^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 1) =============
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 2) CFS Scheduler
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 3) =============
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 4)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 5)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 6) 1. OVERVIEW
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 7) ============
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 8)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 9) CFS stands for "Completely Fair Scheduler," and is the new "desktop" process
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 10) scheduler implemented by Ingo Molnar and merged in Linux 2.6.23. It is the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 11) replacement for the previous vanilla scheduler's SCHED_OTHER interactivity
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 12) code.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 13)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 14) 80% of CFS's design can be summed up in a single sentence: CFS basically models
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 15) an "ideal, precise multi-tasking CPU" on real hardware.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 16)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 17) "Ideal multi-tasking CPU" is a (non-existent :-)) CPU that has 100% physical
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 18) power and which can run each task at precise equal speed, in parallel, each at
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 19) 1/nr_running speed. For example: if there are 2 tasks running, then it runs
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 20) each at 50% physical power --- i.e., actually in parallel.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 21)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 22) On real hardware, we can run only a single task at once, so we have to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 23) introduce the concept of "virtual runtime." The virtual runtime of a task
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 24) specifies when its next timeslice would start execution on the ideal
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 25) multi-tasking CPU described above. In practice, the virtual runtime of a task
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 26) is its actual runtime normalized to the total number of running tasks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 27)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 28)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 29)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 30) 2. FEW IMPLEMENTATION DETAILS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 31) ==============================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 32)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 33) In CFS the virtual runtime is expressed and tracked via the per-task
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 34) p->se.vruntime (nanosec-unit) value. This way, it's possible to accurately
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 35) timestamp and measure the "expected CPU time" a task should have gotten.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 36)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 37) [ small detail: on "ideal" hardware, at any time all tasks would have the same
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 38) p->se.vruntime value --- i.e., tasks would execute simultaneously and no task
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 39) would ever get "out of balance" from the "ideal" share of CPU time. ]
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 40)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 41) CFS's task picking logic is based on this p->se.vruntime value and it is thus
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 42) very simple: it always tries to run the task with the smallest p->se.vruntime
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 43) value (i.e., the task which executed least so far). CFS always tries to split
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 44) up CPU time between runnable tasks as close to "ideal multitasking hardware" as
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 45) possible.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 46)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 47) Most of the rest of CFS's design just falls out of this really simple concept,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 48) with a few add-on embellishments like nice levels, multiprocessing and various
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 49) algorithm variants to recognize sleepers.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 50)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 51)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 52)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 53) 3. THE RBTREE
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 54) ==============
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 55)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 56) CFS's design is quite radical: it does not use the old data structures for the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 57) runqueues, but it uses a time-ordered rbtree to build a "timeline" of future
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 58) task execution, and thus has no "array switch" artifacts (by which both the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 59) previous vanilla scheduler and RSDL/SD are affected).
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 60)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 61) CFS also maintains the rq->cfs.min_vruntime value, which is a monotonic
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 62) increasing value tracking the smallest vruntime among all tasks in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 63) runqueue. The total amount of work done by the system is tracked using
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 64) min_vruntime; that value is used to place newly activated entities on the left
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 65) side of the tree as much as possible.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 66)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 67) The total number of running tasks in the runqueue is accounted through the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 68) rq->cfs.load value, which is the sum of the weights of the tasks queued on the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 69) runqueue.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 70)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 71) CFS maintains a time-ordered rbtree, where all runnable tasks are sorted by the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 72) p->se.vruntime key. CFS picks the "leftmost" task from this tree and sticks to it.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 73) As the system progresses forwards, the executed tasks are put into the tree
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 74) more and more to the right --- slowly but surely giving a chance for every task
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 75) to become the "leftmost task" and thus get on the CPU within a deterministic
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 76) amount of time.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 77)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 78) Summing up, CFS works like this: it runs a task a bit, and when the task
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 79) schedules (or a scheduler tick happens) the task's CPU usage is "accounted
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 80) for": the (small) time it just spent using the physical CPU is added to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 81) p->se.vruntime. Once p->se.vruntime gets high enough so that another task
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 82) becomes the "leftmost task" of the time-ordered rbtree it maintains (plus a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 83) small amount of "granularity" distance relative to the leftmost task so that we
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 84) do not over-schedule tasks and trash the cache), then the new leftmost task is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 85) picked and the current task is preempted.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 86)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 87)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 88)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 89) 4. SOME FEATURES OF CFS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 90) ========================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 91)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 92) CFS uses nanosecond granularity accounting and does not rely on any jiffies or
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 93) other HZ detail. Thus the CFS scheduler has no notion of "timeslices" in the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 94) way the previous scheduler had, and has no heuristics whatsoever. There is
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 95) only one central tunable (you have to switch on CONFIG_SCHED_DEBUG):
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 96)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 97) /proc/sys/kernel/sched_min_granularity_ns
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 98)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 99) which can be used to tune the scheduler from "desktop" (i.e., low latencies) to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 100) "server" (i.e., good batching) workloads. It defaults to a setting suitable
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 101) for desktop workloads. SCHED_BATCH is handled by the CFS scheduler module too.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 102)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 103) Due to its design, the CFS scheduler is not prone to any of the "attacks" that
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 104) exist today against the heuristics of the stock scheduler: fiftyp.c, thud.c,
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 105) chew.c, ring-test.c, massive_intr.c all work fine and do not impact
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 106) interactivity and produce the expected behavior.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 107)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 108) The CFS scheduler has a much stronger handling of nice levels and SCHED_BATCH
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 109) than the previous vanilla scheduler: both types of workloads are isolated much
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 110) more aggressively.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 111)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 112) SMP load-balancing has been reworked/sanitized: the runqueue-walking
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 113) assumptions are gone from the load-balancing code now, and iterators of the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 114) scheduling modules are used. The balancing code got quite a bit simpler as a
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 115) result.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 116)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 117)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 118)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 119) 5. Scheduling policies
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 120) ======================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 121)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 122) CFS implements three scheduling policies:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 123)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 124) - SCHED_NORMAL (traditionally called SCHED_OTHER): The scheduling
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 125) policy that is used for regular tasks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 126)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 127) - SCHED_BATCH: Does not preempt nearly as often as regular tasks
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 128) would, thereby allowing tasks to run longer and make better use of
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 129) caches but at the cost of interactivity. This is well suited for
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 130) batch jobs.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 131)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 132) - SCHED_IDLE: This is even weaker than nice 19, but its not a true
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 133) idle timer scheduler in order to avoid to get into priority
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 134) inversion problems which would deadlock the machine.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 135)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 136) SCHED_FIFO/_RR are implemented in sched/rt.c and are as specified by
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 137) POSIX.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 138)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 139) The command chrt from util-linux-ng 2.13.1.1 can set all of these except
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 140) SCHED_IDLE.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 141)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 142)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 143)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 144) 6. SCHEDULING CLASSES
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 145) ======================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 146)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 147) The new CFS scheduler has been designed in such a way to introduce "Scheduling
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 148) Classes," an extensible hierarchy of scheduler modules. These modules
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 149) encapsulate scheduling policy details and are handled by the scheduler core
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 150) without the core code assuming too much about them.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 151)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 152) sched/fair.c implements the CFS scheduler described above.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 153)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 154) sched/rt.c implements SCHED_FIFO and SCHED_RR semantics, in a simpler way than
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 155) the previous vanilla scheduler did. It uses 100 runqueues (for all 100 RT
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 156) priority levels, instead of 140 in the previous scheduler) and it needs no
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 157) expired array.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 158)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 159) Scheduling classes are implemented through the sched_class structure, which
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 160) contains hooks to functions that must be called whenever an interesting event
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 161) occurs.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 162)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 163) This is the (partial) list of the hooks:
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 164)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 165) - enqueue_task(...)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 166)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 167) Called when a task enters a runnable state.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 168) It puts the scheduling entity (task) into the red-black tree and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 169) increments the nr_running variable.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 170)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 171) - dequeue_task(...)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 172)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 173) When a task is no longer runnable, this function is called to keep the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 174) corresponding scheduling entity out of the red-black tree. It decrements
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 175) the nr_running variable.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 176)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 177) - yield_task(...)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 178)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 179) This function is basically just a dequeue followed by an enqueue, unless the
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 180) compat_yield sysctl is turned on; in that case, it places the scheduling
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 181) entity at the right-most end of the red-black tree.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 182)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 183) - check_preempt_curr(...)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 184)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 185) This function checks if a task that entered the runnable state should
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 186) preempt the currently running task.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 187)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 188) - pick_next_task(...)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 189)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 190) This function chooses the most appropriate task eligible to run next.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 191)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 192) - set_curr_task(...)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 193)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 194) This function is called when a task changes its scheduling class or changes
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 195) its task group.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 196)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 197) - task_tick(...)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 198)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 199) This function is mostly called from time tick functions; it might lead to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 200) process switch. This drives the running preemption.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 201)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 202)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 203)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 204)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 205) 7. GROUP SCHEDULER EXTENSIONS TO CFS
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 206) =====================================
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 207)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 208) Normally, the scheduler operates on individual tasks and strives to provide
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 209) fair CPU time to each task. Sometimes, it may be desirable to group tasks and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 210) provide fair CPU time to each such task group. For example, it may be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 211) desirable to first provide fair CPU time to each user on the system and then to
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 212) each task belonging to a user.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 213)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 214) CONFIG_CGROUP_SCHED strives to achieve exactly that. It lets tasks to be
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 215) grouped and divides CPU time fairly among such groups.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 216)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 217) CONFIG_RT_GROUP_SCHED permits to group real-time (i.e., SCHED_FIFO and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 218) SCHED_RR) tasks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 219)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 220) CONFIG_FAIR_GROUP_SCHED permits to group CFS (i.e., SCHED_NORMAL and
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 221) SCHED_BATCH) tasks.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 222)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 223) These options need CONFIG_CGROUPS to be defined, and let the administrator
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 224) create arbitrary groups of tasks, using the "cgroup" pseudo filesystem. See
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 225) Documentation/admin-guide/cgroup-v1/cgroups.rst for more information about this filesystem.
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 226)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 227) When CONFIG_FAIR_GROUP_SCHED is defined, a "cpu.shares" file is created for each
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 228) group created using the pseudo filesystem. See example steps below to create
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 229) task groups and modify their CPU share using the "cgroups" pseudo filesystem::
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 230)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 231) # mount -t tmpfs cgroup_root /sys/fs/cgroup
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 232) # mkdir /sys/fs/cgroup/cpu
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 233) # mount -t cgroup -ocpu none /sys/fs/cgroup/cpu
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 234) # cd /sys/fs/cgroup/cpu
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 235)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 236) # mkdir multimedia # create "multimedia" group of tasks
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 237) # mkdir browser # create "browser" group of tasks
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 238)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 239) # #Configure the multimedia group to receive twice the CPU bandwidth
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 240) # #that of browser group
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 241)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 242) # echo 2048 > multimedia/cpu.shares
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 243) # echo 1024 > browser/cpu.shares
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 244)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 245) # firefox & # Launch firefox and move it to "browser" group
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 246) # echo <firefox_pid> > browser/tasks
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 247)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 248) # #Launch gmplayer (or your favourite movie player)
^8f3ce5b39 (kx 2023-10-28 12:00:06 +0300 249) # echo <movie_player_pid> > multimedia/tasks
Orange Pi5 kernel
Deprecated Linux kernel 5.10.110 for OrangePi 5/5B/5+ boards
3 Commits
0 Branches
0 Tags