From 740cf8a760b73e8375bfb4bedcbe9746183350f9 Mon Sep 17 00:00:00 2001 From: Dietmar Eggemann Date: Fri, 29 Jul 2022 13:13:03 +0200 Subject: [PATCH 01/41] sched/core: Introduce sched_asym_cpucap_active() Create an inline helper for conditional code to be only executed on asymmetric CPU capacity systems. This makes these (currently ~10 and future) conditions a lot more readable. Signed-off-by: Dietmar Eggemann Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20220729111305.1275158-2-dietmar.eggemann@arm.com --- kernel/sched/cpudeadline.c | 2 +- kernel/sched/deadline.c | 4 ++-- kernel/sched/fair.c | 8 ++++---- kernel/sched/rt.c | 4 ++-- kernel/sched/sched.h | 5 +++++ 5 files changed, 14 insertions(+), 9 deletions(-) diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c index 02d970a879ed..57c92d751bcd 100644 --- a/kernel/sched/cpudeadline.c +++ b/kernel/sched/cpudeadline.c @@ -123,7 +123,7 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p, unsigned long cap, max_cap = 0; int cpu, max_cpu = -1; - if (!static_branch_unlikely(&sched_asym_cpucapacity)) + if (!sched_asym_cpucap_active()) return 1; /* Ensure the capacity of the CPUs fits the task. */ diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 0ab79d819a0d..8bebc36a1b71 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -144,7 +144,7 @@ static inline unsigned long __dl_bw_capacity(int i) */ static inline unsigned long dl_bw_capacity(int i) { - if (!static_branch_unlikely(&sched_asym_cpucapacity) && + if (!sched_asym_cpucap_active() && capacity_orig_of(i) == SCHED_CAPACITY_SCALE) { return dl_bw_cpus(i) << SCHED_CAPACITY_SHIFT; } else { @@ -1849,7 +1849,7 @@ select_task_rq_dl(struct task_struct *p, int cpu, int flags) * Take the capacity of the CPU into account to * ensure it fits the requirement of the task. */ - if (static_branch_unlikely(&sched_asym_cpucapacity)) + if (sched_asym_cpucap_active()) select_rq |= !dl_task_fits_capacity(p, cpu); if (select_rq) { diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 914096c5b1ae..41486d9b0911 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4262,7 +4262,7 @@ static inline int task_fits_capacity(struct task_struct *p, static inline void update_misfit_status(struct task_struct *p, struct rq *rq) { - if (!static_branch_unlikely(&sched_asym_cpucapacity)) + if (!sched_asym_cpucap_active()) return; if (!p || p->nr_cpus_allowed == 1) { @@ -6506,7 +6506,7 @@ select_idle_capacity(struct task_struct *p, struct sched_domain *sd, int target) static inline bool asym_fits_capacity(unsigned long task_util, int cpu) { - if (static_branch_unlikely(&sched_asym_cpucapacity)) + if (sched_asym_cpucap_active()) return fits_capacity(task_util, capacity_of(cpu)); return true; @@ -6526,7 +6526,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) * On asymmetric system, update task utilization because we will check * that the task fits with cpu's capacity. */ - if (static_branch_unlikely(&sched_asym_cpucapacity)) { + if (sched_asym_cpucap_active()) { sync_entity_load_avg(&p->se); task_util = uclamp_task_util(p); } @@ -6580,7 +6580,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) * For asymmetric CPU capacity systems, our domain of interest is * sd_asym_cpucapacity rather than sd_llc. */ - if (static_branch_unlikely(&sched_asym_cpucapacity)) { + if (sched_asym_cpucap_active()) { sd = rcu_dereference(per_cpu(sd_asym_cpucapacity, target)); /* * On an asymmetric CPU capacity system where an exclusive diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 55f39c8f4203..054b6711e961 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -509,7 +509,7 @@ static inline bool rt_task_fits_capacity(struct task_struct *p, int cpu) unsigned int cpu_cap; /* Only heterogeneous systems can benefit from this check */ - if (!static_branch_unlikely(&sched_asym_cpucapacity)) + if (!sched_asym_cpucap_active()) return true; min_cap = uclamp_eff_value(p, UCLAMP_MIN); @@ -1897,7 +1897,7 @@ static int find_lowest_rq(struct task_struct *task) * If we're on asym system ensure we consider the different capacities * of the CPUs when searching for the lowest_mask. */ - if (static_branch_unlikely(&sched_asym_cpucapacity)) { + if (sched_asym_cpucap_active()) { ret = cpupri_find_fitness(&task_rq(task)->rd->cpupri, task, lowest_mask, diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index aad7f5ee9666..86fc36069f38 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1813,6 +1813,11 @@ DECLARE_PER_CPU(struct sched_domain __rcu *, sd_asym_packing); DECLARE_PER_CPU(struct sched_domain __rcu *, sd_asym_cpucapacity); extern struct static_key_false sched_asym_cpucapacity; +static __always_inline bool sched_asym_cpucap_active(void) +{ + return static_branch_unlikely(&sched_asym_cpucapacity); +} + struct sched_group_capacity { atomic_t ref; /* From 6092478bcbf4021d3c52144bd456378de043fe58 Mon Sep 17 00:00:00 2001 From: Dietmar Eggemann Date: Fri, 29 Jul 2022 13:13:04 +0200 Subject: [PATCH 02/41] sched/deadline: Make dl_cpuset_cpumask_can_shrink() capacity-aware dl_cpuset_cpumask_can_shrink() is used to validate whether there is still enough CPU capacity for DL tasks in the reduced cpuset. Currently it still operates on `# remaining CPUs in the cpuset` (1). Change this to use the already capacity-aware DL admission control __dl_overflow() for the `cpumask can shrink` test. dl_b->bw = sched_rt_period << BW_SHIFT / sched_rt_period dl_b->bw * (1) >= currently allocated bandwidth in root_domain (rd) Replace (1) w/ `\Sum CPU capacity in rd >> SCHED_CAPACITY_SHIFT` Adapt __dl_bw_capacity() to take a cpumask instead of a CPU number argument so that `rd->span` and `cpumask of the reduced cpuset` can be used here. Signed-off-by: Dietmar Eggemann Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20220729111305.1275158-3-dietmar.eggemann@arm.com --- kernel/sched/deadline.c | 24 +++++++++++------------- 1 file changed, 11 insertions(+), 13 deletions(-) diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 8bebc36a1b71..1d9c90958baa 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -124,15 +124,12 @@ static inline int dl_bw_cpus(int i) return cpus; } -static inline unsigned long __dl_bw_capacity(int i) +static inline unsigned long __dl_bw_capacity(const struct cpumask *mask) { - struct root_domain *rd = cpu_rq(i)->rd; unsigned long cap = 0; + int i; - RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held(), - "sched RCU must be held"); - - for_each_cpu_and(i, rd->span, cpu_active_mask) + for_each_cpu_and(i, mask, cpu_active_mask) cap += capacity_orig_of(i); return cap; @@ -148,7 +145,10 @@ static inline unsigned long dl_bw_capacity(int i) capacity_orig_of(i) == SCHED_CAPACITY_SCALE) { return dl_bw_cpus(i) << SCHED_CAPACITY_SHIFT; } else { - return __dl_bw_capacity(i); + RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held(), + "sched RCU must be held"); + + return __dl_bw_capacity(cpu_rq(i)->rd->span); } } @@ -3007,17 +3007,15 @@ bool dl_param_changed(struct task_struct *p, const struct sched_attr *attr) int dl_cpuset_cpumask_can_shrink(const struct cpumask *cur, const struct cpumask *trial) { - int ret = 1, trial_cpus; + unsigned long flags, cap; struct dl_bw *cur_dl_b; - unsigned long flags; + int ret = 1; rcu_read_lock_sched(); cur_dl_b = dl_bw_of(cpumask_any(cur)); - trial_cpus = cpumask_weight(trial); - + cap = __dl_bw_capacity(trial); raw_spin_lock_irqsave(&cur_dl_b->lock, flags); - if (cur_dl_b->bw != -1 && - cur_dl_b->bw * trial_cpus < cur_dl_b->total_bw) + if (__dl_overflow(cur_dl_b, cap, 0, 0)) ret = 0; raw_spin_unlock_irqrestore(&cur_dl_b->lock, flags); rcu_read_unlock_sched(); From b3f53daacc74c0ca0922e14a8cb793cb2db4c6d1 Mon Sep 17 00:00:00 2001 From: Dietmar Eggemann Date: Fri, 29 Jul 2022 13:13:05 +0200 Subject: [PATCH 03/41] sched/deadline: Use sched_dl_entity's dl_density in dl_task_fits_capacity() Save a multiplication in dl_task_fits_capacity() by using already maintained per-sched_dl_entity (i.e. per-task) `dl_runtime/dl_deadline` (dl_density). cap_scale(dl_deadline, cap) >= dl_runtime dl_deadline * cap >> SCHED_CAPACITY_SHIFT >= dl_runtime cap >= dl_runtime << SCHED_CAPACITY_SHIFT / dl_deadline cap >= (dl_runtime << BW_SHIFT / dl_deadline) >> BW_SHIFT - SCHED_CAPACITY_SHIFT cap >= dl_density >> BW_SHIFT - SCHED_CAPACITY_SHIFT __sched_setscheduler()->__checkparam_dl() ensures that the 2 corner cases (if conditions) `runtime == RUNTIME_INF (-1)` and `period == 0` of to_ratio(deadline, runtime) are not met when setting dl_density in __sched_setscheduler()-> __setscheduler_params()->__setparam_dl(). Signed-off-by: Dietmar Eggemann Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20220729111305.1275158-4-dietmar.eggemann@arm.com --- kernel/sched/sched.h | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 86fc36069f38..3ccd35c22f0f 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -320,21 +320,6 @@ struct dl_bw { u64 total_bw; }; -/* - * Verify the fitness of task @p to run on @cpu taking into account the - * CPU original capacity and the runtime/deadline ratio of the task. - * - * The function will return true if the CPU original capacity of the - * @cpu scaled by SCHED_CAPACITY_SCALE >= runtime/deadline ratio of the - * task and false otherwise. - */ -static inline bool dl_task_fits_capacity(struct task_struct *p, int cpu) -{ - unsigned long cap = arch_scale_cpu_capacity(cpu); - - return cap_scale(p->dl.dl_deadline, cap) >= p->dl.dl_runtime; -} - extern void init_dl_bw(struct dl_bw *dl_b); extern int sched_dl_global_validate(void); extern void sched_dl_do_global(void); @@ -2899,6 +2884,21 @@ unsigned long effective_cpu_util(int cpu, unsigned long util_cfs, enum cpu_util_type type, struct task_struct *p); +/* + * Verify the fitness of task @p to run on @cpu taking into account the + * CPU original capacity and the runtime/deadline ratio of the task. + * + * The function will return true if the original capacity of @cpu is + * greater than or equal to task's deadline density right shifted by + * (BW_SHIFT - SCHED_CAPACITY_SHIFT) and false otherwise. + */ +static inline bool dl_task_fits_capacity(struct task_struct *p, int cpu) +{ + unsigned long cap = arch_scale_cpu_capacity(cpu); + + return cap >= p->dl.dl_density >> (BW_SHIFT - SCHED_CAPACITY_SHIFT); +} + static inline unsigned long cpu_bw_dl(struct rq *rq) { return (rq->dl.running_bw * SCHED_CAPACITY_SCALE) >> BW_SHIFT; From 0f03d6805bfc454279169a1460abb3f6b3db317f Mon Sep 17 00:00:00 2001 From: Zhen Lei Date: Wed, 27 Jul 2022 14:08:19 +0800 Subject: [PATCH 04/41] sched/debug: Print each field value left-aligned in sched_show_task() Currently, the values of some fields are printed right-aligned, causing the field value to be next to the next field name rather than next to its own field name. So print each field value left-aligned, to make it more readable. Before: stack: 0 pid: 307 ppid: 2 flags:0x00000008 After: stack:0 pid:308 ppid:2 flags:0x0000000a This also makes them print in the same style as the other two fields: task:demo0 state:R running task Signed-off-by: Zhen Lei Signed-off-by: Ingo Molnar Reviewed-by: Valentin Schneider Link: https://lore.kernel.org/r/20220727060819.1085-1-thunder.leizhen@huawei.com --- kernel/sched/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 5555e49c4e12..6785e3be0769 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -8825,7 +8825,7 @@ void sched_show_task(struct task_struct *p) if (pid_alive(p)) ppid = task_pid_nr(rcu_dereference(p->real_parent)); rcu_read_unlock(); - pr_cont(" stack:%5lu pid:%5d ppid:%6d flags:0x%08lx\n", + pr_cont(" stack:%-5lu pid:%-5d ppid:%-6d flags:0x%08lx\n", free, task_pid_nr(p), ppid, read_task_thread_flags(p)); From d985ee9f449aafe45397ae1cf8887e9ac91d2f95 Mon Sep 17 00:00:00 2001 From: Hao Jia Date: Wed, 3 Aug 2022 21:02:23 +0800 Subject: [PATCH 05/41] sched/fair: Remove unused parameter idle of _nohz_idle_balance() After commit 7a82e5f52a35 ("sched/fair: Merge for each idle cpu loop of ILB"), _nohz_idle_balance()'s 'idle' parameter is not used anymore, so we can remove it. Signed-off-by: Hao Jia Signed-off-by: Ingo Molnar Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20220803130223.70419-1-jiahao.os@bytedance.com --- kernel/sched/fair.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 41486d9b0911..d22c5e82d4a6 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -10916,8 +10916,7 @@ static bool update_nohz_stats(struct rq *rq) * can be a simple update of blocked load or a complete load balance with * tasks movement depending of flags. */ -static void _nohz_idle_balance(struct rq *this_rq, unsigned int flags, - enum cpu_idle_type idle) +static void _nohz_idle_balance(struct rq *this_rq, unsigned int flags) { /* Earliest time when we have to do rebalance again */ unsigned long now = jiffies; @@ -11032,7 +11031,7 @@ static bool nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) if (idle != CPU_IDLE) return false; - _nohz_idle_balance(this_rq, flags, idle); + _nohz_idle_balance(this_rq, flags); return true; } @@ -11052,7 +11051,7 @@ void nohz_run_idle_balance(int cpu) * (ie NOHZ_STATS_KICK set) and will do the same. */ if ((flags == NOHZ_NEWILB_KICK) && !need_resched()) - _nohz_idle_balance(cpu_rq(cpu), NOHZ_STATS_KICK, CPU_IDLE); + _nohz_idle_balance(cpu_rq(cpu), NOHZ_STATS_KICK); } static void nohz_newidle_balance(struct rq *this_rq) From 18c31c9711a90b48a77b78afb65012d9feec444c Mon Sep 17 00:00:00 2001 From: Bing Huang Date: Sat, 23 Jul 2022 05:36:09 +0800 Subject: [PATCH 06/41] sched/fair: Make per-cpu cpumasks static The load_balance_mask and select_rq_mask percpu variables are only used in kernel/sched/fair.c. Make them static and move their allocation into init_sched_fair_class(). Replace kzalloc_node() with zalloc_cpumask_var_node() to get rid of the CONFIG_CPUMASK_OFFSTACK #ifdef and to align with per-cpu cpumask allocation for RT (local_cpu_mask in init_sched_rt_class()) and DL class (local_cpu_mask_dl in init_sched_dl_class()). [ mingo: Tidied up changelog & touched up the code. ] Signed-off-by: Bing Huang Signed-off-by: Ingo Molnar Reviewed-by: Dietmar Eggemann Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20220722213609.3901-1-huangbing775@126.com --- kernel/sched/core.c | 11 ----------- kernel/sched/fair.c | 11 +++++++++-- 2 files changed, 9 insertions(+), 13 deletions(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 6785e3be0769..64c08993221b 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -9563,9 +9563,6 @@ LIST_HEAD(task_groups); static struct kmem_cache *task_group_cache __read_mostly; #endif -DECLARE_PER_CPU(cpumask_var_t, load_balance_mask); -DECLARE_PER_CPU(cpumask_var_t, select_rq_mask); - void __init sched_init(void) { unsigned long ptr = 0; @@ -9609,14 +9606,6 @@ void __init sched_init(void) #endif /* CONFIG_RT_GROUP_SCHED */ } -#ifdef CONFIG_CPUMASK_OFFSTACK - for_each_possible_cpu(i) { - per_cpu(load_balance_mask, i) = (cpumask_var_t)kzalloc_node( - cpumask_size(), GFP_KERNEL, cpu_to_node(i)); - per_cpu(select_rq_mask, i) = (cpumask_var_t)kzalloc_node( - cpumask_size(), GFP_KERNEL, cpu_to_node(i)); - } -#endif /* CONFIG_CPUMASK_OFFSTACK */ init_rt_bandwidth(&def_rt_bandwidth, global_rt_period(), global_rt_runtime()); diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index d22c5e82d4a6..da388657d5ac 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5893,8 +5893,8 @@ dequeue_throttle: #ifdef CONFIG_SMP /* Working cpumask for: load_balance, load_balance_newidle. */ -DEFINE_PER_CPU(cpumask_var_t, load_balance_mask); -DEFINE_PER_CPU(cpumask_var_t, select_rq_mask); +static DEFINE_PER_CPU(cpumask_var_t, load_balance_mask); +static DEFINE_PER_CPU(cpumask_var_t, select_rq_mask); #ifdef CONFIG_NO_HZ_COMMON @@ -12074,6 +12074,13 @@ void show_numa_stats(struct task_struct *p, struct seq_file *m) __init void init_sched_fair_class(void) { #ifdef CONFIG_SMP + int i; + + for_each_possible_cpu(i) { + zalloc_cpumask_var_node(&per_cpu(load_balance_mask, i), GFP_KERNEL, cpu_to_node(i)); + zalloc_cpumask_var_node(&per_cpu(select_rq_mask, i), GFP_KERNEL, cpu_to_node(i)); + } + open_softirq(SCHED_SOFTIRQ, run_rebalance_domains); #ifdef CONFIG_NO_HZ_COMMON From 8648f92a66a323ed01903d2cbb248cdbe2f312d9 Mon Sep 17 00:00:00 2001 From: Xin Gao Date: Tue, 19 Jul 2022 19:10:44 +0800 Subject: [PATCH 07/41] sched/core: Remove superfluous semicolon Signed-off-by: Xin Gao Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20220719111044.7095-1-gaoxin@cdjrlc.com --- kernel/sched/core_sched.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sched/core_sched.c b/kernel/sched/core_sched.c index 93878cb2a46d..1ec807fbde56 100644 --- a/kernel/sched/core_sched.c +++ b/kernel/sched/core_sched.c @@ -205,7 +205,7 @@ int sched_core_share_pid(unsigned int cmd, pid_t pid, enum pid_type type, default: err = -EINVAL; goto out; - }; + } if (type == PIDTYPE_PID) { __sched_core_set(task, cookie); From 09348d75a6ce60eec85c86dd0ab7babc4db3caf6 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 11 Aug 2022 08:54:52 +0200 Subject: [PATCH 08/41] sched/all: Change all BUG_ON() instances in the scheduler to WARN_ON_ONCE() There's no good reason to crash a user's system with a BUG_ON(), chances are high that they'll never even see the crash message on Xorg, and it won't make it into the syslog either. By using a WARN_ON_ONCE() we at least give the user a chance to report any bugs triggered here - instead of getting silent hangs. None of these WARN_ON_ONCE()s are supposed to trigger, ever - so we ignore cases where a NULL check is done via a BUG_ON() and we let a NULL pointer through after a WARN_ON_ONCE(). There's one exception: WARN_ON_ONCE() arguments with side-effects, such as locking - in this case we use the return value of the WARN_ON_ONCE(), such as in: - BUG_ON(!lock_task_sighand(p, &flags)); + if (WARN_ON_ONCE(!lock_task_sighand(p, &flags))) + return; Suggested-by: Linus Torvalds Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/YvSsKcAXISmshtHo@gmail.com --- kernel/sched/autogroup.c | 3 ++- kernel/sched/core.c | 2 +- kernel/sched/cpupri.c | 2 +- kernel/sched/deadline.c | 26 +++++++++++++------------- kernel/sched/fair.c | 10 +++++----- kernel/sched/rt.c | 2 +- kernel/sched/sched.h | 6 +++--- 7 files changed, 26 insertions(+), 25 deletions(-) diff --git a/kernel/sched/autogroup.c b/kernel/sched/autogroup.c index 4ebaf97f7bd8..991fc9002535 100644 --- a/kernel/sched/autogroup.c +++ b/kernel/sched/autogroup.c @@ -161,7 +161,8 @@ autogroup_move_group(struct task_struct *p, struct autogroup *ag) struct task_struct *t; unsigned long flags; - BUG_ON(!lock_task_sighand(p, &flags)); + if (WARN_ON_ONCE(!lock_task_sighand(p, &flags))) + return; prev = p->signal->autogroup; if (prev == ag) { diff --git a/kernel/sched/core.c b/kernel/sched/core.c index ee28253c9ac0..813687a5f5cf 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2328,7 +2328,7 @@ static struct rq *move_queued_task(struct rq *rq, struct rq_flags *rf, rq = cpu_rq(new_cpu); rq_lock(rq, rf); - BUG_ON(task_cpu(p) != new_cpu); + WARN_ON_ONCE(task_cpu(p) != new_cpu); activate_task(rq, p, 0); check_preempt_curr(rq, p, 0); diff --git a/kernel/sched/cpupri.c b/kernel/sched/cpupri.c index fa9ce9d83683..a286e726eb4b 100644 --- a/kernel/sched/cpupri.c +++ b/kernel/sched/cpupri.c @@ -147,7 +147,7 @@ int cpupri_find_fitness(struct cpupri *cp, struct task_struct *p, int task_pri = convert_prio(p->prio); int idx, cpu; - BUG_ON(task_pri >= CPUPRI_NR_PRIORITIES); + WARN_ON_ONCE(task_pri >= CPUPRI_NR_PRIORITIES); for (idx = 0; idx < task_pri; idx++) { diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 0ab79d819a0d..962b169b05cf 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -310,7 +310,7 @@ static void dl_change_utilization(struct task_struct *p, u64 new_bw) { struct rq *rq; - BUG_ON(p->dl.flags & SCHED_FLAG_SUGOV); + WARN_ON_ONCE(p->dl.flags & SCHED_FLAG_SUGOV); if (task_on_rq_queued(p)) return; @@ -607,7 +607,7 @@ static void enqueue_pushable_dl_task(struct rq *rq, struct task_struct *p) { struct rb_node *leftmost; - BUG_ON(!RB_EMPTY_NODE(&p->pushable_dl_tasks)); + WARN_ON_ONCE(!RB_EMPTY_NODE(&p->pushable_dl_tasks)); leftmost = rb_add_cached(&p->pushable_dl_tasks, &rq->dl.pushable_dl_tasks_root, @@ -684,7 +684,7 @@ static struct rq *dl_task_offline_migration(struct rq *rq, struct task_struct *p * Failed to find any suitable CPU. * The task will never come back! */ - BUG_ON(dl_bandwidth_enabled()); + WARN_ON_ONCE(dl_bandwidth_enabled()); /* * If admission control is disabled we @@ -830,7 +830,7 @@ static void replenish_dl_entity(struct sched_dl_entity *dl_se) struct dl_rq *dl_rq = dl_rq_of_se(dl_se); struct rq *rq = rq_of_dl_rq(dl_rq); - BUG_ON(pi_of(dl_se)->dl_runtime <= 0); + WARN_ON_ONCE(pi_of(dl_se)->dl_runtime <= 0); /* * This could be the case for a !-dl task that is boosted. @@ -1616,7 +1616,7 @@ static void __enqueue_dl_entity(struct sched_dl_entity *dl_se) { struct dl_rq *dl_rq = dl_rq_of_se(dl_se); - BUG_ON(!RB_EMPTY_NODE(&dl_se->rb_node)); + WARN_ON_ONCE(!RB_EMPTY_NODE(&dl_se->rb_node)); rb_add_cached(&dl_se->rb_node, &dl_rq->root, __dl_less); @@ -1640,7 +1640,7 @@ static void __dequeue_dl_entity(struct sched_dl_entity *dl_se) static void enqueue_dl_entity(struct sched_dl_entity *dl_se, int flags) { - BUG_ON(on_dl_rq(dl_se)); + WARN_ON_ONCE(on_dl_rq(dl_se)); update_stats_enqueue_dl(dl_rq_of_se(dl_se), dl_se, flags); @@ -2017,7 +2017,7 @@ static struct task_struct *pick_task_dl(struct rq *rq) return NULL; dl_se = pick_next_dl_entity(dl_rq); - BUG_ON(!dl_se); + WARN_ON_ONCE(!dl_se); p = dl_task_of(dl_se); return p; @@ -2277,12 +2277,12 @@ static struct task_struct *pick_next_pushable_dl_task(struct rq *rq) p = __node_2_pdl(rb_first_cached(&rq->dl.pushable_dl_tasks_root)); - BUG_ON(rq->cpu != task_cpu(p)); - BUG_ON(task_current(rq, p)); - BUG_ON(p->nr_cpus_allowed <= 1); + WARN_ON_ONCE(rq->cpu != task_cpu(p)); + WARN_ON_ONCE(task_current(rq, p)); + WARN_ON_ONCE(p->nr_cpus_allowed <= 1); - BUG_ON(!task_on_rq_queued(p)); - BUG_ON(!dl_task(p)); + WARN_ON_ONCE(!task_on_rq_queued(p)); + WARN_ON_ONCE(!dl_task(p)); return p; } @@ -2492,7 +2492,7 @@ static void set_cpus_allowed_dl(struct task_struct *p, struct root_domain *src_rd; struct rq *rq; - BUG_ON(!dl_task(p)); + WARN_ON_ONCE(!dl_task(p)); rq = task_rq(p); src_rd = rq->rd; diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 914096c5b1ae..28f10dccd194 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -2600,7 +2600,7 @@ static void task_numa_group(struct task_struct *p, int cpupid, int flags, if (!join) return; - BUG_ON(irqs_disabled()); + WARN_ON_ONCE(irqs_disabled()); double_lock_irq(&my_grp->lock, &grp->lock); for (i = 0; i < NR_NUMA_HINT_FAULT_STATS * nr_node_ids; i++) { @@ -7279,7 +7279,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ return; find_matching_se(&se, &pse); - BUG_ON(!pse); + WARN_ON_ONCE(!pse); cse_is_idle = se_is_idle(se); pse_is_idle = se_is_idle(pse); @@ -8159,7 +8159,7 @@ static void attach_task(struct rq *rq, struct task_struct *p) { lockdep_assert_rq_held(rq); - BUG_ON(task_rq(p) != rq); + WARN_ON_ONCE(task_rq(p) != rq); activate_task(rq, p, ENQUEUE_NOCLOCK); check_preempt_curr(rq, p, 0); } @@ -10134,7 +10134,7 @@ redo: goto out_balanced; } - BUG_ON(busiest == env.dst_rq); + WARN_ON_ONCE(busiest == env.dst_rq); schedstat_add(sd->lb_imbalance[idle], env.imbalance); @@ -10430,7 +10430,7 @@ static int active_load_balance_cpu_stop(void *data) * we need to fix it. Originally reported by * Bjorn Helgaas on a 128-CPU setup. */ - BUG_ON(busiest_rq == target_rq); + WARN_ON_ONCE(busiest_rq == target_rq); /* Search for an sd spanning us and the target CPU. */ rcu_read_lock(); diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 55f39c8f4203..2936fe55cef7 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -843,7 +843,7 @@ static void __disable_runtime(struct rq *rq) * We cannot be left wanting - that would mean some runtime * leaked out of the system. */ - BUG_ON(want); + WARN_ON_ONCE(want); balanced: /* * Disable all the borrow logic by pretending we have inf diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index e26688d387ae..7a44dceeb50a 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -2709,8 +2709,8 @@ static inline void double_rq_lock(struct rq *rq1, struct rq *rq2) __acquires(rq1->lock) __acquires(rq2->lock) { - BUG_ON(!irqs_disabled()); - BUG_ON(rq1 != rq2); + WARN_ON_ONCE(!irqs_disabled()); + WARN_ON_ONCE(rq1 != rq2); raw_spin_rq_lock(rq1); __acquire(rq2->lock); /* Fake it out ;) */ double_rq_clock_clear_update(rq1, rq2); @@ -2726,7 +2726,7 @@ static inline void double_rq_unlock(struct rq *rq1, struct rq *rq2) __releases(rq1->lock) __releases(rq2->lock) { - BUG_ON(rq1 != rq2); + WARN_ON_ONCE(rq1 != rq2); raw_spin_rq_unlock(rq1); __release(rq2->lock); } From 78b6b15770618efb60d84e2d605f6b93dc94051b Mon Sep 17 00:00:00 2001 From: Chengming Zhou Date: Thu, 18 Aug 2022 20:47:57 +0800 Subject: [PATCH 09/41] sched/fair: Maintain task se depth in set_task_rq() Previously we only maintain task se depth in task_move_group_fair(), if a !fair task change task group, its se depth will not be updated, so commit eb7a59b2c888 ("sched/fair: Reset se-depth when task switched to FAIR") fix the problem by updating se depth in switched_to_fair() too. Then commit daa59407b558 ("sched/fair: Unify switched_{from,to}_fair() and task_move_group_fair()") unified these two functions, moved se.depth setting to attach_task_cfs_rq(), which further into attach_entity_cfs_rq() with commit df217913e72e ("sched/fair: Factorize attach/detach entity"). This patch move task se depth maintenance from attach_entity_cfs_rq() to set_task_rq(), which will be called when CPU/cgroup change, so its depth will always be correct. This patch is preparation for the next patch. Signed-off-by: Chengming Zhou Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Dietmar Eggemann Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20220818124805.601-2-zhouchengming@bytedance.com --- kernel/sched/fair.c | 8 -------- kernel/sched/sched.h | 1 + 2 files changed, 1 insertion(+), 8 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index da388657d5ac..a3b0f8b1029e 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -11562,14 +11562,6 @@ static void attach_entity_cfs_rq(struct sched_entity *se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); -#ifdef CONFIG_FAIR_GROUP_SCHED - /* - * Since the real-depth could have been changed (only FAIR - * class maintain depth value), reset depth properly. - */ - se->depth = se->parent ? se->parent->depth + 1 : 0; -#endif - /* Synchronize entity with its cfs_rq */ update_load_avg(cfs_rq, se, sched_feat(ATTACH_AGE_LOAD) ? 0 : SKIP_AGE_LOAD); attach_entity_load_avg(cfs_rq, se); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 3ccd35c22f0f..4c4822141026 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1930,6 +1930,7 @@ static inline void set_task_rq(struct task_struct *p, unsigned int cpu) set_task_rq_fair(&p->se, p->se.cfs_rq, tg->cfs_rq[cpu]); p->se.cfs_rq = tg->cfs_rq[cpu]; p->se.parent = tg->se[cpu]; + p->se.depth = tg->se[cpu] ? tg->se[cpu]->depth + 1 : 0; #endif #ifdef CONFIG_RT_GROUP_SCHED From 39c4261191bf05e7eb310f852980a6d0afe5582a Mon Sep 17 00:00:00 2001 From: Chengming Zhou Date: Thu, 18 Aug 2022 20:47:58 +0800 Subject: [PATCH 10/41] sched/fair: Remove redundant cpu_cgrp_subsys->fork() We use cpu_cgrp_subsys->fork() to set task group for the new fair task in cgroup_post_fork(). Since commit b1e8206582f9 ("sched: Fix yet more sched_fork() races") has already set_task_rq() for the new fair task in sched_cgroup_fork(), so cpu_cgrp_subsys->fork() can be removed. cgroup_can_fork() --> pin parent's sched_task_group sched_cgroup_fork() __set_task_cpu() set_task_rq() cgroup_post_fork() ss->fork() := cpu_cgroup_fork() sched_change_group(..., TASK_SET_GROUP) task_set_group_fair() set_task_rq() --> can be removed After this patch's change, task_change_group_fair() only need to care about task cgroup migration, make the code much simplier. Signed-off-by: Chengming Zhou Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Reviewed-by: Dietmar Eggemann Link: https://lore.kernel.org/r/20220818124805.601-3-zhouchengming@bytedance.com --- kernel/sched/core.c | 27 ++++----------------------- kernel/sched/fair.c | 23 +---------------------- kernel/sched/sched.h | 5 +---- 3 files changed, 6 insertions(+), 49 deletions(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 64c08993221b..e74e79f783af 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -481,8 +481,7 @@ sched_core_dequeue(struct rq *rq, struct task_struct *p, int flags) { } * p->se.load, p->rt_priority, * p->dl.dl_{runtime, deadline, period, flags, bw, density} * - sched_setnuma(): p->numa_preferred_nid - * - sched_move_task()/ - * cpu_cgroup_fork(): p->sched_task_group + * - sched_move_task(): p->sched_task_group * - uclamp_update_active() p->uclamp* * * p->state <- TASK_*: @@ -10114,7 +10113,7 @@ void sched_release_group(struct task_group *tg) spin_unlock_irqrestore(&task_group_lock, flags); } -static void sched_change_group(struct task_struct *tsk, int type) +static void sched_change_group(struct task_struct *tsk) { struct task_group *tg; @@ -10130,7 +10129,7 @@ static void sched_change_group(struct task_struct *tsk, int type) #ifdef CONFIG_FAIR_GROUP_SCHED if (tsk->sched_class->task_change_group) - tsk->sched_class->task_change_group(tsk, type); + tsk->sched_class->task_change_group(tsk); else #endif set_task_rq(tsk, task_cpu(tsk)); @@ -10161,7 +10160,7 @@ void sched_move_task(struct task_struct *tsk) if (running) put_prev_task(rq, tsk); - sched_change_group(tsk, TASK_MOVE_GROUP); + sched_change_group(tsk); if (queued) enqueue_task(rq, tsk, queue_flags); @@ -10239,23 +10238,6 @@ static void cpu_cgroup_css_free(struct cgroup_subsys_state *css) sched_unregister_group(tg); } -/* - * This is called before wake_up_new_task(), therefore we really only - * have to set its group bits, all the other stuff does not apply. - */ -static void cpu_cgroup_fork(struct task_struct *task) -{ - struct rq_flags rf; - struct rq *rq; - - rq = task_rq_lock(task, &rf); - - update_rq_clock(rq); - sched_change_group(task, TASK_SET_GROUP); - - task_rq_unlock(rq, task, &rf); -} - static int cpu_cgroup_can_attach(struct cgroup_taskset *tset) { struct task_struct *task; @@ -11121,7 +11103,6 @@ struct cgroup_subsys cpu_cgrp_subsys = { .css_released = cpu_cgroup_css_released, .css_free = cpu_cgroup_css_free, .css_extra_stat_show = cpu_extra_stat_show, - .fork = cpu_cgroup_fork, .can_attach = cpu_cgroup_can_attach, .attach = cpu_cgroup_attach, .legacy_cftypes = cpu_legacy_files, diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index a3b0f8b1029e..2c0eb2a4e341 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -11657,15 +11657,7 @@ void init_cfs_rq(struct cfs_rq *cfs_rq) } #ifdef CONFIG_FAIR_GROUP_SCHED -static void task_set_group_fair(struct task_struct *p) -{ - struct sched_entity *se = &p->se; - - set_task_rq(p, task_cpu(p)); - se->depth = se->parent ? se->parent->depth + 1 : 0; -} - -static void task_move_group_fair(struct task_struct *p) +static void task_change_group_fair(struct task_struct *p) { detach_task_cfs_rq(p); set_task_rq(p, task_cpu(p)); @@ -11677,19 +11669,6 @@ static void task_move_group_fair(struct task_struct *p) attach_task_cfs_rq(p); } -static void task_change_group_fair(struct task_struct *p, int type) -{ - switch (type) { - case TASK_SET_GROUP: - task_set_group_fair(p); - break; - - case TASK_MOVE_GROUP: - task_move_group_fair(p); - break; - } -} - void free_fair_sched_group(struct task_group *tg) { int i; diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 4c4822141026..74130a69d365 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -2193,11 +2193,8 @@ struct sched_class { void (*update_curr)(struct rq *rq); -#define TASK_SET_GROUP 0 -#define TASK_MOVE_GROUP 1 - #ifdef CONFIG_FAIR_GROUP_SCHED - void (*task_change_group)(struct task_struct *p, int type); + void (*task_change_group)(struct task_struct *p); #endif }; From 5d6da83c44af70ede7bfd0fd6d1ef8a3b3e0402c Mon Sep 17 00:00:00 2001 From: Chengming Zhou Date: Thu, 18 Aug 2022 20:47:59 +0800 Subject: [PATCH 11/41] sched/fair: Reset sched_avg last_update_time before set_task_rq() set_task_rq() -> set_task_rq_fair() will try to synchronize the blocked task's sched_avg when migrate, which is not needed for already detached task. task_change_group_fair() will detached the task sched_avg from prev cfs_rq first, so reset sched_avg last_update_time before set_task_rq() to avoid that. Signed-off-by: Chengming Zhou Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Dietmar Eggemann Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20220818124805.601-4-zhouchengming@bytedance.com --- kernel/sched/fair.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 2c0eb2a4e341..e4c0929a6e71 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -11660,12 +11660,12 @@ void init_cfs_rq(struct cfs_rq *cfs_rq) static void task_change_group_fair(struct task_struct *p) { detach_task_cfs_rq(p); - set_task_rq(p, task_cpu(p)); #ifdef CONFIG_SMP /* Tell se's cfs_rq has been changed -- migrated */ p->se.avg.last_update_time = 0; #endif + set_task_rq(p, task_cpu(p)); attach_task_cfs_rq(p); } From 859f206290f345c151a6005de639ba9677bf3e18 Mon Sep 17 00:00:00 2001 From: Chengming Zhou Date: Thu, 18 Aug 2022 20:48:00 +0800 Subject: [PATCH 12/41] sched/fair: Update comments in enqueue/dequeue_entity() When reading the sched_avg related code, I found the comments in enqueue/dequeue_entity() are not updated with the current code. We don't add/subtract entity's runnable_avg from cfs_rq->runnable_avg during enqueue/dequeue_entity(), those are done only for attach/detach. This patch updates the comments to reflect the current code working. Signed-off-by: Chengming Zhou Signed-off-by: Peter Zijlstra (Intel) Acked-by: Vincent Guittot Link: https://lore.kernel.org/r/20220818124805.601-5-zhouchengming@bytedance.com --- kernel/sched/fair.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index e4c0929a6e71..52de8302b336 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4434,7 +4434,8 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) /* * When enqueuing a sched_entity, we must: * - Update loads to have both entity and cfs_rq synced with now. - * - Add its load to cfs_rq->runnable_avg + * - For group_entity, update its runnable_weight to reflect the new + * h_nr_running of its group cfs_rq. * - For group_entity, update its weight to reflect the new share of * its group cfs_rq * - Add its new weight to cfs_rq->load.weight @@ -4519,7 +4520,8 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) /* * When dequeuing a sched_entity, we must: * - Update loads to have both entity and cfs_rq synced with now. - * - Subtract its load from the cfs_rq->runnable_avg. + * - For group_entity, update its runnable_weight to reflect the new + * h_nr_running of its group cfs_rq. * - Subtract its previous weight from cfs_rq->load.weight. * - For group entity, update its weight to reflect the new share * of its group cfs_rq. From e1f078f50478a51849453341e7356cb298df00cf Mon Sep 17 00:00:00 2001 From: Chengming Zhou Date: Thu, 18 Aug 2022 20:48:01 +0800 Subject: [PATCH 13/41] sched/fair: Combine detach into dequeue when migrating task When we are migrating task out of the CPU, we can combine detach and propagation into dequeue_entity() to save the detach_entity_cfs_rq() in migrate_task_rq_fair(). This optimization is like combining DO_ATTACH in the enqueue_entity() when migrating task to the CPU. So we don't have to traverse the CFS tree extra time to do the detach_entity_cfs_rq() -> propagate_entity_cfs_rq(), which wouldn't be called anymore with this patch's change. detach_task() deactivate_task() dequeue_task_fair() for_each_sched_entity(se) dequeue_entity() update_load_avg() /* (1) */ detach_entity_load_avg() set_task_cpu() migrate_task_rq_fair() detach_entity_cfs_rq() /* (2) */ update_load_avg(); detach_entity_load_avg(); propagate_entity_cfs_rq(); for_each_sched_entity() update_load_avg() This patch save the detach_entity_cfs_rq() called in (2) by doing the detach_entity_load_avg() for a CPU migrating task inside (1) (the task being the first se in the loop) Signed-off-by: Chengming Zhou Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20220818124805.601-6-zhouchengming@bytedance.com --- kernel/sched/fair.c | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 52de8302b336..f52e7dc7f22d 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4003,6 +4003,7 @@ static void detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s #define UPDATE_TG 0x1 #define SKIP_AGE_LOAD 0x2 #define DO_ATTACH 0x4 +#define DO_DETACH 0x8 /* Update task and its cfs_rq load average */ static inline void update_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) @@ -4032,6 +4033,13 @@ static inline void update_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s attach_entity_load_avg(cfs_rq, se); update_tg_load_avg(cfs_rq); + } else if (flags & DO_DETACH) { + /* + * DO_DETACH means we're here from dequeue_entity() + * and we are migrating task out of the CPU. + */ + detach_entity_load_avg(cfs_rq, se); + update_tg_load_avg(cfs_rq); } else if (decayed) { cfs_rq_util_change(cfs_rq, 0); @@ -4292,6 +4300,7 @@ static inline bool cfs_rq_is_decayed(struct cfs_rq *cfs_rq) #define UPDATE_TG 0x0 #define SKIP_AGE_LOAD 0x0 #define DO_ATTACH 0x0 +#define DO_DETACH 0x0 static inline void update_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se, int not_used1) { @@ -4512,6 +4521,11 @@ static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq); static void dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { + int action = UPDATE_TG; + + if (entity_is_task(se) && task_on_rq_migrating(task_of(se))) + action |= DO_DETACH; + /* * Update run-time statistics of the 'current'. */ @@ -4526,7 +4540,7 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * - For group entity, update its weight to reflect the new share * of its group cfs_rq. */ - update_load_avg(cfs_rq, se, UPDATE_TG); + update_load_avg(cfs_rq, se, action); se_update_runnable(se); update_stats_dequeue_fair(cfs_rq, se, flags); @@ -7078,8 +7092,6 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags) return new_cpu; } -static void detach_entity_cfs_rq(struct sched_entity *se); - /* * Called immediately before a task is migrated to a new CPU; task_cpu(p) and * cfs_rq_of(p) references at time of call are still valid and identify the @@ -7101,15 +7113,7 @@ static void migrate_task_rq_fair(struct task_struct *p, int new_cpu) se->vruntime -= u64_u32_load(cfs_rq->min_vruntime); } - if (p->on_rq == TASK_ON_RQ_MIGRATING) { - /* - * In case of TASK_ON_RQ_MIGRATING we in fact hold the 'old' - * rq->lock and can modify state directly. - */ - lockdep_assert_rq_held(task_rq(p)); - detach_entity_cfs_rq(se); - - } else { + if (!task_on_rq_migrating(p)) { remove_entity_load_avg(se); /* From 7e2edaf61814fb6aa363989d718950c023b882d4 Mon Sep 17 00:00:00 2001 From: Chengming Zhou Date: Thu, 18 Aug 2022 20:48:02 +0800 Subject: [PATCH 14/41] sched/fair: Fix another detach on unattached task corner case commit 7dc603c9028e ("sched/fair: Fix PELT integrity for new tasks") fixed two load tracking problems for new task, including detach on unattached new task problem. There still left another detach on unattached task problem for the task which has been woken up by try_to_wake_up() and waiting for actually being woken up by sched_ttwu_pending(). try_to_wake_up(p) cpu = select_task_rq(p) if (task_cpu(p) != cpu) set_task_cpu(p, cpu) migrate_task_rq_fair() remove_entity_load_avg() --> unattached se->avg.last_update_time = 0; __set_task_cpu() ttwu_queue(p, cpu) ttwu_queue_wakelist() __ttwu_queue_wakelist() task_change_group_fair() detach_task_cfs_rq() detach_entity_cfs_rq() detach_entity_load_avg() --> detach on unattached task set_task_rq() attach_task_cfs_rq() attach_entity_cfs_rq() attach_entity_load_avg() The reason of this problem is similar, we should check in detach_entity_cfs_rq() that se->avg.last_update_time != 0, before do detach_entity_load_avg(). Signed-off-by: Chengming Zhou Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20220818124805.601-7-zhouchengming@bytedance.com --- kernel/sched/fair.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index f52e7dc7f22d..e92bc053aff6 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -11557,6 +11557,17 @@ static void detach_entity_cfs_rq(struct sched_entity *se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); +#ifdef CONFIG_SMP + /* + * In case the task sched_avg hasn't been attached: + * - A forked task which hasn't been woken up by wake_up_new_task(). + * - A task which has been woken up by try_to_wake_up() but is + * waiting for actually being woken up by sched_ttwu_pending(). + */ + if (!se->avg.last_update_time) + return; +#endif + /* Catch up with the cfs_rq and remove our load when we leave */ update_load_avg(cfs_rq, se, 0); detach_entity_load_avg(cfs_rq, se); From df16b71c686cb096774e30153c9ce6756450796c Mon Sep 17 00:00:00 2001 From: Chengming Zhou Date: Thu, 18 Aug 2022 20:48:03 +0800 Subject: [PATCH 15/41] sched/fair: Allow changing cgroup of new forked task commit 7dc603c9028e ("sched/fair: Fix PELT integrity for new tasks") introduce a TASK_NEW state and an unnessary limitation that would fail when changing cgroup of new forked task. Because at that time, we can't handle task_change_group_fair() for new forked fair task which hasn't been woken up by wake_up_new_task(), which will cause detach on an unattached task sched_avg problem. This patch delete this unnessary limitation by adding check before do detach or attach in task_change_group_fair(). So cpu_cgrp_subsys.can_attach() has nothing to do for fair tasks, only define it in #ifdef CONFIG_RT_GROUP_SCHED. Signed-off-by: Chengming Zhou Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20220818124805.601-8-zhouchengming@bytedance.com --- kernel/sched/core.c | 25 +++++-------------------- kernel/sched/fair.c | 7 +++++++ 2 files changed, 12 insertions(+), 20 deletions(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index e74e79f783af..603a80ec9b0e 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -10238,36 +10238,19 @@ static void cpu_cgroup_css_free(struct cgroup_subsys_state *css) sched_unregister_group(tg); } +#ifdef CONFIG_RT_GROUP_SCHED static int cpu_cgroup_can_attach(struct cgroup_taskset *tset) { struct task_struct *task; struct cgroup_subsys_state *css; - int ret = 0; cgroup_taskset_for_each(task, css, tset) { -#ifdef CONFIG_RT_GROUP_SCHED if (!sched_rt_can_attach(css_tg(css), task)) return -EINVAL; -#endif - /* - * Serialize against wake_up_new_task() such that if it's - * running, we're sure to observe its full state. - */ - raw_spin_lock_irq(&task->pi_lock); - /* - * Avoid calling sched_move_task() before wake_up_new_task() - * has happened. This would lead to problems with PELT, due to - * move wanting to detach+attach while we're not attached yet. - */ - if (READ_ONCE(task->__state) == TASK_NEW) - ret = -EINVAL; - raw_spin_unlock_irq(&task->pi_lock); - - if (ret) - break; } - return ret; + return 0; } +#endif static void cpu_cgroup_attach(struct cgroup_taskset *tset) { @@ -11103,7 +11086,9 @@ struct cgroup_subsys cpu_cgrp_subsys = { .css_released = cpu_cgroup_css_released, .css_free = cpu_cgroup_css_free, .css_extra_stat_show = cpu_extra_stat_show, +#ifdef CONFIG_RT_GROUP_SCHED .can_attach = cpu_cgroup_can_attach, +#endif .attach = cpu_cgroup_attach, .legacy_cftypes = cpu_legacy_files, .dfl_cftypes = cpu_files, diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index e92bc053aff6..fd1aa4c92b2d 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -11676,6 +11676,13 @@ void init_cfs_rq(struct cfs_rq *cfs_rq) #ifdef CONFIG_FAIR_GROUP_SCHED static void task_change_group_fair(struct task_struct *p) { + /* + * We couldn't detach or attach a forked task which + * hasn't been woken up by wake_up_new_task(). + */ + if (READ_ONCE(p->__state) == TASK_NEW) + return; + detach_task_cfs_rq(p); #ifdef CONFIG_SMP From d6531ab6e50149ab2a144b0f4787cb9277d0893f Mon Sep 17 00:00:00 2001 From: Chengming Zhou Date: Thu, 18 Aug 2022 20:48:04 +0800 Subject: [PATCH 16/41] sched/fair: Move task sched_avg attach to enqueue_task_fair() When wake_up_new_task(), we use post_init_entity_util_avg() to init util_avg/runnable_avg based on cpu's util_avg at that time, and attach task sched_avg to cfs_rq. Since enqueue_task_fair() -> enqueue_entity() -> update_load_avg() loop will do attach, we can move this work to update_load_avg(). wake_up_new_task(p) post_init_entity_util_avg(p) attach_entity_cfs_rq() --> (1) activate_task(rq, p) enqueue_task() := enqueue_task_fair() enqueue_entity() loop update_load_avg(cfs_rq, se, UPDATE_TG | DO_ATTACH) if (!se->avg.last_update_time && (flags & DO_ATTACH)) attach_entity_load_avg() --> (2) This patch move attach from (1) to (2), update related comments too. Signed-off-by: Chengming Zhou Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20220818124805.601-9-zhouchengming@bytedance.com --- kernel/sched/fair.c | 11 +++-------- 1 file changed, 3 insertions(+), 8 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index fd1aa4c92b2d..ef325b561df4 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -799,8 +799,6 @@ void init_entity_runnable_average(struct sched_entity *se) /* when this task enqueue'ed, it will contribute to its cfs_rq's load_avg */ } -static void attach_entity_cfs_rq(struct sched_entity *se); - /* * With new tasks being created, their initial util_avgs are extrapolated * based on the cfs_rq's current util_avg: @@ -863,8 +861,6 @@ void post_init_entity_util_avg(struct task_struct *p) se->avg.last_update_time = cfs_rq_clock_pelt(cfs_rq); return; } - - attach_entity_cfs_rq(se); } #else /* !CONFIG_SMP */ @@ -3838,8 +3834,7 @@ static void migrate_se_pelt_lag(struct sched_entity *se) {} * @cfs_rq: cfs_rq to update * * The cfs_rq avg is the direct sum of all its entities (blocked and runnable) - * avg. The immediate corollary is that all (fair) tasks must be attached, see - * post_init_entity_util_avg(). + * avg. The immediate corollary is that all (fair) tasks must be attached. * * cfs_rq->avg is used for task_h_load() and update_cfs_share() for example. * @@ -4072,8 +4067,8 @@ static void remove_entity_load_avg(struct sched_entity *se) /* * tasks cannot exit without having gone through wake_up_new_task() -> - * post_init_entity_util_avg() which will have added things to the - * cfs_rq, so we can remove unconditionally. + * enqueue_task_fair() which will have added things to the cfs_rq, + * so we can remove unconditionally. */ sync_entity_load_avg(se); From e4fe074d6c359c19b74564fa1364fe48343cfa5d Mon Sep 17 00:00:00 2001 From: Chengming Zhou Date: Thu, 18 Aug 2022 20:48:05 +0800 Subject: [PATCH 17/41] sched/fair: Don't init util/runnable_avg for !fair task post_init_entity_util_avg() init task util_avg according to the cpu util_avg at the time of fork, which will decay when switched_to_fair() some time later, we'd better to not set them at all in the case of !fair task. Suggested-by: Vincent Guittot Signed-off-by: Chengming Zhou Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20220818124805.601-10-zhouchengming@bytedance.com --- kernel/sched/fair.c | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index ef325b561df4..e8c1b889dcbb 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -833,20 +833,6 @@ void post_init_entity_util_avg(struct task_struct *p) long cpu_scale = arch_scale_cpu_capacity(cpu_of(rq_of(cfs_rq))); long cap = (long)(cpu_scale - cfs_rq->avg.util_avg) / 2; - if (cap > 0) { - if (cfs_rq->avg.util_avg != 0) { - sa->util_avg = cfs_rq->avg.util_avg * se->load.weight; - sa->util_avg /= (cfs_rq->avg.load_avg + 1); - - if (sa->util_avg > cap) - sa->util_avg = cap; - } else { - sa->util_avg = cap; - } - } - - sa->runnable_avg = sa->util_avg; - if (p->sched_class != &fair_sched_class) { /* * For !fair tasks do: @@ -861,6 +847,20 @@ void post_init_entity_util_avg(struct task_struct *p) se->avg.last_update_time = cfs_rq_clock_pelt(cfs_rq); return; } + + if (cap > 0) { + if (cfs_rq->avg.util_avg != 0) { + sa->util_avg = cfs_rq->avg.util_avg * se->load.weight; + sa->util_avg /= (cfs_rq->avg.load_avg + 1); + + if (sa->util_avg > cap) + sa->util_avg = cap; + } else { + sa->util_avg = cap; + } + } + + sa->runnable_avg = sa->util_avg; } #else /* !CONFIG_SMP */ From 5531ecffa4b923bc7739e9ea73c552d80af602dc Mon Sep 17 00:00:00 2001 From: Shang XiaoJing Date: Wed, 24 Aug 2022 16:28:56 +0800 Subject: [PATCH 18/41] sched: Add update_current_exec_runtime helper Wrap repeated code in helper function update_current_exec_runtime for update the exec time of the current. Signed-off-by: Shang XiaoJing Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20220824082856.15674-1-shangxiaojing@huawei.com --- kernel/sched/deadline.c | 6 +----- kernel/sched/rt.c | 6 +----- kernel/sched/sched.h | 10 ++++++++++ kernel/sched/stop_task.c | 11 ++++------- 4 files changed, 16 insertions(+), 17 deletions(-) diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 1d9c90958baa..4cbdc0c560da 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -1333,11 +1333,7 @@ static void update_curr_dl(struct rq *rq) trace_sched_stat_runtime(curr, delta_exec, 0); - curr->se.sum_exec_runtime += delta_exec; - account_group_exec_runtime(curr, delta_exec); - - curr->se.exec_start = now; - cgroup_account_cputime(curr, delta_exec); + update_current_exec_runtime(curr, now, delta_exec); if (dl_entity_is_special(dl_se)) return; diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 054b6711e961..4bc84a1135db 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -1062,11 +1062,7 @@ static void update_curr_rt(struct rq *rq) trace_sched_stat_runtime(curr, delta_exec, 0); - curr->se.sum_exec_runtime += delta_exec; - account_group_exec_runtime(curr, delta_exec); - - curr->se.exec_start = now; - cgroup_account_cputime(curr, delta_exec); + update_current_exec_runtime(curr, now, delta_exec); if (!rt_bandwidth_enabled()) return; diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 74130a69d365..da17be6f27fd 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -3158,4 +3158,14 @@ extern int sched_dynamic_mode(const char *str); extern void sched_dynamic_update(int mode); #endif +static inline void update_current_exec_runtime(struct task_struct *curr, + u64 now, u64 delta_exec) +{ + curr->se.sum_exec_runtime += delta_exec; + account_group_exec_runtime(curr, delta_exec); + + curr->se.exec_start = now; + cgroup_account_cputime(curr, delta_exec); +} + #endif /* _KERNEL_SCHED_SCHED_H */ diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c index d04073a93eb4..85590599b4d6 100644 --- a/kernel/sched/stop_task.c +++ b/kernel/sched/stop_task.c @@ -71,20 +71,17 @@ static void yield_task_stop(struct rq *rq) static void put_prev_task_stop(struct rq *rq, struct task_struct *prev) { struct task_struct *curr = rq->curr; - u64 delta_exec; + u64 now, delta_exec; - delta_exec = rq_clock_task(rq) - curr->se.exec_start; + now = rq_clock_task(rq); + delta_exec = now - curr->se.exec_start; if (unlikely((s64)delta_exec < 0)) delta_exec = 0; schedstat_set(curr->stats.exec_max, max(curr->stats.exec_max, delta_exec)); - curr->se.sum_exec_runtime += delta_exec; - account_group_exec_runtime(curr, delta_exec); - - curr->se.exec_start = rq_clock_task(rq); - cgroup_account_cputime(curr, delta_exec); + update_current_exec_runtime(curr, now, delta_exec); } /* From 973bee493a1f75c6c0752a74fb9396cbc34f026e Mon Sep 17 00:00:00 2001 From: Shang XiaoJing Date: Fri, 26 Aug 2022 16:34:53 +0800 Subject: [PATCH 19/41] sched/deadline: Add dl_task_is_earliest_deadline helper Wrap repeated code in helper function dl_task_is_earliest_deadline, which return true if there is no deadline task on the rq at all, or task's deadline earlier than the whole rq. Signed-off-by: Shang XiaoJing Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Daniel Bristot de Oliveira Link: https://lore.kernel.org/r/20220826083453.698-1-shangxiaojing@huawei.com --- kernel/sched/deadline.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index d116d2b9d2f9..34271aff4712 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -1810,6 +1810,14 @@ static void yield_task_dl(struct rq *rq) #ifdef CONFIG_SMP +static inline bool dl_task_is_earliest_deadline(struct task_struct *p, + struct rq *rq) +{ + return (!rq->dl.dl_nr_running || + dl_time_before(p->dl.deadline, + rq->dl.earliest_dl.curr)); +} + static int find_later_rq(struct task_struct *task); static int @@ -1852,9 +1860,7 @@ select_task_rq_dl(struct task_struct *p, int cpu, int flags) int target = find_later_rq(p); if (target != -1 && - (dl_time_before(p->dl.deadline, - cpu_rq(target)->dl.earliest_dl.curr) || - (cpu_rq(target)->dl.dl_nr_running == 0))) + dl_task_is_earliest_deadline(p, cpu_rq(target))) cpu = target; } rcu_read_unlock(); @@ -2221,9 +2227,7 @@ static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq) later_rq = cpu_rq(cpu); - if (later_rq->dl.dl_nr_running && - !dl_time_before(task->dl.deadline, - later_rq->dl.earliest_dl.curr)) { + if (!dl_task_is_earliest_deadline(task, later_rq)) { /* * Target rq has tasks of equal or earlier deadline, * retrying does not release any lock and is unlikely @@ -2251,9 +2255,7 @@ static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq) * its earliest one has a later deadline than our * task, the rq is a good one. */ - if (!later_rq->dl.dl_nr_running || - dl_time_before(task->dl.deadline, - later_rq->dl.earliest_dl.curr)) + if (dl_task_is_earliest_deadline(task, later_rq)) break; /* Otherwise we try again. */ @@ -2424,9 +2426,7 @@ static void pull_dl_task(struct rq *this_rq) * - it will preempt the last one we pulled (if any). */ if (p && dl_time_before(p->dl.deadline, dmin) && - (!this_rq->dl.dl_nr_running || - dl_time_before(p->dl.deadline, - this_rq->dl.earliest_dl.curr))) { + dl_task_is_earliest_deadline(p, this_rq)) { WARN_ON(p == src_rq->curr); WARN_ON(!task_on_rq_queued(p)); From 96458e7f7dc5ad14bd7577cbf1638e1504ad79dd Mon Sep 17 00:00:00 2001 From: Shang XiaoJing Date: Fri, 26 Aug 2022 18:00:37 +0800 Subject: [PATCH 20/41] sched/deadline: Add replenish_dl_new_period helper Wrap repeated code in helper function replenish_dl_new_period, which set the deadline and runtime of input dl_se based on pi_of(dl_se). Note that setup_new_dl_entity originally set the deadline and runtime base on dl_se, which should equals to pi_of(dl_se) for non-boosted task. Signed-off-by: Shang XiaoJing Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Daniel Bristot de Oliveira Link: https://lore.kernel.org/r/20220826100037.12146-1-shangxiaojing@huawei.com --- kernel/sched/deadline.c | 23 +++++++++++++---------- 1 file changed, 13 insertions(+), 10 deletions(-) diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 34271aff4712..3bf4b12ec5b7 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -770,6 +770,14 @@ static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags); static void __dequeue_task_dl(struct rq *rq, struct task_struct *p, int flags); static void check_preempt_curr_dl(struct rq *rq, struct task_struct *p, int flags); +static inline void replenish_dl_new_period(struct sched_dl_entity *dl_se, + struct rq *rq) +{ + /* for non-boosted task, pi_of(dl_se) == dl_se */ + dl_se->deadline = rq_clock(rq) + pi_of(dl_se)->dl_deadline; + dl_se->runtime = pi_of(dl_se)->dl_runtime; +} + /* * We are being explicitly informed that a new instance is starting, * and this means that: @@ -803,8 +811,7 @@ static inline void setup_new_dl_entity(struct sched_dl_entity *dl_se) * future; in fact, we must consider execution overheads (time * spent on hardirq context, etc.). */ - dl_se->deadline = rq_clock(rq) + dl_se->dl_deadline; - dl_se->runtime = dl_se->dl_runtime; + replenish_dl_new_period(dl_se, rq); } /* @@ -836,10 +843,8 @@ static void replenish_dl_entity(struct sched_dl_entity *dl_se) * This could be the case for a !-dl task that is boosted. * Just go with full inherited parameters. */ - if (dl_se->dl_deadline == 0) { - dl_se->deadline = rq_clock(rq) + pi_of(dl_se)->dl_deadline; - dl_se->runtime = pi_of(dl_se)->dl_runtime; - } + if (dl_se->dl_deadline == 0) + replenish_dl_new_period(dl_se, rq); if (dl_se->dl_yielded && dl_se->runtime > 0) dl_se->runtime = 0; @@ -866,8 +871,7 @@ static void replenish_dl_entity(struct sched_dl_entity *dl_se) */ if (dl_time_before(dl_se->deadline, rq_clock(rq))) { printk_deferred_once("sched: DL replenish lagged too much\n"); - dl_se->deadline = rq_clock(rq) + pi_of(dl_se)->dl_deadline; - dl_se->runtime = pi_of(dl_se)->dl_runtime; + replenish_dl_new_period(dl_se, rq); } if (dl_se->dl_yielded) @@ -1024,8 +1028,7 @@ static void update_dl_entity(struct sched_dl_entity *dl_se) return; } - dl_se->deadline = rq_clock(rq) + pi_of(dl_se)->dl_deadline; - dl_se->runtime = pi_of(dl_se)->dl_runtime; + replenish_dl_new_period(dl_se, rq); } } From 33f93525799fa3c841b2ba93a56b2bb32ab11dc9 Mon Sep 17 00:00:00 2001 From: Shang XiaoJing Date: Sat, 27 Aug 2022 10:09:11 +0800 Subject: [PATCH 21/41] sched/deadline: Move __dl_clear_params out of dl_bw lock As members in sched_dl_entity are independent with dl_bw, move __dl_clear_params out of dl_bw lock. Signed-off-by: Shang XiaoJing Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Daniel Bristot de Oliveira Link: https://lore.kernel.org/r/20220827020911.30641-1-shangxiaojing@huawei.com --- kernel/sched/deadline.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 3bf4b12ec5b7..d0fe6a20a9c9 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -431,8 +431,8 @@ static void task_non_contending(struct task_struct *p) sub_rq_bw(&p->dl, &rq->dl); raw_spin_lock(&dl_b->lock); __dl_sub(dl_b, p->dl.dl_bw, dl_bw_cpus(task_cpu(p))); - __dl_clear_params(p); raw_spin_unlock(&dl_b->lock); + __dl_clear_params(p); } return; From 3e6efe87cd5ccabf0f1d4e3ef25881ca0fd337e7 Mon Sep 17 00:00:00 2001 From: Abel Wu Date: Wed, 7 Sep 2022 19:19:56 +0800 Subject: [PATCH 22/41] sched/fair: Remove redundant check in select_idle_smt() If two cpus share LLC cache, then the two cores they belong to are also in the same LLC domain. Signed-off-by: Abel Wu Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Josh Don Acked-by: Mel Gorman Link: https://lore.kernel.org/r/20220907112000.1854-2-wuyun.abel@bytedance.com --- kernel/sched/fair.c | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index efceb670e755..9657c7de5f57 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -6350,14 +6350,11 @@ static int select_idle_core(struct task_struct *p, int core, struct cpumask *cpu /* * Scan the local SMT mask for idle CPUs. */ -static int select_idle_smt(struct task_struct *p, struct sched_domain *sd, int target) +static int select_idle_smt(struct task_struct *p, int target) { int cpu; - for_each_cpu(cpu, cpu_smt_mask(target)) { - if (!cpumask_test_cpu(cpu, p->cpus_ptr) || - !cpumask_test_cpu(cpu, sched_domain_span(sd))) - continue; + for_each_cpu_and(cpu, cpu_smt_mask(target), p->cpus_ptr) { if (available_idle_cpu(cpu) || sched_idle_cpu(cpu)) return cpu; } @@ -6381,7 +6378,7 @@ static inline int select_idle_core(struct task_struct *p, int core, struct cpuma return __select_idle_cpu(core, p); } -static inline int select_idle_smt(struct task_struct *p, struct sched_domain *sd, int target) +static inline int select_idle_smt(struct task_struct *p, int target) { return -1; } @@ -6615,7 +6612,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) has_idle_core = test_idle_cores(target, false); if (!has_idle_core && cpus_share_cache(prev, target)) { - i = select_idle_smt(p, sd, prev); + i = select_idle_smt(p, prev); if ((unsigned int)i < nr_cpumask_bits) return i; } From b9bae70440d21e106fbc098803b5a190df65f2e0 Mon Sep 17 00:00:00 2001 From: Abel Wu Date: Wed, 7 Sep 2022 19:19:57 +0800 Subject: [PATCH 23/41] sched/fair: Avoid double search on same cpu The prev cpu is checked at the beginning of SIS, and it's unlikely to be idle before the second check in select_idle_smt(). So we'd better focus on its SMT siblings. Signed-off-by: Abel Wu Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Josh Don Acked-by: Mel Gorman Link: https://lore.kernel.org/r/20220907112000.1854-3-wuyun.abel@bytedance.com --- kernel/sched/fair.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 9657c7de5f57..1ad79aaaaf93 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -6355,6 +6355,8 @@ static int select_idle_smt(struct task_struct *p, int target) int cpu; for_each_cpu_and(cpu, cpu_smt_mask(target), p->cpus_ptr) { + if (cpu == target) + continue; if (available_idle_cpu(cpu) || sched_idle_cpu(cpu)) return cpu; } From 8eeeed9c4a791f0d1f2ea830eb75a4246c117ae2 Mon Sep 17 00:00:00 2001 From: Abel Wu Date: Wed, 7 Sep 2022 19:19:58 +0800 Subject: [PATCH 24/41] sched/fair: Remove useless check in select_idle_core() The function select_idle_core() only gets called when has_idle_cores is true which can be possible only when sched_smt_present is enabled. This change also aligns select_idle_core() with select_idle_smt() in the way that the caller do the check if necessary. Signed-off-by: Abel Wu Signed-off-by: Peter Zijlstra (Intel) Acked-by: Mel Gorman Link: https://lore.kernel.org/r/20220907112000.1854-4-wuyun.abel@bytedance.com --- kernel/sched/fair.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 1ad79aaaaf93..03ce65068333 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -6321,9 +6321,6 @@ static int select_idle_core(struct task_struct *p, int core, struct cpumask *cpu bool idle = true; int cpu; - if (!static_branch_likely(&sched_smt_present)) - return __select_idle_cpu(core, p); - for_each_cpu(cpu, cpu_smt_mask(core)) { if (!available_idle_cpu(cpu)) { idle = false; From 398ba2b0cc0a43964fe3d2dd19cb2a478f1f220b Mon Sep 17 00:00:00 2001 From: Abel Wu Date: Wed, 7 Sep 2022 19:19:59 +0800 Subject: [PATCH 25/41] sched/fair: Default to false in test_idle_cores() It's uncertain whether idle cores exist or not if shared sched- domains are not ready, so returning "no idle cores" usually makes sense. While __update_idle_core() is an exception, it checks status of this core and set hint to shared sched-domain if necessary. So the whole logic of this function depends on the existence of shared sched-domain, and can certainly bail out early if it is not available. It's somehow a little tricky, and as Josh suggested that it should be transient while the domain isn't ready. So remove the self-defined default value to make things more clearer. Signed-off-by: Abel Wu Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Josh Don Acked-by: Mel Gorman Link: https://lore.kernel.org/r/20220907112000.1854-5-wuyun.abel@bytedance.com --- kernel/sched/fair.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 03ce65068333..23b020c3d3a0 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -1588,11 +1588,11 @@ numa_type numa_classify(unsigned int imbalance_pct, #ifdef CONFIG_SCHED_SMT /* Forward declarations of select_idle_sibling helpers */ -static inline bool test_idle_cores(int cpu, bool def); +static inline bool test_idle_cores(int cpu); static inline int numa_idle_core(int idle_core, int cpu) { if (!static_branch_likely(&sched_smt_present) || - idle_core >= 0 || !test_idle_cores(cpu, false)) + idle_core >= 0 || !test_idle_cores(cpu)) return idle_core; /* @@ -6271,7 +6271,7 @@ static inline void set_idle_cores(int cpu, int val) WRITE_ONCE(sds->has_idle_cores, val); } -static inline bool test_idle_cores(int cpu, bool def) +static inline bool test_idle_cores(int cpu) { struct sched_domain_shared *sds; @@ -6279,7 +6279,7 @@ static inline bool test_idle_cores(int cpu, bool def) if (sds) return READ_ONCE(sds->has_idle_cores); - return def; + return false; } /* @@ -6295,7 +6295,7 @@ void __update_idle_core(struct rq *rq) int cpu; rcu_read_lock(); - if (test_idle_cores(core, true)) + if (test_idle_cores(core)) goto unlock; for_each_cpu(cpu, cpu_smt_mask(core)) { @@ -6367,9 +6367,9 @@ static inline void set_idle_cores(int cpu, int val) { } -static inline bool test_idle_cores(int cpu, bool def) +static inline bool test_idle_cores(int cpu) { - return def; + return false; } static inline int select_idle_core(struct task_struct *p, int core, struct cpumask *cpus, int *idle_cpu) @@ -6608,7 +6608,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) return target; if (sched_smt_active()) { - has_idle_core = test_idle_cores(target, false); + has_idle_core = test_idle_cores(target); if (!has_idle_core && cpus_share_cache(prev, target)) { i = select_idle_smt(p, prev); From 96c1c0cfe493a7ed549169a6f044bbb83e490fb5 Mon Sep 17 00:00:00 2001 From: Abel Wu Date: Wed, 7 Sep 2022 19:20:00 +0800 Subject: [PATCH 26/41] sched/fair: Cleanup for SIS_PROP The sched-domain of this cpu is only used for some heuristics when SIS_PROP is enabled, and it should be irrelevant whether the local sd_llc is valid or not, since all we care about is target sd_llc if !SIS_PROP. Access the local domain only when there is a need. Signed-off-by: Abel Wu Signed-off-by: Peter Zijlstra (Intel) Acked-by: Mel Gorman Link: https://lore.kernel.org/r/20220907112000.1854-6-wuyun.abel@bytedance.com --- kernel/sched/fair.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 23b020c3d3a0..7bad641faef9 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -6396,19 +6396,19 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, bool struct sched_domain_shared *sd_share; struct rq *this_rq = this_rq(); int this = smp_processor_id(); - struct sched_domain *this_sd; + struct sched_domain *this_sd = NULL; u64 time = 0; - this_sd = rcu_dereference(*this_cpu_ptr(&sd_llc)); - if (!this_sd) - return -1; - cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr); if (sched_feat(SIS_PROP) && !has_idle_core) { u64 avg_cost, avg_idle, span_avg; unsigned long now = jiffies; + this_sd = rcu_dereference(*this_cpu_ptr(&sd_llc)); + if (!this_sd) + return -1; + /* * If we're busy, the assumption that the last idle period * predicts the future is flawed; age away the remaining @@ -6462,7 +6462,7 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, bool if (has_idle_core) set_idle_cores(target, false); - if (sched_feat(SIS_PROP) && !has_idle_core) { + if (sched_feat(SIS_PROP) && this_sd && !has_idle_core) { time = cpu_clock(this) - time; /* From 0b9d46fc5ef7a457cc635b30b010081228cb81ac Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 6 Sep 2022 12:33:04 +0200 Subject: [PATCH 27/41] sched: Rename task_running() to task_on_cpu() There is some ambiguity about task_running() in that it is unrelated to TASK_RUNNING but instead tests ->on_cpu. As such, rename the thing task_on_cpu(). Suggested-by: Ingo Molnar Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/Yxhkhn55uHZx+NGl@hirez.programming.kicks-ass.net --- kernel/sched/core.c | 10 +++++----- kernel/sched/core_sched.c | 2 +- kernel/sched/deadline.c | 6 +++--- kernel/sched/fair.c | 2 +- kernel/sched/rt.c | 6 +++--- kernel/sched/sched.h | 2 +- 6 files changed, 14 insertions(+), 14 deletions(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 7d289d87acf7..163018105bf8 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2777,7 +2777,7 @@ static int affine_move_task(struct rq *rq, struct task_struct *p, struct rq_flag return -EINVAL; } - if (task_running(rq, p) || READ_ONCE(p->__state) == TASK_WAKING) { + if (task_on_cpu(rq, p) || READ_ONCE(p->__state) == TASK_WAKING) { /* * MIGRATE_ENABLE gets here because 'p == current', but for * anything else we cannot do is_migration_disabled(), punt @@ -3289,11 +3289,11 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state * * NOTE! Since we don't hold any locks, it's not * even sure that "rq" stays as the right runqueue! - * But we don't care, since "task_running()" will + * But we don't care, since "task_on_cpu()" will * return false if the runqueue has changed and p * is actually now running somewhere else! */ - while (task_running(rq, p)) { + while (task_on_cpu(rq, p)) { if (match_state && unlikely(READ_ONCE(p->__state) != match_state)) return 0; cpu_relax(); @@ -3306,7 +3306,7 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state */ rq = task_rq_lock(p, &rf); trace_sched_wait_task(p); - running = task_running(rq, p); + running = task_on_cpu(rq, p); queued = task_on_rq_queued(p); ncsw = 0; if (!match_state || READ_ONCE(p->__state) == match_state) @@ -8648,7 +8648,7 @@ again: if (curr->sched_class != p->sched_class) goto out_unlock; - if (task_running(p_rq, p) || !task_is_running(p)) + if (task_on_cpu(p_rq, p) || !task_is_running(p)) goto out_unlock; yielded = curr->sched_class->yield_to_task(rq, p); diff --git a/kernel/sched/core_sched.c b/kernel/sched/core_sched.c index 1ec807fbde56..a57fd8f27498 100644 --- a/kernel/sched/core_sched.c +++ b/kernel/sched/core_sched.c @@ -88,7 +88,7 @@ static unsigned long sched_core_update_cookie(struct task_struct *p, * core has now entered/left forced idle state. Defer accounting to the * next scheduling edge, rather than always forcing a reschedule here. */ - if (task_running(rq, p)) + if (task_on_cpu(rq, p)) resched_curr(rq); task_rq_unlock(rq, p, &rf); diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index d0fe6a20a9c9..86dea6a05267 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -2092,7 +2092,7 @@ static void task_fork_dl(struct task_struct *p) static int pick_dl_task(struct rq *rq, struct task_struct *p, int cpu) { - if (!task_running(rq, p) && + if (!task_on_cpu(rq, p) && cpumask_test_cpu(cpu, &p->cpus_mask)) return 1; return 0; @@ -2244,7 +2244,7 @@ static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq) if (double_lock_balance(rq, later_rq)) { if (unlikely(task_rq(task) != rq || !cpumask_test_cpu(later_rq->cpu, &task->cpus_mask) || - task_running(rq, task) || + task_on_cpu(rq, task) || !dl_task(task) || !task_on_rq_queued(task))) { double_unlock_balance(rq, later_rq); @@ -2474,7 +2474,7 @@ skip: */ static void task_woken_dl(struct rq *rq, struct task_struct *p) { - if (!task_running(rq, p) && + if (!task_on_cpu(rq, p) && !test_tsk_need_resched(rq->curr) && p->nr_cpus_allowed > 1 && dl_task(rq->curr) && diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 7bad641faef9..4e5b171b1171 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -7935,7 +7935,7 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) /* Record that we found at least one task that could run on dst_cpu */ env->flags &= ~LBF_ALL_PINNED; - if (task_running(env->src_rq, p)) { + if (task_on_cpu(env->src_rq, p)) { schedstat_inc(p->stats.nr_failed_migrations_running); return 0; } diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 27e694cf3e37..d869bcf898cc 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -1845,7 +1845,7 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p) static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu) { - if (!task_running(rq, p) && + if (!task_on_cpu(rq, p) && cpumask_test_cpu(cpu, &p->cpus_mask)) return 1; @@ -2000,7 +2000,7 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) */ if (unlikely(task_rq(task) != rq || !cpumask_test_cpu(lowest_rq->cpu, &task->cpus_mask) || - task_running(rq, task) || + task_on_cpu(rq, task) || !rt_task(task) || !task_on_rq_queued(task))) { @@ -2458,7 +2458,7 @@ skip: */ static void task_woken_rt(struct rq *rq, struct task_struct *p) { - bool need_to_push = !task_running(rq, p) && + bool need_to_push = !task_on_cpu(rq, p) && !test_tsk_need_resched(rq->curr) && p->nr_cpus_allowed > 1 && (dl_task(rq->curr) || rt_task(rq->curr)) && diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index b75ac74986fb..91b2c7ec53bd 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -2051,7 +2051,7 @@ static inline int task_current(struct rq *rq, struct task_struct *p) return rq->curr == p; } -static inline int task_running(struct rq *rq, struct task_struct *p) +static inline int task_on_cpu(struct rq *rq, struct task_struct *p) { #ifdef CONFIG_SMP return p->on_cpu; From 5950e5d574c636a07dd21a872c2f8b41f6d20c55 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 22 Aug 2022 13:18:17 +0200 Subject: [PATCH 28/41] freezer: Have {,un}lock_system_sleep() save/restore flags Rafael explained that the reason for having both PF_NOFREEZE and PF_FREEZER_SKIP is that {,un}lock_system_sleep() is callable from kthread context that has previously called set_freezable(). In preparation of merging the flags, have {,un}lock_system_slee() save and restore current->flags. Signed-off-by: Peter Zijlstra (Intel) Acked-by: Rafael J. Wysocki Link: https://lore.kernel.org/r/20220822114648.725003428@infradead.org --- drivers/acpi/x86/s2idle.c | 12 +++++++---- drivers/scsi/scsi_transport_spi.c | 7 ++++--- include/linux/suspend.h | 8 +++---- kernel/power/hibernate.c | 35 +++++++++++++++++++------------ kernel/power/main.c | 16 ++++++++------ kernel/power/suspend.c | 12 +++++++---- kernel/power/user.c | 28 ++++++++++++++----------- 7 files changed, 72 insertions(+), 46 deletions(-) diff --git a/drivers/acpi/x86/s2idle.c b/drivers/acpi/x86/s2idle.c index f9ac12b778e6..1d1d6c80f6a8 100644 --- a/drivers/acpi/x86/s2idle.c +++ b/drivers/acpi/x86/s2idle.c @@ -541,12 +541,14 @@ void acpi_s2idle_setup(void) int acpi_register_lps0_dev(struct acpi_s2idle_dev_ops *arg) { + unsigned int sleep_flags; + if (!lps0_device_handle || sleep_no_lps0) return -ENODEV; - lock_system_sleep(); + sleep_flags = lock_system_sleep(); list_add(&arg->list_node, &lps0_s2idle_devops_head); - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); return 0; } @@ -554,12 +556,14 @@ EXPORT_SYMBOL_GPL(acpi_register_lps0_dev); void acpi_unregister_lps0_dev(struct acpi_s2idle_dev_ops *arg) { + unsigned int sleep_flags; + if (!lps0_device_handle || sleep_no_lps0) return; - lock_system_sleep(); + sleep_flags = lock_system_sleep(); list_del(&arg->list_node); - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); } EXPORT_SYMBOL_GPL(acpi_unregister_lps0_dev); diff --git a/drivers/scsi/scsi_transport_spi.c b/drivers/scsi/scsi_transport_spi.c index bd72c38d7bfc..f569cf0095c2 100644 --- a/drivers/scsi/scsi_transport_spi.c +++ b/drivers/scsi/scsi_transport_spi.c @@ -998,8 +998,9 @@ void spi_dv_device(struct scsi_device *sdev) { struct scsi_target *starget = sdev->sdev_target; - u8 *buffer; const int len = SPI_MAX_ECHO_BUFFER_SIZE*2; + unsigned int sleep_flags; + u8 *buffer; /* * Because this function and the power management code both call @@ -1007,7 +1008,7 @@ spi_dv_device(struct scsi_device *sdev) * while suspend or resume is in progress. Hence the * lock/unlock_system_sleep() calls. */ - lock_system_sleep(); + sleep_flags = lock_system_sleep(); if (scsi_autopm_get_device(sdev)) goto unlock_system_sleep; @@ -1058,7 +1059,7 @@ put_autopm: scsi_autopm_put_device(sdev); unlock_system_sleep: - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); } EXPORT_SYMBOL(spi_dv_device); diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 70f2921e2e70..34fb72c0db85 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -510,8 +510,8 @@ extern bool pm_save_wakeup_count(unsigned int count); extern void pm_wakep_autosleep_enabled(bool set); extern void pm_print_active_wakeup_sources(void); -extern void lock_system_sleep(void); -extern void unlock_system_sleep(void); +extern unsigned int lock_system_sleep(void); +extern void unlock_system_sleep(unsigned int); #else /* !CONFIG_PM_SLEEP */ @@ -534,8 +534,8 @@ static inline void pm_system_wakeup(void) {} static inline void pm_wakeup_clear(bool reset) {} static inline void pm_system_irq_wakeup(unsigned int irq_number) {} -static inline void lock_system_sleep(void) {} -static inline void unlock_system_sleep(void) {} +static inline unsigned int lock_system_sleep(void) { return 0; } +static inline void unlock_system_sleep(unsigned int flags) {} #endif /* !CONFIG_PM_SLEEP */ diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index 89c71fce225d..f58a0aa92310 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -92,20 +92,24 @@ bool hibernation_available(void) */ void hibernation_set_ops(const struct platform_hibernation_ops *ops) { + unsigned int sleep_flags; + if (ops && !(ops->begin && ops->end && ops->pre_snapshot && ops->prepare && ops->finish && ops->enter && ops->pre_restore && ops->restore_cleanup && ops->leave)) { WARN_ON(1); return; } - lock_system_sleep(); + + sleep_flags = lock_system_sleep(); + hibernation_ops = ops; if (ops) hibernation_mode = HIBERNATION_PLATFORM; else if (hibernation_mode == HIBERNATION_PLATFORM) hibernation_mode = HIBERNATION_SHUTDOWN; - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); } EXPORT_SYMBOL_GPL(hibernation_set_ops); @@ -713,6 +717,7 @@ static int load_image_and_restore(void) int hibernate(void) { bool snapshot_test = false; + unsigned int sleep_flags; int error; if (!hibernation_available()) { @@ -720,7 +725,7 @@ int hibernate(void) return -EPERM; } - lock_system_sleep(); + sleep_flags = lock_system_sleep(); /* The snapshot device should not be opened while we're running */ if (!hibernate_acquire()) { error = -EBUSY; @@ -794,7 +799,7 @@ int hibernate(void) pm_restore_console(); hibernate_release(); Unlock: - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); pr_info("hibernation exit\n"); return error; @@ -809,9 +814,10 @@ int hibernate(void) */ int hibernate_quiet_exec(int (*func)(void *data), void *data) { + unsigned int sleep_flags; int error; - lock_system_sleep(); + sleep_flags = lock_system_sleep(); if (!hibernate_acquire()) { error = -EBUSY; @@ -891,7 +897,7 @@ restore: hibernate_release(); unlock: - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); return error; } @@ -1100,11 +1106,12 @@ static ssize_t disk_show(struct kobject *kobj, struct kobj_attribute *attr, static ssize_t disk_store(struct kobject *kobj, struct kobj_attribute *attr, const char *buf, size_t n) { + int mode = HIBERNATION_INVALID; + unsigned int sleep_flags; int error = 0; - int i; int len; char *p; - int mode = HIBERNATION_INVALID; + int i; if (!hibernation_available()) return -EPERM; @@ -1112,7 +1119,7 @@ static ssize_t disk_store(struct kobject *kobj, struct kobj_attribute *attr, p = memchr(buf, '\n', n); len = p ? p - buf : n; - lock_system_sleep(); + sleep_flags = lock_system_sleep(); for (i = HIBERNATION_FIRST; i <= HIBERNATION_MAX; i++) { if (len == strlen(hibernation_modes[i]) && !strncmp(buf, hibernation_modes[i], len)) { @@ -1142,7 +1149,7 @@ static ssize_t disk_store(struct kobject *kobj, struct kobj_attribute *attr, if (!error) pm_pr_dbg("Hibernation mode set to '%s'\n", hibernation_modes[mode]); - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); return error ? error : n; } @@ -1158,9 +1165,10 @@ static ssize_t resume_show(struct kobject *kobj, struct kobj_attribute *attr, static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr, const char *buf, size_t n) { - dev_t res; + unsigned int sleep_flags; int len = n; char *name; + dev_t res; if (len && buf[len-1] == '\n') len--; @@ -1173,9 +1181,10 @@ static ssize_t resume_store(struct kobject *kobj, struct kobj_attribute *attr, if (!res) return -EINVAL; - lock_system_sleep(); + sleep_flags = lock_system_sleep(); swsusp_resume_device = res; - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); + pm_pr_dbg("Configured hibernation resume from disk to %u\n", swsusp_resume_device); noresume = 0; diff --git a/kernel/power/main.c b/kernel/power/main.c index e3694034b753..dae3d17919e6 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -21,14 +21,16 @@ #ifdef CONFIG_PM_SLEEP -void lock_system_sleep(void) +unsigned int lock_system_sleep(void) { + unsigned int flags = current->flags; current->flags |= PF_FREEZER_SKIP; mutex_lock(&system_transition_mutex); + return flags; } EXPORT_SYMBOL_GPL(lock_system_sleep); -void unlock_system_sleep(void) +void unlock_system_sleep(unsigned int flags) { /* * Don't use freezer_count() because we don't want the call to @@ -46,7 +48,8 @@ void unlock_system_sleep(void) * Which means, if we use try_to_freeze() here, it would make them * enter the refrigerator, thus causing hibernation to lockup. */ - current->flags &= ~PF_FREEZER_SKIP; + if (!(flags & PF_FREEZER_SKIP)) + current->flags &= ~PF_FREEZER_SKIP; mutex_unlock(&system_transition_mutex); } EXPORT_SYMBOL_GPL(unlock_system_sleep); @@ -263,16 +266,17 @@ static ssize_t pm_test_show(struct kobject *kobj, struct kobj_attribute *attr, static ssize_t pm_test_store(struct kobject *kobj, struct kobj_attribute *attr, const char *buf, size_t n) { + unsigned int sleep_flags; const char * const *s; + int error = -EINVAL; int level; char *p; int len; - int error = -EINVAL; p = memchr(buf, '\n', n); len = p ? p - buf : n; - lock_system_sleep(); + sleep_flags = lock_system_sleep(); level = TEST_FIRST; for (s = &pm_tests[level]; level <= TEST_MAX; s++, level++) @@ -282,7 +286,7 @@ static ssize_t pm_test_store(struct kobject *kobj, struct kobj_attribute *attr, break; } - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); return error ? error : n; } diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 827075944d28..0d5f1fde6a3a 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -75,9 +75,11 @@ EXPORT_SYMBOL_GPL(pm_suspend_default_s2idle); void s2idle_set_ops(const struct platform_s2idle_ops *ops) { - lock_system_sleep(); + unsigned int sleep_flags; + + sleep_flags = lock_system_sleep(); s2idle_ops = ops; - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); } static void s2idle_begin(void) @@ -200,7 +202,9 @@ __setup("mem_sleep_default=", mem_sleep_default_setup); */ void suspend_set_ops(const struct platform_suspend_ops *ops) { - lock_system_sleep(); + unsigned int sleep_flags; + + sleep_flags = lock_system_sleep(); suspend_ops = ops; @@ -216,7 +220,7 @@ void suspend_set_ops(const struct platform_suspend_ops *ops) mem_sleep_current = PM_SUSPEND_MEM; } - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); } EXPORT_SYMBOL_GPL(suspend_set_ops); diff --git a/kernel/power/user.c b/kernel/power/user.c index d43c2aa583b2..3a4e70366f35 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -47,12 +47,13 @@ int is_hibernate_resume_dev(dev_t dev) static int snapshot_open(struct inode *inode, struct file *filp) { struct snapshot_data *data; + unsigned int sleep_flags; int error; if (!hibernation_available()) return -EPERM; - lock_system_sleep(); + sleep_flags = lock_system_sleep(); if (!hibernate_acquire()) { error = -EBUSY; @@ -98,7 +99,7 @@ static int snapshot_open(struct inode *inode, struct file *filp) data->dev = 0; Unlock: - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); return error; } @@ -106,8 +107,9 @@ static int snapshot_open(struct inode *inode, struct file *filp) static int snapshot_release(struct inode *inode, struct file *filp) { struct snapshot_data *data; + unsigned int sleep_flags; - lock_system_sleep(); + sleep_flags = lock_system_sleep(); swsusp_free(); data = filp->private_data; @@ -124,7 +126,7 @@ static int snapshot_release(struct inode *inode, struct file *filp) PM_POST_HIBERNATION : PM_POST_RESTORE); hibernate_release(); - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); return 0; } @@ -132,11 +134,12 @@ static int snapshot_release(struct inode *inode, struct file *filp) static ssize_t snapshot_read(struct file *filp, char __user *buf, size_t count, loff_t *offp) { - struct snapshot_data *data; - ssize_t res; loff_t pg_offp = *offp & ~PAGE_MASK; + struct snapshot_data *data; + unsigned int sleep_flags; + ssize_t res; - lock_system_sleep(); + sleep_flags = lock_system_sleep(); data = filp->private_data; if (!data->ready) { @@ -157,7 +160,7 @@ static ssize_t snapshot_read(struct file *filp, char __user *buf, *offp += res; Unlock: - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); return res; } @@ -165,16 +168,17 @@ static ssize_t snapshot_read(struct file *filp, char __user *buf, static ssize_t snapshot_write(struct file *filp, const char __user *buf, size_t count, loff_t *offp) { - struct snapshot_data *data; - ssize_t res; loff_t pg_offp = *offp & ~PAGE_MASK; + struct snapshot_data *data; + unsigned long sleep_flags; + ssize_t res; if (need_wait) { wait_for_device_probe(); need_wait = false; } - lock_system_sleep(); + sleep_flags = lock_system_sleep(); data = filp->private_data; @@ -196,7 +200,7 @@ static ssize_t snapshot_write(struct file *filp, const char __user *buf, if (res > 0) *offp += res; unlock: - unlock_system_sleep(); + unlock_system_sleep(sleep_flags); return res; } From 1fbcaa923ce2d7e6de17abd74fa076dc1e0be1a2 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 22 Aug 2022 13:18:18 +0200 Subject: [PATCH 29/41] freezer,umh: Clean up freezer/initrd interaction handle_initrd() marks itself as PF_FREEZER_SKIP in order to ensure that the UMH, which is going to freeze the system, doesn't indefinitely wait for it's caller. Rework things by adding UMH_FREEZABLE to indicate the completion is freezable. Signed-off-by: Peter Zijlstra (Intel) Acked-by: Rafael J. Wysocki Link: https://lore.kernel.org/r/20220822114648.791019324@infradead.org --- include/linux/umh.h | 9 +++++---- init/do_mounts_initrd.c | 10 +--------- kernel/umh.c | 8 ++++++++ 3 files changed, 14 insertions(+), 13 deletions(-) diff --git a/include/linux/umh.h b/include/linux/umh.h index 244aff638220..5d1f6129b847 100644 --- a/include/linux/umh.h +++ b/include/linux/umh.h @@ -11,10 +11,11 @@ struct cred; struct file; -#define UMH_NO_WAIT 0 /* don't wait at all */ -#define UMH_WAIT_EXEC 1 /* wait for the exec, but not the process */ -#define UMH_WAIT_PROC 2 /* wait for the process to complete */ -#define UMH_KILLABLE 4 /* wait for EXEC/PROC killable */ +#define UMH_NO_WAIT 0x00 /* don't wait at all */ +#define UMH_WAIT_EXEC 0x01 /* wait for the exec, but not the process */ +#define UMH_WAIT_PROC 0x02 /* wait for the process to complete */ +#define UMH_KILLABLE 0x04 /* wait for EXEC/PROC killable */ +#define UMH_FREEZABLE 0x08 /* wait for EXEC/PROC freezable */ struct subprocess_info { struct work_struct work; diff --git a/init/do_mounts_initrd.c b/init/do_mounts_initrd.c index 327962ea354c..34731241377d 100644 --- a/init/do_mounts_initrd.c +++ b/init/do_mounts_initrd.c @@ -99,19 +99,11 @@ static void __init handle_initrd(void) init_mkdir("/old", 0700); init_chdir("/old"); - /* - * In case that a resume from disk is carried out by linuxrc or one of - * its children, we need to tell the freezer not to wait for us. - */ - current->flags |= PF_FREEZER_SKIP; - info = call_usermodehelper_setup("/linuxrc", argv, envp_init, GFP_KERNEL, init_linuxrc, NULL, NULL); if (!info) return; - call_usermodehelper_exec(info, UMH_WAIT_PROC); - - current->flags &= ~PF_FREEZER_SKIP; + call_usermodehelper_exec(info, UMH_WAIT_PROC|UMH_FREEZABLE); /* move initrd to rootfs' /old */ init_mount("..", ".", NULL, MS_MOVE, NULL); diff --git a/kernel/umh.c b/kernel/umh.c index b989736e8707..8945eaf4c671 100644 --- a/kernel/umh.c +++ b/kernel/umh.c @@ -28,6 +28,7 @@ #include #include #include +#include #include @@ -436,6 +437,9 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info, int wait) if (wait == UMH_NO_WAIT) /* task has freed sub_info */ goto unlock; + if (wait & UMH_FREEZABLE) + freezer_do_not_count(); + if (wait & UMH_KILLABLE) { retval = wait_for_completion_killable(&done); if (!retval) @@ -448,6 +452,10 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info, int wait) } wait_for_completion(&done); + + if (wait & UMH_FREEZABLE) + freezer_count(); + wait_done: retval = sub_info->retval; out: From 9204a97f7ae862fc8a3330ec8335917534c3fb63 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 22 Aug 2022 13:18:19 +0200 Subject: [PATCH 30/41] sched: Change wait_task_inactive()s match_state Make wait_task_inactive()'s @match_state work like ttwu()'s @state. That is, instead of an equal comparison, use it as a mask. This allows matching multiple block conditions. (removes the unlikely; it doesn't make sense how it's only part of the condition) Signed-off-by: Peter Zijlstra (Intel) Link: https://lore.kernel.org/r/20220822114648.856734578@infradead.org --- kernel/sched/core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 163018105bf8..43d71c696125 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -3294,7 +3294,7 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state * is actually now running somewhere else! */ while (task_on_cpu(rq, p)) { - if (match_state && unlikely(READ_ONCE(p->__state) != match_state)) + if (match_state && !(READ_ONCE(p->__state) & match_state)) return 0; cpu_relax(); } @@ -3309,7 +3309,7 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state running = task_on_cpu(rq, p); queued = task_on_rq_queued(p); ncsw = 0; - if (!match_state || READ_ONCE(p->__state) == match_state) + if (!match_state || (READ_ONCE(p->__state) & match_state)) ncsw = p->nvcsw | LONG_MIN; /* sets MSB */ task_rq_unlock(rq, p, &rf); From f9fc8cad9728124cefe8844fb53d1814c92c6bfc Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 6 Sep 2022 12:39:55 +0200 Subject: [PATCH 31/41] sched: Add TASK_ANY for wait_task_inactive() Now that wait_task_inactive()'s @match_state argument is a mask (like ttwu()) it is possible to replace the special !match_state case with an 'all-states' value such that any blocked state will match. Suggested-by: Ingo Molnar (mingo@kernel.org> Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/YxhkzfuFTvRnpUaH@hirez.programming.kicks-ass.net --- drivers/powercap/idle_inject.c | 2 +- fs/coredump.c | 2 +- include/linux/sched.h | 2 ++ kernel/sched/core.c | 16 ++++++++-------- 4 files changed, 12 insertions(+), 10 deletions(-) diff --git a/drivers/powercap/idle_inject.c b/drivers/powercap/idle_inject.c index a20bf12f3ce3..999e218d7793 100644 --- a/drivers/powercap/idle_inject.c +++ b/drivers/powercap/idle_inject.c @@ -254,7 +254,7 @@ void idle_inject_stop(struct idle_inject_device *ii_dev) iit = per_cpu_ptr(&idle_inject_thread, cpu); iit->should_run = 0; - wait_task_inactive(iit->tsk, 0); + wait_task_inactive(iit->tsk, TASK_ANY); } cpu_hotplug_enable(); diff --git a/fs/coredump.c b/fs/coredump.c index 9f4aae202109..3d9054c49a11 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -412,7 +412,7 @@ static int coredump_wait(int exit_code, struct core_state *core_state) */ ptr = core_state->dumper.next; while (ptr != NULL) { - wait_task_inactive(ptr->task, 0); + wait_task_inactive(ptr->task, TASK_ANY); ptr = ptr->next; } } diff --git a/include/linux/sched.h b/include/linux/sched.h index e7b2f8a5c711..0facaadbae71 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -101,6 +101,8 @@ struct task_group; #define TASK_RTLOCK_WAIT 0x1000 #define TASK_STATE_MAX 0x2000 +#define TASK_ANY (TASK_STATE_MAX-1) + /* Convenience macros for the sake of set_current_state: */ #define TASK_KILLABLE (TASK_WAKEKILL | TASK_UNINTERRUPTIBLE) #define TASK_STOPPED (TASK_WAKEKILL | __TASK_STOPPED) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 43d71c696125..ea26526b45bb 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -3253,12 +3253,12 @@ out: /* * wait_task_inactive - wait for a thread to unschedule. * - * If @match_state is nonzero, it's the @p->state value just checked and - * not expected to change. If it changes, i.e. @p might have woken up, - * then return zero. When we succeed in waiting for @p to be off its CPU, - * we return a positive number (its total switch count). If a second call - * a short while later returns the same number, the caller can be sure that - * @p has remained unscheduled the whole time. + * Wait for the thread to block in any of the states set in @match_state. + * If it changes, i.e. @p might have woken up, then return zero. When we + * succeed in waiting for @p to be off its CPU, we return a positive number + * (its total switch count). If a second call a short while later returns the + * same number, the caller can be sure that @p has remained unscheduled the + * whole time. * * The caller must ensure that the task *will* unschedule sometime soon, * else this function might spin for a *long* time. This function can't @@ -3294,7 +3294,7 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state * is actually now running somewhere else! */ while (task_on_cpu(rq, p)) { - if (match_state && !(READ_ONCE(p->__state) & match_state)) + if (!(READ_ONCE(p->__state) & match_state)) return 0; cpu_relax(); } @@ -3309,7 +3309,7 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state running = task_on_cpu(rq, p); queued = task_on_rq_queued(p); ncsw = 0; - if (!match_state || (READ_ONCE(p->__state) & match_state)) + if (READ_ONCE(p->__state) & match_state) ncsw = p->nvcsw | LONG_MIN; /* sets MSB */ task_rq_unlock(rq, p, &rf); From 929659acea03db6411a32de9037abab9f856f586 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 22 Aug 2022 13:18:20 +0200 Subject: [PATCH 32/41] sched/completion: Add wait_for_completion_state() Allows waiting with a custom @state. Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Ingo Molnar Link: https://lore.kernel.org/r/20220822114648.922711674@infradead.org --- include/linux/completion.h | 1 + kernel/sched/completion.c | 12 ++++++++++++ 2 files changed, 13 insertions(+) diff --git a/include/linux/completion.h b/include/linux/completion.h index 51d9ab079629..62b32b19e0a8 100644 --- a/include/linux/completion.h +++ b/include/linux/completion.h @@ -103,6 +103,7 @@ extern void wait_for_completion(struct completion *); extern void wait_for_completion_io(struct completion *); extern int wait_for_completion_interruptible(struct completion *x); extern int wait_for_completion_killable(struct completion *x); +extern int wait_for_completion_state(struct completion *x, unsigned int state); extern unsigned long wait_for_completion_timeout(struct completion *x, unsigned long timeout); extern unsigned long wait_for_completion_io_timeout(struct completion *x, diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c index 35f15c26ed54..d57a5c1c1cd9 100644 --- a/kernel/sched/completion.c +++ b/kernel/sched/completion.c @@ -204,6 +204,7 @@ EXPORT_SYMBOL(wait_for_completion_io_timeout); int __sched wait_for_completion_interruptible(struct completion *x) { long t = wait_for_common(x, MAX_SCHEDULE_TIMEOUT, TASK_INTERRUPTIBLE); + if (t == -ERESTARTSYS) return t; return 0; @@ -241,12 +242,23 @@ EXPORT_SYMBOL(wait_for_completion_interruptible_timeout); int __sched wait_for_completion_killable(struct completion *x) { long t = wait_for_common(x, MAX_SCHEDULE_TIMEOUT, TASK_KILLABLE); + if (t == -ERESTARTSYS) return t; return 0; } EXPORT_SYMBOL(wait_for_completion_killable); +int __sched wait_for_completion_state(struct completion *x, unsigned int state) +{ + long t = wait_for_common(x, MAX_SCHEDULE_TIMEOUT, state); + + if (t == -ERESTARTSYS) + return t; + return 0; +} +EXPORT_SYMBOL(wait_for_completion_state); + /** * wait_for_completion_killable_timeout: - waits for completion of a task (w/(to,killable)) * @x: holds the state of this particular completion From 3f884a10ab41efe2d495bf8ec8d443d70c500a60 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 22 Aug 2022 13:18:21 +0200 Subject: [PATCH 33/41] sched/wait: Add wait_event_state() Allows waiting with a custom @state. Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Ingo Molnar Link: https://lore.kernel.org/r/20220822114648.989212021@infradead.org --- include/linux/wait.h | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/include/linux/wait.h b/include/linux/wait.h index 58cfbf81447c..b926eb9f3e26 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -932,6 +932,34 @@ extern int do_wait_intr_irq(wait_queue_head_t *, wait_queue_entry_t *); __ret; \ }) +#define __wait_event_state(wq, condition, state) \ + ___wait_event(wq, condition, state, 0, 0, schedule()) + +/** + * wait_event_state - sleep until a condition gets true + * @wq_head: the waitqueue to wait on + * @condition: a C expression for the event to wait for + * @state: state to sleep in + * + * The process is put to sleep (@state) until the @condition evaluates to true + * or a signal is received (when allowed by @state). The @condition is checked + * each time the waitqueue @wq_head is woken up. + * + * wake_up() has to be called after changing any variable that could + * change the result of the wait condition. + * + * The function will return -ERESTARTSYS if it was interrupted by a signal + * (when allowed by @state) and 0 if @condition evaluated to true. + */ +#define wait_event_state(wq_head, condition, state) \ +({ \ + int __ret = 0; \ + might_sleep(); \ + if (!(condition)) \ + __ret = __wait_event_state(wq_head, condition, state); \ + __ret; \ +}) + #define __wait_event_killable_timeout(wq_head, condition, timeout) \ ___wait_event(wq_head, ___wait_cond_timeout(condition), \ TASK_KILLABLE, 0, timeout, \ From 9963e444f71e671bcbc30d61cf23a2c686ac7d05 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 6 Sep 2022 13:11:38 +0200 Subject: [PATCH 34/41] sched: Widen TAKS_state literals In preparation of adding more states, add a few 0s to the literals as we've just about ran out. Suggested-by: Ingo Molnar Signed-off-by: Peter Zijlstra (Intel) --- include/linux/sched.h | 30 +++++++++++++++--------------- 1 file changed, 15 insertions(+), 15 deletions(-) diff --git a/include/linux/sched.h b/include/linux/sched.h index 0facaadbae71..9fc23a2d9813 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -81,25 +81,25 @@ struct task_group; */ /* Used in tsk->state: */ -#define TASK_RUNNING 0x0000 -#define TASK_INTERRUPTIBLE 0x0001 -#define TASK_UNINTERRUPTIBLE 0x0002 -#define __TASK_STOPPED 0x0004 -#define __TASK_TRACED 0x0008 +#define TASK_RUNNING 0x00000000 +#define TASK_INTERRUPTIBLE 0x00000001 +#define TASK_UNINTERRUPTIBLE 0x00000002 +#define __TASK_STOPPED 0x00000004 +#define __TASK_TRACED 0x00000008 /* Used in tsk->exit_state: */ -#define EXIT_DEAD 0x0010 -#define EXIT_ZOMBIE 0x0020 +#define EXIT_DEAD 0x00000010 +#define EXIT_ZOMBIE 0x00000020 #define EXIT_TRACE (EXIT_ZOMBIE | EXIT_DEAD) /* Used in tsk->state again: */ -#define TASK_PARKED 0x0040 -#define TASK_DEAD 0x0080 -#define TASK_WAKEKILL 0x0100 -#define TASK_WAKING 0x0200 -#define TASK_NOLOAD 0x0400 -#define TASK_NEW 0x0800 +#define TASK_PARKED 0x00000040 +#define TASK_DEAD 0x00000080 +#define TASK_WAKEKILL 0x00000100 +#define TASK_WAKING 0x00000200 +#define TASK_NOLOAD 0x00000400 +#define TASK_NEW 0x00000800 /* RT specific auxilliary flag to mark RT lock waiters */ -#define TASK_RTLOCK_WAIT 0x1000 -#define TASK_STATE_MAX 0x2000 +#define TASK_RTLOCK_WAIT 0x00001000 +#define TASK_STATE_MAX 0x00002000 #define TASK_ANY (TASK_STATE_MAX-1) From f5d39b020809146cc28e6e73369bf8065e0310aa Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 22 Aug 2022 13:18:22 +0200 Subject: [PATCH 35/41] freezer,sched: Rewrite core freezer logic Rewrite the core freezer to behave better wrt thawing and be simpler in general. By replacing PF_FROZEN with TASK_FROZEN, a special block state, it is ensured frozen tasks stay frozen until thawed and don't randomly wake up early, as is currently possible. As such, it does away with PF_FROZEN and PF_FREEZER_SKIP, freeing up two PF_flags (yay!). Specifically; the current scheme works a little like: freezer_do_not_count(); schedule(); freezer_count(); And either the task is blocked, or it lands in try_to_freezer() through freezer_count(). Now, when it is blocked, the freezer considers it frozen and continues. However, on thawing, once pm_freezing is cleared, freezer_count() stops working, and any random/spurious wakeup will let a task run before its time. That is, thawing tries to thaw things in explicit order; kernel threads and workqueues before doing bringing SMP back before userspace etc.. However due to the above mentioned races it is entirely possible for userspace tasks to thaw (by accident) before SMP is back. This can be a fatal problem in asymmetric ISA architectures (eg ARMv9) where the userspace task requires a special CPU to run. As said; replace this with a special task state TASK_FROZEN and add the following state transitions: TASK_FREEZABLE -> TASK_FROZEN __TASK_STOPPED -> TASK_FROZEN __TASK_TRACED -> TASK_FROZEN The new TASK_FREEZABLE can be set on any state part of TASK_NORMAL (IOW. TASK_INTERRUPTIBLE and TASK_UNINTERRUPTIBLE) -- any such state is already required to deal with spurious wakeups and the freezer causes one such when thawing the task (since the original state is lost). The special __TASK_{STOPPED,TRACED} states *can* be restored since their canonical state is in ->jobctl. With this, frozen tasks need an explicit TASK_FROZEN wakeup and are free of undue (early / spurious) wakeups. Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Ingo Molnar Acked-by: Rafael J. Wysocki Link: https://lore.kernel.org/r/20220822114649.055452969@infradead.org --- drivers/android/binder.c | 4 +- drivers/media/pci/pt3/pt3.c | 4 +- fs/cifs/inode.c | 4 +- fs/cifs/transport.c | 5 +- fs/coredump.c | 5 +- fs/nfs/file.c | 3 +- fs/nfs/inode.c | 12 +- fs/nfs/nfs3proc.c | 3 +- fs/nfs/nfs4proc.c | 14 +- fs/nfs/nfs4state.c | 3 +- fs/nfs/pnfs.c | 4 +- fs/xfs/xfs_trans_ail.c | 8 +- include/linux/freezer.h | 245 ++------------------------------- include/linux/sched.h | 13 +- include/linux/sunrpc/sched.h | 7 +- include/linux/wait.h | 12 +- kernel/cgroup/legacy_freezer.c | 23 ++-- kernel/exit.c | 4 +- kernel/fork.c | 5 +- kernel/freezer.c | 133 +++++++++++++----- kernel/futex/waitwake.c | 8 +- kernel/hung_task.c | 4 +- kernel/power/main.c | 6 +- kernel/power/process.c | 10 +- kernel/ptrace.c | 2 +- kernel/sched/core.c | 2 +- kernel/signal.c | 14 +- kernel/time/hrtimer.c | 4 +- kernel/umh.c | 20 ++- mm/khugepaged.c | 4 +- net/sunrpc/sched.c | 12 +- net/unix/af_unix.c | 8 +- 32 files changed, 210 insertions(+), 395 deletions(-) diff --git a/drivers/android/binder.c b/drivers/android/binder.c index c964d7c8c384..50197ef4b787 100644 --- a/drivers/android/binder.c +++ b/drivers/android/binder.c @@ -4247,10 +4247,9 @@ static int binder_wait_for_work(struct binder_thread *thread, struct binder_proc *proc = thread->proc; int ret = 0; - freezer_do_not_count(); binder_inner_proc_lock(proc); for (;;) { - prepare_to_wait(&thread->wait, &wait, TASK_INTERRUPTIBLE); + prepare_to_wait(&thread->wait, &wait, TASK_INTERRUPTIBLE|TASK_FREEZABLE); if (binder_has_work_ilocked(thread, do_proc_work)) break; if (do_proc_work) @@ -4267,7 +4266,6 @@ static int binder_wait_for_work(struct binder_thread *thread, } finish_wait(&thread->wait, &wait); binder_inner_proc_unlock(proc); - freezer_count(); return ret; } diff --git a/drivers/media/pci/pt3/pt3.c b/drivers/media/pci/pt3/pt3.c index 0d51bdf01f43..f6deac85962e 100644 --- a/drivers/media/pci/pt3/pt3.c +++ b/drivers/media/pci/pt3/pt3.c @@ -445,8 +445,8 @@ static int pt3_fetch_thread(void *data) pt3_proc_dma(adap); delay = ktime_set(0, PT3_FETCH_DELAY * NSEC_PER_MSEC); - set_current_state(TASK_UNINTERRUPTIBLE); - freezable_schedule_hrtimeout_range(&delay, + set_current_state(TASK_UNINTERRUPTIBLE|TASK_FREEZABLE); + schedule_hrtimeout_range(&delay, PT3_FETCH_DELAY_DELTA * NSEC_PER_MSEC, HRTIMER_MODE_REL); } diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index eeeaba3dec05..54e6bfc6a5a6 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -2326,7 +2326,7 @@ cifs_invalidate_mapping(struct inode *inode) static int cifs_wait_bit_killable(struct wait_bit_key *key, int mode) { - freezable_schedule_unsafe(); + schedule(); if (signal_pending_state(mode, current)) return -ERESTARTSYS; return 0; @@ -2344,7 +2344,7 @@ cifs_revalidate_mapping(struct inode *inode) return 0; rc = wait_on_bit_lock_action(flags, CIFS_INO_LOCK, cifs_wait_bit_killable, - TASK_KILLABLE); + TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); if (rc) return rc; diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index de7aeced7e16..eda09fba1550 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -757,8 +757,9 @@ wait_for_response(struct TCP_Server_Info *server, struct mid_q_entry *midQ) { int error; - error = wait_event_freezekillable_unsafe(server->response_q, - midQ->mid_state != MID_REQUEST_SUBMITTED); + error = wait_event_state(server->response_q, + midQ->mid_state != MID_REQUEST_SUBMITTED, + (TASK_KILLABLE|TASK_FREEZABLE_UNSAFE)); if (error < 0) return -ERESTARTSYS; diff --git a/fs/coredump.c b/fs/coredump.c index 3d9054c49a11..9ce59f212a4a 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -402,9 +402,8 @@ static int coredump_wait(int exit_code, struct core_state *core_state) if (core_waiters > 0) { struct core_thread *ptr; - freezer_do_not_count(); - wait_for_completion(&core_state->startup); - freezer_count(); + wait_for_completion_state(&core_state->startup, + TASK_UNINTERRUPTIBLE|TASK_FREEZABLE); /* * Wait for all the threads to become inactive, so that * all the thread context (extended register state, like diff --git a/fs/nfs/file.c b/fs/nfs/file.c index d2bcd4834c0e..e8301a372738 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -570,7 +570,8 @@ static vm_fault_t nfs_vm_page_mkwrite(struct vm_fault *vmf) } wait_on_bit_action(&NFS_I(inode)->flags, NFS_INO_INVALIDATING, - nfs_wait_bit_killable, TASK_KILLABLE); + nfs_wait_bit_killable, + TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); lock_page(page); mapping = page_file_mapping(page); diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index b4e46b0ffa2d..a6058f5787a5 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -72,18 +72,13 @@ nfs_fattr_to_ino_t(struct nfs_fattr *fattr) return nfs_fileid_to_ino_t(fattr->fileid); } -static int nfs_wait_killable(int mode) +int nfs_wait_bit_killable(struct wait_bit_key *key, int mode) { - freezable_schedule_unsafe(); + schedule(); if (signal_pending_state(mode, current)) return -ERESTARTSYS; return 0; } - -int nfs_wait_bit_killable(struct wait_bit_key *key, int mode) -{ - return nfs_wait_killable(mode); -} EXPORT_SYMBOL_GPL(nfs_wait_bit_killable); /** @@ -1331,7 +1326,8 @@ int nfs_clear_invalid_mapping(struct address_space *mapping) */ for (;;) { ret = wait_on_bit_action(bitlock, NFS_INO_INVALIDATING, - nfs_wait_bit_killable, TASK_KILLABLE); + nfs_wait_bit_killable, + TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); if (ret) goto out; spin_lock(&inode->i_lock); diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 1597eef40d54..2e7579626cf0 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -36,7 +36,8 @@ nfs3_rpc_wrapper(struct rpc_clnt *clnt, struct rpc_message *msg, int flags) res = rpc_call_sync(clnt, msg, flags); if (res != -EJUKEBOX) break; - freezable_schedule_timeout_killable_unsafe(NFS_JUKEBOX_RETRY_TIME); + __set_current_state(TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); + schedule_timeout(NFS_JUKEBOX_RETRY_TIME); res = -ERESTARTSYS; } while (!fatal_signal_pending(current)); return res; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 3ed14a2a84a4..4553803538e5 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -416,8 +416,8 @@ static int nfs4_delay_killable(long *timeout) { might_sleep(); - freezable_schedule_timeout_killable_unsafe( - nfs4_update_delay(timeout)); + __set_current_state(TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); + schedule_timeout(nfs4_update_delay(timeout)); if (!__fatal_signal_pending(current)) return 0; return -EINTR; @@ -427,7 +427,8 @@ static int nfs4_delay_interruptible(long *timeout) { might_sleep(); - freezable_schedule_timeout_interruptible_unsafe(nfs4_update_delay(timeout)); + __set_current_state(TASK_INTERRUPTIBLE|TASK_FREEZABLE_UNSAFE); + schedule_timeout(nfs4_update_delay(timeout)); if (!signal_pending(current)) return 0; return __fatal_signal_pending(current) ? -EINTR :-ERESTARTSYS; @@ -7406,7 +7407,8 @@ nfs4_retry_setlk_simple(struct nfs4_state *state, int cmd, status = nfs4_proc_setlk(state, cmd, request); if ((status != -EAGAIN) || IS_SETLK(cmd)) break; - freezable_schedule_timeout_interruptible(timeout); + __set_current_state(TASK_INTERRUPTIBLE|TASK_FREEZABLE); + schedule_timeout(timeout); timeout *= 2; timeout = min_t(unsigned long, NFS4_LOCK_MAXTIMEOUT, timeout); status = -ERESTARTSYS; @@ -7474,10 +7476,8 @@ nfs4_retry_setlk(struct nfs4_state *state, int cmd, struct file_lock *request) break; status = -ERESTARTSYS; - freezer_do_not_count(); - wait_woken(&waiter.wait, TASK_INTERRUPTIBLE, + wait_woken(&waiter.wait, TASK_INTERRUPTIBLE|TASK_FREEZABLE, NFS4_LOCK_MAXTIMEOUT); - freezer_count(); } while (!signalled()); remove_wait_queue(q, &waiter.wait); diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 9bab3e9c702a..7e185f7eb260 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -1314,7 +1314,8 @@ int nfs4_wait_clnt_recover(struct nfs_client *clp) refcount_inc(&clp->cl_count); res = wait_on_bit_action(&clp->cl_state, NFS4CLNT_MANAGER_RUNNING, - nfs_wait_bit_killable, TASK_KILLABLE); + nfs_wait_bit_killable, + TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); if (res) goto out; if (clp->cl_cons_state < 0) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 41a9b6b58fb9..70706db317f3 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1908,7 +1908,7 @@ static int pnfs_prepare_to_retry_layoutget(struct pnfs_layout_hdr *lo) pnfs_layoutcommit_inode(lo->plh_inode, false); return wait_on_bit_action(&lo->plh_flags, NFS_LAYOUT_RETURN, nfs_wait_bit_killable, - TASK_KILLABLE); + TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); } static void nfs_layoutget_begin(struct pnfs_layout_hdr *lo) @@ -3193,7 +3193,7 @@ pnfs_layoutcommit_inode(struct inode *inode, bool sync) status = wait_on_bit_lock_action(&nfsi->flags, NFS_INO_LAYOUTCOMMITTING, nfs_wait_bit_killable, - TASK_KILLABLE); + TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); if (status) goto out; } diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c index d3a97a028560..16fbf2a1144c 100644 --- a/fs/xfs/xfs_trans_ail.c +++ b/fs/xfs/xfs_trans_ail.c @@ -602,9 +602,9 @@ xfsaild( while (1) { if (tout && tout <= 20) - set_current_state(TASK_KILLABLE); + set_current_state(TASK_KILLABLE|TASK_FREEZABLE); else - set_current_state(TASK_INTERRUPTIBLE); + set_current_state(TASK_INTERRUPTIBLE|TASK_FREEZABLE); /* * Check kthread_should_stop() after we set the task state to @@ -653,14 +653,14 @@ xfsaild( ailp->ail_target == ailp->ail_target_prev && list_empty(&ailp->ail_buf_list)) { spin_unlock(&ailp->ail_lock); - freezable_schedule(); + schedule(); tout = 0; continue; } spin_unlock(&ailp->ail_lock); if (tout) - freezable_schedule_timeout(msecs_to_jiffies(tout)); + schedule_timeout(msecs_to_jiffies(tout)); __set_current_state(TASK_RUNNING); diff --git a/include/linux/freezer.h b/include/linux/freezer.h index 0621c5f86c39..b303472255be 100644 --- a/include/linux/freezer.h +++ b/include/linux/freezer.h @@ -8,9 +8,11 @@ #include #include #include +#include #ifdef CONFIG_FREEZER -extern atomic_t system_freezing_cnt; /* nr of freezing conds in effect */ +DECLARE_STATIC_KEY_FALSE(freezer_active); + extern bool pm_freezing; /* PM freezing in effect */ extern bool pm_nosig_freezing; /* PM nosig freezing in effect */ @@ -22,10 +24,7 @@ extern unsigned int freeze_timeout_msecs; /* * Check if a process has been frozen */ -static inline bool frozen(struct task_struct *p) -{ - return p->flags & PF_FROZEN; -} +extern bool frozen(struct task_struct *p); extern bool freezing_slow_path(struct task_struct *p); @@ -34,9 +33,10 @@ extern bool freezing_slow_path(struct task_struct *p); */ static inline bool freezing(struct task_struct *p) { - if (likely(!atomic_read(&system_freezing_cnt))) - return false; - return freezing_slow_path(p); + if (static_branch_unlikely(&freezer_active)) + return freezing_slow_path(p); + + return false; } /* Takes and releases task alloc lock using task_lock() */ @@ -48,23 +48,14 @@ extern int freeze_kernel_threads(void); extern void thaw_processes(void); extern void thaw_kernel_threads(void); -/* - * DO NOT ADD ANY NEW CALLERS OF THIS FUNCTION - * If try_to_freeze causes a lockdep warning it means the caller may deadlock - */ -static inline bool try_to_freeze_unsafe(void) +static inline bool try_to_freeze(void) { might_sleep(); if (likely(!freezing(current))) return false; - return __refrigerator(false); -} - -static inline bool try_to_freeze(void) -{ if (!(current->flags & PF_NOFREEZE)) debug_check_no_locks_held(); - return try_to_freeze_unsafe(); + return __refrigerator(false); } extern bool freeze_task(struct task_struct *p); @@ -79,195 +70,6 @@ static inline bool cgroup_freezing(struct task_struct *task) } #endif /* !CONFIG_CGROUP_FREEZER */ -/* - * The PF_FREEZER_SKIP flag should be set by a vfork parent right before it - * calls wait_for_completion(&vfork) and reset right after it returns from this - * function. Next, the parent should call try_to_freeze() to freeze itself - * appropriately in case the child has exited before the freezing of tasks is - * complete. However, we don't want kernel threads to be frozen in unexpected - * places, so we allow them to block freeze_processes() instead or to set - * PF_NOFREEZE if needed. Fortunately, in the ____call_usermodehelper() case the - * parent won't really block freeze_processes(), since ____call_usermodehelper() - * (the child) does a little before exec/exit and it can't be frozen before - * waking up the parent. - */ - - -/** - * freezer_do_not_count - tell freezer to ignore %current - * - * Tell freezers to ignore the current task when determining whether the - * target frozen state is reached. IOW, the current task will be - * considered frozen enough by freezers. - * - * The caller shouldn't do anything which isn't allowed for a frozen task - * until freezer_cont() is called. Usually, freezer[_do_not]_count() pair - * wrap a scheduling operation and nothing much else. - */ -static inline void freezer_do_not_count(void) -{ - current->flags |= PF_FREEZER_SKIP; -} - -/** - * freezer_count - tell freezer to stop ignoring %current - * - * Undo freezer_do_not_count(). It tells freezers that %current should be - * considered again and tries to freeze if freezing condition is already in - * effect. - */ -static inline void freezer_count(void) -{ - current->flags &= ~PF_FREEZER_SKIP; - /* - * If freezing is in progress, the following paired with smp_mb() - * in freezer_should_skip() ensures that either we see %true - * freezing() or freezer_should_skip() sees !PF_FREEZER_SKIP. - */ - smp_mb(); - try_to_freeze(); -} - -/* DO NOT ADD ANY NEW CALLERS OF THIS FUNCTION */ -static inline void freezer_count_unsafe(void) -{ - current->flags &= ~PF_FREEZER_SKIP; - smp_mb(); - try_to_freeze_unsafe(); -} - -/** - * freezer_should_skip - whether to skip a task when determining frozen - * state is reached - * @p: task in quesion - * - * This function is used by freezers after establishing %true freezing() to - * test whether a task should be skipped when determining the target frozen - * state is reached. IOW, if this function returns %true, @p is considered - * frozen enough. - */ -static inline bool freezer_should_skip(struct task_struct *p) -{ - /* - * The following smp_mb() paired with the one in freezer_count() - * ensures that either freezer_count() sees %true freezing() or we - * see cleared %PF_FREEZER_SKIP and return %false. This makes it - * impossible for a task to slip frozen state testing after - * clearing %PF_FREEZER_SKIP. - */ - smp_mb(); - return p->flags & PF_FREEZER_SKIP; -} - -/* - * These functions are intended to be used whenever you want allow a sleeping - * task to be frozen. Note that neither return any clear indication of - * whether a freeze event happened while in this function. - */ - -/* Like schedule(), but should not block the freezer. */ -static inline void freezable_schedule(void) -{ - freezer_do_not_count(); - schedule(); - freezer_count(); -} - -/* DO NOT ADD ANY NEW CALLERS OF THIS FUNCTION */ -static inline void freezable_schedule_unsafe(void) -{ - freezer_do_not_count(); - schedule(); - freezer_count_unsafe(); -} - -/* - * Like schedule_timeout(), but should not block the freezer. Do not - * call this with locks held. - */ -static inline long freezable_schedule_timeout(long timeout) -{ - long __retval; - freezer_do_not_count(); - __retval = schedule_timeout(timeout); - freezer_count(); - return __retval; -} - -/* - * Like schedule_timeout_interruptible(), but should not block the freezer. Do not - * call this with locks held. - */ -static inline long freezable_schedule_timeout_interruptible(long timeout) -{ - long __retval; - freezer_do_not_count(); - __retval = schedule_timeout_interruptible(timeout); - freezer_count(); - return __retval; -} - -/* DO NOT ADD ANY NEW CALLERS OF THIS FUNCTION */ -static inline long freezable_schedule_timeout_interruptible_unsafe(long timeout) -{ - long __retval; - - freezer_do_not_count(); - __retval = schedule_timeout_interruptible(timeout); - freezer_count_unsafe(); - return __retval; -} - -/* Like schedule_timeout_killable(), but should not block the freezer. */ -static inline long freezable_schedule_timeout_killable(long timeout) -{ - long __retval; - freezer_do_not_count(); - __retval = schedule_timeout_killable(timeout); - freezer_count(); - return __retval; -} - -/* DO NOT ADD ANY NEW CALLERS OF THIS FUNCTION */ -static inline long freezable_schedule_timeout_killable_unsafe(long timeout) -{ - long __retval; - freezer_do_not_count(); - __retval = schedule_timeout_killable(timeout); - freezer_count_unsafe(); - return __retval; -} - -/* - * Like schedule_hrtimeout_range(), but should not block the freezer. Do not - * call this with locks held. - */ -static inline int freezable_schedule_hrtimeout_range(ktime_t *expires, - u64 delta, const enum hrtimer_mode mode) -{ - int __retval; - freezer_do_not_count(); - __retval = schedule_hrtimeout_range(expires, delta, mode); - freezer_count(); - return __retval; -} - -/* - * Freezer-friendly wrappers around wait_event_interruptible(), - * wait_event_killable() and wait_event_interruptible_timeout(), originally - * defined in - */ - -/* DO NOT ADD ANY NEW CALLERS OF THIS FUNCTION */ -#define wait_event_freezekillable_unsafe(wq, condition) \ -({ \ - int __retval; \ - freezer_do_not_count(); \ - __retval = wait_event_killable(wq, (condition)); \ - freezer_count_unsafe(); \ - __retval; \ -}) - #else /* !CONFIG_FREEZER */ static inline bool frozen(struct task_struct *p) { return false; } static inline bool freezing(struct task_struct *p) { return false; } @@ -281,35 +83,8 @@ static inline void thaw_kernel_threads(void) {} static inline bool try_to_freeze(void) { return false; } -static inline void freezer_do_not_count(void) {} -static inline void freezer_count(void) {} -static inline int freezer_should_skip(struct task_struct *p) { return 0; } static inline void set_freezable(void) {} -#define freezable_schedule() schedule() - -#define freezable_schedule_unsafe() schedule() - -#define freezable_schedule_timeout(timeout) schedule_timeout(timeout) - -#define freezable_schedule_timeout_interruptible(timeout) \ - schedule_timeout_interruptible(timeout) - -#define freezable_schedule_timeout_interruptible_unsafe(timeout) \ - schedule_timeout_interruptible(timeout) - -#define freezable_schedule_timeout_killable(timeout) \ - schedule_timeout_killable(timeout) - -#define freezable_schedule_timeout_killable_unsafe(timeout) \ - schedule_timeout_killable(timeout) - -#define freezable_schedule_hrtimeout_range(expires, delta, mode) \ - schedule_hrtimeout_range(expires, delta, mode) - -#define wait_event_freezekillable_unsafe(wq, condition) \ - wait_event_killable(wq, condition) - #endif /* !CONFIG_FREEZER */ #endif /* FREEZER_H_INCLUDED */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 9fc23a2d9813..4c91efd6df82 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -97,12 +97,19 @@ struct task_group; #define TASK_WAKING 0x00000200 #define TASK_NOLOAD 0x00000400 #define TASK_NEW 0x00000800 -/* RT specific auxilliary flag to mark RT lock waiters */ #define TASK_RTLOCK_WAIT 0x00001000 -#define TASK_STATE_MAX 0x00002000 +#define TASK_FREEZABLE 0x00002000 +#define __TASK_FREEZABLE_UNSAFE (0x00004000 * IS_ENABLED(CONFIG_LOCKDEP)) +#define TASK_FROZEN 0x00008000 +#define TASK_STATE_MAX 0x00010000 #define TASK_ANY (TASK_STATE_MAX-1) +/* + * DO NOT ADD ANY NEW USERS ! + */ +#define TASK_FREEZABLE_UNSAFE (TASK_FREEZABLE | __TASK_FREEZABLE_UNSAFE) + /* Convenience macros for the sake of set_current_state: */ #define TASK_KILLABLE (TASK_WAKEKILL | TASK_UNINTERRUPTIBLE) #define TASK_STOPPED (TASK_WAKEKILL | __TASK_STOPPED) @@ -1716,7 +1723,6 @@ extern struct pid *cad_pid; #define PF_NPROC_EXCEEDED 0x00001000 /* set_user() noticed that RLIMIT_NPROC was exceeded */ #define PF_USED_MATH 0x00002000 /* If unset the fpu must be initialized before use */ #define PF_NOFREEZE 0x00008000 /* This thread should not be frozen */ -#define PF_FROZEN 0x00010000 /* Frozen for system suspend */ #define PF_KSWAPD 0x00020000 /* I am kswapd */ #define PF_MEMALLOC_NOFS 0x00040000 /* All allocation requests will inherit GFP_NOFS */ #define PF_MEMALLOC_NOIO 0x00080000 /* All allocation requests will inherit GFP_NOIO */ @@ -1727,7 +1733,6 @@ extern struct pid *cad_pid; #define PF_NO_SETAFFINITY 0x04000000 /* Userland is not allowed to meddle with cpus_mask */ #define PF_MCE_EARLY 0x08000000 /* Early kill for mce process policy */ #define PF_MEMALLOC_PIN 0x10000000 /* Allocation context constrained to zones which allow long term pinning. */ -#define PF_FREEZER_SKIP 0x40000000 /* Freezer should not count it as freezable */ #define PF_SUSPEND_TASK 0x80000000 /* This thread called freeze_processes() and should not be frozen */ /* diff --git a/include/linux/sunrpc/sched.h b/include/linux/sunrpc/sched.h index acc62647317c..baeca2f564dc 100644 --- a/include/linux/sunrpc/sched.h +++ b/include/linux/sunrpc/sched.h @@ -252,7 +252,7 @@ int rpc_malloc(struct rpc_task *); void rpc_free(struct rpc_task *); int rpciod_up(void); void rpciod_down(void); -int __rpc_wait_for_completion_task(struct rpc_task *task, wait_bit_action_f *); +int rpc_wait_for_completion_task(struct rpc_task *task); #if IS_ENABLED(CONFIG_SUNRPC_DEBUG) struct net; void rpc_show_tasks(struct net *); @@ -264,11 +264,6 @@ extern struct workqueue_struct *xprtiod_workqueue; void rpc_prepare_task(struct rpc_task *task); gfp_t rpc_task_gfp_mask(void); -static inline int rpc_wait_for_completion_task(struct rpc_task *task) -{ - return __rpc_wait_for_completion_task(task, NULL); -} - #if IS_ENABLED(CONFIG_SUNRPC_DEBUG) || IS_ENABLED(CONFIG_TRACEPOINTS) static inline const char * rpc_qname(const struct rpc_wait_queue *q) { diff --git a/include/linux/wait.h b/include/linux/wait.h index b926eb9f3e26..14ad8a0e9fac 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -361,8 +361,8 @@ do { \ } while (0) #define __wait_event_freezable(wq_head, condition) \ - ___wait_event(wq_head, condition, TASK_INTERRUPTIBLE, 0, 0, \ - freezable_schedule()) + ___wait_event(wq_head, condition, (TASK_INTERRUPTIBLE|TASK_FREEZABLE), \ + 0, 0, schedule()) /** * wait_event_freezable - sleep (or freeze) until a condition gets true @@ -420,8 +420,8 @@ do { \ #define __wait_event_freezable_timeout(wq_head, condition, timeout) \ ___wait_event(wq_head, ___wait_cond_timeout(condition), \ - TASK_INTERRUPTIBLE, 0, timeout, \ - __ret = freezable_schedule_timeout(__ret)) + (TASK_INTERRUPTIBLE|TASK_FREEZABLE), 0, timeout, \ + __ret = schedule_timeout(__ret)) /* * like wait_event_timeout() -- except it uses TASK_INTERRUPTIBLE to avoid @@ -642,8 +642,8 @@ do { \ #define __wait_event_freezable_exclusive(wq, condition) \ - ___wait_event(wq, condition, TASK_INTERRUPTIBLE, 1, 0, \ - freezable_schedule()) + ___wait_event(wq, condition, (TASK_INTERRUPTIBLE|TASK_FREEZABLE), 1, 0,\ + schedule()) #define wait_event_freezable_exclusive(wq, condition) \ ({ \ diff --git a/kernel/cgroup/legacy_freezer.c b/kernel/cgroup/legacy_freezer.c index 08236798d173..1b6b21851e9d 100644 --- a/kernel/cgroup/legacy_freezer.c +++ b/kernel/cgroup/legacy_freezer.c @@ -113,7 +113,7 @@ static int freezer_css_online(struct cgroup_subsys_state *css) if (parent && (parent->state & CGROUP_FREEZING)) { freezer->state |= CGROUP_FREEZING_PARENT | CGROUP_FROZEN; - atomic_inc(&system_freezing_cnt); + static_branch_inc(&freezer_active); } mutex_unlock(&freezer_mutex); @@ -134,7 +134,7 @@ static void freezer_css_offline(struct cgroup_subsys_state *css) mutex_lock(&freezer_mutex); if (freezer->state & CGROUP_FREEZING) - atomic_dec(&system_freezing_cnt); + static_branch_dec(&freezer_active); freezer->state = 0; @@ -179,6 +179,7 @@ static void freezer_attach(struct cgroup_taskset *tset) __thaw_task(task); } else { freeze_task(task); + /* clear FROZEN and propagate upwards */ while (freezer && (freezer->state & CGROUP_FROZEN)) { freezer->state &= ~CGROUP_FROZEN; @@ -271,16 +272,8 @@ static void update_if_frozen(struct cgroup_subsys_state *css) css_task_iter_start(css, 0, &it); while ((task = css_task_iter_next(&it))) { - if (freezing(task)) { - /* - * freezer_should_skip() indicates that the task - * should be skipped when determining freezing - * completion. Consider it frozen in addition to - * the usual frozen condition. - */ - if (!frozen(task) && !freezer_should_skip(task)) - goto out_iter_end; - } + if (freezing(task) && !frozen(task)) + goto out_iter_end; } freezer->state |= CGROUP_FROZEN; @@ -357,7 +350,7 @@ static void freezer_apply_state(struct freezer *freezer, bool freeze, if (freeze) { if (!(freezer->state & CGROUP_FREEZING)) - atomic_inc(&system_freezing_cnt); + static_branch_inc(&freezer_active); freezer->state |= state; freeze_cgroup(freezer); } else { @@ -366,9 +359,9 @@ static void freezer_apply_state(struct freezer *freezer, bool freeze, freezer->state &= ~state; if (!(freezer->state & CGROUP_FREEZING)) { - if (was_freezing) - atomic_dec(&system_freezing_cnt); freezer->state &= ~CGROUP_FROZEN; + if (was_freezing) + static_branch_dec(&freezer_active); unfreeze_cgroup(freezer); } } diff --git a/kernel/exit.c b/kernel/exit.c index 84021b24f79e..ba683f8d292a 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -374,10 +374,10 @@ static void coredump_task_exit(struct task_struct *tsk) complete(&core_state->startup); for (;;) { - set_current_state(TASK_UNINTERRUPTIBLE); + set_current_state(TASK_UNINTERRUPTIBLE|TASK_FREEZABLE); if (!self.task) /* see coredump_finish() */ break; - freezable_schedule(); + schedule(); } __set_current_state(TASK_RUNNING); } diff --git a/kernel/fork.c b/kernel/fork.c index 90c85b17bf69..a1de2f5e305d 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1420,13 +1420,12 @@ static void complete_vfork_done(struct task_struct *tsk) static int wait_for_vfork_done(struct task_struct *child, struct completion *vfork) { + unsigned int state = TASK_UNINTERRUPTIBLE|TASK_KILLABLE|TASK_FREEZABLE; int killed; - freezer_do_not_count(); cgroup_enter_frozen(); - killed = wait_for_completion_killable(vfork); + killed = wait_for_completion_state(vfork, state); cgroup_leave_frozen(false); - freezer_count(); if (killed) { task_lock(child); diff --git a/kernel/freezer.c b/kernel/freezer.c index 45ab36ffd0e7..4fad0e6fca64 100644 --- a/kernel/freezer.c +++ b/kernel/freezer.c @@ -13,10 +13,11 @@ #include /* total number of freezing conditions in effect */ -atomic_t system_freezing_cnt = ATOMIC_INIT(0); -EXPORT_SYMBOL(system_freezing_cnt); +DEFINE_STATIC_KEY_FALSE(freezer_active); +EXPORT_SYMBOL(freezer_active); -/* indicate whether PM freezing is in effect, protected by +/* + * indicate whether PM freezing is in effect, protected by * system_transition_mutex */ bool pm_freezing; @@ -29,7 +30,7 @@ static DEFINE_SPINLOCK(freezer_lock); * freezing_slow_path - slow path for testing whether a task needs to be frozen * @p: task to be tested * - * This function is called by freezing() if system_freezing_cnt isn't zero + * This function is called by freezing() if freezer_active isn't zero * and tests whether @p needs to enter and stay in frozen state. Can be * called under any context. The freezers are responsible for ensuring the * target tasks see the updated state. @@ -52,41 +53,40 @@ bool freezing_slow_path(struct task_struct *p) } EXPORT_SYMBOL(freezing_slow_path); +bool frozen(struct task_struct *p) +{ + return READ_ONCE(p->__state) & TASK_FROZEN; +} + /* Refrigerator is place where frozen processes are stored :-). */ bool __refrigerator(bool check_kthr_stop) { - /* Hmm, should we be allowed to suspend when there are realtime - processes around? */ + unsigned int state = get_current_state(); bool was_frozen = false; - unsigned int save = get_current_state(); pr_debug("%s entered refrigerator\n", current->comm); + WARN_ON_ONCE(state && !(state & TASK_NORMAL)); + for (;;) { - set_current_state(TASK_UNINTERRUPTIBLE); + bool freeze; + + set_current_state(TASK_FROZEN); spin_lock_irq(&freezer_lock); - current->flags |= PF_FROZEN; - if (!freezing(current) || - (check_kthr_stop && kthread_should_stop())) - current->flags &= ~PF_FROZEN; + freeze = freezing(current) && !(check_kthr_stop && kthread_should_stop()); spin_unlock_irq(&freezer_lock); - if (!(current->flags & PF_FROZEN)) + if (!freeze) break; + was_frozen = true; schedule(); } + __set_current_state(TASK_RUNNING); pr_debug("%s left refrigerator\n", current->comm); - /* - * Restore saved task state before returning. The mb'd version - * needs to be used; otherwise, it might silently break - * synchronization which depends on ordered task state change. - */ - set_current_state(save); - return was_frozen; } EXPORT_SYMBOL(__refrigerator); @@ -101,6 +101,44 @@ static void fake_signal_wake_up(struct task_struct *p) } } +static int __set_task_frozen(struct task_struct *p, void *arg) +{ + unsigned int state = READ_ONCE(p->__state); + + if (p->on_rq) + return 0; + + if (p != current && task_curr(p)) + return 0; + + if (!(state & (TASK_FREEZABLE | __TASK_STOPPED | __TASK_TRACED))) + return 0; + + /* + * Only TASK_NORMAL can be augmented with TASK_FREEZABLE, since they + * can suffer spurious wakeups. + */ + if (state & TASK_FREEZABLE) + WARN_ON_ONCE(!(state & TASK_NORMAL)); + +#ifdef CONFIG_LOCKDEP + /* + * It's dangerous to freeze with locks held; there be dragons there. + */ + if (!(state & __TASK_FREEZABLE_UNSAFE)) + WARN_ON_ONCE(debug_locks && p->lockdep_depth); +#endif + + WRITE_ONCE(p->__state, TASK_FROZEN); + return TASK_FROZEN; +} + +static bool __freeze_task(struct task_struct *p) +{ + /* TASK_FREEZABLE|TASK_STOPPED|TASK_TRACED -> TASK_FROZEN */ + return task_call_func(p, __set_task_frozen, NULL); +} + /** * freeze_task - send a freeze request to given task * @p: task to send the request to @@ -116,20 +154,8 @@ bool freeze_task(struct task_struct *p) { unsigned long flags; - /* - * This check can race with freezer_do_not_count, but worst case that - * will result in an extra wakeup being sent to the task. It does not - * race with freezer_count(), the barriers in freezer_count() and - * freezer_should_skip() ensure that either freezer_count() sees - * freezing == true in try_to_freeze() and freezes, or - * freezer_should_skip() sees !PF_FREEZE_SKIP and freezes the task - * normally. - */ - if (freezer_should_skip(p)) - return false; - spin_lock_irqsave(&freezer_lock, flags); - if (!freezing(p) || frozen(p)) { + if (!freezing(p) || frozen(p) || __freeze_task(p)) { spin_unlock_irqrestore(&freezer_lock, flags); return false; } @@ -137,19 +163,52 @@ bool freeze_task(struct task_struct *p) if (!(p->flags & PF_KTHREAD)) fake_signal_wake_up(p); else - wake_up_state(p, TASK_INTERRUPTIBLE); + wake_up_state(p, TASK_NORMAL); spin_unlock_irqrestore(&freezer_lock, flags); return true; } +/* + * The special task states (TASK_STOPPED, TASK_TRACED) keep their canonical + * state in p->jobctl. If either of them got a wakeup that was missed because + * TASK_FROZEN, then their canonical state reflects that and the below will + * refuse to restore the special state and instead issue the wakeup. + */ +static int __set_task_special(struct task_struct *p, void *arg) +{ + unsigned int state = 0; + + if (p->jobctl & JOBCTL_TRACED) + state = TASK_TRACED; + + else if (p->jobctl & JOBCTL_STOPPED) + state = TASK_STOPPED; + + if (state) + WRITE_ONCE(p->__state, state); + + return state; +} + void __thaw_task(struct task_struct *p) { - unsigned long flags; + unsigned long flags, flags2; spin_lock_irqsave(&freezer_lock, flags); - if (frozen(p)) - wake_up_process(p); + if (WARN_ON_ONCE(freezing(p))) + goto unlock; + + if (lock_task_sighand(p, &flags2)) { + /* TASK_FROZEN -> TASK_{STOPPED,TRACED} */ + bool ret = task_call_func(p, __set_task_special, NULL); + unlock_task_sighand(p, &flags2); + if (ret) + goto unlock; + } + + wake_up_state(p, TASK_FROZEN); +unlock: spin_unlock_irqrestore(&freezer_lock, flags); } diff --git a/kernel/futex/waitwake.c b/kernel/futex/waitwake.c index 4ce0923f1ce3..ba01b9408203 100644 --- a/kernel/futex/waitwake.c +++ b/kernel/futex/waitwake.c @@ -334,7 +334,7 @@ void futex_wait_queue(struct futex_hash_bucket *hb, struct futex_q *q, * futex_queue() calls spin_unlock() upon completion, both serializing * access to the hash list and forcing another memory barrier. */ - set_current_state(TASK_INTERRUPTIBLE); + set_current_state(TASK_INTERRUPTIBLE|TASK_FREEZABLE); futex_queue(q, hb); /* Arm the timer */ @@ -352,7 +352,7 @@ void futex_wait_queue(struct futex_hash_bucket *hb, struct futex_q *q, * is no timeout, or if it has yet to expire. */ if (!timeout || timeout->task) - freezable_schedule(); + schedule(); } __set_current_state(TASK_RUNNING); } @@ -430,7 +430,7 @@ retry: return ret; } - set_current_state(TASK_INTERRUPTIBLE); + set_current_state(TASK_INTERRUPTIBLE|TASK_FREEZABLE); for (i = 0; i < count; i++) { u32 __user *uaddr = (u32 __user *)(unsigned long)vs[i].w.uaddr; @@ -504,7 +504,7 @@ static void futex_sleep_multiple(struct futex_vector *vs, unsigned int count, return; } - freezable_schedule(); + schedule(); } /** diff --git a/kernel/hung_task.c b/kernel/hung_task.c index bb2354f73ded..f1321c03c32a 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -95,8 +95,8 @@ static void check_hung_task(struct task_struct *t, unsigned long timeout) * Ensure the task is not frozen. * Also, skip vfork and any other user process that freezer should skip. */ - if (unlikely(t->flags & (PF_FROZEN | PF_FREEZER_SKIP))) - return; + if (unlikely(READ_ONCE(t->__state) & (TASK_FREEZABLE | TASK_FROZEN))) + return; /* * When a freshly created task is scheduled once, changes its state to diff --git a/kernel/power/main.c b/kernel/power/main.c index dae3d17919e6..31ec4a9b9d70 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -24,7 +24,7 @@ unsigned int lock_system_sleep(void) { unsigned int flags = current->flags; - current->flags |= PF_FREEZER_SKIP; + current->flags |= PF_NOFREEZE; mutex_lock(&system_transition_mutex); return flags; } @@ -48,8 +48,8 @@ void unlock_system_sleep(unsigned int flags) * Which means, if we use try_to_freeze() here, it would make them * enter the refrigerator, thus causing hibernation to lockup. */ - if (!(flags & PF_FREEZER_SKIP)) - current->flags &= ~PF_FREEZER_SKIP; + if (!(flags & PF_NOFREEZE)) + current->flags &= ~PF_NOFREEZE; mutex_unlock(&system_transition_mutex); } EXPORT_SYMBOL_GPL(unlock_system_sleep); diff --git a/kernel/power/process.c b/kernel/power/process.c index 3068601e585a..ddd9988327fe 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -50,8 +50,7 @@ static int try_to_freeze_tasks(bool user_only) if (p == current || !freeze_task(p)) continue; - if (!freezer_should_skip(p)) - todo++; + todo++; } read_unlock(&tasklist_lock); @@ -96,8 +95,7 @@ static int try_to_freeze_tasks(bool user_only) if (!wakeup || pm_debug_messages_on) { read_lock(&tasklist_lock); for_each_process_thread(g, p) { - if (p != current && !freezer_should_skip(p) - && freezing(p) && !frozen(p)) + if (p != current && freezing(p) && !frozen(p)) sched_show_task(p); } read_unlock(&tasklist_lock); @@ -129,7 +127,7 @@ int freeze_processes(void) current->flags |= PF_SUSPEND_TASK; if (!pm_freezing) - atomic_inc(&system_freezing_cnt); + static_branch_inc(&freezer_active); pm_wakeup_clear(0); pr_info("Freezing user space processes ... "); @@ -190,7 +188,7 @@ void thaw_processes(void) trace_suspend_resume(TPS("thaw_processes"), 0, true); if (pm_freezing) - atomic_dec(&system_freezing_cnt); + static_branch_dec(&freezer_active); pm_freezing = false; pm_nosig_freezing = false; diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 1893d909e45c..54482193e1ed 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -269,7 +269,7 @@ static int ptrace_check_attach(struct task_struct *child, bool ignore_state) read_unlock(&tasklist_lock); if (!ret && !ignore_state && - WARN_ON_ONCE(!wait_task_inactive(child, __TASK_TRACED))) + WARN_ON_ONCE(!wait_task_inactive(child, __TASK_TRACED|TASK_FROZEN))) ret = -ESRCH; return ret; diff --git a/kernel/sched/core.c b/kernel/sched/core.c index ea26526b45bb..2b85d1b5fe0c 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -6428,7 +6428,7 @@ static void __sched notrace __schedule(unsigned int sched_mode) prev->sched_contributes_to_load = (prev_state & TASK_UNINTERRUPTIBLE) && !(prev_state & TASK_NOLOAD) && - !(prev->flags & PF_FROZEN); + !(prev_state & TASK_FROZEN); if (prev->sched_contributes_to_load) rq->nr_uninterruptible++; diff --git a/kernel/signal.c b/kernel/signal.c index 6f86fda5e432..c2e90d6ca0d1 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -2304,7 +2304,7 @@ static int ptrace_stop(int exit_code, int why, unsigned long message, read_unlock(&tasklist_lock); cgroup_enter_frozen(); preempt_enable_no_resched(); - freezable_schedule(); + schedule(); cgroup_leave_frozen(true); /* @@ -2473,7 +2473,7 @@ static bool do_signal_stop(int signr) /* Now we don't run again until woken by SIGCONT or SIGKILL */ cgroup_enter_frozen(); - freezable_schedule(); + schedule(); return true; } else { /* @@ -2548,11 +2548,11 @@ static void do_freezer_trap(void) * immediately (if there is a non-fatal signal pending), and * put the task into sleep. */ - __set_current_state(TASK_INTERRUPTIBLE); + __set_current_state(TASK_INTERRUPTIBLE|TASK_FREEZABLE); clear_thread_flag(TIF_SIGPENDING); spin_unlock_irq(¤t->sighand->siglock); cgroup_enter_frozen(); - freezable_schedule(); + schedule(); } static int ptrace_signal(int signr, kernel_siginfo_t *info, enum pid_type type) @@ -3600,9 +3600,9 @@ static int do_sigtimedwait(const sigset_t *which, kernel_siginfo_t *info, recalc_sigpending(); spin_unlock_irq(&tsk->sighand->siglock); - __set_current_state(TASK_INTERRUPTIBLE); - ret = freezable_schedule_hrtimeout_range(to, tsk->timer_slack_ns, - HRTIMER_MODE_REL); + __set_current_state(TASK_INTERRUPTIBLE|TASK_FREEZABLE); + ret = schedule_hrtimeout_range(to, tsk->timer_slack_ns, + HRTIMER_MODE_REL); spin_lock_irq(&tsk->sighand->siglock); __set_task_blocked(tsk, &tsk->real_blocked); sigemptyset(&tsk->real_blocked); diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index 23af5eca11b1..3ae661ab6260 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -2037,11 +2037,11 @@ static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mod struct restart_block *restart; do { - set_current_state(TASK_INTERRUPTIBLE); + set_current_state(TASK_INTERRUPTIBLE|TASK_FREEZABLE); hrtimer_sleeper_start_expires(t, mode); if (likely(t->task)) - freezable_schedule(); + schedule(); hrtimer_cancel(&t->timer); mode = HRTIMER_MODE_ABS; diff --git a/kernel/umh.c b/kernel/umh.c index 8945eaf4c671..850631518665 100644 --- a/kernel/umh.c +++ b/kernel/umh.c @@ -404,6 +404,7 @@ EXPORT_SYMBOL(call_usermodehelper_setup); */ int call_usermodehelper_exec(struct subprocess_info *sub_info, int wait) { + unsigned int state = TASK_UNINTERRUPTIBLE; DECLARE_COMPLETION_ONSTACK(done); int retval = 0; @@ -437,25 +438,22 @@ int call_usermodehelper_exec(struct subprocess_info *sub_info, int wait) if (wait == UMH_NO_WAIT) /* task has freed sub_info */ goto unlock; + if (wait & UMH_KILLABLE) + state |= TASK_KILLABLE; + if (wait & UMH_FREEZABLE) - freezer_do_not_count(); + state |= TASK_FREEZABLE; + + retval = wait_for_completion_state(&done, state); + if (!retval) + goto wait_done; if (wait & UMH_KILLABLE) { - retval = wait_for_completion_killable(&done); - if (!retval) - goto wait_done; - /* umh_complete() will see NULL and free sub_info */ if (xchg(&sub_info->complete, NULL)) goto unlock; - /* fallthrough, umh_complete() was already called */ } - wait_for_completion(&done); - - if (wait & UMH_FREEZABLE) - freezer_count(); - wait_done: retval = sub_info->retval; out: diff --git a/mm/khugepaged.c b/mm/khugepaged.c index 01f71786d530..77162fb7e90b 100644 --- a/mm/khugepaged.c +++ b/mm/khugepaged.c @@ -730,8 +730,8 @@ static void khugepaged_alloc_sleep(void) DEFINE_WAIT(wait); add_wait_queue(&khugepaged_wait, &wait); - freezable_schedule_timeout_interruptible( - msecs_to_jiffies(khugepaged_alloc_sleep_millisecs)); + __set_current_state(TASK_INTERRUPTIBLE|TASK_FREEZABLE); + schedule_timeout(msecs_to_jiffies(khugepaged_alloc_sleep_millisecs)); remove_wait_queue(&khugepaged_wait, &wait); } diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index 25b9221950ff..46cbf151a50b 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -269,7 +269,7 @@ EXPORT_SYMBOL_GPL(rpc_destroy_wait_queue); static int rpc_wait_bit_killable(struct wait_bit_key *key, int mode) { - freezable_schedule_unsafe(); + schedule(); if (signal_pending_state(mode, current)) return -ERESTARTSYS; return 0; @@ -333,14 +333,12 @@ static int rpc_complete_task(struct rpc_task *task) * to enforce taking of the wq->lock and hence avoid races with * rpc_complete_task(). */ -int __rpc_wait_for_completion_task(struct rpc_task *task, wait_bit_action_f *action) +int rpc_wait_for_completion_task(struct rpc_task *task) { - if (action == NULL) - action = rpc_wait_bit_killable; return out_of_line_wait_on_bit(&task->tk_runstate, RPC_TASK_ACTIVE, - action, TASK_KILLABLE); + rpc_wait_bit_killable, TASK_KILLABLE|TASK_FREEZABLE_UNSAFE); } -EXPORT_SYMBOL_GPL(__rpc_wait_for_completion_task); +EXPORT_SYMBOL_GPL(rpc_wait_for_completion_task); /* * Make an RPC task runnable. @@ -964,7 +962,7 @@ static void __rpc_execute(struct rpc_task *task) trace_rpc_task_sync_sleep(task, task->tk_action); status = out_of_line_wait_on_bit(&task->tk_runstate, RPC_TASK_QUEUED, rpc_wait_bit_killable, - TASK_KILLABLE); + TASK_KILLABLE|TASK_FREEZABLE); if (status < 0) { /* * When a sync task receives a signal, it exits with diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index bf338b782fc4..dda9eb1ab41f 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -2543,13 +2543,14 @@ static long unix_stream_data_wait(struct sock *sk, long timeo, struct sk_buff *last, unsigned int last_len, bool freezable) { + unsigned int state = TASK_INTERRUPTIBLE | freezable * TASK_FREEZABLE; struct sk_buff *tail; DEFINE_WAIT(wait); unix_state_lock(sk); for (;;) { - prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE); + prepare_to_wait(sk_sleep(sk), &wait, state); tail = skb_peek_tail(&sk->sk_receive_queue); if (tail != last || @@ -2562,10 +2563,7 @@ static long unix_stream_data_wait(struct sock *sk, long timeo, sk_set_bit(SOCKWQ_ASYNC_WAITDATA, sk); unix_state_unlock(sk); - if (freezable) - timeo = freezable_schedule_timeout(timeo); - else - timeo = schedule_timeout(timeo); + timeo = schedule_timeout(timeo); unix_state_lock(sk); if (sock_flag(sk, SOCK_DEAD)) From fb04563d1cae6f361892b4a339ad92100b1eb0d0 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 6 Sep 2022 13:27:32 +0200 Subject: [PATCH 36/41] sched: Show PF_flag holes Put placeholders in PF_flag so we can see how holey it is. Suggested-by: Ingo Molnar Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Ingo Molnar Link: https://lkml.kernel.org/r/YxctoffFFPXONESt@hirez.programming.kicks-ass.net --- include/linux/sched.h | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/include/linux/sched.h b/include/linux/sched.h index 4c91efd6df82..15e3bd96e4ce 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1722,7 +1722,9 @@ extern struct pid *cad_pid; #define PF_MEMALLOC 0x00000800 /* Allocating memory */ #define PF_NPROC_EXCEEDED 0x00001000 /* set_user() noticed that RLIMIT_NPROC was exceeded */ #define PF_USED_MATH 0x00002000 /* If unset the fpu must be initialized before use */ +#define PF__HOLE__00004000 0x00004000 #define PF_NOFREEZE 0x00008000 /* This thread should not be frozen */ +#define PF__HOLE__00010000 0x00010000 #define PF_KSWAPD 0x00020000 /* I am kswapd */ #define PF_MEMALLOC_NOFS 0x00040000 /* All allocation requests will inherit GFP_NOFS */ #define PF_MEMALLOC_NOIO 0x00080000 /* All allocation requests will inherit GFP_NOIO */ @@ -1730,9 +1732,14 @@ extern struct pid *cad_pid; * I am cleaning dirty pages from some other bdi. */ #define PF_KTHREAD 0x00200000 /* I am a kernel thread */ #define PF_RANDOMIZE 0x00400000 /* Randomize virtual address space */ +#define PF__HOLE__00800000 0x00800000 +#define PF__HOLE__01000000 0x01000000 +#define PF__HOLE__02000000 0x02000000 #define PF_NO_SETAFFINITY 0x04000000 /* Userland is not allowed to meddle with cpus_mask */ #define PF_MCE_EARLY 0x08000000 /* Early kill for mce process policy */ #define PF_MEMALLOC_PIN 0x10000000 /* Allocation context constrained to zones which allow long term pinning. */ +#define PF__HOLE__20000000 0x20000000 +#define PF__HOLE__40000000 0x40000000 #define PF_SUSPEND_TASK 0x80000000 /* This thread called freeze_processes() and should not be frozen */ /* From b0defa7ae03ecf91b8bfd10ede430cff12fcbd06 Mon Sep 17 00:00:00 2001 From: Vincent Guittot Date: Thu, 25 Aug 2022 14:27:23 +0200 Subject: [PATCH 37/41] sched/fair: Make sure to try to detach at least one movable task During load balance, we try at most env->loop_max time to move a task. But it can happen that the loop_max LRU tasks (ie tail of the cfs_tasks list) can't be moved to dst_cpu because of affinity. In this case, loop in the list until we found at least one. The maximum of detached tasks remained the same as before. Signed-off-by: Vincent Guittot Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20220825122726.20819-2-vincent.guittot@linaro.org --- kernel/sched/fair.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 4e5b171b1171..dae3bfae1bc0 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -8049,8 +8049,12 @@ static int detach_tasks(struct lb_env *env) p = list_last_entry(tasks, struct task_struct, se.group_node); env->loop++; - /* We've more or less seen every task there is, call it quits */ - if (env->loop > env->loop_max) + /* + * We've more or less seen every task there is, call it quits + * unless we haven't found any movable task yet. + */ + if (env->loop > env->loop_max && + !(env->flags & LBF_ALL_PINNED)) break; /* take a breather every nr_migrate tasks */ @@ -10179,7 +10183,9 @@ more_balance: if (env.flags & LBF_NEED_BREAK) { env.flags &= ~LBF_NEED_BREAK; - goto more_balance; + /* Stop if we tried all running tasks */ + if (env.loop < busiest->nr_running) + goto more_balance; } /* From c59862f8265f8060b6650ee1dc12159fe5c89779 Mon Sep 17 00:00:00 2001 From: Vincent Guittot Date: Thu, 25 Aug 2022 14:27:24 +0200 Subject: [PATCH 38/41] sched/fair: Cleanup loop_max and loop_break sched_nr_migrate_break is set to a fix value and never changes so we can replace it by a define SCHED_NR_MIGRATE_BREAK. Also, we adjust SCHED_NR_MIGRATE_BREAK to be aligned with the init value of sysctl_sched_nr_migrate which can be init to different values. Then, use SCHED_NR_MIGRATE_BREAK to init sysctl_sched_nr_migrate. The behavior stays unchanged unless you modify sysctl_sched_nr_migrate trough debugfs. Signed-off-by: Vincent Guittot Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20220825122726.20819-3-vincent.guittot@linaro.org --- kernel/sched/core.c | 6 +----- kernel/sched/fair.c | 11 ++++------- kernel/sched/sched.h | 6 ++++++ 3 files changed, 11 insertions(+), 12 deletions(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 2b85d1b5fe0c..4fa4a3ddb4f4 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -142,11 +142,7 @@ __read_mostly int sysctl_resched_latency_warn_once = 1; * Number of tasks to iterate in a single balance run. * Limited because this is done with IRQs disabled. */ -#ifdef CONFIG_PREEMPT_RT -const_debug unsigned int sysctl_sched_nr_migrate = 8; -#else -const_debug unsigned int sysctl_sched_nr_migrate = 32; -#endif +const_debug unsigned int sysctl_sched_nr_migrate = SCHED_NR_MIGRATE_BREAK; __read_mostly int scheduler_running; diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index dae3bfae1bc0..7b3a58fcb436 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -8009,8 +8009,6 @@ static struct task_struct *detach_one_task(struct lb_env *env) return NULL; } -static const unsigned int sched_nr_migrate_break = 32; - /* * detach_tasks() -- tries to detach up to imbalance load/util/tasks from * busiest_rq, as part of a balancing operation within domain "sd". @@ -8059,7 +8057,7 @@ static int detach_tasks(struct lb_env *env) /* take a breather every nr_migrate tasks */ if (env->loop > env->loop_break) { - env->loop_break += sched_nr_migrate_break; + env->loop_break += SCHED_NR_MIGRATE_BREAK; env->flags |= LBF_NEED_BREAK; break; } @@ -10100,14 +10098,13 @@ static int load_balance(int this_cpu, struct rq *this_rq, struct rq *busiest; struct rq_flags rf; struct cpumask *cpus = this_cpu_cpumask_var_ptr(load_balance_mask); - struct lb_env env = { .sd = sd, .dst_cpu = this_cpu, .dst_rq = this_rq, .dst_grpmask = sched_group_span(sd->groups), .idle = idle, - .loop_break = sched_nr_migrate_break, + .loop_break = SCHED_NR_MIGRATE_BREAK, .cpus = cpus, .fbq_type = all, .tasks = LIST_HEAD_INIT(env.tasks), @@ -10216,7 +10213,7 @@ more_balance: env.dst_cpu = env.new_dst_cpu; env.flags &= ~LBF_DST_PINNED; env.loop = 0; - env.loop_break = sched_nr_migrate_break; + env.loop_break = SCHED_NR_MIGRATE_BREAK; /* * Go back to "more_balance" rather than "redo" since we @@ -10248,7 +10245,7 @@ more_balance: */ if (!cpumask_subset(cpus, env.dst_grpmask)) { env.loop = 0; - env.loop_break = sched_nr_migrate_break; + env.loop_break = SCHED_NR_MIGRATE_BREAK; goto redo; } goto out_all_pinned; diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 91b2c7ec53bd..1fc198be1ffd 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -2423,6 +2423,12 @@ extern void deactivate_task(struct rq *rq, struct task_struct *p, int flags); extern void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags); +#ifdef CONFIG_PREEMPT_RT +#define SCHED_NR_MIGRATE_BREAK 8 +#else +#define SCHED_NR_MIGRATE_BREAK 32 +#endif + extern const_debug unsigned int sysctl_sched_nr_migrate; extern const_debug unsigned int sysctl_sched_migration_cost; From 7e9518baed4cef76dbfa07cbffbae1e6dbc87be6 Mon Sep 17 00:00:00 2001 From: Vincent Guittot Date: Thu, 25 Aug 2022 14:27:25 +0200 Subject: [PATCH 39/41] sched/fair: Move call to list_last_entry() in detach_tasks Move the call to list_last_entry() in detach_tasks() after testing loop_max and loop_break. Signed-off-by: Vincent Guittot Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20220825122726.20819-4-vincent.guittot@linaro.org --- kernel/sched/fair.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 7b3a58fcb436..5ffec4370602 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -8044,8 +8044,6 @@ static int detach_tasks(struct lb_env *env) if (env->idle != CPU_NOT_IDLE && env->src_rq->nr_running <= 1) break; - p = list_last_entry(tasks, struct task_struct, se.group_node); - env->loop++; /* * We've more or less seen every task there is, call it quits @@ -8062,6 +8060,8 @@ static int detach_tasks(struct lb_env *env) break; } + p = list_last_entry(tasks, struct task_struct, se.group_node); + if (!can_migrate_task(p, env)) goto next; From 5aec788aeb8eb74282b75ac1b317beb0fbb69a42 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 27 Sep 2022 21:02:34 +0200 Subject: [PATCH 40/41] sched: Fix TASK_state comparisons Task state is fundamentally a bitmask; direct comparisons are probably not working as intended. Specifically the normal wait-state have a number of possible modifiers: TASK_UNINTERRUPTIBLE: TASK_WAKEKILL, TASK_NOLOAD, TASK_FREEZABLE TASK_INTERRUPTIBLE: TASK_FREEZABLE Specifically, the addition of TASK_FREEZABLE wrecked __wait_is_interruptible(). This however led to an audit of direct comparisons yielding the rest of the changes. Fixes: f5d39b020809 ("freezer,sched: Rewrite core freezer logic") Reported-by: Christian Borntraeger Debugged-by: Christian Borntraeger Signed-off-by: Peter Zijlstra (Intel) Tested-by: Christian Borntraeger --- include/linux/wait.h | 2 +- kernel/hung_task.c | 8 ++++++-- kernel/sched/core.c | 2 +- 3 files changed, 8 insertions(+), 4 deletions(-) diff --git a/include/linux/wait.h b/include/linux/wait.h index 14ad8a0e9fac..7f5a51aae0a7 100644 --- a/include/linux/wait.h +++ b/include/linux/wait.h @@ -281,7 +281,7 @@ static inline void wake_up_pollfree(struct wait_queue_head *wq_head) #define ___wait_is_interruptible(state) \ (!__builtin_constant_p(state) || \ - state == TASK_INTERRUPTIBLE || state == TASK_KILLABLE) \ + (state & (TASK_INTERRUPTIBLE | TASK_WAKEKILL))) extern void init_wait_entry(struct wait_queue_entry *wq_entry, int flags); diff --git a/kernel/hung_task.c b/kernel/hung_task.c index f1321c03c32a..3a15169ba2f8 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -191,6 +191,8 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout) hung_task_show_lock = false; rcu_read_lock(); for_each_process_thread(g, t) { + unsigned int state; + if (!max_count--) goto unlock; if (time_after(jiffies, last_break + HUNG_TASK_LOCK_BREAK)) { @@ -198,8 +200,10 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout) goto unlock; last_break = jiffies; } - /* use "==" to skip the TASK_KILLABLE tasks waiting on NFS */ - if (READ_ONCE(t->__state) == TASK_UNINTERRUPTIBLE) + /* skip the TASK_KILLABLE tasks -- these can be killed */ + state = READ_ONCE(t->__state); + if ((state & TASK_UNINTERRUPTIBLE) && + !(state & TASK_WAKEKILL)) check_hung_task(t, timeout); } unlock: diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 4fa4a3ddb4f4..02dc1b8e3cb6 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -8884,7 +8884,7 @@ state_filter_match(unsigned long state_filter, struct task_struct *p) * When looking for TASK_UNINTERRUPTIBLE skip TASK_IDLE (allows * TASK_KILLABLE). */ - if (state_filter == TASK_UNINTERRUPTIBLE && state == TASK_IDLE) + if (state_filter == TASK_UNINTERRUPTIBLE && (state & TASK_NOLOAD)) return false; return true; From fdf756f7127185eeffe00e918e66dfee797f3625 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 30 Sep 2022 16:39:46 +0200 Subject: [PATCH 41/41] sched: Fix more TASK_state comparisons Boris reported hung_task splats after commit 5aec788aeb8e ("sched: Fix TASK_state comparisons"). Upon closer consideration of that change it doesn't only exclude TASK_KILLABLE, but also TASK_IDLE. Update the comment to reflect this fact and add an additional TASK_NOLOAD test to exclude them. Additionally, remove the TASK_FREEZABLE early exit from check_hung_task(), a freezable task is not a frozen task. Fixes: 5aec788aeb8e ("sched: Fix TASK_state comparisons") Reported-by: Borislav Petkov Signed-off-by: Peter Zijlstra (Intel) Tested-by: Borislav Petkov --- kernel/hung_task.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/kernel/hung_task.c b/kernel/hung_task.c index 3a15169ba2f8..c71889f3f3fc 100644 --- a/kernel/hung_task.c +++ b/kernel/hung_task.c @@ -95,7 +95,7 @@ static void check_hung_task(struct task_struct *t, unsigned long timeout) * Ensure the task is not frozen. * Also, skip vfork and any other user process that freezer should skip. */ - if (unlikely(READ_ONCE(t->__state) & (TASK_FREEZABLE | TASK_FROZEN))) + if (unlikely(READ_ONCE(t->__state) & TASK_FROZEN)) return; /* @@ -200,10 +200,14 @@ static void check_hung_uninterruptible_tasks(unsigned long timeout) goto unlock; last_break = jiffies; } - /* skip the TASK_KILLABLE tasks -- these can be killed */ + /* + * skip the TASK_KILLABLE tasks -- these can be killed + * skip the TASK_IDLE tasks -- those are genuinely idle + */ state = READ_ONCE(t->__state); if ((state & TASK_UNINTERRUPTIBLE) && - !(state & TASK_WAKEKILL)) + !(state & TASK_WAKEKILL) && + !(state & TASK_NOLOAD)) check_hung_task(t, timeout); } unlock: