diff options
| -rw-r--r-- | kernel/sched/ext.c | 466 | ||||
| -rw-r--r-- | kernel/sched/ext_idle.c | 208 | ||||
| -rw-r--r-- | kernel/sched/ext_idle.h | 3 | ||||
| -rw-r--r-- | kernel/sched/sched.h | 4 | ||||
| -rw-r--r-- | tools/sched_ext/include/scx/common.bpf.h | 2 | ||||
| -rw-r--r-- | tools/sched_ext/scx_show_state.py | 14 | ||||
| -rw-r--r-- | tools/testing/selftests/sched_ext/Makefile | 1 | ||||
| -rw-r--r-- | tools/testing/selftests/sched_ext/allowed_cpus.bpf.c | 121 | ||||
| -rw-r--r-- | tools/testing/selftests/sched_ext/allowed_cpus.c | 57 |
9 files changed, 584 insertions, 292 deletions
diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index fdbf249d1c68..3cc5d7b5e3a9 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -26,7 +26,7 @@ enum scx_consts { * Iterating all tasks may take a while. Periodically drop * scx_tasks_lock to avoid causing e.g. CSD and RCU stalls. */ - SCX_OPS_TASK_ITER_BATCH = 32, + SCX_TASK_ITER_BATCH = 32, }; enum scx_exit_kind { @@ -44,9 +44,9 @@ enum scx_exit_kind { }; /* - * An exit code can be specified when exiting with scx_bpf_exit() or - * scx_ops_exit(), corresponding to exit_kind UNREG_BPF and UNREG_KERN - * respectively. The codes are 64bit of the format: + * An exit code can be specified when exiting with scx_bpf_exit() or scx_exit(), + * corresponding to exit_kind UNREG_BPF and UNREG_KERN respectively. The codes + * are 64bit of the format: * * Bits: [63 .. 48 47 .. 32 31 .. 0] * [ SYS ACT ] [ SYS RSN ] [ USR ] @@ -465,6 +465,7 @@ struct sched_ext_ops { * idle CPU tracking and the following helpers become unavailable: * * - scx_bpf_select_cpu_dfl() + * - scx_bpf_select_cpu_and() * - scx_bpf_test_and_clear_cpu_idle() * - scx_bpf_pick_idle_cpu() * @@ -838,18 +839,18 @@ enum scx_tg_flags { SCX_TG_INITED = 1U << 1, }; -enum scx_ops_enable_state { - SCX_OPS_ENABLING, - SCX_OPS_ENABLED, - SCX_OPS_DISABLING, - SCX_OPS_DISABLED, +enum scx_enable_state { + SCX_ENABLING, + SCX_ENABLED, + SCX_DISABLING, + SCX_DISABLED, }; -static const char *scx_ops_enable_state_str[] = { - [SCX_OPS_ENABLING] = "enabling", - [SCX_OPS_ENABLED] = "enabled", - [SCX_OPS_DISABLING] = "disabling", - [SCX_OPS_DISABLED] = "disabled", +static const char *scx_enable_state_str[] = { + [SCX_ENABLING] = "enabling", + [SCX_ENABLED] = "enabled", + [SCX_DISABLING] = "disabling", + [SCX_DISABLED] = "disabled", }; /* @@ -907,15 +908,15 @@ static DEFINE_SPINLOCK(scx_tasks_lock); static LIST_HEAD(scx_tasks); /* ops enable/disable */ -static struct kthread_worker *scx_ops_helper; -static DEFINE_MUTEX(scx_ops_enable_mutex); -DEFINE_STATIC_KEY_FALSE(__scx_ops_enabled); +static struct kthread_worker *scx_helper; +static DEFINE_MUTEX(scx_enable_mutex); +DEFINE_STATIC_KEY_FALSE(__scx_enabled); DEFINE_STATIC_PERCPU_RWSEM(scx_fork_rwsem); -static atomic_t scx_ops_enable_state_var = ATOMIC_INIT(SCX_OPS_DISABLED); +static atomic_t scx_enable_state_var = ATOMIC_INIT(SCX_DISABLED); static unsigned long scx_in_softlockup; -static atomic_t scx_ops_breather_depth = ATOMIC_INIT(0); -static int scx_ops_bypass_depth; -static bool scx_ops_init_task_enabled; +static atomic_t scx_breather_depth = ATOMIC_INIT(0); +static int scx_bypass_depth; +static bool scx_init_task_enabled; static bool scx_switching_all; DEFINE_STATIC_KEY_FALSE(__scx_switched_all); @@ -947,7 +948,7 @@ static atomic_long_t scx_enable_seq = ATOMIC_LONG_INIT(0); /* * The maximum amount of time in jiffies that a task may be runnable without * being scheduled on a CPU. If this timeout is exceeded, it will trigger - * scx_ops_error(). + * scx_error(). */ static unsigned long scx_watchdog_timeout; @@ -1043,18 +1044,17 @@ static struct kobject *scx_root_kobj; static void process_ddsp_deferred_locals(struct rq *rq); static void scx_bpf_kick_cpu(s32 cpu, u64 flags); -static __printf(3, 4) void scx_ops_exit_kind(enum scx_exit_kind kind, - s64 exit_code, - const char *fmt, ...); +static __printf(3, 4) void __scx_exit(enum scx_exit_kind kind, s64 exit_code, + const char *fmt, ...); -#define scx_ops_error_kind(err, fmt, args...) \ - scx_ops_exit_kind((err), 0, fmt, ##args) +#define __scx_error(err, fmt, args...) \ + __scx_exit((err), 0, fmt, ##args) -#define scx_ops_exit(code, fmt, args...) \ - scx_ops_exit_kind(SCX_EXIT_UNREG_KERN, (code), fmt, ##args) +#define scx_exit(code, fmt, args...) \ + __scx_exit(SCX_EXIT_UNREG_KERN, (code), fmt, ##args) -#define scx_ops_error(fmt, args...) \ - scx_ops_error_kind(SCX_EXIT_ERROR, fmt, ##args) +#define scx_error(fmt, args...) \ + __scx_error(SCX_EXIT_ERROR, fmt, ##args) #define SCX_HAS_OP(op) static_branch_likely(&scx_has_op[SCX_OP_IDX(op)]) @@ -1187,8 +1187,8 @@ do { \ static __always_inline bool scx_kf_allowed(u32 mask) { if (unlikely(!(current->scx.kf_mask & mask))) { - scx_ops_error("kfunc with mask 0x%x called from an operation only allowing 0x%x", - mask, current->scx.kf_mask); + scx_error("kfunc with mask 0x%x called from an operation only allowing 0x%x", + mask, current->scx.kf_mask); return false; } @@ -1201,13 +1201,13 @@ static __always_inline bool scx_kf_allowed(u32 mask) */ if (unlikely(highest_bit(mask) == SCX_KF_CPU_RELEASE && (current->scx.kf_mask & higher_bits(SCX_KF_CPU_RELEASE)))) { - scx_ops_error("cpu_release kfunc called from a nested operation"); + scx_error("cpu_release kfunc called from a nested operation"); return false; } if (unlikely(highest_bit(mask) == SCX_KF_DISPATCH && (current->scx.kf_mask & higher_bits(SCX_KF_DISPATCH)))) { - scx_ops_error("dispatch kfunc called from a nested operation"); + scx_error("dispatch kfunc called from a nested operation"); return false; } @@ -1223,7 +1223,7 @@ static __always_inline bool scx_kf_allowed_on_arg_tasks(u32 mask, if (unlikely((p != current->scx.kf_tasks[0] && p != current->scx.kf_tasks[1]))) { - scx_ops_error("called on a task not being operated on"); + scx_error("called on a task not being operated on"); return false; } @@ -1402,15 +1402,15 @@ static void scx_task_iter_stop(struct scx_task_iter *iter) * @iter: iterator to walk * * Visit the next task. See scx_task_iter_start() for details. Locks are dropped - * and re-acquired every %SCX_OPS_TASK_ITER_BATCH iterations to avoid causing - * stalls by holding scx_tasks_lock for too long. + * and re-acquired every %SCX_TASK_ITER_BATCH iterations to avoid causing stalls + * by holding scx_tasks_lock for too long. */ static struct task_struct *scx_task_iter_next(struct scx_task_iter *iter) { struct list_head *cursor = &iter->cursor.tasks_node; struct sched_ext_entity *pos; - if (!(++iter->cnt % SCX_OPS_TASK_ITER_BATCH)) { + if (!(++iter->cnt % SCX_TASK_ITER_BATCH)) { scx_task_iter_unlock(iter); cond_resched(); scx_task_iter_relock(iter); @@ -1592,23 +1592,22 @@ static DEFINE_PER_CPU(struct scx_event_stats, event_stats_cpu); static void scx_bpf_events(struct scx_event_stats *events, size_t events__sz); -static enum scx_ops_enable_state scx_ops_enable_state(void) +static enum scx_enable_state scx_enable_state(void) { - return atomic_read(&scx_ops_enable_state_var); + return atomic_read(&scx_enable_state_var); } -static enum scx_ops_enable_state -scx_ops_set_enable_state(enum scx_ops_enable_state to) +static enum scx_enable_state scx_set_enable_state(enum scx_enable_state to) { - return atomic_xchg(&scx_ops_enable_state_var, to); + return atomic_xchg(&scx_enable_state_var, to); } -static bool scx_ops_tryset_enable_state(enum scx_ops_enable_state to, - enum scx_ops_enable_state from) +static bool scx_tryset_enable_state(enum scx_enable_state to, + enum scx_enable_state from) { int from_v = from; - return atomic_try_cmpxchg(&scx_ops_enable_state_var, &from_v, to); + return atomic_try_cmpxchg(&scx_enable_state_var, &from_v, to); } static bool scx_rq_bypassing(struct rq *rq) @@ -1647,8 +1646,7 @@ static bool ops_cpu_valid(s32 cpu, const char *where) if (likely(cpu >= 0 && cpu < nr_cpu_ids && cpu_possible(cpu))) { return true; } else { - scx_ops_error("invalid CPU %d%s%s", cpu, - where ? " " : "", where ?: ""); + scx_error("invalid CPU %d%s%s", cpu, where ? " " : "", where ?: ""); return false; } } @@ -1658,7 +1656,7 @@ static bool ops_cpu_valid(s32 cpu, const char *where) * @ops_name: operation to blame on failure * @err: -errno value to sanitize * - * Verify @err is a valid -errno. If not, trigger scx_ops_error() and return + * Verify @err is a valid -errno. If not, trigger scx_error() and return * -%EPROTO. This is necessary because returning a rogue -errno up the chain can * cause misbehaviors. For an example, a large negative return from * ops.init_task() triggers an oops when passed up the call chain because the @@ -1670,7 +1668,7 @@ static int ops_sanitize_err(const char *ops_name, s32 err) if (err < 0 && err >= -MAX_ERRNO) return err; - scx_ops_error("ops.%s() returned an invalid errno %d", ops_name, err); + scx_error("ops.%s() returned an invalid errno %d", ops_name, err); return -EPROTO; } @@ -1827,7 +1825,7 @@ static void dispatch_enqueue(struct scx_dispatch_q *dsq, struct task_struct *p, if (!is_local) { raw_spin_lock(&dsq->lock); if (unlikely(dsq->id == SCX_DSQ_INVALID)) { - scx_ops_error("attempting to dispatch to a destroyed dsq"); + scx_error("attempting to dispatch to a destroyed dsq"); /* fall back to the global dsq */ raw_spin_unlock(&dsq->lock); dsq = find_global_dsq(p); @@ -1844,7 +1842,7 @@ static void dispatch_enqueue(struct scx_dispatch_q *dsq, struct task_struct *p, * disallow any internal DSQ from doing vtime ordering of * tasks. */ - scx_ops_error("cannot use vtime ordering for built-in DSQs"); + scx_error("cannot use vtime ordering for built-in DSQs"); enq_flags &= ~SCX_ENQ_DSQ_PRIQ; } @@ -1858,8 +1856,8 @@ static void dispatch_enqueue(struct scx_dispatch_q *dsq, struct task_struct *p, */ if (unlikely(RB_EMPTY_ROOT(&dsq->priq) && nldsq_next_task(dsq, NULL, false))) - scx_ops_error("DSQ ID 0x%016llx already had FIFO-enqueued tasks", - dsq->id); + scx_error("DSQ ID 0x%016llx already had FIFO-enqueued tasks", + dsq->id); p->scx.dsq_flags |= SCX_TASK_DSQ_ON_PRIQ; rb_add(&p->scx.dsq_priq, &dsq->priq, scx_dsq_priq_less); @@ -1880,8 +1878,8 @@ static void dispatch_enqueue(struct scx_dispatch_q *dsq, struct task_struct *p, } else { /* a FIFO DSQ shouldn't be using PRIQ enqueuing */ if (unlikely(!RB_EMPTY_ROOT(&dsq->priq))) - scx_ops_error("DSQ ID 0x%016llx already had PRIQ-enqueued tasks", - dsq->id); + scx_error("DSQ ID 0x%016llx already had PRIQ-enqueued tasks", + dsq->id); if (enq_flags & (SCX_ENQ_HEAD | SCX_ENQ_PREEMPT)) list_add(&p->scx.dsq_list.node, &dsq->list); @@ -2019,8 +2017,8 @@ static struct scx_dispatch_q *find_dsq_for_dispatch(struct rq *rq, u64 dsq_id, dsq = find_user_dsq(dsq_id); if (unlikely(!dsq)) { - scx_ops_error("non-existent DSQ 0x%llx for %s[%d]", - dsq_id, p->comm, p->pid); + scx_error("non-existent DSQ 0x%llx for %s[%d]", + dsq_id, p->comm, p->pid); return find_global_dsq(p); } @@ -2041,12 +2039,12 @@ static void mark_direct_dispatch(struct task_struct *ddsp_task, /* @p must match the task on the enqueue path */ if (unlikely(p != ddsp_task)) { if (IS_ERR(ddsp_task)) - scx_ops_error("%s[%d] already direct-dispatched", - p->comm, p->pid); + scx_error("%s[%d] already direct-dispatched", + p->comm, p->pid); else - scx_ops_error("scheduling for %s[%d] but trying to direct-dispatch %s[%d]", - ddsp_task->comm, ddsp_task->pid, - p->comm, p->pid); + scx_error("scheduling for %s[%d] but trying to direct-dispatch %s[%d]", + ddsp_task->comm, ddsp_task->pid, + p->comm, p->pid); return; } @@ -2224,7 +2222,7 @@ static void set_task_runnable(struct rq *rq, struct task_struct *p) } /* - * list_add_tail() must be used. scx_ops_bypass() depends on tasks being + * list_add_tail() must be used. scx_bypass() depends on tasks being * appended to the runnable_list. */ list_add_tail(&p->scx.runnable_node, &rq->scx.runnable_list); @@ -2488,8 +2486,8 @@ static bool task_can_run_on_remote_rq(struct task_struct *p, struct rq *rq, */ if (unlikely(is_migration_disabled(p))) { if (enforce) - scx_ops_error("SCX_DSQ_LOCAL[_ON] cannot move migration disabled %s[%d] from CPU %d to %d", - p->comm, p->pid, task_cpu(p), cpu); + scx_error("SCX_DSQ_LOCAL[_ON] cannot move migration disabled %s[%d] from CPU %d to %d", + p->comm, p->pid, task_cpu(p), cpu); return false; } @@ -2501,8 +2499,8 @@ static bool task_can_run_on_remote_rq(struct task_struct *p, struct rq *rq, */ if (!task_allowed_on_cpu(p, cpu)) { if (enforce) - scx_ops_error("SCX_DSQ_LOCAL[_ON] target CPU %d not allowed for %s[%d]", - cpu, p->comm, p->pid); + scx_error("SCX_DSQ_LOCAL[_ON] target CPU %d not allowed for %s[%d]", + cpu, p->comm, p->pid); return false; } @@ -2658,13 +2656,13 @@ static struct rq *move_task_between_dsqs(struct task_struct *p, u64 enq_flags, * to the bypass mode can take a long time. Inject artificial delays while the * bypass mode is switching to guarantee timely completion. */ -static void scx_ops_breather(struct rq *rq) +static void scx_breather(struct rq *rq) { u64 until; lockdep_assert_rq_held(rq); - if (likely(!atomic_read(&scx_ops_breather_depth))) + if (likely(!atomic_read(&scx_breather_depth))) return; raw_spin_rq_unlock(rq); @@ -2673,9 +2671,9 @@ static void scx_ops_breather(struct rq *rq) do { int cnt = 1024; - while (atomic_read(&scx_ops_breather_depth) && --cnt) + while (atomic_read(&scx_breather_depth) && --cnt) cpu_relax(); - } while (atomic_read(&scx_ops_breather_depth) && + } while (atomic_read(&scx_breather_depth) && time_before64(ktime_get_ns(), until)); raw_spin_rq_lock(rq); @@ -2686,12 +2684,12 @@ static bool consume_dispatch_q(struct rq *rq, struct scx_dispatch_q *dsq) struct task_struct *p; retry: /* - * This retry loop can repeatedly race against scx_ops_bypass() - * dequeueing tasks from @dsq trying to put the system into the bypass - * mode. On some multi-socket machines (e.g. 2x Intel 8480c), this can - * live-lock the machine into soft lockups. Give a breather. + * This retry loop can repeatedly race against scx_bypass() dequeueing + * tasks from @dsq trying to put the system into the bypass mode. On + * some multi-socket machines (e.g. 2x Intel 8480c), this can live-lock + * the machine into soft lockups. Give a breather. */ - scx_ops_breather(rq); + scx_breather(rq); /* * The caller can't expect to successfully consume a task if the task's @@ -2959,8 +2957,8 @@ static int balance_one(struct rq *rq, struct task_struct *prev) * scheduler wants to handle this explicitly, it should * implement ->cpu_release(). * - * See scx_ops_disable_workfn() for the explanation on the - * bypassing test. + * See scx_disable_workfn() for the explanation on the bypassing + * test. */ if (prev_on_rq && prev->scx.slice && !scx_rq_bypassing(rq)) { rq->scx.flags |= SCX_RQ_BAL_KEEP; @@ -3283,7 +3281,7 @@ static struct task_struct *pick_task_scx(struct rq *rq) * Can happen while enabling as SCX_RQ_BAL_PENDING assertion is * conditional on scx_enabled() and may have been skipped. */ - WARN_ON_ONCE(scx_ops_enable_state() == SCX_OPS_ENABLED); + WARN_ON_ONCE(scx_enable_state() == SCX_ENABLED); keep_prev = false; } @@ -3395,7 +3393,7 @@ static int select_task_rq_scx(struct task_struct *p, int prev_cpu, int wake_flag } else { s32 cpu; - cpu = scx_select_cpu_dfl(p, prev_cpu, wake_flags, 0); + cpu = scx_select_cpu_dfl(p, prev_cpu, wake_flags, NULL, 0); if (cpu >= 0) { p->scx.slice = SCX_SLICE_DFL; p->scx.ddsp_dsq_id = SCX_DSQ_LOCAL; @@ -3448,9 +3446,9 @@ static void handle_hotplug(struct rq *rq, bool online) else if (!online && SCX_HAS_OP(cpu_offline)) SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_offline, cpu); else - scx_ops_exit(SCX_ECODE_ACT_RESTART | SCX_ECODE_RSN_HOTPLUG, - "cpu %d going %s, exiting scheduler", cpu, - online ? "online" : "offline"); + scx_exit(SCX_ECODE_ACT_RESTART | SCX_ECODE_RSN_HOTPLUG, + "cpu %d going %s, exiting scheduler", cpu, + online ? "online" : "offline"); } void scx_rq_activate(struct rq *rq) @@ -3489,10 +3487,9 @@ static bool check_rq_for_timeouts(struct rq *rq) last_runnable + scx_watchdog_timeout))) { u32 dur_ms = jiffies_to_msecs(jiffies - last_runnable); - scx_ops_error_kind(SCX_EXIT_ERROR_STALL, - "%s[%d] failed to run for %u.%03us", - p->comm, p->pid, - dur_ms / 1000, dur_ms % 1000); + __scx_error(SCX_EXIT_ERROR_STALL, + "%s[%d] failed to run for %u.%03us", + p->comm, p->pid, dur_ms / 1000, dur_ms % 1000); timed_out = true; break; } @@ -3530,9 +3527,9 @@ void scx_tick(struct rq *rq) last_check + READ_ONCE(scx_watchdog_timeout)))) { u32 dur_ms = jiffies_to_msecs(jiffies - last_check); - scx_ops_error_kind(SCX_EXIT_ERROR_STALL, - "watchdog failed to check in for %u.%03us", - dur_ms / 1000, dur_ms % 1000); + __scx_error(SCX_EXIT_ERROR_STALL, + "watchdog failed to check in for %u.%03us", + dur_ms / 1000, dur_ms % 1000); } update_other_load_avgs(rq); @@ -3615,7 +3612,7 @@ static void scx_set_task_state(struct task_struct *p, enum scx_task_state state) p->scx.flags |= state << SCX_TASK_STATE_SHIFT; } -static int scx_ops_init_task(struct task_struct *p, struct task_group *tg, bool fork) +static int scx_init_task(struct task_struct *p, struct task_group *tg, bool fork) { int ret; @@ -3657,8 +3654,8 @@ static int scx_ops_init_task(struct task_struct *p, struct task_group *tg, bool task_rq_unlock(rq, p, &rf); } else if (p->policy == SCHED_EXT) { - scx_ops_error("ops.init_task() set task->scx.disallow for %s[%d] during fork", - p->comm, p->pid); + scx_error("ops.init_task() set task->scx.disallow for %s[%d] during fork", + p->comm, p->pid); } } @@ -3666,7 +3663,7 @@ static int scx_ops_init_task(struct task_struct *p, struct task_group *tg, bool return 0; } -static void scx_ops_enable_task(struct task_struct *p) +static void scx_enable_task(struct task_struct *p) { u32 weight; @@ -3691,7 +3688,7 @@ static void scx_ops_enable_task(struct task_struct *p) SCX_CALL_OP_TASK(SCX_KF_REST, set_weight, p, p->scx.weight); } -static void scx_ops_disable_task(struct task_struct *p) +static void scx_disable_task(struct task_struct *p) { lockdep_assert_rq_held(task_rq(p)); WARN_ON_ONCE(scx_get_task_state(p) != SCX_TASK_ENABLED); @@ -3701,7 +3698,7 @@ static void scx_ops_disable_task(struct task_struct *p) scx_set_task_state(p, SCX_TASK_READY); } -static void scx_ops_exit_task(struct task_struct *p) +static void scx_exit_task(struct task_struct *p) { struct scx_exit_task_args args = { .cancelled = false, @@ -3718,7 +3715,7 @@ static void scx_ops_exit_task(struct task_struct *p) case SCX_TASK_READY: break; case SCX_TASK_ENABLED: - scx_ops_disable_task(p); + scx_disable_task(p); break; default: WARN_ON_ONCE(true); @@ -3758,15 +3755,15 @@ int scx_fork(struct task_struct *p) { percpu_rwsem_assert_held(&scx_fork_rwsem); - if (scx_ops_init_task_enabled) - return scx_ops_init_task(p, task_group(p), true); + if (scx_init_task_enabled) + return scx_init_task(p, task_group(p), true); else return 0; } void scx_post_fork(struct task_struct *p) { - if (scx_ops_init_task_enabled) { + if (scx_init_task_enabled) { scx_set_task_state(p, SCX_TASK_READY); /* @@ -3779,7 +3776,7 @@ void scx_post_fork(struct task_struct *p) struct rq *rq; rq = task_rq_lock(p, &rf); - scx_ops_enable_task(p); + scx_enable_task(p); task_rq_unlock(rq, p, &rf); } } @@ -3799,7 +3796,7 @@ void scx_cancel_fork(struct task_struct *p) rq = task_rq_lock(p, &rf); WARN_ON_ONCE(scx_get_task_state(p) >= SCX_TASK_READY); - scx_ops_exit_task(p); + scx_exit_task(p); task_rq_unlock(rq, p, &rf); } @@ -3815,15 +3812,15 @@ void sched_ext_free(struct task_struct *p) spin_unlock_irqrestore(&scx_tasks_lock, flags); /* - * @p is off scx_tasks and wholly ours. scx_ops_enable()'s READY -> - * ENABLED transitions can't race us. Disable ops for @p. + * @p is off scx_tasks and wholly ours. scx_enable()'s READY -> ENABLED + * transitions can't race us. Disable ops for @p. */ if (scx_get_task_state(p) != SCX_TASK_NONE) { struct rq_flags rf; struct rq *rq; rq = task_rq_lock(p, &rf); - scx_ops_exit_task(p); + scx_exit_task(p); task_rq_unlock(rq, p, &rf); } } @@ -3844,7 +3841,7 @@ static void prio_changed_scx(struct rq *rq, struct task_struct *p, int oldprio) static void switching_to_scx(struct rq *rq, struct task_struct *p) { - scx_ops_enable_task(p); + scx_enable_task(p); /* * set_cpus_allowed_scx() is not called while @p is associated with a @@ -3857,7 +3854,7 @@ static void switching_to_scx(struct rq *rq, struct task_struct *p) static void switched_from_scx(struct rq *rq, struct task_struct *p) { - scx_ops_disable_task(p); + scx_disable_task(p); } static void wakeup_preempt_scx(struct rq *rq, struct task_struct *p,int wake_flags) {} @@ -4172,8 +4169,8 @@ static void destroy_dsq(u64 dsq_id) raw_spin_lock_irqsave(&dsq->lock, flags); if (dsq->nr) { - scx_ops_error("attempting to destroy in-use dsq 0x%016llx (nr=%u)", - dsq->id, dsq->nr); + scx_error("attempting to destroy in-use dsq 0x%016llx (nr=%u)", + dsq->id, dsq->nr); goto out_unlock_dsq; } @@ -4265,7 +4262,7 @@ static int scx_cgroup_init(void) css->cgroup, &args); if (ret) { css_put(css); - scx_ops_error("ops.cgroup_init() failed (%d)", ret); + scx_error("ops.cgroup_init() failed (%d)", ret); return ret; } tg->scx_flags |= SCX_TG_INITED; @@ -4300,8 +4297,7 @@ static int scx_cgroup_init(void) { return 0; } static ssize_t scx_attr_state_show(struct kobject *kobj, struct kobj_attribute *ka, char *buf) { - return sysfs_emit(buf, "%s\n", - scx_ops_enable_state_str[scx_ops_enable_state()]); + return sysfs_emit(buf, "%s\n", scx_enable_state_str[scx_enable_state()]); } SCX_ATTR(state); @@ -4410,8 +4406,7 @@ static const struct kset_uevent_ops scx_uevent_ops = { */ bool task_should_scx(int policy) { - if (!scx_enabled() || - unlikely(scx_ops_enable_state() == SCX_OPS_DISABLING)) + if (!scx_enabled() || unlikely(scx_enable_state() == SCX_DISABLING)) return false; if (READ_ONCE(scx_switching_all)) return true; @@ -4430,15 +4425,15 @@ bool task_should_scx(int policy) */ void scx_softlockup(u32 dur_s) { - switch (scx_ops_enable_state()) { - case SCX_OPS_ENABLING: - case SCX_OPS_ENABLED: + switch (scx_enable_state()) { + case SCX_ENABLING: + case SCX_ENABLED: break; default: return; } - /* allow only one instance, cleared at the end of scx_ops_bypass() */ + /* allow only one instance, cleared at the end of scx_bypass() */ if (test_and_set_bit(0, &scx_in_softlockup)) return; @@ -4447,23 +4442,21 @@ void scx_softlockup(u32 dur_s) /* * Some CPUs may be trapped in the dispatch paths. Enable breather - * immediately; otherwise, we might even be able to get to - * scx_ops_bypass(). + * immediately; otherwise, we might even be able to get to scx_bypass(). */ - atomic_inc(&scx_ops_breather_depth); + atomic_inc(&scx_breather_depth); - scx_ops_error("soft lockup - CPU#%d stuck for %us", - smp_processor_id(), dur_s); + scx_error("soft lockup - CPU#%d stuck for %us", smp_processor_id(), dur_s); } static void scx_clear_softlockup(void) { if (test_and_clear_bit(0, &scx_in_softlockup)) - atomic_dec(&scx_ops_breather_depth); + atomic_dec(&scx_breather_depth); } /** - * scx_ops_bypass - [Un]bypass scx_ops and guarantee forward progress + * scx_bypass - [Un]bypass scx_ops and guarantee forward progress * @bypass: true for bypass, false for unbypass * * Bypassing guarantees that all runnable tasks make forward progress without @@ -4493,7 +4486,7 @@ static void scx_clear_softlockup(void) * * - scx_prio_less() reverts to the default core_sched_at order. */ -static void scx_ops_bypass(bool bypass) +static void scx_bypass(bool bypass) { static DEFINE_RAW_SPINLOCK(bypass_lock); static unsigned long bypass_timestamp; @@ -4503,22 +4496,22 @@ static void scx_ops_bypass(bool bypass) raw_spin_lock_irqsave(&bypass_lock, flags); if (bypass) { - scx_ops_bypass_depth++; - WARN_ON_ONCE(scx_ops_bypass_depth <= 0); - if (scx_ops_bypass_depth != 1) + scx_bypass_depth++; + WARN_ON_ONCE(scx_bypass_depth <= 0); + if (scx_bypass_depth != 1) goto unlock; bypass_timestamp = ktime_get_ns(); scx_add_event(SCX_EV_BYPASS_ACTIVATE, 1); } else { - scx_ops_bypass_depth--; - WARN_ON_ONCE(scx_ops_bypass_depth < 0); - if (scx_ops_bypass_depth != 0) + scx_bypass_depth--; + WARN_ON_ONCE(scx_bypass_depth < 0); + if (scx_bypass_depth != 0) goto unlock; scx_add_event(SCX_EV_BYPASS_DURATION, ktime_get_ns() - bypass_timestamp); } - atomic_inc(&scx_ops_breather_depth); + atomic_inc(&scx_breather_depth); /* * No task property is changing. We just need to make sure all currently @@ -4576,7 +4569,7 @@ static void scx_ops_bypass(bool bypass) raw_spin_rq_unlock(rq); } - atomic_dec(&scx_ops_breather_depth); + atomic_dec(&scx_breather_depth); unlock: raw_spin_unlock_irqrestore(&bypass_lock, flags); scx_clear_softlockup(); @@ -4632,7 +4625,7 @@ static const char *scx_exit_reason(enum scx_exit_kind kind) } } -static void scx_ops_disable_workfn(struct kthread_work *work) +static void scx_disable_workfn(struct kthread_work *work) { struct scx_exit_info *ei = scx_exit_info; struct scx_task_iter sti; @@ -4657,17 +4650,16 @@ static void scx_ops_disable_workfn(struct kthread_work *work) ei->reason = scx_exit_reason(ei->kind); /* guarantee forward progress by bypassing scx_ops */ - scx_ops_bypass(true); + scx_bypass(true); - switch (scx_ops_set_enable_state(SCX_OPS_DISABLING)) { - case SCX_OPS_DISABLING: + switch (scx_set_enable_state(SCX_DISABLING)) { + case SCX_DISABLING: WARN_ONCE(true, "sched_ext: duplicate disabling instance?"); break; - case SCX_OPS_DISABLED: + case SCX_DISABLED: pr_warn("sched_ext: ops error detected without ops (%s)\n", scx_exit_info->msg); - WARN_ON_ONCE(scx_ops_set_enable_state(SCX_OPS_DISABLED) != - SCX_OPS_DISABLING); + WARN_ON_ONCE(scx_set_enable_state(SCX_DISABLED) != SCX_DISABLING); goto done; default: break; @@ -4678,14 +4670,14 @@ static void scx_ops_disable_workfn(struct kthread_work *work) * we can safely use blocking synchronization constructs. Actually * disable ops. */ - mutex_lock(&scx_ops_enable_mutex); + mutex_lock(&scx_enable_mutex); static_branch_disable(&__scx_switched_all); WRITE_ONCE(scx_switching_all, false); /* * Shut down cgroup support before tasks so that the cgroup attach path - * doesn't race against scx_ops_exit_task(). + * doesn't race against scx_exit_task(). */ scx_cgroup_lock(); scx_cgroup_exit(); @@ -4697,7 +4689,7 @@ static void scx_ops_disable_workfn(struct kthread_work *work) */ percpu_down_write(&scx_fork_rwsem); - scx_ops_init_task_enabled = false; + scx_init_task_enabled = false; scx_task_iter_start(&sti); while ((p = scx_task_iter_next_locked(&sti))) { @@ -4717,7 +4709,7 @@ static void scx_ops_disable_workfn(struct kthread_work *work) sched_enq_and_set_task(&ctx); check_class_changed(task_rq(p), p, old_class, p->prio); - scx_ops_exit_task(p); + scx_exit_task(p); } scx_task_iter_stop(&sti); percpu_up_write(&scx_fork_rwsem); @@ -4732,7 +4724,7 @@ static void scx_ops_disable_workfn(struct kthread_work *work) } /* no task is on scx, turn off all the switches and flush in-progress calls */ - static_branch_disable(&__scx_ops_enabled); + static_branch_disable(&__scx_enabled); for (i = SCX_OPI_BEGIN; i < SCX_OPI_END; i++) static_branch_disable(&scx_has_op[i]); static_branch_disable(&scx_ops_allow_queued_wakeup); @@ -4792,29 +4784,28 @@ static void scx_ops_disable_workfn(struct kthread_work *work) free_exit_info(scx_exit_info); scx_exit_info = NULL; - mutex_unlock(&scx_ops_enable_mutex); + mutex_unlock(&scx_enable_mutex); - WARN_ON_ONCE(scx_ops_set_enable_state(SCX_OPS_DISABLED) != - SCX_OPS_DISABLING); + WARN_ON_ONCE(scx_set_enable_state(SCX_DISABLED) != SCX_DISABLING); done: - scx_ops_bypass(false); + scx_bypass(false); } -static DEFINE_KTHREAD_WORK(scx_ops_disable_work, scx_ops_disable_workfn); +static DEFINE_KTHREAD_WORK(scx_disable_work, scx_disable_workfn); -static void schedule_scx_ops_disable_work(void) +static void schedule_scx_disable_work(void) { - struct kthread_worker *helper = READ_ONCE(scx_ops_helper); + struct kthread_worker *helper = READ_ONCE(scx_helper); /* * We may be called spuriously before the first bpf_sched_ext_reg(). If - * scx_ops_helper isn't set up yet, there's nothing to do. + * scx_helper isn't set up yet, there's nothing to do. */ if (helper) - kthread_queue_work(helper, &scx_ops_disable_work); + kthread_queue_work(helper, &scx_disable_work); } -static void scx_ops_disable(enum scx_exit_kind kind) +static void scx_disable(enum scx_exit_kind kind) { int none = SCX_EXIT_NONE; @@ -4823,7 +4814,7 @@ static void scx_ops_disable(enum scx_exit_kind kind) atomic_try_cmpxchg(&scx_exit_kind, &none, kind); - schedule_scx_ops_disable_work(); + schedule_scx_disable_work(); } static void dump_newline(struct seq_buf *s) @@ -5121,21 +5112,20 @@ static void scx_dump_state(struct scx_exit_info *ei, size_t dump_len) spin_unlock_irqrestore(&dump_lock, flags); } -static void scx_ops_error_irq_workfn(struct irq_work *irq_work) +static void scx_error_irq_workfn(struct irq_work *irq_work) { struct scx_exit_info *ei = scx_exit_info; if (ei->kind >= SCX_EXIT_ERROR) scx_dump_state(ei, scx_ops.exit_dump_len); - schedule_scx_ops_disable_work(); + schedule_scx_disable_work(); } -static DEFINE_IRQ_WORK(scx_ops_error_irq_work, scx_ops_error_irq_workfn); +static DEFINE_IRQ_WORK(scx_error_irq_work, scx_error_irq_workfn); -static __printf(3, 4) void scx_ops_exit_kind(enum scx_exit_kind kind, - s64 exit_code, - const char *fmt, ...) +static __printf(3, 4) void __scx_exit(enum scx_exit_kind kind, s64 exit_code, + const char *fmt, ...) { struct scx_exit_info *ei = scx_exit_info; int none = SCX_EXIT_NONE; @@ -5155,12 +5145,12 @@ static __printf(3, 4) void scx_ops_exit_kind(enum scx_exit_kind kind, /* * Set ei->kind and ->reason for scx_dump_state(). They'll be set again - * in scx_ops_disable_workfn(). + * in scx_disable_workfn(). */ ei->kind = kind; ei->reason = scx_exit_reason(ei->kind); - irq_work_queue(&scx_ops_error_irq_work); + irq_work_queue(&scx_error_irq_work); } static struct kthread_worker *scx_create_rt_helper(const char *name) @@ -5185,9 +5175,9 @@ static void check_hotplug_seq(const struct sched_ext_ops *ops) if (ops->hotplug_seq) { global_hotplug_seq = atomic_long_read(&scx_hotplug_seq); if (ops->hotplug_seq != global_hotplug_seq) { - scx_ops_exit(SCX_ECODE_ACT_RESTART | SCX_ECODE_RSN_HOTPLUG, - "expected hotplug seq %llu did not match actual %llu", - ops->hotplug_seq, global_hotplug_seq); + scx_exit(SCX_ECODE_ACT_RESTART | SCX_ECODE_RSN_HOTPLUG, + "expected hotplug seq %llu did not match actual %llu", + ops->hotplug_seq, global_hotplug_seq); } } } @@ -5199,7 +5189,7 @@ static int validate_ops(const struct sched_ext_ops *ops) * ops.enqueue() callback isn't implemented. */ if ((ops->flags & SCX_OPS_ENQ_LAST) && !ops->enqueue) { - scx_ops_error("SCX_OPS_ENQ_LAST requires ops.enqueue() to be implemented"); + scx_error("SCX_OPS_ENQ_LAST requires ops.enqueue() to be implemented"); return -EINVAL; } @@ -5209,7 +5199,7 @@ static int validate_ops(const struct sched_ext_ops *ops) */ if ((ops->flags & SCX_OPS_BUILTIN_IDLE_PER_NODE) && (ops->update_idle && !(ops->flags & SCX_OPS_KEEP_BUILTIN_IDLE))) { - scx_ops_error("SCX_OPS_BUILTIN_IDLE_PER_NODE requires CPU idle selection enabled"); + scx_error("SCX_OPS_BUILTIN_IDLE_PER_NODE requires CPU idle selection enabled"); return -EINVAL; } @@ -5219,7 +5209,7 @@ static int validate_ops(const struct sched_ext_ops *ops) return 0; } -static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) +static int scx_enable(struct sched_ext_ops *ops, struct bpf_link *link) { struct scx_task_iter sti; struct task_struct *p; @@ -5232,7 +5222,7 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) return -EINVAL; } - mutex_lock(&scx_ops_enable_mutex); + mutex_lock(&scx_enable_mutex); /* * Clear event counters so a new scx scheduler gets @@ -5243,10 +5233,9 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) memset(e, 0, sizeof(*e)); } - if (!scx_ops_helper) { - WRITE_ONCE(scx_ops_helper, - scx_create_rt_helper("sched_ext_ops_helper")); - if (!scx_ops_helper) { + if (!scx_helper) { + WRITE_ONCE(scx_helper, scx_create_rt_helper("sched_ext_helper")); + if (!scx_helper) { ret = -ENOMEM; goto err_unlock; } @@ -5280,7 +5269,7 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) global_dsqs = dsqs; } - if (scx_ops_enable_state() != SCX_OPS_DISABLED) { + if (scx_enable_state() != SCX_DISABLED) { ret = -EBUSY; goto err_unlock; } @@ -5308,8 +5297,7 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) */ scx_ops = *ops; - WARN_ON_ONCE(scx_ops_set_enable_state(SCX_OPS_ENABLING) != - SCX_OPS_DISABLED); + WARN_ON_ONCE(scx_set_enable_state(SCX_ENABLING) != SCX_DISABLED); atomic_set(&scx_exit_kind, SCX_EXIT_NONE); scx_warned_zero_slice = false; @@ -5332,7 +5320,7 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) if (ret) { ret = ops_sanitize_err("init", ret); cpus_read_unlock(); - scx_ops_error("ops.init() failed (%d)", ret); + scx_error("ops.init() failed (%d)", ret); goto err_disable; } } @@ -5371,12 +5359,12 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) scx_watchdog_timeout / 2); /* - * Once __scx_ops_enabled is set, %current can be switched to SCX - * anytime. This can lead to stalls as some BPF schedulers (e.g. - * userspace scheduling) may not function correctly before all tasks are - * switched. Init in bypass mode to guarantee forward progress. + * Once __scx_enabled is set, %current can be switched to SCX anytime. + * This can lead to stalls as some BPF schedulers (e.g. userspace + * scheduling) may not function correctly before all tasks are switched. + * Init in bypass mode to guarantee forward progress. */ - scx_ops_bypass(true); + scx_bypass(true); for (i = SCX_OPI_NORMAL_BEGIN; i < SCX_OPI_NORMAL_END; i++) if (((void (**)(void))ops)[i]) @@ -5399,8 +5387,8 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) */ percpu_down_write(&scx_fork_rwsem); - WARN_ON_ONCE(scx_ops_init_task_enabled); - scx_ops_init_task_enabled = true; + WARN_ON_ONCE(scx_init_task_enabled); + scx_init_task_enabled = true; /* * Enable ops for every task. Fork is excluded by scx_fork_rwsem @@ -5409,11 +5397,11 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) * tasks. Prep all tasks first and then enable them with preemption * disabled. * - * All cgroups should be initialized before scx_ops_init_task() so that - * the BPF scheduler can reliably track each task's cgroup membership - * from scx_ops_init_task(). Lock out cgroup on/offlining and task - * migrations while tasks are being initialized so that - * scx_cgroup_can_attach() never sees uninitialized tasks. + * All cgroups should be initialized before scx_init_task() so that the + * BPF scheduler can reliably track each task's cgroup membership from + * scx_init_task(). Lock out cgroup on/offlining and task migrations + * while tasks are being initialized so that scx_cgroup_can_attach() + * never sees uninitialized tasks. */ scx_cgroup_lock(); ret = scx_cgroup_init(); @@ -5432,13 +5420,13 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) scx_task_iter_unlock(&sti); - ret = scx_ops_init_task(p, task_group(p), false); + ret = scx_init_task(p, task_group(p), false); if (ret) { put_task_struct(p); scx_task_iter_relock(&sti); scx_task_iter_stop(&sti); - scx_ops_error("ops.init_task() failed (%d) for %s[%d]", - ret, p->comm, p->pid); + scx_error("ops.init_task() failed (%d) for %s[%d]", + ret, p->comm, p->pid); goto err_disable_unlock_all; } @@ -5456,7 +5444,7 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) * all eligible tasks. */ WRITE_ONCE(scx_switching_all, !(ops->flags & SCX_OPS_SWITCH_PARTIAL)); - static_branch_enable(&__scx_ops_enabled); + static_branch_enable(&__scx_enabled); /* * We're fully committed and can't fail. The task READY -> ENABLED @@ -5487,9 +5475,9 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) scx_task_iter_stop(&sti); percpu_up_write(&scx_fork_rwsem); - scx_ops_bypass(false); + scx_bypass(false); - if (!scx_ops_tryset_enable_state(SCX_OPS_ENABLED, SCX_OPS_ENABLING)) { + if (!scx_tryset_enable_state(SCX_ENABLED, SCX_ENABLING)) { WARN_ON_ONCE(atomic_read(&scx_exit_kind) == SCX_EXIT_NONE); goto err_disable; } @@ -5500,7 +5488,7 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) pr_info("sched_ext: BPF scheduler \"%s\" enabled%s\n", scx_ops.name, scx_switched_all() ? "" : " (partial)"); kobject_uevent(scx_root_kobj, KOBJ_ADD); - mutex_unlock(&scx_ops_enable_mutex); + mutex_unlock(&scx_enable_mutex); atomic_long_inc(&scx_enable_seq); @@ -5516,26 +5504,26 @@ err: scx_exit_info = NULL; } err_unlock: - mutex_unlock(&scx_ops_enable_mutex); + mutex_unlock(&scx_enable_mutex); return ret; err_disable_unlock_all: scx_cgroup_unlock(); percpu_up_write(&scx_fork_rwsem); - scx_ops_bypass(false); + scx_bypass(false); err_disable: - mutex_unlock(&scx_ops_enable_mutex); + mutex_unlock(&scx_enable_mutex); /* * Returning an error code here would not pass all the error information - * to userspace. Record errno using scx_ops_error() for cases - * scx_ops_error() wasn't already invoked and exit indicating success so - * that the error is notified through ops.exit() with all the details. + * to userspace. Record errno using scx_error() for cases scx_error() + * wasn't already invoked and exit indicating success so that the error + * is notified through ops.exit() with all the details. * - * Flush scx_ops_disable_work to ensure that error is reported before - * init completion. + * Flush scx_disable_work to ensure that error is reported before init + * completion. */ - scx_ops_error("scx_ops_enable() failed (%d)", ret); - kthread_flush_work(&scx_ops_disable_work); + scx_error("scx_enable() failed (%d)", ret); + kthread_flush_work(&scx_disable_work); return 0; } @@ -5679,13 +5667,13 @@ static int bpf_scx_check_member(const struct btf_type *t, static int bpf_scx_reg(void *kdata, struct bpf_link *link) { - return scx_ops_enable(kdata, link); + return scx_enable(kdata, link); } static void bpf_scx_unreg(void *kdata, struct bpf_link *link) { - scx_ops_disable(SCX_EXIT_UNREG); - kthread_flush_work(&scx_ops_disable_work); + scx_disable(SCX_EXIT_UNREG); + kthread_flush_work(&scx_disable_work); } static int bpf_scx_init(struct btf *btf) @@ -5807,8 +5795,8 @@ static struct bpf_struct_ops bpf_sched_ext_ops = { static void sysrq_handle_sched_ext_reset(u8 key) { - if (scx_ops_helper) - scx_ops_disable(SCX_EXIT_SYSRQ); + if (scx_helper) + scx_disable(SCX_EXIT_SYSRQ); else pr_info("sched_ext: BPF scheduler not yet used\n"); } @@ -5958,13 +5946,13 @@ static void kick_cpus_irq_workfn(struct irq_work *irq_work) */ void print_scx_info(const char *log_lvl, struct task_struct *p) { - enum scx_ops_enable_state state = scx_ops_enable_state(); + enum scx_enable_state state = scx_enable_state(); const char *all = READ_ONCE(scx_switching_all) ? "+all" : ""; char runnable_at_buf[22] = "?"; struct sched_class *class; unsigned long runnable_at; - if (state == SCX_OPS_DISABLED) + if (state == SCX_DISABLED) return; /* @@ -5974,7 +5962,7 @@ void print_scx_info(const char *log_lvl, struct task_struct *p) if (copy_from_kernel_nofault(&class, &p->sched_class, sizeof(class)) || class != &ext_sched_class) { printk("%sSched_ext: %s (%s%s)", log_lvl, scx_ops.name, - scx_ops_enable_state_str[state], all); + scx_enable_state_str[state], all); return; } @@ -5985,7 +5973,7 @@ void print_scx_info(const char *log_lvl, struct task_struct *p) /* print everything onto one line to conserve console space */ printk("%sSched_ext: %s (%s%s), task: runnable_at=%s", - log_lvl, scx_ops.name, scx_ops_enable_state_str[state], all, + log_lvl, scx_ops.name, scx_enable_state_str[state], all, runnable_at_buf); } @@ -6001,12 +5989,12 @@ static int scx_pm_handler(struct notifier_block *nb, unsigned long event, void * case PM_HIBERNATION_PREPARE: case PM_SUSPEND_PREPARE: case PM_RESTORE_PREPARE: - scx_ops_bypass(true); + scx_bypass(true); break; case PM_POST_HIBERNATION: case PM_POST_SUSPEND: case PM_POST_RESTORE: - scx_ops_bypass(false); + scx_bypass(false); break; } @@ -6073,12 +6061,12 @@ static bool scx_dsq_insert_preamble(struct task_struct *p, u64 enq_flags) lockdep_assert_irqs_disabled(); if (unlikely(!p)) { - scx_ops_error("called with NULL task"); + scx_error("called with NULL task"); return false; } if (unlikely(enq_flags & __SCX_ENQ_INTERNAL_MASK)) { - scx_ops_error("invalid enq_flags 0x%llx", enq_flags); + scx_error("invalid enq_flags 0x%llx", enq_flags); return false; } @@ -6098,7 +6086,7 @@ static void scx_dsq_insert_commit(struct task_struct *p, u64 dsq_id, } if (unlikely(dspc->cursor >= scx_dsp_max_batch)) { - scx_ops_error("dispatch buffer overflow"); + scx_error("dispatch buffer overflow"); return; } @@ -6264,7 +6252,7 @@ static bool scx_dsq_move(struct bpf_iter_scx_dsq_kern *kit, * cause similar live-lock conditions as consume_dispatch_q(). Insert a * breather if necessary. */ - scx_ops_breather(src_rq); + scx_breather(src_rq); locked_rq = src_rq; raw_spin_lock(&src_dsq->lock); @@ -6343,7 +6331,7 @@ __bpf_kfunc void scx_bpf_dispatch_cancel(void) if (dspc->cursor > 0) dspc->cursor--; else - scx_ops_error("dispatch buffer underflow"); + scx_error("dispatch buffer underflow"); } /** @@ -6372,7 +6360,7 @@ __bpf_kfunc bool scx_bpf_dsq_move_to_local(u64 dsq_id) dsq = find_user_dsq(dsq_id); if (unlikely(!dsq)) { - scx_ops_error("invalid DSQ ID 0x%016llx", dsq_id); + scx_error("invalid DSQ ID 0x%016llx", dsq_id); return false; } @@ -6696,7 +6684,7 @@ __bpf_kfunc void scx_bpf_kick_cpu(s32 cpu, u64 flags) struct rq *target_rq = cpu_rq(cpu); if (unlikely(flags & (SCX_KICK_PREEMPT | SCX_KICK_WAIT))) - scx_ops_error("PREEMPT/WAIT cannot be used with SCX_KICK_IDLE"); + scx_error("PREEMPT/WAIT cannot be used with SCX_KICK_IDLE"); if (raw_spin_rq_trylock(target_rq)) { if (can_skip_idle_kick(target_rq)) { @@ -6884,21 +6872,20 @@ static s32 __bstr_format(u64 *data_buf, char *line_buf, size_t line_size, if (data__sz % 8 || data__sz > MAX_BPRINTF_VARARGS * 8 || (data__sz && !data)) { - scx_ops_error("invalid data=%p and data__sz=%u", - (void *)data, data__sz); + scx_error("invalid data=%p and data__sz=%u", (void *)data, data__sz); return -EINVAL; } ret = copy_from_kernel_nofault(data_buf, data, data__sz); if (ret < 0) { - scx_ops_error("failed to read data fields (%d)", ret); + scx_error("failed to read data fields (%d)", ret); return ret; } ret = bpf_bprintf_prepare(fmt, UINT_MAX, data_buf, data__sz / 8, &bprintf_data); if (ret < 0) { - scx_ops_error("format preparation failed (%d)", ret); + scx_error("format preparation failed (%d)", ret); return ret; } @@ -6906,8 +6893,7 @@ static s32 __bstr_format(u64 *data_buf, char *line_buf, size_t line_size, bprintf_data.bin_args); bpf_bprintf_cleanup(&bprintf_data); if (ret < 0) { - scx_ops_error("(\"%s\", %p, %u) failed to format", - fmt, data, data__sz); + scx_error("(\"%s\", %p, %u) failed to format", fmt, data, data__sz); return ret; } @@ -6940,8 +6926,7 @@ __bpf_kfunc void scx_bpf_exit_bstr(s64 exit_code, char *fmt, raw_spin_lock_irqsave(&scx_exit_bstr_buf_lock, flags); if (bstr_format(&scx_exit_bstr_buf, fmt, data, data__sz) >= 0) - scx_ops_exit_kind(SCX_EXIT_UNREG_BPF, exit_code, "%s", - scx_exit_bstr_buf.line); + __scx_exit(SCX_EXIT_UNREG_BPF, exit_code, "%s", scx_exit_bstr_buf.line); raw_spin_unlock_irqrestore(&scx_exit_bstr_buf_lock, flags); } @@ -6961,8 +6946,7 @@ __bpf_kfunc void scx_bpf_error_bstr(char *fmt, unsigned long long *data, raw_spin_lock_irqsave(&scx_exit_bstr_buf_lock, flags); if (bstr_format(&scx_exit_bstr_buf, fmt, data, data__sz) >= 0) - scx_ops_exit_kind(SCX_EXIT_ERROR_BPF, 0, "%s", - scx_exit_bstr_buf.line); + __scx_exit(SCX_EXIT_ERROR_BPF, 0, "%s", scx_exit_bstr_buf.line); raw_spin_unlock_irqrestore(&scx_exit_bstr_buf_lock, flags); } @@ -6986,7 +6970,7 @@ __bpf_kfunc void scx_bpf_dump_bstr(char *fmt, unsigned long long *data, s32 ret; if (raw_smp_processor_id() != dd->cpu) { - scx_ops_error("scx_bpf_dump() must only be called from ops.dump() and friends"); + scx_error("scx_bpf_dump() must only be called from ops.dump() and friends"); return; } @@ -7072,7 +7056,7 @@ __bpf_kfunc u32 scx_bpf_cpuperf_cur(s32 cpu) __bpf_kfunc void scx_bpf_cpuperf_set(s32 cpu, u32 perf) { if (unlikely(perf > SCX_CPUPERF_ONE)) { - scx_ops_error("Invalid cpuperf target %u for CPU %d", perf, cpu); + scx_error("Invalid cpuperf target %u for CPU %d", perf, cpu); return; } diff --git a/kernel/sched/ext_idle.c b/kernel/sched/ext_idle.c index cb343ca889e0..023ae6df5e8c 100644 --- a/kernel/sched/ext_idle.c +++ b/kernel/sched/ext_idle.c @@ -47,6 +47,13 @@ static struct scx_idle_cpus scx_idle_global_masks; static struct scx_idle_cpus **scx_idle_node_masks; /* + * Local per-CPU cpumasks (used to generate temporary idle cpumasks). + */ +static DEFINE_PER_CPU(cpumask_var_t, local_idle_cpumask); +static DEFINE_PER_CPU(cpumask_var_t, local_llc_idle_cpumask); +static DEFINE_PER_CPU(cpumask_var_t, local_numa_idle_cpumask); + +/* * Return the idle masks associated to a target @node. * * NUMA_NO_NODE identifies the global idle cpumask. @@ -392,6 +399,14 @@ void scx_idle_update_selcpu_topology(struct sched_ext_ops *ops) } /* + * Return true if @p can run on all possible CPUs, false otherwise. + */ +static inline bool task_affinity_all(const struct task_struct *p) +{ + return p->nr_cpus_allowed >= num_possible_cpus(); +} + +/* * Built-in CPU idle selection policy: * * 1. Prioritize full-idle cores: @@ -403,13 +418,15 @@ void scx_idle_update_selcpu_topology(struct sched_ext_ops *ops) * branch prediction optimizations. * * 3. Pick a CPU within the same LLC (Last-Level Cache): - * - if the above conditions aren't met, pick a CPU that shares the same LLC - * to maintain cache locality. + * - if the above conditions aren't met, pick a CPU that shares the same + * LLC, if the LLC domain is a subset of @cpus_allowed, to maintain + * cache locality. * * 4. Pick a CPU within the same NUMA node, if enabled: - * - choose a CPU from the same NUMA node to reduce memory access latency. + * - choose a CPU from the same NUMA node, if the node cpumask is a + * subset of @cpus_allowed, to reduce memory access latency. * - * 5. Pick any idle CPU usable by the task. + * 5. Pick any idle CPU within the @cpus_allowed domain. * * Step 3 and 4 are performed only if the system has, respectively, * multiple LLCs / multiple NUMA nodes (see scx_selcpu_topo_llc and @@ -424,35 +441,77 @@ void scx_idle_update_selcpu_topology(struct sched_ext_ops *ops) * NOTE: tasks that can only run on 1 CPU are excluded by this logic, because * we never call ops.select_cpu() for them, see select_task_rq(). */ -s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags, u64 flags) +s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags, + const struct cpumask *cpus_allowed, u64 flags) { - const struct cpumask *llc_cpus = NULL; - const struct cpumask *numa_cpus = NULL; + const struct cpumask *llc_cpus = NULL, *numa_cpus = NULL; + const struct cpumask *allowed = cpus_allowed ?: p->cpus_ptr; int node = scx_cpu_node_if_enabled(prev_cpu); s32 cpu; + preempt_disable(); + + /* + * Determine the subset of CPUs usable by @p within @cpus_allowed. + */ + if (allowed != p->cpus_ptr) { + struct cpumask *local_cpus = this_cpu_cpumask_var_ptr(local_idle_cpumask); + + if (task_affinity_all(p)) { + allowed = cpus_allowed; + } else if (cpumask_and(local_cpus, cpus_allowed, p->cpus_ptr)) { + allowed = local_cpus; + } else { + cpu = -EBUSY; + goto out_enable; + } + + /* + * If @prev_cpu is not in the allowed CPUs, skip topology + * optimizations and try to pick any idle CPU usable by the + * task. + * + * If %SCX_OPS_BUILTIN_IDLE_PER_NODE is enabled, prioritize + * the current node, as it may optimize some waker->wakee + * workloads. + */ + if (!cpumask_test_cpu(prev_cpu, allowed)) { + node = scx_cpu_node_if_enabled(smp_processor_id()); + cpu = scx_pick_idle_cpu(allowed, node, flags); + goto out_enable; + } + } + /* * This is necessary to protect llc_cpus. */ rcu_read_lock(); /* - * Determine the scheduling domain only if the task is allowed to run - * on all CPUs. + * Determine the subset of CPUs that the task can use in its + * current LLC and node. * - * This is done primarily for efficiency, as it avoids the overhead of - * updating a cpumask every time we need to select an idle CPU (which - * can be costly in large SMP systems), but it also aligns logically: - * if a task's scheduling domain is restricted by user-space (through - * CPU affinity), the task will simply use the flat scheduling domain - * defined by user-space. + * If the task can run on all CPUs, use the node and LLC cpumasks + * directly. */ - if (p->nr_cpus_allowed >= num_possible_cpus()) { - if (static_branch_maybe(CONFIG_NUMA, &scx_selcpu_topo_numa)) - numa_cpus = numa_span(prev_cpu); + if (static_branch_maybe(CONFIG_NUMA, &scx_selcpu_topo_numa)) { + struct cpumask *local_cpus = this_cpu_cpumask_var_ptr(local_numa_idle_cpumask); + const struct cpumask *cpus = numa_span(prev_cpu); + + if (allowed == p->cpus_ptr && task_affinity_all(p)) + numa_cpus = cpus; + else if (cpus && cpumask_and(local_cpus, allowed, cpus)) + numa_cpus = local_cpus; + } + + if (static_branch_maybe(CONFIG_SCHED_MC, &scx_selcpu_topo_llc)) { + struct cpumask *local_cpus = this_cpu_cpumask_var_ptr(local_llc_idle_cpumask); + const struct cpumask *cpus = llc_span(prev_cpu); - if (static_branch_maybe(CONFIG_SCHED_MC, &scx_selcpu_topo_llc)) - llc_cpus = llc_span(prev_cpu); + if (allowed == p->cpus_ptr && task_affinity_all(p)) + llc_cpus = cpus; + else if (cpus && cpumask_and(local_cpus, allowed, cpus)) + llc_cpus = local_cpus; } /* @@ -490,7 +549,7 @@ s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags, u64 cpu_rq(cpu)->scx.local_dsq.nr == 0 && (!(flags & SCX_PICK_IDLE_IN_NODE) || (waker_node == node)) && !cpumask_empty(idle_cpumask(waker_node)->cpu)) { - if (cpumask_test_cpu(cpu, p->cpus_ptr)) + if (cpumask_test_cpu(cpu, allowed)) goto out_unlock; } } @@ -535,7 +594,7 @@ s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags, u64 * begin in prev_cpu's node and proceed to other nodes in * order of increasing distance. */ - cpu = scx_pick_idle_cpu(p->cpus_ptr, node, flags | SCX_PICK_IDLE_CORE); + cpu = scx_pick_idle_cpu(allowed, node, flags | SCX_PICK_IDLE_CORE); if (cpu >= 0) goto out_unlock; @@ -583,10 +642,12 @@ s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags, u64 * in prev_cpu's node and proceed to other nodes in order of * increasing distance. */ - cpu = scx_pick_idle_cpu(p->cpus_ptr, node, flags); + cpu = scx_pick_idle_cpu(allowed, node, flags); out_unlock: rcu_read_unlock(); +out_enable: + preempt_enable(); return cpu; } @@ -596,7 +657,7 @@ out_unlock: */ void scx_idle_init_masks(void) { - int node; + int i; /* Allocate global idle cpumasks */ BUG_ON(!alloc_cpumask_var(&scx_idle_global_masks.cpu, GFP_KERNEL)); @@ -607,13 +668,23 @@ void scx_idle_init_masks(void) sizeof(*scx_idle_node_masks), GFP_KERNEL); BUG_ON(!scx_idle_node_masks); - for_each_node(node) { - scx_idle_node_masks[node] = kzalloc_node(sizeof(**scx_idle_node_masks), - GFP_KERNEL, node); - BUG_ON(!scx_idle_node_masks[node]); + for_each_node(i) { + scx_idle_node_masks[i] = kzalloc_node(sizeof(**scx_idle_node_masks), + GFP_KERNEL, i); + BUG_ON(!scx_idle_node_masks[i]); + + BUG_ON(!alloc_cpumask_var_node(&scx_idle_node_masks[i]->cpu, GFP_KERNEL, i)); + BUG_ON(!alloc_cpumask_var_node(&scx_idle_node_masks[i]->smt, GFP_KERNEL, i)); + } - BUG_ON(!alloc_cpumask_var_node(&scx_idle_node_masks[node]->cpu, GFP_KERNEL, node)); - BUG_ON(!alloc_cpumask_var_node(&scx_idle_node_masks[node]->smt, GFP_KERNEL, node)); + /* Allocate local per-cpu idle cpumasks */ + for_each_possible_cpu(i) { + BUG_ON(!alloc_cpumask_var_node(&per_cpu(local_idle_cpumask, i), + GFP_KERNEL, cpu_to_node(i))); + BUG_ON(!alloc_cpumask_var_node(&per_cpu(local_llc_idle_cpumask, i), + GFP_KERNEL, cpu_to_node(i))); + BUG_ON(!alloc_cpumask_var_node(&per_cpu(local_numa_idle_cpumask, i), + GFP_KERNEL, cpu_to_node(i))); } } @@ -748,7 +819,7 @@ void scx_idle_disable(void) static int validate_node(int node) { if (!static_branch_likely(&scx_builtin_idle_per_node)) { - scx_ops_error("per-node idle tracking is disabled"); + scx_error("per-node idle tracking is disabled"); return -EOPNOTSUPP; } @@ -758,13 +829,13 @@ static int validate_node(int node) /* Make sure node is in a valid range */ if (node < 0 || node >= nr_node_ids) { - scx_ops_error("invalid node %d", node); + scx_error("invalid node %d", node); return -EINVAL; } /* Make sure the node is part of the set of possible nodes */ if (!node_possible(node)) { - scx_ops_error("unavailable node %d", node); + scx_error("unavailable node %d", node); return -EINVAL; } @@ -778,7 +849,7 @@ static bool check_builtin_idle_enabled(void) if (static_branch_likely(&scx_builtin_idle_enabled)) return true; - scx_ops_error("built-in idle tracking is disabled"); + scx_error("built-in idle tracking is disabled"); return false; } @@ -829,7 +900,7 @@ __bpf_kfunc s32 scx_bpf_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, goto prev_cpu; #ifdef CONFIG_SMP - cpu = scx_select_cpu_dfl(p, prev_cpu, wake_flags, 0); + cpu = scx_select_cpu_dfl(p, prev_cpu, wake_flags, NULL, 0); if (cpu >= 0) { *is_idle = true; return cpu; @@ -842,13 +913,67 @@ prev_cpu: } /** + * scx_bpf_select_cpu_and - Pick an idle CPU usable by task @p, + * prioritizing those in @cpus_allowed + * @p: task_struct to select a CPU for + * @prev_cpu: CPU @p was on previously + * @wake_flags: %SCX_WAKE_* flags + * @cpus_allowed: cpumask of allowed CPUs + * @flags: %SCX_PICK_IDLE* flags + * + * Can only be called from ops.select_cpu() or ops.enqueue() if the + * built-in CPU selection is enabled: ops.update_idle() is missing or + * %SCX_OPS_KEEP_BUILTIN_IDLE is set. + * + * @p, @prev_cpu and @wake_flags match ops.select_cpu(). + * + * Returns the selected idle CPU, which will be automatically awakened upon + * returning from ops.select_cpu() and can be used for direct dispatch, or + * a negative value if no idle CPU is available. + */ +__bpf_kfunc s32 scx_bpf_select_cpu_and(struct task_struct *p, s32 prev_cpu, u64 wake_flags, + const struct cpumask *cpus_allowed, u64 flags) +{ + s32 cpu; + + if (!ops_cpu_valid(prev_cpu, NULL)) + return -EINVAL; + + if (!check_builtin_idle_enabled()) + return -EBUSY; + + if (!scx_kf_allowed(SCX_KF_SELECT_CPU | SCX_KF_ENQUEUE)) + return -EPERM; + +#ifdef CONFIG_SMP + /* + * This may also be called from ops.enqueue(), so we need to handle + * per-CPU tasks as well. For these tasks, we can skip all idle CPU + * selection optimizations and simply check whether the previously + * used CPU is idle and within the allowed cpumask. + */ + if (p->nr_cpus_allowed == 1) { + if (cpumask_test_cpu(prev_cpu, cpus_allowed) && + scx_idle_test_and_clear_cpu(prev_cpu)) + return prev_cpu; + return -EBUSY; + } + cpu = scx_select_cpu_dfl(p, prev_cpu, wake_flags, cpus_allowed, flags); +#else + cpu = -EBUSY; +#endif + + return cpu; +} + +/** * scx_bpf_get_idle_cpumask_node - Get a referenced kptr to the * idle-tracking per-CPU cpumask of a target NUMA node. * @node: target NUMA node * * Returns an empty cpumask if idle tracking is not enabled, if @node is * not valid, or running on a UP kernel. In this case the actual error will - * be reported to the BPF scheduler via scx_ops_error(). + * be reported to the BPF scheduler via scx_error(). */ __bpf_kfunc const struct cpumask *scx_bpf_get_idle_cpumask_node(int node) { @@ -873,7 +998,7 @@ __bpf_kfunc const struct cpumask *scx_bpf_get_idle_cpumask_node(int node) __bpf_kfunc const struct cpumask *scx_bpf_get_idle_cpumask(void) { if (static_branch_unlikely(&scx_builtin_idle_per_node)) { - scx_ops_error("SCX_OPS_BUILTIN_IDLE_PER_NODE enabled"); + scx_error("SCX_OPS_BUILTIN_IDLE_PER_NODE enabled"); return cpu_none_mask; } @@ -895,7 +1020,7 @@ __bpf_kfunc const struct cpumask *scx_bpf_get_idle_cpumask(void) * * Returns an empty cpumask if idle tracking is not enabled, if @node is * not valid, or running on a UP kernel. In this case the actual error will - * be reported to the BPF scheduler via scx_ops_error(). + * be reported to the BPF scheduler via scx_error(). */ __bpf_kfunc const struct cpumask *scx_bpf_get_idle_smtmask_node(int node) { @@ -924,7 +1049,7 @@ __bpf_kfunc const struct cpumask *scx_bpf_get_idle_smtmask_node(int node) __bpf_kfunc const struct cpumask *scx_bpf_get_idle_smtmask(void) { if (static_branch_unlikely(&scx_builtin_idle_per_node)) { - scx_ops_error("SCX_OPS_BUILTIN_IDLE_PER_NODE enabled"); + scx_error("SCX_OPS_BUILTIN_IDLE_PER_NODE enabled"); return cpu_none_mask; } @@ -1032,7 +1157,7 @@ __bpf_kfunc s32 scx_bpf_pick_idle_cpu(const struct cpumask *cpus_allowed, u64 flags) { if (static_branch_maybe(CONFIG_NUMA, &scx_builtin_idle_per_node)) { - scx_ops_error("per-node idle tracking is enabled"); + scx_error("per-node idle tracking is enabled"); return -EBUSY; } @@ -1109,7 +1234,7 @@ __bpf_kfunc s32 scx_bpf_pick_any_cpu(const struct cpumask *cpus_allowed, s32 cpu; if (static_branch_maybe(CONFIG_NUMA, &scx_builtin_idle_per_node)) { - scx_ops_error("per-node idle tracking is enabled"); + scx_error("per-node idle tracking is enabled"); return -EBUSY; } @@ -1149,6 +1274,7 @@ static const struct btf_kfunc_id_set scx_kfunc_set_idle = { BTF_KFUNCS_START(scx_kfunc_ids_select_cpu) BTF_ID_FLAGS(func, scx_bpf_select_cpu_dfl, KF_RCU) +BTF_ID_FLAGS(func, scx_bpf_select_cpu_and, KF_RCU) BTF_KFUNCS_END(scx_kfunc_ids_select_cpu) static const struct btf_kfunc_id_set scx_kfunc_set_select_cpu = { diff --git a/kernel/sched/ext_idle.h b/kernel/sched/ext_idle.h index 511cc2221f7a..37be78a7502b 100644 --- a/kernel/sched/ext_idle.h +++ b/kernel/sched/ext_idle.h @@ -27,7 +27,8 @@ static inline s32 scx_pick_idle_cpu(const struct cpumask *cpus_allowed, int node } #endif /* CONFIG_SMP */ -s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags, u64 flags); +s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags, + const struct cpumask *cpus_allowed, u64 flags); void scx_idle_enable(struct sched_ext_ops *ops); void scx_idle_disable(void); int scx_idle_init(void); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 47972f34ea70..ac07f64c8f39 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1717,10 +1717,10 @@ extern struct balance_callback balance_push_callback; #ifdef CONFIG_SCHED_CLASS_EXT extern const struct sched_class ext_sched_class; -DECLARE_STATIC_KEY_FALSE(__scx_ops_enabled); /* SCX BPF scheduler loaded */ +DECLARE_STATIC_KEY_FALSE(__scx_enabled); /* SCX BPF scheduler loaded */ DECLARE_STATIC_KEY_FALSE(__scx_switched_all); /* all fair class tasks on SCX */ -#define scx_enabled() static_branch_unlikely(&__scx_ops_enabled) +#define scx_enabled() static_branch_unlikely(&__scx_enabled) #define scx_switched_all() static_branch_unlikely(&__scx_switched_all) static inline void scx_rq_clock_update(struct rq *rq, u64 clock) diff --git a/tools/sched_ext/include/scx/common.bpf.h b/tools/sched_ext/include/scx/common.bpf.h index 8787048c6762..d4e21558e982 100644 --- a/tools/sched_ext/include/scx/common.bpf.h +++ b/tools/sched_ext/include/scx/common.bpf.h @@ -48,6 +48,8 @@ static inline void ___vmlinux_h_sanity_check___(void) s32 scx_bpf_create_dsq(u64 dsq_id, s32 node) __ksym; s32 scx_bpf_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags, bool *is_idle) __ksym; +s32 scx_bpf_select_cpu_and(struct task_struct *p, s32 prev_cpu, u64 wake_flags, + const struct cpumask *cpus_allowed, u64 flags) __ksym __weak; void scx_bpf_dsq_insert(struct task_struct *p, u64 dsq_id, u64 slice, u64 enq_flags) __ksym __weak; void scx_bpf_dsq_insert_vtime(struct task_struct *p, u64 dsq_id, u64 slice, u64 vtime, u64 enq_flags) __ksym __weak; u32 scx_bpf_dispatch_nr_slots(void) __ksym; diff --git a/tools/sched_ext/scx_show_state.py b/tools/sched_ext/scx_show_state.py index b800d4f5f2e9..7cdcc6729ea4 100644 --- a/tools/sched_ext/scx_show_state.py +++ b/tools/sched_ext/scx_show_state.py @@ -24,19 +24,19 @@ def read_atomic(name): def read_static_key(name): return prog[name].key.enabled.counter.value_() -def ops_state_str(state): - return prog['scx_ops_enable_state_str'][state].string_().decode() +def state_str(state): + return prog['scx_enable_state_str'][state].string_().decode() ops = prog['scx_ops'] -enable_state = read_atomic("scx_ops_enable_state_var") +enable_state = read_atomic("scx_enable_state_var") print(f'ops : {ops.name.string_().decode()}') -print(f'enabled : {read_static_key("__scx_ops_enabled")}') +print(f'enabled : {read_static_key("__scx_enabled")}') print(f'switching_all : {read_int("scx_switching_all")}') print(f'switched_all : {read_static_key("__scx_switched_all")}') -print(f'enable_state : {ops_state_str(enable_state)} ({enable_state})') +print(f'enable_state : {state_str(enable_state)} ({enable_state})') print(f'in_softlockup : {prog["scx_in_softlockup"].value_()}') -print(f'breather_depth: {read_atomic("scx_ops_breather_depth")}') -print(f'bypass_depth : {prog["scx_ops_bypass_depth"].value_()}') +print(f'breather_depth: {read_atomic("scx_breather_depth")}') +print(f'bypass_depth : {prog["scx_bypass_depth"].value_()}') print(f'nr_rejected : {read_atomic("scx_nr_rejected")}') print(f'enable_seq : {read_atomic("scx_enable_seq")}') diff --git a/tools/testing/selftests/sched_ext/Makefile b/tools/testing/selftests/sched_ext/Makefile index f4531327b8e7..e9d5bc575f80 100644 --- a/tools/testing/selftests/sched_ext/Makefile +++ b/tools/testing/selftests/sched_ext/Makefile @@ -173,6 +173,7 @@ auto-test-targets := \ maybe_null \ minimal \ numa \ + allowed_cpus \ prog_run \ reload_loop \ select_cpu_dfl \ diff --git a/tools/testing/selftests/sched_ext/allowed_cpus.bpf.c b/tools/testing/selftests/sched_ext/allowed_cpus.bpf.c new file mode 100644 index 000000000000..39d57f7f7409 --- /dev/null +++ b/tools/testing/selftests/sched_ext/allowed_cpus.bpf.c @@ -0,0 +1,121 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * A scheduler that validates the behavior of scx_bpf_select_cpu_and() by + * selecting idle CPUs strictly within a subset of allowed CPUs. + * + * Copyright (c) 2025 Andrea Righi <arighi@nvidia.com> + */ + +#include <scx/common.bpf.h> + +char _license[] SEC("license") = "GPL"; + +UEI_DEFINE(uei); + +private(PREF_CPUS) struct bpf_cpumask __kptr * allowed_cpumask; + +static void +validate_idle_cpu(const struct task_struct *p, const struct cpumask *allowed, s32 cpu) +{ + if (scx_bpf_test_and_clear_cpu_idle(cpu)) + scx_bpf_error("CPU %d should be marked as busy", cpu); + + if (bpf_cpumask_subset(allowed, p->cpus_ptr) && + !bpf_cpumask_test_cpu(cpu, allowed)) + scx_bpf_error("CPU %d not in the allowed domain for %d (%s)", + cpu, p->pid, p->comm); +} + +s32 BPF_STRUCT_OPS(allowed_cpus_select_cpu, + struct task_struct *p, s32 prev_cpu, u64 wake_flags) +{ + const struct cpumask *allowed; + s32 cpu; + + allowed = cast_mask(allowed_cpumask); + if (!allowed) { + scx_bpf_error("allowed domain not initialized"); + return -EINVAL; + } + + /* + * Select an idle CPU strictly within the allowed domain. + */ + cpu = scx_bpf_select_cpu_and(p, prev_cpu, wake_flags, allowed, 0); + if (cpu >= 0) { + validate_idle_cpu(p, allowed, cpu); + scx_bpf_dsq_insert(p, SCX_DSQ_LOCAL, SCX_SLICE_DFL, 0); + + return cpu; + } + + return prev_cpu; +} + +void BPF_STRUCT_OPS(allowed_cpus_enqueue, struct task_struct *p, u64 enq_flags) +{ + const struct cpumask *allowed; + s32 prev_cpu = scx_bpf_task_cpu(p), cpu; + + scx_bpf_dsq_insert(p, SCX_DSQ_GLOBAL, SCX_SLICE_DFL, 0); + + allowed = cast_mask(allowed_cpumask); + if (!allowed) { + scx_bpf_error("allowed domain not initialized"); + return; + } + + /* + * Use scx_bpf_select_cpu_and() to proactively kick an idle CPU + * within @allowed_cpumask, usable by @p. + */ + cpu = scx_bpf_select_cpu_and(p, prev_cpu, 0, allowed, 0); + if (cpu >= 0) { + validate_idle_cpu(p, allowed, cpu); + scx_bpf_kick_cpu(cpu, SCX_KICK_IDLE); + } +} + +s32 BPF_STRUCT_OPS_SLEEPABLE(allowed_cpus_init) +{ + struct bpf_cpumask *mask; + + mask = bpf_cpumask_create(); + if (!mask) + return -ENOMEM; + + mask = bpf_kptr_xchg(&allowed_cpumask, mask); + if (mask) + bpf_cpumask_release(mask); + + bpf_rcu_read_lock(); + + /* + * Assign the first online CPU to the allowed domain. + */ + mask = allowed_cpumask; + if (mask) { + const struct cpumask *online = scx_bpf_get_online_cpumask(); + + bpf_cpumask_set_cpu(bpf_cpumask_first(online), mask); + scx_bpf_put_cpumask(online); + } + + bpf_rcu_read_unlock(); + + return 0; +} + +void BPF_STRUCT_OPS(allowed_cpus_exit, struct scx_exit_info *ei) +{ + UEI_RECORD(uei, ei); +} + +SEC(".struct_ops.link") +struct sched_ext_ops allowed_cpus_ops = { + .select_cpu = (void *)allowed_cpus_select_cpu, + .enqueue = (void *)allowed_cpus_enqueue, + .init = (void *)allowed_cpus_init, + .exit = (void *)allowed_cpus_exit, + .name = "allowed_cpus", +}; diff --git a/tools/testing/selftests/sched_ext/allowed_cpus.c b/tools/testing/selftests/sched_ext/allowed_cpus.c new file mode 100644 index 000000000000..a001a3a0e9f1 --- /dev/null +++ b/tools/testing/selftests/sched_ext/allowed_cpus.c @@ -0,0 +1,57 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Copyright (c) 2025 Andrea Righi <arighi@nvidia.com> + */ +#include <bpf/bpf.h> +#include <scx/common.h> +#include <sys/wait.h> +#include <unistd.h> +#include "allowed_cpus.bpf.skel.h" +#include "scx_test.h" + +static enum scx_test_status setup(void **ctx) +{ + struct allowed_cpus *skel; + + skel = allowed_cpus__open(); + SCX_FAIL_IF(!skel, "Failed to open"); + SCX_ENUM_INIT(skel); + SCX_FAIL_IF(allowed_cpus__load(skel), "Failed to load skel"); + + *ctx = skel; + + return SCX_TEST_PASS; +} + +static enum scx_test_status run(void *ctx) +{ + struct allowed_cpus *skel = ctx; + struct bpf_link *link; + + link = bpf_map__attach_struct_ops(skel->maps.allowed_cpus_ops); + SCX_FAIL_IF(!link, "Failed to attach scheduler"); + + /* Just sleeping is fine, plenty of scheduling events happening */ + sleep(1); + + SCX_EQ(skel->data->uei.kind, EXIT_KIND(SCX_EXIT_NONE)); + bpf_link__destroy(link); + + return SCX_TEST_PASS; +} + +static void cleanup(void *ctx) +{ + struct allowed_cpus *skel = ctx; + + allowed_cpus__destroy(skel); +} + +struct scx_test allowed_cpus = { + .name = "allowed_cpus", + .description = "Verify scx_bpf_select_cpu_and()", + .setup = setup, + .run = run, + .cleanup = cleanup, +}; +REGISTER_SCX_TEST(&allowed_cpus) |