Skip to contentSkip to conversation

cgroup: fix default notify_on_release setting

PATCH In review Li Zefan 5 participants 5 messages Feb 25, 2008 – Feb 26, 2008 2 messages not in the archive

lore
Share & tools
lore

Li Zefan

author
#0 lore raw

PATCH 1/1 cgroup: fix default notify_on_release setting

The documentation says the default value of notify_on_release of a child cgroup is inherited from its parent, which is reasonable, but the implementation just sets the flag disabled.

Signed-off-by Li Zefan <[email protected]>
---
 kernel/cgroup.c |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)
kernel/cgroup.c +3 -1
index d8abe99..e9c2fb0 100644
@@ -2232,7 +2232,6 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry,
 
 	mutex_lock(&cgroup_mutex);
 
-	cgrp->flags = 0;
 	INIT_LIST_HEAD(&cgrp->sibling);
 	INIT_LIST_HEAD(&cgrp->children);
 	INIT_LIST_HEAD(&cgrp->css_sets);

@@ -2242,6 +2241,9 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry,
 	cgrp->root = parent->root;
 	cgrp->top_cgroup = parent->top_cgroup;
 
+	if (notify_on_release(parent))
+		set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);
+
 	for_each_subsys(root, ss) {
 		struct cgroup_subsys_state *css = ss->create(ss, cgrp);
 		if (IS_ERR(css)) {
Signature · 1 line
1.5.4.rc3

A message that is not in the archive (sent off-list or not yet received): [email protected]

Acked-by
#2 lore raw
Quoted 5 lines · Li Zefan

The documentation says the default value of notify_on_release of
a child cgroup is inherited from its parent, which is reasonable, but the implementation just sets the flag disabled.

Signed-off-by Li Zefan <[email protected]>
Acked-by Paul Menage <[email protected]>

Yes, I guess it makes sense to follow the original cpusets behaviour. I think that got lost when the notify-on-release functionality was temporarily removed during cgroups development.

Quoted 30 lines

---

  kernel/cgroup.c |    4 +++-
  1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index d8abe99..e9c2fb0 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -2232,7 +2232,6 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry,

        mutex_lock(&cgroup_mutex);
 -       cgrp->flags = 0;
        INIT_LIST_HEAD(&cgrp->sibling);
        INIT_LIST_HEAD(&cgrp->children);
        INIT_LIST_HEAD(&cgrp->css_sets);
 @@ -2242,6 +2241,9 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry,
        cgrp->root = parent->root;
        cgrp->top_cgroup = parent->top_cgroup;

+ if (notify_on_release(parent))
+ set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);
+

        for_each_subsys(root, ss) {
                struct cgroup_subsys_state *css = ss->create(ss, cgrp);
                if (IS_ERR(css)) {

--
1.5.4.rc3

Acked-by on v1 1/1
#3 lore raw
In reply to Li Zefan
Quoted 5 lines · Li Zefan

The documentation says the default value of notify_on_release of
a child cgroup is inherited from its parent, which is reasonable, but the implementation just sets the flag disabled.

Signed-off-by Li Zefan <[email protected]>
Acked-by Paul Menage <[email protected]>

Yes, I guess it makes sense to follow the original cpusets behaviour. I think that got lost when the notify-on-release functionality was temporarily removed during cgroups development.

Quoted 30 lines

---

  kernel/cgroup.c |    4 +++-
  1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index d8abe99..e9c2fb0 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -2232,7 +2232,6 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry,

        mutex_lock(&cgroup_mutex);
 -       cgrp->flags = 0;
        INIT_LIST_HEAD(&cgrp->sibling);
        INIT_LIST_HEAD(&cgrp->children);
        INIT_LIST_HEAD(&cgrp->css_sets);
 @@ -2242,6 +2241,9 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry,
        cgrp->root = parent->root;
        cgrp->top_cgroup = parent->top_cgroup;

+ if (notify_on_release(parent))
+ set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);
+

        for_each_subsys(root, ss) {
                struct cgroup_subsys_state *css = ss->create(ss, cgrp);
                if (IS_ERR(css)) {

--
1.5.4.rc3

A message that is not in the archive (sent off-list or not yet received): 6599ad830802242222p7d42d2b2ue7c585eae8f3bed-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org

#5 lore raw
Quoted 6 lines

@@ -2242,6 +2241,9 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry,

        cgrp->root = parent->root;
        cgrp->top_cgroup = parent->top_cgroup;

+ if (notify_on_release(parent))
+ set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);

Good catch, Li Zefan - thanks.

Signature · 3 lines
                  I won't rest till it's the best ...
                  Programmer, Linux Scalability
                  Paul Jackson <pj-sJ/[email protected]> 1.940.382.4214
#6 lore raw
In reply to Paul Menage
Quoted 6 lines

@@ -2242,6 +2241,9 @@ static long cgroup_create(struct cgroup *parent, struct dentry *dentry,

        cgrp->root = parent->root;
        cgrp->top_cgroup = parent->top_cgroup;

+ if (notify_on_release(parent))
+ set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);

Good catch, Li Zefan - thanks.

Signature · 3 lines
                  I won't rest till it's the best ...
                  Programmer, Linux Scalability
                  Paul Jackson <[email protected]> 1.940.382.4214