cgroup: delay the clearing of cgrp->kn->priv

Run these two scripts concurrently:

    for ((; ;))
    {
        mkdir /cgroup/sub
        rmdir /cgroup/sub
    }

    for ((; ;))
    {
        echo $$ > /cgroup/sub/cgroup.procs
        echo $$ > /cgroup/cgroup.procs
    }

A kernel bug will be triggered:

BUG: unable to handle kernel NULL pointer dereference at 00000038
IP: [<c10bbd69>] cgroup_put+0x9/0x80
...
Call Trace:
 [<c10bbe19>] cgroup_kn_unlock+0x39/0x50
 [<c10bbe91>] cgroup_kn_lock_live+0x61/0x70
 [<c10be3c1>] __cgroup_procs_write.isra.26+0x51/0x230
 [<c10be5b2>] cgroup_tasks_write+0x12/0x20
 [<c10bb7b0>] cgroup_file_write+0x40/0x130
 [<c11aee71>] kernfs_fop_write+0xd1/0x160
 [<c1148e58>] vfs_write+0x98/0x1e0
 [<c114934d>] SyS_write+0x4d/0xa0
 [<c16f656b>] sysenter_do_call+0x12/0x12

We clear cgrp->kn->priv in the end of cgroup_rmdir(), but another concurrent thread can access kn->priv after the clearing.

We should move the clearing to css_release_work_fn(). At that time no one is holding reference to the cgroup and no one can gain a new reference to access it.

v2:
- remove RCU_INIT_POINTER() into the else block. (Tejun)
- remove the cgroup_parent() check. (Tejun)
- update the comment in css_tryget_online_from_dir().

---
 kernel/cgroup.c | 21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

index 1c56924..205f793 100644
@@ -4181,6 +4181,15 @@ static void css_release_work_fn(struct work_struct *work)
 		/* cgroup release path */
 		cgroup_idr_remove(&cgrp->root->cgroup_idr, cgrp->id);
 		cgrp->id = -1;
+
+		/*
+		 * There are two control paths which try to determine
+		 * cgroup from dentry without going through kernfs -
+		 * cgroupstats_build() and css_tryget_online_from_dir().
+		 * Those are supported by RCU protecting clearing of
+		 * cgrp->kn->priv backpointer.
+		 */
+		RCU_INIT_POINTER(*(void __rcu __force **)&cgrp->kn->priv, NULL);
 	}
 
 	mutex_unlock(&cgroup_mutex);

@@ -4601,16 +4610,6 @@ static int cgroup_rmdir(struct kernfs_node *kn)
 
 	cgroup_kn_unlock(kn);
 
-	/*
-	 * There are two control paths which try to determine cgroup from
-	 * dentry without going through kernfs - cgroupstats_build() and
-	 * css_tryget_online_from_dir().  Those are supported by RCU
-	 * protecting clearing of cgrp->kn->priv backpointer, which should
-	 * happen after all files under it have been removed.
-	 */
-	if (!ret)
-		RCU_INIT_POINTER(*(void __rcu __force **)&kn->priv, NULL);
-
 	cgroup_put(cgrp);
 	return ret;
 }

@@ -5175,7 +5174,7 @@ struct cgroup_subsys_state *css_tryget_online_from_dir(struct dentry *dentry,
 	/*
 	 * This path doesn't originate from kernfs and @kn could already
 	 * have been or be removed at any point.  @kn->priv is RCU
-	 * protected for this access.  See cgroup_rmdir() for details.
+	 * protected for this access.  See css_release_work_fn() for details.
 	 */
 	cgrp = rcu_dereference(kn->priv);
 	if (cgrp)

1.8.0.2

When cgroup_kn_lock_live() is called through some kernfs operation and another thread is calling cgroup_rmdir(), we'll trigger the warning in cgroup_get().

------------[ cut here ]------------
WARNING: CPU: 1 PID: 1228 at kernel/cgroup.c:1034 cgroup_get+0x89/0xa0()
...
Call Trace:
 [<c16ee73d>] dump_stack+0x41/0x52
 [<c10468ef>] warn_slowpath_common+0x7f/0xa0
 [<c104692d>] warn_slowpath_null+0x1d/0x20
 [<c10bb999>] cgroup_get+0x89/0xa0
 [<c10bbe58>] cgroup_kn_lock_live+0x28/0x70
 [<c10be3c1>] __cgroup_procs_write.isra.26+0x51/0x230
 [<c10be5b2>] cgroup_tasks_write+0x12/0x20
 [<c10bb7b0>] cgroup_file_write+0x40/0x130
 [<c11aee71>] kernfs_fop_write+0xd1/0x160
 [<c1148e58>] vfs_write+0x98/0x1e0
 [<c114934d>] SyS_write+0x4d/0xa0
 [<c16f656b>] sysenter_do_call+0x12/0x12
---[ end trace 6f2e0c38c2108a74 ]---

Fix this by calling css_tryget() instead of cgroup_get().

v2:
- move cgroup_tryget() right below cgroup_get() definition. (Tejun)

---
 kernel/cgroup.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

index 709a6a0..51dd46e 100644
@@ -1031,6 +1031,11 @@ static void cgroup_get(struct cgroup *cgrp)
 	css_get(&cgrp->self);
 }
 
+static bool cgroup_tryget(struct cgroup *cgrp)
+{
+	return css_tryget(&cgrp->self);
+}
+
 static void cgroup_put(struct cgroup *cgrp)
 {
 	css_put(&cgrp->self);

@@ -1091,7 +1096,8 @@ static struct cgroup *cgroup_kn_lock_live(struct kernfs_node *kn)
 	 * protection against removal.  Ensure @cgrp stays accessible and
 	 * break the active_ref protection.
 	 */
-	cgroup_get(cgrp);
+	if (!cgroup_tryget(cgrp))
+		return NULL;
 	kernfs_break_active_protection(kn);
 
 	mutex_lock(&cgroup_mutex);

1.8.0.2

Applied 1-2 to cgroup/for-3.17-fixes.

Thanks.

tejun