[PATCH 00/23] sched: Rewrite sched_domain/sched_group creation

[PATCH 00/23] sched: Rewrite sched_domain/sched_group creation

[Peter Zijlstra]

This series rewrite the sched_domain and sched_group creation code.

While its still not completely finished it does get us a lot of cleanups
and code reduction and seems fairly stable at this point and should thus
be a fairly good base to continue from.

Also available through:
  git://git.kernel.org/pub/scm/linux/kernel/git/peterz/linux-2.6-sched.git sched_domain

---
 include/linux/sched.h |   26 +-
 kernel/cpuset.c       |    2 +-
 kernel/sched.c        |  963 +++++++++++++++----------------------------------
 kernel/sched_fair.c   |   32 ++-
 4 files changed, 326 insertions(+), 697 deletions(-)

[PATCH 01/23] sched: Remove obsolete arch_ prefixes

[Peter Zijlstra]

[-- Attachment #1: sched-remove-arch-prefixes.patch --]
[-- Type: text/plain, Size: 2359 bytes --]

Non weak static functions clearly are not arch specific, so remove the
arch_ prefix.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -232,7 +232,7 @@ static void destroy_rt_bandwidth(struct
 #endif
 
 /*
- * sched_domains_mutex serializes calls to arch_init_sched_domains,
+ * sched_domains_mutex serializes calls to init_sched_domains,
  * detach_destroy_domains and partition_sched_domains.
  */
 static DEFINE_MUTEX(sched_domains_mutex);
@@ -7646,7 +7646,7 @@ void free_sched_domains(cpumask_var_t do
  * For now this just excludes isolated cpus, but could be used to
  * exclude other special cases in the future.
  */
-static int arch_init_sched_domains(const struct cpumask *cpu_map)
+static int init_sched_domains(const struct cpumask *cpu_map)
 {
 	int err;
 
@@ -7663,7 +7663,7 @@ static int arch_init_sched_domains(const
 	return err;
 }
 
-static void arch_destroy_sched_domains(const struct cpumask *cpu_map,
+static void destroy_sched_domains(const struct cpumask *cpu_map,
 				       struct cpumask *tmpmask)
 {
 	free_sched_groups(cpu_map, tmpmask);
@@ -7682,7 +7682,7 @@ static void detach_destroy_domains(const
 	for_each_cpu(i, cpu_map)
 		cpu_attach_domain(NULL, &def_root_domain, i);
 	synchronize_sched();
-	arch_destroy_sched_domains(cpu_map, to_cpumask(tmpmask));
+	destroy_sched_domains(cpu_map, to_cpumask(tmpmask));
 }
 
 /* handle null as "default" */
@@ -7791,7 +7791,7 @@ void partition_sched_domains(int ndoms_n
 }
 
 #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
-static void arch_reinit_sched_domains(void)
+static void reinit_sched_domains(void)
 {
 	get_online_cpus();
 
@@ -7824,7 +7824,7 @@ static ssize_t sched_power_savings_store
 	else
 		sched_mc_power_savings = level;
 
-	arch_reinit_sched_domains();
+	reinit_sched_domains();
 
 	return count;
 }
@@ -7950,7 +7950,7 @@ void __init sched_init_smp(void)
 #endif
 	get_online_cpus();
 	mutex_lock(&sched_domains_mutex);
-	arch_init_sched_domains(cpu_active_mask);
+	init_sched_domains(cpu_active_mask);
 	cpumask_andnot(non_isolated_cpus, cpu_possible_mask, cpu_isolated_map);
 	if (cpumask_empty(non_isolated_cpus))
 		cpumask_set_cpu(smp_processor_id(), non_isolated_cpus);

[tip:sched/domains] sched: Remove obsolete arch_ prefixes

[tip-bot for Peter Zijlstra]

Commit-ID:  c4a8849af939082052d8117f9ea3e170a99ff232
Gitweb:     http://git.kernel.org/tip/c4a8849af939082052d8117f9ea3e170a99ff232
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:42 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:16 +0200

sched: Remove obsolete arch_ prefixes

Non weak static functions clearly are not arch specific, so remove the
arch_ prefix.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122941.820460566@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   14 +++++++-------
 1 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index 4801363..d3e183c 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -231,7 +231,7 @@ static void destroy_rt_bandwidth(struct rt_bandwidth *rt_b)
 #endif
 
 /*
- * sched_domains_mutex serializes calls to arch_init_sched_domains,
+ * sched_domains_mutex serializes calls to init_sched_domains,
  * detach_destroy_domains and partition_sched_domains.
  */
 static DEFINE_MUTEX(sched_domains_mutex);
@@ -7670,7 +7670,7 @@ void free_sched_domains(cpumask_var_t doms[], unsigned int ndoms)
  * For now this just excludes isolated cpus, but could be used to
  * exclude other special cases in the future.
  */
-static int arch_init_sched_domains(const struct cpumask *cpu_map)
+static int init_sched_domains(const struct cpumask *cpu_map)
 {
 	int err;
 
@@ -7687,7 +7687,7 @@ static int arch_init_sched_domains(const struct cpumask *cpu_map)
 	return err;
 }
 
-static void arch_destroy_sched_domains(const struct cpumask *cpu_map,
+static void destroy_sched_domains(const struct cpumask *cpu_map,
 				       struct cpumask *tmpmask)
 {
 	free_sched_groups(cpu_map, tmpmask);
@@ -7706,7 +7706,7 @@ static void detach_destroy_domains(const struct cpumask *cpu_map)
 	for_each_cpu(i, cpu_map)
 		cpu_attach_domain(NULL, &def_root_domain, i);
 	synchronize_sched();
-	arch_destroy_sched_domains(cpu_map, to_cpumask(tmpmask));
+	destroy_sched_domains(cpu_map, to_cpumask(tmpmask));
 }
 
 /* handle null as "default" */
@@ -7815,7 +7815,7 @@ match2:
 }
 
 #if defined(CONFIG_SCHED_MC) || defined(CONFIG_SCHED_SMT)
-static void arch_reinit_sched_domains(void)
+static void reinit_sched_domains(void)
 {
 	get_online_cpus();
 
@@ -7848,7 +7848,7 @@ static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt)
 	else
 		sched_mc_power_savings = level;
 
-	arch_reinit_sched_domains();
+	reinit_sched_domains();
 
 	return count;
 }
@@ -7974,7 +7974,7 @@ void __init sched_init_smp(void)
 #endif
 	get_online_cpus();
 	mutex_lock(&sched_domains_mutex);
-	arch_init_sched_domains(cpu_active_mask);
+	init_sched_domains(cpu_active_mask);
 	cpumask_andnot(non_isolated_cpus, cpu_possible_mask, cpu_isolated_map);
 	if (cpumask_empty(non_isolated_cpus))
 		cpumask_set_cpu(smp_processor_id(), non_isolated_cpus);

[PATCH 02/23] sched: Simplify cpu_power initialization

[Peter Zijlstra]

[-- Attachment #1: sched-simplify-cpu_power.patch --]
[-- Type: text/plain, Size: 2830 bytes --]

The code in update_group_power() does what init_sched_groups_power()
does and more, so remove the special init_ code and call the generic
code instead.

Also move the sd->span_weight initialization because
update_group_power() needs it.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   44 +++++---------------------------------------
 1 file changed, 5 insertions(+), 39 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6655,9 +6655,6 @@ cpu_attach_domain(struct sched_domain *s
 	struct rq *rq = cpu_rq(cpu);
 	struct sched_domain *tmp;
 
-	for (tmp = sd; tmp; tmp = tmp->parent)
-		tmp->span_weight = cpumask_weight(sched_domain_span(tmp));
-
 	/* Remove the sched domains which do not contribute to scheduling. */
 	for (tmp = sd; tmp; ) {
 		struct sched_domain *parent = tmp->parent;
@@ -7135,11 +7132,6 @@ static void free_sched_groups(const stru
  */
 static void init_sched_groups_power(int cpu, struct sched_domain *sd)
 {
-	struct sched_domain *child;
-	struct sched_group *group;
-	long power;
-	int weight;
-
 	WARN_ON(!sd || !sd->groups);
 
 	if (cpu != group_first_cpu(sd->groups))
@@ -7147,36 +7139,7 @@ static void init_sched_groups_power(int
 
 	sd->groups->group_weight = cpumask_weight(sched_group_cpus(sd->groups));
 
-	child = sd->child;
-
-	sd->groups->cpu_power = 0;
-
-	if (!child) {
-		power = SCHED_LOAD_SCALE;
-		weight = cpumask_weight(sched_domain_span(sd));
-		/*
-		 * SMT siblings share the power of a single core.
-		 * Usually multiple threads get a better yield out of
-		 * that one core than a single thread would have,
-		 * reflect that in sd->smt_gain.
-		 */
-		if ((sd->flags & SD_SHARE_CPUPOWER) && weight > 1) {
-			power *= sd->smt_gain;
-			power /= weight;
-			power >>= SCHED_LOAD_SHIFT;
-		}
-		sd->groups->cpu_power += power;
-		return;
-	}
-
-	/*
-	 * Add cpu_power of each child group to this groups cpu_power.
-	 */
-	group = child->groups;
-	do {
-		sd->groups->cpu_power += group->cpu_power;
-		group = group->next;
-	} while (group != child->groups);
+	update_group_power(sd, cpu);
 }
 
 /*
@@ -7483,7 +7446,7 @@ static int __build_sched_domains(const s
 {
 	enum s_alloc alloc_state = sa_none;
 	struct s_data d;
-	struct sched_domain *sd;
+	struct sched_domain *sd, *tmp;
 	int i;
 #ifdef CONFIG_NUMA
 	d.sd_allnodes = 0;
@@ -7506,6 +7469,9 @@ static int __build_sched_domains(const s
 		sd = __build_book_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i);
+
+		for (tmp = sd; tmp; tmp = tmp->parent)
+			tmp->span_weight = cpumask_weight(sched_domain_span(tmp));
 	}
 
 	for_each_cpu(i, cpu_map) {

[tip:sched/domains] sched: Simplify ->cpu_power initialization

[tip-bot for Peter Zijlstra]

Commit-ID:  d274cb30f4a08045492d3f0c47cdf1a25668b1f5
Gitweb:     http://git.kernel.org/tip/d274cb30f4a08045492d3f0c47cdf1a25668b1f5
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:43 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:16 +0200

sched: Simplify ->cpu_power initialization

The code in update_group_power() does what init_sched_groups_power()
does and more, so remove the special init_ code and call the generic
code instead.

Also move the sd->span_weight initialization because
update_group_power() needs it.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122941.875856012@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   44 +++++---------------------------------------
 1 files changed, 5 insertions(+), 39 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index d3e183c..50d5fd3 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6679,9 +6679,6 @@ cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu)
 	struct rq *rq = cpu_rq(cpu);
 	struct sched_domain *tmp;
 
-	for (tmp = sd; tmp; tmp = tmp->parent)
-		tmp->span_weight = cpumask_weight(sched_domain_span(tmp));
-
 	/* Remove the sched domains which do not contribute to scheduling. */
 	for (tmp = sd; tmp; ) {
 		struct sched_domain *parent = tmp->parent;
@@ -7159,11 +7156,6 @@ static void free_sched_groups(const struct cpumask *cpu_map,
  */
 static void init_sched_groups_power(int cpu, struct sched_domain *sd)
 {
-	struct sched_domain *child;
-	struct sched_group *group;
-	long power;
-	int weight;
-
 	WARN_ON(!sd || !sd->groups);
 
 	if (cpu != group_first_cpu(sd->groups))
@@ -7171,36 +7163,7 @@ static void init_sched_groups_power(int cpu, struct sched_domain *sd)
 
 	sd->groups->group_weight = cpumask_weight(sched_group_cpus(sd->groups));
 
-	child = sd->child;
-
-	sd->groups->cpu_power = 0;
-
-	if (!child) {
-		power = SCHED_LOAD_SCALE;
-		weight = cpumask_weight(sched_domain_span(sd));
-		/*
-		 * SMT siblings share the power of a single core.
-		 * Usually multiple threads get a better yield out of
-		 * that one core than a single thread would have,
-		 * reflect that in sd->smt_gain.
-		 */
-		if ((sd->flags & SD_SHARE_CPUPOWER) && weight > 1) {
-			power *= sd->smt_gain;
-			power /= weight;
-			power >>= SCHED_LOAD_SHIFT;
-		}
-		sd->groups->cpu_power += power;
-		return;
-	}
-
-	/*
-	 * Add cpu_power of each child group to this groups cpu_power.
-	 */
-	group = child->groups;
-	do {
-		sd->groups->cpu_power += group->cpu_power;
-		group = group->next;
-	} while (group != child->groups);
+	update_group_power(sd, cpu);
 }
 
 /*
@@ -7507,7 +7470,7 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 {
 	enum s_alloc alloc_state = sa_none;
 	struct s_data d;
-	struct sched_domain *sd;
+	struct sched_domain *sd, *tmp;
 	int i;
 #ifdef CONFIG_NUMA
 	d.sd_allnodes = 0;
@@ -7530,6 +7493,9 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 		sd = __build_book_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i);
+
+		for (tmp = sd; tmp; tmp = tmp->parent)
+			tmp->span_weight = cpumask_weight(sched_domain_span(tmp));
 	}
 
 	for_each_cpu(i, cpu_map) {

[PATCH 03/23] sched: Simplify build_sched_groups

[Peter Zijlstra]

[-- Attachment #1: sched-foo3.patch --]
[-- Type: text/plain, Size: 4708 bytes --]

Notice that the mask being computed is the same as the domain span we
just computed. By using the domain_span we can avoid some mask
allocations and computations.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   52 ++++++++++++++++------------------------------------
 1 file changed, 16 insertions(+), 36 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6842,9 +6842,6 @@ struct s_data {
 	cpumask_var_t		notcovered;
 #endif
 	cpumask_var_t		nodemask;
-	cpumask_var_t		this_sibling_map;
-	cpumask_var_t		this_core_map;
-	cpumask_var_t		this_book_map;
 	cpumask_var_t		send_covered;
 	cpumask_var_t		tmpmask;
 	struct sched_group	**sched_group_nodes;
@@ -6856,9 +6853,6 @@ enum s_alloc {
 	sa_rootdomain,
 	sa_tmpmask,
 	sa_send_covered,
-	sa_this_book_map,
-	sa_this_core_map,
-	sa_this_sibling_map,
 	sa_nodemask,
 	sa_sched_group_nodes,
 #ifdef CONFIG_NUMA
@@ -7201,12 +7195,6 @@ static void __free_domain_allocs(struct
 		free_cpumask_var(d->tmpmask); /* fall through */
 	case sa_send_covered:
 		free_cpumask_var(d->send_covered); /* fall through */
-	case sa_this_book_map:
-		free_cpumask_var(d->this_book_map); /* fall through */
-	case sa_this_core_map:
-		free_cpumask_var(d->this_core_map); /* fall through */
-	case sa_this_sibling_map:
-		free_cpumask_var(d->this_sibling_map); /* fall through */
 	case sa_nodemask:
 		free_cpumask_var(d->nodemask); /* fall through */
 	case sa_sched_group_nodes:
@@ -7245,14 +7233,8 @@ static enum s_alloc __visit_domain_alloc
 #endif
 	if (!alloc_cpumask_var(&d->nodemask, GFP_KERNEL))
 		return sa_sched_group_nodes;
-	if (!alloc_cpumask_var(&d->this_sibling_map, GFP_KERNEL))
-		return sa_nodemask;
-	if (!alloc_cpumask_var(&d->this_core_map, GFP_KERNEL))
-		return sa_this_sibling_map;
-	if (!alloc_cpumask_var(&d->this_book_map, GFP_KERNEL))
-		return sa_this_core_map;
 	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
-		return sa_this_book_map;
+		return sa_nodemask;
 	if (!alloc_cpumask_var(&d->tmpmask, GFP_KERNEL))
 		return sa_send_covered;
 	d->rd = alloc_rootdomain();
@@ -7364,39 +7346,40 @@ static struct sched_domain *__build_smt_
 static void build_sched_groups(struct s_data *d, enum sched_domain_level l,
 			       const struct cpumask *cpu_map, int cpu)
 {
+	struct sched_domain *sd;
+
 	switch (l) {
 #ifdef CONFIG_SCHED_SMT
 	case SD_LV_SIBLING: /* set up CPU (sibling) groups */
-		cpumask_and(d->this_sibling_map, cpu_map,
-			    topology_thread_cpumask(cpu));
-		if (cpu == cpumask_first(d->this_sibling_map))
-			init_sched_build_groups(d->this_sibling_map, cpu_map,
+		sd = &per_cpu(cpu_domains, cpu).sd;
+		if (cpu == cpumask_first(sched_domain_span(sd)))
+			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_cpu_group,
 						d->send_covered, d->tmpmask);
 		break;
 #endif
 #ifdef CONFIG_SCHED_MC
 	case SD_LV_MC: /* set up multi-core groups */
-		cpumask_and(d->this_core_map, cpu_map, cpu_coregroup_mask(cpu));
-		if (cpu == cpumask_first(d->this_core_map))
-			init_sched_build_groups(d->this_core_map, cpu_map,
+		sd = &per_cpu(core_domains, cpu).sd;
+		if (cpu == cpumask_first(sched_domain_span(sd)))
+			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_core_group,
 						d->send_covered, d->tmpmask);
 		break;
 #endif
 #ifdef CONFIG_SCHED_BOOK
 	case SD_LV_BOOK: /* set up book groups */
-		cpumask_and(d->this_book_map, cpu_map, cpu_book_mask(cpu));
-		if (cpu == cpumask_first(d->this_book_map))
-			init_sched_build_groups(d->this_book_map, cpu_map,
+		sd = &per_cpu(book_domains, cpu).sd;
+		if (cpu == cpumask_first(sched_domain_span(sd)))
+			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_book_group,
 						d->send_covered, d->tmpmask);
 		break;
 #endif
 	case SD_LV_CPU: /* set up physical groups */
-		cpumask_and(d->nodemask, cpumask_of_node(cpu), cpu_map);
-		if (!cpumask_empty(d->nodemask))
-			init_sched_build_groups(d->nodemask, cpu_map,
+		sd = &per_cpu(phys_domains, cpu).sd;
+		if (cpu == cpumask_first(sched_domain_span(sd)))
+			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_phys_group,
 						d->send_covered, d->tmpmask);
 		break;
@@ -7452,11 +7435,8 @@ static int __build_sched_domains(const s
 		build_sched_groups(&d, SD_LV_SIBLING, cpu_map, i);
 		build_sched_groups(&d, SD_LV_BOOK, cpu_map, i);
 		build_sched_groups(&d, SD_LV_MC, cpu_map, i);
-	}
-
-	/* Set up physical groups */
-	for (i = 0; i < nr_node_ids; i++)
 		build_sched_groups(&d, SD_LV_CPU, cpu_map, i);
+	}
 
 #ifdef CONFIG_NUMA
 	/* Set up node groups */

[tip:sched/domains] sched: Simplify build_sched_groups()

[tip-bot for Peter Zijlstra]

Commit-ID:  a06dadbec5c5df0bf3a35f33616f67d10ca9ba28
Gitweb:     http://git.kernel.org/tip/a06dadbec5c5df0bf3a35f33616f67d10ca9ba28
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:44 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:17 +0200

sched: Simplify build_sched_groups()

Notice that the mask being computed is the same as the domain span we
just computed. By using the domain_span we can avoid some mask
allocations and computations.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122941.925028189@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   52 ++++++++++++++++------------------------------------
 1 files changed, 16 insertions(+), 36 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index 50d5fd3..e3818f1 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6866,9 +6866,6 @@ struct s_data {
 	cpumask_var_t		notcovered;
 #endif
 	cpumask_var_t		nodemask;
-	cpumask_var_t		this_sibling_map;
-	cpumask_var_t		this_core_map;
-	cpumask_var_t		this_book_map;
 	cpumask_var_t		send_covered;
 	cpumask_var_t		tmpmask;
 	struct sched_group	**sched_group_nodes;
@@ -6880,9 +6877,6 @@ enum s_alloc {
 	sa_rootdomain,
 	sa_tmpmask,
 	sa_send_covered,
-	sa_this_book_map,
-	sa_this_core_map,
-	sa_this_sibling_map,
 	sa_nodemask,
 	sa_sched_group_nodes,
 #ifdef CONFIG_NUMA
@@ -7251,12 +7245,6 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 		free_cpumask_var(d->tmpmask); /* fall through */
 	case sa_send_covered:
 		free_cpumask_var(d->send_covered); /* fall through */
-	case sa_this_book_map:
-		free_cpumask_var(d->this_book_map); /* fall through */
-	case sa_this_core_map:
-		free_cpumask_var(d->this_core_map); /* fall through */
-	case sa_this_sibling_map:
-		free_cpumask_var(d->this_sibling_map); /* fall through */
 	case sa_nodemask:
 		free_cpumask_var(d->nodemask); /* fall through */
 	case sa_sched_group_nodes:
@@ -7295,14 +7283,8 @@ static enum s_alloc __visit_domain_allocation_hell(struct s_data *d,
 #endif
 	if (!alloc_cpumask_var(&d->nodemask, GFP_KERNEL))
 		return sa_sched_group_nodes;
-	if (!alloc_cpumask_var(&d->this_sibling_map, GFP_KERNEL))
-		return sa_nodemask;
-	if (!alloc_cpumask_var(&d->this_core_map, GFP_KERNEL))
-		return sa_this_sibling_map;
-	if (!alloc_cpumask_var(&d->this_book_map, GFP_KERNEL))
-		return sa_this_core_map;
 	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
-		return sa_this_book_map;
+		return sa_nodemask;
 	if (!alloc_cpumask_var(&d->tmpmask, GFP_KERNEL))
 		return sa_send_covered;
 	d->rd = alloc_rootdomain();
@@ -7414,39 +7396,40 @@ static struct sched_domain *__build_smt_sched_domain(struct s_data *d,
 static void build_sched_groups(struct s_data *d, enum sched_domain_level l,
 			       const struct cpumask *cpu_map, int cpu)
 {
+	struct sched_domain *sd;
+
 	switch (l) {
 #ifdef CONFIG_SCHED_SMT
 	case SD_LV_SIBLING: /* set up CPU (sibling) groups */
-		cpumask_and(d->this_sibling_map, cpu_map,
-			    topology_thread_cpumask(cpu));
-		if (cpu == cpumask_first(d->this_sibling_map))
-			init_sched_build_groups(d->this_sibling_map, cpu_map,
+		sd = &per_cpu(cpu_domains, cpu).sd;
+		if (cpu == cpumask_first(sched_domain_span(sd)))
+			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_cpu_group,
 						d->send_covered, d->tmpmask);
 		break;
 #endif
 #ifdef CONFIG_SCHED_MC
 	case SD_LV_MC: /* set up multi-core groups */
-		cpumask_and(d->this_core_map, cpu_map, cpu_coregroup_mask(cpu));
-		if (cpu == cpumask_first(d->this_core_map))
-			init_sched_build_groups(d->this_core_map, cpu_map,
+		sd = &per_cpu(core_domains, cpu).sd;
+		if (cpu == cpumask_first(sched_domain_span(sd)))
+			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_core_group,
 						d->send_covered, d->tmpmask);
 		break;
 #endif
 #ifdef CONFIG_SCHED_BOOK
 	case SD_LV_BOOK: /* set up book groups */
-		cpumask_and(d->this_book_map, cpu_map, cpu_book_mask(cpu));
-		if (cpu == cpumask_first(d->this_book_map))
-			init_sched_build_groups(d->this_book_map, cpu_map,
+		sd = &per_cpu(book_domains, cpu).sd;
+		if (cpu == cpumask_first(sched_domain_span(sd)))
+			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_book_group,
 						d->send_covered, d->tmpmask);
 		break;
 #endif
 	case SD_LV_CPU: /* set up physical groups */
-		cpumask_and(d->nodemask, cpumask_of_node(cpu), cpu_map);
-		if (!cpumask_empty(d->nodemask))
-			init_sched_build_groups(d->nodemask, cpu_map,
+		sd = &per_cpu(phys_domains, cpu).sd;
+		if (cpu == cpumask_first(sched_domain_span(sd)))
+			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_phys_group,
 						d->send_covered, d->tmpmask);
 		break;
@@ -7502,11 +7485,8 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 		build_sched_groups(&d, SD_LV_SIBLING, cpu_map, i);
 		build_sched_groups(&d, SD_LV_BOOK, cpu_map, i);
 		build_sched_groups(&d, SD_LV_MC, cpu_map, i);
-	}
-
-	/* Set up physical groups */
-	for (i = 0; i < nr_node_ids; i++)
 		build_sched_groups(&d, SD_LV_CPU, cpu_map, i);
+	}
 
 #ifdef CONFIG_NUMA
 	/* Set up node groups */

[PATCH 04/23] sched: Change NODE sched_domain group creation

[Peter Zijlstra]

[-- Attachment #1: sched-foo1.patch --]
[-- Type: text/plain, Size: 11410 bytes --]

The NODE sched_domain is 'special' in that it allocates sched_groups
per CPU, instead of sharing the sched_groups between all CPUs.

While this might have some benefits on large NUMA and avoid remote
memory accesses when iterating the sched_groups, this does break
current code that assumes sched_groups are shared between all
sched_domains (since the dynamic cpu_power patches).

So refactor the NODE groups to behave like all other groups.

(The ALLNODES domain again shared its groups across the CPUs for some
reason).

If someone does measure a performance decrease due to this change we
need to revisit this and come up with another way to have both dynamic
cpu_power and NUMA work nice together.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |  231 ++++++++-------------------------------------------------
 1 file changed, 33 insertions(+), 198 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6837,29 +6837,18 @@ struct static_sched_domain {
 struct s_data {
 #ifdef CONFIG_NUMA
 	int			sd_allnodes;
-	cpumask_var_t		domainspan;
-	cpumask_var_t		covered;
-	cpumask_var_t		notcovered;
 #endif
 	cpumask_var_t		nodemask;
 	cpumask_var_t		send_covered;
 	cpumask_var_t		tmpmask;
-	struct sched_group	**sched_group_nodes;
 	struct root_domain	*rd;
 };
 
 enum s_alloc {
-	sa_sched_groups = 0,
 	sa_rootdomain,
 	sa_tmpmask,
 	sa_send_covered,
 	sa_nodemask,
-	sa_sched_group_nodes,
-#ifdef CONFIG_NUMA
-	sa_notcovered,
-	sa_covered,
-	sa_domainspan,
-#endif
 	sa_none,
 };
 
@@ -6955,18 +6944,10 @@ cpu_to_phys_group(int cpu, const struct
 }
 
 #ifdef CONFIG_NUMA
-/*
- * The init_sched_build_groups can't handle what we want to do with node
- * groups, so roll our own. Now each node has its own list of groups which
- * gets dynamically allocated.
- */
 static DEFINE_PER_CPU(struct static_sched_domain, node_domains);
-static struct sched_group ***sched_group_nodes_bycpu;
-
-static DEFINE_PER_CPU(struct static_sched_domain, allnodes_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_group_allnodes);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_node);
 
-static int cpu_to_allnodes_group(int cpu, const struct cpumask *cpu_map,
+static int cpu_to_node_group(int cpu, const struct cpumask *cpu_map,
 				 struct sched_group **sg,
 				 struct cpumask *nodemask)
 {
@@ -6976,142 +6957,27 @@ static int cpu_to_allnodes_group(int cpu
 	group = cpumask_first(nodemask);
 
 	if (sg)
-		*sg = &per_cpu(sched_group_allnodes, group).sg;
+		*sg = &per_cpu(sched_group_node, group).sg;
 	return group;
 }
 
-static void init_numa_sched_groups_power(struct sched_group *group_head)
-{
-	struct sched_group *sg = group_head;
-	int j;
-
-	if (!sg)
-		return;
-	do {
-		for_each_cpu(j, sched_group_cpus(sg)) {
-			struct sched_domain *sd;
-
-			sd = &per_cpu(phys_domains, j).sd;
-			if (j != group_first_cpu(sd->groups)) {
-				/*
-				 * Only add "power" once for each
-				 * physical package.
-				 */
-				continue;
-			}
-
-			sg->cpu_power += sd->groups->cpu_power;
-		}
-		sg = sg->next;
-	} while (sg != group_head);
-}
+static DEFINE_PER_CPU(struct static_sched_domain, allnodes_domains);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_allnodes);
 
-static int build_numa_sched_groups(struct s_data *d,
-				   const struct cpumask *cpu_map, int num)
+static int cpu_to_allnodes_group(int cpu, const struct cpumask *cpu_map,
+				 struct sched_group **sg,
+				 struct cpumask *nodemask)
 {
-	struct sched_domain *sd;
-	struct sched_group *sg, *prev;
-	int n, j;
+	int group;
 
-	cpumask_clear(d->covered);
-	cpumask_and(d->nodemask, cpumask_of_node(num), cpu_map);
-	if (cpumask_empty(d->nodemask)) {
-		d->sched_group_nodes[num] = NULL;
-		goto out;
-	}
-
-	sched_domain_node_span(num, d->domainspan);
-	cpumask_and(d->domainspan, d->domainspan, cpu_map);
-
-	sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(),
-			  GFP_KERNEL, num);
-	if (!sg) {
-		printk(KERN_WARNING "Can not alloc domain group for node %d\n",
-		       num);
-		return -ENOMEM;
-	}
-	d->sched_group_nodes[num] = sg;
-
-	for_each_cpu(j, d->nodemask) {
-		sd = &per_cpu(node_domains, j).sd;
-		sd->groups = sg;
-	}
-
-	sg->cpu_power = 0;
-	cpumask_copy(sched_group_cpus(sg), d->nodemask);
-	sg->next = sg;
-	cpumask_or(d->covered, d->covered, d->nodemask);
-
-	prev = sg;
-	for (j = 0; j < nr_node_ids; j++) {
-		n = (num + j) % nr_node_ids;
-		cpumask_complement(d->notcovered, d->covered);
-		cpumask_and(d->tmpmask, d->notcovered, cpu_map);
-		cpumask_and(d->tmpmask, d->tmpmask, d->domainspan);
-		if (cpumask_empty(d->tmpmask))
-			break;
-		cpumask_and(d->tmpmask, d->tmpmask, cpumask_of_node(n));
-		if (cpumask_empty(d->tmpmask))
-			continue;
-		sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(),
-				  GFP_KERNEL, num);
-		if (!sg) {
-			printk(KERN_WARNING
-			       "Can not alloc domain group for node %d\n", j);
-			return -ENOMEM;
-		}
-		sg->cpu_power = 0;
-		cpumask_copy(sched_group_cpus(sg), d->tmpmask);
-		sg->next = prev->next;
-		cpumask_or(d->covered, d->covered, d->tmpmask);
-		prev->next = sg;
-		prev = sg;
-	}
-out:
-	return 0;
-}
-#endif /* CONFIG_NUMA */
+	cpumask_and(nodemask, cpumask_of_node(cpu_to_node(cpu)), cpu_map);
+	group = cpumask_first(nodemask);
 
-#ifdef CONFIG_NUMA
-/* Free memory allocated for various sched_group structures */
-static void free_sched_groups(const struct cpumask *cpu_map,
-			      struct cpumask *nodemask)
-{
-	int cpu, i;
-
-	for_each_cpu(cpu, cpu_map) {
-		struct sched_group **sched_group_nodes
-			= sched_group_nodes_bycpu[cpu];
-
-		if (!sched_group_nodes)
-			continue;
-
-		for (i = 0; i < nr_node_ids; i++) {
-			struct sched_group *oldsg, *sg = sched_group_nodes[i];
-
-			cpumask_and(nodemask, cpumask_of_node(i), cpu_map);
-			if (cpumask_empty(nodemask))
-				continue;
-
-			if (sg == NULL)
-				continue;
-			sg = sg->next;
-next_sg:
-			oldsg = sg;
-			sg = sg->next;
-			kfree(oldsg);
-			if (oldsg != sched_group_nodes[i])
-				goto next_sg;
-		}
-		kfree(sched_group_nodes);
-		sched_group_nodes_bycpu[cpu] = NULL;
-	}
-}
-#else /* !CONFIG_NUMA */
-static void free_sched_groups(const struct cpumask *cpu_map,
-			      struct cpumask *nodemask)
-{
+	if (sg)
+		*sg = &per_cpu(sched_group_allnodes, group).sg;
+	return group;
 }
+
 #endif /* CONFIG_NUMA */
 
 /*
@@ -7212,9 +7078,6 @@ static void __free_domain_allocs(struct
 				 const struct cpumask *cpu_map)
 {
 	switch (what) {
-	case sa_sched_groups:
-		free_sched_groups(cpu_map, d->tmpmask); /* fall through */
-		d->sched_group_nodes = NULL;
 	case sa_rootdomain:
 		free_rootdomain(d->rd); /* fall through */
 	case sa_tmpmask:
@@ -7223,16 +7086,6 @@ static void __free_domain_allocs(struct
 		free_cpumask_var(d->send_covered); /* fall through */
 	case sa_nodemask:
 		free_cpumask_var(d->nodemask); /* fall through */
-	case sa_sched_group_nodes:
-#ifdef CONFIG_NUMA
-		kfree(d->sched_group_nodes); /* fall through */
-	case sa_notcovered:
-		free_cpumask_var(d->notcovered); /* fall through */
-	case sa_covered:
-		free_cpumask_var(d->covered); /* fall through */
-	case sa_domainspan:
-		free_cpumask_var(d->domainspan); /* fall through */
-#endif
 	case sa_none:
 		break;
 	}
@@ -7241,24 +7094,8 @@ static void __free_domain_allocs(struct
 static enum s_alloc __visit_domain_allocation_hell(struct s_data *d,
 						   const struct cpumask *cpu_map)
 {
-#ifdef CONFIG_NUMA
-	if (!alloc_cpumask_var(&d->domainspan, GFP_KERNEL))
-		return sa_none;
-	if (!alloc_cpumask_var(&d->covered, GFP_KERNEL))
-		return sa_domainspan;
-	if (!alloc_cpumask_var(&d->notcovered, GFP_KERNEL))
-		return sa_covered;
-	/* Allocate the per-node list of sched groups */
-	d->sched_group_nodes = kcalloc(nr_node_ids,
-				      sizeof(struct sched_group *), GFP_KERNEL);
-	if (!d->sched_group_nodes) {
-		printk(KERN_WARNING "Can not alloc sched group node list\n");
-		return sa_notcovered;
-	}
-	sched_group_nodes_bycpu[cpumask_first(cpu_map)] = d->sched_group_nodes;
-#endif
 	if (!alloc_cpumask_var(&d->nodemask, GFP_KERNEL))
-		return sa_sched_group_nodes;
+		return sa_none;
 	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
 		return sa_nodemask;
 	if (!alloc_cpumask_var(&d->tmpmask, GFP_KERNEL))
@@ -7298,6 +7135,7 @@ static struct sched_domain *__build_numa
 	if (parent)
 		parent->child = sd;
 	cpumask_and(sched_domain_span(sd), sched_domain_span(sd), cpu_map);
+	cpu_to_node_group(i, cpu_map, &sd->groups, d->tmpmask);
 #endif
 	return sd;
 }
@@ -7410,6 +7248,13 @@ static void build_sched_groups(struct s_
 						d->send_covered, d->tmpmask);
 		break;
 #ifdef CONFIG_NUMA
+	case SD_LV_NODE:
+		sd = &per_cpu(node_domains, cpu).sd;
+		if (cpu == cpumask_first(sched_domain_span(sd)))
+			init_sched_build_groups(sched_domain_span(sd), cpu_map,
+						&cpu_to_node_group,
+						d->send_covered, d->tmpmask);
+
 	case SD_LV_ALLNODES:
 		init_sched_build_groups(cpu_map, cpu_map, &cpu_to_allnodes_group,
 					d->send_covered, d->tmpmask);
@@ -7438,7 +7283,6 @@ static int __build_sched_domains(const s
 	alloc_state = __visit_domain_allocation_hell(&d, cpu_map);
 	if (alloc_state != sa_rootdomain)
 		goto error;
-	alloc_state = sa_sched_groups;
 
 	/*
 	 * Set up domains for cpus specified by the cpu_map.
@@ -7462,16 +7306,13 @@ static int __build_sched_domains(const s
 		build_sched_groups(&d, SD_LV_BOOK, cpu_map, i);
 		build_sched_groups(&d, SD_LV_MC, cpu_map, i);
 		build_sched_groups(&d, SD_LV_CPU, cpu_map, i);
+		build_sched_groups(&d, SD_LV_NODE, cpu_map, i);
 	}
 
 #ifdef CONFIG_NUMA
 	/* Set up node groups */
 	if (d.sd_allnodes)
 		build_sched_groups(&d, SD_LV_ALLNODES, cpu_map, 0);
-
-	for (i = 0; i < nr_node_ids; i++)
-		if (build_numa_sched_groups(&d, cpu_map, i))
-			goto error;
 #endif
 
 	/* Calculate CPU power for physical packages and nodes */
@@ -7500,15 +7341,16 @@ static int __build_sched_domains(const s
 	}
 
 #ifdef CONFIG_NUMA
-	for (i = 0; i < nr_node_ids; i++)
-		init_numa_sched_groups_power(d.sched_group_nodes[i]);
+	for_each_cpu(i, cpu_map) {
+		sd = &per_cpu(node_domains, i).sd;
+		init_sched_groups_power(i, sd);
+	}
 
 	if (d.sd_allnodes) {
-		struct sched_group *sg;
-
-		cpu_to_allnodes_group(cpumask_first(cpu_map), cpu_map, &sg,
-								d.tmpmask);
-		init_numa_sched_groups_power(sg);
+		for_each_cpu(i, cpu_map) {
+			sd = &per_cpu(allnodes_domains, i).sd;
+			init_sched_groups_power(i, sd);
+		}
 	}
 #endif
 
@@ -7526,7 +7368,6 @@ static int __build_sched_domains(const s
 		cpu_attach_domain(sd, d.rd, i);
 	}
 
-	d.sched_group_nodes = NULL; /* don't free this we still need it */
 	__free_domain_allocs(&d, sa_tmpmask, cpu_map);
 	return 0;
 
@@ -7612,7 +7453,6 @@ static int init_sched_domains(const stru
 static void destroy_sched_domains(const struct cpumask *cpu_map,
 				       struct cpumask *tmpmask)
 {
-	free_sched_groups(cpu_map, tmpmask);
 }
 
 /*
@@ -7889,11 +7729,6 @@ void __init sched_init_smp(void)
 	alloc_cpumask_var(&non_isolated_cpus, GFP_KERNEL);
 	alloc_cpumask_var(&fallback_doms, GFP_KERNEL);
 
-#if defined(CONFIG_NUMA)
-	sched_group_nodes_bycpu = kzalloc(nr_cpu_ids * sizeof(void **),
-								GFP_KERNEL);
-	BUG_ON(sched_group_nodes_bycpu == NULL);
-#endif
 	get_online_cpus();
 	mutex_lock(&sched_domains_mutex);
 	init_sched_domains(cpu_active_mask);

[tip:sched/domains] sched: Change NODE sched_domain group creation

[tip-bot for Peter Zijlstra]

Commit-ID:  cd4ea6ae3982f6861da3b510e69cbc194f331d83
Gitweb:     http://git.kernel.org/tip/cd4ea6ae3982f6861da3b510e69cbc194f331d83
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:45 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:17 +0200

sched: Change NODE sched_domain group creation

The NODE sched_domain is 'special' in that it allocates sched_groups
per CPU, instead of sharing the sched_groups between all CPUs.

While this might have some benefits on large NUMA and avoid remote
memory accesses when iterating the sched_groups, this does break
current code that assumes sched_groups are shared between all
sched_domains (since the dynamic cpu_power patches).

So refactor the NODE groups to behave like all other groups.

(The ALLNODES domain again shared its groups across the CPUs for some
reason).

If someone does measure a performance decrease due to this change we
need to revisit this and come up with another way to have both dynamic
cpu_power and NUMA work nice together.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122941.978111700@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |  229 ++++++++------------------------------------------------
 1 files changed, 32 insertions(+), 197 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index e3818f1..72d561f 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6861,29 +6861,18 @@ struct static_sched_domain {
 struct s_data {
 #ifdef CONFIG_NUMA
 	int			sd_allnodes;
-	cpumask_var_t		domainspan;
-	cpumask_var_t		covered;
-	cpumask_var_t		notcovered;
 #endif
 	cpumask_var_t		nodemask;
 	cpumask_var_t		send_covered;
 	cpumask_var_t		tmpmask;
-	struct sched_group	**sched_group_nodes;
 	struct root_domain	*rd;
 };
 
 enum s_alloc {
-	sa_sched_groups = 0,
 	sa_rootdomain,
 	sa_tmpmask,
 	sa_send_covered,
 	sa_nodemask,
-	sa_sched_group_nodes,
-#ifdef CONFIG_NUMA
-	sa_notcovered,
-	sa_covered,
-	sa_domainspan,
-#endif
 	sa_none,
 };
 
@@ -6979,18 +6968,10 @@ cpu_to_phys_group(int cpu, const struct cpumask *cpu_map,
 }
 
 #ifdef CONFIG_NUMA
-/*
- * The init_sched_build_groups can't handle what we want to do with node
- * groups, so roll our own. Now each node has its own list of groups which
- * gets dynamically allocated.
- */
 static DEFINE_PER_CPU(struct static_sched_domain, node_domains);
-static struct sched_group ***sched_group_nodes_bycpu;
-
-static DEFINE_PER_CPU(struct static_sched_domain, allnodes_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_group_allnodes);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_node);
 
-static int cpu_to_allnodes_group(int cpu, const struct cpumask *cpu_map,
+static int cpu_to_node_group(int cpu, const struct cpumask *cpu_map,
 				 struct sched_group **sg,
 				 struct cpumask *nodemask)
 {
@@ -7000,142 +6981,27 @@ static int cpu_to_allnodes_group(int cpu, const struct cpumask *cpu_map,
 	group = cpumask_first(nodemask);
 
 	if (sg)
-		*sg = &per_cpu(sched_group_allnodes, group).sg;
+		*sg = &per_cpu(sched_group_node, group).sg;
 	return group;
 }
 
-static void init_numa_sched_groups_power(struct sched_group *group_head)
-{
-	struct sched_group *sg = group_head;
-	int j;
-
-	if (!sg)
-		return;
-	do {
-		for_each_cpu(j, sched_group_cpus(sg)) {
-			struct sched_domain *sd;
-
-			sd = &per_cpu(phys_domains, j).sd;
-			if (j != group_first_cpu(sd->groups)) {
-				/*
-				 * Only add "power" once for each
-				 * physical package.
-				 */
-				continue;
-			}
-
-			sg->cpu_power += sd->groups->cpu_power;
-		}
-		sg = sg->next;
-	} while (sg != group_head);
-}
+static DEFINE_PER_CPU(struct static_sched_domain, allnodes_domains);
+static DEFINE_PER_CPU(struct static_sched_group, sched_group_allnodes);
 
-static int build_numa_sched_groups(struct s_data *d,
-				   const struct cpumask *cpu_map, int num)
+static int cpu_to_allnodes_group(int cpu, const struct cpumask *cpu_map,
+				 struct sched_group **sg,
+				 struct cpumask *nodemask)
 {
-	struct sched_domain *sd;
-	struct sched_group *sg, *prev;
-	int n, j;
-
-	cpumask_clear(d->covered);
-	cpumask_and(d->nodemask, cpumask_of_node(num), cpu_map);
-	if (cpumask_empty(d->nodemask)) {
-		d->sched_group_nodes[num] = NULL;
-		goto out;
-	}
-
-	sched_domain_node_span(num, d->domainspan);
-	cpumask_and(d->domainspan, d->domainspan, cpu_map);
-
-	sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(),
-			  GFP_KERNEL, num);
-	if (!sg) {
-		printk(KERN_WARNING "Can not alloc domain group for node %d\n",
-		       num);
-		return -ENOMEM;
-	}
-	d->sched_group_nodes[num] = sg;
-
-	for_each_cpu(j, d->nodemask) {
-		sd = &per_cpu(node_domains, j).sd;
-		sd->groups = sg;
-	}
+	int group;
 
-	sg->cpu_power = 0;
-	cpumask_copy(sched_group_cpus(sg), d->nodemask);
-	sg->next = sg;
-	cpumask_or(d->covered, d->covered, d->nodemask);
+	cpumask_and(nodemask, cpumask_of_node(cpu_to_node(cpu)), cpu_map);
+	group = cpumask_first(nodemask);
 
-	prev = sg;
-	for (j = 0; j < nr_node_ids; j++) {
-		n = (num + j) % nr_node_ids;
-		cpumask_complement(d->notcovered, d->covered);
-		cpumask_and(d->tmpmask, d->notcovered, cpu_map);
-		cpumask_and(d->tmpmask, d->tmpmask, d->domainspan);
-		if (cpumask_empty(d->tmpmask))
-			break;
-		cpumask_and(d->tmpmask, d->tmpmask, cpumask_of_node(n));
-		if (cpumask_empty(d->tmpmask))
-			continue;
-		sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(),
-				  GFP_KERNEL, num);
-		if (!sg) {
-			printk(KERN_WARNING
-			       "Can not alloc domain group for node %d\n", j);
-			return -ENOMEM;
-		}
-		sg->cpu_power = 0;
-		cpumask_copy(sched_group_cpus(sg), d->tmpmask);
-		sg->next = prev->next;
-		cpumask_or(d->covered, d->covered, d->tmpmask);
-		prev->next = sg;
-		prev = sg;
-	}
-out:
-	return 0;
+	if (sg)
+		*sg = &per_cpu(sched_group_allnodes, group).sg;
+	return group;
 }
-#endif /* CONFIG_NUMA */
-
-#ifdef CONFIG_NUMA
-/* Free memory allocated for various sched_group structures */
-static void free_sched_groups(const struct cpumask *cpu_map,
-			      struct cpumask *nodemask)
-{
-	int cpu, i;
 
-	for_each_cpu(cpu, cpu_map) {
-		struct sched_group **sched_group_nodes
-			= sched_group_nodes_bycpu[cpu];
-
-		if (!sched_group_nodes)
-			continue;
-
-		for (i = 0; i < nr_node_ids; i++) {
-			struct sched_group *oldsg, *sg = sched_group_nodes[i];
-
-			cpumask_and(nodemask, cpumask_of_node(i), cpu_map);
-			if (cpumask_empty(nodemask))
-				continue;
-
-			if (sg == NULL)
-				continue;
-			sg = sg->next;
-next_sg:
-			oldsg = sg;
-			sg = sg->next;
-			kfree(oldsg);
-			if (oldsg != sched_group_nodes[i])
-				goto next_sg;
-		}
-		kfree(sched_group_nodes);
-		sched_group_nodes_bycpu[cpu] = NULL;
-	}
-}
-#else /* !CONFIG_NUMA */
-static void free_sched_groups(const struct cpumask *cpu_map,
-			      struct cpumask *nodemask)
-{
-}
 #endif /* CONFIG_NUMA */
 
 /*
@@ -7236,9 +7102,6 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 				 const struct cpumask *cpu_map)
 {
 	switch (what) {
-	case sa_sched_groups:
-		free_sched_groups(cpu_map, d->tmpmask); /* fall through */
-		d->sched_group_nodes = NULL;
 	case sa_rootdomain:
 		free_rootdomain(d->rd); /* fall through */
 	case sa_tmpmask:
@@ -7247,16 +7110,6 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 		free_cpumask_var(d->send_covered); /* fall through */
 	case sa_nodemask:
 		free_cpumask_var(d->nodemask); /* fall through */
-	case sa_sched_group_nodes:
-#ifdef CONFIG_NUMA
-		kfree(d->sched_group_nodes); /* fall through */
-	case sa_notcovered:
-		free_cpumask_var(d->notcovered); /* fall through */
-	case sa_covered:
-		free_cpumask_var(d->covered); /* fall through */
-	case sa_domainspan:
-		free_cpumask_var(d->domainspan); /* fall through */
-#endif
 	case sa_none:
 		break;
 	}
@@ -7265,24 +7118,8 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 static enum s_alloc __visit_domain_allocation_hell(struct s_data *d,
 						   const struct cpumask *cpu_map)
 {
-#ifdef CONFIG_NUMA
-	if (!alloc_cpumask_var(&d->domainspan, GFP_KERNEL))
-		return sa_none;
-	if (!alloc_cpumask_var(&d->covered, GFP_KERNEL))
-		return sa_domainspan;
-	if (!alloc_cpumask_var(&d->notcovered, GFP_KERNEL))
-		return sa_covered;
-	/* Allocate the per-node list of sched groups */
-	d->sched_group_nodes = kcalloc(nr_node_ids,
-				      sizeof(struct sched_group *), GFP_KERNEL);
-	if (!d->sched_group_nodes) {
-		printk(KERN_WARNING "Can not alloc sched group node list\n");
-		return sa_notcovered;
-	}
-	sched_group_nodes_bycpu[cpumask_first(cpu_map)] = d->sched_group_nodes;
-#endif
 	if (!alloc_cpumask_var(&d->nodemask, GFP_KERNEL))
-		return sa_sched_group_nodes;
+		return sa_none;
 	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
 		return sa_nodemask;
 	if (!alloc_cpumask_var(&d->tmpmask, GFP_KERNEL))
@@ -7322,6 +7159,7 @@ static struct sched_domain *__build_numa_sched_domains(struct s_data *d,
 	if (parent)
 		parent->child = sd;
 	cpumask_and(sched_domain_span(sd), sched_domain_span(sd), cpu_map);
+	cpu_to_node_group(i, cpu_map, &sd->groups, d->tmpmask);
 #endif
 	return sd;
 }
@@ -7434,6 +7272,13 @@ static void build_sched_groups(struct s_data *d, enum sched_domain_level l,
 						d->send_covered, d->tmpmask);
 		break;
 #ifdef CONFIG_NUMA
+	case SD_LV_NODE:
+		sd = &per_cpu(node_domains, cpu).sd;
+		if (cpu == cpumask_first(sched_domain_span(sd)))
+			init_sched_build_groups(sched_domain_span(sd), cpu_map,
+						&cpu_to_node_group,
+						d->send_covered, d->tmpmask);
+
 	case SD_LV_ALLNODES:
 		init_sched_build_groups(cpu_map, cpu_map, &cpu_to_allnodes_group,
 					d->send_covered, d->tmpmask);
@@ -7462,7 +7307,6 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 	alloc_state = __visit_domain_allocation_hell(&d, cpu_map);
 	if (alloc_state != sa_rootdomain)
 		goto error;
-	alloc_state = sa_sched_groups;
 
 	/*
 	 * Set up domains for cpus specified by the cpu_map.
@@ -7486,16 +7330,13 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 		build_sched_groups(&d, SD_LV_BOOK, cpu_map, i);
 		build_sched_groups(&d, SD_LV_MC, cpu_map, i);
 		build_sched_groups(&d, SD_LV_CPU, cpu_map, i);
+		build_sched_groups(&d, SD_LV_NODE, cpu_map, i);
 	}
 
 #ifdef CONFIG_NUMA
 	/* Set up node groups */
 	if (d.sd_allnodes)
 		build_sched_groups(&d, SD_LV_ALLNODES, cpu_map, 0);
-
-	for (i = 0; i < nr_node_ids; i++)
-		if (build_numa_sched_groups(&d, cpu_map, i))
-			goto error;
 #endif
 
 	/* Calculate CPU power for physical packages and nodes */
@@ -7524,15 +7365,16 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 	}
 
 #ifdef CONFIG_NUMA
-	for (i = 0; i < nr_node_ids; i++)
-		init_numa_sched_groups_power(d.sched_group_nodes[i]);
+	for_each_cpu(i, cpu_map) {
+		sd = &per_cpu(node_domains, i).sd;
+		init_sched_groups_power(i, sd);
+	}
 
 	if (d.sd_allnodes) {
-		struct sched_group *sg;
-
-		cpu_to_allnodes_group(cpumask_first(cpu_map), cpu_map, &sg,
-								d.tmpmask);
-		init_numa_sched_groups_power(sg);
+		for_each_cpu(i, cpu_map) {
+			sd = &per_cpu(allnodes_domains, i).sd;
+			init_sched_groups_power(i, sd);
+		}
 	}
 #endif
 
@@ -7550,7 +7392,6 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 		cpu_attach_domain(sd, d.rd, i);
 	}
 
-	d.sched_group_nodes = NULL; /* don't free this we still need it */
 	__free_domain_allocs(&d, sa_tmpmask, cpu_map);
 	return 0;
 
@@ -7636,7 +7477,6 @@ static int init_sched_domains(const struct cpumask *cpu_map)
 static void destroy_sched_domains(const struct cpumask *cpu_map,
 				       struct cpumask *tmpmask)
 {
-	free_sched_groups(cpu_map, tmpmask);
 }
 
 /*
@@ -7913,11 +7753,6 @@ void __init sched_init_smp(void)
 	alloc_cpumask_var(&non_isolated_cpus, GFP_KERNEL);
 	alloc_cpumask_var(&fallback_doms, GFP_KERNEL);
 
-#if defined(CONFIG_NUMA)
-	sched_group_nodes_bycpu = kzalloc(nr_cpu_ids * sizeof(void **),
-								GFP_KERNEL);
-	BUG_ON(sched_group_nodes_bycpu == NULL);
-#endif
 	get_online_cpus();
 	mutex_lock(&sched_domains_mutex);
 	init_sched_domains(cpu_active_mask);

Re: [PATCH 04/23] sched: Change NODE sched_domain group creation

[Andreas Herrmann]

On Thu, Apr 07, 2011 at 08:09:45AM -0400, Peter Zijlstra wrote:
> The NODE sched_domain is 'special' in that it allocates sched_groups
> per CPU, instead of sharing the sched_groups between all CPUs.
> 
> While this might have some benefits on large NUMA and avoid remote
> memory accesses when iterating the sched_groups, this does break
> current code that assumes sched_groups are shared between all
> sched_domains (since the dynamic cpu_power patches).
> 
> So refactor the NODE groups to behave like all other groups.
> 
> (The ALLNODES domain again shared its groups across the CPUs for some
> reason).
> 
> If someone does measure a performance decrease due to this change we

[...]

Will do some performance sniff tests to check this.


Andreas

-- 
Operating | Advanced Micro Devices GmbH
  System  | Einsteinring 24, 85609 Dornach b. München, Germany
 Research | Geschäftsführer: Alberto Bozzo, Andrew Bowd
  Center  | Sitz: Dornach, Gemeinde Aschheim, Landkreis München
  (OSRC)  | Registergericht München, HRB Nr. 43632

[PATCH 05/23] sched: Clean up some ALLNODES code

[Peter Zijlstra]

[-- Attachment #1: sched-foo4.patch --]
[-- Type: text/plain, Size: 1194 bytes --]


Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -7256,7 +7256,9 @@ static void build_sched_groups(struct s_
 						d->send_covered, d->tmpmask);
 
 	case SD_LV_ALLNODES:
-		init_sched_build_groups(cpu_map, cpu_map, &cpu_to_allnodes_group,
+		if (cpu == cpumask_first(cpu_map))
+			init_sched_build_groups(cpu_map, cpu_map,
+					&cpu_to_allnodes_group,
 					d->send_covered, d->tmpmask);
 		break;
 #endif
@@ -7307,14 +7309,9 @@ static int __build_sched_domains(const s
 		build_sched_groups(&d, SD_LV_MC, cpu_map, i);
 		build_sched_groups(&d, SD_LV_CPU, cpu_map, i);
 		build_sched_groups(&d, SD_LV_NODE, cpu_map, i);
+		build_sched_groups(&d, SD_LV_ALLNODES, cpu_map, i);
 	}
 
-#ifdef CONFIG_NUMA
-	/* Set up node groups */
-	if (d.sd_allnodes)
-		build_sched_groups(&d, SD_LV_ALLNODES, cpu_map, 0);
-#endif
-
 	/* Calculate CPU power for physical packages and nodes */
 #ifdef CONFIG_SCHED_SMT
 	for_each_cpu(i, cpu_map) {

[tip:sched/domains] sched: Clean up some ALLNODES code

[tip-bot for Peter Zijlstra]

Commit-ID:  3739494e08da50c8a68d65eed5ba3012a54b40d4
Gitweb:     http://git.kernel.org/tip/3739494e08da50c8a68d65eed5ba3012a54b40d4
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:46 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:18 +0200

sched: Clean up some ALLNODES code

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.025636011@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   11 ++++-------
 1 files changed, 4 insertions(+), 7 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index 72d561f..fa10cf7 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -7280,7 +7280,9 @@ static void build_sched_groups(struct s_data *d, enum sched_domain_level l,
 						d->send_covered, d->tmpmask);
 
 	case SD_LV_ALLNODES:
-		init_sched_build_groups(cpu_map, cpu_map, &cpu_to_allnodes_group,
+		if (cpu == cpumask_first(cpu_map))
+			init_sched_build_groups(cpu_map, cpu_map,
+					&cpu_to_allnodes_group,
 					d->send_covered, d->tmpmask);
 		break;
 #endif
@@ -7331,14 +7333,9 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 		build_sched_groups(&d, SD_LV_MC, cpu_map, i);
 		build_sched_groups(&d, SD_LV_CPU, cpu_map, i);
 		build_sched_groups(&d, SD_LV_NODE, cpu_map, i);
+		build_sched_groups(&d, SD_LV_ALLNODES, cpu_map, i);
 	}
 
-#ifdef CONFIG_NUMA
-	/* Set up node groups */
-	if (d.sd_allnodes)
-		build_sched_groups(&d, SD_LV_ALLNODES, cpu_map, 0);
-#endif
-
 	/* Calculate CPU power for physical packages and nodes */
 #ifdef CONFIG_SCHED_SMT
 	for_each_cpu(i, cpu_map) {

[PATCH 06/23] sched: Simplify sched_group creation

[Peter Zijlstra]

[-- Attachment #1: sched-foo5.patch --]
[-- Type: text/plain, Size: 3165 bytes --]

Instead of calling build_sched_groups() for each possible sched_domain
we might have created, note that we can simply iterate the
sched_domain tree and call it for each sched_domain present.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   24 +++++-------------------
 1 file changed, 5 insertions(+), 19 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -7207,15 +7207,12 @@ static struct sched_domain *__build_smt_
 	return sd;
 }
 
-static void build_sched_groups(struct s_data *d, enum sched_domain_level l,
+static void build_sched_groups(struct s_data *d, struct sched_domain *sd,
 			       const struct cpumask *cpu_map, int cpu)
 {
-	struct sched_domain *sd;
-
-	switch (l) {
+	switch (sd->level) {
 #ifdef CONFIG_SCHED_SMT
 	case SD_LV_SIBLING: /* set up CPU (sibling) groups */
-		sd = &per_cpu(cpu_domains, cpu).sd;
 		if (cpu == cpumask_first(sched_domain_span(sd)))
 			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_cpu_group,
@@ -7224,7 +7221,6 @@ static void build_sched_groups(struct s_
 #endif
 #ifdef CONFIG_SCHED_MC
 	case SD_LV_MC: /* set up multi-core groups */
-		sd = &per_cpu(core_domains, cpu).sd;
 		if (cpu == cpumask_first(sched_domain_span(sd)))
 			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_core_group,
@@ -7233,7 +7229,6 @@ static void build_sched_groups(struct s_
 #endif
 #ifdef CONFIG_SCHED_BOOK
 	case SD_LV_BOOK: /* set up book groups */
-		sd = &per_cpu(book_domains, cpu).sd;
 		if (cpu == cpumask_first(sched_domain_span(sd)))
 			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_book_group,
@@ -7241,7 +7236,6 @@ static void build_sched_groups(struct s_
 		break;
 #endif
 	case SD_LV_CPU: /* set up physical groups */
-		sd = &per_cpu(phys_domains, cpu).sd;
 		if (cpu == cpumask_first(sched_domain_span(sd)))
 			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_phys_group,
@@ -7249,7 +7243,6 @@ static void build_sched_groups(struct s_
 		break;
 #ifdef CONFIG_NUMA
 	case SD_LV_NODE:
-		sd = &per_cpu(node_domains, cpu).sd;
 		if (cpu == cpumask_first(sched_domain_span(sd)))
 			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_node_group,
@@ -7299,17 +7292,10 @@ static int __build_sched_domains(const s
 		sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i);
 
-		for (tmp = sd; tmp; tmp = tmp->parent)
+		for (tmp = sd; tmp; tmp = tmp->parent) {
 			tmp->span_weight = cpumask_weight(sched_domain_span(tmp));
-	}
-
-	for_each_cpu(i, cpu_map) {
-		build_sched_groups(&d, SD_LV_SIBLING, cpu_map, i);
-		build_sched_groups(&d, SD_LV_BOOK, cpu_map, i);
-		build_sched_groups(&d, SD_LV_MC, cpu_map, i);
-		build_sched_groups(&d, SD_LV_CPU, cpu_map, i);
-		build_sched_groups(&d, SD_LV_NODE, cpu_map, i);
-		build_sched_groups(&d, SD_LV_ALLNODES, cpu_map, i);
+			build_sched_groups(&d, tmp, cpu_map, i);
+		}
 	}
 
 	/* Calculate CPU power for physical packages and nodes */

[tip:sched/domains] sched: Simplify sched_group creation

[tip-bot for Peter Zijlstra]

Commit-ID:  1cf51902546d60b8a7a6aba2dd557bd4ba8840ea
Gitweb:     http://git.kernel.org/tip/1cf51902546d60b8a7a6aba2dd557bd4ba8840ea
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:47 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:18 +0200

sched: Simplify sched_group creation

Instead of calling build_sched_groups() for each possible sched_domain
we might have created, note that we can simply iterate the
sched_domain tree and call it for each sched_domain present.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.077862519@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   24 +++++-------------------
 1 files changed, 5 insertions(+), 19 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index fa10cf7..e66d24a 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -7231,15 +7231,12 @@ static struct sched_domain *__build_smt_sched_domain(struct s_data *d,
 	return sd;
 }
 
-static void build_sched_groups(struct s_data *d, enum sched_domain_level l,
+static void build_sched_groups(struct s_data *d, struct sched_domain *sd,
 			       const struct cpumask *cpu_map, int cpu)
 {
-	struct sched_domain *sd;
-
-	switch (l) {
+	switch (sd->level) {
 #ifdef CONFIG_SCHED_SMT
 	case SD_LV_SIBLING: /* set up CPU (sibling) groups */
-		sd = &per_cpu(cpu_domains, cpu).sd;
 		if (cpu == cpumask_first(sched_domain_span(sd)))
 			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_cpu_group,
@@ -7248,7 +7245,6 @@ static void build_sched_groups(struct s_data *d, enum sched_domain_level l,
 #endif
 #ifdef CONFIG_SCHED_MC
 	case SD_LV_MC: /* set up multi-core groups */
-		sd = &per_cpu(core_domains, cpu).sd;
 		if (cpu == cpumask_first(sched_domain_span(sd)))
 			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_core_group,
@@ -7257,7 +7253,6 @@ static void build_sched_groups(struct s_data *d, enum sched_domain_level l,
 #endif
 #ifdef CONFIG_SCHED_BOOK
 	case SD_LV_BOOK: /* set up book groups */
-		sd = &per_cpu(book_domains, cpu).sd;
 		if (cpu == cpumask_first(sched_domain_span(sd)))
 			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_book_group,
@@ -7265,7 +7260,6 @@ static void build_sched_groups(struct s_data *d, enum sched_domain_level l,
 		break;
 #endif
 	case SD_LV_CPU: /* set up physical groups */
-		sd = &per_cpu(phys_domains, cpu).sd;
 		if (cpu == cpumask_first(sched_domain_span(sd)))
 			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_phys_group,
@@ -7273,7 +7267,6 @@ static void build_sched_groups(struct s_data *d, enum sched_domain_level l,
 		break;
 #ifdef CONFIG_NUMA
 	case SD_LV_NODE:
-		sd = &per_cpu(node_domains, cpu).sd;
 		if (cpu == cpumask_first(sched_domain_span(sd)))
 			init_sched_build_groups(sched_domain_span(sd), cpu_map,
 						&cpu_to_node_group,
@@ -7323,17 +7316,10 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 		sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i);
 
-		for (tmp = sd; tmp; tmp = tmp->parent)
+		for (tmp = sd; tmp; tmp = tmp->parent) {
 			tmp->span_weight = cpumask_weight(sched_domain_span(tmp));
-	}
-
-	for_each_cpu(i, cpu_map) {
-		build_sched_groups(&d, SD_LV_SIBLING, cpu_map, i);
-		build_sched_groups(&d, SD_LV_BOOK, cpu_map, i);
-		build_sched_groups(&d, SD_LV_MC, cpu_map, i);
-		build_sched_groups(&d, SD_LV_CPU, cpu_map, i);
-		build_sched_groups(&d, SD_LV_NODE, cpu_map, i);
-		build_sched_groups(&d, SD_LV_ALLNODES, cpu_map, i);
+			build_sched_groups(&d, tmp, cpu_map, i);
+		}
 	}
 
 	/* Calculate CPU power for physical packages and nodes */

[PATCH 07/23] sched: Simplify finding the lowest sched_domain

[Peter Zijlstra]

[-- Attachment #1: sched-foo6.patch --]
[-- Type: text/plain, Size: 2423 bytes --]

Instead of relying on knowing the build order and various CONFIG_
flags simply remember the bottom most sched_domain when we created the
domain hierarchy.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 include/asm-generic/percpu.h |    4 ++++
 kernel/sched.c               |   23 +++++++++++++----------
 2 files changed, 17 insertions(+), 10 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6841,11 +6841,13 @@ struct s_data {
 	cpumask_var_t		nodemask;
 	cpumask_var_t		send_covered;
 	cpumask_var_t		tmpmask;
+	struct sched_domain ** __percpu sd;
 	struct root_domain	*rd;
 };
 
 enum s_alloc {
 	sa_rootdomain,
+	sa_sd,
 	sa_tmpmask,
 	sa_send_covered,
 	sa_nodemask,
@@ -7080,6 +7082,8 @@ static void __free_domain_allocs(struct
 	switch (what) {
 	case sa_rootdomain:
 		free_rootdomain(d->rd); /* fall through */
+	case sa_sd:
+		free_percpu(d->sd); /* fall through */
 	case sa_tmpmask:
 		free_cpumask_var(d->tmpmask); /* fall through */
 	case sa_send_covered:
@@ -7100,10 +7104,15 @@ static enum s_alloc __visit_domain_alloc
 		return sa_nodemask;
 	if (!alloc_cpumask_var(&d->tmpmask, GFP_KERNEL))
 		return sa_send_covered;
+	d->sd = alloc_percpu(struct sched_domain *);
+	if (!d->sd) {
+		printk(KERN_WARNING "Cannot alloc per-cpu pointers\n");
+		return sa_tmpmask;
+	}
 	d->rd = alloc_rootdomain();
 	if (!d->rd) {
 		printk(KERN_WARNING "Cannot alloc root domain\n");
-		return sa_tmpmask;
+		return sa_sd;
 	}
 	return sa_rootdomain;
 }
@@ -7292,6 +7301,8 @@ static int __build_sched_domains(const s
 		sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i);
 
+		*per_cpu_ptr(d.sd, i) = sd;
+
 		for (tmp = sd; tmp; tmp = tmp->parent) {
 			tmp->span_weight = cpumask_weight(sched_domain_span(tmp));
 			build_sched_groups(&d, tmp, cpu_map, i);
@@ -7339,15 +7350,7 @@ static int __build_sched_domains(const s
 
 	/* Attach the domains */
 	for_each_cpu(i, cpu_map) {
-#ifdef CONFIG_SCHED_SMT
-		sd = &per_cpu(cpu_domains, i).sd;
-#elif defined(CONFIG_SCHED_MC)
-		sd = &per_cpu(core_domains, i).sd;
-#elif defined(CONFIG_SCHED_BOOK)
-		sd = &per_cpu(book_domains, i).sd;
-#else
-		sd = &per_cpu(phys_domains, i).sd;
-#endif
+		sd = *per_cpu_ptr(d.sd, i);
 		cpu_attach_domain(sd, d.rd, i);
 	}
 

[tip:sched/domains] sched: Simplify finding the lowest sched_domain

[tip-bot for Peter Zijlstra]

Commit-ID:  21d42ccfd6c6c11f96c2acfd32a85cfc33514d3a
Gitweb:     http://git.kernel.org/tip/21d42ccfd6c6c11f96c2acfd32a85cfc33514d3a
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:48 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:19 +0200

sched: Simplify finding the lowest sched_domain

Instead of relying on knowing the build order and various CONFIG_
flags simply remember the bottom most sched_domain when we created the
domain hierarchy.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.134511046@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   23 +++++++++++++----------
 1 files changed, 13 insertions(+), 10 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index e66d24a..d6992bf 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6865,11 +6865,13 @@ struct s_data {
 	cpumask_var_t		nodemask;
 	cpumask_var_t		send_covered;
 	cpumask_var_t		tmpmask;
+	struct sched_domain ** __percpu sd;
 	struct root_domain	*rd;
 };
 
 enum s_alloc {
 	sa_rootdomain,
+	sa_sd,
 	sa_tmpmask,
 	sa_send_covered,
 	sa_nodemask,
@@ -7104,6 +7106,8 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 	switch (what) {
 	case sa_rootdomain:
 		free_rootdomain(d->rd); /* fall through */
+	case sa_sd:
+		free_percpu(d->sd); /* fall through */
 	case sa_tmpmask:
 		free_cpumask_var(d->tmpmask); /* fall through */
 	case sa_send_covered:
@@ -7124,10 +7128,15 @@ static enum s_alloc __visit_domain_allocation_hell(struct s_data *d,
 		return sa_nodemask;
 	if (!alloc_cpumask_var(&d->tmpmask, GFP_KERNEL))
 		return sa_send_covered;
+	d->sd = alloc_percpu(struct sched_domain *);
+	if (!d->sd) {
+		printk(KERN_WARNING "Cannot alloc per-cpu pointers\n");
+		return sa_tmpmask;
+	}
 	d->rd = alloc_rootdomain();
 	if (!d->rd) {
 		printk(KERN_WARNING "Cannot alloc root domain\n");
-		return sa_tmpmask;
+		return sa_sd;
 	}
 	return sa_rootdomain;
 }
@@ -7316,6 +7325,8 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 		sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i);
 
+		*per_cpu_ptr(d.sd, i) = sd;
+
 		for (tmp = sd; tmp; tmp = tmp->parent) {
 			tmp->span_weight = cpumask_weight(sched_domain_span(tmp));
 			build_sched_groups(&d, tmp, cpu_map, i);
@@ -7363,15 +7374,7 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 
 	/* Attach the domains */
 	for_each_cpu(i, cpu_map) {
-#ifdef CONFIG_SCHED_SMT
-		sd = &per_cpu(cpu_domains, i).sd;
-#elif defined(CONFIG_SCHED_MC)
-		sd = &per_cpu(core_domains, i).sd;
-#elif defined(CONFIG_SCHED_BOOK)
-		sd = &per_cpu(book_domains, i).sd;
-#else
-		sd = &per_cpu(phys_domains, i).sd;
-#endif
+		sd = *per_cpu_ptr(d.sd, i);
 		cpu_attach_domain(sd, d.rd, i);
 	}
 

[PATCH 08/23] sched: Simplify sched_groups_power initialization

[Peter Zijlstra]

[-- Attachment #1: sched-foo7.patch --]
[-- Type: text/plain, Size: 1964 bytes --]

Again, instead of relying on knowing the possible domains and their
order, simply rely on the sched_domain tree and whatever domains are
present in there to initialize the sched_group cpu_power.

Note: we need to iterate the CPU mask backwards because of the
cpumask_first() condition for iterating up the tree. By iterating the
mask backwards we ensure all groups of a domain are set-up before
starting on the parent groups that rely on its children to be
completely done.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   39 +++++----------------------------------
 1 file changed, 5 insertions(+), 34 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -7310,43 +7310,14 @@ static int __build_sched_domains(const s
 	}
 
 	/* Calculate CPU power for physical packages and nodes */
-#ifdef CONFIG_SCHED_SMT
-	for_each_cpu(i, cpu_map) {
-		sd = &per_cpu(cpu_domains, i).sd;
-		init_sched_groups_power(i, sd);
-	}
-#endif
-#ifdef CONFIG_SCHED_MC
-	for_each_cpu(i, cpu_map) {
-		sd = &per_cpu(core_domains, i).sd;
-		init_sched_groups_power(i, sd);
-	}
-#endif
-#ifdef CONFIG_SCHED_BOOK
-	for_each_cpu(i, cpu_map) {
-		sd = &per_cpu(book_domains, i).sd;
-		init_sched_groups_power(i, sd);
-	}
-#endif
-
-	for_each_cpu(i, cpu_map) {
-		sd = &per_cpu(phys_domains, i).sd;
-		init_sched_groups_power(i, sd);
-	}
-
-#ifdef CONFIG_NUMA
-	for_each_cpu(i, cpu_map) {
-		sd = &per_cpu(node_domains, i).sd;
-		init_sched_groups_power(i, sd);
-	}
+	for (i = nr_cpumask_bits-1; i >= 0; i--) {
+		if (!cpumask_test_cpu(i, cpu_map))
+			continue;
 
-	if (d.sd_allnodes) {
-		for_each_cpu(i, cpu_map) {
-			sd = &per_cpu(allnodes_domains, i).sd;
+		sd = *per_cpu_ptr(d.sd, i);
+		for (; sd; sd = sd->parent)
 			init_sched_groups_power(i, sd);
-		}
 	}
-#endif
 
 	/* Attach the domains */
 	for_each_cpu(i, cpu_map) {

[tip:sched/domains] sched: Simplify sched_groups_power initialization

[tip-bot for Peter Zijlstra]

Commit-ID:  a9c9a9b6bff27ac9c746344a9c1a19bf3327002c
Gitweb:     http://git.kernel.org/tip/a9c9a9b6bff27ac9c746344a9c1a19bf3327002c
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:49 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:19 +0200

sched: Simplify sched_groups_power initialization

Again, instead of relying on knowing the possible domains and their
order, simply rely on the sched_domain tree and whatever domains are
present in there to initialize the sched_group cpu_power.

Note: we need to iterate the CPU mask backwards because of the
cpumask_first() condition for iterating up the tree. By iterating the
mask backwards we ensure all groups of a domain are set-up before
starting on the parent groups that rely on its children to be
completely done.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.187335414@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   39 +++++----------------------------------
 1 files changed, 5 insertions(+), 34 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index d6992bf..1cca59e 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -7334,43 +7334,14 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 	}
 
 	/* Calculate CPU power for physical packages and nodes */
-#ifdef CONFIG_SCHED_SMT
-	for_each_cpu(i, cpu_map) {
-		sd = &per_cpu(cpu_domains, i).sd;
-		init_sched_groups_power(i, sd);
-	}
-#endif
-#ifdef CONFIG_SCHED_MC
-	for_each_cpu(i, cpu_map) {
-		sd = &per_cpu(core_domains, i).sd;
-		init_sched_groups_power(i, sd);
-	}
-#endif
-#ifdef CONFIG_SCHED_BOOK
-	for_each_cpu(i, cpu_map) {
-		sd = &per_cpu(book_domains, i).sd;
-		init_sched_groups_power(i, sd);
-	}
-#endif
-
-	for_each_cpu(i, cpu_map) {
-		sd = &per_cpu(phys_domains, i).sd;
-		init_sched_groups_power(i, sd);
-	}
-
-#ifdef CONFIG_NUMA
-	for_each_cpu(i, cpu_map) {
-		sd = &per_cpu(node_domains, i).sd;
-		init_sched_groups_power(i, sd);
-	}
+	for (i = nr_cpumask_bits-1; i >= 0; i--) {
+		if (!cpumask_test_cpu(i, cpu_map))
+			continue;
 
-	if (d.sd_allnodes) {
-		for_each_cpu(i, cpu_map) {
-			sd = &per_cpu(allnodes_domains, i).sd;
+		sd = *per_cpu_ptr(d.sd, i);
+		for (; sd; sd = sd->parent)
 			init_sched_groups_power(i, sd);
-		}
 	}
-#endif
 
 	/* Attach the domains */
 	for_each_cpu(i, cpu_map) {

[PATCH 09/23] sched: Dynamically allocate sched_domain/sched_group data-structures

[Peter Zijlstra]

[-- Attachment #1: sched-foo8.patch --]
[-- Type: text/plain, Size: 27558 bytes --]

Instead of relying on static allocations for the sched_domain and
sched_group trees, dynamically allocate and RCU free them.

Allocating this dynamically also allows for some build_sched_groups()
simplification since we can now (like with other simplifications) rely
on the sched_domain tree instead of hard-coded knowledge.

One tricky to note is that detach_destroy_domains() needs to hold
rcu_read_lock() over the entire tear-down, per-cpu is not sufficient
since that can lead to partial sched_group existance (could possibly
be solved by doing the tear-down backwards but this is much more
robust).

A concequence of the above is that we can no longer print the
sched_domain debug stuff from cpu_attach_domain() since that might now
run with preemption disabled (due to classic RCU etc.) and
sched_domain_debug() does some GFP_KERNEL allocations.

Another thing to note is that we now fully rely on normal RCU and not
RCU-sched, this is because with the new and exiting RCU flavours we
grew over the years BH doesn't necessarily hold off RCU-sched grace
periods (-rt is known to break this). This would in fact already cause
us grief since we do sched_domain/sched_group iterations from softirq
context.

This patch is somewhat larger than I would like it to be, but I didn't
find any means of shrinking/splitting this.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 include/linux/sched.h |    5 
 kernel/sched.c        |  480 +++++++++++++++++++-------------------------------
 kernel/sched_fair.c   |   30 ++-
 3 files changed, 219 insertions(+), 296 deletions(-)

Index: linux-2.6/include/linux/sched.h
===================================================================
--- linux-2.6.orig/include/linux/sched.h
+++ linux-2.6/include/linux/sched.h
@@ -868,6 +868,7 @@ static inline int sd_power_saving_flags(
 
 struct sched_group {
 	struct sched_group *next;	/* Must be a circular list */
+	atomic_t ref;
 
 	/*
 	 * CPU power of this group, SCHED_LOAD_SCALE being max power for a
@@ -973,6 +974,10 @@ struct sched_domain {
 #ifdef CONFIG_SCHED_DEBUG
 	char *name;
 #endif
+	union {
+		void *private;		/* used during construction */
+		struct rcu_head rcu;	/* used during destruction */
+	};
 
 	unsigned int span_weight;
 	/*
Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -417,6 +417,7 @@ struct rt_rq {
  */
 struct root_domain {
 	atomic_t refcount;
+	struct rcu_head rcu;
 	cpumask_var_t span;
 	cpumask_var_t online;
 
@@ -571,7 +572,7 @@ static inline int cpu_of(struct rq *rq)
 
 #define rcu_dereference_check_sched_domain(p) \
 	rcu_dereference_check((p), \
-			      rcu_read_lock_sched_held() || \
+			      rcu_read_lock_held() || \
 			      lockdep_is_held(&sched_domains_mutex))
 
 /*
@@ -6558,12 +6559,11 @@ sd_parent_degenerate(struct sched_domain
 	return 1;
 }
 
-static void free_rootdomain(struct root_domain *rd)
+static void free_rootdomain(struct rcu_head *rcu)
 {
-	synchronize_sched();
+	struct root_domain *rd = container_of(rcu, struct root_domain, rcu);
 
 	cpupri_cleanup(&rd->cpupri);
-
 	free_cpumask_var(rd->rto_mask);
 	free_cpumask_var(rd->online);
 	free_cpumask_var(rd->span);
@@ -6604,7 +6604,7 @@ static void rq_attach_root(struct rq *rq
 	raw_spin_unlock_irqrestore(&rq->lock, flags);
 
 	if (old_rd)
-		free_rootdomain(old_rd);
+		call_rcu_sched(&old_rd->rcu, free_rootdomain);
 }
 
 static int init_rootdomain(struct root_domain *rd)
@@ -6655,6 +6655,25 @@ static struct root_domain *alloc_rootdom
 	return rd;
 }
 
+static void free_sched_domain(struct rcu_head *rcu)
+{
+	struct sched_domain *sd = container_of(rcu, struct sched_domain, rcu);
+	if (atomic_dec_and_test(&sd->groups->ref))
+		kfree(sd->groups);
+	kfree(sd);
+}
+
+static void destroy_sched_domain(struct sched_domain *sd, int cpu)
+{
+	call_rcu(&sd->rcu, free_sched_domain);
+}
+
+static void destroy_sched_domains(struct sched_domain *sd, int cpu)
+{
+	for (; sd; sd = sd->parent)
+		destroy_sched_domain(sd, cpu);
+}
+
 /*
  * Attach the domain 'sd' to 'cpu' as its base domain. Callers must
  * hold the hotplug lock.
@@ -6675,20 +6694,25 @@ cpu_attach_domain(struct sched_domain *s
 			tmp->parent = parent->parent;
 			if (parent->parent)
 				parent->parent->child = tmp;
+			destroy_sched_domain(parent, cpu);
 		} else
 			tmp = tmp->parent;
 	}
 
 	if (sd && sd_degenerate(sd)) {
+		tmp = sd;
 		sd = sd->parent;
+		destroy_sched_domain(tmp, cpu);
 		if (sd)
 			sd->child = NULL;
 	}
 
-	sched_domain_debug(sd, cpu);
+//	sched_domain_debug(sd, cpu);
 
 	rq_attach_root(rq, rd);
+	tmp = rq->sd;
 	rcu_assign_pointer(rq->sd, sd);
+	destroy_sched_domains(tmp, cpu);
 }
 
 /* cpus with isolated domains */
@@ -6704,56 +6728,6 @@ static int __init isolated_cpu_setup(cha
 
 __setup("isolcpus=", isolated_cpu_setup);
 
-/*
- * init_sched_build_groups takes the cpumask we wish to span, and a pointer
- * to a function which identifies what group(along with sched group) a CPU
- * belongs to. The return value of group_fn must be a >= 0 and < nr_cpu_ids
- * (due to the fact that we keep track of groups covered with a struct cpumask).
- *
- * init_sched_build_groups will build a circular linked list of the groups
- * covered by the given span, and will set each group's ->cpumask correctly,
- * and ->cpu_power to 0.
- */
-static void
-init_sched_build_groups(const struct cpumask *span,
-			const struct cpumask *cpu_map,
-			int (*group_fn)(int cpu, const struct cpumask *cpu_map,
-					struct sched_group **sg,
-					struct cpumask *tmpmask),
-			struct cpumask *covered, struct cpumask *tmpmask)
-{
-	struct sched_group *first = NULL, *last = NULL;
-	int i;
-
-	cpumask_clear(covered);
-
-	for_each_cpu(i, span) {
-		struct sched_group *sg;
-		int group = group_fn(i, cpu_map, &sg, tmpmask);
-		int j;
-
-		if (cpumask_test_cpu(i, covered))
-			continue;
-
-		cpumask_clear(sched_group_cpus(sg));
-		sg->cpu_power = 0;
-
-		for_each_cpu(j, span) {
-			if (group_fn(j, cpu_map, NULL, tmpmask) != group)
-				continue;
-
-			cpumask_set_cpu(j, covered);
-			cpumask_set_cpu(j, sched_group_cpus(sg));
-		}
-		if (!first)
-			first = sg;
-		if (last)
-			last->next = sg;
-		last = sg;
-	}
-	last->next = first;
-}
-
 #define SD_NODES_PER_DOMAIN 16
 
 #ifdef CONFIG_NUMA
@@ -6844,154 +6818,96 @@ struct static_sched_domain {
 	DECLARE_BITMAP(span, CONFIG_NR_CPUS);
 };
 
+struct sd_data {
+	struct sched_domain **__percpu sd;
+	struct sched_group **__percpu sg;
+};
+
 struct s_data {
 #ifdef CONFIG_NUMA
 	int			sd_allnodes;
 #endif
 	cpumask_var_t		nodemask;
 	cpumask_var_t		send_covered;
-	cpumask_var_t		tmpmask;
 	struct sched_domain ** __percpu sd;
+	struct sd_data 		sdd[SD_LV_MAX];
 	struct root_domain	*rd;
 };
 
 enum s_alloc {
 	sa_rootdomain,
 	sa_sd,
-	sa_tmpmask,
+	sa_sd_storage,
 	sa_send_covered,
 	sa_nodemask,
 	sa_none,
 };
 
 /*
- * SMT sched-domains:
+ * Assumes the sched_domain tree is fully constructed
  */
-#ifdef CONFIG_SCHED_SMT
-static DEFINE_PER_CPU(struct static_sched_domain, cpu_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_groups);
-
-static int
-cpu_to_cpu_group(int cpu, const struct cpumask *cpu_map,
-		 struct sched_group **sg, struct cpumask *unused)
+static int get_group(int cpu, struct sd_data *sdd, struct sched_group **sg)
 {
-	if (sg)
-		*sg = &per_cpu(sched_groups, cpu).sg;
-	return cpu;
-}
-#endif /* CONFIG_SCHED_SMT */
+	struct sched_domain *sd = *per_cpu_ptr(sdd->sd, cpu);
+	struct sched_domain *child = sd->child;
 
-/*
- * multi-core sched-domains:
- */
-#ifdef CONFIG_SCHED_MC
-static DEFINE_PER_CPU(struct static_sched_domain, core_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_group_core);
+	if (child)
+		cpu = cpumask_first(sched_domain_span(child));
 
-static int
-cpu_to_core_group(int cpu, const struct cpumask *cpu_map,
-		  struct sched_group **sg, struct cpumask *mask)
-{
-	int group;
-#ifdef CONFIG_SCHED_SMT
-	cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map);
-	group = cpumask_first(mask);
-#else
-	group = cpu;
-#endif
 	if (sg)
-		*sg = &per_cpu(sched_group_core, group).sg;
-	return group;
+		*sg = *per_cpu_ptr(sdd->sg, cpu);
+
+	return cpu;
 }
-#endif /* CONFIG_SCHED_MC */
 
 /*
- * book sched-domains:
+ * build_sched_groups takes the cpumask we wish to span, and a pointer
+ * to a function which identifies what group(along with sched group) a CPU
+ * belongs to. The return value of group_fn must be a >= 0 and < nr_cpu_ids
+ * (due to the fact that we keep track of groups covered with a struct cpumask).
+ *
+ * build_sched_groups will build a circular linked list of the groups
+ * covered by the given span, and will set each group's ->cpumask correctly,
+ * and ->cpu_power to 0.
  */
-#ifdef CONFIG_SCHED_BOOK
-static DEFINE_PER_CPU(struct static_sched_domain, book_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_group_book);
-
-static int
-cpu_to_book_group(int cpu, const struct cpumask *cpu_map,
-		  struct sched_group **sg, struct cpumask *mask)
-{
-	int group = cpu;
-#ifdef CONFIG_SCHED_MC
-	cpumask_and(mask, cpu_coregroup_mask(cpu), cpu_map);
-	group = cpumask_first(mask);
-#elif defined(CONFIG_SCHED_SMT)
-	cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map);
-	group = cpumask_first(mask);
-#endif
-	if (sg)
-		*sg = &per_cpu(sched_group_book, group).sg;
-	return group;
-}
-#endif /* CONFIG_SCHED_BOOK */
-
-static DEFINE_PER_CPU(struct static_sched_domain, phys_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_group_phys);
-
-static int
-cpu_to_phys_group(int cpu, const struct cpumask *cpu_map,
-		  struct sched_group **sg, struct cpumask *mask)
+static void
+build_sched_groups(struct sched_domain *sd, struct cpumask *covered)
 {
-	int group;
-#ifdef CONFIG_SCHED_BOOK
-	cpumask_and(mask, cpu_book_mask(cpu), cpu_map);
-	group = cpumask_first(mask);
-#elif defined(CONFIG_SCHED_MC)
-	cpumask_and(mask, cpu_coregroup_mask(cpu), cpu_map);
-	group = cpumask_first(mask);
-#elif defined(CONFIG_SCHED_SMT)
-	cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map);
-	group = cpumask_first(mask);
-#else
-	group = cpu;
-#endif
-	if (sg)
-		*sg = &per_cpu(sched_group_phys, group).sg;
-	return group;
-}
-
-#ifdef CONFIG_NUMA
-static DEFINE_PER_CPU(struct static_sched_domain, node_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_group_node);
+	struct sched_group *first = NULL, *last = NULL;
+	struct sd_data *sdd = sd->private;
+	const struct cpumask *span = sched_domain_span(sd);
+	int i;
 
-static int cpu_to_node_group(int cpu, const struct cpumask *cpu_map,
-				 struct sched_group **sg,
-				 struct cpumask *nodemask)
-{
-	int group;
+	cpumask_clear(covered);
 
-	cpumask_and(nodemask, cpumask_of_node(cpu_to_node(cpu)), cpu_map);
-	group = cpumask_first(nodemask);
+	for_each_cpu(i, span) {
+		struct sched_group *sg;
+		int group = get_group(i, sdd, &sg);
+		int j;
 
-	if (sg)
-		*sg = &per_cpu(sched_group_node, group).sg;
-	return group;
-}
+		if (cpumask_test_cpu(i, covered))
+			continue;
 
-static DEFINE_PER_CPU(struct static_sched_domain, allnodes_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_group_allnodes);
+		cpumask_clear(sched_group_cpus(sg));
+		sg->cpu_power = 0;
 
-static int cpu_to_allnodes_group(int cpu, const struct cpumask *cpu_map,
-				 struct sched_group **sg,
-				 struct cpumask *nodemask)
-{
-	int group;
+		for_each_cpu(j, span) {
+			if (get_group(j, sdd, NULL) != group)
+				continue;
 
-	cpumask_and(nodemask, cpumask_of_node(cpu_to_node(cpu)), cpu_map);
-	group = cpumask_first(nodemask);
+			cpumask_set_cpu(j, covered);
+			cpumask_set_cpu(j, sched_group_cpus(sg));
+		}
 
-	if (sg)
-		*sg = &per_cpu(sched_group_allnodes, group).sg;
-	return group;
+		if (!first)
+			first = sg;
+		if (last)
+			last->next = sg;
+		last = sg;
+	}
+	last->next = first;
 }
 
-#endif /* CONFIG_NUMA */
-
 /*
  * Initialize sched groups cpu_power.
  *
@@ -7025,15 +6941,15 @@ static void init_sched_groups_power(int 
 # define SD_INIT_NAME(sd, type)		do { } while (0)
 #endif
 
-#define	SD_INIT(sd, type)	sd_init_##type(sd)
-
-#define SD_INIT_FUNC(type)	\
-static noinline void sd_init_##type(struct sched_domain *sd)	\
-{								\
-	memset(sd, 0, sizeof(*sd));				\
-	*sd = SD_##type##_INIT;					\
-	sd->level = SD_LV_##type;				\
-	SD_INIT_NAME(sd, type);					\
+#define SD_INIT_FUNC(type)						       \
+static noinline struct sched_domain *sd_init_##type(struct s_data *d, int cpu) \
+{									       \
+	struct sched_domain *sd = *per_cpu_ptr(d->sdd[SD_LV_##type].sd, cpu);  \
+	*sd = SD_##type##_INIT;						       \
+	sd->level = SD_LV_##type;					       \
+	SD_INIT_NAME(sd, type);						       \
+	sd->private = &d->sdd[SD_LV_##type];				       \
+	return sd;							       \
 }
 
 SD_INIT_FUNC(CPU)
@@ -7089,13 +7005,22 @@ static void set_domain_attribute(struct 
 static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 				 const struct cpumask *cpu_map)
 {
+	int i, j;
+
 	switch (what) {
 	case sa_rootdomain:
-		free_rootdomain(d->rd); /* fall through */
+		free_rootdomain(&d->rd->rcu); /* fall through */
 	case sa_sd:
 		free_percpu(d->sd); /* fall through */
-	case sa_tmpmask:
-		free_cpumask_var(d->tmpmask); /* fall through */
+	case sa_sd_storage:
+		for (i = 0; i < SD_LV_MAX; i++) {
+			for_each_cpu(j, cpu_map) {
+				kfree(*per_cpu_ptr(d->sdd[i].sd, j));
+				kfree(*per_cpu_ptr(d->sdd[i].sg, j));
+			}
+			free_percpu(d->sdd[i].sd);
+			free_percpu(d->sdd[i].sg);
+		} /* fall through */
 	case sa_send_covered:
 		free_cpumask_var(d->send_covered); /* fall through */
 	case sa_nodemask:
@@ -7108,25 +7033,70 @@ static void __free_domain_allocs(struct 
 static enum s_alloc __visit_domain_allocation_hell(struct s_data *d,
 						   const struct cpumask *cpu_map)
 {
+	int i, j;
+
+	memset(d, 0, sizeof(*d));
+
 	if (!alloc_cpumask_var(&d->nodemask, GFP_KERNEL))
 		return sa_none;
 	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
 		return sa_nodemask;
-	if (!alloc_cpumask_var(&d->tmpmask, GFP_KERNEL))
-		return sa_send_covered;
-	d->sd = alloc_percpu(struct sched_domain *);
-	if (!d->sd) {
-		printk(KERN_WARNING "Cannot alloc per-cpu pointers\n");
-		return sa_tmpmask;
+	for (i = 0; i < SD_LV_MAX; i++) {
+		d->sdd[i].sd = alloc_percpu(struct sched_domain *);
+		if (!d->sdd[i].sd)
+			return sa_sd_storage;
+
+		d->sdd[i].sg = alloc_percpu(struct sched_group *);
+		if (!d->sdd[i].sg)
+			return sa_sd_storage;
+
+		for_each_cpu(j, cpu_map) {
+			struct sched_domain *sd;
+			struct sched_group *sg;
+
+		       	sd = kzalloc_node(sizeof(struct sched_domain) + cpumask_size(),
+					GFP_KERNEL, cpu_to_node(j));
+			if (!sd)
+				return sa_sd_storage;
+
+			*per_cpu_ptr(d->sdd[i].sd, j) = sd;
+
+			sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(),
+					GFP_KERNEL, cpu_to_node(j));
+			if (!sg)
+				return sa_sd_storage;
+
+			*per_cpu_ptr(d->sdd[i].sg, j) = sg;
+		}
 	}
+	d->sd = alloc_percpu(struct sched_domain *);
+	if (!d->sd)
+		return sa_sd_storage;
 	d->rd = alloc_rootdomain();
-	if (!d->rd) {
-		printk(KERN_WARNING "Cannot alloc root domain\n");
+	if (!d->rd)
 		return sa_sd;
-	}
 	return sa_rootdomain;
 }
 
+/*
+ * NULL the sd_data elements we've used to build the sched_domain and
+ * sched_group structure so that the subsequent __free_domain_allocs()
+ * will not free the data we're using.
+ */
+static void claim_allocations(int cpu, struct sched_domain *sd)
+{
+	struct sd_data *sdd = sd->private;
+	struct sched_group *sg = sd->groups;
+
+	WARN_ON_ONCE(*per_cpu_ptr(sdd->sd, cpu) != sd);
+	*per_cpu_ptr(sdd->sd, cpu) = NULL;
+
+	if (cpu == cpumask_first(sched_group_cpus(sg))) {
+		WARN_ON_ONCE(*per_cpu_ptr(sdd->sg, cpu) != sg);
+		*per_cpu_ptr(sdd->sg, cpu) = NULL;
+	}
+}
+
 static struct sched_domain *__build_numa_sched_domains(struct s_data *d,
 	const struct cpumask *cpu_map, struct sched_domain_attr *attr, int i)
 {
@@ -7137,24 +7107,20 @@ static struct sched_domain *__build_numa
 	d->sd_allnodes = 0;
 	if (cpumask_weight(cpu_map) >
 	    SD_NODES_PER_DOMAIN * cpumask_weight(d->nodemask)) {
-		sd = &per_cpu(allnodes_domains, i).sd;
-		SD_INIT(sd, ALLNODES);
+		sd = sd_init_ALLNODES(d, i);
 		set_domain_attribute(sd, attr);
 		cpumask_copy(sched_domain_span(sd), cpu_map);
-		cpu_to_allnodes_group(i, cpu_map, &sd->groups, d->tmpmask);
 		d->sd_allnodes = 1;
 	}
 	parent = sd;
 
-	sd = &per_cpu(node_domains, i).sd;
-	SD_INIT(sd, NODE);
+	sd = sd_init_NODE(d, i);
 	set_domain_attribute(sd, attr);
 	sched_domain_node_span(cpu_to_node(i), sched_domain_span(sd));
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;
 	cpumask_and(sched_domain_span(sd), sched_domain_span(sd), cpu_map);
-	cpu_to_node_group(i, cpu_map, &sd->groups, d->tmpmask);
 #endif
 	return sd;
 }
@@ -7164,14 +7130,12 @@ static struct sched_domain *__build_cpu_
 	struct sched_domain *parent, int i)
 {
 	struct sched_domain *sd;
-	sd = &per_cpu(phys_domains, i).sd;
-	SD_INIT(sd, CPU);
+	sd = sd_init_CPU(d, i);
 	set_domain_attribute(sd, attr);
 	cpumask_copy(sched_domain_span(sd), d->nodemask);
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;
-	cpu_to_phys_group(i, cpu_map, &sd->groups, d->tmpmask);
 	return sd;
 }
 
@@ -7181,13 +7145,11 @@ static struct sched_domain *__build_book
 {
 	struct sched_domain *sd = parent;
 #ifdef CONFIG_SCHED_BOOK
-	sd = &per_cpu(book_domains, i).sd;
-	SD_INIT(sd, BOOK);
+	sd = sd_init_BOOK(d, i);
 	set_domain_attribute(sd, attr);
 	cpumask_and(sched_domain_span(sd), cpu_map, cpu_book_mask(i));
 	sd->parent = parent;
 	parent->child = sd;
-	cpu_to_book_group(i, cpu_map, &sd->groups, d->tmpmask);
 #endif
 	return sd;
 }
@@ -7198,13 +7160,11 @@ static struct sched_domain *__build_mc_s
 {
 	struct sched_domain *sd = parent;
 #ifdef CONFIG_SCHED_MC
-	sd = &per_cpu(core_domains, i).sd;
-	SD_INIT(sd, MC);
+	sd = sd_init_MC(d, i);
 	set_domain_attribute(sd, attr);
 	cpumask_and(sched_domain_span(sd), cpu_map, cpu_coregroup_mask(i));
 	sd->parent = parent;
 	parent->child = sd;
-	cpu_to_core_group(i, cpu_map, &sd->groups, d->tmpmask);
 #endif
 	return sd;
 }
@@ -7215,92 +7175,32 @@ static struct sched_domain *__build_smt_
 {
 	struct sched_domain *sd = parent;
 #ifdef CONFIG_SCHED_SMT
-	sd = &per_cpu(cpu_domains, i).sd;
-	SD_INIT(sd, SIBLING);
+	sd = sd_init_SIBLING(d, i);
 	set_domain_attribute(sd, attr);
 	cpumask_and(sched_domain_span(sd), cpu_map, topology_thread_cpumask(i));
 	sd->parent = parent;
 	parent->child = sd;
-	cpu_to_cpu_group(i, cpu_map, &sd->groups, d->tmpmask);
 #endif
 	return sd;
 }
 
-static void build_sched_groups(struct s_data *d, struct sched_domain *sd,
-			       const struct cpumask *cpu_map, int cpu)
-{
-	switch (sd->level) {
-#ifdef CONFIG_SCHED_SMT
-	case SD_LV_SIBLING: /* set up CPU (sibling) groups */
-		if (cpu == cpumask_first(sched_domain_span(sd)))
-			init_sched_build_groups(sched_domain_span(sd), cpu_map,
-						&cpu_to_cpu_group,
-						d->send_covered, d->tmpmask);
-		break;
-#endif
-#ifdef CONFIG_SCHED_MC
-	case SD_LV_MC: /* set up multi-core groups */
-		if (cpu == cpumask_first(sched_domain_span(sd)))
-			init_sched_build_groups(sched_domain_span(sd), cpu_map,
-						&cpu_to_core_group,
-						d->send_covered, d->tmpmask);
-		break;
-#endif
-#ifdef CONFIG_SCHED_BOOK
-	case SD_LV_BOOK: /* set up book groups */
-		if (cpu == cpumask_first(sched_domain_span(sd)))
-			init_sched_build_groups(sched_domain_span(sd), cpu_map,
-						&cpu_to_book_group,
-						d->send_covered, d->tmpmask);
-		break;
-#endif
-	case SD_LV_CPU: /* set up physical groups */
-		if (cpu == cpumask_first(sched_domain_span(sd)))
-			init_sched_build_groups(sched_domain_span(sd), cpu_map,
-						&cpu_to_phys_group,
-						d->send_covered, d->tmpmask);
-		break;
-#ifdef CONFIG_NUMA
-	case SD_LV_NODE:
-		if (cpu == cpumask_first(sched_domain_span(sd)))
-			init_sched_build_groups(sched_domain_span(sd), cpu_map,
-						&cpu_to_node_group,
-						d->send_covered, d->tmpmask);
-
-	case SD_LV_ALLNODES:
-		if (cpu == cpumask_first(cpu_map))
-			init_sched_build_groups(cpu_map, cpu_map,
-					&cpu_to_allnodes_group,
-					d->send_covered, d->tmpmask);
-		break;
-#endif
-	default:
-		break;
-	}
-}
-
 /*
  * Build sched domains for a given set of cpus and attach the sched domains
  * to the individual cpus
  */
-static int __build_sched_domains(const struct cpumask *cpu_map,
-				 struct sched_domain_attr *attr)
+static int build_sched_domains(const struct cpumask *cpu_map,
+			       struct sched_domain_attr *attr)
 {
 	enum s_alloc alloc_state = sa_none;
+	struct sched_domain *sd;
 	struct s_data d;
-	struct sched_domain *sd, *tmp;
 	int i;
-#ifdef CONFIG_NUMA
-	d.sd_allnodes = 0;
-#endif
 
 	alloc_state = __visit_domain_allocation_hell(&d, cpu_map);
 	if (alloc_state != sa_rootdomain)
 		goto error;
 
-	/*
-	 * Set up domains for cpus specified by the cpu_map.
-	 */
+	/* Set up domains for cpus specified by the cpu_map. */
 	for_each_cpu(i, cpu_map) {
 		cpumask_and(d.nodemask, cpumask_of_node(cpu_to_node(i)),
 			    cpu_map);
@@ -7312,10 +7212,19 @@ static int __build_sched_domains(const s
 		sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i);
 
 		*per_cpu_ptr(d.sd, i) = sd;
+	}
+
+	/* Build the groups for the domains */
+	for_each_cpu(i, cpu_map) {
+		for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) {
+			sd->span_weight = cpumask_weight(sched_domain_span(sd));
+			get_group(i, sd->private, &sd->groups);
+			atomic_inc(&sd->groups->ref);
 
-		for (tmp = sd; tmp; tmp = tmp->parent) {
-			tmp->span_weight = cpumask_weight(sched_domain_span(tmp));
-			build_sched_groups(&d, tmp, cpu_map, i);
+			if (i != cpumask_first(sched_domain_span(sd)))
+				continue;
+
+			build_sched_groups(sd, d.send_covered);
 		}
 	}
 
@@ -7324,18 +7233,22 @@ static int __build_sched_domains(const s
 		if (!cpumask_test_cpu(i, cpu_map))
 			continue;
 
-		sd = *per_cpu_ptr(d.sd, i);
-		for (; sd; sd = sd->parent)
+		for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) {
+			claim_allocations(i, sd);
 			init_sched_groups_power(i, sd);
+		}
 	}
 
 	/* Attach the domains */
+	rcu_read_lock();
 	for_each_cpu(i, cpu_map) {
 		sd = *per_cpu_ptr(d.sd, i);
 		cpu_attach_domain(sd, d.rd, i);
+//		sched_domain_debug(sd, i);
 	}
+	rcu_read_unlock();
 
-	__free_domain_allocs(&d, sa_tmpmask, cpu_map);
+	__free_domain_allocs(&d, sa_sd, cpu_map);
 	return 0;
 
 error:
@@ -7343,11 +7256,6 @@ static int __build_sched_domains(const s
 	return -ENOMEM;
 }
 
-static int build_sched_domains(const struct cpumask *cpu_map)
-{
-	return __build_sched_domains(cpu_map, NULL);
-}
-
 static cpumask_var_t *doms_cur;	/* current sched domains */
 static int ndoms_cur;		/* number of sched domains in 'doms_cur' */
 static struct sched_domain_attr *dattr_cur;
@@ -7411,31 +7319,24 @@ static int init_sched_domains(const stru
 		doms_cur = &fallback_doms;
 	cpumask_andnot(doms_cur[0], cpu_map, cpu_isolated_map);
 	dattr_cur = NULL;
-	err = build_sched_domains(doms_cur[0]);
+	err = build_sched_domains(doms_cur[0], NULL);
 	register_sched_domain_sysctl();
 
 	return err;
 }
 
-static void destroy_sched_domains(const struct cpumask *cpu_map,
-				       struct cpumask *tmpmask)
-{
-}
-
 /*
  * Detach sched domains from a group of cpus specified in cpu_map
  * These cpus will now be attached to the NULL domain
  */
 static void detach_destroy_domains(const struct cpumask *cpu_map)
 {
-	/* Save because hotplug lock held. */
-	static DECLARE_BITMAP(tmpmask, CONFIG_NR_CPUS);
 	int i;
 
+	rcu_read_lock();
 	for_each_cpu(i, cpu_map)
 		cpu_attach_domain(NULL, &def_root_domain, i);
-	synchronize_sched();
-	destroy_sched_domains(cpu_map, to_cpumask(tmpmask));
+	rcu_read_unlock();
 }
 
 /* handle null as "default" */
@@ -7524,8 +7425,7 @@ void partition_sched_domains(int ndoms_n
 				goto match2;
 		}
 		/* no match - add a new doms_new */
-		__build_sched_domains(doms_new[i],
-					dattr_new ? dattr_new + i : NULL);
+		build_sched_domains(doms_new[i], dattr_new ? dattr_new + i : NULL);
 match2:
 		;
 	}
Index: linux-2.6/kernel/sched_fair.c
===================================================================
--- linux-2.6.orig/kernel/sched_fair.c
+++ linux-2.6/kernel/sched_fair.c
@@ -1621,6 +1621,7 @@ static int select_idle_sibling(struct ta
 	/*
 	 * Otherwise, iterate the domains and find an elegible idle cpu.
 	 */
+	rcu_read_lock();
 	for_each_domain(target, sd) {
 		if (!(sd->flags & SD_SHARE_PKG_RESOURCES))
 			break;
@@ -1640,6 +1641,7 @@ static int select_idle_sibling(struct ta
 		    cpumask_test_cpu(prev_cpu, sched_domain_span(sd)))
 			break;
 	}
+	rcu_read_unlock();
 
 	return target;
 }
@@ -1672,6 +1674,7 @@ select_task_rq_fair(struct rq *rq, struc
 		new_cpu = prev_cpu;
 	}
 
+	rcu_read_lock();
 	for_each_domain(cpu, tmp) {
 		if (!(tmp->flags & SD_LOAD_BALANCE))
 			continue;
@@ -1722,9 +1725,10 @@ select_task_rq_fair(struct rq *rq, struc
 
 	if (affine_sd) {
 		if (cpu == prev_cpu || wake_affine(affine_sd, p, sync))
-			return select_idle_sibling(p, cpu);
-		else
-			return select_idle_sibling(p, prev_cpu);
+			prev_cpu = cpu;
+
+		new_cpu = select_idle_sibling(p, prev_cpu);
+		goto unlock;
 	}
 
 	while (sd) {
@@ -1765,6 +1769,8 @@ select_task_rq_fair(struct rq *rq, struc
 		}
 		/* while loop will break here if sd == NULL */
 	}
+unlock:
+	rcu_read_unlock();
 
 	return new_cpu;
 }
@@ -3466,6 +3472,7 @@ static void idle_balance(int this_cpu, s
 	raw_spin_unlock(&this_rq->lock);
 
 	update_shares(this_cpu);
+	rcu_read_lock();
 	for_each_domain(this_cpu, sd) {
 		unsigned long interval;
 		int balance = 1;
@@ -3487,6 +3494,7 @@ static void idle_balance(int this_cpu, s
 			break;
 		}
 	}
+	rcu_read_unlock();
 
 	raw_spin_lock(&this_rq->lock);
 
@@ -3535,6 +3543,7 @@ static int active_load_balance_cpu_stop(
 	double_lock_balance(busiest_rq, target_rq);
 
 	/* Search for an sd spanning us and the target CPU. */
+	rcu_read_lock();
 	for_each_domain(target_cpu, sd) {
 		if ((sd->flags & SD_LOAD_BALANCE) &&
 		    cpumask_test_cpu(busiest_cpu, sched_domain_span(sd)))
@@ -3550,6 +3559,7 @@ static int active_load_balance_cpu_stop(
 		else
 			schedstat_inc(sd, alb_failed);
 	}
+	rcu_read_unlock();
 	double_unlock_balance(busiest_rq, target_rq);
 out_unlock:
 	busiest_rq->active_balance = 0;
@@ -3676,6 +3686,7 @@ static int find_new_ilb(int cpu)
 {
 	struct sched_domain *sd;
 	struct sched_group *ilb_group;
+	int ilb = nr_cpu_ids;
 
 	/*
 	 * Have idle load balancer selection from semi-idle packages only
@@ -3691,20 +3702,25 @@ static int find_new_ilb(int cpu)
 	if (cpumask_weight(nohz.idle_cpus_mask) < 2)
 		goto out_done;
 
+	rcu_read_lock();
 	for_each_flag_domain(cpu, sd, SD_POWERSAVINGS_BALANCE) {
 		ilb_group = sd->groups;
 
 		do {
-			if (is_semi_idle_group(ilb_group))
-				return cpumask_first(nohz.grp_idle_mask);
+			if (is_semi_idle_group(ilb_group)) {
+				ilb = cpumask_first(nohz.grp_idle_mask);
+				goto unlock;
+			}
 
 			ilb_group = ilb_group->next;
 
 		} while (ilb_group != sd->groups);
 	}
+unlock:
+	rcu_read_unlock();
 
 out_done:
-	return nr_cpu_ids;
+	return ilb;
 }
 #else /*  (CONFIG_SCHED_MC || CONFIG_SCHED_SMT) */
 static inline int find_new_ilb(int call_cpu)
@@ -3838,6 +3854,7 @@ static void rebalance_domains(int cpu, e
 
 	update_shares(cpu);
 
+	rcu_read_lock();
 	for_each_domain(cpu, sd) {
 		if (!(sd->flags & SD_LOAD_BALANCE))
 			continue;
@@ -3886,6 +3903,7 @@ static void rebalance_domains(int cpu, e
 		if (!balance)
 			break;
 	}
+	rcu_read_unlock();
 
 	/*
 	 * next_balance will be updated only when there is a need.

[tip:sched/domains] sched: Dynamically allocate sched_domain/sched_group data-structures

[tip-bot for Peter Zijlstra]

Commit-ID:  dce840a08702bd13a9a186e07e63d1ef82256b5e
Gitweb:     http://git.kernel.org/tip/dce840a08702bd13a9a186e07e63d1ef82256b5e
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:50 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:19 +0200

sched: Dynamically allocate sched_domain/sched_group data-structures

Instead of relying on static allocations for the sched_domain and
sched_group trees, dynamically allocate and RCU free them.

Allocating this dynamically also allows for some build_sched_groups()
simplification since we can now (like with other simplifications) rely
on the sched_domain tree instead of hard-coded knowledge.

One tricky to note is that detach_destroy_domains() needs to hold
rcu_read_lock() over the entire tear-down, per-cpu is not sufficient
since that can lead to partial sched_group existance (could possibly
be solved by doing the tear-down backwards but this is much more
robust).

A concequence of the above is that we can no longer print the
sched_domain debug stuff from cpu_attach_domain() since that might now
run with preemption disabled (due to classic RCU etc.) and
sched_domain_debug() does some GFP_KERNEL allocations.

Another thing to note is that we now fully rely on normal RCU and not
RCU-sched, this is because with the new and exiting RCU flavours we
grew over the years BH doesn't necessarily hold off RCU-sched grace
periods (-rt is known to break this). This would in fact already cause
us grief since we do sched_domain/sched_group iterations from softirq
context.

This patch is somewhat larger than I would like it to be, but I didn't
find any means of shrinking/splitting this.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.245307941@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |    5 +
 kernel/sched.c        |  479 +++++++++++++++++++------------------------------
 kernel/sched_fair.c   |   30 +++-
 3 files changed, 218 insertions(+), 296 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4ec2c02..020b79d 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -868,6 +868,7 @@ static inline int sd_power_saving_flags(void)
 
 struct sched_group {
 	struct sched_group *next;	/* Must be a circular list */
+	atomic_t ref;
 
 	/*
 	 * CPU power of this group, SCHED_LOAD_SCALE being max power for a
@@ -973,6 +974,10 @@ struct sched_domain {
 #ifdef CONFIG_SCHED_DEBUG
 	char *name;
 #endif
+	union {
+		void *private;		/* used during construction */
+		struct rcu_head rcu;	/* used during destruction */
+	};
 
 	unsigned int span_weight;
 	/*
diff --git a/kernel/sched.c b/kernel/sched.c
index 1cca59e..6520484 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -417,6 +417,7 @@ struct rt_rq {
  */
 struct root_domain {
 	atomic_t refcount;
+	struct rcu_head rcu;
 	cpumask_var_t span;
 	cpumask_var_t online;
 
@@ -571,7 +572,7 @@ static inline int cpu_of(struct rq *rq)
 
 #define rcu_dereference_check_sched_domain(p) \
 	rcu_dereference_check((p), \
-			      rcu_read_lock_sched_held() || \
+			      rcu_read_lock_held() || \
 			      lockdep_is_held(&sched_domains_mutex))
 
 /*
@@ -6572,12 +6573,11 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent)
 	return 1;
 }
 
-static void free_rootdomain(struct root_domain *rd)
+static void free_rootdomain(struct rcu_head *rcu)
 {
-	synchronize_sched();
+	struct root_domain *rd = container_of(rcu, struct root_domain, rcu);
 
 	cpupri_cleanup(&rd->cpupri);
-
 	free_cpumask_var(rd->rto_mask);
 	free_cpumask_var(rd->online);
 	free_cpumask_var(rd->span);
@@ -6618,7 +6618,7 @@ static void rq_attach_root(struct rq *rq, struct root_domain *rd)
 	raw_spin_unlock_irqrestore(&rq->lock, flags);
 
 	if (old_rd)
-		free_rootdomain(old_rd);
+		call_rcu_sched(&old_rd->rcu, free_rootdomain);
 }
 
 static int init_rootdomain(struct root_domain *rd)
@@ -6669,6 +6669,25 @@ static struct root_domain *alloc_rootdomain(void)
 	return rd;
 }
 
+static void free_sched_domain(struct rcu_head *rcu)
+{
+	struct sched_domain *sd = container_of(rcu, struct sched_domain, rcu);
+	if (atomic_dec_and_test(&sd->groups->ref))
+		kfree(sd->groups);
+	kfree(sd);
+}
+
+static void destroy_sched_domain(struct sched_domain *sd, int cpu)
+{
+	call_rcu(&sd->rcu, free_sched_domain);
+}
+
+static void destroy_sched_domains(struct sched_domain *sd, int cpu)
+{
+	for (; sd; sd = sd->parent)
+		destroy_sched_domain(sd, cpu);
+}
+
 /*
  * Attach the domain 'sd' to 'cpu' as its base domain. Callers must
  * hold the hotplug lock.
@@ -6689,20 +6708,25 @@ cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu)
 			tmp->parent = parent->parent;
 			if (parent->parent)
 				parent->parent->child = tmp;
+			destroy_sched_domain(parent, cpu);
 		} else
 			tmp = tmp->parent;
 	}
 
 	if (sd && sd_degenerate(sd)) {
+		tmp = sd;
 		sd = sd->parent;
+		destroy_sched_domain(tmp, cpu);
 		if (sd)
 			sd->child = NULL;
 	}
 
-	sched_domain_debug(sd, cpu);
+	/* sched_domain_debug(sd, cpu); */
 
 	rq_attach_root(rq, rd);
+	tmp = rq->sd;
 	rcu_assign_pointer(rq->sd, sd);
+	destroy_sched_domains(tmp, cpu);
 }
 
 /* cpus with isolated domains */
@@ -6718,56 +6742,6 @@ static int __init isolated_cpu_setup(char *str)
 
 __setup("isolcpus=", isolated_cpu_setup);
 
-/*
- * init_sched_build_groups takes the cpumask we wish to span, and a pointer
- * to a function which identifies what group(along with sched group) a CPU
- * belongs to. The return value of group_fn must be a >= 0 and < nr_cpu_ids
- * (due to the fact that we keep track of groups covered with a struct cpumask).
- *
- * init_sched_build_groups will build a circular linked list of the groups
- * covered by the given span, and will set each group's ->cpumask correctly,
- * and ->cpu_power to 0.
- */
-static void
-init_sched_build_groups(const struct cpumask *span,
-			const struct cpumask *cpu_map,
-			int (*group_fn)(int cpu, const struct cpumask *cpu_map,
-					struct sched_group **sg,
-					struct cpumask *tmpmask),
-			struct cpumask *covered, struct cpumask *tmpmask)
-{
-	struct sched_group *first = NULL, *last = NULL;
-	int i;
-
-	cpumask_clear(covered);
-
-	for_each_cpu(i, span) {
-		struct sched_group *sg;
-		int group = group_fn(i, cpu_map, &sg, tmpmask);
-		int j;
-
-		if (cpumask_test_cpu(i, covered))
-			continue;
-
-		cpumask_clear(sched_group_cpus(sg));
-		sg->cpu_power = 0;
-
-		for_each_cpu(j, span) {
-			if (group_fn(j, cpu_map, NULL, tmpmask) != group)
-				continue;
-
-			cpumask_set_cpu(j, covered);
-			cpumask_set_cpu(j, sched_group_cpus(sg));
-		}
-		if (!first)
-			first = sg;
-		if (last)
-			last->next = sg;
-		last = sg;
-	}
-	last->next = first;
-}
-
 #define SD_NODES_PER_DOMAIN 16
 
 #ifdef CONFIG_NUMA
@@ -6858,154 +6832,96 @@ struct static_sched_domain {
 	DECLARE_BITMAP(span, CONFIG_NR_CPUS);
 };
 
+struct sd_data {
+	struct sched_domain **__percpu sd;
+	struct sched_group **__percpu sg;
+};
+
 struct s_data {
 #ifdef CONFIG_NUMA
 	int			sd_allnodes;
 #endif
 	cpumask_var_t		nodemask;
 	cpumask_var_t		send_covered;
-	cpumask_var_t		tmpmask;
 	struct sched_domain ** __percpu sd;
+	struct sd_data 		sdd[SD_LV_MAX];
 	struct root_domain	*rd;
 };
 
 enum s_alloc {
 	sa_rootdomain,
 	sa_sd,
-	sa_tmpmask,
+	sa_sd_storage,
 	sa_send_covered,
 	sa_nodemask,
 	sa_none,
 };
 
 /*
- * SMT sched-domains:
+ * Assumes the sched_domain tree is fully constructed
  */
-#ifdef CONFIG_SCHED_SMT
-static DEFINE_PER_CPU(struct static_sched_domain, cpu_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_groups);
-
-static int
-cpu_to_cpu_group(int cpu, const struct cpumask *cpu_map,
-		 struct sched_group **sg, struct cpumask *unused)
+static int get_group(int cpu, struct sd_data *sdd, struct sched_group **sg)
 {
-	if (sg)
-		*sg = &per_cpu(sched_groups, cpu).sg;
-	return cpu;
-}
-#endif /* CONFIG_SCHED_SMT */
+	struct sched_domain *sd = *per_cpu_ptr(sdd->sd, cpu);
+	struct sched_domain *child = sd->child;
 
-/*
- * multi-core sched-domains:
- */
-#ifdef CONFIG_SCHED_MC
-static DEFINE_PER_CPU(struct static_sched_domain, core_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_group_core);
+	if (child)
+		cpu = cpumask_first(sched_domain_span(child));
 
-static int
-cpu_to_core_group(int cpu, const struct cpumask *cpu_map,
-		  struct sched_group **sg, struct cpumask *mask)
-{
-	int group;
-#ifdef CONFIG_SCHED_SMT
-	cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map);
-	group = cpumask_first(mask);
-#else
-	group = cpu;
-#endif
 	if (sg)
-		*sg = &per_cpu(sched_group_core, group).sg;
-	return group;
+		*sg = *per_cpu_ptr(sdd->sg, cpu);
+
+	return cpu;
 }
-#endif /* CONFIG_SCHED_MC */
 
 /*
- * book sched-domains:
+ * build_sched_groups takes the cpumask we wish to span, and a pointer
+ * to a function which identifies what group(along with sched group) a CPU
+ * belongs to. The return value of group_fn must be a >= 0 and < nr_cpu_ids
+ * (due to the fact that we keep track of groups covered with a struct cpumask).
+ *
+ * build_sched_groups will build a circular linked list of the groups
+ * covered by the given span, and will set each group's ->cpumask correctly,
+ * and ->cpu_power to 0.
  */
-#ifdef CONFIG_SCHED_BOOK
-static DEFINE_PER_CPU(struct static_sched_domain, book_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_group_book);
-
-static int
-cpu_to_book_group(int cpu, const struct cpumask *cpu_map,
-		  struct sched_group **sg, struct cpumask *mask)
-{
-	int group = cpu;
-#ifdef CONFIG_SCHED_MC
-	cpumask_and(mask, cpu_coregroup_mask(cpu), cpu_map);
-	group = cpumask_first(mask);
-#elif defined(CONFIG_SCHED_SMT)
-	cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map);
-	group = cpumask_first(mask);
-#endif
-	if (sg)
-		*sg = &per_cpu(sched_group_book, group).sg;
-	return group;
-}
-#endif /* CONFIG_SCHED_BOOK */
-
-static DEFINE_PER_CPU(struct static_sched_domain, phys_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_group_phys);
-
-static int
-cpu_to_phys_group(int cpu, const struct cpumask *cpu_map,
-		  struct sched_group **sg, struct cpumask *mask)
+static void
+build_sched_groups(struct sched_domain *sd, struct cpumask *covered)
 {
-	int group;
-#ifdef CONFIG_SCHED_BOOK
-	cpumask_and(mask, cpu_book_mask(cpu), cpu_map);
-	group = cpumask_first(mask);
-#elif defined(CONFIG_SCHED_MC)
-	cpumask_and(mask, cpu_coregroup_mask(cpu), cpu_map);
-	group = cpumask_first(mask);
-#elif defined(CONFIG_SCHED_SMT)
-	cpumask_and(mask, topology_thread_cpumask(cpu), cpu_map);
-	group = cpumask_first(mask);
-#else
-	group = cpu;
-#endif
-	if (sg)
-		*sg = &per_cpu(sched_group_phys, group).sg;
-	return group;
-}
-
-#ifdef CONFIG_NUMA
-static DEFINE_PER_CPU(struct static_sched_domain, node_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_group_node);
+	struct sched_group *first = NULL, *last = NULL;
+	struct sd_data *sdd = sd->private;
+	const struct cpumask *span = sched_domain_span(sd);
+	int i;
 
-static int cpu_to_node_group(int cpu, const struct cpumask *cpu_map,
-				 struct sched_group **sg,
-				 struct cpumask *nodemask)
-{
-	int group;
+	cpumask_clear(covered);
 
-	cpumask_and(nodemask, cpumask_of_node(cpu_to_node(cpu)), cpu_map);
-	group = cpumask_first(nodemask);
+	for_each_cpu(i, span) {
+		struct sched_group *sg;
+		int group = get_group(i, sdd, &sg);
+		int j;
 
-	if (sg)
-		*sg = &per_cpu(sched_group_node, group).sg;
-	return group;
-}
+		if (cpumask_test_cpu(i, covered))
+			continue;
 
-static DEFINE_PER_CPU(struct static_sched_domain, allnodes_domains);
-static DEFINE_PER_CPU(struct static_sched_group, sched_group_allnodes);
+		cpumask_clear(sched_group_cpus(sg));
+		sg->cpu_power = 0;
 
-static int cpu_to_allnodes_group(int cpu, const struct cpumask *cpu_map,
-				 struct sched_group **sg,
-				 struct cpumask *nodemask)
-{
-	int group;
+		for_each_cpu(j, span) {
+			if (get_group(j, sdd, NULL) != group)
+				continue;
 
-	cpumask_and(nodemask, cpumask_of_node(cpu_to_node(cpu)), cpu_map);
-	group = cpumask_first(nodemask);
+			cpumask_set_cpu(j, covered);
+			cpumask_set_cpu(j, sched_group_cpus(sg));
+		}
 
-	if (sg)
-		*sg = &per_cpu(sched_group_allnodes, group).sg;
-	return group;
+		if (!first)
+			first = sg;
+		if (last)
+			last->next = sg;
+		last = sg;
+	}
+	last->next = first;
 }
 
-#endif /* CONFIG_NUMA */
-
 /*
  * Initialize sched groups cpu_power.
  *
@@ -7039,15 +6955,15 @@ static void init_sched_groups_power(int cpu, struct sched_domain *sd)
 # define SD_INIT_NAME(sd, type)		do { } while (0)
 #endif
 
-#define	SD_INIT(sd, type)	sd_init_##type(sd)
-
-#define SD_INIT_FUNC(type)	\
-static noinline void sd_init_##type(struct sched_domain *sd)	\
-{								\
-	memset(sd, 0, sizeof(*sd));				\
-	*sd = SD_##type##_INIT;					\
-	sd->level = SD_LV_##type;				\
-	SD_INIT_NAME(sd, type);					\
+#define SD_INIT_FUNC(type)						       \
+static noinline struct sched_domain *sd_init_##type(struct s_data *d, int cpu) \
+{									       \
+	struct sched_domain *sd = *per_cpu_ptr(d->sdd[SD_LV_##type].sd, cpu);  \
+	*sd = SD_##type##_INIT;						       \
+	sd->level = SD_LV_##type;					       \
+	SD_INIT_NAME(sd, type);						       \
+	sd->private = &d->sdd[SD_LV_##type];				       \
+	return sd;							       \
 }
 
 SD_INIT_FUNC(CPU)
@@ -7103,13 +7019,22 @@ static void set_domain_attribute(struct sched_domain *sd,
 static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 				 const struct cpumask *cpu_map)
 {
+	int i, j;
+
 	switch (what) {
 	case sa_rootdomain:
-		free_rootdomain(d->rd); /* fall through */
+		free_rootdomain(&d->rd->rcu); /* fall through */
 	case sa_sd:
 		free_percpu(d->sd); /* fall through */
-	case sa_tmpmask:
-		free_cpumask_var(d->tmpmask); /* fall through */
+	case sa_sd_storage:
+		for (i = 0; i < SD_LV_MAX; i++) {
+			for_each_cpu(j, cpu_map) {
+				kfree(*per_cpu_ptr(d->sdd[i].sd, j));
+				kfree(*per_cpu_ptr(d->sdd[i].sg, j));
+			}
+			free_percpu(d->sdd[i].sd);
+			free_percpu(d->sdd[i].sg);
+		} /* fall through */
 	case sa_send_covered:
 		free_cpumask_var(d->send_covered); /* fall through */
 	case sa_nodemask:
@@ -7122,25 +7047,70 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 static enum s_alloc __visit_domain_allocation_hell(struct s_data *d,
 						   const struct cpumask *cpu_map)
 {
+	int i, j;
+
+	memset(d, 0, sizeof(*d));
+
 	if (!alloc_cpumask_var(&d->nodemask, GFP_KERNEL))
 		return sa_none;
 	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
 		return sa_nodemask;
-	if (!alloc_cpumask_var(&d->tmpmask, GFP_KERNEL))
-		return sa_send_covered;
-	d->sd = alloc_percpu(struct sched_domain *);
-	if (!d->sd) {
-		printk(KERN_WARNING "Cannot alloc per-cpu pointers\n");
-		return sa_tmpmask;
+	for (i = 0; i < SD_LV_MAX; i++) {
+		d->sdd[i].sd = alloc_percpu(struct sched_domain *);
+		if (!d->sdd[i].sd)
+			return sa_sd_storage;
+
+		d->sdd[i].sg = alloc_percpu(struct sched_group *);
+		if (!d->sdd[i].sg)
+			return sa_sd_storage;
+
+		for_each_cpu(j, cpu_map) {
+			struct sched_domain *sd;
+			struct sched_group *sg;
+
+		       	sd = kzalloc_node(sizeof(struct sched_domain) + cpumask_size(),
+					GFP_KERNEL, cpu_to_node(j));
+			if (!sd)
+				return sa_sd_storage;
+
+			*per_cpu_ptr(d->sdd[i].sd, j) = sd;
+
+			sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(),
+					GFP_KERNEL, cpu_to_node(j));
+			if (!sg)
+				return sa_sd_storage;
+
+			*per_cpu_ptr(d->sdd[i].sg, j) = sg;
+		}
 	}
+	d->sd = alloc_percpu(struct sched_domain *);
+	if (!d->sd)
+		return sa_sd_storage;
 	d->rd = alloc_rootdomain();
-	if (!d->rd) {
-		printk(KERN_WARNING "Cannot alloc root domain\n");
+	if (!d->rd)
 		return sa_sd;
-	}
 	return sa_rootdomain;
 }
 
+/*
+ * NULL the sd_data elements we've used to build the sched_domain and
+ * sched_group structure so that the subsequent __free_domain_allocs()
+ * will not free the data we're using.
+ */
+static void claim_allocations(int cpu, struct sched_domain *sd)
+{
+	struct sd_data *sdd = sd->private;
+	struct sched_group *sg = sd->groups;
+
+	WARN_ON_ONCE(*per_cpu_ptr(sdd->sd, cpu) != sd);
+	*per_cpu_ptr(sdd->sd, cpu) = NULL;
+
+	if (cpu == cpumask_first(sched_group_cpus(sg))) {
+		WARN_ON_ONCE(*per_cpu_ptr(sdd->sg, cpu) != sg);
+		*per_cpu_ptr(sdd->sg, cpu) = NULL;
+	}
+}
+
 static struct sched_domain *__build_numa_sched_domains(struct s_data *d,
 	const struct cpumask *cpu_map, struct sched_domain_attr *attr, int i)
 {
@@ -7151,24 +7121,20 @@ static struct sched_domain *__build_numa_sched_domains(struct s_data *d,
 	d->sd_allnodes = 0;
 	if (cpumask_weight(cpu_map) >
 	    SD_NODES_PER_DOMAIN * cpumask_weight(d->nodemask)) {
-		sd = &per_cpu(allnodes_domains, i).sd;
-		SD_INIT(sd, ALLNODES);
+		sd = sd_init_ALLNODES(d, i);
 		set_domain_attribute(sd, attr);
 		cpumask_copy(sched_domain_span(sd), cpu_map);
-		cpu_to_allnodes_group(i, cpu_map, &sd->groups, d->tmpmask);
 		d->sd_allnodes = 1;
 	}
 	parent = sd;
 
-	sd = &per_cpu(node_domains, i).sd;
-	SD_INIT(sd, NODE);
+	sd = sd_init_NODE(d, i);
 	set_domain_attribute(sd, attr);
 	sched_domain_node_span(cpu_to_node(i), sched_domain_span(sd));
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;
 	cpumask_and(sched_domain_span(sd), sched_domain_span(sd), cpu_map);
-	cpu_to_node_group(i, cpu_map, &sd->groups, d->tmpmask);
 #endif
 	return sd;
 }
@@ -7178,14 +7144,12 @@ static struct sched_domain *__build_cpu_sched_domain(struct s_data *d,
 	struct sched_domain *parent, int i)
 {
 	struct sched_domain *sd;
-	sd = &per_cpu(phys_domains, i).sd;
-	SD_INIT(sd, CPU);
+	sd = sd_init_CPU(d, i);
 	set_domain_attribute(sd, attr);
 	cpumask_copy(sched_domain_span(sd), d->nodemask);
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;
-	cpu_to_phys_group(i, cpu_map, &sd->groups, d->tmpmask);
 	return sd;
 }
 
@@ -7195,13 +7159,11 @@ static struct sched_domain *__build_book_sched_domain(struct s_data *d,
 {
 	struct sched_domain *sd = parent;
 #ifdef CONFIG_SCHED_BOOK
-	sd = &per_cpu(book_domains, i).sd;
-	SD_INIT(sd, BOOK);
+	sd = sd_init_BOOK(d, i);
 	set_domain_attribute(sd, attr);
 	cpumask_and(sched_domain_span(sd), cpu_map, cpu_book_mask(i));
 	sd->parent = parent;
 	parent->child = sd;
-	cpu_to_book_group(i, cpu_map, &sd->groups, d->tmpmask);
 #endif
 	return sd;
 }
@@ -7212,13 +7174,11 @@ static struct sched_domain *__build_mc_sched_domain(struct s_data *d,
 {
 	struct sched_domain *sd = parent;
 #ifdef CONFIG_SCHED_MC
-	sd = &per_cpu(core_domains, i).sd;
-	SD_INIT(sd, MC);
+	sd = sd_init_MC(d, i);
 	set_domain_attribute(sd, attr);
 	cpumask_and(sched_domain_span(sd), cpu_map, cpu_coregroup_mask(i));
 	sd->parent = parent;
 	parent->child = sd;
-	cpu_to_core_group(i, cpu_map, &sd->groups, d->tmpmask);
 #endif
 	return sd;
 }
@@ -7229,92 +7189,32 @@ static struct sched_domain *__build_smt_sched_domain(struct s_data *d,
 {
 	struct sched_domain *sd = parent;
 #ifdef CONFIG_SCHED_SMT
-	sd = &per_cpu(cpu_domains, i).sd;
-	SD_INIT(sd, SIBLING);
+	sd = sd_init_SIBLING(d, i);
 	set_domain_attribute(sd, attr);
 	cpumask_and(sched_domain_span(sd), cpu_map, topology_thread_cpumask(i));
 	sd->parent = parent;
 	parent->child = sd;
-	cpu_to_cpu_group(i, cpu_map, &sd->groups, d->tmpmask);
 #endif
 	return sd;
 }
 
-static void build_sched_groups(struct s_data *d, struct sched_domain *sd,
-			       const struct cpumask *cpu_map, int cpu)
-{
-	switch (sd->level) {
-#ifdef CONFIG_SCHED_SMT
-	case SD_LV_SIBLING: /* set up CPU (sibling) groups */
-		if (cpu == cpumask_first(sched_domain_span(sd)))
-			init_sched_build_groups(sched_domain_span(sd), cpu_map,
-						&cpu_to_cpu_group,
-						d->send_covered, d->tmpmask);
-		break;
-#endif
-#ifdef CONFIG_SCHED_MC
-	case SD_LV_MC: /* set up multi-core groups */
-		if (cpu == cpumask_first(sched_domain_span(sd)))
-			init_sched_build_groups(sched_domain_span(sd), cpu_map,
-						&cpu_to_core_group,
-						d->send_covered, d->tmpmask);
-		break;
-#endif
-#ifdef CONFIG_SCHED_BOOK
-	case SD_LV_BOOK: /* set up book groups */
-		if (cpu == cpumask_first(sched_domain_span(sd)))
-			init_sched_build_groups(sched_domain_span(sd), cpu_map,
-						&cpu_to_book_group,
-						d->send_covered, d->tmpmask);
-		break;
-#endif
-	case SD_LV_CPU: /* set up physical groups */
-		if (cpu == cpumask_first(sched_domain_span(sd)))
-			init_sched_build_groups(sched_domain_span(sd), cpu_map,
-						&cpu_to_phys_group,
-						d->send_covered, d->tmpmask);
-		break;
-#ifdef CONFIG_NUMA
-	case SD_LV_NODE:
-		if (cpu == cpumask_first(sched_domain_span(sd)))
-			init_sched_build_groups(sched_domain_span(sd), cpu_map,
-						&cpu_to_node_group,
-						d->send_covered, d->tmpmask);
-
-	case SD_LV_ALLNODES:
-		if (cpu == cpumask_first(cpu_map))
-			init_sched_build_groups(cpu_map, cpu_map,
-					&cpu_to_allnodes_group,
-					d->send_covered, d->tmpmask);
-		break;
-#endif
-	default:
-		break;
-	}
-}
-
 /*
  * Build sched domains for a given set of cpus and attach the sched domains
  * to the individual cpus
  */
-static int __build_sched_domains(const struct cpumask *cpu_map,
-				 struct sched_domain_attr *attr)
+static int build_sched_domains(const struct cpumask *cpu_map,
+			       struct sched_domain_attr *attr)
 {
 	enum s_alloc alloc_state = sa_none;
+	struct sched_domain *sd;
 	struct s_data d;
-	struct sched_domain *sd, *tmp;
 	int i;
-#ifdef CONFIG_NUMA
-	d.sd_allnodes = 0;
-#endif
 
 	alloc_state = __visit_domain_allocation_hell(&d, cpu_map);
 	if (alloc_state != sa_rootdomain)
 		goto error;
 
-	/*
-	 * Set up domains for cpus specified by the cpu_map.
-	 */
+	/* Set up domains for cpus specified by the cpu_map. */
 	for_each_cpu(i, cpu_map) {
 		cpumask_and(d.nodemask, cpumask_of_node(cpu_to_node(i)),
 			    cpu_map);
@@ -7326,10 +7226,19 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 		sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i);
 
 		*per_cpu_ptr(d.sd, i) = sd;
+	}
+
+	/* Build the groups for the domains */
+	for_each_cpu(i, cpu_map) {
+		for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) {
+			sd->span_weight = cpumask_weight(sched_domain_span(sd));
+			get_group(i, sd->private, &sd->groups);
+			atomic_inc(&sd->groups->ref);
 
-		for (tmp = sd; tmp; tmp = tmp->parent) {
-			tmp->span_weight = cpumask_weight(sched_domain_span(tmp));
-			build_sched_groups(&d, tmp, cpu_map, i);
+			if (i != cpumask_first(sched_domain_span(sd)))
+				continue;
+
+			build_sched_groups(sd, d.send_covered);
 		}
 	}
 
@@ -7338,18 +7247,21 @@ static int __build_sched_domains(const struct cpumask *cpu_map,
 		if (!cpumask_test_cpu(i, cpu_map))
 			continue;
 
-		sd = *per_cpu_ptr(d.sd, i);
-		for (; sd; sd = sd->parent)
+		for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) {
+			claim_allocations(i, sd);
 			init_sched_groups_power(i, sd);
+		}
 	}
 
 	/* Attach the domains */
+	rcu_read_lock();
 	for_each_cpu(i, cpu_map) {
 		sd = *per_cpu_ptr(d.sd, i);
 		cpu_attach_domain(sd, d.rd, i);
 	}
+	rcu_read_unlock();
 
-	__free_domain_allocs(&d, sa_tmpmask, cpu_map);
+	__free_domain_allocs(&d, sa_sd, cpu_map);
 	return 0;
 
 error:
@@ -7357,11 +7269,6 @@ error:
 	return -ENOMEM;
 }
 
-static int build_sched_domains(const struct cpumask *cpu_map)
-{
-	return __build_sched_domains(cpu_map, NULL);
-}
-
 static cpumask_var_t *doms_cur;	/* current sched domains */
 static int ndoms_cur;		/* number of sched domains in 'doms_cur' */
 static struct sched_domain_attr *dattr_cur;
@@ -7425,31 +7332,24 @@ static int init_sched_domains(const struct cpumask *cpu_map)
 		doms_cur = &fallback_doms;
 	cpumask_andnot(doms_cur[0], cpu_map, cpu_isolated_map);
 	dattr_cur = NULL;
-	err = build_sched_domains(doms_cur[0]);
+	err = build_sched_domains(doms_cur[0], NULL);
 	register_sched_domain_sysctl();
 
 	return err;
 }
 
-static void destroy_sched_domains(const struct cpumask *cpu_map,
-				       struct cpumask *tmpmask)
-{
-}
-
 /*
  * Detach sched domains from a group of cpus specified in cpu_map
  * These cpus will now be attached to the NULL domain
  */
 static void detach_destroy_domains(const struct cpumask *cpu_map)
 {
-	/* Save because hotplug lock held. */
-	static DECLARE_BITMAP(tmpmask, CONFIG_NR_CPUS);
 	int i;
 
+	rcu_read_lock();
 	for_each_cpu(i, cpu_map)
 		cpu_attach_domain(NULL, &def_root_domain, i);
-	synchronize_sched();
-	destroy_sched_domains(cpu_map, to_cpumask(tmpmask));
+	rcu_read_unlock();
 }
 
 /* handle null as "default" */
@@ -7538,8 +7438,7 @@ match1:
 				goto match2;
 		}
 		/* no match - add a new doms_new */
-		__build_sched_domains(doms_new[i],
-					dattr_new ? dattr_new + i : NULL);
+		build_sched_domains(doms_new[i], dattr_new ? dattr_new + i : NULL);
 match2:
 		;
 	}
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 4ee50f0..4a8ac7c 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1622,6 +1622,7 @@ static int select_idle_sibling(struct task_struct *p, int target)
 	/*
 	 * Otherwise, iterate the domains and find an elegible idle cpu.
 	 */
+	rcu_read_lock();
 	for_each_domain(target, sd) {
 		if (!(sd->flags & SD_SHARE_PKG_RESOURCES))
 			break;
@@ -1641,6 +1642,7 @@ static int select_idle_sibling(struct task_struct *p, int target)
 		    cpumask_test_cpu(prev_cpu, sched_domain_span(sd)))
 			break;
 	}
+	rcu_read_unlock();
 
 	return target;
 }
@@ -1673,6 +1675,7 @@ select_task_rq_fair(struct rq *rq, struct task_struct *p, int sd_flag, int wake_
 		new_cpu = prev_cpu;
 	}
 
+	rcu_read_lock();
 	for_each_domain(cpu, tmp) {
 		if (!(tmp->flags & SD_LOAD_BALANCE))
 			continue;
@@ -1723,9 +1726,10 @@ select_task_rq_fair(struct rq *rq, struct task_struct *p, int sd_flag, int wake_
 
 	if (affine_sd) {
 		if (cpu == prev_cpu || wake_affine(affine_sd, p, sync))
-			return select_idle_sibling(p, cpu);
-		else
-			return select_idle_sibling(p, prev_cpu);
+			prev_cpu = cpu;
+
+		new_cpu = select_idle_sibling(p, prev_cpu);
+		goto unlock;
 	}
 
 	while (sd) {
@@ -1766,6 +1770,8 @@ select_task_rq_fair(struct rq *rq, struct task_struct *p, int sd_flag, int wake_
 		}
 		/* while loop will break here if sd == NULL */
 	}
+unlock:
+	rcu_read_unlock();
 
 	return new_cpu;
 }
@@ -3462,6 +3468,7 @@ static void idle_balance(int this_cpu, struct rq *this_rq)
 	raw_spin_unlock(&this_rq->lock);
 
 	update_shares(this_cpu);
+	rcu_read_lock();
 	for_each_domain(this_cpu, sd) {
 		unsigned long interval;
 		int balance = 1;
@@ -3483,6 +3490,7 @@ static void idle_balance(int this_cpu, struct rq *this_rq)
 			break;
 		}
 	}
+	rcu_read_unlock();
 
 	raw_spin_lock(&this_rq->lock);
 
@@ -3531,6 +3539,7 @@ static int active_load_balance_cpu_stop(void *data)
 	double_lock_balance(busiest_rq, target_rq);
 
 	/* Search for an sd spanning us and the target CPU. */
+	rcu_read_lock();
 	for_each_domain(target_cpu, sd) {
 		if ((sd->flags & SD_LOAD_BALANCE) &&
 		    cpumask_test_cpu(busiest_cpu, sched_domain_span(sd)))
@@ -3546,6 +3555,7 @@ static int active_load_balance_cpu_stop(void *data)
 		else
 			schedstat_inc(sd, alb_failed);
 	}
+	rcu_read_unlock();
 	double_unlock_balance(busiest_rq, target_rq);
 out_unlock:
 	busiest_rq->active_balance = 0;
@@ -3672,6 +3682,7 @@ static int find_new_ilb(int cpu)
 {
 	struct sched_domain *sd;
 	struct sched_group *ilb_group;
+	int ilb = nr_cpu_ids;
 
 	/*
 	 * Have idle load balancer selection from semi-idle packages only
@@ -3687,20 +3698,25 @@ static int find_new_ilb(int cpu)
 	if (cpumask_weight(nohz.idle_cpus_mask) < 2)
 		goto out_done;
 
+	rcu_read_lock();
 	for_each_flag_domain(cpu, sd, SD_POWERSAVINGS_BALANCE) {
 		ilb_group = sd->groups;
 
 		do {
-			if (is_semi_idle_group(ilb_group))
-				return cpumask_first(nohz.grp_idle_mask);
+			if (is_semi_idle_group(ilb_group)) {
+				ilb = cpumask_first(nohz.grp_idle_mask);
+				goto unlock;
+			}
 
 			ilb_group = ilb_group->next;
 
 		} while (ilb_group != sd->groups);
 	}
+unlock:
+	rcu_read_unlock();
 
 out_done:
-	return nr_cpu_ids;
+	return ilb;
 }
 #else /*  (CONFIG_SCHED_MC || CONFIG_SCHED_SMT) */
 static inline int find_new_ilb(int call_cpu)
@@ -3845,6 +3861,7 @@ static void rebalance_domains(int cpu, enum cpu_idle_type idle)
 
 	update_shares(cpu);
 
+	rcu_read_lock();
 	for_each_domain(cpu, sd) {
 		if (!(sd->flags & SD_LOAD_BALANCE))
 			continue;
@@ -3890,6 +3907,7 @@ out:
 		if (!balance)
 			break;
 	}
+	rcu_read_unlock();
 
 	/*
 	 * next_balance will be updated only when there is a need.

[PATCH 10/23] sched: Simplify the free path some

[Peter Zijlstra]

[-- Attachment #1: sched-foo9.patch --]
[-- Type: text/plain, Size: 1372 bytes --]

If we check the root_domain reference count we can see if its been
used or not, use this observation to simplify some of the return
paths.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -7009,7 +7009,8 @@ static void __free_domain_allocs(struct 
 
 	switch (what) {
 	case sa_rootdomain:
-		free_rootdomain(&d->rd->rcu); /* fall through */
+		if (!atomic_read(&d->rd->refcount))
+			free_rootdomain(&d->rd->rcu); /* fall through */
 	case sa_sd:
 		free_percpu(d->sd); /* fall through */
 	case sa_sd_storage:
@@ -7194,7 +7195,7 @@ static int build_sched_domains(const str
 	enum s_alloc alloc_state = sa_none;
 	struct sched_domain *sd;
 	struct s_data d;
-	int i;
+	int i, ret = -ENOMEM;
 
 	alloc_state = __visit_domain_allocation_hell(&d, cpu_map);
 	if (alloc_state != sa_rootdomain)
@@ -7248,12 +7249,10 @@ static int build_sched_domains(const str
 	}
 	rcu_read_unlock();
 
-	__free_domain_allocs(&d, sa_sd, cpu_map);
-	return 0;
-
+	ret = 0;
 error:
 	__free_domain_allocs(&d, alloc_state, cpu_map);
-	return -ENOMEM;
+	return ret;
 }
 
 static cpumask_var_t *doms_cur;	/* current sched domains */

[tip:sched/domains] sched: Simplify the free path some

[tip-bot for Peter Zijlstra]

Commit-ID:  822ff793c34a5d4c8b5f3f9ce932602233d96464
Gitweb:     http://git.kernel.org/tip/822ff793c34a5d4c8b5f3f9ce932602233d96464
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:51 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:20 +0200

sched: Simplify the free path some

If we check the root_domain reference count we can see if its been
used or not, use this observation to simplify some of the return
paths.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.298339503@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   11 +++++------
 1 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index 6520484..72c194c 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -7023,7 +7023,8 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 
 	switch (what) {
 	case sa_rootdomain:
-		free_rootdomain(&d->rd->rcu); /* fall through */
+		if (!atomic_read(&d->rd->refcount))
+			free_rootdomain(&d->rd->rcu); /* fall through */
 	case sa_sd:
 		free_percpu(d->sd); /* fall through */
 	case sa_sd_storage:
@@ -7208,7 +7209,7 @@ static int build_sched_domains(const struct cpumask *cpu_map,
 	enum s_alloc alloc_state = sa_none;
 	struct sched_domain *sd;
 	struct s_data d;
-	int i;
+	int i, ret = -ENOMEM;
 
 	alloc_state = __visit_domain_allocation_hell(&d, cpu_map);
 	if (alloc_state != sa_rootdomain)
@@ -7261,12 +7262,10 @@ static int build_sched_domains(const struct cpumask *cpu_map,
 	}
 	rcu_read_unlock();
 
-	__free_domain_allocs(&d, sa_sd, cpu_map);
-	return 0;
-
+	ret = 0;
 error:
 	__free_domain_allocs(&d, alloc_state, cpu_map);
-	return -ENOMEM;
+	return ret;
 }
 
 static cpumask_var_t *doms_cur;	/* current sched domains */

[PATCH 11/23] sched: Avoid using sd->level

[Peter Zijlstra]

[-- Attachment #1: sched-foo14.patch --]
[-- Type: text/plain, Size: 618 bytes --]

Don't use sd->level for identifying properties of the domain.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched_fair.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Index: linux-2.6/kernel/sched_fair.c
===================================================================
--- linux-2.6.orig/kernel/sched_fair.c
+++ linux-2.6/kernel/sched_fair.c
@@ -2657,7 +2657,7 @@ fix_small_capacity(struct sched_domain *
 	/*
 	 * Only siblings can have significantly less than SCHED_LOAD_SCALE
 	 */
-	if (sd->level != SD_LV_SIBLING)
+	if (!(sd->flags & SD_SHARE_CPUPOWER))
 		return 0;
 
 	/*

[tip:sched/domains] sched: Avoid using sd->level

[tip-bot for Peter Zijlstra]

Commit-ID:  a6c75f2f8d988ecfecf971f98f1cb6fc4de522fe
Gitweb:     http://git.kernel.org/tip/a6c75f2f8d988ecfecf971f98f1cb6fc4de522fe
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:52 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:20 +0200

sched: Avoid using sd->level

Don't use sd->level for identifying properties of the domain.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.350174079@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched_fair.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 4a8ac7c..9c5679c 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -2651,7 +2651,7 @@ fix_small_capacity(struct sched_domain *sd, struct sched_group *group)
 	/*
 	 * Only siblings can have significantly less than SCHED_LOAD_SCALE
 	 */
-	if (sd->level != SD_LV_SIBLING)
+	if (!(sd->flags & SD_SHARE_CPUPOWER))
 		return 0;
 
 	/*

[PATCH 12/23] sched: Reduce some allocation pressure

[Peter Zijlstra]

[-- Attachment #1: sched-foo10.patch --]
[-- Type: text/plain, Size: 958 bytes --]

Since we now allocate SD_LV_MAX * nr_cpu_ids sched_domain/sched_group
structures when rebuilding the scheduler toplogy it might make sense
to shrink that depending on the CONFIG_ options.

This is only needed until we get rid of SD_LV_* alltogether and
provide a full dynamic topology interface.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 include/linux/sched.h |    8 ++++++++
 1 file changed, 8 insertions(+)

Index: linux-2.6/include/linux/sched.h
===================================================================
--- linux-2.6.orig/include/linux/sched.h
+++ linux-2.6/include/linux/sched.h
@@ -895,12 +895,20 @@ static inline struct cpumask *sched_grou
 
 enum sched_domain_level {
 	SD_LV_NONE = 0,
+#ifdef CONFIG_SCHED_SMT
 	SD_LV_SIBLING,
+#endif
+#ifdef CONFIG_SCHED_MC
 	SD_LV_MC,
+#endif
+#ifdef CONFIG_SCHED_BOOK
 	SD_LV_BOOK,
+#endif
 	SD_LV_CPU,
+#ifdef CONFIG_NUMA
 	SD_LV_NODE,
 	SD_LV_ALLNODES,
+#endif
 	SD_LV_MAX
 };
 

[tip:sched/domains] sched: Reduce some allocation pressure

[tip-bot for Peter Zijlstra]

Commit-ID:  3859173d43658d51a749bc0201b943922577d39c
Gitweb:     http://git.kernel.org/tip/3859173d43658d51a749bc0201b943922577d39c
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:53 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:21 +0200

sched: Reduce some allocation pressure

Since we now allocate SD_LV_MAX * nr_cpu_ids sched_domain/sched_group
structures when rebuilding the scheduler toplogy it might make sense
to shrink that depending on the CONFIG_ options.

This is only needed until we get rid of SD_LV_* alltogether and
provide a full dynamic topology interface.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.406226449@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |    8 ++++++++
 1 files changed, 8 insertions(+), 0 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 020b79d..5a9168b 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -897,12 +897,20 @@ static inline struct cpumask *sched_group_cpus(struct sched_group *sg)
 
 enum sched_domain_level {
 	SD_LV_NONE = 0,
+#ifdef CONFIG_SCHED_SMT
 	SD_LV_SIBLING,
+#endif
+#ifdef CONFIG_SCHED_MC
 	SD_LV_MC,
+#endif
+#ifdef CONFIG_SCHED_BOOK
 	SD_LV_BOOK,
+#endif
 	SD_LV_CPU,
+#ifdef CONFIG_NUMA
 	SD_LV_NODE,
 	SD_LV_ALLNODES,
+#endif
 	SD_LV_MAX
 };
 

[PATCH 13/23] sched: Simplify NODE/ALLNODES domain creation

[Peter Zijlstra]

[-- Attachment #1: sched-foo11.patch --]
[-- Type: text/plain, Size: 2806 bytes --]

Don't treat ALLNODES/NODE different for difference's sake. Simply
always create the ALLNODES domain and let the sd_degenerate() checks
kill it when its redundant. This simplifies the code flow.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   40 ++++++++++++++++++++++------------------
 1 file changed, 22 insertions(+), 18 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6814,9 +6814,6 @@ struct sd_data {
 };
 
 struct s_data {
-#ifdef CONFIG_NUMA
-	int			sd_allnodes;
-#endif
 	cpumask_var_t		nodemask;
 	cpumask_var_t		send_covered;
 	struct sched_domain ** __percpu sd;
@@ -7088,30 +7085,35 @@ static void claim_allocations(int cpu, s
 	}
 }
 
-static struct sched_domain *__build_numa_sched_domains(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr, int i)
+static struct sched_domain *__build_allnodes_sched_domain(struct s_data *d,
+	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
+	struct sched_domain *parent, int i)
 {
 	struct sched_domain *sd = NULL;
 #ifdef CONFIG_NUMA
-	struct sched_domain *parent;
-
-	d->sd_allnodes = 0;
-	if (cpumask_weight(cpu_map) >
-	    SD_NODES_PER_DOMAIN * cpumask_weight(d->nodemask)) {
-		sd = sd_init_ALLNODES(d, i);
-		set_domain_attribute(sd, attr);
-		cpumask_copy(sched_domain_span(sd), cpu_map);
-		d->sd_allnodes = 1;
-	}
-	parent = sd;
+	sd = sd_init_ALLNODES(d, i);
+	set_domain_attribute(sd, attr);
+	cpumask_copy(sched_domain_span(sd), cpu_map);
+	sd->parent = parent;
+	if (parent)
+		parent->child = sd;
+#endif
+	return sd;
+}
 
+static struct sched_domain *__build_node_sched_domain(struct s_data *d,
+	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
+	struct sched_domain *parent, int i)
+{
+	struct sched_domain *sd = NULL;
+#ifdef CONFIG_NUMA
 	sd = sd_init_NODE(d, i);
 	set_domain_attribute(sd, attr);
 	sched_domain_node_span(cpu_to_node(i), sched_domain_span(sd));
+	cpumask_and(sched_domain_span(sd), sched_domain_span(sd), cpu_map);
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;
-	cpumask_and(sched_domain_span(sd), sched_domain_span(sd), cpu_map);
 #endif
 	return sd;
 }
@@ -7196,7 +7198,9 @@ static int __build_sched_domains(const s
 		cpumask_and(d.nodemask, cpumask_of_node(cpu_to_node(i)),
 			    cpu_map);
 
-		sd = __build_numa_sched_domains(&d, cpu_map, attr, i);
+		sd = NULL;
+		sd = __build_allnodes_sched_domain(&d, cpu_map, attr, sd, i);
+		sd = __build_node_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_cpu_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_book_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i);

[tip:sched/domains] sched: Simplify NODE/ALLNODES domain creation

[tip-bot for Peter Zijlstra]

Commit-ID:  3bd65a80affb9768b91f03c56dba46ee79525f9b
Gitweb:     http://git.kernel.org/tip/3bd65a80affb9768b91f03c56dba46ee79525f9b
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:54 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:21 +0200

sched: Simplify NODE/ALLNODES domain creation

Don't treat ALLNODES/NODE different for difference's sake. Simply
always create the ALLNODES domain and let the sd_degenerate() checks
kill it when its redundant. This simplifies the code flow.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.455464579@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   40 ++++++++++++++++++++++------------------
 1 files changed, 22 insertions(+), 18 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index 72c194c..d395fe5 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6838,9 +6838,6 @@ struct sd_data {
 };
 
 struct s_data {
-#ifdef CONFIG_NUMA
-	int			sd_allnodes;
-#endif
 	cpumask_var_t		nodemask;
 	cpumask_var_t		send_covered;
 	struct sched_domain ** __percpu sd;
@@ -7112,30 +7109,35 @@ static void claim_allocations(int cpu, struct sched_domain *sd)
 	}
 }
 
-static struct sched_domain *__build_numa_sched_domains(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr, int i)
+static struct sched_domain *__build_allnodes_sched_domain(struct s_data *d,
+	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
+	struct sched_domain *parent, int i)
 {
 	struct sched_domain *sd = NULL;
 #ifdef CONFIG_NUMA
-	struct sched_domain *parent;
-
-	d->sd_allnodes = 0;
-	if (cpumask_weight(cpu_map) >
-	    SD_NODES_PER_DOMAIN * cpumask_weight(d->nodemask)) {
-		sd = sd_init_ALLNODES(d, i);
-		set_domain_attribute(sd, attr);
-		cpumask_copy(sched_domain_span(sd), cpu_map);
-		d->sd_allnodes = 1;
-	}
-	parent = sd;
+	sd = sd_init_ALLNODES(d, i);
+	set_domain_attribute(sd, attr);
+	cpumask_copy(sched_domain_span(sd), cpu_map);
+	sd->parent = parent;
+	if (parent)
+		parent->child = sd;
+#endif
+	return sd;
+}
 
+static struct sched_domain *__build_node_sched_domain(struct s_data *d,
+	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
+	struct sched_domain *parent, int i)
+{
+	struct sched_domain *sd = NULL;
+#ifdef CONFIG_NUMA
 	sd = sd_init_NODE(d, i);
 	set_domain_attribute(sd, attr);
 	sched_domain_node_span(cpu_to_node(i), sched_domain_span(sd));
+	cpumask_and(sched_domain_span(sd), sched_domain_span(sd), cpu_map);
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;
-	cpumask_and(sched_domain_span(sd), sched_domain_span(sd), cpu_map);
 #endif
 	return sd;
 }
@@ -7220,7 +7222,9 @@ static int build_sched_domains(const struct cpumask *cpu_map,
 		cpumask_and(d.nodemask, cpumask_of_node(cpu_to_node(i)),
 			    cpu_map);
 
-		sd = __build_numa_sched_domains(&d, cpu_map, attr, i);
+		sd = NULL;
+		sd = __build_allnodes_sched_domain(&d, cpu_map, attr, sd, i);
+		sd = __build_node_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_cpu_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_book_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i);

[PATCH 14/23] sched: Remove nodemask allocation

[Peter Zijlstra]

[-- Attachment #1: sched-foo12.patch --]
[-- Type: text/plain, Size: 2122 bytes --]

There's only one nodemask user left so remove it with a direct
computation and save some memory and reduce some code-flow
complexity.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   14 +++-----------
 1 file changed, 3 insertions(+), 11 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6814,7 +6814,6 @@ struct sd_data {
 };
 
 struct s_data {
-	cpumask_var_t		nodemask;
 	cpumask_var_t		send_covered;
 	struct sched_domain ** __percpu sd;
 	struct sd_data 		sdd[SD_LV_MAX];
@@ -6826,7 +6825,6 @@ enum s_alloc {
 	sa_sd,
 	sa_sd_storage,
 	sa_send_covered,
-	sa_nodemask,
 	sa_none,
 };
 
@@ -7011,8 +7009,6 @@ static void __free_domain_allocs(struct 
 		} /* fall through */
 	case sa_send_covered:
 		free_cpumask_var(d->send_covered); /* fall through */
-	case sa_nodemask:
-		free_cpumask_var(d->nodemask); /* fall through */
 	case sa_none:
 		break;
 	}
@@ -7025,10 +7021,8 @@ static enum s_alloc __visit_domain_alloc
 
 	memset(d, 0, sizeof(*d));
 
-	if (!alloc_cpumask_var(&d->nodemask, GFP_KERNEL))
-		return sa_none;
 	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
-		return sa_nodemask;
+		return sa_none;
 	for (i = 0; i < SD_LV_MAX; i++) {
 		d->sdd[i].sd = alloc_percpu(struct sched_domain *);
 		if (!d->sdd[i].sd)
@@ -7125,7 +7119,8 @@ static struct sched_domain *__build_cpu_
 	struct sched_domain *sd;
 	sd = sd_init_CPU(d, i);
 	set_domain_attribute(sd, attr);
-	cpumask_copy(sched_domain_span(sd), d->nodemask);
+	cpumask_and(sched_domain_span(sd),
+			cpumask_of_node(cpu_to_node(i)), cpu_map);
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;
@@ -7195,9 +7190,6 @@ static int __build_sched_domains(const s
 
 	/* Set up domains for cpus specified by the cpu_map. */
 	for_each_cpu(i, cpu_map) {
-		cpumask_and(d.nodemask, cpumask_of_node(cpu_to_node(i)),
-			    cpu_map);
-
 		sd = NULL;
 		sd = __build_allnodes_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_node_sched_domain(&d, cpu_map, attr, sd, i);

[tip:sched/domains] sched: Remove nodemask allocation

[tip-bot for Peter Zijlstra]

Commit-ID:  bf28b253266ebd73c331dde24d64606afde32ceb
Gitweb:     http://git.kernel.org/tip/bf28b253266ebd73c331dde24d64606afde32ceb
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:55 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:22 +0200

sched: Remove nodemask allocation

There's only one nodemask user left so remove it with a direct
computation and save some memory and reduce some code-flow
complexity.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.505608966@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   14 +++-----------
 1 files changed, 3 insertions(+), 11 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index d395fe5..f4d3a62 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6838,7 +6838,6 @@ struct sd_data {
 };
 
 struct s_data {
-	cpumask_var_t		nodemask;
 	cpumask_var_t		send_covered;
 	struct sched_domain ** __percpu sd;
 	struct sd_data 		sdd[SD_LV_MAX];
@@ -6850,7 +6849,6 @@ enum s_alloc {
 	sa_sd,
 	sa_sd_storage,
 	sa_send_covered,
-	sa_nodemask,
 	sa_none,
 };
 
@@ -7035,8 +7033,6 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 		} /* fall through */
 	case sa_send_covered:
 		free_cpumask_var(d->send_covered); /* fall through */
-	case sa_nodemask:
-		free_cpumask_var(d->nodemask); /* fall through */
 	case sa_none:
 		break;
 	}
@@ -7049,10 +7045,8 @@ static enum s_alloc __visit_domain_allocation_hell(struct s_data *d,
 
 	memset(d, 0, sizeof(*d));
 
-	if (!alloc_cpumask_var(&d->nodemask, GFP_KERNEL))
-		return sa_none;
 	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
-		return sa_nodemask;
+		return sa_none;
 	for (i = 0; i < SD_LV_MAX; i++) {
 		d->sdd[i].sd = alloc_percpu(struct sched_domain *);
 		if (!d->sdd[i].sd)
@@ -7149,7 +7143,8 @@ static struct sched_domain *__build_cpu_sched_domain(struct s_data *d,
 	struct sched_domain *sd;
 	sd = sd_init_CPU(d, i);
 	set_domain_attribute(sd, attr);
-	cpumask_copy(sched_domain_span(sd), d->nodemask);
+	cpumask_and(sched_domain_span(sd),
+			cpumask_of_node(cpu_to_node(i)), cpu_map);
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;
@@ -7219,9 +7214,6 @@ static int build_sched_domains(const struct cpumask *cpu_map,
 
 	/* Set up domains for cpus specified by the cpu_map. */
 	for_each_cpu(i, cpu_map) {
-		cpumask_and(d.nodemask, cpumask_of_node(cpu_to_node(i)),
-			    cpu_map);
-
 		sd = NULL;
 		sd = __build_allnodes_sched_domain(&d, cpu_map, attr, sd, i);
 		sd = __build_node_sched_domain(&d, cpu_map, attr, sd, i);

[PATCH 15/23] sched: Remove some dead code

[Peter Zijlstra]

[-- Attachment #1: sched-foo13.patch --]
[-- Type: text/plain, Size: 1934 bytes --]


Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 include/linux/sched.h |    6 ------
 kernel/sched.c        |   16 ----------------
 2 files changed, 22 deletions(-)

Index: linux-2.6/include/linux/sched.h
===================================================================
--- linux-2.6.orig/include/linux/sched.h
+++ linux-2.6/include/linux/sched.h
@@ -881,9 +881,6 @@ struct sched_group {
 	 * NOTE: this field is variable length. (Allocated dynamically
 	 * by attaching extra space to the end of the structure,
 	 * depending on how many CPUs the kernel has booted up with)
-	 *
-	 * It is also be embedded into static data structures at build
-	 * time. (See 'struct static_sched_group' in kernel/sched.c)
 	 */
 	unsigned long cpumask[0];
 };
@@ -992,9 +989,6 @@ struct sched_domain {
 	 * NOTE: this field is variable length. (Allocated dynamically
 	 * by attaching extra space to the end of the structure,
 	 * depending on how many CPUs the kernel has booted up with)
-	 *
-	 * It is also be embedded into static data structures at build
-	 * time. (See 'struct static_sched_domain' in kernel/sched.c)
 	 */
 	unsigned long span[0];
 };
Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6793,22 +6793,6 @@ static void sched_domain_node_span(int n
 
 int sched_smt_power_savings = 0, sched_mc_power_savings = 0;
 
-/*
- * The cpus mask in sched_group and sched_domain hangs off the end.
- *
- * ( See the the comments in include/linux/sched.h:struct sched_group
- *   and struct sched_domain. )
- */
-struct static_sched_group {
-	struct sched_group sg;
-	DECLARE_BITMAP(cpus, CONFIG_NR_CPUS);
-};
-
-struct static_sched_domain {
-	struct sched_domain sd;
-	DECLARE_BITMAP(span, CONFIG_NR_CPUS);
-};
-
 struct sd_data {
 	struct sched_domain **__percpu sd;
 	struct sched_group **__percpu sg;

[tip:sched/domains] sched: Remove some dead code

[tip-bot for Peter Zijlstra]

Commit-ID:  7dd04b730749f957c116f363524fd622b05e5141
Gitweb:     http://git.kernel.org/tip/7dd04b730749f957c116f363524fd622b05e5141
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:56 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:22 +0200

sched: Remove some dead code

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.553814623@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |    6 ------
 kernel/sched.c        |   16 ----------------
 2 files changed, 0 insertions(+), 22 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 5a9168b..09d9e02 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -883,9 +883,6 @@ struct sched_group {
 	 * NOTE: this field is variable length. (Allocated dynamically
 	 * by attaching extra space to the end of the structure,
 	 * depending on how many CPUs the kernel has booted up with)
-	 *
-	 * It is also be embedded into static data structures at build
-	 * time. (See 'struct static_sched_group' in kernel/sched.c)
 	 */
 	unsigned long cpumask[0];
 };
@@ -994,9 +991,6 @@ struct sched_domain {
 	 * NOTE: this field is variable length. (Allocated dynamically
 	 * by attaching extra space to the end of the structure,
 	 * depending on how many CPUs the kernel has booted up with)
-	 *
-	 * It is also be embedded into static data structures at build
-	 * time. (See 'struct static_sched_domain' in kernel/sched.c)
 	 */
 	unsigned long span[0];
 };
diff --git a/kernel/sched.c b/kernel/sched.c
index f4d3a62..5ec685c 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6816,22 +6816,6 @@ static void sched_domain_node_span(int node, struct cpumask *span)
 
 int sched_smt_power_savings = 0, sched_mc_power_savings = 0;
 
-/*
- * The cpus mask in sched_group and sched_domain hangs off the end.
- *
- * ( See the the comments in include/linux/sched.h:struct sched_group
- *   and struct sched_domain. )
- */
-struct static_sched_group {
-	struct sched_group sg;
-	DECLARE_BITMAP(cpus, CONFIG_NR_CPUS);
-};
-
-struct static_sched_domain {
-	struct sched_domain sd;
-	DECLARE_BITMAP(span, CONFIG_NR_CPUS);
-};
-
 struct sd_data {
 	struct sched_domain **__percpu sd;
 	struct sched_group **__percpu sg;

[PATCH 16/23] sched: Create persistent sched_domains_tmpmask

[Peter Zijlstra]

[-- Attachment #1: sched-foo15.patch --]
[-- Type: text/plain, Size: 2746 bytes --]

Since sched domain creation is fully serialized by the
sched_domains_mutex we can create a single persistent tmpmask to use
during domain creation.

This removes the need for s_data::send_covered.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6791,7 +6791,6 @@ struct sd_data {
 };
 
 struct s_data {
-	cpumask_var_t		send_covered;
 	struct sched_domain ** __percpu sd;
 	struct sd_data 		sdd[SD_LV_MAX];
 	struct root_domain	*rd;
@@ -6801,7 +6800,6 @@ enum s_alloc {
 	sa_rootdomain,
 	sa_sd,
 	sa_sd_storage,
-	sa_send_covered,
 	sa_none,
 };
 
@@ -6822,6 +6820,8 @@ static int get_group(int cpu, struct sd_
 	return cpu;
 }
 
+static cpumask_var_t sched_domains_tmpmask; /* sched_domains_mutex */
+
 /*
  * build_sched_groups takes the cpumask we wish to span, and a pointer
  * to a function which identifies what group(along with sched group) a CPU
@@ -6833,13 +6833,17 @@ static int get_group(int cpu, struct sd_
  * and ->cpu_power to 0.
  */
 static void
-build_sched_groups(struct sched_domain *sd, struct cpumask *covered)
+build_sched_groups(struct sched_domain *sd)
 {
 	struct sched_group *first = NULL, *last = NULL;
 	struct sd_data *sdd = sd->private;
 	const struct cpumask *span = sched_domain_span(sd);
+	struct cpumask *covered;
 	int i;
 
+	lockdep_assert_held(&sched_domains_mutex);
+	covered = sched_domains_tmpmask;
+
 	cpumask_clear(covered);
 
 	for_each_cpu(i, span) {
@@ -6984,8 +6988,6 @@ static void __free_domain_allocs(struct 
 			free_percpu(d->sdd[i].sd);
 			free_percpu(d->sdd[i].sg);
 		} /* fall through */
-	case sa_send_covered:
-		free_cpumask_var(d->send_covered); /* fall through */
 	case sa_none:
 		break;
 	}
@@ -6998,8 +7000,6 @@ static enum s_alloc __visit_domain_alloc
 
 	memset(d, 0, sizeof(*d));
 
-	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
-		return sa_none;
 	for (i = 0; i < SD_LV_MAX; i++) {
 		d->sdd[i].sd = alloc_percpu(struct sched_domain *);
 		if (!d->sdd[i].sd)
@@ -7188,7 +7188,7 @@ static int __build_sched_domains(const s
 			if (i != cpumask_first(sched_domain_span(sd)))
 				continue;
 
-			build_sched_groups(sd, d.send_covered);
+			build_sched_groups(sd);
 		}
 	}
 
@@ -7870,6 +7870,7 @@ void __init sched_init(void)
 
 	/* Allocate the nohz_cpu_mask if CONFIG_CPUMASK_OFFSTACK */
 	zalloc_cpumask_var(&nohz_cpu_mask, GFP_NOWAIT);
+	zalloc_cpumask_var(&sched_domains_tmpmask, GFP_NOWAIT);
 #ifdef CONFIG_SMP
 #ifdef CONFIG_NO_HZ
 	zalloc_cpumask_var(&nohz.idle_cpus_mask, GFP_NOWAIT);

[tip:sched/domains] sched: Create persistent sched_domains_tmpmask

[tip-bot for Peter Zijlstra]

Commit-ID:  f96225fd51893b6650cffd5427f13f6b1b356488
Gitweb:     http://git.kernel.org/tip/f96225fd51893b6650cffd5427f13f6b1b356488
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:57 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 12:58:23 +0200

sched: Create persistent sched_domains_tmpmask

Since sched domain creation is fully serialized by the
sched_domains_mutex we can create a single persistent tmpmask to use
during domain creation.

This removes the need for s_data::send_covered.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.607287405@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   17 +++++++++--------
 1 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index 5ec685c..fd73e91 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6822,7 +6822,6 @@ struct sd_data {
 };
 
 struct s_data {
-	cpumask_var_t		send_covered;
 	struct sched_domain ** __percpu sd;
 	struct sd_data 		sdd[SD_LV_MAX];
 	struct root_domain	*rd;
@@ -6832,7 +6831,6 @@ enum s_alloc {
 	sa_rootdomain,
 	sa_sd,
 	sa_sd_storage,
-	sa_send_covered,
 	sa_none,
 };
 
@@ -6853,6 +6851,8 @@ static int get_group(int cpu, struct sd_data *sdd, struct sched_group **sg)
 	return cpu;
 }
 
+static cpumask_var_t sched_domains_tmpmask; /* sched_domains_mutex */
+
 /*
  * build_sched_groups takes the cpumask we wish to span, and a pointer
  * to a function which identifies what group(along with sched group) a CPU
@@ -6864,13 +6864,17 @@ static int get_group(int cpu, struct sd_data *sdd, struct sched_group **sg)
  * and ->cpu_power to 0.
  */
 static void
-build_sched_groups(struct sched_domain *sd, struct cpumask *covered)
+build_sched_groups(struct sched_domain *sd)
 {
 	struct sched_group *first = NULL, *last = NULL;
 	struct sd_data *sdd = sd->private;
 	const struct cpumask *span = sched_domain_span(sd);
+	struct cpumask *covered;
 	int i;
 
+	lockdep_assert_held(&sched_domains_mutex);
+	covered = sched_domains_tmpmask;
+
 	cpumask_clear(covered);
 
 	for_each_cpu(i, span) {
@@ -7015,8 +7019,6 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 			free_percpu(d->sdd[i].sd);
 			free_percpu(d->sdd[i].sg);
 		} /* fall through */
-	case sa_send_covered:
-		free_cpumask_var(d->send_covered); /* fall through */
 	case sa_none:
 		break;
 	}
@@ -7029,8 +7031,6 @@ static enum s_alloc __visit_domain_allocation_hell(struct s_data *d,
 
 	memset(d, 0, sizeof(*d));
 
-	if (!alloc_cpumask_var(&d->send_covered, GFP_KERNEL))
-		return sa_none;
 	for (i = 0; i < SD_LV_MAX; i++) {
 		d->sdd[i].sd = alloc_percpu(struct sched_domain *);
 		if (!d->sdd[i].sd)
@@ -7219,7 +7219,7 @@ static int build_sched_domains(const struct cpumask *cpu_map,
 			if (i != cpumask_first(sched_domain_span(sd)))
 				continue;
 
-			build_sched_groups(sd, d.send_covered);
+			build_sched_groups(sd);
 		}
 	}
 
@@ -7896,6 +7896,7 @@ void __init sched_init(void)
 
 	/* Allocate the nohz_cpu_mask if CONFIG_CPUMASK_OFFSTACK */
 	zalloc_cpumask_var(&nohz_cpu_mask, GFP_NOWAIT);
+	zalloc_cpumask_var(&sched_domains_tmpmask, GFP_NOWAIT);
 #ifdef CONFIG_SMP
 #ifdef CONFIG_NO_HZ
 	zalloc_cpumask_var(&nohz.idle_cpus_mask, GFP_NOWAIT);

[PATCH 17/23] sched: Avoid allocations in sched_domain_debug()

[Peter Zijlstra]

[-- Attachment #1: sched-foo23.patch --]
[-- Type: text/plain, Size: 2356 bytes --]

Since we're all serialized by sched_domains_mutex we can use
sched_domains_tmpmask and avoid having to do allocations. This means
we can use sched_domains_debug() for cpu_attach_domain() again.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6393,6 +6393,8 @@ static int __init sched_domain_debug_set
 }
 early_param("sched_debug", sched_domain_debug_setup);
 
+static cpumask_var_t sched_domains_tmpmask; /* sched_domains_mutex */
+
 static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
 				  struct cpumask *groupmask)
 {
@@ -6476,7 +6478,6 @@ static int sched_domain_debug_one(struct
 
 static void sched_domain_debug(struct sched_domain *sd, int cpu)
 {
-	cpumask_var_t groupmask;
 	int level = 0;
 
 	if (!sched_domain_debug_enabled)
@@ -6489,20 +6490,14 @@ static void sched_domain_debug(struct sc
 
 	printk(KERN_DEBUG "CPU%d attaching sched-domain:\n", cpu);
 
-	if (!alloc_cpumask_var(&groupmask, GFP_KERNEL)) {
-		printk(KERN_DEBUG "Cannot load-balance (out of memory)\n");
-		return;
-	}
-
 	for (;;) {
-		if (sched_domain_debug_one(sd, cpu, level, groupmask))
+		if (sched_domain_debug_one(sd, cpu, level, sched_domains_tmpmask))
 			break;
 		level++;
 		sd = sd->parent;
 		if (!sd)
 			break;
 	}
-	free_cpumask_var(groupmask);
 }
 #else /* !CONFIG_SCHED_DEBUG */
 # define sched_domain_debug(sd, cpu) do { } while (0)
@@ -6707,7 +6702,7 @@ cpu_attach_domain(struct sched_domain *s
 			sd->child = NULL;
 	}
 
-//	sched_domain_debug(sd, cpu);
+	sched_domain_debug(sd, cpu);
 
 	rq_attach_root(rq, rd);
 	tmp = rq->sd;
@@ -6837,8 +6832,6 @@ static int get_group(int cpu, struct sd_
 	return cpu;
 }
 
-static cpumask_var_t sched_domains_tmpmask; /* sched_domains_mutex */
-
 /*
  * build_sched_groups takes the cpumask we wish to span, and a pointer
  * to a function which identifies what group(along with sched group) a CPU
@@ -7225,7 +7218,6 @@ static int build_sched_domains(const str
 	for_each_cpu(i, cpu_map) {
 		sd = *per_cpu_ptr(d.sd, i);
 		cpu_attach_domain(sd, d.rd, i);
-//		sched_domain_debug(sd, i);
 	}
 	rcu_read_unlock();
 

[tip:sched/domains] sched: Avoid allocations in sched_domain_debug()

[tip-bot for Peter Zijlstra]

Commit-ID:  4cb988395da6e16627a8be69729e50cd72ebb23e
Gitweb:     http://git.kernel.org/tip/4cb988395da6e16627a8be69729e50cd72ebb23e
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:58 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 14:05:00 +0200

sched: Avoid allocations in sched_domain_debug()

Since we're all serialized by sched_domains_mutex we can use
sched_domains_tmpmask and avoid having to do allocations. This means
we can use sched_domains_debug() for cpu_attach_domain() again.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.664347467@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   17 +++++------------
 1 files changed, 5 insertions(+), 12 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index fd73e91..35fc995 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6395,6 +6395,8 @@ early_initcall(migration_init);
 
 #ifdef CONFIG_SMP
 
+static cpumask_var_t sched_domains_tmpmask; /* sched_domains_mutex */
+
 #ifdef CONFIG_SCHED_DEBUG
 
 static __read_mostly int sched_domain_debug_enabled;
@@ -6490,7 +6492,6 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
 
 static void sched_domain_debug(struct sched_domain *sd, int cpu)
 {
-	cpumask_var_t groupmask;
 	int level = 0;
 
 	if (!sched_domain_debug_enabled)
@@ -6503,20 +6504,14 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu)
 
 	printk(KERN_DEBUG "CPU%d attaching sched-domain:\n", cpu);
 
-	if (!alloc_cpumask_var(&groupmask, GFP_KERNEL)) {
-		printk(KERN_DEBUG "Cannot load-balance (out of memory)\n");
-		return;
-	}
-
 	for (;;) {
-		if (sched_domain_debug_one(sd, cpu, level, groupmask))
+		if (sched_domain_debug_one(sd, cpu, level, sched_domains_tmpmask))
 			break;
 		level++;
 		sd = sd->parent;
 		if (!sd)
 			break;
 	}
-	free_cpumask_var(groupmask);
 }
 #else /* !CONFIG_SCHED_DEBUG */
 # define sched_domain_debug(sd, cpu) do { } while (0)
@@ -6721,7 +6716,7 @@ cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu)
 			sd->child = NULL;
 	}
 
-	/* sched_domain_debug(sd, cpu); */
+	sched_domain_debug(sd, cpu);
 
 	rq_attach_root(rq, rd);
 	tmp = rq->sd;
@@ -6851,8 +6846,6 @@ static int get_group(int cpu, struct sd_data *sdd, struct sched_group **sg)
 	return cpu;
 }
 
-static cpumask_var_t sched_domains_tmpmask; /* sched_domains_mutex */
-
 /*
  * build_sched_groups takes the cpumask we wish to span, and a pointer
  * to a function which identifies what group(along with sched group) a CPU
@@ -7896,8 +7889,8 @@ void __init sched_init(void)
 
 	/* Allocate the nohz_cpu_mask if CONFIG_CPUMASK_OFFSTACK */
 	zalloc_cpumask_var(&nohz_cpu_mask, GFP_NOWAIT);
-	zalloc_cpumask_var(&sched_domains_tmpmask, GFP_NOWAIT);
 #ifdef CONFIG_SMP
+	zalloc_cpumask_var(&sched_domains_tmpmask, GFP_NOWAIT);
 #ifdef CONFIG_NO_HZ
 	zalloc_cpumask_var(&nohz.idle_cpus_mask, GFP_NOWAIT);
 	alloc_cpumask_var(&nohz.grp_idle_mask, GFP_NOWAIT);

[PATCH 18/23] sched: Create proper cpu_$DOM_mask() functions

[Peter Zijlstra]

[-- Attachment #1: sched-foo16.patch --]
[-- Type: text/plain, Size: 2156 bytes --]

In order to unify the sched domain creation more, create proper
cpu_$DOM_mask() functions for those domains that didn't already have
one.

Use the sched_domains_tmpmask for the weird NUMA domain span.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   22 +++++++++++++++++-----
 1 file changed, 17 insertions(+), 5 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6793,8 +6793,22 @@ static void sched_domain_node_span(int n
 		cpumask_or(span, span, cpumask_of_node(next_node));
 	}
 }
+
+static const struct cpumask *cpu_node_mask(int cpu)
+{
+	lockdep_assert_held(&sched_domains_mutex);
+
+	sched_domain_node_span(cpu_to_node(cpu), sched_domains_tmpmask);
+
+	return sched_domains_tmpmask;
+}
 #endif /* CONFIG_NUMA */
 
+static const struct cpumask *cpu_cpu_mask(int cpu)
+{
+	return cpumask_of_node(cpu_to_node(cpu));
+}
+
 int sched_smt_power_savings = 0, sched_mc_power_savings = 0;
 
 struct sd_data {
@@ -7074,7 +7088,7 @@ static struct sched_domain *__build_alln
 #ifdef CONFIG_NUMA
 	sd = sd_init_ALLNODES(d, i);
 	set_domain_attribute(sd, attr);
-	cpumask_copy(sched_domain_span(sd), cpu_map);
+	cpumask_and(sched_domain_span(sd), cpu_map, cpu_possible_mask);
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;
@@ -7090,8 +7104,7 @@ static struct sched_domain *__build_node
 #ifdef CONFIG_NUMA
 	sd = sd_init_NODE(d, i);
 	set_domain_attribute(sd, attr);
-	sched_domain_node_span(cpu_to_node(i), sched_domain_span(sd));
-	cpumask_and(sched_domain_span(sd), sched_domain_span(sd), cpu_map);
+	cpumask_and(sched_domain_span(sd), cpu_map, cpu_node_mask(i));
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;
@@ -7106,8 +7119,7 @@ static struct sched_domain *__build_cpu_
 	struct sched_domain *sd;
 	sd = sd_init_CPU(d, i);
 	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd),
-			cpumask_of_node(cpu_to_node(i)), cpu_map);
+	cpumask_and(sched_domain_span(sd), cpu_map, cpu_cpu_mask(i));
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;

[tip:sched/domains] sched: Create proper cpu_$DOM_mask() functions

[tip-bot for Peter Zijlstra]

Commit-ID:  d3081f52f29da1ba6c27685519a9222b39eac763
Gitweb:     http://git.kernel.org/tip/d3081f52f29da1ba6c27685519a9222b39eac763
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:09:59 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 14:09:24 +0200

sched: Create proper cpu_$DOM_mask() functions

In order to unify the sched domain creation more, create proper
cpu_$DOM_mask() functions for those domains that didn't already have
one.

Use the sched_domains_tmpmask for the weird NUMA domain span.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.717702108@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   22 +++++++++++++++++-----
 1 files changed, 17 insertions(+), 5 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index 35fc995..3ae1e02 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6807,8 +6807,22 @@ static void sched_domain_node_span(int node, struct cpumask *span)
 		cpumask_or(span, span, cpumask_of_node(next_node));
 	}
 }
+
+static const struct cpumask *cpu_node_mask(int cpu)
+{
+	lockdep_assert_held(&sched_domains_mutex);
+
+	sched_domain_node_span(cpu_to_node(cpu), sched_domains_tmpmask);
+
+	return sched_domains_tmpmask;
+}
 #endif /* CONFIG_NUMA */
 
+static const struct cpumask *cpu_cpu_mask(int cpu)
+{
+	return cpumask_of_node(cpu_to_node(cpu));
+}
+
 int sched_smt_power_savings = 0, sched_mc_power_savings = 0;
 
 struct sd_data {
@@ -7088,7 +7102,7 @@ static struct sched_domain *__build_allnodes_sched_domain(struct s_data *d,
 #ifdef CONFIG_NUMA
 	sd = sd_init_ALLNODES(d, i);
 	set_domain_attribute(sd, attr);
-	cpumask_copy(sched_domain_span(sd), cpu_map);
+	cpumask_and(sched_domain_span(sd), cpu_map, cpu_possible_mask);
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;
@@ -7104,8 +7118,7 @@ static struct sched_domain *__build_node_sched_domain(struct s_data *d,
 #ifdef CONFIG_NUMA
 	sd = sd_init_NODE(d, i);
 	set_domain_attribute(sd, attr);
-	sched_domain_node_span(cpu_to_node(i), sched_domain_span(sd));
-	cpumask_and(sched_domain_span(sd), sched_domain_span(sd), cpu_map);
+	cpumask_and(sched_domain_span(sd), cpu_map, cpu_node_mask(i));
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;
@@ -7120,8 +7133,7 @@ static struct sched_domain *__build_cpu_sched_domain(struct s_data *d,
 	struct sched_domain *sd;
 	sd = sd_init_CPU(d, i);
 	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd),
-			cpumask_of_node(cpu_to_node(i)), cpu_map);
+	cpumask_and(sched_domain_span(sd), cpu_map, cpu_cpu_mask(i));
 	sd->parent = parent;
 	if (parent)
 		parent->child = sd;

[PATCH 19/23] sched: Stuff the sched_domain creation in a data-structure

[Peter Zijlstra]

[-- Attachment #1: sched-foo17.patch --]
[-- Type: text/plain, Size: 2201 bytes --]

In order to make the topology contruction fully dynamic, remove the
still hard-coded list of possible domains and stuck them in a
data-structure.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   32 ++++++++++++++++++++++++++------
 1 file changed, 26 insertions(+), 6 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6819,6 +6819,16 @@ enum s_alloc {
 	sa_none,
 };
 
+typedef struct sched_domain *(*sched_domain_build_f)(struct s_data *d,
+		const struct cpumask *cpu_map, struct sched_domain_attr *attr,
+		struct sched_domain *parent, int cpu);
+
+typedef const struct cpumask *(*sched_domain_mask_f)(int cpu);
+
+struct sched_domain_topology_level {
+	sched_domain_build_f build;
+};
+
 /*
  * Assumes the sched_domain tree is fully constructed
  */
@@ -7161,6 +7171,18 @@ static struct sched_domain *__build_smt_
 	return sd;
 }
 
+static struct sched_domain_topology_level default_topology[] = {
+	{ __build_allnodes_sched_domain, },
+	{ __build_node_sched_domain, },
+	{ __build_cpu_sched_domain, },
+	{ __build_book_sched_domain, },
+	{ __build_mc_sched_domain, },
+	{ __build_smt_sched_domain, },
+	{ NULL, },
+};
+
+static struct sched_domain_topology_level *sched_domain_topology = default_topology;
+
 /*
  * Build sched domains for a given set of cpus and attach the sched domains
  * to the individual cpus
@@ -7179,13 +7201,11 @@ static int __build_sched_domains(const s
 
 	/* Set up domains for cpus specified by the cpu_map. */
 	for_each_cpu(i, cpu_map) {
+		struct sched_domain_topology_level *tl;
+
 		sd = NULL;
-		sd = __build_allnodes_sched_domain(&d, cpu_map, attr, sd, i);
-		sd = __build_node_sched_domain(&d, cpu_map, attr, sd, i);
-		sd = __build_cpu_sched_domain(&d, cpu_map, attr, sd, i);
-		sd = __build_book_sched_domain(&d, cpu_map, attr, sd, i);
-		sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i);
-		sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i);
+		for (tl = sched_domain_topology; tl->build; tl++)
+			sd = tl->build(&d, cpu_map, attr, sd, i);
 
 		*per_cpu_ptr(d.sd, i) = sd;
 	}

[tip:sched/domains] sched: Stuff the sched_domain creation in a data-structure

[tip-bot for Peter Zijlstra]

Commit-ID:  eb7a74e6cd936c00749e2921b9e058631d986648
Gitweb:     http://git.kernel.org/tip/eb7a74e6cd936c00749e2921b9e058631d986648
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:10:00 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 14:09:26 +0200

sched: Stuff the sched_domain creation in a data-structure

In order to make the topology contruction fully dynamic, remove the
still hard-coded list of possible domains and stick them in a
data-structure.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.770335383@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   32 ++++++++++++++++++++++++++------
 1 files changed, 26 insertions(+), 6 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index 3ae1e02..f0e1821 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6843,6 +6843,16 @@ enum s_alloc {
 	sa_none,
 };
 
+typedef struct sched_domain *(*sched_domain_build_f)(struct s_data *d,
+		const struct cpumask *cpu_map, struct sched_domain_attr *attr,
+		struct sched_domain *parent, int cpu);
+
+typedef const struct cpumask *(*sched_domain_mask_f)(int cpu);
+
+struct sched_domain_topology_level {
+	sched_domain_build_f build;
+};
+
 /*
  * Assumes the sched_domain tree is fully constructed
  */
@@ -7185,6 +7195,18 @@ static struct sched_domain *__build_smt_sched_domain(struct s_data *d,
 	return sd;
 }
 
+static struct sched_domain_topology_level default_topology[] = {
+	{ __build_allnodes_sched_domain, },
+	{ __build_node_sched_domain, },
+	{ __build_cpu_sched_domain, },
+	{ __build_book_sched_domain, },
+	{ __build_mc_sched_domain, },
+	{ __build_smt_sched_domain, },
+	{ NULL, },
+};
+
+static struct sched_domain_topology_level *sched_domain_topology = default_topology;
+
 /*
  * Build sched domains for a given set of cpus and attach the sched domains
  * to the individual cpus
@@ -7203,13 +7225,11 @@ static int build_sched_domains(const struct cpumask *cpu_map,
 
 	/* Set up domains for cpus specified by the cpu_map. */
 	for_each_cpu(i, cpu_map) {
+		struct sched_domain_topology_level *tl;
+
 		sd = NULL;
-		sd = __build_allnodes_sched_domain(&d, cpu_map, attr, sd, i);
-		sd = __build_node_sched_domain(&d, cpu_map, attr, sd, i);
-		sd = __build_cpu_sched_domain(&d, cpu_map, attr, sd, i);
-		sd = __build_book_sched_domain(&d, cpu_map, attr, sd, i);
-		sd = __build_mc_sched_domain(&d, cpu_map, attr, sd, i);
-		sd = __build_smt_sched_domain(&d, cpu_map, attr, sd, i);
+		for (tl = sched_domain_topology; tl->build; tl++)
+			sd = tl->build(&d, cpu_map, attr, sd, i);
 
 		*per_cpu_ptr(d.sd, i) = sd;
 	}

[PATCH 20/23] sched: Unify the sched_domain build functions

[Peter Zijlstra]

[-- Attachment #1: sched-foo18.patch --]
[-- Type: text/plain, Size: 5687 bytes --]

Since all the __build_$DOM_sched_domain() functions do pretty much the
same thing, unify them.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |  133 ++++++++++++++++-----------------------------------------
 1 file changed, 39 insertions(+), 94 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6792,6 +6792,11 @@ static const struct cpumask *cpu_node_ma
 
 	return sched_domains_tmpmask;
 }
+
+static const struct cpumask *cpu_allnodes_mask(int cpu)
+{
+	return cpu_possible_mask;
+}
 #endif /* CONFIG_NUMA */
 
 static const struct cpumask *cpu_cpu_mask(int cpu)
@@ -6819,14 +6824,12 @@ enum s_alloc {
 	sa_none,
 };
 
-typedef struct sched_domain *(*sched_domain_build_f)(struct s_data *d,
-		const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-		struct sched_domain *parent, int cpu);
-
+typedef struct sched_domain *(*sched_domain_init_f)(struct s_data *d, int cpu);
 typedef const struct cpumask *(*sched_domain_mask_f)(int cpu);
 
 struct sched_domain_topology_level {
-	sched_domain_build_f build;
+	sched_domain_init_f init;
+	sched_domain_mask_f mask;
 };
 
 /*
@@ -7080,109 +7083,51 @@ static void claim_allocations(int cpu, s
 	}
 }
 
-static struct sched_domain *__build_allnodes_sched_domain(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-	struct sched_domain *parent, int i)
+#ifdef CONFIG_SCHED_SMT
+static const struct cpumask *cpu_smt_mask(int cpu)
 {
-	struct sched_domain *sd = NULL;
-#ifdef CONFIG_NUMA
-	sd = sd_init_ALLNODES(d, i);
-	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd), cpu_map, cpu_possible_mask);
-	sd->parent = parent;
-	if (parent)
-		parent->child = sd;
-#endif
-	return sd;
+	return topology_thread_cpumask(cpu);
 }
+#endif
 
-static struct sched_domain *__build_node_sched_domain(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-	struct sched_domain *parent, int i)
-{
-	struct sched_domain *sd = NULL;
+static struct sched_domain_topology_level default_topology[] = {
 #ifdef CONFIG_NUMA
-	sd = sd_init_NODE(d, i);
-	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd), cpu_map, cpu_node_mask(i));
-	sd->parent = parent;
-	if (parent)
-		parent->child = sd;
+	{ sd_init_ALLNODES, cpu_allnodes_mask, },
+	{ sd_init_NODE, cpu_node_mask, },
 #endif
-	return sd;
-}
-
-static struct sched_domain *__build_cpu_sched_domain(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-	struct sched_domain *parent, int i)
-{
-	struct sched_domain *sd;
-	sd = sd_init_CPU(d, i);
-	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd), cpu_map, cpu_cpu_mask(i));
-	sd->parent = parent;
-	if (parent)
-		parent->child = sd;
-	return sd;
-}
-
-static struct sched_domain *__build_book_sched_domain(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-	struct sched_domain *parent, int i)
-{
-	struct sched_domain *sd = parent;
+	{ sd_init_CPU, cpu_cpu_mask, },
 #ifdef CONFIG_SCHED_BOOK
-	sd = sd_init_BOOK(d, i);
-	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd), cpu_map, cpu_book_mask(i));
-	sd->parent = parent;
-	parent->child = sd;
+	{ sd_init_BOOK, cpu_book_mask, },
 #endif
-	return sd;
-}
-
-static struct sched_domain *__build_mc_sched_domain(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-	struct sched_domain *parent, int i)
-{
-	struct sched_domain *sd = parent;
 #ifdef CONFIG_SCHED_MC
-	sd = sd_init_MC(d, i);
-	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd), cpu_map, cpu_coregroup_mask(i));
-	sd->parent = parent;
-	parent->child = sd;
+	{ sd_init_MC, cpu_coregroup_mask, },
 #endif
-	return sd;
-}
-
-static struct sched_domain *__build_smt_sched_domain(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-	struct sched_domain *parent, int i)
-{
-	struct sched_domain *sd = parent;
 #ifdef CONFIG_SCHED_SMT
-	sd = sd_init_SIBLING(d, i);
-	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd), cpu_map, topology_thread_cpumask(i));
-	sd->parent = parent;
-	parent->child = sd;
+	{ sd_init_SIBLING, cpu_smt_mask, },
 #endif
-	return sd;
-}
-
-static struct sched_domain_topology_level default_topology[] = {
-	{ __build_allnodes_sched_domain, },
-	{ __build_node_sched_domain, },
-	{ __build_cpu_sched_domain, },
-	{ __build_book_sched_domain, },
-	{ __build_mc_sched_domain, },
-	{ __build_smt_sched_domain, },
 	{ NULL, },
 };
 
 static struct sched_domain_topology_level *sched_domain_topology = default_topology;
 
+struct sched_domain *build_sched_domain(struct sched_domain_topology_level *tl,
+		struct s_data *d, const struct cpumask *cpu_map,
+		struct sched_domain_attr *attr, struct sched_domain *parent,
+		int cpu)
+{
+	struct sched_domain *sd = tl->init(d, cpu);
+	if (!sd)
+		return parent;
+
+	set_domain_attribute(sd, attr);
+	cpumask_and(sched_domain_span(sd), cpu_map, tl->mask(cpu));
+	sd->parent = parent;
+	if (parent)
+		parent->child = sd;
+
+	return sd;
+}
+
 /*
  * Build sched domains for a given set of cpus and attach the sched domains
  * to the individual cpus
@@ -7204,8 +7149,8 @@ static int __build_sched_domains(const s
 		struct sched_domain_topology_level *tl;
 
 		sd = NULL;
-		for (tl = sched_domain_topology; tl->build; tl++)
-			sd = tl->build(&d, cpu_map, attr, sd, i);
+		for (tl = sched_domain_topology; tl->init; tl++)
+			sd = build_sched_domain(tl, &d, cpu_map, attr, sd, i);
 
 		*per_cpu_ptr(d.sd, i) = sd;
 	}

[tip:sched/domains] sched: Unify the sched_domain build functions

[tip-bot for Peter Zijlstra]

Commit-ID:  2c402dc3bb502e9dd74fce72c14d293fcef4719d
Gitweb:     http://git.kernel.org/tip/2c402dc3bb502e9dd74fce72c14d293fcef4719d
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:10:01 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 14:09:27 +0200

sched: Unify the sched_domain build functions

Since all the __build_$DOM_sched_domain() functions do pretty much the
same thing, unify them.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.826347257@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |  133 ++++++++++++++++---------------------------------------
 1 files changed, 39 insertions(+), 94 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index f0e1821..00d1e37 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6816,6 +6816,11 @@ static const struct cpumask *cpu_node_mask(int cpu)
 
 	return sched_domains_tmpmask;
 }
+
+static const struct cpumask *cpu_allnodes_mask(int cpu)
+{
+	return cpu_possible_mask;
+}
 #endif /* CONFIG_NUMA */
 
 static const struct cpumask *cpu_cpu_mask(int cpu)
@@ -6843,14 +6848,12 @@ enum s_alloc {
 	sa_none,
 };
 
-typedef struct sched_domain *(*sched_domain_build_f)(struct s_data *d,
-		const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-		struct sched_domain *parent, int cpu);
-
+typedef struct sched_domain *(*sched_domain_init_f)(struct s_data *d, int cpu);
 typedef const struct cpumask *(*sched_domain_mask_f)(int cpu);
 
 struct sched_domain_topology_level {
-	sched_domain_build_f build;
+	sched_domain_init_f init;
+	sched_domain_mask_f mask;
 };
 
 /*
@@ -7104,109 +7107,51 @@ static void claim_allocations(int cpu, struct sched_domain *sd)
 	}
 }
 
-static struct sched_domain *__build_allnodes_sched_domain(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-	struct sched_domain *parent, int i)
+#ifdef CONFIG_SCHED_SMT
+static const struct cpumask *cpu_smt_mask(int cpu)
 {
-	struct sched_domain *sd = NULL;
-#ifdef CONFIG_NUMA
-	sd = sd_init_ALLNODES(d, i);
-	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd), cpu_map, cpu_possible_mask);
-	sd->parent = parent;
-	if (parent)
-		parent->child = sd;
-#endif
-	return sd;
+	return topology_thread_cpumask(cpu);
 }
+#endif
 
-static struct sched_domain *__build_node_sched_domain(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-	struct sched_domain *parent, int i)
-{
-	struct sched_domain *sd = NULL;
+static struct sched_domain_topology_level default_topology[] = {
 #ifdef CONFIG_NUMA
-	sd = sd_init_NODE(d, i);
-	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd), cpu_map, cpu_node_mask(i));
-	sd->parent = parent;
-	if (parent)
-		parent->child = sd;
+	{ sd_init_ALLNODES, cpu_allnodes_mask, },
+	{ sd_init_NODE, cpu_node_mask, },
 #endif
-	return sd;
-}
-
-static struct sched_domain *__build_cpu_sched_domain(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-	struct sched_domain *parent, int i)
-{
-	struct sched_domain *sd;
-	sd = sd_init_CPU(d, i);
-	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd), cpu_map, cpu_cpu_mask(i));
-	sd->parent = parent;
-	if (parent)
-		parent->child = sd;
-	return sd;
-}
-
-static struct sched_domain *__build_book_sched_domain(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-	struct sched_domain *parent, int i)
-{
-	struct sched_domain *sd = parent;
+	{ sd_init_CPU, cpu_cpu_mask, },
 #ifdef CONFIG_SCHED_BOOK
-	sd = sd_init_BOOK(d, i);
-	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd), cpu_map, cpu_book_mask(i));
-	sd->parent = parent;
-	parent->child = sd;
+	{ sd_init_BOOK, cpu_book_mask, },
 #endif
-	return sd;
-}
-
-static struct sched_domain *__build_mc_sched_domain(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-	struct sched_domain *parent, int i)
-{
-	struct sched_domain *sd = parent;
 #ifdef CONFIG_SCHED_MC
-	sd = sd_init_MC(d, i);
-	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd), cpu_map, cpu_coregroup_mask(i));
-	sd->parent = parent;
-	parent->child = sd;
+	{ sd_init_MC, cpu_coregroup_mask, },
 #endif
-	return sd;
-}
-
-static struct sched_domain *__build_smt_sched_domain(struct s_data *d,
-	const struct cpumask *cpu_map, struct sched_domain_attr *attr,
-	struct sched_domain *parent, int i)
-{
-	struct sched_domain *sd = parent;
 #ifdef CONFIG_SCHED_SMT
-	sd = sd_init_SIBLING(d, i);
-	set_domain_attribute(sd, attr);
-	cpumask_and(sched_domain_span(sd), cpu_map, topology_thread_cpumask(i));
-	sd->parent = parent;
-	parent->child = sd;
+	{ sd_init_SIBLING, cpu_smt_mask, },
 #endif
-	return sd;
-}
-
-static struct sched_domain_topology_level default_topology[] = {
-	{ __build_allnodes_sched_domain, },
-	{ __build_node_sched_domain, },
-	{ __build_cpu_sched_domain, },
-	{ __build_book_sched_domain, },
-	{ __build_mc_sched_domain, },
-	{ __build_smt_sched_domain, },
 	{ NULL, },
 };
 
 static struct sched_domain_topology_level *sched_domain_topology = default_topology;
 
+struct sched_domain *build_sched_domain(struct sched_domain_topology_level *tl,
+		struct s_data *d, const struct cpumask *cpu_map,
+		struct sched_domain_attr *attr, struct sched_domain *parent,
+		int cpu)
+{
+	struct sched_domain *sd = tl->init(d, cpu);
+	if (!sd)
+		return parent;
+
+	set_domain_attribute(sd, attr);
+	cpumask_and(sched_domain_span(sd), cpu_map, tl->mask(cpu));
+	sd->parent = parent;
+	if (parent)
+		parent->child = sd;
+
+	return sd;
+}
+
 /*
  * Build sched domains for a given set of cpus and attach the sched domains
  * to the individual cpus
@@ -7228,8 +7173,8 @@ static int build_sched_domains(const struct cpumask *cpu_map,
 		struct sched_domain_topology_level *tl;
 
 		sd = NULL;
-		for (tl = sched_domain_topology; tl->build; tl++)
-			sd = tl->build(&d, cpu_map, attr, sd, i);
+		for (tl = sched_domain_topology; tl->init; tl++)
+			sd = build_sched_domain(tl, &d, cpu_map, attr, sd, i);
 
 		*per_cpu_ptr(d.sd, i) = sd;
 	}

[PATCH 21/23] sched: Reverse the topology list

[Peter Zijlstra]

[-- Attachment #1: sched-foo19.patch --]
[-- Type: text/plain, Size: 2174 bytes --]

In order to get rid of static sched_domain::level assignments, reverse
the topology iteration.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |   34 ++++++++++++++++++++--------------
 1 file changed, 20 insertions(+), 14 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -7090,20 +7090,23 @@ static const struct cpumask *cpu_smt_mas
 }
 #endif
 
+/*
+ * Topology list, bottom-up.
+ */
 static struct sched_domain_topology_level default_topology[] = {
-#ifdef CONFIG_NUMA
-	{ sd_init_ALLNODES, cpu_allnodes_mask, },
-	{ sd_init_NODE, cpu_node_mask, },
-#endif
-	{ sd_init_CPU, cpu_cpu_mask, },
-#ifdef CONFIG_SCHED_BOOK
-	{ sd_init_BOOK, cpu_book_mask, },
+#ifdef CONFIG_SCHED_SMT
+	{ sd_init_SIBLING, cpu_smt_mask, },
 #endif
 #ifdef CONFIG_SCHED_MC
 	{ sd_init_MC, cpu_coregroup_mask, },
 #endif
-#ifdef CONFIG_SCHED_SMT
-	{ sd_init_SIBLING, cpu_smt_mask, },
+#ifdef CONFIG_SCHED_BOOK
+	{ sd_init_BOOK, cpu_book_mask, },
+#endif
+	{ sd_init_CPU, cpu_cpu_mask, },
+#ifdef CONFIG_NUMA
+	{ sd_init_NODE, cpu_node_mask, },
+	{ sd_init_ALLNODES, cpu_allnodes_mask, },
 #endif
 	{ NULL, },
 };
@@ -7112,18 +7115,18 @@ static struct sched_domain_topology_leve
 
 struct sched_domain *build_sched_domain(struct sched_domain_topology_level *tl,
 		struct s_data *d, const struct cpumask *cpu_map,
-		struct sched_domain_attr *attr, struct sched_domain *parent,
+		struct sched_domain_attr *attr, struct sched_domain *child,
 		int cpu)
 {
 	struct sched_domain *sd = tl->init(d, cpu);
 	if (!sd)
-		return parent;
+		return child;
 
 	set_domain_attribute(sd, attr);
 	cpumask_and(sched_domain_span(sd), cpu_map, tl->mask(cpu));
-	sd->parent = parent;
-	if (parent)
-		parent->child = sd;
+	if (child)
+		child->parent = sd;
+	sd->child = child;
 
 	return sd;
 }
@@ -7152,6 +7155,9 @@ static int __build_sched_domains(const s
 		for (tl = sched_domain_topology; tl->init; tl++)
 			sd = build_sched_domain(tl, &d, cpu_map, attr, sd, i);
 
+		while (sd->child)
+			sd = sd->child;
+
 		*per_cpu_ptr(d.sd, i) = sd;
 	}
 

[tip:sched/domains] sched: Reverse the topology list

[tip-bot for Peter Zijlstra]

Commit-ID:  d069b916f7b50021d41d6ce498f86da32a7afaec
Gitweb:     http://git.kernel.org/tip/d069b916f7b50021d41d6ce498f86da32a7afaec
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:10:02 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 14:09:29 +0200

sched: Reverse the topology list

In order to get rid of static sched_domain::level assignments, reverse
the topology iteration.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.876506131@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |   34 ++++++++++++++++++++--------------
 1 files changed, 20 insertions(+), 14 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index 00d1e37..38bc53b 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -7114,20 +7114,23 @@ static const struct cpumask *cpu_smt_mask(int cpu)
 }
 #endif
 
+/*
+ * Topology list, bottom-up.
+ */
 static struct sched_domain_topology_level default_topology[] = {
-#ifdef CONFIG_NUMA
-	{ sd_init_ALLNODES, cpu_allnodes_mask, },
-	{ sd_init_NODE, cpu_node_mask, },
-#endif
-	{ sd_init_CPU, cpu_cpu_mask, },
-#ifdef CONFIG_SCHED_BOOK
-	{ sd_init_BOOK, cpu_book_mask, },
+#ifdef CONFIG_SCHED_SMT
+	{ sd_init_SIBLING, cpu_smt_mask, },
 #endif
 #ifdef CONFIG_SCHED_MC
 	{ sd_init_MC, cpu_coregroup_mask, },
 #endif
-#ifdef CONFIG_SCHED_SMT
-	{ sd_init_SIBLING, cpu_smt_mask, },
+#ifdef CONFIG_SCHED_BOOK
+	{ sd_init_BOOK, cpu_book_mask, },
+#endif
+	{ sd_init_CPU, cpu_cpu_mask, },
+#ifdef CONFIG_NUMA
+	{ sd_init_NODE, cpu_node_mask, },
+	{ sd_init_ALLNODES, cpu_allnodes_mask, },
 #endif
 	{ NULL, },
 };
@@ -7136,18 +7139,18 @@ static struct sched_domain_topology_level *sched_domain_topology = default_topol
 
 struct sched_domain *build_sched_domain(struct sched_domain_topology_level *tl,
 		struct s_data *d, const struct cpumask *cpu_map,
-		struct sched_domain_attr *attr, struct sched_domain *parent,
+		struct sched_domain_attr *attr, struct sched_domain *child,
 		int cpu)
 {
 	struct sched_domain *sd = tl->init(d, cpu);
 	if (!sd)
-		return parent;
+		return child;
 
 	set_domain_attribute(sd, attr);
 	cpumask_and(sched_domain_span(sd), cpu_map, tl->mask(cpu));
-	sd->parent = parent;
-	if (parent)
-		parent->child = sd;
+	if (child)
+		child->parent = sd;
+	sd->child = child;
 
 	return sd;
 }
@@ -7176,6 +7179,9 @@ static int build_sched_domains(const struct cpumask *cpu_map,
 		for (tl = sched_domain_topology; tl->init; tl++)
 			sd = build_sched_domain(tl, &d, cpu_map, attr, sd, i);
 
+		while (sd->child)
+			sd = sd->child;
+
 		*per_cpu_ptr(d.sd, i) = sd;
 	}
 

[PATCH 22/23] sched: Move sched domain storage into the topology list

[Peter Zijlstra]

[-- Attachment #1: sched-foo20.patch --]
[-- Type: text/plain, Size: 5713 bytes --]

In order to remove the last dependency on the statid domain levels,
move the sd_data storage into the topology structure.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/sched.c |  129 ++++++++++++++++++++++++++++++++++-----------------------
 1 file changed, 77 insertions(+), 52 deletions(-)

Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6813,7 +6813,6 @@ struct sd_data {
 
 struct s_data {
 	struct sched_domain ** __percpu sd;
-	struct sd_data 		sdd[SD_LV_MAX];
 	struct root_domain	*rd;
 };
 
@@ -6824,12 +6823,15 @@ enum s_alloc {
 	sa_none,
 };
 
-typedef struct sched_domain *(*sched_domain_init_f)(struct s_data *d, int cpu);
+struct sched_domain_topology_level;
+
+typedef struct sched_domain *(*sched_domain_init_f)(struct sched_domain_topology_level *tl, int cpu);
 typedef const struct cpumask *(*sched_domain_mask_f)(int cpu);
 
 struct sched_domain_topology_level {
 	sched_domain_init_f init;
 	sched_domain_mask_f mask;
+	struct sd_data      data;
 };
 
 /*
@@ -6934,15 +6936,16 @@ static void init_sched_groups_power(int 
 # define SD_INIT_NAME(sd, type)		do { } while (0)
 #endif
 
-#define SD_INIT_FUNC(type)						       \
-static noinline struct sched_domain *sd_init_##type(struct s_data *d, int cpu) \
-{									       \
-	struct sched_domain *sd = *per_cpu_ptr(d->sdd[SD_LV_##type].sd, cpu);  \
-	*sd = SD_##type##_INIT;						       \
-	sd->level = SD_LV_##type;					       \
-	SD_INIT_NAME(sd, type);						       \
-	sd->private = &d->sdd[SD_LV_##type];				       \
-	return sd;							       \
+#define SD_INIT_FUNC(type)						\
+static noinline struct sched_domain *					\
+sd_init_##type(struct sched_domain_topology_level *tl, int cpu) 	\
+{									\
+	struct sched_domain *sd = *per_cpu_ptr(tl->data.sd, cpu);	\
+	*sd = SD_##type##_INIT;						\
+	sd->level = SD_LV_##type;					\
+	SD_INIT_NAME(sd, type);						\
+	sd->private = &tl->data;					\
+	return sd;							\
 }
 
 SD_INIT_FUNC(CPU)
@@ -6995,11 +6998,12 @@ static void set_domain_attribute(struct 
 	}
 }
 
+static void __sdt_free(const struct cpumask *cpu_map);
+static int __sdt_alloc(const struct cpumask *cpu_map);
+
 static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 				 const struct cpumask *cpu_map)
 {
-	int i, j;
-
 	switch (what) {
 	case sa_rootdomain:
 		if (!atomic_read(&d->rd->refcount))
@@ -7007,14 +7011,7 @@ static void __free_domain_allocs(struct 
 	case sa_sd:
 		free_percpu(d->sd); /* fall through */
 	case sa_sd_storage:
-		for (i = 0; i < SD_LV_MAX; i++) {
-			for_each_cpu(j, cpu_map) {
-				kfree(*per_cpu_ptr(d->sdd[i].sd, j));
-				kfree(*per_cpu_ptr(d->sdd[i].sg, j));
-			}
-			free_percpu(d->sdd[i].sd);
-			free_percpu(d->sdd[i].sg);
-		} /* fall through */
+		__sdt_free(cpu_map); /* fall through */
 	case sa_none:
 		break;
 	}
@@ -7023,38 +7020,10 @@ static void __free_domain_allocs(struct 
 static enum s_alloc __visit_domain_allocation_hell(struct s_data *d,
 						   const struct cpumask *cpu_map)
 {
-	int i, j;
-
 	memset(d, 0, sizeof(*d));
 
-	for (i = 0; i < SD_LV_MAX; i++) {
-		d->sdd[i].sd = alloc_percpu(struct sched_domain *);
-		if (!d->sdd[i].sd)
-			return sa_sd_storage;
-
-		d->sdd[i].sg = alloc_percpu(struct sched_group *);
-		if (!d->sdd[i].sg)
-			return sa_sd_storage;
-
-		for_each_cpu(j, cpu_map) {
-			struct sched_domain *sd;
-			struct sched_group *sg;
-
-		       	sd = kzalloc_node(sizeof(struct sched_domain) + cpumask_size(),
-					GFP_KERNEL, cpu_to_node(j));
-			if (!sd)
-				return sa_sd_storage;
-
-			*per_cpu_ptr(d->sdd[i].sd, j) = sd;
-
-			sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(),
-					GFP_KERNEL, cpu_to_node(j));
-			if (!sg)
-				return sa_sd_storage;
-
-			*per_cpu_ptr(d->sdd[i].sg, j) = sg;
-		}
-	}
+	if (__sdt_alloc(cpu_map))
+		return sa_sd_storage;
 	d->sd = alloc_percpu(struct sched_domain *);
 	if (!d->sd)
 		return sa_sd_storage;
@@ -7113,12 +7082,68 @@ static struct sched_domain_topology_leve
 
 static struct sched_domain_topology_level *sched_domain_topology = default_topology;
 
+static int __sdt_alloc(const struct cpumask *cpu_map)
+{
+	struct sched_domain_topology_level *tl;
+	int j;
+
+	for (tl = sched_domain_topology; tl->init; tl++) {
+		struct sd_data *sdd = &tl->data;
+
+		sdd->sd = alloc_percpu(struct sched_domain *);
+		if (!sdd->sd)
+			return -ENOMEM;
+
+		sdd->sg = alloc_percpu(struct sched_group *);
+		if (!sdd->sg)
+			return -ENOMEM;
+
+		for_each_cpu(j, cpu_map) {
+			struct sched_domain *sd;
+			struct sched_group *sg;
+
+		       	sd = kzalloc_node(sizeof(struct sched_domain) + cpumask_size(),
+					GFP_KERNEL, cpu_to_node(j));
+			if (!sd)
+				return -ENOMEM;
+
+			*per_cpu_ptr(sdd->sd, j) = sd;
+
+			sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(),
+					GFP_KERNEL, cpu_to_node(j));
+			if (!sg)
+				return -ENOMEM;
+
+			*per_cpu_ptr(sdd->sg, j) = sg;
+		}
+	}
+
+	return 0;
+}
+
+static void __sdt_free(const struct cpumask *cpu_map)
+{
+	struct sched_domain_topology_level *tl;
+	int j;
+
+	for (tl = sched_domain_topology; tl->init; tl++) {
+		struct sd_data *sdd = &tl->data;
+
+		for_each_cpu(j, cpu_map) {
+			kfree(*per_cpu_ptr(sdd->sd, j));
+			kfree(*per_cpu_ptr(sdd->sg, j));
+		}
+		free_percpu(sdd->sd);
+		free_percpu(sdd->sg);
+	}
+}
+
 struct sched_domain *build_sched_domain(struct sched_domain_topology_level *tl,
 		struct s_data *d, const struct cpumask *cpu_map,
 		struct sched_domain_attr *attr, struct sched_domain *child,
 		int cpu)
 {
-	struct sched_domain *sd = tl->init(d, cpu);
+	struct sched_domain *sd = tl->init(tl, cpu);
 	if (!sd)
 		return child;
 

[tip:sched/domains] sched: Move sched domain storage into the topology list

[tip-bot for Peter Zijlstra]

Commit-ID:  54ab4ff4316eb329d2c1acc110fbc623d2966931
Gitweb:     http://git.kernel.org/tip/54ab4ff4316eb329d2c1acc110fbc623d2966931
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:10:03 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 14:09:31 +0200

sched: Move sched domain storage into the topology list

In order to remove the last dependency on the statid domain levels,
move the sd_data storage into the topology structure.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.924926412@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched.c |  129 +++++++++++++++++++++++++++++++++----------------------
 1 files changed, 77 insertions(+), 52 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index 38bc53b..3231e19 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6837,7 +6837,6 @@ struct sd_data {
 
 struct s_data {
 	struct sched_domain ** __percpu sd;
-	struct sd_data 		sdd[SD_LV_MAX];
 	struct root_domain	*rd;
 };
 
@@ -6848,12 +6847,15 @@ enum s_alloc {
 	sa_none,
 };
 
-typedef struct sched_domain *(*sched_domain_init_f)(struct s_data *d, int cpu);
+struct sched_domain_topology_level;
+
+typedef struct sched_domain *(*sched_domain_init_f)(struct sched_domain_topology_level *tl, int cpu);
 typedef const struct cpumask *(*sched_domain_mask_f)(int cpu);
 
 struct sched_domain_topology_level {
 	sched_domain_init_f init;
 	sched_domain_mask_f mask;
+	struct sd_data      data;
 };
 
 /*
@@ -6958,15 +6960,16 @@ static void init_sched_groups_power(int cpu, struct sched_domain *sd)
 # define SD_INIT_NAME(sd, type)		do { } while (0)
 #endif
 
-#define SD_INIT_FUNC(type)						       \
-static noinline struct sched_domain *sd_init_##type(struct s_data *d, int cpu) \
-{									       \
-	struct sched_domain *sd = *per_cpu_ptr(d->sdd[SD_LV_##type].sd, cpu);  \
-	*sd = SD_##type##_INIT;						       \
-	sd->level = SD_LV_##type;					       \
-	SD_INIT_NAME(sd, type);						       \
-	sd->private = &d->sdd[SD_LV_##type];				       \
-	return sd;							       \
+#define SD_INIT_FUNC(type)						\
+static noinline struct sched_domain *					\
+sd_init_##type(struct sched_domain_topology_level *tl, int cpu) 	\
+{									\
+	struct sched_domain *sd = *per_cpu_ptr(tl->data.sd, cpu);	\
+	*sd = SD_##type##_INIT;						\
+	sd->level = SD_LV_##type;					\
+	SD_INIT_NAME(sd, type);						\
+	sd->private = &tl->data;					\
+	return sd;							\
 }
 
 SD_INIT_FUNC(CPU)
@@ -7019,11 +7022,12 @@ static void set_domain_attribute(struct sched_domain *sd,
 	}
 }
 
+static void __sdt_free(const struct cpumask *cpu_map);
+static int __sdt_alloc(const struct cpumask *cpu_map);
+
 static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 				 const struct cpumask *cpu_map)
 {
-	int i, j;
-
 	switch (what) {
 	case sa_rootdomain:
 		if (!atomic_read(&d->rd->refcount))
@@ -7031,14 +7035,7 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 	case sa_sd:
 		free_percpu(d->sd); /* fall through */
 	case sa_sd_storage:
-		for (i = 0; i < SD_LV_MAX; i++) {
-			for_each_cpu(j, cpu_map) {
-				kfree(*per_cpu_ptr(d->sdd[i].sd, j));
-				kfree(*per_cpu_ptr(d->sdd[i].sg, j));
-			}
-			free_percpu(d->sdd[i].sd);
-			free_percpu(d->sdd[i].sg);
-		} /* fall through */
+		__sdt_free(cpu_map); /* fall through */
 	case sa_none:
 		break;
 	}
@@ -7047,38 +7044,10 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what,
 static enum s_alloc __visit_domain_allocation_hell(struct s_data *d,
 						   const struct cpumask *cpu_map)
 {
-	int i, j;
-
 	memset(d, 0, sizeof(*d));
 
-	for (i = 0; i < SD_LV_MAX; i++) {
-		d->sdd[i].sd = alloc_percpu(struct sched_domain *);
-		if (!d->sdd[i].sd)
-			return sa_sd_storage;
-
-		d->sdd[i].sg = alloc_percpu(struct sched_group *);
-		if (!d->sdd[i].sg)
-			return sa_sd_storage;
-
-		for_each_cpu(j, cpu_map) {
-			struct sched_domain *sd;
-			struct sched_group *sg;
-
-		       	sd = kzalloc_node(sizeof(struct sched_domain) + cpumask_size(),
-					GFP_KERNEL, cpu_to_node(j));
-			if (!sd)
-				return sa_sd_storage;
-
-			*per_cpu_ptr(d->sdd[i].sd, j) = sd;
-
-			sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(),
-					GFP_KERNEL, cpu_to_node(j));
-			if (!sg)
-				return sa_sd_storage;
-
-			*per_cpu_ptr(d->sdd[i].sg, j) = sg;
-		}
-	}
+	if (__sdt_alloc(cpu_map))
+		return sa_sd_storage;
 	d->sd = alloc_percpu(struct sched_domain *);
 	if (!d->sd)
 		return sa_sd_storage;
@@ -7137,12 +7106,68 @@ static struct sched_domain_topology_level default_topology[] = {
 
 static struct sched_domain_topology_level *sched_domain_topology = default_topology;
 
+static int __sdt_alloc(const struct cpumask *cpu_map)
+{
+	struct sched_domain_topology_level *tl;
+	int j;
+
+	for (tl = sched_domain_topology; tl->init; tl++) {
+		struct sd_data *sdd = &tl->data;
+
+		sdd->sd = alloc_percpu(struct sched_domain *);
+		if (!sdd->sd)
+			return -ENOMEM;
+
+		sdd->sg = alloc_percpu(struct sched_group *);
+		if (!sdd->sg)
+			return -ENOMEM;
+
+		for_each_cpu(j, cpu_map) {
+			struct sched_domain *sd;
+			struct sched_group *sg;
+
+		       	sd = kzalloc_node(sizeof(struct sched_domain) + cpumask_size(),
+					GFP_KERNEL, cpu_to_node(j));
+			if (!sd)
+				return -ENOMEM;
+
+			*per_cpu_ptr(sdd->sd, j) = sd;
+
+			sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(),
+					GFP_KERNEL, cpu_to_node(j));
+			if (!sg)
+				return -ENOMEM;
+
+			*per_cpu_ptr(sdd->sg, j) = sg;
+		}
+	}
+
+	return 0;
+}
+
+static void __sdt_free(const struct cpumask *cpu_map)
+{
+	struct sched_domain_topology_level *tl;
+	int j;
+
+	for (tl = sched_domain_topology; tl->init; tl++) {
+		struct sd_data *sdd = &tl->data;
+
+		for_each_cpu(j, cpu_map) {
+			kfree(*per_cpu_ptr(sdd->sd, j));
+			kfree(*per_cpu_ptr(sdd->sg, j));
+		}
+		free_percpu(sdd->sd);
+		free_percpu(sdd->sg);
+	}
+}
+
 struct sched_domain *build_sched_domain(struct sched_domain_topology_level *tl,
 		struct s_data *d, const struct cpumask *cpu_map,
 		struct sched_domain_attr *attr, struct sched_domain *child,
 		int cpu)
 {
-	struct sched_domain *sd = tl->init(d, cpu);
+	struct sched_domain *sd = tl->init(tl, cpu);
 	if (!sd)
 		return child;
 

Re: [PATCH 22/23] sched: Move sched domain storage into the topology list

[Andreas Herrmann]

On Thu, Apr 07, 2011 at 08:10:03AM -0400, Peter Zijlstra wrote:
> In order to remove the last dependency on the statid domain levels,

typo:

s/statid/static/


Andreas

[PATCH 23/23] sched: Dynamic sched_domain::level

[Peter Zijlstra]

[-- Attachment #1: sched-foo21.patch --]
[-- Type: text/plain, Size: 3038 bytes --]

Remove the SD_LV_ enum and use dynamic level assignments.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 include/linux/sched.h |   23 +++--------------------
 kernel/cpuset.c       |    2 +-
 kernel/sched.c        |    9 ++++++---
 3 files changed, 10 insertions(+), 24 deletions(-)

Index: linux-2.6/include/linux/sched.h
===================================================================
--- linux-2.6.orig/include/linux/sched.h
+++ linux-2.6/include/linux/sched.h
@@ -891,25 +891,6 @@ static inline struct cpumask *sched_grou
 	return to_cpumask(sg->cpumask);
 }
 
-enum sched_domain_level {
-	SD_LV_NONE = 0,
-#ifdef CONFIG_SCHED_SMT
-	SD_LV_SIBLING,
-#endif
-#ifdef CONFIG_SCHED_MC
-	SD_LV_MC,
-#endif
-#ifdef CONFIG_SCHED_BOOK
-	SD_LV_BOOK,
-#endif
-	SD_LV_CPU,
-#ifdef CONFIG_NUMA
-	SD_LV_NODE,
-	SD_LV_ALLNODES,
-#endif
-	SD_LV_MAX
-};
-
 struct sched_domain_attr {
 	int relax_domain_level;
 };
@@ -918,6 +899,8 @@ struct sched_domain_attr {
 	.relax_domain_level = -1,			\
 }
 
+extern int sched_domain_level_max;
+
 struct sched_domain {
 	/* These fields must be setup */
 	struct sched_domain *parent;	/* top domain must be null terminated */
@@ -935,7 +918,7 @@ struct sched_domain {
 	unsigned int forkexec_idx;
 	unsigned int smt_gain;
 	int flags;			/* See SD_* */
-	enum sched_domain_level level;
+	int level;
 
 	/* Runtime fields. */
 	unsigned long last_balance;	/* init to jiffies. units in jiffies */
Index: linux-2.6/kernel/cpuset.c
===================================================================
--- linux-2.6.orig/kernel/cpuset.c
+++ linux-2.6/kernel/cpuset.c
@@ -1164,7 +1164,7 @@ int current_cpuset_is_being_rebound(void
 static int update_relax_domain_level(struct cpuset *cs, s64 val)
 {
 #ifdef CONFIG_SMP
-	if (val < -1 || val >= SD_LV_MAX)
+	if (val < -1 || val >= sched_domain_level_max)
 		return -EINVAL;
 #endif
 
Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6942,7 +6942,6 @@ sd_init_##type(struct sched_domain_topol
 {									\
 	struct sched_domain *sd = *per_cpu_ptr(tl->data.sd, cpu);	\
 	*sd = SD_##type##_INIT;						\
-	sd->level = SD_LV_##type;					\
 	SD_INIT_NAME(sd, type);						\
 	sd->private = &tl->data;					\
 	return sd;							\
@@ -6964,13 +6963,14 @@ SD_INIT_FUNC(CPU)
 #endif
 
 static int default_relax_domain_level = -1;
+int sched_domain_level_max;
 
 static int __init setup_relax_domain_level(char *str)
 {
 	unsigned long val;
 
 	val = simple_strtoul(str, NULL, 0);
-	if (val < SD_LV_MAX)
+	if (val < sched_domain_level_max)
 		default_relax_domain_level = val;
 
 	return 1;
@@ -7149,8 +7149,11 @@ struct sched_domain *build_sched_domain(
 
 	set_domain_attribute(sd, attr);
 	cpumask_and(sched_domain_span(sd), cpu_map, tl->mask(cpu));
-	if (child)
+	if (child) {
+		sd->level = child->level + 1;
+		sched_domain_level_max = max(sched_domain_level_max, sd->level);
 		child->parent = sd;
+	}
 	sd->child = child;
 
 	return sd;

[tip:sched/domains] sched: Dynamic sched_domain::level

[tip-bot for Peter Zijlstra]

Commit-ID:  60495e7760d8ee364695006af37309b0755e0e17
Gitweb:     http://git.kernel.org/tip/60495e7760d8ee364695006af37309b0755e0e17
Author:     Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Thu, 7 Apr 2011 14:10:04 +0200
Committer:  Ingo Molnar <mingo@elte.hu>
CommitDate: Mon, 11 Apr 2011 14:09:32 +0200

sched: Dynamic sched_domain::level

Remove the SD_LV_ enum and use dynamic level assignments.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Nick Piggin <npiggin@kernel.dk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20110407122942.969433965@chello.nl
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/sched.h |   23 +++--------------------
 kernel/cpuset.c       |    2 +-
 kernel/sched.c        |    9 ++++++---
 3 files changed, 10 insertions(+), 24 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 09d9e02..e43e5b0 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -892,25 +892,6 @@ static inline struct cpumask *sched_group_cpus(struct sched_group *sg)
 	return to_cpumask(sg->cpumask);
 }
 
-enum sched_domain_level {
-	SD_LV_NONE = 0,
-#ifdef CONFIG_SCHED_SMT
-	SD_LV_SIBLING,
-#endif
-#ifdef CONFIG_SCHED_MC
-	SD_LV_MC,
-#endif
-#ifdef CONFIG_SCHED_BOOK
-	SD_LV_BOOK,
-#endif
-	SD_LV_CPU,
-#ifdef CONFIG_NUMA
-	SD_LV_NODE,
-	SD_LV_ALLNODES,
-#endif
-	SD_LV_MAX
-};
-
 struct sched_domain_attr {
 	int relax_domain_level;
 };
@@ -919,6 +900,8 @@ struct sched_domain_attr {
 	.relax_domain_level = -1,			\
 }
 
+extern int sched_domain_level_max;
+
 struct sched_domain {
 	/* These fields must be setup */
 	struct sched_domain *parent;	/* top domain must be null terminated */
@@ -936,7 +919,7 @@ struct sched_domain {
 	unsigned int forkexec_idx;
 	unsigned int smt_gain;
 	int flags;			/* See SD_* */
-	enum sched_domain_level level;
+	int level;
 
 	/* Runtime fields. */
 	unsigned long last_balance;	/* init to jiffies. units in jiffies */
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 33eee16..2bb8c2e 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -1159,7 +1159,7 @@ int current_cpuset_is_being_rebound(void)
 static int update_relax_domain_level(struct cpuset *cs, s64 val)
 {
 #ifdef CONFIG_SMP
-	if (val < -1 || val >= SD_LV_MAX)
+	if (val < -1 || val >= sched_domain_level_max)
 		return -EINVAL;
 #endif
 
diff --git a/kernel/sched.c b/kernel/sched.c
index 3231e19..506cb81 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6966,7 +6966,6 @@ sd_init_##type(struct sched_domain_topology_level *tl, int cpu) 	\
 {									\
 	struct sched_domain *sd = *per_cpu_ptr(tl->data.sd, cpu);	\
 	*sd = SD_##type##_INIT;						\
-	sd->level = SD_LV_##type;					\
 	SD_INIT_NAME(sd, type);						\
 	sd->private = &tl->data;					\
 	return sd;							\
@@ -6988,13 +6987,14 @@ SD_INIT_FUNC(CPU)
 #endif
 
 static int default_relax_domain_level = -1;
+int sched_domain_level_max;
 
 static int __init setup_relax_domain_level(char *str)
 {
 	unsigned long val;
 
 	val = simple_strtoul(str, NULL, 0);
-	if (val < SD_LV_MAX)
+	if (val < sched_domain_level_max)
 		default_relax_domain_level = val;
 
 	return 1;
@@ -7173,8 +7173,11 @@ struct sched_domain *build_sched_domain(struct sched_domain_topology_level *tl,
 
 	set_domain_attribute(sd, attr);
 	cpumask_and(sched_domain_span(sd), cpu_map, tl->mask(cpu));
-	if (child)
+	if (child) {
+		sd->level = child->level + 1;
+		sched_domain_level_max = max(sched_domain_level_max, sd->level);
 		child->parent = sd;
+	}
 	sd->child = child;
 
 	return sd;

Re: [PATCH 00/23] sched: Rewrite sched_domain/sched_group creation

[Mike Galbraith]

On Thu, 2011-04-07 at 14:09 +0200, Peter Zijlstra wrote:
> This series rewrite the sched_domain and sched_group creation code.
> 
> While its still not completely finished it does get us a lot of cleanups
> and code reduction and seems fairly stable at this point and should thus
> be a fairly good base to continue from.

Looks and sounds like good plan to me.

> Also available through:
>   git://git.kernel.org/pub/scm/linux/kernel/git/peterz/linux-2.6-sched.git sched_domain
> 
> ---
>  include/linux/sched.h |   26 +-
>  kernel/cpuset.c       |    2 +-
>  kernel/sched.c        |  963 +++++++++++++++----------------------------------
>  kernel/sched_fair.c   |   32 ++-
>  4 files changed, 326 insertions(+), 697 deletions(-)

We could use a few more diffstats like this one :)

	-Mike

[RFC][PATCH 24/23] sched: Rewrite CONFIG_NUMA support

[Peter Zijlstra]

The below is proven to be broken on a non-trivial NUMA setup (4 socekt
magny-cours, tested by Andreas) but shows the direction I'm wanting to
take for NUMA.

The current stuff 16 nodes in a domain and one top-level domain to rule
them all just doesn't sound right, esp for these small
non-fully-connected systems of today.

So what it attempts is to sort the numa-distance table and create
domains for each grouping resulting from that. This should then match
the actual machine topology much better.

The problem with it is that its quite possible to generate overlapping
groups for a domain and the magny-cours topology makes that happen. I've
still not quite figured out wth to do about that though :-/


---
Subject: sched: Rewrite CONFIG_NUMA support
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Fri Mar 25 21:52:17 CET 2011

Rewrite the CONFIG_NUMA sched domain support.

The current code groups up to 16 nodes in a level and then puts an
ALLNODES domain spanning the entire tree on top of that. This doesn't
reflect the numa topology and esp for the smaller not-fully-connected
machines out there today this might make a difference.

Therefore, build a proper numa topology based on node_distance().

TODO: figure out a way to set SD_flags based on distance such that
      we disable various expensive load-balancing features at some
      point and increase the balance interval prop. to the distance.

XXX: remove debug prints

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
---
 include/linux/topology.h |   25 -----
 kernel/sched.c           |  201 +++++++++++++++++++++++++----------------------
 2 files changed, 110 insertions(+), 116 deletions(-)

Index: linux-2.6/include/linux/topology.h
===================================================================
--- linux-2.6.orig/include/linux/topology.h
+++ linux-2.6/include/linux/topology.h
@@ -176,31 +176,6 @@ int arch_update_cpu_topology(void);
 }
 #endif
 
-/* sched_domains SD_ALLNODES_INIT for NUMA machines */
-#define SD_ALLNODES_INIT (struct sched_domain) {			\
-	.min_interval		= 64,					\
-	.max_interval		= 64*num_online_cpus(),			\
-	.busy_factor		= 128,					\
-	.imbalance_pct		= 133,					\
-	.cache_nice_tries	= 1,					\
-	.busy_idx		= 3,					\
-	.idle_idx		= 3,					\
-	.flags			= 1*SD_LOAD_BALANCE			\
-				| 1*SD_BALANCE_NEWIDLE			\
-				| 0*SD_BALANCE_EXEC			\
-				| 0*SD_BALANCE_FORK			\
-				| 0*SD_BALANCE_WAKE			\
-				| 0*SD_WAKE_AFFINE			\
-				| 0*SD_SHARE_CPUPOWER			\
-				| 0*SD_POWERSAVINGS_BALANCE		\
-				| 0*SD_SHARE_PKG_RESOURCES		\
-				| 1*SD_SERIALIZE			\
-				| 0*SD_PREFER_SIBLING			\
-				,					\
-	.last_balance		= jiffies,				\
-	.balance_interval	= 64,					\
-}
-
 #ifdef CONFIG_SCHED_BOOK
 #ifndef SD_BOOK_INIT
 #error Please define an appropriate SD_BOOK_INIT in include/asm/topology.h!!!
Index: linux-2.6/kernel/sched.c
===================================================================
--- linux-2.6.orig/kernel/sched.c
+++ linux-2.6/kernel/sched.c
@@ -6723,92 +6723,6 @@ static int __init isolated_cpu_setup(cha
 
 __setup("isolcpus=", isolated_cpu_setup);
 
-#define SD_NODES_PER_DOMAIN 16
-
-#ifdef CONFIG_NUMA
-
-/**
- * find_next_best_node - find the next node to include in a sched_domain
- * @node: node whose sched_domain we're building
- * @used_nodes: nodes already in the sched_domain
- *
- * Find the next node to include in a given scheduling domain. Simply
- * finds the closest node not already in the @used_nodes map.
- *
- * Should use nodemask_t.
- */
-static int find_next_best_node(int node, nodemask_t *used_nodes)
-{
-	int i, n, val, min_val, best_node = 0;
-
-	min_val = INT_MAX;
-
-	for (i = 0; i < nr_node_ids; i++) {
-		/* Start at @node */
-		n = (node + i) % nr_node_ids;
-
-		if (!nr_cpus_node(n))
-			continue;
-
-		/* Skip already used nodes */
-		if (node_isset(n, *used_nodes))
-			continue;
-
-		/* Simple min distance search */
-		val = node_distance(node, n);
-
-		if (val < min_val) {
-			min_val = val;
-			best_node = n;
-		}
-	}
-
-	node_set(best_node, *used_nodes);
-	return best_node;
-}
-
-/**
- * sched_domain_node_span - get a cpumask for a node's sched_domain
- * @node: node whose cpumask we're constructing
- * @span: resulting cpumask
- *
- * Given a node, construct a good cpumask for its sched_domain to span. It
- * should be one that prevents unnecessary balancing, but also spreads tasks
- * out optimally.
- */
-static void sched_domain_node_span(int node, struct cpumask *span)
-{
-	nodemask_t used_nodes;
-	int i;
-
-	cpumask_clear(span);
-	nodes_clear(used_nodes);
-
-	cpumask_or(span, span, cpumask_of_node(node));
-	node_set(node, used_nodes);
-
-	for (i = 1; i < SD_NODES_PER_DOMAIN; i++) {
-		int next_node = find_next_best_node(node, &used_nodes);
-
-		cpumask_or(span, span, cpumask_of_node(next_node));
-	}
-}
-
-static const struct cpumask *cpu_node_mask(int cpu)
-{
-	lockdep_assert_held(&sched_domains_mutex);
-
-	sched_domain_node_span(cpu_to_node(cpu), sched_domains_tmpmask);
-
-	return sched_domains_tmpmask;
-}
-
-static const struct cpumask *cpu_allnodes_mask(int cpu)
-{
-	return cpu_possible_mask;
-}
-#endif /* CONFIG_NUMA */
-
 static const struct cpumask *cpu_cpu_mask(int cpu)
 {
 	return cpumask_of_node(cpu_to_node(cpu));
@@ -6841,6 +6755,7 @@ typedef const struct cpumask *(*sched_do
 struct sched_domain_topology_level {
 	sched_domain_init_f init;
 	sched_domain_mask_f mask;
+	int		    numa_level;
 	struct sd_data      data;
 };
 
@@ -6959,7 +6874,6 @@ sd_init_##type(struct sched_domain_topol
 
 SD_INIT_FUNC(CPU)
 #ifdef CONFIG_NUMA
- SD_INIT_FUNC(ALLNODES)
  SD_INIT_FUNC(NODE)
 #endif
 #ifdef CONFIG_SCHED_SMT
@@ -7083,15 +6997,118 @@ static struct sched_domain_topology_leve
 	{ sd_init_BOOK, cpu_book_mask, },
 #endif
 	{ sd_init_CPU, cpu_cpu_mask, },
-#ifdef CONFIG_NUMA
-	{ sd_init_NODE, cpu_node_mask, },
-	{ sd_init_ALLNODES, cpu_allnodes_mask, },
-#endif
 	{ NULL, },
 };
 
 static struct sched_domain_topology_level *sched_domain_topology = default_topology;
 
+#ifdef CONFIG_NUMA
+
+static int sched_domains_numa_levels;
+static int *sched_domains_numa_distance;
+static struct cpumask ** __percpu sched_domains_numa_masks;
+static int sched_domains_curr_level;
+
+static struct sched_domain *
+sd_init_NUMA(struct sched_domain_topology_level *tl, int cpu)
+{
+	sched_domains_curr_level = tl->numa_level;
+	return sd_init_NODE(tl, cpu);
+}
+
+static const struct cpumask *sd_numa_mask(int cpu)
+{
+	return per_cpu_ptr(sched_domains_numa_masks[sched_domains_curr_level], cpu);
+}
+
+static void sched_init_numa(void)
+{
+	int next_distance, curr_distance = node_distance(0, 0);
+	struct sched_domain_topology_level *tl;
+	int level = 0;
+	int i, j, k;
+	char str[256];
+
+	sched_domains_numa_distance =
+		kzalloc(sizeof(int) * nr_node_ids, GFP_KERNEL);
+	if (!sched_domains_numa_distance)
+		return;
+
+	next_distance = curr_distance;
+	for (i = 0; i < nr_node_ids; i++) {
+		for (j = 0; j < nr_node_ids; j++) {
+			int distance = node_distance(0, j);
+			printk("distance(0,%d): %d\n", j, distance);
+			if (distance > curr_distance &&
+					(distance < next_distance ||
+					 next_distance == curr_distance))
+				next_distance = distance;
+		}
+		if (next_distance != curr_distance) {
+			sched_domains_numa_distance[level++] = next_distance;
+			sched_domains_numa_levels = level;
+			curr_distance = next_distance;
+		} else break;
+	}
+
+	sched_domains_numa_masks = kzalloc(sizeof(void *) * level, GFP_KERNEL);
+	if (!sched_domains_numa_masks)
+		return;
+
+	printk("numa levels: %d\n", level);
+	for (i = 0; i < level; i++) {
+		printk("numa distance(%d): %d\n", i, sched_domains_numa_distance[i]);
+
+		sched_domains_numa_masks[i] = alloc_percpu(cpumask_t);
+		if (!sched_domains_numa_masks[i])
+			return;
+
+		for_each_possible_cpu(j) {
+			struct cpumask *mask =
+				per_cpu_ptr(sched_domains_numa_masks[i], j);
+
+			for (k = 0; k < nr_node_ids; k++) {
+				if (node_distance(cpu_to_node(j), k) >
+						sched_domains_numa_distance[i])
+					continue;
+
+				cpumask_or(mask, mask, cpumask_of_node(k));
+			}
+
+			cpulist_scnprintf(str, sizeof(str), mask);
+			printk("numa cpu(%d) mask: %s\n", j, str);
+		}
+	}
+
+	tl = kzalloc((ARRAY_SIZE(default_topology) + level) *
+			sizeof(struct sched_domain_topology_level), GFP_KERNEL);
+	if (!tl)
+		return;
+
+	sched_domain_topology = tl;
+	for (i = 0; default_topology[i].init; i++)
+		tl[i] = default_topology[i];
+
+	for (j = 0; j < level; i++, j++) {
+		tl[i] = (struct sched_domain_topology_level){
+			.init = sd_init_NUMA,
+			.mask = sd_numa_mask,
+			.numa_level = j,
+		};
+	}
+
+	for (tl = sched_domain_topology; tl->init; tl++) {
+		printk("Topology: %pF\n", tl->init);
+	}
+
+	return;
+}
+#else
+static inline void sched_init_numa(void)
+{
+}
+#endif /* CONFIG_NUMA */
+
 static int __sdt_alloc(const struct cpumask *cpu_map)
 {
 	struct sched_domain_topology_level *tl;
@@ -7578,6 +7595,8 @@ void __init sched_init_smp(void)
 	alloc_cpumask_var(&non_isolated_cpus, GFP_KERNEL);
 	alloc_cpumask_var(&fallback_doms, GFP_KERNEL);
 
+	sched_init_numa();
+
 	get_online_cpus();
 	mutex_lock(&sched_domains_mutex);
 	init_sched_domains(cpu_active_mask);

Re: [PATCH 00/23] sched: Rewrite sched_domain/sched_group creation

[Andreas Herrmann]

On Thu, Apr 07, 2011 at 08:09:41AM -0400, Peter Zijlstra wrote:
> This series rewrite the sched_domain and sched_group creation code.
> 
> While its still not completely finished it does get us a lot of cleanups
> and code reduction and seems fairly stable at this point and should thus
> be a fairly good base to continue from.

Hi Peter,

finally I've reviewed the entire patch set.
Didn't find suspicious things.
So if you care you can add a
Reviewed-by: Andreas Herrmann <andreas.herrmann3@amd.com>
to each of the 23 patches.

> Also available through:
>   git://git.kernel.org/pub/scm/linux/kernel/git/peterz/linux-2.6-sched.git sched_domain

I've also tested that stuff on a Magny-Cours and other systems.  Seems
to be quite stable so far. (Especially also on multi-node CPUs;-)
(I've used your branch rebased (w/o issues) on -rc5.)


Regards,

Andreas

-- 
Operating | Advanced Micro Devices GmbH
  System  | Einsteinring 24, 85609 Dornach b. München, Germany
 Research | Geschäftsführer: Alberto Bozzo, Andrew Bowd
  Center  | Sitz: Dornach, Gemeinde Aschheim, Landkreis München
  (OSRC)  | Registergericht München, HRB Nr. 43632