Received-SPF: pass (google.com: domain of linux-ext4-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Mon, 4 May 2020 15:39:00 -0700
From:   "Darrick J. Wong" <darrick.wong@oracle.com>
To:     Andreas Dilger <adilger@dilger.ca>
Cc:     Jonny Grant <jg@jguk.org>,
        Ext4 Developers List <linux-ext4@vger.kernel.org>
Subject: Re: /fs/ext4/ext4.h add a comment to ext4_dir_entry_2
Message-ID: <20200504223900.GA5691@magnolia>
References: <bf50e54f-2a0c-17a4-89c3-4afcc298daeb@jguk.org>
 <1B91A6E6-7F4A-4C58-93E7-394217C1631C@dilger.ca>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1B91A6E6-7F4A-4C58-93E7-394217C1631C@dilger.ca>
Sender: linux-ext4-owner@vger.kernel.org
Precedence: bulk

On Mon, May 04, 2020 at 04:26:35PM -0600, Andreas Dilger wrote:
> 
> > On May 3, 2020, at 6:52 AM, Jonny Grant <jg@jguk.org> wrote:
> > 
> > Hello
> > 
> > Could a comment be added to clarify 'file_type' ?
> > 
> > struct ext4_dir_entry_2 {
> >    __le32    inode;            /* Inode number */
> >    __le16    rec_len;        /* Directory entry length */
> >    __u8    name_len;        /* Name length */
> >    __u8    file_type;
> >    char    name[EXT4_NAME_LEN];    /* File name */
> > };
> > 
> > 
> > 
> > This what I am proposing to add:
> > 
> >    __u8    file_type;        /* See directory file type macros below */
> 
> For this kind of structure field, it makes sense to reference the macro
> names directly, like:
> 
> 	__u8	file_type;	/* See EXT4_FT_* type macros below */
> 
> since "macros below" may be ambiguous as the header changes over time.
> 
> 
> Even better (IMHO) is to use a named enum for this, like:
> 
>         enum ext4_file_type file_type:8; /* See EXT4_FT_ types below */
> 
> /*
>  * Ext4 directory file types.  Only the low 3 bits are used.  The
>  * other bits are reserved for now.
>  */
> enum ext4_file_type {
> 	EXT4_FT_UNKNOWN		= 0,
> 	EXT4_FT_REG_FILE	= 1,
> 	EXT4_FT_DIR		= 2,
> 	EXT4_FT_CHRDEV		= 3,
> 	EXT4_FT_BLKDEV		= 4,
> 	EXT4_FT_FIFO		= 5,
> 	EXT4_FT_SOCK		= 6,
> 	EXT4_FT_SYMLINK		= 7,
> 	EXT4_FT_MAX,
> 	EXT4_FT_DIR_CSUM	= 0xDE
> };
> 
> so that the allowed values for this field are clear from the definition.
> However, the use of a fixed-with bitfield (enum :8) is a GCC-ism and Ted
> may be against that for portability reasons, since the kernel and
> userspace headers should be as similar as possible.

This is an on-disk structure.  Do /not/ make this an enum because that
would replace a __u8 with an int, which will break directories.

--D

> Cheers, Andreas
> 
> 
> 
> 
>