gpiolib: Discourage to use formatting strings in line names
PATCH 1/1 gpiolib: Discourage to use formatting strings in line names
Currently the documentation for line names allows to use %u inside the alternative name. This is broken in character device approach from day 1 and being in use solely in sysfs.
Character device interface has a line number as a part of its address, so the users better rely on it. Hence remove the misleading documentation.
On top of that, there are no in-kernel users (out of 6, if I'm correct) for such names and moreover if one exists it won't help in distinguishing lines with the same naming as '%u' will also be in them and we will get a warning in gpiochip_set_desc_names() for such cases.
--- include/linux/gpio/driver.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
include/linux/gpio/driver.h +1 -3
@@ -376,9 +376,7 @@ struct gpio_irq_chip {* @names: if set, must be an array of strings to use as alternative * names for the GPIOs in this chip. Any entry in the array * may be NULL if there is no alias for the GPIO, however the -* array must be @ngpio entries long. A name can include a single printk -* format specifier for an unsigned int. It is substituted by the actual -* number of the gpio. +* array must be @ngpio entries long. * @can_sleep: flag must be set iff get()/set() methods sleep, as they * must while accessing GPIO expander chips over I2C or SPI. This * implies that if the chip supports IRQs, these IRQs need to be threaded
Signature · 1 line
2.45.0
Quoted 7 lines · Andy Shevchenko
Currently the documentation for line names allows to use %u inside the alternative name. This is broken in character device approach from day 1 and being in use solely in sysfs.
Character device interface has a line number as a part of its address, so the users better rely on it. Hence remove the misleading documentation.
Quoted 6 lines
On top of that, there are no in-kernel users (out of 6, if I'm correct) for such names and moreover if one exists it won't help in distinguishing lines with the same naming as '%u' will also be in them and we will get a warning in
gpiochip_set_desc_names()for such cases.Signed-off-by Andy Shevchenko <[email protected]>
Yours,
Linus Walleij
Quoted 10 lines · Linus Walleij
Quoted 7 lines · Andy Shevchenko
Currently the documentation for line names allows to use %u inside the alternative name. This is broken in character device approach from day 1 and being in use solely in sysfs.
Character device interface has a line number as a part of its address, so the users better rely on it. Hence remove the misleading documentation.
Quoted 8 lines
Quoted 6 lines
On top of that, there are no in-kernel users (out of 6, if I'm correct) for such names and moreover if one exists it won't help in distinguishing lines with the same naming as '%u' will also be in them and we will get a warning in
gpiochip_set_desc_names()for such cases.Signed-off-by Andy Shevchenko <[email protected]>Reviewed-by Linus Walleij <[email protected]>
Thank you!
Meanwhile, Cc'ing to Kent as well.
Signature · 2 lines
With Best Regards, Andy Shevchenko
Kent Gibson
Reviewed-byQuoted 16 lines · Andy Shevchenko
Quoted 15 lines · Linus Walleij
Quoted 12 lines · Andy Shevchenko
Currently the documentation for line names allows to use %u inside the alternative name. This is broken in character device approach from day 1 and being in use solely in sysfs.
Character device interface has a line number as a part of its address, so the users better rely on it. Hence remove the misleading documentation.
On top of that, there are no in-kernel users (out of 6, if I'm correct) for such names and moreover if one exists it won't help in distinguishing lines with the same naming as '%u' will also be in them and we will get a warning in
gpiochip_set_desc_names()for such cases.
Quoted 8 lines
Quoted 3 lines
Quoted 1 line
Signed-off-by Andy Shevchenko <[email protected]>Reviewed-by Linus Walleij <[email protected]>Thank you!
Meanwhile, Cc'ing to Kent as well.
I saw it - makes total sense to me too.
Quoted 8 lines · Andy Shevchenko
Quoted 7 lines · Linus Walleij
Quoted 4 lines · Andy Shevchenko
Currently the documentation for line names allows to use %u inside the alternative name. This is broken in character device approach from day 1 and being in use solely in sysfs.
Quoted 7 lines
Quoted 7 lines
Quoted 7 lines
Character device interface has a line number as a part of its address, so the users better rely on it. Hence remove the misleading documentation.
On top of that, there are no in-kernel users (out of 6, if I'm correct) for such names and moreover if one exists it won't help in distinguishing lines with the same naming as '%u' will also be in them and we will get a warning in
gpiochip_set_desc_names()for such cases.
Dunno if I need to elaborate this more, but just in case here is one:
Even if one puts '%u' to one line and avoids putting it into other:
"gpio%u.foo"
"gpioX.foo"
it means that it was already in mind to distinguish them beforehand, diminishing the '%u' appearance in the first place. I.e. one may do
"foo X"
"foo Y"
to begin with. Besides that repetitive namings are discouraged and most likely have no value but confusion.
For example,
"gpio%u.SPI CS"
"gpio%u.SPI CS"
would be rather
"SPI CS 0"
"SPI CS 1"
which is much more clearer to the user.
Quoted 11 lines
Quoted 3 lines
Quoted 1 line
Signed-off-by Andy Shevchenko <[email protected]>Reviewed-by Linus Walleij <[email protected]>Thank you!
Meanwhile, Cc'ing to Kent as well.
--
With Best Regards,
Andy Shevchenko
Signature · 2 lines
With Best Regards, Andy Shevchenko
Bartosz Golaszewski
AppliedFrom: Bartosz Golaszewski <[email protected]>
Quoted 8 lines · Andy Shevchenko
Currently the documentation for line names allows to use %u inside the alternative name. This is broken in character device approach from day 1 and being in use solely in sysfs.
Character device interface has a line number as a part of its address, so the users better rely on it. Hence remove the misleading documentation.
[...]
Applied, thanks!
[1/1] gpiolib: Discourage to use formatting strings in line names
commit: 2b5ae9c7d9e5ef4bc52c932fdf10328feb5167c6
Best regards,
Signature · 1 line
Bartosz Golaszewski <[email protected]>