€•“      Œsphinx.addnodes”Œdocument”“”)”}”(Œ	rawsource”Œ ”Œchildren”]”(Œtranslations”ŒLanguagesNode”“”)”}”(hhh]”(h Œpending_xref”“”)”}”(hhh]”Œdocutils.nodes”ŒText”“”ŒChinese (Simplified)”…””}”Œparent”hsbaŒ
attributes”}”(Œids”]”Œclasses”]”Œnames”]”Œdupnames”]”Œbackrefs”]”Œ	refdomain”Œstd”Œreftype”Œdoc”Œ	reftarget”Œ"/translations/zh_CN/timers/highres”Œmodname”NŒ	classname”NŒrefexplicit”ˆuŒtagname”hhhubh)”}”(hhh]”hŒChinese (Traditional)”…””}”hh2sbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ"/translations/zh_TW/timers/highres”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒItalian”…””}”hhFsbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ"/translations/it_IT/timers/highres”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒJapanese”…””}”hhZsbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ"/translations/ja_JP/timers/highres”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒKorean”…””}”hhnsbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ"/translations/ko_KR/timers/highres”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒPortuguese (Brazilian)”…””}”hh‚sbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ"/translations/pt_BR/timers/highres”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubh)”}”(hhh]”hŒSpanish”…””}”hh–sbah}”(h]”h ]”h"]”h$]”h&]”Œ	refdomain”h)Œreftype”h+Œ	reftarget”Œ"/translations/sp_SP/timers/highres”Œmodname”NŒ	classname”NŒrefexplicit”ˆuh1hhhubeh}”(h]”h ]”h"]”h$]”h&]”Œcurrent_language”ŒEnglish”uh1h
hhŒ	_document”hŒsource”NŒline”NubhŒsection”“”)”}”(hhh]”(hŒtitle”“”)”}”(hŒ5High resolution timers and dynamic ticks design notes”h]”hŒ5High resolution timers and dynamic ticks design notes”…””}”(hh¼h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hºhh·h²hh³Œ</var/lib/git/docbuild/linux/Documentation/timers/highres.rst”h´KubhŒ	paragraph”“”)”}”(hŒøFurther information can be found in the paper of the OLS 2006 talk "hrtimers
and beyond". The paper is part of the OLS 2006 Proceedings Volume 1, which can
be found on the OLS website:
https://www.kernel.org/doc/ols/2006/ols2006v1-pages-333-346.pdf”h]”(hŒ½Further information can be found in the paper of the OLS 2006 talk â€œhrtimers
and beyondâ€. The paper is part of the OLS 2006 Proceedings Volume 1, which can
be found on the OLS website:
”…””}”(hhÍh²hh³Nh´NubhŒ	reference”“”)”}”(hŒ?https://www.kernel.org/doc/ols/2006/ols2006v1-pages-333-346.pdf”h]”hŒ?https://www.kernel.org/doc/ols/2006/ols2006v1-pages-333-346.pdf”…””}”(hh×h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”hÙuh1hÕhhÍubeh}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Khh·h²hubhÌ)”}”(hŒvThe slides to this talk are available from:
http://www.cs.columbia.edu/~nahum/w6998/papers/ols2006-hrtimers-slides.pdf”h]”(hŒ,The slides to this talk are available from:
”…””}”(hhìh²hh³Nh´NubhÖ)”}”(hŒJhttp://www.cs.columbia.edu/~nahum/w6998/papers/ols2006-hrtimers-slides.pdf”h]”hŒJhttp://www.cs.columbia.edu/~nahum/w6998/papers/ols2006-hrtimers-slides.pdf”…””}”(hhôh²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”höuh1hÕhhìubeh}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K
hh·h²hubhÌ)”}”(hX  The slides contain five figures (pages 2, 15, 18, 20, 22), which illustrate the
changes in the time(r) related Linux subsystems. Figure #1 (p. 2) shows the
design of the Linux time(r) system before hrtimers and other building blocks
got merged into mainline.”h]”hX  The slides contain five figures (pages 2, 15, 18, 20, 22), which illustrate the
changes in the time(r) related Linux subsystems. Figure #1 (p. 2) shows the
design of the Linux time(r) system before hrtimers and other building blocks
got merged into mainline.”…””}”(hj	  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Khh·h²hubhÌ)”}”(hŒ‡Note: the paper and the slides are talking about "clock event source", while we
switched to the name "clock event devices" in meantime.”h]”hŒNote: the paper and the slides are talking about â€œclock event sourceâ€, while we
switched to the name â€œclock event devicesâ€ in meantime.”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Khh·h²hubhÌ)”}”(hŒ8The design contains the following basic building blocks:”h]”hŒ8The design contains the following basic building blocks:”…””}”(hj%  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Khh·h²hubhŒbullet_list”“”)”}”(hhh]”(hŒ	list_item”“”)”}”(hŒhrtimer base infrastructure”h]”hÌ)”}”(hj<  h]”hŒhrtimer base infrastructure”…””}”(hj>  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Khj:  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hj5  h²hh³hÊh´Nubj9  )”}”(hŒ%timeofday and clock source management”h]”hÌ)”}”(hjS  h]”hŒ%timeofday and clock source management”…””}”(hjU  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KhjQ  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hj5  h²hh³hÊh´Nubj9  )”}”(hŒclock event management”h]”hÌ)”}”(hjj  h]”hŒclock event management”…””}”(hjl  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Khjh  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hj5  h²hh³hÊh´Nubj9  )”}”(hŒ#high resolution timer functionality”h]”hÌ)”}”(hj  h]”hŒ#high resolution timer functionality”…””}”(hjƒ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Khj  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hj5  h²hh³hÊh´Nubj9  )”}”(hŒdynamic ticks

”h]”hÌ)”}”(hŒdynamic ticks”h]”hŒdynamic ticks”…””}”(hjš  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Khj–  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hj5  h²hh³hÊh´Nubeh}”(h]”h ]”h"]”h$]”h&]”Œbullet”Œ-”uh1j3  h³hÊh´Khh·h²hubh¶)”}”(hhh]”(h»)”}”(hŒhrtimer base infrastructure”h]”hŒhrtimer base infrastructure”…””}”(hj¹  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hºhj¶  h²hh³hÊh´KubhÌ)”}”(hŒ½The hrtimer base infrastructure was merged into the 2.6.16 kernel. Details of
the base implementation are covered in Documentation/timers/hrtimers.rst. See
also figure #2 (OLS slides p. 15)”h]”hŒ½The hrtimer base infrastructure was merged into the 2.6.16 kernel. Details of
the base implementation are covered in Documentation/timers/hrtimers.rst. See
also figure #2 (OLS slides p. 15)”…””}”(hjÇ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K!hj¶  h²hubhÌ)”}”(hŒZThe main differences to the timer wheel, which holds the armed timer_list type
timers are:”h]”hŒZThe main differences to the timer wheel, which holds the armed timer_list type
timers are:”…””}”(hjÕ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K%hj¶  h²hubhŒblock_quote”“”)”}”(hŒj- time ordered enqueueing into a rb-tree
- independent of ticks (the processing is based on nanoseconds)

”h]”j4  )”}”(hhh]”(j9  )”}”(hŒ&time ordered enqueueing into a rb-tree”h]”hÌ)”}”(hjî  h]”hŒ&time ordered enqueueing into a rb-tree”…””}”(hjð  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K(hjì  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hjé  ubj9  )”}”(hŒ?independent of ticks (the processing is based on nanoseconds)

”h]”hÌ)”}”(hŒ=independent of ticks (the processing is based on nanoseconds)”h]”hŒ=independent of ticks (the processing is based on nanoseconds)”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K)hj  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hjé  ubeh}”(h]”h ]”h"]”h$]”h&]”j´  jµ  uh1j3  h³hÊh´K(hjå  ubah}”(h]”h ]”h"]”h$]”h&]”uh1jã  h³hÊh´K(hj¶  h²hubeh}”(h]”Œhrtimer-base-infrastructure”ah ]”h"]”Œhrtimer base infrastructure”ah$]”h&]”uh1hµhh·h²hh³hÊh´Kubh¶)”}”(hhh]”(h»)”}”(hŒ%timeofday and clock source management”h]”hŒ%timeofday and clock source management”…””}”(hj2  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hºhj/  h²hh³hÊh´K-ubhÌ)”}”(hXÒ  John Stultz's Generic Time Of Day (GTOD) framework moves a large portion of
code out of the architecture-specific areas into a generic management
framework, as illustrated in figure #3 (OLS slides p. 18). The architecture
specific portion is reduced to the low level hardware details of the clock
sources, which are registered in the framework and selected on a quality based
decision. The low level code provides hardware setup and readout routines and
initializes data structures, which are used by the generic time keeping code to
convert the clock ticks to nanosecond based time values. All other time keeping
related functionality is moved into the generic code. The GTOD base patch got
merged into the 2.6.18 kernel.”h]”hXÔ  John Stultzâ€™s Generic Time Of Day (GTOD) framework moves a large portion of
code out of the architecture-specific areas into a generic management
framework, as illustrated in figure #3 (OLS slides p. 18). The architecture
specific portion is reduced to the low level hardware details of the clock
sources, which are registered in the framework and selected on a quality based
decision. The low level code provides hardware setup and readout routines and
initializes data structures, which are used by the generic time keeping code to
convert the clock ticks to nanosecond based time values. All other time keeping
related functionality is moved into the generic code. The GTOD base patch got
merged into the 2.6.18 kernel.”…””}”(hj@  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K/hj/  h²hubhÌ)”}”(hŒnFurther information about the Generic Time Of Day framework is available in the
OLS 2005 Proceedings Volume 1:”h]”hŒnFurther information about the Generic Time Of Day framework is available in the
OLS 2005 Proceedings Volume 1:”…””}”(hjN  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K:hj/  h²hubjä  )”}”(hŒ=http://www.linuxsymposium.org/2005/linuxsymposium_procv1.pdf
”h]”hÌ)”}”(hŒ<http://www.linuxsymposium.org/2005/linuxsymposium_procv1.pdf”h]”hÖ)”}”(hjb  h]”hŒ<http://www.linuxsymposium.org/2005/linuxsymposium_procv1.pdf”…””}”(hjd  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”Œrefuri”jb  uh1hÕhj`  ubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K=hj\  ubah}”(h]”h ]”h"]”h$]”h&]”uh1jã  h³hÊh´K=hj/  h²hubhÌ)”}”(hŒƒThe paper "We Are Not Getting Any Younger: A New Approach to Time and
Timers" was written by J. Stultz, D.V. Hart, & N. Aravamudan.”h]”hŒ‡The paper â€œWe Are Not Getting Any Younger: A New Approach to Time and
Timersâ€ was written by J. Stultz, D.V. Hart, & N. Aravamudan.”…””}”(hj~  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K?hj/  h²hubhÌ)”}”(hŒ;Figure #3 (OLS slides p.18) illustrates the transformation.”h]”hŒ;Figure #3 (OLS slides p.18) illustrates the transformation.”…””}”(hjŒ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KBhj/  h²hubeh}”(h]”Œ%timeofday-and-clock-source-management”ah ]”h"]”Œ%timeofday and clock source management”ah$]”h&]”uh1hµhh·h²hh³hÊh´K-ubh¶)”}”(hhh]”(h»)”}”(hŒclock event management”h]”hŒclock event management”…””}”(hj¥  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hºhj¢  h²hh³hÊh´KFubhÌ)”}”(hX$  While clock sources provide read access to the monotonically increasing time
value, clock event devices are used to schedule the next event
interrupt(s). The next event is currently defined to be periodic, with its
period defined at compile time. The setup and selection of the event device
for various event driven functionalities is hardwired into the architecture
dependent code. This results in duplicated code across all architectures and
makes it extremely difficult to change the configuration of the system to use
event interrupt devices other than those already built into the
architecture. Another implication of the current design is that it is necessary
to touch all the architecture-specific implementations in order to provide new
functionality like high resolution timers or dynamic ticks.”h]”hX$  While clock sources provide read access to the monotonically increasing time
value, clock event devices are used to schedule the next event
interrupt(s). The next event is currently defined to be periodic, with its
period defined at compile time. The setup and selection of the event device
for various event driven functionalities is hardwired into the architecture
dependent code. This results in duplicated code across all architectures and
makes it extremely difficult to change the configuration of the system to use
event interrupt devices other than those already built into the
architecture. Another implication of the current design is that it is necessary
to touch all the architecture-specific implementations in order to provide new
functionality like high resolution timers or dynamic ticks.”…””}”(hj³  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KHhj¢  h²hubhÌ)”}”(hXQ  The clock events subsystem tries to address this problem by providing a generic
solution to manage clock event devices and their usage for the various clock
event driven kernel functionalities. The goal of the clock event subsystem is
to minimize the clock event related architecture dependent code to the pure
hardware related handling and to allow easy addition and utilization of new
clock event devices. It also minimizes the duplicated code across the
architectures as it provides generic functionality down to the interrupt
service handler, which is almost inherently hardware dependent.”h]”hXQ  The clock events subsystem tries to address this problem by providing a generic
solution to manage clock event devices and their usage for the various clock
event driven kernel functionalities. The goal of the clock event subsystem is
to minimize the clock event related architecture dependent code to the pure
hardware related handling and to allow easy addition and utilization of new
clock event devices. It also minimizes the duplicated code across the
architectures as it provides generic functionality down to the interrupt
service handler, which is almost inherently hardware dependent.”…””}”(hjÁ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KThj¢  h²hubhÌ)”}”(hXÅ  Clock event devices are registered either by the architecture dependent boot
code or at module insertion time. Each clock event device fills a data
structure with clock-specific property parameters and callback functions. The
clock event management decides, by using the specified property parameters, the
set of system functions a clock event device will be used to support. This
includes the distinction of per-CPU and per-system global event devices.”h]”hXÅ  Clock event devices are registered either by the architecture dependent boot
code or at module insertion time. Each clock event device fills a data
structure with clock-specific property parameters and callback functions. The
clock event management decides, by using the specified property parameters, the
set of system functions a clock event device will be used to support. This
includes the distinction of per-CPU and per-system global event devices.”…””}”(hjÏ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K]hj¢  h²hubhÌ)”}”(hŒÌSystem-level global event devices are used for the Linux periodic tick. Per-CPU
event devices are used to provide local CPU functionality such as process
accounting, profiling, and high resolution timers.”h]”hŒÌSystem-level global event devices are used for the Linux periodic tick. Per-CPU
event devices are used to provide local CPU functionality such as process
accounting, profiling, and high resolution timers.”…””}”(hjÝ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kdhj¢  h²hubhÌ)”}”(hŒ\The management layer assigns one or more of the following functions to a clock
event device:”h]”hŒ\The management layer assigns one or more of the following functions to a clock
event device:”…””}”(hjë  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Khhj¢  h²hubjä  )”}”(hŒ›- system global periodic tick (jiffies update)
- cpu local update_process_times
- cpu local profiling
- cpu local next event interrupt (non periodic mode)
”h]”j4  )”}”(hhh]”(j9  )”}”(hŒ,system global periodic tick (jiffies update)”h]”hÌ)”}”(hj  h]”hŒ,system global periodic tick (jiffies update)”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kkhj   ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hjý  ubj9  )”}”(hŒcpu local update_process_times”h]”hÌ)”}”(hj  h]”hŒcpu local update_process_times”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Klhj  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hjý  ubj9  )”}”(hŒcpu local profiling”h]”hÌ)”}”(hj0  h]”hŒcpu local profiling”…””}”(hj2  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kmhj.  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hjý  ubj9  )”}”(hŒ3cpu local next event interrupt (non periodic mode)
”h]”hÌ)”}”(hŒ2cpu local next event interrupt (non periodic mode)”h]”hŒ2cpu local next event interrupt (non periodic mode)”…””}”(hjI  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KnhjE  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hjý  ubeh}”(h]”h ]”h"]”h$]”h&]”j´  jµ  uh1j3  h³hÊh´Kkhjù  ubah}”(h]”h ]”h"]”h$]”h&]”uh1jã  h³hÊh´Kkhj¢  h²hubhÌ)”}”(hXî  The clock event device delegates the selection of those timer interrupt related
functions completely to the management layer. The clock management layer stores
a function pointer in the device description structure, which has to be called
from the hardware level handler. This removes a lot of duplicated code from the
architecture specific timer interrupt handlers and hands the control over the
clock event devices and the assignment of timer interrupt related functionality
to the core code.”h]”hXî  The clock event device delegates the selection of those timer interrupt related
functions completely to the management layer. The clock management layer stores
a function pointer in the device description structure, which has to be called
from the hardware level handler. This removes a lot of duplicated code from the
architecture specific timer interrupt handlers and hands the control over the
clock event devices and the assignment of timer interrupt related functionality
to the core code.”…””}”(hji  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kphj¢  h²hubhÌ)”}”(hŒëThe clock event layer API is rather small. Aside from the clock event device
registration interface it provides functions to schedule the next event
interrupt, clock event device notification service and support for suspend and
resume.”h]”hŒëThe clock event layer API is rather small. Aside from the clock event device
registration interface it provides functions to schedule the next event
interrupt, clock event device notification service and support for suspend and
resume.”…””}”(hjw  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kxhj¢  h²hubhÌ)”}”(hXk  The framework adds about 700 lines of code which results in a 2KB increase of
the kernel binary size. The conversion of i386 removes about 100 lines of
code. The binary size decrease is in the range of 400 byte. We believe that the
increase of flexibility and the avoidance of duplicated code across
architectures justifies the slight increase of the binary size.”h]”hXk  The framework adds about 700 lines of code which results in a 2KB increase of
the kernel binary size. The conversion of i386 removes about 100 lines of
code. The binary size decrease is in the range of 400 byte. We believe that the
increase of flexibility and the avoidance of duplicated code across
architectures justifies the slight increase of the binary size.”…””}”(hj…  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K}hj¢  h²hubhÌ)”}”(hXþ  The conversion of an architecture has no functional impact, but allows to
utilize the high resolution and dynamic tick functionalities without any change
to the clock event device and timer interrupt code. After the conversion the
enabling of high resolution timers and dynamic ticks is simply provided by
adding the kernel/time/Kconfig file to the architecture specific Kconfig and
adding the dynamic tick specific calls to the idle routine (a total of 3 lines
added to the idle function and the Kconfig file)”h]”hXþ  The conversion of an architecture has no functional impact, but allows to
utilize the high resolution and dynamic tick functionalities without any change
to the clock event device and timer interrupt code. After the conversion the
enabling of high resolution timers and dynamic ticks is simply provided by
adding the kernel/time/Kconfig file to the architecture specific Kconfig and
adding the dynamic tick specific calls to the idle routine (a total of 3 lines
added to the idle function and the Kconfig file)”…””}”(hj“  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kƒhj¢  h²hubhÌ)”}”(hŒ;Figure #4 (OLS slides p.20) illustrates the transformation.”h]”hŒ;Figure #4 (OLS slides p.20) illustrates the transformation.”…””}”(hj¡  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K‹hj¢  h²hubeh}”(h]”Œclock-event-management”ah ]”h"]”Œclock event management”ah$]”h&]”uh1hµhh·h²hh³hÊh´KFubh¶)”}”(hhh]”(h»)”}”(hŒ#high resolution timer functionality”h]”hŒ#high resolution timer functionality”…””}”(hjº  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hºhj·  h²hh³hÊh´KubhÌ)”}”(hX«  During system boot it is not possible to use the high resolution timer
functionality, while making it possible would be difficult and would serve no
useful function. The initialization of the clock event device framework, the
clock source framework (GTOD) and hrtimers itself has to be done and
appropriate clock sources and clock event devices have to be registered before
the high resolution functionality can work. Up to the point where hrtimers are
initialized, the system works in the usual low resolution periodic mode. The
clock source and the clock event device layers provide notification functions
which inform hrtimers about availability of new hardware. hrtimers validates
the usability of the registered clock sources and clock event devices before
switching to high resolution mode. This ensures also that a kernel which is
configured for high resolution timers can run on a system which lacks the
necessary hardware support.”h]”hX«  During system boot it is not possible to use the high resolution timer
functionality, while making it possible would be difficult and would serve no
useful function. The initialization of the clock event device framework, the
clock source framework (GTOD) and hrtimers itself has to be done and
appropriate clock sources and clock event devices have to be registered before
the high resolution functionality can work. Up to the point where hrtimers are
initialized, the system works in the usual low resolution periodic mode. The
clock source and the clock event device layers provide notification functions
which inform hrtimers about availability of new hardware. hrtimers validates
the usability of the registered clock sources and clock event devices before
switching to high resolution mode. This ensures also that a kernel which is
configured for high resolution timers can run on a system which lacks the
necessary hardware support.”…””}”(hjÈ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K‘hj·  h²hubhÌ)”}”(hXË  The high resolution timer code does not support SMP machines which have only
global clock event devices. The support of such hardware would involve IPI
calls when an interrupt happens. The overhead would be much larger than the
benefit. This is the reason why we currently disable high resolution and
dynamic ticks on i386 SMP systems which stop the local APIC in C3 power
state. A workaround is available as an idea, but the problem has not been
tackled yet.”h]”hXË  The high resolution timer code does not support SMP machines which have only
global clock event devices. The support of such hardware would involve IPI
calls when an interrupt happens. The overhead would be much larger than the
benefit. This is the reason why we currently disable high resolution and
dynamic ticks on i386 SMP systems which stop the local APIC in C3 power
state. A workaround is available as an idea, but the problem has not been
tackled yet.”…””}”(hjÖ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KŸhj·  h²hubhÌ)”}”(hX´  The time ordered insertion of timers provides all the infrastructure to decide
whether the event device has to be reprogrammed when a timer is added. The
decision is made per timer base and synchronized across per-cpu timer bases in
a support function. The design allows the system to utilize separate per-CPU
clock event devices for the per-CPU timer bases, but currently only one
reprogrammable clock event device per-CPU is utilized.”h]”hX´  The time ordered insertion of timers provides all the infrastructure to decide
whether the event device has to be reprogrammed when a timer is added. The
decision is made per timer base and synchronized across per-cpu timer bases in
a support function. The design allows the system to utilize separate per-CPU
clock event devices for the per-CPU timer bases, but currently only one
reprogrammable clock event device per-CPU is utilized.”…””}”(hjä  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K§hj·  h²hubhÌ)”}”(hXý  When the timer interrupt happens, the next event interrupt handler is called
from the clock event distribution code and moves expired timers from the
red-black tree to a separate double linked list and invokes the softirq
handler. An additional mode field in the hrtimer structure allows the system to
execute callback functions directly from the next event interrupt handler. This
is restricted to code which can safely be executed in the hard interrupt
context. This applies, for example, to the common case of a wakeup function as
used by nanosleep. The advantage of executing the handler in the interrupt
context is the avoidance of up to two context switches - from the interrupted
context to the softirq and to the task which is woken up by the expired
timer.”h]”hXý  When the timer interrupt happens, the next event interrupt handler is called
from the clock event distribution code and moves expired timers from the
red-black tree to a separate double linked list and invokes the softirq
handler. An additional mode field in the hrtimer structure allows the system to
execute callback functions directly from the next event interrupt handler. This
is restricted to code which can safely be executed in the hard interrupt
context. This applies, for example, to the common case of a wakeup function as
used by nanosleep. The advantage of executing the handler in the interrupt
context is the avoidance of up to two context switches - from the interrupted
context to the softirq and to the task which is woken up by the expired
timer.”…””}”(hjò  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K®hj·  h²hubhÌ)”}”(hŒºOnce a system has switched to high resolution mode, the periodic tick is
switched off. This disables the per system global periodic clock event device -
e.g. the PIT on i386 SMP systems.”h]”hŒºOnce a system has switched to high resolution mode, the periodic tick is
switched off. This disables the per system global periodic clock event device -
e.g. the PIT on i386 SMP systems.”…””}”(hj   h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kºhj·  h²hubhÌ)”}”(hX,  The periodic tick functionality is provided by an per-cpu hrtimer. The callback
function is executed in the next event interrupt context and updates jiffies
and calls update_process_times and profiling. The implementation of the hrtimer
based periodic tick is designed to be extended with dynamic tick functionality.
This allows to use a single clock event device to schedule high resolution
timer and periodic events (jiffies tick, profiling, process accounting) on UP
systems. This has been proved to work with the PIT on i386 and the Incrementer
on PPC.”h]”hX,  The periodic tick functionality is provided by an per-cpu hrtimer. The callback
function is executed in the next event interrupt context and updates jiffies
and calls update_process_times and profiling. The implementation of the hrtimer
based periodic tick is designed to be extended with dynamic tick functionality.
This allows to use a single clock event device to schedule high resolution
timer and periodic events (jiffies tick, profiling, process accounting) on UP
systems. This has been proved to work with the PIT on i386 and the Incrementer
on PPC.”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K¾hj·  h²hubhÌ)”}”(hX€  The softirq for running the hrtimer queues and executing the callbacks has been
separated from the tick bound timer softirq to allow accurate delivery of high
resolution timer signals which are used by itimer and POSIX interval
timers. The execution of this softirq can still be delayed by other softirqs,
but the overall latencies have been significantly improved by this separation.”h]”hX€  The softirq for running the hrtimer queues and executing the callbacks has been
separated from the tick bound timer softirq to allow accurate delivery of high
resolution timer signals which are used by itimer and POSIX interval
timers. The execution of this softirq can still be delayed by other softirqs,
but the overall latencies have been significantly improved by this separation.”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KÇhj·  h²hubhÌ)”}”(hŒ;Figure #5 (OLS slides p.22) illustrates the transformation.”h]”hŒ;Figure #5 (OLS slides p.22) illustrates the transformation.”…””}”(hj*  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KÍhj·  h²hubeh}”(h]”Œ#high-resolution-timer-functionality”ah ]”h"]”Œ#high resolution timer functionality”ah$]”h&]”uh1hµhh·h²hh³hÊh´Kubh¶)”}”(hhh]”(h»)”}”(hŒdynamic ticks”h]”hŒdynamic ticks”…””}”(hjC  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hºhj@  h²hh³hÊh´KÑubhÌ)”}”(hŒ²Dynamic ticks are the logical consequence of the hrtimer based periodic tick
replacement (sched_tick). The functionality of the sched_tick hrtimer is
extended by three functions:”h]”hŒ²Dynamic ticks are the logical consequence of the hrtimer based periodic tick
replacement (sched_tick). The functionality of the sched_tick hrtimer is
extended by three functions:”…””}”(hjQ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KÓhj@  h²hubj4  )”}”(hhh]”(j9  )”}”(hŒhrtimer_stop_sched_tick”h]”hÌ)”}”(hjd  h]”hŒhrtimer_stop_sched_tick”…””}”(hjf  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´K×hjb  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hj_  h²hh³hÊh´Nubj9  )”}”(hŒhrtimer_restart_sched_tick”h]”hÌ)”}”(hj{  h]”hŒhrtimer_restart_sched_tick”…””}”(hj}  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KØhjy  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hj_  h²hh³hÊh´Nubj9  )”}”(hŒhrtimer_update_jiffies
”h]”hÌ)”}”(hŒhrtimer_update_jiffies”h]”hŒhrtimer_update_jiffies”…””}”(hj”  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KÙhj  ubah}”(h]”h ]”h"]”h$]”h&]”uh1j8  hj_  h²hh³hÊh´Nubeh}”(h]”h ]”h"]”h$]”h&]”j´  jµ  uh1j3  h³hÊh´K×hj@  h²hubhÌ)”}”(hX£  hrtimer_stop_sched_tick() is called when a CPU goes into idle state. The code
evaluates the next scheduled timer event (from both hrtimers and the timer
wheel) and in case that the next event is further away than the next tick it
reprograms the sched_tick to this future event, to allow longer idle sleeps
without worthless interruption by the periodic tick. The function is also
called when an interrupt happens during the idle period, which does not cause a
reschedule. The call is necessary as the interrupt handler might have armed a
new timer whose expiry time is before the time which was identified as the
nearest event in the previous call to hrtimer_stop_sched_tick.”h]”hX£  hrtimer_stop_sched_tick() is called when a CPU goes into idle state. The code
evaluates the next scheduled timer event (from both hrtimers and the timer
wheel) and in case that the next event is further away than the next tick it
reprograms the sched_tick to this future event, to allow longer idle sleeps
without worthless interruption by the periodic tick. The function is also
called when an interrupt happens during the idle period, which does not cause a
reschedule. The call is necessary as the interrupt handler might have armed a
new timer whose expiry time is before the time which was identified as the
nearest event in the previous call to hrtimer_stop_sched_tick.”…””}”(hj®  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´KÛhj@  h²hubhÌ)”}”(hŒährtimer_restart_sched_tick() is called when the CPU leaves the idle state before
it calls schedule(). hrtimer_restart_sched_tick() resumes the periodic tick,
which is kept active until the next call to hrtimer_stop_sched_tick().”h]”hŒährtimer_restart_sched_tick() is called when the CPU leaves the idle state before
it calls schedule(). hrtimer_restart_sched_tick() resumes the periodic tick,
which is kept active until the next call to hrtimer_stop_sched_tick().”…””}”(hj¼  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kåhj@  h²hubhÌ)”}”(hŒÙhrtimer_update_jiffies() is called from irq_enter() when an interrupt happens
in the idle period to make sure that jiffies are up to date and the interrupt
handler has not to deal with an eventually stale jiffy value.”h]”hŒÙhrtimer_update_jiffies() is called from irq_enter() when an interrupt happens
in the idle period to make sure that jiffies are up to date and the interrupt
handler has not to deal with an eventually stale jiffy value.”…””}”(hjÊ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kéhj@  h²hubhÌ)”}”(hŒ¤The dynamic tick feature provides statistical values which are exported to
userspace via /proc/stat and can be made available for enhanced power
management control.”h]”hŒ¤The dynamic tick feature provides statistical values which are exported to
userspace via /proc/stat and can be made available for enhanced power
management control.”…””}”(hjØ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kíhj@  h²hubhÌ)”}”(hŒÖThe implementation leaves room for further development like full tickless
systems, where the time slice is controlled by the scheduler, variable
frequency profiling, and a complete removal of jiffies in the future.”h]”hŒÖThe implementation leaves room for further development like full tickless
systems, where the time slice is controlled by the scheduler, variable
frequency profiling, and a complete removal of jiffies in the future.”…””}”(hjæ  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kñhj@  h²hubhÌ)”}”(hŒºAside the current initial submission of i386 support, the patchset has been
extended to x86_64 and ARM already. Initial (work in progress) support is also
available for MIPS and PowerPC.”h]”hŒºAside the current initial submission of i386 support, the patchset has been
extended to x86_64 and ARM already. Initial (work in progress) support is also
available for MIPS and PowerPC.”…””}”(hjô  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Köhj@  h²hubjä  )”}”(hŒThomas, Ingo”h]”hÌ)”}”(hj  h]”hŒThomas, Ingo”…””}”(hj  h²hh³Nh´Nubah}”(h]”h ]”h"]”h$]”h&]”uh1hËh³hÊh´Kúhj  ubah}”(h]”h ]”h"]”h$]”h&]”uh1jã  h³hÊh´Kúhj@  h²hubeh}”(h]”Œdynamic-ticks”ah ]”h"]”Œdynamic ticks”ah$]”h&]”uh1hµhh·h²hh³hÊh´KÑubeh}”(h]”Œ5high-resolution-timers-and-dynamic-ticks-design-notes”ah ]”h"]”Œ5high resolution timers and dynamic ticks design notes”ah$]”h&]”uh1hµhhh²hh³hÊh´Kubeh}”(h]”h ]”h"]”h$]”h&]”Œsource”hÊuh1hŒcurrent_source”NŒcurrent_line”NŒsettings”Œdocutils.frontend”ŒValues”“”)”}”(hºNŒ	generator”NŒ	datestamp”NŒsource_link”NŒ
source_url”NŒtoc_backlinks”Œentry”Œfootnote_backlinks”KŒsectnum_xform”KŒstrip_comments”NŒstrip_elements_with_classes”NŒstrip_classes”NŒreport_level”KŒ
halt_level”KŒexit_status_level”KŒdebug”NŒwarning_stream”NŒ	traceback”ˆŒinput_encoding”Œ	utf-8-sig”Œinput_encoding_error_handler”Œstrict”Œoutput_encoding”Œutf-8”Œoutput_encoding_error_handler”jL  Œerror_encoding”Œutf-8”Œerror_encoding_error_handler”Œbackslashreplace”Œlanguage_code”Œen”Œrecord_dependencies”NŒconfig”NŒ	id_prefix”hŒauto_id_prefix”Œid”Œdump_settings”NŒdump_internals”NŒdump_transforms”NŒdump_pseudo_xml”NŒexpose_internals”NŒstrict_visitor”NŒ_disable_config”NŒ_source”hÊŒ_destination”NŒ_config_files”]”Œ7/var/lib/git/docbuild/linux/Documentation/docutils.conf”aŒfile_insertion_enabled”ˆŒraw_enabled”KŒline_length_limit”M'Œpep_references”NŒpep_base_url”Œhttps://peps.python.org/”Œpep_file_url_template”Œpep-%04d”Œrfc_references”NŒrfc_base_url”Œ&https://datatracker.ietf.org/doc/html/”Œ	tab_width”KŒtrim_footnote_reference_space”‰Œsyntax_highlight”Œlong”Œsmart_quotes”ˆŒsmartquotes_locales”]”Œcharacter_level_inline_markup”‰Œdoctitle_xform”‰Œdocinfo_xform”KŒsectsubtitle_xform”‰Œimage_loading”Œlink”Œembed_stylesheet”‰Œcloak_email_addresses”ˆŒsection_self_link”‰Œenv”NubŒreporter”NŒindirect_targets”]”Œsubstitution_defs”}”Œsubstitution_names”}”Œrefnames”}”Œrefids”}”Œnameids”}”(j&  j#  j,  j)  jŸ  jœ  j´  j±  j=  j:  j  j  uŒ	nametypes”}”(j&  ‰j,  ‰jŸ  ‰j´  ‰j=  ‰j  ‰uh}”(j#  h·j)  j¶  jœ  j/  j±  j¢  j:  j·  j  j@  uŒfootnote_refs”}”Œcitation_refs”}”Œautofootnotes”]”Œautofootnote_refs”]”Œsymbol_footnotes”]”Œsymbol_footnote_refs”]”Œ	footnotes”]”Œ	citations”]”Œautofootnote_start”KŒsymbol_footnote_start”K Œ
id_counter”Œcollections”ŒCounter”“”}”…”R”Œparse_messages”]”Œtransform_messages”]”Œtransformer”NŒinclude_log”]”Œ
decoration”Nh²hub.