RE: [PATCH] drivers/serial/8250_pci.c (add support for '8-port RS-232 MIC-3620 from advantech'

["Tosoni" <jp.tosoni@acksys.fr>]

> [mailto:linux-serial-owner@vger.kernel.org]On Behalf Of Michael Bramer
>
> On Wed, Jan 21, 2009 at 12:32:15PM +0000, Alan Cox wrote:
> > > >>  static struct pci_device_id serial_pci_tbl[] = {
> > > >> +       {       PCI_VENDOR_ID_ADVANTECH,
> PCI_DEVICE_ID_ADVANTECH_PCI3620,
> > > >> +               0x3620, PCI_ANY_ID, 0, 0,
> > >
> > > Why not use PCI_VENDOR_ID_ADVANTECH as PCI_SUBVENDOR_ID too?
> >
> > The Advantech vendor id is not 0x3620. This confused me as
> well which is
> > why I asked for an lspci. Advantech has stuck the device id in the
> > subvendor bits and '1' in the subdevice (so it should be 1 not
> > PCI_ANY_ID).
>
> is this better?
>
> +       {       PCI_VENDOR_ID_ADVANTECH,
> PCI_DEVICE_ID_ADVANTECH_PCI3620,
> +               PCI_DEVICE_ID_ADVANTECH_PCI3620, 1, 0, 0,

Since the name describes a device id where it should be a (sub)vendor id,
I would suggest that you add a line of comment to explain the case.
So that no one will be tempted to change it back to PCI_VENDOR_ID_ADVANTECH
in the future.

Regards