aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorSrivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>2007-10-15 17:00:09 +0200
committerIngo Molnar <mingo@elte.hu>2007-10-15 17:00:09 +0200
commit9b5b77512dce239fa168183fa71896712232e95a (patch)
tree2a95cf9151b4b6bd745236858f5b73b639dbc0d9
parent75c28ace9f2b2f403674e045939424a77c95b47c (diff)
downloadmrst-s0i3-test-9b5b77512dce239fa168183fa71896712232e95a.tar.gz
mrst-s0i3-test-9b5b77512dce239fa168183fa71896712232e95a.tar.xz
mrst-s0i3-test-9b5b77512dce239fa168183fa71896712232e95a.zip
sched: clean up code under CONFIG_FAIR_GROUP_SCHED
With the view of supporting user-id based fair scheduling (and not just container-based fair scheduling), this patch renames several functions and makes them independent of whether they are being used for container or user-id based fair scheduling. Also fix a problem reported by KAMEZAWA Hiroyuki (wrt allocating less-sized array for tg->cfs_rq[] and tf->se[]). Signed-off-by: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com> Signed-off-by: Dhaval Giani <dhaval@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
-rw-r--r--include/linux/sched.h12
-rw-r--r--init/Kconfig11
-rw-r--r--kernel/sched.c172
-rw-r--r--kernel/sched_fair.c5
4 files changed, 83 insertions, 117 deletions
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 66169005f00..03c13b663e4 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -136,6 +136,7 @@ extern unsigned long weighted_cpuload(const int cpu);
struct seq_file;
struct cfs_rq;
+struct task_grp;
#ifdef CONFIG_SCHED_DEBUG
extern void proc_sched_show_task(struct task_struct *p, struct seq_file *m);
extern void proc_sched_set_task(struct task_struct *p);
@@ -1834,6 +1835,17 @@ extern int sched_mc_power_savings, sched_smt_power_savings;
extern void normalize_rt_tasks(void);
+#ifdef CONFIG_FAIR_GROUP_SCHED
+
+extern struct task_grp init_task_grp;
+
+extern struct task_grp *sched_create_group(void);
+extern void sched_destroy_group(struct task_grp *tg);
+extern void sched_move_task(struct task_struct *tsk);
+extern int sched_group_set_shares(struct task_grp *tg, unsigned long shares);
+
+#endif
+
#ifdef CONFIG_TASK_XACCT
static inline void add_rchar(struct task_struct *tsk, ssize_t amt)
{
diff --git a/init/Kconfig b/init/Kconfig
index 11c6762a652..ef90a154dd9 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -282,13 +282,12 @@ config CPUSETS
Say N if unsure.
config FAIR_GROUP_SCHED
- bool "Fair group scheduler"
- depends on EXPERIMENTAL && CONTAINERS
+ bool "Fair group cpu scheduler"
+ default n
+ depends on EXPERIMENTAL
help
- This option enables you to group tasks and control CPU resource
- allocation to such groups.
-
- Say N if unsure.
+ This feature lets cpu scheduler recognize task groups and control cpu
+ bandwidth allocation to such task groups.
config SYSFS_DEPRECATED
bool "Create deprecated sysfs files"
diff --git a/kernel/sched.c b/kernel/sched.c
index ee7ac71b12f..e10c403b121 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -173,13 +173,10 @@ struct rt_prio_array {
#ifdef CONFIG_FAIR_GROUP_SCHED
-#include <linux/container.h>
-
struct cfs_rq;
/* task group related information */
struct task_grp {
- struct container_subsys_state css;
/* schedulable entities of this group on each cpu */
struct sched_entity **se;
/* runqueue "owned" by this group on each cpu */
@@ -192,22 +189,28 @@ static DEFINE_PER_CPU(struct sched_entity, init_sched_entity);
/* Default task group's cfs_rq on each cpu */
static DEFINE_PER_CPU(struct cfs_rq, init_cfs_rq) ____cacheline_aligned_in_smp;
-static struct sched_entity *init_sched_entity_p[CONFIG_NR_CPUS];
-static struct cfs_rq *init_cfs_rq_p[CONFIG_NR_CPUS];
+static struct sched_entity *init_sched_entity_p[NR_CPUS];
+static struct cfs_rq *init_cfs_rq_p[NR_CPUS];
/* Default task group.
* Every task in system belong to this group at bootup.
*/
-static struct task_grp init_task_grp = {
- .se = init_sched_entity_p,
- .cfs_rq = init_cfs_rq_p,
- };
+struct task_grp init_task_grp = {
+ .se = init_sched_entity_p,
+ .cfs_rq = init_cfs_rq_p,
+ };
+
+#define INIT_TASK_GRP_LOAD NICE_0_LOAD
+static int init_task_grp_load = INIT_TASK_GRP_LOAD;
/* return group to which a task belongs */
static inline struct task_grp *task_grp(struct task_struct *p)
{
- return container_of(task_subsys_state(p, cpu_subsys_id),
- struct task_grp, css);
+ struct task_grp *tg;
+
+ tg = &init_task_grp;
+
+ return tg;
}
/* Change a task's cfs_rq and parent entity if it moves across CPUs/groups */
@@ -250,6 +253,7 @@ struct cfs_rq {
*/
struct list_head leaf_cfs_rq_list; /* Better name : task_cfs_rq_list? */
struct task_grp *tg; /* group that "owns" this runqueue */
+ struct rcu_head rcu;
#endif
};
@@ -6513,11 +6517,12 @@ void __init sched_init(void)
init_sched_entity_p[i] = se;
se->cfs_rq = &rq->cfs;
se->my_q = cfs_rq;
- se->load.weight = NICE_0_LOAD;
- se->load.inv_weight = div64_64(1ULL<<32, NICE_0_LOAD);
+ se->load.weight = init_task_grp_load;
+ se->load.inv_weight =
+ div64_64(1ULL<<32, init_task_grp_load);
se->parent = NULL;
}
- init_task_grp.shares = NICE_0_LOAD;
+ init_task_grp.shares = init_task_grp_load;
#endif
for (j = 0; j < CPU_LOAD_IDX_MAX; j++)
@@ -6707,45 +6712,28 @@ void set_curr_task(int cpu, struct task_struct *p)
#ifdef CONFIG_FAIR_GROUP_SCHED
-/* return corresponding task_grp object of a container */
-static inline struct task_grp *container_tg(struct container *cont)
-{
- return container_of(container_subsys_state(cont, cpu_subsys_id),
- struct task_grp, css);
-}
-
/* allocate runqueue etc for a new task group */
-static struct container_subsys_state *
-sched_create_group(struct container_subsys *ss, struct container *cont)
+struct task_grp *sched_create_group(void)
{
struct task_grp *tg;
struct cfs_rq *cfs_rq;
struct sched_entity *se;
+ struct rq *rq;
int i;
- if (!cont->parent) {
- /* This is early initialization for the top container */
- init_task_grp.css.container = cont;
- return &init_task_grp.css;
- }
-
- /* we support only 1-level deep hierarchical scheduler atm */
- if (cont->parent->parent)
- return ERR_PTR(-EINVAL);
-
tg = kzalloc(sizeof(*tg), GFP_KERNEL);
if (!tg)
return ERR_PTR(-ENOMEM);
- tg->cfs_rq = kzalloc(sizeof(cfs_rq) * num_possible_cpus(), GFP_KERNEL);
+ tg->cfs_rq = kzalloc(sizeof(cfs_rq) * NR_CPUS, GFP_KERNEL);
if (!tg->cfs_rq)
goto err;
- tg->se = kzalloc(sizeof(se) * num_possible_cpus(), GFP_KERNEL);
+ tg->se = kzalloc(sizeof(se) * NR_CPUS, GFP_KERNEL);
if (!tg->se)
goto err;
for_each_possible_cpu(i) {
- struct rq *rq = cpu_rq(i);
+ rq = cpu_rq(i);
cfs_rq = kmalloc_node(sizeof(struct cfs_rq), GFP_KERNEL,
cpu_to_node(i));
@@ -6763,7 +6751,6 @@ sched_create_group(struct container_subsys *ss, struct container *cont)
tg->cfs_rq[i] = cfs_rq;
init_cfs_rq(cfs_rq, rq);
cfs_rq->tg = tg;
- list_add_rcu(&cfs_rq->leaf_cfs_rq_list, &rq->leaf_cfs_rq_list);
tg->se[i] = se;
se->cfs_rq = &rq->cfs;
@@ -6773,12 +6760,15 @@ sched_create_group(struct container_subsys *ss, struct container *cont)
se->parent = NULL;
}
- tg->shares = NICE_0_LOAD;
+ for_each_possible_cpu(i) {
+ rq = cpu_rq(i);
+ cfs_rq = tg->cfs_rq[i];
+ list_add_rcu(&cfs_rq->leaf_cfs_rq_list, &rq->leaf_cfs_rq_list);
+ }
- /* Bind the container to task_grp object we just created */
- tg->css.container = cont;
+ tg->shares = NICE_0_LOAD;
- return &tg->css;
+ return tg;
err:
for_each_possible_cpu(i) {
@@ -6797,24 +6787,14 @@ err:
return ERR_PTR(-ENOMEM);
}
-
-/* destroy runqueue etc associated with a task group */
-static void sched_destroy_group(struct container_subsys *ss,
- struct container *cont)
+/* rcu callback to free various structures associated with a task group */
+static void free_sched_group(struct rcu_head *rhp)
{
- struct task_grp *tg = container_tg(cont);
- struct cfs_rq *cfs_rq;
+ struct cfs_rq *cfs_rq = container_of(rhp, struct cfs_rq, rcu);
+ struct task_grp *tg = cfs_rq->tg;
struct sched_entity *se;
int i;
- for_each_possible_cpu(i) {
- cfs_rq = tg->cfs_rq[i];
- list_del_rcu(&cfs_rq->leaf_cfs_rq_list);
- }
-
- /* wait for possible concurrent references to cfs_rqs complete */
- synchronize_sched();
-
/* now it should be safe to free those cfs_rqs */
for_each_possible_cpu(i) {
cfs_rq = tg->cfs_rq[i];
@@ -6829,19 +6809,29 @@ static void sched_destroy_group(struct container_subsys *ss,
kfree(tg);
}
-static int sched_can_attach(struct container_subsys *ss,
- struct container *cont, struct task_struct *tsk)
+/* Destroy runqueue etc associated with a task group */
+void sched_destroy_group(struct task_grp *tg)
{
- /* We don't support RT-tasks being in separate groups */
- if (tsk->sched_class != &fair_sched_class)
- return -EINVAL;
+ struct cfs_rq *cfs_rq;
+ int i;
- return 0;
+ for_each_possible_cpu(i) {
+ cfs_rq = tg->cfs_rq[i];
+ list_del_rcu(&cfs_rq->leaf_cfs_rq_list);
+ }
+
+ cfs_rq = tg->cfs_rq[0];
+
+ /* wait for possible concurrent references to cfs_rqs complete */
+ call_rcu(&cfs_rq->rcu, free_sched_group);
}
-/* change task's runqueue when it moves between groups */
-static void sched_move_task(struct container_subsys *ss, struct container *cont,
- struct container *old_cont, struct task_struct *tsk)
+/* change task's runqueue when it moves between groups.
+ * The caller of this function should have put the task in its new group
+ * by now. This function just updates tsk->se.cfs_rq and tsk->se.parent to
+ * reflect its new group.
+ */
+void sched_move_task(struct task_struct *tsk)
{
int on_rq, running;
unsigned long flags;
@@ -6896,58 +6886,20 @@ static void set_se_shares(struct sched_entity *se, unsigned long shares)
spin_unlock_irq(&rq->lock);
}
-static ssize_t cpu_shares_write(struct container *cont, struct cftype *cftype,
- struct file *file, const char __user *userbuf,
- size_t nbytes, loff_t *ppos)
+int sched_group_set_shares(struct task_grp *tg, unsigned long shares)
{
int i;
- unsigned long shareval;
- struct task_grp *tg = container_tg(cont);
- char buffer[2*sizeof(unsigned long) + 1];
-
- if (nbytes > 2*sizeof(unsigned long)) /* safety check */
- return -E2BIG;
- if (copy_from_user(buffer, userbuf, nbytes))
- return -EFAULT;
+ if (tg->shares == shares)
+ return 0;
- buffer[nbytes] = 0; /* nul-terminate */
- shareval = simple_strtoul(buffer, NULL, 10);
+ /* return -EINVAL if the new value is not sane */
- tg->shares = shareval;
+ tg->shares = shares;
for_each_possible_cpu(i)
- set_se_shares(tg->se[i], shareval);
-
- return nbytes;
-}
-
-static u64 cpu_shares_read_uint(struct container *cont, struct cftype *cft)
-{
- struct task_grp *tg = container_tg(cont);
-
- return (u64) tg->shares;
-}
+ set_se_shares(tg->se[i], shares);
-struct cftype cpuctl_share = {
- .name = "shares",
- .read_uint = cpu_shares_read_uint,
- .write = cpu_shares_write,
-};
-
-static int sched_populate(struct container_subsys *ss, struct container *cont)
-{
- return container_add_file(cont, ss, &cpuctl_share);
+ return 0;
}
-struct container_subsys cpu_subsys = {
- .name = "cpu",
- .create = sched_create_group,
- .destroy = sched_destroy_group,
- .can_attach = sched_can_attach,
- .attach = sched_move_task,
- .populate = sched_populate,
- .subsys_id = cpu_subsys_id,
- .early_init = 1,
-};
-
-#endif /* CONFIG_FAIR_GROUP_SCHED */
+#endif /* CONFIG_FAIR_GROUP_SCHED */
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 556942cf260..abd65ed9f2a 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -877,7 +877,10 @@ static int cfs_rq_best_prio(struct cfs_rq *cfs_rq)
if (!cfs_rq->nr_running)
return MAX_PRIO;
- curr = __pick_next_entity(cfs_rq);
+ curr = cfs_rq->curr;
+ if (!curr)
+ curr = __pick_next_entity(cfs_rq);
+
p = task_of(curr);
return p->prio;