~ [ source navigation ] ~ [ diff markup ] ~ [ identifier search ] ~

TOMOYO Linux Cross Reference
Linux/Documentation/timers/no_hz.rst

Version: ~ [ linux-6.11.5 ] ~ [ linux-6.10.14 ] ~ [ linux-6.9.12 ] ~ [ linux-6.8.12 ] ~ [ linux-6.7.12 ] ~ [ linux-6.6.58 ] ~ [ linux-6.5.13 ] ~ [ linux-6.4.16 ] ~ [ linux-6.3.13 ] ~ [ linux-6.2.16 ] ~ [ linux-6.1.114 ] ~ [ linux-6.0.19 ] ~ [ linux-5.19.17 ] ~ [ linux-5.18.19 ] ~ [ linux-5.17.15 ] ~ [ linux-5.16.20 ] ~ [ linux-5.15.169 ] ~ [ linux-5.14.21 ] ~ [ linux-5.13.19 ] ~ [ linux-5.12.19 ] ~ [ linux-5.11.22 ] ~ [ linux-5.10.228 ] ~ [ linux-5.9.16 ] ~ [ linux-5.8.18 ] ~ [ linux-5.7.19 ] ~ [ linux-5.6.19 ] ~ [ linux-5.5.19 ] ~ [ linux-5.4.284 ] ~ [ linux-5.3.18 ] ~ [ linux-5.2.21 ] ~ [ linux-5.1.21 ] ~ [ linux-5.0.21 ] ~ [ linux-4.20.17 ] ~ [ linux-4.19.322 ] ~ [ linux-4.18.20 ] ~ [ linux-4.17.19 ] ~ [ linux-4.16.18 ] ~ [ linux-4.15.18 ] ~ [ linux-4.14.336 ] ~ [ linux-4.13.16 ] ~ [ linux-4.12.14 ] ~ [ linux-4.11.12 ] ~ [ linux-4.10.17 ] ~ [ linux-4.9.337 ] ~ [ linux-4.4.302 ] ~ [ linux-3.10.108 ] ~ [ linux-2.6.32.71 ] ~ [ linux-2.6.0 ] ~ [ linux-2.4.37.11 ] ~ [ unix-v6-master ] ~ [ ccs-tools-1.8.9 ] ~ [ policy-sample ] ~
Architecture: ~ [ i386 ] ~ [ alpha ] ~ [ m68k ] ~ [ mips ] ~ [ ppc ] ~ [ sparc ] ~ [ sparc64 ] ~

  1 ======================================
  2 NO_HZ: Reducing Scheduling-Clock Ticks
  3 ======================================
  4 
  5 
  6 This document describes Kconfig options and boot parameters that can
  7 reduce the number of scheduling-clock interrupts, thereby improving energy
  8 efficiency and reducing OS jitter.  Reducing OS jitter is important for
  9 some types of computationally intensive high-performance computing (HPC)
 10 applications and for real-time applications.
 11 
 12 There are three main ways of managing scheduling-clock interrupts
 13 (also known as "scheduling-clock ticks" or simply "ticks"):
 14 
 15 1.      Never omit scheduling-clock ticks (CONFIG_HZ_PERIODIC=y or
 16         CONFIG_NO_HZ=n for older kernels).  You normally will -not-
 17         want to choose this option.
 18 
 19 2.      Omit scheduling-clock ticks on idle CPUs (CONFIG_NO_HZ_IDLE=y or
 20         CONFIG_NO_HZ=y for older kernels).  This is the most common
 21         approach, and should be the default.
 22 
 23 3.      Omit scheduling-clock ticks on CPUs that are either idle or that
 24         have only one runnable task (CONFIG_NO_HZ_FULL=y).  Unless you
 25         are running realtime applications or certain types of HPC
 26         workloads, you will normally -not- want this option.
 27 
 28 These three cases are described in the following three sections, followed
 29 by a third section on RCU-specific considerations, a fourth section
 30 discussing testing, and a fifth and final section listing known issues.
 31 
 32 
 33 Never Omit Scheduling-Clock Ticks
 34 =================================
 35 
 36 Very old versions of Linux from the 1990s and the very early 2000s
 37 are incapable of omitting scheduling-clock ticks.  It turns out that
 38 there are some situations where this old-school approach is still the
 39 right approach, for example, in heavy workloads with lots of tasks
 40 that use short bursts of CPU, where there are very frequent idle
 41 periods, but where these idle periods are also quite short (tens or
 42 hundreds of microseconds).  For these types of workloads, scheduling
 43 clock interrupts will normally be delivered any way because there
 44 will frequently be multiple runnable tasks per CPU.  In these cases,
 45 attempting to turn off the scheduling clock interrupt will have no effect
 46 other than increasing the overhead of switching to and from idle and
 47 transitioning between user and kernel execution.
 48 
 49 This mode of operation can be selected using CONFIG_HZ_PERIODIC=y (or
 50 CONFIG_NO_HZ=n for older kernels).
 51 
 52 However, if you are instead running a light workload with long idle
 53 periods, failing to omit scheduling-clock interrupts will result in
 54 excessive power consumption.  This is especially bad on battery-powered
 55 devices, where it results in extremely short battery lifetimes.  If you
 56 are running light workloads, you should therefore read the following
 57 section.
 58 
 59 In addition, if you are running either a real-time workload or an HPC
 60 workload with short iterations, the scheduling-clock interrupts can
 61 degrade your applications performance.  If this describes your workload,
 62 you should read the following two sections.
 63 
 64 
 65 Omit Scheduling-Clock Ticks For Idle CPUs
 66 =========================================
 67 
 68 If a CPU is idle, there is little point in sending it a scheduling-clock
 69 interrupt.  After all, the primary purpose of a scheduling-clock interrupt
 70 is to force a busy CPU to shift its attention among multiple duties,
 71 and an idle CPU has no duties to shift its attention among.
 72 
 73 An idle CPU that is not receiving scheduling-clock interrupts is said to
 74 be "dyntick-idle", "in dyntick-idle mode", "in nohz mode", or "running
 75 tickless".  The remainder of this document will use "dyntick-idle mode".
 76 
 77 The CONFIG_NO_HZ_IDLE=y Kconfig option causes the kernel to avoid sending
 78 scheduling-clock interrupts to idle CPUs, which is critically important
 79 both to battery-powered devices and to highly virtualized mainframes.
 80 A battery-powered device running a CONFIG_HZ_PERIODIC=y kernel would
 81 drain its battery very quickly, easily 2-3 times as fast as would the
 82 same device running a CONFIG_NO_HZ_IDLE=y kernel.  A mainframe running
 83 1,500 OS instances might find that half of its CPU time was consumed by
 84 unnecessary scheduling-clock interrupts.  In these situations, there
 85 is strong motivation to avoid sending scheduling-clock interrupts to
 86 idle CPUs.  That said, dyntick-idle mode is not free:
 87 
 88 1.      It increases the number of instructions executed on the path
 89         to and from the idle loop.
 90 
 91 2.      On many architectures, dyntick-idle mode also increases the
 92         number of expensive clock-reprogramming operations.
 93 
 94 Therefore, systems with aggressive real-time response constraints often
 95 run CONFIG_HZ_PERIODIC=y kernels (or CONFIG_NO_HZ=n for older kernels)
 96 in order to avoid degrading from-idle transition latencies.
 97 
 98 There is also a boot parameter "nohz=" that can be used to disable
 99 dyntick-idle mode in CONFIG_NO_HZ_IDLE=y kernels by specifying "nohz=off".
100 By default, CONFIG_NO_HZ_IDLE=y kernels boot with "nohz=on", enabling
101 dyntick-idle mode.
102 
103 
104 Omit Scheduling-Clock Ticks For CPUs With Only One Runnable Task
105 ================================================================
106 
107 If a CPU has only one runnable task, there is little point in sending it
108 a scheduling-clock interrupt because there is no other task to switch to.
109 Note that omitting scheduling-clock ticks for CPUs with only one runnable
110 task implies also omitting them for idle CPUs.
111 
112 The CONFIG_NO_HZ_FULL=y Kconfig option causes the kernel to avoid
113 sending scheduling-clock interrupts to CPUs with a single runnable task,
114 and such CPUs are said to be "adaptive-ticks CPUs".  This is important
115 for applications with aggressive real-time response constraints because
116 it allows them to improve their worst-case response times by the maximum
117 duration of a scheduling-clock interrupt.  It is also important for
118 computationally intensive short-iteration workloads:  If any CPU is
119 delayed during a given iteration, all the other CPUs will be forced to
120 wait idle while the delayed CPU finishes.  Thus, the delay is multiplied
121 by one less than the number of CPUs.  In these situations, there is
122 again strong motivation to avoid sending scheduling-clock interrupts.
123 
124 By default, no CPU will be an adaptive-ticks CPU.  The "nohz_full="
125 boot parameter specifies the adaptive-ticks CPUs.  For example,
126 "nohz_full=1,6-8" says that CPUs 1, 6, 7, and 8 are to be adaptive-ticks
127 CPUs.  Note that you are prohibited from marking all of the CPUs as
128 adaptive-tick CPUs:  At least one non-adaptive-tick CPU must remain
129 online to handle timekeeping tasks in order to ensure that system
130 calls like gettimeofday() returns accurate values on adaptive-tick CPUs.
131 (This is not an issue for CONFIG_NO_HZ_IDLE=y because there are no running
132 user processes to observe slight drifts in clock rate.) Note that this
133 means that your system must have at least two CPUs in order for
134 CONFIG_NO_HZ_FULL=y to do anything for you.
135 
136 Finally, adaptive-ticks CPUs must have their RCU callbacks offloaded.
137 This is covered in the "RCU IMPLICATIONS" section below.
138 
139 Normally, a CPU remains in adaptive-ticks mode as long as possible.
140 In particular, transitioning to kernel mode does not automatically change
141 the mode.  Instead, the CPU will exit adaptive-ticks mode only if needed,
142 for example, if that CPU enqueues an RCU callback.
143 
144 Just as with dyntick-idle mode, the benefits of adaptive-tick mode do
145 not come for free:
146 
147 1.      CONFIG_NO_HZ_FULL selects CONFIG_NO_HZ_COMMON, so you cannot run
148         adaptive ticks without also running dyntick idle.  This dependency
149         extends down into the implementation, so that all of the costs
150         of CONFIG_NO_HZ_IDLE are also incurred by CONFIG_NO_HZ_FULL.
151 
152 2.      The user/kernel transitions are slightly more expensive due
153         to the need to inform kernel subsystems (such as RCU) about
154         the change in mode.
155 
156 3.      POSIX CPU timers prevent CPUs from entering adaptive-tick mode.
157         Real-time applications needing to take actions based on CPU time
158         consumption need to use other means of doing so.
159 
160 4.      If there are more perf events pending than the hardware can
161         accommodate, they are normally round-robined so as to collect
162         all of them over time.  Adaptive-tick mode may prevent this
163         round-robining from happening.  This will likely be fixed by
164         preventing CPUs with large numbers of perf events pending from
165         entering adaptive-tick mode.
166 
167 5.      Scheduler statistics for adaptive-tick CPUs may be computed
168         slightly differently than those for non-adaptive-tick CPUs.
169         This might in turn perturb load-balancing of real-time tasks.
170 
171 Although improvements are expected over time, adaptive ticks is quite
172 useful for many types of real-time and compute-intensive applications.
173 However, the drawbacks listed above mean that adaptive ticks should not
174 (yet) be enabled by default.
175 
176 
177 RCU Implications
178 ================
179 
180 There are situations in which idle CPUs cannot be permitted to
181 enter either dyntick-idle mode or adaptive-tick mode, the most
182 common being when that CPU has RCU callbacks pending.
183 
184 Avoid this by offloading RCU callback processing to "rcuo" kthreads
185 using the CONFIG_RCU_NOCB_CPU=y Kconfig option.  The specific CPUs to
186 offload may be selected using The "rcu_nocbs=" kernel boot parameter,
187 which takes a comma-separated list of CPUs and CPU ranges, for example,
188 "1,3-5" selects CPUs 1, 3, 4, and 5.  Note that CPUs specified by
189 the "nohz_full" kernel boot parameter are also offloaded.
190 
191 The offloaded CPUs will never queue RCU callbacks, and therefore RCU
192 never prevents offloaded CPUs from entering either dyntick-idle mode
193 or adaptive-tick mode.  That said, note that it is up to userspace to
194 pin the "rcuo" kthreads to specific CPUs if desired.  Otherwise, the
195 scheduler will decide where to run them, which might or might not be
196 where you want them to run.
197 
198 
199 Testing
200 =======
201 
202 So you enable all the OS-jitter features described in this document,
203 but do not see any change in your workload's behavior.  Is this because
204 your workload isn't affected that much by OS jitter, or is it because
205 something else is in the way?  This section helps answer this question
206 by providing a simple OS-jitter test suite, which is available on branch
207 master of the following git archive:
208 
209 git://git.kernel.org/pub/scm/linux/kernel/git/frederic/dynticks-testing.git
210 
211 Clone this archive and follow the instructions in the README file.
212 This test procedure will produce a trace that will allow you to evaluate
213 whether or not you have succeeded in removing OS jitter from your system.
214 If this trace shows that you have removed OS jitter as much as is
215 possible, then you can conclude that your workload is not all that
216 sensitive to OS jitter.
217 
218 Note: this test requires that your system have at least two CPUs.
219 We do not currently have a good way to remove OS jitter from single-CPU
220 systems.
221 
222 
223 Known Issues
224 ============
225 
226 *       Dyntick-idle slows transitions to and from idle slightly.
227         In practice, this has not been a problem except for the most
228         aggressive real-time workloads, which have the option of disabling
229         dyntick-idle mode, an option that most of them take.  However,
230         some workloads will no doubt want to use adaptive ticks to
231         eliminate scheduling-clock interrupt latencies.  Here are some
232         options for these workloads:
233 
234         a.      Use PMQOS from userspace to inform the kernel of your
235                 latency requirements (preferred).
236 
237         b.      On x86 systems, use the "idle=mwait" boot parameter.
238 
239         c.      On x86 systems, use the "intel_idle.max_cstate=" to limit
240         `       the maximum C-state depth.
241 
242         d.      On x86 systems, use the "idle=poll" boot parameter.
243                 However, please note that use of this parameter can cause
244                 your CPU to overheat, which may cause thermal throttling
245                 to degrade your latencies -- and that this degradation can
246                 be even worse than that of dyntick-idle.  Furthermore,
247                 this parameter effectively disables Turbo Mode on Intel
248                 CPUs, which can significantly reduce maximum performance.
249 
250 *       Adaptive-ticks slows user/kernel transitions slightly.
251         This is not expected to be a problem for computationally intensive
252         workloads, which have few such transitions.  Careful benchmarking
253         will be required to determine whether or not other workloads
254         are significantly affected by this effect.
255 
256 *       Adaptive-ticks does not do anything unless there is only one
257         runnable task for a given CPU, even though there are a number
258         of other situations where the scheduling-clock tick is not
259         needed.  To give but one example, consider a CPU that has one
260         runnable high-priority SCHED_FIFO task and an arbitrary number
261         of low-priority SCHED_OTHER tasks.  In this case, the CPU is
262         required to run the SCHED_FIFO task until it either blocks or
263         some other higher-priority task awakens on (or is assigned to)
264         this CPU, so there is no point in sending a scheduling-clock
265         interrupt to this CPU.  However, the current implementation
266         nevertheless sends scheduling-clock interrupts to CPUs having a
267         single runnable SCHED_FIFO task and multiple runnable SCHED_OTHER
268         tasks, even though these interrupts are unnecessary.
269 
270         And even when there are multiple runnable tasks on a given CPU,
271         there is little point in interrupting that CPU until the current
272         running task's timeslice expires, which is almost always way
273         longer than the time of the next scheduling-clock interrupt.
274 
275         Better handling of these sorts of situations is future work.
276 
277 *       A reboot is required to reconfigure both adaptive idle and RCU
278         callback offloading.  Runtime reconfiguration could be provided
279         if needed, however, due to the complexity of reconfiguring RCU at
280         runtime, there would need to be an earthshakingly good reason.
281         Especially given that you have the straightforward option of
282         simply offloading RCU callbacks from all CPUs and pinning them
283         where you want them whenever you want them pinned.
284 
285 *       Additional configuration is required to deal with other sources
286         of OS jitter, including interrupts and system-utility tasks
287         and processes.  This configuration normally involves binding
288         interrupts and tasks to particular CPUs.
289 
290 *       Some sources of OS jitter can currently be eliminated only by
291         constraining the workload.  For example, the only way to eliminate
292         OS jitter due to global TLB shootdowns is to avoid the unmapping
293         operations (such as kernel module unload operations) that
294         result in these shootdowns.  For another example, page faults
295         and TLB misses can be reduced (and in some cases eliminated) by
296         using huge pages and by constraining the amount of memory used
297         by the application.  Pre-faulting the working set can also be
298         helpful, especially when combined with the mlock() and mlockall()
299         system calls.
300 
301 *       Unless all CPUs are idle, at least one CPU must keep the
302         scheduling-clock interrupt going in order to support accurate
303         timekeeping.
304 
305 *       If there might potentially be some adaptive-ticks CPUs, there
306         will be at least one CPU keeping the scheduling-clock interrupt
307         going, even if all CPUs are otherwise idle.
308 
309         Better handling of this situation is ongoing work.
310 
311 *       Some process-handling operations still require the occasional
312         scheduling-clock tick.  These operations include calculating CPU
313         load, maintaining sched average, computing CFS entity vruntime,
314         computing avenrun, and carrying out load balancing.  They are
315         currently accommodated by scheduling-clock tick every second
316         or so.  On-going work will eliminate the need even for these
317         infrequent scheduling-clock ticks.

~ [ source navigation ] ~ [ diff markup ] ~ [ identifier search ] ~

kernel.org | git.kernel.org | LWN.net | Project Home | SVN repository | Mail admin

Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.

sflogo.php