cgroups: add an owner to the mm_struct

Remove the mem_cgroup member from mm_struct and instead adds an owner.

This approach was suggested by Paul Menage.  The advantage of this approach
is that, once the mm->owner is known, using the subsystem id, the cgroup
can be determined.  It also allows several control groups that are
virtually grouped by mm_struct, to exist independent of the memory
controller i.e., without adding mem_cgroup's for each controller, to
mm_struct.

A new config option CONFIG_MM_OWNER is added and the memory resource
controller selects this config option.

This patch also adds cgroup callbacks to notify subsystems when mm->owner
changes.  The mm_cgroup_changed callback is called with the task_lock() of
the new task held and is called just prior to changing the mm->owner.

I am indebted to Paul Menage for the several reviews of this patchset and
helping me make it lighter and simpler.

This patch was tested on a powerpc box, it was compiled with both the
MM_OWNER config turned on and off.

After the thread group leader exits, it's moved to init_css_state by
cgroup_exit(), thus all future charges from runnings threads would be
redirected to the init_css_set's subsystem.

Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Pavel Emelianov <xemul@openvz.org>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Sudhir Kumar <skumar@linux.vnet.ibm.com>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: Hirokazu Takahashi <taka@valinux.co.jp>
Cc: David Rientjes <rientjes@google.com>,
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Reviewed-by: Paul Menage <menage@google.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

3 files changed, 121 insertions, 3 deletions

diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index abc433772e5a..b9d467d83fc1 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -119,6 +119,7 @@ static int root_count;
  * be called.
  */
 static int need_forkexit_callback;
+static int need_mm_owner_callback __read_mostly;
 
 /* convenient tests for these bits */
 inline int cgroup_is_removed(const struct cgroup *cgrp)
@@ -2498,6 +2499,7 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss)
 	init_css_set.subsys[ss->subsys_id] = dummytop->subsys[ss->subsys_id];
 
 	need_forkexit_callback |= ss->fork || ss->exit;
+	need_mm_owner_callback |= !!ss->mm_owner_changed;
 
 	/* At system boot, before all subsystems have been
 	 * registered, no tasks have been forked, so we don't
@@ -2748,6 +2750,34 @@ void cgroup_fork_callbacks(struct task_struct *child)
 	}
 }
 
+#ifdef CONFIG_MM_OWNER
+/**
+ * cgroup_mm_owner_callbacks - run callbacks when the mm->owner changes
+ * @p: the new owner
+ *
+ * Called on every change to mm->owner. mm_init_owner() does not
+ * invoke this routine, since it assigns the mm->owner the first time
+ * and does not change it.
+ */
+void cgroup_mm_owner_callbacks(struct task_struct *old, struct task_struct *new)
+{
+	struct cgroup *oldcgrp, *newcgrp;
+
+	if (need_mm_owner_callback) {
+		int i;
+		for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) {
+			struct cgroup_subsys *ss = subsys[i];
+			oldcgrp = task_cgroup(old, ss->subsys_id);
+			newcgrp = task_cgroup(new, ss->subsys_id);
+			if (oldcgrp == newcgrp)
+				continue;
+			if (ss->mm_owner_changed)
+				ss->mm_owner_changed(ss, oldcgrp, newcgrp);
+		}
+	}
+}
+#endif /* CONFIG_MM_OWNER */
+
 /**
  * cgroup_post_fork - called on a new task after adding it to the task list
  * @child: the task in question
diff --git a/kernel/exit.c b/kernel/exit.c
index 2a9d98c641ac..ae0f2c4e452b 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -557,6 +557,88 @@ void exit_fs(struct task_struct *tsk)
 
 EXPORT_SYMBOL_GPL(exit_fs);
 
+#ifdef CONFIG_MM_OWNER
+/*
+ * Task p is exiting and it owned mm, lets find a new owner for it
+ */
+static inline int
+mm_need_new_owner(struct mm_struct *mm, struct task_struct *p)
+{
+	/*
+	 * If there are other users of the mm and the owner (us) is exiting
+	 * we need to find a new owner to take on the responsibility.
+	 */
+	if (!mm)
+		return 0;
+	if (atomic_read(&mm->mm_users) <= 1)
+		return 0;
+	if (mm->owner != p)
+		return 0;
+	return 1;
+}
+
+void mm_update_next_owner(struct mm_struct *mm)
+{
+	struct task_struct *c, *g, *p = current;
+
+retry:
+	if (!mm_need_new_owner(mm, p))
+		return;
+
+	read_lock(&tasklist_lock);
+	/*
+	 * Search in the children
+	 */
+	list_for_each_entry(c, &p->children, sibling) {
+		if (c->mm == mm)
+			goto assign_new_owner;
+	}
+
+	/*
+	 * Search in the siblings
+	 */
+	list_for_each_entry(c, &p->parent->children, sibling) {
+		if (c->mm == mm)
+			goto assign_new_owner;
+	}
+
+	/*
+	 * Search through everything else. We should not get
+	 * here often
+	 */
+	do_each_thread(g, c) {
+		if (c->mm == mm)
+			goto assign_new_owner;
+	} while_each_thread(g, c);
+
+	read_unlock(&tasklist_lock);
+	return;
+
+assign_new_owner:
+	BUG_ON(c == p);
+	get_task_struct(c);
+	/*
+	 * The task_lock protects c->mm from changing.
+	 * We always want mm->owner->mm == mm
+	 */
+	task_lock(c);
+	/*
+	 * Delay read_unlock() till we have the task_lock()
+	 * to ensure that c does not slip away underneath us
+	 */
+	read_unlock(&tasklist_lock);
+	if (c->mm != mm) {
+		task_unlock(c);
+		put_task_struct(c);
+		goto retry;
+	}
+	cgroup_mm_owner_callbacks(mm->owner, c);
+	mm->owner = c;
+	task_unlock(c);
+	put_task_struct(c);
+}
+#endif /* CONFIG_MM_OWNER */
+
 /*
  * Turn us into a lazy TLB process if we
  * aren't already..
@@ -596,6 +678,7 @@ static void exit_mm(struct task_struct * tsk)
 	/* We don't want this task to be frozen prematurely */
 	clear_freeze_flag(tsk);
 	task_unlock(tsk);
+	mm_update_next_owner(mm);
 	mmput(mm);
 }
 
diff --git a/kernel/fork.c b/kernel/fork.c
index 6067e429f281..156db96ff754 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -381,14 +381,13 @@ static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p)
 	mm->ioctx_list = NULL;
 	mm->free_area_cache = TASK_UNMAPPED_BASE;
 	mm->cached_hole_size = ~0UL;
-	mm_init_cgroup(mm, p);
+	mm_init_owner(mm, p);
 
 	if (likely(!mm_alloc_pgd(mm))) {
 		mm->def_flags = 0;
 		return mm;
 	}
 
-	mm_free_cgroup(mm);
 	free_mm(mm);
 	return NULL;
 }
@@ -438,7 +437,6 @@ void mmput(struct mm_struct *mm)
 			spin_unlock(&mmlist_lock);
 		}
 		put_swap_token(mm);
-		mm_free_cgroup(mm);
 		mmdrop(mm);
 	}
 }
@@ -982,6 +980,13 @@ static void rt_mutex_init_task(struct task_struct *p)
 #endif
 }
 
+#ifdef CONFIG_MM_OWNER
+void mm_init_owner(struct mm_struct *mm, struct task_struct *p)
+{
+	mm->owner = p;
+}
+#endif /* CONFIG_MM_OWNER */
+
 /*
  * This creates a new process as a copy of the old one,
  * but does not actually start it yet.