X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=kernel%2Fcgroup.c;h=b6eadfe30e7be1627e81b8158e1d581c4575b11f;hb=8177e6d6dfb9cd03d9bdeb647c32161f8f58f686;hp=068f58da855a166a0f6704b95f8a5b28e78ae4ca;hpb=4fe91d518e4958af7edebbeb112a3272b2be232d;p=safe%2Fjmp%2Flinux-2.6 diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 068f58d..b6eadfe 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -44,6 +44,10 @@ #include #include #include +#include +#include +#include +#include #include @@ -82,21 +86,16 @@ struct cgroupfs_root { /* Tracks how many cgroups are currently defined in hierarchy.*/ int number_of_cgroups; - /* A list running through the mounted hierarchies */ + /* A list running through the active hierarchies */ struct list_head root_list; /* Hierarchy-specific flags */ unsigned long flags; - /* The path to use for release notifications. No locking - * between setting and use - so if userspace updates this - * while child cgroups exist, you could miss a - * notification. We ensure that it's always a valid - * NUL-terminated string */ + /* The path to use for release notifications. */ char release_agent_path[PATH_MAX]; }; - /* * The "rootnode" hierarchy is the "dummy hierarchy", reserved for the * subsystems that are otherwise unattached - it never has more than a @@ -104,6 +103,39 @@ struct cgroupfs_root { */ static struct cgroupfs_root rootnode; +/* + * CSS ID -- ID per subsys's Cgroup Subsys State(CSS). used only when + * cgroup_subsys->use_id != 0. + */ +#define CSS_ID_MAX (65535) +struct css_id { + /* + * The css to which this ID points. This pointer is set to valid value + * after cgroup is populated. If cgroup is removed, this will be NULL. + * This pointer is expected to be RCU-safe because destroy() + * is called after synchronize_rcu(). But for safe use, css_is_removed() + * css_tryget() should be used for avoiding race. + */ + struct cgroup_subsys_state *css; + /* + * ID of this css. + */ + unsigned short id; + /* + * Depth in hierarchy which this ID belongs to. + */ + unsigned short depth; + /* + * ID is freed by RCU. (and lookup routine is RCU safe.) + */ + struct rcu_head rcu_head; + /* + * Hierarchy of CSS ID belongs to. + */ + unsigned short stack[0]; /* Array of Length (depth+1) */ +}; + + /* The list of hierarchy roots */ static LIST_HEAD(roots); @@ -117,18 +149,7 @@ static int root_count; * extra work in the fork/exit path if none of the subsystems need to * be called. */ -static int need_forkexit_callback; - -/* bits in struct cgroup flags field */ -enum { - /* Control Group is dead */ - CGRP_REMOVED, - /* Control Group has previously had a child cgroup or a task, - * but no longer (only if CGRP_NOTIFY_ON_RELEASE is set) */ - CGRP_RELEASABLE, - /* Control Group requires release notifications to userspace */ - CGRP_NOTIFY_ON_RELEASE, -}; +static int need_forkexit_callback __read_mostly; /* convenient tests for these bits */ inline int cgroup_is_removed(const struct cgroup *cgrp) @@ -161,8 +182,8 @@ static int notify_on_release(const struct cgroup *cgrp) #define for_each_subsys(_root, _ss) \ list_for_each_entry(_ss, &_root->subsys_list, sibling) -/* for_each_root() allows you to iterate across the active hierarchies */ -#define for_each_root(_root) \ +/* for_each_active_root() allows you to iterate across the active hierarchies */ +#define for_each_active_root(_root) \ list_for_each_entry(_root, &roots, root_list) /* the list of cgroups eligible for automatic release. Protected by @@ -198,17 +219,40 @@ struct cg_cgroup_link { static struct css_set init_css_set; static struct cg_cgroup_link init_css_set_link; +static int cgroup_subsys_init_idr(struct cgroup_subsys *ss); + /* css_set_lock protects the list of css_set objects, and the * chain of tasks off each css_set. Nests outside task->alloc_lock * due to cgroup_iter_start() */ static DEFINE_RWLOCK(css_set_lock); static int css_set_count; +/* hash table for cgroup groups. This improves the performance to + * find an existing css_set */ +#define CSS_SET_HASH_BITS 7 +#define CSS_SET_TABLE_SIZE (1 << CSS_SET_HASH_BITS) +static struct hlist_head css_set_table[CSS_SET_TABLE_SIZE]; + +static struct hlist_head *css_set_hash(struct cgroup_subsys_state *css[]) +{ + int i; + int index; + unsigned long tmp = 0UL; + + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) + tmp += (unsigned long)css[i]; + tmp = (tmp >> 16) ^ tmp; + + index = hash_long(tmp, CSS_SET_HASH_BITS); + + return &css_set_table[index]; +} + /* We don't maintain the lists running through each css_set to its * task until after the first call to cgroup_iter_start(). This * reduces the fork()/exit() overhead for people who have cgroups * compiled into their kernel but not actually in use */ -static int use_task_css_set_links; +static int use_task_css_set_links __read_mostly; /* When we create or destroy a css_set, the operation simply * takes/releases a reference count on all the cgroups referenced @@ -229,30 +273,41 @@ static int use_task_css_set_links; */ static void unlink_css_set(struct css_set *cg) { - write_lock(&css_set_lock); - list_del(&cg->list); + struct cg_cgroup_link *link; + struct cg_cgroup_link *saved_link; + + hlist_del(&cg->hlist); css_set_count--; - while (!list_empty(&cg->cg_links)) { - struct cg_cgroup_link *link; - link = list_entry(cg->cg_links.next, - struct cg_cgroup_link, cg_link_list); + + list_for_each_entry_safe(link, saved_link, &cg->cg_links, + cg_link_list) { list_del(&link->cg_link_list); list_del(&link->cgrp_link_list); kfree(link); } - write_unlock(&css_set_lock); } -static void __release_css_set(struct kref *k, int taskexit) +static void __put_css_set(struct css_set *cg, int taskexit) { int i; - struct css_set *cg = container_of(k, struct css_set, ref); - + /* + * Ensure that the refcount doesn't hit zero while any readers + * can see it. Similar to atomic_dec_and_lock(), but for an + * rwlock + */ + if (atomic_add_unless(&cg->refcount, -1, 1)) + return; + write_lock(&css_set_lock); + if (!atomic_dec_and_test(&cg->refcount)) { + write_unlock(&css_set_lock); + return; + } unlink_css_set(cg); + write_unlock(&css_set_lock); rcu_read_lock(); for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { - struct cgroup *cgrp = cg->subsys[i]->cgroup; + struct cgroup *cgrp = rcu_dereference(cg->subsys[i]->cgroup); if (atomic_dec_and_test(&cgrp->count) && notify_on_release(cgrp)) { if (taskexit) @@ -264,40 +319,28 @@ static void __release_css_set(struct kref *k, int taskexit) kfree(cg); } -static void release_css_set(struct kref *k) -{ - __release_css_set(k, 0); -} - -static void release_css_set_taskexit(struct kref *k) -{ - __release_css_set(k, 1); -} - /* * refcounted get/put for css_set objects */ static inline void get_css_set(struct css_set *cg) { - kref_get(&cg->ref); + atomic_inc(&cg->refcount); } static inline void put_css_set(struct css_set *cg) { - kref_put(&cg->ref, release_css_set); + __put_css_set(cg, 0); } static inline void put_css_set_taskexit(struct css_set *cg) { - kref_put(&cg->ref, release_css_set_taskexit); + __put_css_set(cg, 1); } /* * find_existing_css_set() is a helper for * find_css_set(), and checks to see whether an existing - * css_set is suitable. This currently walks a linked-list for - * simplicity; a later patch will use a hash table for better - * performance + * css_set is suitable. * * oldcg: the cgroup group that we're using before the cgroup * transition @@ -314,7 +357,9 @@ static struct css_set *find_existing_css_set( { int i; struct cgroupfs_root *root = cgrp->root; - struct list_head *l = &init_css_set.list; + struct hlist_head *hhead; + struct hlist_node *node; + struct css_set *cg; /* Built the set of subsystem state objects that we want to * see in the new css_set */ @@ -331,23 +376,29 @@ static struct css_set *find_existing_css_set( } } - /* Look through existing cgroup groups to find one to reuse */ - do { - struct css_set *cg = - list_entry(l, struct css_set, list); - + hhead = css_set_hash(template); + hlist_for_each_entry(cg, node, hhead, hlist) { if (!memcmp(template, cg->subsys, sizeof(cg->subsys))) { /* All subsystems matched */ return cg; } - /* Try the next cgroup group */ - l = l->next; - } while (l != &init_css_set.list); + } /* No existing cgroup group matched */ return NULL; } +static void free_cg_links(struct list_head *tmp) +{ + struct cg_cgroup_link *link; + struct cg_cgroup_link *saved_link; + + list_for_each_entry_safe(link, saved_link, tmp, cgrp_link_list) { + list_del(&link->cgrp_link_list); + kfree(link); + } +} + /* * allocate_cg_links() allocates "count" cg_cgroup_link structures * and chains them on tmp through their cgrp_link_list fields. Returns 0 on @@ -361,13 +412,7 @@ static int allocate_cg_links(int count, struct list_head *tmp) for (i = 0; i < count; i++) { link = kmalloc(sizeof(*link), GFP_KERNEL); if (!link) { - while (!list_empty(tmp)) { - link = list_entry(tmp->next, - struct cg_cgroup_link, - cgrp_link_list); - list_del(&link->cgrp_link_list); - kfree(link); - } + free_cg_links(tmp); return -ENOMEM; } list_add(&link->cgrp_link_list, tmp); @@ -375,16 +420,23 @@ static int allocate_cg_links(int count, struct list_head *tmp) return 0; } -static void free_cg_links(struct list_head *tmp) +/** + * link_css_set - a helper function to link a css_set to a cgroup + * @tmp_cg_links: cg_cgroup_link objects allocated by allocate_cg_links() + * @cg: the css_set to be linked + * @cgrp: the destination cgroup + */ +static void link_css_set(struct list_head *tmp_cg_links, + struct css_set *cg, struct cgroup *cgrp) { - while (!list_empty(tmp)) { - struct cg_cgroup_link *link; - link = list_entry(tmp->next, - struct cg_cgroup_link, - cgrp_link_list); - list_del(&link->cgrp_link_list); - kfree(link); - } + struct cg_cgroup_link *link; + + BUG_ON(list_empty(tmp_cg_links)); + link = list_first_entry(tmp_cg_links, struct cg_cgroup_link, + cgrp_link_list); + link->cg = cg; + list_move(&link->cgrp_link_list, &cgrp->css_sets); + list_add(&link->cg_link_list, &cg->cg_links); } /* @@ -402,15 +454,16 @@ static struct css_set *find_css_set( int i; struct list_head tmp_cg_links; - struct cg_cgroup_link *link; + + struct hlist_head *hhead; /* First see if we already have a cgroup group that matches * the desired set */ - write_lock(&css_set_lock); + read_lock(&css_set_lock); res = find_existing_css_set(oldcg, cgrp, template); if (res) get_css_set(res); - write_unlock(&css_set_lock); + read_unlock(&css_set_lock); if (res) return res; @@ -425,9 +478,10 @@ static struct css_set *find_css_set( return NULL; } - kref_init(&res->ref); + atomic_set(&res->refcount, 1); INIT_LIST_HEAD(&res->cg_links); INIT_LIST_HEAD(&res->tasks); + INIT_HLIST_NODE(&res->hlist); /* Copy the set of subsystem state objects generated in * find_existing_css_set() */ @@ -444,32 +498,20 @@ static struct css_set *find_css_set( * only do it for the first subsystem in each * hierarchy */ - if (ss->root->subsys_list.next == &ss->sibling) { - BUG_ON(list_empty(&tmp_cg_links)); - link = list_entry(tmp_cg_links.next, - struct cg_cgroup_link, - cgrp_link_list); - list_del(&link->cgrp_link_list); - list_add(&link->cgrp_link_list, &cgrp->css_sets); - link->cg = res; - list_add(&link->cg_link_list, &res->cg_links); - } - } - if (list_empty(&rootnode.subsys_list)) { - link = list_entry(tmp_cg_links.next, - struct cg_cgroup_link, - cgrp_link_list); - list_del(&link->cgrp_link_list); - list_add(&link->cgrp_link_list, &dummytop->css_sets); - link->cg = res; - list_add(&link->cg_link_list, &res->cg_links); + if (ss->root->subsys_list.next == &ss->sibling) + link_css_set(&tmp_cg_links, res, cgrp); } + if (list_empty(&rootnode.subsys_list)) + link_css_set(&tmp_cg_links, res, dummytop); BUG_ON(!list_empty(&tmp_cg_links)); - /* Link this cgroup group into the list */ - list_add(&res->list, &init_css_set.list); css_set_count++; + + /* Add this cgroup group to the hash table */ + hhead = css_set_hash(res->subsys); + hlist_add_head(&res->hlist, hhead); + write_unlock(&css_set_lock); return res; @@ -494,10 +536,6 @@ static struct css_set *find_css_set( * knows that the cgroup won't be removed, as cgroup_rmdir() * needs that mutex. * - * The cgroup_common_file_write handler for operations that modify - * the cgroup hierarchy holds cgroup_mutex across the entire operation, - * single threading all such cgroup modifications across the system. - * * The fork and exit callbacks cgroup_fork() and cgroup_exit(), don't * (usually) take cgroup_mutex. These are the two most performance * critical pieces of code here. The exception occurs on cgroup_exit(), @@ -562,18 +600,20 @@ static struct inode_operations cgroup_dir_inode_operations; static struct file_operations proc_cgroupstats_operations; static struct backing_dev_info cgroup_backing_dev_info = { - .capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK, + .capabilities = BDI_CAP_NO_ACCT_AND_WRITEBACK, }; +static int alloc_css_id(struct cgroup_subsys *ss, + struct cgroup *parent, struct cgroup *child); + static struct inode *cgroup_new_inode(mode_t mode, struct super_block *sb) { struct inode *inode = new_inode(sb); if (inode) { inode->i_mode = mode; - inode->i_uid = current->fsuid; - inode->i_gid = current->fsgid; - inode->i_blocks = 0; + inode->i_uid = current_fsuid(); + inode->i_gid = current_fsgid(); inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_mapping->backing_dev_info = &cgroup_backing_dev_info; } @@ -584,13 +624,25 @@ static struct inode *cgroup_new_inode(mode_t mode, struct super_block *sb) * Call subsys's pre_destroy handler. * This is called before css refcnt check. */ -static void cgroup_call_pre_destroy(struct cgroup *cgrp) +static int cgroup_call_pre_destroy(struct cgroup *cgrp) { struct cgroup_subsys *ss; + int ret = 0; + for_each_subsys(cgrp->root, ss) - if (ss->pre_destroy && cgrp->subsys[ss->subsys_id]) - ss->pre_destroy(ss, cgrp); - return; + if (ss->pre_destroy) { + ret = ss->pre_destroy(ss, cgrp); + if (ret) + break; + } + return ret; +} + +static void free_cgroup_rcu(struct rcu_head *obj) +{ + struct cgroup *cgrp = container_of(obj, struct cgroup, rcu_head); + + kfree(cgrp); } static void cgroup_diput(struct dentry *dentry, struct inode *inode) @@ -612,19 +664,19 @@ static void cgroup_diput(struct dentry *dentry, struct inode *inode) /* * Release the subsystem state objects. */ - for_each_subsys(cgrp->root, ss) { - if (cgrp->subsys[ss->subsys_id]) - ss->destroy(ss, cgrp); - } + for_each_subsys(cgrp->root, ss) + ss->destroy(ss, cgrp); cgrp->root->number_of_cgroups--; mutex_unlock(&cgroup_mutex); - /* Drop the active superblock reference that we took when we - * created the cgroup */ + /* + * Drop the active superblock reference that we took when we + * created the cgroup + */ deactivate_super(cgrp->root->sb); - kfree(cgrp); + call_rcu(&cgrp->rcu_head, free_cgroup_rcu); } iput(inode); } @@ -677,6 +729,34 @@ static void cgroup_d_remove_dir(struct dentry *dentry) remove_dir(dentry); } +/* + * A queue for waiters to do rmdir() cgroup. A tasks will sleep when + * cgroup->count == 0 && list_empty(&cgroup->children) && subsys has some + * reference to css->refcnt. In general, this refcnt is expected to goes down + * to zero, soon. + * + * CGRP_WAIT_ON_RMDIR flag is set under cgroup's inode->i_mutex; + */ +DECLARE_WAIT_QUEUE_HEAD(cgroup_rmdir_waitq); + +static void cgroup_wakeup_rmdir_waiter(struct cgroup *cgrp) +{ + if (unlikely(test_and_clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags))) + wake_up_all(&cgroup_rmdir_waitq); +} + +void cgroup_exclude_rmdir(struct cgroup_subsys_state *css) +{ + css_get(css); +} + +void cgroup_release_and_wakeup_rmdir(struct cgroup_subsys_state *css) +{ + cgroup_wakeup_rmdir_waiter(css->cgroup); + css_put(css); +} + + static int rebind_subsystems(struct cgroupfs_root *root, unsigned long final_bits) { @@ -702,7 +782,7 @@ static int rebind_subsystems(struct cgroupfs_root *root, * any child cgroups exist. This is theoretically supportable * but involves complex error handling, so it's being left until * later */ - if (!list_empty(&cgrp->children)) + if (root->number_of_cgroups > 1) return -EBUSY; /* Process each subsystem */ @@ -714,23 +794,26 @@ static int rebind_subsystems(struct cgroupfs_root *root, BUG_ON(cgrp->subsys[i]); BUG_ON(!dummytop->subsys[i]); BUG_ON(dummytop->subsys[i]->cgroup != dummytop); + mutex_lock(&ss->hierarchy_mutex); cgrp->subsys[i] = dummytop->subsys[i]; cgrp->subsys[i]->cgroup = cgrp; - list_add(&ss->sibling, &root->subsys_list); - rcu_assign_pointer(ss->root, root); + list_move(&ss->sibling, &root->subsys_list); + ss->root = root; if (ss->bind) ss->bind(ss, cgrp); - + mutex_unlock(&ss->hierarchy_mutex); } else if (bit & removed_bits) { /* We're removing this subsystem */ BUG_ON(cgrp->subsys[i] != dummytop->subsys[i]); BUG_ON(cgrp->subsys[i]->cgroup != cgrp); + mutex_lock(&ss->hierarchy_mutex); if (ss->bind) ss->bind(ss, dummytop); dummytop->subsys[i]->cgroup = dummytop; cgrp->subsys[i] = NULL; - rcu_assign_pointer(subsys[i]->root, &rootnode); - list_del(&ss->sibling); + subsys[i]->root = &rootnode; + list_move(&ss->sibling, &rootnode.subsys_list); + mutex_unlock(&ss->hierarchy_mutex); } else if (bit & final_bits) { /* Subsystem state should already exist */ BUG_ON(!cgrp->subsys[i]); @@ -773,6 +856,11 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) { char *token, *o = data ?: "all"; + unsigned long mask = (unsigned long)-1; + +#ifdef CONFIG_CPUSETS + mask = ~(1UL << cpuset_subsys_id); +#endif opts->subsys_bits = 0; opts->flags = 0; @@ -817,6 +905,15 @@ static int parse_cgroupfs_options(char *data, } } + /* + * Option noprefix was introduced just for backward compatibility + * with the old cpuset, so we allow noprefix only if mounting just + * the cpuset subsystem. + */ + if (test_bit(ROOT_NOPREFIX, &opts->flags) && + (opts->subsys_bits & mask)) + return -EINVAL; + /* We can't have an empty hierarchy */ if (!opts->subsys_bits) return -EINVAL; @@ -831,6 +928,7 @@ static int cgroup_remount(struct super_block *sb, int *flags, char *data) struct cgroup *cgrp = &root->top_cgroup; struct cgroup_sb_opts opts; + lock_kernel(); mutex_lock(&cgrp->dentry->d_inode->i_mutex); mutex_lock(&cgroup_mutex); @@ -846,18 +944,19 @@ static int cgroup_remount(struct super_block *sb, int *flags, char *data) } ret = rebind_subsystems(root, opts.subsys_bits); + if (ret) + goto out_unlock; /* (re)populate subsystem files */ - if (!ret) - cgroup_populate_dir(cgrp); + cgroup_populate_dir(cgrp); if (opts.release_agent) strcpy(root->release_agent_path, opts.release_agent); out_unlock: - if (opts.release_agent) - kfree(opts.release_agent); + kfree(opts.release_agent); mutex_unlock(&cgroup_mutex); mutex_unlock(&cgrp->dentry->d_inode->i_mutex); + unlock_kernel(); return ret; } @@ -868,6 +967,15 @@ static struct super_operations cgroup_ops = { .remount_fs = cgroup_remount, }; +static void init_cgroup_housekeeping(struct cgroup *cgrp) +{ + INIT_LIST_HEAD(&cgrp->sibling); + INIT_LIST_HEAD(&cgrp->children); + INIT_LIST_HEAD(&cgrp->css_sets); + INIT_LIST_HEAD(&cgrp->release_list); + INIT_LIST_HEAD(&cgrp->pids_list); + init_rwsem(&cgrp->pids_mutex); +} static void init_cgroup_root(struct cgroupfs_root *root) { struct cgroup *cgrp = &root->top_cgroup; @@ -876,10 +984,7 @@ static void init_cgroup_root(struct cgroupfs_root *root) root->number_of_cgroups = 1; cgrp->root = root; cgrp->top_cgroup = cgrp; - INIT_LIST_HEAD(&cgrp->sibling); - INIT_LIST_HEAD(&cgrp->children); - INIT_LIST_HEAD(&cgrp->css_sets); - INIT_LIST_HEAD(&cgrp->release_list); + init_cgroup_housekeeping(cgrp); } static int cgroup_test_super(struct super_block *sb, void *data) @@ -948,21 +1053,18 @@ static int cgroup_get_sb(struct file_system_type *fs_type, int ret = 0; struct super_block *sb; struct cgroupfs_root *root; - struct list_head tmp_cg_links, *l; - INIT_LIST_HEAD(&tmp_cg_links); + struct list_head tmp_cg_links; /* First find the desired set of subsystems */ ret = parse_cgroupfs_options(data, &opts); if (ret) { - if (opts.release_agent) - kfree(opts.release_agent); + kfree(opts.release_agent); return ret; } root = kzalloc(sizeof(*root), GFP_KERNEL); if (!root) { - if (opts.release_agent) - kfree(opts.release_agent); + kfree(opts.release_agent); return -ENOMEM; } @@ -988,8 +1090,9 @@ static int cgroup_get_sb(struct file_system_type *fs_type, root = NULL; } else { /* New superblock */ - struct cgroup *cgrp = &root->top_cgroup; + struct cgroup *root_cgrp = &root->top_cgroup; struct inode *inode; + int i; BUG_ON(sb->s_root != NULL); @@ -1019,7 +1122,7 @@ static int cgroup_get_sb(struct file_system_type *fs_type, if (ret == -EBUSY) { mutex_unlock(&cgroup_mutex); mutex_unlock(&inode->i_mutex); - goto drop_new_super; + goto free_cg_links; } /* EBUSY should be the only error here */ @@ -1028,47 +1131,40 @@ static int cgroup_get_sb(struct file_system_type *fs_type, list_add(&root->root_list, &roots); root_count++; - sb->s_root->d_fsdata = &root->top_cgroup; + sb->s_root->d_fsdata = root_cgrp; root->top_cgroup.dentry = sb->s_root; /* Link the top cgroup in this hierarchy into all * the css_set objects */ write_lock(&css_set_lock); - l = &init_css_set.list; - do { + for (i = 0; i < CSS_SET_TABLE_SIZE; i++) { + struct hlist_head *hhead = &css_set_table[i]; + struct hlist_node *node; struct css_set *cg; - struct cg_cgroup_link *link; - cg = list_entry(l, struct css_set, list); - BUG_ON(list_empty(&tmp_cg_links)); - link = list_entry(tmp_cg_links.next, - struct cg_cgroup_link, - cgrp_link_list); - list_del(&link->cgrp_link_list); - link->cg = cg; - list_add(&link->cgrp_link_list, - &root->top_cgroup.css_sets); - list_add(&link->cg_link_list, &cg->cg_links); - l = l->next; - } while (l != &init_css_set.list); + + hlist_for_each_entry(cg, node, hhead, hlist) + link_css_set(&tmp_cg_links, cg, root_cgrp); + } write_unlock(&css_set_lock); free_cg_links(&tmp_cg_links); - BUG_ON(!list_empty(&cgrp->sibling)); - BUG_ON(!list_empty(&cgrp->children)); + BUG_ON(!list_empty(&root_cgrp->sibling)); + BUG_ON(!list_empty(&root_cgrp->children)); BUG_ON(root->number_of_cgroups != 1); - cgroup_populate_dir(cgrp); + cgroup_populate_dir(root_cgrp); mutex_unlock(&inode->i_mutex); mutex_unlock(&cgroup_mutex); } - return simple_set_mnt(mnt, sb); + simple_set_mnt(mnt, sb); + return 0; - drop_new_super: - up_write(&sb->s_umount); - deactivate_super(sb); + free_cg_links: free_cg_links(&tmp_cg_links); + drop_new_super: + deactivate_locked_super(sb); return ret; } @@ -1076,6 +1172,8 @@ static void cgroup_kill_sb(struct super_block *sb) { struct cgroupfs_root *root = sb->s_fs_info; struct cgroup *cgrp = &root->top_cgroup; int ret; + struct cg_cgroup_link *link; + struct cg_cgroup_link *saved_link; BUG_ON(!root); @@ -1095,10 +1193,9 @@ static void cgroup_kill_sb(struct super_block *sb) { * root cgroup */ write_lock(&css_set_lock); - while (!list_empty(&cgrp->css_sets)) { - struct cg_cgroup_link *link; - link = list_entry(cgrp->css_sets.next, - struct cg_cgroup_link, cgrp_link_list); + + list_for_each_entry_safe(link, saved_link, &cgrp->css_sets, + cgrp_link_list) { list_del(&link->cg_link_list); list_del(&link->cgrp_link_list); kfree(link); @@ -1109,10 +1206,11 @@ static void cgroup_kill_sb(struct super_block *sb) { list_del(&root->root_list); root_count--; } + mutex_unlock(&cgroup_mutex); - kfree(root); kill_litter_super(sb); + kfree(root); } static struct file_system_type cgroup_fs_type = { @@ -1137,14 +1235,16 @@ static inline struct cftype *__d_cft(struct dentry *dentry) * @buf: the buffer to write the path into * @buflen: the length of the buffer * - * Called with cgroup_mutex held. Writes path of cgroup into buf. - * Returns 0 on success, -errno on error. + * Called with cgroup_mutex held or else with an RCU-protected cgroup + * reference. Writes path of cgroup into buf. Returns 0 on success, + * -errno on error. */ int cgroup_path(const struct cgroup *cgrp, char *buf, int buflen) { char *start; + struct dentry *dentry = rcu_dereference(cgrp->dentry); - if (cgrp == dummytop) { + if (!dentry || cgrp == dummytop) { /* * Inactive subsystems have no dentry for their root * cgroup @@ -1157,13 +1257,14 @@ int cgroup_path(const struct cgroup *cgrp, char *buf, int buflen) *--start = '\0'; for (;;) { - int len = cgrp->dentry->d_name.len; + int len = dentry->d_name.len; if ((start -= len) < buf) return -ENAMETOOLONG; memcpy(start, cgrp->dentry->d_name.name, len); cgrp = cgrp->parent; if (!cgrp) break; + dentry = rcu_dereference(cgrp->dentry); if (!cgrp->parent) continue; if (--start < buf) @@ -1208,7 +1309,7 @@ int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk) int retval = 0; struct cgroup_subsys *ss; struct cgroup *oldcgrp; - struct css_set *cg = tsk->cgroups; + struct css_set *cg; struct css_set *newcg; struct cgroupfs_root *root = cgrp->root; int subsys_id; @@ -1228,11 +1329,16 @@ int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk) } } + task_lock(tsk); + cg = tsk->cgroups; + get_css_set(cg); + task_unlock(tsk); /* * Locate or allocate a new css_set for this task, * based on its final set of cgroups */ newcg = find_css_set(cg, cgrp); + put_css_set(cg); if (!newcg) return -ENOMEM; @@ -1260,22 +1366,25 @@ int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk) set_bit(CGRP_RELEASABLE, &oldcgrp->flags); synchronize_rcu(); put_css_set(cg); + + /* + * wake up rmdir() waiter. the rmdir should fail since the cgroup + * is no longer empty. + */ + cgroup_wakeup_rmdir_waiter(cgrp); return 0; } /* - * Attach task with pid 'pid' to cgroup 'cgrp'. Call with - * cgroup_mutex, may take task_lock of task + * Attach task with pid 'pid' to cgroup 'cgrp'. Call with cgroup_mutex + * held. May take task_lock of task */ -static int attach_task_by_pid(struct cgroup *cgrp, char *pidbuf) +static int attach_task_by_pid(struct cgroup *cgrp, u64 pid) { - pid_t pid; struct task_struct *tsk; + const struct cred *cred = current_cred(), *tcred; int ret; - if (sscanf(pidbuf, "%d", &pid) != 1) - return -EIO; - if (pid) { rcu_read_lock(); tsk = find_task_by_vpid(pid); @@ -1283,14 +1392,16 @@ static int attach_task_by_pid(struct cgroup *cgrp, char *pidbuf) rcu_read_unlock(); return -ESRCH; } - get_task_struct(tsk); - rcu_read_unlock(); - if ((current->euid) && (current->euid != tsk->uid) - && (current->euid != tsk->suid)) { - put_task_struct(tsk); + tcred = __task_cred(tsk); + if (cred->euid && + cred->euid != tcred->uid && + cred->euid != tcred->suid) { + rcu_read_unlock(); return -EACCES; } + get_task_struct(tsk); + rcu_read_unlock(); } else { tsk = current; get_task_struct(tsk); @@ -1301,24 +1412,74 @@ static int attach_task_by_pid(struct cgroup *cgrp, char *pidbuf) return ret; } +static int cgroup_tasks_write(struct cgroup *cgrp, struct cftype *cft, u64 pid) +{ + int ret; + if (!cgroup_lock_live_group(cgrp)) + return -ENODEV; + ret = attach_task_by_pid(cgrp, pid); + cgroup_unlock(); + return ret; +} + /* The various types of files and directories in a cgroup file system */ enum cgroup_filetype { FILE_ROOT, FILE_DIR, FILE_TASKLIST, FILE_NOTIFY_ON_RELEASE, - FILE_RELEASABLE, FILE_RELEASE_AGENT, }; -static ssize_t cgroup_write_uint(struct cgroup *cgrp, struct cftype *cft, - struct file *file, - const char __user *userbuf, - size_t nbytes, loff_t *unused_ppos) +/** + * cgroup_lock_live_group - take cgroup_mutex and check that cgrp is alive. + * @cgrp: the cgroup to be checked for liveness + * + * On success, returns true; the lock should be later released with + * cgroup_unlock(). On failure returns false with no lock held. + */ +bool cgroup_lock_live_group(struct cgroup *cgrp) +{ + mutex_lock(&cgroup_mutex); + if (cgroup_is_removed(cgrp)) { + mutex_unlock(&cgroup_mutex); + return false; + } + return true; +} + +static int cgroup_release_agent_write(struct cgroup *cgrp, struct cftype *cft, + const char *buffer) +{ + BUILD_BUG_ON(sizeof(cgrp->root->release_agent_path) < PATH_MAX); + if (!cgroup_lock_live_group(cgrp)) + return -ENODEV; + strcpy(cgrp->root->release_agent_path, buffer); + cgroup_unlock(); + return 0; +} + +static int cgroup_release_agent_show(struct cgroup *cgrp, struct cftype *cft, + struct seq_file *seq) { - char buffer[64]; + if (!cgroup_lock_live_group(cgrp)) + return -ENODEV; + seq_puts(seq, cgrp->root->release_agent_path); + seq_putc(seq, '\n'); + cgroup_unlock(); + return 0; +} + +/* A buffer size big enough for numbers or short strings */ +#define CGROUP_LOCAL_BUFFER_SIZE 64 + +static ssize_t cgroup_write_X64(struct cgroup *cgrp, struct cftype *cft, + struct file *file, + const char __user *userbuf, + size_t nbytes, loff_t *unused_ppos) +{ + char buffer[CGROUP_LOCAL_BUFFER_SIZE]; int retval = 0; - u64 val; char *end; if (!nbytes) @@ -1329,83 +1490,56 @@ static ssize_t cgroup_write_uint(struct cgroup *cgrp, struct cftype *cft, return -EFAULT; buffer[nbytes] = 0; /* nul-terminate */ - - /* strip newline if necessary */ - if (nbytes && (buffer[nbytes-1] == '\n')) - buffer[nbytes-1] = 0; - val = simple_strtoull(buffer, &end, 0); - if (*end) - return -EINVAL; - - /* Pass to subsystem */ - retval = cft->write_uint(cgrp, cft, val); + strstrip(buffer); + if (cft->write_u64) { + u64 val = simple_strtoull(buffer, &end, 0); + if (*end) + return -EINVAL; + retval = cft->write_u64(cgrp, cft, val); + } else { + s64 val = simple_strtoll(buffer, &end, 0); + if (*end) + return -EINVAL; + retval = cft->write_s64(cgrp, cft, val); + } if (!retval) retval = nbytes; return retval; } -static ssize_t cgroup_common_file_write(struct cgroup *cgrp, - struct cftype *cft, - struct file *file, - const char __user *userbuf, - size_t nbytes, loff_t *unused_ppos) +static ssize_t cgroup_write_string(struct cgroup *cgrp, struct cftype *cft, + struct file *file, + const char __user *userbuf, + size_t nbytes, loff_t *unused_ppos) { - enum cgroup_filetype type = cft->private; - char *buffer; + char local_buffer[CGROUP_LOCAL_BUFFER_SIZE]; int retval = 0; + size_t max_bytes = cft->max_write_len; + char *buffer = local_buffer; - if (nbytes >= PATH_MAX) + if (!max_bytes) + max_bytes = sizeof(local_buffer) - 1; + if (nbytes >= max_bytes) return -E2BIG; - - /* +1 for nul-terminator */ - buffer = kmalloc(nbytes + 1, GFP_KERNEL); - if (buffer == NULL) - return -ENOMEM; - - if (copy_from_user(buffer, userbuf, nbytes)) { + /* Allocate a dynamic buffer if we need one */ + if (nbytes >= sizeof(local_buffer)) { + buffer = kmalloc(nbytes + 1, GFP_KERNEL); + if (buffer == NULL) + return -ENOMEM; + } + if (nbytes && copy_from_user(buffer, userbuf, nbytes)) { retval = -EFAULT; - goto out1; + goto out; } - buffer[nbytes] = 0; /* nul-terminate */ - strstrip(buffer); /* strip -just- trailing whitespace */ - mutex_lock(&cgroup_mutex); - - /* - * This was already checked for in cgroup_file_write(), but - * check again now we're holding cgroup_mutex. - */ - if (cgroup_is_removed(cgrp)) { - retval = -ENODEV; - goto out2; - } - - switch (type) { - case FILE_TASKLIST: - retval = attach_task_by_pid(cgrp, buffer); - break; - case FILE_NOTIFY_ON_RELEASE: - clear_bit(CGRP_RELEASABLE, &cgrp->flags); - if (simple_strtoul(buffer, NULL, 10) != 0) - set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); - else - clear_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); - break; - case FILE_RELEASE_AGENT: - BUILD_BUG_ON(sizeof(cgrp->root->release_agent_path) < PATH_MAX); - strcpy(cgrp->root->release_agent_path, buffer); - break; - default: - retval = -EINVAL; - goto out2; - } - - if (retval == 0) + buffer[nbytes] = 0; /* nul-terminate */ + strstrip(buffer); + retval = cft->write_string(cgrp, cft, buffer); + if (!retval) retval = nbytes; -out2: - mutex_unlock(&cgroup_mutex); -out1: - kfree(buffer); +out: + if (buffer != local_buffer) + kfree(buffer); return retval; } @@ -1415,68 +1549,43 @@ static ssize_t cgroup_file_write(struct file *file, const char __user *buf, struct cftype *cft = __d_cft(file->f_dentry); struct cgroup *cgrp = __d_cgrp(file->f_dentry->d_parent); - if (!cft || cgroup_is_removed(cgrp)) + if (cgroup_is_removed(cgrp)) return -ENODEV; if (cft->write) return cft->write(cgrp, cft, file, buf, nbytes, ppos); - if (cft->write_uint) - return cgroup_write_uint(cgrp, cft, file, buf, nbytes, ppos); + if (cft->write_u64 || cft->write_s64) + return cgroup_write_X64(cgrp, cft, file, buf, nbytes, ppos); + if (cft->write_string) + return cgroup_write_string(cgrp, cft, file, buf, nbytes, ppos); + if (cft->trigger) { + int ret = cft->trigger(cgrp, (unsigned int)cft->private); + return ret ? ret : nbytes; + } return -EINVAL; } -static ssize_t cgroup_read_uint(struct cgroup *cgrp, struct cftype *cft, - struct file *file, - char __user *buf, size_t nbytes, - loff_t *ppos) +static ssize_t cgroup_read_u64(struct cgroup *cgrp, struct cftype *cft, + struct file *file, + char __user *buf, size_t nbytes, + loff_t *ppos) { - char tmp[64]; - u64 val = cft->read_uint(cgrp, cft); + char tmp[CGROUP_LOCAL_BUFFER_SIZE]; + u64 val = cft->read_u64(cgrp, cft); int len = sprintf(tmp, "%llu\n", (unsigned long long) val); return simple_read_from_buffer(buf, nbytes, ppos, tmp, len); } -static ssize_t cgroup_common_file_read(struct cgroup *cgrp, - struct cftype *cft, - struct file *file, - char __user *buf, - size_t nbytes, loff_t *ppos) +static ssize_t cgroup_read_s64(struct cgroup *cgrp, struct cftype *cft, + struct file *file, + char __user *buf, size_t nbytes, + loff_t *ppos) { - enum cgroup_filetype type = cft->private; - char *page; - ssize_t retval = 0; - char *s; - - if (!(page = (char *)__get_free_page(GFP_KERNEL))) - return -ENOMEM; - - s = page; - - switch (type) { - case FILE_RELEASE_AGENT: - { - struct cgroupfs_root *root; - size_t n; - mutex_lock(&cgroup_mutex); - root = cgrp->root; - n = strnlen(root->release_agent_path, - sizeof(root->release_agent_path)); - n = min(n, (size_t) PAGE_SIZE); - strncpy(s, root->release_agent_path, n); - mutex_unlock(&cgroup_mutex); - s += n; - break; - } - default: - retval = -EINVAL; - goto out; - } - *s++ = '\n'; + char tmp[CGROUP_LOCAL_BUFFER_SIZE]; + s64 val = cft->read_s64(cgrp, cft); + int len = sprintf(tmp, "%lld\n", (long long) val); - retval = simple_read_from_buffer(buf, nbytes, ppos, page, s - page); -out: - free_page((unsigned long)page); - return retval; + return simple_read_from_buffer(buf, nbytes, ppos, tmp, len); } static ssize_t cgroup_file_read(struct file *file, char __user *buf, @@ -1485,16 +1594,62 @@ static ssize_t cgroup_file_read(struct file *file, char __user *buf, struct cftype *cft = __d_cft(file->f_dentry); struct cgroup *cgrp = __d_cgrp(file->f_dentry->d_parent); - if (!cft || cgroup_is_removed(cgrp)) + if (cgroup_is_removed(cgrp)) return -ENODEV; if (cft->read) return cft->read(cgrp, cft, file, buf, nbytes, ppos); - if (cft->read_uint) - return cgroup_read_uint(cgrp, cft, file, buf, nbytes, ppos); + if (cft->read_u64) + return cgroup_read_u64(cgrp, cft, file, buf, nbytes, ppos); + if (cft->read_s64) + return cgroup_read_s64(cgrp, cft, file, buf, nbytes, ppos); return -EINVAL; } +/* + * seqfile ops/methods for returning structured data. Currently just + * supports string->u64 maps, but can be extended in future. + */ + +struct cgroup_seqfile_state { + struct cftype *cft; + struct cgroup *cgroup; +}; + +static int cgroup_map_add(struct cgroup_map_cb *cb, const char *key, u64 value) +{ + struct seq_file *sf = cb->state; + return seq_printf(sf, "%s %llu\n", key, (unsigned long long)value); +} + +static int cgroup_seqfile_show(struct seq_file *m, void *arg) +{ + struct cgroup_seqfile_state *state = m->private; + struct cftype *cft = state->cft; + if (cft->read_map) { + struct cgroup_map_cb cb = { + .fill = cgroup_map_add, + .state = m, + }; + return cft->read_map(state->cgroup, cft, &cb); + } + return cft->read_seq_string(state->cgroup, cft, m); +} + +static int cgroup_seqfile_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq = file->private_data; + kfree(seq->private); + return single_release(inode, file); +} + +static struct file_operations cgroup_seqfile_operations = { + .read = seq_read, + .write = cgroup_file_write, + .llseek = seq_lseek, + .release = cgroup_seqfile_release, +}; + static int cgroup_file_open(struct inode *inode, struct file *file) { int err; @@ -1503,11 +1658,20 @@ static int cgroup_file_open(struct inode *inode, struct file *file) err = generic_file_open(inode, file); if (err) return err; - cft = __d_cft(file->f_dentry); - if (!cft) - return -ENODEV; - if (cft->open) + + if (cft->read_map || cft->read_seq_string) { + struct cgroup_seqfile_state *state = + kzalloc(sizeof(*state), GFP_USER); + if (!state) + return -ENOMEM; + state->cft = cft; + state->cgroup = __d_cgrp(file->f_dentry->d_parent); + file->f_op = &cgroup_seqfile_operations; + err = single_open(file, cgroup_seqfile_show, state); + if (err < 0) + kfree(state); + } else if (cft->open) err = cft->open(inode, file); else err = 0; @@ -1553,10 +1717,10 @@ static struct inode_operations cgroup_dir_inode_operations = { .rename = cgroup_rename, }; -static int cgroup_create_file(struct dentry *dentry, int mode, +static int cgroup_create_file(struct dentry *dentry, mode_t mode, struct super_block *sb) { - static struct dentry_operations cgroup_dops = { + static const struct dentry_operations cgroup_dops = { .d_iput = cgroup_diput, }; @@ -1599,7 +1763,7 @@ static int cgroup_create_file(struct dentry *dentry, int mode, * @mode: mode to set on new directory. */ static int cgroup_create_dir(struct cgroup *cgrp, struct dentry *dentry, - int mode) + mode_t mode) { struct dentry *parent; int error = 0; @@ -1609,7 +1773,7 @@ static int cgroup_create_dir(struct cgroup *cgrp, struct dentry *dentry, if (!error) { dentry->d_fsdata = cgrp; inc_nlink(parent->d_inode); - cgrp->dentry = dentry; + rcu_assign_pointer(cgrp->dentry, dentry); dget(dentry); } dput(dentry); @@ -1617,13 +1781,41 @@ static int cgroup_create_dir(struct cgroup *cgrp, struct dentry *dentry, return error; } -int cgroup_add_file(struct cgroup *cgrp, - struct cgroup_subsys *subsys, +/** + * cgroup_file_mode - deduce file mode of a control file + * @cft: the control file in question + * + * returns cft->mode if ->mode is not 0 + * returns S_IRUGO|S_IWUSR if it has both a read and a write handler + * returns S_IRUGO if it has only a read handler + * returns S_IWUSR if it has only a write hander + */ +static mode_t cgroup_file_mode(const struct cftype *cft) +{ + mode_t mode = 0; + + if (cft->mode) + return cft->mode; + + if (cft->read || cft->read_u64 || cft->read_s64 || + cft->read_map || cft->read_seq_string) + mode |= S_IRUGO; + + if (cft->write || cft->write_u64 || cft->write_s64 || + cft->write_string || cft->trigger) + mode |= S_IWUSR; + + return mode; +} + +int cgroup_add_file(struct cgroup *cgrp, + struct cgroup_subsys *subsys, const struct cftype *cft) { struct dentry *dir = cgrp->dentry; struct dentry *dentry; int error; + mode_t mode; char name[MAX_CGROUP_TYPE_NAMELEN + MAX_CFTYPE_NAME + 2] = { 0 }; if (subsys && !test_bit(ROOT_NOPREFIX, &cgrp->root->flags)) { @@ -1634,7 +1826,8 @@ int cgroup_add_file(struct cgroup *cgrp, BUG_ON(!mutex_is_locked(&dir->d_inode->i_mutex)); dentry = lookup_one_len(name, dir, strlen(name)); if (!IS_ERR(dentry)) { - error = cgroup_create_file(dentry, 0644 | S_IFREG, + mode = cgroup_file_mode(cft); + error = cgroup_create_file(dentry, mode | S_IFREG, cgrp->root->sb); if (!error) dentry->d_fsdata = (void *)cft; @@ -1667,15 +1860,11 @@ int cgroup_add_files(struct cgroup *cgrp, int cgroup_task_count(const struct cgroup *cgrp) { int count = 0; - struct list_head *l; + struct cg_cgroup_link *link; read_lock(&css_set_lock); - l = cgrp->css_sets.next; - while (l != &cgrp->css_sets) { - struct cg_cgroup_link *link = - list_entry(l, struct cg_cgroup_link, cgrp_link_list); - count += atomic_read(&link->cg->ref.refcount); - l = l->next; + list_for_each_entry(link, &cgrp->css_sets, cgrp_link_list) { + count += atomic_read(&link->cg->refcount); } read_unlock(&css_set_lock); return count; @@ -1754,6 +1943,7 @@ struct task_struct *cgroup_iter_next(struct cgroup *cgrp, { struct task_struct *res; struct list_head *l = it->task; + struct cg_cgroup_link *link; /* If the iterator cg is NULL, we have no tasks */ if (!it->cg_link) @@ -1761,7 +1951,8 @@ struct task_struct *cgroup_iter_next(struct cgroup *cgrp, res = list_entry(l, struct task_struct, cg_list); /* Advance iterator to find next entry */ l = l->next; - if (l == &res->cgroups->tasks) { + link = list_entry(it->cg_link, struct cg_cgroup_link, cgrp_link_list); + if (l == &link->cg->tasks) { /* We reached the end of this task list - move on to * the next cg_cgroup_link */ cgroup_advance_iter(cgrp, it); @@ -1944,16 +2135,7 @@ int cgroup_scan_tasks(struct cgroup_scanner *scan) * but we cannot guarantee that the information we produce is correct * unless we produce it entirely atomically. * - * Upon tasks file open(), a struct ctr_struct is allocated, that - * will have a pointer to an array (also allocated here). The struct - * ctr_struct * is stored in file->private_data. Its resources will - * be freed by release() when the file is closed. The array is used - * to sprintf the PIDs and then used by read(). */ -struct ctr_struct { - char *buf; - int bufsz; -}; /* * Load into 'pidarray' up to 'npids' of the tasks using cgroup @@ -1964,14 +2146,16 @@ struct ctr_struct { */ static int pid_array_load(pid_t *pidarray, int npids, struct cgroup *cgrp) { - int n = 0; + int n = 0, pid; struct cgroup_iter it; struct task_struct *tsk; cgroup_iter_start(cgrp, &it); while ((tsk = cgroup_iter_next(cgrp, &it))) { if (unlikely(n == npids)) break; - pidarray[n++] = task_pid_vnr(tsk); + pid = task_pid_vnr(tsk); + if (pid > 0) + pidarray[n++] = pid; } cgroup_iter_end(cgrp, &it); return n; @@ -1992,15 +2176,17 @@ int cgroupstats_build(struct cgroupstats *stats, struct dentry *dentry) struct cgroup *cgrp; struct cgroup_iter it; struct task_struct *tsk; + /* - * Validate dentry by checking the superblock operations + * Validate dentry by checking the superblock operations, + * and make sure it's a directory. */ - if (dentry->d_sb->s_op != &cgroup_ops) + if (dentry->d_sb->s_op != &cgroup_ops || + !S_ISDIR(dentry->d_inode->i_mode)) goto err; ret = 0; cgrp = dentry->d_fsdata; - rcu_read_lock(); cgroup_iter_start(cgrp, &it); while ((tsk = cgroup_iter_next(cgrp, &it))) { @@ -2025,52 +2211,170 @@ int cgroupstats_build(struct cgroupstats *stats, struct dentry *dentry) } cgroup_iter_end(cgrp, &it); - rcu_read_unlock(); err: return ret; } +/* + * Cache pids for all threads in the same pid namespace that are + * opening the same "tasks" file. + */ +struct cgroup_pids { + /* The node in cgrp->pids_list */ + struct list_head list; + /* The cgroup those pids belong to */ + struct cgroup *cgrp; + /* The namepsace those pids belong to */ + struct pid_namespace *ns; + /* Array of process ids in the cgroup */ + pid_t *tasks_pids; + /* How many files are using the this tasks_pids array */ + int use_count; + /* Length of the current tasks_pids array */ + int length; +}; + static int cmppid(const void *a, const void *b) { return *(pid_t *)a - *(pid_t *)b; } /* - * Convert array 'a' of 'npids' pid_t's to a string of newline separated - * decimal pids in 'buf'. Don't write more than 'sz' chars, but return - * count 'cnt' of how many chars would be written if buf were large enough. + * seq_file methods for the "tasks" file. The seq_file position is the + * next pid to display; the seq_file iterator is a pointer to the pid + * in the cgroup->tasks_pids array. */ -static int pid_array_to_buf(char *buf, int sz, pid_t *a, int npids) + +static void *cgroup_tasks_start(struct seq_file *s, loff_t *pos) { - int cnt = 0; - int i; + /* + * Initially we receive a position value that corresponds to + * one more than the last pid shown (or 0 on the first call or + * after a seek to the start). Use a binary-search to find the + * next pid to display, if any + */ + struct cgroup_pids *cp = s->private; + struct cgroup *cgrp = cp->cgrp; + int index = 0, pid = *pos; + int *iter; + + down_read(&cgrp->pids_mutex); + if (pid) { + int end = cp->length; + + while (index < end) { + int mid = (index + end) / 2; + if (cp->tasks_pids[mid] == pid) { + index = mid; + break; + } else if (cp->tasks_pids[mid] <= pid) + index = mid + 1; + else + end = mid; + } + } + /* If we're off the end of the array, we're done */ + if (index >= cp->length) + return NULL; + /* Update the abstract position to be the actual pid that we found */ + iter = cp->tasks_pids + index; + *pos = *iter; + return iter; +} + +static void cgroup_tasks_stop(struct seq_file *s, void *v) +{ + struct cgroup_pids *cp = s->private; + struct cgroup *cgrp = cp->cgrp; + up_read(&cgrp->pids_mutex); +} + +static void *cgroup_tasks_next(struct seq_file *s, void *v, loff_t *pos) +{ + struct cgroup_pids *cp = s->private; + int *p = v; + int *end = cp->tasks_pids + cp->length; + + /* + * Advance to the next pid in the array. If this goes off the + * end, we're done + */ + p++; + if (p >= end) { + return NULL; + } else { + *pos = *p; + return p; + } +} + +static int cgroup_tasks_show(struct seq_file *s, void *v) +{ + return seq_printf(s, "%d\n", *(int *)v); +} + +static struct seq_operations cgroup_tasks_seq_operations = { + .start = cgroup_tasks_start, + .stop = cgroup_tasks_stop, + .next = cgroup_tasks_next, + .show = cgroup_tasks_show, +}; - for (i = 0; i < npids; i++) - cnt += snprintf(buf + cnt, max(sz - cnt, 0), "%d\n", a[i]); - return cnt; +static void release_cgroup_pid_array(struct cgroup_pids *cp) +{ + struct cgroup *cgrp = cp->cgrp; + + down_write(&cgrp->pids_mutex); + BUG_ON(!cp->use_count); + if (!--cp->use_count) { + list_del(&cp->list); + put_pid_ns(cp->ns); + kfree(cp->tasks_pids); + kfree(cp); + } + up_write(&cgrp->pids_mutex); +} + +static int cgroup_tasks_release(struct inode *inode, struct file *file) +{ + struct seq_file *seq; + struct cgroup_pids *cp; + + if (!(file->f_mode & FMODE_READ)) + return 0; + + seq = file->private_data; + cp = seq->private; + + release_cgroup_pid_array(cp); + return seq_release(inode, file); } +static struct file_operations cgroup_tasks_operations = { + .read = seq_read, + .llseek = seq_lseek, + .write = cgroup_file_write, + .release = cgroup_tasks_release, +}; + /* - * Handle an open on 'tasks' file. Prepare a buffer listing the + * Handle an open on 'tasks' file. Prepare an array containing the * process id's of tasks currently attached to the cgroup being opened. - * - * Does not require any specific cgroup mutexes, and does not take any. */ + static int cgroup_tasks_open(struct inode *unused, struct file *file) { struct cgroup *cgrp = __d_cgrp(file->f_dentry->d_parent); - struct ctr_struct *ctr; + struct pid_namespace *ns = current->nsproxy->pid_ns; + struct cgroup_pids *cp; pid_t *pidarray; int npids; - char c; + int retval; + /* Nothing to do for write-only files */ if (!(file->f_mode & FMODE_READ)) return 0; - ctr = kmalloc(sizeof(*ctr), GFP_KERNEL); - if (!ctr) - goto err0; - /* * If cgroup gets more users after we read count, we won't have * enough space - tough. This race is indistinguishable to the @@ -2078,57 +2382,48 @@ static int cgroup_tasks_open(struct inode *unused, struct file *file) * show up until sometime later on. */ npids = cgroup_task_count(cgrp); - if (npids) { - pidarray = kmalloc(npids * sizeof(pid_t), GFP_KERNEL); - if (!pidarray) - goto err1; - - npids = pid_array_load(pidarray, npids, cgrp); - sort(pidarray, npids, sizeof(pid_t), cmppid, NULL); + pidarray = kmalloc(npids * sizeof(pid_t), GFP_KERNEL); + if (!pidarray) + return -ENOMEM; + npids = pid_array_load(pidarray, npids, cgrp); + sort(pidarray, npids, sizeof(pid_t), cmppid, NULL); - /* Call pid_array_to_buf() twice, first just to get bufsz */ - ctr->bufsz = pid_array_to_buf(&c, sizeof(c), pidarray, npids) + 1; - ctr->buf = kmalloc(ctr->bufsz, GFP_KERNEL); - if (!ctr->buf) - goto err2; - ctr->bufsz = pid_array_to_buf(ctr->buf, ctr->bufsz, pidarray, npids); + /* + * Store the array in the cgroup, freeing the old + * array if necessary + */ + down_write(&cgrp->pids_mutex); - kfree(pidarray); - } else { - ctr->buf = NULL; - ctr->bufsz = 0; + list_for_each_entry(cp, &cgrp->pids_list, list) { + if (ns == cp->ns) + goto found; } - file->private_data = ctr; - return 0; -err2: - kfree(pidarray); -err1: - kfree(ctr); -err0: - return -ENOMEM; -} - -static ssize_t cgroup_tasks_read(struct cgroup *cgrp, - struct cftype *cft, - struct file *file, char __user *buf, - size_t nbytes, loff_t *ppos) -{ - struct ctr_struct *ctr = file->private_data; - - return simple_read_from_buffer(buf, nbytes, ppos, ctr->buf, ctr->bufsz); -} - -static int cgroup_tasks_release(struct inode *unused_inode, - struct file *file) -{ - struct ctr_struct *ctr; - - if (file->f_mode & FMODE_READ) { - ctr = file->private_data; - kfree(ctr->buf); - kfree(ctr); + cp = kzalloc(sizeof(*cp), GFP_KERNEL); + if (!cp) { + up_write(&cgrp->pids_mutex); + kfree(pidarray); + return -ENOMEM; } + cp->cgrp = cgrp; + cp->ns = ns; + get_pid_ns(ns); + list_add(&cp->list, &cgrp->pids_list); +found: + kfree(cp->tasks_pids); + cp->tasks_pids = pidarray; + cp->length = npids; + cp->use_count++; + up_write(&cgrp->pids_mutex); + + file->f_op = &cgroup_tasks_operations; + + retval = seq_open(file, &cgroup_tasks_seq_operations); + if (retval) { + release_cgroup_pid_array(cp); + return retval; + } + ((struct seq_file *)file->private_data)->private = cp; return 0; } @@ -2138,9 +2433,16 @@ static u64 cgroup_read_notify_on_release(struct cgroup *cgrp, return notify_on_release(cgrp); } -static u64 cgroup_read_releasable(struct cgroup *cgrp, struct cftype *cft) +static int cgroup_write_notify_on_release(struct cgroup *cgrp, + struct cftype *cft, + u64 val) { - return test_bit(CGRP_RELEASABLE, &cgrp->flags); + clear_bit(CGRP_RELEASABLE, &cgrp->flags); + if (val) + set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); + else + clear_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags); + return 0; } /* @@ -2150,30 +2452,25 @@ static struct cftype files[] = { { .name = "tasks", .open = cgroup_tasks_open, - .read = cgroup_tasks_read, - .write = cgroup_common_file_write, + .write_u64 = cgroup_tasks_write, .release = cgroup_tasks_release, .private = FILE_TASKLIST, + .mode = S_IRUGO | S_IWUSR, }, { .name = "notify_on_release", - .read_uint = cgroup_read_notify_on_release, - .write = cgroup_common_file_write, + .read_u64 = cgroup_read_notify_on_release, + .write_u64 = cgroup_write_notify_on_release, .private = FILE_NOTIFY_ON_RELEASE, }, - - { - .name = "releasable", - .read_uint = cgroup_read_releasable, - .private = FILE_RELEASABLE, - } }; static struct cftype cft_release_agent = { .name = "release_agent", - .read = cgroup_common_file_read, - .write = cgroup_common_file_write, + .read_seq_string = cgroup_release_agent_show, + .write_string = cgroup_release_agent_write, + .max_write_len = PATH_MAX, .private = FILE_RELEASE_AGENT, }; @@ -2198,6 +2495,17 @@ static int cgroup_populate_dir(struct cgroup *cgrp) if (ss->populate && (err = ss->populate(ss, cgrp)) < 0) return err; } + /* This cgroup is ready now */ + for_each_subsys(cgrp->root, ss) { + struct cgroup_subsys_state *css = cgrp->subsys[ss->subsys_id]; + /* + * Update id->css pointer and make this css visible from + * CSS ID functions. This pointer will be dereferened + * from RCU-read-side without locks. + */ + if (css->id) + rcu_assign_pointer(css->id->css, css); + } return 0; } @@ -2207,14 +2515,38 @@ static void init_cgroup_css(struct cgroup_subsys_state *css, struct cgroup *cgrp) { css->cgroup = cgrp; - atomic_set(&css->refcnt, 0); + atomic_set(&css->refcnt, 1); css->flags = 0; + css->id = NULL; if (cgrp == dummytop) set_bit(CSS_ROOT, &css->flags); BUG_ON(cgrp->subsys[ss->subsys_id]); cgrp->subsys[ss->subsys_id] = css; } +static void cgroup_lock_hierarchy(struct cgroupfs_root *root) +{ + /* We need to take each hierarchy_mutex in a consistent order */ + int i; + + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { + struct cgroup_subsys *ss = subsys[i]; + if (ss->root == root) + mutex_lock(&ss->hierarchy_mutex); + } +} + +static void cgroup_unlock_hierarchy(struct cgroupfs_root *root) +{ + int i; + + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { + struct cgroup_subsys *ss = subsys[i]; + if (ss->root == root) + mutex_unlock(&ss->hierarchy_mutex); + } +} + /* * cgroup_create - create a cgroup * @parent: cgroup that will be parent of the new cgroup @@ -2224,7 +2556,7 @@ static void init_cgroup_css(struct cgroup_subsys_state *css, * Must be called with the mutex on the parent inode held */ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, - int mode) + mode_t mode) { struct cgroup *cgrp; struct cgroupfs_root *root = parent->root; @@ -2245,10 +2577,7 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, mutex_lock(&cgroup_mutex); - INIT_LIST_HEAD(&cgrp->sibling); - INIT_LIST_HEAD(&cgrp->children); - INIT_LIST_HEAD(&cgrp->css_sets); - INIT_LIST_HEAD(&cgrp->release_list); + init_cgroup_housekeeping(cgrp); cgrp->parent = parent; cgrp->root = parent->root; @@ -2264,9 +2593,15 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, goto err_destroy; } init_cgroup_css(css, ss, cgrp); + if (ss->use_id) + if (alloc_css_id(ss, parent, cgrp)) + goto err_destroy; + /* At error, ->destroy() callback has to free assigned ID. */ } + cgroup_lock_hierarchy(root); list_add(&cgrp->sibling, &cgrp->parent->children); + cgroup_unlock_hierarchy(root); root->number_of_cgroups++; err = cgroup_create_dir(cgrp, dentry, mode); @@ -2286,7 +2621,9 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry, err_remove: + cgroup_lock_hierarchy(root); list_del(&cgrp->sibling); + cgroup_unlock_hierarchy(root); root->number_of_cgroups--; err_destroy: @@ -2313,11 +2650,11 @@ static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, int mode) return cgroup_create(c_parent, dentry, mode | S_IFDIR); } -static inline int cgroup_has_css_refs(struct cgroup *cgrp) +static int cgroup_has_css_refs(struct cgroup *cgrp) { /* Check the reference count on each subsystem. Since we * already established that there are no tasks in the - * cgroup, if the css refcount is also 0, then there should + * cgroup, if the css refcount is also 1, then there should * be no outstanding references, so the subsystem is safe to * destroy. We scan across all subsystems rather than using * the per-hierarchy linked list of mounted subsystems since @@ -2338,22 +2675,75 @@ static inline int cgroup_has_css_refs(struct cgroup *cgrp) * matter, since it can only happen if the cgroup * has been deleted and hence no longer needs the * release agent to be called anyway. */ - if (css && atomic_read(&css->refcnt)) + if (css && (atomic_read(&css->refcnt) > 1)) return 1; } return 0; } +/* + * Atomically mark all (or else none) of the cgroup's CSS objects as + * CSS_REMOVED. Return true on success, or false if the cgroup has + * busy subsystems. Call with cgroup_mutex held + */ + +static int cgroup_clear_css_refs(struct cgroup *cgrp) +{ + struct cgroup_subsys *ss; + unsigned long flags; + bool failed = false; + local_irq_save(flags); + for_each_subsys(cgrp->root, ss) { + struct cgroup_subsys_state *css = cgrp->subsys[ss->subsys_id]; + int refcnt; + while (1) { + /* We can only remove a CSS with a refcnt==1 */ + refcnt = atomic_read(&css->refcnt); + if (refcnt > 1) { + failed = true; + goto done; + } + BUG_ON(!refcnt); + /* + * Drop the refcnt to 0 while we check other + * subsystems. This will cause any racing + * css_tryget() to spin until we set the + * CSS_REMOVED bits or abort + */ + if (atomic_cmpxchg(&css->refcnt, refcnt, 0) == refcnt) + break; + cpu_relax(); + } + } + done: + for_each_subsys(cgrp->root, ss) { + struct cgroup_subsys_state *css = cgrp->subsys[ss->subsys_id]; + if (failed) { + /* + * Restore old refcnt if we previously managed + * to clear it from 1 to 0 + */ + if (!atomic_read(&css->refcnt)) + atomic_set(&css->refcnt, 1); + } else { + /* Commit the fact that the CSS is removed */ + set_bit(CSS_REMOVED, &css->flags); + } + } + local_irq_restore(flags); + return !failed; +} + static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry) { struct cgroup *cgrp = dentry->d_fsdata; struct dentry *d; struct cgroup *parent; - struct super_block *sb; - struct cgroupfs_root *root; + DEFINE_WAIT(wait); + int ret; /* the vfs holds both inode->i_mutex already */ - +again: mutex_lock(&cgroup_mutex); if (atomic_read(&cgrp->count) != 0) { mutex_unlock(&cgroup_mutex); @@ -2363,32 +2753,68 @@ static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry) mutex_unlock(&cgroup_mutex); return -EBUSY; } + mutex_unlock(&cgroup_mutex); - parent = cgrp->parent; - root = cgrp->root; - sb = root->sb; + /* + * In general, subsystem has no css->refcnt after pre_destroy(). But + * in racy cases, subsystem may have to get css->refcnt after + * pre_destroy() and it makes rmdir return with -EBUSY. This sometimes + * make rmdir return -EBUSY too often. To avoid that, we use waitqueue + * for cgroup's rmdir. CGRP_WAIT_ON_RMDIR is for synchronizing rmdir + * and subsystem's reference count handling. Please see css_get/put + * and css_tryget() and cgroup_wakeup_rmdir_waiter() implementation. + */ + set_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags); /* * Call pre_destroy handlers of subsys. Notify subsystems * that rmdir() request comes. */ - cgroup_call_pre_destroy(cgrp); + ret = cgroup_call_pre_destroy(cgrp); + if (ret) { + clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags); + return ret; + } - if (cgroup_has_css_refs(cgrp)) { + mutex_lock(&cgroup_mutex); + parent = cgrp->parent; + if (atomic_read(&cgrp->count) || !list_empty(&cgrp->children)) { + clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags); mutex_unlock(&cgroup_mutex); return -EBUSY; } + prepare_to_wait(&cgroup_rmdir_waitq, &wait, TASK_INTERRUPTIBLE); + if (!cgroup_clear_css_refs(cgrp)) { + mutex_unlock(&cgroup_mutex); + /* + * Because someone may call cgroup_wakeup_rmdir_waiter() before + * prepare_to_wait(), we need to check this flag. + */ + if (test_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags)) + schedule(); + finish_wait(&cgroup_rmdir_waitq, &wait); + clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags); + if (signal_pending(current)) + return -EINTR; + goto again; + } + /* NO css_tryget() can success after here. */ + finish_wait(&cgroup_rmdir_waitq, &wait); + clear_bit(CGRP_WAIT_ON_RMDIR, &cgrp->flags); spin_lock(&release_list_lock); set_bit(CGRP_REMOVED, &cgrp->flags); if (!list_empty(&cgrp->release_list)) list_del(&cgrp->release_list); spin_unlock(&release_list_lock); - /* delete my sibling from parent->children */ + + cgroup_lock_hierarchy(cgrp->root); + /* delete this cgroup from parent->children */ list_del(&cgrp->sibling); + cgroup_unlock_hierarchy(cgrp->root); + spin_lock(&cgrp->dentry->d_lock); d = dget(cgrp->dentry); - cgrp->dentry = NULL; spin_unlock(&d->d_lock); cgroup_d_remove_dir(d); @@ -2401,49 +2827,35 @@ static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry) return 0; } -static void cgroup_init_subsys(struct cgroup_subsys *ss) +static void __init cgroup_init_subsys(struct cgroup_subsys *ss) { struct cgroup_subsys_state *css; - struct list_head *l; printk(KERN_INFO "Initializing cgroup subsys %s\n", ss->name); /* Create the top cgroup state for this subsystem */ + list_add(&ss->sibling, &rootnode.subsys_list); ss->root = &rootnode; css = ss->create(ss, dummytop); /* We don't handle early failures gracefully */ BUG_ON(IS_ERR(css)); init_cgroup_css(css, ss, dummytop); - /* Update all cgroup groups to contain a subsys + /* Update the init_css_set to contain a subsys * pointer to this state - since the subsystem is - * newly registered, all tasks and hence all cgroup - * groups are in the subsystem's top cgroup. */ - write_lock(&css_set_lock); - l = &init_css_set.list; - do { - struct css_set *cg = - list_entry(l, struct css_set, list); - cg->subsys[ss->subsys_id] = dummytop->subsys[ss->subsys_id]; - l = l->next; - } while (l != &init_css_set.list); - write_unlock(&css_set_lock); - - /* If this subsystem requested that it be notified with fork - * events, we should send it one now for every process in the - * system */ - if (ss->fork) { - struct task_struct *g, *p; - - read_lock(&tasklist_lock); - do_each_thread(g, p) { - ss->fork(ss, p); - } while_each_thread(g, p); - read_unlock(&tasklist_lock); - } + * newly registered, all tasks and hence the + * init_css_set is in the subsystem's top cgroup. */ + init_css_set.subsys[ss->subsys_id] = dummytop->subsys[ss->subsys_id]; need_forkexit_callback |= ss->fork || ss->exit; + /* At system boot, before all subsystems have been + * registered, no tasks have been forked, so we don't + * need to invoke fork callbacks here. */ + BUG_ON(!list_empty(&init_task.tasks)); + + mutex_init(&ss->hierarchy_mutex); + lockdep_set_class(&ss->hierarchy_mutex, &ss->subsys_key); ss->active = 1; } @@ -2456,14 +2868,12 @@ static void cgroup_init_subsys(struct cgroup_subsys *ss) int __init cgroup_init_early(void) { int i; - kref_init(&init_css_set.ref); - kref_get(&init_css_set.ref); - INIT_LIST_HEAD(&init_css_set.list); + atomic_set(&init_css_set.refcount, 1); INIT_LIST_HEAD(&init_css_set.cg_links); INIT_LIST_HEAD(&init_css_set.tasks); + INIT_HLIST_NODE(&init_css_set.hlist); css_set_count = 1; init_cgroup_root(&rootnode); - list_add(&rootnode.root_list, &roots); root_count = 1; init_task.cgroups = &init_css_set; @@ -2473,6 +2883,9 @@ int __init cgroup_init_early(void) list_add(&init_css_set_link.cg_link_list, &init_css_set.cg_links); + for (i = 0; i < CSS_SET_TABLE_SIZE; i++) + INIT_HLIST_HEAD(&css_set_table[i]); + for (i = 0; i < CGROUP_SUBSYS_COUNT; i++) { struct cgroup_subsys *ss = subsys[i]; @@ -2502,7 +2915,7 @@ int __init cgroup_init(void) { int err; int i; - struct proc_dir_entry *entry; + struct hlist_head *hhead; err = bdi_init(&cgroup_backing_dev_info); if (err) @@ -2512,15 +2925,19 @@ int __init cgroup_init(void) struct cgroup_subsys *ss = subsys[i]; if (!ss->early_init) cgroup_init_subsys(ss); + if (ss->use_id) + cgroup_subsys_init_idr(ss); } + /* Add init_css_set to the hash table */ + hhead = css_set_hash(init_css_set.subsys); + hlist_add_head(&init_css_set.hlist, hhead); + err = register_filesystem(&cgroup_fs_type); if (err < 0) goto out; - entry = create_proc_entry("cgroups", 0, NULL); - if (entry) - entry->proc_fops = &proc_cgroupstats_operations; + proc_create("cgroups", 0, NULL, &proc_cgroupstats_operations); out: if (err) @@ -2565,15 +2982,12 @@ static int proc_cgroup_show(struct seq_file *m, void *v) mutex_lock(&cgroup_mutex); - for_each_root(root) { + for_each_active_root(root) { struct cgroup_subsys *ss; struct cgroup *cgrp; int subsys_id; int count = 0; - /* Skip this hierarchy if it has no active subsystems */ - if (!root->actual_subsys_bits) - continue; seq_printf(m, "%lu:", root->subsys_bits); for_each_subsys(root, ss) seq_printf(m, "%s%s", count++ ? "," : "", ss->name); @@ -2696,8 +3110,10 @@ void cgroup_post_fork(struct task_struct *child) { if (use_task_css_set_links) { write_lock(&css_set_lock); + task_lock(child); if (list_empty(&child->cg_list)) list_add(&child->cg_list, &child->cgroups->tasks); + task_unlock(child); write_unlock(&css_set_lock); } } @@ -2774,16 +3190,17 @@ void cgroup_exit(struct task_struct *tsk, int run_callbacks) * cgroup_clone - clone the cgroup the given subsystem is attached to * @tsk: the task to be moved * @subsys: the given subsystem + * @nodename: the name for the new cgroup * * Duplicate the current cgroup in the hierarchy that the given * subsystem is attached to, and move this task into the new * child. */ -int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys) +int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys, + char *nodename) { struct dentry *dentry; int ret = 0; - char nodename[MAX_CGROUP_TYPE_NAMELEN]; struct cgroup *parent, *child; struct inode *inode; struct css_set *cg; @@ -2799,22 +3216,24 @@ int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys) again: root = subsys->root; if (root == &rootnode) { - printk(KERN_INFO - "Not cloning cgroup for unused subsystem %s\n", - subsys->name); mutex_unlock(&cgroup_mutex); return 0; } - cg = tsk->cgroups; - parent = task_cgroup(tsk, subsys->subsys_id); - - snprintf(nodename, MAX_CGROUP_TYPE_NAMELEN, "node_%d", tsk->pid); /* Pin the hierarchy */ - atomic_inc(&parent->root->sb->s_active); + if (!atomic_inc_not_zero(&root->sb->s_active)) { + /* We race with the final deactivate_super() */ + mutex_unlock(&cgroup_mutex); + return 0; + } /* Keep the cgroup alive */ + task_lock(tsk); + parent = task_cgroup(tsk, subsys->subsys_id); + cg = tsk->cgroups; get_css_set(cg); + task_unlock(tsk); + mutex_unlock(&cgroup_mutex); /* Now do the VFS work to create a cgroup */ @@ -2833,7 +3252,7 @@ int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys) } /* Create the cgroup directory, which also creates the cgroup */ - ret = vfs_mkdir(inode, dentry, S_IFDIR | 0755); + ret = vfs_mkdir(inode, dentry, 0755); child = __d_cgrp(dentry); dput(dentry); if (ret) { @@ -2843,13 +3262,6 @@ int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys) goto out_release; } - if (!child) { - printk(KERN_INFO - "Couldn't find new cgroup %s\n", nodename); - ret = -ENOMEM; - goto out_release; - } - /* The cgroup now exists. Retake cgroup_mutex and check * that we're still in the same state that we thought we * were. */ @@ -2860,7 +3272,7 @@ int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys) mutex_unlock(&inode->i_mutex); put_css_set(cg); - deactivate_super(parent->root->sb); + deactivate_super(root->sb); /* The cgroup is still accessible in the VFS, but * we're not going to try to rmdir() it at this * point. */ @@ -2886,23 +3298,24 @@ int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys) mutex_lock(&cgroup_mutex); put_css_set(cg); mutex_unlock(&cgroup_mutex); - deactivate_super(parent->root->sb); + deactivate_super(root->sb); return ret; } /** - * cgroup_is_descendant - see if @cgrp is a descendant of current task's cgrp + * cgroup_is_descendant - see if @cgrp is a descendant of @task's cgrp * @cgrp: the cgroup in question + * @task: the task in question * - * See if @cgrp is a descendant of the current task's cgroup in - * the appropriate hierarchy. + * See if @cgrp is a descendant of @task's cgroup in the appropriate + * hierarchy. * * If we are sending in dummytop, then presumably we are creating * the top cgroup in the subsystem. * * Called only by the ns (nsproxy) cgroup. */ -int cgroup_is_descendant(const struct cgroup *cgrp) +int cgroup_is_descendant(const struct cgroup *cgrp, struct task_struct *task) { int ret; struct cgroup *target; @@ -2912,7 +3325,7 @@ int cgroup_is_descendant(const struct cgroup *cgrp) return 1; get_first_subsys(cgrp, NULL, &subsys_id); - target = task_cgroup(current, subsys_id); + target = task_cgroup(task, subsys_id); while (cgrp != target && cgrp!= cgrp->top_cgroup) cgrp = cgrp->parent; ret = (cgrp == target); @@ -2945,9 +3358,12 @@ void __css_put(struct cgroup_subsys_state *css) { struct cgroup *cgrp = css->cgroup; rcu_read_lock(); - if (atomic_dec_and_test(&css->refcnt) && notify_on_release(cgrp)) { - set_bit(CGRP_RELEASABLE, &cgrp->flags); - check_for_release(cgrp); + if (atomic_dec_return(&css->refcnt) == 1) { + if (notify_on_release(cgrp)) { + set_bit(CGRP_RELEASABLE, &cgrp->flags); + check_for_release(cgrp); + } + cgroup_wakeup_rmdir_waiter(cgrp); } rcu_read_unlock(); } @@ -2983,27 +3399,24 @@ static void cgroup_release_agent(struct work_struct *work) while (!list_empty(&release_list)) { char *argv[3], *envp[3]; int i; - char *pathbuf; + char *pathbuf = NULL, *agentbuf = NULL; struct cgroup *cgrp = list_entry(release_list.next, struct cgroup, release_list); list_del_init(&cgrp->release_list); spin_unlock(&release_list_lock); pathbuf = kmalloc(PAGE_SIZE, GFP_KERNEL); - if (!pathbuf) { - spin_lock(&release_list_lock); - continue; - } - - if (cgroup_path(cgrp, pathbuf, PAGE_SIZE) < 0) { - kfree(pathbuf); - spin_lock(&release_list_lock); - continue; - } + if (!pathbuf) + goto continue_free; + if (cgroup_path(cgrp, pathbuf, PAGE_SIZE) < 0) + goto continue_free; + agentbuf = kstrdup(cgrp->root->release_agent_path, GFP_KERNEL); + if (!agentbuf) + goto continue_free; i = 0; - argv[i++] = cgrp->root->release_agent_path; - argv[i++] = (char *)pathbuf; + argv[i++] = agentbuf; + argv[i++] = pathbuf; argv[i] = NULL; i = 0; @@ -3017,8 +3430,10 @@ static void cgroup_release_agent(struct work_struct *work) * be a slow process */ mutex_unlock(&cgroup_mutex); call_usermodehelper(argv[0], argv, envp, UMH_WAIT_EXEC); - kfree(pathbuf); mutex_lock(&cgroup_mutex); + continue_free: + kfree(pathbuf); + kfree(agentbuf); spin_lock(&release_list_lock); } spin_unlock(&release_list_lock); @@ -3048,3 +3463,232 @@ static int __init cgroup_disable(char *str) return 1; } __setup("cgroup_disable=", cgroup_disable); + +/* + * Functons for CSS ID. + */ + +/* + *To get ID other than 0, this should be called when !cgroup_is_removed(). + */ +unsigned short css_id(struct cgroup_subsys_state *css) +{ + struct css_id *cssid = rcu_dereference(css->id); + + if (cssid) + return cssid->id; + return 0; +} + +unsigned short css_depth(struct cgroup_subsys_state *css) +{ + struct css_id *cssid = rcu_dereference(css->id); + + if (cssid) + return cssid->depth; + return 0; +} + +bool css_is_ancestor(struct cgroup_subsys_state *child, + const struct cgroup_subsys_state *root) +{ + struct css_id *child_id = rcu_dereference(child->id); + struct css_id *root_id = rcu_dereference(root->id); + + if (!child_id || !root_id || (child_id->depth < root_id->depth)) + return false; + return child_id->stack[root_id->depth] == root_id->id; +} + +static void __free_css_id_cb(struct rcu_head *head) +{ + struct css_id *id; + + id = container_of(head, struct css_id, rcu_head); + kfree(id); +} + +void free_css_id(struct cgroup_subsys *ss, struct cgroup_subsys_state *css) +{ + struct css_id *id = css->id; + /* When this is called before css_id initialization, id can be NULL */ + if (!id) + return; + + BUG_ON(!ss->use_id); + + rcu_assign_pointer(id->css, NULL); + rcu_assign_pointer(css->id, NULL); + spin_lock(&ss->id_lock); + idr_remove(&ss->idr, id->id); + spin_unlock(&ss->id_lock); + call_rcu(&id->rcu_head, __free_css_id_cb); +} + +/* + * This is called by init or create(). Then, calls to this function are + * always serialized (By cgroup_mutex() at create()). + */ + +static struct css_id *get_new_cssid(struct cgroup_subsys *ss, int depth) +{ + struct css_id *newid; + int myid, error, size; + + BUG_ON(!ss->use_id); + + size = sizeof(*newid) + sizeof(unsigned short) * (depth + 1); + newid = kzalloc(size, GFP_KERNEL); + if (!newid) + return ERR_PTR(-ENOMEM); + /* get id */ + if (unlikely(!idr_pre_get(&ss->idr, GFP_KERNEL))) { + error = -ENOMEM; + goto err_out; + } + spin_lock(&ss->id_lock); + /* Don't use 0. allocates an ID of 1-65535 */ + error = idr_get_new_above(&ss->idr, newid, 1, &myid); + spin_unlock(&ss->id_lock); + + /* Returns error when there are no free spaces for new ID.*/ + if (error) { + error = -ENOSPC; + goto err_out; + } + if (myid > CSS_ID_MAX) + goto remove_idr; + + newid->id = myid; + newid->depth = depth; + return newid; +remove_idr: + error = -ENOSPC; + spin_lock(&ss->id_lock); + idr_remove(&ss->idr, myid); + spin_unlock(&ss->id_lock); +err_out: + kfree(newid); + return ERR_PTR(error); + +} + +static int __init cgroup_subsys_init_idr(struct cgroup_subsys *ss) +{ + struct css_id *newid; + struct cgroup_subsys_state *rootcss; + + spin_lock_init(&ss->id_lock); + idr_init(&ss->idr); + + rootcss = init_css_set.subsys[ss->subsys_id]; + newid = get_new_cssid(ss, 0); + if (IS_ERR(newid)) + return PTR_ERR(newid); + + newid->stack[0] = newid->id; + newid->css = rootcss; + rootcss->id = newid; + return 0; +} + +static int alloc_css_id(struct cgroup_subsys *ss, struct cgroup *parent, + struct cgroup *child) +{ + int subsys_id, i, depth = 0; + struct cgroup_subsys_state *parent_css, *child_css; + struct css_id *child_id, *parent_id = NULL; + + subsys_id = ss->subsys_id; + parent_css = parent->subsys[subsys_id]; + child_css = child->subsys[subsys_id]; + depth = css_depth(parent_css) + 1; + parent_id = parent_css->id; + + child_id = get_new_cssid(ss, depth); + if (IS_ERR(child_id)) + return PTR_ERR(child_id); + + for (i = 0; i < depth; i++) + child_id->stack[i] = parent_id->stack[i]; + child_id->stack[depth] = child_id->id; + /* + * child_id->css pointer will be set after this cgroup is available + * see cgroup_populate_dir() + */ + rcu_assign_pointer(child_css->id, child_id); + + return 0; +} + +/** + * css_lookup - lookup css by id + * @ss: cgroup subsys to be looked into. + * @id: the id + * + * Returns pointer to cgroup_subsys_state if there is valid one with id. + * NULL if not. Should be called under rcu_read_lock() + */ +struct cgroup_subsys_state *css_lookup(struct cgroup_subsys *ss, int id) +{ + struct css_id *cssid = NULL; + + BUG_ON(!ss->use_id); + cssid = idr_find(&ss->idr, id); + + if (unlikely(!cssid)) + return NULL; + + return rcu_dereference(cssid->css); +} + +/** + * css_get_next - lookup next cgroup under specified hierarchy. + * @ss: pointer to subsystem + * @id: current position of iteration. + * @root: pointer to css. search tree under this. + * @foundid: position of found object. + * + * Search next css under the specified hierarchy of rootid. Calling under + * rcu_read_lock() is necessary. Returns NULL if it reaches the end. + */ +struct cgroup_subsys_state * +css_get_next(struct cgroup_subsys *ss, int id, + struct cgroup_subsys_state *root, int *foundid) +{ + struct cgroup_subsys_state *ret = NULL; + struct css_id *tmp; + int tmpid; + int rootid = css_id(root); + int depth = css_depth(root); + + if (!rootid) + return NULL; + + BUG_ON(!ss->use_id); + /* fill start point for scan */ + tmpid = id; + while (1) { + /* + * scan next entry from bitmap(tree), tmpid is updated after + * idr_get_next(). + */ + spin_lock(&ss->id_lock); + tmp = idr_get_next(&ss->idr, &tmpid); + spin_unlock(&ss->id_lock); + + if (!tmp) + break; + if (tmp->depth >= depth && tmp->stack[depth] == rootid) { + ret = rcu_dereference(tmp->css); + if (ret) { + *foundid = tmpid; + break; + } + } + /* continue to scan from next id */ + tmpid = tmpid + 1; + } + return ret; +} +