ax-task 0.7.1

OS-independent IRQ-safe SMP task scheduling core
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173

//! Policy under the owning scheduler transaction.

use super::*;

impl TaskSystem {
    pub(in crate::sched::system::task_system) fn apply_owner_policy_update_locked(
        &self,
        remote: &CpuRemote,
        core: &Arc<ThreadCore>,
        sched: &mut ThreadSchedState,
        generation: u64,
    ) -> Result<OwnerPolicyApply, TaskError> {
        let owner = remote.owner();
        let donor = sched.pi.donors.first_entry();
        Self::validate_owner_policy_generation(sched, generation)?;

        let mut transaction = OwnerRqTxn::begin(self, remote);
        let owner_now_ns = transaction.clock().wall().as_nanos();
        if transaction.current().is_some() {
            let _settled = transaction.settle_current(0);
        }
        let rq_state = transaction.task_state(core.id(), &sched.placement);
        let fair_placement = match sched.policy.requested_policy() {
            SchedulePolicy::Fair { .. } => {
                let _source_entity = core
                    .sched()
                    .active_option(sched)
                    .map(|active| active.base_entity().clone())
                    .or_else(|| transaction.base_scheduling_entity(core.id()))
                    .unwrap_or_else(|| {
                        task_runtime::fatal_invariant(0x5251_1202, core.id().as_u64() as usize)
                    });
                Some(FairPolicyPlacement {
                    source_virtual_time: transaction.virtual_time(),
                    destination_virtual_time: transaction.virtual_time(),
                })
            }
            _ => None,
        };
        let mut active = match rq_state {
            OwnerRqTaskState::Current => transaction.detach_current_schedule(core.id()),
            OwnerRqTaskState::Queued { outgoing } => {
                let detached = transaction.reclassify_task(core.id());
                if !outgoing {
                    sched.placement.deactivate(owner);
                }
                detached.into_active()
            }
            OwnerRqTaskState::DelayedFair { .. } => transaction
                .take_delayed_fair_for_update(core.id())
                .into_active(),
            OwnerRqTaskState::Inactive => core.sched().take_active(sched),
        };
        Self::detach_owner_deadline_bandwidth_in_rq(core, sched, remote, &mut transaction);
        let commit = self
            .apply_policy_generation_locked(
                sched,
                &mut active,
                generation,
                owner_now_ns,
                fair_placement,
                PolicyApplication::from_rq_state(rq_state, owner_now_ns),
            )
            .unwrap_or_else(|_| {
                task_runtime::fatal_invariant(0x5251_1203, core.id().as_u64() as usize)
            })
            .unwrap_or_else(|| {
                task_runtime::fatal_invariant(0x5251_1204, core.id().as_u64() as usize)
            });
        let base_entity = active.base_entity().clone();
        let pi_update = self
            .resolved_pi_schedule_update(
                sched.policy.base,
                base_entity,
                donor,
                sched.policy.dispatch_generation,
            )
            .unwrap_or_else(|_| {
                task_runtime::fatal_invariant(0x5251_1206, core.id().as_u64() as usize)
            });
        active = apply_pi_schedule_update(sched, active, pi_update, owner_now_ns, fair_placement)
            .unwrap_or_else(|_| {
                task_runtime::fatal_invariant(0x5251_1207, core.id().as_u64() as usize)
            });
        let effective_policy = active.policy();
        let effective_entity = active.entity().clone();
        let enqueue = match rq_state {
            OwnerRqTaskState::Current => {
                Self::activate_deadline_bandwidth_locked(core, sched, &mut transaction, owner);
                let rt_quota_exempt = sched.is_pi_boosted_rt_owner_for(active.policy());
                let migration_capable = sched.affinity.affinity.is_migration_capable();
                let metadata = sched.rq_task_metadata().unwrap_or_else(|_| {
                    task_runtime::fatal_invariant(0x5251_1205, core.id().as_u64() as usize)
                });
                transaction.install_current_schedule(
                    core.id(),
                    active,
                    Arc::clone(core),
                    rt_quota_exempt,
                    migration_capable,
                    metadata,
                );
                dispatch::OwnerReadyEnqueue {
                    reschedule: Some(RescheduleKind::Immediate),
                    scheduler_deadline_refresh_required: false,
                }
            }
            OwnerRqTaskState::Queued { .. } => {
                core.sched().install_active(sched, active);
                self.link_owner_ready_thread_locked(
                    owner,
                    &mut transaction,
                    core,
                    sched,
                    EnqueueReason::PolicyChanged,
                )
            }
            OwnerRqTaskState::DelayedFair { .. } => {
                if active.entity().fair().is_some_and(|fair| fair.is_delayed()) {
                    let metadata = sched.rq_task_metadata().unwrap_or_else(|_| {
                        task_runtime::fatal_invariant(0x5251_1209, core.id().as_u64() as usize)
                    });
                    let queued = QueuedThread::new(
                        core.id(),
                        active,
                        Arc::clone(core),
                        false,
                        sched.affinity.affinity.is_migration_capable(),
                        metadata,
                    );
                    let _entity = transaction.restore_delayed_fair_after_update(queued);
                } else {
                    transaction.finish_detached_delayed_fair(
                        &mut active,
                        self.config.timing_granularity_ns(),
                    );
                    core.sched().install_active(sched, active);
                    sched.placement.finish_delayed_dequeue(owner);
                }
                dispatch::OwnerReadyEnqueue {
                    reschedule: None,
                    // Removing/reinserting a delayed node can move a Fair
                    // runtime boundary in either direction or remove it.
                    scheduler_deadline_refresh_required: true,
                }
            }
            OwnerRqTaskState::Inactive => {
                core.sched().install_active(sched, active);
                dispatch::OwnerReadyEnqueue {
                    reschedule: None,
                    scheduler_deadline_refresh_required: false,
                }
            }
        };
        // Publish policy while the same rq still serializes the applied
        // entity against rq-only schedule-out writers.
        core.publish_base_policy(sched.policy.base);
        core.publish_effective_schedule(effective_policy, &effective_entity);
        transaction.commit();
        // Linux starts rt_bandwidth when sched_setscheduler() re-enqueues an
        // RT entity. Current tasks are detached and reinstalled rather than
        // passing through the ordinary wake/enqueue completion path, so the
        // policy transaction owns the equivalent activation edge.
        let rt_period_started = rq_state.is_runnable()
            && self.activate_owner_rt_period_for_policy(owner, effective_policy);
        Ok(OwnerPolicyApply {
            commit,
            reschedule: enqueue.reschedule,
            scheduler_deadline_refresh_required: enqueue.scheduler_deadline_refresh_required,
            rt_period_started,
        })
    }
}