[PATCH] gpiolib: enable CONFIG_GPIOLIB_LEGACY even for !GPIOLIB

[PATCH] gpiolib: enable CONFIG_GPIOLIB_LEGACY even for !GPIOLIB

[Arnd Bergmann]

From: Arnd Bergmann <arnd@arndb.de>

A few drivers that use the legacy GPIOLIB interfaces can be enabled
even when GPIOLIB is disabled entirely. With my previous patch this
now causes build failures like:

   drivers/nfc/s3fwrn5/uart.c: In function 's3fwrn82_uart_parse_dt':
        drivers/nfc/s3fwrn5/uart.c:100:14: error: implicit declaration of function 'gpio_is_valid'; did you mean 'uuid_is_valid'? [-Werror=implicit-function-declaration]

These did not show up in my randconfig tests because randconfig almost
always has GPIOLIB selected by some other driver, and I did most
of the testing with follow-up patches that address the failures
properly.

Move the symbol outside of the 'if CONFIG_GPIOLIB' block for the moment
to avoid the build failures. It can be moved back and turned off by
default once all the driver specific changes are merged.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202507261934.yIHeUuEQ-lkp@intel.com/
Fixes: 678bae2eaa81 ("gpiolib: make legacy interfaces optional")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 drivers/gpio/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
index 8bda3c9d47b4..c48f9badb513 100644
--- a/drivers/gpio/Kconfig
+++ b/drivers/gpio/Kconfig
@@ -12,11 +12,11 @@ menuconfig GPIOLIB
 
 	  If unsure, say N.
 
-if GPIOLIB
-
 config GPIOLIB_LEGACY
 	def_bool y
 
+if GPIOLIB
+
 config GPIOLIB_FASTPATH_LIMIT
 	int "Maximum number of GPIOs for fast path"
 	range 32 512
-- 
2.39.5

Re: [PATCH] gpiolib: enable CONFIG_GPIOLIB_LEGACY even for !GPIOLIB

[Bartosz Golaszewski]

From: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>


On Sat, 26 Jul 2025 23:10:43 +0200, Arnd Bergmann wrote:
> A few drivers that use the legacy GPIOLIB interfaces can be enabled
> even when GPIOLIB is disabled entirely. With my previous patch this
> now causes build failures like:
> 
>    drivers/nfc/s3fwrn5/uart.c: In function 's3fwrn82_uart_parse_dt':
>         drivers/nfc/s3fwrn5/uart.c:100:14: error: implicit declaration of function 'gpio_is_valid'; did you mean 'uuid_is_valid'? [-Werror=implicit-function-declaration]
> 
> [...]

Applied, thanks!

[1/1] gpiolib: enable CONFIG_GPIOLIB_LEGACY even for !GPIOLIB
      https://git.kernel.org/brgl/linux/c/a86240a37d43fc22b4e4953242fca8d90df2c555

Best regards,
-- 
Bartosz Golaszewski <bartosz.golaszewski@linaro.org>

Re: [PATCH] gpiolib: enable CONFIG_GPIOLIB_LEGACY even for !GPIOLIB

[Geert Uytterhoeven]

Hi Arnd,

On Sat, 26 Jul 2025 at 23:11, Arnd Bergmann <arnd@kernel.org> wrote:
> From: Arnd Bergmann <arnd@arndb.de>
>
> A few drivers that use the legacy GPIOLIB interfaces can be enabled
> even when GPIOLIB is disabled entirely. With my previous patch this
> now causes build failures like:
>
>    drivers/nfc/s3fwrn5/uart.c: In function 's3fwrn82_uart_parse_dt':
>         drivers/nfc/s3fwrn5/uart.c:100:14: error: implicit declaration of function 'gpio_is_valid'; did you mean 'uuid_is_valid'? [-Werror=implicit-function-declaration]
>
> These did not show up in my randconfig tests because randconfig almost
> always has GPIOLIB selected by some other driver, and I did most
> of the testing with follow-up patches that address the failures
> properly.
>
> Move the symbol outside of the 'if CONFIG_GPIOLIB' block for the moment
> to avoid the build failures. It can be moved back and turned off by
> default once all the driver specific changes are merged.
>
> Reported-by: kernel test robot <lkp@intel.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202507261934.yIHeUuEQ-lkp@intel.com/
> Fixes: 678bae2eaa81 ("gpiolib: make legacy interfaces optional")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>

Thanks for your patch, which is now commit a86240a37d43fc22 ("gpiolib:
enable CONFIG_GPIOLIB_LEGACY even for !GPIOLIB") in linus/master

> --- a/drivers/gpio/Kconfig
> +++ b/drivers/gpio/Kconfig
> @@ -12,11 +12,11 @@ menuconfig GPIOLIB
>
>           If unsure, say N.
>
> -if GPIOLIB
> -
>  config GPIOLIB_LEGACY
>         def_bool y
>
> +if GPIOLIB
> +
>  config GPIOLIB_FASTPATH_LIMIT
>         int "Maximum number of GPIOs for fast path"
>         range 32 512

This won't work for everything.
While I now get CONFIG_GPIOLIB_LEGACY=y in all m68k defconfigs, and
simple inline functions like gpio_is_valid() are now available, more
complex functions will still fail, as drivers/gpio/gpiolib-legacy.c
is not built.

drivers/Makefile:

    obj-$(CONFIG_GPIOLIB)           += gpio/

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

Re: [PATCH] gpiolib: enable CONFIG_GPIOLIB_LEGACY even for !GPIOLIB

[Arnd Bergmann]

On Tue, Jul 29, 2025, at 12:47, Geert Uytterhoeven wrote:
>>
>> -if GPIOLIB
>> -
>>  config GPIOLIB_LEGACY
>>         def_bool y
>>
>> +if GPIOLIB
>> +
>>  config GPIOLIB_FASTPATH_LIMIT
>>         int "Maximum number of GPIOs for fast path"
>>         range 32 512
>
> This won't work for everything.
> While I now get CONFIG_GPIOLIB_LEGACY=y in all m68k defconfigs, and
> simple inline functions like gpio_is_valid() are now available, more
> complex functions will still fail, as drivers/gpio/gpiolib-legacy.c
> is not built.
>
> drivers/Makefile:
>
>     obj-$(CONFIG_GPIOLIB)           += gpio/
>

Hi Geert,

Do you have an example config that shows this problem? 
I've tried a couple of configurations on m68k now but are unable
to reproduce this, using 'defconfig' (without GPIOLIB) and
'm5475evb_defconfig' (with GPIOLIB).

The intention of this patch (in combination with the previous one)
was that the legacy interfaces would still behave exactly like
before, falling back to the stubs when GPIOLIB is disabled.

    Arnd

Re: [PATCH] gpiolib: enable CONFIG_GPIOLIB_LEGACY even for !GPIOLIB

[Geert Uytterhoeven]

Hi Arnd,

On Tue, 29 Jul 2025 at 13:58, Arnd Bergmann <arnd@arndb.de> wrote:
> On Tue, Jul 29, 2025, at 12:47, Geert Uytterhoeven wrote:
> >>
> >> -if GPIOLIB
> >> -
> >>  config GPIOLIB_LEGACY
> >>         def_bool y
> >>
> >> +if GPIOLIB
> >> +
> >>  config GPIOLIB_FASTPATH_LIMIT
> >>         int "Maximum number of GPIOs for fast path"
> >>         range 32 512
> >
> > This won't work for everything.
> > While I now get CONFIG_GPIOLIB_LEGACY=y in all m68k defconfigs, and
> > simple inline functions like gpio_is_valid() are now available, more
> > complex functions will still fail, as drivers/gpio/gpiolib-legacy.c
> > is not built.
> >
> > drivers/Makefile:
> >
> >     obj-$(CONFIG_GPIOLIB)           += gpio/

> Do you have an example config that shows this problem?
> I've tried a couple of configurations on m68k now but are unable
> to reproduce this, using 'defconfig' (without GPIOLIB) and
> 'm5475evb_defconfig' (with GPIOLIB).
>
> The intention of this patch (in combination with the previous one)
> was that the legacy interfaces would still behave exactly like
> before, falling back to the stubs when GPIOLIB is disabled.

I haven't seen any actual failures.  When discovering
CONFIG_GPIOLIB_LEGACY=y in all m68k defconfigs, my initial worry
was that it would increase kernel size by needlessly including
gpiolib-legacy.o. When that didn't turn out to be true, I started
wondering how your commit would fix anything without including
gpiolib-legacy.o.  Looks like any users just uses the simple static
inlines...
Sorry for confusing you.

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

Re: [PATCH] gpiolib: enable CONFIG_GPIOLIB_LEGACY even for !GPIOLIB

[Arnd Bergmann]

On Tue, Jul 29, 2025, at 14:12, Geert Uytterhoeven wrote:
> On Tue, 29 Jul 2025 at 13:58, Arnd Bergmann <arnd@arndb.de> wrote:
>> On Tue, Jul 29, 2025, at 12:47, Geert Uytterhoeven wrote:
>> Do you have an example config that shows this problem?
>> I've tried a couple of configurations on m68k now but are unable
>> to reproduce this, using 'defconfig' (without GPIOLIB) and
>> 'm5475evb_defconfig' (with GPIOLIB).
>>
>> The intention of this patch (in combination with the previous one)
>> was that the legacy interfaces would still behave exactly like
>> before, falling back to the stubs when GPIOLIB is disabled.
>
> I haven't seen any actual failures.  When discovering
> CONFIG_GPIOLIB_LEGACY=y in all m68k defconfigs, my initial worry
> was that it would increase kernel size by needlessly including
> gpiolib-legacy.o. When that didn't turn out to be true, I started
> wondering how your commit would fix anything without including
> gpiolib-legacy.o.  Looks like any users just uses the simple static
> inlines...
> Sorry for confusing you.

No worries, thanks for paying attention to incoming changes!

If you want to see what the actual plan is, have a look at

https://git.kernel.org/pub/scm/linux/kernel/git/arnd/playground.git/log/?h=config-gpio-legacy

which ends with making GPIOLIB_LEGACY actually optional.

Any driver that actually uses the legacy gpiolib interfaces
at that point is already specific to one of the few platforms
that still have legacy gpiochips (sh, sa1100, pxa, s3c64xx,
orion5x, mv78xx0, coldfire, alchemy, txx9, bcm47xx, bcm63xx,
rb532, olpc, and a few x86 atom boards), or it has an explicit
dependency on GPIOLIB_LEGACY.

The arm boards are already on their way out, but the others
could probably use some help converting to gpio descriptors.

     Arnd

Re: [PATCH] gpiolib: enable CONFIG_GPIOLIB_LEGACY even for !GPIOLIB

[Rob Herring]

On Sat, Jul 26, 2025 at 11:10:43PM +0200, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
> 
> A few drivers that use the legacy GPIOLIB interfaces can be enabled
> even when GPIOLIB is disabled entirely. With my previous patch this
> now causes build failures like:
> 
>    drivers/nfc/s3fwrn5/uart.c: In function 's3fwrn82_uart_parse_dt':
>         drivers/nfc/s3fwrn5/uart.c:100:14: error: implicit declaration of function 'gpio_is_valid'; did you mean 'uuid_is_valid'? [-Werror=implicit-function-declaration]
> 
> These did not show up in my randconfig tests because randconfig almost
> always has GPIOLIB selected by some other driver, and I did most
> of the testing with follow-up patches that address the failures
> properly.
> 
> Move the symbol outside of the 'if CONFIG_GPIOLIB' block for the moment
> to avoid the build failures. It can be moved back and turned off by
> default once all the driver specific changes are merged.
> 
> Reported-by: kernel test robot <lkp@intel.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202507261934.yIHeUuEQ-lkp@intel.com/
> Fixes: 678bae2eaa81 ("gpiolib: make legacy interfaces optional")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>  drivers/gpio/Kconfig | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)

This change causes all of the GPIO submenu to show up directly in the 
already way too long 'Device Drivers' menu.

> 
> diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
> index 8bda3c9d47b4..c48f9badb513 100644
> --- a/drivers/gpio/Kconfig
> +++ b/drivers/gpio/Kconfig
> @@ -12,11 +12,11 @@ menuconfig GPIOLIB
>  
>  	  If unsure, say N.
>  
> -if GPIOLIB
> -
>  config GPIOLIB_LEGACY
>  	def_bool y

Perhaps this has to be before "menuconfig GPIOLIB"?

>  
> +if GPIOLIB
> +
>  config GPIOLIB_FASTPATH_LIMIT
>  	int "Maximum number of GPIOs for fast path"
>  	range 32 512
> -- 
> 2.39.5
> 

Re: [PATCH] gpiolib: enable CONFIG_GPIOLIB_LEGACY even for !GPIOLIB

[Bartosz Golaszewski]

On Thu, Aug 14, 2025 at 12:26 AM Rob Herring <robh@kernel.org> wrote:
>
> On Sat, Jul 26, 2025 at 11:10:43PM +0200, Arnd Bergmann wrote:
> > From: Arnd Bergmann <arnd@arndb.de>
> >
> > A few drivers that use the legacy GPIOLIB interfaces can be enabled
> > even when GPIOLIB is disabled entirely. With my previous patch this
> > now causes build failures like:
> >
> >    drivers/nfc/s3fwrn5/uart.c: In function 's3fwrn82_uart_parse_dt':
> >         drivers/nfc/s3fwrn5/uart.c:100:14: error: implicit declaration of function 'gpio_is_valid'; did you mean 'uuid_is_valid'? [-Werror=implicit-function-declaration]
> >
> > These did not show up in my randconfig tests because randconfig almost
> > always has GPIOLIB selected by some other driver, and I did most
> > of the testing with follow-up patches that address the failures
> > properly.
> >
> > Move the symbol outside of the 'if CONFIG_GPIOLIB' block for the moment
> > to avoid the build failures. It can be moved back and turned off by
> > default once all the driver specific changes are merged.
> >
> > Reported-by: kernel test robot <lkp@intel.com>
> > Closes: https://lore.kernel.org/oe-kbuild-all/202507261934.yIHeUuEQ-lkp@intel.com/
> > Fixes: 678bae2eaa81 ("gpiolib: make legacy interfaces optional")
> > Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> > ---
> >  drivers/gpio/Kconfig | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
>
> This change causes all of the GPIO submenu to show up directly in the
> already way too long 'Device Drivers' menu.
>
> >
> > diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
> > index 8bda3c9d47b4..c48f9badb513 100644
> > --- a/drivers/gpio/Kconfig
> > +++ b/drivers/gpio/Kconfig
> > @@ -12,11 +12,11 @@ menuconfig GPIOLIB
> >
> >         If unsure, say N.
> >
> > -if GPIOLIB
> > -
> >  config GPIOLIB_LEGACY
> >       def_bool y
>
> Perhaps this has to be before "menuconfig GPIOLIB"?
>

Hi Arnd,

This is still broken in next, will you have some time to spend on it
or should I just go with what Rob proposed?

Bart

> >
> > +if GPIOLIB
> > +
> >  config GPIOLIB_FASTPATH_LIMIT
> >       int "Maximum number of GPIOs for fast path"
> >       range 32 512
> > --
> > 2.39.5
> >