linux-kernel - [RFC PATCH] configfs: make it not be a module any more

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Tue, 9 Nov 2010 17:38:43 +0800
From:	Américo Wang <xiyou.wangcong@...il.com>
To:	Américo Wang <xiyou.wangcong@...il.com>
Cc:	Mike Waychison <mikew@...gle.com>, simon.kagstrom@...insight.net,
	davem@...emloft.net, nhorman@...driver.com,
	Matt Mackall <mpm@...enic.com>, adurbin@...gle.com,
	linux-kernel@...r.kernel.org, chavey@...gle.com,
	Greg KH <greg@...ah.com>, akpm@...ux-foundation.org,
	linux-api@...r.kernel.org
Subject: [RFC PATCH] configfs: make it not be a module any more

On Tue, Nov 09, 2010 at 05:06:45PM +0800, Américo Wang wrote:
>On Tue, Nov 09, 2010 at 12:34:24AM -0800, Mike Waychison wrote:
>>On Mon, Nov 8, 2010 at 8:27 PM, Américo Wang <xiyou.wangcong@...il.com> wrote:
>>> On Tue, Nov 09, 2010 at 11:30:24AM +0800, Américo Wang wrote:
>>>>
>>> ....
>>>>
>>>>So, either we need to de-modulize configfs or replace configfs API
>>>>with sysfs API. Personally, I prefer the former one, I don't think
>>>>configfs should be a module as long as it can provide API's
>>>>for other subsystems, like debugfs.
>>>>
>>>
>>> To clarify, I meant "as long as the API it provides can be used by
>>> other core subsystems".
>>>
>>
>>Ya, I see the problem with it being a tristate.
>>
>>Why not just make netconsole support being compiled in force configfs
>>to be compiled in?  Or does that just set bad precedent?
>
>That is what netconsole does now, and this is fine, since netconsole is
>a module too, however, after you move that code into netpoll, then netpoll
>will have a dependence on it, we will have problems.
>
>I think we can let NETPOLL_TARGET depend on CONFIGFS_FS=y, but I still
>see no reason why CONFIGFS_FS should be a module.
>

Okay, here we go.

------------------>

Netpoll will use configfs API's to provide generic
netpoll target support. These API's provided by
configfs could also be used by other core subsystems.
So, make configfs not a module any more, like debugfs.

Signed-off-by: WANG Cong <xiyou.wangcong@...il.com>

---
diff --git a/fs/configfs/Kconfig b/fs/configfs/Kconfig
index 13587cc..4bdfe1e 100644
--- a/fs/configfs/Kconfig
+++ b/fs/configfs/Kconfig
@@ -1,5 +1,5 @@
 config CONFIGFS_FS
-	tristate "Userspace-driven configuration filesystem"
+	bool "Userspace-driven configuration filesystem"
 	depends on SYSFS
 	help
 	  configfs is a ram-based filesystem that provides the converse
diff --git a/fs/configfs/configfs_internal.h b/fs/configfs/configfs_internal.h
index da6061a..714a192 100644
--- a/fs/configfs/configfs_internal.h
+++ b/fs/configfs/configfs_internal.h
@@ -66,7 +66,6 @@ extern int configfs_is_root(struct config_item *item);
 extern struct inode * configfs_new_inode(mode_t mode, struct configfs_dirent *);
 extern int configfs_create(struct dentry *, int mode, int (*init)(struct inode *));
 extern int configfs_inode_init(void);
-extern void configfs_inode_exit(void);
 
 extern int configfs_create_file(struct config_item *, const struct configfs_attribute *);
 extern int configfs_make_dirent(struct configfs_dirent *,
diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c
index 253476d..3508bc1 100644
--- a/fs/configfs/inode.c
+++ b/fs/configfs/inode.c
@@ -295,7 +295,3 @@ int __init configfs_inode_init(void)
 	return bdi_init(&configfs_backing_dev_info);
 }
 
-void __exit configfs_inode_exit(void)
-{
-	bdi_destroy(&configfs_backing_dev_info);
-}
diff --git a/fs/configfs/mount.c b/fs/configfs/mount.c
index 7d3607f..1cf57b3 100644
--- a/fs/configfs/mount.c
+++ b/fs/configfs/mount.c
@@ -111,7 +111,6 @@ static struct dentry *configfs_do_mount(struct file_system_type *fs_type,
 }
 
 static struct file_system_type configfs_fs_type = {
-	.owner		= THIS_MODULE,
 	.name		= "configfs",
 	.mount		= configfs_do_mount,
 	.kill_sb	= kill_litter_super,
@@ -167,20 +166,5 @@ static int __init configfs_init(void)
 out:
 	return err;
 }
+core_initcall(configfs_init);
 
-static void __exit configfs_exit(void)
-{
-	unregister_filesystem(&configfs_fs_type);
-	kobject_put(config_kobj);
-	kmem_cache_destroy(configfs_dir_cachep);
-	configfs_dir_cachep = NULL;
-	configfs_inode_exit();
-}
-
-MODULE_AUTHOR("Oracle");
-MODULE_LICENSE("GPL");
-MODULE_VERSION("0.0.2");
-MODULE_DESCRIPTION("Simple RAM filesystem for user driven kernel subsystem configuration.");
-
-module_init(configfs_init);
-module_exit(configfs_exit);

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