xfs
[Top] [All Lists]

Re: [REVIEW] cleanup - remove bhv_vname_t

To: Barry Naujok <bnaujok@xxxxxxx>
Subject: Re: [REVIEW] cleanup - remove bhv_vname_t
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Tue, 8 Apr 2008 02:38:22 -0400
Cc: "xfs@xxxxxxxxxxx" <xfs@xxxxxxxxxxx>, xfs-dev <xfs-dev@xxxxxxx>
In-reply-to: <op.t89zp3j63jf8g2@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>
References: <op.t89zp3j63jf8g2@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>
Sender: xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.5.17 (2007-11-01)
On Tue, Apr 08, 2008 at 02:50:17PM +1000, Barry Naujok wrote:
> -
> +

What's this?  both lines look empty.

>       return -dm_send_namesp_event(event, mp ? mp->m_super : NULL,
>                                   ip1->i_vnode, vp1_right,
>                                   ip2 ? ip2->i_vnode : NULL, vp2_right,
> -                                 name1, name2,
> +                                 (char *)name1, (char *)name2,

please change dm_send_namesp_event to take a const char aswell.

> +static inline struct xfs_name *
> +xfs_dentry_name(
> +     struct xfs_name *namep,
> +     struct dentry   *dentry)
> +{
> +     namep->name = dentry->d_name.name;
> +     namep->len = dentry->d_name.len;
> +     return namep;
> +}

As mentioned in my comment to the CI series:  shouldn't you just use
a struct qstr instead of adding a new struct xfs_name?

Also please don't add inline for this.
> +xfs_name_t xfs_name_dotdot = {"..", 2};

const?

> +     args.name = name->name;
> +     args.namelen = name->len;
> +     args.hashval = xfs_da_hashname(name->name, name->len);

What about just putting a pointer to the xfs_name/qstr into args?

>  /*
> + * Counted string for file names.
> + */
> +typedef struct xfs_name {
> +     const uchar_t   *name;
> +     int             len;
> +} xfs_name_t;

Please try to avoid the typedef for newly added types.


<Prev in Thread] Current Thread [Next in Thread>