Messages in this thread | | | Date | Thu, 12 Dec 2013 15:40:32 +0100 | Subject | Re: [PATCH 2/3] pinctrl: at91: initialize config parameter to 0 | From | Linus Walleij <> |
| |
On Mon, Dec 9, 2013 at 10:55 AM, Alexandre Belloni <alexandre.belloni@free-electrons.com> wrote:
> I'm actually wondering if the checks for the ops->pin_config_get are not > a bit overkill. We check for that function in: > - pinconf_check_ops() > - before calling it in pin_config_get_for_pin() which is only used > once, in the same path : dump using debugfs and having ops->is_generic > == true > - in pinconf_pins_show() which is the function calling also in the same > path > > What I would do is: > - remove all the checks in pinconf_check_ops() and pinconf_pins_show() > so that people are not pressured to implement a function that is simply > never used. > - modify pin_config_get_for_pin() by removing the dev_err() call and > returning -ENOTSUPP instead of -EINVAL (it doesn't change the behaviour > but I feel -ENOTSUPP is more appropriate) > > I have a patch ready but I can't test it as I don't own any of the > is_generic platforms.
Mail it out with a [CFT: ] "call for testing" prefix.
Yours, Linus Walleij
| |