[RFC,4/5] usb: cdc-acm: Enable serdev support

[RFC,4/5] usb: cdc-acm: Enable serdev support

[Andreas Färber]

Switch from tty_port_register_device() to tty_port_register_device_serdev()
and from tty_unregister_device() to tty_port_unregister_device().

On removal of a serdev driver sometimes count mismatch warnings were seen:

  ttyACM ttyACM0: tty_hangup: tty->count(1) != (#fd's(0) + #kopen's(0))
  ttyACM ttyACM0: tty_port_close_start: tty->count = 1 port count = 0

Note: The serdev drivers appear to probe asynchronously as soon as they
are registered. Should the USB quirks in probe be moved before registration?
No noticeable difference for the devices at hand.

Cc: Rob Herring <robh@kernel.org>
Cc: Johan Hovold <johan@kernel.org>
Signed-off-by: Andreas Färber <afaerber@suse.de>
---
 drivers/usb/class/cdc-acm.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c
index ed8c62b2d9d1..c225a586c524 100644
--- a/drivers/usb/class/cdc-acm.c
+++ b/drivers/usb/class/cdc-acm.c
@@ -1460,7 +1460,7 @@ static int acm_probe(struct usb_interface *intf,
 	usb_set_intfdata(data_interface, acm);
 
 	usb_get_intf(control_interface);
-	tty_dev = tty_port_register_device(&acm->port, acm_tty_driver, minor,
+	tty_dev = tty_port_register_device_serdev(&acm->port, acm_tty_driver, minor,
 			&control_interface->dev);
 	if (IS_ERR(tty_dev)) {
 		rv = PTR_ERR(tty_dev);
@@ -1534,7 +1534,7 @@ static void acm_disconnect(struct usb_interface *intf)
 	acm_kill_urbs(acm);
 	cancel_work_sync(&acm->work);
 
-	tty_unregister_device(acm_tty_driver, acm->minor);
+	tty_port_unregister_device(&acm->port, acm_tty_driver, acm->minor);
 
 	usb_free_urb(acm->ctrlurb);
 	for (i = 0; i < ACM_NW; i++)

[RFC,4/5] usb: cdc-acm: Enable serdev support

[Oliver Neukum]

On Fr, 2019-01-04 at 12:21 +0100, Andreas Färber  wrote:
> Switch from tty_port_register_device() to tty_port_register_device_serdev()
> and from tty_unregister_device() to tty_port_unregister_device().
> 
> On removal of a serdev driver sometimes count mismatch warnings were seen:
> 
>   ttyACM ttyACM0: tty_hangup: tty->count(1) != (#fd's(0) + #kopen's(0))
>   ttyACM ttyACM0: tty_port_close_start: tty->count = 1 port count = 0
> 
> Note: The serdev drivers appear to probe asynchronously as soon as they
> are registered. Should the USB quirks in probe be moved before registration?
> No noticeable difference for the devices at hand.

That is quite drastic a change.
Johan, how complete in terms of features is serdev?

Are you refering to CLEAR_HALT_CONDITIONS in terms of quirks?

	Regards
		Oliver

[RFC,4/5] usb: cdc-acm: Enable serdev support

[Johan Hovold]

On Mon, Jan 07, 2019 at 02:48:26PM +0100, Oliver Neukum wrote:
> On Fr, 2019-01-04 at 12:21 +0100, Andreas Färber  wrote:
> > Switch from tty_port_register_device() to tty_port_register_device_serdev()
> > and from tty_unregister_device() to tty_port_unregister_device().
> > 
> > On removal of a serdev driver sometimes count mismatch warnings were seen:
> > 
> >   ttyACM ttyACM0: tty_hangup: tty->count(1) != (#fd's(0) + #kopen's(0))
> >   ttyACM ttyACM0: tty_port_close_start: tty->count = 1 port count = 0
> > 
> > Note: The serdev drivers appear to probe asynchronously as soon as they
> > are registered. Should the USB quirks in probe be moved before registration?
> > No noticeable difference for the devices at hand.
> 
> That is quite drastic a change.
> Johan, how complete in terms of features is serdev?

serdev doesn't support hangups yet, and that's precisely why it's not
enabled for hotpluggable buses. That would need to be fixed before
accepting something like this.

Johan