Re: [PATCH] gpio: Drop CONFIG_DEBUG_GPIO

[Brian Norris]

On Mon, Mar 14, 2022 at 3:23 PM Linus Walleij <linus.walleij@xxxxxxxxxx> wrote:
> On Mon, Mar 14, 2022 at 4:00 PM Bartosz Golaszewski <brgl@xxxxxxxx> wrote:
> > I like it. It's true we don't see many of those DEBUG constructs
> > anymore nowadays and overhead for might_sleep() and WARN_ON() is
> > negligible.
>
> I agree.  I have something similar for pinctrl, maybe that needs to
> go too.

Huh, yeah, CONFIG_DEBUG_PINCTRL does look awfully similar, and I just
didn't notice because we don't happen to have it enabled for Chromium
kernels. We happen to have CONFIG_DEBUG_GPIO enabled though, and the
"new" rockchip-gpio log messages triggered me :)

I guess one difference is that CONFIG_DEBUG_PINCTRL is almost
exclusively (aside from some renesas drivers?) about extra logging and
less about interesting checks that one might want to enable in more
general settings. So it's a clearer call to make that people generally
want it disabled.

In other words, a -DDEBUG construct in itself isn't necessarily
terrible (even if it's a little redundant with dynamic debug?), if
it's not conflated with stuff that might be more generally useful.

Regards,
Brian