Skip to contentSkip to conversation

sched: Core Scheduling

PATCH Applied Peter Zijlstra 13 participants 103 messages Apr 22, 2021 – Aug 23, 2021 Page 1 of 2

Subsystems (5)
lore
Share & tools
lore

Peter Zijlstra

author maintainer
#0 lore raw

cover sched: Core Scheduling

Hai,

This is an agressive fold of all the core-scheduling work so far. I've stripped a whole bunch of tags along the way (hopefully not too many, please yell if you feel I made a mistake), including tested-by. Please retest.

Changes since the last partial post is dropping all the cgroup stuff and PR_SCHED_CORE_CLEAR as well as that exec() behaviour in order to later resolve the cgroup issue.

Since we're really rather late for the coming merge window, my plan was to merge the lot right after the merge window.

Again, please test.

These patches should shortly be available in my queue.git.

---
 b/kernel/sched/core_sched.c                     |  229 ++++++
 b/tools/testing/selftests/sched/.gitignore      |    1 
 b/tools/testing/selftests/sched/Makefile        |   14 
 b/tools/testing/selftests/sched/config          |    1 
 b/tools/testing/selftests/sched/cs_prctl_test.c |  338 +++++++++
 include/linux/sched.h                           |   19 
 include/uapi/linux/prctl.h                      |    8 
 kernel/Kconfig.preempt                          |    6 
 kernel/fork.c                                   |    4 
 kernel/sched/Makefile                           |    1 
 kernel/sched/core.c                             |  858 ++++++++++++++++++++++--
 kernel/sched/cpuacct.c                          |   12 
 kernel/sched/deadline.c                         |   38 -
 kernel/sched/debug.c                            |    4 
 kernel/sched/fair.c                             |  276 +++++--
 kernel/sched/idle.c                             |   13 
 kernel/sched/pelt.h                             |    2 
 kernel/sched/rt.c                               |   31 
 kernel/sched/sched.h                            |  393 ++++++++--
 kernel/sched/stop_task.c                        |   14 
 kernel/sched/topology.c                         |    4 
 kernel/sys.c                                    |    5 
 tools/include/uapi/linux/prctl.h                |    8 
 23 files changed, 2057 insertions(+), 222 deletions(-)
author maintainer
#1 lore raw

PATCH 1/19 sched/fair: Add a few assertions

---
 kernel/sched/fair.c |   11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)
kernel/sched/fair.c +9 -2
@@ -6246,6 +6246,11 @@ static int select_idle_sibling(struct ta
 		task_util = uclamp_task_util(p);
 	}
 
+	/*
+	 * per-cpu select_idle_mask usage
+	 */
+	lockdep_assert_irqs_disabled();
+
 	if ((available_idle_cpu(target) || sched_idle_cpu(target)) &&
 	    asym_fits_capacity(task_util, target))
 		return target;
@@ -6711,8 +6716,6 @@ static int find_energy_efficient_cpu(str
  * certain conditions an idle sibling CPU if the domain has SD_WAKE_AFFINE set.
  *
  * Returns the target CPU number.
- *
- * preempt must be disabled.
  */
 static int
 select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
@@ -6725,6 +6728,10 @@ select_task_rq_fair(struct task_struct *
 	/* SD_flags and WF_flags share the first nibble */
 	int sd_flag = wake_flags & 0xF;
 
+	/*
+	 * required for stable ->cpus_allowed
+	 */
+	lockdep_assert_held(&p->pi_lock);
 	if (wake_flags & WF_TTWU) {
 		record_wakee(p);
 
merge bot Applied
#2 lore raw

[tip: sched/core] sched/fair: Add a few assertions

The following commit has been merged into the sched/core branch of tip:

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     9099a14708ce1dfecb6002605594a0daa319b555
Gitweb:        https://git.kernel.org/tip/9099a14708ce1dfecb6002605594a0daa319b555
Author:        Peter Zijlstra <[email protected]>
AuthorDate:    Tue, 17 Nov 2020 18:19:35 -05:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 12 May 2021 11:43:26 +02:00

sched/fair: Add a few assertions

---
 kernel/sched/fair.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)
kernel/sched/fair.c +9 -2
index c209f68..6bdbb7b 100644
@@ -6288,6 +6288,11 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target)
 		task_util = uclamp_task_util(p);
 	}
 
+	/*
+	 * per-cpu select_idle_mask usage
+	 */
+	lockdep_assert_irqs_disabled();
+
 	if ((available_idle_cpu(target) || sched_idle_cpu(target)) &&
 	    asym_fits_capacity(task_util, target))
 		return target;
@@ -6781,8 +6786,6 @@ unlock:
  * certain conditions an idle sibling CPU if the domain has SD_WAKE_AFFINE set.
  *
  * Returns the target CPU number.
- *
- * preempt must be disabled.
  */
 static int
 select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
@@ -6795,6 +6798,10 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
 	/* SD_flags and WF_flags share the first nibble */
 	int sd_flag = wake_flags & 0xF;
 
+	/*
+	 * required for stable ->cpus_allowed
+	 */
+	lockdep_assert_held(&p->pi_lock);
 	if (wake_flags & WF_TTWU) {
 		record_wakee(p);
 
#3 lore raw

Re: [tip: sched/core] sched/fair: Add a few assertions

Quoted 11 lines · On 2021/5/12 18:28, tip-bot2 for Peter Zijlstra wrote:

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     9099a14708ce1dfecb6002605594a0daa319b555
Gitweb:        https://git.kernel.org/tip/9099a14708ce1dfecb6002605594a0daa319b555
Author:        Peter Zijlstra <[email protected]>
AuthorDate:    Tue, 17 Nov 2020 18:19:35 -05:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 12 May 2021 11:43:26 +02:00

sched/fair: Add a few assertions

Quoted 9 lines
---
 kernel/sched/fair.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

Add quick test results based on tip tree sched/core merge commit:

====TEST INFO====
- kernel under test:

	-- tip tree sched/core merge: https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?id=60208dac643e24cbc62317de4e486fdcbbf05215
	-- coresched_v10 kernel source: https://github.com/digitalocean/linux-coresched/commits/coresched/v10-v5.10.y

- test machine setup:
CPU(s): 192
On-line CPU(s) list: 0-191
Thread(s) per core: 2
Core(s) per socket: 48
Socket(s): 2
NUMA node(s): 4

- performance test workloads:
-- A. sysbench cpu (192 threads) + sysbench cpu (192 threads) -- B. sysbench cpu (192 threads) + sysbench mysql (192 threads) -- C. uperf netperf.xml (192 threads over TCP or UDP protocol separately) -- D. will-it-scale context_switch via pipe (192 threads)

- negative test:
-- A. continuously toggle coresched (enable/disable) thru prctl on task cookies of PGID, during full loading of uperf workload with coresched on -- B. continuously toggle smt (on/off) via /sys/devices/system/cpu/smt/control, during full loading of uperf workload with coresched on

====TEST RESULTS====
- performance change key info:
--workload B: coresched (cs_on), sysbench mysql performance drop around 20% vs coresched_v10
--workload C, coresched (cs_on), uperf performance increased almost double vs coresched_v10
--workload C, default (cs_off), uperf performance drop over 25% vs coresched_v10, same issue seen on v5.13-rc1 base (w/o coresched patchset) --workload D, coresched (cs_on), wis performance increased almost double vs coresched_v10

- negative test summary:
no platform hang or kernel panic observed for both test

- performance info of workloads, normalized based on coresched_v10 results -- performance workload A:
Note:
* no performance change compared to coresched_v10

+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
|                                       | **   | coresched_tip_merge_base_v5.13-rc1   | coresched_tip_merge_base_v5.13-rc1     | **    | coresched_v10_base_v5.10.11   | coresched_v10_base_v5.10.11     |
+=======================================+======+======================================+========================================+=======+===============================+=================================+
| workload                              | **   | sysbench cpu * 192                   | sysbench cpu * 192                     | **    | sysbench cpu * 192            | sysbench cpu * 192              |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| prctl/cgroup                          | **   | prctl on workload cpu_0              | prctl on workload cpu_1                | **    | cg_sysbench_cpu_0             | cg_sysbench_cpu_1               |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| record_item                           | **   | Tput_avg (events/s)                  | Tput_avg (events/s)                    | **    | Tput_avg (events/s)           | Tput_avg (events/s)             |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| coresched normalized vs coresched_v10 | **   | 0.97                                 | 1.05                                   | **    | 1                             | 1                               |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| default normalized vs coresched_v10   | **   | 1.03                                 | 0.95                                   | **    | 1                             | 1                               |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| smtoff normalized vs coresched_v10    | **   | 0.96                                 | 1.04                                   | **    | 1                             | 1                               |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+

-- performance workload B:
Note:
* coresched (cs_on), sysbench mysql performance drop around 20% vs coresched_v10

+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
|                                       | **   | coresched_tip_merge_base_v5.13-rc1   | coresched_tip_merge_base_v5.13-rc1     | **    | coresched_v10_base_v5.10.11   | coresched_v10_base_v5.10.11     |
+=======================================+======+======================================+========================================+=======+===============================+=================================+
| workload                              | **   | sysbench cpu * 192                   | sysbench mysql * 192                   | **    | sysbench cpu * 192            | sysbench mysql * 192            |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| prctl/cgroup                          | **   | prctl on workload cpu_0              | prctl on workload mysql_0              | **    | cg_sysbench_cpu_0             | cg_sysbench_mysql_0             |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| record_item                           | **   | Tput_avg (events/s)                  | Tput_avg (events/s)                    | **    | Tput_avg (events/s)           | Tput_avg (events/s)             |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| coresched normalized vs coresched_v10 | **   | 1.02                                 | 0.81                                   | **    | 1                             | 1                               |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| default normalized vs coresched_v10   | **   | 1.01                                 | 0.94                                   | **    | 1                             | 1                               |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| smtoff normalized vs coresched_v10    | **   | 0.93                                 | 1.18                                   | **    | 1                             | 1                               |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+

-- performance workload C:
Note:
* coresched (cs_on), uperf performance increased almost double vs coresched_v10
* default (cs_off), uperf performance drop over 25% vs coresched_v10, same issue seen on v5.13-rc1 base (w/o coresched patchset)

+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
|                                       | **   | coresched_tip_merge_base_v5.13-rc1   | coresched_tip_merge_base_v5.13-rc1     | **    | coresched_v10_base_v5.10.11   | coresched_v10_base_v5.10.11     |
+=======================================+======+======================================+========================================+=======+===============================+=================================+
| workload                              | **   | uperf netperf TCP * 192              | uperf netperf UDP * 192                | **    | uperf netperf TCP * 192       | uperf netperf UDP * 192         |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| prctl/cgroup                          | **   | prctl on workload uperf              | prctl on workload uperf                | **    | cg_uperf                      | cg_uperf                        |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| record_item                           | **   | Tput_avg (Gb/s)                      | Tput_avg (Gb/s)                        | **    | Tput_avg (Gb/s)               | Tput_avg (Gb/s)                 |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| coresched normalized vs coresched_v10 | **   | 1.83                                 | 1.93                                   | **    | 1                             | 1                               |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| default normalized vs coresched_v10   | **   | 0.75                                 | 0.71                                   | **    | 1                             | 1                               |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+
| smtoff normalized vs coresched_v10    | **   | 1                                    | 1.06                                   | **    | 1                             | 1                               |
+---------------------------------------+------+--------------------------------------+----------------------------------------+-------+-------------------------------+---------------------------------+

-- performance workload D:
Note:
* coresched (cs_on), wis performance increased almost double vs coresched_v10
* default (cs_off) and smtoff, wis performance is better vs coresched_v10

+---------------------------------------+------+--------------------------------------+-------+-------------------------------+
|                                       | **   | coresched_tip_merge_base_v5.13-rc1   | **    | coresched_v10_base_v5.10.11   |
+=======================================+======+======================================+=======+===============================+
| workload                              | **   | will-it-scale  * 192                 | **    | will-it-scale  * 192          |
|                                       |      | (pipe based context_switch)          |       | (pipe based context_switch)   |
+---------------------------------------+------+--------------------------------------+-------+-------------------------------+
| prctl/cgroup                          | **   | prctl on workload wis                | **    | cg_wis                        |
+---------------------------------------+------+--------------------------------------+-------+-------------------------------+
| record_item                           | **   | threads_avg                          | **    | threads_avg                   |
+---------------------------------------+------+--------------------------------------+-------+-------------------------------+
| coresched normalized vs coresched_v10 | **   | 2.01                                 | **    | 1.00                          |
+---------------------------------------+------+--------------------------------------+-------+-------------------------------+
| default normalized vs coresched_v10   | **   | 1.13                                 | **    | 1.00                          |
+---------------------------------------+------+--------------------------------------+-------+-------------------------------+
| smtoff normalized vs coresched_v10    | **   | 1.29                                 | **    | 1.00                          |
+---------------------------------------+------+--------------------------------------+-------+-------------------------------+

-- notes on record_item:
* coresched normalized vs coresched_v10: smton, cs enabled, test result normalized by result of coresched_v10 under same config
* default normalized vs coresched_v10: smton, cs disabled, test result normalized by result of coresched_v10 under same config
* smtoff normalized vs coresched_v10: smtoff, test result normalized by result of coresched_v10 under same config

-- Hongyu Ning

author maintainer
#4 lore raw

PATCH 2/19 sched: Provide raw_spin_rq_*lock*() helpers

In reply to Peter Zijlstra

In prepration for playing games with rq->lock, add some rq_lock wrappers.

---
 kernel/sched/core.c  |   15 +++++++++++++++
 kernel/sched/sched.h |   50 ++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 65 insertions(+)
kernel/sched/core.c +15 -0
@@ -184,6 +184,21 @@ int sysctl_sched_rt_runtime = 950000;
  *
  */
 
+void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
+{
+	raw_spin_lock_nested(rq_lockp(rq), subclass);
+}
+
+bool raw_spin_rq_trylock(struct rq *rq)
+{
+	return raw_spin_trylock(rq_lockp(rq));
+}
+
+void raw_spin_rq_unlock(struct rq *rq)
+{
+	raw_spin_unlock(rq_lockp(rq));
+}
+
 /*
  * __task_rq_lock - lock the rq @p resides on.
  */
kernel/sched/sched.h +50 -0

56 lines not on this page · raw message

merge bot Applied
#5 lore raw

[tip: sched/core] sched: Provide raw_spin_rq_*lock*() helpers

The following commit has been merged into the sched/core branch of tip:

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     39d371b7c0c299d489041884d005aacc4bba8c15
Gitweb:        https://git.kernel.org/tip/39d371b7c0c299d489041884d005aacc4bba8c15
Author:        Peter Zijlstra <[email protected]>
AuthorDate:    Tue, 02 Mar 2021 12:13:13 +01:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 12 May 2021 11:43:26 +02:00

sched: Provide raw_spin_rq_*lock*() helpers

In prepration for playing games with rq->lock, add some rq_lock wrappers.

---
 kernel/sched/core.c  | 15 +++++++++++++-
 kernel/sched/sched.h | 50 +++++++++++++++++++++++++++++++++++++++++++-
 2 files changed, 65 insertions(+)
kernel/sched/core.c +15 -0
index 660120d..5568018 100644
@@ -184,6 +184,21 @@ int sysctl_sched_rt_runtime = 950000;
  *
  */
 
+void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
+{
+	raw_spin_lock_nested(rq_lockp(rq), subclass);
+}
+
+bool raw_spin_rq_trylock(struct rq *rq)
+{
+	return raw_spin_trylock(rq_lockp(rq));
+}
+
+void raw_spin_rq_unlock(struct rq *rq)
+{
+	raw_spin_unlock(rq_lockp(rq));
+}
+
 /*
  * __task_rq_lock - lock the rq @p resides on.
  */
kernel/sched/sched.h +50 -0

56 lines not on this page · raw message

author maintainer
#6 lore raw

PATCH 3/19 sched: Wrap rq::lock access

In reply to Peter Zijlstra

In preparation of playing games with rq->lock, abstract the thing using an accessor.

---
 kernel/sched/core.c     |   70 ++++++++++++++++----------------
 kernel/sched/cpuacct.c  |   12 ++---
 kernel/sched/deadline.c |   22 +++++-----
 kernel/sched/debug.c    |    4 -
 kernel/sched/fair.c     |   35 +++++++---------
 kernel/sched/idle.c     |    4 -
 kernel/sched/pelt.h     |    2 
 kernel/sched/rt.c       |   16 +++----
 kernel/sched/sched.h    |  103 +++++++++++++++++++++++-------------------------
 kernel/sched/topology.c |    4 -
 10 files changed, 135 insertions(+), 137 deletions(-)
kernel/sched/core.c +35 -35
@@ -211,12 +211,12 @@ struct rq *__task_rq_lock(struct task_st
 
 	for (;;) {
 		rq = task_rq(p);
-		raw_spin_lock(&rq->lock);
+		raw_spin_rq_lock(rq);
 		if (likely(rq == task_rq(p) && !task_on_rq_migrating(p))) {
 			rq_pin_lock(rq, rf);
 			return rq;
 		}
-		raw_spin_unlock(&rq->lock);
+		raw_spin_rq_unlock(rq);
 
 		while (unlikely(task_on_rq_migrating(p)))
 			cpu_relax();
@@ -235,7 +235,7 @@ struct rq *task_rq_lock(struct task_stru
 	for (;;) {
 		raw_spin_lock_irqsave(&p->pi_lock, rf->flags);
 		rq = task_rq(p);
-		raw_spin_lock(&rq->lock);
+		raw_spin_rq_lock(rq);
 		/*
 		 *	move_queued_task()		task_rq_lock()
 		 *
@@ -257,7 +257,7 @@ struct rq *task_rq_lock(struct task_stru
 			rq_pin_lock(rq, rf);
 			return rq;
 		}
-		raw_spin_unlock(&rq->lock);
+		raw_spin_rq_unlock(rq);
 		raw_spin_unlock_irqrestore(&p->pi_lock, rf->flags);
 
 		while (unlikely(task_on_rq_migrating(p)))
@@ -327,7 +327,7 @@ void update_rq_clock(struct rq *rq)
 {
 	s64 delta;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	if (rq->clock_update_flags & RQCF_ACT_SKIP)
 		return;
@@ -626,7 +626,7 @@ void resched_curr(struct rq *rq)
 	struct task_struct *curr = rq->curr;
 	int cpu;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	if (test_tsk_need_resched(curr))
 		return;
@@ -650,10 +650,10 @@ void resched_cpu(int cpu)
 	struct rq *rq = cpu_rq(cpu);
 	unsigned long flags;
 
-	raw_spin_lock_irqsave(&rq->lock, flags);
+	raw_spin_rq_lock_irqsave(rq, flags);
 	if (cpu_online(cpu) || cpu == smp_processor_id())
 		resched_curr(rq);
-	raw_spin_unlock_irqrestore(&rq->lock, flags);
+	raw_spin_rq_unlock_irqrestore(rq, flags);
 }
 
 #ifdef CONFIG_SMP
@@ -1152,7 +1152,7 @@ static inline void uclamp_rq_inc_id(stru
 	struct uclamp_se *uc_se = &p->uclamp[clamp_id];
 	struct uclamp_bucket *bucket;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	/* Update task effective clamp */
 	p->uclamp[clamp_id] = uclamp_eff_get(p, clamp_id);
@@ -1192,7 +1192,7 @@ static inline void uclamp_rq_dec_id(stru
 	unsigned int bkt_clamp;
 	unsigned int rq_clamp;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	/*
 	 * If sched_uclamp_used was enabled after task @p was enqueued,
@@ -1865,7 +1865,7 @@ static inline bool is_cpu_allowed(struct
 static struct rq *move_queued_task(struct rq *rq, struct rq_flags *rf,
 				   struct task_struct *p, int new_cpu)
 {
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	deactivate_task(rq, p, DEQUEUE_NOCLOCK);
 	set_task_cpu(p, new_cpu);
@@ -2039,7 +2039,7 @@ int push_cpu_stop(void *arg)
 	struct task_struct *p = arg;
 
 	raw_spin_lock_irq(&p->pi_lock);
-	raw_spin_lock(&rq->lock);
+	raw_spin_rq_lock(rq);
 
 	if (task_rq(p) != rq)
 		goto out_unlock;
@@ -2069,7 +2069,7 @@ int push_cpu_stop(void *arg)
 
 out_unlock:
 	rq->push_busy = false;
-	raw_spin_unlock(&rq->lock);
+	raw_spin_rq_unlock(rq);
 	raw_spin_unlock_irq(&p->pi_lock);
 
 	put_task_struct(p);
@@ -2122,7 +2122,7 @@ __do_set_cpus_allowed(struct task_struct
 		 * Because __kthread_bind() calls this on blocked tasks without
 		 * holding rq->lock.
 		 */
-		lockdep_assert_held(&rq->lock);
+		lockdep_assert_rq_held(rq);
 		dequeue_task(rq, p, DEQUEUE_SAVE | DEQUEUE_NOCLOCK);
 	}
 	if (running)
@@ -2463,7 +2463,7 @@ void set_task_cpu(struct task_struct *p,
 	 * task_rq_lock().
 	 */
 	WARN_ON_ONCE(debug_locks && !(lockdep_is_held(&p->pi_lock) ||
-				      lockdep_is_held(&task_rq(p)->lock)));
+				      lockdep_is_held(rq_lockp(task_rq(p)))));
 #endif
 	/*
 	 * Clearly, migrating tasks to offline CPUs is a fairly daft thing.
@@ -3005,7 +3005,7 @@ ttwu_do_activate(struct rq *rq, struct t
 {
 	int en_flags = ENQUEUE_WAKEUP | ENQUEUE_NOCLOCK;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	if (p->sched_contributes_to_load)
 		rq->nr_uninterruptible--;
@@ -4016,7 +4016,7 @@ static void do_balance_callbacks(struct
 	void (*func)(struct rq *rq);
 	struct callback_head *next;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	while (head) {
 		func = (void (*)(struct rq *))head->func;
@@ -4039,7 +4039,7 @@ static inline struct callback_head *spli
 {
 	struct callback_head *head = rq->balance_callback;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 	if (head)
 		rq->balance_callback = NULL;
 
@@ -4056,9 +4056,9 @@ static inline void balance_callbacks(str
 	unsigned long flags;
 
 	if (unlikely(head)) {
-		raw_spin_lock_irqsave(&rq->lock, flags);
+		raw_spin_rq_lock_irqsave(rq, flags);
 		do_balance_callbacks(rq, head);
-		raw_spin_unlock_irqrestore(&rq->lock, flags);
+		raw_spin_rq_unlock_irqrestore(rq, flags);
 	}
 }
 
@@ -4089,10 +4089,10 @@ prepare_lock_switch(struct rq *rq, struc
 	 * do an early lockdep release here:
 	 */
 	rq_unpin_lock(rq, rf);
-	spin_release(&rq->lock.dep_map, _THIS_IP_);
+	spin_release(&rq_lockp(rq)->dep_map, _THIS_IP_);
 #ifdef CONFIG_DEBUG_SPINLOCK
 	/* this is a valid case when another task releases the spinlock */
-	rq->lock.owner = next;
+	rq_lockp(rq)->owner = next;
 #endif
 }
 
@@ -4103,9 +4103,9 @@ static inline void finish_lock_switch(st
 	 * fix up the runqueue lock - which gets 'carried over' from
 	 * prev into current:
 	 */
-	spin_acquire(&rq->lock.dep_map, 0, 0, _THIS_IP_);
+	spin_acquire(&rq_lockp(rq)->dep_map, 0, 0, _THIS_IP_);
 	__balance_callbacks(rq);
-	raw_spin_unlock_irq(&rq->lock);
+	raw_spin_rq_unlock_irq(rq);
 }
 
 /*
@@ -5165,7 +5165,7 @@ static void __sched notrace __schedule(b
 
 		rq_unpin_lock(rq, &rf);
 		__balance_callbacks(rq);
-		raw_spin_unlock_irq(&rq->lock);
+		raw_spin_rq_unlock_irq(rq);
 	}
 }
 
@@ -5707,7 +5707,7 @@ void rt_mutex_setprio(struct task_struct
 
 	rq_unpin_lock(rq, &rf);
 	__balance_callbacks(rq);
-	raw_spin_unlock(&rq->lock);
+	raw_spin_rq_unlock(rq);
 
 	preempt_enable();
 }
@@ -7456,7 +7456,7 @@ void init_idle(struct task_struct *idle,
 	__sched_fork(0, idle);
 
 	raw_spin_lock_irqsave(&idle->pi_lock, flags);
-	raw_spin_lock(&rq->lock);
+	raw_spin_rq_lock(rq);
 
 	idle->state = TASK_RUNNING;
 	idle->se.exec_start = sched_clock();
@@ -7494,7 +7494,7 @@ void init_idle(struct task_struct *idle,
 #ifdef CONFIG_SMP
 	idle->on_cpu = 1;
 #endif
-	raw_spin_unlock(&rq->lock);
+	raw_spin_rq_unlock(rq);
 	raw_spin_unlock_irqrestore(&idle->pi_lock, flags);
 
 	/* Set the preempt count _outside_ the spinlocks! */
@@ -7660,7 +7660,7 @@ static void balance_push(struct rq *rq)
 {
 	struct task_struct *push_task = rq->curr;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 	SCHED_WARN_ON(rq->cpu != smp_processor_id());
 
 	/*
@@ -7698,9 +7698,9 @@ static void balance_push(struct rq *rq)
 		 */
 		if (!rq->nr_running && !rq_has_pinned_tasks(rq) &&
 		    rcuwait_active(&rq->hotplug_wait)) {
-			raw_spin_unlock(&rq->lock);
+			raw_spin_rq_unlock(rq);
 			rcuwait_wake_up(&rq->hotplug_wait);
-			raw_spin_lock(&rq->lock);
+			raw_spin_rq_lock(rq);
 		}
 		return;
 	}
@@ -7710,7 +7710,7 @@ static void balance_push(struct rq *rq)
 	 * Temporarily drop rq->lock such that we can wake-up the stop task.
 	 * Both preemption and IRQs are still disabled.
 	 */
-	raw_spin_unlock(&rq->lock);
+	raw_spin_rq_unlock(rq);
 	stop_one_cpu_nowait(rq->cpu, __balance_push_cpu_stop, push_task,
 			    this_cpu_ptr(&push_work));
 	/*
@@ -7718,7 +7718,7 @@ static void balance_push(struct rq *rq)
 	 * schedule(). The next pick is obviously going to be the stop task
 	 * which kthread_is_per_cpu() and will push this task away.
 	 */
-	raw_spin_lock(&rq->lock);
+	raw_spin_rq_lock(rq);
 }
 
 static void balance_push_set(int cpu, bool on)
@@ -8008,7 +8008,7 @@ static void dump_rq_tasks(struct rq *rq,
 	struct task_struct *g, *p;
 	int cpu = cpu_of(rq);
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	printk("%sCPU%d enqueued tasks (%u total):\n", loglvl, cpu, rq->nr_running);
 	for_each_process_thread(g, p) {
@@ -8181,7 +8181,7 @@ void __init sched_init(void)
 		struct rq *rq;
 
 		rq = cpu_rq(i);
-		raw_spin_lock_init(&rq->lock);
+		raw_spin_lock_init(&rq->__lock);
 		rq->nr_running = 0;
 		rq->calc_load_active = 0;
 		rq->calc_load_update = jiffies + LOAD_FREQ;
kernel/sched/cpuacct.c +6 -6
@@ -112,7 +112,7 @@ static u64 cpuacct_cpuusage_read(struct
 	/*
 	 * Take rq->lock to make 64-bit read safe on 32-bit platforms.
 	 */
-	raw_spin_lock_irq(&cpu_rq(cpu)->lock);
+	raw_spin_rq_lock_irq(cpu_rq(cpu));
 #endif
 
 	if (index == CPUACCT_STAT_NSTATS) {
@@ -126,7 +126,7 @@ static u64 cpuacct_cpuusage_read(struct
 	}
 
 #ifndef CONFIG_64BIT
-	raw_spin_unlock_irq(&cpu_rq(cpu)->lock);
+	raw_spin_rq_unlock_irq(cpu_rq(cpu));
 #endif
 
 	return data;
@@ -141,14 +141,14 @@ static void cpuacct_cpuusage_write(struc
 	/*
 	 * Take rq->lock to make 64-bit write safe on 32-bit platforms.
 	 */
-	raw_spin_lock_irq(&cpu_rq(cpu)->lock);
+	raw_spin_rq_lock_irq(cpu_rq(cpu));
 #endif
 
 	for (i = 0; i < CPUACCT_STAT_NSTATS; i++)
 		cpuusage->usages[i] = val;
 
 #ifndef CONFIG_64BIT
-	raw_spin_unlock_irq(&cpu_rq(cpu)->lock);
+	raw_spin_rq_unlock_irq(cpu_rq(cpu));
 #endif
 }
 
@@ -253,13 +253,13 @@ static int cpuacct_all_seq_show(struct s
 			 * Take rq->lock to make 64-bit read safe on 32-bit
 			 * platforms.
 			 */
-			raw_spin_lock_irq(&cpu_rq(cpu)->lock);
+			raw_spin_rq_lock_irq(cpu_rq(cpu));
 #endif
 
 			seq_printf(m, " %llu", cpuusage->usages[index]);
 
 #ifndef CONFIG_64BIT
-			raw_spin_unlock_irq(&cpu_rq(cpu)->lock);
+			raw_spin_rq_unlock_irq(cpu_rq(cpu));
 #endif
 		}
 		seq_puts(m, "\n");
kernel/sched/deadline.c +11 -11
@@ -157,7 +157,7 @@ void __add_running_bw(u64 dl_bw, struct
 {
 	u64 old = dl_rq->running_bw;
 
-	lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock);
+	lockdep_assert_rq_held(rq_of_dl_rq(dl_rq));
 	dl_rq->running_bw += dl_bw;
 	SCHED_WARN_ON(dl_rq->running_bw < old); /* overflow */
 	SCHED_WARN_ON(dl_rq->running_bw > dl_rq->this_bw);
@@ -170,7 +170,7 @@ void __sub_running_bw(u64 dl_bw, struct
 {
 	u64 old = dl_rq->running_bw;
 
-	lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock);
+	lockdep_assert_rq_held(rq_of_dl_rq(dl_rq));
 	dl_rq->running_bw -= dl_bw;
 	SCHED_WARN_ON(dl_rq->running_bw > old); /* underflow */
 	if (dl_rq->running_bw > old)
@@ -184,7 +184,7 @@ void __add_rq_bw(u64 dl_bw, struct dl_rq
 {
 	u64 old = dl_rq->this_bw;
 
-	lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock);
+	lockdep_assert_rq_held(rq_of_dl_rq(dl_rq));
 	dl_rq->this_bw += dl_bw;
 	SCHED_WARN_ON(dl_rq->this_bw < old); /* overflow */
 }
@@ -194,7 +194,7 @@ void __sub_rq_bw(u64 dl_bw, struct dl_rq
 {
 	u64 old = dl_rq->this_bw;
 
-	lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock);
+	lockdep_assert_rq_held(rq_of_dl_rq(dl_rq));
 	dl_rq->this_bw -= dl_bw;
 	SCHED_WARN_ON(dl_rq->this_bw > old); /* underflow */
 	if (dl_rq->this_bw > old)
@@ -987,7 +987,7 @@ static int start_dl_timer(struct task_st
 	ktime_t now, act;
 	s64 delta;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	/*
 	 * We want the timer to fire at the deadline, but considering
@@ -1097,9 +1097,9 @@ static enum hrtimer_restart dl_task_time
 		 * If the runqueue is no longer available, migrate the
 		 * task elsewhere. This necessarily changes rq.
 		 */
-		lockdep_unpin_lock(&rq->lock, rf.cookie);
+		lockdep_unpin_lock(rq_lockp(rq), rf.cookie);
 		rq = dl_task_offline_migration(rq, p);
-		rf.cookie = lockdep_pin_lock(&rq->lock);
+		rf.cookie = lockdep_pin_lock(rq_lockp(rq));
 		update_rq_clock(rq);
 
 		/*
@@ -1731,7 +1731,7 @@ static void migrate_task_rq_dl(struct ta
 	 * from try_to_wake_up(). Hence, p->pi_lock is locked, but
 	 * rq->lock is not... So, lock it
 	 */
-	raw_spin_lock(&rq->lock);
+	raw_spin_rq_lock(rq);
 	if (p->dl.dl_non_contending) {
 		sub_running_bw(&p->dl, &rq->dl);
 		p->dl.dl_non_contending = 0;
@@ -1746,7 +1746,7 @@ static void migrate_task_rq_dl(struct ta
 			put_task_struct(p);
 	}
 	sub_rq_bw(&p->dl, &rq->dl);
-	raw_spin_unlock(&rq->lock);
+	raw_spin_rq_unlock(rq);
 }
 
 static void check_preempt_equal_dl(struct rq *rq, struct task_struct *p)
@@ -2291,10 +2291,10 @@ static void pull_dl_task(struct rq *this
 		double_unlock_balance(this_rq, src_rq);
 
 		if (push_task) {
-			raw_spin_unlock(&this_rq->lock);
+			raw_spin_rq_unlock(this_rq);
 			stop_one_cpu_nowait(src_rq->cpu, push_cpu_stop,
 					    push_task, &src_rq->push_work);
-			raw_spin_lock(&this_rq->lock);
+			raw_spin_rq_lock(this_rq);
 		}
 	}
 
kernel/sched/debug.c +2 -2
@@ -576,7 +576,7 @@ void print_cfs_rq(struct seq_file *m, in
 	SEQ_printf(m, "  .%-30s: %Ld.%06ld\n", "exec_clock",
 			SPLIT_NS(cfs_rq->exec_clock));
 
-	raw_spin_lock_irqsave(&rq->lock, flags);
+	raw_spin_rq_lock_irqsave(rq, flags);
 	if (rb_first_cached(&cfs_rq->tasks_timeline))
 		MIN_vruntime = (__pick_first_entity(cfs_rq))->vruntime;
 	last = __pick_last_entity(cfs_rq);
@@ -584,7 +584,7 @@ void print_cfs_rq(struct seq_file *m, in
 		max_vruntime = last->vruntime;
 	min_vruntime = cfs_rq->min_vruntime;
 	rq0_min_vruntime = cpu_rq(0)->cfs.min_vruntime;
-	raw_spin_unlock_irqrestore(&rq->lock, flags);
+	raw_spin_rq_unlock_irqrestore(rq, flags);
 	SEQ_printf(m, "  .%-30s: %Ld.%06ld\n", "MIN_vruntime",
 			SPLIT_NS(MIN_vruntime));
 	SEQ_printf(m, "  .%-30s: %Ld.%06ld\n", "min_vruntime",
kernel/sched/fair.c +17 -18
@@ -1107,7 +1107,7 @@ struct numa_group {
 static struct numa_group *deref_task_numa_group(struct task_struct *p)
 {
 	return rcu_dereference_check(p->numa_group, p == current ||
-		(lockdep_is_held(&task_rq(p)->lock) && !READ_ONCE(p->on_cpu)));
+		(lockdep_is_held(rq_lockp(task_rq(p))) && !READ_ONCE(p->on_cpu)));
 }
 
 static struct numa_group *deref_curr_numa_group(struct task_struct *p)
@@ -5328,7 +5328,7 @@ static void __maybe_unused update_runtim
 {
 	struct task_group *tg;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(tg, &task_groups, list) {
@@ -5347,7 +5347,7 @@ static void __maybe_unused unthrottle_of
 {
 	struct task_group *tg;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(tg, &task_groups, list) {
@@ -6874,7 +6874,7 @@ static void migrate_task_rq_fair(struct
 		 * In case of TASK_ON_RQ_MIGRATING we in fact hold the 'old'
 		 * rq->lock and can modify state directly.
 		 */
-		lockdep_assert_held(&task_rq(p)->lock);
+		lockdep_assert_rq_held(task_rq(p));
 		detach_entity_cfs_rq(&p->se);
 
 	} else {
@@ -7501,7 +7501,7 @@ static int task_hot(struct task_struct *
 {
 	s64 delta;
 
-	lockdep_assert_held(&env->src_rq->lock);
+	lockdep_assert_rq_held(env->src_rq);
 
 	if (p->sched_class != &fair_sched_class)
 		return 0;
@@ -7599,7 +7599,7 @@ int can_migrate_task(struct task_struct
 {
 	int tsk_cache_hot;
 
-	lockdep_assert_held(&env->src_rq->lock);
+	lockdep_assert_rq_held(env->src_rq);
 
 	/*
 	 * We do not migrate tasks that are:
@@ -7688,7 +7688,7 @@ int can_migrate_task(struct task_struct
  */
 static void detach_task(struct task_struct *p, struct lb_env *env)
 {
-	lockdep_assert_held(&env->src_rq->lock);
+	lockdep_assert_rq_held(env->src_rq);
 
 	deactivate_task(env->src_rq, p, DEQUEUE_NOCLOCK);
 	set_task_cpu(p, env->dst_cpu);
@@ -7704,7 +7704,7 @@ static struct task_struct *detach_one_ta
 {
 	struct task_struct *p;
 
-	lockdep_assert_held(&env->src_rq->lock);
+	lockdep_assert_rq_held(env->src_rq);
 
 	list_for_each_entry_reverse(p,
 			&env->src_rq->cfs_tasks, se.group_node) {
@@ -7740,7 +7740,7 @@ static int detach_tasks(struct lb_env *e
 	struct task_struct *p;
 	int detached = 0;
 
-	lockdep_assert_held(&env->src_rq->lock);
+	lockdep_assert_rq_held(env->src_rq);
 
 	/*
 	 * Source run queue has been emptied by another CPU, clear
@@ -7870,7 +7870,7 @@ static int detach_tasks(struct lb_env *e
  */
 static void attach_task(struct rq *rq, struct task_struct *p)
 {
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	BUG_ON(task_rq(p) != rq);
 	activate_task(rq, p, ENQUEUE_NOCLOCK);
@@ -9781,7 +9781,7 @@ static int load_balance(int this_cpu, st
 		if (need_active_balance(&env)) {
 			unsigned long flags;
 
-			raw_spin_lock_irqsave(&busiest->lock, flags);
+			raw_spin_rq_lock_irqsave(busiest, flags);
 
 			/*
 			 * Don't kick the active_load_balance_cpu_stop,
@@ -9789,8 +9789,7 @@ static int load_balance(int this_cpu, st
 			 * moved to this_cpu:
 			 */
 			if (!cpumask_test_cpu(this_cpu, busiest->curr->cpus_ptr)) {
-				raw_spin_unlock_irqrestore(&busiest->lock,
-							    flags);
+				raw_spin_rq_unlock_irqrestore(busiest, flags);
 				goto out_one_pinned;
 			}
 
@@ -9807,7 +9806,7 @@ static int load_balance(int this_cpu, st
 				busiest->push_cpu = this_cpu;
 				active_balance = 1;
 			}
-			raw_spin_unlock_irqrestore(&busiest->lock, flags);
+			raw_spin_rq_unlock_irqrestore(busiest, flags);
 
 			if (active_balance) {
 				stop_one_cpu_nowait(cpu_of(busiest),
@@ -10624,7 +10623,7 @@ static int newidle_balance(struct rq *th
 		goto out;
 	}
 
-	raw_spin_unlock(&this_rq->lock);
+	raw_spin_rq_unlock(this_rq);
 
 	update_blocked_averages(this_cpu);
 	rcu_read_lock();
@@ -10662,7 +10661,7 @@ static int newidle_balance(struct rq *th
 	}
 	rcu_read_unlock();
 
-	raw_spin_lock(&this_rq->lock);
+	raw_spin_rq_lock(this_rq);
 
 	if (curr_cost > this_rq->max_idle_balance_cost)
 		this_rq->max_idle_balance_cost = curr_cost;
@@ -11143,9 +11142,9 @@ void unregister_fair_sched_group(struct
 
 		rq = cpu_rq(cpu);
 
-		raw_spin_lock_irqsave(&rq->lock, flags);
+		raw_spin_rq_lock_irqsave(rq, flags);
 		list_del_leaf_cfs_rq(tg->cfs_rq[cpu]);
-		raw_spin_unlock_irqrestore(&rq->lock, flags);
+		raw_spin_rq_unlock_irqrestore(rq, flags);
 	}
 }
 
kernel/sched/idle.c +2 -2
@@ -455,10 +455,10 @@ struct task_struct *pick_next_task_idle(
 static void
 dequeue_task_idle(struct rq *rq, struct task_struct *p, int flags)
 {
-	raw_spin_unlock_irq(&rq->lock);
+	raw_spin_rq_unlock_irq(rq);
 	printk(KERN_ERR "bad: scheduling from the idle thread!\n");
 	dump_stack();
-	raw_spin_lock_irq(&rq->lock);
+	raw_spin_rq_lock_irq(rq);
 }
 
 /*
kernel/sched/pelt.h +1 -1
@@ -141,7 +141,7 @@ static inline void update_idle_rq_clock_
 
 static inline u64 rq_clock_pelt(struct rq *rq)
 {
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 	assert_clock_updated(rq);
 
 	return rq->clock_pelt - rq->lost_idle_time;
kernel/sched/rt.c +8 -8
@@ -888,7 +888,7 @@ static int do_sched_rt_period_timer(stru
 		if (skip)
 			continue;
 
-		raw_spin_lock(&rq->lock);
+		raw_spin_rq_lock(rq);
 		update_rq_clock(rq);
 
 		if (rt_rq->rt_time) {
@@ -926,7 +926,7 @@ static int do_sched_rt_period_timer(stru
 
 		if (enqueue)
 			sched_rt_rq_enqueue(rt_rq);
-		raw_spin_unlock(&rq->lock);
+		raw_spin_rq_unlock(rq);
 	}
 
 	if (!throttled && (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF))
@@ -1894,10 +1894,10 @@ static int push_rt_task(struct rq *rq, b
 		 */
 		push_task = get_push_task(rq);
 		if (push_task) {
-			raw_spin_unlock(&rq->lock);
+			raw_spin_rq_unlock(rq);
 			stop_one_cpu_nowait(rq->cpu, push_cpu_stop,
 					    push_task, &rq->push_work);
-			raw_spin_lock(&rq->lock);
+			raw_spin_rq_lock(rq);
 		}
 
 		return 0;
@@ -2122,10 +2122,10 @@ void rto_push_irq_work_func(struct irq_w
 	 * When it gets updated, a check is made if a push is possible.
 	 */
 	if (has_pushable_tasks(rq)) {
-		raw_spin_lock(&rq->lock);
+		raw_spin_rq_lock(rq);
 		while (push_rt_task(rq, true))
 			;
-		raw_spin_unlock(&rq->lock);
+		raw_spin_rq_unlock(rq);
 	}
 
 	raw_spin_lock(&rd->rto_lock);
@@ -2243,10 +2243,10 @@ static void pull_rt_task(struct rq *this
 		double_unlock_balance(this_rq, src_rq);
 
 		if (push_task) {
-			raw_spin_unlock(&this_rq->lock);
+			raw_spin_rq_unlock(this_rq);
 			stop_one_cpu_nowait(src_rq->cpu, push_cpu_stop,
 					    push_task, &src_rq->push_work);
-			raw_spin_lock(&this_rq->lock);
+			raw_spin_rq_lock(this_rq);
 		}
 	}
 
kernel/sched/sched.h +51 -52

274 lines not on this page · raw message

kernel/sched/topology.c +2 -2
@@ -467,7 +467,7 @@ void rq_attach_root(struct rq *rq, struc
 	struct root_domain *old_rd = NULL;
 	unsigned long flags;
 
-	raw_spin_lock_irqsave(&rq->lock, flags);
+	raw_spin_rq_lock_irqsave(rq, flags);
 
 	if (rq->rd) {
 		old_rd = rq->rd;
@@ -493,7 +493,7 @@ void rq_attach_root(struct rq *rq, struc
 	if (cpumask_test_cpu(rq->cpu, cpu_active_mask))
 		set_rq_online(rq);
 
-	raw_spin_unlock_irqrestore(&rq->lock, flags);
+	raw_spin_rq_unlock_irqrestore(rq, flags);
 
 	if (old_rd)
 		call_rcu(&old_rd->rcu, free_rootdomain);
merge bot Applied
#7 lore raw

[tip: sched/core] sched: Wrap rq::lock access

The following commit has been merged into the sched/core branch of tip:

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     5cb9eaa3d274f75539077a28cf01e3563195fa53
Gitweb:        https://git.kernel.org/tip/5cb9eaa3d274f75539077a28cf01e3563195fa53
Author:        Peter Zijlstra <[email protected]>
AuthorDate:    Tue, 17 Nov 2020 18:19:31 -05:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 12 May 2021 11:43:26 +02:00

sched: Wrap rq::lock access

In preparation of playing games with rq->lock, abstract the thing using an accessor.

---
 kernel/sched/core.c     |  70 +++++++++++++-------------
 kernel/sched/cpuacct.c  |  12 ++--
 kernel/sched/deadline.c |  22 ++++----
 kernel/sched/debug.c    |   4 +-
 kernel/sched/fair.c     |  35 ++++++-------
 kernel/sched/idle.c     |   4 +-
 kernel/sched/pelt.h     |   2 +-
 kernel/sched/rt.c       |  16 +++---
 kernel/sched/sched.h    | 105 +++++++++++++++++++--------------------
 kernel/sched/topology.c |   4 +-
 10 files changed, 136 insertions(+), 138 deletions(-)
kernel/sched/core.c +35 -35
index 5568018..5e6f5f5 100644
@@ -211,12 +211,12 @@ struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags *rf)
 
 	for (;;) {
 		rq = task_rq(p);
-		raw_spin_lock(&rq->lock);
+		raw_spin_rq_lock(rq);
 		if (likely(rq == task_rq(p) && !task_on_rq_migrating(p))) {
 			rq_pin_lock(rq, rf);
 			return rq;
 		}
-		raw_spin_unlock(&rq->lock);
+		raw_spin_rq_unlock(rq);
 
 		while (unlikely(task_on_rq_migrating(p)))
 			cpu_relax();
@@ -235,7 +235,7 @@ struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf)
 	for (;;) {
 		raw_spin_lock_irqsave(&p->pi_lock, rf->flags);
 		rq = task_rq(p);
-		raw_spin_lock(&rq->lock);
+		raw_spin_rq_lock(rq);
 		/*
 		 *	move_queued_task()		task_rq_lock()
 		 *
@@ -257,7 +257,7 @@ struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf)
 			rq_pin_lock(rq, rf);
 			return rq;
 		}
-		raw_spin_unlock(&rq->lock);
+		raw_spin_rq_unlock(rq);
 		raw_spin_unlock_irqrestore(&p->pi_lock, rf->flags);
 
 		while (unlikely(task_on_rq_migrating(p)))
@@ -327,7 +327,7 @@ void update_rq_clock(struct rq *rq)
 {
 	s64 delta;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	if (rq->clock_update_flags & RQCF_ACT_SKIP)
 		return;
@@ -625,7 +625,7 @@ void resched_curr(struct rq *rq)
 	struct task_struct *curr = rq->curr;
 	int cpu;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	if (test_tsk_need_resched(curr))
 		return;
@@ -649,10 +649,10 @@ void resched_cpu(int cpu)
 	struct rq *rq = cpu_rq(cpu);
 	unsigned long flags;
 
-	raw_spin_lock_irqsave(&rq->lock, flags);
+	raw_spin_rq_lock_irqsave(rq, flags);
 	if (cpu_online(cpu) || cpu == smp_processor_id())
 		resched_curr(rq);
-	raw_spin_unlock_irqrestore(&rq->lock, flags);
+	raw_spin_rq_unlock_irqrestore(rq, flags);
 }
 
 #ifdef CONFIG_SMP
@@ -1151,7 +1151,7 @@ static inline void uclamp_rq_inc_id(struct rq *rq, struct task_struct *p,
 	struct uclamp_se *uc_se = &p->uclamp[clamp_id];
 	struct uclamp_bucket *bucket;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	/* Update task effective clamp */
 	p->uclamp[clamp_id] = uclamp_eff_get(p, clamp_id);
@@ -1191,7 +1191,7 @@ static inline void uclamp_rq_dec_id(struct rq *rq, struct task_struct *p,
 	unsigned int bkt_clamp;
 	unsigned int rq_clamp;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	/*
 	 * If sched_uclamp_used was enabled after task @p was enqueued,
@@ -1864,7 +1864,7 @@ static inline bool is_cpu_allowed(struct task_struct *p, int cpu)
 static struct rq *move_queued_task(struct rq *rq, struct rq_flags *rf,
 				   struct task_struct *p, int new_cpu)
 {
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	deactivate_task(rq, p, DEQUEUE_NOCLOCK);
 	set_task_cpu(p, new_cpu);
@@ -2038,7 +2038,7 @@ int push_cpu_stop(void *arg)
 	struct task_struct *p = arg;
 
 	raw_spin_lock_irq(&p->pi_lock);
-	raw_spin_lock(&rq->lock);
+	raw_spin_rq_lock(rq);
 
 	if (task_rq(p) != rq)
 		goto out_unlock;
@@ -2068,7 +2068,7 @@ int push_cpu_stop(void *arg)
 
 out_unlock:
 	rq->push_busy = false;
-	raw_spin_unlock(&rq->lock);
+	raw_spin_rq_unlock(rq);
 	raw_spin_unlock_irq(&p->pi_lock);
 
 	put_task_struct(p);
@@ -2121,7 +2121,7 @@ __do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask, u32
 		 * Because __kthread_bind() calls this on blocked tasks without
 		 * holding rq->lock.
 		 */
-		lockdep_assert_held(&rq->lock);
+		lockdep_assert_rq_held(rq);
 		dequeue_task(rq, p, DEQUEUE_SAVE | DEQUEUE_NOCLOCK);
 	}
 	if (running)
@@ -2462,7 +2462,7 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
 	 * task_rq_lock().
 	 */
 	WARN_ON_ONCE(debug_locks && !(lockdep_is_held(&p->pi_lock) ||
-				      lockdep_is_held(&task_rq(p)->lock)));
+				      lockdep_is_held(rq_lockp(task_rq(p)))));
 #endif
 	/*
 	 * Clearly, migrating tasks to offline CPUs is a fairly daft thing.
@@ -3004,7 +3004,7 @@ ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags,
 {
 	int en_flags = ENQUEUE_WAKEUP | ENQUEUE_NOCLOCK;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	if (p->sched_contributes_to_load)
 		rq->nr_uninterruptible--;
@@ -4015,7 +4015,7 @@ static void do_balance_callbacks(struct rq *rq, struct callback_head *head)
 	void (*func)(struct rq *rq);
 	struct callback_head *next;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	while (head) {
 		func = (void (*)(struct rq *))head->func;
@@ -4038,7 +4038,7 @@ static inline struct callback_head *splice_balance_callbacks(struct rq *rq)
 {
 	struct callback_head *head = rq->balance_callback;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 	if (head)
 		rq->balance_callback = NULL;
 
@@ -4055,9 +4055,9 @@ static inline void balance_callbacks(struct rq *rq, struct callback_head *head)
 	unsigned long flags;
 
 	if (unlikely(head)) {
-		raw_spin_lock_irqsave(&rq->lock, flags);
+		raw_spin_rq_lock_irqsave(rq, flags);
 		do_balance_callbacks(rq, head);
-		raw_spin_unlock_irqrestore(&rq->lock, flags);
+		raw_spin_rq_unlock_irqrestore(rq, flags);
 	}
 }
 
@@ -4088,10 +4088,10 @@ prepare_lock_switch(struct rq *rq, struct task_struct *next, struct rq_flags *rf
 	 * do an early lockdep release here:
 	 */
 	rq_unpin_lock(rq, rf);
-	spin_release(&rq->lock.dep_map, _THIS_IP_);
+	spin_release(&rq_lockp(rq)->dep_map, _THIS_IP_);
 #ifdef CONFIG_DEBUG_SPINLOCK
 	/* this is a valid case when another task releases the spinlock */
-	rq->lock.owner = next;
+	rq_lockp(rq)->owner = next;
 #endif
 }
 
@@ -4102,9 +4102,9 @@ static inline void finish_lock_switch(struct rq *rq)
 	 * fix up the runqueue lock - which gets 'carried over' from
 	 * prev into current:
 	 */
-	spin_acquire(&rq->lock.dep_map, 0, 0, _THIS_IP_);
+	spin_acquire(&rq_lockp(rq)->dep_map, 0, 0, _THIS_IP_);
 	__balance_callbacks(rq);
-	raw_spin_unlock_irq(&rq->lock);
+	raw_spin_rq_unlock_irq(rq);
 }
 
 /*
@@ -5164,7 +5164,7 @@ static void __sched notrace __schedule(bool preempt)
 
 		rq_unpin_lock(rq, &rf);
 		__balance_callbacks(rq);
-		raw_spin_unlock_irq(&rq->lock);
+		raw_spin_rq_unlock_irq(rq);
 	}
 }
 
@@ -5706,7 +5706,7 @@ out_unlock:
 
 	rq_unpin_lock(rq, &rf);
 	__balance_callbacks(rq);
-	raw_spin_unlock(&rq->lock);
+	raw_spin_rq_unlock(rq);
 
 	preempt_enable();
 }
@@ -7456,7 +7456,7 @@ void init_idle(struct task_struct *idle, int cpu)
 	__sched_fork(0, idle);
 
 	raw_spin_lock_irqsave(&idle->pi_lock, flags);
-	raw_spin_lock(&rq->lock);
+	raw_spin_rq_lock(rq);
 
 	idle->state = TASK_RUNNING;
 	idle->se.exec_start = sched_clock();
@@ -7494,7 +7494,7 @@ void init_idle(struct task_struct *idle, int cpu)
 #ifdef CONFIG_SMP
 	idle->on_cpu = 1;
 #endif
-	raw_spin_unlock(&rq->lock);
+	raw_spin_rq_unlock(rq);
 	raw_spin_unlock_irqrestore(&idle->pi_lock, flags);
 
 	/* Set the preempt count _outside_ the spinlocks! */
@@ -7660,7 +7660,7 @@ static void balance_push(struct rq *rq)
 {
 	struct task_struct *push_task = rq->curr;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 	SCHED_WARN_ON(rq->cpu != smp_processor_id());
 
 	/*
@@ -7698,9 +7698,9 @@ static void balance_push(struct rq *rq)
 		 */
 		if (!rq->nr_running && !rq_has_pinned_tasks(rq) &&
 		    rcuwait_active(&rq->hotplug_wait)) {
-			raw_spin_unlock(&rq->lock);
+			raw_spin_rq_unlock(rq);
 			rcuwait_wake_up(&rq->hotplug_wait);
-			raw_spin_lock(&rq->lock);
+			raw_spin_rq_lock(rq);
 		}
 		return;
 	}
@@ -7710,7 +7710,7 @@ static void balance_push(struct rq *rq)
 	 * Temporarily drop rq->lock such that we can wake-up the stop task.
 	 * Both preemption and IRQs are still disabled.
 	 */
-	raw_spin_unlock(&rq->lock);
+	raw_spin_rq_unlock(rq);
 	stop_one_cpu_nowait(rq->cpu, __balance_push_cpu_stop, push_task,
 			    this_cpu_ptr(&push_work));
 	/*
@@ -7718,7 +7718,7 @@ static void balance_push(struct rq *rq)
 	 * schedule(). The next pick is obviously going to be the stop task
 	 * which kthread_is_per_cpu() and will push this task away.
 	 */
-	raw_spin_lock(&rq->lock);
+	raw_spin_rq_lock(rq);
 }
 
 static void balance_push_set(int cpu, bool on)
@@ -8008,7 +8008,7 @@ static void dump_rq_tasks(struct rq *rq, const char *loglvl)
 	struct task_struct *g, *p;
 	int cpu = cpu_of(rq);
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	printk("%sCPU%d enqueued tasks (%u total):\n", loglvl, cpu, rq->nr_running);
 	for_each_process_thread(g, p) {
@@ -8181,7 +8181,7 @@ void __init sched_init(void)
 		struct rq *rq;
 
 		rq = cpu_rq(i);
-		raw_spin_lock_init(&rq->lock);
+		raw_spin_lock_init(&rq->__lock);
 		rq->nr_running = 0;
 		rq->calc_load_active = 0;
 		rq->calc_load_update = jiffies + LOAD_FREQ;
kernel/sched/cpuacct.c +6 -6
index 104a1ba..893eece 100644
@@ -112,7 +112,7 @@ static u64 cpuacct_cpuusage_read(struct cpuacct *ca, int cpu,
 	/*
 	 * Take rq->lock to make 64-bit read safe on 32-bit platforms.
 	 */
-	raw_spin_lock_irq(&cpu_rq(cpu)->lock);
+	raw_spin_rq_lock_irq(cpu_rq(cpu));
 #endif
 
 	if (index == CPUACCT_STAT_NSTATS) {
@@ -126,7 +126,7 @@ static u64 cpuacct_cpuusage_read(struct cpuacct *ca, int cpu,
 	}
 
 #ifndef CONFIG_64BIT
-	raw_spin_unlock_irq(&cpu_rq(cpu)->lock);
+	raw_spin_rq_unlock_irq(cpu_rq(cpu));
 #endif
 
 	return data;
@@ -141,14 +141,14 @@ static void cpuacct_cpuusage_write(struct cpuacct *ca, int cpu, u64 val)
 	/*
 	 * Take rq->lock to make 64-bit write safe on 32-bit platforms.
 	 */
-	raw_spin_lock_irq(&cpu_rq(cpu)->lock);
+	raw_spin_rq_lock_irq(cpu_rq(cpu));
 #endif
 
 	for (i = 0; i < CPUACCT_STAT_NSTATS; i++)
 		cpuusage->usages[i] = val;
 
 #ifndef CONFIG_64BIT
-	raw_spin_unlock_irq(&cpu_rq(cpu)->lock);
+	raw_spin_rq_unlock_irq(cpu_rq(cpu));
 #endif
 }
 
@@ -253,13 +253,13 @@ static int cpuacct_all_seq_show(struct seq_file *m, void *V)
 			 * Take rq->lock to make 64-bit read safe on 32-bit
 			 * platforms.
 			 */
-			raw_spin_lock_irq(&cpu_rq(cpu)->lock);
+			raw_spin_rq_lock_irq(cpu_rq(cpu));
 #endif
 
 			seq_printf(m, " %llu", cpuusage->usages[index]);
 
 #ifndef CONFIG_64BIT
-			raw_spin_unlock_irq(&cpu_rq(cpu)->lock);
+			raw_spin_rq_unlock_irq(cpu_rq(cpu));
 #endif
 		}
 		seq_puts(m, "\n");
kernel/sched/deadline.c +11 -11
index 9a29897..6e99b8b 100644
@@ -157,7 +157,7 @@ void __add_running_bw(u64 dl_bw, struct dl_rq *dl_rq)
 {
 	u64 old = dl_rq->running_bw;
 
-	lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock);
+	lockdep_assert_rq_held(rq_of_dl_rq(dl_rq));
 	dl_rq->running_bw += dl_bw;
 	SCHED_WARN_ON(dl_rq->running_bw < old); /* overflow */
 	SCHED_WARN_ON(dl_rq->running_bw > dl_rq->this_bw);
@@ -170,7 +170,7 @@ void __sub_running_bw(u64 dl_bw, struct dl_rq *dl_rq)
 {
 	u64 old = dl_rq->running_bw;
 
-	lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock);
+	lockdep_assert_rq_held(rq_of_dl_rq(dl_rq));
 	dl_rq->running_bw -= dl_bw;
 	SCHED_WARN_ON(dl_rq->running_bw > old); /* underflow */
 	if (dl_rq->running_bw > old)
@@ -184,7 +184,7 @@ void __add_rq_bw(u64 dl_bw, struct dl_rq *dl_rq)
 {
 	u64 old = dl_rq->this_bw;
 
-	lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock);
+	lockdep_assert_rq_held(rq_of_dl_rq(dl_rq));
 	dl_rq->this_bw += dl_bw;
 	SCHED_WARN_ON(dl_rq->this_bw < old); /* overflow */
 }
@@ -194,7 +194,7 @@ void __sub_rq_bw(u64 dl_bw, struct dl_rq *dl_rq)
 {
 	u64 old = dl_rq->this_bw;
 
-	lockdep_assert_held(&(rq_of_dl_rq(dl_rq))->lock);
+	lockdep_assert_rq_held(rq_of_dl_rq(dl_rq));
 	dl_rq->this_bw -= dl_bw;
 	SCHED_WARN_ON(dl_rq->this_bw > old); /* underflow */
 	if (dl_rq->this_bw > old)
@@ -987,7 +987,7 @@ static int start_dl_timer(struct task_struct *p)
 	ktime_t now, act;
 	s64 delta;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	/*
 	 * We want the timer to fire at the deadline, but considering
@@ -1097,9 +1097,9 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer)
 		 * If the runqueue is no longer available, migrate the
 		 * task elsewhere. This necessarily changes rq.
 		 */
-		lockdep_unpin_lock(&rq->lock, rf.cookie);
+		lockdep_unpin_lock(rq_lockp(rq), rf.cookie);
 		rq = dl_task_offline_migration(rq, p);
-		rf.cookie = lockdep_pin_lock(&rq->lock);
+		rf.cookie = lockdep_pin_lock(rq_lockp(rq));
 		update_rq_clock(rq);
 
 		/*
@@ -1731,7 +1731,7 @@ static void migrate_task_rq_dl(struct task_struct *p, int new_cpu __maybe_unused
 	 * from try_to_wake_up(). Hence, p->pi_lock is locked, but
 	 * rq->lock is not... So, lock it
 	 */
-	raw_spin_lock(&rq->lock);
+	raw_spin_rq_lock(rq);
 	if (p->dl.dl_non_contending) {
 		sub_running_bw(&p->dl, &rq->dl);
 		p->dl.dl_non_contending = 0;
@@ -1746,7 +1746,7 @@ static void migrate_task_rq_dl(struct task_struct *p, int new_cpu __maybe_unused
 			put_task_struct(p);
 	}
 	sub_rq_bw(&p->dl, &rq->dl);
-	raw_spin_unlock(&rq->lock);
+	raw_spin_rq_unlock(rq);
 }
 
 static void check_preempt_equal_dl(struct rq *rq, struct task_struct *p)
@@ -2291,10 +2291,10 @@ skip:
 		double_unlock_balance(this_rq, src_rq);
 
 		if (push_task) {
-			raw_spin_unlock(&this_rq->lock);
+			raw_spin_rq_unlock(this_rq);
 			stop_one_cpu_nowait(src_rq->cpu, push_cpu_stop,
 					    push_task, &src_rq->push_work);
-			raw_spin_lock(&this_rq->lock);
+			raw_spin_rq_lock(this_rq);
 		}
 	}
 
kernel/sched/debug.c +2 -2
index 9c882f2..3bdee5f 100644
@@ -576,7 +576,7 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq)
 	SEQ_printf(m, "  .%-30s: %Ld.%06ld\n", "exec_clock",
 			SPLIT_NS(cfs_rq->exec_clock));
 
-	raw_spin_lock_irqsave(&rq->lock, flags);
+	raw_spin_rq_lock_irqsave(rq, flags);
 	if (rb_first_cached(&cfs_rq->tasks_timeline))
 		MIN_vruntime = (__pick_first_entity(cfs_rq))->vruntime;
 	last = __pick_last_entity(cfs_rq);
@@ -584,7 +584,7 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq)
 		max_vruntime = last->vruntime;
 	min_vruntime = cfs_rq->min_vruntime;
 	rq0_min_vruntime = cpu_rq(0)->cfs.min_vruntime;
-	raw_spin_unlock_irqrestore(&rq->lock, flags);
+	raw_spin_rq_unlock_irqrestore(rq, flags);
 	SEQ_printf(m, "  .%-30s: %Ld.%06ld\n", "MIN_vruntime",
 			SPLIT_NS(MIN_vruntime));
 	SEQ_printf(m, "  .%-30s: %Ld.%06ld\n", "min_vruntime",
kernel/sched/fair.c +17 -18
index 6bdbb7b..e50bd75 100644
@@ -1107,7 +1107,7 @@ struct numa_group {
 static struct numa_group *deref_task_numa_group(struct task_struct *p)
 {
 	return rcu_dereference_check(p->numa_group, p == current ||
-		(lockdep_is_held(&task_rq(p)->lock) && !READ_ONCE(p->on_cpu)));
+		(lockdep_is_held(rq_lockp(task_rq(p))) && !READ_ONCE(p->on_cpu)));
 }
 
 static struct numa_group *deref_curr_numa_group(struct task_struct *p)
@@ -5328,7 +5328,7 @@ static void __maybe_unused update_runtime_enabled(struct rq *rq)
 {
 	struct task_group *tg;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(tg, &task_groups, list) {
@@ -5347,7 +5347,7 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq)
 {
 	struct task_group *tg;
 
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	rcu_read_lock();
 	list_for_each_entry_rcu(tg, &task_groups, list) {
@@ -6891,7 +6891,7 @@ static void migrate_task_rq_fair(struct task_struct *p, int new_cpu)
 		 * In case of TASK_ON_RQ_MIGRATING we in fact hold the 'old'
 		 * rq->lock and can modify state directly.
 		 */
-		lockdep_assert_held(&task_rq(p)->lock);
+		lockdep_assert_rq_held(task_rq(p));
 		detach_entity_cfs_rq(&p->se);
 
 	} else {
@@ -7518,7 +7518,7 @@ static int task_hot(struct task_struct *p, struct lb_env *env)
 {
 	s64 delta;
 
-	lockdep_assert_held(&env->src_rq->lock);
+	lockdep_assert_rq_held(env->src_rq);
 
 	if (p->sched_class != &fair_sched_class)
 		return 0;
@@ -7616,7 +7616,7 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env)
 {
 	int tsk_cache_hot;
 
-	lockdep_assert_held(&env->src_rq->lock);
+	lockdep_assert_rq_held(env->src_rq);
 
 	/*
 	 * We do not migrate tasks that are:
@@ -7705,7 +7705,7 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env)
  */
 static void detach_task(struct task_struct *p, struct lb_env *env)
 {
-	lockdep_assert_held(&env->src_rq->lock);
+	lockdep_assert_rq_held(env->src_rq);
 
 	deactivate_task(env->src_rq, p, DEQUEUE_NOCLOCK);
 	set_task_cpu(p, env->dst_cpu);
@@ -7721,7 +7721,7 @@ static struct task_struct *detach_one_task(struct lb_env *env)
 {
 	struct task_struct *p;
 
-	lockdep_assert_held(&env->src_rq->lock);
+	lockdep_assert_rq_held(env->src_rq);
 
 	list_for_each_entry_reverse(p,
 			&env->src_rq->cfs_tasks, se.group_node) {
@@ -7757,7 +7757,7 @@ static int detach_tasks(struct lb_env *env)
 	struct task_struct *p;
 	int detached = 0;
 
-	lockdep_assert_held(&env->src_rq->lock);
+	lockdep_assert_rq_held(env->src_rq);
 
 	/*
 	 * Source run queue has been emptied by another CPU, clear
@@ -7887,7 +7887,7 @@ next:
  */
 static void attach_task(struct rq *rq, struct task_struct *p)
 {
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 
 	BUG_ON(task_rq(p) != rq);
 	activate_task(rq, p, ENQUEUE_NOCLOCK);
@@ -9798,7 +9798,7 @@ more_balance:
 		if (need_active_balance(&env)) {
 			unsigned long flags;
 
-			raw_spin_lock_irqsave(&busiest->lock, flags);
+			raw_spin_rq_lock_irqsave(busiest, flags);
 
 			/*
 			 * Don't kick the active_load_balance_cpu_stop,
@@ -9806,8 +9806,7 @@ more_balance:
 			 * moved to this_cpu:
 			 */
 			if (!cpumask_test_cpu(this_cpu, busiest->curr->cpus_ptr)) {
-				raw_spin_unlock_irqrestore(&busiest->lock,
-							    flags);
+				raw_spin_rq_unlock_irqrestore(busiest, flags);
 				goto out_one_pinned;
 			}
 
@@ -9824,7 +9823,7 @@ more_balance:
 				busiest->push_cpu = this_cpu;
 				active_balance = 1;
 			}
-			raw_spin_unlock_irqrestore(&busiest->lock, flags);
+			raw_spin_rq_unlock_irqrestore(busiest, flags);
 
 			if (active_balance) {
 				stop_one_cpu_nowait(cpu_of(busiest),
@@ -10649,7 +10648,7 @@ static int newidle_balance(struct rq *this_rq, struct rq_flags *rf)
 		goto out;
 	}
 
-	raw_spin_unlock(&this_rq->lock);
+	raw_spin_rq_unlock(this_rq);
 
 	update_blocked_averages(this_cpu);
 	rcu_read_lock();
@@ -10688,7 +10687,7 @@ static int newidle_balance(struct rq *this_rq, struct rq_flags *rf)
 	}
 	rcu_read_unlock();
 
-	raw_spin_lock(&this_rq->lock);
+	raw_spin_rq_lock(this_rq);
 
 	if (curr_cost > this_rq->max_idle_balance_cost)
 		this_rq->max_idle_balance_cost = curr_cost;
@@ -11175,9 +11174,9 @@ void unregister_fair_sched_group(struct task_group *tg)
 
 		rq = cpu_rq(cpu);
 
-		raw_spin_lock_irqsave(&rq->lock, flags);
+		raw_spin_rq_lock_irqsave(rq, flags);
 		list_del_leaf_cfs_rq(tg->cfs_rq[cpu]);
-		raw_spin_unlock_irqrestore(&rq->lock, flags);
+		raw_spin_rq_unlock_irqrestore(rq, flags);
 	}
 }
 
kernel/sched/idle.c +2 -2
index 7ca3d3d..0194768 100644
@@ -455,10 +455,10 @@ struct task_struct *pick_next_task_idle(struct rq *rq)
 static void
 dequeue_task_idle(struct rq *rq, struct task_struct *p, int flags)
 {
-	raw_spin_unlock_irq(&rq->lock);
+	raw_spin_rq_unlock_irq(rq);
 	printk(KERN_ERR "bad: scheduling from the idle thread!\n");
 	dump_stack();
-	raw_spin_lock_irq(&rq->lock);
+	raw_spin_rq_lock_irq(rq);
 }
 
 /*
kernel/sched/pelt.h +1 -1
index 1462846..9ed6d8c 100644
@@ -141,7 +141,7 @@ static inline void update_idle_rq_clock_pelt(struct rq *rq)
 
 static inline u64 rq_clock_pelt(struct rq *rq)
 {
-	lockdep_assert_held(&rq->lock);
+	lockdep_assert_rq_held(rq);
 	assert_clock_updated(rq);
 
 	return rq->clock_pelt - rq->lost_idle_time;
kernel/sched/rt.c +8 -8
index c286e5b..b3d39c3 100644
@@ -888,7 +888,7 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
 		if (skip)
 			continue;
 
-		raw_spin_lock(&rq->lock);
+		raw_spin_rq_lock(rq);
 		update_rq_clock(rq);
 
 		if (rt_rq->rt_time) {
@@ -926,7 +926,7 @@ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun)
 
 		if (enqueue)
 			sched_rt_rq_enqueue(rt_rq);
-		raw_spin_unlock(&rq->lock);
+		raw_spin_rq_unlock(rq);
 	}
 
 	if (!throttled && (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF))
@@ -1894,10 +1894,10 @@ retry:
 		 */
 		push_task = get_push_task(rq);
 		if (push_task) {
-			raw_spin_unlock(&rq->lock);
+			raw_spin_rq_unlock(rq);
 			stop_one_cpu_nowait(rq->cpu, push_cpu_stop,
 					    push_task, &rq->push_work);
-			raw_spin_lock(&rq->lock);
+			raw_spin_rq_lock(rq);
 		}
 
 		return 0;
@@ -2122,10 +2122,10 @@ void rto_push_irq_work_func(struct irq_work *work)
 	 * When it gets updated, a check is made if a push is possible.
 	 */
 	if (has_pushable_tasks(rq)) {
-		raw_spin_lock(&rq->lock);
+		raw_spin_rq_lock(rq);
 		while (push_rt_task(rq, true))
 			;
-		raw_spin_unlock(&rq->lock);
+		raw_spin_rq_unlock(rq);
 	}
 
 	raw_spin_lock(&rd->rto_lock);
@@ -2243,10 +2243,10 @@ skip:
 		double_unlock_balance(this_rq, src_rq);
 
 		if (push_task) {
-			raw_spin_unlock(&this_rq->lock);
+			raw_spin_rq_unlock(this_rq);
 			stop_one_cpu_nowait(src_rq->cpu, push_cpu_stop,
 					    push_task, &src_rq->push_work);
-			raw_spin_lock(&this_rq->lock);
+			raw_spin_rq_lock(this_rq);
 		}
 	}
 
kernel/sched/sched.h +52 -53

275 lines not on this page · raw message

kernel/sched/topology.c +2 -2
index 55a0a24..053115b 100644
@@ -467,7 +467,7 @@ void rq_attach_root(struct rq *rq, struct root_domain *rd)
 	struct root_domain *old_rd = NULL;
 	unsigned long flags;
 
-	raw_spin_lock_irqsave(&rq->lock, flags);
+	raw_spin_rq_lock_irqsave(rq, flags);
 
 	if (rq->rd) {
 		old_rd = rq->rd;
@@ -493,7 +493,7 @@ void rq_attach_root(struct rq *rq, struct root_domain *rd)
 	if (cpumask_test_cpu(rq->cpu, cpu_active_mask))
 		set_rq_online(rq);
 
-	raw_spin_unlock_irqrestore(&rq->lock, flags);
+	raw_spin_rq_unlock_irqrestore(rq, flags);
 
 	if (old_rd)
 		call_rcu(&old_rd->rcu, free_rootdomain);
author maintainer
#8 lore raw

PATCH 4/19 sched: Prepare for Core-wide rq->lock

In reply to Peter Zijlstra

When switching on core-sched, CPUs need to agree which lock to use for their RQ.

The new rule will be that rq->core_enabled will be toggled while holding all rq->__locks that belong to a core. This means we need to double check the rq->core_enabled value after each lock acquire and retry if it changed.

This also has implications for those sites that take multiple RQ locks, they need to be careful that the second lock doesn't end up being the first lock.

Verify the lock pointer after acquiring the first lock, because if they're on the same core, holding any of the rq->__lock instances will pin the core state.

While there, change the rq->__lock order to CPU number, instead of rq address, this greatly simplifies the next patch.

---
 kernel/sched/core.c  |   48 ++++++++++++++++++++++++++++++++++++++++++++++--
 kernel/sched/sched.h |   41 +++++++++++------------------------------
 2 files changed, 57 insertions(+), 32 deletions(-)
kernel/sched/core.c +46 -2
@@ -186,12 +186,37 @@ int sysctl_sched_rt_runtime = 950000;
 
 void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
-	raw_spin_lock_nested(rq_lockp(rq), subclass);
+	raw_spinlock_t *lock;
+
+	if (sched_core_disabled()) {
+		raw_spin_lock_nested(&rq->__lock, subclass);
+		return;
+	}
+
+	for (;;) {
+		lock = rq_lockp(rq);
+		raw_spin_lock_nested(lock, subclass);
+		if (likely(lock == rq_lockp(rq)))
+			return;
+		raw_spin_unlock(lock);
+	}
 }
 
 bool raw_spin_rq_trylock(struct rq *rq)
 {
-	return raw_spin_trylock(rq_lockp(rq));
+	raw_spinlock_t *lock;
+	bool ret;
+
+	if (sched_core_disabled())
+		return raw_spin_trylock(&rq->__lock);
+
+	for (;;) {
+		lock = rq_lockp(rq);
+		ret = raw_spin_trylock(lock);
+		if (!ret || (likely(lock == rq_lockp(rq))))
+			return ret;
+		raw_spin_unlock(lock);
+	}
 }
 
 void raw_spin_rq_unlock(struct rq *rq)
@@ -199,6 +224,25 @@ void raw_spin_rq_unlock(struct rq *rq)
 	raw_spin_unlock(rq_lockp(rq));
 }
 
+#ifdef CONFIG_SMP
+/*
+ * double_rq_lock - safely lock two runqueues
+ */
+void double_rq_lock(struct rq *rq1, struct rq *rq2)
+{
+	lockdep_assert_irqs_disabled();
+
+	if (rq1->cpu > rq2->cpu)
+		swap(rq1, rq2);
+
+	raw_spin_rq_lock(rq1);
+	if (rq_lockp(rq1) == rq_lockp(rq2))
+		return;
+
+	raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+}
+#endif
+
 /*
  * __task_rq_lock - lock the rq @p resides on.
  */
kernel/sched/sched.h +11 -30

72 lines not on this page · raw message

#9 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Hi Peter,

kernel/sched/core.c int sysctl_sched_rt_runtime = 950000; · line 191

Quoted 10 lines
kernel/sched/core.c +3 -1
@@ -186,12 +186,37 @@ int sysctl_sched_rt_runtime = 950000;
 
 void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
-       raw_spin_lock_nested(rq_lockp(rq), subclass);
+       raw_spinlock_t *lock;
+
+       if (sched_core_disabled()) {

Nothing to stop sched_core from being enabled right here? Leading to us potentially taking the wrong lock.

Quoted 12 lines
kernel/sched/core.c +11 -0
+               raw_spin_lock_nested(&rq->__lock, subclass);
+               return;
+       }
+
+       for (;;) {
+               lock = rq_lockp(rq);
+               raw_spin_lock_nested(lock, subclass);
+               if (likely(lock == rq_lockp(rq)))
+                       return;
+               raw_spin_unlock(lock);
+       }
 }
author maintainer
#10 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 6 lines · Josh Don

Hi Peter,

Quoted 4 lines
kernel/sched/core.c +0 -0
@@ -186,12 +186,37 @@ int sysctl_sched_rt_runtime = 950000;
 
Quoted 22 lines

kernel/sched/core.c int sysctl_sched_rt_runtime = 950000; · line 191

Quoted 6 lines
kernel/sched/core.c +3 -1
 void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
-       raw_spin_lock_nested(rq_lockp(rq), subclass);
+       raw_spinlock_t *lock;
+
+       if (sched_core_disabled()) {

Nothing to stop sched_core from being enabled right here? Leading to us potentially taking the wrong lock.

Quoted 12 lines
kernel/sched/core.c +11 -0
+               raw_spin_lock_nested(&rq->__lock, subclass);
+               return;
+       }
+
+       for (;;) {
+               lock = rq_lockp(rq);
+               raw_spin_lock_nested(lock, subclass);
+               if (likely(lock == rq_lockp(rq)))
+                       return;
+               raw_spin_unlock(lock);
+       }
 }

Very good; something like the below seems to be the best I can make of it..

kernel/sched/core.c +20 -4
index f732642e3e09..1a81e9cc9e5d 100644
@@ -290,6 +290,10 @@ static void sched_core_assert_empty(void)
 static void __sched_core_enable(void)
 {
 	static_branch_enable(&__sched_core_enabled);
+	/*
+	 * Ensure raw_spin_rq_*lock*() have completed before flipping.
+	 */
+	synchronize_sched();
 	__sched_core_flip(true);
 	sched_core_assert_empty();
 }
@@ -449,16 +453,22 @@ void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
 	raw_spinlock_t *lock;
 
+	preempt_disable();
 	if (sched_core_disabled()) {
 		raw_spin_lock_nested(&rq->__lock, subclass);
+		/* preempt *MUST* still be disabled here */
+		preempt_enable_no_resched();
 		return;
 	}
 
 	for (;;) {
 		lock = __rq_lockp(rq);
 		raw_spin_lock_nested(lock, subclass);
-		if (likely(lock == __rq_lockp(rq)))
+		if (likely(lock == __rq_lockp(rq))) {
+			/* preempt *MUST* still be disabled here */
+			preempt_enable_no_resched();
 			return;
+		}
 		raw_spin_unlock(lock);
 	}
 }
@@ -468,14 +478,20 @@ bool raw_spin_rq_trylock(struct rq *rq)
 	raw_spinlock_t *lock;
 	bool ret;
 
-	if (sched_core_disabled())
-		return raw_spin_trylock(&rq->__lock);
+	preempt_disable();
+	if (sched_core_disabled()) {
+		ret = raw_spin_trylock(&rq->__lock);
+		preempt_enable();
+		return ret;
+	}
 
 	for (;;) {
 		lock = __rq_lockp(rq);
 		ret = raw_spin_trylock(lock);
-		if (!ret || (likely(lock == __rq_lockp(rq))))
+		if (!ret || (likely(lock == __rq_lockp(rq)))) {
+			preempt_enable();
 			return ret;
+		}
 		raw_spin_unlock(lock);
 	}
 }
#11 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

kernel/sched/core.c static void sched_core_assert_empty(void) · line 296

Quoted 12 lines
kernel/sched/core.c +4 -0
index f732642e3e09..1a81e9cc9e5d 100644
@@ -290,6 +290,10 @@ static void sched_core_assert_empty(void)
 static void __sched_core_enable(void)
 {
        static_branch_enable(&__sched_core_enabled);
+       /*
+        * Ensure raw_spin_rq_*lock*() have completed before flipping.
+        */
+       synchronize_sched();

synchronize_rcu()

Quoted 4 lines
kernel/sched/core.c +0 -0
        __sched_core_flip(true);
        sched_core_assert_empty();
 }

kernel/sched/core.c void raw_spin_rq_lock_nested(struct rq *rq, int subclass) · line 462

Quoted 10 lines
kernel/sched/core.c +3 -0
@@ -449,16 +453,22 @@ void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
        raw_spinlock_t *lock;
 
+       preempt_disable();
        if (sched_core_disabled()) {
                raw_spin_lock_nested(&rq->__lock, subclass);
+               /* preempt *MUST* still be disabled here */
+               preempt_enable_no_resched();
                return;
        }

This approach looks good to me. I'm guessing you went this route instead of doing the re-check after locking in order to optimize the disabled case?

Recommend a comment that the preempt_disable() here pairs with the synchronize_rcu() in __sched_core_enable().

Quoted 13 lines
kernel/sched/core.c +4 -1
        for (;;) {
                lock = __rq_lockp(rq);
                raw_spin_lock_nested(lock, subclass);
-               if (likely(lock == __rq_lockp(rq)))
+               if (likely(lock == __rq_lockp(rq))) {
+                       /* preempt *MUST* still be disabled here */
+                       preempt_enable_no_resched();
                        return;
+               }
                raw_spin_unlock(lock);
        }
 }
#12 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 23 lines · Josh Don

kernel/sched/core.c static void sched_core_assert_empty(void) · line 296

Quoted 13 lines
kernel/sched/core.c +4 -0
index f732642e3e09..1a81e9cc9e5d 100644
@@ -290,6 +290,10 @@ static void sched_core_assert_empty(void)
 static void __sched_core_enable(void)
 {
        static_branch_enable(&__sched_core_enabled);
+       /*
+        * Ensure raw_spin_rq_*lock*() have completed before flipping.
+        */
+       synchronize_sched();

synchronize_rcu()

Quoted 7 lines
kernel/sched/core.c +0 -0
        __sched_core_flip(true);
        sched_core_assert_empty();
 }
@@ -449,16 +453,22 @@ void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
        raw_spinlock_t *lock;
 
Quoted 12 lines

kernel/sched/core.c void raw_spin_rq_lock_nested(struct rq *rq, int subclass) · line 462

Quoted 7 lines
kernel/sched/core.c +3 -0
+       preempt_disable();
        if (sched_core_disabled()) {
                raw_spin_lock_nested(&rq->__lock, subclass);
+               /* preempt *MUST* still be disabled here */
+               preempt_enable_no_resched();
                return;
        }

This approach looks good to me. I'm guessing you went this route instead of doing the re-check after locking in order to optimize the disabled case?

Hi Josh and Peter,

I've been running into soft lookups and hard lockups when running a script that just cycles setting the cookie of a group of processes over and over again.

Unfortunately the only way I can reproduce this is by setting the cookies on qemu. I've tried sysbench, stress-ng but those seem to work just fine.

I'm running Peter's branch and even tried the suggested changes here but still see the same behavior. I enabled panic on hard lockup and here below is a snippet of the log.

Is there anything you'd like me to try or have any debugging you'd like me to do? I'd certainly like to get to the bottom of this.

Thanks and have a great day,

Don

[ 1164.129706] NMI backtrace for cpu 10
[ 1164.129707] CPU: 10 PID: 6994 Comm: CPU 2/KVM Kdump: loaded
Tainted: G          I  L    5.12.0-rc8+ #4
[ 1164.129707] Hardware name: Dell Inc. PowerEdge R640/0W23H8, BIOS
2.9.4 11/06/2020
[ 1164.129708] RIP: 0010:native_queued_spin_lock_slowpath+0x69/0x1e0
[ 1164.129708] Code: 0f 92 c0 0f b6 c0 c1 e0 08 89 c2 8b 07 30 e4 09
d0 a9 00 01 ff ff 75 20 85 c0 75 0c b8 01 00 00 00 66 89 07 5d c3 f3
90 8b 07 <84> c0 75 f8 b8 01 00 00 00 66 89 07 eb ec f6 c4 01 75 04 c6
47 01
[ 1164.129709] RSP: 0018:ffffb35d591e0e20 EFLAGS: 00000002
[ 1164.129710] RAX: 0000000000200101 RBX: ffff9ecb0096c180 RCX: 0000000000000001
[ 1164.129710] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff9ecb0096c180
[ 1164.129711] RBP: ffffb35d591e0e20 R08: 0000000000000000 R09: 000000010002e2a6
[ 1164.129711] R10: 0000000000000000 R11: ffffb35d591e0ff8 R12: ffff9ecb0096c180
[ 1164.129712] R13: 000000000000000a R14: 000000000000000a R15: ffff9e6e8dc88000
[ 1164.129712] FS:  00007f91bd79f700(0000) GS:ffff9ecb00940000(0000)
knlGS:0000000000000000
[ 1164.129713] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 1164.129713] CR2: 000000000255d05c CR3: 000000022eb8c005 CR4: 00000000007726e0
[ 1164.129714] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 1164.129715] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 1164.129715] PKRU: 55555554
[ 1164.129715] Call Trace:
[ 1164.129716]  <IRQ>
[ 1164.129716]  _raw_spin_lock+0x1f/0x30
[ 1164.129717]  raw_spin_rq_lock_nested.part.93+0x2b/0x60
[ 1164.129717]  raw_spin_rq_lock_nested.constprop.126+0x1a/0x20
[ 1164.129718]  scheduler_tick+0x4c/0x240
[ 1164.129718]  ? tick_sched_handle.isra.17+0x60/0x60
[ 1164.129719]  update_process_times+0xab/0xc0
[ 1164.129719]  tick_sched_handle.isra.17+0x21/0x60
[ 1164.129720]  tick_sched_timer+0x6b/0x80
[ 1164.129720]  __hrtimer_run_queues+0x10d/0x230
[ 1164.129721]  hrtimer_interrupt+0xe7/0x230
[ 1164.129721]  __sysvec_apic_timer_interrupt+0x62/0x100
[ 1164.129721]  sysvec_apic_timer_interrupt+0x51/0xa0
[ 1164.129722]  </IRQ>
[ 1164.129722]  asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 1164.129722] RIP: 0010:smp_call_function_single+0xf6/0x130
[ 1164.129723] Code: 00 75 4d 48 83 c4 48 41 5a 5d 49 8d 62 f8 c3 48
89 75 c0 48 8d 75 b0 48 89 55 c8 e8 84 fe ff ff 8b 55 b8 83 e2 01 74
0a f3 90 <8b> 55 b8 83 e2 01 75 f6 eb c0 8b 05 4a d8 b0 01 85 c0 0f 85
67 ff
[ 1164.129724] RSP: 0018:ffffb35d5e20fb20 EFLAGS: 00000202
[ 1164.129725] RAX: 0000000000000000 RBX: ffff9e6e8df12600 RCX: ffffb35d5f2ffcc0
[ 1164.129725] RDX: 0000000000000001 RSI: ffffb35d5e20fb20 RDI: ffffb35d5e20fb20
[ 1164.129725] RBP: ffffb35d5e20fb70 R08: 0000000000001000 R09: 0000000000001907
[ 1164.129726] R10: ffffb35d5e20fb98 R11: 00000000003d0900 R12: 000000000000000a
[ 1164.129726] R13: 000000000000000c R14: 0000000000000000 R15: 0000000000000000
[ 1164.129727]  ? vmclear_error+0x150/0x390 [kvm_intel]
[ 1164.129727]  vmx_vcpu_load_vmcs+0x1b2/0x330 [kvm_intel]
[ 1164.129728]  ? update_load_avg+0x1b8/0x640
[ 1164.129728]  ? vmx_vcpu_load_vmcs+0x1b2/0x330 [kvm_intel]
[ 1164.129728]  ? load_fixmap_gdt+0x23/0x30
[ 1164.129729]  vmx_vcpu_load_vmcs+0x309/0x330 [kvm_intel]
[ 1164.129729]  kvm_arch_vcpu_load+0x41/0x260 [kvm]
[ 1164.129729]  ? __switch_to_xtra+0xe3/0x4e0
[ 1164.129730]  kvm_io_bus_write+0xbc/0xf0 [kvm]
[ 1164.129730]  ? kvm_io_bus_write+0xbc/0xf0 [kvm]
[ 1164.129730]  finish_task_switch+0x184/0x290
[ 1164.129731]  __schedule+0x30a/0x12f0
[ 1164.129731]  schedule+0x40/0xb0
[ 1164.129732]  kvm_vcpu_block+0x57/0x4b0 [kvm]
[ 1164.129732]  kvm_arch_vcpu_ioctl_run+0x3fe/0x1b60 [kvm]
[ 1164.129732]  kvm_vcpu_kick+0x59c/0xa10 [kvm]
[ 1164.129733]  ? kvm_vcpu_kick+0x59c/0xa10 [kvm]
[ 1164.129733]  __x64_sys_ioctl+0x96/0xd0
[ 1164.129733]  do_syscall_64+0x37/0x80
[ 1164.129734]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[ 1164.129734] RIP: 0033:0x7fb1c8a4c317
[ 1164.129735] Code: b3 66 90 48 8b 05 71 4b 2d 00 64 c7 00 26 00 00
00 48 c7 c0 ff ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 b8 10 00 00
00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 41 4b 2d 00 f7 d8 64 89
01 48
[ 1164.129735] RSP: 002b:00007f91bd79e7e8 EFLAGS: 00000246 ORIG_RAX:
0000000000000010
[ 1164.129736] RAX: ffffffffffffffda RBX: 000000000000ae80 RCX: 00007fb1c8a4c317
[ 1164.129737] RDX: 0000000000000000 RSI: 000000000000ae80 RDI: 000000000000001a
[ 1164.129737] RBP: 0000000000000000 R08: 000055f3e6d027c0 R09: 00007f91a80040b8
[ 1164.129738] R10: 00007f91a80050a0 R11: 0000000000000246 R12: 00007fb1c4023010
[ 1164.129739] R13: 000055f3e569d880 R14: 00007fb1cb8e7000 R15: 00007ffd86806f70
Quoted 16 lines

Recommend a comment that the preempt_disable() here pairs with the synchronize_rcu() in __sched_core_enable().

Quoted 13 lines
kernel/sched/core.c +4 -1
        for (;;) {
                lock = __rq_lockp(rq);
                raw_spin_lock_nested(lock, subclass);
-               if (likely(lock == __rq_lockp(rq)))
+               if (likely(lock == __rq_lockp(rq))) {
+                       /* preempt *MUST* still be disabled here */
+                       preempt_enable_no_resched();
                        return;
+               }
                raw_spin_unlock(lock);
        }
 }
#13 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 8 lines · Don Hiatt

Hi Josh and Peter,

I've been running into soft lookups and hard lockups when running a script that just cycles setting the cookie of a group of processes over and over again.

Unfortunately the only way I can reproduce this is by setting the cookies on qemu. I've tried sysbench, stress-ng but those seem to work just fine.

Quoted 6 lines

I'm running Peter's branch and even tried the suggested changes here but still see the same behavior. I enabled panic on hard lockup and here below is a snippet of the log.

Is there anything you'd like me to try or have any debugging you'd like me to do? I'd certainly like to get to the bottom of this.

Hi Don,

I tried to repro using qemu, but did not generate a lockup. Could you provide more details on what your script is doing (or better yet, share the script directly)? I would have expected you to potentially hit a lockup if you were cycling sched_core being enabled and disabled, but it sounds like you are just recreating the cookie for a process group over and over?

Best,
Josh

#14 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 17 lines · Josh Don
Quoted 8 lines · Don Hiatt

Hi Josh and Peter,

I've been running into soft lookups and hard lockups when running a script that just cycles setting the cookie of a group of processes over and over again.

Unfortunately the only way I can reproduce this is by setting the cookies on qemu. I've tried sysbench, stress-ng but those seem to work just fine.

Quoted 6 lines

I'm running Peter's branch and even tried the suggested changes here but still see the same behavior. I enabled panic on hard lockup and here below is a snippet of the log.

Is there anything you'd like me to try or have any debugging you'd like me to do? I'd certainly like to get to the bottom of this.

Quoted 9 lines

Hi Don,

I tried to repro using qemu, but did not generate a lockup. Could you provide more details on what your script is doing (or better yet, share the script directly)? I would have expected you to potentially hit a lockup if you were cycling sched_core being enabled and disabled, but it sounds like you are just recreating the cookie for a process group over and over?

I saw something similar on a bare metal hardware. Also tried the suggested patch here and no luck. Panic stack attached with softlockup_all_cpu_backtrace=1.
(sorry, my system has 192 cpus and somehow putting 184 cpus offline causes system hang without any message...)

My script created the core cookie for two different process groups. The one is for sysbench cpu, the other is for sysbench mysql, mysqld(cookie=0) is
also on the same machine. The number of tasks in each category is the same as the number of CPUs on the system. And cookie is created just during task startup.

Please let me know if the script is needed, I'll push it to github with some cleanup.

Thanks,
-Aubrey

��[BEGIN] 2021/4/28 1:00:11
[ 153.988458] watchdog: BUG: soft lockup - CPU#40 stuck for 22s! [sysbench:7127]
[ 153.995832] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.068739] irq event stamp: 78908
[ 154.072154] hardirqs last enabled at (78907): [<ffffffff81c00d02>] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.082078] hardirqs last disabled at (78908): [<ffffffff81af1d6a>] sysvec_apic_timer_interrupt+0xa/0x90
[ 154.091567] softirqs last enabled at (78906): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.100016] softirqs last disabled at (78901): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.108291] CPU: 40 PID: 7127 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.117000] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.127532] RIP: 0010:smp_call_function_many_cond+0x2b1/0x2e0
[ 154.133288] Code: c7 e8 13 97 47 00 3b 05 e1 b4 b5 01 0f 83 e8 fd ff ff 48 63 c8 48 8b 13 48 03 14 cd 20 e9 79 82 8b 4a 08 83 e1 01 74 0a f3 90 <8b> 4a 08 83 e1 01 75 f6 eb c6 0f 0b e9 97 fd ff ff 48 c7 c2 e0 cc
[ 154.152065] RSP: 0000:ffffc90019377ca0 EFLAGS: 00000202
[ 154.157300] RAX: 0000000000000001 RBX: ffff88b007fff4c0 RCX: 0000000000000001
[ 154.164444] RDX: ffff889804c04be0 RSI: 0000000000000000 RDI: ffff88b007fff4c8
[ 154.171594] RBP: 00000000000000bf R08: 0000000000000000 R09: 0000000000000001
[ 154.178743] R10: ffffc90019377c20 R11: 0000000000000008 R12: ffffffff810801a0
[ 154.185893] R13: 0000000000000001 R14: ffff88b007ffdac0 R15: 0000000000000200
[ 154.193040] FS: 00007f0453fb8700(0000) GS:ffff88b007e00000(0000) knlGS:0000000000000000
[ 154.201136] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.206895] CR2: 00007f0456ee1150 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.214039] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.221179] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.228321] PKRU: 55555554
[ 154.231043] Call Trace:
[ 154.233503] ? flush_tlb_func_common.constprop.10+0x300/0x300
[ 154.239264] ? x86_configure_nx+0x50/0x50
[ 154.243283] ? flush_tlb_func_common.constprop.10+0x300/0x300
[ 154.249039] on_each_cpu_cond_mask+0x36/0xc0
[ 154.253328] flush_tlb_mm_range+0xb2/0x130
[ 154.257435] change_protection+0x178/0xce0
[ 154.261541] ? __lock_acquire+0x90f/0x1a80
[ 154.265650] change_prot_numa+0x15/0x30
[ 154.269506] task_numa_work+0x1b7/0x310
[ 154.273354] task_work_run+0x6a/0xa0
[ 154.276938] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.282176] exit_to_user_mode_prepare+0x1ed/0x200
[ 154.286976] irqentry_exit_to_user_mode+0x5/0x30
[ 154.291603] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.297285] RIP: 0033:0x555bc1620f87
[ 154.301356] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.321091] RSP: 002b:00007f0453fb7d40 EFLAGS: 00000246
[ 154.326800] RAX: 0000000000000e2e RBX: 0000000000008dd3 RCX: 000000000000000a
[ 154.334402] RDX: 0000000000000007 RSI: 0000000000000091 RDI: 00007f0453fb7dc0
[ 154.341996] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.349574] R10: 00007f0453fb7d60 R11: 000000000000a884 R12: 0000000000000001
[ 154.357137] R13: 0000000000004880 R14: 00007f0453fb7d90 R15: 0000555bc39a6d00
[ 154.364691] Sending NMI from CPU 40 to CPUs 0-39,41-191:
[ 154.372345] NMI backtrace for cpu 27
[ 154.372347] CPU: 27 PID: 7048 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.372348] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
** 3514 printk messages dropped **
[ 154.389620] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389620] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389621] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389621] RSP: 0000:ffffc9001862b988 EFLAGS: 00000046
[ 154.389621] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000940000
[ 154.389622] RDX: ffff88b006fff1c0 RSI: 0000000000000080 RDI: ffff88c0293fe380
[ 154.389622] RBP: ffff88c0293fe380 R08: 0000000000940000 R09: 00000000001fd9c0
[ 154.389623] R10: ffffc9001862b938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.389623] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.389623] FS: 00007f045465c700(0000) GS:ffff88b006e00000(0000) knlGS:0000000000000000
[ 154.389623] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389624] CR2: 0000555bc18acb48 CR3: 000000011a18a004 CR4: 00000000007706e0
[ 154.389624] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389624] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389625] PKRU: 55555554
[ 154.389625] Call Trace:
[ 154.389625] do_raw_spin_lock+0xb1/0xc0
[ 154.389625] _raw_spin_lock_nested+0x50/0x60
[ 154.389625] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389626] load_balance+0x2f7/0xd60
[ 154.389626] ? update_blocked_averages+0x500/0x820
[ 154.389626] newidle_balance+0x3d4/0x670
[ 154.389626] balance_fair+0x1b/0x30
[ 154.389627] put_prev_task_balance+0x4b/0x80
[ 154.389627] __schedule+0x57b/0x14d0
[ 154.389627] schedule+0x51/0xc0
[ 154.389627] io_schedule+0x12/0x40
[ 154.389627] __lock_page_killable+0x167/0x250
[ 154.389628] ? file_fdatawait_range+0x20/0x20
[ 154.389628] filemap_fault+0x9b5/0xfa0
[ 154.389628] ? lock_acquire+0x266/0x3d0
[ 154.389628] ? ext4_filemap_fault+0x27/0x40
[ 154.389629] ? ext4_filemap_fault+0x27/0x40
[ 154.389629] ext4_filemap_fault+0x2f/0x40
[ 154.389629] __do_fault+0x30/0xd0
[ 154.389629] __handle_mm_fault+0xf0d/0x12c0
[ 154.389629] handle_mm_fault+0x8d/0x240
[ 154.389630] do_user_addr_fault+0x21d/0x730
[ 154.389630] ? asm_exc_page_fault+0x8/0x30
[ 154.389630] exc_page_fault+0x6e/0x270
[ 154.389630] ? asm_exc_page_fault+0x8/0x30
[ 154.389631] asm_exc_page_fault+0x1e/0x30
[ 154.389631] RIP: 0033:0x555bc1610b55
[ 154.389631] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389632] RSP: 002b:00007f045465bd60 EFLAGS: 00010202
[ 154.389632] RAX: 0000000000000000 RBX: 0000000000001680 RCX: 0000000000000001
[ 154.389632] RDX: 0000000000000000 RSI: 000000000000002d RDI: 000000000000002d
[ 154.389633] RBP: 000000000000002d R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389633] R10: 00007f045465bd60 R11: 000000000000ba24 R12: 0000000000000001
[ 154.389633] R13: 0000000000001680 R14: 00007f045465bd90 R15: 0000000000000000
[ 154.389634] NMI backtrace for cpu 36
[ 154.389634] CPU: 36 PID: 7009 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389634] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389635] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389635] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389635] RSP: 0000:ffffc9001862b988 EFLAGS: 00000046
[ 154.389636] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000940000
[ 154.389636] RDX: ffff88b006fff1c0 RSI: 0000000000000080 RDI: ffff88c0293fe380
[ 154.389636] RBP: ffff88c0293fe380 R08: 0000000000940000 R09: 00000000001fd9c0
[ 154.389637] R10: ffffc9001862b938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.389637] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.389637] FS: 00007f045465c700(0000) GS:ffff88b006e00000(0000) knlGS:0000000000000000
[ 154.389638] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389638] CR2: 0000555bc18acb48 CR3: 000000011a18a004 CR4: 00000000007706e0
[ 154.389638] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389639] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389639] PKRU: 55555554
[ 154.389639] Call Trace:
[ 154.389639] do_raw_spin_lock+0xb1/0xc0
[ 154.389639] _raw_spin_lock_nested+0x50/0x60
[ 154.389640] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389640] load_balance+0x2f7/0xd60
[ 154.389640] ? update_blocked_averages+0x500/0x820
[ 154.389640] newidle_balance+0x3d4/0x670
[ 154.389641] balance_fair+0x1b/0x30
[ 154.389641] put_prev_task_balance+0x4b/0x80
[ 154.389641] __schedule+0x57b/0x14d0
[ 154.389641] schedule+0x51/0xc0
[ 154.389641] io_schedule+0x12/0x40
[ 154.389642] __lock_page_killable+0x167/0x250
[ 154.389642] ? file_fdatawait_range+0x20/0x20
[ 154.389642] filemap_fault+0x9b5/0xfa0
[ 154.389642] ? lock_acquire+0x266/0x3d0
[ 154.389643] ? ext4_filemap_fault+0x27/0x40
[ 154.389643] ? ext4_filemap_fault+0x27/0x40
[ 154.389643] ext4_filemap_fault+0x2f/0x40
[ 154.389643] __do_fault+0x30/0xd0
[ 154.389644] __handle_mm_fault+0xf0d/0x12c0
[ 154.389644] handle_mm_fault+0x8d/0x240
[ 154.389644] do_user_addr_fault+0x21d/0x730
[ 154.389644] ? asm_exc_page_fault+0x8/0x30
[ 154.389644] exc_page_fault+0x6e/0x270
[ 154.389645] ? asm_exc_page_fault+0x8/0x30
[ 154.389645] asm_exc_page_fault+0x1e/0x30
[ 154.389645] RIP: 0033:0x555bc1610b55
[ 154.389646] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389646] RSP: 002b:00007f045465bd60 EFLAGS: 00010202
[ 154.389646] RAX: 0000000000000000 RBX: 0000000000001680 RCX: 0000000000000001
[ 154.389647] RDX: 0000000000000000 RSI: 000000000000002d RDI: 000000000000002d
[ 154.389647] RBP: 000000000000002d R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389647] R10: 00007f045465bd60 R11: 000000000000ba24 R12: 0000000000000001
[ 154.389648] R13: 0000000000001680 R14: 00007f045465bd90 R15: 0000000000000000
[ 154.389648] NMI watchdog: Watchdog detected hard LOCKUP on cpu 37
[ 154.389649] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.389663] irq event stamp: 48450
[ 154.389664] hardirqs last enabled at (48449): [<ffffffff81b05ab4>] _raw_spin_unlock_irq+0x24/0x40
[ 154.389664] hardirqs last disabled at (48450): [<ffffffff81afcf5a>] __schedule+0xd7a/0x14d0
[ 154.389664] softirqs last enabled at (48442): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.389665] softirqs last disabled at (48437): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.389665] CPU: 37 PID: 7160 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389665] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389666] RIP: 0010:native_queued_spin_lock_slowpath+0x194/0x1e0
[ 154.389666] Code: 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 8b 42 08 <85> c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07 85 c0 75
[ 154.389666] RSP: 0000:ffffc90019507988 EFLAGS: 00000046
[ 154.389667] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000980000
[ 154.389667] RDX: ffff88b0073ff1c0 RSI: 0000000000000085 RDI: ffff88c0293fe380
[ 154.389667] RBP: ffff88c0293fe380 R08: 0000000000980000 R09: 00000000001fd9c0
[ 154.389668] R10: ffffc90019507938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.389668] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.389668] FS: 00007f0453ddc700(0000) GS:ffff88b007200000(0000) knlGS:0000000000000000
[ 154.389669] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389669] CR2: 0000555bc18acb48 CR3: 000000011a18a005 CR4: 00000000007706e0
[ 154.389669] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389670] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389670] PKRU: 55555554
[ 154.389670] Call Trace:
[ 154.389670] do_raw_spin_lock+0xb1/0xc0
[ 154.389670] _raw_spin_lock_nested+0x50/0x60
[ 154.389671] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389671] load_balance+0x2f7/0xd60
[ 154.389671] ? update_blocked_averages+0x500/0x820
[ 154.389671] newidle_balance+0x3d4/0x670
[ 154.389672] balance_fair+0x1b/0x30
[ 154.389672] put_prev_task_balance+0x4b/0x80
[ 154.389672] __schedule+0x57b/0x14d0
[ 154.389672] schedule+0x51/0xc0
[ 154.389672] io_schedule+0x12/0x40
[ 154.389673] __lock_page_killable+0x167/0x250
[ 154.389673] ? file_fdatawait_range+0x20/0x20
[ 154.389673] filemap_fault+0x9b5/0xfa0
[ 154.389673] ? lock_acquire+0x266/0x3d0
[ 154.389674] ? ext4_filemap_fault+0x27/0x40
[ 154.389674] ? ext4_filemap_fault+0x27/0x40
[ 154.389674] ext4_filemap_fault+0x2f/0x40
[ 154.389674] __do_fault+0x30/0xd0
[ 154.389675] __handle_mm_fault+0xf0d/0x12c0
[ 154.389675] handle_mm_fault+0x8d/0x240
[ 154.389675] do_user_addr_fault+0x21d/0x730
[ 154.389675] ? asm_exc_page_fault+0x8/0x30
[ 154.389675] exc_page_fault+0x6e/0x270
[ 154.389676] ? asm_exc_page_fault+0x8/0x30
[ 154.389676] asm_exc_page_fault+0x1e/0x30
[ 154.389676] RIP: 0033:0x555bc1610b55
[ 154.389676] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389677] RSP: 002b:00007f0453ddbd60 EFLAGS: 00010202
[ 154.389677] RAX: 0000000000000000 RBX: 0000000000005680 RCX: 0000000000000001
[ 154.389678] RDX: 0000000000000000 RSI: 00000000000000ad RDI: 00000000000000ad
[ 154.389678] RBP: 00000000000000ad R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389678] R10: 00007f0453ddbd60 R11: 000000000000ba24 R12: 0000000000000001
[ 154.389679] R13: 0000000000005680 R14: 00007f0453ddbd90 R15: 0000000000000000
[ 154.389679] NMI backtrace for cpu 37
[ 154.389679] CPU: 37 PID: 7160 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389679] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389680] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389680] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389681] RSP: 0000:ffffc90019507988 EFLAGS: 00000046
[ 154.389681] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000980000
[ 154.389681] RDX: ffff88b0073ff1c0 RSI: 0000000000000085 RDI: ffff88c0293fe380
[ 154.389682] RBP: ffff88c0293fe380 R08: 0000000000980000 R09: 00000000001fd9c0
[ 154.389682] R10: ffffc90019507938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.389682] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.389683] FS: 00007f0453ddc700(0000) GS:ffff88b007200000(0000) knlGS:0000000000000000
[ 154.389683] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389683] CR2: 0000555bc18acb48 CR3: 000000011a18a005 CR4: 00000000007706e0
[ 154.389683] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389684] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389684] PKRU: 55555554
[ 154.389684] Call Trace:
[ 154.389684] do_raw_spin_lock+0xb1/0xc0
[ 154.389685] _raw_spin_lock_nested+0x50/0x60
[ 154.389685] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389685] load_balance+0x2f7/0xd60
[ 154.389685] ? update_blocked_averages+0x500/0x820
[ 154.389686] newidle_balance+0x3d4/0x670
[ 154.389686] balance_fair+0x1b/0x30
[ 154.389686] put_prev_task_balance+0x4b/0x80
[ 154.389686] __schedule+0x57b/0x14d0
[ 154.389686] schedule+0x51/0xc0
[ 154.389687] io_schedule+0x12/0x40
[ 154.389687] __lock_page_killable+0x167/0x250
[ 154.389687] ? file_fdatawait_range+0x20/0x20
[ 154.389687] filemap_fault+0x9b5/0xfa0
[ 154.389688] ? lock_acquire+0x266/0x3d0
[ 154.389688] ? ext4_filemap_fault+0x27/0x40
[ 154.389688] ? ext4_filemap_fault+0x27/0x40
[ 154.389688] ext4_filemap_fault+0x2f/0x40
[ 154.389689] __do_fault+0x30/0xd0
[ 154.389689] __handle_mm_fault+0xf0d/0x12c0
[ 154.389689] handle_mm_fault+0x8d/0x240
[ 154.389689] do_user_addr_fault+0x21d/0x730
[ 154.389690] ? asm_exc_page_fault+0x8/0x30
[ 154.389690] exc_page_fault+0x6e/0x270
[ 154.389690] ? asm_exc_page_fault+0x8/0x30
[ 154.389690] asm_exc_page_fault+0x1e/0x30
[ 154.389690] RIP: 0033:0x555bc1610b55
[ 154.389691] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389691] RSP: 002b:00007f0453ddbd60 EFLAGS: 00010202
[ 154.389692] RAX: 0000000000000000 RBX: 0000000000005680 RCX: 0000000000000001
[ 154.389692] RDX: 0000000000000000 RSI: 00000000000000ad RDI: 00000000000000ad
[ 154.389692] RBP: 00000000000000ad R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389693] R10: 00007f0453ddbd60 R11: 000000000000ba24 R12: 0000000000000001
[ 154.389693] R13: 0000000000005680 R14: 00007f0453ddbd90 R15: 0000000000000000
[ 154.389694] NMI watchdog: Watchdog detected hard LOCKUP on cpu 38
[ 154.389694] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.389708] irq event stamp: 45920
[ 154.389709] hardirqs last enabled at (45919): [<ffffffff81b05ab4>] _raw_spin_unlock_irq+0x24/0x40
[ 154.389709] hardirqs last disabled at (45920): [<ffffffff81afcf5a>] __schedule+0xd7a/0x14d0
[ 154.389709] softirqs last enabled at (45912): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.389710] softirqs last disabled at (45907): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.389710] CPU: 38 PID: 7154 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389710] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389711] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389711] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389712] RSP: 0000:ffffc900194c7988 EFLAGS: 00000046
[ 154.389712] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 00000000009c0000
[ 154.389712] RDX: ffff88b0077ff1c0 RSI: 0000000000000059 RDI: ffff88c0293fe380
[ 154.389713] RBP: ffff88c0293fe380 R08: 00000000009c0000 R09: 00000000001fd9c0
[ 154.389713] R10: ffffc900194c7938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.389713] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.389714] FS: 00007f0453e31700(0000) GS:ffff88b007600000(0000) knlGS:0000000000000000
[ 154.389714] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389714] CR2: 0000555bc18acb48 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.389714] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389715] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389715] PKRU: 55555554
[ 154.389715] Call Trace:
[ 154.389715] do_raw_spin_lock+0xb1/0xc0
[ 154.389716] _raw_spin_lock_nested+0x50/0x60
[ 154.389716] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389716] load_balance+0x2f7/0xd60
[ 154.389716] ? update_blocked_averages+0x500/0x820
[ 154.389717] newidle_balance+0x3d4/0x670
[ 154.389717] balance_fair+0x1b/0x30
[ 154.389717] put_prev_task_balance+0x4b/0x80
[ 154.389717] __schedule+0x57b/0x14d0
[ 154.389718] schedule+0x51/0xc0
[ 154.389718] io_schedule+0x12/0x40
[ 154.389718] __lock_page_killable+0x167/0x250
[ 154.389718] ? file_fdatawait_range+0x20/0x20
[ 154.389718] filemap_fault+0x9b5/0xfa0
[ 154.389719] ? lock_acquire+0x266/0x3d0
[ 154.389719] ? ext4_filemap_fault+0x27/0x40
[ 154.389719] ? ext4_filemap_fault+0x27/0x40
[ 154.389719] ext4_filemap_fault+0x2f/0x40
[ 154.389720] __do_fault+0x30/0xd0
[ 154.389720] __handle_mm_fault+0xf0d/0x12c0
[ 154.389720] handle_mm_fault+0x8d/0x240
[ 154.389720] do_user_addr_fault+0x21d/0x730
[ 154.389721] ? asm_exc_page_fault+0x8/0x30
[ 154.389721] exc_page_fault+0x6e/0x270
[ 154.389721] ? asm_exc_page_fault+0x8/0x30
[ 154.389721] asm_exc_page_fault+0x1e/0x30
[ 154.389721] RIP: 0033:0x555bc1610b55
[ 154.389722] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389722] RSP: 002b:00007f0453e30d60 EFLAGS: 00010206
[ 154.389723] RAX: 0000000000000000 RBX: 0000000000005400 RCX: 0000000000000001
[ 154.389723] RDX: 0000000000000000 RSI: 00000000000000a8 RDI: 00000000000000a8
[ 154.389723] RBP: 00000000000000a8 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389724] R10: 00007f0453e30d60 R11: 000000000000ba1a R12: 0000000000000001
[ 154.389724] R13: 0000000000005400 R14: 00007f0453e30d90 R15: 0000000000000000
[ 154.389724] NMI backtrace for cpu 38
[ 154.389724] CPU: 38 PID: 7154 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389725] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389725] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389726] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389726] RSP: 0000:ffffc900194c7988 EFLAGS: 00000046
[ 154.389726] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 00000000009c0000
[ 154.389727] RDX: ffff88b0077ff1c0 RSI: 0000000000000059 RDI: ffff88c0293fe380
[ 154.389727] RBP: ffff88c0293fe380 R08: 00000000009c0000 R09: 00000000001fd9c0
[ 154.389727] R10: ffffc900194c7938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.389728] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.389728] FS: 00007f0453e31700(0000) GS:ffff88b007600000(0000) knlGS:0000000000000000
[ 154.389728] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389729] CR2: 0000555bc18acb48 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.389729] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389729] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389730] PKRU: 55555554
[ 154.389730] Call Trace:
[ 154.389730] do_raw_spin_lock+0xb1/0xc0
[ 154.389730] _raw_spin_lock_nested+0x50/0x60
[ 154.389730] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389731] load_balance+0x2f7/0xd60
[ 154.389731] ? update_blocked_averages+0x500/0x820
[ 154.389731] newidle_balance+0x3d4/0x670
[ 154.389731] balance_fair+0x1b/0x30
[ 154.389732] put_prev_task_balance+0x4b/0x80
[ 154.389732] __schedule+0x57b/0x14d0
[ 154.389732] schedule+0x51/0xc0
[ 154.389732] io_schedule+0x12/0x40
[ 154.389733] __lock_page_killable+0x167/0x250
[ 154.389733] ? file_fdatawait_range+0x20/0x20
[ 154.389733] filemap_fault+0x9b5/0xfa0
[ 154.389733] ? lock_acquire+0x266/0x3d0
[ 154.389733] ? ext4_filemap_fault+0x27/0x40
[ 154.389734] ? ext4_filemap_fault+0x27/0x40
[ 154.389734] ext4_filemap_fault+0x2f/0x40
[ 154.389734] __do_fault+0x30/0xd0
[ 154.389734] __handle_mm_fault+0xf0d/0x12c0
[ 154.389735] handle_mm_fault+0x8d/0x240
[ 154.389735] do_user_addr_fault+0x21d/0x730
[ 154.389735] ? asm_exc_page_fault+0x8/0x30
[ 154.389735] exc_page_fault+0x6e/0x270
[ 154.389736] ? asm_exc_page_fault+0x8/0x30
[ 154.389736] asm_exc_page_fault+0x1e/0x30
[ 154.389736] RIP: 0033:0x555bc1610b55
[ 154.389736] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389737] RSP: 002b:00007f0453e30d60 EFLAGS: 00010206
[ 154.389737] RAX: 0000000000000000 RBX: 0000000000005400 RCX: 0000000000000001
[ 154.389737] RDX: 0000000000000000 RSI: 00000000000000a8 RDI: 00000000000000a8
[ 154.389738] RBP: 00000000000000a8 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389738] R10: 00007f0453e30d60 R11: 000000000000ba1a R12: 0000000000000001
[ 154.389738] R13: 0000000000005400 R14: 00007f0453e30d90 R15: 0000000000000000
[ 154.389739] NMI watchdog: Watchdog detected hard LOCKUP on cpu 39
[ 154.389739] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.389754] irq event stamp: 145164
[ 154.389754] hardirqs last enabled at (145163): [<ffffffff81af2ebc>] syscall_enter_from_user_mode+0x1c/0x50
[ 154.389754] hardirqs last disabled at (145164): [<ffffffff81afcf5a>] __schedule+0xd7a/0x14d0
[ 154.389755] softirqs last enabled at (143734): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.389755] softirqs last disabled at (143729): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.389755] CPU: 39 PID: 7491 Comm: mysqld Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389756] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389756] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389757] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389757] RSP: 0018:ffffc900166a3a08 EFLAGS: 00000046
[ 154.389757] RAX: 0000000000000000 RBX: ffff889804ffe380 RCX: 0000000000a00000
[ 154.389758] RDX: ffff88b007bff1c0 RSI: 0000000000000050 RDI: ffff889804ffe380
[ 154.389758] RBP: ffff889804ffe380 R08: 0000000000a00000 R09: 00000000001fd9c0
[ 154.389758] R10: ffffc900166a39b8 R11: 00000000046fb5a6 R12: ffff889804ffe380
[ 154.389759] R13: 0000000000000400 R14: 0000000000000000 R15: ffff889804ffe380
[ 154.389759] FS: 00007f9fa7253700(0000) GS:ffff88b007a00000(0000) knlGS:0000000000000000
[ 154.389759] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389759] CR2: 0000557160ee00a0 CR3: 0000000127f0c002 CR4: 00000000007706e0
[ 154.389760] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389760] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389760] PKRU: 55555554
[ 154.389761] Call Trace:
[ 154.389761] do_raw_spin_lock+0xb1/0xc0
[ 154.389761] _raw_spin_lock_nested+0x50/0x60
[ 154.389761] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389761] load_balance+0x2f7/0xd60
[ 154.389762] ? update_blocked_averages+0x500/0x820
[ 154.389762] newidle_balance+0x3d4/0x670
[ 154.389762] balance_fair+0x1b/0x30
[ 154.389762] put_prev_task_balance+0x4b/0x80
[ 154.389763] __schedule+0x57b/0x14d0
[ 154.389763] schedule+0x51/0xc0
[ 154.389763] futex_wait_queue_me+0xb0/0x120
[ 154.389763] futex_wait+0xd9/0x220
[ 154.389763] ? __lock_acquire+0x90f/0x1a80
[ 154.389764] do_futex+0x132/0xa70
[ 154.389764] ? lockdep_hardirqs_on_prepare+0xd9/0x180
[ 154.389764] ? finish_task_switch+0xb1/0x290
[ 154.389764] ? finish_task_switch+0xb1/0x290
[ 154.389765] ? finish_task_switch+0x63/0x290
[ 154.389765] ? __schedule+0x3aa/0x14d0
[ 154.389765] __x64_sys_futex+0x13b/0x1f0
[ 154.389765] ? lockdep_hardirqs_on_prepare+0xd9/0x180
[ 154.389766] do_syscall_64+0x33/0x80
[ 154.389766] entry_SYSCALL_64_after_hwframe+0x44/0xae
[ 154.389766] RIP: 0033:0x7fa0dc7721fd
[ 154.389766] Code: f0 5a 41 5a c3 0f 1f 84 00 00 00 00 00 41 52 52 4d 31 d2 ba 02 00 00 00 81 f6 80 00 00 00 39 d0 75 08 90 b8 ca 00 00 00 0f 05 <89> d0 87 07 85 c0 75 f0 5a 41 5a c3 0f 1f 80 00 00 00 00 48 83 3a
[ 154.389767] RSP: 002b:00007f9fa7247718 EFLAGS: 00000246 ORIG_RAX: 00000000000000ca
[ 154.389767] RAX: ffffffffffffffda RBX: 0000000003f474a8 RCX: 00007fa0dc7721fd
[ 154.389768] RDX: 0000000000000002 RSI: 0000000000000080 RDI: 0000000003f474b8
[ 154.389768] RBP: 00007f9fa72477d0 R08: 0000000003f474b8 R09: 00007ffe81beb080
[ 154.389768] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000003f474b8
[ 154.389769] R13: 00007fa0d0f38f68 R14: 0000000003c3b668 R15: 0000000003f46fc8
[ 154.389769] NMI backtrace for cpu 39
[ 154.389769] CPU: 39 PID: 7491 Comm: mysqld Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389770] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389770] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389770] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389771] RSP: 0018:ffffc900166a3a08 EFLAGS: 00000046
[ 154.389771] RAX: 0000000000000000 RBX: ffff889804ffe380 RCX: 0000000000a00000
[ 154.389771] RDX: ffff88b007bff1c0 RSI: 0000000000000050 RDI: ffff889804ffe380
[ 154.389772] RBP: ffff889804ffe380 R08: 0000000000a00000 R09: 00000000001fd9c0
[ 154.389772] R10: ffffc900166a39b8 R11: 00000000046fb5a6 R12: ffff889804ffe380
[ 154.389772] R13: 0000000000000400 R14: 0000000000000000 R15: ffff889804ffe380
[ 154.389773] FS: 00007f9fa7253700(0000) GS:ffff88b007a00000(0000) knlGS:0000000000000000
[ 154.389773] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389773] CR2: 0000557160ee00a0 CR3: 0000000127f0c002 CR4: 00000000007706e0
[ 154.389774] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389774] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389774] PKRU: 55555554
[ 154.389774] Call Trace:
[ 154.389774] do_raw_spin_lock+0xb1/0xc0
[ 154.389775] _raw_spin_lock_nested+0x50/0x60
[ 154.389775] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389775] load_balance+0x2f7/0xd60
[ 154.389775] ? update_blocked_averages+0x500/0x820
[ 154.389776] newidle_balance+0x3d4/0x670
[ 154.389776] balance_fair+0x1b/0x30
[ 154.389776] put_prev_task_balance+0x4b/0x80
[ 154.389776] __schedule+0x57b/0x14d0
[ 154.389777] schedule+0x51/0xc0
[ 154.389777] futex_wait_queue_me+0xb0/0x120
[ 154.389777] futex_wait+0xd9/0x220
[ 154.389777] ? __lock_acquire+0x90f/0x1a80
[ 154.389778] do_futex+0x132/0xa70
[ 154.389778] ? lockdep_hardirqs_on_prepare+0xd9/0x180
[ 154.389778] ? finish_task_switch+0xb1/0x290
[ 154.389778] ? finish_task_switch+0xb1/0x290
[ 154.389778] ? finish_task_switch+0x63/0x290
[ 154.389779] ? __schedule+0x3aa/0x14d0
[ 154.389779] __x64_sys_futex+0x13b/0x1f0
[ 154.389779] ? lockdep_hardirqs_on_prepare+0xd9/0x180
[ 154.389779] do_syscall_64+0x33/0x80
[ 154.389780] entry_SYSCALL_64_after_hwframe+0x44/0xae
[ 154.389780] RIP: 0033:0x7fa0dc7721fd
[ 154.389780] Code: f0 5a 41 5a c3 0f 1f 84 00 00 00 00 00 41 52 52 4d 31 d2 ba 02 00 00 00 81 f6 80 00 00 00 39 d0 75 08 90 b8 ca 00 00 00 0f 05 <89> d0 87 07 85 c0 75 f0 5a 41 5a c3 0f 1f 80 00 00 00 00 48 83 3a
[ 154.389781] RSP: 002b:00007f9fa7247718 EFLAGS: 00000246 ORIG_RAX: 00000000000000ca
[ 154.389781] RAX: ffffffffffffffda RBX: 0000000003f474a8 RCX: 00007fa0dc7721fd
[ 154.389782] RDX: 0000000000000002 RSI: 0000000000000080 RDI: 0000000003f474b8
[ 154.389782] RBP: 00007f9fa72477d0 R08: 0000000003f474b8 R09: 00007ffe81beb080
[ 154.389782] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000003f474b8
[ 154.389783] R13: 00007fa0d0f38f68 R14: 0000000003c3b668 R15: 0000000003f46fc8
[ 154.389783]
[ 154.389783] =============================
[ 154.389784] [ BUG: Invalid wait context ]
[ 154.389784] 5.12.0-rc8-coresched+ #3 Not tainted
[ 154.389784] -----------------------------
[ 154.389784] sysbench/7127 is trying to lock:
[ 154.389785] ffffffff840b3898 (&port_lock_key){-.-.}-{3:3}, at: serial8250_console_write+0xa2/0x370
[ 154.389786] other info that might help us debug this:
[ 154.389786] context-{2:2}
[ 154.389786] 3 locks held by sysbench/7127:
[ 154.389787] #0: ffff88810be54070 (&mm->mmap_lock#2){++++}-{4:4}, at: task_numa_work+0xe5/0x310
[ 154.389788] #1: ffffffff82b56d40 (console_lock){+.+.}-{0:0}, at: vprintk_emit+0xfe/0x3e0
[ 154.389789] #2: ffffffff82a769e0 (console_owner){-...}-{0:0}, at: console_unlock+0x16b/0x610
[ 154.389790] stack backtrace:
[ 154.389791] CPU: 40 PID: 7127 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389791] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389791] Call Trace:
[ 154.389792] <IRQ>
[ 154.389792] dump_stack+0x96/0xcf
[ 154.389792] __lock_acquire+0x672/0x1a80
[ 154.389792] lock_acquire+0x266/0x3d0
[ 154.389792] ? serial8250_console_write+0xa2/0x370
[ 154.389793] ? console_unlock+0x16b/0x610
[ 154.389793] _raw_spin_lock_irqsave+0x4c/0x90
[ 154.389793] ? serial8250_console_write+0xa2/0x370
[ 154.389793] serial8250_console_write+0xa2/0x370
[ 154.389794] ? console_unlock+0x1c4/0x610
[ 154.389794] ? console_unlock+0x16b/0x610
[ 154.389794] console_unlock+0x5a5/0x610
[ 154.389794] vprintk_emit+0x10b/0x3e0
[ 154.389795] ? cyc2ns_read_begin+0x50/0x50
[ 154.389795] printk+0x52/0x6e
[ 154.389795] watchdog_timer_fn+0x1bb/0x220
[ 154.389795] ? softlockup_fn+0x30/0x30
[ 154.389796] __hrtimer_run_queues+0x1c4/0x440
[ 154.389796] hrtimer_interrupt+0xe5/0x240
[ 154.389796] ? __do_softirq+0x30f/0x432
[ 154.389796] __sysvec_apic_timer_interrupt+0x74/0x1f0
[ 154.389797] sysvec_apic_timer_interrupt+0x4b/0x90
[ 154.389797] </IRQ>
[ 154.389797] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.389797] RIP: 0010:smp_call_function_many_cond+0x2b1/0x2e0
[ 154.389798] Code: c7 e8 13 97 47 00 3b 05 e1 b4 b5 01 0f 83 e8 fd ff ff 48 63 c8 48 8b 13 48 03 14 cd 20 e9 79 82 8b 4a 08 83 e1 01 74 0a f3 90 <8b> 4a 08 83 e1 01 75 f6 eb c6 0f 0b e9 97 fd ff ff 48 c7 c2 e0 cc
[ 154.389798] RSP: 0000:ffffc90019377ca0 EFLAGS: 00000202
[ 154.389798] RAX: 0000000000000001 RBX: ffff88b007fff4c0 RCX: 0000000000000001
[ 154.389799] RDX: ffff889804c04be0 RSI: 0000000000000000 RDI: ffff88b007fff4c8
[ 154.389799] RBP: 00000000000000bf R08: 0000000000000000 R09: 0000000000000001
[ 154.389799] R10: ffffc90019377c20 R11: 0000000000000008 R12: ffffffff810801a0
[ 154.389800] R13: 0000000000000001 R14: ffff88b007ffdac0 R15: 0000000000000200
[ 154.389800] ? x86_configure_nx+0x50/0x50
[ 154.389800] ? flush_tlb_func_common.constprop.10+0x300/0x300
[ 154.389800] ? x86_configure_nx+0x50/0x50
[ 154.389801] ? flush_tlb_func_common.constprop.10+0x300/0x300
[ 154.389801] on_each_cpu_cond_mask+0x36/0xc0
[ 154.389801] flush_tlb_mm_range+0xb2/0x130
[ 154.389801] change_protection+0x178/0xce0
[ 154.389802] ? __lock_acquire+0x90f/0x1a80
[ 154.389802] change_prot_numa+0x15/0x30
[ 154.389802] task_numa_work+0x1b7/0x310
[ 154.389802] task_work_run+0x6a/0xa0
[ 154.389803] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.389803] exit_to_user_mode_prepare+0x1ed/0x200
[ 154.389803] irqentry_exit_to_user_mode+0x5/0x30
[ 154.389803] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.389804] RIP: 0033:0x555bc1620f87
[ 154.389804] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.389804] RSP: 002b:00007f0453fb7d40 EFLAGS: 00000246
[ 154.389805] RAX: 0000000000000e2e RBX: 0000000000008dd3 RCX: 000000000000000a
[ 154.389805] RDX: 0000000000000007 RSI: 0000000000000091 RDI: 00007f0453fb7dc0
[ 154.389805] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389806] R10: 00007f0453fb7d60 R11: 000000000000a884 R12: 0000000000000001
[ 154.389806] R13: 0000000000004880 R14: 00007f0453fb7d90 R15: 0000555bc39a6d00
[ 154.389806] NMI watchdog: Watchdog detected hard LOCKUP on cpu 41
[ 154.389807] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.389821] irq event stamp: 54916
[ 154.389822] hardirqs last enabled at (54915): [<ffffffff81b05ab4>] _raw_spin_unlock_irq+0x24/0x40
[ 154.389822] hardirqs last disabled at (54916): [<ffffffff81afcf5a>] __schedule+0xd7a/0x14d0
[ 154.389822] softirqs last enabled at (54908): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.389823] softirqs last disabled at (54903): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.389823] CPU: 41 PID: 7016 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389823] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389824] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389824] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389824] RSP: 0000:ffffc900186a7988 EFLAGS: 00000046
[ 154.389825] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000a80000
[ 154.389825] RDX: ffff88b0083ff1c0 RSI: 0000000000000021 RDI: ffff88c0293fe380
[ 154.389825] RBP: ffff88c0293fe380 R08: 0000000000a80000 R09: 00000000001fd9c0
[ 154.389826] R10: ffffc900186a7938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.389826] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.389826] FS: 00007f04545f6700(0000) GS:ffff88b008200000(0000) knlGS:0000000000000000
[ 154.389827] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389827] CR2: 0000555bc18acb48 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.389827] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389828] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389828] PKRU: 55555554
[ 154.389828] Call Trace:
[ 154.389828] do_raw_spin_lock+0xb1/0xc0
[ 154.389829] _raw_spin_lock_nested+0x50/0x60
[ 154.389829] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389829] load_balance+0x2f7/0xd60
[ 154.389829] ? update_blocked_averages+0x500/0x820
[ 154.389830] newidle_balance+0x3d4/0x670
[ 154.389830] balance_fair+0x1b/0x30
[ 154.389830] put_prev_task_balance+0x4b/0x80
[ 154.389830] __schedule+0x57b/0x14d0
[ 154.389830] schedule+0x51/0xc0
[ 154.389831] io_schedule+0x12/0x40
[ 154.389831] __lock_page_killable+0x167/0x250
[ 154.389831] ? file_fdatawait_range+0x20/0x20
[ 154.389831] filemap_fault+0x9b5/0xfa0
[ 154.389832] ? lock_acquire+0x266/0x3d0
[ 154.389832] ? ext4_filemap_fault+0x27/0x40
[ 154.389832] ? ext4_filemap_fault+0x27/0x40
[ 154.389832] ext4_filemap_fault+0x2f/0x40
[ 154.389833] __do_fault+0x30/0xd0
[ 154.389833] __handle_mm_fault+0xf0d/0x12c0
[ 154.389833] handle_mm_fault+0x8d/0x240
[ 154.389833] do_user_addr_fault+0x21d/0x730
[ 154.389833] ? asm_exc_page_fault+0x8/0x30
[ 154.389834] exc_page_fault+0x6e/0x270
[ 154.389834] ? asm_exc_page_fault+0x8/0x30
[ 154.389834] asm_exc_page_fault+0x1e/0x30
[ 154.389834] RIP: 0033:0x555bc1610b55
[ 154.389835] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389835] RSP: 002b:00007f04545f5d60 EFLAGS: 00010202
[ 154.389836] RAX: 0000000000000000 RBX: 0000000000001980 RCX: 0000000000000001
[ 154.389836] RDX: 0000000000000000 RSI: 0000000000000033 RDI: 0000000000000033
[ 154.389836] RBP: 0000000000000033 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389836] R10: 00007f04545f5d60 R11: 000000000000ba22 R12: 0000000000000001
[ 154.389837] R13: 0000000000001980 R14: 00007f04545f5d90 R15: 0000000000000000
[ 154.389837] NMI backtrace for cpu 41
[ 154.389837] CPU: 41 PID: 7016 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389838] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389838] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389838] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389839] RSP: 0000:ffffc900186a7988 EFLAGS: 00000046
[ 154.389839] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000a80000
[ 154.389840] RDX: ffff88b0083ff1c0 RSI: 0000000000000021 RDI: ffff88c0293fe380
[ 154.389840] RBP: ffff88c0293fe380 R08: 0000000000a80000 R09: 00000000001fd9c0
[ 154.389840] R10: ffffc900186a7938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.389841] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.389841] FS: 00007f04545f6700(0000) GS:ffff88b008200000(0000) knlGS:0000000000000000
[ 154.389841] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389841] CR2: 0000555bc18acb48 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.389842] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389842] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389842] PKRU: 55555554
[ 154.389842] Call Trace:
[ 154.389843] do_raw_spin_lock+0xb1/0xc0
[ 154.389843] _raw_spin_lock_nested+0x50/0x60
[ 154.389843] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389843] load_balance+0x2f7/0xd60
[ 154.389844] ? update_blocked_averages+0x500/0x820
[ 154.389844] newidle_balance+0x3d4/0x670
[ 154.389844] balance_fair+0x1b/0x30
[ 154.389844] put_prev_task_balance+0x4b/0x80
[ 154.389845] __schedule+0x57b/0x14d0
[ 154.389845] schedule+0x51/0xc0
[ 154.389845] io_schedule+0x12/0x40
[ 154.389845] __lock_page_killable+0x167/0x250
[ 154.389845] ? file_fdatawait_range+0x20/0x20
[ 154.389846] filemap_fault+0x9b5/0xfa0
[ 154.389846] ? lock_acquire+0x266/0x3d0
[ 154.389846] ? ext4_filemap_fault+0x27/0x40
[ 154.389846] ? ext4_filemap_fault+0x27/0x40
[ 154.389847] ext4_filemap_fault+0x2f/0x40
[ 154.389847] __do_fault+0x30/0xd0
[ 154.389847] __handle_mm_fault+0xf0d/0x12c0
[ 154.389847] handle_mm_fault+0x8d/0x240
[ 154.389847] do_user_addr_fault+0x21d/0x730
[ 154.389848] ? asm_exc_page_fault+0x8/0x30
[ 154.389848] exc_page_fault+0x6e/0x270
[ 154.389848] ? asm_exc_page_fault+0x8/0x30
[ 154.389848] asm_exc_page_fault+0x1e/0x30
[ 154.389849] RIP: 0033:0x555bc1610b55
[ 154.389849] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389849] RSP: 002b:00007f04545f5d60 EFLAGS: 00010202
[ 154.389850] RAX: 0000000000000000 RBX: 0000000000001980 RCX: 0000000000000001
[ 154.389850] RDX: 0000000000000000 RSI: 0000000000000033 RDI: 0000000000000033
[ 154.389850] RBP: 0000000000000033 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389851] R10: 00007f04545f5d60 R11: 000000000000ba22 R12: 0000000000000001
[ 154.389851] R13: 0000000000001980 R14: 00007f04545f5d90 R15: 0000000000000000
[ 154.389852] NMI watchdog: Watchdog detected hard LOCKUP on cpu 42
[ 154.389852] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.389867] irq event stamp: 46192
[ 154.389867] hardirqs last enabled at (46191): [<ffffffff81b05ab4>] _raw_spin_unlock_irq+0x24/0x40
[ 154.389867] hardirqs last disabled at (46192): [<ffffffff81afcf5a>] __schedule+0xd7a/0x14d0
[ 154.389868] softirqs last enabled at (46184): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.389868] softirqs last disabled at (46179): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.389868] CPU: 42 PID: 7012 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389869] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389869] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389869] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389870] RSP: 0000:ffffc90018663988 EFLAGS: 00000046
[ 154.389870] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000ac0000
[ 154.389870] RDX: ffff88b0087ff1c0 RSI: 00000000000000ad RDI: ffff88c0293fe380
[ 154.389871] RBP: ffff88c0293fe380 R08: 0000000000ac0000 R09: 00000000001fd9c0
[ 154.389871] R10: ffffc90018663938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.389871] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.389872] FS: 00007f045463a700(0000) GS:ffff88b008600000(0000) knlGS:0000000000000000
[ 154.389872] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389872] CR2: 0000555bc18acb48 CR3: 000000011a18a001 CR4: 00000000007706e0
[ 154.389873] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389873] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389873] PKRU: 55555554
[ 154.389873] Call Trace:
[ 154.389874] do_raw_spin_lock+0xb1/0xc0
[ 154.389874] _raw_spin_lock_nested+0x50/0x60
[ 154.389874] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389874] load_balance+0x2f7/0xd60
[ 154.389874] ? update_blocked_averages+0x500/0x820
[ 154.389875] newidle_balance+0x3d4/0x670
[ 154.389875] balance_fair+0x1b/0x30
[ 154.389875] put_prev_task_balance+0x4b/0x80
[ 154.389875] __schedule+0x57b/0x14d0
[ 154.389876] schedule+0x51/0xc0
[ 154.389876] io_schedule+0x12/0x40
[ 154.389876] __lock_page_killable+0x167/0x250
[ 154.389876] ? file_fdatawait_range+0x20/0x20
[ 154.389876] filemap_fault+0x9b5/0xfa0
[ 154.389877] ? lock_acquire+0x266/0x3d0
[ 154.389877] ? ext4_filemap_fault+0x27/0x40
[ 154.389877] ? ext4_filemap_fault+0x27/0x40
[ 154.389877] ext4_filemap_fault+0x2f/0x40
[ 154.389878] __do_fault+0x30/0xd0
[ 154.389878] __handle_mm_fault+0xf0d/0x12c0
[ 154.389878] handle_mm_fault+0x8d/0x240
[ 154.389878] do_user_addr_fault+0x21d/0x730
[ 154.389879] ? asm_exc_page_fault+0x8/0x30
[ 154.389879] exc_page_fault+0x6e/0x270
[ 154.389879] ? asm_exc_page_fault+0x8/0x30
[ 154.389879] asm_exc_page_fault+0x1e/0x30
[ 154.389879] RIP: 0033:0x555bc1610b55
[ 154.389880] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389880] RSP: 002b:00007f0454639d60 EFLAGS: 00010202
[ 154.389881] RAX: 0000000000000000 RBX: 0000000000001780 RCX: 0000000000000001
[ 154.389881] RDX: 0000000000000000 RSI: 000000000000002f RDI: 000000000000002f
[ 154.389881] RBP: 000000000000002f R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389882] R10: 00007f0454639d60 R11: 000000000000ba22 R12: 0000000000000001
[ 154.389882] R13: 0000000000001780 R14: 00007f0454639d90 R15: 0000000000000000
[ 154.389882] NMI backtrace for cpu 42
[ 154.389882] CPU: 42 PID: 7012 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389883] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389883] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389884] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389884] RSP: 0000:ffffc90018663988 EFLAGS: 00000046
[ 154.389884] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000ac0000
[ 154.389885] RDX: ffff88b0087ff1c0 RSI: 00000000000000ad RDI: ffff88c0293fe380
[ 154.389885] RBP: ffff88c0293fe380 R08: 0000000000ac0000 R09: 00000000001fd9c0
[ 154.389885] R10: ffffc90018663938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.389886] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.389886] FS: 00007f045463a700(0000) GS:ffff88b008600000(0000) knlGS:0000000000000000
[ 154.389886] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389886] CR2: 0000555bc18acb48 CR3: 000000011a18a001 CR4: 00000000007706e0
[ 154.389887] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389887] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389887] PKRU: 55555554
[ 154.389887] Call Trace:
[ 154.389888] do_raw_spin_lock+0xb1/0xc0
[ 154.389888] _raw_spin_lock_nested+0x50/0x60
[ 154.389888] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389888] load_balance+0x2f7/0xd60
[ 154.389889] ? update_blocked_averages+0x500/0x820
[ 154.389889] newidle_balance+0x3d4/0x670
[ 154.389889] balance_fair+0x1b/0x30
[ 154.389889] put_prev_task_balance+0x4b/0x80
[ 154.389889] __schedule+0x57b/0x14d0
[ 154.389890] schedule+0x51/0xc0
[ 154.389890] io_schedule+0x12/0x40
[ 154.389890] __lock_page_killable+0x167/0x250
[ 154.389890] ? file_fdatawait_range+0x20/0x20
[ 154.389891] filemap_fault+0x9b5/0xfa0
[ 154.389891] ? lock_acquire+0x266/0x3d0
[ 154.389891] ? ext4_filemap_fault+0x27/0x40
[ 154.389891] ? ext4_filemap_fault+0x27/0x40
[ 154.389892] ext4_filemap_fault+0x2f/0x40
[ 154.389892] __do_fault+0x30/0xd0
[ 154.389892] __handle_mm_fault+0xf0d/0x12c0
[ 154.389892] handle_mm_fault+0x8d/0x240
[ 154.389893] do_user_addr_fault+0x21d/0x730
[ 154.389893] ? asm_exc_page_fault+0x8/0x30
[ 154.389893] exc_page_fault+0x6e/0x270
[ 154.389893] ? asm_exc_page_fault+0x8/0x30
[ 154.389894] asm_exc_page_fault+0x1e/0x30
[ 154.389894] RIP: 0033:0x555bc1610b55
[ 154.389894] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389895] RSP: 002b:00007f0454639d60 EFLAGS: 00010202
[ 154.389895] RAX: 0000000000000000 RBX: 0000000000001780 RCX: 0000000000000001
[ 154.389895] RDX: 0000000000000000 RSI: 000000000000002f RDI: 000000000000002f
[ 154.389896] RBP: 000000000000002f R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389896] R10: 00007f0454639d60 R11: 000000000000ba22 R12: 0000000000000001
[ 154.389896] R13: 0000000000001780 R14: 00007f0454639d90 R15: 0000000000000000
[ 154.389897] NMI watchdog: Watchdog detected hard LOCKUP on cpu 43
[ 154.389897] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.389912] irq event stamp: 56354
[ 154.389912] hardirqs last enabled at (56353): [<ffffffff81b05ab4>] _raw_spin_unlock_irq+0x24/0x40
[ 154.389912] hardirqs last disabled at (56354): [<ffffffff81afcf5a>] __schedule+0xd7a/0x14d0
[ 154.389913] softirqs last enabled at (56346): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.389913] softirqs last disabled at (56341): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.389913] CPU: 43 PID: 6971 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389914] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389914] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389915] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389915] RSP: 0000:ffffc9001843f988 EFLAGS: 00000046
[ 154.389915] RAX: 0000000000000000 RBX: ffff88b004ffe380 RCX: 0000000000b00000
[ 154.389916] RDX: ffff88b008bff1c0 RSI: 000000000000007c RDI: ffff88b004ffe380
[ 154.389916] RBP: ffff88b004ffe380 R08: 0000000000b00000 R09: 00000000001fd9c0
[ 154.389916] R10: ffffc9001843f938 R11: 000000001e2f75eb R12: ffff88b004ffe380
[ 154.389917] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88b004ffe380
[ 154.389917] FS: 00007f0456d41700(0000) GS:ffff88b008a00000(0000) knlGS:0000000000000000
[ 154.389917] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389917] CR2: 0000555bc18acb48 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.389918] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389918] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389918] PKRU: 55555554
[ 154.389918] Call Trace:
[ 154.389919] do_raw_spin_lock+0xb1/0xc0
[ 154.389919] _raw_spin_lock_nested+0x50/0x60
[ 154.389919] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389919] load_balance+0x2f7/0xd60
[ 154.389920] ? update_blocked_averages+0x500/0x820
[ 154.389920] newidle_balance+0x3d4/0x670
[ 154.389920] balance_fair+0x1b/0x30
[ 154.389920] put_prev_task_balance+0x4b/0x80
[ 154.389921] __schedule+0x57b/0x14d0
[ 154.389921] schedule+0x51/0xc0
[ 154.389921] io_schedule+0x12/0x40
[ 154.389921] __lock_page_killable+0x167/0x250
[ 154.389922] ? file_fdatawait_range+0x20/0x20
[ 154.389922] filemap_fault+0x9b5/0xfa0
[ 154.389922] ? lock_acquire+0x266/0x3d0
[ 154.389922] ? ext4_filemap_fault+0x27/0x40
[ 154.389923] ? ext4_filemap_fault+0x27/0x40
[ 154.389923] ext4_filemap_fault+0x2f/0x40
[ 154.389923] __do_fault+0x30/0xd0
[ 154.389923] __handle_mm_fault+0xf0d/0x12c0
[ 154.389923] handle_mm_fault+0x8d/0x240
[ 154.389924] do_user_addr_fault+0x21d/0x730
[ 154.389924] ? asm_exc_page_fault+0x8/0x30
[ 154.389924] exc_page_fault+0x6e/0x270
[ 154.389924] ? asm_exc_page_fault+0x8/0x30
[ 154.389925] asm_exc_page_fault+0x1e/0x30
[ 154.389925] RIP: 0033:0x555bc1610b55
[ 154.389925] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389926] RSP: 002b:00007f0456d40d60 EFLAGS: 00010206
[ 154.389926] RAX: 0000000000000000 RBX: 0000000000000600 RCX: 0000000000000001
[ 154.389926] RDX: 0000000000000000 RSI: 000000000000000c RDI: 000000000000000c
[ 154.389927] RBP: 000000000000000c R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389927] R10: 00007f0456d40d60 R11: 000000000000ba24 R12: 0000000000000001
[ 154.389927] R13: 0000000000000600 R14: 00007f0456d40d90 R15: 0000000000000000
[ 154.389927] NMI backtrace for cpu 43
[ 154.389928] CPU: 43 PID: 6971 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389928] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389928] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389929] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389929] RSP: 0000:ffffc9001843f988 EFLAGS: 00000046
[ 154.389930] RAX: 0000000000000000 RBX: ffff88b004ffe380 RCX: 0000000000b00000
[ 154.389930] RDX: ffff88b008bff1c0 RSI: 000000000000007c RDI: ffff88b004ffe380
[ 154.389930] RBP: ffff88b004ffe380 R08: 0000000000b00000 R09: 00000000001fd9c0
[ 154.389931] R10: ffffc9001843f938 R11: 000000001e2f75eb R12: ffff88b004ffe380
[ 154.389931] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88b004ffe380
[ 154.389931] FS: 00007f0456d41700(0000) GS:ffff88b008a00000(0000) knlGS:0000000000000000
[ 154.389931] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389932] CR2: 0000555bc18acb48 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.389932] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389932] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389933] PKRU: 55555554
[ 154.389933] Call Trace:
[ 154.389933] do_raw_spin_lock+0xb1/0xc0
[ 154.389933] _raw_spin_lock_nested+0x50/0x60
[ 154.389934] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389934] load_balance+0x2f7/0xd60
[ 154.389934] ? update_blocked_averages+0x500/0x820
[ 154.389934] newidle_balance+0x3d4/0x670
[ 154.389935] balance_fair+0x1b/0x30
[ 154.389935] put_prev_task_balance+0x4b/0x80
[ 154.389935] __schedule+0x57b/0x14d0
[ 154.389935] schedule+0x51/0xc0
[ 154.389935] io_schedule+0x12/0x40
[ 154.389936] __lock_page_killable+0x167/0x250
[ 154.389936] ? file_fdatawait_range+0x20/0x20
[ 154.389936] filemap_fault+0x9b5/0xfa0
[ 154.389936] ? lock_acquire+0x266/0x3d0
[ 154.389937] ? ext4_filemap_fault+0x27/0x40
[ 154.389937] ? ext4_filemap_fault+0x27/0x40
[ 154.389937] ext4_filemap_fault+0x2f/0x40
[ 154.389937] __do_fault+0x30/0xd0
[ 154.389938] __handle_mm_fault+0xf0d/0x12c0
[ 154.389938] handle_mm_fault+0x8d/0x240
[ 154.389938] do_user_addr_fault+0x21d/0x730
[ 154.389938] ? asm_exc_page_fault+0x8/0x30
[ 154.389938] exc_page_fault+0x6e/0x270
[ 154.389939] ? asm_exc_page_fault+0x8/0x30
[ 154.389939] asm_exc_page_fault+0x1e/0x30
[ 154.389939] RIP: 0033:0x555bc1610b55
[ 154.389940] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389940] RSP: 002b:00007f0456d40d60 EFLAGS: 00010206
[ 154.389940] RAX: 0000000000000000 RBX: 0000000000000600 RCX: 0000000000000001
[ 154.389941] RDX: 0000000000000000 RSI: 000000000000000c RDI: 000000000000000c
[ 154.389941] RBP: 000000000000000c R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389941] R10: 00007f0456d40d60 R11: 000000000000ba24 R12: 0000000000000001
[ 154.389942] R13: 0000000000000600 R14: 00007f0456d40d90 R15: 0000000000000000
[ 154.389942] NMI watchdog: Watchdog detected hard LOCKUP on cpu 44
[ 154.389942] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.389957] irq event stamp: 67272
[ 154.389957] hardirqs last enabled at (67271): [<ffffffff81b05ab4>] _raw_spin_unlock_irq+0x24/0x40
[ 154.389958] hardirqs last disabled at (67272): [<ffffffff81afcf5a>] __schedule+0xd7a/0x14d0
[ 154.389958] softirqs last enabled at (67264): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.389958] softirqs last disabled at (67259): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.389959] CPU: 44 PID: 7069 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389959] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389959] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389960] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389960] RSP: 0000:ffffc900190a3988 EFLAGS: 00000046
[ 154.389961] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000b40000
[ 154.389961] RDX: ffff88b008fff1c0 RSI: 0000000000000098 RDI: ffff88c0293fe380
[ 154.389961] RBP: ffff88c0293fe380 R08: 0000000000b40000 R09: 00000000001fd9c0
[ 154.389961] R10: ffffc900190a3938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.389962] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.389962] FS: 00007f04542e8700(0000) GS:ffff88b008e00000(0000) knlGS:0000000000000000
[ 154.389962] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389963] CR2: 0000555bc18acb48 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.389963] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389963] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389964] PKRU: 55555554
[ 154.389964] Call Trace:
[ 154.389964] do_raw_spin_lock+0xb1/0xc0
[ 154.389964] _raw_spin_lock_nested+0x50/0x60
[ 154.389964] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389965] load_balance+0x2f7/0xd60
[ 154.389965] ? update_blocked_averages+0x500/0x820
[ 154.389965] newidle_balance+0x3d4/0x670
[ 154.389965] balance_fair+0x1b/0x30
[ 154.389966] put_prev_task_balance+0x4b/0x80
[ 154.389966] __schedule+0x57b/0x14d0
[ 154.389966] schedule+0x51/0xc0
[ 154.389966] io_schedule+0x12/0x40
[ 154.389966] __lock_page_killable+0x167/0x250
[ 154.389967] ? file_fdatawait_range+0x20/0x20
[ 154.389967] filemap_fault+0x9b5/0xfa0
[ 154.389967] ? lock_acquire+0x266/0x3d0
[ 154.389967] ? ext4_filemap_fault+0x27/0x40
[ 154.389968] ? ext4_filemap_fault+0x27/0x40
[ 154.389968] ext4_filemap_fault+0x2f/0x40
[ 154.389968] __do_fault+0x30/0xd0
[ 154.389968] __handle_mm_fault+0xf0d/0x12c0
[ 154.389968] handle_mm_fault+0x8d/0x240
[ 154.389969] do_user_addr_fault+0x21d/0x730
[ 154.389969] ? asm_exc_page_fault+0x8/0x30
[ 154.389969] exc_page_fault+0x6e/0x270
[ 154.389969] ? asm_exc_page_fault+0x8/0x30
[ 154.389970] asm_exc_page_fault+0x1e/0x30
[ 154.389970] RIP: 0033:0x555bc1610b55
[ 154.389970] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389971] RSP: 002b:00007f04542e7d60 EFLAGS: 00010202
[ 154.389971] RAX: 0000000000000000 RBX: 0000000000003080 RCX: 0000000000000001
[ 154.389971] RDX: 0000000000000000 RSI: 0000000000000061 RDI: 0000000000000061
[ 154.389972] RBP: 0000000000000061 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389972] R10: 00007f04542e7d60 R11: 000000000000ba24 R12: 0000000000000001
[ 154.389972] R13: 0000000000003080 R14: 00007f04542e7d90 R15: 0000000000000000
[ 154.389973] NMI backtrace for cpu 44
[ 154.389973] CPU: 44 PID: 7069 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.389973] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.389974] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.389974] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.389975] RSP: 0000:ffffc900190a3988 EFLAGS: 00000046
[ 154.389975] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000b40000
[ 154.389975] RDX: ffff88b008fff1c0 RSI: 0000000000000098 RDI: ffff88c0293fe380
[ 154.389976] RBP: ffff88c0293fe380 R08: 0000000000b40000 R09: 00000000001fd9c0
[ 154.389976] R10: ffffc900190a3938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.389976] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.389976] FS: 00007f04542e8700(0000) GS:ffff88b008e00000(0000) knlGS:0000000000000000
[ 154.389977] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.389977] CR2: 0000555bc18acb48 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.389977] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.389978] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.389978] PKRU: 55555554
[ 154.389978] Call Trace:
[ 154.389978] do_raw_spin_lock+0xb1/0xc0
[ 154.389978] _raw_spin_lock_nested+0x50/0x60
[ 154.389979] raw_spin_rq_lock_nested+0x51/0x80
[ 154.389979] load_balance+0x2f7/0xd60
[ 154.389979] ? update_blocked_averages+0x500/0x820
[ 154.389979] newidle_balance+0x3d4/0x670
[ 154.389980] balance_fair+0x1b/0x30
[ 154.389980] put_prev_task_balance+0x4b/0x80
[ 154.389980] __schedule+0x57b/0x14d0
[ 154.389980] schedule+0x51/0xc0
[ 154.389981] io_schedule+0x12/0x40
[ 154.389981] __lock_page_killable+0x167/0x250
[ 154.389981] ? file_fdatawait_range+0x20/0x20
[ 154.389981] filemap_fault+0x9b5/0xfa0
[ 154.389982] ? lock_acquire+0x266/0x3d0
[ 154.389982] ? ext4_filemap_fault+0x27/0x40
[ 154.389982] ? ext4_filemap_fault+0x27/0x40
[ 154.389982] ext4_filemap_fault+0x2f/0x40
[ 154.389982] __do_fault+0x30/0xd0
[ 154.389983] __handle_mm_fault+0xf0d/0x12c0
[ 154.389983] handle_mm_fault+0x8d/0x240
[ 154.389983] do_user_addr_fault+0x21d/0x730
[ 154.389983] ? asm_exc_page_fault+0x8/0x30
[ 154.389984] exc_page_fault+0x6e/0x270
[ 154.389984] ? asm_exc_page_fault+0x8/0x30
[ 154.389984] asm_exc_page_fault+0x1e/0x30
[ 154.389984] RIP: 0033:0x555bc1610b55
[ 154.389985] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.389985] RSP: 002b:00007f04542e7d60 EFLAGS: 00010202
[ 154.389985] RAX: 0000000000000000 RBX: 0000000000003080 RCX: 0000000000000001
[ 154.389986] RDX: 0000000000000000 RSI: 0000000000000061 RDI: 0000000000000061
[ 154.389986] RBP: 0000000000000061 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.389987] R10: 00007f04542e7d60 R11: 000000000000ba24 R12: 0000000000000001
[ 154.389987] R13: 0000000000003080 R14: 00007f04542e7d90 R15: 0000000000000000
[ 154.389987] NMI watchdog: Watchdog detected hard LOCKUP on cpu 45
[ 154.389988] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390002] irq event stamp: 45120
[ 154.390003] hardirqs last enabled at (45119): [<ffffffff81b05ab4>] _raw_spin_unlock_irq+0x24/0x40
[ 154.390003] hardirqs last disabled at (45120): [<ffffffff81afcf5a>] __schedule+0xd7a/0x14d0
[ 154.390003] softirqs last enabled at (45112): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390004] softirqs last disabled at (45107): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390004] CPU: 45 PID: 7158 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390004] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390005] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390005] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390005] RSP: 0000:ffffc900194df988 EFLAGS: 00000046
[ 154.390006] RAX: 0000000000000000 RBX: ffff88b004ffe380 RCX: 0000000000b80000
[ 154.390006] RDX: ffff88b0093ff1c0 RSI: 000000000000002b RDI: ffff88b004ffe380
[ 154.390006] RBP: ffff88b004ffe380 R08: 0000000000b80000 R09: 00000000001fd9c0
[ 154.390007] R10: ffffc900194df938 R11: 000000001e2f75eb R12: ffff88b004ffe380
[ 154.390007] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88b004ffe380
[ 154.390007] FS: 00007f0453dfe700(0000) GS:ffff88b009200000(0000) knlGS:0000000000000000
[ 154.390008] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390008] CR2: 0000555bc18acb48 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.390008] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390008] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390009] PKRU: 55555554
[ 154.390009] Call Trace:
[ 154.390009] do_raw_spin_lock+0xb1/0xc0
[ 154.390009] _raw_spin_lock_nested+0x50/0x60
[ 154.390010] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390010] load_balance+0x2f7/0xd60
[ 154.390010] ? update_blocked_averages+0x500/0x820
[ 154.390010] newidle_balance+0x3d4/0x670
[ 154.390010] balance_fair+0x1b/0x30
[ 154.390011] put_prev_task_balance+0x4b/0x80
[ 154.390011] __schedule+0x57b/0x14d0
[ 154.390011] schedule+0x51/0xc0
[ 154.390011] io_schedule+0x12/0x40
[ 154.390012] __lock_page_killable+0x167/0x250
[ 154.390012] ? file_fdatawait_range+0x20/0x20
[ 154.390012] filemap_fault+0x9b5/0xfa0
[ 154.390012] ? lock_acquire+0x266/0x3d0
[ 154.390013] ? ext4_filemap_fault+0x27/0x40
[ 154.390013] ? ext4_filemap_fault+0x27/0x40
[ 154.390013] ext4_filemap_fault+0x2f/0x40
[ 154.390013] __do_fault+0x30/0xd0
[ 154.390013] __handle_mm_fault+0xf0d/0x12c0
[ 154.390014] handle_mm_fault+0x8d/0x240
[ 154.390014] do_user_addr_fault+0x21d/0x730
[ 154.390014] ? asm_exc_page_fault+0x8/0x30
[ 154.390014] exc_page_fault+0x6e/0x270
[ 154.390015] ? asm_exc_page_fault+0x8/0x30
[ 154.390015] asm_exc_page_fault+0x1e/0x30
[ 154.390015] RIP: 0033:0x555bc1610b55
[ 154.390015] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.390016] RSP: 002b:00007f0453dfdd60 EFLAGS: 00010202
[ 154.390016] RAX: 0000000000000000 RBX: 0000000000005580 RCX: 0000000000000001
[ 154.390017] RDX: 0000000000000000 RSI: 00000000000000ab RDI: 00000000000000ab
[ 154.390017] RBP: 00000000000000ab R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390017] R10: 00007f0453dfdd60 R11: 000000000000ba24 R12: 0000000000000001
[ 154.390017] R13: 0000000000005580 R14: 00007f0453dfdd90 R15: 0000000000000000
[ 154.390018] NMI backtrace for cpu 45
[ 154.390018] CPU: 45 PID: 7158 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390018] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390019] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390019] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390020] RSP: 0000:ffffc900194df988 EFLAGS: 00000046
[ 154.390020] RAX: 0000000000000000 RBX: ffff88b004ffe380 RCX: 0000000000b80000
[ 154.390020] RDX: ffff88b0093ff1c0 RSI: 000000000000002b RDI: ffff88b004ffe380
[ 154.390021] RBP: ffff88b004ffe380 R08: 0000000000b80000 R09: 00000000001fd9c0
[ 154.390021] R10: ffffc900194df938 R11: 000000001e2f75eb R12: ffff88b004ffe380
[ 154.390021] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88b004ffe380
[ 154.390021] FS: 00007f0453dfe700(0000) GS:ffff88b009200000(0000) knlGS:0000000000000000
[ 154.390022] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390022] CR2: 0000555bc18acb48 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.390022] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390023] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390023] PKRU: 55555554
[ 154.390023] Call Trace:
[ 154.390023] do_raw_spin_lock+0xb1/0xc0
[ 154.390023] _raw_spin_lock_nested+0x50/0x60
[ 154.390024] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390024] load_balance+0x2f7/0xd60
[ 154.390024] ? update_blocked_averages+0x500/0x820
[ 154.390024] newidle_balance+0x3d4/0x670
[ 154.390025] balance_fair+0x1b/0x30
[ 154.390025] put_prev_task_balance+0x4b/0x80
[ 154.390025] __schedule+0x57b/0x14d0
[ 154.390025] schedule+0x51/0xc0
[ 154.390026] io_schedule+0x12/0x40
[ 154.390026] __lock_page_killable+0x167/0x250
[ 154.390026] ? file_fdatawait_range+0x20/0x20
[ 154.390026] filemap_fault+0x9b5/0xfa0
[ 154.390026] ? lock_acquire+0x266/0x3d0
[ 154.390027] ? ext4_filemap_fault+0x27/0x40
[ 154.390027] ? ext4_filemap_fault+0x27/0x40
[ 154.390027] ext4_filemap_fault+0x2f/0x40
[ 154.390028] __do_fault+0x30/0xd0
[ 154.390028] __handle_mm_fault+0xf0d/0x12c0
[ 154.390028] handle_mm_fault+0x8d/0x240
[ 154.390028] do_user_addr_fault+0x21d/0x730
[ 154.390028] ? asm_exc_page_fault+0x8/0x30
[ 154.390029] exc_page_fault+0x6e/0x270
[ 154.390029] ? asm_exc_page_fault+0x8/0x30
[ 154.390029] asm_exc_page_fault+0x1e/0x30
[ 154.390029] RIP: 0033:0x555bc1610b55
[ 154.390030] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.390030] RSP: 002b:00007f0453dfdd60 EFLAGS: 00010202
[ 154.390031] RAX: 0000000000000000 RBX: 0000000000005580 RCX: 0000000000000001
[ 154.390031] RDX: 0000000000000000 RSI: 00000000000000ab RDI: 00000000000000ab
[ 154.390031] RBP: 00000000000000ab R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390031] R10: 00007f0453dfdd60 R11: 000000000000ba24 R12: 0000000000000001
[ 154.390032] R13: 0000000000005580 R14: 00007f0453dfdd90 R15: 0000000000000000
[ 154.390032] NMI watchdog: Watchdog detected hard LOCKUP on cpu 46
[ 154.390033] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390047] irq event stamp: 46730
[ 154.390048] hardirqs last enabled at (46729): [<ffffffff81b05ab4>] _raw_spin_unlock_irq+0x24/0x40
[ 154.390048] hardirqs last disabled at (46730): [<ffffffff81afcf5a>] __schedule+0xd7a/0x14d0
[ 154.390048] softirqs last enabled at (46722): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390049] softirqs last disabled at (46717): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390049] CPU: 46 PID: 7155 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390049] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390050] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390050] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390051] RSP: 0000:ffffc900194cf988 EFLAGS: 00000046
[ 154.390051] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000bc0000
[ 154.390051] RDX: ffff88b0097ff1c0 RSI: 00000000000000bb RDI: ffff88c0293fe380
[ 154.390052] RBP: ffff88c0293fe380 R08: 0000000000bc0000 R09: 00000000001fd9c0
[ 154.390052] R10: ffffc900194cf938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.390052] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.390052] FS: 00007f0453e20700(0000) GS:ffff88b009600000(0000) knlGS:0000000000000000
[ 154.390053] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390053] CR2: 0000555bc18acb48 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390053] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390054] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390054] PKRU: 55555554
[ 154.390054] Call Trace:
[ 154.390054] do_raw_spin_lock+0xb1/0xc0
[ 154.390054] _raw_spin_lock_nested+0x50/0x60
[ 154.390055] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390055] load_balance+0x2f7/0xd60
[ 154.390055] ? update_blocked_averages+0x500/0x820
[ 154.390056] newidle_balance+0x3d4/0x670
[ 154.390056] balance_fair+0x1b/0x30
[ 154.390056] put_prev_task_balance+0x4b/0x80
[ 154.390056] __schedule+0x57b/0x14d0
[ 154.390056] schedule+0x51/0xc0
[ 154.390057] io_schedule+0x12/0x40
[ 154.390057] __lock_page_killable+0x167/0x250
[ 154.390057] ? file_fdatawait_range+0x20/0x20
[ 154.390057] filemap_fault+0x9b5/0xfa0
[ 154.390058] ? lock_acquire+0x266/0x3d0
[ 154.390058] ? ext4_filemap_fault+0x27/0x40
[ 154.390058] ? ext4_filemap_fault+0x27/0x40
[ 154.390058] ext4_filemap_fault+0x2f/0x40
[ 154.390059] __do_fault+0x30/0xd0
[ 154.390059] __handle_mm_fault+0xf0d/0x12c0
[ 154.390059] handle_mm_fault+0x8d/0x240
[ 154.390059] do_user_addr_fault+0x21d/0x730
[ 154.390059] ? asm_exc_page_fault+0x8/0x30
[ 154.390060] exc_page_fault+0x6e/0x270
[ 154.390060] ? asm_exc_page_fault+0x8/0x30
[ 154.390060] asm_exc_page_fault+0x1e/0x30
[ 154.390060] RIP: 0033:0x555bc1610b55
[ 154.390061] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.390061] RSP: 002b:00007f0453e1fd60 EFLAGS: 00010202
[ 154.390062] RAX: 0000000000000000 RBX: 0000000000005480 RCX: 0000000000000001
[ 154.390062] RDX: 0000000000000000 RSI: 00000000000000a9 RDI: 00000000000000a9
[ 154.390062] RBP: 00000000000000a9 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390063] R10: 00007f0453e1fd60 R11: 000000000000ba1e R12: 0000000000000001
[ 154.390063] R13: 0000000000005480 R14: 00007f0453e1fd90 R15: 0000000000000000
[ 154.390063] NMI backtrace for cpu 46
[ 154.390063] CPU: 46 PID: 7155 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390064] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390064] RIP: 0010:native_queued_spin_lock_slowpath+0x194/0x1e0
[ 154.390064] Code: 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 8b 42 08 <85> c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07 85 c0 75
[ 154.390065] RSP: 0000:ffffc900194cf988 EFLAGS: 00000046
[ 154.390065] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000bc0000
[ 154.390066] RDX: ffff88b0097ff1c0 RSI: 00000000000000bb RDI: ffff88c0293fe380
[ 154.390066] RBP: ffff88c0293fe380 R08: 0000000000bc0000 R09: 00000000001fd9c0
[ 154.390066] R10: ffffc900194cf938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.390066] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.390067] FS: 00007f0453e20700(0000) GS:ffff88b009600000(0000) knlGS:0000000000000000
[ 154.390067] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390067] CR2: 0000555bc18acb48 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390068] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390068] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390068] PKRU: 55555554
[ 154.390068] Call Trace:
[ 154.390069] do_raw_spin_lock+0xb1/0xc0
[ 154.390069] _raw_spin_lock_nested+0x50/0x60
[ 154.390069] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390069] load_balance+0x2f7/0xd60
[ 154.390070] ? update_blocked_averages+0x500/0x820
[ 154.390070] newidle_balance+0x3d4/0x670
[ 154.390070] balance_fair+0x1b/0x30
[ 154.390070] put_prev_task_balance+0x4b/0x80
[ 154.390070] __schedule+0x57b/0x14d0
[ 154.390071] schedule+0x51/0xc0
[ 154.390071] io_schedule+0x12/0x40
[ 154.390071] __lock_page_killable+0x167/0x250
[ 154.390071] ? file_fdatawait_range+0x20/0x20
[ 154.390072] filemap_fault+0x9b5/0xfa0
[ 154.390072] ? lock_acquire+0x266/0x3d0
[ 154.390072] ? ext4_filemap_fault+0x27/0x40
[ 154.390072] ? ext4_filemap_fault+0x27/0x40
[ 154.390073] ext4_filemap_fault+0x2f/0x40
[ 154.390073] __do_fault+0x30/0xd0
[ 154.390073] __handle_mm_fault+0xf0d/0x12c0
[ 154.390073] handle_mm_fault+0x8d/0x240
[ 154.390073] do_user_addr_fault+0x21d/0x730
[ 154.390074] ? asm_exc_page_fault+0x8/0x30
[ 154.390074] exc_page_fault+0x6e/0x270
[ 154.390074] ? asm_exc_page_fault+0x8/0x30
[ 154.390074] asm_exc_page_fault+0x1e/0x30
[ 154.390075] RIP: 0033:0x555bc1610b55
[ 154.390075] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.390075] RSP: 002b:00007f0453e1fd60 EFLAGS: 00010202
[ 154.390076] RAX: 0000000000000000 RBX: 0000000000005480 RCX: 0000000000000001
[ 154.390076] RDX: 0000000000000000 RSI: 00000000000000a9 RDI: 00000000000000a9
[ 154.390076] RBP: 00000000000000a9 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390077] R10: 00007f0453e1fd60 R11: 000000000000ba1e R12: 0000000000000001
[ 154.390077] R13: 0000000000005480 R14: 00007f0453e1fd90 R15: 0000000000000000
[ 154.390078] NMI watchdog: Watchdog detected hard LOCKUP on cpu 47
[ 154.390078] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390092] irq event stamp: 47178
[ 154.390093] hardirqs last enabled at (47177): [<ffffffff81b05ab4>] _raw_spin_unlock_irq+0x24/0x40
[ 154.390093] hardirqs last disabled at (47178): [<ffffffff81afcf5a>] __schedule+0xd7a/0x14d0
[ 154.390093] softirqs last enabled at (47170): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390094] softirqs last disabled at (47165): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390094] CPU: 47 PID: 6975 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390094] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390095] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390095] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390096] RSP: 0000:ffffc90018457988 EFLAGS: 00000046
[ 154.390096] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000c00000
[ 154.390096] RDX: ffff88b009bff1c0 RSI: 0000000000000079 RDI: ffff88c0293fe380
[ 154.390097] RBP: ffff88c0293fe380 R08: 0000000000c00000 R09: 00000000001fd9c0
[ 154.390097] R10: ffffc90018457938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.390097] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.390097] FS: 00007f045485a700(0000) GS:ffff88b009a00000(0000) knlGS:0000000000000000
[ 154.390098] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390098] CR2: 0000555bc18acb48 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390098] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390099] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390099] PKRU: 55555554
[ 154.390099] Call Trace:
[ 154.390099] do_raw_spin_lock+0xb1/0xc0
[ 154.390100] _raw_spin_lock_nested+0x50/0x60
[ 154.390100] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390100] load_balance+0x2f7/0xd60
[ 154.390100] ? update_blocked_averages+0x500/0x820
[ 154.390100] newidle_balance+0x3d4/0x670
[ 154.390101] balance_fair+0x1b/0x30
[ 154.390101] put_prev_task_balance+0x4b/0x80
[ 154.390101] __schedule+0x57b/0x14d0
[ 154.390101] schedule+0x51/0xc0
[ 154.390102] io_schedule+0x12/0x40
[ 154.390102] __lock_page_killable+0x167/0x250
[ 154.390102] ? file_fdatawait_range+0x20/0x20
[ 154.390102] filemap_fault+0x9b5/0xfa0
[ 154.390102] ? lock_acquire+0x266/0x3d0
[ 154.390103] ? ext4_filemap_fault+0x27/0x40
[ 154.390103] ? ext4_filemap_fault+0x27/0x40
[ 154.390103] ext4_filemap_fault+0x2f/0x40
[ 154.390103] __do_fault+0x30/0xd0
[ 154.390104] __handle_mm_fault+0xf0d/0x12c0
[ 154.390104] handle_mm_fault+0x8d/0x240
[ 154.390104] do_user_addr_fault+0x21d/0x730
[ 154.390104] ? asm_exc_page_fault+0x8/0x30
[ 154.390105] exc_page_fault+0x6e/0x270
[ 154.390105] ? asm_exc_page_fault+0x8/0x30
[ 154.390105] asm_exc_page_fault+0x1e/0x30
[ 154.390105] RIP: 0033:0x555bc1610b55
[ 154.390106] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.390106] RSP: 002b:00007f0454859d60 EFLAGS: 00010202
[ 154.390106] RAX: 0000000000000000 RBX: 0000000000000780 RCX: 0000000000000001
[ 154.390107] RDX: 0000000000000000 RSI: 000000000000000f RDI: 000000000000000f
[ 154.390107] RBP: 000000000000000f R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390107] R10: 00007f0454859d60 R11: 000000000000ba16 R12: 0000000000000001
[ 154.390108] R13: 0000000000000780 R14: 00007f0454859d90 R15: 0000000000000000
[ 154.390108] NMI backtrace for cpu 47
[ 154.390108] CPU: 47 PID: 6975 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390109] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390109] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390110] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390110] RSP: 0000:ffffc90018457988 EFLAGS: 00000046
[ 154.390110] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000c00000
[ 154.390111] RDX: ffff88b009bff1c0 RSI: 0000000000000079 RDI: ffff88c0293fe380
[ 154.390111] RBP: ffff88c0293fe380 R08: 0000000000c00000 R09: 00000000001fd9c0
[ 154.390111] R10: ffffc90018457938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.390112] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.390112] FS: 00007f045485a700(0000) GS:ffff88b009a00000(0000) knlGS:0000000000000000
[ 154.390112] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390113] CR2: 0000555bc18acb48 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390113] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390113] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390114] PKRU: 55555554
[ 154.390114] Call Trace:
[ 154.390114] do_raw_spin_lock+0xb1/0xc0
[ 154.390114] _raw_spin_lock_nested+0x50/0x60
[ 154.390114] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390115] load_balance+0x2f7/0xd60
[ 154.390115] ? update_blocked_averages+0x500/0x820
[ 154.390115] newidle_balance+0x3d4/0x670
[ 154.390115] balance_fair+0x1b/0x30
[ 154.390116] put_prev_task_balance+0x4b/0x80
[ 154.390116] __schedule+0x57b/0x14d0
[ 154.390116] schedule+0x51/0xc0
[ 154.390116] io_schedule+0x12/0x40
[ 154.390116] __lock_page_killable+0x167/0x250
[ 154.390117] ? file_fdatawait_range+0x20/0x20
[ 154.390117] filemap_fault+0x9b5/0xfa0
[ 154.390117] ? lock_acquire+0x266/0x3d0
[ 154.390117] ? ext4_filemap_fault+0x27/0x40
[ 154.390118] ? ext4_filemap_fault+0x27/0x40
[ 154.390118] ext4_filemap_fault+0x2f/0x40
[ 154.390118] __do_fault+0x30/0xd0
[ 154.390118] __handle_mm_fault+0xf0d/0x12c0
[ 154.390119] handle_mm_fault+0x8d/0x240
[ 154.390119] do_user_addr_fault+0x21d/0x730
[ 154.390119] ? asm_exc_page_fault+0x8/0x30
[ 154.390119] exc_page_fault+0x6e/0x270
[ 154.390119] ? asm_exc_page_fault+0x8/0x30
[ 154.390120] asm_exc_page_fault+0x1e/0x30
[ 154.390120] RIP: 0033:0x555bc1610b55
[ 154.390120] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.390121] RSP: 002b:00007f0454859d60 EFLAGS: 00010202
[ 154.390121] RAX: 0000000000000000 RBX: 0000000000000780 RCX: 0000000000000001
[ 154.390122] RDX: 0000000000000000 RSI: 000000000000000f RDI: 000000000000000f
[ 154.390122] RBP: 000000000000000f R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390122] R10: 00007f0454859d60 R11: 000000000000ba16 R12: 0000000000000001
[ 154.390122] R13: 0000000000000780 R14: 00007f0454859d90 R15: 0000000000000000
[ 154.390123] NMI watchdog: Watchdog detected hard LOCKUP on cpu 48
[ 154.390123] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390138] irq event stamp: 24021
[ 154.390138] hardirqs last enabled at (24020): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390139] hardirqs last disabled at (24021): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390139] softirqs last enabled at (24010): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390139] softirqs last disabled at (24017): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390140] CPU: 48 PID: 7088 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390140] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390140] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390141] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390141] RSP: 0000:ffffc9001915fce8 EFLAGS: 00000046
[ 154.390142] RAX: 0000000000000000 RBX: ffff889804ffe380 RCX: 0000000000c40000
[ 154.390142] RDX: ffff88c023fff1c0 RSI: 0000000000000013 RDI: ffff889804ffe380
[ 154.390142] RBP: ffff889804ffe380 R08: 0000000000c40000 R09: 00000000001fd9c0
[ 154.390143] R10: ffffc9001915fc98 R11: 00000000e455304a R12: ffff889804ffe380
[ 154.390143] R13: 0000000000000400 R14: 0000000000000200 R15: ffff889804ffe380
[ 154.390143] FS: 00007f04541c7700(0000) GS:ffff88c023e00000(0000) knlGS:0000000000000000
[ 154.390143] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390144] CR2: 00005570f6d9affd CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.390144] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390144] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390145] PKRU: 55555554
[ 154.390145] Call Trace:
[ 154.390145] do_raw_spin_lock+0xb1/0xc0
[ 154.390145] _raw_spin_lock_nested+0x50/0x60
[ 154.390145] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390146] load_balance+0x2f7/0xd60
[ 154.390146] rebalance_domains+0x2f6/0x4d0
[ 154.390146] __do_softirq+0xd4/0x432
[ 154.390146] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390147] irq_exit_rcu+0xbc/0xc0
[ 154.390147] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390147] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390147] RIP: 0033:0x555bc1620f5a
[ 154.390148] Code: 00 00 00 66 0f ef db 66 0f 1f 84 00 00 00 00 00 48 83 c3 01 48 39 eb 74 67 66 0f ef c0 f2 48 0f 2a c3 66 0f 2e d8 f2 0f 51 c8 <77> 5d 66 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90
[ 154.390148] RSP: 002b:00007f04541c6d40 EFLAGS: 00000203
[ 154.390149] RAX: 00000000000056b9 RBX: 000000000001042d RCX: 0000000000000003
[ 154.390149] RDX: 0000000000000000 RSI: 0000000000000072 RDI: 00007f04541c6dc0
[ 154.390149] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390150] R10: 00007f04541c6d60 R11: 000000000000901e R12: 0000000000000001
[ 154.390150] R13: 0000000000003900 R14: 00007f04541c6d90 R15: 0000555bc39a5d80
[ 154.390150] NMI backtrace for cpu 48
[ 154.390151] CPU: 48 PID: 7088 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390151] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390151] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390152] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390152] RSP: 0000:ffffc9001915fce8 EFLAGS: 00000046
[ 154.390152] RAX: 0000000000000000 RBX: ffff889804ffe380 RCX: 0000000000c40000
[ 154.390153] RDX: ffff88c023fff1c0 RSI: 0000000000000013 RDI: ffff889804ffe380
[ 154.390153] RBP: ffff889804ffe380 R08: 0000000000c40000 R09: 00000000001fd9c0
[ 154.390153] R10: ffffc9001915fc98 R11: 00000000e455304a R12: ffff889804ffe380
[ 154.390154] R13: 0000000000000400 R14: 0000000000000200 R15: ffff889804ffe380
[ 154.390154] FS: 00007f04541c7700(0000) GS:ffff88c023e00000(0000) knlGS:0000000000000000
[ 154.390154] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390155] CR2: 00005570f6d9affd CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.390155] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390155] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390155] PKRU: 55555554
[ 154.390156] Call Trace:
[ 154.390156] do_raw_spin_lock+0xb1/0xc0
[ 154.390156] _raw_spin_lock_nested+0x50/0x60
[ 154.390156] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390157] load_balance+0x2f7/0xd60
[ 154.390157] rebalance_domains+0x2f6/0x4d0
[ 154.390157] __do_softirq+0xd4/0x432
[ 154.390157] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390157] irq_exit_rcu+0xbc/0xc0
[ 154.390158] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390158] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390158] RIP: 0033:0x555bc1620f5a
[ 154.390159] Code: 00 00 00 66 0f ef db 66 0f 1f 84 00 00 00 00 00 48 83 c3 01 48 39 eb 74 67 66 0f ef c0 f2 48 0f 2a c3 66 0f 2e d8 f2 0f 51 c8 <77> 5d 66 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90
[ 154.390159] RSP: 002b:00007f04541c6d40 EFLAGS: 00000203
[ 154.390159] RAX: 00000000000056b9 RBX: 000000000001042d RCX: 0000000000000003
[ 154.390160] RDX: 0000000000000000 RSI: 0000000000000072 RDI: 00007f04541c6dc0
[ 154.390160] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390160] R10: 00007f04541c6d60 R11: 000000000000901e R12: 0000000000000001
[ 154.390161] R13: 0000000000003900 R14: 00007f04541c6d90 R15: 0000555bc39a5d80
[ 154.390161] NMI watchdog: Watchdog detected hard LOCKUP on cpu 49
[ 154.390162] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390176] irq event stamp: 23705
[ 154.390177] hardirqs last enabled at (23704): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390177] hardirqs last disabled at (23705): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390177] softirqs last enabled at (23692): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390178] softirqs last disabled at (23701): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390178] CPU: 49 PID: 7083 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390178] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390179] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390179] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390180] RSP: 0000:ffffc9001912fce8 EFLAGS: 00000046
[ 154.390180] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000c80000
[ 154.390180] RDX: ffff88c0243ff1c0 RSI: 0000000000000042 RDI: ffff88c0293fe380
[ 154.390181] RBP: ffff88c0293fe380 R08: 0000000000c80000 R09: 00000000001fd9c0
[ 154.390181] R10: ffffc9001912fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390181] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390182] FS: 00007f045421c700(0000) GS:ffff88c024200000(0000) knlGS:0000000000000000
[ 154.390182] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390182] CR2: 00007f0456e1b160 CR3: 000000011a18a004 CR4: 00000000007706e0
[ 154.390182] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390183] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390183] PKRU: 55555554
[ 154.390183] Call Trace:
[ 154.390183] do_raw_spin_lock+0xb1/0xc0
[ 154.390184] _raw_spin_lock_nested+0x50/0x60
[ 154.390184] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390184] load_balance+0x2f7/0xd60
[ 154.390184] rebalance_domains+0x2f6/0x4d0
[ 154.390184] __do_softirq+0xd4/0x432
[ 154.390185] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390185] irq_exit_rcu+0xbc/0xc0
[ 154.390185] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390185] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390186] RIP: 0033:0x555bc1620f87
[ 154.390186] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390186] RSP: 002b:00007f045421bd40 EFLAGS: 00000246
[ 154.390187] RAX: 0000000000000479 RBX: 000000000000f199 RCX: 0000000000000036
[ 154.390187] RDX: 0000000000000013 RSI: 000000000000006d RDI: 00007f045421bdc0
[ 154.390188] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390188] R10: 00007f045421bd60 R11: 0000000000008cd4 R12: 0000000000000001
[ 154.390188] R13: 0000000000003680 R14: 00007f045421bd90 R15: 0000555bc39a5b00
[ 154.390188] NMI backtrace for cpu 49
[ 154.390189] CPU: 49 PID: 7083 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390189] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390189] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390190] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390190] RSP: 0000:ffffc9001912fce8 EFLAGS: 00000046
[ 154.390191] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000c80000
[ 154.390191] RDX: ffff88c0243ff1c0 RSI: 0000000000000042 RDI: ffff88c0293fe380
[ 154.390191] RBP: ffff88c0293fe380 R08: 0000000000c80000 R09: 00000000001fd9c0
[ 154.390192] R10: ffffc9001912fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390192] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390192] FS: 00007f045421c700(0000) GS:ffff88c024200000(0000) knlGS:0000000000000000
[ 154.390192] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390193] CR2: 00007f0456e1b160 CR3: 000000011a18a004 CR4: 00000000007706e0
[ 154.390193] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390193] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390194] PKRU: 55555554
[ 154.390194] Call Trace:
[ 154.390194] do_raw_spin_lock+0xb1/0xc0
[ 154.390194] _raw_spin_lock_nested+0x50/0x60
[ 154.390195] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390195] load_balance+0x2f7/0xd60
[ 154.390195] rebalance_domains+0x2f6/0x4d0
[ 154.390195] __do_softirq+0xd4/0x432
[ 154.390196] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390196] irq_exit_rcu+0xbc/0xc0
[ 154.390196] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390196] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390196] RIP: 0033:0x555bc1620f87
[ 154.390197] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390197] RSP: 002b:00007f045421bd40 EFLAGS: 00000246
[ 154.390198] RAX: 0000000000000479 RBX: 000000000000f199 RCX: 0000000000000036
[ 154.390198] RDX: 0000000000000013 RSI: 000000000000006d RDI: 00007f045421bdc0
[ 154.390198] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390199] R10: 00007f045421bd60 R11: 0000000000008cd4 R12: 0000000000000001
[ 154.390199] R13: 0000000000003680 R14: 00007f045421bd90 R15: 0000555bc39a5b00
[ 154.390200] NMI watchdog: Watchdog detected hard LOCKUP on cpu 50
[ 154.390200] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390214] irq event stamp: 24553
[ 154.390215] hardirqs last enabled at (24552): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390215] hardirqs last disabled at (24553): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390215] softirqs last enabled at (24530): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390216] softirqs last disabled at (24549): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390216] CPU: 50 PID: 7093 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390217] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390217] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390217] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390218] RSP: 0000:ffffc9001918bce8 EFLAGS: 00000046
[ 154.390218] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000cc0000
[ 154.390218] RDX: ffff88c0247ff1c0 RSI: 0000000000000036 RDI: ffff88c0293fe380
[ 154.390219] RBP: ffff88c0293fe380 R08: 0000000000cc0000 R09: 00000000001fd9c0
[ 154.390219] R10: ffffc9001918bc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390219] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390220] FS: 00007f0454194700(0000) GS:ffff88c024600000(0000) knlGS:0000000000000000
[ 154.390220] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390220] CR2: 00007f0456e61158 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390220] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390221] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390221] PKRU: 55555554
[ 154.390221] Call Trace:
[ 154.390221] do_raw_spin_lock+0xb1/0xc0
[ 154.390222] _raw_spin_lock_nested+0x50/0x60
[ 154.390222] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390222] load_balance+0x2f7/0xd60
[ 154.390222] rebalance_domains+0x2f6/0x4d0
[ 154.390223] __do_softirq+0xd4/0x432
[ 154.390223] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390223] irq_exit_rcu+0xbc/0xc0
[ 154.390223] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390224] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390224] RIP: 0033:0x555bc1620f5a
[ 154.390224] Code: 00 00 00 66 0f ef db 66 0f 1f 84 00 00 00 00 00 48 83 c3 01 48 39 eb 74 67 66 0f ef c0 f2 48 0f 2a c3 66 0f 2e d8 f2 0f 51 c8 <77> 5d 66 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90
[ 154.390225] RSP: 002b:00007f0454193d40 EFLAGS: 00000203
[ 154.390225] RAX: 0000000000000e95 RBX: 0000000000002bc0 RCX: 0000000000000003
[ 154.390225] RDX: 0000000000000000 RSI: 0000000000000075 RDI: 00007f0454193dc0
[ 154.390226] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390226] R10: 00007f0454193d60 R11: 0000000000008cda R12: 0000000000000001
[ 154.390226] R13: 0000000000003a80 R14: 00007f0454193d90 R15: 0000555bc39a5f00
[ 154.390227] NMI backtrace for cpu 50
[ 154.390227] CPU: 50 PID: 7093 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390227] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390228] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390228] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390228] RSP: 0000:ffffc9001918bce8 EFLAGS: 00000046
[ 154.390229] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000cc0000
[ 154.390229] RDX: ffff88c0247ff1c0 RSI: 0000000000000036 RDI: ffff88c0293fe380
[ 154.390229] RBP: ffff88c0293fe380 R08: 0000000000cc0000 R09: 00000000001fd9c0
[ 154.390230] R10: ffffc9001918bc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390230] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390230] FS: 00007f0454194700(0000) GS:ffff88c024600000(0000) knlGS:0000000000000000
[ 154.390231] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390231] CR2: 00007f0456e61158 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390231] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390231] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390232] PKRU: 55555554
[ 154.390232] Call Trace:
[ 154.390232] do_raw_spin_lock+0xb1/0xc0
[ 154.390232] _raw_spin_lock_nested+0x50/0x60
[ 154.390233] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390233] load_balance+0x2f7/0xd60
[ 154.390233] rebalance_domains+0x2f6/0x4d0
[ 154.390233] __do_softirq+0xd4/0x432
[ 154.390233] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390234] irq_exit_rcu+0xbc/0xc0
[ 154.390234] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390234] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390234] RIP: 0033:0x555bc1620f5a
[ 154.390235] Code: 00 00 00 66 0f ef db 66 0f 1f 84 00 00 00 00 00 48 83 c3 01 48 39 eb 74 67 66 0f ef c0 f2 48 0f 2a c3 66 0f 2e d8 f2 0f 51 c8 <77> 5d 66 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90
[ 154.390235] RSP: 002b:00007f0454193d40 EFLAGS: 00000203
[ 154.390236] RAX: 0000000000000e95 RBX: 0000000000002bc0 RCX: 0000000000000003
[ 154.390236] RDX: 0000000000000000 RSI: 0000000000000075 RDI: 00007f0454193dc0
[ 154.390236] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390237] R10: 00007f0454193d60 R11: 0000000000008cda R12: 0000000000000001
[ 154.390237] R13: 0000000000003a80 R14: 00007f0454193d90 R15: 0000555bc39a5f00
[ 154.390238] NMI watchdog: Watchdog detected hard LOCKUP on cpu 51
[ 154.390238] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390252] irq event stamp: 23165
[ 154.390253] hardirqs last enabled at (23164): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390253] hardirqs last disabled at (23165): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390253] softirqs last enabled at (23144): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390254] softirqs last disabled at (23161): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390254] CPU: 51 PID: 7094 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390254] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390255] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390255] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390256] RSP: 0000:ffffc90019193ce8 EFLAGS: 00000046
[ 154.390256] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000d00000
[ 154.390256] RDX: ffff88c024bff1c0 RSI: 0000000000000031 RDI: ffff88c0293fe380
[ 154.390257] RBP: ffff88c0293fe380 R08: 0000000000d00000 R09: 00000000001fd9c0
[ 154.390257] R10: ffffc90019193c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390257] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390258] FS: 00007f0454183700(0000) GS:ffff88c024a00000(0000) knlGS:0000000000000000
[ 154.390258] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390258] CR2: 00007f0456e95160 CR3: 000000011a18a005 CR4: 00000000007706e0
[ 154.390258] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390259] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390259] PKRU: 55555554
[ 154.390259] Call Trace:
[ 154.390259] do_raw_spin_lock+0xb1/0xc0
[ 154.390260] _raw_spin_lock_nested+0x50/0x60
[ 154.390260] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390260] load_balance+0x2f7/0xd60
[ 154.390260] rebalance_domains+0x2f6/0x4d0
[ 154.390260] __do_softirq+0xd4/0x432
[ 154.390261] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390261] irq_exit_rcu+0xbc/0xc0
[ 154.390261] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390261] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390262] RIP: 0033:0x555bc1620f87
[ 154.390262] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390262] RSP: 002b:00007f0454182d40 EFLAGS: 00000246
[ 154.390263] RAX: 0000000000000181 RBX: 0000000000010dbd RCX: 00000000000000b3
[ 154.390263] RDX: 000000000000008a RSI: 0000000000000076 RDI: 00007f0454182dc0
[ 154.390264] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390264] R10: 00007f0454182d60 R11: 0000000000008cd4 R12: 0000000000000001
[ 154.390264] R13: 0000000000003b00 R14: 00007f0454182d90 R15: 0000555bc39a5f80
[ 154.390265] NMI backtrace for cpu 51
[ 154.390265] CPU: 51 PID: 7094 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390265] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390265] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390266] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390266] RSP: 0000:ffffc90019193ce8 EFLAGS: 00000046
[ 154.390267] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000d00000
[ 154.390267] RDX: ffff88c024bff1c0 RSI: 0000000000000031 RDI: ffff88c0293fe380
[ 154.390267] RBP: ffff88c0293fe380 R08: 0000000000d00000 R09: 00000000001fd9c0
[ 154.390268] R10: ffffc90019193c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390268] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390268] FS: 00007f0454183700(0000) GS:ffff88c024a00000(0000) knlGS:0000000000000000
[ 154.390269] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390269] CR2: 00007f0456e95160 CR3: 000000011a18a005 CR4: 00000000007706e0
[ 154.390269] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390269] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390270] PKRU: 55555554
[ 154.390270] Call Trace:
[ 154.390270] do_raw_spin_lock+0xb1/0xc0
[ 154.390270] _raw_spin_lock_nested+0x50/0x60
[ 154.390271] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390271] load_balance+0x2f7/0xd60
[ 154.390271] rebalance_domains+0x2f6/0x4d0
[ 154.390271] __do_softirq+0xd4/0x432
[ 154.390271] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390272] irq_exit_rcu+0xbc/0xc0
[ 154.390272] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390272] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390272] RIP: 0033:0x555bc1620f87
[ 154.390273] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390273] RSP: 002b:00007f0454182d40 EFLAGS: 00000246
[ 154.390274] RAX: 0000000000000181 RBX: 0000000000010dbd RCX: 00000000000000b3
[ 154.390274] RDX: 000000000000008a RSI: 0000000000000076 RDI: 00007f0454182dc0
[ 154.390274] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390275] R10: 00007f0454182d60 R11: 0000000000008cd4 R12: 0000000000000001
[ 154.390275] R13: 0000000000003b00 R14: 00007f0454182d90 R15: 0000555bc39a5f80
[ 154.390276] NMI watchdog: Watchdog detected hard LOCKUP on cpu 52
[ 154.390276] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390290] irq event stamp: 15997
[ 154.390291] hardirqs last enabled at (15996): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390291] hardirqs last disabled at (15997): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390291] softirqs last enabled at (15986): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390292] softirqs last disabled at (15993): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390292] CPU: 52 PID: 7090 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390293] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390293] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390293] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390294] RSP: 0000:ffffc9001917bce8 EFLAGS: 00000046
[ 154.390294] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000d40000
[ 154.390295] RDX: ffff88c024fff1c0 RSI: 0000000000000047 RDI: ffff88c0293fe380
[ 154.390295] RBP: ffff88c0293fe380 R08: 0000000000d40000 R09: 00000000001fd9c0
[ 154.390295] R10: ffffc9001917bc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390295] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390296] FS: 00007f04541b6700(0000) GS:ffff88c024e00000(0000) knlGS:0000000000000000
[ 154.390296] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390296] CR2: 00007f0456e7b158 CR3: 000000011a18a001 CR4: 00000000007706e0
[ 154.390297] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390297] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390297] PKRU: 55555554
[ 154.390297] Call Trace:
[ 154.390298] do_raw_spin_lock+0xb1/0xc0
[ 154.390298] _raw_spin_lock_nested+0x50/0x60
[ 154.390298] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390298] load_balance+0x2f7/0xd60
[ 154.390299] rebalance_domains+0x2f6/0x4d0
[ 154.390299] __do_softirq+0xd4/0x432
[ 154.390299] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390299] irq_exit_rcu+0xbc/0xc0
[ 154.390299] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390300] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390300] RIP: 0033:0x555bc1620f87
[ 154.390300] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390301] RSP: 002b:00007f04541b5d40 EFLAGS: 00000246
[ 154.390301] RAX: 0000000000002938 RBX: 000000000000ce1b RCX: 0000000000000005
[ 154.390301] RDX: 0000000000000003 RSI: 0000000000000073 RDI: 00007f04541b5dc0
[ 154.390302] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390302] R10: 00007f04541b5d60 R11: 0000000000008cd2 R12: 0000000000000001
[ 154.390302] R13: 0000000000003980 R14: 00007f04541b5d90 R15: 0000555bc39a5e00
[ 154.390303] NMI backtrace for cpu 52
[ 154.390303] CPU: 52 PID: 7090 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390303] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390304] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390304] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390304] RSP: 0000:ffffc9001917bce8 EFLAGS: 00000046
[ 154.390305] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000d40000
[ 154.390305] RDX: ffff88c024fff1c0 RSI: 0000000000000047 RDI: ffff88c0293fe380
[ 154.390305] RBP: ffff88c0293fe380 R08: 0000000000d40000 R09: 00000000001fd9c0
[ 154.390306] R10: ffffc9001917bc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390306] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390306] FS: 00007f04541b6700(0000) GS:ffff88c024e00000(0000) knlGS:0000000000000000
[ 154.390307] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390307] CR2: 00007f0456e7b158 CR3: 000000011a18a001 CR4: 00000000007706e0
[ 154.390307] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390308] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390308] PKRU: 55555554
[ 154.390308] Call Trace:
[ 154.390308] do_raw_spin_lock+0xb1/0xc0
[ 154.390309] _raw_spin_lock_nested+0x50/0x60
[ 154.390309] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390309] load_balance+0x2f7/0xd60
[ 154.390309] rebalance_domains+0x2f6/0x4d0
[ 154.390309] __do_softirq+0xd4/0x432
[ 154.390310] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390310] irq_exit_rcu+0xbc/0xc0
[ 154.390310] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390310] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390311] RIP: 0033:0x555bc1620f87
[ 154.390311] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390311] RSP: 002b:00007f04541b5d40 EFLAGS: 00000246
[ 154.390312] RAX: 0000000000002938 RBX: 000000000000ce1b RCX: 0000000000000005
[ 154.390312] RDX: 0000000000000003 RSI: 0000000000000073 RDI: 00007f04541b5dc0
[ 154.390313] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390313] R10: 00007f04541b5d60 R11: 0000000000008cd2 R12: 0000000000000001
[ 154.390313] R13: 0000000000003980 R14: 00007f04541b5d90 R15: 0000555bc39a5e00
[ 154.390314] NMI watchdog: Watchdog detected hard LOCKUP on cpu 53
[ 154.390314] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390329] irq event stamp: 16503
[ 154.390329] hardirqs last enabled at (16502): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390329] hardirqs last disabled at (16503): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390330] softirqs last enabled at (16486): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390330] softirqs last disabled at (16499): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390330] CPU: 53 PID: 7138 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390331] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390331] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390331] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390332] RSP: 0000:ffffc90019407ce8 EFLAGS: 00000046
[ 154.390332] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000d80000
[ 154.390333] RDX: ffff88c0253ff1c0 RSI: 000000000000003e RDI: ffff88c0293fe380
[ 154.390333] RBP: ffff88c0293fe380 R08: 0000000000d80000 R09: 00000000001fd9c0
[ 154.390333] R10: ffffc90019407c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390334] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390334] FS: 00007f0453f30700(0000) GS:ffff88c025200000(0000) knlGS:0000000000000000
[ 154.390334] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390334] CR2: 00007f0456e41158 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.390335] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390335] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390335] PKRU: 55555554
[ 154.390335] Call Trace:
[ 154.390336] do_raw_spin_lock+0xb1/0xc0
[ 154.390336] _raw_spin_lock_nested+0x50/0x60
[ 154.390336] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390336] load_balance+0x2f7/0xd60
[ 154.390337] rebalance_domains+0x2f6/0x4d0
[ 154.390337] __do_softirq+0xd4/0x432
[ 154.390337] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390337] irq_exit_rcu+0xbc/0xc0
[ 154.390337] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390338] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390338] RIP: 0033:0x555bc1620f87
[ 154.390338] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390339] RSP: 002b:00007f0453f2fd40 EFLAGS: 00000246
[ 154.390339] RAX: 00000000000011ea RBX: 000000000000b327 RCX: 000000000000000a
[ 154.390339] RDX: 0000000000000003 RSI: 0000000000000099 RDI: 00007f0453f2fdc0
[ 154.390340] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390340] R10: 00007f0453f2fd60 R11: 0000000000008cd6 R12: 0000000000000001
[ 154.390340] R13: 0000000000004c80 R14: 00007f0453f2fd90 R15: 0000555bc39a7100
[ 154.390341] NMI backtrace for cpu 53
[ 154.390341] CPU: 53 PID: 7138 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390341] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390342] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390342] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390342] RSP: 0000:ffffc90019407ce8 EFLAGS: 00000046
[ 154.390343] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000d80000
[ 154.390343] RDX: ffff88c0253ff1c0 RSI: 000000000000003e RDI: ffff88c0293fe380
[ 154.390343] RBP: ffff88c0293fe380 R08: 0000000000d80000 R09: 00000000001fd9c0
[ 154.390344] R10: ffffc90019407c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390344] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390344] FS: 00007f0453f30700(0000) GS:ffff88c025200000(0000) knlGS:0000000000000000
[ 154.390345] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390345] CR2: 00007f0456e41158 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.390345] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390346] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390346] PKRU: 55555554
[ 154.390346] Call Trace:
[ 154.390346] do_raw_spin_lock+0xb1/0xc0
[ 154.390346] _raw_spin_lock_nested+0x50/0x60
[ 154.390347] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390347] load_balance+0x2f7/0xd60
[ 154.390347] rebalance_domains+0x2f6/0x4d0
[ 154.390347] __do_softirq+0xd4/0x432
[ 154.390348] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390348] irq_exit_rcu+0xbc/0xc0
[ 154.390348] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390348] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390349] RIP: 0033:0x555bc1620f87
[ 154.390349] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390350] RSP: 002b:00007f0453f2fd40 EFLAGS: 00000246
[ 154.390350] RAX: 00000000000011ea RBX: 000000000000b327 RCX: 000000000000000a
[ 154.390350] RDX: 0000000000000003 RSI: 0000000000000099 RDI: 00007f0453f2fdc0
[ 154.390351] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390351] R10: 00007f0453f2fd60 R11: 0000000000008cd6 R12: 0000000000000001
[ 154.390351] R13: 0000000000004c80 R14: 00007f0453f2fd90 R15: 0000555bc39a7100
[ 154.390352] NMI watchdog: Watchdog detected hard LOCKUP on cpu 54
[ 154.390352] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390367] irq event stamp: 20909
[ 154.390367] hardirqs last enabled at (20908): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390367] hardirqs last disabled at (20909): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390368] softirqs last enabled at (20886): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390368] softirqs last disabled at (20905): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390369] CPU: 54 PID: 7096 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390369] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390370] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390370] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390370] RSP: 0000:ffffc900191cfce8 EFLAGS: 00000046
[ 154.390371] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000dc0000
[ 154.390371] RDX: ffff88c0257ff1c0 RSI: 0000000000000033 RDI: ffff88c0293fe380
[ 154.390371] RBP: ffff88c0293fe380 R08: 0000000000dc0000 R09: 00000000001fd9c0
[ 154.390372] R10: ffffc900191cfc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390372] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390372] FS: 00007f0454172700(0000) GS:ffff88c025600000(0000) knlGS:0000000000000000
[ 154.390373] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390373] CR2: 00007f0456eb1160 CR3: 000000011a18a006 CR4: 00000000007706e0
[ 154.390373] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390373] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390374] PKRU: 55555554
[ 154.390374] Call Trace:
[ 154.390374] do_raw_spin_lock+0xb1/0xc0
[ 154.390374] _raw_spin_lock_nested+0x50/0x60
[ 154.390375] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390375] load_balance+0x2f7/0xd60
[ 154.390375] rebalance_domains+0x2f6/0x4d0
[ 154.390375] __do_softirq+0xd4/0x432
[ 154.390375] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390376] irq_exit_rcu+0xbc/0xc0
[ 154.390376] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390376] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390376] RIP: 0033:0x555bc1620f74
[ 154.390377] Code: 0f ef c0 f2 48 0f 2a c3 66 0f 2e d8 f2 0f 51 c8 77 5d 66 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 <f2> 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 48 85 d2
[ 154.390377] RSP: 002b:00007f0454171d40 EFLAGS: 00000202
[ 154.390378] RAX: 0000000000000155 RBX: 0000000000012dc1 RCX: 00000000000000e3
[ 154.390378] RDX: 00000000000000b7 RSI: 0000000000000077 RDI: 00007f0454171dc0
[ 154.390378] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390379] R10: 00007f0454171d60 R11: 0000000000008cd4 R12: 0000000000000001
[ 154.390379] R13: 0000000000003b80 R14: 00007f0454171d90 R15: 0000555bc39a6000
[ 154.390379] NMI backtrace for cpu 54
[ 154.390380] CPU: 54 PID: 7096 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390380] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390380] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390381] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390381] RSP: 0000:ffffc900191cfce8 EFLAGS: 00000046
[ 154.390381] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000dc0000
[ 154.390382] RDX: ffff88c0257ff1c0 RSI: 0000000000000033 RDI: ffff88c0293fe380
[ 154.390382] RBP: ffff88c0293fe380 R08: 0000000000dc0000 R09: 00000000001fd9c0
[ 154.390382] R10: ffffc900191cfc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390383] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390383] FS: 00007f0454172700(0000) GS:ffff88c025600000(0000) knlGS:0000000000000000
[ 154.390383] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390384] CR2: 00007f0456eb1160 CR3: 000000011a18a006 CR4: 00000000007706e0
[ 154.390384] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390384] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390384] PKRU: 55555554
[ 154.390385] Call Trace:
[ 154.390385] do_raw_spin_lock+0xb1/0xc0
[ 154.390385] _raw_spin_lock_nested+0x50/0x60
[ 154.390385] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390386] load_balance+0x2f7/0xd60
[ 154.390386] rebalance_domains+0x2f6/0x4d0
[ 154.390386] __do_softirq+0xd4/0x432
[ 154.390386] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390386] irq_exit_rcu+0xbc/0xc0
[ 154.390387] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390387] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390387] RIP: 0033:0x555bc1620f74
[ 154.390388] Code: 0f ef c0 f2 48 0f 2a c3 66 0f 2e d8 f2 0f 51 c8 77 5d 66 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 <f2> 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 48 85 d2
[ 154.390388] RSP: 002b:00007f0454171d40 EFLAGS: 00000202
[ 154.390388] RAX: 0000000000000155 RBX: 0000000000012dc1 RCX: 00000000000000e3
[ 154.390389] RDX: 00000000000000b7 RSI: 0000000000000077 RDI: 00007f0454171dc0
[ 154.390389] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390389] R10: 00007f0454171d60 R11: 0000000000008cd4 R12: 0000000000000001
[ 154.390390] R13: 0000000000003b80 R14: 00007f0454171d90 R15: 0000555bc39a6000
[ 154.390390] NMI watchdog: Watchdog detected hard LOCKUP on cpu 55
[ 154.390391] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390405] irq event stamp: 16065
[ 154.390405] hardirqs last enabled at (16064): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390406] hardirqs last disabled at (16065): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390406] softirqs last enabled at (16054): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390407] softirqs last disabled at (16061): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390407] CPU: 55 PID: 7106 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390407] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390408] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390408] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390409] RSP: 0000:ffffc9001926fce8 EFLAGS: 00000046
[ 154.390409] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000e00000
[ 154.390409] RDX: ffff88c025bff1c0 RSI: 0000000000000038 RDI: ffff88c0293fe380
[ 154.390410] RBP: ffff88c0293fe380 R08: 0000000000e00000 R09: 00000000001fd9c0
[ 154.390410] R10: ffffc9001926fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390410] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390410] FS: 00007f04540ea700(0000) GS:ffff88c025a00000(0000) knlGS:0000000000000000
[ 154.390411] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390411] CR2: 00007f04540ea730 CR3: 000000011a18a006 CR4: 00000000007706e0
[ 154.390411] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390412] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390412] PKRU: 55555554
[ 154.390412] Call Trace:
[ 154.390412] do_raw_spin_lock+0xb1/0xc0
[ 154.390412] _raw_spin_lock_nested+0x50/0x60
[ 154.390413] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390413] load_balance+0x2f7/0xd60
[ 154.390413] rebalance_domains+0x2f6/0x4d0
[ 154.390413] __do_softirq+0xd4/0x432
[ 154.390414] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390414] irq_exit_rcu+0xbc/0xc0
[ 154.390414] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390414] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390415] RIP: 0033:0x555bc1620f87
[ 154.390415] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390415] RSP: 002b:00007f04540e9d40 EFLAGS: 00000246
[ 154.390416] RAX: 0000000000000343 RBX: 000000000000d0eb RCX: 0000000000000040
[ 154.390416] RDX: 000000000000002b RSI: 000000000000007f RDI: 00007f04540e9dc0
[ 154.390416] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390417] R10: 00007f04540e9d60 R11: 0000000000008cd2 R12: 0000000000000001
[ 154.390417] R13: 0000000000003f80 R14: 00007f04540e9d90 R15: 0000555bc39a6400
[ 154.390417] NMI backtrace for cpu 55
[ 154.390418] CPU: 55 PID: 7106 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390418] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390418] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390419] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390419] RSP: 0000:ffffc9001926fce8 EFLAGS: 00000046
[ 154.390419] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000e00000
[ 154.390420] RDX: ffff88c025bff1c0 RSI: 0000000000000038 RDI: ffff88c0293fe380
[ 154.390420] RBP: ffff88c0293fe380 R08: 0000000000e00000 R09: 00000000001fd9c0
[ 154.390420] R10: ffffc9001926fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390421] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390421] FS: 00007f04540ea700(0000) GS:ffff88c025a00000(0000) knlGS:0000000000000000
[ 154.390421] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390422] CR2: 00007f04540ea730 CR3: 000000011a18a006 CR4: 00000000007706e0
[ 154.390422] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390422] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390423] PKRU: 55555554
[ 154.390423] Call Trace:
[ 154.390423] do_raw_spin_lock+0xb1/0xc0
[ 154.390423] _raw_spin_lock_nested+0x50/0x60
[ 154.390423] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390424] load_balance+0x2f7/0xd60
[ 154.390424] rebalance_domains+0x2f6/0x4d0
[ 154.390424] __do_softirq+0xd4/0x432
[ 154.390424] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390425] irq_exit_rcu+0xbc/0xc0
[ 154.390425] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390425] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390425] RIP: 0033:0x555bc1620f87
[ 154.390426] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390426] RSP: 002b:00007f04540e9d40 EFLAGS: 00000246
[ 154.390427] RAX: 0000000000000343 RBX: 000000000000d0eb RCX: 0000000000000040
[ 154.390427] RDX: 000000000000002b RSI: 000000000000007f RDI: 00007f04540e9dc0
[ 154.390427] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390428] R10: 00007f04540e9d60 R11: 0000000000008cd2 R12: 0000000000000001
[ 154.390428] R13: 0000000000003f80 R14: 00007f04540e9d90 R15: 0000555bc39a6400
[ 154.390428] NMI watchdog: Watchdog detected hard LOCKUP on cpu 56
[ 154.390429] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390443] irq event stamp: 16523
[ 154.390444] hardirqs last enabled at (16522): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390444] hardirqs last disabled at (16523): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390444] softirqs last enabled at (16494): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390445] softirqs last disabled at (16519): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390445] CPU: 56 PID: 7118 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390445] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390446] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390446] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390447] RSP: 0000:ffffc90019307ce8 EFLAGS: 00000046
[ 154.390447] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000e40000
[ 154.390447] RDX: ffff88c025fff1c0 RSI: 0000000000000039 RDI: ffff88c0293fe380
[ 154.390448] RBP: ffff88c0293fe380 R08: 0000000000e40000 R09: 00000000001fd9c0
[ 154.390448] R10: ffffc90019307c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390448] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390449] FS: 00007f0454040700(0000) GS:ffff88c025e00000(0000) knlGS:0000000000000000
[ 154.390449] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390449] CR2: 00007f0456e6b158 CR3: 000000011a18a006 CR4: 00000000007706e0
[ 154.390450] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390450] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390450] PKRU: 55555554
[ 154.390450] Call Trace:
[ 154.390450] do_raw_spin_lock+0xb1/0xc0
[ 154.390451] _raw_spin_lock_nested+0x50/0x60
[ 154.390451] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390451] load_balance+0x2f7/0xd60
[ 154.390451] rebalance_domains+0x2f6/0x4d0
[ 154.390452] __do_softirq+0xd4/0x432
[ 154.390452] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390452] irq_exit_rcu+0xbc/0xc0
[ 154.390452] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390453] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390453] RIP: 0033:0x555bc1620f87
[ 154.390453] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390454] RSP: 002b:00007f045403fd40 EFLAGS: 00000246
[ 154.390454] RAX: 000000000000028f RBX: 0000000000005c29 RCX: 0000000000000024
[ 154.390454] RDX: 000000000000000d RSI: 0000000000000089 RDI: 00007f045403fdc0
[ 154.390455] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390455] R10: 00007f045403fd60 R11: 0000000000008cd6 R12: 0000000000000001
[ 154.390455] R13: 0000000000004480 R14: 00007f045403fd90 R15: 0000555bc39a6900
[ 154.390456] NMI backtrace for cpu 56
[ 154.390456] CPU: 56 PID: 7118 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390456] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390456] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390457] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390457] RSP: 0000:ffffc90019307ce8 EFLAGS: 00000046
[ 154.390458] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000e40000
[ 154.390458] RDX: ffff88c025fff1c0 RSI: 0000000000000039 RDI: ffff88c0293fe380
[ 154.390458] RBP: ffff88c0293fe380 R08: 0000000000e40000 R09: 00000000001fd9c0
[ 154.390459] R10: ffffc90019307c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390459] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390459] FS: 00007f0454040700(0000) GS:ffff88c025e00000(0000) knlGS:0000000000000000
[ 154.390460] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390460] CR2: 00007f0456e6b158 CR3: 000000011a18a006 CR4: 00000000007706e0
[ 154.390460] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390460] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390461] PKRU: 55555554
[ 154.390461] Call Trace:
[ 154.390461] do_raw_spin_lock+0xb1/0xc0
[ 154.390461] _raw_spin_lock_nested+0x50/0x60
[ 154.390462] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390462] load_balance+0x2f7/0xd60
[ 154.390462] rebalance_domains+0x2f6/0x4d0
[ 154.390462] __do_softirq+0xd4/0x432
[ 154.390462] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390463] irq_exit_rcu+0xbc/0xc0
[ 154.390463] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390463] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390464] RIP: 0033:0x555bc1620f87
[ 154.390464] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390464] RSP: 002b:00007f045403fd40 EFLAGS: 00000246
[ 154.390465] RAX: 000000000000028f RBX: 0000000000005c29 RCX: 0000000000000024
[ 154.390465] RDX: 000000000000000d RSI: 0000000000000089 RDI: 00007f045403fdc0
[ 154.390465] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390466] R10: 00007f045403fd60 R11: 0000000000008cd6 R12: 0000000000000001
[ 154.390466] R13: 0000000000004480 R14: 00007f045403fd90 R15: 0000555bc39a6900
[ 154.390467] NMI watchdog: Watchdog detected hard LOCKUP on cpu 57
[ 154.390467] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390481] irq event stamp: 16049
[ 154.390482] hardirqs last enabled at (16048): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390482] hardirqs last disabled at (16049): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390482] softirqs last enabled at (16034): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390483] softirqs last disabled at (16045): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390483] CPU: 57 PID: 7099 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390484] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390484] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390484] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390485] RSP: 0000:ffffc900191ffce8 EFLAGS: 00000046
[ 154.390485] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000e80000
[ 154.390485] RDX: ffff88c0263ff1c0 RSI: 00000000000000a5 RDI: ffff88c0293fe380
[ 154.390486] RBP: ffff88c0293fe380 R08: 0000000000e80000 R09: 00000000001fd9c0
[ 154.390486] R10: ffffc900191ffc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390486] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390487] FS: 00007f045413f700(0000) GS:ffff88c026200000(0000) knlGS:0000000000000000
[ 154.390487] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390487] CR2: 00007f045413f730 CR3: 000000011a18a005 CR4: 00000000007706e0
[ 154.390487] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390488] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390488] PKRU: 55555554
[ 154.390488] Call Trace:
[ 154.390488] do_raw_spin_lock+0xb1/0xc0
[ 154.390489] _raw_spin_lock_nested+0x50/0x60
[ 154.390489] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390489] load_balance+0x2f7/0xd60
[ 154.390489] rebalance_domains+0x2f6/0x4d0
[ 154.390490] __do_softirq+0xd4/0x432
[ 154.390490] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390490] irq_exit_rcu+0xbc/0xc0
[ 154.390490] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390491] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390491] RIP: 0033:0x555bc1620f87
[ 154.390491] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390492] RSP: 002b:00007f045413ed40 EFLAGS: 00000246
[ 154.390492] RAX: 0000000000000382 RBX: 000000000000f91d RCX: 0000000000000047
[ 154.390492] RDX: 000000000000000f RSI: 000000000000007a RDI: 00007f045413edc0
[ 154.390493] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390493] R10: 00007f045413ed60 R11: 0000000000008cd2 R12: 0000000000000001
[ 154.390493] R13: 0000000000003d00 R14: 00007f045413ed90 R15: 0000555bc39a6180
[ 154.390493] NMI backtrace for cpu 57
[ 154.390494] CPU: 57 PID: 7099 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390494] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390495] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390495] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390495] RSP: 0000:ffffc900191ffce8 EFLAGS: 00000046
[ 154.390496] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000e80000
[ 154.390496] RDX: ffff88c0263ff1c0 RSI: 00000000000000a5 RDI: ffff88c0293fe380
[ 154.390496] RBP: ffff88c0293fe380 R08: 0000000000e80000 R09: 00000000001fd9c0
[ 154.390497] R10: ffffc900191ffc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390497] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390497] FS: 00007f045413f700(0000) GS:ffff88c026200000(0000) knlGS:0000000000000000
[ 154.390498] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390498] CR2: 00007f045413f730 CR3: 000000011a18a005 CR4: 00000000007706e0
[ 154.390498] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390499] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390499] PKRU: 55555554
[ 154.390499] Call Trace:
[ 154.390499] do_raw_spin_lock+0xb1/0xc0
[ 154.390499] _raw_spin_lock_nested+0x50/0x60
[ 154.390500] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390500] load_balance+0x2f7/0xd60
[ 154.390500] rebalance_domains+0x2f6/0x4d0
[ 154.390500] __do_softirq+0xd4/0x432
[ 154.390501] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390501] irq_exit_rcu+0xbc/0xc0
[ 154.390501] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390501] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390502] RIP: 0033:0x555bc1620f87
[ 154.390502] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390502] RSP: 002b:00007f045413ed40 EFLAGS: 00000246
[ 154.390503] RAX: 0000000000000382 RBX: 000000000000f91d RCX: 0000000000000047
[ 154.390503] RDX: 000000000000000f RSI: 000000000000007a RDI: 00007f045413edc0
[ 154.390503] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390504] R10: 00007f045413ed60 R11: 0000000000008cd2 R12: 0000000000000001
[ 154.390504] R13: 0000000000003d00 R14: 00007f045413ed90 R15: 0000555bc39a6180
[ 154.390505] NMI watchdog: Watchdog detected hard LOCKUP on cpu 58
[ 154.390505] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390520] irq event stamp: 21005
[ 154.390520] hardirqs last enabled at (21004): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390520] hardirqs last disabled at (21005): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390521] softirqs last enabled at (20978): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390521] softirqs last disabled at (21001): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390521] CPU: 58 PID: 7098 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390522] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390522] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390522] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390523] RSP: 0000:ffffc900191f7ce8 EFLAGS: 00000046
[ 154.390523] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000ec0000
[ 154.390524] RDX: ffff88c0267ff1c0 RSI: 0000000000000032 RDI: ffff88c0293fe380
[ 154.390524] RBP: ffff88c0293fe380 R08: 0000000000ec0000 R09: 00000000001fd9c0
[ 154.390524] R10: ffffc900191f7c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390525] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390525] FS: 00007f0454150700(0000) GS:ffff88c026600000(0000) knlGS:0000000000000000
[ 154.390525] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390525] CR2: 00007f0456e19158 CR3: 000000011a18a001 CR4: 00000000007706e0
[ 154.390526] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390526] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390526] PKRU: 55555554
[ 154.390526] Call Trace:
[ 154.390527] do_raw_spin_lock+0xb1/0xc0
[ 154.390527] _raw_spin_lock_nested+0x50/0x60
[ 154.390527] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390527] load_balance+0x2f7/0xd60
[ 154.390528] rebalance_domains+0x2f6/0x4d0
[ 154.390528] __do_softirq+0xd4/0x432
[ 154.390528] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390528] irq_exit_rcu+0xbc/0xc0
[ 154.390529] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390529] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390529] RIP: 0033:0x555bc1620f87
[ 154.390529] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390530] RSP: 002b:00007f045414fd40 EFLAGS: 00000246
[ 154.390530] RAX: 0000000000000087 RBX: 0000000000002d8b RCX: 0000000000000056
[ 154.390531] RDX: 0000000000000031 RSI: 0000000000000079 RDI: 00007f045414fdc0
[ 154.390531] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390531] R10: 00007f045414fd60 R11: 0000000000008cda R12: 0000000000000001
[ 154.390531] R13: 0000000000003c80 R14: 00007f045414fd90 R15: 0000555bc39a6100
[ 154.390532] NMI backtrace for cpu 58
[ 154.390532] CPU: 58 PID: 7098 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390532] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390533] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390533] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390534] RSP: 0000:ffffc900191f7ce8 EFLAGS: 00000046
[ 154.390534] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000ec0000
[ 154.390534] RDX: ffff88c0267ff1c0 RSI: 0000000000000032 RDI: ffff88c0293fe380
[ 154.390535] RBP: ffff88c0293fe380 R08: 0000000000ec0000 R09: 00000000001fd9c0
[ 154.390535] R10: ffffc900191f7c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390535] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390536] FS: 00007f0454150700(0000) GS:ffff88c026600000(0000) knlGS:0000000000000000
[ 154.390536] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390536] CR2: 00007f0456e19158 CR3: 000000011a18a001 CR4: 00000000007706e0
[ 154.390537] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390537] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390537] PKRU: 55555554
[ 154.390537] Call Trace:
[ 154.390537] do_raw_spin_lock+0xb1/0xc0
[ 154.390538] _raw_spin_lock_nested+0x50/0x60
[ 154.390538] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390538] load_balance+0x2f7/0xd60
[ 154.390538] rebalance_domains+0x2f6/0x4d0
[ 154.390539] __do_softirq+0xd4/0x432
[ 154.390539] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390539] irq_exit_rcu+0xbc/0xc0
[ 154.390539] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390540] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390540] RIP: 0033:0x555bc1620f87
[ 154.390540] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390541] RSP: 002b:00007f045414fd40 EFLAGS: 00000246
[ 154.390541] RAX: 0000000000000087 RBX: 0000000000002d8b RCX: 0000000000000056
[ 154.390541] RDX: 0000000000000031 RSI: 0000000000000079 RDI: 00007f045414fdc0
[ 154.390542] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390542] R10: 00007f045414fd60 R11: 0000000000008cda R12: 0000000000000001
[ 154.390542] R13: 0000000000003c80 R14: 00007f045414fd90 R15: 0000555bc39a6100
[ 154.390543] NMI watchdog: Watchdog detected hard LOCKUP on cpu 59
[ 154.390543] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390558] irq event stamp: 16215
[ 154.390558] hardirqs last enabled at (16214): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390558] hardirqs last disabled at (16215): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390559] softirqs last enabled at (16200): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390559] softirqs last disabled at (16211): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390560] CPU: 59 PID: 7101 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390560] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390560] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390561] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390561] RSP: 0000:ffffc9001920fce8 EFLAGS: 00000046
[ 154.390561] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000f00000
[ 154.390562] RDX: ffff88c026bff1c0 RSI: 0000000000000034 RDI: ffff88c0293fe380
[ 154.390562] RBP: ffff88c0293fe380 R08: 0000000000f00000 R09: 00000000001fd9c0
[ 154.390562] R10: ffffc9001920fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390563] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390563] FS: 00007f045412e700(0000) GS:ffff88c026a00000(0000) knlGS:0000000000000000
[ 154.390563] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390564] CR2: 00007f045412e730 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390564] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390564] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390565] PKRU: 55555554
[ 154.390565] Call Trace:
[ 154.390565] do_raw_spin_lock+0xb1/0xc0
[ 154.390565] _raw_spin_lock_nested+0x50/0x60
[ 154.390565] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390566] load_balance+0x2f7/0xd60
[ 154.390566] rebalance_domains+0x2f6/0x4d0
[ 154.390566] __do_softirq+0xd4/0x432
[ 154.390566] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390567] irq_exit_rcu+0xbc/0xc0
[ 154.390567] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390567] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390567] RIP: 0033:0x555bc1620f87
[ 154.390568] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390568] RSP: 002b:00007f045412dd40 EFLAGS: 00000246
[ 154.390568] RAX: 00000000000001c1 RBX: 0000000000004efd RCX: 000000000000002d
[ 154.390569] RDX: 0000000000000010 RSI: 000000000000007b RDI: 00007f045412ddc0
[ 154.390569] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390569] R10: 00007f045412dd60 R11: 0000000000008cd6 R12: 0000000000000001
[ 154.390570] R13: 0000000000003d80 R14: 00007f045412dd90 R15: 0000555bc39a6200
[ 154.390570] NMI backtrace for cpu 59
[ 154.390570] CPU: 59 PID: 7101 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390571] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390571] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390571] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390572] RSP: 0000:ffffc9001920fce8 EFLAGS: 00000046
[ 154.390572] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000f00000
[ 154.390572] RDX: ffff88c026bff1c0 RSI: 0000000000000034 RDI: ffff88c0293fe380
[ 154.390573] RBP: ffff88c0293fe380 R08: 0000000000f00000 R09: 00000000001fd9c0
[ 154.390573] R10: ffffc9001920fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390573] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390574] FS: 00007f045412e700(0000) GS:ffff88c026a00000(0000) knlGS:0000000000000000
[ 154.390574] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390574] CR2: 00007f045412e730 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390574] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390575] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390575] PKRU: 55555554
[ 154.390575] Call Trace:
[ 154.390575] do_raw_spin_lock+0xb1/0xc0
[ 154.390576] _raw_spin_lock_nested+0x50/0x60
[ 154.390576] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390576] load_balance+0x2f7/0xd60
[ 154.390576] rebalance_domains+0x2f6/0x4d0
[ 154.390577] __do_softirq+0xd4/0x432
[ 154.390577] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390577] irq_exit_rcu+0xbc/0xc0
[ 154.390577] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390578] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390578] RIP: 0033:0x555bc1620f87
[ 154.390578] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390579] RSP: 002b:00007f045412dd40 EFLAGS: 00000246
[ 154.390579] RAX: 00000000000001c1 RBX: 0000000000004efd RCX: 000000000000002d
[ 154.390579] RDX: 0000000000000010 RSI: 000000000000007b RDI: 00007f045412ddc0
[ 154.390580] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390580] R10: 00007f045412dd60 R11: 0000000000008cd6 R12: 0000000000000001
[ 154.390580] R13: 0000000000003d80 R14: 00007f045412dd90 R15: 0000555bc39a6200
[ 154.390581] NMI watchdog: Watchdog detected hard LOCKUP on cpu 60
[ 154.390581] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390596] irq event stamp: 15855
[ 154.390596] hardirqs last enabled at (15854): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390596] hardirqs last disabled at (15855): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390597] softirqs last enabled at (15844): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390597] softirqs last disabled at (15851): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390598] CPU: 60 PID: 7108 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390598] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390598] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390599] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390599] RSP: 0000:ffffc90019287ce8 EFLAGS: 00000046
[ 154.390600] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000f40000
[ 154.390600] RDX: ffff88c026fff1c0 RSI: 000000000000003d RDI: ffff88c0293fe380
[ 154.390600] RBP: ffff88c0293fe380 R08: 0000000000f40000 R09: 00000000001fd9c0
[ 154.390600] R10: ffffc90019287c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390601] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390601] FS: 00007f04540d9700(0000) GS:ffff88c026e00000(0000) knlGS:0000000000000000
[ 154.390601] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390602] CR2: 00007f04540d9730 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390602] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390602] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390602] PKRU: 55555554
[ 154.390603] Call Trace:
[ 154.390603] do_raw_spin_lock+0xb1/0xc0
[ 154.390603] _raw_spin_lock_nested+0x50/0x60
[ 154.390603] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390604] load_balance+0x2f7/0xd60
[ 154.390604] rebalance_domains+0x2f6/0x4d0
[ 154.390604] __do_softirq+0xd4/0x432
[ 154.390604] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390604] irq_exit_rcu+0xbc/0xc0
[ 154.390605] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390605] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390605] RIP: 0033:0x555bc1620f87
[ 154.390606] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390606] RSP: 002b:00007f04540d8d40 EFLAGS: 00000246
[ 154.390607] RAX: 00000000000000c0 RBX: 00000000000043c7 RCX: 000000000000005a
[ 154.390607] RDX: 0000000000000047 RSI: 0000000000000080 RDI: 00007f04540d8dc0
[ 154.390607] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390607] R10: 00007f04540d8d60 R11: 0000000000008cd6 R12: 0000000000000001
[ 154.390608] R13: 0000000000004000 R14: 00007f04540d8d90 R15: 0000555bc39a6480
[ 154.390608] NMI backtrace for cpu 60
[ 154.390608] CPU: 60 PID: 7108 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390609] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390609] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390609] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390610] RSP: 0000:ffffc90019287ce8 EFLAGS: 00000046
[ 154.390610] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000f40000
[ 154.390611] RDX: ffff88c026fff1c0 RSI: 000000000000003d RDI: ffff88c0293fe380
[ 154.390611] RBP: ffff88c0293fe380 R08: 0000000000f40000 R09: 00000000001fd9c0
[ 154.390611] R10: ffffc90019287c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390612] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390612] FS: 00007f04540d9700(0000) GS:ffff88c026e00000(0000) knlGS:0000000000000000
[ 154.390612] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390613] CR2: 00007f04540d9730 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390613] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390613] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390613] PKRU: 55555554
[ 154.390614] Call Trace:
[ 154.390614] do_raw_spin_lock+0xb1/0xc0
[ 154.390614] _raw_spin_lock_nested+0x50/0x60
[ 154.390614] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390614] load_balance+0x2f7/0xd60
[ 154.390615] rebalance_domains+0x2f6/0x4d0
[ 154.390615] __do_softirq+0xd4/0x432
[ 154.390615] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390615] irq_exit_rcu+0xbc/0xc0
[ 154.390616] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390616] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390616] RIP: 0033:0x555bc1620f87
[ 154.390616] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390617] RSP: 002b:00007f04540d8d40 EFLAGS: 00000246
[ 154.390617] RAX: 00000000000000c0 RBX: 00000000000043c7 RCX: 000000000000005a
[ 154.390618] RDX: 0000000000000047 RSI: 0000000000000080 RDI: 00007f04540d8dc0
[ 154.390618] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390618] R10: 00007f04540d8d60 R11: 0000000000008cd6 R12: 0000000000000001
[ 154.390619] R13: 0000000000004000 R14: 00007f04540d8d90 R15: 0000555bc39a6480
[ 154.390619] NMI watchdog: Watchdog detected hard LOCKUP on cpu 61
[ 154.390620] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390634] irq event stamp: 21743
[ 154.390634] hardirqs last enabled at (21742): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390635] hardirqs last disabled at (21743): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390635] softirqs last enabled at (21728): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390636] softirqs last disabled at (21739): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390636] CPU: 61 PID: 7128 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390636] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390637] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390637] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390637] RSP: 0000:ffffc9001938fce8 EFLAGS: 00000046
[ 154.390638] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000f80000
[ 154.390638] RDX: ffff88c0273ff1c0 RSI: 0000000000000037 RDI: ffff88c0293fe380
[ 154.390639] RBP: ffff88c0293fe380 R08: 0000000000f80000 R09: 00000000001fd9c0
[ 154.390639] R10: ffffc9001938fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390639] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390639] FS: 00007f0453fa7700(0000) GS:ffff88c027200000(0000) knlGS:0000000000000000
[ 154.390640] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390640] CR2: 00007f045424ed68 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390640] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390641] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390641] PKRU: 55555554
[ 154.390641] Call Trace:
[ 154.390641] do_raw_spin_lock+0xb1/0xc0
[ 154.390641] _raw_spin_lock_nested+0x50/0x60
[ 154.390642] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390642] load_balance+0x2f7/0xd60
[ 154.390642] rebalance_domains+0x2f6/0x4d0
[ 154.390642] __do_softirq+0xd4/0x432
[ 154.390643] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390643] irq_exit_rcu+0xbc/0xc0
[ 154.390643] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390643] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390644] RIP: 0033:0x555bc1620f87
[ 154.390644] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390644] RSP: 002b:00007f0453fa6d40 EFLAGS: 00000246
[ 154.390645] RAX: 000000000000013f RBX: 000000000001640f RCX: 000000000000011d
[ 154.390645] RDX: 00000000000000ec RSI: 0000000000000092 RDI: 00007f0453fa6dc0
[ 154.390645] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390646] R10: 00007f0453fa6d60 R11: 0000000000008cce R12: 0000000000000001
[ 154.390646] R13: 0000000000004900 R14: 00007f0453fa6d90 R15: 0000555bc39a6d80
[ 154.390646] NMI backtrace for cpu 61
[ 154.390647] CPU: 61 PID: 7128 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390647] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390647] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390648] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390648] RSP: 0000:ffffc9001938fce8 EFLAGS: 00000046
[ 154.390648] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000f80000
[ 154.390649] RDX: ffff88c0273ff1c0 RSI: 0000000000000037 RDI: ffff88c0293fe380
[ 154.390649] RBP: ffff88c0293fe380 R08: 0000000000f80000 R09: 00000000001fd9c0
[ 154.390650] R10: ffffc9001938fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390650] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390650] FS: 00007f0453fa7700(0000) GS:ffff88c027200000(0000) knlGS:0000000000000000
[ 154.390650] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390651] CR2: 00007f045424ed68 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390651] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390651] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390652] PKRU: 55555554
[ 154.390652] Call Trace:
[ 154.390652] do_raw_spin_lock+0xb1/0xc0
[ 154.390652] _raw_spin_lock_nested+0x50/0x60
[ 154.390652] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390653] load_balance+0x2f7/0xd60
[ 154.390653] rebalance_domains+0x2f6/0x4d0
[ 154.390653] __do_softirq+0xd4/0x432
[ 154.390653] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390654] irq_exit_rcu+0xbc/0xc0
[ 154.390654] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390654] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390654] RIP: 0033:0x555bc1620f87
[ 154.390655] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390655] RSP: 002b:00007f0453fa6d40 EFLAGS: 00000246
[ 154.390655] RAX: 000000000000013f RBX: 000000000001640f RCX: 000000000000011d
[ 154.390656] RDX: 00000000000000ec RSI: 0000000000000092 RDI: 00007f0453fa6dc0
[ 154.390656] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390657] R10: 00007f0453fa6d60 R11: 0000000000008cce R12: 0000000000000001
[ 154.390657] R13: 0000000000004900 R14: 00007f0453fa6d90 R15: 0000555bc39a6d80
[ 154.390657] NMI watchdog: Watchdog detected hard LOCKUP on cpu 62
[ 154.390658] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390672] irq event stamp: 15889
[ 154.390673] hardirqs last enabled at (15888): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390673] hardirqs last disabled at (15889): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390673] softirqs last enabled at (15878): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390674] softirqs last disabled at (15885): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390674] CPU: 62 PID: 7109 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390674] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390675] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390675] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390676] RSP: 0000:ffffc9001928fce8 EFLAGS: 00000046
[ 154.390676] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000fc0000
[ 154.390676] RDX: ffff88c0277ff1c0 RSI: 000000000000003f RDI: ffff88c0293fe380
[ 154.390677] RBP: ffff88c0293fe380 R08: 0000000000fc0000 R09: 00000000001fd9c0
[ 154.390677] R10: ffffc9001928fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390677] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390677] FS: 00007f04540c8700(0000) GS:ffff88c027600000(0000) knlGS:0000000000000000
[ 154.390678] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390678] CR2: 00007f0456e53158 CR3: 000000011a18a005 CR4: 00000000007706e0
[ 154.390678] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390679] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390679] PKRU: 55555554
[ 154.390679] Call Trace:
[ 154.390679] do_raw_spin_lock+0xb1/0xc0
[ 154.390680] _raw_spin_lock_nested+0x50/0x60
[ 154.390680] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390680] load_balance+0x2f7/0xd60
[ 154.390680] rebalance_domains+0x2f6/0x4d0
[ 154.390680] __do_softirq+0xd4/0x432
[ 154.390681] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390681] irq_exit_rcu+0xbc/0xc0
[ 154.390681] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390681] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390682] RIP: 0033:0x555bc1620f87
[ 154.390682] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390682] RSP: 002b:00007f04540c7d40 EFLAGS: 00000246
[ 154.390683] RAX: 0000000000000173 RBX: 000000000000f995 RCX: 00000000000000ac
[ 154.390683] RDX: 0000000000000051 RSI: 0000000000000081 RDI: 00007f04540c7dc0
[ 154.390683] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390684] R10: 00007f04540c7d60 R11: 0000000000008cd2 R12: 0000000000000001
[ 154.390684] R13: 0000000000004080 R14: 00007f04540c7d90 R15: 0000555bc39a6500
[ 154.390684] NMI backtrace for cpu 62
[ 154.390685] CPU: 62 PID: 7109 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390685] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390685] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390686] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390686] RSP: 0000:ffffc9001928fce8 EFLAGS: 00000046
[ 154.390687] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000000fc0000
[ 154.390687] RDX: ffff88c0277ff1c0 RSI: 000000000000003f RDI: ffff88c0293fe380
[ 154.390687] RBP: ffff88c0293fe380 R08: 0000000000fc0000 R09: 00000000001fd9c0
[ 154.390687] R10: ffffc9001928fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390688] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390688] FS: 00007f04540c8700(0000) GS:ffff88c027600000(0000) knlGS:0000000000000000
[ 154.390688] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390689] CR2: 00007f0456e53158 CR3: 000000011a18a005 CR4: 00000000007706e0
[ 154.390689] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390689] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390689] PKRU: 55555554
[ 154.390690] Call Trace:
[ 154.390690] do_raw_spin_lock+0xb1/0xc0
[ 154.390690] _raw_spin_lock_nested+0x50/0x60
[ 154.390690] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390691] load_balance+0x2f7/0xd60
[ 154.390691] rebalance_domains+0x2f6/0x4d0
[ 154.390691] __do_softirq+0xd4/0x432
[ 154.390691] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390692] irq_exit_rcu+0xbc/0xc0
[ 154.390692] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390692] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390692] RIP: 0033:0x555bc1620f87
[ 154.390693] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390693] RSP: 002b:00007f04540c7d40 EFLAGS: 00000246
[ 154.390694] RAX: 0000000000000173 RBX: 000000000000f995 RCX: 00000000000000ac
[ 154.390694] RDX: 0000000000000051 RSI: 0000000000000081 RDI: 00007f04540c7dc0
[ 154.390694] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390695] R10: 00007f04540c7d60 R11: 0000000000008cd2 R12: 0000000000000001
[ 154.390695] R13: 0000000000004080 R14: 00007f04540c7d90 R15: 0000555bc39a6500
[ 154.390696] NMI watchdog: Watchdog detected hard LOCKUP on cpu 63
[ 154.390696] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390710] irq event stamp: 16025
[ 154.390711] hardirqs last enabled at (16024): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390711] hardirqs last disabled at (16025): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390711] softirqs last enabled at (16010): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390712] softirqs last disabled at (16021): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390712] CPU: 63 PID: 7110 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390712] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390713] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390713] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390714] RSP: 0000:ffffc90019297ce8 EFLAGS: 00000046
[ 154.390714] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001000000
[ 154.390714] RDX: ffff88c027bff1c0 RSI: 000000000000003b RDI: ffff88c0293fe380
[ 154.390715] RBP: ffff88c0293fe380 R08: 0000000001000000 R09: 00000000001fd9c0
[ 154.390715] R10: ffffc90019297c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390715] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390715] FS: 00007f04540b7700(0000) GS:ffff88c027a00000(0000) knlGS:0000000000000000
[ 154.390716] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390716] CR2: 00007f04540b7730 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390716] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390717] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390717] PKRU: 55555554
[ 154.390717] Call Trace:
[ 154.390717] do_raw_spin_lock+0xb1/0xc0
[ 154.390718] _raw_spin_lock_nested+0x50/0x60
[ 154.390718] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390718] load_balance+0x2f7/0xd60
[ 154.390718] rebalance_domains+0x2f6/0x4d0
[ 154.390718] __do_softirq+0xd4/0x432
[ 154.390719] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390719] irq_exit_rcu+0xbc/0xc0
[ 154.390719] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390719] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390720] RIP: 0033:0x555bc1620f87
[ 154.390720] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390721] RSP: 002b:00007f04540b6d40 EFLAGS: 00000246
[ 154.390721] RAX: 000000000000013a RBX: 0000000000006749 RCX: 0000000000000054
[ 154.390721] RDX: 0000000000000041 RSI: 0000000000000082 RDI: 00007f04540b6dc0
[ 154.390722] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390722] R10: 00007f04540b6d60 R11: 0000000000008cd6 R12: 0000000000000001
[ 154.390722] R13: 0000000000004100 R14: 00007f04540b6d90 R15: 0000555bc39a6580
[ 154.390722] NMI backtrace for cpu 63
[ 154.390723] CPU: 63 PID: 7110 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390723] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390723] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390724] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390724] RSP: 0000:ffffc90019297ce8 EFLAGS: 00000046
[ 154.390725] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001000000
[ 154.390725] RDX: ffff88c027bff1c0 RSI: 000000000000003b RDI: ffff88c0293fe380
[ 154.390725] RBP: ffff88c0293fe380 R08: 0000000001000000 R09: 00000000001fd9c0
[ 154.390726] R10: ffffc90019297c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390726] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390726] FS: 00007f04540b7700(0000) GS:ffff88c027a00000(0000) knlGS:0000000000000000
[ 154.390726] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390727] CR2: 00007f04540b7730 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390727] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390727] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390728] PKRU: 55555554
[ 154.390728] Call Trace:
[ 154.390728] do_raw_spin_lock+0xb1/0xc0
[ 154.390728] _raw_spin_lock_nested+0x50/0x60
[ 154.390728] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390729] load_balance+0x2f7/0xd60
[ 154.390729] rebalance_domains+0x2f6/0x4d0
[ 154.390729] __do_softirq+0xd4/0x432
[ 154.390729] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390730] irq_exit_rcu+0xbc/0xc0
[ 154.390730] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390730] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390730] RIP: 0033:0x555bc1620f87
[ 154.390731] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390731] RSP: 002b:00007f04540b6d40 EFLAGS: 00000246
[ 154.390732] RAX: 000000000000013a RBX: 0000000000006749 RCX: 0000000000000054
[ 154.390732] RDX: 0000000000000041 RSI: 0000000000000082 RDI: 00007f04540b6dc0
[ 154.390732] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390733] R10: 00007f04540b6d60 R11: 0000000000008cd6 R12: 0000000000000001
[ 154.390733] R13: 0000000000004100 R14: 00007f04540b6d90 R15: 0000555bc39a6580
[ 154.390733] NMI watchdog: Watchdog detected hard LOCKUP on cpu 64
[ 154.390734] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390748] irq event stamp: 38615
[ 154.390749] hardirqs last enabled at (38614): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390749] hardirqs last disabled at (38615): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390749] softirqs last enabled at (38584): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390750] softirqs last disabled at (38611): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390750] CPU: 64 PID: 6974 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390750] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390751] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390751] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390751] RSP: 0000:ffffc9001844fce8 EFLAGS: 00000046
[ 154.390752] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001040000
[ 154.390752] RDX: ffff88c027fff1c0 RSI: 0000000000000041 RDI: ffff88c0293fe380
[ 154.390753] RBP: ffff88c0293fe380 R08: 0000000001040000 R09: 00000000001fd9c0
[ 154.390753] R10: ffffc9001844fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390753] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390753] FS: 00007f045486b700(0000) GS:ffff88c027e00000(0000) knlGS:0000000000000000
[ 154.390754] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390754] CR2: 00007f0456d51d68 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.390754] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390755] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390755] PKRU: 55555554
[ 154.390755] Call Trace:
[ 154.390755] do_raw_spin_lock+0xb1/0xc0
[ 154.390755] _raw_spin_lock_nested+0x50/0x60
[ 154.390756] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390756] load_balance+0x2f7/0xd60
[ 154.390756] rebalance_domains+0x2f6/0x4d0
[ 154.390756] __do_softirq+0xd4/0x432
[ 154.390757] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390757] irq_exit_rcu+0xbc/0xc0
[ 154.390757] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390757] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390758] RIP: 0033:0x555bc1620f87
[ 154.390758] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390758] RSP: 002b:00007f045486ad40 EFLAGS: 00000246
[ 154.390759] RAX: 0000000000000398 RBX: 000000000000c941 RCX: 0000000000000038
[ 154.390759] RDX: 0000000000000001 RSI: 000000000000000e RDI: 00007f045486adc0
[ 154.390759] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390760] R10: 00007f045486ad60 R11: 0000000000008fd0 R12: 0000000000000001
[ 154.390760] R13: 0000000000000700 R14: 00007f045486ad90 R15: 0000555bc39a2b80
[ 154.390760] NMI backtrace for cpu 64
[ 154.390761] CPU: 64 PID: 6974 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390761] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390761] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390762] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390762] RSP: 0000:ffffc9001844fce8 EFLAGS: 00000046
[ 154.390763] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001040000
[ 154.390763] RDX: ffff88c027fff1c0 RSI: 0000000000000041 RDI: ffff88c0293fe380
[ 154.390763] RBP: ffff88c0293fe380 R08: 0000000001040000 R09: 00000000001fd9c0
[ 154.390764] R10: ffffc9001844fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390764] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390764] FS: 00007f045486b700(0000) GS:ffff88c027e00000(0000) knlGS:0000000000000000
[ 154.390764] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390765] CR2: 00007f0456d51d68 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.390765] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390765] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390766] PKRU: 55555554
[ 154.390766] Call Trace:
[ 154.390766] do_raw_spin_lock+0xb1/0xc0
[ 154.390766] _raw_spin_lock_nested+0x50/0x60
[ 154.390766] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390767] load_balance+0x2f7/0xd60
[ 154.390767] rebalance_domains+0x2f6/0x4d0
[ 154.390767] __do_softirq+0xd4/0x432
[ 154.390767] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390768] irq_exit_rcu+0xbc/0xc0
[ 154.390768] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390768] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390768] RIP: 0033:0x555bc1620f87
[ 154.390769] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390769] RSP: 002b:00007f045486ad40 EFLAGS: 00000246
[ 154.390770] RAX: 0000000000000398 RBX: 000000000000c941 RCX: 0000000000000038
[ 154.390770] RDX: 0000000000000001 RSI: 000000000000000e RDI: 00007f045486adc0
[ 154.390770] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390771] R10: 00007f045486ad60 R11: 0000000000008fd0 R12: 0000000000000001
[ 154.390771] R13: 0000000000000700 R14: 00007f045486ad90 R15: 0000555bc39a2b80
[ 154.390772] NMI watchdog: Watchdog detected hard LOCKUP on cpu 65
[ 154.390772] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390786] irq event stamp: 16975
[ 154.390787] hardirqs last enabled at (16974): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390787] hardirqs last disabled at (16975): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390787] softirqs last enabled at (16948): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390788] softirqs last disabled at (16971): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390788] CPU: 65 PID: 7112 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390788] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390789] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390789] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390790] RSP: 0000:ffffc900192a7ce8 EFLAGS: 00000046
[ 154.390790] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001080000
[ 154.390790] RDX: ffff88c0283ff1c0 RSI: 0000000000000046 RDI: ffff88c0293fe380
[ 154.390791] RBP: ffff88c0293fe380 R08: 0000000001080000 R09: 00000000001fd9c0
[ 154.390791] R10: ffffc900192a7c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390791] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390792] FS: 00007f0454095700(0000) GS:ffff88c028200000(0000) knlGS:0000000000000000
[ 154.390792] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390792] CR2: 00007f0456e23148 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390792] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390793] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390793] PKRU: 55555554
[ 154.390793] Call Trace:
[ 154.390793] do_raw_spin_lock+0xb1/0xc0
[ 154.390794] _raw_spin_lock_nested+0x50/0x60
[ 154.390794] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390794] load_balance+0x2f7/0xd60
[ 154.390794] rebalance_domains+0x2f6/0x4d0
[ 154.390795] __do_softirq+0xd4/0x432
[ 154.390795] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390795] irq_exit_rcu+0xbc/0xc0
[ 154.390795] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390795] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390796] RIP: 0033:0x555bc1620f87
[ 154.390796] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390797] RSP: 002b:00007f0454094d40 EFLAGS: 00000246
[ 154.390797] RAX: 0000000000000440 RBX: 00000000000061c7 RCX: 0000000000000017
[ 154.390797] RDX: 0000000000000007 RSI: 0000000000000084 RDI: 00007f0454094dc0
[ 154.390798] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390798] R10: 00007f0454094d60 R11: 0000000000008fd4 R12: 0000000000000001
[ 154.390798] R13: 0000000000004200 R14: 00007f0454094d90 R15: 0000555bc39a6680
[ 154.390798] NMI backtrace for cpu 65
[ 154.390799] CPU: 65 PID: 7112 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390799] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390799] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390800] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390800] RSP: 0000:ffffc900192a7ce8 EFLAGS: 00000046
[ 154.390801] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001080000
[ 154.390801] RDX: ffff88c0283ff1c0 RSI: 0000000000000046 RDI: ffff88c0293fe380
[ 154.390801] RBP: ffff88c0293fe380 R08: 0000000001080000 R09: 00000000001fd9c0
[ 154.390802] R10: ffffc900192a7c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390802] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390802] FS: 00007f0454095700(0000) GS:ffff88c028200000(0000) knlGS:0000000000000000
[ 154.390802] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390803] CR2: 00007f0456e23148 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.390803] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390803] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390804] PKRU: 55555554
[ 154.390804] Call Trace:
[ 154.390804] do_raw_spin_lock+0xb1/0xc0
[ 154.390804] _raw_spin_lock_nested+0x50/0x60
[ 154.390804] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390805] load_balance+0x2f7/0xd60
[ 154.390805] rebalance_domains+0x2f6/0x4d0
[ 154.390805] __do_softirq+0xd4/0x432
[ 154.390805] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390806] irq_exit_rcu+0xbc/0xc0
[ 154.390806] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390806] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390806] RIP: 0033:0x555bc1620f87
[ 154.390807] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390807] RSP: 002b:00007f0454094d40 EFLAGS: 00000246
[ 154.390808] RAX: 0000000000000440 RBX: 00000000000061c7 RCX: 0000000000000017
[ 154.390808] RDX: 0000000000000007 RSI: 0000000000000084 RDI: 00007f0454094dc0
[ 154.390808] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390809] R10: 00007f0454094d60 R11: 0000000000008fd4 R12: 0000000000000001
[ 154.390809] R13: 0000000000004200 R14: 00007f0454094d90 R15: 0000555bc39a6680
[ 154.390810] NMI watchdog: Watchdog detected hard LOCKUP on cpu 66
[ 154.390810] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390825] irq event stamp: 23593
[ 154.390825] hardirqs last enabled at (23592): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390825] hardirqs last disabled at (23593): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390825] softirqs last enabled at (23574): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390826] softirqs last disabled at (23589): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390826] CPU: 66 PID: 6960 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390827] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390827] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390827] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390828] RSP: 0000:ffffc90017fe7ce8 EFLAGS: 00000046
[ 154.390828] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 00000000010c0000
[ 154.390828] RDX: ffff88c0287ff1c0 RSI: 0000000000000043 RDI: ffff88c0293fe380
[ 154.390829] RBP: ffff88c0293fe380 R08: 00000000010c0000 R09: 00000000001fd9c0
[ 154.390829] R10: ffffc90017fe7c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390829] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390830] FS: 00007f0456dc9700(0000) GS:ffff88c028600000(0000) knlGS:0000000000000000
[ 154.390830] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390830] CR2: 00007f0456ea7160 CR3: 000000011a18a006 CR4: 00000000007706e0
[ 154.390831] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390831] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390831] PKRU: 55555554
[ 154.390831] Call Trace:
[ 154.390831] do_raw_spin_lock+0xb1/0xc0
[ 154.390832] _raw_spin_lock_nested+0x50/0x60
[ 154.390832] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390832] load_balance+0x2f7/0xd60
[ 154.390832] rebalance_domains+0x2f6/0x4d0
[ 154.390833] __do_softirq+0xd4/0x432
[ 154.390833] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390833] irq_exit_rcu+0xbc/0xc0
[ 154.390833] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390834] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390834] RIP: 0033:0x555bc1620f87
[ 154.390834] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390835] RSP: 002b:00007f0456dc8d40 EFLAGS: 00000246
[ 154.390835] RAX: 0000000000000505 RBX: 000000000001284b RCX: 000000000000003b
[ 154.390835] RDX: 0000000000000024 RSI: 0000000000000004 RDI: 00007f0456dc8dc0
[ 154.390836] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390836] R10: 00007f0456dc8d60 R11: 0000000000008cd2 R12: 0000000000000001
[ 154.390836] R13: 0000000000000200 R14: 00007f0456dc8d90 R15: 0000555bc39a2680
[ 154.390837] NMI backtrace for cpu 66
[ 154.390837] CPU: 66 PID: 6960 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390837] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390837] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390838] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390838] RSP: 0000:ffffc90017fe7ce8 EFLAGS: 00000046
[ 154.390839] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 00000000010c0000
[ 154.390839] RDX: ffff88c0287ff1c0 RSI: 0000000000000043 RDI: ffff88c0293fe380
[ 154.390839] RBP: ffff88c0293fe380 R08: 00000000010c0000 R09: 00000000001fd9c0
[ 154.390840] R10: ffffc90017fe7c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390840] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390840] FS: 00007f0456dc9700(0000) GS:ffff88c028600000(0000) knlGS:0000000000000000
[ 154.390841] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390841] CR2: 00007f0456ea7160 CR3: 000000011a18a006 CR4: 00000000007706e0
[ 154.390841] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390841] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390842] PKRU: 55555554
[ 154.390842] Call Trace:
[ 154.390842] do_raw_spin_lock+0xb1/0xc0
[ 154.390842] _raw_spin_lock_nested+0x50/0x60
[ 154.390843] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390843] load_balance+0x2f7/0xd60
[ 154.390843] rebalance_domains+0x2f6/0x4d0
[ 154.390843] __do_softirq+0xd4/0x432
[ 154.390843] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390844] irq_exit_rcu+0xbc/0xc0
[ 154.390844] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390844] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390844] RIP: 0033:0x555bc1620f87
[ 154.390845] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390845] RSP: 002b:00007f0456dc8d40 EFLAGS: 00000246
[ 154.390846] RAX: 0000000000000505 RBX: 000000000001284b RCX: 000000000000003b
[ 154.390846] RDX: 0000000000000024 RSI: 0000000000000004 RDI: 00007f0456dc8dc0
[ 154.390846] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390846] R10: 00007f0456dc8d60 R11: 0000000000008cd2 R12: 0000000000000001
[ 154.390847] R13: 0000000000000200 R14: 00007f0456dc8d90 R15: 0000555bc39a2680
[ 154.390847] NMI watchdog: Watchdog detected hard LOCKUP on cpu 67
[ 154.390848] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390862] irq event stamp: 16461
[ 154.390863] hardirqs last enabled at (16460): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390863] hardirqs last disabled at (16461): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390863] softirqs last enabled at (16444): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390864] softirqs last disabled at (16457): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390864] CPU: 67 PID: 7173 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390865] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390865] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390865] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390866] RSP: 0000:ffffc9001956fce8 EFLAGS: 00000046
[ 154.390866] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001100000
[ 154.390866] RDX: ffff88c028bff1c0 RSI: 0000000000000035 RDI: ffff88c0293fe380
[ 154.390867] RBP: ffff88c0293fe380 R08: 0000000001100000 R09: 00000000001fd9c0
[ 154.390867] R10: ffffc9001956fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390867] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390868] FS: 00007f0453d10700(0000) GS:ffff88c028a00000(0000) knlGS:0000000000000000
[ 154.390868] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390868] CR2: 00007f0456e77160 CR3: 000000011a18a006 CR4: 00000000007706e0
[ 154.390869] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390869] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390869] PKRU: 55555554
[ 154.390869] Call Trace:
[ 154.390870] do_raw_spin_lock+0xb1/0xc0
[ 154.390870] _raw_spin_lock_nested+0x50/0x60
[ 154.390870] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390870] load_balance+0x2f7/0xd60
[ 154.390870] rebalance_domains+0x2f6/0x4d0
[ 154.390871] __do_softirq+0xd4/0x432
[ 154.390871] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390871] irq_exit_rcu+0xbc/0xc0
[ 154.390871] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390872] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390872] RIP: 0033:0x555bc1620f7f
[ 154.390872] Code: d8 f2 0f 51 c8 77 5d 66 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 <31> d2 48 89 d8 48 f7 f1 48 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8
[ 154.390873] RSP: 002b:00007f0453d0fd40 EFLAGS: 00000202
[ 154.390873] RAX: 0000000000000121 RBX: 0000000000008563 RCX: 0000000000000077
[ 154.390873] RDX: 000000000000002d RSI: 00000000000000b9 RDI: 00007f0453d0fdc0
[ 154.390874] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390874] R10: 00007f0453d0fd60 R11: 0000000000008cd6 R12: 0000000000000001
[ 154.390874] R13: 0000000000005c80 R14: 00007f0453d0fd90 R15: 0000555bc39a8100
[ 154.390875] NMI backtrace for cpu 67
[ 154.390875] CPU: 67 PID: 7173 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390875] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390876] RIP: 0010:native_queued_spin_lock_slowpath+0x194/0x1e0
[ 154.390876] Code: 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 8b 42 08 <85> c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07 85 c0 75
[ 154.390876] RSP: 0000:ffffc9001956fce8 EFLAGS: 00000046
[ 154.390877] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001100000
[ 154.390877] RDX: ffff88c028bff1c0 RSI: 0000000000000035 RDI: ffff88c0293fe380
[ 154.390878] RBP: ffff88c0293fe380 R08: 0000000001100000 R09: 00000000001fd9c0
[ 154.390878] R10: ffffc9001956fc98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390878] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390879] FS: 00007f0453d10700(0000) GS:ffff88c028a00000(0000) knlGS:0000000000000000
[ 154.390879] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390879] CR2: 00007f0456e77160 CR3: 000000011a18a006 CR4: 00000000007706e0
[ 154.390880] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390880] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390880] PKRU: 55555554
[ 154.390880] Call Trace:
[ 154.390880] do_raw_spin_lock+0xb1/0xc0
[ 154.390881] _raw_spin_lock_nested+0x50/0x60
[ 154.390881] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390881] load_balance+0x2f7/0xd60
[ 154.390881] rebalance_domains+0x2f6/0x4d0
[ 154.390882] __do_softirq+0xd4/0x432
[ 154.390882] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390882] irq_exit_rcu+0xbc/0xc0
[ 154.390882] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390883] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390883] RIP: 0033:0x555bc1620f7f
[ 154.390883] Code: d8 f2 0f 51 c8 77 5d 66 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 <31> d2 48 89 d8 48 f7 f1 48 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8
[ 154.390884] RSP: 002b:00007f0453d0fd40 EFLAGS: 00000202
[ 154.390884] RAX: 0000000000000121 RBX: 0000000000008563 RCX: 0000000000000077
[ 154.390884] RDX: 000000000000002d RSI: 00000000000000b9 RDI: 00007f0453d0fdc0
[ 154.390885] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390885] R10: 00007f0453d0fd60 R11: 0000000000008cd6 R12: 0000000000000001
[ 154.390885] R13: 0000000000005c80 R14: 00007f0453d0fd90 R15: 0000555bc39a8100
[ 154.390886] NMI watchdog: Watchdog detected hard LOCKUP on cpu 68
[ 154.390886] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390901] irq event stamp: 23769
[ 154.390901] hardirqs last enabled at (23768): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390901] hardirqs last disabled at (23769): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390902] softirqs last enabled at (23750): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390902] softirqs last disabled at (23765): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390903] CPU: 68 PID: 6967 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390903] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390903] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390904] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390904] RSP: 0000:ffffc90018427ce8 EFLAGS: 00000046
[ 154.390904] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001140000
[ 154.390905] RDX: ffff88c028fff1c0 RSI: 000000000000003a RDI: ffff88c0293fe380
[ 154.390905] RBP: ffff88c0293fe380 R08: 0000000001140000 R09: 00000000001fd9c0
[ 154.390905] R10: ffffc90018427c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390906] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390906] FS: 00007f0456d74700(0000) GS:ffff88c028e00000(0000) knlGS:0000000000000000
[ 154.390906] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390907] CR2: 00007f0456ebb160 CR3: 000000011a18a005 CR4: 00000000007706e0
[ 154.390907] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390907] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390908] PKRU: 55555554
[ 154.390908] Call Trace:
[ 154.390908] do_raw_spin_lock+0xb1/0xc0
[ 154.390908] _raw_spin_lock_nested+0x50/0x60
[ 154.390909] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390909] load_balance+0x2f7/0xd60
[ 154.390909] rebalance_domains+0x2f6/0x4d0
[ 154.390909] __do_softirq+0xd4/0x432
[ 154.390909] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390910] irq_exit_rcu+0xbc/0xc0
[ 154.390910] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390910] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390910] RIP: 0033:0x555bc1620f87
[ 154.390911] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390911] RSP: 002b:00007f0456d73d40 EFLAGS: 00000246
[ 154.390912] RAX: 00000000000005ce RBX: 0000000000013f61 RCX: 0000000000000037
[ 154.390912] RDX: 000000000000001f RSI: 0000000000000009 RDI: 00007f0456d73dc0
[ 154.390912] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390913] R10: 00007f0456d73d60 R11: 0000000000008cd4 R12: 0000000000000001
[ 154.390913] R13: 0000000000000480 R14: 00007f0456d73d90 R15: 0000555bc39a2900
[ 154.390913] NMI backtrace for cpu 68
[ 154.390913] CPU: 68 PID: 6967 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390914] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390914] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390914] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390915] RSP: 0000:ffffc90018427ce8 EFLAGS: 00000046
[ 154.390915] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001140000
[ 154.390916] RDX: ffff88c028fff1c0 RSI: 000000000000003a RDI: ffff88c0293fe380
[ 154.390916] RBP: ffff88c0293fe380 R08: 0000000001140000 R09: 00000000001fd9c0
[ 154.390916] R10: ffffc90018427c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390917] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390917] FS: 00007f0456d74700(0000) GS:ffff88c028e00000(0000) knlGS:0000000000000000
[ 154.390917] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390917] CR2: 00007f0456ebb160 CR3: 000000011a18a005 CR4: 00000000007706e0
[ 154.390918] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390918] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390918] PKRU: 55555554
[ 154.390918] Call Trace:
[ 154.390919] do_raw_spin_lock+0xb1/0xc0
[ 154.390919] _raw_spin_lock_nested+0x50/0x60
[ 154.390919] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390919] load_balance+0x2f7/0xd60
[ 154.390920] rebalance_domains+0x2f6/0x4d0
[ 154.390920] __do_softirq+0xd4/0x432
[ 154.390920] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390920] irq_exit_rcu+0xbc/0xc0
[ 154.390920] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390921] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390921] RIP: 0033:0x555bc1620f87
[ 154.390921] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.390922] RSP: 002b:00007f0456d73d40 EFLAGS: 00000246
[ 154.390922] RAX: 00000000000005ce RBX: 0000000000013f61 RCX: 0000000000000037
[ 154.390922] RDX: 000000000000001f RSI: 0000000000000009 RDI: 00007f0456d73dc0
[ 154.390923] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390923] R10: 00007f0456d73d60 R11: 0000000000008cd4 R12: 0000000000000001
[ 154.390923] R13: 0000000000000480 R14: 00007f0456d73d90 R15: 0000555bc39a2900
[ 154.390924] NMI watchdog: Watchdog detected hard LOCKUP on cpu 69
[ 154.390924] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390939] irq event stamp: 80340
[ 154.390939] hardirqs last enabled at (80339): [<ffffffff810da5c2>] sched_core_balance+0x182/0x3d0
[ 154.390940] hardirqs last disabled at (80340): [<ffffffff810da647>] sched_core_balance+0x207/0x3d0
[ 154.390940] softirqs last enabled at (79702): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390940] softirqs last disabled at (79695): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390941] CPU: 69 PID: 0 Comm: swapper/69 Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390941] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390941] RIP: 0010:native_queued_spin_lock_slowpath+0x62/0x1e0
[ 154.390942] Code: ba 2f 08 0f 92 c0 0f b6 c0 c1 e0 08 89 c2 8b 07 30 e4 09 d0 a9 00 01 ff ff 75 1e 85 c0 75 0b b8 01 00 00 00 66 89 07 c3 f3 90 <8b> 07 84 c0 75 f8 b8 01 00 00 00 66 89 07 c3 f6 c4 01 75 04 c6 47
[ 154.390942] RSP: 0018:ffffc9000078fd18 EFLAGS: 00000002
[ 154.390943] RAX: 0000000001f40101 RBX: ffff88b004ffe380 RCX: 00000000000024b9
[ 154.390943] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff88b004ffe380
[ 154.390943] RBP: ffff88b004ffe380 R08: 0000000000000002 R09: 00000000001fd9c0
[ 154.390943] R10: ffffc9000078fcc8 R11: 00000000df3b7c2e R12: ffff88b004ffe380
[ 154.390944] R13: 0000000000000000 R14: ffff888105805f40 R15: ffff88c0293fe380
[ 154.390944] FS: 0000000000000000(0000) GS:ffff88c029200000(0000) knlGS:0000000000000000
[ 154.390944] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390945] CR2: 00005570aa6dbfff CR3: 000000013c89c003 CR4: 00000000007706e0
[ 154.390945] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390945] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390946] PKRU: 55555554
[ 154.390946] Call Trace:
[ 154.390946] do_raw_spin_lock+0xb1/0xc0
[ 154.390946] _raw_spin_lock_nested+0x50/0x60
[ 154.390946] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390947] sched_core_balance+0x115/0x3d0
[ 154.390947] do_balance_callbacks+0x59/0x70
[ 154.390947] finish_task_switch+0xa4/0x290
[ 154.390947] ? finish_task_switch+0x63/0x290
[ 154.390948] __schedule+0x3aa/0x14d0
[ 154.390948] ? lockdep_hardirqs_on_prepare+0xd9/0x180
[ 154.390948] schedule_idle+0x1e/0x40
[ 154.390948] do_idle+0x1b5/0x2f0
[ 154.390949] cpu_startup_entry+0x19/0x20
[ 154.390949] start_secondary+0x12a/0x170
[ 154.390949] secondary_startup_64_no_verify+0xc2/0xcb
[ 154.390949] NMI backtrace for cpu 69
[ 154.390950] CPU: 69 PID: 0 Comm: swapper/69 Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390950] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390950] RIP: 0010:native_queued_spin_lock_slowpath+0x62/0x1e0
[ 154.390951] Code: ba 2f 08 0f 92 c0 0f b6 c0 c1 e0 08 89 c2 8b 07 30 e4 09 d0 a9 00 01 ff ff 75 1e 85 c0 75 0b b8 01 00 00 00 66 89 07 c3 f3 90 <8b> 07 84 c0 75 f8 b8 01 00 00 00 66 89 07 c3 f6 c4 01 75 04 c6 47
[ 154.390951] RSP: 0018:ffffc9000078fd18 EFLAGS: 00000002
[ 154.390952] RAX: 0000000001f00101 RBX: ffff88b004ffe380 RCX: 00000000000024b9
[ 154.390952] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff88b004ffe380
[ 154.390952] RBP: ffff88b004ffe380 R08: 0000000000000002 R09: 00000000001fd9c0
[ 154.390953] R10: ffffc9000078fcc8 R11: 00000000df3b7c2e R12: ffff88b004ffe380
[ 154.390953] R13: 0000000000000000 R14: ffff888105805f40 R15: ffff88c0293fe380
[ 154.390953] FS: 0000000000000000(0000) GS:ffff88c029200000(0000) knlGS:0000000000000000
[ 154.390953] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390954] CR2: 00005570aa6dbfff CR3: 000000013c89c003 CR4: 00000000007706e0
[ 154.390954] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390954] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390955] PKRU: 55555554
[ 154.390955] Call Trace:
[ 154.390955] do_raw_spin_lock+0xb1/0xc0
[ 154.390955] _raw_spin_lock_nested+0x50/0x60
[ 154.390955] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390956] sched_core_balance+0x115/0x3d0
[ 154.390956] do_balance_callbacks+0x59/0x70
[ 154.390956] finish_task_switch+0xa4/0x290
[ 154.390956] ? finish_task_switch+0x63/0x290
[ 154.390957] __schedule+0x3aa/0x14d0
[ 154.390957] ? lockdep_hardirqs_on_prepare+0xd9/0x180
[ 154.390957] schedule_idle+0x1e/0x40
[ 154.390957] do_idle+0x1b5/0x2f0
[ 154.390958] cpu_startup_entry+0x19/0x20
[ 154.390958] start_secondary+0x12a/0x170
[ 154.390958] secondary_startup_64_no_verify+0xc2/0xcb
[ 154.390959] NMI watchdog: Watchdog detected hard LOCKUP on cpu 70
[ 154.390959] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.390974] irq event stamp: 19129
[ 154.390974] hardirqs last enabled at (19128): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.390974] hardirqs last disabled at (19129): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.390974] softirqs last enabled at (19098): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.390975] softirqs last disabled at (19125): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.390975] CPU: 70 PID: 7086 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390976] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390976] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390977] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390977] RSP: 0000:ffffc90019147ce8 EFLAGS: 00000046
[ 154.390977] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 00000000011c0000
[ 154.390978] RDX: ffff88c0297ff1c0 RSI: 0000000000000044 RDI: ffff88c0293fe380
[ 154.390978] RBP: ffff88c0293fe380 R08: 00000000011c0000 R09: 00000000001fd9c0
[ 154.390978] R10: ffffc90019147c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390979] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390979] FS: 00007f04541e9700(0000) GS:ffff88c029600000(0000) knlGS:0000000000000000
[ 154.390979] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390979] CR2: 00007f0456ea3158 CR3: 000000011a18a001 CR4: 00000000007706e0
[ 154.390980] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390980] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390980] PKRU: 55555554
[ 154.390980] Call Trace:
[ 154.390981] do_raw_spin_lock+0xb1/0xc0
[ 154.390981] _raw_spin_lock_nested+0x50/0x60
[ 154.390981] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390981] load_balance+0x2f7/0xd60
[ 154.390982] rebalance_domains+0x2f6/0x4d0
[ 154.390982] __do_softirq+0xd4/0x432
[ 154.390982] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390982] irq_exit_rcu+0xbc/0xc0
[ 154.390983] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390983] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390983] RIP: 0033:0x555bc1620f5a
[ 154.390983] Code: 00 00 00 66 0f ef db 66 0f 1f 84 00 00 00 00 00 48 83 c3 01 48 39 eb 74 67 66 0f ef c0 f2 48 0f 2a c3 66 0f 2e d8 f2 0f 51 c8 <77> 5d 66 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90
[ 154.390984] RSP: 002b:00007f04541e8d40 EFLAGS: 00000203
[ 154.390984] RAX: 0000000000001d5f RBX: 0000000000017dd4 RCX: 000000000000000d
[ 154.390985] RDX: 0000000000000000 RSI: 0000000000000070 RDI: 00007f04541e8dc0
[ 154.390985] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390985] R10: 00007f04541e8d60 R11: 0000000000008cfc R12: 0000000000000001
[ 154.390985] R13: 0000000000003800 R14: 00007f04541e8d90 R15: 0000555bc39a5c80
[ 154.390986] NMI backtrace for cpu 70
[ 154.390986] CPU: 70 PID: 7086 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.390986] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.390987] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.390987] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.390988] RSP: 0000:ffffc90019147ce8 EFLAGS: 00000046
[ 154.390988] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 00000000011c0000
[ 154.390988] RDX: ffff88c0297ff1c0 RSI: 0000000000000044 RDI: ffff88c0293fe380
[ 154.390989] RBP: ffff88c0293fe380 R08: 00000000011c0000 R09: 00000000001fd9c0
[ 154.390989] R10: ffffc90019147c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.390989] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.390989] FS: 00007f04541e9700(0000) GS:ffff88c029600000(0000) knlGS:0000000000000000
[ 154.390990] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.390990] CR2: 00007f0456ea3158 CR3: 000000011a18a001 CR4: 00000000007706e0
[ 154.390990] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.390991] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.390991] PKRU: 55555554
[ 154.390991] Call Trace:
[ 154.390991] do_raw_spin_lock+0xb1/0xc0
[ 154.390991] _raw_spin_lock_nested+0x50/0x60
[ 154.390992] raw_spin_rq_lock_nested+0x51/0x80
[ 154.390992] load_balance+0x2f7/0xd60
[ 154.390992] rebalance_domains+0x2f6/0x4d0
[ 154.390992] __do_softirq+0xd4/0x432
[ 154.390993] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.390993] irq_exit_rcu+0xbc/0xc0
[ 154.390993] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.390993] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.390994] RIP: 0033:0x555bc1620f5a
[ 154.390994] Code: 00 00 00 66 0f ef db 66 0f 1f 84 00 00 00 00 00 48 83 c3 01 48 39 eb 74 67 66 0f ef c0 f2 48 0f 2a c3 66 0f 2e d8 f2 0f 51 c8 <77> 5d 66 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90
[ 154.390994] RSP: 002b:00007f04541e8d40 EFLAGS: 00000203
[ 154.390995] RAX: 0000000000001d5f RBX: 0000000000017dd4 RCX: 000000000000000d
[ 154.390995] RDX: 0000000000000000 RSI: 0000000000000070 RDI: 00007f04541e8dc0
[ 154.390995] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.390996] R10: 00007f04541e8d60 R11: 0000000000008cfc R12: 0000000000000001
[ 154.390996] R13: 0000000000003800 R14: 00007f04541e8d90 R15: 0000555bc39a5c80
[ 154.390997] NMI watchdog: Watchdog detected hard LOCKUP on cpu 71
[ 154.390997] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.391011] irq event stamp: 16131
[ 154.391012] hardirqs last enabled at (16130): [<ffffffff810e5925>] update_blocked_averages+0x795/0x820
[ 154.391012] hardirqs last disabled at (16131): [<ffffffff810ee42d>] load_balance+0x2ed/0xd60
[ 154.391012] softirqs last enabled at (16120): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.391013] softirqs last disabled at (16127): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.391013] CPU: 71 PID: 7146 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.391014] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.391014] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.391014] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.391015] RSP: 0000:ffffc90019467ce8 EFLAGS: 00000046
[ 154.391015] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001200000
[ 154.391015] RDX: ffff88c029bff1c0 RSI: 000000000000003c RDI: ffff88c0293fe380
[ 154.391016] RBP: ffff88c0293fe380 R08: 0000000001200000 R09: 00000000001fd9c0
[ 154.391016] R10: ffffc90019467c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.391016] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.391017] FS: 00007f0453ea8700(0000) GS:ffff88c029a00000(0000) knlGS:0000000000000000
[ 154.391017] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.391017] CR2: 00007f0456ee5160 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.391017] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.391018] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.391018] PKRU: 55555554
[ 154.391018] Call Trace:
[ 154.391018] do_raw_spin_lock+0xb1/0xc0
[ 154.391019] _raw_spin_lock_nested+0x50/0x60
[ 154.391019] raw_spin_rq_lock_nested+0x51/0x80
[ 154.391019] load_balance+0x2f7/0xd60
[ 154.391019] rebalance_domains+0x2f6/0x4d0
[ 154.391020] __do_softirq+0xd4/0x432
[ 154.391020] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.391020] irq_exit_rcu+0xbc/0xc0
[ 154.391020] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.391021] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.391021] RIP: 0033:0x555bc1620f87
[ 154.391021] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.391022] RSP: 002b:00007f0453ea7d40 EFLAGS: 00000246
[ 154.391022] RAX: 00000000000000c7 RBX: 000000000000990d RCX: 00000000000000c4
[ 154.391022] RDX: 00000000000000b1 RSI: 00000000000000a1 RDI: 00007f0453ea7dc0
[ 154.391023] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.391023] R10: 00007f0453ea7d60 R11: 0000000000008cd4 R12: 0000000000000001
[ 154.391023] R13: 0000000000005080 R14: 00007f0453ea7d90 R15: 0000555bc39a7500
[ 154.391023] NMI backtrace for cpu 71
[ 154.391024] CPU: 71 PID: 7146 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.391024] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.391024] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.391025] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.391025] RSP: 0000:ffffc90019467ce8 EFLAGS: 00000046
[ 154.391026] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001200000
[ 154.391026] RDX: ffff88c029bff1c0 RSI: 000000000000003c RDI: ffff88c0293fe380
[ 154.391026] RBP: ffff88c0293fe380 R08: 0000000001200000 R09: 00000000001fd9c0
[ 154.391027] R10: ffffc90019467c98 R11: 00000000046fb5a6 R12: ffff88c0293fe380
[ 154.391027] R13: 0000000000000400 R14: 0000000000000200 R15: ffff88c0293fe380
[ 154.391027] FS: 00007f0453ea8700(0000) GS:ffff88c029a00000(0000) knlGS:0000000000000000
[ 154.391027] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.391028] CR2: 00007f0456ee5160 CR3: 000000011a18a002 CR4: 00000000007706e0
[ 154.391028] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.391028] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.391029] PKRU: 55555554
[ 154.391029] Call Trace:
[ 154.391029] do_raw_spin_lock+0xb1/0xc0
[ 154.391029] _raw_spin_lock_nested+0x50/0x60
[ 154.391029] raw_spin_rq_lock_nested+0x51/0x80
[ 154.391030] load_balance+0x2f7/0xd60
[ 154.391030] rebalance_domains+0x2f6/0x4d0
[ 154.391030] __do_softirq+0xd4/0x432
[ 154.391030] ? asm_sysvec_apic_timer_interrupt+0xa/0x20
[ 154.391031] irq_exit_rcu+0xbc/0xc0
[ 154.391031] sysvec_apic_timer_interrupt+0x77/0x90
[ 154.391031] asm_sysvec_apic_timer_interrupt+0x12/0x20
[ 154.391032] RIP: 0033:0x555bc1620f87
[ 154.391032] Code: 0f 2e ca 72 de f6 c3 01 74 d9 b9 02 00 00 00 eb 1e 66 90 66 0f ef c0 f2 48 0f 2a c1 66 0f 2e c8 72 c1 31 d2 48 89 d8 48 f7 f1 <48> 85 d2 74 b4 48 83 c1 01 79 de 48 89 c8 48 89 ca 66 0f ef c0 48
[ 154.391032] RSP: 002b:00007f0453ea7d40 EFLAGS: 00000246
[ 154.391033] RAX: 00000000000000c7 RBX: 000000000000990d RCX: 00000000000000c4
[ 154.391033] RDX: 00000000000000b1 RSI: 00000000000000a1 RDI: 00007f0453ea7dc0
[ 154.391033] RBP: 00000000000186a0 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.391034] R10: 00007f0453ea7d60 R11: 0000000000008cd4 R12: 0000000000000001
[ 154.391034] R13: 0000000000005080 R14: 00007f0453ea7d90 R15: 0000555bc39a7500
[ 154.391035] NMI watchdog: Watchdog detected hard LOCKUP on cpu 72
[ 154.391035] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.391050] irq event stamp: 67400
[ 154.391050] hardirqs last enabled at (67399): [<ffffffff81b05ab4>] _raw_spin_unlock_irq+0x24/0x40
[ 154.391050] hardirqs last disabled at (67400): [<ffffffff81afcf5a>] __schedule+0xd7a/0x14d0
[ 154.391050] softirqs last enabled at (67392): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.391051] softirqs last disabled at (67387): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.391051] CPU: 72 PID: 6964 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.391052] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.391052] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.391052] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.391053] RSP: 0000:ffffc9001840f988 EFLAGS: 00000046
[ 154.391053] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001240000
[ 154.391053] RDX: ffff88d01d3ff1c0 RSI: 0000000000000084 RDI: ffff88c0293fe380
[ 154.391054] RBP: ffff88c0293fe380 R08: 0000000001240000 R09: 00000000001fd9c0
[ 154.391054] R10: ffffc9001840f938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.391054] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.391055] FS: 00007f0456d96700(0000) GS:ffff88d01d200000(0000) knlGS:0000000000000000
[ 154.391055] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.391055] CR2: 0000555bc18acb48 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.391056] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.391056] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.391056] PKRU: 55555554
[ 154.391056] Call Trace:
[ 154.391057] do_raw_spin_lock+0xb1/0xc0
[ 154.391057] _raw_spin_lock_nested+0x50/0x60
[ 154.391057] raw_spin_rq_lock_nested+0x51/0x80
[ 154.391057] load_balance+0x2f7/0xd60
[ 154.391058] ? update_blocked_averages+0x500/0x820
[ 154.391058] newidle_balance+0x3d4/0x670
[ 154.391058] balance_fair+0x1b/0x30
[ 154.391058] put_prev_task_balance+0x4b/0x80
[ 154.391058] __schedule+0x57b/0x14d0
[ 154.391059] schedule+0x51/0xc0
[ 154.391059] io_schedule+0x12/0x40
[ 154.391059] __lock_page_killable+0x167/0x250
[ 154.391059] ? file_fdatawait_range+0x20/0x20
[ 154.391060] filemap_fault+0x9b5/0xfa0
[ 154.391060] ? lock_acquire+0x266/0x3d0
[ 154.391060] ? ext4_filemap_fault+0x27/0x40
[ 154.391060] ? ext4_filemap_fault+0x27/0x40
[ 154.391061] ext4_filemap_fault+0x2f/0x40
[ 154.391061] __do_fault+0x30/0xd0
[ 154.391061] __handle_mm_fault+0xf0d/0x12c0
[ 154.391061] handle_mm_fault+0x8d/0x240
[ 154.391062] do_user_addr_fault+0x21d/0x730
[ 154.391062] ? asm_exc_page_fault+0x8/0x30
[ 154.391062] exc_page_fault+0x6e/0x270
[ 154.391062] ? asm_exc_page_fault+0x8/0x30
[ 154.391062] asm_exc_page_fault+0x1e/0x30
[ 154.391063] RIP: 0033:0x555bc1610b55
[ 154.391063] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.391064] RSP: 002b:00007f0456d95d60 EFLAGS: 00010202
[ 154.391064] RAX: 0000000000000000 RBX: 0000000000000380 RCX: 0000000000000001
[ 154.391064] RDX: 0000000000000000 RSI: 0000000000000007 RDI: 0000000000000007
[ 154.391065] RBP: 0000000000000007 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.391065] R10: 00007f0456d95d60 R11: 000000000000ba1e R12: 0000000000000001
[ 154.391065] R13: 0000000000000380 R14: 00007f0456d95d90 R15: 0000000000000000
[ 154.391065] NMI backtrace for cpu 72
[ 154.391066] CPU: 72 PID: 6964 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.391066] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.391066] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.391067] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.391067] RSP: 0000:ffffc9001840f988 EFLAGS: 00000046
[ 154.391068] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001240000
[ 154.391068] RDX: ffff88d01d3ff1c0 RSI: 0000000000000084 RDI: ffff88c0293fe380
[ 154.391068] RBP: ffff88c0293fe380 R08: 0000000001240000 R09: 00000000001fd9c0
[ 154.391069] R10: ffffc9001840f938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.391069] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.391069] FS: 00007f0456d96700(0000) GS:ffff88d01d200000(0000) knlGS:0000000000000000
[ 154.391069] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.391070] CR2: 0000555bc18acb48 CR3: 000000011a18a003 CR4: 00000000007706e0
[ 154.391070] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.391070] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.391071] PKRU: 55555554
[ 154.391071] Call Trace:
[ 154.391071] do_raw_spin_lock+0xb1/0xc0
[ 154.391071] _raw_spin_lock_nested+0x50/0x60
[ 154.391071] raw_spin_rq_lock_nested+0x51/0x80
[ 154.391072] load_balance+0x2f7/0xd60
[ 154.391072] ? update_blocked_averages+0x500/0x820
[ 154.391072] newidle_balance+0x3d4/0x670
[ 154.391072] balance_fair+0x1b/0x30
[ 154.391073] put_prev_task_balance+0x4b/0x80
[ 154.391073] __schedule+0x57b/0x14d0
[ 154.391073] schedule+0x51/0xc0
[ 154.391073] io_schedule+0x12/0x40
[ 154.391073] __lock_page_killable+0x167/0x250
[ 154.391074] ? file_fdatawait_range+0x20/0x20
[ 154.391074] filemap_fault+0x9b5/0xfa0
[ 154.391074] ? lock_acquire+0x266/0x3d0
[ 154.391074] ? ext4_filemap_fault+0x27/0x40
[ 154.391075] ? ext4_filemap_fault+0x27/0x40
[ 154.391075] ext4_filemap_fault+0x2f/0x40
[ 154.391075] __do_fault+0x30/0xd0
[ 154.391075] __handle_mm_fault+0xf0d/0x12c0
[ 154.391075] handle_mm_fault+0x8d/0x240
[ 154.391076] do_user_addr_fault+0x21d/0x730
[ 154.391076] ? asm_exc_page_fault+0x8/0x30
[ 154.391076] exc_page_fault+0x6e/0x270
[ 154.391076] ? asm_exc_page_fault+0x8/0x30
[ 154.391077] asm_exc_page_fault+0x1e/0x30
[ 154.391077] RIP: 0033:0x555bc1610b55
[ 154.391077] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.391078] RSP: 002b:00007f0456d95d60 EFLAGS: 00010202
[ 154.391078] RAX: 0000000000000000 RBX: 0000000000000380 RCX: 0000000000000001
[ 154.391078] RDX: 0000000000000000 RSI: 0000000000000007 RDI: 0000000000000007
[ 154.391079] RBP: 0000000000000007 R08: 0000000000000000 R09: 00007fff06be1080
[ 154.391079] R10: 00007f0456d95d60 R11: 000000000000ba1e R12: 0000000000000001
[ 154.391079] R13: 0000000000000380 R14: 00007f0456d95d90 R15: 0000000000000000
[ 154.391080] NMI watchdog: Watchdog detected hard LOCKUP on cpu 73
[ 154.391080] Modules linked in: binfmt_misc intel_rapl_msr intel_rapl_common skx_edac nfit x86_pkg_temp_thermal intel_powerclamp ipmi_ssif input_leds kvm_intel kvm irqbypass efi_pstore mei_me mei lpc_ich intel_pch_thermal ioatdma acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel iscsi_tcp libiscsi_tcp libiscsi scsi_transport_iscsi ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath linear ixgbe ast drm_vram_helper drm_ttm_helper crct10dif_pclmul crc32_pclmul ttm ghash_clmulni_intel drm_kms_helper aesni_intel syscopyarea igb sysfillrect crypto_simd sysimgblt dca cryptd fb_sys_fops ptp ahci pps_core mdio drm i2c_algo_bit libahci wmi hid_generic usbhid hid [ 154.391095] irq event stamp: 81922
[ 154.391095] hardirqs last enabled at (81921): [<ffffffff81b05ab4>] _raw_spin_unlock_irq+0x24/0x40
[ 154.391095] hardirqs last disabled at (81922): [<ffffffff81afcf5a>] __schedule+0xd7a/0x14d0
[ 154.391096] softirqs last enabled at (81914): [<ffffffff81e0030f>] __do_softirq+0x30f/0x432
[ 154.391096] softirqs last disabled at (81909): [<ffffffff8109f53c>] irq_exit_rcu+0xbc/0xc0
[ 154.391097] CPU: 73 PID: 7045 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.391097] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.391097] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.391098] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.391098] RSP: 0000:ffffc90018b2f988 EFLAGS: 00000046
[ 154.391098] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001280000
[ 154.391099] RDX: ffff88d01d7ff1c0 RSI: 000000000000007f RDI: ffff88c0293fe380
[ 154.391099] RBP: ffff88c0293fe380 R08: 0000000001280000 R09: 00000000001fd9c0
[ 154.391099] R10: ffffc90018b2f938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.391100] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.391100] FS: 00007f045444d700(0000) GS:ffff88d01d600000(0000) knlGS:0000000000000000
[ 154.391100] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.391101] CR2: 0000555bc18acb48 CR3: 000000011a18a006 CR4: 00000000007706e0
[ 154.391101] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.391101] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.391102] PKRU: 55555554
[ 154.391102] Call Trace:
[ 154.391102] do_raw_spin_lock+0xb1/0xc0
[ 154.391102] _raw_spin_lock_nested+0x50/0x60
[ 154.391102] raw_spin_rq_lock_nested+0x51/0x80
[ 154.391103] load_balance+0x2f7/0xd60
[ 154.391103] ? update_blocked_averages+0x500/0x820
[ 154.391103] newidle_balance+0x3d4/0x670
[ 154.391103] balance_fair+0x1b/0x30
[ 154.391104] put_prev_task_balance+0x4b/0x80
[ 154.391104] __schedule+0x57b/0x14d0
[ 154.391104] schedule+0x51/0xc0
[ 154.391104] io_schedule+0x12/0x40
[ 154.391104] __lock_page_killable+0x167/0x250
[ 154.391105] ? file_fdatawait_range+0x20/0x20
[ 154.391105] filemap_fault+0x9b5/0xfa0
[ 154.391105] ? lock_acquire+0x266/0x3d0
[ 154.391106] ? ext4_filemap_fault+0x27/0x40
[ 154.391106] ? ext4_filemap_fault+0x27/0x40
[ 154.391106] ext4_filemap_fault+0x2f/0x40
[ 154.391106] __do_fault+0x30/0xd0
[ 154.391106] __handle_mm_fault+0xf0d/0x12c0
[ 154.391107] handle_mm_fault+0x8d/0x240
[ 154.391107] do_user_addr_fault+0x21d/0x730
[ 154.391107] ? asm_exc_page_fault+0x8/0x30
[ 154.391107] exc_page_fault+0x6e/0x270
[ 154.391108] ? asm_exc_page_fault+0x8/0x30
[ 154.391108] asm_exc_page_fault+0x1e/0x30
[ 154.391108] RIP: 0033:0x555bc1610b55
[ 154.391108] Code: ff ff c7 43 48 00 00 00 00 5b c3 66 2e 0f 1f 84 00 00 00 00 00 55 48 63 ef b9 01 00 00 00 53 48 89 eb 48 83 ec 08 48 c1 e3 07 <48> 03 1d ec bf 29 00 4c 8d 5b 48 89 c8 41 87 03 83 f8 01 75 16 66
[ 154.391109] RSP: 002b:00007f045444cd60 EFLAGS: 00010206
[ 154.391109] RAX: 0000000000000000 RBX: 0000000000002600 RCX: 0000000000000001
[ 154.391110] RDX: 0000000000000000 RSI: 000000000000004c RDI: 000000000000004c
[ 154.391110] RBP: 000000000000004c R08: 0000000000000000 R09: 00007fff06be1080
[ 154.391110] R10: 00007f045444cd60 R11: 000000000000ba24 R12: 0000000000000001
[ 154.391110] R13: 0000000000002600 R14: 00007f045444cd90 R15: 0000000000000000
[ 154.391111] NMI backtrace for cpu 73
[ 154.391111] CPU: 73 PID: 7045 Comm: sysbench Kdump: loaded Not tainted 5.12.0-rc8-coresched+ #3
[ 154.391111] Hardware name: Intel Corporation CooperCity/CooperCity, BIOS WLYDCRB1.SYS.0016.P38.2006170234 06/17/2020
[ 154.391112] RIP: 0010:native_queued_spin_lock_slowpath+0x191/0x1e0
[ 154.391112] Code: c1 ee 12 83 e0 03 83 ee 01 48 c1 e0 05 48 63 f6 48 05 c0 f1 1f 00 48 03 04 f5 20 e9 79 82 48 89 10 8b 42 08 85 c0 75 09 f3 90 <8b> 42 08 85 c0 74 f7 48 8b 32 48 85 f6 74 94 0f 0d 0e eb 8f 8b 07
[ 154.391113] RSP: 0000:ffffc90018b2f988 EFLAGS: 00000046
[ 154.391113] RAX: 0000000000000000 RBX: ffff88c0293fe380 RCX: 0000000001280000
[ 154.391113] RDX: ffff88d01d7ff1c0 RSI: 000000000000007f RDI: ffff88c0293fe380
[ 154.391114] RBP: ffff88c0293fe380 R08: 0000000001280000 R09: 00000000001fd9c0
[ 154.391114] R10: ffffc90018b2f938 R11: 000000001e2f75eb R12: ffff88c0293fe380
[ 154.391114] R13: 0000000000000001 R14: 0000000000000000 R15: ffff88c0293fe380
[ 154.391115] FS: 00007f045444d700(0000) GS:ffff88d01d600000(0000) knlGS:0000000000000000
[ 154.391115] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 154.391115] CR2: 0000555bc18acb48 CR3: 000000011a18a006 CR4: 00000000007706e0
[ 154.391115] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 154.391116] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 154.391116] PKRU: 55555554
[ 154.391116] Call Trace:
[ 154.391116] do_raw_spin_lock+0xb1/0xc0
[ 154.391117] _raw_spin_lock_nested+0x50/0x60
[ 154.391117] raw_spin_rq_lock_nested+0x51/0x80
[ 154.391117] load_balance+0x2f7/0xd60
[ 154.391117] ? update_blocked_averages+0x500/0x820
[ 154.391118] newidle_balance+0x3d4/0x670
[ 154.391118] balance_fair+0x1b/0x30
[ 154.391118] put_prev_task_balance+0x4b/0x80
[ 154.391118] __schedule+0x57b/0x14d0
[ 154.391119] schedule+0x51/0xc0
[ 154.391119] io_schedule+0x12/0x40
[ 154.391119] __lock_page_killable+0x167/0x250
[ 154.391119] ? file_fdatawait_range+0x20/0x20
[ 154.391120] filemap_fault+0x9b5/0xfa0
[ 154.391120] ? lock_acquire+0x266/0x3d0
[ 154.391120] ? ext4_filemap_fault+0x27/0x40

#15 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 28 lines · Aubrey Li
Quoted 17 lines · Josh Don
Quoted 8 lines · Don Hiatt

Hi Josh and Peter,

I've been running into soft lookups and hard lockups when running a script that just cycles setting the cookie of a group of processes over and over again.

Unfortunately the only way I can reproduce this is by setting the cookies on qemu. I've tried sysbench, stress-ng but those seem to work just fine.

Quoted 6 lines

I'm running Peter's branch and even tried the suggested changes here but still see the same behavior. I enabled panic on hard lockup and here below is a snippet of the log.

Is there anything you'd like me to try or have any debugging you'd like me to do? I'd certainly like to get to the bottom of this.

Quoted 9 lines

Hi Don,

I tried to repro using qemu, but did not generate a lockup. Could you provide more details on what your script is doing (or better yet, share the script directly)? I would have expected you to potentially hit a lockup if you were cycling sched_core being enabled and disabled, but it sounds like you are just recreating the cookie for a process group over and over?

Quoted 5 lines

I saw something similar on a bare metal hardware. Also tried the suggested patch here and no luck. Panic stack attached with softlockup_all_cpu_backtrace=1.
(sorry, my system has 192 cpus and somehow putting 184 cpus offline causes system hang without any message...)

Can you please try the following change to see if the problem is gone on your side?

Thanks,
-Aubrey

kernel/sched/core.c +11 -8
index f732642e3e09..1ef13b50dfcd 100644
@@ -493,14 +493,17 @@ void double_rq_lock(struct rq *rq1, struct rq *rq2)
 {
 	lockdep_assert_irqs_disabled();
 
-	if (rq1->cpu > rq2->cpu)
-		swap(rq1, rq2);
-
-	raw_spin_rq_lock(rq1);
-	if (__rq_lockp(rq1) == __rq_lockp(rq2))
-		return;
-
-	raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+	if (__rq_lockp(rq1) == __rq_lockp(rq2)) {
+		raw_spin_rq_lock(rq1);
+	} else {
+		if (__rq_lockp(rq1) < __rq_lockp(rq2)) {
+			raw_spin_rq_lock(rq1);
+			raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+		} else {
+			raw_spin_rq_lock(rq2);
+			raw_spin_rq_lock_nested(rq1, SINGLE_DEPTH_NESTING);
+		}
+	}
 }
 #endif
 
#16 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 30 lines · Aubrey Li
Quoted 28 lines · Aubrey Li
Quoted 26 lines · Josh Don
Quoted 8 lines · Don Hiatt

Hi Josh and Peter,

I've been running into soft lookups and hard lockups when running a script that just cycles setting the cookie of a group of processes over and over again.

Unfortunately the only way I can reproduce this is by setting the cookies on qemu. I've tried sysbench, stress-ng but those seem to work just fine.

Quoted 6 lines

I'm running Peter's branch and even tried the suggested changes here but still see the same behavior. I enabled panic on hard lockup and here below is a snippet of the log.

Is there anything you'd like me to try or have any debugging you'd like me to do? I'd certainly like to get to the bottom of this.

Hi Don,

I tried to repro using qemu, but did not generate a lockup. Could you provide more details on what your script is doing (or better yet, share the script directly)? I would have expected you to potentially hit a lockup if you were cycling sched_core being enabled and disabled, but it sounds like you are just recreating the cookie for a process group over and over?

Quoted 8 lines
Quoted 5 lines

I saw something similar on a bare metal hardware. Also tried the suggested patch here and no luck. Panic stack attached with softlockup_all_cpu_backtrace=1.
(sorry, my system has 192 cpus and somehow putting 184 cpus offline causes system hang without any message...)

Can you please try the following change to see if the problem is gone on your side?

Please ignore this patch, as the change of double_rq_unlock() in Peter's last patch
fixed the problem.

Thanks,
-Aubrey

Quoted 31 lines
kernel/sched/core.c +11 -8
index f732642e3e09..1ef13b50dfcd 100644
@@ -493,14 +493,17 @@ void double_rq_lock(struct rq *rq1, struct rq *rq2)
 {
        lockdep_assert_irqs_disabled();
 
-       if (rq1->cpu > rq2->cpu)
-               swap(rq1, rq2);
-
-       raw_spin_rq_lock(rq1);
-       if (__rq_lockp(rq1) == __rq_lockp(rq2))
-               return;
-
-       raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+       if (__rq_lockp(rq1) == __rq_lockp(rq2)) {
+               raw_spin_rq_lock(rq1);
+       } else {
+               if (__rq_lockp(rq1) < __rq_lockp(rq2)) {
+                       raw_spin_rq_lock(rq1);
+                       raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+               } else {
+                       raw_spin_rq_lock(rq2);
+                       raw_spin_rq_lock_nested(rq1, SINGLE_DEPTH_NESTING);
+               }
+       }
 }
 #endif
 
#17 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 41 lines · Aubrey Li
Quoted 30 lines · Aubrey Li
Quoted 28 lines · Aubrey Li
Quoted 26 lines · Josh Don
Quoted 8 lines · Don Hiatt

Hi Josh and Peter,

I've been running into soft lookups and hard lockups when running a script that just cycles setting the cookie of a group of processes over and over again.

Unfortunately the only way I can reproduce this is by setting the cookies on qemu. I've tried sysbench, stress-ng but those seem to work just fine.

Quoted 6 lines

I'm running Peter's branch and even tried the suggested changes here but still see the same behavior. I enabled panic on hard lockup and here below is a snippet of the log.

Is there anything you'd like me to try or have any debugging you'd like me to do? I'd certainly like to get to the bottom of this.

Hi Don,

I tried to repro using qemu, but did not generate a lockup. Could you provide more details on what your script is doing (or better yet, share the script directly)? I would have expected you to potentially hit a lockup if you were cycling sched_core being enabled and disabled, but it sounds like you are just recreating the cookie for a process group over and over?

Quoted 8 lines
Quoted 5 lines

I saw something similar on a bare metal hardware. Also tried the suggested patch here and no luck. Panic stack attached with softlockup_all_cpu_backtrace=1.
(sorry, my system has 192 cpus and somehow putting 184 cpus offline causes system hang without any message...)

Can you please try the following change to see if the problem is gone on your side?

Quoted 7 lines

Please ignore this patch, as the change of double_rq_unlock() in Peter's last patch
fixed the problem.

Thanks,
-Aubrey

I'm still seeing hard lockups while repeatedly setting cookies on qemu processes even with
the updated patch. If there is any debug you'd like me to turn on, just let me know.

Thanks!

don

Quoted 31 lines
Quoted 31 lines
kernel/sched/core.c +11 -8
index f732642e3e09..1ef13b50dfcd 100644
@@ -493,14 +493,17 @@ void double_rq_lock(struct rq *rq1, struct rq *rq2)
 {
        lockdep_assert_irqs_disabled();
 
-       if (rq1->cpu > rq2->cpu)
-               swap(rq1, rq2);
-
-       raw_spin_rq_lock(rq1);
-       if (__rq_lockp(rq1) == __rq_lockp(rq2))
-               return;
-
-       raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+       if (__rq_lockp(rq1) == __rq_lockp(rq2)) {
+               raw_spin_rq_lock(rq1);
+       } else {
+               if (__rq_lockp(rq1) < __rq_lockp(rq2)) {
+                       raw_spin_rq_lock(rq1);
+                       raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+               } else {
+                       raw_spin_rq_lock(rq2);
+                       raw_spin_rq_lock_nested(rq1, SINGLE_DEPTH_NESTING);
+               }
+       }
 }
 #endif
 
#18 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 9 lines · Don Hiatt

I'm still seeing hard lockups while repeatedly setting cookies on qemu processes even with
the updated patch. If there is any debug you'd like me to turn on, just let me know.

Thanks!

don

Thanks for the added context on your repro configuration. In addition to the updated patch from earlier, could you try the modification to double_rq_lock() from
https://lkml.kernel.org/r/CABk29NuS-B3n4sbmavo0NDA1OCCsz6Zf2VDjjFQvAxBMQoJ_Lg@mail.gmail.com
? I have a feeling this is what's causing your lockup.

Best,
Josh

#19 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 12 lines · Josh Don
Quoted 9 lines · Don Hiatt

I'm still seeing hard lockups while repeatedly setting cookies on qemu processes even with
the updated patch. If there is any debug you'd like me to turn on, just let me know.

Thanks!

don

Quoted 8 lines

Thanks for the added context on your repro configuration. In addition to the updated patch from earlier, could you try the modification to double_rq_lock() from
https://lkml.kernel.org/r/CABk29NuS-B3n4sbmavo0NDA1OCCsz6Zf2VDjjFQvAxBMQoJ_Lg@mail.gmail.com
? I have a feeling this is what's causing your lockup.

Best,
Josh

Hi Josh,

I've been running Aubrey+Peter's patch (attached) for almost 5 hours and haven't had a single issue. :)

I'm running a set-cookie script every 5 seconds on the two VMs (each vm is running
'sysbench --threads=1 --time=0 cpu run' to generate some load in the vm) and I'm running two of the same sysbench runs on the HV while setting cookies every 5 seconds.

Unless I jinxed us it looks like a great fix. :)

Let me know if there is anything else you'd like me to try. I'm going to leave the tests running
and see what happens. I update with what I find.

Thanks!

don

kernel/sched/core.c +33 -12
index f732642e3e09..1d7bc87007cd 100644
@@ -290,6 +290,10 @@ static void sched_core_assert_empty(void)
 static void __sched_core_enable(void)
 {
 	static_branch_enable(&__sched_core_enabled);
+	/*
+	 * Ensure raw_spin_rq_*lock*() have completed before flipping.
+	 */
+	synchronize_rcu();
 	__sched_core_flip(true);
 	sched_core_assert_empty();
 }
@@ -449,16 +453,23 @@ void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
 	raw_spinlock_t *lock;
 
+	/* Matches synchronize_sched() in __sched_core_enabled() */
+	preempt_disable();
 	if (sched_core_disabled()) {
 		raw_spin_lock_nested(&rq->__lock, subclass);
+		/* preempt-count *MUST* be > 1 */
+		preempt_enable_no_resched();
 		return;
 	}
 
 	for (;;) {
 		lock = __rq_lockp(rq);
 		raw_spin_lock_nested(lock, subclass);
-		if (likely(lock == __rq_lockp(rq)))
+		if (likely(lock == __rq_lockp(rq))) {
+			/* preempt-count *MUST* be > 1 */
+			preempt_enable_no_resched();
 			return;
+		}
 		raw_spin_unlock(lock);
 	}
 }
@@ -468,14 +479,21 @@ bool raw_spin_rq_trylock(struct rq *rq)
 	raw_spinlock_t *lock;
 	bool ret;
 
-	if (sched_core_disabled())
-		return raw_spin_trylock(&rq->__lock);
+	/* Matches synchronize_sched() in __sched_core_enabled() */
+	preempt_disable();
+	if (sched_core_disabled()) {
+		ret = raw_spin_trylock(&rq->__lock);
+		preempt_enable();
+		return ret;
+	}
 
 	for (;;) {
 		lock = __rq_lockp(rq);
 		ret = raw_spin_trylock(lock);
-		if (!ret || (likely(lock == __rq_lockp(rq))))
+		if (!ret || (likely(lock == __rq_lockp(rq)))) {
+			preempt_enable();
 			return ret;
+		}
 		raw_spin_unlock(lock);
 	}
 }
@@ -493,14 +511,17 @@ void double_rq_lock(struct rq *rq1, struct rq *rq2)
 {
 	lockdep_assert_irqs_disabled();
 
-	if (rq1->cpu > rq2->cpu)
-		swap(rq1, rq2);
-
-	raw_spin_rq_lock(rq1);
-	if (__rq_lockp(rq1) == __rq_lockp(rq2))
-		return;
-
-	raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+	if (__rq_lockp(rq1) == __rq_lockp(rq2)) {
+		raw_spin_rq_lock(rq1);
+	} else {
+		if (__rq_lockp(rq1) < __rq_lockp(rq2)) {
+			raw_spin_rq_lock(rq1);
+			raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+		} else {
+			raw_spin_rq_lock(rq2);
+			raw_spin_rq_lock_nested(rq1, SINGLE_DEPTH_NESTING);
+		}
+	}
 }
 #endif
#20 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 36 lines · Don Hiatt
Quoted 12 lines · Josh Don
Quoted 9 lines · Don Hiatt

I'm still seeing hard lockups while repeatedly setting cookies on qemu processes even with
the updated patch. If there is any debug you'd like me to turn on, just let me know.

Thanks!

don

Quoted 8 lines

Thanks for the added context on your repro configuration. In addition to the updated patch from earlier, could you try the modification to double_rq_lock() from
https://lkml.kernel.org/r/CABk29NuS-B3n4sbmavo0NDA1OCCsz6Zf2VDjjFQvAxBMQoJ_Lg@mail.gmail.com
? I have a feeling this is what's causing your lockup.

Best,
Josh

Hi Josh,

I've been running Aubrey+Peter's patch (attached) for almost 5 hours and haven't had a single issue. :)

I'm running a set-cookie script every 5 seconds on the two VMs (each vm is running
'sysbench --threads=1 --time=0 cpu run' to generate some load in the vm) and I'm running two of the same sysbench runs on the HV while setting cookies every 5 seconds.

Unless I jinxed us it looks like a great fix. :)

Quoted 7 lines

Let me know if there is anything else you'd like me to try. I'm going to leave the tests running
and see what happens. I update with what I find.

Thanks!

don

That's awesome news, thanks for validating. Note that with Aubrey's patch there is still a race window if sched core is being enabled/disabled (ie. if you alternate between there being some cookies in the system, and no cookies). In my reply I posted an alternative version to avoid that. If your script were to do the on-off flipping with the old patch, you'd might eventually see another lockup.

Tested-by
#21 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 46 lines · Josh Don
Quoted 36 lines · Don Hiatt
Quoted 12 lines · Josh Don
Quoted 9 lines · Don Hiatt

I'm still seeing hard lockups while repeatedly setting cookies on qemu processes even with
the updated patch. If there is any debug you'd like me to turn on, just let me know.

Thanks!

don

Quoted 8 lines

Thanks for the added context on your repro configuration. In addition to the updated patch from earlier, could you try the modification to double_rq_lock() from
https://lkml.kernel.org/r/CABk29NuS-B3n4sbmavo0NDA1OCCsz6Zf2VDjjFQvAxBMQoJ_Lg@mail.gmail.com
? I have a feeling this is what's causing your lockup.

Best,
Josh

Hi Josh,

I've been running Aubrey+Peter's patch (attached) for almost 5 hours and haven't had a single issue. :)

I'm running a set-cookie script every 5 seconds on the two VMs (each vm is running
'sysbench --threads=1 --time=0 cpu run' to generate some load in the vm) and I'm running two of the same sysbench runs on the HV while setting cookies every 5 seconds.

Unless I jinxed us it looks like a great fix. :)

Quoted 7 lines

Let me know if there is anything else you'd like me to try. I'm going to leave the tests running
and see what happens. I update with what I find.

Thanks!

don

Quoted 7 lines

That's awesome news, thanks for validating. Note that with Aubrey's patch there is still a race window if sched core is being enabled/disabled (ie. if you alternate between there being some cookies in the system, and no cookies). In my reply I posted an alternative version to avoid that. If your script were to do the on-off flipping with the old patch, you'd might eventually see another lockup.

My tests have been running 24 hours now and all is good. I'll do another test with your changes next as well as continue to test the core-sched queue.

This and all the other patches in Peter's repo:

Have a great day and thanks again.

don

#22 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

In reply to Don Hiatt
Quoted 34 lines · Don Hiatt
Quoted 12 lines · Josh Don
Quoted 9 lines · Don Hiatt

I'm still seeing hard lockups while repeatedly setting cookies on qemu processes even with
the updated patch. If there is any debug you'd like me to turn on, just let me know.

Thanks!

don

Quoted 8 lines

Thanks for the added context on your repro configuration. In addition to the updated patch from earlier, could you try the modification to double_rq_lock() from
https://lkml.kernel.org/r/CABk29NuS-B3n4sbmavo0NDA1OCCsz6Zf2VDjjFQvAxBMQoJ_Lg@mail.gmail.com
? I have a feeling this is what's causing your lockup.

Best,
Josh

Hi Josh,

I've been running Aubrey+Peter's patch (attached) for almost 5 hours and haven't had a single issue. :)

I'm running a set-cookie script every 5 seconds on the two VMs (each vm is running
'sysbench --threads=1 --time=0 cpu run' to generate some load in the vm) and I'm running two of the same sysbench runs on the HV while setting cookies every 5 seconds.

Quoted 6 lines

Unless I jinxed us it looks like a great fix. :)

Let me know if there is anything else you'd like me to try. I'm going to leave the tests running
and see what happens. I update with what I find.

This sounds great, Thanks Don.

However, I guess we have the same problem in _double_lock_balance(), which needs to be fixed as well.

I or Josh can make a better patch for this after Peter's comments.

Thanks,
-Aubrey

#23 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

In reply to Josh Don
Quoted 26 lines · Josh Don
Quoted 8 lines · Don Hiatt

Hi Josh and Peter,

I've been running into soft lookups and hard lockups when running a script that just cycles setting the cookie of a group of processes over and over again.

Unfortunately the only way I can reproduce this is by setting the cookies on qemu. I've tried sysbench, stress-ng but those seem to work just fine.

Quoted 6 lines

I'm running Peter's branch and even tried the suggested changes here but still see the same behavior. I enabled panic on hard lockup and here below is a snippet of the log.

Is there anything you'd like me to try or have any debugging you'd like me to do? I'd certainly like to get to the bottom of this.

Hi Don,

I tried to repro using qemu, but did not generate a lockup. Could you provide more details on what your script is doing (or better yet, share the script directly)? I would have expected you to potentially hit a lockup if you were cycling sched_core being enabled and disabled, but it sounds like you are just recreating the cookie for a process group over and over?

Quoted 2 lines

Best,
Josh

Hi Josh,

Sorry if I wasn't clear, but I'm running on bare metal (Peter's 5.12-rc8 repo) and have two qemu-system-x86_64 vms (1 with 8vcpu, the other with 24 but it doesn't really matter). I then run a script [1] that cycles setting the cookie for all the processes given the pid of qemu-system-x86_64.

I also just found a lockup when I set the cookies for those two vms, pinned them both to the same processor pair, and ran sysbench within each vm to generate some load [2]. This is without cycling the cookies, just setting once.

Thanks!

Don

----[1]---
This is a little test harness (I can provide the source but it is based on your kselftests)

dhiatt@s2r5node34:~/do_coresched$ ./do_coresched -h
Usage for ./do_coresched
   -c Create sched cookie for <pid>
   -f Share sched cookie from <pid>
   -g Get sched cookie from <pid>
   -t Share sched cookie to <pid>
   -z Clear sched cookie from <pid>
   -p PID

Create: _prctl(PR_SCHED_CORE, PR_SCHED_CORE_CREATE, pid, PIDTYPE_PID, 0)
Share: _prctl(PR_SCHED_CORE, PR_SCHED_CORE_SHARE_TO, pid, PIDTYPE_PID, 0)
Get: prctl(PR_SCHED_CORE, PR_SCHED_CORE_GET, pid, PIDTYPE_PID,

                    (unsigned long)&cookie);

--

dhiatt@s2r5node34:~/do_coresched$ cat set.sh
#!/bin/bash
# usage: set.sh pid

target=$1
echo target pid: $target
pids=ps -eL | grep $target | awk '{print $2}' | sort -g

echo "Setting cookies for $target"
./do_coresched -c -p $BASHPID. #
for i in $pids
do
./do_coresched -t -p $i
./do_coresched -g -p $i
done
---

---[2]---

[ 8911.926989] watchdog: BUG: soft lockup - CPU#53 stuck for 22s! [CPU
1/KVM:19539]
[ 8911.935727] NMI watchdog: Watchdog detected hard LOCKUP on cpu 6
[ 8911.935791] NMI watchdog: Watchdog detected hard LOCKUP on cpu 7
[ 8911.935908] NMI watchdog: Watchdog detected hard LOCKUP on cpu 12
[ 8911.935967] NMI watchdog: Watchdog detected hard LOCKUP on cpu 13
[ 8911.936070] NMI watchdog: Watchdog detected hard LOCKUP on cpu 19
[ 8911.936145] NMI watchdog: Watchdog detected hard LOCKUP on cpu 21
[ 8911.936220] NMI watchdog: Watchdog detected hard LOCKUP on cpu 23
[ 8911.936361] NMI watchdog: Watchdog detected hard LOCKUP on cpu 31
[ 8911.936453] NMI watchdog: Watchdog detected hard LOCKUP on cpu 34
[ 8911.936567] NMI watchdog: Watchdog detected hard LOCKUP on cpu 42
[ 8911.936627] NMI watchdog: Watchdog detected hard LOCKUP on cpu 46
[ 8911.936712] NMI watchdog: Watchdog detected hard LOCKUP on cpu 49
[ 8911.936827] NMI watchdog: Watchdog detected hard LOCKUP on cpu 58
[ 8911.936887] NMI watchdog: Watchdog detected hard LOCKUP on cpu 60
[ 8911.936969] NMI watchdog: Watchdog detected hard LOCKUP on cpu 70
[ 8915.926847] rcu: INFO: rcu_sched detected stalls on CPUs/tasks:
[ 8915.932904] rcu: 2-...!: (7 ticks this GP)
idle=38e/1/0x4000000000000000 softirq=181538/181540 fqs=1178
[ 8915.942617] rcu: 14-...!: (2 GPs behind)
idle=d0e/1/0x4000000000000000 softirq=44825/44825 fqs=1178
[ 8915.954034] rcu: rcu_sched kthread timer wakeup didn't happen for
12568 jiffies! g462469 f0x0 RCU_GP_WAIT_FQS(5) ->state=0x402
[ 8915.965775] rcu: Possible timer handling issue on cpu=6 timer-softirq=10747
[ 8915.973021] rcu: rcu_sched kthread starved for 12572 jiffies!
g462469 f0x0 RCU_GP_WAIT_FQS(5) ->state=0x402 ->cpu=6
[ 8915.983681] rcu: Unless rcu_sched kthread gets sufficient CPU time,
OOM is now expected behavior.
[ 8915.992879] rcu: RCU grace-period kthread stack dump:
[ 8915.998211] rcu: Stack dump where RCU GP kthread last ran:
[ 8939.925995] watchdog: BUG: soft lockup - CPU#53 stuck for 22s! [CPU
1/KVM:19539]
[ 8939.935274] NMI watchdog: Watchdog detected hard LOCKUP on cpu 25
[ 8939.935351] NMI watchdog: Watchdog detected hard LOCKUP on cpu 27
[ 8939.935425] NMI watchdog: Watchdog detected hard LOCKUP on cpu 29
[ 8939.935653] NMI watchdog: Watchdog detected hard LOCKUP on cpu 44
[ 8939.935845] NMI watchdog: Watchdog detected hard LOCKUP on cpu 63
[ 8963.997140] watchdog: BUG: soft lockup - CPU#71 stuck for 22s!

[SchedulerRunner:4405]

-----

#24 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

In reply to Josh Don
Quoted 35 lines · Josh Don

kernel/sched/core.c static void sched_core_assert_empty(void) · line 296

Quoted 13 lines
kernel/sched/core.c +4 -0
index f732642e3e09..1a81e9cc9e5d 100644
@@ -290,6 +290,10 @@ static void sched_core_assert_empty(void)
 static void __sched_core_enable(void)
 {
        static_branch_enable(&__sched_core_enabled);
+       /*
+        * Ensure raw_spin_rq_*lock*() have completed before flipping.
+        */
+       synchronize_sched();

synchronize_rcu()

Quoted 4 lines
kernel/sched/core.c +0 -0
        __sched_core_flip(true);
        sched_core_assert_empty();
 }

kernel/sched/core.c void raw_spin_rq_lock_nested(struct rq *rq, int subclass) · line 462

Quoted 10 lines
kernel/sched/core.c +3 -0
@@ -449,16 +453,22 @@ void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
        raw_spinlock_t *lock;
 
+       preempt_disable();
        if (sched_core_disabled()) {
                raw_spin_lock_nested(&rq->__lock, subclass);
+               /* preempt *MUST* still be disabled here */
+               preempt_enable_no_resched();
                return;
        }

This approach looks good to me. I'm guessing you went this route instead of doing the re-check after locking in order to optimize the disabled case?

Quoted 16 lines

Recommend a comment that the preempt_disable() here pairs with the synchronize_rcu() in __sched_core_enable().

Quoted 13 lines
kernel/sched/core.c +4 -1
        for (;;) {
                lock = __rq_lockp(rq);
                raw_spin_lock_nested(lock, subclass);
-               if (likely(lock == __rq_lockp(rq)))
+               if (likely(lock == __rq_lockp(rq))) {
+                       /* preempt *MUST* still be disabled here */
+                       preempt_enable_no_resched();
                        return;
+               }
                raw_spin_unlock(lock);
        }
 }

Also, did you mean to have a preempt_enable_no_resched() rather than prempt_enable() in raw_spin_rq_trylock?

I went over the rq_lockp stuff again after Don's reported lockup. Most uses are safe due to already holding an rq lock. However, double_rq_unlock() is prone to race:

double_rq_unlock(rq1, rq2):
/* Initial state: core sched enabled, and rq1 and rq2 are smt siblings. So, double_rq_lock(rq1, rq2) only took a single rq lock */ raw_spin_rq_unlock(rq1);
/* now not holding any rq lock */
/* sched core disabled. Now __rq_lockp(rq1) != __rq_lockp(rq2), so we falsely unlock rq2 */
if (__rq_lockp(rq1) != __rq_lockp(rq2))

        raw_spin_rq_unlock(rq2);
else
        __release(rq2->lock);

Instead we can cache __rq_lockp(rq1) and __rq_lockp(rq2) before releasing the lock, in order to prevent this. FWIW I think it is likely that Don is seeing a different issue.

author maintainer
#25 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 2 lines · Josh Don

Also, did you mean to have a preempt_enable_no_resched() rather than prempt_enable() in raw_spin_rq_trylock?

No, trylock really needs to be preempt_enable(), because it can have failed, at which point it will not have incremented the preemption count and our decrement can hit 0, at which point we really should reschedule.

Quoted 12 lines

I went over the rq_lockp stuff again after Don's reported lockup. Most uses are safe due to already holding an rq lock. However, double_rq_unlock() is prone to race:

double_rq_unlock(rq1, rq2):
/* Initial state: core sched enabled, and rq1 and rq2 are smt siblings. So, double_rq_lock(rq1, rq2) only took a single rq lock */ raw_spin_rq_unlock(rq1);
/* now not holding any rq lock */
/* sched core disabled. Now __rq_lockp(rq1) != __rq_lockp(rq2), so we falsely unlock rq2 */
if (__rq_lockp(rq1) != __rq_lockp(rq2))

Quoted 7 lines
        raw_spin_rq_unlock(rq2);
else
        __release(rq2->lock);

Instead we can cache __rq_lockp(rq1) and __rq_lockp(rq2) before releasing the lock, in order to prevent this. FWIW I think it is likely that Don is seeing a different issue.

Ah, indeed so.. rq_lockp() could do with an assertion, not sure how to sanely do that. Anyway, double_rq_unlock() is simple enough to fix, we can simply flip the unlock()s.

( I'm suffering a cold and am really quite slow atm )

How's this then?

---

kernel/sched/core.c +22 -4
index f732642e3e09..3a534c0c1c46 100644
@@ -290,6 +290,10 @@ static void sched_core_assert_empty(void)
 static void __sched_core_enable(void)
 {
 	static_branch_enable(&__sched_core_enabled);
+	/*
+	 * Ensure raw_spin_rq_*lock*() have completed before flipping.
+	 */
+	synchronize_sched();
 	__sched_core_flip(true);
 	sched_core_assert_empty();
 }
@@ -449,16 +453,23 @@ void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
 	raw_spinlock_t *lock;
 
+	/* Matches synchronize_sched() in __sched_core_enabled() */
+	preempt_disable();
 	if (sched_core_disabled()) {
 		raw_spin_lock_nested(&rq->__lock, subclass);
+		/* preempt-count *MUST* be > 1 */
+		preempt_enable_no_resched();
 		return;
 	}
 
 	for (;;) {
 		lock = __rq_lockp(rq);
 		raw_spin_lock_nested(lock, subclass);
-		if (likely(lock == __rq_lockp(rq)))
+		if (likely(lock == __rq_lockp(rq))) {
+			/* preempt-count *MUST* be > 1 */
+			preempt_enable_no_resched();
 			return;
+		}
 		raw_spin_unlock(lock);
 	}
 }
@@ -468,14 +479,21 @@ bool raw_spin_rq_trylock(struct rq *rq)
 	raw_spinlock_t *lock;
 	bool ret;
 
-	if (sched_core_disabled())
-		return raw_spin_trylock(&rq->__lock);
+	/* Matches synchronize_sched() in __sched_core_enabled() */
+	preempt_disable();
+	if (sched_core_disabled()) {
+		ret = raw_spin_trylock(&rq->__lock);
+		preempt_enable();
+		return ret;
+	}
 
 	for (;;) {
 		lock = __rq_lockp(rq);
 		ret = raw_spin_trylock(lock);
-		if (!ret || (likely(lock == __rq_lockp(rq))))
+		if (!ret || (likely(lock == __rq_lockp(rq)))) {
+			preempt_enable();
 			return ret;
+		}
 		raw_spin_unlock(lock);
 	}
 }
kernel/sched/sched.h +2 -1
index 6a905fe19eef..c9a52231d58a 100644
@@ -2568,11 +2568,12 @@ static inline void double_rq_unlock(struct rq *rq1, struct rq *rq2)
 	__releases(rq1->lock)
 	__releases(rq2->lock)
 {
-	raw_spin_rq_unlock(rq1);
 	if (__rq_lockp(rq1) != __rq_lockp(rq2))
 		raw_spin_rq_unlock(rq2);
 	else
 		__release(rq2->lock);
+
+	raw_spin_rq_unlock(rq1);
 }
 
 extern void set_rq_online (struct rq *rq);
#26 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 44 lines · Peter Zijlstra
Quoted 2 lines · Josh Don

Also, did you mean to have a preempt_enable_no_resched() rather than prempt_enable() in raw_spin_rq_trylock?

No, trylock really needs to be preempt_enable(), because it can have failed, at which point it will not have incremented the preemption count and our decrement can hit 0, at which point we really should reschedule.

Quoted 12 lines

I went over the rq_lockp stuff again after Don's reported lockup. Most uses are safe due to already holding an rq lock. However, double_rq_unlock() is prone to race:

double_rq_unlock(rq1, rq2):
/* Initial state: core sched enabled, and rq1 and rq2 are smt siblings. So, double_rq_lock(rq1, rq2) only took a single rq lock */ raw_spin_rq_unlock(rq1);
/* now not holding any rq lock */
/* sched core disabled. Now __rq_lockp(rq1) != __rq_lockp(rq2), so we falsely unlock rq2 */
if (__rq_lockp(rq1) != __rq_lockp(rq2))

Quoted 7 lines
        raw_spin_rq_unlock(rq2);
else
        __release(rq2->lock);

Instead we can cache __rq_lockp(rq1) and __rq_lockp(rq2) before releasing the lock, in order to prevent this. FWIW I think it is likely that Don is seeing a different issue.

Ah, indeed so.. rq_lockp() could do with an assertion, not sure how to sanely do that. Anyway, double_rq_unlock() is simple enough to fix, we can simply flip the unlock()s.

( I'm suffering a cold and am really quite slow atm )

How's this then?

---

kernel/sched/core.c +0 -0
index f732642e3e09..3a534c0c1c46 100644

kernel/sched/core.c static void sched_core_assert_empty(void) · line 296

Quoted 7 lines
kernel/sched/core.c +4 -0
@@ -290,6 +290,10 @@ static void sched_core_assert_empty(void)
 static void __sched_core_enable(void)
 {
        static_branch_enable(&__sched_core_enabled);
+       /*
+        * Ensure raw_spin_rq_*lock*() have completed before flipping.
+        */
+       synchronize_sched();

synchronize_sched() seems no longer exist...

Thanks,
-Aubrey

author maintainer
#27 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 10 lines · Aubrey Li
Quoted 8 lines · Peter Zijlstra

Ah, indeed so.. rq_lockp() could do with an assertion, not sure how to sanely do that. Anyway, double_rq_unlock() is simple enough to fix, we can simply flip the unlock()s.

( I'm suffering a cold and am really quite slow atm )

How's this then?

Quoted 15 lines

kernel/sched/core.c static void sched_core_assert_empty(void) · line 296

Quoted 13 lines

---

kernel/sched/core.c +4 -0
index f732642e3e09..3a534c0c1c46 100644
@@ -290,6 +290,10 @@ static void sched_core_assert_empty(void)
 static void __sched_core_enable(void)
 {
        static_branch_enable(&__sched_core_enabled);
+       /*
+        * Ensure raw_spin_rq_*lock*() have completed before flipping.
+        */
+       synchronize_sched();

synchronize_sched() seems no longer exist...

Bah.. Paul, why did that go away? I realize RCU merged in the sched and bh flavours, but I still find it expressive to use sync_sched() vs preempt_disable().

Anyway, just use sync_rcu().

#28 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 27 lines · Peter Zijlstra
Quoted 10 lines · Aubrey Li
Quoted 8 lines · Peter Zijlstra

Ah, indeed so.. rq_lockp() could do with an assertion, not sure how to sanely do that. Anyway, double_rq_unlock() is simple enough to fix, we can simply flip the unlock()s.

( I'm suffering a cold and am really quite slow atm )

How's this then?

Quoted 15 lines

kernel/sched/core.c static void sched_core_assert_empty(void) · line 296

Quoted 13 lines

---

kernel/sched/core.c +4 -0
index f732642e3e09..3a534c0c1c46 100644
@@ -290,6 +290,10 @@ static void sched_core_assert_empty(void)
 static void __sched_core_enable(void)
 {
        static_branch_enable(&__sched_core_enabled);
+       /*
+        * Ensure raw_spin_rq_*lock*() have completed before flipping.
+        */
+       synchronize_sched();

synchronize_sched() seems no longer exist...

Quoted 3 lines

Bah.. Paul, why did that go away? I realize RCU merged in the sched and bh flavours, but I still find it expressive to use sync_sched() vs preempt_disable().

I could have made synchronize_sched() a synonym for synchronize_rcu(), but that would be more likely to mislead than to help.

Quoted 1 line

Anyway, just use sync_rcu().

And yes, just use synchronize_rcu().

Thanx, Paul

Reviewed-by
#29 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

In reply to Peter Zijlstra
Quoted 9 lines · Peter Zijlstra
Quoted 2 lines · Josh Don

Also, did you mean to have a preempt_enable_no_resched() rather than prempt_enable() in raw_spin_rq_trylock?

No, trylock really needs to be preempt_enable(), because it can have failed, at which point it will not have incremented the preemption count and our decrement can hit 0, at which point we really should reschedule.

Ah yes, makes sense. Did you want to use preempt_enable_no_resched() at all then? No chance of preempt_count() being 1 at the point of enable, as you comment below.

Quoted 20 lines
Quoted 12 lines

I went over the rq_lockp stuff again after Don's reported lockup. Most uses are safe due to already holding an rq lock. However, double_rq_unlock() is prone to race:

double_rq_unlock(rq1, rq2):
/* Initial state: core sched enabled, and rq1 and rq2 are smt siblings. So, double_rq_lock(rq1, rq2) only took a single rq lock */ raw_spin_rq_unlock(rq1);
/* now not holding any rq lock */
/* sched core disabled. Now __rq_lockp(rq1) != __rq_lockp(rq2), so we falsely unlock rq2 */
if (__rq_lockp(rq1) != __rq_lockp(rq2))

Quoted 7 lines
        raw_spin_rq_unlock(rq2);
else
        __release(rq2->lock);

Instead we can cache __rq_lockp(rq1) and __rq_lockp(rq2) before releasing the lock, in order to prevent this. FWIW I think it is likely that Don is seeing a different issue.

Quoted 5 lines

Ah, indeed so.. rq_lockp() could do with an assertion, not sure how to sanely do that. Anyway, double_rq_unlock() is simple enough to fix, we can simply flip the unlock()s.

( I'm suffering a cold and am really quite slow atm )

No worries, hope it's a mild one.

Quoted 1 line

How's this then?

Looks good to me (other than the synchronize_sched()->synchronize_rcu()).

For these locking patches,

I'll see if I can repro that lockup.

Quoted 84 lines

---

kernel/sched/core.c +22 -4
index f732642e3e09..3a534c0c1c46 100644
@@ -290,6 +290,10 @@ static void sched_core_assert_empty(void)
 static void __sched_core_enable(void)
 {
        static_branch_enable(&__sched_core_enabled);
+       /*
+        * Ensure raw_spin_rq_*lock*() have completed before flipping.
+        */
+       synchronize_sched();
        __sched_core_flip(true);
        sched_core_assert_empty();
 }
@@ -449,16 +453,23 @@ void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
        raw_spinlock_t *lock;
 
+       /* Matches synchronize_sched() in __sched_core_enabled() */
+       preempt_disable();
        if (sched_core_disabled()) {
                raw_spin_lock_nested(&rq->__lock, subclass);
+               /* preempt-count *MUST* be > 1 */
+               preempt_enable_no_resched();
                return;
        }
 
        for (;;) {
                lock = __rq_lockp(rq);
                raw_spin_lock_nested(lock, subclass);
-               if (likely(lock == __rq_lockp(rq)))
+               if (likely(lock == __rq_lockp(rq))) {
+                       /* preempt-count *MUST* be > 1 */
+                       preempt_enable_no_resched();
                        return;
+               }
                raw_spin_unlock(lock);
        }
 }
@@ -468,14 +479,21 @@ bool raw_spin_rq_trylock(struct rq *rq)
        raw_spinlock_t *lock;
        bool ret;
 
-       if (sched_core_disabled())
-               return raw_spin_trylock(&rq->__lock);
+       /* Matches synchronize_sched() in __sched_core_enabled() */
+       preempt_disable();
+       if (sched_core_disabled()) {
+               ret = raw_spin_trylock(&rq->__lock);
+               preempt_enable();
+               return ret;
+       }
 
        for (;;) {
                lock = __rq_lockp(rq);
                ret = raw_spin_trylock(lock);
-               if (!ret || (likely(lock == __rq_lockp(rq))))
+               if (!ret || (likely(lock == __rq_lockp(rq)))) {
+                       preempt_enable();
                        return ret;
+               }
                raw_spin_unlock(lock);
        }
 }
kernel/sched/sched.h +2 -1
index 6a905fe19eef..c9a52231d58a 100644
@@ -2568,11 +2568,12 @@ static inline void double_rq_unlock(struct rq *rq1, struct rq *rq2)
        __releases(rq1->lock)
        __releases(rq2->lock)
 {
-       raw_spin_rq_unlock(rq1);
        if (__rq_lockp(rq1) != __rq_lockp(rq2))
                raw_spin_rq_unlock(rq2);
        else
                __release(rq2->lock);
+
+       raw_spin_rq_unlock(rq1);
 }
 
 extern void set_rq_online (struct rq *rq);
author maintainer
#30 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 7 lines · Josh Don
Quoted 6 lines · Peter Zijlstra
Quoted 2 lines · Josh Don

Also, did you mean to have a preempt_enable_no_resched() rather than prempt_enable() in raw_spin_rq_trylock?

Quoted 7 lines
Quoted 3 lines

No, trylock really needs to be preempt_enable(), because it can have failed, at which point it will not have incremented the preemption count and our decrement can hit 0, at which point we really should reschedule.

Ah yes, makes sense. Did you want to use preempt_enable_no_resched() at all then? No chance of preempt_count() being 1 at the point of enable, as you comment below.

preempt_enable_no_resched() avoids the branch which we know we'll never take. It generates slightly saner code. It's not much, but every little bit helps, right? ;-)

Quoted 3 lines
Quoted 1 line

( I'm suffering a cold and am really quite slow atm )

No worries, hope it's a mild one.

It's not the super popular one from '19, and the family seems to be mostly recovered again, so all's well.

Quoted 6 lines
Quoted 1 line

How's this then?

Looks good to me (other than the synchronize_sched()->synchronize_rcu()).

For these locking patches,

Thanks!, I'll go fold them into the proper place and update the repo.

author maintainer
#31 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

In reply to Josh Don
Quoted 6 lines · Josh Don
Quoted 6 lines
kernel/sched/core.c +0 -0
index f732642e3e09..1a81e9cc9e5d 100644
@@ -290,6 +290,10 @@ static void sched_core_assert_empty(void)
 static void __sched_core_enable(void)
Quoted 8 lines

kernel/sched/core.c static void sched_core_assert_empty(void) · line 296

Quoted 6 lines
kernel/sched/core.c +4 -0
 {
        static_branch_enable(&__sched_core_enabled);
+       /*
+        * Ensure raw_spin_rq_*lock*() have completed before flipping.
+        */
+       synchronize_sched();

synchronize_rcu()

Moo, I actually like synchronize_sched() because it indicates it matches a preempt disabled region.

Quoted 7 lines
Quoted 7 lines
kernel/sched/core.c +0 -0
        __sched_core_flip(true);
        sched_core_assert_empty();
 }
@@ -449,16 +453,22 @@ void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
        raw_spinlock_t *lock;
 
Quoted 11 lines

kernel/sched/core.c void raw_spin_rq_lock_nested(struct rq *rq, int subclass) · line 462

Quoted 7 lines
kernel/sched/core.c +3 -0
+       preempt_disable();
        if (sched_core_disabled()) {
                raw_spin_lock_nested(&rq->__lock, subclass);
+               /* preempt *MUST* still be disabled here */
+               preempt_enable_no_resched();
                return;
        }

This approach looks good to me. I'm guessing you went this route instead of doing the re-check after locking in order to optimize the disabled case?

Exactly.

Quoted 2 lines

Recommend a comment that the preempt_disable() here pairs with the synchronize_rcu() in __sched_core_enable().

Fair enough.

#32 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

In reply to Peter Zijlstra
Quoted 71 lines · Peter Zijlstra

When switching on core-sched, CPUs need to agree which lock to use for their RQ.

The new rule will be that rq->core_enabled will be toggled while holding all rq->__locks that belong to a core. This means we need to double check the rq->core_enabled value after each lock acquire and retry if it changed.

This also has implications for those sites that take multiple RQ locks, they need to be careful that the second lock doesn't end up being the first lock.

Verify the lock pointer after acquiring the first lock, because if they're on the same core, holding any of the rq->__lock instances will pin the core state.

While there, change the rq->__lock order to CPU number, instead of rq address, this greatly simplifies the next patch.

---
 kernel/sched/core.c  |   48 ++++++++++++++++++++++++++++++++++++++++++++++--
 kernel/sched/sched.h |   41 +++++++++++------------------------------
 2 files changed, 57 insertions(+), 32 deletions(-)
kernel/sched/core.c +27 -2
@@ -186,12 +186,37 @@ int sysctl_sched_rt_runtime = 950000;
 
 void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
-	raw_spin_lock_nested(rq_lockp(rq), subclass);
+	raw_spinlock_t *lock;
+
+	if (sched_core_disabled()) {
+		raw_spin_lock_nested(&rq->__lock, subclass);
+		return;
+	}
+
+	for (;;) {
+		lock = rq_lockp(rq);
+		raw_spin_lock_nested(lock, subclass);
+		if (likely(lock == rq_lockp(rq)))
+			return;
+		raw_spin_unlock(lock);
+	}
 }
 
 bool raw_spin_rq_trylock(struct rq *rq)
 {
-	return raw_spin_trylock(rq_lockp(rq));
+	raw_spinlock_t *lock;
+	bool ret;
+
+	if (sched_core_disabled())
+		return raw_spin_trylock(&rq->__lock);
+
+	for (;;) {
+		lock = rq_lockp(rq);
+		ret = raw_spin_trylock(lock);
+		if (!ret || (likely(lock == rq_lockp(rq))))
+			return ret;
+		raw_spin_unlock(lock);
+	}
 }
 
 void raw_spin_rq_unlock(struct rq *rq)
@@ -199,6 +224,25 @@ void raw_spin_rq_unlock(struct rq *rq)
 	raw_spin_unlock(rq_lockp(rq));
 }
 

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 236

Quoted 10 lines
kernel/sched/core.c +10 -0
+#ifdef CONFIG_SMP
+/*
+ * double_rq_lock - safely lock two runqueues
+ */
+void double_rq_lock(struct rq *rq1, struct rq *rq2)
+{
+	lockdep_assert_irqs_disabled();
+
+	if (rq1->cpu > rq2->cpu)
+		swap(rq1, rq2);

I'm not sure why swap rq here instead of rq lock? This swaps dst rq and src rq and causes the subsequent logic wrong at least in try_steal_cookie().

Thanks,
-Aubrey

#33 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

In reply to Peter Zijlstra

kernel/sched/core.c

Quoted 28 lines · Peter Zijlstra

When switching on core-sched, CPUs need to agree which lock to use for their RQ.

The new rule will be that rq->core_enabled will be toggled while holding all rq->__locks that belong to a core. This means we need to double check the rq->core_enabled value after each lock acquire and retry if it changed.

This also has implications for those sites that take multiple RQ locks, they need to be careful that the second lock doesn't end up being the first lock.

Verify the lock pointer after acquiring the first lock, because if they're on the same core, holding any of the rq->__lock instances will pin the core state.

While there, change the rq->__lock order to CPU number, instead of rq address, this greatly simplifies the next patch.

---
 kernel/sched/core.c  |   48 ++++++++++++++++++++++++++++++++++++++++++++++--
 kernel/sched/sched.h |   41 +++++++++++------------------------------
 2 files changed, 57 insertions(+), 32 deletions(-)
kernel/sched/core.c +0 -0

----snip----

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 235

Quoted 13 lines
kernel/sched/core.c +9 -0
@@ -199,6 +224,25 @@ void raw_spin_rq_unlock(struct rq *rq)
        raw_spin_unlock(rq_lockp(rq));
 }
 
+#ifdef CONFIG_SMP
+/*
+ * double_rq_lock - safely lock two runqueues
+ */
+void double_rq_lock(struct rq *rq1, struct rq *rq2)
+{
+       lockdep_assert_irqs_disabled();
+
+       if (rq1->cpu > rq2->cpu)

It's still a bit hard for me to digest this function, I guess using (rq->cpu) can't guarantee the sequence of locking when coresched is enabled.

- cpu1 and cpu7 shares lockA
- cpu2 and cpu8 shares lockB

double_rq_lock(1,8) leads to lock(A) and lock(B)
double_rq_lock(7,2) leads to lock(B) and lock(A)

change to below to avoid ABBA?
+ if (__rq_lockp(rq1) > __rq_lockp(rq2))

Please correct me if I was wrong.

Thanks,
-Aubrey

Quoted 10 lines
kernel/sched/core.c +10 -0
+               swap(rq1, rq2);
+
+       raw_spin_rq_lock(rq1);
+       if (rq_lockp(rq1) == rq_lockp(rq2))
+               return;
+
+       raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+}
+#endif
+
#34 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 23 lines · Aubrey Li

On Thu, Apr 22, 2021 at 8:39 PM Peter Zijlstra <[email protected]> wrote:
----snip----

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 235

Quoted 13 lines
kernel/sched/core.c +9 -0
@@ -199,6 +224,25 @@ void raw_spin_rq_unlock(struct rq *rq)
        raw_spin_unlock(rq_lockp(rq));
 }
 
+#ifdef CONFIG_SMP
+/*
+ * double_rq_lock - safely lock two runqueues
+ */
+void double_rq_lock(struct rq *rq1, struct rq *rq2)
+{
+       lockdep_assert_irqs_disabled();
+
+       if (rq1->cpu > rq2->cpu)

It's still a bit hard for me to digest this function, I guess using (rq->cpu) can't guarantee the sequence of locking when coresched is enabled.

- cpu1 and cpu7 shares lockA
- cpu2 and cpu8 shares lockB

Quoted 7 lines

double_rq_lock(1,8) leads to lock(A) and lock(B)
double_rq_lock(7,2) leads to lock(B) and lock(A)

change to below to avoid ABBA?
+ if (__rq_lockp(rq1) > __rq_lockp(rq2))

Please correct me if I was wrong.

Great catch Aubrey. This is possibly what is causing the lockups that Don is seeing.

The proposed usage of __rq_lockp() is prone to race with sched core being enabled/disabled. It also won't order properly if we do double_rq_lock(smt0, smt1) vs double_rq_lock(smt1, smt0), since these would have equivalent __rq_lockp(). I'd propose an alternative but similar idea: order by core, then break ties by ordering on cpu.

+#ifdef CONFIG_SCHED_CORE
+ if (rq1->core->cpu > rq2->core->cpu)
+ swap(rq1, rq2);
+ else if (rq1->core->cpu == rq2->core->cpu && rq1->cpu > rq2->cpu)
+ swap(rq1, rq2);
+#else

        if (rq1->cpu > rq2->cpu)
                swap(rq1, rq2);

+#endif

#35 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 33 lines · Josh Don
Quoted 23 lines · Aubrey Li

On Thu, Apr 22, 2021 at 8:39 PM Peter Zijlstra <[email protected]> wrote:
----snip----

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 235

Quoted 13 lines
kernel/sched/core.c +9 -0
@@ -199,6 +224,25 @@ void raw_spin_rq_unlock(struct rq *rq)
        raw_spin_unlock(rq_lockp(rq));
 }
 
+#ifdef CONFIG_SMP
+/*
+ * double_rq_lock - safely lock two runqueues
+ */
+void double_rq_lock(struct rq *rq1, struct rq *rq2)
+{
+       lockdep_assert_irqs_disabled();
+
+       if (rq1->cpu > rq2->cpu)

It's still a bit hard for me to digest this function, I guess using (rq->cpu) can't guarantee the sequence of locking when coresched is enabled.

- cpu1 and cpu7 shares lockA
- cpu2 and cpu8 shares lockB

Quoted 7 lines

double_rq_lock(1,8) leads to lock(A) and lock(B)
double_rq_lock(7,2) leads to lock(B) and lock(A)

change to below to avoid ABBA?
+ if (__rq_lockp(rq1) > __rq_lockp(rq2))

Please correct me if I was wrong.

Quoted 7 lines

Great catch Aubrey. This is possibly what is causing the lockups that Don is seeing.

The proposed usage of __rq_lockp() is prone to race with sched core being enabled/disabled.It also won't order properly if we do double_rq_lock(smt0, smt1) vs double_rq_lock(smt1, smt0), since these would have equivalent __rq_lockp()

If __rq_lockp(smt0) == __rq_lockp(smt1), rq0 and rq1 won't swap, Later only one rq is locked and just returns. I'm not sure how does it not order properly?

.> I'd propose an alternative but similar idea: order by core, then break ties

Review

Quoted 7 lines

by ordering on cpu.

(quoted diff) +5 -0
+#ifdef CONFIG_SCHED_CORE
+       if (rq1->core->cpu > rq2->core->cpu)
+               swap(rq1, rq2);
+       else if (rq1->core->cpu == rq2->core->cpu && rq1->cpu > rq2->cpu)
+               swap(rq1, rq2);

That is, why the "else if" branch is needed?

Quoted 4 lines
(quoted diff) +2 -0
+#else
        if (rq1->cpu > rq2->cpu)
                swap(rq1, rq2);
+#endif
#36 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 38 lines · Aubrey Li
Quoted 36 lines · Josh Don
Quoted 23 lines · Aubrey Li

On Thu, Apr 22, 2021 at 8:39 PM Peter Zijlstra <[email protected]> wrote:
----snip----

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 235

Quoted 13 lines
kernel/sched/core.c +9 -0
@@ -199,6 +224,25 @@ void raw_spin_rq_unlock(struct rq *rq)
        raw_spin_unlock(rq_lockp(rq));
 }
 
+#ifdef CONFIG_SMP
+/*
+ * double_rq_lock - safely lock two runqueues
+ */
+void double_rq_lock(struct rq *rq1, struct rq *rq2)
+{
+       lockdep_assert_irqs_disabled();
+
+       if (rq1->cpu > rq2->cpu)

It's still a bit hard for me to digest this function, I guess using (rq->cpu) can't guarantee the sequence of locking when coresched is enabled.

- cpu1 and cpu7 shares lockA
- cpu2 and cpu8 shares lockB

Quoted 7 lines

double_rq_lock(1,8) leads to lock(A) and lock(B)
double_rq_lock(7,2) leads to lock(B) and lock(A)

change to below to avoid ABBA?
+ if (__rq_lockp(rq1) > __rq_lockp(rq2))

Please correct me if I was wrong.

Great catch Aubrey. This is possibly what is causing the lockups that Don is seeing.

Quoted 8 lines
Quoted 4 lines

The proposed usage of __rq_lockp() is prone to race with sched core being enabled/disabled.It also won't order properly if we do double_rq_lock(smt0, smt1) vs double_rq_lock(smt1, smt0), since these would have equivalent __rq_lockp()

If __rq_lockp(smt0) == __rq_lockp(smt1), rq0 and rq1 won't swap, Later only one rq is locked and just returns. I'm not sure how does it not order properly?

If there is a concurrent switch from sched_core enable <-> disable, the value of __rq_lockp() will race.

In the version you posted directly above, where we swap rq1 and rq2 if __rq_lockp(rq1) > __rqlockp(rq2) rather than comparing the cpu, the following can happen:

cpu 1 and cpu 7 share a core lock when coresched is enabled

- schedcore enabled
- double_lock(7, 1)
- __rq_lockp compares equal for 7 and 1; no swap is done
- schedcore disabled; now __rq_lockp returns the per-rq lock
- lock(__rq_lockp(7)) => lock(7)
- lock(__rq_lockp(1)) => lock(1)

Then we can also have

- schedcore disabled
- double_lock(1, 7)
- __rq_lock(1) < rq_lock(7), so no swap
- lock(__rqlockp(1)) => lock(1)
- lock(__rq_lockp(7)) => lock(7)

So we have in the first 7->1 and in the second 1->7

Quoted 11 lines

.> I'd propose an alternative but similar idea: order by core, then break ties

Review

Quoted 7 lines

by ordering on cpu.

(quoted diff) +5 -0
+#ifdef CONFIG_SCHED_CORE
+       if (rq1->core->cpu > rq2->core->cpu)
+               swap(rq1, rq2);
+       else if (rq1->core->cpu == rq2->core->cpu && rq1->cpu > rq2->cpu)
+               swap(rq1, rq2);

That is, why the "else if" branch is needed?

Ensuring that core siblings always take their locks in the same order if coresched is disabled.

Quoted 5 lines

Review

Quoted 5 lines
(quoted diff) +2 -0
+#else
        if (rq1->cpu > rq2->cpu)
                swap(rq1, rq2);
+#endif

Best,
Josh

#37 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 85 lines · Josh Don
Quoted 38 lines · Aubrey Li
Quoted 36 lines · Josh Don
Quoted 23 lines · Aubrey Li

On Thu, Apr 22, 2021 at 8:39 PM Peter Zijlstra <[email protected]> wrote:
----snip----

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 235

Quoted 13 lines
kernel/sched/core.c +9 -0
@@ -199,6 +224,25 @@ void raw_spin_rq_unlock(struct rq *rq)
        raw_spin_unlock(rq_lockp(rq));
 }
 
+#ifdef CONFIG_SMP
+/*
+ * double_rq_lock - safely lock two runqueues
+ */
+void double_rq_lock(struct rq *rq1, struct rq *rq2)
+{
+       lockdep_assert_irqs_disabled();
+
+       if (rq1->cpu > rq2->cpu)

It's still a bit hard for me to digest this function, I guess using (rq->cpu) can't guarantee the sequence of locking when coresched is enabled.

- cpu1 and cpu7 shares lockA
- cpu2 and cpu8 shares lockB

Quoted 7 lines

double_rq_lock(1,8) leads to lock(A) and lock(B)
double_rq_lock(7,2) leads to lock(B) and lock(A)

change to below to avoid ABBA?
+ if (__rq_lockp(rq1) > __rq_lockp(rq2))

Please correct me if I was wrong.

Great catch Aubrey. This is possibly what is causing the lockups that Don is seeing.

Quoted 8 lines
Quoted 4 lines

The proposed usage of __rq_lockp() is prone to race with sched core being enabled/disabled.It also won't order properly if we do double_rq_lock(smt0, smt1) vs double_rq_lock(smt1, smt0), since these would have equivalent __rq_lockp()

If __rq_lockp(smt0) == __rq_lockp(smt1), rq0 and rq1 won't swap, Later only one rq is locked and just returns. I'm not sure how does it not order properly?

If there is a concurrent switch from sched_core enable <-> disable, the value of __rq_lockp() will race.

In the version you posted directly above, where we swap rq1 and rq2 if __rq_lockp(rq1) > __rqlockp(rq2) rather than comparing the cpu, the following can happen:

cpu 1 and cpu 7 share a core lock when coresched is enabled

- schedcore enabled
- double_lock(7, 1)
- __rq_lockp compares equal for 7 and 1; no swap is done
- schedcore disabled; now __rq_lockp returns the per-rq lock
- lock(__rq_lockp(7)) => lock(7)
- lock(__rq_lockp(1)) => lock(1)

Then we can also have

- schedcore disabled
- double_lock(1, 7)
- __rq_lock(1) < rq_lock(7), so no swap
- lock(__rqlockp(1)) => lock(1)
- lock(__rq_lockp(7)) => lock(7)

So we have in the first 7->1 and in the second 1->7

Quoted 10 lines

.> I'd propose an alternative but similar idea: order by core, then break ties

Quoted 7 lines

by ordering on cpu.

(quoted diff) +5 -0
+#ifdef CONFIG_SCHED_CORE
+       if (rq1->core->cpu > rq2->core->cpu)
+               swap(rq1, rq2);
+       else if (rq1->core->cpu == rq2->core->cpu && rq1->cpu > rq2->cpu)
+               swap(rq1, rq2);
Quoted 5 lines
Quoted 1 line

That is, why the "else if" branch is needed?

Ensuring that core siblings always take their locks in the same order if coresched is disabled.

Both this and above make sense to me. Thanks for the great elaboration, Josh!

Thanks,
-Aubrey

author maintainer
#38 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

In reply to Josh Don
Quoted 9 lines · Josh Don
Quoted 9 lines

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 235

Quoted 5 lines
kernel/sched/core.c +5 -0
+void double_rq_lock(struct rq *rq1, struct rq *rq2)
+{
+       lockdep_assert_irqs_disabled();
+
+       if (rq1->cpu > rq2->cpu)

It's still a bit hard for me to digest this function, I guess using (rq->cpu) can't guarantee the sequence of locking when coresched is enabled.

Quoted 5 lines
Quoted 5 lines

- cpu1 and cpu7 shares lockA
- cpu2 and cpu8 shares lockB

double_rq_lock(1,8) leads to lock(A) and lock(B)
double_rq_lock(7,2) leads to lock(B) and lock(A)

Good one!

Quoted 2 lines
Quoted 2 lines

change to below to avoid ABBA?
+ if (__rq_lockp(rq1) > __rq_lockp(rq2))

This, however, is broken badly, not only does it suffer the problem Josh pointed out, it also breaks the rq->__lock ordering vs __sched_core_flip(), which was the whole reason the ordering needed changing in the first place.

Review

Quoted 12 lines

I'd propose an alternative but
similar idea: order by core, then break ties by ordering on cpu.

(quoted diff) +7 -0
+#ifdef CONFIG_SCHED_CORE
+       if (rq1->core->cpu > rq2->core->cpu)
+               swap(rq1, rq2);
+       else if (rq1->core->cpu == rq2->core->cpu && rq1->cpu > rq2->cpu)
+               swap(rq1, rq2);
+#else
        if (rq1->cpu > rq2->cpu)
                swap(rq1, rq2);
+#endif

I've written it like so:

static inline bool rq_order_less(struct rq *rq1, struct rq *rq2) {

#ifdef CONFIG_SCHED_CORE
	if (rq1->core->cpu < rq2->core->cpu)
		return true;
	if (rq1->core->cpu > rq2->core->cpu)
		return false;
#endif
	return rq1->cpu < rq2->cpu;
}

/*
* double_rq_lock - safely lock two runqueues */
void double_rq_lock(struct rq *rq1, struct rq *rq2)

{
	lockdep_assert_irqs_disabled();

	if (rq_order_less(rq2, rq1))
		swap(rq1, rq2);

	raw_spin_rq_lock(rq1);
	if (rq_lockp(rq1) == rq_lockp(rq2))
		return;

	raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
}
#39 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 12 lines · Peter Zijlstra
Quoted 9 lines · Josh Don
Quoted 9 lines

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 235

Quoted 5 lines
kernel/sched/core.c +5 -0
+void double_rq_lock(struct rq *rq1, struct rq *rq2)
+{
+       lockdep_assert_irqs_disabled();
+
+       if (rq1->cpu > rq2->cpu)

It's still a bit hard for me to digest this function, I guess using (rq->cpu) can't guarantee the sequence of locking when coresched is enabled.

Quoted 7 lines
Quoted 5 lines
Quoted 5 lines

- cpu1 and cpu7 shares lockA
- cpu2 and cpu8 shares lockB

double_rq_lock(1,8) leads to lock(A) and lock(B)
double_rq_lock(7,2) leads to lock(B) and lock(A)

Good one!

Hi Peter,

I've been running the same set-cookie tests on your latest repo for the last 24 hours and haven't had a single lockup. Thank you very much!

don

author maintainer
#40 lore raw

Re: [PATCH 04/19] sched: Prepare for Core-wide rq->lock

Quoted 19 lines · Don Hiatt
Quoted 18 lines · Peter Zijlstra
Quoted 14 lines · Josh Don
Quoted 14 lines

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 235

Quoted 5 lines
kernel/sched/core.c +5 -0
+void double_rq_lock(struct rq *rq1, struct rq *rq2)
+{
+       lockdep_assert_irqs_disabled();
+
+       if (rq1->cpu > rq2->cpu)

It's still a bit hard for me to digest this function, I guess using (rq->cpu) can't guarantee the sequence of locking when coresched is enabled.

- cpu1 and cpu7 shares lockA
- cpu2 and cpu8 shares lockB

double_rq_lock(1,8) leads to lock(A) and lock(B)
double_rq_lock(7,2) leads to lock(B) and lock(A)

Quoted 7 lines
Quoted 1 line

Good one!

Hi Peter,

I've been running the same set-cookie tests on your latest repo for the last 24 hours and haven't had a single lockup. Thank you very much!

Excellent, applied your Tested-by, thanks!

author maintainer
#41 lore raw

PATCH v2 4/19 sched: Prepare for Core-wide rq->lock

In reply to Peter Zijlstra

When switching on core-sched, CPUs need to agree which lock to use for their RQ.

The new rule will be that rq->core_enabled will be toggled while holding all rq->__locks that belong to a core. This means we need to double check the rq->core_enabled value after each lock acquire and retry if it changed.

This also has implications for those sites that take multiple RQ locks, they need to be careful that the second lock doesn't end up being the first lock.

Verify the lock pointer after acquiring the first lock, because if they're on the same core, holding any of the rq->__lock instances will pin the core state.

While there, change the rq->__lock order to CPU number, instead of rq address, this greatly simplifies the next patch.

---
 kernel/sched/core.c  |   48 ++++++++++++++++++++++++++++++++++++++++++++++--
 kernel/sched/sched.h |   48 +++++++++++++++++-------------------------------
 2 files changed, 63 insertions(+), 33 deletions(-)
kernel/sched/core.c +46 -2
@@ -186,12 +186,37 @@ int sysctl_sched_rt_runtime = 950000;
 
 void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
-	raw_spin_lock_nested(rq_lockp(rq), subclass);
+	raw_spinlock_t *lock;
+
+	if (sched_core_disabled()) {
+		raw_spin_lock_nested(&rq->__lock, subclass);
+		return;
+	}
+
+	for (;;) {
+		lock = rq_lockp(rq);
+		raw_spin_lock_nested(lock, subclass);
+		if (likely(lock == rq_lockp(rq)))
+			return;
+		raw_spin_unlock(lock);
+	}
 }
 
 bool raw_spin_rq_trylock(struct rq *rq)
 {
-	return raw_spin_trylock(rq_lockp(rq));
+	raw_spinlock_t *lock;
+	bool ret;
+
+	if (sched_core_disabled())
+		return raw_spin_trylock(&rq->__lock);
+
+	for (;;) {
+		lock = rq_lockp(rq);
+		ret = raw_spin_trylock(lock);
+		if (!ret || (likely(lock == rq_lockp(rq))))
+			return ret;
+		raw_spin_unlock(lock);
+	}
 }
 
 void raw_spin_rq_unlock(struct rq *rq)
@@ -199,6 +224,25 @@ void raw_spin_rq_unlock(struct rq *rq)
 	raw_spin_unlock(rq_lockp(rq));
 }
 
+#ifdef CONFIG_SMP
+/*
+ * double_rq_lock - safely lock two runqueues
+ */
+void double_rq_lock(struct rq *rq1, struct rq *rq2)
+{
+	lockdep_assert_irqs_disabled();
+
+	if (rq_order_less(rq2, rq1))
+		swap(rq1, rq2);
+
+	raw_spin_rq_lock(rq1);
+	if (rq_lockp(rq1) == rq_lockp(rq2))
+		return;
+
+	raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+}
+#endif
+
 /*
  * __task_rq_lock - lock the rq @p resides on.
  */
kernel/sched/sched.h +17 -31

89 lines not on this page · raw message

#42 lore raw

Re: [PATCH v2 04/19] sched: Prepare for Core-wide rq->lock

Quoted 72 lines · Peter Zijlstra

When switching on core-sched, CPUs need to agree which lock to use for their RQ.

The new rule will be that rq->core_enabled will be toggled while holding all rq->__locks that belong to a core. This means we need to double check the rq->core_enabled value after each lock acquire and retry if it changed.

This also has implications for those sites that take multiple RQ locks, they need to be careful that the second lock doesn't end up being the first lock.

Verify the lock pointer after acquiring the first lock, because if they're on the same core, holding any of the rq->__lock instances will pin the core state.

While there, change the rq->__lock order to CPU number, instead of rq address, this greatly simplifies the next patch.

---
 kernel/sched/core.c  |   48 ++++++++++++++++++++++++++++++++++++++++++++++--
 kernel/sched/sched.h |   48 +++++++++++++++++-------------------------------
 2 files changed, 63 insertions(+), 33 deletions(-)
kernel/sched/core.c +27 -2
@@ -186,12 +186,37 @@ int sysctl_sched_rt_runtime = 950000;
 
 void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 {
-       raw_spin_lock_nested(rq_lockp(rq), subclass);
+       raw_spinlock_t *lock;
+
+       if (sched_core_disabled()) {
+               raw_spin_lock_nested(&rq->__lock, subclass);
+               return;
+       }
+
+       for (;;) {
+               lock = rq_lockp(rq);
+               raw_spin_lock_nested(lock, subclass);
+               if (likely(lock == rq_lockp(rq)))
+                       return;
+               raw_spin_unlock(lock);
+       }
 }
 
 bool raw_spin_rq_trylock(struct rq *rq)
 {
-       return raw_spin_trylock(rq_lockp(rq));
+       raw_spinlock_t *lock;
+       bool ret;
+
+       if (sched_core_disabled())
+               return raw_spin_trylock(&rq->__lock);
+
+       for (;;) {
+               lock = rq_lockp(rq);
+               ret = raw_spin_trylock(lock);
+               if (!ret || (likely(lock == rq_lockp(rq))))
+                       return ret;
+               raw_spin_unlock(lock);
+       }
 }
 
 void raw_spin_rq_unlock(struct rq *rq)

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 231

Quoted 8 lines
kernel/sched/core.c +5 -0
@@ -199,6 +224,25 @@ void raw_spin_rq_unlock(struct rq *rq)
        raw_spin_unlock(rq_lockp(rq));
 }
 
+#ifdef CONFIG_SMP
+/*
+ * double_rq_lock - safely lock two runqueues
+ */
+void double_rq_lock(struct rq *rq1, struct rq *rq2)

Do we need the static lock checking here?

        __acquires(rq1->lock)
        __acquires(rq2->lock)

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq)

Quoted 8 lines
kernel/sched/core.c +8 -0
+{
+       lockdep_assert_irqs_disabled();
+
+       if (rq_order_less(rq2, rq1))
+               swap(rq1, rq2);
+
+       raw_spin_rq_lock(rq1);
+       if (rq_lockp(rq1) == rq_lockp(rq2)) {

And here?

                __acquire(rq2->lock);

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 240

Quoted 1 line
kernel/sched/core.c +1 -0
+               return;

}

Quoted 92 lines
kernel/sched/core.c +5 -0
+
+       raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+}
+#endif
+
 /*
  * __task_rq_lock - lock the rq @p resides on.
  */
kernel/sched/sched.h +16 -30
@@ -1113,6 +1113,11 @@ static inline bool is_migration_disabled
 #endif
 }
 
+static inline bool sched_core_disabled(void)
+{
+       return true;
+}
+
 static inline raw_spinlock_t *rq_lockp(struct rq *rq)
 {
        return &rq->__lock;
@@ -2231,10 +2236,17 @@ unsigned long arch_scale_freq_capacity(i
 }
 #endif
 
+
 #ifdef CONFIG_SMP
-#ifdef CONFIG_PREEMPTION
 
-static inline void double_rq_lock(struct rq *rq1, struct rq *rq2);
+static inline bool rq_order_less(struct rq *rq1, struct rq *rq2)
+{
+       return rq1->cpu < rq2->cpu;
+}
+
+extern void double_rq_lock(struct rq *rq1, struct rq *rq2);
+
+#ifdef CONFIG_PREEMPTION
 
 /*
  * fair double_lock_balance: Safely acquires both rq->locks in a fair
@@ -2274,14 +2286,13 @@ static inline int _double_lock_balance(s
        if (likely(raw_spin_rq_trylock(busiest)))
                return 0;
 
-       if (rq_lockp(busiest) >= rq_lockp(this_rq)) {
+       if (rq_order_less(this_rq, busiest)) {
                raw_spin_rq_lock_nested(busiest, SINGLE_DEPTH_NESTING);
                return 0;
        }
 
        raw_spin_rq_unlock(this_rq);
-       raw_spin_rq_lock(busiest);
-       raw_spin_rq_lock_nested(this_rq, SINGLE_DEPTH_NESTING);
+       double_rq_lock(this_rq, busiest);
 
        return 1;
 }
@@ -2334,31 +2345,6 @@ static inline void double_raw_lock(raw_s
 }
 
 /*
- * double_rq_lock - safely lock two runqueues
- *
- * Note this does not disable interrupts like task_rq_lock,
- * you need to do so manually before calling.
- */
-static inline void double_rq_lock(struct rq *rq1, struct rq *rq2)
-       __acquires(rq1->lock)
-       __acquires(rq2->lock)
-{
-       BUG_ON(!irqs_disabled());
-       if (rq_lockp(rq1) == rq_lockp(rq2)) {
-               raw_spin_rq_lock(rq1);
-               __acquire(rq2->lock);   /* Fake it out ;) */
-       } else {
-               if (rq_lockp(rq1) < rq_lockp(rq2)) {
-                       raw_spin_rq_lock(rq1);
-                       raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
-               } else {
-                       raw_spin_rq_lock(rq2);
-                       raw_spin_rq_lock_nested(rq1, SINGLE_DEPTH_NESTING);
-               }
-       }
-}
-
-/*
  * double_rq_unlock - safely unlock two runqueues
  *
  * Note this does not restore interrupts like task_rq_unlock,

kernel/sched/sched.h static inline void double_rq_unlock(stru · line 2361

Quoted 9 lines
kernel/sched/sched.h +1 -1
@@ -2368,11 +2354,11 @@ static inline void double_rq_unlock(stru
        __releases(rq1->lock)
        __releases(rq2->lock)
 {
-       raw_spin_rq_unlock(rq1);
        if (rq_lockp(rq1) != rq_lockp(rq2))
                raw_spin_rq_unlock(rq2);
        else
                __release(rq2->lock);
+       raw_spin_rq_unlock(rq1);

This change seems not necessary, as the softlockup root cause is not the misorder lock release.

Thanks,
-Aubrey

author maintainer
#43 lore raw

Re: [PATCH v2 04/19] sched: Prepare for Core-wide rq->lock

Quoted 20 lines · Aubrey Li

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 231

Quoted 5 lines
kernel/sched/core.c +5 -0
+#ifdef CONFIG_SMP
+/*
+ * double_rq_lock - safely lock two runqueues
+ */
+void double_rq_lock(struct rq *rq1, struct rq *rq2)

Do we need the static lock checking here?

        __acquires(rq1->lock)
        __acquires(rq2->lock)

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq)

Quoted 8 lines
kernel/sched/core.c +8 -0
+{
+       lockdep_assert_irqs_disabled();
+
+       if (rq_order_less(rq2, rq1))
+               swap(rq1, rq2);
+
+       raw_spin_rq_lock(rq1);
+       if (rq_lockp(rq1) == rq_lockp(rq2)) {

And here?

Quoted 8 lines
                __acquire(rq2->lock);
Quoted 1 line

+ return;

}

kernel/sched/core.c void raw_spin_rq_unlock(struct rq *rq) · line 244

Quoted 4 lines
kernel/sched/core.c +4 -0
+
+       raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
+}
+#endif

I'd as soon rip out all that sparse annotation crud; I don't think I've ever had any benefit from it.

Quoted 5 lines
Quoted 5 lines
kernel/sched/sched.h +0 -1
@@ -2368,11 +2354,11 @@ static inline void double_rq_unlock(stru
        __releases(rq1->lock)
        __releases(rq2->lock)
 {
-       raw_spin_rq_unlock(rq1);
Quoted 8 lines

kernel/sched/sched.h static inline void double_rq_unlock(stru · line 2361

Quoted 5 lines
kernel/sched/sched.h +1 -0
        if (rq_lockp(rq1) != rq_lockp(rq2))
                raw_spin_rq_unlock(rq2);
        else
                __release(rq2->lock);
+       raw_spin_rq_unlock(rq1);

This change seems not necessary, as the softlockup root cause is not the misorder lock release.

No, it really is needed; rq_lockp() is not stable if we don't hold a lock.

author maintainer
#44 lore raw

PATCH 5/19 sched: Core-wide rq->lock

In reply to Peter Zijlstra

Introduce the basic infrastructure to have a core wide rq->lock.

This relies on the rq->__lock order being in increasing CPU number. It is also constrained to SMT8 per lockdep (and SMT256 per preempt_count).

Luckily SMT8 is the max supported SMT count for Linux (Mips, Sparc and Power are known to have this).

---
 kernel/Kconfig.preempt |    6 ++
 kernel/sched/core.c    |  139 +++++++++++++++++++++++++++++++++++++++++++++++++
 kernel/sched/sched.h   |   37 +++++++++++++
 3 files changed, 182 insertions(+)
kernel/Kconfig.preempt +6 -0
@@ -99,3 +99,9 @@ config PREEMPT_DYNAMIC
 
 	  Interesting if you want the same pre-built kernel should be used for
 	  both Server and Desktop workloads.
+
+config SCHED_CORE
+	bool "Core Scheduling for SMT"
+	default y
+	depends on SCHED_SMT
+
kernel/sched/core.c +137 -0

161 lines not on this page · raw message

kernel/sched/sched.h +37 -0
@@ -1075,6 +1075,12 @@ struct rq {
 #endif
 	unsigned int		push_busy;
 	struct cpu_stop_work	push_work;
+
+#ifdef CONFIG_SCHED_CORE
+	/* per rq */
+	struct rq		*core;
+	unsigned int		core_enabled;
+#endif
 };
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
@@ -1113,6 +1119,35 @@ static inline bool is_migration_disabled
 #endif
 }
 
+#ifdef CONFIG_SCHED_CORE
+
+DECLARE_STATIC_KEY_FALSE(__sched_core_enabled);
+
+static inline bool sched_core_enabled(struct rq *rq)
+{
+	return static_branch_unlikely(&__sched_core_enabled) && rq->core_enabled;
+}
+
+static inline bool sched_core_disabled(void)
+{
+	return !static_branch_unlikely(&__sched_core_enabled);
+}
+
+static inline raw_spinlock_t *rq_lockp(struct rq *rq)
+{
+	if (sched_core_enabled(rq))
+		return &rq->core->__lock;
+
+	return &rq->__lock;
+}
+
+#else /* !CONFIG_SCHED_CORE */
+
+static inline bool sched_core_enabled(struct rq *rq)
+{
+	return false;
+}
+
 static inline bool sched_core_disabled(void)
 {
 	return true;
@@ -1123,6 +1158,8 @@ static inline raw_spinlock_t *rq_lockp(s
 	return &rq->__lock;
 }
 
+#endif /* CONFIG_SCHED_CORE */
+
 static inline void lockdep_assert_rq_held(struct rq *rq)
 {
 	lockdep_assert_held(rq_lockp(rq));
author maintainer
#45 lore raw

PATCH v2 5/19 sched: Core-wide rq->lock

Introduce the basic infrastructure to have a core wide rq->lock.

This relies on the rq->__lock order being in increasing CPU number (inside a core). It is also constrained to SMT8 per lockdep (and SMT256 per preempt_count).

Luckily SMT8 is the max supported SMT count for Linux (Mips, Sparc and Power are known to have this).

---
 kernel/Kconfig.preempt |    6 +
 kernel/sched/core.c    |  164 +++++++++++++++++++++++++++++++++++++++++++++++--
 kernel/sched/sched.h   |   58 +++++++++++++++++
 3 files changed, 224 insertions(+), 4 deletions(-)
kernel/Kconfig.preempt +6 -0
@@ -99,3 +99,9 @@ config PREEMPT_DYNAMIC
 
 	  Interesting if you want the same pre-built kernel should be used for
 	  both Server and Desktop workloads.
+
+config SCHED_CORE
+	bool "Core Scheduling for SMT"
+	default y
+	depends on SCHED_SMT
+
kernel/sched/core.c +160 -4

214 lines not on this page · raw message

kernel/sched/sched.h +58 -0

82 lines not on this page · raw message

merge bot Applied
#46 lore raw

[tip: sched/core] sched: Core-wide rq->lock

The following commit has been merged into the sched/core branch of tip:

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     9edeaea1bc452372718837ed2ba775811baf1ba1
Gitweb:        https://git.kernel.org/tip/9edeaea1bc452372718837ed2ba775811baf1ba1
Author:        Peter Zijlstra <[email protected]>
AuthorDate:    Tue, 17 Nov 2020 18:19:34 -05:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 12 May 2021 11:43:27 +02:00

sched: Core-wide rq->lock

Introduce the basic infrastructure to have a core wide rq->lock.

This relies on the rq->__lock order being in increasing CPU number (inside a core). It is also constrained to SMT8 per lockdep (and SMT256 per preempt_count).

Luckily SMT8 is the max supported SMT count for Linux (Mips, Sparc and Power are known to have this).

---
 kernel/Kconfig.preempt |   6 +-
 kernel/sched/core.c    | 164 +++++++++++++++++++++++++++++++++++++++-
 kernel/sched/sched.h   |  58 ++++++++++++++-
 3 files changed, 224 insertions(+), 4 deletions(-)
kernel/Kconfig.preempt +6 -0
index 4160173..ea1e333 100644
@@ -99,3 +99,9 @@ config PREEMPT_DYNAMIC
 
 	  Interesting if you want the same pre-built kernel should be used for
 	  both Server and Desktop workloads.
+
+config SCHED_CORE
+	bool "Core Scheduling for SMT"
+	default y
+	depends on SCHED_SMT
+
kernel/sched/core.c +160 -4

214 lines not on this page · raw message

kernel/sched/sched.h +58 -0

82 lines not on this page · raw message

author maintainer
#47 lore raw

PATCH 6/19 sched: Optimize rq_lockp() usage

In reply to Peter Zijlstra

rq_lockp() includes a static_branch(), which is asm-goto, which is asm volatile which defeats regular CSE. This means that:

	if (!static_branch(&foo))
		return simple;

	if (static_branch(&foo) && cond)
		return complex;

Doesn't fold and we get horrible code. Introduce __rq_lockp() without the static_branch() on.

---
 kernel/sched/core.c     |   16 ++++++++--------
 kernel/sched/deadline.c |    4 ++--
 kernel/sched/fair.c     |    2 +-
 kernel/sched/sched.h    |   33 +++++++++++++++++++++++++--------
 4 files changed, 36 insertions(+), 19 deletions(-)
kernel/sched/core.c +8 -8
@@ -281,9 +281,9 @@ void raw_spin_rq_lock_nested(struct rq *
 	}
 
 	for (;;) {
-		lock = rq_lockp(rq);
+		lock = __rq_lockp(rq);
 		raw_spin_lock_nested(lock, subclass);
-		if (likely(lock == rq_lockp(rq)))
+		if (likely(lock == __rq_lockp(rq)))
 			return;
 		raw_spin_unlock(lock);
 	}
@@ -298,9 +298,9 @@ bool raw_spin_rq_trylock(struct rq *rq)
 		return raw_spin_trylock(&rq->__lock);
 
 	for (;;) {
-		lock = rq_lockp(rq);
+		lock = __rq_lockp(rq);
 		ret = raw_spin_trylock(lock);
-		if (!ret || (likely(lock == rq_lockp(rq))))
+		if (!ret || (likely(lock == __rq_lockp(rq))))
 			return ret;
 		raw_spin_unlock(lock);
 	}
@@ -323,7 +323,7 @@ void double_rq_lock(struct rq *rq1, stru
 		swap(rq1, rq2);
 
 	raw_spin_rq_lock(rq1);
-	if (rq_lockp(rq1) == rq_lockp(rq2))
+	if (__rq_lockp(rq1) == __rq_lockp(rq2))
 		return;
 
 	raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
@@ -2594,7 +2594,7 @@ void set_task_cpu(struct task_struct *p,
 	 * task_rq_lock().
 	 */
 	WARN_ON_ONCE(debug_locks && !(lockdep_is_held(&p->pi_lock) ||
-				      lockdep_is_held(rq_lockp(task_rq(p)))));
+				      lockdep_is_held(__rq_lockp(task_rq(p)))));
 #endif
 	/*
 	 * Clearly, migrating tasks to offline CPUs is a fairly daft thing.
@@ -4220,7 +4220,7 @@ prepare_lock_switch(struct rq *rq, struc
 	 * do an early lockdep release here:
 	 */
 	rq_unpin_lock(rq, rf);
-	spin_release(&rq_lockp(rq)->dep_map, _THIS_IP_);
+	spin_release(&__rq_lockp(rq)->dep_map, _THIS_IP_);
 #ifdef CONFIG_DEBUG_SPINLOCK
 	/* this is a valid case when another task releases the spinlock */
 	rq_lockp(rq)->owner = next;
@@ -4234,7 +4234,7 @@ static inline void finish_lock_switch(st
 	 * fix up the runqueue lock - which gets 'carried over' from
 	 * prev into current:
 	 */
-	spin_acquire(&rq_lockp(rq)->dep_map, 0, 0, _THIS_IP_);
+	spin_acquire(&__rq_lockp(rq)->dep_map, 0, 0, _THIS_IP_);
 	__balance_callbacks(rq);
 	raw_spin_rq_unlock_irq(rq);
 }
kernel/sched/deadline.c +2 -2
@@ -1097,9 +1097,9 @@ static enum hrtimer_restart dl_task_time
 		 * If the runqueue is no longer available, migrate the
 		 * task elsewhere. This necessarily changes rq.
 		 */
-		lockdep_unpin_lock(rq_lockp(rq), rf.cookie);
+		lockdep_unpin_lock(__rq_lockp(rq), rf.cookie);
 		rq = dl_task_offline_migration(rq, p);
-		rf.cookie = lockdep_pin_lock(rq_lockp(rq));
+		rf.cookie = lockdep_pin_lock(__rq_lockp(rq));
 		update_rq_clock(rq);
 
 		/*
kernel/sched/fair.c +1 -1
@@ -1107,7 +1107,7 @@ struct numa_group {
 static struct numa_group *deref_task_numa_group(struct task_struct *p)
 {
 	return rcu_dereference_check(p->numa_group, p == current ||
-		(lockdep_is_held(rq_lockp(task_rq(p))) && !READ_ONCE(p->on_cpu)));
+		(lockdep_is_held(__rq_lockp(task_rq(p))) && !READ_ONCE(p->on_cpu)));
 }
 
 static struct numa_group *deref_curr_numa_group(struct task_struct *p)
kernel/sched/sched.h +25 -8
@@ -1127,6 +1127,10 @@ static inline bool sched_core_disabled(v
 	return !static_branch_unlikely(&__sched_core_enabled);
 }
 
+/*
+ * Be careful with this function; not for general use. The return value isn't
+ * stable unless you actually hold a relevant rq->__lock.
+ */
 static inline raw_spinlock_t *rq_lockp(struct rq *rq)
 {
 	if (sched_core_enabled(rq))
@@ -1135,6 +1139,14 @@ static inline raw_spinlock_t *rq_lockp(s
 	return &rq->__lock;
 }
 
+static inline raw_spinlock_t *__rq_lockp(struct rq *rq)
+{
+	if (rq->core_enabled)
+		return &rq->core->__lock;
+
+	return &rq->__lock;
+}
+
 #else /* !CONFIG_SCHED_CORE */
 
 static inline bool sched_core_enabled(struct rq *rq)
@@ -1152,11 +1164,16 @@ static inline raw_spinlock_t *rq_lockp(s
 	return &rq->__lock;
 }
 
+static inline raw_spinlock_t *__rq_lockp(struct rq *rq)
+{
+	return &rq->__lock;
+}
+
 #endif /* CONFIG_SCHED_CORE */
 
 static inline void lockdep_assert_rq_held(struct rq *rq)
 {
-	lockdep_assert_held(rq_lockp(rq));
+	lockdep_assert_held(__rq_lockp(rq));
 }
 
 extern void raw_spin_rq_lock_nested(struct rq *rq, int subclass);
@@ -1340,7 +1357,7 @@ extern struct callback_head balance_push
  */
 static inline void rq_pin_lock(struct rq *rq, struct rq_flags *rf)
 {
-	rf->cookie = lockdep_pin_lock(rq_lockp(rq));
+	rf->cookie = lockdep_pin_lock(__rq_lockp(rq));
 
 #ifdef CONFIG_SCHED_DEBUG
 	rq->clock_update_flags &= (RQCF_REQ_SKIP|RQCF_ACT_SKIP);
@@ -1358,12 +1375,12 @@ static inline void rq_unpin_lock(struct
 		rf->clock_update_flags = RQCF_UPDATED;
 #endif
 
-	lockdep_unpin_lock(rq_lockp(rq), rf->cookie);
+	lockdep_unpin_lock(__rq_lockp(rq), rf->cookie);
 }
 
 static inline void rq_repin_lock(struct rq *rq, struct rq_flags *rf)
 {
-	lockdep_repin_lock(rq_lockp(rq), rf->cookie);
+	lockdep_repin_lock(__rq_lockp(rq), rf->cookie);
 
 #ifdef CONFIG_SCHED_DEBUG
 	/*
@@ -2306,7 +2323,7 @@ static inline int _double_lock_balance(s
 	__acquires(busiest->lock)
 	__acquires(this_rq->lock)
 {
-	if (rq_lockp(this_rq) == rq_lockp(busiest))
+	if (__rq_lockp(this_rq) == __rq_lockp(busiest))
 		return 0;
 
 	if (likely(raw_spin_rq_trylock(busiest)))
@@ -2338,9 +2355,9 @@ static inline int double_lock_balance(st
 static inline void double_unlock_balance(struct rq *this_rq, struct rq *busiest)
 	__releases(busiest->lock)
 {
-	if (rq_lockp(this_rq) != rq_lockp(busiest))
+	if (__rq_lockp(this_rq) != __rq_lockp(busiest))
 		raw_spin_rq_unlock(busiest);
-	lock_set_subclass(&rq_lockp(this_rq)->dep_map, 0, _RET_IP_);
+	lock_set_subclass(&__rq_lockp(this_rq)->dep_map, 0, _RET_IP_);
 }
 
 static inline void double_lock(spinlock_t *l1, spinlock_t *l2)
@@ -2381,7 +2398,7 @@ static inline void double_rq_unlock(stru
 	__releases(rq2->lock)
 {
 	raw_spin_rq_unlock(rq1);
-	if (rq_lockp(rq1) != rq_lockp(rq2))
+	if (__rq_lockp(rq1) != __rq_lockp(rq2))
 		raw_spin_rq_unlock(rq2);
 	else
 		__release(rq2->lock);
merge bot Applied
#48 lore raw

[tip: sched/core] sched: Optimize rq_lockp() usage

The following commit has been merged into the sched/core branch of tip:

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     9ef7e7e33bcdb57be1afb28884053c28b5f05240
Gitweb:        https://git.kernel.org/tip/9ef7e7e33bcdb57be1afb28884053c28b5f05240
Author:        Peter Zijlstra <[email protected]>
AuthorDate:    Wed, 03 Mar 2021 16:45:41 +01:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 12 May 2021 11:43:27 +02:00

sched: Optimize rq_lockp() usage

rq_lockp() includes a static_branch(), which is asm-goto, which is asm volatile which defeats regular CSE. This means that:

	if (!static_branch(&foo))
		return simple;

	if (static_branch(&foo) && cond)
		return complex;

Doesn't fold and we get horrible code. Introduce __rq_lockp() without the static_branch() on.

---
 kernel/sched/core.c     | 16 ++++++++--------
 kernel/sched/deadline.c |  4 ++--
 kernel/sched/fair.c     |  2 +-
 kernel/sched/sched.h    | 33 +++++++++++++++++++++++++--------
 4 files changed, 36 insertions(+), 19 deletions(-)
kernel/sched/core.c +8 -8
index 384b793..42c1c88 100644
@@ -300,9 +300,9 @@ void raw_spin_rq_lock_nested(struct rq *rq, int subclass)
 	}
 
 	for (;;) {
-		lock = rq_lockp(rq);
+		lock = __rq_lockp(rq);
 		raw_spin_lock_nested(lock, subclass);
-		if (likely(lock == rq_lockp(rq))) {
+		if (likely(lock == __rq_lockp(rq))) {
 			/* preempt_count *MUST* be > 1 */
 			preempt_enable_no_resched();
 			return;
@@ -325,9 +325,9 @@ bool raw_spin_rq_trylock(struct rq *rq)
 	}
 
 	for (;;) {
-		lock = rq_lockp(rq);
+		lock = __rq_lockp(rq);
 		ret = raw_spin_trylock(lock);
-		if (!ret || (likely(lock == rq_lockp(rq)))) {
+		if (!ret || (likely(lock == __rq_lockp(rq)))) {
 			preempt_enable();
 			return ret;
 		}
@@ -352,7 +352,7 @@ void double_rq_lock(struct rq *rq1, struct rq *rq2)
 		swap(rq1, rq2);
 
 	raw_spin_rq_lock(rq1);
-	if (rq_lockp(rq1) == rq_lockp(rq2))
+	if (__rq_lockp(rq1) == __rq_lockp(rq2))
 		return;
 
 	raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING);
@@ -2622,7 +2622,7 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
 	 * task_rq_lock().
 	 */
 	WARN_ON_ONCE(debug_locks && !(lockdep_is_held(&p->pi_lock) ||
-				      lockdep_is_held(rq_lockp(task_rq(p)))));
+				      lockdep_is_held(__rq_lockp(task_rq(p)))));
 #endif
 	/*
 	 * Clearly, migrating tasks to offline CPUs is a fairly daft thing.
@@ -4248,7 +4248,7 @@ prepare_lock_switch(struct rq *rq, struct task_struct *next, struct rq_flags *rf
 	 * do an early lockdep release here:
 	 */
 	rq_unpin_lock(rq, rf);
-	spin_release(&rq_lockp(rq)->dep_map, _THIS_IP_);
+	spin_release(&__rq_lockp(rq)->dep_map, _THIS_IP_);
 #ifdef CONFIG_DEBUG_SPINLOCK
 	/* this is a valid case when another task releases the spinlock */
 	rq_lockp(rq)->owner = next;
@@ -4262,7 +4262,7 @@ static inline void finish_lock_switch(struct rq *rq)
 	 * fix up the runqueue lock - which gets 'carried over' from
 	 * prev into current:
 	 */
-	spin_acquire(&rq_lockp(rq)->dep_map, 0, 0, _THIS_IP_);
+	spin_acquire(&__rq_lockp(rq)->dep_map, 0, 0, _THIS_IP_);
 	__balance_callbacks(rq);
 	raw_spin_rq_unlock_irq(rq);
 }
kernel/sched/deadline.c +2 -2
index 6e99b8b..fb0eb9a 100644
@@ -1097,9 +1097,9 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer)
 		 * If the runqueue is no longer available, migrate the
 		 * task elsewhere. This necessarily changes rq.
 		 */
-		lockdep_unpin_lock(rq_lockp(rq), rf.cookie);
+		lockdep_unpin_lock(__rq_lockp(rq), rf.cookie);
 		rq = dl_task_offline_migration(rq, p);
-		rf.cookie = lockdep_pin_lock(rq_lockp(rq));
+		rf.cookie = lockdep_pin_lock(__rq_lockp(rq));
 		update_rq_clock(rq);
 
 		/*
kernel/sched/fair.c +1 -1
index e50bd75..18960d0 100644
@@ -1107,7 +1107,7 @@ struct numa_group {
 static struct numa_group *deref_task_numa_group(struct task_struct *p)
 {
 	return rcu_dereference_check(p->numa_group, p == current ||
-		(lockdep_is_held(rq_lockp(task_rq(p))) && !READ_ONCE(p->on_cpu)));
+		(lockdep_is_held(__rq_lockp(task_rq(p))) && !READ_ONCE(p->on_cpu)));
 }
 
 static struct numa_group *deref_curr_numa_group(struct task_struct *p)
kernel/sched/sched.h +25 -8
index 29418b8..ca30af3 100644
@@ -1133,6 +1133,10 @@ static inline bool sched_core_disabled(void)
 	return !static_branch_unlikely(&__sched_core_enabled);
 }
 
+/*
+ * Be careful with this function; not for general use. The return value isn't
+ * stable unless you actually hold a relevant rq->__lock.
+ */
 static inline raw_spinlock_t *rq_lockp(struct rq *rq)
 {
 	if (sched_core_enabled(rq))
@@ -1141,6 +1145,14 @@ static inline raw_spinlock_t *rq_lockp(struct rq *rq)
 	return &rq->__lock;
 }
 
+static inline raw_spinlock_t *__rq_lockp(struct rq *rq)
+{
+	if (rq->core_enabled)
+		return &rq->core->__lock;
+
+	return &rq->__lock;
+}
+
 #else /* !CONFIG_SCHED_CORE */
 
 static inline bool sched_core_enabled(struct rq *rq)
@@ -1158,11 +1170,16 @@ static inline raw_spinlock_t *rq_lockp(struct rq *rq)
 	return &rq->__lock;
 }
 
+static inline raw_spinlock_t *__rq_lockp(struct rq *rq)
+{
+	return &rq->__lock;
+}
+
 #endif /* CONFIG_SCHED_CORE */
 
 static inline void lockdep_assert_rq_held(struct rq *rq)
 {
-	lockdep_assert_held(rq_lockp(rq));
+	lockdep_assert_held(__rq_lockp(rq));
 }
 
 extern void raw_spin_rq_lock_nested(struct rq *rq, int subclass);
@@ -1346,7 +1363,7 @@ extern struct callback_head balance_push_callback;
  */
 static inline void rq_pin_lock(struct rq *rq, struct rq_flags *rf)
 {
-	rf->cookie = lockdep_pin_lock(rq_lockp(rq));
+	rf->cookie = lockdep_pin_lock(__rq_lockp(rq));
 
 #ifdef CONFIG_SCHED_DEBUG
 	rq->clock_update_flags &= (RQCF_REQ_SKIP|RQCF_ACT_SKIP);
@@ -1364,12 +1381,12 @@ static inline void rq_unpin_lock(struct rq *rq, struct rq_flags *rf)
 		rf->clock_update_flags = RQCF_UPDATED;
 #endif
 
-	lockdep_unpin_lock(rq_lockp(rq), rf->cookie);
+	lockdep_unpin_lock(__rq_lockp(rq), rf->cookie);
 }
 
 static inline void rq_repin_lock(struct rq *rq, struct rq_flags *rf)
 {
-	lockdep_repin_lock(rq_lockp(rq), rf->cookie);
+	lockdep_repin_lock(__rq_lockp(rq), rf->cookie);
 
 #ifdef CONFIG_SCHED_DEBUG
 	/*
@@ -2338,7 +2355,7 @@ static inline int _double_lock_balance(struct rq *this_rq, struct rq *busiest)
 	__acquires(busiest->lock)
 	__acquires(this_rq->lock)
 {
-	if (rq_lockp(this_rq) == rq_lockp(busiest))
+	if (__rq_lockp(this_rq) == __rq_lockp(busiest))
 		return 0;
 
 	if (likely(raw_spin_rq_trylock(busiest)))
@@ -2370,9 +2387,9 @@ static inline int double_lock_balance(struct rq *this_rq, struct rq *busiest)
 static inline void double_unlock_balance(struct rq *this_rq, struct rq *busiest)
 	__releases(busiest->lock)
 {
-	if (rq_lockp(this_rq) != rq_lockp(busiest))
+	if (__rq_lockp(this_rq) != __rq_lockp(busiest))
 		raw_spin_rq_unlock(busiest);
-	lock_set_subclass(&rq_lockp(this_rq)->dep_map, 0, _RET_IP_);
+	lock_set_subclass(&__rq_lockp(this_rq)->dep_map, 0, _RET_IP_);
 }
 
 static inline void double_lock(spinlock_t *l1, spinlock_t *l2)
@@ -2412,7 +2429,7 @@ static inline void double_rq_unlock(struct rq *rq1, struct rq *rq2)
 	__releases(rq1->lock)
 	__releases(rq2->lock)
 {
-	if (rq_lockp(rq1) != rq_lockp(rq2))
+	if (__rq_lockp(rq1) != __rq_lockp(rq2))
 		raw_spin_rq_unlock(rq2);
 	else
 		__release(rq2->lock);
author maintainer
#49 lore raw

PATCH 7/19 sched: Allow sched_core_put() from atomic context

In reply to Peter Zijlstra

Stuff the meat of sched_core_put() into a work such that we can use sched_core_put() from atomic context.

---
 kernel/sched/core.c |   33 +++++++++++++++++++++++++++------
 1 file changed, 27 insertions(+), 6 deletions(-)
kernel/sched/core.c +27 -6
@@ -102,7 +102,7 @@ DEFINE_STATIC_KEY_FALSE(__sched_core_ena
  */
 
 static DEFINE_MUTEX(sched_core_mutex);
-static int sched_core_count;
+static atomic_t sched_core_count;
 static struct cpumask sched_core_mask;
 
 static void __sched_core_flip(bool enabled)
@@ -165,18 +165,39 @@ static void __sched_core_disable(void)
 
 void sched_core_get(void)
 {
+	if (atomic_inc_not_zero(&sched_core_count))
+		return;
+
 	mutex_lock(&sched_core_mutex);
-	if (!sched_core_count++)
+	if (!atomic_read(&sched_core_count))
 		__sched_core_enable();
+
+	smp_mb__before_atomic();
+	atomic_inc(&sched_core_count);
 	mutex_unlock(&sched_core_mutex);
 }
 
-void sched_core_put(void)
+static void __sched_core_put(struct work_struct *work)
 {
-	mutex_lock(&sched_core_mutex);
-	if (!--sched_core_count)
+	if (atomic_dec_and_mutex_lock(&sched_core_count, &sched_core_mutex)) {
 		__sched_core_disable();
-	mutex_unlock(&sched_core_mutex);
+		mutex_unlock(&sched_core_mutex);
+	}
+}
+
+void sched_core_put(void)
+{
+	static DECLARE_WORK(_work, __sched_core_put);
+
+	/*
+	 * "There can be only one"
+	 *
+	 * Either this is the last one, or we don't actually need to do any
+	 * 'work'. If it is the last *again*, we rely on
+	 * WORK_STRUCT_PENDING_BIT.
+	 */
+	if (!atomic_add_unless(&sched_core_count, -1, 1))
+		schedule_work(&_work);
 }
 
 #endif /* CONFIG_SCHED_CORE */
merge bot Applied
#50 lore raw

[tip: sched/core] sched: Allow sched_core_put() from atomic context

The following commit has been merged into the sched/core branch of tip:

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     875feb41fd20f6bd6054c9e79a5bcd9da6d8d2b2
Gitweb:        https://git.kernel.org/tip/875feb41fd20f6bd6054c9e79a5bcd9da6d8d2b2
Author:        Peter Zijlstra <[email protected]>
AuthorDate:    Mon, 29 Mar 2021 10:08:58 +02:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 12 May 2021 11:43:27 +02:00

sched: Allow sched_core_put() from atomic context

Stuff the meat of sched_core_put() into a work such that we can use sched_core_put() from atomic context.

---
 kernel/sched/core.c | 33 +++++++++++++++++++++++++++------
 1 file changed, 27 insertions(+), 6 deletions(-)
kernel/sched/core.c +27 -6
index 42c1c88..85147be 100644
@@ -102,7 +102,7 @@ DEFINE_STATIC_KEY_FALSE(__sched_core_enabled);
  */
 
 static DEFINE_MUTEX(sched_core_mutex);
-static int sched_core_count;
+static atomic_t sched_core_count;
 static struct cpumask sched_core_mask;
 
 static void __sched_core_flip(bool enabled)
@@ -170,18 +170,39 @@ static void __sched_core_disable(void)
 
 void sched_core_get(void)
 {
+	if (atomic_inc_not_zero(&sched_core_count))
+		return;
+
 	mutex_lock(&sched_core_mutex);
-	if (!sched_core_count++)
+	if (!atomic_read(&sched_core_count))
 		__sched_core_enable();
+
+	smp_mb__before_atomic();
+	atomic_inc(&sched_core_count);
 	mutex_unlock(&sched_core_mutex);
 }
 
-void sched_core_put(void)
+static void __sched_core_put(struct work_struct *work)
 {
-	mutex_lock(&sched_core_mutex);
-	if (!--sched_core_count)
+	if (atomic_dec_and_mutex_lock(&sched_core_count, &sched_core_mutex)) {
 		__sched_core_disable();
-	mutex_unlock(&sched_core_mutex);
+		mutex_unlock(&sched_core_mutex);
+	}
+}
+
+void sched_core_put(void)
+{
+	static DECLARE_WORK(_work, __sched_core_put);
+
+	/*
+	 * "There can be only one"
+	 *
+	 * Either this is the last one, or we don't actually need to do any
+	 * 'work'. If it is the last *again*, we rely on
+	 * WORK_STRUCT_PENDING_BIT.
+	 */
+	if (!atomic_add_unless(&sched_core_count, -1, 1))
+		schedule_work(&_work);
 }
 
 #endif /* CONFIG_SCHED_CORE */
author maintainer
#51 lore raw

PATCH 8/19 sched: Introduce sched_class::pick_task()

In reply to Peter Zijlstra

Because sched_class::pick_next_task() also implies sched_class::set_next_task() (and possibly put_prev_task() and newidle_balance) it is not state invariant. This makes it unsuitable for remote task selection.

---
 kernel/sched/deadline.c  |   16 ++++++++++++++--
 kernel/sched/fair.c      |   40 +++++++++++++++++++++++++++++++++++++---
 kernel/sched/idle.c      |    8 ++++++++
 kernel/sched/rt.c        |   15 +++++++++++++--
 kernel/sched/sched.h     |    3 +++
 kernel/sched/stop_task.c |   14 ++++++++++++--
 6 files changed, 87 insertions(+), 9 deletions(-)
kernel/sched/deadline.c +14 -2
@@ -1873,7 +1873,7 @@ static struct sched_dl_entity *pick_next
 	return rb_entry(left, struct sched_dl_entity, rb_node);
 }
 
-static struct task_struct *pick_next_task_dl(struct rq *rq)
+static struct task_struct *pick_task_dl(struct rq *rq)
 {
 	struct sched_dl_entity *dl_se;
 	struct dl_rq *dl_rq = &rq->dl;
@@ -1885,7 +1885,18 @@ static struct task_struct *pick_next_tas
 	dl_se = pick_next_dl_entity(rq, dl_rq);
 	BUG_ON(!dl_se);
 	p = dl_task_of(dl_se);
-	set_next_task_dl(rq, p, true);
+
+	return p;
+}
+
+static struct task_struct *pick_next_task_dl(struct rq *rq)
+{
+	struct task_struct *p;
+
+	p = pick_task_dl(rq);
+	if (p)
+		set_next_task_dl(rq, p, true);
+
 	return p;
 }
 
@@ -2557,6 +2568,7 @@ DEFINE_SCHED_CLASS(dl) = {
 
 #ifdef CONFIG_SMP
 	.balance		= balance_dl,
+	.pick_task		= pick_task_dl,
 	.select_task_rq		= select_task_rq_dl,
 	.migrate_task_rq	= migrate_task_rq_dl,
 	.set_cpus_allowed       = set_cpus_allowed_dl,
kernel/sched/fair.c +37 -3
@@ -4400,6 +4400,8 @@ check_preempt_tick(struct cfs_rq *cfs_rq
 static void
 set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
+	clear_buddies(cfs_rq, se);
+
 	/* 'current' is not kept within the tree. */
 	if (se->on_rq) {
 		/*
@@ -4459,7 +4461,7 @@ pick_next_entity(struct cfs_rq *cfs_rq,
 	 * Avoid running the skip buddy, if running something else can
 	 * be done without getting too unfair.
 	 */
-	if (cfs_rq->skip == se) {
+	if (cfs_rq->skip && cfs_rq->skip == se) {
 		struct sched_entity *second;
 
 		if (se == curr) {
@@ -4486,8 +4488,6 @@ pick_next_entity(struct cfs_rq *cfs_rq,
 		se = cfs_rq->last;
 	}
 
-	clear_buddies(cfs_rq, se);
-
 	return se;
 }
 
@@ -7018,6 +7018,39 @@ static void check_preempt_wakeup(struct
 		set_last_buddy(se);
 }
 
+#ifdef CONFIG_SMP
+static struct task_struct *pick_task_fair(struct rq *rq)
+{
+	struct sched_entity *se;
+	struct cfs_rq *cfs_rq;
+
+again:
+	cfs_rq = &rq->cfs;
+	if (!cfs_rq->nr_running)
+		return NULL;
+
+	do {
+		struct sched_entity *curr = cfs_rq->curr;
+
+		/* When we pick for a remote RQ, we'll not have done put_prev_entity() */
+		if (curr) {
+			if (curr->on_rq)
+				update_curr(cfs_rq);
+			else
+				curr = NULL;
+
+			if (unlikely(check_cfs_rq_runtime(cfs_rq)))
+				goto again;
+		}
+
+		se = pick_next_entity(cfs_rq, curr);
+		cfs_rq = group_cfs_rq(se);
+	} while (cfs_rq);
+
+	return task_of(se);
+}
+#endif
+
 struct task_struct *
 pick_next_task_fair(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 {
@@ -11220,6 +11253,7 @@ DEFINE_SCHED_CLASS(fair) = {
 
 #ifdef CONFIG_SMP
 	.balance		= balance_fair,
+	.pick_task		= pick_task_fair,
 	.select_task_rq		= select_task_rq_fair,
 	.migrate_task_rq	= migrate_task_rq_fair,
 
kernel/sched/idle.c +8 -0
@@ -406,6 +406,13 @@ static void set_next_task_idle(struct rq
 	schedstat_inc(rq->sched_goidle);
 }
 
+#ifdef CONFIG_SMP
+static struct task_struct *pick_task_idle(struct rq *rq)
+{
+	return rq->idle;
+}
+#endif
+
 struct task_struct *pick_next_task_idle(struct rq *rq)
 {
 	struct task_struct *next = rq->idle;
@@ -473,6 +480,7 @@ DEFINE_SCHED_CLASS(idle) = {
 
 #ifdef CONFIG_SMP
 	.balance		= balance_idle,
+	.pick_task		= pick_task_idle,
 	.select_task_rq		= select_task_rq_idle,
 	.set_cpus_allowed	= set_cpus_allowed_common,
 #endif
kernel/sched/rt.c +13 -2
@@ -1626,7 +1626,7 @@ static struct task_struct *_pick_next_ta
 	return rt_task_of(rt_se);
 }
 
-static struct task_struct *pick_next_task_rt(struct rq *rq)
+static struct task_struct *pick_task_rt(struct rq *rq)
 {
 	struct task_struct *p;
 
@@ -1634,7 +1634,17 @@ static struct task_struct *pick_next_tas
 		return NULL;
 
 	p = _pick_next_task_rt(rq);
-	set_next_task_rt(rq, p, true);
+
+	return p;
+}
+
+static struct task_struct *pick_next_task_rt(struct rq *rq)
+{
+	struct task_struct *p = pick_task_rt(rq);
+
+	if (p)
+		set_next_task_rt(rq, p, true);
+
 	return p;
 }
 
@@ -2483,6 +2493,7 @@ DEFINE_SCHED_CLASS(rt) = {
 
 #ifdef CONFIG_SMP
 	.balance		= balance_rt,
+	.pick_task		= pick_task_rt,
 	.select_task_rq		= select_task_rq_rt,
 	.set_cpus_allowed       = set_cpus_allowed_common,
 	.rq_online              = rq_online_rt,
kernel/sched/sched.h +3 -0
@@ -1839,6 +1839,9 @@ struct sched_class {
 #ifdef CONFIG_SMP
 	int (*balance)(struct rq *rq, struct task_struct *prev, struct rq_flags *rf);
 	int  (*select_task_rq)(struct task_struct *p, int task_cpu, int flags);
+
+	struct task_struct * (*pick_task)(struct rq *rq);
+
 	void (*migrate_task_rq)(struct task_struct *p, int new_cpu);
 
 	void (*task_woken)(struct rq *this_rq, struct task_struct *task);
kernel/sched/stop_task.c +12 -2
@@ -34,15 +34,24 @@ static void set_next_task_stop(struct rq
 	stop->se.exec_start = rq_clock_task(rq);
 }
 
-static struct task_struct *pick_next_task_stop(struct rq *rq)
+static struct task_struct *pick_task_stop(struct rq *rq)
 {
 	if (!sched_stop_runnable(rq))
 		return NULL;
 
-	set_next_task_stop(rq, rq->stop, true);
 	return rq->stop;
 }
 
+static struct task_struct *pick_next_task_stop(struct rq *rq)
+{
+	struct task_struct *p = pick_task_stop(rq);
+
+	if (p)
+		set_next_task_stop(rq, p, true);
+
+	return p;
+}
+
 static void
 enqueue_task_stop(struct rq *rq, struct task_struct *p, int flags)
 {
@@ -123,6 +132,7 @@ DEFINE_SCHED_CLASS(stop) = {
 
 #ifdef CONFIG_SMP
 	.balance		= balance_stop,
+	.pick_task		= pick_task_stop,
 	.select_task_rq		= select_task_rq_stop,
 	.set_cpus_allowed	= set_cpus_allowed_common,
 #endif
merge bot Applied
#52 lore raw

[tip: sched/core] sched: Introduce sched_class::pick_task()

The following commit has been merged into the sched/core branch of tip:

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     21f56ffe4482e501b9e83737612493eeaac21f5a
Gitweb:        https://git.kernel.org/tip/21f56ffe4482e501b9e83737612493eeaac21f5a
Author:        Peter Zijlstra <[email protected]>
AuthorDate:    Tue, 17 Nov 2020 18:19:32 -05:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 12 May 2021 11:43:28 +02:00

sched: Introduce sched_class::pick_task()

Because sched_class::pick_next_task() also implies sched_class::set_next_task() (and possibly put_prev_task() and newidle_balance) it is not state invariant. This makes it unsuitable for remote task selection.

---
 kernel/sched/deadline.c  | 16 ++++++++++++++--
 kernel/sched/fair.c      | 40 ++++++++++++++++++++++++++++++++++++---
 kernel/sched/idle.c      |  8 ++++++++-
 kernel/sched/rt.c        | 15 +++++++++++++--
 kernel/sched/sched.h     |  3 +++-
 kernel/sched/stop_task.c | 14 ++++++++++++--
 6 files changed, 87 insertions(+), 9 deletions(-)
kernel/sched/deadline.c +14 -2
index fb0eb9a..3829c5a 100644
@@ -1852,7 +1852,7 @@ static struct sched_dl_entity *pick_next_dl_entity(struct rq *rq,
 	return rb_entry(left, struct sched_dl_entity, rb_node);
 }
 
-static struct task_struct *pick_next_task_dl(struct rq *rq)
+static struct task_struct *pick_task_dl(struct rq *rq)
 {
 	struct sched_dl_entity *dl_se;
 	struct dl_rq *dl_rq = &rq->dl;
@@ -1864,7 +1864,18 @@ static struct task_struct *pick_next_task_dl(struct rq *rq)
 	dl_se = pick_next_dl_entity(rq, dl_rq);
 	BUG_ON(!dl_se);
 	p = dl_task_of(dl_se);
-	set_next_task_dl(rq, p, true);
+
+	return p;
+}
+
+static struct task_struct *pick_next_task_dl(struct rq *rq)
+{
+	struct task_struct *p;
+
+	p = pick_task_dl(rq);
+	if (p)
+		set_next_task_dl(rq, p, true);
+
 	return p;
 }
 
@@ -2539,6 +2550,7 @@ DEFINE_SCHED_CLASS(dl) = {
 
 #ifdef CONFIG_SMP
 	.balance		= balance_dl,
+	.pick_task		= pick_task_dl,
 	.select_task_rq		= select_task_rq_dl,
 	.migrate_task_rq	= migrate_task_rq_dl,
 	.set_cpus_allowed       = set_cpus_allowed_dl,
kernel/sched/fair.c +37 -3
index 18960d0..51d72ab 100644
@@ -4419,6 +4419,8 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr)
 static void
 set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
+	clear_buddies(cfs_rq, se);
+
 	/* 'current' is not kept within the tree. */
 	if (se->on_rq) {
 		/*
@@ -4478,7 +4480,7 @@ pick_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *curr)
 	 * Avoid running the skip buddy, if running something else can
 	 * be done without getting too unfair.
 	 */
-	if (cfs_rq->skip == se) {
+	if (cfs_rq->skip && cfs_rq->skip == se) {
 		struct sched_entity *second;
 
 		if (se == curr) {
@@ -4505,8 +4507,6 @@ pick_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *curr)
 		se = cfs_rq->last;
 	}
 
-	clear_buddies(cfs_rq, se);
-
 	return se;
 }
 
@@ -7095,6 +7095,39 @@ preempt:
 		set_last_buddy(se);
 }
 
+#ifdef CONFIG_SMP
+static struct task_struct *pick_task_fair(struct rq *rq)
+{
+	struct sched_entity *se;
+	struct cfs_rq *cfs_rq;
+
+again:
+	cfs_rq = &rq->cfs;
+	if (!cfs_rq->nr_running)
+		return NULL;
+
+	do {
+		struct sched_entity *curr = cfs_rq->curr;
+
+		/* When we pick for a remote RQ, we'll not have done put_prev_entity() */
+		if (curr) {
+			if (curr->on_rq)
+				update_curr(cfs_rq);
+			else
+				curr = NULL;
+
+			if (unlikely(check_cfs_rq_runtime(cfs_rq)))
+				goto again;
+		}
+
+		se = pick_next_entity(cfs_rq, curr);
+		cfs_rq = group_cfs_rq(se);
+	} while (cfs_rq);
+
+	return task_of(se);
+}
+#endif
+
 struct task_struct *
 pick_next_task_fair(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 {
@@ -11298,6 +11331,7 @@ DEFINE_SCHED_CLASS(fair) = {
 
 #ifdef CONFIG_SMP
 	.balance		= balance_fair,
+	.pick_task		= pick_task_fair,
 	.select_task_rq		= select_task_rq_fair,
 	.migrate_task_rq	= migrate_task_rq_fair,
 
kernel/sched/idle.c +8 -0
index 0194768..43646e7 100644
@@ -439,6 +439,13 @@ static void set_next_task_idle(struct rq *rq, struct task_struct *next, bool fir
 	schedstat_inc(rq->sched_goidle);
 }
 
+#ifdef CONFIG_SMP
+static struct task_struct *pick_task_idle(struct rq *rq)
+{
+	return rq->idle;
+}
+#endif
+
 struct task_struct *pick_next_task_idle(struct rq *rq)
 {
 	struct task_struct *next = rq->idle;
@@ -506,6 +513,7 @@ DEFINE_SCHED_CLASS(idle) = {
 
 #ifdef CONFIG_SMP
 	.balance		= balance_idle,
+	.pick_task		= pick_task_idle,
 	.select_task_rq		= select_task_rq_idle,
 	.set_cpus_allowed	= set_cpus_allowed_common,
 #endif
kernel/sched/rt.c +13 -2
index b3d39c3..a525447 100644
@@ -1626,7 +1626,7 @@ static struct task_struct *_pick_next_task_rt(struct rq *rq)
 	return rt_task_of(rt_se);
 }
 
-static struct task_struct *pick_next_task_rt(struct rq *rq)
+static struct task_struct *pick_task_rt(struct rq *rq)
 {
 	struct task_struct *p;
 
@@ -1634,7 +1634,17 @@ static struct task_struct *pick_next_task_rt(struct rq *rq)
 		return NULL;
 
 	p = _pick_next_task_rt(rq);
-	set_next_task_rt(rq, p, true);
+
+	return p;
+}
+
+static struct task_struct *pick_next_task_rt(struct rq *rq)
+{
+	struct task_struct *p = pick_task_rt(rq);
+
+	if (p)
+		set_next_task_rt(rq, p, true);
+
 	return p;
 }
 
@@ -2483,6 +2493,7 @@ DEFINE_SCHED_CLASS(rt) = {
 
 #ifdef CONFIG_SMP
 	.balance		= balance_rt,
+	.pick_task		= pick_task_rt,
 	.select_task_rq		= select_task_rq_rt,
 	.set_cpus_allowed       = set_cpus_allowed_common,
 	.rq_online              = rq_online_rt,
kernel/sched/sched.h +3 -0
index ca30af3..fa990cd 100644
@@ -1953,6 +1953,9 @@ struct sched_class {
 #ifdef CONFIG_SMP
 	int (*balance)(struct rq *rq, struct task_struct *prev, struct rq_flags *rf);
 	int  (*select_task_rq)(struct task_struct *p, int task_cpu, int flags);
+
+	struct task_struct * (*pick_task)(struct rq *rq);
+
 	void (*migrate_task_rq)(struct task_struct *p, int new_cpu);
 
 	void (*task_woken)(struct rq *this_rq, struct task_struct *task);
kernel/sched/stop_task.c +12 -2
index 55f3912..f988ebe 100644
@@ -34,15 +34,24 @@ static void set_next_task_stop(struct rq *rq, struct task_struct *stop, bool fir
 	stop->se.exec_start = rq_clock_task(rq);
 }
 
-static struct task_struct *pick_next_task_stop(struct rq *rq)
+static struct task_struct *pick_task_stop(struct rq *rq)
 {
 	if (!sched_stop_runnable(rq))
 		return NULL;
 
-	set_next_task_stop(rq, rq->stop, true);
 	return rq->stop;
 }
 
+static struct task_struct *pick_next_task_stop(struct rq *rq)
+{
+	struct task_struct *p = pick_task_stop(rq);
+
+	if (p)
+		set_next_task_stop(rq, p, true);
+
+	return p;
+}
+
 static void
 enqueue_task_stop(struct rq *rq, struct task_struct *p, int flags)
 {
@@ -123,6 +132,7 @@ DEFINE_SCHED_CLASS(stop) = {
 
 #ifdef CONFIG_SMP
 	.balance		= balance_stop,
+	.pick_task		= pick_task_stop,
 	.select_task_rq		= select_task_rq_stop,
 	.set_cpus_allowed	= set_cpus_allowed_common,
 #endif
author maintainer
#53 lore raw

PATCH 9/19 sched: Basic tracking of matching tasks

In reply to Peter Zijlstra

Introduce task_struct::core_cookie as an opaque identifier for core scheduling. When enabled; core scheduling will only allow matching task to be on the core; where idle matches everything.

When task_struct::core_cookie is set (and core scheduling is enabled) these tasks are indexed in a second RB-tree, first on cookie value then on scheduling function, such that matching task selection always finds the most elegible match.

NOTE: *shudder* at the overhead...

NOTE: *sigh*, a 3rd copy of the scheduling function; the alternative is per class tracking of cookies and that just duplicates a lot of stuff for no raisin (the 2nd copy lives in the rt-mutex PI code).

---
 include/linux/sched.h |    8 ++
 kernel/sched/core.c   |  152 ++++++++++++++++++++++++++++++++++++++++++++++++--
 kernel/sched/fair.c   |   46 ---------------
 kernel/sched/sched.h  |   55 ++++++++++++++++++
 4 files changed, 210 insertions(+), 51 deletions(-)
include/linux/sched.h +7 -1
@@ -699,10 +699,16 @@ struct task_struct {
 	const struct sched_class	*sched_class;
 	struct sched_entity		se;
 	struct sched_rt_entity		rt;
+	struct sched_dl_entity		dl;
+
+#ifdef CONFIG_SCHED_CORE
+	struct rb_node			core_node;
+	unsigned long			core_cookie;
+#endif
+
 #ifdef CONFIG_CGROUP_SCHED
 	struct task_group		*sched_task_group;
 #endif
-	struct sched_dl_entity		dl;
 
 #ifdef CONFIG_UCLAMP_TASK
 	/*
kernel/sched/core.c +148 -4

188 lines not on this page · raw message

kernel/sched/fair.c +0 -46

66 lines not on this page · raw message

kernel/sched/sched.h +55 -0

67 lines not on this page · raw message

merge bot Applied
#54 lore raw

[tip: sched/core] sched: Basic tracking of matching tasks

The following commit has been merged into the sched/core branch of tip:

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     8a311c740b53324ec584e0e3bb7077d56b123c28
Gitweb:        https://git.kernel.org/tip/8a311c740b53324ec584e0e3bb7077d56b123c28
Author:        Peter Zijlstra <[email protected]>
AuthorDate:    Tue, 17 Nov 2020 18:19:36 -05:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 12 May 2021 11:43:28 +02:00

sched: Basic tracking of matching tasks

Introduce task_struct::core_cookie as an opaque identifier for core scheduling. When enabled; core scheduling will only allow matching task to be on the core; where idle matches everything.

When task_struct::core_cookie is set (and core scheduling is enabled) these tasks are indexed in a second RB-tree, first on cookie value then on scheduling function, such that matching task selection always finds the most elegible match.

NOTE: *shudder* at the overhead...

NOTE: *sigh*, a 3rd copy of the scheduling function; the alternative is per class tracking of cookies and that just duplicates a lot of stuff for no raisin (the 2nd copy lives in the rt-mutex PI code).

---
 include/linux/sched.h |   8 +-
 kernel/sched/core.c   | 152 +++++++++++++++++++++++++++++++++++++++--
 kernel/sched/fair.c   |  46 +------------
 kernel/sched/sched.h  |  55 +++++++++++++++-
 4 files changed, 210 insertions(+), 51 deletions(-)
include/linux/sched.h +7 -1
index d2c8813..45eedcc 100644
@@ -700,10 +700,16 @@ struct task_struct {
 	const struct sched_class	*sched_class;
 	struct sched_entity		se;
 	struct sched_rt_entity		rt;
+	struct sched_dl_entity		dl;
+
+#ifdef CONFIG_SCHED_CORE
+	struct rb_node			core_node;
+	unsigned long			core_cookie;
+#endif
+
 #ifdef CONFIG_CGROUP_SCHED
 	struct task_group		*sched_task_group;
 #endif
-	struct sched_dl_entity		dl;
 
 #ifdef CONFIG_UCLAMP_TASK
 	/*
kernel/sched/core.c +148 -4

192 lines not on this page · raw message

kernel/sched/fair.c +0 -46

66 lines not on this page · raw message

kernel/sched/sched.h +55 -0

67 lines not on this page · raw message

author maintainer
#55 lore raw

PATCH 10/19 sched: Add core wide task selection and scheduling

In reply to Peter Zijlstra

Instead of only selecting a local task, select a task for all SMT siblings for every reschedule on the core (irrespective which logical CPU does the reschedule).

---
 kernel/sched/core.c  |  301 ++++++++++++++++++++++++++++++++++++++++++++++++++-
 kernel/sched/sched.h |    6 -
 2 files changed, 305 insertions(+), 2 deletions(-)
kernel/sched/core.c +300 -1
@@ -5077,7 +5077,7 @@ static void put_prev_task_balance(struct
  * Pick up the highest-prio task:
  */
 static inline struct task_struct *
-pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
+__pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 {
 	const struct sched_class *class;
 	struct task_struct *p;
@@ -5118,6 +5118,294 @@ pick_next_task(struct rq *rq, struct tas
 }
 
 #ifdef CONFIG_SCHED_CORE
+static inline bool is_task_rq_idle(struct task_struct *t)
+{
+	return (task_rq(t)->idle == t);
+}
+
+static inline bool cookie_equals(struct task_struct *a, unsigned long cookie)
+{
+	return is_task_rq_idle(a) || (a->core_cookie == cookie);
+}
+
+static inline bool cookie_match(struct task_struct *a, struct task_struct *b)
+{
+	if (is_task_rq_idle(a) || is_task_rq_idle(b))
+		return true;
+
+	return a->core_cookie == b->core_cookie;
+}
+
+// XXX fairness/fwd progress conditions
+/*
+ * Returns
+ * - NULL if there is no runnable task for this class.
+ * - the highest priority task for this runqueue if it matches
+ *   rq->core->core_cookie or its priority is greater than max.
+ * - Else returns idle_task.
+ */
+static struct task_struct *
+pick_task(struct rq *rq, const struct sched_class *class, struct task_struct *max)
+{
+	struct task_struct *class_pick, *cookie_pick;
+	unsigned long cookie = rq->core->core_cookie;
+
+	class_pick = class->pick_task(rq);
+	if (!class_pick)
+		return NULL;
+
+	if (!cookie) {
+		/*
+		 * If class_pick is tagged, return it only if it has
+		 * higher priority than max.
+		 */
+		if (max && class_pick->core_cookie &&
+		    prio_less(class_pick, max))
+			return idle_sched_class.pick_task(rq);
+
+		return class_pick;
+	}
+
+	/*
+	 * If class_pick is idle or matches cookie, return early.
+	 */
+	if (cookie_equals(class_pick, cookie))
+		return class_pick;
+
+	cookie_pick = sched_core_find(rq, cookie);
+
+	/*
+	 * If class > max && class > cookie, it is the highest priority task on
+	 * the core (so far) and it must be selected, otherwise we must go with
+	 * the cookie pick in order to satisfy the constraint.
+	 */
+	if (prio_less(cookie_pick, class_pick) &&
+	    (!max || prio_less(max, class_pick)))
+		return class_pick;
+
+	return cookie_pick;
+}
+
+static struct task_struct *
+pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
+{
+	struct task_struct *next, *max = NULL;
+	const struct sched_class *class;
+	const struct cpumask *smt_mask;
+	bool need_sync;
+	int i, j, cpu;
+
+	if (!sched_core_enabled(rq))
+		return __pick_next_task(rq, prev, rf);
+
+	cpu = cpu_of(rq);
+
+	/* Stopper task is switching into idle, no need core-wide selection. */
+	if (cpu_is_offline(cpu)) {
+		/*
+		 * Reset core_pick so that we don't enter the fastpath when
+		 * coming online. core_pick would already be migrated to
+		 * another cpu during offline.
+		 */
+		rq->core_pick = NULL;
+		return __pick_next_task(rq, prev, rf);
+	}
+
+	/*
+	 * If there were no {en,de}queues since we picked (IOW, the task
+	 * pointers are all still valid), and we haven't scheduled the last
+	 * pick yet, do so now.
+	 *
+	 * rq->core_pick can be NULL if no selection was made for a CPU because
+	 * it was either offline or went offline during a sibling's core-wide
+	 * selection. In this case, do a core-wide selection.
+	 */
+	if (rq->core->core_pick_seq == rq->core->core_task_seq &&
+	    rq->core->core_pick_seq != rq->core_sched_seq &&
+	    rq->core_pick) {
+		WRITE_ONCE(rq->core_sched_seq, rq->core->core_pick_seq);
+
+		next = rq->core_pick;
+		if (next != prev) {
+			put_prev_task(rq, prev);
+			set_next_task(rq, next);
+		}
+
+		rq->core_pick = NULL;
+		return next;
+	}
+
+	put_prev_task_balance(rq, prev, rf);
+
+	smt_mask = cpu_smt_mask(cpu);
+
+	/*
+	 * core->core_task_seq, core->core_pick_seq, rq->core_sched_seq
+	 *
+	 * @task_seq guards the task state ({en,de}queues)
+	 * @pick_seq is the @task_seq we did a selection on
+	 * @sched_seq is the @pick_seq we scheduled
+	 *
+	 * However, preemptions can cause multiple picks on the same task set.
+	 * 'Fix' this by also increasing @task_seq for every pick.
+	 */
+	rq->core->core_task_seq++;
+	need_sync = !!rq->core->core_cookie;
+
+	/* reset state */
+	rq->core->core_cookie = 0UL;
+	for_each_cpu(i, smt_mask) {
+		struct rq *rq_i = cpu_rq(i);
+
+		rq_i->core_pick = NULL;
+
+		if (rq_i->core_forceidle) {
+			need_sync = true;
+			rq_i->core_forceidle = false;
+		}
+
+		if (i != cpu)
+			update_rq_clock(rq_i);
+	}
+
+	/*
+	 * Try and select tasks for each sibling in decending sched_class
+	 * order.
+	 */
+	for_each_class(class) {
+again:
+		for_each_cpu_wrap(i, smt_mask, cpu) {
+			struct rq *rq_i = cpu_rq(i);
+			struct task_struct *p;
+
+			if (rq_i->core_pick)
+				continue;
+
+			/*
+			 * If this sibling doesn't yet have a suitable task to
+			 * run; ask for the most elegible task, given the
+			 * highest priority task already selected for this
+			 * core.
+			 */
+			p = pick_task(rq_i, class, max);
+			if (!p) {
+				/*
+				 * If there weren't no cookies; we don't need to
+				 * bother with the other siblings.
+				 * If the rest of the core is not running a tagged
+				 * task, i.e.  need_sync == 0, and the current CPU
+				 * which called into the schedule() loop does not
+				 * have any tasks for this class, skip selecting for
+				 * other siblings since there's no point. We don't skip
+				 * for RT/DL because that could make CFS force-idle RT.
+				 */
+				if (i == cpu && !need_sync && class == &fair_sched_class)
+					goto next_class;
+
+				continue;
+			}
+
+			/*
+			 * Optimize the 'normal' case where there aren't any
+			 * cookies and we don't need to sync up.
+			 */
+			if (i == cpu && !need_sync && !p->core_cookie) {
+				next = p;
+				goto done;
+			}
+
+			rq_i->core_pick = p;
+
+			/*
+			 * If this new candidate is of higher priority than the
+			 * previous; and they're incompatible; we need to wipe
+			 * the slate and start over. pick_task makes sure that
+			 * p's priority is more than max if it doesn't match
+			 * max's cookie.
+			 *
+			 * NOTE: this is a linear max-filter and is thus bounded
+			 * in execution time.
+			 */
+			if (!max || !cookie_match(max, p)) {
+				struct task_struct *old_max = max;
+
+				rq->core->core_cookie = p->core_cookie;
+				max = p;
+
+				if (old_max) {
+					for_each_cpu(j, smt_mask) {
+						if (j == i)
+							continue;
+
+						cpu_rq(j)->core_pick = NULL;
+					}
+					goto again;
+				} else {
+					/*
+					 * Once we select a task for a cpu, we
+					 * should not be doing an unconstrained
+					 * pick because it might starve a task
+					 * on a forced idle cpu.
+					 */
+					need_sync = true;
+				}
+
+			}
+		}
+next_class:;
+	}
+
+	rq->core->core_pick_seq = rq->core->core_task_seq;
+	next = rq->core_pick;
+	rq->core_sched_seq = rq->core->core_pick_seq;
+
+	/* Something should have been selected for current CPU */
+	WARN_ON_ONCE(!next);
+
+	/*
+	 * Reschedule siblings
+	 *
+	 * NOTE: L1TF -- at this point we're no longer running the old task and
+	 * sending an IPI (below) ensures the sibling will no longer be running
+	 * their task. This ensures there is no inter-sibling overlap between
+	 * non-matching user state.
+	 */
+	for_each_cpu(i, smt_mask) {
+		struct rq *rq_i = cpu_rq(i);
+
+		/*
+		 * An online sibling might have gone offline before a task
+		 * could be picked for it, or it might be offline but later
+		 * happen to come online, but its too late and nothing was
+		 * picked for it.  That's Ok - it will pick tasks for itself,
+		 * so ignore it.
+		 */
+		if (!rq_i->core_pick)
+			continue;
+
+		if (is_task_rq_idle(rq_i->core_pick) && rq_i->nr_running)
+			rq_i->core_forceidle = true;
+
+		if (i == cpu) {
+			rq_i->core_pick = NULL;
+			continue;
+		}
+
+		/* Did we break L1TF mitigation requirements? */
+		WARN_ON_ONCE(!cookie_match(next, rq_i->core_pick));
+
+		if (rq_i->curr == rq_i->core_pick) {
+			rq_i->core_pick = NULL;
+			continue;
+		}
+
+		resched_curr(rq_i);
+	}
+
+done:
+	set_next_task(rq, next);
+	return next;
+}
 
 static inline void sched_core_cpu_starting(unsigned int cpu)
 {
@@ -5151,6 +5439,12 @@ static inline void sched_core_cpu_starti
 
 static inline void sched_core_cpu_starting(unsigned int cpu) {}
 
+static struct task_struct *
+pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
+{
+	return __pick_next_task(rq, prev, rf);
+}
+
 #endif /* CONFIG_SCHED_CORE */
 
 /*
@@ -8048,7 +8342,12 @@ void __init sched_init(void)
 
 #ifdef CONFIG_SCHED_CORE
 		rq->core = NULL;
+		rq->core_pick = NULL;
 		rq->core_enabled = 0;
+		rq->core_tree = RB_ROOT;
+		rq->core_forceidle = false;
+
+		rq->core_cookie = 0UL;
 #endif
 	}
 
kernel/sched/sched.h +5 -1
@@ -1065,11 +1065,16 @@ struct rq {
 #ifdef CONFIG_SCHED_CORE
 	/* per rq */
 	struct rq		*core;
+	struct task_struct	*core_pick;
 	unsigned int		core_enabled;
+	unsigned int		core_sched_seq;
 	struct rb_root		core_tree;
+	unsigned char		core_forceidle;
 
 	/* shared state */
 	unsigned int		core_task_seq;
+	unsigned int		core_pick_seq;
+	unsigned long		core_cookie;
 #endif
 };
 
@@ -1977,7 +1982,6 @@ static inline void put_prev_task(struct
 
 static inline void set_next_task(struct rq *rq, struct task_struct *next)
 {
-	WARN_ON_ONCE(rq->curr != next);
 	next->sched_class->set_next_task(rq, next, false);
 }
 
merge bot Applied
#56 lore raw

[tip: sched/core] sched: Add core wide task selection and scheduling

The following commit has been merged into the sched/core branch of tip:

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     539f65125d20aacab54d02d77f10a839f45b09dc
Gitweb:        https://git.kernel.org/tip/539f65125d20aacab54d02d77f10a839f45b09dc
Author:        Peter Zijlstra <[email protected]>
AuthorDate:    Tue, 17 Nov 2020 18:19:37 -05:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 12 May 2021 11:43:28 +02:00

sched: Add core wide task selection and scheduling

Instead of only selecting a local task, select a task for all SMT siblings for every reschedule on the core (irrespective which logical CPU does the reschedule).

---
 kernel/sched/core.c  | 301 +++++++++++++++++++++++++++++++++++++++++-
 kernel/sched/sched.h |   6 +-
 2 files changed, 305 insertions(+), 2 deletions(-)
kernel/sched/core.c +300 -1
index c057d47..db763f4 100644
@@ -5282,7 +5282,7 @@ static void put_prev_task_balance(struct rq *rq, struct task_struct *prev,
  * Pick up the highest-prio task:
  */
 static inline struct task_struct *
-pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
+__pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 {
 	const struct sched_class *class;
 	struct task_struct *p;
@@ -5323,6 +5323,294 @@ restart:
 }
 
 #ifdef CONFIG_SCHED_CORE
+static inline bool is_task_rq_idle(struct task_struct *t)
+{
+	return (task_rq(t)->idle == t);
+}
+
+static inline bool cookie_equals(struct task_struct *a, unsigned long cookie)
+{
+	return is_task_rq_idle(a) || (a->core_cookie == cookie);
+}
+
+static inline bool cookie_match(struct task_struct *a, struct task_struct *b)
+{
+	if (is_task_rq_idle(a) || is_task_rq_idle(b))
+		return true;
+
+	return a->core_cookie == b->core_cookie;
+}
+
+// XXX fairness/fwd progress conditions
+/*
+ * Returns
+ * - NULL if there is no runnable task for this class.
+ * - the highest priority task for this runqueue if it matches
+ *   rq->core->core_cookie or its priority is greater than max.
+ * - Else returns idle_task.
+ */
+static struct task_struct *
+pick_task(struct rq *rq, const struct sched_class *class, struct task_struct *max)
+{
+	struct task_struct *class_pick, *cookie_pick;
+	unsigned long cookie = rq->core->core_cookie;
+
+	class_pick = class->pick_task(rq);
+	if (!class_pick)
+		return NULL;
+
+	if (!cookie) {
+		/*
+		 * If class_pick is tagged, return it only if it has
+		 * higher priority than max.
+		 */
+		if (max && class_pick->core_cookie &&
+		    prio_less(class_pick, max))
+			return idle_sched_class.pick_task(rq);
+
+		return class_pick;
+	}
+
+	/*
+	 * If class_pick is idle or matches cookie, return early.
+	 */
+	if (cookie_equals(class_pick, cookie))
+		return class_pick;
+
+	cookie_pick = sched_core_find(rq, cookie);
+
+	/*
+	 * If class > max && class > cookie, it is the highest priority task on
+	 * the core (so far) and it must be selected, otherwise we must go with
+	 * the cookie pick in order to satisfy the constraint.
+	 */
+	if (prio_less(cookie_pick, class_pick) &&
+	    (!max || prio_less(max, class_pick)))
+		return class_pick;
+
+	return cookie_pick;
+}
+
+static struct task_struct *
+pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
+{
+	struct task_struct *next, *max = NULL;
+	const struct sched_class *class;
+	const struct cpumask *smt_mask;
+	bool need_sync;
+	int i, j, cpu;
+
+	if (!sched_core_enabled(rq))
+		return __pick_next_task(rq, prev, rf);
+
+	cpu = cpu_of(rq);
+
+	/* Stopper task is switching into idle, no need core-wide selection. */
+	if (cpu_is_offline(cpu)) {
+		/*
+		 * Reset core_pick so that we don't enter the fastpath when
+		 * coming online. core_pick would already be migrated to
+		 * another cpu during offline.
+		 */
+		rq->core_pick = NULL;
+		return __pick_next_task(rq, prev, rf);
+	}
+
+	/*
+	 * If there were no {en,de}queues since we picked (IOW, the task
+	 * pointers are all still valid), and we haven't scheduled the last
+	 * pick yet, do so now.
+	 *
+	 * rq->core_pick can be NULL if no selection was made for a CPU because
+	 * it was either offline or went offline during a sibling's core-wide
+	 * selection. In this case, do a core-wide selection.
+	 */
+	if (rq->core->core_pick_seq == rq->core->core_task_seq &&
+	    rq->core->core_pick_seq != rq->core_sched_seq &&
+	    rq->core_pick) {
+		WRITE_ONCE(rq->core_sched_seq, rq->core->core_pick_seq);
+
+		next = rq->core_pick;
+		if (next != prev) {
+			put_prev_task(rq, prev);
+			set_next_task(rq, next);
+		}
+
+		rq->core_pick = NULL;
+		return next;
+	}
+
+	put_prev_task_balance(rq, prev, rf);
+
+	smt_mask = cpu_smt_mask(cpu);
+
+	/*
+	 * core->core_task_seq, core->core_pick_seq, rq->core_sched_seq
+	 *
+	 * @task_seq guards the task state ({en,de}queues)
+	 * @pick_seq is the @task_seq we did a selection on
+	 * @sched_seq is the @pick_seq we scheduled
+	 *
+	 * However, preemptions can cause multiple picks on the same task set.
+	 * 'Fix' this by also increasing @task_seq for every pick.
+	 */
+	rq->core->core_task_seq++;
+	need_sync = !!rq->core->core_cookie;
+
+	/* reset state */
+	rq->core->core_cookie = 0UL;
+	for_each_cpu(i, smt_mask) {
+		struct rq *rq_i = cpu_rq(i);
+
+		rq_i->core_pick = NULL;
+
+		if (rq_i->core_forceidle) {
+			need_sync = true;
+			rq_i->core_forceidle = false;
+		}
+
+		if (i != cpu)
+			update_rq_clock(rq_i);
+	}
+
+	/*
+	 * Try and select tasks for each sibling in decending sched_class
+	 * order.
+	 */
+	for_each_class(class) {
+again:
+		for_each_cpu_wrap(i, smt_mask, cpu) {
+			struct rq *rq_i = cpu_rq(i);
+			struct task_struct *p;
+
+			if (rq_i->core_pick)
+				continue;
+
+			/*
+			 * If this sibling doesn't yet have a suitable task to
+			 * run; ask for the most elegible task, given the
+			 * highest priority task already selected for this
+			 * core.
+			 */
+			p = pick_task(rq_i, class, max);
+			if (!p) {
+				/*
+				 * If there weren't no cookies; we don't need to
+				 * bother with the other siblings.
+				 * If the rest of the core is not running a tagged
+				 * task, i.e.  need_sync == 0, and the current CPU
+				 * which called into the schedule() loop does not
+				 * have any tasks for this class, skip selecting for
+				 * other siblings since there's no point. We don't skip
+				 * for RT/DL because that could make CFS force-idle RT.
+				 */
+				if (i == cpu && !need_sync && class == &fair_sched_class)
+					goto next_class;
+
+				continue;
+			}
+
+			/*
+			 * Optimize the 'normal' case where there aren't any
+			 * cookies and we don't need to sync up.
+			 */
+			if (i == cpu && !need_sync && !p->core_cookie) {
+				next = p;
+				goto done;
+			}
+
+			rq_i->core_pick = p;
+
+			/*
+			 * If this new candidate is of higher priority than the
+			 * previous; and they're incompatible; we need to wipe
+			 * the slate and start over. pick_task makes sure that
+			 * p's priority is more than max if it doesn't match
+			 * max's cookie.
+			 *
+			 * NOTE: this is a linear max-filter and is thus bounded
+			 * in execution time.
+			 */
+			if (!max || !cookie_match(max, p)) {
+				struct task_struct *old_max = max;
+
+				rq->core->core_cookie = p->core_cookie;
+				max = p;
+
+				if (old_max) {
+					for_each_cpu(j, smt_mask) {
+						if (j == i)
+							continue;
+
+						cpu_rq(j)->core_pick = NULL;
+					}
+					goto again;
+				} else {
+					/*
+					 * Once we select a task for a cpu, we
+					 * should not be doing an unconstrained
+					 * pick because it might starve a task
+					 * on a forced idle cpu.
+					 */
+					need_sync = true;
+				}
+
+			}
+		}
+next_class:;
+	}
+
+	rq->core->core_pick_seq = rq->core->core_task_seq;
+	next = rq->core_pick;
+	rq->core_sched_seq = rq->core->core_pick_seq;
+
+	/* Something should have been selected for current CPU */
+	WARN_ON_ONCE(!next);
+
+	/*
+	 * Reschedule siblings
+	 *
+	 * NOTE: L1TF -- at this point we're no longer running the old task and
+	 * sending an IPI (below) ensures the sibling will no longer be running
+	 * their task. This ensures there is no inter-sibling overlap between
+	 * non-matching user state.
+	 */
+	for_each_cpu(i, smt_mask) {
+		struct rq *rq_i = cpu_rq(i);
+
+		/*
+		 * An online sibling might have gone offline before a task
+		 * could be picked for it, or it might be offline but later
+		 * happen to come online, but its too late and nothing was
+		 * picked for it.  That's Ok - it will pick tasks for itself,
+		 * so ignore it.
+		 */
+		if (!rq_i->core_pick)
+			continue;
+
+		if (is_task_rq_idle(rq_i->core_pick) && rq_i->nr_running)
+			rq_i->core_forceidle = true;
+
+		if (i == cpu) {
+			rq_i->core_pick = NULL;
+			continue;
+		}
+
+		/* Did we break L1TF mitigation requirements? */
+		WARN_ON_ONCE(!cookie_match(next, rq_i->core_pick));
+
+		if (rq_i->curr == rq_i->core_pick) {
+			rq_i->core_pick = NULL;
+			continue;
+		}
+
+		resched_curr(rq_i);
+	}
+
+done:
+	set_next_task(rq, next);
+	return next;
+}
 
 static inline void sched_core_cpu_starting(unsigned int cpu)
 {
@@ -5354,6 +5642,12 @@ static inline void sched_core_cpu_starting(unsigned int cpu)
 
 static inline void sched_core_cpu_starting(unsigned int cpu) {}
 
+static struct task_struct *
+pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
+{
+	return __pick_next_task(rq, prev, rf);
+}
+
 #endif /* CONFIG_SCHED_CORE */
 
 /*
@@ -8609,7 +8903,12 @@ void __init sched_init(void)
 
 #ifdef CONFIG_SCHED_CORE
 		rq->core = NULL;
+		rq->core_pick = NULL;
 		rq->core_enabled = 0;
+		rq->core_tree = RB_ROOT;
+		rq->core_forceidle = false;
+
+		rq->core_cookie = 0UL;
 #endif
 	}
 
kernel/sched/sched.h +5 -1
index e43a217..dd44a31 100644
@@ -1079,11 +1079,16 @@ struct rq {
 #ifdef CONFIG_SCHED_CORE
 	/* per rq */
 	struct rq		*core;
+	struct task_struct	*core_pick;
 	unsigned int		core_enabled;
+	unsigned int		core_sched_seq;
 	struct rb_root		core_tree;
+	unsigned char		core_forceidle;
 
 	/* shared state */
 	unsigned int		core_task_seq;
+	unsigned int		core_pick_seq;
+	unsigned long		core_cookie;
 #endif
 };
 
@@ -2060,7 +2065,6 @@ static inline void put_prev_task(struct rq *rq, struct task_struct *prev)
 
 static inline void set_next_task(struct rq *rq, struct task_struct *next)
 {
-	WARN_ON_ONCE(rq->curr != next);
 	next->sched_class->set_next_task(rq, next, false);
 }
 
author maintainer
#57 lore raw

PATCH 11/19 sched/fair: Fix forced idle sibling starvation corner case

In reply to Peter Zijlstra

From: Vineeth Pillai <[email protected]>

If there is only one long running local task and the sibling is forced idle, it might not get a chance to run until a schedule event happens on any cpu in the core.

So we check for this condition during a tick to see if a sibling is starved and then give it a chance to schedule.

---
 kernel/sched/core.c  | 15 ++++++++-------
 kernel/sched/fair.c  | 40 ++++++++++++++++++++++++++++++++++++++++
 kernel/sched/sched.h |  2 +-
 3 files changed, 49 insertions(+), 8 deletions(-)
kernel/sched/core.c +8 -7
index 1bd0b0bbb040..52d0e83072a4 100644
@@ -5206,16 +5206,15 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 
 	/* reset state */
 	rq->core->core_cookie = 0UL;
+	if (rq->core->core_forceidle) {
+		need_sync = true;
+		rq->core->core_forceidle = false;
+	}
 	for_each_cpu(i, smt_mask) {
 		struct rq *rq_i = cpu_rq(i);
 
 		rq_i->core_pick = NULL;
 
-		if (rq_i->core_forceidle) {
-			need_sync = true;
-			rq_i->core_forceidle = false;
-		}
-
 		if (i != cpu)
 			update_rq_clock(rq_i);
 	}
@@ -5335,8 +5334,10 @@ next_class:;
 		if (!rq_i->core_pick)
 			continue;
 
-		if (is_task_rq_idle(rq_i->core_pick) && rq_i->nr_running)
-			rq_i->core_forceidle = true;
+		if (is_task_rq_idle(rq_i->core_pick) && rq_i->nr_running &&
+		    !rq_i->core->core_forceidle) {
+			rq_i->core->core_forceidle = true;
+		}
 
 		if (i == cpu) {
 			rq_i->core_pick = NULL;
kernel/sched/fair.c +40 -0
index f53681cd263e..42965c4fd71f 100644
@@ -10692,6 +10692,44 @@ static void rq_offline_fair(struct rq *rq)
 
 #endif /* CONFIG_SMP */
 
+#ifdef CONFIG_SCHED_CORE
+static inline bool
+__entity_slice_used(struct sched_entity *se, int min_nr_tasks)
+{
+	u64 slice = sched_slice(cfs_rq_of(se), se);
+	u64 rtime = se->sum_exec_runtime - se->prev_sum_exec_runtime;
+
+	return (rtime * min_nr_tasks > slice);
+}
+
+#define MIN_NR_TASKS_DURING_FORCEIDLE	2
+static inline void task_tick_core(struct rq *rq, struct task_struct *curr)
+{
+	if (!sched_core_enabled(rq))
+		return;
+
+	/*
+	 * If runqueue has only one task which used up its slice and
+	 * if the sibling is forced idle, then trigger schedule to
+	 * give forced idle task a chance.
+	 *
+	 * sched_slice() considers only this active rq and it gets the
+	 * whole slice. But during force idle, we have siblings acting
+	 * like a single runqueue and hence we need to consider runnable
+	 * tasks on this cpu and the forced idle cpu. Ideally, we should
+	 * go through the forced idle rq, but that would be a perf hit.
+	 * We can assume that the forced idle cpu has atleast
+	 * MIN_NR_TASKS_DURING_FORCEIDLE - 1 tasks and use that to check
+	 * if we need to give up the cpu.
+	 */
+	if (rq->core->core_forceidle && rq->cfs.nr_running == 1 &&
+	    __entity_slice_used(&curr->se, MIN_NR_TASKS_DURING_FORCEIDLE))
+		resched_curr(rq);
+}
+#else
+static inline void task_tick_core(struct rq *rq, struct task_struct *curr) {}
+#endif
+
 /*
  * scheduler tick hitting a task of our scheduling class.
  *
@@ -10715,6 +10753,8 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued)
 
 	update_misfit_status(curr, rq);
 	update_overutilized_status(task_rq(curr));
+
+	task_tick_core(rq, curr);
 }
 
 /*
kernel/sched/sched.h +1 -1
index 63b28e1843ee..be656ca8693d 100644
@@ -1069,12 +1069,12 @@ struct rq {
 	unsigned int		core_enabled;
 	unsigned int		core_sched_seq;
 	struct rb_root		core_tree;
-	unsigned char		core_forceidle;
 
 	/* shared state */
 	unsigned int		core_task_seq;
 	unsigned int		core_pick_seq;
 	unsigned long		core_cookie;
+	unsigned char		core_forceidle;
 #endif
 };
 
Signature · 1 line
2.29.2.299.gdc1121823c-goog
merge bot Applied
#58 lore raw

[tip: sched/core] sched/fair: Fix forced idle sibling starvation corner case

The following commit has been merged into the sched/core branch of tip:

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     8039e96fcc1de30d5bcaf05da9ca2de46a800826
Gitweb:        https://git.kernel.org/tip/8039e96fcc1de30d5bcaf05da9ca2de46a800826
Author:        Vineeth Pillai <[email protected]>
AuthorDate:    Tue, 17 Nov 2020 18:19:38 -05:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 12 May 2021 11:43:29 +02:00

sched/fair: Fix forced idle sibling starvation corner case

If there is only one long running local task and the sibling is forced idle, it might not get a chance to run until a schedule event happens on any cpu in the core.

So we check for this condition during a tick to see if a sibling is starved and then give it a chance to schedule.

---
 kernel/sched/core.c  | 15 ++++++++-------
 kernel/sched/fair.c  | 40 ++++++++++++++++++++++++++++++++++++++++
 kernel/sched/sched.h |  2 +-
 3 files changed, 49 insertions(+), 8 deletions(-)
kernel/sched/core.c +8 -7
index db763f4..f5e1e6f 100644
@@ -5459,16 +5459,15 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 
 	/* reset state */
 	rq->core->core_cookie = 0UL;
+	if (rq->core->core_forceidle) {
+		need_sync = true;
+		rq->core->core_forceidle = false;
+	}
 	for_each_cpu(i, smt_mask) {
 		struct rq *rq_i = cpu_rq(i);
 
 		rq_i->core_pick = NULL;
 
-		if (rq_i->core_forceidle) {
-			need_sync = true;
-			rq_i->core_forceidle = false;
-		}
-
 		if (i != cpu)
 			update_rq_clock(rq_i);
 	}
@@ -5588,8 +5587,10 @@ next_class:;
 		if (!rq_i->core_pick)
 			continue;
 
-		if (is_task_rq_idle(rq_i->core_pick) && rq_i->nr_running)
-			rq_i->core_forceidle = true;
+		if (is_task_rq_idle(rq_i->core_pick) && rq_i->nr_running &&
+		    !rq_i->core->core_forceidle) {
+			rq_i->core->core_forceidle = true;
+		}
 
 		if (i == cpu) {
 			rq_i->core_pick = NULL;
kernel/sched/fair.c +40 -0
index 08be7a2..4d1ecab 100644
@@ -10767,6 +10767,44 @@ static void rq_offline_fair(struct rq *rq)
 
 #endif /* CONFIG_SMP */
 
+#ifdef CONFIG_SCHED_CORE
+static inline bool
+__entity_slice_used(struct sched_entity *se, int min_nr_tasks)
+{
+	u64 slice = sched_slice(cfs_rq_of(se), se);
+	u64 rtime = se->sum_exec_runtime - se->prev_sum_exec_runtime;
+
+	return (rtime * min_nr_tasks > slice);
+}
+
+#define MIN_NR_TASKS_DURING_FORCEIDLE	2
+static inline void task_tick_core(struct rq *rq, struct task_struct *curr)
+{
+	if (!sched_core_enabled(rq))
+		return;
+
+	/*
+	 * If runqueue has only one task which used up its slice and
+	 * if the sibling is forced idle, then trigger schedule to
+	 * give forced idle task a chance.
+	 *
+	 * sched_slice() considers only this active rq and it gets the
+	 * whole slice. But during force idle, we have siblings acting
+	 * like a single runqueue and hence we need to consider runnable
+	 * tasks on this cpu and the forced idle cpu. Ideally, we should
+	 * go through the forced idle rq, but that would be a perf hit.
+	 * We can assume that the forced idle cpu has atleast
+	 * MIN_NR_TASKS_DURING_FORCEIDLE - 1 tasks and use that to check
+	 * if we need to give up the cpu.
+	 */
+	if (rq->core->core_forceidle && rq->cfs.nr_running == 1 &&
+	    __entity_slice_used(&curr->se, MIN_NR_TASKS_DURING_FORCEIDLE))
+		resched_curr(rq);
+}
+#else
+static inline void task_tick_core(struct rq *rq, struct task_struct *curr) {}
+#endif
+
 /*
  * scheduler tick hitting a task of our scheduling class.
  *
@@ -10790,6 +10828,8 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued)
 
 	update_misfit_status(curr, rq);
 	update_overutilized_status(task_rq(curr));
+
+	task_tick_core(rq, curr);
 }
 
 /*
kernel/sched/sched.h +1 -1
index dd44a31..db55514 100644
@@ -1083,12 +1083,12 @@ struct rq {
 	unsigned int		core_enabled;
 	unsigned int		core_sched_seq;
 	struct rb_root		core_tree;
-	unsigned char		core_forceidle;
 
 	/* shared state */
 	unsigned int		core_task_seq;
 	unsigned int		core_pick_seq;
 	unsigned long		core_cookie;
+	unsigned char		core_forceidle;
 #endif
 };
 
author maintainer
#59 lore raw

PATCH 12/19 sched: Fix priority inversion of cookied task with sibling

In reply to Peter Zijlstra

From: "Joel Fernandes (Google)" <[email protected]>

The rationale is as follows. In the core-wide pick logic, even if need_sync == false, we need to go look at other CPUs (non-local CPUs) to see if they could be running RT.

Say the RQs in a particular core look like this:

Let CFS1 and CFS2 be 2 tagged CFS tags.
Let RT1 be an untagged RT task.

	rq0		rq1
	CFS1 (tagged)	RT1 (no tag)
	CFS2 (tagged)

Say schedule() runs on rq0. Now, it will enter the above loop and pick_task(RT) will return NULL for 'p'. It will enter the above if() block and see that need_sync == false and will skip RT entirely.

The end result of the selection will be (say prio(CFS1) > prio(CFS2)):

rq0 rq1
CFS1 IDLE

When it should have selected:

rq0 rq1
IDLE RT

---
 kernel/sched/core.c |   65 ++++++++++++++++++++--------------------------------
 1 file changed, 26 insertions(+), 39 deletions(-)
kernel/sched/core.c +26 -39
@@ -5425,6 +5425,15 @@ pick_next_task(struct rq *rq, struct tas
 	put_prev_task_balance(rq, prev, rf);
 
 	smt_mask = cpu_smt_mask(cpu);
+	need_sync = !!rq->core->core_cookie;
+
+	/* reset state */
+	rq->core->core_cookie = 0UL;
+	if (rq->core->core_forceidle) {
+		need_sync = true;
+		fi_before = true;
+		rq->core->core_forceidle = false;
+	}
 
 	/*
 	 * core->core_task_seq, core->core_pick_seq, rq->core_sched_seq
@@ -5437,14 +5446,25 @@ pick_next_task(struct rq *rq, struct tas
 	 * 'Fix' this by also increasing @task_seq for every pick.
 	 */
 	rq->core->core_task_seq++;
-	need_sync = !!rq->core->core_cookie;
 
-	/* reset state */
-	rq->core->core_cookie = 0UL;
-	if (rq->core->core_forceidle) {
+	/*
+	 * Optimize for common case where this CPU has no cookies
+	 * and there are no cookied tasks running on siblings.
+	 */
+	if (!need_sync) {
+		for_each_class(class) {
+			next = class->pick_task(rq);
+			if (next)
+				break;
+		}
+
+		if (!next->core_cookie) {
+			rq->core_pick = NULL;
+			goto done;
+		}
 		need_sync = true;
-		rq->core->core_forceidle = false;
 	}
+
 	for_each_cpu(i, smt_mask) {
 		struct rq *rq_i = cpu_rq(i);
 
@@ -5474,31 +5494,8 @@ pick_next_task(struct rq *rq, struct tas
 			 * core.
 			 */
 			p = pick_task(rq_i, class, max);
-			if (!p) {
-				/*
-				 * If there weren't no cookies; we don't need to
-				 * bother with the other siblings.
-				 * If the rest of the core is not running a tagged
-				 * task, i.e.  need_sync == 0, and the current CPU
-				 * which called into the schedule() loop does not
-				 * have any tasks for this class, skip selecting for
-				 * other siblings since there's no point. We don't skip
-				 * for RT/DL because that could make CFS force-idle RT.
-				 */
-				if (i == cpu && !need_sync && class == &fair_sched_class)
-					goto next_class;
-
+			if (!p)
 				continue;
-			}
-
-			/*
-			 * Optimize the 'normal' case where there aren't any
-			 * cookies and we don't need to sync up.
-			 */
-			if (i == cpu && !need_sync && !p->core_cookie) {
-				next = p;
-				goto done;
-			}
 
 			rq_i->core_pick = p;
 
@@ -5526,19 +5523,9 @@ pick_next_task(struct rq *rq, struct tas
 						cpu_rq(j)->core_pick = NULL;
 					}
 					goto again;
-				} else {
-					/*
-					 * Once we select a task for a cpu, we
-					 * should not be doing an unconstrained
-					 * pick because it might starve a task
-					 * on a forced idle cpu.
-					 */
-					need_sync = true;
 				}
-
 			}
 		}
-next_class:;
 	}
 
 	rq->core->core_pick_seq = rq->core->core_task_seq;
#60 lore raw

[tip: sched/core] sched: Fix priority inversion of cookied task with sibling

The following commit has been merged into the sched/core branch of tip:

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     7afbba119f0da09824d723f8081608ea1f74ff57
Gitweb:        https://git.kernel.org/tip/7afbba119f0da09824d723f8081608ea1f74ff57
Author:        Joel Fernandes (Google) <[email protected]>
AuthorDate:    Tue, 17 Nov 2020 18:19:42 -05:00
Committer:     Peter Zijlstra <[email protected]>
CommitterDate: Wed, 12 May 2021 11:43:29 +02:00

sched: Fix priority inversion of cookied task with sibling

The rationale is as follows. In the core-wide pick logic, even if need_sync == false, we need to go look at other CPUs (non-local CPUs) to see if they could be running RT.

Say the RQs in a particular core look like this:

Let CFS1 and CFS2 be 2 tagged CFS tags.
Let RT1 be an untagged RT task.

	rq0		rq1
	CFS1 (tagged)	RT1 (no tag)
	CFS2 (tagged)

Say schedule() runs on rq0. Now, it will enter the above loop and pick_task(RT) will return NULL for 'p'. It will enter the above if() block and see that need_sync == false and will skip RT entirely.

The end result of the selection will be (say prio(CFS1) > prio(CFS2)):

rq0 rq1
CFS1 IDLE

When it should have selected:

rq0 rq1
IDLE RT

---
 kernel/sched/core.c | 65 +++++++++++++++++---------------------------
 1 file changed, 26 insertions(+), 39 deletions(-)
kernel/sched/core.c +26 -39
index f5e1e6f..e506d9d 100644
@@ -5443,6 +5443,15 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 	put_prev_task_balance(rq, prev, rf);
 
 	smt_mask = cpu_smt_mask(cpu);
+	need_sync = !!rq->core->core_cookie;
+
+	/* reset state */
+	rq->core->core_cookie = 0UL;
+	if (rq->core->core_forceidle) {
+		need_sync = true;
+		fi_before = true;
+		rq->core->core_forceidle = false;
+	}
 
 	/*
 	 * core->core_task_seq, core->core_pick_seq, rq->core_sched_seq
@@ -5455,14 +5464,25 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 	 * 'Fix' this by also increasing @task_seq for every pick.
 	 */
 	rq->core->core_task_seq++;
-	need_sync = !!rq->core->core_cookie;
 
-	/* reset state */
-	rq->core->core_cookie = 0UL;
-	if (rq->core->core_forceidle) {
+	/*
+	 * Optimize for common case where this CPU has no cookies
+	 * and there are no cookied tasks running on siblings.
+	 */
+	if (!need_sync) {
+		for_each_class(class) {
+			next = class->pick_task(rq);
+			if (next)
+				break;
+		}
+
+		if (!next->core_cookie) {
+			rq->core_pick = NULL;
+			goto done;
+		}
 		need_sync = true;
-		rq->core->core_forceidle = false;
 	}
+
 	for_each_cpu(i, smt_mask) {
 		struct rq *rq_i = cpu_rq(i);
 
@@ -5492,31 +5512,8 @@ again:
 			 * core.
 			 */
 			p = pick_task(rq_i, class, max);
-			if (!p) {
-				/*
-				 * If there weren't no cookies; we don't need to
-				 * bother with the other siblings.
-				 * If the rest of the core is not running a tagged
-				 * task, i.e.  need_sync == 0, and the current CPU
-				 * which called into the schedule() loop does not
-				 * have any tasks for this class, skip selecting for
-				 * other siblings since there's no point. We don't skip
-				 * for RT/DL because that could make CFS force-idle RT.
-				 */
-				if (i == cpu && !need_sync && class == &fair_sched_class)
-					goto next_class;
-
+			if (!p)
 				continue;
-			}
-
-			/*
-			 * Optimize the 'normal' case where there aren't any
-			 * cookies and we don't need to sync up.
-			 */
-			if (i == cpu && !need_sync && !p->core_cookie) {
-				next = p;
-				goto done;
-			}
 
 			rq_i->core_pick = p;
 
@@ -5544,19 +5541,9 @@ again:
 						cpu_rq(j)->core_pick = NULL;
 					}
 					goto again;
-				} else {
-					/*
-					 * Once we select a task for a cpu, we
-					 * should not be doing an unconstrained
-					 * pick because it might starve a task
-					 * on a forced idle cpu.
-					 */
-					need_sync = true;
 				}
-
 			}
 		}
-next_class:;
 	}
 
 	rq->core->core_pick_seq = rq->core->core_task_seq;
author maintainer
#61 lore raw

PATCH 13/19 sched/fair: Snapshot the min_vruntime of CPUs on force idle

In reply to Peter Zijlstra

From: "Joel Fernandes (Google)" <[email protected]>

During force-idle, we end up doing cross-cpu comparison of vruntimes during pick_next_task. If we simply compare (vruntime-min_vruntime) across CPUs, and if the CPUs only have 1 task each, we will always end up comparing 0 with 0 and pick just one of the tasks all the time. This starves the task that was not picked. To fix this, take a snapshot of the min_vruntime when entering force idle and use it for comparison. This min_vruntime snapshot will only be used for cross-CPU vruntime comparison, and nothing else.

A note about the min_vruntime snapshot and force idling:

During selection:

When we're not fi, we need to update snapshot.
when we're fi and we were not fi, we must update snapshot. When we're fi and we were already fi, we must not update snapshot.

Which gives:

  fib     fi      update
  0       0       1
  0       1       1
  1       0       1
  1       1       0

Where:

fi: force-idled now
fib: force-idled before

So the min_vruntime snapshot needs to be updated when: !(fib && fi).

Also, the cfs_prio_less() function needs to be aware of whether the core is in force idle or not, since it will be use this information to know whether to advance a cfs_rq's min_vruntime_fi in the hierarchy. So pass this information along via pick_task() -> prio_less().

---
 kernel/sched/core.c  |   59 +++++++++++++++++++++++-----------------
 kernel/sched/fair.c  |   75 +++++++++++++++++++++++++++++++++++++++++++++++++++
 kernel/sched/sched.h |    8 +++++
 3 files changed, 117 insertions(+), 25 deletions(-)
kernel/sched/core.c +34 -25
@@ -101,7 +101,7 @@ static inline int __task_prio(struct tas
  */
 
 /* real prio, less is less */
-static inline bool prio_less(struct task_struct *a, struct task_struct *b)
+static inline bool prio_less(struct task_struct *a, struct task_struct *b, bool in_fi)
 {
 
 	int pa = __task_prio(a), pb = __task_prio(b);
@@ -115,19 +115,8 @@ static inline bool prio_less(struct task
 	if (pa == -1) /* dl_prio() doesn't work because of stop_class above */
 		return !dl_time_before(a->dl.deadline, b->dl.deadline);
 
-	if (pa == MAX_RT_PRIO + MAX_NICE)  { /* fair */
-		u64 vruntime = b->se.vruntime;
-
-		/*
-		 * Normalize the vruntime if tasks are in different cpus.
-		 */
-		if (task_cpu(a) != task_cpu(b)) {
-			vruntime -= task_cfs_rq(b)->min_vruntime;
-			vruntime += task_cfs_rq(a)->min_vruntime;
-		}
-
-		return !((s64)(a->se.vruntime - vruntime) <= 0);
-	}
+	if (pa == MAX_RT_PRIO + MAX_NICE)	/* fair */
+		return cfs_prio_less(a, b, in_fi);
 
 	return false;
 }
@@ -141,7 +130,7 @@ static inline bool __sched_core_less(str
 		return false;
 
 	/* flip prio, so high prio is leftmost */
-	if (prio_less(b, a))
+	if (prio_less(b, a, task_rq(a)->core->core_forceidle))
 		return true;
 
 	return false;
@@ -5145,7 +5134,7 @@ static inline bool cookie_match(struct t
  * - Else returns idle_task.
  */
 static struct task_struct *
-pick_task(struct rq *rq, const struct sched_class *class, struct task_struct *max)
+pick_task(struct rq *rq, const struct sched_class *class, struct task_struct *max, bool in_fi)
 {
 	struct task_struct *class_pick, *cookie_pick;
 	unsigned long cookie = rq->core->core_cookie;
@@ -5160,7 +5149,7 @@ pick_task(struct rq *rq, const struct sc
 		 * higher priority than max.
 		 */
 		if (max && class_pick->core_cookie &&
-		    prio_less(class_pick, max))
+		    prio_less(class_pick, max, in_fi))
 			return idle_sched_class.pick_task(rq);
 
 		return class_pick;
@@ -5179,19 +5168,22 @@ pick_task(struct rq *rq, const struct sc
 	 * the core (so far) and it must be selected, otherwise we must go with
 	 * the cookie pick in order to satisfy the constraint.
 	 */
-	if (prio_less(cookie_pick, class_pick) &&
-	    (!max || prio_less(max, class_pick)))
+	if (prio_less(cookie_pick, class_pick, in_fi) &&
+	    (!max || prio_less(max, class_pick, in_fi)))
 		return class_pick;
 
 	return cookie_pick;
 }
 
+extern void task_vruntime_update(struct rq *rq, struct task_struct *p, bool in_fi);
+
 static struct task_struct *
 pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf)
 {
 	struct task_struct *next, *max = NULL;
 	const struct sched_class *class;
 	const struct cpumask *smt_mask;
+	bool fi_before = false;
 	bool need_sync;
 	int i, j, cpu;
 
@@ -5273,9 +5265,14 @@ pick_next_task(struct rq *rq, struct tas
 
 		if (!next->core_cookie) {
 			rq->core_pick = NULL;
+			/*
+			 * For robustness, update the min_vruntime_fi for
+			 * unconstrained picks as well.
+			 */
+			WARN_ON_ONCE(fi_before);
+			task_vruntime_update(rq, next, false);
 			goto done;
 		}
-		need_sync = true;
 	}
 
 	for_each_cpu(i, smt_mask) {
@@ -5306,11 +5303,16 @@ pick_next_task(struct rq *rq, struct tas
 			 * highest priority task already selected for this
 			 * core.
 			 */
-			p = pick_task(rq_i, class, max);
+			p = pick_task(rq_i, class, max, fi_before);
 			if (!p)
 				continue;
 
 			rq_i->core_pick = p;
+			if (rq_i->idle == p && rq_i->nr_running) {
+				rq->core->core_forceidle = true;
+				if (!fi_before)
+					rq->core->core_forceidle_seq++;
+			}
 
 			/*
 			 * If this new candidate is of higher priority than the
@@ -5329,6 +5331,7 @@ pick_next_task(struct rq *rq, struct tas
 				max = p;
 
 				if (old_max) {
+					rq->core->core_forceidle = false;
 					for_each_cpu(j, smt_mask) {
 						if (j == i)
 							continue;
@@ -5369,10 +5372,16 @@ pick_next_task(struct rq *rq, struct tas
 		if (!rq_i->core_pick)
 			continue;
 
-		if (is_task_rq_idle(rq_i->core_pick) && rq_i->nr_running &&
-		    !rq_i->core->core_forceidle) {
-			rq_i->core->core_forceidle = true;
-		}
+		/*
+		 * Update for new !FI->FI transitions, or if continuing to be in !FI:
+		 * fi_before     fi      update?
+		 *  0            0       1
+		 *  0            1       1
+		 *  1            0       1
+		 *  1            1       0
+		 */
+		if (!(fi_before && rq->core->core_forceidle))
+			task_vruntime_update(rq_i, rq_i->core_pick, rq->core->core_forceidle);
 
 		if (i == cpu) {
 			rq_i->core_pick = NULL;
kernel/sched/fair.c +75 -0

81 lines not on this page · raw message

kernel/sched/sched.h +8 -0
@@ -517,6 +517,11 @@ struct cfs_rq {
 
 	u64			exec_clock;
 	u64			min_vruntime;
+#ifdef CONFIG_SCHED_CORE
+	unsigned int		forceidle_seq;
+	u64			min_vruntime_fi;
+#endif
+
 #ifndef CONFIG_64BIT
 	u64			min_vruntime_copy;
 #endif
@@ -1075,6 +1080,7 @@ struct rq {
 	unsigned int		core_pick_seq;
 	unsigned long		core_cookie;
 	unsigned char		core_forceidle;
+	unsigned int		core_forceidle_seq;
 #endif
 };
 
@@ -1130,6 +1136,8 @@ static inline raw_spinlock_t *rq_lockp(s
 	return &rq->__lock;
 }
 
+bool cfs_prio_less(struct task_struct *a, struct task_struct *b, bool fi);
+
 #else /* !CONFIG_SCHED_CORE */
 
 static inline bool sched_core_enabled(struct rq *rq)