Update Linux to v5.4.2
Change-Id: Idf6911045d9d382da2cfe01b1edff026404ac8fd
diff --git a/kernel/cgroup/Makefile b/kernel/cgroup/Makefile
index bfcdae8..5d7a76b 100644
--- a/kernel/cgroup/Makefile
+++ b/kernel/cgroup/Makefile
@@ -1,7 +1,7 @@
# SPDX-License-Identifier: GPL-2.0
-obj-y := cgroup.o rstat.o namespace.o cgroup-v1.o
+obj-y := cgroup.o rstat.o namespace.o cgroup-v1.o freezer.o
-obj-$(CONFIG_CGROUP_FREEZER) += freezer.o
+obj-$(CONFIG_CGROUP_FREEZER) += legacy_freezer.o
obj-$(CONFIG_CGROUP_PIDS) += pids.o
obj-$(CONFIG_CGROUP_RDMA) += rdma.o
obj-$(CONFIG_CPUSETS) += cpuset.o
diff --git a/kernel/cgroup/cgroup-internal.h b/kernel/cgroup/cgroup-internal.h
index 75568fc..809e34a 100644
--- a/kernel/cgroup/cgroup-internal.h
+++ b/kernel/cgroup/cgroup-internal.h
@@ -7,10 +7,13 @@
#include <linux/workqueue.h>
#include <linux/list.h>
#include <linux/refcount.h>
+#include <linux/fs_context.h>
#define TRACE_CGROUP_PATH_LEN 1024
extern spinlock_t trace_cgroup_path_lock;
extern char trace_cgroup_path[TRACE_CGROUP_PATH_LEN];
+extern bool cgroup_debug;
+extern void __init enable_debug_cgroup(void);
/*
* cgroup_path() takes a spin lock. It is good practice not to take
@@ -25,16 +28,44 @@
#define TRACE_CGROUP_PATH(type, cgrp, ...) \
do { \
if (trace_cgroup_##type##_enabled()) { \
- spin_lock(&trace_cgroup_path_lock); \
+ unsigned long flags; \
+ spin_lock_irqsave(&trace_cgroup_path_lock, \
+ flags); \
cgroup_path(cgrp, trace_cgroup_path, \
TRACE_CGROUP_PATH_LEN); \
trace_cgroup_##type(cgrp, trace_cgroup_path, \
##__VA_ARGS__); \
- spin_unlock(&trace_cgroup_path_lock); \
+ spin_unlock_irqrestore(&trace_cgroup_path_lock, \
+ flags); \
} \
} while (0)
/*
+ * The cgroup filesystem superblock creation/mount context.
+ */
+struct cgroup_fs_context {
+ struct kernfs_fs_context kfc;
+ struct cgroup_root *root;
+ struct cgroup_namespace *ns;
+ unsigned int flags; /* CGRP_ROOT_* flags */
+
+ /* cgroup1 bits */
+ bool cpuset_clone_children;
+ bool none; /* User explicitly requested empty subsystem */
+ bool all_ss; /* Seen 'all' option */
+ u16 subsys_mask; /* Selected subsystems */
+ char *name; /* Hierarchy name */
+ char *release_agent; /* Path for release notifications */
+};
+
+static inline struct cgroup_fs_context *cgroup_fc2context(struct fs_context *fc)
+{
+ struct kernfs_fs_context *kfc = fc->fs_private;
+
+ return container_of(kfc, struct cgroup_fs_context, kfc);
+}
+
+/*
* A cgroup can be associated with multiple css_sets as different tasks may
* belong to different cgroups on different hierarchies. In the other
* direction, a css_set is naturally associated with multiple cgroups.
@@ -115,16 +146,6 @@
#define DEFINE_CGROUP_MGCTX(name) \
struct cgroup_mgctx name = CGROUP_MGCTX_INIT(name)
-struct cgroup_sb_opts {
- u16 subsys_mask;
- unsigned int flags;
- char *release_agent;
- bool cpuset_clone_children;
- char *name;
- /* User explicitly requested empty subsystem */
- bool none;
-};
-
extern struct mutex cgroup_mutex;
extern spinlock_t css_set_lock;
extern struct cgroup_subsys *cgroup_subsys[];
@@ -195,12 +216,10 @@
struct cgroup_namespace *ns);
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);
+void init_cgroup_root(struct cgroup_fs_context *ctx);
+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,
- struct cgroup_namespace *ns);
+int cgroup_do_get_tree(struct fs_context *fc);
int cgroup_migrate_vet_dst(struct cgroup *dst_cgrp);
void cgroup_migrate_finish(struct cgroup_mgctx *mgctx);
@@ -224,6 +243,7 @@
int cgroup_show_path(struct seq_file *sf, struct kernfs_node *kf_node,
struct kernfs_root *kf_root);
+int __cgroup_task_count(const struct cgroup *cgrp);
int cgroup_task_count(const struct cgroup *cgrp);
/*
@@ -244,14 +264,15 @@
*/
extern struct cftype cgroup1_base_files[];
extern struct kernfs_syscall_ops cgroup1_kf_syscall_ops;
+extern const struct fs_parameter_description cgroup1_fs_parameters;
int proc_cgroupstats_show(struct seq_file *m, void *v);
bool cgroup1_ssid_disabled(int ssid);
void cgroup1_pidlist_destroy_all(struct cgroup *cgrp);
void cgroup1_release_agent(struct work_struct *work);
void cgroup1_check_for_release(struct cgroup *cgrp);
-struct dentry *cgroup1_mount(struct file_system_type *fs_type, int flags,
- void *data, unsigned long magic,
- struct cgroup_namespace *ns);
+int cgroup1_parse_param(struct fs_context *fc, struct fs_parameter *param);
+int cgroup1_get_tree(struct fs_context *fc);
+int cgroup1_reconfigure(struct fs_context *ctx);
#endif /* __CGROUP_INTERNAL_H */
diff --git a/kernel/cgroup/cgroup-v1.c b/kernel/cgroup/cgroup-v1.c
index 51063e7..7f83f41 100644
--- a/kernel/cgroup/cgroup-v1.c
+++ b/kernel/cgroup/cgroup-v1.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
#include "cgroup-internal.h"
#include <linux/ctype.h>
@@ -13,9 +14,12 @@
#include <linux/delayacct.h>
#include <linux/pid_namespace.h>
#include <linux/cgroupstats.h>
+#include <linux/fs_parser.h>
#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
@@ -27,6 +31,9 @@
/* Controllers blocked by the commandline in v1 */
static u16 cgroup_no_v1_mask;
+/* disable named v1 mounts */
+static bool cgroup_no_v1_named;
+
/*
* pidlist destructions need to be flushed on cgroup destruction. Use a
* separate workqueue as flush domain.
@@ -187,25 +194,6 @@
};
/*
- * The following two functions "fix" the issue where there are more pids
- * than kmalloc will give memory for; in such cases, we use vmalloc/vfree.
- * TODO: replace with a kernel-wide solution to this problem
- */
-#define PIDLIST_TOO_LARGE(c) ((c) * sizeof(pid_t) > (PAGE_SIZE * 2))
-static void *pidlist_allocate(int count)
-{
- if (PIDLIST_TOO_LARGE(count))
- return vmalloc(array_size(count, sizeof(pid_t)));
- else
- return kmalloc_array(count, sizeof(pid_t), GFP_KERNEL);
-}
-
-static void pidlist_free(void *p)
-{
- kvfree(p);
-}
-
-/*
* Used to destroy all pidlists lingering waiting for destroy timer. None
* should be left afterwards.
*/
@@ -237,7 +225,7 @@
*/
if (!delayed_work_pending(dwork)) {
list_del(&l->links);
- pidlist_free(l->list);
+ kvfree(l->list);
put_pid_ns(l->key.ns);
tofree = l;
}
@@ -336,22 +324,6 @@
return l;
}
-/**
- * cgroup_task_count - count the number of tasks in a cgroup.
- * @cgrp: the cgroup in question
- */
-int cgroup_task_count(const struct cgroup *cgrp)
-{
- int count = 0;
- struct cgrp_cset_link *link;
-
- spin_lock_irq(&css_set_lock);
- list_for_each_entry(link, &cgrp->cset_links, cset_link)
- count += link->cset->nr_tasks;
- spin_unlock_irq(&css_set_lock);
- return count;
-}
-
/*
* Load a cgroup's pidarray with either procs' tgids or tasks' pids
*/
@@ -374,7 +346,7 @@
* show up until sometime later on.
*/
length = cgroup_task_count(cgrp);
- array = pidlist_allocate(length);
+ array = kvmalloc_array(length, sizeof(pid_t), GFP_KERNEL);
if (!array)
return -ENOMEM;
/* now, populate the array */
@@ -399,12 +371,12 @@
l = cgroup_pidlist_find_create(cgrp, type);
if (!l) {
- pidlist_free(array);
+ kvfree(array);
return -ENOMEM;
}
/* store array, freeing old if necessary */
- pidlist_free(l->list);
+ kvfree(l->list);
l->list = array;
l->length = length;
*lp = l;
@@ -903,168 +875,195 @@
return 0;
}
-static int parse_cgroupfs_options(char *data, struct cgroup_sb_opts *opts)
+enum cgroup1_param {
+ Opt_all,
+ Opt_clone_children,
+ Opt_cpuset_v2_mode,
+ Opt_name,
+ Opt_none,
+ Opt_noprefix,
+ Opt_release_agent,
+ Opt_xattr,
+};
+
+static const struct fs_parameter_spec cgroup1_param_specs[] = {
+ fsparam_flag ("all", Opt_all),
+ fsparam_flag ("clone_children", Opt_clone_children),
+ fsparam_flag ("cpuset_v2_mode", Opt_cpuset_v2_mode),
+ fsparam_string("name", Opt_name),
+ fsparam_flag ("none", Opt_none),
+ fsparam_flag ("noprefix", Opt_noprefix),
+ fsparam_string("release_agent", Opt_release_agent),
+ fsparam_flag ("xattr", Opt_xattr),
+ {}
+};
+
+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)
{
- char *token, *o = data;
- bool all_ss = false, one_ss = false;
- u16 mask = U16_MAX;
+ struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
struct cgroup_subsys *ss;
- int nr_opts = 0;
+ struct fs_parse_result result;
+ int opt, i;
+
+ opt = fs_parse(fc, &cgroup1_fs_parameters, param, &result);
+ if (opt == -ENOPARAM) {
+ if (strcmp(param->key, "source") == 0) {
+ fc->source = param->string;
+ param->string = NULL;
+ return 0;
+ }
+ for_each_subsys(ss, i) {
+ if (strcmp(param->key, ss->legacy_name))
+ continue;
+ ctx->subsys_mask |= (1 << i);
+ return 0;
+ }
+ return cg_invalf(fc, "cgroup1: Unknown subsys name '%s'", param->key);
+ }
+ if (opt < 0)
+ return opt;
+
+ switch (opt) {
+ case Opt_none:
+ /* Explicitly have no subsystems */
+ ctx->none = true;
+ break;
+ case Opt_all:
+ ctx->all_ss = true;
+ break;
+ case Opt_noprefix:
+ ctx->flags |= CGRP_ROOT_NOPREFIX;
+ break;
+ case Opt_clone_children:
+ ctx->cpuset_clone_children = true;
+ break;
+ case Opt_cpuset_v2_mode:
+ ctx->flags |= CGRP_ROOT_CPUSET_V2_MODE;
+ break;
+ case Opt_xattr:
+ ctx->flags |= CGRP_ROOT_XATTR;
+ break;
+ case Opt_release_agent:
+ /* Specifying two release agents is forbidden */
+ if (ctx->release_agent)
+ return cg_invalf(fc, "cgroup1: release_agent respecified");
+ ctx->release_agent = param->string;
+ param->string = NULL;
+ break;
+ case Opt_name:
+ /* blocked by boot param? */
+ if (cgroup_no_v1_named)
+ return -ENOENT;
+ /* Can't specify an empty name */
+ if (!param->size)
+ return cg_invalf(fc, "cgroup1: Empty name");
+ if (param->size > MAX_CGROUP_ROOT_NAMELEN - 1)
+ return cg_invalf(fc, "cgroup1: Name too long");
+ /* Must match [\w.-]+ */
+ for (i = 0; i < param->size; i++) {
+ char c = param->string[i];
+ if (isalnum(c))
+ continue;
+ if ((c == '.') || (c == '-') || (c == '_'))
+ continue;
+ return cg_invalf(fc, "cgroup1: Invalid name");
+ }
+ /* Specifying two names is forbidden */
+ if (ctx->name)
+ return cg_invalf(fc, "cgroup1: name respecified");
+ ctx->name = param->string;
+ param->string = NULL;
+ break;
+ }
+ return 0;
+}
+
+static int check_cgroupfs_options(struct fs_context *fc)
+{
+ struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
+ u16 mask = U16_MAX;
+ u16 enabled = 0;
+ struct cgroup_subsys *ss;
int i;
#ifdef CONFIG_CPUSETS
mask = ~((u16)1 << cpuset_cgrp_id);
#endif
+ for_each_subsys(ss, i)
+ if (cgroup_ssid_enabled(i) && !cgroup1_ssid_disabled(i))
+ enabled |= 1 << i;
- memset(opts, 0, sizeof(*opts));
-
- while ((token = strsep(&o, ",")) != NULL) {
- nr_opts++;
-
- if (!*token)
- return -EINVAL;
- if (!strcmp(token, "none")) {
- /* Explicitly have no subsystems */
- opts->none = true;
- continue;
- }
- if (!strcmp(token, "all")) {
- /* Mutually exclusive option 'all' + subsystem name */
- if (one_ss)
- return -EINVAL;
- all_ss = true;
- continue;
- }
- if (!strcmp(token, "noprefix")) {
- opts->flags |= CGRP_ROOT_NOPREFIX;
- continue;
- }
- if (!strcmp(token, "clone_children")) {
- opts->cpuset_clone_children = true;
- continue;
- }
- if (!strcmp(token, "cpuset_v2_mode")) {
- opts->flags |= CGRP_ROOT_CPUSET_V2_MODE;
- continue;
- }
- if (!strcmp(token, "xattr")) {
- opts->flags |= CGRP_ROOT_XATTR;
- continue;
- }
- if (!strncmp(token, "release_agent=", 14)) {
- /* Specifying two release agents is forbidden */
- if (opts->release_agent)
- return -EINVAL;
- opts->release_agent =
- kstrndup(token + 14, PATH_MAX - 1, GFP_KERNEL);
- if (!opts->release_agent)
- return -ENOMEM;
- continue;
- }
- if (!strncmp(token, "name=", 5)) {
- const char *name = token + 5;
- /* Can't specify an empty name */
- if (!strlen(name))
- return -EINVAL;
- /* Must match [\w.-]+ */
- for (i = 0; i < strlen(name); i++) {
- char c = name[i];
- if (isalnum(c))
- continue;
- if ((c == '.') || (c == '-') || (c == '_'))
- continue;
- return -EINVAL;
- }
- /* Specifying two names is forbidden */
- if (opts->name)
- return -EINVAL;
- opts->name = kstrndup(name,
- MAX_CGROUP_ROOT_NAMELEN - 1,
- GFP_KERNEL);
- if (!opts->name)
- return -ENOMEM;
-
- continue;
- }
-
- for_each_subsys(ss, i) {
- if (strcmp(token, ss->legacy_name))
- continue;
- if (!cgroup_ssid_enabled(i))
- continue;
- if (cgroup1_ssid_disabled(i))
- continue;
-
- /* Mutually exclusive option 'all' + subsystem name */
- if (all_ss)
- return -EINVAL;
- opts->subsys_mask |= (1 << i);
- one_ss = true;
-
- break;
- }
- if (i == CGROUP_SUBSYS_COUNT)
- return -ENOENT;
- }
+ ctx->subsys_mask &= enabled;
/*
- * If the 'all' option was specified select all the subsystems,
- * otherwise if 'none', 'name=' and a subsystem name options were
- * not specified, let's default to 'all'
+ * In absense of 'none', 'name=' or subsystem name options,
+ * let's default to 'all'.
*/
- if (all_ss || (!one_ss && !opts->none && !opts->name))
- for_each_subsys(ss, i)
- if (cgroup_ssid_enabled(i) && !cgroup1_ssid_disabled(i))
- opts->subsys_mask |= (1 << i);
+ if (!ctx->subsys_mask && !ctx->none && !ctx->name)
+ ctx->all_ss = true;
+
+ if (ctx->all_ss) {
+ /* Mutually exclusive option 'all' + subsystem name */
+ if (ctx->subsys_mask)
+ return cg_invalf(fc, "cgroup1: subsys name conflicts with all");
+ /* 'all' => select all the subsystems */
+ ctx->subsys_mask = enabled;
+ }
/*
* We either have to specify by name or by subsystems. (So all
* empty hierarchies must have a name).
*/
- if (!opts->subsys_mask && !opts->name)
- return -EINVAL;
+ if (!ctx->subsys_mask && !ctx->name)
+ return cg_invalf(fc, "cgroup1: Need name or subsystem set");
/*
* Option noprefix was introduced just for backward compatibility
* with the old cpuset, so we allow noprefix only if mounting just
* the cpuset subsystem.
*/
- if ((opts->flags & CGRP_ROOT_NOPREFIX) && (opts->subsys_mask & mask))
- return -EINVAL;
+ if ((ctx->flags & CGRP_ROOT_NOPREFIX) && (ctx->subsys_mask & mask))
+ return cg_invalf(fc, "cgroup1: noprefix used incorrectly");
/* Can't specify "none" and some subsystems */
- if (opts->subsys_mask && opts->none)
- return -EINVAL;
+ if (ctx->subsys_mask && ctx->none)
+ return cg_invalf(fc, "cgroup1: none used incorrectly");
return 0;
}
-static int cgroup1_remount(struct kernfs_root *kf_root, int *flags, char *data)
+int cgroup1_reconfigure(struct fs_context *fc)
{
- int ret = 0;
+ struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
+ struct kernfs_root *kf_root = kernfs_root_from_sb(fc->root->d_sb);
struct cgroup_root *root = cgroup_root_from_kf(kf_root);
- struct cgroup_sb_opts opts;
+ int ret = 0;
u16 added_mask, removed_mask;
cgroup_lock_and_drain_offline(&cgrp_dfl_root.cgrp);
/* See what subsystems are wanted */
- ret = parse_cgroupfs_options(data, &opts);
+ ret = check_cgroupfs_options(fc);
if (ret)
goto out_unlock;
- if (opts.subsys_mask != root->subsys_mask || opts.release_agent)
+ if (ctx->subsys_mask != root->subsys_mask || ctx->release_agent)
pr_warn("option changes via remount are deprecated (pid=%d comm=%s)\n",
task_tgid_nr(current), current->comm);
- added_mask = opts.subsys_mask & ~root->subsys_mask;
- removed_mask = root->subsys_mask & ~opts.subsys_mask;
+ added_mask = ctx->subsys_mask & ~root->subsys_mask;
+ removed_mask = root->subsys_mask & ~ctx->subsys_mask;
/* Don't allow flags or name to change at remount */
- if ((opts.flags ^ root->flags) ||
- (opts.name && strcmp(opts.name, root->name))) {
- pr_err("option or name mismatch, new: 0x%x \"%s\", old: 0x%x \"%s\"\n",
- opts.flags, opts.name ?: "", root->flags, root->name);
+ 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\"",
+ ctx->flags, ctx->name ?: "", root->flags, root->name);
ret = -EINVAL;
goto out_unlock;
}
@@ -1081,17 +1080,15 @@
WARN_ON(rebind_subsystems(&cgrp_dfl_root, removed_mask));
- if (opts.release_agent) {
+ if (ctx->release_agent) {
spin_lock(&release_agent_path_lock);
- strcpy(root->release_agent_path, opts.release_agent);
+ strcpy(root->release_agent_path, ctx->release_agent);
spin_unlock(&release_agent_path_lock);
}
trace_cgroup_remount(root);
out_unlock:
- kfree(opts.release_agent);
- kfree(opts.name);
mutex_unlock(&cgroup_mutex);
return ret;
}
@@ -1099,30 +1096,30 @@
struct kernfs_syscall_ops cgroup1_kf_syscall_ops = {
.rename = cgroup1_rename,
.show_options = cgroup1_show_options,
- .remount_fs = cgroup1_remount,
.mkdir = cgroup_mkdir,
.rmdir = cgroup_rmdir,
.show_path = cgroup_show_path,
};
-struct dentry *cgroup1_mount(struct file_system_type *fs_type, int flags,
- void *data, unsigned long magic,
- struct cgroup_namespace *ns)
+/*
+ * The guts of cgroup1 mount - find or create cgroup_root to use.
+ * Called with cgroup_mutex held; returns 0 on success, -E... on
+ * error and positive - in case when the candidate is busy dying.
+ * On success it stashes a reference to cgroup_root into given
+ * cgroup_fs_context; that reference is *NOT* counting towards the
+ * cgroup_root refcount.
+ */
+static int cgroup1_root_to_use(struct fs_context *fc)
{
- struct super_block *pinned_sb = NULL;
- struct cgroup_sb_opts opts;
+ struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
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);
/* First find the desired set of subsystems */
- ret = parse_cgroupfs_options(data, &opts);
+ ret = check_cgroupfs_options(fc);
if (ret)
- goto out_unlock;
+ return ret;
/*
* Destruction of cgroup root is asynchronous, so subsystems may
@@ -1132,16 +1129,12 @@
* starting. Testing ref liveliness is good enough.
*/
for_each_subsys(ss, i) {
- if (!(opts.subsys_mask & (1 << i)) ||
+ if (!(ctx->subsys_mask & (1 << i)) ||
ss->root == &cgrp_dfl_root)
continue;
- if (!percpu_ref_tryget_live(&ss->root->cgrp.self.refcnt)) {
- mutex_unlock(&cgroup_mutex);
- msleep(10);
- ret = restart_syscall();
- goto out_free;
- }
+ if (!percpu_ref_tryget_live(&ss->root->cgrp.self.refcnt))
+ return 1; /* restart */
cgroup_put(&ss->root->cgrp);
}
@@ -1156,8 +1149,8 @@
* name matches but sybsys_mask doesn't, we should fail.
* Remember whether name matched.
*/
- if (opts.name) {
- if (strcmp(opts.name, root->name))
+ if (ctx->name) {
+ if (strcmp(ctx->name, root->name))
continue;
name_match = true;
}
@@ -1166,42 +1159,18 @@
* If we asked for subsystems (or explicitly for no
* subsystems) then they must match.
*/
- if ((opts.subsys_mask || opts.none) &&
- (opts.subsys_mask != root->subsys_mask)) {
+ if ((ctx->subsys_mask || ctx->none) &&
+ (ctx->subsys_mask != root->subsys_mask)) {
if (!name_match)
continue;
- ret = -EBUSY;
- goto out_unlock;
+ return -EBUSY;
}
- if (root->flags ^ opts.flags)
+ if (root->flags ^ ctx->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;
+ ctx->root = root;
+ return 0;
}
/*
@@ -1209,62 +1178,58 @@
* specification is allowed for already existing hierarchies but we
* can't create new one without subsys specification.
*/
- if (!opts.subsys_mask && !opts.none) {
- ret = -EINVAL;
- goto out_unlock;
- }
+ if (!ctx->subsys_mask && !ctx->none)
+ return cg_invalf(fc, "cgroup1: No subsys list or none specified");
/* Hierarchies may only be created in the initial cgroup namespace. */
- if (ns != &init_cgroup_ns) {
- ret = -EPERM;
- goto out_unlock;
- }
+ if (ctx->ns != &init_cgroup_ns)
+ return -EPERM;
root = kzalloc(sizeof(*root), GFP_KERNEL);
- if (!root) {
- ret = -ENOMEM;
- goto out_unlock;
- }
- new_root = true;
+ if (!root)
+ return -ENOMEM;
- init_cgroup_root(root, &opts);
+ ctx->root = root;
+ init_cgroup_root(ctx);
- ret = cgroup_setup_root(root, opts.subsys_mask, PERCPU_REF_INIT_DEAD);
+ ret = cgroup_setup_root(root, ctx->subsys_mask);
if (ret)
cgroup_free_root(root);
+ return ret;
+}
-out_unlock:
+int cgroup1_get_tree(struct fs_context *fc)
+{
+ struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
+ int ret;
+
+ /* Check if the caller has permission to mount. */
+ if (!ns_capable(ctx->ns->user_ns, CAP_SYS_ADMIN))
+ return -EPERM;
+
+ cgroup_lock_and_drain_offline(&cgrp_dfl_root.cgrp);
+
+ ret = cgroup1_root_to_use(fc);
+ if (!ret && !percpu_ref_tryget_live(&ctx->root->cgrp.self.refcnt))
+ ret = 1; /* restart */
+
mutex_unlock(&cgroup_mutex);
-out_free:
- kfree(opts.release_agent);
- kfree(opts.name);
- if (ret)
- return ERR_PTR(ret);
+ if (!ret)
+ ret = cgroup_do_get_tree(fc);
- 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 (!ret && percpu_ref_is_dying(&ctx->root->cgrp.self.refcnt)) {
+ struct super_block *sb = fc->root->d_sb;
+ dput(fc->root);
+ deactivate_locked_super(sb);
+ ret = 1;
}
- /*
- * 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;
+ if (unlikely(ret > 0)) {
+ msleep(10);
+ return restart_syscall();
+ }
+ return ret;
}
static int __init cgroup1_wq_init(void)
@@ -1292,7 +1257,12 @@
if (!strcmp(token, "all")) {
cgroup_no_v1_mask = U16_MAX;
- break;
+ continue;
+ }
+
+ if (!strcmp(token, "named")) {
+ cgroup_no_v1_named = true;
+ continue;
}
for_each_subsys(ss, i) {
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 4a3dae2..ef4242e 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -54,7 +54,9 @@
#include <linux/proc_ns.h>
#include <linux/nsproxy.h>
#include <linux/file.h>
+#include <linux/fs_parser.h>
#include <linux/sched/cputime.h>
+#include <linux/psi.h>
#include <net/sock.h>
#define CREATE_TRACE_POINTS
@@ -85,6 +87,7 @@
DEFINE_SPINLOCK(trace_cgroup_path_lock);
char trace_cgroup_path[TRACE_CGROUP_PATH_LEN];
+bool cgroup_debug __read_mostly;
/*
* Protects cgroup_idr and css_idr so that IDs can be released without
@@ -98,7 +101,7 @@
*/
static DEFINE_SPINLOCK(cgroup_file_kn_lock);
-struct percpu_rw_semaphore cgroup_threadgroup_rwsem;
+DEFINE_PERCPU_RWSEM(cgroup_threadgroup_rwsem);
#define cgroup_assert_mutex_or_rcu_locked() \
RCU_LOCKDEP_WARN(!rcu_read_lock_held() && \
@@ -195,7 +198,7 @@
*/
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 */
@@ -212,7 +215,8 @@
static int cgroup_apply_control(struct cgroup *cgrp);
static void cgroup_finalize_control(struct cgroup *cgrp, int ret);
-static void css_task_iter_advance(struct css_task_iter *it);
+static void css_task_iter_skip(struct css_task_iter *it,
+ struct task_struct *task);
static int cgroup_destroy_locked(struct cgroup *cgrp);
static struct cgroup_subsys_state *css_create(struct cgroup *cgrp,
struct cgroup_subsys *ss);
@@ -484,7 +488,7 @@
rcu_read_lock();
css = cgroup_css(cgrp, ss);
- if (!css || !css_tryget_online(css))
+ if (css && !css_tryget_online(css))
css = NULL;
rcu_read_unlock();
@@ -492,7 +496,7 @@
}
/**
- * cgroup_e_css - obtain a cgroup's effective css for the specified subsystem
+ * cgroup_e_css_by_mask - obtain a cgroup's effective css for the specified ss
* @cgrp: the cgroup of interest
* @ss: the subsystem of interest (%NULL returns @cgrp->self)
*
@@ -501,8 +505,8 @@
* enabled. If @ss is associated with the hierarchy @cgrp is on, this
* function is guaranteed to return non-NULL css.
*/
-static struct cgroup_subsys_state *cgroup_e_css(struct cgroup *cgrp,
- struct cgroup_subsys *ss)
+static struct cgroup_subsys_state *cgroup_e_css_by_mask(struct cgroup *cgrp,
+ struct cgroup_subsys *ss)
{
lockdep_assert_held(&cgroup_mutex);
@@ -523,6 +527,35 @@
}
/**
+ * cgroup_e_css - obtain a cgroup's effective css for the specified subsystem
+ * @cgrp: the cgroup of interest
+ * @ss: the subsystem of interest
+ *
+ * Find and get the effective css of @cgrp for @ss. The effective css is
+ * defined as the matching css of the nearest ancestor including self which
+ * has @ss enabled. If @ss is not mounted on the hierarchy @cgrp is on,
+ * the root css is returned, so this function always returns a valid css.
+ *
+ * The returned css is not guaranteed to be online, and therefore it is the
+ * callers responsiblity to tryget a reference for it.
+ */
+struct cgroup_subsys_state *cgroup_e_css(struct cgroup *cgrp,
+ struct cgroup_subsys *ss)
+{
+ struct cgroup_subsys_state *css;
+
+ do {
+ css = cgroup_css(cgrp, ss);
+
+ if (css)
+ return css;
+ cgrp = cgroup_parent(cgrp);
+ } while (cgrp);
+
+ return init_css_set.subsys[ss->id];
+}
+
+/**
* cgroup_get_e_css - get a cgroup's effective css for the specified subsystem
* @cgrp: the cgroup of interest
* @ss: the subsystem of interest
@@ -561,6 +594,39 @@
css_get(&cgrp->self);
}
+/**
+ * __cgroup_task_count - count the number of tasks in a cgroup. The caller
+ * is responsible for taking the css_set_lock.
+ * @cgrp: the cgroup in question
+ */
+int __cgroup_task_count(const struct cgroup *cgrp)
+{
+ int count = 0;
+ struct cgrp_cset_link *link;
+
+ lockdep_assert_held(&css_set_lock);
+
+ list_for_each_entry(link, &cgrp->cset_links, cset_link)
+ count += link->cset->nr_tasks;
+
+ return count;
+}
+
+/**
+ * cgroup_task_count - count the number of tasks in a cgroup.
+ * @cgrp: the cgroup in question
+ */
+int cgroup_task_count(const struct cgroup *cgrp)
+{
+ int count;
+
+ spin_lock_irq(&css_set_lock);
+ count = __cgroup_task_count(cgrp);
+ spin_unlock_irq(&css_set_lock);
+
+ return count;
+}
+
struct cgroup_subsys_state *of_css(struct kernfs_open_file *of)
{
struct cgroup *cgrp = of->kn->parent->priv;
@@ -604,10 +670,11 @@
*
* Should be called under cgroup_[tree_]mutex.
*/
-#define for_each_e_css(css, ssid, cgrp) \
- for ((ssid) = 0; (ssid) < CGROUP_SUBSYS_COUNT; (ssid)++) \
- if (!((css) = cgroup_e_css(cgrp, cgroup_subsys[(ssid)]))) \
- ; \
+#define for_each_e_css(css, ssid, cgrp) \
+ for ((ssid) = 0; (ssid) < CGROUP_SUBSYS_COUNT; (ssid)++) \
+ if (!((css) = cgroup_e_css_by_mask(cgrp, \
+ cgroup_subsys[(ssid)]))) \
+ ; \
else
/**
@@ -672,6 +739,7 @@
.dom_cset = &init_css_set,
.tasks = LIST_HEAD_INIT(init_css_set.tasks),
.mg_tasks = LIST_HEAD_INIT(init_css_set.mg_tasks),
+ .dying_tasks = LIST_HEAD_INIT(init_css_set.dying_tasks),
.task_iters = LIST_HEAD_INIT(init_css_set.task_iters),
.threaded_csets = LIST_HEAD_INIT(init_css_set.threaded_csets),
.cgrp_links = LIST_HEAD_INIT(init_css_set.cgrp_links),
@@ -750,6 +818,8 @@
break;
cgroup1_check_for_release(cgrp);
+ TRACE_CGROUP_PATH(notify_populated, cgrp,
+ cgroup_is_populated(cgrp));
cgroup_file_notify(&cgrp->events_file);
child = cgrp;
@@ -775,6 +845,21 @@
cgroup_update_populated(link->cgrp, populated);
}
+/*
+ * @task is leaving, advance task iterators which are pointing to it so
+ * that they can resume at the next position. Advancing an iterator might
+ * remove it from the list, use safe walk. See css_task_iter_skip() for
+ * details.
+ */
+static void css_set_skip_task_iters(struct css_set *cset,
+ struct task_struct *task)
+{
+ struct css_task_iter *it, *pos;
+
+ list_for_each_entry_safe(it, pos, &cset->task_iters, iters_node)
+ css_task_iter_skip(it, task);
+}
+
/**
* css_set_move_task - move a task from one css_set to another
* @task: task being moved
@@ -800,22 +885,9 @@
css_set_update_populated(to_cset, true);
if (from_cset) {
- struct css_task_iter *it, *pos;
-
WARN_ON_ONCE(list_empty(&task->cg_list));
- /*
- * @task is leaving, advance task iterators which are
- * pointing to it so that they can resume at the next
- * position. Advancing an iterator might remove it from
- * the list, use safe walk. See css_task_iter_advance*()
- * for details.
- */
- list_for_each_entry_safe(it, pos, &from_cset->task_iters,
- iters_node)
- if (it->task_pos == &task->cg_list)
- css_task_iter_advance(it);
-
+ css_set_skip_task_iters(from_cset, task);
list_del_init(&task->cg_list);
if (!css_set_populated(from_cset))
css_set_update_populated(from_cset, false);
@@ -832,7 +904,7 @@
*/
WARN_ON_ONCE(task->flags & PF_EXITING);
- rcu_assign_pointer(task->cgroups, to_cset);
+ cgroup_move_task(task, to_cset);
list_add_tail(&task->cg_list, use_mg_tasks ? &to_cset->mg_tasks :
&to_cset->tasks);
}
@@ -1006,7 +1078,7 @@
* @ss is in this hierarchy, so we want the
* effective css from @cgrp.
*/
- template[i] = cgroup_e_css(cgrp, ss);
+ template[i] = cgroup_e_css_by_mask(cgrp, ss);
} else {
/*
* @ss is not in this hierarchy, so we don't want
@@ -1142,6 +1214,7 @@
cset->dom_cset = cset;
INIT_LIST_HEAD(&cset->tasks);
INIT_LIST_HEAD(&cset->mg_tasks);
+ INIT_LIST_HEAD(&cset->dying_tasks);
INIT_LIST_HEAD(&cset->task_iters);
INIT_LIST_HEAD(&cset->threaded_csets);
INIT_HLIST_NODE(&cset->hlist);
@@ -1398,12 +1471,15 @@
struct cgroup_subsys *ss = cft->ss;
if (cft->ss && !(cft->flags & CFTYPE_NO_PREFIX) &&
- !(cgrp->root->flags & CGRP_ROOT_NOPREFIX))
- snprintf(buf, CGROUP_FILE_NAME_MAX, "%s.%s",
- cgroup_on_dfl(cgrp) ? ss->name : ss->legacy_name,
+ !(cgrp->root->flags & CGRP_ROOT_NOPREFIX)) {
+ const char *dbg = (cft->flags & CFTYPE_DEBUG) ? ".__DEBUG__." : "";
+
+ snprintf(buf, CGROUP_FILE_NAME_MAX, "%s%s.%s",
+ dbg, cgroup_on_dfl(cgrp) ? ss->name : ss->legacy_name,
cft->name);
- else
+ } else {
strscpy(buf, cft->name, CGROUP_FILE_NAME_MAX);
+ }
return buf;
}
@@ -1737,26 +1813,42 @@
return len;
}
-static int parse_cgroup_root_flags(char *data, unsigned int *root_flags)
+enum cgroup2_param {
+ Opt_nsdelegate,
+ Opt_memory_localevents,
+ nr__cgroup2_params
+};
+
+static const struct fs_parameter_spec cgroup2_param_specs[] = {
+ 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)
{
- char *token;
+ struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
+ struct fs_parse_result result;
+ int opt;
- *root_flags = 0;
+ opt = fs_parse(fc, &cgroup2_fs_parameters, param, &result);
+ if (opt < 0)
+ return opt;
- if (!data)
+ switch (opt) {
+ case Opt_nsdelegate:
+ ctx->flags |= CGRP_ROOT_NS_DELEGATE;
return 0;
-
- while ((token = strsep(&data, ",")) != NULL) {
- if (!strcmp(token, "nsdelegate")) {
- *root_flags |= CGRP_ROOT_NS_DELEGATE;
- continue;
- }
-
- pr_err("cgroup2: unknown option \"%s\"\n", token);
- return -EINVAL;
+ case Opt_memory_localevents:
+ ctx->flags |= CGRP_ROOT_MEMORY_LOCAL_EVENTS;
+ return 0;
}
-
- return 0;
+ return -EINVAL;
}
static void apply_cgroup_root_flags(unsigned int root_flags)
@@ -1766,6 +1858,11 @@
cgrp_dfl_root.flags |= CGRP_ROOT_NS_DELEGATE;
else
cgrp_dfl_root.flags &= ~CGRP_ROOT_NS_DELEGATE;
+
+ if (root_flags & CGRP_ROOT_MEMORY_LOCAL_EVENTS)
+ cgrp_dfl_root.flags |= CGRP_ROOT_MEMORY_LOCAL_EVENTS;
+ else
+ cgrp_dfl_root.flags &= ~CGRP_ROOT_MEMORY_LOCAL_EVENTS;
}
}
@@ -1773,19 +1870,16 @@
{
if (cgrp_dfl_root.flags & CGRP_ROOT_NS_DELEGATE)
seq_puts(seq, ",nsdelegate");
+ if (cgrp_dfl_root.flags & CGRP_ROOT_MEMORY_LOCAL_EVENTS)
+ seq_puts(seq, ",memory_localevents");
return 0;
}
-static int cgroup_remount(struct kernfs_root *kf_root, int *flags, char *data)
+static int cgroup_reconfigure(struct fs_context *fc)
{
- unsigned int root_flags;
- int ret;
+ struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
- ret = parse_cgroup_root_flags(data, &root_flags);
- if (ret)
- return ret;
-
- apply_cgroup_root_flags(root_flags);
+ apply_cgroup_root_flags(ctx->flags);
return 0;
}
@@ -1797,7 +1891,7 @@
*/
static bool use_task_css_set_links __read_mostly;
-static void cgroup_enable_task_cg_lists(void)
+void cgroup_enable_task_cg_lists(void)
{
struct task_struct *p, *g;
@@ -1873,8 +1967,9 @@
INIT_WORK(&cgrp->release_agent_work, cgroup1_release_agent);
}
-void init_cgroup_root(struct cgroup_root *root, struct cgroup_sb_opts *opts)
+void init_cgroup_root(struct cgroup_fs_context *ctx)
{
+ struct cgroup_root *root = ctx->root;
struct cgroup *cgrp = &root->cgrp;
INIT_LIST_HEAD(&root->root_list);
@@ -1883,16 +1978,16 @@
init_cgroup_housekeeping(cgrp);
idr_init(&root->cgroup_idr);
- root->flags = opts->flags;
- if (opts->release_agent)
- strscpy(root->release_agent_path, opts->release_agent, PATH_MAX);
- if (opts->name)
- strscpy(root->name, opts->name, MAX_CGROUP_ROOT_NAMELEN);
- if (opts->cpuset_clone_children)
+ root->flags = ctx->flags;
+ if (ctx->release_agent)
+ strscpy(root->release_agent_path, ctx->release_agent, PATH_MAX);
+ if (ctx->name)
+ strscpy(root->name, ctx->name, MAX_CGROUP_ROOT_NAMELEN);
+ if (ctx->cpuset_clone_children)
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;
@@ -1909,7 +2004,7 @@
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;
@@ -1993,57 +2088,105 @@
return ret;
}
-struct dentry *cgroup_do_mount(struct file_system_type *fs_type, int flags,
- struct cgroup_root *root, unsigned long magic,
- struct cgroup_namespace *ns)
+int cgroup_do_get_tree(struct fs_context *fc)
{
- struct dentry *dentry;
- bool new_sb;
+ struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
+ int ret;
- dentry = kernfs_mount(fs_type, flags, root->kf_root, magic, &new_sb);
+ ctx->kfc.root = ctx->root->kf_root;
+ if (fc->fs_type == &cgroup2_fs_type)
+ ctx->kfc.magic = CGROUP2_SUPER_MAGIC;
+ else
+ ctx->kfc.magic = CGROUP_SUPER_MAGIC;
+ ret = kernfs_get_tree(fc);
/*
* In non-init cgroup namespace, instead of root cgroup's dentry,
* we return the dentry corresponding to the cgroupns->root_cgrp.
*/
- if (!IS_ERR(dentry) && ns != &init_cgroup_ns) {
+ if (!ret && ctx->ns != &init_cgroup_ns) {
struct dentry *nsdentry;
+ struct super_block *sb = fc->root->d_sb;
struct cgroup *cgrp;
mutex_lock(&cgroup_mutex);
spin_lock_irq(&css_set_lock);
- cgrp = cset_cgroup_from_root(ns->root_cset, root);
+ cgrp = cset_cgroup_from_root(ctx->ns->root_cset, ctx->root);
spin_unlock_irq(&css_set_lock);
mutex_unlock(&cgroup_mutex);
- nsdentry = kernfs_node_dentry(cgrp->kn, dentry->d_sb);
- dput(dentry);
- dentry = nsdentry;
+ nsdentry = kernfs_node_dentry(cgrp->kn, sb);
+ dput(fc->root);
+ if (IS_ERR(nsdentry)) {
+ deactivate_locked_super(sb);
+ ret = PTR_ERR(nsdentry);
+ nsdentry = NULL;
+ }
+ fc->root = nsdentry;
}
- if (IS_ERR(dentry) || !new_sb)
- cgroup_put(&root->cgrp);
+ if (!ctx->kfc.new_sb_created)
+ cgroup_put(&ctx->root->cgrp);
- return dentry;
+ return ret;
}
-static struct dentry *cgroup_mount(struct file_system_type *fs_type,
- int flags, const char *unused_dev_name,
- void *data)
+/*
+ * Destroy a cgroup filesystem context.
+ */
+static void cgroup_fs_context_free(struct fs_context *fc)
{
- struct cgroup_namespace *ns = current->nsproxy->cgroup_ns;
- struct dentry *dentry;
+ struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
+
+ kfree(ctx->name);
+ kfree(ctx->release_agent);
+ put_cgroup_ns(ctx->ns);
+ kernfs_free_fs_context(fc);
+ kfree(ctx);
+}
+
+static int cgroup_get_tree(struct fs_context *fc)
+{
+ struct cgroup_fs_context *ctx = cgroup_fc2context(fc);
int ret;
- get_cgroup_ns(ns);
+ cgrp_dfl_visible = true;
+ cgroup_get_live(&cgrp_dfl_root.cgrp);
+ ctx->root = &cgrp_dfl_root;
- /* Check if the caller has permission to mount. */
- if (!ns_capable(ns->user_ns, CAP_SYS_ADMIN)) {
- put_cgroup_ns(ns);
- return ERR_PTR(-EPERM);
- }
+ ret = cgroup_do_get_tree(fc);
+ if (!ret)
+ apply_cgroup_root_flags(ctx->flags);
+ return ret;
+}
+
+static const struct fs_context_operations cgroup_fs_context_ops = {
+ .free = cgroup_fs_context_free,
+ .parse_param = cgroup2_parse_param,
+ .get_tree = cgroup_get_tree,
+ .reconfigure = cgroup_reconfigure,
+};
+
+static const struct fs_context_operations cgroup1_fs_context_ops = {
+ .free = cgroup_fs_context_free,
+ .parse_param = cgroup1_parse_param,
+ .get_tree = cgroup1_get_tree,
+ .reconfigure = cgroup1_reconfigure,
+};
+
+/*
+ * Initialise the cgroup filesystem creation/reconfiguration context. Notably,
+ * we select the namespace we're going to use.
+ */
+static int cgroup_init_fs_context(struct fs_context *fc)
+{
+ struct cgroup_fs_context *ctx;
+
+ ctx = kzalloc(sizeof(struct cgroup_fs_context), GFP_KERNEL);
+ if (!ctx)
+ return -ENOMEM;
/*
* The first time anyone tries to mount a cgroup, enable the list
@@ -2052,29 +2195,17 @@
if (!use_task_css_set_links)
cgroup_enable_task_cg_lists();
- if (fs_type == &cgroup2_fs_type) {
- unsigned int root_flags;
-
- ret = parse_cgroup_root_flags(data, &root_flags);
- if (ret) {
- put_cgroup_ns(ns);
- return ERR_PTR(ret);
- }
-
- cgrp_dfl_visible = true;
- cgroup_get_live(&cgrp_dfl_root.cgrp);
-
- dentry = cgroup_do_mount(&cgroup2_fs_type, flags, &cgrp_dfl_root,
- CGROUP2_SUPER_MAGIC, ns);
- if (!IS_ERR(dentry))
- apply_cgroup_root_flags(root_flags);
- } else {
- dentry = cgroup1_mount(&cgroup_fs_type, flags, data,
- CGROUP_SUPER_MAGIC, ns);
- }
-
- put_cgroup_ns(ns);
- return dentry;
+ ctx->ns = current->nsproxy->cgroup_ns;
+ get_cgroup_ns(ctx->ns);
+ fc->fs_private = &ctx->kfc;
+ if (fc->fs_type == &cgroup2_fs_type)
+ fc->ops = &cgroup_fs_context_ops;
+ else
+ fc->ops = &cgroup1_fs_context_ops;
+ put_user_ns(fc->user_ns);
+ fc->user_ns = get_user_ns(ctx->ns->user_ns);
+ fc->global = true;
+ return 0;
}
static void cgroup_kill_sb(struct super_block *sb)
@@ -2083,35 +2214,79 @@
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);
}
struct file_system_type cgroup_fs_type = {
- .name = "cgroup",
- .mount = cgroup_mount,
- .kill_sb = cgroup_kill_sb,
- .fs_flags = FS_USERNS_MOUNT,
+ .name = "cgroup",
+ .init_fs_context = cgroup_init_fs_context,
+ .parameters = &cgroup1_fs_parameters,
+ .kill_sb = cgroup_kill_sb,
+ .fs_flags = FS_USERNS_MOUNT,
};
static struct file_system_type cgroup2_fs_type = {
- .name = "cgroup2",
- .mount = cgroup_mount,
- .kill_sb = cgroup_kill_sb,
- .fs_flags = FS_USERNS_MOUNT,
+ .name = "cgroup2",
+ .init_fs_context = cgroup_init_fs_context,
+ .parameters = &cgroup2_fs_parameters,
+ .kill_sb = cgroup_kill_sb,
+ .fs_flags = FS_USERNS_MOUNT,
};
+#ifdef CONFIG_CPUSETS
+static const struct fs_context_operations cpuset_fs_context_ops = {
+ .get_tree = cgroup1_get_tree,
+ .free = cgroup_fs_context_free,
+};
+
+/*
+ * This is ugly, but preserves the userspace API for existing cpuset
+ * users. If someone tries to mount the "cpuset" filesystem, we
+ * silently switch it to mount "cgroup" instead
+ */
+static int cpuset_init_fs_context(struct fs_context *fc)
+{
+ char *agent = kstrdup("/sbin/cpuset_release_agent", GFP_USER);
+ struct cgroup_fs_context *ctx;
+ int err;
+
+ err = cgroup_init_fs_context(fc);
+ if (err) {
+ kfree(agent);
+ return err;
+ }
+
+ fc->ops = &cpuset_fs_context_ops;
+
+ ctx = cgroup_fc2context(fc);
+ ctx->subsys_mask = 1 << cpuset_cgrp_id;
+ ctx->flags |= CGRP_ROOT_NOPREFIX;
+ ctx->release_agent = agent;
+
+ get_filesystem(&cgroup_fs_type);
+ put_filesystem(fc->fs_type);
+ fc->fs_type = &cgroup_fs_type;
+
+ return 0;
+}
+
+static struct file_system_type cpuset_fs_type = {
+ .name = "cpuset",
+ .init_fs_context = cpuset_init_fs_context,
+ .fs_flags = FS_USERNS_MOUNT,
+};
+#endif
+
int cgroup_path_ns_locked(struct cgroup *cgrp, char *buf, size_t buflen,
struct cgroup_namespace *ns)
{
@@ -2323,8 +2498,15 @@
get_css_set(to_cset);
to_cset->nr_tasks++;
css_set_move_task(task, from_cset, to_cset, true);
- put_css_set_locked(from_cset);
from_cset->nr_tasks--;
+ /*
+ * If the source or destination cgroup is frozen,
+ * the task might require to change its state.
+ */
+ cgroup_freezer_migrate_task(task, from_cset->dfl_cgrp,
+ to_cset->dfl_cgrp);
+ put_css_set_locked(from_cset);
+
}
}
spin_unlock_irq(&css_set_lock);
@@ -2523,7 +2705,7 @@
dst_cset = find_css_set(src_cset, src_cset->mg_dst_cgrp);
if (!dst_cset)
- goto err;
+ return -ENOMEM;
WARN_ON_ONCE(src_cset->mg_dst_cset || dst_cset->mg_dst_cset);
@@ -2555,9 +2737,6 @@
}
return 0;
-err:
- cgroup_migrate_finish(mgctx);
- return -ENOMEM;
}
/**
@@ -2716,7 +2895,7 @@
do_each_subsys_mask(ss, ssid, ss_mask) {
if (printed)
seq_putc(seq, ' ');
- seq_printf(seq, "%s", ss->name);
+ seq_puts(seq, ss->name);
printed = true;
} while_each_subsys_mask();
if (printed)
@@ -3023,7 +3202,7 @@
return ret;
/*
- * At this point, cgroup_e_css() results reflect the new csses
+ * At this point, cgroup_e_css_by_mask() results reflect the new csses
* making the following cgroup_update_dfl_csses() properly update
* css associations of all tasks in the subtree.
*/
@@ -3368,8 +3547,11 @@
static int cgroup_events_show(struct seq_file *seq, void *v)
{
- seq_printf(seq, "populated %d\n",
- cgroup_is_populated(seq_css(seq)->cgroup));
+ struct cgroup *cgrp = seq_css(seq)->cgroup;
+
+ seq_printf(seq, "populated %d\n", cgroup_is_populated(cgrp));
+ seq_printf(seq, "frozen %d\n", test_bit(CGRP_FROZEN, &cgrp->flags));
+
return 0;
}
@@ -3416,6 +3598,122 @@
return ret;
}
+#ifdef CONFIG_PSI
+static int cgroup_io_pressure_show(struct seq_file *seq, void *v)
+{
+ struct cgroup *cgroup = seq_css(seq)->cgroup;
+ struct psi_group *psi = cgroup->id == 1 ? &psi_system : &cgroup->psi;
+
+ return psi_show(seq, psi, PSI_IO);
+}
+static int cgroup_memory_pressure_show(struct seq_file *seq, void *v)
+{
+ struct cgroup *cgroup = seq_css(seq)->cgroup;
+ struct psi_group *psi = cgroup->id == 1 ? &psi_system : &cgroup->psi;
+
+ return psi_show(seq, psi, PSI_MEM);
+}
+static int cgroup_cpu_pressure_show(struct seq_file *seq, void *v)
+{
+ struct cgroup *cgroup = seq_css(seq)->cgroup;
+ struct psi_group *psi = cgroup->id == 1 ? &psi_system : &cgroup->psi;
+
+ return psi_show(seq, psi, PSI_CPU);
+}
+
+static ssize_t cgroup_pressure_write(struct kernfs_open_file *of, char *buf,
+ size_t nbytes, enum psi_res res)
+{
+ struct psi_trigger *new;
+ struct cgroup *cgrp;
+
+ cgrp = cgroup_kn_lock_live(of->kn, false);
+ if (!cgrp)
+ return -ENODEV;
+
+ cgroup_get(cgrp);
+ cgroup_kn_unlock(of->kn);
+
+ new = psi_trigger_create(&cgrp->psi, buf, nbytes, res);
+ if (IS_ERR(new)) {
+ cgroup_put(cgrp);
+ return PTR_ERR(new);
+ }
+
+ psi_trigger_replace(&of->priv, new);
+
+ cgroup_put(cgrp);
+
+ return nbytes;
+}
+
+static ssize_t cgroup_io_pressure_write(struct kernfs_open_file *of,
+ char *buf, size_t nbytes,
+ loff_t off)
+{
+ return cgroup_pressure_write(of, buf, nbytes, PSI_IO);
+}
+
+static ssize_t cgroup_memory_pressure_write(struct kernfs_open_file *of,
+ char *buf, size_t nbytes,
+ loff_t off)
+{
+ return cgroup_pressure_write(of, buf, nbytes, PSI_MEM);
+}
+
+static ssize_t cgroup_cpu_pressure_write(struct kernfs_open_file *of,
+ char *buf, size_t nbytes,
+ loff_t off)
+{
+ return cgroup_pressure_write(of, buf, nbytes, PSI_CPU);
+}
+
+static __poll_t cgroup_pressure_poll(struct kernfs_open_file *of,
+ poll_table *pt)
+{
+ return psi_trigger_poll(&of->priv, of->file, pt);
+}
+
+static void cgroup_pressure_release(struct kernfs_open_file *of)
+{
+ psi_trigger_replace(&of->priv, NULL);
+}
+#endif /* CONFIG_PSI */
+
+static int cgroup_freeze_show(struct seq_file *seq, void *v)
+{
+ struct cgroup *cgrp = seq_css(seq)->cgroup;
+
+ seq_printf(seq, "%d\n", cgrp->freezer.freeze);
+
+ return 0;
+}
+
+static ssize_t cgroup_freeze_write(struct kernfs_open_file *of,
+ char *buf, size_t nbytes, loff_t off)
+{
+ struct cgroup *cgrp;
+ ssize_t ret;
+ int freeze;
+
+ ret = kstrtoint(strstrip(buf), 0, &freeze);
+ if (ret)
+ return ret;
+
+ if (freeze < 0 || freeze > 1)
+ return -ERANGE;
+
+ cgrp = cgroup_kn_lock_live(of->kn, false);
+ if (!cgrp)
+ return -ENOENT;
+
+ cgroup_freeze(cgrp, freeze);
+
+ cgroup_kn_unlock(of->kn);
+
+ return nbytes;
+}
+
static int cgroup_file_open(struct kernfs_open_file *of)
{
struct cftype *cft = of->kn->priv;
@@ -3483,6 +3781,16 @@
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);
@@ -3521,6 +3829,7 @@
.open = cgroup_file_open,
.release = cgroup_file_release,
.write = cgroup_file_write,
+ .poll = cgroup_file_poll,
.seq_show = cgroup_seqfile_show,
};
@@ -3529,6 +3838,7 @@
.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,
@@ -3623,7 +3933,8 @@
continue;
if ((cft->flags & CFTYPE_ONLY_ON_ROOT) && cgroup_parent(cgrp))
continue;
-
+ if ((cft->flags & CFTYPE_DEBUG) && !cgroup_debug)
+ continue;
if (is_add) {
ret = cgroup_add_file(css, cgrp, cft);
if (ret) {
@@ -3959,6 +4270,7 @@
return NULL;
}
+EXPORT_SYMBOL_GPL(css_next_descendant_pre);
/**
* css_rightmost_descendant - return the rightmost descendant of a css
@@ -4146,15 +4458,18 @@
it->task_pos = NULL;
return;
}
- } while (!css_set_populated(cset));
+ } while (!css_set_populated(cset) && list_empty(&cset->dying_tasks));
if (!list_empty(&cset->tasks))
it->task_pos = cset->tasks.next;
- else
+ else if (!list_empty(&cset->mg_tasks))
it->task_pos = cset->mg_tasks.next;
+ else
+ it->task_pos = cset->dying_tasks.next;
it->tasks_head = &cset->tasks;
it->mg_tasks_head = &cset->mg_tasks;
+ it->dying_tasks_head = &cset->dying_tasks;
/*
* We don't keep css_sets locked across iteration steps and thus
@@ -4180,32 +4495,63 @@
list_add(&it->iters_node, &cset->task_iters);
}
+static void css_task_iter_skip(struct css_task_iter *it,
+ struct task_struct *task)
+{
+ lockdep_assert_held(&css_set_lock);
+
+ if (it->task_pos == &task->cg_list) {
+ it->task_pos = it->task_pos->next;
+ it->flags |= CSS_TASK_ITER_SKIPPED;
+ }
+}
+
static void css_task_iter_advance(struct css_task_iter *it)
{
- struct list_head *next;
+ struct task_struct *task;
lockdep_assert_held(&css_set_lock);
repeat:
- /*
- * Advance iterator to find next entry. cset->tasks is consumed
- * first and then ->mg_tasks. After ->mg_tasks, we move onto the
- * next cset.
- */
- next = it->task_pos->next;
+ if (it->task_pos) {
+ /*
+ * Advance iterator to find next entry. cset->tasks is
+ * consumed first and then ->mg_tasks. After ->mg_tasks,
+ * we move onto the next cset.
+ */
+ if (it->flags & CSS_TASK_ITER_SKIPPED)
+ it->flags &= ~CSS_TASK_ITER_SKIPPED;
+ else
+ it->task_pos = it->task_pos->next;
- if (next == it->tasks_head)
- next = it->mg_tasks_head->next;
-
- if (next == it->mg_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->task_pos = it->dying_tasks_head->next;
+ if (it->task_pos == it->dying_tasks_head)
+ css_task_iter_advance_css_set(it);
+ } else {
+ /* called from start, proceed to the first cset */
css_task_iter_advance_css_set(it);
- else
- it->task_pos = next;
+ }
- /* if PROCS, skip over tasks which aren't group leaders */
- if ((it->flags & CSS_TASK_ITER_PROCS) && it->task_pos &&
- !thread_group_leader(list_entry(it->task_pos, struct task_struct,
- cg_list)))
- goto repeat;
+ if (!it->task_pos)
+ return;
+
+ task = list_entry(it->task_pos, struct task_struct, cg_list);
+
+ if (it->flags & CSS_TASK_ITER_PROCS) {
+ /* if PROCS, skip over tasks which aren't group leaders */
+ if (!thread_group_leader(task))
+ goto repeat;
+
+ /* and dying leaders w/o live member threads */
+ if (!atomic_read(&task->signal->live))
+ goto repeat;
+ } else {
+ /* skip all dying ones */
+ if (task->flags & PF_EXITING)
+ goto repeat;
+ }
}
/**
@@ -4239,7 +4585,7 @@
it->cset_head = it->cset_pos;
- css_task_iter_advance_css_set(it);
+ css_task_iter_advance(it);
spin_unlock_irq(&css_set_lock);
}
@@ -4261,6 +4607,10 @@
spin_lock_irq(&css_set_lock);
+ /* @it may be half-advanced by skips, finish advancing */
+ if (it->flags & CSS_TASK_ITER_SKIPPED)
+ css_task_iter_advance(it);
+
if (it->task_pos) {
it->cur_task = list_entry(it->task_pos, struct task_struct,
cg_list);
@@ -4542,10 +4892,39 @@
.seq_show = cgroup_stat_show,
},
{
+ .name = "cgroup.freeze",
+ .flags = CFTYPE_NOT_ON_ROOT,
+ .seq_show = cgroup_freeze_show,
+ .write = cgroup_freeze_write,
+ },
+ {
.name = "cpu.stat",
.flags = CFTYPE_NOT_ON_ROOT,
.seq_show = cpu_stat_show,
},
+#ifdef CONFIG_PSI
+ {
+ .name = "io.pressure",
+ .seq_show = cgroup_io_pressure_show,
+ .write = cgroup_io_pressure_write,
+ .poll = cgroup_pressure_poll,
+ .release = cgroup_pressure_release,
+ },
+ {
+ .name = "memory.pressure",
+ .seq_show = cgroup_memory_pressure_show,
+ .write = cgroup_memory_pressure_write,
+ .poll = cgroup_pressure_poll,
+ .release = cgroup_pressure_release,
+ },
+ {
+ .name = "cpu.pressure",
+ .seq_show = cgroup_cpu_pressure_show,
+ .write = cgroup_cpu_pressure_write,
+ .poll = cgroup_pressure_poll,
+ .release = cgroup_pressure_release,
+ },
+#endif /* CONFIG_PSI */
{ } /* terminate */
};
@@ -4606,6 +4985,7 @@
*/
cgroup_put(cgroup_parent(cgrp));
kernfs_put(cgrp->kn);
+ psi_cgroup_free(cgrp);
if (cgroup_on_dfl(cgrp))
cgroup_rstat_exit(cgrp);
kfree(cgrp);
@@ -4651,9 +5031,11 @@
if (cgroup_on_dfl(cgrp))
cgroup_rstat_flush(cgrp);
+ spin_lock_irq(&css_set_lock);
for (tcgrp = cgroup_parent(cgrp); tcgrp;
tcgrp = cgroup_parent(tcgrp))
tcgrp->nr_dying_descendants--;
+ spin_unlock_irq(&css_set_lock);
cgroup_idr_remove(&cgrp->root->cgroup_idr, cgrp->id);
cgrp->id = -1;
@@ -4668,8 +5050,6 @@
if (cgrp->kn)
RCU_INIT_POINTER(*(void __rcu __force **)&cgrp->kn->priv,
NULL);
-
- cgroup_bpf_put(cgrp);
}
mutex_unlock(&cgroup_mutex);
@@ -4862,16 +5242,48 @@
cgrp->self.parent = &parent->self;
cgrp->root = root;
cgrp->level = level;
- ret = cgroup_bpf_inherit(cgrp);
+
+ ret = psi_cgroup_alloc(cgrp);
if (ret)
goto out_idr_free;
+ ret = cgroup_bpf_inherit(cgrp);
+ if (ret)
+ goto out_psi_free;
+
+ /*
+ * New cgroup inherits effective freeze counter, and
+ * if the parent has to be frozen, the child has too.
+ */
+ cgrp->freezer.e_freeze = parent->freezer.e_freeze;
+ if (cgrp->freezer.e_freeze) {
+ /*
+ * Set the CGRP_FREEZE flag, so when a process will be
+ * attached to the child cgroup, it will become frozen.
+ * At this point the new cgroup is unpopulated, so we can
+ * consider it frozen immediately.
+ */
+ set_bit(CGRP_FREEZE, &cgrp->flags);
+ set_bit(CGRP_FROZEN, &cgrp->flags);
+ }
+
+ spin_lock_irq(&css_set_lock);
for (tcgrp = cgrp; tcgrp; tcgrp = cgroup_parent(tcgrp)) {
cgrp->ancestor_ids[tcgrp->level] = tcgrp->id;
- if (tcgrp != cgrp)
+ if (tcgrp != cgrp) {
tcgrp->nr_descendants++;
+
+ /*
+ * If the new cgroup is frozen, all ancestor cgroups
+ * get a new frozen descendant, but their state can't
+ * change because of this.
+ */
+ if (cgrp->freezer.e_freeze)
+ tcgrp->freezer.nr_frozen_descendants++;
+ }
}
+ spin_unlock_irq(&css_set_lock);
if (notify_on_release(parent))
set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);
@@ -4903,6 +5315,8 @@
return cgrp;
+out_psi_free:
+ psi_cgroup_free(cgrp);
out_idr_free:
cgroup_idr_remove(&root->cgroup_idr, cgrp->id);
out_stat_exit:
@@ -5154,13 +5568,23 @@
if (parent && cgroup_is_threaded(cgrp))
parent->nr_threaded_children--;
+ spin_lock_irq(&css_set_lock);
for (tcgrp = cgroup_parent(cgrp); tcgrp; tcgrp = cgroup_parent(tcgrp)) {
tcgrp->nr_descendants--;
tcgrp->nr_dying_descendants++;
+ /*
+ * If the dying cgroup is frozen, decrease frozen descendants
+ * counters of ancestor cgroups.
+ */
+ if (test_bit(CGRP_FROZEN, &cgrp->flags))
+ tcgrp->freezer.nr_frozen_descendants--;
}
+ spin_unlock_irq(&css_set_lock);
cgroup1_check_for_release(parent);
+ cgroup_bpf_offline(cgrp);
+
/* put the base reference */
percpu_ref_kill(&cgrp->self.refcnt);
@@ -5186,7 +5610,6 @@
static struct kernfs_syscall_ops cgroup_kf_syscall_ops = {
.show_options = cgroup_show_options,
- .remount_fs = cgroup_remount,
.mkdir = cgroup_mkdir,
.rmdir = cgroup_rmdir,
.show_path = cgroup_show_path,
@@ -5232,7 +5655,7 @@
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
@@ -5253,11 +5676,12 @@
*/
int __init cgroup_init_early(void)
{
- static struct cgroup_sb_opts __initdata opts;
+ static struct cgroup_fs_context __initdata ctx;
struct cgroup_subsys *ss;
int i;
- init_cgroup_root(&cgrp_dfl_root, &opts);
+ ctx.root = &cgrp_dfl_root;
+ init_cgroup_root(&ctx);
cgrp_dfl_root.cgrp.self.flags |= CSS_NO_REF;
RCU_INIT_POINTER(init_task.cgroups, &init_css_set);
@@ -5295,14 +5719,13 @@
int ssid;
BUILD_BUG_ON(CGROUP_SUBSYS_COUNT > 16);
- BUG_ON(percpu_init_rwsem(&cgroup_threadgroup_rwsem));
BUG_ON(cgroup_init_cftypes(NULL, cgroup_base_files));
BUG_ON(cgroup_init_cftypes(NULL, cgroup1_base_files));
cgroup_rstat_boot();
/*
- * The latency of the synchronize_sched() is too high for cgroups,
+ * The latency of the synchronize_rcu() is too high for cgroups,
* avoid it at the cost of forcing all readers into the slow path.
*/
rcu_sync_enter_start(&cgroup_threadgroup_rwsem.rss);
@@ -5318,7 +5741,7 @@
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);
@@ -5390,6 +5813,9 @@
WARN_ON(register_filesystem(&cgroup_fs_type));
WARN_ON(register_filesystem(&cgroup2_fs_type));
WARN_ON(!proc_create_single("cgroups", 0, NULL, proc_cgroupstats_show));
+#ifdef CONFIG_CPUSETS
+ WARN_ON(register_filesystem(&cpuset_fs_type));
+#endif
return 0;
}
@@ -5609,6 +6035,26 @@
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 the JOBCTL_TRAP_FREEZE jobctl bit to get
+ * the task into the frozen state.
+ */
+ if (unlikely(cgroup_task_freeze(child))) {
+ spin_lock(&child->sighand->siglock);
+ WARN_ON_ONCE(child->frozen);
+ child->jobctl |= JOBCTL_TRAP_FREEZE;
+ spin_unlock(&child->sighand->siglock);
+
+ /*
+ * Calling cgroup_update_frozen() isn't required here,
+ * because it will be called anyway a bit later
+ * from do_freezer_trap(). So we avoid cgroup's
+ * transient switch from the frozen state and back.
+ */
+ }
+
spin_unlock_irq(&css_set_lock);
}
@@ -5656,7 +6102,13 @@
if (!list_empty(&tsk->cg_list)) {
spin_lock_irq(&css_set_lock);
css_set_move_task(tsk, cset, NULL, false);
+ list_add_tail(&tsk->cg_list, &cset->dying_tasks);
cset->nr_tasks--;
+
+ WARN_ON_ONCE(cgroup_task_frozen(tsk));
+ if (unlikely(cgroup_task_freeze(tsk)))
+ cgroup_update_frozen(task_dfl_cgroup(tsk));
+
spin_unlock_irq(&css_set_lock);
} else {
get_css_set(cset);
@@ -5668,16 +6120,26 @@
} 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();
+ if (use_task_css_set_links) {
+ spin_lock_irq(&css_set_lock);
+ css_set_skip_task_iters(task_css_set(task), task);
+ list_del_init(&task->cg_list);
+ spin_unlock_irq(&css_set_lock);
+ }
+}
+
+void cgroup_free(struct task_struct *task)
+{
+ struct css_set *cset = task_css_set(task);
put_css_set(cset);
}
@@ -5702,6 +6164,16 @@
}
__setup("cgroup_disable=", cgroup_disable);
+void __init __weak enable_debug_cgroup(void) { }
+
+static int __init enable_cgroup_debug(char *str)
+{
+ cgroup_debug = true;
+ enable_debug_cgroup();
+ return 1;
+}
+__setup("cgroup_debug", enable_cgroup_debug);
+
/**
* css_tryget_online_from_dir - get corresponding css from a cgroup dentry
* @dentry: directory dentry of interest
@@ -5824,6 +6296,48 @@
}
EXPORT_SYMBOL_GPL(cgroup_get_from_fd);
+static u64 power_of_ten(int power)
+{
+ u64 v = 1;
+ while (power--)
+ v *= 10;
+ return v;
+}
+
+/**
+ * cgroup_parse_float - parse a floating number
+ * @input: input string
+ * @dec_shift: number of decimal digits to shift
+ * @v: output
+ *
+ * Parse a decimal floating point number in @input and store the result in
+ * @v with decimal point right shifted @dec_shift times. For example, if
+ * @input is "12.3456" and @dec_shift is 3, *@v will be set to 12345.
+ * Returns 0 on success, -errno otherwise.
+ *
+ * There's nothing cgroup specific about this function except that it's
+ * currently the only user.
+ */
+int cgroup_parse_float(const char *input, unsigned dec_shift, s64 *v)
+{
+ s64 whole, frac = 0;
+ int fstart = 0, fend = 0, flen;
+
+ if (!sscanf(input, "%lld.%n%lld%n", &whole, &fstart, &frac, &fend))
+ return -EINVAL;
+ if (frac < 0)
+ return -EINVAL;
+
+ flen = fend > fstart ? fend - fstart : 0;
+ if (flen < dec_shift)
+ frac *= power_of_ten(dec_shift - flen);
+ else
+ frac = DIV_ROUND_CLOSEST_ULL(frac, power_of_ten(flen - dec_shift));
+
+ *v = whole * power_of_ten(dec_shift) + frac;
+ return 0;
+}
+
/*
* sock->sk_cgrp_data handling. For more info, see sock_cgroup_data
* definition in cgroup-defs.h.
@@ -5862,6 +6376,7 @@
* Don't use cgroup_get_live().
*/
cgroup_get(sock_cgroup_ptr(skcd));
+ cgroup_bpf_get(sock_cgroup_ptr(skcd));
return;
}
@@ -5873,6 +6388,7 @@
cset = task_css_set(current);
if (likely(cgroup_tryget(cset->dfl_cgrp))) {
skcd->val = (unsigned long)cset->dfl_cgrp;
+ cgroup_bpf_get(cset->dfl_cgrp);
break;
}
cpu_relax();
@@ -5883,7 +6399,10 @@
void cgroup_sk_free(struct sock_cgroup_data *skcd)
{
- cgroup_put(sock_cgroup_ptr(skcd));
+ struct cgroup *cgrp = sock_cgroup_ptr(skcd);
+
+ cgroup_bpf_put(cgrp);
+ cgroup_put(cgrp);
}
#endif /* CONFIG_SOCK_CGROUP_DATA */
@@ -5905,7 +6424,7 @@
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;
}
@@ -5937,10 +6456,8 @@
ret += snprintf(buf + ret, size - ret, "%s\n", cft->name);
- if (unlikely(ret >= size)) {
- WARN_ON(1);
+ if (WARN_ON(ret >= size))
break;
- }
}
return ret;
@@ -5968,7 +6485,7 @@
static ssize_t features_show(struct kobject *kobj, struct kobj_attribute *attr,
char *buf)
{
- return snprintf(buf, PAGE_SIZE, "nsdelegate\n");
+ return snprintf(buf, PAGE_SIZE, "nsdelegate\nmemory_localevents\n");
}
static struct kobj_attribute cgroup_features_attr = __ATTR_RO(features);
@@ -5988,4 +6505,5 @@
return sysfs_create_group(kernel_kobj, &cgroup_sysfs_attr_group);
}
subsys_initcall(cgroup_sysfs_init);
+
#endif /* CONFIG_SYSFS */
diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
index 266f10c..c87ee64 100644
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -39,11 +39,13 @@
#include <linux/memory.h>
#include <linux/export.h>
#include <linux/mount.h>
+#include <linux/fs_context.h>
#include <linux/namei.h>
#include <linux/pagemap.h>
#include <linux/proc_fs.h>
#include <linux/rcupdate.h>
#include <linux/sched.h>
+#include <linux/sched/deadline.h>
#include <linux/sched/mm.h>
#include <linux/sched/task.h>
#include <linux/seq_file.h>
@@ -110,6 +112,16 @@
nodemask_t effective_mems;
/*
+ * CPUs allocated to child sub-partitions (default hierarchy only)
+ * - CPUs granted by the parent = effective_cpus U subparts_cpus
+ * - effective_cpus and subparts_cpus are mutually exclusive.
+ *
+ * effective_cpus contains only onlined CPUs, but subparts_cpus
+ * may have offlined ones.
+ */
+ cpumask_var_t subparts_cpus;
+
+ /*
* This is old Memory Nodes tasks took on.
*
* - top_cpuset.old_mems_allowed is initialized to mems_allowed.
@@ -134,6 +146,47 @@
/* for custom sched domain */
int relax_domain_level;
+
+ /* number of CPUs in subparts_cpus */
+ int nr_subparts_cpus;
+
+ /* partition root state */
+ int partition_root_state;
+
+ /*
+ * Default hierarchy only:
+ * use_parent_ecpus - set if using parent's effective_cpus
+ * child_ecpus_count - # of children with use_parent_ecpus set
+ */
+ int use_parent_ecpus;
+ int child_ecpus_count;
+};
+
+/*
+ * Partition root states:
+ *
+ * 0 - not a partition root
+ *
+ * 1 - partition root
+ *
+ * -1 - invalid partition root
+ * None of the cpus in cpus_allowed can be put into the parent's
+ * subparts_cpus. In this case, the cpuset is not a real partition
+ * root anymore. However, the CPU_EXCLUSIVE bit will still be set
+ * and the cpuset can be restored back to a partition root if the
+ * parent cpuset can give more CPUs back to this child cpuset.
+ */
+#define PRS_DISABLED 0
+#define PRS_ENABLED 1
+#define PRS_ERROR -1
+
+/*
+ * Temporary cpumasks for working with partitions that are passed among
+ * functions to avoid memory allocation in inner functions.
+ */
+struct tmpmasks {
+ cpumask_var_t addmask, delmask; /* For partition root */
+ cpumask_var_t new_cpus; /* For update_cpumasks_hier() */
};
static inline struct cpuset *css_cs(struct cgroup_subsys_state *css)
@@ -152,19 +205,6 @@
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,
@@ -218,9 +258,15 @@
return test_bit(CS_SPREAD_SLAB, &cs->flags);
}
+static inline int is_partition_root(const struct cpuset *cs)
+{
+ return cs->partition_root_state > 0;
+}
+
static struct cpuset top_cpuset = {
.flags = ((1 << CS_ONLINE) | (1 << CS_CPU_EXCLUSIVE) |
(1 << CS_MEM_EXCLUSIVE)),
+ .partition_root_state = PRS_ENABLED,
};
/**
@@ -287,7 +333,18 @@
* guidelines for accessing subsystem state in kernel/cgroup.c
*/
-static DEFINE_MUTEX(cpuset_mutex);
+DEFINE_STATIC_PERCPU_RWSEM(cpuset_rwsem);
+
+void cpuset_read_lock(void)
+{
+ percpu_down_read(&cpuset_rwsem);
+}
+
+void cpuset_read_unlock(void)
+{
+ percpu_up_read(&cpuset_rwsem);
+}
+
static DEFINE_SPINLOCK(callback_lock);
static struct workqueue_struct *cpuset_migrate_mm_wq;
@@ -311,32 +368,6 @@
}
/*
- * This is ugly, but preserves the userspace API for existing cpuset
- * users. If someone tries to mount the "cpuset" filesystem, we
- * silently switch it to mount "cgroup" instead
- */
-static struct dentry *cpuset_mount(struct file_system_type *fs_type,
- int flags, const char *unused_dev_name, void *data)
-{
- struct file_system_type *cgroup_fs = get_fs_type("cgroup");
- struct dentry *ret = ERR_PTR(-ENODEV);
- if (cgroup_fs) {
- char mountopts[] =
- "cpuset,noprefix,"
- "release_agent=/sbin/cpuset_release_agent";
- ret = cgroup_fs->mount(cgroup_fs, flags,
- unused_dev_name, mountopts);
- put_filesystem(cgroup_fs);
- }
- return ret;
-}
-
-static struct file_system_type cpuset_fs_type = {
- .name = "cpuset",
- .mount = cpuset_mount,
-};
-
-/*
* Return in pmask the portion of a cpusets's cpus_allowed that
* are online. If none are online, walk up the cpuset hierarchy
* until we find one that does have some online cpus.
@@ -419,6 +450,65 @@
}
/**
+ * alloc_cpumasks - allocate three cpumasks for cpuset
+ * @cs: the cpuset that have cpumasks to be allocated.
+ * @tmp: the tmpmasks structure pointer
+ * Return: 0 if successful, -ENOMEM otherwise.
+ *
+ * Only one of the two input arguments should be non-NULL.
+ */
+static inline int alloc_cpumasks(struct cpuset *cs, struct tmpmasks *tmp)
+{
+ cpumask_var_t *pmask1, *pmask2, *pmask3;
+
+ if (cs) {
+ pmask1 = &cs->cpus_allowed;
+ pmask2 = &cs->effective_cpus;
+ pmask3 = &cs->subparts_cpus;
+ } else {
+ pmask1 = &tmp->new_cpus;
+ pmask2 = &tmp->addmask;
+ pmask3 = &tmp->delmask;
+ }
+
+ if (!zalloc_cpumask_var(pmask1, GFP_KERNEL))
+ return -ENOMEM;
+
+ if (!zalloc_cpumask_var(pmask2, GFP_KERNEL))
+ goto free_one;
+
+ if (!zalloc_cpumask_var(pmask3, GFP_KERNEL))
+ goto free_two;
+
+ return 0;
+
+free_two:
+ free_cpumask_var(*pmask2);
+free_one:
+ free_cpumask_var(*pmask1);
+ return -ENOMEM;
+}
+
+/**
+ * free_cpumasks - free cpumasks in a tmpmasks structure
+ * @cs: the cpuset that have cpumasks to be free.
+ * @tmp: the tmpmasks structure pointer
+ */
+static inline void free_cpumasks(struct cpuset *cs, struct tmpmasks *tmp)
+{
+ if (cs) {
+ free_cpumask_var(cs->cpus_allowed);
+ free_cpumask_var(cs->effective_cpus);
+ free_cpumask_var(cs->subparts_cpus);
+ }
+ if (tmp) {
+ free_cpumask_var(tmp->new_cpus);
+ free_cpumask_var(tmp->addmask);
+ free_cpumask_var(tmp->delmask);
+ }
+}
+
+/**
* alloc_trial_cpuset - allocate a trial cpuset
* @cs: the cpuset that the trial cpuset duplicates
*/
@@ -430,31 +520,24 @@
if (!trial)
return NULL;
- if (!alloc_cpumask_var(&trial->cpus_allowed, GFP_KERNEL))
- goto free_cs;
- if (!alloc_cpumask_var(&trial->effective_cpus, GFP_KERNEL))
- goto free_cpus;
+ if (alloc_cpumasks(trial, NULL)) {
+ kfree(trial);
+ return NULL;
+ }
cpumask_copy(trial->cpus_allowed, cs->cpus_allowed);
cpumask_copy(trial->effective_cpus, cs->effective_cpus);
return trial;
-
-free_cpus:
- free_cpumask_var(trial->cpus_allowed);
-free_cs:
- kfree(trial);
- return NULL;
}
/**
- * free_trial_cpuset - free the trial cpuset
- * @trial: the trial cpuset to be freed
+ * free_cpuset - free the cpuset
+ * @cs: the cpuset to be freed
*/
-static void free_trial_cpuset(struct cpuset *trial)
+static inline void free_cpuset(struct cpuset *cs)
{
- free_cpumask_var(trial->effective_cpus);
- free_cpumask_var(trial->cpus_allowed);
- kfree(trial);
+ free_cpumasks(cs, NULL);
+ kfree(cs);
}
/*
@@ -605,7 +688,7 @@
* load balancing domains (sched domains) as specified by that partial
* partition.
*
- * See "What is sched_load_balance" in Documentation/cgroup-v1/cpusets.txt
+ * See "What is sched_load_balance" in Documentation/admin-guide/cgroup-v1/cpusets.rst
* for a background explanation of this.
*
* Does not return errors, on the theory that the callers of this
@@ -616,11 +699,10 @@
* Must be called with cpuset_mutex held.
*
* The three key local variables below are:
- * q - a linked-list queue of cpuset pointers, used to implement a
- * top-down scan of all cpusets. This scan loads a pointer
- * to each cpuset marked is_sched_load_balance into the
- * array 'csa'. For our purposes, rebuilding the schedulers
- * sched domains, we can ignore !is_sched_load_balance cpusets.
+ * cp - cpuset pointer, used (together with pos_css) to perform a
+ * top-down scan of all cpusets. For our purposes, rebuilding
+ * the schedulers sched domains, we can ignore !is_sched_load_
+ * balance cpusets.
* csa - (for CpuSet Array) Array of pointers to all the cpusets
* that need to be load balanced, for convenient iterative
* access by the subsequent code that finds the best partition,
@@ -651,7 +733,7 @@
static int generate_sched_domains(cpumask_var_t **domains,
struct sched_domain_attr **attributes)
{
- struct cpuset *cp; /* scans q */
+ struct cpuset *cp; /* top-down scan of cpusets */
struct cpuset **csa; /* array of all cpuset ptrs */
int csn; /* how many cpuset ptrs in csa so far */
int i, j, k; /* indices for partition finding loops */
@@ -660,13 +742,14 @@
int ndoms = 0; /* number of sched domains in result */
int nslot; /* next empty doms[] struct cpumask slot */
struct cgroup_subsys_state *pos_css;
+ bool root_load_balance = is_sched_load_balance(&top_cpuset);
doms = NULL;
dattr = NULL;
csa = NULL;
/* Special case for the 99% of systems with one, full, sched domain */
- if (is_sched_load_balance(&top_cpuset)) {
+ if (root_load_balance && !top_cpuset.nr_subparts_cpus) {
ndoms = 1;
doms = alloc_sched_domains(ndoms);
if (!doms)
@@ -689,6 +772,8 @@
csn = 0;
rcu_read_lock();
+ if (root_load_balance)
+ csa[csn++] = &top_cpuset;
cpuset_for_each_descendant_pre(cp, pos_css, &top_cpuset) {
if (cp == &top_cpuset)
continue;
@@ -699,6 +784,9 @@
* parent's cpus, so just skip them, and then we call
* update_domain_attr_tree() to calc relax_domain_level of
* the corresponding sched domain.
+ *
+ * If root is load-balancing, we can skip @cp if it
+ * is a subset of the root's effective_cpus.
*/
if (!cpumask_empty(cp->cpus_allowed) &&
!(is_sched_load_balance(cp) &&
@@ -706,11 +794,17 @@
housekeeping_cpumask(HK_FLAG_DOMAIN))))
continue;
- if (is_sched_load_balance(cp))
+ if (root_load_balance &&
+ cpumask_subset(cp->cpus_allowed, top_cpuset.effective_cpus))
+ continue;
+
+ if (is_sched_load_balance(cp) &&
+ !cpumask_empty(cp->effective_cpus))
csa[csn++] = cp;
- /* skip @cp's subtree */
- pos_css = css_rightmost_descendant(pos_css);
+ /* skip @cp's subtree if not a partition root */
+ if (!is_partition_root(cp))
+ pos_css = css_rightmost_descendant(pos_css);
}
rcu_read_unlock();
@@ -813,6 +907,67 @@
return ndoms;
}
+static void update_tasks_root_domain(struct cpuset *cs)
+{
+ struct css_task_iter it;
+ struct task_struct *task;
+
+ css_task_iter_start(&cs->css, 0, &it);
+
+ while ((task = css_task_iter_next(&it)))
+ dl_add_task_root_domain(task);
+
+ css_task_iter_end(&it);
+}
+
+static void rebuild_root_domains(void)
+{
+ struct cpuset *cs = NULL;
+ struct cgroup_subsys_state *pos_css;
+
+ percpu_rwsem_assert_held(&cpuset_rwsem);
+ lockdep_assert_cpus_held();
+ lockdep_assert_held(&sched_domains_mutex);
+
+ cgroup_enable_task_cg_lists();
+
+ rcu_read_lock();
+
+ /*
+ * Clear default root domain DL accounting, it will be computed again
+ * if a task belongs to it.
+ */
+ dl_clear_root_domain(&def_root_domain);
+
+ cpuset_for_each_descendant_pre(cs, pos_css, &top_cpuset) {
+
+ if (cpumask_empty(cs->effective_cpus)) {
+ pos_css = css_rightmost_descendant(pos_css);
+ continue;
+ }
+
+ css_get(&cs->css);
+
+ rcu_read_unlock();
+
+ update_tasks_root_domain(cs);
+
+ rcu_read_lock();
+ css_put(&cs->css);
+ }
+ rcu_read_unlock();
+}
+
+static void
+partition_and_rebuild_sched_domains(int ndoms_new, cpumask_var_t doms_new[],
+ struct sched_domain_attr *dattr_new)
+{
+ mutex_lock(&sched_domains_mutex);
+ partition_sched_domains_locked(ndoms_new, doms_new, dattr_new);
+ rebuild_root_domains();
+ mutex_unlock(&sched_domains_mutex);
+}
+
/*
* Rebuild scheduler domains.
*
@@ -830,24 +985,27 @@
cpumask_var_t *doms;
int ndoms;
- lockdep_assert_held(&cpuset_mutex);
- get_online_cpus();
+ lockdep_assert_cpus_held();
+ percpu_rwsem_assert_held(&cpuset_rwsem);
/*
* We have raced with CPU hotplug. Don't do anything to avoid
* passing doms with offlined cpu to partition_sched_domains().
* Anyways, hotplug work item will rebuild sched domains.
*/
- if (!cpumask_equal(top_cpuset.effective_cpus, cpu_active_mask))
- goto out;
+ if (!top_cpuset.nr_subparts_cpus &&
+ !cpumask_equal(top_cpuset.effective_cpus, cpu_active_mask))
+ return;
+
+ if (top_cpuset.nr_subparts_cpus &&
+ !cpumask_subset(top_cpuset.effective_cpus, cpu_active_mask))
+ return;
/* Generate domain masks and attrs */
ndoms = generate_sched_domains(&doms, &attr);
/* Have scheduler rebuild the domains */
- partition_sched_domains(ndoms, doms, attr);
-out:
- put_online_cpus();
+ partition_and_rebuild_sched_domains(ndoms, doms, attr);
}
#else /* !CONFIG_SMP */
static void rebuild_sched_domains_locked(void)
@@ -857,9 +1015,11 @@
void rebuild_sched_domains(void)
{
- mutex_lock(&cpuset_mutex);
+ get_online_cpus();
+ percpu_down_write(&cpuset_rwsem);
rebuild_sched_domains_locked();
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
+ put_online_cpus();
}
/**
@@ -881,10 +1041,248 @@
css_task_iter_end(&it);
}
+/**
+ * compute_effective_cpumask - Compute the effective cpumask of the cpuset
+ * @new_cpus: the temp variable for the new effective_cpus mask
+ * @cs: the cpuset the need to recompute the new effective_cpus mask
+ * @parent: the parent cpuset
+ *
+ * If the parent has subpartition CPUs, include them in the list of
+ * allowable CPUs in computing the new effective_cpus mask. Since offlined
+ * CPUs are not removed from subparts_cpus, we have to use cpu_active_mask
+ * to mask those out.
+ */
+static void compute_effective_cpumask(struct cpumask *new_cpus,
+ struct cpuset *cs, struct cpuset *parent)
+{
+ if (parent->nr_subparts_cpus) {
+ cpumask_or(new_cpus, parent->effective_cpus,
+ parent->subparts_cpus);
+ cpumask_and(new_cpus, new_cpus, cs->cpus_allowed);
+ cpumask_and(new_cpus, new_cpus, cpu_active_mask);
+ } else {
+ cpumask_and(new_cpus, cs->cpus_allowed, parent->effective_cpus);
+ }
+}
+
+/*
+ * Commands for update_parent_subparts_cpumask
+ */
+enum subparts_cmd {
+ partcmd_enable, /* Enable partition root */
+ partcmd_disable, /* Disable partition root */
+ partcmd_update, /* Update parent's subparts_cpus */
+};
+
+/**
+ * update_parent_subparts_cpumask - update subparts_cpus mask of parent cpuset
+ * @cpuset: The cpuset that requests change in partition root state
+ * @cmd: Partition root state change command
+ * @newmask: Optional new cpumask for partcmd_update
+ * @tmp: Temporary addmask and delmask
+ * Return: 0, 1 or an error code
+ *
+ * For partcmd_enable, the cpuset is being transformed from a non-partition
+ * root to a partition root. The cpus_allowed mask of the given cpuset will
+ * be put into parent's subparts_cpus and taken away from parent's
+ * effective_cpus. The function will return 0 if all the CPUs listed in
+ * cpus_allowed can be granted or an error code will be returned.
+ *
+ * For partcmd_disable, the cpuset is being transofrmed from a partition
+ * root back to a non-partition root. any CPUs in cpus_allowed that are in
+ * parent's subparts_cpus will be taken away from that cpumask and put back
+ * into parent's effective_cpus. 0 should always be returned.
+ *
+ * For partcmd_update, if the optional newmask is specified, the cpu
+ * list is to be changed from cpus_allowed to newmask. Otherwise,
+ * cpus_allowed is assumed to remain the same. The cpuset should either
+ * be a partition root or an invalid partition root. The partition root
+ * state may change if newmask is NULL and none of the requested CPUs can
+ * be granted by the parent. The function will return 1 if changes to
+ * parent's subparts_cpus and effective_cpus happen or 0 otherwise.
+ * Error code should only be returned when newmask is non-NULL.
+ *
+ * The partcmd_enable and partcmd_disable commands are used by
+ * update_prstate(). The partcmd_update command is used by
+ * update_cpumasks_hier() with newmask NULL and update_cpumask() with
+ * newmask set.
+ *
+ * The checking is more strict when enabling partition root than the
+ * other two commands.
+ *
+ * Because of the implicit cpu exclusive nature of a partition root,
+ * cpumask changes that violates the cpu exclusivity rule will not be
+ * permitted when checked by validate_change(). The validate_change()
+ * function will also prevent any changes to the cpu list if it is not
+ * a superset of children's cpu lists.
+ */
+static int update_parent_subparts_cpumask(struct cpuset *cpuset, int cmd,
+ struct cpumask *newmask,
+ struct tmpmasks *tmp)
+{
+ struct cpuset *parent = parent_cs(cpuset);
+ int adding; /* Moving cpus from effective_cpus to subparts_cpus */
+ int deleting; /* Moving cpus from subparts_cpus to effective_cpus */
+ bool part_error = false; /* Partition error? */
+
+ percpu_rwsem_assert_held(&cpuset_rwsem);
+
+ /*
+ * The parent must be a partition root.
+ * The new cpumask, if present, or the current cpus_allowed must
+ * not be empty.
+ */
+ if (!is_partition_root(parent) ||
+ (newmask && cpumask_empty(newmask)) ||
+ (!newmask && cpumask_empty(cpuset->cpus_allowed)))
+ return -EINVAL;
+
+ /*
+ * Enabling/disabling partition root is not allowed if there are
+ * online children.
+ */
+ if ((cmd != partcmd_update) && css_has_online_children(&cpuset->css))
+ return -EBUSY;
+
+ /*
+ * Enabling partition root is not allowed if not all the CPUs
+ * can be granted from parent's effective_cpus or at least one
+ * CPU will be left after that.
+ */
+ if ((cmd == partcmd_enable) &&
+ (!cpumask_subset(cpuset->cpus_allowed, parent->effective_cpus) ||
+ cpumask_equal(cpuset->cpus_allowed, parent->effective_cpus)))
+ return -EINVAL;
+
+ /*
+ * A cpumask update cannot make parent's effective_cpus become empty.
+ */
+ adding = deleting = false;
+ if (cmd == partcmd_enable) {
+ cpumask_copy(tmp->addmask, cpuset->cpus_allowed);
+ adding = true;
+ } else if (cmd == partcmd_disable) {
+ deleting = cpumask_and(tmp->delmask, cpuset->cpus_allowed,
+ parent->subparts_cpus);
+ } else if (newmask) {
+ /*
+ * partcmd_update with newmask:
+ *
+ * delmask = cpus_allowed & ~newmask & parent->subparts_cpus
+ * addmask = newmask & parent->effective_cpus
+ * & ~parent->subparts_cpus
+ */
+ cpumask_andnot(tmp->delmask, cpuset->cpus_allowed, newmask);
+ deleting = cpumask_and(tmp->delmask, tmp->delmask,
+ parent->subparts_cpus);
+
+ cpumask_and(tmp->addmask, newmask, parent->effective_cpus);
+ adding = cpumask_andnot(tmp->addmask, tmp->addmask,
+ parent->subparts_cpus);
+ /*
+ * Return error if the new effective_cpus could become empty.
+ */
+ if (adding &&
+ cpumask_equal(parent->effective_cpus, tmp->addmask)) {
+ if (!deleting)
+ return -EINVAL;
+ /*
+ * As some of the CPUs in subparts_cpus might have
+ * been offlined, we need to compute the real delmask
+ * to confirm that.
+ */
+ if (!cpumask_and(tmp->addmask, tmp->delmask,
+ cpu_active_mask))
+ return -EINVAL;
+ cpumask_copy(tmp->addmask, parent->effective_cpus);
+ }
+ } else {
+ /*
+ * partcmd_update w/o newmask:
+ *
+ * addmask = cpus_allowed & parent->effectiveb_cpus
+ *
+ * Note that parent's subparts_cpus may have been
+ * pre-shrunk in case there is a change in the cpu list.
+ * So no deletion is needed.
+ */
+ adding = cpumask_and(tmp->addmask, cpuset->cpus_allowed,
+ parent->effective_cpus);
+ part_error = cpumask_equal(tmp->addmask,
+ parent->effective_cpus);
+ }
+
+ if (cmd == partcmd_update) {
+ int prev_prs = cpuset->partition_root_state;
+
+ /*
+ * Check for possible transition between PRS_ENABLED
+ * and PRS_ERROR.
+ */
+ switch (cpuset->partition_root_state) {
+ case PRS_ENABLED:
+ if (part_error)
+ cpuset->partition_root_state = PRS_ERROR;
+ break;
+ case PRS_ERROR:
+ if (!part_error)
+ cpuset->partition_root_state = PRS_ENABLED;
+ break;
+ }
+ /*
+ * Set part_error if previously in invalid state.
+ */
+ part_error = (prev_prs == PRS_ERROR);
+ }
+
+ if (!part_error && (cpuset->partition_root_state == PRS_ERROR))
+ return 0; /* Nothing need to be done */
+
+ if (cpuset->partition_root_state == PRS_ERROR) {
+ /*
+ * Remove all its cpus from parent's subparts_cpus.
+ */
+ adding = false;
+ deleting = cpumask_and(tmp->delmask, cpuset->cpus_allowed,
+ parent->subparts_cpus);
+ }
+
+ if (!adding && !deleting)
+ return 0;
+
+ /*
+ * Change the parent's subparts_cpus.
+ * Newly added CPUs will be removed from effective_cpus and
+ * newly deleted ones will be added back to effective_cpus.
+ */
+ spin_lock_irq(&callback_lock);
+ if (adding) {
+ cpumask_or(parent->subparts_cpus,
+ parent->subparts_cpus, tmp->addmask);
+ cpumask_andnot(parent->effective_cpus,
+ parent->effective_cpus, tmp->addmask);
+ }
+ if (deleting) {
+ cpumask_andnot(parent->subparts_cpus,
+ parent->subparts_cpus, tmp->delmask);
+ /*
+ * Some of the CPUs in subparts_cpus might have been offlined.
+ */
+ cpumask_and(tmp->delmask, tmp->delmask, cpu_active_mask);
+ cpumask_or(parent->effective_cpus,
+ parent->effective_cpus, tmp->delmask);
+ }
+
+ parent->nr_subparts_cpus = cpumask_weight(parent->subparts_cpus);
+ spin_unlock_irq(&callback_lock);
+
+ return cmd == partcmd_update;
+}
+
/*
* update_cpumasks_hier - Update effective cpumasks and tasks in the subtree
- * @cs: the cpuset to consider
- * @new_cpus: temp variable for calculating new effective_cpus
+ * @cs: the cpuset to consider
+ * @tmp: temp variables for calculating effective_cpus & partition setup
*
* When congifured cpumask is changed, the effective cpumasks of this cpuset
* and all its descendants need to be updated.
@@ -893,7 +1291,7 @@
*
* Called with cpuset_mutex held
*/
-static void update_cpumasks_hier(struct cpuset *cs, struct cpumask *new_cpus)
+static void update_cpumasks_hier(struct cpuset *cs, struct tmpmasks *tmp)
{
struct cpuset *cp;
struct cgroup_subsys_state *pos_css;
@@ -903,27 +1301,115 @@
cpuset_for_each_descendant_pre(cp, pos_css, cs) {
struct cpuset *parent = parent_cs(cp);
- cpumask_and(new_cpus, cp->cpus_allowed, parent->effective_cpus);
+ compute_effective_cpumask(tmp->new_cpus, cp, parent);
/*
* If it becomes empty, inherit the effective mask of the
* parent, which is guaranteed to have some CPUs.
*/
- if (is_in_v2_mode() && cpumask_empty(new_cpus))
- cpumask_copy(new_cpus, parent->effective_cpus);
+ if (is_in_v2_mode() && cpumask_empty(tmp->new_cpus)) {
+ cpumask_copy(tmp->new_cpus, parent->effective_cpus);
+ if (!cp->use_parent_ecpus) {
+ cp->use_parent_ecpus = true;
+ parent->child_ecpus_count++;
+ }
+ } else if (cp->use_parent_ecpus) {
+ cp->use_parent_ecpus = false;
+ WARN_ON_ONCE(!parent->child_ecpus_count);
+ parent->child_ecpus_count--;
+ }
- /* Skip the whole subtree if the cpumask remains the same. */
- if (cpumask_equal(new_cpus, cp->effective_cpus)) {
+ /*
+ * Skip the whole subtree if the cpumask remains the same
+ * and has no partition root state.
+ */
+ if (!cp->partition_root_state &&
+ cpumask_equal(tmp->new_cpus, cp->effective_cpus)) {
pos_css = css_rightmost_descendant(pos_css);
continue;
}
+ /*
+ * update_parent_subparts_cpumask() should have been called
+ * for cs already in update_cpumask(). We should also call
+ * update_tasks_cpumask() again for tasks in the parent
+ * cpuset if the parent's subparts_cpus changes.
+ */
+ if ((cp != cs) && cp->partition_root_state) {
+ switch (parent->partition_root_state) {
+ case PRS_DISABLED:
+ /*
+ * If parent is not a partition root or an
+ * invalid partition root, clear the state
+ * state and the CS_CPU_EXCLUSIVE flag.
+ */
+ WARN_ON_ONCE(cp->partition_root_state
+ != PRS_ERROR);
+ cp->partition_root_state = 0;
+
+ /*
+ * clear_bit() is an atomic operation and
+ * readers aren't interested in the state
+ * of CS_CPU_EXCLUSIVE anyway. So we can
+ * just update the flag without holding
+ * the callback_lock.
+ */
+ clear_bit(CS_CPU_EXCLUSIVE, &cp->flags);
+ break;
+
+ case PRS_ENABLED:
+ if (update_parent_subparts_cpumask(cp, partcmd_update, NULL, tmp))
+ update_tasks_cpumask(parent);
+ break;
+
+ case PRS_ERROR:
+ /*
+ * When parent is invalid, it has to be too.
+ */
+ cp->partition_root_state = PRS_ERROR;
+ if (cp->nr_subparts_cpus) {
+ cp->nr_subparts_cpus = 0;
+ cpumask_clear(cp->subparts_cpus);
+ }
+ break;
+ }
+ }
+
if (!css_tryget_online(&cp->css))
continue;
rcu_read_unlock();
spin_lock_irq(&callback_lock);
- cpumask_copy(cp->effective_cpus, new_cpus);
+
+ cpumask_copy(cp->effective_cpus, tmp->new_cpus);
+ if (cp->nr_subparts_cpus &&
+ (cp->partition_root_state != PRS_ENABLED)) {
+ cp->nr_subparts_cpus = 0;
+ cpumask_clear(cp->subparts_cpus);
+ } else if (cp->nr_subparts_cpus) {
+ /*
+ * Make sure that effective_cpus & subparts_cpus
+ * are mutually exclusive.
+ *
+ * In the unlikely event that effective_cpus
+ * becomes empty. we clear cp->nr_subparts_cpus and
+ * let its child partition roots to compete for
+ * CPUs again.
+ */
+ cpumask_andnot(cp->effective_cpus, cp->effective_cpus,
+ cp->subparts_cpus);
+ if (cpumask_empty(cp->effective_cpus)) {
+ cpumask_copy(cp->effective_cpus, tmp->new_cpus);
+ cpumask_clear(cp->subparts_cpus);
+ cp->nr_subparts_cpus = 0;
+ } else if (!cpumask_subset(cp->subparts_cpus,
+ tmp->new_cpus)) {
+ cpumask_andnot(cp->subparts_cpus,
+ cp->subparts_cpus, tmp->new_cpus);
+ cp->nr_subparts_cpus
+ = cpumask_weight(cp->subparts_cpus);
+ }
+ }
spin_unlock_irq(&callback_lock);
WARN_ON(!is_in_v2_mode() &&
@@ -932,11 +1418,15 @@
update_tasks_cpumask(cp);
/*
- * If the effective cpumask of any non-empty cpuset is changed,
- * we need to rebuild sched domains.
+ * On legacy hierarchy, if the effective cpumask of any non-
+ * empty cpuset is changed, we need to rebuild sched domains.
+ * On default hierarchy, the cpuset needs to be a partition
+ * root as well.
*/
if (!cpumask_empty(cp->cpus_allowed) &&
- is_sched_load_balance(cp))
+ is_sched_load_balance(cp) &&
+ (!cgroup_subsys_on_dfl(cpuset_cgrp_subsys) ||
+ is_partition_root(cp)))
need_rebuild_sched_domains = true;
rcu_read_lock();
@@ -949,6 +1439,35 @@
}
/**
+ * update_sibling_cpumasks - Update siblings cpumasks
+ * @parent: Parent cpuset
+ * @cs: Current cpuset
+ * @tmp: Temp variables
+ */
+static void update_sibling_cpumasks(struct cpuset *parent, struct cpuset *cs,
+ struct tmpmasks *tmp)
+{
+ struct cpuset *sibling;
+ struct cgroup_subsys_state *pos_css;
+
+ /*
+ * Check all its siblings and call update_cpumasks_hier()
+ * if their use_parent_ecpus flag is set in order for them
+ * to use the right effective_cpus value.
+ */
+ rcu_read_lock();
+ cpuset_for_each_child(sibling, pos_css, parent) {
+ if (sibling == cs)
+ continue;
+ if (!sibling->use_parent_ecpus)
+ continue;
+
+ update_cpumasks_hier(sibling, tmp);
+ }
+ rcu_read_unlock();
+}
+
+/**
* update_cpumask - update the cpus_allowed mask of a cpuset and all tasks in it
* @cs: the cpuset to consider
* @trialcs: trial cpuset
@@ -958,6 +1477,7 @@
const char *buf)
{
int retval;
+ struct tmpmasks tmp;
/* top_cpuset.cpus_allowed tracks cpu_online_mask; it's read-only */
if (cs == &top_cpuset)
@@ -989,12 +1509,50 @@
if (retval < 0)
return retval;
+#ifdef CONFIG_CPUMASK_OFFSTACK
+ /*
+ * Use the cpumasks in trialcs for tmpmasks when they are pointers
+ * to allocated cpumasks.
+ */
+ tmp.addmask = trialcs->subparts_cpus;
+ tmp.delmask = trialcs->effective_cpus;
+ tmp.new_cpus = trialcs->cpus_allowed;
+#endif
+
+ if (cs->partition_root_state) {
+ /* Cpumask of a partition root cannot be empty */
+ if (cpumask_empty(trialcs->cpus_allowed))
+ return -EINVAL;
+ if (update_parent_subparts_cpumask(cs, partcmd_update,
+ trialcs->cpus_allowed, &tmp) < 0)
+ return -EINVAL;
+ }
+
spin_lock_irq(&callback_lock);
cpumask_copy(cs->cpus_allowed, trialcs->cpus_allowed);
+
+ /*
+ * Make sure that subparts_cpus is a subset of cpus_allowed.
+ */
+ if (cs->nr_subparts_cpus) {
+ cpumask_andnot(cs->subparts_cpus, cs->subparts_cpus,
+ cs->cpus_allowed);
+ cs->nr_subparts_cpus = cpumask_weight(cs->subparts_cpus);
+ }
spin_unlock_irq(&callback_lock);
- /* use trialcs->cpus_allowed as a temp variable */
- update_cpumasks_hier(cs, trialcs->cpus_allowed);
+ update_cpumasks_hier(cs, &tmp);
+
+ if (cs->partition_root_state) {
+ struct cpuset *parent = parent_cs(cs);
+
+ /*
+ * For partition root, update the cpumasks of sibling
+ * cpusets if they use parent's effective_cpus.
+ */
+ if (parent->child_ecpus_count)
+ update_sibling_cpumasks(parent, cs, &tmp);
+ }
return 0;
}
@@ -1348,7 +1906,95 @@
if (spread_flag_changed)
update_tasks_flags(cs);
out:
- free_trial_cpuset(trialcs);
+ free_cpuset(trialcs);
+ return err;
+}
+
+/*
+ * update_prstate - update partititon_root_state
+ * cs: the cpuset to update
+ * val: 0 - disabled, 1 - enabled
+ *
+ * Call with cpuset_mutex held.
+ */
+static int update_prstate(struct cpuset *cs, int val)
+{
+ int err;
+ struct cpuset *parent = parent_cs(cs);
+ struct tmpmasks tmp;
+
+ if ((val != 0) && (val != 1))
+ return -EINVAL;
+ if (val == cs->partition_root_state)
+ return 0;
+
+ /*
+ * Cannot force a partial or invalid partition root to a full
+ * partition root.
+ */
+ if (val && cs->partition_root_state)
+ return -EINVAL;
+
+ if (alloc_cpumasks(NULL, &tmp))
+ return -ENOMEM;
+
+ err = -EINVAL;
+ if (!cs->partition_root_state) {
+ /*
+ * Turning on partition root requires setting the
+ * CS_CPU_EXCLUSIVE bit implicitly as well and cpus_allowed
+ * cannot be NULL.
+ */
+ if (cpumask_empty(cs->cpus_allowed))
+ goto out;
+
+ err = update_flag(CS_CPU_EXCLUSIVE, cs, 1);
+ if (err)
+ goto out;
+
+ err = update_parent_subparts_cpumask(cs, partcmd_enable,
+ NULL, &tmp);
+ if (err) {
+ update_flag(CS_CPU_EXCLUSIVE, cs, 0);
+ goto out;
+ }
+ cs->partition_root_state = PRS_ENABLED;
+ } else {
+ /*
+ * Turning off partition root will clear the
+ * CS_CPU_EXCLUSIVE bit.
+ */
+ if (cs->partition_root_state == PRS_ERROR) {
+ cs->partition_root_state = 0;
+ update_flag(CS_CPU_EXCLUSIVE, cs, 0);
+ err = 0;
+ goto out;
+ }
+
+ err = update_parent_subparts_cpumask(cs, partcmd_disable,
+ NULL, &tmp);
+ if (err)
+ goto out;
+
+ cs->partition_root_state = 0;
+
+ /* Turning off CS_CPU_EXCLUSIVE will not return error */
+ update_flag(CS_CPU_EXCLUSIVE, cs, 0);
+ }
+
+ /*
+ * Update cpumask of parent's tasks except when it is the top
+ * cpuset as some system daemons cannot be mapped to other CPUs.
+ */
+ if (parent != &top_cpuset)
+ update_tasks_cpumask(parent);
+
+ if (parent->child_ecpus_count)
+ update_sibling_cpumasks(parent, cs, &tmp);
+
+ rebuild_sched_domains_locked();
+out:
+ free_cpumasks(NULL, &tmp);
return err;
}
@@ -1467,7 +2113,7 @@
cpuset_attach_old_cs = task_cs(cgroup_taskset_first(tset, &css));
cs = css_cs(css);
- mutex_lock(&cpuset_mutex);
+ percpu_down_write(&cpuset_rwsem);
/* allow moving tasks into an empty cpuset if on default hierarchy */
ret = -ENOSPC;
@@ -1491,21 +2137,19 @@
cs->attach_in_progress++;
ret = 0;
out_unlock:
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
return ret;
}
static void cpuset_cancel_attach(struct cgroup_taskset *tset)
{
struct cgroup_subsys_state *css;
- struct cpuset *cs;
cgroup_taskset_first(tset, &css);
- cs = css_cs(css);
- mutex_lock(&cpuset_mutex);
+ percpu_down_write(&cpuset_rwsem);
css_cs(css)->attach_in_progress--;
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
}
/*
@@ -1528,7 +2172,7 @@
cgroup_taskset_first(tset, &css);
cs = css_cs(css);
- mutex_lock(&cpuset_mutex);
+ percpu_down_write(&cpuset_rwsem);
/* prepare for attach */
if (cs == &top_cpuset)
@@ -1582,7 +2226,7 @@
if (!cs->attach_in_progress)
wake_up(&cpuset_attach_wq);
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
}
/* The various types of files and directories in a cpuset file system */
@@ -1593,10 +2237,12 @@
FILE_MEMLIST,
FILE_EFFECTIVE_CPULIST,
FILE_EFFECTIVE_MEMLIST,
+ FILE_SUBPARTS_CPULIST,
FILE_CPU_EXCLUSIVE,
FILE_MEM_EXCLUSIVE,
FILE_MEM_HARDWALL,
FILE_SCHED_LOAD_BALANCE,
+ FILE_PARTITION_ROOT,
FILE_SCHED_RELAX_DOMAIN_LEVEL,
FILE_MEMORY_PRESSURE_ENABLED,
FILE_MEMORY_PRESSURE,
@@ -1611,7 +2257,8 @@
cpuset_filetype_t type = cft->private;
int retval = 0;
- mutex_lock(&cpuset_mutex);
+ get_online_cpus();
+ percpu_down_write(&cpuset_rwsem);
if (!is_cpuset_online(cs)) {
retval = -ENODEV;
goto out_unlock;
@@ -1647,7 +2294,8 @@
break;
}
out_unlock:
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
+ put_online_cpus();
return retval;
}
@@ -1658,7 +2306,8 @@
cpuset_filetype_t type = cft->private;
int retval = -ENODEV;
- mutex_lock(&cpuset_mutex);
+ get_online_cpus();
+ percpu_down_write(&cpuset_rwsem);
if (!is_cpuset_online(cs))
goto out_unlock;
@@ -1671,7 +2320,8 @@
break;
}
out_unlock:
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
+ put_online_cpus();
return retval;
}
@@ -1710,7 +2360,8 @@
kernfs_break_active_protection(of->kn);
flush_work(&cpuset_hotplug_work);
- mutex_lock(&cpuset_mutex);
+ get_online_cpus();
+ percpu_down_write(&cpuset_rwsem);
if (!is_cpuset_online(cs))
goto out_unlock;
@@ -1732,9 +2383,10 @@
break;
}
- free_trial_cpuset(trialcs);
+ free_cpuset(trialcs);
out_unlock:
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
+ put_online_cpus();
kernfs_unbreak_active_protection(of->kn);
css_put(&cs->css);
flush_workqueue(cpuset_migrate_mm_wq);
@@ -1770,6 +2422,9 @@
case FILE_EFFECTIVE_MEMLIST:
seq_printf(sf, "%*pbl\n", nodemask_pr_args(&cs->effective_mems));
break;
+ case FILE_SUBPARTS_CPULIST:
+ seq_printf(sf, "%*pbl\n", cpumask_pr_args(cs->subparts_cpus));
+ break;
default:
ret = -EINVAL;
}
@@ -1824,12 +2479,62 @@
return 0;
}
+static int sched_partition_show(struct seq_file *seq, void *v)
+{
+ struct cpuset *cs = css_cs(seq_css(seq));
+
+ switch (cs->partition_root_state) {
+ case PRS_ENABLED:
+ seq_puts(seq, "root\n");
+ break;
+ case PRS_DISABLED:
+ seq_puts(seq, "member\n");
+ break;
+ case PRS_ERROR:
+ seq_puts(seq, "root invalid\n");
+ break;
+ }
+ return 0;
+}
+
+static ssize_t sched_partition_write(struct kernfs_open_file *of, char *buf,
+ size_t nbytes, loff_t off)
+{
+ struct cpuset *cs = css_cs(of_css(of));
+ int val;
+ int retval = -ENODEV;
+
+ buf = strstrip(buf);
+
+ /*
+ * Convert "root" to ENABLED, and convert "member" to DISABLED.
+ */
+ if (!strcmp(buf, "root"))
+ val = PRS_ENABLED;
+ else if (!strcmp(buf, "member"))
+ val = PRS_DISABLED;
+ else
+ return -EINVAL;
+
+ css_get(&cs->css);
+ get_online_cpus();
+ percpu_down_write(&cpuset_rwsem);
+ if (!is_cpuset_online(cs))
+ goto out_unlock;
+
+ retval = update_prstate(cs, val);
+out_unlock:
+ percpu_up_write(&cpuset_rwsem);
+ put_online_cpus();
+ css_put(&cs->css);
+ return retval ?: nbytes;
+}
/*
* for the common functions, 'private' gives the type of file
*/
-static struct cftype files[] = {
+static struct cftype legacy_files[] = {
{
.name = "cpus",
.seq_show = cpuset_common_seq_show,
@@ -1932,6 +2637,60 @@
};
/*
+ * This is currently a minimal set for the default hierarchy. It can be
+ * expanded later on by migrating more features and control files from v1.
+ */
+static struct cftype dfl_files[] = {
+ {
+ .name = "cpus",
+ .seq_show = cpuset_common_seq_show,
+ .write = cpuset_write_resmask,
+ .max_write_len = (100U + 6 * NR_CPUS),
+ .private = FILE_CPULIST,
+ .flags = CFTYPE_NOT_ON_ROOT,
+ },
+
+ {
+ .name = "mems",
+ .seq_show = cpuset_common_seq_show,
+ .write = cpuset_write_resmask,
+ .max_write_len = (100U + 6 * MAX_NUMNODES),
+ .private = FILE_MEMLIST,
+ .flags = CFTYPE_NOT_ON_ROOT,
+ },
+
+ {
+ .name = "cpus.effective",
+ .seq_show = cpuset_common_seq_show,
+ .private = FILE_EFFECTIVE_CPULIST,
+ },
+
+ {
+ .name = "mems.effective",
+ .seq_show = cpuset_common_seq_show,
+ .private = FILE_EFFECTIVE_MEMLIST,
+ },
+
+ {
+ .name = "cpus.partition",
+ .seq_show = sched_partition_show,
+ .write = sched_partition_write,
+ .private = FILE_PARTITION_ROOT,
+ .flags = CFTYPE_NOT_ON_ROOT,
+ },
+
+ {
+ .name = "cpus.subpartitions",
+ .seq_show = cpuset_common_seq_show,
+ .private = FILE_SUBPARTS_CPULIST,
+ .flags = CFTYPE_DEBUG,
+ },
+
+ { } /* terminate */
+};
+
+
+/*
* cpuset_css_alloc - allocate a cpuset css
* cgrp: control group that the new cpuset will be part of
*/
@@ -1947,26 +2706,19 @@
cs = kzalloc(sizeof(*cs), GFP_KERNEL);
if (!cs)
return ERR_PTR(-ENOMEM);
- if (!alloc_cpumask_var(&cs->cpus_allowed, GFP_KERNEL))
- goto free_cs;
- if (!alloc_cpumask_var(&cs->effective_cpus, GFP_KERNEL))
- goto free_cpus;
+
+ if (alloc_cpumasks(cs, NULL)) {
+ kfree(cs);
+ return ERR_PTR(-ENOMEM);
+ }
set_bit(CS_SCHED_LOAD_BALANCE, &cs->flags);
- cpumask_clear(cs->cpus_allowed);
nodes_clear(cs->mems_allowed);
- cpumask_clear(cs->effective_cpus);
nodes_clear(cs->effective_mems);
fmeter_init(&cs->fmeter);
cs->relax_domain_level = -1;
return &cs->css;
-
-free_cpus:
- free_cpumask_var(cs->cpus_allowed);
-free_cs:
- kfree(cs);
- return ERR_PTR(-ENOMEM);
}
static int cpuset_css_online(struct cgroup_subsys_state *css)
@@ -1979,7 +2731,8 @@
if (!parent)
return 0;
- mutex_lock(&cpuset_mutex);
+ get_online_cpus();
+ percpu_down_write(&cpuset_rwsem);
set_bit(CS_ONLINE, &cs->flags);
if (is_spread_page(parent))
@@ -1993,6 +2746,8 @@
if (is_in_v2_mode()) {
cpumask_copy(cs->effective_cpus, parent->effective_cpus);
cs->effective_mems = parent->effective_mems;
+ cs->use_parent_ecpus = true;
+ parent->child_ecpus_count++;
}
spin_unlock_irq(&callback_lock);
@@ -2028,43 +2783,60 @@
cpumask_copy(cs->effective_cpus, parent->cpus_allowed);
spin_unlock_irq(&callback_lock);
out_unlock:
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
+ put_online_cpus();
return 0;
}
/*
* If the cpuset being removed has its flag 'sched_load_balance'
* enabled, then simulate turning sched_load_balance off, which
- * will call rebuild_sched_domains_locked().
+ * will call rebuild_sched_domains_locked(). That is not needed
+ * in the default hierarchy where only changes in partition
+ * will cause repartitioning.
+ *
+ * If the cpuset has the 'sched.partition' flag enabled, simulate
+ * turning 'sched.partition" off.
*/
static void cpuset_css_offline(struct cgroup_subsys_state *css)
{
struct cpuset *cs = css_cs(css);
- mutex_lock(&cpuset_mutex);
+ get_online_cpus();
+ percpu_down_write(&cpuset_rwsem);
- if (is_sched_load_balance(cs))
+ if (is_partition_root(cs))
+ update_prstate(cs, 0);
+
+ if (!cgroup_subsys_on_dfl(cpuset_cgrp_subsys) &&
+ is_sched_load_balance(cs))
update_flag(CS_SCHED_LOAD_BALANCE, cs, 0);
+ if (cs->use_parent_ecpus) {
+ struct cpuset *parent = parent_cs(cs);
+
+ cs->use_parent_ecpus = false;
+ parent->child_ecpus_count--;
+ }
+
cpuset_dec();
clear_bit(CS_ONLINE, &cs->flags);
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
+ put_online_cpus();
}
static void cpuset_css_free(struct cgroup_subsys_state *css)
{
struct cpuset *cs = css_cs(css);
- free_cpumask_var(cs->effective_cpus);
- free_cpumask_var(cs->cpus_allowed);
- kfree(cs);
+ free_cpuset(cs);
}
static void cpuset_bind(struct cgroup_subsys_state *root_css)
{
- mutex_lock(&cpuset_mutex);
+ percpu_down_write(&cpuset_rwsem);
spin_lock_irq(&callback_lock);
if (is_in_v2_mode()) {
@@ -2077,7 +2849,7 @@
}
spin_unlock_irq(&callback_lock);
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
}
/*
@@ -2090,7 +2862,7 @@
if (task_css_is_root(task, cpuset_cgrp_id))
return;
- set_cpus_allowed_ptr(task, ¤t->cpus_allowed);
+ set_cpus_allowed_ptr(task, current->cpus_ptr);
task->mems_allowed = current->mems_allowed;
}
@@ -2105,22 +2877,25 @@
.post_attach = cpuset_post_attach,
.bind = cpuset_bind,
.fork = cpuset_fork,
- .legacy_cftypes = files,
+ .legacy_cftypes = legacy_files,
+ .dfl_cftypes = dfl_files,
.early_init = true,
+ .threaded = true,
};
/**
* cpuset_init - initialize cpusets at system boot
*
- * Description: Initialize top_cpuset and the cpuset internal file system,
+ * Description: Initialize top_cpuset
**/
int __init cpuset_init(void)
{
- int err = 0;
+ BUG_ON(percpu_init_rwsem(&cpuset_rwsem));
BUG_ON(!alloc_cpumask_var(&top_cpuset.cpus_allowed, GFP_KERNEL));
BUG_ON(!alloc_cpumask_var(&top_cpuset.effective_cpus, GFP_KERNEL));
+ BUG_ON(!zalloc_cpumask_var(&top_cpuset.subparts_cpus, GFP_KERNEL));
cpumask_setall(top_cpuset.cpus_allowed);
nodes_setall(top_cpuset.mems_allowed);
@@ -2131,10 +2906,6 @@
set_bit(CS_SCHED_LOAD_BALANCE, &top_cpuset.flags);
top_cpuset.relax_domain_level = -1;
- err = register_filesystem(&cpuset_fs_type);
- if (err < 0)
- return err;
-
BUG_ON(!alloc_cpumask_var(&cpus_attach, GFP_KERNEL));
return 0;
@@ -2193,7 +2964,7 @@
is_empty = cpumask_empty(cs->cpus_allowed) ||
nodes_empty(cs->mems_allowed);
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
/*
* Move tasks to the nearest ancestor with execution resources,
@@ -2203,7 +2974,7 @@
if (is_empty)
remove_tasks_in_empty_cpuset(cs);
- mutex_lock(&cpuset_mutex);
+ percpu_down_write(&cpuset_rwsem);
}
static void
@@ -2227,37 +2998,97 @@
update_tasks_nodemask(cs);
}
+static bool force_rebuild;
+
+void cpuset_force_rebuild(void)
+{
+ force_rebuild = true;
+}
+
/**
* cpuset_hotplug_update_tasks - update tasks in a cpuset for hotunplug
* @cs: cpuset in interest
+ * @tmp: the tmpmasks structure pointer
*
* Compare @cs's cpu and mem masks against top_cpuset and if some have gone
* offline, update @cs accordingly. If @cs ends up with no CPU or memory,
* all its tasks are moved to the nearest ancestor with both resources.
*/
-static void cpuset_hotplug_update_tasks(struct cpuset *cs)
+static void cpuset_hotplug_update_tasks(struct cpuset *cs, struct tmpmasks *tmp)
{
static cpumask_t new_cpus;
static nodemask_t new_mems;
bool cpus_updated;
bool mems_updated;
+ struct cpuset *parent;
retry:
wait_event(cpuset_attach_wq, cs->attach_in_progress == 0);
- mutex_lock(&cpuset_mutex);
+ percpu_down_write(&cpuset_rwsem);
/*
* We have raced with task attaching. We wait until attaching
* is finished, so we won't attach a task to an empty cpuset.
*/
if (cs->attach_in_progress) {
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
goto retry;
}
- cpumask_and(&new_cpus, cs->cpus_allowed, parent_cs(cs)->effective_cpus);
- nodes_and(new_mems, cs->mems_allowed, parent_cs(cs)->effective_mems);
+ parent = parent_cs(cs);
+ compute_effective_cpumask(&new_cpus, cs, parent);
+ nodes_and(new_mems, cs->mems_allowed, parent->effective_mems);
+ if (cs->nr_subparts_cpus)
+ /*
+ * Make sure that CPUs allocated to child partitions
+ * do not show up in effective_cpus.
+ */
+ cpumask_andnot(&new_cpus, &new_cpus, cs->subparts_cpus);
+
+ if (!tmp || !cs->partition_root_state)
+ goto update_tasks;
+
+ /*
+ * In the unlikely event that a partition root has empty
+ * effective_cpus or its parent becomes erroneous, we have to
+ * transition it to the erroneous state.
+ */
+ if (is_partition_root(cs) && (cpumask_empty(&new_cpus) ||
+ (parent->partition_root_state == PRS_ERROR))) {
+ if (cs->nr_subparts_cpus) {
+ cs->nr_subparts_cpus = 0;
+ cpumask_clear(cs->subparts_cpus);
+ compute_effective_cpumask(&new_cpus, cs, parent);
+ }
+
+ /*
+ * If the effective_cpus is empty because the child
+ * partitions take away all the CPUs, we can keep
+ * the current partition and let the child partitions
+ * fight for available CPUs.
+ */
+ if ((parent->partition_root_state == PRS_ERROR) ||
+ cpumask_empty(&new_cpus)) {
+ update_parent_subparts_cpumask(cs, partcmd_disable,
+ NULL, tmp);
+ cs->partition_root_state = PRS_ERROR;
+ }
+ cpuset_force_rebuild();
+ }
+
+ /*
+ * On the other hand, an erroneous partition root may be transitioned
+ * back to a regular one or a partition root with no CPU allocated
+ * from the parent may change to erroneous.
+ */
+ if (is_partition_root(parent) &&
+ ((cs->partition_root_state == PRS_ERROR) ||
+ !cpumask_intersects(&new_cpus, parent->subparts_cpus)) &&
+ update_parent_subparts_cpumask(cs, partcmd_update, NULL, tmp))
+ cpuset_force_rebuild();
+
+update_tasks:
cpus_updated = !cpumask_equal(&new_cpus, cs->effective_cpus);
mems_updated = !nodes_equal(new_mems, cs->effective_mems);
@@ -2268,14 +3099,7 @@
hotplug_update_tasks_legacy(cs, &new_cpus, &new_mems,
cpus_updated, mems_updated);
- mutex_unlock(&cpuset_mutex);
-}
-
-static bool force_rebuild;
-
-void cpuset_force_rebuild(void)
-{
- force_rebuild = true;
+ percpu_up_write(&cpuset_rwsem);
}
/**
@@ -2300,13 +3124,22 @@
static nodemask_t new_mems;
bool cpus_updated, mems_updated;
bool on_dfl = is_in_v2_mode();
+ struct tmpmasks tmp, *ptmp = NULL;
- mutex_lock(&cpuset_mutex);
+ if (on_dfl && !alloc_cpumasks(NULL, &tmp))
+ ptmp = &tmp;
+
+ percpu_down_write(&cpuset_rwsem);
/* fetch the available cpus/mems and find out which changed how */
cpumask_copy(&new_cpus, cpu_active_mask);
new_mems = node_states[N_MEMORY];
+ /*
+ * If subparts_cpus is populated, it is likely that the check below
+ * will produce a false positive on cpus_updated when the cpu list
+ * isn't changed. It is extra work, but it is better to be safe.
+ */
cpus_updated = !cpumask_equal(top_cpuset.effective_cpus, &new_cpus);
mems_updated = !nodes_equal(top_cpuset.effective_mems, new_mems);
@@ -2315,6 +3148,22 @@
spin_lock_irq(&callback_lock);
if (!on_dfl)
cpumask_copy(top_cpuset.cpus_allowed, &new_cpus);
+ /*
+ * Make sure that CPUs allocated to child partitions
+ * do not show up in effective_cpus. If no CPU is left,
+ * we clear the subparts_cpus & let the child partitions
+ * fight for the CPUs again.
+ */
+ if (top_cpuset.nr_subparts_cpus) {
+ if (cpumask_subset(&new_cpus,
+ top_cpuset.subparts_cpus)) {
+ top_cpuset.nr_subparts_cpus = 0;
+ cpumask_clear(top_cpuset.subparts_cpus);
+ } else {
+ cpumask_andnot(&new_cpus, &new_cpus,
+ top_cpuset.subparts_cpus);
+ }
+ }
cpumask_copy(top_cpuset.effective_cpus, &new_cpus);
spin_unlock_irq(&callback_lock);
/* we don't mess with cpumasks of tasks in top_cpuset */
@@ -2330,7 +3179,7 @@
update_tasks_nodemask(&top_cpuset);
}
- mutex_unlock(&cpuset_mutex);
+ percpu_up_write(&cpuset_rwsem);
/* if cpus or mems changed, we need to propagate to descendants */
if (cpus_updated || mems_updated) {
@@ -2343,7 +3192,7 @@
continue;
rcu_read_unlock();
- cpuset_hotplug_update_tasks(cs);
+ cpuset_hotplug_update_tasks(cs, ptmp);
rcu_read_lock();
css_put(&cs->css);
@@ -2356,6 +3205,8 @@
force_rebuild = false;
rebuild_sched_domains();
}
+
+ free_cpumasks(NULL, ptmp);
}
void cpuset_update_active_cpus(void)
@@ -2432,10 +3283,23 @@
spin_unlock_irqrestore(&callback_lock, flags);
}
+/**
+ * cpuset_cpus_allowed_fallback - final fallback before complete catastrophe.
+ * @tsk: pointer to task_struct with which the scheduler is struggling
+ *
+ * Description: In the case that the scheduler cannot find an allowed cpu in
+ * tsk->cpus_allowed, we fall back to task_cs(tsk)->cpus_allowed. In legacy
+ * mode however, this value is the same as task_cs(tsk)->effective_cpus,
+ * which will not contain a sane cpumask during cases such as cpu hotplugging.
+ * This is the absolute last resort for the scheduler and it is only used if
+ * _every_ other avenue has been traveled.
+ **/
+
void cpuset_cpus_allowed_fallback(struct task_struct *tsk)
{
rcu_read_lock();
- do_set_cpus_allowed(tsk, task_cs(tsk)->effective_cpus);
+ do_set_cpus_allowed(tsk, is_in_v2_mode() ?
+ task_cs(tsk)->cpus_allowed : cpu_possible_mask);
rcu_read_unlock();
/*
@@ -2666,9 +3530,9 @@
rcu_read_lock();
cgrp = task_cs(current)->css.cgroup;
- pr_info("%s cpuset=", current->comm);
+ pr_cont(",cpuset=");
pr_cont_cgroup_name(cgrp);
- pr_cont(" mems_allowed=%*pbl\n",
+ pr_cont(",mems_allowed=%*pbl",
nodemask_pr_args(¤t->mems_allowed));
rcu_read_unlock();
diff --git a/kernel/cgroup/debug.c b/kernel/cgroup/debug.c
index 9caeda6..80aa3f0 100644
--- a/kernel/cgroup/debug.c
+++ b/kernel/cgroup/debug.c
@@ -64,8 +64,8 @@
css = cset->subsys[ss->id];
if (!css)
continue;
- seq_printf(seq, "%2d: %-4s\t- %lx[%d]\n", ss->id, ss->name,
- (unsigned long)css, css->id);
+ seq_printf(seq, "%2d: %-4s\t- %p[%d]\n", ss->id, ss->name,
+ css, css->id);
}
rcu_read_unlock();
spin_unlock_irq(&css_set_lock);
@@ -224,8 +224,8 @@
if (css->parent)
snprintf(pbuf, sizeof(pbuf) - 1, " P=%d",
css->parent->id);
- seq_printf(seq, "%2d: %-4s\t- %lx[%d] %d%s\n", ss->id, ss->name,
- (unsigned long)css, css->id,
+ seq_printf(seq, "%2d: %-4s\t- %p[%d] %d%s\n", ss->id, ss->name,
+ css, css->id,
atomic_read(&css->online_cnt), pbuf);
}
@@ -373,11 +373,9 @@
* On v2, debug is an implicit controller enabled by "cgroup_debug" boot
* parameter.
*/
-static int __init enable_cgroup_debug(char *str)
+void __init enable_debug_cgroup(void)
{
debug_cgrp_subsys.dfl_cftypes = debug_files;
debug_cgrp_subsys.implicit_on_dfl = true;
debug_cgrp_subsys.threaded = true;
- return 1;
}
-__setup("cgroup_debug", enable_cgroup_debug);
diff --git a/kernel/cgroup/freezer.c b/kernel/cgroup/freezer.c
index 0823679..8cf0106 100644
--- a/kernel/cgroup/freezer.c
+++ b/kernel/cgroup/freezer.c
@@ -1,481 +1,314 @@
-/*
- * cgroup_freezer.c - control group freezer subsystem
- *
- * Copyright IBM Corporation, 2007
- *
- * Author : Cedric Le Goater <clg@fr.ibm.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of version 2.1 of the GNU Lesser General Public License
- * as published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it would be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
- */
-
-#include <linux/export.h>
-#include <linux/slab.h>
+//SPDX-License-Identifier: GPL-2.0
#include <linux/cgroup.h>
-#include <linux/fs.h>
-#include <linux/uaccess.h>
-#include <linux/freezer.h>
-#include <linux/seq_file.h>
-#include <linux/mutex.h>
+#include <linux/sched.h>
+#include <linux/sched/task.h>
+#include <linux/sched/signal.h>
+
+#include "cgroup-internal.h"
+
+#include <trace/events/cgroup.h>
/*
- * A cgroup is freezing if any FREEZING flags are set. FREEZING_SELF is
- * set if "FROZEN" is written to freezer.state cgroupfs file, and cleared
- * for "THAWED". FREEZING_PARENT is set if the parent freezer is FREEZING
- * for whatever reason. IOW, a cgroup has FREEZING_PARENT set if one of
- * its ancestors has FREEZING_SELF set.
+ * Propagate the cgroup frozen state upwards by the cgroup tree.
*/
-enum freezer_state_flags {
- CGROUP_FREEZER_ONLINE = (1 << 0), /* freezer is fully online */
- CGROUP_FREEZING_SELF = (1 << 1), /* this freezer is freezing */
- CGROUP_FREEZING_PARENT = (1 << 2), /* the parent freezer is freezing */
- CGROUP_FROZEN = (1 << 3), /* this and its descendants frozen */
-
- /* mask for all FREEZING flags */
- CGROUP_FREEZING = CGROUP_FREEZING_SELF | CGROUP_FREEZING_PARENT,
-};
-
-struct freezer {
- struct cgroup_subsys_state css;
- unsigned int state;
-};
-
-static DEFINE_MUTEX(freezer_mutex);
-
-static inline struct freezer *css_freezer(struct cgroup_subsys_state *css)
+static void cgroup_propagate_frozen(struct cgroup *cgrp, bool frozen)
{
- return css ? container_of(css, struct freezer, css) : NULL;
-}
-
-static inline struct freezer *task_freezer(struct task_struct *task)
-{
- return css_freezer(task_css(task, freezer_cgrp_id));
-}
-
-static struct freezer *parent_freezer(struct freezer *freezer)
-{
- return css_freezer(freezer->css.parent);
-}
-
-bool cgroup_freezing(struct task_struct *task)
-{
- bool ret;
-
- rcu_read_lock();
- ret = task_freezer(task)->state & CGROUP_FREEZING;
- rcu_read_unlock();
-
- return ret;
-}
-
-static const char *freezer_state_strs(unsigned int state)
-{
- if (state & CGROUP_FROZEN)
- return "FROZEN";
- if (state & CGROUP_FREEZING)
- return "FREEZING";
- return "THAWED";
-};
-
-static struct cgroup_subsys_state *
-freezer_css_alloc(struct cgroup_subsys_state *parent_css)
-{
- struct freezer *freezer;
-
- freezer = kzalloc(sizeof(struct freezer), GFP_KERNEL);
- if (!freezer)
- return ERR_PTR(-ENOMEM);
-
- return &freezer->css;
-}
-
-/**
- * freezer_css_online - commit creation of a freezer css
- * @css: css being created
- *
- * We're committing to creation of @css. Mark it online and inherit
- * parent's freezing state while holding both parent's and our
- * freezer->lock.
- */
-static int freezer_css_online(struct cgroup_subsys_state *css)
-{
- struct freezer *freezer = css_freezer(css);
- struct freezer *parent = parent_freezer(freezer);
-
- mutex_lock(&freezer_mutex);
-
- freezer->state |= CGROUP_FREEZER_ONLINE;
-
- if (parent && (parent->state & CGROUP_FREEZING)) {
- freezer->state |= CGROUP_FREEZING_PARENT | CGROUP_FROZEN;
- atomic_inc(&system_freezing_cnt);
- }
-
- mutex_unlock(&freezer_mutex);
- return 0;
-}
-
-/**
- * freezer_css_offline - initiate destruction of a freezer css
- * @css: css being destroyed
- *
- * @css is going away. Mark it dead and decrement system_freezing_count if
- * it was holding one.
- */
-static void freezer_css_offline(struct cgroup_subsys_state *css)
-{
- struct freezer *freezer = css_freezer(css);
-
- mutex_lock(&freezer_mutex);
-
- if (freezer->state & CGROUP_FREEZING)
- atomic_dec(&system_freezing_cnt);
-
- freezer->state = 0;
-
- mutex_unlock(&freezer_mutex);
-}
-
-static void freezer_css_free(struct cgroup_subsys_state *css)
-{
- kfree(css_freezer(css));
-}
-
-/*
- * Tasks can be migrated into a different freezer anytime regardless of its
- * current state. freezer_attach() is responsible for making new tasks
- * conform to the current state.
- *
- * Freezer state changes and task migration are synchronized via
- * @freezer->lock. freezer_attach() makes the new tasks conform to the
- * current state and all following state changes can see the new tasks.
- */
-static void freezer_attach(struct cgroup_taskset *tset)
-{
- struct task_struct *task;
- struct cgroup_subsys_state *new_css;
-
- mutex_lock(&freezer_mutex);
+ int desc = 1;
/*
- * Make the new tasks conform to the current state of @new_css.
- * For simplicity, when migrating any task to a FROZEN cgroup, we
- * revert it to FREEZING and let update_if_frozen() determine the
- * correct state later.
+ * If the new state is frozen, some freezing ancestor cgroups may change
+ * their state too, depending on if all their descendants are frozen.
*
- * Tasks in @tset are on @new_css but may not conform to its
- * current state before executing the following - !frozen tasks may
- * be visible in a FROZEN cgroup and frozen tasks in a THAWED one.
+ * Otherwise, all ancestor cgroups are forced into the non-frozen state.
*/
- cgroup_taskset_for_each(task, new_css, tset) {
- struct freezer *freezer = css_freezer(new_css);
-
- if (!(freezer->state & CGROUP_FREEZING)) {
- __thaw_task(task);
+ while ((cgrp = cgroup_parent(cgrp))) {
+ if (frozen) {
+ cgrp->freezer.nr_frozen_descendants += desc;
+ if (!test_bit(CGRP_FROZEN, &cgrp->flags) &&
+ test_bit(CGRP_FREEZE, &cgrp->flags) &&
+ cgrp->freezer.nr_frozen_descendants ==
+ cgrp->nr_descendants) {
+ set_bit(CGRP_FROZEN, &cgrp->flags);
+ cgroup_file_notify(&cgrp->events_file);
+ TRACE_CGROUP_PATH(notify_frozen, cgrp, 1);
+ desc++;
+ }
} else {
- freeze_task(task);
- /* clear FROZEN and propagate upwards */
- while (freezer && (freezer->state & CGROUP_FROZEN)) {
- freezer->state &= ~CGROUP_FROZEN;
- freezer = parent_freezer(freezer);
+ cgrp->freezer.nr_frozen_descendants -= desc;
+ if (test_bit(CGRP_FROZEN, &cgrp->flags)) {
+ clear_bit(CGRP_FROZEN, &cgrp->flags);
+ cgroup_file_notify(&cgrp->events_file);
+ TRACE_CGROUP_PATH(notify_frozen, cgrp, 0);
+ desc++;
}
}
}
-
- mutex_unlock(&freezer_mutex);
}
-/**
- * freezer_fork - cgroup post fork callback
- * @task: a task which has just been forked
- *
- * @task has just been created and should conform to the current state of
- * the cgroup_freezer it belongs to. This function may race against
- * freezer_attach(). Losing to freezer_attach() means that we don't have
- * to do anything as freezer_attach() will put @task into the appropriate
- * state.
+/*
+ * Revisit the cgroup frozen state.
+ * Checks if the cgroup is really frozen and perform all state transitions.
*/
-static void freezer_fork(struct task_struct *task)
+void cgroup_update_frozen(struct cgroup *cgrp)
{
- struct freezer *freezer;
+ bool frozen;
+
+ lockdep_assert_held(&css_set_lock);
/*
- * The root cgroup is non-freezable, so we can skip locking the
- * freezer. This is safe regardless of race with task migration.
- * If we didn't race or won, skipping is obviously the right thing
- * to do. If we lost and root is the new cgroup, noop is still the
- * right thing to do.
+ * If the cgroup has to be frozen (CGRP_FREEZE bit set),
+ * and all tasks are frozen and/or stopped, let's consider
+ * the cgroup frozen. Otherwise it's not frozen.
*/
- if (task_css_is_root(task, freezer_cgrp_id))
- return;
+ frozen = test_bit(CGRP_FREEZE, &cgrp->flags) &&
+ cgrp->freezer.nr_frozen_tasks == __cgroup_task_count(cgrp);
- mutex_lock(&freezer_mutex);
- rcu_read_lock();
-
- freezer = task_freezer(task);
- if (freezer->state & CGROUP_FREEZING)
- freeze_task(task);
-
- rcu_read_unlock();
- mutex_unlock(&freezer_mutex);
-}
-
-/**
- * update_if_frozen - update whether a cgroup finished freezing
- * @css: css of interest
- *
- * Once FREEZING is initiated, transition to FROZEN is lazily updated by
- * calling this function. If the current state is FREEZING but not FROZEN,
- * this function checks whether all tasks of this cgroup and the descendant
- * cgroups finished freezing and, if so, sets FROZEN.
- *
- * The caller is responsible for grabbing RCU read lock and calling
- * update_if_frozen() on all descendants prior to invoking this function.
- *
- * Task states and freezer state might disagree while tasks are being
- * migrated into or out of @css, so we can't verify task states against
- * @freezer state here. See freezer_attach() for details.
- */
-static void update_if_frozen(struct cgroup_subsys_state *css)
-{
- struct freezer *freezer = css_freezer(css);
- struct cgroup_subsys_state *pos;
- struct css_task_iter it;
- struct task_struct *task;
-
- lockdep_assert_held(&freezer_mutex);
-
- if (!(freezer->state & CGROUP_FREEZING) ||
- (freezer->state & CGROUP_FROZEN))
- return;
-
- /* are all (live) children frozen? */
- rcu_read_lock();
- css_for_each_child(pos, css) {
- struct freezer *child = css_freezer(pos);
-
- if ((child->state & CGROUP_FREEZER_ONLINE) &&
- !(child->state & CGROUP_FROZEN)) {
- rcu_read_unlock();
+ if (frozen) {
+ /* Already there? */
+ if (test_bit(CGRP_FROZEN, &cgrp->flags))
return;
- }
+
+ set_bit(CGRP_FROZEN, &cgrp->flags);
+ } else {
+ /* Already there? */
+ if (!test_bit(CGRP_FROZEN, &cgrp->flags))
+ return;
+
+ clear_bit(CGRP_FROZEN, &cgrp->flags);
}
- rcu_read_unlock();
+ cgroup_file_notify(&cgrp->events_file);
+ TRACE_CGROUP_PATH(notify_frozen, cgrp, frozen);
- /* are all tasks frozen? */
- css_task_iter_start(css, 0, &it);
-
- while ((task = css_task_iter_next(&it))) {
- if (freezing(task)) {
- /*
- * freezer_should_skip() indicates that the task
- * should be skipped when determining freezing
- * completion. Consider it frozen in addition to
- * the usual frozen condition.
- */
- if (!frozen(task) && !freezer_should_skip(task))
- goto out_iter_end;
- }
- }
-
- freezer->state |= CGROUP_FROZEN;
-out_iter_end:
- css_task_iter_end(&it);
+ /* Update the state of ancestor cgroups. */
+ cgroup_propagate_frozen(cgrp, frozen);
}
-static int freezer_read(struct seq_file *m, void *v)
-{
- struct cgroup_subsys_state *css = seq_css(m), *pos;
-
- mutex_lock(&freezer_mutex);
- rcu_read_lock();
-
- /* update states bottom-up */
- css_for_each_descendant_post(pos, css) {
- if (!css_tryget_online(pos))
- continue;
- rcu_read_unlock();
-
- update_if_frozen(pos);
-
- rcu_read_lock();
- css_put(pos);
- }
-
- rcu_read_unlock();
- mutex_unlock(&freezer_mutex);
-
- seq_puts(m, freezer_state_strs(css_freezer(css)->state));
- seq_putc(m, '\n');
- return 0;
-}
-
-static void freeze_cgroup(struct freezer *freezer)
-{
- struct css_task_iter it;
- struct task_struct *task;
-
- css_task_iter_start(&freezer->css, 0, &it);
- while ((task = css_task_iter_next(&it)))
- freeze_task(task);
- css_task_iter_end(&it);
-}
-
-static void unfreeze_cgroup(struct freezer *freezer)
-{
- struct css_task_iter it;
- struct task_struct *task;
-
- css_task_iter_start(&freezer->css, 0, &it);
- while ((task = css_task_iter_next(&it)))
- __thaw_task(task);
- css_task_iter_end(&it);
-}
-
-/**
- * freezer_apply_state - apply state change to a single cgroup_freezer
- * @freezer: freezer to apply state change to
- * @freeze: whether to freeze or unfreeze
- * @state: CGROUP_FREEZING_* flag to set or clear
- *
- * Set or clear @state on @cgroup according to @freeze, and perform
- * freezing or thawing as necessary.
+/*
+ * Increment cgroup's nr_frozen_tasks.
*/
-static void freezer_apply_state(struct freezer *freezer, bool freeze,
- unsigned int state)
+static void cgroup_inc_frozen_cnt(struct cgroup *cgrp)
{
- /* also synchronizes against task migration, see freezer_attach() */
- lockdep_assert_held(&freezer_mutex);
+ cgrp->freezer.nr_frozen_tasks++;
+}
- if (!(freezer->state & CGROUP_FREEZER_ONLINE))
+/*
+ * Decrement cgroup's nr_frozen_tasks.
+ */
+static void cgroup_dec_frozen_cnt(struct cgroup *cgrp)
+{
+ cgrp->freezer.nr_frozen_tasks--;
+ WARN_ON_ONCE(cgrp->freezer.nr_frozen_tasks < 0);
+}
+
+/*
+ * Enter frozen/stopped state, if not yet there. Update cgroup's counters,
+ * and revisit the state of the cgroup, if necessary.
+ */
+void cgroup_enter_frozen(void)
+{
+ struct cgroup *cgrp;
+
+ if (current->frozen)
+ return;
+
+ spin_lock_irq(&css_set_lock);
+ current->frozen = true;
+ cgrp = task_dfl_cgroup(current);
+ cgroup_inc_frozen_cnt(cgrp);
+ cgroup_update_frozen(cgrp);
+ spin_unlock_irq(&css_set_lock);
+}
+
+/*
+ * Conditionally leave frozen/stopped state. Update cgroup's counters,
+ * and revisit the state of the cgroup, if necessary.
+ *
+ * If always_leave is not set, and the cgroup is freezing,
+ * we're racing with the cgroup freezing. In this case, we don't
+ * drop the frozen counter to avoid a transient switch to
+ * the unfrozen state.
+ */
+void cgroup_leave_frozen(bool always_leave)
+{
+ struct cgroup *cgrp;
+
+ spin_lock_irq(&css_set_lock);
+ cgrp = task_dfl_cgroup(current);
+ if (always_leave || !test_bit(CGRP_FREEZE, &cgrp->flags)) {
+ cgroup_dec_frozen_cnt(cgrp);
+ cgroup_update_frozen(cgrp);
+ WARN_ON_ONCE(!current->frozen);
+ current->frozen = false;
+ } else if (!(current->jobctl & JOBCTL_TRAP_FREEZE)) {
+ spin_lock(¤t->sighand->siglock);
+ current->jobctl |= JOBCTL_TRAP_FREEZE;
+ set_thread_flag(TIF_SIGPENDING);
+ spin_unlock(¤t->sighand->siglock);
+ }
+ spin_unlock_irq(&css_set_lock);
+}
+
+/*
+ * Freeze or unfreeze the task by setting or clearing the JOBCTL_TRAP_FREEZE
+ * jobctl bit.
+ */
+static void cgroup_freeze_task(struct task_struct *task, bool freeze)
+{
+ unsigned long flags;
+
+ /* If the task is about to die, don't bother with freezing it. */
+ if (!lock_task_sighand(task, &flags))
return;
if (freeze) {
- if (!(freezer->state & CGROUP_FREEZING))
- atomic_inc(&system_freezing_cnt);
- freezer->state |= state;
- freeze_cgroup(freezer);
+ task->jobctl |= JOBCTL_TRAP_FREEZE;
+ signal_wake_up(task, false);
} else {
- bool was_freezing = freezer->state & CGROUP_FREEZING;
-
- freezer->state &= ~state;
-
- if (!(freezer->state & CGROUP_FREEZING)) {
- if (was_freezing)
- atomic_dec(&system_freezing_cnt);
- freezer->state &= ~CGROUP_FROZEN;
- unfreeze_cgroup(freezer);
- }
+ task->jobctl &= ~JOBCTL_TRAP_FREEZE;
+ wake_up_process(task);
}
+
+ unlock_task_sighand(task, &flags);
}
-/**
- * freezer_change_state - change the freezing state of a cgroup_freezer
- * @freezer: freezer of interest
- * @freeze: whether to freeze or thaw
- *
- * Freeze or thaw @freezer according to @freeze. The operations are
- * recursive - all descendants of @freezer will be affected.
+/*
+ * Freeze or unfreeze all tasks in the given cgroup.
*/
-static void freezer_change_state(struct freezer *freezer, bool freeze)
+static void cgroup_do_freeze(struct cgroup *cgrp, bool freeze)
{
- struct cgroup_subsys_state *pos;
+ struct css_task_iter it;
+ struct task_struct *task;
+
+ lockdep_assert_held(&cgroup_mutex);
+
+ spin_lock_irq(&css_set_lock);
+ if (freeze)
+ set_bit(CGRP_FREEZE, &cgrp->flags);
+ else
+ clear_bit(CGRP_FREEZE, &cgrp->flags);
+ spin_unlock_irq(&css_set_lock);
+
+ if (freeze)
+ TRACE_CGROUP_PATH(freeze, cgrp);
+ else
+ TRACE_CGROUP_PATH(unfreeze, cgrp);
+
+ css_task_iter_start(&cgrp->self, 0, &it);
+ while ((task = css_task_iter_next(&it))) {
+ /*
+ * Ignore kernel threads here. Freezing cgroups containing
+ * kthreads isn't supported.
+ */
+ if (task->flags & PF_KTHREAD)
+ continue;
+ cgroup_freeze_task(task, freeze);
+ }
+ css_task_iter_end(&it);
/*
- * Update all its descendants in pre-order traversal. Each
- * descendant will try to inherit its parent's FREEZING state as
- * CGROUP_FREEZING_PARENT.
+ * Cgroup state should be revisited here to cover empty leaf cgroups
+ * and cgroups which descendants are already in the desired state.
*/
- mutex_lock(&freezer_mutex);
- rcu_read_lock();
- css_for_each_descendant_pre(pos, &freezer->css) {
- struct freezer *pos_f = css_freezer(pos);
- struct freezer *parent = parent_freezer(pos_f);
+ spin_lock_irq(&css_set_lock);
+ if (cgrp->nr_descendants == cgrp->freezer.nr_frozen_descendants)
+ cgroup_update_frozen(cgrp);
+ spin_unlock_irq(&css_set_lock);
+}
- if (!css_tryget_online(pos))
- continue;
- rcu_read_unlock();
+/*
+ * Adjust the task state (freeze or unfreeze) and revisit the state of
+ * source and destination cgroups.
+ */
+void cgroup_freezer_migrate_task(struct task_struct *task,
+ struct cgroup *src, struct cgroup *dst)
+{
+ lockdep_assert_held(&css_set_lock);
- if (pos_f == freezer)
- freezer_apply_state(pos_f, freeze,
- CGROUP_FREEZING_SELF);
- else
- freezer_apply_state(pos_f,
- parent->state & CGROUP_FREEZING,
- CGROUP_FREEZING_PARENT);
+ /*
+ * Kernel threads are not supposed to be frozen at all.
+ */
+ if (task->flags & PF_KTHREAD)
+ return;
- rcu_read_lock();
- css_put(pos);
+ /*
+ * Adjust counters of freezing and frozen tasks.
+ * Note, that if the task is frozen, but the destination cgroup is not
+ * frozen, we bump both counters to keep them balanced.
+ */
+ if (task->frozen) {
+ cgroup_inc_frozen_cnt(dst);
+ cgroup_dec_frozen_cnt(src);
}
- rcu_read_unlock();
- mutex_unlock(&freezer_mutex);
+ cgroup_update_frozen(dst);
+ cgroup_update_frozen(src);
+
+ /*
+ * Force the task to the desired state.
+ */
+ cgroup_freeze_task(task, test_bit(CGRP_FREEZE, &dst->flags));
}
-static ssize_t freezer_write(struct kernfs_open_file *of,
- char *buf, size_t nbytes, loff_t off)
+void cgroup_freeze(struct cgroup *cgrp, bool freeze)
{
- bool freeze;
+ struct cgroup_subsys_state *css;
+ struct cgroup *dsct;
+ bool applied = false;
- buf = strstrip(buf);
+ lockdep_assert_held(&cgroup_mutex);
- if (strcmp(buf, freezer_state_strs(0)) == 0)
- freeze = false;
- else if (strcmp(buf, freezer_state_strs(CGROUP_FROZEN)) == 0)
- freeze = true;
- else
- return -EINVAL;
+ /*
+ * Nothing changed? Just exit.
+ */
+ if (cgrp->freezer.freeze == freeze)
+ return;
- freezer_change_state(css_freezer(of_css(of)), freeze);
- return nbytes;
+ cgrp->freezer.freeze = freeze;
+
+ /*
+ * Propagate changes downwards the cgroup tree.
+ */
+ css_for_each_descendant_pre(css, &cgrp->self) {
+ dsct = css->cgroup;
+
+ if (cgroup_is_dead(dsct))
+ continue;
+
+ if (freeze) {
+ dsct->freezer.e_freeze++;
+ /*
+ * Already frozen because of ancestor's settings?
+ */
+ if (dsct->freezer.e_freeze > 1)
+ continue;
+ } else {
+ dsct->freezer.e_freeze--;
+ /*
+ * Still frozen because of ancestor's settings?
+ */
+ if (dsct->freezer.e_freeze > 0)
+ continue;
+
+ WARN_ON_ONCE(dsct->freezer.e_freeze < 0);
+ }
+
+ /*
+ * Do change actual state: freeze or unfreeze.
+ */
+ cgroup_do_freeze(dsct, freeze);
+ applied = true;
+ }
+
+ /*
+ * Even if the actual state hasn't changed, let's notify a user.
+ * The state can be enforced by an ancestor cgroup: the cgroup
+ * can already be in the desired state or it can be locked in the
+ * opposite state, so that the transition will never happen.
+ * In both cases it's better to notify a user, that there is
+ * nothing to wait for.
+ */
+ if (!applied) {
+ TRACE_CGROUP_PATH(notify_frozen, cgrp,
+ test_bit(CGRP_FROZEN, &cgrp->flags));
+ cgroup_file_notify(&cgrp->events_file);
+ }
}
-
-static u64 freezer_self_freezing_read(struct cgroup_subsys_state *css,
- struct cftype *cft)
-{
- struct freezer *freezer = css_freezer(css);
-
- return (bool)(freezer->state & CGROUP_FREEZING_SELF);
-}
-
-static u64 freezer_parent_freezing_read(struct cgroup_subsys_state *css,
- struct cftype *cft)
-{
- struct freezer *freezer = css_freezer(css);
-
- return (bool)(freezer->state & CGROUP_FREEZING_PARENT);
-}
-
-static struct cftype files[] = {
- {
- .name = "state",
- .flags = CFTYPE_NOT_ON_ROOT,
- .seq_show = freezer_read,
- .write = freezer_write,
- },
- {
- .name = "self_freezing",
- .flags = CFTYPE_NOT_ON_ROOT,
- .read_u64 = freezer_self_freezing_read,
- },
- {
- .name = "parent_freezing",
- .flags = CFTYPE_NOT_ON_ROOT,
- .read_u64 = freezer_parent_freezing_read,
- },
- { } /* terminate */
-};
-
-struct cgroup_subsys freezer_cgrp_subsys = {
- .css_alloc = freezer_css_alloc,
- .css_online = freezer_css_online,
- .css_offline = freezer_css_offline,
- .css_free = freezer_css_free,
- .attach = freezer_attach,
- .fork = freezer_fork,
- .legacy_cftypes = files,
-};
diff --git a/kernel/cgroup/legacy_freezer.c b/kernel/cgroup/legacy_freezer.c
new file mode 100644
index 0000000..0823679
--- /dev/null
+++ b/kernel/cgroup/legacy_freezer.c
@@ -0,0 +1,481 @@
+/*
+ * cgroup_freezer.c - control group freezer subsystem
+ *
+ * Copyright IBM Corporation, 2007
+ *
+ * Author : Cedric Le Goater <clg@fr.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of version 2.1 of the GNU Lesser General Public License
+ * as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it would be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+ */
+
+#include <linux/export.h>
+#include <linux/slab.h>
+#include <linux/cgroup.h>
+#include <linux/fs.h>
+#include <linux/uaccess.h>
+#include <linux/freezer.h>
+#include <linux/seq_file.h>
+#include <linux/mutex.h>
+
+/*
+ * A cgroup is freezing if any FREEZING flags are set. FREEZING_SELF is
+ * set if "FROZEN" is written to freezer.state cgroupfs file, and cleared
+ * for "THAWED". FREEZING_PARENT is set if the parent freezer is FREEZING
+ * for whatever reason. IOW, a cgroup has FREEZING_PARENT set if one of
+ * its ancestors has FREEZING_SELF set.
+ */
+enum freezer_state_flags {
+ CGROUP_FREEZER_ONLINE = (1 << 0), /* freezer is fully online */
+ CGROUP_FREEZING_SELF = (1 << 1), /* this freezer is freezing */
+ CGROUP_FREEZING_PARENT = (1 << 2), /* the parent freezer is freezing */
+ CGROUP_FROZEN = (1 << 3), /* this and its descendants frozen */
+
+ /* mask for all FREEZING flags */
+ CGROUP_FREEZING = CGROUP_FREEZING_SELF | CGROUP_FREEZING_PARENT,
+};
+
+struct freezer {
+ struct cgroup_subsys_state css;
+ unsigned int state;
+};
+
+static DEFINE_MUTEX(freezer_mutex);
+
+static inline struct freezer *css_freezer(struct cgroup_subsys_state *css)
+{
+ return css ? container_of(css, struct freezer, css) : NULL;
+}
+
+static inline struct freezer *task_freezer(struct task_struct *task)
+{
+ return css_freezer(task_css(task, freezer_cgrp_id));
+}
+
+static struct freezer *parent_freezer(struct freezer *freezer)
+{
+ return css_freezer(freezer->css.parent);
+}
+
+bool cgroup_freezing(struct task_struct *task)
+{
+ bool ret;
+
+ rcu_read_lock();
+ ret = task_freezer(task)->state & CGROUP_FREEZING;
+ rcu_read_unlock();
+
+ return ret;
+}
+
+static const char *freezer_state_strs(unsigned int state)
+{
+ if (state & CGROUP_FROZEN)
+ return "FROZEN";
+ if (state & CGROUP_FREEZING)
+ return "FREEZING";
+ return "THAWED";
+};
+
+static struct cgroup_subsys_state *
+freezer_css_alloc(struct cgroup_subsys_state *parent_css)
+{
+ struct freezer *freezer;
+
+ freezer = kzalloc(sizeof(struct freezer), GFP_KERNEL);
+ if (!freezer)
+ return ERR_PTR(-ENOMEM);
+
+ return &freezer->css;
+}
+
+/**
+ * freezer_css_online - commit creation of a freezer css
+ * @css: css being created
+ *
+ * We're committing to creation of @css. Mark it online and inherit
+ * parent's freezing state while holding both parent's and our
+ * freezer->lock.
+ */
+static int freezer_css_online(struct cgroup_subsys_state *css)
+{
+ struct freezer *freezer = css_freezer(css);
+ struct freezer *parent = parent_freezer(freezer);
+
+ mutex_lock(&freezer_mutex);
+
+ freezer->state |= CGROUP_FREEZER_ONLINE;
+
+ if (parent && (parent->state & CGROUP_FREEZING)) {
+ freezer->state |= CGROUP_FREEZING_PARENT | CGROUP_FROZEN;
+ atomic_inc(&system_freezing_cnt);
+ }
+
+ mutex_unlock(&freezer_mutex);
+ return 0;
+}
+
+/**
+ * freezer_css_offline - initiate destruction of a freezer css
+ * @css: css being destroyed
+ *
+ * @css is going away. Mark it dead and decrement system_freezing_count if
+ * it was holding one.
+ */
+static void freezer_css_offline(struct cgroup_subsys_state *css)
+{
+ struct freezer *freezer = css_freezer(css);
+
+ mutex_lock(&freezer_mutex);
+
+ if (freezer->state & CGROUP_FREEZING)
+ atomic_dec(&system_freezing_cnt);
+
+ freezer->state = 0;
+
+ mutex_unlock(&freezer_mutex);
+}
+
+static void freezer_css_free(struct cgroup_subsys_state *css)
+{
+ kfree(css_freezer(css));
+}
+
+/*
+ * Tasks can be migrated into a different freezer anytime regardless of its
+ * current state. freezer_attach() is responsible for making new tasks
+ * conform to the current state.
+ *
+ * Freezer state changes and task migration are synchronized via
+ * @freezer->lock. freezer_attach() makes the new tasks conform to the
+ * current state and all following state changes can see the new tasks.
+ */
+static void freezer_attach(struct cgroup_taskset *tset)
+{
+ struct task_struct *task;
+ struct cgroup_subsys_state *new_css;
+
+ mutex_lock(&freezer_mutex);
+
+ /*
+ * Make the new tasks conform to the current state of @new_css.
+ * For simplicity, when migrating any task to a FROZEN cgroup, we
+ * revert it to FREEZING and let update_if_frozen() determine the
+ * correct state later.
+ *
+ * Tasks in @tset are on @new_css but may not conform to its
+ * current state before executing the following - !frozen tasks may
+ * be visible in a FROZEN cgroup and frozen tasks in a THAWED one.
+ */
+ cgroup_taskset_for_each(task, new_css, tset) {
+ struct freezer *freezer = css_freezer(new_css);
+
+ if (!(freezer->state & CGROUP_FREEZING)) {
+ __thaw_task(task);
+ } else {
+ freeze_task(task);
+ /* clear FROZEN and propagate upwards */
+ while (freezer && (freezer->state & CGROUP_FROZEN)) {
+ freezer->state &= ~CGROUP_FROZEN;
+ freezer = parent_freezer(freezer);
+ }
+ }
+ }
+
+ mutex_unlock(&freezer_mutex);
+}
+
+/**
+ * freezer_fork - cgroup post fork callback
+ * @task: a task which has just been forked
+ *
+ * @task has just been created and should conform to the current state of
+ * the cgroup_freezer it belongs to. This function may race against
+ * freezer_attach(). Losing to freezer_attach() means that we don't have
+ * to do anything as freezer_attach() will put @task into the appropriate
+ * state.
+ */
+static void freezer_fork(struct task_struct *task)
+{
+ struct freezer *freezer;
+
+ /*
+ * The root cgroup is non-freezable, so we can skip locking the
+ * freezer. This is safe regardless of race with task migration.
+ * If we didn't race or won, skipping is obviously the right thing
+ * to do. If we lost and root is the new cgroup, noop is still the
+ * right thing to do.
+ */
+ if (task_css_is_root(task, freezer_cgrp_id))
+ return;
+
+ mutex_lock(&freezer_mutex);
+ rcu_read_lock();
+
+ freezer = task_freezer(task);
+ if (freezer->state & CGROUP_FREEZING)
+ freeze_task(task);
+
+ rcu_read_unlock();
+ mutex_unlock(&freezer_mutex);
+}
+
+/**
+ * update_if_frozen - update whether a cgroup finished freezing
+ * @css: css of interest
+ *
+ * Once FREEZING is initiated, transition to FROZEN is lazily updated by
+ * calling this function. If the current state is FREEZING but not FROZEN,
+ * this function checks whether all tasks of this cgroup and the descendant
+ * cgroups finished freezing and, if so, sets FROZEN.
+ *
+ * The caller is responsible for grabbing RCU read lock and calling
+ * update_if_frozen() on all descendants prior to invoking this function.
+ *
+ * Task states and freezer state might disagree while tasks are being
+ * migrated into or out of @css, so we can't verify task states against
+ * @freezer state here. See freezer_attach() for details.
+ */
+static void update_if_frozen(struct cgroup_subsys_state *css)
+{
+ struct freezer *freezer = css_freezer(css);
+ struct cgroup_subsys_state *pos;
+ struct css_task_iter it;
+ struct task_struct *task;
+
+ lockdep_assert_held(&freezer_mutex);
+
+ if (!(freezer->state & CGROUP_FREEZING) ||
+ (freezer->state & CGROUP_FROZEN))
+ return;
+
+ /* are all (live) children frozen? */
+ rcu_read_lock();
+ css_for_each_child(pos, css) {
+ struct freezer *child = css_freezer(pos);
+
+ if ((child->state & CGROUP_FREEZER_ONLINE) &&
+ !(child->state & CGROUP_FROZEN)) {
+ rcu_read_unlock();
+ return;
+ }
+ }
+ rcu_read_unlock();
+
+ /* are all tasks frozen? */
+ css_task_iter_start(css, 0, &it);
+
+ while ((task = css_task_iter_next(&it))) {
+ if (freezing(task)) {
+ /*
+ * freezer_should_skip() indicates that the task
+ * should be skipped when determining freezing
+ * completion. Consider it frozen in addition to
+ * the usual frozen condition.
+ */
+ if (!frozen(task) && !freezer_should_skip(task))
+ goto out_iter_end;
+ }
+ }
+
+ freezer->state |= CGROUP_FROZEN;
+out_iter_end:
+ css_task_iter_end(&it);
+}
+
+static int freezer_read(struct seq_file *m, void *v)
+{
+ struct cgroup_subsys_state *css = seq_css(m), *pos;
+
+ mutex_lock(&freezer_mutex);
+ rcu_read_lock();
+
+ /* update states bottom-up */
+ css_for_each_descendant_post(pos, css) {
+ if (!css_tryget_online(pos))
+ continue;
+ rcu_read_unlock();
+
+ update_if_frozen(pos);
+
+ rcu_read_lock();
+ css_put(pos);
+ }
+
+ rcu_read_unlock();
+ mutex_unlock(&freezer_mutex);
+
+ seq_puts(m, freezer_state_strs(css_freezer(css)->state));
+ seq_putc(m, '\n');
+ return 0;
+}
+
+static void freeze_cgroup(struct freezer *freezer)
+{
+ struct css_task_iter it;
+ struct task_struct *task;
+
+ css_task_iter_start(&freezer->css, 0, &it);
+ while ((task = css_task_iter_next(&it)))
+ freeze_task(task);
+ css_task_iter_end(&it);
+}
+
+static void unfreeze_cgroup(struct freezer *freezer)
+{
+ struct css_task_iter it;
+ struct task_struct *task;
+
+ css_task_iter_start(&freezer->css, 0, &it);
+ while ((task = css_task_iter_next(&it)))
+ __thaw_task(task);
+ css_task_iter_end(&it);
+}
+
+/**
+ * freezer_apply_state - apply state change to a single cgroup_freezer
+ * @freezer: freezer to apply state change to
+ * @freeze: whether to freeze or unfreeze
+ * @state: CGROUP_FREEZING_* flag to set or clear
+ *
+ * Set or clear @state on @cgroup according to @freeze, and perform
+ * freezing or thawing as necessary.
+ */
+static void freezer_apply_state(struct freezer *freezer, bool freeze,
+ unsigned int state)
+{
+ /* also synchronizes against task migration, see freezer_attach() */
+ lockdep_assert_held(&freezer_mutex);
+
+ if (!(freezer->state & CGROUP_FREEZER_ONLINE))
+ return;
+
+ if (freeze) {
+ if (!(freezer->state & CGROUP_FREEZING))
+ atomic_inc(&system_freezing_cnt);
+ freezer->state |= state;
+ freeze_cgroup(freezer);
+ } else {
+ bool was_freezing = freezer->state & CGROUP_FREEZING;
+
+ freezer->state &= ~state;
+
+ if (!(freezer->state & CGROUP_FREEZING)) {
+ if (was_freezing)
+ atomic_dec(&system_freezing_cnt);
+ freezer->state &= ~CGROUP_FROZEN;
+ unfreeze_cgroup(freezer);
+ }
+ }
+}
+
+/**
+ * freezer_change_state - change the freezing state of a cgroup_freezer
+ * @freezer: freezer of interest
+ * @freeze: whether to freeze or thaw
+ *
+ * Freeze or thaw @freezer according to @freeze. The operations are
+ * recursive - all descendants of @freezer will be affected.
+ */
+static void freezer_change_state(struct freezer *freezer, bool freeze)
+{
+ struct cgroup_subsys_state *pos;
+
+ /*
+ * Update all its descendants in pre-order traversal. Each
+ * descendant will try to inherit its parent's FREEZING state as
+ * CGROUP_FREEZING_PARENT.
+ */
+ mutex_lock(&freezer_mutex);
+ rcu_read_lock();
+ css_for_each_descendant_pre(pos, &freezer->css) {
+ struct freezer *pos_f = css_freezer(pos);
+ struct freezer *parent = parent_freezer(pos_f);
+
+ if (!css_tryget_online(pos))
+ continue;
+ rcu_read_unlock();
+
+ if (pos_f == freezer)
+ freezer_apply_state(pos_f, freeze,
+ CGROUP_FREEZING_SELF);
+ else
+ freezer_apply_state(pos_f,
+ parent->state & CGROUP_FREEZING,
+ CGROUP_FREEZING_PARENT);
+
+ rcu_read_lock();
+ css_put(pos);
+ }
+ rcu_read_unlock();
+ mutex_unlock(&freezer_mutex);
+}
+
+static ssize_t freezer_write(struct kernfs_open_file *of,
+ char *buf, size_t nbytes, loff_t off)
+{
+ bool freeze;
+
+ buf = strstrip(buf);
+
+ if (strcmp(buf, freezer_state_strs(0)) == 0)
+ freeze = false;
+ else if (strcmp(buf, freezer_state_strs(CGROUP_FROZEN)) == 0)
+ freeze = true;
+ else
+ return -EINVAL;
+
+ freezer_change_state(css_freezer(of_css(of)), freeze);
+ return nbytes;
+}
+
+static u64 freezer_self_freezing_read(struct cgroup_subsys_state *css,
+ struct cftype *cft)
+{
+ struct freezer *freezer = css_freezer(css);
+
+ return (bool)(freezer->state & CGROUP_FREEZING_SELF);
+}
+
+static u64 freezer_parent_freezing_read(struct cgroup_subsys_state *css,
+ struct cftype *cft)
+{
+ struct freezer *freezer = css_freezer(css);
+
+ return (bool)(freezer->state & CGROUP_FREEZING_PARENT);
+}
+
+static struct cftype files[] = {
+ {
+ .name = "state",
+ .flags = CFTYPE_NOT_ON_ROOT,
+ .seq_show = freezer_read,
+ .write = freezer_write,
+ },
+ {
+ .name = "self_freezing",
+ .flags = CFTYPE_NOT_ON_ROOT,
+ .read_u64 = freezer_self_freezing_read,
+ },
+ {
+ .name = "parent_freezing",
+ .flags = CFTYPE_NOT_ON_ROOT,
+ .read_u64 = freezer_parent_freezing_read,
+ },
+ { } /* terminate */
+};
+
+struct cgroup_subsys freezer_cgrp_subsys = {
+ .css_alloc = freezer_css_alloc,
+ .css_online = freezer_css_online,
+ .css_offline = freezer_css_offline,
+ .css_free = freezer_css_free,
+ .attach = freezer_attach,
+ .fork = freezer_fork,
+ .legacy_cftypes = files,
+};
diff --git a/kernel/cgroup/pids.c b/kernel/cgroup/pids.c
index 9829c67..8e513a5 100644
--- a/kernel/cgroup/pids.c
+++ b/kernel/cgroup/pids.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
/*
* Process number limiting controller for cgroups.
*
@@ -25,10 +26,6 @@
* a superset of parent/child/pids.current.
*
* Copyright (C) 2015 Aleksa Sarai <cyphar@cyphar.com>
- *
- * This file is subject to the terms and conditions of version 2 of the GNU
- * General Public License. See the file COPYING in the main directory of the
- * Linux distribution for more details.
*/
#include <linux/kernel.h>
@@ -247,7 +244,7 @@
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 +339,7 @@
.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/rdma.c b/kernel/cgroup/rdma.c
index d3bbb75..ae042c3 100644
--- a/kernel/cgroup/rdma.c
+++ b/kernel/cgroup/rdma.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
/*
* RDMA resource limiting controller for cgroups.
*
@@ -5,10 +6,6 @@
* additional RDMA resources after a certain limit is reached.
*
* Copyright (C) 2016 Parav Pandit <pandit.parav@gmail.com>
- *
- * This file is subject to the terms and conditions of version 2 of the GNU
- * General Public License. See the file COPYING in the main directory of the
- * Linux distribution for more details.
*/
#include <linux/bitops.h>
@@ -313,10 +310,8 @@
* If IB stack wish a device to participate in rdma cgroup resource
* tracking, it must invoke this API to register with rdma cgroup before
* any user space application can start using the RDMA resources.
- * Returns 0 on success or EINVAL when table length given is beyond
- * supported size.
*/
-int rdmacg_register_device(struct rdmacg_device *device)
+void rdmacg_register_device(struct rdmacg_device *device)
{
INIT_LIST_HEAD(&device->dev_node);
INIT_LIST_HEAD(&device->rpools);
@@ -324,7 +319,6 @@
mutex_lock(&rdmacg_mutex);
list_add_tail(&device->dev_node, &rdmacg_devices);
mutex_unlock(&rdmacg_mutex);
- return 0;
}
EXPORT_SYMBOL(rdmacg_register_device);
diff --git a/kernel/cgroup/rstat.c b/kernel/cgroup/rstat.c
index d503d1a..ca19b4c 100644
--- a/kernel/cgroup/rstat.c
+++ b/kernel/cgroup/rstat.c
@@ -1,3 +1,4 @@
+// SPDX-License-Identifier: GPL-2.0-only
#include "cgroup-internal.h"
#include <linux/sched/cputime.h>
@@ -87,7 +88,6 @@
struct cgroup *root, int cpu)
{
struct cgroup_rstat_cpu *rstatc;
- struct cgroup *parent;
if (pos == root)
return NULL;
@@ -115,8 +115,8 @@
* 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 +140,12 @@
* updated stat.
*/
smp_mb();
+
+ return pos;
}
- return pos;
+ /* only happens for @root */
+ return NULL;
}
/* see cgroup_rstat_flush() */