Skip to contentSkip to conversation

kernel/ksysfs.c:fix dependence on CONFIG_NET

PATCH In review [email protected] 2 participants 4 messages Nov 16, 2008 – Nov 19, 2008 Driver core

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 kernel/ksysfs.c:fix dependence on CONFIG_NET

From: Ming Lei <[email protected]>

Access to uevent_seqnum and uevent_helper does not need to depend on CONFIG_NET, so remove it.

---
 kernel/ksysfs.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)
kernel/ksysfs.c +2 -2
index 08dd8ed..528dd78 100644
@@ -24,7 +24,7 @@ static struct kobj_attribute _name##_attr = __ATTR_RO(_name)
 static struct kobj_attribute _name##_attr = \
 	__ATTR(_name, 0644, _name##_show, _name##_store)
 
-#if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET)
+#if defined(CONFIG_HOTPLUG)
 /* current uevent sequence number */
 static ssize_t uevent_seqnum_show(struct kobject *kobj,
 				  struct kobj_attribute *attr, char *buf)
@@ -137,7 +137,7 @@ struct kobject *kernel_kobj;
 EXPORT_SYMBOL_GPL(kernel_kobj);
 
 static struct attribute * kernel_attrs[] = {
-#if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET)
+#if defined(CONFIG_HOTPLUG)
 	&uevent_seqnum_attr.attr,
 	&uevent_helper_attr.attr,
 #endif
Signature · 1 line
1.6.0
#1 lore raw
Quoted 24 lines · On Sun, 16 Nov 2008 18:22:09 +0800 [email protected] wrote:

From: Ming Lei <[email protected]>

Access to uevent_seqnum and uevent_helper does not need to depend on CONFIG_NET, so remove it.

---
 kernel/ksysfs.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)
kernel/ksysfs.c +1 -1
index 08dd8ed..528dd78 100644
@@ -24,7 +24,7 @@ static struct kobj_attribute _name##_attr = __ATTR_RO(_name)
 static struct kobj_attribute _name##_attr = \
 	__ATTR(_name, 0644, _name##_show, _name##_store)
 
-#if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET)
+#if defined(CONFIG_HOTPLUG)
 /* current uevent sequence number */
 static ssize_t uevent_seqnum_show(struct kobject *kobj,
 				  struct kobj_attribute *attr, char *buf)

kernel/ksysfs.c struct kobject *kernel_kobj; · line 143

Quoted 8 lines
kernel/ksysfs.c +1 -1
@@ -137,7 +137,7 @@ struct kobject *kernel_kobj;
 EXPORT_SYMBOL_GPL(kernel_kobj);
 
 static struct attribute * kernel_attrs[] = {
-#if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET)
+#if defined(CONFIG_HOTPLUG)
 	&uevent_seqnum_attr.attr,
 	&uevent_helper_attr.attr,
 #endif

I think the story here is that uevent uses netlink, and with CONFIG_NET=n, we have no netlink, so that code couldn't do anything anyway.

#2 lore raw
Quoted 29 lines · On Tue, 18 Nov 2008 15:38:18 -0800 Andrew Morton <[email protected]>…
Quoted 26 lines · On Sun, 16 Nov 2008 18:22:09 +0800 [email protected] wrote:

From: Ming Lei <[email protected]>

Access to uevent_seqnum and uevent_helper does not need to depend on CONFIG_NET, so remove it.

---
 kernel/ksysfs.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)
kernel/ksysfs.c +1 -1
index 08dd8ed..528dd78 100644
@@ -24,7 +24,7 @@ static struct kobj_attribute _name##_attr = __ATTR_RO(_name)
 static struct kobj_attribute _name##_attr = \
 	__ATTR(_name, 0644, _name##_show, _name##_store)
 
-#if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET)
+#if defined(CONFIG_HOTPLUG)
 /* current uevent sequence number */
 static ssize_t uevent_seqnum_show(struct kobject *kobj,
 				  struct kobj_attribute *attr, char *buf)
@@ -137,7 +137,7 @@ struct kobject *kernel_kobj;
 EXPORT_SYMBOL_GPL(kernel_kobj);
 
Quoted 11 lines

kernel/ksysfs.c struct kobject *kernel_kobj; · line 143

Quoted 6 lines
kernel/ksysfs.c +1 -1
 static struct attribute * kernel_attrs[] = {
-#if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET)
+#if defined(CONFIG_HOTPLUG)
 	&uevent_seqnum_attr.attr,
 	&uevent_helper_attr.attr,
 #endif

I think the story here is that uevent uses netlink, and with CONFIG_NET=n, we have no netlink, so that code couldn't do anything anyway.

Or not. Perhaps we can still use uevent_helper in this case?

(Getting a bit tired of writing people's changelogs for them!)

author
#3 lore raw

2008/11/19 Andrew Morton <[email protected]>:

Quoted 44 lines
Quoted 29 lines · On Tue, 18 Nov 2008 15:38:18 -0800 Andrew Morton <[email protected]>…
Quoted 26 lines · On Sun, 16 Nov 2008 18:22:09 +0800 [email protected] wrote:

From: Ming Lei <[email protected]>

Access to uevent_seqnum and uevent_helper does not need to depend on CONFIG_NET, so remove it.

---
 kernel/ksysfs.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)
kernel/ksysfs.c +1 -1
index 08dd8ed..528dd78 100644
@@ -24,7 +24,7 @@ static struct kobj_attribute _name##_attr = __ATTR_RO(_name)
 static struct kobj_attribute _name##_attr = \
    __ATTR(_name, 0644, _name##_show, _name##_store)
 
-#if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET)
+#if defined(CONFIG_HOTPLUG)
 /* current uevent sequence number */
 static ssize_t uevent_seqnum_show(struct kobject *kobj,
                              struct kobj_attribute *attr, char *buf)
@@ -137,7 +137,7 @@ struct kobject *kernel_kobj;
 EXPORT_SYMBOL_GPL(kernel_kobj);
 
Quoted 11 lines

kernel/ksysfs.c struct kobject *kernel_kobj; · line 143

Quoted 6 lines
kernel/ksysfs.c +1 -1
 static struct attribute * kernel_attrs[] = {
-#if defined(CONFIG_HOTPLUG) && defined(CONFIG_NET)
+#if defined(CONFIG_HOTPLUG)
    &uevent_seqnum_attr.attr,
    &uevent_helper_attr.attr,
 #endif

I think the story here is that uevent uses netlink, and with CONFIG_NET=n, we have no netlink, so that code couldn't do anything anyway.

Quoted 1 line

Or not. Perhaps we can still use uevent_helper in this case?

yes.

Quoted 3 lines

(Getting a bit tired of writing people's changelogs for them!)

Signature · 1 line
Lei Ming