aboutsummaryrefslogtreecommitdiff
path: root/kernel
diff options
context:
space:
mode:
Diffstat (limited to 'kernel')
-rw-r--r--kernel/Kconfig.locks6
-rw-r--r--kernel/audit.c2
-rw-r--r--kernel/audit_tree.c1
-rw-r--r--kernel/auditfilter.c3
-rw-r--r--kernel/cgroup.c1565
-rw-r--r--kernel/context_tracking.c41
-rw-r--r--kernel/cpu.c55
-rw-r--r--kernel/cpu/idle.c19
-rw-r--r--kernel/cpuset.c482
-rw-r--r--kernel/events/core.c751
-rw-r--r--kernel/events/hw_breakpoint.c193
-rw-r--r--kernel/events/internal.h4
-rw-r--r--kernel/exit.c15
-rw-r--r--kernel/fork.c66
-rw-r--r--kernel/freezer.c12
-rw-r--r--kernel/futex.c6
-rw-r--r--kernel/hrtimer.c35
-rw-r--r--kernel/irq/chip.c13
-rw-r--r--kernel/irq/generic-chip.c314
-rw-r--r--kernel/irq/irqdomain.c586
-rw-r--r--kernel/irq/manage.c17
-rw-r--r--kernel/irq/proc.c2
-rw-r--r--kernel/kmod.c6
-rw-r--r--kernel/kprobes.c33
-rw-r--r--kernel/lockdep.c17
-rw-r--r--kernel/module.c21
-rw-r--r--kernel/mutex.c384
-rw-r--r--kernel/pid.c14
-rw-r--r--kernel/posix-cpu-timers.c395
-rw-r--r--kernel/power/Kconfig21
-rw-r--r--kernel/power/main.c6
-rw-r--r--kernel/power/process.c26
-rw-r--r--kernel/power/qos.c14
-rw-r--r--kernel/power/snapshot.c9
-rw-r--r--kernel/power/suspend.c2
-rw-r--r--kernel/printk.c91
-rw-r--r--kernel/ptrace.c64
-rw-r--r--kernel/range.c19
-rw-r--r--kernel/rcupdate.c29
-rw-r--r--kernel/rcutiny.c21
-rw-r--r--kernel/rcutiny_plugin.h1009
-rw-r--r--kernel/rcutorture.c39
-rw-r--r--kernel/rcutree.c191
-rw-r--r--kernel/rcutree.h17
-rw-r--r--kernel/rcutree_plugin.h85
-rw-r--r--kernel/resource.c2
-rw-r--r--kernel/rtmutex.c13
-rw-r--r--kernel/sched/Makefile2
-rw-r--r--kernel/sched/auto_group.c3
-rw-r--r--kernel/sched/core.c660
-rw-r--r--kernel/sched/cputime.c11
-rw-r--r--kernel/sched/debug.c37
-rw-r--r--kernel/sched/fair.c175
-rw-r--r--kernel/sched/proc.c591
-rw-r--r--kernel/sched/rt.c132
-rw-r--r--kernel/sched/sched.h71
-rw-r--r--kernel/sched/stats.h47
-rw-r--r--kernel/sched/stop_task.c8
-rw-r--r--kernel/signal.c2
-rw-r--r--kernel/softirq.c23
-rw-r--r--kernel/sys.c47
-rw-r--r--kernel/sysctl.c12
-rw-r--r--kernel/time.c2
-rw-r--r--kernel/time/Kconfig5
-rw-r--r--kernel/time/Makefile2
-rw-r--r--kernel/time/alarmtimer.c47
-rw-r--r--kernel/time/clockevents.c271
-rw-r--r--kernel/time/clocksource.c266
-rw-r--r--kernel/time/ntp.c1
-rw-r--r--kernel/time/sched_clock.c212
-rw-r--r--kernel/time/tick-broadcast.c155
-rw-r--r--kernel/time/tick-common.c197
-rw-r--r--kernel/time/tick-internal.h17
-rw-r--r--kernel/time/tick-sched.c5
-rw-r--r--kernel/time/timekeeping.c73
-rw-r--r--kernel/time/timekeeping_debug.c72
-rw-r--r--kernel/time/timekeeping_internal.h14
-rw-r--r--kernel/timer.c10
-rw-r--r--kernel/trace/ftrace.c18
-rw-r--r--kernel/trace/ring_buffer.c3
-rw-r--r--kernel/trace/trace.c27
-rw-r--r--kernel/trace/trace.h2
-rw-r--r--kernel/trace/trace_events.c4
-rw-r--r--kernel/trace/trace_events_filter.c4
-rw-r--r--kernel/trace/trace_kprobe.c53
-rw-r--r--kernel/trace/trace_selftest.c2
-rw-r--r--kernel/wait.c88
-rw-r--r--kernel/workqueue.c45
-rw-r--r--kernel/workqueue_internal.h2
89 files changed, 5517 insertions, 4617 deletions
diff --git a/kernel/Kconfig.locks b/kernel/Kconfig.locks
index 44511d100ea..d2b32ac27a3 100644
--- a/kernel/Kconfig.locks
+++ b/kernel/Kconfig.locks
@@ -138,7 +138,7 @@ config INLINE_SPIN_UNLOCK_BH
config INLINE_SPIN_UNLOCK_IRQ
def_bool y
- depends on !PREEMPT || ARCH_INLINE_SPIN_UNLOCK_BH
+ depends on !PREEMPT || ARCH_INLINE_SPIN_UNLOCK_IRQ
config INLINE_SPIN_UNLOCK_IRQRESTORE
def_bool y
@@ -175,7 +175,7 @@ config INLINE_READ_UNLOCK_BH
config INLINE_READ_UNLOCK_IRQ
def_bool y
- depends on !PREEMPT || ARCH_INLINE_READ_UNLOCK_BH
+ depends on !PREEMPT || ARCH_INLINE_READ_UNLOCK_IRQ
config INLINE_READ_UNLOCK_IRQRESTORE
def_bool y
@@ -212,7 +212,7 @@ config INLINE_WRITE_UNLOCK_BH
config INLINE_WRITE_UNLOCK_IRQ
def_bool y
- depends on !PREEMPT || ARCH_INLINE_WRITE_UNLOCK_BH
+ depends on !PREEMPT || ARCH_INLINE_WRITE_UNLOCK_IRQ
config INLINE_WRITE_UNLOCK_IRQRESTORE
def_bool y
diff --git a/kernel/audit.c b/kernel/audit.c
index 21c7fa615bd..91e53d04b6a 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -1056,7 +1056,7 @@ static inline void audit_get_stamp(struct audit_context *ctx,
static void wait_for_auditd(unsigned long sleep_time)
{
DECLARE_WAITQUEUE(wait, current);
- set_current_state(TASK_INTERRUPTIBLE);
+ set_current_state(TASK_UNINTERRUPTIBLE);
add_wait_queue(&audit_backlog_wait, &wait);
if (audit_backlog_limit &&
diff --git a/kernel/audit_tree.c b/kernel/audit_tree.c
index a291aa23fb3..43c307dc945 100644
--- a/kernel/audit_tree.c
+++ b/kernel/audit_tree.c
@@ -658,6 +658,7 @@ int audit_add_tree_rule(struct audit_krule *rule)
struct vfsmount *mnt;
int err;
+ rule->tree = NULL;
list_for_each_entry(tree, &tree_list, list) {
if (!strcmp(seed->pathname, tree->pathname)) {
put_tree(seed);
diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c
index 83a2970295d..6bd4a90d199 100644
--- a/kernel/auditfilter.c
+++ b/kernel/auditfilter.c
@@ -1021,9 +1021,6 @@ static void audit_log_rule_change(char *action, struct audit_krule *rule, int re
* @seq: netlink audit message sequence (serial) number
* @data: payload data
* @datasz: size of payload data
- * @loginuid: loginuid of sender
- * @sessionid: sessionid for netlink audit message
- * @sid: SE Linux Security ID of sender
*/
int audit_receive_filter(int type, int pid, int seq, void *data, size_t datasz)
{
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 2a9926275f8..e5583d10a32 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -63,9 +63,6 @@
#include <linux/atomic.h>
-/* css deactivation bias, makes css->refcnt negative to deny new trygets */
-#define CSS_DEACT_BIAS INT_MIN
-
/*
* cgroup_mutex is the master lock. Any modification to cgroup or its
* hierarchy must be performed while holding it.
@@ -99,16 +96,19 @@ static DEFINE_MUTEX(cgroup_root_mutex);
*/
#define SUBSYS(_x) [_x ## _subsys_id] = &_x ## _subsys,
#define IS_SUBSYS_ENABLED(option) IS_BUILTIN(option)
-static struct cgroup_subsys *subsys[CGROUP_SUBSYS_COUNT] = {
+static struct cgroup_subsys *cgroup_subsys[CGROUP_SUBSYS_COUNT] = {
#include <linux/cgroup_subsys.h>
};
/*
- * The "rootnode" hierarchy is the "dummy hierarchy", reserved for the
- * subsystems that are otherwise unattached - it never has more than a
- * single cgroup, and all tasks are part of that cgroup.
+ * The dummy hierarchy, reserved for the subsystems that are otherwise
+ * unattached - it never has more than a single cgroup, and all tasks are
+ * part of that cgroup.
*/
-static struct cgroupfs_root rootnode;
+static struct cgroupfs_root cgroup_dummy_root;
+
+/* dummy_top is a shorthand for the dummy hierarchy's top cgroup */
+static struct cgroup * const cgroup_dummy_top = &cgroup_dummy_root.top_cgroup;
/*
* cgroupfs file entry, pointed to from leaf dentry->d_fsdata.
@@ -186,18 +186,28 @@ struct cgroup_event {
/* The list of hierarchy roots */
-static LIST_HEAD(roots);
-static int root_count;
+static LIST_HEAD(cgroup_roots);
+static int cgroup_root_count;
-static DEFINE_IDA(hierarchy_ida);
-static int next_hierarchy_id;
-static DEFINE_SPINLOCK(hierarchy_id_lock);
-
-/* dummytop is a shorthand for the dummy hierarchy's top cgroup */
-#define dummytop (&rootnode.top_cgroup)
+/*
+ * Hierarchy ID allocation and mapping. It follows the same exclusion
+ * rules as other root ops - both cgroup_mutex and cgroup_root_mutex for
+ * writes, either for reads.
+ */
+static DEFINE_IDR(cgroup_hierarchy_idr);
static struct cgroup_name root_cgroup_name = { .name = "/" };
+/*
+ * Assign a monotonically increasing serial number to cgroups. It
+ * guarantees cgroups with bigger numbers are newer than those with smaller
+ * numbers. Also, as cgroups are always appended to the parent's
+ * ->children list, it guarantees that sibling cgroups are always sorted in
+ * the ascending serial number order on the list. Protected by
+ * cgroup_mutex.
+ */
+static u64 cgroup_serial_nr_next = 1;
+
/* This flag indicates whether tasks in the fork and exit paths should
* check for fork/exit handlers to call. This avoids us having to do
* extra work in the fork/exit path if none of the subsystems need to
@@ -205,27 +215,15 @@ static struct cgroup_name root_cgroup_name = { .name = "/" };
*/
static int need_forkexit_callback __read_mostly;
+static void cgroup_offline_fn(struct work_struct *work);
static int cgroup_destroy_locked(struct cgroup *cgrp);
static int cgroup_addrm_files(struct cgroup *cgrp, struct cgroup_subsys *subsys,
struct cftype cfts[], bool is_add);
-static int css_unbias_refcnt(int refcnt)
-{
- return refcnt >= 0 ? refcnt : refcnt - CSS_DEACT_BIAS;
-}
-
-/* the current nr of refs, always >= 0 whether @css is deactivated or not */
-static int css_refcnt(struct cgroup_subsys_state *css)
-{
- int v = atomic_read(&css->refcnt);
-
- return css_unbias_refcnt(v);
-}
-
/* convenient tests for these bits */
-inline int cgroup_is_removed(const struct cgroup *cgrp)
+static inline bool cgroup_is_dead(const struct cgroup *cgrp)
{
- return test_bit(CGRP_REMOVED, &cgrp->flags);
+ return test_bit(CGRP_DEAD, &cgrp->flags);
}
/**
@@ -261,16 +259,38 @@ static int notify_on_release(const struct cgroup *cgrp)
return test_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);
}
-/*
- * for_each_subsys() allows you to iterate on each subsystem attached to
- * an active hierarchy
+/**
+ * for_each_subsys - iterate all loaded cgroup subsystems
+ * @ss: the iteration cursor
+ * @i: the index of @ss, CGROUP_SUBSYS_COUNT after reaching the end
+ *
+ * Should be called under cgroup_mutex.
*/
-#define for_each_subsys(_root, _ss) \
-list_for_each_entry(_ss, &_root->subsys_list, sibling)
+#define for_each_subsys(ss, i) \
+ for ((i) = 0; (i) < CGROUP_SUBSYS_COUNT; (i)++) \
+ if (({ lockdep_assert_held(&cgroup_mutex); \
+ !((ss) = cgroup_subsys[i]); })) { } \
+ else
-/* for_each_active_root() allows you to iterate across the active hierarchies */
-#define for_each_active_root(_root) \
-list_for_each_entry(_root, &roots, root_list)
+/**
+ * for_each_builtin_subsys - iterate all built-in cgroup subsystems
+ * @ss: the iteration cursor
+ * @i: the index of @ss, CGROUP_BUILTIN_SUBSYS_COUNT after reaching the end
+ *
+ * Bulit-in subsystems are always present and iteration itself doesn't
+ * require any synchronization.
+ */
+#define for_each_builtin_subsys(ss, i) \
+ for ((i) = 0; (i) < CGROUP_BUILTIN_SUBSYS_COUNT && \
+ (((ss) = cgroup_subsys[i]) || true); (i)++)
+
+/* iterate each subsystem attached to a hierarchy */
+#define for_each_root_subsys(root, ss) \
+ list_for_each_entry((ss), &(root)->subsys_list, sibling)
+
+/* iterate across the active hierarchies */
+#define for_each_active_root(root) \
+ list_for_each_entry((root), &cgroup_roots, root_list)
static inline struct cgroup *__d_cgrp(struct dentry *dentry)
{
@@ -297,7 +317,7 @@ static inline struct cftype *__d_cft(struct dentry *dentry)
static bool cgroup_lock_live_group(struct cgroup *cgrp)
{
mutex_lock(&cgroup_mutex);
- if (cgroup_is_removed(cgrp)) {
+ if (cgroup_is_dead(cgrp)) {
mutex_unlock(&cgroup_mutex);
return false;
}
@@ -312,20 +332,24 @@ static void cgroup_release_agent(struct work_struct *work);
static DECLARE_WORK(release_agent_work, cgroup_release_agent);
static void check_for_release(struct cgroup *cgrp);
-/* Link structure for associating css_set objects with cgroups */
-struct cg_cgroup_link {
- /*
- * List running through cg_cgroup_links associated with a
- * cgroup, anchored on cgroup->css_sets
- */
- struct list_head cgrp_link_list;
- struct cgroup *cgrp;
- /*
- * List running through cg_cgroup_links pointing at a
- * single css_set object, anchored on css_set->cg_links
- */
- struct list_head cg_link_list;
- struct css_set *cg;
+/*
+ * 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.
+ * This M:N relationship is represented by the following link structure
+ * which exists for each association and allows traversing the associations
+ * from both sides.
+ */
+struct cgrp_cset_link {
+ /* the cgroup and css_set this link associates */
+ struct cgroup *cgrp;
+ struct css_set *cset;
+
+ /* list of cgrp_cset_links anchored at cgrp->cset_links */
+ struct list_head cset_link;
+
+ /* list of cgrp_cset_links anchored at css_set->cgrp_links */
+ struct list_head cgrp_link;
};
/* The default css_set - used by init and its children prior to any
@@ -336,7 +360,7 @@ struct cg_cgroup_link {
*/
static struct css_set init_css_set;
-static struct cg_cgroup_link init_css_set_link;
+static struct cgrp_cset_link init_cgrp_cset_link;
static int cgroup_init_idr(struct cgroup_subsys *ss,
struct cgroup_subsys_state *css);
@@ -357,10 +381,11 @@ static DEFINE_HASHTABLE(css_set_table, CSS_SET_HASH_BITS);
static unsigned long css_set_hash(struct cgroup_subsys_state *css[])
{
- int i;
unsigned long key = 0UL;
+ struct cgroup_subsys *ss;
+ int i;
- for (i = 0; i < CGROUP_SUBSYS_COUNT; i++)
+ for_each_subsys(ss, i)
key += (unsigned long)css[i];
key = (key >> 16) ^ key;
@@ -373,90 +398,83 @@ static unsigned long css_set_hash(struct cgroup_subsys_state *css[])
* compiled into their kernel but not actually in use */
static int use_task_css_set_links __read_mostly;
-static void __put_css_set(struct css_set *cg, int taskexit)
+static void __put_css_set(struct css_set *cset, int taskexit)
{
- struct cg_cgroup_link *link;
- struct cg_cgroup_link *saved_link;
+ struct cgrp_cset_link *link, *tmp_link;
+
/*
* Ensure that the refcount doesn't hit zero while any readers
* can see it. Similar to atomic_dec_and_lock(), but for an
* rwlock
*/
- if (atomic_add_unless(&cg->refcount, -1, 1))
+ if (atomic_add_unless(&cset->refcount, -1, 1))
return;
write_lock(&css_set_lock);
- if (!atomic_dec_and_test(&cg->refcount)) {
+ if (!atomic_dec_and_test(&cset->refcount)) {
write_unlock(&css_set_lock);
return;
}
/* This css_set is dead. unlink it and release cgroup refcounts */
- hash_del(&cg->hlist);
+ hash_del(&cset->hlist);
css_set_count--;
- list_for_each_entry_safe(link, saved_link, &cg->cg_links,
- cg_link_list) {
+ list_for_each_entry_safe(link, tmp_link, &cset->cgrp_links, cgrp_link) {
struct cgroup *cgrp = link->cgrp;
- list_del(&link->cg_link_list);
- list_del(&link->cgrp_link_list);
- /*
- * We may not be holding cgroup_mutex, and if cgrp->count is
- * dropped to 0 the cgroup can be destroyed at any time, hence
- * rcu_read_lock is used to keep it alive.
- */
- rcu_read_lock();
- if (atomic_dec_and_test(&cgrp->count) &&
- notify_on_release(cgrp)) {
+ list_del(&link->cset_link);
+ list_del(&link->cgrp_link);
+
+ /* @cgrp can't go away while we're holding css_set_lock */
+ if (list_empty(&cgrp->cset_links) && notify_on_release(cgrp)) {
if (taskexit)
set_bit(CGRP_RELEASABLE, &cgrp->flags);
check_for_release(cgrp);
}
- rcu_read_unlock();
kfree(link);
}
write_unlock(&css_set_lock);
- kfree_rcu(cg, rcu_head);