diff options
Diffstat (limited to 'kernel/cgroup.c')
-rw-r--r-- | kernel/cgroup.c | 283 |
1 files changed, 150 insertions, 133 deletions
diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 469dd547770c..f89d9292eee6 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -46,6 +46,7 @@ #include <linux/slab.h> #include <linux/spinlock.h> #include <linux/rwsem.h> +#include <linux/percpu-rwsem.h> #include <linux/string.h> #include <linux/sort.h> #include <linux/kmod.h> @@ -103,6 +104,8 @@ static DEFINE_SPINLOCK(cgroup_idr_lock); */ static DEFINE_SPINLOCK(release_agent_path_lock); +struct percpu_rw_semaphore cgroup_threadgroup_rwsem; + #define cgroup_assert_mutex_or_rcu_locked() \ rcu_lockdep_assert(rcu_read_lock_held() || \ lockdep_is_held(&cgroup_mutex), \ @@ -156,7 +159,7 @@ static bool cgrp_dfl_root_visible; static bool cgroup_legacy_files_on_dfl; /* some controllers are not supported in the default hierarchy */ -static unsigned int cgrp_dfl_root_inhibit_ss_mask; +static unsigned long cgrp_dfl_root_inhibit_ss_mask; /* The list of hierarchy roots */ @@ -175,18 +178,19 @@ static DEFINE_IDR(cgroup_hierarchy_idr); */ static u64 css_serial_nr_next = 1; -/* This flag indicates whether tasks in the fork and exit paths should - * check for fork/exit handlers to call. This avoids us having to do - * extra work in the fork/exit path if none of the subsystems need to - * be called. +/* + * These bitmask flags indicate whether tasks in the fork and exit paths have + * fork/exit handlers to call. This avoids us having to do extra work in the + * fork/exit path to check which subsystems have fork/exit callbacks. */ -static int need_forkexit_callback __read_mostly; +static unsigned long have_fork_callback __read_mostly; +static unsigned long have_exit_callback __read_mostly; static struct cftype cgroup_dfl_base_files[]; static struct cftype cgroup_legacy_base_files[]; static int rebind_subsystems(struct cgroup_root *dst_root, - unsigned int ss_mask); + unsigned long ss_mask); static int cgroup_destroy_locked(struct cgroup *cgrp); static int create_css(struct cgroup *cgrp, struct cgroup_subsys *ss, bool visible); @@ -261,7 +265,7 @@ static struct cgroup_subsys_state *cgroup_css(struct cgroup *cgrp, * @cgrp: the cgroup of interest * @ss: the subsystem of interest (%NULL returns @cgrp->self) * - * Similar to cgroup_css() but returns the effctive css, which is defined + * Similar to cgroup_css() but returns the effective css, which is defined * as the matching css of the nearest ancestor including self which has @ss * enabled. If @ss is associated with the hierarchy @cgrp is on, this * function is guaranteed to return non-NULL css. @@ -409,6 +413,24 @@ static int notify_on_release(const struct cgroup *cgrp) for ((ssid) = 0; (ssid) < CGROUP_SUBSYS_COUNT && \ (((ss) = cgroup_subsys[ssid]) || true); (ssid)++) +/** + * for_each_subsys_which - filter for_each_subsys with a bitmask + * @ss: the iteration cursor + * @ssid: the index of @ss, CGROUP_SUBSYS_COUNT after reaching the end + * @ss_maskp: a pointer to the bitmask + * + * The block will only run for cases where the ssid-th bit (1 << ssid) of + * mask is set to 1. + */ +#define for_each_subsys_which(ss, ssid, ss_maskp) \ + if (!CGROUP_SUBSYS_COUNT) /* to avoid spurious gcc warning */ \ + (ssid) = 0; \ + else \ + for_each_set_bit(ssid, ss_maskp, CGROUP_SUBSYS_COUNT) \ + if (((ss) = cgroup_subsys[ssid]) && false) \ + break; \ + else + /* iterate across the hierarchies */ #define for_each_root(root) \ list_for_each_entry((root), &cgroup_roots, root_list) @@ -882,7 +904,7 @@ static void cgroup_exit_root_id(struct cgroup_root *root) static void cgroup_free_root(struct cgroup_root *root) { if (root) { - /* hierarhcy ID shoulid already have been released */ + /* hierarchy ID should already have been released */ WARN_ON_ONCE(root->hierarchy_id); idr_destroy(&root->cgroup_idr); @@ -998,7 +1020,7 @@ static struct cgroup *task_cgroup_from_root(struct task_struct *task, * update of a tasks cgroup pointer by cgroup_attach_task() */ -static int cgroup_populate_dir(struct cgroup *cgrp, unsigned int subsys_mask); +static int cgroup_populate_dir(struct cgroup *cgrp, unsigned long subsys_mask); static struct kernfs_syscall_ops cgroup_kf_syscall_ops; static const struct file_operations proc_cgroupstats_operations; @@ -1068,11 +1090,11 @@ static void cgroup_put(struct cgroup *cgrp) * @subtree_control is to be applied to @cgrp. The returned mask is always * a superset of @subtree_control and follows the usual hierarchy rules. */ -static unsigned int cgroup_calc_child_subsys_mask(struct cgroup *cgrp, - unsigned int subtree_control) +static unsigned long cgroup_calc_child_subsys_mask(struct cgroup *cgrp, + unsigned long subtree_control) { struct cgroup *parent = cgroup_parent(cgrp); - unsigned int cur_ss_mask = subtree_control; + unsigned long cur_ss_mask = subtree_control; struct cgroup_subsys *ss; int ssid; @@ -1082,11 +1104,10 @@ static unsigned int cgroup_calc_child_subsys_mask(struct cgroup *cgrp, return cur_ss_mask; while (true) { - unsigned int new_ss_mask = cur_ss_mask; + unsigned long new_ss_mask = cur_ss_mask; - for_each_subsys(ss, ssid) - if (cur_ss_mask & (1 << ssid)) - new_ss_mask |= ss->depends_on; + for_each_subsys_which(ss, ssid, &cur_ss_mask) + new_ss_mask |= ss->depends_on; /* * Mask out subsystems which aren't available. This can @@ -1200,7 +1221,7 @@ static void cgroup_rm_file(struct cgroup *cgrp, const struct cftype *cft) * @cgrp: target cgroup * @subsys_mask: mask of the subsystem ids whose files should be removed */ -static void cgroup_clear_dir(struct cgroup *cgrp, unsigned int subsys_mask) +static void cgroup_clear_dir(struct cgroup *cgrp, unsigned long subsys_mask) { struct cgroup_subsys *ss; int i; @@ -1215,18 +1236,16 @@ static void cgroup_clear_dir(struct cgroup *cgrp, unsigned int subsys_mask) } } -static int rebind_subsystems(struct cgroup_root *dst_root, unsigned int ss_mask) +static int rebind_subsystems(struct cgroup_root *dst_root, + unsigned long ss_mask) { struct cgroup_subsys *ss; - unsigned int tmp_ss_mask; + unsigned long tmp_ss_mask; int ssid, i, ret; lockdep_assert_held(&cgroup_mutex); - for_each_subsys(ss, ssid) { - if (!(ss_mask & (1 << ssid))) - continue; - + for_each_subsys_which(ss, ssid, &ss_mask) { /* if @ss has non-root csses attached to it, can't move */ if (css_next_child(NULL, cgroup_css(&ss->root->cgrp, ss))) return -EBUSY; @@ -1253,7 +1272,7 @@ static int rebind_subsystems(struct cgroup_root *dst_root, unsigned int ss_mask) * Just warn about it and continue. */ if (cgrp_dfl_root_visible) { - pr_warn("failed to create files (%d) while rebinding 0x%x to default root\n", + pr_warn("failed to create files (%d) while rebinding 0x%lx to default root\n", ret, ss_mask); pr_warn("you may retry by moving them to a different hierarchy and unbinding\n"); } @@ -1263,18 +1282,14 @@ static int rebind_subsystems(struct cgroup_root *dst_root, unsigned int ss_mask) * Nothing can fail from this point on. Remove files for the * removed subsystems and rebind each subsystem. */ - for_each_subsys(ss, ssid) - if (ss_mask & (1 << ssid)) - cgroup_clear_dir(&ss->root->cgrp, 1 << ssid); + for_each_subsys_which(ss, ssid, &ss_mask) + cgroup_clear_dir(&ss->root->cgrp, 1 << ssid); - for_each_subsys(ss, ssid) { + for_each_subsys_which(ss, ssid, &ss_mask) { struct cgroup_root *src_root; struct cgroup_subsys_state *css; struct css_set *cset; - if (!(ss_mask & (1 << ssid))) - continue; - src_root = ss->root; css = cgroup_css(&src_root->cgrp, ss); @@ -1338,7 +1353,7 @@ static int cgroup_show_options(struct seq_file *seq, } struct cgroup_sb_opts { - unsigned int subsys_mask; + unsigned long subsys_mask; unsigned int flags; char *release_agent; bool cpuset_clone_children; @@ -1351,7 +1366,7 @@ static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts) { char *token, *o = data; bool all_ss = false, one_ss = false; - unsigned int mask = -1U; + unsigned long mask = -1UL; struct cgroup_subsys *ss; int nr_opts = 0; int i; @@ -1495,7 +1510,7 @@ static int cgroup_remount(struct kernfs_root *kf_root, int *flags, char *data) int ret = 0; struct cgroup_root *root = cgroup_root_from_kf(kf_root); struct cgroup_sb_opts opts; - unsigned int added_mask, removed_mask; + unsigned long added_mask, removed_mask; if (root == &cgrp_dfl_root) { pr_err("remount is not allowed\n"); @@ -1641,7 +1656,7 @@ static void init_cgroup_root(struct cgroup_root *root, set_bit(CGRP_CPUSET_CLONE_CHILDREN, &root->cgrp.flags); } -static int cgroup_setup_root(struct cgroup_root *root, unsigned int ss_mask) +static int cgroup_setup_root(struct cgroup_root *root, unsigned long ss_mask) { LIST_HEAD(tmp_links); struct cgroup *root_cgrp = &root->cgrp; @@ -1924,8 +1939,6 @@ static struct file_system_type cgroup_fs_type = { .kill_sb = cgroup_kill_sb, }; -static struct kobject *cgroup_kobj; - /** * task_cgroup_path - cgroup path of a task in the first cgroup hierarchy * @task: target task @@ -2052,9 +2065,9 @@ static void cgroup_task_migrate(struct cgroup *old_cgrp, lockdep_assert_held(&css_set_rwsem); /* - * We are synchronized through threadgroup_lock() against PF_EXITING - * setting such that we can't race against cgroup_exit() changing the - * css_set to init_css_set and dropping the old one. + * We are synchronized through cgroup_threadgroup_rwsem against + * PF_EXITING setting such that we can't race against cgroup_exit() + * changing the css_set to init_css_set and dropping the old one. */ WARN_ON_ONCE(tsk->flags & PF_EXITING); old_cset = task_css_set(tsk); @@ -2111,10 +2124,11 @@ static void cgroup_migrate_finish(struct list_head *preloaded_csets) * @src_cset and add it to @preloaded_csets, which should later be cleaned * up by cgroup_migrate_finish(). * - * This function may be called without holding threadgroup_lock even if the - * target is a process. Threads may be created and destroyed but as long - * as cgroup_mutex is not dropped, no new css_set can be put into play and - * the preloaded css_sets are guaranteed to cover all migrations. + * This function may be called without holding cgroup_threadgroup_rwsem + * even if the target is a process. Threads may be created and destroyed + * but as long as cgroup_mutex is not dropped, no new css_set can be put + * into play and the preloaded css_sets are guaranteed to cover all + * migrations. */ static void cgroup_migrate_add_src(struct css_set *src_cset, struct cgroup *dst_cgrp, @@ -2217,7 +2231,7 @@ err: * @threadgroup: whether @leader points to the whole process or a single task * * Migrate a process or task denoted by @leader to @cgrp. If migrating a - * process, the caller must be holding threadgroup_lock of @leader. The + * process, the caller must be holding cgroup_threadgroup_rwsem. The * caller is also responsible for invoking cgroup_migrate_add_src() and * cgroup_migrate_prepare_dst() on the targets before invoking this * function and following up with cgroup_migrate_finish(). @@ -2345,7 +2359,7 @@ out_release_tset: * @leader: the task or the leader of the threadgroup to be attached * @threadgroup: attach the whole threadgroup? * - * Call holding cgroup_mutex and threadgroup_lock of @leader. + * Call holding cgroup_mutex and cgroup_threadgroup_rwsem. */ static int cgroup_attach_task(struct cgroup *dst_cgrp, struct task_struct *leader, bool threadgroup) @@ -2376,6 +2390,47 @@ static int cgroup_attach_task(struct cgroup *dst_cgrp, return ret; } +static int cgroup_procs_write_permission(struct task_struct *task, + struct cgroup *dst_cgrp, + struct kernfs_open_file *of) +{ + const struct cred *cred = current_cred(); + const struct cred *tcred = get_task_cred(task); + int ret = 0; + + /* + * even if we're attaching all tasks in the thread group, we only + * need to check permissions on one of them. + */ + if (!uid_eq(cred->euid, GLOBAL_ROOT_UID) && + !uid_eq(cred->euid, tcred->uid) && + !uid_eq(cred->euid, tcred->suid)) + ret = -EACCES; + + if (!ret && cgroup_on_dfl(dst_cgrp)) { + struct super_block *sb = of->file->f_path.dentry->d_sb; + struct cgroup *cgrp; + struct inode *inode; + + down_read(&css_set_rwsem); + cgrp = task_cgroup_from_root(task, &cgrp_dfl_root); + up_read(&css_set_rwsem); + + while (!cgroup_is_descendant(dst_cgrp, cgrp)) + cgrp = cgroup_parent(cgrp); + + ret = -ENOMEM; + inode = kernfs_get_inode(sb, cgrp->procs_kn); + if (inode) { + ret = inode_permission(inode, MAY_WRITE); + iput(inode); + } + } + + put_cred(tcred); + return ret; +} + /* * Find the task_struct of the task to attach by vpid and pass it along to the * function to attach either it or all tasks in its threadgroup. Will lock @@ -2385,7 +2440,6 @@ static ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf, size_t nbytes, loff_t off, bool threadgroup) { struct task_struct *tsk; - const struct cred *cred = current_cred(), *tcred; struct cgroup *cgrp; pid_t pid; int ret; @@ -2397,29 +2451,17 @@ static ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf, if (!cgrp) return -ENODEV; -retry_find_task: + percpu_down_write(&cgroup_threadgroup_rwsem); rcu_read_lock(); if (pid) { tsk = find_task_by_vpid(pid); if (!tsk) { - rcu_read_unlock(); ret = -ESRCH; - goto out_unlock_cgroup; - } - /* - * even if we're attaching all tasks in the thread group, we - * only need to check permissions on one of them. - */ - tcred = __task_cred(tsk); - if (!uid_eq(cred->euid, GLOBAL_ROOT_UID) && - !uid_eq(cred->euid, tcred->uid) && - !uid_eq(cred->euid, tcred->suid)) { - rcu_read_unlock(); - ret = -EACCES; - goto out_unlock_cgroup; + goto out_unlock_rcu; } - } else + } else { tsk = current; + } if (threadgroup) tsk = tsk->group_leader; @@ -2431,35 +2473,23 @@ retry_find_task: */ if (tsk == kthreadd_task || (tsk->flags & PF_NO_SETAFFINITY)) { ret = -EINVAL; - rcu_read_unlock(); - goto out_unlock_cgroup; + goto out_unlock_rcu; } get_task_struct(tsk); rcu_read_unlock(); - threadgroup_lock(tsk); - if (threadgroup) { - if (!thread_group_leader(tsk)) { - /* - * a race with de_thread from another thread's exec() - * may strip us of our leadership, if this happens, - * there is no choice but to throw this task away and - * try again; this is - * "double-double-toil-and-trouble-check locking". - */ - threadgroup_unlock(tsk); - put_task_struct(tsk); - goto retry_find_task; - } - } - - ret = cgroup_attach_task(cgrp, tsk, threadgroup); - - threadgroup_unlock(tsk); + ret = cgroup_procs_write_permission(tsk, cgrp, of); + if (!ret) + ret = cgroup_attach_task(cgrp, tsk, threadgroup); put_task_struct(tsk); -out_unlock_cgroup: + goto out_unlock_threadgroup; + +out_unlock_rcu: + rcu_read_unlock(); +out_unlock_threadgroup: + percpu_up_write(&cgroup_threadgroup_rwsem); cgroup_kn_unlock(of->kn); return ret ?: nbytes; } @@ -2542,19 +2572,17 @@ static int cgroup_sane_behavior_show(struct seq_file *seq, void *v) return 0; } -static void cgroup_print_ss_mask(struct seq_file *seq, unsigned int ss_mask) +static void cgroup_print_ss_mask(struct seq_file *seq, unsigned long ss_mask) { struct cgroup_subsys *ss; bool printed = false; int ssid; - for_each_subsys(ss, ssid) { - if (ss_mask & (1 << ssid)) { - if (printed) - seq_putc(seq, ' '); - seq_printf(seq, "%s", ss->name); - printed = true; - } + for_each_subsys_which(ss, ssid, &ss_mask) { + if (printed) + seq_putc(seq, ' '); + seq_printf(seq, "%s", ss->name); + printed = true; } if (printed) seq_putc(seq, '\n'); @@ -2606,6 +2634,8 @@ static int cgroup_update_dfl_csses(struct cgroup *cgrp) lockdep_assert_held(&cgroup_mutex); + percpu_down_write(&cgroup_threadgroup_rwsem); + /* look up all csses currently attached to @cgrp's subtree */ down_read(&css_set_rwsem); css_for_each_descendant_pre(css, cgroup_css(cgrp, NULL)) { @@ -2661,17 +2691,8 @@ static int cgroup_update_dfl_csses(struct cgroup *cgrp) goto out_finish; last_task = task; - threadgroup_lock(task); - /* raced against de_thread() from another thread? */ - if (!thread_group_leader(task)) { - threadgroup_unlock(task); - put_task_struct(task); - continue; - } - ret = cgroup_migrate(src_cset->dfl_cgrp, task, true); - threadgroup_unlock(task); put_task_struct(task); if (WARN(ret, "cgroup: failed to update controllers for the default hierarchy (%d), further operations may crash or hang\n", ret)) @@ -2681,6 +2702,7 @@ static int cgroup_update_dfl_csses(struct cgroup *cgrp) out_finish: cgroup_migrate_finish(&preloaded_csets); + percpu_up_write(&cgroup_threadgroup_rwsem); return ret; } @@ -2689,8 +2711,8 @@ static ssize_t cgroup_subtree_control_write(struct kernfs_open_file *of, char *buf, size_t nbytes, loff_t off) { - unsigned int enable = 0, disable = 0; - unsigned int css_enable, css_disable, old_sc, new_sc, old_ss, new_ss; + unsigned long enable = 0, disable = 0; + unsigned long css_enable, css_disable, old_sc, new_sc, old_ss, new_ss; struct cgroup *cgrp, *child; struct cgroup_subsys *ss; char *tok; @@ -2702,11 +2724,12 @@ static ssize_t cgroup_subtree_control_write(struct kernfs_open_file *of, */ buf = strstrip(buf); while ((tok = strsep(&buf, " "))) { + unsigned long tmp_ss_mask = ~cgrp_dfl_root_inhibit_ss_mask; + if (tok[0] == '\0') continue; - for_each_subsys(ss, ssid) { - if (ss->disabled || strcmp(tok + 1, ss->name) || - ((1 << ss->id) & cgrp_dfl_root_inhibit_ss_mask)) + for_each_subsys_which(ss, ssid, &tmp_ss_mask) { + if (ss->disabled || strcmp(tok + 1, ss->name)) continue; if (*tok == '+') { @@ -2793,10 +2816,7 @@ static ssize_t cgroup_subtree_control_write(struct kernfs_open_file *of, * still around. In such cases, wait till it's gone using * offline_waitq. */ - for_each_subsys(ss, ssid) { - if (!(css_enable & (1 << ssid))) - continue; - + for_each_subsys_which(ss, ssid, &css_enable) { cgroup_for_each_live_child(child, cgrp) { DEFINE_WAIT(wait); @@ -3087,7 +3107,9 @@ static int cgroup_add_file(struct cgroup *cgrp, struct cftype *cft) return ret; } - if (cft->seq_show == cgroup_populated_show) + if (cft->write == cgroup_procs_write) + cgrp->procs_kn = kn; + else if (cft->seq_show == cgroup_populated_show) cgrp->populated_kn = kn; return 0; } @@ -4322,7 +4344,7 @@ static struct cftype cgroup_legacy_base_files[] = { * * On failure, no file is added. */ -static int cgroup_populate_dir(struct cgroup *cgrp, unsigned int subsys_mask) +static int cgroup_populate_dir(struct cgroup *cgrp, unsigned long subsys_mask) { struct cgroup_subsys *ss; int i, ret = 0; @@ -4931,7 +4953,8 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss, bool early) * init_css_set is in the subsystem's root cgroup. */ init_css_set.subsys[ss->id] = css; - need_forkexit_callback |= ss->fork || ss->exit; + have_fork_callback |= (bool)ss->fork << ss->id; + have_exit_callback |= (bool)ss->exit << ss->id; /* At system boot, before all subsystems have been * registered, no tasks have been forked, so we don't @@ -4989,6 +5012,7 @@ int __init cgroup_init(void) unsigned long key; int ssid, err; + BUG_ON(percpu_init_rwsem(&cgroup_threadgroup_rwsem)); BUG_ON(cgroup_init_cftypes(NULL, cgroup_dfl_base_files)); BUG_ON(cgroup_init_cftypes(NULL, cgroup_legacy_base_files)); @@ -5044,13 +5068,13 @@ int __init cgroup_init(void) ss->bind(init_css_set.subsys[ssid]); } - cgroup_kobj = kobject_create_and_add("cgroup", fs_kobj); - if (!cgroup_kobj) - return -ENOMEM; + err = sysfs_create_mount_point(fs_kobj, "cgroup"); + if (err) + return err; err = register_filesystem(&cgroup_fs_type); if (err < 0) { - kobject_put(cgroup_kobj); + sysfs_remove_mount_point(fs_kobj, "cgroup"); return err; } @@ -5241,11 +5265,8 @@ void cgroup_post_fork(struct task_struct *child) * css_set; otherwise, @child might change state between ->fork() * and addition to css_set. */ - if (need_forkexit_callback) { - for_each_subsys(ss, i) - if (ss->fork) - ss->fork(child); - } + for_each_subsys_which(ss, i, &have_fork_callback) + ss->fork(child); } /** @@ -5289,16 +5310,12 @@ void cgroup_exit(struct task_struct *tsk) cset = task_css_set(tsk); RCU_INIT_POINTER(tsk->cgroups, &init_css_set); - if (need_forkexit_callback) { - /* see cgroup_post_fork() for details */ - for_each_subsys(ss, i) { - if (ss->exit) { - struct cgroup_subsys_state *old_css = cset->subsys[i]; - struct cgroup_subsys_state *css = task_css(tsk, i); + /* see cgroup_post_fork() for details */ + for_each_subsys_which(ss, i, &have_exit_callback) { + struct cgroup_subsys_state *old_css = cset->subsys[i]; + struct cgroup_subsys_state *css = task_css(tsk, i); - ss->exit(css, old_css, tsk); - } - } + ss->exit(css, old_css, tsk); } if (put_cset) |