From: Prateek Sood <prsood@codeaurora.org>
To: peterz@infradead.org, tj@kernel.org, lizefan@huawei.com,
mingo@kernel.org, longman@redhat.com, boqun.feng@gmail.com,
tglx@linutronix.de
Cc: Prateek Sood <prsood@codeaurora.org>,
cgroups@vger.kernel.org, sramana@codeaurora.org,
linux-kernel@vger.kernel.org
Subject: [PATCH] cgroup/cpuset: remove circular dependency deadlock
Date: Thu, 26 Oct 2017 17:22:02 +0530 [thread overview]
Message-ID: <1509018722-30359-1-git-send-email-prsood@codeaurora.org> (raw)
In-Reply-To: <20171025093041.GO3165@worktop.lehotels.local>
Remove circular dependency deadlock in a scenario where hotplug of CPU is
being done while there is updation in cgroup and cpuset triggered from
userspace.
Process A => kthreadd => Process B => Process C => Process A
Process A
cpu_subsys_offline();
cpu_down();
_cpu_down();
percpu_down_write(&cpu_hotplug_lock); //held
cpuhp_invoke_callback();
workqueue_offline_cpu();
wq_update_unbound_numa();
kthread_create_on_node();
wake_up_process(); //wakeup kthreadd
flush_work();
wait_for_completion();
kthreadd
kthreadd();
kernel_thread();
do_fork();
copy_process();
percpu_down_read(&cgroup_threadgroup_rwsem);
__rwsem_down_read_failed_common(); //waiting
Process B
kernfs_fop_write();
cgroup_file_write();
cgroup_procs_write();
percpu_down_write(&cgroup_threadgroup_rwsem); //held
cgroup_attach_task();
cgroup_migrate();
cgroup_migrate_execute();
cpuset_can_attach();
mutex_lock(&cpuset_mutex); //waiting
Process C
kernfs_fop_write();
cgroup_file_write();
cpuset_write_resmask();
mutex_lock(&cpuset_mutex); //held
update_cpumask();
update_cpumasks_hier();
rebuild_sched_domains_locked();
get_online_cpus();
percpu_down_read(&cpu_hotplug_lock); //waiting
Eliminating deadlock by reversing the locking order for cpuset_mutex and
cpu_hotplug_lock.
Signed-off-by: Prateek Sood <prsood@codeaurora.org>
---
include/linux/cpuset.h | 6 -----
kernel/cgroup/cpuset.c | 70 ++++++++++++++++++++++++++------------------------
kernel/power/process.c | 2 --
kernel/sched/core.c | 1 -
4 files changed, 36 insertions(+), 43 deletions(-)
diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index a1e6a33..e74655d 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -51,9 +51,7 @@ static inline void cpuset_dec(void)
extern int cpuset_init(void);
extern void cpuset_init_smp(void);
-extern void cpuset_force_rebuild(void);
extern void cpuset_update_active_cpus(void);
-extern void cpuset_wait_for_hotplug(void);
extern void cpuset_cpus_allowed(struct task_struct *p, struct cpumask *mask);
extern void cpuset_cpus_allowed_fallback(struct task_struct *p);
extern nodemask_t cpuset_mems_allowed(struct task_struct *p);
@@ -166,15 +164,11 @@ static inline void set_mems_allowed(nodemask_t nodemask)
static inline int cpuset_init(void) { return 0; }
static inline void cpuset_init_smp(void) {}
-static inline void cpuset_force_rebuild(void) { }
-
static inline void cpuset_update_active_cpus(void)
{
partition_sched_domains(1, NULL, NULL);
}
-static inline void cpuset_wait_for_hotplug(void) { }
-
static inline void cpuset_cpus_allowed(struct task_struct *p,
struct cpumask *mask)
{
diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
index 4657e29..a8213c2 100644
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -826,16 +826,14 @@ static int generate_sched_domains(cpumask_var_t **domains,
* 'cpus' is removed, then call this routine to rebuild the
* scheduler's dynamic sched domains.
*
- * Call with cpuset_mutex held. Takes get_online_cpus().
*/
-static void rebuild_sched_domains_locked(void)
+static void rebuild_sched_domains_cpuslocked(void)
{
struct sched_domain_attr *attr;
cpumask_var_t *doms;
int ndoms;
lockdep_assert_held(&cpuset_mutex);
- get_online_cpus();
/*
* We have raced with CPU hotplug. Don't do anything to avoid
@@ -843,27 +841,27 @@ static void rebuild_sched_domains_locked(void)
* Anyways, hotplug work item will rebuild sched domains.
*/
if (!cpumask_equal(top_cpuset.effective_cpus, cpu_active_mask))
- goto out;
+ return;
/* Generate domain masks and attrs */
ndoms = generate_sched_domains(&doms, &attr);
/* Have scheduler rebuild the domains */
partition_sched_domains(ndoms, doms, attr);
-out:
- put_online_cpus();
}
#else /* !CONFIG_SMP */
-static void rebuild_sched_domains_locked(void)
+static void rebuild_sched_domains_cpuslocked(void)
{
}
#endif /* CONFIG_SMP */
void rebuild_sched_domains(void)
{
+ cpus_read_lock();
mutex_lock(&cpuset_mutex);
- rebuild_sched_domains_locked();
+ rebuild_sched_domains_cpuslocked();
mutex_unlock(&cpuset_mutex);
+ cpus_read_unlock();
}
/**
@@ -949,7 +947,7 @@ static void update_cpumasks_hier(struct cpuset *cs, struct cpumask *new_cpus)
rcu_read_unlock();
if (need_rebuild_sched_domains)
- rebuild_sched_domains_locked();
+ rebuild_sched_domains_cpuslocked();
}
/**
@@ -1281,7 +1279,7 @@ static int update_relax_domain_level(struct cpuset *cs, s64 val)
cs->relax_domain_level = val;
if (!cpumask_empty(cs->cpus_allowed) &&
is_sched_load_balance(cs))
- rebuild_sched_domains_locked();
+ rebuild_sched_domains_cpuslocked();
}
return 0;
@@ -1314,7 +1312,6 @@ static void update_tasks_flags(struct cpuset *cs)
*
* Call with cpuset_mutex held.
*/
-
static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs,
int turning_on)
{
@@ -1347,7 +1344,7 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs,
spin_unlock_irq(&callback_lock);
if (!cpumask_empty(trialcs->cpus_allowed) && balance_flag_changed)
- rebuild_sched_domains_locked();
+ rebuild_sched_domains_cpuslocked();
if (spread_flag_changed)
update_tasks_flags(cs);
@@ -1615,6 +1612,7 @@ static int cpuset_write_u64(struct cgroup_subsys_state *css, struct cftype *cft,
cpuset_filetype_t type = cft->private;
int retval = 0;
+ cpus_read_lock();
mutex_lock(&cpuset_mutex);
if (!is_cpuset_online(cs)) {
retval = -ENODEV;
@@ -1652,6 +1650,7 @@ static int cpuset_write_u64(struct cgroup_subsys_state *css, struct cftype *cft,
}
out_unlock:
mutex_unlock(&cpuset_mutex);
+ cpus_read_unlock();
return retval;
}
@@ -1662,6 +1661,7 @@ static int cpuset_write_s64(struct cgroup_subsys_state *css, struct cftype *cft,
cpuset_filetype_t type = cft->private;
int retval = -ENODEV;
+ cpus_read_lock();
mutex_lock(&cpuset_mutex);
if (!is_cpuset_online(cs))
goto out_unlock;
@@ -1676,6 +1676,7 @@ static int cpuset_write_s64(struct cgroup_subsys_state *css, struct cftype *cft,
}
out_unlock:
mutex_unlock(&cpuset_mutex);
+ cpus_read_unlock();
return retval;
}
@@ -1714,6 +1715,7 @@ static ssize_t cpuset_write_resmask(struct kernfs_open_file *of,
kernfs_break_active_protection(of->kn);
flush_work(&cpuset_hotplug_work);
+ cpus_read_lock();
mutex_lock(&cpuset_mutex);
if (!is_cpuset_online(cs))
goto out_unlock;
@@ -1739,6 +1741,7 @@ static ssize_t cpuset_write_resmask(struct kernfs_open_file *of,
free_trial_cpuset(trialcs);
out_unlock:
mutex_unlock(&cpuset_mutex);
+ cpus_read_unlock();
kernfs_unbreak_active_protection(of->kn);
css_put(&cs->css);
flush_workqueue(cpuset_migrate_mm_wq);
@@ -2039,13 +2042,14 @@ static int cpuset_css_online(struct cgroup_subsys_state *css)
/*
* If the cpuset being removed has its flag 'sched_load_balance'
* enabled, then simulate turning sched_load_balance off, which
- * will call rebuild_sched_domains_locked().
+ * will call rebuild_sched_domains_cpuslocked().
*/
static void cpuset_css_offline(struct cgroup_subsys_state *css)
{
struct cpuset *cs = css_cs(css);
+ cpus_read_lock();
mutex_lock(&cpuset_mutex);
if (is_sched_load_balance(cs))
@@ -2055,6 +2059,7 @@ static void cpuset_css_offline(struct cgroup_subsys_state *css)
clear_bit(CS_ONLINE, &cs->flags);
mutex_unlock(&cpuset_mutex);
+ cpus_read_unlock();
}
static void cpuset_css_free(struct cgroup_subsys_state *css)
@@ -2275,15 +2280,8 @@ static void cpuset_hotplug_update_tasks(struct cpuset *cs)
mutex_unlock(&cpuset_mutex);
}
-static bool force_rebuild;
-
-void cpuset_force_rebuild(void)
-{
- force_rebuild = true;
-}
-
/**
- * cpuset_hotplug_workfn - handle CPU/memory hotunplug for a cpuset
+ * cpuset_hotplug - handle CPU/memory hotunplug for a cpuset
*
* This function is called after either CPU or memory configuration has
* changed and updates cpuset accordingly. The top_cpuset is always
@@ -2298,7 +2296,7 @@ void cpuset_force_rebuild(void)
* Note that CPU offlining during suspend is ignored. We don't modify
* cpusets across suspend/resume cycles at all.
*/
-static void cpuset_hotplug_workfn(struct work_struct *work)
+static void cpuset_hotplug(bool use_cpu_hp_lock)
{
static cpumask_t new_cpus;
static nodemask_t new_mems;
@@ -2356,25 +2354,29 @@ static void cpuset_hotplug_workfn(struct work_struct *work)
}
/* rebuild sched domains if cpus_allowed has changed */
- if (cpus_updated || force_rebuild) {
- force_rebuild = false;
- rebuild_sched_domains();
+ if (cpus_updated) {
+ if (use_cpu_hp_lock)
+ rebuild_sched_domains();
+ else {
+ /* When called during cpu hotplug cpu_hotplug_lock
+ * is held by the calling thread, not
+ * not cpuhp_thread_fun
+ */
+ mutex_lock(&cpuset_mutex);
+ rebuild_sched_domains_cpuslocked();
+ mutex_unlock(&cpuset_mutex);
+ }
}
}
-void cpuset_update_active_cpus(void)
+static void cpuset_hotplug_workfn(struct work_struct *work)
{
- /*
- * We're inside cpu hotplug critical region which usually nests
- * inside cgroup synchronization. Bounce actual hotplug processing
- * to a work item to avoid reverse locking order.
- */
- schedule_work(&cpuset_hotplug_work);
+ cpuset_hotplug(true);
}
-void cpuset_wait_for_hotplug(void)
+void cpuset_update_active_cpus(void)
{
- flush_work(&cpuset_hotplug_work);
+ cpuset_hotplug(false);
}
/*
diff --git a/kernel/power/process.c b/kernel/power/process.c
index 50f25cb..28772b405 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -203,8 +203,6 @@ void thaw_processes(void)
__usermodehelper_set_disable_depth(UMH_FREEZING);
thaw_workqueues();
- cpuset_wait_for_hotplug();
-
read_lock(&tasklist_lock);
for_each_process_thread(g, p) {
/* No other threads should have PF_SUSPEND_TASK set */
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index d17c5da..25b8717 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -5590,7 +5590,6 @@ static void cpuset_cpu_active(void)
* restore the original sched domains by considering the
* cpuset configurations.
*/
- cpuset_force_rebuild();
}
cpuset_update_active_cpus();
}
--
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc.,
is a member of Code Aurora Forum, a Linux Foundation Collaborative Project.
next prev parent reply other threads:[~2017-10-26 11:52 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-09-07 13:56 [PATCH] cgroup/cpuset: remove circular dependency deadlock Prateek Sood
2017-09-07 17:45 ` Peter Zijlstra
2017-09-08 2:13 ` Prateek Sood
2017-09-07 17:51 ` Peter Zijlstra
2017-10-09 13:27 ` Prateek Sood
2017-10-11 9:48 ` Peter Zijlstra
2017-10-25 8:39 ` Prateek Sood
2017-10-25 9:30 ` Peter Zijlstra
2017-10-26 11:52 ` Prateek Sood [this message]
2017-10-26 14:05 ` Waiman Long
2017-10-27 8:03 ` Prateek Sood
2017-10-30 7:16 ` [PATCH v2] " Prateek Sood
2017-11-06 4:01 ` Prateek Sood
2017-11-15 10:26 ` Prateek Sood
2017-11-15 10:37 ` Peter Zijlstra
2017-11-15 14:20 ` [PATCH v3 0/2] Invert cpu_hotplug_lock and cpuset_mutex locking order Prateek Sood
2017-11-15 14:20 ` [PATCH v3 1/2] cgroup/cpuset: remove circular dependency deadlock Prateek Sood
2017-11-15 14:20 ` [PATCH v3 2/2] cpuset: Make cpuset hotplug synchronous Prateek Sood
2017-11-27 16:48 ` [PATCH v3 0/2] Invert cpu_hotplug_lock and cpuset_mutex locking order Tejun Heo
2017-11-15 17:05 ` [PATCH v2] cgroup/cpuset: remove circular dependency deadlock Tejun Heo
2017-11-15 17:18 ` Prateek Sood
-- strict thread matches above, loose matches on Subject: below --
2017-09-07 6:04 [PATCH] " Prateek Sood
2017-09-07 7:28 ` Peter Zijlstra
2017-09-07 8:56 ` Boqun Feng
2017-09-07 9:07 ` Prateek Sood
2017-09-07 9:05 ` Prateek Sood
2017-09-06 11:48 Prateek Sood
2017-09-06 12:56 ` Waiman Long
2017-09-06 14:23 ` Prateek Sood
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1509018722-30359-1-git-send-email-prsood@codeaurora.org \
--to=prsood@codeaurora.org \
--cc=boqun.feng@gmail.com \
--cc=cgroups@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lizefan@huawei.com \
--cc=longman@redhat.com \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=sramana@codeaurora.org \
--cc=tglx@linutronix.de \
--cc=tj@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).