Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20210123205516.2738060-1-zenczykowski@gmail.com> <YBFu9WkVMdsfy51I@kroah.com>
In-Reply-To: <YBFu9WkVMdsfy51I@kroah.com>
From:   =?UTF-8?Q?Maciej_=C5=BBenczykowski?= <zenczykowski@gmail.com>
Date:   Wed, 27 Jan 2021 11:56:03 -0800
Message-ID: <CANP3RGcoXV+gDXjhYV5USYURHOyiZE=t6ptwp2hZb=vni3RP7g@mail.gmail.com>
Subject: Re: [PATCH] configfs: make directories inherit uid/gid from creator
To:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc:     Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Joel Becker <jlbec@evilplan.org>,
        Christoph Hellwig <hch@lst.de>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

> > Currently a non-root process can create directories, but cannot
> > create stuff in the directories it creates.
>
> Isn't that on purpose?

Is it?  What's the use case of being able to create empty directories
you can't use?
Why allow their creation in the first place then?

> > +             (void)configfs_setattr(dentry, &ia);
>
> No need for (void), here, right?

Just being clear we're ignoring any potential error return.

> And this feels like a big user-visible change, what is going to break
> with this?

That I don't know, but it's unlikely to break anything, since it's virtually
impossible to use as it is now.  If non-root creates the directory, it's
currently root-owned, so can only be used by root (or something with
appropriate caps).
Something with capabilities will be able to use it even if it is no
longer owned by root.