From d6e486ee0ef2f99a4069d9186e53dac61b28cb3c Mon Sep 17 00:00:00 2001 From: Shakeel Butt Date: Wed, 3 Apr 2019 16:03:54 -0700 Subject: [PATCH 01/15] cgroup: remove extra cgroup_migrate_finish() call The callers of cgroup_migrate_prepare_dst() correctly call cgroup_migrate_finish() for success and failure cases both. No need to call it in cgroup_migrate_prepare_dst() in failure case. Signed-off-by: Shakeel Butt Reviewed-by: Daniel Jordan Signed-off-by: Tejun Heo --- kernel/cgroup/cgroup.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 3f2b4bde0f9c..f219c195a9a5 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -2602,7 +2602,7 @@ int cgroup_migrate_prepare_dst(struct cgroup_mgctx *mgctx) dst_cset = find_css_set(src_cset, src_cset->mg_dst_cgrp); if (!dst_cset) - goto err; + return -ENOMEM; WARN_ON_ONCE(src_cset->mg_dst_cset || dst_cset->mg_dst_cset); @@ -2634,9 +2634,6 @@ int cgroup_migrate_prepare_dst(struct cgroup_mgctx *mgctx) } return 0; -err: - cgroup_migrate_finish(mgctx); - return -ENOMEM; } /** From 50943f3e136adfc421f9768d6ae09ba7b83aaefd Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Fri, 19 Apr 2019 10:03:01 -0700 Subject: [PATCH 02/15] cgroup: rename freezer.c into legacy_freezer.c Freezer.c will contain an implementation of cgroup v2 freezer, so let's rename the v1 freezer to avoid naming conflicts. Signed-off-by: Roman Gushchin Signed-off-by: Tejun Heo Cc: kernel-team@fb.com --- kernel/cgroup/Makefile | 2 +- kernel/cgroup/{freezer.c => legacy_freezer.c} | 0 2 files changed, 1 insertion(+), 1 deletion(-) rename kernel/cgroup/{freezer.c => legacy_freezer.c} (100%) diff --git a/kernel/cgroup/Makefile b/kernel/cgroup/Makefile index bfcdae896122..8d5689ca94b9 100644 --- a/kernel/cgroup/Makefile +++ b/kernel/cgroup/Makefile @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0 obj-y := cgroup.o rstat.o namespace.o cgroup-v1.o -obj-$(CONFIG_CGROUP_FREEZER) += freezer.o +obj-$(CONFIG_CGROUP_FREEZER) += legacy_freezer.o obj-$(CONFIG_CGROUP_PIDS) += pids.o obj-$(CONFIG_CGROUP_RDMA) += rdma.o obj-$(CONFIG_CPUSETS) += cpuset.o diff --git a/kernel/cgroup/freezer.c b/kernel/cgroup/legacy_freezer.c similarity index 100% rename from kernel/cgroup/freezer.c rename to kernel/cgroup/legacy_freezer.c From aade7f9efba098859681f8e88d81a5b44ad09b12 Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Fri, 19 Apr 2019 10:03:02 -0700 Subject: [PATCH 03/15] cgroup: implement __cgroup_task_count() helper The helper is identical to the existing cgroup_task_count() except it doesn't take the css_set_lock by itself, assuming that the caller does. Also, move cgroup_task_count() implementation into kernel/cgroup/cgroup.c, as there is nothing specific to cgroup v1. Signed-off-by: Roman Gushchin Signed-off-by: Tejun Heo Cc: kernel-team@fb.com --- kernel/cgroup/cgroup-internal.h | 1 + kernel/cgroup/cgroup-v1.c | 16 ---------------- kernel/cgroup/cgroup.c | 33 +++++++++++++++++++++++++++++++++ 3 files changed, 34 insertions(+), 16 deletions(-) diff --git a/kernel/cgroup/cgroup-internal.h b/kernel/cgroup/cgroup-internal.h index 30e39f3932ad..02c001ffe2e2 100644 --- a/kernel/cgroup/cgroup-internal.h +++ b/kernel/cgroup/cgroup-internal.h @@ -240,6 +240,7 @@ int cgroup_rmdir(struct kernfs_node *kn); int cgroup_show_path(struct seq_file *sf, struct kernfs_node *kf_node, struct kernfs_root *kf_root); +int __cgroup_task_count(const struct cgroup *cgrp); int cgroup_task_count(const struct cgroup *cgrp); /* diff --git a/kernel/cgroup/cgroup-v1.c b/kernel/cgroup/cgroup-v1.c index c126b34fd4ff..68ca5de7ec27 100644 --- a/kernel/cgroup/cgroup-v1.c +++ b/kernel/cgroup/cgroup-v1.c @@ -342,22 +342,6 @@ static struct cgroup_pidlist *cgroup_pidlist_find_create(struct cgroup *cgrp, return l; } -/** - * cgroup_task_count - count the number of tasks in a cgroup. - * @cgrp: the cgroup in question - */ -int cgroup_task_count(const struct cgroup *cgrp) -{ - int count = 0; - struct cgrp_cset_link *link; - - spin_lock_irq(&css_set_lock); - list_for_each_entry(link, &cgrp->cset_links, cset_link) - count += link->cset->nr_tasks; - spin_unlock_irq(&css_set_lock); - return count; -} - /* * Load a cgroup's pidarray with either procs' tgids or tasks' pids */ diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index f219c195a9a5..3008ea684aa0 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -593,6 +593,39 @@ static void cgroup_get_live(struct cgroup *cgrp) css_get(&cgrp->self); } +/** + * __cgroup_task_count - count the number of tasks in a cgroup. The caller + * is responsible for taking the css_set_lock. + * @cgrp: the cgroup in question + */ +int __cgroup_task_count(const struct cgroup *cgrp) +{ + int count = 0; + struct cgrp_cset_link *link; + + lockdep_assert_held(&css_set_lock); + + list_for_each_entry(link, &cgrp->cset_links, cset_link) + count += link->cset->nr_tasks; + + return count; +} + +/** + * cgroup_task_count - count the number of tasks in a cgroup. + * @cgrp: the cgroup in question + */ +int cgroup_task_count(const struct cgroup *cgrp) +{ + int count; + + spin_lock_irq(&css_set_lock); + count = __cgroup_task_count(cgrp); + spin_unlock_irq(&css_set_lock); + + return count; +} + struct cgroup_subsys_state *of_css(struct kernfs_open_file *of) { struct cgroup *cgrp = of->kn->parent->priv; From 4dcabece4c3a9f9522127be12cc12cc120399b2f Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Fri, 19 Apr 2019 10:03:03 -0700 Subject: [PATCH 04/15] cgroup: protect cgroup->nr_(dying_)descendants by css_set_lock The number of descendant cgroups and the number of dying descendant cgroups are currently synchronized using the cgroup_mutex. The number of descendant cgroups will be required by the cgroup v2 freezer, which will use it to determine if a cgroup is frozen (depending on total number of descendants and number of frozen descendants). It's not always acceptable to grab the cgroup_mutex, especially from quite hot paths (e.g. exit()). To avoid this, let's additionally synchronize these counters using the css_set_lock. So, it's safe to read these counters with either cgroup_mutex or css_set_lock locked, and for changing both locks should be acquired. Signed-off-by: Roman Gushchin Signed-off-by: Tejun Heo Cc: kernel-team@fb.com --- include/linux/cgroup-defs.h | 5 +++++ kernel/cgroup/cgroup.c | 6 ++++++ 2 files changed, 11 insertions(+) diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h index 1c70803e9f77..7d57890cec67 100644 --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -349,6 +349,11 @@ struct cgroup { * Dying cgroups are cgroups which were deleted by a user, * but are still existing because someone else is holding a reference. * max_descendants is a maximum allowed number of descent cgroups. + * + * nr_descendants and nr_dying_descendants are protected + * by cgroup_mutex and css_set_lock. It's fine to read them holding + * any of cgroup_mutex and css_set_lock; for writing both locks + * should be held. */ int nr_descendants; int nr_dying_descendants; diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 3008ea684aa0..786ceef2f222 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -4811,9 +4811,11 @@ static void css_release_work_fn(struct work_struct *work) if (cgroup_on_dfl(cgrp)) cgroup_rstat_flush(cgrp); + spin_lock_irq(&css_set_lock); for (tcgrp = cgroup_parent(cgrp); tcgrp; tcgrp = cgroup_parent(tcgrp)) tcgrp->nr_dying_descendants--; + spin_unlock_irq(&css_set_lock); cgroup_idr_remove(&cgrp->root->cgroup_idr, cgrp->id); cgrp->id = -1; @@ -5031,12 +5033,14 @@ static struct cgroup *cgroup_create(struct cgroup *parent) if (ret) goto out_psi_free; + spin_lock_irq(&css_set_lock); for (tcgrp = cgrp; tcgrp; tcgrp = cgroup_parent(tcgrp)) { cgrp->ancestor_ids[tcgrp->level] = tcgrp->id; if (tcgrp != cgrp) tcgrp->nr_descendants++; } + spin_unlock_irq(&css_set_lock); if (notify_on_release(parent)) set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); @@ -5321,10 +5325,12 @@ static int cgroup_destroy_locked(struct cgroup *cgrp) if (parent && cgroup_is_threaded(cgrp)) parent->nr_threaded_children--; + spin_lock_irq(&css_set_lock); for (tcgrp = cgroup_parent(cgrp); tcgrp; tcgrp = cgroup_parent(tcgrp)) { tcgrp->nr_descendants--; tcgrp->nr_dying_descendants++; } + spin_unlock_irq(&css_set_lock); cgroup1_check_for_release(parent); From 76f969e8948d82e78e1bc4beb6b9465908e74873 Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Fri, 19 Apr 2019 10:03:04 -0700 Subject: [PATCH 05/15] cgroup: cgroup v2 freezer Cgroup v1 implements the freezer controller, which provides an ability to stop the workload in a cgroup and temporarily free up some resources (cpu, io, network bandwidth and, potentially, memory) for some other tasks. Cgroup v2 lacks this functionality. This patch implements freezer for cgroup v2. Cgroup v2 freezer tries to put tasks into a state similar to jobctl stop. This means that tasks can be killed, ptraced (using PTRACE_SEIZE*), and interrupted. It is possible to attach to a frozen task, get some information (e.g. read registers) and detach. It's also possible to migrate a frozen tasks to another cgroup. This differs cgroup v2 freezer from cgroup v1 freezer, which mostly tried to imitate the system-wide freezer. However uninterruptible sleep is fine when all tasks are going to be frozen (hibernation case), it's not the acceptable state for some subset of the system. Cgroup v2 freezer is not supporting freezing kthreads. If a non-root cgroup contains kthread, the cgroup still can be frozen, but the kthread will remain running, the cgroup will be shown as non-frozen, and the notification will not be delivered. * PTRACE_ATTACH is not working because non-fatal signal delivery is blocked in frozen state. There are some interface differences between cgroup v1 and cgroup v2 freezer too, which are required to conform the cgroup v2 interface design principles: 1) There is no separate controller, which has to be turned on: the functionality is always available and is represented by cgroup.freeze and cgroup.events cgroup control files. 2) The desired state is defined by the cgroup.freeze control file. Any hierarchical configuration is allowed. 3) The interface is asynchronous. The actual state is available using cgroup.events control file ("frozen" field). There are no dedicated transitional states. 4) It's allowed to make any changes with the cgroup hierarchy (create new cgroups, remove old cgroups, move tasks between cgroups) no matter if some cgroups are frozen. Signed-off-by: Roman Gushchin Signed-off-by: Tejun Heo No-objection-from-me-by: Oleg Nesterov Cc: kernel-team@fb.com --- include/linux/cgroup-defs.h | 28 ++++ include/linux/cgroup.h | 43 +++++ include/linux/sched.h | 2 + include/linux/sched/jobctl.h | 2 + kernel/cgroup/Makefile | 2 +- kernel/cgroup/cgroup.c | 110 +++++++++++- kernel/cgroup/freezer.c | 317 +++++++++++++++++++++++++++++++++++ kernel/fork.c | 2 + kernel/signal.c | 70 +++++++- 9 files changed, 566 insertions(+), 10 deletions(-) create mode 100644 kernel/cgroup/freezer.c diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h index 7d57890cec67..77258d276f93 100644 --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -65,6 +65,12 @@ enum { * specified at mount time and thus is implemented here. */ CGRP_CPUSET_CLONE_CHILDREN, + + /* Control group has to be frozen. */ + CGRP_FREEZE, + + /* Cgroup is frozen. */ + CGRP_FROZEN, }; /* cgroup_root->flags */ @@ -317,6 +323,25 @@ struct cgroup_rstat_cpu { struct cgroup *updated_next; /* NULL iff not on the list */ }; +struct cgroup_freezer_state { + /* Should the cgroup and its descendants be frozen. */ + bool freeze; + + /* Should the cgroup actually be frozen? */ + int e_freeze; + + /* Fields below are protected by css_set_lock */ + + /* Number of frozen descendant cgroups */ + int nr_frozen_descendants; + + /* + * Number of tasks, which are counted as frozen: + * frozen, SIGSTOPped, and PTRACEd. + */ + int nr_frozen_tasks; +}; + struct cgroup { /* self css with NULL ->ss, points back to this cgroup */ struct cgroup_subsys_state self; @@ -453,6 +478,9 @@ struct cgroup { /* If there is block congestion on this cgroup. */ atomic_t congestion_count; + /* Used to store internal freezer state */ + struct cgroup_freezer_state freezer; + /* ids of the ancestors at each level including self */ int ancestor_ids[]; }; diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 81f58b4a5418..3e2efd412dfa 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -881,4 +881,47 @@ static inline void put_cgroup_ns(struct cgroup_namespace *ns) free_cgroup_ns(ns); } +#ifdef CONFIG_CGROUPS + +void cgroup_enter_frozen(void); +void cgroup_leave_frozen(bool always_leave); +void cgroup_update_frozen(struct cgroup *cgrp); +void cgroup_freeze(struct cgroup *cgrp, bool freeze); +void cgroup_freezer_migrate_task(struct task_struct *task, struct cgroup *src, + struct cgroup *dst); +void cgroup_freezer_frozen_exit(struct task_struct *task); +static inline bool cgroup_task_freeze(struct task_struct *task) +{ + bool ret; + + if (task->flags & PF_KTHREAD) + return false; + + rcu_read_lock(); + ret = test_bit(CGRP_FREEZE, &task_dfl_cgroup(task)->flags); + rcu_read_unlock(); + + return ret; +} + +static inline bool cgroup_task_frozen(struct task_struct *task) +{ + return task->frozen; +} + +#else /* !CONFIG_CGROUPS */ + +static inline void cgroup_enter_frozen(void) { } +static inline void cgroup_leave_frozen(bool always_leave) { } +static inline bool cgroup_task_freeze(struct task_struct *task) +{ + return false; +} +static inline bool cgroup_task_frozen(struct task_struct *task) +{ + return false; +} + +#endif /* !CONFIG_CGROUPS */ + #endif /* _LINUX_CGROUP_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 1549584a1538..45b2199bd683 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -726,6 +726,8 @@ struct task_struct { #ifdef CONFIG_CGROUPS /* disallow userland-initiated cgroup migration */ unsigned no_cgroup_migration:1; + /* task is frozen/stopped (used by the cgroup freezer) */ + unsigned frozen:1; #endif #ifdef CONFIG_BLK_CGROUP /* to be used once the psi infrastructure lands upstream. */ diff --git a/include/linux/sched/jobctl.h b/include/linux/sched/jobctl.h index 98228bd48aee..fa067de9f1a9 100644 --- a/include/linux/sched/jobctl.h +++ b/include/linux/sched/jobctl.h @@ -18,6 +18,7 @@ struct task_struct; #define JOBCTL_TRAP_NOTIFY_BIT 20 /* trap for NOTIFY */ #define JOBCTL_TRAPPING_BIT 21 /* switching to TRACED */ #define JOBCTL_LISTENING_BIT 22 /* ptracer is listening for events */ +#define JOBCTL_TRAP_FREEZE_BIT 23 /* trap for cgroup freezer */ #define JOBCTL_STOP_DEQUEUED (1UL << JOBCTL_STOP_DEQUEUED_BIT) #define JOBCTL_STOP_PENDING (1UL << JOBCTL_STOP_PENDING_BIT) @@ -26,6 +27,7 @@ struct task_struct; #define JOBCTL_TRAP_NOTIFY (1UL << JOBCTL_TRAP_NOTIFY_BIT) #define JOBCTL_TRAPPING (1UL << JOBCTL_TRAPPING_BIT) #define JOBCTL_LISTENING (1UL << JOBCTL_LISTENING_BIT) +#define JOBCTL_TRAP_FREEZE (1UL << JOBCTL_TRAP_FREEZE_BIT) #define JOBCTL_TRAP_MASK (JOBCTL_TRAP_STOP | JOBCTL_TRAP_NOTIFY) #define JOBCTL_PENDING_MASK (JOBCTL_STOP_PENDING | JOBCTL_TRAP_MASK) diff --git a/kernel/cgroup/Makefile b/kernel/cgroup/Makefile index 8d5689ca94b9..5d7a76bfbbb7 100644 --- a/kernel/cgroup/Makefile +++ b/kernel/cgroup/Makefile @@ -1,5 +1,5 @@ # SPDX-License-Identifier: GPL-2.0 -obj-y := cgroup.o rstat.o namespace.o cgroup-v1.o +obj-y := cgroup.o rstat.o namespace.o cgroup-v1.o freezer.o obj-$(CONFIG_CGROUP_FREEZER) += legacy_freezer.o obj-$(CONFIG_CGROUP_PIDS) += pids.o diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 786ceef2f222..6895464b54c6 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -2435,8 +2435,15 @@ static int cgroup_migrate_execute(struct cgroup_mgctx *mgctx) get_css_set(to_cset); to_cset->nr_tasks++; css_set_move_task(task, from_cset, to_cset, true); - put_css_set_locked(from_cset); from_cset->nr_tasks--; + /* + * If the source or destination cgroup is frozen, + * the task might require to change its state. + */ + cgroup_freezer_migrate_task(task, from_cset->dfl_cgrp, + to_cset->dfl_cgrp); + put_css_set_locked(from_cset); + } } spin_unlock_irq(&css_set_lock); @@ -3477,8 +3484,11 @@ static ssize_t cgroup_max_depth_write(struct kernfs_open_file *of, static int cgroup_events_show(struct seq_file *seq, void *v) { - seq_printf(seq, "populated %d\n", - cgroup_is_populated(seq_css(seq)->cgroup)); + struct cgroup *cgrp = seq_css(seq)->cgroup; + + seq_printf(seq, "populated %d\n", cgroup_is_populated(cgrp)); + seq_printf(seq, "frozen %d\n", test_bit(CGRP_FROZEN, &cgrp->flags)); + return 0; } @@ -3540,6 +3550,40 @@ static int cgroup_cpu_pressure_show(struct seq_file *seq, void *v) } #endif +static int cgroup_freeze_show(struct seq_file *seq, void *v) +{ + struct cgroup *cgrp = seq_css(seq)->cgroup; + + seq_printf(seq, "%d\n", cgrp->freezer.freeze); + + return 0; +} + +static ssize_t cgroup_freeze_write(struct kernfs_open_file *of, + char *buf, size_t nbytes, loff_t off) +{ + struct cgroup *cgrp; + ssize_t ret; + int freeze; + + ret = kstrtoint(strstrip(buf), 0, &freeze); + if (ret) + return ret; + + if (freeze < 0 || freeze > 1) + return -ERANGE; + + cgrp = cgroup_kn_lock_live(of->kn, false); + if (!cgrp) + return -ENOENT; + + cgroup_freeze(cgrp, freeze); + + cgroup_kn_unlock(of->kn); + + return nbytes; +} + static int cgroup_file_open(struct kernfs_open_file *of) { struct cftype *cft = of->kn->priv; @@ -4683,6 +4727,12 @@ static struct cftype cgroup_base_files[] = { .name = "cgroup.stat", .seq_show = cgroup_stat_show, }, + { + .name = "cgroup.freeze", + .flags = CFTYPE_NOT_ON_ROOT, + .seq_show = cgroup_freeze_show, + .write = cgroup_freeze_write, + }, { .name = "cpu.stat", .flags = CFTYPE_NOT_ON_ROOT, @@ -5033,12 +5083,29 @@ static struct cgroup *cgroup_create(struct cgroup *parent) if (ret) goto out_psi_free; + /* + * New cgroup inherits effective freeze counter, and + * if the parent has to be frozen, the child has too. + */ + cgrp->freezer.e_freeze = parent->freezer.e_freeze; + if (cgrp->freezer.e_freeze) + set_bit(CGRP_FROZEN, &cgrp->flags); + spin_lock_irq(&css_set_lock); for (tcgrp = cgrp; tcgrp; tcgrp = cgroup_parent(tcgrp)) { cgrp->ancestor_ids[tcgrp->level] = tcgrp->id; - if (tcgrp != cgrp) + if (tcgrp != cgrp) { tcgrp->nr_descendants++; + + /* + * If the new cgroup is frozen, all ancestor cgroups + * get a new frozen descendant, but their state can't + * change because of this. + */ + if (cgrp->freezer.e_freeze) + tcgrp->freezer.nr_frozen_descendants++; + } } spin_unlock_irq(&css_set_lock); @@ -5329,6 +5396,12 @@ static int cgroup_destroy_locked(struct cgroup *cgrp) for (tcgrp = cgroup_parent(cgrp); tcgrp; tcgrp = cgroup_parent(tcgrp)) { tcgrp->nr_descendants--; tcgrp->nr_dying_descendants++; + /* + * If the dying cgroup is frozen, decrease frozen descendants + * counters of ancestor cgroups. + */ + if (test_bit(CGRP_FROZEN, &cgrp->flags)) + tcgrp->freezer.nr_frozen_descendants--; } spin_unlock_irq(&css_set_lock); @@ -5782,6 +5855,29 @@ void cgroup_post_fork(struct task_struct *child) cset->nr_tasks++; css_set_move_task(child, NULL, cset, false); } + + /* + * If the cgroup has to be frozen, the new task has too. + * Let's set the JOBCTL_TRAP_FREEZE jobctl bit to get + * the task into the frozen state. + */ + if (unlikely(cgroup_task_freeze(child))) { + struct cgroup *cgrp; + + spin_lock(&child->sighand->siglock); + WARN_ON_ONCE(child->frozen); + cgrp = cset->dfl_cgrp; + child->jobctl |= JOBCTL_TRAP_FREEZE; + spin_unlock(&child->sighand->siglock); + + /* + * Calling cgroup_update_frozen() isn't required here, + * because it will be called anyway a bit later + * from do_freezer_trap(). So we avoid cgroup's + * transient switch from the frozen state and back. + */ + } + spin_unlock_irq(&css_set_lock); } @@ -5830,6 +5926,12 @@ void cgroup_exit(struct task_struct *tsk) spin_lock_irq(&css_set_lock); css_set_move_task(tsk, cset, NULL, false); cset->nr_tasks--; + + if (unlikely(cgroup_task_frozen(tsk))) + cgroup_freezer_frozen_exit(tsk); + else if (unlikely(cgroup_task_freeze(tsk))) + cgroup_update_frozen(task_dfl_cgroup(tsk)); + spin_unlock_irq(&css_set_lock); } else { get_css_set(cset); diff --git a/kernel/cgroup/freezer.c b/kernel/cgroup/freezer.c new file mode 100644 index 000000000000..9d8cda478fc9 --- /dev/null +++ b/kernel/cgroup/freezer.c @@ -0,0 +1,317 @@ +//SPDX-License-Identifier: GPL-2.0 +#include +#include +#include +#include + +#include "cgroup-internal.h" + +/* + * Propagate the cgroup frozen state upwards by the cgroup tree. + */ +static void cgroup_propagate_frozen(struct cgroup *cgrp, bool frozen) +{ + int desc = 1; + + /* + * If the new state is frozen, some freezing ancestor cgroups may change + * their state too, depending on if all their descendants are frozen. + * + * Otherwise, all ancestor cgroups are forced into the non-frozen state. + */ + while ((cgrp = cgroup_parent(cgrp))) { + if (frozen) { + cgrp->freezer.nr_frozen_descendants += desc; + if (!test_bit(CGRP_FROZEN, &cgrp->flags) && + test_bit(CGRP_FREEZE, &cgrp->flags) && + cgrp->freezer.nr_frozen_descendants == + cgrp->nr_descendants) { + set_bit(CGRP_FROZEN, &cgrp->flags); + cgroup_file_notify(&cgrp->events_file); + desc++; + } + } else { + cgrp->freezer.nr_frozen_descendants -= desc; + if (test_bit(CGRP_FROZEN, &cgrp->flags)) { + clear_bit(CGRP_FROZEN, &cgrp->flags); + cgroup_file_notify(&cgrp->events_file); + desc++; + } + } + } +} + +/* + * Revisit the cgroup frozen state. + * Checks if the cgroup is really frozen and perform all state transitions. + */ +void cgroup_update_frozen(struct cgroup *cgrp) +{ + bool frozen; + + lockdep_assert_held(&css_set_lock); + + /* + * If the cgroup has to be frozen (CGRP_FREEZE bit set), + * and all tasks are frozen and/or stopped, let's consider + * the cgroup frozen. Otherwise it's not frozen. + */ + frozen = test_bit(CGRP_FREEZE, &cgrp->flags) && + cgrp->freezer.nr_frozen_tasks == __cgroup_task_count(cgrp); + + if (frozen) { + /* Already there? */ + if (test_bit(CGRP_FROZEN, &cgrp->flags)) + return; + + set_bit(CGRP_FROZEN, &cgrp->flags); + } else { + /* Already there? */ + if (!test_bit(CGRP_FROZEN, &cgrp->flags)) + return; + + clear_bit(CGRP_FROZEN, &cgrp->flags); + } + cgroup_file_notify(&cgrp->events_file); + + /* Update the state of ancestor cgroups. */ + cgroup_propagate_frozen(cgrp, frozen); +} + +/* + * Increment cgroup's nr_frozen_tasks. + */ +static void cgroup_inc_frozen_cnt(struct cgroup *cgrp) +{ + cgrp->freezer.nr_frozen_tasks++; +} + +/* + * Decrement cgroup's nr_frozen_tasks. + */ +static void cgroup_dec_frozen_cnt(struct cgroup *cgrp) +{ + cgrp->freezer.nr_frozen_tasks--; + WARN_ON_ONCE(cgrp->freezer.nr_frozen_tasks < 0); +} + +/* + * Enter frozen/stopped state, if not yet there. Update cgroup's counters, + * and revisit the state of the cgroup, if necessary. + */ +void cgroup_enter_frozen(void) +{ + struct cgroup *cgrp; + + if (current->frozen) + return; + + spin_lock_irq(&css_set_lock); + current->frozen = true; + cgrp = task_dfl_cgroup(current); + cgroup_inc_frozen_cnt(cgrp); + cgroup_update_frozen(cgrp); + spin_unlock_irq(&css_set_lock); +} + +/* + * Conditionally leave frozen/stopped state. Update cgroup's counters, + * and revisit the state of the cgroup, if necessary. + * + * If always_leave is not set, and the cgroup is freezing, + * we're racing with the cgroup freezing. In this case, we don't + * drop the frozen counter to avoid a transient switch to + * the unfrozen state. + */ +void cgroup_leave_frozen(bool always_leave) +{ + struct cgroup *cgrp; + + spin_lock_irq(&css_set_lock); + cgrp = task_dfl_cgroup(current); + if (always_leave || !test_bit(CGRP_FREEZE, &cgrp->flags)) { + cgroup_dec_frozen_cnt(cgrp); + cgroup_update_frozen(cgrp); + WARN_ON_ONCE(!current->frozen); + current->frozen = false; + } + spin_unlock_irq(&css_set_lock); + + if (unlikely(current->frozen)) { + /* + * If the task remained in the frozen state, + * make sure it won't reach userspace without + * entering the signal handling loop. + */ + spin_lock_irq(¤t->sighand->siglock); + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + } +} + +/* + * Freeze or unfreeze the task by setting or clearing the JOBCTL_TRAP_FREEZE + * jobctl bit. + */ +static void cgroup_freeze_task(struct task_struct *task, bool freeze) +{ + unsigned long flags; + + /* If the task is about to die, don't bother with freezing it. */ + if (!lock_task_sighand(task, &flags)) + return; + + if (freeze) { + task->jobctl |= JOBCTL_TRAP_FREEZE; + signal_wake_up(task, false); + } else { + task->jobctl &= ~JOBCTL_TRAP_FREEZE; + wake_up_process(task); + } + + unlock_task_sighand(task, &flags); +} + +/* + * Freeze or unfreeze all tasks in the given cgroup. + */ +static void cgroup_do_freeze(struct cgroup *cgrp, bool freeze) +{ + struct css_task_iter it; + struct task_struct *task; + + lockdep_assert_held(&cgroup_mutex); + + spin_lock_irq(&css_set_lock); + if (freeze) + set_bit(CGRP_FREEZE, &cgrp->flags); + else + clear_bit(CGRP_FREEZE, &cgrp->flags); + spin_unlock_irq(&css_set_lock); + + css_task_iter_start(&cgrp->self, 0, &it); + while ((task = css_task_iter_next(&it))) { + /* + * Ignore kernel threads here. Freezing cgroups containing + * kthreads isn't supported. + */ + if (task->flags & PF_KTHREAD) + continue; + cgroup_freeze_task(task, freeze); + } + css_task_iter_end(&it); + + /* + * Cgroup state should be revisited here to cover empty leaf cgroups + * and cgroups which descendants are already in the desired state. + */ + spin_lock_irq(&css_set_lock); + if (cgrp->nr_descendants == cgrp->freezer.nr_frozen_descendants) + cgroup_update_frozen(cgrp); + spin_unlock_irq(&css_set_lock); +} + +/* + * Adjust the task state (freeze or unfreeze) and revisit the state of + * source and destination cgroups. + */ +void cgroup_freezer_migrate_task(struct task_struct *task, + struct cgroup *src, struct cgroup *dst) +{ + lockdep_assert_held(&css_set_lock); + + /* + * Kernel threads are not supposed to be frozen at all. + */ + if (task->flags & PF_KTHREAD) + return; + + /* + * Adjust counters of freezing and frozen tasks. + * Note, that if the task is frozen, but the destination cgroup is not + * frozen, we bump both counters to keep them balanced. + */ + if (task->frozen) { + cgroup_inc_frozen_cnt(dst); + cgroup_dec_frozen_cnt(src); + } + cgroup_update_frozen(dst); + cgroup_update_frozen(src); + + /* + * Force the task to the desired state. + */ + cgroup_freeze_task(task, test_bit(CGRP_FREEZE, &dst->flags)); +} + +void cgroup_freezer_frozen_exit(struct task_struct *task) +{ + struct cgroup *cgrp = task_dfl_cgroup(task); + + lockdep_assert_held(&css_set_lock); + + cgroup_dec_frozen_cnt(cgrp); + cgroup_update_frozen(cgrp); +} + +void cgroup_freeze(struct cgroup *cgrp, bool freeze) +{ + struct cgroup_subsys_state *css; + struct cgroup *dsct; + bool applied = false; + + lockdep_assert_held(&cgroup_mutex); + + /* + * Nothing changed? Just exit. + */ + if (cgrp->freezer.freeze == freeze) + return; + + cgrp->freezer.freeze = freeze; + + /* + * Propagate changes downwards the cgroup tree. + */ + css_for_each_descendant_pre(css, &cgrp->self) { + dsct = css->cgroup; + + if (cgroup_is_dead(dsct)) + continue; + + if (freeze) { + dsct->freezer.e_freeze++; + /* + * Already frozen because of ancestor's settings? + */ + if (dsct->freezer.e_freeze > 1) + continue; + } else { + dsct->freezer.e_freeze--; + /* + * Still frozen because of ancestor's settings? + */ + if (dsct->freezer.e_freeze > 0) + continue; + + WARN_ON_ONCE(dsct->freezer.e_freeze < 0); + } + + /* + * Do change actual state: freeze or unfreeze. + */ + cgroup_do_freeze(dsct, freeze); + applied = true; + } + + /* + * Even if the actual state hasn't changed, let's notify a user. + * The state can be enforced by an ancestor cgroup: the cgroup + * can already be in the desired state or it can be locked in the + * opposite state, so that the transition will never happen. + * In both cases it's better to notify a user, that there is + * nothing to wait for. + */ + if (!applied) + cgroup_file_notify(&cgrp->events_file); +} diff --git a/kernel/fork.c b/kernel/fork.c index 9dcd18aa210b..8097a0cce4db 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1222,7 +1222,9 @@ static int wait_for_vfork_done(struct task_struct *child, int killed; freezer_do_not_count(); + cgroup_enter_frozen(); killed = wait_for_completion_killable(vfork); + cgroup_leave_frozen(false); freezer_count(); if (killed) { diff --git a/kernel/signal.c b/kernel/signal.c index f98448cf2def..095e0fc57b25 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -43,6 +43,7 @@ #include #include #include +#include #define CREATE_TRACE_POINTS #include @@ -146,9 +147,10 @@ static inline bool has_pending_signals(sigset_t *signal, sigset_t *blocked) static bool recalc_sigpending_tsk(struct task_struct *t) { - if ((t->jobctl & JOBCTL_PENDING_MASK) || + if ((t->jobctl & (JOBCTL_PENDING_MASK | JOBCTL_TRAP_FREEZE)) || PENDING(&t->pending, &t->blocked) || - PENDING(&t->signal->shared_pending, &t->blocked)) { + PENDING(&t->signal->shared_pending, &t->blocked) || + cgroup_task_frozen(t)) { set_tsk_thread_flag(t, TIF_SIGPENDING); return true; } @@ -2108,6 +2110,7 @@ static void ptrace_stop(int exit_code, int why, int clear_code, kernel_siginfo_t preempt_disable(); read_unlock(&tasklist_lock); preempt_enable_no_resched(); + cgroup_enter_frozen(); freezable_schedule(); } else { /* @@ -2286,6 +2289,7 @@ static bool do_signal_stop(int signr) } /* Now we don't run again until woken by SIGCONT or SIGKILL */ + cgroup_enter_frozen(); freezable_schedule(); return true; } else { @@ -2332,6 +2336,43 @@ static void do_jobctl_trap(void) } } +/** + * do_freezer_trap - handle the freezer jobctl trap + * + * Puts the task into frozen state, if only the task is not about to quit. + * In this case it drops JOBCTL_TRAP_FREEZE. + * + * CONTEXT: + * Must be called with @current->sighand->siglock held, + * which is always released before returning. + */ +static void do_freezer_trap(void) + __releases(¤t->sighand->siglock) +{ + /* + * If there are other trap bits pending except JOBCTL_TRAP_FREEZE, + * let's make another loop to give it a chance to be handled. + * In any case, we'll return back. + */ + if ((current->jobctl & (JOBCTL_PENDING_MASK | JOBCTL_TRAP_FREEZE)) != + JOBCTL_TRAP_FREEZE) { + spin_unlock_irq(¤t->sighand->siglock); + return; + } + + /* + * Now we're sure that there is no pending fatal signal and no + * pending traps. Clear TIF_SIGPENDING to not get out of schedule() + * immediately (if there is a non-fatal signal pending), and + * put the task into sleep. + */ + __set_current_state(TASK_INTERRUPTIBLE); + clear_thread_flag(TIF_SIGPENDING); + spin_unlock_irq(¤t->sighand->siglock); + cgroup_enter_frozen(); + freezable_schedule(); +} + static int ptrace_signal(int signr, kernel_siginfo_t *info) { /* @@ -2442,6 +2483,10 @@ relock: ksig->info.si_signo = signr = SIGKILL; sigdelset(¤t->pending.signal, SIGKILL); recalc_sigpending(); + current->jobctl &= ~JOBCTL_TRAP_FREEZE; + spin_unlock_irq(&sighand->siglock); + if (unlikely(cgroup_task_frozen(current))) + cgroup_leave_frozen(true); goto fatal; } @@ -2452,9 +2497,24 @@ relock: do_signal_stop(0)) goto relock; - if (unlikely(current->jobctl & JOBCTL_TRAP_MASK)) { - do_jobctl_trap(); + if (unlikely(current->jobctl & + (JOBCTL_TRAP_MASK | JOBCTL_TRAP_FREEZE))) { + if (current->jobctl & JOBCTL_TRAP_MASK) { + do_jobctl_trap(); + spin_unlock_irq(&sighand->siglock); + } else if (current->jobctl & JOBCTL_TRAP_FREEZE) + do_freezer_trap(); + + goto relock; + } + + /* + * If the task is leaving the frozen state, let's update + * cgroup counters and reset the frozen bit. + */ + if (unlikely(cgroup_task_frozen(current))) { spin_unlock_irq(&sighand->siglock); + cgroup_leave_frozen(true); goto relock; } @@ -2548,8 +2608,8 @@ relock: continue; } - fatal: spin_unlock_irq(&sighand->siglock); + fatal: /* * Anything else is fatal, maybe with a core dump. From ff9fb7cb515b32ac8d479b086c7c8c565d6905fb Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Fri, 19 Apr 2019 10:03:05 -0700 Subject: [PATCH 06/15] kselftests: cgroup: don't fail on cg_kill_all() error in cg_destroy() If the cgroup destruction races with an exit() of a belonging process(es), cg_kill_all() may fail. It's not a good reason to make cg_destroy() fail and leave the cgroup in place, potentially causing next test runs to fail. Signed-off-by: Roman Gushchin Signed-off-by: Tejun Heo Cc: Shuah Khan Cc: kernel-team@fb.com Cc: linux-kselftest@vger.kernel.org --- tools/testing/selftests/cgroup/cgroup_util.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/tools/testing/selftests/cgroup/cgroup_util.c b/tools/testing/selftests/cgroup/cgroup_util.c index 14c9fe284806..eba06f94433b 100644 --- a/tools/testing/selftests/cgroup/cgroup_util.c +++ b/tools/testing/selftests/cgroup/cgroup_util.c @@ -227,9 +227,7 @@ int cg_destroy(const char *cgroup) retry: ret = rmdir(cgroup); if (ret && errno == EBUSY) { - ret = cg_killall(cgroup); - if (ret) - return ret; + cg_killall(cgroup); usleep(100); goto retry; } From 5313bfe425c8aadc582356f575100f3235a6d638 Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Fri, 19 Apr 2019 10:03:06 -0700 Subject: [PATCH 07/15] kselftests: cgroup: add freezer controller self-tests This patch implements 9 tests for the freezer controller for cgroup v2: 1) a simple test, which aims to freeze and unfreeze a cgroup with 100 processes 2) a more complicated tree test, which creates a hierarchy of cgroups, puts some processes in some cgroups, and tries to freeze and unfreeze different parts of the subtree 3) a forkbomb test: the test aims to freeze a forkbomb running in a cgroup, kill all tasks in the cgroup and remove the cgroup without the unfreezing. 4) rmdir test: the test creates two nested cgroups, freezes the parent one, checks that the child can be successfully removed, and a new child can be created 5) migration tests: the test checks migration of a task between frozen cgroups: from a frozen to a running, from a running to a frozen, and from a frozen to a frozen. 6) ptrace test: the test checks that it's possible to attach to a process in a frozen cgroup, get some information and detach, and the cgroup will remain frozen. 7) stopped test: the test checks that it's possible to freeze a cgroup with a stopped task 8) ptraced test: the test checks that it's possible to freeze a cgroup with a ptraced task 9) vfork test: the test checks that it's possible to freeze a cgroup with a parent process waiting for the child process in vfork() Expected output: $ ./test_freezer ok 1 test_cgfreezer_simple ok 2 test_cgfreezer_tree ok 3 test_cgfreezer_forkbomb ok 4 test_cgrreezer_rmdir ok 5 test_cgfreezer_migrate ok 6 test_cgfreezer_ptrace ok 7 test_cgfreezer_stopped ok 8 test_cgfreezer_ptraced ok 9 test_cgfreezer_vfork Signed-off-by: Roman Gushchin Signed-off-by: Tejun Heo Cc: Shuah Khan Cc: kernel-team@fb.com Cc: linux-kselftest@vger.kernel.org --- tools/testing/selftests/cgroup/.gitignore | 1 + tools/testing/selftests/cgroup/Makefile | 2 + tools/testing/selftests/cgroup/cgroup_util.c | 54 +- tools/testing/selftests/cgroup/cgroup_util.h | 5 + tools/testing/selftests/cgroup/test_freezer.c | 851 ++++++++++++++++++ 5 files changed, 912 insertions(+), 1 deletion(-) create mode 100644 tools/testing/selftests/cgroup/test_freezer.c diff --git a/tools/testing/selftests/cgroup/.gitignore b/tools/testing/selftests/cgroup/.gitignore index adacda50a4b2..7f9835624793 100644 --- a/tools/testing/selftests/cgroup/.gitignore +++ b/tools/testing/selftests/cgroup/.gitignore @@ -1,2 +1,3 @@ test_memcontrol test_core +test_freezer diff --git a/tools/testing/selftests/cgroup/Makefile b/tools/testing/selftests/cgroup/Makefile index 23fbaa4a9630..8d369b6a2069 100644 --- a/tools/testing/selftests/cgroup/Makefile +++ b/tools/testing/selftests/cgroup/Makefile @@ -5,8 +5,10 @@ all: TEST_GEN_PROGS = test_memcontrol TEST_GEN_PROGS += test_core +TEST_GEN_PROGS += test_freezer include ../lib.mk $(OUTPUT)/test_memcontrol: cgroup_util.c $(OUTPUT)/test_core: cgroup_util.c +$(OUTPUT)/test_freezer: cgroup_util.c diff --git a/tools/testing/selftests/cgroup/cgroup_util.c b/tools/testing/selftests/cgroup/cgroup_util.c index eba06f94433b..4c223266299a 100644 --- a/tools/testing/selftests/cgroup/cgroup_util.c +++ b/tools/testing/selftests/cgroup/cgroup_util.c @@ -74,6 +74,16 @@ char *cg_name_indexed(const char *root, const char *name, int index) return ret; } +char *cg_control(const char *cgroup, const char *control) +{ + size_t len = strlen(cgroup) + strlen(control) + 2; + char *ret = malloc(len); + + snprintf(ret, len, "%s/%s", cgroup, control); + + return ret; +} + int cg_read(const char *cgroup, const char *control, char *buf, size_t len) { char path[PATH_MAX]; @@ -196,7 +206,32 @@ int cg_create(const char *cgroup) return mkdir(cgroup, 0644); } -static int cg_killall(const char *cgroup) +int cg_wait_for_proc_count(const char *cgroup, int count) +{ + char buf[10 * PAGE_SIZE] = {0}; + int attempts; + char *ptr; + + for (attempts = 10; attempts >= 0; attempts--) { + int nr = 0; + + if (cg_read(cgroup, "cgroup.procs", buf, sizeof(buf))) + break; + + for (ptr = buf; *ptr; ptr++) + if (*ptr == '\n') + nr++; + + if (nr >= count) + return 0; + + usleep(100000); + } + + return -1; +} + +int cg_killall(const char *cgroup) { char buf[PAGE_SIZE]; char *ptr = buf; @@ -238,6 +273,14 @@ retry: return ret; } +int cg_enter(const char *cgroup, int pid) +{ + char pidbuf[64]; + + snprintf(pidbuf, sizeof(pidbuf), "%d", pid); + return cg_write(cgroup, "cgroup.procs", pidbuf); +} + int cg_enter_current(const char *cgroup) { char pidbuf[64]; @@ -367,3 +410,12 @@ int set_oom_adj_score(int pid, int score) close(fd); return 0; } + +char proc_read_text(int pid, const char *item, char *buf, size_t size) +{ + char path[PATH_MAX]; + + snprintf(path, sizeof(path), "/proc/%d/%s", pid, item); + + return read_text(path, buf, size); +} diff --git a/tools/testing/selftests/cgroup/cgroup_util.h b/tools/testing/selftests/cgroup/cgroup_util.h index 9ac8b7958f83..c72f28046bfa 100644 --- a/tools/testing/selftests/cgroup/cgroup_util.h +++ b/tools/testing/selftests/cgroup/cgroup_util.h @@ -18,6 +18,7 @@ static inline int values_close(long a, long b, int err) extern int cg_find_unified_root(char *root, size_t len); extern char *cg_name(const char *root, const char *name); extern char *cg_name_indexed(const char *root, const char *name, int index); +extern char *cg_control(const char *cgroup, const char *control); extern int cg_create(const char *cgroup); extern int cg_destroy(const char *cgroup); extern int cg_read(const char *cgroup, const char *control, @@ -32,6 +33,7 @@ extern int cg_write(const char *cgroup, const char *control, char *buf); extern int cg_run(const char *cgroup, int (*fn)(const char *cgroup, void *arg), void *arg); +extern int cg_enter(const char *cgroup, int pid); extern int cg_enter_current(const char *cgroup); extern int cg_run_nowait(const char *cgroup, int (*fn)(const char *cgroup, void *arg), @@ -41,3 +43,6 @@ extern int alloc_pagecache(int fd, size_t size); extern int alloc_anon(const char *cgroup, void *arg); extern int is_swap_enabled(void); extern int set_oom_adj_score(int pid, int score); +extern int cg_wait_for_proc_count(const char *cgroup, int count); +extern int cg_killall(const char *cgroup); +extern char proc_read_text(int pid, const char *item, char *buf, size_t size); diff --git a/tools/testing/selftests/cgroup/test_freezer.c b/tools/testing/selftests/cgroup/test_freezer.c new file mode 100644 index 000000000000..2bfddb6d6d3b --- /dev/null +++ b/tools/testing/selftests/cgroup/test_freezer.c @@ -0,0 +1,851 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "../kselftest.h" +#include "cgroup_util.h" + +#define DEBUG +#ifdef DEBUG +#define debug(args...) fprintf(stderr, args) +#else +#define debug(args...) +#endif + +/* + * Check if the cgroup is frozen by looking at the cgroup.events::frozen value. + */ +static int cg_check_frozen(const char *cgroup, bool frozen) +{ + if (frozen) { + if (cg_read_strstr(cgroup, "cgroup.events", "frozen 1") != 0) { + debug("Cgroup %s isn't frozen\n", cgroup); + return -1; + } + } else { + /* + * Check the cgroup.events::frozen value. + */ + if (cg_read_strstr(cgroup, "cgroup.events", "frozen 0") != 0) { + debug("Cgroup %s is frozen\n", cgroup); + return -1; + } + } + + return 0; +} + +/* + * Freeze the given cgroup. + */ +static int cg_freeze_nowait(const char *cgroup, bool freeze) +{ + return cg_write(cgroup, "cgroup.freeze", freeze ? "1" : "0"); +} + +/* + * Prepare for waiting on cgroup.events file. + */ +static int cg_prepare_for_wait(const char *cgroup) +{ + int fd, ret = -1; + + fd = inotify_init1(0); + if (fd == -1) { + debug("Error: inotify_init1() failed\n"); + return fd; + } + + ret = inotify_add_watch(fd, cg_control(cgroup, "cgroup.events"), + IN_MODIFY); + if (ret == -1) { + debug("Error: inotify_add_watch() failed\n"); + close(fd); + } + + return fd; +} + +/* + * Wait for an event. If there are no events for 10 seconds, + * treat this an error. + */ +static int cg_wait_for(int fd) +{ + int ret = -1; + struct pollfd fds = { + .fd = fd, + .events = POLLIN, + }; + + while (true) { + ret = poll(&fds, 1, 10000); + + if (ret == -1) { + if (errno == EINTR) + continue; + debug("Error: poll() failed\n"); + break; + } + + if (ret > 0 && fds.revents & POLLIN) { + ret = 0; + break; + } + } + + return ret; +} + +/* + * Attach a task to the given cgroup and wait for a cgroup frozen event. + * All transient events (e.g. populated) are ignored. + */ +static int cg_enter_and_wait_for_frozen(const char *cgroup, int pid, + bool frozen) +{ + int fd, ret = -1; + int attempts; + + fd = cg_prepare_for_wait(cgroup); + if (fd < 0) + return fd; + + ret = cg_enter(cgroup, pid); + if (ret) + goto out; + + for (attempts = 0; attempts < 10; attempts++) { + ret = cg_wait_for(fd); + if (ret) + break; + + ret = cg_check_frozen(cgroup, frozen); + if (ret) + continue; + } + +out: + close(fd); + return ret; +} + +/* + * Freeze the given cgroup and wait for the inotify signal. + * If there are no events in 10 seconds, treat this as an error. + * Then check that the cgroup is in the desired state. + */ +static int cg_freeze_wait(const char *cgroup, bool freeze) +{ + int fd, ret = -1; + + fd = cg_prepare_for_wait(cgroup); + if (fd < 0) + return fd; + + ret = cg_freeze_nowait(cgroup, freeze); + if (ret) { + debug("Error: cg_freeze_nowait() failed\n"); + goto out; + } + + ret = cg_wait_for(fd); + if (ret) + goto out; + + ret = cg_check_frozen(cgroup, freeze); +out: + close(fd); + return ret; +} + +/* + * A simple process running in a sleep loop until being + * re-parented. + */ +static int child_fn(const char *cgroup, void *arg) +{ + int ppid = getppid(); + + while (getppid() == ppid) + usleep(1000); + + return getppid() == ppid; +} + +/* + * A simple test for the cgroup freezer: populated the cgroup with 100 + * running processes and freeze it. Then unfreeze it. Then it kills all + * processes and destroys the cgroup. + */ +static int test_cgfreezer_simple(const char *root) +{ + int ret = KSFT_FAIL; + char *cgroup = NULL; + int i; + + cgroup = cg_name(root, "cg_test_simple"); + if (!cgroup) + goto cleanup; + + if (cg_create(cgroup)) + goto cleanup; + + for (i = 0; i < 100; i++) + cg_run_nowait(cgroup, child_fn, NULL); + + if (cg_wait_for_proc_count(cgroup, 100)) + goto cleanup; + + if (cg_check_frozen(cgroup, false)) + goto cleanup; + + if (cg_freeze_wait(cgroup, true)) + goto cleanup; + + if (cg_freeze_wait(cgroup, false)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (cgroup) + cg_destroy(cgroup); + free(cgroup); + return ret; +} + +/* + * The test creates the following hierarchy: + * A + * / / \ \ + * B E I K + * /\ | + * C D F + * | + * G + * | + * H + * + * with a process in C, H and 3 processes in K. + * Then it tries to freeze and unfreeze the whole tree. + */ +static int test_cgfreezer_tree(const char *root) +{ + char *cgroup[10] = {0}; + int ret = KSFT_FAIL; + int i; + + cgroup[0] = cg_name(root, "cg_test_tree_A"); + if (!cgroup[0]) + goto cleanup; + + cgroup[1] = cg_name(cgroup[0], "B"); + if (!cgroup[1]) + goto cleanup; + + cgroup[2] = cg_name(cgroup[1], "C"); + if (!cgroup[2]) + goto cleanup; + + cgroup[3] = cg_name(cgroup[1], "D"); + if (!cgroup[3]) + goto cleanup; + + cgroup[4] = cg_name(cgroup[0], "E"); + if (!cgroup[4]) + goto cleanup; + + cgroup[5] = cg_name(cgroup[4], "F"); + if (!cgroup[5]) + goto cleanup; + + cgroup[6] = cg_name(cgroup[5], "G"); + if (!cgroup[6]) + goto cleanup; + + cgroup[7] = cg_name(cgroup[6], "H"); + if (!cgroup[7]) + goto cleanup; + + cgroup[8] = cg_name(cgroup[0], "I"); + if (!cgroup[8]) + goto cleanup; + + cgroup[9] = cg_name(cgroup[0], "K"); + if (!cgroup[9]) + goto cleanup; + + for (i = 0; i < 10; i++) + if (cg_create(cgroup[i])) + goto cleanup; + + cg_run_nowait(cgroup[2], child_fn, NULL); + cg_run_nowait(cgroup[7], child_fn, NULL); + cg_run_nowait(cgroup[9], child_fn, NULL); + cg_run_nowait(cgroup[9], child_fn, NULL); + cg_run_nowait(cgroup[9], child_fn, NULL); + + /* + * Wait until all child processes will enter + * corresponding cgroups. + */ + + if (cg_wait_for_proc_count(cgroup[2], 1) || + cg_wait_for_proc_count(cgroup[7], 1) || + cg_wait_for_proc_count(cgroup[9], 3)) + goto cleanup; + + /* + * Freeze B. + */ + if (cg_freeze_wait(cgroup[1], true)) + goto cleanup; + + /* + * Freeze F. + */ + if (cg_freeze_wait(cgroup[5], true)) + goto cleanup; + + /* + * Freeze G. + */ + if (cg_freeze_wait(cgroup[6], true)) + goto cleanup; + + /* + * Check that A and E are not frozen. + */ + if (cg_check_frozen(cgroup[0], false)) + goto cleanup; + + if (cg_check_frozen(cgroup[4], false)) + goto cleanup; + + /* + * Freeze A. Check that A, B and E are frozen. + */ + if (cg_freeze_wait(cgroup[0], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[1], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[4], true)) + goto cleanup; + + /* + * Unfreeze B, F and G + */ + if (cg_freeze_nowait(cgroup[1], false)) + goto cleanup; + + if (cg_freeze_nowait(cgroup[5], false)) + goto cleanup; + + if (cg_freeze_nowait(cgroup[6], false)) + goto cleanup; + + /* + * Check that C and H are still frozen. + */ + if (cg_check_frozen(cgroup[2], true)) + goto cleanup; + + if (cg_check_frozen(cgroup[7], true)) + goto cleanup; + + /* + * Unfreeze A. Check that A, C and K are not frozen. + */ + if (cg_freeze_wait(cgroup[0], false)) + goto cleanup; + + if (cg_check_frozen(cgroup[2], false)) + goto cleanup; + + if (cg_check_frozen(cgroup[9], false)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + for (i = 9; i >= 0 && cgroup[i]; i--) { + cg_destroy(cgroup[i]); + free(cgroup[i]); + } + + return ret; +} + +/* + * A fork bomb emulator. + */ +static int forkbomb_fn(const char *cgroup, void *arg) +{ + int ppid; + + fork(); + fork(); + + ppid = getppid(); + + while (getppid() == ppid) + usleep(1000); + + return getppid() == ppid; +} + +/* + * The test runs a fork bomb in a cgroup and tries to freeze it. + * Then it kills all processes and checks that cgroup isn't populated + * anymore. + */ +static int test_cgfreezer_forkbomb(const char *root) +{ + int ret = KSFT_FAIL; + char *cgroup = NULL; + + cgroup = cg_name(root, "cg_forkbomb_test"); + if (!cgroup) + goto cleanup; + + if (cg_create(cgroup)) + goto cleanup; + + cg_run_nowait(cgroup, forkbomb_fn, NULL); + + usleep(100000); + + if (cg_freeze_wait(cgroup, true)) + goto cleanup; + + if (cg_killall(cgroup)) + goto cleanup; + + if (cg_wait_for_proc_count(cgroup, 0)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (cgroup) + cg_destroy(cgroup); + free(cgroup); + return ret; +} + +/* + * The test creates two nested cgroups, freezes the parent + * and removes the child. Then it checks that the parent cgroup + * remains frozen and it's possible to create a new child + * without unfreezing. The new child is frozen too. + */ +static int test_cgfreezer_rmdir(const char *root) +{ + int ret = KSFT_FAIL; + char *parent, *child = NULL; + + parent = cg_name(root, "cg_test_rmdir_A"); + if (!parent) + goto cleanup; + + child = cg_name(parent, "cg_test_rmdir_B"); + if (!child) + goto cleanup; + + if (cg_create(parent)) + goto cleanup; + + if (cg_create(child)) + goto cleanup; + + if (cg_freeze_wait(parent, true)) + goto cleanup; + + if (cg_destroy(child)) + goto cleanup; + + if (cg_check_frozen(parent, true)) + goto cleanup; + + if (cg_create(child)) + goto cleanup; + + if (cg_check_frozen(child, true)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (child) + cg_destroy(child); + free(child); + if (parent) + cg_destroy(parent); + free(parent); + return ret; +} + +/* + * The test creates two cgroups: A and B, runs a process in A + * and performs several migrations: + * 1) A (running) -> B (frozen) + * 2) B (frozen) -> A (running) + * 3) A (frozen) -> B (frozen) + * + * On each step it checks the actual state of both cgroups. + */ +static int test_cgfreezer_migrate(const char *root) +{ + int ret = KSFT_FAIL; + char *cgroup[2] = {0}; + int pid; + + cgroup[0] = cg_name(root, "cg_test_migrate_A"); + if (!cgroup[0]) + goto cleanup; + + cgroup[1] = cg_name(root, "cg_test_migrate_B"); + if (!cgroup[1]) + goto cleanup; + + if (cg_create(cgroup[0])) + goto cleanup; + + if (cg_create(cgroup[1])) + goto cleanup; + + pid = cg_run_nowait(cgroup[0], child_fn, NULL); + if (pid < 0) + goto cleanup; + + if (cg_wait_for_proc_count(cgroup[0], 1)) + goto cleanup; + + /* + * Migrate from A (running) to B (frozen) + */ + if (cg_freeze_wait(cgroup[1], true)) + goto cleanup; + + if (cg_enter_and_wait_for_frozen(cgroup[1], pid, true)) + goto cleanup; + + if (cg_check_frozen(cgroup[0], false)) + goto cleanup; + + /* + * Migrate from B (frozen) to A (running) + */ + if (cg_enter_and_wait_for_frozen(cgroup[0], pid, false)) + goto cleanup; + + if (cg_check_frozen(cgroup[1], true)) + goto cleanup; + + /* + * Migrate from A (frozen) to B (frozen) + */ + if (cg_freeze_wait(cgroup[0], true)) + goto cleanup; + + if (cg_enter_and_wait_for_frozen(cgroup[1], pid, true)) + goto cleanup; + + if (cg_check_frozen(cgroup[0], true)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (cgroup[0]) + cg_destroy(cgroup[0]); + free(cgroup[0]); + if (cgroup[1]) + cg_destroy(cgroup[1]); + free(cgroup[1]); + return ret; +} + +/* + * The test checks that ptrace works with a tracing process in a frozen cgroup. + */ +static int test_cgfreezer_ptrace(const char *root) +{ + int ret = KSFT_FAIL; + char *cgroup = NULL; + siginfo_t siginfo; + int pid; + + cgroup = cg_name(root, "cg_test_ptrace"); + if (!cgroup) + goto cleanup; + + if (cg_create(cgroup)) + goto cleanup; + + pid = cg_run_nowait(cgroup, child_fn, NULL); + if (pid < 0) + goto cleanup; + + if (cg_wait_for_proc_count(cgroup, 1)) + goto cleanup; + + if (cg_freeze_wait(cgroup, true)) + goto cleanup; + + if (ptrace(PTRACE_SEIZE, pid, NULL, NULL)) + goto cleanup; + + if (ptrace(PTRACE_INTERRUPT, pid, NULL, NULL)) + goto cleanup; + + waitpid(pid, NULL, 0); + + /* + * Cgroup has to remain frozen, however the test task + * is in traced state. + */ + if (cg_check_frozen(cgroup, true)) + goto cleanup; + + if (ptrace(PTRACE_GETSIGINFO, pid, NULL, &siginfo)) + goto cleanup; + + if (ptrace(PTRACE_DETACH, pid, NULL, NULL)) + goto cleanup; + + if (cg_check_frozen(cgroup, true)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (cgroup) + cg_destroy(cgroup); + free(cgroup); + return ret; +} + +/* + * Check if the process is stopped. + */ +static int proc_check_stopped(int pid) +{ + char buf[PAGE_SIZE]; + int len; + + len = proc_read_text(pid, "stat", buf, sizeof(buf)); + if (len == -1) { + debug("Can't get %d stat\n", pid); + return -1; + } + + if (strstr(buf, "(test_freezer) T ") == NULL) { + debug("Process %d in the unexpected state: %s\n", pid, buf); + return -1; + } + + return 0; +} + +/* + * Test that it's possible to freeze a cgroup with a stopped process. + */ +static int test_cgfreezer_stopped(const char *root) +{ + int pid, ret = KSFT_FAIL; + char *cgroup = NULL; + + cgroup = cg_name(root, "cg_test_stopped"); + if (!cgroup) + goto cleanup; + + if (cg_create(cgroup)) + goto cleanup; + + pid = cg_run_nowait(cgroup, child_fn, NULL); + + if (cg_wait_for_proc_count(cgroup, 1)) + goto cleanup; + + if (kill(pid, SIGSTOP)) + goto cleanup; + + if (cg_check_frozen(cgroup, false)) + goto cleanup; + + if (cg_freeze_wait(cgroup, true)) + goto cleanup; + + if (cg_freeze_wait(cgroup, false)) + goto cleanup; + + if (proc_check_stopped(pid)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (cgroup) + cg_destroy(cgroup); + free(cgroup); + return ret; +} + +/* + * Test that it's possible to freeze a cgroup with a ptraced process. + */ +static int test_cgfreezer_ptraced(const char *root) +{ + int pid, ret = KSFT_FAIL; + char *cgroup = NULL; + siginfo_t siginfo; + + cgroup = cg_name(root, "cg_test_ptraced"); + if (!cgroup) + goto cleanup; + + if (cg_create(cgroup)) + goto cleanup; + + pid = cg_run_nowait(cgroup, child_fn, NULL); + + if (cg_wait_for_proc_count(cgroup, 1)) + goto cleanup; + + if (ptrace(PTRACE_SEIZE, pid, NULL, NULL)) + goto cleanup; + + if (ptrace(PTRACE_INTERRUPT, pid, NULL, NULL)) + goto cleanup; + + waitpid(pid, NULL, 0); + + if (cg_check_frozen(cgroup, false)) + goto cleanup; + + if (cg_freeze_wait(cgroup, true)) + goto cleanup; + + /* + * cg_check_frozen(cgroup, true) will fail here, + * because the task in in the TRACEd state. + */ + if (cg_freeze_wait(cgroup, false)) + goto cleanup; + + if (ptrace(PTRACE_GETSIGINFO, pid, NULL, &siginfo)) + goto cleanup; + + if (ptrace(PTRACE_DETACH, pid, NULL, NULL)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (cgroup) + cg_destroy(cgroup); + free(cgroup); + return ret; +} + +static int vfork_fn(const char *cgroup, void *arg) +{ + int pid = vfork(); + + if (pid == 0) + while (true) + sleep(1); + + return pid; +} + +/* + * Test that it's possible to freeze a cgroup with a process, + * which called vfork() and is waiting for a child. + */ +static int test_cgfreezer_vfork(const char *root) +{ + int ret = KSFT_FAIL; + char *cgroup = NULL; + + cgroup = cg_name(root, "cg_test_vfork"); + if (!cgroup) + goto cleanup; + + if (cg_create(cgroup)) + goto cleanup; + + cg_run_nowait(cgroup, vfork_fn, NULL); + + if (cg_wait_for_proc_count(cgroup, 2)) + goto cleanup; + + if (cg_freeze_wait(cgroup, true)) + goto cleanup; + + ret = KSFT_PASS; + +cleanup: + if (cgroup) + cg_destroy(cgroup); + free(cgroup); + return ret; +} + +#define T(x) { x, #x } +struct cgfreezer_test { + int (*fn)(const char *root); + const char *name; +} tests[] = { + T(test_cgfreezer_simple), + T(test_cgfreezer_tree), + T(test_cgfreezer_forkbomb), + T(test_cgfreezer_rmdir), + T(test_cgfreezer_migrate), + T(test_cgfreezer_ptrace), + T(test_cgfreezer_stopped), + T(test_cgfreezer_ptraced), + T(test_cgfreezer_vfork), +}; +#undef T + +int main(int argc, char *argv[]) +{ + char root[PATH_MAX]; + int i, ret = EXIT_SUCCESS; + + if (cg_find_unified_root(root, sizeof(root))) + ksft_exit_skip("cgroup v2 isn't mounted\n"); + for (i = 0; i < ARRAY_SIZE(tests); i++) { + switch (tests[i].fn(root)) { + case KSFT_PASS: + ksft_test_result_pass("%s\n", tests[i].name); + break; + case KSFT_SKIP: + ksft_test_result_skip("%s\n", tests[i].name); + break; + default: + ret = EXIT_FAILURE; + ksft_test_result_fail("%s\n", tests[i].name); + break; + } + } + + return ret; +} From 712e35178754bbb785d00d5fcf5abaf32699bf11 Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Fri, 19 Apr 2019 10:03:07 -0700 Subject: [PATCH 08/15] cgroup: make TRACE_CGROUP_PATH irq-safe To use the TRACE_CGROUP_PATH() macro with css_set_lock locked, let's make the macro irq-safe. It's necessary in order to trace cgroup freezer state transitions (frozen/not frozen), which are happening with css_set_lock locked. Signed-off-by: Roman Gushchin Signed-off-by: Tejun Heo --- kernel/cgroup/cgroup-internal.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/kernel/cgroup/cgroup-internal.h b/kernel/cgroup/cgroup-internal.h index 02c001ffe2e2..809e34a3c017 100644 --- a/kernel/cgroup/cgroup-internal.h +++ b/kernel/cgroup/cgroup-internal.h @@ -28,12 +28,15 @@ extern void __init enable_debug_cgroup(void); #define TRACE_CGROUP_PATH(type, cgrp, ...) \ do { \ if (trace_cgroup_##type##_enabled()) { \ - spin_lock(&trace_cgroup_path_lock); \ + unsigned long flags; \ + spin_lock_irqsave(&trace_cgroup_path_lock, \ + flags); \ cgroup_path(cgrp, trace_cgroup_path, \ TRACE_CGROUP_PATH_LEN); \ trace_cgroup_##type(cgrp, trace_cgroup_path, \ ##__VA_ARGS__); \ - spin_unlock(&trace_cgroup_path_lock); \ + spin_unlock_irqrestore(&trace_cgroup_path_lock, \ + flags); \ } \ } while (0) From 4c476d8cff48853645abc822154aaad208faebcc Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Fri, 19 Apr 2019 10:03:08 -0700 Subject: [PATCH 09/15] cgroup: add tracing points for cgroup v2 freezer Add cgroup:cgroup_freeze and cgroup:cgroup_unfreeze events, which are using the existing cgroup tracing infrastructure. Add the cgroup_event event class, which is similar to the cgroup class, but contains an additional integer field to store a new value (the level field is dropped). Also add two tracing events: cgroup_notify_populated and cgroup_notify_frozen, which are raised in a generic way using the TRACE_CGROUP_PATH() macro. This allows to trace cgroup state transitions and is generally helpful for debugging the cgroup freezer code. Signed-off-by: Roman Gushchin Signed-off-by: Tejun Heo --- include/trace/events/cgroup.h | 55 +++++++++++++++++++++++++++++++++++ kernel/cgroup/cgroup.c | 2 ++ kernel/cgroup/freezer.c | 15 +++++++++- 3 files changed, 71 insertions(+), 1 deletion(-) diff --git a/include/trace/events/cgroup.h b/include/trace/events/cgroup.h index a401ff5e7847..a566cc521476 100644 --- a/include/trace/events/cgroup.h +++ b/include/trace/events/cgroup.h @@ -103,6 +103,20 @@ DEFINE_EVENT(cgroup, cgroup_rename, TP_ARGS(cgrp, path) ); +DEFINE_EVENT(cgroup, cgroup_freeze, + + TP_PROTO(struct cgroup *cgrp, const char *path), + + TP_ARGS(cgrp, path) +); + +DEFINE_EVENT(cgroup, cgroup_unfreeze, + + TP_PROTO(struct cgroup *cgrp, const char *path), + + TP_ARGS(cgrp, path) +); + DECLARE_EVENT_CLASS(cgroup_migrate, TP_PROTO(struct cgroup *dst_cgrp, const char *path, @@ -149,6 +163,47 @@ DEFINE_EVENT(cgroup_migrate, cgroup_transfer_tasks, TP_ARGS(dst_cgrp, path, task, threadgroup) ); +DECLARE_EVENT_CLASS(cgroup_event, + + TP_PROTO(struct cgroup *cgrp, const char *path, int val), + + TP_ARGS(cgrp, path, val), + + TP_STRUCT__entry( + __field( int, root ) + __field( int, id ) + __field( int, level ) + __string( path, path ) + __field( int, val ) + ), + + TP_fast_assign( + __entry->root = cgrp->root->hierarchy_id; + __entry->id = cgrp->id; + __entry->level = cgrp->level; + __assign_str(path, path); + __entry->val = val; + ), + + TP_printk("root=%d id=%d level=%d path=%s val=%d", + __entry->root, __entry->id, __entry->level, __get_str(path), + __entry->val) +); + +DEFINE_EVENT(cgroup_event, cgroup_notify_populated, + + TP_PROTO(struct cgroup *cgrp, const char *path, int val), + + TP_ARGS(cgrp, path, val) +); + +DEFINE_EVENT(cgroup_event, cgroup_notify_frozen, + + TP_PROTO(struct cgroup *cgrp, const char *path, int val), + + TP_ARGS(cgrp, path, val) +); + #endif /* _TRACE_CGROUP_H */ /* This part must be outside protection */ diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 6895464b54c6..57edcf398d71 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -816,6 +816,8 @@ static void cgroup_update_populated(struct cgroup *cgrp, bool populated) break; cgroup1_check_for_release(cgrp); + TRACE_CGROUP_PATH(notify_populated, cgrp, + cgroup_is_populated(cgrp)); cgroup_file_notify(&cgrp->events_file); child = cgrp; diff --git a/kernel/cgroup/freezer.c b/kernel/cgroup/freezer.c index 9d8cda478fc9..3bfbb3c8baf3 100644 --- a/kernel/cgroup/freezer.c +++ b/kernel/cgroup/freezer.c @@ -6,6 +6,8 @@ #include "cgroup-internal.h" +#include + /* * Propagate the cgroup frozen state upwards by the cgroup tree. */ @@ -28,6 +30,7 @@ static void cgroup_propagate_frozen(struct cgroup *cgrp, bool frozen) cgrp->nr_descendants) { set_bit(CGRP_FROZEN, &cgrp->flags); cgroup_file_notify(&cgrp->events_file); + TRACE_CGROUP_PATH(notify_frozen, cgrp, 1); desc++; } } else { @@ -35,6 +38,7 @@ static void cgroup_propagate_frozen(struct cgroup *cgrp, bool frozen) if (test_bit(CGRP_FROZEN, &cgrp->flags)) { clear_bit(CGRP_FROZEN, &cgrp->flags); cgroup_file_notify(&cgrp->events_file); + TRACE_CGROUP_PATH(notify_frozen, cgrp, 0); desc++; } } @@ -73,6 +77,7 @@ void cgroup_update_frozen(struct cgroup *cgrp) clear_bit(CGRP_FROZEN, &cgrp->flags); } cgroup_file_notify(&cgrp->events_file); + TRACE_CGROUP_PATH(notify_frozen, cgrp, frozen); /* Update the state of ancestor cgroups. */ cgroup_propagate_frozen(cgrp, frozen); @@ -189,6 +194,11 @@ static void cgroup_do_freeze(struct cgroup *cgrp, bool freeze) clear_bit(CGRP_FREEZE, &cgrp->flags); spin_unlock_irq(&css_set_lock); + if (freeze) + TRACE_CGROUP_PATH(freeze, cgrp); + else + TRACE_CGROUP_PATH(unfreeze, cgrp); + css_task_iter_start(&cgrp->self, 0, &it); while ((task = css_task_iter_next(&it))) { /* @@ -312,6 +322,9 @@ void cgroup_freeze(struct cgroup *cgrp, bool freeze) * In both cases it's better to notify a user, that there is * nothing to wait for. */ - if (!applied) + if (!applied) { + TRACE_CGROUP_PATH(notify_frozen, cgrp, + test_bit(CGRP_FROZEN, &cgrp->flags)); cgroup_file_notify(&cgrp->events_file); + } } From afe471ea2d6fcd439c3a7a1fe2df0389e07a67c8 Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Fri, 19 Apr 2019 10:03:09 -0700 Subject: [PATCH 10/15] cgroup: document cgroup v2 freezer interface Describe cgroup v2 freezer interface in the cgroup v2 admin guide. Signed-off-by: Roman Gushchin Reviewed-by: Mike Rapoport Signed-off-by: Tejun Heo Cc: linux-doc@vger.kernel.org Cc: kernel-team@fb.com --- Documentation/admin-guide/cgroup-v2.rst | 27 +++++++++++++++++++++++++ 1 file changed, 27 insertions(+) diff --git a/Documentation/admin-guide/cgroup-v2.rst b/Documentation/admin-guide/cgroup-v2.rst index 20f92c16ffbf..88e746074252 100644 --- a/Documentation/admin-guide/cgroup-v2.rst +++ b/Documentation/admin-guide/cgroup-v2.rst @@ -864,6 +864,8 @@ All cgroup core files are prefixed with "cgroup." populated 1 if the cgroup or its descendants contains any live processes; otherwise, 0. + frozen + 1 if the cgroup is frozen; otherwise, 0. cgroup.max.descendants A read-write single value files. The default is "max". @@ -897,6 +899,31 @@ All cgroup core files are prefixed with "cgroup." A dying cgroup can consume system resources not exceeding limits, which were active at the moment of cgroup deletion. + cgroup.freeze + A read-write single value file which exists on non-root cgroups. + Allowed values are "0" and "1". The default is "0". + + Writing "1" to the file causes freezing of the cgroup and all + descendant cgroups. This means that all belonging processes will + be stopped and will not run until the cgroup will be explicitly + unfrozen. Freezing of the cgroup may take some time; when this action + is completed, the "frozen" value in the cgroup.events control file + will be updated to "1" and the corresponding notification will be + issued. + + A cgroup can be frozen either by its own settings, or by settings + of any ancestor cgroups. If any of ancestor cgroups is frozen, the + cgroup will remain frozen. + + Processes in the frozen cgroup can be killed by a fatal signal. + They also can enter and leave a frozen cgroup: either by an explicit + move by a user, or if freezing of the cgroup races with fork(). + If a process is moved to a frozen cgroup, it stops. If a process is + moved out of a frozen cgroup, it becomes running. + + Frozen status of a cgroup doesn't affect any cgroup tree operations: + it's possible to delete a frozen (and empty) cgroup, as well as + create new sub-cgroups. Controllers =========== From 533307dc20a9e84a0687d4ca24aeb669516c0243 Mon Sep 17 00:00:00 2001 From: Shaokun Zhang Date: Tue, 30 Apr 2019 17:57:29 +0800 Subject: [PATCH 11/15] cgroup: Remove unused cgrp variable The 'cgrp' is set but not used in commit <76f969e8948d8> ("cgroup: cgroup v2 freezer"). Remove it to avoid [-Wunused-but-set-variable] warning. Cc: Tejun Heo Signed-off-by: Shaokun Zhang Acked-by: Roman Gushchin Signed-off-by: Tejun Heo --- kernel/cgroup/cgroup.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 57edcf398d71..4fe9f7f1a3fa 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -5864,11 +5864,8 @@ void cgroup_post_fork(struct task_struct *child) * the task into the frozen state. */ if (unlikely(cgroup_task_freeze(child))) { - struct cgroup *cgrp; - spin_lock(&child->sighand->siglock); WARN_ON_ONCE(child->frozen); - cgrp = cset->dfl_cgrp; child->jobctl |= JOBCTL_TRAP_FREEZE; spin_unlock(&child->sighand->siglock); From cb2c4cd87874a7975b7b8615866b3a87bae10aab Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Fri, 26 Apr 2019 10:59:44 -0700 Subject: [PATCH 12/15] cgroup: prevent spurious transition into non-frozen state If freezing of a cgroup races with waking of a task from the frozen state (like waiting in vfork() or in do_signal_stop()), a spurious transition of the cgroup state can happen. The task enters cgroup_leave_frozen(true), the cgroup->nr_frozen_tasks counter decrements, and the cgroup is switched to the unfrozen state. To prevent it, let's reserve cgroup_leave_frozen(true) for terminating processes and use cgroup_leave_frozen(false) otherwise. To avoid busy-looping in the signal handling loop waiting for JOBCTL_TRAP_FREEZE set from the cgroup freezing path, let's do it explicitly in cgroup_leave_frozen(), if the task is going to stay frozen. Suggested-by: Oleg Nesterov Signed-off-by: Roman Gushchin Signed-off-by: Tejun Heo --- kernel/cgroup/freezer.c | 16 +++++----------- kernel/signal.c | 2 +- 2 files changed, 6 insertions(+), 12 deletions(-) diff --git a/kernel/cgroup/freezer.c b/kernel/cgroup/freezer.c index 3bfbb3c8baf3..c321e768f8d3 100644 --- a/kernel/cgroup/freezer.c +++ b/kernel/cgroup/freezer.c @@ -139,19 +139,13 @@ void cgroup_leave_frozen(bool always_leave) cgroup_update_frozen(cgrp); WARN_ON_ONCE(!current->frozen); current->frozen = false; + } else if (!(current->jobctl & JOBCTL_TRAP_FREEZE)) { + spin_lock(¤t->sighand->siglock); + current->jobctl |= JOBCTL_TRAP_FREEZE; + set_thread_flag(TIF_SIGPENDING); + spin_unlock(¤t->sighand->siglock); } spin_unlock_irq(&css_set_lock); - - if (unlikely(current->frozen)) { - /* - * If the task remained in the frozen state, - * make sure it won't reach userspace without - * entering the signal handling loop. - */ - spin_lock_irq(¤t->sighand->siglock); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - } } /* diff --git a/kernel/signal.c b/kernel/signal.c index 095e0fc57b25..16b72f4f14df 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -2514,7 +2514,7 @@ relock: */ if (unlikely(cgroup_task_frozen(current))) { spin_unlock_irq(&sighand->siglock); - cgroup_leave_frozen(true); + cgroup_leave_frozen(false); goto relock; } From 96b9c592def5d7203bdad1337d9c92a2183de5cb Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Fri, 26 Apr 2019 10:59:45 -0700 Subject: [PATCH 13/15] cgroup: get rid of cgroup_freezer_frozen_exit() A task should never enter the exit path with the task->frozen bit set. Any frozen task must enter the signal handling loop and the only way to escape is through cgroup_leave_frozen(true), which unconditionally drops the task->frozen bit. So it means that cgroyp_freezer_frozen_exit() has zero chances to be called and has to be removed. Let's put a WARN_ON_ONCE() instead of the cgroup_freezer_frozen_exit() call to catch any potential leak of the task's frozen bit. Suggested-by: Oleg Nesterov Signed-off-by: Roman Gushchin Signed-off-by: Tejun Heo --- include/linux/cgroup.h | 2 +- kernel/cgroup/cgroup.c | 5 ++--- kernel/cgroup/freezer.c | 10 ---------- 3 files changed, 3 insertions(+), 14 deletions(-) diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 3e2efd412dfa..c0077adeea83 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -889,7 +889,7 @@ void cgroup_update_frozen(struct cgroup *cgrp); void cgroup_freeze(struct cgroup *cgrp, bool freeze); void cgroup_freezer_migrate_task(struct task_struct *task, struct cgroup *src, struct cgroup *dst); -void cgroup_freezer_frozen_exit(struct task_struct *task); + static inline bool cgroup_task_freeze(struct task_struct *task) { bool ret; diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 4fe9f7f1a3fa..327f37c9fdfa 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -5926,9 +5926,8 @@ void cgroup_exit(struct task_struct *tsk) css_set_move_task(tsk, cset, NULL, false); cset->nr_tasks--; - if (unlikely(cgroup_task_frozen(tsk))) - cgroup_freezer_frozen_exit(tsk); - else if (unlikely(cgroup_task_freeze(tsk))) + WARN_ON_ONCE(cgroup_task_frozen(tsk)); + if (unlikely(cgroup_task_freeze(tsk))) cgroup_update_frozen(task_dfl_cgroup(tsk)); spin_unlock_irq(&css_set_lock); diff --git a/kernel/cgroup/freezer.c b/kernel/cgroup/freezer.c index c321e768f8d3..8cf010680678 100644 --- a/kernel/cgroup/freezer.c +++ b/kernel/cgroup/freezer.c @@ -248,16 +248,6 @@ void cgroup_freezer_migrate_task(struct task_struct *task, cgroup_freeze_task(task, test_bit(CGRP_FREEZE, &dst->flags)); } -void cgroup_freezer_frozen_exit(struct task_struct *task) -{ - struct cgroup *cgrp = task_dfl_cgroup(task); - - lockdep_assert_held(&css_set_lock); - - cgroup_dec_frozen_cnt(cgrp); - cgroup_update_frozen(cgrp); -} - void cgroup_freeze(struct cgroup *cgrp, bool freeze) { struct cgroup_subsys_state *css; From 1900da520c9fdc3a7cf00d21638f7c8721d5ac7f Mon Sep 17 00:00:00 2001 From: Fuqian Huang Date: Sun, 21 Apr 2019 19:47:27 +0800 Subject: [PATCH 14/15] kernel: cgroup: fix misuse of %x Pointers should be printed with %p or %px rather than cast to unsigned long type and printed with %lx. Change %lx to %p to print the pointers. Signed-off-by: Fuqian Huang Signed-off-by: Tejun Heo --- kernel/cgroup/debug.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/kernel/cgroup/debug.c b/kernel/cgroup/debug.c index 5f1b87330bee..80aa3f027ac3 100644 --- a/kernel/cgroup/debug.c +++ b/kernel/cgroup/debug.c @@ -64,8 +64,8 @@ static int current_css_set_read(struct seq_file *seq, void *v) css = cset->subsys[ss->id]; if (!css) continue; - seq_printf(seq, "%2d: %-4s\t- %lx[%d]\n", ss->id, ss->name, - (unsigned long)css, css->id); + seq_printf(seq, "%2d: %-4s\t- %p[%d]\n", ss->id, ss->name, + css, css->id); } rcu_read_unlock(); spin_unlock_irq(&css_set_lock); @@ -224,8 +224,8 @@ static int cgroup_subsys_states_read(struct seq_file *seq, void *v) if (css->parent) snprintf(pbuf, sizeof(pbuf) - 1, " P=%d", css->parent->id); - seq_printf(seq, "%2d: %-4s\t- %lx[%d] %d%s\n", ss->id, ss->name, - (unsigned long)css, css->id, + seq_printf(seq, "%2d: %-4s\t- %p[%d] %d%s\n", ss->id, ss->name, + css, css->id, atomic_read(&css->online_cnt), pbuf); } From f2b31bb598248c04721cb8485e6091a9feb045ac Mon Sep 17 00:00:00 2001 From: Roman Gushchin Date: Wed, 8 May 2019 13:34:20 -0700 Subject: [PATCH 15/15] cgroup: never call do_group_exit() with task->frozen bit set I've got two independent reports that cgroup_task_frozen() check in cgroup_exit() has been triggered by lkp libhugetlbfs-test and LTP ptrace01 tests. For example: [ 44.576072] WARNING: CPU: 1 PID: 3028 at kernel/cgroup/cgroup.c:5932 cgroup_exit+0x148/0x160 [ 44.577724] Modules linked in: crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel sr_mod cdrom bochs_drm sg ttm ata_generic pata_acpi ppdev drm_kms_helper snd_pcm syscopyarea aesni_intel snd_timer sysfillrect sysimgblt snd crypto_simd cryptd glue_helper soundcore fb_sys_fops joydev drm serio_raw pcspkr ata_piix libata i2c_piix4 floppy parport_pc parport ip_tables [ 44.583106] CPU: 1 PID: 3028 Comm: ptrace-write-hu Not tainted 5.1.0-rc3-00053-g9262503 #5 [ 44.584600] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1 04/01/2014 [ 44.586116] RIP: 0010:cgroup_exit+0x148/0x160 [ 44.587135] Code: 0f 84 50 ff ff ff 48 8b 85 c8 0c 00 00 48 8b 78 70 e8 ec 2e 00 00 e9 3b ff ff ff f0 ff 43 60 0f 88 72 21 89 00 e9 48 ff ff ff <0f> 0b e9 1b ff ff ff e8 3c 73 f4 ff 66 90 66 2e 0f 1f 84 00 00 00 [ 44.590113] RSP: 0018:ffffb25702dcfd30 EFLAGS: 00010002 [ 44.591167] RAX: ffff96a7fee32410 RBX: ffff96a7ff1d6000 RCX: dead000000000200 [ 44.592446] RDX: ffff96a7ff1d6080 RSI: ffff96a7fec75290 RDI: ffff96a7fec75290 [ 44.593715] RBP: ffff96a7fec745c0 R08: ffff96a7fec74658 R09: 0000000000000000 [ 44.594985] R10: 0000000000000000 R11: 0000000000000001 R12: ffff96a7fec75101 [ 44.596266] R13: ffff96a7fec745c0 R14: ffff96a7ff3bde30 R15: ffff96a7fec75130 [ 44.597550] FS: 0000000000000000(0000) GS:ffff96a7dd700000(0000) knlGS:0000000000000000 [ 44.598950] CS: 0010 DS: 002b ES: 002b CR0: 0000000080050033 [ 44.600098] CR2: 00000000f7a00000 CR3: 000000000d20e000 CR4: 00000000000406e0 [ 44.601417] Call Trace: [ 44.602777] do_exit+0x337/0xc40 [ 44.603677] do_group_exit+0x3a/0xa0 [ 44.604610] get_signal+0x12e/0x8d0 [ 44.605533] ? __switch_to_asm+0x40/0x70 [ 44.606503] do_signal+0x36/0x650 [ 44.607409] ? __switch_to_asm+0x40/0x70 [ 44.608383] ? __schedule+0x267/0x860 [ 44.609329] exit_to_usermode_loop+0x89/0xf0 [ 44.610349] do_fast_syscall_32+0x251/0x2e3 [ 44.611357] entry_SYSENTER_compat+0x7f/0x91 [ 44.612376] ---[ end trace e4ca5cfc4b7f7964 ]--- The problem is caused by the ptrace_signal() call in the for loop in get_signal(). There is a cgroup_enter_frozen() call inside ptrace_signal(), so after exit from ptrace_signal() the task->frozen bit might be set. In this case do_group_exit() can be called with the task->frozen bit set and trigger the warning. This is only place where we can leave the loop with the task->frozen bit set and without setting JOBCTL_TRAP_FREEZE and TIF_SIGPENDING. To resolve this problem, let's move cgroup_leave_frozen(true) call to just after the fatal label. If the task is going to die, the frozen bit must be cleared no matter how we get into this point. Reported-by: kernel test robot Reported-by: Qian Cai Cc: Oleg Nesterov Cc: Tejun Heo Signed-off-by: Roman Gushchin Signed-off-by: Tejun Heo --- kernel/signal.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/kernel/signal.c b/kernel/signal.c index 16b72f4f14df..8607b11ff936 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -2483,10 +2483,6 @@ relock: ksig->info.si_signo = signr = SIGKILL; sigdelset(¤t->pending.signal, SIGKILL); recalc_sigpending(); - current->jobctl &= ~JOBCTL_TRAP_FREEZE; - spin_unlock_irq(&sighand->siglock); - if (unlikely(cgroup_task_frozen(current))) - cgroup_leave_frozen(true); goto fatal; } @@ -2608,8 +2604,10 @@ relock: continue; } - spin_unlock_irq(&sighand->siglock); fatal: + spin_unlock_irq(&sighand->siglock); + if (unlikely(cgroup_task_frozen(current))) + cgroup_leave_frozen(true); /* * Anything else is fatal, maybe with a core dump.