Re: [PATCH 2/2 v2] [POWERPC] Ignore disabled serial ports

[Josh Boyer <jwboyer@linux.vnet.ibm.com>]

On Mon, 3 Mar 2008 04:43:42 +0100
Arnd Bergmann <arnd@arndb.de> wrote:

> On Saturday 01 March 2008, Josh Boyer wrote:
> > --- linux-2.6.orig/drivers/serial/of_serial.c
> > +++ linux-2.6/drivers/serial/of_serial.c
> > @@ -72,6 +72,11 @@ static int __devinit of_platform_serial_
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0int port_type;
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0int ret;
> > =C2=A0
> > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0if (!of_device_is_available(=
ofdev->node)) {
> > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0dev_info(&ofdev->dev, "Disabled serial port. =C2=A0Ign=
ored\n");
> > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0return -ENODEV;
> > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0}
> > +
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0if (of_find_property(of=
dev->node, "used-by-rtas", NULL))
> > =C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0return -EBUSY;
>=20
> I wonder whether we should move the check for "used-by-rtas" into the
> of_device_is_available function. I understand that used-by-rtas is
> another way of expressing the idea that the kernel is not supposed to
> access the specific device. In this case, the device is physically
> present, but is not available to the OS.

I'd rather not at the moment.  My intention was to only look at the
status property for now.  I'd like to avoid this function growing into
a huge switch statement for $random_firmware's way of flagging
something as "don't touch".

josh