Re: [PATCH 1/1] driver:garmin gps: use setup_timer

[Johan Hovold <johan@kernel.org>]

On Mon, Oct 09, 2017 at 09:34:05AM -0700, Kees Cook wrote:
> On Mon, Oct 9, 2017 at 6:59 AM, Johan Hovold <johan@kernel.org> wrote:
> > On Thu, Sep 21, 2017 at 04:54:24PM +0530, Allen Pais wrote:
> >>     Use setup_timer function instead of initializing timer with the
> >>     function and data fields.
> >
> > Why the odd indentation here?
> >
> >> Signed-off-by: Allen Pais <allen.lkml@gmail.com>
> >> ---
> >>  drivers/usb/serial/garmin_gps.c | 5 ++---
> >>  1 file changed, 2 insertions(+), 3 deletions(-)
> >>
> >> diff --git a/drivers/usb/serial/garmin_gps.c b/drivers/usb/serial/garmin_gps.c
> >> index b2f2e87..2dbc1bed 100644
> >> --- a/drivers/usb/serial/garmin_gps.c
> >> +++ b/drivers/usb/serial/garmin_gps.c
> >> @@ -1391,12 +1391,11 @@ static int garmin_port_probe(struct usb_serial_port *port)
> >>       if (!garmin_data_p)
> >>               return -ENOMEM;
> >>
> >> -     init_timer(&garmin_data_p->timer);
> >> +     setup_timer(&garmin_data_p->timer, timeout_handler,
> >> +                 (unsigned long)garmin_data_p);
> >>       spin_lock_init(&garmin_data_p->lock);
> >>       INIT_LIST_HEAD(&garmin_data_p->pktlist);
> >>       /* garmin_data_p->timer.expires = jiffies + session_timeout; */
> >> -     garmin_data_p->timer.data = (unsigned long)garmin_data_p;
> >> -     garmin_data_p->timer.function = timeout_handler;
> >>       garmin_data_p->port = port;
> >>       garmin_data_p->state = 0;
> >>       garmin_data_p->flags = 0;
> >
> > Kees submitted the exact same change the day before you as part of a
> > larger patch resulting from running coccinelle and replacing all
> > open-coded versions of setup_timer in go, something which seems
> > preferable.
> >
> > Not sure who is expected to pick that patch up, but I'll hold off on
> > this one for a while still waiting to see what happens to Kees's series.
> 
> Hi, actually, "setup_timer" is the old api ("init_timer" is the
> ancient api), "timer_setup" (to match hrtimer_setup) is the current
> future approach here...

So your patch from 20/9 ("[PATCH v2 02/31] timer: Convert open-coded
init_timer() to setup_timer()") has been superseded? Or do you still
expect that one to be applied (by tglx?)?

Johan