summaryrefslogtreecommitdiffstats
path: root/kernel/cgroup
diff options
context:
space:
mode:
Diffstat (limited to 'kernel/cgroup')
-rw-r--r--kernel/cgroup/cgroup-internal.h4
-rw-r--r--kernel/cgroup/cgroup-v1.c38
-rw-r--r--kernel/cgroup/cgroup.c81
-rw-r--r--kernel/cgroup/rstat.c2
4 files changed, 66 insertions, 59 deletions
diff --git a/kernel/cgroup/cgroup-internal.h b/kernel/cgroup/cgroup-internal.h
index 90d1710fef6c..bfbeabc17a9d 100644
--- a/kernel/cgroup/cgroup-internal.h
+++ b/kernel/cgroup/cgroup-internal.h
@@ -7,7 +7,7 @@
#include <linux/workqueue.h>
#include <linux/list.h>
#include <linux/refcount.h>
-#include <linux/fs_context.h>
+#include <linux/fs_parser.h>
#define TRACE_CGROUP_PATH_LEN 1024
extern spinlock_t trace_cgroup_path_lock;
@@ -265,7 +265,7 @@ extern const struct proc_ns_operations cgroupns_operations;
*/
extern struct cftype cgroup1_base_files[];
extern struct kernfs_syscall_ops cgroup1_kf_syscall_ops;
-extern const struct fs_parameter_description cgroup1_fs_parameters;
+extern const struct fs_parameter_spec cgroup1_fs_parameters[];
int proc_cgroupstats_show(struct seq_file *m, void *v);
bool cgroup1_ssid_disabled(int ssid);
diff --git a/kernel/cgroup/cgroup-v1.c b/kernel/cgroup/cgroup-v1.c
index 09f3a413f6f8..f2d7cea86ffe 100644
--- a/kernel/cgroup/cgroup-v1.c
+++ b/kernel/cgroup/cgroup-v1.c
@@ -18,8 +18,6 @@
#include <trace/events/cgroup.h>
-#define cg_invalf(fc, fmt, ...) invalf(fc, fmt, ## __VA_ARGS__)
-
/*
* pidlists linger the following amount before being destroyed. The goal
* is avoiding frequent destruction in the middle of consecutive read calls
@@ -473,6 +471,7 @@ static void *cgroup_pidlist_next(struct seq_file *s, void *v, loff_t *pos)
*/
p++;
if (p >= end) {
+ (*pos)++;
return NULL;
} else {
*pos = *p;
@@ -784,7 +783,7 @@ void cgroup1_release_agent(struct work_struct *work)
pathbuf = kmalloc(PATH_MAX, GFP_KERNEL);
agentbuf = kstrdup(cgrp->root->release_agent_path, GFP_KERNEL);
- if (!pathbuf || !agentbuf)
+ if (!pathbuf || !agentbuf || !strlen(agentbuf))
goto out;
spin_lock_irq(&css_set_lock);
@@ -887,7 +886,7 @@ enum cgroup1_param {
Opt_xattr,
};
-static const struct fs_parameter_spec cgroup1_param_specs[] = {
+const struct fs_parameter_spec cgroup1_fs_parameters[] = {
fsparam_flag ("all", Opt_all),
fsparam_flag ("clone_children", Opt_clone_children),
fsparam_flag ("cpuset_v2_mode", Opt_cpuset_v2_mode),
@@ -899,11 +898,6 @@ static const struct fs_parameter_spec cgroup1_param_specs[] = {
{}
};
-const struct fs_parameter_description cgroup1_fs_parameters = {
- .name = "cgroup1",
- .specs = cgroup1_param_specs,
-};
-
int cgroup1_parse_param(struct fs_context *fc, struct fs_parameter *param)
{
struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
@@ -911,7 +905,7 @@ int cgroup1_parse_param(struct fs_context *fc, struct fs_parameter *param)
struct fs_parse_result result;
int opt, i;
- opt = fs_parse(fc, &cgroup1_fs_parameters, param, &result);
+ opt = fs_parse(fc, cgroup1_fs_parameters, param, &result);
if (opt == -ENOPARAM) {
if (strcmp(param->key, "source") == 0) {
fc->source = param->string;
@@ -924,7 +918,7 @@ int cgroup1_parse_param(struct fs_context *fc, struct fs_parameter *param)
ctx->subsys_mask |= (1 << i);
return 0;
}
- return cg_invalf(fc, "cgroup1: Unknown subsys name '%s'", param->key);
+ return invalfc(fc, "Unknown subsys name '%s'", param->key);
}
if (opt < 0)
return opt;
@@ -952,7 +946,7 @@ int cgroup1_parse_param(struct fs_context *fc, struct fs_parameter *param)
case Opt_release_agent:
/* Specifying two release agents is forbidden */
if (ctx->release_agent)
- return cg_invalf(fc, "cgroup1: release_agent respecified");
+ return invalfc(fc, "release_agent respecified");
ctx->release_agent = param->string;
param->string = NULL;
break;
@@ -962,9 +956,9 @@ int cgroup1_parse_param(struct fs_context *fc, struct fs_parameter *param)
return -ENOENT;
/* Can't specify an empty name */
if (!param->size)
- return cg_invalf(fc, "cgroup1: Empty name");
+ return invalfc(fc, "Empty name");
if (param->size > MAX_CGROUP_ROOT_NAMELEN - 1)
- return cg_invalf(fc, "cgroup1: Name too long");
+ return invalfc(fc, "Name too long");
/* Must match [\w.-]+ */
for (i = 0; i < param->size; i++) {
char c = param->string[i];
@@ -972,11 +966,11 @@ int cgroup1_parse_param(struct fs_context *fc, struct fs_parameter *param)
continue;
if ((c == '.') || (c == '-') || (c == '_'))
continue;
- return cg_invalf(fc, "cgroup1: Invalid name");
+ return invalfc(fc, "Invalid name");
}
/* Specifying two names is forbidden */
if (ctx->name)
- return cg_invalf(fc, "cgroup1: name respecified");
+ return invalfc(fc, "name respecified");
ctx->name = param->string;
param->string = NULL;
break;
@@ -1011,7 +1005,7 @@ static int check_cgroupfs_options(struct fs_context *fc)
if (ctx->all_ss) {
/* Mutually exclusive option 'all' + subsystem name */
if (ctx->subsys_mask)
- return cg_invalf(fc, "cgroup1: subsys name conflicts with all");
+ return invalfc(fc, "subsys name conflicts with all");
/* 'all' => select all the subsystems */
ctx->subsys_mask = enabled;
}
@@ -1021,7 +1015,7 @@ static int check_cgroupfs_options(struct fs_context *fc)
* empty hierarchies must have a name).
*/
if (!ctx->subsys_mask && !ctx->name)
- return cg_invalf(fc, "cgroup1: Need name or subsystem set");
+ return invalfc(fc, "Need name or subsystem set");
/*
* Option noprefix was introduced just for backward compatibility
@@ -1029,11 +1023,11 @@ static int check_cgroupfs_options(struct fs_context *fc)
* the cpuset subsystem.
*/
if ((ctx->flags & CGRP_ROOT_NOPREFIX) && (ctx->subsys_mask & mask))
- return cg_invalf(fc, "cgroup1: noprefix used incorrectly");
+ return invalfc(fc, "noprefix used incorrectly");
/* Can't specify "none" and some subsystems */
if (ctx->subsys_mask && ctx->none)
- return cg_invalf(fc, "cgroup1: none used incorrectly");
+ return invalfc(fc, "none used incorrectly");
return 0;
}
@@ -1063,7 +1057,7 @@ int cgroup1_reconfigure(struct fs_context *fc)
/* Don't allow flags or name to change at remount */
if ((ctx->flags ^ root->flags) ||
(ctx->name && strcmp(ctx->name, root->name))) {
- cg_invalf(fc, "option or name mismatch, new: 0x%x \"%s\", old: 0x%x \"%s\"",
+ errorfc(fc, "option or name mismatch, new: 0x%x \"%s\", old: 0x%x \"%s\"",
ctx->flags, ctx->name ?: "", root->flags, root->name);
ret = -EINVAL;
goto out_unlock;
@@ -1180,7 +1174,7 @@ static int cgroup1_root_to_use(struct fs_context *fc)
* can't create new one without subsys specification.
*/
if (!ctx->subsys_mask && !ctx->none)
- return cg_invalf(fc, "cgroup1: No subsys list or none specified");
+ return invalfc(fc, "No subsys list or none specified");
/* Hierarchies may only be created in the initial cgroup namespace. */
if (ctx->ns != &init_cgroup_ns)
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 735af8f15f95..7a39dc882095 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -1816,24 +1816,19 @@ enum cgroup2_param {
nr__cgroup2_params
};
-static const struct fs_parameter_spec cgroup2_param_specs[] = {
+static const struct fs_parameter_spec cgroup2_fs_parameters[] = {
fsparam_flag("nsdelegate", Opt_nsdelegate),
fsparam_flag("memory_localevents", Opt_memory_localevents),
{}
};
-static const struct fs_parameter_description cgroup2_fs_parameters = {
- .name = "cgroup2",
- .specs = cgroup2_param_specs,
-};
-
static int cgroup2_parse_param(struct fs_context *fc, struct fs_parameter *param)
{
struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
struct fs_parse_result result;
int opt;
- opt = fs_parse(fc, &cgroup2_fs_parameters, param, &result);
+ opt = fs_parse(fc, cgroup2_fs_parameters, param, &result);
if (opt < 0)
return opt;
@@ -2156,7 +2151,7 @@ static void cgroup_kill_sb(struct super_block *sb)
struct file_system_type cgroup_fs_type = {
.name = "cgroup",
.init_fs_context = cgroup_init_fs_context,
- .parameters = &cgroup1_fs_parameters,
+ .parameters = cgroup1_fs_parameters,
.kill_sb = cgroup_kill_sb,
.fs_flags = FS_USERNS_MOUNT,
};
@@ -2164,7 +2159,7 @@ struct file_system_type cgroup_fs_type = {
static struct file_system_type cgroup2_fs_type = {
.name = "cgroup2",
.init_fs_context = cgroup_init_fs_context,
- .parameters = &cgroup2_fs_parameters,
+ .parameters = cgroup2_fs_parameters,
.kill_sb = cgroup_kill_sb,
.fs_flags = FS_USERNS_MOUNT,
};
@@ -3055,8 +3050,6 @@ static int cgroup_apply_control_enable(struct cgroup *cgrp)
for_each_subsys(ss, ssid) {
struct cgroup_subsys_state *css = cgroup_css(dsct, ss);
- WARN_ON_ONCE(css && percpu_ref_is_dying(&css->refcnt));
-
if (!(cgroup_ss_mask(dsct) & (1 << ss->id)))
continue;
@@ -3066,6 +3059,8 @@ static int cgroup_apply_control_enable(struct cgroup *cgrp)
return PTR_ERR(css);
}
+ WARN_ON_ONCE(percpu_ref_is_dying(&css->refcnt));
+
if (css_visible(css)) {
ret = css_populate_dir(css);
if (ret)
@@ -3101,11 +3096,11 @@ static void cgroup_apply_control_disable(struct cgroup *cgrp)
for_each_subsys(ss, ssid) {
struct cgroup_subsys_state *css = cgroup_css(dsct, ss);
- WARN_ON_ONCE(css && percpu_ref_is_dying(&css->refcnt));
-
if (!css)
continue;
+ WARN_ON_ONCE(percpu_ref_is_dying(&css->refcnt));
+
if (css->parent &&
!(cgroup_ss_mask(dsct) & (1 << ss->id))) {
kill_css(css);
@@ -3392,7 +3387,8 @@ static ssize_t cgroup_type_write(struct kernfs_open_file *of, char *buf,
if (strcmp(strstrip(buf), "threaded"))
return -EINVAL;
- cgrp = cgroup_kn_lock_live(of->kn, false);
+ /* drain dying csses before we re-apply (threaded) subtree control */
+ cgrp = cgroup_kn_lock_live(of->kn, true);
if (!cgrp)
return -ENOENT;
@@ -3546,21 +3542,21 @@ static int cpu_stat_show(struct seq_file *seq, void *v)
static int cgroup_io_pressure_show(struct seq_file *seq, void *v)
{
struct cgroup *cgrp = seq_css(seq)->cgroup;
- struct psi_group *psi = cgroup_id(cgrp) == 1 ? &psi_system : &cgrp->psi;
+ struct psi_group *psi = cgroup_ino(cgrp) == 1 ? &psi_system : &cgrp->psi;
return psi_show(seq, psi, PSI_IO);
}
static int cgroup_memory_pressure_show(struct seq_file *seq, void *v)
{
struct cgroup *cgrp = seq_css(seq)->cgroup;
- struct psi_group *psi = cgroup_id(cgrp) == 1 ? &psi_system : &cgrp->psi;
+ struct psi_group *psi = cgroup_ino(cgrp) == 1 ? &psi_system : &cgrp->psi;
return psi_show(seq, psi, PSI_MEM);
}
static int cgroup_cpu_pressure_show(struct seq_file *seq, void *v)
{
struct cgroup *cgrp = seq_css(seq)->cgroup;
- struct psi_group *psi = cgroup_id(cgrp) == 1 ? &psi_system : &cgrp->psi;
+ struct psi_group *psi = cgroup_ino(cgrp) == 1 ? &psi_system : &cgrp->psi;
return psi_show(seq, psi, PSI_CPU);
}
@@ -4404,12 +4400,16 @@ static void css_task_iter_advance_css_set(struct css_task_iter *it)
}
} while (!css_set_populated(cset) && list_empty(&cset->dying_tasks));
- if (!list_empty(&cset->tasks))
+ if (!list_empty(&cset->tasks)) {
it->task_pos = cset->tasks.next;
- else if (!list_empty(&cset->mg_tasks))
+ it->cur_tasks_head = &cset->tasks;
+ } else if (!list_empty(&cset->mg_tasks)) {
it->task_pos = cset->mg_tasks.next;
- else
+ it->cur_tasks_head = &cset->mg_tasks;
+ } else {
it->task_pos = cset->dying_tasks.next;
+ it->cur_tasks_head = &cset->dying_tasks;
+ }
it->tasks_head = &cset->tasks;
it->mg_tasks_head = &cset->mg_tasks;
@@ -4467,10 +4467,14 @@ repeat:
else
it->task_pos = it->task_pos->next;
- if (it->task_pos == it->tasks_head)
+ if (it->task_pos == it->tasks_head) {
it->task_pos = it->mg_tasks_head->next;
- if (it->task_pos == it->mg_tasks_head)
+ it->cur_tasks_head = it->mg_tasks_head;
+ }
+ if (it->task_pos == it->mg_tasks_head) {
it->task_pos = it->dying_tasks_head->next;
+ it->cur_tasks_head = it->dying_tasks_head;
+ }
if (it->task_pos == it->dying_tasks_head)
css_task_iter_advance_css_set(it);
} else {
@@ -4489,11 +4493,12 @@ repeat:
goto repeat;
/* and dying leaders w/o live member threads */
- if (!atomic_read(&task->signal->live))
+ if (it->cur_tasks_head == it->dying_tasks_head &&
+ !atomic_read(&task->signal->live))
goto repeat;
} else {
/* skip all dying ones */
- if (task->flags & PF_EXITING)
+ if (it->cur_tasks_head == it->dying_tasks_head)
goto repeat;
}
}
@@ -4599,6 +4604,9 @@ static void *cgroup_procs_next(struct seq_file *s, void *v, loff_t *pos)
struct kernfs_open_file *of = s->private;
struct css_task_iter *it = of->priv;
+ if (pos)
+ (*pos)++;
+
return css_task_iter_next(it);
}
@@ -4614,7 +4622,7 @@ static void *__cgroup_procs_start(struct seq_file *s, loff_t *pos,
* from position 0, so we can simply keep iterating on !0 *pos.
*/
if (!it) {
- if (WARN_ON_ONCE((*pos)++))
+ if (WARN_ON_ONCE((*pos)))
return ERR_PTR(-EINVAL);
it = kzalloc(sizeof(*it), GFP_KERNEL);
@@ -4622,10 +4630,11 @@ static void *__cgroup_procs_start(struct seq_file *s, loff_t *pos,
return ERR_PTR(-ENOMEM);
of->priv = it;
css_task_iter_start(&cgrp->self, iter_flags, it);
- } else if (!(*pos)++) {
+ } else if (!(*pos)) {
css_task_iter_end(it);
css_task_iter_start(&cgrp->self, iter_flags, it);
- }
+ } else
+ return it->cur_task;
return cgroup_procs_next(s, NULL, NULL);
}
@@ -5931,11 +5940,14 @@ void cgroup_post_fork(struct task_struct *child)
spin_lock_irq(&css_set_lock);
- WARN_ON_ONCE(!list_empty(&child->cg_list));
- cset = task_css_set(current); /* current is @child's parent */
- get_css_set(cset);
- cset->nr_tasks++;
- css_set_move_task(child, NULL, cset, false);
+ /* init tasks are special, only link regular threads */
+ if (likely(child->pid)) {
+ WARN_ON_ONCE(!list_empty(&child->cg_list));
+ cset = task_css_set(current); /* current is @child's parent */
+ get_css_set(cset);
+ cset->nr_tasks++;
+ css_set_move_task(child, NULL, cset, false);
+ }
/*
* If the cgroup has to be frozen, the new task has too. Let's set
@@ -6288,12 +6300,13 @@ void cgroup_sk_free(struct sock_cgroup_data *skcd)
#ifdef CONFIG_CGROUP_BPF
int cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog,
- enum bpf_attach_type type, u32 flags)
+ struct bpf_prog *replace_prog, enum bpf_attach_type type,
+ u32 flags)
{
int ret;
mutex_lock(&cgroup_mutex);
- ret = __cgroup_bpf_attach(cgrp, prog, type, flags);
+ ret = __cgroup_bpf_attach(cgrp, prog, replace_prog, type, flags);
mutex_unlock(&cgroup_mutex);
return ret;
}
diff --git a/kernel/cgroup/rstat.c b/kernel/cgroup/rstat.c
index b48b22d4deb6..6f87352f8219 100644
--- a/kernel/cgroup/rstat.c
+++ b/kernel/cgroup/rstat.c
@@ -33,7 +33,7 @@ void cgroup_rstat_updated(struct cgroup *cgrp, int cpu)
return;
/*
- * Paired with the one in cgroup_rstat_cpu_pop_upated(). Either we
+ * Paired with the one in cgroup_rstat_cpu_pop_updated(). Either we
* see NULL updated_next or they see our updated stat.
*/
smp_mb();
generated by cgit v1.2.3 (git 2.39.1) at 2024-08-02 13:26:29 +0000