Documentation/kernel-per-CPU-kthreads.txt

   1 REDUCING OS JITTER DUE TO PER-CPU KTHREADS
   2
   3 This document lists per-CPU kthreads in the Linux kernel and presents
   4 options to control their OS jitter.  Note that non-per-CPU kthreads are
   5 not listed here.  To reduce OS jitter from non-per-CPU kthreads, bind
   6 them to a "housekeeping" CPU dedicated to such work.
   7
   8
   9 REFERENCES
  10
  11 o       Documentation/IRQ-affinity.txt:  Binding interrupts to sets of CPUs.
  12
  13 o       Documentation/cgroups:  Using cgroups to bind tasks to sets of CPUs.
  14
  15 o       man taskset:  Using the taskset command to bind tasks to sets
  16         of CPUs.
  17
  18 o       man sched_setaffinity:  Using the sched_setaffinity() system
  19         call to bind tasks to sets of CPUs.
  20
  21 o       /sys/devices/system/cpu/cpuN/online:  Control CPU N's hotplug state,
  22         writing "0" to offline and "1" to online.
  23
  24 o       In order to locate kernel-generated OS jitter on CPU N:
  25
  26                 cd /sys/kernel/debug/tracing
  27                 echo 1 > max_graph_depth # Increase the "1" for more detail
  28                 echo function_graph > current_tracer
  29                 # run workload
  30                 cat per_cpu/cpuN/trace
  31
  32
  33 KTHREADS
  34
  35 Name: ehca_comp/%u
  36 Purpose: Periodically process Infiniband-related work.
  37 To reduce its OS jitter, do any of the following:
  38 1.      Don't use eHCA Infiniband hardware, instead choosing hardware
  39         that does not require per-CPU kthreads.  This will prevent these
  40         kthreads from being created in the first place.  (This will
  41         work for most people, as this hardware, though important, is
  42         relatively old and is produced in relatively low unit volumes.)
  43 2.      Do all eHCA-Infiniband-related work on other CPUs, including
  44         interrupts.
  45 3.      Rework the eHCA driver so that its per-CPU kthreads are
  46         provisioned only on selected CPUs.
  47
  48
  49 Name: irq/%d-%s
  50 Purpose: Handle threaded interrupts.
  51 To reduce its OS jitter, do the following:
  52 1.      Use irq affinity to force the irq threads to execute on
  53         some other CPU.
  54
  55 Name: kcmtpd_ctr_%d
  56 Purpose: Handle Bluetooth work.
  57 To reduce its OS jitter, do one of the following:
  58 1.      Don't use Bluetooth, in which case these kthreads won't be
  59         created in the first place.
  60 2.      Use irq affinity to force Bluetooth-related interrupts to
  61         occur on some other CPU and furthermore initiate all
  62         Bluetooth activity on some other CPU.
  63
  64 Name: ksoftirqd/%u
  65 Purpose: Execute softirq handlers when threaded or when under heavy load.
  66 To reduce its OS jitter, each softirq vector must be handled
  67 separately as follows:
  68 TIMER_SOFTIRQ:  Do all of the following:
  69 1.      To the extent possible, keep the CPU out of the kernel when it
  70         is non-idle, for example, by avoiding system calls and by forcing
  71         both kernel threads and interrupts to execute elsewhere.
  72 2.      Build with CONFIG_HOTPLUG_CPU=y.  After boot completes, force
  73         the CPU offline, then bring it back online.  This forces
  74         recurring timers to migrate elsewhere.  If you are concerned
  75         with multiple CPUs, force them all offline before bringing the
  76         first one back online.  Once you have onlined the CPUs in question,
  77         do not offline any other CPUs, because doing so could force the
  78         timer back onto one of the CPUs in question.
  79 NET_TX_SOFTIRQ and NET_RX_SOFTIRQ:  Do all of the following:
  80 1.      Force networking interrupts onto other CPUs.
  81 2.      Initiate any network I/O on other CPUs.
  82 3.      Once your application has started, prevent CPU-hotplug operations
  83         from being initiated from tasks that might run on the CPU to
  84         be de-jittered.  (It is OK to force this CPU offline and then
  85         bring it back online before you start your application.)
  86 BLOCK_SOFTIRQ:  Do all of the following:
  87 1.      Force block-device interrupts onto some other CPU.
  88 2.      Initiate any block I/O on other CPUs.
  89 3.      Once your application has started, prevent CPU-hotplug operations
  90         from being initiated from tasks that might run on the CPU to
  91         be de-jittered.  (It is OK to force this CPU offline and then
  92         bring it back online before you start your application.)
  93 BLOCK_IOPOLL_SOFTIRQ:  Do all of the following:
  94 1.      Force block-device interrupts onto some other CPU.
  95 2.      Initiate any block I/O and block-I/O polling on other CPUs.
  96 3.      Once your application has started, prevent CPU-hotplug operations
  97         from being initiated from tasks that might run on the CPU to
  98         be de-jittered.  (It is OK to force this CPU offline and then
  99         bring it back online before you start your application.)
 100 TASKLET_SOFTIRQ: Do one or more of the following:
 101 1.      Avoid use of drivers that use tasklets.  (Such drivers will contain
 102         calls to things like tasklet_schedule().)
 103 2.      Convert all drivers that you must use from tasklets to workqueues.
 104 3.      Force interrupts for drivers using tasklets onto other CPUs,
 105         and also do I/O involving these drivers on other CPUs.
 106 SCHED_SOFTIRQ: Do all of the following:
 107 1.      Avoid sending scheduler IPIs to the CPU to be de-jittered,
 108         for example, ensure that at most one runnable kthread is present
 109         on that CPU.  If a thread that expects to run on the de-jittered
 110         CPU awakens, the scheduler will send an IPI that can result in
 111         a subsequent SCHED_SOFTIRQ.
 112 2.      Build with CONFIG_RCU_NOCB_CPU=y, CONFIG_RCU_NOCB_CPU_ALL=y,
 113         CONFIG_NO_HZ_FULL=y, and, in addition, ensure that the CPU
 114         to be de-jittered is marked as an adaptive-ticks CPU using the
 115         "nohz_full=" boot parameter.  This reduces the number of
 116         scheduler-clock interrupts that the de-jittered CPU receives,
 117         minimizing its chances of being selected to do the load balancing
 118         work that runs in SCHED_SOFTIRQ context.
 119 3.      To the extent possible, keep the CPU out of the kernel when it
 120         is non-idle, for example, by avoiding system calls and by
 121         forcing both kernel threads and interrupts to execute elsewhere.
 122         This further reduces the number of scheduler-clock interrupts
 123         received by the de-jittered CPU.
 124 HRTIMER_SOFTIRQ:  Do all of the following:
 125 1.      To the extent possible, keep the CPU out of the kernel when it
 126         is non-idle.  For example, avoid system calls and force both
 127         kernel threads and interrupts to execute elsewhere.
 128 2.      Build with CONFIG_HOTPLUG_CPU=y.  Once boot completes, force the
 129         CPU offline, then bring it back online.  This forces recurring
 130         timers to migrate elsewhere.  If you are concerned with multiple
 131         CPUs, force them all offline before bringing the first one
 132         back online.  Once you have onlined the CPUs in question, do not
 133         offline any other CPUs, because doing so could force the timer
 134         back onto one of the CPUs in question.
 135 RCU_SOFTIRQ:  Do at least one of the following:
 136 1.      Offload callbacks and keep the CPU in either dyntick-idle or
 137         adaptive-ticks state by doing all of the following:
 138         a.      Build with CONFIG_RCU_NOCB_CPU=y, CONFIG_RCU_NOCB_CPU_ALL=y,
 139                 CONFIG_NO_HZ_FULL=y, and, in addition ensure that the CPU
 140                 to be de-jittered is marked as an adaptive-ticks CPU using
 141                 the "nohz_full=" boot parameter.  Bind the rcuo kthreads
 142                 to housekeeping CPUs, which can tolerate OS jitter.
 143         b.      To the extent possible, keep the CPU out of the kernel
 144                 when it is non-idle, for example, by avoiding system
 145                 calls and by forcing both kernel threads and interrupts
 146                 to execute elsewhere.
 147 2.      Enable RCU to do its processing remotely via dyntick-idle by
 148         doing all of the following:
 149         a.      Build with CONFIG_NO_HZ=y and CONFIG_RCU_FAST_NO_HZ=y.
 150         b.      Ensure that the CPU goes idle frequently, allowing other
 151                 CPUs to detect that it has passed through an RCU quiescent
 152                 state.  If the kernel is built with CONFIG_NO_HZ_FULL=y,
 153                 userspace execution also allows other CPUs to detect that
 154                 the CPU in question has passed through a quiescent state.
 155         c.      To the extent possible, keep the CPU out of the kernel
 156                 when it is non-idle, for example, by avoiding system
 157                 calls and by forcing both kernel threads and interrupts
 158                 to execute elsewhere.
 159
 160 Name: kworker/%u:%d%s (cpu, id, priority)
 161 Purpose: Execute workqueue requests
 162 To reduce its OS jitter, do any of the following:
 163 1.      Run your workload at a real-time priority, which will allow
 164         preempting the kworker daemons.
 165 2.      Do any of the following needed to avoid jitter that your
 166         application cannot tolerate:
 167         a.      Build your kernel with CONFIG_SLUB=y rather than
 168                 CONFIG_SLAB=y, thus avoiding the slab allocator's periodic
 169                 use of each CPU's workqueues to run its cache_reap()
 170                 function.
 171         b.      Avoid using oprofile, thus avoiding OS jitter from
 172                 wq_sync_buffer().
 173         c.      Limit your CPU frequency so that a CPU-frequency
 174                 governor is not required, possibly enlisting the aid of
 175                 special heatsinks or other cooling technologies.  If done
 176                 correctly, and if you CPU architecture permits, you should
 177                 be able to build your kernel with CONFIG_CPU_FREQ=n to
 178                 avoid the CPU-frequency governor periodically running
 179                 on each CPU, including cs_dbs_timer() and od_dbs_timer().
 180                 WARNING:  Please check your CPU specifications to
 181                 make sure that this is safe on your particular system.
 182         d.      It is not possible to entirely get rid of OS jitter
 183                 from vmstat_update() on CONFIG_SMP=y systems, but you
 184                 can decrease its frequency by writing a large value
 185                 to /proc/sys/vm/stat_interval.  The default value is
 186                 HZ, for an interval of one second.  Of course, larger
 187                 values will make your virtual-memory statistics update
 188                 more slowly.  Of course, you can also run your workload
 189                 at a real-time priority, thus preempting vmstat_update(),
 190                 but if your workload is CPU-bound, this is a bad idea.
 191                 However, there is an RFC patch from Christoph Lameter
 192                 (based on an earlier one from Gilad Ben-Yossef) that
 193                 reduces or even eliminates vmstat overhead for some
 194                 workloads at https://lkml.org/lkml/2013/9/4/379.
 195         e.      If running on high-end powerpc servers, build with
 196                 CONFIG_PPC_RTAS_DAEMON=n.  This prevents the RTAS
 197                 daemon from running on each CPU every second or so.
 198                 (This will require editing Kconfig files and will defeat
 199                 this platform's RAS functionality.)  This avoids jitter
 200                 due to the rtas_event_scan() function.
 201                 WARNING:  Please check your CPU specifications to
 202                 make sure that this is safe on your particular system.
 203         f.      If running on Cell Processor, build your kernel with
 204                 CBE_CPUFREQ_SPU_GOVERNOR=n to avoid OS jitter from
 205                 spu_gov_work().
 206                 WARNING:  Please check your CPU specifications to
 207                 make sure that this is safe on your particular system.
 208         g.      If running on PowerMAC, build your kernel with
 209                 CONFIG_PMAC_RACKMETER=n to disable the CPU-meter,
 210                 avoiding OS jitter from rackmeter_do_timer().
 211
 212 Name: rcuc/%u
 213 Purpose: Execute RCU callbacks in CONFIG_RCU_BOOST=y kernels.
 214 To reduce its OS jitter, do at least one of the following:
 215 1.      Build the kernel with CONFIG_PREEMPT=n.  This prevents these
 216         kthreads from being created in the first place, and also obviates
 217         the need for RCU priority boosting.  This approach is feasible
 218         for workloads that do not require high degrees of responsiveness.
 219 2.      Build the kernel with CONFIG_RCU_BOOST=n.  This prevents these
 220         kthreads from being created in the first place.  This approach
 221         is feasible only if your workload never requires RCU priority
 222         boosting, for example, if you ensure frequent idle time on all
 223         CPUs that might execute within the kernel.
 224 3.      Build with CONFIG_RCU_NOCB_CPU=y and CONFIG_RCU_NOCB_CPU_ALL=y,
 225         which offloads all RCU callbacks to kthreads that can be moved
 226         off of CPUs susceptible to OS jitter.  This approach prevents the
 227         rcuc/%u kthreads from having any work to do, so that they are
 228         never awakened.
 229 4.      Ensure that the CPU never enters the kernel, and, in particular,
 230         avoid initiating any CPU hotplug operations on this CPU.  This is
 231         another way of preventing any callbacks from being queued on the
 232         CPU, again preventing the rcuc/%u kthreads from having any work
 233         to do.
 234
 235 Name: rcuob/%d, rcuop/%d, and rcuos/%d
 236 Purpose: Offload RCU callbacks from the corresponding CPU.
 237 To reduce its OS jitter, do at least one of the following:
 238 1.      Use affinity, cgroups, or other mechanism to force these kthreads
 239         to execute on some other CPU.
 240 2.      Build with CONFIG_RCU_NOCB_CPU=n, which will prevent these
 241         kthreads from being created in the first place.  However, please
 242         note that this will not eliminate OS jitter, but will instead
 243         shift it to RCU_SOFTIRQ.
 244
 245 Name: watchdog/%u
 246 Purpose: Detect software lockups on each CPU.
 247 To reduce its OS jitter, do at least one of the following:
 248 1.      Build with CONFIG_LOCKUP_DETECTOR=n, which will prevent these
 249         kthreads from being created in the first place.
 250 2.      Echo a zero to /proc/sys/kernel/watchdog to disable the
 251         watchdog timer.
 252 3.      Echo a large number of /proc/sys/kernel/watchdog_thresh in
 253         order to reduce the frequency of OS jitter due to the watchdog
 254         timer down to a level that is acceptable for your workload.