From a26169ba86c2fe819f58275235652bbae9ae8000 Mon Sep 17 00:00:00 2001 From: Con Kolivas Date: Fri, 25 Oct 2019 14:00:52 +1100 Subject: [PATCH 01/16] MultiQueue Skiplist Scheduler v0.202 --- .../admin-guide/kernel-parameters.txt | 8 + Documentation/admin-guide/sysctl/kernel.rst | 34 + Documentation/scheduler/sched-BFS.txt | 351 + Documentation/scheduler/sched-MuQSS.txt | 373 + arch/alpha/Kconfig | 2 + arch/arm/Kconfig | 2 + arch/arm64/Kconfig | 2 + arch/powerpc/Kconfig | 2 + arch/powerpc/platforms/cell/spufs/sched.c | 5 - arch/x86/Kconfig | 18 + fs/proc/base.c | 2 +- include/linux/init_task.h | 4 + include/linux/ioprio.h | 2 + include/linux/sched.h | 60 +- include/linux/sched/deadline.h | 9 + include/linux/sched/nohz.h | 2 +- include/linux/sched/prio.h | 12 + include/linux/sched/rt.h | 2 + include/linux/sched/task.h | 2 +- include/linux/skip_list.h | 33 + include/uapi/linux/sched.h | 9 +- init/Kconfig | 25 +- init/init_task.c | 10 + init/main.c | 2 + kernel/Kconfig.MuQSS | 105 + kernel/Makefile | 2 +- kernel/delayacct.c | 2 +- kernel/exit.c | 4 +- kernel/kthread.c | 30 +- kernel/livepatch/transition.c | 6 +- kernel/sched/Makefile | 10 +- kernel/sched/MuQSS.c | 7614 +++++++++++++++++ kernel/sched/MuQSS.h | 1051 +++ kernel/sched/cpufreq_schedutil.c | 12 +- kernel/sched/cpupri.h | 2 + kernel/sched/cputime.c | 22 +- kernel/sched/idle.c | 2 + kernel/sched/sched.h | 35 + kernel/sched/topology.c | 8 + kernel/skip_list.c | 148 + kernel/sysctl.c | 54 +- kernel/time/Kconfig | 2 +- kernel/time/clockevents.c | 5 + kernel/time/posix-cpu-timers.c | 4 +- kernel/time/timer.c | 7 +- kernel/trace/trace_selftest.c | 5 + 46 files changed, 10049 insertions(+), 52 deletions(-) create mode 100644 Documentation/scheduler/sched-BFS.txt create mode 100644 Documentation/scheduler/sched-MuQSS.txt create mode 100644 include/linux/skip_list.h create mode 100644 kernel/Kconfig.MuQSS create mode 100644 kernel/sched/MuQSS.c create mode 100644 kernel/sched/MuQSS.h create mode 100644 kernel/skip_list.c diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 7bc83f3d9bdf..2f9e8cdf5fec 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -4429,6 +4429,14 @@ Memory area to be used by remote processor image, managed by CMA. + rqshare= [X86] Select the MuQSS scheduler runqueue sharing type. + Format: + smt -- Share SMT (hyperthread) sibling runqueues + mc -- Share MC (multicore) sibling runqueues + smp -- Share SMP runqueues + none -- So not share any runqueues + Default value is mc + rw [KNL] Mount root device read-write on boot S [KNL] Run init in single mode diff --git a/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/admin-guide/sysctl/kernel.rst index 0d427fd10941..5b3406a3d76f 100644 --- a/Documentation/admin-guide/sysctl/kernel.rst +++ b/Documentation/admin-guide/sysctl/kernel.rst @@ -344,6 +344,16 @@ Controls whether the panic kmsg data should be reported to Hyper-V. = ========================================================= +iso_cpu: (MuQSS CPU scheduler only) +=================================== + +This sets the percentage cpu that the unprivileged SCHED_ISO tasks can +run effectively at realtime priority, averaged over a rolling five +seconds over the -whole- system, meaning all cpus. + +Set to 70 (percent) by default. + + kexec_load_disabled =================== @@ -922,6 +932,20 @@ ROM/Flash boot loader. Maybe to tell it what to do after rebooting. ??? +rr_interval: (MuQSS CPU scheduler only) +======================================= + +This is the smallest duration that any cpu process scheduling unit +will run for. Increasing this value can increase throughput of cpu +bound tasks substantially but at the expense of increased latencies +overall. Conversely decreasing it will decrease average and maximum +latencies but at the expense of throughput. This value is in +milliseconds and the default value chosen depends on the number of +cpus available at scheduler initialisation with a minimum of 6. + +Valid values are from 1-1000. + + sched_energy_aware ================== @@ -1230,3 +1254,13 @@ is 10 seconds. The softlockup threshold is (``2 * watchdog_thresh``). Setting this tunable to zero will disable lockup detection altogether. + + +yield_type: (MuQSS CPU scheduler only) +====================================== + +This determines what type of yield calls to sched_yield will perform. + + 0: No yield. + 1: Yield only to better priority/deadline tasks. (default) + 2: Expire timeslice and recalculate deadline. diff --git a/Documentation/scheduler/sched-BFS.txt b/Documentation/scheduler/sched-BFS.txt new file mode 100644 index 000000000000..c0282002a079 --- /dev/null +++ b/Documentation/scheduler/sched-BFS.txt @@ -0,0 +1,351 @@ +BFS - The Brain Fuck Scheduler by Con Kolivas. + +Goals. + +The goal of the Brain Fuck Scheduler, referred to as BFS from here on, is to +completely do away with the complex designs of the past for the cpu process +scheduler and instead implement one that is very simple in basic design. +The main focus of BFS is to achieve excellent desktop interactivity and +responsiveness without heuristics and tuning knobs that are difficult to +understand, impossible to model and predict the effect of, and when tuned to +one workload cause massive detriment to another. + + +Design summary. + +BFS is best described as a single runqueue, O(n) lookup, earliest effective +virtual deadline first design, loosely based on EEVDF (earliest eligible virtual +deadline first) and my previous Staircase Deadline scheduler. Each component +shall be described in order to understand the significance of, and reasoning for +it. The codebase when the first stable version was released was approximately +9000 lines less code than the existing mainline linux kernel scheduler (in +2.6.31). This does not even take into account the removal of documentation and +the cgroups code that is not used. + +Design reasoning. + +The single runqueue refers to the queued but not running processes for the +entire system, regardless of the number of CPUs. The reason for going back to +a single runqueue design is that once multiple runqueues are introduced, +per-CPU or otherwise, there will be complex interactions as each runqueue will +be responsible for the scheduling latency and fairness of the tasks only on its +own runqueue, and to achieve fairness and low latency across multiple CPUs, any +advantage in throughput of having CPU local tasks causes other disadvantages. +This is due to requiring a very complex balancing system to at best achieve some +semblance of fairness across CPUs and can only maintain relatively low latency +for tasks bound to the same CPUs, not across them. To increase said fairness +and latency across CPUs, the advantage of local runqueue locking, which makes +for better scalability, is lost due to having to grab multiple locks. + +A significant feature of BFS is that all accounting is done purely based on CPU +used and nowhere is sleep time used in any way to determine entitlement or +interactivity. Interactivity "estimators" that use some kind of sleep/run +algorithm are doomed to fail to detect all interactive tasks, and to falsely tag +tasks that aren't interactive as being so. The reason for this is that it is +close to impossible to determine that when a task is sleeping, whether it is +doing it voluntarily, as in a userspace application waiting for input in the +form of a mouse click or otherwise, or involuntarily, because it is waiting for +another thread, process, I/O, kernel activity or whatever. Thus, such an +estimator will introduce corner cases, and more heuristics will be required to +cope with those corner cases, introducing more corner cases and failed +interactivity detection and so on. Interactivity in BFS is built into the design +by virtue of the fact that tasks that are waking up have not used up their quota +of CPU time, and have earlier effective deadlines, thereby making it very likely +they will preempt any CPU bound task of equivalent nice level. See below for +more information on the virtual deadline mechanism. Even if they do not preempt +a running task, because the rr interval is guaranteed to have a bound upper +limit on how long a task will wait for, it will be scheduled within a timeframe +that will not cause visible interface jitter. + + +Design details. + +Task insertion. + +BFS inserts tasks into each relevant queue as an O(1) insertion into a double +linked list. On insertion, *every* running queue is checked to see if the newly +queued task can run on any idle queue, or preempt the lowest running task on the +system. This is how the cross-CPU scheduling of BFS achieves significantly lower +latency per extra CPU the system has. In this case the lookup is, in the worst +case scenario, O(n) where n is the number of CPUs on the system. + +Data protection. + +BFS has one single lock protecting the process local data of every task in the +global queue. Thus every insertion, removal and modification of task data in the +global runqueue needs to grab the global lock. However, once a task is taken by +a CPU, the CPU has its own local data copy of the running process' accounting +information which only that CPU accesses and modifies (such as during a +timer tick) thus allowing the accounting data to be updated lockless. Once a +CPU has taken a task to run, it removes it from the global queue. Thus the +global queue only ever has, at most, + + (number of tasks requesting cpu time) - (number of logical CPUs) + 1 + +tasks in the global queue. This value is relevant for the time taken to look up +tasks during scheduling. This will increase if many tasks with CPU affinity set +in their policy to limit which CPUs they're allowed to run on if they outnumber +the number of CPUs. The +1 is because when rescheduling a task, the CPU's +currently running task is put back on the queue. Lookup will be described after +the virtual deadline mechanism is explained. + +Virtual deadline. + +The key to achieving low latency, scheduling fairness, and "nice level" +distribution in BFS is entirely in the virtual deadline mechanism. The one +tunable in BFS is the rr_interval, or "round robin interval". This is the +maximum time two SCHED_OTHER (or SCHED_NORMAL, the common scheduling policy) +tasks of the same nice level will be running for, or looking at it the other +way around, the longest duration two tasks of the same nice level will be +delayed for. When a task requests cpu time, it is given a quota (time_slice) +equal to the rr_interval and a virtual deadline. The virtual deadline is +offset from the current time in jiffies by this equation: + + jiffies + (prio_ratio * rr_interval) + +The prio_ratio is determined as a ratio compared to the baseline of nice -20 +and increases by 10% per nice level. The deadline is a virtual one only in that +no guarantee is placed that a task will actually be scheduled by this time, but +it is used to compare which task should go next. There are three components to +how a task is next chosen. First is time_slice expiration. If a task runs out +of its time_slice, it is descheduled, the time_slice is refilled, and the +deadline reset to that formula above. Second is sleep, where a task no longer +is requesting CPU for whatever reason. The time_slice and deadline are _not_ +adjusted in this case and are just carried over for when the task is next +scheduled. Third is preemption, and that is when a newly waking task is deemed +higher priority than a currently running task on any cpu by virtue of the fact +that it has an earlier virtual deadline than the currently running task. The +earlier deadline is the key to which task is next chosen for the first and +second cases. Once a task is descheduled, it is put back on the queue, and an +O(n) lookup of all queued-but-not-running tasks is done to determine which has +the earliest deadline and that task is chosen to receive CPU next. + +The CPU proportion of different nice tasks works out to be approximately the + + (prio_ratio difference)^2 + +The reason it is squared is that a task's deadline does not change while it is +running unless it runs out of time_slice. Thus, even if the time actually +passes the deadline of another task that is queued, it will not get CPU time +unless the current running task deschedules, and the time "base" (jiffies) is +constantly moving. + +Task lookup. + +BFS has 103 priority queues. 100 of these are dedicated to the static priority +of realtime tasks, and the remaining 3 are, in order of best to worst priority, +SCHED_ISO (isochronous), SCHED_NORMAL, and SCHED_IDLEPRIO (idle priority +scheduling). When a task of these priorities is queued, a bitmap of running +priorities is set showing which of these priorities has tasks waiting for CPU +time. When a CPU is made to reschedule, the lookup for the next task to get +CPU time is performed in the following way: + +First the bitmap is checked to see what static priority tasks are queued. If +any realtime priorities are found, the corresponding queue is checked and the +first task listed there is taken (provided CPU affinity is suitable) and lookup +is complete. If the priority corresponds to a SCHED_ISO task, they are also +taken in FIFO order (as they behave like SCHED_RR). If the priority corresponds +to either SCHED_NORMAL or SCHED_IDLEPRIO, then the lookup becomes O(n). At this +stage, every task in the runlist that corresponds to that priority is checked +to see which has the earliest set deadline, and (provided it has suitable CPU +affinity) it is taken off the runqueue and given the CPU. If a task has an +expired deadline, it is taken and the rest of the lookup aborted (as they are +chosen in FIFO order). + +Thus, the lookup is O(n) in the worst case only, where n is as described +earlier, as tasks may be chosen before the whole task list is looked over. + + +Scalability. + +The major limitations of BFS will be that of scalability, as the separate +runqueue designs will have less lock contention as the number of CPUs rises. +However they do not scale linearly even with separate runqueues as multiple +runqueues will need to be locked concurrently on such designs to be able to +achieve fair CPU balancing, to try and achieve some sort of nice-level fairness +across CPUs, and to achieve low enough latency for tasks on a busy CPU when +other CPUs would be more suited. BFS has the advantage that it requires no +balancing algorithm whatsoever, as balancing occurs by proxy simply because +all CPUs draw off the global runqueue, in priority and deadline order. Despite +the fact that scalability is _not_ the prime concern of BFS, it both shows very +good scalability to smaller numbers of CPUs and is likely a more scalable design +at these numbers of CPUs. + +It also has some very low overhead scalability features built into the design +when it has been deemed their overhead is so marginal that they're worth adding. +The first is the local copy of the running process' data to the CPU it's running +on to allow that data to be updated lockless where possible. Then there is +deference paid to the last CPU a task was running on, by trying that CPU first +when looking for an idle CPU to use the next time it's scheduled. Finally there +is the notion of cache locality beyond the last running CPU. The sched_domains +information is used to determine the relative virtual "cache distance" that +other CPUs have from the last CPU a task was running on. CPUs with shared +caches, such as SMT siblings, or multicore CPUs with shared caches, are treated +as cache local. CPUs without shared caches are treated as not cache local, and +CPUs on different NUMA nodes are treated as very distant. This "relative cache +distance" is used by modifying the virtual deadline value when doing lookups. +Effectively, the deadline is unaltered between "cache local" CPUs, doubled for +"cache distant" CPUs, and quadrupled for "very distant" CPUs. The reasoning +behind the doubling of deadlines is as follows. The real cost of migrating a +task from one CPU to another is entirely dependant on the cache footprint of +the task, how cache intensive the task is, how long it's been running on that +CPU to take up the bulk of its cache, how big the CPU cache is, how fast and +how layered the CPU cache is, how fast a context switch is... and so on. In +other words, it's close to random in the real world where we do more than just +one sole workload. The only thing we can be sure of is that it's not free. So +BFS uses the principle that an idle CPU is a wasted CPU and utilising idle CPUs +is more important than cache locality, and cache locality only plays a part +after that. Doubling the effective deadline is based on the premise that the +"cache local" CPUs will tend to work on the same tasks up to double the number +of cache local CPUs, and once the workload is beyond that amount, it is likely +that none of the tasks are cache warm anywhere anyway. The quadrupling for NUMA +is a value I pulled out of my arse. + +When choosing an idle CPU for a waking task, the cache locality is determined +according to where the task last ran and then idle CPUs are ranked from best +to worst to choose the most suitable idle CPU based on cache locality, NUMA +node locality and hyperthread sibling business. They are chosen in the +following preference (if idle): + +* Same core, idle or busy cache, idle threads +* Other core, same cache, idle or busy cache, idle threads. +* Same node, other CPU, idle cache, idle threads. +* Same node, other CPU, busy cache, idle threads. +* Same core, busy threads. +* Other core, same cache, busy threads. +* Same node, other CPU, busy threads. +* Other node, other CPU, idle cache, idle threads. +* Other node, other CPU, busy cache, idle threads. +* Other node, other CPU, busy threads. + +This shows the SMT or "hyperthread" awareness in the design as well which will +choose a real idle core first before a logical SMT sibling which already has +tasks on the physical CPU. + +Early benchmarking of BFS suggested scalability dropped off at the 16 CPU mark. +However this benchmarking was performed on an earlier design that was far less +scalable than the current one so it's hard to know how scalable it is in terms +of both CPUs (due to the global runqueue) and heavily loaded machines (due to +O(n) lookup) at this stage. Note that in terms of scalability, the number of +_logical_ CPUs matters, not the number of _physical_ CPUs. Thus, a dual (2x) +quad core (4X) hyperthreaded (2X) machine is effectively a 16X. Newer benchmark +results are very promising indeed, without needing to tweak any knobs, features +or options. Benchmark contributions are most welcome. + + +Features + +As the initial prime target audience for BFS was the average desktop user, it +was designed to not need tweaking, tuning or have features set to obtain benefit +from it. Thus the number of knobs and features has been kept to an absolute +minimum and should not require extra user input for the vast majority of cases. +There are precisely 2 tunables, and 2 extra scheduling policies. The rr_interval +and iso_cpu tunables, and the SCHED_ISO and SCHED_IDLEPRIO policies. In addition +to this, BFS also uses sub-tick accounting. What BFS does _not_ now feature is +support for CGROUPS. The average user should neither need to know what these +are, nor should they need to be using them to have good desktop behaviour. + +rr_interval + +There is only one "scheduler" tunable, the round robin interval. This can be +accessed in + + /proc/sys/kernel/rr_interval + +The value is in milliseconds, and the default value is set to 6 on a +uniprocessor machine, and automatically set to a progressively higher value on +multiprocessor machines. The reasoning behind increasing the value on more CPUs +is that the effective latency is decreased by virtue of there being more CPUs on +BFS (for reasons explained above), and increasing the value allows for less +cache contention and more throughput. Valid values are from 1 to 1000 +Decreasing the value will decrease latencies at the cost of decreasing +throughput, while increasing it will improve throughput, but at the cost of +worsening latencies. The accuracy of the rr interval is limited by HZ resolution +of the kernel configuration. Thus, the worst case latencies are usually slightly +higher than this actual value. The default value of 6 is not an arbitrary one. +It is based on the fact that humans can detect jitter at approximately 7ms, so +aiming for much lower latencies is pointless under most circumstances. It is +worth noting this fact when comparing the latency performance of BFS to other +schedulers. Worst case latencies being higher than 7ms are far worse than +average latencies not being in the microsecond range. + +Isochronous scheduling. + +Isochronous scheduling is a unique scheduling policy designed to provide +near-real-time performance to unprivileged (ie non-root) users without the +ability to starve the machine indefinitely. Isochronous tasks (which means +"same time") are set using, for example, the schedtool application like so: + + schedtool -I -e amarok + +This will start the audio application "amarok" as SCHED_ISO. How SCHED_ISO works +is that it has a priority level between true realtime tasks and SCHED_NORMAL +which would allow them to preempt all normal tasks, in a SCHED_RR fashion (ie, +if multiple SCHED_ISO tasks are running, they purely round robin at rr_interval +rate). However if ISO tasks run for more than a tunable finite amount of time, +they are then demoted back to SCHED_NORMAL scheduling. This finite amount of +time is the percentage of _total CPU_ available across the machine, configurable +as a percentage in the following "resource handling" tunable (as opposed to a +scheduler tunable): + + /proc/sys/kernel/iso_cpu + +and is set to 70% by default. It is calculated over a rolling 5 second average +Because it is the total CPU available, it means that on a multi CPU machine, it +is possible to have an ISO task running as realtime scheduling indefinitely on +just one CPU, as the other CPUs will be available. Setting this to 100 is the +equivalent of giving all users SCHED_RR access and setting it to 0 removes the +ability to run any pseudo-realtime tasks. + +A feature of BFS is that it detects when an application tries to obtain a +realtime policy (SCHED_RR or SCHED_FIFO) and the caller does not have the +appropriate privileges to use those policies. When it detects this, it will +give the task SCHED_ISO policy instead. Thus it is transparent to the user. +Because some applications constantly set their policy as well as their nice +level, there is potential for them to undo the override specified by the user +on the command line of setting the policy to SCHED_ISO. To counter this, once +a task has been set to SCHED_ISO policy, it needs superuser privileges to set +it back to SCHED_NORMAL. This will ensure the task remains ISO and all child +processes and threads will also inherit the ISO policy. + +Idleprio scheduling. + +Idleprio scheduling is a scheduling policy designed to give out CPU to a task +_only_ when the CPU would be otherwise idle. The idea behind this is to allow +ultra low priority tasks to be run in the background that have virtually no +effect on the foreground tasks. This is ideally suited to distributed computing +clients (like setiathome, folding, mprime etc) but can also be used to start +a video encode or so on without any slowdown of other tasks. To avoid this +policy from grabbing shared resources and holding them indefinitely, if it +detects a state where the task is waiting on I/O, the machine is about to +suspend to ram and so on, it will transiently schedule them as SCHED_NORMAL. As +per the Isochronous task management, once a task has been scheduled as IDLEPRIO, +it cannot be put back to SCHED_NORMAL without superuser privileges. Tasks can +be set to start as SCHED_IDLEPRIO with the schedtool command like so: + + schedtool -D -e ./mprime + +Subtick accounting. + +It is surprisingly difficult to get accurate CPU accounting, and in many cases, +the accounting is done by simply determining what is happening at the precise +moment a timer tick fires off. This becomes increasingly inaccurate as the +timer tick frequency (HZ) is lowered. It is possible to create an application +which uses almost 100% CPU, yet by being descheduled at the right time, records +zero CPU usage. While the main problem with this is that there are possible +security implications, it is also difficult to determine how much CPU a task +really does use. BFS tries to use the sub-tick accounting from the TSC clock, +where possible, to determine real CPU usage. This is not entirely reliable, but +is far more likely to produce accurate CPU usage data than the existing designs +and will not show tasks as consuming no CPU usage when they actually are. Thus, +the amount of CPU reported as being used by BFS will more accurately represent +how much CPU the task itself is using (as is shown for example by the 'time' +application), so the reported values may be quite different to other schedulers. +Values reported as the 'load' are more prone to problems with this design, but +per process values are closer to real usage. When comparing throughput of BFS +to other designs, it is important to compare the actual completed work in terms +of total wall clock time taken and total work done, rather than the reported +"cpu usage". + + +Con Kolivas Fri Aug 27 2010 diff --git a/Documentation/scheduler/sched-MuQSS.txt b/Documentation/scheduler/sched-MuQSS.txt new file mode 100644 index 000000000000..ae28b85c9995 --- /dev/null +++ b/Documentation/scheduler/sched-MuQSS.txt @@ -0,0 +1,373 @@ +MuQSS - The Multiple Queue Skiplist Scheduler by Con Kolivas. + +MuQSS is a per-cpu runqueue variant of the original BFS scheduler with +one 8 level skiplist per runqueue, and fine grained locking for much more +scalability. + + +Goals. + +The goal of the Multiple Queue Skiplist Scheduler, referred to as MuQSS from +here on (pronounced mux) is to completely do away with the complex designs of +the past for the cpu process scheduler and instead implement one that is very +simple in basic design. The main focus of MuQSS is to achieve excellent desktop +interactivity and responsiveness without heuristics and tuning knobs that are +difficult to understand, impossible to model and predict the effect of, and when +tuned to one workload cause massive detriment to another, while still being +scalable to many CPUs and processes. + + +Design summary. + +MuQSS is best described as per-cpu multiple runqueue, O(log n) insertion, O(1) +lookup, earliest effective virtual deadline first tickless design, loosely based +on EEVDF (earliest eligible virtual deadline first) and my previous Staircase +Deadline scheduler, and evolved from the single runqueue O(n) BFS scheduler. +Each component shall be described in order to understand the significance of, +and reasoning for it. + + +Design reasoning. + +In BFS, the use of a single runqueue across all CPUs meant that each CPU would +need to scan the entire runqueue looking for the process with the earliest +deadline and schedule that next, regardless of which CPU it originally came +from. This made BFS deterministic with respect to latency and provided +guaranteed latencies dependent on number of processes and CPUs. The single +runqueue, however, meant that all CPUs would compete for the single lock +protecting it, which would lead to increasing lock contention as the number of +CPUs rose and appeared to limit scalability of common workloads beyond 16 +logical CPUs. Additionally, the O(n) lookup of the runqueue list obviously +increased overhead proportionate to the number of queued proecesses and led to +cache thrashing while iterating over the linked list. + +MuQSS is an evolution of BFS, designed to maintain the same scheduling +decision mechanism and be virtually deterministic without relying on the +constrained design of the single runqueue by splitting out the single runqueue +to be per-CPU and use skiplists instead of linked lists. + +The original reason for going back to a single runqueue design for BFS was that +once multiple runqueues are introduced, per-CPU or otherwise, there will be +complex interactions as each runqueue will be responsible for the scheduling +latency and fairness of the tasks only on its own runqueue, and to achieve +fairness and low latency across multiple CPUs, any advantage in throughput of +having CPU local tasks causes other disadvantages. This is due to requiring a +very complex balancing system to at best achieve some semblance of fairness +across CPUs and can only maintain relatively low latency for tasks bound to the +same CPUs, not across them. To increase said fairness and latency across CPUs, +the advantage of local runqueue locking, which makes for better scalability, is +lost due to having to grab multiple locks. + +MuQSS works around the problems inherent in multiple runqueue designs by +making its skip lists priority ordered and through novel use of lockless +examination of each other runqueue it can decide if it should take the earliest +deadline task from another runqueue for latency reasons, or for CPU balancing +reasons. It still does not have a balancing system, choosing to allow the +next task scheduling decision and task wakeup CPU choice to allow balancing to +happen by virtue of its choices. + +As a further evolution of the design, MuQSS normally configures sharing of +runqueues in a logical fashion for when CPU resources are shared for improved +latency and throughput. By default it shares runqueues and locks between +multicore siblings. Optionally it can be configured to run with sharing of +SMT siblings only, all SMP packages or no sharing at all. Additionally it can +be selected at boot time. + + +Design details. + +Custom skip list implementation: + +To avoid the overhead of building up and tearing down skip list structures, +the variant used by MuQSS has a number of optimisations making it specific for +its use case in the scheduler. It uses static arrays of 8 'levels' instead of +building up and tearing down structures dynamically. This makes each runqueue +only scale O(log N) up to 64k tasks. However as there is one runqueue per CPU +it means that it scales O(log N) up to 64k x number of logical CPUs which is +far beyond the realistic task limits each CPU could handle. By being 8 levels +it also makes the array exactly one cacheline in size. Additionally, each +skip list node is bidirectional making insertion and removal amortised O(1), +being O(k) where k is 1-8. Uniquely, we are only ever interested in the very +first entry in each list at all times with MuQSS, so there is never a need to +do a search and thus look up is always O(1). In interactive mode, the queues +will be searched beyond their first entry if the first task is not suitable +for affinity or SMT nice reasons. + +Task insertion: + +MuQSS inserts tasks into a per CPU runqueue as an O(log N) insertion into +a custom skip list as described above (based on the original design by William +Pugh). Insertion is ordered in such a way that there is never a need to do a +search by ordering tasks according to static priority primarily, and then +virtual deadline at the time of insertion. + +Niffies: + +Niffies are a monotonic forward moving timer not unlike the "jiffies" but are +of nanosecond resolution. Niffies are calculated per-runqueue from the high +resolution TSC timers, and in order to maintain fairness are synchronised +between CPUs whenever both runqueues are locked concurrently. + +Virtual deadline: + +The key to achieving low latency, scheduling fairness, and "nice level" +distribution in MuQSS is entirely in the virtual deadline mechanism. The one +tunable in MuQSS is the rr_interval, or "round robin interval". This is the +maximum time two SCHED_OTHER (or SCHED_NORMAL, the common scheduling policy) +tasks of the same nice level will be running for, or looking at it the other +way around, the longest duration two tasks of the same nice level will be +delayed for. When a task requests cpu time, it is given a quota (time_slice) +equal to the rr_interval and a virtual deadline. The virtual deadline is +offset from the current time in niffies by this equation: + + niffies + (prio_ratio * rr_interval) + +The prio_ratio is determined as a ratio compared to the baseline of nice -20 +and increases by 10% per nice level. The deadline is a virtual one only in that +no guarantee is placed that a task will actually be scheduled by this time, but +it is used to compare which task should go next. There are three components to +how a task is next chosen. First is time_slice expiration. If a task runs out +of its time_slice, it is descheduled, the time_slice is refilled, and the +deadline reset to that formula above. Second is sleep, where a task no longer +is requesting CPU for whatever reason. The time_slice and deadline are _not_ +adjusted in this case and are just carried over for when the task is next +scheduled. Third is preemption, and that is when a newly waking task is deemed +higher priority than a currently running task on any cpu by virtue of the fact +that it has an earlier virtual deadline than the currently running task. The +earlier deadline is the key to which task is next chosen for the first and +second cases. + +The CPU proportion of different nice tasks works out to be approximately the + + (prio_ratio difference)^2 + +The reason it is squared is that a task's deadline does not change while it is +running unless it runs out of time_slice. Thus, even if the time actually +passes the deadline of another task that is queued, it will not get CPU time +unless the current running task deschedules, and the time "base" (niffies) is +constantly moving. + +Task lookup: + +As tasks are already pre-ordered according to anticipated scheduling order in +the skip lists, lookup for the next suitable task per-runqueue is always a +matter of simply selecting the first task in the 0th level skip list entry. +In order to maintain optimal latency and fairness across CPUs, MuQSS does a +novel examination of every other runqueue in cache locality order, choosing the +best task across all runqueues. This provides near-determinism of how long any +task across the entire system may wait before receiving CPU time. The other +runqueues are first examine lockless and then trylocked to minimise the +potential lock contention if they are likely to have a suitable better task. +Each other runqueue lock is only held for as long as it takes to examine the +entry for suitability. In "interactive" mode, the default setting, MuQSS will +look for the best deadline task across all CPUs, while in !interactive mode, +it will only select a better deadline task from another CPU if it is more +heavily laden than the current one. + +Lookup is therefore O(k) where k is number of CPUs. + + +Latency. + +Through the use of virtual deadlines to govern the scheduling order of normal +tasks, queue-to-activation latency per runqueue is guaranteed to be bound by +the rr_interval tunable which is set to 6ms by default. This means that the +longest a CPU bound task will wait for more CPU is proportional to the number +of running tasks and in the common case of 0-2 running tasks per CPU, will be +under the 7ms threshold for human perception of jitter. Additionally, as newly +woken tasks will have an early deadline from their previous runtime, the very +tasks that are usually latency sensitive will have the shortest interval for +activation, usually preempting any existing CPU bound tasks. + +Tickless expiry: + +A feature of MuQSS is that it is not tied to the resolution of the chosen tick +rate in Hz, instead depending entirely on the high resolution timers where +possible for sub-millisecond accuracy on timeouts regarless of the underlying +tick rate. This allows MuQSS to be run with the low overhead of low Hz rates +such as 100 by default, benefiting from the improved throughput and lower +power usage it provides. Another advantage of this approach is that in +combination with the Full No HZ option, which disables ticks on running task +CPUs instead of just idle CPUs, the tick can be disabled at all times +regardless of how many tasks are running instead of being limited to just one +running task. Note that this option is NOT recommended for regular desktop +users. + + +Scalability and balancing. + +Unlike traditional approaches where balancing is a combination of CPU selection +at task wakeup and intermittent balancing based on a vast array of rules set +according to architecture, busyness calculations and special case management, +MuQSS indirectly balances on the fly at task wakeup and next task selection. +During initialisation, MuQSS creates a cache coherency ordered list of CPUs for +each logical CPU and uses this to aid task/CPU selection when CPUs are busy. +Additionally it selects any idle CPUs, if they are available, at any time over +busy CPUs according to the following preference: + + * Same thread, idle or busy cache, idle or busy threads + * Other core, same cache, idle or busy cache, idle threads. + * Same node, other CPU, idle cache, idle threads. + * Same node, other CPU, busy cache, idle threads. + * Other core, same cache, busy threads. + * Same node, other CPU, busy threads. + * Other node, other CPU, idle cache, idle threads. + * Other node, other CPU, busy cache, idle threads. + * Other node, other CPU, busy threads. + +Mux is therefore SMT, MC and Numa aware without the need for extra +intermittent balancing to maintain CPUs busy and make the most of cache +coherency. + + +Features + +As the initial prime target audience for MuQSS was the average desktop user, it +was designed to not need tweaking, tuning or have features set to obtain benefit +from it. Thus the number of knobs and features has been kept to an absolute +minimum and should not require extra user input for the vast majority of cases. +There are 3 optional tunables, and 2 extra scheduling policies. The rr_interval, +interactive, and iso_cpu tunables, and the SCHED_ISO and SCHED_IDLEPRIO +policies. In addition to this, MuQSS also uses sub-tick accounting. What MuQSS +does _not_ now feature is support for CGROUPS. The average user should neither +need to know what these are, nor should they need to be using them to have good +desktop behaviour. However since some applications refuse to work without +cgroups, one can enable them with MuQSS as a stub and the filesystem will be +created which will allow the applications to work. + +rr_interval: + + /proc/sys/kernel/rr_interval + +The value is in milliseconds, and the default value is set to 6. Valid values +are from 1 to 1000 Decreasing the value will decrease latencies at the cost of +decreasing throughput, while increasing it will improve throughput, but at the +cost of worsening latencies. It is based on the fact that humans can detect +jitter at approximately 7ms, so aiming for much lower latencies is pointless +under most circumstances. It is worth noting this fact when comparing the +latency performance of MuQSS to other schedulers. Worst case latencies being +higher than 7ms are far worse than average latencies not being in the +microsecond range. + +interactive: + + /proc/sys/kernel/interactive + +The value is a simple boolean of 1 for on and 0 for off and is set to on by +default. Disabling this will disable the near-determinism of MuQSS when +selecting the next task by not examining all CPUs for the earliest deadline +task, or which CPU to wake to, instead prioritising CPU balancing for improved +throughput. Latency will still be bound by rr_interval, but on a per-CPU basis +instead of across the whole system. + +Runqueue sharing. + +By default MuQSS chooses to share runqueue resources (specifically the skip +list and locking) between multicore siblings. It is configurable at build time +to select between None, SMT, MC and SMP, corresponding to no sharing, sharing +only between simultaneous mulithreading siblings, multicore siblings, or +symmetric multiprocessing physical packages. Additionally it can be se at +bootime with the use of the rqshare parameter. The reason for configurability +is that some architectures have CPUs with many multicore siblings (>= 16) +where it may be detrimental to throughput to share runqueues and another +sharing option may be desirable. Additionally, more sharing than usual can +improve latency on a system-wide level at the expense of throughput if desired. + +The options are: +none, smt, mc, smp + +eg: + rqshare=mc + +Isochronous scheduling: + +Isochronous scheduling is a unique scheduling policy designed to provide +near-real-time performance to unprivileged (ie non-root) users without the +ability to starve the machine indefinitely. Isochronous tasks (which means +"same time") are set using, for example, the schedtool application like so: + + schedtool -I -e amarok + +This will start the audio application "amarok" as SCHED_ISO. How SCHED_ISO works +is that it has a priority level between true realtime tasks and SCHED_NORMAL +which would allow them to preempt all normal tasks, in a SCHED_RR fashion (ie, +if multiple SCHED_ISO tasks are running, they purely round robin at rr_interval +rate). However if ISO tasks run for more than a tunable finite amount of time, +they are then demoted back to SCHED_NORMAL scheduling. This finite amount of +time is the percentage of CPU available per CPU, configurable as a percentage in +the following "resource handling" tunable (as opposed to a scheduler tunable): + +iso_cpu: + + /proc/sys/kernel/iso_cpu + +and is set to 70% by default. It is calculated over a rolling 5 second average +Because it is the total CPU available, it means that on a multi CPU machine, it +is possible to have an ISO task running as realtime scheduling indefinitely on +just one CPU, as the other CPUs will be available. Setting this to 100 is the +equivalent of giving all users SCHED_RR access and setting it to 0 removes the +ability to run any pseudo-realtime tasks. + +A feature of MuQSS is that it detects when an application tries to obtain a +realtime policy (SCHED_RR or SCHED_FIFO) and the caller does not have the +appropriate privileges to use those policies. When it detects this, it will +give the task SCHED_ISO policy instead. Thus it is transparent to the user. + + +Idleprio scheduling: + +Idleprio scheduling is a scheduling policy designed to give out CPU to a task +_only_ when the CPU would be otherwise idle. The idea behind this is to allow +ultra low priority tasks to be run in the background that have virtually no +effect on the foreground tasks. This is ideally suited to distributed computing +clients (like setiathome, folding, mprime etc) but can also be used to start a +video encode or so on without any slowdown of other tasks. To avoid this policy +from grabbing shared resources and holding them indefinitely, if it detects a +state where the task is waiting on I/O, the machine is about to suspend to ram +and so on, it will transiently schedule them as SCHED_NORMAL. Once a task has +been scheduled as IDLEPRIO, it cannot be put back to SCHED_NORMAL without +superuser privileges since it is effectively a lower scheduling policy. Tasks +can be set to start as SCHED_IDLEPRIO with the schedtool command like so: + +schedtool -D -e ./mprime + +Subtick accounting: + +It is surprisingly difficult to get accurate CPU accounting, and in many cases, +the accounting is done by simply determining what is happening at the precise +moment a timer tick fires off. This becomes increasingly inaccurate as the timer +tick frequency (HZ) is lowered. It is possible to create an application which +uses almost 100% CPU, yet by being descheduled at the right time, records zero +CPU usage. While the main problem with this is that there are possible security +implications, it is also difficult to determine how much CPU a task really does +use. Mux uses sub-tick accounting from the TSC clock to determine real CPU +usage. Thus, the amount of CPU reported as being used by MuQSS will more +accurately represent how much CPU the task itself is using (as is shown for +example by the 'time' application), so the reported values may be quite +different to other schedulers. When comparing throughput of MuQSS to other +designs, it is important to compare the actual completed work in terms of total +wall clock time taken and total work done, rather than the reported "cpu usage". + +Symmetric MultiThreading (SMT) aware nice: + +SMT, a.k.a. hyperthreading, is a very common feature on modern CPUs. While the +logical CPU count rises by adding thread units to each CPU core, allowing more +than one task to be run simultaneously on the same core, the disadvantage of it +is that the CPU power is shared between the tasks, not summating to the power +of two CPUs. The practical upshot of this is that two tasks running on +separate threads of the same core run significantly slower than if they had one +core each to run on. While smart CPU selection allows each task to have a core +to itself whenever available (as is done on MuQSS), it cannot offset the +slowdown that occurs when the cores are all loaded and only a thread is left. +Most of the time this is harmless as the CPU is effectively overloaded at this +point and the extra thread is of benefit. However when running a niced task in +the presence of an un-niced task (say nice 19 v nice 0), the nice task gets +precisely the same amount of CPU power as the unniced one. MuQSS has an +optional configuration feature known as SMT-NICE which selectively idles the +secondary niced thread for a period proportional to the nice difference, +allowing CPU distribution according to nice level to be maintained, at the +expense of a small amount of extra overhead. If this is configured in on a +machine without SMT threads, the overhead is minimal. + + +Con Kolivas Sat, 29th October 2016 diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig index ef179033a7c2..14b576a531ad 100644 --- a/arch/alpha/Kconfig +++ b/arch/alpha/Kconfig @@ -665,6 +665,8 @@ config HZ default 1200 if HZ_1200 default 1024 +source "kernel/Kconfig.MuQSS" + config SRM_ENV tristate "SRM environment through procfs" depends on PROC_FS diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index c77c93c485a0..c16a89549ff2 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -1237,6 +1237,8 @@ config SCHED_SMT MultiThreading at a cost of slightly increased overhead in some places. If unsure say N here. +source "kernel/Kconfig.MuQSS" + config HAVE_ARM_SCU bool help diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 5d513f461957..7cb8456280be 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -942,6 +942,8 @@ config SCHED_SMT MultiThreading at a cost of slightly increased overhead in some places. If unsure say N here. +source "kernel/Kconfig.MuQSS" + config NR_CPUS int "Maximum number of CPUs (2-4096)" range 2 4096 diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index b29d7cb38368..3af947541fdc 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -879,6 +879,8 @@ config SCHED_SMT when dealing with POWER5 cpus at a cost of slightly increased overhead in some places. If unsure say N here. +source "kernel/Kconfig.MuQSS" + config PPC_DENORMALISATION bool "PowerPC denormalisation exception handling" depends on PPC_BOOK3S_64 diff --git a/arch/powerpc/platforms/cell/spufs/sched.c b/arch/powerpc/platforms/cell/spufs/sched.c index f18d5067cd0f..fe489fc01c73 100644 --- a/arch/powerpc/platforms/cell/spufs/sched.c +++ b/arch/powerpc/platforms/cell/spufs/sched.c @@ -51,11 +51,6 @@ static struct task_struct *spusched_task; static struct timer_list spusched_timer; static struct timer_list spuloadavg_timer; -/* - * Priority of a normal, non-rt, non-niced'd process (aka nice level 0). - */ -#define NORMAL_PRIO 120 - /* * Frequency of the spu scheduler tick. By default we do one SPU scheduler * tick for every 10 CPU scheduler ticks. diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 2d3f963fd6f1..3ab85854565c 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1006,6 +1006,22 @@ config NR_CPUS config SCHED_SMT def_bool y if SMP +config SMT_NICE + bool "SMT (Hyperthreading) aware nice priority and policy support" + depends on SCHED_MUQSS && SCHED_SMT + default y + ---help--- + Enabling Hyperthreading on Intel CPUs decreases the effectiveness + of the use of 'nice' levels and different scheduling policies + (e.g. realtime) due to sharing of CPU power between hyperthreads. + SMT nice support makes each logical CPU aware of what is running on + its hyperthread siblings, maintaining appropriate distribution of + CPU according to nice levels and scheduling policies at the expense + of slightly increased overhead. + + If unsure say Y here. + + config SCHED_MC def_bool y prompt "Multi-core scheduler support" @@ -1036,6 +1052,8 @@ config SCHED_MC_PRIO If unsure say Y here. +source "kernel/Kconfig.MuQSS" + config UP_LATE_INIT def_bool y depends on !SMP && X86_LOCAL_APIC diff --git a/fs/proc/base.c b/fs/proc/base.c index eb2255e95f62..62b8cedbccb6 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -479,7 +479,7 @@ static int proc_pid_schedstat(struct seq_file *m, struct pid_namespace *ns, seq_puts(m, "0 0 0\n"); else seq_printf(m, "%llu %llu %lu\n", - (unsigned long long)task->se.sum_exec_runtime, + (unsigned long long)tsk_seruntime(task), (unsigned long long)task->sched_info.run_delay, task->sched_info.pcount); diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 2c620d7ac432..73417df5daa2 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -36,7 +36,11 @@ extern struct cred init_cred; #define INIT_PREV_CPUTIME(x) #endif +#ifdef CONFIG_SCHED_MUQSS +#define INIT_TASK_COMM "MuQSS" +#else #define INIT_TASK_COMM "swapper" +#endif /* Attach to the init_task data structure for proper alignment */ #ifdef CONFIG_ARCH_TASK_STRUCT_ON_STACK diff --git a/include/linux/ioprio.h b/include/linux/ioprio.h index e9bfe6972aed..16ba1c7e5bde 100644 --- a/include/linux/ioprio.h +++ b/include/linux/ioprio.h @@ -53,6 +53,8 @@ enum { */ static inline int task_nice_ioprio(struct task_struct *task) { + if (iso_task(task)) + return 0; return (task_nice(task) + 20) / 5; } diff --git a/include/linux/sched.h b/include/linux/sched.h index 4418f5cb8324..793cdd00d6e0 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -31,6 +31,9 @@ #include #include #include +#ifdef CONFIG_SCHED_MUQSS +#include +#endif /* task_struct member predeclarations (sorted alphabetically): */ struct audit_context; @@ -652,9 +655,11 @@ struct task_struct { unsigned int flags; unsigned int ptrace; +#if defined(CONFIG_SMP) || defined(CONFIG_SCHED_MUQSS) + int on_cpu; +#endif #ifdef CONFIG_SMP struct llist_node wake_entry; - int on_cpu; #ifdef CONFIG_THREAD_INFO_IN_TASK /* Current CPU: */ unsigned int cpu; @@ -679,10 +684,25 @@ struct task_struct { int static_prio; int normal_prio; unsigned int rt_priority; +#ifdef CONFIG_SCHED_MUQSS + int time_slice; + u64 deadline; + skiplist_node node; /* Skip list node */ + u64 last_ran; + u64 sched_time; /* sched_clock time spent running */ +#ifdef CONFIG_SMT_NICE + int smt_bias; /* Policy/nice level bias across smt siblings */ +#endif +#ifdef CONFIG_HOTPLUG_CPU + bool zerobound; /* Bound to CPU0 for hotplug */ +#endif + unsigned long rt_timeout; +#else /* CONFIG_SCHED_MUQSS */ const struct sched_class *sched_class; struct sched_entity se; struct sched_rt_entity rt; +#endif #ifdef CONFIG_CGROUP_SCHED struct task_group *sched_task_group; #endif @@ -850,6 +870,10 @@ struct task_struct { #ifdef CONFIG_ARCH_HAS_SCALED_CPUTIME u64 utimescaled; u64 stimescaled; +#endif +#ifdef CONFIG_SCHED_MUQSS + /* Unbanked cpu time */ + unsigned long utime_ns, stime_ns; #endif u64 gtime; struct prev_cputime prev_cputime; @@ -1306,6 +1330,40 @@ struct task_struct { */ }; +#ifdef CONFIG_SCHED_MUQSS +#define tsk_seruntime(t) ((t)->sched_time) +#define tsk_rttimeout(t) ((t)->rt_timeout) + +static inline void tsk_cpus_current(struct task_struct *p) +{ +} + +void print_scheduler_version(void); + +static inline bool iso_task(struct task_struct *p) +{ + return (p->policy == SCHED_ISO); +} +#else /* CFS */ +#define tsk_seruntime(t) ((t)->se.sum_exec_runtime) +#define tsk_rttimeout(t) ((t)->rt.timeout) + +static inline void tsk_cpus_current(struct task_struct *p) +{ + p->nr_cpus_allowed = current->nr_cpus_allowed; +} + +static inline void print_scheduler_version(void) +{ + printk(KERN_INFO "CFS CPU scheduler.\n"); +} + +static inline bool iso_task(struct task_struct *p) +{ + return false; +} +#endif /* CONFIG_SCHED_MUQSS */ + static inline struct pid *task_pid(struct task_struct *task) { return task->thread_pid; diff --git a/include/linux/sched/deadline.h b/include/linux/sched/deadline.h index 1aff00b65f3c..73d6319a856a 100644 --- a/include/linux/sched/deadline.h +++ b/include/linux/sched/deadline.h @@ -28,7 +28,16 @@ static inline bool dl_time_before(u64 a, u64 b) #ifdef CONFIG_SMP struct root_domain; +#ifdef CONFIG_SCHED_MUQSS +static inline void dl_clear_root_domain(struct root_domain *rd) +{ +} +static inline void dl_add_task_root_domain(struct task_struct *p) +{ +} +#else /* CONFIG_SCHED_MUQSS */ extern void dl_add_task_root_domain(struct task_struct *p); extern void dl_clear_root_domain(struct root_domain *rd); +#endif /* CONFIG_SCHED_MUQSS */ #endif /* CONFIG_SMP */ diff --git a/include/linux/sched/nohz.h b/include/linux/sched/nohz.h index 6d67e9a5af6b..101fe470aa8f 100644 --- a/include/linux/sched/nohz.h +++ b/include/linux/sched/nohz.h @@ -13,7 +13,7 @@ extern int get_nohz_timer_target(void); static inline void nohz_balance_enter_idle(int cpu) { } #endif -#ifdef CONFIG_NO_HZ_COMMON +#if defined(CONFIG_NO_HZ_COMMON) && !defined(CONFIG_SCHED_MUQSS) void calc_load_nohz_start(void); void calc_load_nohz_remote(struct rq *rq); void calc_load_nohz_stop(void); diff --git a/include/linux/sched/prio.h b/include/linux/sched/prio.h index 7d64feafc408..43c9d9e50c09 100644 --- a/include/linux/sched/prio.h +++ b/include/linux/sched/prio.h @@ -20,8 +20,20 @@ */ #define MAX_USER_RT_PRIO 100 + +#ifdef CONFIG_SCHED_MUQSS +/* Note different MAX_RT_PRIO */ +#define MAX_RT_PRIO (MAX_USER_RT_PRIO + 1) + +#define ISO_PRIO (MAX_RT_PRIO) +#define NORMAL_PRIO (MAX_RT_PRIO + 1) +#define IDLE_PRIO (MAX_RT_PRIO + 2) +#define PRIO_LIMIT ((IDLE_PRIO) + 1) +#else /* CONFIG_SCHED_MUQSS */ #define MAX_RT_PRIO MAX_USER_RT_PRIO +#endif /* CONFIG_SCHED_MUQSS */ + #define MAX_PRIO (MAX_RT_PRIO + NICE_WIDTH) #define DEFAULT_PRIO (MAX_RT_PRIO + NICE_WIDTH / 2) diff --git a/include/linux/sched/rt.h b/include/linux/sched/rt.h index e5af028c08b4..010b2244e0b6 100644 --- a/include/linux/sched/rt.h +++ b/include/linux/sched/rt.h @@ -24,8 +24,10 @@ static inline bool task_is_realtime(struct task_struct *tsk) if (policy == SCHED_FIFO || policy == SCHED_RR) return true; +#ifndef CONFIG_SCHED_MUQSS if (policy == SCHED_DEADLINE) return true; +#endif return false; } diff --git a/include/linux/sched/task.h b/include/linux/sched/task.h index 38359071236a..e2ebedb6512c 100644 --- a/include/linux/sched/task.h +++ b/include/linux/sched/task.h @@ -106,7 +106,7 @@ extern long kernel_wait4(pid_t, int __user *, int, struct rusage *); extern void free_task(struct task_struct *tsk); /* sched_exec is called by processes performing an exec */ -#ifdef CONFIG_SMP +#if defined(CONFIG_SMP) && !defined(CONFIG_SCHED_MUQSS) extern void sched_exec(void); #else #define sched_exec() {} diff --git a/include/linux/skip_list.h b/include/linux/skip_list.h new file mode 100644 index 000000000000..d4be84ba273b --- /dev/null +++ b/include/linux/skip_list.h @@ -0,0 +1,33 @@ +#ifndef _LINUX_SKIP_LISTS_H +#define _LINUX_SKIP_LISTS_H +typedef u64 keyType; +typedef void *valueType; + +typedef struct nodeStructure skiplist_node; + +struct nodeStructure { + int level; /* Levels in this structure */ + keyType key; + valueType value; + skiplist_node *next[8]; + skiplist_node *prev[8]; +}; + +typedef struct listStructure { + int entries; + int level; /* Maximum level of the list + (1 more than the number of levels in the list) */ + skiplist_node *header; /* pointer to header */ +} skiplist; + +void skiplist_init(skiplist_node *slnode); +skiplist *new_skiplist(skiplist_node *slnode); +void free_skiplist(skiplist *l); +void skiplist_node_init(skiplist_node *node); +void skiplist_insert(skiplist *l, skiplist_node *node, keyType key, valueType value, unsigned int randseed); +void skiplist_delete(skiplist *l, skiplist_node *node); + +static inline bool skiplist_node_empty(skiplist_node *node) { + return (!node->next[0]); +} +#endif /* _LINUX_SKIP_LISTS_H */ diff --git a/include/uapi/linux/sched.h b/include/uapi/linux/sched.h index 3bac0a8ceab2..f48c5c5da651 100644 --- a/include/uapi/linux/sched.h +++ b/include/uapi/linux/sched.h @@ -115,9 +115,16 @@ struct clone_args { #define SCHED_FIFO 1 #define SCHED_RR 2 #define SCHED_BATCH 3 -/* SCHED_ISO: reserved but not implemented yet */ +/* SCHED_ISO: Implemented on MuQSS only */ #define SCHED_IDLE 5 +#ifdef CONFIG_SCHED_MUQSS +#define SCHED_ISO 4 +#define SCHED_IDLEPRIO SCHED_IDLE +#define SCHED_MAX (SCHED_IDLEPRIO) +#define SCHED_RANGE(policy) ((policy) <= SCHED_MAX) +#else /* CONFIG_SCHED_MUQSS */ #define SCHED_DEADLINE 6 +#endif /* CONFIG_SCHED_MUQSS */ /* Can be ORed in to make sure the process is reverted back to SCHED_NORMAL on fork */ #define SCHED_RESET_ON_FORK 0x40000000 diff --git a/init/Kconfig b/init/Kconfig index 74a5ac65644f..44bba84664f3 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -61,6 +61,18 @@ config THREAD_INFO_IN_TASK menu "General setup" +config SCHED_MUQSS + bool "MuQSS cpu scheduler" + select HIGH_RES_TIMERS + ---help--- + The Multiple Queue Skiplist Scheduler for excellent interactivity and + responsiveness on the desktop and highly scalable deterministic + low latency on any hardware. + + Say Y here. + default y + + config BROKEN bool @@ -440,7 +452,7 @@ config HAVE_SCHED_AVG_IRQ config SCHED_THERMAL_PRESSURE bool "Enable periodic averaging of thermal pressure" - depends on SMP + depends on SMP && !SCHED_MUQSS config BSD_PROCESS_ACCT bool "BSD Process Accounting" @@ -777,6 +789,7 @@ config NUMA_BALANCING depends on ARCH_SUPPORTS_NUMA_BALANCING depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY depends on SMP && NUMA && MIGRATION + depends on !SCHED_MUQSS help This option adds support for automatic NUMA aware memory/task placement. The mechanism is quite primitive and is based on migrating memory when @@ -876,9 +889,13 @@ menuconfig CGROUP_SCHED help This feature lets CPU scheduler recognize task groups and control CPU bandwidth allocation to such task groups. It uses cgroups to group - tasks. + tasks. In combination with MuQSS this is purely a STUB to create the + files associated with the CPU controller cgroup but most of the + controls do nothing. This is useful for working in environments and + with applications that will only work if this control group is + present. -if CGROUP_SCHED +if CGROUP_SCHED && !SCHED_MUQSS config FAIR_GROUP_SCHED bool "Group scheduling for SCHED_OTHER" depends on CGROUP_SCHED @@ -1007,6 +1024,7 @@ config CGROUP_DEVICE config CGROUP_CPUACCT bool "Simple CPU accounting controller" + depends on !SCHED_MUQSS help Provides a simple controller for monitoring the total CPU consumed by the tasks in a cgroup. @@ -1134,6 +1152,7 @@ config CHECKPOINT_RESTORE config SCHED_AUTOGROUP bool "Automatic process group scheduling" + depends on !SCHED_MUQSS select CGROUPS select CGROUP_SCHED select FAIR_GROUP_SCHED diff --git a/init/init_task.c b/init/init_task.c index bd403ed3e418..5df65b2578eb 100644 --- a/init/init_task.c +++ b/init/init_task.c @@ -67,9 +67,17 @@ struct task_struct init_task .stack = init_stack, .usage = REFCOUNT_INIT(2), .flags = PF_KTHREAD, +#ifdef CONFIG_SCHED_MUQSS + .prio = NORMAL_PRIO, + .static_prio = MAX_PRIO - 20, + .normal_prio = NORMAL_PRIO, + .deadline = 0, + .time_slice = 1000000, +#else .prio = MAX_PRIO - 20, .static_prio = MAX_PRIO - 20, .normal_prio = MAX_PRIO - 20, +#endif .policy = SCHED_NORMAL, .cpus_ptr = &init_task.cpus_mask, .cpus_mask = CPU_MASK_ALL, @@ -79,6 +87,7 @@ struct task_struct init_task .restart_block = { .fn = do_no_restart_syscall, }, +#ifndef CONFIG_SCHED_MUQSS .se = { .group_node = LIST_HEAD_INIT(init_task.se.group_node), }, @@ -86,6 +95,7 @@ struct task_struct init_task .run_list = LIST_HEAD_INIT(init_task.rt.run_list), .time_slice = RR_TIMESLICE, }, +#endif .tasks = LIST_HEAD_INIT(init_task.tasks), #ifdef CONFIG_SMP .pushable_tasks = PLIST_NODE_INIT(init_task.pushable_tasks, MAX_PRIO), diff --git a/init/main.c b/init/main.c index 03371976d387..63243a24de9b 100644 --- a/init/main.c +++ b/init/main.c @@ -1411,6 +1411,8 @@ static int __ref kernel_init(void *unused) rcu_end_inkernel_boot(); + print_scheduler_version(); + if (ramdisk_execute_command) { ret = run_init_process(ramdisk_execute_command); if (!ret) diff --git a/kernel/Kconfig.MuQSS b/kernel/Kconfig.MuQSS new file mode 100644 index 000000000000..a6a58781ef91 --- /dev/null +++ b/kernel/Kconfig.MuQSS @@ -0,0 +1,105 @@ +choice + prompt "CPU scheduler runqueue sharing" + default RQ_MC if SCHED_MUQSS + default RQ_NONE + +config RQ_NONE + bool "No sharing" + help + This is the default behaviour where the CPU scheduler has one runqueue + per CPU, whether it is a physical or logical CPU (hyperthread). + + This can still be enabled runtime with the boot parameter + rqshare=none + + If unsure, say N. + +config RQ_SMT + bool "SMT (hyperthread) siblings" + depends on SCHED_SMT && SCHED_MUQSS + + help + With this option enabled, the CPU scheduler will have one runqueue + shared by SMT (hyperthread) siblings. As these logical cores share + one physical core, sharing the runqueue resource can lead to decreased + overhead, lower latency and higher throughput. + + This can still be enabled runtime with the boot parameter + rqshare=smt + + If unsure, say N. + +config RQ_MC + bool "Multicore siblings" + depends on SCHED_MC && SCHED_MUQSS + help + With this option enabled, the CPU scheduler will have one runqueue + shared by multicore siblings in addition to any SMT siblings. + As these physical cores share caches, sharing the runqueue resource + will lead to lower latency, but its effects on overhead and throughput + are less predictable. As a general rule, 6 or fewer cores will likely + benefit from this, while larger CPUs will only derive a latency + benefit. If your workloads are primarily single threaded, this will + possibly worsen throughput. If you are only concerned about latency + then enable this regardless of how many cores you have. + + This can still be enabled runtime with the boot parameter + rqshare=mc + + If unsure, say Y. + +config RQ_MC_LLC + bool "Multicore siblings (LLC)" + depends on SCHED_MC && SCHED_MUQSS + help + With this option enabled, the CPU scheduler will behave similarly as + with "Multicore siblings". + This option takes LLC cache into account when scheduling tasks. + Option may benefit CPUs with multiple LLC caches, such as Ryzen + and Xeon CPUs. + + This can still be enabled runtime with the boot parameter + rqshare=llc + + If unsure, say N. + +config RQ_SMP + bool "Symmetric Multi-Processing" + depends on SMP && SCHED_MUQSS + help + With this option enabled, the CPU scheduler will have one runqueue + shared by all physical CPUs unless they are on separate NUMA nodes. + As physical CPUs usually do not share resources, sharing the runqueue + will normally worsen throughput but improve latency. If you only + care about latency enable this. + + This can still be enabled runtime with the boot parameter + rqshare=smp + + If unsure, say N. + +config RQ_ALL + bool "NUMA" + depends on SMP && SCHED_MUQSS + help + With this option enabled, the CPU scheduler will have one runqueue + regardless of the architecture configuration, including across NUMA + nodes. This can substantially decrease throughput in NUMA + configurations, but light NUMA designs will not be dramatically + affected. This option should only be chosen if latency is the prime + concern. + + This can still be enabled runtime with the boot parameter + rqshare=all + + If unsure, say N. +endchoice + +config SHARERQ + int + default 0 if RQ_NONE + default 1 if RQ_SMT + default 2 if RQ_MC + default 3 if RQ_MC_LLC + default 4 if RQ_SMP + default 5 if RQ_ALL diff --git a/kernel/Makefile b/kernel/Makefile index 4cb4130ced32..b11afae9eea8 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -10,7 +10,7 @@ obj-y = fork.o exec_domain.o panic.o \ extable.o params.o \ kthread.o sys_ni.o nsproxy.o \ notifier.o ksysfs.o cred.o reboot.o \ - async.o range.o smpboot.o ucount.o + async.o range.o smpboot.o ucount.o skip_list.o obj-$(CONFIG_MODULES) += kmod.o obj-$(CONFIG_MULTIUSER) += groups.o diff --git a/kernel/delayacct.c b/kernel/delayacct.c index 27725754ac99..769d773c7182 100644 --- a/kernel/delayacct.c +++ b/kernel/delayacct.c @@ -106,7 +106,7 @@ int __delayacct_add_tsk(struct taskstats *d, struct task_struct *tsk) */ t1 = tsk->sched_info.pcount; t2 = tsk->sched_info.run_delay; - t3 = tsk->se.sum_exec_runtime; + t3 = tsk_seruntime(tsk); d->cpu_count += t1; diff --git a/kernel/exit.c b/kernel/exit.c index ce2a75bc0ade..f0f864bc1ab9 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -122,7 +122,7 @@ static void __exit_signal(struct task_struct *tsk) sig->curr_target = next_thread(tsk); } - add_device_randomness((const void*) &tsk->se.sum_exec_runtime, + add_device_randomness((const void*) &tsk_seruntime(tsk), sizeof(unsigned long long)); /* @@ -143,7 +143,7 @@ static void __exit_signal(struct task_struct *tsk) sig->inblock += task_io_get_inblock(tsk); sig->oublock += task_io_get_oublock(tsk); task_io_accounting_add(&sig->ioac, &tsk->ioac); - sig->sum_sched_runtime += tsk->se.sum_exec_runtime; + sig->sum_sched_runtime += tsk_seruntime(tsk); sig->nr_threads--; __unhash_process(tsk, group_dead); write_sequnlock(&sig->stats_lock); diff --git a/kernel/kthread.c b/kernel/kthread.c index bfbfa481be3a..f5942fb29ba8 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -446,6 +446,34 @@ void kthread_bind(struct task_struct *p, unsigned int cpu) } EXPORT_SYMBOL(kthread_bind); +#if defined(CONFIG_SCHED_MUQSS) && defined(CONFIG_SMP) +extern void __do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask); + +/* + * new_kthread_bind is a special variant of __kthread_bind_mask. + * For new threads to work on muqss we want to call do_set_cpus_allowed + * without the task_cpu being set and the task rescheduled until they're + * rescheduled on their own so we call __do_set_cpus_allowed directly which + * only changes the cpumask. This is particularly important for smpboot threads + * to work. + */ +static void new_kthread_bind(struct task_struct *p, unsigned int cpu) +{ + unsigned long flags; + + if (WARN_ON(!wait_task_inactive(p, TASK_UNINTERRUPTIBLE))) + return; + + /* It's safe because the task is inactive. */ + raw_spin_lock_irqsave(&p->pi_lock, flags); + __do_set_cpus_allowed(p, cpumask_of(cpu)); + p->flags |= PF_NO_SETAFFINITY; + raw_spin_unlock_irqrestore(&p->pi_lock, flags); +} +#else +#define new_kthread_bind(p, cpu) kthread_bind(p, cpu) +#endif + /** * kthread_create_on_cpu - Create a cpu bound kthread * @threadfn: the function to run until signal_pending(current). @@ -467,7 +495,7 @@ struct task_struct *kthread_create_on_cpu(int (*threadfn)(void *data), cpu); if (IS_ERR(p)) return p; - kthread_bind(p, cpu); + new_kthread_bind(p, cpu); /* CPU hotplug need to bind once again when unparking the thread. */ set_bit(KTHREAD_IS_PER_CPU, &to_kthread(p)->flags); to_kthread(p)->cpu = cpu; diff --git a/kernel/livepatch/transition.c b/kernel/livepatch/transition.c index f6310f848f34..825f9b8e228f 100644 --- a/kernel/livepatch/transition.c +++ b/kernel/livepatch/transition.c @@ -282,7 +282,7 @@ static bool klp_try_switch_task(struct task_struct *task) { static char err_buf[STACK_ERR_BUF_SIZE]; struct rq *rq; - struct rq_flags flags; + struct rq_flags rf; int ret; bool success = false; @@ -304,7 +304,7 @@ static bool klp_try_switch_task(struct task_struct *task) * functions. If all goes well, switch the task to the target patch * state. */ - rq = task_rq_lock(task, &flags); + rq = task_rq_lock(task, &rf); if (task_running(rq, task) && task != current) { snprintf(err_buf, STACK_ERR_BUF_SIZE, @@ -323,7 +323,7 @@ static bool klp_try_switch_task(struct task_struct *task) task->patch_state = klp_target_state; done: - task_rq_unlock(rq, task, &flags); + task_rq_unlock(rq, task, &rf); /* * Due to console deadlock issues, pr_debug() can't be used while diff --git a/kernel/sched/Makefile b/kernel/sched/Makefile index 21fb5a5662b5..a04ffebc6b7a 100644 --- a/kernel/sched/Makefile +++ b/kernel/sched/Makefile @@ -16,15 +16,23 @@ ifneq ($(CONFIG_SCHED_OMIT_FRAME_POINTER),y) CFLAGS_core.o := $(PROFILING) -fno-omit-frame-pointer endif +ifdef CONFIG_SCHED_MUQSS +obj-y += MuQSS.o clock.o cputime.o +obj-y += idle.o +obj-y += wait.o wait_bit.o swait.o completion.o + +obj-$(CONFIG_SMP) += topology.o +else obj-y += core.o loadavg.o clock.o cputime.o obj-y += idle.o fair.o rt.o deadline.o obj-y += wait.o wait_bit.o swait.o completion.o obj-$(CONFIG_SMP) += cpupri.o cpudeadline.o topology.o stop_task.o pelt.o obj-$(CONFIG_SCHED_AUTOGROUP) += autogroup.o -obj-$(CONFIG_SCHEDSTATS) += stats.o obj-$(CONFIG_SCHED_DEBUG) += debug.o obj-$(CONFIG_CGROUP_CPUACCT) += cpuacct.o +endif +obj-$(CONFIG_SCHEDSTATS) += stats.o obj-$(CONFIG_CPU_FREQ) += cpufreq.o obj-$(CONFIG_CPU_FREQ_GOV_SCHEDUTIL) += cpufreq_schedutil.o obj-$(CONFIG_MEMBARRIER) += membarrier.o diff --git a/kernel/sched/MuQSS.c b/kernel/sched/MuQSS.c new file mode 100644 index 000000000000..18a9b4a23e44 --- /dev/null +++ b/kernel/sched/MuQSS.c @@ -0,0 +1,7614 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * kernel/sched/MuQSS.c, was kernel/sched.c + * + * Kernel scheduler and related syscalls + * + * Copyright (C) 1991-2002 Linus Torvalds + * + * 1996-12-23 Modified by Dave Grothe to fix bugs in semaphores and + * make semaphores SMP safe + * 1998-11-19 Implemented schedule_timeout() and related stuff + * by Andrea Arcangeli + * 2002-01-04 New ultra-scalable O(1) scheduler by Ingo Molnar: + * hybrid priority-list and round-robin design with + * an array-switch method of distributing timeslices + * and per-CPU runqueues. Cleanups and useful suggestions + * by Davide Libenzi, preemptible kernel bits by Robert Love. + * 2003-09-03 Interactivity tuning by Con Kolivas. + * 2004-04-02 Scheduler domains code by Nick Piggin + * 2007-04-15 Work begun on replacing all interactivity tuning with a + * fair scheduling design by Con Kolivas. + * 2007-05-05 Load balancing (smp-nice) and other improvements + * by Peter Williams + * 2007-05-06 Interactivity improvements to CFS by Mike Galbraith + * 2007-07-01 Group scheduling enhancements by Srivatsa Vaddagiri + * 2007-11-29 RT balancing improvements by Steven Rostedt, Gregory Haskins, + * Thomas Gleixner, Mike Kravetz + * 2009-08-13 Brainfuck deadline scheduling policy by Con Kolivas deletes + * a whole lot of those previous things. + * 2016-10-01 Multiple Queue Skiplist Scheduler scalable evolution of BFS + * scheduler by Con Kolivas. + * 2019-08-31 LLC bits by Eduards Bezverhijs + */ + +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include "../workqueue_internal.h" +#include "../../fs/io-wq.h" +#include "../smpboot.h" + +#define CREATE_TRACE_POINTS +#include + +#include "MuQSS.h" + +#define rt_prio(prio) unlikely((prio) < MAX_RT_PRIO) +#define rt_task(p) rt_prio((p)->prio) +#define batch_task(p) (unlikely((p)->policy == SCHED_BATCH)) +#define is_rt_policy(policy) ((policy) == SCHED_FIFO || \ + (policy) == SCHED_RR) +#define has_rt_policy(p) unlikely(is_rt_policy((p)->policy)) + +#define is_idle_policy(policy) ((policy) == SCHED_IDLEPRIO) +#define idleprio_task(p) unlikely(is_idle_policy((p)->policy)) +#define task_running_idle(p) unlikely((p)->prio == IDLE_PRIO) + +#define is_iso_policy(policy) ((policy) == SCHED_ISO) +#define iso_task(p) unlikely(is_iso_policy((p)->policy)) +#define task_running_iso(p) unlikely((p)->prio == ISO_PRIO) + +#define rq_idle(rq) ((rq)->rq_prio == PRIO_LIMIT) + +#define ISO_PERIOD (5 * HZ) + +#define STOP_PRIO (MAX_RT_PRIO - 1) + +/* + * Some helpers for converting to/from various scales. Use shifts to get + * approximate multiples of ten for less overhead. + */ +#define APPROX_NS_PS (1073741824) /* Approximate ns per second */ +#define JIFFIES_TO_NS(TIME) ((TIME) * (APPROX_NS_PS / HZ)) +#define JIFFY_NS (APPROX_NS_PS / HZ) +#define JIFFY_US (1048576 / HZ) +#define NS_TO_JIFFIES(TIME) ((TIME) / JIFFY_NS) +#define HALF_JIFFY_NS (APPROX_NS_PS / HZ / 2) +#define HALF_JIFFY_US (1048576 / HZ / 2) +#define MS_TO_NS(TIME) ((TIME) << 20) +#define MS_TO_US(TIME) ((TIME) << 10) +#define NS_TO_MS(TIME) ((TIME) >> 20) +#define NS_TO_US(TIME) ((TIME) >> 10) +#define US_TO_NS(TIME) ((TIME) << 10) +#define TICK_APPROX_NS ((APPROX_NS_PS+HZ/2)/HZ) + +#define RESCHED_US (100) /* Reschedule if less than this many μs left */ + +void print_scheduler_version(void) +{ + printk(KERN_INFO "MuQSS CPU scheduler v0.202 by Con Kolivas.\n"); +} + +/* Define RQ share levels */ +#define RQSHARE_NONE 0 +#define RQSHARE_SMT 1 +#define RQSHARE_MC 2 +#define RQSHARE_MC_LLC 3 +#define RQSHARE_SMP 4 +#define RQSHARE_ALL 5 + +/* Define locality levels */ +#define LOCALITY_SAME 0 +#define LOCALITY_SMT 1 +#define LOCALITY_MC_LLC 2 +#define LOCALITY_MC 3 +#define LOCALITY_SMP 4 +#define LOCALITY_DISTANT 5 + +/* + * This determines what level of runqueue sharing will be done and is + * configurable at boot time with the bootparam rqshare = + */ +static int rqshare __read_mostly = CONFIG_SHARERQ; /* Default RQSHARE_MC */ + +static int __init set_rqshare(char *str) +{ + if (!strncmp(str, "none", 4)) { + rqshare = RQSHARE_NONE; + return 0; + } + if (!strncmp(str, "smt", 3)) { + rqshare = RQSHARE_SMT; + return 0; + } + if (!strncmp(str, "mc", 2)) { + rqshare = RQSHARE_MC; + return 0; + } + if (!strncmp(str, "llc", 3)) { + rqshare = RQSHARE_MC_LLC; + return 0; + } + if (!strncmp(str, "smp", 3)) { + rqshare = RQSHARE_SMP; + return 0; + } + if (!strncmp(str, "all", 3)) { + rqshare = RQSHARE_ALL; + return 0; + } + return 1; +} +__setup("rqshare=", set_rqshare); + +/* + * This is the time all tasks within the same priority round robin. + * Value is in ms and set to a minimum of 6ms. + * Tunable via /proc interface. + */ +int rr_interval __read_mostly = 6; + +/* + * Tunable to choose whether to prioritise latency or throughput, simple + * binary yes or no + */ +int sched_interactive __read_mostly = 1; + +/* + * sched_iso_cpu - sysctl which determines the cpu percentage SCHED_ISO tasks + * are allowed to run five seconds as real time tasks. This is the total over + * all online cpus. + */ +int sched_iso_cpu __read_mostly = 70; + +/* + * sched_yield_type - Choose what sort of yield sched_yield will perform. + * 0: No yield. + * 1: Yield only to better priority/deadline tasks. (default) + * 2: Expire timeslice and recalculate deadline. + */ +int sched_yield_type __read_mostly = 1; + +/* + * The relative length of deadline for each priority(nice) level. + */ +static int prio_ratios[NICE_WIDTH] __read_mostly; + + +/* + * The quota handed out to tasks of all priority levels when refilling their + * time_slice. + */ +static inline int timeslice(void) +{ + return MS_TO_US(rr_interval); +} + +DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); + +#ifdef CONFIG_SMP +/* + * Total number of runqueues. Equals number of CPUs when there is no runqueue + * sharing but is usually less with SMT/MC sharing of runqueues. + */ +static int total_runqueues __read_mostly = 1; + +static cpumask_t cpu_idle_map ____cacheline_aligned_in_smp; + +struct rq *cpu_rq(int cpu) +{ + return &per_cpu(runqueues, (cpu)); +} +#define cpu_curr(cpu) (cpu_rq(cpu)->curr) + +/* + * For asym packing, by default the lower numbered cpu has higher priority. + */ +int __weak arch_asym_cpu_priority(int cpu) +{ + return -cpu; +} + +int __weak arch_sd_sibling_asym_packing(void) +{ + return 0*SD_ASYM_PACKING; +} + +#ifdef CONFIG_SCHED_SMT +DEFINE_STATIC_KEY_FALSE(sched_smt_present); +EXPORT_SYMBOL_GPL(sched_smt_present); +#endif + +#else +struct rq *uprq; +#endif /* CONFIG_SMP */ + +#include "stats.h" + +/* + * All common locking functions performed on rq->lock. rq->clock is local to + * the CPU accessing it so it can be modified just with interrupts disabled + * when we're not updating niffies. + * Looking up task_rq must be done under rq->lock to be safe. + */ + +/* + * RQ-clock updating methods: + */ + +#ifdef HAVE_SCHED_AVG_IRQ +static void update_irq_load_avg(struct rq *rq, long delta); +#else +static inline void update_irq_load_avg(struct rq *rq, long delta) {} +#endif + +static void update_rq_clock_task(struct rq *rq, s64 delta) +{ +/* + * In theory, the compile should just see 0 here, and optimize out the call + * to sched_rt_avg_update. But I don't trust it... + */ + s64 __maybe_unused steal = 0, irq_delta = 0; +#ifdef CONFIG_IRQ_TIME_ACCOUNTING + irq_delta = irq_time_read(cpu_of(rq)) - rq->prev_irq_time; + + /* + * Since irq_time is only updated on {soft,}irq_exit, we might run into + * this case when a previous update_rq_clock() happened inside a + * {soft,}irq region. + * + * When this happens, we stop ->clock_task and only update the + * prev_irq_time stamp to account for the part that fit, so that a next + * update will consume the rest. This ensures ->clock_task is + * monotonic. + * + * It does however cause some slight miss-attribution of {soft,}irq + * time, a more accurate solution would be to update the irq_time using + * the current rq->clock timestamp, except that would require using + * atomic ops. + */ + if (irq_delta > delta) + irq_delta = delta; + + rq->prev_irq_time += irq_delta; + delta -= irq_delta; +#endif +#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING + if (static_key_false((¶virt_steal_rq_enabled))) { + steal = paravirt_steal_clock(cpu_of(rq)); + steal -= rq->prev_steal_time_rq; + + if (unlikely(steal > delta)) + steal = delta; + + rq->prev_steal_time_rq += steal; + delta -= steal; + } +#endif + rq->clock_task += delta; + +#ifdef CONFIG_HAVE_SCHED_AVG_IRQ + if (irq_delta + steal) + update_irq_load_avg(rq, irq_delta + steal); +#endif +} + +static inline void update_rq_clock(struct rq *rq) +{ + s64 delta = sched_clock_cpu(cpu_of(rq)) - rq->clock; + + if (unlikely(delta < 0)) + return; + rq->clock += delta; + update_rq_clock_task(rq, delta); +} + +/* + * Niffies are a globally increasing nanosecond counter. They're only used by + * update_load_avg and time_slice_expired, however deadlines are based on them + * across CPUs. Update them whenever we will call one of those functions, and + * synchronise them across CPUs whenever we hold both runqueue locks. + */ +static inline void update_clocks(struct rq *rq) +{ + s64 ndiff, minndiff; + long jdiff; + + update_rq_clock(rq); + ndiff = rq->clock - rq->old_clock; + rq->old_clock = rq->clock; + jdiff = jiffies - rq->last_jiffy; + + /* Subtract any niffies added by balancing with other rqs */ + ndiff -= rq->niffies - rq->last_niffy; + minndiff = JIFFIES_TO_NS(jdiff) - rq->niffies + rq->last_jiffy_niffies; + if (minndiff < 0) + minndiff = 0; + ndiff = max(ndiff, minndiff); + rq->niffies += ndiff; + rq->last_niffy = rq->niffies; + if (jdiff) { + rq->last_jiffy += jdiff; + rq->last_jiffy_niffies = rq->niffies; + } +} + +/* + * Any time we have two runqueues locked we use that as an opportunity to + * synchronise niffies to the highest value as idle ticks may have artificially + * kept niffies low on one CPU and the truth can only be later. + */ +static inline void synchronise_niffies(struct rq *rq1, struct rq *rq2) +{ + if (rq1->niffies > rq2->niffies) + rq2->niffies = rq1->niffies; + else + rq1->niffies = rq2->niffies; +} + +/* + * double_rq_lock - safely lock two runqueues + * + * Note this does not disable interrupts like task_rq_lock, + * you need to do so manually before calling. + */ + +/* For when we know rq1 != rq2 */ +static inline void __double_rq_lock(struct rq *rq1, struct rq *rq2) + __acquires(rq1->lock) + __acquires(rq2->lock) +{ + if (rq1 < rq2) { + raw_spin_lock(rq1->lock); + raw_spin_lock_nested(rq2->lock, SINGLE_DEPTH_NESTING); + } else { + raw_spin_lock(rq2->lock); + raw_spin_lock_nested(rq1->lock, SINGLE_DEPTH_NESTING); + } +} + +static inline void double_rq_lock(struct rq *rq1, struct rq *rq2) + __acquires(rq1->lock) + __acquires(rq2->lock) +{ + BUG_ON(!irqs_disabled()); + if (rq1->lock == rq2->lock) { + raw_spin_lock(rq1->lock); + __acquire(rq2->lock); /* Fake it out ;) */ + } else + __double_rq_lock(rq1, rq2); + synchronise_niffies(rq1, rq2); +} + +/* + * double_rq_unlock - safely unlock two runqueues + * + * Note this does not restore interrupts like task_rq_unlock, + * you need to do so manually after calling. + */ +static inline void double_rq_unlock(struct rq *rq1, struct rq *rq2) + __releases(rq1->lock) + __releases(rq2->lock) +{ + raw_spin_unlock(rq1->lock); + if (rq1->lock != rq2->lock) + raw_spin_unlock(rq2->lock); + else + __release(rq2->lock); +} + +static inline void lock_all_rqs(void) +{ + int cpu; + + preempt_disable(); + for_each_possible_cpu(cpu) { + struct rq *rq = cpu_rq(cpu); + + do_raw_spin_lock(rq->lock); + } +} + +static inline void unlock_all_rqs(void) +{ + int cpu; + + for_each_possible_cpu(cpu) { + struct rq *rq = cpu_rq(cpu); + + do_raw_spin_unlock(rq->lock); + } + preempt_enable(); +} + +/* Specially nest trylock an rq */ +static inline bool trylock_rq(struct rq *this_rq, struct rq *rq) +{ + if (unlikely(!do_raw_spin_trylock(rq->lock))) + return false; + spin_acquire(&rq->lock->dep_map, SINGLE_DEPTH_NESTING, 1, _RET_IP_); + synchronise_niffies(this_rq, rq); + return true; +} + +/* Unlock a specially nested trylocked rq */ +static inline void unlock_rq(struct rq *rq) +{ + spin_release(&rq->lock->dep_map, _RET_IP_); + do_raw_spin_unlock(rq->lock); +} + +/* + * cmpxchg based fetch_or, macro so it works for different integer types + */ +#define fetch_or(ptr, mask) \ + ({ \ + typeof(ptr) _ptr = (ptr); \ + typeof(mask) _mask = (mask); \ + typeof(*_ptr) _old, _val = *_ptr; \ + \ + for (;;) { \ + _old = cmpxchg(_ptr, _val, _val | _mask); \ + if (_old == _val) \ + break; \ + _val = _old; \ + } \ + _old; \ +}) + +#if defined(CONFIG_SMP) && defined(TIF_POLLING_NRFLAG) +/* + * Atomically set TIF_NEED_RESCHED and test for TIF_POLLING_NRFLAG, + * this avoids any races wrt polling state changes and thereby avoids + * spurious IPIs. + */ +static bool set_nr_and_not_polling(struct task_struct *p) +{ + struct thread_info *ti = task_thread_info(p); + return !(fetch_or(&ti->flags, _TIF_NEED_RESCHED) & _TIF_POLLING_NRFLAG); +} + +/* + * Atomically set TIF_NEED_RESCHED if TIF_POLLING_NRFLAG is set. + * + * If this returns true, then the idle task promises to call + * sched_ttwu_pending() and reschedule soon. + */ +static bool set_nr_if_polling(struct task_struct *p) +{ + struct thread_info *ti = task_thread_info(p); + typeof(ti->flags) old, val = READ_ONCE(ti->flags); + + for (;;) { + if (!(val & _TIF_POLLING_NRFLAG)) + return false; + if (val & _TIF_NEED_RESCHED) + return true; + old = cmpxchg(&ti->flags, val, val | _TIF_NEED_RESCHED); + if (old == val) + break; + val = old; + } + return true; +} + +#else +static bool set_nr_and_not_polling(struct task_struct *p) +{ + set_tsk_need_resched(p); + return true; +} + +#ifdef CONFIG_SMP +static bool set_nr_if_polling(struct task_struct *p) +{ + return false; +} +#endif +#endif + +static bool __wake_q_add(struct wake_q_head *head, struct task_struct *task) +{ + struct wake_q_node *node = &task->wake_q; + + /* + * Atomically grab the task, if ->wake_q is !nil already it means + * its already queued (either by us or someone else) and will get the + * wakeup due to that. + * + * In order to ensure that a pending wakeup will observe our pending + * state, even in the failed case, an explicit smp_mb() must be used. + */ + smp_mb__before_atomic(); + if (unlikely(cmpxchg_relaxed(&node->next, NULL, WAKE_Q_TAIL))) + return false; + + /* + * The head is context local, there can be no concurrency. + */ + *head->lastp = node; + head->lastp = &node->next; + return true; +} + +/** + * wake_q_add() - queue a wakeup for 'later' waking. + * @head: the wake_q_head to add @task to + * @task: the task to queue for 'later' wakeup + * + * Queue a task for later wakeup, most likely by the wake_up_q() call in the + * same context, _HOWEVER_ this is not guaranteed, the wakeup can come + * instantly. + * + * This function must be used as-if it were wake_up_process(); IOW the task + * must be ready to be woken at this location. + */ +void wake_q_add(struct wake_q_head *head, struct task_struct *task) +{ + if (__wake_q_add(head, task)) + get_task_struct(task); +} + +/** + * wake_q_add_safe() - safely queue a wakeup for 'later' waking. + * @head: the wake_q_head to add @task to + * @task: the task to queue for 'later' wakeup + * + * Queue a task for later wakeup, most likely by the wake_up_q() call in the + * same context, _HOWEVER_ this is not guaranteed, the wakeup can come + * instantly. + * + * This function must be used as-if it were wake_up_process(); IOW the task + * must be ready to be woken at this location. + * + * This function is essentially a task-safe equivalent to wake_q_add(). Callers + * that already hold reference to @task can call the 'safe' version and trust + * wake_q to do the right thing depending whether or not the @task is already + * queued for wakeup. + */ +void wake_q_add_safe(struct wake_q_head *head, struct task_struct *task) +{ + if (!__wake_q_add(head, task)) + put_task_struct(task); +} + +void wake_up_q(struct wake_q_head *head) +{ + struct wake_q_node *node = head->first; + + while (node != WAKE_Q_TAIL) { + struct task_struct *task; + + task = container_of(node, struct task_struct, wake_q); + BUG_ON(!task); + /* Task can safely be re-inserted now */ + node = node->next; + task->wake_q.next = NULL; + + /* + * wake_up_process() executes a full barrier, which pairs with + * the queueing in wake_q_add() so as not to miss wakeups. + */ + wake_up_process(task); + put_task_struct(task); + } +} + +static inline void smp_sched_reschedule(int cpu) +{ + if (likely(cpu_online(cpu))) + smp_send_reschedule(cpu); +} + +/* + * resched_task - mark a task 'to be rescheduled now'. + * + * On UP this means the setting of the need_resched flag, on SMP it + * might also involve a cross-CPU call to trigger the scheduler on + * the target CPU. + */ +void resched_task(struct task_struct *p) +{ + int cpu; +#ifdef CONFIG_LOCKDEP + /* Kernel threads call this when creating workqueues while still + * inactive from __kthread_bind_mask, holding only the pi_lock */ + if (!(p->flags & PF_KTHREAD)) { + struct rq *rq = task_rq(p); + + lockdep_assert_held(rq->lock); + } +#endif + if (test_tsk_need_resched(p)) + return; + + cpu = task_cpu(p); + if (cpu == smp_processor_id()) { + set_tsk_need_resched(p); + set_preempt_need_resched(); + return; + } + + if (set_nr_and_not_polling(p)) + smp_sched_reschedule(cpu); + else + trace_sched_wake_idle_without_ipi(cpu); +} + +/* + * A task that is not running or queued will not have a node set. + * A task that is queued but not running will have a node set. + * A task that is currently running will have ->on_cpu set but no node set. + */ +static inline bool task_queued(struct task_struct *p) +{ + return !skiplist_node_empty(&p->node); +} + +static void enqueue_task(struct rq *rq, struct task_struct *p, int flags); +static inline void resched_if_idle(struct rq *rq); + +static inline bool deadline_before(u64 deadline, u64 time) +{ + return (deadline < time); +} + +/* + * Deadline is "now" in niffies + (offset by priority). Setting the deadline + * is the key to everything. It distributes cpu fairly amongst tasks of the + * same nice value, it proportions cpu according to nice level, it means the + * task that last woke up the longest ago has the earliest deadline, thus + * ensuring that interactive tasks get low latency on wake up. The CPU + * proportion works out to the square of the virtual deadline difference, so + * this equation will give nice 19 3% CPU compared to nice 0. + */ +static inline u64 prio_deadline_diff(int user_prio) +{ + return (prio_ratios[user_prio] * rr_interval * (MS_TO_NS(1) / 128)); +} + +static inline u64 task_deadline_diff(struct task_struct *p) +{ + return prio_deadline_diff(TASK_USER_PRIO(p)); +} + +static inline u64 static_deadline_diff(int static_prio) +{ + return prio_deadline_diff(USER_PRIO(static_prio)); +} + +static inline int longest_deadline_diff(void) +{ + return prio_deadline_diff(39); +} + +static inline int ms_longest_deadline_diff(void) +{ + return NS_TO_MS(longest_deadline_diff()); +} + +static inline bool rq_local(struct rq *rq); + +#ifndef SCHED_CAPACITY_SCALE +#define SCHED_CAPACITY_SCALE 1024 +#endif + +static inline int rq_load(struct rq *rq) +{ + return rq->nr_running; +} + +/* + * Update the load average for feeding into cpu frequency governors. Use a + * rough estimate of a rolling average with ~ time constant of 32ms. + * 80/128 ~ 0.63. * 80 / 32768 / 128 == * 5 / 262144 + * Make sure a call to update_clocks has been made before calling this to get + * an updated rq->niffies. + */ +static void update_load_avg(struct rq *rq, unsigned int flags) +{ + long us_interval, load; + + us_interval = NS_TO_US(rq->niffies - rq->load_update); + if (unlikely(us_interval <= 0)) + return; + + load = rq->load_avg - (rq->load_avg * us_interval * 5 / 262144); + if (unlikely(load < 0)) + load = 0; + load += rq_load(rq) * SCHED_CAPACITY_SCALE * us_interval * 5 / 262144; + rq->load_avg = load; + + rq->load_update = rq->niffies; + update_irq_load_avg(rq, 0); + if (likely(rq_local(rq))) + cpufreq_trigger(rq, flags); +} + +#ifdef HAVE_SCHED_AVG_IRQ +/* + * IRQ variant of update_load_avg below. delta is actually time in nanoseconds + * here so we scale curload to how long it's been since the last update. + */ +static void update_irq_load_avg(struct rq *rq, long delta) +{ + long us_interval, load; + + us_interval = NS_TO_US(rq->niffies - rq->irq_load_update); + if (unlikely(us_interval <= 0)) + return; + + load = rq->irq_load_avg - (rq->irq_load_avg * us_interval * 5 / 262144); + if (unlikely(load < 0)) + load = 0; + load += NS_TO_US(delta) * SCHED_CAPACITY_SCALE * 5 / 262144; + rq->irq_load_avg = load; + + rq->irq_load_update = rq->niffies; +} +#endif + +/* + * Removing from the runqueue. Enter with rq locked. Deleting a task + * from the skip list is done via the stored node reference in the task struct + * and does not require a full look up. Thus it occurs in O(k) time where k + * is the "level" of the list the task was stored at - usually < 4, max 8. + */ +static void dequeue_task(struct rq *rq, struct task_struct *p, int flags) +{ + skiplist_delete(rq->sl, &p->node); + rq->best_key = rq->node->next[0]->key; + update_clocks(rq); + + if (!(flags & DEQUEUE_SAVE)) { + sched_info_dequeued(rq, p); + psi_dequeue(p, flags & DEQUEUE_SLEEP); + } + rq->nr_running--; + if (rt_task(p)) + rq->rt_nr_running--; + update_load_avg(rq, flags); +} + +#ifdef CONFIG_PREEMPT_RCU +static bool rcu_read_critical(struct task_struct *p) +{ + return p->rcu_read_unlock_special.b.blocked; +} +#else /* CONFIG_PREEMPT_RCU */ +#define rcu_read_critical(p) (false) +#endif /* CONFIG_PREEMPT_RCU */ + +/* + * To determine if it's safe for a task of SCHED_IDLEPRIO to actually run as + * an idle task, we ensure none of the following conditions are met. + */ +static bool idleprio_suitable(struct task_struct *p) +{ + return (!(task_contributes_to_load(p)) && !(p->flags & (PF_EXITING)) && + !signal_pending(p) && !rcu_read_critical(p) && !freezing(p)); +} + +/* + * To determine if a task of SCHED_ISO can run in pseudo-realtime, we check + * that the iso_refractory flag is not set. + */ +static inline bool isoprio_suitable(struct rq *rq) +{ + return !rq->iso_refractory; +} + +/* + * Adding to the runqueue. Enter with rq locked. + */ +static void enqueue_task(struct rq *rq, struct task_struct *p, int flags) +{ + unsigned int randseed, cflags = 0; + u64 sl_id; + + if (!rt_task(p)) { + /* Check it hasn't gotten rt from PI */ + if ((idleprio_task(p) && idleprio_suitable(p)) || + (iso_task(p) && isoprio_suitable(rq))) + p->prio = p->normal_prio; + else + p->prio = NORMAL_PRIO; + } else + rq->rt_nr_running++; + /* + * The sl_id key passed to the skiplist generates a sorted list. + * Realtime and sched iso tasks run FIFO so they only need be sorted + * according to priority. The skiplist will put tasks of the same + * key inserted later in FIFO order. Tasks of sched normal, batch + * and idleprio are sorted according to their deadlines. Idleprio + * tasks are offset by an impossibly large deadline value ensuring + * they get sorted into last positions, but still according to their + * own deadlines. This creates a "landscape" of skiplists running + * from priority 0 realtime in first place to the lowest priority + * idleprio tasks last. Skiplist insertion is an O(log n) process. + */ + if (p->prio <= ISO_PRIO) { + sl_id = p->prio; + } else { + sl_id = p->deadline; + if (idleprio_task(p)) { + if (p->prio == IDLE_PRIO) + sl_id |= 0xF000000000000000; + else + sl_id += longest_deadline_diff(); + } + } + /* + * Some architectures don't have better than microsecond resolution + * so mask out ~microseconds as the random seed for skiplist insertion. + */ + update_clocks(rq); + if (!(flags & ENQUEUE_RESTORE)) { + sched_info_queued(rq, p); + psi_enqueue(p, flags & ENQUEUE_WAKEUP); + } + + randseed = (rq->niffies >> 10) & 0xFFFFFFFF; + skiplist_insert(rq->sl, &p->node, sl_id, p, randseed); + rq->best_key = rq->node->next[0]->key; + if (p->in_iowait) + cflags |= SCHED_CPUFREQ_IOWAIT; + rq->nr_running++; + update_load_avg(rq, cflags); +} + +/* + * Returns the relative length of deadline all compared to the shortest + * deadline which is that of nice -20. + */ +static inline int task_prio_ratio(struct task_struct *p) +{ + return prio_ratios[TASK_USER_PRIO(p)]; +} + +/* + * task_timeslice - all tasks of all priorities get the exact same timeslice + * length. CPU distribution is handled by giving different deadlines to + * tasks of different priorities. Use 128 as the base value for fast shifts. + */ +static inline int task_timeslice(struct task_struct *p) +{ + return (rr_interval * task_prio_ratio(p) / 128); +} + +#ifdef CONFIG_SMP +/* Entered with rq locked */ +static inline void resched_if_idle(struct rq *rq) +{ + if (rq_idle(rq)) + resched_task(rq->curr); +} + +static inline bool rq_local(struct rq *rq) +{ + return (rq->cpu == smp_processor_id()); +} +#ifdef CONFIG_SMT_NICE +static const cpumask_t *thread_cpumask(int cpu); + +/* Find the best real time priority running on any SMT siblings of cpu and if + * none are running, the static priority of the best deadline task running. + * The lookups to the other runqueues is done lockless as the occasional wrong + * value would be harmless. */ +static int best_smt_bias(struct rq *this_rq) +{ + int other_cpu, best_bias = 0; + + for_each_cpu(other_cpu, &this_rq->thread_mask) { + struct rq *rq = cpu_rq(other_cpu); + + if (rq_idle(rq)) + continue; + if (unlikely(!rq->online)) + continue; + if (!rq->rq_mm) + continue; + if (likely(rq->rq_smt_bias > best_bias)) + best_bias = rq->rq_smt_bias; + } + return best_bias; +} + +static int task_prio_bias(struct task_struct *p) +{ + if (rt_task(p)) + return 1 << 30; + else if (task_running_iso(p)) + return 1 << 29; + else if (task_running_idle(p)) + return 0; + return MAX_PRIO - p->static_prio; +} + +static bool smt_always_schedule(struct task_struct __maybe_unused *p, struct rq __maybe_unused *this_rq) +{ + return true; +} + +static bool (*smt_schedule)(struct task_struct *p, struct rq *this_rq) = &smt_always_schedule; + +/* We've already decided p can run on CPU, now test if it shouldn't for SMT + * nice reasons. */ +static bool smt_should_schedule(struct task_struct *p, struct rq *this_rq) +{ + int best_bias, task_bias; + + /* Kernel threads always run */ + if (unlikely(!p->mm)) + return true; + if (rt_task(p)) + return true; + if (!idleprio_suitable(p)) + return true; + best_bias = best_smt_bias(this_rq); + /* The smt siblings are all idle or running IDLEPRIO */ + if (best_bias < 1) + return true; + task_bias = task_prio_bias(p); + if (task_bias < 1) + return false; + if (task_bias >= best_bias) + return true; + /* Dither 25% cpu of normal tasks regardless of nice difference */ + if (best_bias % 4 == 1) + return true; + /* Sorry, you lose */ + return false; +} +#else /* CONFIG_SMT_NICE */ +#define smt_schedule(p, this_rq) (true) +#endif /* CONFIG_SMT_NICE */ + +static inline void atomic_set_cpu(int cpu, cpumask_t *cpumask) +{ + set_bit(cpu, (volatile unsigned long *)cpumask); +} + +/* + * The cpu_idle_map stores a bitmap of all the CPUs currently idle to + * allow easy lookup of whether any suitable idle CPUs are available. + * It's cheaper to maintain a binary yes/no if there are any idle CPUs on the + * idle_cpus variable than to do a full bitmask check when we are busy. The + * bits are set atomically but read locklessly as occasional false positive / + * negative is harmless. + */ +static inline void set_cpuidle_map(int cpu) +{ + if (likely(cpu_online(cpu))) + atomic_set_cpu(cpu, &cpu_idle_map); +} + +static inline void atomic_clear_cpu(int cpu, cpumask_t *cpumask) +{ + clear_bit(cpu, (volatile unsigned long *)cpumask); +} + +static inline void clear_cpuidle_map(int cpu) +{ + atomic_clear_cpu(cpu, &cpu_idle_map); +} + +static bool suitable_idle_cpus(struct task_struct *p) +{ + return (cpumask_intersects(p->cpus_ptr, &cpu_idle_map)); +} + +/* + * Resched current on rq. We don't know if rq is local to this CPU nor if it + * is locked so we do not use an intermediate variable for the task to avoid + * having it dereferenced. + */ +static void resched_curr(struct rq *rq) +{ + int cpu; + + if (test_tsk_need_resched(rq->curr)) + return; + + rq->preempt = rq->curr; + cpu = rq->cpu; + + /* We're doing this without holding the rq lock if it's not task_rq */ + + if (cpu == smp_processor_id()) { + set_tsk_need_resched(rq->curr); + set_preempt_need_resched(); + return; + } + + if (set_nr_and_not_polling(rq->curr)) + smp_sched_reschedule(cpu); + else + trace_sched_wake_idle_without_ipi(cpu); +} + +#define CPUIDLE_DIFF_THREAD (1) +#define CPUIDLE_DIFF_CORE_LLC (2) +#define CPUIDLE_DIFF_CORE (4) +#define CPUIDLE_CACHE_BUSY (8) +#define CPUIDLE_DIFF_CPU (16) +#define CPUIDLE_THREAD_BUSY (32) +#define CPUIDLE_DIFF_NODE (64) + +/* + * The best idle CPU is chosen according to the CPUIDLE ranking above where the + * lowest value would give the most suitable CPU to schedule p onto next. The + * order works out to be the following: + * + * Same thread, idle or busy cache, idle or busy threads + * Other core, same cache, idle or busy cache, idle threads. + * Same node, other CPU, idle cache, idle threads. + * Same node, other CPU, busy cache, idle threads. + * Other core, same cache, busy threads. + * Same node, other CPU, busy threads. + * Other node, other CPU, idle cache, idle threads. + * Other node, other CPU, busy cache, idle threads. + * Other node, other CPU, busy threads. + */ +static int best_mask_cpu(int best_cpu, struct rq *rq, cpumask_t *tmpmask) +{ + int best_ranking = CPUIDLE_DIFF_NODE | CPUIDLE_THREAD_BUSY | + CPUIDLE_DIFF_CPU | CPUIDLE_CACHE_BUSY | CPUIDLE_DIFF_CORE | + CPUIDLE_DIFF_CORE_LLC | CPUIDLE_DIFF_THREAD; + int cpu_tmp; + + if (cpumask_test_cpu(best_cpu, tmpmask)) + goto out; + + for_each_cpu(cpu_tmp, tmpmask) { + int ranking, locality; + struct rq *tmp_rq; + + ranking = 0; + tmp_rq = cpu_rq(cpu_tmp); + + locality = rq->cpu_locality[cpu_tmp]; +#ifdef CONFIG_NUMA + if (locality > LOCALITY_SMP) + ranking |= CPUIDLE_DIFF_NODE; + else +#endif + if (locality > LOCALITY_MC) + ranking |= CPUIDLE_DIFF_CPU; +#ifdef CONFIG_SCHED_MC + else if (locality == LOCALITY_MC_LLC) + ranking |= CPUIDLE_DIFF_CORE_LLC; + else if (locality == LOCALITY_MC) + ranking |= CPUIDLE_DIFF_CORE; + if (!(tmp_rq->cache_idle(tmp_rq))) + ranking |= CPUIDLE_CACHE_BUSY; +#endif +#ifdef CONFIG_SCHED_SMT + if (locality == LOCALITY_SMT) + ranking |= CPUIDLE_DIFF_THREAD; +#endif + if (ranking < best_ranking +#ifdef CONFIG_SCHED_SMT + || (ranking == best_ranking && (tmp_rq->siblings_idle(tmp_rq))) +#endif + ) { + best_cpu = cpu_tmp; + best_ranking = ranking; + } + } +out: + return best_cpu; +} + +bool cpus_share_cache(int this_cpu, int that_cpu) +{ + struct rq *this_rq = cpu_rq(this_cpu); + + return (this_rq->cpu_locality[that_cpu] < LOCALITY_SMP); +} + +/* As per resched_curr but only will resched idle task */ +static inline void resched_idle(struct rq *rq) +{ + if (test_tsk_need_resched(rq->idle)) + return; + + rq->preempt = rq->idle; + + set_tsk_need_resched(rq->idle); + + if (rq_local(rq)) { + set_preempt_need_resched(); + return; + } + + smp_sched_reschedule(rq->cpu); +} + +DEFINE_PER_CPU(cpumask_t, idlemask); + +static struct rq *resched_best_idle(struct task_struct *p, int cpu) +{ + cpumask_t *tmpmask = &(per_cpu(idlemask, cpu)); + struct rq *rq; + int best_cpu; + + cpumask_and(tmpmask, p->cpus_ptr, &cpu_idle_map); + best_cpu = best_mask_cpu(cpu, task_rq(p), tmpmask); + rq = cpu_rq(best_cpu); + if (!smt_schedule(p, rq)) + return NULL; + rq->preempt = p; + resched_idle(rq); + return rq; +} + +static inline void resched_suitable_idle(struct task_struct *p) +{ + if (suitable_idle_cpus(p)) + resched_best_idle(p, task_cpu(p)); +} + +static inline struct rq *rq_order(struct rq *rq, int cpu) +{ + return rq->rq_order[cpu]; +} +#else /* CONFIG_SMP */ +static inline void set_cpuidle_map(int cpu) +{ +} + +static inline void clear_cpuidle_map(int cpu) +{ +} + +static inline bool suitable_idle_cpus(struct task_struct *p) +{ + return uprq->curr == uprq->idle; +} + +static inline void resched_suitable_idle(struct task_struct *p) +{ +} + +static inline void resched_curr(struct rq *rq) +{ + resched_task(rq->curr); +} + +static inline void resched_if_idle(struct rq *rq) +{ +} + +static inline bool rq_local(struct rq *rq) +{ + return true; +} + +static inline struct rq *rq_order(struct rq *rq, int cpu) +{ + return rq; +} + +static inline bool smt_schedule(struct task_struct *p, struct rq *rq) +{ + return true; +} +#endif /* CONFIG_SMP */ + +static inline int normal_prio(struct task_struct *p) +{ + if (has_rt_policy(p)) + return MAX_RT_PRIO - 1 - p->rt_priority; + if (idleprio_task(p)) + return IDLE_PRIO; + if (iso_task(p)) + return ISO_PRIO; + return NORMAL_PRIO; +} + +/* + * Calculate the current priority, i.e. the priority + * taken into account by the scheduler. This value might + * be boosted by RT tasks as it will be RT if the task got + * RT-boosted. If not then it returns p->normal_prio. + */ +static int effective_prio(struct task_struct *p) +{ + p->normal_prio = normal_prio(p); + /* + * If we are RT tasks or we were boosted to RT priority, + * keep the priority unchanged. Otherwise, update priority + * to the normal priority: + */ + if (!rt_prio(p->prio)) + return p->normal_prio; + return p->prio; +} + +/* + * activate_task - move a task to the runqueue. Enter with rq locked. + */ +static void activate_task(struct rq *rq, struct task_struct *p, int flags) +{ + resched_if_idle(rq); + + /* + * Sleep time is in units of nanosecs, so shift by 20 to get a + * milliseconds-range estimation of the amount of time that the task + * spent sleeping: + */ + if (unlikely(prof_on == SLEEP_PROFILING)) { + if (p->state == TASK_UNINTERRUPTIBLE) + profile_hits(SLEEP_PROFILING, (void *)get_wchan(p), + (rq->niffies - p->last_ran) >> 20); + } + + p->prio = effective_prio(p); + if (task_contributes_to_load(p)) + rq->nr_uninterruptible--; + + enqueue_task(rq, p, flags); + p->on_rq = TASK_ON_RQ_QUEUED; +} + +/* + * deactivate_task - If it's running, it's not on the runqueue and we can just + * decrement the nr_running. Enter with rq locked. + */ +static inline void deactivate_task(struct task_struct *p, struct rq *rq) +{ + if (task_contributes_to_load(p)) + rq->nr_uninterruptible++; + + p->on_rq = 0; + sched_info_dequeued(rq, p); + /* deactivate_task is always DEQUEUE_SLEEP in muqss */ + psi_dequeue(p, DEQUEUE_SLEEP); +} + +#ifdef CONFIG_SMP +void set_task_cpu(struct task_struct *p, unsigned int new_cpu) +{ + struct rq *rq; + + if (task_cpu(p) == new_cpu) + return; + + /* Do NOT call set_task_cpu on a currently queued task as we will not + * be reliably holding the rq lock after changing CPU. */ + BUG_ON(task_queued(p)); + rq = task_rq(p); + +#ifdef CONFIG_LOCKDEP + /* + * The caller should hold either p->pi_lock or rq->lock, when changing + * a task's CPU. ->pi_lock for waking tasks, rq->lock for runnable tasks. + * + * Furthermore, all task_rq users should acquire both locks, see + * task_rq_lock(). + */ + WARN_ON_ONCE(debug_locks && !(lockdep_is_held(&p->pi_lock) || + lockdep_is_held(rq->lock))); +#endif + + trace_sched_migrate_task(p, new_cpu); + rseq_migrate(p); + perf_event_task_migrate(p); + + /* + * After ->cpu is set up to a new value, task_rq_lock(p, ...) can be + * successfully executed on another CPU. We must ensure that updates of + * per-task data have been completed by this moment. + */ + smp_wmb(); + + p->wake_cpu = new_cpu; + + if (task_running(rq, p)) { + /* + * We should only be calling this on a running task if we're + * holding rq lock. + */ + lockdep_assert_held(rq->lock); + + /* + * We can't change the task_thread_info CPU on a running task + * as p will still be protected by the rq lock of the CPU it + * is still running on so we only set the wake_cpu for it to be + * lazily updated once off the CPU. + */ + return; + } + +#ifdef CONFIG_THREAD_INFO_IN_TASK + WRITE_ONCE(p->cpu, new_cpu); +#else + WRITE_ONCE(task_thread_info(p)->cpu, new_cpu); +#endif + /* We're no longer protecting p after this point since we're holding + * the wrong runqueue lock. */ +} +#endif /* CONFIG_SMP */ + +/* + * Move a task off the runqueue and take it to a cpu for it will + * become the running task. + */ +static inline void take_task(struct rq *rq, int cpu, struct task_struct *p) +{ + struct rq *p_rq = task_rq(p); + + dequeue_task(p_rq, p, DEQUEUE_SAVE); + if (p_rq != rq) { + sched_info_dequeued(p_rq, p); + sched_info_queued(rq, p); + } + set_task_cpu(p, cpu); +} + +/* + * Returns a descheduling task to the runqueue unless it is being + * deactivated. + */ +static inline void return_task(struct task_struct *p, struct rq *rq, + int cpu, bool deactivate) +{ + if (deactivate) + deactivate_task(p, rq); + else { +#ifdef CONFIG_SMP + /* + * set_task_cpu was called on the running task that doesn't + * want to deactivate so it has to be enqueued to a different + * CPU and we need its lock. Tag it to be moved with as the + * lock is dropped in finish_lock_switch. + */ + if (unlikely(p->wake_cpu != cpu)) + WRITE_ONCE(p->on_rq, TASK_ON_RQ_MIGRATING); + else +#endif + enqueue_task(rq, p, ENQUEUE_RESTORE); + } +} + +/* Enter with rq lock held. We know p is on the local cpu */ +static inline void __set_tsk_resched(struct task_struct *p) +{ + set_tsk_need_resched(p); + set_preempt_need_resched(); +} + +/** + * task_curr - is this task currently executing on a CPU? + * @p: the task in question. + * + * Return: 1 if the task is currently executing. 0 otherwise. + */ +inline int task_curr(const struct task_struct *p) +{ + return cpu_curr(task_cpu(p)) == p; +} + +#ifdef CONFIG_SMP +/* + * wait_task_inactive - wait for a thread to unschedule. + * + * If @match_state is nonzero, it's the @p->state value just checked and + * not expected to change. If it changes, i.e. @p might have woken up, + * then return zero. When we succeed in waiting for @p to be off its CPU, + * we return a positive number (its total switch count). If a second call + * a short while later returns the same number, the caller can be sure that + * @p has remained unscheduled the whole time. + * + * The caller must ensure that the task *will* unschedule sometime soon, + * else this function might spin for a *long* time. This function can't + * be called with interrupts off, or it may introduce deadlock with + * smp_call_function() if an IPI is sent by the same process we are + * waiting to become inactive. + */ +unsigned long wait_task_inactive(struct task_struct *p, long match_state) +{ + int running, queued; + struct rq_flags rf; + unsigned long ncsw; + struct rq *rq; + + for (;;) { + rq = task_rq(p); + + /* + * If the task is actively running on another CPU + * still, just relax and busy-wait without holding + * any locks. + * + * NOTE! Since we don't hold any locks, it's not + * even sure that "rq" stays as the right runqueue! + * But we don't care, since this will return false + * if the runqueue has changed and p is actually now + * running somewhere else! + */ + while (task_running(rq, p)) { + if (match_state && unlikely(p->state != match_state)) + return 0; + cpu_relax(); + } + + /* + * Ok, time to look more closely! We need the rq + * lock now, to be *sure*. If we're wrong, we'll + * just go back and repeat. + */ + rq = task_rq_lock(p, &rf); + trace_sched_wait_task(p); + running = task_running(rq, p); + queued = task_on_rq_queued(p); + ncsw = 0; + if (!match_state || p->state == match_state) + ncsw = p->nvcsw | LONG_MIN; /* sets MSB */ + task_rq_unlock(rq, p, &rf); + + /* + * If it changed from the expected state, bail out now. + */ + if (unlikely(!ncsw)) + break; + + /* + * Was it really running after all now that we + * checked with the proper locks actually held? + * + * Oops. Go back and try again.. + */ + if (unlikely(running)) { + cpu_relax(); + continue; + } + + /* + * It's not enough that it's not actively running, + * it must be off the runqueue _entirely_, and not + * preempted! + * + * So if it was still runnable (but just not actively + * running right now), it's preempted, and we should + * yield - it could be a while. + */ + if (unlikely(queued)) { + ktime_t to = NSEC_PER_SEC / HZ; + + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_hrtimeout(&to, HRTIMER_MODE_REL); + continue; + } + + /* + * Ahh, all good. It wasn't running, and it wasn't + * runnable, which means that it will never become + * running in the future either. We're all done! + */ + break; + } + + return ncsw; +} + +/*** + * kick_process - kick a running thread to enter/exit the kernel + * @p: the to-be-kicked thread + * + * Cause a process which is running on another CPU to enter + * kernel-mode, without any delay. (to get signals handled.) + * + * NOTE: this function doesn't have to take the runqueue lock, + * because all it wants to ensure is that the remote task enters + * the kernel. If the IPI races and the task has been migrated + * to another CPU then no harm is done and the purpose has been + * achieved as well. + */ +void kick_process(struct task_struct *p) +{ + int cpu; + + preempt_disable(); + cpu = task_cpu(p); + if ((cpu != smp_processor_id()) && task_curr(p)) + smp_sched_reschedule(cpu); + preempt_enable(); +} +EXPORT_SYMBOL_GPL(kick_process); +#endif + +/* + * RT tasks preempt purely on priority. SCHED_NORMAL tasks preempt on the + * basis of earlier deadlines. SCHED_IDLEPRIO don't preempt anything else or + * between themselves, they cooperatively multitask. An idle rq scores as + * prio PRIO_LIMIT so it is always preempted. + */ +static inline bool +can_preempt(struct task_struct *p, int prio, u64 deadline) +{ + /* Better static priority RT task or better policy preemption */ + if (p->prio < prio) + return true; + if (p->prio > prio) + return false; + if (p->policy == SCHED_BATCH) + return false; + /* SCHED_NORMAL and ISO will preempt based on deadline */ + if (!deadline_before(p->deadline, deadline)) + return false; + return true; +} + +#ifdef CONFIG_SMP + +/* + * Per-CPU kthreads are allowed to run on !active && online CPUs, see + * __set_cpus_allowed_ptr(). + */ +static inline bool is_cpu_allowed(struct task_struct *p, int cpu) +{ + if (!cpumask_test_cpu(cpu, p->cpus_ptr)) + return false; + + if (is_per_cpu_kthread(p)) + return cpu_online(cpu); + + return cpu_active(cpu); +} + +/* + * Check to see if p can run on cpu, and if not, whether there are any online + * CPUs it can run on instead. This only happens with the hotplug threads that + * bring up the CPUs. + */ +static inline bool sched_other_cpu(struct task_struct *p, int cpu) +{ + if (likely(cpumask_test_cpu(cpu, p->cpus_ptr))) + return false; + if (p->nr_cpus_allowed == 1) { + cpumask_t valid_mask; + + cpumask_and(&valid_mask, p->cpus_ptr, cpu_online_mask); + if (unlikely(cpumask_empty(&valid_mask))) + return false; + } + return true; +} + +static inline bool needs_other_cpu(struct task_struct *p, int cpu) +{ + if (cpumask_test_cpu(cpu, p->cpus_ptr)) + return false; + return true; +} + +#define cpu_online_map (*(cpumask_t *)cpu_online_mask) + +static void try_preempt(struct task_struct *p, struct rq *this_rq) +{ + int i, this_entries = rq_load(this_rq); + cpumask_t tmp; + + if (suitable_idle_cpus(p) && resched_best_idle(p, task_cpu(p))) + return; + + /* IDLEPRIO tasks never preempt anything but idle */ + if (p->policy == SCHED_IDLEPRIO) + return; + + cpumask_and(&tmp, &cpu_online_map, p->cpus_ptr); + + for (i = 0; i < num_online_cpus(); i++) { + struct rq *rq = this_rq->cpu_order[i]; + + if (!cpumask_test_cpu(rq->cpu, &tmp)) + continue; + + if (!sched_interactive && rq != this_rq && rq_load(rq) <= this_entries) + continue; + if (smt_schedule(p, rq) && can_preempt(p, rq->rq_prio, rq->rq_deadline)) { + /* We set rq->preempting lockless, it's a hint only */ + rq->preempting = p; + resched_curr(rq); + return; + } + } +} + +static int __set_cpus_allowed_ptr(struct task_struct *p, + const struct cpumask *new_mask, bool check); +#else /* CONFIG_SMP */ +static inline bool needs_other_cpu(struct task_struct *p, int cpu) +{ + return false; +} + +static void try_preempt(struct task_struct *p, struct rq *this_rq) +{ + if (p->policy == SCHED_IDLEPRIO) + return; + if (can_preempt(p, uprq->rq_prio, uprq->rq_deadline)) + resched_curr(uprq); +} + +static inline int __set_cpus_allowed_ptr(struct task_struct *p, + const struct cpumask *new_mask, bool check) +{ + return set_cpus_allowed_ptr(p, new_mask); +} +#endif /* CONFIG_SMP */ + +static void +ttwu_stat(struct task_struct *p, int cpu, int wake_flags) +{ + struct rq *rq; + + if (!schedstat_enabled()) + return; + + rq = this_rq(); + +#ifdef CONFIG_SMP + if (cpu == rq->cpu) { + __schedstat_inc(rq->ttwu_local); + } else { + struct sched_domain *sd; + + rcu_read_lock(); + for_each_domain(rq->cpu, sd) { + if (cpumask_test_cpu(cpu, sched_domain_span(sd))) { + __schedstat_inc(sd->ttwu_wake_remote); + break; + } + } + rcu_read_unlock(); + } + +#endif /* CONFIG_SMP */ + + __schedstat_inc(rq->ttwu_count); +} + +/* + * Mark the task runnable and perform wakeup-preemption. + */ +static void ttwu_do_wakeup(struct rq *rq, struct task_struct *p, int wake_flags) +{ + /* + * Sync wakeups (i.e. those types of wakeups where the waker + * has indicated that it will leave the CPU in short order) + * don't trigger a preemption if there are no idle cpus, + * instead waiting for current to deschedule. + */ + if (wake_flags & WF_SYNC) + resched_suitable_idle(p); + else + try_preempt(p, rq); + p->state = TASK_RUNNING; + trace_sched_wakeup(p); +} + +static void +ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags) +{ + int en_flags = ENQUEUE_WAKEUP; + + lockdep_assert_held(rq->lock); + +#ifdef CONFIG_SMP + if (p->sched_contributes_to_load) + rq->nr_uninterruptible--; + + if (wake_flags & WF_MIGRATED) + en_flags |= ENQUEUE_MIGRATED; +#endif + + activate_task(rq, p, en_flags); + ttwu_do_wakeup(rq, p, wake_flags); +} + +/* + * Called in case the task @p isn't fully descheduled from its runqueue, + * in this case we must do a remote wakeup. Its a 'light' wakeup though, + * since all we need to do is flip p->state to TASK_RUNNING, since + * the task is still ->on_rq. + */ +static int ttwu_remote(struct task_struct *p, int wake_flags) +{ + struct rq *rq; + int ret = 0; + + rq = __task_rq_lock(p, NULL); + if (likely(task_on_rq_queued(p))) { + ttwu_do_wakeup(rq, p, wake_flags); + ret = 1; + } + __task_rq_unlock(rq, NULL); + + return ret; +} + +#ifdef CONFIG_SMP +void sched_ttwu_pending(void) +{ + struct rq *rq = this_rq(); + struct llist_node *llist = llist_del_all(&rq->wake_list); + struct task_struct *p, *t; + struct rq_flags rf; + + if (!llist) + return; + + rq_lock_irqsave(rq, &rf); + + llist_for_each_entry_safe(p, t, llist, wake_entry) + ttwu_do_activate(rq, p, 0); + + rq_unlock_irqrestore(rq, &rf); +} + +void scheduler_ipi(void) +{ + /* + * Fold TIF_NEED_RESCHED into the preempt_count; anybody setting + * TIF_NEED_RESCHED remotely (for the first time) will also send + * this IPI. + */ + preempt_fold_need_resched(); + + if (llist_empty(&this_rq()->wake_list) && (!idle_cpu(smp_processor_id()) || need_resched())) + return; + + /* + * Not all reschedule IPI handlers call irq_enter/irq_exit, since + * traditionally all their work was done from the interrupt return + * path. Now that we actually do some work, we need to make sure + * we do call them. + * + * Some archs already do call them, luckily irq_enter/exit nest + * properly. + * + * Arguably we should visit all archs and update all handlers, + * however a fair share of IPIs are still resched only so this would + * somewhat pessimize the simple resched case. + */ + irq_enter(); + sched_ttwu_pending(); + irq_exit(); +} + +static void ttwu_queue_remote(struct task_struct *p, int cpu, int wake_flags) +{ + struct rq *rq = cpu_rq(cpu); + + if (llist_add(&p->wake_entry, &cpu_rq(cpu)->wake_list)) { + if (!set_nr_if_polling(rq->idle)) + smp_sched_reschedule(cpu); + else + trace_sched_wake_idle_without_ipi(cpu); + } +} + +void wake_up_if_idle(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + struct rq_flags rf; + + rcu_read_lock(); + + if (!is_idle_task(rcu_dereference(rq->curr))) + goto out; + + if (set_nr_if_polling(rq->idle)) { + trace_sched_wake_idle_without_ipi(cpu); + } else { + rq_lock_irqsave(rq, &rf); + if (likely(is_idle_task(rq->curr))) + smp_sched_reschedule(cpu); + /* Else cpu is not in idle, do nothing here */ + rq_unlock_irqrestore(rq, &rf); + } + +out: + rcu_read_unlock(); +} + +static int valid_task_cpu(struct task_struct *p) +{ + cpumask_t valid_mask; + + if (p->flags & PF_KTHREAD) + cpumask_and(&valid_mask, p->cpus_ptr, cpu_all_mask); + else + cpumask_and(&valid_mask, p->cpus_ptr, cpu_active_mask); + + if (unlikely(!cpumask_weight(&valid_mask))) { + /* We shouldn't be hitting this any more */ + printk(KERN_WARNING "SCHED: No cpumask for %s/%d weight %d\n", p->comm, + p->pid, cpumask_weight(p->cpus_ptr)); + return cpumask_any(p->cpus_ptr); + } + return cpumask_any(&valid_mask); +} + +/* + * For a task that's just being woken up we have a valuable balancing + * opportunity so choose the nearest cache most lightly loaded runqueue. + * Entered with rq locked and returns with the chosen runqueue locked. + */ +static inline int select_best_cpu(struct task_struct *p) +{ + unsigned int idlest = ~0U; + struct rq *rq = NULL; + int i; + + if (suitable_idle_cpus(p)) { + int cpu = task_cpu(p); + + if (unlikely(needs_other_cpu(p, cpu))) + cpu = valid_task_cpu(p); + rq = resched_best_idle(p, cpu); + if (likely(rq)) + return rq->cpu; + } + + for (i = 0; i < num_online_cpus(); i++) { + struct rq *other_rq = task_rq(p)->cpu_order[i]; + int entries; + + if (!other_rq->online) + continue; + if (needs_other_cpu(p, other_rq->cpu)) + continue; + entries = rq_load(other_rq); + if (entries >= idlest) + continue; + idlest = entries; + rq = other_rq; + } + if (unlikely(!rq)) + return task_cpu(p); + return rq->cpu; +} +#else /* CONFIG_SMP */ +static int valid_task_cpu(struct task_struct *p) +{ + return 0; +} + +static inline int select_best_cpu(struct task_struct *p) +{ + return 0; +} + +static struct rq *resched_best_idle(struct task_struct *p, int cpu) +{ + return NULL; +} +#endif /* CONFIG_SMP */ + +static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags) +{ + struct rq *rq = cpu_rq(cpu); + +#if defined(CONFIG_SMP) + if (!cpus_share_cache(smp_processor_id(), cpu)) { + sched_clock_cpu(cpu); /* Sync clocks across CPUs */ + ttwu_queue_remote(p, cpu, wake_flags); + return; + } +#endif + rq_lock(rq); + ttwu_do_activate(rq, p, wake_flags); + rq_unlock(rq); +} + +/*** + * try_to_wake_up - wake up a thread + * @p: the thread to be awakened + * @state: the mask of task states that can be woken + * @wake_flags: wake modifier flags (WF_*) + * + * Put it on the run-queue if it's not already there. The "current" + * thread is always on the run-queue (except when the actual + * re-schedule is in progress), and as such you're allowed to do + * the simpler "current->state = TASK_RUNNING" to mark yourself + * runnable without the overhead of this. + * + * Return: %true if @p was woken up, %false if it was already running. + * or @state didn't match @p's state. + */ +static int +try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) +{ + unsigned long flags; + int cpu, success = 0; + + preempt_disable(); + if (p == current) { + /* + * We're waking current, this means 'p->on_rq' and 'task_cpu(p) + * == smp_processor_id()'. Together this means we can special + * case the whole 'p->on_rq && ttwu_remote()' case below + * without taking any locks. + * + * In particular: + * - we rely on Program-Order guarantees for all the ordering, + * - we're serialized against set_special_state() by virtue of + * it disabling IRQs (this allows not taking ->pi_lock). + */ + if (!(p->state & state)) + goto out; + + success = 1; + cpu = task_cpu(p); + trace_sched_waking(p); + p->state = TASK_RUNNING; + trace_sched_wakeup(p); + goto out; + } + + /* + * If we are going to wake up a thread waiting for CONDITION we + * need to ensure that CONDITION=1 done by the caller can not be + * reordered with p->state check below. This pairs with mb() in + * set_current_state() the waiting thread does. + */ + raw_spin_lock_irqsave(&p->pi_lock, flags); + smp_mb__after_spinlock(); + if (!(p->state & state)) + goto unlock; + + trace_sched_waking(p); + + /* We're going to change ->state: */ + success = 1; + cpu = task_cpu(p); + + /* + * Ensure we load p->on_rq _after_ p->state, otherwise it would + * be possible to, falsely, observe p->on_rq == 0 and get stuck + * in smp_cond_load_acquire() below. + * + * sched_ttwu_pending() try_to_wake_up() + * STORE p->on_rq = 1 LOAD p->state + * UNLOCK rq->lock + * + * __schedule() (switch to task 'p') + * LOCK rq->lock smp_rmb(); + * smp_mb__after_spinlock(); + * UNLOCK rq->lock + * + * [task p] + * STORE p->state = UNINTERRUPTIBLE LOAD p->on_rq + * + * Pairs with the LOCK+smp_mb__after_spinlock() on rq->lock in + * __schedule(). See the comment for smp_mb__after_spinlock(). + */ + smp_rmb(); + if (p->on_rq && ttwu_remote(p, wake_flags)) + goto unlock; + +#ifdef CONFIG_SMP + /* + * Ensure we load p->on_cpu _after_ p->on_rq, otherwise it would be + * possible to, falsely, observe p->on_cpu == 0. + * + * One must be running (->on_cpu == 1) in order to remove oneself + * from the runqueue. + * + * __schedule() (switch to task 'p') try_to_wake_up() + * STORE p->on_cpu = 1 LOAD p->on_rq + * UNLOCK rq->lock + * + * __schedule() (put 'p' to sleep) + * LOCK rq->lock smp_rmb(); + * smp_mb__after_spinlock(); + * STORE p->on_rq = 0 LOAD p->on_cpu + * + * Pairs with the LOCK+smp_mb__after_spinlock() on rq->lock in + * __schedule(). See the comment for smp_mb__after_spinlock(). + */ + smp_rmb(); + + /* + * If the owning (remote) CPU is still in the middle of schedule() with + * this task as prev, wait until its done referencing the task. + * + * Pairs with the smp_store_release() in finish_task(). + * + * This ensures that tasks getting woken will be fully ordered against + * their previous state and preserve Program Order. + */ + smp_cond_load_acquire(&p->on_cpu, !VAL); + + p->sched_contributes_to_load = !!task_contributes_to_load(p); + p->state = TASK_WAKING; + + if (p->in_iowait) { + delayacct_blkio_end(p); + atomic_dec(&task_rq(p)->nr_iowait); + } + + cpu = select_best_cpu(p); + if (task_cpu(p) != cpu) { + wake_flags |= WF_MIGRATED; + psi_ttwu_dequeue(p); + set_task_cpu(p, cpu); + } + +#else /* CONFIG_SMP */ + + if (p->in_iowait) { + delayacct_blkio_end(p); + atomic_dec(&task_rq(p)->nr_iowait); + } + +#endif /* CONFIG_SMP */ + + ttwu_queue(p, cpu, wake_flags); +unlock: + raw_spin_unlock_irqrestore(&p->pi_lock, flags); +out: + if (success) + ttwu_stat(p, cpu, wake_flags); + preempt_enable(); + + return success; +} + +/** + * wake_up_process - Wake up a specific process + * @p: The process to be woken up. + * + * Attempt to wake up the nominated process and move it to the set of runnable + * processes. + * + * Return: 1 if the process was woken up, 0 if it was already running. + * + * This function executes a full memory barrier before accessing the task state. + */ +int wake_up_process(struct task_struct *p) +{ + return try_to_wake_up(p, TASK_NORMAL, 0); +} +EXPORT_SYMBOL(wake_up_process); + +int wake_up_state(struct task_struct *p, unsigned int state) +{ + return try_to_wake_up(p, state, 0); +} + +static void time_slice_expired(struct task_struct *p, struct rq *rq); + +/* + * Perform scheduler related setup for a newly forked process p. + * p is forked by current. + */ +int sched_fork(unsigned long __maybe_unused clone_flags, struct task_struct *p) +{ + unsigned long flags; + +#ifdef CONFIG_PREEMPT_NOTIFIERS + INIT_HLIST_HEAD(&p->preempt_notifiers); +#endif + +#ifdef CONFIG_COMPACTION + p->capture_control = NULL; +#endif + + /* + * We mark the process as NEW here. This guarantees that + * nobody will actually run it, and a signal or other external + * event cannot wake it up and insert it on the runqueue either. + */ + p->state = TASK_NEW; + + /* + * The process state is set to the same value of the process executing + * do_fork() code. That is running. This guarantees that nobody will + * actually run it, and a signal or other external event cannot wake + * it up and insert it on the runqueue either. + */ + + /* Should be reset in fork.c but done here for ease of MuQSS patching */ + p->on_cpu = + p->on_rq = + p->utime = + p->stime = + p->sched_time = + p->stime_ns = + p->utime_ns = 0; + skiplist_node_init(&p->node); + + /* + * Revert to default priority/policy on fork if requested. + */ + if (unlikely(p->sched_reset_on_fork)) { + if (p->policy == SCHED_FIFO || p->policy == SCHED_RR || p-> policy == SCHED_ISO) { + p->policy = SCHED_NORMAL; + p->normal_prio = normal_prio(p); + } + + if (PRIO_TO_NICE(p->static_prio) < 0) { + p->static_prio = NICE_TO_PRIO(0); + p->normal_prio = p->static_prio; + } + + /* + * We don't need the reset flag anymore after the fork. It has + * fulfilled its duty: + */ + p->sched_reset_on_fork = 0; + } + + /* + * Silence PROVE_RCU. + */ + raw_spin_lock_irqsave(&p->pi_lock, flags); + set_task_cpu(p, smp_processor_id()); + raw_spin_unlock_irqrestore(&p->pi_lock, flags); + +#ifdef CONFIG_SCHED_INFO + if (unlikely(sched_info_on())) + memset(&p->sched_info, 0, sizeof(p->sched_info)); +#endif + init_task_preempt_count(p); + + return 0; +} + +#ifdef CONFIG_SCHEDSTATS + +DEFINE_STATIC_KEY_FALSE(sched_schedstats); +static bool __initdata __sched_schedstats = false; + +static void set_schedstats(bool enabled) +{ + if (enabled) + static_branch_enable(&sched_schedstats); + else + static_branch_disable(&sched_schedstats); +} + +void force_schedstat_enabled(void) +{ + if (!schedstat_enabled()) { + pr_info("kernel profiling enabled schedstats, disable via kernel.sched_schedstats.\n"); + static_branch_enable(&sched_schedstats); + } +} + +static int __init setup_schedstats(char *str) +{ + int ret = 0; + if (!str) + goto out; + + /* + * This code is called before jump labels have been set up, so we can't + * change the static branch directly just yet. Instead set a temporary + * variable so init_schedstats() can do it later. + */ + if (!strcmp(str, "enable")) { + __sched_schedstats = true; + ret = 1; + } else if (!strcmp(str, "disable")) { + __sched_schedstats = false; + ret = 1; + } +out: + if (!ret) + pr_warn("Unable to parse schedstats=\n"); + + return ret; +} +__setup("schedstats=", setup_schedstats); + +static void __init init_schedstats(void) +{ + set_schedstats(__sched_schedstats); +} + +#ifdef CONFIG_PROC_SYSCTL +int sysctl_schedstats(struct ctl_table *table, int write, + void __user *buffer, size_t *lenp, loff_t *ppos) +{ + struct ctl_table t; + int err; + int state = static_branch_likely(&sched_schedstats); + + if (write && !capable(CAP_SYS_ADMIN)) + return -EPERM; + + t = *table; + t.data = &state; + err = proc_dointvec_minmax(&t, write, buffer, lenp, ppos); + if (err < 0) + return err; + if (write) + set_schedstats(state); + return err; +} +#endif /* CONFIG_PROC_SYSCTL */ +#else /* !CONFIG_SCHEDSTATS */ +static inline void init_schedstats(void) {} +#endif /* CONFIG_SCHEDSTATS */ + +static void update_cpu_clock_switch(struct rq *rq, struct task_struct *p); + +static void account_task_cpu(struct rq *rq, struct task_struct *p) +{ + update_clocks(rq); + /* This isn't really a context switch but accounting is the same */ + update_cpu_clock_switch(rq, p); + p->last_ran = rq->niffies; +} + +bool sched_smp_initialized __read_mostly; + +static inline int hrexpiry_enabled(struct rq *rq) +{ + if (unlikely(!cpu_active(cpu_of(rq)) || !sched_smp_initialized)) + return 0; + return hrtimer_is_hres_active(&rq->hrexpiry_timer); +} + +/* + * Use HR-timers to deliver accurate preemption points. + */ +static inline void hrexpiry_clear(struct rq *rq) +{ + if (!hrexpiry_enabled(rq)) + return; + if (hrtimer_active(&rq->hrexpiry_timer)) + hrtimer_cancel(&rq->hrexpiry_timer); +} + +/* + * High-resolution time_slice expiry. + * Runs from hardirq context with interrupts disabled. + */ +static enum hrtimer_restart hrexpiry(struct hrtimer *timer) +{ + struct rq *rq = container_of(timer, struct rq, hrexpiry_timer); + struct task_struct *p; + + /* This can happen during CPU hotplug / resume */ + if (unlikely(cpu_of(rq) != smp_processor_id())) + goto out; + + /* + * We're doing this without the runqueue lock but this should always + * be run on the local CPU. Time slice should run out in __schedule + * but we set it to zero here in case niffies is slightly less. + */ + p = rq->curr; + p->time_slice = 0; + __set_tsk_resched(p); +out: + return HRTIMER_NORESTART; +} + +/* + * Called to set the hrexpiry timer state. + * + * called with irqs disabled from the local CPU only + */ +static void hrexpiry_start(struct rq *rq, u64 delay) +{ + if (!hrexpiry_enabled(rq)) + return; + + hrtimer_start(&rq->hrexpiry_timer, ns_to_ktime(delay), + HRTIMER_MODE_REL_PINNED); +} + +static void init_rq_hrexpiry(struct rq *rq) +{ + hrtimer_init(&rq->hrexpiry_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + rq->hrexpiry_timer.function = hrexpiry; +} + +static inline int rq_dither(struct rq *rq) +{ + if (!hrexpiry_enabled(rq)) + return HALF_JIFFY_US; + return 0; +} + +/* + * wake_up_new_task - wake up a newly created task for the first time. + * + * This function will do some initial scheduler statistics housekeeping + * that must be done for every newly created context, then puts the task + * on the runqueue and wakes it. + */ +void wake_up_new_task(struct task_struct *p) +{ + struct task_struct *parent, *rq_curr; + struct rq *rq, *new_rq; + unsigned long flags; + + parent = p->parent; + + raw_spin_lock_irqsave(&p->pi_lock, flags); + p->state = TASK_RUNNING; + /* Task_rq can't change yet on a new task */ + new_rq = rq = task_rq(p); + if (unlikely(needs_other_cpu(p, task_cpu(p)))) { + set_task_cpu(p, valid_task_cpu(p)); + new_rq = task_rq(p); + } + + double_rq_lock(rq, new_rq); + rq_curr = rq->curr; + + /* + * Make sure we do not leak PI boosting priority to the child. + */ + p->prio = rq_curr->normal_prio; + + trace_sched_wakeup_new(p); + + /* + * Share the timeslice between parent and child, thus the + * total amount of pending timeslices in the system doesn't change, + * resulting in more scheduling fairness. If it's negative, it won't + * matter since that's the same as being 0. rq->rq_deadline is only + * modified within schedule() so it is always equal to + * current->deadline. + */ + account_task_cpu(rq, rq_curr); + p->last_ran = rq_curr->last_ran; + if (likely(rq_curr->policy != SCHED_FIFO)) { + rq_curr->time_slice /= 2; + if (rq_curr->time_slice < RESCHED_US) { + /* + * Forking task has run out of timeslice. Reschedule it and + * start its child with a new time slice and deadline. The + * child will end up running first because its deadline will + * be slightly earlier. + */ + __set_tsk_resched(rq_curr); + time_slice_expired(p, new_rq); + if (suitable_idle_cpus(p)) + resched_best_idle(p, task_cpu(p)); + else if (unlikely(rq != new_rq)) + try_preempt(p, new_rq); + } else { + p->time_slice = rq_curr->time_slice; + if (rq_curr == parent && rq == new_rq && !suitable_idle_cpus(p)) { + /* + * The VM isn't cloned, so we're in a good position to + * do child-runs-first in anticipation of an exec. This + * usually avoids a lot of COW overhead. + */ + __set_tsk_resched(rq_curr); + } else { + /* + * Adjust the hrexpiry since rq_curr will keep + * running and its timeslice has been shortened. + */ + hrexpiry_start(rq, US_TO_NS(rq_curr->time_slice)); + try_preempt(p, new_rq); + } + } + } else { + time_slice_expired(p, new_rq); + try_preempt(p, new_rq); + } + activate_task(new_rq, p, 0); + double_rq_unlock(rq, new_rq); + raw_spin_unlock_irqrestore(&p->pi_lock, flags); +} + +#ifdef CONFIG_PREEMPT_NOTIFIERS + +static DEFINE_STATIC_KEY_FALSE(preempt_notifier_key); + +void preempt_notifier_inc(void) +{ + static_branch_inc(&preempt_notifier_key); +} +EXPORT_SYMBOL_GPL(preempt_notifier_inc); + +void preempt_notifier_dec(void) +{ + static_branch_dec(&preempt_notifier_key); +} +EXPORT_SYMBOL_GPL(preempt_notifier_dec); + +/** + * preempt_notifier_register - tell me when current is being preempted & rescheduled + * @notifier: notifier struct to register + */ +void preempt_notifier_register(struct preempt_notifier *notifier) +{ + if (!static_branch_unlikely(&preempt_notifier_key)) + WARN(1, "registering preempt_notifier while notifiers disabled\n"); + + hlist_add_head(¬ifier->link, ¤t->preempt_notifiers); +} +EXPORT_SYMBOL_GPL(preempt_notifier_register); + +/** + * preempt_notifier_unregister - no longer interested in preemption notifications + * @notifier: notifier struct to unregister + * + * This is *not* safe to call from within a preemption notifier. + */ +void preempt_notifier_unregister(struct preempt_notifier *notifier) +{ + hlist_del(¬ifier->link); +} +EXPORT_SYMBOL_GPL(preempt_notifier_unregister); + +static void __fire_sched_in_preempt_notifiers(struct task_struct *curr) +{ + struct preempt_notifier *notifier; + + hlist_for_each_entry(notifier, &curr->preempt_notifiers, link) + notifier->ops->sched_in(notifier, raw_smp_processor_id()); +} + +static __always_inline void fire_sched_in_preempt_notifiers(struct task_struct *curr) +{ + if (static_branch_unlikely(&preempt_notifier_key)) + __fire_sched_in_preempt_notifiers(curr); +} + +static void +__fire_sched_out_preempt_notifiers(struct task_struct *curr, + struct task_struct *next) +{ + struct preempt_notifier *notifier; + + hlist_for_each_entry(notifier, &curr->preempt_notifiers, link) + notifier->ops->sched_out(notifier, next); +} + +static __always_inline void +fire_sched_out_preempt_notifiers(struct task_struct *curr, + struct task_struct *next) +{ + if (static_branch_unlikely(&preempt_notifier_key)) + __fire_sched_out_preempt_notifiers(curr, next); +} + +#else /* !CONFIG_PREEMPT_NOTIFIERS */ + +static inline void fire_sched_in_preempt_notifiers(struct task_struct *curr) +{ +} + +static inline void +fire_sched_out_preempt_notifiers(struct task_struct *curr, + struct task_struct *next) +{ +} + +#endif /* CONFIG_PREEMPT_NOTIFIERS */ + +static inline void prepare_task(struct task_struct *next) +{ + /* + * Claim the task as running, we do this before switching to it + * such that any running task will have this set. + */ + next->on_cpu = 1; +} + +static inline void finish_task(struct task_struct *prev) +{ +#ifdef CONFIG_SMP + /* + * After ->on_cpu is cleared, the task can be moved to a different CPU. + * We must ensure this doesn't happen until the switch is completely + * finished. + * + * In particular, the load of prev->state in finish_task_switch() must + * happen before this. + * + * Pairs with the smp_cond_load_acquire() in try_to_wake_up(). + */ + smp_store_release(&prev->on_cpu, 0); +#endif +} + +static inline void +prepare_lock_switch(struct rq *rq, struct task_struct *next) +{ + /* + * Since the runqueue lock will be released by the next + * task (which is an invalid locking op but in the case + * of the scheduler it's an obvious special-case), so we + * do an early lockdep release here: + */ + spin_release(&rq->lock->dep_map, _THIS_IP_); +#ifdef CONFIG_DEBUG_SPINLOCK + /* this is a valid case when another task releases the spinlock */ + rq->lock->owner = next; +#endif +} + +static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) +{ + /* + * If we are tracking spinlock dependencies then we have to + * fix up the runqueue lock - which gets 'carried over' from + * prev into current: + */ + spin_acquire(&rq->lock->dep_map, 0, 0, _THIS_IP_); + +#ifdef CONFIG_SMP + /* + * If prev was marked as migrating to another CPU in return_task, drop + * the local runqueue lock but leave interrupts disabled and grab the + * remote lock we're migrating it to before enabling them. + */ + if (unlikely(task_on_rq_migrating(prev))) { + sched_info_dequeued(rq, prev); + /* + * We move the ownership of prev to the new cpu now. ttwu can't + * activate prev to the wrong cpu since it has to grab this + * runqueue in ttwu_remote. + */ +#ifdef CONFIG_THREAD_INFO_IN_TASK + prev->cpu = prev->wake_cpu; +#else + task_thread_info(prev)->cpu = prev->wake_cpu; +#endif + raw_spin_unlock(rq->lock); + + raw_spin_lock(&prev->pi_lock); + rq = __task_rq_lock(prev, NULL); + /* Check that someone else hasn't already queued prev */ + if (likely(!task_queued(prev))) { + enqueue_task(rq, prev, 0); + prev->on_rq = TASK_ON_RQ_QUEUED; + /* Wake up the CPU if it's not already running */ + resched_if_idle(rq); + } + raw_spin_unlock(&prev->pi_lock); + } +#endif + rq_unlock(rq); + local_irq_enable(); +} + +#ifndef prepare_arch_switch +# define prepare_arch_switch(next) do { } while (0) +#endif +#ifndef finish_arch_switch +# define finish_arch_switch(prev) do { } while (0) +#endif +#ifndef finish_arch_post_lock_switch +# define finish_arch_post_lock_switch() do { } while (0) +#endif + +/** + * prepare_task_switch - prepare to switch tasks + * @rq: the runqueue preparing to switch + * @next: the task we are going to switch to. + * + * This is called with the rq lock held and interrupts off. It must + * be paired with a subsequent finish_task_switch after the context + * switch. + * + * prepare_task_switch sets up locking and calls architecture specific + * hooks. + */ +static inline void +prepare_task_switch(struct rq *rq, struct task_struct *prev, + struct task_struct *next) +{ + kcov_prepare_switch(prev); + sched_info_switch(rq, prev, next); + perf_event_task_sched_out(prev, next); + rseq_preempt(prev); + fire_sched_out_preempt_notifiers(prev, next); + prepare_task(next); + prepare_arch_switch(next); +} + +/** + * finish_task_switch - clean up after a task-switch + * @rq: runqueue associated with task-switch + * @prev: the thread we just switched away from. + * + * finish_task_switch must be called after the context switch, paired + * with a prepare_task_switch call before the context switch. + * finish_task_switch will reconcile locking set up by prepare_task_switch, + * and do any other architecture-specific cleanup actions. + * + * Note that we may have delayed dropping an mm in context_switch(). If + * so, we finish that here outside of the runqueue lock. (Doing it + * with the lock held can cause deadlocks; see schedule() for + * details.) + * + * The context switch have flipped the stack from under us and restored the + * local variables which were saved when this task called schedule() in the + * past. prev == current is still correct but we need to recalculate this_rq + * because prev may have moved to another CPU. + */ +static void finish_task_switch(struct task_struct *prev) + __releases(rq->lock) +{ + struct rq *rq = this_rq(); + struct mm_struct *mm = rq->prev_mm; + long prev_state; + + /* + * The previous task will have left us with a preempt_count of 2 + * because it left us after: + * + * schedule() + * preempt_disable(); // 1 + * __schedule() + * raw_spin_lock_irq(rq->lock) // 2 + * + * Also, see FORK_PREEMPT_COUNT. + */ + if (WARN_ONCE(preempt_count() != 2*PREEMPT_DISABLE_OFFSET, + "corrupted preempt_count: %s/%d/0x%x\n", + current->comm, current->pid, preempt_count())) + preempt_count_set(FORK_PREEMPT_COUNT); + + rq->prev_mm = NULL; + + /* + * A task struct has one reference for the use as "current". + * If a task dies, then it sets TASK_DEAD in tsk->state and calls + * schedule one last time. The schedule call will never return, and + * the scheduled task must drop that reference. + * + * We must observe prev->state before clearing prev->on_cpu (in + * finish_task), otherwise a concurrent wakeup can get prev + * running on another CPU and we could rave with its RUNNING -> DEAD + * transition, resulting in a double drop. + */ + prev_state = prev->state; + vtime_task_switch(prev); + perf_event_task_sched_in(prev, current); + finish_task(prev); + finish_lock_switch(rq, prev); + finish_arch_post_lock_switch(); + kcov_finish_switch(current); + + fire_sched_in_preempt_notifiers(current); + /* + * When switching through a kernel thread, the loop in + * membarrier_{private,global}_expedited() may have observed that + * kernel thread and not issued an IPI. It is therefore possible to + * schedule between user->kernel->user threads without passing though + * switch_mm(). Membarrier requires a barrier after storing to + * rq->curr, before returning to userspace, so provide them here: + * + * - a full memory barrier for {PRIVATE,GLOBAL}_EXPEDITED, implicitly + * provided by mmdrop(), + * - a sync_core for SYNC_CORE. + */ + if (mm) { + membarrier_mm_sync_core_before_usermode(mm); + mmdrop(mm); + } + if (unlikely(prev_state == TASK_DEAD)) { + /* + * Remove function-return probe instances associated with this + * task and put them back on the free list. + */ + kprobe_flush_task(prev); + + /* Task is done with its stack. */ + put_task_stack(prev); + + put_task_struct_rcu_user(prev); + } +} + +/** + * schedule_tail - first thing a freshly forked thread must call. + * @prev: the thread we just switched away from. + */ +asmlinkage __visible void schedule_tail(struct task_struct *prev) +{ + /* + * New tasks start with FORK_PREEMPT_COUNT, see there and + * finish_task_switch() for details. + * + * finish_task_switch() will drop rq->lock() and lower preempt_count + * and the preempt_enable() will end up enabling preemption (on + * PREEMPT_COUNT kernels). + */ + + finish_task_switch(prev); + preempt_enable(); + + if (current->set_child_tid) + put_user(task_pid_vnr(current), current->set_child_tid); + + calculate_sigpending(); +} + +/* + * context_switch - switch to the new MM and the new thread's register state. + */ +static __always_inline void +context_switch(struct rq *rq, struct task_struct *prev, + struct task_struct *next) +{ + prepare_task_switch(rq, prev, next); + + /* + * For paravirt, this is coupled with an exit in switch_to to + * combine the page table reload and the switch backend into + * one hypercall. + */ + arch_start_context_switch(prev); + + /* + * kernel -> kernel lazy + transfer active + * user -> kernel lazy + mmgrab() active + * + * kernel -> user switch + mmdrop() active + * user -> user switch + */ + if (!next->mm) { // to kernel + enter_lazy_tlb(prev->active_mm, next); + + next->active_mm = prev->active_mm; + if (prev->mm) // from user + mmgrab(prev->active_mm); + else + prev->active_mm = NULL; + } else { // to user + membarrier_switch_mm(rq, prev->active_mm, next->mm); + /* + * sys_membarrier() requires an smp_mb() between setting + * rq->curr / membarrier_switch_mm() and returning to userspace. + * + * The below provides this either through switch_mm(), or in + * case 'prev->active_mm == next->mm' through + * finish_task_switch()'s mmdrop(). + */ + switch_mm_irqs_off(prev->active_mm, next->mm, next); + + if (!prev->mm) { // from kernel + /* will mmdrop() in finish_task_switch(). */ + rq->prev_mm = prev->active_mm; + prev->active_mm = NULL; + } + } + prepare_lock_switch(rq, next); + + /* Here we just switch the register state and the stack. */ + switch_to(prev, next, prev); + barrier(); + + finish_task_switch(prev); +} + +/* + * nr_running, nr_uninterruptible and nr_context_switches: + * + * externally visible scheduler statistics: current number of runnable + * threads, total number of context switches performed since bootup. + */ +unsigned long nr_running(void) +{ + unsigned long i, sum = 0; + + for_each_online_cpu(i) + sum += cpu_rq(i)->nr_running; + + return sum; +} + +static unsigned long nr_uninterruptible(void) +{ + unsigned long i, sum = 0; + + for_each_online_cpu(i) + sum += cpu_rq(i)->nr_uninterruptible; + + return sum; +} + +/* + * Check if only the current task is running on the CPU. + * + * Caution: this function does not check that the caller has disabled + * preemption, thus the result might have a time-of-check-to-time-of-use + * race. The caller is responsible to use it correctly, for example: + * + * - from a non-preemptible section (of course) + * + * - from a thread that is bound to a single CPU + * + * - in a loop with very short iterations (e.g. a polling loop) + */ +bool single_task_running(void) +{ + if (rq_load(raw_rq()) == 1) + return true; + else + return false; +} +EXPORT_SYMBOL(single_task_running); + +unsigned long long nr_context_switches(void) +{ + int cpu; + unsigned long long sum = 0; + + for_each_possible_cpu(cpu) + sum += cpu_rq(cpu)->nr_switches; + + return sum; +} + +/* + * Consumers of these two interfaces, like for example the cpufreq menu + * governor are using nonsensical data. Boosting frequency for a CPU that has + * IO-wait which might not even end up running the task when it does become + * runnable. + */ + +unsigned long nr_iowait_cpu(int cpu) +{ + return atomic_read(&cpu_rq(cpu)->nr_iowait); +} + +/* + * IO-wait accounting, and how its mostly bollocks (on SMP). + * + * The idea behind IO-wait account is to account the idle time that we could + * have spend running if it were not for IO. That is, if we were to improve the + * storage performance, we'd have a proportional reduction in IO-wait time. + * + * This all works nicely on UP, where, when a task blocks on IO, we account + * idle time as IO-wait, because if the storage were faster, it could've been + * running and we'd not be idle. + * + * This has been extended to SMP, by doing the same for each CPU. This however + * is broken. + * + * Imagine for instance the case where two tasks block on one CPU, only the one + * CPU will have IO-wait accounted, while the other has regular idle. Even + * though, if the storage were faster, both could've ran at the same time, + * utilising both CPUs. + * + * This means, that when looking globally, the current IO-wait accounting on + * SMP is a lower bound, by reason of under accounting. + * + * Worse, since the numbers are provided per CPU, they are sometimes + * interpreted per CPU, and that is nonsensical. A blocked task isn't strictly + * associated with any one particular CPU, it can wake to another CPU than it + * blocked on. This means the per CPU IO-wait number is meaningless. + * + * Task CPU affinities can make all that even more 'interesting'. + */ + +unsigned long nr_iowait(void) +{ + unsigned long cpu, sum = 0; + + for_each_possible_cpu(cpu) + sum += nr_iowait_cpu(cpu); + + return sum; +} + +unsigned long nr_active(void) +{ + return nr_running() + nr_uninterruptible(); +} + +/* Variables and functions for calc_load */ +static unsigned long calc_load_update; +unsigned long avenrun[3]; +EXPORT_SYMBOL(avenrun); + +/** + * get_avenrun - get the load average array + * @loads: pointer to dest load array + * @offset: offset to add + * @shift: shift count to shift the result left + * + * These values are estimates at best, so no need for locking. + */ +void get_avenrun(unsigned long *loads, unsigned long offset, int shift) +{ + loads[0] = (avenrun[0] + offset) << shift; + loads[1] = (avenrun[1] + offset) << shift; + loads[2] = (avenrun[2] + offset) << shift; +} + +/* + * calc_load - update the avenrun load estimates every LOAD_FREQ seconds. + */ +void calc_global_load(unsigned long ticks) +{ + long active; + + if (time_before(jiffies, READ_ONCE(calc_load_update))) + return; + active = nr_active() * FIXED_1; + + avenrun[0] = calc_load(avenrun[0], EXP_1, active); + avenrun[1] = calc_load(avenrun[1], EXP_5, active); + avenrun[2] = calc_load(avenrun[2], EXP_15, active); + + calc_load_update = jiffies + LOAD_FREQ; +} + +/** + * fixed_power_int - compute: x^n, in O(log n) time + * + * @x: base of the power + * @frac_bits: fractional bits of @x + * @n: power to raise @x to. + * + * By exploiting the relation between the definition of the natural power + * function: x^n := x*x*...*x (x multiplied by itself for n times), and + * the binary encoding of numbers used by computers: n := \Sum n_i * 2^i, + * (where: n_i \elem {0, 1}, the binary vector representing n), + * we find: x^n := x^(\Sum n_i * 2^i) := \Prod x^(n_i * 2^i), which is + * of course trivially computable in O(log_2 n), the length of our binary + * vector. + */ +static unsigned long +fixed_power_int(unsigned long x, unsigned int frac_bits, unsigned int n) +{ + unsigned long result = 1UL << frac_bits; + + if (n) { + for (;;) { + if (n & 1) { + result *= x; + result += 1UL << (frac_bits - 1); + result >>= frac_bits; + } + n >>= 1; + if (!n) + break; + x *= x; + x += 1UL << (frac_bits - 1); + x >>= frac_bits; + } + } + + return result; +} + +/* + * a1 = a0 * e + a * (1 - e) + * + * a2 = a1 * e + a * (1 - e) + * = (a0 * e + a * (1 - e)) * e + a * (1 - e) + * = a0 * e^2 + a * (1 - e) * (1 + e) + * + * a3 = a2 * e + a * (1 - e) + * = (a0 * e^2 + a * (1 - e) * (1 + e)) * e + a * (1 - e) + * = a0 * e^3 + a * (1 - e) * (1 + e + e^2) + * + * ... + * + * an = a0 * e^n + a * (1 - e) * (1 + e + ... + e^n-1) [1] + * = a0 * e^n + a * (1 - e) * (1 - e^n)/(1 - e) + * = a0 * e^n + a * (1 - e^n) + * + * [1] application of the geometric series: + * + * n 1 - x^(n+1) + * S_n := \Sum x^i = ------------- + * i=0 1 - x + */ +unsigned long +calc_load_n(unsigned long load, unsigned long exp, + unsigned long active, unsigned int n) +{ + return calc_load(load, fixed_power_int(exp, FSHIFT, n), active); +} + +DEFINE_PER_CPU(struct kernel_stat, kstat); +DEFINE_PER_CPU(struct kernel_cpustat, kernel_cpustat); + +EXPORT_PER_CPU_SYMBOL(kstat); +EXPORT_PER_CPU_SYMBOL(kernel_cpustat); + +#ifdef CONFIG_PARAVIRT +static inline u64 steal_ticks(u64 steal) +{ + if (unlikely(steal > NSEC_PER_SEC)) + return div_u64(steal, TICK_NSEC); + + return __iter_div_u64_rem(steal, TICK_NSEC, &steal); +} +#endif + +#ifndef nsecs_to_cputime +# define nsecs_to_cputime(__nsecs) nsecs_to_jiffies(__nsecs) +#endif + +/* + * On each tick, add the number of nanoseconds to the unbanked variables and + * once one tick's worth has accumulated, account it allowing for accurate + * sub-tick accounting and totals. Use the TICK_APPROX_NS to match the way we + * deduct nanoseconds. + */ +static void pc_idle_time(struct rq *rq, struct task_struct *idle, unsigned long ns) +{ + u64 *cpustat = kcpustat_this_cpu->cpustat; + unsigned long ticks; + + if (atomic_read(&rq->nr_iowait) > 0) { + rq->iowait_ns += ns; + if (rq->iowait_ns >= JIFFY_NS) { + ticks = NS_TO_JIFFIES(rq->iowait_ns); + cpustat[CPUTIME_IOWAIT] += (__force u64)TICK_APPROX_NS * ticks; + rq->iowait_ns %= JIFFY_NS; + } + } else { + rq->idle_ns += ns; + if (rq->idle_ns >= JIFFY_NS) { + ticks = NS_TO_JIFFIES(rq->idle_ns); + cpustat[CPUTIME_IDLE] += (__force u64)TICK_APPROX_NS * ticks; + rq->idle_ns %= JIFFY_NS; + } + } + acct_update_integrals(idle); +} + +static void pc_system_time(struct rq *rq, struct task_struct *p, + int hardirq_offset, unsigned long ns) +{ + u64 *cpustat = kcpustat_this_cpu->cpustat; + unsigned long ticks; + + p->stime_ns += ns; + if (p->stime_ns >= JIFFY_NS) { + ticks = NS_TO_JIFFIES(p->stime_ns); + p->stime_ns %= JIFFY_NS; + p->stime += (__force u64)TICK_APPROX_NS * ticks; + account_group_system_time(p, TICK_APPROX_NS * ticks); + } + p->sched_time += ns; + account_group_exec_runtime(p, ns); + + if (hardirq_count() - hardirq_offset) { + rq->irq_ns += ns; + if (rq->irq_ns >= JIFFY_NS) { + ticks = NS_TO_JIFFIES(rq->irq_ns); + cpustat[CPUTIME_IRQ] += (__force u64)TICK_APPROX_NS * ticks; + rq->irq_ns %= JIFFY_NS; + } + } else if (in_serving_softirq()) { + rq->softirq_ns += ns; + if (rq->softirq_ns >= JIFFY_NS) { + ticks = NS_TO_JIFFIES(rq->softirq_ns); + cpustat[CPUTIME_SOFTIRQ] += (__force u64)TICK_APPROX_NS * ticks; + rq->softirq_ns %= JIFFY_NS; + } + } else { + rq->system_ns += ns; + if (rq->system_ns >= JIFFY_NS) { + ticks = NS_TO_JIFFIES(rq->system_ns); + cpustat[CPUTIME_SYSTEM] += (__force u64)TICK_APPROX_NS * ticks; + rq->system_ns %= JIFFY_NS; + } + } + acct_update_integrals(p); +} + +static void pc_user_time(struct rq *rq, struct task_struct *p, unsigned long ns) +{ + u64 *cpustat = kcpustat_this_cpu->cpustat; + unsigned long ticks; + + p->utime_ns += ns; + if (p->utime_ns >= JIFFY_NS) { + ticks = NS_TO_JIFFIES(p->utime_ns); + p->utime_ns %= JIFFY_NS; + p->utime += (__force u64)TICK_APPROX_NS * ticks; + account_group_user_time(p, TICK_APPROX_NS * ticks); + } + p->sched_time += ns; + account_group_exec_runtime(p, ns); + + if (this_cpu_ksoftirqd() == p) { + /* + * ksoftirqd time do not get accounted in cpu_softirq_time. + * So, we have to handle it separately here. + */ + rq->softirq_ns += ns; + if (rq->softirq_ns >= JIFFY_NS) { + ticks = NS_TO_JIFFIES(rq->softirq_ns); + cpustat[CPUTIME_SOFTIRQ] += (__force u64)TICK_APPROX_NS * ticks; + rq->softirq_ns %= JIFFY_NS; + } + } + + if (task_nice(p) > 0 || idleprio_task(p)) { + rq->nice_ns += ns; + if (rq->nice_ns >= JIFFY_NS) { + ticks = NS_TO_JIFFIES(rq->nice_ns); + cpustat[CPUTIME_NICE] += (__force u64)TICK_APPROX_NS * ticks; + rq->nice_ns %= JIFFY_NS; + } + } else { + rq->user_ns += ns; + if (rq->user_ns >= JIFFY_NS) { + ticks = NS_TO_JIFFIES(rq->user_ns); + cpustat[CPUTIME_USER] += (__force u64)TICK_APPROX_NS * ticks; + rq->user_ns %= JIFFY_NS; + } + } + acct_update_integrals(p); +} + +/* + * This is called on clock ticks. + * Bank in p->sched_time the ns elapsed since the last tick or switch. + * CPU scheduler quota accounting is also performed here in microseconds. + */ +static void update_cpu_clock_tick(struct rq *rq, struct task_struct *p) +{ + s64 account_ns = rq->niffies - p->last_ran; + struct task_struct *idle = rq->idle; + + /* Accurate tick timekeeping */ + if (user_mode(get_irq_regs())) + pc_user_time(rq, p, account_ns); + else if (p != idle || (irq_count() != HARDIRQ_OFFSET)) { + pc_system_time(rq, p, HARDIRQ_OFFSET, account_ns); + } else + pc_idle_time(rq, idle, account_ns); + + /* time_slice accounting is done in usecs to avoid overflow on 32bit */ + if (p->policy != SCHED_FIFO && p != idle) + p->time_slice -= NS_TO_US(account_ns); + + p->last_ran = rq->niffies; +} + +/* + * This is called on context switches. + * Bank in p->sched_time the ns elapsed since the last tick or switch. + * CPU scheduler quota accounting is also performed here in microseconds. + */ +static void update_cpu_clock_switch(struct rq *rq, struct task_struct *p) +{ + s64 account_ns = rq->niffies - p->last_ran; + struct task_struct *idle = rq->idle; + + /* Accurate subtick timekeeping */ + if (p != idle) + pc_user_time(rq, p, account_ns); + else + pc_idle_time(rq, idle, account_ns); + + /* time_slice accounting is done in usecs to avoid overflow on 32bit */ + if (p->policy != SCHED_FIFO && p != idle) + p->time_slice -= NS_TO_US(account_ns); +} + +/* + * Return any ns on the sched_clock that have not yet been accounted in + * @p in case that task is currently running. + * + * Called with task_rq_lock(p) held. + */ +static inline u64 do_task_delta_exec(struct task_struct *p, struct rq *rq) +{ + u64 ns = 0; + + /* + * Must be ->curr _and_ ->on_rq. If dequeued, we would + * project cycles that may never be accounted to this + * thread, breaking clock_gettime(). + */ + if (p == rq->curr && task_on_rq_queued(p)) { + update_clocks(rq); + ns = rq->niffies - p->last_ran; + } + + return ns; +} + +/* + * Return accounted runtime for the task. + * Return separately the current's pending runtime that have not been + * accounted yet. + */ +unsigned long long task_sched_runtime(struct task_struct *p) +{ + struct rq_flags rf; + struct rq *rq; + u64 ns; + +#if defined(CONFIG_64BIT) && defined(CONFIG_SMP) + /* + * 64-bit doesn't need locks to atomically read a 64-bit value. + * So we have a optimisation chance when the task's delta_exec is 0. + * Reading ->on_cpu is racy, but this is ok. + * + * If we race with it leaving CPU, we'll take a lock. So we're correct. + * If we race with it entering CPU, unaccounted time is 0. This is + * indistinguishable from the read occurring a few cycles earlier. + * If we see ->on_cpu without ->on_rq, the task is leaving, and has + * been accounted, so we're correct here as well. + */ + if (!p->on_cpu || !task_on_rq_queued(p)) + return tsk_seruntime(p); +#endif + + rq = task_rq_lock(p, &rf); + ns = p->sched_time + do_task_delta_exec(p, rq); + task_rq_unlock(rq, p, &rf); + + return ns; +} + +/* + * Functions to test for when SCHED_ISO tasks have used their allocated + * quota as real time scheduling and convert them back to SCHED_NORMAL. All + * data is modified only by the local runqueue during scheduler_tick with + * interrupts disabled. + */ + +/* + * Test if SCHED_ISO tasks have run longer than their alloted period as RT + * tasks and set the refractory flag if necessary. There is 10% hysteresis + * for unsetting the flag. 115/128 is ~90/100 as a fast shift instead of a + * slow division. + */ +static inline void iso_tick(struct rq *rq) +{ + rq->iso_ticks = rq->iso_ticks * (ISO_PERIOD - 1) / ISO_PERIOD; + rq->iso_ticks += 100; + if (rq->iso_ticks > ISO_PERIOD * sched_iso_cpu) { + rq->iso_refractory = true; + if (unlikely(rq->iso_ticks > ISO_PERIOD * 100)) + rq->iso_ticks = ISO_PERIOD * 100; + } +} + +/* No SCHED_ISO task was running so decrease rq->iso_ticks */ +static inline void no_iso_tick(struct rq *rq, int ticks) +{ + if (rq->iso_ticks > 0 || rq->iso_refractory) { + rq->iso_ticks = rq->iso_ticks * (ISO_PERIOD - ticks) / ISO_PERIOD; + if (rq->iso_ticks < ISO_PERIOD * (sched_iso_cpu * 115 / 128)) { + rq->iso_refractory = false; + if (unlikely(rq->iso_ticks < 0)) + rq->iso_ticks = 0; + } + } +} + +/* This manages tasks that have run out of timeslice during a scheduler_tick */ +static void task_running_tick(struct rq *rq) +{ + struct task_struct *p = rq->curr; + + /* + * If a SCHED_ISO task is running we increment the iso_ticks. In + * order to prevent SCHED_ISO tasks from causing starvation in the + * presence of true RT tasks we account those as iso_ticks as well. + */ + if (rt_task(p) || task_running_iso(p)) + iso_tick(rq); + else + no_iso_tick(rq, 1); + + /* SCHED_FIFO tasks never run out of timeslice. */ + if (p->policy == SCHED_FIFO) + return; + + if (iso_task(p)) { + if (task_running_iso(p)) { + if (rq->iso_refractory) { + /* + * SCHED_ISO task is running as RT and limit + * has been hit. Force it to reschedule as + * SCHED_NORMAL by zeroing its time_slice + */ + p->time_slice = 0; + } + } else if (!rq->iso_refractory) { + /* Can now run again ISO. Reschedule to pick up prio */ + goto out_resched; + } + } + + /* + * Tasks that were scheduled in the first half of a tick are not + * allowed to run into the 2nd half of the next tick if they will + * run out of time slice in the interim. Otherwise, if they have + * less than RESCHED_US μs of time slice left they will be rescheduled. + * Dither is used as a backup for when hrexpiry is disabled or high res + * timers not configured in. + */ + if (p->time_slice - rq->dither >= RESCHED_US) + return; +out_resched: + rq_lock(rq); + __set_tsk_resched(p); + rq_unlock(rq); +} + +static inline void task_tick(struct rq *rq) +{ + if (!rq_idle(rq)) + task_running_tick(rq); + else if (rq->last_jiffy > rq->last_scheduler_tick) + no_iso_tick(rq, rq->last_jiffy - rq->last_scheduler_tick); +} + +#ifdef CONFIG_NO_HZ_FULL +/* + * We can stop the timer tick any time highres timers are active since + * we rely entirely on highres timeouts for task expiry rescheduling. + */ +static void sched_stop_tick(struct rq *rq, int cpu) +{ + if (!hrexpiry_enabled(rq)) + return; + if (!tick_nohz_full_enabled()) + return; + if (!tick_nohz_full_cpu(cpu)) + return; + tick_nohz_dep_clear_cpu(cpu, TICK_DEP_BIT_SCHED); +} + +static inline void sched_start_tick(struct rq *rq, int cpu) +{ + tick_nohz_dep_set_cpu(cpu, TICK_DEP_BIT_SCHED); +} + +struct tick_work { + int cpu; + atomic_t state; + struct delayed_work work; +}; +/* Values for ->state, see diagram below. */ +#define TICK_SCHED_REMOTE_OFFLINE 0 +#define TICK_SCHED_REMOTE_OFFLINING 1 +#define TICK_SCHED_REMOTE_RUNNING 2 + +/* + * State diagram for ->state: + * + * + * TICK_SCHED_REMOTE_OFFLINE + * | ^ + * | | + * | | sched_tick_remote() + * | | + * | | + * +--TICK_SCHED_REMOTE_OFFLINING + * | ^ + * | | + * sched_tick_start() | | sched_tick_stop() + * | | + * V | + * TICK_SCHED_REMOTE_RUNNING + * + * + * Other transitions get WARN_ON_ONCE(), except that sched_tick_remote() + * and sched_tick_start() are happy to leave the state in RUNNING. + */ + +static struct tick_work __percpu *tick_work_cpu; + +static void sched_tick_remote(struct work_struct *work) +{ + struct delayed_work *dwork = to_delayed_work(work); + struct tick_work *twork = container_of(dwork, struct tick_work, work); + int cpu = twork->cpu; + struct rq *rq = cpu_rq(cpu); + struct task_struct *curr; + u64 delta; + int os; + + /* + * Handle the tick only if it appears the remote CPU is running in full + * dynticks mode. The check is racy by nature, but missing a tick or + * having one too much is no big deal because the scheduler tick updates + * statistics and checks timeslices in a time-independent way, regardless + * of when exactly it is running. + */ + if (!tick_nohz_tick_stopped_cpu(cpu)) + goto out_requeue; + + rq_lock_irq(rq); + if (cpu_is_offline(cpu)) + goto out_unlock; + + curr = rq->curr; + update_rq_clock(rq); + + if (!is_idle_task(curr)) { + /* + * Make sure the next tick runs within a reasonable + * amount of time. + */ + delta = rq_clock_task(rq) - curr->last_ran; + WARN_ON_ONCE(delta > (u64)NSEC_PER_SEC * 3); + } + task_tick(rq); + +out_unlock: + rq_unlock_irq(rq, NULL); + +out_requeue: + + /* + * Run the remote tick once per second (1Hz). This arbitrary + * frequency is large enough to avoid overload but short enough + * to keep scheduler internal stats reasonably up to date. But + * first update state to reflect hotplug activity if required. + */ + os = atomic_fetch_add_unless(&twork->state, -1, TICK_SCHED_REMOTE_RUNNING); + WARN_ON_ONCE(os == TICK_SCHED_REMOTE_OFFLINE); + if (os == TICK_SCHED_REMOTE_RUNNING) + queue_delayed_work(system_unbound_wq, dwork, HZ); +} + +static void sched_tick_start(int cpu) +{ + struct tick_work *twork; + int os; + + if (housekeeping_cpu(cpu, HK_FLAG_TICK)) + return; + + WARN_ON_ONCE(!tick_work_cpu); + + twork = per_cpu_ptr(tick_work_cpu, cpu); + os = atomic_xchg(&twork->state, TICK_SCHED_REMOTE_RUNNING); + WARN_ON_ONCE(os == TICK_SCHED_REMOTE_RUNNING); + if (os == TICK_SCHED_REMOTE_OFFLINE) { + twork->cpu = cpu; + INIT_DELAYED_WORK(&twork->work, sched_tick_remote); + queue_delayed_work(system_unbound_wq, &twork->work, HZ); + } +} + +#ifdef CONFIG_HOTPLUG_CPU +static void sched_tick_stop(int cpu) +{ + struct tick_work *twork; + int os; + + if (housekeeping_cpu(cpu, HK_FLAG_TICK)) + return; + + WARN_ON_ONCE(!tick_work_cpu); + + twork = per_cpu_ptr(tick_work_cpu, cpu); + /* There cannot be competing actions, but don't rely on stop-machine. */ + os = atomic_xchg(&twork->state, TICK_SCHED_REMOTE_OFFLINING); + WARN_ON_ONCE(os != TICK_SCHED_REMOTE_RUNNING); + /* Don't cancel, as this would mess up the state machine. */ +} +#endif /* CONFIG_HOTPLUG_CPU */ + +int __init sched_tick_offload_init(void) +{ + tick_work_cpu = alloc_percpu(struct tick_work); + BUG_ON(!tick_work_cpu); + return 0; +} + +#else /* !CONFIG_NO_HZ_FULL */ +static inline void sched_stop_tick(struct rq *rq, int cpu) {} +static inline void sched_start_tick(struct rq *rq, int cpu) {} +static inline void sched_tick_start(int cpu) { } +static inline void sched_tick_stop(int cpu) { } +#endif + +/* + * This function gets called by the timer code, with HZ frequency. + * We call it with interrupts disabled. + */ +void scheduler_tick(void) +{ + int cpu __maybe_unused = smp_processor_id(); + struct rq *rq = cpu_rq(cpu); + + arch_scale_freq_tick(); + sched_clock_tick(); + update_clocks(rq); + update_load_avg(rq, 0); + update_cpu_clock_tick(rq, rq->curr); + task_tick(rq); + rq->last_scheduler_tick = rq->last_jiffy; + rq->last_tick = rq->clock; + psi_task_tick(rq); + perf_event_task_tick(); + sched_stop_tick(rq, cpu); +} + +#if defined(CONFIG_PREEMPTION) && (defined(CONFIG_DEBUG_PREEMPT) || \ + defined(CONFIG_TRACE_PREEMPT_TOGGLE)) +/* + * If the value passed in is equal to the current preempt count + * then we just disabled preemption. Start timing the latency. + */ +static inline void preempt_latency_start(int val) +{ + if (preempt_count() == val) { + unsigned long ip = get_lock_parent_ip(); +#ifdef CONFIG_DEBUG_PREEMPT + current->preempt_disable_ip = ip; +#endif + trace_preempt_off(CALLER_ADDR0, ip); + } +} + +void preempt_count_add(int val) +{ +#ifdef CONFIG_DEBUG_PREEMPT + /* + * Underflow? + */ + if (DEBUG_LOCKS_WARN_ON((preempt_count() < 0))) + return; +#endif + __preempt_count_add(val); +#ifdef CONFIG_DEBUG_PREEMPT + /* + * Spinlock count overflowing soon? + */ + DEBUG_LOCKS_WARN_ON((preempt_count() & PREEMPT_MASK) >= + PREEMPT_MASK - 10); +#endif + preempt_latency_start(val); +} +EXPORT_SYMBOL(preempt_count_add); +NOKPROBE_SYMBOL(preempt_count_add); + +/* + * If the value passed in equals to the current preempt count + * then we just enabled preemption. Stop timing the latency. + */ +static inline void preempt_latency_stop(int val) +{ + if (preempt_count() == val) + trace_preempt_on(CALLER_ADDR0, get_lock_parent_ip()); +} + +void preempt_count_sub(int val) +{ +#ifdef CONFIG_DEBUG_PREEMPT + /* + * Underflow? + */ + if (DEBUG_LOCKS_WARN_ON(val > preempt_count())) + return; + /* + * Is the spinlock portion underflowing? + */ + if (DEBUG_LOCKS_WARN_ON((val < PREEMPT_MASK) && + !(preempt_count() & PREEMPT_MASK))) + return; +#endif + + preempt_latency_stop(val); + __preempt_count_sub(val); +} +EXPORT_SYMBOL(preempt_count_sub); +NOKPROBE_SYMBOL(preempt_count_sub); + +#else +static inline void preempt_latency_start(int val) { } +static inline void preempt_latency_stop(int val) { } +#endif + +static inline unsigned long get_preempt_disable_ip(struct task_struct *p) +{ +#ifdef CONFIG_DEBUG_PREEMPT + return p->preempt_disable_ip; +#else + return 0; +#endif +} + +/* + * The time_slice is only refilled when it is empty and that is when we set a + * new deadline. Make sure update_clocks has been called recently to update + * rq->niffies. + */ +static void time_slice_expired(struct task_struct *p, struct rq *rq) +{ + p->time_slice = timeslice(); + p->deadline = rq->niffies + task_deadline_diff(p); +#ifdef CONFIG_SMT_NICE + if (!p->mm) + p->smt_bias = 0; + else if (rt_task(p)) + p->smt_bias = 1 << 30; + else if (task_running_iso(p)) + p->smt_bias = 1 << 29; + else if (idleprio_task(p)) { + if (task_running_idle(p)) + p->smt_bias = 0; + else + p->smt_bias = 1; + } else if (--p->smt_bias < 1) + p->smt_bias = MAX_PRIO - p->static_prio; +#endif +} + +/* + * Timeslices below RESCHED_US are considered as good as expired as there's no + * point rescheduling when there's so little time left. SCHED_BATCH tasks + * have been flagged be not latency sensitive and likely to be fully CPU + * bound so every time they're rescheduled they have their time_slice + * refilled, but get a new later deadline to have little effect on + * SCHED_NORMAL tasks. + + */ +static inline void check_deadline(struct task_struct *p, struct rq *rq) +{ + if (p->time_slice < RESCHED_US || batch_task(p)) + time_slice_expired(p, rq); +} + +/* + * Task selection with skiplists is a simple matter of picking off the first + * task in the sorted list, an O(1) operation. The lookup is amortised O(1) + * being bound to the number of processors. + * + * Runqueues are selectively locked based on their unlocked data and then + * unlocked if not needed. At most 3 locks will be held at any time and are + * released as soon as they're no longer needed. All balancing between CPUs + * is thus done here in an extremely simple first come best fit manner. + * + * This iterates over runqueues in cache locality order. In interactive mode + * it iterates over all CPUs and finds the task with the best key/deadline. + * In non-interactive mode it will only take a task if it's from the current + * runqueue or a runqueue with more tasks than the current one with a better + * key/deadline. + */ +#ifdef CONFIG_SMP +static inline struct task_struct +*earliest_deadline_task(struct rq *rq, int cpu, struct task_struct *idle) +{ + struct rq *locked = NULL, *chosen = NULL; + struct task_struct *edt = idle; + int i, best_entries = 0; + u64 best_key = ~0ULL; + + for (i = 0; i < total_runqueues; i++) { + struct rq *other_rq = rq_order(rq, i); + skiplist_node *next; + int entries; + + entries = other_rq->sl->entries; + /* + * Check for queued entres lockless first. The local runqueue + * is locked so entries will always be accurate. + */ + if (!sched_interactive) { + /* + * Don't reschedule balance across nodes unless the CPU + * is idle. + */ + if (edt != idle && rq->cpu_locality[other_rq->cpu] > LOCALITY_SMP) + break; + if (entries <= best_entries) + continue; + } else if (!entries) + continue; + + /* if (i) implies other_rq != rq */ + if (i) { + /* Check for best id queued lockless first */ + if (other_rq->best_key >= best_key) + continue; + + if (unlikely(!trylock_rq(rq, other_rq))) + continue; + + /* Need to reevaluate entries after locking */ + entries = other_rq->sl->entries; + if (unlikely(!entries)) { + unlock_rq(other_rq); + continue; + } + } + + next = other_rq->node; + /* + * In interactive mode we check beyond the best entry on other + * runqueues if we can't get the best for smt or affinity + * reasons. + */ + while ((next = next->next[0]) != other_rq->node) { + struct task_struct *p; + u64 key = next->key; + + /* Reevaluate key after locking */ + if (key >= best_key) + break; + + p = next->value; + if (!smt_schedule(p, rq)) { + if (i && !sched_interactive) + break; + continue; + } + + if (sched_other_cpu(p, cpu)) { + if (sched_interactive || !i) + continue; + break; + } + /* Make sure affinity is ok */ + if (i) { + /* From this point on p is the best so far */ + if (locked) + unlock_rq(locked); + chosen = locked = other_rq; + } + best_entries = entries; + best_key = key; + edt = p; + break; + } + /* rq->preempting is a hint only as the state may have changed + * since it was set with the resched call but if we have met + * the condition we can break out here. */ + if (edt == rq->preempting) + break; + if (i && other_rq != chosen) + unlock_rq(other_rq); + } + + if (likely(edt != idle)) + take_task(rq, cpu, edt); + + if (locked) + unlock_rq(locked); + + rq->preempting = NULL; + + return edt; +} +#else /* CONFIG_SMP */ +static inline struct task_struct +*earliest_deadline_task(struct rq *rq, int cpu, struct task_struct *idle) +{ + struct task_struct *edt; + + if (unlikely(!rq->sl->entries)) + return idle; + edt = rq->node->next[0]->value; + take_task(rq, cpu, edt); + return edt; +} +#endif /* CONFIG_SMP */ + +/* + * Print scheduling while atomic bug: + */ +static noinline void __schedule_bug(struct task_struct *prev) +{ + /* Save this before calling printk(), since that will clobber it */ + unsigned long preempt_disable_ip = get_preempt_disable_ip(current); + + if (oops_in_progress) + return; + + printk(KERN_ERR "BUG: scheduling while atomic: %s/%d/0x%08x\n", + prev->comm, prev->pid, preempt_count()); + + debug_show_held_locks(prev); + print_modules(); + if (irqs_disabled()) + print_irqtrace_events(prev); + if (IS_ENABLED(CONFIG_DEBUG_PREEMPT) + && in_atomic_preempt_off()) { + pr_err("Preemption disabled at:"); + print_ip_sym(preempt_disable_ip); + pr_cont("\n"); + } + dump_stack(); + add_taint(TAINT_WARN, LOCKDEP_STILL_OK); +} + +/* + * Various schedule()-time debugging checks and statistics: + */ +static inline void schedule_debug(struct task_struct *prev, bool preempt) +{ +#ifdef CONFIG_SCHED_STACK_END_CHECK + if (task_stack_end_corrupted(prev)) + panic("corrupted stack end detected inside scheduler\n"); +#endif + +#ifdef CONFIG_DEBUG_ATOMIC_SLEEP + if (!preempt && prev->state && prev->non_block_count) { + printk(KERN_ERR "BUG: scheduling in a non-blocking section: %s/%d/%i\n", + prev->comm, prev->pid, prev->non_block_count); + dump_stack(); + add_taint(TAINT_WARN, LOCKDEP_STILL_OK); + } +#endif + + if (unlikely(in_atomic_preempt_off())) { + __schedule_bug(prev); + preempt_count_set(PREEMPT_DISABLED); + } + rcu_sleep_check(); + + profile_hit(SCHED_PROFILING, __builtin_return_address(0)); + + schedstat_inc(this_rq()->sched_count); +} + +/* + * The currently running task's information is all stored in rq local data + * which is only modified by the local CPU. + */ +static inline void set_rq_task(struct rq *rq, struct task_struct *p) +{ + if (p == rq->idle || p->policy == SCHED_FIFO) + hrexpiry_clear(rq); + else + hrexpiry_start(rq, US_TO_NS(p->time_slice)); + if (rq->clock - rq->last_tick > HALF_JIFFY_NS) + rq->dither = 0; + else + rq->dither = rq_dither(rq); + + rq->rq_deadline = p->deadline; + rq->rq_prio = p->prio; +#ifdef CONFIG_SMT_NICE + rq->rq_mm = p->mm; + rq->rq_smt_bias = p->smt_bias; +#endif +} + +#ifdef CONFIG_SMT_NICE +static void check_no_siblings(struct rq __maybe_unused *this_rq) {} +static void wake_no_siblings(struct rq __maybe_unused *this_rq) {} +static void (*check_siblings)(struct rq *this_rq) = &check_no_siblings; +static void (*wake_siblings)(struct rq *this_rq) = &wake_no_siblings; + +/* Iterate over smt siblings when we've scheduled a process on cpu and decide + * whether they should continue running or be descheduled. */ +static void check_smt_siblings(struct rq *this_rq) +{ + int other_cpu; + + for_each_cpu(other_cpu, &this_rq->thread_mask) { + struct task_struct *p; + struct rq *rq; + + rq = cpu_rq(other_cpu); + if (rq_idle(rq)) + continue; + p = rq->curr; + if (!smt_schedule(p, this_rq)) + resched_curr(rq); + } +} + +static void wake_smt_siblings(struct rq *this_rq) +{ + int other_cpu; + + for_each_cpu(other_cpu, &this_rq->thread_mask) { + struct rq *rq; + + rq = cpu_rq(other_cpu); + if (rq_idle(rq)) + resched_idle(rq); + } +} +#else +static void check_siblings(struct rq __maybe_unused *this_rq) {} +static void wake_siblings(struct rq __maybe_unused *this_rq) {} +#endif + +/* + * schedule() is the main scheduler function. + * + * The main means of driving the scheduler and thus entering this function are: + * + * 1. Explicit blocking: mutex, semaphore, waitqueue, etc. + * + * 2. TIF_NEED_RESCHED flag is checked on interrupt and userspace return + * paths. For example, see arch/x86/entry_64.S. + * + * To drive preemption between tasks, the scheduler sets the flag in timer + * interrupt handler scheduler_tick(). + * + * 3. Wakeups don't really cause entry into schedule(). They add a + * task to the run-queue and that's it. + * + * Now, if the new task added to the run-queue preempts the current + * task, then the wakeup sets TIF_NEED_RESCHED and schedule() gets + * called on the nearest possible occasion: + * + * - If the kernel is preemptible (CONFIG_PREEMPTION=y): + * + * - in syscall or exception context, at the next outmost + * preempt_enable(). (this might be as soon as the wake_up()'s + * spin_unlock()!) + * + * - in IRQ context, return from interrupt-handler to + * preemptible context + * + * - If the kernel is not preemptible (CONFIG_PREEMPTION is not set) + * then at the next: + * + * - cond_resched() call + * - explicit schedule() call + * - return from syscall or exception to user-space + * - return from interrupt-handler to user-space + * + * WARNING: must be called with preemption disabled! + */ +static void __sched notrace __schedule(bool preempt) +{ + struct task_struct *prev, *next, *idle; + unsigned long *switch_count; + bool deactivate = false; + struct rq *rq; + u64 niffies; + int cpu; + + cpu = smp_processor_id(); + rq = cpu_rq(cpu); + prev = rq->curr; + idle = rq->idle; + + schedule_debug(prev, preempt); + + local_irq_disable(); + rcu_note_context_switch(preempt); + + /* + * Make sure that signal_pending_state()->signal_pending() below + * can't be reordered with __set_current_state(TASK_INTERRUPTIBLE) + * done by the caller to avoid the race with signal_wake_up(). + * + * The membarrier system call requires a full memory barrier + * after coming from user-space, before storing to rq->curr. + */ + rq_lock(rq); + smp_mb__after_spinlock(); +#ifdef CONFIG_SMP + if (rq->preempt) { + /* + * Make sure resched_curr hasn't triggered a preemption + * locklessly on a task that has since scheduled away. Spurious + * wakeup of idle is okay though. + */ + if (unlikely(preempt && prev != idle && !test_tsk_need_resched(prev))) { + rq->preempt = NULL; + clear_preempt_need_resched(); + rq_unlock_irq(rq, NULL); + return; + } + rq->preempt = NULL; + } +#endif + + switch_count = &prev->nivcsw; + if (!preempt && prev->state) { + if (signal_pending_state(prev->state, prev)) { + prev->state = TASK_RUNNING; + } else { + deactivate = true; + + if (prev->in_iowait) { + atomic_inc(&rq->nr_iowait); + delayacct_blkio_start(); + } + } + switch_count = &prev->nvcsw; + } + + /* + * Store the niffy value here for use by the next task's last_ran + * below to avoid losing niffies due to update_clocks being called + * again after this point. + */ + update_clocks(rq); + niffies = rq->niffies; + update_cpu_clock_switch(rq, prev); + + clear_tsk_need_resched(prev); + clear_preempt_need_resched(); + + if (idle != prev) { + check_deadline(prev, rq); + return_task(prev, rq, cpu, deactivate); + } + + next = earliest_deadline_task(rq, cpu, idle); + if (likely(next->prio != PRIO_LIMIT)) + clear_cpuidle_map(cpu); + else { + set_cpuidle_map(cpu); + update_load_avg(rq, 0); + } + + set_rq_task(rq, next); + next->last_ran = niffies; + + if (likely(prev != next)) { + /* + * Don't reschedule an idle task or deactivated tasks + */ + if (prev == idle) { + rq->nr_running++; + if (rt_task(next)) + rq->rt_nr_running++; + } else if (!deactivate) + resched_suitable_idle(prev); + if (unlikely(next == idle)) { + rq->nr_running--; + if (rt_task(prev)) + rq->rt_nr_running--; + wake_siblings(rq); + } else + check_siblings(rq); + rq->nr_switches++; + /* + * RCU users of rcu_dereference(rq->curr) may not see + * changes to task_struct made by pick_next_task(). + */ + RCU_INIT_POINTER(rq->curr, next); + /* + * The membarrier system call requires each architecture + * to have a full memory barrier after updating + * rq->curr, before returning to user-space. + * + * Here are the schemes providing that barrier on the + * various architectures: + * - mm ? switch_mm() : mmdrop() for x86, s390, sparc, PowerPC. + * switch_mm() rely on membarrier_arch_switch_mm() on PowerPC. + * - finish_lock_switch() for weakly-ordered + * architectures where spin_unlock is a full barrier, + * - switch_to() for arm64 (weakly-ordered, spin_unlock + * is a RELEASE barrier), + */ + ++*switch_count; + + psi_sched_switch(prev, next, !task_on_rq_queued(prev)); + + trace_sched_switch(preempt, prev, next); + context_switch(rq, prev, next); /* unlocks the rq */ + } else { + check_siblings(rq); + rq_unlock(rq); + local_irq_enable(); + } +} + +void __noreturn do_task_dead(void) +{ + /* Causes final put_task_struct in finish_task_switch(). */ + set_special_state(TASK_DEAD); + + /* Tell freezer to ignore us: */ + current->flags |= PF_NOFREEZE; + __schedule(false); + BUG(); + + /* Avoid "noreturn function does return" - but don't continue if BUG() is a NOP: */ + for (;;) + cpu_relax(); +} + +static inline void sched_submit_work(struct task_struct *tsk) +{ + if (!tsk->state) + return; + + /* + * If a worker went to sleep, notify and ask workqueue whether + * it wants to wake up a task to maintain concurrency. + * As this function is called inside the schedule() context, + * we disable preemption to avoid it calling schedule() again + * in the possible wakeup of a kworker and because wq_worker_sleeping() + * requires it. + */ + if (tsk->flags & (PF_WQ_WORKER | PF_IO_WORKER)) { + preempt_disable(); + if (tsk->flags & PF_WQ_WORKER) + wq_worker_sleeping(tsk); + else + io_wq_worker_sleeping(tsk); + preempt_enable_no_resched(); + } + + if (tsk_is_pi_blocked(tsk)) + return; + + /* + * If we are going to sleep and we have plugged IO queued, + * make sure to submit it to avoid deadlocks. + */ + if (blk_needs_flush_plug(tsk)) + blk_schedule_flush_plug(tsk); +} + +static inline void sched_update_worker(struct task_struct *tsk) +{ + if (tsk->flags & (PF_WQ_WORKER | PF_IO_WORKER)) { + if (tsk->flags & PF_WQ_WORKER) + wq_worker_running(tsk); + else + io_wq_worker_running(tsk); + } +} + +asmlinkage __visible void __sched schedule(void) +{ + struct task_struct *tsk = current; + + sched_submit_work(tsk); + do { + preempt_disable(); + __schedule(false); + sched_preempt_enable_no_resched(); + } while (need_resched()); + sched_update_worker(tsk); +} + +EXPORT_SYMBOL(schedule); + +/* + * synchronize_rcu_tasks() makes sure that no task is stuck in preempted + * state (have scheduled out non-voluntarily) by making sure that all + * tasks have either left the run queue or have gone into user space. + * As idle tasks do not do either, they must not ever be preempted + * (schedule out non-voluntarily). + * + * schedule_idle() is similar to schedule_preempt_disable() except that it + * never enables preemption because it does not call sched_submit_work(). + */ +void __sched schedule_idle(void) +{ + /* + * As this skips calling sched_submit_work(), which the idle task does + * regardless because that function is a nop when the task is in a + * TASK_RUNNING state, make sure this isn't used someplace that the + * current task can be in any other state. Note, idle is always in the + * TASK_RUNNING state. + */ + WARN_ON_ONCE(current->state); + do { + __schedule(false); + } while (need_resched()); +} + +#ifdef CONFIG_CONTEXT_TRACKING +asmlinkage __visible void __sched schedule_user(void) +{ + /* + * If we come here after a random call to set_need_resched(), + * or we have been woken up remotely but the IPI has not yet arrived, + * we haven't yet exited the RCU idle mode. Do it here manually until + * we find a better solution. + * + * NB: There are buggy callers of this function. Ideally we + * should warn if prev_state != IN_USER, but that will trigger + * too frequently to make sense yet. + */ + enum ctx_state prev_state = exception_enter(); + schedule(); + exception_exit(prev_state); +} +#endif + +/** + * schedule_preempt_disabled - called with preemption disabled + * + * Returns with preemption disabled. Note: preempt_count must be 1 + */ +void __sched schedule_preempt_disabled(void) +{ + sched_preempt_enable_no_resched(); + schedule(); + preempt_disable(); +} + +static void __sched notrace preempt_schedule_common(void) +{ + do { + /* + * Because the function tracer can trace preempt_count_sub() + * and it also uses preempt_enable/disable_notrace(), if + * NEED_RESCHED is set, the preempt_enable_notrace() called + * by the function tracer will call this function again and + * cause infinite recursion. + * + * Preemption must be disabled here before the function + * tracer can trace. Break up preempt_disable() into two + * calls. One to disable preemption without fear of being + * traced. The other to still record the preemption latency, + * which can also be traced by the function tracer. + */ + preempt_disable_notrace(); + preempt_latency_start(1); + __schedule(true); + preempt_latency_stop(1); + preempt_enable_no_resched_notrace(); + + /* + * Check again in case we missed a preemption opportunity + * between schedule and now. + */ + } while (need_resched()); +} + +#ifdef CONFIG_PREEMPTION +/* + * This is the entry point to schedule() from in-kernel preemption + * off of preempt_enable. + */ +asmlinkage __visible void __sched notrace preempt_schedule(void) +{ + /* + * If there is a non-zero preempt_count or interrupts are disabled, + * we do not want to preempt the current task. Just return.. + */ + if (likely(!preemptible())) + return; + + preempt_schedule_common(); +} +NOKPROBE_SYMBOL(preempt_schedule); +EXPORT_SYMBOL(preempt_schedule); + +/** + * preempt_schedule_notrace - preempt_schedule called by tracing + * + * The tracing infrastructure uses preempt_enable_notrace to prevent + * recursion and tracing preempt enabling caused by the tracing + * infrastructure itself. But as tracing can happen in areas coming + * from userspace or just about to enter userspace, a preempt enable + * can occur before user_exit() is called. This will cause the scheduler + * to be called when the system is still in usermode. + * + * To prevent this, the preempt_enable_notrace will use this function + * instead of preempt_schedule() to exit user context if needed before + * calling the scheduler. + */ +asmlinkage __visible void __sched notrace preempt_schedule_notrace(void) +{ + enum ctx_state prev_ctx; + + if (likely(!preemptible())) + return; + + do { + /* + * Because the function tracer can trace preempt_count_sub() + * and it also uses preempt_enable/disable_notrace(), if + * NEED_RESCHED is set, the preempt_enable_notrace() called + * by the function tracer will call this function again and + * cause infinite recursion. + * + * Preemption must be disabled here before the function + * tracer can trace. Break up preempt_disable() into two + * calls. One to disable preemption without fear of being + * traced. The other to still record the preemption latency, + * which can also be traced by the function tracer. + */ + preempt_disable_notrace(); + preempt_latency_start(1); + /* + * Needs preempt disabled in case user_exit() is traced + * and the tracer calls preempt_enable_notrace() causing + * an infinite recursion. + */ + prev_ctx = exception_enter(); + __schedule(true); + exception_exit(prev_ctx); + + preempt_latency_stop(1); + preempt_enable_no_resched_notrace(); + } while (need_resched()); +} +EXPORT_SYMBOL_GPL(preempt_schedule_notrace); + +#endif /* CONFIG_PREEMPTION */ + +/* + * This is the entry point to schedule() from kernel preemption + * off of irq context. + * Note, that this is called and return with irqs disabled. This will + * protect us against recursive calling from irq. + */ +asmlinkage __visible void __sched preempt_schedule_irq(void) +{ + enum ctx_state prev_state; + + /* Catch callers which need to be fixed */ + BUG_ON(preempt_count() || !irqs_disabled()); + + prev_state = exception_enter(); + + do { + preempt_disable(); + local_irq_enable(); + __schedule(true); + local_irq_disable(); + sched_preempt_enable_no_resched(); + } while (need_resched()); + + exception_exit(prev_state); +} + +int default_wake_function(wait_queue_entry_t *curr, unsigned mode, int wake_flags, + void *key) +{ + return try_to_wake_up(curr->private, mode, wake_flags); +} +EXPORT_SYMBOL(default_wake_function); + +#ifdef CONFIG_RT_MUTEXES + +static inline int __rt_effective_prio(struct task_struct *pi_task, int prio) +{ + if (pi_task) + prio = min(prio, pi_task->prio); + + return prio; +} + +static inline int rt_effective_prio(struct task_struct *p, int prio) +{ + struct task_struct *pi_task = rt_mutex_get_top_task(p); + + return __rt_effective_prio(pi_task, prio); +} + +/* + * rt_mutex_setprio - set the current priority of a task + * @p: task to boost + * @pi_task: donor task + * + * This function changes the 'effective' priority of a task. It does + * not touch ->normal_prio like __setscheduler(). + * + * Used by the rt_mutex code to implement priority inheritance + * logic. Call site only calls if the priority of the task changed. + */ +void rt_mutex_setprio(struct task_struct *p, struct task_struct *pi_task) +{ + int prio, oldprio; + struct rq *rq; + + /* XXX used to be waiter->prio, not waiter->task->prio */ + prio = __rt_effective_prio(pi_task, p->normal_prio); + + /* + * If nothing changed; bail early. + */ + if (p->pi_top_task == pi_task && prio == p->prio) + return; + + rq = __task_rq_lock(p, NULL); + update_rq_clock(rq); + /* + * Set under pi_lock && rq->lock, such that the value can be used under + * either lock. + * + * Note that there is loads of tricky to make this pointer cache work + * right. rt_mutex_slowunlock()+rt_mutex_postunlock() work together to + * ensure a task is de-boosted (pi_task is set to NULL) before the + * task is allowed to run again (and can exit). This ensures the pointer + * points to a blocked task -- which guaratees the task is present. + */ + p->pi_top_task = pi_task; + + /* + * For FIFO/RR we only need to set prio, if that matches we're done. + */ + if (prio == p->prio) + goto out_unlock; + + /* + * Idle task boosting is a nono in general. There is one + * exception, when PREEMPT_RT and NOHZ is active: + * + * The idle task calls get_next_timer_interrupt() and holds + * the timer wheel base->lock on the CPU and another CPU wants + * to access the timer (probably to cancel it). We can safely + * ignore the boosting request, as the idle CPU runs this code + * with interrupts disabled and will complete the lock + * protected section without being interrupted. So there is no + * real need to boost. + */ + if (unlikely(p == rq->idle)) { + WARN_ON(p != rq->curr); + WARN_ON(p->pi_blocked_on); + goto out_unlock; + } + + trace_sched_pi_setprio(p, pi_task); + oldprio = p->prio; + p->prio = prio; + if (task_running(rq, p)){ + if (prio > oldprio) + resched_task(p); + } else if (task_queued(p)) { + dequeue_task(rq, p, DEQUEUE_SAVE); + enqueue_task(rq, p, ENQUEUE_RESTORE); + if (prio < oldprio) + try_preempt(p, rq); + } +out_unlock: + __task_rq_unlock(rq, NULL); +} +#else +static inline int rt_effective_prio(struct task_struct *p, int prio) +{ + return prio; +} +#endif + +/* + * Adjust the deadline for when the priority is to change, before it's + * changed. + */ +static inline void adjust_deadline(struct task_struct *p, int new_prio) +{ + p->deadline += static_deadline_diff(new_prio) - task_deadline_diff(p); +} + +void set_user_nice(struct task_struct *p, long nice) +{ + int new_static, old_static; + struct rq_flags rf; + struct rq *rq; + + if (task_nice(p) == nice || nice < MIN_NICE || nice > MAX_NICE) + return; + new_static = NICE_TO_PRIO(nice); + /* + * We have to be careful, if called from sys_setpriority(), + * the task might be in the middle of scheduling on another CPU. + */ + rq = task_rq_lock(p, &rf); + update_rq_clock(rq); + + /* + * The RT priorities are set via sched_setscheduler(), but we still + * allow the 'normal' nice value to be set - but as expected + * it wont have any effect on scheduling until the task is + * not SCHED_NORMAL/SCHED_BATCH: + */ + if (has_rt_policy(p)) { + p->static_prio = new_static; + goto out_unlock; + } + + adjust_deadline(p, new_static); + old_static = p->static_prio; + p->static_prio = new_static; + p->prio = effective_prio(p); + + if (task_queued(p)) { + dequeue_task(rq, p, DEQUEUE_SAVE); + enqueue_task(rq, p, ENQUEUE_RESTORE); + if (new_static < old_static) + try_preempt(p, rq); + } else if (task_running(rq, p)) { + set_rq_task(rq, p); + if (old_static < new_static) + resched_task(p); + } +out_unlock: + task_rq_unlock(rq, p, &rf); +} +EXPORT_SYMBOL(set_user_nice); + +/* + * can_nice - check if a task can reduce its nice value + * @p: task + * @nice: nice value + */ +int can_nice(const struct task_struct *p, const int nice) +{ + /* Convert nice value [19,-20] to rlimit style value [1,40] */ + int nice_rlim = nice_to_rlimit(nice); + + return (nice_rlim <= task_rlimit(p, RLIMIT_NICE) || + capable(CAP_SYS_NICE)); +} + +#ifdef __ARCH_WANT_SYS_NICE + +/* + * sys_nice - change the priority of the current process. + * @increment: priority increment + * + * sys_setpriority is a more generic, but much slower function that + * does similar things. + */ +SYSCALL_DEFINE1(nice, int, increment) +{ + long nice, retval; + + /* + * Setpriority might change our priority at the same moment. + * We don't have to worry. Conceptually one call occurs first + * and we have a single winner. + */ + + increment = clamp(increment, -NICE_WIDTH, NICE_WIDTH); + nice = task_nice(current) + increment; + + nice = clamp_val(nice, MIN_NICE, MAX_NICE); + if (increment < 0 && !can_nice(current, nice)) + return -EPERM; + + retval = security_task_setnice(current, nice); + if (retval) + return retval; + + set_user_nice(current, nice); + return 0; +} + +#endif + +/** + * task_prio - return the priority value of a given task. + * @p: the task in question. + * + * Return: The priority value as seen by users in /proc. + * RT tasks are offset by -100. Normal tasks are centered around 1, value goes + * from 0 (SCHED_ISO) up to 82 (nice +19 SCHED_IDLEPRIO). + */ +int task_prio(const struct task_struct *p) +{ + int delta, prio = p->prio - MAX_RT_PRIO; + + /* rt tasks and iso tasks */ + if (prio <= 0) + goto out; + + /* Convert to ms to avoid overflows */ + delta = NS_TO_MS(p->deadline - task_rq(p)->niffies); + if (unlikely(delta < 0)) + delta = 0; + delta = delta * 40 / ms_longest_deadline_diff(); + if (delta <= 80) + prio += delta; + if (idleprio_task(p)) + prio += 40; +out: + return prio; +} + +/** + * idle_cpu - is a given CPU idle currently? + * @cpu: the processor in question. + * + * Return: 1 if the CPU is currently idle. 0 otherwise. + */ +int idle_cpu(int cpu) +{ + return cpu_curr(cpu) == cpu_rq(cpu)->idle; +} + +/** + * available_idle_cpu - is a given CPU idle for enqueuing work. + * @cpu: the CPU in question. + * + * Return: 1 if the CPU is currently idle. 0 otherwise. + */ +int available_idle_cpu(int cpu) +{ + if (!idle_cpu(cpu)) + return 0; + + if (vcpu_is_preempted(cpu)) + return 0; + + return 1; +} + +/** + * idle_task - return the idle task for a given CPU. + * @cpu: the processor in question. + * + * Return: The idle task for the CPU @cpu. + */ +struct task_struct *idle_task(int cpu) +{ + return cpu_rq(cpu)->idle; +} + +/** + * find_process_by_pid - find a process with a matching PID value. + * @pid: the pid in question. + * + * The task of @pid, if found. %NULL otherwise. + */ +static inline struct task_struct *find_process_by_pid(pid_t pid) +{ + return pid ? find_task_by_vpid(pid) : current; +} + +/* Actually do priority change: must hold rq lock. */ +static void __setscheduler(struct task_struct *p, struct rq *rq, int policy, + int prio, const struct sched_attr *attr, + bool keep_boost) +{ + int oldrtprio, oldprio; + + /* + * If params can't change scheduling class changes aren't allowed + * either. + */ + if (attr->sched_flags & SCHED_FLAG_KEEP_PARAMS) + return; + + p->policy = policy; + oldrtprio = p->rt_priority; + p->rt_priority = prio; + p->normal_prio = normal_prio(p); + oldprio = p->prio; + /* + * Keep a potential priority boosting if called from + * sched_setscheduler(). + */ + p->prio = normal_prio(p); + if (keep_boost) + p->prio = rt_effective_prio(p, p->prio); + + if (task_running(rq, p)) { + set_rq_task(rq, p); + resched_task(p); + } else if (task_queued(p)) { + dequeue_task(rq, p, DEQUEUE_SAVE); + enqueue_task(rq, p, ENQUEUE_RESTORE); + if (p->prio < oldprio || p->rt_priority > oldrtprio) + try_preempt(p, rq); + } +} + +/* + * Check the target process has a UID that matches the current process's + */ +static bool check_same_owner(struct task_struct *p) +{ + const struct cred *cred = current_cred(), *pcred; + bool match; + + rcu_read_lock(); + pcred = __task_cred(p); + match = (uid_eq(cred->euid, pcred->euid) || + uid_eq(cred->euid, pcred->uid)); + rcu_read_unlock(); + return match; +} + +static int __sched_setscheduler(struct task_struct *p, + const struct sched_attr *attr, + bool user, bool pi) +{ + int retval, policy = attr->sched_policy, oldpolicy = -1, priority = attr->sched_priority; + unsigned long rlim_rtprio = 0; + struct rq_flags rf; + int reset_on_fork; + struct rq *rq; + + /* The pi code expects interrupts enabled */ + BUG_ON(pi && in_interrupt()); + + if (is_rt_policy(policy) && !capable(CAP_SYS_NICE)) { + unsigned long lflags; + + if (!lock_task_sighand(p, &lflags)) + return -ESRCH; + rlim_rtprio = task_rlimit(p, RLIMIT_RTPRIO); + unlock_task_sighand(p, &lflags); + if (rlim_rtprio) + goto recheck; + /* + * If the caller requested an RT policy without having the + * necessary rights, we downgrade the policy to SCHED_ISO. + * We also set the parameter to zero to pass the checks. + */ + policy = SCHED_ISO; + priority = 0; + } +recheck: + /* Double check policy once rq lock held */ + if (policy < 0) { + reset_on_fork = p->sched_reset_on_fork; + policy = oldpolicy = p->policy; + } else { + reset_on_fork = !!(policy & SCHED_RESET_ON_FORK); + policy &= ~SCHED_RESET_ON_FORK; + + if (!SCHED_RANGE(policy)) + return -EINVAL; + } + + if (attr->sched_flags & ~(SCHED_FLAG_ALL | SCHED_FLAG_SUGOV)) + return -EINVAL; + + /* + * Valid priorities for SCHED_FIFO and SCHED_RR are + * 1..MAX_USER_RT_PRIO-1, valid priority for SCHED_NORMAL and + * SCHED_BATCH is 0. + */ + if (priority < 0 || + (p->mm && priority > MAX_USER_RT_PRIO - 1) || + (!p->mm && priority > MAX_RT_PRIO - 1)) + return -EINVAL; + if (is_rt_policy(policy) != (priority != 0)) + return -EINVAL; + + /* + * Allow unprivileged RT tasks to decrease priority: + */ + if (user && !capable(CAP_SYS_NICE)) { + if (is_rt_policy(policy)) { + unsigned long rlim_rtprio = + task_rlimit(p, RLIMIT_RTPRIO); + + /* Can't set/change the rt policy */ + if (policy != p->policy && !rlim_rtprio) + return -EPERM; + + /* Can't increase priority */ + if (priority > p->rt_priority && + priority > rlim_rtprio) + return -EPERM; + } else { + switch (p->policy) { + /* + * Can only downgrade policies but not back to + * SCHED_NORMAL + */ + case SCHED_ISO: + if (policy == SCHED_ISO) + goto out; + if (policy != SCHED_NORMAL) + return -EPERM; + break; + case SCHED_BATCH: + if (policy == SCHED_BATCH) + goto out; + if (policy != SCHED_IDLEPRIO) + return -EPERM; + break; + case SCHED_IDLEPRIO: + if (policy == SCHED_IDLEPRIO) + goto out; + return -EPERM; + default: + break; + } + } + + /* Can't change other user's priorities */ + if (!check_same_owner(p)) + return -EPERM; + + /* Normal users shall not reset the sched_reset_on_fork flag: */ + if (p->sched_reset_on_fork && !reset_on_fork) + return -EPERM; + } + + if (user) { + retval = security_task_setscheduler(p); + if (retval) + return retval; + } + + if (pi) + cpuset_read_lock(); + + /* + * Make sure no PI-waiters arrive (or leave) while we are + * changing the priority of the task: + * + * To be able to change p->policy safely, the runqueue lock must be + * held. + */ + rq = task_rq_lock(p, &rf); + update_rq_clock(rq); + + /* + * Changing the policy of the stop threads its a very bad idea: + */ + if (p == rq->stop) { + retval = -EINVAL; + goto unlock; + } + + /* + * If not changing anything there's no need to proceed further: + */ + if (unlikely(policy == p->policy && (!is_rt_policy(policy) || + priority == p->rt_priority))) { + retval = 0; + goto unlock; + } + + /* Re-check policy now with rq lock held */ + if (unlikely(oldpolicy != -1 && oldpolicy != p->policy)) { + policy = oldpolicy = -1; + task_rq_unlock(rq, p, &rf); + if (pi) + cpuset_read_unlock(); + goto recheck; + } + p->sched_reset_on_fork = reset_on_fork; + + __setscheduler(p, rq, policy, priority, attr, pi); + + /* Avoid rq from going away on us: */ + preempt_disable(); + task_rq_unlock(rq, p, &rf); + + if (pi) { + cpuset_read_unlock(); + rt_mutex_adjust_pi(p); + } + preempt_enable(); +out: + return 0; + +unlock: + task_rq_unlock(rq, p, &rf); + if (pi) + cpuset_read_unlock(); + return retval; +} + +static int _sched_setscheduler(struct task_struct *p, int policy, + const struct sched_param *param, bool check) +{ + struct sched_attr attr = { + .sched_policy = policy, + .sched_priority = param->sched_priority, + .sched_nice = PRIO_TO_NICE(p->static_prio), + }; + + return __sched_setscheduler(p, &attr, check, true); +} +/** + * sched_setscheduler - change the scheduling policy and/or RT priority of a thread. + * @p: the task in question. + * @policy: new policy. + * @param: structure containing the new RT priority. + * + * Return: 0 on success. An error code otherwise. + * + * NOTE that the task may be already dead. + */ +int sched_setscheduler(struct task_struct *p, int policy, + const struct sched_param *param) +{ + return _sched_setscheduler(p, policy, param, true); +} + +EXPORT_SYMBOL_GPL(sched_setscheduler); + +int sched_setattr(struct task_struct *p, const struct sched_attr *attr) +{ + return __sched_setscheduler(p, attr, true, true); +} +EXPORT_SYMBOL_GPL(sched_setattr); + +int sched_setattr_nocheck(struct task_struct *p, const struct sched_attr *attr) +{ + return __sched_setscheduler(p, attr, false, true); +} + +/** + * sched_setscheduler_nocheck - change the scheduling policy and/or RT priority of a thread from kernelspace. + * @p: the task in question. + * @policy: new policy. + * @param: structure containing the new RT priority. + * + * Just like sched_setscheduler, only don't bother checking if the + * current context has permission. For example, this is needed in + * stop_machine(): we create temporary high priority worker threads, + * but our caller might not have that capability. + * + * Return: 0 on success. An error code otherwise. + */ +int sched_setscheduler_nocheck(struct task_struct *p, int policy, + const struct sched_param *param) +{ + return _sched_setscheduler(p, policy, param, false); +} +EXPORT_SYMBOL_GPL(sched_setscheduler_nocheck); + +static int +do_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param) +{ + struct sched_param lparam; + struct task_struct *p; + int retval; + + if (!param || pid < 0) + return -EINVAL; + if (copy_from_user(&lparam, param, sizeof(struct sched_param))) + return -EFAULT; + + rcu_read_lock(); + retval = -ESRCH; + p = find_process_by_pid(pid); + if (likely(p)) + get_task_struct(p); + rcu_read_unlock(); + + if (likely(p)) { + retval = sched_setscheduler(p, policy, &lparam); + put_task_struct(p); + } + + return retval; +} + +/* + * Mimics kernel/events/core.c perf_copy_attr(). + */ +static int sched_copy_attr(struct sched_attr __user *uattr, + struct sched_attr *attr) +{ + u32 size; + int ret; + + /* Zero the full structure, so that a short copy will be nice: */ + memset(attr, 0, sizeof(*attr)); + + ret = get_user(size, &uattr->size); + if (ret) + return ret; + + /* ABI compatibility quirk: */ + if (!size) + size = SCHED_ATTR_SIZE_VER0; + + if (size < SCHED_ATTR_SIZE_VER0 || size > PAGE_SIZE) + goto err_size; + + ret = copy_struct_from_user(attr, sizeof(*attr), uattr, size); + if (ret) { + if (ret == -E2BIG) + goto err_size; + return ret; + } + + /* + * XXX: Do we want to be lenient like existing syscalls; or do we want + * to be strict and return an error on out-of-bounds values? + */ + attr->sched_nice = clamp(attr->sched_nice, -20, 19); + + /* sched/core.c uses zero here but we already know ret is zero */ + return 0; + +err_size: + put_user(sizeof(*attr), &uattr->size); + return -E2BIG; +} + +/* + * sched_setparam() passes in -1 for its policy, to let the functions + * it calls know not to change it. + */ +#define SETPARAM_POLICY -1 + +/** + * sys_sched_setscheduler - set/change the scheduler policy and RT priority + * @pid: the pid in question. + * @policy: new policy. + * @param: structure containing the new RT priority. + * + * Return: 0 on success. An error code otherwise. + */ +SYSCALL_DEFINE3(sched_setscheduler, pid_t, pid, int, policy, struct sched_param __user *, param) +{ + if (policy < 0) + return -EINVAL; + + return do_sched_setscheduler(pid, policy, param); +} + +/** + * sys_sched_setparam - set/change the RT priority of a thread + * @pid: the pid in question. + * @param: structure containing the new RT priority. + * + * Return: 0 on success. An error code otherwise. + */ +SYSCALL_DEFINE2(sched_setparam, pid_t, pid, struct sched_param __user *, param) +{ + return do_sched_setscheduler(pid, SETPARAM_POLICY, param); +} + +/** + * sys_sched_setattr - same as above, but with extended sched_attr + * @pid: the pid in question. + * @uattr: structure containing the extended parameters. + */ +SYSCALL_DEFINE3(sched_setattr, pid_t, pid, struct sched_attr __user *, uattr, + unsigned int, flags) +{ + struct sched_attr attr; + struct task_struct *p; + int retval; + + if (!uattr || pid < 0 || flags) + return -EINVAL; + + retval = sched_copy_attr(uattr, &attr); + if (retval) + return retval; + + if ((int)attr.sched_policy < 0) + return -EINVAL; + if (attr.sched_flags & SCHED_FLAG_KEEP_POLICY) + attr.sched_policy = SETPARAM_POLICY; + + rcu_read_lock(); + retval = -ESRCH; + p = find_process_by_pid(pid); + if (likely(p)) + get_task_struct(p); + rcu_read_unlock(); + + if (likely(p)) { + retval = sched_setattr(p, &attr); + put_task_struct(p); + } + + return retval; +} + +/** + * sys_sched_getscheduler - get the policy (scheduling class) of a thread + * @pid: the pid in question. + * + * Return: On success, the policy of the thread. Otherwise, a negative error + * code. + */ +SYSCALL_DEFINE1(sched_getscheduler, pid_t, pid) +{ + struct task_struct *p; + int retval = -EINVAL; + + if (pid < 0) + goto out_nounlock; + + retval = -ESRCH; + rcu_read_lock(); + p = find_process_by_pid(pid); + if (p) { + retval = security_task_getscheduler(p); + if (!retval) + retval = p->policy; + } + rcu_read_unlock(); + +out_nounlock: + return retval; +} + +/** + * sys_sched_getscheduler - get the RT priority of a thread + * @pid: the pid in question. + * @param: structure containing the RT priority. + * + * Return: On success, 0 and the RT priority is in @param. Otherwise, an error + * code. + */ +SYSCALL_DEFINE2(sched_getparam, pid_t, pid, struct sched_param __user *, param) +{ + struct sched_param lp = { .sched_priority = 0 }; + struct task_struct *p; + int retval = -EINVAL; + + if (!param || pid < 0) + goto out_nounlock; + + rcu_read_lock(); + p = find_process_by_pid(pid); + retval = -ESRCH; + if (!p) + goto out_unlock; + + retval = security_task_getscheduler(p); + if (retval) + goto out_unlock; + + if (has_rt_policy(p)) + lp.sched_priority = p->rt_priority; + rcu_read_unlock(); + + /* + * This one might sleep, we cannot do it with a spinlock held ... + */ + retval = copy_to_user(param, &lp, sizeof(*param)) ? -EFAULT : 0; + +out_nounlock: + return retval; + +out_unlock: + rcu_read_unlock(); + return retval; +} + +/* + * Copy the kernel size attribute structure (which might be larger + * than what user-space knows about) to user-space. + * + * Note that all cases are valid: user-space buffer can be larger or + * smaller than the kernel-space buffer. The usual case is that both + * have the same size. + */ +static int +sched_attr_copy_to_user(struct sched_attr __user *uattr, + struct sched_attr *kattr, + unsigned int usize) +{ + unsigned int ksize = sizeof(*kattr); + + if (!access_ok(uattr, usize)) + return -EFAULT; + + /* + * sched_getattr() ABI forwards and backwards compatibility: + * + * If usize == ksize then we just copy everything to user-space and all is good. + * + * If usize < ksize then we only copy as much as user-space has space for, + * this keeps ABI compatibility as well. We skip the rest. + * + * If usize > ksize then user-space is using a newer version of the ABI, + * which part the kernel doesn't know about. Just ignore it - tooling can + * detect the kernel's knowledge of attributes from the attr->size value + * which is set to ksize in this case. + */ + kattr->size = min(usize, ksize); + + if (copy_to_user(uattr, kattr, kattr->size)) + return -EFAULT; + + return 0; +} + +/** + * sys_sched_getattr - similar to sched_getparam, but with sched_attr + * @pid: the pid in question. + * @uattr: structure containing the extended parameters. + * @usize: sizeof(attr) for fwd/bwd comp. + * @flags: for future extension. + */ +SYSCALL_DEFINE4(sched_getattr, pid_t, pid, struct sched_attr __user *, uattr, + unsigned int, usize, unsigned int, flags) +{ + struct sched_attr kattr = { }; + struct task_struct *p; + int retval; + + if (!uattr || pid < 0 || usize > PAGE_SIZE || + usize < SCHED_ATTR_SIZE_VER0 || flags) + return -EINVAL; + + rcu_read_lock(); + p = find_process_by_pid(pid); + retval = -ESRCH; + if (!p) + goto out_unlock; + + retval = security_task_getscheduler(p); + if (retval) + goto out_unlock; + + kattr.sched_policy = p->policy; + if (rt_task(p)) + kattr.sched_priority = p->rt_priority; + else + kattr.sched_nice = task_nice(p); + + rcu_read_unlock(); + + return sched_attr_copy_to_user(uattr, &kattr, usize); + +out_unlock: + rcu_read_unlock(); + return retval; +} + +long sched_setaffinity(pid_t pid, const struct cpumask *in_mask) +{ + cpumask_var_t cpus_allowed, new_mask; + struct task_struct *p; + int retval; + + rcu_read_lock(); + + p = find_process_by_pid(pid); + if (!p) { + rcu_read_unlock(); + return -ESRCH; + } + + /* Prevent p going away */ + get_task_struct(p); + rcu_read_unlock(); + + if (p->flags & PF_NO_SETAFFINITY) { + retval = -EINVAL; + goto out_put_task; + } + if (!alloc_cpumask_var(&cpus_allowed, GFP_KERNEL)) { + retval = -ENOMEM; + goto out_put_task; + } + if (!alloc_cpumask_var(&new_mask, GFP_KERNEL)) { + retval = -ENOMEM; + goto out_free_cpus_allowed; + } + retval = -EPERM; + if (!check_same_owner(p)) { + rcu_read_lock(); + if (!ns_capable(__task_cred(p)->user_ns, CAP_SYS_NICE)) { + rcu_read_unlock(); + goto out_unlock; + } + rcu_read_unlock(); + } + + retval = security_task_setscheduler(p); + if (retval) + goto out_unlock; + + cpuset_cpus_allowed(p, cpus_allowed); + cpumask_and(new_mask, in_mask, cpus_allowed); +again: + retval = __set_cpus_allowed_ptr(p, new_mask, true); + + if (!retval) { + cpuset_cpus_allowed(p, cpus_allowed); + if (!cpumask_subset(new_mask, cpus_allowed)) { + /* + * We must have raced with a concurrent cpuset + * update. Just reset the cpus_allowed to the + * cpuset's cpus_allowed + */ + cpumask_copy(new_mask, cpus_allowed); + goto again; + } + } +out_unlock: + free_cpumask_var(new_mask); +out_free_cpus_allowed: + free_cpumask_var(cpus_allowed); +out_put_task: + put_task_struct(p); + return retval; +} + +static int get_user_cpu_mask(unsigned long __user *user_mask_ptr, unsigned len, + cpumask_t *new_mask) +{ + if (len < cpumask_size()) + cpumask_clear(new_mask); + else if (len > cpumask_size()) + len = cpumask_size(); + + return copy_from_user(new_mask, user_mask_ptr, len) ? -EFAULT : 0; +} + + +/** + * sys_sched_setaffinity - set the CPU affinity of a process + * @pid: pid of the process + * @len: length in bytes of the bitmask pointed to by user_mask_ptr + * @user_mask_ptr: user-space pointer to the new CPU mask + * + * Return: 0 on success. An error code otherwise. + */ +SYSCALL_DEFINE3(sched_setaffinity, pid_t, pid, unsigned int, len, + unsigned long __user *, user_mask_ptr) +{ + cpumask_var_t new_mask; + int retval; + + if (!alloc_cpumask_var(&new_mask, GFP_KERNEL)) + return -ENOMEM; + + retval = get_user_cpu_mask(user_mask_ptr, len, new_mask); + if (retval == 0) + retval = sched_setaffinity(pid, new_mask); + free_cpumask_var(new_mask); + return retval; +} + +long sched_getaffinity(pid_t pid, cpumask_t *mask) +{ + struct task_struct *p; + unsigned long flags; + int retval; + + get_online_cpus(); + rcu_read_lock(); + + retval = -ESRCH; + p = find_process_by_pid(pid); + if (!p) + goto out_unlock; + + retval = security_task_getscheduler(p); + if (retval) + goto out_unlock; + + raw_spin_lock_irqsave(&p->pi_lock, flags); + cpumask_and(mask, &p->cpus_mask, cpu_active_mask); + raw_spin_unlock_irqrestore(&p->pi_lock, flags); + +out_unlock: + rcu_read_unlock(); + put_online_cpus(); + + return retval; +} + +/** + * sys_sched_getaffinity - get the CPU affinity of a process + * @pid: pid of the process + * @len: length in bytes of the bitmask pointed to by user_mask_ptr + * @user_mask_ptr: user-space pointer to hold the current CPU mask + * + * Return: 0 on success. An error code otherwise. + */ +SYSCALL_DEFINE3(sched_getaffinity, pid_t, pid, unsigned int, len, + unsigned long __user *, user_mask_ptr) +{ + int ret; + cpumask_var_t mask; + + if ((len * BITS_PER_BYTE) < nr_cpu_ids) + return -EINVAL; + if (len & (sizeof(unsigned long)-1)) + return -EINVAL; + + if (!alloc_cpumask_var(&mask, GFP_KERNEL)) + return -ENOMEM; + + ret = sched_getaffinity(pid, mask); + if (ret == 0) { + unsigned int retlen = min(len, cpumask_size()); + + if (copy_to_user(user_mask_ptr, mask, retlen)) + ret = -EFAULT; + else + ret = retlen; + } + free_cpumask_var(mask); + + return ret; +} + +/** + * sys_sched_yield - yield the current processor to other threads. + * + * This function yields the current CPU to other tasks. It does this by + * scheduling away the current task. If it still has the earliest deadline + * it will be scheduled again as the next task. + * + * Return: 0. + */ +static void do_sched_yield(void) +{ + struct rq *rq; + + if (!sched_yield_type) + return; + + local_irq_disable(); + rq = this_rq(); + rq_lock(rq); + + if (sched_yield_type > 1) + time_slice_expired(current, rq); + schedstat_inc(rq->yld_count); + + /* + * Since we are going to call schedule() anyway, there's + * no need to preempt or enable interrupts: + */ + preempt_disable(); + rq_unlock(rq); + sched_preempt_enable_no_resched(); + + schedule(); +} + +SYSCALL_DEFINE0(sched_yield) +{ + do_sched_yield(); + return 0; +} + +#ifndef CONFIG_PREEMPTION +int __sched _cond_resched(void) +{ + if (should_resched(0)) { + preempt_schedule_common(); + return 1; + } + rcu_all_qs(); + return 0; +} +EXPORT_SYMBOL(_cond_resched); +#endif + +/* + * __cond_resched_lock() - if a reschedule is pending, drop the given lock, + * call schedule, and on return reacquire the lock. + * + * This works OK both with and without CONFIG_PREEMPTION. We do strange low-level + * operations here to prevent schedule() from being called twice (once via + * spin_unlock(), once by hand). + */ +int __cond_resched_lock(spinlock_t *lock) +{ + int resched = should_resched(PREEMPT_LOCK_OFFSET); + int ret = 0; + + lockdep_assert_held(lock); + + if (spin_needbreak(lock) || resched) { + spin_unlock(lock); + if (resched) + preempt_schedule_common(); + else + cpu_relax(); + ret = 1; + spin_lock(lock); + } + return ret; +} +EXPORT_SYMBOL(__cond_resched_lock); + +/** + * yield - yield the current processor to other threads. + * + * Do not ever use this function, there's a 99% chance you're doing it wrong. + * + * The scheduler is at all times free to pick the calling task as the most + * eligible task to run, if removing the yield() call from your code breaks + * it, its already broken. + * + * Typical broken usage is: + * + * while (!event) + * yield(); + * + * where one assumes that yield() will let 'the other' process run that will + * make event true. If the current task is a SCHED_FIFO task that will never + * happen. Never use yield() as a progress guarantee!! + * + * If you want to use yield() to wait for something, use wait_event(). + * If you want to use yield() to be 'nice' for others, use cond_resched(). + * If you still want to use yield(), do not! + */ +void __sched yield(void) +{ + set_current_state(TASK_RUNNING); + do_sched_yield(); +} +EXPORT_SYMBOL(yield); + +/** + * yield_to - yield the current processor to another thread in + * your thread group, or accelerate that thread toward the + * processor it's on. + * @p: target task + * @preempt: whether task preemption is allowed or not + * + * It's the caller's job to ensure that the target task struct + * can't go away on us before we can do any checks. + * + * Return: + * true (>0) if we indeed boosted the target task. + * false (0) if we failed to boost the target. + * -ESRCH if there's no task to yield to. + */ +int __sched yield_to(struct task_struct *p, bool preempt) +{ + struct task_struct *rq_p; + struct rq *rq, *p_rq; + unsigned long flags; + int yielded = 0; + + local_irq_save(flags); + rq = this_rq(); + +again: + p_rq = task_rq(p); + /* + * If we're the only runnable task on the rq and target rq also + * has only one task, there's absolutely no point in yielding. + */ + if (task_running(p_rq, p) || p->state) { + yielded = -ESRCH; + goto out_irq; + } + + double_rq_lock(rq, p_rq); + if (unlikely(task_rq(p) != p_rq)) { + double_rq_unlock(rq, p_rq); + goto again; + } + + yielded = 1; + schedstat_inc(rq->yld_count); + rq_p = rq->curr; + if (p->deadline > rq_p->deadline) + p->deadline = rq_p->deadline; + p->time_slice += rq_p->time_slice; + if (p->time_slice > timeslice()) + p->time_slice = timeslice(); + time_slice_expired(rq_p, rq); + if (preempt && rq != p_rq) + resched_task(p_rq->curr); + double_rq_unlock(rq, p_rq); +out_irq: + local_irq_restore(flags); + + if (yielded > 0) + schedule(); + return yielded; +} +EXPORT_SYMBOL_GPL(yield_to); + +int io_schedule_prepare(void) +{ + int old_iowait = current->in_iowait; + + current->in_iowait = 1; + blk_schedule_flush_plug(current); + + return old_iowait; +} + +void io_schedule_finish(int token) +{ + current->in_iowait = token; +} + +/* + * This task is about to go to sleep on IO. Increment rq->nr_iowait so + * that process accounting knows that this is a task in IO wait state. + * + * But don't do that if it is a deliberate, throttling IO wait (this task + * has set its backing_dev_info: the queue against which it should throttle) + */ + +long __sched io_schedule_timeout(long timeout) +{ + int token; + long ret; + + token = io_schedule_prepare(); + ret = schedule_timeout(timeout); + io_schedule_finish(token); + + return ret; +} +EXPORT_SYMBOL(io_schedule_timeout); + +void __sched io_schedule(void) +{ + int token; + + token = io_schedule_prepare(); + schedule(); + io_schedule_finish(token); +} +EXPORT_SYMBOL(io_schedule); + +/** + * sys_sched_get_priority_max - return maximum RT priority. + * @policy: scheduling class. + * + * Return: On success, this syscall returns the maximum + * rt_priority that can be used by a given scheduling class. + * On failure, a negative error code is returned. + */ +SYSCALL_DEFINE1(sched_get_priority_max, int, policy) +{ + int ret = -EINVAL; + + switch (policy) { + case SCHED_FIFO: + case SCHED_RR: + ret = MAX_USER_RT_PRIO-1; + break; + case SCHED_NORMAL: + case SCHED_BATCH: + case SCHED_ISO: + case SCHED_IDLEPRIO: + ret = 0; + break; + } + return ret; +} + +/** + * sys_sched_get_priority_min - return minimum RT priority. + * @policy: scheduling class. + * + * Return: On success, this syscall returns the minimum + * rt_priority that can be used by a given scheduling class. + * On failure, a negative error code is returned. + */ +SYSCALL_DEFINE1(sched_get_priority_min, int, policy) +{ + int ret = -EINVAL; + + switch (policy) { + case SCHED_FIFO: + case SCHED_RR: + ret = 1; + break; + case SCHED_NORMAL: + case SCHED_BATCH: + case SCHED_ISO: + case SCHED_IDLEPRIO: + ret = 0; + break; + } + return ret; +} + +static int sched_rr_get_interval(pid_t pid, struct timespec64 *t) +{ + struct task_struct *p; + unsigned int time_slice; + struct rq_flags rf; + struct rq *rq; + int retval; + + if (pid < 0) + return -EINVAL; + + retval = -ESRCH; + rcu_read_lock(); + p = find_process_by_pid(pid); + if (!p) + goto out_unlock; + + retval = security_task_getscheduler(p); + if (retval) + goto out_unlock; + + rq = task_rq_lock(p, &rf); + time_slice = p->policy == SCHED_FIFO ? 0 : MS_TO_NS(task_timeslice(p)); + task_rq_unlock(rq, p, &rf); + + rcu_read_unlock(); + *t = ns_to_timespec64(time_slice); + return 0; + +out_unlock: + rcu_read_unlock(); + return retval; +} + +/** + * sys_sched_rr_get_interval - return the default timeslice of a process. + * @pid: pid of the process. + * @interval: userspace pointer to the timeslice value. + * + * this syscall writes the default timeslice value of a given process + * into the user-space timespec buffer. A value of '0' means infinity. + * + * Return: On success, 0 and the timeslice is in @interval. Otherwise, + * an error code. + */ +SYSCALL_DEFINE2(sched_rr_get_interval, pid_t, pid, + struct __kernel_timespec __user *, interval) +{ + struct timespec64 t; + int retval = sched_rr_get_interval(pid, &t); + + if (retval == 0) + retval = put_timespec64(&t, interval); + + return retval; +} + +#ifdef CONFIG_COMPAT_32BIT_TIME +SYSCALL_DEFINE2(sched_rr_get_interval_time32, pid_t, pid, + struct old_timespec32 __user *, interval) +{ + struct timespec64 t; + int retval = sched_rr_get_interval(pid, &t); + + if (retval == 0) + retval = put_old_timespec32(&t, interval); + return retval; +} +#endif + +void sched_show_task(struct task_struct *p) +{ + unsigned long free = 0; + int ppid; + + if (!try_get_task_stack(p)) + return; + + printk(KERN_INFO "%-15.15s %c", p->comm, task_state_to_char(p)); + + if (p->state == TASK_RUNNING) + printk(KERN_CONT " running task "); +#ifdef CONFIG_DEBUG_STACK_USAGE + free = stack_not_used(p); +#endif + ppid = 0; + rcu_read_lock(); + if (pid_alive(p)) + ppid = task_pid_nr(rcu_dereference(p->real_parent)); + rcu_read_unlock(); + printk(KERN_CONT "%5lu %5d %6d 0x%08lx\n", free, + task_pid_nr(p), ppid, + (unsigned long)task_thread_info(p)->flags); + + print_worker_info(KERN_INFO, p); + show_stack(p, NULL); + put_task_stack(p); +} +EXPORT_SYMBOL_GPL(sched_show_task); + +static inline bool +state_filter_match(unsigned long state_filter, struct task_struct *p) +{ + /* no filter, everything matches */ + if (!state_filter) + return true; + + /* filter, but doesn't match */ + if (!(p->state & state_filter)) + return false; + + /* + * When looking for TASK_UNINTERRUPTIBLE skip TASK_IDLE (allows + * TASK_KILLABLE). + */ + if (state_filter == TASK_UNINTERRUPTIBLE && p->state == TASK_IDLE) + return false; + + return true; +} + +void show_state_filter(unsigned long state_filter) +{ + struct task_struct *g, *p; + +#if BITS_PER_LONG == 32 + printk(KERN_INFO + " task PC stack pid father\n"); +#else + printk(KERN_INFO + " task PC stack pid father\n"); +#endif + rcu_read_lock(); + for_each_process_thread(g, p) { + /* + * reset the NMI-timeout, listing all files on a slow + * console might take a lot of time: + * Also, reset softlockup watchdogs on all CPUs, because + * another CPU might be blocked waiting for us to process + * an IPI. + */ + touch_nmi_watchdog(); + touch_all_softlockup_watchdogs(); + if (state_filter_match(state_filter, p)) + sched_show_task(p); + } + + rcu_read_unlock(); + /* + * Only show locks if all tasks are dumped: + */ + if (!state_filter) + debug_show_all_locks(); +} + +void dump_cpu_task(int cpu) +{ + pr_info("Task dump for CPU %d:\n", cpu); + sched_show_task(cpu_curr(cpu)); +} + +#ifdef CONFIG_SMP +void set_cpus_allowed_common(struct task_struct *p, const struct cpumask *new_mask) +{ + cpumask_copy(&p->cpus_mask, new_mask); + p->nr_cpus_allowed = cpumask_weight(new_mask); +} + +void __do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) +{ + struct rq *rq = task_rq(p); + + lockdep_assert_held(&p->pi_lock); + + cpumask_copy(&p->cpus_mask, new_mask); + + if (task_queued(p)) { + /* + * Because __kthread_bind() calls this on blocked tasks without + * holding rq->lock. + */ + lockdep_assert_held(rq->lock); + } +} + +/* + * Calling do_set_cpus_allowed from outside the scheduler code should not be + * called on a running or queued task. We should be holding pi_lock. + */ +void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) +{ + __do_set_cpus_allowed(p, new_mask); + if (needs_other_cpu(p, task_cpu(p))) { + struct rq *rq; + + rq = __task_rq_lock(p, NULL); + set_task_cpu(p, valid_task_cpu(p)); + resched_task(p); + __task_rq_unlock(rq, NULL); + } +} +#endif + +/** + * init_idle - set up an idle thread for a given CPU + * @idle: task in question + * @cpu: cpu the idle task belongs to + * + * NOTE: this function does not set the idle thread's NEED_RESCHED + * flag, to make booting more robust. + */ +void init_idle(struct task_struct *idle, int cpu) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long flags; + + raw_spin_lock_irqsave(&idle->pi_lock, flags); + raw_spin_lock(rq->lock); + idle->last_ran = rq->niffies; + time_slice_expired(idle, rq); + idle->state = TASK_RUNNING; + /* Setting prio to illegal value shouldn't matter when never queued */ + idle->prio = PRIO_LIMIT; + idle->flags |= PF_IDLE; + + kasan_unpoison_task_stack(idle); + +#ifdef CONFIG_SMP + /* + * It's possible that init_idle() gets called multiple times on a task, + * in that case do_set_cpus_allowed() will not do the right thing. + * + * And since this is boot we can forgo the serialisation. + */ + set_cpus_allowed_common(idle, cpumask_of(cpu)); +#ifdef CONFIG_SMT_NICE + idle->smt_bias = 0; +#endif +#endif + set_rq_task(rq, idle); + + /* Silence PROVE_RCU */ + rcu_read_lock(); + set_task_cpu(idle, cpu); + rcu_read_unlock(); + + rq->idle = idle; + rcu_assign_pointer(rq->curr, idle); + idle->on_rq = TASK_ON_RQ_QUEUED; + raw_spin_unlock(rq->lock); + raw_spin_unlock_irqrestore(&idle->pi_lock, flags); + + /* Set the preempt count _outside_ the spinlocks! */ + init_idle_preempt_count(idle, cpu); + + ftrace_graph_init_idle_task(idle, cpu); + vtime_init_idle(idle, cpu); +#ifdef CONFIG_SMP + sprintf(idle->comm, "%s/%d", INIT_TASK_COMM, cpu); +#endif +} + +int cpuset_cpumask_can_shrink(const struct cpumask __maybe_unused *cur, + const struct cpumask __maybe_unused *trial) +{ + return 1; +} + +int task_can_attach(struct task_struct *p, + const struct cpumask *cs_cpus_allowed) +{ + int ret = 0; + + /* + * Kthreads which disallow setaffinity shouldn't be moved + * to a new cpuset; we don't want to change their CPU + * affinity and isolating such threads by their set of + * allowed nodes is unnecessary. Thus, cpusets are not + * applicable for such threads. This prevents checking for + * success of set_cpus_allowed_ptr() on all attached tasks + * before cpus_mask may be changed. + */ + if (p->flags & PF_NO_SETAFFINITY) + ret = -EINVAL; + + return ret; +} + +void resched_cpu(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + struct rq_flags rf; + + rq_lock_irqsave(rq, &rf); + if (cpu_online(cpu) || cpu == smp_processor_id()) + resched_curr(rq); + rq_unlock_irqrestore(rq, &rf); +} + +#ifdef CONFIG_SMP +#ifdef CONFIG_NO_HZ_COMMON +void select_nohz_load_balancer(int stop_tick) +{ +} + +void set_cpu_sd_state_idle(void) {} +void nohz_balance_enter_idle(int cpu) {} + +/* + * In the semi idle case, use the nearest busy CPU for migrating timers + * from an idle CPU. This is good for power-savings. + * + * We don't do similar optimization for completely idle system, as + * selecting an idle CPU will add more delays to the timers than intended + * (as that CPU's timer base may not be uptodate wrt jiffies etc). + */ +int get_nohz_timer_target(void) +{ + int i, cpu = smp_processor_id(), default_cpu = -1; + struct sched_domain *sd; + + if (housekeeping_cpu(cpu, HK_FLAG_TIMER)) { + if (!idle_cpu(cpu)) + return cpu; + default_cpu = cpu; + } + + rcu_read_lock(); + for_each_domain(cpu, sd) { + for_each_cpu_and(i, sched_domain_span(sd), + housekeeping_cpumask(HK_FLAG_TIMER)) { + if (cpu == i) + continue; + + if (!idle_cpu(i)) { + cpu = i; + goto unlock; + } + } + } + + if (default_cpu == -1) + default_cpu = housekeeping_any_cpu(HK_FLAG_TIMER); + cpu = default_cpu; +unlock: + rcu_read_unlock(); + return cpu; +} + +/* + * When add_timer_on() enqueues a timer into the timer wheel of an + * idle CPU then this timer might expire before the next timer event + * which is scheduled to wake up that CPU. In case of a completely + * idle system the next event might even be infinite time into the + * future. wake_up_idle_cpu() ensures that the CPU is woken up and + * leaves the inner idle loop so the newly added timer is taken into + * account when the CPU goes back to idle and evaluates the timer + * wheel for the next timer event. + */ +void wake_up_idle_cpu(int cpu) +{ + if (cpu == smp_processor_id()) + return; + + if (set_nr_and_not_polling(cpu_rq(cpu)->idle)) + smp_sched_reschedule(cpu); + else + trace_sched_wake_idle_without_ipi(cpu); +} + +static bool wake_up_full_nohz_cpu(int cpu) +{ + /* + * We just need the target to call irq_exit() and re-evaluate + * the next tick. The nohz full kick at least implies that. + * If needed we can still optimize that later with an + * empty IRQ. + */ + if (cpu_is_offline(cpu)) + return true; /* Don't try to wake offline CPUs. */ + if (tick_nohz_full_cpu(cpu)) { + if (cpu != smp_processor_id() || + tick_nohz_tick_stopped()) + tick_nohz_full_kick_cpu(cpu); + return true; + } + + return false; +} + +/* + * Wake up the specified CPU. If the CPU is going offline, it is the + * caller's responsibility to deal with the lost wakeup, for example, + * by hooking into the CPU_DEAD notifier like timers and hrtimers do. + */ +void wake_up_nohz_cpu(int cpu) +{ + if (!wake_up_full_nohz_cpu(cpu)) + wake_up_idle_cpu(cpu); +} +#endif /* CONFIG_NO_HZ_COMMON */ + +/* + * Change a given task's CPU affinity. Migrate the thread to a + * proper CPU and schedule it away if the CPU it's executing on + * is removed from the allowed bitmask. + * + * NOTE: the caller must have a valid reference to the task, the + * task must not exit() & deallocate itself prematurely. The + * call is not atomic; no spinlocks may be held. + */ +static int __set_cpus_allowed_ptr(struct task_struct *p, + const struct cpumask *new_mask, bool check) +{ + const struct cpumask *cpu_valid_mask = cpu_active_mask; + bool queued = false, running_wrong = false, kthread; + unsigned int dest_cpu; + struct rq_flags rf; + struct rq *rq; + int ret = 0; + + rq = task_rq_lock(p, &rf); + update_rq_clock(rq); + + kthread = !!(p->flags & PF_KTHREAD); + if (kthread) { + /* + * Kernel threads are allowed on online && !active CPUs + */ + cpu_valid_mask = cpu_online_mask; + } + + /* + * Must re-check here, to close a race against __kthread_bind(), + * sched_setaffinity() is not guaranteed to observe the flag. + */ + if (check && (p->flags & PF_NO_SETAFFINITY)) { + ret = -EINVAL; + goto out; + } + + if (cpumask_equal(p->cpus_ptr, new_mask)) + goto out; + + /* + * Picking a ~random cpu helps in cases where we are changing affinity + * for groups of tasks (ie. cpuset), so that load balancing is not + * immediately required to distribute the tasks within their new mask. + */ + dest_cpu = cpumask_any_and_distribute(cpu_valid_mask, new_mask); + if (dest_cpu >= nr_cpu_ids) { + ret = -EINVAL; + goto out; + } + + queued = task_queued(p); + __do_set_cpus_allowed(p, new_mask); + + if (kthread) { + /* + * For kernel threads that do indeed end up on online && + * !active we want to ensure they are strict per-CPU threads. + */ + WARN_ON(cpumask_intersects(new_mask, cpu_online_mask) && + !cpumask_intersects(new_mask, cpu_active_mask) && + p->nr_cpus_allowed != 1); + } + + /* Can the task run on the task's current CPU? If so, we're done */ + if (cpumask_test_cpu(task_cpu(p), new_mask)) + goto out; + + if (task_running(rq, p)) { + /* Task is running on the wrong cpu now, reschedule it. */ + if (rq == this_rq()) { + set_task_cpu(p, dest_cpu); + set_tsk_need_resched(p); + running_wrong = true; + } else + resched_task(p); + } else { + if (queued) { + /* + * Switch runqueue locks after dequeueing the task + * here while still holding the pi_lock to be holding + * the correct lock for enqueueing. + */ + dequeue_task(rq, p, 0); + rq_unlock(rq); + + rq = cpu_rq(dest_cpu); + rq_lock(rq); + } + set_task_cpu(p, dest_cpu); + if (queued) + enqueue_task(rq, p, 0); + } + if (queued) + try_preempt(p, rq); + if (running_wrong) + preempt_disable(); +out: + task_rq_unlock(rq, p, &rf); + + if (running_wrong) { + __schedule(true); + preempt_enable(); + } + + return ret; +} + +int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask) +{ + return __set_cpus_allowed_ptr(p, new_mask, false); +} +EXPORT_SYMBOL_GPL(set_cpus_allowed_ptr); + +#ifdef CONFIG_HOTPLUG_CPU +/* + * Run through task list and find tasks affined to the dead cpu, then remove + * that cpu from the list, enable cpu0 and set the zerobound flag. Must hold + * cpu 0 and src_cpu's runqueue locks. We should be holding both rq lock and + * pi_lock to change cpus_mask but it's not going to matter here. + */ +static void bind_zero(int src_cpu) +{ + struct task_struct *p, *t; + struct rq *rq0; + int bound = 0; + + if (src_cpu == 0) + return; + + rq0 = cpu_rq(0); + + do_each_thread(t, p) { + if (cpumask_test_cpu(src_cpu, p->cpus_ptr)) { + bool local = (task_cpu(p) == src_cpu); + struct rq *rq = task_rq(p); + + /* task_running is the cpu stopper thread */ + if (local && task_running(rq, p)) + continue; + atomic_clear_cpu(src_cpu, &p->cpus_mask); + atomic_set_cpu(0, &p->cpus_mask); + p->zerobound = true; + bound++; + if (local) { + bool queued = task_queued(p); + + if (queued) + dequeue_task(rq, p, 0); + set_task_cpu(p, 0); + if (queued) + enqueue_task(rq0, p, 0); + } + } + } while_each_thread(t, p); + + if (bound) { + printk(KERN_INFO "MuQSS removed affinity for %d processes to cpu %d\n", + bound, src_cpu); + } +} + +/* Find processes with the zerobound flag and reenable their affinity for the + * CPU coming alive. */ +static void unbind_zero(int src_cpu) +{ + int unbound = 0, zerobound = 0; + struct task_struct *p, *t; + + if (src_cpu == 0) + return; + + do_each_thread(t, p) { + if (!p->mm) + p->zerobound = false; + if (p->zerobound) { + unbound++; + cpumask_set_cpu(src_cpu, &p->cpus_mask); + /* Once every CPU affinity has been re-enabled, remove + * the zerobound flag */ + if (cpumask_subset(cpu_possible_mask, p->cpus_ptr)) { + p->zerobound = false; + zerobound++; + } + } + } while_each_thread(t, p); + + if (unbound) { + printk(KERN_INFO "MuQSS added affinity for %d processes to cpu %d\n", + unbound, src_cpu); + } + if (zerobound) { + printk(KERN_INFO "MuQSS released forced binding to cpu0 for %d processes\n", + zerobound); + } +} + +/* + * Ensure that the idle task is using init_mm right before its cpu goes + * offline. + */ +void idle_task_exit(void) +{ + struct mm_struct *mm = current->active_mm; + + BUG_ON(cpu_online(smp_processor_id())); + + if (mm != &init_mm) { + switch_mm(mm, &init_mm, current); + current->active_mm = &init_mm; + finish_arch_post_lock_switch(); + } + mmdrop(mm); +} +#else /* CONFIG_HOTPLUG_CPU */ +static void unbind_zero(int src_cpu) {} +#endif /* CONFIG_HOTPLUG_CPU */ + +void sched_set_stop_task(int cpu, struct task_struct *stop) +{ + struct sched_param stop_param = { .sched_priority = STOP_PRIO }; + struct sched_param start_param = { .sched_priority = 0 }; + struct task_struct *old_stop = cpu_rq(cpu)->stop; + + if (stop) { + /* + * Make it appear like a SCHED_FIFO task, its something + * userspace knows about and won't get confused about. + * + * Also, it will make PI more or less work without too + * much confusion -- but then, stop work should not + * rely on PI working anyway. + */ + sched_setscheduler_nocheck(stop, SCHED_FIFO, &stop_param); + } + + cpu_rq(cpu)->stop = stop; + + if (old_stop) { + /* + * Reset it back to a normal scheduling policy so that + * it can die in pieces. + */ + sched_setscheduler_nocheck(old_stop, SCHED_NORMAL, &start_param); + } +} + +#if defined(CONFIG_SCHED_DEBUG) && defined(CONFIG_SYSCTL) + +static struct ctl_table sd_ctl_dir[] = { + { + .procname = "sched_domain", + .mode = 0555, + }, + {} +}; + +static struct ctl_table sd_ctl_root[] = { + { + .procname = "kernel", + .mode = 0555, + .child = sd_ctl_dir, + }, + {} +}; + +static struct ctl_table *sd_alloc_ctl_entry(int n) +{ + struct ctl_table *entry = + kcalloc(n, sizeof(struct ctl_table), GFP_KERNEL); + + return entry; +} + +static void sd_free_ctl_entry(struct ctl_table **tablep) +{ + struct ctl_table *entry; + + /* + * In the intermediate directories, both the child directory and + * procname are dynamically allocated and could fail but the mode + * will always be set. In the lowest directory the names are + * static strings and all have proc handlers. + */ + for (entry = *tablep; entry->mode; entry++) { + if (entry->child) + sd_free_ctl_entry(&entry->child); + if (entry->proc_handler == NULL) + kfree(entry->procname); + } + + kfree(*tablep); + *tablep = NULL; +} + +static void +set_table_entry(struct ctl_table *entry, + const char *procname, void *data, int maxlen, + umode_t mode, proc_handler *proc_handler) +{ + entry->procname = procname; + entry->data = data; + entry->maxlen = maxlen; + entry->mode = mode; + entry->proc_handler = proc_handler; +} + +static struct ctl_table * +sd_alloc_ctl_domain_table(struct sched_domain *sd) +{ + struct ctl_table *table = sd_alloc_ctl_entry(9); + + if (table == NULL) + return NULL; + + set_table_entry(&table[0], "min_interval", &sd->min_interval, sizeof(long), 0644, proc_doulongvec_minmax); + set_table_entry(&table[1], "max_interval", &sd->max_interval, sizeof(long), 0644, proc_doulongvec_minmax); + set_table_entry(&table[2], "busy_factor", &sd->busy_factor, sizeof(int), 0644, proc_dointvec_minmax); + set_table_entry(&table[3], "imbalance_pct", &sd->imbalance_pct, sizeof(int), 0644, proc_dointvec_minmax); + set_table_entry(&table[4], "cache_nice_tries", &sd->cache_nice_tries, sizeof(int), 0644, proc_dointvec_minmax); + set_table_entry(&table[5], "flags", &sd->flags, sizeof(int), 0644, proc_dointvec_minmax); + set_table_entry(&table[6], "max_newidle_lb_cost", &sd->max_newidle_lb_cost, sizeof(long), 0644, proc_doulongvec_minmax); + set_table_entry(&table[7], "name", sd->name, CORENAME_MAX_SIZE, 0444, proc_dostring); + /* &table[8] is terminator */ + + return table; +} + +static struct ctl_table *sd_alloc_ctl_cpu_table(int cpu) +{ + struct ctl_table *entry, *table; + struct sched_domain *sd; + int domain_num = 0, i; + char buf[32]; + + for_each_domain(cpu, sd) + domain_num++; + entry = table = sd_alloc_ctl_entry(domain_num + 1); + if (table == NULL) + return NULL; + + i = 0; + for_each_domain(cpu, sd) { + snprintf(buf, 32, "domain%d", i); + entry->procname = kstrdup(buf, GFP_KERNEL); + entry->mode = 0555; + entry->child = sd_alloc_ctl_domain_table(sd); + entry++; + i++; + } + return table; +} + +static cpumask_var_t sd_sysctl_cpus; +static struct ctl_table_header *sd_sysctl_header; + +void register_sched_domain_sysctl(void) +{ + static struct ctl_table *cpu_entries; + static struct ctl_table **cpu_idx; + char buf[32]; + int i; + + if (!cpu_entries) { + cpu_entries = sd_alloc_ctl_entry(num_possible_cpus() + 1); + if (!cpu_entries) + return; + + WARN_ON(sd_ctl_dir[0].child); + sd_ctl_dir[0].child = cpu_entries; + } + + if (!cpu_idx) { + struct ctl_table *e = cpu_entries; + + cpu_idx = kcalloc(nr_cpu_ids, sizeof(struct ctl_table*), GFP_KERNEL); + if (!cpu_idx) + return; + + /* deal with sparse possible map */ + for_each_possible_cpu(i) { + cpu_idx[i] = e; + e++; + } + } + + if (!cpumask_available(sd_sysctl_cpus)) { + if (!alloc_cpumask_var(&sd_sysctl_cpus, GFP_KERNEL)) + return; + + /* init to possible to not have holes in @cpu_entries */ + cpumask_copy(sd_sysctl_cpus, cpu_possible_mask); + } + + for_each_cpu(i, sd_sysctl_cpus) { + struct ctl_table *e = cpu_idx[i]; + + if (e->child) + sd_free_ctl_entry(&e->child); + + if (!e->procname) { + snprintf(buf, 32, "cpu%d", i); + e->procname = kstrdup(buf, GFP_KERNEL); + } + e->mode = 0555; + e->child = sd_alloc_ctl_cpu_table(i); + + __cpumask_clear_cpu(i, sd_sysctl_cpus); + } + + WARN_ON(sd_sysctl_header); + sd_sysctl_header = register_sysctl_table(sd_ctl_root); +} + +void dirty_sched_domain_sysctl(int cpu) +{ + if (cpumask_available(sd_sysctl_cpus)) + __cpumask_set_cpu(cpu, sd_sysctl_cpus); +} + +/* may be called multiple times per register */ +void unregister_sched_domain_sysctl(void) +{ + unregister_sysctl_table(sd_sysctl_header); + sd_sysctl_header = NULL; +} +#endif /* CONFIG_SYSCTL */ + +void set_rq_online(struct rq *rq) +{ + if (!rq->online) { + cpumask_set_cpu(cpu_of(rq), rq->rd->online); + rq->online = true; + } +} + +void set_rq_offline(struct rq *rq) +{ + if (rq->online) { + int cpu = cpu_of(rq); + + cpumask_clear_cpu(cpu, rq->rd->online); + rq->online = false; + clear_cpuidle_map(cpu); + } +} + +/* + * used to mark begin/end of suspend/resume: + */ +static int num_cpus_frozen; + +/* + * Update cpusets according to cpu_active mask. If cpusets are + * disabled, cpuset_update_active_cpus() becomes a simple wrapper + * around partition_sched_domains(). + * + * If we come here as part of a suspend/resume, don't touch cpusets because we + * want to restore it back to its original state upon resume anyway. + */ +static void cpuset_cpu_active(void) +{ + if (cpuhp_tasks_frozen) { + /* + * num_cpus_frozen tracks how many CPUs are involved in suspend + * resume sequence. As long as this is not the last online + * operation in the resume sequence, just build a single sched + * domain, ignoring cpusets. + */ + partition_sched_domains(1, NULL, NULL); + if (--num_cpus_frozen) + return; + /* + * This is the last CPU online operation. So fall through and + * restore the original sched domains by considering the + * cpuset configurations. + */ + cpuset_force_rebuild(); + } + + cpuset_update_active_cpus(); +} + +static int cpuset_cpu_inactive(unsigned int cpu) +{ + if (!cpuhp_tasks_frozen) { + cpuset_update_active_cpus(); + } else { + num_cpus_frozen++; + partition_sched_domains(1, NULL, NULL); + } + return 0; +} + +int sched_cpu_activate(unsigned int cpu) +{ + struct rq *rq = cpu_rq(cpu); + struct rq_flags rf; + +#ifdef CONFIG_SCHED_SMT + /* + * When going up, increment the number of cores with SMT present. + */ + if (cpumask_weight(cpu_smt_mask(cpu)) == 2) + static_branch_inc_cpuslocked(&sched_smt_present); +#endif + set_cpu_active(cpu, true); + + if (sched_smp_initialized) { + sched_domains_numa_masks_set(cpu); + cpuset_cpu_active(); + } + + /* + * Put the rq online, if not already. This happens: + * + * 1) In the early boot process, because we build the real domains + * after all CPUs have been brought up. + * + * 2) At runtime, if cpuset_cpu_active() fails to rebuild the + * domains. + */ + rq_lock_irqsave(rq, &rf); + if (rq->rd) { + BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); + set_rq_online(rq); + } + unbind_zero(cpu); + rq_unlock_irqrestore(rq, &rf); + + return 0; +} + +int sched_cpu_deactivate(unsigned int cpu) +{ + int ret; + + set_cpu_active(cpu, false); + /* + * We've cleared cpu_active_mask, wait for all preempt-disabled and RCU + * users of this state to go away such that all new such users will + * observe it. + * + * Do sync before park smpboot threads to take care the rcu boost case. + */ + synchronize_rcu(); + +#ifdef CONFIG_SCHED_SMT + /* + * When going down, decrement the number of cores with SMT present. + */ + if (cpumask_weight(cpu_smt_mask(cpu)) == 2) + static_branch_dec_cpuslocked(&sched_smt_present); +#endif + + if (!sched_smp_initialized) + return 0; + + ret = cpuset_cpu_inactive(cpu); + if (ret) { + set_cpu_active(cpu, true); + return ret; + } + sched_domains_numa_masks_clear(cpu); + return 0; +} + +int sched_cpu_starting(unsigned int cpu) +{ + sched_tick_start(cpu); + return 0; +} + +#ifdef CONFIG_HOTPLUG_CPU +int sched_cpu_dying(unsigned int cpu) +{ + struct rq *rq = cpu_rq(cpu); + unsigned long flags; + + /* Handle pending wakeups and then migrate everything off */ + sched_ttwu_pending(); + sched_tick_stop(cpu); + + local_irq_save(flags); + double_rq_lock(rq, cpu_rq(0)); + if (rq->rd) { + BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); + set_rq_offline(rq); + } + bind_zero(cpu); + double_rq_unlock(rq, cpu_rq(0)); + sched_start_tick(rq, cpu); + hrexpiry_clear(rq); + local_irq_restore(flags); + + return 0; +} +#endif + +#if defined(CONFIG_SCHED_SMT) || defined(CONFIG_SCHED_MC) +/* + * Cheaper version of the below functions in case support for SMT and MC is + * compiled in but CPUs have no siblings. + */ +static bool sole_cpu_idle(struct rq *rq) +{ + return rq_idle(rq); +} +#endif +#ifdef CONFIG_SCHED_SMT +static const cpumask_t *thread_cpumask(int cpu) +{ + return topology_sibling_cpumask(cpu); +} +/* All this CPU's SMT siblings are idle */ +static bool siblings_cpu_idle(struct rq *rq) +{ + return cpumask_subset(&rq->thread_mask, &cpu_idle_map); +} +#endif +#ifdef CONFIG_SCHED_MC +static const cpumask_t *core_cpumask(int cpu) +{ + return topology_core_cpumask(cpu); +} +/* All this CPU's shared cache siblings are idle */ +static bool cache_cpu_idle(struct rq *rq) +{ + return cpumask_subset(&rq->core_mask, &cpu_idle_map); +} +/* MC siblings CPU mask which share the same LLC */ +static const cpumask_t *llc_core_cpumask(int cpu) +{ +#ifdef CONFIG_X86 + return per_cpu(cpu_llc_shared_map, cpu); +#else + return topology_core_cpumask(cpu); +#endif +} +#endif + +enum sched_domain_level { + SD_LV_NONE = 0, + SD_LV_SIBLING, + SD_LV_MC, + SD_LV_BOOK, + SD_LV_CPU, + SD_LV_NODE, + SD_LV_ALLNODES, + SD_LV_MAX +}; + +/* + * Set up the relative cache distance of each online cpu from each + * other in a simple array for quick lookup. Locality is determined + * by the closest sched_domain that CPUs are separated by. CPUs with + * shared cache in SMT and MC are treated as local. Separate CPUs + * (within the same package or physically) within the same node are + * treated as not local. CPUs not even in the same domain (different + * nodes) are treated as very distant. + */ +static void __init select_leaders(void) +{ + struct rq *rq, *other_rq, *leader; + struct sched_domain *sd; + int cpu, other_cpu; +#ifdef CONFIG_SCHED_SMT + bool smt_threads = false; +#endif + + for (cpu = 0; cpu < num_online_cpus(); cpu++) { + rq = cpu_rq(cpu); + leader = NULL; + /* First check if this cpu is in the same node */ + for_each_domain(cpu, sd) { + if (sd->level > SD_LV_MC) + continue; + if (rqshare != RQSHARE_ALL) + leader = NULL; + /* Set locality to local node if not already found lower */ + for_each_cpu(other_cpu, sched_domain_span(sd)) { + if (rqshare >= RQSHARE_SMP) { + other_rq = cpu_rq(other_cpu); + + /* Set the smp_leader to the first CPU */ + if (!leader) + leader = rq; + if (!other_rq->smp_leader) + other_rq->smp_leader = leader; + } + if (rq->cpu_locality[other_cpu] > LOCALITY_SMP) + rq->cpu_locality[other_cpu] = LOCALITY_SMP; + } + } + + /* + * Each runqueue has its own function in case it doesn't have + * siblings of its own allowing mixed topologies. + */ +#ifdef CONFIG_SCHED_MC + leader = NULL; + if (cpumask_weight(core_cpumask(cpu)) > 1) { + cpumask_copy(&rq->core_mask, llc_core_cpumask(cpu)); + cpumask_clear_cpu(cpu, &rq->core_mask); + for_each_cpu(other_cpu, core_cpumask(cpu)) { + if (rqshare == RQSHARE_MC || + (rqshare == RQSHARE_MC_LLC && cpumask_test_cpu(other_cpu, llc_core_cpumask(cpu)))) { + other_rq = cpu_rq(other_cpu); + + /* Set the mc_leader to the first CPU */ + if (!leader) + leader = rq; + if (!other_rq->mc_leader) + other_rq->mc_leader = leader; + } + if (rq->cpu_locality[other_cpu] > LOCALITY_MC) { + /* this is to get LLC into play even in case LLC sharing is not used */ + if (cpumask_test_cpu(other_cpu, llc_core_cpumask(cpu))) + rq->cpu_locality[other_cpu] = LOCALITY_MC_LLC; + else + rq->cpu_locality[other_cpu] = LOCALITY_MC; + } + } + rq->cache_idle = cache_cpu_idle; + } +#endif +#ifdef CONFIG_SCHED_SMT + leader = NULL; + if (cpumask_weight(thread_cpumask(cpu)) > 1) { + cpumask_copy(&rq->thread_mask, thread_cpumask(cpu)); + cpumask_clear_cpu(cpu, &rq->thread_mask); + for_each_cpu(other_cpu, thread_cpumask(cpu)) { + if (rqshare == RQSHARE_SMT) { + other_rq = cpu_rq(other_cpu); + + /* Set the smt_leader to the first CPU */ + if (!leader) + leader = rq; + if (!other_rq->smt_leader) + other_rq->smt_leader = leader; + } + if (rq->cpu_locality[other_cpu] > LOCALITY_SMT) + rq->cpu_locality[other_cpu] = LOCALITY_SMT; + } + rq->siblings_idle = siblings_cpu_idle; + smt_threads = true; + } +#endif + } + +#ifdef CONFIG_SMT_NICE + if (smt_threads) { + check_siblings = &check_smt_siblings; + wake_siblings = &wake_smt_siblings; + smt_schedule = &smt_should_schedule; + } +#endif + + for_each_online_cpu(cpu) { + rq = cpu_rq(cpu); + for_each_online_cpu(other_cpu) { + printk(KERN_DEBUG "MuQSS locality CPU %d to %d: %d\n", cpu, other_cpu, rq->cpu_locality[other_cpu]); + } + } +} + +/* FIXME freeing locked spinlock */ +static void __init share_and_free_rq(struct rq *leader, struct rq *rq) +{ + WARN_ON(rq->nr_running > 0); + + kfree(rq->node); + kfree(rq->sl); + kfree(rq->lock); + rq->node = leader->node; + rq->sl = leader->sl; + rq->lock = leader->lock; + rq->is_leader = false; + barrier(); + /* To make up for not unlocking the freed runlock */ + preempt_enable(); +} + +static void __init share_rqs(void) +{ + struct rq *rq, *leader; + int cpu; + + for_each_online_cpu(cpu) { + rq = cpu_rq(cpu); + leader = rq->smp_leader; + + rq_lock(rq); + if (leader && rq != leader) { + printk(KERN_INFO "MuQSS sharing SMP runqueue from CPU %d to CPU %d\n", + leader->cpu, rq->cpu); + share_and_free_rq(leader, rq); + } else + rq_unlock(rq); + } + +#ifdef CONFIG_SCHED_MC + for_each_online_cpu(cpu) { + rq = cpu_rq(cpu); + leader = rq->mc_leader; + + rq_lock(rq); + if (leader && rq != leader) { + printk(KERN_INFO "MuQSS sharing MC runqueue from CPU %d to CPU %d\n", + leader->cpu, rq->cpu); + share_and_free_rq(leader, rq); + } else + rq_unlock(rq); + } +#endif /* CONFIG_SCHED_MC */ + +#ifdef CONFIG_SCHED_SMT + for_each_online_cpu(cpu) { + rq = cpu_rq(cpu); + leader = rq->smt_leader; + + rq_lock(rq); + if (leader && rq != leader) { + printk(KERN_INFO "MuQSS sharing SMT runqueue from CPU %d to CPU %d\n", + leader->cpu, rq->cpu); + share_and_free_rq(leader, rq); + } else + rq_unlock(rq); + } +#endif /* CONFIG_SCHED_SMT */ +} + +static void __init setup_rq_orders(void) +{ + int *selected_cpus, *ordered_cpus; + struct rq *rq, *other_rq; + int cpu, other_cpu, i; + + selected_cpus = kmalloc(sizeof(int) * NR_CPUS, GFP_ATOMIC); + ordered_cpus = kmalloc(sizeof(int) * NR_CPUS, GFP_ATOMIC); + + total_runqueues = 0; + for_each_online_cpu(cpu) { + int locality, total_rqs = 0, total_cpus = 0; + + rq = cpu_rq(cpu); + if (rq->is_leader) + total_runqueues++; + + for (locality = LOCALITY_SAME; locality <= LOCALITY_DISTANT; locality++) { + int selected_cpu_cnt, selected_cpu_idx, test_cpu_idx, cpu_idx, best_locality, test_cpu; + int ordered_cpus_idx; + + ordered_cpus_idx = -1; + selected_cpu_cnt = 0; + + for_each_online_cpu(test_cpu) { + if (cpu < num_online_cpus() / 2) + other_cpu = cpu + test_cpu; + else + other_cpu = cpu - test_cpu; + if (other_cpu < 0) + other_cpu += num_online_cpus(); + else + other_cpu %= num_online_cpus(); + /* gather CPUs of the same locality */ + if (rq->cpu_locality[other_cpu] == locality) { + selected_cpus[selected_cpu_cnt] = other_cpu; + selected_cpu_cnt++; + } + } + + /* reserve first CPU as starting point */ + if (selected_cpu_cnt > 0) { + ordered_cpus_idx++; + ordered_cpus[ordered_cpus_idx] = selected_cpus[ordered_cpus_idx]; + selected_cpus[ordered_cpus_idx] = -1; + } + + /* take each CPU and sort it within the same locality based on each inter-CPU localities */ + for(test_cpu_idx = 1; test_cpu_idx < selected_cpu_cnt; test_cpu_idx++) { + /* starting point with worst locality and current CPU */ + best_locality = LOCALITY_DISTANT; + selected_cpu_idx = test_cpu_idx; + + /* try to find the best locality within group */ + for(cpu_idx = 1; cpu_idx < selected_cpu_cnt; cpu_idx++) { + /* if CPU has not been used and locality is better */ + if (selected_cpus[cpu_idx] > -1) { + other_rq = cpu_rq(ordered_cpus[ordered_cpus_idx]); + if (best_locality > other_rq->cpu_locality[selected_cpus[cpu_idx]]) { + /* assign best locality and best CPU idx in array */ + best_locality = other_rq->cpu_locality[selected_cpus[cpu_idx]]; + selected_cpu_idx = cpu_idx; + } + } + } + + /* add our next best CPU to ordered list */ + ordered_cpus_idx++; + ordered_cpus[ordered_cpus_idx] = selected_cpus[selected_cpu_idx]; + /* mark this CPU as used */ + selected_cpus[selected_cpu_idx] = -1; + } + + /* set up RQ and CPU orders */ + for (test_cpu = 0; test_cpu <= ordered_cpus_idx; test_cpu++) { + other_rq = cpu_rq(ordered_cpus[test_cpu]); + /* set up cpu orders */ + rq->cpu_order[total_cpus++] = other_rq; + if (other_rq->is_leader) { + /* set up RQ orders */ + rq->rq_order[total_rqs++] = other_rq; + } + } + } + } + + kfree(selected_cpus); + kfree(ordered_cpus); + +#ifdef CONFIG_X86 + for_each_online_cpu(cpu) { + rq = cpu_rq(cpu); + for (i = 0; i < total_runqueues; i++) { + printk(KERN_DEBUG "MuQSS CPU %d llc %d RQ order %d RQ %d llc %d\n", cpu, per_cpu(cpu_llc_id, cpu), i, + rq->rq_order[i]->cpu, per_cpu(cpu_llc_id, rq->rq_order[i]->cpu)); + } + } + + for_each_online_cpu(cpu) { + rq = cpu_rq(cpu); + for (i = 0; i < num_online_cpus(); i++) { + printk(KERN_DEBUG "MuQSS CPU %d llc %d CPU order %d RQ %d llc %d\n", cpu, per_cpu(cpu_llc_id, cpu), i, + rq->cpu_order[i]->cpu, per_cpu(cpu_llc_id, rq->cpu_order[i]->cpu)); + } + } +#endif +} + +void __init sched_init_smp(void) +{ + sched_init_numa(); + + /* + * There's no userspace yet to cause hotplug operations; hence all the + * cpu masks are stable and all blatant races in the below code cannot + * happen. + */ + mutex_lock(&sched_domains_mutex); + sched_init_domains(cpu_active_mask); + mutex_unlock(&sched_domains_mutex); + + /* Move init over to a non-isolated CPU */ + if (set_cpus_allowed_ptr(current, housekeeping_cpumask(HK_FLAG_DOMAIN)) < 0) + BUG(); + + local_irq_disable(); + mutex_lock(&sched_domains_mutex); + lock_all_rqs(); + + printk(KERN_INFO "MuQSS possible/present/online CPUs: %d/%d/%d\n", + num_possible_cpus(), num_present_cpus(), num_online_cpus()); + + select_leaders(); + + unlock_all_rqs(); + mutex_unlock(&sched_domains_mutex); + + share_rqs(); + + local_irq_enable(); + + setup_rq_orders(); + + switch (rqshare) { + case RQSHARE_ALL: + /* This should only ever read 1 */ + printk(KERN_INFO "MuQSS runqueue share type ALL total runqueues: %d\n", + total_runqueues); + break; + case RQSHARE_SMP: + printk(KERN_INFO "MuQSS runqueue share type SMP total runqueues: %d\n", + total_runqueues); + break; + case RQSHARE_MC: + printk(KERN_INFO "MuQSS runqueue share type MC total runqueues: %d\n", + total_runqueues); + break; + case RQSHARE_MC_LLC: + printk(KERN_INFO "MuQSS runqueue share type LLC total runqueues: %d\n", + total_runqueues); + break; + case RQSHARE_SMT: + printk(KERN_INFO "MuQSS runqueue share type SMT total runqueues: %d\n", + total_runqueues); + break; + case RQSHARE_NONE: + printk(KERN_INFO "MuQSS runqueue share type NONE total runqueues: %d\n", + total_runqueues); + break; + } + + sched_smp_initialized = true; +} +#else +void __init sched_init_smp(void) +{ + sched_smp_initialized = true; +} +#endif /* CONFIG_SMP */ + +int in_sched_functions(unsigned long addr) +{ + return in_lock_functions(addr) || + (addr >= (unsigned long)__sched_text_start + && addr < (unsigned long)__sched_text_end); +} + +#ifdef CONFIG_CGROUP_SCHED +/* task group related information */ +struct task_group { + struct cgroup_subsys_state css; + + struct rcu_head rcu; + struct list_head list; + + struct task_group *parent; + struct list_head siblings; + struct list_head children; +}; + +/* + * Default task group. + * Every task in system belongs to this group at bootup. + */ +struct task_group root_task_group; +LIST_HEAD(task_groups); + +/* Cacheline aligned slab cache for task_group */ +static struct kmem_cache *task_group_cache __read_mostly; +#endif /* CONFIG_CGROUP_SCHED */ + +void __init sched_init(void) +{ +#ifdef CONFIG_SMP + int cpu_ids; +#endif + int i; + struct rq *rq; + + wait_bit_init(); + + prio_ratios[0] = 128; + for (i = 1 ; i < NICE_WIDTH ; i++) + prio_ratios[i] = prio_ratios[i - 1] * 11 / 10; + + skiplist_node_init(&init_task.node); + +#ifdef CONFIG_SMP + init_defrootdomain(); + cpumask_clear(&cpu_idle_map); +#else + uprq = &per_cpu(runqueues, 0); +#endif + +#ifdef CONFIG_CGROUP_SCHED + task_group_cache = KMEM_CACHE(task_group, 0); + + list_add(&root_task_group.list, &task_groups); + INIT_LIST_HEAD(&root_task_group.children); + INIT_LIST_HEAD(&root_task_group.siblings); +#endif /* CONFIG_CGROUP_SCHED */ + for_each_possible_cpu(i) { + rq = cpu_rq(i); + rq->node = kmalloc(sizeof(skiplist_node), GFP_ATOMIC); + skiplist_init(rq->node); + rq->sl = new_skiplist(rq->node); + rq->lock = kmalloc(sizeof(raw_spinlock_t), GFP_ATOMIC); + raw_spin_lock_init(rq->lock); + rq->nr_running = 0; + rq->nr_uninterruptible = 0; + rq->nr_switches = 0; + rq->clock = rq->old_clock = rq->last_niffy = rq->niffies = 0; + rq->last_jiffy = jiffies; + rq->user_ns = rq->nice_ns = rq->softirq_ns = rq->system_ns = + rq->iowait_ns = rq->idle_ns = 0; + rq->dither = 0; + set_rq_task(rq, &init_task); + rq->iso_ticks = 0; + rq->iso_refractory = false; +#ifdef CONFIG_SMP + rq->is_leader = true; + rq->smp_leader = NULL; +#ifdef CONFIG_SCHED_MC + rq->mc_leader = NULL; +#endif +#ifdef CONFIG_SCHED_SMT + rq->smt_leader = NULL; +#endif + rq->sd = NULL; + rq->rd = NULL; + rq->online = false; + rq->cpu = i; + rq_attach_root(rq, &def_root_domain); +#endif + init_rq_hrexpiry(rq); + atomic_set(&rq->nr_iowait, 0); + } + +#ifdef CONFIG_SMP + cpu_ids = i; + /* + * Set the base locality for cpu cache distance calculation to + * "distant" (3). Make sure the distance from a CPU to itself is 0. + */ + for_each_possible_cpu(i) { + int j; + + rq = cpu_rq(i); +#ifdef CONFIG_SCHED_SMT + rq->siblings_idle = sole_cpu_idle; +#endif +#ifdef CONFIG_SCHED_MC + rq->cache_idle = sole_cpu_idle; +#endif + rq->cpu_locality = kmalloc(cpu_ids * sizeof(int *), GFP_ATOMIC); + for_each_possible_cpu(j) { + if (i == j) + rq->cpu_locality[j] = LOCALITY_SAME; + else + rq->cpu_locality[j] = LOCALITY_DISTANT; + } + rq->rq_order = kmalloc(cpu_ids * sizeof(struct rq *), GFP_ATOMIC); + rq->cpu_order = kmalloc(cpu_ids * sizeof(struct rq *), GFP_ATOMIC); + rq->rq_order[0] = rq->cpu_order[0] = rq; + for (j = 1; j < cpu_ids; j++) + rq->rq_order[j] = rq->cpu_order[j] = cpu_rq(j); + } +#endif + + /* + * The boot idle thread does lazy MMU switching as well: + */ + mmgrab(&init_mm); + enter_lazy_tlb(&init_mm, current); + + /* + * Make us the idle thread. Technically, schedule() should not be + * called from this thread, however somewhere below it might be, + * but because we are the idle thread, we just pick up running again + * when this runqueue becomes "idle". + */ + init_idle(current, smp_processor_id()); + +#ifdef CONFIG_SMP + idle_thread_set_boot_cpu(); +#endif /* SMP */ + + init_schedstats(); + + psi_init(); +} + +#ifdef CONFIG_DEBUG_ATOMIC_SLEEP +static inline int preempt_count_equals(int preempt_offset) +{ + int nested = preempt_count() + rcu_preempt_depth(); + + return (nested == preempt_offset); +} + +void __might_sleep(const char *file, int line, int preempt_offset) +{ + /* + * Blocking primitives will set (and therefore destroy) current->state, + * since we will exit with TASK_RUNNING make sure we enter with it, + * otherwise we will destroy state. + */ + WARN_ONCE(current->state != TASK_RUNNING && current->task_state_change, + "do not call blocking ops when !TASK_RUNNING; " + "state=%lx set at [<%p>] %pS\n", + current->state, + (void *)current->task_state_change, + (void *)current->task_state_change); + + ___might_sleep(file, line, preempt_offset); +} +EXPORT_SYMBOL(__might_sleep); + +void __cant_sleep(const char *file, int line, int preempt_offset) +{ + static unsigned long prev_jiffy; + + if (irqs_disabled()) + return; + + if (!IS_ENABLED(CONFIG_PREEMPT_COUNT)) + return; + + if (preempt_count() > preempt_offset) + return; + + if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy) + return; + prev_jiffy = jiffies; + + printk(KERN_ERR "BUG: assuming atomic context at %s:%d\n", file, line); + printk(KERN_ERR "in_atomic(): %d, irqs_disabled(): %d, pid: %d, name: %s\n", + in_atomic(), irqs_disabled(), + current->pid, current->comm); + + debug_show_held_locks(current); + dump_stack(); + add_taint(TAINT_WARN, LOCKDEP_STILL_OK); +} +EXPORT_SYMBOL_GPL(__cant_sleep); + +void ___might_sleep(const char *file, int line, int preempt_offset) +{ + /* Ratelimiting timestamp: */ + static unsigned long prev_jiffy; + + unsigned long preempt_disable_ip; + + /* WARN_ON_ONCE() by default, no rate limit required: */ + rcu_sleep_check(); + + if ((preempt_count_equals(preempt_offset) && !irqs_disabled() && + !is_idle_task(current) && !current->non_block_count) || + system_state == SYSTEM_BOOTING || system_state > SYSTEM_RUNNING || + oops_in_progress) + return; + + if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy) + return; + prev_jiffy = jiffies; + + /* Save this before calling printk(), since that will clobber it: */ + preempt_disable_ip = get_preempt_disable_ip(current); + + printk(KERN_ERR + "BUG: sleeping function called from invalid context at %s:%d\n", + file, line); + printk(KERN_ERR + "in_atomic(): %d, irqs_disabled(): %d, non_block: %d, pid: %d, name: %s\n", + in_atomic(), irqs_disabled(), current->non_block_count, + current->pid, current->comm); + + if (task_stack_end_corrupted(current)) + printk(KERN_EMERG "Thread overran stack, or stack corrupted\n"); + + debug_show_held_locks(current); + if (irqs_disabled()) + print_irqtrace_events(current); + if (IS_ENABLED(CONFIG_DEBUG_PREEMPT) + && !preempt_count_equals(preempt_offset)) { + pr_err("Preemption disabled at:"); + print_ip_sym(preempt_disable_ip); + pr_cont("\n"); + } + dump_stack(); + add_taint(TAINT_WARN, LOCKDEP_STILL_OK); +} +EXPORT_SYMBOL(___might_sleep); +#endif + +#ifdef CONFIG_MAGIC_SYSRQ +static inline void normalise_rt_tasks(void) +{ + struct sched_attr attr = {}; + struct task_struct *g, *p; + struct rq_flags rf; + struct rq *rq; + + read_lock(&tasklist_lock); + for_each_process_thread(g, p) { + /* + * Only normalize user tasks: + */ + if (p->flags & PF_KTHREAD) + continue; + + if (!rt_task(p) && !iso_task(p)) + continue; + + rq = task_rq_lock(p, &rf); + __setscheduler(p, rq, SCHED_NORMAL, 0, &attr, false); + task_rq_unlock(rq, p, &rf); + } + read_unlock(&tasklist_lock); +} + +void normalize_rt_tasks(void) +{ + normalise_rt_tasks(); +} +#endif /* CONFIG_MAGIC_SYSRQ */ + +#if defined(CONFIG_IA64) || defined(CONFIG_KGDB_KDB) +/* + * These functions are only useful for the IA64 MCA handling, or kdb. + * + * They can only be called when the whole system has been + * stopped - every CPU needs to be quiescent, and no scheduling + * activity can take place. Using them for anything else would + * be a serious bug, and as a result, they aren't even visible + * under any other configuration. + */ + +/** + * curr_task - return the current task for a given CPU. + * @cpu: the processor in question. + * + * ONLY VALID WHEN THE WHOLE SYSTEM IS STOPPED! + * + * Return: The current task for @cpu. + */ +struct task_struct *curr_task(int cpu) +{ + return cpu_curr(cpu); +} + +#endif /* defined(CONFIG_IA64) || defined(CONFIG_KGDB_KDB) */ + +#ifdef CONFIG_IA64 +/** + * ia64_set_curr_task - set the current task for a given CPU. + * @cpu: the processor in question. + * @p: the task pointer to set. + * + * Description: This function must only be used when non-maskable interrupts + * are serviced on a separate stack. It allows the architecture to switch the + * notion of the current task on a CPU in a non-blocking manner. This function + * must be called with all CPU's synchronised, and interrupts disabled, the + * and caller must save the original value of the current task (see + * curr_task() above) and restore that value before reenabling interrupts and + * re-starting the system. + * + * ONLY VALID WHEN THE WHOLE SYSTEM IS STOPPED! + */ +void ia64_set_curr_task(int cpu, struct task_struct *p) +{ + cpu_curr(cpu) = p; +} + +#endif + +void init_idle_bootup_task(struct task_struct *idle) +{} + +#ifdef CONFIG_SCHED_DEBUG +__read_mostly bool sched_debug_enabled; + +void proc_sched_show_task(struct task_struct *p, struct pid_namespace *ns, + struct seq_file *m) +{ + seq_printf(m, "%s (%d, #threads: %d)\n", p->comm, task_pid_nr_ns(p, ns), + get_nr_threads(p)); +} + +void proc_sched_set_task(struct task_struct *p) +{} +#endif + +#ifdef CONFIG_CGROUP_SCHED +static void sched_free_group(struct task_group *tg) +{ + kmem_cache_free(task_group_cache, tg); +} + +/* allocate runqueue etc for a new task group */ +struct task_group *sched_create_group(struct task_group *parent) +{ + struct task_group *tg; + + tg = kmem_cache_alloc(task_group_cache, GFP_KERNEL | __GFP_ZERO); + if (!tg) + return ERR_PTR(-ENOMEM); + + return tg; +} + +void sched_online_group(struct task_group *tg, struct task_group *parent) +{ +} + +/* rcu callback to free various structures associated with a task group */ +static void sched_free_group_rcu(struct rcu_head *rhp) +{ + /* Now it should be safe to free those cfs_rqs */ + sched_free_group(container_of(rhp, struct task_group, rcu)); +} + +void sched_destroy_group(struct task_group *tg) +{ + /* Wait for possible concurrent references to cfs_rqs complete */ + call_rcu(&tg->rcu, sched_free_group_rcu); +} + +void sched_offline_group(struct task_group *tg) +{ +} + +static inline struct task_group *css_tg(struct cgroup_subsys_state *css) +{ + return css ? container_of(css, struct task_group, css) : NULL; +} + +static struct cgroup_subsys_state * +cpu_cgroup_css_alloc(struct cgroup_subsys_state *parent_css) +{ + struct task_group *parent = css_tg(parent_css); + struct task_group *tg; + + if (!parent) { + /* This is early initialization for the top cgroup */ + return &root_task_group.css; + } + + tg = sched_create_group(parent); + if (IS_ERR(tg)) + return ERR_PTR(-ENOMEM); + return &tg->css; +} + +/* Expose task group only after completing cgroup initialization */ +static int cpu_cgroup_css_online(struct cgroup_subsys_state *css) +{ + struct task_group *tg = css_tg(css); + struct task_group *parent = css_tg(css->parent); + + if (parent) + sched_online_group(tg, parent); + return 0; +} + +static void cpu_cgroup_css_released(struct cgroup_subsys_state *css) +{ + struct task_group *tg = css_tg(css); + + sched_offline_group(tg); +} + +static void cpu_cgroup_css_free(struct cgroup_subsys_state *css) +{ + struct task_group *tg = css_tg(css); + + /* + * Relies on the RCU grace period between css_released() and this. + */ + sched_free_group(tg); +} + +static void cpu_cgroup_fork(struct task_struct *task) +{ +} + +static int cpu_cgroup_can_attach(struct cgroup_taskset *tset) +{ + return 0; +} + +static void cpu_cgroup_attach(struct cgroup_taskset *tset) +{ +} + +static struct cftype cpu_legacy_files[] = { + { } /* Terminate */ +}; + +static struct cftype cpu_files[] = { + { } /* terminate */ +}; + +static int cpu_extra_stat_show(struct seq_file *sf, + struct cgroup_subsys_state *css) +{ + return 0; +} + +struct cgroup_subsys cpu_cgrp_subsys = { + .css_alloc = cpu_cgroup_css_alloc, + .css_online = cpu_cgroup_css_online, + .css_released = cpu_cgroup_css_released, + .css_free = cpu_cgroup_css_free, + .css_extra_stat_show = cpu_extra_stat_show, + .fork = cpu_cgroup_fork, + .can_attach = cpu_cgroup_can_attach, + .attach = cpu_cgroup_attach, + .legacy_cftypes = cpu_files, + .legacy_cftypes = cpu_legacy_files, + .dfl_cftypes = cpu_files, + .early_init = true, + .threaded = true, +}; +#endif /* CONFIG_CGROUP_SCHED */ + +#undef CREATE_TRACE_POINTS diff --git a/kernel/sched/MuQSS.h b/kernel/sched/MuQSS.h new file mode 100644 index 000000000000..b34f2797e44f --- /dev/null +++ b/kernel/sched/MuQSS.h @@ -0,0 +1,1051 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef MUQSS_SCHED_H +#define MUQSS_SCHED_H + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#ifdef CONFIG_PARAVIRT +#include +#endif + +#include "cpupri.h" + +#ifdef CONFIG_SCHED_DEBUG +# define SCHED_WARN_ON(x) WARN_ONCE(x, #x) +#else +# define SCHED_WARN_ON(x) ((void)(x)) +#endif + +/* + * wake flags + */ +#define WF_SYNC 0x01 /* waker goes to sleep after wakeup */ +#define WF_FORK 0x02 /* child wakeup after fork */ +#define WF_MIGRATED 0x04 /* internal use, task got migrated */ + +/* task_struct::on_rq states: */ +#define TASK_ON_RQ_QUEUED 1 +#define TASK_ON_RQ_MIGRATING 2 + +struct rq; + +#ifdef CONFIG_SMP + +static inline bool sched_asym_prefer(int a, int b) +{ + return arch_asym_cpu_priority(a) > arch_asym_cpu_priority(b); +} + +struct perf_domain { + struct em_perf_domain *em_pd; + struct perf_domain *next; + struct rcu_head rcu; +}; + +/* Scheduling group status flags */ +#define SG_OVERLOAD 0x1 /* More than one runnable task on a CPU. */ +#define SG_OVERUTILIZED 0x2 /* One or more CPUs are over-utilized. */ + +/* + * We add the notion of a root-domain which will be used to define per-domain + * variables. Each exclusive cpuset essentially defines an island domain by + * fully partitioning the member cpus from any other cpuset. Whenever a new + * exclusive cpuset is created, we also create and attach a new root-domain + * object. + * + */ +struct root_domain { + atomic_t refcount; + atomic_t rto_count; + struct rcu_head rcu; + cpumask_var_t span; + cpumask_var_t online; + + /* + * Indicate pullable load on at least one CPU, e.g: + * - More than one runnable task + * - Running task is misfit + */ + int overload; + + /* Indicate one or more cpus over-utilized (tipping point) */ + int overutilized; + + /* + * The bit corresponding to a CPU gets set here if such CPU has more + * than one runnable -deadline task (as it is below for RT tasks). + */ + cpumask_var_t dlo_mask; + atomic_t dlo_count; + /* Replace unused CFS structures with void */ + //struct dl_bw dl_bw; + //struct cpudl cpudl; + void *dl_bw; + void *cpudl; + + /* + * The "RT overload" flag: it gets set if a CPU has more than + * one runnable RT task. + */ + cpumask_var_t rto_mask; + //struct cpupri cpupri; + void *cpupri; + + unsigned long max_cpu_capacity; + + /* + * NULL-terminated list of performance domains intersecting with the + * CPUs of the rd. Protected by RCU. + */ + struct perf_domain *pd; +}; + +extern void init_defrootdomain(void); +extern int sched_init_domains(const struct cpumask *cpu_map); +extern void rq_attach_root(struct rq *rq, struct root_domain *rd); + +static inline void cpupri_cleanup(void __maybe_unused *cpupri) +{ +} + +static inline void cpudl_cleanup(void __maybe_unused *cpudl) +{ +} + +static inline void init_dl_bw(void __maybe_unused *dl_bw) +{ +} + +static inline int cpudl_init(void __maybe_unused *dl_bw) +{ + return 0; +} + +static inline int cpupri_init(void __maybe_unused *cpupri) +{ + return 0; +} +#endif /* CONFIG_SMP */ + +/* + * This is the main, per-CPU runqueue data structure. + * This data should only be modified by the local cpu. + */ +struct rq { + raw_spinlock_t *lock; + raw_spinlock_t *orig_lock; + + struct task_struct __rcu *curr; + struct task_struct *idle; + struct task_struct *stop; + struct mm_struct *prev_mm; + + unsigned int nr_running; + /* + * This is part of a global counter where only the total sum + * over all CPUs matters. A task can increase this counter on + * one CPU and if it got migrated afterwards it may decrease + * it on another CPU. Always updated under the runqueue lock: + */ + unsigned long nr_uninterruptible; + u64 nr_switches; + + /* Stored data about rq->curr to work outside rq lock */ + u64 rq_deadline; + int rq_prio; + + /* Best queued id for use outside lock */ + u64 best_key; + + unsigned long last_scheduler_tick; /* Last jiffy this RQ ticked */ + unsigned long last_jiffy; /* Last jiffy this RQ updated rq clock */ + u64 niffies; /* Last time this RQ updated rq clock */ + u64 last_niffy; /* Last niffies as updated by local clock */ + u64 last_jiffy_niffies; /* Niffies @ last_jiffy */ + + u64 load_update; /* When we last updated load */ + unsigned long load_avg; /* Rolling load average */ +#ifdef CONFIG_HAVE_SCHED_AVG_IRQ + u64 irq_load_update; /* When we last updated IRQ load */ + unsigned long irq_load_avg; /* Rolling IRQ load average */ +#endif +#ifdef CONFIG_SMT_NICE + struct mm_struct *rq_mm; + int rq_smt_bias; /* Policy/nice level bias across smt siblings */ +#endif + /* Accurate timekeeping data */ + unsigned long user_ns, nice_ns, irq_ns, softirq_ns, system_ns, + iowait_ns, idle_ns; + atomic_t nr_iowait; + +#ifdef CONFIG_MEMBARRIER + int membarrier_state; +#endif + + skiplist_node *node; + skiplist *sl; +#ifdef CONFIG_SMP + struct task_struct *preempt; /* Preempt triggered on this task */ + struct task_struct *preempting; /* Hint only, what task is preempting */ + + int cpu; /* cpu of this runqueue */ + bool online; + + struct root_domain *rd; + struct sched_domain *sd; + + unsigned long cpu_capacity_orig; + + int *cpu_locality; /* CPU relative cache distance */ + struct rq **rq_order; /* Shared RQs ordered by relative cache distance */ + struct rq **cpu_order; /* RQs of discrete CPUs ordered by distance */ + + bool is_leader; + struct rq *smp_leader; /* First physical CPU per node */ +#ifdef CONFIG_SCHED_THERMAL_PRESSURE + struct sched_avg avg_thermal; +#endif /* CONFIG_SCHED_THERMAL_PRESSURE */ +#ifdef CONFIG_SCHED_SMT + struct rq *smt_leader; /* First logical CPU in SMT siblings */ + cpumask_t thread_mask; + bool (*siblings_idle)(struct rq *rq); + /* See if all smt siblings are idle */ +#endif /* CONFIG_SCHED_SMT */ +#ifdef CONFIG_SCHED_MC + struct rq *mc_leader; /* First logical CPU in MC siblings */ + cpumask_t core_mask; + bool (*cache_idle)(struct rq *rq); + /* See if all cache siblings are idle */ +#endif /* CONFIG_SCHED_MC */ +#endif /* CONFIG_SMP */ + +#ifdef CONFIG_IRQ_TIME_ACCOUNTING + u64 prev_irq_time; +#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ +#ifdef CONFIG_PARAVIRT + u64 prev_steal_time; +#endif /* CONFIG_PARAVIRT */ +#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING + u64 prev_steal_time_rq; +#endif /* CONFIG_PARAVIRT_TIME_ACCOUNTING */ + + u64 clock, old_clock, last_tick; + /* Ensure that all clocks are in the same cache line */ + u64 clock_task ____cacheline_aligned; + int dither; + + int iso_ticks; + bool iso_refractory; + +#ifdef CONFIG_HIGH_RES_TIMERS + struct hrtimer hrexpiry_timer; +#endif + + int rt_nr_running; /* Number real time tasks running */ +#ifdef CONFIG_SCHEDSTATS + + /* latency stats */ + struct sched_info rq_sched_info; + unsigned long long rq_cpu_time; + /* could above be rq->cfs_rq.exec_clock + rq->rt_rq.rt_runtime ? */ + + /* sys_sched_yield() stats */ + unsigned int yld_count; + + /* schedule() stats */ + unsigned int sched_switch; + unsigned int sched_count; + unsigned int sched_goidle; + + /* try_to_wake_up() stats */ + unsigned int ttwu_count; + unsigned int ttwu_local; +#endif /* CONFIG_SCHEDSTATS */ + +#ifdef CONFIG_SMP + struct llist_head wake_list; +#endif + +#ifdef CONFIG_CPU_IDLE + /* Must be inspected within a rcu lock section */ + struct cpuidle_state *idle_state; +#endif +}; + +static inline u64 __rq_clock_broken(struct rq *rq) +{ + return READ_ONCE(rq->clock); +} + +static inline u64 rq_clock(struct rq *rq) +{ + lockdep_assert_held(rq->lock); + + return rq->clock; +} + +static inline u64 rq_clock_task(struct rq *rq) +{ + lockdep_assert_held(rq->lock); + + return rq->clock_task; +} + +/** + * By default the decay is the default pelt decay period. + * The decay shift can change the decay period in + * multiples of 32. + * Decay shift Decay period(ms) + * 0 32 + * 1 64 + * 2 128 + * 3 256 + * 4 512 + */ +extern int sched_thermal_decay_shift; + +static inline u64 rq_clock_thermal(struct rq *rq) +{ + return rq_clock_task(rq) >> sched_thermal_decay_shift; +} + +struct rq_flags { + unsigned long flags; +}; + +#ifdef CONFIG_SMP +struct rq *cpu_rq(int cpu); +#endif + +#ifndef CONFIG_SMP +extern struct rq *uprq; +#define cpu_rq(cpu) (uprq) +#define this_rq() (uprq) +#define raw_rq() (uprq) +#define task_rq(p) (uprq) +#define cpu_curr(cpu) ((uprq)->curr) +#else /* CONFIG_SMP */ +DECLARE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); +#define this_rq() this_cpu_ptr(&runqueues) +#define raw_rq() raw_cpu_ptr(&runqueues) +#define task_rq(p) cpu_rq(task_cpu(p)) +#endif /* CONFIG_SMP */ + +static inline int task_current(struct rq *rq, struct task_struct *p) +{ + return rq->curr == p; +} + +static inline int task_running(struct rq *rq, struct task_struct *p) +{ +#ifdef CONFIG_SMP + return p->on_cpu; +#else + return task_current(rq, p); +#endif +} + +static inline int task_on_rq_queued(struct task_struct *p) +{ + return p->on_rq == TASK_ON_RQ_QUEUED; +} + +static inline int task_on_rq_migrating(struct task_struct *p) +{ + return READ_ONCE(p->on_rq) == TASK_ON_RQ_MIGRATING; +} + +static inline void rq_lock(struct rq *rq) + __acquires(rq->lock) +{ + raw_spin_lock(rq->lock); +} + +static inline void rq_unlock(struct rq *rq) + __releases(rq->lock) +{ + raw_spin_unlock(rq->lock); +} + +static inline void rq_lock_irq(struct rq *rq) + __acquires(rq->lock) +{ + raw_spin_lock_irq(rq->lock); +} + +static inline void rq_unlock_irq(struct rq *rq, struct rq_flags __always_unused *rf) + __releases(rq->lock) +{ + raw_spin_unlock_irq(rq->lock); +} + +static inline void rq_lock_irqsave(struct rq *rq, struct rq_flags *rf) + __acquires(rq->lock) +{ + raw_spin_lock_irqsave(rq->lock, rf->flags); +} + +static inline void rq_unlock_irqrestore(struct rq *rq, struct rq_flags *rf) + __releases(rq->lock) +{ + raw_spin_unlock_irqrestore(rq->lock, rf->flags); +} + +static inline struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf) + __acquires(p->pi_lock) + __acquires(rq->lock) +{ + struct rq *rq; + + while (42) { + raw_spin_lock_irqsave(&p->pi_lock, rf->flags); + rq = task_rq(p); + raw_spin_lock(rq->lock); + if (likely(rq == task_rq(p))) + break; + raw_spin_unlock(rq->lock); + raw_spin_unlock_irqrestore(&p->pi_lock, rf->flags); + } + return rq; +} + +static inline void task_rq_unlock(struct rq *rq, struct task_struct *p, struct rq_flags *rf) + __releases(rq->lock) + __releases(p->pi_lock) +{ + rq_unlock(rq); + raw_spin_unlock_irqrestore(&p->pi_lock, rf->flags); +} + +static inline struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags __always_unused *rf) + __acquires(rq->lock) +{ + struct rq *rq; + + lockdep_assert_held(&p->pi_lock); + + while (42) { + rq = task_rq(p); + raw_spin_lock(rq->lock); + if (likely(rq == task_rq(p))) + break; + raw_spin_unlock(rq->lock); + } + return rq; +} + +static inline void __task_rq_unlock(struct rq *rq, struct rq_flags __always_unused *rf) +{ + rq_unlock(rq); +} + +static inline struct rq * +this_rq_lock_irq(struct rq_flags *rf) + __acquires(rq->lock) +{ + struct rq *rq; + + local_irq_disable(); + rq = this_rq(); + rq_lock(rq); + return rq; +} + +/* + * {de,en}queue flags: Most not used on MuQSS. + * + * DEQUEUE_SLEEP - task is no longer runnable + * ENQUEUE_WAKEUP - task just became runnable + * + * SAVE/RESTORE - an otherwise spurious dequeue/enqueue, done to ensure tasks + * are in a known state which allows modification. Such pairs + * should preserve as much state as possible. + * + * MOVE - paired with SAVE/RESTORE, explicitly does not preserve the location + * in the runqueue. + * + * ENQUEUE_HEAD - place at front of runqueue (tail if not specified) + * ENQUEUE_REPLENISH - CBS (replenish runtime and postpone deadline) + * ENQUEUE_MIGRATED - the task was migrated during wakeup + * + */ + +#define DEQUEUE_SLEEP 0x01 +#define DEQUEUE_SAVE 0x02 /* matches ENQUEUE_RESTORE */ + +#define ENQUEUE_WAKEUP 0x01 +#define ENQUEUE_RESTORE 0x02 + +#ifdef CONFIG_SMP +#define ENQUEUE_MIGRATED 0x40 +#else +#define ENQUEUE_MIGRATED 0x00 +#endif + +#ifdef CONFIG_NUMA +enum numa_topology_type { + NUMA_DIRECT, + NUMA_GLUELESS_MESH, + NUMA_BACKPLANE, +}; +extern enum numa_topology_type sched_numa_topology_type; +extern int sched_max_numa_distance; +extern bool find_numa_distance(int distance); +extern void sched_init_numa(void); +extern void sched_domains_numa_masks_set(unsigned int cpu); +extern void sched_domains_numa_masks_clear(unsigned int cpu); +extern int sched_numa_find_closest(const struct cpumask *cpus, int cpu); +#else +static inline void sched_init_numa(void) { } +static inline void sched_domains_numa_masks_set(unsigned int cpu) { } +static inline void sched_domains_numa_masks_clear(unsigned int cpu) { } +static inline int sched_numa_find_closest(const struct cpumask *cpus, int cpu) +{ + return nr_cpu_ids; +} +#endif + +extern struct mutex sched_domains_mutex; +extern struct static_key_false sched_schedstats; + +#define rcu_dereference_check_sched_domain(p) \ + rcu_dereference_check((p), \ + lockdep_is_held(&sched_domains_mutex)) + +#ifdef CONFIG_SMP + +/* + * The domain tree (rq->sd) is protected by RCU's quiescent state transition. + * See destroy_sched_domains: call_rcu for details. + * + * The domain tree of any CPU may only be accessed from within + * preempt-disabled sections. + */ +#define for_each_domain(cpu, __sd) \ + for (__sd = rcu_dereference_check_sched_domain(cpu_rq(cpu)->sd); \ + __sd; __sd = __sd->parent) + +/** + * highest_flag_domain - Return highest sched_domain containing flag. + * @cpu: The cpu whose highest level of sched domain is to + * be returned. + * @flag: The flag to check for the highest sched_domain + * for the given cpu. + * + * Returns the highest sched_domain of a cpu which contains the given flag. + */ +static inline struct sched_domain *highest_flag_domain(int cpu, int flag) +{ + struct sched_domain *sd, *hsd = NULL; + + for_each_domain(cpu, sd) { + if (!(sd->flags & flag)) + break; + hsd = sd; + } + + return hsd; +} + +static inline struct sched_domain *lowest_flag_domain(int cpu, int flag) +{ + struct sched_domain *sd; + + for_each_domain(cpu, sd) { + if (sd->flags & flag) + break; + } + + return sd; +} + +DECLARE_PER_CPU(struct sched_domain *, sd_llc); +DECLARE_PER_CPU(int, sd_llc_size); +DECLARE_PER_CPU(int, sd_llc_id); +DECLARE_PER_CPU(struct sched_domain_shared *, sd_llc_shared); +DECLARE_PER_CPU(struct sched_domain *, sd_numa); +DECLARE_PER_CPU(struct sched_domain *, sd_asym_packing); +DECLARE_PER_CPU(struct sched_domain *, sd_asym_cpucapacity); + +struct sched_group_capacity { + atomic_t ref; + /* + * CPU capacity of this group, SCHED_CAPACITY_SCALE being max capacity + * for a single CPU. + */ + unsigned long capacity; + unsigned long min_capacity; /* Min per-CPU capacity in group */ + unsigned long max_capacity; /* Max per-CPU capacity in group */ + unsigned long next_update; + int imbalance; /* XXX unrelated to capacity but shared group state */ + +#ifdef CONFIG_SCHED_DEBUG + int id; +#endif + + unsigned long cpumask[0]; /* balance mask */ +}; + +struct sched_group { + struct sched_group *next; /* Must be a circular list */ + atomic_t ref; + + unsigned int group_weight; + struct sched_group_capacity *sgc; + int asym_prefer_cpu; /* cpu of highest priority in group */ + + /* + * The CPUs this group covers. + * + * NOTE: this field is variable length. (Allocated dynamically + * by attaching extra space to the end of the structure, + * depending on how many CPUs the kernel has booted up with) + */ + unsigned long cpumask[0]; +}; + +static inline struct cpumask *sched_group_span(struct sched_group *sg) +{ + return to_cpumask(sg->cpumask); +} + +/* + * See build_balance_mask(). + */ +static inline struct cpumask *group_balance_mask(struct sched_group *sg) +{ + return to_cpumask(sg->sgc->cpumask); +} + +/** + * group_first_cpu - Returns the first cpu in the cpumask of a sched_group. + * @group: The group whose first cpu is to be returned. + */ +static inline unsigned int group_first_cpu(struct sched_group *group) +{ + return cpumask_first(sched_group_span(group)); +} + + +#if defined(CONFIG_SCHED_DEBUG) && defined(CONFIG_SYSCTL) +void register_sched_domain_sysctl(void); +void dirty_sched_domain_sysctl(int cpu); +void unregister_sched_domain_sysctl(void); +#else +static inline void register_sched_domain_sysctl(void) +{ +} +static inline void dirty_sched_domain_sysctl(int cpu) +{ +} +static inline void unregister_sched_domain_sysctl(void) +{ +} +#endif + +extern void sched_ttwu_pending(void); +extern void set_cpus_allowed_common(struct task_struct *p, const struct cpumask *new_mask); +extern void set_rq_online (struct rq *rq); +extern void set_rq_offline(struct rq *rq); +extern bool sched_smp_initialized; + +static inline void update_group_capacity(struct sched_domain *sd, int cpu) +{ +} + +static inline void trigger_load_balance(struct rq *rq) +{ +} + +#define sched_feat(x) 0 + +#else /* CONFIG_SMP */ + +static inline void sched_ttwu_pending(void) { } + +#endif /* CONFIG_SMP */ + +#ifdef CONFIG_CPU_IDLE +static inline void idle_set_state(struct rq *rq, + struct cpuidle_state *idle_state) +{ + rq->idle_state = idle_state; +} + +static inline struct cpuidle_state *idle_get_state(struct rq *rq) +{ + SCHED_WARN_ON(!rcu_read_lock_held()); + return rq->idle_state; +} +#else +static inline void idle_set_state(struct rq *rq, + struct cpuidle_state *idle_state) +{ +} + +static inline struct cpuidle_state *idle_get_state(struct rq *rq) +{ + return NULL; +} +#endif + +#ifdef CONFIG_SCHED_DEBUG +extern bool sched_debug_enabled; +#endif + +extern void schedule_idle(void); + +#ifdef CONFIG_IRQ_TIME_ACCOUNTING +struct irqtime { + u64 total; + u64 tick_delta; + u64 irq_start_time; + struct u64_stats_sync sync; +}; + +DECLARE_PER_CPU(struct irqtime, cpu_irqtime); + +/* + * Returns the irqtime minus the softirq time computed by ksoftirqd. + * Otherwise ksoftirqd's sum_exec_runtime is substracted its own runtime + * and never move forward. + */ +static inline u64 irq_time_read(int cpu) +{ + struct irqtime *irqtime = &per_cpu(cpu_irqtime, cpu); + unsigned int seq; + u64 total; + + do { + seq = __u64_stats_fetch_begin(&irqtime->sync); + total = irqtime->total; + } while (__u64_stats_fetch_retry(&irqtime->sync, seq)); + + return total; +} +#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ + +static inline bool sched_stop_runnable(struct rq *rq) +{ + return rq->stop && task_on_rq_queued(rq->stop); +} + +#ifdef CONFIG_SMP +static inline int cpu_of(struct rq *rq) +{ + return rq->cpu; +} +#else /* CONFIG_SMP */ +static inline int cpu_of(struct rq *rq) +{ + return 0; +} +#endif + +#ifdef CONFIG_CPU_FREQ +DECLARE_PER_CPU(struct update_util_data *, cpufreq_update_util_data); + +static inline void cpufreq_trigger(struct rq *rq, unsigned int flags) +{ + struct update_util_data *data; + + data = rcu_dereference_sched(*per_cpu_ptr(&cpufreq_update_util_data, + cpu_of(rq))); + + if (data) + data->func(data, rq->niffies, flags); +} +#else +static inline void cpufreq_trigger(struct rq *rq, unsigned int flag) +{ +} +#endif /* CONFIG_CPU_FREQ */ + +static __always_inline +unsigned int uclamp_rq_util_with(struct rq __maybe_unused *rq, unsigned int util, + struct task_struct __maybe_unused *p) +{ + return util; +} + +#ifndef arch_scale_freq_tick +static __always_inline +void arch_scale_freq_tick(void) +{ +} +#endif + +#ifdef arch_scale_freq_capacity +#ifndef arch_scale_freq_invariant +#define arch_scale_freq_invariant() (true) +#endif +#else /* arch_scale_freq_capacity */ +#define arch_scale_freq_invariant() (false) +#endif + +#ifdef CONFIG_64BIT +static inline u64 read_sum_exec_runtime(struct task_struct *t) +{ + return tsk_seruntime(t); +} +#else +static inline u64 read_sum_exec_runtime(struct task_struct *t) +{ + struct rq_flags rf; + u64 ns; + struct rq *rq; + + rq = task_rq_lock(t, &rf); + ns = tsk_seruntime(t); + task_rq_unlock(rq, t, &rf); + + return ns; +} +#endif + +#ifndef arch_scale_freq_capacity +static __always_inline +unsigned long arch_scale_freq_capacity(int cpu) +{ + return SCHED_CAPACITY_SCALE; +} +#endif + +#ifdef CONFIG_NO_HZ_FULL +extern bool sched_can_stop_tick(struct rq *rq); +extern int __init sched_tick_offload_init(void); + +/* + * Tick may be needed by tasks in the runqueue depending on their policy and + * requirements. If tick is needed, lets send the target an IPI to kick it out of + * nohz mode if necessary. + */ +static inline void sched_update_tick_dependency(struct rq *rq) +{ + int cpu; + + if (!tick_nohz_full_enabled()) + return; + + cpu = cpu_of(rq); + + if (!tick_nohz_full_cpu(cpu)) + return; + + if (sched_can_stop_tick(rq)) + tick_nohz_dep_clear_cpu(cpu, TICK_DEP_BIT_SCHED); + else + tick_nohz_dep_set_cpu(cpu, TICK_DEP_BIT_SCHED); +} +#else +static inline int sched_tick_offload_init(void) { return 0; } +static inline void sched_update_tick_dependency(struct rq *rq) { } +#endif + +#define SCHED_FLAG_SUGOV 0x10000000 + +static inline bool rt_rq_is_runnable(struct rq *rt_rq) +{ + return rt_rq->rt_nr_running; +} + +/** + * enum schedutil_type - CPU utilization type + * @FREQUENCY_UTIL: Utilization used to select frequency + * @ENERGY_UTIL: Utilization used during energy calculation + * + * The utilization signals of all scheduling classes (CFS/RT/DL) and IRQ time + * need to be aggregated differently depending on the usage made of them. This + * enum is used within schedutil_freq_util() to differentiate the types of + * utilization expected by the callers, and adjust the aggregation accordingly. + */ +enum schedutil_type { + FREQUENCY_UTIL, + ENERGY_UTIL, +}; + +#ifdef CONFIG_CPU_FREQ_GOV_SCHEDUTIL + +unsigned long schedutil_cpu_util(int cpu, unsigned long util_cfs, + unsigned long max, enum schedutil_type type, + struct task_struct *p); + +static inline unsigned long cpu_bw_dl(struct rq *rq) +{ + return 0; +} + +static inline unsigned long cpu_util_dl(struct rq *rq) +{ + return 0; +} + +static inline unsigned long cpu_util_cfs(struct rq *rq) +{ + unsigned long ret = READ_ONCE(rq->load_avg); + + if (ret > SCHED_CAPACITY_SCALE) + ret = SCHED_CAPACITY_SCALE; + return ret; +} + +static inline unsigned long cpu_util_rt(struct rq *rq) +{ + unsigned long ret = READ_ONCE(rq->rt_nr_running); + + if (ret > SCHED_CAPACITY_SCALE) + ret = SCHED_CAPACITY_SCALE; + return ret; +} + +#ifdef CONFIG_HAVE_SCHED_AVG_IRQ +static inline unsigned long cpu_util_irq(struct rq *rq) +{ + unsigned long ret = READ_ONCE(rq->irq_load_avg); + + if (ret > SCHED_CAPACITY_SCALE) + ret = SCHED_CAPACITY_SCALE; + return ret; +} + +static inline +unsigned long scale_irq_capacity(unsigned long util, unsigned long irq, unsigned long max) +{ + util *= (max - irq); + util /= max; + + return util; + +} +#else +static inline unsigned long cpu_util_irq(struct rq *rq) +{ + return 0; +} + +static inline +unsigned long scale_irq_capacity(unsigned long util, unsigned long irq, unsigned long max) +{ + return util; +} +#endif +#endif + +#if defined(CONFIG_ENERGY_MODEL) && defined(CONFIG_CPU_FREQ_GOV_SCHEDUTIL) +#define perf_domain_span(pd) (to_cpumask(((pd)->em_pd->cpus))) + +DECLARE_STATIC_KEY_FALSE(sched_energy_present); + +static inline bool sched_energy_enabled(void) +{ + return static_branch_unlikely(&sched_energy_present); +} + +#else /* ! (CONFIG_ENERGY_MODEL && CONFIG_CPU_FREQ_GOV_SCHEDUTIL) */ + +#define perf_domain_span(pd) NULL +static inline bool sched_energy_enabled(void) { return false; } + +#endif /* CONFIG_ENERGY_MODEL && CONFIG_CPU_FREQ_GOV_SCHEDUTIL */ + +#ifdef CONFIG_MEMBARRIER +/* + * The scheduler provides memory barriers required by membarrier between: + * - prior user-space memory accesses and store to rq->membarrier_state, + * - store to rq->membarrier_state and following user-space memory accesses. + * In the same way it provides those guarantees around store to rq->curr. + */ +static inline void membarrier_switch_mm(struct rq *rq, + struct mm_struct *prev_mm, + struct mm_struct *next_mm) +{ + int membarrier_state; + + if (prev_mm == next_mm) + return; + + membarrier_state = atomic_read(&next_mm->membarrier_state); + if (READ_ONCE(rq->membarrier_state) == membarrier_state) + return; + + WRITE_ONCE(rq->membarrier_state, membarrier_state); +} +#else +static inline void membarrier_switch_mm(struct rq *rq, + struct mm_struct *prev_mm, + struct mm_struct *next_mm) +{ +} +#endif + +#ifdef CONFIG_SMP +static inline bool is_per_cpu_kthread(struct task_struct *p) +{ + if (!(p->flags & PF_KTHREAD)) + return false; + + if (p->nr_cpus_allowed != 1) + return false; + + return true; +} +#endif + +void swake_up_all_locked(struct swait_queue_head *q); +void __prepare_to_swait(struct swait_queue_head *q, struct swait_queue *wait); + +/* pelt.h compat CONFIG_SCHED_THERMAL_PRESSURE impossible with MUQSS */ +static inline int +update_thermal_load_avg(u64 now, struct rq *rq, u64 capacity) +{ + return 0; +} + +static inline u64 thermal_load_avg(struct rq *rq) +{ + return 0; +} + +#endif /* MUQSS_SCHED_H */ diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index 7fbaee24c824..15d274af9b1c 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -183,6 +183,12 @@ static unsigned int get_next_freq(struct sugov_policy *sg_policy, return cpufreq_driver_resolve_freq(policy, freq); } +#ifdef CONFIG_SCHED_MUQSS +#define rt_rq_runnable(rq_rt) rt_rq_is_runnable(rq) +#else +#define rt_rq_runnable(rq_rt) rt_rq_is_runnable(&rq->rt) +#endif + /* * This function computes an effective utilization for the given CPU, to be * used for frequency selection given the linear relation: f = u * f_max. @@ -211,7 +217,7 @@ unsigned long schedutil_cpu_util(int cpu, unsigned long util_cfs, struct rq *rq = cpu_rq(cpu); if (!IS_BUILTIN(CONFIG_UCLAMP_TASK) && - type == FREQUENCY_UTIL && rt_rq_is_runnable(&rq->rt)) { + type == FREQUENCY_UTIL && rt_rq_runnable(rq)) { return max; } @@ -656,7 +662,11 @@ static int sugov_kthread_create(struct sugov_policy *sg_policy) struct task_struct *thread; struct sched_attr attr = { .size = sizeof(struct sched_attr), +#ifdef CONFIG_SCHED_MUQSS + .sched_policy = SCHED_RR, +#else .sched_policy = SCHED_DEADLINE, +#endif .sched_flags = SCHED_FLAG_SUGOV, .sched_nice = 0, .sched_priority = 0, diff --git a/kernel/sched/cpupri.h b/kernel/sched/cpupri.h index efbb492bb94c..f0288c32ab17 100644 --- a/kernel/sched/cpupri.h +++ b/kernel/sched/cpupri.h @@ -17,6 +17,7 @@ struct cpupri { int *cpu_to_pri; }; +#ifndef CONFIG_SCHED_MUQSS #ifdef CONFIG_SMP int cpupri_find(struct cpupri *cp, struct task_struct *p, struct cpumask *lowest_mask); @@ -27,3 +28,4 @@ void cpupri_set(struct cpupri *cp, int cpu, int pri); int cpupri_init(struct cpupri *cp); void cpupri_cleanup(struct cpupri *cp); #endif +#endif diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c index ff9435dee1df..d7bd67204d65 100644 --- a/kernel/sched/cputime.c +++ b/kernel/sched/cputime.c @@ -266,26 +266,6 @@ static inline u64 account_other_time(u64 max) return accounted; } -#ifdef CONFIG_64BIT -static inline u64 read_sum_exec_runtime(struct task_struct *t) -{ - return t->se.sum_exec_runtime; -} -#else -static u64 read_sum_exec_runtime(struct task_struct *t) -{ - u64 ns; - struct rq_flags rf; - struct rq *rq; - - rq = task_rq_lock(t, &rf); - ns = t->se.sum_exec_runtime; - task_rq_unlock(rq, t, &rf); - - return ns; -} -#endif - /* * Accumulate raw cputime values of dead tasks (sig->[us]time) and live * tasks (sum on group iteration) belonging to @tsk's group. @@ -658,7 +638,7 @@ void cputime_adjust(struct task_cputime *curr, struct prev_cputime *prev, void task_cputime_adjusted(struct task_struct *p, u64 *ut, u64 *st) { struct task_cputime cputime = { - .sum_exec_runtime = p->se.sum_exec_runtime, + .sum_exec_runtime = tsk_seruntime(p), }; task_cputime(p, &cputime.utime, &cputime.stime); diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c index b743bf38f08f..c769795d726b 100644 --- a/kernel/sched/idle.c +++ b/kernel/sched/idle.c @@ -361,6 +361,7 @@ void cpu_startup_entry(enum cpuhp_state state) do_idle(); } +#ifndef CONFIG_SCHED_MUQSS /* * idle-task scheduling class. */ @@ -481,3 +482,4 @@ const struct sched_class idle_sched_class = { .switched_to = switched_to_idle, .update_curr = update_curr_idle, }; +#endif /* CONFIG_SCHED_MUQSS */ diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index db3a57675ccf..1f11cefe8d20 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -2,6 +2,19 @@ /* * Scheduler internal types and methods: */ +#ifdef CONFIG_SCHED_MUQSS +#include "MuQSS.h" + +/* Begin compatibility wrappers for MuQSS/CFS differences */ +#define rq_rt_nr_running(rq) ((rq)->rt_nr_running) +#define rq_h_nr_running(rq) ((rq)->nr_running) + +#else /* CONFIG_SCHED_MUQSS */ + +#define rq_rt_nr_running(rq) ((rq)->rt.rt_nr_running) +#define rq_h_nr_running(rq) ((rq)->cfs.h_nr_running) + + #include #include @@ -2546,3 +2559,25 @@ static inline bool is_per_cpu_kthread(struct task_struct *p) void swake_up_all_locked(struct swait_queue_head *q); void __prepare_to_swait(struct swait_queue_head *q, struct swait_queue *wait); + +/* MuQSS compatibility functions */ +#ifdef CONFIG_64BIT +static inline u64 read_sum_exec_runtime(struct task_struct *t) +{ + return t->se.sum_exec_runtime; +} +#else +static inline u64 read_sum_exec_runtime(struct task_struct *t) +{ + u64 ns; + struct rq_flags rf; + struct rq *rq; + + rq = task_rq_lock(t, &rf); + ns = t->se.sum_exec_runtime; + task_rq_unlock(rq, t, &rf); + + return ns; +} +#endif +#endif /* CONFIG_SCHED_MUQSS */ diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c index 8344757bba6e..d819af35a770 100644 --- a/kernel/sched/topology.c +++ b/kernel/sched/topology.c @@ -450,7 +450,11 @@ void rq_attach_root(struct rq *rq, struct root_domain *rd) struct root_domain *old_rd = NULL; unsigned long flags; +#ifdef CONFIG_SCHED_MUQSS + raw_spin_lock_irqsave(rq->lock, flags); +#else raw_spin_lock_irqsave(&rq->lock, flags); +#endif if (rq->rd) { old_rd = rq->rd; @@ -476,7 +480,11 @@ void rq_attach_root(struct rq *rq, struct root_domain *rd) if (cpumask_test_cpu(rq->cpu, cpu_active_mask)) set_rq_online(rq); +#ifdef CONFIG_SCHED_MUQSS + raw_spin_unlock_irqrestore(rq->lock, flags); +#else raw_spin_unlock_irqrestore(&rq->lock, flags); +#endif if (old_rd) call_rcu(&old_rd->rcu, free_rootdomain); diff --git a/kernel/skip_list.c b/kernel/skip_list.c new file mode 100644 index 000000000000..bf5c6e97e139 --- /dev/null +++ b/kernel/skip_list.c @@ -0,0 +1,148 @@ +/* + Copyright (C) 2011,2016 Con Kolivas. + + Code based on example originally by William Pugh. + +Skip Lists are a probabilistic alternative to balanced trees, as +described in the June 1990 issue of CACM and were invented by +William Pugh in 1987. + +A couple of comments about this implementation: +The routine randomLevel has been hard-coded to generate random +levels using p=0.25. It can be easily changed. + +The insertion routine has been implemented so as to use the +dirty hack described in the CACM paper: if a random level is +generated that is more than the current maximum level, the +current maximum level plus one is used instead. + +Levels start at zero and go up to MaxLevel (which is equal to +MaxNumberOfLevels-1). + +The routines defined in this file are: + +init: defines slnode + +new_skiplist: returns a new, empty list + +randomLevel: Returns a random level based on a u64 random seed passed to it. +In MuQSS, the "niffy" time is used for this purpose. + +insert(l,key, value): inserts the binding (key, value) into l. This operation +occurs in O(log n) time. + +delnode(slnode, l, node): deletes any binding of key from the l based on the +actual node value. This operation occurs in O(k) time where k is the +number of levels of the node in question (max 8). The original delete +function occurred in O(log n) time and involved a search. + +MuQSS Notes: In this implementation of skiplists, there are bidirectional +next/prev pointers and the insert function returns a pointer to the actual +node the value is stored. The key here is chosen by the scheduler so as to +sort tasks according to the priority list requirements and is no longer used +by the scheduler after insertion. The scheduler lookup, however, occurs in +O(1) time because it is always the first item in the level 0 linked list. +Since the task struct stores a copy of the node pointer upon skiplist_insert, +it can also remove it much faster than the original implementation with the +aid of prev<->next pointer manipulation and no searching. + +*/ + +#include +#include + +#define MaxNumberOfLevels 8 +#define MaxLevel (MaxNumberOfLevels - 1) + +void skiplist_init(skiplist_node *slnode) +{ + int i; + + slnode->key = 0xFFFFFFFFFFFFFFFF; + slnode->level = 0; + slnode->value = NULL; + for (i = 0; i < MaxNumberOfLevels; i++) + slnode->next[i] = slnode->prev[i] = slnode; +} + +skiplist *new_skiplist(skiplist_node *slnode) +{ + skiplist *l = kzalloc(sizeof(skiplist), GFP_ATOMIC); + + BUG_ON(!l); + l->header = slnode; + return l; +} + +void free_skiplist(skiplist *l) +{ + skiplist_node *p, *q; + + p = l->header; + do { + q = p->next[0]; + p->next[0]->prev[0] = q->prev[0]; + skiplist_node_init(p); + p = q; + } while (p != l->header); + kfree(l); +} + +void skiplist_node_init(skiplist_node *node) +{ + memset(node, 0, sizeof(skiplist_node)); +} + +static inline unsigned int randomLevel(const long unsigned int randseed) +{ + return find_first_bit(&randseed, MaxLevel) / 2; +} + +void skiplist_insert(skiplist *l, skiplist_node *node, keyType key, valueType value, unsigned int randseed) +{ + skiplist_node *update[MaxNumberOfLevels]; + skiplist_node *p, *q; + int k = l->level; + + p = l->header; + do { + while (q = p->next[k], q->key <= key) + p = q; + update[k] = p; + } while (--k >= 0); + + ++l->entries; + k = randomLevel(randseed); + if (k > l->level) { + k = ++l->level; + update[k] = l->header; + } + + node->level = k; + node->key = key; + node->value = value; + do { + p = update[k]; + node->next[k] = p->next[k]; + p->next[k] = node; + node->prev[k] = p; + node->next[k]->prev[k] = node; + } while (--k >= 0); +} + +void skiplist_delete(skiplist *l, skiplist_node *node) +{ + int k, m = node->level; + + for (k = 0; k <= m; k++) { + node->prev[k]->next[k] = node->next[k]; + node->next[k]->prev[k] = node->prev[k]; + } + skiplist_node_init(node); + if (m == l->level) { + while (l->header->next[m] == l->header && l->header->prev[m] == l->header && m > 0) + m--; + l->level = m; + } + l->entries--; +} diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 8a176d8727a3..7c58ed33e54b 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -130,8 +130,16 @@ static int __maybe_unused four = 4; static unsigned long zero_ul; static unsigned long one_ul = 1; static unsigned long long_max = LONG_MAX; -static int one_hundred = 100; -static int one_thousand = 1000; +static int __read_mostly one_hundred = 100; +static int __read_mostly one_thousand = 1000; +static int zero = 0; +static int one = 1; +#ifdef CONFIG_SCHED_MUQSS +extern int rr_interval; +extern int sched_interactive; +extern int sched_iso_cpu; +extern int sched_yield_type; +#endif #ifdef CONFIG_PRINTK static int ten_thousand = 10000; #endif @@ -288,7 +296,7 @@ static struct ctl_table sysctl_base_table[] = { { } }; -#ifdef CONFIG_SCHED_DEBUG +#if defined(CONFIG_SCHED_DEBUG) && !defined(CONFIG_SCHED_MUQSS) static int min_sched_granularity_ns = 100000; /* 100 usecs */ static int max_sched_granularity_ns = NSEC_PER_SEC; /* 1 second */ static int min_wakeup_granularity_ns; /* 0 usecs */ @@ -305,6 +313,7 @@ static int max_extfrag_threshold = 1000; #endif static struct ctl_table kern_table[] = { +#ifndef CONFIG_SCHED_MUQSS { .procname = "sched_child_runs_first", .data = &sysctl_sched_child_runs_first, @@ -486,6 +495,7 @@ static struct ctl_table kern_table[] = { .extra2 = SYSCTL_ONE, }, #endif +#endif /* !CONFIG_SCHED_MUQSS */ #ifdef CONFIG_PROVE_LOCKING { .procname = "prove_locking", @@ -1049,6 +1059,44 @@ static struct ctl_table kern_table[] = { .proc_handler = proc_dointvec, }, #endif +#ifdef CONFIG_SCHED_MUQSS + { + .procname = "rr_interval", + .data = &rr_interval, + .maxlen = sizeof (int), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .extra1 = &one, + .extra2 = &one_thousand, + }, + { + .procname = "interactive", + .data = &sched_interactive, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, + }, + { + .procname = "iso_cpu", + .data = &sched_iso_cpu, + .maxlen = sizeof (int), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one_hundred, + }, + { + .procname = "yield_type", + .data = &sched_yield_type, + .maxlen = sizeof (int), + .mode = 0644, + .proc_handler = &proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &two, + }, +#endif #if defined(CONFIG_S390) && defined(CONFIG_SMP) { .procname = "spin_retry", diff --git a/kernel/time/Kconfig b/kernel/time/Kconfig index fcc42353f125..60b4fd254c80 100644 --- a/kernel/time/Kconfig +++ b/kernel/time/Kconfig @@ -123,7 +123,7 @@ config CONTEXT_TRACKING config CONTEXT_TRACKING_FORCE bool "Force context tracking" - depends on CONTEXT_TRACKING + depends on CONTEXT_TRACKING && !SCHED_MUQSS default y if !NO_HZ_FULL help The major pre-requirement for full dynticks to work is to diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index f5490222e134..7a61971cca74 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -190,8 +190,13 @@ int clockevents_tick_resume(struct clock_event_device *dev) #ifdef CONFIG_GENERIC_CLOCKEVENTS_MIN_ADJUST +#ifdef CONFIG_SCHED_MUQSS +/* Limit min_delta to 100us */ +#define MIN_DELTA_LIMIT (NSEC_PER_SEC / 10000) +#else /* Limit min_delta to a jiffie */ #define MIN_DELTA_LIMIT (NSEC_PER_SEC / HZ) +#endif /** * clockevents_increase_min_delta - raise minimum delta of a clock event device diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 2fd3b3fa68bf..1202d7fe5d8e 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -236,7 +236,7 @@ static void task_sample_cputime(struct task_struct *p, u64 *samples) u64 stime, utime; task_cputime(p, &utime, &stime); - store_samples(samples, stime, utime, p->se.sum_exec_runtime); + store_samples(samples, stime, utime, tsk_seruntime(p)); } static void proc_sample_cputime_atomic(struct task_cputime_atomic *at, @@ -855,7 +855,7 @@ static void check_thread_timers(struct task_struct *tsk, soft = task_rlimit(tsk, RLIMIT_RTTIME); if (soft != RLIM_INFINITY) { /* Task RT timeout is accounted in jiffies. RTTIME is usec */ - unsigned long rttime = tsk->rt.timeout * (USEC_PER_SEC / HZ); + unsigned long rttime = tsk_rttimeout(tsk) * (USEC_PER_SEC / HZ); unsigned long hard = task_rlimit_max(tsk, RLIMIT_RTTIME); /* At the hard limit, send SIGKILL. No further action. */ diff --git a/kernel/time/timer.c b/kernel/time/timer.c index a5221abb4594..daa431373f55 100644 --- a/kernel/time/timer.c +++ b/kernel/time/timer.c @@ -1568,7 +1568,7 @@ static unsigned long __next_timer_interrupt(struct timer_base *base) * Check, if the next hrtimer event is before the next timer wheel * event: */ -static u64 cmp_next_hrtimer_event(u64 basem, u64 expires) +static u64 cmp_next_hrtimer_event(struct timer_base *base, u64 basem, u64 expires) { u64 nextevt = hrtimer_get_next_event(); @@ -1586,6 +1586,9 @@ static u64 cmp_next_hrtimer_event(u64 basem, u64 expires) if (nextevt <= basem) return basem; + if (nextevt < expires && nextevt - basem <= TICK_NSEC) + base->is_idle = false; + /* * Round up to the next jiffie. High resolution timers are * off, so the hrtimers are expired in the tick and we need to @@ -1655,7 +1658,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem) } raw_spin_unlock(&base->lock); - return cmp_next_hrtimer_event(basem, expires); + return cmp_next_hrtimer_event(base, basem, expires); } /** diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index b5e3496cf803..68930e7f4d28 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -1048,10 +1048,15 @@ static int trace_wakeup_test_thread(void *data) { /* Make this a -deadline thread */ static const struct sched_attr attr = { +#ifdef CONFIG_SCHED_MUQSS + /* No deadline on MuQSS, use RR */ + .sched_policy = SCHED_RR, +#else .sched_policy = SCHED_DEADLINE, .sched_runtime = 100000ULL, .sched_deadline = 10000000ULL, .sched_period = 10000000ULL +#endif }; struct wakeup_test_data *x = data; -- 2.25.1