From 5edee61edeaaebafe584f8fb7074c1ef4658596b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 16 Oct 2012 15:03:14 -0700 Subject: cgroup: cgroup_subsys->fork() should be called after the task is added to css_set cgroup core has a bug which violates a basic rule about event notifications - when a new entity needs to be added, you add that to the notification list first and then make the new entity conform to the current state. If done in the reverse order, an event happening inbetween will be lost. cgroup_subsys->fork() is invoked way before the new task is added to the css_set. Currently, cgroup_freezer is the only user of ->fork() and uses it to make new tasks conform to the current state of the freezer. If FROZEN state is requested while fork is in progress between cgroup_fork_callbacks() and cgroup_post_fork(), the child could escape freezing - the cgroup isn't frozen when ->fork() is called and the freezer couldn't see the new task on the css_set. This patch moves cgroup_subsys->fork() invocation to cgroup_post_fork() after the new task is added to the css_set. cgroup_fork_callbacks() is removed. Because now a task may be migrated during cgroup_subsys->fork(), freezer_fork() is updated so that it adheres to the usual RCU locking and the rather pointless comment on why locking can be different there is removed (if it doesn't make anything simpler, why even bother?). Signed-off-by: Tejun Heo Cc: Oleg Nesterov Cc: Rafael J. Wysocki Cc: stable@vger.kernel.org --- include/linux/cgroup.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index f8a030ced0c7..4cd1d0fd2542 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -34,7 +34,6 @@ extern int cgroup_lock_is_held(void); extern bool cgroup_lock_live_group(struct cgroup *cgrp); extern void cgroup_unlock(void); extern void cgroup_fork(struct task_struct *p); -extern void cgroup_fork_callbacks(struct task_struct *p); extern void cgroup_post_fork(struct task_struct *p); extern void cgroup_exit(struct task_struct *p, int run_callbacks); extern int cgroupstats_build(struct cgroupstats *stats, -- cgit v1.2.3 From dd67d32dbc5de299d70cc9e10c6c1e29ffa56b92 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 16 Oct 2012 15:03:14 -0700 Subject: freezer: add missing mb's to freezer_count() and freezer_should_skip() A task is considered frozen enough between freezer_do_not_count() and freezer_count() and freezers use freezer_should_skip() to test this condition. This supposedly works because freezer_count() always calls try_to_freezer() after clearing %PF_FREEZER_SKIP. However, there currently is nothing which guarantees that freezer_count() sees %true freezing() after clearing %PF_FREEZER_SKIP when freezing is in progress, and vice-versa. A task can escape the freezing condition in effect by freezer_count() seeing !freezing() and freezer_should_skip() seeing %PF_FREEZER_SKIP. This patch adds smp_mb()'s to freezer_count() and freezer_should_skip() such that either %true freezing() is visible to freezer_count() or !PF_FREEZER_SKIP is visible to freezer_should_skip(). Signed-off-by: Tejun Heo Cc: Oleg Nesterov Cc: Rafael J. Wysocki Cc: stable@vger.kernel.org --- include/linux/freezer.h | 50 +++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 42 insertions(+), 8 deletions(-) (limited to 'include/linux') diff --git a/include/linux/freezer.h b/include/linux/freezer.h index d09af4b67cf1..ee899329e65a 100644 --- a/include/linux/freezer.h +++ b/include/linux/freezer.h @@ -75,28 +75,62 @@ static inline bool cgroup_freezing(struct task_struct *task) */ -/* Tell the freezer not to count the current task as freezable. */ +/** + * freezer_do_not_count - tell freezer to ignore %current + * + * Tell freezers to ignore the current task when determining whether the + * target frozen state is reached. IOW, the current task will be + * considered frozen enough by freezers. + * + * The caller shouldn't do anything which isn't allowed for a frozen task + * until freezer_cont() is called. Usually, freezer[_do_not]_count() pair + * wrap a scheduling operation and nothing much else. + */ static inline void freezer_do_not_count(void) { current->flags |= PF_FREEZER_SKIP; } -/* - * Tell the freezer to count the current task as freezable again and try to - * freeze it. +/** + * freezer_count - tell freezer to stop ignoring %current + * + * Undo freezer_do_not_count(). It tells freezers that %current should be + * considered again and tries to freeze if freezing condition is already in + * effect. */ static inline void freezer_count(void) { current->flags &= ~PF_FREEZER_SKIP; + /* + * If freezing is in progress, the following paired with smp_mb() + * in freezer_should_skip() ensures that either we see %true + * freezing() or freezer_should_skip() sees !PF_FREEZER_SKIP. + */ + smp_mb(); try_to_freeze(); } -/* - * Check if the task should be counted as freezable by the freezer +/** + * freezer_should_skip - whether to skip a task when determining frozen + * state is reached + * @p: task in quesion + * + * This function is used by freezers after establishing %true freezing() to + * test whether a task should be skipped when determining the target frozen + * state is reached. IOW, if this function returns %true, @p is considered + * frozen enough. */ -static inline int freezer_should_skip(struct task_struct *p) +static inline bool freezer_should_skip(struct task_struct *p) { - return !!(p->flags & PF_FREEZER_SKIP); + /* + * The following smp_mb() paired with the one in freezer_count() + * ensures that either freezer_count() sees %true freezing() or we + * see cleared %PF_FREEZER_SKIP and return %false. This makes it + * impossible for a task to slip frozen state testing after + * clearing %PF_FREEZER_SKIP. + */ + smp_mb(); + return p->flags & PF_FREEZER_SKIP; } /* -- cgit v1.2.3 From 5d8f72b55c275677865de670fa147ed318191d81 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Fri, 26 Oct 2012 19:46:06 +0200 Subject: freezer: change ptrace_stop/do_signal_stop to use freezable_schedule() try_to_freeze_tasks() and cgroup_freezer rely on scheduler locks to ensure that a task doing STOPPED/TRACED -> RUNNING transition can't escape freezing. This mostly works, but ptrace_stop() does not necessarily call schedule(), it can change task->state back to RUNNING and check freezing() without any lock/barrier in between. We could add the necessary barrier, but this patch changes ptrace_stop() and do_signal_stop() to use freezable_schedule(). This fixes the race, freezer_count() and freezer_should_skip() carefully avoid the race. And this simplifies the code, try_to_freeze_tasks/update_if_frozen no longer need to use task_is_stopped_or_traced() checks with the non trivial assumptions. We can rely on the mechanism which was specially designed to mark the sleeping task as "frozen enough". v2: As Tejun pointed out, we can also change get_signal_to_deliver() and move try_to_freeze() up before 'relock' label. Signed-off-by: Oleg Nesterov Signed-off-by: Tejun Heo --- include/linux/freezer.h | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/freezer.h b/include/linux/freezer.h index ee899329e65a..8039893bc3ec 100644 --- a/include/linux/freezer.h +++ b/include/linux/freezer.h @@ -134,10 +134,9 @@ static inline bool freezer_should_skip(struct task_struct *p) } /* - * These macros are intended to be used whenever you want allow a task that's - * sleeping in TASK_UNINTERRUPTIBLE or TASK_KILLABLE state to be frozen. Note - * that neither return any clear indication of whether a freeze event happened - * while in this function. + * These macros are intended to be used whenever you want allow a sleeping + * task to be frozen. Note that neither return any clear indication of + * whether a freeze event happened while in this function. */ /* Like schedule(), but should not block the freezer. */ -- cgit v1.2.3 From ed95779340b50e362245c81b5dec0d11a1debfa8 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 5 Nov 2012 09:16:58 -0800 Subject: cgroup: kill cgroup_subsys->__DEPRECATED_clear_css_refs 2ef37d3fe4 ("memcg: Simplify mem_cgroup_force_empty_list error handling") removed the last user of __DEPRECATED_clear_css_refs. This patch removes __DEPRECATED_clear_css_refs and mechanisms to support it. * Conditionals dependent on __DEPRECATED_clear_css_refs removed. * cgroup_clear_css_refs() can no longer fail. All that needs to be done are deactivating refcnts, setting CSS_REMOVED and putting the base reference on each css. Remove cgroup_clear_css_refs() and the failure path, and open-code the loops into cgroup_rmdir(). This patch keeps the two for_each_subsys() loops separate while open coding them. They can be merged now but there are scheduled changes which need them to be separate, so keep them separate to reduce the amount of churn. local_irq_save/restore() from cgroup_clear_css_refs() are replaced with local_irq_disable/enable() for simplicity. This is safe as cgroup_rmdir() is always called with IRQ enabled. Note that this IRQ switching is necessary to ensure that css_tryget() isn't called from IRQ context on the same CPU while lower context is between CSS deactivation and setting CSS_REMOVED as css_tryget() would hang forever in such cases waiting for CSS to be re-activated or CSS_REMOVED set. This will go away soon. v2: cgroup_call_pre_destroy() removal dropped per Michal. Commit message updated to explain local_irq_disable/enable() conversion. Signed-off-by: Tejun Heo Reviewed-by: Michal Hocko Reviewed-by: KAMEZAWA Hiroyuki Acked-by: Li Zefan --- include/linux/cgroup.h | 12 ------------ 1 file changed, 12 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index c90eaa803440..02e09c0e98ab 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -86,7 +86,6 @@ struct cgroup_subsys_state { enum { CSS_ROOT, /* This CSS is the root of the subsystem */ CSS_REMOVED, /* This CSS is dead */ - CSS_CLEAR_CSS_REFS, /* @ss->__DEPRECATED_clear_css_refs */ }; /* Caller must verify that the css is not for root cgroup */ @@ -485,17 +484,6 @@ struct cgroup_subsys { */ bool use_id; - /* - * If %true, cgroup removal will try to clear css refs by retrying - * ss->pre_destroy() until there's no css ref left. This behavior - * is strictly for backward compatibility and will be removed as - * soon as the current user (memcg) is updated. - * - * If %false, ss->pre_destroy() can't fail and cgroup removal won't - * wait for css refs to drop to zero before proceeding. - */ - bool __DEPRECATED_clear_css_refs; - #define MAX_CGROUP_TYPE_NAMELEN 32 const char *name; -- cgit v1.2.3 From e93160803ffda2e67d9ff9cacb63bb6868c8398f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 5 Nov 2012 09:16:58 -0800 Subject: cgroup: kill CSS_REMOVED CSS_REMOVED is one of the several contortions which were necessary to support css reference draining on cgroup removal. All css->refcnts which need draining should be deactivated and verified to equal zero atomically w.r.t. css_tryget(). If any one isn't zero, all refcnts needed to be re-activated and css_tryget() shouldn't fail in the process. This was achieved by letting css_tryget() busy-loop until either the refcnt is reactivated (failed removal attempt) or CSS_REMOVED is set (committing to removal). Now that css refcnt draining is no longer used, there's no need for atomic rollback mechanism. css_tryget() simply can look at the reference count and fail if it's deactivated - it's never getting re-activated. This patch removes CSS_REMOVED and updates __css_tryget() to fail if the refcnt is deactivated. As deactivation and removal are a single step now, they no longer need to be protected against css_tryget() happening from irq context. Remove local_irq_disable/enable() from cgroup_rmdir(). Note that this removes css_is_removed() whose only user is VM_BUG_ON() in memcontrol.c. We can replace it with a check on the refcnt but given that the only use case is a debug assert, I think it's better to simply unexport it. v2: Comment updated and explanation on local_irq_disable/enable() added per Michal Hocko. Signed-off-by: Tejun Heo Reviewed-by: Michal Hocko Reviewed-by: KAMEZAWA Hiroyuki Acked-by: Li Zefan Cc: Johannes Weiner Cc: Balbir Singh --- include/linux/cgroup.h | 6 ------ 1 file changed, 6 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 02e09c0e98ab..a3098046250b 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -85,7 +85,6 @@ struct cgroup_subsys_state { /* bits in struct cgroup_subsys_state flags field */ enum { CSS_ROOT, /* This CSS is the root of the subsystem */ - CSS_REMOVED, /* This CSS is dead */ }; /* Caller must verify that the css is not for root cgroup */ @@ -108,11 +107,6 @@ static inline void css_get(struct cgroup_subsys_state *css) __css_get(css, 1); } -static inline bool css_is_removed(struct cgroup_subsys_state *css) -{ - return test_bit(CSS_REMOVED, &css->flags); -} - /* * Call css_tryget() to take a reference on a css if your existing * (known-valid) reference isn't already ref-counted. Returns false if -- cgit v1.2.3 From b25ed609d0eecf077db607e88ea70bae83b6adb2 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 5 Nov 2012 09:16:59 -0800 Subject: cgroup: remove CGRP_WAIT_ON_RMDIR, cgroup_exclude_rmdir() and cgroup_release_and_wakeup_rmdir() CGRP_WAIT_ON_RMDIR is another kludge which was added to make cgroup destruction rollback somewhat working. cgroup_rmdir() used to drain CSS references and CGRP_WAIT_ON_RMDIR and the associated waitqueue and helpers were used to allow the task performing rmdir to wait for the next relevant event. Unfortunately, the wait is visible to controllers too and the mechanism got exposed to memcg by 887032670d ("cgroup avoid permanent sleep at rmdir"). Now that the draining and retries are gone, CGRP_WAIT_ON_RMDIR is unnecessary. Remove it and all the mechanisms supporting it. Note that memcontrol.c changes are essentially revert of 887032670d ("cgroup avoid permanent sleep at rmdir"). Signed-off-by: Tejun Heo Reviewed-by: Michal Hocko Reviewed-by: KAMEZAWA Hiroyuki Acked-by: Li Zefan Cc: Balbir Singh --- include/linux/cgroup.h | 21 --------------------- 1 file changed, 21 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index a3098046250b..47868a86ba2b 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -144,10 +144,6 @@ enum { CGRP_RELEASABLE, /* Control Group requires release notifications to userspace */ CGRP_NOTIFY_ON_RELEASE, - /* - * A thread in rmdir() is wating for this cgroup. - */ - CGRP_WAIT_ON_RMDIR, /* * Clone cgroup values when creating a new child cgroup */ @@ -411,23 +407,6 @@ int cgroup_task_count(const struct cgroup *cgrp); /* Return true if cgrp is a descendant of the task's cgroup */ int cgroup_is_descendant(const struct cgroup *cgrp, struct task_struct *task); -/* - * When the subsys has to access css and may add permanent refcnt to css, - * it should take care of racy conditions with rmdir(). Following set of - * functions, is for stop/restart rmdir if necessary. - * Because these will call css_get/put, "css" should be alive css. - * - * cgroup_exclude_rmdir(); - * ...do some jobs which may access arbitrary empty cgroup - * cgroup_release_and_wakeup_rmdir(); - * - * When someone removes a cgroup while cgroup_exclude_rmdir() holds it, - * it sleeps and cgroup_release_and_wakeup_rmdir() will wake him up. - */ - -void cgroup_exclude_rmdir(struct cgroup_subsys_state *css); -void cgroup_release_and_wakeup_rmdir(struct cgroup_subsys_state *css); - /* * Control Group taskset, used to pass around set of tasks to cgroup_subsys * methods. -- cgit v1.2.3 From bcf6de1b9129531215d26dd9af8331e84973bc52 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 5 Nov 2012 09:16:59 -0800 Subject: cgroup: make ->pre_destroy() return void All ->pre_destory() implementations return 0 now, which is the only allowed return value. Make it return void. Signed-off-by: Tejun Heo Reviewed-by: Michal Hocko Acked-by: KAMEZAWA Hiroyuki Acked-by: Li Zefan Cc: Balbir Singh Cc: Vivek Goyal --- include/linux/cgroup.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 47868a86ba2b..adb2adc8ec1a 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -436,7 +436,7 @@ int cgroup_taskset_size(struct cgroup_taskset *tset); struct cgroup_subsys { struct cgroup_subsys_state *(*create)(struct cgroup *cgrp); - int (*pre_destroy)(struct cgroup *cgrp); + void (*pre_destroy)(struct cgroup *cgrp); void (*destroy)(struct cgroup *cgrp); int (*can_attach)(struct cgroup *cgrp, struct cgroup_taskset *tset); void (*cancel_attach)(struct cgroup *cgrp, struct cgroup_taskset *tset); -- cgit v1.2.3 From a8638030f668884720b8f4456448d0ce33952b05 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 9 Nov 2012 09:12:29 -0800 Subject: cgroup: add cgroup_subsys->post_create() Currently, there's no way for a controller to find out whether a new cgroup finished all ->create() allocatinos successfully and is considered "live" by cgroup. This becomes a problem later when we add generic descendants walking to cgroup which can be used by controllers as controllers don't have a synchronization point where it can synchronize against new cgroups appearing in such walks. This patch adds ->post_create(). It's called after all ->create() succeeded and the cgroup is linked into the generic cgroup hierarchy. This plays the counterpart of ->pre_destroy(). When used in combination with the to-be-added generic descendant iterators, ->post_create() can be used to implement reliable state inheritance. It will be explained with the descendant iterators. v2: Added a paragraph about its future use w/ descendant iterators per Michal. v3: Forgot to add ->post_create() invocation to cgroup_load_subsys(). Fixed. Signed-off-by: Tejun Heo Acked-by: Michal Hocko Reviewed-by: KAMEZAWA Hiroyuki Acked-by: Li Zefan Cc: Glauber Costa --- include/linux/cgroup.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index fe876a77031a..b4421221111d 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -438,6 +438,7 @@ int cgroup_taskset_size(struct cgroup_taskset *tset); struct cgroup_subsys { struct cgroup_subsys_state *(*create)(struct cgroup *cgrp); + void (*post_create)(struct cgroup *cgrp); void (*pre_destroy)(struct cgroup *cgrp); void (*destroy)(struct cgroup *cgrp); int (*can_attach)(struct cgroup *cgrp, struct cgroup_taskset *tset); -- cgit v1.2.3 From eb6fd5040ee799009173daa49c3e7aa0362167c9 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 9 Nov 2012 09:12:29 -0800 Subject: cgroup: use rculist ops for cgroup->children Use RCU safe list operations for cgroup->children. This will be used to implement cgroup children / descendant walking which can be used by controllers. Note that cgroup_create() now puts a new cgroup at the end of the ->children list instead of head. This isn't strictly necessary but is done so that the iteration order is more conventional. Signed-off-by: Tejun Heo Reviewed-by: Michal Hocko Reviewed-by: KAMEZAWA Hiroyuki Acked-by: Li Zefan --- include/linux/cgroup.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index b4421221111d..90c33ebdd6bc 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include -- cgit v1.2.3 From 574bd9f7c7c1d32f52dea5488018a6ff79031e22 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 9 Nov 2012 09:12:29 -0800 Subject: cgroup: implement generic child / descendant walk macros Currently, cgroup doesn't provide any generic helper for walking a given cgroup's children or descendants. This patch adds the following three macros. * cgroup_for_each_child() - walk immediate children of a cgroup. * cgroup_for_each_descendant_pre() - visit all descendants of a cgroup in pre-order tree traversal. * cgroup_for_each_descendant_post() - visit all descendants of a cgroup in post-order tree traversal. All three only require the user to hold RCU read lock during traversal. Verifying that each iterated cgroup is online is the responsibility of the user. When used with proper synchronization, cgroup_for_each_descendant_pre() can be used to propagate state updates to descendants in reliable way. See comments for details. v2: s/config/state/ in commit message and comments per Michal. More documentation on synchronization rules. Signed-off-by: Tejun Heo Reviewed-by: KAMEZAWA Hiroyuki Reviewed-by: Michal Hocko Acked-by: Li Zefan --- include/linux/cgroup.h | 94 ++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 94 insertions(+) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 90c33ebdd6bc..8f64b459fbd4 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -534,6 +534,100 @@ static inline struct cgroup* task_cgroup(struct task_struct *task, return task_subsys_state(task, subsys_id)->cgroup; } +/** + * cgroup_for_each_child - iterate through children of a cgroup + * @pos: the cgroup * to use as the loop cursor + * @cgroup: cgroup whose children to walk + * + * Walk @cgroup's children. Must be called under rcu_read_lock(). A child + * cgroup which hasn't finished ->post_create() or already has finished + * ->pre_destroy() may show up during traversal and it's each subsystem's + * responsibility to verify that each @pos is alive. + * + * If a subsystem synchronizes against the parent in its ->post_create() + * and before starting iterating, a cgroup which finished ->post_create() + * is guaranteed to be visible in the future iterations. + */ +#define cgroup_for_each_child(pos, cgroup) \ + list_for_each_entry_rcu(pos, &(cgroup)->children, sibling) + +struct cgroup *cgroup_next_descendant_pre(struct cgroup *pos, + struct cgroup *cgroup); + +/** + * cgroup_for_each_descendant_pre - pre-order walk of a cgroup's descendants + * @pos: the cgroup * to use as the loop cursor + * @cgroup: cgroup whose descendants to walk + * + * Walk @cgroup's descendants. Must be called under rcu_read_lock(). A + * descendant cgroup which hasn't finished ->post_create() or already has + * finished ->pre_destroy() may show up during traversal and it's each + * subsystem's responsibility to verify that each @pos is alive. + * + * If a subsystem synchronizes against the parent in its ->post_create() + * and before starting iterating, and synchronizes against @pos on each + * iteration, any descendant cgroup which finished ->post_create() is + * guaranteed to be visible in the future iterations. + * + * In other words, the following guarantees that a descendant can't escape + * state updates of its ancestors. + * + * my_post_create(@cgrp) + * { + * Lock @cgrp->parent and @cgrp; + * Inherit state from @cgrp->parent; + * Unlock both. + * } + * + * my_update_state(@cgrp) + * { + * Lock @cgrp; + * Update @cgrp's state; + * Unlock @cgrp; + * + * cgroup_for_each_descendant_pre(@pos, @cgrp) { + * Lock @pos; + * Verify @pos is alive and inherit state from @pos->parent; + * Unlock @pos; + * } + * } + * + * As long as the inheriting step, including checking the parent state, is + * enclosed inside @pos locking, double-locking the parent isn't necessary + * while inheriting. The state update to the parent is guaranteed to be + * visible by walking order and, as long as inheriting operations to the + * same @pos are atomic to each other, multiple updates racing each other + * still result in the correct state. It's guaranateed that at least one + * inheritance happens for any cgroup after the latest update to its + * parent. + * + * If checking parent's state requires locking the parent, each inheriting + * iteration should lock and unlock both @pos->parent and @pos. + * + * Alternatively, a subsystem may choose to use a single global lock to + * synchronize ->post_create() and ->pre_destroy() against tree-walking + * operations. + */ +#define cgroup_for_each_descendant_pre(pos, cgroup) \ + for (pos = cgroup_next_descendant_pre(NULL, (cgroup)); (pos); \ + pos = cgroup_next_descendant_pre((pos), (cgroup))) + +struct cgroup *cgroup_next_descendant_post(struct cgroup *pos, + struct cgroup *cgroup); + +/** + * cgroup_for_each_descendant_post - post-order walk of a cgroup's descendants + * @pos: the cgroup * to use as the loop cursor + * @cgroup: cgroup whose descendants to walk + * + * Similar to cgroup_for_each_descendant_pre() but performs post-order + * traversal instead. Note that the walk visibility guarantee described in + * pre-order walk doesn't apply the same to post-order walks. + */ +#define cgroup_for_each_descendant_post(pos, cgroup) \ + for (pos = cgroup_next_descendant_post(NULL, (cgroup)); (pos); \ + pos = cgroup_next_descendant_post((pos), (cgroup))) + /* A cgroup_iter should be treated as an opaque object */ struct cgroup_iter { struct list_head *cg_link; -- cgit v1.2.3 From febfcef60d4f9457785b45aab548bc7ee5ea158f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 19 Nov 2012 08:13:36 -0800 Subject: cgroup: cgroup->dentry isn't a RCU pointer cgroup->dentry is marked and used as a RCU pointer; however, it isn't one - the final dentry put doesn't go through call_rcu(). cgroup and dentry share the same RCU freeing rule via synchronize_rcu() in cgroup_diput() (kfree_rcu() used on cgrp is unnecessary). If cgrp is accessible under RCU read lock, so is its dentry and dereferencing cgrp->dentry doesn't need any further RCU protection or annotation. While not being accurate, before the previous patch, the RCU accessors served a purpose as memory barriers - cgroup->dentry used to be assigned after the cgroup was made visible to cgroup_path(), so the assignment and dereferencing in cgroup_path() needed the memory barrier pair. Now that list_add_tail_rcu() happens after cgroup->dentry is assigned, this no longer is necessary. Remove the now unnecessary and misleading RCU annotations from cgroup->dentry. To make up for the removal of rcu_dereference_check() in cgroup_path(), add an explicit rcu_lockdep_assert(), which asserts the dereference rule of @cgrp, not cgrp->dentry. Signed-off-by: Tejun Heo Acked-by: Li Zefan --- include/linux/cgroup.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 8f64b459fbd4..d605857c4bf3 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -165,7 +165,7 @@ struct cgroup { struct list_head files; /* my files */ struct cgroup *parent; /* my parent */ - struct dentry __rcu *dentry; /* cgroup fs entry, RCU protected */ + struct dentry *dentry; /* cgroup fs entry, RCU protected */ /* Private pointers for each registered subsystem */ struct cgroup_subsys_state *subsys[CGROUP_SUBSYS_COUNT]; -- cgit v1.2.3 From 38b53abaa3e0c7e750ef73eee919cf42eee6b134 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 19 Nov 2012 08:13:36 -0800 Subject: cgroup: make CSS_* flags bit masks instead of bit positions Currently, CSS_* flags are defined as bit positions and manipulated using atomic bitops. There's no reason to use atomic bitops for them and bit positions are clunkier to deal with than bit masks. Make CSS_* bit masks instead and use the usual C bitwise operators to access them. Signed-off-by: Tejun Heo Acked-by: Li Zefan --- include/linux/cgroup.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index d605857c4bf3..a0fc64167129 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -81,7 +81,7 @@ struct cgroup_subsys_state { /* bits in struct cgroup_subsys_state flags field */ enum { - CSS_ROOT, /* This CSS is the root of the subsystem */ + CSS_ROOT = (1 << 0), /* this CSS is the root of the subsystem */ }; /* Caller must verify that the css is not for root cgroup */ @@ -100,7 +100,7 @@ static inline void __css_get(struct cgroup_subsys_state *css, int count) static inline void css_get(struct cgroup_subsys_state *css) { /* We don't need to reference count the root state */ - if (!test_bit(CSS_ROOT, &css->flags)) + if (!(css->flags & CSS_ROOT)) __css_get(css, 1); } @@ -113,7 +113,7 @@ static inline void css_get(struct cgroup_subsys_state *css) extern bool __css_tryget(struct cgroup_subsys_state *css); static inline bool css_tryget(struct cgroup_subsys_state *css) { - if (test_bit(CSS_ROOT, &css->flags)) + if (css->flags & CSS_ROOT) return true; return __css_tryget(css); } @@ -126,7 +126,7 @@ static inline bool css_tryget(struct cgroup_subsys_state *css) extern void __css_put(struct cgroup_subsys_state *css); static inline void css_put(struct cgroup_subsys_state *css) { - if (!test_bit(CSS_ROOT, &css->flags)) + if (!(css->flags & CSS_ROOT)) __css_put(css); } -- cgit v1.2.3 From a31f2d3ff7fe20cbe2a143515a7d7c408b29dd0d Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 19 Nov 2012 08:13:37 -0800 Subject: cgroup: introduce CSS_ONLINE flag and on/offline_css() helpers New helpers on/offline_css() respectively wrap ->post_create() and ->pre_destroy() invocations. online_css() sets CSS_ONLINE after ->post_create() is complete and offline_css() invokes ->pre_destroy() iff CSS_ONLINE is set and clears it while also handling the temporary dropping of cgroup_mutex. This patch doesn't introduce any behavior change at the moment but will be used to improve cgroup_create() failure path and allow ->post_create() to fail. Signed-off-by: Tejun Heo Acked-by: Li Zefan --- include/linux/cgroup.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index a0fc64167129..f4a9c9836906 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -82,6 +82,7 @@ struct cgroup_subsys_state { /* bits in struct cgroup_subsys_state flags field */ enum { CSS_ROOT = (1 << 0), /* this CSS is the root of the subsystem */ + CSS_ONLINE = (1 << 1), /* between ->post_create() and ->pre_destroy() */ }; /* Caller must verify that the css is not for root cgroup */ -- cgit v1.2.3 From b1929db42f8a649d9a9e397119f628c27fd4021f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 19 Nov 2012 08:13:38 -0800 Subject: cgroup: allow ->post_create() to fail There could be cases where controllers want to do initialization operations which may fail from ->post_create(). This patch makes ->post_create() return -errno to indicate failure and online_css() relay such failures. Signed-off-by: Tejun Heo Acked-by: Li Zefan Cc: Glauber Costa --- include/linux/cgroup.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index f4a9c9836906..03d8a92786da 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -440,7 +440,7 @@ int cgroup_taskset_size(struct cgroup_taskset *tset); struct cgroup_subsys { struct cgroup_subsys_state *(*create)(struct cgroup *cgrp); - void (*post_create)(struct cgroup *cgrp); + int (*post_create)(struct cgroup *cgrp); void (*pre_destroy)(struct cgroup *cgrp); void (*destroy)(struct cgroup *cgrp); int (*can_attach)(struct cgroup *cgrp, struct cgroup_taskset *tset); -- cgit v1.2.3 From 92fb97487a7e41b222c1417cabd1d1ab7cc3a48c Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 19 Nov 2012 08:13:38 -0800 Subject: cgroup: rename ->create/post_create/pre_destroy/destroy() to ->css_alloc/online/offline/free() Rename cgroup_subsys css lifetime related callbacks to better describe what their roles are. Also, update documentation. Signed-off-by: Tejun Heo Acked-by: Li Zefan --- include/linux/cgroup.h | 35 ++++++++++++++++++----------------- 1 file changed, 18 insertions(+), 17 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 03d8a92786da..7a2189ca8327 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -82,7 +82,7 @@ struct cgroup_subsys_state { /* bits in struct cgroup_subsys_state flags field */ enum { CSS_ROOT = (1 << 0), /* this CSS is the root of the subsystem */ - CSS_ONLINE = (1 << 1), /* between ->post_create() and ->pre_destroy() */ + CSS_ONLINE = (1 << 1), /* between ->css_online() and ->css_offline() */ }; /* Caller must verify that the css is not for root cgroup */ @@ -439,10 +439,11 @@ int cgroup_taskset_size(struct cgroup_taskset *tset); */ struct cgroup_subsys { - struct cgroup_subsys_state *(*create)(struct cgroup *cgrp); - int (*post_create)(struct cgroup *cgrp); - void (*pre_destroy)(struct cgroup *cgrp); - void (*destroy)(struct cgroup *cgrp); + struct cgroup_subsys_state *(*css_alloc)(struct cgroup *cgrp); + int (*css_online)(struct cgroup *cgrp); + void (*css_offline)(struct cgroup *cgrp); + void (*css_free)(struct cgroup *cgrp); + int (*can_attach)(struct cgroup *cgrp, struct cgroup_taskset *tset); void (*cancel_attach)(struct cgroup *cgrp, struct cgroup_taskset *tset); void (*attach)(struct cgroup *cgrp, struct cgroup_taskset *tset); @@ -541,13 +542,13 @@ static inline struct cgroup* task_cgroup(struct task_struct *task, * @cgroup: cgroup whose children to walk * * Walk @cgroup's children. Must be called under rcu_read_lock(). A child - * cgroup which hasn't finished ->post_create() or already has finished - * ->pre_destroy() may show up during traversal and it's each subsystem's + * cgroup which hasn't finished ->css_online() or already has finished + * ->css_offline() may show up during traversal and it's each subsystem's * responsibility to verify that each @pos is alive. * - * If a subsystem synchronizes against the parent in its ->post_create() - * and before starting iterating, a cgroup which finished ->post_create() - * is guaranteed to be visible in the future iterations. + * If a subsystem synchronizes against the parent in its ->css_online() and + * before starting iterating, a cgroup which finished ->css_online() is + * guaranteed to be visible in the future iterations. */ #define cgroup_for_each_child(pos, cgroup) \ list_for_each_entry_rcu(pos, &(cgroup)->children, sibling) @@ -561,19 +562,19 @@ struct cgroup *cgroup_next_descendant_pre(struct cgroup *pos, * @cgroup: cgroup whose descendants to walk * * Walk @cgroup's descendants. Must be called under rcu_read_lock(). A - * descendant cgroup which hasn't finished ->post_create() or already has - * finished ->pre_destroy() may show up during traversal and it's each + * descendant cgroup which hasn't finished ->css_online() or already has + * finished ->css_offline() may show up during traversal and it's each * subsystem's responsibility to verify that each @pos is alive. * - * If a subsystem synchronizes against the parent in its ->post_create() - * and before starting iterating, and synchronizes against @pos on each - * iteration, any descendant cgroup which finished ->post_create() is + * If a subsystem synchronizes against the parent in its ->css_online() and + * before starting iterating, and synchronizes against @pos on each + * iteration, any descendant cgroup which finished ->css_offline() is * guaranteed to be visible in the future iterations. * * In other words, the following guarantees that a descendant can't escape * state updates of its ancestors. * - * my_post_create(@cgrp) + * my_online(@cgrp) * { * Lock @cgrp->parent and @cgrp; * Inherit state from @cgrp->parent; @@ -606,7 +607,7 @@ struct cgroup *cgroup_next_descendant_pre(struct cgroup *pos, * iteration should lock and unlock both @pos->parent and @pos. * * Alternatively, a subsystem may choose to use a single global lock to - * synchronize ->post_create() and ->pre_destroy() against tree-walking + * synchronize ->css_online() and ->css_offline() against tree-walking * operations. */ #define cgroup_for_each_descendant_pre(pos, cgroup) \ -- cgit v1.2.3 From 2260e7fc1f18ad815324605c1ce7d5c6fd9b19a2 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 19 Nov 2012 08:13:38 -0800 Subject: cgroup: s/CGRP_CLONE_CHILDREN/CGRP_CPUSET_CLONE_CHILDREN/ clone_children is only meaningful for cpuset and will stay that way. Rename the flag to reflect that and update documentation. Also, drop clone_children() wrapper in cgroup.c. The thin wrapper is used only a few times and one of them will go away soon. Signed-off-by: Tejun Heo Acked-by: Serge E. Hallyn Acked-by: Li Zefan Cc: Glauber Costa --- include/linux/cgroup.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 7a2189ca8327..d2f82979f6c1 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -143,9 +143,11 @@ enum { /* Control Group requires release notifications to userspace */ CGRP_NOTIFY_ON_RELEASE, /* - * Clone cgroup values when creating a new child cgroup + * Clone the parent's configuration when creating a new child + * cpuset cgroup. For historical reasons, this option can be + * specified at mount time and thus is implemented here. */ - CGRP_CLONE_CHILDREN, + CGRP_CPUSET_CLONE_CHILDREN, }; struct cgroup { -- cgit v1.2.3 From 033fa1c5f5f73833598a0beb022c0048fb769dad Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 19 Nov 2012 08:13:39 -0800 Subject: cgroup, cpuset: remove cgroup_subsys->post_clone() Currently CGRP_CPUSET_CLONE_CHILDREN triggers ->post_clone(). Now that clone_children is cpuset specific, there's no reason to have this rather odd option activation mechanism in cgroup core. cpuset can check the flag from its ->css_allocate() and take the necessary action. Move cpuset_post_clone() logic to the end of cpuset_css_alloc() and remove cgroup_subsys->post_clone(). Loosely based on Glauber's "generalize post_clone into post_create" patch. Signed-off-by: Tejun Heo Original-patch-by: Glauber Costa Original-patch: <1351686554-22592-2-git-send-email-glommer@parallels.com> Acked-by: Serge E. Hallyn Acked-by: Li Zefan Cc: Glauber Costa --- include/linux/cgroup.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index d2f82979f6c1..c798997e5011 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -452,7 +452,6 @@ struct cgroup_subsys { void (*fork)(struct task_struct *task); void (*exit)(struct cgroup *cgrp, struct cgroup *old_cgrp, struct task_struct *task); - void (*post_clone)(struct cgroup *cgrp); void (*bind)(struct cgroup *root); int subsys_id; -- cgit v1.2.3 From 0a950f65e1e64f4e82b4b5507773848ea88bcb8e Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 19 Nov 2012 09:02:12 -0800 Subject: cgroup: add cgroup->id With the introduction of generic cgroup hierarchy iterators, css_id is being phased out. It was unnecessarily complex, id'ing the wrong thing (cgroups need IDs, not CSSes) and has other oddities like not being available at ->css_alloc(). This patch adds cgroup->id, which is a simple per-hierarchy ida-allocated ID which is assigned before ->css_alloc() and released after ->css_free(). Signed-off-by: Tejun Heo Acked-by: Li Zefan Acked-by: Neil Horman --- include/linux/cgroup.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index c798997e5011..82cf9e6fc77b 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -159,6 +159,8 @@ struct cgroup { */ atomic_t count; + int id; /* ida allocated in-hierarchy ID */ + /* * We link our 'sibling' struct into our parent's 'children'. * Our children link their 'sibling' into our 'children'. -- cgit v1.2.3 From d0b2fdd2a51203f04ea0a5d716e033c15e0231af Mon Sep 17 00:00:00 2001 From: Tao Ma Date: Tue, 20 Nov 2012 22:06:18 +0800 Subject: cgroup: remove obsolete guarantee from cgroup_task_migrate. 'guarantee' is already removed from cgroup_task_migrate, so remove the corresponding comments. Some other typos in cgroup are also changed. Cc: Tejun Heo Cc: Li Zefan Signed-off-by: Tao Ma Signed-off-by: Tejun Heo --- include/linux/cgroup.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 82cf9e6fc77b..7d73905dcba2 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -66,7 +66,7 @@ struct cgroup_subsys_state { /* * State maintained by the cgroup system to allow subsystems * to be "busy". Should be accessed via css_get(), - * css_tryget() and and css_put(). + * css_tryget() and css_put(). */ atomic_t refcnt; @@ -276,7 +276,7 @@ struct cgroup_map_cb { /* cftype->flags */ #define CFTYPE_ONLY_ON_ROOT (1U << 0) /* only create on root cg */ -#define CFTYPE_NOT_ON_ROOT (1U << 1) /* don't create onp root cg */ +#define CFTYPE_NOT_ON_ROOT (1U << 1) /* don't create on root cg */ #define MAX_CFTYPE_NAME 64 -- cgit v1.2.3