aboutsummaryrefslogtreecommitdiffstats
path: root/kernel/cgroup
diff options
context:
space:
mode:
Diffstat (limited to 'kernel/cgroup')
-rw-r--r--kernel/cgroup/cgroup-internal.h2
-rw-r--r--kernel/cgroup/cgroup-v1.c58
-rw-r--r--kernel/cgroup/cgroup.c54
-rw-r--r--kernel/cgroup/cpuset.c13
-rw-r--r--kernel/cgroup/pids.c4
-rw-r--r--kernel/cgroup/rstat.c10
6 files changed, 57 insertions, 84 deletions
diff --git a/kernel/cgroup/cgroup-internal.h b/kernel/cgroup/cgroup-internal.h
index c950864016e2..c9a35f09e4b9 100644
--- a/kernel/cgroup/cgroup-internal.h
+++ b/kernel/cgroup/cgroup-internal.h
@@ -198,7 +198,7 @@ int cgroup_path_ns_locked(struct cgroup *cgrp, char *buf, size_t buflen,
void cgroup_free_root(struct cgroup_root *root);
void init_cgroup_root(struct cgroup_root *root, struct cgroup_sb_opts *opts);
-int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask, int ref_flags);
+int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask);
int rebind_subsystems(struct cgroup_root *dst_root, u16 ss_mask);
struct dentry *cgroup_do_mount(struct file_system_type *fs_type, int flags,
struct cgroup_root *root, unsigned long magic,
diff --git a/kernel/cgroup/cgroup-v1.c b/kernel/cgroup/cgroup-v1.c
index 583b969b0c0e..f94a7229974e 100644
--- a/kernel/cgroup/cgroup-v1.c
+++ b/kernel/cgroup/cgroup-v1.c
@@ -1116,13 +1116,11 @@ struct dentry *cgroup1_mount(struct file_system_type *fs_type, int flags,
void *data, unsigned long magic,
struct cgroup_namespace *ns)
{
- struct super_block *pinned_sb = NULL;
struct cgroup_sb_opts opts;
struct cgroup_root *root;
struct cgroup_subsys *ss;
struct dentry *dentry;
int i, ret;
- bool new_root = false;
cgroup_lock_and_drain_offline(&cgrp_dfl_root.cgrp);
@@ -1184,29 +1182,6 @@ struct dentry *cgroup1_mount(struct file_system_type *fs_type, int flags,
if (root->flags ^ opts.flags)
pr_warn("new mount options do not match the existing superblock, will be ignored\n");
- /*
- * We want to reuse @root whose lifetime is governed by its
- * ->cgrp. Let's check whether @root is alive and keep it
- * that way. As cgroup_kill_sb() can happen anytime, we
- * want to block it by pinning the sb so that @root doesn't
- * get killed before mount is complete.
- *
- * With the sb pinned, tryget_live can reliably indicate
- * whether @root can be reused. If it's being killed,
- * drain it. We can use wait_queue for the wait but this
- * path is super cold. Let's just sleep a bit and retry.
- */
- pinned_sb = kernfs_pin_sb(root->kf_root, NULL);
- if (IS_ERR(pinned_sb) ||
- !percpu_ref_tryget_live(&root->cgrp.self.refcnt)) {
- mutex_unlock(&cgroup_mutex);
- if (!IS_ERR_OR_NULL(pinned_sb))
- deactivate_super(pinned_sb);
- msleep(10);
- ret = restart_syscall();
- goto out_free;
- }
-
ret = 0;
goto out_unlock;
}
@@ -1232,15 +1207,20 @@ struct dentry *cgroup1_mount(struct file_system_type *fs_type, int flags,
ret = -ENOMEM;
goto out_unlock;
}
- new_root = true;
init_cgroup_root(root, &opts);
- ret = cgroup_setup_root(root, opts.subsys_mask, PERCPU_REF_INIT_DEAD);
+ ret = cgroup_setup_root(root, opts.subsys_mask);
if (ret)
cgroup_free_root(root);
out_unlock:
+ if (!ret && !percpu_ref_tryget_live(&root->cgrp.self.refcnt)) {
+ mutex_unlock(&cgroup_mutex);
+ msleep(10);
+ ret = restart_syscall();
+ goto out_free;
+ }
mutex_unlock(&cgroup_mutex);
out_free:
kfree(opts.release_agent);
@@ -1252,25 +1232,13 @@ out_free:
dentry = cgroup_do_mount(&cgroup_fs_type, flags, root,
CGROUP_SUPER_MAGIC, ns);
- /*
- * There's a race window after we release cgroup_mutex and before
- * allocating a superblock. Make sure a concurrent process won't
- * be able to re-use the root during this window by delaying the
- * initialization of root refcnt.
- */
- if (new_root) {
- mutex_lock(&cgroup_mutex);
- percpu_ref_reinit(&root->cgrp.self.refcnt);
- mutex_unlock(&cgroup_mutex);
+ if (!IS_ERR(dentry) && percpu_ref_is_dying(&root->cgrp.self.refcnt)) {
+ struct super_block *sb = dentry->d_sb;
+ dput(dentry);
+ deactivate_locked_super(sb);
+ msleep(10);
+ dentry = ERR_PTR(restart_syscall());
}
-
- /*
- * If @pinned_sb, we're reusing an existing root and holding an
- * extra ref on its sb. Mount is complete. Put the extra ref.
- */
- if (pinned_sb)
- deactivate_super(pinned_sb);
-
return dentry;
}
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index f31bd61c9466..eef24a25bda7 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -197,7 +197,7 @@ static u64 css_serial_nr_next = 1;
*/
static u16 have_fork_callback __read_mostly;
static u16 have_exit_callback __read_mostly;
-static u16 have_free_callback __read_mostly;
+static u16 have_release_callback __read_mostly;
static u16 have_canfork_callback __read_mostly;
/* cgroup namespace for init task */
@@ -1927,7 +1927,7 @@ void init_cgroup_root(struct cgroup_root *root, struct cgroup_sb_opts *opts)
set_bit(CGRP_CPUSET_CLONE_CHILDREN, &root->cgrp.flags);
}
-int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask, int ref_flags)
+int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask)
{
LIST_HEAD(tmp_links);
struct cgroup *root_cgrp = &root->cgrp;
@@ -1944,7 +1944,7 @@ int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask, int ref_flags)
root_cgrp->ancestor_ids[0] = ret;
ret = percpu_ref_init(&root_cgrp->self.refcnt, css_release,
- ref_flags, GFP_KERNEL);
+ 0, GFP_KERNEL);
if (ret)
goto out;
@@ -2033,7 +2033,7 @@ struct dentry *cgroup_do_mount(struct file_system_type *fs_type, int flags,
struct cgroup_namespace *ns)
{
struct dentry *dentry;
- bool new_sb;
+ bool new_sb = false;
dentry = kernfs_mount(fs_type, flags, root->kf_root, magic, &new_sb);
@@ -2043,6 +2043,7 @@ struct dentry *cgroup_do_mount(struct file_system_type *fs_type, int flags,
*/
if (!IS_ERR(dentry) && ns != &init_cgroup_ns) {
struct dentry *nsdentry;
+ struct super_block *sb = dentry->d_sb;
struct cgroup *cgrp;
mutex_lock(&cgroup_mutex);
@@ -2053,12 +2054,14 @@ struct dentry *cgroup_do_mount(struct file_system_type *fs_type, int flags,
spin_unlock_irq(&css_set_lock);
mutex_unlock(&cgroup_mutex);
- nsdentry = kernfs_node_dentry(cgrp->kn, dentry->d_sb);
+ nsdentry = kernfs_node_dentry(cgrp->kn, sb);
dput(dentry);
+ if (IS_ERR(nsdentry))
+ deactivate_locked_super(sb);
dentry = nsdentry;
}
- if (IS_ERR(dentry) || !new_sb)
+ if (!new_sb)
cgroup_put(&root->cgrp);
return dentry;
@@ -2118,18 +2121,16 @@ static void cgroup_kill_sb(struct super_block *sb)
struct cgroup_root *root = cgroup_root_from_kf(kf_root);
/*
- * If @root doesn't have any mounts or children, start killing it.
+ * If @root doesn't have any children, start killing it.
* This prevents new mounts by disabling percpu_ref_tryget_live().
* cgroup_mount() may wait for @root's release.
*
* And don't kill the default root.
*/
- if (!list_empty(&root->cgrp.self.children) ||
- root == &cgrp_dfl_root)
- cgroup_put(&root->cgrp);
- else
+ if (list_empty(&root->cgrp.self.children) && root != &cgrp_dfl_root &&
+ !percpu_ref_is_dying(&root->cgrp.self.refcnt))
percpu_ref_kill(&root->cgrp.self.refcnt);
-
+ cgroup_put(&root->cgrp);
kernfs_kill_sb(sb);
}
@@ -3533,6 +3534,16 @@ static ssize_t cgroup_file_write(struct kernfs_open_file *of, char *buf,
return ret ?: nbytes;
}
+static __poll_t cgroup_file_poll(struct kernfs_open_file *of, poll_table *pt)
+{
+ struct cftype *cft = of->kn->priv;
+
+ if (cft->poll)
+ return cft->poll(of, pt);
+
+ return kernfs_generic_poll(of, pt);
+}
+
static void *cgroup_seqfile_start(struct seq_file *seq, loff_t *ppos)
{
return seq_cft(seq)->seq_start(seq, ppos);
@@ -3571,6 +3582,7 @@ static struct kernfs_ops cgroup_kf_single_ops = {
.open = cgroup_file_open,
.release = cgroup_file_release,
.write = cgroup_file_write,
+ .poll = cgroup_file_poll,
.seq_show = cgroup_seqfile_show,
};
@@ -3579,6 +3591,7 @@ static struct kernfs_ops cgroup_kf_ops = {
.open = cgroup_file_open,
.release = cgroup_file_release,
.write = cgroup_file_write,
+ .poll = cgroup_file_poll,
.seq_start = cgroup_seqfile_start,
.seq_next = cgroup_seqfile_next,
.seq_stop = cgroup_seqfile_stop,
@@ -5313,7 +5326,7 @@ static void __init cgroup_init_subsys(struct cgroup_subsys *ss, bool early)
have_fork_callback |= (bool)ss->fork << ss->id;
have_exit_callback |= (bool)ss->exit << ss->id;
- have_free_callback |= (bool)ss->free << ss->id;
+ have_release_callback |= (bool)ss->release << ss->id;
have_canfork_callback |= (bool)ss->can_fork << ss->id;
/* At system boot, before all subsystems have been
@@ -5399,7 +5412,7 @@ int __init cgroup_init(void)
hash_add(css_set_table, &init_css_set.hlist,
css_set_hash(init_css_set.subsys));
- BUG_ON(cgroup_setup_root(&cgrp_dfl_root, 0, 0));
+ BUG_ON(cgroup_setup_root(&cgrp_dfl_root, 0));
mutex_unlock(&cgroup_mutex);
@@ -5749,16 +5762,19 @@ void cgroup_exit(struct task_struct *tsk)
} while_each_subsys_mask();
}
-void cgroup_free(struct task_struct *task)
+void cgroup_release(struct task_struct *task)
{
- struct css_set *cset = task_css_set(task);
struct cgroup_subsys *ss;
int ssid;
- do_each_subsys_mask(ss, ssid, have_free_callback) {
- ss->free(task);
+ do_each_subsys_mask(ss, ssid, have_release_callback) {
+ ss->release(task);
} while_each_subsys_mask();
+}
+void cgroup_free(struct task_struct *task)
+{
+ struct css_set *cset = task_css_set(task);
put_css_set(cset);
}
@@ -5996,7 +6012,7 @@ int cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog,
int ret;
mutex_lock(&cgroup_mutex);
- ret = __cgroup_bpf_detach(cgrp, prog, type, flags);
+ ret = __cgroup_bpf_detach(cgrp, prog, type);
mutex_unlock(&cgroup_mutex);
return ret;
}
diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
index 479743db6c37..72afd55f70c6 100644
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -203,19 +203,6 @@ static inline struct cpuset *parent_cs(struct cpuset *cs)
return css_cs(cs->css.parent);
}
-#ifdef CONFIG_NUMA
-static inline bool task_has_mempolicy(struct task_struct *task)
-{
- return task->mempolicy;
-}
-#else
-static inline bool task_has_mempolicy(struct task_struct *task)
-{
- return false;
-}
-#endif
-
-
/* bits in struct cpuset flags field */
typedef enum {
CS_ONLINE,
diff --git a/kernel/cgroup/pids.c b/kernel/cgroup/pids.c
index 9829c67ebc0a..c9960baaa14f 100644
--- a/kernel/cgroup/pids.c
+++ b/kernel/cgroup/pids.c
@@ -247,7 +247,7 @@ static void pids_cancel_fork(struct task_struct *task)
pids_uncharge(pids, 1);
}
-static void pids_free(struct task_struct *task)
+static void pids_release(struct task_struct *task)
{
struct pids_cgroup *pids = css_pids(task_css(task, pids_cgrp_id));
@@ -342,7 +342,7 @@ struct cgroup_subsys pids_cgrp_subsys = {
.cancel_attach = pids_cancel_attach,
.can_fork = pids_can_fork,
.cancel_fork = pids_cancel_fork,
- .free = pids_free,
+ .release = pids_release,
.legacy_cftypes = pids_files,
.dfl_cftypes = pids_files,
.threaded = true,
diff --git a/kernel/cgroup/rstat.c b/kernel/cgroup/rstat.c
index d503d1a9007c..bb95a35e8c2d 100644
--- a/kernel/cgroup/rstat.c
+++ b/kernel/cgroup/rstat.c
@@ -87,7 +87,6 @@ static struct cgroup *cgroup_rstat_cpu_pop_updated(struct cgroup *pos,
struct cgroup *root, int cpu)
{
struct cgroup_rstat_cpu *rstatc;
- struct cgroup *parent;
if (pos == root)
return NULL;
@@ -115,8 +114,8 @@ static struct cgroup *cgroup_rstat_cpu_pop_updated(struct cgroup *pos,
* However, due to the way we traverse, @pos will be the first
* child in most cases. The only exception is @root.
*/
- parent = cgroup_parent(pos);
- if (parent && rstatc->updated_next) {
+ if (rstatc->updated_next) {
+ struct cgroup *parent = cgroup_parent(pos);
struct cgroup_rstat_cpu *prstatc = cgroup_rstat_cpu(parent, cpu);
struct cgroup_rstat_cpu *nrstatc;
struct cgroup **nextp;
@@ -140,9 +139,12 @@ static struct cgroup *cgroup_rstat_cpu_pop_updated(struct cgroup *pos,
* updated stat.
*/
smp_mb();
+
+ return pos;
}
- return pos;
+ /* only happens for @root */
+ return NULL;
}
/* see cgroup_rstat_flush() */
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.