--k1lZvvs/B4yU6o8G
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Wed, Jun 01, 2005 at 12:25:51PM +1000, matthew green wrote:
>   =20
>    Why not just use the device? Define a standard ioctl that all=20
>    firmware-loading devices support and have one tool that will itterativ=
ely=20
>    upload.
>=20
> there is currently no device node for my "iwi0" card.  seems wasteful
> to add one just for this, rather than pass the name in as part of the
> ioctl (like many other network ioctls, i think..)

Oh, yeah, that.

For now, I think a second variant of the firmware upload that takes a NIC=
=20
name would be fine. While it strikes me as a bit of a hack, it's well in=20
line with other ioctls now.

To be honest, I wouldn't mind a device node for network interfaces, but I=
=20
think it'd be better to have that as a separate change; changing firmware=
=20
uploading shouldn't be gated on this.

I also think that a device node for a NIC would do more than support
firmware update. NICs are devices, and it would be nice to be able to
control all devices (things with a struct device) in a coherent manner. By
that I mean that as we add common "device" functionality (like detach
yourself, report your autoconfig attach point..), it would be nice for it=
=20
to work on all devices the same.

Take care,

Bill

--k1lZvvs/B4yU6o8G
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (NetBSD)

iD8DBQFCndwfWz+3JHUci9cRAhSCAKCVKpvTR66n3c4uTroF/PkRzrThIwCfTO0N
kaPSnK8sktcmzi/guSBSyXA=
=Fhjx
-----END PGP SIGNATURE-----

--k1lZvvs/B4yU6o8G--