 
 	exit_creds(tsk);
 	delayacct_tsk_free(tsk);
+	put_signal_struct(tsk->signal);
 
 	if (!profile_handoff_task(tsk))
 		free_task(tsk);
@@ -895,8 +908,9 @@
 	if (!sig)
 		return -ENOMEM;
 
-	atomic_set(&sig->count, 1);
+	sig->nr_threads = 1;
 	atomic_set(&sig->live, 1);
+	atomic_set(&sig->sigcnt, 1);
 	init_waitqueue_head(&sig->wait_chldexit);
 	sig->flags = 0;
 	if (clone_flags & CLONE_NEWPID)
@@ -949,9 +963,2 @@
 
-void __cleanup_signal(struct signal_struct *sig)
-{
-	thread_group_cputime_free(sig);
-	tty_kref_put(sig->tty);
-	kmem_cache_free(signal_cachep, sig);
-}
-
 static void copy_flags(unsigned long clone_flags, struct task_struct *p)
@@ -1302,8 +1309,9 @@
 	}
 
 	if (clone_flags & CLONE_THREAD) {
-		atomic_inc(&current->signal->count);
+		current->signal->nr_threads++;
 		atomic_inc(&current->signal->live);
+		atomic_inc(&current->signal->sigcnt);
 		p->group_leader = current->group_leader;
 		list_add_tail_rcu(&p->thread_group, &p->group_leader->thread_group);
 	}
@@ -1317,7 +1325,6 @@
 				p->nsproxy->pid_ns->child_reaper = p;
 
 			p->signal->leader_pid = pid;
-			tty_kref_put(p->signal->tty);
 			p->signal->tty = tty_kref_get(current->signal->tty);
 			attach_pid(p, PIDTYPE_PGID, task_pgrp(current));
 			attach_pid(p, PIDTYPE_SID, task_session(current));
@@ -1349,7 +1356,7 @@
 		mmput(p->mm);
 bad_fork_cleanup_signal:
 	if (!(clone_flags & CLONE_THREAD))
-		__cleanup_signal(p->signal);
+		free_signal_struct(p->signal);
 bad_fork_cleanup_sighand:
 	__cleanup_sighand(p->sighand);
 bad_fork_cleanup_fs:
@@ -1568,14 +1575,6 @@
 		*flags_ptr |= CLONE_SIGHAND;
 
 	/*
-	 * If unsharing signal handlers and the task was created
-	 * using CLONE_THREAD, then must unshare the thread
-	 */
-	if ((*flags_ptr & CLONE_SIGHAND) &&
-	    (atomic_read(&current->signal->count) > 1))
-		*flags_ptr |= CLONE_THREAD;
-
-	/*
 	 * If unsharing namespace, must also unshare filesystem information.
 	 */
 	if (*flags_ptr & CLONE_NEWNS)
diff -u b/kernel/signal.c b/kernel/signal.c
--- b/kernel/signal.c
+++ b/kernel/signal.c
@@ -1054,23 +1054,24 @@
 /*
  * Nuke all other threads in the group.
  */
-void zap_other_threads(struct task_struct *p)
+int zap_other_threads(struct task_struct *p)
 {
-	struct task_struct *t;
+	struct task_struct *t = p;
+	int count = 0;
 
 	p->signal->group_stop_count = 0;
 
-	for (t = next_thread(p); t != p; t = next_thread(t)) {
-		/*
-		 * Don't bother with already dead threads
-		 */
+	while_each_thread(p, t) {
+		count++;
+
+		/* Don't bother with already dead threads */
 		if (t->exit_state)
 			continue;
-
-		/* SIGKILL will be handled before any pending SIGSTOP */
 		sigaddset(&t->pending.signal, SIGKILL);
 		signal_wake_up(t, 1);
 	}
+
+	return count;
 }
 
 struct sighand_struct *lock_task_sighand(struct task_struct *tsk, unsigned long *flags)
only in patch2:
unchanged:
--- a/kernel/sched_debug.c
+++ b/kernel/sched_debug.c
@@ -369,15 +369,9 @@ __initcall(init_sched_debug_procfs);
 void proc_sched_show_task(struct task_struct *p, struct seq_file *m)
 {
 	unsigned long nr_switches;
-	unsigned long flags;
-	int num_threads = 1;
 
-	if (lock_task_sighand(p, &flags)) {
-		num_threads = atomic_read(&p->signal->count);
-		unlock_task_sighand(p, &flags);
-	}
-
-	SEQ_printf(m, "%s (%d, #threads: %d)\n", p->comm, p->pid, num_threads);
+	SEQ_printf(m, "%s (%d, #threads: %d)\n", p->comm, p->pid,
+						get_nr_threads(p));
 	SEQ_printf(m,
 		"---------------------------------------------------------\n");
 #define __P(F) \
only in patch2:
unchanged:
--- a/security/keys/keyctl.c
+++ b/security/keys/keyctl.c
@@ -1270,7 +1270,7 @@ long keyctl_session_to_parent(void)
 		goto not_permitted;
 
 	/* the parent must be single threaded */
-	if (atomic_read(&parent->signal->count) != 1)
+	if (!thread_group_empty(parent))
 		goto not_permitted;
 
 	/* the parent and the child must have different session keyrings or