On Tue, Feb 16, 2021 at 08:37:31PM -0800, Dmitry Torokhov wrote:
> Hi Dan,
>
> On Thu, Jan 28, 2021 at 12:35:51PM +0300, Dan Carpenter wrote:
> > If elo_setup_10() fails then this should return an error code instead
> > of success.
>
> Thank you for the patch.
>
> >
> > Fixes: fae3006e4b42 ("Input: elo - add support for non-pressure-sensitive touchscreens")
> > Signed-off-by: Dan Carpenter <dan.carpenter@xxxxxxxxxx>
> > ---
> > drivers/input/touchscreen/elo.c | 4 +++-
> > 1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/input/touchscreen/elo.c b/drivers/input/touchscreen/elo.c
> > index e0bacd34866a..4b2fb73da5e2 100644
> > --- a/drivers/input/touchscreen/elo.c
> > +++ b/drivers/input/touchscreen/elo.c
> > @@ -341,8 +341,10 @@ static int elo_connect(struct serio *serio, struct serio_driver *drv)
> > switch (elo->id) {
> >
> > case 0: /* 10-byte protocol */
> > - if (elo_setup_10(elo))
> > + if (elo_setup_10(elo)) {
> > + err = -EINVAL;
>
> Ideally we'd propagate error from elo_setup_10() and underlying code,
> but we are not ready for it, as most serio code simply uses -1 for
> errors. However I think that -EIO would suit better here. Please let me
> know if you disagree, otherwise I'll fix it up on my side.
>
-EIO is fine. Thanks!
regards,
dan carpenter
