From: Prateek Sood <prsood@codeaurora.org>
To: Peter Zijlstra <peterz@infradead.org>
Cc: tj@kernel.org, lizefan@huawei.com, cgroups@vger.kernel.org,
mingo@kernel.org, longman@redhat.com, boqun.feng@gmail.com,
tglx@linutronix.de, linux-kernel@vger.kernel.org,
sramana@codeaurora.org
Subject: Re: [PATCH] cgroup/cpuset: remove circular dependency deadlock
Date: Wed, 25 Oct 2017 14:09:54 +0530 [thread overview]
Message-ID: <00043e79-3647-3979-07bb-27039921efb4@codeaurora.org> (raw)
In-Reply-To: <20171011094833.pdp4torvotvjdmkt@hirez.programming.kicks-ass.net>
On 10/11/2017 03:18 PM, Peter Zijlstra wrote:
> On Mon, Oct 09, 2017 at 06:57:46PM +0530, Prateek Sood wrote:
>> On 09/07/2017 11:21 PM, Peter Zijlstra wrote:
>
>>> But if you invert these locks, the need for cpuset_hotplug_workfn() goes
>>> away, at least for the CPU part, and we can make in synchronous again.
>>> Yay!!
>
>> The callback making a call to cpuset_hotplug_workfn()in hotplug path are
>> [CPUHP_AP_ACTIVE] = {
>> .name = "sched:active",
>> .startup.single = sched_cpu_activate,
>> .teardown.single = sched_cpu_deactivate,
>> },
>>
>> if we make cpuset_hotplug_workfn() synchronous, deadlock might happen:
>> _cpu_down()
>> cpus_write_lock() //held
>> cpuhp_kick_ap_work()
>> cpuhp_kick_ap()
>> __cpuhp_kick_ap()
>> wake_up_process() //cpuhp_thread_fun
>> wait_for_ap_thread() //wait for complete from cpuhp_thread_fun()
>>
>> cpuhp_thread_fun()
>> cpuhp_invoke_callback()
>> sched_cpu_deactivate()
>> cpuset_cpu_inactive()
>> cpuset_update_active_cpus()
>> cpuset_hotplug_work()
>> rebuild_sched_domains()
>> cpus_read_lock() //waiting as acquired in _cpu_down()
>
> Well, duh, don't use rebuild_sched_domains() 'obviously' :-) use
> rebuild_sched_domains_cpuslocked() instead and it works just fine.
>
> After applying your patch, the below boots and survives a hotplug.
>
> ---
> include/linux/cpuset.h | 6 ------
> kernel/cgroup/cpuset.c | 30 +++++++++---------------------
> kernel/power/process.c | 2 --
> kernel/sched/core.c | 1 -
> 4 files changed, 9 insertions(+), 30 deletions(-)
>
> --- 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 bool cpusets_enabled(void)
> 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)
> {
> --- a/kernel/cgroup/cpuset.c
> +++ b/kernel/cgroup/cpuset.c
> @@ -833,7 +833,12 @@ static void rebuild_sched_domains_cpuslo
> cpumask_var_t *doms;
> int ndoms;
>
> + /*
> + * When called during hotplug, this lock is held by the calling
> + * thread, not cpuhp_thread_fun :/
> + *
> lockdep_assert_cpus_held();
> + */
> lockdep_assert_held(&cpuset_mutex);
>
> /*
> @@ -2281,13 +2286,6 @@ static void cpuset_hotplug_update_tasks(
> 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
> *
> @@ -2362,25 +2360,15 @@ static void cpuset_hotplug_workfn(struct
> }
>
> /* rebuild sched domains if cpus_allowed has changed */
> - if (cpus_updated || force_rebuild) {
> - force_rebuild = false;
> + if (cpus_updated)
> rebuild_sched_domains();
> - }
> }
>
> void cpuset_update_active_cpus(void)
> {
> - /*
> - * 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);
> -}
> -
> -void cpuset_wait_for_hotplug(void)
> -{
> - flush_work(&cpuset_hotplug_work);
> + mutex_lock(&cpuset_mutex);
> + rebuild_sched_domains_cpuslocked();
> + mutex_unlock(&cpuset_mutex);
> }
>
> /*
> --- 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 */
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -5598,7 +5598,6 @@ static void cpuset_cpu_active(void)
> * restore the original sched domains by considering the
> * cpuset configurations.
> */
> - cpuset_force_rebuild();
> }
> cpuset_update_active_cpus();
> }
>
Thanks Peter for sharing the patch and test results.
> void cpuset_update_active_cpus(void)
> {
> - /*
> - * 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);
> -}
> -
> -void cpuset_wait_for_hotplug(void)
> -{
> - flush_work(&cpuset_hotplug_work);
> + mutex_lock(&cpuset_mutex);
> + rebuild_sched_domains_cpuslocked();
> + mutex_unlock(&cpuset_mutex);
> }
In the above patch rebuild_sched_domains_cpuslocked() has been
used directly. Earlier cpuset_hotplug_update_tasks() was also
called from cpuset_hotplug_workfn(). So migration of tasks
related to cgroup which has empty cpuset would not happen
during cpu hotplug.
Could you please help in understanding more on this.
--
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:[~2017-10-25 8:40 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 [this message]
2017-10-25 9:30 ` Peter Zijlstra
2017-10-26 11:52 ` Prateek Sood
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=00043e79-3647-3979-07bb-27039921efb4@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).