Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1205051 > unrolled thread

[PATCH v2 0/3] Disable sched_numa_balancing on uma systems

Back to article view | Back to linux.kernel

Contents

  [PATCH v2 0/3] Disable sched_numa_balancing on uma systems Srikar Dronamraju <srikar@linux.vnet.ibm.com> - 2015-08-11 13:10 +0200
    [PATCH v2 3/3] sched/numa: Remove NUMA sched feature Srikar Dronamraju <srikar@linux.vnet.ibm.com> - 2015-08-11 13:10 +0200
      Re: [PATCH v2 3/3] sched/numa: Remove NUMA sched feature Peter Zijlstra <peterz@infradead.org> - 2015-08-11 13:20 +0200
    [PATCH v2 2/3] sched/numa: Disable sched_numa_balancing on uma systems Srikar Dronamraju <srikar@linux.vnet.ibm.com> - 2015-08-11 13:10 +0200
    [PATCH v2 4/4] sched/numa: Convert sched_numa_balancing to a static_branch Srikar Dronamraju <srikar@linux.vnet.ibm.com> - 2015-08-11 18:30 +0200

#1205051 — [PATCH v2 0/3] Disable sched_numa_balancing on uma systems

With recent commit 2a1ed24 ("sched/numa: Prefer NUMA hotness over cache
hotness") sets sched feature NUMA to true. This can enable numa hinting
faults on a uma system.

This patchset ensures that numa hinting faults occur only on a numa system
by setting/resetting sched_numa_balancing.

This patchset
- Renames numabalancing_enabled to sched_numa_balancing
- Makes sched_numa_balancing common to CONFIG_SCHED_DEBUG and
  !CONFIG_SCHED_DEBUG. Earlier it was only in !CONFIG_SCHED_DEBUG
- Checks for sched_numa_balancing instead of sched_feat(NUMA)
- Removes NUMA sched feature

Srikar Dronamraju (3):
  sched/numa: Rename numabalancing_enabled to sched_numa_balancing
  sched/numa: Disable sched_numa_balancing on uma systems
  sched/numa: Remove NUMA sched feature

 kernel/sched/core.c     | 16 +++-------------
 kernel/sched/fair.c     |  8 ++++----
 kernel/sched/features.h | 16 ----------------
 kernel/sched/sched.h    | 10 ++--------
 4 files changed, 9 insertions(+), 41 deletions(-)

-- 
1.8.3.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1205054 — [PATCH v2 3/3] sched/numa: Remove NUMA sched feature

Variable sched_numa_balancing is available for both CONFIG_SCHED_DEBUG
and !CONFIG_SCHED_DEBUG. All code paths now check for
sched_numa_balancing. Hence remove sched_feat(NUMA).

Suggested-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
---
 kernel/sched/core.c     |  6 ------
 kernel/sched/features.h | 16 ----------------
 2 files changed, 22 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 7cbdf44..d02570b 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2063,12 +2063,6 @@ __read_mostly bool sched_numa_balancing;
 void set_numabalancing_state(bool enabled)
 {
 	sched_numa_balancing = enabled;
-#ifdef CONFIG_SCHED_DEBUG
-	if (enabled)
-		sched_feat_set("NUMA");
-	else
-		sched_feat_set("NO_NUMA");
-#endif /* CONFIG_SCHED_DEBUG */
 }
 
 #ifdef CONFIG_PROC_SYSCTL
diff --git a/kernel/sched/features.h b/kernel/sched/features.h
index 83a50e7..8baa708 100644
--- a/kernel/sched/features.h
+++ b/kernel/sched/features.h
@@ -72,19 +72,3 @@ SCHED_FEAT(RT_PUSH_IPI, true)
 SCHED_FEAT(FORCE_SD_OVERLAP, false)
 SCHED_FEAT(RT_RUNTIME_SHARE, true)
 SCHED_FEAT(LB_MIN, false)
-
-/*
- * Apply the automatic NUMA scheduling policy. Enabled automatically
- * at runtime if running on a NUMA machine. Can be controlled via
- * numa_balancing=
- */
-#ifdef CONFIG_NUMA_BALANCING
-
-/*
- * NUMA will favor moving tasks towards nodes where a higher number of
- * hinting faults are recorded during active load balancing. It will
- * resist moving tasks towards nodes where a lower number of hinting
- * faults have been recorded.
- */
-SCHED_FEAT(NUMA,	true)
-#endif
-- 
1.8.3.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1205061 — Re: [PATCH v2 3/3] sched/numa: Remove NUMA sched feature

On Tue, Aug 11, 2015 at 04:30:13PM +0530, Srikar Dronamraju wrote:
> Variable sched_numa_balancing is available for both CONFIG_SCHED_DEBUG
> and !CONFIG_SCHED_DEBUG. All code paths now check for
> sched_numa_balancing. Hence remove sched_feat(NUMA).
> 
> Suggested-by: Ingo Molnar <mingo@kernel.org>
> Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
> ---
>  kernel/sched/core.c     |  6 ------
>  kernel/sched/features.h | 16 ----------------
>  2 files changed, 22 deletions(-)
> 
> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index 7cbdf44..d02570b 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -2063,12 +2063,6 @@ __read_mostly bool sched_numa_balancing;
>  void set_numabalancing_state(bool enabled)
>  {
>  	sched_numa_balancing = enabled;
> -#ifdef CONFIG_SCHED_DEBUG
> -	if (enabled)
> -		sched_feat_set("NUMA");
> -	else
> -		sched_feat_set("NO_NUMA");
> -#endif /* CONFIG_SCHED_DEBUG */
>  }

Could you at least replace sched_numa_balancing with a static_key
thingy?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1205055 — [PATCH v2 2/3] sched/numa: Disable sched_numa_balancing on uma systems

Commit 2a1ed24 ("sched/numa: Prefer NUMA hotness over cache hotness")
sets sched feature NUMA to true. However this can enable numa hinting
faults on a uma system.

This commit ensures that numa hinting faults occur only on a numa system
by setting/resetting sched_numa_balancing.

This commit
- Makes sched_numa_balancing common to CONFIG_SCHED_DEBUG and
  !CONFIG_SCHED_DEBUG. Earlier it was only in !CONFIG_SCHED_DEBUG
- Checks for sched_numa_balancing instead of sched_feat(NUMA)

Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
---
 kernel/sched/core.c  | 14 +++++---------
 kernel/sched/fair.c  |  4 ++--
 kernel/sched/sched.h |  6 ------
 3 files changed, 7 insertions(+), 17 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 71c1d25..7cbdf44 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2058,22 +2058,18 @@ static void __sched_fork(unsigned long clone_flags, struct task_struct *p)
 }
 
 #ifdef CONFIG_NUMA_BALANCING
-#ifdef CONFIG_SCHED_DEBUG
+__read_mostly bool sched_numa_balancing;
+
 void set_numabalancing_state(bool enabled)
 {
+	sched_numa_balancing = enabled;
+#ifdef CONFIG_SCHED_DEBUG
 	if (enabled)
 		sched_feat_set("NUMA");
 	else
 		sched_feat_set("NO_NUMA");
-}
-#else
-__read_mostly bool sched_numa_balancing;
-
-void set_numabalancing_state(bool enabled)
-{
-	sched_numa_balancing = enabled;
-}
 #endif /* CONFIG_SCHED_DEBUG */
+}
 
 #ifdef CONFIG_PROC_SYSCTL
 int sysctl_numa_balancing(struct ctl_table *table, int write,
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 3ec9b0b..f67f2bc 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5524,10 +5524,10 @@ static int migrate_degrades_locality(struct task_struct *p, struct lb_env *env)
 	unsigned long src_faults, dst_faults;
 	int src_nid, dst_nid;
 
-	if (!p->numa_faults || !(env->sd->flags & SD_NUMA))
+	if (!sched_numa_balancing)
 		return -1;
 
-	if (!sched_feat(NUMA))
+	if (!p->numa_faults || !(env->sd->flags & SD_NUMA))
 		return -1;
 
 	src_nid = cpu_to_node(env->src_cpu);
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index a02bd8d..953be0f 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1004,14 +1004,8 @@ extern struct static_key sched_feat_keys[__SCHED_FEAT_NR];
 #endif /* SCHED_DEBUG && HAVE_JUMP_LABEL */
 
 #ifdef CONFIG_NUMA_BALANCING
-#define sched_feat_numa(x) sched_feat(x)
-#ifdef CONFIG_SCHED_DEBUG
-#define sched_numa_balancing sched_feat_numa(NUMA)
-#else
 extern bool sched_numa_balancing;
-#endif /* CONFIG_SCHED_DEBUG */
 #else
-#define sched_feat_numa(x) (0)
 #define sched_numa_balancing (0)
 #endif /* CONFIG_NUMA_BALANCING */
 
-- 
1.8.3.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1205263 — [PATCH v2 4/4] sched/numa: Convert sched_numa_balancing to a static_branch

Variable sched_numa_balancing toggles numa_balancing feature. Hence
moving from a simple read mostly variable to a more apt static_branch.

Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
---
 kernel/sched/core.c  | 10 +++++++---
 kernel/sched/fair.c  |  6 +++---
 kernel/sched/sched.h |  6 +-----
 3 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index d02570b..5f330d5 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2057,12 +2057,16 @@ static void __sched_fork(unsigned long clone_flags, struct task_struct *p)
 #endif /* CONFIG_NUMA_BALANCING */
 }
 
+DEFINE_STATIC_KEY_FALSE(sched_numa_balancing);
+
 #ifdef CONFIG_NUMA_BALANCING
-__read_mostly bool sched_numa_balancing;
 
 void set_numabalancing_state(bool enabled)
 {
-	sched_numa_balancing = enabled;
+	if (enabled)
+		static_branch_enable(&sched_numa_balancing);
+	else
+		static_branch_disable(&sched_numa_balancing);
 }
 
 #ifdef CONFIG_PROC_SYSCTL
@@ -2071,7 +2075,7 @@ int sysctl_numa_balancing(struct ctl_table *table, int write,
 {
 	struct ctl_table t;
 	int err;
-	int state = sched_numa_balancing;
+	int state = static_branch_likely(&sched_numa_balancing);
 
 	if (write && !capable(CAP_SYS_ADMIN))
 		return -EPERM;
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index f67f2bc..aab730e 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -2069,7 +2069,7 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags)
 	int local = !!(flags & TNF_FAULT_LOCAL);
 	int priv;
 
-	if (!sched_numa_balancing)
+	if (!static_branch_likely(&sched_numa_balancing))
 		return;
 
 	/* for example, ksmd faulting in a user's mm */
@@ -5524,7 +5524,7 @@ static int migrate_degrades_locality(struct task_struct *p, struct lb_env *env)
 	unsigned long src_faults, dst_faults;
 	int src_nid, dst_nid;
 
-	if (!sched_numa_balancing)
+	if (!static_branch_likely(&sched_numa_balancing))
 		return -1;
 
 	if (!p->numa_faults || !(env->sd->flags & SD_NUMA))
@@ -7810,7 +7810,7 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued)
 		entity_tick(cfs_rq, se, queued);
 	}
 
-	if (sched_numa_balancing)
+	if (!static_branch_unlikely(&sched_numa_balancing))
 		task_tick_numa(rq, curr);
 }
 
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 953be0f..72cae5d 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1003,11 +1003,7 @@ extern struct static_key sched_feat_keys[__SCHED_FEAT_NR];
 #define sched_feat(x) (sysctl_sched_features & (1UL << __SCHED_FEAT_##x))
 #endif /* SCHED_DEBUG && HAVE_JUMP_LABEL */
 
-#ifdef CONFIG_NUMA_BALANCING
-extern bool sched_numa_balancing;
-#else
-#define sched_numa_balancing (0)
-#endif /* CONFIG_NUMA_BALANCING */
+extern struct static_key_false sched_numa_balancing;
 
 static inline u64 global_rt_period(void)
 {
-- 
1.8.3.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web