[PATCH 1/4] SCHED: Add sched-domain roots

[Gregory Haskins <ghaskins@novell.com>]

We add the notion of a root-domain which will be used later to rescope
global variables to per-domain variables.  Each exclusive cpuset
essentially defines an island domain by fully partitioning the member cpus
from any other cpuset.  However, we currently still maintain some
policy/state as global variables which transcend all cpusets.  Consider,
for instance, rt-overload state.

Whenever a new exclusive cpuset is created, we also create a new
root-domain object and move each cpu member to the root-domain's span.
By default the system creates a single root-domain with all cpus as
members (mimicking the global state we have today).

We add some plumbing for storing class specific data in our root-domain.
Whenever a RQ is switching root-domains (because of repartitioning) we
give each sched_class the opportunity to remove any state from its old
domain and add state to the new one.  This logic doesn't have any clients
yet but it will later in the series.

Signed-off-by: Gregory Haskins <ghaskins@novell.com>
CC: Christoph Lameter <clameter@sgi.com>
CC: Paul Jackson <pj@sgi.com>
CC: Simon Derr <simon.derr@bull.net>
---

 include/linux/sched.h |    3 ++
 kernel/sched.c        |   89 +++++++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 89 insertions(+), 3 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 19d9f3f..0ba221e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -845,6 +845,9 @@ struct sched_class {
 	void (*task_tick) (struct rq *rq, struct task_struct *p);
 	void (*task_new) (struct rq *rq, struct task_struct *p);
 	void (*set_cpus_allowed)(struct task_struct *p, cpumask_t *newmask);
+
+	void (*join_domain)(struct rq *rq);
+	void (*leave_domain)(struct rq *rq);
 };
 
 struct load_weight {
diff --git a/kernel/sched.c b/kernel/sched.c
index e685402..e5c902c 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -276,6 +276,17 @@ struct rt_rq {
 	int overloaded;
 };
 
+#ifdef CONFIG_SMP
+
+struct root_domain {
+	atomic_t refcount;
+	cpumask_t span;
+};
+
+static struct root_domain def_root_domain;
+
+#endif
+
 /*
  * This is the main, per-CPU runqueue data structure.
  *
@@ -333,6 +344,7 @@ struct rq {
 	atomic_t nr_iowait;
 
 #ifdef CONFIG_SMP
+	struct root_domain  *rd;
 	struct sched_domain *sd;
 
 	/* For active balancing */
@@ -5718,11 +5730,68 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent)
 	return 1;
 }
 
+static void rq_attach_root(struct rq *rq, struct root_domain *rd)
+{
+	unsigned long flags;
+	const struct sched_class *class;
+
+	spin_lock_irqsave(&rq->lock, flags);
+
+	if (rq->rd) {
+		struct root_domain *old_rd = rq->rd;
+
+		for (class = sched_class_highest; class; class = class->next)
+			if (class->leave_domain)
+				class->leave_domain(rq);
+
+		if (atomic_dec_and_test(&old_rd->refcount))
+			kfree(old_rd);
+	}
+
+	atomic_inc(&rd->refcount);
+	rq->rd = rd;
+
+	for (class = sched_class_highest; class; class = class->next)
+		if (class->join_domain)
+			class->join_domain(rq);
+
+	spin_unlock_irqrestore(&rq->lock, flags);
+}
+
+static void init_rootdomain(struct root_domain *rd, cpumask_t *map)
+{
+	memset(rd, 0, sizeof(*rd));
+
+	rd->span = *map;
+}
+
+static void init_defrootdomain(void)
+{
+	cpumask_t cpus = CPU_MASK_ALL;
+
+	init_rootdomain(&def_root_domain, &cpus);
+	atomic_set(&def_root_domain.refcount, 1);
+}
+
+static struct root_domain *alloc_rootdomain(cpumask_t *map)
+{
+	struct root_domain *rd;
+
+	rd = kmalloc(sizeof(*rd), GFP_KERNEL);
+	if (!rd)
+		return NULL;
+
+	init_rootdomain(rd, map);
+
+	return rd;
+}
+
 /*
  * Attach the domain 'sd' to 'cpu' as its base domain.  Callers must
  * hold the hotplug lock.
  */
-static void cpu_attach_domain(struct sched_domain *sd, int cpu)
+static void cpu_attach_domain(struct sched_domain *sd,
+			      struct root_domain *rd, int cpu)
 {
 	struct rq *rq = cpu_rq(cpu);
 	struct sched_domain *tmp;
@@ -5747,6 +5816,7 @@ static void cpu_attach_domain(struct sched_domain *sd, int cpu)
 
 	sched_domain_debug(sd, cpu);
 
+	rq_attach_root(rq, rd);
 	rcu_assign_pointer(rq->sd, sd);
 }
 
@@ -6115,6 +6185,7 @@ static void init_sched_groups_power(int cpu, struct sched_domain *sd)
 static int build_sched_domains(const cpumask_t *cpu_map)
 {
 	int i;
+	struct root_domain *rd;
 #ifdef CONFIG_NUMA
 	struct sched_group **sched_group_nodes = NULL;
 	int sd_allnodes = 0;
@@ -6131,6 +6202,12 @@ static int build_sched_domains(const cpumask_t *cpu_map)
 	sched_group_nodes_bycpu[first_cpu(*cpu_map)] = sched_group_nodes;
 #endif
 
+	rd = alloc_rootdomain(cpu_map);
+	if (!rd) {
+		printk(KERN_WARNING "Cannot alloc root domain\n");
+		return -ENOMEM;
+	}
+
 	/*
 	 * Set up domains for cpus specified by the cpu_map.
 	 */
@@ -6347,7 +6424,7 @@ static int build_sched_domains(const cpumask_t *cpu_map)
 #else
 		sd = &per_cpu(phys_domains, i);
 #endif
-		cpu_attach_domain(sd, i);
+		cpu_attach_domain(sd, rd, i);
 	}
 
 	return 0;
@@ -6405,7 +6482,7 @@ static void detach_destroy_domains(const cpumask_t *cpu_map)
 	unregister_sched_domain_sysctl();
 
 	for_each_cpu_mask(i, *cpu_map)
-		cpu_attach_domain(NULL, i);
+		cpu_attach_domain(NULL, &def_root_domain, i);
 	synchronize_sched();
 	arch_destroy_sched_domains(cpu_map);
 }
@@ -6641,6 +6718,10 @@ void __init sched_init(void)
 	int highest_cpu = 0;
 	int i, j;
 
+#ifdef CONFIG_SMP
+	init_defrootdomain();
+#endif
+
 	for_each_possible_cpu(i) {
 		struct rt_prio_array *array;
 		struct rq *rq;
@@ -6680,6 +6761,8 @@ void __init sched_init(void)
 			rq->cpu_load[j] = 0;
 #ifdef CONFIG_SMP
 		rq->sd = NULL;
+		rq->rd = NULL;
+		rq_attach_root(rq, &def_root_domain);
 		rq->active_balance = 0;
 		rq->next_balance = jiffies;
 		rq->push_cpu = 0;