| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <20220913140357.359330349@linuxfoundation.org>
Date: Tue, 13 Sep 2022 16:06:58 +0200
From: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To: linux-kernel@...r.kernel.org
Cc: Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
stable@...r.kernel.org, Tejun Heo <tj@...nel.org>,
Sasha Levin <sashal@...nel.org>,
Imran Khan <imran.f.khan@...cle.com>,
Xuewen Yan <xuewen.yan@...soc.com>
Subject: [PATCH 5.4 087/108] cgroup: Fix threadgroup_rwsem <-> cpus_read_lock() deadlock
From: Tejun Heo <tj@...nel.org>
[ Upstream commit 4f7e7236435ca0abe005c674ebd6892c6e83aeb3 ]
Bringing up a CPU may involve creating and destroying tasks which requires
read-locking threadgroup_rwsem, so threadgroup_rwsem nests inside
cpus_read_lock(). However, cpuset's ->attach(), which may be called with
thredagroup_rwsem write-locked, also wants to disable CPU hotplug and
acquires cpus_read_lock(), leading to a deadlock.
Fix it by guaranteeing that ->attach() is always called with CPU hotplug
disabled and removing cpus_read_lock() call from cpuset_attach().
Signed-off-by: Tejun Heo <tj@...nel.org>
Reviewed-and-tested-by: Imran Khan <imran.f.khan@...cle.com>
Reported-and-tested-by: Xuewen Yan <xuewen.yan@...soc.com>
Fixes: 05c7b7a92cc8 ("cgroup/cpuset: Fix a race between cpuset_attach() and cpu hotplug")
Cc: stable@...r.kernel.org # v5.17+
Signed-off-by: Sasha Levin <sashal@...nel.org>
---
kernel/cgroup/cgroup.c | 78 ++++++++++++++++++++++++++++++++++---------------
kernel/cgroup/cpuset.c | 3 -
2 files changed, 56 insertions(+), 25 deletions(-)
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -30,6 +30,7 @@
#include "cgroup-internal.h"
+#include <linux/cpu.h>
#include <linux/cred.h>
#include <linux/errno.h>
#include <linux/init_task.h>
@@ -2377,6 +2378,47 @@ int task_cgroup_path(struct task_struct
EXPORT_SYMBOL_GPL(task_cgroup_path);
/**
+ * cgroup_attach_lock - Lock for ->attach()
+ * @lock_threadgroup: whether to down_write cgroup_threadgroup_rwsem
+ *
+ * cgroup migration sometimes needs to stabilize threadgroups against forks and
+ * exits by write-locking cgroup_threadgroup_rwsem. However, some ->attach()
+ * implementations (e.g. cpuset), also need to disable CPU hotplug.
+ * Unfortunately, letting ->attach() operations acquire cpus_read_lock() can
+ * lead to deadlocks.
+ *
+ * Bringing up a CPU may involve creating and destroying tasks which requires
+ * read-locking threadgroup_rwsem, so threadgroup_rwsem nests inside
+ * cpus_read_lock(). If we call an ->attach() which acquires the cpus lock while
+ * write-locking threadgroup_rwsem, the locking order is reversed and we end up
+ * waiting for an on-going CPU hotplug operation which in turn is waiting for
+ * the threadgroup_rwsem to be released to create new tasks. For more details:
+ *
+ * http://lkml.kernel.org/r/20220711174629.uehfmqegcwn2lqzu@wubuntu
+ *
+ * Resolve the situation by always acquiring cpus_read_lock() before optionally
+ * write-locking cgroup_threadgroup_rwsem. This allows ->attach() to assume that
+ * CPU hotplug is disabled on entry.
+ */
+static void cgroup_attach_lock(bool lock_threadgroup)
+{
+ cpus_read_lock();
+ if (lock_threadgroup)
+ percpu_down_write(&cgroup_threadgroup_rwsem);
+}
+
+/**
+ * cgroup_attach_unlock - Undo cgroup_attach_lock()
+ * @lock_threadgroup: whether to up_write cgroup_threadgroup_rwsem
+ */
+static void cgroup_attach_unlock(bool lock_threadgroup)
+{
+ if (lock_threadgroup)
+ percpu_up_write(&cgroup_threadgroup_rwsem);
+ cpus_read_unlock();
+}
+
+/**
* cgroup_migrate_add_task - add a migration target task to a migration context
* @task: target task
* @mgctx: target migration context
@@ -2857,8 +2899,7 @@ int cgroup_attach_task(struct cgroup *ds
}
struct task_struct *cgroup_procs_write_start(char *buf, bool threadgroup,
- bool *locked)
- __acquires(&cgroup_threadgroup_rwsem)
+ bool *threadgroup_locked)
{
struct task_struct *tsk;
pid_t pid;
@@ -2875,12 +2916,8 @@ struct task_struct *cgroup_procs_write_s
* Therefore, we can skip the global lock.
*/
lockdep_assert_held(&cgroup_mutex);
- if (pid || threadgroup) {
- percpu_down_write(&cgroup_threadgroup_rwsem);
- *locked = true;
- } else {
- *locked = false;
- }
+ *threadgroup_locked = pid || threadgroup;
+ cgroup_attach_lock(*threadgroup_locked);
rcu_read_lock();
if (pid) {
@@ -2911,17 +2948,14 @@ struct task_struct *cgroup_procs_write_s
goto out_unlock_rcu;
out_unlock_threadgroup:
- if (*locked) {
- percpu_up_write(&cgroup_threadgroup_rwsem);
- *locked = false;
- }
+ cgroup_attach_unlock(*threadgroup_locked);
+ *threadgroup_locked = false;
out_unlock_rcu:
rcu_read_unlock();
return tsk;
}
-void cgroup_procs_write_finish(struct task_struct *task, bool locked)
- __releases(&cgroup_threadgroup_rwsem)
+void cgroup_procs_write_finish(struct task_struct *task, bool threadgroup_locked)
{
struct cgroup_subsys *ss;
int ssid;
@@ -2929,8 +2963,8 @@ void cgroup_procs_write_finish(struct ta
/* release reference from cgroup_procs_write_start() */
put_task_struct(task);
- if (locked)
- percpu_up_write(&cgroup_threadgroup_rwsem);
+ cgroup_attach_unlock(threadgroup_locked);
+
for_each_subsys(ss, ssid)
if (ss->post_attach)
ss->post_attach();
@@ -3007,8 +3041,7 @@ static int cgroup_update_dfl_csses(struc
* write-locking can be skipped safely.
*/
has_tasks = !list_empty(&mgctx.preloaded_src_csets);
- if (has_tasks)
- percpu_down_write(&cgroup_threadgroup_rwsem);
+ cgroup_attach_lock(has_tasks);
/* NULL dst indicates self on default hierarchy */
ret = cgroup_migrate_prepare_dst(&mgctx);
@@ -3029,8 +3062,7 @@ static int cgroup_update_dfl_csses(struc
ret = cgroup_migrate_execute(&mgctx);
out_finish:
cgroup_migrate_finish(&mgctx);
- if (has_tasks)
- percpu_up_write(&cgroup_threadgroup_rwsem);
+ cgroup_attach_unlock(has_tasks);
return ret;
}
@@ -4859,13 +4891,13 @@ static ssize_t cgroup_procs_write(struct
struct task_struct *task;
const struct cred *saved_cred;
ssize_t ret;
- bool locked;
+ bool threadgroup_locked;
dst_cgrp = cgroup_kn_lock_live(of->kn, false);
if (!dst_cgrp)
return -ENODEV;
- task = cgroup_procs_write_start(buf, true, &locked);
+ task = cgroup_procs_write_start(buf, true, &threadgroup_locked);
ret = PTR_ERR_OR_ZERO(task);
if (ret)
goto out_unlock;
@@ -4891,7 +4923,7 @@ static ssize_t cgroup_procs_write(struct
ret = cgroup_attach_task(dst_cgrp, task, true);
out_finish:
- cgroup_procs_write_finish(task, locked);
+ cgroup_procs_write_finish(task, threadgroup_locked);
out_unlock:
cgroup_kn_unlock(of->kn);
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -2204,7 +2204,7 @@ static void cpuset_attach(struct cgroup_
cgroup_taskset_first(tset, &css);
cs = css_cs(css);
- cpus_read_lock();
+ lockdep_assert_cpus_held(); /* see cgroup_attach_lock() */
percpu_down_write(&cpuset_rwsem);
/* prepare for attach */
@@ -2260,7 +2260,6 @@ static void cpuset_attach(struct cgroup_
wake_up(&cpuset_attach_wq);
percpu_up_write(&cpuset_rwsem);
- cpus_read_unlock();
}
/* The various types of files and directories in a cpuset file system */
Powered by blists - more mailing lists