This document lists per-CPU kthreads in the Linux kernel and presents
options to control their OS jitter. Note that non-per-CPU kthreads are
not listed here. To reduce OS jitter from non-per-CPU kthreads, bind
them to a "housekeeping" CPU dedicated to such work.
This document lists per-CPU kthreads in the Linux kernel and presents
options to control their OS jitter. Note that non-per-CPU kthreads are
not listed here. To reduce OS jitter from non-per-CPU kthreads, bind
them to a "housekeeping" CPU dedicated to such work.
1. Don't use eHCA Infiniband hardware, instead choosing hardware
that does not require per-CPU kthreads. This will prevent these
kthreads from being created in the first place. (This will
1. Don't use eHCA Infiniband hardware, instead choosing hardware
that does not require per-CPU kthreads. This will prevent these
kthreads from being created in the first place. (This will
1. Don't use Bluetooth, in which case these kthreads won't be
created in the first place.
2. Use irq affinity to force Bluetooth-related interrupts to
occur on some other CPU and furthermore initiate all
Bluetooth activity on some other CPU.
1. Don't use Bluetooth, in which case these kthreads won't be
created in the first place.
2. Use irq affinity to force Bluetooth-related interrupts to
occur on some other CPU and furthermore initiate all
Bluetooth activity on some other CPU.
1. To the extent possible, keep the CPU out of the kernel when it
is non-idle, for example, by avoiding system calls and by forcing
both kernel threads and interrupts to execute elsewhere.
1. To the extent possible, keep the CPU out of the kernel when it
is non-idle, for example, by avoiding system calls and by forcing
both kernel threads and interrupts to execute elsewhere.
first one back online. Once you have onlined the CPUs in question,
do not offline any other CPUs, because doing so could force the
timer back onto one of the CPUs in question.
first one back online. Once you have onlined the CPUs in question,
do not offline any other CPUs, because doing so could force the
timer back onto one of the CPUs in question.
1. Force networking interrupts onto other CPUs.
2. Initiate any network I/O on other CPUs.
3. Once your application has started, prevent CPU-hotplug operations
from being initiated from tasks that might run on the CPU to
be de-jittered. (It is OK to force this CPU offline and then
bring it back online before you start your application.)
1. Force networking interrupts onto other CPUs.
2. Initiate any network I/O on other CPUs.
3. Once your application has started, prevent CPU-hotplug operations
from being initiated from tasks that might run on the CPU to
be de-jittered. (It is OK to force this CPU offline and then
bring it back online before you start your application.)
1. Force block-device interrupts onto some other CPU.
2. Initiate any block I/O on other CPUs.
3. Once your application has started, prevent CPU-hotplug operations
from being initiated from tasks that might run on the CPU to
be de-jittered. (It is OK to force this CPU offline and then
bring it back online before you start your application.)
1. Force block-device interrupts onto some other CPU.
2. Initiate any block I/O on other CPUs.
3. Once your application has started, prevent CPU-hotplug operations
from being initiated from tasks that might run on the CPU to
be de-jittered. (It is OK to force this CPU offline and then
bring it back online before you start your application.)
1. Force block-device interrupts onto some other CPU.
2. Initiate any block I/O and block-I/O polling on other CPUs.
3. Once your application has started, prevent CPU-hotplug operations
from being initiated from tasks that might run on the CPU to
be de-jittered. (It is OK to force this CPU offline and then
bring it back online before you start your application.)
1. Force block-device interrupts onto some other CPU.
2. Initiate any block I/O and block-I/O polling on other CPUs.
3. Once your application has started, prevent CPU-hotplug operations
from being initiated from tasks that might run on the CPU to
be de-jittered. (It is OK to force this CPU offline and then
bring it back online before you start your application.)
1. Avoid use of drivers that use tasklets. (Such drivers will contain
calls to things like tasklet_schedule().)
2. Convert all drivers that you must use from tasklets to workqueues.
3. Force interrupts for drivers using tasklets onto other CPUs,
and also do I/O involving these drivers on other CPUs.
1. Avoid use of drivers that use tasklets. (Such drivers will contain
calls to things like tasklet_schedule().)
2. Convert all drivers that you must use from tasklets to workqueues.
3. Force interrupts for drivers using tasklets onto other CPUs,
and also do I/O involving these drivers on other CPUs.
1. Avoid sending scheduler IPIs to the CPU to be de-jittered,
for example, ensure that at most one runnable kthread is present
on that CPU. If a thread that expects to run on the de-jittered
1. Avoid sending scheduler IPIs to the CPU to be de-jittered,
for example, ensure that at most one runnable kthread is present
on that CPU. If a thread that expects to run on the de-jittered
forcing both kernel threads and interrupts to execute elsewhere.
This further reduces the number of scheduler-clock interrupts
received by the de-jittered CPU.
forcing both kernel threads and interrupts to execute elsewhere.
This further reduces the number of scheduler-clock interrupts
received by the de-jittered CPU.
1. To the extent possible, keep the CPU out of the kernel when it
is non-idle. For example, avoid system calls and force both
kernel threads and interrupts to execute elsewhere.
1. To the extent possible, keep the CPU out of the kernel when it
is non-idle. For example, avoid system calls and force both
kernel threads and interrupts to execute elsewhere.
back online. Once you have onlined the CPUs in question, do not
offline any other CPUs, because doing so could force the timer
back onto one of the CPUs in question.
back online. Once you have onlined the CPUs in question, do not
offline any other CPUs, because doing so could force the timer
back onto one of the CPUs in question.
1. Offload callbacks and keep the CPU in either dyntick-idle or
adaptive-ticks state by doing all of the following:
1. Offload callbacks and keep the CPU in either dyntick-idle or
adaptive-ticks state by doing all of the following:
a. CONFIG_NO_HZ_FULL=y and ensure that the CPU to be
de-jittered is marked as an adaptive-ticks CPU using the
"nohz_full=" boot parameter. Bind the rcuo kthreads to
a. CONFIG_NO_HZ_FULL=y and ensure that the CPU to be
de-jittered is marked as an adaptive-ticks CPU using the
"nohz_full=" boot parameter. Bind the rcuo kthreads to
when it is non-idle, for example, by avoiding system
calls and by forcing both kernel threads and interrupts
to execute elsewhere.
when it is non-idle, for example, by avoiding system
calls and by forcing both kernel threads and interrupts
to execute elsewhere.
a. Build with CONFIG_NO_HZ=y and CONFIG_RCU_FAST_NO_HZ=y.
b. Ensure that the CPU goes idle frequently, allowing other
CPUs to detect that it has passed through an RCU quiescent
a. Build with CONFIG_NO_HZ=y and CONFIG_RCU_FAST_NO_HZ=y.
b. Ensure that the CPU goes idle frequently, allowing other
CPUs to detect that it has passed through an RCU quiescent
-Name: kworker/%u:%d%s (cpu, id, priority)
-Purpose: Execute workqueue requests
+Name:
+ kworker/%u:%d%s (cpu, id, priority)
+
+Purpose:
+ Execute workqueue requests
+
1. Run your workload at a real-time priority, which will allow
preempting the kworker daemons.
2. A given workqueue can be made visible in the sysfs filesystem
by passing the WQ_SYSFS to that workqueue's alloc_workqueue().
Such a workqueue can be confined to a given subset of the
1. Run your workload at a real-time priority, which will allow
preempting the kworker daemons.
2. A given workqueue can be made visible in the sysfs filesystem
by passing the WQ_SYSFS to that workqueue's alloc_workqueue().
Such a workqueue can be confined to a given subset of the
files. The set of WQ_SYSFS workqueues can be displayed using
"ls sys/devices/virtual/workqueue". That said, the workqueues
maintainer would like to caution people against indiscriminately
files. The set of WQ_SYSFS workqueues can be displayed using
"ls sys/devices/virtual/workqueue". That said, the workqueues
maintainer would like to caution people against indiscriminately
to remove it, even if its addition was a mistake.
3. Do any of the following needed to avoid jitter that your
application cannot tolerate:
to remove it, even if its addition was a mistake.
3. Do any of the following needed to avoid jitter that your
application cannot tolerate:
a. Build your kernel with CONFIG_SLUB=y rather than
CONFIG_SLAB=y, thus avoiding the slab allocator's periodic
use of each CPU's workqueues to run its cache_reap()
a. Build your kernel with CONFIG_SLUB=y rather than
CONFIG_SLAB=y, thus avoiding the slab allocator's periodic
use of each CPU's workqueues to run its cache_reap()
be able to build your kernel with CONFIG_CPU_FREQ=n to
avoid the CPU-frequency governor periodically running
on each CPU, including cs_dbs_timer() and od_dbs_timer().
be able to build your kernel with CONFIG_CPU_FREQ=n to
avoid the CPU-frequency governor periodically running
on each CPU, including cs_dbs_timer() and od_dbs_timer().
WARNING: Please check your CPU specifications to
make sure that this is safe on your particular system.
d. As of v3.18, Christoph Lameter's on-demand vmstat workers
WARNING: Please check your CPU specifications to
make sure that this is safe on your particular system.
d. As of v3.18, Christoph Lameter's on-demand vmstat workers
CONFIG_PMAC_RACKMETER=n to disable the CPU-meter,
avoiding OS jitter from rackmeter_do_timer().
CONFIG_PMAC_RACKMETER=n to disable the CPU-meter,
avoiding OS jitter from rackmeter_do_timer().
1. Build the kernel with CONFIG_PREEMPT=n. This prevents these
kthreads from being created in the first place, and also obviates
the need for RCU priority boosting. This approach is feasible
1. Build the kernel with CONFIG_PREEMPT=n. This prevents these
kthreads from being created in the first place, and also obviates
the need for RCU priority boosting. This approach is feasible
-Name: rcuob/%d, rcuop/%d, and rcuos/%d
-Purpose: Offload RCU callbacks from the corresponding CPU.
+Name:
+ rcuob/%d, rcuop/%d, and rcuos/%d
+
+Purpose:
+ Offload RCU callbacks from the corresponding CPU.
+
1. Use affinity, cgroups, or other mechanism to force these kthreads
to execute on some other CPU.
2. Build with CONFIG_RCU_NOCB_CPU=n, which will prevent these
1. Use affinity, cgroups, or other mechanism to force these kthreads
to execute on some other CPU.
2. Build with CONFIG_RCU_NOCB_CPU=n, which will prevent these
1. Build with CONFIG_LOCKUP_DETECTOR=n, which will prevent these
kthreads from being created in the first place.
2. Boot with "nosoftlockup=0", which will also prevent these kthreads
1. Build with CONFIG_LOCKUP_DETECTOR=n, which will prevent these
kthreads from being created in the first place.
2. Boot with "nosoftlockup=0", which will also prevent these kthreads