summaryrefslogtreecommitdiff
path: root/kernel/sched/sched.h
AgeCommit message (Collapse)Author
2023-10-10sched/headers: Remove comment referring to rq::cpu_load, since this has been ↵Colin Ian King
removed There is a comment that refers to cpu_load, however, this cpu_load was removed with: 55627e3cd22c ("sched/core: Remove rq->cpu_load[]") ... back in 2019. The comment does not make sense with respect to this removed array, so remove the comment. Signed-off-by: Colin Ian King <colin.i.king@gmail.com> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://lore.kernel.org/r/20231010155744.1381065-1-colin.i.king@gmail.com
2023-10-09sched/topology: Move the declaration of 'schedutil_gov' to kernel/sched/sched.hIngo Molnar
Move it out of the .c file into the shared scheduler-internal header file, to gain type-checking. Signed-off-by: Ingo Molnar <mingo@kernel.org> Cc: Shrikanth Hegde <sshegde@linux.vnet.ibm.com> Cc: Valentin Schneider <vschneid@redhat.com> Link: https://lore.kernel.org/r/20231009060037.170765-3-sshegde@linux.vnet.ibm.com
2023-10-09sched/topology: Consolidate and clean up access to a CPU's max compute capacityVincent Guittot
Remove the rq::cpu_capacity_orig field and use arch_scale_cpu_capacity() instead. The scheduler uses 3 methods to get access to a CPU's max compute capacity: - arch_scale_cpu_capacity(cpu) which is the default way to get a CPU's capacity. - cpu_capacity_orig field which is periodically updated with arch_scale_cpu_capacity(). - capacity_orig_of(cpu) which encapsulates rq->cpu_capacity_orig. There is no real need to save the value returned by arch_scale_cpu_capacity() in struct rq. arch_scale_cpu_capacity() returns: - either a per_cpu variable. - or a const value for systems which have only one capacity. Remove rq::cpu_capacity_orig and use arch_scale_cpu_capacity() everywhere. No functional changes. Some performance tests on Arm64: - small SMP device (hikey): no noticeable changes - HMP device (RB5): hackbench shows minor improvement (1-2%) - large smp (thx2): hackbench and tbench shows minor improvement (1%) Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com> Link: https://lore.kernel.org/r/20231009103621.374412-2-vincent.guittot@linaro.org
2023-10-09sched/rt: Change the type of 'sysctl_sched_rt_period' from 'unsigned int' to ↵Yajun Deng
'int' Doing this matches the natural type of 'int' based calculus in sched_rt_handler(), and also enables the adding in of a correct upper bounds check on the sysctl interface. [ mingo: Rewrote the changelog. ] Signed-off-by: Yajun Deng <yajun.deng@linux.dev> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://lore.kernel.org/r/20231008021538.3063250-1-yajun.deng@linux.dev
2023-09-29sched/deadline: Make dl_rq->pushable_dl_tasks update drive dl_rq->overloadedValentin Schneider
dl_rq->dl_nr_migratory is increased whenever a DL entity is enqueued and it has nr_cpus_allowed > 1. Unlike the pushable_dl_tasks tree, dl_rq->dl_nr_migratory includes a dl_rq's current task. This means a dl_rq can have a migratable current, N non-migratable queued tasks, and be flagged as overloaded and have its CPU set in the dlo_mask, despite having an empty pushable_tasks tree. Make an dl_rq's overload logic be driven by {enqueue,dequeue}_pushable_dl_task(), in other words make DL RQs only be flagged as overloaded if they have at least one runnable-but-not-current migratable task. o push_dl_task() is unaffected, as it is a no-op if there are no pushable tasks. o pull_dl_task() now no longer scans runqueues whose sole migratable task is their current one, which it can't do anything about anyway. It may also now pull tasks to a DL RQ with dl_nr_running > 1 if only its current task is migratable. Since dl_rq->dl_nr_migratory becomes unused, remove it. RT had the exact same mechanism (rt_rq->rt_nr_migratory) which was dropped in favour of relying on rt_rq->pushable_tasks, see: 612f769edd06 ("sched/rt: Make rt_rq->pushable_tasks updates drive rto_mask") Signed-off-by: Valentin Schneider <vschneid@redhat.com> Signed-off-by: Ingo Molnar <mingo@kernel.org> Acked-by: Juri Lelli <juri.lelli@redhat.com> Link: https://lore.kernel.org/r/20230928150251.463109-1-vschneid@redhat.com
2023-09-25sched/rt: Make rt_rq->pushable_tasks updates drive rto_maskValentin Schneider
Sebastian noted that the rto_push_work IRQ work can be queued for a CPU that has an empty pushable_tasks list, which means nothing useful will be done in the IPI other than queue the work for the next CPU on the rto_mask. rto_push_irq_work_func() only operates on tasks in the pushable_tasks list, but the conditions for that irq_work to be queued (and for a CPU to be added to the rto_mask) rely on rq_rt->nr_migratory instead. nr_migratory is increased whenever an RT task entity is enqueued and it has nr_cpus_allowed > 1. Unlike the pushable_tasks list, nr_migratory includes a rt_rq's current task. This means a rt_rq can have a migratible current, N non-migratible queued tasks, and be flagged as overloaded / have its CPU set in the rto_mask, despite having an empty pushable_tasks list. Make an rt_rq's overload logic be driven by {enqueue,dequeue}_pushable_task(). Since rt_rq->{rt_nr_migratory,rt_nr_total} become unused, remove them. Note that the case where the current task is pushed away to make way for a migration-disabled task remains unchanged: the migration-disabled task has to be in the pushable_tasks list in the first place, which means it has nr_cpus_allowed > 1. Reported-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Valentin Schneider <vschneid@redhat.com> Signed-off-by: Ingo Molnar <mingo@kernel.org> Tested-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Link: https://lore.kernel.org/r/20230811112044.3302588-1-vschneid@redhat.com
2023-09-24sched/fair: Make cfs_rq->throttled_csd_list available on !SMPJosh Don
This makes the following patch cleaner by avoiding extra CONFIG_SMP conditionals on the availability of rq->throttled_csd_list. Signed-off-by: Josh Don <joshdon@google.com> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://lore.kernel.org/r/20230922230535.296350-1-joshdon@google.com
2023-09-21sched/debug: Remove the /proc/sys/kernel/sched_child_runs_first sysctlSebastian Andrzej Siewior
The /proc/sys/kernel/sched_child_runs_first knob is no longer connected since: 5e963f2bd4654 ("sched/fair: Commit to EEVDF") Remove it. Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://lore.kernel.org/r/20230920130025.412071-2-bigeasy@linutronix.de
2023-09-19sched/fair: Rename check_preempt_curr() to wakeup_preempt()Ingo Molnar
The name is a bit opaque - make it clear that this is about wakeup preemption. Also rename the ->check_preempt_curr() methods similarly. Signed-off-by: Ingo Molnar <mingo@kernel.org> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
2023-09-18sched/headers: Remove duplicated includes in kernel/sched/sched.hGUO Zihua
Remove duplicated includes of linux/cgroup.h and linux/psi.h. Both of these includes are included regardless of the config and they are all protected by ifndef, so no point including them again. Signed-off-by: GUO Zihua <guozihua@huawei.com> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://lore.kernel.org/r/20230818015633.18370-1-guozihua@huawei.com
2023-09-18sched/fair: Ratelimit update to tg->load_avgAaron Lu
When using sysbench to benchmark Postgres in a single docker instance with sysbench's nr_threads set to nr_cpu, it is observed there are times update_cfs_group() and update_load_avg() shows noticeable overhead on a 2sockets/112core/224cpu Intel Sapphire Rapids(SPR): 13.75% 13.74% [kernel.vmlinux] [k] update_cfs_group 10.63% 10.04% [kernel.vmlinux] [k] update_load_avg Annotate shows the cycles are mostly spent on accessing tg->load_avg with update_load_avg() being the write side and update_cfs_group() being the read side. tg->load_avg is per task group and when different tasks of the same taskgroup running on different CPUs frequently access tg->load_avg, it can be heavily contended. E.g. when running postgres_sysbench on a 2sockets/112cores/224cpus Intel Sappire Rapids, during a 5s window, the wakeup number is 14millions and migration number is 11millions and with each migration, the task's load will transfer from src cfs_rq to target cfs_rq and each change involves an update to tg->load_avg. Since the workload can trigger as many wakeups and migrations, the access(both read and write) to tg->load_avg can be unbound. As a result, the two mentioned functions showed noticeable overhead. With netperf/nr_client=nr_cpu/UDP_RR, the problem is worse: during a 5s window, wakeup number is 21millions and migration number is 14millions; update_cfs_group() costs ~25% and update_load_avg() costs ~16%. Reduce the overhead by limiting updates to tg->load_avg to at most once per ms. The update frequency is a tradeoff between tracking accuracy and overhead. 1ms is chosen because PELT window is roughly 1ms and it delivered good results for the tests that I've done. After this change, the cost of accessing tg->load_avg is greatly reduced and performance improved. Detailed test results below. ============================== postgres_sysbench on SPR: 25% base: 42382±19.8% patch: 50174±9.5% (noise) 50% base: 67626±1.3% patch: 67365±3.1% (noise) 75% base: 100216±1.2% patch: 112470±0.1% +12.2% 100% base: 93671±0.4% patch: 113563±0.2% +21.2% ============================== hackbench on ICL: group=1 base: 114912±5.2% patch: 117857±2.5% (noise) group=4 base: 359902±1.6% patch: 361685±2.7% (noise) group=8 base: 461070±0.8% patch: 491713±0.3% +6.6% group=16 base: 309032±5.0% patch: 378337±1.3% +22.4% ============================= hackbench on SPR: group=1 base: 100768±2.9% patch: 103134±2.9% (noise) group=4 base: 413830±12.5% patch: 378660±16.6% (noise) group=8 base: 436124±0.6% patch: 490787±3.2% +12.5% group=16 base: 457730±3.2% patch: 680452±1.3% +48.8% ============================ netperf/udp_rr on ICL 25% base: 114413±0.1% patch: 115111±0.0% +0.6% 50% base: 86803±0.5% patch: 86611±0.0% (noise) 75% base: 35959±5.3% patch: 49801±0.6% +38.5% 100% base: 61951±6.4% patch: 70224±0.8% +13.4% =========================== netperf/udp_rr on SPR 25% base: 104954±1.3% patch: 107312±2.8% (noise) 50% base: 55394±4.6% patch: 54940±7.4% (noise) 75% base: 13779±3.1% patch: 36105±1.1% +162% 100% base: 9703±3.7% patch: 28011±0.2% +189% ============================================== netperf/tcp_stream on ICL (all in noise range) 25% base: 43092±0.1% patch: 42891±0.5% 50% base: 19278±14.9% patch: 22369±7.2% 75% base: 16822±3.0% patch: 17086±2.3% 100% base: 18216±0.6% patch: 18078±2.9% =============================================== netperf/tcp_stream on SPR (all in noise range) 25% base: 34491±0.3% patch: 34886±0.5% 50% base: 19278±14.9% patch: 22369±7.2% 75% base: 16822±3.0% patch: 17086±2.3% 100% base: 18216±0.6% patch: 18078±2.9% Reported-by: Nitin Tekchandani <nitin.tekchandani@intel.com> Suggested-by: Vincent Guittot <vincent.guittot@linaro.org> Signed-off-by: Aaron Lu <aaron.lu@intel.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org> Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Reviewed-by: David Vernet <void@manifault.com> Tested-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Tested-by: Swapnil Sapkal <Swapnil.Sapkal@amd.com> Link: https://lkml.kernel.org/r/20230912065808.2530-2-aaron.lu@intel.com
2023-09-13sched: Simplify set_user_nice()Peter Zijlstra
Use guards to reduce gotos and simplify control flow. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org>
2023-08-28Merge tag 'x86-cleanups-2023-08-28' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip Pull misc x86 cleanups from Ingo Molnar: "The following commit deserves special mention: 22dc02f81cddd Revert "sched/fair: Move unused stub functions to header" This is in x86/cleanups, because the revert is a re-application of a number of cleanups that got removed inadvertedly" [ This also effectively undoes the amd_check_microcode() microcode declaration change I had done in my microcode loader merge in commit 42a7f6e3ffe0 ("Merge tag 'x86_microcode_for_v6.6_rc1' [...]"). I picked the declaration change by Arnd from this branch instead, which put it in <asm/processor.h> instead of <asm/microcode.h> like I had done in my merge resolution - Linus ] * tag 'x86-cleanups-2023-08-28' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86/platform/uv: Refactor code using deprecated strncpy() interface to use strscpy() x86/hpet: Refactor code using deprecated strncpy() interface to use strscpy() x86/platform/uv: Refactor code using deprecated strcpy()/strncpy() interfaces to use strscpy() x86/qspinlock-paravirt: Fix missing-prototype warning x86/paravirt: Silence unused native_pv_lock_init() function warning x86/alternative: Add a __alt_reloc_selftest() prototype x86/purgatory: Include header for warn() declaration x86/asm: Avoid unneeded __div64_32 function definition Revert "sched/fair: Move unused stub functions to header" x86/apic: Hide unused safe_smp_processor_id() on 32-bit UP x86/cpu: Fix amd_check_microcode() declaration
2023-08-28Merge tag 'sched-core-2023-08-28' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip Pull scheduler updates from Ingo Molnar: - The biggest change is introduction of a new iteration of the SCHED_FAIR interactivity code: the EEVDF ("Earliest Eligible Virtual Deadline First") scheduler EEVDF too is a virtual-time scheduler, with two parameters (weight and relative deadline), compared to CFS that had weight only. It completely reworks the base scheduler: placement, preemption, picking -- everything LWN.net, as usual, has a terrific writeup about EEVDF: https://lwn.net/Articles/925371/ Preemption (both tick and wakeup) is driven by testing against a fresh pick. Because the tree is now effectively an interval tree, and the selection is no longer the 'leftmost' task, over-scheduling is less of a problem. A lot of the CFS heuristics are removed or replaced by more natural latency-space parameters & constructs In terms of expected performance regressions: we will and can fix everything where a 'good' workload misbehaves with the new scheduler, but EEVDF inevitably changes workload scheduling in a binary fashion, hopefully for the better in the overwhelming majority of cases, but in some cases it won't, especially in adversarial loads that got lucky with the previous code, such as some variants of hackbench. We are trying hard to err on the side of fixing all performance regressions, but we expect some inevitable post-release iterations of that process - Improve load-balancing on hybrid x86 systems: enable cluster scheduling (again) - Improve & fix bandwidth-scheduling on nohz systems - Improve bandwidth-throttling - Use lock guards to simplify and de-goto-ify control flow - Misc improvements, cleanups and fixes * tag 'sched-core-2023-08-28' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (43 commits) sched/eevdf/doc: Modify the documented knob to base_slice_ns as well sched/eevdf: Curb wakeup-preemption sched: Simplify sched_core_cpu_{starting,deactivate}() sched: Simplify try_steal_cookie() sched: Simplify sched_tick_remote() sched: Simplify sched_exec() sched: Simplify ttwu() sched: Simplify wake_up_if_idle() sched: Simplify: migrate_swap_stop() sched: Simplify sysctl_sched_uclamp_handler() sched: Simplify get_nohz_timer_target() sched/rt: sysctl_sched_rr_timeslice show default timeslice after reset sched/rt: Fix sysctl_sched_rr_timeslice intial value sched/fair: Block nohz tick_stop when cfs bandwidth in use sched, cgroup: Restore meaning to hierarchical_quota MAINTAINERS: Add Peter explicitly to the psi section sched/psi: Select KERNFS as needed sched/topology: Align group flags when removing degenerate domain sched/fair: remove util_est boosting sched/fair: Propagate enqueue flags into place_entity() ...
2023-08-14sched: Simplify wake_up_if_idle()Peter Zijlstra
Use guards to reduce gotos and simplify control flow. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Valentin Schneider <vschneid@redhat.com> Link: https://lore.kernel.org/r/20230801211812.032678917@infradead.org
2023-08-14sched: Simplify: migrate_swap_stop()Peter Zijlstra
Use guards to reduce gotos and simplify control flow. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Valentin Schneider <vschneid@redhat.com> Link: https://lore.kernel.org/r/20230801211811.964370836@infradead.org
2023-08-10Merge branch 'sched/eevdf' into sched/coreIngo Molnar
Pick up the EEVDF work into the main branch - it's looking good so far. Conflicts: kernel/sched/features.h Signed-off-by: Ingo Molnar <mingo@kernel.org>
2023-08-02sched/fair: Block nohz tick_stop when cfs bandwidth in usePhil Auld
CFS bandwidth limits and NOHZ full don't play well together. Tasks can easily run well past their quotas before a remote tick does accounting. This leads to long, multi-period stalls before such tasks can run again. Currently, when presented with these conflicting requirements the scheduler is favoring nohz_full and letting the tick be stopped. However, nohz tick stopping is already best-effort, there are a number of conditions that can prevent it, whereas cfs runtime bandwidth is expected to be enforced. Make the scheduler favor bandwidth over stopping the tick by setting TICK_DEP_BIT_SCHED when the only running task is a cfs task with runtime limit enabled. We use cfs_b->hierarchical_quota to determine if the task requires the tick. Add check in pick_next_task_fair() as well since that is where we have a handle on the task that is actually going to be running. Add check in sched_can_stop_tick() to cover some edge cases such as nr_running going from 2->1 and the 1 remains the running task. Reviewed-By: Ben Segall <bsegall@google.com> Signed-off-by: Phil Auld <pauld@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lore.kernel.org/r/20230712133357.381137-3-pauld@redhat.com
2023-08-02sched, cgroup: Restore meaning to hierarchical_quotaPhil Auld
In cgroupv2 cfs_b->hierarchical_quota is set to -1 for all task groups due to the previous fix simply taking the min. It should reflect a limit imposed at that level or by an ancestor. Even though cgroupv2 does not require child quota to be less than or equal to that of its ancestors the task group will still be constrained by such a quota so this should be shown here. Cgroupv1 continues to set this correctly. In both cases, add initialization when a new task group is created based on the current parent's value (or RUNTIME_INF in the case of root_task_group). Otherwise, the field is wrong until a quota is changed after creation and __cfs_schedulable() is called. Fixes: c53593e5cb69 ("sched, cgroup: Don't reject lower cpu.max on ancestors") Signed-off-by: Phil Auld <pauld@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Ben Segall <bsegall@google.com> Acked-by: Tejun Heo <tj@kernel.org> Link: https://lore.kernel.org/r/20230714125746.812891-1-pauld@redhat.com
2023-07-31Revert "sched/fair: Move unused stub functions to header"Peter Zijlstra
Revert commit 7aa55f2a5902 ("sched/fair: Move unused stub functions to header"), for while it has the right Changelog, the actual patch content a revert of the previous 4 patches: f7df852ad6db ("sched: Make task_vruntime_update() prototype visible") c0bdfd72fbfb ("sched/fair: Hide unused init_cfs_bandwidth() stub") 378be384e01f ("sched: Add schedule_user() declaration") d55ebae3f312 ("sched: Hide unused sched_update_scaling()") So in effect this is a revert of a revert and re-applies those patches. Fixes: 7aa55f2a5902 ("sched/fair: Move unused stub functions to header") Reported-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
2023-07-19sched/fair: Propagate enqueue flags into place_entity()Peter Zijlstra
This allows place_entity() to consider ENQUEUE_WAKEUP and ENQUEUE_MIGRATED. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://lore.kernel.org/r/20230531124604.274010996@infradead.org
2023-07-19sched/debug: Rename sysctl_sched_min_granularity to sysctl_sched_base_slicePeter Zijlstra
EEVDF uses this tunable as the base request/slice -- make sure the name reflects this. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://lore.kernel.org/r/20230531124604.205287511@infradead.org
2023-07-19sched/fair: Commit to EEVDFPeter Zijlstra
EEVDF is a better defined scheduling policy, as a result it has less heuristics/tunables. There is no compelling reason to keep CFS around. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://lore.kernel.org/r/20230531124604.137187212@infradead.org
2023-07-19sched/fair: Implement an EEVDF-like scheduling policyPeter Zijlstra
Where CFS is currently a WFQ based scheduler with only a single knob, the weight. The addition of a second, latency oriented parameter, makes something like WF2Q or EEVDF based a much better fit. Specifically, EEVDF does EDF like scheduling in the left half of the tree -- those entities that are owed service. Except because this is a virtual time scheduler, the deadlines are in virtual time as well, which is what allows over-subscription. EEVDF has two parameters: - weight, or time-slope: which is mapped to nice just as before - request size, or slice length: which is used to compute the virtual deadline as: vd_i = ve_i + r_i/w_i Basically, by setting a smaller slice, the deadline will be earlier and the task will be more eligible and ran earlier. Tick driven preemption is driven by request/slice completion; while wakeup preemption is driven by the deadline. Because the tree is now effectively an interval tree, and the selection is no longer 'leftmost', over-scheduling is less of a problem. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://lore.kernel.org/r/20230531124603.931005524@infradead.org
2023-07-19sched/fair: Add cfs_rq::avg_vruntimePeter Zijlstra
In order to move to an eligibility based scheduling policy, we need to have a better approximation of the ideal scheduler. Specifically, for a virtual time weighted fair queueing based scheduler the ideal scheduler will be the weighted average of the individual virtual runtimes (math in the comment). As such, compute the weighted average to approximate the ideal scheduler -- note that the approximation is in the individual task behaviour, which isn't strictly conformant. Specifically consider adding a task with a vruntime left of center, in this case the average will move backwards in time -- something the ideal scheduler would of course never do. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://lore.kernel.org/r/20230531124603.654144274@infradead.org
2023-07-17sched: add WF_CURRENT_CPU and externise ttwuPeter Oskolkov
Add WF_CURRENT_CPU wake flag that advices the scheduler to move the wakee to the current CPU. This is useful for fast on-CPU context switching use cases. In addition, make ttwu external rather than static so that the flag could be passed to it from outside of sched/core.c. Signed-off-by: Peter Oskolkov <posk@google.com> Signed-off-by: Andrei Vagin <avagin@google.com> Acked-by: "Peter Zijlstra (Intel)" <peterz@infradead.org> Link: https://lore.kernel.org/r/20230308073201.3102738-3-avagin@google.com Signed-off-by: Kees Cook <keescook@chromium.org>
2023-07-13sched/topology: Record number of cores in sched groupTim C Chen
When balancing sibling domains that have different number of cores, tasks in respective sibling domain should be proportional to the number of cores in each domain. In preparation of implementing such a policy, record the number of cores in a scheduling group. Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lore.kernel.org/r/04641eeb0e95c21224352f5743ecb93dfac44654.1688770494.git.tim.c.chen@linux.intel.com
2023-07-13sched: add throttled time stat for throttled childrenJosh Don
We currently export the total throttled time for cgroups that are given a bandwidth limit. This patch extends this accounting to also account the total time that each children cgroup has been throttled. This is useful to understand the degree to which children have been affected by the throttling control. Children which are not runnable during the entire throttled period, for example, will not show any self-throttling time during this period. Expose this in a new interface, 'cpu.stat.local', which is similar to how non-hierarchical events are accounted in 'memory.events.local'. Signed-off-by: Josh Don <joshdon@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by: Tejun Heo <tj@kernel.org> Link: https://lore.kernel.org/r/20230620183247.737942-2-joshdon@google.com
2023-06-27Merge tag 'cgroup-for-6.5' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup Pull cgroup updates from Tejun Heo: - Whenever cpuset needs to rebuild sched_domain, it walked all tasks looking for DEADLINE tasks as they need to be accounted on the new domain. Walking all tasks can be expensive and there may not be any DEADLINE tasks at all. Task iteration is now omitted if there are no DEADLINE tasks - Fixes DEADLINE bandwidth misaccounting after task migration failures - When no controller is enabled, -Wstringop-overflow warning is triggered. The fix patch added an early exit which is too eager and got reverted for now. Will fix later - Everything else is minor cleanups * tag 'cgroup-for-6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup: Revert "cgroup: Avoid -Wstringop-overflow warnings" cgroup/misc: Expose misc.current on cgroup v2 root cgroup: Avoid -Wstringop-overflow warnings cgroup: remove obsolete comment on cgroup_on_dfl() cgroup: remove unused task_cgroup_path() cgroup/cpuset: remove unneeded header files cgroup: make cgroup_is_threaded() and cgroup_is_thread_root() static rdmacg: fix kernel-doc warnings in rdmacg cgroup: Replace the css_set call with cgroup_get cgroup: remove unused macro for_each_e_css() cgroup: Update out-of-date comment in cgroup_migrate() cgroup: Replace all non-returning strlcpy with strscpy cgroup/cpuset: remove unneeded header files cgroup/cpuset: Free DL BW in case can_attach() fails sched/deadline: Create DL BW alloc, free & check overflow interface cgroup/cpuset: Iterate only if DEADLINE tasks are present sched/cpuset: Keep track of SCHED_DEADLINE task in cpusets sched/cpuset: Bring back cpuset_mutex cgroup/cpuset: Rename functions dealing with DEADLINE accounting
2023-06-16sched/core: Avoid multiple calling update_rq_clock() in __cfsb_csd_unthrottle()Hao Jia
After commit 8ad075c2eb1f ("sched: Async unthrottling for cfs bandwidth"), we may update the rq clock multiple times in the loop of __cfsb_csd_unthrottle(). A prior (although less common) instance of this problem exists in unthrottle_offline_cfs_rqs(). Cure both by ensuring update_rq_clock() is called before the loop and setting RQCF_ACT_SKIP during the loop, to supress further updates. The alternative would be pulling update_rq_clock() out of unthrottle_cfs_rq(), but that gives an even bigger mess. Fixes: 8ad075c2eb1f ("sched: Async unthrottling for cfs bandwidth") Reviewed-By: Ben Segall <bsegall@google.com> Suggested-by: Vincent Guittot <vincent.guittot@linaro.org> Signed-off-by: Hao Jia <jiahao.os@bytedance.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org> Link: https://lkml.kernel.org/r/20230613082012.49615-4-jiahao.os@bytedance.com
2023-06-16sched/deadline: Fix bandwidth reclaim equation in GRUBVineeth Pillai
According to the GRUB[1] rule, the runtime is depreciated as: "dq = -max{u, (1 - Uinact - Uextra)} dt" (1) To guarantee that deadline tasks doesn't starve lower class tasks, we do not allocate the full bandwidth of the cpu to deadline tasks. Maximum bandwidth usable by deadline tasks is denoted by "Umax". Considering Umax, equation (1) becomes: "dq = -(max{u, (Umax - Uinact - Uextra)} / Umax) dt" (2) Current implementation has a minor bug in equation (2), which this patch fixes. The reclamation logic is verified by a sample program which creates multiple deadline threads and observing their utilization. The tests were run on an isolated cpu(isolcpus=3) on a 4 cpu system. Tests on 6.3.0 ============== RUN 1: runtime=7ms, deadline=period=10ms, RT capacity = 95% TID[693]: RECLAIM=1, (r=7ms, d=10ms, p=10ms), Util: 93.33 TID[693]: RECLAIM=1, (r=7ms, d=10ms, p=10ms), Util: 93.35 RUN 2: runtime=1ms, deadline=period=100ms, RT capacity = 95% TID[708]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 16.69 TID[708]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 16.69 RUN 3: 2 tasks Task 1: runtime=1ms, deadline=period=10ms Task 2: runtime=1ms, deadline=period=100ms TID[631]: RECLAIM=1, (r=1ms, d=10ms, p=10ms), Util: 62.67 TID[632]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 6.37 TID[631]: RECLAIM=1, (r=1ms, d=10ms, p=10ms), Util: 62.38 TID[632]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 6.23 As seen above, the reclamation doesn't reclaim the maximum allowed bandwidth and as the bandwidth of tasks gets smaller, the reclaimed bandwidth also comes down. Tests with this patch applied ============================= RUN 1: runtime=7ms, deadline=period=10ms, RT capacity = 95% TID[608]: RECLAIM=1, (r=7ms, d=10ms, p=10ms), Util: 95.19 TID[608]: RECLAIM=1, (r=7ms, d=10ms, p=10ms), Util: 95.16 RUN 2: runtime=1ms, deadline=period=100ms, RT capacity = 95% TID[616]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 95.27 TID[616]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 95.21 RUN 3: 2 tasks Task 1: runtime=1ms, deadline=period=10ms Task 2: runtime=1ms, deadline=period=100ms TID[620]: RECLAIM=1, (r=1ms, d=10ms, p=10ms), Util: 86.64 TID[621]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 8.66 TID[620]: RECLAIM=1, (r=1ms, d=10ms, p=10ms), Util: 86.45 TID[621]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 8.73 Running tasks on all cpus allowing for migration also showed that the utilization is reclaimed to the maximum. Running 10 tasks on 3 cpus SCHED_FLAG_RECLAIM - top shows: %Cpu0 : 94.6 us, 0.0 sy, 0.0 ni, 5.4 id, 0.0 wa %Cpu1 : 95.2 us, 0.0 sy, 0.0 ni, 4.8 id, 0.0 wa %Cpu2 : 95.8 us, 0.0 sy, 0.0 ni, 4.2 id, 0.0 wa [1]: Abeni, Luca & Lipari, Giuseppe & Parri, Andrea & Sun, Youcheng. (2015). Parallel and sequential reclaiming in multicore real-time global scheduling. Signed-off-by: Vineeth Pillai (Google) <vineeth@bitbyteword.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Daniel Bristot de Oliveira <bristot@kernel.org> Acked-by: Juri Lelli <juri.lelli@redhat.com> Link: https://lore.kernel.org/r/20230530135526.2385378-1-vineeth@bitbyteword.org
2023-06-05sched/fair, cpufreq: Introduce 'runnable boosting'Dietmar Eggemann
The responsiveness of the Per Entity Load Tracking (PELT) util_avg in mobile devices is still considered too low for utilization changes during task ramp-up. In Android this manifests in the fact that the first frames of a UI activity are very prone to be jankframes (a frame which doesn't meet the required frame rendering time, e.g. 16ms@60Hz) since the CPU frequency is normally low at this point and has to ramp up quickly. The beginning of an UI activity is also characterized by the occurrence of CPU contention, especially on little CPUs. Current little CPUs can have an original CPU capacity of only ~ 150 which means that the actual CPU capacity at lower frequency can even be much smaller. Schedutil maps CPU util_avg into CPU frequency request via: util = effective_cpu_util(..., cpu_util_cfs(cpu), ...) -> util = map_util_perf(util) -> freq = map_util_freq(util, ...) CPU contention for CFS tasks can be detected by 'CPU runnable > CPU utililization' in cpu_util_cfs_boost() -> cpu_util(..., boost = 1). Schedutil uses 'runnable boosting' by calling cpu_util_cfs_boost(). To be in sync with schedutil's CPU frequency selection, Energy Aware Scheduling (EAS) also calls cpu_util(..., boost = 1) during max util detection. Moreover, 'runnable boosting' is also used in load-balance for busiest CPU selection when the migration type is 'migrate_util', i.e. only at sched domains which don't have the SD_SHARE_PKG_RESOURCES flag set. Suggested-by: Vincent Guittot <vincent.guittot@linaro.org> Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org> Link: https://lore.kernel.org/r/20230515115735.296329-3-dietmar.eggemann@arm.com
2023-06-05sched/fair: Refactor CPU utilization functionsDietmar Eggemann
There is a lot of code duplication in cpu_util_next() & cpu_util_cfs(). Remove this by allowing cpu_util_next() to be called with p = NULL. Rename cpu_util_next() to cpu_util() since the '_next' suffix is no longer necessary to distinct cpu utilization related functions. Implement cpu_util_cfs(cpu) as cpu_util(cpu, p = NULL, -1). This will allow to code future related cpu util changes only in one place, namely in cpu_util(). Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org> Link: https://lore.kernel.org/r/20230515115735.296329-2-dietmar.eggemann@arm.com
2023-05-30sched/deadline: remove unused dl_bandwidthMiaohe Lin
The default deadline bandwidth control structure has been removed since commit eb77cf1c151c ("sched/deadline: Remove unused def_dl_bandwidth") leading to unused init_dl_bandwidth() and struct dl_bandwidth. Remove them to clean up the code. Signed-off-by: Miaohe Lin <linmiaohe@huawei.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by: Juri Lelli <juri.lelli@redhat.com> Link: https://lore.kernel.org/r/20230524102514.407486-1-linmiaohe@huawei.com
2023-05-30sched/fair: Move unused stub functions to headerArnd Bergmann
These four functions have a normal definition for CONFIG_FAIR_GROUP_SCHED, and empty one that is only referenced when FAIR_GROUP_SCHED is disabled but CGROUP_SCHED is still enabled. If both are turned off, the functions are still defined but the misisng prototype causes a W=1 warning: kernel/sched/fair.c:12544:6: error: no previous prototype for 'free_fair_sched_group' kernel/sched/fair.c:12546:5: error: no previous prototype for 'alloc_fair_sched_group' kernel/sched/fair.c:12553:6: error: no previous prototype for 'online_fair_sched_group' kernel/sched/fair.c:12555:6: error: no previous prototype for 'unregister_fair_sched_group' Move the alternatives into the header as static inline functions with the correct combination of #ifdef checks to avoid the warning without adding even more complexity. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org> Link: https://lore.kernel.org/r/20230522195021.3456768-6-arnd@kernel.org
2023-05-30sched: Make task_vruntime_update() prototype visibleArnd Bergmann
Having the prototype next to the caller but not visible to the callee causes a W=1 warning: kernel/sched/fair.c:11985:6: error: no previous prototype for 'task_vruntime_update' [-Werror=missing-prototypes] Move this to a header, as we do for all other function declarations. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org> Link: https://lore.kernel.org/r/20230522195021.3456768-5-arnd@kernel.org
2023-05-30sched: Add schedule_user() declarationArnd Bergmann
The schedule_user() function is used on powerpc and sparc architectures, but only ever called from assembler, so it has no prototype, causing a harmless W=1 warning: kernel/sched/core.c:6730:35: error: no previous prototype for 'schedule_user' [-Werror=missing-prototypes] Add a prototype in sched/sched.h to shut up the warning. Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org> Link: https://lore.kernel.org/r/20230522195021.3456768-3-arnd@kernel.org
2023-05-08sched/deadline: Create DL BW alloc, free & check overflow interfaceDietmar Eggemann
While moving a set of tasks between exclusive cpusets, cpuset_can_attach() -> task_can_attach() calls dl_cpu_busy(..., p) for DL BW overflow checking and per-task DL BW allocation on the destination root_domain for the DL tasks in this set. This approach has the issue of not freeing already allocated DL BW in the following error cases: (1) The set of tasks includes multiple DL tasks and DL BW overflow checking fails for one of the subsequent DL tasks. (2) Another controller next to the cpuset controller which is attached to the same cgroup fails in its can_attach(). To address this problem rework dl_cpu_busy(): (1) Split it into dl_bw_check_overflow() & dl_bw_alloc() and add a dedicated dl_bw_free(). (2) dl_bw_alloc() & dl_bw_free() take a `u64 dl_bw` parameter instead of a `struct task_struct *p` used in dl_cpu_busy(). This allows to allocate DL BW for a set of tasks too rather than only for a single task. Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com> Signed-off-by: Juri Lelli <juri.lelli@redhat.com> Signed-off-by: Tejun Heo <tj@kernel.org>
2023-05-08sched/topology: Check SDF_SHARED_CHILD in highest_flag_domain()Ricardo Neri
Do not assume that all the children of a scheduling domain have a given flag. Check whether it has the SDF_SHARED_CHILD meta flag. Suggested-by: Ionela Voinescu <ionela.voinescu@arm.com> Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lore.kernel.org/r/20230406203148.19182-9-ricardo.neri-calderon@linux.intel.com
2023-04-21sched: Fix performance regression introduced by mm_cidMathieu Desnoyers
Introduce per-mm/cpu current concurrency id (mm_cid) to fix a PostgreSQL sysbench regression reported by Aaron Lu. Keep track of the currently allocated mm_cid for each mm/cpu rather than freeing them immediately on context switch. This eliminates most atomic operations when context switching back and forth between threads belonging to different memory spaces in multi-threaded scenarios (many processes, each with many threads). The per-mm/per-cpu mm_cid values are serialized by their respective runqueue locks. Thread migration is handled by introducing invocation to sched_mm_cid_migrate_to() (with destination runqueue lock held) in activate_task() for migrating tasks. If the destination cpu's mm_cid is unset, and if the source runqueue is not actively using its mm_cid, then the source cpu's mm_cid is moved to the destination cpu on migration. Introduce a task-work executed periodically, similarly to NUMA work, which delays reclaim of cid values when they are unused for a period of time. Keep track of the allocation time for each per-cpu cid, and let the task work clear them when they are observed to be older than SCHED_MM_CID_PERIOD_NS and unused. This task work also clears all mm_cids which are greater or equal to the Hamming weight of the mm cidmask to keep concurrency ids compact. Because we want to ensure the mm_cid converges towards the smaller values as migrations happen, the prior optimization that was done when context switching between threads belonging to the same mm is removed, because it could delay the lazy release of the destination runqueue mm_cid after it has been replaced by a migration. Removing this prior optimization is not an issue performance-wise because the introduced per-mm/per-cpu mm_cid tracking also covers this more specific case. Fixes: af7f588d8f73 ("sched: Introduce per-memory-map concurrency ID") Reported-by: Aaron Lu <aaron.lu@intel.com> Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Tested-by: Aaron Lu <aaron.lu@intel.com> Link: https://lore.kernel.org/lkml/20230327080502.GA570847@ziqianlu-desk2/
2023-03-22sched/core: Avoid selecting the task that is throttled to run when ↵Hao Jia
core-sched enable When {rt, cfs}_rq or dl task is throttled, since cookied tasks are not dequeued from the core tree, So sched_core_find() and sched_core_next() may return throttled task, which may cause throttled task to run on the CPU. So we add checks in sched_core_find() and sched_core_next() to make sure that the return is a runnable task that is not throttled. Co-developed-by: Cruz Zhao <CruzZhao@linux.alibaba.com> Signed-off-by: Cruz Zhao <CruzZhao@linux.alibaba.com> Signed-off-by: Hao Jia <jiahao.os@bytedance.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20230316081806.69544-1-jiahao.os@bytedance.com
2023-02-11sched/fair: Remove capacity inversion detectionVincent Guittot
Remove the capacity inversion detection which is now handled by util_fits_cpu() returning -1 when we need to continue to look for a potential CPU with better performance. This ends up almost reverting patches below except for some comments: commit da07d2f9c153 ("sched/fair: Fixes for capacity inversion detection") commit aa69c36f31aa ("sched/fair: Consider capacity inversion in util_fits_cpu()") commit 44c7b80bffc3 ("sched/fair: Detect capacity inversion") Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lore.kernel.org/r/20230201143628.270912-3-vincent.guittot@linaro.org
2022-12-27sched: Make const-safeMatthew Wilcox (Oracle)
With a modified container_of() that preserves constness, the compiler finds some pointers which should have been marked as const. task_of() also needs to become const-preserving for the !FAIR_GROUP_SCHED case so that cfs_rq_of() can take a const argument. No change to generated code. Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20221212144946.2657785-1-willy@infradead.org
2022-12-27sched: Introduce per-memory-map concurrency IDMathieu Desnoyers
This feature allows the scheduler to expose a per-memory map concurrency ID to user-space. This concurrency ID is within the possible cpus range, and is temporarily (and uniquely) assigned while threads are actively running within a memory map. If a memory map has fewer threads than cores, or is limited to run on few cores concurrently through sched affinity or cgroup cpusets, the concurrency IDs will be values close to 0, thus allowing efficient use of user-space memory for per-cpu data structures. This feature is meant to be exposed by a new rseq thread area field. The primary purpose of this feature is to do the heavy-lifting needed by memory allocators to allow them to use per-cpu data structures efficiently in the following situations: - Single-threaded applications, - Multi-threaded applications on large systems (many cores) with limited cpu affinity mask, - Multi-threaded applications on large systems (many cores) with restricted cgroup cpuset per container. One of the key concern from scheduler maintainers is the overhead associated with additional spin locks or atomic operations in the scheduler fast-path. This is why the following optimization is implemented. On context switch between threads belonging to the same memory map, transfer the mm_cid from prev to next without any atomic ops. This takes care of use-cases involving frequent context switch between threads belonging to the same memory map. Additional optimizations can be done if the spin locks added when context switching between threads belonging to different memory maps end up being a performance bottleneck. Those are left out of this patch though. A performance impact would have to be clearly demonstrated to justify the added complexity. The credit goes to Paul Turner (Google) for the original virtual cpu id idea. This feature is implemented based on the discussions with Paul Turner and Peter Oskolkov (Google), but I took the liberty to implement scheduler fast-path optimizations and my own NUMA-awareness scheme. The rumor has it that Google have been running a rseq vcpu_id extension internally in production for a year. The tcmalloc source code indeed has comments hinting at a vcpu_id prototype extension to the rseq system call [1]. The following benchmarks do not show any significant overhead added to the scheduler context switch by this feature: * perf bench sched messaging (process) Baseline: 86.5±0.3 ms With mm_cid: 86.7±2.6 ms * perf bench sched messaging (threaded) Baseline: 84.3±3.0 ms With mm_cid: 84.7±2.6 ms * hackbench (process) Baseline: 82.9±2.7 ms With mm_cid: 82.9±2.9 ms * hackbench (threaded) Baseline: 85.2±2.6 ms With mm_cid: 84.4±2.9 ms [1] https://github.com/google/tcmalloc/blob/master/tcmalloc/internal/linux_syscall_support.h#L26 Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lore.kernel.org/r/20221122203932.231377-8-mathieu.desnoyers@efficios.com
2022-12-27sched: Async unthrottling for cfs bandwidthJosh Don
CFS bandwidth currently distributes new runtime and unthrottles cfs_rq's inline in an hrtimer callback. Runtime distribution is a per-cpu operation, and unthrottling is a per-cgroup operation, since a tg walk is required. On machines with a large number of cpus and large cgroup hierarchies, this cpus*cgroups work can be too much to do in a single hrtimer callback: since IRQ are disabled, hard lockups may easily occur. Specifically, we've found this scalability issue on configurations with 256 cpus, O(1000) cgroups in the hierarchy being throttled, and high memory bandwidth usage. To fix this, we can instead unthrottle cfs_rq's asynchronously via a CSD. Each cpu is responsible for unthrottling itself, thus sharding the total work more fairly across the system, and avoiding hard lockups. Signed-off-by: Josh Don <joshdon@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20221117005418.3499691-1-joshdon@google.com
2022-10-27sched: Enforce user requested affinityWaiman Long
It was found that the user requested affinity via sched_setaffinity() can be easily overwritten by other kernel subsystems without an easy way to reset it back to what the user requested. For example, any change to the current cpuset hierarchy may reset the cpumask of the tasks in the affected cpusets to the default cpuset value even if those tasks have pre-existing user requested affinity. That is especially easy to trigger under a cgroup v2 environment where writing "+cpuset" to the root cgroup's cgroup.subtree_control file will reset the cpus affinity of all the processes in the system. That is problematic in a nohz_full environment where the tasks running in the nohz_full CPUs usually have their cpus affinity explicitly set and will behave incorrectly if cpus affinity changes. Fix this problem by looking at user_cpus_ptr in __set_cpus_allowed_ptr() and use it to restrcit the given cpumask unless there is no overlap. In that case, it will fallback to the given one. The SCA_USER flag is reused to indicate intent to set user_cpus_ptr and so user_cpus_ptr masking should be skipped. In addition, masking should also be skipped if any of the SCA_MIGRATE_* flag is set. All callers of set_cpus_allowed_ptr() will be affected by this change. A scratch cpumask is added to percpu runqueues structure for doing additional masking when user_cpus_ptr is set. Signed-off-by: Waiman Long <longman@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20220922180041.1768141-4-longman@redhat.com
2022-10-27sched: Always preserve the user requested cpumaskWaiman Long
Unconditionally preserve the user requested cpumask on sched_setaffinity() calls. This allows using it outside of the fairly narrow restrict_cpus_allowed_ptr() use-case and fix some cpuset issues that currently suffer destruction of cpumasks. Signed-off-by: Waiman Long <longman@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20220922180041.1768141-3-longman@redhat.com
2022-10-27sched: Introduce affinity_contextWaiman Long
In order to prepare for passing through additional data through the affinity call-chains, convert the mask and flags argument into a structure. Suggested-by: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Waiman Long <longman@redhat.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20220922180041.1768141-5-longman@redhat.com
2022-10-27sched/fair: Detect capacity inversionQais Yousef
Check each performance domain to see if thermal pressure is causing its capacity to be lower than another performance domain. We assume that each performance domain has CPUs with the same capacities, which is similar to an assumption made in energy_model.c We also assume that thermal pressure impacts all CPUs in a performance domain equally. If there're multiple performance domains with the same capacity_orig, we will trigger a capacity inversion if the domain is under thermal pressure. The new cpu_in_capacity_inversion() should help users to know when information about capacity_orig are not reliable and can opt in to use the inverted capacity as the 'actual' capacity_orig. Signed-off-by: Qais Yousef <qais.yousef@arm.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lore.kernel.org/r/20220804143609.515789-9-qais.yousef@arm.com
2022-10-27sched/uclamp: Fix fits_capacity() check in feec()Qais Yousef
As reported by Yun Hsiang [1], if a task has its uclamp_min >= 0.8 * 1024, it'll always pick the previous CPU because fits_capacity() will always return false in this case. The new util_fits_cpu() logic should handle this correctly for us beside more corner cases where similar failures could occur, like when using UCLAMP_MAX. We open code uclamp_rq_util_with() except for the clamp() part, util_fits_cpu() needs the 'raw' values to be passed to it. Also introduce uclamp_rq_{set, get}() shorthand accessors to get uclamp value for the rq. Makes the code more readable and ensures the right rules (use READ_ONCE/WRITE_ONCE) are respected transparently. [1] https://lists.linaro.org/pipermail/eas-dev/2020-July/001488.html Fixes: 1d42509e475c ("sched/fair: Make EAS wakeup placement consider uclamp restrictions") Reported-by: Yun Hsiang <hsiang023167@gmail.com> Signed-off-by: Qais Yousef <qais.yousef@arm.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lore.kernel.org/r/20220804143609.515789-4-qais.yousef@arm.com