diff options
Diffstat (limited to 'kernel/cgroup')
| -rw-r--r-- | kernel/cgroup/cgroup-v1.c | 16 | ||||
| -rw-r--r-- | kernel/cgroup/cgroup.c | 86 | ||||
| -rw-r--r-- | kernel/cgroup/cpuset.c | 208 | ||||
| -rw-r--r-- | kernel/cgroup/legacy_freezer.c | 7 | ||||
| -rw-r--r-- | kernel/cgroup/rstat.c | 8 |
5 files changed, 226 insertions, 99 deletions
diff --git a/kernel/cgroup/cgroup-v1.c b/kernel/cgroup/cgroup-v1.c index 52bb5a74a23b..aeef06c465ef 100644 --- a/kernel/cgroup/cgroup-v1.c +++ b/kernel/cgroup/cgroup-v1.c @@ -58,7 +58,7 @@ int cgroup_attach_task_all(struct task_struct *from, struct task_struct *tsk) struct cgroup_root *root; int retval = 0; - mutex_lock(&cgroup_mutex); + cgroup_lock(); cgroup_attach_lock(true); for_each_root(root) { struct cgroup *from_cgrp; @@ -72,7 +72,7 @@ int cgroup_attach_task_all(struct task_struct *from, struct task_struct *tsk) break; } cgroup_attach_unlock(true); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); return retval; } @@ -106,7 +106,7 @@ int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from) if (ret) return ret; - mutex_lock(&cgroup_mutex); + cgroup_lock(); percpu_down_write(&cgroup_threadgroup_rwsem); @@ -145,7 +145,7 @@ int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from) out_err: cgroup_migrate_finish(&mgctx); percpu_up_write(&cgroup_threadgroup_rwsem); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); return ret; } @@ -847,13 +847,13 @@ static int cgroup1_rename(struct kernfs_node *kn, struct kernfs_node *new_parent kernfs_break_active_protection(new_parent); kernfs_break_active_protection(kn); - mutex_lock(&cgroup_mutex); + cgroup_lock(); ret = kernfs_rename(kn, new_parent, new_name_str); if (!ret) TRACE_CGROUP_PATH(rename, cgrp); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); kernfs_unbreak_active_protection(kn); kernfs_unbreak_active_protection(new_parent); @@ -1119,7 +1119,7 @@ int cgroup1_reconfigure(struct fs_context *fc) trace_cgroup_remount(root); out_unlock: - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); return ret; } @@ -1246,7 +1246,7 @@ int cgroup1_get_tree(struct fs_context *fc) if (!ret && !percpu_ref_tryget_live(&ctx->root->cgrp.self.refcnt)) ret = 1; /* restart */ - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); if (!ret) ret = cgroup_do_get_tree(fc); diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 935e8121b21e..625d7483951c 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -1391,7 +1391,7 @@ static void cgroup_destroy_root(struct cgroup_root *root) cgroup_favor_dynmods(root, false); cgroup_exit_root_id(root); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); cgroup_rstat_exit(cgrp); kernfs_destroy_root(root->kf_root); @@ -1465,8 +1465,18 @@ static struct cgroup *current_cgns_cgroup_dfl(void) { struct css_set *cset; - cset = current->nsproxy->cgroup_ns->root_cset; - return __cset_cgroup_from_root(cset, &cgrp_dfl_root); + if (current->nsproxy) { + cset = current->nsproxy->cgroup_ns->root_cset; + return __cset_cgroup_from_root(cset, &cgrp_dfl_root); + } else { + /* + * NOTE: This function may be called from bpf_cgroup_from_id() + * on a task which has already passed exit_task_namespaces() and + * nsproxy == NULL. Fall back to cgrp_dfl_root which will make all + * cgroups visible for lookups. + */ + return &cgrp_dfl_root.cgrp; + } } /* look up cgroup associated with given css_set on the specified hierarchy */ @@ -1625,7 +1635,7 @@ void cgroup_kn_unlock(struct kernfs_node *kn) else cgrp = kn->parent->priv; - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); kernfs_unbreak_active_protection(kn); cgroup_put(cgrp); @@ -1670,7 +1680,7 @@ struct cgroup *cgroup_kn_lock_live(struct kernfs_node *kn, bool drain_offline) if (drain_offline) cgroup_lock_and_drain_offline(cgrp); else - mutex_lock(&cgroup_mutex); + cgroup_lock(); if (!cgroup_is_dead(cgrp)) return cgrp; @@ -2167,13 +2177,13 @@ int cgroup_do_get_tree(struct fs_context *fc) struct super_block *sb = fc->root->d_sb; struct cgroup *cgrp; - mutex_lock(&cgroup_mutex); + cgroup_lock(); spin_lock_irq(&css_set_lock); cgrp = cset_cgroup_from_root(ctx->ns->root_cset, ctx->root); spin_unlock_irq(&css_set_lock); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); nsdentry = kernfs_node_dentry(cgrp->kn, sb); dput(fc->root); @@ -2356,13 +2366,13 @@ int cgroup_path_ns(struct cgroup *cgrp, char *buf, size_t buflen, { int ret; - mutex_lock(&cgroup_mutex); + cgroup_lock(); spin_lock_irq(&css_set_lock); ret = cgroup_path_ns_locked(cgrp, buf, buflen, ns); spin_unlock_irq(&css_set_lock); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); return ret; } @@ -2388,7 +2398,7 @@ int task_cgroup_path(struct task_struct *task, char *buf, size_t buflen) int hierarchy_id = 1; int ret; - mutex_lock(&cgroup_mutex); + cgroup_lock(); spin_lock_irq(&css_set_lock); root = idr_get_next(&cgroup_hierarchy_idr, &hierarchy_id); @@ -2402,7 +2412,7 @@ int task_cgroup_path(struct task_struct *task, char *buf, size_t buflen) } spin_unlock_irq(&css_set_lock); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); return ret; } EXPORT_SYMBOL_GPL(task_cgroup_path); @@ -3111,7 +3121,7 @@ void cgroup_lock_and_drain_offline(struct cgroup *cgrp) int ssid; restart: - mutex_lock(&cgroup_mutex); + cgroup_lock(); cgroup_for_each_live_descendant_post(dsct, d_css, cgrp) { for_each_subsys(ss, ssid) { @@ -3125,7 +3135,7 @@ restart: prepare_to_wait(&dsct->offline_waitq, &wait, TASK_UNINTERRUPTIBLE); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); schedule(); finish_wait(&dsct->offline_waitq, &wait); @@ -3761,7 +3771,7 @@ static ssize_t pressure_write(struct kernfs_open_file *of, char *buf, } psi = cgroup_psi(cgrp); - new = psi_trigger_create(psi, buf, res); + new = psi_trigger_create(psi, buf, res, of->file); if (IS_ERR(new)) { cgroup_put(cgrp); return PTR_ERR(new); @@ -4374,9 +4384,9 @@ int cgroup_rm_cftypes(struct cftype *cfts) if (!(cfts[0].flags & __CFTYPE_ADDED)) return -ENOENT; - mutex_lock(&cgroup_mutex); + cgroup_lock(); ret = cgroup_rm_cftypes_locked(cfts); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); return ret; } @@ -4408,14 +4418,14 @@ static int cgroup_add_cftypes(struct cgroup_subsys *ss, struct cftype *cfts) if (ret) return ret; - mutex_lock(&cgroup_mutex); + cgroup_lock(); list_add_tail(&cfts->node, &ss->cfts); ret = cgroup_apply_cftypes(cfts, true); if (ret) cgroup_rm_cftypes_locked(cfts); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); return ret; } @@ -5385,7 +5395,7 @@ static void css_release_work_fn(struct work_struct *work) struct cgroup_subsys *ss = css->ss; struct cgroup *cgrp = css->cgroup; - mutex_lock(&cgroup_mutex); + cgroup_lock(); css->flags |= CSS_RELEASED; list_del_rcu(&css->sibling); @@ -5426,7 +5436,7 @@ static void css_release_work_fn(struct work_struct *work) NULL); } - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); INIT_RCU_WORK(&css->destroy_rwork, css_free_rwork_fn); queue_rcu_work(cgroup_destroy_wq, &css->destroy_rwork); @@ -5774,7 +5784,7 @@ static void css_killed_work_fn(struct work_struct *work) struct cgroup_subsys_state *css = container_of(work, struct cgroup_subsys_state, destroy_work); - mutex_lock(&cgroup_mutex); + cgroup_lock(); do { offline_css(css); @@ -5783,7 +5793,7 @@ static void css_killed_work_fn(struct work_struct *work) css = css->parent; } while (css && atomic_dec_and_test(&css->online_cnt)); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); } /* css kill confirmation processing requires process context, bounce */ @@ -5967,7 +5977,7 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss, bool early) pr_debug("Initializing cgroup subsys %s\n", ss->name); - mutex_lock(&cgroup_mutex); + cgroup_lock(); idr_init(&ss->css_idr); INIT_LIST_HEAD(&ss->cfts); @@ -6011,7 +6021,7 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss, bool early) BUG_ON(online_css(css)); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); } /** @@ -6071,7 +6081,7 @@ int __init cgroup_init(void) get_user_ns(init_cgroup_ns.user_ns); - mutex_lock(&cgroup_mutex); + cgroup_lock(); /* * Add init_css_set to the hash table so that dfl_root can link to @@ -6082,7 +6092,7 @@ int __init cgroup_init(void) BUG_ON(cgroup_setup_root(&cgrp_dfl_root, 0)); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); for_each_subsys(ss, ssid) { if (ss->early_init) { @@ -6134,9 +6144,9 @@ int __init cgroup_init(void) if (ss->bind) ss->bind(init_css_set.subsys[ssid]); - mutex_lock(&cgroup_mutex); + cgroup_lock(); css_populate_dir(init_css_set.subsys[ssid]); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); } /* init_css_set.subsys[] has been updated, re-hash */ @@ -6241,7 +6251,7 @@ int proc_cgroup_show(struct seq_file *m, struct pid_namespace *ns, if (!buf) goto out; - mutex_lock(&cgroup_mutex); + cgroup_lock(); spin_lock_irq(&css_set_lock); for_each_root(root) { @@ -6296,7 +6306,7 @@ int proc_cgroup_show(struct seq_file *m, struct pid_namespace *ns, retval = 0; out_unlock: spin_unlock_irq(&css_set_lock); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); kfree(buf); out: return retval; @@ -6380,7 +6390,7 @@ static int cgroup_css_set_fork(struct kernel_clone_args *kargs) struct file *f; if (kargs->flags & CLONE_INTO_CGROUP) - mutex_lock(&cgroup_mutex); + cgroup_lock(); cgroup_threadgroup_change_begin(current); @@ -6455,7 +6465,7 @@ static int cgroup_css_set_fork(struct kernel_clone_args *kargs) err: cgroup_threadgroup_change_end(current); - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); if (f) fput(f); if (dst_cgrp) @@ -6482,7 +6492,7 @@ static void cgroup_css_set_put_fork(struct kernel_clone_args *kargs) struct cgroup *cgrp = kargs->cgrp; struct css_set *cset = kargs->cset; - mutex_unlock(&cgroup_mutex); + cgroup_unlock(); if (cset) { put_css_set(cset); @@ -6856,14 +6866,12 @@ EXPORT_SYMBOL_GPL(cgroup_get_from_path); struct cgroup *cgroup_v1v2_get_from_fd(int fd) { struct cgroup *cgrp; - struct file *f; - - f = fget_raw(fd); - if (!f) + struct fd f = fdget_raw(fd); + if (!f.file) return ERR_PTR(-EBADF); - cgrp = cgroup_v1v2_get_from_file(f); - fput(f); + cgrp = cgroup_v1v2_get_from_file(f.file); + fdput(f); return cgrp; } diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index 636f1c682ac0..e4ca2dd2b764 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -1209,7 +1209,9 @@ void rebuild_sched_domains(void) * * Iterate through each task of @cs updating its cpus_allowed to the * effective cpuset's. As this function is called with cpuset_rwsem held, - * cpuset membership stays stable. + * cpuset membership stays stable. For top_cpuset, task_cpu_possible_mask() + * is used instead of effective_cpus to make sure all offline CPUs are also + * included as hotplug code won't update cpumasks for tasks in top_cpuset. */ static void update_tasks_cpumask(struct cpuset *cs, struct cpumask *new_cpus) { @@ -1219,15 +1221,18 @@ static void update_tasks_cpumask(struct cpuset *cs, struct cpumask *new_cpus) css_task_iter_start(&cs->css, 0, &it); while ((task = css_task_iter_next(&it))) { - /* - * Percpu kthreads in top_cpuset are ignored - */ - if (top_cs && (task->flags & PF_KTHREAD) && - kthread_is_per_cpu(task)) - continue; + const struct cpumask *possible_mask = task_cpu_possible_mask(task); - cpumask_and(new_cpus, cs->effective_cpus, - task_cpu_possible_mask(task)); + if (top_cs) { + /* + * Percpu kthreads in top_cpuset are ignored + */ + if ((task->flags & PF_KTHREAD) && kthread_is_per_cpu(task)) + continue; + cpumask_andnot(new_cpus, possible_mask, cs->subparts_cpus); + } else { + cpumask_and(new_cpus, possible_mask, cs->effective_cpus); + } set_cpus_allowed_ptr(task, new_cpus); } css_task_iter_end(&it); @@ -1513,7 +1518,7 @@ static int update_parent_subparts_cpumask(struct cpuset *cs, int cmd, spin_unlock_irq(&callback_lock); if (adding || deleting) - update_tasks_cpumask(parent, tmp->new_cpus); + update_tasks_cpumask(parent, tmp->addmask); /* * Set or clear CS_SCHED_LOAD_BALANCE when partcmd_update, if necessary. @@ -1770,10 +1775,13 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs, /* * Use the cpumasks in trialcs for tmpmasks when they are pointers * to allocated cpumasks. + * + * Note that update_parent_subparts_cpumask() uses only addmask & + * delmask, but not new_cpus. */ tmp.addmask = trialcs->subparts_cpus; tmp.delmask = trialcs->effective_cpus; - tmp.new_cpus = trialcs->cpus_allowed; + tmp.new_cpus = NULL; #endif retval = validate_change(cs, trialcs); @@ -1838,6 +1846,11 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs, } spin_unlock_irq(&callback_lock); +#ifdef CONFIG_CPUMASK_OFFSTACK + /* Now trialcs->cpus_allowed is available */ + tmp.new_cpus = trialcs->cpus_allowed; +#endif + /* effective_cpus will be updated here */ update_cpumasks_hier(cs, &tmp, false); @@ -2445,6 +2458,20 @@ static int fmeter_getrate(struct fmeter *fmp) static struct cpuset *cpuset_attach_old_cs; +/* + * Check to see if a cpuset can accept a new task + * For v1, cpus_allowed and mems_allowed can't be empty. + * For v2, effective_cpus can't be empty. + * Note that in v1, effective_cpus = cpus_allowed. + */ +static int cpuset_can_attach_check(struct cpuset *cs) +{ + if (cpumask_empty(cs->effective_cpus) || + (!is_in_v2_mode() && nodes_empty(cs->mems_allowed))) + return -ENOSPC; + return 0; +} + /* Called by cgroups to determine if a cpuset is usable; cpuset_rwsem held */ static int cpuset_can_attach(struct cgroup_taskset *tset) { @@ -2459,16 +2486,9 @@ static int cpuset_can_attach(struct cgroup_taskset *tset) percpu_down_write(&cpuset_rwsem); - /* allow moving tasks into an empty cpuset if on default hierarchy */ - ret = -ENOSPC; - if (!is_in_v2_mode() && - (cpumask_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed))) - goto out_unlock; - - /* - * Task cannot be moved to a cpuset with empty effective cpus. - */ - if (cpumask_empty(cs->effective_cpus)) + /* Check to see if task is allowed in the cpuset */ + ret = cpuset_can_attach_check(cs); + if (ret) goto out_unlock; cgroup_taskset_for_each(task, css, tset) { @@ -2485,7 +2505,6 @@ static int cpuset_can_attach(struct cgroup_taskset *tset) * changes which zero cpus/mems_allowed. */ cs->attach_in_progress++; - ret = 0; out_unlock: percpu_up_write(&cpuset_rwsem); return ret; @@ -2494,25 +2513,47 @@ out_unlock: static void cpuset_cancel_attach(struct cgroup_taskset *tset) { struct cgroup_subsys_state *css; + struct cpuset *cs; cgroup_taskset_first(tset, &css); + cs = css_cs(css); percpu_down_write(&cpuset_rwsem); - css_cs(css)->attach_in_progress--; + cs->attach_in_progress--; + if (!cs->attach_in_progress) + wake_up(&cpuset_attach_wq); percpu_up_write(&cpuset_rwsem); } /* - * Protected by cpuset_rwsem. cpus_attach is used only by cpuset_attach() + * Protected by cpuset_rwsem. cpus_attach is used only by cpuset_attach_task() * but we can't allocate it dynamically there. Define it global and * allocate from cpuset_init(). */ static cpumask_var_t cpus_attach; +static nodemask_t cpuset_attach_nodemask_to; + +static void cpuset_attach_task(struct cpuset *cs, struct task_struct *task) +{ + percpu_rwsem_assert_held(&cpuset_rwsem); + + if (cs != &top_cpuset) + guarantee_online_cpus(task, cpus_attach); + else + cpumask_andnot(cpus_attach, task_cpu_possible_mask(task), + cs->subparts_cpus); + /* + * can_attach beforehand should guarantee that this doesn't + * fail. TODO: have a better way to handle failure here + */ + WARN_ON_ONCE(set_cpus_allowed_ptr(task, cpus_attach)); + + cpuset_change_task_nodemask(task, &cpuset_attach_nodemask_to); + cpuset_update_task_spread_flags(cs, task); +} static void cpuset_attach(struct cgroup_taskset *tset) { - /* static buf protected by cpuset_rwsem */ - static nodemask_t cpuset_attach_nodemask_to; struct task_struct *task; struct task_struct *leader; struct cgroup_subsys_state *css; @@ -2543,20 +2584,8 @@ static void cpuset_attach(struct cgroup_taskset *tset) guarantee_online_mems(cs, &cpuset_attach_nodemask_to); - cgroup_taskset_for_each(task, css, tset) { - if (cs != &top_cpuset) - guarantee_online_cpus(task, cpus_attach); - else - cpumask_copy(cpus_attach, task_cpu_possible_mask(task)); - /* - * can_attach beforehand should guarantee that this doesn't - * fail. TODO: have a better way to handle failure here - */ - WARN_ON_ONCE(set_cpus_allowed_ptr(task, cpus_attach)); - - cpuset_change_task_nodemask(task, &cpuset_attach_nodemask_to); - cpuset_update_task_spread_flags(cs, task); - } + cgroup_taskset_for_each(task, css, tset) + cpuset_attach_task(cs, task); /* * Change mm for all threadgroup leaders. This is expensive and may @@ -3248,17 +3277,101 @@ static void cpuset_bind(struct cgroup_subsys_state *root_css) } /* + * In case the child is cloned into a cpuset different from its parent, + * additional checks are done to see if the move is allowed. + */ +static int cpuset_can_fork(struct task_struct *task, struct css_set *cset) +{ + struct cpuset *cs = css_cs(cset->subsys[cpuset_cgrp_id]); + bool same_cs; + int ret; + + rcu_read_lock(); + same_cs = (cs == task_cs(current)); + rcu_read_unlock(); + + if (same_cs) + return 0; + + lockdep_assert_held(&cgroup_mutex); + percpu_down_write(&cpuset_rwsem); + + /* Check to see if task is allowed in the cpuset */ + ret = cpuset_can_attach_check(cs); + if (ret) + goto out_unlock; + + ret = task_can_attach(task, cs->effective_cpus); + if (ret) + goto out_unlock; + + ret = security_task_setscheduler(task); + if (ret) + goto out_unlock; + + /* + * Mark attach is in progress. This makes validate_change() fail + * changes which zero cpus/mems_allowed. + */ + cs->attach_in_progress++; +out_unlock: + percpu_up_write(&cpuset_rwsem); + return ret; +} + +static void cpuset_cancel_fork(struct task_struct *task, struct css_set *cset) +{ + struct cpuset *cs = css_cs(cset->subsys[cpuset_cgrp_id]); + bool same_cs; + + rcu_read_lock(); + same_cs = (cs == task_cs(current)); + rcu_read_unlock(); + + if (same_cs) + return; + + percpu_down_write(&cpuset_rwsem); + cs->attach_in_progress--; + if (!cs->attach_in_progress) + wake_up(&cpuset_attach_wq); + percpu_up_write(&cpuset_rwsem); +} + +/* * Make sure the new task conform to the current state of its parent, * which could have been changed by cpuset just after it inherits the * state from the parent and before it sits on the cgroup's task list. */ static void cpuset_fork(struct task_struct *task) { - if (task_css_is_root(task, cpuset_cgrp_id)) + struct cpuset *cs; + bool same_cs; + + rcu_read_lock(); + cs = task_cs(task); + same_cs = (cs == task_cs(current)); + rcu_read_unlock(); + + if (same_cs) { + if (cs == &top_cpuset) + return; + + set_cpus_allowed_ptr(task, current->cpus_ptr); + task->mems_allowed = current->mems_allowed; return; + } - set_cpus_allowed_ptr(task, current->cpus_ptr); - task->mems_allowed = current->mems_allowed; + /* CLONE_INTO_CGROUP */ + percpu_down_write(&cpuset_rwsem); + guarantee_online_mems(cs, &cpuset_attach_nodemask_to); + cpuset_attach_task(cs, task); + + cs->attach_in_progress--; + if (!cs->attach_in_progress) + wake_up(&cpuset_attach_wq); + + percpu_up_write(&cpuset_rwsem); } struct cgroup_subsys cpuset_cgrp_subsys = { @@ -3271,6 +3384,8 @@ struct cgroup_subsys cpuset_cgrp_subsys = { .attach = cpuset_attach, .post_attach = cpuset_post_attach, .bind = cpuset_bind, + .can_fork = cpuset_can_fork, + .cancel_fork = cpuset_cancel_fork, .fork = cpuset_fork, .legacy_cftypes = legacy_files, .dfl_cftypes = dfl_files, @@ -3508,6 +3623,8 @@ retry: update_tasks: cpus_updated = !cpumask_equal(&new_cpus, cs->effective_cpus); mems_updated = !nodes_equal(new_mems, cs->effective_mems); + if (!cpus_updated && !mems_updated) + goto unlock; /* Hotplug doesn't affect this cpuset */ if (mems_updated) check_insane_mems_config(&new_mems); @@ -3519,6 +3636,7 @@ update_tasks: hotplug_update_tasks_legacy(cs, &new_cpus, &new_mems, cpus_updated, mems_updated); +unlock: percpu_up_write(&cpuset_rwsem); } @@ -3831,7 +3949,7 @@ static struct cpuset *nearest_hardwall_ancestor(struct cpuset *cs) } /* - * __cpuset_node_allowed - Can we allocate on a memory node? + * cpuset_node_allowed - Can we allocate on a memory node? * @node: is this an allowed node? * @gfp_mask: memory allocation flags * @@ -3870,7 +3988,7 @@ static struct cpuset *nearest_hardwall_ancestor(struct cpuset *cs) * GFP_KERNEL - any node in enclosing hardwalled cpuset ok * GFP_USER - only nodes in current tasks mems allowed ok. */ -bool __cpuset_node_allowed(int node, gfp_t gfp_mask) +bool cpuset_node_allowed(int node, gfp_t gfp_mask) { struct cpuset *cs; /* current cpuset ancestors */ bool allowed; /* is allocation in zone z allowed? */ diff --git a/kernel/cgroup/legacy_freezer.c b/kernel/cgroup/legacy_freezer.c index 1b6b21851e9d..936473203a6b 100644 --- a/kernel/cgroup/legacy_freezer.c +++ b/kernel/cgroup/legacy_freezer.c @@ -22,6 +22,7 @@ #include <linux/freezer.h> #include <linux/seq_file.h> #include <linux/mutex.h> +#include <linux/cpu.h> /* * A cgroup is freezing if any FREEZING flags are set. FREEZING_SELF is @@ -350,7 +351,7 @@ static void freezer_apply_state(struct freezer *freezer, bool freeze, if (freeze) { if (!(freezer->state & CGROUP_FREEZING)) - static_branch_inc(&freezer_active); + static_branch_inc_cpuslocked(&freezer_active); freezer->state |= state; freeze_cgroup(freezer); } else { @@ -361,7 +362,7 @@ static void freezer_apply_state(struct freezer *freezer, bool freeze, if (!(freezer->state & CGROUP_FREEZING)) { freezer->state &= ~CGROUP_FROZEN; if (was_freezing) - static_branch_dec(&freezer_active); + static_branch_dec_cpuslocked(&freezer_active); unfreeze_cgroup(freezer); } } @@ -379,6 +380,7 @@ static void freezer_change_state(struct freezer *freezer, bool freeze) { struct cgroup_subsys_state *pos; + cpus_read_lock(); /* * Update all its descendants in pre-order traversal. Each * descendant will try to inherit its parent's FREEZING state as @@ -407,6 +409,7 @@ static void freezer_change_state(struct freezer *freezer, bool freeze) } rcu_read_unlock(); mutex_unlock(&freezer_mutex); + cpus_read_unlock(); } static ssize_t freezer_write(struct kernfs_open_file *of, diff --git a/kernel/cgroup/rstat.c b/kernel/cgroup/rstat.c index 831f1f472bb8..9c4c55228567 100644 --- a/kernel/cgroup/rstat.c +++ b/kernel/cgroup/rstat.c @@ -241,12 +241,12 @@ __bpf_kfunc void cgroup_rstat_flush(struct cgroup *cgrp) } /** - * cgroup_rstat_flush_irqsafe - irqsafe version of cgroup_rstat_flush() + * cgroup_rstat_flush_atomic- atomic version of cgroup_rstat_flush() * @cgrp: target cgroup * * This function can be called from any context. */ -void cgroup_rstat_flush_irqsafe(struct cgroup *cgrp) +void cgroup_rstat_flush_atomic(struct cgroup *cgrp) { unsigned long flags; @@ -457,9 +457,7 @@ static void root_cgroup_cputime(struct cgroup_base_stat *bstat) struct task_cputime *cputime = &bstat->cputime; int i; - cputime->stime = 0; - cputime->utime = 0; - cputime->sum_exec_runtime = 0; + memset(bstat, 0, sizeof(*bstat)); for_each_possible_cpu(i) { struct kernel_cpustat kcpustat; u64 *cpustat = kcpustat.cpustat; |