mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Li Zefan <lizefan@huawei.com>
To: Tejun Heo <tj@kernel.org>
Cc: LKML <linux-kernel@vger.kernel.org>, cgroups <cgroups@vger.kernel.org>
Subject: [PATCH v2 03/12] cpuset: update cs->effective_{cpus,mems} when config changes
Date: Fri, 11 Oct 2013 17:50:04 +0800	[thread overview]
Message-ID: <5257C9CC.3080806@huawei.com> (raw)
In-Reply-To: <5257C9A1.1030404@huawei.com>

We're going to have separate user-configured masks and effective ones.

Eventually configured masks can only be changed by writing cpuset.cpus
and cpuset.mems, and they won't be restricted by parent cpuset. While
effective masks reflect cpu/memory hotplug and hierachical restriction,
and these are the real masks that apply to the tasks in the cpuset.

We calculate effective mask this way:
  - top cpuset's effective_mask == online_mask, otherwise
  - cpuset's effective_mask == configured_mask & parent effective_mask,
    if the result is empty, it inherits parent effective mask.

Those behavior changes are for sane_behavior only. For !sane_behavior
effective_mask and configured_mask are the same, so we won't break old
interfaces.

To make cs->effective_{cpus,mems} to be effective masks, we need to
  - change the effective masks at hotplug
  - change the effective masks at config change
  - take on ancestor's mask when the effective mask is empty

The second item is done here. We don't need to treat root_cs specially
in update_cpumasks_hier(). While at it, remove the redundant variable
is_load_balanced.

This won't introduce behavior change.

v2:
- revise the comment in update_{cpu,node}masks_hier(), suggested by Tejun.
- fix to use @cp instead of @cs in these two functions.

Signed-off-by: Li Zefan <lizefan@huawei.com>
---
 kernel/cpuset.c | 115 ++++++++++++++++++++++++++++++++------------------------
 1 file changed, 66 insertions(+), 49 deletions(-)

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index d0ccde2..bdc6047 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -879,39 +879,49 @@ static void update_tasks_cpumask(struct cpuset *cs, struct ptr_heap *heap)
 	css_scan_tasks(&cs->css, NULL, cpuset_change_cpumask, cs, heap);
 }
 
-/*
- * update_tasks_cpumask_hier - Update the cpumasks of tasks in the hierarchy.
- * @root_cs: the root cpuset of the hierarchy
- * @update_root: update root cpuset or not?
+/**
+ * update_cpumasks_hier - Update effective cpumasks and tasks in the subtree
+ * @cs: the cpuset to consider
+ * @trialcs: the trial cpuset
  * @heap: the heap used by css_scan_tasks()
  *
- * This will update cpumasks of tasks in @root_cs and all other empty cpusets
- * which take on cpumask of @root_cs.
- *
- * Called with cpuset_mutex held
+ * When configured cpumask is changed, the effective cpumasks of this cpuset
+ * and all its descendants need to be updated.
  */
-static void update_tasks_cpumask_hier(struct cpuset *root_cs,
-				      bool update_root, struct ptr_heap *heap)
+static void update_cpumasks_hier(struct cpuset *cs, struct cpuset *trialcs,
+				 struct ptr_heap *heap)
 {
-	struct cpuset *cp;
 	struct cgroup_subsys_state *pos_css;
+	struct cpuset *cp;
 
 	rcu_read_lock();
-	cpuset_for_each_descendant_pre(cp, pos_css, root_cs) {
-		if (cp == root_cs) {
-			if (!update_root)
-				continue;
-		} else {
-			/* skip the whole subtree if @cp have some CPU */
-			if (!cpumask_empty(cp->cpus_allowed)) {
-				pos_css = css_rightmost_descendant(pos_css);
-				continue;
-			}
+	cpuset_for_each_descendant_pre(cp, pos_css, cs) {
+		struct cpuset *parent = parent_cs(cp);
+		struct cpumask *new_cpus = trialcs->effective_cpus;
+
+		cpumask_and(new_cpus, cp->cpus_allowed,
+			    parent->effective_cpus);
+
+		/*
+		 * Skip the whole subtree if the cpumask remains the same
+		 * and isn't empty. If it's empty, we need to update tasks
+		 * to take on an ancestor's cpumask.
+		 */
+		if (cpumask_equal(new_cpus, cp->effective_cpus) &&
+		    ((cp == cs) || !cpumask_empty(new_cpus))) {
+			pos_css = css_rightmost_descendant(pos_css);
+			continue;
 		}
+
 		if (!css_tryget(&cp->css))
 			continue;
+
 		rcu_read_unlock();
 
+		mutex_lock(&callback_mutex);
+		cpumask_copy(cp->effective_cpus, new_cpus);
+		mutex_unlock(&callback_mutex);
+
 		update_tasks_cpumask(cp, heap);
 
 		rcu_read_lock();
@@ -930,7 +940,6 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
 {
 	struct ptr_heap heap;
 	int retval;
-	int is_load_balanced;
 
 	/* top_cpuset.cpus_allowed tracks cpu_online_mask; it's read-only */
 	if (cs == &top_cpuset)
@@ -965,17 +974,15 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
 	if (retval)
 		return retval;
 
-	is_load_balanced = is_sched_load_balance(trialcs);
-
 	mutex_lock(&callback_mutex);
 	cpumask_copy(cs->cpus_allowed, trialcs->cpus_allowed);
 	mutex_unlock(&callback_mutex);
 
-	update_tasks_cpumask_hier(cs, true, &heap);
+	update_cpumasks_hier(cs, trialcs, &heap);
 
 	heap_free(&heap);
 
-	if (is_load_balanced)
+	if (is_sched_load_balance(cs))
 		rebuild_sched_domains_locked();
 	return 0;
 }
@@ -1136,40 +1143,50 @@ static void update_tasks_nodemask(struct cpuset *cs, struct ptr_heap *heap)
 	cpuset_being_rebound = NULL;
 }
 
-/*
- * update_tasks_nodemask_hier - Update the nodemasks of tasks in the hierarchy.
- * @cs: the root cpuset of the hierarchy
- * @update_root: update the root cpuset or not?
+/**
+ * update_nodesmasks_hier - Update effective nodemasks and tasks in the subtree
+ * @cs: the cpuset to consider
+ * @trialcs: the trial cpuset
  * @heap: the heap used by css_scan_tasks()
  *
- * This will update nodemasks of tasks in @root_cs and all other empty cpusets
- * which take on nodemask of @root_cs.
- *
- * Called with cpuset_mutex held
+ * When configured nodemask is changed, the effective nodemasks of this cpuset
+ * and all its descendants need to be updated.
  */
-static void update_tasks_nodemask_hier(struct cpuset *root_cs,
-				       bool update_root, struct ptr_heap *heap)
+static void update_nodemasks_hier(struct cpuset *cs, struct cpuset *trialcs,
+				 struct ptr_heap *heap)
 {
-	struct cpuset *cp;
 	struct cgroup_subsys_state *pos_css;
+	struct cpuset *cp;
 
 	rcu_read_lock();
-	cpuset_for_each_descendant_pre(cp, pos_css, root_cs) {
-		if (cp == root_cs) {
-			if (!update_root)
-				continue;
-		} else {
-			/* skip the whole subtree if @cp have some CPU */
-			if (!nodes_empty(cp->mems_allowed)) {
-				pos_css = css_rightmost_descendant(pos_css);
-				continue;
-			}
+	cpuset_for_each_descendant_pre(cp, pos_css, cs) {
+		struct cpuset *parent = parent_cs(cp);
+		nodemask_t *new_mems = &trialcs->effective_mems;
+
+		nodes_and(*new_mems, cp->mems_allowed,
+			  parent->effective_mems);
+
+		/*
+		 * Skip the whole subtree if the nodemask remains the same
+		 * and isn't empty. If it's empty, we need to update tasks
+		 * to take on an ancestor's nodemask.
+		 */
+		if (nodes_equal(*new_mems, cp->effective_mems) &&
+		    ((cp == cs) || !nodes_empty(*new_mems))) {
+			pos_css = css_rightmost_descendant(pos_css);
+			continue;
 		}
+
 		if (!css_tryget(&cp->css))
 			continue;
+
 		rcu_read_unlock();
 
-		update_tasks_nodemask(cp, heap);
+		mutex_lock(&callback_mutex);
+		cp->effective_mems = *new_mems;
+		mutex_unlock(&callback_mutex);
+
+		update_tasks_cpumask(cp, heap);
 
 		rcu_read_lock();
 		css_put(&cp->css);
@@ -1241,7 +1258,7 @@ static int update_nodemask(struct cpuset *cs, struct cpuset *trialcs,
 	cs->mems_allowed = trialcs->mems_allowed;
 	mutex_unlock(&callback_mutex);
 
-	update_tasks_nodemask_hier(cs, true, &heap);
+	update_nodemasks_hier(cs, trialcs, &heap);
 
 	heap_free(&heap);
 done:
-- 
1.8.0.2


  parent reply	other threads:[~2013-10-11  9:50 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-11  9:49 [PATCH v2 00/12] cpuset: separate configured masks and effective masks Li Zefan
2013-10-11  9:49 ` [PATCH v2 01/12] cpuset: add cs->effective_cpus and cs->effective_mems Li Zefan
2013-10-11  9:49 ` [PATCH v2 02/12] cpuset: update cpuset->effective_{cpus,mems} at hotplug Li Zefan
2013-10-11  9:50 ` Li Zefan [this message]
2013-10-15 15:18   ` [PATCH v2 03/12] cpuset: update cs->effective_{cpus,mems} when config changes Tejun Heo
2013-10-11  9:50 ` [PATCH v2 04/12] cpuset: inherit ancestor's masks if effective_{cpus,mems} becomes empty Li Zefan
2013-10-11  9:50 ` [PATCH v2 05/12] cpuset: use effective cpumask to build sched domains Li Zefan
2013-10-15 15:25   ` Tejun Heo
2013-10-11  9:50 ` [PATCH v2 06/12] cpuset: initialize top_cpuset's configured masks at mount Li Zefan
2013-10-11  9:51 ` [PATCH v2 07/12] cpuset: apply cs->effective_{cpus,mems} Li Zefan
2013-10-11  9:51 ` [PATCH v2 08/12] cpuset: make cs->{cpus,mems}_allowed as user-configured masks Li Zefan
2013-10-11  9:51 ` [PATCH v2 09/12] cpuset: refactor cpuset_hotplug_update_tasks() Li Zefan
2013-10-11  9:51 ` [PATCH v2 10/12] cpuset: enable onlined cpu/node in effective masks Li Zefan
2013-10-11  9:51 ` [PATCH v2 11/12] cpuset: allow writing offlined masks to cpuset.cpus/mems Li Zefan
2013-10-15 15:36   ` Tejun Heo
2013-10-11  9:52 ` [PATCH v2 12/12] cpuset: export effective masks to userspace Li Zefan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=5257C9CC.3080806@huawei.com \
    --to=lizefan@huawei.com \
    --cc=cgroups@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=tj@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

Powered by JetHome