path: root/Documentation/scheduler/sched-design-CFS.txt
diff options
Diffstat (limited to 'Documentation/scheduler/sched-design-CFS.txt')
1 files changed, 186 insertions, 0 deletions
diff --git a/Documentation/scheduler/sched-design-CFS.txt b/Documentation/scheduler/sched-design-CFS.txt
new file mode 100644
index 000000000000..88bcb8767335
--- /dev/null
+++ b/Documentation/scheduler/sched-design-CFS.txt
@@ -0,0 +1,186 @@
+This is the CFS scheduler.
+80% of CFS's design can be summed up in a single sentence: CFS basically
+models an "ideal, precise multi-tasking CPU" on real hardware.
+"Ideal multi-tasking CPU" is a (non-existent :-)) CPU that has 100%
+physical power and which can run each task at precise equal speed, in
+parallel, each at 1/nr_running speed. For example: if there are 2 tasks
+running then it runs each at 50% physical power - totally in parallel.
+On real hardware, we can run only a single task at once, so while that
+one task runs, the other tasks that are waiting for the CPU are at a
+disadvantage - the current task gets an unfair amount of CPU time. In
+CFS this fairness imbalance is expressed and tracked via the per-task
+p->wait_runtime (nanosec-unit) value. "wait_runtime" is the amount of
+time the task should now run on the CPU for it to become completely fair
+and balanced.
+( small detail: on 'ideal' hardware, the p->wait_runtime value would
+ always be zero - no task would ever get 'out of balance' from the
+ 'ideal' share of CPU time. )
+CFS's task picking logic is based on this p->wait_runtime value and it
+is thus very simple: it always tries to run the task with the largest
+p->wait_runtime value. In other words, CFS tries to run the task with
+the 'gravest need' for more CPU time. So CFS always tries to split up
+CPU time between runnable tasks as close to 'ideal multitasking
+hardware' as possible.
+Most of the rest of CFS's design just falls out of this really simple
+concept, with a few add-on embellishments like nice levels,
+multiprocessing and various algorithm variants to recognize sleepers.
+In practice it works like this: the system runs a task a bit, and when
+the task schedules (or a scheduler tick happens) the task's CPU usage is
+'accounted for': the (small) time it just spent using the physical CPU
+is deducted from p->wait_runtime. [minus the 'fair share' it would have
+gotten anyway]. Once p->wait_runtime gets low enough so that another
+task becomes the 'leftmost task' of the time-ordered rbtree it maintains
+(plus a small amount of 'granularity' distance relative to the leftmost
+task so that we do not over-schedule tasks and trash the cache) then the
+new leftmost task is picked and the current task is preempted.
+The rq->fair_clock value tracks the 'CPU time a runnable task would have
+fairly gotten, had it been runnable during that time'. So by using
+rq->fair_clock values we can accurately timestamp and measure the
+'expected CPU time' a task should have gotten. All runnable tasks are
+sorted in the rbtree by the "rq->fair_clock - p->wait_runtime" key, and
+CFS picks the 'leftmost' task and sticks to it. As the system progresses
+forwards, newly woken tasks are put into the tree more and more to the
+right - slowly but surely giving a chance for every task to become the
+'leftmost task' and thus get on the CPU within a deterministic amount of
+Some implementation details:
+ - the introduction of Scheduling Classes: an extensible hierarchy of
+ scheduler modules. These modules encapsulate scheduling policy
+ details and are handled by the scheduler core without the core
+ code assuming about them too much.
+ - sched_fair.c implements the 'CFS desktop scheduler': it is a
+ replacement for the vanilla scheduler's SCHED_OTHER interactivity
+ code.
+ I'd like to give credit to Con Kolivas for the general approach here:
+ he has proven via RSDL/SD that 'fair scheduling' is possible and that
+ it results in better desktop scheduling. Kudos Con!
+ The CFS patch uses a completely different approach and implementation
+ from RSDL/SD. My goal was to make CFS's interactivity quality exceed
+ that of RSDL/SD, which is a high standard to meet :-) Testing
+ feedback is welcome to decide this one way or another. [ and, in any
+ case, all of SD's logic could be added via a kernel/sched_sd.c module
+ as well, if Con is interested in such an approach. ]
+ CFS's design is quite radical: it does not use runqueues, it uses a
+ time-ordered rbtree to build a 'timeline' of future task execution,
+ and thus has no 'array switch' artifacts (by which both the vanilla
+ scheduler and RSDL/SD are affected).
+ CFS uses nanosecond granularity accounting and does not rely on any
+ jiffies or other HZ detail. Thus the CFS scheduler has no notion of
+ 'timeslices' and has no heuristics whatsoever. There is only one
+ central tunable (you have to switch on CONFIG_SCHED_DEBUG):
+ /proc/sys/kernel/sched_granularity_ns
+ which can be used to tune the scheduler from 'desktop' (low
+ latencies) to 'server' (good batching) workloads. It defaults to a
+ setting suitable for desktop workloads. SCHED_BATCH is handled by the
+ CFS scheduler module too.
+ Due to its design, the CFS scheduler is not prone to any of the
+ 'attacks' that exist today against the heuristics of the stock
+ scheduler: fiftyp.c, thud.c, chew.c, ring-test.c, massive_intr.c all
+ work fine and do not impact interactivity and produce the expected
+ behavior.
+ the CFS scheduler has a much stronger handling of nice levels and
+ SCHED_BATCH: both types of workloads should be isolated much more
+ agressively than under the vanilla scheduler.
+ ( another detail: due to nanosec accounting and timeline sorting,
+ sched_yield() support is very simple under CFS, and in fact under
+ CFS sched_yield() behaves much better than under any other
+ scheduler i have tested so far. )
+ - sched_rt.c implements SCHED_FIFO and SCHED_RR semantics, in a simpler
+ way than the vanilla scheduler does. It uses 100 runqueues (for all
+ 100 RT priority levels, instead of 140 in the vanilla scheduler)
+ and it needs no expired array.
+ - reworked/sanitized SMP load-balancing: the runqueue-walking
+ assumptions are gone from the load-balancing code now, and
+ iterators of the scheduling modules are used. The balancing code got
+ quite a bit simpler as a result.
+Group scheduler extension to CFS
+Normally the scheduler operates on individual tasks and strives to provide
+fair CPU time to each task. Sometimes, it may be desirable to group tasks
+and provide fair CPU time to each such task group. For example, it may
+be desirable to first provide fair CPU time to each user on the system
+and then to each task belonging to a user.
+CONFIG_FAIR_GROUP_SCHED strives to achieve exactly that. It lets
+SCHED_NORMAL/BATCH tasks be be grouped and divides CPU time fairly among such
+groups. At present, there are two (mutually exclusive) mechanisms to group
+tasks for CPU bandwidth control purpose:
+ - Based on user id (CONFIG_FAIR_USER_SCHED)
+ In this option, tasks are grouped according to their user id.
+ - Based on "cgroup" pseudo filesystem (CONFIG_FAIR_CGROUP_SCHED)
+ This options lets the administrator create arbitrary groups
+ of tasks, using the "cgroup" pseudo filesystem. See
+ Documentation/cgroups.txt for more information about this
+ filesystem.
+Only one of these options to group tasks can be chosen and not both.
+Group scheduler tunables:
+When CONFIG_FAIR_USER_SCHED is defined, a directory is created in sysfs for
+each new user and a "cpu_share" file is added in that directory.
+ # cd /sys/kernel/uids
+ # cat 512/cpu_share # Display user 512's CPU share
+ 1024
+ # echo 2048 > 512/cpu_share # Modify user 512's CPU share
+ # cat 512/cpu_share # Display user 512's CPU share
+ 2048
+ #
+CPU bandwidth between two users are divided in the ratio of their CPU shares.
+For ex: if you would like user "root" to get twice the bandwidth of user
+"guest", then set the cpu_share for both the users such that "root"'s
+cpu_share is twice "guest"'s cpu_share
+When CONFIG_FAIR_CGROUP_SCHED is defined, a "cpu.shares" file is created
+for each group created using the pseudo filesystem. See example steps
+below to create task groups and modify their CPU share using the "cgroups"
+pseudo filesystem
+ # mkdir /dev/cpuctl
+ # mount -t cgroup -ocpu none /dev/cpuctl
+ # cd /dev/cpuctl
+ # mkdir multimedia # create "multimedia" group of tasks
+ # mkdir browser # create "browser" group of tasks
+ # #Configure the multimedia group to receive twice the CPU bandwidth
+ # #that of browser group
+ # echo 2048 > multimedia/cpu.shares
+ # echo 1024 > browser/cpu.shares
+ # firefox & # Launch firefox and move it to "browser" group
+ # echo <firefox_pid> > browser/tasks
+ # #Launch gmplayer (or your favourite movie player)
+ # echo <movie_player_pid> > multimedia/tasks

Privacy Policy