Merge branch 'linus' into xen-64bit

10 files changed, 442 insertions, 292 deletions

diff --git a/kernel/Makefile b/kernel/Makefile
index 0a7ed838984b..985ddb7da4d0 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -11,8 +11,6 @@ obj-y     = sched.o fork.o exec_domain.o panic.o printk.o profile.o \
 	    hrtimer.o rwsem.o nsproxy.o srcu.o semaphore.o \
 	    notifier.o ksysfs.o pm_qos_params.o sched_clock.o
 
-CFLAGS_REMOVE_sched.o = -mno-spe
-
 ifdef CONFIG_FTRACE
 # Do not trace debug files and internal ftrace files
 CFLAGS_REMOVE_lockdep.o = -pg
@@ -21,6 +19,7 @@ CFLAGS_REMOVE_mutex-debug.o = -pg
 CFLAGS_REMOVE_rtmutex-debug.o = -pg
 CFLAGS_REMOVE_cgroup-debug.o = -pg
 CFLAGS_REMOVE_sched_clock.o = -pg
+CFLAGS_REMOVE_sched.o = -mno-spe -pg
 endif
 
 obj-$(CONFIG_SYSCTL_SYSCALL_CHECK) += sysctl_check.o
diff --git a/kernel/cpu.c b/kernel/cpu.c
index b11f06dc149a..cfb1d43ab801 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -299,6 +299,7 @@ int __ref cpu_down(unsigned int cpu)
 	cpu_maps_update_done();
 	return err;
 }
+EXPORT_SYMBOL(cpu_down);
 #endif /*CONFIG_HOTPLUG_CPU*/
 
 /* Requires cpu_add_remove_lock to be held */
diff --git a/kernel/exit.c b/kernel/exit.c
index ceb258782835..93d2711b9381 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -71,7 +71,7 @@ static void __unhash_process(struct task_struct *p)
 		__get_cpu_var(process_counts)--;
 	}
 	list_del_rcu(&p->thread_group);
-	remove_parent(p);
+	list_del_init(&p->sibling);
 }
 
 /*
@@ -152,6 +152,18 @@ static void delayed_put_task_struct(struct rcu_head *rhp)
 	put_task_struct(container_of(rhp, struct task_struct, rcu));
 }
 
+/*
+ * Do final ptrace-related cleanup of a zombie being reaped.
+ *
+ * Called with write_lock(&tasklist_lock) held.
+ */
+static void ptrace_release_task(struct task_struct *p)
+{
+	BUG_ON(!list_empty(&p->ptraced));
+	ptrace_unlink(p);
+	BUG_ON(!list_empty(&p->ptrace_entry));
+}
+
 void release_task(struct task_struct * p)
 {
 	struct task_struct *leader;
@@ -160,8 +172,7 @@ repeat:
 	atomic_dec(&p->user->processes);
 	proc_flush_task(p);
 	write_lock_irq(&tasklist_lock);
-	ptrace_unlink(p);
-	BUG_ON(!list_empty(&p->ptrace_list) || !list_empty(&p->ptrace_children));
+	ptrace_release_task(p);
 	__exit_signal(p);
 
 	/*
@@ -315,9 +326,8 @@ static void reparent_to_kthreadd(void)
 
 	ptrace_unlink(current);
 	/* Reparent to init */
-	remove_parent(current);
 	current->real_parent = current->parent = kthreadd_task;
-	add_parent(current);
+	list_move_tail(&current->sibling, &current->real_parent->children);
 
 	/* Set the exit signal to SIGCHLD so we signal init on exit */
 	current->exit_signal = SIGCHLD;
@@ -692,37 +702,97 @@ static void exit_mm(struct task_struct * tsk)
 	mmput(mm);
 }
 
-static void
-reparent_thread(struct task_struct *p, struct task_struct *father, int traced)
+/*
+ * Return nonzero if @parent's children should reap themselves.
+ *
+ * Called with write_lock_irq(&tasklist_lock) held.
+ */
+static int ignoring_children(struct task_struct *parent)
 {
-	if (p->pdeath_signal)
-		/* We already hold the tasklist_lock here.  */
-		group_send_sig_info(p->pdeath_signal, SEND_SIG_NOINFO, p);
+	int ret;
+	struct sighand_struct *psig = parent->sighand;
+	unsigned long flags;
+	spin_lock_irqsave(&psig->siglock, flags);
+	ret = (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN ||
+	       (psig->action[SIGCHLD-1].sa.sa_flags & SA_NOCLDWAIT));
+	spin_unlock_irqrestore(&psig->siglock, flags);
+	return ret;
+}
 
-	/* Move the child from its dying parent to the new one.  */
-	if (unlikely(traced)) {
-		/* Preserve ptrace links if someone else is tracing this child.  */
-		list_del_init(&p->ptrace_list);
-		if (ptrace_reparented(p))
-			list_add(&p->ptrace_list, &p->real_parent->ptrace_children);
-	} else {
-		/* If this child is being traced, then we're the one tracing it
-		 * anyway, so let go of it.
+/*
+ * Detach all tasks we were using ptrace on.
+ * Any that need to be release_task'd are put on the @dead list.
+ *
+ * Called with write_lock(&tasklist_lock) held.
+ */
+static void ptrace_exit(struct task_struct *parent, struct list_head *dead)
+{
+	struct task_struct *p, *n;
+	int ign = -1;
+
+	list_for_each_entry_safe(p, n, &parent->ptraced, ptrace_entry) {
+		__ptrace_unlink(p);
+
+		if (p->exit_state != EXIT_ZOMBIE)
+			continue;
+
+		/*
+		 * If it's a zombie, our attachedness prevented normal
+		 * parent notification or self-reaping.  Do notification
+		 * now if it would have happened earlier.  If it should
+		 * reap itself, add it to the @dead list.  We can't call
+		 * release_task() here because we already hold tasklist_lock.
+		 *
+		 * If it's our own child, there is no notification to do.
+		 * But if our normal children self-reap, then this child
+		 * was prevented by ptrace and we must reap it now.
 		 */
-		p->ptrace = 0;
-		remove_parent(p);
-		p->parent = p->real_parent;
-		add_parent(p);
+		if (!task_detached(p) && thread_group_empty(p)) {
+			if (!same_thread_group(p->real_parent, parent))
+				do_notify_parent(p, p->exit_signal);
+			else {
+				if (ign < 0)
+					ign = ignoring_children(parent);
+				if (ign)
+					p->exit_signal = -1;
+			}
+		}
 
-		if (task_is_traced(p)) {
+		if (task_detached(p)) {
 			/*
-			 * If it was at a trace stop, turn it into
-			 * a normal stop since it's no longer being
-			 * traced.
+			 * Mark it as in the process of being reaped.
 			 */
-			ptrace_untrace(p);
+			p->exit_state = EXIT_DEAD;
+			list_add(&p->ptrace_entry, dead);
 		}
 	}
+}
+
+/*
+ * Finish up exit-time ptrace cleanup.
+ *
+ * Called without locks.
+ */
+static void ptrace_exit_finish(struct task_struct *parent,
+			       struct list_head *dead)
+{
+	struct task_struct *p, *n;
+
+	BUG_ON(!list_empty(&parent->ptraced));
+
+	list_for_each_entry_safe(p, n, dead, ptrace_entry) {
+		list_del_init(&p->ptrace_entry);
+		release_task(p);
+	}
+}
+
+static void reparent_thread(struct task_struct *p, struct task_struct *father)
+{
+	if (p->pdeath_signal)
+		/* We already hold the tasklist_lock here.  */
+		group_send_sig_info(p->pdeath_signal, SEND_SIG_NOINFO, p);
+
+	list_move_tail(&p->sibling, &p->real_parent->children);
 
 	/* If this is a threaded reparent there is no need to
 	 * notify anyone anything has happened.
@@ -737,7 +807,8 @@ reparent_thread(struct task_struct *p, struct task_struct *father, int traced)
 	/* If we'd notified the old parent about this child's death,
 	 * also notify the new parent.
 	 */
-	if (!traced && p->exit_state == EXIT_ZOMBIE &&
+	if (!ptrace_reparented(p) &&
+	    p->exit_state == EXIT_ZOMBIE &&
 	    !task_detached(p) && thread_group_empty(p))
 		do_notify_parent(p, p->exit_signal);
 
@@ -754,12 +825,15 @@ reparent_thread(struct task_struct *p, struct task_struct *father, int traced)
 static void forget_original_parent(struct task_struct *father)
 {
 	struct task_struct *p, *n, *reaper = father;
-	struct list_head ptrace_dead;
-
-	INIT_LIST_HEAD(&ptrace_dead);
+	LIST_HEAD(ptrace_dead);
 
 	write_lock_irq(&tasklist_lock);
 
+	/*
+	 * First clean up ptrace if we were using it.
+	 */
+	ptrace_exit(father, &ptrace_dead);
+
 	do {
 		reaper = next_thread(reaper);
 		if (reaper == father) {
@@ -768,58 +842,19 @@ static void forget_original_parent(struct task_struct *father)
 		}
 	} while (reaper->flags & PF_EXITING);
 
-	/*
-	 * There are only two places where our children can be:
-	 *
-	 * - in our child list
-	 * - in our ptraced child list
-	 *
-	 * Search them and reparent children.
-	 */
 	list_for_each_entry_safe(p, n, &father->children, sibling) {
-		int ptrace;
-
-		ptrace = p->ptrace;
-
-		/* if father isn't the real parent, then ptrace must be enabled */
-		BUG_ON(father != p->real_parent && !ptrace);
-
-		if (father == p->real_parent) {
-			/* reparent with a reaper, real father it's us */
-			p->real_parent = reaper;
-			reparent_thread(p, father, 0);
-		} else {
-			/* reparent ptraced task to its real parent */
-			__ptrace_unlink (p);
-			if (p->exit_state == EXIT_ZOMBIE && !task_detached(p) &&
-			    thread_group_empty(p))
-				do_notify_parent(p, p->exit_signal);
-		}
-
-		/*
-		 * if the ptraced child is a detached zombie we must collect
-		 * it before we exit, or it will remain zombie forever since
-		 * we prevented it from self-reap itself while it was being
-		 * traced by us, to be able to see it in wait4.
-		 */
-		if (unlikely(ptrace && p->exit_state == EXIT_ZOMBIE && task_detached(p)))
-			list_add(&p->ptrace_list, &ptrace_dead);
-	}
-
-	list_for_each_entry_safe(p, n, &father->ptrace_children, ptrace_list) {
 		p->real_parent = reaper;
-		reparent_thread(p, father, 1);
+		if (p->parent == father) {
+			BUG_ON(p->ptrace);
+			p->parent = p->real_parent;
+		}
+		reparent_thread(p, father);
 	}
 
 	write_unlock_irq(&tasklist_lock);
 	BUG_ON(!list_empty(&father->children));
-	BUG_ON(!list_empty(&father->ptrace_children));
-
-	list_for_each_entry_safe(p, n, &ptrace_dead, ptrace_list) {
-		list_del_init(&p->ptrace_list);
-		release_task(p);
-	}
 
+	ptrace_exit_finish(father, &ptrace_dead);
 }
 
 /*
@@ -1180,13 +1215,6 @@ static int eligible_child(enum pid_type type, struct pid *pid, int options,
 			return 0;
 	}
 
-	/*
-	 * Do not consider detached threads that are
-	 * not ptraced:
-	 */
-	if (task_detached(p) && !p->ptrace)
-		return 0;
-
 	/* Wait for all children (clone and not) if __WALL is set;
 	 * otherwise, wait for clone children *only* if __WCLONE is
 	 * set; otherwise, wait for non-clone children *only*.  (Note:
@@ -1197,14 +1225,10 @@ static int eligible_child(enum pid_type type, struct pid *pid, int options,
 		return 0;
 
 	err = security_task_wait(p);
-	if (likely(!err))
-		return 1;
+	if (err)
+		return err;
 
-	if (type != PIDTYPE_PID)
-		return 0;
-	/* This child was explicitly requested, abort */
-	read_unlock(&tasklist_lock);
-	return err;
+	return 1;
 }
 
 static int wait_noreap_copyout(struct task_struct *p, pid_t pid, uid_t uid,
@@ -1238,7 +1262,7 @@ static int wait_noreap_copyout(struct task_struct *p, pid_t pid, uid_t uid,
  * the lock and this task is uninteresting.  If we return nonzero, we have
  * released the lock and the system call should return.
  */
-static int wait_task_zombie(struct task_struct *p, int noreap,
+static int wait_task_zombie(struct task_struct *p, int options,
 			    struct siginfo __user *infop,
 			    int __user *stat_addr, struct rusage __user *ru)
 {
@@ -1246,7 +1270,10 @@ static int wait_task_zombie(struct task_struct *p, int noreap,
 	int retval, status, traced;
 	pid_t pid = task_pid_vnr(p);
 
-	if (unlikely(noreap)) {
+	if (!likely(options & WEXITED))
+		return 0;
+
+	if (unlikely(options & WNOWAIT)) {
 		uid_t uid = p->uid;
 		int exit_code = p->exit_code;
 		int why, status;
@@ -1396,21 +1423,24 @@ static int wait_task_zombie(struct task_struct *p, int noreap,
  * the lock and this task is uninteresting.  If we return nonzero, we have
  * released the lock and the system call should return.
  */
-static int wait_task_stopped(struct task_struct *p,
-			     int noreap, struct siginfo __user *infop,
+static int wait_task_stopped(int ptrace, struct task_struct *p,
+			     int options, struct siginfo __user *infop,
 			     int __user *stat_addr, struct rusage __user *ru)
 {
 	int retval, exit_code, why;
 	uid_t uid = 0; /* unneeded, required by compiler */
 	pid_t pid;
 
+	if (!(options & WUNTRACED))
+		return 0;
+
 	exit_code = 0;
 	spin_lock_irq(&p->sighand->siglock);
 
 	if (unlikely(!task_is_stopped_or_traced(p)))
 		goto unlock_sig;
 
-	if (!(p->ptrace & PT_PTRACED) && p->signal->group_stop_count > 0)
+	if (!ptrace && p->signal->group_stop_count > 0)
 		/*
 		 * A group stop is in progress and this is the group leader.
 		 * We won't report until all threads have stopped.
@@ -1421,7 +1451,7 @@ static int wait_task_stopped(struct task_struct *p,
 	if (!exit_code)
 		goto unlock_sig;
 
-	if (!noreap)
+	if (!unlikely(options & WNOWAIT))
 		p->exit_code = 0;
 
 	uid = p->uid;
@@ -1439,10 +1469,10 @@ unlock_sig:
 	 */
 	get_task_struct(p);
 	pid = task_pid_vnr(p);
-	why = (p->ptrace & PT_PTRACED) ? CLD_TRAPPED : CLD_STOPPED;
+	why = ptrace ? CLD_TRAPPED : CLD_STOPPED;
 	read_unlock(&tasklist_lock);
 
-	if (unlikely(noreap))
+	if (unlikely(options & WNOWAIT))
 		return wait_noreap_copyout(p, pid, uid,
 					   why, exit_code,
 					   infop, ru);
@@ -1476,7 +1506,7 @@ unlock_sig:
  * the lock and this task is uninteresting.  If we return nonzero, we have
  * released the lock and the system call should return.
  */
-static int wait_task_continued(struct task_struct *p, int noreap,
+static int wait_task_continued(struct task_struct *p, int options,
 			       struct siginfo __user *infop,
 			       int __user *stat_addr, struct rusage __user *ru)
 {
@@ -1484,6 +1514,9 @@ static int wait_task_continued(struct task_struct *p, int noreap,
 	pid_t pid;
 	uid_t uid;
 
+	if (!unlikely(options & WCONTINUED))
+		return 0;
+
 	if (!(p->signal->flags & SIGNAL_STOP_CONTINUED))
 		return 0;
 
@@ -1493,7 +1526,7 @@ static int wait_task_continued(struct task_struct *p, int noreap,
 		spin_unlock_irq(&p->sighand->siglock);
 		return 0;
 	}
-	if (!noreap)
+	if (!unlikely(options & WNOWAIT))
 		p->signal->flags &= ~SIGNAL_STOP_CONTINUED;
 	spin_unlock_irq(&p->sighand->siglock);
 
@@ -1519,89 +1552,161 @@ static int wait_task_continued(struct task_struct *p, int noreap,
 	return retval;
 }
 
+/*
+ * Consider @p for a wait by @parent.
+ *
+ * -ECHILD should be in *@notask_error before the first call.
+ * Returns nonzero for a final return, when we have unlocked tasklist_lock.
+ * Returns zero if the search for a child should continue;
+ * then *@notask_error is 0 if @p is an eligible child,
+ * or another error from security_task_wait(), or still -ECHILD.
+ */
+static int wait_consider_task(struct task_struct *parent, int ptrace,
+			      struct task_struct *p, int *notask_error,
+			      enum pid_type type, struct pid *pid, int options,
+			      struct siginfo __user *infop,
+			      int __user *stat_addr, struct rusage __user *ru)
+{
+	int ret = eligible_child(type, pid, options, p);
+	if (!ret)
+		return ret;
+
+	if (unlikely(ret < 0)) {
+		/*
+		 * If we have not yet seen any eligible child,
+		 * then let this error code replace -ECHILD.
+		 * A permission error will give the user a clue
+		 * to look for security policy problems, rather
+		 * than for mysterious wait bugs.
+		 */
+		if (*notask_error)
+			*notask_error = ret;
+	}
+
+	if (likely(!ptrace) && unlikely(p->ptrace)) {
+		/*
+		 * This child is hidden by ptrace.
+		 * We aren't allowed to see it now, but eventually we will.
+		 */
+		*notask_error = 0;
+		return 0;
+	}
+
+	if (p->exit_state == EXIT_DEAD)
+		return 0;
+
+	/*
+	 * We don't reap group leaders with subthreads.
+	 */
+	if (p->exit_state == EXIT_ZOMBIE && !delay_group_leader(p))
+		return wait_task_zombie(p, options, infop, stat_addr, ru);
+
+	/*
+	 * It's stopped or running now, so it might
+	 * later continue, exit, or stop again.
+	 */
+	*notask_error = 0;
+
+	if (task_is_stopped_or_traced(p))
+		return wait_task_stopped(ptrace, p, options,
+					 infop, stat_addr, ru);
+
+	return wait_task_continued(p, options, infop, stat_addr, ru);
+}
+
+/*
+ * Do the work of do_wait() for one thread in the group, @tsk.
+ *
+ * -ECHILD should be in *@notask_error before the first call.
+ * Returns nonzero for a final return, when we have unlocked tasklist_lock.
+ * Returns zero if the search for a child should continue; then
+ * *@notask_error is 0 if there were any eligible children,
+ * or another error from security_task_wait(), or still -ECHILD.
+ */
+static int do_wait_thread(struct task_struct *tsk, int *notask_error,
+			  enum pid_type type, struct pid *pid, int options,
+			  struct siginfo __user *infop, int __user *stat_addr,
+			  struct rusage __user *ru)
+{
+	struct task_struct *p;
+
+	list_for_each_entry(p, &tsk->children, sibling) {
+		/*
+		 * Do not consider detached threads.
+		 */
+		if (!task_detached(p)) {
+			int ret = wait_consider_task(tsk, 0, p, notask_error,
+						     type, pid, options,
+						     infop, stat_addr, ru);
+			if (ret)
+				return ret;
+		}
+	}
+
+	return 0;
+}
+
+static int ptrace_do_wait(struct task_struct *tsk, int *notask_error,
+			  enum pid_type type, struct pid *pid, int options,
+			  struct siginfo __user *infop, int __user *stat_addr,
+			  struct rusage __user *ru)
+{
+	struct task_struct *p;
+
+	/*
+	 * Traditionally we see ptrace'd stopped tasks regardless of options.
+	 */
+	options |= WUNTRACED;
+
+	list_for_each_entry(p, &tsk->ptraced, ptrace_entry) {
+		int ret = wait_consider_task(tsk, 1, p, notask_error,
+					     type, pid, options,
+					     infop, stat_addr, ru);
+		if (ret)
+			return ret;
+	}
+
+	return 0;
+}
+
 static long do_wait(enum pid_type type, struct pid *pid, int options,
 		    struct siginfo __user *infop, int __user *stat_addr,
 		    struct rusage __user *ru)
 {
 	DECLARE_WAITQUEUE(wait, current);
 	struct task_struct *tsk;
-	int flag, retval;
+	int retval;
 
 	add_wait_queue(&current->signal->wait_chldexit,&wait);
 repeat:
-	/* If there is nothing that can match our critier just get out */
+	/*
+	 * If there is nothing that can match our critiera just get out.
+	 * We will clear @retval to zero if we see any child that might later
+	 * match our criteria, even if we are not able to reap it yet.
+	 */
 	retval = -ECHILD;
 	if ((type < PIDTYPE_MAX) && (!pid || hlist_empty(&pid->tasks[type])))
 		goto end;
 
-	/*
-	 * We will set this flag if we see any child that might later
-	 * match our criteria, even if we are not able to reap it yet.
-	 */
-	flag = retval = 0;
 	current->state = TASK_INTERRUPTIBLE;
 	read_lock(&tasklist_lock);
 	tsk = current;
 	do {
-		struct task_struct *p;
-
-		list_for_each_entry(p, &tsk->children, sibling) {
-			int ret = eligible_child(type, pid, options, p);
-			if (!ret)
-				continue;
-
-			if (unlikely(ret < 0)) {
-				retval = ret;
-			} else if (task_is_stopped_or_traced(p)) {
-				/*
-				 * It's stopped now, so it might later
-				 * continue, exit, or stop again.
-				 */
-				flag = 1;
-				if (!(p->ptrace & PT_PTRACED) &&
-				    !(options & WUNTRACED))
-					continue;
-
-				retval = wait_task_stopped(p,
-						(options & WNOWAIT), infop,
-						stat_addr, ru);
-			} else if (p->exit_state == EXIT_ZOMBIE &&
-					!delay_group_leader(p)) {
-				/*
-				 * We don't reap group leaders with subthreads.
-				 */
-				if (!likely(options & WEXITED))
-					continue;
-				retval = wait_task_zombie(p,
-						(options & WNOWAIT), infop,
-						stat_addr, ru);
-			} else if (p->exit_state != EXIT_DEAD) {
-				/*
-				 * It's running now, so it might later
-				 * exit, stop, or stop and then continue.
-				 */
-				flag = 1;
-				if (!unlikely(options & WCONTINUED))
-					continue;
-				retval = wait_task_continued(p,
-						(options & WNOWAIT), infop,
-						stat_addr, ru);
-			}
-			if (retval != 0) /* tasklist_lock released */
-				goto end;
-		}
-		if (!flag) {
-			list_for_each_entry(p, &tsk->ptrace_children,
-								ptrace_list) {
-				flag = eligible_child(type, pid, options, p);
-				if (!flag)
-					continue;
-				if (likely(flag > 0))
-					break;
-				retval = flag;
-				goto end;
-			}
+		int tsk_result = do_wait_thread(tsk, &retval,
+						type, pid, options,
+						infop, stat_addr, ru);
+		if (!tsk_result)
+			tsk_result = ptrace_do_wait(tsk, &retval,
+						    type, pid, options,
+						    infop, stat_addr, ru);
+		if (tsk_result) {
+			/*
+			 * tasklist_lock is unlocked and we have a final result.
+			 */
+			retval = tsk_result;
+			goto end;
 		}
+
 		if (options & __WNOTHREAD)
 			break;
 		tsk = next_thread(tsk);
@@ -1609,16 +1714,14 @@ repeat:
 	} while (tsk != current);
 	read_unlock(&tasklist_lock);
 
-	if (flag) {
-		if (options & WNOHANG)
-			goto end;
+	if (!retval && !(options & WNOHANG)) {
 		retval = -ERESTARTSYS;
-		if (signal_pending(current))
-			goto end;
-		schedule();
-		goto repeat;
+		if (!signal_pending(current)) {
+			schedule();
+			goto repeat;
+		}
 	}
-	retval = -ECHILD;
+
 end:
 	current->state = TASK_RUNNING;
 	remove_wait_queue(&current->signal->wait_chldexit,&wait);
diff --git a/kernel/fork.c b/kernel/fork.c
index 4bd2f516401f..adefc1131f27 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1125,8 +1125,8 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	 */
 	p->group_leader = p;
 	INIT_LIST_HEAD(&p->thread_group);
-	INIT_LIST_HEAD(&p->ptrace_children);
-	INIT_LIST_HEAD(&p->ptrace_list);
+	INIT_LIST_HEAD(&p->ptrace_entry);
+	INIT_LIST_HEAD(&p->ptraced);
 
 	/* Now that the task is set up, run cgroup callbacks if
 	 * necessary. We need to run them before the task is visible
@@ -1198,7 +1198,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	}
 
 	if (likely(p->pid)) {
-		add_parent(p);
+		list_add_tail(&p->sibling, &p->real_parent->children);
 		if (unlikely(p->ptrace & PT_PTRACED))
 			__ptrace_link(p, current->parent);
 
diff --git a/kernel/kthread.c b/kernel/kthread.c
index 97747cdd37c9..ac3fb7326641 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -235,7 +235,7 @@ int kthreadd(void *unused)
 	set_user_nice(tsk, KTHREAD_NICE_LEVEL);
 	set_cpus_allowed(tsk, CPU_MASK_ALL);
 
-	current->flags |= PF_NOFREEZE;
+	current->flags |= PF_NOFREEZE | PF_FREEZER_NOSIG;
 
 	for (;;) {
 		set_current_state(TASK_INTERRUPTIBLE);
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index 14a656cdc652..f011e0870b52 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -180,6 +180,17 @@ static void platform_restore_cleanup(int platform_mode)
 }
 
 /**
+ *	platform_recover - recover the platform from a failure to suspend
+ *	devices.
+ */
+
+static void platform_recover(int platform_mode)
+{
+	if (platform_mode && hibernation_ops && hibernation_ops->recover)
+		hibernation_ops->recover();
+}
+
+/**
  *	create_image - freeze devices that need to be frozen with interrupts
  *	off, create the hibernation image and thaw those devices.  Control
  *	reappears in this routine after a restore.
@@ -193,6 +204,7 @@ static int create_image(int platform_mode)
 	if (error)
 		return error;
 
+	device_pm_lock();
 	local_irq_disable();
 	/* At this point, device_suspend() has been called, but *not*
 	 * device_power_down(). We *must* call device_power_down() now.
@@ -224,9 +236,11 @@ static int create_image(int platform_mode)
 	/* NOTE:  device_power_up() is just a resume() for devices
 	 * that suspended with irqs off ... no overall powerup.
 	 */
-	device_power_up();
+	device_power_up(in_suspend ?
+		(error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE);
  Enable_irqs:
 	local_irq_enable();
+	device_pm_unlock();
 	return error;
 }
 
@@ -255,10 +269,10 @@ int hibernation_snapshot(int platform_mode)
 	suspend_console();
 	error = device_suspend(PMSG_FREEZE);
 	if (error)
-		goto Resume_console;
+		goto Recover_platform;
 
 	if (hibernation_test(TEST_DEVICES))
-		goto Resume_devices;
+		goto Recover_platform;
 
 	error = platform_pre_snapshot(platform_mode);
 	if (error || hibernation_test(TEST_PLATFORM))
@@ -280,12 +294,16 @@ int hibernation_snapshot(int platform_mode)
  Finish:
 	platform_finish(platform_mode);
  Resume_devices:
-	device_resume();
- Resume_console:
+	device_resume(in_suspend ?
+		(error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE);
 	resume_console();
  Close:
 	platform_end(platform_mode);
 	return error;
+
+ Recover_platform:
+	platform_recover(platform_mode);
+	goto Resume_devices;
 }
 
 /**
@@ -300,8 +318,9 @@ static int resume_target_kernel(void)
 {
 	int error;
 
+	device_pm_lock();
 	local_irq_disable();
-	error = device_power_down(PMSG_PRETHAW);
+	error = device_power_down(PMSG_QUIESCE);
 	if (error) {
 		printk(KERN_ERR "PM: Some devices failed to power down, "
 			"aborting resume\n");
@@ -329,9 +348,10 @@ static int resume_target_kernel(void)
 	swsusp_free();
 	restore_processor_state();
 	touch_softlockup_watchdog();
-	device_power_up();
+	device_power_up(PMSG_RECOVER);
  Enable_irqs:
 	local_irq_enable();
+	device_pm_unlock();
 	return error;
 }
 
@@ -350,7 +370,7 @@ int hibernation_restore(int platform_mode)
 
 	pm_prepare_console();
 	suspend_console();
-	error = device_suspend(PMSG_PRETHAW);
+	error = device_suspend(PMSG_QUIESCE);
 	if (error)
 		goto Finish;
 
@@ -362,7 +382,7 @@ int hibernation_restore(int platform_mode)
 		enable_nonboot_cpus();
 	}
 	platform_restore_cleanup(platform_mode);
-	device_resume();
+	device_resume(PMSG_RECOVER);
  Finish:
 	resume_console();
 	pm_restore_console();
@@ -392,8 +412,11 @@ int hibernation_platform_enter(void)
 
 	suspend_console();
 	error = device_suspend(PMSG_HIBERNATE);
-	if (error)
-		goto Resume_console;
+	if (error) {
+		if (hibernation_ops->recover)
+			hibernation_ops->recover();
+		goto Resume_devices;
+	}
 
 	error = hibernation_ops->prepare();
 	if (error)
@@ -403,6 +426,7 @@ int hibernation_platform_enter(void)
 	if (error)
 		goto Finish;
 
+	device_pm_lock();
 	local_irq_disable();
 	error = device_power_down(PMSG_HIBERNATE);
 	if (!error) {
@@ -411,6 +435,7 @@ int hibernation_platform_enter(void)
 		while (1);
 	}
 	local_irq_enable();
+	device_pm_unlock();
 
 	/*
 	 * We don't need to reenable the nonboot CPUs or resume consoles, since
@@ -419,8 +444,7 @@ int hibernation_platform_enter(void)
  Finish:
 	hibernation_ops->finish();
  Resume_devices:
-	device_resume();
- Resume_console:
+	device_resume(PMSG_RESTORE);
 	resume_console();
  Close:
 	hibernation_ops->end();
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 6a6d5eb3524e..3398f4651aa1 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -228,6 +228,7 @@ static int suspend_enter(suspend_state_t state)
 {
 	int error = 0;
 
+	device_pm_lock();
 	arch_suspend_disable_irqs();
 	BUG_ON(!irqs_disabled());
 
@@ -239,10 +240,11 @@ static int suspend_enter(suspend_state_t state)
 	if (!suspend_test(TEST_CORE))
 		error = suspend_ops->enter(state);
 
-	device_power_up();
+	device_power_up(PMSG_RESUME);
  Done:
 	arch_suspend_enable_irqs();
 	BUG_ON(irqs_disabled());
+	device_pm_unlock();
 	return error;
 }
 
@@ -267,11 +269,11 @@ int suspend_devices_and_enter(suspend_state_t state)
 	error = device_suspend(PMSG_SUSPEND);
 	if (error) {
 		printk(KERN_ERR "PM: Some devices failed to suspend\n");
-		goto Resume_console;
+		goto Recover_platform;
 	}
 
 	if (suspend_test(TEST_DEVICES))
-		goto Resume_devices;
+		goto Recover_platform;
 
 	if (suspend_ops->prepare) {
 		error = suspend_ops->prepare();
@@ -291,13 +293,17 @@ int suspend_devices_and_enter(suspend_state_t state)
 	if (suspend_ops->finish)
 		suspend_ops->finish();
  Resume_devices:
-	device_resume();
- Resume_console:
+	device_resume(PMSG_RESUME);
 	resume_console();
  Close:
 	if (suspend_ops->end)
 		suspend_ops->end();
 	return error;
+
+ Recover_platform:
+	if (suspend_ops->recover)
+		suspend_ops->recover();
+	goto Resume_devices;
 }
 
 /**
diff --git a/kernel/power/process.c b/kernel/power/process.c
index f1d0b345c9ba..5fb87652f214 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -19,9 +19,6 @@
  */
 #define TIMEOUT	(20 * HZ)
 
-#define FREEZER_KERNEL_THREADS 0
-#define FREEZER_USER_SPACE 1
-
 static inline int freezeable(struct task_struct * p)
 {
 	if ((p == current) ||
@@ -84,63 +81,53 @@ static void fake_signal_wake_up(struct task_struct *p)
 	spin_unlock_irqrestore(&p->sighand->siglock, flags);
 }
 
-static int has_mm(struct task_struct *p)
+static inline bool should_send_signal(struct task_struct *p)
 {
-	return (p->mm && !(p->flags & PF_BORROWED_MM));
+	return !(p->flags & PF_FREEZER_NOSIG);
 }
 
 /**
  *	freeze_task - send a freeze request to given task
  *	@p: task to send the request to
- *	@with_mm_only: if set, the request will only be sent if the task has its
- *		own mm
- *	Return value: 0, if @with_mm_only is set and the task has no mm of its
- *		own or the task is frozen, 1, otherwise
+ *	@sig_only: if set, the request will only be sent if the task has the
+ *		PF_FREEZER_NOSIG flag unset
+ *	Return value: 'false', if @sig_only is set and the task has
+ *		PF_FREEZER_NOSIG set or the task is frozen, 'true', otherwise
  *
- *	The freeze request is sent by seting the tasks's TIF_FREEZE flag and
+ *	The freeze request is sent by setting the tasks's TIF_FREEZE flag and
  *	either sending a fake signal to it or waking it up, depending on whether
- *	or not it has its own mm (ie. it is a user land task).  If @with_mm_only
- *	is set and the task has no mm of its own (ie. it is a kernel thread),
- *	its TIF_FREEZE flag should not be set.
- *
- *	The task_lock() is necessary to prevent races with exit_mm() or
- *	use_mm()/unuse_mm() from occuring.
+ *	or not it has PF_FREEZER_NOSIG set.  If @sig_only is set and the task
+ *	has PF_FREEZER_NOSIG set (ie. it is a typical kernel thread), its
+ *	TIF_FREEZE flag will not be set.
  */
-static int freeze_task(struct task_struct *p, int with_mm_only)
+static bool freeze_task(struct task_struct *p, bool sig_only)
 {
-	int ret = 1;
+	/*
+	 * We first check if the task is freezing and next if it has already
+	 * been frozen to avoid the race with frozen_process() which first marks
+	 * the task as frozen and next clears its TIF_FREEZE.
+	 */
+	if (!freezing(p)) {
+		rmb();
+		if (frozen(p))
+			return false;
 
-	task_lock(p);
-	if (freezing(p)) {
-		if (has_mm(p)) {
-			if (!signal_pending(p))
-				fake_signal_wake_up(p);
-		} else {
-			if (with_mm_only)
-				ret = 0;
-			else
-				wake_up_state(p, TASK_INTERRUPTIBLE);
-		}
+		if (!sig_only || should_send_signal(p))
+			set_freeze_flag(p);
+		else
+			return false;
+	}
+
+	if (should_send_signal(p)) {
+		if (!signal_pending(p))
+			fake_signal_wake_up(p);
+	} else if (sig_only) {
+		return false;
 	} else {
-		rmb();
-		if (frozen(p)) {
-			ret = 0;
-		} else {
-			if (has_mm(p)) {
-				set_freeze_flag(p);
-				fake_signal_wake_up(p);
-			} else {
-				if (with_mm_only) {
-					ret = 0;
-				} else {
-					set_freeze_flag(p);
-					wake_up_state(p, TASK_INTERRUPTIBLE);
-				}
-			}
-		}
+		wake_up_state(p, TASK_INTERRUPTIBLE);
 	}
-	task_unlock(p);
-	return ret;
+
+	return true;
 }
 
 static void cancel_freezing(struct task_struct *p)
@@ -156,7 +143,7 @@ static void cancel_freezing(struct task_struct *p)
 	}
 }
 
-static int try_to_freeze_tasks(int freeze_user_space)
+static int try_to_freeze_tasks(bool sig_only)
 {
 	struct task_struct *g, *p;
 	unsigned long end_time;
@@ -175,7 +162,7 @@ static int try_to_freeze_tasks(int freeze_user_space)
 			if (frozen(p) || !freezeable(p))
 				continue;
 
-			if (!freeze_task(p, freeze_user_space))
+			if (!freeze_task(p, sig_only))
 				continue;
 
 			/*
@@ -235,13 +222,13 @@ int freeze_processes(void)
 	int error;
 
 	printk("Freezing user space processes ... ");
-	error = try_to_freeze_tasks(FREEZER_USER_SPACE);
+	error = try_to_freeze_tasks(true);
 	if (error)
 		goto Exit;
 	printk("done.\n");
 
 	printk("Freezing remaining freezable tasks ... ");
-	error = try_to_freeze_tasks(FREEZER_KERNEL_THREADS);
+	error = try_to_freeze_tasks(false);
 	if (error)
 		goto Exit;
 	printk("done.");
@@ -251,7 +238,7 @@ int freeze_processes(void)
 	return error;
 }
 
-static void thaw_tasks(int thaw_user_space)
+static void thaw_tasks(bool nosig_only)
 {
 	struct task_struct *g, *p;
 
@@ -260,7 +247,7 @@ static void thaw_tasks(int thaw_user_space)
 		if (!freezeable(p))
 			continue;
 
-		if (!p->mm == thaw_user_space)
+		if (nosig_only && should_send_signal(p))
 			continue;
 
 		thaw_process(p);
@@ -271,8 +258,8 @@ static void thaw_tasks(int thaw_user_space)
 void thaw_processes(void)
 {
 	printk("Restarting tasks ... ");
-	thaw_tasks(FREEZER_KERNEL_THREADS);
-	thaw_tasks(FREEZER_USER_SPACE);
+	thaw_tasks(true);
+	thaw_tasks(false);
 	schedule();
 	printk("done.\n");
 }
diff --git a/kernel/power/user.c b/kernel/power/user.c
index f5512cb3aa86..a6332a313262 100644
--- a/kernel/power/user.c
+++ b/kernel/power/user.c
@@ -23,6 +23,7 @@
 #include <linux/console.h>
 #include <linux/cpu.h>
 #include <linux/freezer.h>
+#include <linux/smp_lock.h>
 
 #include <asm/uaccess.h>
 
@@ -69,16 +70,22 @@ static int snapshot_open(struct inode *inode, struct file *filp)
 	struct snapshot_data *data;
 	int error;
 
-	if (!atomic_add_unless(&snapshot_device_available, -1, 0))
-		return -EBUSY;
+	mutex_lock(&pm_mutex);
+
+	if (!atomic_add_unless(&snapshot_device_available, -1, 0)) {
+		error = -EBUSY;
+		goto Unlock;
+	}
 
 	if ((filp->f_flags & O_ACCMODE) == O_RDWR) {
 		atomic_inc(&snapshot_device_available);
-		return -ENOSYS;
+		error = -ENOSYS;
+		goto Unlock;
 	}
 	if(create_basic_memory_bitmaps()) {
 		atomic_inc(&snapshot_device_available);
-		return -ENOMEM;
+		error = -ENOMEM;
+		goto Unlock;
 	}
 	nonseekable_open(inode, filp);
 	data = &snapshot_state;
@@ -98,33 +105,36 @@ static int snapshot_open(struct inode *inode, struct file *filp)
 		if (error)
 			pm_notifier_call_chain(PM_POST_HIBERNATION);
 	}
-	if (error) {
+	if (error)
 		atomic_inc(&snapshot_device_available);
-		return error;
-	}
 	data->frozen = 0;
 	data->ready = 0;
 	data->platform_support = 0;
 
-	return 0;
+ Unlock:
+	mutex_unlock(&pm_mutex);
+
+	return error;
 }
 
 static int snapshot_release(struct inode *inode, struct file *filp)
 {
 	struct snapshot_data *data;
 
+	mutex_lock(&pm_mutex);
+
 	swsusp_free();
 	free_basic_memory_bitmaps();
 	data = filp->private_data;
 	free_all_swap_pages(data->swap);
-	if (data->frozen) {
-		mutex_lock(&pm_mutex);
+	if (data->frozen)
 		thaw_processes();
-		mutex_unlock(&pm_mutex);
-	}
 	pm_notifier_call_chain(data->mode == O_WRONLY ?
 			PM_POST_HIBERNATION : PM_POST_RESTORE);
 	atomic_inc(&snapshot_device_available);
+
+	mutex_unlock(&pm_mutex);
+
 	return 0;
 }
 
@@ -134,9 +144,13 @@ static ssize_t snapshot_read(struct file *filp, char __user *buf,
 	struct snapshot_data *data;
 	ssize_t res;
 
+	mutex_lock(&pm_mutex);
+
 	data = filp->private_data;
-	if (!data->ready)
-		return -ENODATA;
+	if (!data->ready) {
+		res = -ENODATA;
+		goto Unlock;
+	}
 	res = snapshot_read_next(&data->handle, count);
 	if (res > 0) {
 		if (copy_to_user(buf, data_of(data->handle), res))
@@ -144,6 +158,10 @@ static ssize_t snapshot_read(struct file *filp, char __user *buf,
 		else
 			*offp = data->handle.offset;
 	}
+
+ Unlock:
+	mutex_unlock(&pm_mutex);
+
 	return res;
 }
 
@@ -153,6 +171,8 @@ static ssize_t snapshot_write(struct file *filp, const char __user *buf,
 	struct snapshot_data *data;
 	ssize_t res;
 
+	mutex_lock(&pm_mutex);
+
 	data = filp->private_data;
 	res = snapshot_write_next(&data->handle, count);
 	if (res > 0) {
@@ -161,11 +181,14 @@ static ssize_t snapshot_write(struct file *filp, const char __user *buf,
 		else
 			*offp = data->handle.offset;
 	}
+
+	mutex_unlock(&pm_mutex);
+
 	return res;
 }
 
-static int snapshot_ioctl(struct inode *inode, struct file *filp,
-                          unsigned int cmd, unsigned long arg)
+static long snapshot_ioctl(struct file *filp, unsigned int cmd,
+							unsigned long arg)
 {
 	int error = 0;
 	struct snapshot_data *data;
@@ -179,6 +202,9 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
+	if (!mutex_trylock(&pm_mutex))
+		return -EBUSY;
+
 	data = filp->private_data;
 
 	switch (cmd) {
@@ -186,7 +212,6 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 	case SNAPSHOT_FREEZE:
 		if (data->frozen)
 			break;
-		mutex_lock(&pm_mutex);
 		printk("Syncing filesystems ... ");
 		sys_sync();
 		printk("done.\n");
@@ -194,7 +219,6 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 		error = freeze_processes();
 		if (error)
 			thaw_processes();
-		mutex_unlock(&pm_mutex);
 		if (!error)
 			data->frozen = 1;
 		break;
@@ -202,9 +226,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 	case SNAPSHOT_UNFREEZE:
 		if (!data->frozen || data->ready)
 			break;
-		mutex_lock(&pm_mutex);
 		thaw_processes();
-		mutex_unlock(&pm_mutex);
 		data->frozen = 0;
 		break;
 
@@ -307,16 +329,11 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 			error = -EPERM;
 			break;
 		}
-		if (!mutex_trylock(&pm_mutex)) {
-			error = -EBUSY;
-			break;
-		}
 		/*
 		 * Tasks are frozen and the notifiers have been called with
 		 * PM_HIBERNATION_PREPARE
 		 */
 		error = suspend_devices_and_enter(PM_SUSPEND_MEM);
-		mutex_unlock(&pm_mutex);
 		break;
 
 	case SNAPSHOT_PLATFORM_SUPPORT:
@@ -390,6 +407,8 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp,
 
 	}
 
+	mutex_unlock(&pm_mutex);
+
 	return error;
 }
 
@@ -399,7 +418,7 @@ static const struct file_operations snapshot_fops = {
 	.read = snapshot_read,
 	.write = snapshot_write,
 	.llseek = no_llseek,
-	.ioctl = snapshot_ioctl,
+	.unlocked_ioctl = snapshot_ioctl,
 };
 
 static struct miscdevice snapshot_device = {
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index e337390fce01..8392a9da6450 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -33,13 +33,9 @@
  */
 void __ptrace_link(struct task_struct *child, struct task_struct *new_parent)
 {
-	BUG_ON(!list_empty(&child->ptrace_list));
-	if (child->parent == new_parent)
-		return;
-	list_add(&child->ptrace_list, &child->parent->ptrace_children);
-	remove_parent(child);
+	BUG_ON(!list_empty(&child->ptrace_entry));
+	list_add(&child->ptrace_entry, &new_parent->ptraced);
 	child->parent = new_parent;
-	add_parent(child);
 }
  
 /*
@@ -73,12 +69,8 @@ void __ptrace_unlink(struct task_struct *child)
 	BUG_ON(!child->ptrace);
 
 	child->ptrace = 0;
-	if (ptrace_reparented(child)) {
-		list_del_init(&child->ptrace_list);
-		remove_parent(child);
-		child->parent = child->real_parent;
-		add_parent(child);
-	}
+	child->parent = child->real_parent;
+	list_del_init(&child->ptrace_entry);
 
 	if (task_is_traced(child))
 		ptrace_untrace(child);
@@ -492,15 +484,34 @@ int ptrace_traceme(void)
 	/*
 	 * Are we already being traced?
 	 */
+repeat:
 	task_lock(current);
 	if (!(current->ptrace & PT_PTRACED)) {
+		/*
+		 * See ptrace_attach() comments about the locking here.
+		 */
+		unsigned long flags;
+		if (!write_trylock_irqsave(&tasklist_lock, flags)) {
+			task_unlock(current);
+			do {
+				cpu_relax();
+			} while (!write_can_lock(&tasklist_lock));
+			goto repeat;
+		}
+
 		ret = security_ptrace(current->parent, current,
 				      PTRACE_MODE_ATTACH);
+
 		/*
 		 * Set the ptrace bit in the process ptrace flags.
+		 * Then link us on our parent's ptraced list.
 		 */
-		if (!ret)
+		if (!ret) {
 			current->ptrace |= PT_PTRACED;
+			__ptrace_link(current, current->real_parent);
+		}
+
+		write_unlock_irqrestore(&tasklist_lock, flags);
 	}
 	task_unlock(current);
 	return ret;