Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 7 Apr 2008 23:40:27 +0200
From:      "=?ISO-8859-1?Q?Ermal_Lu=E7i?=" <ermal.luci@gmail.com>
To:        "Brooks Davis" <brooks@freebsd.org>
Cc:        freebsd-arch@freebsd.org
Subject:   Re: Question about network interface naming and devfs(5)!
Message-ID:  <9a542da30804071440i1d7b0c6cwbdc44625b94fb90e@mail.gmail.com>
In-Reply-To: <20080407202527.GG28511@lor.one-eyed-alien.net>
References:  <9a542da30804071201o520812f8w4fc413131c4140b0@mail.gmail.com> <20080407202527.GG28511@lor.one-eyed-alien.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Apr 7, 2008 at 10:25 PM, Brooks Davis <brooks@freebsd.org> wrote:
>
> On Mon, Apr 07, 2008 at 09:01:18PM +0200, Ermal Lu?i wrote:
>  > Hello,
>  >
>  > it seems that devfs(5) is not consistent with network interface naming feature.
>  >
>  > Network interface naming is offered by ifconfig $interface name $name.
>  > This feature is really helpful to distinguish interfaces properly in a
>  > machine and helps in scripting things.
>  > The problem is that the ioctl that renames the device advertises it as
>  > an interface arrival event which devfs does not handle and such you
>  > have inconsistence between ifconfig output and the information in
>  > devfs.
>  >
>  > Are there any plans to fix this or is there any issue for not making
>  > devfs(5) aware of network interfaces events?!
>
>  I haven't dealt with it because a) there's no way to rename entries
>  which currently means you've have to destroy the current entry and add
>  a completely new one which would break applications using the devfs
>  entries across renames and b) I think the devfs entries are a mistake and
>  should die.

Can't it be made to behave like a real file system?!
Meaning to have reference count for its devices/entries so the
entry/descriptor is kept open till the last close() is performed!

Ermal

>
>  -- Brooks
>



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9a542da30804071440i1d7b0c6cwbdc44625b94fb90e>