Skip to contentSkip to conversation

affine kernel threads to nohz_full= cpumask (v4)

PATCH In review Marcelo Tosatti 3 participants 8 messages Apr 1, 2020 – Apr 20, 2020

Subsystems (2)
lore
Share & tools
lore
#1 lore raw

PATCH 1/4 kthread: switch to cpu_possible_mask

In reply to Marcelo Tosatti

Next patch will switch unbound kernel threads mask to housekeeping_cpumask(), a subset of cpu_possible_mask.

Switch from cpu_all_mask to cpu_possible_mask separately, to ease bisection.

---
 kernel/kthread.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
kernel/kthread.c +2 -2
Index: linux-2.6/kernel/kthread.c
===================================================================
@@ -347,7 +347,7 @@ struct task_struct *__kthread_create_on_
 		 * The kernel thread should not inherit these properties.
 		 */
 		sched_setscheduler_nocheck(task, SCHED_NORMAL, &param);
-		set_cpus_allowed_ptr(task, cpu_all_mask);
+		set_cpus_allowed_ptr(task, cpu_possible_mask);
 	}
 	kfree(create);
 	return task;
@@ -572,7 +572,7 @@ int kthreadd(void *unused)
 	/* Setup a clean context for our children to inherit. */
 	set_task_comm(tsk, "kthreadd");
 	ignore_signals(tsk);
-	set_cpus_allowed_ptr(tsk, cpu_all_mask);
+	set_cpus_allowed_ptr(tsk, cpu_possible_mask);
 	set_mems_allowed(node_states[N_MEMORY]);
 
 	current->flags |= PF_NOFREEZE;
#2 lore raw

PATCH 2/4 isolation: set HK_FLAG_SCHED on nohz_full CPUs

In reply to Marcelo Tosatti

Avoid idle load balancing on nohz_full CPUs.

This avoids assigning tasks to such CPUs, when they enter idle.

kernel/sched/isolation.c +2 -1
Index: linux-2.6/kernel/sched/isolation.c
===================================================================
@@ -140,7 +140,8 @@ static int __init housekeeping_nohz_full
 {
 	unsigned int flags;
 
-	flags = HK_FLAG_TICK | HK_FLAG_WQ | HK_FLAG_TIMER | HK_FLAG_RCU | HK_FLAG_MISC;
+	flags = HK_FLAG_TICK | HK_FLAG_WQ | HK_FLAG_TIMER | HK_FLAG_RCU |
+		HK_FLAG_MISC | HK_FLAG_SCHED;
 
 	return housekeeping_setup(str, flags);
 }
#3 lore raw

PATCH 3/4 isolcpus: affine kernel threads to housekeeping cpus

In reply to Marcelo Tosatti

This is a kernel enhancement that configures the cpu affinity of kernel threads via kernel boot option nohz_full=.

When this option is specified, the cpumask is immediately applied upon thread launch. This does not affect kernel threads that specify cpu and node.

This allows CPU isolation (that is not allowing certain threads to execute on certain CPUs) without using the isolcpus=domain parameter, making it possible to enable load balancing on such CPUs during runtime (see kernel-parameters.txt).

Note-1: this is based off on Wind River's patch at
https://github.com/starlingx-staging/stx-integ/blob/master/kernel/kernel-std/centos/patches/affine-compute-kernel-threads.patch

Difference being that this patch is limited to modifying kernel thread cpumask: Behaviour of other threads can be controlled via cgroups or sched_setaffinity.

Note-2: Wind River's patch was based off Christoph Lameter's patch at
https://lwn.net/Articles/565932/ with the only difference being the kernel parameter changed from kthread to kthread_cpus.

---
 Documentation/admin-guide/kernel-parameters.txt |    8 ++++++++
 include/linux/sched/isolation.h                 |    1 +
 kernel/kthread.c                                |    6 ++++--
 kernel/sched/isolation.c                        |    6 ++++++
 4 files changed, 19 insertions(+), 2 deletions(-)
include/linux/sched/isolation.h +1 -0
Index: linux-2.6/include/linux/sched/isolation.h
===================================================================
@@ -14,6 +14,7 @@ enum hk_flags {
 	HK_FLAG_DOMAIN		= (1 << 5),
 	HK_FLAG_WQ		= (1 << 6),
 	HK_FLAG_MANAGED_IRQ	= (1 << 7),
+	HK_FLAG_KTHREAD		= (1 << 8),
 };
 
 #ifdef CONFIG_CPU_ISOLATION
kernel/kthread.c +4 -2
Index: linux-2.6/kernel/kthread.c
===================================================================
@@ -23,6 +23,7 @@
 #include <linux/ptrace.h>
 #include <linux/uaccess.h>
 #include <linux/numa.h>
+#include <linux/sched/isolation.h>
 #include <trace/events/sched.h>
 
 static DEFINE_SPINLOCK(kthread_create_lock);
@@ -347,7 +348,8 @@ struct task_struct *__kthread_create_on_
 		 * The kernel thread should not inherit these properties.
 		 */
 		sched_setscheduler_nocheck(task, SCHED_NORMAL, &param);
-		set_cpus_allowed_ptr(task, cpu_possible_mask);
+		set_cpus_allowed_ptr(task,
+				     housekeeping_cpumask(HK_FLAG_KTHREAD));
 	}
 	kfree(create);
 	return task;
@@ -572,7 +574,7 @@ int kthreadd(void *unused)
 	/* Setup a clean context for our children to inherit. */
 	set_task_comm(tsk, "kthreadd");
 	ignore_signals(tsk);
-	set_cpus_allowed_ptr(tsk, cpu_possible_mask);
+	set_cpus_allowed_ptr(tsk, housekeeping_cpumask(HK_FLAG_KTHREAD));
 	set_mems_allowed(node_states[N_MEMORY]);
 
 	current->flags |= PF_NOFREEZE;
kernel/sched/isolation.c +1 -1
Index: linux-2.6/kernel/sched/isolation.c
===================================================================
@@ -141,7 +141,7 @@ static int __init housekeeping_nohz_full
 	unsigned int flags;
 
 	flags = HK_FLAG_TICK | HK_FLAG_WQ | HK_FLAG_TIMER | HK_FLAG_RCU |
-		HK_FLAG_MISC | HK_FLAG_SCHED;
+		HK_FLAG_MISC | HK_FLAG_SCHED | HK_FLAG_KTHREAD;
 
 	return housekeeping_setup(str, flags);
 }
#6 lore raw

PATCH 4/4 isolcpus: undeprecate on documentation

In reply to Marcelo Tosatti

isolcpus is used to control steering of interrupts to managed_irqs and kernel threads, therefore its incorrect to state that its deprecated.

Remove deprecation warning.

---
 Documentation/admin-guide/kernel-parameters.txt |    1 -
 1 file changed, 1 deletion(-)
Documentation/admin-guide/kernel-parameters.txt +0 -1
Index: linux-2.6/Documentation/admin-guide/kernel-parameters.txt
===================================================================
@@ -1926,7 +1926,6 @@
 			Format: <RDP>,<reset>,<pci_scan>,<verbosity>
 
 	isolcpus=	[KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance.
-			[Deprecated - use cpusets instead]
 			Format: [flag-list,]<cpu-list>
 
 			Specify one or more CPUs to isolate from disturbances
#0 lore raw

cover affine kernel threads to nohz_full= cpumask (v4)

This is a kernel enhancement that configures the cpu affinity of kernel threads via kernel boot option nohz_full=.

When this option is specified, the cpumask is immediately applied upon thread launch. This does not affect kernel threads that specify cpu and node.

This allows CPU isolation (that is not allowing certain threads to execute on certain CPUs) without using the isolcpus=domain parameter, making it possible to enable load balancing on such CPUs during runtime (see kernel-parameters.txt).

Note-1: this is based off on Wind River's patch at
https://github.com/starlingx-staging/stx-integ/blob/master/kernel/kernel-std/centos/patches/affine-compute-kernel-threads.patch

Difference being that this patch is limited to modifying kernel thread cpumask: Behaviour of other threads can be controlled via cgroups or sched_setaffinity.

Note-2: Wind River's patch was based off Christoph Lameter's patch at
https://lwn.net/Articles/565932/ with the only difference being the kernel parameter changed from kthread to kthread_cpus.

v2: use isolcpus= subcommand (Thomas Gleixner)

v3: s/MontaVista/Wind River/ on changelog (Chris Friesen)

    documentation updates                 (Chris Friesen)
    undeprecate isolcpus                  (Chris Friesen)
    general cleanups                      (Frederic Weisbecker)

separate cpu_possible_mask kthread
mask change (Frederic Weisbecker)

v4: disable idle load balancing for nohz_full=
use nohz_full= option for kthread isolation (Frederic Weisbecker)

reviewer
#4 lore raw

Re: [patch 3/4] isolcpus: affine kernel threads to housekeeping cpus

In reply to Marcelo Tosatti
Quoted 19 lines · Marcelo Tosatti

This is a kernel enhancement that configures the cpu affinity of kernel threads via kernel boot option nohz_full=.

When this option is specified, the cpumask is immediately applied upon thread launch. This does not affect kernel threads that specify cpu and node.

This allows CPU isolation (that is not allowing certain threads to execute on certain CPUs) without using the isolcpus=domain parameter, making it possible to enable load balancing on such CPUs during runtime (see kernel-parameters.txt).

Note-1: this is based off on Wind River's patch at
https://github.com/starlingx-staging/stx-integ/blob/master/kernel/kernel-std/centos/patches/affine-compute-kernel-threads.patch

Difference being that this patch is limited to modifying kernel thread cpumask: Behaviour of other threads can be controlled via cgroups or sched_setaffinity.

Quoted 6 lines

Note-2: Wind River's patch was based off Christoph Lameter's patch at
https://lwn.net/Articles/565932/ with the only difference being the kernel parameter changed from kthread to kthread_cpus.

Hi Marcelo,

Quoted 2 lines
---
 Documentation/admin-guide/kernel-parameters.txt |    8 ++++++++

Patch is missing those Docum bits.

Quoted 55 lines
 include/linux/sched/isolation.h                 |    1 +
 kernel/kthread.c                                |    6 ++++--
 kernel/sched/isolation.c                        |    6 ++++++
 4 files changed, 19 insertions(+), 2 deletions(-)

Index: linux-2.6/include/linux/sched/isolation.h
===================================================================

include/linux/sched/isolation.h +1 -0
@@ -14,6 +14,7 @@ enum hk_flags {
 	HK_FLAG_DOMAIN		= (1 << 5),
 	HK_FLAG_WQ		= (1 << 6),
 	HK_FLAG_MANAGED_IRQ	= (1 << 7),
+	HK_FLAG_KTHREAD		= (1 << 8),
 };
 
 #ifdef CONFIG_CPU_ISOLATION

Index: linux-2.6/kernel/kthread.c
===================================================================

kernel/kthread.c +4 -2
@@ -23,6 +23,7 @@
 #include <linux/ptrace.h>
 #include <linux/uaccess.h>
 #include <linux/numa.h>
+#include <linux/sched/isolation.h>
 #include <trace/events/sched.h>
 
 static DEFINE_SPINLOCK(kthread_create_lock);
@@ -347,7 +348,8 @@ struct task_struct *__kthread_create_on_
 		 * The kernel thread should not inherit these properties.
 		 */
 		sched_setscheduler_nocheck(task, SCHED_NORMAL, &param);
-		set_cpus_allowed_ptr(task, cpu_possible_mask);
+		set_cpus_allowed_ptr(task,
+				     housekeeping_cpumask(HK_FLAG_KTHREAD));
 	}
 	kfree(create);
 	return task;
@@ -572,7 +574,7 @@ int kthreadd(void *unused)
 	/* Setup a clean context for our children to inherit. */
 	set_task_comm(tsk, "kthreadd");
 	ignore_signals(tsk);
-	set_cpus_allowed_ptr(tsk, cpu_possible_mask);
+	set_cpus_allowed_ptr(tsk, housekeeping_cpumask(HK_FLAG_KTHREAD));
 	set_mems_allowed(node_states[N_MEMORY]);
 
 	current->flags |= PF_NOFREEZE;

Index: linux-2.6/kernel/sched/isolation.c
===================================================================

kernel/sched/isolation.c static int __init housekeeping_nohz_full · line 147

Quoted 8 lines
kernel/sched/isolation.c +1 -1
@@ -141,7 +141,7 @@ static int __init housekeeping_nohz_full
 	unsigned int flags;
 
 	flags = HK_FLAG_TICK | HK_FLAG_WQ | HK_FLAG_TIMER | HK_FLAG_RCU |
-		HK_FLAG_MISC | HK_FLAG_SCHED;
+		HK_FLAG_MISC | HK_FLAG_SCHED | HK_FLAG_KTHREAD;
 
 	return housekeeping_setup(str, flags);
 }

thanks.

Signature · 1 line
~Randy
#5 lore raw

PATCH v2 3/4 isolcpus: affine kernel threads to housekeeping cpus

In reply to Marcelo Tosatti

This is a kernel enhancement that configures the cpu affinity of kernel threads via kernel boot option nohz_full=.

When this option is specified, the cpumask is immediately applied upon thread launch. This does not affect kernel threads that specify cpu and node.

This allows CPU isolation (that is not allowing certain threads to execute on certain CPUs) without using the isolcpus=domain parameter, making it possible to enable load balancing on such CPUs during runtime (see kernel-parameters.txt).

Note-1: this is based off on Wind River's patch at
https://github.com/starlingx-staging/stx-integ/blob/master/kernel/kernel-std/centos/patches/affine-compute-kernel-threads.patch

Difference being that this patch is limited to modifying kernel thread cpumask: Behaviour of other threads can be controlled via cgroups or sched_setaffinity.

Note-2: Wind River's patch was based off Christoph Lameter's patch at
https://lwn.net/Articles/565932/ with the only difference being the kernel parameter changed from kthread to kthread_cpus.

---
 include/linux/sched/isolation.h                 |    1 +
 kernel/kthread.c                                |    6 ++++--
 kernel/sched/isolation.c                        |    6 ++++++
 4 files changed, 19 insertions(+), 2 deletions(-)
include/linux/sched/isolation.h +1 -0
Index: linux-2.6/include/linux/sched/isolation.h
===================================================================
@@ -14,6 +14,7 @@ enum hk_flags {
 	HK_FLAG_DOMAIN		= (1 << 5),
 	HK_FLAG_WQ		= (1 << 6),
 	HK_FLAG_MANAGED_IRQ	= (1 << 7),
+	HK_FLAG_KTHREAD		= (1 << 8),
 };
 
 #ifdef CONFIG_CPU_ISOLATION
kernel/kthread.c +4 -2
Index: linux-2.6/kernel/kthread.c
===================================================================
@@ -23,6 +23,7 @@
 #include <linux/ptrace.h>
 #include <linux/uaccess.h>
 #include <linux/numa.h>
+#include <linux/sched/isolation.h>
 #include <trace/events/sched.h>
 
 static DEFINE_SPINLOCK(kthread_create_lock);
@@ -347,7 +348,8 @@ struct task_struct *__kthread_create_on_
 		 * The kernel thread should not inherit these properties.
 		 */
 		sched_setscheduler_nocheck(task, SCHED_NORMAL, &param);
-		set_cpus_allowed_ptr(task, cpu_possible_mask);
+		set_cpus_allowed_ptr(task,
+				     housekeeping_cpumask(HK_FLAG_KTHREAD));
 	}
 	kfree(create);
 	return task;
@@ -572,7 +574,7 @@ int kthreadd(void *unused)
 	/* Setup a clean context for our children to inherit. */
 	set_task_comm(tsk, "kthreadd");
 	ignore_signals(tsk);
-	set_cpus_allowed_ptr(tsk, cpu_possible_mask);
+	set_cpus_allowed_ptr(tsk, housekeeping_cpumask(HK_FLAG_KTHREAD));
 	set_mems_allowed(node_states[N_MEMORY]);
 
 	current->flags |= PF_NOFREEZE;
kernel/sched/isolation.c +1 -1
Index: linux-2.6/kernel/sched/isolation.c
===================================================================
@@ -141,7 +141,7 @@ static int __init housekeeping_nohz_full
 	unsigned int flags;
 
 	flags = HK_FLAG_TICK | HK_FLAG_WQ | HK_FLAG_TIMER | HK_FLAG_RCU |
-		HK_FLAG_MISC | HK_FLAG_SCHED;
+		HK_FLAG_MISC | HK_FLAG_SCHED | HK_FLAG_KTHREAD;
 
 	return housekeeping_setup(str, flags);
 }
#7 lore raw
In reply to Marcelo Tosatti
Quoted 26 lines · Marcelo Tosatti

This is a kernel enhancement that configures the cpu affinity of kernel threads via kernel boot option nohz_full=.

When this option is specified, the cpumask is immediately applied upon thread launch. This does not affect kernel threads that specify cpu and node.

This allows CPU isolation (that is not allowing certain threads to execute on certain CPUs) without using the isolcpus=domain parameter, making it possible to enable load balancing on such CPUs during runtime (see kernel-parameters.txt).

Note-1: this is based off on Wind River's patch at
https://github.com/starlingx-staging/stx-integ/blob/master/kernel/kernel-std/centos/patches/affine-compute-kernel-threads.patch

Difference being that this patch is limited to modifying kernel thread cpumask: Behaviour of other threads can be controlled via cgroups or sched_setaffinity.

Note-2: Wind River's patch was based off Christoph Lameter's patch at
https://lwn.net/Articles/565932/ with the only difference being the kernel parameter changed from kthread to kthread_cpus.

v2: use isolcpus= subcommand (Thomas Gleixner)

v3: s/MontaVista/Wind River/ on changelog (Chris Friesen)

Quoted 9 lines
    documentation updates                 (Chris Friesen)
    undeprecate isolcpus                  (Chris Friesen)
    general cleanups                      (Frederic Weisbecker)

separate cpu_possible_mask kthread
mask change (Frederic Weisbecker)

v4: disable idle load balancing for nohz_full=
use nohz_full= option for kthread isolation (Frederic Weisbecker)

Thanks for the patches. I'm applying them, I may add a few details to the changelogs and stuff.