Merge branch 'for-4.15' of git://git.kernel.org/pub/scm/linux/kernel/…

…git/tj/cgroup

Pull cgroup updates from Tejun Heo:
 "Cgroup2 cpu controller support is finally merged.

   - Basic cpu statistics support to allow monitoring by default without
     the CPU controller enabled.

   - cgroup2 cpu controller support.

   - /sys/kernel/cgroup files to help dealing with new / optional
     features"

* 'for-4.15' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
  cgroup: export list of cgroups v2 features using sysfs
  cgroup: export list of delegatable control files using sysfs
  cgroup: mark @CGRP __maybe_unused in cpu_stat_show()
  MAINTAINERS: relocate cpuset.c
  cgroup, sched: Move basic cpu stats from cgroup.stat to cpu.stat
  sched: Implement interface for cgroup unified hierarchy
  sched: Misc preps for cgroup unified hierarchy interface
  sched/cputime: Add dummy cputime_adjust() implementation for CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
  cgroup: statically initialize init_css_set->dfl_cgrp
  cgroup: Implement cgroup2 basic CPU usage accounting
  cpuacct: Introduce cgroup_account_cputime[_field]()
  sched/cputime: Expose cputime_adjust()

Documentation/cgroup-v2.txt

@@ -893,10 +893,6 @@ Controllers
 CPU
 ---
 
-.. note::
-
-	The interface for the cpu controller hasn't been merged yet
-
 The "cpu" controllers regulates distribution of CPU cycles.  This
 controller implements weight and absolute bandwidth limit models for
 normal scheduling policy and absolute bandwidth allocation model for
@@ -910,12 +906,16 @@ All time durations are in microseconds.
 
   cpu.stat
 	A read-only flat-keyed file which exists on non-root cgroups.
+	This file exists whether the controller is enabled or not.
 
-	It reports the following six stats:
+	It always reports the following three stats:
 
 	- usage_usec
 	- user_usec
 	- system_usec
+
+	and the following three when the controller is enabled:
+
 	- nr_periods
 	- nr_throttled
 	- throttled_usec
@@ -926,6 +926,18 @@ All time durations are in microseconds.
 
 	The weight in the range [1, 10000].
 
+  cpu.weight.nice
+	A read-write single value file which exists on non-root
+	cgroups.  The default is "0".
+
+	The nice value is in the range [-20, 19].
+
+	This interface file is an alternative interface for
+	"cpu.weight" and allows reading and setting weight using the
+	same values used by nice(2).  Because the range is smaller and
+	granularity is coarser for the nice values, the read value is
+	the closest approximation of the current weight.
+
   cpu.max
 	A read-write two value file which exists on non-root cgroups.
 	The default is "max 100000".
@@ -938,26 +950,6 @@ All time durations are in microseconds.
 	$PERIOD duration.  "max" for $MAX indicates no limit.  If only
 	one number is written, $MAX is updated.
 
-  cpu.rt.max
-	.. note::
-
-	   The semantics of this file is still under discussion and the
-	   interface hasn't been merged yet
-
-	A read-write two value file which exists on all cgroups.
-	The default is "0 100000".
-
-	The maximum realtime runtime allocation.  Over-committing
-	configurations are disallowed and process migrations are
-	rejected if not enough bandwidth is available.  It's in the
-	following format::
-
-	  $MAX $PERIOD
-
-	which indicates that the group may consume upto $MAX in each
-	$PERIOD duration.  If only one number is written, $MAX is
-	updated.
-
 
 Memory
 ------

MAINTAINERS

@@ -3592,7 +3592,7 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup.git
 S:	Maintained
 F:	Documentation/cgroup-v1/cpusets.txt
 F:	include/linux/cpuset.h
-F:	kernel/cpuset.c
+F:	kernel/cgroup/cpuset.c
 
 CONTROL GROUP - MEMORY RESOURCE CONTROLLER (MEMCG)
 M:	Johannes Weiner <[email protected]>

include/linux/cgroup-defs.h

@@ -17,6 +17,7 @@
 #include <linux/refcount.h>
 #include <linux/percpu-refcount.h>
 #include <linux/percpu-rwsem.h>
+#include <linux/u64_stats_sync.h>
 #include <linux/workqueue.h>
 #include <linux/bpf-cgroup.h>
 
@@ -255,6 +256,57 @@ struct css_set {
 	struct rcu_head rcu_head;
 };
 
+/*
+ * cgroup basic resource usage statistics.  Accounting is done per-cpu in
+ * cgroup_cpu_stat which is then lazily propagated up the hierarchy on
+ * reads.
+ *
+ * When a stat gets updated, the cgroup_cpu_stat and its ancestors are
+ * linked into the updated tree.  On the following read, propagation only
+ * considers and consumes the updated tree.  This makes reading O(the
+ * number of descendants which have been active since last read) instead of
+ * O(the total number of descendants).
+ *
+ * This is important because there can be a lot of (draining) cgroups which
+ * aren't active and stat may be read frequently.  The combination can
+ * become very expensive.  By propagating selectively, increasing reading
+ * frequency decreases the cost of each read.
+ */
+struct cgroup_cpu_stat {
+	/*
+	 * ->sync protects all the current counters.  These are the only
+	 * fields which get updated in the hot path.
+	 */
+	struct u64_stats_sync sync;
+	struct task_cputime cputime;
+
+	/*
+	 * Snapshots at the last reading.  These are used to calculate the
+	 * deltas to propagate to the global counters.
+	 */
+	struct task_cputime last_cputime;
+
+	/*
+	 * Child cgroups with stat updates on this cpu since the last read
+	 * are linked on the parent's ->updated_children through
+	 * ->updated_next.
+	 *
+	 * In addition to being more compact, singly-linked list pointing
+	 * to the cgroup makes it unnecessary for each per-cpu struct to
+	 * point back to the associated cgroup.
+	 *
+	 * Protected by per-cpu cgroup_cpu_stat_lock.
+	 */
+	struct cgroup *updated_children;	/* terminated by self cgroup */
+	struct cgroup *updated_next;		/* NULL iff not on the list */
+};
+
+struct cgroup_stat {
+	/* per-cpu statistics are collected into the folowing global counters */
+	struct task_cputime cputime;
+	struct prev_cputime prev_cputime;
+};
+
 struct cgroup {
 	/* self css with NULL ->ss, points back to this cgroup */
 	struct cgroup_subsys_state self;
@@ -354,6 +406,11 @@ struct cgroup {
 	 */
 	struct cgroup *dom_cgrp;
 
+	/* cgroup basic resource statistics */
+	struct cgroup_cpu_stat __percpu *cpu_stat;
+	struct cgroup_stat pending_stat;	/* pending from children */
+	struct cgroup_stat stat;
+
 	/*
 	 * list of pidlists, up to two for each namespace (one for procs, one
 	 * for tasks); created on demand.
@@ -513,6 +570,8 @@ struct cgroup_subsys {
 	void (*css_released)(struct cgroup_subsys_state *css);
 	void (*css_free)(struct cgroup_subsys_state *css);
 	void (*css_reset)(struct cgroup_subsys_state *css);
+	int (*css_extra_stat_show)(struct seq_file *seq,
+				   struct cgroup_subsys_state *css);
 
 	int (*can_attach)(struct cgroup_taskset *tset);
 	void (*cancel_attach)(struct cgroup_taskset *tset);

include/linux/cgroup.h

@@ -23,6 +23,7 @@
 #include <linux/nsproxy.h>
 #include <linux/user_namespace.h>
 #include <linux/refcount.h>
+#include <linux/kernel_stat.h>
 
 #include <linux/cgroup-defs.h>
 
@@ -689,6 +690,63 @@ static inline void cgroup_path_from_kernfs_id(const union kernfs_node_id *id,
 	char *buf, size_t buflen) {}
 #endif /* !CONFIG_CGROUPS */
 
+/*
+ * Basic resource stats.
+ */
+#ifdef CONFIG_CGROUPS
+
+#ifdef CONFIG_CGROUP_CPUACCT
+void cpuacct_charge(struct task_struct *tsk, u64 cputime);
+void cpuacct_account_field(struct task_struct *tsk, int index, u64 val);
+#else
+static inline void cpuacct_charge(struct task_struct *tsk, u64 cputime) {}
+static inline void cpuacct_account_field(struct task_struct *tsk, int index,
+					 u64 val) {}
+#endif
+
+void __cgroup_account_cputime(struct cgroup *cgrp, u64 delta_exec);
+void __cgroup_account_cputime_field(struct cgroup *cgrp,
+				    enum cpu_usage_stat index, u64 delta_exec);
+
+static inline void cgroup_account_cputime(struct task_struct *task,
+					  u64 delta_exec)
+{
+	struct cgroup *cgrp;
+
+	cpuacct_charge(task, delta_exec);
+
+	rcu_read_lock();
+	cgrp = task_dfl_cgroup(task);
+	if (cgroup_parent(cgrp))
+		__cgroup_account_cputime(cgrp, delta_exec);
+	rcu_read_unlock();
+}
+
+static inline void cgroup_account_cputime_field(struct task_struct *task,
+						enum cpu_usage_stat index,
+						u64 delta_exec)
+{
+	struct cgroup *cgrp;
+
+	cpuacct_account_field(task, index, delta_exec);
+
+	rcu_read_lock();
+	cgrp = task_dfl_cgroup(task);
+	if (cgroup_parent(cgrp))
+		__cgroup_account_cputime_field(cgrp, index, delta_exec);
+	rcu_read_unlock();
+}
+
+#else	/* CONFIG_CGROUPS */
+
+static inline void cgroup_account_cputime(struct task_struct *task,
+					  u64 delta_exec) {}
+static inline void cgroup_account_cputime_field(struct task_struct *task,
+						enum cpu_usage_stat index,
+						u64 delta_exec) {}
+
+#endif	/* CONFIG_CGROUPS */
+
 /*
  * sock->sk_cgrp_data handling.  For more info, see sock_cgroup_data
  * definition in cgroup-defs.h.

include/linux/sched/cputime.h

@@ -54,7 +54,8 @@ static inline void task_cputime_scaled(struct task_struct *t,
 
 extern void task_cputime_adjusted(struct task_struct *p, u64 *ut, u64 *st);
 extern void thread_group_cputime_adjusted(struct task_struct *p, u64 *ut, u64 *st);
-
+extern void cputime_adjust(struct task_cputime *curr, struct prev_cputime *prev,
+			   u64 *ut, u64 *st);
 
 /*
  * Thread group CPU time accounting.

kernel/cgroup/Makefile

@@ -1,5 +1,5 @@
 # SPDX-License-Identifier: GPL-2.0
-obj-y := cgroup.o namespace.o cgroup-v1.o
+obj-y := cgroup.o stat.o namespace.o cgroup-v1.o
 
 obj-$(CONFIG_CGROUP_FREEZER) += freezer.o
 obj-$(CONFIG_CGROUP_PIDS) += pids.o

kernel/cgroup/cgroup-internal.h

@@ -200,6 +200,15 @@ int cgroup_show_path(struct seq_file *sf, struct kernfs_node *kf_node,
 
 int cgroup_task_count(const struct cgroup *cgrp);
 
+/*
+ * stat.c
+ */
+void cgroup_stat_flush(struct cgroup *cgrp);
+int cgroup_stat_init(struct cgroup *cgrp);
+void cgroup_stat_exit(struct cgroup *cgrp);
+void cgroup_stat_show_cputime(struct seq_file *seq);
+void cgroup_stat_boot(void);
+
 /*
  * namespace.c
  */