diff mbox

[4/4] tty/serial: sh-sci: remove uneeded IS_ERR_OR_NULL calls

Message ID 20170324080006.tbhyqlgk35xybsna@pengutronix.de
State New
Headers show

Commit Message

Uwe Kleine-König March 24, 2017, 8 a.m. UTC 
Hello Dmitry,

On Thu, Mar 23, 2017 at 12:58:04PM -0700, Dmitry Torokhov wrote:
> On Thu, Mar 23, 2017 at 08:10:20PM +0100, Uwe Kleine-König wrote:
> > On Thu, Mar 23, 2017 at 08:44:41AM -0700, Dmitry Torokhov wrote:
> > > On Thu, Mar 23, 2017 at 07:43:25AM -0700, Dmitry Torokhov wrote:
> > > > On Thu, Mar 23, 2017 at 02:41:53PM +0100, Linus Walleij wrote:
> > > > > On Thu, Mar 23, 2017 at 1:34 PM, Uwe Kleine-König
> > > > > <u.kleine-koenig@pengutronix.de> wrote:
> > > > > 
> > > > > > Maybe we can make gpiod_get_optional look like this:
> > > > > >
> > > > > >         if (!dev->of_node && isnt_a_acpi_device(dev) && !IS_ENABLED(GPIOLIB))
> > > > > >                 return NULL;
> > > > > >         else
> > > > > >                 return -ENOSYS;
> > > > > >
> > > > > > I don't know how isnt_a_acpi_device looks like, probably it involves
> > > > > > CONFIG_ACPI and/or dev->acpi_node.
> > > > > >
> > > > > > This should be safe and still comfortable for legacy platforms, isn't it?
> > > > > 
> > > > > I like the looks of this.
> > > > > 
> > > > > Can we revert Dmitry's patch and apply something like this instead?
> > > > > 
> > > > > Dmitry, how do you feel about this?
> > > > 
> > > > I frankly do not see the point. It still makes driver code more complex
> > 
> > Note that this code is in the gpio header, and not in driver code. So
> > the driver just does
> > 
> > 	gpiod = gpiod_get_optional(...)
> > 	if (IS_ERR(gpiod))
> > 		return PTR_ERR(gpiod);
> > 
> > (as it is supposed to do now). I think that's nice.
> 
> Except that it breaks if !CONFIG_GPIOLIB which is legitimate config in
> many cases. Can I have a DT platform or ACPI platform that does not
> expose any GPIOs for kernel to control and thus want to disable GPIOLIB?
> I can't see why not.
> 
> > 
> > > > for no good reason. I also think that not having optional GPIO is not an
> > > > error, so returning value from error space is not correct. NULL is value
> > > > from another space altogether.
> > 
> > It seems you didn't understand my concern.
> 
> Likewise.

OK, lets agree that we don't agree then. You think that if someone
disabled GPIOLIB it should be safe to assume that there is no GPIO, I
think that's wrong.

Still I think it should be possible that we agree on my suggestion to
return NULL in some cases only. Here is a prototype (not even compile
tested without GPIOLIB):


------->8--------
From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Subject: [PATCH] gpiod: let get_optional return NULL in some cases with GPIOLIB disabled

People disagree if gpiod_get_optional should return NULL or
ERR_PTR(-ENOSYS) if GPIOLIB is disabled. The argument for NULL is that
the person who decided to disable GPIOLIB is assumed to know that there
is no GPIO. The reason to stick to ERR_PTR(-ENOSYS) is that it might
introduce hard to debug problems if that decision is wrong.

So this patch introduces a compromise and let gpiod_get_optional (and
its variants) return NULL if the device in question cannot have an
associated GPIO because it is neither instantiated by a device tree nor
by ACPI.

This should handle most cases that are argued about.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
---
 include/linux/gpio/consumer.h | 55 ++++++++++++++++++++++++++++++++++++-------
 1 file changed, 46 insertions(+), 9 deletions(-)


Best regards
Uwe

Comments

Geert Uytterhoeven March 24, 2017, 8:29 a.m. UTC | #1 
Hi Uwe,

On Fri, Mar 24, 2017 at 9:00 AM, Uwe Kleine-König
<u.kleine-koenig@pengutronix.de> wrote:
> From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> Subject: [PATCH] gpiod: let get_optional return NULL in some cases with GPIOLIB disabled
>
> People disagree if gpiod_get_optional should return NULL or
> ERR_PTR(-ENOSYS) if GPIOLIB is disabled. The argument for NULL is that
> the person who decided to disable GPIOLIB is assumed to know that there
> is no GPIO. The reason to stick to ERR_PTR(-ENOSYS) is that it might
> introduce hard to debug problems if that decision is wrong.
>
> So this patch introduces a compromise and let gpiod_get_optional (and
> its variants) return NULL if the device in question cannot have an
> associated GPIO because it is neither instantiated by a device tree nor
> by ACPI.
>
> This should handle most cases that are argued about.
>
> Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> ---
>  include/linux/gpio/consumer.h | 55 ++++++++++++++++++++++++++++++++++++-------
>  1 file changed, 46 insertions(+), 9 deletions(-)
>
> diff --git a/include/linux/gpio/consumer.h b/include/linux/gpio/consumer.h
> index fb0fde686cb1..0ca29889290d 100644
> --- a/include/linux/gpio/consumer.h
> +++ b/include/linux/gpio/consumer.h
> @@ -161,20 +161,48 @@ gpiod_get_index(struct device *dev,
>         return ERR_PTR(-ENOSYS);
>  }
>
> -static inline struct gpio_desc *__must_check
> -gpiod_get_optional(struct device *dev, const char *con_id,
> -                  enum gpiod_flags flags)
> +static inline bool __gpiod_no_optional_possible(struct device *dev)
>  {
> -       return ERR_PTR(-ENOSYS);
> +       /*
> +        * gpiod_get_optional et al can only provide a GPIO if at least one of
> +        * the backends for specifing a GPIO is available. These are device
> +        * tree, ACPI and gpiolib's lookup tables. The latter isn't available if
> +        * GPIOLIB is disabled (which is the case here).
> +        * So if the provided device is unrelated to device tree and ACPI, we
> +        * can be sure that there is no optional GPIO and let gpiod_get_optional
> +        * safely return NULL.
> +        * Otherwise there is still a chance that there is no GPIO but we cannot
> +        * be sure without having to enable a part of GPIOLIB (i.e. the lookup
> +        * part). So lets play safe and return an error. (Though there are also
> +        * arguments that returning NULL then would be beneficial.)
> +        */
> +
> +       if (IS_ENABLED(CONFIG_OF) && dev && dev->of_node)
> +               return false;

At first sight, I though this was OK:

  1. On ARM with DT, we can assume CONFIG_GPIOLOB=y.

  2. I managed to configure an SH kernel with CONFIG_GPIOLOB=n, CONFIG_OF=y,
     and CONFIG_SERIAL_SH_SCI=y, but since SH boards with SH-SCI UARTs do
     not use DT (yet), the check for dev->of_node (false) should handle
     that.

  3. However, I managed to do the same for h8300, which does use DT. Hence
     if mctrl_gpio would start relying on gpiod_get_optional(), this would
     break the sh-sci driver on h8300 :-(
     Note that h8300 doesn't have any GPIO drivers (yet?), so
     CONFIG_GPIPOLIB=n makes perfect sense!

So I'm afraid the only option is to always return NULL, and put the
responsability on the shoulders of the system integrator...

> +       if (IS_ENABLED(CONFIG_ACPI) && dev && ACPI_COMPANION(dev))
> +               return false;

No comments about the ACPI case.

>  static inline struct gpio_desc *__must_check
>  gpiod_get_index_optional(struct device *dev, const char *con_id,
>                          unsigned int index, enum gpiod_flags flags)
>  {
> +       if (__gpiod_no_optional_possible(dev))
> +               return NULL;
> +
>         return ERR_PTR(-ENOSYS);

Regardless of the above, given you use the exact same construct in four
locations, what about letting __gpiod_no_optional_possible() return the NULL
or ERR_PTR itself, and renaming it to e.g. __gpiod_no_optional_return_value()?

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
--
To unsubscribe from this list: send the line "unsubscribe linux-gpio" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Uwe Kleine-König March 24, 2017, 8:39 a.m. UTC | #2 
Hello Geert,

On Fri, Mar 24, 2017 at 09:29:02AM +0100, Geert Uytterhoeven wrote:
> On Fri, Mar 24, 2017 at 9:00 AM, Uwe Kleine-König
> <u.kleine-koenig@pengutronix.de> wrote:
> > From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> > Subject: [PATCH] gpiod: let get_optional return NULL in some cases with GPIOLIB disabled
> >
> > People disagree if gpiod_get_optional should return NULL or
> > ERR_PTR(-ENOSYS) if GPIOLIB is disabled. The argument for NULL is that
> > the person who decided to disable GPIOLIB is assumed to know that there
> > is no GPIO. The reason to stick to ERR_PTR(-ENOSYS) is that it might
> > introduce hard to debug problems if that decision is wrong.
> >
> > So this patch introduces a compromise and let gpiod_get_optional (and
> > its variants) return NULL if the device in question cannot have an
> > associated GPIO because it is neither instantiated by a device tree nor
> > by ACPI.
> >
> > This should handle most cases that are argued about.
> >
> > Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> > ---
> >  include/linux/gpio/consumer.h | 55 ++++++++++++++++++++++++++++++++++++-------
> >  1 file changed, 46 insertions(+), 9 deletions(-)
> >
> > diff --git a/include/linux/gpio/consumer.h b/include/linux/gpio/consumer.h
> > index fb0fde686cb1..0ca29889290d 100644
> > --- a/include/linux/gpio/consumer.h
> > +++ b/include/linux/gpio/consumer.h
> > @@ -161,20 +161,48 @@ gpiod_get_index(struct device *dev,
> >         return ERR_PTR(-ENOSYS);
> >  }
> >
> > -static inline struct gpio_desc *__must_check
> > -gpiod_get_optional(struct device *dev, const char *con_id,
> > -                  enum gpiod_flags flags)
> > +static inline bool __gpiod_no_optional_possible(struct device *dev)
> >  {
> > -       return ERR_PTR(-ENOSYS);
> > +       /*
> > +        * gpiod_get_optional et al can only provide a GPIO if at least one of
> > +        * the backends for specifing a GPIO is available. These are device
> > +        * tree, ACPI and gpiolib's lookup tables. The latter isn't available if
> > +        * GPIOLIB is disabled (which is the case here).
> > +        * So if the provided device is unrelated to device tree and ACPI, we
> > +        * can be sure that there is no optional GPIO and let gpiod_get_optional
> > +        * safely return NULL.
> > +        * Otherwise there is still a chance that there is no GPIO but we cannot
> > +        * be sure without having to enable a part of GPIOLIB (i.e. the lookup
> > +        * part). So lets play safe and return an error. (Though there are also
> > +        * arguments that returning NULL then would be beneficial.)
> > +        */
> > +
> > +       if (IS_ENABLED(CONFIG_OF) && dev && dev->of_node)
> > +               return false;
> 
> At first sight, I though this was OK:
> 
>   1. On ARM with DT, we can assume CONFIG_GPIOLOB=y.
> 
>   2. I managed to configure an SH kernel with CONFIG_GPIOLOB=n, CONFIG_OF=y,
>      and CONFIG_SERIAL_SH_SCI=y, but since SH boards with SH-SCI UARTs do
>      not use DT (yet), the check for dev->of_node (false) should handle
>      that.
> 
>   3. However, I managed to do the same for h8300, which does use DT. Hence
>      if mctrl_gpio would start relying on gpiod_get_optional(), this would
>      break the sh-sci driver on h8300 :-(
>      Note that h8300 doesn't have any GPIO drivers (yet?), so
>      CONFIG_GPIPOLIB=n makes perfect sense!

Thanks for your efforts.

> So I'm afraid the only option is to always return NULL, and put the
> responsability on the shoulders of the system integrator...

The gpio lines could be provided by an i2c gpio adapter, right? So IMHO
you don't need platform gpios to justify -ENODEV. So I guess that's a
case where we don't come to an agreement.

> > +       if (IS_ENABLED(CONFIG_ACPI) && dev && ACPI_COMPANION(dev))
> > +               return false;
> 
> No comments about the ACPI case.
> 
> >  static inline struct gpio_desc *__must_check
> >  gpiod_get_index_optional(struct device *dev, const char *con_id,
> >                          unsigned int index, enum gpiod_flags flags)
> >  {
> > +       if (__gpiod_no_optional_possible(dev))
> > +               return NULL;
> > +
> >         return ERR_PTR(-ENOSYS);
> 
> Regardless of the above, given you use the exact same construct in four
> locations, what about letting __gpiod_no_optional_possible() return the NULL
> or ERR_PTR itself, and renaming it to e.g. __gpiod_no_optional_return_value()?

I thought about that but didn't find a good name and so considered it
more clear this way. Another optimisation would be to unconditionally
define get_optional in terms of get_index_optional which would simplify
my patch a bit.

I'd consider __gpiod_optional_return_value a better name than
__gpiod_no_optional_return_value but I'm still not convinced.

Best regards
Uwe
Geert Uytterhoeven March 24, 2017, 8:59 a.m. UTC | #3 
Hi Uwe,

On Fri, Mar 24, 2017 at 9:39 AM, Uwe Kleine-König
<u.kleine-koenig@pengutronix.de> wrote:
> On Fri, Mar 24, 2017 at 09:29:02AM +0100, Geert Uytterhoeven wrote:
>> On Fri, Mar 24, 2017 at 9:00 AM, Uwe Kleine-König
>> <u.kleine-koenig@pengutronix.de> wrote:
>> > From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
>> > Subject: [PATCH] gpiod: let get_optional return NULL in some cases with GPIOLIB disabled
>> >
>> > People disagree if gpiod_get_optional should return NULL or
>> > ERR_PTR(-ENOSYS) if GPIOLIB is disabled. The argument for NULL is that
>> > the person who decided to disable GPIOLIB is assumed to know that there
>> > is no GPIO. The reason to stick to ERR_PTR(-ENOSYS) is that it might
>> > introduce hard to debug problems if that decision is wrong.
>> >
>> > So this patch introduces a compromise and let gpiod_get_optional (and
>> > its variants) return NULL if the device in question cannot have an
>> > associated GPIO because it is neither instantiated by a device tree nor
>> > by ACPI.
>> >
>> > This should handle most cases that are argued about.
>> >
>> > Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
>> > ---
>> >  include/linux/gpio/consumer.h | 55 ++++++++++++++++++++++++++++++++++++-------
>> >  1 file changed, 46 insertions(+), 9 deletions(-)
>> >
>> > diff --git a/include/linux/gpio/consumer.h b/include/linux/gpio/consumer.h
>> > index fb0fde686cb1..0ca29889290d 100644
>> > --- a/include/linux/gpio/consumer.h
>> > +++ b/include/linux/gpio/consumer.h
>> > @@ -161,20 +161,48 @@ gpiod_get_index(struct device *dev,
>> >         return ERR_PTR(-ENOSYS);
>> >  }
>> >
>> > -static inline struct gpio_desc *__must_check
>> > -gpiod_get_optional(struct device *dev, const char *con_id,
>> > -                  enum gpiod_flags flags)
>> > +static inline bool __gpiod_no_optional_possible(struct device *dev)
>> >  {
>> > -       return ERR_PTR(-ENOSYS);
>> > +       /*
>> > +        * gpiod_get_optional et al can only provide a GPIO if at least one of
>> > +        * the backends for specifing a GPIO is available. These are device
>> > +        * tree, ACPI and gpiolib's lookup tables. The latter isn't available if
>> > +        * GPIOLIB is disabled (which is the case here).
>> > +        * So if the provided device is unrelated to device tree and ACPI, we
>> > +        * can be sure that there is no optional GPIO and let gpiod_get_optional
>> > +        * safely return NULL.
>> > +        * Otherwise there is still a chance that there is no GPIO but we cannot
>> > +        * be sure without having to enable a part of GPIOLIB (i.e. the lookup
>> > +        * part). So lets play safe and return an error. (Though there are also
>> > +        * arguments that returning NULL then would be beneficial.)
>> > +        */
>> > +
>> > +       if (IS_ENABLED(CONFIG_OF) && dev && dev->of_node)
>> > +               return false;
>>
>> At first sight, I though this was OK:
>>
>>   1. On ARM with DT, we can assume CONFIG_GPIOLOB=y.
>>
>>   2. I managed to configure an SH kernel with CONFIG_GPIOLOB=n, CONFIG_OF=y,
>>      and CONFIG_SERIAL_SH_SCI=y, but since SH boards with SH-SCI UARTs do
>>      not use DT (yet), the check for dev->of_node (false) should handle
>>      that.
>>
>>   3. However, I managed to do the same for h8300, which does use DT. Hence
>>      if mctrl_gpio would start relying on gpiod_get_optional(), this would
>>      break the sh-sci driver on h8300 :-(
>>      Note that h8300 doesn't have any GPIO drivers (yet?), so
>>      CONFIG_GPIPOLIB=n makes perfect sense!
>
> Thanks for your efforts.

You're welcome.

>> So I'm afraid the only option is to always return NULL, and put the
>> responsability on the shoulders of the system integrator...
>
> The gpio lines could be provided by an i2c gpio adapter, right? So IMHO
> you don't need platform gpios to justify -ENODEV. So I guess that's a
> case where we don't come to an agreement.

While you can enable I2C without further dependencies, no I2C GPIO expander
will be offered... unless you have enabled CONFIG_GPIOLIB first.

>> >  static inline struct gpio_desc *__must_check
>> >  gpiod_get_index_optional(struct device *dev, const char *con_id,
>> >                          unsigned int index, enum gpiod_flags flags)
>> >  {
>> > +       if (__gpiod_no_optional_possible(dev))
>> > +               return NULL;
>> > +
>> >         return ERR_PTR(-ENOSYS);
>>
>> Regardless of the above, given you use the exact same construct in four
>> locations, what about letting __gpiod_no_optional_possible() return the NULL
>> or ERR_PTR itself, and renaming it to e.g. __gpiod_no_optional_return_value()?
>
> I thought about that but didn't find a good name and so considered it
> more clear this way. Another optimisation would be to unconditionally
> define get_optional in terms of get_index_optional which would simplify
> my patch a bit.
>
> I'd consider __gpiod_optional_return_value a better name than
> __gpiod_no_optional_return_value but I'm still not convinced.

No hard feelings about the name from my side.

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
--
To unsubscribe from this list: send the line "unsubscribe linux-gpio" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Uwe Kleine-König March 24, 2017, 9:15 a.m. UTC | #4 
On Fri, Mar 24, 2017 at 09:59:04AM +0100, Geert Uytterhoeven wrote:
> Hi Uwe,
> 
> On Fri, Mar 24, 2017 at 9:39 AM, Uwe Kleine-König
> <u.kleine-koenig@pengutronix.de> wrote:
> > On Fri, Mar 24, 2017 at 09:29:02AM +0100, Geert Uytterhoeven wrote:
> >> On Fri, Mar 24, 2017 at 9:00 AM, Uwe Kleine-König
> >> <u.kleine-koenig@pengutronix.de> wrote:
> >> > From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> >> > Subject: [PATCH] gpiod: let get_optional return NULL in some cases with GPIOLIB disabled
> >> >
> >> > People disagree if gpiod_get_optional should return NULL or
> >> > ERR_PTR(-ENOSYS) if GPIOLIB is disabled. The argument for NULL is that
> >> > the person who decided to disable GPIOLIB is assumed to know that there
> >> > is no GPIO. The reason to stick to ERR_PTR(-ENOSYS) is that it might
> >> > introduce hard to debug problems if that decision is wrong.
> >> >
> >> > So this patch introduces a compromise and let gpiod_get_optional (and
> >> > its variants) return NULL if the device in question cannot have an
> >> > associated GPIO because it is neither instantiated by a device tree nor
> >> > by ACPI.
> >> >
> >> > This should handle most cases that are argued about.
> >> >
> >> > Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> >> > ---
> >> >  include/linux/gpio/consumer.h | 55 ++++++++++++++++++++++++++++++++++++-------
> >> >  1 file changed, 46 insertions(+), 9 deletions(-)
> >> >
> >> > diff --git a/include/linux/gpio/consumer.h b/include/linux/gpio/consumer.h
> >> > index fb0fde686cb1..0ca29889290d 100644
> >> > --- a/include/linux/gpio/consumer.h
> >> > +++ b/include/linux/gpio/consumer.h
> >> > @@ -161,20 +161,48 @@ gpiod_get_index(struct device *dev,
> >> >         return ERR_PTR(-ENOSYS);
> >> >  }
> >> >
> >> > -static inline struct gpio_desc *__must_check
> >> > -gpiod_get_optional(struct device *dev, const char *con_id,
> >> > -                  enum gpiod_flags flags)
> >> > +static inline bool __gpiod_no_optional_possible(struct device *dev)
> >> >  {
> >> > -       return ERR_PTR(-ENOSYS);
> >> > +       /*
> >> > +        * gpiod_get_optional et al can only provide a GPIO if at least one of
> >> > +        * the backends for specifing a GPIO is available. These are device
> >> > +        * tree, ACPI and gpiolib's lookup tables. The latter isn't available if
> >> > +        * GPIOLIB is disabled (which is the case here).
> >> > +        * So if the provided device is unrelated to device tree and ACPI, we
> >> > +        * can be sure that there is no optional GPIO and let gpiod_get_optional
> >> > +        * safely return NULL.
> >> > +        * Otherwise there is still a chance that there is no GPIO but we cannot
> >> > +        * be sure without having to enable a part of GPIOLIB (i.e. the lookup
> >> > +        * part). So lets play safe and return an error. (Though there are also
> >> > +        * arguments that returning NULL then would be beneficial.)
> >> > +        */
> >> > +
> >> > +       if (IS_ENABLED(CONFIG_OF) && dev && dev->of_node)
> >> > +               return false;
> >>
> >> At first sight, I though this was OK:
> >>
> >>   1. On ARM with DT, we can assume CONFIG_GPIOLOB=y.
> >>
> >>   2. I managed to configure an SH kernel with CONFIG_GPIOLOB=n, CONFIG_OF=y,
> >>      and CONFIG_SERIAL_SH_SCI=y, but since SH boards with SH-SCI UARTs do
> >>      not use DT (yet), the check for dev->of_node (false) should handle
> >>      that.
> >>
> >>   3. However, I managed to do the same for h8300, which does use DT. Hence
> >>      if mctrl_gpio would start relying on gpiod_get_optional(), this would
> >>      break the sh-sci driver on h8300 :-(
> >>      Note that h8300 doesn't have any GPIO drivers (yet?), so
> >>      CONFIG_GPIPOLIB=n makes perfect sense!
> >
> > Thanks for your efforts.
> 
> You're welcome.
> 
> >> So I'm afraid the only option is to always return NULL, and put the
> >> responsability on the shoulders of the system integrator...
> >
> > The gpio lines could be provided by an i2c gpio adapter, right? So IMHO
> > you don't need platform gpios to justify -ENODEV. So I guess that's a
> > case where we don't come to an agreement.
> 
> While you can enable I2C without further dependencies, no I2C GPIO expander
> will be offered... unless you have enabled CONFIG_GPIOLIB first.

And that is expected, still the device tree could reference such a GPIO
and thus create a situation where Dmitry's and my judgement disagree.

So I think my suggestion is the best we could do now. It minimizes the
number of cases where we disagree. The next best thing would be to
implement that half gpiolib stuff (i.e. do the full lookup to be sure
there is no gpio) but this comes at a price: We need some time to
implement it and it adds a bit to the kernel size.

Best regards
Uwe
Geert Uytterhoeven March 24, 2017, 9:44 a.m. UTC | #5 
Hi Uwe,

On Fri, Mar 24, 2017 at 10:15 AM, Uwe Kleine-König
<u.kleine-koenig@pengutronix.de> wrote:
> On Fri, Mar 24, 2017 at 09:59:04AM +0100, Geert Uytterhoeven wrote:
>> On Fri, Mar 24, 2017 at 9:39 AM, Uwe Kleine-König
>> <u.kleine-koenig@pengutronix.de> wrote:
>> > On Fri, Mar 24, 2017 at 09:29:02AM +0100, Geert Uytterhoeven wrote:
>> >> On Fri, Mar 24, 2017 at 9:00 AM, Uwe Kleine-König
>> >> <u.kleine-koenig@pengutronix.de> wrote:
>> >> > From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
>> >> > Subject: [PATCH] gpiod: let get_optional return NULL in some cases with GPIOLIB disabled
>> >> >
>> >> > People disagree if gpiod_get_optional should return NULL or
>> >> > ERR_PTR(-ENOSYS) if GPIOLIB is disabled. The argument for NULL is that
>> >> > the person who decided to disable GPIOLIB is assumed to know that there
>> >> > is no GPIO. The reason to stick to ERR_PTR(-ENOSYS) is that it might
>> >> > introduce hard to debug problems if that decision is wrong.
>> >> >
>> >> > So this patch introduces a compromise and let gpiod_get_optional (and
>> >> > its variants) return NULL if the device in question cannot have an
>> >> > associated GPIO because it is neither instantiated by a device tree nor
>> >> > by ACPI.
>> >> >
>> >> > This should handle most cases that are argued about.
>> >> >
>> >> > Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
>> >> > ---
>> >> >  include/linux/gpio/consumer.h | 55 ++++++++++++++++++++++++++++++++++++-------
>> >> >  1 file changed, 46 insertions(+), 9 deletions(-)
>> >> >
>> >> > diff --git a/include/linux/gpio/consumer.h b/include/linux/gpio/consumer.h
>> >> > index fb0fde686cb1..0ca29889290d 100644
>> >> > --- a/include/linux/gpio/consumer.h
>> >> > +++ b/include/linux/gpio/consumer.h
>> >> > @@ -161,20 +161,48 @@ gpiod_get_index(struct device *dev,
>> >> >         return ERR_PTR(-ENOSYS);
>> >> >  }
>> >> >
>> >> > -static inline struct gpio_desc *__must_check
>> >> > -gpiod_get_optional(struct device *dev, const char *con_id,
>> >> > -                  enum gpiod_flags flags)
>> >> > +static inline bool __gpiod_no_optional_possible(struct device *dev)
>> >> >  {
>> >> > -       return ERR_PTR(-ENOSYS);
>> >> > +       /*
>> >> > +        * gpiod_get_optional et al can only provide a GPIO if at least one of
>> >> > +        * the backends for specifing a GPIO is available. These are device
>> >> > +        * tree, ACPI and gpiolib's lookup tables. The latter isn't available if
>> >> > +        * GPIOLIB is disabled (which is the case here).
>> >> > +        * So if the provided device is unrelated to device tree and ACPI, we
>> >> > +        * can be sure that there is no optional GPIO and let gpiod_get_optional
>> >> > +        * safely return NULL.
>> >> > +        * Otherwise there is still a chance that there is no GPIO but we cannot
>> >> > +        * be sure without having to enable a part of GPIOLIB (i.e. the lookup
>> >> > +        * part). So lets play safe and return an error. (Though there are also
>> >> > +        * arguments that returning NULL then would be beneficial.)
>> >> > +        */
>> >> > +
>> >> > +       if (IS_ENABLED(CONFIG_OF) && dev && dev->of_node)
>> >> > +               return false;
>> >>
>> >> At first sight, I though this was OK:
>> >>
>> >>   1. On ARM with DT, we can assume CONFIG_GPIOLOB=y.
>> >>
>> >>   2. I managed to configure an SH kernel with CONFIG_GPIOLOB=n, CONFIG_OF=y,
>> >>      and CONFIG_SERIAL_SH_SCI=y, but since SH boards with SH-SCI UARTs do
>> >>      not use DT (yet), the check for dev->of_node (false) should handle
>> >>      that.
>> >>
>> >>   3. However, I managed to do the same for h8300, which does use DT. Hence
>> >>      if mctrl_gpio would start relying on gpiod_get_optional(), this would
>> >>      break the sh-sci driver on h8300 :-(
>> >>      Note that h8300 doesn't have any GPIO drivers (yet?), so
>> >>      CONFIG_GPIPOLIB=n makes perfect sense!
>> >
>> > Thanks for your efforts.
>>
>> You're welcome.
>>
>> >> So I'm afraid the only option is to always return NULL, and put the
>> >> responsability on the shoulders of the system integrator...
>> >
>> > The gpio lines could be provided by an i2c gpio adapter, right? So IMHO
>> > you don't need platform gpios to justify -ENODEV. So I guess that's a
>> > case where we don't come to an agreement.
>>
>> While you can enable I2C without further dependencies, no I2C GPIO expander
>> will be offered... unless you have enabled CONFIG_GPIOLIB first.
>
> And that is expected, still the device tree could reference such a GPIO
> and thus create a situation where Dmitry's and my judgement disagree.

If the device tree references such a GPIO, and CONFIG_GPIOLIB is not set,
the I2C GPIO expander device will not be bound.
Frank Rowand's DT scripts (http://elinux.org/Device_Tree_frowand) will come
to the rescue, and inform the user which driver(s) to enable.

> So I think my suggestion is the best we could do now. It minimizes the
> number of cases where we disagree. The next best thing would be to
> implement that half gpiolib stuff (i.e. do the full lookup to be sure
> there is no gpio) but this comes at a price: We need some time to
> implement it and it adds a bit to the kernel size.

So I still have to handle -ENOSYS in sh-sci.c, to avoid regressions...

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
--
To unsubscribe from this list: send the line "unsubscribe linux-gpio" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Uwe Kleine-König March 24, 2017, 10:01 a.m. UTC | #6 
Hello Geert,

On Fri, Mar 24, 2017 at 10:44:50AM +0100, Geert Uytterhoeven wrote:
> On Fri, Mar 24, 2017 at 10:15 AM, Uwe Kleine-König
> <u.kleine-koenig@pengutronix.de> wrote:
> > On Fri, Mar 24, 2017 at 09:59:04AM +0100, Geert Uytterhoeven wrote:
> >> On Fri, Mar 24, 2017 at 9:39 AM, Uwe Kleine-König
> >> <u.kleine-koenig@pengutronix.de> wrote:
> >> > On Fri, Mar 24, 2017 at 09:29:02AM +0100, Geert Uytterhoeven wrote:
> >> >> On Fri, Mar 24, 2017 at 9:00 AM, Uwe Kleine-König
> >> >> <u.kleine-koenig@pengutronix.de> wrote:
> >> >> > From: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> >> >> > Subject: [PATCH] gpiod: let get_optional return NULL in some cases with GPIOLIB disabled
> >> >> >
> >> >> > People disagree if gpiod_get_optional should return NULL or
> >> >> > ERR_PTR(-ENOSYS) if GPIOLIB is disabled. The argument for NULL is that
> >> >> > the person who decided to disable GPIOLIB is assumed to know that there
> >> >> > is no GPIO. The reason to stick to ERR_PTR(-ENOSYS) is that it might
> >> >> > introduce hard to debug problems if that decision is wrong.
> >> >> >
> >> >> > So this patch introduces a compromise and let gpiod_get_optional (and
> >> >> > its variants) return NULL if the device in question cannot have an
> >> >> > associated GPIO because it is neither instantiated by a device tree nor
> >> >> > by ACPI.
> >> >> >
> >> >> > This should handle most cases that are argued about.
> >> >> >
> >> >> > Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
> >> >> > ---
> >> >> >  include/linux/gpio/consumer.h | 55 ++++++++++++++++++++++++++++++++++++-------
> >> >> >  1 file changed, 46 insertions(+), 9 deletions(-)
> >> >> >
> >> >> > diff --git a/include/linux/gpio/consumer.h b/include/linux/gpio/consumer.h
> >> >> > index fb0fde686cb1..0ca29889290d 100644
> >> >> > --- a/include/linux/gpio/consumer.h
> >> >> > +++ b/include/linux/gpio/consumer.h
> >> >> > @@ -161,20 +161,48 @@ gpiod_get_index(struct device *dev,
> >> >> >         return ERR_PTR(-ENOSYS);
> >> >> >  }
> >> >> >
> >> >> > -static inline struct gpio_desc *__must_check
> >> >> > -gpiod_get_optional(struct device *dev, const char *con_id,
> >> >> > -                  enum gpiod_flags flags)
> >> >> > +static inline bool __gpiod_no_optional_possible(struct device *dev)
> >> >> >  {
> >> >> > -       return ERR_PTR(-ENOSYS);
> >> >> > +       /*
> >> >> > +        * gpiod_get_optional et al can only provide a GPIO if at least one of
> >> >> > +        * the backends for specifing a GPIO is available. These are device
> >> >> > +        * tree, ACPI and gpiolib's lookup tables. The latter isn't available if
> >> >> > +        * GPIOLIB is disabled (which is the case here).
> >> >> > +        * So if the provided device is unrelated to device tree and ACPI, we
> >> >> > +        * can be sure that there is no optional GPIO and let gpiod_get_optional
> >> >> > +        * safely return NULL.
> >> >> > +        * Otherwise there is still a chance that there is no GPIO but we cannot
> >> >> > +        * be sure without having to enable a part of GPIOLIB (i.e. the lookup
> >> >> > +        * part). So lets play safe and return an error. (Though there are also
> >> >> > +        * arguments that returning NULL then would be beneficial.)
> >> >> > +        */
> >> >> > +
> >> >> > +       if (IS_ENABLED(CONFIG_OF) && dev && dev->of_node)
> >> >> > +               return false;
> >> >>
> >> >> At first sight, I though this was OK:
> >> >>
> >> >>   1. On ARM with DT, we can assume CONFIG_GPIOLOB=y.
> >> >>
> >> >>   2. I managed to configure an SH kernel with CONFIG_GPIOLOB=n, CONFIG_OF=y,
> >> >>      and CONFIG_SERIAL_SH_SCI=y, but since SH boards with SH-SCI UARTs do
> >> >>      not use DT (yet), the check for dev->of_node (false) should handle
> >> >>      that.
> >> >>
> >> >>   3. However, I managed to do the same for h8300, which does use DT. Hence
> >> >>      if mctrl_gpio would start relying on gpiod_get_optional(), this would
> >> >>      break the sh-sci driver on h8300 :-(
> >> >>      Note that h8300 doesn't have any GPIO drivers (yet?), so
> >> >>      CONFIG_GPIPOLIB=n makes perfect sense!
> >>
> >> >> So I'm afraid the only option is to always return NULL, and put the
> >> >> responsability on the shoulders of the system integrator...
> >> >
> >> > The gpio lines could be provided by an i2c gpio adapter, right? So IMHO
> >> > you don't need platform gpios to justify -ENODEV. So I guess that's a
> >> > case where we don't come to an agreement.
> >>
> >> While you can enable I2C without further dependencies, no I2C GPIO expander
> >> will be offered... unless you have enabled CONFIG_GPIOLIB first.
> >
> > And that is expected, still the device tree could reference such a GPIO
> > and thus create a situation where Dmitry's and my judgement disagree.
> 
> If the device tree references such a GPIO, and CONFIG_GPIOLIB is not set,
> the I2C GPIO expander device will not be bound.
> Frank Rowand's DT scripts (http://elinux.org/Device_Tree_frowand) will come
> to the rescue, and inform the user which driver(s) to enable.
> 
> > So I think my suggestion is the best we could do now. It minimizes the
> > number of cases where we disagree. The next best thing would be to
> > implement that half gpiolib stuff (i.e. do the full lookup to be sure
> > there is no gpio) but this comes at a price: We need some time to
> > implement it and it adds a bit to the kernel size.
> 
> So I still have to handle -ENOSYS in sh-sci.c, to avoid regressions...

How much would it hurt to enable GPIOLIB there now? I assume that the
platform has GPIOs and it's only an intermediate step that there is no
driver for it at present? At some point you have to bite the bullet.
Given that mctrl_gpio is a usage of GPIOs that might be now?

Best regards
Uwe
diff mbox

Patch

diff --git a/include/linux/gpio/consumer.h b/include/linux/gpio/consumer.h
index fb0fde686cb1..0ca29889290d 100644
--- a/include/linux/gpio/consumer.h
+++ b/include/linux/gpio/consumer.h
@@ -161,20 +161,48 @@  gpiod_get_index(struct device *dev,
 	return ERR_PTR(-ENOSYS);
 }
 
-static inline struct gpio_desc *__must_check
-gpiod_get_optional(struct device *dev, const char *con_id,
-		   enum gpiod_flags flags)
+static inline bool __gpiod_no_optional_possible(struct device *dev)
 {
-	return ERR_PTR(-ENOSYS);
+	/*
+	 * gpiod_get_optional et al can only provide a GPIO if at least one of
+	 * the backends for specifing a GPIO is available. These are device
+	 * tree, ACPI and gpiolib's lookup tables. The latter isn't available if
+	 * GPIOLIB is disabled (which is the case here).
+	 * So if the provided device is unrelated to device tree and ACPI, we
+	 * can be sure that there is no optional GPIO and let gpiod_get_optional
+	 * safely return NULL.
+	 * Otherwise there is still a chance that there is no GPIO but we cannot
+	 * be sure without having to enable a part of GPIOLIB (i.e. the lookup
+	 * part). So lets play safe and return an error. (Though there are also
+	 * arguments that returning NULL then would be beneficial.)
+	 */
+
+	if (IS_ENABLED(CONFIG_OF) && dev && dev->of_node)
+		return false;
+
+	if (IS_ENABLED(CONFIG_ACPI) && dev && ACPI_COMPANION(dev))
+		return false;
+
+	return true;
 }
 
 static inline struct gpio_desc *__must_check
 gpiod_get_index_optional(struct device *dev, const char *con_id,
 			 unsigned int index, enum gpiod_flags flags)
 {
+	if (__gpiod_no_optional_possible(dev))
+		return NULL;
+
 	return ERR_PTR(-ENOSYS);
 }
 
+static inline struct gpio_desc *__must_check
+gpiod_get_optional(struct device *dev, const char *con_id,
+		   enum gpiod_flags flags)
+{
+	return gpiod_get_index_optional(dev, con_id, 0, flags);
+}
+
 static inline struct gpio_descs *__must_check
 gpiod_get_array(struct device *dev, const char *con_id,
 		enum gpiod_flags flags)
@@ -186,6 +214,9 @@  static inline struct gpio_descs *__must_check
 gpiod_get_array_optional(struct device *dev, const char *con_id,
 			 enum gpiod_flags flags)
 {
+	if (__gpiod_no_optional_possible(dev))
+		return NULL;
+
 	return ERR_PTR(-ENOSYS);
 }
 
@@ -223,17 +254,20 @@  devm_gpiod_get_index(struct device *dev,
 }
 
 static inline struct gpio_desc *__must_check
-devm_gpiod_get_optional(struct device *dev, const char *con_id,
-			  enum gpiod_flags flags)
+devm_gpiod_get_index_optional(struct device *dev, const char *con_id,
+			      unsigned int index, enum gpiod_flags flags)
 {
+	if (__gpiod_no_optional_possible(dev))
+		return NULL;
+
 	return ERR_PTR(-ENOSYS);
 }
 
 static inline struct gpio_desc *__must_check
-devm_gpiod_get_index_optional(struct device *dev, const char *con_id,
-				unsigned int index, enum gpiod_flags flags)
+devm_gpiod_get_optional(struct device *dev, const char *con_id,
+			enum gpiod_flags flags)
 {
-	return ERR_PTR(-ENOSYS);
+	return devm_gpiod_get_index_optional(dev, con_id, 0, flags);
 }
 
 static inline struct gpio_descs *__must_check
@@ -247,6 +281,9 @@  static inline struct gpio_descs *__must_check
 devm_gpiod_get_array_optional(struct device *dev, const char *con_id,
 			      enum gpiod_flags flags)
 {
+	if (__gpiod_no_optional_possible(dev))
+		return NULL;
+
 	return ERR_PTR(-ENOSYS);
 }