sched: group scheduling, sysfs tunables

Add tunables in sysfs to modify a user's cpu share.

A directory is created in sysfs for each new user in the system.

	/sys/kernel/uids/<uid>/cpu_share

Reading this file returns the cpu shares granted for the user.
Writing into this file modifies the cpu share for the user. Only an
administrator is allowed to modify a user's cpu share.

Ex:
	# cd /sys/kernel/uids/
	# cat 512/cpu_share
	1024
	# echo 2048 > 512/cpu_share
	# cat 512/cpu_share
	2048
	#

Signed-off-by: Srivatsa Vaddagiri <[email protected]>
Signed-off-by: Dhaval Giani <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>

Documentation/sched-design-CFS.txt

@@ -117,3 +117,70 @@ Some implementation details:
    iterators of the scheduling modules are used. The balancing code got
    quite a bit simpler as a result.
 
+
+Group scheduler extension to CFS
+================================
+
+Normally the scheduler operates on individual tasks and strives to provide
+fair CPU time to each task. Sometimes, it may be desirable to group tasks
+and provide fair CPU time to each such task group. For example, it may
+be desirable to first provide fair CPU time to each user on the system
+and then to each task belonging to a user.
+
+CONFIG_FAIR_GROUP_SCHED strives to achieve exactly that. It lets
+SCHED_NORMAL/BATCH tasks be be grouped and divides CPU time fairly among such
+groups. At present, there are two (mutually exclusive) mechanisms to group
+tasks for CPU bandwidth control purpose:
+
+	- Based on user id (CONFIG_FAIR_USER_SCHED)
+		In this option, tasks are grouped according to their user id.
+	- Based on "cgroup" pseudo filesystem (CONFIG_FAIR_CGROUP_SCHED)
+		This options lets the administrator create arbitrary groups
+		of tasks, using the "cgroup" pseudo filesystem. See
+		Documentation/cgroups.txt for more information about this
+		filesystem.
+
+Only one of these options to group tasks can be chosen and not both.
+
+Group scheduler tunables:
+
+When CONFIG_FAIR_USER_SCHED is defined, a directory is created in sysfs for
+each new user and a "cpu_share" file is added in that directory.
+
+	# cd /sys/kernel/uids
+	# cat 512/cpu_share		# Display user 512's CPU share
+	1024
+	# echo 2048 > 512/cpu_share	# Modify user 512's CPU share
+	# cat 512/cpu_share		# Display user 512's CPU share
+	2048
+	#
+
+CPU bandwidth between two users are divided in the ratio of their CPU shares.
+For ex: if you would like user "root" to get twice the bandwidth of user
+"guest", then set the cpu_share for both the users such that "root"'s
+cpu_share is twice "guest"'s cpu_share
+
+
+When CONFIG_FAIR_CGROUP_SCHED is defined, a "cpu.shares" file is created
+for each group created using the pseudo filesystem. See example steps
+below to create task groups and modify their CPU share using the "cgroups"
+pseudo filesystem
+
+	# mkdir /dev/cpuctl
+	# mount -t cgroup -ocpu none /dev/cpuctl
+	# cd /dev/cpuctl
+
+	# mkdir multimedia	# create "multimedia" group of tasks
+	# mkdir browser		# create "browser" group of tasks
+
+	# #Configure the multimedia group to receive twice the CPU bandwidth
+	# #that of browser group
+
+	# echo 2048 > multimedia/cpu.shares
+	# echo 1024 > browser/cpu.shares
+
+	# firefox &	# Launch firefox and move it to "browser" group
+	# echo <firefox_pid> > browser/tasks
+
+	# #Launch gmplayer (or your favourite movie player)
+	# echo <movie_player_pid> > multimedia/tasks

include/linux/sched.h

@@ -87,6 +87,7 @@ struct sched_param {
 #include <linux/timer.h>
 #include <linux/hrtimer.h>
 #include <linux/task_io_accounting.h>
+#include <linux/kobject.h>
 
 #include <asm/processor.h>
 
@@ -599,9 +600,18 @@ struct user_struct {
 
 #ifdef CONFIG_FAIR_USER_SCHED
 	struct task_group *tg;
+	struct kset kset;
+	struct subsys_attribute user_attr;
+	struct work_struct work;
 #endif
 };
 
+#ifdef CONFIG_FAIR_USER_SCHED
+extern int uids_kobject_init(void);
+#else
+static inline int uids_kobject_init(void) { return 0; }
+#endif
+
 extern struct user_struct *find_user(uid_t);
 
 extern struct user_struct root_user;
@@ -1848,6 +1858,7 @@ extern struct task_group *sched_create_group(void);
 extern void sched_destroy_group(struct task_group *tg);
 extern void sched_move_task(struct task_struct *tsk);
 extern int sched_group_set_shares(struct task_group *tg, unsigned long shares);
+extern unsigned long sched_group_shares(struct task_group *tg);
 
 #endif
 

kernel/ksysfs.c

@@ -14,6 +14,7 @@
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/kexec.h>
+#include <linux/sched.h>
 
 #define KERNEL_ATTR_RO(_name) \
 static struct subsys_attribute _name##_attr = __ATTR_RO(_name)
@@ -116,6 +117,13 @@ static int __init ksysfs_init(void)
 					      &notes_attr);
 	}
 
+	/*
+	 * Create "/sys/kernel/uids" directory and corresponding root user's
+	 * directory under it.
+	 */
+	if (!error)
+		error = uids_kobject_init();
+
 	return error;
 }
 

kernel/sched.c

@@ -162,6 +162,8 @@ struct task_group {
 	/* runqueue "owned" by this group on each cpu */
 	struct cfs_rq **cfs_rq;
 	unsigned long shares;
+	/* spinlock to serialize modification to shares */
+	spinlock_t lock;
 };
 
 /* Default task group's sched entity on each cpu */
@@ -6533,6 +6535,7 @@ void __init sched_init(void)
 			se->parent = NULL;
 		}
 		init_task_group.shares = init_task_group_load;
+		spin_lock_init(&init_task_group.lock);
 #endif
 
 		for (j = 0; j < CPU_LOAD_IDX_MAX; j++)
@@ -6777,6 +6780,7 @@ struct task_group *sched_create_group(void)
 	}
 
 	tg->shares = NICE_0_LOAD;
+	spin_lock_init(&tg->lock);
 
 	return tg;
 
@@ -6897,16 +6901,24 @@ int sched_group_set_shares(struct task_group *tg, unsigned long shares)
 {
 	int i;
 
+	spin_lock(&tg->lock);
 	if (tg->shares == shares)
-		return 0;
+		goto done;
 
 	/* return -EINVAL if the new value is not sane */
 
 	tg->shares = shares;
 	for_each_possible_cpu(i)
 		set_se_shares(tg->se[i], shares);
 
+done:
+	spin_unlock(&tg->lock);
 	return 0;
 }
 
+unsigned long sched_group_shares(struct task_group *tg)
+{
+	return tg->shares;
+}
+
 #endif	/* CONFIG_FAIR_GROUP_SCHED */

kernel/sched_debug.c

@@ -231,45 +231,6 @@ static void sysrq_sched_debug_show(void)
 	sched_debug_show(NULL, NULL);
 }
 
-#ifdef CONFIG_FAIR_USER_SCHED
-
-static DEFINE_MUTEX(root_user_share_mutex);
-
-static int
-root_user_share_read_proc(char *page, char **start, off_t off, int count,
-				 int *eof, void *data)
-{
-	return sprintf(page, "%d\n", init_task_group_load);
-}
-
-static int
-root_user_share_write_proc(struct file *file, const char __user *buffer,
-				 unsigned long count, void *data)
-{
-	unsigned long shares;
-	char kbuf[sizeof(unsigned long)+1];
-	int rc = 0;
-
-	if (copy_from_user(kbuf, buffer, sizeof(kbuf)))
-		return -EFAULT;
-
-	shares = simple_strtoul(kbuf, NULL, 0);
-
-	if (!shares)
-		shares = NICE_0_LOAD;
-
-	mutex_lock(&root_user_share_mutex);
-
-	init_task_group_load = shares;
-	rc = sched_group_set_shares(&init_task_group, shares);
-
-	mutex_unlock(&root_user_share_mutex);
-
-	return (rc < 0 ? rc : count);
-}
-
-#endif	/* CONFIG_FAIR_USER_SCHED */
-
 static int sched_debug_open(struct inode *inode, struct file *filp)
 {
 	return single_open(filp, sched_debug_show, NULL);
@@ -292,15 +253,6 @@ static int __init init_sched_debug_procfs(void)
 
 	pe->proc_fops = &sched_debug_fops;
 
-#ifdef CONFIG_FAIR_USER_SCHED
-	pe = create_proc_entry("root_user_cpu_share", 0644, NULL);
-	if (!pe)
-		return -ENOMEM;
-
-	pe->read_proc = root_user_share_read_proc;
-	pe->write_proc = root_user_share_write_proc;
-#endif
-
 	return 0;
 }