#include <linux/pid_namespace.h>
#include <linux/idr.h>
#include <linux/vmalloc.h> /* TODO: replace with more sophisticated array */
-#include <linux/eventfd.h>
-#include <linux/poll.h>
#include <linux/flex_array.h> /* used in cgroup_attach_task */
#include <linux/kthread.h>
-#include <linux/file.h>
#include <linux/atomic.h>
/*
+ * pidlists linger the following amount before being destroyed. The goal
+ * is avoiding frequent destruction in the middle of consecutive read calls
+ * Expiring in the middle is a performance problem not a correctness one.
+ * 1 sec should be enough.
+ */
+#define CGROUP_PIDLIST_DESTROY_DELAY HZ
+
+/*
* cgroup_mutex is the master lock. Any modification to cgroup or its
* hierarchy must be performed while holding it.
*
static DEFINE_MUTEX(cgroup_root_mutex);
/*
+ * cgroup destruction makes heavy use of work items and there can be a lot
+ * of concurrent destructions. Use a separate workqueue so that cgroup
+ * destruction work items don't end up filling up max_active of system_wq
+ * which may lead to deadlock.
+ */
+static struct workqueue_struct *cgroup_destroy_wq;
+
+/*
+ * pidlist destructions need to be flushed on cgroup destruction. Use a
+ * separate workqueue as flush domain.
+ */
+static struct workqueue_struct *cgroup_pidlist_destroy_wq;
+
+/*
* Generate an array of cgroup subsystem pointers. At boot time, this is
* populated with the built in subsystems, and modular subsystems are
* registered after that. The mutable section of this array is protected by
struct simple_xattrs xattrs;
};
-/*
- * cgroup_event represents events which userspace want to receive.
- */
-struct cgroup_event {
- /*
- * css which the event belongs to.
- */
- struct cgroup_subsys_state *css;
- /*
- * Control file which the event associated.
- */
- struct cftype *cft;
- /*
- * eventfd to signal userspace about the event.
- */
- struct eventfd_ctx *eventfd;
- /*
- * Each of these stored in a list by the cgroup.
- */
- struct list_head list;
- /*
- * All fields below needed to unregister event when
- * userspace closes eventfd.
- */
- poll_table pt;
- wait_queue_head_t *wqh;
- wait_queue_t wait;
- struct work_struct remove;
-};
-
/* The list of hierarchy roots */
static LIST_HEAD(cgroup_roots);
static int cgroup_destroy_locked(struct cgroup *cgrp);
static int cgroup_addrm_files(struct cgroup *cgrp, struct cftype cfts[],
bool is_add);
+static int cgroup_file_release(struct inode *inode, struct file *file);
+static void cgroup_pidlist_destroy_all(struct cgroup *cgrp);
/**
* cgroup_css - obtain a cgroup's css for the specified subsystem
*/
deactivate_super(cgrp->root->sb);
- /*
- * if we're getting rid of the cgroup, refcount should ensure
- * that there are no pidlists left.
- */
- BUG_ON(!list_empty(&cgrp->pidlists));
+ cgroup_pidlist_destroy_all(cgrp);
simple_xattrs_free(&cgrp->xattrs);
struct cgroup *cgrp = container_of(head, struct cgroup, rcu_head);
INIT_WORK(&cgrp->destroy_work, cgroup_free_fn);
- schedule_work(&cgrp->destroy_work);
+ queue_work(cgroup_destroy_wq, &cgrp->destroy_work);
}
static void cgroup_diput(struct dentry *dentry, struct inode *inode)
iput(inode);
}
-static int cgroup_delete(const struct dentry *d)
-{
- return 1;
-}
-
static void remove_dir(struct dentry *d)
{
struct dentry *parent = dget(d->d_parent);
INIT_LIST_HEAD(&cgrp->pidlists);
mutex_init(&cgrp->pidlist_mutex);
cgrp->dummy_css.cgroup = cgrp;
- INIT_LIST_HEAD(&cgrp->event_list);
- spin_lock_init(&cgrp->event_list_lock);
simple_xattrs_init(&cgrp->xattrs);
}
{
static const struct dentry_operations cgroup_dops = {
.d_iput = cgroup_diput,
- .d_delete = cgroup_delete,
+ .d_delete = always_delete_dentry,
};
struct inode *inode =
struct cftype *cft = __d_cft(file->f_dentry);
struct cgroup_subsys_state *css = cfe->css;
- if (cft->write)
- return cft->write(css, cft, file, buf, nbytes, ppos);
if (cft->write_u64 || cft->write_s64)
return cgroup_write_X64(css, cft, file, buf, nbytes, ppos);
if (cft->write_string)
struct cftype *cft = __d_cft(file->f_dentry);
struct cgroup_subsys_state *css = cfe->css;
- if (cft->read)
- return cft->read(css, cft, file, buf, nbytes, ppos);
if (cft->read_u64)
return cgroup_read_u64(css, cft, file, buf, nbytes, ppos);
if (cft->read_s64)
* supports string->u64 maps, but can be extended in future.
*/
-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 cfent *cfe = m->private;
struct cftype *cft = cfe->type;
struct cgroup_subsys_state *css = cfe->css;
- if (cft->read_map) {
- struct cgroup_map_cb cb = {
- .fill = cgroup_map_add,
- .state = m,
- };
- return cft->read_map(css, cft, &cb);
- }
return cft->read_seq_string(css, cft, m);
}
.read = seq_read,
.write = cgroup_file_write,
.llseek = seq_lseek,
- .release = single_release,
+ .release = cgroup_file_release,
};
static int cgroup_file_open(struct inode *inode, struct file *file)
WARN_ON_ONCE(cfe->css && cfe->css != css);
cfe->css = css;
- if (cft->read_map || cft->read_seq_string) {
+ if (cft->read_seq_string) {
file->f_op = &cgroup_seqfile_operations;
err = single_open(file, cgroup_seqfile_show, cfe);
} else if (cft->open) {
static int cgroup_file_release(struct inode *inode, struct file *file)
{
struct cfent *cfe = __d_cfe(file->f_dentry);
- struct cftype *cft = __d_cft(file->f_dentry);
struct cgroup_subsys_state *css = cfe->css;
- int ret = 0;
- if (cft->release)
- ret = cft->release(inode, file);
if (css->ss)
css_put(css);
- return ret;
+ if (file->f_op == &cgroup_seqfile_operations)
+ single_release(inode, file);
+ return 0;
}
/*
.removexattr = cgroup_removexattr,
};
-/*
- * Check if a file is a control file
- */
-static inline struct cftype *__file_cft(struct file *file)
-{
- if (file_inode(file)->i_fop != &cgroup_file_operations)
- return ERR_PTR(-EINVAL);
- return __d_cft(file->f_dentry);
-}
-
static int cgroup_create_file(struct dentry *dentry, umode_t mode,
struct super_block *sb)
{
if (cft->mode)
return cft->mode;
- if (cft->read || cft->read_u64 || cft->read_s64 ||
- cft->read_map || cft->read_seq_string)
+ if (cft->read_u64 || cft->read_s64 || cft->read_seq_string)
mode |= S_IRUGO;
- if (cft->write || cft->write_u64 || cft->write_s64 ||
- cft->write_string || cft->trigger)
+ if (cft->write_u64 || cft->write_s64 || cft->write_string ||
+ cft->trigger)
mode |= S_IWUSR;
return mode;
pid_t *list;
/* how many elements the above list has */
int length;
- /* how many files are using the current array */
- int use_count;
/* each of these stored in a list by its cgroup */
struct list_head links;
/* pointer to the cgroup we belong to, for list removal purposes */
struct cgroup *owner;
- /* protects the other fields */
- struct rw_semaphore rwsem;
+ /* for delayed destruction */
+ struct delayed_work destroy_dwork;
+};
+
+/* seq_file->private points to the following */
+struct cgroup_pidlist_open_file {
+ enum cgroup_filetype type;
+ struct cgroup *cgrp;
+ struct cgroup_pidlist *pidlist;
};
/*
else
return kmalloc(count * sizeof(pid_t), GFP_KERNEL);
}
+
static void pidlist_free(void *p)
{
if (is_vmalloc_addr(p))
}
/*
+ * Used to destroy all pidlists lingering waiting for destroy timer. None
+ * should be left afterwards.
+ */
+static void cgroup_pidlist_destroy_all(struct cgroup *cgrp)
+{
+ struct cgroup_pidlist *l, *tmp_l;
+
+ mutex_lock(&cgrp->pidlist_mutex);
+ list_for_each_entry_safe(l, tmp_l, &cgrp->pidlists, links)
+ mod_delayed_work(cgroup_pidlist_destroy_wq, &l->destroy_dwork, 0);
+ mutex_unlock(&cgrp->pidlist_mutex);
+
+ flush_workqueue(cgroup_pidlist_destroy_wq);
+ BUG_ON(!list_empty(&cgrp->pidlists));
+}
+
+static void cgroup_pidlist_destroy_work_fn(struct work_struct *work)
+{
+ struct delayed_work *dwork = to_delayed_work(work);
+ struct cgroup_pidlist *l = container_of(dwork, struct cgroup_pidlist,
+ destroy_dwork);
+ struct cgroup_pidlist *tofree = NULL;
+
+ mutex_lock(&l->owner->pidlist_mutex);
+
+ /*
+ * Destroy iff we didn't get queued again. The state won't change
+ * as destroy_dwork can only be queued while locked.
+ */
+ if (!delayed_work_pending(dwork)) {
+ list_del(&l->links);
+ pidlist_free(l->list);
+ put_pid_ns(l->key.ns);
+ tofree = l;
+ }
+
+ mutex_unlock(&l->owner->pidlist_mutex);
+ kfree(tofree);
+}
+
+/*
* pidlist_uniq - given a kmalloc()ed list, strip out all duplicate entries
* Returns the number of unique elements.
*/
return dest;
}
+/*
+ * The two pid files - task and cgroup.procs - guaranteed that the result
+ * is sorted, which forced this whole pidlist fiasco. As pid order is
+ * different per namespace, each namespace needs differently sorted list,
+ * making it impossible to use, for example, single rbtree of member tasks
+ * sorted by task pointer. As pidlists can be fairly large, allocating one
+ * per open file is dangerous, so cgroup had to implement shared pool of
+ * pidlists keyed by cgroup and namespace.
+ *
+ * All this extra complexity was caused by the original implementation
+ * committing to an entirely unnecessary property. In the long term, we
+ * want to do away with it. Explicitly scramble sort order if
+ * sane_behavior so that no such expectation exists in the new interface.
+ *
+ * Scrambling is done by swapping every two consecutive bits, which is
+ * non-identity one-to-one mapping which disturbs sort order sufficiently.
+ */
+static pid_t pid_fry(pid_t pid)
+{
+ unsigned a = pid & 0x55555555;
+ unsigned b = pid & 0xAAAAAAAA;
+
+ return (a << 1) | (b >> 1);
+}
+
+static pid_t cgroup_pid_fry(struct cgroup *cgrp, pid_t pid)
+{
+ if (cgroup_sane_behavior(cgrp))
+ return pid_fry(pid);
+ else
+ return pid;
+}
+
static int cmppid(const void *a, const void *b)
{
return *(pid_t *)a - *(pid_t *)b;
}
+static int fried_cmppid(const void *a, const void *b)
+{
+ return pid_fry(*(pid_t *)a) - pid_fry(*(pid_t *)b);
+}
+
+static struct cgroup_pidlist *cgroup_pidlist_find(struct cgroup *cgrp,
+ enum cgroup_filetype type)
+{
+ struct cgroup_pidlist *l;
+ /* don't need task_nsproxy() if we're looking at ourself */
+ struct pid_namespace *ns = task_active_pid_ns(current);
+
+ lockdep_assert_held(&cgrp->pidlist_mutex);
+
+ list_for_each_entry(l, &cgrp->pidlists, links)
+ if (l->key.type == type && l->key.ns == ns)
+ return l;
+ return NULL;
+}
+
/*
* find the appropriate pidlist for our purpose (given procs vs tasks)
* returns with the lock on that pidlist already held, and takes care
* of the use count, or returns NULL with no locks held if we're out of
* memory.
*/
-static struct cgroup_pidlist *cgroup_pidlist_find(struct cgroup *cgrp,
- enum cgroup_filetype type)
+static struct cgroup_pidlist *cgroup_pidlist_find_create(struct cgroup *cgrp,
+ enum cgroup_filetype type)
{
struct cgroup_pidlist *l;
- /* don't need task_nsproxy() if we're looking at ourself */
- struct pid_namespace *ns = task_active_pid_ns(current);
- /*
- * We can't drop the pidlist_mutex before taking the l->rwsem in case
- * the last ref-holder is trying to remove l from the list at the same
- * time. Holding the pidlist_mutex precludes somebody taking whichever
- * list we find out from under us - compare release_pid_array().
- */
- mutex_lock(&cgrp->pidlist_mutex);
- list_for_each_entry(l, &cgrp->pidlists, links) {
- if (l->key.type == type && l->key.ns == ns) {
- /* make sure l doesn't vanish out from under us */
- down_write(&l->rwsem);
- mutex_unlock(&cgrp->pidlist_mutex);
- return l;
- }
- }
+ lockdep_assert_held(&cgrp->pidlist_mutex);
+
+ l = cgroup_pidlist_find(cgrp, type);
+ if (l)
+ return l;
+
/* entry not found; create a new one */
l = kzalloc(sizeof(struct cgroup_pidlist), GFP_KERNEL);
- if (!l) {
- mutex_unlock(&cgrp->pidlist_mutex);
+ if (!l)
return l;
- }
- init_rwsem(&l->rwsem);
- down_write(&l->rwsem);
+
+ INIT_DELAYED_WORK(&l->destroy_dwork, cgroup_pidlist_destroy_work_fn);
l->key.type = type;
- l->key.ns = get_pid_ns(ns);
+ /* don't need task_nsproxy() if we're looking at ourself */
+ l->key.ns = get_pid_ns(task_active_pid_ns(current));
l->owner = cgrp;
list_add(&l->links, &cgrp->pidlists);
- mutex_unlock(&cgrp->pidlist_mutex);
return l;
}
struct task_struct *tsk;
struct cgroup_pidlist *l;
+ lockdep_assert_held(&cgrp->pidlist_mutex);
+
/*
* If cgroup gets more users after we read count, we won't have
* enough space - tough. This race is indistinguishable to the
css_task_iter_end(&it);
length = n;
/* now sort & (if procs) strip out duplicates */
- sort(array, length, sizeof(pid_t), cmppid, NULL);
+ if (cgroup_sane_behavior(cgrp))
+ sort(array, length, sizeof(pid_t), fried_cmppid, NULL);
+ else
+ sort(array, length, sizeof(pid_t), cmppid, NULL);
if (type == CGROUP_FILE_PROCS)
length = pidlist_uniq(array, length);
- l = cgroup_pidlist_find(cgrp, type);
+
+ l = cgroup_pidlist_find_create(cgrp, type);
if (!l) {
+ mutex_unlock(&cgrp->pidlist_mutex);
pidlist_free(array);
return -ENOMEM;
}
- /* store array, freeing old if necessary - lock already held */
+
+ /* store array, freeing old if necessary */
pidlist_free(l->list);
l->list = array;
l->length = length;
- l->use_count++;
- up_write(&l->rwsem);
*lp = l;
return 0;
}
* after a seek to the start). Use a binary-search to find the
* next pid to display, if any
*/
- struct cgroup_pidlist *l = s->private;
+ struct cgroup_pidlist_open_file *of = s->private;
+ struct cgroup *cgrp = of->cgrp;
+ struct cgroup_pidlist *l;
int index = 0, pid = *pos;
- int *iter;
+ int *iter, ret;
+
+ mutex_lock(&cgrp->pidlist_mutex);
+
+ /*
+ * !NULL @of->pidlist indicates that this isn't the first start()
+ * after open. If the matching pidlist is around, we can use that.
+ * Look for it. Note that @of->pidlist can't be used directly. It
+ * could already have been destroyed.
+ */
+ if (of->pidlist)
+ of->pidlist = cgroup_pidlist_find(cgrp, of->type);
+
+ /*
+ * Either this is the first start() after open or the matching
+ * pidlist has been destroyed inbetween. Create a new one.
+ */
+ if (!of->pidlist) {
+ ret = pidlist_array_load(of->cgrp, of->type, &of->pidlist);
+ if (ret)
+ return ERR_PTR(ret);
+ }
+ l = of->pidlist;
- down_read(&l->rwsem);
if (pid) {
int end = l->length;
while (index < end) {
int mid = (index + end) / 2;
- if (l->list[mid] == pid) {
+ if (cgroup_pid_fry(cgrp, l->list[mid]) == pid) {
index = mid;
break;
- } else if (l->list[mid] <= pid)
+ } else if (cgroup_pid_fry(cgrp, l->list[mid]) <= pid)
index = mid + 1;
else
end = mid;
return NULL;
/* Update the abstract position to be the actual pid that we found */
iter = l->list + index;
- *pos = *iter;
+ *pos = cgroup_pid_fry(cgrp, *iter);
return iter;
}
static void cgroup_pidlist_stop(struct seq_file *s, void *v)
{
- struct cgroup_pidlist *l = s->private;
- up_read(&l->rwsem);
+ struct cgroup_pidlist_open_file *of = s->private;
+
+ if (of->pidlist)
+ mod_delayed_work(cgroup_pidlist_destroy_wq,
+ &of->pidlist->destroy_dwork,
+ CGROUP_PIDLIST_DESTROY_DELAY);
+ mutex_unlock(&of->cgrp->pidlist_mutex);
}
static void *cgroup_pidlist_next(struct seq_file *s, void *v, loff_t *pos)
{
- struct cgroup_pidlist *l = s->private;
+ struct cgroup_pidlist_open_file *of = s->private;
+ struct cgroup_pidlist *l = of->pidlist;
pid_t *p = v;
pid_t *end = l->list + l->length;
/*
if (p >= end) {
return NULL;
} else {
- *pos = *p;
+ *pos = cgroup_pid_fry(of->cgrp, *p);
return p;
}
}
.show = cgroup_pidlist_show,
};
-static void cgroup_release_pid_array(struct cgroup_pidlist *l)
-{
- /*
- * the case where we're the last user of this particular pidlist will
- * have us remove it from the cgroup's list, which entails taking the
- * mutex. since in pidlist_find the pidlist->lock depends on cgroup->
- * pidlist_mutex, we have to take pidlist_mutex first.
- */
- mutex_lock(&l->owner->pidlist_mutex);
- down_write(&l->rwsem);
- BUG_ON(!l->use_count);
- if (!--l->use_count) {
- /* we're the last user if refcount is 0; remove and free */
- list_del(&l->links);
- mutex_unlock(&l->owner->pidlist_mutex);
- pidlist_free(l->list);
- put_pid_ns(l->key.ns);
- up_write(&l->rwsem);
- kfree(l);
- return;
- }
- mutex_unlock(&l->owner->pidlist_mutex);
- up_write(&l->rwsem);
-}
-
-static int cgroup_pidlist_release(struct inode *inode, struct file *file)
-{
- struct cgroup_pidlist *l;
- if (!(file->f_mode & FMODE_READ))
- return 0;
- /*
- * the seq_file will only be initialized if the file was opened for
- * reading; hence we check if it's not null only in that case.
- */
- l = ((struct seq_file *)file->private_data)->private;
- cgroup_release_pid_array(l);
- return seq_release(inode, file);
-}
-
static const struct file_operations cgroup_pidlist_operations = {
.read = seq_read,
.llseek = seq_lseek,
.write = cgroup_file_write,
- .release = cgroup_pidlist_release,
+ .release = seq_release_private,
};
/*
static int cgroup_pidlist_open(struct file *file, enum cgroup_filetype type)
{
struct cgroup *cgrp = __d_cgrp(file->f_dentry->d_parent);
- struct cgroup_pidlist *l;
- int retval;
-
- /* Nothing to do for write-only files */
- if (!(file->f_mode & FMODE_READ))
- return 0;
+ struct cgroup_pidlist_open_file *of;
- /* have the array populated */
- retval = pidlist_array_load(cgrp, type, &l);
- if (retval)
- return retval;
/* configure file information */
file->f_op = &cgroup_pidlist_operations;
- retval = seq_open(file, &cgroup_pidlist_seq_operations);
- if (retval) {
- cgroup_release_pid_array(l);
- return retval;
- }
- ((struct seq_file *)file->private_data)->private = l;
+ of = __seq_open_private(file, &cgroup_pidlist_seq_operations,
+ sizeof(*of));
+ if (!of)
+ return -ENOMEM;
+
+ of->type = type;
+ of->cgrp = cgrp;
return 0;
}
static int cgroup_tasks_open(struct inode *unused, struct file *file)
deactivate_super(sb);
}
-/*
- * Unregister event and free resources.
- *
- * Gets called from workqueue.
- */
-static void cgroup_event_remove(struct work_struct *work)
-{
- struct cgroup_event *event = container_of(work, struct cgroup_event,
- remove);
- struct cgroup_subsys_state *css = event->css;
-
- remove_wait_queue(event->wqh, &event->wait);
-
- event->cft->unregister_event(css, event->cft, event->eventfd);
-
- /* Notify userspace the event is going away. */
- eventfd_signal(event->eventfd, 1);
-
- eventfd_ctx_put(event->eventfd);
- kfree(event);
- css_put(css);
-}
-
-/*
- * Gets called on POLLHUP on eventfd when user closes it.
- *
- * Called with wqh->lock held and interrupts disabled.
- */
-static int cgroup_event_wake(wait_queue_t *wait, unsigned mode,
- int sync, void *key)
-{
- struct cgroup_event *event = container_of(wait,
- struct cgroup_event, wait);
- struct cgroup *cgrp = event->css->cgroup;
- unsigned long flags = (unsigned long)key;
-
- if (flags & POLLHUP) {
- /*
- * If the event has been detached at cgroup removal, we
- * can simply return knowing the other side will cleanup
- * for us.
- *
- * We can't race against event freeing since the other
- * side will require wqh->lock via remove_wait_queue(),
- * which we hold.
- */
- spin_lock(&cgrp->event_list_lock);
- if (!list_empty(&event->list)) {
- list_del_init(&event->list);
- /*
- * We are in atomic context, but cgroup_event_remove()
- * may sleep, so we have to call it in workqueue.
- */
- schedule_work(&event->remove);
- }
- spin_unlock(&cgrp->event_list_lock);
- }
-
- return 0;
-}
-
-static void cgroup_event_ptable_queue_proc(struct file *file,
- wait_queue_head_t *wqh, poll_table *pt)
-{
- struct cgroup_event *event = container_of(pt,
- struct cgroup_event, pt);
-
- event->wqh = wqh;
- add_wait_queue(wqh, &event->wait);
-}
-
-/*
- * Parse input and register new cgroup event handler.
- *
- * Input must be in format '<event_fd> <control_fd> <args>'.
- * Interpretation of args is defined by control file implementation.
- */
-static int cgroup_write_event_control(struct cgroup_subsys_state *dummy_css,
- struct cftype *cft, const char *buffer)
-{
- struct cgroup *cgrp = dummy_css->cgroup;
- struct cgroup_event *event;
- struct cgroup_subsys_state *cfile_css;
- unsigned int efd, cfd;
- struct fd efile;
- struct fd cfile;
- char *endp;
- int ret;
-
- efd = simple_strtoul(buffer, &endp, 10);
- if (*endp != ' ')
- return -EINVAL;
- buffer = endp + 1;
-
- cfd = simple_strtoul(buffer, &endp, 10);
- if ((*endp != ' ') && (*endp != '\0'))
- return -EINVAL;
- buffer = endp + 1;
-
- event = kzalloc(sizeof(*event), GFP_KERNEL);
- if (!event)
- return -ENOMEM;
-
- INIT_LIST_HEAD(&event->list);
- init_poll_funcptr(&event->pt, cgroup_event_ptable_queue_proc);
- init_waitqueue_func_entry(&event->wait, cgroup_event_wake);
- INIT_WORK(&event->remove, cgroup_event_remove);
-
- efile = fdget(efd);
- if (!efile.file) {
- ret = -EBADF;
- goto out_kfree;
- }
-
- event->eventfd = eventfd_ctx_fileget(efile.file);
- if (IS_ERR(event->eventfd)) {
- ret = PTR_ERR(event->eventfd);
- goto out_put_efile;
- }
-
- cfile = fdget(cfd);
- if (!cfile.file) {
- ret = -EBADF;
- goto out_put_eventfd;
- }
-
- /* the process need read permission on control file */
- /* AV: shouldn't we check that it's been opened for read instead? */
- ret = inode_permission(file_inode(cfile.file), MAY_READ);
- if (ret < 0)
- goto out_put_cfile;
-
- event->cft = __file_cft(cfile.file);
- if (IS_ERR(event->cft)) {
- ret = PTR_ERR(event->cft);
- goto out_put_cfile;
- }
-
- if (!event->cft->ss) {
- ret = -EBADF;
- goto out_put_cfile;
- }
-
- /*
- * Determine the css of @cfile, verify it belongs to the same
- * cgroup as cgroup.event_control, and associate @event with it.
- * Remaining events are automatically removed on cgroup destruction
- * but the removal is asynchronous, so take an extra ref.
- */
- rcu_read_lock();
-
- ret = -EINVAL;
- event->css = cgroup_css(cgrp, event->cft->ss);
- cfile_css = css_from_dir(cfile.file->f_dentry->d_parent, event->cft->ss);
- if (event->css && event->css == cfile_css && css_tryget(event->css))
- ret = 0;
-
- rcu_read_unlock();
- if (ret)
- goto out_put_cfile;
-
- if (!event->cft->register_event || !event->cft->unregister_event) {
- ret = -EINVAL;
- goto out_put_css;
- }
-
- ret = event->cft->register_event(event->css, event->cft,
- event->eventfd, buffer);
- if (ret)
- goto out_put_css;
-
- efile.file->f_op->poll(efile.file, &event->pt);
-
- spin_lock(&cgrp->event_list_lock);
- list_add(&event->list, &cgrp->event_list);
- spin_unlock(&cgrp->event_list_lock);
-
- fdput(cfile);
- fdput(efile);
-
- return 0;
-
-out_put_css:
- css_put(event->css);
-out_put_cfile:
- fdput(cfile);
-out_put_eventfd:
- eventfd_ctx_put(event->eventfd);
-out_put_efile:
- fdput(efile);
-out_kfree:
- kfree(event);
-
- return ret;
-}
-
static u64 cgroup_clone_children_read(struct cgroup_subsys_state *css,
struct cftype *cft)
{
.name = "cgroup.procs",
.open = cgroup_procs_open,
.write_u64 = cgroup_procs_write,
- .release = cgroup_pidlist_release,
.mode = S_IRUGO | S_IWUSR,
},
{
- .name = "cgroup.event_control",
- .write_string = cgroup_write_event_control,
- .mode = S_IWUGO,
- },
- {
.name = "cgroup.clone_children",
.flags = CFTYPE_INSANE,
.read_u64 = cgroup_clone_children_read,
.flags = CFTYPE_INSANE, /* use "procs" instead */
.open = cgroup_tasks_open,
.write_u64 = cgroup_tasks_write,
- .release = cgroup_pidlist_release,
.mode = S_IRUGO | S_IWUSR,
},
{
* css_put(). dput() requires process context which we don't have.
*/
INIT_WORK(&css->destroy_work, css_free_work_fn);
- schedule_work(&css->destroy_work);
+ queue_work(cgroup_destroy_wq, &css->destroy_work);
}
static void css_release(struct percpu_ref *ref)
container_of(ref, struct cgroup_subsys_state, refcnt);
INIT_WORK(&css->destroy_work, css_killed_work_fn);
- schedule_work(&css->destroy_work);
+ queue_work(cgroup_destroy_wq, &css->destroy_work);
}
/**
__releases(&cgroup_mutex) __acquires(&cgroup_mutex)
{
struct dentry *d = cgrp->dentry;
- struct cgroup_event *event, *tmp;
struct cgroup_subsys *ss;
struct cgroup *child;
bool empty;
dget(d);
cgroup_d_remove_dir(d);
- /*
- * Unregister events and notify userspace.
- * Notify userspace about cgroup removing only after rmdir of cgroup
- * directory to avoid race between userspace and kernelspace.
- */
- spin_lock(&cgrp->event_list_lock);
- list_for_each_entry_safe(event, tmp, &cgrp->event_list, list) {
- list_del_init(&event->list);
- schedule_work(&event->remove);
- }
- spin_unlock(&cgrp->event_list_lock);
-
return 0;
};
return err;
}
+static int __init cgroup_wq_init(void)
+{
+ /*
+ * There isn't much point in executing destruction path in
+ * parallel. Good chunk is serialized with cgroup_mutex anyway.
+ * Use 1 for @max_active.
+ *
+ * We would prefer to do this in cgroup_init() above, but that
+ * is called before init_workqueues(): so leave this until after.
+ */
+ cgroup_destroy_wq = alloc_workqueue("cgroup_destroy", 0, 1);
+ BUG_ON(!cgroup_destroy_wq);
+
+ /*
+ * Used to destroy pidlists and separate to serve as flush domain.
+ * Cap @max_active to 1 too.
+ */
+ cgroup_pidlist_destroy_wq = alloc_workqueue("cgroup_pidlist_destroy",
+ 0, 1);
+ BUG_ON(!cgroup_pidlist_destroy_wq);
+
+ return 0;
+}
+core_initcall(cgroup_wq_init);
+
/*
* proc_cgroup_show()
* - Print task's cgroup paths into seq_file, one line for each hierarchy