From da9cbc87395308a21465bd25441297bbba0477e1 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Mon, 30 Jun 2008 20:42:08 +0200
Subject: block: blkdev.h cleanup, move iocontext stuff to iocontext.h

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
---
 kernel/exit.c | 1 +
 1 file changed, 1 insertion(+)

(limited to 'kernel/exit.c')

diff --git a/kernel/exit.c b/kernel/exit.c
index 8f6185e69b6..ceb25878283 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -13,6 +13,7 @@
 #include <linux/personality.h>
 #include <linux/tty.h>
 #include <linux/mnt_namespace.h>
+#include <linux/iocontext.h>
 #include <linux/key.h>
 #include <linux/security.h>
 #include <linux/cpu.h>
-- 
cgit v1.2.3-70-g09d2


From 98abed02007b19bbfd68b6d06a5485afc3eeb01b Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Wed, 19 Mar 2008 19:24:59 -0700
Subject: do_wait reorganization

This breaks out the guts of do_wait into three subfunctions.
The control flow is less nonobvious without so much goto.
do_wait_thread and ptrace_do_wait contain the main work of the outer loop.
wait_consider_task contains the main work of the inner loop.

Signed-off-by: Roland McGrath <roland@redhat.com>
---
 kernel/exit.c | 215 ++++++++++++++++++++++++++++++++++++----------------------
 1 file changed, 135 insertions(+), 80 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/kernel/exit.c b/kernel/exit.c
index ceb25878283..7453356a961 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -1238,7 +1238,7 @@ 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 +1246,10 @@ 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;
@@ -1397,13 +1400,16 @@ static int wait_task_zombie(struct task_struct *p, int noreap,
  * released the lock and the system call should return.
  */
 static int wait_task_stopped(struct task_struct *p,
-			     int noreap, struct siginfo __user *infop,
+			     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 (!(p->ptrace & PT_PTRACED) && !(options & WUNTRACED))
+		return 0;
+
 	exit_code = 0;
 	spin_lock_irq(&p->sighand->siglock);
 
@@ -1421,7 +1427,7 @@ 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;
@@ -1442,7 +1448,7 @@ unlock_sig:
 	why = (p->ptrace & PT_PTRACED) ? 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 +1482,7 @@ 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 +1490,9 @@ 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 +1502,7 @@ 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 +1528,137 @@ 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 still -ECHILD.
+ */
+static int wait_consider_task(struct task_struct *parent,
+			      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 <= 0)
+		return ret;
+
+	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(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 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) {
+		int ret = wait_consider_task(tsk, 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;
+
+	/*
+	 * If we never saw an eligile child, check for children stolen by
+	 * ptrace.  We don't leave -ECHILD in *@notask_error if there are any,
+	 * because we will eventually be allowed to wait for them again.
+	 */
+	if (!*notask_error)
+		return 0;
+
+	list_for_each_entry(p, &tsk->ptrace_children, ptrace_list) {
+		int ret = eligible_child(type, pid, options, p);
+		if (unlikely(ret < 0))
+			return ret;
+		if (ret) {
+			*notask_error = 0;
+			return 0;
+		}
+	}
+
+	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 +1666,14 @@ 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);
-- 
cgit v1.2.3-70-g09d2


From f470021adb9190819c03d6d8c5c860a17480aa6d Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Mon, 24 Mar 2008 18:36:23 -0700
Subject: ptrace children revamp

ptrace no longer fiddles with the children/sibling links, and the
old ptrace_children list is gone.  Now ptrace, whether of one's own
children or another's via PTRACE_ATTACH, just uses the new ptraced
list instead.

There should be no user-visible difference that matters.  The only
change is the order in which do_wait() sees multiple stopped
children and stopped ptrace attachees.  Since wait_task_stopped()
was changed earlier so it no longer reorders the children list, we
already know this won't cause any new problems.

Signed-off-by: Roland McGrath <roland@redhat.com>
---
 include/linux/init_task.h |   4 +-
 include/linux/sched.h     |  26 +++---
 kernel/exit.c             | 226 ++++++++++++++++++++++++----------------------
 kernel/fork.c             |   6 +-
 kernel/ptrace.c           |  37 +++++---
 5 files changed, 160 insertions(+), 139 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 9927a88674a..93c45acf249 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -140,8 +140,8 @@ extern struct group_info init_groups;
 		.nr_cpus_allowed = NR_CPUS,				\
 	},								\
 	.tasks		= LIST_HEAD_INIT(tsk.tasks),			\
-	.ptrace_children= LIST_HEAD_INIT(tsk.ptrace_children),		\
-	.ptrace_list	= LIST_HEAD_INIT(tsk.ptrace_list),		\
+	.ptraced	= LIST_HEAD_INIT(tsk.ptraced),			\
+	.ptrace_entry	= LIST_HEAD_INIT(tsk.ptrace_entry),		\
 	.real_parent	= &tsk,						\
 	.parent		= &tsk,						\
 	.children	= LIST_HEAD_INIT(tsk.children),			\
diff --git a/include/linux/sched.h b/include/linux/sched.h
index ba2f859c6e4..1941d8b5cf1 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1062,12 +1062,6 @@ struct task_struct {
 #endif
 
 	struct list_head tasks;
-	/*
-	 * ptrace_list/ptrace_children forms the list of my children
-	 * that were stolen by a ptracer.
-	 */
-	struct list_head ptrace_children;
-	struct list_head ptrace_list;
 
 	struct mm_struct *mm, *active_mm;
 
@@ -1089,18 +1083,25 @@ struct task_struct {
 	/* 
 	 * pointers to (original) parent process, youngest child, younger sibling,
 	 * older sibling, respectively.  (p->father can be replaced with 
-	 * p->parent->pid)
+	 * p->real_parent->pid)
 	 */
-	struct task_struct *real_parent; /* real parent process (when being debugged) */
-	struct task_struct *parent;	/* parent process */
+	struct task_struct *real_parent; /* real parent process */
+	struct task_struct *parent; /* recipient of SIGCHLD, wait4() reports */
 	/*
-	 * children/sibling forms the list of my children plus the
-	 * tasks I'm ptracing.
+	 * children/sibling forms the list of my natural children
 	 */
 	struct list_head children;	/* list of my children */
 	struct list_head sibling;	/* linkage in my parent's children list */
 	struct task_struct *group_leader;	/* threadgroup leader */
 
+	/*
+	 * ptraced is the list of tasks this task is using ptrace on.
+	 * This includes both natural children and PTRACE_ATTACH targets.
+	 * p->ptrace_entry is p's link on the p->parent->ptraced list.
+	 */
+	struct list_head ptraced;
+	struct list_head ptrace_entry;
+
 	/* PID/PID hash table linkage. */
 	struct pid_link pids[PIDTYPE_MAX];
 	struct list_head thread_group;
@@ -1876,9 +1877,6 @@ extern void wait_task_inactive(struct task_struct * p);
 #define wait_task_inactive(p)	do { } while (0)
 #endif
 
-#define remove_parent(p)	list_del_init(&(p)->sibling)
-#define add_parent(p)		list_add_tail(&(p)->sibling,&(p)->parent->children)
-
 #define next_task(p)	list_entry(rcu_dereference((p)->tasks.next), struct task_struct, tasks)
 
 #define for_each_process(p) \
diff --git a/kernel/exit.c b/kernel/exit.c
index 7453356a961..1e909826a80 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 +702,71 @@ static void exit_mm(struct task_struct * tsk)
 	mmput(mm);
 }
 
-static void
-reparent_thread(struct task_struct *p, struct task_struct *father, int traced)
+/*
+ * 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)
 {
-	if (p->pdeath_signal)
-		/* We already hold the tasklist_lock here.  */
-		group_send_sig_info(p->pdeath_signal, SEND_SIG_NOINFO, p);
+	struct task_struct *p, *n;
 
-	/* 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.
+	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.
 		 */
-		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);
+		}
 
-		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 +781,8 @@ 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 +799,15 @@ 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 +816,19 @@ 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 +1189,6 @@ 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:
@@ -1399,7 +1401,7 @@ static int wait_task_zombie(struct task_struct *p, int options,
  * 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,
+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)
 {
@@ -1407,7 +1409,7 @@ static int wait_task_stopped(struct task_struct *p,
 	uid_t uid = 0; /* unneeded, required by compiler */
 	pid_t pid;
 
-	if (!(p->ptrace & PT_PTRACED) && !(options & WUNTRACED))
+	if (!(options & WUNTRACED))
 		return 0;
 
 	exit_code = 0;
@@ -1416,7 +1418,7 @@ static int wait_task_stopped(struct task_struct *p,
 	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.
@@ -1445,7 +1447,7 @@ 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(options & WNOWAIT))
@@ -1536,7 +1538,7 @@ static int wait_task_continued(struct task_struct *p, int options,
  * Returns zero if the search for a child should continue;
  * then *@notask_error is 0 if @p is an eligible child, or still -ECHILD.
  */
-static int wait_consider_task(struct task_struct *parent,
+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,
@@ -1546,6 +1548,15 @@ static int wait_consider_task(struct task_struct *parent,
 	if (ret <= 0)
 		return 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;
 
@@ -1562,7 +1573,8 @@ static int wait_consider_task(struct task_struct *parent,
 	*notask_error = 0;
 
 	if (task_is_stopped_or_traced(p))
-		return wait_task_stopped(p, options, infop, stat_addr, ru);
+		return wait_task_stopped(ptrace, p, options,
+					 infop, stat_addr, ru);
 
 	return wait_task_continued(p, options, infop, stat_addr, ru);
 }
@@ -1583,11 +1595,16 @@ static int do_wait_thread(struct task_struct *tsk, int *notask_error,
 	struct task_struct *p;
 
 	list_for_each_entry(p, &tsk->children, sibling) {
-		int ret = wait_consider_task(tsk, p, notask_error,
-					     type, pid, options,
-					     infop, stat_addr, ru);
-		if (ret)
-			return ret;
+		/*
+		 * 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;
@@ -1601,21 +1618,16 @@ static int ptrace_do_wait(struct task_struct *tsk, int *notask_error,
 	struct task_struct *p;
 
 	/*
-	 * If we never saw an eligile child, check for children stolen by
-	 * ptrace.  We don't leave -ECHILD in *@notask_error if there are any,
-	 * because we will eventually be allowed to wait for them again.
+	 * Traditionally we see ptrace'd stopped tasks regardless of options.
 	 */
-	if (!*notask_error)
-		return 0;
+	options |= WUNTRACED;
 
-	list_for_each_entry(p, &tsk->ptrace_children, ptrace_list) {
-		int ret = eligible_child(type, pid, options, p);
-		if (unlikely(ret < 0))
+	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;
-		if (ret) {
-			*notask_error = 0;
-			return 0;
-		}
 	}
 
 	return 0;
diff --git a/kernel/fork.c b/kernel/fork.c
index 4bd2f516401..adefc1131f2 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/ptrace.c b/kernel/ptrace.c
index e337390fce0..8392a9da645 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;
-- 
cgit v1.2.3-70-g09d2


From 14dd0b81414a58caf0296dbeace016bb0a5d11ab Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Sun, 30 Mar 2008 18:41:25 -0700
Subject: do_wait: return security_task_wait() error code in place of -ECHILD

This reverts the effect of commit f2cc3eb133baa2e9dc8efd40f417106b2ee520f3
"do_wait: fix security checks".  That change reverted the effect of commit
73243284463a761e04d69d22c7516b2be7de096c.  The rationale for the original
commit still stands.  The inconsistent treatment of children hidden by
ptrace was an unintended omission in the original change and in no way
invalidates its purpose.

This makes do_wait return the error returned by security_task_wait()
(usually -EACCES) in place of -ECHILD when there are some children the
caller would be able to wait for if not for the permission failure.  A
permission error will give the user a clue to look for security policy
problems, rather than for mysterious wait bugs.

Signed-off-by: Roland McGrath <roland@redhat.com>
---
 kernel/exit.c | 30 ++++++++++++++++++++----------
 1 file changed, 20 insertions(+), 10 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/kernel/exit.c b/kernel/exit.c
index 1e909826a80..a2af6cac823 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -1199,14 +1199,10 @@ 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,
@@ -1536,7 +1532,8 @@ static int wait_task_continued(struct task_struct *p, int options,
  * -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 still -ECHILD.
+ * 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,
@@ -1545,9 +1542,21 @@ static int wait_consider_task(struct task_struct *parent, int ptrace,
 			      int __user *stat_addr, struct rusage __user *ru)
 {
 	int ret = eligible_child(type, pid, options, p);
-	if (ret <= 0)
+	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.
@@ -1585,7 +1594,8 @@ static int wait_consider_task(struct task_struct *parent, int ptrace,
  * -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 still -ECHILD.
+ * *@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,
-- 
cgit v1.2.3-70-g09d2


From 666f164f4fbfa78bd00fb4b74788b42a39842c64 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Tue, 8 Apr 2008 23:12:30 -0700
Subject: fix dangling zombie when new parent ignores children

This fixes an arcane bug that we think was a regression introduced
by commit b2b2cbc4b2a2f389442549399a993a8306420baf.  When a parent
ignores SIGCHLD (or uses SA_NOCLDWAIT), its children would self-reap
but they don't if it's using ptrace on them.  When the parent thread
later exits and ceases to ptrace a child but leaves other live
threads in the parent's thread group, any zombie children are left
dangling.  The fix makes them self-reap then, as they would have
done earlier if ptrace had not been in use.

Signed-off-by: Roland McGrath <roland@redhat.com>
---
 kernel/exit.c | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

(limited to 'kernel/exit.c')

diff --git a/kernel/exit.c b/kernel/exit.c
index a2af6cac823..93d2711b938 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -702,6 +702,23 @@ static void exit_mm(struct task_struct * tsk)
 	mmput(mm);
 }
 
+/*
+ * 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)
+{
+	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;
+}
+
 /*
  * Detach all tasks we were using ptrace on.
  * Any that need to be release_task'd are put on the @dead list.
@@ -711,6 +728,7 @@ static void exit_mm(struct task_struct * tsk)
 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);
@@ -726,10 +744,18 @@ static void ptrace_exit(struct task_struct *parent, struct list_head *dead)
 		 * 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.
 		 */
 		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_detached(p)) {
-- 
cgit v1.2.3-70-g09d2


From 3854a771821c970065e3203a0b40ddc4101538cc Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:29 -0700
Subject: __exit_signal: don't take rcu lock

There is no reason for rcu_read_lock() in __exit_signal().  tsk->sighand
can only be changed if tsk does exec, obviously this is not possible.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/exit.c | 2 --
 1 file changed, 2 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/kernel/exit.c b/kernel/exit.c
index 93d2711b938..a7799d8a640 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -85,7 +85,6 @@ static void __exit_signal(struct task_struct *tsk)
 	BUG_ON(!sig);
 	BUG_ON(!atomic_read(&sig->count));
 
-	rcu_read_lock();
 	sighand = rcu_dereference(tsk->sighand);
 	spin_lock(&sighand->siglock);
 
@@ -136,7 +135,6 @@ static void __exit_signal(struct task_struct *tsk)
 	tsk->signal = NULL;
 	tsk->sighand = NULL;
 	spin_unlock(&sighand->siglock);
-	rcu_read_unlock();
 
 	__cleanup_sighand(sighand);
 	clear_tsk_thread_flag(tsk,TIF_SIGPENDING);
-- 
cgit v1.2.3-70-g09d2


From 7b34e4283c685f5cc6ba6d30e939906eee0d4bcf Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:37 -0700
Subject: introduce PF_KTHREAD flag

Introduce the new PF_KTHREAD flag to mark the kernel threads.  It is set
by INIT_TASK() and copied to the forked childs (we could set it in
kthreadd() along with PF_NOFREEZE instead).

daemonize() was changed as well.  In that case testing of PF_KTHREAD is
racy, but daemonize() is hopeless anyway.

This flag is cleared in do_execve(), before search_binary_handler().
Probably not the best place, we can do this in exec_mmap() or in
start_thread(), or clear it along with PF_FORKNOEXEC.  But I think this
doesn't matter in practice, and if do_execve() fails kthread should die
soon.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                 | 1 +
 include/linux/init_task.h | 2 +-
 include/linux/sched.h     | 1 +
 kernel/exit.c             | 2 +-
 4 files changed, 4 insertions(+), 2 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/fs/exec.c b/fs/exec.c
index af249af4cca..cd2e8c9b124 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1326,6 +1326,7 @@ int do_execve(char * filename,
 	if (retval < 0)
 		goto out;
 
+	current->flags &= ~PF_KTHREAD;
 	retval = search_binary_handler(bprm,regs);
 	if (retval >= 0) {
 		/* execve success */
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 93c45acf249..021d8e720c7 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -122,7 +122,7 @@ extern struct group_info init_groups;
 	.state		= 0,						\
 	.stack		= &init_thread_info,				\
 	.usage		= ATOMIC_INIT(2),				\
-	.flags		= 0,						\
+	.flags		= PF_KTHREAD,					\
 	.lock_depth	= -1,						\
 	.prio		= MAX_PRIO-20,					\
 	.static_prio	= MAX_PRIO-20,					\
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 79e749dbf81..eec64a4adb9 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1483,6 +1483,7 @@ static inline void put_task_struct(struct task_struct *t)
 #define PF_EXITING	0x00000004	/* getting shut down */
 #define PF_EXITPIDONE	0x00000008	/* pi exit done on shut down */
 #define PF_VCPU		0x00000010	/* I'm a virtual CPU */
+#define PF_KTHREAD	0x00000020	/* I am a kernel thread */
 #define PF_FORKNOEXEC	0x00000040	/* forked but didn't exec */
 #define PF_SUPERPRIV	0x00000100	/* used super-user privileges */
 #define PF_DUMPCORE	0x00000200	/* dumped core */
diff --git a/kernel/exit.c b/kernel/exit.c
index a7799d8a640..28a44a2612d 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -430,7 +430,7 @@ void daemonize(const char *name, ...)
 	 * We don't want to have TIF_FREEZE set if the system-wide hibernation
 	 * or suspend transition begins right now.
 	 */
-	current->flags |= PF_NOFREEZE;
+	current->flags |= (PF_NOFREEZE | PF_KTHREAD);
 
 	if (current->nsproxy != &init_nsproxy) {
 		get_nsproxy(&init_nsproxy);
-- 
cgit v1.2.3-70-g09d2


From 32ecb1f26dd50eeaac4e3f4dea4541c97848e459 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:41 -0700
Subject: coredump: turn mm->core_startup_done into the pointer to struct
 core_state

mm->core_startup_done points to "struct completion startup_done" allocated
on the coredump_wait()'s stack.  Introduce the new structure, core_state,
which holds this "struct completion".  This way we can add more info
visible to the threads participating in coredump without enlarging
mm_struct.

No changes in affected .o files.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                | 8 ++++----
 include/linux/mm_types.h | 7 ++++++-
 kernel/exit.c            | 2 +-
 3 files changed, 11 insertions(+), 6 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/fs/exec.c b/fs/exec.c
index e347e6ed161..71734568f01 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1597,13 +1597,13 @@ static int coredump_wait(int exit_code)
 {
 	struct task_struct *tsk = current;
 	struct mm_struct *mm = tsk->mm;
-	struct completion startup_done;
+	struct core_state core_state;
 	struct completion *vfork_done;
 	int core_waiters;
 
 	init_completion(&mm->core_done);
-	init_completion(&startup_done);
-	mm->core_startup_done = &startup_done;
+	init_completion(&core_state.startup);
+	mm->core_state = &core_state;
 
 	core_waiters = zap_threads(tsk, mm, exit_code);
 	up_write(&mm->mmap_sem);
@@ -1622,7 +1622,7 @@ static int coredump_wait(int exit_code)
 	}
 
 	if (core_waiters)
-		wait_for_completion(&startup_done);
+		wait_for_completion(&core_state.startup);
 fail:
 	BUG_ON(mm->core_waiters);
 	return core_waiters;
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 02a27ae7853..97819efd233 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -159,6 +159,10 @@ struct vm_area_struct {
 #endif
 };
 
+struct core_state {
+	struct completion startup;
+};
+
 struct mm_struct {
 	struct vm_area_struct * mmap;		/* list of VMAs */
 	struct rb_root mm_rb;
@@ -220,7 +224,8 @@ struct mm_struct {
 	unsigned long flags; /* Must use atomic bitops to access the bits */
 
 	/* coredumping support */
-	struct completion *core_startup_done, core_done;
+	struct core_state *core_state;
+	struct completion core_done;
 
 	/* aio bits */
 	rwlock_t		ioctx_list_lock;	/* aio lock */
diff --git a/kernel/exit.c b/kernel/exit.c
index 28a44a2612d..f7fa21dbced 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -680,7 +680,7 @@ static void exit_mm(struct task_struct * tsk)
 		up_read(&mm->mmap_sem);
 		down_write(&mm->mmap_sem);
 		if (!--mm->core_waiters)
-			complete(mm->core_startup_done);
+			complete(&mm->core_state->startup);
 		up_write(&mm->mmap_sem);
 
 		wait_for_completion(&mm->core_done);
-- 
cgit v1.2.3-70-g09d2


From 999d9fc1670bc082928b93b11d1f2e0e417d973c Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:41 -0700
Subject: coredump: move mm->core_waiters into struct core_state

Move mm->core_waiters into "struct core_state" allocated on stack.  This
shrinks mm_struct a little bit and allows further changes.

This patch mostly does s/core_waiters/core_state.  The only essential
change is that coredump_wait() must clear mm->core_state before return.

The coredump_wait()'s path is uglified and .text grows by 30 bytes, this
is fixed by the next patch.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                | 21 +++++++++++----------
 include/linux/mm_types.h |  2 +-
 kernel/exit.c            |  8 ++++----
 kernel/fork.c            |  2 +-
 kernel/signal.c          |  4 ++--
 5 files changed, 19 insertions(+), 18 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/fs/exec.c b/fs/exec.c
index 71734568f01..50de3aaff4d 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -722,12 +722,10 @@ static int exec_mmap(struct mm_struct *mm)
 		 * Make sure that if there is a core dump in progress
 		 * for the old mm, we get out and die instead of going
 		 * through with the exec.  We must hold mmap_sem around
-		 * checking core_waiters and changing tsk->mm.  The
-		 * core-inducing thread will increment core_waiters for
-		 * each thread whose ->mm == old_mm.
+		 * checking core_state and changing tsk->mm.
 		 */
 		down_read(&old_mm->mmap_sem);
-		if (unlikely(old_mm->core_waiters)) {
+		if (unlikely(old_mm->core_state)) {
 			up_read(&old_mm->mmap_sem);
 			return -EINTR;
 		}
@@ -1514,7 +1512,7 @@ static void zap_process(struct task_struct *start)
 	t = start;
 	do {
 		if (t != current && t->mm) {
-			t->mm->core_waiters++;
+			t->mm->core_state->nr_threads++;
 			sigaddset(&t->pending.signal, SIGKILL);
 			signal_wake_up(t, 1);
 		}
@@ -1538,11 +1536,11 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
 	if (err)
 		return err;
 
-	if (atomic_read(&mm->mm_users) == mm->core_waiters + 1)
+	if (atomic_read(&mm->mm_users) == mm->core_state->nr_threads + 1)
 		goto done;
 	/*
 	 * We should find and kill all tasks which use this mm, and we should
-	 * count them correctly into mm->core_waiters. We don't take tasklist
+	 * count them correctly into ->nr_threads. We don't take tasklist
 	 * lock, but this is safe wrt:
 	 *
 	 * fork:
@@ -1590,7 +1588,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
 	}
 	rcu_read_unlock();
 done:
-	return mm->core_waiters;
+	return mm->core_state->nr_threads;
 }
 
 static int coredump_wait(int exit_code)
@@ -1603,9 +1601,12 @@ static int coredump_wait(int exit_code)
 
 	init_completion(&mm->core_done);
 	init_completion(&core_state.startup);
+	core_state.nr_threads = 0;
 	mm->core_state = &core_state;
 
 	core_waiters = zap_threads(tsk, mm, exit_code);
+	if (core_waiters < 0)
+		mm->core_state = NULL;
 	up_write(&mm->mmap_sem);
 
 	if (unlikely(core_waiters < 0))
@@ -1623,8 +1624,8 @@ static int coredump_wait(int exit_code)
 
 	if (core_waiters)
 		wait_for_completion(&core_state.startup);
+	mm->core_state = NULL;
 fail:
-	BUG_ON(mm->core_waiters);
 	return core_waiters;
 }
 
@@ -1702,7 +1703,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs)
 	/*
 	 * If another thread got here first, or we are not dumpable, bail out.
 	 */
-	if (mm->core_waiters || !get_dumpable(mm)) {
+	if (mm->core_state || !get_dumpable(mm)) {
 		up_write(&mm->mmap_sem);
 		goto fail;
 	}
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 97819efd233..c0b1747b61a 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -160,6 +160,7 @@ struct vm_area_struct {
 };
 
 struct core_state {
+	int nr_threads;
 	struct completion startup;
 };
 
@@ -179,7 +180,6 @@ struct mm_struct {
 	atomic_t mm_users;			/* How many users with user space? */
 	atomic_t mm_count;			/* How many references to "struct mm_struct" (users count as 1) */
 	int map_count;				/* number of VMAs */
-	int core_waiters;
 	struct rw_semaphore mmap_sem;
 	spinlock_t page_table_lock;		/* Protects page tables and some counters */
 
diff --git a/kernel/exit.c b/kernel/exit.c
index f7fa21dbced..988e232254e 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -670,16 +670,16 @@ static void exit_mm(struct task_struct * tsk)
 		return;
 	/*
 	 * Serialize with any possible pending coredump.
-	 * We must hold mmap_sem around checking core_waiters
+	 * We must hold mmap_sem around checking core_state
 	 * and clearing tsk->mm.  The core-inducing thread
-	 * will increment core_waiters for each thread in the
+	 * will increment ->nr_threads for each thread in the
 	 * group with ->mm != NULL.
 	 */
 	down_read(&mm->mmap_sem);
-	if (mm->core_waiters) {
+	if (mm->core_state) {
 		up_read(&mm->mmap_sem);
 		down_write(&mm->mmap_sem);
-		if (!--mm->core_waiters)
+		if (!--mm->core_state->nr_threads)
 			complete(&mm->core_state->startup);
 		up_write(&mm->mmap_sem);
 
diff --git a/kernel/fork.c b/kernel/fork.c
index eeaec6893b0..813d5c89b9d 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -400,7 +400,7 @@ static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p)
 	INIT_LIST_HEAD(&mm->mmlist);
 	mm->flags = (current->mm) ? current->mm->flags
 				  : MMF_DUMP_FILTER_DEFAULT;
-	mm->core_waiters = 0;
+	mm->core_state = NULL;
 	mm->nr_ptes = 0;
 	set_mm_counter(mm, file_rss, 0);
 	set_mm_counter(mm, anon_rss, 0);
diff --git a/kernel/signal.c b/kernel/signal.c
index 39c1706edf0..5c7b7eaa0dc 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1480,10 +1480,10 @@ static inline int may_ptrace_stop(void)
 	 * is a deadlock situation, and pointless because our tracer
 	 * is dead so don't allow us to stop.
 	 * If SIGKILL was already sent before the caller unlocked
-	 * ->siglock we must see ->core_waiters != 0. Otherwise it
+	 * ->siglock we must see ->core_state != NULL. Otherwise it
 	 * is safe to enter schedule().
 	 */
-	if (unlikely(current->mm->core_waiters) &&
+	if (unlikely(current->mm->core_state) &&
 	    unlikely(current->mm == current->parent->mm))
 		return 0;
 
-- 
cgit v1.2.3-70-g09d2


From c5f1cc8c1828486a61ab3e575da6e2c62b34d399 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:42 -0700
Subject: coredump: turn core_state->nr_threads into atomic_t

Turn core_state->nr_threads into atomic_t and kill now unneeded
down_write(&mm->mmap_sem) in exit_mm().

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                | 2 +-
 include/linux/mm_types.h | 2 +-
 kernel/exit.c            | 5 ++---
 3 files changed, 4 insertions(+), 5 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/fs/exec.c b/fs/exec.c
index c74bb34eeef..15d493fe8aa 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1591,7 +1591,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
 	}
 	rcu_read_unlock();
 done:
-	core_state->nr_threads = nr;
+	atomic_set(&core_state->nr_threads, nr);
 	return nr;
 }
 
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index c0b1747b61a..ae99a28ba6a 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -160,7 +160,7 @@ struct vm_area_struct {
 };
 
 struct core_state {
-	int nr_threads;
+	atomic_t nr_threads;
 	struct completion startup;
 };
 
diff --git a/kernel/exit.c b/kernel/exit.c
index 988e232254e..63d82957baa 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -678,10 +678,9 @@ static void exit_mm(struct task_struct * tsk)
 	down_read(&mm->mmap_sem);
 	if (mm->core_state) {
 		up_read(&mm->mmap_sem);
-		down_write(&mm->mmap_sem);
-		if (!--mm->core_state->nr_threads)
+
+		if (atomic_dec_and_test(&mm->core_state->nr_threads))
 			complete(&mm->core_state->startup);
-		up_write(&mm->mmap_sem);
 
 		wait_for_completion(&mm->core_done);
 		down_read(&mm->mmap_sem);
-- 
cgit v1.2.3-70-g09d2


From b564daf806d492dd4f7afe9b6c83b8d35d137669 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:44 -0700
Subject: coredump: construct the list of coredumping threads at startup time

binfmt->core_dump() has to iterate over the all threads in system in order
to find the coredumping threads and construct the list using the
GFP_ATOMIC allocations.

With this patch each thread allocates the list node on exit_mm()'s stack and
adds itself to the list.

This allows us to do further changes:

	- simplify ->core_dump()

	- change exit_mm() to clear ->mm first, then wait for ->core_done.
	  this makes the coredumping process visible to oom_kill

	- kill mm->core_done

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                |  2 ++
 include/linux/mm_types.h |  6 ++++++
 kernel/exit.c            | 15 ++++++++++++---
 3 files changed, 20 insertions(+), 3 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/fs/exec.c b/fs/exec.c
index b8ee842d93c..fe2873b8037 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1604,6 +1604,8 @@ static int coredump_wait(int exit_code, struct core_state *core_state)
 
 	init_completion(&mm->core_done);
 	init_completion(&core_state->startup);
+	core_state->dumper.task = tsk;
+	core_state->dumper.next = NULL;
 	core_waiters = zap_threads(tsk, mm, core_state, exit_code);
 	up_write(&mm->mmap_sem);
 
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index ae99a28ba6a..4d0d0abc79f 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -159,8 +159,14 @@ struct vm_area_struct {
 #endif
 };
 
+struct core_thread {
+	struct task_struct *task;
+	struct core_thread *next;
+};
+
 struct core_state {
 	atomic_t nr_threads;
+	struct core_thread dumper;
 	struct completion startup;
 };
 
diff --git a/kernel/exit.c b/kernel/exit.c
index 63d82957baa..b66f0d55c79 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -664,6 +664,7 @@ assign_new_owner:
 static void exit_mm(struct task_struct * tsk)
 {
 	struct mm_struct *mm = tsk->mm;
+	struct core_state *core_state;
 
 	mm_release(tsk, mm);
 	if (!mm)
@@ -676,11 +677,19 @@ static void exit_mm(struct task_struct * tsk)
 	 * group with ->mm != NULL.
 	 */
 	down_read(&mm->mmap_sem);
-	if (mm->core_state) {
+	core_state = mm->core_state;
+	if (core_state) {
+		struct core_thread self;
 		up_read(&mm->mmap_sem);
 
-		if (atomic_dec_and_test(&mm->core_state->nr_threads))
-			complete(&mm->core_state->startup);
+		self.task = tsk;
+		self.next = xchg(&core_state->dumper.next, &self);
+		/*
+		 * Implies mb(), the result of xchg() must be visible
+		 * to core_state->dumper.
+		 */
+		if (atomic_dec_and_test(&core_state->nr_threads))
+			complete(&core_state->startup);
 
 		wait_for_completion(&mm->core_done);
 		down_read(&mm->mmap_sem);
-- 
cgit v1.2.3-70-g09d2


From a94e2d408eaedbd85aae259621d46fafc10479a2 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:46 -0700
Subject: coredump: kill mm->core_done

Now that we have core_state->dumper list we can use it to wake up the
sub-threads waiting for the coredump completion.

This uglifies the code and .text grows by 47 bytes, but otoh mm_struct
lessens by sizeof(struct completion).  Also, with this change we can
decouple exit_mm() from the coredumping code.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                | 25 ++++++++++++++++++++++---
 include/linux/mm_types.h |  4 +---
 kernel/exit.c            |  8 +++++++-
 3 files changed, 30 insertions(+), 7 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/fs/exec.c b/fs/exec.c
index fe2873b8037..bff43aeb235 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1602,7 +1602,6 @@ static int coredump_wait(int exit_code, struct core_state *core_state)
 	struct completion *vfork_done;
 	int core_waiters;
 
-	init_completion(&mm->core_done);
 	init_completion(&core_state->startup);
 	core_state->dumper.task = tsk;
 	core_state->dumper.next = NULL;
@@ -1628,6 +1627,27 @@ fail:
 	return core_waiters;
 }
 
+static void coredump_finish(struct mm_struct *mm)
+{
+	struct core_thread *curr, *next;
+	struct task_struct *task;
+
+	next = mm->core_state->dumper.next;
+	while ((curr = next) != NULL) {
+		next = curr->next;
+		task = curr->task;
+		/*
+		 * see exit_mm(), curr->task must not see
+		 * ->task == NULL before we read ->next.
+		 */
+		smp_mb();
+		curr->task = NULL;
+		wake_up_process(task);
+	}
+
+	mm->core_state = NULL;
+}
+
 /*
  * set_dumpable converts traditional three-value dumpable to two flags and
  * stores them into mm->flags.  It modifies lower two bits of mm->flags, but
@@ -1812,8 +1832,7 @@ fail_unlock:
 		argv_free(helper_argv);
 
 	current->fsuid = fsuid;
-	complete_all(&mm->core_done);
-	mm->core_state = NULL;
+	coredump_finish(mm);
 fail:
 	return retval;
 }
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 4d0d0abc79f..746f975b58e 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -229,9 +229,7 @@ struct mm_struct {
 
 	unsigned long flags; /* Must use atomic bitops to access the bits */
 
-	/* coredumping support */
-	struct core_state *core_state;
-	struct completion core_done;
+	struct core_state *core_state; /* coredumping support */
 
 	/* aio bits */
 	rwlock_t		ioctx_list_lock;	/* aio lock */
diff --git a/kernel/exit.c b/kernel/exit.c
index b66f0d55c79..8a4d4d12e29 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -691,7 +691,13 @@ static void exit_mm(struct task_struct * tsk)
 		if (atomic_dec_and_test(&core_state->nr_threads))
 			complete(&core_state->startup);
 
-		wait_for_completion(&mm->core_done);
+		for (;;) {
+			set_task_state(tsk, TASK_UNINTERRUPTIBLE);
+			if (!self.task) /* see coredump_finish() */
+				break;
+			schedule();
+		}
+		__set_task_state(tsk, TASK_RUNNING);
 		down_read(&mm->mmap_sem);
 	}
 	atomic_inc(&mm->mm_count);
-- 
cgit v1.2.3-70-g09d2


From 297c5d92634c809cef23d73e7b2556f2528ff7e2 Mon Sep 17 00:00:00 2001
From: Andrea Righi <righi.andrea@gmail.com>
Date: Fri, 25 Jul 2008 01:48:49 -0700
Subject: task IO accounting: provide distinct tgid/tid I/O statistics

Report per-thread I/O statistics in /proc/pid/task/tid/io and aggregate
parent I/O statistics in /proc/pid/io.  This approach follows the same
model used to account per-process and per-thread CPU times.

As a practial application, this allows for example to quickly find the top
I/O consumer when a process spawns many child threads that perform the
actual I/O work, because the aggregated I/O statistics can always be found
in /proc/pid/io.

[ Oleg Nesterov points out that we should check that the task is still
  alive before we iterate over the threads, but also says that we can do
  that fixup on top of this later.  - Linus ]

Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: Andrea Righi <righi.andrea@gmail.com>
Cc: Matt Heaton <matt@hostmonster.com>
Cc: Shailabh Nagar <nagar@watson.ibm.com>
Acked-by-with-comments: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/base.c        | 86 ++++++++++++++++++++++++++++++++++++++++++---------
 include/linux/sched.h |  4 +++
 kernel/exit.c         | 27 ++++++++++++++++
 kernel/fork.c         |  6 ++++
 4 files changed, 108 insertions(+), 15 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/fs/proc/base.c b/fs/proc/base.c
index 58c3e6a8e15..a891fe4cb43 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2376,29 +2376,82 @@ static int proc_base_fill_cache(struct file *filp, void *dirent,
 }
 
 #ifdef CONFIG_TASK_IO_ACCOUNTING
-static int proc_pid_io_accounting(struct task_struct *task, char *buffer)
-{
+static int do_io_accounting(struct task_struct *task, char *buffer, int whole)
+{
+	u64 rchar, wchar, syscr, syscw;
+	struct task_io_accounting ioac;
+
+	if (!whole) {
+		rchar = task->rchar;
+		wchar = task->wchar;
+		syscr = task->syscr;
+		syscw = task->syscw;
+		memcpy(&ioac, &task->ioac, sizeof(ioac));
+	} else {
+		unsigned long flags;
+		struct task_struct *t = task;
+		rchar = wchar = syscr = syscw = 0;
+		memset(&ioac, 0, sizeof(ioac));
+
+		rcu_read_lock();
+		do {
+			rchar += t->rchar;
+			wchar += t->wchar;
+			syscr += t->syscr;
+			syscw += t->syscw;
+
+			ioac.read_bytes += t->ioac.read_bytes;
+			ioac.write_bytes += t->ioac.write_bytes;
+			ioac.cancelled_write_bytes +=
+					t->ioac.cancelled_write_bytes;
+			t = next_thread(t);
+		} while (t != task);
+		rcu_read_unlock();
+
+		if (lock_task_sighand(task, &flags)) {
+			struct signal_struct *sig = task->signal;
+
+			rchar += sig->rchar;
+			wchar += sig->wchar;
+			syscr += sig->syscr;
+			syscw += sig->syscw;
+
+			ioac.read_bytes += sig->ioac.read_bytes;
+			ioac.write_bytes += sig->ioac.write_bytes;
+			ioac.cancelled_write_bytes +=
+					sig->ioac.cancelled_write_bytes;
+
+			unlock_task_sighand(task, &flags);
+		}
+	}
+
 	return sprintf(buffer,
-#ifdef CONFIG_TASK_XACCT
 			"rchar: %llu\n"
 			"wchar: %llu\n"
 			"syscr: %llu\n"
 			"syscw: %llu\n"
-#endif
 			"read_bytes: %llu\n"
 			"write_bytes: %llu\n"
 			"cancelled_write_bytes: %llu\n",
-#ifdef CONFIG_TASK_XACCT
-			(unsigned long long)task->rchar,
-			(unsigned long long)task->wchar,
-			(unsigned long long)task->syscr,
-			(unsigned long long)task->syscw,
-#endif
-			(unsigned long long)task->ioac.read_bytes,
-			(unsigned long long)task->ioac.write_bytes,
-			(unsigned long long)task->ioac.cancelled_write_bytes);
+			(unsigned long long)rchar,
+			(unsigned long long)wchar,
+			(unsigned long long)syscr,
+			(unsigned long long)syscw,
+			(unsigned long long)ioac.read_bytes,
+			(unsigned long long)ioac.write_bytes,
+			(unsigned long long)ioac.cancelled_write_bytes);
+}
+
+static int proc_tid_io_accounting(struct task_struct *task, char *buffer)
+{
+	return do_io_accounting(task, buffer, 0);
 }
-#endif
+
+static int proc_tgid_io_accounting(struct task_struct *task, char *buffer)
+{
+	return do_io_accounting(task, buffer, 1);
+}
+#endif /* CONFIG_TASK_IO_ACCOUNTING */
 
 /*
  * Thread groups
@@ -2470,7 +2523,7 @@ static const struct pid_entry tgid_base_stuff[] = {
 	REG("coredump_filter", S_IRUGO|S_IWUSR, coredump_filter),
 #endif
 #ifdef CONFIG_TASK_IO_ACCOUNTING
-	INF("io",	S_IRUGO, pid_io_accounting),
+	INF("io",	S_IRUGO, tgid_io_accounting),
 #endif
 };
 
@@ -2797,6 +2850,9 @@ static const struct pid_entry tid_base_stuff[] = {
 #ifdef CONFIG_FAULT_INJECTION
 	REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject),
 #endif
+#ifdef CONFIG_TASK_IO_ACCOUNTING
+	INF("io",	S_IRUGO, tid_io_accounting),
+#endif
 };
 
 static int proc_tid_base_readdir(struct file * filp,
diff --git a/include/linux/sched.h b/include/linux/sched.h
index af780f299c7..d22ffe06d0e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -506,6 +506,10 @@ struct signal_struct {
 	unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw;
 	unsigned long min_flt, maj_flt, cmin_flt, cmaj_flt;
 	unsigned long inblock, oublock, cinblock, coublock;
+#ifdef CONFIG_TASK_XACCT
+	u64 rchar, wchar, syscr, syscw;
+#endif
+	struct task_io_accounting ioac;
 
 	/*
 	 * Cumulative ns of scheduled CPU time for dead threads in the
diff --git a/kernel/exit.c b/kernel/exit.c
index 8a4d4d12e29..ad933bb29ec 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -120,6 +120,18 @@ static void __exit_signal(struct task_struct *tsk)
 		sig->nivcsw += tsk->nivcsw;
 		sig->inblock += task_io_get_inblock(tsk);
 		sig->oublock += task_io_get_oublock(tsk);
+#ifdef CONFIG_TASK_XACCT
+		sig->rchar += tsk->rchar;
+		sig->wchar += tsk->wchar;
+		sig->syscr += tsk->syscr;
+		sig->syscw += tsk->syscw;
+#endif /* CONFIG_TASK_XACCT */
+#ifdef CONFIG_TASK_IO_ACCOUNTING
+		sig->ioac.read_bytes += tsk->ioac.read_bytes;
+		sig->ioac.write_bytes += tsk->ioac.write_bytes;
+		sig->ioac.cancelled_write_bytes +=
+					tsk->ioac.cancelled_write_bytes;
+#endif /* CONFIG_TASK_IO_ACCOUNTING */
 		sig->sum_sched_runtime += tsk->se.sum_exec_runtime;
 		sig = NULL; /* Marker for below. */
 	}
@@ -1366,6 +1378,21 @@ static int wait_task_zombie(struct task_struct *p, int options,
 		psig->coublock +=
 			task_io_get_oublock(p) +
 			sig->oublock + sig->coublock;
+#ifdef CONFIG_TASK_XACCT
+		psig->rchar += p->rchar + sig->rchar;
+		psig->wchar += p->wchar + sig->wchar;
+		psig->syscr += p->syscr + sig->syscr;
+		psig->syscw += p->syscw + sig->syscw;
+#endif /* CONFIG_TASK_XACCT */
+#ifdef CONFIG_TASK_IO_ACCOUNTING
+		psig->ioac.read_bytes +=
+			p->ioac.read_bytes + sig->ioac.read_bytes;
+		psig->ioac.write_bytes +=
+			p->ioac.write_bytes + sig->ioac.write_bytes;
+		psig->ioac.cancelled_write_bytes +=
+				p->ioac.cancelled_write_bytes +
+				sig->ioac.cancelled_write_bytes;
+#endif /* CONFIG_TASK_IO_ACCOUNTING */
 		spin_unlock_irq(&p->parent->sighand->siglock);
 	}
 
diff --git a/kernel/fork.c b/kernel/fork.c
index 813d5c89b9d..b99d73e971a 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -812,6 +812,12 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
 	sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0;
 	sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0;
 	sig->inblock = sig->oublock = sig->cinblock = sig->coublock = 0;
+#ifdef CONFIG_TASK_XACCT
+	sig->rchar = sig->wchar = sig->syscr = sig->syscw = 0;
+#endif
+#ifdef CONFIG_TASK_IO_ACCOUNTING
+	memset(&sig->ioac, 0, sizeof(sig->ioac));
+#endif
 	sig->sum_sched_runtime = 0;
 	INIT_LIST_HEAD(&sig->cpu_timers[0]);
 	INIT_LIST_HEAD(&sig->cpu_timers[1]);
-- 
cgit v1.2.3-70-g09d2


From 30199f5a46aee204bf437a4f5b0740f3efe448b7 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:46 -0700
Subject: tracehook: exit

This moves the PTRACE_EVENT_EXIT tracing into a tracehook.h inline,
tracehook_report_exec().  The change has no effect, just clean-up.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tracehook.h | 15 +++++++++++++++
 kernel/exit.c             |  6 ++----
 2 files changed, 17 insertions(+), 4 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 6276353709c..967ab473afb 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -95,4 +95,19 @@ static inline void tracehook_report_exec(struct linux_binfmt *fmt,
 		send_sig(SIGTRAP, current, 0);
 }
 
+/**
+ * tracehook_report_exit - task has begun to exit
+ * @exit_code:		pointer to value destined for @current->exit_code
+ *
+ * @exit_code points to the value passed to do_exit(), which tracing
+ * might change here.  This is almost the first thing in do_exit(),
+ * before freeing any resources or setting the %PF_EXITING flag.
+ *
+ * Called with no locks held.
+ */
+static inline void tracehook_report_exit(long *exit_code)
+{
+	ptrace_event(PT_TRACE_EXIT, PTRACE_EVENT_EXIT, *exit_code);
+}
+
 #endif	/* <linux/tracehook.h> */
diff --git a/kernel/exit.c b/kernel/exit.c
index ad933bb29ec..c3691cbc220 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -46,6 +46,7 @@
 #include <linux/resource.h>
 #include <linux/blkdev.h>
 #include <linux/task_io_accounting_ops.h>
+#include <linux/tracehook.h>
 
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
@@ -1029,10 +1030,7 @@ NORET_TYPE void do_exit(long code)
 	if (unlikely(!tsk->pid))
 		panic("Attempted to kill the idle task!");
 
-	if (unlikely(current->ptrace & PT_TRACE_EXIT)) {
-		current->ptrace_message = code;
-		ptrace_notify((PTRACE_EVENT_EXIT << 8) | SIGTRAP);
-	}
+	tracehook_report_exit(&code);
 
 	/*
 	 * We're taking recursive faults here in do_exit. Safest is to just
-- 
cgit v1.2.3-70-g09d2


From dae33574dcf5211e1f43c7e45fa29f73ba3e00cb Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:48 -0700
Subject: tracehook: release_task

This moves the ptrace-related logic from release_task into tracehook.h and
ptrace.h inlines.  It provides clean hooks both before and after locking
tasklist_lock, for future tracing logic to do more cleanup without the
lock.

This also changes release_task() itself in the rare "zap_leader" case to
set the leader to EXIT_DEAD before iterating.  This maintains the
invariant that release_task() only ever handles a task in EXIT_DEAD.  This
is a common-sense invariant that is already always true except in this one
arcane case of zombie leader whose parent ignores SIGCHLD.

This change is harmless and only costs one store in this one rare case.
It keeps the expected state more consisently sane, which is nicer when
debugging weirdness in release_task().  It also lets some future code in
the tracehook entry points rely on this invariant for bookkeeping.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/ptrace.h    | 13 +++++++++++++
 include/linux/tracehook.h | 28 ++++++++++++++++++++++++++++
 kernel/exit.c             | 21 +++++++++------------
 3 files changed, 50 insertions(+), 12 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
index dae6d85520f..ed69c03692d 100644
--- a/include/linux/ptrace.h
+++ b/include/linux/ptrace.h
@@ -176,6 +176,19 @@ static inline void ptrace_init_task(struct task_struct *child, bool ptrace)
 	}
 }
 
+/**
+ * ptrace_release_task - final ptrace-related cleanup of a zombie being reaped
+ * @task:	task in %EXIT_DEAD state
+ *
+ * Called with write_lock(&tasklist_lock) held.
+ */
+static inline void ptrace_release_task(struct task_struct *task)
+{
+	BUG_ON(!list_empty(&task->ptraced));
+	ptrace_unlink(task);
+	BUG_ON(!list_empty(&task->ptrace_entry));
+}
+
 #ifndef force_successful_syscall_return
 /*
  * System call handlers that, upon successful completion, need to return a
diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 830e6e16097..9a5b3be2503 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -228,4 +228,32 @@ static inline void tracehook_report_vfork_done(struct task_struct *child,
 	ptrace_event(PT_TRACE_VFORK_DONE, PTRACE_EVENT_VFORK_DONE, pid);
 }
 
+/**
+ * tracehook_prepare_release_task - task is being reaped, clean up tracing
+ * @task:		task in %EXIT_DEAD state
+ *
+ * This is called in release_task() just before @task gets finally reaped
+ * and freed.  This would be the ideal place to remove and clean up any
+ * tracing-related state for @task.
+ *
+ * Called with no locks held.
+ */
+static inline void tracehook_prepare_release_task(struct task_struct *task)
+{
+}
+
+/**
+ * tracehook_finish_release_task - task is being reaped, clean up tracing
+ * @task:		task in %EXIT_DEAD state
+ *
+ * This is called in release_task() when @task is being in the middle of
+ * being reaped.  After this, there must be no tracing entanglements.
+ *
+ * Called with write_lock_irq(&tasklist_lock) held.
+ */
+static inline void tracehook_finish_release_task(struct task_struct *task)
+{
+	ptrace_release_task(task);
+}
+
 #endif	/* <linux/tracehook.h> */
diff --git a/kernel/exit.c b/kernel/exit.c
index c3691cbc220..da28745f7c3 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -163,27 +163,17 @@ 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;
 	int zap_leader;
 repeat:
+	tracehook_prepare_release_task(p);
 	atomic_dec(&p->user->processes);
 	proc_flush_task(p);
 	write_lock_irq(&tasklist_lock);
-	ptrace_release_task(p);
+	tracehook_finish_release_task(p);
 	__exit_signal(p);
 
 	/*
@@ -205,6 +195,13 @@ repeat:
 		 * that case.
 		 */
 		zap_leader = task_detached(leader);
+
+		/*
+		 * This maintains the invariant that release_task()
+		 * only runs on a task in EXIT_DEAD, just for sanity.
+		 */
+		if (zap_leader)
+			leader->exit_state = EXIT_DEAD;
 	}
 
 	write_unlock_irq(&tasklist_lock);
-- 
cgit v1.2.3-70-g09d2


From 2b2a1ff64afbadac842bbc58c5166962cf4f7664 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:54 -0700
Subject: tracehook: death

This moves the ptrace logic in task death (exit_notify) into tracehook.h
inlines.  Some code is rearranged slightly to make things nicer.  There is
no change, only cleanup.

There is one hook called with the tasklist_lock write-locked, as ptrace
needs.  There is also a new hook called after exit_state changes and
without locks.  This is a better place for tracing work to be in the
future, since it doesn't delay the whole system with locking.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sched.h     |  2 +-
 include/linux/tracehook.h | 52 +++++++++++++++++++++++++++++++++++++++++++++++
 kernel/exit.c             | 26 ++++++++----------------
 kernel/signal.c           | 10 ++++++---
 4 files changed, 69 insertions(+), 21 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/include/linux/sched.h b/include/linux/sched.h
index adb8077dc46..a95d84d0da9 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1796,7 +1796,7 @@ extern int kill_pid_info_as_uid(int, struct siginfo *, struct pid *, uid_t, uid_
 extern int kill_pgrp(struct pid *pid, int sig, int priv);
 extern int kill_pid(struct pid *pid, int sig, int priv);
 extern int kill_proc_info(int, struct siginfo *, pid_t);
-extern void do_notify_parent(struct task_struct *, int);
+extern int do_notify_parent(struct task_struct *, int);
 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);
diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 6dc428dd2f3..4c50e1b5734 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -471,4 +471,56 @@ static inline int tracehook_notify_jctl(int notify, int why)
 	return notify || (current->ptrace & PT_PTRACED);
 }
 
+/**
+ * tracehook_notify_death - task is dead, ready to notify parent
+ * @task:		@current task now exiting
+ * @death_cookie:	value to pass to tracehook_report_death()
+ * @group_dead:		nonzero if this was the last thread in the group to die
+ *
+ * Return the signal number to send our parent with do_notify_parent(), or
+ * zero to send no signal and leave a zombie, or -1 to self-reap right now.
+ *
+ * Called with write_lock_irq(&tasklist_lock) held.
+ */
+static inline int tracehook_notify_death(struct task_struct *task,
+					 void **death_cookie, int group_dead)
+{
+	if (task->exit_signal == -1)
+		return task->ptrace ? SIGCHLD : -1;
+
+	/*
+	 * If something other than our normal parent is ptracing us, then
+	 * send it a SIGCHLD instead of honoring exit_signal.  exit_signal
+	 * only has special meaning to our real parent.
+	 */
+	if (thread_group_empty(task) && !ptrace_reparented(task))
+		return task->exit_signal;
+
+	return task->ptrace ? SIGCHLD : 0;
+}
+
+/**
+ * tracehook_report_death - task is dead and ready to be reaped
+ * @task:		@current task now exiting
+ * @signal:		signal number sent to parent, or 0 or -1
+ * @death_cookie:	value passed back from tracehook_notify_death()
+ * @group_dead:		nonzero if this was the last thread in the group to die
+ *
+ * Thread has just become a zombie or is about to self-reap.  If positive,
+ * @signal is the signal number just sent to the parent (usually %SIGCHLD).
+ * If @signal is -1, this thread will self-reap.  If @signal is 0, this is
+ * a delayed_group_leader() zombie.  The @death_cookie was passed back by
+ * tracehook_notify_death().
+ *
+ * If normal reaping is not inhibited, @task->exit_state might be changing
+ * in parallel.
+ *
+ * Called without locks.
+ */
+static inline void tracehook_report_death(struct task_struct *task,
+					  int signal, void *death_cookie,
+					  int group_dead)
+{
+}
+
 #endif	/* <linux/tracehook.h> */
diff --git a/kernel/exit.c b/kernel/exit.c
index da28745f7c3..6cdf60712bd 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -885,7 +885,8 @@ static void forget_original_parent(struct task_struct *father)
  */
 static void exit_notify(struct task_struct *tsk, int group_dead)
 {
-	int state;
+	int signal;
+	void *cookie;
 
 	/*
 	 * This does two things:
@@ -922,22 +923,11 @@ static void exit_notify(struct task_struct *tsk, int group_dead)
 	    !capable(CAP_KILL))
 		tsk->exit_signal = SIGCHLD;
 
-	/* If something other than our normal parent is ptracing us, then
-	 * send it a SIGCHLD instead of honoring exit_signal.  exit_signal
-	 * only has special meaning to our real parent.
-	 */
-	if (!task_detached(tsk) && thread_group_empty(tsk)) {
-		int signal = ptrace_reparented(tsk) ?
-				SIGCHLD : tsk->exit_signal;
-		do_notify_parent(tsk, signal);
-	} else if (tsk->ptrace) {
-		do_notify_parent(tsk, SIGCHLD);
-	}
+	signal = tracehook_notify_death(tsk, &cookie, group_dead);
+	if (signal > 0)
+		signal = do_notify_parent(tsk, signal);
 
-	state = EXIT_ZOMBIE;
-	if (task_detached(tsk) && likely(!tsk->ptrace))
-		state = EXIT_DEAD;
-	tsk->exit_state = state;
+	tsk->exit_state = signal < 0 ? EXIT_DEAD : EXIT_ZOMBIE;
 
 	/* mt-exec, de_thread() is waiting for us */
 	if (thread_group_leader(tsk) &&
@@ -947,8 +937,10 @@ static void exit_notify(struct task_struct *tsk, int group_dead)
 
 	write_unlock_irq(&tasklist_lock);
 
+	tracehook_report_death(tsk, signal, cookie, group_dead);
+
 	/* If the process is dead, release it - nobody will wait for it */
-	if (state == EXIT_DEAD)
+	if (signal < 0)
 		release_task(tsk);
 }
 
diff --git a/kernel/signal.c b/kernel/signal.c
index e9e699f4b1b..0e862d3130f 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1326,9 +1326,11 @@ static inline void __wake_up_parent(struct task_struct *p,
 /*
  * Let a parent know about the death of a child.
  * For a stopped/continued status change, use do_notify_parent_cldstop instead.
+ *
+ * Returns -1 if our parent ignored us and so we've switched to
+ * self-reaping, or else @sig.
  */
-
-void do_notify_parent(struct task_struct *tsk, int sig)
+int do_notify_parent(struct task_struct *tsk, int sig)
 {
 	struct siginfo info;
 	unsigned long flags;
@@ -1399,12 +1401,14 @@ void do_notify_parent(struct task_struct *tsk, int sig)
 		 */
 		tsk->exit_signal = -1;
 		if (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN)
-			sig = 0;
+			sig = -1;
 	}
 	if (valid_signal(sig) && sig > 0)
 		__group_send_sig_info(sig, &info, tsk->parent);
 	__wake_up_parent(tsk, tsk->parent);
 	spin_unlock_irqrestore(&psig->siglock, flags);
+
+	return sig;
 }
 
 static void do_notify_parent_cldstop(struct task_struct *tsk, int why)
-- 
cgit v1.2.3-70-g09d2


From 7f2da1e7d0330395e5e9e350b879b98a1ea495df Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 10 May 2008 20:44:54 -0400
Subject: [PATCH] kill altroot

long overdue...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c                    | 89 +------------------------------------------
 fs/namespace.c                |  8 +---
 fs/open.c                     |  3 +-
 include/asm-alpha/namei.h     | 17 ---------
 include/asm-arm/namei.h       | 25 ------------
 include/asm-avr32/namei.h     |  7 ----
 include/asm-blackfin/namei.h  | 19 ---------
 include/asm-cris/namei.h      | 17 ---------
 include/asm-frv/namei.h       | 18 ---------
 include/asm-h8300/namei.h     | 17 ---------
 include/asm-ia64/namei.h      | 25 ------------
 include/asm-m32r/namei.h      | 17 ---------
 include/asm-m68k/namei.h      | 17 ---------
 include/asm-m68knommu/namei.h |  1 -
 include/asm-mips/namei.h      | 11 ------
 include/asm-mn10300/namei.h   | 22 -----------
 include/asm-parisc/namei.h    | 17 ---------
 include/asm-powerpc/namei.h   | 20 ----------
 include/asm-s390/namei.h      | 21 ----------
 include/asm-sh/namei.h        | 17 ---------
 include/asm-sparc/namei.h     |  8 ----
 include/asm-sparc64/namei.h   |  1 -
 include/asm-um/namei.h        |  6 ---
 include/asm-v850/namei.h      | 17 ---------
 include/asm-x86/namei.h       | 11 ------
 include/asm-xtensa/namei.h    | 26 -------------
 include/linux/fs_struct.h     |  3 +-
 include/linux/namei.h         |  1 -
 kernel/exec_domain.c          |  1 -
 kernel/exit.c                 |  2 -
 kernel/fork.c                 |  7 ----
 31 files changed, 5 insertions(+), 466 deletions(-)
 delete mode 100644 include/asm-alpha/namei.h
 delete mode 100644 include/asm-arm/namei.h
 delete mode 100644 include/asm-avr32/namei.h
 delete mode 100644 include/asm-blackfin/namei.h
 delete mode 100644 include/asm-cris/namei.h
 delete mode 100644 include/asm-frv/namei.h
 delete mode 100644 include/asm-h8300/namei.h
 delete mode 100644 include/asm-ia64/namei.h
 delete mode 100644 include/asm-m32r/namei.h
 delete mode 100644 include/asm-m68k/namei.h
 delete mode 100644 include/asm-m68knommu/namei.h
 delete mode 100644 include/asm-mips/namei.h
 delete mode 100644 include/asm-mn10300/namei.h
 delete mode 100644 include/asm-parisc/namei.h
 delete mode 100644 include/asm-powerpc/namei.h
 delete mode 100644 include/asm-s390/namei.h
 delete mode 100644 include/asm-sh/namei.h
 delete mode 100644 include/asm-sparc/namei.h
 delete mode 100644 include/asm-sparc64/namei.h
 delete mode 100644 include/asm-um/namei.h
 delete mode 100644 include/asm-v850/namei.h
 delete mode 100644 include/asm-x86/namei.h
 delete mode 100644 include/asm-xtensa/namei.h

(limited to 'kernel/exit.c')

diff --git a/fs/namei.c b/fs/namei.c
index 6c76e1ee9c4..095818089ac 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -31,7 +31,6 @@
 #include <linux/file.h>
 #include <linux/fcntl.h>
 #include <linux/device_cgroup.h>
-#include <asm/namei.h>
 #include <asm/uaccess.h>
 
 #define ACC_MODE(x) ("\000\004\002\006"[(x)&O_ACCMODE])
@@ -562,27 +561,16 @@ out_unlock:
 	return result;
 }
 
-static int __emul_lookup_dentry(const char *, struct nameidata *);
-
 /* SMP-safe */
-static __always_inline int
+static __always_inline void
 walk_init_root(const char *name, struct nameidata *nd)
 {
 	struct fs_struct *fs = current->fs;
 
 	read_lock(&fs->lock);
-	if (fs->altroot.dentry && !(nd->flags & LOOKUP_NOALT)) {
-		nd->path = fs->altroot;
-		path_get(&fs->altroot);
-		read_unlock(&fs->lock);
-		if (__emul_lookup_dentry(name,nd))
-			return 0;
-		read_lock(&fs->lock);
-	}
 	nd->path = fs->root;
 	path_get(&fs->root);
 	read_unlock(&fs->lock);
-	return 1;
 }
 
 /*
@@ -623,12 +611,9 @@ static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *l
 
 	if (*link == '/') {
 		path_put(&nd->path);
-		if (!walk_init_root(link, nd))
-			/* weird __emul_prefix() stuff did it */
-			goto out;
+		walk_init_root(link, nd);
 	}
 	res = link_path_walk(link, nd);
-out:
 	if (nd->depth || res || nd->last_type!=LAST_NORM)
 		return res;
 	/*
@@ -1077,67 +1062,6 @@ static int path_walk(const char *name, struct nameidata *nd)
 	return link_path_walk(name, nd);
 }
 
-/* 
- * SMP-safe: Returns 1 and nd will have valid dentry and mnt, if
- * everything is done. Returns 0 and drops input nd, if lookup failed;
- */
-static int __emul_lookup_dentry(const char *name, struct nameidata *nd)
-{
-	if (path_walk(name, nd))
-		return 0;		/* something went wrong... */
-
-	if (!nd->path.dentry->d_inode ||
-	    S_ISDIR(nd->path.dentry->d_inode->i_mode)) {
-		struct path old_path = nd->path;
-		struct qstr last = nd->last;
-		int last_type = nd->last_type;
-		struct fs_struct *fs = current->fs;
-
-		/*
-		 * NAME was not found in alternate root or it's a directory.
-		 * Try to find it in the normal root:
-		 */
-		nd->last_type = LAST_ROOT;
-		read_lock(&fs->lock);
-		nd->path = fs->root;
-		path_get(&fs->root);
-		read_unlock(&fs->lock);
-		if (path_walk(name, nd) == 0) {
-			if (nd->path.dentry->d_inode) {
-				path_put(&old_path);
-				return 1;
-			}
-			path_put(&nd->path);
-		}
-		nd->path = old_path;
-		nd->last = last;
-		nd->last_type = last_type;
-	}
-	return 1;
-}
-
-void set_fs_altroot(void)
-{
-	char *emul = __emul_prefix();
-	struct nameidata nd;
-	struct path path = {}, old_path;
-	int err;
-	struct fs_struct *fs = current->fs;
-
-	if (!emul)
-		goto set_it;
-	err = path_lookup(emul, LOOKUP_FOLLOW|LOOKUP_DIRECTORY|LOOKUP_NOALT, &nd);
-	if (!err)
-		path = nd.path;
-set_it:
-	write_lock(&fs->lock);
-	old_path = fs->altroot;
-	fs->altroot = path;
-	write_unlock(&fs->lock);
-	if (old_path.dentry)
-		path_put(&old_path);
-}
-
 /* Returns 0 and nd will be valid on success; Retuns error, otherwise. */
 static int do_path_lookup(int dfd, const char *name,
 				unsigned int flags, struct nameidata *nd)
@@ -1153,14 +1077,6 @@ static int do_path_lookup(int dfd, const char *name,
 
 	if (*name=='/') {
 		read_lock(&fs->lock);
-		if (fs->altroot.dentry && !(nd->flags & LOOKUP_NOALT)) {
-			nd->path = fs->altroot;
-			path_get(&fs->altroot);
-			read_unlock(&fs->lock);
-			if (__emul_lookup_dentry(name,nd))
-				goto out; /* found in altroot */
-			read_lock(&fs->lock);
-		}
 		nd->path = fs->root;
 		path_get(&fs->root);
 		read_unlock(&fs->lock);
@@ -1194,7 +1110,6 @@ static int do_path_lookup(int dfd, const char *name,
 	}
 
 	retval = path_walk(name, nd);
-out:
 	if (unlikely(!retval && !audit_dummy_context() && nd->path.dentry &&
 				nd->path.dentry->d_inode))
 		audit_inode(name, nd->path.dentry);
diff --git a/fs/namespace.c b/fs/namespace.c
index f30b11e2240..c4fcf48acef 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1972,7 +1972,7 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns,
 		struct fs_struct *fs)
 {
 	struct mnt_namespace *new_ns;
-	struct vfsmount *rootmnt = NULL, *pwdmnt = NULL, *altrootmnt = NULL;
+	struct vfsmount *rootmnt = NULL, *pwdmnt = NULL;
 	struct vfsmount *p, *q;
 
 	new_ns = kmalloc(sizeof(struct mnt_namespace), GFP_KERNEL);
@@ -2015,10 +2015,6 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns,
 				pwdmnt = p;
 				fs->pwd.mnt = mntget(q);
 			}
-			if (p == fs->altroot.mnt) {
-				altrootmnt = p;
-				fs->altroot.mnt = mntget(q);
-			}
 		}
 		p = next_mnt(p, mnt_ns->root);
 		q = next_mnt(q, new_ns->root);
@@ -2029,8 +2025,6 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns,
 		mntput(rootmnt);
 	if (pwdmnt)
 		mntput(pwdmnt);
-	if (altrootmnt)
-		mntput(altrootmnt);
 
 	return new_ns;
 }
diff --git a/fs/open.c b/fs/open.c
index 8e02d42bfe4..d3a2a00f52d 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -548,7 +548,7 @@ asmlinkage long sys_chroot(const char __user * filename)
 	struct nameidata nd;
 	int error;
 
-	error = __user_walk(filename, LOOKUP_FOLLOW | LOOKUP_DIRECTORY | LOOKUP_NOALT, &nd);
+	error = __user_walk(filename, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &nd);
 	if (error)
 		goto out;
 
@@ -561,7 +561,6 @@ asmlinkage long sys_chroot(const char __user * filename)
 		goto dput_and_out;
 
 	set_fs_root(current->fs, &nd.path);
-	set_fs_altroot();
 	error = 0;
 dput_and_out:
 	path_put(&nd.path);
diff --git a/include/asm-alpha/namei.h b/include/asm-alpha/namei.h
deleted file mode 100644
index 5cc9bb39499..00000000000
--- a/include/asm-alpha/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/* $Id: namei.h,v 1.1 1996/12/13 14:48:21 jj Exp $
- * linux/include/asm-alpha/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __ALPHA_NAMEI_H
-#define __ALPHA_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __ALPHA_NAMEI_H */
diff --git a/include/asm-arm/namei.h b/include/asm-arm/namei.h
deleted file mode 100644
index a402d3b9d0f..00000000000
--- a/include/asm-arm/namei.h
+++ /dev/null
@@ -1,25 +0,0 @@
-/* 
- * linux/include/asm-arm/namei.h
- *
- * Routines to handle famous /usr/gnemul
- * Derived from the Sparc version of this file
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __ASMARM_NAMEI_H
-#define __ASMARM_NAMEI_H
-
-#define ARM_BSD_EMUL "usr/gnemul/bsd/"
-
-static inline char *__emul_prefix(void)
-{
-	switch (current->personality) {
-	case PER_BSD:
-		return ARM_BSD_EMUL;
-	default:
-		return NULL;
-	}
-}
-
-#endif /* __ASMARM_NAMEI_H */
diff --git a/include/asm-avr32/namei.h b/include/asm-avr32/namei.h
deleted file mode 100644
index f0a26de06ca..00000000000
--- a/include/asm-avr32/namei.h
+++ /dev/null
@@ -1,7 +0,0 @@
-#ifndef __ASM_AVR32_NAMEI_H
-#define __ASM_AVR32_NAMEI_H
-
-/* This dummy routine may be changed to something useful */
-#define __emul_prefix() NULL
-
-#endif /* __ASM_AVR32_NAMEI_H */
diff --git a/include/asm-blackfin/namei.h b/include/asm-blackfin/namei.h
deleted file mode 100644
index 8b89a2d65cb..00000000000
--- a/include/asm-blackfin/namei.h
+++ /dev/null
@@ -1,19 +0,0 @@
-/*
- * linux/include/asm/namei.h
- *
- * Included from linux/fs/namei.c
- *
- * Changes made by Lineo Inc.    May 2001
- */
-
-#ifndef __BFIN_NAMEI_H
-#define __BFIN_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif
diff --git a/include/asm-cris/namei.h b/include/asm-cris/namei.h
deleted file mode 100644
index 8a3be7a6d9f..00000000000
--- a/include/asm-cris/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/* $Id: namei.h,v 1.1 2000/07/10 16:32:31 bjornw Exp $
- * linux/include/asm-cris/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __CRIS_NAMEI_H
-#define __CRIS_NAMEI_H
-
-/* used to find file-system prefixes for doing emulations
- * see for example asm-sparc/namei.h
- * we don't use it...
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __CRIS_NAMEI_H */
diff --git a/include/asm-frv/namei.h b/include/asm-frv/namei.h
deleted file mode 100644
index 4ea57171d95..00000000000
--- a/include/asm-frv/namei.h
+++ /dev/null
@@ -1,18 +0,0 @@
-/*
- * include/asm-frv/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __ASM_NAMEI_H
-#define __ASM_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif
-
diff --git a/include/asm-h8300/namei.h b/include/asm-h8300/namei.h
deleted file mode 100644
index ab6f196db6e..00000000000
--- a/include/asm-h8300/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/*
- * linux/include/asm-h8300/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __H8300_NAMEI_H
-#define __H8300_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif
diff --git a/include/asm-ia64/namei.h b/include/asm-ia64/namei.h
deleted file mode 100644
index 78e76807908..00000000000
--- a/include/asm-ia64/namei.h
+++ /dev/null
@@ -1,25 +0,0 @@
-#ifndef _ASM_IA64_NAMEI_H
-#define _ASM_IA64_NAMEI_H
-
-/*
- * Modified 1998, 1999, 2001
- *	David Mosberger-Tang <davidm@hpl.hp.com>, Hewlett-Packard Co
- */
-
-#include <asm/ptrace.h>
-#include <asm/system.h>
-
-#define EMUL_PREFIX_LINUX_IA32 "/emul/ia32-linux/"
-
-static inline char *
-__emul_prefix (void)
-{
-	switch (current->personality) {
-	      case PER_LINUX32:
-		return EMUL_PREFIX_LINUX_IA32;
-	      default:
-		return NULL;
-	}
-}
-
-#endif /* _ASM_IA64_NAMEI_H */
diff --git a/include/asm-m32r/namei.h b/include/asm-m32r/namei.h
deleted file mode 100644
index 210f8056b80..00000000000
--- a/include/asm-m32r/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-#ifndef _ASM_M32R_NAMEI_H
-#define _ASM_M32R_NAMEI_H
-
-/*
- * linux/include/asm-m32r/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* _ASM_M32R_NAMEI_H */
diff --git a/include/asm-m68k/namei.h b/include/asm-m68k/namei.h
deleted file mode 100644
index f33f243b644..00000000000
--- a/include/asm-m68k/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/*
- * linux/include/asm-m68k/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __M68K_NAMEI_H
-#define __M68K_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif
diff --git a/include/asm-m68knommu/namei.h b/include/asm-m68knommu/namei.h
deleted file mode 100644
index 31a85d27b93..00000000000
--- a/include/asm-m68knommu/namei.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-m68k/namei.h>
diff --git a/include/asm-mips/namei.h b/include/asm-mips/namei.h
deleted file mode 100644
index a6605a75246..00000000000
--- a/include/asm-mips/namei.h
+++ /dev/null
@@ -1,11 +0,0 @@
-#ifndef _ASM_NAMEI_H
-#define _ASM_NAMEI_H
-
-/*
- * This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* _ASM_NAMEI_H */
diff --git a/include/asm-mn10300/namei.h b/include/asm-mn10300/namei.h
deleted file mode 100644
index bd9ce94aeb6..00000000000
--- a/include/asm-mn10300/namei.h
+++ /dev/null
@@ -1,22 +0,0 @@
-/* Emulation stuff
- *
- * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public Licence
- * as published by the Free Software Foundation; either version
- * 2 of the Licence, or (at your option) any later version.
- */
-
-#ifndef _ASM_NAMEI_H
-#define _ASM_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* _ASM_NAMEI_H */
diff --git a/include/asm-parisc/namei.h b/include/asm-parisc/namei.h
deleted file mode 100644
index 8d29b3d9fb3..00000000000
--- a/include/asm-parisc/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/* $Id: namei.h,v 1.1 1996/12/13 14:48:21 jj Exp $
- * linux/include/asm-parisc/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __PARISC_NAMEI_H
-#define __PARISC_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __PARISC_NAMEI_H */
diff --git a/include/asm-powerpc/namei.h b/include/asm-powerpc/namei.h
deleted file mode 100644
index 657443474a6..00000000000
--- a/include/asm-powerpc/namei.h
+++ /dev/null
@@ -1,20 +0,0 @@
-#ifndef _ASM_POWERPC_NAMEI_H
-#define _ASM_POWERPC_NAMEI_H
-
-#ifdef __KERNEL__
-
-/*
- * Adapted from include/asm-alpha/namei.h
- *
- * Included from fs/namei.c
- */
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif	/* __KERNEL__ */
-#endif	/* _ASM_POWERPC_NAMEI_H */
diff --git a/include/asm-s390/namei.h b/include/asm-s390/namei.h
deleted file mode 100644
index 3e286bdde4b..00000000000
--- a/include/asm-s390/namei.h
+++ /dev/null
@@ -1,21 +0,0 @@
-/*
- *  include/asm-s390/namei.h
- *
- *  S390 version
- *
- *  Derived from "include/asm-i386/namei.h"
- *
- *  Included from linux/fs/namei.c
- */
-
-#ifndef __S390_NAMEI_H
-#define __S390_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __S390_NAMEI_H */
diff --git a/include/asm-sh/namei.h b/include/asm-sh/namei.h
deleted file mode 100644
index 338a5d94714..00000000000
--- a/include/asm-sh/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/* $Id: namei.h,v 1.3 2000/07/04 06:24:49 gniibe Exp $
- * linux/include/asm-sh/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __ASM_SH_NAMEI_H
-#define __ASM_SH_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __ASM_SH_NAMEI_H */
diff --git a/include/asm-sparc/namei.h b/include/asm-sparc/namei.h
deleted file mode 100644
index eff944b8e32..00000000000
--- a/include/asm-sparc/namei.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef ___ASM_SPARC_NAMEI_H
-#define ___ASM_SPARC_NAMEI_H
-#if defined(__sparc__) && defined(__arch64__)
-#include <asm-sparc/namei_64.h>
-#else
-#include <asm-sparc/namei_32.h>
-#endif
-#endif
diff --git a/include/asm-sparc64/namei.h b/include/asm-sparc64/namei.h
deleted file mode 100644
index 1344a910ba2..00000000000
--- a/include/asm-sparc64/namei.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-sparc/namei.h>
diff --git a/include/asm-um/namei.h b/include/asm-um/namei.h
deleted file mode 100644
index 002984d5bc8..00000000000
--- a/include/asm-um/namei.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __UM_NAMEI_H
-#define __UM_NAMEI_H
-
-#include "asm/arch/namei.h"
-
-#endif
diff --git a/include/asm-v850/namei.h b/include/asm-v850/namei.h
deleted file mode 100644
index ee8339b2384..00000000000
--- a/include/asm-v850/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/*
- * linux/include/asm-v850/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __V850_NAMEI_H__
-#define __V850_NAMEI_H__
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __V850_NAMEI_H__ */
diff --git a/include/asm-x86/namei.h b/include/asm-x86/namei.h
deleted file mode 100644
index 415ef5d9550..00000000000
--- a/include/asm-x86/namei.h
+++ /dev/null
@@ -1,11 +0,0 @@
-#ifndef _ASM_X86_NAMEI_H
-#define _ASM_X86_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* _ASM_X86_NAMEI_H */
diff --git a/include/asm-xtensa/namei.h b/include/asm-xtensa/namei.h
deleted file mode 100644
index 3fdff039d27..00000000000
--- a/include/asm-xtensa/namei.h
+++ /dev/null
@@ -1,26 +0,0 @@
-/*
- * include/asm-xtensa/namei.h
- *
- * Included from linux/fs/namei.c
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2001 - 2005 Tensilica Inc.
- */
-
-#ifndef _XTENSA_NAMEI_H
-#define _XTENSA_NAMEI_H
-
-#ifdef __KERNEL__
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif	/* __KERNEL__ */
-#endif	/* _XTENSA_NAMEI_H */
diff --git a/include/linux/fs_struct.h b/include/linux/fs_struct.h
index 282f5421912..9e5a06e78d0 100644
--- a/include/linux/fs_struct.h
+++ b/include/linux/fs_struct.h
@@ -7,7 +7,7 @@ struct fs_struct {
 	atomic_t count;
 	rwlock_t lock;
 	int umask;
-	struct path root, pwd, altroot;
+	struct path root, pwd;
 };
 
 #define INIT_FS {				\
@@ -19,7 +19,6 @@ struct fs_struct {
 extern struct kmem_cache *fs_cachep;
 
 extern void exit_fs(struct task_struct *);
-extern void set_fs_altroot(void);
 extern void set_fs_root(struct fs_struct *, struct path *);
 extern void set_fs_pwd(struct fs_struct *, struct path *);
 extern struct fs_struct *copy_fs_struct(struct fs_struct *);
diff --git a/include/linux/namei.h b/include/linux/namei.h
index 3cf62d26d49..768773d5785 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -47,7 +47,6 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND};
 #define LOOKUP_DIRECTORY	 2
 #define LOOKUP_CONTINUE		 4
 #define LOOKUP_PARENT		16
-#define LOOKUP_NOALT		32
 #define LOOKUP_REVAL		64
 /*
  * Intent data
diff --git a/kernel/exec_domain.c b/kernel/exec_domain.c
index c1ef192aa65..0d407e88673 100644
--- a/kernel/exec_domain.c
+++ b/kernel/exec_domain.c
@@ -168,7 +168,6 @@ __set_personality(u_long personality)
 	current->personality = personality;
 	oep = current_thread_info()->exec_domain;
 	current_thread_info()->exec_domain = ep;
-	set_fs_altroot();
 
 	module_put(oep->module);
 	return 0;
diff --git a/kernel/exit.c b/kernel/exit.c
index 6cdf60712bd..0caf590548a 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -565,8 +565,6 @@ void put_fs_struct(struct fs_struct *fs)
 	if (atomic_dec_and_test(&fs->count)) {
 		path_put(&fs->root);
 		path_put(&fs->pwd);
-		if (fs->altroot.dentry)
-			path_put(&fs->altroot);
 		kmem_cache_free(fs_cachep, fs);
 	}
 }
diff --git a/kernel/fork.c b/kernel/fork.c
index abb3ed6298f..5e050c1317c 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -657,13 +657,6 @@ static struct fs_struct *__copy_fs_struct(struct fs_struct *old)
 		path_get(&old->root);
 		fs->pwd = old->pwd;
 		path_get(&old->pwd);
-		if (old->altroot.dentry) {
-			fs->altroot = old->altroot;
-			path_get(&old->altroot);
-		} else {
-			fs->altroot.mnt = NULL;
-			fs->altroot.dentry = NULL;
-		}
 		read_unlock(&old->lock);
 	}
 	return fs;
-- 
cgit v1.2.3-70-g09d2


From 5995477ab7f3522c497c9c4a1c55373e9d655574 Mon Sep 17 00:00:00 2001
From: Andrea Righi <righi.andrea@gmail.com>
Date: Sun, 27 Jul 2008 17:29:15 +0200
Subject: task IO accounting: improve code readability

Put all i/o statistics in struct proc_io_accounting and use inline functions to
initialize and increment statistics, removing a lot of single variable
assignments.

This also reduces the kernel size as following (with CONFIG_TASK_XACCT=y and
CONFIG_TASK_IO_ACCOUNTING=y).

    text    data     bss     dec     hex filename
   11651       0       0   11651    2d83 kernel/exit.o.before
   11619       0       0   11619    2d63 kernel/exit.o.after
   10886     132     136   11154    2b92 kernel/fork.o.before
   10758     132     136   11026    2b12 kernel/fork.o.after

 3082029  807968 4818600 8708597  84e1f5 vmlinux.o.before
 3081869  807968 4818600 8708437  84e155 vmlinux.o.after

Signed-off-by: Andrea Righi <righi.andrea@gmail.com>
Acked-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/base.c                         | 57 ++++++++++------------------------
 include/linux/sched.h                  | 19 ++++--------
 include/linux/task_io_accounting.h     | 27 ++++++++++++++--
 include/linux/task_io_accounting_ops.h | 56 +++++++++++++++++++++++++++------
 kernel/exit.c                          | 30 ++----------------
 kernel/fork.c                          | 15 ++-------
 kernel/tsacct.c                        | 14 ++++-----
 7 files changed, 104 insertions(+), 114 deletions(-)

(limited to 'kernel/exit.c')

diff --git a/fs/proc/base.c b/fs/proc/base.c
index e74308bdabd..3d94906c7aa 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -53,6 +53,7 @@
 #include <linux/time.h>
 #include <linux/proc_fs.h>
 #include <linux/stat.h>
+#include <linux/task_io_accounting_ops.h>
 #include <linux/init.h>
 #include <linux/capability.h>
 #include <linux/file.h>
@@ -2402,44 +2403,17 @@ static int proc_base_fill_cache(struct file *filp, void *dirent,
 #ifdef CONFIG_TASK_IO_ACCOUNTING
 static int do_io_accounting(struct task_struct *task, char *buffer, int whole)
 {
-	u64 rchar, wchar, syscr, syscw;
-	struct task_io_accounting ioac;
-
-	rchar = task->rchar;
-	wchar = task->wchar;
-	syscr = task->syscr;
-	syscw = task->syscw;
-	memcpy(&ioac, &task->ioac, sizeof(ioac));
-
-	if (whole) {
-		unsigned long flags;
-
-		if (lock_task_sighand(task, &flags)) {
-			struct signal_struct *sig = task->signal;
-			struct task_struct *t = task;
-
-			rchar += sig->rchar;
-			wchar += sig->wchar;
-			syscr += sig->syscr;
-			syscw += sig->syscw;
-
-			ioac.read_bytes += sig->ioac.read_bytes;
-			ioac.write_bytes += sig->ioac.write_bytes;
-			ioac.cancelled_write_bytes +=
-					sig->ioac.cancelled_write_bytes;
-			while_each_thread(task, t) {
-				rchar += t->rchar;
-				wchar += t->wchar;
-				syscr += t->syscr;
-				syscw += t->syscw;
-
-				ioac.read_bytes += t->ioac.read_bytes;
-				ioac.write_bytes += t->ioac.write_bytes;
-				ioac.cancelled_write_bytes +=
-					t->ioac.cancelled_write_bytes;
-			}
-			unlock_task_sighand(task, &flags);
-		}
+	struct proc_io_accounting acct = task->ioac;
+	unsigned long flags;
+
+	if (whole && lock_task_sighand(task, &flags)) {
+		struct task_struct *t = task;
+
+		task_io_accounting_add(&acct, &task->signal->ioac);
+		while_each_thread(task, t)
+			task_io_accounting_add(&acct, &t->ioac);
+
+		unlock_task_sighand(task, &flags);
 	}
 	return sprintf(buffer,
 			"rchar: %llu\n"
@@ -2449,9 +2423,10 @@ static int do_io_accounting(struct task_struct *task, char *buffer, int whole)
 			"read_bytes: %llu\n"
 			"write_bytes: %llu\n"
 			"cancelled_write_bytes: %llu\n",
-			rchar, wchar, syscr, syscw,
-			ioac.read_bytes, ioac.write_bytes,
-			ioac.cancelled_write_bytes);
+			acct.chr.rchar, acct.chr.wchar,
+			acct.chr.syscr, acct.chr.syscw,
+			acct.blk.read_bytes, acct.blk.write_bytes,
+			acct.blk.cancelled_write_bytes);
 }
 
 static int proc_tid_io_accounting(struct task_struct *task, char *buffer)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index f59318a0099..034c1ca6b33 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -505,10 +505,7 @@ struct signal_struct {
 	unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw;
 	unsigned long min_flt, maj_flt, cmin_flt, cmaj_flt;
 	unsigned long inblock, oublock, cinblock, coublock;
-#ifdef CONFIG_TASK_XACCT
-	u64 rchar, wchar, syscr, syscw;
-#endif
-	struct task_io_accounting ioac;
+	struct proc_io_accounting ioac;
 
 	/*
 	 * Cumulative ns of scheduled CPU time for dead threads in the
@@ -1256,11 +1253,7 @@ struct task_struct {
 
 	unsigned long ptrace_message;
 	siginfo_t *last_siginfo; /* For ptrace use.  */
-#ifdef CONFIG_TASK_XACCT
-/* i/o counters(bytes read/written, #syscalls */
-	u64 rchar, wchar, syscr, syscw;
-#endif
-	struct task_io_accounting ioac;
+	struct proc_io_accounting ioac;
 #if defined(CONFIG_TASK_XACCT)
 	u64 acct_rss_mem1;	/* accumulated rss usage */
 	u64 acct_vm_mem1;	/* accumulated virtual memory usage */
@@ -2190,22 +2183,22 @@ extern long sched_group_rt_period(struct task_group *tg);
 #ifdef CONFIG_TASK_XACCT
 static inline void add_rchar(struct task_struct *tsk, ssize_t amt)
 {
-	tsk->rchar += amt;
+	tsk->ioac.chr.rchar += amt;
 }
 
 static inline void add_wchar(struct task_struct *tsk, ssize_t amt)
 {
-	tsk->wchar += amt;
+	tsk->ioac.chr.wchar += amt;
 }
 
 static inline void inc_syscr(struct task_struct *tsk)
 {
-	tsk->syscr++;
+	tsk->ioac.chr.syscr++;
 }
 
 static inline void inc_syscw(struct task_struct *tsk)
 {
-	tsk->syscw++;
+	tsk->ioac.chr.syscw++;
 }
 #else
 static inline void add_rchar(struct task_struct *tsk, ssize_t amt)
diff --git a/include/linux/task_io_accounting.h b/include/linux/task_io_accounting.h
index 44d00e9ccee..165390f8b93 100644
--- a/include/linux/task_io_accounting.h
+++ b/include/linux/task_io_accounting.h
@@ -1,5 +1,5 @@
 /*
- * task_io_accounting: a structure which is used for recording a single task's
+ * proc_io_accounting: a structure which is used for recording a single task's
  * IO statistics.
  *
  * Don't include this header file directly - it is designed to be dragged in via
@@ -8,6 +8,22 @@
  * Blame akpm@osdl.org for all this.
  */
 
+#ifdef CONFIG_TASK_XACCT
+struct task_chr_io_accounting {
+	/* bytes read */
+	u64 rchar;
+	/*  bytes written */
+	u64 wchar;
+	/* # of read syscalls */
+	u64 syscr;
+	/* # of write syscalls */
+	u64 syscw;
+};
+#else /* CONFIG_TASK_XACCT */
+struct task_chr_io_accounting {
+};
+#endif /* CONFIG_TASK_XACCT */
+
 #ifdef CONFIG_TASK_IO_ACCOUNTING
 struct task_io_accounting {
 	/*
@@ -31,7 +47,12 @@ struct task_io_accounting {
 	 */
 	u64 cancelled_write_bytes;
 };
-#else
+#else /* CONFIG_TASK_IO_ACCOUNTING */
 struct task_io_accounting {
 };
-#endif
+#endif /* CONFIG_TASK_IO_ACCOUNTING */
+
+struct proc_io_accounting {
+	struct task_chr_io_accounting chr;
+	struct task_io_accounting blk;
+};
diff --git a/include/linux/task_io_accounting_ops.h b/include/linux/task_io_accounting_ops.h
index ff46c6fad79..e6f958ebe97 100644
--- a/include/linux/task_io_accounting_ops.h
+++ b/include/linux/task_io_accounting_ops.h
@@ -9,7 +9,7 @@
 #ifdef CONFIG_TASK_IO_ACCOUNTING
 static inline void task_io_account_read(size_t bytes)
 {
-	current->ioac.read_bytes += bytes;
+	current->ioac.blk.read_bytes += bytes;
 }
 
 /*
@@ -18,12 +18,12 @@ static inline void task_io_account_read(size_t bytes)
  */
 static inline unsigned long task_io_get_inblock(const struct task_struct *p)
 {
-	return p->ioac.read_bytes >> 9;
+	return p->ioac.blk.read_bytes >> 9;
 }
 
 static inline void task_io_account_write(size_t bytes)
 {
-	current->ioac.write_bytes += bytes;
+	current->ioac.blk.write_bytes += bytes;
 }
 
 /*
@@ -32,17 +32,25 @@ static inline void task_io_account_write(size_t bytes)
  */
 static inline unsigned long task_io_get_oublock(const struct task_struct *p)
 {
-	return p->ioac.write_bytes >> 9;
+	return p->ioac.blk.write_bytes >> 9;
 }
 
 static inline void task_io_account_cancelled_write(size_t bytes)
 {
-	current->ioac.cancelled_write_bytes += bytes;
+	current->ioac.blk.cancelled_write_bytes += bytes;
 }
 
-static inline void task_io_accounting_init(struct task_struct *tsk)
+static inline void task_io_accounting_init(struct proc_io_accounting *ioac)
 {
-	memset(&tsk->ioac, 0, sizeof(tsk->ioac));
+	memset(ioac, 0, sizeof(*ioac));
+}
+
+static inline void task_blk_io_accounting_add(struct proc_io_accounting *dst,
+						struct proc_io_accounting *src)
+{
+	dst->blk.read_bytes += src->blk.read_bytes;
+	dst->blk.write_bytes += src->blk.write_bytes;
+	dst->blk.cancelled_write_bytes += src->blk.cancelled_write_bytes;
 }
 
 #else
@@ -69,9 +77,37 @@ static inline void task_io_account_cancelled_write(size_t bytes)
 {
 }
 
-static inline void task_io_accounting_init(struct task_struct *tsk)
+static inline void task_io_accounting_init(struct proc_io_accounting *ioac)
+{
+}
+
+static inline void task_blk_io_accounting_add(struct proc_io_accounting *dst,
+						struct proc_io_accounting *src)
 {
 }
 
-#endif		/* CONFIG_TASK_IO_ACCOUNTING */
-#endif		/* __TASK_IO_ACCOUNTING_OPS_INCLUDED */
+#endif /* CONFIG_TASK_IO_ACCOUNTING */
+
+#ifdef CONFIG_TASK_XACCT
+static inline void task_chr_io_accounting_add(struct proc_io_accounting *dst,
+						struct proc_io_accounting *src)
+{
+	dst->chr.rchar += src->chr.rchar;
+	dst->chr.wchar += src->chr.wchar;
+	dst->chr.syscr += src->chr.syscr;
+	dst->chr.syscw += src->chr.syscw;
+}
+#else
+static inline void task_chr_io_accounting_add(struct proc_io_accounting *dst,
+						struct proc_io_accounting *src)
+{
+}
+#endif /* CONFIG_TASK_XACCT */
+
+static inline void task_io_accounting_add(struct proc_io_accounting *dst,
+						struct proc_io_accounting *src)
+{
+	task_chr_io_accounting_add(dst, src);
+	task_blk_io_accounting_add(dst, src);
+}
+#endif /* __TASK_IO_ACCOUNTING_OPS_INCLUDED */
diff --git a/kernel/exit.c b/kernel/exit.c
index 0caf590548a..eb4d6470d1d 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -121,18 +121,7 @@ static void __exit_signal(struct task_struct *tsk)
 		sig->nivcsw += tsk->nivcsw;
 		sig->inblock += task_io_get_inblock(tsk);
 		sig->oublock += task_io_get_oublock(tsk);
-#ifdef CONFIG_TASK_XACCT
-		sig->rchar += tsk->rchar;
-		sig->wchar += tsk->wchar;
-		sig->syscr += tsk->syscr;
-		sig->syscw += tsk->syscw;
-#endif /* CONFIG_TASK_XACCT */
-#ifdef CONFIG_TASK_IO_ACCOUNTING
-		sig->ioac.read_bytes += tsk->ioac.read_bytes;
-		sig->ioac.write_bytes += tsk->ioac.write_bytes;
-		sig->ioac.cancelled_write_bytes +=
-					tsk->ioac.cancelled_write_bytes;
-#endif /* CONFIG_TASK_IO_ACCOUNTING */
+		task_io_accounting_add(&sig->ioac, &tsk->ioac);
 		sig->sum_sched_runtime += tsk->se.sum_exec_runtime;
 		sig = NULL; /* Marker for below. */
 	}
@@ -1363,21 +1352,8 @@ static int wait_task_zombie(struct task_struct *p, int options,
 		psig->coublock +=
 			task_io_get_oublock(p) +
 			sig->oublock + sig->coublock;
-#ifdef CONFIG_TASK_XACCT
-		psig->rchar += p->rchar + sig->rchar;
-		psig->wchar += p->wchar + sig->wchar;
-		psig->syscr += p->syscr + sig->syscr;
-		psig->syscw += p->syscw + sig->syscw;
-#endif /* CONFIG_TASK_XACCT */
-#ifdef CONFIG_TASK_IO_ACCOUNTING
-		psig->ioac.read_bytes +=
-			p->ioac.read_bytes + sig->ioac.read_bytes;
-		psig->ioac.write_bytes +=
-			p->ioac.write_bytes + sig->ioac.write_bytes;
-		psig->ioac.cancelled_write_bytes +=
-				p->ioac.cancelled_write_bytes +
-				sig->ioac.cancelled_write_bytes;
-#endif /* CONFIG_TASK_IO_ACCOUNTING */
+		task_io_accounting_add(&psig->ioac, &p->ioac);
+		task_io_accounting_add(&psig->ioac, &sig->ioac);
 		spin_unlock_irq(&p->parent->sighand->siglock);
 	}
 
diff --git a/kernel/fork.c b/kernel/fork.c
index 5e050c1317c..8214ba7c8bb 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -806,12 +806,7 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
 	sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0;
 	sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0;
 	sig->inblock = sig->oublock = sig->cinblock = sig->coublock = 0;
-#ifdef CONFIG_TASK_XACCT
-	sig->rchar = sig->wchar = sig->syscr = sig->syscw = 0;
-#endif
-#ifdef CONFIG_TASK_IO_ACCOUNTING
-	memset(&sig->ioac, 0, sizeof(sig->ioac));
-#endif
+	task_io_accounting_init(&sig->ioac);
 	sig->sum_sched_runtime = 0;
 	INIT_LIST_HEAD(&sig->cpu_timers[0]);
 	INIT_LIST_HEAD(&sig->cpu_timers[1]);
@@ -994,13 +989,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	p->last_switch_timestamp = 0;
 #endif
 
-#ifdef CONFIG_TASK_XACCT
-	p->rchar = 0;		/* I/O counter: bytes read */
-	p->wchar = 0;		/* I/O counter: bytes written */
-	p->syscr = 0;		/* I/O counter: read syscalls */
-	p->syscw = 0;		/* I/O counter: write syscalls */
-#endif
-	task_io_accounting_init(p);
+	task_io_accounting_init(&p->ioac);
 	acct_clear_integrals(p);
 
 	p->it_virt_expires = cputime_zero;
diff --git a/kernel/tsacct.c b/kernel/tsacct.c
index 3da47ccdc5e..f9cd2561689 100644
--- a/kernel/tsacct.c
+++ b/kernel/tsacct.c
@@ -94,14 +94,14 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p)
 		stats->hiwater_vm    = mm->hiwater_vm * PAGE_SIZE / KB;
 		mmput(mm);
 	}
-	stats->read_char	= p->rchar;
-	stats->write_char	= p->wchar;
-	stats->read_syscalls	= p->syscr;
-	stats->write_syscalls	= p->syscw;
+	stats->read_char	= p->ioac.chr.rchar;
+	stats->write_char	= p->ioac.chr.wchar;
+	stats->read_syscalls	= p->ioac.chr.syscr;
+	stats->write_syscalls	= p->ioac.chr.syscw;
 #ifdef CONFIG_TASK_IO_ACCOUNTING
-	stats->read_bytes	= p->ioac.read_bytes;
-	stats->write_bytes	= p->ioac.write_bytes;
-	stats->cancelled_write_bytes = p->ioac.cancelled_write_bytes;
+	stats->read_bytes	= p->ioac.blk.read_bytes;
+	stats->write_bytes	= p->ioac.blk.write_bytes;
+	stats->cancelled_write_bytes = p->ioac.blk.cancelled_write_bytes;
 #else
 	stats->read_bytes	= 0;
 	stats->write_bytes	= 0;
-- 
cgit v1.2.3-70-g09d2