From mboxrd@z Thu Jan  1 00:00:00 1970
From: Lars-Peter Clausen <lars@metafoo.de>
Subject: Re: [PATCH v7 1/5] configfs: Allow dynamic group (un)registration
Date: Mon, 31 Aug 2015 18:02:43 +0200
Message-ID: <55E47AA3.3040001@metafoo.de>
References: <1439246562-17515-1-git-send-email-daniel.baluta@intel.com> <1439246562-17515-2-git-send-email-daniel.baluta@intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 8bit
Cc: knaack.h@gmx.de, linux-kernel@vger.kernel.org,
	octavian.purdila@intel.com, pebolle@tiscali.nl,
	patrick.porlan@intel.com, adriana.reus@intel.com,
	constantin.musca@intel.com, marten@intuitiveaerial.com,
	cristina.opriceana@gmail.com, pmeerw@pmeerw.net, hch@lst.de,
	viro@zeniv.linux.org.uk, akpm@linux-foundation.org
To: Daniel Baluta <daniel.baluta@intel.com>, jic23@kernel.org,
	jlbec@evilplan.org, linux-iio@vger.kernel.org,
	linux-fsdevel@vger.kernel.org
Return-path: <linux-fsdevel-owner@vger.kernel.org>
Received: from smtp-out-225.synserver.de ([212.40.185.225]:1101 "EHLO
	smtp-out-200.synserver.de" rhost-flags-OK-OK-OK-FAIL)
	by vger.kernel.org with ESMTP id S1752792AbbHaQCt (ORCPT
	<rfc822;linux-fsdevel@vger.kernel.org>);
	Mon, 31 Aug 2015 12:02:49 -0400
In-Reply-To: <1439246562-17515-2-git-send-email-daniel.baluta@intel.com>
Sender: linux-fsdevel-owner@vger.kernel.org
List-ID: <linux-fsdevel.vger.kernel.org>

[...]
> +int configfs_register_group(struct config_group *parent_group,
> +	struct config_group *group)
> +{
> +	struct dentry *dentry;
> +	int ret;
> +
> +	link_group(parent_group, group);

This needs to hold subsys->su_mutex

I think from here to before the return we need to hold the parent dentry
mutex with I_MUTEX_PARENT

> +
> +	ret = create_default_group(parent_group, group, &dentry);
> +	if (ret == 0)
> +		configfs_dir_set_ready(dentry->d_fsdata);

This needs to hold configfs_dirent_lock

> +
> +	return ret;
> +}
> +EXPORT_SYMBOL(configfs_register_group);
> +
> +void configfs_unregister_group(struct config_group *group)
> +{
> +	struct dentry *dentry = group->cg_item.ci_dentry;
> +

We probably need to do the detach_prep stuff somewhere in here.


I think this needs to hold the parent dentry mutex with I_MUTEX_PARENT from
here to ...

> +	mutex_lock(&d_inode(dentry)->i_mutex);
> +	configfs_detach_group(&group->cg_item);
> +	d_inode(dentry)->i_flags |= S_DEAD;
> +	dont_mount(dentry);
> +	mutex_unlock(&d_inode(dentry)->i_mutex);
> +
> +	d_delete(dentry);

 ... here

> +	dput(dentry);
> +	unlink_group(group);

needs subsys->su_mutex

> +}
> +EXPORT_SYMBOL(configfs_unregister_group);
> +
>  int configfs_register_subsystem(struct configfs_subsystem *subsys)
>  {
>  	int err;
> diff --git a/include/linux/configfs.h b/include/linux/configfs.h
> index c9e5c57..726980e 100644
> --- a/include/linux/configfs.h
> +++ b/include/linux/configfs.h
> @@ -251,6 +251,10 @@ static inline struct configfs_subsystem *to_configfs_subsystem(struct config_gro
>  int configfs_register_subsystem(struct configfs_subsystem *subsys);
>  void configfs_unregister_subsystem(struct configfs_subsystem *subsys);
>  
> +int configfs_register_group(struct config_group *parent_group,
> +			    struct config_group *group);
> +void configfs_unregister_group(struct config_group *group);
> +
>  /* These functions can sleep and can alloc with GFP_KERNEL */
>  /* WARNING: These cannot be called underneath configfs callbacks!! */
>  int configfs_depend_item(struct configfs_subsystem *subsys, struct config_item *target);
>