On Tue, May 28, 2013 at 09:41:08PM +0200, Andrew Lunn wrote:
> On Sun, May 26, 2013 at 06:04:24PM +0200, Johan Hovold wrote:
> > On Sun, May 26, 2013 at 01:00:51PM +0200, Andrew Lunn wrote:
> > > +/*
> > > + * mxuport_write_room
> > > + *
> > > + * Return how much space is available in the buffer.
> > > + */
> > > +static int mxuport_write_room(struct tty_struct *tty)
> > > +{
> > > + struct usb_serial_port *port = tty->driver_data;
> > > + unsigned long flags;
> > > + int room;
> > > +
> > > + spin_lock_irqsave(&port->lock, flags);
> > > + room = kfifo_avail(&port->write_fifo);
> > > + spin_unlock_irqrestore(&port->lock, flags);
> > > +
> > > + dev_dbg(&port->dev, "%s - returns %d\n", __func__, room);
> > > + return room;
> > > +}
>
> ....
>
> > It seems it could be possible to reuse the generic write implementations
> > of all functions but prepare_write_buffer above rather than the modified
> > copies.
>
> Hi Johan
>
> If there a reason why usb_serial_generic_write_room() is not exported
> as a GPL symbol. It looks like i should be able to use it once i sort
> my 'fake' ports/endpoints out.
No, I don't think so, but perhaps it should be. Most usb-serial drivers
still use it. Simply leave the operation unset and usb-serial core will
set it for you.
Johan
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
